Textbook of Stereotactic and Functional Neurosurgery
Andres M. Lozano, Philip L. Gildenberg, Ronald R. Tasker (Eds.)
Textbook of Stereotactic and Functional Neurosurgery With 1088 Figures and 232 Tables
Editors: Andres M. Lozano Professor of Surgery and RR Tasker Chair in Functional Neurosurgery University of Toronto Senior Scientist, Toronto Western Research Institute President, World Society for Stereotactic and Functional Neurosurgery Canadian Research Chair in Neuroscience (Tier 1) 399 Bathurst St. WW 4-447 Toronto, Ontario M5T 2S8 Canada
[email protected] Philip L. Gildenberg Houston Stereotactic Concepts, Inc. 2260 West Holcombe Boulevard Suite 309 Houston, Texas 77030 USA
[email protected] Ronald R. Tasker University of Toronto 399 Bathurst St. WW 4-447 Toronto, Ontario M5T 2S8 Canada A C.I.P. Catalog record for this book is available from the Library of Congress ISBN: 978-3-540-69959-0 This publication is available also as: Electronic publication under ISBN 978-3-540-69960-6 and Print and electronic bundle under ISBN 978-3-540-70779-0 This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in other ways, and storage in data banks. Duplication of this publication or parts thereof is only permitted under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer‐Verlag. Violations are liable for prosecution under the German Copyright Law. ß Springer‐Verlag Berlin Heidelberg 2009 The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Product liability: The publishers cannot guarantee the accuracy of any information about the application of operative techniques and medications contained in this book. In every individual case the user must check such information by consulting the relevant literature. Springer is part of Springer Science+Business Media springer.com Publishing Editor: Gabriele Schro¨der MRW Editor: Sandra Fabiani Printed on acid‐free paper
SPIN: 11534778
2109 – 5 4 3 2 1 0
Andres Lozano dedicates this work To Marie, Alexander and Christopher
Philip L. Gildenberg, MD, PhD, dedicates this ‘‘Textbook of Stereotactic and Functional Neurosurgery’’ to Patricia Franklin Gildenberg, who has been a significant collaborator in everything he has written since 1973. Her contributions and dedication to both the American and World Societies for Stereotactic and Functional Neurosurgery have been important in maintaining the integrity of those organizations throughout the years.
I would like to dedicate my contribution to this book to my late wife, Mary Morris Craig Tasker, who made my career possible. Ronald R. Tasker OC, MD, FRCSC
Preface
This second edition of the Textbook of Stereotactic and Functional Neurosurgery appears just 11 years after the first. The first edition of the Textbook marked the 50th anniversary of the birth of human stereotactic neurosurgery. After an initial period of rapid growth, the field became quiescent in the 1960s when treatment of Parkinson’s disease migrated from stereotactic surgery to medication, primarily L-dopa. When it became apparent that L-dopa was not the final answer to the management of Parkinson’S disease, human stereotaxis had a rebirth, and the first edition followed a few years later. Many patients returned once again to surgical management in the early 1990s, so many neurosurgeons returned to the fields of stereotactic and functional neurosurgery. In the meantime, computer science and imaging techniques had advanced to foster the development of frameless image guided surgery, as well as stereotactic radiosurgery. Consequently, many neurosurgeons were becoming involved with stereotactic surgery, either returning to the field or being introduced to it for the first time. Both groups needed a comprehensive review of the then-current state of the field which was provided by the first edition of the Textbook of Stereotactic and Functional Neurosurgery. During the past 11 years there have been additional significant advances in this field. Computers have advanced to the point where exotic complex surgical programs have become commonplace and computers have become commonly used in the operating room, both for image guidance and physiological verification of targets. Image guided surgery is being used by most neurosurgeons, many of whom may not have had a prior background in stereotactic surgery. Deep brain stimulation for the treatment of motor disorders has become technically feasible and is commonly practiced at many stereotactic centers, using some techniques that were still under development a decade ago and technically advanced stimulating devices. Radiosurgical techniques involve more complex dosimetry than was available earlier, again because of advances in computer technology. Many new possibilities are being opened for both neurosurgical guidance and functional surgery involving targets that had not previously been recognized. These advances have proceeded so rapidly and broadly across several scientific fields that we have a need for a new edition of the Textbook of Stereotactic and Functional Neurosurgery. It is meant to provide a broad background for neurosurgeons, neurologists, and radiation oncologists, radiation physicists, and other scientists in the ever expanding fields of stereotactic neurosurgery, functional neurosurgery, and stereotactic radiosurgery. These fields have become so broad that no other publication covers the information necessary for all those specialists in a single reference source. This book will provide an update on the significant advances, information, and knowledge that have developed during this past decade has and have been scattered through the literature of these disparate specialties. The first edition was edited by Philip L. Gildenberg and Ronald Tasker. Both those stereotactic neurosurgeons retired from active clinical practice in the past 10 years, so they recruited Andres Lozano to be the primary editor. They felt that he brings to this position a unique combination of clinical #
Springer-Verlag Berlin/Heidelberg 2009
xii
Preface
expertise, extensive knowledge of background laboratory information, a record of providing innovation to functional neurosurgery, and has provided leadership for a new generation of stereotactic neurosurgeons. The editors wish to thank McGraw-Hill, who published the first edition, for relinquishing the copyright to Dr. Gildenberg, which opened the door for Springer to publish the second edition. The difference in the editorial process between the first edition and the second edition in some ways reflects the development of computers during that time, and consequently the advances in the fields of stereotactic and functional neurosurgery. The authors who contributed to the first edition sent hard copy printed manuscripts to the editors who would then read for content and copy edit, a timeconsuming and inefficient process. Manuscripts were then mailed back to the authors for final approval and/or correction and sent back to the Editors, possibly several times, who relayed them to the publisher when they were perfected. All manuscript preparation and submission for the second edition were done electronically, so no paper changed hands, which sometimes introduced new complexities, but ultimately proved more efficient, not to mention friendly to the environment. It is noteworthy that between the time of the first edition and the second edition, the operating room also became digitized, with consequent improvement in the display of diagnostic studies, processing information from those studies, and integrating that information into the surgical plan. The editors thank Garbiele Schro¨der and Stephanie Benko of Springer, who made this Textbook possible. The editors especially wish to thank Andrew Spencer and his staff at Springer, who served as the publisher’s representative to this project and maintained the flow of digital information between authors, editors and publisher. The editors also wish to thank Joan Richardson, who served as a liaison between the authors, editors, and publisher’s staff, and provided an accurate update on the progress and problems of each chapter and author. She managed the organization of 195 manuscripts with over 250 authors, with occasional changes in authors or topics, to facilitate communication between the editors and authors.
Table of Contents
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxix
Section 1 History of Stereotactic Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 History of Stereotactic Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 P. L. Gildenberg . J. K. Krauss
2 History of the Stereotactic Societies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 P. L. Gildenberg . J. K. Krauss
3 History of Stereotactic Surgery in US . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 P. L. Gildenberg
4 History of Stereotactic Surgery in Germany . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 J. K. Krauss
5 History of Stereotactic Surgery in Japan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 C. Ohye
6 History of Stereotactic Neurosurgery in the Nordic Countries . . . . . . . . . . . . . . . . . . 65 B. A. Meyerson . B. Linderoth
7 A Brief History of Stereotactic Neurosurgery in Switzerland . . . . . . . . . . . . . . . . . . . 73 E. Taub
8 History of Stereotactic Surgery in Great Britain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 E. A. C. Pereira . A. L. Green . D. Nandi . T. Z. Aziz
9 History of Stereotactic Surgery in France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 A. L. Benabid . S. Chabardes . E. Seigneuret
10 History of Stereotactic and Functional Neurosurgery in Canada . . . . . . . . . . . . . . . 113 A. G. Parrent
xiv
Table of contents
11 History of Stereotactic Surgery in China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 F.-C. Lee . B. Sun . J. Zhang . K. Zhang . F.-G. Meng
12 History of Stereotactic Surgery in India . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 P. K. Doshi
13 History of Stereotactic Surgery in Korea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171 S. S. Chung
14 History of Stereotactic Surgery in Spain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 J. Guridi . M. Manrique
15 History of Stereotactic Neurosurgery in Italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 A. Franzini . V. A. Sironi . G. Broggi
16 History of Stereotactic and Functional Neurosurgery in Brazil . . . . . . . . . . . . . . . . 197 O. Vilela Filho
Section 2 Imaging in Stereotactic Surgery . . . . . . . . . . . . . . . . . . . . . . . 249 17 General Imaging Modalities: Basic Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251 A. A. Gorgulho . W. Ishida . A. A. F. De Salles
18 CT/MRI Technology: Basic Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269 M. I. Hariz . L. Zrinzo
19 CT/MRI Safety in Functional Neurosurgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279 M. Schulder . A. Oubre´
20 Functional MRI in Image Guided Neurosurgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287 T. Sankar . G. R. Cosgrove
21 Angiography, MRA in Image Guided Neurosurgery . . . . . . . . . . . . . . . . . . . . . . . . . 299 P. Jabbour . S. Tjoumakaris . R. Rosenwasser
22 Diagnostic PET in Image Guided Neurosurgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307 B. Ballanger . T. van Eimeren . A. P. Strafella
23 Neurophysiologic Mapping for Glioma Surgery: Preservation of Functional Areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325 R. M. Richardson . M. S. Berger
Table of contents
24 Image Reconstruction and Fusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335 B. A. Kall
Section 3 Stereotactic Targeting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345 25 Printed Stereotactic Atlases, Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347 R. J. Coffey
26 Electronic Stereotactic Atlases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373 J. Yelnik . E. Bardinet . D. Dormont
27 Anatomical and Probabilistic Functional Atlases in Stereotactic and Functional Neurosurgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395 W. L. Nowinski
28 Accuracy in Stereotactic and Image Guidance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443 A. Hartov . D. W. Roberts
29 Development of a Classic: The Todd-Wells Apparatus, the BRW, and the CRW Stereotactic Frames . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453 J. Arle
30 Leksell Stereotactic Apparatus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 469 L. D. Lunsford . D. Kondziolka . D. Leksell
31 The Riechert/Mundinger Stereotactic Apparatus . . . . . . . . . . . . . . . . . . . . . . . . . . . 487 J. K. Krauss
32 The Talairach Stereotactic System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 495 A. L. Benabid . S. Chabardes . E. Seigneuret . D. Hoffmann . J. F. LeBas
33 Laitinen Stereotactic Apparatus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 511 M. I. Hariz . L. V. Laitinen
34 Miniframe Stereotactic Apparatus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 521 M. A. Madera . W. D. Tobler
Section 4 Image Guided Neurosurgery . . . . . . . . . . . . . . . . . . . . . . . . . . 533 35 Engineering Aspects of Electromagnetic Localization in Image Guided Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 535 E. C. Parker . P. J. Kelly
xv
xvi
Table of contents
36 The History, Current Status, and Future of the StealthStation Treatment Guidance System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 543 R. Bucholz . L. McDurmont
37 BrainLab Image Guided System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 567 J. F. Fraser . T. H. Schwartz . M. G. Kaplitt
38 Robotic Neurosurgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 583 P. L. Gildenberg
39 MRI in Image Guided Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 599 M. Schulder . L. Jarchin
40 CT in Image Guided Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 619 D. Kondziolka . L. D. Lunsford
41 Impedance Recording in Central Nervous System Surgery . . . . . . . . . . . . . . . . . . . 631 R. J. Andrews . J. Li . S. A. Kuhn . J. Walter . R. Reichart
42 Stereotactic and Image-Guided Biopsy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 645 J. B. Elder . A. P. Amar . M. L. J. Apuzzo
43 Pathology Techniques in Stereotactic and Image Guided Biopsy . . . . . . . . . . . . . . 663 P. T. Chandrasoma . N. E. Klipfel
44 Stereotactic and Image Guided Craniotomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 679 E. C. Parker . P. J. Kelly
45 Image Guided Craniotomy for Brain Tumor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 699 I. E. McCutcheon
46 Virtual Reality in the Operating Room . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 725 P. L. Gildenberg
47 Comprehensive Brain Tumor Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 735 M. Tamber . M. Bernstein
48 Novel Therapies for Brain Tumors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 749 G. Al-Shamy . R. Sawaya
49 Image-Guided Management of Brain Abscess . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 769 E. Taub . A. M. Lozano
Table of contents
50 Image-Guided Management of Brain Stem Lesions . . . . . . . . . . . . . . . . . . . . . . . . . 779 M. Levivier
51 Stereotactic Approaches to the Brain Stem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 789 L. U. Zrinzo . D. G. T. Thomas
52 Image Guided Management of Intracerebral Hematoma . . . . . . . . . . . . . . . . . . . . . 797 A. Losiniecki . G. Mandybur
53 Technical Aspects of Image-Guided Neuroendoscopy . . . . . . . . . . . . . . . . . . . . . . . 807 J. D. Caird . J. M. Drake
54 Intraoperative Image Guidance in Skull Base Tumors . . . . . . . . . . . . . . . . . . . . . . . 815 D. Omahen . F. Doglietto . D. Mukherjee . F. Gentili
55 Image Guided Management of Cerebral Metastases . . . . . . . . . . . . . . . . . . . . . . . . 831 P. Kongkham . M. Bernstein
Section 5 Stereotactic Radiosurgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . 851 56 Radiobiology of Stereotactic Radiosurgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 853 D. C. Shrieve . J. S. Loeffler
57 Overview of Radiosurgery Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 867 M. Schulder
58 Gamma Knife: Technical Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 897 D. J. Schlesinger . C. P. Yen . C. Lindquist . L. Steiner
59 Linac Radiosurgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 929 W. A. Friedman . F. J. Bova
60 CyberKnife: Technical Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 949 J. R. Adler . D. W. Schaal . A. Muacevic
61 Proton Beam Radiotherapy: Technical and Clinical Aspects . . . . . . . . . . . . . . . . . . 957 S. Y. Woo
62 IMRT: Technical and Clinical Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 965 M. P. Carol
63 What Every Neurosurgeon Should Know About Stereotactic Radiosurgery . . . . . . 977 P. M. Black . F. Tariq
xvii
xviii
Table of contents
64 Radiosensitizers in Neurooncology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 987 D. Khuntia . A. Chakravarti . H. I. Robins . K. Palanichamy . M. P. Mehta
65 Gamma Knife: Clinical Experience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1007 A. Niranjan . L. D. Lunsford . J. C. Flickinger . J. Novotny . J. Bhatnagar . D. Kondziolka
66 Gamma Knife: Clinical Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1037 L. Steiner . C. P. Yen . J. Jagannathan . D. Schlesinger . M. Steiner
67 Linac Radiosurgery: Technical Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1087 F. J. Bova . W. A. Friedman
68 Cyberknife: Clinical Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1111 F. C. Henderson Sr . W. Jean . N. Nasr . G. Gagnon
69 Proton Beam Radiosurgery: Clinical Experience . . . . . . . . . . . . . . . . . . . . . . . . . . . 1131 H. A. Shih . P. H. Chapman . J. S. Loeffler
70 Radiosurgery for Metastases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1139 M. Maarouf . C. Bu¨hrle . M. Kocher . V. Sturm
71 Focused and Conventional Radiation for Acoustic Nerve Tumors . . . . . . . . . . . . . 1151 R. Den . S. H. Paek . D. W. Andrews
72 Radiosurgery for Pituitary Tumors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1171 J. P. Sheehan . J. Jagannathan . W. J. Elias . E. R. Laws
73 Radiosurgery for Functional Neurosurgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1191 D. Kondziolka
74 Whole Body and Spinal Radiosurgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1203 P. C. Gerszten
75 Gamma Knife Radiosurgery: Technical Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1225 D. Kondziolka . A. Niranjan . J. Novotny . J. Bhatanagar . L. D. Lunsford
Section 6 Functional Neurosurgery – Technical Aspects . . . . . . . . . 1237 76 Image Guided Functional Neurosurgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1239 S. Khan . N. K. Patel . E. White . P. Plaha . S. Ashton . S. S. Gill
77 Evoked Potentials in Functional Neurosurgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1255 J. L. Shils . J. E. Arle
Table of contents
78 Microelectrode Recording in Functional Neurosurgery . . . . . . . . . . . . . . . . . . . . . 1283 W. D. Hutchison . J. O. Dostrovsky . M. Hodaie . K. D. Davis . A. M. Lozano . R. R. Tasker
79 Impedance Recording in Functional Neurosurgery . . . . . . . . . . . . . . . . . . . . . . . . . 1325 L. Zrinzo . M. I. Hariz
80 Anesthesia for Functional Neurosurgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1331 P. H. Manninen . N. Apichatibutra
81 Lesions Versus Implanted Stimulators in Functional Neurosurgery . . . . . . . . . . . 1349 W. S. Anderson . R. E. Clatterbuck . K. Kobayashi . J.-H. Kim . F. A. Lenz
82 Radiofrequency Lesions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1359 E. R. Cosman Sr. . E. R. Cosman Jr.
83 Stimulation Physiology in Functional Neurosurgery . . . . . . . . . . . . . . . . . . . . . . . 1383 A. W. Laxton . J. O. Dostrovsky . A. M. Lozano
84 Stimulation Technology in Functional Neurosurgery . . . . . . . . . . . . . . . . . . . . . . . 1401 B. H. Kopell . A. Machado . C. Butson
85 Therapeutic Lesions Through Chronically Implanted Deep Brain Stimulation Electrodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1427 S. Raoul . D. Leduc . C. Deligny . Y. Lajat
Section 7 Functional Neurosurgery for Movement and Motor Disorders – Clinical Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . 1443 86 Surgery for Movement Disorders: An Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . 1445 K. M. Prakash . A. E. Lang
87 History of Surgery for Movement Disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1467 A. G. Parrent
88 Psychiatric Considerations in Management of Movement Disorders . . . . . . . . . . 1487 M. Zurowski . V. Voon . V. Valerie
89 Pathophysiology of Parkinson’s Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1497 M. R. DeLong . T. Wichmann
90 Medical Management of Parkinson’s Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1507 E. V. Encarnacion . R. A. Hauser
xix
xx
Table of contents
91 Patient Selection for Surgery for Parkinson’s Disease . . . . . . . . . . . . . . . . . . . . . . 1529 E. K. Tan . J. Jankovic
92 Pallidotomy for Parkinson’s Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1539 M. I. Hariz
93 Selective Thalamotomy and Gamma Thalamotomy for Parkinson Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1549 C. Ohye
94 Subthalamotomy for Parkinson’s Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1569 J. A. Obeso . L. Alvarez . R. Macias . N. Pavon . G. Lopez . R. Rodriguez-Rojas . M. C. Rodriguez-Oroz . J. Guridi
95 Globus Pallidus Stimulation for Parkinson’s Disease . . . . . . . . . . . . . . . . . . . . . . . 1577 M. Deogaonkar . J. L. Vitek
96 Subthalamic Nucleus Stimulation for Parkinson’s Disease . . . . . . . . . . . . . . . . . . . 1603 A. L. Benabid . J. Mitrofanis . S. Chabardes . E. Seigneuret . N. Torres . B. Piallat . A. Benazzouz . V. Fraix . P. Krack . P. Pollak . S. Grand . J. F. LeBas
97 Thalamic Stimulation for Parkinson’s Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1631 R. E. Wharen . R. J. Uitti . J. A. Lucas
98 PPN Stimulation for Parkinson’s Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1649 S. Stone . C. Hamani . A. M. Lozano
99 Other Targets to Treat Parkinson’s Disease (Posterior Subthalamic Targets and Motor Cortex) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1665 F. Velasco . S. Palfi . F. Jime´nez . J. D. Carrillo-Ruiz . G. Castro . Y. Keravel
100 Motor Cortex Stimulation for Parkinson’s Disease . . . . . . . . . . . . . . . . . . . . . . . . . 1679 M. Meglio . B. Cioni
101 Tissue Transplantation for Parkinson’s Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . 1691 K. Mukhida . M. Hong . I. Mendez
102 Gene Transfer for Parkinson’s Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1719 P. A. Starr . K. S. Bankiewicz
103 Intraparenchymal Drug Delivery for Parkinson’s Disease . . . . . . . . . . . . . . . . . . . 1731 R. D. Penn . A. A. Linninger
Table of contents
104 Management of Essential Tremor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1743 J. M. Nazzaro . K. E. Lyons . R. Pahwa
105 Management of Tremors other than Essential Tremor and Parkinson’s Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1757 J. P. Nguyen . S. Raoul . C. Deligny . V. Roualdes . Y. Keravel
106 Diagnosis and Medical Management of Dystonia . . . . . . . . . . . . . . . . . . . . . . . . . . 1767 S. Fahn
107 Pathophysiology of Dystonia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1779 J. A. Bajwa . M. D. Johnson . J. L. Vitek
108 Central Procedures for Primary Dystonia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1801 X. A. Vasques . L. Cif . B. Biolsi . P. Coubes
109 Functional Stereotactic Procedures for Treatment of Secondary Dystonia . . . . . 1835 H-H. Capelle . J. K. Krauss
110 Diagnosis and Medical Management of Cervical Dystonia . . . . . . . . . . . . . . . . . . . 1857 R. Bhidayasiri . D. Tarsy
111 Central Procedures for Cervical Dystonia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1871 J. Q. Oropilla . Z. H. T. Kiss
112 Peripheral Procedures for Cervical Dystonia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1885 T. Taira
113 Microvascular Decompression for Trigeminal Neuralgia . . . . . . . . . . . . . . . . . . . . 1911 J. R. Pagura
114 History and Current Neurosurgical Management of Spasticity . . . . . . . . . . . . . . . 1925 R. D. Penn
115 Destructive Neurosurgical Procedures for Spasticity . . . . . . . . . . . . . . . . . . . . . . . 1935 M. Sindou . P. Mertens
116 Surgery in the Dorsal Root Entry Zone for Spasticity . . . . . . . . . . . . . . . . . . . . . . . 1959 M. P Sindou . P. Mertens
117 Intrathecal Drugs for Spasticity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1973 R. D. Penn
xxi
xxii
Table of contents
Section 8 Functional Neurosurgery for Pain . . . . . . . . . . . . . . . . . . . . 1983 118 Anatomy and Physiology of Cancer Pain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1985 W. D. Willis Jr. . K. N. Westlund
119 Neuroimaging and Pain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2019 R. Peyron
120 What have PET Studies Taught us about Cerebral Mechanisms Involved in Analgesic Effect of DBS? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2031 R. Kupers . J. Gybels
121 History of DBS for Pain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2049 D. Richardson
122 Comprehensive Management of Cancer Pain Including Surgery . . . . . . . . . . . . . . 2061 P. S. Kalanithi . J. M. Henderson
123 The Central Lateral Thalamotomy for Neuropathic Pain . . . . . . . . . . . . . . . . . . . . 2081 D. Jeanmonod . A. Morel
124 Technique of Trigeminal Nucleotractotomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2097 M. J. Teixeira . E. T. Fonoff
125 Bulbar DREZ Procedures for Facial Pain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2125 J. P. Gorecki
126 Percutaneous Cordotomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2137 R. R. Tasker
127 CT-Guided Percutaneous Cervical Cordotomy for Cancer Pain . . . . . . . . . . . . . . . 2149 Y. Kanpolat
128 Ablative Spinal Cord Procedures for Cancer Pain . . . . . . . . . . . . . . . . . . . . . . . . . . 2159 P. L. Gildenberg
129 Intrathecal Opiates for Cancer Pain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2171 J. C. Sol . J. C. Verdie . Y. Lazorthes
130 Management of Pain of Benign Versus Cancer Origin . . . . . . . . . . . . . . . . . . . . . . 2197 P. L. Gildenberg . R. A. DeVaul
Table of contents
131 DBS for Persistent Non-Cancer Pain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2227 C. Hamani . D. Fontaine . A. Lozano
132 Motor Cortex Stimulation for Persistent Non-Cancer Pain . . . . . . . . . . . . . . . . . . . 2239 A. G. Machado . A. Y. Mogilner . A. R. Rezai
133 Radiofrequency Dorsal Root Entry Zone Lesions for Pain . . . . . . . . . . . . . . . . . . . 2251 P. Konrad . F. Caputi . A. O. El-Naggar
134 Surgical Dorsal Root Entry Zone Lesions for Pain . . . . . . . . . . . . . . . . . . . . . . . . . . 2269 M. P. Sindou
135 Facet Denervation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2291 R. R. Tasker . Wen Ching Tzaan
136 Sympathectomy for Pain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2297 C. R. Telles-Ribeiro . L. F. de Oliveira
137 Spinal Cord Stimulation. Techniques, Indications and Outcome . . . . . . . . . . . . . . 2305 B. Linderoth . B. A. Meyerson
138 Mechanisms of Action of Spinal Cord Stimulation . . . . . . . . . . . . . . . . . . . . . . . . . 2331 B. Linderoth . R. D. Foreman . B. A. Meyerson
139 Peripheral Nerve Stimulation for Neuropathic Pain . . . . . . . . . . . . . . . . . . . . . . . . 2349 A. G. Shetter
140 The Pathophysiology of Trigeminal Neuralgia . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2359 R. W. Hurt
141 Radiofrequency Rhizotomy for Trigeminal Neuralgia . . . . . . . . . . . . . . . . . . . . . . . 2421 E. Taub
142 Retrogasserian Glycerol Injection for Trigeminal Neuralgia . . . . . . . . . . . . . . . . . 2429 B. Linderoth . G. Lind
143 Balloon Compression for Trigeminal Neuralgia . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2457 J. A. Brown . J. G. Pilitsis
144 Microvascular Decompression for Trigeminal Neuralgia . . . . . . . . . . . . . . . . . . . . 2465 S. Sup Chung
145 Gamma Knife Surgery for Trigeminal Neuralgia and Facial Pain . . . . . . . . . . . . . . 2475 A. C. J. de Lotbinie`re
xxiii
xxiv
Table of contents
146 Treatment of Headache . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2483 N. T. Mathew
147 Occipital Neuralgia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2507 D. B. Cohen . M. Y. Oh . D. M. Whiting
148 Hypothalamic Stimulation for Cluster Headache . . . . . . . . . . . . . . . . . . . . . . . . . . 2517 A. Franzini . M. Leone . G. Messina . R. Cordella . C. Marras . G. Bussone . G. Broggi
149 Surgical Treatment of Chronic Cluster Headache . . . . . . . . . . . . . . . . . . . . . . . . . . 2525 J. M. Castilla
150 Mesencephalotomy for Cancer Pain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2533 P. L. Gildenberg
Section 9 Management of Epilepsy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2541 151 Indications for Surgical Management of Epilepsy . . . . . . . . . . . . . . . . . . . . . . . . . . 2543 H. G. Wieser . D. Zumsteg
152 Classification of Epileptic Seizures and Epilepsies . . . . . . . . . . . . . . . . . . . . . . . . . 2561 H. O. Lu¨ders . S. Noachtar
153 EEG in Epilepsy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2575 M. Hoppe . R. Wennberg . P. Tai . B. Pohlmann-Eden
154 The Wada Test-60th Year Anniversary Update-In Epilepsy Surgery . . . . . . . . . . . 2587 J. A. Wada . B. Kosaka
155 Imaging Evaluation of Epilepsy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2617 D. Madhavan . R. Kuzniecky
156 Image Guided Epilepsy Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2633 Y. G. Comair . R. B. Chamoun
157 Intraoperative Monitoring in Epilepsy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2651 G. Ojemann
158 MEG in Epilepsy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2661 A. Fujimoto . T. Akiyama . H. Otsubo
159 Medical Management of Epilepsy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2669 M. E. Newmark
Table of contents
160 Selective Amygdalo-Hypocampectomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2677 T. A. Valiante
161 Subpial Transection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2715 Z. S. Tovar-Spinoza . J. T. Rutka
162 Corpus Callosotomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2723 R. E. Maxwell
163 Hemispherectomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2741 J.- G. Villemure . R. T. Daniel
164 Radiosurgery in Epilepsy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2761 I. Yang . N. M. Barbaro
165 Centromedian Thalamic Stimulation for Epilepsy . . . . . . . . . . . . . . . . . . . . . . . . . . 2777 F. Velasco . A. L. Velasco . M. Velasco . F. Jime´nez . J. D. Carrillo-Ruiz . G. Castro
166 Anterior Nucleus DBS in Epilepsy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2793 M. Hodaie . C. Hamani . R. Wennberg . W. Hutchison . J. Dostrovky . A. M. Lozano
167 Vagal Nerve Stimulation for Seizures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2801 A. P. Amar . J. B. Elder . M. L. J. Apuzzo
168 Cerebellar Stimulation for Seizure Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2823 R. Davis
169 Stimulation of the Hippocampus and the Seizure Focus . . . . . . . . . . . . . . . . . . . . 2839 A. L. Velasco . F. Velasco . M. Velasco . G. Castro . J. D. Carrillo-Ruiz . J. M. Nu´n˜ez D. Trejo
Section 10 Psychiatric Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2853 170 Ethical Considerations in Psychiatric Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2855 B. S. Appleby . P. V. Rabins
171 Psychosurgery – A Historical Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2867 C. R. Bjarkam . J. C. Sørensen
172 Cingulotomy for Depression and OCD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2887 G. R. Cosgrove
xxv
xxvi
Table of contents
173 DBS for OCD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2897 L. Gabrie¨ls . P. Cosyns . K. van Kuyck . B. Nuttin
174 Medical Management and Indications for Surgery in Depression . . . . . . . . . . . . . 2925 P. Giacobbe . S. Kennedy
175 Ablative Procedures for Depression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2943 V. A. Coenen . C. R. Honey
176 Deep Brain Stimulation for Depression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2953 C. Hamani . B. Snyder . A. Laxton . A. Lozano
177 Surgical Procedures for Tourette’s Syndrome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2963 V. Visser-Vandewalle
178 Treatment of Aggressive Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2971 G. Broggi . A. Franzini
Section 11 Special and Emerging Applications . . . . . . . . . . . . . . . . . . 2979 179 DBS Disorders of Consciousness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2981 N. D. Schiff
180 Apnea: Phrenic Nerve Stimulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2991 S. Rehncrona . G. Sedin . H. Fodstad
181 DBS for Bladder Dysfunction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2999 R. Almusa . M. M. Hassouna
182 Impaired Vision: Visual Prosthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3009 J. P. Girvin . A. G. Martins
183 Impaired Hearing: Auditory Prosthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3021 J. K. Niparko . A. Marlowe . H. W. Francis
184 Impaired Motor Function: Functional Electrical Stimulation . . . . . . . . . . . . . . . . . 3047 R. B. Stein . A. Prochazka
185 Gene Therapy for Neurological Disorders (Except Oncology) . . . . . . . . . . . . . . . . 3061 M. G. Kaplitt
Table of contents
186 Gene Therapy for Brain Tumors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3083 M. L. M. Lamfers . E. A. Chiocca
187 Microdialysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3117 M. J. Smith . M. G. Kaplitt
Section 12 The Future of Stereotactic and Functional Neurosurgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3129 188 The Future of Computers and Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3131 B. A. Kall
189 The Future of Neuronavigation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3137 D. W. Roberts
190 The Future of Radiosurgery and Radiotherapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3143 L. Ma . P. K. Sneed
191 The Future of Infusion Systems in Neurosurgery . . . . . . . . . . . . . . . . . . . . . . . . . . 3155 R. D. Penn
192 The Future of Cell Transplantation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3161 M. B. Newman . R. A. E. Bakay
193 The Future of Neural Interface Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3185 M. C. Park . M. A. Goldman . T. W. Belknap . G. M. Friehs
194 The Future of Molecular Neuro-Oncology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3201 J. A. J. King . M. D. Taylor
195 Future Ethical Challenges in Neurosurgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3229 N. Lipsman . M. Bernstein
Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3239
xxvii
List of Contributors
John R. Adler, Jr. Department of Neurosurgery, Stanford University, Stanford, CA, USA
Women’s Hospital, 75 Francis Street, CA 138F, Boston, MA 02115, USA Email:
[email protected]
Tomoyuki Akiyama Division of Neurology, Department of Pediatrics, The Hospital for Sick Children, University of Toronto, 555 University Avenue, Toronto, Ontario, M5G 1X8, Canada
David W. Andrews Department of Neurosurgery, Jefferson Medical College, 834 Walnut St. Suite 650, PA 19107, Philadelphia, USA Email:
[email protected]
Riyad Almusa Toronto Western Hospital, 399 Bathurst St, M5T 2S8, Toronto, Ontario, Canada George Al-Shamy Department of Neurosurgery - 442, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd., Houston, TX 77030, USA L. Alvarez Movement Disorders, Functional Neurosurgery and Neurophysiology Units, Centro Internacional de Restauracion Neurologica (CIREN), Havana, Cuba Arun Paul Amar Departments of Neurosurgery, University of California San Francisco and the Permanente Medical Group 2025 Morse Avenue, Sacramento, CA 95825, USA Email:
[email protected] William S. Anderson Instructor in Neurosurgery, Harvard Medical School, Department of Neurological Surgery, Brigham and
Russell J. Andrews Ames Associate (Smart Systems and Nanotechnology), NASA Ames Research Center, CA 94035, Moffett Field, USA Email:
[email protected] Narisa Apichatibutra Department of Anesthesia, Toronto Western Hospital, University Health Network, University of Toronto, Toronto, Ontario, Canada Brian S. Appleby John Hopkins Hospital, Meyer 279, 600 N. Wolfe St., 21287–7279, MD, Baltimore, USA Michael L. J. Apuzzo Edwin M. Todd/Trent H. Wells, Jr., Professor of Neurological Surgery and Professor of Radiation Oncology, Biology, and Physics Keck School of Medicine, University of Southern California, 1200 North State Street, Suite 5046, Los Angeles, CA 90033, USA Email:
[email protected]
xxx
List of contributors
Jeffrey E. Arle Department of Neurosurgery, Lahey Clinic, 41 Mall Road, 01805, Burlington, MA, USA
Eric Bardinet INSERM U679, Hoˆpital de la Salpeˆtrie`re, 47, Bd de l’Hoˆpital, 75013, Paris, France
Sharon Ashton Renishaw plc, New Mills, Wotton-under-Edge, Gloucestershire GL12 8JR, UK
Thomas Belknap Department of Clinical Neurosciences, Program in Neurosurgery, The Warren Alpert Medical School of Brown University, Rhode Island Hospital, Providence, RI 02903, USA
Tipu Z. Aziz Oxford Functional Neurosurgery, Nuffield Department of Surgery, University of Oxford, and Department of Neurological Surgery, The West Wing, The John Radcliffe Hospital, Oxford. Imperial College London and Charing Cross Hospital, London Jawad A. Bajwa Capistrant Parkinson and Movement Disorder Center, Bethesda Hospital and Nasseff Neuroscience Center, United Hospital, Saint Paul, MN, USA Neurological Associates of Saint Paul, Maplewood, MN, USA Roy A. E. Bakay Department of Neurosurgery, Rush University Medical Center, Chicago IL, USA Email:
[email protected] B. Ballanger PET Imaging Centre, Center of Addiction Mental Health, University of Toronto, Toronto, Ontario, Canada Krystof S. Bankiewicz Department of Neurological Surgery, University of California, San Francisco, USA Nicholas M. Barbaro Department of Neurological Surgery, University of California, San Francisco, USA
Alim-Louis Benabid Joseph Fourier University, Pavillon B - Grenoble University Hospital, Grenoble, France Email:
[email protected] Abdelhamid Benazzouz University of Bordeaux, France Mitchel S. Berger Department of Neurological Surgery, University of California, 505 Panassus Ave, Box 0112, San Francisco, CA, USA Email:
[email protected] Mark Bernstein Division of Neurosurgery, Toronto Western Hospital and University of Toronto, Toronto, Canada Email:
[email protected] Jagdish Bhatnagar The Center for Image-Guided Neurosurgery, University of Pittsburgh Medical Center, Pittsburgh, PA 15213, USA Roongroj Bhidayasiri Chulalongkorn Comprehensive Movement Disorders Center, Chulalongkorn University Hospital, 1873 Rama 4 Road, Bangkok 10330, Thailand Email:
[email protected]
List of contributors
B. Biolsi CHRU Montpellier, Service de Neurochirurgie, Universite´ Montpellier I, Montpellier, France
Chris Butson Medical School of Wisconsin, 8701 Watertown Plank Road, WI 53226, Milwaukee, USA
Carsten Reidies Bjarkam Department of Neurobiology, Institute of Anatomy, University of Aarhus, DK–8000 Aarhus C, Denmark Email:
[email protected]
John D. Caird Division of Neurosurgery, Department of Surgery, The Hospital for Sick Children, University of Toronto, 555 University Avenue, M5G 1X8, Toronto, Ontario, Canada
Peter M. Black Department of Neurosurgery, Brigham and Women’s Hospital, 75 Francis St, Boston, MA 02115, USA Email:
[email protected] Frank J. Bova Department of Neurological Surgery, University of Florida, PO Box 100265, Gainesville, FL 32610, USA Giovanni Broggi Department of Neurosurgery, Fondazione Istituto Neurologico “C. Besta”, Milan, Italy Email:
[email protected] Jeffrey A. Brown 600 Northern Boulevard #118, Great Neck, NY 11021, USA Email:
[email protected] Richard Bucholz Division of Neurosurgery, Department of Surgery, Saint Louis University School of Medicine, Saint Louis, Missouri, USA Email:
[email protected] Christian Bu¨hrle Associate Professor, Department of Stereotactic and Functional Neurosurgery, University Hospital, 50924 Cologne, Germany G. Bussone Department of Neurosurgery, Neurological Institute Foundation “C. Besta”, Milan, Italy
Hans-Holger Capelle Department of Neurosurgery, Medical School Hannover, MHH, Hannover, Germany Franco Caputi Department of Neurosurgery, Rome, Italy Email:
[email protected] Mark P. Carol 744 Lexington Way, Burlingame, CA 94010, USA Email:
[email protected] Jose´ Damia´n Carrillo Epilepsy Surgery Clinic; Unit for Stereotactic, Functional Neurosurgery and Radiosurgery, General Hospital of Mexico City, Mexico Jose´ Manuel Castilla Servicio de Neurocirugı´a, Hospital “General Yagu¨e”, Avenida de Cid 96, 09005, Burgos, Spain Email:
[email protected] Guillermo Castro Epilepsy Surgery Clinic; Unit for Stereotactic, Functional Neurosurgery and Radiosurgery, General Hospital of Mexico City, Mexico Stephan Chabardes Joseph Fourier University, Pavillon B - Grenoble University Hospital, 38700, Grenoble, France Arnab Chakravarti Massachusetts General Hospital, Department of
xxxi
xxxii
List of contributors
Radiation Oncology, Cox 3, 100 Blossom Street, Boston, MA, USA Email:
[email protected] Roukoz B. Chamoun Department of Neurosurgery, Baylor College of Medicine, Houston, Texas, USA Parakrama T. Chandrasoma Department of Pathology, Keck School of Medicine University of Southern California, GNH 2900, 1200 North State St., CA, USA Paul H. Chapman Departments of Radiation Oncology and Neurosurgery, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA Email:
[email protected] E. Antonio Chiocca James Cancer Hospital/Solove Research Institute, The Ohio State University Medical Center, Columbus, OH 43210–1240, USA Email:
[email protected] Sang Sup Chung Bundang CHA General Hospital, 351 Yatap-dong, 463–712, Bundang-gu Sungnam-se, Kyong-doSungnam, South Korea Email:
[email protected] L. Cif CHRU Montpellier, Service de Neurochirurgie, Universite´ Montpellier I, Montpellier, France Beatrice Cioni Neurochirurgia Funzionale e Spinale, Universita` Cattolica, Roma, Italy Richard E. Clatterbuck Hattiesburg Clinic, 4155 28th Avenue, Hattiesburg,
MS 39401, USA Email:
[email protected] Volker A. Coenen Surgical Center for Movement Disorders, Division of Neurosurgery, University of British Columbia, Vancouver, BC, Canada Robert J. Coffey Medtronic Neurological, 4000 Lexington Avenue North, MN, USA Email:
[email protected] David B. Cohen Drexel University College of Medicine Department of Neurosurgery, Allegheny General Hospital, 420 East North Ave, PA 15212, Pittsburgh, USA Youssef G. Comair Department of Neurosurgery, Baylor College of Medicine, 1709 Dryden, TX 77030, Houston, Texas, USA Email:
[email protected] R. Cordella Department of Neurosurgery, Neurological Institute Foundation “C. Besta”, Milan, Italy G. Rees Cosgrove Department of Neurosurgery, Lahey Clinic Medical Center and Tufts University School of Medicine, Boston, Massachusetts, USA Email:
[email protected] Eric R. Cosman, Jr. Cosman Medical, 76 Cambridge Street, MA, Burlington, USA Email:
[email protected] Eric R. Cosman, Sr. Cosman Medical, 76 Cambridge Street, MA, Burlington, USA Email:
[email protected]
List of contributors
Paul Cosyns Department of Psychiatry, University Hospital Antwerp, Antwerp, Belgium
Woodruff Memorial Research Building, 101 Woodruff Circle, Atlanta, GA 30322, USA Email:
[email protected]
Philippe Coubes Unite´ de Recherche sur les Mouvements Anormaux, Hoˆpital Gui de Chauliac, Service de Neurochirurgie, 80 Avenue Augustin Fliche, 34295 Montpellier cedex 05, France Email:
[email protected]
Robert Den Department of Neurosurgery, Jefferson Medical College, 834 Walnut St. Suite 650, PA 19107, Philadelphia, USA
Karen D. Davis Department of Physiology, University of Toronto, 1 King’s College Circle, M5S 1S8, Toronto, Canada
Milind Deogaonkar Functional Neuroscience Research Center, Cleveland Clinic Lerner Research Center, 9500 Euclid Avenue, OH 44195, Cleveland, USA
Ross Davis Neural Engineering Clinic, 330 Hammock Shore Dr., Melbourne Beach, FL, 32951, USA Email:
[email protected]
Richard A. DeVaul Prof. Psychiatry, College of Medicine, Texas A & M System Health Science Center, College Station, Texas, USA
Alain C. J. de Lotbinie`re FACS Brain & Spine Surgeons of New York, 244 Westchester Avenue, NY 10604, White Plains, USA Email:
[email protected]
John P. Donoghue Department of Neuroscience and Brain Science Program, Brown University, Providence, RI 02912, USA
Luiz F. de Oliveira Neurodor - Neurosurgery and Pain Clinic, Rua Visconde de Piraja 351/509, Cep: 22471–002, Ipanema, Rio de Janeiro, RJ, Brazil Antonio A. F. De Salles Department of Neurosurgery, David Geffen School of Medicine, University of California Los Angeles, 10495 Le Conte Avenue, Suite 2120, Los Angeles, California 90095, USA Email:
[email protected] Ce´line Deligny Department of Neurology, CHU de Nantes, Nantes, France Mahlon R. DeLong Department of Neurology, Emory University
Paresh K. Doshi Department of Neuroscience, Jaslok Hospital and Research Centre, 15 - Dr. Deshmukh Marg, Pedder Road, Mumbai 400 026, Maharashtra, India Email:
[email protected] Jonathan O. Dostrovsky Department of Physiology, University of Toronto, 1 King’s College Circle, M5S 1S8, Toronto, Canada Email:
[email protected] James M. Drake Division of Neurosurgery, Department of Surgery, The Hospital for Sick Children, University of Toronto, 555 University Avenue, M5G 1X8, Toronto, Ontario, Canada Email:
[email protected]
xxxiii
xxxiv
List of contributors
James B. Elder Department of Neurological Surgery, Keck School of Medicine, University of Southern California, 1200 North State Street, Suite 5046, Los Angeles, CA 90033, USA Email:
[email protected]
Sa˜o Paulo Medical School, Rua Mena Barreto 765 - Itaim Bibi, SP- 014033–010, Sa˜o Paulo, Brazil
Jeff Elias Department of Neurosurgery, University of Virginia, Box 800212, HSC, VA, 22908–0212, Charlottesville, USA Email:
[email protected]
Robert D. Foreman Department of Physiology, University of Oklahoma Health Sciences Center, Oklahoma City, OK, USA
Amr O. El-Naggar Lake Cumberland Neurosurgery, Somerset, KY, USA Email:
[email protected] Elmyra V. Encarnacion Plummmer Movement Disorders Center, Texas A & M Health Sciences Center/Scott & White, Temple, Texas, USA Stanley Fahn Department of Neurology, Columbia University College of Physicians & Surgeons, New York, NY 10032, USA Email:
[email protected] Osvaldo Vilela Filho Goiaˆnia Neurological Institute, Goiaˆnia, GO, Brazil Email:
[email protected] John C. Flickinger The Departments of Neurological Surgery, and Radiation Oncology The Center for Image-Guided Neurosurgery, University of Pittsburgh Medical Center, Pittsburgh, PA 15213, USA Harald Fodstad University of Uppsala, Uppsala, Sweden Erich Talamoni Fonoff Division of Functional Neurosurgery, University of
Denys Fontaine Service de Neurochirurgie, Hoˆpital Pasteur, Centre Hospitalier Universitaire de Nice, Nice, France
Valerie Fraix Joseph Fourier University, Pavillon B - Grenoble University Hospital, 38700, Grenoble, France Doglietto Francesco Department of Neurosurgery, Toronto Western Hospital, 399 Bathurst Street, M5T 2S8, Toronto, Canada Howard W. Francis Department of Otolaryngology, Head & Neck Surgery, The Johns Hopkins School of Medicine, Baltimore, USA Angelo Franzini Department of Neurosurgery, Fondazione Istituto Neurologico “C. Besta”, Milan, Italy Justin F. Fraser Department of Neurological Surgery, Weill Medical College of Cornell University, New YorkPresbyterian Hospital, New York, NY, USA William A. Friedman Department of Neurological Surgery, University of Florida, PO Box 100265, Gainesville, FL 32610, USA Email:
[email protected] Gerhard M. Friehs Department of Clinical Neurosciences, Program in Neurosurgery, The Warren Alpert Medical School of Brown University, Rhode Island Hospital, Providence, RI 02903, USA
List of contributors
Ayataka Fujimoto Division of Neurology, Department of Pediatrics, The Hospital for Sick Children, University of Toronto, 555 University Avenue, Toronto, Ontario, M5G 1X8, Canada
John P. Girvin 1078 The Parkway, ON, N6A 2W0, London, Canada Email:
[email protected]
Loes Gabrie¨ls Department of Psychiatry, University Hospital Gasthuisberg, Leuven, Belgium
Marc A. Goldman Department of Clinical Neurosciences, Program in Neurosurgery, The Warren Alpert Medical School of Brown University, Rhode Island Hospital, Providence, RI 02903, USA
Gregory Gagnon Department of Neurosurgery and Radiation Medicine, Georgetown University Hospital, 3800 Reservoir Rd, 20007, Washington DC, USA
John P. Gorecki Department of Neurosurgery, Wichita Surgical Specialists, P.A. The Heritage Plaza, 818 N. Emporia, Suite 200, 67214–3788, Wichita, Kansas, USA Email:
[email protected]
Fred Gentili Department of Neurosurgery, Toronto Western Hospital, 399 Bathurst Street, M5T 2S8, Toronto, Canada Email:
[email protected] Peter C. Gerszten Department of Neurological Surgery, University of Pittsburgh, UPMC Presbyterian B–400, 200 Lothrop Street, PA 15213, Pittsburgh, USA Email:
[email protected]
Alessandra A. Gorgulho Department of Neurosurgery, David Geffen School of Medicine, University of California Los Angeles, 10495 Le Conte Avenue, Suite 2120, Los Angeles, California 90095, USA Sylvie Grand Joseph Fourier University, Pavillon B - Grenoble University Hospital, 38700, Grenoble, France
Peter Giacobbe University Health Network, University of Toronto, 200 Elizabeth Street, M5G 2C4, Toronto, Ontario, Canada
Alexander L. Green Oxford Functional Neurosurgery, Nuffield Department of Surgery, University of Oxford and Department of Neurological Surgery, The West Wing, The John Radcliffe Hospital, Oxford, UK
Philip L. Gildenberg Houston Stereotactic Concepts, 3776 Darcus St., 77005, Houston, Texas, USA Email:
[email protected]
Jorge Guridi Division of Neurosurgery, Clinica Universitaria, Universidad de Navarra, Centro de Investigacio´n Medica Aplicada. CIMA, Pamplona, Spain
Steven S. Gill Department of Neurosurgery, Frenchay Hospital, Bristol BS16 1LE, UK Email:
[email protected]
Jan Gybels Department of Neurosurgery, Gasthuisberg University Hospital, University of Leuven, Leuven, Belgium
xxxv
xxxvi
List of contributors
Sun Ha Paek Department of Neurosurgery, Jefferson Medical College, 834 Walnut St. Suite 650, PA 19107, Philadelphia, USA Clement Hamani Division of Neurosurgery, Department of Surgery, Toronto Western Hospital, University of Toronto, Toronto, Ontario, Canada Marwan I. Hariz Edmond J. Safra Chair of Functional Neurosurgery, Unit of Functional Neurosurgery, Institute of Neurology, London, UK Email:
[email protected] Alex Hartov Thayer School of Engineering, Dartmouth College, Hannover, NH 03755, USA Magdy M. Hassouna Toronto Western Hospital, 399 Bathurst St, M5T 2S8, Toronto, Ontario, Canada Email:
[email protected] Robert A. Hauser Parkinson’s Disease and Movement Disorders Center, University of South Florida, 4 Columbia Drive, Suite 410, Tampa, Florida 33606, USA Email:
[email protected] Jaimie M. Henderson Stanford University Medical Center, 300 Pasteur Dr, CA 94305, Stanford, USA Email:
[email protected] Fraser Cummins Henderson, Sr. Georgetown University Medical Center, 3800 Reservoir Rd, 20007, Washington DC, USA Email:
[email protected] Leigh R. Hochberg Department of Neuroscience and Brain Science
Program, Brown University, Providence, RI 02912, USA Department of Neurology, Massachusetts General Hospital, Brigham and Women’s Hospital, Spaulding Rehabilitation Hospital, Harvard Medical School, Boston, MA, USA Center for Restorative and Regenerative Medicine, Rehabilitation Research and Development Service, Department of Veterans Affairs, Veterans Health Administration, Providence, RI 02908, USA Mojgan Hodaie Department of Physiology, University of Toronto, 1 King’s College Circle, M5S 1S8, Toronto, Canada Dominique Hoffmann Joseph Fourier University, Pavillon B - Grenoble University Hospital, Grenoble, France Christopher R. Honey Surgical Center for Movement Disorders - Division of Neurosurgery – University of British Columbia Vancouver, BC, Canada Email:
[email protected] Murray Hong Departments of Anatomy & Neurobiology and Surgery (Neurosurgery), Dalhousie University, Halifax, Nova Scotia, Canada Matthias Hoppe Bethel Epilepsy Centre, Bielefeld, Germany R. Wayne Hurt Chief of Neurological Surgery, St. Joseph Medical Center and Assistant Clinical Professor of Neurological Surgery, Baylor College of Medicine and University of Texas Medical Branch, Houston, Texas, USA William Hutchison Department of Neurosurgery, Toronto Western Hospital, 399 Bathurst Street, M5S 2T8, Toronto, Ontario, Canada
List of contributors
Warren Ishida Department of Neurosurgery, David Geffen School of Medicine, University of California Los Angeles, 10495 Le Conte Avenue, Suite 2120, Los Angeles, California 90095, USA Pascal Jabbour Department of Neurosurgery, Philadelphia, PA, USA E-mail:
[email protected] Jay Jagannathan Lars Leksell Center for Gamma Surgery, Department of Neurosurgery, University of Virginia, Charlottesville, VA, USA Joseph Jankovic Parkinson’s Disease Center and Movement Disorders Clinic, Department of Neurology, Baylor College of Medicine, 6550 Fannin, Suite 1801, Houston, Texas 77030, USA Email:
[email protected]
Matthew D. Johnson Department of Biomedical Engineering, Cleveland Clinic Foundation, Cleveland, Ohio, USA Paul Kalanithi Stanford University Medical Center, 300 Pasteur Dr, CA 94305, Stanford, USA Bruce A. Kall Departments of Neurologic Surgery and Information Technology, Mayo Clinic, Rochester, MN, USA Email:
[email protected] Yu¨cel Kanpolat Inkilap Sokak No: 24/2, 06640, Kizilay, Turkey Email:
[email protected] Michael G. Kaplitt Department of Neurological Surgery, Weill Cornell Medical College, New York, NY, USA Email:
[email protected]
Lauren Jarchin Department of Neurosurgery, North Shore LIJ, Manhasset, NY 11030, USA
Patrick J. Kelly Department of Neurological Surgery, NYU School of Medicine, 530 First Ave, Suite 8R, NY, USA Email:
[email protected]
Walter Jean Department of Neurosurgery and Radiation Medicine, Georgetown University Hospital, 3800 Reservoir Rd, 20007, Washington DC, USA Email:
[email protected]
Sidney Kennedy University Health Network, University of Toronto, 200 Elizabeth Street, M5G 2C4, Toronto, Ontario, Canada Email:
[email protected]
Daniel Jeanmonod Department of Functional Neurosurgery, University Hospital Zu¨rich, Zu¨rich, Switzerland Email:
[email protected]
Yves Keravel Service de Neurochirurgie, Hoˆpital Henri Mondor, Cre´teil, France
Fiacro Jime´nez Epilepsy Surgery Clinic; Unit for Stereotactic, Functional Neurosurgery and Radiosurgery, General Hospital of Mexico City, Mexico
Sadaquate Khan Institute of Neurosciences, Frenchay Hospital, Bristol BS16 1LE, UK Deepak Khuntia Department of Human Oncology, University of
xxxvii
xxxviii
List of contributors
Wisconsin, 600 Highland Ave, K4-B100, Madison, WI, USA Email:
[email protected]
Paul Kongkham Division of Neurosurgery, Toronto Western Hospital and University of Toronto, Toronto, Canada
Jong-Hyun Kim Fellow in Neurosurgery, Department of Neurosurgery, 600 North Wolfe Street, Meyer 8–181, Baltimore, MD 21287, USA Email:
[email protected]
Peter Konrad Director of Functional Neurosurgery, Vanderbilt University Medical Center, Department of Neurological Surgery, Rm T–4224; MCN, 37232–2380, Nashville, Tennessee, USA Email:
[email protected]
James A. J. King Sick kids Hospital, 555 University Avenue, M5G 1X8, Toronto, Ontario, Canada Zelma H. T. Kiss Associate Professor, Neurosurgery University of Calgary, Calgary, Alberta T2N 4N1, Canada Email:
[email protected] Nancy E. Klipfel Department of Pathology, Keck School of Medicine University of Southern California, GNH 2900, 1200 North State St., CA, USA Email:
[email protected] Kazu Kobayashi Fellow in Neurosurgery, Department of Neurosurgery, 600 North Wolfe Street, Meyer 8–181, Baltimore, MD 21287, USA Martin Kocher Professor, Department of Radiation Oncology, University Hospital, 50924 Cologne, Germany Douglas Kondziolka The Departments of Neurological Surgery, and Radiation Oncology, The University of Pittsburgh, PA, USA The Center for Image-Guided Neurosurgery, University of Pittsburgh Medical Center, Pittsburgh, PA 15213, USA E-mail:
[email protected]
Brian Harris Kopell Medical School of Wisconsin, 8701 Watertown Plank Road, WI 53226, Milwaukee, USA Email:
[email protected] Brenda Kosaka Division of Neurosciences, University of British Columbia, 2329 West Mall, V6T 1Z4, Vancouver, Canada Paul Krack Joseph Fourier University, Pavillon B - Grenoble University Hospital, 38700, Grenoble, France Joachim K. Krauss Chairman and Director, Department of Neurosurgery, Medical University, MHH CarlNeuberg-Str. 1, 30625 Hannover, Germany Email
[email protected] S. A. Kuhn Ames Associate (Smart Systems and Nanotechnology), NASA Ames Research Center, CA 94035, Moffett Field, USA Ron Kupers PET Unit & Department of Surgical Pathophysiology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark Email:
[email protected]
List of contributors
Ruben Kuzniecky Professor, NYU Comprehensive Epilepsy Center, Department of Neurology, New York University Medical Center, NY, USA Email:
[email protected] Lauri V. Laitinen Unit of Functional Neurosurgery, Institute of Neurology, Box 146, Queen Square, WC1N 3BG, London, UK Martine L. M. Lamfers Department of Neurosurgery, Erasmus Medical Center, Rotterdam, The Netherlands Anthony E. Lang Division of Patient Based Clinical Research, Toronto Western Hospital McLaughlin Pavilion, 7th Floor Rm 7–403, 399 Bathurst Street, M5T 2S8, Toronto, Ontario, Canada Email:
[email protected] Edward R. Laws Department of Neurosurgery, University of Virginia, Box 800212, HSC, VA, 22908–0212, Charlottesville, USA Email:
[email protected] Adrian Laxton Division of Neurosurgery, Toronto Western Hospital, University of Toronto, UHN, Toronto, Ontario, Canada Yves Lazorthes Service de Neurochirurgie, Hospital LarreyRangueil, Avenue Jean Poulhes, 31403, Toulouse, France Email:
[email protected] Jean Franc¸ois LeBas Joseph Fourier University, Pavillon B - Grenoble University Hospital, 38700, Grenoble, France
Dominique Leduc IREENA, Faculte´ des Sciences de Nantes, Nantes, France Foo-Chiang Lee Consultant Neurosurgeon, Division of Neuroscience Sunway Medical Center, No. 5 Jln Lagoon Selatan, Bandar Sunway, 46150, Selangor, Subang Jaya, Malaysia Email:
[email protected] Dan Leksell Lars Leksell Professor of Neurological Surgery, University of Pittsburgh, B400 UPMC, Pittsburgh, PA 15213, USA Frederick A. Lenz Professor of Neurosurgery, The Johns Hopkins University School of Medicine, Department of Neurosurgery, 600 North Wolfe Street, Meyer 8–181, Baltimore, MD 21287, USA Email:
[email protected] M. Leone Department of Neurosurgery, Neurological Institute Foundation “C. Besta”, Milan, Italy Marc Levivier Department of Neurosurgery, CHUV, and Centre Universitaire Romand de Neurochirurgie, Lausanne, Switzerland Email:
[email protected] J. Li Ames Associate (Smart Systems and Nanotechnology), NASA Ames Research Center, CA 94035, Moffett Field, USA Go¨ran Lind Department of Neurosurgery, Karolinska Institutet and Karolinska University Hospital Stockholm, Sweden
xxxix
xl
List of contributors
Bengt Linderoth Department of Neurosurgery, Karolinska University Hospital, SE 171–76 Stockholm, Sweden Email:
[email protected]
John A. Lucas Associate Professor of Psychology, Mayo Clinic College of Medicine, 4500 San Pablo Rd, FL 32224, Jacksonville, USA
Christer Lindquist Director, Gamma Knife Centre, Cromwell Hospital, London, England
L. Dade Lunsford Department of Neurological Surgery, Suit B–400, University of Pittsburgh Medical Center, 200 Lothrop Street, Pittsburgh, PA 15213, USA Email:
[email protected]
Andreas A. Linninger Associate Professor of Chemical Engineering and Bioengineering, University of Illinois at Chicago, Laboratory for Product and Process Design, M/C 063, 851 S. Morgan St. - 218 SEO, Chicago, Illinois 60607–7000, USA Email:
[email protected]
Hans O. Lu¨ders Department of Neurology, Cleveland Clinic, 9500 Euclid Ave, OH 44195, Cleveland, USA Email:
[email protected]
Nir Lipsman Division of Neurosurgery, Toronto Western Hospital, University Health Network, University of Toronto, Toronto, Ontario, Canada
Kelly E. Lyons University of Kansas Medical Center, Department of Neurology, 3599 Rainbow Blvd, Mailstop 2012, Kansas City, KS 66160, USA Email:
[email protected]
Jay S. Loeffler Department of Radiation Oncology, Massachusetts General Hospital, 55 Fruit Street, MA 02114, Boston, USA Email:
[email protected]
Lijun Ma Department of Radiation Oncology, University of California-San Francisco, 505 Parnassus Avenue, CA 94143–0226, San Francisco, USA
G. Lopez Movement Disorders, Functional Neurosurgery and Neurophysiology Units, Centro Internacional de Restauracion Neurologica (CIREN), Havana, Cuba
Mohammad Maarouf Department of Stereotactic and Functional Neurosurgery, University Hospital, 50924 Cologne, Germany
Andrew Losiniecki Department of Neurosurgery, University of Cincinnati College of Medicine, Cincinnati, OH, USA
Andre G. Machado Center for Neurological Restoration, Department of Neurosurgery, Department of Biomedical Engineering, Cleveland Clinic, Cleveland, OH, USA
Andres M. Lozano Division of Neurosurgery, Department of Surgery, Toronto Western Hospital, University of Toronto, 399 Bathurst Street, Toronto, Ontario, Canada Email:
[email protected]
R. Macias Movement Disorders, Functional Neurosurgery and Neurophysiology Units, Centro Internacional de Restauracion Neurologica (CIREN), Havana, Cuba
List of contributors
Marcella A. Madera University of Cincinnati, Department of Neurosurgery, Cincinnati, Ohio, USA
MMC 96, 420 E. Delaware St., MN 554455–0374, Minneapolis, USA Email:
[email protected]
Deepak Madhavan Department of Neurological Sciences, University of Nebraska Medical Center, Omaha, NE, USA Email:
[email protected]
Ian E. McCutcheon Department of Neurosurgery, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd., Box 44, Houston, Texas 77030–4009, USA Email:
[email protected]
George Mandybur The Neuroscience Institute: Department of Neurosurgery, University of Cincinnati College of Medicine and Mayfield Clinic, Cincinnati, OH, USA Email:
[email protected] Pirjo Manninen Department of Anesthesia, Toronto Western Hospital, 399 Bathurst Street, Toronto, Ontario, Canada Email:
[email protected] Miguel Manrique Division of Neurosurgery, Clinica Universitaria, Universidad de Navarra, Centro de Investigacio´n Medica Aplicada. CIMA, Pamplona, Spain Andrea Marlowe Department of Otolaryngology, Head & Neck Surgery, The Johns Hopkins School of Medicine, Baltimore, USA C. Marras Department of Neurosurgery, Neurological Institute Foundation “C. Besta”, Milan, Italy Antonio G. Martins Ninan T. Mathew Houston Headache Clinic, Houston, Texas, USA Email:
[email protected] Robert Maxwell Department of Neurosurgery, Univ of Minnesota/S.E.
Lee McDurmont Division of Neurosurgery, Department of Surgery, Saint Louis University School of Medicine, Saint Louis, Missouri, USA Mario Meglio Istituto di Neurochirurgia, Policlinico Agostino Gemelli, Lg Gemelli, 00168, Roma, Italy Email:
[email protected] Minesh P. Mehta Department of Human Oncology, University of Wisconsin, 600 Highland Ave, K4-B100, Madison, WI 53792, USA Email:
[email protected] Ivar Mendez Departments of Anatomy & Neurobiology and Surgery (Neurosurgery), Dalhousie University, Halifax, Nova Scotia, Canada Fang-Gang Meng Center for Functional Neurosurgery (BMS), Shanghai Jiao Tong University Ruijin Hospital and Department of Functional Neurosurgery (JGZ,KZ, FGM), Beijing Tiantan Hospital, People’s Republic of China Patrick Mertens Department of Neurosurgery, Hopital Neurologique P. Wertheimer, University of Lyon, 59 Bd Pinel, F–69003 Lyon, France
xli
xlii
List of contributors
G. Messina Department of Neurosurgery, Neurological Institute Foundation “C. Besta”, Milan, Italy
Nadim Nasr Department of Neurosurgery and Radiation Medicine, Georgetown University Hospital, 3800 Reservoir Rd, 20007, Washington DC, USA
Bjo¨rn A. Meyerson Department of Neurosurgery, Karolinska Institute and Karolinska University Hospital Stockholm, Stockholm, Sweden Email:
[email protected]
Jules M. Nazzaro University of Kansas Medical Center, Department of Neurology, 3599 Rainbow Blvd, Mailstop 2012, Kansas City, KS 66160, USA
John Mitrofanis University of New South Wales, Australia
Mary B. Newman Department of Neurosurgery, Rush University Medical Center, Chicago, IL, USA
Alon Y. Mogilner Section of Functional and Restorative Neurosurgery, North Shore-LIJ Health System, Manhasset, New York, USA Anne Morel Department of Functional Neurosurgery, University Hospital Zu¨rich, Zu¨rich, Switzerland Alexander Muacevic European CyberKnife™ Center Munich, Max Lebsche Platz 31, 81377 Munich, Germany Email:
[email protected] Debabrada Mukherjee Department of Neurosurgery, Toronto Western Hospital, 399 Bathurst Street, M5T 2S8, Toronto, Canada Karim Mukhida Cell Restoration Laboratory, Room 12H1, Departments of Anatomy & Neurobiology and Surgery (Neurosurgery), Dalhousie University, 5850 College Street, Halifax, Nova Scotia, Canada B3H 1X5 Email:
[email protected] Dipankar Nandi Imperial College London and Charing Cross Hospital, London, UK
Michael E. Newmark Kelsey-Seybold Clinic, 2727 W. Holcombe Blvd, 2nd floor Neurology, TX 77025, Houston, USA Email:
[email protected] Jean Paul Nguyen Department of neurosciences, CHU Laennec, Nantes, France Email:
[email protected] John K. Niparko Department of Otolaryngology, Head & Neck Surgery, The Johns Hopkins School of Medicine, Baltimore, USA Email:
[email protected] Ajay Niranjan The Departments of Neurological Surgery, and Radiation Oncology, The University of Pittsburgh, PA, USA The Center for Image-Guided Neurosurgery, University of Pittsburgh Medical Center, Pittsburgh, PA 15213, USA Soheyl Noachtar Department of Neurology, Cleveland Clinic, 9500 Euclid Ave, OH 44195, Cleveland, USA Josef Novotny The Departments of Neurological Surgery, and
List of contributors
Radiation Oncology, The University of Pittsburgh, PA, USA The Center for Image-Guided Neurosurgery, University of Pittsburgh Medical Center, Pittsburgh, PA 15213, USA Wieslaw L. Nowinski Biomedical Imaging Lab, Agency for Science, Technology & Research (ASTAR), 30 Biopolis Street, 138671, The Matrix, Singapore Email:
[email protected] Jose´ Marı´a Nu´n˜ez Epilepsy Surgery Clinic; Unit for Stereotactic, Functional Neurosurgery and Radiosurgery, General Hospital of Mexico City, Mexico Bart Nuttin Department of Neurosurgery, University Hospital Gasthuisberg and Katholieke Universiteit Leuven, Herestraat 49, B–3000, Leuven, Belgium Email:
[email protected] Jose A. Obeso Department of Neurology and Neurosurgery, Clinica Universitaria and Medical School and Neuroscience Division, CIMA Centro de Investigacio´n Biome´dica en Red sobre Enfermedades Neurodegenerativas (CIBERNED), University of Navarra, Pamplona, Spain Email:
[email protected] Michael Y. Oh Assistant Professor of Neurosurgery, Drexel University College of Medicine Co-Director, Division of Neuromodulation Department of Neurosurgery, Allegheny General Hospital, 420 East North Ave, PA 15212, Pittsburgh, USA Email:
[email protected] Chihiro Ohye Functional & Gamma Knife Surgery Center, 886
Nakao-machi, 370–0001, Takasaki, Gunma, Japan Email:
[email protected] George Ojemann Department of Neurological Surgery RI–20, University of Washington Medical Center, Campus Box 356470, 1959 N.E. Pacific Street, WA 98195, Seattle, USA Email:
[email protected] David Omahen Department of Neurosurgery, Toronto Western Hospital, 399 Bathurst Street, M5T 2S8, Toronto, Canada Hiroshi Otsubo Division of Neurology, Department of Pediatrics, The Hospital for Sick Children, University of Toronto, 555 University Avenue, Toronto, Ontario, M5G 1X8, Canada Email:
[email protected] Alondra Oubre´ Department of Neurosurgery, North Shore University Hospital, 9 Tower, Manhassett, NY 11030, USA Jorge Roberto Pagura Centro Integrado de Dor, Rua Baltazar da Veiga 490, S.P. 04510, Sao Paolo, Brazil Email:
[email protected] Rajesh Pahwa University of Kansas Medical Center, Department of Neurology, 3599 Rainbow Blvd, Mailstop 2012, Kansas City, KS 66160, USA Kamalakannan Palanichamy Massachusetts General Hospital, Department of Radiation Oncology, Cox 3, 100 Blossom Street, Boston, MA 02114, USA
xliii
xliv
List of contributors
Stephan Palfi Service de Neurochirurgie, Hoˆpital Henri Mondor, Cre´teil, France Michael C. Park Department of Clinical Neurosciences, Program in Neurosurgery, The Warren Alpert Medical School of Brown University, Rhode Island Hospital, Providence, RI 02903, USA E-mail:
[email protected] Erik C. Parker Department of Neurological Surgery, NYU School of Medicine, 530 First Ave, Suite 8R, NY, USA Andrew G. Parrent Department of Clinical Neurosciences, University Hospital, 339 Windemere Rd., N6A 5A5, London, Ontario, Canada Email:
[email protected] Nikunj K. Patel Institute of Neurosciences, Frenchay Hospital, Bristol BS16 1LE, UK N. Pavon Movement Disorders, Functional Neurosurgery and Neurophysiology Units, Centro Internacional de Restauracion Neurologica (CIREN), Havana, Cuba Richard D. Penn Professor of Neurosurgery, The University of Chicago Medical Center Section of Neurosurgery, MC 3026 5841 South Maryland Avenue, Chicago, Illinois 60637, USA E-mail:
[email protected] Erlick A. C. Pereira Oxford Functional Neurosurgery, Nuffield Department of Surgery, University of Oxford and Department of Neurological Surgery, The West Wing, The John Radcliffe Hospital, Oxford, OX3 9DU, UK Email:
[email protected]
Roland Peyron CHU de Saint-Etienne & INSERM U879, University of St. Etienne, Lyon, France Email:
[email protected] Brigitte Piallat Joseph Fourier University, Pavillon B - Grenoble University Hospital, 38700, Grenoble, France Julie G. Pilitsis Assistant Professor, University of Massachusetts, School of Medicine, Worcester, MA, USA Email:
[email protected] Puneet Plaha Institute of Neurosciences, Frenchay Hospital, Bristol BS16 1LE, UK Bernd Pohlmann-Eden Bethel Epilepsy Centre, Bielefeld, Germany Epilepsy Service, Division of Neurology, Dalhousie University, Halifax, Canada Pierre Pollak Joseph Fourier University, Pavillon B - Grenoble University Hospital, 38700, Grenoble, France Kumar M. Prakash Division of Patient Based Clinical Research, Toronto Western Hospital McLaughlin Pavilion, 7th Floor Rm 7–403, 399 Bathurst Street, M5T 2S8, Toronto, Ontario, Canada Arthur Prochazka Department of Physiology and Centre for Neuroscience, University of Alberta, Edmonton AB T6G 2S2, Canada Jean Quint Oropilla Consultant in Neurosurgery, Makati Medical Center, Philippines Email:
[email protected]
List of contributors
Peter Rabins John Hopkins Hospital, Meyer 279, 600 N. Wolfe St., 21287–7279, MD, Baltimore, USA Email:
[email protected] Sylvie Raoul Service de Neurochirurgie, Hoˆpital G. et R. Lae¨nnec, Bd J. Monod, 44093 NANTES cedex, France Email:
[email protected] Stig Rehncrona Lunds Universitet, Box 117, SE-221 00 Lund, Sweden Email:
[email protected] R. Reichart Ames Associate (Smart Systems and Nanotechnology), NASA Ames Research Center, CA 94035, Moffett Field, USA Ali Rezai Director, Center for Neurological Restoration, Jane and Lee Seidman Chair in Functional Neurosurgery, Department of Neurosurgery, Cleveland Clinic, Cleveland, OH, USA Email:
[email protected]
H. Ian Robins Department of Medicine and Human Oncology, University of Wisconsin, 600 Highland Ave, K4-B100, Madison, WI 53792, USA Email:
[email protected] M. C. Rodriguez-Oroz Department of Neurology and Neurosurgery, Clinica Universitaria and Medical School and Neuroscience Division, CIMA Centro de Investigacio´n Biome´dica en Red sobre Enfermedades Neurodegenerativas (CIBERNED), University of Navarra, Pamplona, Spain R. Rodriguez-Rojas Movement Disorders, Functional Neurosurgery and Neurophysiology Units, Centro Internacional de Restauracion Neurologica (CIREN), Havana, Cuba Robert Rosenwasser Department of Neurosurgery, Thomas Jefferson University Hospital, Philadelphia, PA, USA V. Roualdes Department of neurosciences, CHU Laennec, Nantes, France
Donald Richardson Emeritus Professor Neurosurgery, Tulane University Health Science Center, New Orleans, Louisiana, USA Email:
[email protected]
James T. Rutka Division of Neurosurgery, Suite 1503, The Hospital for Sick Children, 555 University Avenue, Toronto, Ontario, Canada M5G 1X8 Email:
[email protected]
R. Mark Richardson Department of Neurological Surgery, University of California, San Francisco, 505 Panassus Ave, Box 0112, CA, USA
Tejas Sankar Department of Neurosurgery, Lahey Clinic Medical Center, Tufts University School of Medicine, Boston, Massachusetts, USA
David W. Roberts Section of Neurosurgery, Dartmouth Medical School, Hannover, NH 03755, USA Email:
[email protected]
Raymond Sawaya Department of Neurosurgery - 442, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd., Houston, TX 77030, USA Email:
[email protected]
xlv
xlvi
List of contributors
David W. Schaal Clinical Development, Accuray Incorporated, Sunnyvale, CA, USA Nicholas D. Schiff Department of Neurology and Neuroscience, Weill Medical College of Cornell University, 1300 York Avenue Room F610, New York, New York, 10021, USA Email:
[email protected] David Schlesinger Lars Leksell Center for Gamma Surgery, Department of Neurosurgery, University of Virginia, Charlottesville, VA, USA Michael Schulder Department of Neurosurgery, North Shore University Hospital, 9 Tower, Manhassett, NY 11030, USA Email:
[email protected] Theodore H. Schwartz Department of Neurological Surgery, Weill Medical College of Cornell University, New YorkPresbyterian Hospital, New York, NY, USA Gunnar Sedin University of Uppsala, Uppsala, Sweden Email:
[email protected]
Helen A. Shih Departments of Radiation Oncology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA Email:
[email protected] Jay L. Shils Department of Neurosurgery, Lahey Clinic, 41 Mall Road, 01805, Burlington, MA, USA Email:
[email protected] Dennis C. Shrieve Department of Radiation Oncology, Massachusetts General Hospital, 55 Fruit Street, MA 02114, Boston, USA Email:
[email protected] Marc P. Sindou Chairman Department of Neurosurgery, University Claude-Bernard of Lyon Hoˆpital Neurologique Pierre Wertheimer, 59 Bd Pinel, 69677, BRON CEDEX, France Email:
[email protected] V.A. Sironi Fondazione Istituto Neurologico, “C. Besta”, Milan, Italy
Eric Seigneuret Joseph Fourier University, Pavillon B - Grenoble University Hospital, 38700, Grenoble, France
Michelle J. Smith Department of Neurological Surgery, Weill Cornell Medical College, New York, NY, USA
Jason P. Sheehan Box 800–212, Department of Neurological Surgery, Health Sciences Center, Charlottesville, VA 22908, USA
Penny K. Sneed Department of Radiation Oncology, University of California-San Francisco, 505 Parnassus Avenue, CA 94143–0226, San Francisco, USA Email:
[email protected]
Andrew G. Shetter Division of Neurological Surgery, Barrow Neurological Institute, St. Joseph’s Hospital and Medical Center, Phoenix, Arizona 85013, USA Email:
[email protected]
Brian Snyder Division of Neurosurgery, Toronto Western Hospital, University of Toronto, UHN, Toronto, Ontario, Canada
List of contributors
J. C. Sol Multidisciplinary Pain Center Unit of Stereotactic and Functional Neurosurgery, Department of Neuroscience, University Hospital of Toulouse, Toulouse, France Jens Christian Sørensen Department of Neurosurgery, Aarhus University Hospital, DK–8000 Aarhus C, Denmark Philip A. Starr Department of Neurological Surgery, University of California, San Francisco, USA Email:
[email protected] Richard B. Stein Department of Physiology and Centre for Neuroscience, University of Alberta, Edmonton AB T6G 2S2, Canada Email:
[email protected] Ladislau Steiner Lars Leksell Center for Gamma Surgery, Department of Neurosurgery, University of Virginia, Charlottesville, VA, USA Email:
[email protected] Melita Steiner Lars Leksell Center for Gamma Surgery, Department of Neurosurgery, University of Virginia, Charlottesville, VA, USA Scellig Stone Division of Neurosurgery, Department of Surgery, Toronto Western Hospital, University of Toronto, Toronto, Ontario, Canada A. P. Strafella PET Imaging Centre, Center of Addiction Mental Health, University of Toronto, Toronto, Ontario, Canada Movement Disorders Center, Toronto Western Hospital & Research Institute, University of Toronto,
Toronto, Ontario, Canada Email:
[email protected] Volker Sturm Professor, Department of Stereotactic and Functional Neurosurgery, University Hospital, 50924 Cologne, Germany Email:
[email protected] Bomin Sun Center for Functional Neurosurgery (BMS), Shanghai Jiao Tong University Ruijin Hospital and Department of Functional Neurosurgery (JGZ,KZ,FGM), Beijing Tiantan Hospital, People’s Republic of China Peter Tai Epilepsy Program, Toronto Western Hospital, University of Toronto, Toronto, Canada Takaomi Taira Director of Stereotactic and Functional Neurosurgery, Department of Neurosurgery, Tokyo Women’s Medical University, 8–1 Kawada, Shinjuku, Tokyo 1628666, Japan Email:
[email protected] Mandeep Tamber Division of Neurosurgery, Toronto Western Hospital and University of Toronto, Toronto, Canada Eng-King Tan National Neuroscience Institute, Singapore General Hospital, Singapore Farzana Tariq Department of Neurosurgery, Brigham and Women’s Hospital, 75 Francis St, Boston, MA 02115, USA Daniel Tarsy Department of Neurology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts, USA
xlvii
xlviii
List of contributors
Ronald R. Tasker Department of Physiology, University of Toronto, 1 King’s College Circle, M5S 1S8, Toronto, Canada Ethan Taub Oberarzt, Department of Neurosurgery, Basel University Hospital, Spitalstrasse 21, CH–4031 Basel, Switzerland Email:
[email protected] Michael D. Taylor Sick Kids Hospital, 555 University Avenue, M5G 1X8, Toronto, Ontario, Canada Email:
[email protected] Manoel Jacobsen Teixeira Division of Functional Neurosurgery, University of Sa˜o Paulo Medical School, Rua Mena Barreto 765 - Itaim Bibi, SP- 014033–010, Sa˜o Paulo, Brazil Email:
[email protected]
Ohio, USA Email:
[email protected] Napoleon Torres Joseph Fourier University, Pavillon B - Grenoble University Hospital, 38700, Grenoble, France Zulma Tovar-Spinoza Division of Neurosurgery, Suite 1503, The Hospital for Sick Children, 555 University Avenue, Toronto, Ontario, Canada M5G 1X8 David Trejo Epilepsy Surgery Clinic; Unit for Stereotactic, Functional Neurosurgery and Radiosurgery, General Hospital of Mexico City, Mexico Wen-Ching Tzaan Department of Neurosurgery, Chang Gung Medical College and Memorial Hospital, 5 Zu Shung St., Kweishan, Taoynan 333, Taiwan
Carlos R. Telles-Ribeiro Neurodor - Neurosurgery and Pain Clinic, Rua Visconde de Piraja 351/509, Cep: 22471–002, Ipanema, Rio de Janeiro, RJ, Brazil Email:
[email protected]
Ryan J. Uitti Mayo Clinic College of Medicine, 4500 San Pablo Rd, FL 32224, Jacksonville, USA
David G. T. Thomas Division of Neurosurgery, The National Hospital for Neurology & Neurosurgery, Queen Square, Box 32, WC1N 3BG, London, UK
Taufik A. Valiante Univeristy of Toronto, Department of Surgery, CoDirector, Epilepsy Program, Krembil Neuroscience Center, UHN, 399 Bathurst Street 4W–436, Toronto, Ontario, Canada, M4S3H6 Email:
[email protected]
Daniel Roy Thomas Department of Neurological Sciences, Christian Medical College, Vellore, Tamil Nadu 632004, India Stavropoula Tjoumakaris Department of Neurosurgery, Thomas Jefferson University Hospital, Philadelphia, PA, USA William D. Tobler Department of Neurosurgery, University of Cincinnati College of Medicine, ML 0515, Cincinnati,
T. van Eimeren PET Imaging Centre, Center of Addiction Mental Health, University of Toronto, Toronto, Ontario, Canada Kris van Kuyck Department of Neurosurgery, Laboratory of Experimental Neurosurgery and Neuroanatomy, Katholieke Universiteit Leuven, Belgium
List of contributors
X. A. Vasques CHRU Montpellier, Service de Neurochirurgie, Universite´ Montpellier I, Montpellier, France
Ana Luisa Velasco Cerrada Bosques de Moctezuma 55, La Herradura Huixquilucan, Estado de Me´xico 52780, Me´xico Email:
[email protected] Francisco Velasco Epilepsy Surgery Clinic; Unit for Stereotactic, Functional Neurosurgery and Radiosurgery, General Hospital of Mexico City, Mexico Email:
[email protected]
Jerrold L. Vitek Department of Neurosciences, Cleveland Clinic Foundation, 9500 Euclid Ave, NC30, Cleveland, OH 44195, USA Email:
[email protected] Valerie Voon NINDS/NIH, 10 Center Dr, 20892–1428, Bethesda, Maryland, USA Email:
[email protected] Juhn A. Wada Division of Neurosciences, University of British Columbia, 2329 West Mall, V6T 1Z4, Vancouver, Canada Email:
[email protected]
Marcos Velasco Epilepsy Surgery Clinic; Unit for Stereotactic, Functional Neurosurgery and Radiosurgery, General Hospital of Mexico City, Mexico
J. Walter Ames Associate (Smart Systems and Nanotechnology), NASA Ames Research Center, CA 94035, Moffett Field, USA
J. C. Verdie Multidisciplinary Pain Center Unit of Stereotactic and Functional Neurosurgery, Department of Neuroscience, University Hospital of Toulouse, France
Richard Wennberg Department of Neurosurgery, Toronto Western Hospital, 399 Bathurst Street, M5S 2T8, Toronto, Ontario, Canada Email:
[email protected]
Jean-Guy Villemure Director, Division of Neurosurgery, University of Montreal, Professor and Chairman, Neurosurgery Service, Centre hospitalier de l’universite´ de Montre´al, 1560 Sherbrooke East, Montreal, Canada H2L 4M1 Email:
[email protected]
Karin N. Westlund Marine Biomedical Institute, University of Texas Medical Branch, 2925 Beluche, TX 77551, Galveston, USA
Veerle Visser-Vandewalle Department of Neurosurgery, Academic Hospital Maastricht, P.O. Box 5800, 6202, AZ Maastricht, The Netherlands Email:
[email protected]
Robert E. Wharen, Jr. Mayo Clinic College of Medicine, 4500 San Pablo Rd, FL 32224, Jacksonville, USA Email:
[email protected] Edward White Institute of Neurosciences, Frenchay Hospital, Bristol BS16 1LE, UK
xlix
l
List of contributors
Donald M. Whiting Drexel University College of Medicine Director, Division of Neuromodulation Vice-Chairman, Department of Neurosurgery, Allegheny General Hospital, 420 East North Ave, PA 15212, Pittsburgh, USA Thomas Wichmann Department of Neurology, Emory University Woodruff Memorial Research Building 101 Woodruff Circle, Atlanta, GA 30322, USA Heinz Gregor Wieser Neurology Department, University Hospital Zurich, Frauenklinikstrasse 26, 8091 Zurich, Switzerland Email:
[email protected] William D. Willis, Jr. Marine Biomedical Institute, University of Texas Medical Branch, 2925 Beluche, TX 77551, Galveston, USA Email:
[email protected] Shiao Y. Woo The University of Texas MD Anderson Cancer Center, Proton Therapy Center- Houston, 1840 Old Spanish Trail, TX 77054, Houston, USA Email:
[email protected] Isaac Yang Department of Neurological Surgery, University of California San Francisco, USA Email:
[email protected] Je´roˆme Yelnik INSERM U679, Hoˆpital de la Salpeˆtrie`re, 47, Bd de
l’Hoˆpital, 75013, Paris, France Email:
[email protected] Chun-Po Yen Lars Leksell Center for Gamma Surgery, Department of Neurosurgery, University of Virginia, Charlottesville, VA, USA Jianguo Zhang Center for Functional Neurosurgery (BMS), Shanghai Jiao Tong University Ruijin Hospital and Department of Functional Neurosurgery (JGZ,KZ, FGM), Beijing Tiantan Hospital, People’s Republic of China Kai Zhang Center for Functional Neurosurgery (BMS), Shanghai Jiao Tong University Ruijin Hospital and Department of Functional Neurosurgery (JGZ,KZ, FGM), Beijing Tiantan Hospital, People’s Republic of China Ludvic U. Zrinzo Division of Neurosurgery, The National Hospital for Neurology & Neurosurgery, Queen Square, Box 32, WC1N 3BG, London, UK Email:
[email protected] Dominik Zumsteg Neurology Department, University Hospital Zurich, Frauenklinikstrasse 26, 8091 Zurich, Switzerland
7 A Brief History of Stereotactic Neurosurgery in Switzerland E. Taub
Though he was not a neurosurgeon, the first great stereotactician in Switzerland was surely Walter Rudolf Hess (1881–1973), Professor of Physiology at the University of Zurich from 1917 to 1951. From the 1920s onward, he performed basic studies on the functional organization of the diencephalon, using electrical stimulation through stereotactically implanted depth electrodes in freely moving cats [1]. For this work, he was given the Nobel Prize in Physiology or Medicine in 1949, sharing the award with Anto´nio Egas Moniz (1874–1955), the Portuguese inventor of leukotomy and of cerebral angiography. Hugo Krayenbu¨hl (1902–1985), who had been trained as a neurosurgeon by Sir Hugh Cairns in London, became Professor of Neurosurgery in Zurich in 1948 – the first such chair in the country. He was active in functional neurosurgery from the beginning, publishing a study of prefrontal leukotomy and topectomy for the treatment of pain in 1950 [2] and a case series of temporal cortical excision and lobectomy for epilepsy in 1953 [3]. It is reported that he performed the first stereotactic procedures in Switzerland in 1958, in collaboration with Mehmet Gazi Yas¸argil, operating with a Riechert frame [4]. (It was, of course, Yas¸argil who was to succeed Krayenbu¨hl as chief of neurosurgery upon his retirement in 1973). In these early years of functional (lesional) stereotaxy, Krayenbu¨hl and Yas¸argil used equipment designed by their Zurich physiologist colleague O. A. M. Wyss [5] to perform both thalamotomies and #
Springer-Verlag Berlin/Heidelberg 2009
pallidotomies for the treatment of Parkinson’s disease and other extrapyramidal disorders [6]. In the 1960s, functional and stereotactic neurosurgery in Krayenbu¨hl’s department became the domain of Jean Siegfried, who had developed his interests in the field as a Fellow under William H. Sweet at Massachusetts General Hospital in 1962. In the late 1960s, Siegfried wrote on percutaneous cordotomy and the surgical treatment of spasmodic torticollis and proposed stereotactic dentatotomy for the treatment of movement disorders [7,8,9]. Toward the end of the decade, with the introduction of L-DOPA, Siegfried took on the question of medical versus surgical treatment of Parkinson’s disease, writing extensively on the subject and gaining a very large clinical experience in stereotactic lesionmaking [10]. At the same time, he became Switzerland’s leading expert in Sweet’s technique of radiofrequency thermocoagulation of the Gasserian ganglion for the treatment of trigeminal neuralgia, eventually amassing a huge case series: a report on the first 500 cases, published in 1977, was followed by a report on 1,000 cases in 1981 [11,12]. Siegfried went on to become a pioneer of deep brain stimulation and one of the prime movers in the transformation of functional neurosurgery – now nearly, though not entirely, complete – from a ‘‘lesioning’’ to a ‘‘stimulating’’ discipline. He stimulated the sensory thalamus for the treatment of chronic pain in a series of 89 patients from 1978 to 1985 [13] and began to perform thalamic stimulation for tremor in the
74
7
A brief history of stereotactic neurosurgery in switzerland
late 1980s [14]. After the reintroduction of pallidotomy by Laitinen and colleagues in the early 1990s, Siegfried was the first to describe the achievement of comparable effects by pallidal stimulation [15]. After Siegfried’s move from the University across the Lake of Zurich to the Klinik Im Park (originally called AMI Klinik Im Park), a private clinic that had been opened in 1986, he also had the distinction of introducing Gamma Knife radiosurgery to Switzerland. He founded the Gamma Knife Center at the Klinik Im Park in 1994 and headed it until his retirement in 2001. Siegfried’s activities at the Klinik Im Park were continued by Thomas Mindermann (2000-present) and Ethan Taub (2001– 2007), a former fellow of Ronald Tasker and Andres Lozano at the University of Toronto. Epilepsy surgery had already been an interest of Krayenbu¨hl’s in the 1950s, as mentioned above, and was further developed in Zurich by both Siegfried and Yas¸argil. In 1970, Siegfried and the epileptologist Christoph Bernoulli brought the technique of stereoencephalography (SEEG) for the localization of epileptic discharges to Zurich, 5 years after its description by Bancaud and Talairach in Paris [16]. In the early 1980s, together with the epileptologist Heinz Gregor Wieser, Yas¸argil pioneered the technique of selective amygdalohippocampectomy for the treatment of mesiobasal limbic epilepsy [17]. At around the same time, SEEG gave way to semi-invasive localization via foramen ovale electrodes, which were first described by Siegfried, Wieser, and H. R. Stodieck [18]. From 1993 onward, the Zurich tradition of selective amygdalohippocampectomy was continued by Yas¸argil’s successor as chairman, Yasuhiro Yonekawa. Also at the University of Zurich, Daniel Jeanmonod has headed the Laboratory for Functional Neurosurgery for some years, continuing into the present. Known in the clinical sphere as one of the few current proponents of deep brain lesioning, as opposed to stimulation, he has pursued basic research on the pathophysiology of
movement disorders and other neurological conditions. His research group has published a detailed stereotactic atlas of the thalamus [19]. Off to the west, in the French-speaking part of Switzerland, both epilepsy surgery and functional stereotaxy became major areas of clinical activity in Lausanne after Jean-Guy Villemure moved there from the Montreal Neurological Institute to become chairman in 1997. Villemure’s department achieved high case numbers and became a renowned center of excellence in both areas. In particular, he and his group amassed a large series of patients who underwent subthalamic stimulation for Parkinson’s disease and published very important findings about both the usefulness of this procedure and its complications [20,21]. After Villemure’s retirement in 2006, his activities in the functional field were continued by his disciples Jocelyne Bloch and Claudio Pollo under his successor as chairman, Marc Levivier. Joachim-Kurt Krauss, currently chief of neurosurgery at the Medizinische Hochschule in Hanover (Germany), briefly headed the functional neurosurgery unit at the University of Berne in the 1990s. He was succeeded in this capacity (1999–2001) by Ethan Taub, who later brought deep brain stimulation procedures to Basel (2007present) in collaboration with Morten Wasner. Deep brain stimulation in Berne is currently the responsibility of Alexander Stibal. Heinz Fankhauser began a deep brain stimulation program at the private Clinique Cecil in Lausanne in 2002, while Ronald Bauer, a former fellow of Tipu Aziz at Oxford, has led the functional neurosurgery unit at the Cantonal Hospital in St. Gallen since 2006. The current state of functional neurosurgery in Switzerland can thus be briefly summarized: There are active clinical programs in deep brain stimulation at the University Hospitals of Lausanne, Berne, and Basel, as well as at the Cantonal Hospital in St. Gallen and at a private clinic in Lausanne. The Gamma Knife Center at the Klinik Im Park, long the only dedicated facility for stereotactic radiosurgery
A brief history of stereotactic neurosurgery in switzerland
in Switzerland, is still in operation as of this writing and is likely to be joined by other centers in this field in the near future. Across the country, scientific presentations for the medical public and articles in the lay press are gradually heightening awareness of the uses of functional neurosurgery among referring physicians and prospective patients, and activity in the field can be expected to expand and to flourish.
References 1. Hess WR. Die funktionelle Organisation des vegetativen Nervensystems. Basel: Schwabe; 1948. 2. Krayenbu¨hl H, Stoll W. [Prefrontal leucotomy and topectomy for the treatment of irreducible pain]. Rev Neurol (Paris) 1950;83(1):40-1. 3. Krayenbu¨hl H, Hess R, Weber G. [Electroencephalographic, corticographic, and surgical considerations on 21 cases of temporal epilepsy treated by cortical excision and lobectomy]. Rev Neurol (Paris) 1953;88(6):564-7. 4. Siegfried J, quoted in Nashold BS. The history of stereotactic neurosurgery. Stereotact Funct Neurosurg 1994;62:29-40. 5. Wyss OAM. Hochfrequenzkoagulationsgera¨t zur reizlosen Ausschaltung. Helv Physiol Pharmacol Acta 1945;3:437-43. 6. Krayenbu¨hl H, Wyss OAM, Yas¸argil MG. Bilateral thalamotomy and pallidotomy as treatment for bilateral Parkinsonism. J Neurosurg 1961;18:429-44. 7. Siegfried J. [Percutaneous cordotomy]. Schweiz Med Wochenschr 1967;97(40):1325-6. 8. Siegfried J. [Surgery for spasmodic torticollis]. Schweiz Med Wochenschr 1967;97(40):1325. 9. Siegfried J, Perret E. [Stereotaxic dentatotomy. New method of surgical treatment of hyperkinesis]. Rev Otoneuroophtalmol 1968;40:(7)341-3.
7
10. Krayenbu¨hl H, Siegfried J. [Treatment of Parkinson’s disease: L-dopa or stereotaxic technics?]. Neurochirurgie 1970;16(1):71-6. 11. Siegfried J. 500 Percutaneous thermocoagulations of the Gasserian ganglion for trigeminal pain. Surg Neurol 1977;8(2):126-31. 12. Siegfried J. Percutaneous controlled thermocoagulation of Gasserian ganglion in trigeminal neuralgia: experiences with 1000 cases. In: Samii M, Jannetta P, editor. The cranial nerves. Berlin: Springer; 1981. p. 322-30. 13. Siegfried J. Sensory thalamic neurostimulation for chronic pain. Pacing Clin Electrophysiol 1987;10(1 Pt 2):209-12. 14. Blond S, Siegfried J. Thalamic stimulation for the treatment of tremor and other movement disorders. Acta Neurochir Suppl (Wien) 1991;52:109-11. 15. Siegfried J, Lippitz B. Bilateral chronic electrostimulation of ventroposterolateral pallidum: a new therapeutic approach for alleviating all parkinsonian symptoms. Neurosurgery 1994;35(6):1126-9. 16. Lu¨ders H, Comair YG, (eds). Epilepsy surgery. 2nd ed. Philadelphia, PA: Lippincott Williams and Wilkins; 2000. p. 48. 17. Wieser HG, Yas¸argil MG. Selective amygdalohippocampectomy as a surgical treatment of mesiobasal limbic epilepsy. Surg Neurol 1982;17(6):445-57. 18. Siegfried J, Wieser HG, Stodieck SR. Foramen ovale electrodes: a new technique enabling presurgical evaluation of patients with mesiobasal temporal lobe seizures. Appl Neurophysiol 1985;48(1–6):408-17. 19. Morel A, Magnin M, Jeanmonod D. Multiarchitectonic and stereotactic atlas of the human thalamus. J Comp Neurol 1997;387(4):588-630. 20. Vingerhoets FJ, Villemure JG, Temperli P, Pollo C, Pralong E, Ghika J. Subthalamic DBS replaces levodopa in Parkinson’s disease: two-year follow-up. Neurology 2002;58(3):396-401. 21. Burkhard PR, Vingerhoets FJ, Berney A, Bogousslavsky J, Villemure JG, Ghika J. Suicide after successful deep brain stimulation for movement disorders. Neurology 2004;63(11):2170-2.
75
16 History of Stereotactic and Functional Neurosurgery in Brazil O. Vilela Filho
When Andres Lozano invited me to write this chapter, although honored, I declined the invitation at first. After all, very little had been written about this issue. The neurosurgeon Sebastia˜o Gusma˜o, the person who has most studied the history of Brazilian neurosurgery, has already written a significant number of papers about this subject, but basically nothing about the history of the Brazilian stereotactic and functional neurosurgery. Therefore, being the president of the Brazilian Society for Stereotactic and Functional Neurosurgery, I felt impelled to accept the challenge and write this chapter, but not without some extra pressure from Lozano. It was a tough job. I called and emailed colleagues throughout this country innumerable times, at any time, and I suppose they must be tired of me, obviously not without reason. In the case of colleagues already dead, their relatives, friends and colleagues were my target. Not very infrequently, the same fact was presented with different versions, obliging me to dig even deeper. But here it comes at last, the first paper on this issue. I have done my very best and, I must say, it truly fascinated me. If any incongruity has occurred, it was not my fault, since most of the times I had to base myself on told stories instead of recorded information. Dear editors and friends, please accept my apologies for having missed the deadline so many times. The beginning of the history of Brazilian general and functional neurosurgery is closely intermingled, being humanly impossible to separate them, and that is how I decided to start this chapter. #
Springer-Verlag Berlin/Heidelberg 2009
A Brief History of Brazilian Neurosurgery and the First Functional Procedures Performed in Brazil According to Horrax [1], the history of neurosurgery in modern times may be divided into three periods: pre-Lister (1710–1846), preHorsley (1846–1890), and modern neurosurgery (from 1890 until today). The first known neurosurgical procedure performed in Brazil was carried out by the surgeon Luis Gomes Ferreyra, in 1710, in the vicinities of Sabara´, a town located in the countryside of Minas Gerais State. The patient was a slave presenting with a compound depressed skull fracture caused by the natural fall of a tree branch. Ferreyra removed the bone fragments, performed hemostasis using silk threads from cobwebs soaked in egg white, and covered the bone defect with a piece of calabash shell until bone healing. The patient eventually recovered and returned to work, presenting as only sequel a very mild expression dysphasia [2]. The pre-Horsley period started with the clinical introduction of general anesthesia by Morton, in 1846, and of antisepsis by Lister, in 1867. In Brazil, differently from Europe, this period extended until 1928. During this phase, neurosurgical interventions were almost completely limited to the treatment of depressed skull fractures and drainage of extracerebral hematomas and abscesses [2,3].
198
16
History of stereotactic and functional neurosurgery in brazil
Curiously, many functional procedures were started in Brazil in the end of the pre-Horsley period. Following are the pioneers of some of these operations [3]. Paolo Josetti, in 1896, performed a gasserian ganglion resection for the treatment of trigeminal neuralgia. In 1903, Joa˜o Burnier and Jose´ Hungria Jr performed periarterial sympathectomy for the treatment of causalgia. Paes Leme, in an imprecise date, but certainly before 1910, carried out the first surgical procedure for epilepsy in a patient presenting with Jacksonian seizures, starting in the right hemiface, caused by encephalomalacia of the central part of the left middle frontal gyrus, which was secondary to head injury and subsequent brain abscess. In 1913, Nascimento Gurgel employed peripheral neurectomies (Stoffel’s technique) to treat spasticity. From this period, worth mentioning was the contribution from Augusto Paulino de Souza and Ame´rico Vale´rio [3]. At that time, the prevalent idea was that the entire rolandic area (pre- and postcentral gyri) was related to motor control. In 1907, these surgeons operated on a victim from cranial missile wound presenting with contralateral hemiplegia. During the operation, being the brain widely exposed through a large craniectomy, they observed the complete integrity of the posterior rolandic area (postcentral gyrus). Based on this surgical finding, they hypothesized that the posterior rolandic area was not involved in motor control. This hypothesis, presented at the Fourth Latin American Medical Congress, held in Rio de Janeiro in 1909, and mentioned in the 1922 paper of the authors [3], was confirmed by Horsley, Campbell, and Dana 2 years later (1909). The above mentioned authors, among many others, are considered the predecessors of neurosurgery in our country [2]. Two surgeons are regarded as the precursors of Brazilian neurosurgery: Augusto Branda˜o Filho, known as the prince of the surgeons, and Alfredo Monteiro, another formidable general
surgeon, both from Rio de Janeiro, at that time the capital of the country [2]. Branda˜o Filho, in 1924, was the first to perform ventriculography in Brazil, and on 6 August 1928, directly supervised by Egas Moniz, he performed the first cerebral angiography in the Americas, in fact, only 14 months after Egas Moniz, assisted by Almeida Lima, invented this procedure (28 June 1927) [2]. Branda˜o Filho also performed some open cordotomies (according to Jose´ Portugal), two gasserian ganglion resections for trigeminal neuralgia (though the first patient died, an excellent result was achieved in the second), as well as brain tumor resection in seven patients, all of which died according to his publication of 1931 [2]. Noteworthy is the fact that the initial experimental researches performed by Egas Moniz and Almeida Lima that culminated with the development of human cerebral angiography were carried out at Instituto Rocha Cabral, in Lisbon. This institute was founded and maintained by the Portuguese Bento da Rocha Cabral, greatuncle of one of the most important Brazilian neurosurgeons, still very active, Guilherme Cabral, from Belo Horizonte, capital of Minas Gerais State [2]. In 1928, Antoˆnio Austrege´silo Rodrigues Lima, an eminent neurologist and the first Professor of Neurology in Brazil (1912, Federal University of Rio de Janeiro, as it is called today), visited many neurosurgical services in USA, including those headed by Cushing, Dandy, Adson, and Frazier, and was very impressed with what he saw. Back to Brazil, he urged the brilliant and very skillful general surgeon, Alfredo Monteiro, and his young and recently graduated (1927) assistant, Jose´ Ribe Portugal, to start neurosurgery in Brazil. To do so, he inaugurated the Neurosurgical Service of the Department of Neurology of Federal University of Rio de Janeiro (Universidade Federal do Rio de Janeiro – UFRJ). In 1932, this service became a department, and Alfredo Monteiro was
History of stereotactic and functional neurosurgery in brazil
invested its head and the first Professor of Neurosurgery in Brazil. In 1935, Monteiro, under his own request, was transferred to the Department of Operative Techniques and Experimental Surgery, and Portugal, stimulated by his former boss, took over the chair of Neurosurgery [2]. In the early 1930s, both Branda˜o Filho and Alfredo Monteiro, probably disenchanted with their results, mainly in brain tumor operations, and realizing that surgery of the nervous system should be performed by people with adequate knowledge and formation in neurological sciences, including anatomy, pathophysiology, semiology, neurology, and neurosurgical techniques, that is, by true neurosurgeons, decided to abandon neurosurgery. They really never considered themselves as neurosurgeons, but general surgeons performing neurosurgery [2]. For the reasons aforementioned, Antoˆnio Austrege´silo Rodrigues Lima may be regarded as the godfather of Brazilian neurosurgery. The inauguration of modern neurosurgery in the world, in 1890, was led by the association of the following essential technological innovations: general anesthesia, antisepsis, Broca’s theory of cerebral localization (further corroborated and enlarged by the findings of Jackson, Fritsch & Hitzig, and Ferrier) and the studies on cranioencephalic topography carried out by Broca (still very important, but even more in a time when neuroimaging studies were not available) [2]. In the late 1920s, being neurosurgery well established in the world, neurology and general surgery well developed in Brazil, and considering the fact that state-of-the-art neuroimaging techniques (ventriculography and cerebral angiography) had already been introduced in the country, the moment was favorable for the birth of modern neurosurgery in Brazil [2]. Modern Brazilian neurosurgery was then inaugurated [2], having as pioneers Jose´ Ribe Portugal (1901–1992) and Elyseu Paglioli
16
(1898–1985), and in our opinion, also Antoˆnio Carlos Gama Rodrigues, or Carlos Gama (1904– 1963), as he was known. All of them started neurosurgery in their respective states, Rio de Janeiro, Rio Grande do Sul, and Sa˜o Paulo, approximately at the same time. Jose´ Portugal and Carlos Gama were, initially, self-taught, and just sometime later in their carriers they visited the services of the icons of world neurosurgery. Portugal, in 1945, spent 4 months with John Scarff, in New York, returning in 1947 for 6 more months. Gama, on the other hand, visited the service of Harvey Cushing. Differently, Elyseu Paglioli only started his neurosurgical practice after spending 8 months in Paris (1930) with De Martel, one of the pioneers of French neurosurgery [2]. Elyseu Paglioli and Jose´ Portugal founded two of the most important Brazilian neurosurgical schools. Of interest for this chapter, Paglioli formed Manoel Krimberg, one of the beginners of stereotactic and functional neurosurgery in Rio Grande do Sul State, and Djacir Figueiredo, the pioneer of stereotactic and functional neurosurgery in Ceara´ State. From Portugal’s school emerged Renato Barbosa, regarded as the pioneer of stereotactic and functional neurosurgery in Brazil, Gianni Temponi, another very well known Brazilian functional neurosurgeon, and Joffre Lima, the beginner of stereotactic and functional neurosurgery in Para´ State. Carlos Gama did not found a neurosurgical school, but he did form Rolando Tenuto, who headed the Neurosurgical Service at the University of Sa˜o Paulo (Universidade de Sa˜o Paulo – USP) Medical School teaching hospital (Clinic Hospital) from its inauguration, in 1945, until his retirement, in 1970, having formed, among many others, Jose´ Zaclis, one of the pioneers in stereotactic and functional neurosurgery in Sa˜o Paulo State, and Raul Marino Jr, the most widely known Brazilian functional neurosurgeon; this service became also one of the most important Brazilian neurosurgical schools [2].
199
200
16
History of stereotactic and functional neurosurgery in brazil
The Brazilian Society of Neurosurgery (Sociedade Brasileira de Neurocirurgia – SBN) was founded by the twelve Brazilian neurosurgeons that attended the First International Congress of Neurological Surgeons in Brussels, Belgium, on 26 July 1957 [2], and became affiliated to the World Federation of Neurosurgical Societies in the same year. Jose´ Portugal and Elyseu Paglioli were among the twelve founding members, and were elected, respectively, the first and third president of the SBN. Both were later reelected for another term. According to the last census, performed in 2005, there are currently 2,251 neurosurgeons in Brazil, 1,735 of which are members of the SBN, making it the third in the world in number of members [2]. Interestingly, the three pioneers of neurosurgery in Brazil were in some way involved with functional neurosurgery, making relevant contributions to the field. Graduated from medical school in 1927, 2 years later Jose´ Portugal underwent examination for full professorship of the Department of Surgical Techniques and Experimental Surgery, Medical School, Federal University of Rio de Janeiro, occasion in which he presented the thesis entitled Contribuic¸a˜o a` Neurectomia Retrogasseriana (Contribution to Retrogasserian Neurectomy). This work was itself already a register of his main contributions to the field. His interest in the surgical treatment of trigeminal neuralgia, in fact, started during the last 3 years (total duration of medical school, including internship: 6 years) of his medical school, when he was a monitor of Anatomy (in the Brazilian university system, the monitor is a student who helps teaching a discipline he/she has already attended; to get this academic position, those interested should undergo an examination, when the best candidate is chosen), and remained his primary interest throughout his whole carrier. He was really a superb neurosurgeon [2]. According to Mario Brock, one of his disciples, Portugal used to perform a retrogasserian neurectomy in
roughly 20 min [4]. This procedure was so perfected by him that, in 1938, Leriche, the pioneer of pain surgery, came to Rio de Janeiro to observe one of his operations [2]. Regarding his contributions to the field of functional neurosurgery, the first one was registered when he was still a medical student and monitor of Anatomy. At that time, the prevailing idea, as shown in the first edition of Hovelaque’s book, was that the motor root of the trigeminal nerve crossed the gasserian ganglion though its outer aspect and joined the mandibular division of the trigeminal nerve externally. While making an anatomical preparation of the Fifth cranial nerve, however, Portugal noted that its motor root crossed the gasserian ganglion through its middle part and joined the mandibular division internally. Puzzled with this finding, which was in contradiction with the ‘‘bible’’ of Anatomy, he dissected many other specimens, confirming his initial observation. His findings were published in 1926, and again in 1929, being cited in Rouviere’s Anatomy [2]. His second contribution was in regard to the relative distribution of the three divisions of the trigeminal nerve in its sensory root. According to Stookey, the lateral, intermediate and medial parts of the trigeminal sensory root corresponded, respectively, to V3, V2 and V1. Portugal, however, showed that the entire external half of the sensory root corresponded to V3, and that the preservation of the medial 1/5 of the sensory root was enough to maintain corneal innervation and so prevent keratitis and consequent visual loss [2]. Finally, the third and most important contribution from Portugal to this field was the original clinical introduction of intradural retrogasserian neurectomy for the treatment of trigeminal neuralgia (1929) in a time when everybody else used the extradural approach [2]. Carlos Gama was also very fond of trigeminal procedures. In 1929 he improved the technique of gasserian ganglion alcoholization and published his first papers on this subject. Later, in 1938, Gama became Professor of the Department
History of stereotactic and functional neurosurgery in brazil
of Neurology at the USP Medical School presenting his thesis entitled Neuralgias do Trigeˆmeo (Trigeminal Neuralgias) [2]. Elyseu Paglioli was a neurosurgeon of many publications, two of which are of particular interest to the field of functional neurosurgery. The first was the thesis Circulac¸a˜o Venosa dos Nu´cleos Centrais do Ce´rebro (Venous Circulation of the Basal Ganglia), which was presented in 1927 to secure the chair of Anatomy at the Federal University of Rio Grande do Sul, and published in 1929 [2]. The second was the thesis entitled Ventriculografia (Ventriculography), prefaced by his former chief, De Martel, and presented in 1938 to obtain the chair of Surgical Clinic at the Federal University of Rio Grande do Sul Medical School [2]. The apex of his stupendous academic carrier was reached in 1950, when he became the dean of the university, a position he occupied for twelve consecutive years. Besides his neurosurgical and academic carriers, he was also engaged in politics, being mayor of Porto Alegre, the capital of Rio Grande do Sul State, and Minister of Health during the administration of the Brazilian President Joa˜o Goulart [2].
Stereotactic and Functional Neurosurgery in Brazil: From Early Days to Present Time Brazil is a large country, the fifth in the world in extension. It is divided into 26 states and the federal district, which are agrouped in five regions: North, Northeast, Midwest, Southeast, and South. For didatical reasons, we decided to describe the history of stereotactic and functional neurosurgery in Brazil considering its different regions.
Southeast Region The Southeast region is made up by the states of Rio de Janeiro, Sa˜o Paulo, Minas Gerais, and Espı´rito Santo.
16
It is hard to ascertain who the pioneer of stereotactic and functional neurosurgery in the Southeast and in Brazil was. It depends on a number of parameters not clearly defined, but certainly one of the following will fulfill the criteria: Renato Tavares Barbosa, Paulo Niemeyer, both from Rio de Janeiro, or Aloysio Mattos Pimenta, from Sa˜o Paulo.
Rio de Janeiro State Renato Tavares Barbosa (1912–2003) (> Figure 16-1) graduated from the Federal University of Minas Gerais Medical School. After that, he went to Rio de Janeiro, being the first disciple formed by Jose´ Portugal, pioneer of Brazilian neurosurgery. Initially, he worked as an assistant to Portugal [2]. In the 1950s, Barbosa went to Freiburg, Germany, where he got his training in stereotactic and functional neurosurgery with Riechert and Mundinger [5]. His first stereotactic operation was performed at Rio de Janeiro Neurological Institute, Federal University of Rio de Janeiro, . Figure 16-1 Renato Barbosa
201
202
16
History of stereotactic and functional neurosurgery in brazil
in 1958, using the Riechert and Mundinger apparatus [2]. Throughout his life he accumulated a huge experience with stereotactic treatment of movement disorders, mainly Parkinson’s disease, and psychosurgery (open procedures such as prefrontal leucotomy and cingulectomy), initially at Rio de Janeiro Neurological Institute, and later at Lagoa Hospital, where he was head of the Neurosurgery Service for 27 years, and at Sorocaba Clinic, a private hospital. According to Luiz Fernando Martins, in 1971, at that time doing his internship at Lagoa Hospital, Barbosa was the first in Brazil to use a radiofrequency generator to perform thermocoagulation lesions. Until then, thermocoagulation lesions were made by using electrocautery: placing the stethoscope on the skull, one could ‘‘hear’’ the lesion being produced, which was stopped when tremor arrest was achieved or side effects started to appear. He was one of the founders of the Brazilian Society of Neurosurgery, being elected its president for the biennium 1970/1972 [2]. He was also one of the founders of Brazilian Society for Stereotactic and Functional Neurosurgery (1980), becoming its first president [6]. Renato Barbosa, one of the pioneers of stereotactic and functional neurosurgery in our country, was the first to dedicate most of his practice to this field, which he did during his whole career. For these reasons, he is regarded as the father of Brazilian stereotactic and functional neurosurgery [2]. Paulo Niemeyer (1914–2004) (> Figure 16-2), brother of the worldwide known architect, Oscar Niemeyer (a 100 years old and still very active, the planner of the most important buildings in Brası´lia, remarried his 60-year-old secretary when he was already 98) and father of Paulo Niemeyer Filho, at present one of the most renowned Brazilian neurosurgeons, was actually the first to perform stereotactic surgery in Brazil. Niemeyer graduated from the Federal University of Rio de Janeiro Medical School in 1936. His initial training was in general surgery, which he did under the supervision of Augusto Paulino,
. Figure 16-2 Paulo Niemeyer
Professor of Surgery and Head of the Surgical Clinic at Santa Casa de Miserico´rdia do Rio de Janeiro Hospital (at that time the teaching hospital of the Federal University of Rio de Janeiro Medical School), and Alfredo Monteiro, Professor of the Department of Operative Techniques and Experimental Surgery at the same institution and one of the precursors of Brazilian neurosurgery [2,7,8]. His interest in neurosurgery became apparent in 1939, when he started dedicating to the surgery of head injury, until then part of the general surgery practice [2,7]. At the beginning a self-taught neurosurgeon, just like Jose´ Portugal, later on he visited important and well-established neurosurgical centers of other countries [2]. Since 1943 he became interested in functional neurosurgery, particularly in the surgical treatment of movement disorders and epilepsy [2,7,8]. Before the stereotactic era, he treated many patients harboring Parkinson’s disease, athetosis
History of stereotactic and functional neurosurgery in brazil
and other hyperkinesias employing pyramidotomy or cortical resection [7,9], and on 14 April 1954, he performed the first stereotactic surgery (left pallidotomy for PD in a 60-year-old patient) in South America, using the frame he was given by Riechert [7]. The report of his experience with stereotactic treatment of dyskinesias was the first publication of this kind in South America [10]. According to Niemeyer, when he was still an 8-year-old child, while alone with his beloved sister, Judith, she presented a seizure, the first of a series to follow. It was this event that later led him to Medicine, in general, and to epilepsy surgery, in particular [8]. In 1946 he was the first to perform electrocorticography in Brazil [7]. Initially, he used this technique in an attempt to demonstrate, in humans, the existence of cortical spreading depression, as previously reported in animals by Aristides Lea˜o, a Brazilian scientist [7,8,11]. The paper derived from this research was widely recognized. Subsequently, he used not only electrocorticography, but also electrographic exploration with implanted electrodes, to determine epileptic foci [7,8]. In 1949, together with the neurologist Abraham Ackerman, he founded the Brazilian League Against Epilepsy [2,7,8]. Some meetings on epilepsy and electrocorticography were then organized. The French epileptologist Henri Gastaut was invited for two of these meetings (1954 and 1955). During his lectures, Gastaut showed evidence suggesting the involvement of amygdala and hippocampus in the genesis of temporal lobe epilepsy. At that time already experienced with Falconer’s en bloc temporal lobectomy technique (26 cases), Niemeyer, in 1956, inspired by Gastaut’s findings, was the first in the world to perform selective amygdalo-hippocampectomy for treating temporal lobe epilepsy. He approached the mesial temporal structures through a 2-cm incision placed in the second temporal gyrus, and named his technique transventricular amygdalohippocampectomy [2,7,8,12].
16
Recommended by Paul Bucy, he was invited by the Department of Health, Education, and Welfare of the American Government to participate in the Second International Colloquium on Temporal Lobe Epilepsy, in Washington, organized by the National Institute of Neurological Diseases and Blindnesses, in 1957, to show his technique, which was published in the following year [7,8,12]. Another important contribution from Niemeyer was his original observation that immediately after selective amygdalo-hippocampectomy, the abnormal cortical activity, as shown by pre-resection electrocorticography, becomes surprisingly worse, and that this abnormality subsides in a few days, as demonstrated through electrodes left in the subdural space [8,12]. This finding was reported in his 1958 paper [12]. More recently, two Brazilians, during their fellowship at Montreal Neurological Institute, the epileptologist Fernando Cendes and the epilepsy surgeon Arthur Cukiert, came to the same conclusion and called the aforementioned phenomenon ‘‘Niemeyer’s effect’’ [13]. In 1953, Niemeyer became head of the Department of Neurosurgery of Santa Casa de Miserico´rdia do Rio de Janeiro Hospital [7]. In 1957, he was among the 12 Brazilian neurosurgeons that founded the Brazilian Society of Neurosurgery [2]. In 1981, in Munich, he was elected second Vice-President of the World Federation of Neurosurgical Societies (WFNS), and in 1997, in Netherlands, he received the Honor Medal of the WFNS, which was awarded to five living neurosurgeons of the world that most contributed for the progress of neurosurgery [7]. The single thing he was most proud of, though, was the large number of neurosurgeons that he trained, now spread all over the country [7]. After 1958, unfortunately, his interest in functional neurosurgery progressively declined, and he became dedicated almost exclusively to general neurosurgery, in particular to microsurgery for
203
204
16
History of stereotactic and functional neurosurgery in brazil
intracranial aneurisms and transesphenoidal approach for pituitary tumors [2,7]. On 10 March 2004, Paulo Niemeyer passed away [7,8]. Undoubtedly, a great loss to Brazilian and World neurosurgery. Gianni Maure´lio Temponi (1934–2007), another pioneer of stereotactic and functional neurosurgery in Brazil, graduated from the Federal University of Triaˆngulo Mineiro Medical School in 1959 [2,14]. His neurosurgical training (1961/1964) was carried out at Rio de Janeiro Neurological Institute under Jose´ Portugal. Renato Barbosa was the functional neurosurgeon in charge at the same institution, and it was from him that he got his training in stereotactic and functional neurosurgery. When Barbosa moved to Lagoa Hospital, Temponi was left in charge of functional neurosurgery. Posteriorly, he became Director of Rio de Janeiro Neurological Institute, Full Professor of Neurosurgery at Federal University of Rio de Janeiro Medical School (1972), and founded the Neurosurgical Service at Clementino Fraga Filho Hospital, the teaching hospital at Federal University of Rio de Janeiro. As head of the Service, he formed 45 neurosurgeons. Like his master in stereotactic and functional neurosurgery, Renato Barbosa, his main areas of interest were stereotactic treatment of movement disorders and psychosurgery. During his career, always using the Riechert and Mundinger apparatus, Temponi performed 478 ventriculographyguided procedures, mainly thalamotomy for Parkinson’s disease [15], dentatotomy for cerebral palsy (dystonia, chorea and spasticity), the theme of his PhD thesis defended in 1971 (‘‘Stereoencephalotomy for cerebral palsy in children’’), and psychosurgery (especially for heretic oligophrenia). He was so confident and satisfied with his results that, contrarily to almost everyone else, he never replaced ventriculography with computed tomography and/or magnetic resonance imaging as a tool for determining the stereotactic coordinates of functional targets. Temponi left four offspring, two of
them (Vicente and Gianni) neurosurgeons in Rio de Janeiro. Carlos Roberto Telles Ribeiro or simply Carlos Telles (> Figure 16-3), as he is known, is certainly one of the most respected and renowned functional neurosurgeons in the country nowadays. Telles graduated from the State University of Rio de Janeiro Medical Sciences School in 1969. In 1973 he finished his residence training at Brası´lia District Hospital, in Brası´lia, under Paulo Mello. From 1976 to June of 1979, he stayed as a fellow at University of Hannover, doing his PhD under Wolfhard Winkelmu¨ller. Invited by Mario Brock, in July 1979 he moved to the University of Berlin, where he inaugurated the Pain Clinic and, in 1980, defended his PhD thesis, supervised by Winkelmu¨ller, entitled ‘‘Treatment of chronic pain by electrical stimulation of the spinal cord.’’ After almost 1 year in Berlin, he went to Duke University, spending 1 month with Blaine Nashold Jr. Returning to Brazil in 1980, by invitation of Pedro Sampaio, Professor of Neurosurgery and Head of the Division of Neurosurgery at
. Figure 16-3 Carlos Telles
History of stereotactic and functional neurosurgery in brazil
Pedro Ernesto Hospital, teaching hospital at the State University of Rio de Janeiro, Telles inaugurated the first multidisciplinary pain clinic in Brazil, which was probably one of his most important contributions to Brazilian stereotactic and functional neurosurgery. He was also the first to perform spinal cord stimulation for the treatment of pain in our country. Still in 1980, as will be mentioned later in this chapter, his role was fundamental for the foundation of the Brazilian Society for Stereotactic and Functional Neurosurgery. His main contribution to stereotactic and functional neurosurgery was the development of the technique he called combined solitary tractotomy (RF solitary tractotomy + RF lesioning of the subnucleus caudalis of the trigeminal nerve + C2 and C3 dorsal rhizotomy), which was performed through an open procedure, and designed to treat pain secondary to invasive cancer of the head and neck in the distribution of the V, IX and X cranial nerves and cervical region [16–18]. This technique was performed in more than 50 patients, providing excellent pain relief in 80% of them. Though never published, it was presented in three important meetings: 14th Brazilian Congress of Neurosurgery (Belo Horizonte, Brazil, 1982) [16], 20th Latin American Congress of Neurosurgery (Sa˜o Paulo, Brazil, 1983) [17], and Annual Meeting of the American Society for Stereotactic and Functional Neurosurgery (North Caroline, US, 1983) [18]. In 1991, he spent 6 months at the University of Freiburg, under Christopher Ostertag, doing his post-doctorate thesis (‘‘Stereotactic biopsy for the diagnosis of deep brain lesions’’). In 1992, after the retirement of Pedro Sampaio, he became Professor and Head of the Service of Neurosurgery at Pedro Ernesto Hospital, State University of Rio de Janeiro. Still very active, and having dedicated most of his life to the treatment of pain, he supervised the formation of more than 40 physicians in the area of pain, many of them illustrious pain specialists in our country. Other of his areas of expertise are
16
stereotactic biopsy of deep brain lesions, percutaneous cordotomy for cancer pain (> 800 cases), sympathetic procedures for the treatment of complex regional pain syndromes, DREZotomy, and trigeminal procedures (> 1,300 cases using RF retrogasserian rhizotomy, microvascular decompression, or balloon microcompression of the Gasserian ganglion). Carlos Telles was president of both the Brazilian Society for Stereotactic and Functional Neurosurgery (1988/1990) and Brazilian Society of Neurosurgery (1996/1998). Currently, there are some other colleagues in Rio de Janeiro with an impressive experience in stereotactic and functional neurosurgery, among them: Alexandre Amaral, Servidores Pu´blicos Estaduais Hospital (training with Manoel Teixeira); Ce´sar Fantezia Andrauss, Rio de Janeiro Neurological Institute (training with Gianni Temponi); Eduardo Carlos Barreto, Quinta D’Or Hospital (training with Carlos Telles and Mario Brock); Paulo L. C. Cruz, Santa Casa de Miserico´rdia do Rio de Janeiro Hospital (training with Manoel Teixeira); Ney Jose´ Monteiro, Lagoa Hospital (training with Renato Barbosa); Jose´ Augusto Nasser dos Santos, Esta´cio de Sa´ University (training at Columbia University); and Marcello Reis da Silva, Clementino Fraga Filho Hospital (training with Jean Claude Peragut). Eduardo Barreto was one of the founders of the Brazilian Chapter of the International Neuromodulation Society – Brazilian Neuromodulation Society (May 2007), occasion in which he was elected its first president. Ce´sar Andrauss, in 1997, helped by an engineer, built the TCA stereotactic apparatus, a modification of the Leksell frame; until 2005, when the last unity was built, 28 apparatuses had been manufactured by hand.
Sa˜o Paulo State Two of the pioneers in stereotactic and functional neurosurgery in Brazil are from Sa˜o Paulo State: Aloysio Mattos Pimenta and Jose´ Zaclis.
205
206
16
History of stereotactic and functional neurosurgery in brazil
Aloysio Mattos Pimenta (1912–1987) (> Figure 16-4) graduated from the University of Sa˜o Paulo (Universidade de Sa˜o Paulo – USP) Medical School in 1935. During his internship, in 1935, he divided his time between the Surgical Clinic, headed by Alves Lima; Neurological Clinic, headed by Enjolras Vampre´ (the pioneer of Neurology in Sa˜o Paulo), both at Santa Casa de Miserico´rdia de Sa˜o Paulo Hospital, at that time the teaching hospital at the USP Medical School; and Juqueri Hospital, a psychiatric institution. We assume that, since Carlos Gama, pioneer of neurosurgery in Sa˜o Paulo and Brazil, was the only neurosurgeon of the Neurological Clinic at the time, Mattos Pimenta accompanied him also. In 1936, he was part of the delegation of the USP Medical School that visited Argentina, spending most of his time with Manuel Balado, then one of the most prominent South American neurosurgeons. The combination of these different experiences, we believe, enabled Mattos Pimenta to start his neurosurgical practice [2,19]. In 1936, the same year Egas Moniz published the initial results of prefrontal leucotomy, Mattos . Figure 16-4 Aloysio Mattos Pimenta
Pimenta, oriented by Pacheco e Silva (Professor of Psychiatry at the University of Sa˜o Paulo Medical School), was one of the first in the world and the first in the Americas to perform the prefrontal leucotomy of Egas Moniz, which took place at Juqueri Hospital, in fact only a few days before Freeman and Watts performed their first psychosurgical procedure in the USA [2,19]. Until the late 1940s, assisted by the psychiatrist Mario Yahn and the neurosurgeon Afonso Sette Jr, Mattos Pimenta performed 279 operations of this type. The technique of Egas Moniz was used in the first 161 patients (before 1945), the technique proposed by Freeman and Watts in the 48 intermediate patients of their series, and the staged Freeman and Watts procedure (operation performed in three stages), as proposed by Mario Yahn, in the last 70 patients. The results obtained in the 279 patients were compiled in a book authored by Yahn, Mattos Pimenta and Sette Jr, and was prefaced by Egas Moniz and published in 1951 [20]. In August 1948, Egas Moniz chaired the First International Congress of Psychosurgery, which was held in Lisbon. During its closing session, the Brazilian delegation (Mattos Pimenta, Paulino Longo, Pacheco e Silva, Ma´rio Yahn, Anı´bal Silveira, E´lio Simo˜es, and Antoˆnio Carlos Barreto), led by Paulino Longo (head of the Neurological Division, Paulista Medical School, Federal University of Sa˜o Paulo) proposed the official nomination of Egas Moniz for the Nobel Prize considering his two great and universally accepted achievements, cerebral angiography and prefrontal leucotomy. The proposal was enthusiastically accepted and embodied by the members of the delegations of the 26 other countries that attended the meeting [2]. He had already been nominated for this award twice before, but in both occasions his nomination was declined. In 1949 Egas Moniz was at last awarded the Noble Prize in Physiology and Medicine for his invaluable contribution with the prefrontal leucotomy, no mention made, however, of his discovery of the cerebral angiography.
History of stereotactic and functional neurosurgery in brazil
At the beginning, without formal training, Mattos Pimenta performed only simpler procedures. In order to polish his neurosurgical formation, he spent 2 years (1938 and 1939) in Europe, most of the time with To¨nnis, in Berlin, but also with Zu¨lch, in Berlin (2 months), Fo¨erster, in Breslau (1 month), Olivercrona, in Stockholm (1 month), Busch, in Copenhagen, Leriche, in Strassbourg, and Clovis Vincent, in Paris. Due to the eclosion of World War II, however, he had to return to Brazil much earlier than he had planned. Always a restless creature, in January 1942 he went to the New York Neurological Institute, spending 4 months with Stookey and Putnam, and then moved on to the University of Michigan, where he stayed for 20 months under the supervision of Max Peet [2,19]. Back to Brazil, in 1944 he started working at the Neurological Clinic of Paulista Medical School, Federal University of Sa˜o Paulo, headed by Paulino Longo. In 1947, the Neurosurgical Division of this institution was officially inaugurated and Mattos Pimenta was invested its head, position that he held until his retirement (1982). In the 1960s he became Full Professor of Neurosurgery of Paulista Medical School, and in 1980 he started up the post-graduation course in neurosurgery (master degree and PhD), still the only one in the country [2,19]. Eventually he became involved with stereotactic surgery for Parkinson’s disease (PD). The date, though, is uncertain. We had the opportunity to analyze his curriculum vitae prepared in November, 1959 [19]. It is a very detailed CV, but no mention is made concerning any training in stereotactic surgery. Fernando Braga, who started his neurosurgical training under Mattos Pimenta in 1961, however, told us that his uncle, presenting with Parkinson’s disease, was operated on by his chief, who used the Riechert and Mundinger apparatus, in the same year he started his training. For this very reason, we inferred that Mattos Pimenta spent some time with Riechert and Mundinger in Freiburg in 1960, and after
16
returning to Brazil, started performing stereotactic surgery in 1961. Since then he operated on Parkinson’s disease with certain regularity for many years, but always as an appendage of his neurosurgical career. Neuro-oncology was always the apple of his eye. His associative activity was also intense. He was one of the twelve founders and the second president of Brazilian Society of Neurosurgery (1958/1959), presiding its second congress in 1959 (Campos do Jorda˜o, Sa˜o Paulo State) [2]. In 1965 he was president of the 11th Latin-America Congress of Neurosurgery, held in Sa˜o Paulo, having as official topics only stereotactic surgery and urgencies in neurosurgery [21]. Maybe even more important were his activities in the World Federation of Neurosurgical Societies: First vice-president (1973/1977), honorary president (elected in 1977), and president of its congress held in Sa˜o Paulo in 1977. Two other neurosurgeons performed stereotactic and functional neurosurgery at Sa˜o Paulo Hospital, the teaching hospital at Federal University of Sa˜o Paulo: Fernando Menezes Braga and Fernando Antonio Patriani Ferraz. Fernando Braga, once finished his training at Paulista Medical School, spent 1 year (1965/ 1966) in Edinburgh with Gillinghan, doing mostly general neurosurgery, but also some stereotactic and functional neurosurgery. Before coming back to Brazil, he purchased the McCaul apparatus, and in 1967, performed his first thalamotomy in a parkinsonian patient at Sa˜o Paulo Hospital. Along the years, mainly until 1970, he accumulated a significant experience with stereotactic treatment of movement disorders. In 1992, as Division Head and Professor of Neurosurgery, Braga entrusted Fernando Ferraz the task of creating the Stereotactic and Functional Neurosurgery Service at the Division of Neurosurgery at Federal University of Sa˜o Paulo. Ferraz, an ex-resident of the institution, already had some training in stereotactic and functional neurosurgery, which he had acquired
207
208
16
History of stereotactic and functional neurosurgery in brazil
from Mattos Pimenta and Braga. Before starting this new task, though, he went to UCLA School of Medicine and spent a short time with Antonio De Salles. A CT/MR-compatible stereotactic apparatus was acquired (Micromar stereotactic system), and then, still in 1992, the service, almost completely dedicated to the treatment of movement disorders, was finally inaugurated. Ferraz did a great job. The service went remarkably well until the beginning of the 2000s, when, due to a number of institutional problems, it faded away. It is very unfortunate, since Federal University of Sa˜o Paulo is considered one of the centers of excellence for neurosurgical training in Brazil. In this regard, it is probably wise mentioning that, just as in the case of Mattos Pimenta, stereotactic and functional neurosurgery was only an appendage of the neurosurgical practice of Braga and Ferraz. None of them was actually a functional neurosurgeon by ‘‘birth’’ and this can have made a real difference. The most important center of training in stereotactic and functional neurosurgery in Sa˜o Paulo is, in fact, the USP Medical School, which started by the hands of Jose´ Zaclis, another pioneer of the field. Graduated from the University of Sa˜o Paulo (USP) Medical School in 1944, Jose´ Zaclis (1917–1983) (> Figure 16-5a) was formed neurosurgeon by Rolando Tenuto, Professor of Neurosurgery and Head of the Division of Neurosurgery at the Clinic Hospital, teaching hospital at the USP Medical School [22]. The main assistant to Tenuto, in 1947 he was given the task of organizing the Neuroradiology Service of the institution [2,22]. From this time on, he dedicated almost exclusively to this area, performing ventriculography, pneumoencephalography, myelography, and cerebral angiography. He was the first to describe the technique of cerebral pan-angiography by means of injection of contrast medium into a single artery while simultaneously producing transient intrathoracic hypertension [23,24]. His service was the first
entirely dedicated to neuroradiology in Brazil. Since at that time stereotactic surgery relied basically on ventriculography, in which he was an expert, after a trip to Germany, Russia, and USA, places where he became acquainted with different stereotactic apparatuses, and with the help of the engineer Koralek, he built, in 1961, the HC stereotactic apparatus (HC stands for Hospital das Clı´nicas) (> Figure 16-5b) [25]. The equipment was set in a room of the neuroradiology unit completely dedicated to stereotactic surgery, where he used part of his time to perform stereotactic procedures for movement disorders and pain. The operations were carried out with the patient in a sitting position. This was the second Brazilian made stereotactic frame, in fact the first really elaborated and arc-centered apparatus. Only two of these frames were produced: one for the USP Medical School, and another for the Rio de Janeiro Neurological Institute. From 1971 to 1974, he was Head of the Division of Neurosurgery, and Professor at the USP Medical School. After Zaclis came Raul Marino Jr and Manoel Teixeira. Raul Marino Jr (> Figure 16-6) graduated from the USP Medical School in 1960, and after finishing his neurosurgical training in the same institution in 1964, under Rolando Tenuto, he went to North America to brush up his formation. In 1965, while in Boston, he divided his time between the Lahey Clinic, accompanying James Poppen and Charles Fager, who introduced him to stereotaxis and surgery for movement disorders, and the laboratory of Walle Nauta at the Massachusetts Institute of Technology (MIT), where he was for the first time really exposed to the limbic system, basal ganglia, and functional neuroanatomy. In 1966, still in Boston, he moved to the Massachusetts General Hospital (MGH), and was extremely fortunate to have as simultaneous supervisors professionals like Thomas Ballantine Jr (psychiatric surgery), William Sweet (pain surgery), and Raymond Kjellberg (surgery for movement disorders, brachytherapy, and
History of stereotactic and functional neurosurgery in brazil
16
. Figure 16-5 (a) Jose´ Zaclis (b) HC stereotactic apparatus, built by Zaclis and Koralek in 1961
. Figure 16-6 Raul Marino Jr
cyclotron radiosurgery for pituitary tumors and other endocrine diseases), all luminaries of the field of stereotactic and functional neurosurgery. Learning of Marino’s previous work on limbic system with Nauta at MIT, Ballantine Jr invited him to join his team and work on the cingulotomy project for mental illnesses and pain [26]. In 1967, this team published their pioneer paper on stereotactic anterior cingulotomy for psychiatric
disorders and pain [27]. According to Marino Jr (personal communication), in the 1965 publication of Ballantine’s group [28] on the same subject, the free-hand technique was used instead of the stereotactic technique and the number of cases was still too small. Marino spent the year of 1967 in Montreal. At Montreal Neurological Institute (MNI), McGill University, where he spent most of his time, he learned electroencephalography, electrocorticography, and the Wada test with Peter Gloor and epilepsy surgery with Theodore Rasmussen. There he had the opportunity to meet Wilder Penfield and Herbert Jasper. His free time was spent at the Notre Dame Hospital, University of Montreal, accompanying the stereotactic procedures of Claude Bertrand and the transsphenoidal surgeries of Jules Hardy, who later on became his brother-in-law. In 1968 he returned to the USA for an extra year at the National Institutes of Health (Bethesda, Maryland) supervised by Paul MacLean (lab work), the father of the limbic system, and Van Buren, at the time in charge of the stereotactic program [26]. After 4 years abroad, Marino finally returned to Brazil. In 1969, requested by Rolando Tenuto, his former chief, he resumed his work at the USP Medical School. Initially, he performed his stereotactic operations using the Zaclis’ HC apparatus. In 1971, after a substantial reform of
209
210
16
History of stereotactic and functional neurosurgery in brazil
the operating rooms and infirmaries of the Psychiatric Institute at the USP Medical School, Raul Marino inaugurated the first integrated, multidisciplinary Stereotactic and Functional Neurosurgery Service in Brazil, a service with 51 beds completely devoted to stereotactic and functional neurosurgery [26]. Well prepared as he was, he performed the most varied functional procedures, although his main interests, from the beginning until his retirement, always remained the same, that is, surgical treatment of epilepsy and psychiatric disorders, from which areas derived his most important contributions. Aiming to improve the electrophysiological investigation of refractory epilepsy, in 1972 he spent 6 months in Paris with Jean Talairach and Gabor Szikla, masters of the stereotactic electroencephalography technique [26]. Back to Brazil, a Talairach frame was acquired by his institution, finally replacing the Zaclis apparatus. A member of the American Branch of the International Society for Research in Stereoencephalotomy since its foundation (1968), in 1973, during the Sixth Symposium of the International Society in Tokyo, he became its member and was elected local Chairman of the next meeting (1977), which would be held in Sa˜o Paulo, Brazil, as a satellite meeting of that of the World Federation of Neurosurgical Societies (WFNS) [26]. Still during the Tokyo meeting, when was proposed a change in the name of the International Society, Marino suggested adding the term functional, which was accepted and the new designation became World Society for Stereotactic and Functional Neurosurgery. In 1977, during the Seventh Meeting of the World Society for Stereotactic and Functional Neurosurgery, Marino was elected Vice President of the Eighth Meeting of the World Society for Stereotactic and Functional Neurosurgery (the President elected was Jean Siegfried) and appointed member of the editorial board of the official journal of the society, Applied Neurophysiology (previously – 1938/1975 – called Confinia
Neurologica), later (1988) renamed as Stereotactic and Functional Neurosurgery, position that he held until his retirement in 2007 [26]. As a result of the 1977 meeting, the book Functional Neurosurgery, edited by Theodore Rasmussen and Raul Marino Jr, having as contributors an impressive number of outstanding functional neurosurgeons, was published in 1978 [29]. In his introductory chapter, Marino explains the reasons that led him to propose the term functional neurosurgery for this specialty, making a parallel with the bees behavior: ‘‘General neurosurgery tends to concentrate on the lesion, rather than on the symptoms, while functional neurosurgery focus on the symptoms, that is, on the abnormal functions, which are often hyperfunctional states that appear at a distance as a consequence of the primary lesion. Many specialties deal with the human brain, as many insects alight on the prairie flowers. However, only the bees know how to extract the honey. The bees alone are able to do that job and leave the flowers intact, without hurting or making them lose their freshness, allowing them to remain exactly as they were before. This is the hope and aim of functional neurosurgery’’ [29]. In 1980, Marino also played a very important role in the foundation of the Brazilian Society for Stereotactic and Functional Neurosurgery, and it was his close relationship with Blaine Nashold, who was present during this event, that enabled the Brazilian Society for Stereotactic and Functional Neurosurgery to be regarded as the Brazilian branch of World Society for Stereotactic and Functional Neurosurgery since its creation [26]. Raul Marino was elected the second President of Brazilian Society for Stereotactic and Functional Neurosurgery. Undoubtedly due to his great fondness for and expertise in psychiatric surgery, Marino also became Vice President of the International Society of Psychosurgery. In the 1980s (Raul Marino, personal communication), Marino was the first and the only (to the best of our knowledge) Brazilian neurosurgeon
History of stereotactic and functional neurosurgery in brazil
to use neural (dopaminergic cells) grafts in humans for the treatment of Parkinson’s disease. Autologous adrenal medullary grafts were attempted in three patients (Beneficeˆncia Portuguesa Hospital), but no benefit was achieved and this technique was abandoned. Homologous substantia nigra grafts derived from fetuses of legally approved abortions (pregnancy following rape) were tried in three other patients. However, soon afterwards these previously legal abortions became prohibited and the trials were terminated. In 1990, Marino became Full Professor of Neurosurgery and Chief of the Division of Neurosurgery at the USP Medical School, one of the most prestigious Brazilian schools of Medicine, position that he held until his retirement in 2007 [26]. During his years at the University (1971–2007) he helped to form a large number of neurosurgeons. In 1993, after spending some time at the Temple University (Philadelphia, USA), he inaugurated the second LINAC Radiosurgery Service in the country at Beneficeˆncia Portuguesa Hospital (Sa˜o Paulo). The first X-Knife service, although quite rudimentary, had already been installed by Jack Beraha at Oswaldo Cruz Hospital (Sa˜o Paulo) in 1983, having being closed in 1989. His most important contributions were in the areas of epilepsy surgery and psychosurgery [27,29–37]. As already mentioned, as part of Ballantine’s team, he played a very important role in the development of stereotactic anterior cingulotomy for the treatment of psychiatric disorders (major depression and obsessive-compulsive disorder) and pain, a technique he later spread in our country [27]. Concerning epilepsy surgery, his contributions were various, standing out: he was the first in Brazil to implement and diffuse the need of a multidisciplinary team to better evaluate the surgical candidates; his team was pioneer in establishing the electrophysiological parameter to determine the extension of the
16
callosotomy, that is, the disruption of bilateral synchrony of spike and wave discharges [30]; finally, Marino Jr was the first to propose and use the stereotactic callosotomy technique [31]. Raul Marino Jr was the first Brazilian neurosurgeon with formal training in stereotactic and functional neurosurgery and also the first to introduce the concept of a multidisciplinary stereotactic and functional neurosurgery team in our country. From the beginning, his career was almost completely devoted to the field of stereotactic and functional neurosurgery. Besides, he is by far the most renowned and prominent Brazilian functional neurosurgeon. For all these reasons, Marino Jr should be the one to be regarded as the real father of stereotactic and functional neurosurgery in Brazil. Graduated from the USP Medical School in 1972, Manoel Jacobsen Teixeira (> Figure 16-7a) finished his residence training in the same institution in 1976. During his training, he was for the first time exposed to stereotactic surgery by Jose´ Zaclis. Still in 1976, he spent 3 months at Goiaˆnia Neurological Institute under Luiz Fernando Martins, a pupil of Wilhelm Umbach, deepening his knowledge and practice in stereotactic and functional neurosurgery. In 1977 he went to Europe, where he stayed for more than 2 years. Initially, Teixeira spent 1 year and a half in Edinburgh with Edward Hitchcock, his most beloved chief and icon. Afterwards, he moved on to Zurich, spending 3 months with Jean Siegfried, and then, finally, to Freiburg, where he accompanied Mundinger during 4 months. During his outstanding career, he has visited a number of other functional neurosurgery services around the world. Teixeira started working at the Division of Neurosurgery of the Department of Neurology at the USP Medical School in 1984. In 1985, he obtained his master degree, and in 1990, his PhD. From 1997 to 2004, he was Technical Director of the Health Service of the Division of Functional Neurosurgery of the Institute of
211
212
16
History of stereotactic and functional neurosurgery in brazil
. Figure 16-7 (a) Manoel Teixeira (b) Micromar stereotactic system, model TM-03B
Psychiatry at the USP Medical School, and since 2004 he became Technical Director of the Division of Functional Neurosurgery. In March 2007, after the retirement of Raul Marino, there was an examination for full professorship in neurosurgery at the USP Medical School, and Teixeira was approved in first place, becoming Full Professor of the discipline and Head of the Division of Neurosurgery. His academic career has been quite profitable. Teixeira and his group have published a large number of papers, book chapters and books, and presented innumerable oral abstracts and posters both in national and international meetings, the vast majority in the area of pain. The multidisciplinary Pain Clinic of USP Medical School, which he runs, is most probably the busiest pain center in the country. Along the years he has formed a large number of functional neurosurgeons and pain specialists not only from Brazil but also from many other South American countries. Manoel Teixeira is a very versatile functional neurosurgeon, being experienced in all areas of the field, including brachytherapy and radiosurgery; his expertise in epilepsy and psychiatric disorders surgery, however, is less impressive. But the apple of his eye is really pain, anything in this area, from pathophysiology to the most
refined surgical technique. In fact, he is regarded by many as a living encyclopedia in this field. Also, he is certainly one of the most experienced active pain surgeons in the world. Accordingly, his most important contributions were on the surgical treatment of pain. A disciple of Edward Hitchcock, he probably is the still active neurosurgeon with the largest experience in medullary stereotactic trigeminal nucleotractotomy [38,39], a technique developed by Hitchcock in 1970 [40]. Hitchcock and Teixeira, in 1987, described the technique of pontine stereotactic trigeminal nucleotractotomy, indicated in cases of deafferentation extending through a long distance in the trigeminal nuclear complex, when the pain can not be adequately alleviated by medullary tractotomy alone [38,41]. In 2005, his group showed in rats that the motor cortex presents an antinociceptive function even under physiological circumstances and that motor cortex stimulation-induced analgesia is mediated by the opioid system [42]; for this research, Erich Fonoff received the Young Investigator Award from the World Society for Stereotactic and Functional Neurosurgery [43]. Teixeira et al., in 2007, demonstrated that the deafferentation pain associated with actinic brachial plexopathy and trigeminal neuropathy, usually unresponsive to
History of stereotactic and functional neurosurgery in brazil
various therapeutic approaches, both conservative and surgical, can be successfully relieved by, respectively, cervical DREZotomy and stereotactic trigeminal nucleotractotomy [44]. Also in 2007, using transcranial magnetic stimulation to map the motor cortex, Teixeira et al demonstrated that the response to motor cortex stimulation in patients harboring neuropathic pain secondary to brachial plexus avulsion could be predicted by the size of the field of representation of the deafferented limb in the motor cortex, being the result good or excellent in patients with larger fields and poor in those with smaller or undetectable fields; this research was awarded the NeuPSIG Poster Prize (category: clinical studies) during the Second International Congress on Neuropathic Pain (Berlin, June, 2007) [45]. Recently, his group proposed a technique for stereotactic disconnection of the hypothalamic hamartoma and so preventing gelastic and secondary seizures and aggressive behavior [46], and phrenic nerve stimulation for the treatment of refractory singultus [47]. Other important but unpublished contributions from his group were the use of computer-assisted stereotactic fenestration of aqueductal cysts for the treatment of hydrocephalus [48] and the stereotactic intracavitary instillation of amphotericin B through an indwelling catheter to treat deep intracerebral paracoccidioidomycosis cysts [49]. In our opinion, these are the most relevant contributions from Teixeira and his group, among others [50–58]. A great amount of the development of stereotactic and functional neurosurgery in our country was due to the integrated actions of Teixeira and the neurosurgical industry. In 1985, Teixeira and the engineer Antoˆnio Martos, CEO of Micromar, built the first commercially available Brazilian stereotactic apparatus, the Micromar stereotactic system, model TM-01B (Teixeira & Martos), a modification of that of Hitchcock’s. Continuous refinement led to the production of new versions of the first frame,
16
the models TM-02B, in 1991, and TM-03B, in 1996 (> Figure 16-7b). In 1998, Teixeira, Martos and the physicist Armando Alaminos produced the first linear accelerator- compatible radiosurgery system in Brazil, the Micromar radiosurgery system. Manoel Teixeira, a very well known and admired functional neurosurgeon in our country, was president of the Brazilian Society for Stereotactic and Functional Neurosurgery for two terms: 1990/1992 and 1998/2000. Other important names of stereotactic and functional neurosurgery in Sa˜o Paulo State are Jorge Roberto Pagura, Jose´ Oswaldo Oliveira Jr, Cla´udio Fernandes Correˆa, Arthur Cukiert, Jack Beraha, and Nilton Luis Latuf. Jorge Roberto Pagura (> Figure 16-8) graduated from the ABC Medical School in 1974. Once finished his neurosurgical residence (1975/1978) at Nove de Julho Hospital, in Sa˜o Paulo, under the supervision of Gilberto Machado de Almeida (also Professor of Neurosurgery and Head of the Division of Neurosurgery at the USP Medical School), he went to . Figure 16-8 Jorge Pagura
213
214
16
History of stereotactic and functional neurosurgery in brazil
Germany, where he spent 15 months (1978/1979) with Wolfhard Winkelmu¨ller at the University of Hannover. In 1980 Pagura returned to Hannover and stayed with Majid Samii for 1 month; at that time, Peter Janetta was also there showing his technique of neurovascular decompression for trigeminal neuralgia. This procedure became the one he is most fond of. In 1983 he obtained his PhD with the thesis entitled ‘‘Percutaneous radiofrequency spinal rhizotomy’’ at the Paulista Medical School, Federal University of Sa˜o Paulo, which was later published [59]. In the same year, he went to Duke University where he spent 40 days with Blaine Nashold learning the DREZ procedure. In 1996 Pagura became Full Professor and Head of the Division of Neurosurgery at ABC Medical School. His main areas of expertise are trigeminal neurovascular decompression [60], percutaneous radiofrequency glossopharyngeal rhizotomy [61], DREZotomy for pain and spasticity, and neuronavigation. His most important contributions to the literature are exactly in the two first topics above mentioned. His paper on glossopharyngeal rhizotomy, a classic in this issue, is still frequently cited [61]. Jorge Pagura was president of the Brazilian Society for Stereotactic and Functional Neurosurgery in the biennium 2002/2004. Also worth mentioning was his role in the creation of the Brazilian Society for the Study of Pain (Sociedade Brasileira para o Estudo da Dor – SBED), the Brazilian branch of the International Association for the Study of Pain. The SBED was created by initiative of Pagura and the neurologist Moacir Schnapp. Pagura was elected the second president of this society. Jose´ Oswaldo Oliveira Jr (> Figure 16-9a) graduated from the USP Medical School in 1977, institution where he also did his residence training from 1978 to 1982. In 1984 he obtained his master degree, and in 1986, his PhD, both from the USP Medical School. His training in functional neurosurgery was done in Edinburgh,
under the supervision of Edward Hitchcock, in 1985 (7 months), and in Paris, supervised by Gabor Szikla, in 1986 (9 months). A very inventive person, in the 1980s, time in which the radiofrequency generators (RFG) in Brazil had to be imported, with technical assistance, he built his own equipment, not for commercial purposes, but for his personal use. Two other versions of this RFG were later developed. The main drawback of his equipment was the lack of temperature monitoring, which had to be inferred based on the amperage. Placement of the DBS electrode is, to say the least, a quite boring and time consuming maneuver, as everyone involved with this technique knows. In 2005, Oliveira Jr developed the DBS electrode placer (> Figure 16-9b), later commercialized by Micromar. This very simple equipment makes DBS electrode placement an easy and fast procedure. Besides, he also developed a head and electrode holder for percutaneous cordotomy, a kit for transsphenoidal chemical hypophysectomy, and a number of electrode kits. Oliveira Jr heads the Department of Functional Neurosurgery, Antalgic Therapy and Palliative Care at AC Camargo Hospital, the most important and antique institution for cancer treatment in Brazil, and the Functional Neurosurgery Service at the Division of Neurosurgery at Servidores Pu´blicos Estaduais Hospital, both in Sa˜o Paulo. Although also a very versatile functional neurosurgeon, his main area of interest and expertise has always been pain surgery. During his very active career, he has formed more than 20 functional neurosurgeons. Cla´udio Fernandes Correˆa (> Figure 16-10) graduated from the Federal University of Triaˆngulo Mineiro in 1978 and did his neurosurgical residence (1979/1982) at Beneficeˆncia Portuguesa Hospital, in Santos, Sa˜o Paulo State. Correˆa obtained both his master degree (spinal cord stimulation for refractory pain, 1994) and PhD (brachytherapy for brain tumors, 1999) at the Paulista Medical School, Federal University
History of stereotactic and functional neurosurgery in brazil
16
. Figure 16-9 (a) Jose´ Oswaldo Oliveira Jr (b) Micromar electrode placer
. Figure 16-10 Cla´udio Correˆa
of Sa˜o Paulo. His formation in stereotactic and functional neurosurgery was initially supervised by Carlos Telles (1986/1987), and later by Manoel Teixeira. At present he heads the Stereotactic and Functional Neurosurgery Service at Nove de Julho Hospital, in Sa˜o Paulo. His main areas of expertise are pain surgery and stereotactic neurooncology. So far Correˆa has performed more than 1,300 microcompressions of the gasserian ganglion [62], which probably represents one of the largest series in the world. His experience in brachytherapy is also impressive [52], seconding only that of
Teixeira in our country. Correˆa was president of the Brazilian Society for Stereotactic and Functional Neurosurgery in the biennium 1996/1998 and the founder of SIMBIDOR, a biennial multidisciplinary pain meeting, with already eight versions (first version: 1994), undoubtedly one of the most important Brazilian pain meetings. Arthur Cukiert graduated from the USP Medical School in 1986, completing his neurosurgical training at the same institution in 1990 (1987/1990). He then moved to Montreal Neurological Institute, where he trained in epilepsy and stereotactic surgery with Andre Olivier (1991/1992). Since he returned to Brazil Cukiert has headed the Division of Neurology and Neurosurgery at Brigadeiro Hospital, in Sa˜o Paulo. In 1996, he obtained his PhD at the USP. Almost all his practice has been dedicated to epilepsy and pituitary tumor surgery. So far he has performed more than 1,200 epilepsy surgeries. Cukiert has published extensively on all aspects of epilepsy [63–68]. His main interests in this area are the normal and abnormal function of the corpus callosum and the development of clinical and flowchart paradigms for the medical and surgical treatment of epilepsy. He is one of the authors of the first epidemiological paper on
215
216
16
History of stereotactic and functional neurosurgery in brazil
epilepsy released in Brazil [63]. Cukiert was also the first in the country to perform deep brain stimulation for the treatment of refractory epilepsy (anterior thalamic nucleus). He is author of five books: three on epilepsy and two on neuroendocrinology. Jack Beraha graduated from the Pontifical Catholic University of Sa˜o Paulo Medical School in 1978. He did his neurosurgical training at Oswaldo Cruz Hospital (Sa˜o Paulo), supervised by Darcy Vellutini, going later to Geneva, where he accompanied the services of A. Werner and J. Berney. He then moved on to the University of Genebra, where he obtained his PhD. As he became interested in radiosurgery he went to the University of Valencia Medical School, staying with Barcia-Salorio in 1982/1983, and afterwards to Buenos Aires and Charlottesville, accompanying Ladislau Steiner for a period of 21=2 months. In 1987, Beraha became Full Professor at the Pontifical Catholic University of Sa˜o Paulo Medical School. Until 1983 radiosurgery was performed by using gamma-knife, telecobaltotherapy, proton beam, and betatron. During his stay with Barcia-Salorio, he learned about telecobaltotherapy radiosurgery. Using a principle similar to that of Barcia-Salorio, Beraha, helped by the physicists Luiz Scaff and Dirceu Vizeu, modified a 4 MeV linear accelerator (LINAC) at Oswaldo Cruz Hospital, in Sa˜o Paulo, to make it radiosurgery-compatible, and in July of 1983 Beraha et al performed the first LINAC radiosurgery in the world in a patient harboring a craniopharyngioma [69]. From then until 1989 he treated more than 100 patients presenting with intracranial tumors or arteriovenous malformations [69,70]. One of these patients, who had a vestibular schwannoma that apparently failed to respond to radiosurgery, was subsequently operated on by a very skillful and famous neurosurgeon. After the operation, this patient presented significant complications, and the neurosurgeon, seemingly in an attempt to get rid of the possible
implications of the poor result, told the patient’s family that the sequel was consequence of the prior radiosurgery. Beraha was sued by the patient’s family, but even having been proved innocent, as expected, the number of referrals declined drastically, and his radiosurgery service was finally closed in 1989. It must be remembered, however, that the world owes Jack Beraha the clinical introduction of LINAC radiosurgery. Also worth mentioning is the fact that this story was told us not by Beraha, but by a number of other colleagues. Nilton Luis Latuf (1937–2005) graduated from the USP Ribeira˜o Preto Medical School in 1963, where he also attended his residence in neurosurgery (1964/1966). In 1967 Latuf went to Paris and spent 1 year with Gerard Guiot at Foch Hospital [71]. Back to Brazil and to Ribeira˜o Preto, in Sa˜o Paulo State, he was the first functional neurosurgeon in Brazil to establish his practice in the countryside. His first functional procedure, a thalamotomy in a parkinsonian patient, was performed by using the Guiot frame, and it took place at Santa Casa de Miserico´rdia Hospital, in Ribeira˜o Preto, in October 1968. In 1969 Latuf inaugurated the Division of Neurology and Neurosurgery in the same institution, heading it for 25 years. He formed 48 residents, and performed more than eight thousand neurological surgeries throughout his life [71], a significant amount of these constituted by stereotactic and functional neurosurgery procedures. Latuf was a member of the executive council of the Brazilian Society of Neurosurgery for 21 years, as well as the South American Representative and Honorary President of the World Association of Lebanese Neurosurgeons for many years. A number of other colleagues in Sa˜o Paulo State have established a busy practice in the field of stereotactic and functional neurosurgery, among them: Erich Fonoff (Sa˜o Paulo), Nilton Lara (Sa˜o Paulo), Salomon Benabou (Sa˜o Paulo), Valter Cescato (Sa˜o Paulo), Antoˆnio de Almeida
History of stereotactic and functional neurosurgery in brazil
(Sa˜o Paulo), Evandro de Souza (Sa˜o Paulo), Soraya Cecı´lio (Sa˜o Paulo), Jose´ Cla´udio Marinho da No´brega (Sa˜o Paulo), Luis Augusto Rogano (Sa˜o Paulo), Edson Amaˆncio (Santos), Arthur Ungaretti Jr (Santo Andre´), Joa˜o Alberto Assirati Jr (Ribeira˜o Preto), Marcus Colbachini (Ribeira˜o Preto), Sebastia˜o Carlos da Silva Jr (Sa˜o Jose´ do Rio Preto), Kleber Duarte (Sa˜o Jose´ do Rio Preto), Carlos Tadeu Parisi Oliveira (Braganc¸a Paulista), Guilherme Cantatore Castro (Braganc¸a Paulista), Jose´ Paulo Montemor (Campinas), and Edmur Piza Filho (Jundiaı´).
Minas Gerais State The pioneer of stereotactic and functional neurosurgery in Minas Gerais State was Jose´ de Arau´jo Barros. Jose´ de Arau´jo Barros (> Figure 16-11a) graduated from the Federal University of Minas Gerais Medical School and in 1953 he finished his initial training in neurosurgery with Francisco Rocha, a disciple of Herbert Olivecrona and one of the pioneers of neurosurgery in Minas Gerais State. In 1954 he moved on to the University of Illinois in Chicago, spending 2 years under Percival Bailey. During his fellowship, he went several times to New York to learn stereotactic surgery with Irving Cooper. Back to Belo Horizonte, capital of Minas Gerais State, Barros constructed his own frame (> Figure 16-11b), based on that of Cooper, and started performing stereotactic surgery in 1960. This was the first stereotactic apparatus built in Brazil. Along the years he performed 248 procedures in 186 patients harboring movement disorders, mainly Parkinson’s disease, but also dystonia, hemiballism, and choreoathetosis. In 1961 he became Full Professor of Neurology at the Minas Gerais Medical Sciences School, position that he occupied until his retirement in 1998. A very active neurosurgeon at Sa˜o Jose´ Hospital, the teaching hospital of the Medical
16
Sciences School, Barros formed 58 neurosurgeons. One of his ex-residents, Lourenc¸o de Freitas Neto, replicated his frame and was the pioneer in the stereotactic field in Espı´rito Santo State. At present, five are the best known and active functional neurosurgeons in Minas Gerais: Jose´ Maurı´cio Siqueira, Gerva´sio Teles Cardoso de Carvalho, Sebastia˜o Nataniel Silva Gusma˜o, Rodrigo de Mattos Labruna, and Gilberto de Almeida Fonseca Filho. Jose´ Maurı´cio Siqueira graduated from the Federal University of Minas Gerais Medical School in 1976 and attended neurosurgical residence at Felı´cio Roxo Hospital (1977/1979), in Belo Horizonte, previously one of the two teaching hospitals at the Federal University of Minas Gerais. In 1981 he went to the University of Toronto, spending 1 year as a fellow under Ronald Tasker. Consecutively, he accompanied Claude Bertrand at the Notre Dame Hospital for 2 months and Patrick Kelly, at the New York State University at Buffalo, for 1 month. Some years later he also spent a short period (1 month) with Erik Backlund at the Karolinska Institute in Stockholm. Siqueira made two major contributions to the field of stereotactic and functional neurosurgery. In 1985 Siqueira was the first in the literature to describe the technique of open bulbospinal trigeminal nucleotomy through radiofrequency lesioning for the treatment of refractory facial deafferentation pain [72]. This paper, in which he reported the 1-year follow-up results obtained in two patients operated on in 1983 (complete abolition of pain was achieved in both patients), had been previously presented as a poster at the IX Meeting of the World Society for Stereotactic and Functional Neurosurgery in Toronto, in 1985, chaired by Ronald Tasker. In 1986, Blaine Nashold et al published an abstract in Neurosurgery describing a technique identical to that of Siqueira, claiming that this ‘‘new surgery’’ had
217
218
16
History of stereotactic and functional neurosurgery in brazil
. Figure 16-11 (a) Jose´ de Arau´jo Barros (b) Stereotactic frame built by Barros
been developed by his group [73]. Actually, Nashold has never mentioned Siqueira’s pioneer paper in any of his publications. In 1987, in a letter sent to the editor of Neurosurgery, Schvarcz contested the idea that Nashold was the original author of the technique and stated that Hitchcock and himself, in 1972, were the first to report that technique, using the stereotactic approach, and mentioned Siqueira’s poster, who had performed the procedure through an open operation [74]. Schvarcz, however, failed to mention that the very first to perform this procedure by employing the stereotactic technique was Edward Hitchcock alone in 1970 [40]. In 1997, during the 11th Congress of the World Federation of Neurosurgical Societies, held in Amsterdam, Tasker introduced Siqueira to John Gorecki, an ex-resident of Nashold’s, and told him about the pioneer paper of Siqueira. Afterwards, Siqueira sent Gorecki a reprint of his paper, as well as the certificate of his poster presentation of 1985 at the World Society for Stereotactic and Functional Neurosurgery Meeting in Toronto. In 2004 Gorecki, in his chapter on DREZ and brainstem ablative procedures for Youmans Neurological Surgery fifth edition, finally gave Siqueira the deserved credit, reputing his report in this issue as the first published in the literature [75].
During his fellowship at the University of Toronto, Siqueira helped to review Tasker’s experience on unilateral Vim thalamotomy for the treatment of severe drug-resistant parkinsonian tremor. The conclusions derived from this study, published in 1983, when very few was being said about the surgical treatment of Parkinson’s disease, were very instrumental for the renascence of interest in this modality of treatment [76]. Currently, Siqueira heads the Services of Functional and Epilepsy Surgery of the Division of Neurology and Neurosurgery at Felı´cio Roxo Hospital. Gerva´sio Teles Cardoso de Carvalho graduated from the Minas Gerais Medical Sciences School in 1981. He finished his residence training at Santa Casa de Miserico´rdia de Belo Horizonte Hospital in 1985 (1982/1985), and spent 1 year (1986) with Carlos Telles at Pedro Ernesto Hospital, the teaching hospital at the State University of Rio de Janeiro. In 1987, following the lead of Siqueira, he went to the University of Toronto and spent 1 year under Ronald Tasker (1987/ 1988). In 1995 he obtained his master degree from the Federal University of Sa˜o Paulo. Carvalho has a vast experience in the surgical treatment of pain [77–79], psychiatric and movement disorders [80], and coordinates the
History of stereotactic and functional neurosurgery in brazil
Stereotactic and Functional Neurosurgery Service at the Division of Neurosurgery at Santa Casa de Miserico´rdia de Belo Horizonte Hospital. During the time he was a fellow under Tasker, he was part of the team that demonstrated that, in patients suffering from cord injury pain, the intermittent neuralgic and evoked elements of neuropathic pain differed from the steady causalgic/dysesthetic element in their response to destructive and modulatory procedures (the latter, contrarily to the first two elements, which are frequently relieved by interruption of the pain pathways, responds better to stimulation of the spinal cord or the brain), suggesting the presence of distinct mechanisms underlying these different modalities of pain [77–79]. Sebastia˜o Nataniel Silva Gusma˜o graduated from the Minas Gerais Medical Sciences School in 1973 and did his neurosurgical residence in the same institution (1974/1976). Three years later, to polish his initial formation, he spent 1 year at the Department of Neurosurgery at the National Hospital for Nervous Diseases, in London, and a shorter period at the Louis Pasteur University Medical School, in Strasbourg. Gusma˜o obtained his master degree from the Federal University of Minas Gerais in 1984, and his PhD from the Federal University of Sa˜o Paulo in 1993. His training in stereotactic and functional neurosurgery was done under Carlos Telles, in Rio de Janeiro, and also, for a shorter period, at Stockholm University. Gusma˜o was president of the Brazilian Society for Stereotactic and Functional Neurosurgery in the biennium 1992/1994 and is currently Full Professor of Neurosurgery and Head of the Division of Neurosurgery at the Federal University of Minas Gerais. Pain surgery constitutes his main area of expertise in the field of stereotactic and functional neurosurgery, and from this area derived his most important contribution. Gusma˜o was the first in the literature to propose the use of intraoperative computed tomography to guide percutaneous
16
radiofrequency trigeminal rhizotomy, greatly facilitating the identification of the foramen ovale [81]. Two other very experienced colleagues from Minas Gerais are Rodrigo de Mattos Labruna, trained by Teixeira, and Gilberto de Almeida Fonseca Filho, trained by Teixeira, Dieckmann and Ostertag. Their main areas of expertise are pain and spasticity surgery. Labruna has also a significant experience in surgery for psychiatric disorders.
Espı´rito do Santo State The pioneer of stereotactic and functional neurosurgery in Espı´rito Santo State was Lourenc¸o de Freitas Neto. Freitas Neto did his residence training in Belo Horizonte under Jose´ de Arau´jo Barros, who introduced him to stereotactic surgery. He then moved back to Vito´ria, the capital of Espı´rito Santo, in 1968, where, after replicating the frame developed by Barros, he started performing stereotactic surgery for movement disorders. Until 1984 he performed 32 procedures, basically for Parkinson’s disease, occasion in which he abandoned functional neurosurgery due to the lack of candidates for surgery. Se´rgio Otoni, after a 6-month training with Edward Hitchcock, returned to Vito´ria in 1978, but never really dedicated to stereotactic and functional neurosurgery. Pedro Menezes, trained by Teixeira, worked part of the time in Sa˜o Paulo and part in Vito´ria. Between 2000 and 2006, while in Vito´ria, he performed a significant number of functional procedures. Only recently (2003), though, Walter Fagundes-Pereyra for the first time established a really active and ongoing practice in functional neurosurgery in Espı´rito Santo (Vito´ria). Fagundes-Pereyra was first exposed to stereotactic and functional neurosurgery by Gerva´sio Carvalho, during his neurosurgical residence at Santa Casa de Miserico´rdia de Belo Horizonte Hospital. In 2001 he obtained his master degree from the same institution. He
219
220
16
History of stereotactic and functional neurosurgery in brazil
then moved on to Lille, where, during 1 year and a half, he refined his training in stereotactic and functional neurosurgery with Serge Blond. In 2007 he obtained his PhD from the Federal University of Sa˜o Paulo.
Midwest Region The Midwest region is composed by the states of Goia´s, Mato Grosso, and Mato Grosso do Sul and the Federal District.
Goia´s State The pioneer of stereotactic and functional neurosurgery in the Midwest region was Luiz Fernando Martins (> Figure 16-12), undoubtedly one of the most renowned and important functional neurosurgeons in Brazil and South America. Martins graduated from the Federal University of Goia´s Medical School in 1971. His internship
. Figure 16-12 Luiz Fernando Martins
(1971), the last year (sixth) of Medical School in the Brazilian university system, was spent with Renato Barbosa, at Lagoa Hospital, in Rio de Janeiro. Barbosa was the one who first introduced him to stereotaxis. Since then, and certainly forever, Martins became a victim of a platonic love for this marvelous field of neurosurgery. In 1972 he went to the University of Berlin, where he stayed for a period of 4 years, doing his neurosurgical residence under Wilhelm Umbach, at the time one of the most important functional neurosurgeons in Germany. In 1975 Martins obtained his PhD from the University of Berlin, and in 1976, back to Brazil, he started his practice in Goiaˆnia, the capital of Goia´s, joining the team of the Goiaˆnia Neurological Institute, a private hospital founded the year before, then and now one of the most important neurosurgical institutions in the country. On 13 February 1976, using a Riechert and Mundinger apparatus, he performed his first stereotactic procedure, lesioning the zona incerta in a patient presenting with Parkinson’s disease. In 1982 he returned to Europe, spending more than 1 year: 3 months with Brock, in Berlin; 2 months with Dieckmann, in Go¨ttingen; 7 months with Siegfried, in Zurich; and 2 months with Mundinger, in Freiburg. A team player, Martins has always proclaimed the need of a multidisciplinary group to evaluate surgical candidates in all areas of functional neurosurgery. Though performed in an irregular basis since 1976, only in 1989 the epilepsy surgery program was inaugurated at Goiaˆnia Neurological Institute. Thanks to this government-supported program, for which implementation were instrumental Martins, the neurosurgeon Orlando Arruda, and the epileptologist Paulo Ragazzo (trained at the Montreal Neurological Institute), 1,782 patients, not only from the Midwest, but also from the North and Northeast of the country could be operated on in this institution. These operations should be credited to the following team of neurosurgeons: Luiz Fernando Martins, Orlando Arruda (already
History of stereotactic and functional neurosurgery in brazil
dead), Valter da Costa, Joa˜o Arruda, Joaquim da Costa, and more recently, Henrique Lobo, Marcelo Martins, and Osvaldo Vilela Filho. In 1999 Martins innovated again, creating the first radiosurgery service in the Midwest region, and one of the first in the country. This service is a joint project of Goiaˆnia Neurological Institute and CEBROM (Brazilian Center of Radiotherapy, Oncology and Mastology), a private outpatient clinic dedicated to oncology, radiotherapy and chemotherapy and owner of the LINAC used in the procedures. Since then, three hundred and 16 radiosurgeries have been performed. Both epilepsy surgery and radiosurgery programs are clear examples of a multidisciplinary team work. Martins is a highly experienced neurosurgeon in all areas of stereotactic and functional neurosurgery, but spasticity. Throughout his outstanding career he has performed, besides those aforementioned, 402 surgeries for movement disorders (mainly Parkinson’s disease); 714 for pain (mainly trigeminal neuralgia); 468 for psychiatric disorders (mainly aggressiveness and OCD), probably one of the largest series in the world at present; 186 stereotactic biopsies; and 36 stereotactic-guided craniotomies. Martins was one of the founders of the Brazilian Society for Stereotactic and Functional Neurosurgery, being its president for two terms (1984/1986 and 2000/2002), and of the recently created Brazilian Society of Radiosurgery (2007), which has as president the functional neurosurgeon Salomon Benabou, and is now president of the SLANFE (Latin American Society for Stereotactic and Functional Neurosurgery). Though author of just a few papers [82–88], he made two significant scientific contributions: the determination of the radiographic stereotactic coordinates of the foramen ovale [82], and the correlation between the position and size of medial thalamic lesions and the degree of pain relief obtained by patients presenting cancer pain (an autopsy study) [83,84]. Another important contribution, although not published, was his
16
proposal of combined amygdalotomy and anterior capsulotomy for the treatment of refractory aggressiveness (1985). But as he says, ‘‘I am not a writer, but a surgeon. I like to operate and also to lecture, this is how I have chosen to pass along my experience.’’ A superb lecturer, he played a very important role spreading the knowledge concerning surgical treatment of psychiatric disorders, movement disorders, and epilepsy across the country. Unfortunately, the lack of habit of writing papers is a characteristic shared by the vast majority of Brazilian neurosurgeons, which seems to be a cultural deficiency. How much more could someone like him have contributed to the field by writing down such a huge experience. There are six other functional neurosurgeons in Goia´s State, five of them from Goiaˆnia and one from Ana´polis (Nivaldo Evangelista Teles). Four of those from Goiaˆnia are part of the team of the Goiaˆnia Neurological Institute (Joaquim da Costa, Joa˜o Batista Arruda, Vladimir Arruda Zaccariotti, and Osvaldo Vilela Filho), and the other one has his Service at Santa Helena Hospital (Hamilton Ayres da Silva). Hamilton da Silva, once finished his 5-year residence in neurosurgery (1987/1991) at Servidores Pu´blicos Estaduais Hospital (Sa˜o Paulo), decided to dedicate his career to stereotactic and functional neurosurgery and spent 4 years (1992/ 1995) with Teixeira at USP Medical School and 6 months (1996) with Ronald Tasker at University of Toronto. Eventually he returned to Goiaˆnia and established his practice at Santa Helena Hospital and Goiaˆnia General Hospital. Most of his practice in stereotactic and functional neurosurgery has been devoted to the surgical treatment of pain. Joaquim da Costa had his initial contact with stereotaxis as a resident under Jose´ de Arau´jo Barros, the pioneer of stereotactic and functional neurosurgery in Minas Gerais State, but his true master was Martins, with whom he works until now. In 1983 he spent a short period with Falk Oppel in Berlin learning epilepsy surgery.
221
222
16
History of stereotactic and functional neurosurgery in brazil
Like Martins, he is a very experienced functional neurosurgeon. Joa˜o Arruda and Vladimir Zaccariotti did their neurosurgical residence at Goiaˆnia Neurological Institute. Both had Martins as their first master in stereotactic and functional neurosurgery. Arruda completed his training with Blaine Nashold at Duke University (1 year), and Zaccariotti, with Falk Oppel (17 months), in Bielefeld, and Antonio De Salles, at UCLA (7 months). Arruda works both at Goiaˆnia Neurological Institute and Arau´jo Jorge Hospital (Cancer Hospital), where he runs the Neurosurgical Service, devoted mainly to neurooncology, treatment of pain and radiosurgery, areas in which he has a large experience. Besides, he is also a very experienced epilepsy surgeon. Zaccariotti works with him in both institutions. Osvaldo Vilela Filho (> Figure 16-13) graduated from the Federal University of Goia´s (Universidade Federal de Goia´s – UFG) Medical School in 1984. During his last 2 years of Medical School, he accompanied many functional procedures performed by Luiz Fernando Martins at Goiaˆnia Neurological Institute, one of his . Figure 16-13 Osvaldo Vilela Filho
personal icons. This simple fact would later influence his whole career. His neurosurgical residence (1985/1988) was done at the Brası´lia District Hospital, in Brası´lia, being chief resident of neurosurgery in 1988. Unfortunately, stereotactic and functional neurosurgery was not performed in this institution at that time. Worried about the difficulty to approach some lesions of difficult location, mainly those deeply placed, Vilela Filho (1987) developed a technique that allowed the transposition of the lesions seen on CT-scan to the scalp, greatly easing the surgical procedure [89]. Later, before his stereotactic training, he used the same technique to perform CT- or ultrasound-guided free-hand biopsy of intracranial lesions [90]. In February of 1991 he went to the Toronto Western Hospital, University of Toronto, to do a fellowship in stereotactic and functional neurosurgery under Ronald Tasker, staying until 1992. While at UT he reviewed all patients with neuropathic pain submitted to DBS by Ronald Tasker. This research led to a number of important publications [91–95]. Tasker became another of his icons, with whom he established a close friendship that remains until now, someone that he gladly reputes as his Canadian father. In 1993 he returned to Canada, this time to the University of Western Ontario, in London, to learn epilepsy surgery under the supervision of Andrew Parrent and John Girvin. Later he went to other centers of excellence in stereotactic and functional neurosurgery for shortperiod visits, like those of Roy Bakay, at the time at Emory University (2000); Andres Lozano, at University of Toronto (2000); and Ali Rezai, at Cleveland Clinic (2004). In 2006 he obtained his PhD from the Paulista Medical School, Federal University of Sa˜o Paulo. Back to his hometown, Vilela Filho inaugurated and headed the Goiaˆnia Brain Institute, devoted to neurosurgery and neurology, in general, but particularly to stereotactic and functional neurosurgery. Initially located at the Goiaˆnia Orthopedic Institute (1993/2001), it
History of stereotactic and functional neurosurgery in brazil
was later moved to Lucio Rebelo Hospital (2001/ 2003). In 1994 he underwent a competitive examination for neurosurgeon of the UFG Medical School, achieving the highest score, and since 1995 he has worked as an Invited Professor of Neurosurgery in the Department of Surgery of the same institution. After 3 years of continuous insistence, the Clinic Hospital, teaching hospital of the UFG Medical School, finally acquired the necessary equipment and in 1998 he created the Stereotactic and Functional Neurosurgery Service at the Division of Neurosurgery at the Clinic Hospital, which he has chaired since then. This was the first university and public hospital in the Midwest region of the country to offer stereotactic and functional neurosurgery. In 2001 Vilela Filho inaugurated the Laboratory of Experimental Stereotaxis at the UFG Medical School, which he still runs. In 2002 he became a consultant in neurosurgery and neurology to the Council of Healthcare Advisors, which was presided by the Nobel Laureate James Watson (dead in 2004), and in 2003 he became part of the team at the Goiaˆnia Neurological Institute. Since 2006 he has worked as an Invited Professor of the Department of Medicine at the Catholic University of Goia´s, becoming Associate Professor of Neurosciences after a competitive examination in 2007, once more achieving the highest score. Vilela Filho has had a very intense associative activity. Still as a student, he became a member of the Society for the Defense of Natural Resources (1977), a society made up almost exclusively by professors and researchers of the local universities (Federal and Catholic Universities of Goia´s), serving as its first secretary (1980/1982) and president (1982/1984). In 1997 he founded the Young Neurosurgeons Committee of the Brazilian Society of Neurosurgery, becoming its coordinator for three consecutive terms (1997/2002), and in the same year he was the chairman of the Scientific Committee of the VII National Congress of the Brazilian Academy of Neurosurgery. Again in 1997 he was one of the
16
founders and the first president of the Goia´s Association for the Study of Pain, position that he held for three terms (1997/2002), period in which he presided two important international congresses on pain, the I and II INDOR (International Congress on Pain). In 1999 he was also one of the founders and the first president of the Goia´s Society of Neurosurgery (1999/2002), presiding in 2001 its first congress, the first joint meeting Brazil & Canada on neurosurgery. For two consecutive terms he served as member of the International Relations Committee of the Brazilian Society of Neurosurgery (2000/2004), and in 2002 he was elected member of the Executive Council of this society (2002/2008). In 2004 he chaired the Scientific Committee of the XXV Brazilian Congress of Neurosurgery, organized by the Brazilian Society of Neurosurgery. Vilela Filho has been a true soldier of the Brazilian Society for Stereotactic and Functional Neurosurgery, having served as its secretary (2000/2002), vice-president (2004/2006), and president (2006/2008). In 2007 he presided the Eighth Congress of the Brazilian Society for Stereotactic and Functional Neurosurgery & First International Joint Meeting on Stereotactic and Functional Neurosurgery, undoubtedly the most important meeting organized by the Brazilian society. During this event was founded the Brazilian Neuromodulation Society, the Brazilian branch of the International Neuromodulation Society, and Vilela Filho was elected its vicepresident. In 2008 he became secretary of the Latin American Society for Stereotactic and Functional Neurosurgery (SLANFE). He is also member of a number of other societies, including the World and American Societies for Stereotactic and Functional Neurosurgery, International Neuromodulation Society, Brazilian Society for the Study of Pain, and American Association of Neurological Surgeons (1995/2002). Vilela Filho is member of the editorial board of three international journals (Neuromodulation; Brazilian Contemporary Neurosurgery, the
223
224
16
History of stereotactic and functional neurosurgery in brazil
Brazilian supplement of Surgical Neurology; and Neurotarget, the official journal of the Latin American Society for Stereotactic and Functional Neurosurgery) and six Brazilian journals (Neurocirurgia Contemporaˆnea Brasileira; Jornal Brasileiro de Neurocirurgia; Dor – 2002/2004; Simbidor; Jovem Me´dico; and Pra´tica Hospitalar), and was Associate Editor of the SBN Newsletter, the bulletin of the Brazilian Society of Neurosurgery (2002/2004). Throughout his career, Vilela Filho and his team have made numerous original scientific contributions, besides those aforementioned. Reviewing Tasker’s 16 patients with neuropathic pain of spinal cord origin submitted to DBS (thalamic ventrocaudal nucleus – VC and periventricular gray matter – PVG), he identified four patients with complete cord injury. PVGDBS was performed in three of these patients to treat the evoked and intermittent components of their pain, being unsuccessful in all them. VC-DBS, however, was successful in three out of four patients. Based on these results he contested previously proposed hypotheses, according which VC-DBS produces pain relief by inhibition of dorsal horn nociceptive neurons through a variety of pathways, which could not be at work in patients presenting complete cord transection [96,97]. Alternatively, he proposed that VC-DBS induces pain relief by inhibition of medial thalamus nociceptive neurons through a polysynaptic pathway, as follows: VC stimulation » somatosensory cortex excitation (» motor cortex activation) » anterior putamen excitation » peptidergic (substance P) activation of GPi/ SNR » medial thalamus inhibition [97]. Vilela Filho later proposed that this polysynaptic pathway would be operative even under physiological circumstances and could work as a modulatory pain center, which led him to name it as prosencephalo-mesencephalic modulatory circuit [98,99]. He also suggested that the interruption of this circuit or of its activating pathway, which he supposed to be the anterior spinothalamic
tract, by determining overactivation of the medial pain system, would be the pathophysiological substrate of the steady dysesthetic component of neuropathic pain [98,99]. According to his hypothesis, stimulation of various structures such as the peripheral nerves, spinal cord, medial lemniscus, VC, internal capsule and motor cortex induces pain relief by activation of this circuit [100,101]. Still in the field of pain surgery, Vilela Filho introduced the CT-guided percutaneous technique to perform punctuate midline myelotomy [102]; proposed the use of doppler-scan for localization of the occipital artery, which lies just lateral to the greater occipital nerve, greatly facilitating the realization of percutaneous radiofrequency occipital neurotomy and occipital nerve stimulation [103]; and was the first to propose the use of MRI for postoperative evaluation of percutaneous cordotomy [104]. It is well known that functional neuroimaging studies may frequently show abnormalities in patients harboring psychiatric disorders and that these abnormalities tend to disappear after successful surgical treatment. Vilela Filho, however, was the first to propose the use of functional studies to determine the best target to treat each patient. The idea was to surgically deactivate the areas shown to be hyperactive in the functional studies [105]. Some other studies performed by other authors, based on the principle suggested by Vilela Filho, led to the discovery of Brodmann area 25 as a target for depression [106] and of the posteromedial hypothalamus as a target for cluster headache [107]. Vilela Filho and Souza, in 1996 [108], and again in 1998 [109], originally proposed that Tourette syndrome is the clinical manifestation of the hyperactivity of the globus pallidus externus and prefrontal area. Based on this hypothesis, Vilela Filho et al, in 2004, were the first to perform bilateral GPe-DBS for the treatment of Tourette syndrome refractory to conservative management [110]. Four patients have been operated on so far. The excellent results obtained
History of stereotactic and functional neurosurgery in brazil
lend support to the presented hypothesis [111– 113]. In 2001, Vilela Filho proposed a modification in the technique of stereotactic subcaudate tractotomy, using landmarks seen on CT-scan instead of those seen in the skull X-ray, as it was usually performed [114]. Vilela Filho and Silva current series of 54 patients harboring Parkinson’s disease submitted to unilateral subthalamotomy is one of the largest in the literature. They have shown that STN lesioning is a highly effective and safe procedure [115–117]. A possible complication of this procedure is hemiballism, which occurred in 11% of the patients. These authors demonstrated that damage to the dorsolateral territory of STN and concomitant sparing of the zona incerta seem to be essential for the development of hemiballism, and that the presence of intraoperative stimulation-induced dyskinesia and, possibly, levodopainduced dyskinesia apparently are significant risk factors for the development of this complication [117,118]. In 2001 Vilela Filho et al reported the case of a patient with unilateral essential tremor whose tremor completely disappeared after a stroke restricted to the contralateral posterior putamen, leading him to originally propose a role for the basal ganglia in the genesis of essential tremor, possibly through the hyperactivity of the posterior putamen [119]. To test this hypothesis, they used an animal model of essential tremor (harmaline-induced tremor) and showed that the harmaline-induced tremor could be greatly reduced by an electrolytic lesion placed in the ipsilateral posterior striatum of the rat, giving further support to the proposed hypothesis [120]. Vilela Filho et al., in 1997, introduced an apparently low-risk and accurate technique for CT-guided free-hand percutaneous biopsy of spinal cord tumors [121]. In 2001 Vilela Fiho and his colleagues developed a new product that did not increase the inherent MRI distortion and could be seen in the main MRI sequences used for stereotactic surgery (T1-weighted,
16
T2-weighted and inversion recovery). This product, however, was never patented or commercialized [122]. Vilela Filho and Carneiro Filho, in 2002, were the first to introduce the gamma probe in neurosurgery, which was initially used for intraoperative detection of brain tumors and to ascertain its complete removal [123]. In 2006 the same authors again pioneered using the gamma probe for intraoperative detection of epileptogenic areas during epilepsy surgery, a technique still in investigation that seems particularly helpful in non-mesial temporal lobe epilepsy [124]. Finally, Vilela Filho and Carneiro Jr developed an MRI-based frameless technique to determine the stereotactic coordinates. A very precise technique, it allows to perform sort of an image fusion (CT and MRI) without the necessity of any software other than those of the MRI and CT scanners [125].
Mato Grosso State The pioneer of stereotactic and functional neurosurgery in Mato Grosso State was Jony Soares Ramos. Ramos did his residence training at Lagoa Hospital, in Rio de Janeiro. The Head of the Neurosurgery Service, Ney Monteiro, who supervised his residence, was one of the disciples of Renato Barbosa, the pioneer of stereotactic and functional neurosurgery in Brazil. During the last year of his training (2001), he completed his formation in pain surgery accompanying Carlos Telles at Pedro Ernesto Hospital, the teaching hospital at the State University of Rio de Janeiro. In 2002 he started his practice in Cuiaba´, the capital of Mato Grosso, initially using an antique Riechert and Mundinger apparatus, which he had bought from Renato Barbosa, already retired for a long time, and later he acquired the Micromar stereotactic system. From the beginning to 2004, he worked with Bruno Silveira, who trained with Teixeira, but Silveira stopped functional neurosurgery in 2004, and Ramos carried on by
225
226
16
History of stereotactic and functional neurosurgery in brazil
himself. Already performing stereotactic procedures for movement disorders, pain and biopsy for some years, in 2007, oriented by Martins, Ramos performed his first operation for the treatment of a psychiatric disorder. He is still the only active functional neurosurgeon in the state.
Mato Grosso do Sul State Two general neurosurgeons have recently started performing stereotactic and functional neurosurgery in Campo Grande, capital of Mato Grosso do Sul State. Cesar Nicolatte did his residence in neurosurgery at the Beneficeˆncia Portuguesa Hospital (1996/2000), in Sa˜o Paulo, where he learned stereotactic and functional neurosurgery with Paulo Dorsa, arriving in Campo Grande in 2001. Joaquim Oliveira Vieira Jr did his neurosurgical training at USP Medical School (1995/ 1999), accompanying the functional procedures performed by Teixeira, Cescato, and Oliveira Jr. After his residency, he did his PhD at the Brigadeiro Hospital, in Sa˜o Paulo, oriented by Arthur Cukiert, and started his practice in Campo Grande in 2003. Nicolatte and Vieira Jr, working together, have recently started to perform stereotactic and functional neurosurgery procedures, so far restricted to percutaneous procedures for pain, stereotactic biopsy and stereotactic-guided craniotomy.
Federal District The Federal District, located in Goia´s State, houses Brası´lia, the capital of the country. It seems unbelievable that stereotactic and functional neurosurgery was inaugurated in Brası´lia only in the 1990s, but there are some explanations for that. First, Brası´lia is a very young city, having been founded only in 1960, when it became the country’s capital (until then Rio de Janeiro was the Brazilian capital). Second, Paulo Mello, the
pioneer of neurosurgery in Brası´lia (1961), went to England a few years after its inauguration. Mello spent 4 years (1964/1967) at the Newcastle Upon Tyne Hospitals, becoming acquainted with stereotactic surgery by the hands of John Hankinson. Returning to Brası´lia, he was appointed Head of the Division of Neurosurgery at Brası´lia District Hospital (1971). In 1973 he went to Freiburg and spent 1 month with Mundinger. Still in the seventies the hospital acquired a Riechert and Mundinger apparatus and a radiofrequency generator, but at the time few stereotactic operations were being performed everywhere and Mello, already involved with vascular neurosurgery, felt his interest in stereotactic and functional neurosurgery almost completely vanish. Besides, the surgical room equipped with the system for orthogonal teleradiography has never been built. Third, Carlos Telles, who did his neurosurgical residence at Brası´lia District Hospital, was incited by Paulo Mello to dedicate himself to stereotactic and functional neurosurgery. Accepting the suggestion, he went to Germany to attend a fellowship in this field, where he spent almost 5 years. During his last year there (1980), Telles was visited by Pedro Sampaio, Professor of Neurosurgery and Head of the Division of Neurosurgery at Pedro Ernesto Hospital, the teaching hospital at Rio de Janeiro State University, who invited him to join his team. Telles accepted his invitation and established his practice in Rio de Janeiro in 1980. Brası´lia, then, remained without its so needed functional neurosurgeon. In 1994 the Division of Neurosurgery at ´ Brasılia District Hospital was headed by Carlos Silve´rio de Almeida, who had attended a fellowship in England. Almeida was strongly determined to unearth the old Riechert and Mundinger apparatus and put it at work, and so he did. Orienting his 4th year neurosurgical resident Luiz Cla´udio Modesto Pereira, they transposed the lesion seen on the CT-scan to the stereotactic skull radiography and finally performed the first stereotactic
History of stereotactic and functional neurosurgery in brazil
procedure in Brası´lia, the drainage of a deeply placed brain abscess. Luiz Cla´udio Modesto Pereira, excited with this first step, decided to dedicate to the field and became the pioneer in stereotactic and functional neurosurgery in the Federal District. In 1995 he visited the services of Christopher Ostertag, in Freiburg (2 months); Robert Goodman, in New York (1 month); and Manoel Teixeira, in Sa˜o Paulo (1 month). In 1996 he spent 1 month with Rees Cosgrove, in Boston, and 15 days with Ronald Tasker, in Toronto. From 1995 to 1998 his practice in stereotactic and functional neurosurgery was constituted by sporadic procedures and restricted to anesthetic blockades for pain, stereotactic biopsy, stereotactic-guided surgery and spinal cord stimulation. Even though, Pereira was apparently the first in Brazil to perform stereotactic fibrinolysis with rTPA to treat intracerebral hematomas (1995). In 1999 he finally went to Toronto, spending 1 year under Andres Lozano at the University of Toronto, and since his return to Brası´lia (2000) he has established a full and busy practice in stereotactic and functional neurosurgery [126,127]. The second and last functional neurosurgeon in the Federal District so far is Tiago Freitas, who, like Pereira, was a resident at the Brası´lia District Hospital (2002/2005). Graduated from the Federal University of Goia´s Medical School (2001), during his last year of neurosurgical training he intermittently accompanied Vilela Filho at Goiaˆnia Neurological Institute. In 2006 Freitas went to USP Medical School, accompanying Manoel Teixeira for 3 months, and then to the Cleveland Clinic for a 7-month fellowship under Ali Rezai. Still in the same year he started his practice in Brası´lia.
South Region The South region is made up by the states of Parana´, Rio Grande do Sul and Santa Catarina.
16
The pioneer of stereotactic and functional neurosurgery in the South was Renato de Muggiati, from Parana´.
Parana´ State Renato de Muggiati (1932–1996) (> Figure 16-14) graduated from the Federal University of Parana´ in 1955. During his last 2 years at Medical School, he had the opportunity to accompany the neurosurgeon Jose´ Portugal Pinto, nephew of Jose´ Ribe Portugal, who had recently arrived in Curitiba (state capital) after 6 years at Cleveland Clinic under Gardner, inaugurating neurosurgery in this state. As an assistant to Pinto, he helped him to perform a large number of prefrontal lobotomies at Adauto Botelho Hospital. Finished his graduation, he continued at that institution until 1959, occasion in which he went to Freiburg, spending 2 years under Riechert. A Riechert and Mundinger apparatus was the most important piece of his luggage on his trip back to Curitiba. In 1961 he performed the first stereotactic procedure in Parana´ State, a
. Figure 16-14 Renato de Muggiati
227
228
16
History of stereotactic and functional neurosurgery in brazil
thalamotomy in a parkinsonian patient [128]. At the beginning, he built his own electrodes using the spokes from bike wheels. Still in 1961 he was invited to create and head the Division of Neurosurgery at the Clinic Hospital, the teaching hospital of the Federal University of Parana´, in fact the first university neurosurgical service in Parana´. In 1962 he inaugurated the Neurosurgery Service at Santa Casa de Miserico´rdia de Curitiba Hospital, which later became one of the teaching hospitals of the Pontifical Catholic University of Parana´ [128]. Despite being a general neurosurgeon, stereotactic surgery always remained as one of the most important activities in his outstanding career, along which he performed 304 stereotactic procedures, mainly for Parkinson’s disease [129]. The oldest functional neurosurgeon in Parana´ today is Alceu Correia, who trained with Wolfhard Winkelmu¨ller at the University of Hannover for 5 years (1971/1975). Correia has a vast experience in the surgical treatment of pain and was president of the Brazilian Society for Stereotactic and Functional Neurosurgery in the biennium 1994/1996. At present there are eleven other colleagues in Parana´ performing functional neurosurgery. However, stereotactic and functional neurosurgery constitutes the main area of activity of only four of them: Alexandre Novicki Francisco, Murilo Sousa Meneses, Danel Benzecry de Almeida, and Helve´rcio Fernando Polsaque Alves. Alexandre Novicki Francisco graduated from the Federal University of Parana´ in 1992. After finishing his neurosurgical residence at Santa Casa de Miserico´rdia de Curitiba Hospital in 1996, he went to the USP Medical School, spending 6 months with Teixeira (1997). The next step was the University of Toronto, where he did a 1-year fellowship under Ronald Tasker and Andres Lozano (1997/1998). Since his return to Curitiba in 1998 he has coordinated the Stereotactic and Functional Service at the Division of Neurosurgery at Cajuru Hospital, a
teaching hospital at the Pontifical Catholic University of Parana´. His main areas of expertise are the surgical treatment of pain [95,130], spasticity and movement disorders. Francisco was secretary of the Brazilian Society for Stereotactic and Functional Neurosurgery (2006/2008), and is now vice-president (president-elect) of the same society and secretary of the Brazilian Neuromodulation Society, the Brazilian branch of the International Neuromodulation Society (INS). Murilo Sousa Meneses graduated from the Federal University of Parana´ in 1980. He then went to France for a 6-year residence in neurosurgery. During his last year (1987), at Lariboisiere Hospital (Paris) under Claude Thurel, Meneses got most of his training in stereotactic and functional neurosurgery. In 1987 he obtained his master degree, and in 1990 his PhD, both from the University of Picardie, in Amiens. In 1988, back to Curitiba, he became Assistant Professor of Anatomy (neuroanatomy) at the Federal University of Parana´; he is now Associate Professor of the same discipline. His initial formation in stereotactic and functional neurosurgery was later complemented with short-period (1 month) visits to other centers of excellence in the field, such as Mayo Clinic, with Patrick Kelly (1991); Saint Anne Hospital (Paris), with Chudkiewicz (1993); University of South California, with Michael Appuzzo (1994); and in Bordeaux, with Alain Rougier (1998). The areas in which he is most experienced are epilepsy [131–133] and movement disorders [134] surgery. Meneses is the author of three books, one in Parkinson’s disease (two editions), and two in neuroanatomy (one of them with two editions). Daniel Benzecry de Almeida graduated from the Federal University of Para´ in 1990. His neurosurgical residence was done at Paulista Medical School, Federal University of Sa˜o Paulo (1991/ 1994). In 1995 he started his training in stereotactic and functional neurosurgery, spending 1 year with Teixeira at USP Medical School (1995/1996), and after that, another year with
History of stereotactic and functional neurosurgery in brazil
Oliveira Jr at AC Camargo Hospital (1996/1997). In 1998 he went to Lyon for a 3-month fellowship under Marc Sindou. Later he visited Richard North at Johns Hopkins for a short period (2 weeks). Almeida dedicates most of his practice to pain surgery [135,136]. In 2007 he obtained his master degree from the Federal University of Sa˜o Paulo. Helve´rcio Fernando Polsaque Alves graduated from the State University of Maringa´ in 1997. Once finished his residency at Santa Casa de Miserico´rdia de Ribeira˜o Preto Hospital (2001), where he was first exposed to stereotactic surgery by Nilton Latuf and Marcus Colbachini, he spent 15 months at AC Camargo Hospital under Oliveira Jr. In 2003 he went to Maringa´, a beautiful and relatively small town in the interior of Parana´, where he established his busy practice in stereotactic and functional neurosurgery, consisting predominantly of surgical treatment of pain and movement disorders. The other colleagues performing stereotactic and functional neurosurgery in Parana´ to a lesser extent are: Ce´sar Vinicius Grande, Sonival Caˆndido Hunhevicz, Sı´lvio Machado, and Samir Ale Bark, in Curitiba; Stenio H. de Souza, in Cascavel; and Pedro Garcia Lopes and Wander Miguel Tamburus, in Londrina.
Rio Grande do Sul State The pioneer of stereotactic and functional neurosurgery in Rio Grande do Sul was Paris Ferreira Souza. Paris Ferreira Souza (> Figure 16-15) graduated from the Federal University of Rio Grande do Sul in 1952. His neurosurgical training was done at USP Medical School, under Rolando Tenuto, from 1954 to 1958. Willing to polish his neurosurgical technique, Souza went to Mainz, accompanying Schu¨rmann for 2 years (1961/1963). Before leaving Germany, he decided to go to Freiburg, where he spent 2 months with Riechert and Mundinger to learn stereotactic
16
. Figure 16-15 Paris Ferreira Souza
surgery and purchased a Riechert and Mundinger apparatus. Back to Brazil, he decided to return to Porto Alegre, capital of Rio Grande do Sul, and start his practice. Still in 1963 Souza performed his first stereotactic procedure to treat a patient with Parkinson’s disease at the Beneficeˆncia Portuguesa Hospital, in Porto Alegre, becoming the pioneer of stereotactic and functional neurosurgery in this state. Due to some problems of political nature, however, he had to leave Porto Alegre in 1964. In 1965 he restarted his stereotactic practice in Sa˜o Paulo, at Oswaldo Cruz Hospital, a private institution. A joint project of Oswaldo Cruz Hospital and Servidores Pu´blicos Estaduais Hospital, a public institution, allowed him to operate on patients of the public hospital in the private hospital. Until the beginning of the 1970s he operated on a lot of parkinsonian patients, occasion in which he abandoned functional neurosurgery due to the sharp decline in surgical indications and went on as a general neurosurgeon. Paris Souza is now 84 years old, the oldest Brazilian functional neurosurgeon alive. He spends his lifetime
229
230
16
History of stereotactic and functional neurosurgery in brazil
between Sa˜o Paulo and his ranch in Rio Grande do Sul, and still flies his ultralight plane like a young man (previously, he used to fly a Cessna 182, the skyline), despite the number of forced landings he has endured. The void left with the sudden exit of Paris Souza was filled by his prior classmate from the Medical School in Porto Alegre, Manoel Krimberg. Krimberg was formed neurosurgeon by Elyseu Paglioli at Sa˜o Jose´ Hospital. After 2 years in France with Talairach and Guiot (1963/1964), he returned to Porto Alegre, and in 1965 performed his first stereotactic procedure, a thalamotomy in a parkinsonian patient, using the Talairach apparatus. His stereotactic practice was going remarkably well until he associated with the anatomist Paolo Contu, an expert in glia. Apparently incited by Contu, as we were told (Ney Azambuja, personal communication), Krimberg started performing autologous sciatic nerve grafting to treat spinal cord injury patients, proclaiming spectacular results. Paglioli, his chief, became upset and in 1969 Krimberg left Sa˜o Jose´ Hospital and started working in two other places: Brigada Militar Hospital, in Porto Alegre, and Nossa Senhora das Grac¸as Hospital, in Canoas, in the countryside. At this point, we lost his track. We were recently told about his death years ago (uncertain date). Once again practically orphan of stereotactic and functional neurosurgery, the next in the row to restart the specialty in Porto Alegre was Telmo Tonetto Reis. Reis graduated from the Federal University of Rio Grande do Sul in 1964 and finished his neurosurgical training at Maia Filho Hospital, in Porto Alegre, in 1966. In 1967, unable to afford an air trip, and with one thousand dollars in his pocket, he endured a 15-day ship trip to London, where he spent 1 year at the National Hospital Queen Square under Valentine Logue. He then moved on to Edinburgh, where he accompanied John Gillingham and Edward Hitchcock for another year (1968). Part of this period (1 month and a
half) was spent in France with Guiot. Returning to Porto Alegre, he performed his first stereotactic procedure (thalamotomy) to treat a parkinsonian patient in 1969, by using the Guiot and Gillingham apparatus, at Moinho de Ventos Hospital. Still an active functional neurosurgeon, his main area of expertise is movement disorders. Graduated from the Federal University of Parana´ in 1974, Jose´ Vitor Pinto did his neurosurgical training at Maia Filho Hospital and Moinho de Ventos Hospital (1975/1976), in Porto Alegre, under Telmo Reis, who introduced him to stereotaxis. Enchanted with stereotactic and functional neurosurgery, Pinto decided to follow the steps of his chief. In 1977 he went to Edinburgh, spending 2 years under Edward Hitchcock and John Gillingham. In 1979, when Hitchcock moved to Birmingham, Pinto did the same, staying for another year. Returning to Porto Alegre with a Hitchcock apparatus and a Radionics radiofrequency generator in his luggage, he started his practice working with Telmo Reis at Moinho de Ventos Hospital, a partnership that still remains. In the 1980s, Pinto was probably the first to perform pallidotomy (Leksell’s target) for Parkinson’s disease in our country. His initial series of 30 cases was first presented in the XVIII Congress of the Brazilian Society of Neurosurgery in 1990. Pinto was president of the Brazilian Society for Stereotactic and Functional Neurosurgery in the biennium 1988/1990. Two other superb functional neurosurgeons started their practice in Porto Alegre before Pinto: Paulo Petry Oppitz and Ney Arthur Vilamil de Castro Azambuja. Paulo Oppitz graduated from the Catholic University of Pelotas in 1969. Some time after his 2-year neurosurgical training at Sa˜o Jose´ Hospital with Nelson Ferreira, in Porto Alegre (1970/ 1971), Oppitz went to Buenos Aires, spending 1 year with Jorge Schvarcz, and then to Edinburgh, accompanying Hitchcock for 6 months. Soon after his arrival in Edinburgh, he purchased the fifth Hitchcock apparatus sold in the world,
History of stereotactic and functional neurosurgery in brazil
which costed him five thousand pounds. In fact, Hitchcock even used Oppitz recently acquired frame in some of his operations. Returning to Porto Alegre, Oppitz started his practice at Cristo Redentor Hospital in 1976, and like almost all functional neurosurgeons of his state, his main area of expertise is the surgical treatment of movement disorders. Ney Azambuja graduated from the Catholic Medical School of Porto Alegre (now a federal foundation) in 1974, and like Oppitz, did his 3-year neurosurgical residence at Sa˜o Jose´ Hospital (1975/1977). In 1978 Azambuja went to Zurich, spending 1 year under Jean Siegfried. Before returning to Brazil, he visited Hitchcock for 1 month. Back to Porto Alegre, he started his practice at Sa˜o Jose´ Hospital in 1980. Later, Schvarcz introduced him to Oppitz, and since then they have developed an associated stereotactic practice, first at Cristo Redentor and Sa˜o Jose´ Hospitals, and later at Moinho de Ventos and Sa˜o Lucas Hospitals. Azambuja, in the beginning of the 1980s, was probably the first in Brazil to perform periaqueductal gray matter DBS for the treatment of pain. In 1985 he obtained his master degree from the Federal University of Rio Grande do Sul. As he became interested in epilepsy surgery, Azambuja moved from Sa˜o Jose´ Hospital to Sa˜o Lucas Hospital, the teaching hospital at the Pontifical Catholic University of Rio Grande do Sul, and undoubtedly one of the most renowned centers of epilepsy surgery in the country. Azambuja is now Associate Professor and regent of the discipline of neurosurgery at the Pontifical Catholic University of Rio Grande do Sul, as well as researcher of the Institute of Biomedical Research at the same institution. His main areas of expertise are the surgical treatment for epilepsy [137–139], movement disorders and pain [140], in particular trigeminal neuralgia. Other colleagues devoting part of their practice to stereotactic and functional neurosurgery in Rio Grande do Sul State are: Elyseu Paglioli Neto
16
(Porto Alegre, epilepsy surgery only), Jorge Bizzi (Porto Alegre), and Leonardo Frighetto (Passo Fundo, mainly radiosurgery).
Santa Catarina State Marcelo Neves Linhares was the pioneer of stereotactic and functional neurosurgery in Santa Catarina State, starting his practice in Floriano´polis, the state capital, in 1999. Curiously, he is still the only functional neurosurgeon in his state. Linhares graduated from the Federal University of Santa Catarina in 1992. After finishing his neurosurgical residence (1994/1997), he stayed at the USP Medical School under Teixeira for 5 months (1998), and then moved on to Toronto, doing a 1-year fellowship under Ronald Tasker and Andres Lozano (1998/1999) [141]. Linhares obtained his PhD from the USP Medical School in Ribeira˜o Preto, Sa˜o Paulo State, in 2005, and in 2007 he became Associate Professor of Neurosurgery at the Federal University of Santa Catarina.
Northeast Region The Northeast region is composed by the following states: Pernambuco, Ceara´, Bahia, Paraı´ba, Alagoas, Sergipe, Rio Grande do Norte, Maranha˜o, and Piauı´. There is a certain degree of uncertainty regarding the place, neurosurgeon and date that stereotactic and functional neurosurgery was inaugurated in the Northeast. We will try to clear this up along the next paragraphs. Manoel Caetano de Barros, the oldest Brazilian neurosurgeon still alive (sadly, Caetano de Barros died on 31 October 2008, sometime after we finished writing this chapter), was the pioneer of neurosurgery in the Northeast region of the country, starting his career in Recife, capital of Pernambuco, in 1947, initially as a self-taught neurosurgeon. Still in 1947 he went to Paris, spending 2 years with Clovis Vincent at Salpetrie`re
231
232
16
History of stereotactic and functional neurosurgery in brazil
Hospital. In 1950 he returned to Europe, this time to London, spending 1 year under McKissock at the National Hospital Queen Square. In 1958 Caetano de Barros became Full Professor of Neurosurgery and Head of the Department of Neurology and Neurosurgery at the Federal University of Pernambuco. In his service at Pedro II Hospital, at that time the teaching hospital at the Federal University of Pernambuco, there was a Talairach apparatus acquired in the 1960s. For this reason, it has been assumed that he was the pioneer also of stereotactic and functional neurosurgery in the Northeast region of the country. We tried to contact him, but he is already very old (94 years old) and unfortunately unable to answer our questions. His son, Alex Caetano de Barros, a very active neurosurgeon in Recife, was unaware if his father had any involvement with stereotactic and functional neurosurgery in the early years of his career. Many other colleagues were consulted, but none had the answer to our question. At last, after a tough search, we were able to come up with the name of the neurologist and clinical neurophysiologist Salustiano Gomes Lins, a colleague and friend of Caetano de Barros from the old days. Salustiano Lins did his training in clinical neurophysiology in Montreal, where he had the opportunity to observe a number of stereotactic procedures. Back to Recife, he incited Caetano de Barros to purchase a stereotactic frame and start performing functional neurosurgery. The stereotactic apparatus was acquired, but, according to Lins, Caetano de Barros never really started his stereotactic practice. Once cleared up the aforementioned issue, it seems to us that the pioneer of stereotactic and functional neurosurgery in the Northeast was Djacir Figueiredo, from Fortaleza, capital of Ceara´. Graduated from the Federal University of Ceara´ in 1955, Figueiredo was formed neurosurgeon by Elyseu Paglioli, in Porto Alegre (Rio Grande do Sul State), at Sa˜o Jose´ Hospital, one of the two first Brazilian neurosurgical schools. In 1964 he went to New York, spending 1 year under
Lawrence Pool at the Neurological Institute of Columbia University. During his time in the USA, Figueiredo visited Earl Walker at Johns Hopkins for 2 months, where he first became involved with stereotactic and functional neurosurgery. At that time Walker used the McKinney stereotactic instrument. Before returning to Fortaleza, Figueiredo purchased the improved version of the McKinney apparatus, and even not being a genuine functional neurosurgeon, he was the first to perform a stereotactic procedure in the Northeast. From 1965 to 1968 he operated on five parkinsonian patients, and then quit his stereotactic practice. With the vacuum left by Figueiredo, someone else wishing to dedicate to stereotactic and functional neurosurgery was needed. This space was occupied by Vicente de Paula Lobo, who worked in the team coordinated by Djacir Figueiredo. Vicente de Paula Lobo (1930–2007) graduated from the Federal University of Ceara´ in 1960 and finished his neurosurgical training at Rio de Janeiro Neurological Institute, under Jose´ Ribe Portugal, in 1963. During his residency he was first exposed to stereotactic and functional neurosurgery by Renato Barbosa. In 1972 Lobo returned to Rio de Janeiro to update his training in stereotactic and functional neurosurgery, again under Renato Barbosa, but at this time at the Lagoa Hospital, since Barbosa had already left the Neurological Institute. Returning to Fortaleza, after the acquisition of a Riechert and Mundinger apparatus, Lobo performed his first procedure for Parkinson’s disease at Fortaleza General Hospital in May 1973. Approximately in 1980, due to the pronounced declining in the surgical indication for Parkinson’s disease, Lobo also abandoned the field of stereotaxis. In 1985, Fla´vio Belmino Barbosa Evangelista, who did his residence training at the Rio Janeiro Neurological Institute under Gianni Temponi, one of the pioneers of stereotactic and functional neurosurgery in Rio de Janeiro and Brazil, started performing stereotactic surgery in Fortaleza, mainly for Parkinson’s disease and only in private clinic.
History of stereotactic and functional neurosurgery in brazil
The pioneer of stereotactic and functional neurosurgery in Pernambuco was Gla´ucio Veras. Veras trained with G. Dieckmann, in Go¨ttingen, from 1979 to 1985. Back to Brazil he stayed in Foz do Iguac¸u, Parana´, until 1989 and then moved on to Recife, capital of Pernambuco. In 1991 he inaugurated the Pain Clinic at the Cancer Hospital, and in 1993 he started performing stereotactic surgery in the same institution. His practice has been dedicated almost exclusively to pain surgery [142] and brachytherapy, having performed only a few procedures for other functional disorders. The most active functional neurosurgeon in this state, however, is Paulo Thadeu Brainer-Lima. Brainer-Lima graduated from the University of Pernambuco Medical School in 1989 and did his residence training at Restaurac¸a˜o Hospital, in Recife, from 1991 to 1994. In 1995 he went to the USP Medical School, spending 1 year and a half under Marino Jr. During this period he also had the privilege to accompany Teixeira, Cukiert, and Oliveira Jr. He then went to Oxford University (1998), staying with Tipu Aziz for 9 months. Returning to Recife, he started performing stereotactic and functional neurosurgery in 1999, and since then he has coordinated the Stereotactic and Functional Neurosurgery Service of the Division of Neurosurgery at Restaurac¸a˜o Hospital, the most important neurosurgical school in the Northeastern region. Brainer-Lima obtained his master degree in 1999, and his PhD in 2003, both from the same institution, the Federal University of Pernambuco. In September 2008, he became president of the Brazilian Society for Stereotactic and Functional Neurosurgery. The main foci of his career have been surgical treatment of pain, movement disorders and epilepsy [143–145]. Maria da Glo´ria S. Pabst was the pioneer of stereotactic and functional neurosurgery in Bahia State. Trained by Teixeira at the USP Medical School in 1991, she started her practice in Salvador (state capital) in 1992, performing her first procedure for Parkinson’s disease (helped by
16
Teixeira) in the same year. Due to a number of difficulties, however, her activities in the field have been restricted to stereotactic biopsy and stereotactic guided surgery since then. Stereotactic and functional neurosurgery in Bahia only flourished very recently (2006), though, with the arrival of Yuri Andrade-Souza in Salvador. Andrade-Souza did his neurosurgical training (1999/2002) at Sa˜o Jose´ Hospital, in Porto Alegre, under Nelson Ferreira (Ferreira, though a general neurosurgeon, was the first in Brazil to perform trigeminal retrogasserian rhizothomy by using radiofrequency thermocoagulation). In 2003 he went to the University of Toronto, spending 1 year under Andres Lozano at the Toronto Western Hospital, and 1 year and a half under Michael Schwartz at Sunnybrook Hospital. He then moved on to the University Hospital, University of Western Ontario, in London (Canada), staying 1 year under Andrew Parrent to learn epilepsy surgery. During his fellowships he published a number of significant papers [146–153]. Back to his hometown, Salvador, in 2006, Andrade-Souza has established a growing and active practice devoted to functional neurosurgery. The pioneer of stereotactic and functional neurosurgery in Paraı´ba was Valdir Delmiro Neves. Neves started his practice in the field in Joa˜o Pessoa, state capital, in 1997, after being trained by Teixeira. Some years later (2000) the colleague Ussaˆnio Mororo´ Meira [154], trained by Oliveira Jr, arrived in Joa˜o Pessoa. Both have a busy practice in the field of stereotactic and functional neurosurgery. Heider Lopes de Souza was the first to perform stereotactic and functional neurosurgery in Rio Grande do Norte. Souza did his neurosurgical training at Paulista Medical School, Federal University of Sa˜o Paulo, period in which he intermittently accompanied Teixeira at USP Medical School. He obtained his master degree in 1993, and his PhD (trigeminal neuralgia) in 2000, both from the Federal University of
233
234
16
History of stereotactic and functional neurosurgery in brazil
Sa˜o Paulo. In 1996 he started his practice in Natal, capital of the state. Se´rgio Adrian Fernandes Dantas, in 2002, was though the first neurosurgeon with formal training in stereotactic and functional neurosurgery to establish his practice in Natal. Dantas initially spent 1 year with Oliveira Jr at AC Camargo Hospital, in Sa˜o Paulo. He then went to Lille and spent two more years under Serge Blond [155]. The first neurosurgeon to perform a stereotactic procedure in Maranha˜o, more precisely in its capital, Sa˜o Luı´s, was Osmir de Ca´ssia Sampaio. Sampaio, in 1995, after a short period (2 months) with Teixeira, started performing stereotactic biopsy, but he never went beyond this. He is truly a general neurosurgeon, and just wanted to widen his surgical possibilities. The first ‘‘authentic’’ functional neurosurgeon in Maranha˜o was in fact Jose´ Roberto Pereira Guimara˜es, who started his practice in 1999 after a formal training with Oliveira Jr in Sa˜o Paulo. Reynaldo Mendes de Carvalho Jr is the pioneer and still the only functional neurosurgeon in Piauı´, having established his practice in Teresina, state capital, in 1998, after a regular training with Teixeira. Likewise, Paulo Roberto Santos Mendonc¸a, from Aracaju, the capital of Sergipe State, is the only functional neurosurgeon in his state. He trained with Oliveira Jr for 2 years and only recently (2006) started his practice. According to Abynada´ de Siqueira Lyro, the oldest neurosurgeon from Alagoas, there is no one in his state dedicated to stereotactic and functional neurosurgery; some colleagues, though, do perform stereotactic biopsy, like Ricardo Camelo and Aldo Calac¸a. Other colleagues from the Northeast performing stereotactic and functional neurosurgery to a lesser extent are: Antoˆnio Marcos de Albuquerque (Pernambuco), Ju´lio Augusto Lustosa Nogueira (Pernambuco), Joa˜o Carlos Soares de Souza Jr (Maranha˜o), Ro´dio Luiz Branda˜o Caˆmara (Rio Grande do Norte), Ricardo Rodrigues de Carvalho (Paraı´ba), and Orlando Espinheiro Freire de Carvalho Filho (Bahia).
North Region Though a vast region, the largest in the country, Para´, to the best of our knowledge, is the only state of the North region (made up by the states of Amazonas, Para´, Roraima, Rondoˆnia, Acre, Amapa´, and Tocantins) in which stereotactic and functional neurosurgery has been implanted. The first neurosurgeon to perform stereotactic and functional neurosurgery in the North region was Joffre Moreira Lima. Joffre Moreira Lima (1918–1990) graduated from the Federal University of Para´ Medical School in 1955. After finishing his neurosurgical training under Jose´ Ribe Portugal at Rio de Janeiro Neurological Institute (1956 and 1957), he returned to Bele´m, capital of Para´, becoming together with Eloy Simo˜es Bona one of the pioneers of neurosurgery in this state [2]. In the next 2 years (1958 and 1959), he returned frequently to Rio de Janeiro Neurological Institute, keeping in close touch with Portugal and Renato Barbosa, the father of stereotactic and functional neurosurgery in Brazil. While doing the research for this chapter, we were told by Ce´sar Neves, one of the most active and renowned neurosurgeons of Para´, that Joffre Lima had performed stereotactic surgery for Parkinson’s disease a long time before. The physical therapist Lila Janahu, Lima’s daughter, under our request consulted very old operative recordings of Santa Casa de Miserico´rdia de Bele´m Hospital, where her father worked, and discovered that Joffre Lima operated on a patient with epilepsy on 22 September 1958, and a patient with Parkinson’s disease on 22 July 1959. Therefore, it seems that Lima was also the pioneer of stereotactic and functional neurosurgery in Para´. Unfortunately, we could not determine for how long he kept his stereotactic practice or any other details in this regard. More recently, stereotactic and functional neurosurgery was restarted in Para´ by Scylla Lage Silva Neto. Silva Neto, during the last year of his residency in Sa˜o Paulo (1988), accompanied
History of stereotactic and functional neurosurgery in brazil
Teixeira part of the time at USP Medical School. He then moved on to Freiburg, spending 2 years (1989/1990) with Christopher Ostertag. Returning to Bele´m, Silva Neto performed his first functional procedure in March, 1991. Albedy Moreira Bastos, also trained by Teixeira, started his practice in stereotactic and functional neurosurgery in 1994. Though very experienced in the areas of pain surgery and stereotactic biopsy, and despite their adequate training, little have they done in the other areas of stereotactic and functional neurosurgery. Kleber Duarte, another disciple of Teixeira, stayed in Bele´m from 1995 to 1999. During this period he performed a variety of functional procedures, including 32 operations for Parkinson’s disease, but then he moved on to Rio de Janeiro and later to Sa˜o Jose´ do Rio Preto (Sa˜o Paulo State), where he keeps a busy practice. More recently, two other colleagues, Mauro Almeida and Jose´ Cla´udio Rodrigues, both trained by Oliveira Jr, initiated their stereotactic and functional neurosurgery practice in Bele´m.
The Brazilian Functional Neurosurgeons Abroad At present there are five Brazilian functional neurosurgeons working abroad (USA, 3; Canada, 1; and Germany, 1), three of them already well established. The first to follow this trajectory was Antonio De Salles, 25 years ago. Antonio Afonso Ferreira De Salles (Antonio De Salles) (> Figure 16-16) graduated from the Federal University of Goia´s Medical School in 1978. After his 4-year residency at Goiaˆnia Neurological Institute (1979/1982), he went to the Division of Neurosurgery at the Medical College of Virginia for a research fellowship in head injury (1983/1985), and in 1986 he got his PhD at the Virginia Commonwealth University. Both activities were supervised by Donald Becker. His primary interest was on head injury, and he
16
. Figure 16-16 Antonio Afonso De Salles
certainly made very important contributions to this field, which are beyond the scope of this chapter. After that, his interest turned to stereotactic and functional neurosurgery and he moved to Massachusetts General Hospital, Harvard Medical School, for a clinical and research fellowship under Raymond Kjellberg (1986/1988). He then spent 1.5 year at Umea University with Lauri Laitinen (1988/1989). Back to the USA, he was invited by his former boss, Donald Becker, to the UCLA School of Medicine, where he was appointed Assistant Professor, Head of the Stereotactic Surgery Section, and co-director of the radiosurgery program, positions that he still holds. In 1999 he became Full Professor at the Division of Neurosurgery and Department of Radiation Oncology at UCLA School of Medicine. His contributions to stereotactic and functional neurosurgery have been substantial. He was the first to demonstrate that low-frequency/ low-voltage electrical stimulation of the pontine parabrachial region in cats produces a nonopiate-mediated pain suppression, as well as the connections of this region with the frontal fields
235
236
16
History of stereotactic and functional neurosurgery in brazil
[156–158]. De Salles demonstrated for the first time that MRI and CT scans could be matched in the stereotactic space, which prompted the development of the image fusion technique, widely used nowadays [159]. Over the years, his team did several firsts: MR-guided radiofrequency ablation of brain tumors [160]; performance of functional neurosurgery in the MRI operating room, showing the feasibility of electrophysiological studies in such an environment [161,162]; LINAC radiosurgery for the treatment of trigeminal neuralgia [163]; demonstration of the changes in the neurotransmitters environment following low-dose LINAC radiosurgery in non-human primates, which may be implicated in the response to radiosurgery [164]; the use of shaped beam radiosurgery in a multitude of applications [165]; and the impact of ventromedial hypothalamus DBS on food intake in freely moving non-human primates [166]. Once graduated from the Paulista Medical School, Federal University of Sa˜o Paulo, in 1993, Clement Hamani did his PhD in Neurosciences in the same institution (1994/1996), under Luiz Mello, studying the reorganization of hippocampal circuitry in a pilocarpine-induced rat model of epilepsy [167,168]. During this period, he spent 8 months at the University of Illinois in Chicago as a research fellow, working with microdialysis in the brain [169]. From 1997 to 2002, he did his residence training at the USP Medical School, under Raul Marino Jr, and after a 6-month fellowship in functional neurosurgery with Manoel Teixeira, in the same institution, he went to the University of Toronto, where, under the supervision of Andres Lozano, he did his postdoctorate fellowship. In 2005 he became Invited Professor (neurophysiology) of the Federal University of Sa˜o Paulo. In 2007, he was offered the position of Associate Researcher of the Center of Addiction and Mental Health of the University of Toronto, and in July, 2008, became Assistant Professor, Division of Neurosurgery, Toronto Western Hospital, of this great university. As a member of Andres Lozano’s team, he has been part of a
number of very important researches, both clinical and laboratorial, such as subgenual cingulum DBS for depression [106], anterior thalamic nucleus DBS for epilepsy [170,171], hypothalamic/ fornix DBS for memory enhancement [172], and pedunculopontine nucleus DBS for progressive supranuclear palsy and Parkinson’s disease [173], among others [147,148,174–180]. After finishing his residency (1998/2002) and preceptorship (2003/2004) at the University of Sa˜o Paulo Medical School, under Raul Marino Jr, Andre´ Machado did a 2-year fellowship at the Center of Neurological Restoration, Department of Neurosurgery, Cleveland Clinic, under the supervision of Ali Rezai, and since then he became a staff member of the same institution. He runs a lab at the Cleveland Clinic Lemer Research Institute focused on motor rehabilitation in animal models of stroke [181]. As part of Ali Rezai’s team, he has greatly contributed to clinical researches in motor cortex [182] and gasserian ganglion [183] stimulation for pain, DBS for movement disorders [184,185] and psychiatric illnesses [186,187], and thalamic stimulation for minimally conscious state [188]. The two other colleagues working abroad are Alessandra Gorgulho and Guilherme Lepski. Gorgulho attended a 1-year fellowship with Teixeira, in Sa˜o Paulo, and a 3-year research fellowship with the Salles, at UCLA. At present, she is an assistant researcher at the Stereotactic Surgery Section at UCLA, but is still not working as a functional neurosurgeon. Lepski, of German origin, an ex-fellow and assistant to Teixeira, in Sa˜o Paulo, has just recently (October 2007) moved to Tu¨bingen, and is now working at the University of Eberhard-Karls under Marcos Tatagiba.
The Brazilian Society for Stereotactic and Functional Neurosurgery Despite its constant growth, the Brazilian Society of Neurosurgery (Sociedade Brasileira de
History of stereotactic and functional neurosurgery in brazil
Neurocirurgia – SBN) started to be departmentalized into its various subspecialties only in 1992, during the presidency of Carlos Batista Alves de Souza. The history of Brazilian stereotactic and functional neurosurgery, however, is somewhat different from the other neurosurgical subspecialties. It started as an independent society, although connected to the SBN. After the departmentalization of SBN, it also became its Department of Functional Surgery and Pain. During the 13th Brazilian Congress of Neurosurgery, chaired by Laelio Lucas and held in Guarapari, Espı´rito Santo State, by initiative of Carlos Telles, who had recently arrived from Germany after a 4-year fellowship, the Brazilian Society for Stereotactic and Functional Neurosurgery was founded on 16 September 1980. Before the conference, Telles had written to colleagues around the country who were dealing with functional neurosurgery and invited them for a meeting dedicated to the foundation of the Brazilian Society for Stereotactic and Functional Neurosurgery. The following colleagues attended the meeting, becoming founding members of the Brazilian Society for Stereotactic and Functional Neurosurgery: Renato Barbosa, Jose´ Barros, Alceu Correia, Bernardo Couto, Djacir Figueiredo, Lourenc¸o de Freitas Neto, Nilton Latuf, Raul Marino Jr, Luiz Fernando Martins, Pedro Motta, Delfim Nunes Neto, Sergio Ottoni, Jorge Pagura, Otoı´de Pinheiro, Jose´ Vitor Pinto, Telmo Reis, Manoel Teixeira, Carlos Telles, and Gianni Temponi. Blaine Nashold Jr and Edward Hitchcock, invited speakers to the Brazilian Congress, were also present, and thankful to the invaluable support lent by Nashold, the Brazilian Society for Stereotactic and Functional Neurosurgery became affiliated to the World Society for Stereotactic and Functional Neurosurgery since its birth. Renato Barbosa, Raul Marino Jr, Carlos Telles, Otoı´de Pinheiro, and Jose´ Barros were elected, respectively, as president, vice-president, secretary, assistant secretary, and treasurer of the first board of directors of our Society. Although there
16
is some controversy about to whom should be attributed the initiative to create Brazilian Society for Stereotactic and Functional Neurosurgery, the data here reported were collected from the blotter of Brazilian Society for Stereotactic and Functional Neurosurgery. The leadership of the Brazilian Society for Stereotactic and Functional Neurosurgery is renewed every 2 years, and the following were the presidents from its birth up to now: 1980/ 1982, Renato Barbosa; 1982/1984, Raul Marino Jr; 1984/1986, Luiz Fernando Martins; 1986/ 1988, Carlos Telles; 1988/1990, Jose´ Vitor Pinto; 1990/1992, Manoel Teixeira; 1992/1994, Sebastia˜o Gusma˜o; 1994/1996, Alceu Correia; 1996/ 1998, Cla´udio Correˆa; 1998/2000, Manoel Teixeira; 2000/2002, Luiz Fernando Martins; 2002/ 2004, Jorge Pagura; 2004/2006, Edvaldo Cardoso; 2006/2008, Osvaldo Vilela Filho; and 2008/2010, Paulo Brainer-Lima. The main focus of Brazilian Society for Stereotactic and Functional Neurosurgery has always been educational. Approximately three or four meetings are organized every year, usually as a parallel activity within another greater meeting such as those of the Brazilian Society of Neurosurgery, Brazilian Academy of Neurosurgery, SIMBIDOR – Brazilian Symposium and International Meeting on Pain, and CINDOR – Interdisciplinary Congress on Pain. Besides, every 2 years, in Sa˜o Paulo, there is a meeting co-organized by Brazilian Society for Stereotactic and Functional Neurosurgery and Micromar. The first time the Brazilian Society for Stereotactic and Functional Neurosurgery organized a completely independent meeting, was on 16–19 May 2007, the 1st International Joint Meeting on Stereotactic and Functional Neurosurgery & 8th Congress of the Brazilian Society for Stereotactic and Functional Neurosurgery: The Future Today!, held at Rio Quente Resorts, Rio Quente, Goia´s State, and chaired by Osvaldo Vilela Filho, president of the Brazilian Society for Stereotactic and Functional Neurosurgery. The conference venue, with its hot springs, was magnificent, the
237
238
16
History of stereotactic and functional neurosurgery in brazil
scientific level was the highest possible, and the social activities were superb. The faculty was made up by 42 speakers, 18 from Brazil and 24 from other countries (US, Canada, Mexico, France, Italy, Sweden, China, Argentina, Chile, Venezuela, and Colombia), all well-known functional neurosurgeons. The Brazilian Society for Stereotactic and Functional Neurosurgery did not fund any speaker. Everyone graciously accepted to come on their own, in a gesture of great consideration and friendship never to be forgotten. The official clothing was casual wear, being suits and ties officially prohibited, which was instrumental in establishing a very informal atmosphere. The great innovation of the meeting was the ‘‘spicy-session,’’ in which controversial topics were fully debated by experts in an attempt to reach a consensus. One hundred and forty three participants attended the meeting, half from Brazil and half from various other countries. The Brazilian Society for Stereotactic and Functional Neurosurgery paid homage to two friends and colleagues, Ronald Tasker, International Honored President, and Luiz Fernando Martins, Brazilian Honored President, for their invaluable contributions to, respectively, world and Brazilian stereotactic and functional neurosurgery. Finally, by initiative of Eduardo Barreto and Osvaldo Vilela Filho, there was a session during the meeting, coordinated by Elliot Krames, president of the International Neuromodulation Society (INS), destined to the foundation of the Brazilian Neuromodulation Society, the Brazilian chapter of the INS, which happened on 18 May 2007. Its first board of directors was then elected: Eduardo Barreto, president; Osvaldo Vilela Filho, vice-president; and Alexandre Francisco, secretary. The Brazilian Society for Stereotactic and Functional Neurosurgery has 98 active members (85 are adequately formed functional neurosurgeons, nine of which are either currently inactive or never performed functional neurosurgery, and 13 are non-functional neurosurgeons or
neurologists) and five honorary members, who were chosen based on their great contribution to the field and relationship with Brazilian functional neurosurgeons: Ronald Tasker, Antoˆnio de Salles, Edward Hitchcock, Blaine Nashold Jr, and Wolfhard Winkelmu¨ller. Besides, there are 61 other neurosurgeons performing at least some of the stereotactic and/or functional procedures, like epilepsy alone or pain surgery and stereotactic biopsy, which gives a total of 137 (98–9–13 = 76 + 61 = 137) neurosurgeons at least partially involved with functional neurosurgery (stereotactic biopsy alone was not considered), as shown in > Figure 16-17. However, only 60 Brazilian neurosurgeons (41 members and 19 non-members) are truly functional neurosurgeons, that is, most of their practice is dedicated to stereotactic and functional neurosurgery. Finally, also worth mentioning is the creation of the Brazilian Society for the Study of Pain, which happened on 29 August 1983, by initiative of the functional neurosurgeon Jorge Pagura and the neurologist Moacir Schnapp. One year later it was affiliated to the International Association for the Study of Pain. Pagura became its second president, in fact the only functional neurosurgeon elected for president of this society.
Non-Medical Contributors for the Development of Stereotactic and Functional Neurosurgery in Brazil Until 1985, with the exception of the stereotactic frames built by the neurosurgeons Jose´ Barros and Jose´ Zaclis for their own use, all other stereotactic apparatuses in the country were imported, at a very high cost, being those of Riechert-Mundinger and Hitchcock the most commonly found. In 1985, the engineer Antonio Martos (CEO of Micromar), with the supervision of Manoel Teixeira, built the first commercially available Brazilian stereotactic apparatus,
History of stereotactic and functional neurosurgery in brazil
16
. Figure 16-17 Distribution of Brazilian neurosurgeons performing functional procedures per regions and states
. Figure 16-18 Micromar radiofrequency generator
the Micromar stereotactic system, a modification of that of Hitchcock’s. Since then, this frame has been continuously refined, achieving a very high quality. Micromar is still the only Brazilian company dedicated to the stereotactic field.
It was also very instrumental to this field manufacturing electrode kits for the different neurosurgical procedures, radiofrequency generators (the first version, released in 1996, was basically analogical and received a series of refinements along the years; in 2008, an entirely digital version was released) (> Figure 16-18), and the radiosurgery system for linear accelerator (its production was again supervised by Manoel Teixeira, being released for clinical use in 1998). Another very important person in this regard has been the physicist Armando Alaminos Bouza, with a master degree in neurosciences. He was the one responsible for the development of the software (MNPS) used in association with the Micromar stereotactic system and Micromar radiosurgery system, allowing the determination of stereotactic coordinates, detailed planning of the electrode trajectory, overlaying CT and MR
239
240
16
History of stereotactic and functional neurosurgery in brazil
. Figure 16-19 Neuroaugmentative procedures performed in Brazil from 2002 to 2008. SCS = spinal cord stimulation; DBS = deep brain stimulation; MCS = motor cortex stimulation; VNS = vagal nerve stimulation
images with digitized maps of stereotactic atlas (Schaltenbrand & Wahren and Talairach), shrunk or stretched as necessary to match the patient’s intercomissural distance, multimodality image fusion, and planning for radiosurgery and brachytherapy. Also important for the development of the stereotactic field in our country were Sandra Ferraz and Victor Dabah, from Dabasons Importac¸a˜o, Exportac¸a˜o e Come´rcio Ltda, local representative of Medtronic, and Joaquim Cordeiro and Ma´rcio Sossai, from JM Come´rcio e Importac¸a˜o Ltda, local representative of ANS, for their role in directly or indirectly (through neurosurgeons) spreading the reasoning for the use of neuroaugmentative techniques across the country and sponsoring the stereotactic and functional neurosurgery meetings. Sandra Ferraz should also be accounted for tirelessly backing the functional neurosurgeons in their fight against the health insurance companies in an attempt to make them pay for the procedures and necessary hardware. The results of their combined efforts can be better appreciated in > Figure 16-19, which shows a significant increase in the number of neuromodulatory procedures in our country.
Recent developments will most probably change forever this panorama, strengthening the tendency shown in > Figure 16-19. Until recently, neuromodulatory surgeries were not officially recognized by the National Health Agency (Ageˆncia Nacional de Sau´de – ANS). On 2 April 2008, though, according to the normative resolution number 167 of ANS, published in the Official Diary of the Union, neuroaugmentative surgeries became part of the roll of procedures recognized by ANS, which means that, from now on, private health insurance companies will be obliged to authorize and pay for these procedures. The Brazilian functional neurosurgical community will always be in great debt to all these people.
Acknowledgments Besides those already mentioned in the introductory part of the chapter, the author would like to express his deep gratitude to Fernanda Vilela for proofreading this manuscript.
History of stereotactic and functional neurosurgery in brazil
References 1. Horrax G. Neurosurgery: a historical sketch. Springfield, IL: Thomas; 1952. p. 10-15. 2. Gusma˜o SS, de Souza JG. Histo´ria da Neurocirurgia no Brasil [History of Brazilian neurosurgery]. Joinville: Letra Me´dica; 2000. 3. Souza APS, Vale´rio AG. A cirurgia nervosa no Brasil [Neurological surgery in Brazil]. Anais do Segundo Congresso Brasileiro de Neurologia-Psychiatria e Medicina Legal. Arch Neurol Psiq e Med Legal 1922;6:30-67. 4. Mario Brock. Jose´ Ribeiro Portugal. J Bras Neurocir 2008;19(1):38-41. 5. Coutinho MF. Obitua´rio: Renato Tavares Barbosa. J Bras Neurocir 2003;14(1):29. 6. Livro de atas da Sociedade Brasileira de Neurocirurgia Funcional e Estereota´xica (Blotter of the Brazilian Society for Stereotactic and Functional Neurosurgery). 7. Filho PN. Paulo Niemeyer. Arq Bras Neurocir 2004;23 (1):8-11. 8. Yacubian EM, Cavalheiro EA. In memoriam: Paulo Niemeyer Soares. Epilepsia 2004;45(9):1166-8. 9. Niemeyer P. Tratamento ciru´rgico do tremor parkinsoniano, da atetose e outras hipercinesias. Med Cirurg Farm 1945;116:1-16. 10. Niemeyer P. Tratamento ciru´rgico das discinesias. Rev Bras Cir 1955;30(4):211-48. 11. Niemeyer P. Supressa˜o da resposta motora e da atividade cortical ele´trica no homem. Arq Neuropsiquiatr 1946;4:109-17. 12. Niemeyer P. The transventricular amygdala-hippocampectomy in temporal lobe epilepsy. In: Baldwin MBP, editor. Temporal lobe epilepsy. Springfield, IL: Thomas; 1958. p. 461-82. 13. Cendes F, Dubeau F, Olivier A, Cukiert A, Andermann E, Quesney LF, et al. Increased neocortical spiking and surgical outcome after selective amygdalo-hippocampectomy. Epilepsy Res 1993;16(3):195-206. 14. Lynch JC. Gianni Maure´lio Temponi (1934–2007). Boletim da SBN (SBN Newsletter). Outubro; 2007. p. 22. 15. Temponi G. Estudo sobre 100 intervenc¸o˜es estereota´xicas cerebrais no tratamento do parkinsonismo (Abstr.). In: Mattos Pimenta A, editor. XI Congresso Latino-Americano de Neurocirurgia (Relato´rios Oficiais e Comunicac¸o˜es Livres), Sa˜o Paulo; 1965. p. 272. 16. Telles CR. Nova te´cnica para o tratamento das algias cervicofaciais neopla´sicas. Presented at the 14th Brazilian Congress of Neurosurgery, Belo Horizonte, Sept 1982. 17. Telles CR. Tratamento ciru´rgico das dores cervicofaciais de origem neopla´sica. Presented at the 20th Latin American Congress of Neurosurgery, Sa˜o Paulo, Mar 1983. 18. Telles CR. Facial pain due to cancer of head and neck. Presented at the Annual Meeting of the American Society for Stereotactic and Functional Neurosurgery, North Caroline, Apr 1983. 19. Aloysio Mattos Pimenta. Curriculum Vitae, 1959.
16
20. Yahn M, Mattos Pimenta A, Sette A, Jr. Tratamento ciru´rgico das mole´stias mentais (Leucotomia), Sa˜o Paulo; 1951. 21. Mattos Pimenta A. XI Congresso Latino-Americano de Neurocirurgia (Relato´rios Oficiais e Comunicac¸o˜es Livres). Sa˜o Paulo; 1965. 22. Barros NG. In memoriam: Jose´ Zaclis (1917–1983). Arq Neuropsiquiatr 1984;42:84-5. 23. Zaclis J. [X-ray visualization of the entire cerebrovascular system by means of contrast medium injection into one single artery; cerebral pan-angiography.] Arq Neuropsiquiatr 1959;17(1):1-22. (Portuguese). 24. Zaclis J, Longo PH, Cruz OR. Intrathoracic hyperpressure in neuroradiology; results in (1) cerebral panangiography and (2) arteriography of the thoracocervical region. J Neurosurg 1964;21:1087-94. 25. Zaclis J, Koralek JM, Longo PH, Almeida GM. Novo aparelho para neurocirurgia estereota´xica (Abstr.). In: Mattos Pimenta A (Ed). XI Congresso Latino-Americano de Neurocirurgia Sa˜o Paulo; (Relato´rios Oficiais e Comunicac¸o˜es Livres). 1965. p. 272. 26. Marino R. Jr. Reflections on a stereotactic and functional journey: the blessing and inspiration of working among giants. Neurosurgery 2005;56(1):172-7. 27. Ballantine HT, Jr, Cassidy WL, Flanagan NB, Marino R, Jr. Stereotaxic anterior cingulotomy for neuropsychiatric illness and intractable pain. J Neurosurg 1967;26 (5):488-95. 28. Cassidy WL, Ballantine HT, Jr, Flanagan NB. Frontal cingulotomy for affective disorders. Recent Adv Biol Psychiatry 1965;8:269-82. 29. Rasmussen T, Marino R, Jr. Functional neurosurgery. New York: Raven Press; 1978. 30. Ragazzo PC, Manzano GM, Marino R, Jr. Functional microsurgical partial callosotomy in patients with secondary generalized epilepsies. I. Disruption of bilateral synchrony of spike and wave discharges. Appl Neurophysiol 1988;51(6):297-306. 31. Marino Ju´nior R, Gronich G. Corpus callosum stimulation and stereotactic callosotomy in the management of refractory generalized epilepsy. Arq Neuropsiquiat 1989;47(3):320-5. 32. Marino Ju´nior R. Surgery for epilepsy. Selective partial microsurgical callosotomy for intractable multiform seizures: criteria for clinical selection and results. Appl Neurophysiol 1985;48(1–6):404-7. 33. Cukiert A, Puglia P, Scapolan HB, Vilela MM, Marino Ju´nior R. Congruence of the topography of intracranial calcifications and epileptic foci. Arq Neuropsiquiatr 1994;52(3):289-94. 34. Marino Ju´nior R, Cosgrove GR. Neurosurgical treatment of neuropsychiatric illness. Psychiatr Clin North Am 1997;20(4):933-43. 35. Wen HT, Rhoton AL, Jr, Marino R, Jr. Anatomical landmarks for hemispherotomy and their clinical application. J Neurosurg 2004;101(5):747-55.
241
242
16
History of stereotactic and functional neurosurgery in brazil
36. de Almeida AN, Marino R, Jr. The early years of hemispherectomy. Pediatr Neurosurg 2005;41(3): 137-40. 37. Wen HT, Rhoton AL, Jr, Marino R, Jr. Gray matter overlying anterior basal temporal sulci as an intraoperative landmark for locating the temporal horn in amygdalohippocampectomies. Neurosurgery 2006;59 (4 Suppl 2):ONS221-7. 38. Teixeira MJ. Various functional procedures for pain. Part II: Facial pain. In: Gildenberg PL, Tasker RR, editors. Textbook of stereotactic and functional neurosurgery. New York: McGraw-Hill; 1998. p. 1389-402. 39. Teixeira MJ, Lepski G, Aguiar PH, Cescato VA, Rogano L, Alaminos AB. Bulbar trigeminal stereotactic nucleotractotomy for treatment of facial pain. Stereotact Funct Neurosurg 2003;81(1–4):37-42. 40. Hitchcock E. Stereotactic trigeminal tractotomy. Farmatsiia 1970;19(3):131-5. 41. Hitchcock E, Teixeira MJ. Pontine stereotactic surgery and facial nociception. Neurol Res 1987;9(2):113-7. 42. Fonoff ET, Dale CS, Pagano RL, Paccola CC, Ballester G, Teixeira MJ, et al. Antinociception induced by epidural motor cortex stimulation in naive conscious rats is mediated by the opioid system. Behav Brain Res (2008). doi:10.1016/j.bbr.2008.07.027. 43. Fonoff ET, Dale CS, Pagano RL, Teixeira MJ, Ballester G, Giorgi R. Antinociceptive effect induced by sensorimotor cortex stimulation in normal rats. Young Investigator Award. Presented at the 14th Meeting of the World Society for Stereotactic and Functional Neurosurgery, Rome, Italy, June 2005. 44. Teixeira MJ, Fonoff ET, Montenegro MC. Dorsal root entry zone lesions for treatment of pain-related to radiation-induced plexopathy. Spine 2007;32(10): E316-9. 45. Teixeira MJ, Fonoff ET, Macri F, Picarelli H, Lin TY, Marcolin MA. Prediction of Motor Cortex Simulation in Treatment of Brachial Plexus Avulsion Pain by Transcranial Magnetic Stimulation. NeuPSIG Poster Prize (category: clinical studies) presented at the Second International Congress on Neuropathic Pain, Berlin, Germany, June 2007. 46. de Almeida AN, Fonoff ET, Ballester G, Teixeira MJ, Marino R, Jr. Stereotactic disconnection of hypothalamic hamartoma to control seizure and behavior disturbance: case report and literature review. Neurosurg Rev 2008;31 (3):343-9. 47. Fonoff ET, Lara N, Teixeira MJ. Phrenic nerve stimulation for the treatment of refractory hiccup. Presented at the Meeting of the European Society for Stereotactic and Functional Neurosurgery, Rimini, Italy, Oct 2008. 48. Teixeira MJ, Fonoff ET, Santana PA, Padilha PM, Cescato VAS. Computer-assisted stereotactic fenestration of aqueductal cyst for treatment of hydrocephalus: a technical report. Presented at the, 14th Meeting of the World Society for Stereotactic and Functional Neurosurgery, Rome, Italy, June 2005.
49. Cescato VAS, Teixeira MJ, No´brega, JPS, Almeida GM, Oliveira JO, Jr., Butori S. Tratamento de granuloma blastomico´tico cı´stico talaˆmico por meio de instilac¸a˜o estereota´xica de quimiotera´picos atrave´s de caˆnula intralesional. Presented at the XVI Brazilian Congress of Neurosurgery, Rio de Janeiro, Brazil, Sept 1986. 50. Hitchcock ER, Teixeira MJ. A comparison of results from center-median and basal thalamotomies for pain. Surg Neurol 1981;15(5):341-51. 51. Teixeira MJ, De Souza EC, Yeng LT, Pereira WC. [Lesion of the Lissauer tract and of the posterior horn of the gray substance of the spinal cord and the electrical stimulation of the central nervous system for the treatment of brachial plexus avulsion pain]. Arq Neuropsiquiatr 1999;57 (1):56-62. (Portuguese). 52. Teixeira MJ, Lepski G, Correia C, Aguiar PH. Interstitial irradiation for CNS lesions. Stereotact Funct Neurosurg 2003;81(1–4):24-9. 53. Rogano L, Teixeira MJ, Lepski G. Chronic pain after spinal cord injury: clinical characteristics. Stereotact Funct Neurosurg 2003;81(1–4):65-9. 54. Rogano LA, Greve JM, Teixeira MJ. Use of intrathecal morphine infusion for spasticity. Arq Neuropsiquiatr 2004;62(2B):403-5. 55. de Siqueira SR, da No´brega JC, de Siqueira JT, Teixeira MJ. Frequency of postoperative complications after balloon compression for idiopathic trigeminal neuralgia: prospective study. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2006;102(5):e39-45. 56. Arantes PR, Cardoso EF, Barreiros MA, Teixeira MJ, Gonc¸alves MR, Barbosa ER, et al. Performing functional magnetic resonance imaging in patients with Parkinson’s disease treated with deep brain stimulation. Mov Disord 2006;21(8):1154-62. 57. Teixeira MJ, de Siqueira SR, Bor-Seng-Shu E. Glossopharyngeal neuralgia: neurosurgical treatment and differential diagnosis. Acta Neurochir (Wien) 2008;150 (5):471-5. 58. de Almeida AN, Teixeira MJ, Feindel WH. From lateral to mesial: the quest for a surgical cure for temporal lobe epilepsy. Epilepsia 2008;49(1):98-107. 59. Pagura JR. Percutaneous radiofrequency spinal rhizotomy. Appl Neurophysiol 1983;46(1–4):138-46. 60. Pagura JR, Rabello JP, Lima WC. Microvascular decompression for trigeminal neuralgia. In: Gildenberg PL, Tasker RR, editors. Textbook of stereotactic and functional neurosurgery. New York: McGraw-Hill; 1998. p. 1715-21. 61. Pagura JR, Schnapp M, Passarelli P. Percutaneous radiofrequency glossopharyngeal rhizotomy for cancer pain. Appl Neurophysiol 1983;46(1–4):154-9. 62. Correˆa CF, Teixeira MJ. Balloon compression of the Gasserian ganglion for the treatment of trigeminal neuralgia. Stereotact Funct Neurosurg 1998;71(2):83-9. 63. Marino R, Jr., Cukiert A, Pinho E. Epidemiological aspects of epilepsy in Sa˜o Paulo, Brazil. A prevalence rate study. In:
History of stereotactic and functional neurosurgery in brazil
64.
65.
66.
67.
68.
69.
70.
71. 72.
73.
74. 75.
76.
77.
78.
Wolf P, Dam M, Janz D, Dreifuss FE, editors. Advances in epileptology. New York: Raven Press; 1987. p. 759-64. Cukiert A, Baumel SW, Andrioli MSD, Marino R, Jr. Effects of corpus callosum stimulation on the morphology and frequency of the epileptic bursts in the feline topical penicillin generalized model. Stereotact Funct Neurosurg 1989;52:18-25. Cendes F, Dubeau F, Olivier A, Cukiert A, Andermann E, Quesney LF, et al. Increased neocortical spiking and surgical outcome after selective amygdalo-hippocampectomy. Epilepsy Res 1993;16(3):195-206. Olivier A, Germano I, Cukiert A, Peters T. Frameless stereotaxy for surgery of the epilepsies: preliminary experience. Technical note. J Neurosurg 1994;81:629-33. Cukiert A, Burattini JA, Machado E, Sousa A, Vieira JO, Argentoni M, et al. Results of surgery in patients with refractory extratemporal epilepsy with normal or nonlocalizing magnetic resonance findings investigated with subdural grids. Epilepsia 2001;42:889-94. Cukiert A, Burattini JA, Machado E, Sousa A, Vieira JO, Forster C, et al. Seizure-related outcome after corticoamygdalohippocampectomy in patients with refractory temporal lobe epilepsy and mesial temporal sclerosis evaluated by magnetic resonance alone. Neurosurg Focus 2002;13(4): ecp2 Beraha J, Barcia Salorio JL, Bordes V, Hernandez G. Radiocirurgia estereota´xica./Stereotaxic radiosurgery. Arq Bras Neurocir 1984;3(1):15-22. Beraha J, Feriancic CV, Scaff LA. [Stereotaxic radiosurgery in the treatment of cerebral arteriovenous malformations]. Arq Neuropsiquiatr 1987;45(4):391-6. In memoriam: Dr. Nilton Luiz Latuf (1937–2005). Boletim da SBN (SBN Newsletter). Dezembro; 2005. Siqueira JM. A method for bulbospinal trigeminal nucleotomy in the treatment of facial deafferentation pain. Appl Neurophysiol 1985;48(1–6):277-80. Nashold BS, Jr, Caputti F, Bernard E. Trigeminal DREZ: Caudalis nuclear lesions for relief of facial pain (Abstr). Neurosurgery 1986;19(1):150. Schvarcz JR. Trigeminal DREZ (correspondence to the editor). Neurosurgery 1987;20(2):348. Gorecki JP. Dorsal root entry zone and brainstem ablative procedures. In: Winn HR, editor. Youmans neurological surgery. 5th ed. Philadelphia, PA: Saunders; 2004. p. 3045-58. Tasker RR, Siqueira J, Hawrylyshyn P, Organ LW. What happened to VIM thalamotomy for Parkinson’s disease? Appl Neurophysiol 1983;46(1–4):68-83. Tasker RR, DeCarvalho GTC, Dolan EJ. Intractable pain of spinal cord injury: clinical features and implications for surgery. J Neurosurg 1992;77:373-8. Tasker RR, de Carvalho G, Dostrovsky JO. The history of central pain syndromes, with observations concerning pathophysiology and treatment. In: Casey KL, editor. Pain and central nervous system disease: the central pain syndromes. New York: Raven Press; 1991. p. 31-58.
16
79. Tasker RR, De Carvalho GTC. Central pain of spinal cord injury. In: North RB, Levy RM, editors. Neurosurgical management of pain. New York: Springer; 1997. p. 110-6. 80. Tasker RR, DeCarvalho GC, Li CS, Kestle JRW. Does thalamotomy alter the course of Parkinson’s disease? Adv Neurol 1996;69:563-83. 81. Gusma˜o S, Oliveira M, Tazinaffo U, Honey CR. Percutaneous trigeminal nerve radiofrequency rhizotomy guided by computerized tomography fluoroscopy. Technical note. J Neurosurg 2003;99(4):785-6. 82. Martins LF, Umbach W. Simple determination of the foramen in trigeminus coagulation. Neurochirurgia (Stuttg) 1975;18(5):163-6. 83. Martins LF, Umbach W. [Clinico-anatomical correlation between sighting of the lesion and alleviation of pain following stereotactic surgery]. Neurochirurgia (Stuttg) 1974;17(3):77-83. (German). 84. Martins LF, Umbach W. Size and position of stereotaxic lesions in comparison with clinical pain relief. Confin Neurol 1975;37(1–3):80-5. 85. Damasceno BP, Campos Junior SS, Martins LF, de Melo-Souza SE. [Clinical treatment of resistant epilepsy]. Arq Neuropsiquiatr 1988;46(4):351-8. (Portuguese). 86. Cendes F, Ragazzo PC, da Costa V, Martins LF. [Interhemispheric disconnection syndrome following total callosotomy associated to anterior commissurotomy for the treatment of intractable epilepsy: a case report]. Arq Neuropsiquiatr 1990;48(3):385-8. (Portuguese). 87. Cendes F, Ragazzo PC, da Costa V, Martins LF. Corpus callosotomy in treatment of medically resistant epilepsy: preliminary results in a pediatric population. Epilepsia 1993;34(5):910-7. 88. Zaccariotti VA, Martins LF, da Costa V, Silva NA, da Casas AA, de Melo-Souza SE. [Sneddon syndrome. Report of 3 cases]. Arq Neuropsiquiatr 1995;53(1):82-7. (Portuguese). 89. Vilela Filho O. Estrate´gia para a abordagem de leso˜es intracranianas. Rev Goiana de Medicina 1989;35:59-70. 90. Vilela Filho O, Rocha JCN, Almeida WC, Leite MSB, Silva DJ, Cavalcante JE. Bio´psia de leso˜es intracranianas a ma˜o livre assistida por tomografia computorizada ou ultra-sonografia. Arq Bras Neurocir 1991;10(2):103-19. 91. Tasker RR, Vilela Filho O. Deep brain stimulation for neuropathic pain. Stereotact Funct Neurosurg 1995;65 (1–4):122-4. 92. Tasker RR, Vilela Filho O. Deep brain stimulation for the control of intractable pain. In: Youmans JR, editor. Neurological surgery. 4th ed. Philadelphia, PA: Saunders; 1996. p. 3512-27. 93. Vilela Filho O. Risk factors for unpleasant paresthesiae induced by parethesiae-producing deep brain stimulation. Arq Neuropsiquiatr 1996;54(1):57-63. 94. Vilela Filho O. Risk factors for paresthesia-producing deep brain stimulation (Abstr.). Resumos de poˆsteres do 3 Simpo´sio Internacional de Dor (Simbidor). Sa˜o Paulo; p. 19-20.
243
244
16
History of stereotactic and functional neurosurgery in brazil
95. Rezai AR, Francisco AN, Vilela Filho O, Lozano AM, Tasker RR. Deep brain stimulation for intractable neuropathic pain: contemporary management and outcome in 80 patients (Abstr.). J Neurosurg 1998;88:416A. 96. Vilela Filho O, Tasker RR. ‘‘Pathways involved in thalamic ventrobasal stimulation for pain relief: evidence against the hypothesis VB stimulation ! rostroventral medulla excitation ! dorsal horn inhibition’’. Arq Neuropsiquiatr 1994;52(3):386-91. 97. Vilela Filho O. Thalamic ventrobasal stimulation for pain relief: probable mechanisms, pathways and neurotransmitters. Arq Neuropsiquiatr 1994;52(4): 578-84. 98. Vilela Filho O. Pathophysiology of the steady burning, dysesthetic component of neuropathic pain: a hypothesis (Abstr). Stereotact Funct Neurosurg 2001;77:205. 99. Vilela Filho O. Tratamiento Neuroquiru´rgico del Dolor: Estado Actual. Neurotarget 2006;1(2):6-20. 100. Vilela Filho O, Correˆa CF. Dor e neuroestimulac¸a˜o: Estimulac¸a˜o ele´trica do sistema nervoso para o manejo da dor croˆnica intrata´vel. Biotecnologia 1998;6:56-66. 101. Vilela Filho O, Correˆa CF. Tratamento da dor croˆnica refrata´ria por neuroestimulac¸a˜o. Neurocirurgia Contemporaˆnea Brasileira 1999;2(12): 1–8. 102. Vilela Filho O, Araujo MR, Florencio RS, Silva MAC, Silveira MT. CT-guided percutaneous punctate midline myelotomy for the treatment of intractable visceral pain: technical note. Stereotact Funct Neurosurg 2001;77: 177-82. 103. Vilela Filho O, Pereira CE. Establishment of a reliable landmark to guide percutaneousradiofrequency occipital neurotomy (Abstr). Stereotact Funct Neurosurg 2001;77:204. 104. Vilela Filho O. In vivo evaluation of cordotomy lesions by magnetic resonance imaging: a new approach (Abstr). Stereotact Funct Neurosurg 2001;77:203-4. 105. Vilela Filho O, Carneiro Filho O, Souza HAO, Machado DC, Rodrigues Filho S, Campos JA. SPECT-based tailoring of psychosurgical procedures: is it possible? Stereotact Funct Neurosurg 2001; 76:256-61. 106. Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, et al. Deep brain stimulation for treatment-resistant depression. Neuron 2005;45 (5):651-60. 107. Franzini A, Ferroli P, Leone M, Broggi G. Stimulation of the posterior hypothalamus for treatment of chronic intractable cluster headaches: first reported series. Neurosurgery 2003;52(5):1095-9. 108. Vilela Filho O, Souza JT. Sı´ndrome de Tourette: uma decorreˆncia da hiperatividade do pa´lido lateral? Rev Psiquiat Clin 1996;22(4):138 (Abstr.). 109. Vilela Filho O, Souza JT. Sı´ndrome de Tourette: Anatomofisiologia dos gaˆnglios da base e sistema lı´mbico,
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
circuita´ria e fisiopatologia. In: Santos MGP, editor. Sı´ndrome de Gilles de la Tourette: tiques nervosos e transtornos de comportamento associados na infaˆncia e adolesceˆncia. Sa˜o Paulo: Lemos Editorial; 1998. p. 137-58. Vilela Filho O, Ragazzo PC, Silva DJ, Souza JT, Oliveira PM, Carneiro Filho O, et al. Globus Pallidus externus (GPe) deep brain stimulation (DBS) for the treatment of Tourette Syndrome (TS): evaluation of a new target in a prospective Controlled Study (Abstr.). In Abstract Book of the ASSFN (American Society for Stereotactic and Functional Neurosurgery) 2004 Biennial Meeting – Neuromodulation: Defining the Future. Cleveland: Cleveland Clinic Foundation; 2004. Vilela Filho O, RagazzoPC, Silva DJ, Souza JT, Oliveira PM, Ribeiro TMC. Bilateral Globus Pallidus Externus Deep Brain Stimulation (GPe-DBS) for the Treatment of Tourette Sqndrome: an ongoing prospective controlled study (Abstr.). Stereotact Funct Neurosurg 2007;85: 42-3. Vilela Filho O. DBS for Tourette’s syndrome (Abstr.). In Syllabus – CME Conference of the eighth World Congress of the International Neuromodulation Society & 11th Annual Meeting of the North American Neuromodulation Society Acapulco; (Neuromodulation: Technology at the Neural Interface). 2007. p. 21. Vilela Filho O, Ragazzo PC, Silva DJ, Souza JT, Oliveira PM, Ribeiro TM. Bilateral GPe-DBS for Tourette’s Syndrome (Abstr.). Neurotarget 2008;3(1):66. Vilela Filho O. CT-based stereotactic subcaudate tractotomy (SST): technical note (Abstr). Stereotact Funct Neurosurg 2001;76:287-8. Vilela Filho O, Silva DJ, Souza HAO, Cavalcante JES, Souza JT, Ferraz FP, et al. Stereotactic subthalamic nucleus (STN) lesioning for the treatment of Parkinsons’s disease (PD). Stereotact Funct Neurosurg 2001;77: 79-86. Vilela Filho O, Silva DJ. Unilateral subthalamic nucleus lesioning: a safe and effective treatment for Parkinson’s disease. Arq Neuropsiquiatr 2002;60(4):935-48. Vilela Filho O, Silva DJ. STN lesioning for the treatment of Parkinson’s disease: risks and benefits in a long-term follow-up study (Abstr.). Stereotact Funct Neurosurg 2007;85:24. Vilela Filho O, Silva DJ, Souza JT, Ferraz FAP, Cavalcante JES. STN lesioning-induced dyskinesia in patients with Parkinson’s disease: What is behind it? In Abstract Book of the 2003 Quadrennial Meeting of the American Society for Stereotactic and Functional Neurosurgery. New York: New York University Medical School; 2003. p. 198. Vilela Filho O, Silva DJ, Ferraz FP. Evidence for the involvement of the basal ganglia in the genesis of essential tremor (ET) (Abstr.). Stereotact Funct Neurosurg 2001;77:149-50. Vilela Filho O, Ferraz FAP, Barros BA, Barreto PG, Nobrega MDA, Atayde IB, et al. Effects of Unilateral
History of stereotactic and functional neurosurgery in brazil
121.
122.
123.
124.
125.
126.
127.
128. 129.
130.
131.
132.
Posterior Striatotomy on Harmaline-induced Tremor in Rats (Abstr.). Stereotact Funct Neurosurg 2007;85: 29-30. Vilela Filho O, Souza HAO, Cavalcante JES, Dias PCM, Leite MSB. Bio´psia percutaˆnea assistida por tomografia computadorizada de tumores intramedulares (Abstr.). Programa Final do VII Congresso Nacional da Academia Brasileira de Neurocirurgia. Goiaˆnia: Cegraf; 1997. p. 93. Vilela Filho O, Botter W, Jr, Oliveira SB, Teixeira KS. OVF MRI fiducial: a new and long-lasting material for stereotactic MRI (Abstr.). Stereotact Funct Neurosurg 2001;76:191-2. Vilela Filho O, Carneiro Filho O. Gamma probeassisted brain tumor microsurgical resection: a new technique. Arq Neuropsiquiatr 2002;60(4):1042-37. Vilela Filho O, Carneiro Filho O, Arratia JIC, Ragazzo PC, Arruda JB, Costa V, et al. Gamma probe assisted epilepsy surgery: preliminary results of a new technique (Abstr.). Stereotact Funct Neurosurg 2007;85:46. Vilela Filho O, Carneiro R, Jr. MRI-based frameless technique to determine target coordinates (Abstr.). Stereotact Funct Neurosurg 2007;85:62. Pereira LC, Palter VN, Lang AE, Hutchison WD, Lozano AM, Dostrovsky JO. Neuronal activity in the globus pallidus of multiple system atrophy patients. Mov Disord 2004;19(12):1485-92. Saint-Cyr JA, Hoque T, Pereira LC, Dostrovsky JO, Hutchison WD, Mikulis DJ, et al. Localization of clinically effective stimulating electrodes in the human subthalamic nucleus on magnetic resonance imaging. J Neurosurg 2002;97(5):1152-66. Aguiar LR. Biografia do Prof. Dr. Renato de Muggiati (1932–1996). J Bras Neurocirurg 1996;7(3): 56-7. Muggiati R, Ratton JF, Machado S. Resultado sobre 100 pacientes portadores de sı´ndrome de Parkinson tratados cirurgicamente pela talamotomia (Abstr.). In Mattos Pimenta A (Ed). XI Congresso Latino-Americano de Neurocirurgia (Relato´rios Oficiais e Comunicac¸o˜es Livres). Sa˜o Paulo; 1965. p. 272. Francisco AN, Loba˜o CA, Sassaki VS, Garbossa MC, Aguiar LR. Punctate midline myelotomy for the treatment of oncologic visceral pain: analysis of three cases]. Arq Neuropsiquiatr 2006;64(2B):446-50. (Portuguese). Meneses MS, Rocha SB, Kowacs PA, Andrade NO, Santos HL, Narata AP, et al. [Surgical treatment of temporal lobe epilepsy: a series of forty-three cases analysis]. Arq Neuropsiquiatr 2005;63(3A):618-24. (Portuguese). Meneses MS, Rocha SF, Blood MR, Trentin A, Jr, Benites Filho PR, Kowacs PA, et al. [Functional magnetic resonance imaging in the determination of dominant language cerebral area]. Arq Neuropsiquiatr 2004;62(1):61-7. (Portuguese).
16
133. Meneses MS, Follador FR, Arruda WO, Santos HL, Yonesawa D, Hunhevicz SC. [Stereotactic implantation of depth electrodes by magnetic resonance image in epilepsy surgery]. Arq Neuropsiquiatr 1999;57 (3A):628-35. (Portuguese). 134. Meneses MS, Arruda WO, Hunhevicz SC, Ramina R, Pedrozo AA, Tsubouchi MH. Comparison of MRIguided and ventriculography-based stereotactic surgery for Parkinson’s disease. Arq Neuropsiquiatr 1997;55 (3B):547-52. 135. Almeida DB, Poletto PH, Milano JB, Leal AG, Ramina R. Is preoperative occupation related to longterm pain in patients operated for lumbar disc herniation? Arq Neuropsiquiatr 2007;65(3B):758-63. 136. Almeida DB, Hunhevicz S, Bordignon K, Barros E, Mehl AA, Burak Mehl AC, et al. A model for foramen ovale puncture training: Technical note. Acta Neurochir (Wien) 2006;148(8):881-3. 137. Salamoni SD, da Costa JC, Palma MS, Konno K, Nihei K, Azambuja NA, et al. The antiepileptic activity of JSTX-3 is mediated by N-methyl-D-aspartate receptors in human hippocampal neurons. Neuroreport 2005;16 (16):1869-73. 138. Paglioli E, Palmini A, Portuguez M, Paglioli E, Azambuja N, da Costa JC, et al. Seizure and memory outcome following temporal lobe surgery: selective compared with nonselective approaches for hippocampal sclerosis. J Neurosurg 2006;104(1):70-8. 139. Isolan GR, Azambuja N, Paglioli Neto E, Paglioli E. [Hippocampal microsurgical anatomy regarding the selective amygdalohippocampectomy in the Niemeyer’s technique perspective and preoperative method to maximize the corticotomy]. Arq Neuropsiquiatr 2007;65 (4A):1062-9. 140. Ferreira NP, Azambuja NA. [Percutaneous cordotomy]. Arq Neuropsiquiatr 1980;38(1):7-17. 141. Linhares MN, Tasker RR. Microelectrode-guided thalamotomy for Parkinson’s disease. Neurosurgery 2000;46 (2):390-5. 142. Dieckmann G, Veras G. High frequency coagulation of dorsal root entry zone in patients with deafferentation pain. Acta Neurochir 1984; Suppl 33:445–50. 143. Brainer-Lima PT, Rao S, Cukiert A, Yacubian EM, Gronich G, Marino Ju´nior R. Surgical treatment of refractory epilepsy associated with space occupying lesions. Experience and review. Arq Neuropsiquiatr 1996;54(3):384-92. 144. Brainer-Lima PT, Brainer-Lima AM, Azevedo Filho H, Cukiert A. [Partial epilepsy associated to primary brain tumors]. Arq Neuropsiquiatr 2002;60(3-B):797-800. (Portuguese). 145. Brainer-Lima PT, Brainer-Lima AM, Brandt CT, Carneiro GS, Azevedo HC. [Intraoperative mapping of motor areas during brain tumor surgery: electrical stimulation patterns]. Arq Neuropsiquiatr 2005;63 (1):55-60. (Portuguese).
245
246
16
History of stereotactic and functional neurosurgery in brazil
146. Andrade-Souza YM, Zadeh G, Ramani M, Scora D, Tsao MN, Schwartz ML. Testing the radiosurgerybased arteriovenous malformation score and the modified Spetzler-Martin grading system to predict radiosurgical outcome. J Neurosurg 2005;103(4):642-8. 147. Andrade-Souza YM, Schwalb JM, Hamani C, Hoque T, Saint-Cyr J, Lozano AM. Comparison of 2-dimensional magnetic resonance imaging and 3-planar reconstruction methods for targeting the subthalamic nucleus in Parkinson disease. Surg Neurol 2005;63(4):357-62. 148. Andrade-Souza YM, Schwalb JM, Hamani C, Eltahawy H, Hoque T, Saint-Cyr J, et al. Comparison of three methods of targeting the subthalamic nucleus for chronic stimulation in Parkinson’s disease. Neurosurgery 2005;56 (2 Suppl):360-8. 149. Andrade-Souza YM, Zadeh G, Scora D, Tsao MN, Schwartz ML. Radiosurgery for basal ganglia, internal capsule, and thalamus arteriovenous malformation: clinical outcome. Neurosurgery 2005;56(1):56-63. 150. Andrade-Souza YM, Ramani M, Scora D, Tsao MN, TerBrugge K, Schwartz ML. Radiosurgical treatment for rolandic arteriovenous malformations. J Neurosurg 2006;105(5):689-97. 151. Andrade-Souza YM, Ramani M, Scora D, Tsao MN, terBrugge K, Schwartz ML. Embolization before radiosurgery reduces the obliteration rate of arteriovenous malformations. Neurosurgery 2007;60(3):443-51. 152. Steven DA, Andrade-Souza YM, Burneo JG, McLachlan RS, Parrent AG. Insertion of subdural strip electrodes for the investigation of temporal lobe epilepsy. Technical note. J Neurosurg 2007;106(6):1102-6. 153. Andrade-Souza YM, Ramani M, Beachey DJ, Scora D, Tsao MN, Terbrugge K, et al. Liquid embolisation material reduces the delivered radiation dose: a physical experiment. Acta Neurochir (Wien) 2008;150(2):161-4. 154. Holanda MM, Meira UM, Magalha˜es FN, da Silva JA. [Surgical treatment of meralgia paresthetica: case report]. Arq Neuropsiquiatr 2003;61(2A):288-90. (Portuguese). 155. Blond S, Touzet G, Reyns N, Dantas S, Pruvo JP. [Clinical applications of stereotaxic methology]. Ann Fr Anesth Reanim 2002;21(2):162-9. (French). 156. De Salles AA, Katayama Y, Becker DP, Hayes RL. Pain suppression induced by electrical stimulation of the pontine parabrachial region. Experimental study in cats. J Neurosurg 1985;62:397-407. 157. Leichnetz GR, Gonzalo-Ruiz A, De Salles AAF, Hayes RL. The frontal eye field and prefrontal cortex project to the paramedian pontine reticular formation in the cat. Brain Res 1987;416:195-9. 158. Leichnetz GR, Gonzalo-Ruiz A, De Salles AAF, Hayes RL. The origin of brainstem afferents of the paramedian pontine reticular formation in the cat. Brain Res 1987;422:389-97. 159. De Salles AAF, Asfora WT, Abe M, Kjelberg RN. Transposition of target information from the magnetic
160.
161.
162.
163.
164.
165.
166.
167.
168.
169.
170.
171.
172.
resonance and CT-scan images to the conventional x-ray stereotactic space. Appl Neurophysiol 1987; 50:23-32. Anzai Y, Lufkin R, De Salles AAF, Hamilton DR, Farahani K, Black KL. Preliminary experience with MR-guided thermal ablation of brain tumors. AJNR Am J Neuroradiol 1995;16:39-48. De Salles AAF, Frighetto L, Behnke E, Sinha S, Bronstein J, Torres R, et al. Functional neurosurgery in the MRI environment. Minim Invasive Neurosurg 2004;47(5):284-9. Lee WY M, De Salles AAF, Frighetto L, Torres R, Behnke E, Bronstein J. Imaging techniques and electrode fixation methods for deep brain stimulation – Intraoperative MRI 0.2T, 1.5 T and fluoroscopy. Minim Invas Neurosurg 2005;48:1-6. De Salles AAF, Solberg T, Medin P, Vassilev V, Cabatan-Awang C, Selch M. Linear accelerator radiosurgery for trigeminal neuralgia. Radiosurgery 1997;2:173-82. De Salles AAF, Melega WP, Lacan G, Steele LH, Solberg TD. Radiosurgery with a 3 mm collimator in the subthalamic nucleus and substantia nigra of the vervet monkey. J Neurosurg 2001;95:990-7. De Salles AAF, Gorgulho A, Selch M, De Marco J, Agazaryan N. Radiosurgery from the brain to the spine: 20 years experience. Acta Neurochir Suppl 2008;101:163-8. Lacan G, De Salles AA, Gorgulho AA, Krahl SE, Frighetto L, Behnke EJ, et al. Modulation of food intake following deep brain stimulation of the ventromedial hypothalamus in the vervet monkey. Laboratory investigation. J Neurosurg 2008;108(2):336-42. Hamani C, Teno´rio F, Mendez-Otero R, Mello LE. Loss of NADPH diaphorase-positive neurons in the hippocampal formation of chronic pilocarpine-epileptic rats. Hippocampus 1999;9(3):303-13. Hamani C, Mello LE. Status epilepticus induced by pilocarpine and picrotoxin. Epilepsy Res 1997;28 (1):73-82. Hamani C, Luer MS, Dujovny M. Microdialysis in the human brain: review of its applications. Neurol Res 1997;19(3):281-8. Andrade DM, Zumsteg D, Hamani C, Hodaie M, Sarkissian S, Lozano AM, et al. Long-term follow-up of patients with thalamic deep brain stimulation for epilepsy. Neurology 2006;66(10):1571-3. Hamani C, Hodaie M, Chiang J, del Campo M, Andrade DM, Sherman D, et al. Deep brain stimulation of the anterior nucleus of the thalamus: effects of electrical stimulation on pilocarpine-induced seizures and status epilepticus. Epilepsy Res 2008;78(2–3):117-23. Hamani C, McAndrews MP, Cohn M, Oh M, Zumsteg D, Shapiro CM, et al. Memory enhancement induced by hypothalamic/fornix deep brain stimulation. Ann Neurol 2008;63(1):119-23.
History of stereotactic and functional neurosurgery in brazil
173. Weinberger M, Hamani C, Hutchinson WD, Moro E, Lozano AM, Dostrovsky JO. Pedunculopontine nucleus microelectrode recordings in movement disorder patients. Exp Brain Res 2008;188(2):165-74. 174. Toda H, Hamani C, Fawcett AP, Hutchison WD, Lozano AM. The regulation of adult rodent hippocampal neurogenesis by deep brain stimulation. J Neurosurg 2008;108(1):132-8. 175. Hamani C, Saint-Cyr JA, Fraser J, Kaplitt M, Lozano AM. The subthalamic nucleus in the context of movement disorders. Brain 2004;127(Pt 1):4-20. 176. Hamani C, Richter E, Schwalb JM, Lozano AM. Bilateral subthalamic nucleus stimulation for Parkinson’s disease: a systematic review of the clinical literature. Neurosurgery 2005;56(6):1313-21. 177. Hamani C, Neimat J, Lozano AM. Deep brain stimulation for the treatment of Parkinson’s disease. J Neural Transm Suppl 2006;(70):393–9. 178. Hung SW, Hamani C, Lozano AM, Poon YY, Piboolnurak P, Miyasaki JM, et al. Long-term outcome of bilateral pallidal deep brain stimulation for primary cervical dystonia. Neurology 2007;68(6):457-9. 179. Hamani C, Schwalb JM, Rezai AR, Dostrovsky JO, Davis KD, Lozano AM. Deep brain stimulation for chronic neuropathic pain: long-term outcome and the incidence of insertional effect. Pain 2006;125 (1–2):188-96. 180. Richter EO, Davis KD, Hamani C, Hutchison WD, Dostrovsky JO, Lozano AM. Cingulotomy for psychiatric disease: microelectrode guidance, a callosal reference system for documenting lesion location, and clinical results. Neurosurgery 2004;54(3):622-28.
16
181. Machado AG, Baker K, Godwin E, Schuster D, Rezai AR. Electrical stimulation of the cerebellar dentate nucleus enhances recovery of motor function following adult ischemic strokes in adult rats. San Diego; SFN, 2007. 182. Machado AG, Azmi H, Rezai AR. Motor cortex stimulation for chronic benign pain. Clin Neurosurg 2005;54:70-7. 183. Machado A, Ogrin M, Rosenow JM, Henderson JM. A 12-month prospective study of gasserian ganglion stimulation for trigeminal neuropathic pain. Stereotact Funct Neurosurg 2007;85(5):216-24. 184. Machado A, Rezai AR, Kopell BH, Gross RE, Sharan AD, Benabid AL. Deep brain stimulation for Parkinson’s disease: surgical technique and perioperative management. Mov Disord 2006;21 Suppl 14: S247-58. 185. Rezai AR, Machado AG, Deogaonkar M, Azmi H, Kubu C, Boulis NM. Surgery for movement disorders. Neurosurgery 2008;62 Suppl 2:SHC809-39. 186. Kopell BH, Machado AG, Rezai AR. Not your father’s lobotomy: psychiatric surgery revisited. Clin Neurosurg 2005;52:315-30. 187. Machado AG, Malone D, Kopell BH, Greenberg B, Rezai AR. Deep brain stimulation for the treatment of refractory obsessive compulsive disorder: assessment of optimal therapeutic cathode location. San Francisco; Gilbenberg Award. AANS Meeting, 2006. 188. Schiff ND, Giacino JT, Kalmar K, Victor JD, Baker K, Gerber M, et al. Behavioural improvements with thalamic stimulation after severe traumatic brain injury. Nature 2007;448(7153):600-3.
247
10 History of Stereotactic and Functional Neurosurgery in Canada A. G. Parrent
Introduction Functional neurosurgery in its broadest sense includes the surgical treatment of pain, movement disorders, epilepsy and psychiatric conditions. Canada has played a rich and pivotal role in the development of these areas. Wilder Penfield, Herbert Jasper, Ted Rasmussen, Claude Bertrand and Ron Tasker – to mention a few – are names that are familiar to most neurosurgeons who are involved in the practice of functional neurosurgery. This chapter will review some highlights in the development of functional neurosurgery in Canada with an emphasis on stereotactic surgery and epilepsy surgery.
Epilepsy Surgery and the Montreal Neurological Institute In 1928 Edward Archibald, professor of surgery at McGill University, brought Wilder Penfield and William Cone to Montreal, to develop neurosurgery at McGill University. Their clinical work started at the Royal Victoria Hospital and the Montreal General Hospital in Quebec [1]. Penfield [2] had wanted to establish an institute for the scientific study and treatment of neurological disorders and in 1932, the Rockefeller Foundation granted 1.2 million dollars to McGill University for this purpose. This money along with contributions from the city, province and other donors allowed construction of the Montreal Neurological Institute (MNI), which opened #
Springer-Verlag Berlin/Heidelberg 2009
in 1934 as a 50 bed hospital for the investigation and treatment of brain disorders. Penfield’s background included study with Sherrington and Osler at Oxford, Holmes and Greenfield at Queen Square, Cajal and Hortega in Madrid, and Foerster in Breslau. With Foerster he learned the cortical stimulation and excision techniques that he would apply to the treatment of epilepsy. Penfield carried out his first procedure for focal epilepsy at the Royal Victoria Hospital in November 1928, on a young man with posttraumatic epilepsy resulting from a head injury with subdural hematoma and brain contusion after a fall from a horse. At the first operation the motor strip was identified by cortical stimulation and a small adjacent corticectomy was performed. Seizures continued and this man underwent a total of three epilepsy procedures culminating in a wide temporal resection – Penfield’s first temporal lobe resection for epilepsy [3]. Penfield meticulously studied his patients, both clinically and in the operating room and kept detailed records of the results of cortical stimulation. In 1937, Penfield and Boldrey [4] published their results of cortical stimulation mapping in 163 patients marking the first appearance of their motor-sensory homunculus (> Figure 10‐1a, b). Many of our current concepts of cortical localization are based on Penfield’s work [5]. By the early 1950s, Penfield [6] had recognized the unique nature of temporal lobe epilepsy, and introduced the concept of mesial temporal
114
10
History of stereotactic and functional neurosurgery in canada
. Figure 10‐1 (a) Penfield and his surgical team operating on an epilepsy patient who is awake under local anesthesia, in order to identify the epileptic focus by brain mapping. Herbert Jasper, in the glazed gallery, records the electrocorticogram (from Brain vol. 130, figure 10, 2007). (b) The ‘‘homunculus’’ showing areas of cortex related to body movements as evoked by stimulation of the cortex (from Penfield and Boldrey [4])
sclerosis; at that time attributed to herniation of the mesial temporal structures over the edge of the tentorium during difficult childbirth. Penfield and Jasper’s book ‘‘Epilepsy and the functional anatomy of the human brain’’ was published in 1954 and represented a wealth of information on the clinical presentation of epilepsy, observation related to memory, localization of language and surgical aspects of epilepsy treatment [7] (> Figure 10‐2). Penfield was responsible for the recognition of epilepsy as a surgically treatable disease. He utilized electrical stimulation of the cortex in patients undergoing craniotomy under local anesthetic and thereby improved the safety of cortical excisions near eloquent cortex. His technique of subpial dissection resulted in less residual damage and gliosis after corticectomy. This is now a standard technique for epilepsy surgeons. Penfield [1] was Director of the MNI from 1934 to 1960, succeeded by Theodore Rasmussen from 1960 to1972. Rasmussen became Professor
. Figure 10‐2 Penfield and Jasper [7] with manuscript of their book ‘‘Epilepsy and the Functional Anatomy of the Human Brain’’(from Can J Neurol Sci18:540 figure 6)
and Chairman of Neurology and Neurosurgery at the MNI in 1954, coming from the University of Chicago where he was had been Professor of Surgery since 1947 [8].
History of stereotactic and functional neurosurgery in canada
Rasmussen [9–11] added to the MNI’s contributions to the surgical treatment of epilepsy. In his years there he probably performed more operations for epilepsy than any other surgeon of his time. These cases were meticulously followed and his outcome publications report some of the longest follow-ups of epilepsy surgery patients. Rasmussen [12,13] together with Juhn Wada validated the application of the carotid amytal test for lateralization of language function and extended its value by using it for the assessment of memory in candidates for temporal lobe surgery. This is still an important means of testing memory adequacy and reserve in candidates for temporal lobectomy. Rasmussen’s name became attached to the entity of chronic localized encephalitis associated with intractable focal seizures (Rasmussen’s encephalitis); first described by Rasmussen et al. [14] in 1958. Rasmussen and others [15–17] described the condition of cerebral hemisiderosis
10
as a late complication of anatomical hemispherectomy, and introduced functional hemispherectomy as a strategy for preventing it. The MNI continues in its role, not only in the surgical treatment of epilepsy, but also in its role as educator with the training of numerous Canadian and international neurosurgeons, and neurosurgical fellows in these techniques.
Stereotactic and Functional Neurosurgery at the MNI Gilles Bertrand and John Blundell started the functional stereotactic program at the MNI when they performed the first stereotactic pallidotomy in 1958 and the first thalamotomy in 1959 [18]. Gill Bertrand started at the MNI as a resident in 1951 (> Figure 10‐3). John Blundell was invited to join the MNI group in 1957. He was sent to study with Lars Leksell in Sweden and to bring
. Figure 10‐3 (a) Dr. Gilles Bertrand inserting a microelectrode into a patient with Parkinson’s disease. (from Can J Neurol Sci 14:542 figure 8). (b) Gilles Bertrand (left) with Herbert Jasper (right) in 1988. They collaborated from 1964 to 1966 on pioneer studies of microelectrodes recording from basal ganglia neurons in patients undergoing stereotactic surgery for Parkinson’s disease. (from Can J Neurol Sci 26:228 figure 8)
115
116
10
History of stereotactic and functional neurosurgery in canada
. Figure 10‐4 The Jasper and Hunter stereotactic instrument. (from Neurosurgery 2004, 54(5):1246 figure 3)
back a Leksell stereotactic instrument. Prior to this, Drs. Herbert Jasper and John Hunter had constructed a stereotactic frame for human use – but this was never used for clinical purposes. This frame was designed to mount on the operating table with the patient in a sitting position. It was held in place by earplugs, orbital bars and a plate under the upper teeth rather than by skull pins (> Figure 10‐4). They settled on the leukotome of Claude Bertrand for lesion production after studying thermocoagulative lesions in animals and finding great variability of lesion size with identical lesion parameters. At the suggestion of Jasper in 1963 they started using microelectrode recording, and the group adopted a system with a straight microelectrode and one with a curved side arm to record from areas around the central exploring electrode (> Figure 10‐5). This group was one of the first to develop a computer program to assist with operative mapping. It allowed operative stereotactic data to be mapped onto digitized versions of various brain atlases that were expanded or shrunk to match the intercommissural distances of the respective patients [19,20]. Oblique trajectories could be followed interactively through the different
. Figure 10‐5 Leukotome (far left) used for lesion making, curved stimulating electrode (second from left) for performing macrostimulation, three-step electrode (center) used for macrostimulation and evoked potentials, straight and curved tungsten microelectrodes (right). (from Neurosurgery 2004, 54(4):1246 figure 4)
parasagittal planes, and operative physiologic data could be registered to stimulation points for later reference. The leukotome was modeled on the computer screen to allow lesions to be accurately planned and mapped [21,22]. As with all neurosurgical centers, the MNI saw a decline in the number of stereotactic procedures performed for Parkinson’s disease after the introduction of L-dopa. Surgery for other
History of stereotactic and functional neurosurgery in canada
movement disorders patients, mainly essential tremor and dystonia, continued. Abbas Sadikot joined Gilles Bertrand in 1993 and they worked together for a year prior to Bertrand’s retirement from clinical practice. The MNI group continues it work on computerized brain atlases for use in functional neurosurgery. The Leksell stereotactic frame, which was used for their stereotactic procedures, underwent modifications over time. Initial modifications by Bertrand allowed x-rays to be taken through the frame. Later, Andre Olivier made changes to facilitate the insertion of orthogonal electrodes and to make it compatible with angiography and MRI. This became the Olivier–Bertrand–Tipal or OBT frame (OBT frame, Tipal Instruments, Montreal).
10
. Figure 10‐6 Photograph of Claude Bertrand’s pneumotaxic guide with lineated prism to avoid parallax while taking preliminary measurements outside the patient’s head before control x-ray. Also, on the guiding bar, some of the collars could be angulated to allow penetration at an angle or phantom measurements. (from Neurosurgery 2004, 55(3):699 figure 2)
Functional Neurosurgery in Montreal – Hoˆpital Noˆtre-Dame Claude Bertrand was one of the pioneers of stereotactic and functional neurosurgery in Canada. Born in 1917, he obtained his BA (1934), then MD (1940) at the Universite´ de Montre´al. Neurosurgical training at the Montreal Neurological Institute was completed in1946. In 1947 Claude Bertrand established the neurosurgical department at Hoˆpital Noˆtre-Dame in Montreal. Very early in his career he was interested in the treatment of involuntary movement disorders and performed anterior choroidal artery ligation on a number of patients, following Irving Cooper’s lead. In 1953, he carried out the first stereotactic lesioning (pallidotomy) for Parkinson’s disease in Canada. Although stereotactic frames designed by Speigel and Wycis, Riechert and Leksell were available; they usually required general anesthetic for application and Bertrand was looking for something better. He designed a stereotactic device called a pneumotaxic guide that could be applied under local anesthetic, and was used for his procedures [23] (> Figure 10‐6). Unsatisfied
with the inconsistency of lesions produced by cauterization or injection of substances into the brain, he used a leukotome to produce lesions. The original Moniz leukotome with a sharp blade was fairly quickly replaced with a leukotome made with blunt fine piano wire after an intracerebral hemorrhage occurred during one of his procedures [24]. The foramen of Munro was initially used as the anterior landmark for localization but Bertrand was convinced by Guiot to change to the anterior commissure. Influenced by Riechert’s reports [25], Bertrand, like many of his contemporaries moved from the pallidum to the thalamus as the target for Parkinsonian tremor. Bertrand was joined by Sonis N. Martinez in 1956, Jules Hardy in 1962 and Pedro MolinaNegro in 1967. Jules Hardy had completed his neurosurgical training in 1960 and took advantage of a McLaughlin traveling fellowship to work with
117
118
10
History of stereotactic and functional neurosurgery in canada
Gerard Guiot and Denise Albe-Fessard in Paris. There they carried out the some of the first microelectrode recordings in Parkinson’s patients undergoing stereotactic surgery. Following his return to Montreal in 1962, he introduced these techniques to the group and microelectrode recording was used routinely for their functional neurosurgical procedures [24,26] (> Figure 10‐3a). Molina-Negro was the neurophysiologist in the group and involved in the analysis of recordings obtained during stereotactic procedures. He suggested using a proportional measurement system to compensate for variations of individual patient anatomy, and they adopted one in which the AC-PC line was divided into ten equal parts. Using this system, a subthalamic, prelemniscal area target could be delineated with such accuracy that the mere introduction of an electrode was sufficient to arrest tremor [27,28]. One of the major contributions to functional neurosurgery coming from this group was the inception and development of the technique of selective peripheral denervation for the treatment of torticollis [48]. They were unhappy with the results of stereotactic thalamic interventions and wondered whether peripheral denervation could improve the results [18]. They noted that significant though temporary improvement could be produced when active muscles were injected with 1% lidocaine under EMG guidance. Their results of combined thalamotomy and peripheral denervation were reported in 1978 [29]. But they ultimately found that peripheral denervation alone was quite effective for treating torticollis.
The Toronto School The Toronto Neurosurgical program started with Kenneth McKenzie, who broke off from general practice in 1923 to take up a fellowship at the Peter Brent Brigham in Boston. While there he published an article describing sectioning of the anterior cervical spinal nerves along with the
spinal accessory for the treatment of torticollis – a procedure we still call the McKenzie procedure in Canada [30]. He returned to Toronto in 1924 as a surgical resident and went on from there to develop neurosurgery at the Toronto General Hospital. After his retirement in 1952, Dr. Harry Botterell took over the program until 1963 when he left to become Dean of Medicine at Queen’s University in Kingston. The program was then taken over by Dr. Tom Morley. Dr. Ron Tasker (> Figure 10‐7a, b) was a trainee of the Toronto Neurosurgical Program, obtaining his Royal College Fellowship in Neurosurgery in 1959. Tasker spent the following year with Clinton Woolsey at the University of Wisconsin’s Laboratory of Neurophysiology. Woolsey was studying the evolution of the sensorimotor cortex, and carried out cortical mapping in a variety of animals. Tasker developed a great deal of admiration for the ‘simple expressive beauty’ of the figurine charts that were used to represent the results of cortical mapping. Additional time was spent at the National Hospital in Queen Square, London (as was the tradition for Neurosurgical trainees at that time) and in Europe observing Albe-Fessard, Guiot, Riechert, Leksell, Talairach and others. Tasker returned to Toronto September 1961 and started his neurosurgical practice. At that time, functional neurosurgery in Toronto consisted of open cordotomy, the Frazier operation for tic douloureux, open dorsal rhizotomy for pain and the McKenzie operation for torticollis. At that time, stereotactic surgery was used primarily for the treatment of tremor associated with Parkinson’s disease. Pallidotomy was the favored procedure, carried out using ventriculographic guidance and macrostimulation to identify internal capsule. Early on, lesions were made by the inflation of a balloon and instillation of alcohol, or by cryolesioning. Tasker (May 2008, personal communication) rapidly adopted radiofrequency lesioning after its introduction in the early 1960s.
History of stereotactic and functional neurosurgery in canada
10
. Figure 10‐7 (a) Ron Tasker performing stereotactic surgery with one of the earlier Leksell steterotactic frames (b) Tasker (right) with Hirotaro Narabayashi from Japan
Tasker followed Riechert’s lead in moving the tremor target from pallidum to thalamus sometime in the early 1960s [25]. Tasker espoused the philosophy that the stereotactic operating room offered an opportunity to study the physiology of the brain. He established early collaboration with Raimond Emmers, a physiologist at the College of Physicians and Surgeons of Columbia University and later Leslie Organ of the Department of Physiology at the University of Toronto to study the somatosensory pathways of the thalamus and upper midbrain [31]. A computer program was developed to allow mapping of the stimulation responses using Woolsey-type figurines [32]. In 1982, Tasker et al. [33] published this work – ‘‘The Thalamus and Midbrain of Man’’ – which serves as a remarkable body of data and solid reference material for stereotactic and functional neurosurgeons.
Microelectrode mapping was brought into the operating room sometime in the late 1970s. Dostrovsky and Tasker studied the detailed physiology of the motor and sensory thalamus, mentoring numerous masters and PhD students as well as clinical and research fellows. Between 1964 and 1998, 48 such students and fellows were trained by Tasker, many going on to start up stereotactic programs of their own. Tasker took the surgical treatment of movement disorders in Toronto from pallidotomy to thalamotomy then through the reintroduction pallidotomy in the early 1990s. Tasker suggested that one of the great oversights in functional neurosurgery was the failure to recognize that pallidotomy was effective for bradykinesia in Parkinson’s disease. Deep brain stimulation was introduced in the early 1990s following the work of Benabid in Grenoble.
119
120
10
History of stereotactic and functional neurosurgery in canada
Tasker (May 2008, personal communication) was interested in the physiology and treatment of chronic pain. When he started practice pain was categorized as benign or malignant, and there were few operations available for the treatment of pain. Cordotomy was one of the favored procedures, was open and performed bilaterally at T1–2. The other procedure was open mesencephalic tractotomy. One of Tasker’s most significant contributions to pain management was the recognition of the difference between nociceptive pain and deafferentation or neuropathic pain and the breakdown of neuropathic pain into its different components: steady pain, shooting pain and allodynia [34,35]. Tasker has been honored with numerous awards in recognition of his contributions to functional neurosurgery including: the Spiegel & Wycis Medal, awarded by the WSSFN in 1993; the Distinguished Service Award by the ASSFN in 1995; Award for Distinguished Contributions to Pain Research and Management in Canada by the Canadian Pain society in 1997. The Tasker Chair in Functional Neurosurgery was created at the University of Toronto in 1999. In the early 1990s Andres Lozano was recruited to the University of Toronto after completing his neurosurgical training at the MNI. Lozano has expanded the functional neurosurgical program in Toronto and has gone on to be a world leader in stereotactic and functional neurosurgery. Lozano now holds the Tasker Chair at the University of Toronto.
Western Canada Edmonton Howard H. Hepburn started the Neurosurgical program in Edmonton, Alberta in 1920, having completed his medical training at McGill University in 1910. His plans for neurosurgical training in Germany were interrupted by the outbreak of
the first world war. In 1919, he returned to North American and obtained training in neurosurgery at the Mayo Clinic in Rochester. Hepburn was joined in 1945 by Guy Morton who, ultimately assumed leadership of the Neurosurgical Division with Dr. Hepburn’s retirement in 1951. Dr. Thomas Speakman joined the division in 1952. Speakman was trained at the MNI and strongly influenced by Penfield. He started performing stereotactic thalamotomy for Parkinson’s disease in the late 1950s [36] and continued until his sudden death in 1969 at the age of 45.
Saskatchewan Krishna (Kris) Kumar started the functional neurosurgical program in Regina, Saskatchewan in 1961. He came to Canada in 1959 after completing his medical training and surgical residency in Indore, India. He obtained Canadian neurosurgical training under W. D. Stevenson in Halifax and Dr. Morton in Edmonton. In Edmonton, Kumar received training in stereotactic surgery under Dr. Speakman. At the start of his career Dr. Kumar (May 2008, personal communication) was involved in the stereotactic treatment of movement disorders, mainly thalamotomy for Parkinson’s disease. After the introduction of L-dopa and the decline in candidates for movement disorder surgery he concentrated his efforts on the treatment of intractable pain. He was involved in the stereotactic implantation of DBS electrodes in periaqueductal grey (PAG) and sensory thalamus from its start in the late 1960s, and was an advocate of the intravenous pain test for assessing surgical candidacy. This was carried out by injecting small repeated amounts of intravenous morphine to assess the degree of pain relief, and the utilization of naloxone to assess reversibility of the response. Patients responding in a reversible manner to i.v. morphine were considered candidates for PAG stimulation, those not responding were
History of stereotactic and functional neurosurgery in canada
considered better candidates for sensory thalamic stimulation. In general, most patients received simultaneous implantations into both sites followed by percutaneous testing. The most effective electrode was ultimately internalized. The DBS system at the time consisted of a ring electrode activated by a radiofrequencycoupled stimulator (Medtronic). These systems developed slowly; although approved for the treatment of pain in Canada, DBS for pain never obtained FDA approval in the United States. Kumar (May 2008, personal communication) was also involved in spinal stimulation for pain. Starting with the open implantation of intradural and interdural electrodes and through to the epidural systems. The initial percutaneous monopolar electrodes came in straight, sigmashaped and tined varieties. Radiofrequencycoupled systems were eventually supplanted by implanted battery operated systems. Kumar was influential in the development of the fully implantable spinal cord stimulation devices and implantable pumps. Significant contributions to the area of functional neurosurgery include cost-effectiveness studies for spinal cord stimulation and intrathecal drug therapy [37–39], as well as long-term follow-up studies of SCS and DBS for pain [40–42].
British Columbia Frank Turnbull was one of the pioneers of neurosurgery in Canada and the first neurosurgeon in British Columbia. He received his neurosurgical training under K. McKenzie in Toronto, followed by postgraduate work at Queen Square in London, and further work with Foerster in Breslau. He started work at the Vancouver General Hospital in 1933 and devoted special attention to the surgical treatment of intractable pain. Turnbull [43–45] attempted to look at pain syndromes as distinct entities so as to better rationalize management. He published on the pain syndromes associated with
10
pelvic carcinoma and specifically looked at the role of cordotomy in the surgical management of these pain syndromes [45]. The early cordotomies were open procedures, carried out at T1–2. In the mid to late 1960s the switch was made to anterior cervical percutaneous cordotomy. Peter Lehmann joined Frank Turnbull in the late 1940s and started performing thalamotomy for parkinsonian tremor in the late 1950s using the Cooper apparatus and leukotome. Ian Turnbull joined the group in 1966. Ian Turnbull (July 2008, personal communication) obtained his medical training in Vancouver and neurosurgical training in Toronto. While in Toronto he was exposed to Tasker and his work. Residency was followed by a traveling fellowship in Europe gaining exposure to Leksell, Gillingham and others. After his return to Vancouver he rapidly adopted the Todd Wells stereotactic apparatus and used radiofrequency to carry out thalamotomy for tremor disorders. The stereotactic movement disorder program remained active through to his retirement from clinical practice in 1999. In the early 1990s he, like most neurosurgeons switched to pallidotomy for Parkinson’s disease. Ian Turnbull also established and maintained a large pain practice, initially learning the percutaneous cordotomy technique from his father Frank. He favored stimulation of the ventrobasal thalamus for neuropathic pain [46] and reported on combined thalamic or midbrain lesions with cingulumotomy in some patients with nociceptive pain [46].
The Canadian Stereotactic and Functional Neurosurgery Group In 2001 the Canadian Neurosurgical Society formed a Section of Stereotactic and Functional Neurosurgery. The Canadian group is small with only 15 or so neurosurgeons carrying out functional stereotactic procedures in the country. As of 2008, there are active programs in Halifax Nova Scotia,
121
122
10
History of stereotactic and functional neurosurgery in canada
Quebec City and Montreal in Quebec, Toronto and London Ontario, Winnipeg Manitoba, Regina Saskatchewan, Calgary and Edmonton Alberta and Vancouver British Columbia. The relatively small size of the group has been conducive to collaborative work and has resulted in a multicentre study of DBS for cervical dystonia [47]. Other projects are ongoing.
References 1. Feindel W. Development of surgical therapy for epilepsy at the Montreal Neurological Institute. Can J Neurol Sci 1991;18:549-53. 2. Penfield W. No man alone: a neurosurgeon’s life. Boston: Little, Brown & Co; 1977. 3. Foerster O, Penfield W. The structural basis of traumatic epilepsy and results of radical operation. Brain 1930;53 (2):99-119. 4. Penfield W, Boldrey E. Somatic motor and sensory representation in cerebral cortex of man as studied by electrical stimulation. Brain 1937;60(4):389-443. 5. Penfield W, Erickson TC. Epilepsy and cerebral localization. Springfield, IL: Charles C Thomas; 1941. 6. Penfield W, Earle KM, Baldwin M. Incisural sclerosis and temporal lobe seizures produced by hippocampal herniation at birth. Arch Neurol Psychiatry1953;69:17-42. 7. Penfield W, Jasper H. Epilepsy and the functional anatomy of the human brain. Boston: Little, Brown and Company; 1954. 8. Feindel W. Theodore Brown Rasmussen (1910–2002): epilepsy surgeon, scientist and teacher. J Neurosurg 2003;98:631-7. 9. Rasmussen T. Surgical treatment of complex partial seizures: results, lessons and problems. Epilepsia 1983;24 Suppl 1:S65-76. 10. Feindel W, Rasmussen T. Temporal lobectomy with amygdalectomy and minimal hippocampal resection: review of 100 cases. Can J Neurol Sci 1991;18 Suppl 4:603-5. 11. Rasmussen T, Feindel W. Temporal lobectomy: review of 100 cases with major hippocampectomy. Can J Neurol Sci 1991;18 Suppl 4:601-2. 12. Wada J, Rasmussen T. Intracarotid injection of sodium amytal for the lateralization of cerebral speech dominance. Experimental and clinical observations. J Neurosurg 1960;17:266-82. 13. Milner B, Branch C, Rasmussen T. Study of short-term memory after intracarotid injection of sodium amytal. Trans Am Neurol Assoc 1962;87:224-6. 14. Rasmussen T, Olszewski J, Lloyd-Smith D. Focal seizures due to chronic localized encephalitis. Neurology 1958;8: 435-45.
15. Rasmussen T. Postoperative superficial hemosiderosis of the brain, its diagnosis, treatment and prevention. Trans Am Neurol Assoc 1973;98:133-7. 16. Rasmussen T. Hemispherectomy for seizures revisited. Can J Neurol Sci 1983;10:71-8. 17. Falconer MA, Wilson PJ. Complications related to delayed hemorrhage after hemispherectomy. J Neurosurg 1969;30(4):413-26. 18. Bertrand G. Stereotactic surgery at McGill: the early years. Neurosurgery 2004;54:1244-52. 19. Bertrand G, Olivier A, Thompson CJ. Computer display of stereotaxic brain maps and probe tracts. Acta Neurochir Suppl (Wien) 1974;21:235-43. 20. Thompson CJ, Bertrand G. A computer program to aid the neurosurgeon to locate probes used during stereotactic surgery of deep cerebral structures. Comput Programs Biomed 1972;2:265-76. 21. Bertrand G. Computers in functional neurosurgery. In: Rasmussen T, Marino R Jr, editors. Functional neurosurgery. NY: Raven; 1979. 22. Thompson CJ, Bertrand G. A computer program to aid the neurosurgeon to locate probes used during stereotaxic surgery on deep cerebral structures. Comput Programs Biomed 1972;2(4):265-76. 23. Bertrand CM. A pneumotaxic technique for producing localized cerebral lesions, and its use in the treatment of Parkinson’s disease. J Neurosurg 1958;15 (3):251-64. 24. Bertrand CM. Surgery of involuntary movements, particularly stereotactic surgery: reminences. Neurosurgery 2004;55:698-704. 25. Hassler R, Riechert T. lndikationen und lolalisationcmethode der gezielton hirnoperationen. Nervenarzt 1954; 25:441-7. 26. Hardy J. Historical background of stereotactic surgery: reflections on stereotactic surgery and the introduction of microelectrode recording in Montreal. Neurosurgery 2004;54:1508-11. 27. Velasco F, Molina-Negro P, Bertrand C, Hardy J. Further definition of the sub-thalamic target for the arrest of tremor. J Neurosurg 1972;36:184-91. 28. Bertrand C, Hardy J, Molina-Negro P, Martinez SN. Optimum physiological target for the arrest of tremor. In 3rd Symposium on Parkinson’s Disease, Edinburgh, Livingstone, 1969, 251-9. 29. Bertrand C, Molina-Negro P, Martinez SN. Combinedstereotactic and peripheral surgical approach for spasmodic torticollis. Appl Neurophysiol 1978;41: 122-33. 30. McKenzie KG. Intrameningeal division of the spinal accessory and roots of the upper cervical nerves for the treatment of spasmodic torticollis. Surg Gynecol Obstet 1924;39:5-10. 31. Emmers R, Tasker RR. The human somesthetic thalamus. NY: Raven; 1975.
History of stereotactic and functional neurosurgery in canada
32. Tasker RR, Rowe IH, Hawrlyshyn P, Organ LW. Computer mapping of human subcortical sensory pathways during stereotaxis. J Neurol Neurosurg Psychiatry 1975;38 (4):408. 33. Tasker RR, Organ LW, Hawrylyshyn PA. The thalamus and midbrain of man. A physiological atlas using electrical stimulation. Springfield, IL: Charles C Thomas; 1982. 34. Takser RR. Deafferentation. In: Wall PD, Melzak R, editors. Textbook of pain. London: Churchill Livingstone; 1984. p. 639-55. 35. Tasker RR. Management of nociceptive, deafferentation and central pain by surgical intervention. In: Fields HL editor. Pain syndromes in neurology. 2nd ed. London: Butterworths; 1990. p. 143-200. 36. Speakman TJ. Results of thalamotomy for Parkinson’s disease. Can Med Assoc J 1963;89:652-66. 37. Kumar K, Malik S, Demeria D. Treatment of chronic pain with spinal cord stimulation versus alternative therapies: cost-effectiveness analysis. Neurosurgery 2002;51(1): 106-15. 38. Kumar K, Hunter G, Demeria DD. Treatment of chronic pain by using intrathecal drug therapy compared with conventional pain therapies: a cost-effectiveness analysis. J Neurosurg 2002;97(4):803-10. 39. Hornberger J, Kumar K, Verhulst E, Clark MA, Hernandez J. Rechargeable spinal cord stimulation versus nonrechargeable system for patients with failed back surgery syndrome: a cost-consequences analysis. Clin J Pain 2008;24(3):244-52.
10
40. Kumar K, Toth C, Nath RK, Laing P. Epidural spinal cord stimulation for treatment of chronic pain - some predictors of success. A 15-year experience. Surg Neurol 1998;50(2): 110-20. 41. Kumar K, Hunter G, Demeria D. Spinal cord stimulation in treatment of chronic benign pain: challenges in treatment planning and present status, a 22-year experience. Neurosurgery 2006;58(3):481-96. 42. Kumar K, Toth C, Nath RK. Deep brain stimulation for intractable pain: a 15-year experience. Neurosurgery 1997;40(4):736-46. 43. Turnbull F. The nature of pain in the late stages of cancer. Surg Gynecol Obstet 1960;110:665-8. 44. Turnbull F. Intractable pain. Proc R Soc Med 1954;47(2): 155-6. 45. Turnbull F. A basis for decision about cordotomy in cases of pelvic carcinoma. J Neurosurg 1959;16:595-9. 46. Turnbull IM, Shulman R, Woodhurst WB. Thalamic stimulation for neuropathic pain. J Neurosurg 1980;52(4): 486-93. 47. Kiss ZH, Doig-Beyaert K, Eliasziw M, Tsui J, Haffenden A, Suchowersky O. Functional and Stereotactic Section of the Canadian Neurosurgical Society, Canadian Movement Disorders Group. The Canadian multicentre study of deep brain stimulation for cervical dystonia. Brain 2007;130 Pt 11:2879-86. 48. Bertrand C, Molina-Negro P, Martinez SN. Technical aspects of selective peripheral denervation for spasmodic torticollis. Appl Neurophysiol 1982;45:326-39.
123
15 History of Stereotactic Neurosurgery in Italy A. Franzini . V. A. Sironi . G. Broggi
The history of stereotactic neurosurgery in Italy is strictly linked to the development of the mayor European schools of Paris, Stokholm and Freiburg. In the sixties almost any neurosurgical department in Italy had a stereotactic frame dedicated to thalamotomy for Parkinson disease (the Talairach frame, Reichert frame, Guiot frame, Leksell frame and the Cooper frame were the most popular steretactic devices). From this ‘‘pneumoencephalography era’’ we can remember Franco Migliavacca in Milan, Dalle Ore in Verona, Faust D’Andrea in Naple and Elio Tartarini in Genoa who performed thousands of stereotactic operations for Parkinson disease, mental illness, and pain [4]. After the LDopa discovery and its wide therapeutical application, stereotactic surgery seems to have disappeared from Italy except for a few Institutes which still continued to perform stereotactic operations for tremor, pain, dystonia, and epilepsy. So the real first generation of surgeons mainly devoted to functional neurosurgery in Italy in the ‘‘ventriculography era’’ includes Franco Marossero and Paolo Emilio. Maspes in Milan [12,13,14], Victor Aldo Fasano in Turin [7], Gianfranco Rossi [16] and Beniamino Guidetti [11] in Rome. An original stereotactic frame was also developed in Turin and utilized for the treatment of cerebral palsy in adults and children (> Figure 15-1) utilizing alcoholic lesions and later cryothalamotomy. The main interests of the schools of Milan (F. Marossero) and Rome (G. Rossi) were the functional exploration of the brain by acute and chronic implanted electrodes recording deep EEG activity (SEEG) in epilectic patients candidated to tailored resection of the epileptic focus or to #
Springer-Verlag Berlin/Heidelberg 2009
deep radiofrequency lesions of nuclei and tracts involved in the origin and diffusion of the epileptic discharge [13]. Also if this methodology was imported from the Paris School of Talairach and Bancaud, they developed many original observations and contributed to the definition of criteria still utilized worldwide for surgical treatment of epilepsy [12,14]. In those years particularly Gianfranco Rossi had an eminent role in the development of functional and stereotactic neurosurgery in Italy; he therefore deserves a more detailed history. Prof. Rossi worked for 4 years in the field of experimental neurophysiology under the leadership of Prof. Giuseppe Moruzzi and his research interest was centered on the anatomic and functional organization of brainstem reticular formation and sleep physiology. Later when he became chairman of the neurosurgical Department at the Catholic University of Rome his neurophisyological background had a strong influence on clinical practice and research projects and original criteria were proposed to improve the interpretation of electrocerebral epileptic signals. The rationales, indications, and relative efficacy of classic surgical resection approaches including callosotomy, multiple subpial transection, and the so-called lesionectomy were studied [17]. In the same years C. A. Pagni in Milan wrote an universally appreciated book and publications about the surgical treatment of central pain [16]. In that period B. Guidetti in Rome performed operations to treat spasticity (Dentatectomy) and pain (Pulvinotomy) [11]. The main interest of the Turin school was the treatment of cerebral palsy and many thalamotomies have been
194
15
History of stereotactic neurosurgery in italy
. Figure 15-1 Stereotactic thalamotomy performed in the ‘‘pneumoencephalographic era’’ with the Italian frame named FasanoSguazzi. Note the contrast medium (lipiodol)in alcoholic solution injected at the target sites on both sides. (Courtesy of Sergio Zeme M.D., University of Turin, Italy)
performed in dystonic children [7]. The second generation of functional neurosurgeons at the beginning of the ‘‘CT era’’ includes De Divitis in Naples who heralded stereotactic surgery for Gilles de la Tourette syndrome [5]; the senior author (G.B.) who published milestone studies about impedance guided biopsy [1], cell kinetics of deep brain tumors, stereotactic treatment of brain abscesses and pioneered in Europe the treatment of cystic components of craniopharingiomas by intracavitary Bleomycin; founded in Milan the first Italian neurosurgical Department dedicated to functional and stereotactic neurosurgery providing fuel for future development of original treatments such as Deep Brain Stimulation for the treatment of chronic refractory cluster headache and disruptive behaviour. Another master of stereotactic neurosurgery was Franco Frank in Bologna who pioneered surgery of mesencephalic structure to treat cancer pain [8]. Prof. Frank was trained in Freiburg by Mundinger and founded in Bologna the school of functional and stereotactic neurosurgery. His
original approach included nearly all fields of interest of neurosurgery with particular regard to management of brain tumors by intracavitary irradiation and management of pain by stimulation of the periacqueductal gray and thalamus [9]. Massimo Scerrati in Rome and Pierligi De Riu in Sassari [6,18] introduced in Italy the stereotactic brachiterapy of brain tumors. Prof. Scerrati developed the ‘‘Scerrati’s arc’’ to transform the Talairach frame in an isocentric frame [19]. Prof. Mario Meglio in Rome introduced the use of spinal cord stimulation to improve blood flow in peripheral vascular diseases [15]. Belonging to this generation is also Claudio Munari who worked in Paris and Grenoble and founded in Italy the first department entirely dedicated to surgery of epilepsy and functional exploration of the brain. Also in the field of radiosurgery the Italian contribution was significant, particularly due to the contribution of Federico Colombo who applied the linear accelerator to the stereotactic frame and widened considerably the field of application of radiosurgery [3]. Dr. Colombo
History of stereotactic neurosurgery in italy
built up a huge casuistic which includes more than 800 arteriovenous malformations treated by radiosurgery and is one of the larger series of the world. Also in the field of neuroimaging, the Italian contribution was highly represented by Cesare Giorgi who built one the first digitalized stereotactic atlases [10] and developed original equipments in the field of robotics and tridimensional neuronavigation. The ‘‘neuromodulation era’’ heralded at the beginning of the eighties by the senior author [2] belong to the present.
References 1. Broggi G, Franzini A. Value of serial stereotactic biopsies and impedance monitoring in the treatment of deep brain tumours. J Neurol Neurosurg Psychiatry 1981;44(5): 397-401. 2. Broggi G, Franzini A, Giorgi C, Servello D, Spreafico R. Preliminary Results of Specific Thalamic Stimulation for Deafferentation Pain. Acta Neurochir (Wien) Suppl 1984;33:497-500. 3. Colombo F, Benedetti A, Pozza F, Zanardo A, Avanzo RC, Chierego G, Marchetti C. Stereotactic radiosurgery utilizing a linear accelerator. Appl Neurophysiol 1985;48 (1–6):133-45. 4. Dalle Ore G, Da Pian R. Intractable pains and destruction of the latero-ventral nucleus of the thalamus. Presentation of a case of ‘‘phantom limb’’ operated upon with complete success. Minerva Med 1960;29;51:2771-3. 5. de Divitiis E, D’Errico A, Cerillo A. Stereotactic surgery in Gilles de la Tourette syndrome. Acta Neurochir (Wien) 1977;(Suppl 24):73. 6. De Riu PL, Rocca A. Interstitial irradiation therapy of supratentorial gliomas by stereotaxic technique. Long term results. Ital J Neurol Sci 1988;9(3):243-8. 7. Fasano VA, Broggi G, Schiffer D, Urciuoli R. Experimental lesions induced by localized cooling of the brain. Morphologic and histochemical study. Neurochirurgie 1965;11(6):519-28.
15
8. Frank F, Fabrizi AP, Gaist G. Stereotactic mesencephalic tractotomy in the treatment of chronic cancer pain. Acta Neurochir (Wien) 1989;99(1–2):38-40. 9. Frank F, Frank G, Gaist G, Galassi E, Sturiale G, Fabrizi A. Deep brain stimulation in the treatment of chronic pain syndromes. Riv Neurobiol 1982;28 (3–4):309-16. 10. Giorgi C, Garibotto G, Cerchiari U, Broggi G, Franzini A, Koslow M. Neuroanatomical digital image processing in CT-guided stereotactic operations. Appl Neurophysiol 1983;46(1–4):236-9. 11. Guidetti B, Fraioli B. Neurosurgical treatment of spasticity and dyskinesias. Acta Neurochir (Wien) 1977; (Suppl 24):27-39. 12. Marossero F, Ettorre G, Infuso L, Pagni CA. Surgical treatment of non-tumoral epilepsy with stereotaxic methods. Minerva Neurochir 1966;10(4):342-3. 13. Marossero F, Ravagnati L, Sironi VA, Miserocchi G, Franzini A, Ettorre G, Cabrini GP. Late results of stereotactic radiofrequency lesions in epilepsy. Acta Neurochir Suppl (Wien) 1980;30:145-9. 14. Marossero F, Ettorre G, Franzini A, Motti DF. Chronic depth electrodes study of one case of bitemporal epilepsy due to glial tumour. Some physiopathological considerations. Acta Neurochir (Wien) 1978;45(1–2): 123-31. 15. Meglio M, Cioni B, Dal Lago A, De Santis M, Pola P, Serricchio M. Pain control and improvement of peripheral blood flow following epidural spinal cord stimulation: case report. J Neurosurg 1981;54(6):821-3. 16. Pagni CA. Central pain and painful anesthesia. In: Progress in Neurological Surgery vol.8:132-157. (Karger, Basel 1976). 17. Rossi GF. Surgical treatment of partial epilepsy: remarks on the relative role of stereo-EEG and other diagnostic examinations. J Neurosurg Sci 1975;19(1–2):89-94. 18. Scerrati M, Roselli R, Iacoangeli M, Montemaggi P, Cellini N, Falcinelli R, Rossi GF. Comments on brachycurie therapy of cerebral tumours. Acta Neurochir Suppl (Wien) 1989;46:94-6. 19. Scerrati M, Fiorentino A, Fiorentino M, Pola P. Stereotaxic device for polar approaches in orthogonal systems. Technical note. J. Neurosurg 1984;61(6):1146-7.
195
6 History of Stereotactic Neurosurgery in the Nordic Countries B. A. Meyerson . B. Linderoth
In the sixth volume of the classical and comprehensive textbook of neurosurgery by Olivecrona and To¨nnis (1957), Lars Leksell contributed a chapter on ‘‘Targeted Brain Operations’’ (Gezielte Hirnoperationen). In the introduction he refers to the development of his stereotactic instrument – he always preferred to describe his stereotactic frame as a surgical instrument – that followed the pioneering work by Spiegel and Wycis in 1947–1948. The first presentation of Leksell’s sterotactic apparatus dates from 1949 and already in the early 1950s he had gained experience in targeted lesional surgery using electrical current and different types of X-ray beams – radiosurgery. There is no doubt that this makes Leksell, together with a few others, one of the founders of stereotactic surgery. It should also be noted that the principal mode of function and general features of his original instrument have survived in the modern version that is still one of the most commonly used stereotactic systems in the world. That is why the advancement of stereotactic neurosurgery in the Nordic countries is so intimately associated with the name of Lars Leksell and his contributions. Lars Leksell (1907–1986) started his neurosurgical training with Herbert Olivecrona in 1935 at the Serafimer Hospital, one of the oldest hospitals in Sweden founded in 1752. The Olivecrona neurosurgical service enjoyed a solid international reputation and attracted a large number of trainees from all over the world. For a short period Leksell served as a volunteer #
Springer-Verlag Berlin/Heidelberg 2009
medical doctor in Finland when it was attacked by the Soviet Union in November 1939. Later he told that he during that war often speculated on the possibility of extracting bullets from the brain with minimal damage to the surrounding brain tissue using a mechanically guided instrument. In the early 1940s Leksell joined Ragnar Granit, Nobel Laureate 1967, for experimental studies in neurophysiology. In 1945 he presented a PhD dissertation, a monograph on the motor gamma system titled ‘‘The action potential and excitatory effects of the small ventral root fibers to skeletal muscle.’’ This was a major milestone in the understanding of muscle control and has now become part of basic neurophysiology. It should be noted that during these years he, together with Granit and Skoglund, made another major contribution by describing the phenomenon of ephapsis, ‘‘artificial synapses,’’ caused by local pressure on a nerve, as a possible mechanism involved in trigeminal neuralgia. After resuming clinical work, he started work on the development of a stereotactic instrument and in 1947 he visited Wycis in Philadelphia. Leksell described his instrument in a publication in 1949 and this was the first example of a stereotactic system based on the principle of ‘‘center-of-arc’’ in contrast to the Spiegel and Wycis orthogonal, rectangular frame (> Figure 6-1). The use of a movable semi-arc with an electrode carrier implies that the tip of a probe can reach the target regardless of the position of the carrier or the angling of the arc relative to the skull
66
6
History of stereotactic neurosurgery in the nordic countries
. Figure 6-1 The first Leksell stereotactic instrument, based on the ‘‘center-of-arc’’ principle, was described in a publication from 1949
fixation device, i.e., a frame or base plate with bars for bone fixation screws. This construction permits also transphenoidal, straight lateral and suboccipital probe approaches. Leksell was in many respects a perfectionist and for the rest of his life he continued to change and revise the design of virtually every small part of his instrument though the basic semicircular frame was retained. He focused not only on upgrading the function of the instrument but also on its aesthetic appearance. An important feature was that ‘‘the apparatus should be easy to handle and practical in routine clinical work’’ and ‘‘a high degree of exactitude is necessary.’’ An oft-cited quotation is ‘‘Tools used by the surgeon must be adapted to the task and where the human brain is concerned, no tool can be too refined.’’ The first, documented clinical application of Leksell’s stereotactic system was a case of a craniopharyngioma cyst that was punctured and treated with injection of radioactive phosphorus – that patient was probably the first patient in the world to undergo this form of therapy (1948) (> Figure 6-2). Before the advent of modern imaging techniques (CT, MRI), ventriculography was, and in some centers still is, routinely utilized for target
. Figure 6-2 The first practical application of the stereotactic instrument was for puncture of craniopharyngioma cysts
coordinate determination. Already in the late 1940s neuroradiology was a well-developed speciality at the Serafimer Hospital and angiography and pneumoencephalography were routinely practiced. Leksell performed pneumoencephalography, first in the sitting and then in the supine position to visualize the anterior and posterior commissures, respectively. In order to compensate for the divergence of the X-rays, he constructed a diagram of tightly packed concentric circles, approximated to spirals, geometrically related to the divergence and the distance between the X-ray tube and the film, and frame planes, for determining the target coordinates; it has to be admitted, however, that in contrast to Leksell’s other inventions many surgeons found it difficult to understand and use this diagram. Beside the passionate interest in the technical aspects of stereotaxy, Leksell was in the 1950s and 1960s very active in the operation theatre. He performed a large number of pallidotomies, and later also thalamotomies, in Parkinson’s disease and capsulotomies in various forms of
History of stereotactic neurosurgery in the nordic countries
mental disorders (> Figure 6-3). The results of a series of 81 patients subjected to pallidotomy was published in 1960, and 116 patients treated with capsulotomy were reported in 1961. The term and concept of radiosurgery were introduced by Leksell already in 1951 when he reasoned that the ‘‘center-of-arc’’ principle and his first stereotactic instrument were suitable for replacing a probe (needle electrode) by cross-firing intracerebral structures with narrow beams of radiant energy. X-rays were first tried but both gamma rays and ultrasonics were included as alternatives (> Figure 6-4). Initial experiments were performed on cats and then a few patients with pain and chronic psychosis were treated with a 280 kV X-ray tube attached to the arc. Of particular interest is that in 1953 two cases of trigeminal neuralgia were treated and at followup in 1971 they were still free of pain. In 1946 Leksell was appointed head of a neurosurgical unit in Lund in southern Sweden where he became professor in 1958 and remained so until 1960. In those days there were very few neurosurgeons around the world who were active in stereotactic surgery and the international network was very small; it is of interest that the Schaltenbrand and Bailey’s Stereotactic Atlas was partly based on some brain specimens supplied by Leksell. While in Lund, Leksell was apparently able to evade many of his clinical obligations because he was able to initiate a close collaboration with a team of physicists led by Bo¨rje Larsson at the University of Uppsala (north of Stockholm) where a synchrocyclotron was available. They conducted experiments with stereotactic high-energy proton irradiation in goats resulting in a seminal publication in Nature in 1958 (> Figure 6-5). This technique was also applied in a few patients with Parkinson’s disease (pallidotomy), psychiatric disorder (capsulotomy) and pain (mesencephalotomy). Although precisely placed and well-limited lesions could be produced by the focused proton beams, as demonstrated in a few autopsy cases, the synchrocyclotron proved to be too complicated
6
. Figure 6-3 Leksell in the early 1960s with the second generation of his stereotactic frame
. Figure 6-4 Application of the first stereotactic instrument for radiosurgery using cross-firing of 280 kV X-rays. Photo from the early 1950s
for general clinical use. This compelled Leksell to consider other radiation sources and he started designing the 60Co Gamma Unit that was fully integrated with the stereotactic system. The
67
68
6
History of stereotactic neurosurgery in the nordic countries
. Figure 6-5 The first clinical trials with radiosurgery using proton beams from the cyclotron at Uppsala University were performed in the late 1950s
. Figure 6-6 Leksell treating the first case of acoustic neurinoma with the first version of the Gamma Knife in 1968
development of the ‘‘Beam-knife’’ took place after Leksell had been appointed successor to Olivecrona in 1960 and the first unit was inaugurated in 1967 (> Figure 6-6). Later the same year reports of the two first cases, patients with cancer related pain subjected to radiosurgical thalamotomy, were published.
Originally, radiosurgery and the Gamma Unit were developed with the hope that it would offer a bloodless, and less risky, method to be applied principally in functional neurosurgery, for example in thalamotomy for Parkinson’s disease. On the other hand, Leksell had always considered his sterotactic instrument a
History of stereotactic neurosurgery in the nordic countries
surgical tool that should also be utilized in general neurosurgery in order to enhance precision and minimize hazards. This idea had to some extent been realized by the extensive use of stereotactic technique in puncturing cysts and also in performing biopsies in critical regions. However, the Gamma Unit soon proved to be very useful in the treatment of some typical diseases requiring neurosurgery, such as pituitary adenomas, acoustic neurinomas and arteriovenous malformations. This successful application of radiosurgery has indeed revolutionized the management of these conditions but was also met with much skepticism from the neurosurgical community. Albeit the topic of this chapter is the history of stereotaxis, two other examples of Leksell’s exceptionally innovative mind deserve to be mentioned. He was the first to apply ultrasound in neurosurgical diagnosis by the development of echoencephalography as early as 1955. Moreover, his elegantly designed, double action rongeur has become an indispensable tool in the hands of most neurosurgeons.
Leksell’s Disciples Erik-Olof Backlund (1931) started his neurosurgical training in 1960 at the Serafimer Hospital and he soon became a pupil and a close associate of Leksell. At that time neuroradiology could visualize pathological mass processes only indirectly, and therefore stereotactic biopsy was rarely performed, the exceptions being trials to obtain specimens of the cystic wall when performing craniopharyngioma cyst punctures. This became part of Backlund’s doctoral thesis and to facilitate biopsy sampling he constructed a new device consisting of a spiral that could be screwed into the often quite tough craniopharyngioma tissue from which a specimen could then be harvested by using an outer sharp cutting needle as a sleeve punch.
6
This biopsy needle proved to be very useful and is still routinely used. Backlund also contributed other innovative needle instruments for aspiration biopsy, hematoma evacuation, and aqueduct reconstruction. Backlund’s interest in craniopharyngiomas led him to reconsider the choice of radioactive phosphorus for cystic intracavity irradiation and he demonstrated that a colloid 90 Y was more effective. Subsequently, he developed a multimodal treatment program for craniopharyngiomas, and the evacuation of cysts and installation of radioactive substances were in some cases supplemented by radiosurgery of the solid parts (> Figure 6-7). This treatment strategy is still practiced at the Karolinska University Hospital. Another contribution was the application of radiosurgery for capsulotomy in patients with anxiety and OCD; this was first published, with Leksell as co-author, in 1978. As noted by Backlund himself in an autobiographical vignette, the most spectacular phase of his career was the first neurotransplantation of adrenal chromaffin cells in a patient with Parkinson’s disease. This trial was initiated by a group of researchers in basic sciences at the . Figure 6-7 Erik-Olof Backlund (right) and the research engineer Bengt Jernberg (left) preparing a patient for capsulotomy. Note, at that time the head was fixed with a Orthoplast cap. Photo from the mid 1970s
69
70
6
History of stereotactic neurosurgery in the nordic countries
Karolinska Institute and Backlund developed the stereotactic technique for precise depositing of the cellular specimen. Ladislau Steiner (1920) is presumably the most well known of Leksell’s pupils. From 1962 he first trained with Olivecrona who was then still active in spite of retirement. His prime interest was vascular and tumor microsurgery rather than endeavors within the stereotactic and functional field. His first exposure to radiosurgery was when he took part in a major study thalamotomy for cancer-related pain. It then appeared that radiosurgery might be useful for some common neurosurgical diseases also, and together with Leksell he became the first to demonstrate that centrally located arteriovenous malformations, otherwise inaccessible to surgery, could be miraculously obliterated. Steiner, along with Christer Lindquist (1944) who later became one of the leading names in radiosurgery with many important publications, subsequently performed a large number of AVM treatments. Following the first publication many patients from all over the world were referred to him. The AVM experiences triggered a surge of trials on new applications for radiosurgery (acoustic neurinomas, pituitary adenomas, pituitary hormone suppression in Cushing’s disease, pinealomas, meningiomas, and later cerebral metastases). In 1987 Steiner left the Karolinska University Hospital and became professor and director of the Lars Leksell Gamma Knife Center at the University of Virginia. There he created a very active team that has made this center one of the leading ones for radiosurgery. Scientifically also, he has continued to be very active and is co-author of numerous publications covering the entire field of radiosurgery. A few others of Leksell’s disciples who were active in the field of radiosurgery and have made important, more or less independent contributions should be mentioned. Georg Nore´n (1943) was the first to demonstrate, in an extensive and long-lasting study, that the outcome of
radiosurgical treatment of acoustic neurinomas is at least as favorable as that with open surgery performed by the most experienced neurosurgeons. Since 1992, Nore´n has been director of the Gamma Knife Center at Brown University, Providence. Tiit Ra¨hn (1940) became involved in stereotactic techniques and radiosurgery in the late 1960s and he pioneered the management of hormone secreting pituitary adenomas. His career, until retirement, has been entirely devoted to radiosurgery. The senior author of this paper (B.A.M 1933), Meyerson joined Leksell’s team in 1968, shortly after the inauguration of the first Gamma Unit. He soon became involved in ‘‘gammathalamotomy’’ for cancer-related pain and the preliminary results were published in 1969. Because of the relatively unsatisfactory outcome of this form of surgery a project with intracerebral electrical stimulation (deep brain stimulation, DBS) for pain was initiated in 1974, and with electrodes of our own design stimulation was applied simultaneously in several brain targets. This was the beginning of a life-long career in pain research. At an early stage Meyerson was also introduced to surgical treatment of mental disorders and subsequently he performed the great majority of capsulotomies, as well as a large number of interventions for movement disorders, at the Karolinska University Hospital. The second most prominent stereotactic neurosurgeon in the Nordic countries was Lauri Laitinen (1928–2005) and it is notable that he pursued his own career independent of Leksell. He had his basic neurosurgical training in Helsinki and was confronted with stereotactic surgery when working in London, and visiting Gillingham in Edinburgh. He started operating on Parkinsonian patients in 1961 using first the Cooper instrument and later a RiechertMundinger frame. He was, however, dissatisfied with both these instruments and towards the end of the 1960s he had constructed his own system
History of stereotactic neurosurgery in the nordic countries
that was later acquired by many stereotactic centers (> Figure 6-8). This instrument has the Leksell center-of-arc design combined with some features of the Riechert-Mundinger frame. Throughout his career he was very active in clinical practice and performed about 200 thalamotomies annually. During these operations he made extensive trials with impedance monitoring which he found very helpful for differentiating white from grey matter, for example to identify the border between the internal capsule and pallidal tissue. In 1963 Laitinen published his first paper in the field of stereotactic surgery, on the treatment of torticollis. Later he joined with a young neuropsychologist, Juhani Vilkki, and in a series of papers they documented, for the first time, the deterioration of motor, verbal, and visuospatial functions that may occur after thalamotomy. Laitinen was also very interested in pain, and inspired by Gabriel Mazars in Paris, he performed the first trial of implanting a stimulating electrode in a patient with phantom limb pain already in 1968. In 1980 Laitinen moved to Umea˚, a university town in northern Sweden, where he revived stereotactic and functional neurosurgery to such an extent that Umea˚ became a center of excellence. There he developed a device (the
6
‘‘Stereoadapter’’), a light frame that in a reproducible way could be repeatedly attached to the skull during CT and MRI examinations and then utilized for coordinate determination together with the original frame. In 1984 he, along with Marwan Hariz (who later became his successor) and Tommy Bergenheim, began to test Leksell’s old and almost forgotten version of pallidotomy from the early 1950s. For many of the younger neurosurgeons this endeavor is directly associated with the name of Laitinen. It should be recalled that with the advent of L-dopa in the early 1970s the use of surgery for movement disorders, in particular for Parkinson’s disease, was reduced to just a few cases each year even in the major centers. The first presentation in 1992 of the favorable outcome of Laitinen’s pallidotomy cases resulted in a dramatic change in attitude toward surgery among the neurologists. A contributing, important factor was that this type of intervention had proven to be effective for L-dopa induced hyperkinesias also, which by that time had become a major problem. No doubt, the acceptance of pallidal surgery following Laitinen’s reports paved the way for the ensuing development of stimulation of central brain areas (DBS) as a novel mode of therapy for Parkinson’s disease.
. Figure 6-8 Lauri Laitinen standing by a case displaying his own frame and stereoadapter. Photo from 2003
71
72
6
History of stereotactic neurosurgery in the nordic countries
During the 1960–1970s stereotactic surgery was practised occasionally in a few other neurosurgical units in the Nordic countries but these activities cannot be regarded as pioneering or innovative. One exception was the contributions made by Kjeld Vaernet (1920–2006) in Copenhagen who in 1974 reported on a trial with ‘‘stereotactic stimulation and electrocoagulaton of the lateral hypothamus’’ in five patients with gross obesity. He also performed many stereotactic surgeries for epilepsy and mental disorders. Even though the contributions of Leksell are exceptional and outstanding, the importance of the achievements of his disciples and of Laitinen in making the Nordic countries strong in stereotactic open neurosurgery and radiosurgery should not be underestimated.
Selected Bibliography 1. Backlund EO, Granberg PO, Hamberger B, Knutsson E, Martensson A, Sedvall G, Seiger A, Olson L. Transplantation of adrenal medullary tissue to striatum in parkinsonism: first clinical trials. J Neurosurg 1985;62: 169-73. 2. Backlund EO. Reflections: a historical vignette. Neurosurgery 2004;54:734-41. 3. Backlund EO, Johansson L, Sarby B. Studies on craniopharyngiomas. II. Treatment by stereotaxis and radiosurgery. Acta Chir Scand 1972;138:749-59. 4. Bingley T, Leksell L, Meyerson BA, Rylander G. Stereotaxic anterior capsulotomy in anxiety and obsessivecompulsive states. Third World Congress of Psychosurgery, Cambridge: University of Cambridge Press; 1972. p. 159-64. 5. Boe¨thius J, Lindblom U, Meyerson BA, Wide´n L. Effects of multifocal brain stimulation on pain and somatosensory functions. In: Zotterman Y, editor. Sensory functions of the skin in primates with special reference to man. New York: Pergamon Press; 1976. p. 531-48. 6. Hirsch A, Nore´n G, Anderson H. Audiologic findings after stereotactic radiosurgery in nine cases of acoustic neurinomas. Acta Otolaryngol 1979;88:155-60. 7. Laitinen LV. A new stereoencephalotome. Zentralbl Neurochir 1971; 32: 67-73. 8. Laitinen LV, Bergenheim AT, Hariz MI: Leksell’s posteroventral pallidotomy in the treatment of Parkinson’s disease. J Neurosurg 1992;76:53-61.
9. Laitinen LV. Personal memories of the history of stereotactic neurosurgery. Neurosurgery 2004;55:1420-8. 10. Larsson B, Leksell L, Rexed B, Sourander P, Mair W, Andersson B. The high-energy proton beam as a neurosurgical tool. Nature 1958;182(4644):1222-3. 11. Lindquist C, Kihlstro¨m L. Department of Neurosurgery, Karolinska Institute: 60 years. Neurosurgery 1996;39: 1016-21. 12. Leksell L. The action potential and excitatory effects of the small ventral root fibres to skeletal muscle. Acta Physiol Scand 1945;10 Suppl 31:1-79. 13. Leksell L. A stereotaxic apparatus for intracerebral surgery. Acta Chir Scand 1949;99:229-33. 14. Leksell L. The stereotaxic method and radiosurgery of the brain. Acta Chir Scand 1951;102:316-9. 15. Leksell L, Lide´n K. A therapeutic trial with radioactive isotopes in cystic brain tumor. In: Radioisotope techniques. Proceedings of the isotope techniques conference, Vol 1. Oxford: H.M. Stationary Office 1953; 1951. p. 76-8. 16. Leksell L. Gezielte Hirnoperationen. In: Olivecrona H and To¨nnis W, editors. Handbuch der Neurochirurgie, Vol 6., Berlin: Springer-Verlag; 1957. p. 178-99. 17. Leksell L. Some principles and technical aspects of stereotaxic surgery. In: Knighton RS and Dumke PR, editors. Pain. Boston: Little, Brown and Company; 1966. p. 493-502. 18. Leksell L. Cerebral radiosurgery. I. Gammathalamotomy in two cases of intractable pain. Acta Chir Scand 1968;134:585-95. 19. Leksell L, Backlund EO, Johansson L. Treatment of craniopharyngiomas. Acta Chir Scand 1967;133:345-50. 20. Leksell L. A note on the treatment of acoustic tumours. Acta Chir Scand 1971;137:763-5. 21. Leksell L. Sterotaxic radiosurgery in trigeminal neuralgia. Acta Chir Scand 1971;137:311-4. 22. LeksellL, Backlund EO. [Radiosurgical capsulotomy – a closed surgical method for psychiatric surgery] Lakartidningen 1978;75:546-7 (Swedish). 23. Nore´n G, Arndt J, Hindmarsh T. Stereotactic radiosurgery in cases of acoustic neurinoma: further experiences. Neurosurgery 1983;13:12-22. 24. Quaade F, Vaernet K, Larsson S. Stereotaxic stimulation and electrocoagulation of the lateral hypothalamus in obese humans. Acta Neurochir (Wien) 1974;30:111-7. 25. Ra¨hn T, Thore´n M, Hall K, Backlund EO. Stereotactic radiosurgery in Cushing’s syndrome: acute radiation effects. Surg Neurol 1980;14:85-92. 26. Steiner L, Leksell L, Greitz T, Forster DMC, Backlund EO. Stereotaxic radiosurgery for cerebral arteriovenous malformations: report of a case. Acta Chir Scand 1972; 138:459-64. 27. Svennilson E, Torvik A, Lowe R, Leksell L. Treatment of parkinsonism by stereotatic thermolesions in the pallidal region. A clinical evaluation of 81 cases. Acta Psychiatr Scand 1960; 35: 358-77.
History of Stereotactic Surgery
1 History of Stereotactic Surgery P. L. Gildenberg . J. K. Krauss
Preface In order to define when stereotactic surgery started, it is first necessary to define what stereotactic surgery is [1]. ‘‘Stereotaxic’’ surgery began in 1908, when Sir Victor Horsley and Robert Clarke [2] introduced their new apparatus to allow them to insert a probe, blade, or needle under accurate control into a subcortical structure of a monkey or other experimental animal. They specified the use of a Cartesian coordinate system [3] that makes it possible to define a point in space by specifying three coordinates, anterior-posterior (AP), lateral and vertical, and that Cartesian system remains the hallmark of stereotactic systems. Sir Victor Horsley was both a neurosurgeon and neurophysiologist, and is generally recognized as the father of human functional neurosurgery. He had collaborated with Robert Clarke, a mathematician and also a surgeon, and recruited him to help him design their device. The design and mathematics of the instrument was mainly Clarke’s, and the details of how it might be used were Horsley’s [4]. Their initial report of an experiment conducted with this apparatus concerned making electrolytic lesions in the dentate nucleus of the cerebellum of the monkey to study its structure and functions [2] (> Figure 1-1). The 1908 article that introduces stereotaxic surgery is a magnificent illustration of a literary scientific journal article, and the original should be read by anyone with an appreciation of the history of medical literature. The article is divided into four parts. One part describes the first animal stereotaxic apparatus, which has remained a model for most such devices, both for animals and patients, over this past century (> Figure 1-1). #
Springer-Verlag Berlin/Heidelberg 2009
In order to define a target point, they used Cartesian coordinates, that is, defining a point in space by its relationship to three planes, each at right angles to the other two, and all meeting at a common point as shown in > Figure 1-2. A location can be defined by three coordinates, each of which tells the relationship (in this case in millimeters) to one of the three planes – x mm anterior or posterior to the coronal plane, y mm lateral to the midsagittal plane, and z mm above or below the horizontal plane. For that to have relevance to the structures within the animal brain, the three planes must be registered, or accurately aligned, to the position of the head of the experimental animal. This was accomplished by relating the three planes to the same parts of the device that held the head securely – [6] a horizontal plane (similar to the Frankfort plane used in anthropology [6]) that passed through both ear plugs and a tab that held the left inferior orbital rim (or both if there is no asymmetry) from which to define the vertical coordinate – [7] a mid-sagittal plane that passes through the mid-point between the ear plugs at right angles to the horizontal plane to define the left or right lateral coordinate – and [8] a coronal plane that passes through the ear plugs and is perpendicular to the other two planes to define the AP coordinate. The second part of the Horsley-Clarke paper is a technique for making a stereotactic atlas registered to this same Cartesian system, in which the stereotaxic coordinates of a particular structure could be found. The third part was an excellent treatise about making direct current electrolytic lesions. The fourth and least remembered part was the experiment itself. Clarke suggested to Horsley that the technique might be useful in humans, and even
4
1
History of stereotactic surgery
. Figure 1-1 The original Horsley-Clarke apparatus [2]
. Figure 1-2 The concept of Cartesian planes to measure coordinates aligned with third ventricle
patented the idea of a human stereotactic apparatus, still based on using the boney landmarks to establish the landmarks from which Cartesian coordinates might be determined [9,10]. The
pair separated because of other disagreements, so neither pursued the idea of a human stereotactic any further [11]. In 1918, however, Mussen, an engineer who had been involved in making the Horsley-Clarke apparatus designed a similar device for use with the human skull [12]. There is no evidence that he persuaded any of his neurosurgical colleagues to use it, so he eventually wrapped it in newspaper and stored it in a box in his attic. When his family discovered the box almost 60 years after the Mussen frame had been made, they could determine when it had been stored by the date on the newspaper wrapping. It is perhaps best that it was not used, because it has since become known that the relationship between the boney landmarks on the human skull and the cerebral structures within are so variable that accurate targeting would not have been possible. It took most of the 40 years after the 1908 report that concepts of human physiology became well enough known to have determined the appropriate intracerebral subcortical targets. Let us digress for a moment to confront two other issues involving [6] devices prior to Horsley and Clarke and [7] the spelling of ‘‘stereotaxic’’ versus ‘‘stereotactic.’’ There were reports of a variety of devices to guide a probe into the brain or spinal cord prior to 1908, with consequent claims of a variety of ‘‘first’’ stereotactic devices. We are unaware of any prior system that used a Cartesian coordinate system. Most used overlying landmarks to approximate the location of various structures. Probably the first such system was that presented by Dittmar [13] in 1873, in which he used a guided probe to insert a blade into the medulla oblongata of the rat to perform physiological studies. From the Russian perspective, the beginnings of human stereotactic surgery are ascribed to Zernov [14], who in 1889 described an ‘‘encephalometer’’ that helped localize cortical areas, and Altukhov [15] who used it clinically 2 years later. This was not Cartesian in principle and did
History of stereotactic surgery
not address the localization of deep subcortical structures. In regard to the spelling of the technology addressed throughout this book, there is a logic to it. Horsley and Clarke called their technique ‘‘stereotaxic,’’ from the Greek ‘‘stereos,’’ threedimensional, and ‘‘taxus,’’ an arrangement, as in taxonomy. In a meeting in 1973 when it was decided to change the name of the ‘‘International Society for Research in Stereoencephalotomy’’ (meaning three-dimensional study of the encephalon or brain), it became necessary to agree on a spelling. Since the advent of human stereotactic surgery, some authors, mainly in Europe, used the spelling ‘‘stereotactic,’’ rather than ‘‘stereotaxic,’’ although the origins of that spelling have been lost. A vote was taken as to whether the newly named society would be spelled ‘‘stereotaxic’’ (‘‘three-dimensional’’ and ‘‘arrangement,’’ both from Greek), or ‘‘stereotactic’’ (a mongrel word meaning ‘‘three-dimensional’’ from Greek and ‘‘tactus’’ from Latin ‘‘to touch’’). It was felt that the object of the surgery was actually to touch the desired structure with a probe or electrode, so by vote the decision was made to spell the society ‘‘Stereotactic.’’ In a prescient moment, the decision was also made to call the societies which followed the International Society for Research in Stereoencephalotomy, the World Society for Stereotactic and Functional Neurosurgery, the American Society for Stereotactic and Functional Neurosurgery, and the European Society for Stereotactic and Functional Neurosurgery, ‘‘stereotactic’’ to denote the manner of localization (how you get to the target) and ‘‘functional,’’ meaning to change the function of the brain (what you do after you get there) [16]. The word ‘‘stereotaxis’’ refers to both animal and human procedures. The convention of referring to human surgery as ‘‘stereotactic’’ and animal surgery as ‘‘stereotaxic’’ has been generally adopted and will be maintained throughout this review. History is not a straight line. Different topics that relate to stereotactic surgery developed at
1
the same time, but often with different chronology. More often, advances in other fields became incorporated into stereotactic technique at various stages of development while other things were happening within stereotaxis itself. Consequently history cannot be presented as a continuum. We will digress as occasion demands, take side trips if they appear to be interesting, and double back on the calendar as we switch from one topic to another. The major topics will have separate titles, and subtopics will be announced in bold type, which signifies a change in gears and often a change in direction. After introducing the technology, we will take movement disorder surgery up to 1968, where we can pause that story, discuss other indications for functional neurosurgery, and then come back to the reawakening of stereotactic movement disorder surgery in 1992.
Birth of Human Stereotactic Surgery The problem of establishing accurate intracerebral landmarks from which specific brain structures could be measured and building a device to insert an electrode accurately to a chosen anatomical structure is credited to Ernest A. Spiegel and Henry T. Wycis (> Figure 1-3). Spiegel was a neurologist and neurophysiologist who fled the Nazis in Vienna and emigrated to Temple Medical School in Philadelphia as Professor of Experimental Neurology [17]. (His wife, Mona Spiegel-Adolph, was simultaneously recruited as Professor of Colloid Chemistry). Wycis began to work in Spiegel’s laboratory when he was a medical student, and their collaboration developed throughout his neurosurgical residency and appointment to the faculty [18]. In 1947, they reported on the first use of a human stereotactic apparatus they had designed (still spelled ‘‘stereotaxic’’ at that time) [5].
5
6
1
History of stereotactic surgery
. Figure 1-3 Spiegel and Wycis with their Model V apparatus [19]
The time was right for development of human stereotactic surgery (> Figures 1-4 and > 1-5). The motivation was there: Prefrontal lobotomy had been popular and was indicated in many cases, prior to the development of psychotropic medication, but then it became abused by some [20]. One main motivation to develop stereotactic surgery was to provide a controlled lobotomy effect with less risk of unintended neurological deficit. Once stereotaxis was available, however, the first patients were those with movement disorders, and its use in psychosurgery was not reported for several years. The technology had just been developed: One key to this field was the first introduction of X-ray equipment into the operating room with rapid film development. The key to human stereotactic surgery is to identify landmarks within the brain by X-ray or by other imaging and calculating where a target lay in relation to those landmarks. Ventriculogram X-rays could be taken to visualize internal cerebral landmarks about the third ventricle, from which measurement could be made to localize any cerebral
. Figure 1-4 The original Spiegel-Wycis stereotactic apparatus [5]
. Figure 1-5 Spiegel and Wycis in the operating room. Note the Faraday cage to shield electrical noise so recording could be done
History of stereotactic surgery
structure [5]. This made it possible to use intracerebral landmarks, usually around the third ventricle, from which to measure the three coordinates, each of which was based on one of the three Cartesian planes, hence the AP, lateral, and vertical coordinate (> Figure 1-2). The location of a specific structure required reference to a human stereotactic atlas, which they published soon after stereoencephalotomy was introduced [21] (> Figure 1-6). The science was there: The field of neurophysiology had advanced to the point where targets could be selected on a rational basis. An appreciation of the extra-pyramidal system was emerging, and it was recognized that many movement disorders involved those circuits. Surgical feasibility was demonstrated by Meyers [22] (> Figure 1-7) – series of 38 patients reported in 1942 (> Figure 1-8). There were also attempts to interrupt pathways in the brain for management of pain. Many of these techniques involved surgical interruption of conveniently superficial pathways and are still in use [24–26]. There was a good concept
. Figure 1-6 The orientation of slices in a stereotactic atlas
1
of the primary pain pathway as it ascends through the brainstem, and the concept that the limbic system participated in pain perception was becoming well developed [27,28]. The engineering was there: Spiegel and Wycis [21] called their new technique ‘‘stereoencephalotomy,’’ a three-dimensional system based on brain measurements. The device was essentially a Horsley-Clarke apparatus suspended over the patient’s head, secured to a ring held by a plaster cap made for each patient individually (> Figure 1-3). The ring was aligned with the horizontal plane by ear plugs so that AP and lateral X-rays could be taken accurately and repeatedly. A microdrive supported by the ring was mounted above the patient’s head to hold the electrode and advance it either vertically or obliquely into the target within the brain. They originally selected the anterior commissure and the pineal gland as the two basic landmarks, and produced a stereotactic atlas based on measurements between those structures and the desired target [21]. The use of the anterior and posterior commissure was made a decade later
7
8
1
History of stereotactic surgery
. Figure 1-7 Russel Meyers, 1904–1999 [23]
. Figure 1-8 Meyers’ open surgical approach to the basal ganglia [23]
by Talairach and his associates [29], which, with some variations, were adopted by most stereotactic neurosurgeons. Originally, X-ray visualization of the landmarks about the ventricle required a pneumoencephalogram [21,30], which made it mandatory that the stereotactic apparatus could be accurately reapplied, which was the protocol when
I (plg) first started working as a medical student with Spiegel and Wycis [18,31] in 1956. Patients had the device attached on Tuesday, pneumoencephalogram performed, and measurement made. The apparatus was removed, because the patient by that time was too sick to have a procedure performed under local anesthesia. The patient was returned to the operating room 2 days later to replace the head ring and for the actual surgery. When positive contrast agents became available, it became possible to do the surgery at a single session [18]. A philosophy that was begun with the first Spiegel-Wycis procedure and maintained through the years by most stereotacticians is that each time an electrode is inserted into the brain it produces a unique opportunity to study human neurophysiology, so each procedure included both physiological confirmation of the electrode position and neurophysiologic studies [32]. Not only did this provide a confirmation of the intended target, but led to advances in our understanding of the human brain and the diseases being treated. Originally, the lesions were made with alcohol injection, which theoretically was more likely to affect the neurons in the intended nucleus while sparing the fibres en passage. The spread of the alcohol was unpredictable, however, so soon lesions were made with the same electrolytic direct current that Horsley and Clarke had reported almost half a century before [2]. However, that carried the risk of a sudden sharp stimulation if the current varied. Other techniques were soon developed, such as oil-procaine or oil-procaine-wax injection [33,34], alcohol injection, sometimes with a balloon cannula or coagulating substance [35,36], mechanical damage with a leukotome [37], and later radiofrequency [38] and cryoprobe methods [39]. The original report by Spiegel and Wycis [5] ended by enumerating a list of potential indications for stereotactic surgery. It is almost certain that they had used this new procedure for all those indications prior to their first publication.
History of stereotactic surgery
The interest was there: The first decade after human stereotactic surgery was introduced was particularly productive. There was a stream of neurosurgeons from throughout the world visiting Spiegel and Wycis’ service at Temple Medical School in Philadelphia, learning this new technique and returning home to enter this new field. It was necessary for each to design and manufacture his own stereotactic frame, since there were none commercially available. A variety of apparatus was rapidly introduced, including several advanced designs by Spiegel and Wycis, so that they did most of their work with the Model V [19]. All told, during the 1950’s, at least 40 other stereotactic apparatus were designed along the three basic types. The Spiegel-Wycis system required translational adjustments, as did the Horsley-Clarke apparatus, in that the position of the electrode was changed by sliding the electrode carrier anterior-posterior and laterally along a base plate, to adjust to the proper coordinates, the vertical adjustment was made by a microdrive system holding the electrode. The predetermined trajectory might require two separate angular adjustments. Other devices rapidly followed [40]. Lars Leksell [41] returned home to Sweden after a trip to Philadelphia and designed the first arc centered device in 1948. The three coordinates indicated the center of a semicircular arc along which an electrode carrier moved, so it always pointed to the isocenter at which the target lay. Since the target always lies in the center of the arc, insertion along any angle would bring the probe to the target. The following year, Talairach [42] in Paris designed a system which involved the insertion of electrodes through a fixed grid system, which in turn invited an elaboration of the cerebral circulation. In Germany, Riechert and Wolff [43] reported in 1951 their translational system, which included the first phantom base to verify the settings mechanically.
1
Bailey and Stein [44] exhibited a burr hole mounted apparatus in the United States that same year. It was essentially a pointing device mounted on a ball and socket joint. The electrode was aligned with the intended trajectory drawn on both the AP and lateral films, resulting in an angular adjustment pointing at the target. Parallax was corrected by repeatedly adjusting to repeated films as the electrode was further advanced, a system of successive approximations. Special note must be made of the development of Narabayashi’s [45,46] system in Japan. He had been cut off from the western literature at the end of World War II, but independently developed an apparatus under difficult circumstances [47]. After finding out that his system was not unique, he conceded that Spiegel and Wycis had preceded his contribution. In the first decade after stereotactic surgery was born, a number of centers were developed throughout the world. Leksell in Sweden [41]. Talairach and associates [48] and Guiot and colleagues [49] in France. Riechert and Mundinger [50] in Germany. Gillingham [51] in Great Britain, Laitinen and Toivakka [52] in Finland, Rossi [53] in Italy, Bertand and colleagues [54] in Canada, Velasco Suarez and Escobedo [55] in Mexico, Obrador [37] in Spain, and Bechtereva and colleagues [56] and Kandel [57] in Russia, among others. Each investigator in turn added his own embellishments and indications, and the field grew rapidly. Within 20 years, stereotactic surgery was practiced throughout the world [58,59]. It is estimated that by 1965 more than 25,000 stereotactic treatments had been done worldwide [60], and 37,000 patients had been treated by 1969 [61]. The need for communication among this handful of pioneers required meeting together, and the stereotactic societies were born. The first meeting of the International Society for Research in Stereoencephalotomy (which became the World Society for Stereotactic and Functional Neurosurgery in 1973) was held in Philadelphia
9
10
1
History of stereotactic surgery
in 1966, 20 years after the first human stereotactic procedure. The emphasis was on instrumentation, since there were still few commercially available apparatus. Everyone was invited to bring their stereotactic apparatus, or at least the portable portion. Over 40 systems were laid out on the benches in the chemistry laboratory and demonstrated. The second meeting was held in Atlantic City the following year and the emphasis was on the growing number of indications for stereotactic surgery. During the 1960’s, the Todd-Wells stereotactic apparatus became the most popular in the United States, and the Leksell and RiechertMundinger systems in Europe [62,63]. There was little further modification of apparatus during the next decade, when the emphasis shifted to indications and results. The indications were there: There are four fields in which stereotactic functional neurosurgery has found a place, and all four were established during the first decade after stereotactic surgery was introduced: movement disorders, pain, epilepsy and psychiatric abnormalities. The most cited is the use of stereotactic techniques for treatment of movement disorders. Historically stereotactic surgery was developed to offer a more refined way of doing psychosurgery, without the gross inaccuracies and complications of pre-frontal lobotomy. Treatment of pain was addressed very early in the history of this new technique. Although there were early attempts at treating epilepsy with stereotactic surgery, they were less successful and became more important later, after imaging and invasive monitoring were introduced.
Movement Disorders – the Early Years Although the origin of functional neurosurgery was motivated by a need to improve psychosurgery, the first patient had a motor disorder (Huntington’s chorea).
There was growing understanding that interruption of the extrapyramidal system might benefit patients with movement disorders, particularly Parkinson’s disease, but pre-existing surgical techniques had unacceptable morbidity and mortality. The evolution of surgery for Parkinson’s disease deserves mention. Before the introduction of stereotactic techniques, the most common surgery for movement disorders involved resection of the motor or pre-motor cortex, since it was proposed as early as 1932 by Bucy and Buchanan [64] that it was necessary to interrupt the pyramidal system to obtain relief of athetosis. Shortly after that, Bucy [65,66] also claimed relief of tremor after interruption of the pyramidal system, even though the procedure was fraught with severe post-operative deficits. In 1933, Putnam [67] interrupted the proprioceptive input by posterolateral cordotomy in an attempt to lessen tremor and rigidity with less risk. Walker [68], on the other hand, who had reported mesencephalotomy for pain in 1942, attacked the extrapyramidal pathways at that same mesencephalic level in 1949 by incising the peduncle [68,69], which was reported also about that same time by Guiot and Pecker [70] (and will have some historical importance later in this story). It was the often under-appreciated Russel Meyers [22,23] at the University of Iowa who provided the surgical observations that led to the definition of targets that would soon thereafter become targets for stereotactic surgery. Up to the early 1940s it was thought that a lesion involving the extrapyramidal system would result in irreversible coma, a conclusion that was based on Dandy’s observations that frontal lobe stroke involving the basal ganglia was invariably lethal [71,72]; because Dandy has so stated it was not challenged. Meyers [73] defied that convention and reported in 1939 benefit in a single patient with Parkinson’s disease from surgical extirpation of the head of the caudate nucleus. He was encouraged to investigate further, and he devised
History of stereotactic surgery
intricate non-stereotactic interhemispheric and transventricular approaches to interrupt the ansa lenticularis at the base of the globus pallidus, which he reported in 1942. Since there were no drugs effective against the manifestations of Parkinson’s disease, adventurous surgery was often indicated, but even Meyers cautioned against its general use, as he reported a mortality rate of 15.7%. In contrast, within a decade, after stereotactic surgery was introduced, Spiegel and Wycis [74] reported operative mortality of 2% to attain an even more accurate interruption of the extrapyramidal system. Riechert and Mundinger [75] soon after reported a mortality rate less than 1%, where it remains today. Although the first reported stereotactic case involved pallidotomy for Huntington’s chorea [5], Spiegel and Wycis were initially reluctant to lesion that structure for Parkinson’s disease for fear that the hypokinesia that is seen after experimental pallidotomy might make Parkinson akinesia worse. However, Hassler and Riechert [76] reported in 1954 that they had successfully treated Parkinson’s disease as early as in 1952 by ventrolateral thalamotomy. This encouraged Speigel and Wycis [77] to lesion the ansa lenticularis fibers as they emerged from the pallidum, which they termed pallido-ansotomy. At about the same time, Narabayashi and Okuma [33] made a lesion within the pallidum with procaine-oil injection. The first patient who had stereotactic surgery in 1947 had two lesions made by alcohol injection, one in the globus pallidus and one in the dorsomedial thalamus [5]. The reasons for using two targets were [6] to interrupt the extrapyramidal circuit in the pallidum and [7] to lessen the emotional tone of the patient by interrupting the thalamic projections to the frontal lobe, since it was recognized that tension made the chorea worse. Although the patient had only moderate but temporary improvement, it had been demonstrated that pathways could be interrupted with minimal risk and provide
1
improvement in motor control while alleviating involuntary movements. During the next 15 years, a number of targets were identified for a variety of movement disorders. Patients with intention tremor and Parkinson tremor were often considered together. Intention tremor was treated with lesions in the ventrolateral nucleus [78,79], as was hemiballismus [80]. Substantia nigra lesions were also used for hemiballismus [19,81] and hypertonus [82]. There were reports of improvement with targets in the globus pallidus [74,83]. The most common target for tremor at that time was in the ventrolateral thalamus [78,84,85]. Hyperkinesia was treated with a variety of targets [75,86,87]. Hassler [88] had not yet published his nomenclature of the subdivisions of the thalamus. The most common indication for stereotactic surgery was Parkinson’s disease. This was prior to the age of l-dopa, and pharmacologic management was not adequate. Consequently, patients were referred to the surgeon earlier in the course of the disease than now, although with more severe symptoms. The later stage bradykinesia was seen less, and patients more often presented with tremor as the primary symptom. There was no such thing as dopamine dyskinesia, and one must remember in reading the literature that the patients who were treated up to the end of the 1960s were different from the patients we see today [40]. Also some patients suffered from postencephalitic parkinsonism after the big Spanish flu epidemic in 1918, a condition which is seen only exceptionally nowadays. Irving Cooper [89] first appeared in the functional neurosurgery literature in the early 1950s. During a Walker pedunculotomy [69] for Parkinson’s disease, Cooper inadvertently interrupted the anterior choroidal artery in a now famous ‘‘surgical accident,’’ so he clipped the vessel and stopped the surgery. The patient awoke with relief of symptoms and no deficit, so, as he reported in 1953, Cooper advocated anterior choroidal ligation as a treatment for parkinsonism [89]. However, both
11
12
1
History of stereotactic surgery
the results and complications were very variable. He correctly concluded that he had often produced a stroke in an area involved with the production of parkinsonian symptoms, and the globus pallidus had been cited by others as a stereotactic target, so in 1955 he began to inject free-hand that structure with alcohol, so-called chemopallidectomy [90]. He did not use stereotactic localization (his injection guide which he introduced later was not Cartesian nor based on cerebral localization) but he reported anecdotally a relatively large series of patients. The trajectory that Cooper had selected for chemopallidectomy was through the temporal lobe, aiming upwards toward the pallidum. One patient who had a good result died of other causes, and the autopsy revealed that the needle had advanced further than intended and the lesion was actually in the ventrolateral thalamus, so Cooper moved his intended target to that area which had been introduced earlier by Hassler and Riechert. In 1958 Cooper reported a series of 700 patients contrasting chemopallidectomy and chemothalamectomy [91,92] and the series had grown to 1,000 patients by 1960 [93]. He still used alcohol injection, which was a problem, since it would spread in an uncontrolled fashion. Consequently, he changed from injection of alcohol with a brain needle to a cannula with a balloon at the tip that was intended to leave a cavity that would retain the injection material, but it did not [36]. This fared no better and its blunt tip caused damage throughout the insertion tract, so he introduced a thick injection material, Etopalin [93], but still encountered the problem of uncontrolled spread. This led him to introduce the Cryoprobe [39] in 1961 – although it had a relatively large diameter and blunt tip, it made a lesion by destroying the tissue by freezing a ball of tissue at the end of the probe, which he continued to use for the remainder of his surgical career. If we might digress, the story of the Cryoprobe provides a little-known but fascinating coincidence. The original Spiegel-Wycis report introducing human stereotactic surgery [5] in
1947 had four coauthors. In addition to Spiegel and Wycis, there was Marks, the machinist from Temple Medical School who actually made the device. The fourth author was A.St.J.Lee, who was a college student who worked as a handyman in Spiegel’s laboratory. He soon after quit his job and returned to school, completed an engineering degree and became a freelance engineering consultant. The second author on the original Cooper report of ‘‘cryostatic congelation’’ is the same Arnold St. J. Lee [39], who was coincidentally hired by Cooper to design and build the Cryoprobe. I (plg) understand that a dispute arose over who owned the patent, which was never fully resolved. By 1954, Hassler and Riechert [76] had defined their thalamic targets more precisely, with the Vop recommended for tremor and the Voa recommended for rigidity. This was made possible by Hassler’s [88] subdivision of the thalamic nuclei, so that the results of lesioning more precisely localized targets allowed better correlation with clinical results. Spiegel and Wycis [74] still preferred the pallidum as the target, and in 1958 they observed that the lesion would be more effective against rigidity if it were placed more posteriorly than the tremor target in the emerging ansa lenticularis fibers. In 1959, Svennilson and coauthors [52] reviewed Leksell’s cases and also advocated a pallidal lesion more ventral and posterior than that used by other authors (in a study later cited by Laitinen [94] that led to the rebirth of VP pallidotomy). Another interesting study was published in 1959 by Levy who was a stipendiary of the Swiss Academy of Science spending several years under the guidance of Riechert in Freiburg [95]. He compared the differences in the pallidal target used by various groups at that time. This study did not only consider the target coordinates but also the angle of the trajectory in relation to the line between the foramen of Monro and the posterior commissure. There was a large variety considering both the target in the pallidum itself
History of stereotactic surgery
but also in the attempts to include pallidofugal fibers. Hassler and colleagues considered pallidotomy to be not effective for alleviation of hypokinetic symptoms because it was thought that a negative motor symptom could not be improved by a lesion: ‘‘for theoretical reasons alone, one could hardly expect an improvement of akinesia by a stereotactic operation’’ [7,96]. Nevertheless, despite the theoretical difficulties it was occasionally acknowledged that pallidotomy indeed improved akinesia. In the late 1950s, however, most stereotacticians followed the lead of Hassler and Riechert to the thalamus, since the most dramatic effect on the early Parkinson’s disease before the l-dopa era was prompt and complete relief of tremor, the dominant symptom seen at that time. The optimal target for tremor was eventually accepted to be the Vim nucleus [75,97–99], and the pallidum was almost abandoned [100]. Meanwhile, Spiegel and Wycis [101–103] moved their lesion for Parkinson’s disease to Forel’s field, a procedure they called ‘‘campotomy,’’ for campus Foreli. In 1961, microelectrode recording (MER) from the human brain was introduced by AlbeFessard [104]. It was adopted as a surgical tool to the point where many consider it an essential technique [105–107]. A number of authors used MER to confirm localization in the Vim nucleus [54,108,109], but MER was not yet a routine part of stereotactic surgery, in part for the lack of adequate recording in the unfriendly operating room electrical environment. (It later became important in defining the subthalamic nucleus [110,111]). Both thalamotomies and pallidotomies were also popular for the treatment of dystonia in the 1960s. The results were more variable than those reported for parkinsonism [8,112] but overall beneficial results were achieved in more than 50% of patients including generalized and cervical dystonia. There is published experience with more than 300 patients with cervical dystonia [104,113]. Hassler elaborated a complex model
1
to explain the deviation of the head along its different axes including the interstitial nucleus of Cajal, the pallidum and thalamic nuclei [114,115]. According to his pathophysiological concepts, Hassler suggested choosing different targets depending on the pattern of cervical dystonia in the individual patient. In cases with rotatory torticollis, the primary target was the nucleus ventralis oralis internus and its interstitiothalamic pathway, while for treatment of horizontal torticollis pallidofugal pathways in Forel’s field H1 and the ventro-oral thalamic nuclei were approached. Working further along this concept Sano even started to use the interstitial nucleus of Cajal as a target for cervical dystonia [15,116]. The elaborate selection of surgical targets within the basal ganglia circuitry according to the phenomenology of dystonia in individual patients certainly was fascinating, although it was never conclusively shown that such superselective approaches indeed resulted in improved clinical outcome. It was estimated that by 1965 more than 25,000 functional stereotactic procedures for Parkinsonism had been performed worldwide. Things changed drastically in 1968, when l-dopa became generally available [59]. Within a few months, the number of patients with Parkinson’s disease presenting for surgery plummeted. Only a few patients with primary tremor went to stereotactic surgeons during the next few years. The number of neurosurgeons doing stereotactic surgery declined, and the field was mainly practiced by a few neurosurgeons in academic centers who enjoyed the challenge. There were three principal factors that lead to the abandonment of functional stereotactic surgery for dystonia in the late 1970s, about 10 years later than for parkinsonism. First, the general decline of movement disorders surgery at that time, second, the introduction of selective peripheral denervation by Bertrand, and third, the widespread use of botulinum toxin thereafter. It was not until the introduction of CT-directed targeting,
13
14
1
History of stereotactic surgery
then MRI targeting, and later the reintroduction of deep brain stimulation that stereotactic surgery rebounded, and it is now more active than ever [59]. Eventually, it was recognized that l-dopa is not a permanent answer to Parkinson’s disease. As the disease progresses with long-term l-dopa administration, the symptoms become more refractory and dopa dyskinesias may limit the tolerated dose [117,118]. The response to medication may vary significantly and abruptly, producing repeated on-off episodes during the day. Rigidity and akinesia may progress, and freezing episodes may intervene. By the early 1980s, a search was on for a surgical answer to these problems. In 1985, Backlund and associates [119] reported their first clinical trials of autologous transplantation of adrenal medullary tissue into the head of the caudate nucleus in two patients. They were encouraged enough to conclude that ‘‘the results merit further clinical trials.’’ Two years later, Madrazo and colleagues [120,121] reported two additional patients and cited more favorable results. Later that year they reported that they had operated on 18 patients [120]. This provoked great interest and programs were begun at a number of institutions. Unfortunately, the results were modest and of limited duration, [122–124], the surgery to retrieve one adrenal gland was stressful to this fragile group of patients, and the adrenal was sometimes found to be atrophic [125]. In addition, there were a number of complications resulting from the craniotomy, if open surgery were used [126], but few complications arose when the adrenal tissue was injected using stereotactic methods [127]. By 1991 the procedure was essentially abandoned. The idea of tissue transplantation continued to garner interest, and attention was shifted to the transplantation of fetal nigral tissue. As early as 1984, a symposium was devoted to the potential neurosurgical use of fetal cells [128]. The experimental groundwork had been done in rats in the
late 1980s [129] or MPTP-injected monkeys [130, 131]. In 1989, the first two patients to receive fetal cell transplantation for the treatment of Parkinson’s disease were reported by Lindvall [132], a multi-institutional consortium was established and others patients followed soon afterward [124]. The problems that existed with such a complex program caused interest to wane. There was considerable difficulty obtaining fetal tissue at the proper age and in the proper amounts, identification of the proper tissue in the fetal brain was difficult, coordination between the service obtaining the tissue and the implantation surgery team was expensive and elaborate, and the ultimate goal of the tissue remained in doubt [122,123,133]. The field of functional neurosurgery was otherwise quiet, through the 1970s and 1980s, except for long-established procedures primarily for pain and epilepsy [59]. There was little psychosurgery being done, mostly because of a public campaign against it [134] and the development of new psychotropic drugs, rather than any scientific reason. Let us pause in the story about stereotactic surgery for movement disorders, as we might catch up to this date the story of other fields of progress, to a great extent in technology. Stereotactic and functional neurosurgery has by its nature been dependent on advances in other fields, such as radiology and later imaging, radiotherapy, computer science, and miniaturization and implantation of electronic devices.
Persistent Pain Let us return to the birth of stereotactic surgery in 1947. Many of the earliest patients had intractable pain [135]. An aside – Spiegel always referred to such pain as ‘‘so-called intractable pain.’’ If there were a chance it might be treated, it would not be intractable, by definition, so truly ‘‘intractable’’ pain was not an indication for the treatment
History of stereotactic surgery
under discussion. I (plg) prefer not to use the term ‘‘chronic pain,’’ since it includes all longlasting pain, whether from cancer or from a noncancer (or unknown) etiology, which are two different clinical conditions managed in very different ways. Instead, I insist on the use of ‘‘persistent pain’’ if both cancer and non-cancer etiologies are discussed together [136]. Functional neurosurgical management of pain had been considered for almost a century prior to the advent of stereotactic surgery [137]. One of the prime indications was facial pain, particularly trigeminal neuralgia. In 1853, Trousseau [138] suggested that the paroxysmal activity of trigeminal neuralgia resembled epilepsy and suggested gasserian ganglionectomy, but because surgery in that area was too adventurous at that time, it was not reported formally until 1890 by Rose [139]. Horsley and coauthors [140] advocated epidural total gasserian ganglionectomy by a transsphenoidal approach in 1891, but a mortality rate of 20–25% was prohibitive, even in his excellent hands. In 1900, Cushing [141] modified the procedure, and in 1920 he reported 298 consecutive cases without mortality [142]. Ramonede [143] had already introduced the suboccipital approach in 1903, and Dandy [144] modified it in 1925. In 1931, Kirschner [145,146] introduced electrocoagulation of the gasserian ganglion. He developed an apparatus to guide the electrode through the foramen ovale, which was not Cartesian and therefore not stereotactic. His use of electrocoagulation in the human was encouraging to those who later used it in the brain. At the time that stereotactic surgery was introduced, the philosophy for pain management was to interrupt the primary pain pathway, and the most inviting target was the spinothalamic fiber bundle. Although the first patient operated with stereotactic surgery had a motor disorder, Huntington’s chorea, the second had persistent pain. Walker [26] had earlier described the surgical section of the pain pathway at the level of the mesencephalon, and Spiegel and
1
Wycis made lesions in a mesencephalotomy target [135,147], although again they made a second lesion in the dorsomedial nucleus of the thalamus, as well. Walker had reported dysesthesia from the medial lemniscus being involved with his mesencephalotomy incision, but stereotactic techniques avoided that complication. If the lateral spinothalamic tract is sectioned by open surgery in the spinal cord, the procedure is anterolateral cordotomy [148–150]. If sectioned in the brain, the spinothalamic fibers that have not already synapsed in the lower brain stem lie just behind the medial lemniscus at the level of the mesencephalon. Unfortunately, when the surgical section included the medial lemniscus, severe dysesthesia might occur [151], and that was more easily avoided with stereotactic than with open surgical techniques [152,153]. The quintothalamic fibers, which descended from the trigeminal area to join the spinothalamic bundle, provided a more compact target, so a stereotactic procedure was done for facial pain quite early [135]. Again, dorsomedial thalamotomy was often recommended to relieve the emotional component of persistent or chronic pain [154]. Thalamic pain was managed with lesions in the ventral posteromedial nucleus and the centre median nucleus [42]. In 1949, He´caen, Talairach and associates [42,155] reported on a thalamotomy for persistent pain, but they interrupted the diffuse pain projection system in the centrum medianum plus a lesion in the ventrobasal complex, demonstrating successful relief of pain without interrupting the primary pain pathway. Spiegel and Wycis [156] reported similar findings for facial pain in 1953 and further refined the procedure in 1964 by defining several distinct targets within the mediobasal thalamus [157]. The concept of interrupting the reticulospinothalamic tract for pain management led Nashold [158] in 1969 to extend the mesencephalic lesion into that bundle. This in turn led him to explore the mesencephalon with implanted electrodes to define the multisynaptic pathway [159].
15
16
1
History of stereotactic surgery
The result was to leave the primary pain pathway intact and to confine the lesion to the extralemniscal area, which provided equivalent pain relief with minimal risk of dysesthesia [151,160]. This demonstration led to the limbic system becoming a primary target for pain management. Foltz and White [161] (as reported by Sweet and Gybels [136], who previously had considerable experience with the use of cingulotomy for psychiatric disorders, began in 1962 to use that procedure for pain management. Long-term relief was reported in over half the patients with cancer pain and one-third with ‘‘failed back syndrome.’’ Although it is functional but not ordinarily stereotactic, percutaneous cervical cordotomy must be mentioned. In 1959, Mullan [162] introduced the implantation of a radioactive seed as a method for producing lesions for motor disorders, such as Parkinson’s disease. In 1963, he introduced the use of a radioactive strontium needle inserted percutaneously between the arch of the first and the lamina of the second cervical vertebrae, lain against the anterolateral surface of the spinal cord for a measured duration, in order to make a cordotomy lesion to manage pain without the need for surgical exposure [163]. This approach was used in 1965 by Rosomoff [164] who made a controlled lesion with a radiofrequency electrode, which was much more practical. This more practical procedure was adopted by many neurosurgeons. There was one grave risk, however, that of Ondine’s curse or sleep-induced apnea after bilateral lesions were made [165]. Unilateral percutaneous cervical cordotomy, however, had good results with reasonable risk, particularly in cancer patients but not particularly in patients with other persistent pain. Nevertheless, since many of the cancer patients required bilateral pain management, Lin, Polakoff and I [166] introduced in 1966 a technique to introduce the electrode through a disk to the lower cervical spinal cord, which provided good relief of bilateral pain but with occasional weakness of the hand.
Back to 1937, when Sjo¨qvist [167] presented a technique, later modified by White and Sweet [168] for surgical interruption of the descending trigeminal tract at the level of the medulla, limiting the analgesia to the distribution of the trigeminal nerve. In 1970, Hitchcock [169] made the procedure stereotactic by inserting the electrode through the foramen magnum, an approach that was uniquely suited to his stereotactic apparatus, and reported good results in postherpetic facial neuralgia. He also used that approach to modify percutaneous cervical cordotomy by inserting the electrode from dorsally through the uppermost spinal cord into the anterolateral spinothalamic tract [170], which was similar to the approach Crue and his colleagues [171] reported in 1968. This led to an inadvertent discovery of a new pain pathway [172]. One of Hitchcock’s patients reportedly moved as the electrode was inserted, and the lesion was made in the midpoint of the spinal cord at the cervico-medullary junction, with an excellent clinical result and no complications [173]. He theorized that he had interrupted a pain pathway, probably the spinoreticulothalamic multisynaptic path, and began to use the same target for pain management. He and later Schvarcz [174] termed the procedure extralemniscal myelotomy. In an attempt to avoid making a cervical lesion to treat pain confined to the pelvic area, Gildenberg and Hirshberg [175] made a mechanical lesion under surgical exposure at the thoracolumbar spinal cord level with good relief of particularly visceral cancer pain, a procedure they dubbed limited myelotomy. The spinal cord of one of the patients who had later died from his cancer was studied by Al-Cher, Willis and his group [8], who demonstrated that there was indeed a pathway, previously undescribed, not multisynaptic, transmitting pelvic visceral pain. This brings us to the point in the discussion where pain management intersects with chronic stimulation of the nervous system, the beginning of so-called neuromodulation.
History of stereotactic surgery
Neuromodulation-Stimulation There is a long history of application of electrical stimulation to the nervous system. It is reported that Scribonius treated gout pain by application of an electric torpedo fish as early as 15 AD [176]. One of the earliest to perform controlled stimulation with close observation of muscle contraction was Benjamin Franklin [177] as early as 1774, several years before Galvani [178] demonstrated electrical contraction of frog muscle in 1780. Perhaps the earliest treatise on potential physiologic use of electricity was written by Mary Shelly [179] in her novel Frankenstein, which was based on scientific speculation previously published by Dr. Erasmus Darwin, Charles Darwin’s grandfather. In 1870, Fritz and Hitzig [180] demonstrated that limb movement occurred on stimulation of the motor cortex of the dog. Electrical stimulation was applied to the brain of an awake patient soon after, in 1874, when Barthalow [181] stimulated the motor cortex that was exposed after debridement for osteomyelitis. It was left to Sir Victor Horsley [182] to first use intraoperative stimulation in 1884, when he stimulated an occipital encephalocele and noted conjugate eye movement. The first electrical stimulator designed specifically to treat pain appeared in the early 1900s when the Electreat was advertised to relieve not only pain, but innumerable physical maladies, as well. It was battery operated and bore an uncanny resemblance to TENS units that appeared 70 years later. Stimulation during stereotactic surgery was used by Spiegel and Wycis from the very first case in order to obtain physiologic localization of the electrode. Since the first lesions were in the pallidum and soon after the thalamus, stimulation was used to assure that the electrode or needle did not lie in the internal capsule by observing for a contralateral involuntary motor response. Although various frequencies were used, there
1
was not originally a distinction made between high frequency and low frequency stimulation, nor were the responses to stimulation recorded with detail. Hassler [183] was the first to suggest that observations made on stimulation in the operating room might have long-term effects. He had been a graduate student of Rudolph Hess in Switzerland during the late 1940s, where chronic stimulation was routinely administered to cats with permanently implanted electrodes. They recognized that stimulation might produce the same effects as making a lesion at the same site, but changing the frequency might give the contrasting effect. Throughout the stereotactic community, little attention was paid to high frequency versus low frequency stimulation. Those terms were not universally defined, so some institutions tested the patients with significantly different parameters. In addition, stimulation caused seizures in some rare instances, so higher frequencies were sometimes avoided. Even so, intraoperative stimulation became the norm prior to lesion production. For instance, Spiegel and Wycis [103] made lesions in Forel’s field for the treatment of Parkinson’s disease, but stimulated critically prior to lesion production. The desired target lay just above the emerging oculomotor fibers, so they stimulated at successively deeper levels until uni-ocular deviation demonstrated that they had impinged on the oculomotor fibers. The electrode was then withdrawn 2 mm and a lesion was made. Tasker [184] performed an extensive stimulation study of the human thalamus, which he published as a physiological atlas in 1982. The technique was very similar to that he used as a graduate student with Clinton Woolsey from 1961 to 1963 to map the cortex and subcortex of animals [185]. Since there was no way to apply stimulation for prolonged periods, the therapeutic potential of stimulation of the brain was rarely pursued.
17
18
1
History of stereotactic surgery
Perhaps the earliest was Pool [186] who stimulated frontal tracts rather than performing prefrontal lobotomy for psychosurgery in older patients as early as 1948. In the early 1950s, Heath [187] stimulated a variety of subcortical areas and made detailed observations about behavioral changes. In 1954, Olds and Milner [188] observed that rats would aggressively seek stimulation of the septal area, which they concluded provided the rats with intense pleasure. That same year, Heath [189] concluded that, since pleasure is the opposite of pain, septal stimulation might be used to treat pain. There were no implantable stimulators, so he stimulated the septum for intervals of 15 min daily to weekly and was able to alleviate cancer pain in one patient. A decade later, when chronic stimulation was available, Gol [190] implanted stimulators in the septal area and had relief of cancer pain in several patients. In 1972, Bechtereva [191] reported what may have been the first therapeutic use of stimulators in motor disorders. However, implantable stimulators were not available in Russia, so again the stimulation was applied to the pallidum or thalamus only intermittently. In the meantime, significant advances had been made in the understanding of pain perception. The concept that provided the basis for the use of stimulation in persistent pain was the presentation of the Melzack-Wall [192] gate theory in 1965. They proposed that pain perception involves a ‘‘gate’’ that can be opened or closed to allow pain sensation to pass through or to block it. If the activity of the small pain fibers predominates, the gate will open. If the large non-pain proprioception and touch fibers predominate, te gate will close and pain will decrease (‘‘If you rub it, it feels better!’’). Since the large non-pain fibers are isolated in the dorsal columns, stimulation of that part of the spinal cord should close the gate and provide relief of pain. Stimulating the skin with TENS unit electrodes applied to the skin and adjusting the sensation so it is below
pain threshold [193]. It could also be done by stimulating through an electrode placed behind the spinal cord (and in other positions, as it turned out), which provided the impetus to provide an implantable stimulator. The gate theory was tested in 1967 by Wall and Sweet [194] who stimulated their own infraorbital nerves. Sweet recruited Roger Avery, an engineering colleague at MIT, to make an implantable stimulator, which he and Wepsic [195] used to provide peripheral nerve stimulation for pain management. At about the same time in 1967, Norm Shealy [196] stimulated the large nerves where they were uniquely gathered in the dorsal columns of the spinal cord. He theorized that the impulse would travel retrograde down the dorsal columns to inhibit the small nerve input at each level of the spinal cord to close the gate and diminish pain sensation. He recruited Thomas Mortimer, a graduate engineering student, to design an implantable stimulator. The first mode required an external power supply connected to the stimulator by needles inserted through the skin. By coincidence, Mortimer had interviewed for a job at Medtronic 2 years before that. He contacted Norm Hagfors, one of the engineers he had met, to see if their cardiovascular stimulator might be adapted to stimulate the spinal cord. Mortimer designed an implantable electrode that Shealy used with the Medtronic cardiac stimulator, which provided relief of pain for the last several months of a cancer patient’s life. Shealy contacted Medtronic to improve and provide the system for more patients. Medtronic had previously been in the business of manufacturing implantable cardiovascular stimulators. In 1963, they were making the Barostat, which was used to stimulate the carotid sinus for treatment of hypertension. In 1965, they released the Angiostat, which stimulated the carotid sinus for angina. Shealy’s second patient used that stimulator attached to the electrode that Mortimer had designed and had good relief of chronic pain for
History of stereotactic surgery
4 years. In 1968, Medtronic provided this system for spinal cord stimulation as the Myelostat. The early spinal cord stimulators came in two parts. An internal implantable system had no internal power supply, since implantable batteries were just being developed. It was attached to a circular wound antenna implanted subcutaneously. The external unit had the controls and a battery, and transmitted both the power and the control signal transcutaneously. By 1981, battery technology had advanced to the point where the entire unit could be implanted. Avery had kept pace with Medtronic until that time, but when Roger Avery retired, the company no longer provided implantable spinal cord stimulators. In 1971, I (plg) was working at the Cleveland Clinic just down the street from where Shealy and Mortimer had introduced spinal cord stimulation. I was able to obtain stimulators from Medtronic modified to provide frequencies of 800–1,200 Hz. They were implanted to stimulate at the C2 level for the treatment of spasmodic torticollis. Half the patients had significant relief, which was the first use of implanted stimulators for motor disorders [197]. In 1976, both Cook [198] and Dooley [199] recognized improvement in spasticity in multiple sclerosis patients who had had spinal cord stimulators implanted for pain of muscle spasm. In addition, Dooley [200] recognized improvement in peripheral blood flow in patients undergoing spinal cord stimulation for pain management. It was in 1973 that Hosobuchi [201] implanted a stimulator attached to an electrode in the somatosensory thalamus for treatment of denervation pain with anesthesia dolorosa, and the field of deep brain stimulation (DBS) was born. There were few attempts during the 1970s to introduce DBS also for the treatment of movement disorders. In 1977, Mundinger reported on the benefits of unilateral thalamic DBS in seven patients with cervical dystonia [202]. Intermittent stimulation for 30–40 min with frequencies up to 150 Hz resulted in improvement of dystonia
1
for up to 7 h. The longest follow-up record was 9 months. The results were never published in the English literature and the study was completely forgotten for decades. Let us go back to 1969. Reynolds [203] demonstrated in the rat that periventricular stimulation produced profound analgesia, sufficient to perform surgery with no apparent pain perception. In 1977, Richardson and Akil [204] provided chronic periventricular stimulation to patients and produced marked reduction in pain. The following year, they demonstrated that application of such stimulation was associated with release of endorphin into the ventricular fluid [7]. Also in 1977, spinal cord stimulation and DBS were being used so extensively that the Food and Drug Administration held a symposium on safety and efficacy [205]. It was felt that the use of stimulators for pain had been documented, but not for other uses, which at that time included movement disorders, epilepsy, cerebral palsy and bladder control. At about that same time, the FDA was given the charge to regulate devices as well as their historical charge to regulate drugs. They felt that relief of pain by deep brain stimulation had not been sufficiently documented and gave the manufacturers several years to provide the data to document that such devices should be continued, especially for pain management. Of the three manufacturers, only Avery provided data, but just then Roger Avery retired so that product was discontinued, and the use of DBS for pain management was de-approved. In 1991, Tsubokawa [206] reported on the stimulation of the motor cortex, but not the sensory cortex, for the management of central pain. In 1995, Migita [207] reported on the use of extracranial magnetic stimulation of the motor cortex for pain management. As a general rule, ablative procedures are more usually indicated for cancer pain, but stimulation procedures for non-cancer persistent pain [208,209].
19
20
1
History of stereotactic surgery
In 1979, the senior author commented that the engineers could give us any stimulation parameters we wanted, but it was up to the surgeons and scientists to let them know what we need [210]. That has not changed.
The Return of Movement Disorder Surgery When l-dopa became widely used for treatment of Parkinson’s disease in 1968, stereotactic surgery was rarely used for that indication. Although there were other indications for stereotaxis, they were not enough to maintain a critical mass of centers, so stereotactic surgery went into a marked decline [59]. It was a seminal report by Laitinen [94,211] in 1992 that re-awakened interest once again in the use of pallidotomy for Parkinson’s disease, and with that awakened the field of stereotactic and functional neurosurgery. It had become obvious that many of the patients who took l-dopa for years were still disabled and needed a more aggressive therapy. Even more, it was recognized that chronic l-dopa treatment resulted in disabling dyskinesias and unpredictable motor fluctuations. Laitinen recalled the report of Leksell’s pallidotomy that had been published in 1960 by Svennilson and associates [52] who had reported more improvement in bradykinesia and rigidity with a lesion that Leksell had made in the ventral posterior area of the pallidum than with the more central lesion. He attributed this observation to the inclusion of the pallidofugal fibers in the lesion, similar to Spiegel and Wycis [212] pallido-ansotomy. After a cautious start in 1987, Laitinen and his colleagues [94,211] reported in 1992 a significant improvement in rigidity, bradykinesia and dopa-induced dyskinesia in patients who had been taking l-dopa for years for their Parkinson’s disease. There was an enthusiastic return to stereotactic pallidotomy for Parkinson’s disease [213]. The setting was more advanced than had
existed in the original years of surgical management of Parkinson’s disease. The evaluation of results was more sophisticated [214], localization of the target had presumably become more accurate with the introduction of imaging techniques [215], and microelectrode confirmation of the target [216] may have helped to localize the lesion. In addition, it was found that pallidotomy managed not only the symptoms of Parkinson’s disease, but the side-effects of l-dopa, such as dyskinesia. The first ‘‘modern’’ pallidotomy in the United States was performed in December, 1991, in the Hospital for Joint Diseases, New York, by Michael Dogali [217] in the presence of Laitinen and Tasker. Dogali initially considered the possibility of placing lesions in the subthalamic nucleus after reading reports on the experimental studies of the Atlanta group and discussed this issue with Ransohoff. As a neurosurgeon who had actively participated in the first wave of functional stereotactic surgery, however, Ransohoff strongly opposed using the subthalamic nucleus because of the fear of producing hemiballism and instead suggested using the pallidum. Many pallidotomy series were reported over the next 5 years [213,218]. Many of the neurologists who referred such patients became involved in the intraoperative microelectrode evaluation of targets [216], further encouraging referral of patients. The resurgence in stereotactic activity also led to a search for new targets. Improvements in imaging during the prior decade made target visualization more feasible. Improvements in physiological localization allowed the surgeons to be more secure in identifying targets. Prior to the lull, lesions had been placed by a few adventurous neurosurgeons in the zona incerta [219] and the subthalamic nucleus [220], which became a target for DBS. In 1980, during the lull, Brice and McLellan [221] used thalamic stimulators in two patients for management of tremor of multiple sclerosis. Also during the lull, the following year, Benabid and his colleagues [222] reported on management of tremor by deep brain stimulation
History of stereotactic surgery
(DBS) in the Vim nucleus. In 1994, when Siegfried and Lippitz [223] reported on chronic electrical stimulation of the VL-VPL complex and of the pallidum, they indicated that this represented their experience since 1982. Benabid carried the banner for the use of DBS in motor disorders and popularized the field by his results and the study of the effects of DBS on various targets. Based on laboratory studies that stimulation of the subthalamic nucleus improves the symptoms of Parkinson’s disease, they have concentrated on this target, which has become the primary target for such stimulation [224]. Since the effects of moderately high frequency stimulation are equivalent, the targets for stimulation are for the most part the same than the targets where lesions are made. In those areas where DBS is not generally available, the success of subthalamic stimulation has led to renewed interest in lesioning targets [225]. Other targets have presented themselves, in part because the risk of permanent complications are less with stimulation than with lesions, so surgeons can be more adventurous. Velasco and his group [226] stimulate the prelemniscal radiations with success, a target just lateral to the old campotomy target [227]. DBS has become increasingly popular in Europe during the past decade and in the US since it was approved in 2002 [72,228]. The renaissance of functional stereotactic surgery for dystonia occurred with a delay of about a decade after that of Parkinson’s disease. With that consideration, we have to remember that also the decline of dystonia surgery happened a decade later than that for Parkinson’s disease in the 1970s. Encouraged by the improvement of dyskinesias in Parkinson’s disease patients after pallidotomy, the GPi was reintroduced as a target for radiofrequency lesioning in the early and mid 1990s [112], While it was shown that pallidotomy yielded beneficial results, it became also clear very soon, however, that bilateral surgery which is necessary in dystonia
1
was burdened with a higher frequency of side effects and that the improvement of dystonia lessened after years. Therefore, pallidal DBS was introduced in the mid 1990s for cervical dystonia [229] and generalized dystonia [230,231].
Epilepsy Surgery Let us go back in time again. Epilepsy also invited a variety of approaches [232–234]. Lesions were often made in the targets that had been defined from movement disorders, on the theory that both represented uncontrolled activation that might be propagated through known pathways. Forel’s field was one of those areas, with significant decrease in seizures being reported as early as 1963 [235]. Thalamotomy was used for myoclonic epilepsy [236]. Fornicotomy was used for epilepsy by several investigators [232,237]. Starting in the 1960s, electrodes were inserted for prolonged recording over several days to identify epileptic foci [238]. The hallmark of surgery for epilepsy remained temporal lobectomy in selected patients [239,240], or the more selective amygdalotomy [241], which was made even more precise with image guidance [242], which is discussed below. With image guidance, it was possible to insert a multi-contact depth electrode from occipitally just lateral to the hippocampus through the temporal lobe to its tip, as well as subdural electrode strips over the anterolateral surface and just beneath the temporal lobe. Prolonged recording identified which contact lay at the origin of the seizures. That contact could then be approached also with image guidance to obtain the optimal temporal lobe resection [243].
Psychiatric Surgery Again we go back in time. In the late 1930s and 1940s, pre-frontal lobotomy was introduced as being beneficial for psychiatric disorders [20].
21
22
1
History of stereotactic surgery
The indications, however, were not clear and it is very likely that many of those patients had schizophrenia, which does not respond to psychosurgery. There were no psychotropic medications other than barbiturates, and often the only alternative to surgery was for permanent commitment to a mental institution. Pre-frontal lobotomy produced poorly controlled damage to the frontal lobes, and it became over-used and abused [244]. During the 1950s, pre-frontal lobotomy became less desirable than it had been, in part because new medications became available and stereotaxis provided a better controlled alternative. The first stereotactic patient had dorsomedian thalamotomy [5]. Dorsomedian lesions were combined with lesions in the anterior nuclei for anxiety and aggressive disorders [245]. Lesions of the internal medullary lamina were again combined with hypothalamic lesions for aggressive behavior [246–248]. Surgery for psychiatric symptoms almost came to a halt in the late 1970s and through the 1980s for what appeared to a great extent for political rather than scientific reasons. As occurred during that period, protests were used to interrupt scientific meetings where psychiatric surgery was being discussed, and all surgery for psychiatric symptoms was equated in the media with the non-anatomical and abused psychosurgery of the pre-stereotactic era [134,249]. Although psychiatric surgery has resumed on a more limited basis, the availability of psychotropics and more specific psychiatric diagnosis has put it on a more scientific basis. The main indications are now recognized as obsessive-compulsive disorder [250,251] and intractable depression [252,253]. As with motor disorders, there appears to be a significant role for deep brain stimulation [254,255].
Image Guided Surgery After the first generation of stereotactic frames, there was little further modification until the late
1970s, with the introduction of image guided surgery [59]. The idea of using X-ray visualization to identify the location of radio-opaque masses is older than you think. Perhaps the birth of image guided surgery can be traced to the work of Elizabeth Fleishmann in San Francisco in 1900 [256]. When she was 28 years old, Roentgen’s discovery of X-rays captured the public’s attention, and do-it-yourself articles appeared in the newspapers. She quit her job as a bookkeeper and somehow managed to set up a complete X-ray system in an office in Sutter Street the following year. She became perhaps the best radiologist on the west coast. Soldiers returning from the Spanish-American War in 1900 were referred to her to identify bullets and shrapnel that they still carried. She developed the skill to triangulate on bullets lodged in the skull or body and to steer surgeons to them so they might be removed. She certainly qualifies as being the mother of image guided surgery. Up until CT scanning was introduced, intracerebral masses were deduced from shifts in blood vessels in angiography or the position of the ventricles in ventriculography. Now for the first time it was possible to see the mass itself, which invited stereotactic intervention as early as 1956 [257]. New techniques had to be developed to relate the target to stereotactic coordinates in order to guide the stereotactic frame with CT images rather than X-ray [258,259]. Each slice of the CT scan presented a measurable two-dimensional picture from which AP and lateral coordinates could be measured. The relative position in space of the target slice provided the third vertical coordinate. By localizing that slice on the head in order to register it to the stereotactic apparatus, the third coordinate could be determined. A biopsy cannula or other probe could be inserted in order to direct the surgeon to the pathology [260]. New stereotactic devices were developed to automate that process. The Leksell apparatus was
History of stereotactic surgery
modified with a base plate that could be secured to the scanner and then to the head frame in the operating room to transpose the coordinates to the stereotactic apparatus [261]. Another system employed a series of wires of varying lengths incorporated into plastic plates attached to the head frame. The number of wires appearing in cross-section in the target slice indicated its vertical position [262]. One ingenious system used three acrylic screws with lead markers that were secured to the patient’s skull prior to scanning in order to establish a reference plane. The base ring of the stereotactic apparatus was attached to the screws, and coordinates were calculated [263,264]. Several stereotactic devices were developed especially for use with CT (and later MRI) scanning. Some depended on the calibrated movement of the CT scanner table to establish the vertical coordinate [262,265]. A CT scanner that incorporated a stereotactic apparatus was designed [266]. In some busy neurosurgical services, a CT scanner was installed in the operating room [267]. The breakthrough in marrying to stereotactic frames CT or MRI scans, as well as more recent scanners, involved the development in 1980 of a fiducial system that contains all the three-dimensional information for targeting independently on each CT slice. The idea was invented by a medical student, who is the ‘‘B’’ in the BRW apparatus [268]. Three sets of three rods, with each set in an N-shaped configuration, are attached to a frame that is in turn attached to the stereotactic head ring. Since the center rod of each set is diagonal, the height of the slice scan be determined by the position of the center rod relative to the vertical rods on each side of it. Since there are three sets of rods, three points are used to determine the position of the plane bearing the target, and the AP and lateral coordinates can be established on that target slice. This system was first incorporated into the BRW system, a frame consisting of interlocking arcs, and
1
similar systems have been used with other apparatus as well [115,269,270]. The availability of image-based stereotactic surgery opened up new possibilities for management of brain tumors and other masses, and brought many neurosurgeons into the stereotactic arena. The most common stereotactic procedure became biopsy [63]. In addition, aspiration of abscesses [271], hematomas [272,273], and cysts [274] became commonplace. A stereotactic frame could be used to guide the patient to a tumor at craniotomy [275], with a virtual reality program visualizing the tumor beneath the surface prior to making the skin incision. Visualization of structures such as the ventricle and blood vessels can guide the surgeon. The image is upgraded as the resection proceeds, so the surgeon can see when the optimal resection has been done. Cannulas [276] or isotope seeds [277] can provide brachytherapy of various types. Kelly [269,278] is considered by many to be the father of image guided brain tumor resection. His development between 1980 and 1983 of a stereotactic frame that integrated visual guidance technology antedated frameless systems and opened the door to guided resection. The data from a CT scan or MRI was fed into a computer workstation. The computer reconstructed the volume of the target, ordinarily a tumor, into a three-dimensional volumetric object that could be registered to the stereotactic head frame, rather than a target point-in-space, as had been the practice until then. The digital manipulations of those data were even more remarkable when one considers the capability of computers of that time, which required an entire room to manage the data for a single surgery. The registration of the microscope to the patient was done by aligning the view with a cylindrical retractor. The outline of the tumor was seen on a heads-up display superimposed on the microscope view. The image was manipulated so only the crosssection of the part of the tumor being resected was seen on the display. Because of the restricted
23
24
1
History of stereotactic surgery
access through the cylinder retractor, the resection was usually done with the CO2 laser [279–281]. Several technologies for frameless image guided surgery emerged at about the same time. Since most were developed or at least used commercially early in development, much of the relevant information does not appear in any detail in the scientific literature. The first frameless systems, which became generally available about 1990, consisted of an articulated arm attached to the operating table. Each of the joints contained a potentiometer to measure the angle of each single joint. The information from all the joints could then be used to calculate the position of the tip of the pointer at the end of the arm. It was used first to register the position of the patient’s head to the arm, after which the location of the tip of the pointer in relation to the head and the target could be displayed to the surgeon in real time [282]. The original use of the operating microscope as a frameless guidance instrument was developed by Roberts [283] as early as 1986. It localized its position in space by ultrasound triangulation. However, ultrasound localization had the drawback of being subject to error from air currents or temperature gradients in the operating room. Kelly [284] introduced frameless electromagnetic localization to image guided surgery, first with the Regulus Navigator, then with his Compass system which eventually required only a lap top computer for neuronavigation. The most commonly used localization system at present involves the localization in space of a series of computer identifiable fiducials by a pair of video-like cameras overlooking the surgical field, reported by Heilbrun [285] in 1992 as machine vision. By 1994, his team reported the use of this system for multiple imaging modalities [286,287]. This system was used by Smith and Bucholz [288] in 1994 in a system they called the NeuroStation that grew up to be the Stealth Station,
now the most widely used neuronavigation system in the US (the most widely used system in Europe is the BrainLab system). Fiducials are secured to the patient’s head prior to scanning. A three-dimensional reconstruction of the head, including the fiducials, is computed. In the operating room, the surgeon localizes each of the fiducials sequentially by touching each in turn with the image localized pointer, which registers the location of the patients head to the preoperative volumetric scan. Thereafter, the surgeon can use the tip of the pointer to localize any structure of interest or to localize the tip of the pointer in relation to the target even before the incision is made. More recently, the development of computer-generated skin surface rendering has become an alternative to the use of fiducials. Other authors have used similar threedimensional volume reconstruction as early as 1994 to guide craniotomy and tumor resection by a system that visualizes the entire volume of the target, volume-in-space rather than a pointin-space [275,289–291]. The system that has been used by Gildenberg [275,292,293] localizes a video camera stereotactically, so that the video image of the surgical field is registered to the reconstructed volume of the tumor. The superimposition of both images provides virtual reality to surgical resection.
Further Steps The history of stereotactic and functional neurosurgery has been documented in the ‘‘stereotactic journal’’ as it occurred. When Spiegel first reported the birth of human stereotactic surgery in 1947, he was editor of the journal Confinia neurologica, where a preponderance of early articles was published. When Spiegel retired in 1968, Gildenberg became editor and changed the name to Applied Neurophysiology, since that described almost all of the functional procedures that constituted the field. Within a few years,
History of stereotactic surgery
image guidance, computers, and stereotactic radiosurgery became part of the field, and the name was changed to Stereotactic and Functional Neurosurgery. Half of the issues between 1968 and 2001 were devoted to proceedings of the World and American Societies for Stereotactic and Functional Neurosurgery. The philosophy was based on the feeling that if anyone wanted to keep up with the field, he or she would know what was happening in the field by reading just this one journal, a repository of stereotactic history as it happened [294]. The journal is stronger than ever under the editorial leadership of David Roberts since 2001. Stereotactic and functional neurosurgery nowadays is a major column supporting the building of neurosurgery. Its concepts have penetrated essentially all subspecialties of neurosurgery and they reach far beyond to other disciplines. Therefore, the question arises how the core of contemporary stereotactic and functional neurosurgery would be defined. In an attempt to establish a Training Chart in Movement Disorders Surgery Added Competence by the ESSFN the first request by the public authorities was to give a precise and pragmatic definition of the field nowadays. After a long discussion the following definition came up as a consensus based on the input of Drs. Blond, Broggi, Gildenberg, Hariz, Krauss, Lazorthes and Lozano: "
Stereotactic and functional stereotactic surgery is a branch of neurosurgery that utilizes dedicated structural and functional neuroimaging to identify and target discrete areas of the nervous system and to perform specific interventions (for example neuroablation, neurostimulation, neuromodulation, neurotransplantation, and others) using dedicated instruments and machinery in order to relieve a variety of symptoms of neurological and other disorders and to improve function of both the structurally normal and abnormal nervous system. The practice of stereotactic and functional neurosurgery mainly extends
1
into the fields of movement disorders, pain, epilepsy, psychoaffective disorders, neoplastic diseases of the nervous system and the restoration of function in degenerative disorders.
References 1. Gildenberg PL. General concepts of stereotactic surgery. In: Lunsford LD, editor. Modern stereotactic neurosurgery. Boston: Martinus Nijhoff; 1988. p. 3-12. 2. Horsley V, Clarke RH. The structure and functions of the cerebellum examined by a new method. Brain 1908;31:45-124. 3. Descartes R. Discours de la me´thod pour bien conduire sa raison et chercher la ve´rite´ dans les sciences. Leyden, Holland: Maire; 1637. 4. Pereria EAC, Green AL, Nandi D, Aziz TZ. Stereotactic surgery in the United Kingdom: the hundred years from Horsley to Hariz. Neurosurgery 2008 (in press). 5. Spiegel EA, Wycis HT, Marks M, Lee AS. Stereotaxic apparatus for operations on the human brain. Science 1947;106:349-50. 6. Frankfort plane. Dorlandaˆ s Illustrated Medical Dictionary. Philadelphia, PA: W.B.Saunders; 1994, p 1300. 7. Akil H, Richardson DE, Hughes J, Barchas JD. Enkephalin-like material elevated in ventricular cerebrospinal fluid of pain patients after analgetic focal stimulation. Science 1978;201:463-5. 8. Al Chaer ED, Lawand NB, Westlund KN, Willis WD. Pelvic visceral input into the nucleus gracilis is largely mediated by the postsynaptic dorsal column pathway. J Neurophysiol 1996;76:2675-90. 9. Clarke RH. Part I. Investigation of the central nervous system, methods and instruments. Baltimore: Johns Hopkins; 1920. 10. Levy R. A short history of stereotactic neurosurgery. Park Ridge, IL: American Association of Neurological Surgeons; 1992. 11. Schaltenbrand G. Personal observations on the development of stereotaxy. Confin Neurol 1975;37:410-6. 12. Olivier A, Bertrand G, Picard C. Discovery of the first human stereotactic instrument. Appl Neurophysiol 1983;46:84-91. 13. Dittmar C. Ueber die Lage des sogenannten Gefaesszentrums in der Medulla oblongata. Ber saechs Ges Wiss Leipzig (Math Phys) 1873;25:449-69. 14. Zernov DN. Encephalometer. Device for estimation of parts of brain in human (Russian). Proc Soc Physicomed, Moscow Univ 1889;2:70-80. 15. Altukhov NV. Encephalometric investigations of the brain relative to the sex, age and skull indexes. Moscow; 1891.
25
26
1
History of stereotactic surgery
16. Gildenberg PL. ‘‘Stereotaxic’’ versus ‘‘stereotactic’’. Neurosurgery 1993;32:965-6. 17. Gildenberg PL. Ernest A. Spiegel 1895–1985. Appl Neurophysiol 1985;48:I-VI. 18. Gildenberg PL. Spiegel and Wycis – the early years. Stereotact Funct Neurosurg 2001;77:11-16. 19. Spiegel EA, Wycis HT. Sterencephalotomy, Part II. Philadelphia, PA: Grune & Stratton; 1962. 20. Freeman W, Watts JW. Psychosurgery: intelligence, emotional and social behavior following prefrontal lobotomy for mental disorders. Springfield, IL: Charles C Thomas; 1942. 21. Spiegel EA, Wycis HT. Stereoencephalotomy, Part I. New York: Grune & Stratton; 1952. 22. Meyers R. The modification of alternating tremors, rigidity and festination by surgery of the basal ganglia. Res Publ Ass Res Nerv Ment Dis 1942;21:602-65. 23. Meyers R. Historical background and personal experiences in the surgical relief of hyperkinesia and hypertonus. In: Fields W editor. Pathogenesis and treatment of parkinsonism. Springfield, IL: Charles C Thomas; 1958. p. 229-70. 24. Dogliotti M. First surgical sections, in man, of the lemniscal lateralis (pain-temperature path) at the brain stem, for the treatment of diffuse rebellious pain. Anesth Analg 1938;17:143-5. 25. Schwartz HG, O’Leary JL. Section of the spinothalamic tract in the medulla with observations on the pathways for pain. Surgery 1941;9:183-93. 26. Walker AE. Relief of pain by mesencephalic tractotomy. Arch Neurol Psychiat 1942;48:865-83. 27. Keele KD. Anatomies of pain. Baltimore, MD: Williams & Wilkins; 1949. 28. Livingston WK. Pain mechanisms. New York: Macmillan; 1943. 29. Talairach J, David M, Tournoux P, Corredor H, Kvasina T. Atlas d’anatomie stereotaxique. Paris: Masson; 1957. 30. Spiegel EA, Wycis HT. Pallido-thalamotomy in chorea. Arch Neurol Psychiatry (Chicago) 1950;64:495-6. 31. Gildenberg PL. The birth of stereotactic surgery: a personal retrospective. Neurosurgery 2004;54:199-208. 32. Spiegel EA. Guided brain operations. Basel: Karger; 1982. 33. Narabayashi H, Okuma T. Procaine oil blocking of the globus pallidus for the treatment of rigidity and tremor of parkinsonism. Proc Jpn Acad 1953;29:310-8. 34. Narabayashi H, Shimazu H, Fujita Y, Shikiba S, Nagao T, Nagahata M. Procaine-oil-wax pallidotomy for double athetosis and spastic states in infantile cerebral palsy. Neurology 1960;10:61-9. 35. Cooper IS. Chemopallidectomy and chemothalamectomy for parkinsonism and dystonia. Proc R Soc Med 1959;52:47-60. 36. Gildenberg PL. Studies in stereoencephalotomy. X. Variability of subcortical lesions produced by a heating electrode and Cooper’s balloon-cannula. Confin Neurol 1960;20:53-65.
37. Obrador S, Dierssen G. Cirurgia de la region palidal. Revta Clin Esp Ano XVII 1956;61:229-37. 38. Cosman ER, Nashold BS, Jr, Bedenbaugh P. Stereotactic radiofrequency lesion making. Appl Neurophysiol 1983;46:160-6. 39. Cooper IS, St Lee AJ. Cryostatic congelation. J Nerv Ment Dis 1961;133:259-63. 40. Gildenberg PL. Functional neurosurgery. In: Schmidek HH, Sweet WH, editors. Operative neurosurgical techniques, vol. II. New York: Grune & Stratton; 1982. p. 993-1043. 41. Leksell L. A stereotaxic apparatus for intracerebral surgery. Acta chir scand 1949;99:229-33. 42. He´caen H, Talairach J, David M, Dell MD. Coagulations limite´es du thalamus dans les algies du syndrome thalamique. Rev Neurol (Paris) 1949;81:917-31. ¨ ber ein neues Zielgeraet zur 43. Riechert T, Wolff M. U intrakraniellen elektrischen Abteilung und Ausschaltung. Arch Psychiatr Z Neurol 1951;186:225-30. 44. Bailey P, Stein SN. A stereotaxic apparatus for use on the human brain. Atlantic City: AMA Scientific Exhibit; 1951. 45. Narabayashi H. Stereotaxic instrument for operation on the human basal ganglia. Psychiatr Neurol Jpn 1952;54:669-71. 46. Uchimura Y, Narabayashi H. Stereoencephalotom. Psychiatr Neurol Jpn 1951;52:265. 47. Ohye C, Fodstad H. Forty years with Professor Narabayashi. Neurosurgery 2004;55:222-6. 48. Talairach J, He´caen M, David M, Monnier M, Ajuriaguerra J. Recherches sur la coagulation therapeutique des structures sous-corticales chez l’homme. Rev Neurol 1949;81:4-24. 49. Guiot G, Brion S, Fardeau M, Bettaier A, Molina P. Dyskinesie volitionelle d’attitude suppremee par la coagulation thalamo-capsulaire. Rev Neurol 1960; 102:220-9. 50. Riechert T, Wolff M. Die Entwicklung u. klinische Bedeutung der gezielten Hirnoperationen. Med Klin 1951; 46:609-11. 51. Gillingham FJ. Small localized lesions of the internal capsule in the treatment of the dyskinesias. Conf Neurol 1962;22:385-92. 52. Svennilson E, Torvik A, Lowe R, Leksell L. Treatment of parkinsonism by stereotactic thermolesions in the pallidal region. A clinical evaluation of 81 cases. Acta Psychiatr Neurol Scand 1960;35:358-77. 53. Rossi GF, Colicchio G, Gentilomo A, Scerrati M. Discussion on the causes of failure of surgical treatment of partial epilepsies. Appl Neurophysiol 1978;41:29-37. 54. Bertrand G, Jasper H. Microelectrode recording of unit activity in the human thalamus. Confin Neurol 1965;26:205-8. 55. Velasco Suarez MM, Escobedo FR. Stereotaxic intracerebral instillation of dopa. Confin Neurol 1970; 32:149-57.
History of stereotactic surgery
56. Bechtereva NP, Bondartchuk AN, Smirnov VM, Meliutcheva LA, Shandurina AN. Method of electrostimulation of the deep brain structures in treatment of some chronic diseases. Confin Neurol 1975;37:136-40. 57. Kandel EI. Experience with the cryosurgical method in production of lesions of the extrapyramidal system. Confin Neurol 1965;26:306-9. 58. Bullard DE, Nashold BSJ. Evolution of principles of stereotactic neurosurgery. Neurosurg Clin N Am 1995;6:27-41. 59. Gildenberg PL. Whatever happened to stereotactic surgery? Neurosurgery 1987;20:983-7. 60. Spiegel EA. Methodological problems in stereoencephalotomy. Confin Neurol 1965;26:125-32. 61. Spiegel EA. History of human stereotaxy (stereoencephalotomy). In: Schaltenbrand G, Walker AE, editors. Stereotaxy of the human brain. Anatomical, physiological and clinical applications. Stuttgart/New York: Georg Thieme Verlag; 1982. p. 3-10. 62. Gildenberg PL. Survey of stereotactic and functional neurosurgery in the United States and Canada. Appl Neurophysiol 1975;38:31-7. 63. Gildenberg PL, Franklin P. Survey of CT-guided stereotactic surgery. Appl Neurophysiol 1985;48:477-80. 64. Bucy PC, Buchanan DN. Athetosis. Brain 1932;55:479-92. 65. Bucy PC. Cortical extirpation in the treatment of involuntary movements. In: Putnam TJ, editor. The diseases of the basal ganglia. New York: Hafner Publishing Co; 1966. p. 551-95. 66. Bucy PC, Case TJ. Tremor. Physiologic mechanism and abolition by surgical means. Arch Neurol Psychiat 1939;41:721-46. 67. Putnam TJ. Treatment of athetosis and dystonia by section of the extrapyramidal motor tracts. Arch Neurol Psychiatry 1933;29:504-21. 68. Walker AE. Cerebral pedunculotomy for the relief of involuntary movements. I. Hemiballismus. Acta Psychiatr Neurol Scand 1949;24:712-29. 69. Walker AE. Cerebral pedunculotomy for the relief of involuntary movements. Parkinsonian tremor. J Nerv Ment Dis 1952;116:766-75. 70. Guiot G, Pecker J. Tractotomie mesencephalique anterieure pour tremblement parkinsonien. Rev Neurol Suppl 1949;1:387. 71. Dandy WE. Changes in our conceptions of localization of certain functions in the brain. Am J Physiol 1930;93:643-7. 72. Gildenberg PL. Evolution of basal ganglia surgery for movement disorders. Stereotact Funct Neurosurg 2006;84:131-5. 73. Meyers R. Dandy’s striatal theory of the ‘‘center of consciousness’’; surgical evidence and logical analysis indicating its improbability. Arch Neurol Psychiat 1951;65:659-71. 74. Spiegel EA, Wycis HT, Baird HW, III. Long range effects of electropallido-ansotomy in extrapyramidal and convulsive disorders. Neurology 1958;8:734-40.
1
75. Mundinger F, Riechert T. Die stereotaktischen Hirnoperationen zur Behandlung extrapyramidaler Bewegungssto¨rungen (Parkinsonismus und Hyperkinesen) und ihre Resultate. Postoperative und Langzeitergebnisse der stereotaktischen Hirnoperationen bei extrapyramidalmotorischen Bewedunasstro¨rungen. Teil B. Fortschr Neurol Psychiatr 1963;31:69-120. 76. Hassler R, Riechert T. Indikationen und Lokalisationsmethode der gezielten Hirnoperationen. Nervenarzt 1954;25:441-7. 77. Spiegel EA, Wycis HT. Ansotomy in paralysis agitans; demonstration of results by motion pictures and electromyograms. Trans Am Neurol Assoc 1953;3:178-80. 78. Krayenbuhl H, Yasargil MG. Relief of intention tremor due to multiple sclerosis by stereotaxic thalamotomy. Conf Neurol 1962;22:368-74. 79. Obrador S, Dierssen G. Observations on the treatment of intentional and postural tremor by subcortical stereotaxic lesions. Confin Neurol 1965;26:250-3. 80. Tsubokawa T, Moriyasu N. Lateral pallidotomy for relief of ballistic movement – its basic evidences and clinical application. Confin Neurol 1975;37:10-15. 81. Andy OJ. Diencephalic coagulation in the treatment of hemiballism. Conf Neurol 1962;22:346-50. 82. Meyers R, Fry WJ, Fry FJ, Dreyer LL, Schultz DF, Noyes RF. Early experiences with ultrasonic irradiation of the pallidofugal and nigral complexes in hyperkinetic and hypertonic disorders. J Neurosurg 1959;16:32-54. 83. Obrador S. A simplified neurosurgical technique for approaching and damaging the region of the globus pallidus in Parkinson’s disease. J Neurol Neurosurg Psychiatr 1957;20:47-9. 84. Van Buren JM, Li CL, Shapiro DY, Henderson WG, Sadowsky DA. A qualitative and quantitative evaluation of parkinsonians three to six years following thalamotomy. Confin Neurol 1973;35:202-35. 85. Yoshida M, Okada K, Watanabe M, Kuramoto S. Analysis of tremulous movements after thalamotomy correlated to intrathalamic therapeutic lesions. Appl Neurophysiol 1976;39:311-4. 86. Krayenbuhl H, Siegfried J. Dentatotomies or thalamotomies in the treatment of hyperkinesia. Confin Neurol 1972;34:29-33. 87. Riechert T. The stereotactic technique and its application in extrapyramidal hyperkinesia. Confin Neurol 1972;34:325-30. 88. Hassler R. Architectonic organization of the thalamic nuclei. In: Schaltenbrand G, Walker AE, editors. Stereotaxy of the human brain. Stuttgart/New York: Georg Thieme Verlag; 1982. p. 140 -80. 89. Cooper IS. Ligation of the anterior choroidal artery for involuntary movements of parkinsonism. Psychiat Quart 1953;27:317-9. 90. Cooper IS. Chemopallidectomy. Science 1955;121:217. 91. Cooper IS. Clinical results and follow-up studies in a personal series of 300 operations for parkinsonism. J Am Geriat Soc 1956;4:1171-81.
27
28
1
History of stereotactic surgery
92. Cooper IS, Bravo GJ. Implications of a five year study of 700 basal ganglia operations. Neurology 1958;8:701-17. 93. Cooper IS. Results of 1000 consecutive basal ganglia operations for parkinsonism. Ann Intern Med 1960;52:483-99. 94. Laitinen LV, Bergenheim AT, Hariz MI. Leksell’s posteroventral pallidotomy in the treatment of Parkinson’s disease. J Neurosurg 1992;76:53-61. 95. Levy A. Die Pallidotomie beim Parkinson syndrom. Eine vergleichende anatomoradiologische Studie. Arch Psychiatr Nervenkr 1959;199:487-507. 96. Hassler R, Mundinger F, Riechert T. The future of therapy in parkinsonism. In: Hassler R, Mundinger F, Riechert T, editors. Stereotaxis in Parkinson syndrome. New York: Springer; 1979. p. 236-48. 97. Kelly PJ, Ahlskog JE, Goerss SJ, Daube JR, Duffy JR, Kall BA. Computer-assisted stereotactic ventralis lateralis thalamotomy with microelectrode recording control in patients with Parkinson’s disease. Mayo Clin Proc 1987;62:655-64. 98. Riechert T. Long term follow-up of results of stereotaxic treatment of extrapyramidal disorders. Conf Neurol 1962;22:356-63. 99. Tasker RR, Siqueira J, Hawrylshyn P, Organ LW. What happened to Vim thalamotomy for Parkinson’s disease? Appl Neurophysiol 1983;46:68-83. 100. Levy A. Stereotaxic brain operations in Parkinson’s syndrome and related motor disturbances. Comparison of lesions in the pallidum and thalamus with those in the internal capsule. Confin Neurol 1967;291-70. 101. Spiegel EA. Significance of Forel’s field for neurosurgery and neurophysiology. Acta Neurochir (Wien) 1965;13: 292-304. 102. Spiegel EA, Wycis HT, Szekely EG, Baird HW, III, Adams J, Flanagan M. Campotomy. Trans Am Neurol Assoc 1962;87:240-2. 103. Spiegel EA, Wycis HT, Szekely EG, Soloff L, Adams J, Gildenberg P, Zanes C. Stimulation of Forel’s field during stereotaxic operations in the human brain. Electroencephalogr Clin Neurophysiol 1964;16:537-48. 104. Albe-Fessard D, Arfel G, Guiot G, Hardy J, Hertzog E, Aleonard P. Identification et de´limitation pre´cise de certaines structures souscorticales de l’homme par l’electrophysiologie. CR Acad Sci (Paris) 1961;243:2412-4. 105. Narabayashi H, Gildenberg PL, Franklin PO (eds.). Proceedings of the meeting on the use of microphysiological recordings during stereotactic neurosurgery. Stereotac Funct Neurosurg 1989;52:77-261. 106. Narabayashi H, Ohye C. Importance of microstereoencephalotomy for tremor alleviation. Appl Neurophysiol 1980;43:222-7. 107. Ohye C, Narabayashi H. Physiological study of presumed ventralis intermedius neurons in the human thalamus. J Neurosurg 1979;50:290-7. 108. Kelly PJ. Microelectrode recording for the somatotopic placement of stereotactic thalamic lesions in the
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
121.
122.
treatment of parkinsonian and cerebellar intention tremor. Appl Neurophysiol 1980;43:262-6. Tasker RR, Lenz F, Yamashiro K, Gorecki J, Hirayama T, Dostrovsky JO. Microelectrode techniques in localization of stereotactic targets. Neurol Res 1987;9:105-12. Falkenberg JH, McNames J, Favre J, Burchiel KJ. Automatic analysis and visualization of microelectrode recording trajectories to the subthalamic nucleus: preliminary results. Stereotact Funct Neurosurg 2006;84:35-44. Starr PA, Christine CW, Theodosopoulos PV, Lindsey N, Byrd D, Mosley A, Marks WJ, Jr. Implantation of deep brain stimulators into the subthalamic nucleus: technical approach and magnetic resonance imagingverified lead locations. J Neurosurg 2002;97:370-87. Ondo WG, Krauss JK. Surgical therapies for dystonia. In: Brin MF, Comella CL, Jankovic J, editors. Dystonia: etiology, clinical features and treatment. Philadelphia, PA: Lippincott Williams & Wilkins; 2004. p. 125-47. Loher TJ, Pohle T, Krauss JK. Functional stereotactic surgery for treatment of cervical dystonia: review of the experience from the lesional era. Stereotact Funct Neurosurg 2004;82:1-13. Hassler R, Dieckmann G. Stereotactic treatment of different kinds of spasmodic torticollis. Confin Neurol 1970;32:135-43. Alker G, Kelly PJ. An overview of CT based stereotactic systems for the localization of intracranial lesions. Comput Radiol 1984;8:193-6. Sano K, Sekino H, Tsukamoto Y, Yoshimasu N, Ishijima B. Stimulation and destruction of the region of the interstitial nucleus in cases of torticollis and see-saw nystagmus. Confin Neurol 1972;34:331-8. Crossman AR. A hypothesis on the pathophysiological mechanisms that underlie levodopa- or dopamine agonist-induced dyskinesia in Parkinson’s disease: implications for future strategies in treatment. Mov Disord 1990;5:100-8. Narabayashi H, Yokochi F, Nakajima Y. Levodopa-induced dyskinesia and thalamotomy. J Neurol Neurosurg Psychiatr 1984;47:831-9. Backlund EO, Granberg PO, Hamberger B, Knutsson E, Martensson A, Sedvall G, Seiger A˚. Transplantation of adrenal medullary tissue to striatum in parkinsonism. First clinical trials. J Neurosurg 1985;62:169-73. Drucker-Colin R, Madrazo I, Diaz V. Open microsurgical autograft of adrenal medulla to caudate nucleus of patients with Parkinson’s disease. Schmitt Neurological Sciences Symposium, Rochester, NY, June 30–July 3, 1987. Madrazo I, Drucker-Colin R, Diaz V. Open microsurgical autograft of adrenal medulla to the right caudate nucleus in two patients with intractable Parkinson’s disease. N Engl J Med 1987;316:831-4. Bakay RA, Herring CJ. Central nervous system grafting in the treatment of parkinsonism. Stereotact Funct Neurosurg 1989;53:1-20.
History of stereotactic surgery
123. Goetz CG, De Long MR, Penn RD, Bakay RA. Neurosurgical horizons in Parkinson’s disease. Neurology 1993;43:1-7. 124. Hitchcock ER, Clough CG, Hughes RC, Kenny BG. Transplantation in Parkinson’s disease: stereotactic implantation of adrenal medulla and foetal mesencephalon. Acta Neurochir Suppl (Wien) 1989;46:48-50. 125. Stoddard SL, Ahlskog JE, Kelly PJ, Tyce GM, van Heerden JA, Zinsmeister AR, Carmichael SW. Decreased adrenal medullary catecholamines in adrenal transplanted parkinsonian patients compared to nephrectomy patients. Exp Neurol 1989;104:218-22. 126. Jankovic J, Grossman R, Goodman C, Pirozzolo F, Schneider L, Zhu Z, Scardino P, Garber AJ, Jhingran SG, Martin S. Clinical, biochemical, and neuropathologic findings following transplantation of adrenal medulla to the caudate nucleus for treatment of Parkinson’s disease. Neurology 1989;39:1227-34. 127. Gildenberg PL, Pettigrew LC, Merrell R, Butler I, Conklin R, Katz J, DeFrance J. Transplantation of adrenal medullary tissue to caudate nucleus using stereotactic techniques. Stereotact Funct Neurosurg 1990;54:268-71. 128. Mark VH, Gildenberg PL, Franklin PO (eds.). Proceedings of the colloquium on the use of embryonic cell transplantation for correction of CNS disorders. Appl Neurophysiol 1984;47:1-76. 129. Freed WJ, Cannon-Spoor HE, Krauthamer E. Intrastriatal adrenal medulla grafts in rats. Long-term survival and behavioral effects J Neurosurg 1986;65(5):664-70. 130. Bakay RA, Fiandaca MS, Barrow DL, Schiff A, Collins DC. Preliminary report on the use of fetal tissue transplantation to correct MPTP-induced Parkinson-like syndrome in primates. Appl Neurophysiol 1985;48:358-61. 131. Redmond DE, Jr, Roth RH, Spencer DD, Naftolin F, Leranth C, Robbins RJ, Marek KL, Elsworth JD, Taylor JR, Sass KJ. Neural transplantation for neurodegenerative diseases: past, present, and future. Ann NY Acad Sci 1993;695:258-66. 132. Lindvall O, Rehncrona S, Brundin P, Gustavii B, Astedt B, Widner H, Lindholm T, Bjorklund A, Tan M. Human fetal dopamine neurons grafted into the striatum in two patients with severe Parkinson’s disease. A detailed account of methodology and a 6-month followup. Arch Neurol 1989;46:615-31. 133. Bakay RA. Neurotransplantation: a clinical update. Acta Neurochir Suppl (Wien) 1993;58:8-16. 134. Breggin PR. The return of lobotomy and psychosurgery. Congressional Record 1972;118:24-6. 135. Spiegel EA, Wycis HT. Mesencephalotomy in intractable facial pain. Trans Am Neurol Assoc 1952;56:186-9. 136. Gybels JM, Sweet WH. Neurosurgical treatment of persistent pain. Pain and headache, vol. 11. Basel: Karger; 1989. 137. Gildenberg PL. History of pain management. In: Greenblatt SH, Dagi TF, editors. A history of neurosurgery. Park Ridge, IL: American Association of Neurological Surgeons; 1997. p. 465-88.
1
138. Trousseau A. De la ne´vralgie ´epileptiforme. Arch Gen Med 1853;1:33-44. 139. Rose W. Removal of the gasserian ganglion for severe neuralgia. Trans Med Soc Lond 1890;14:35-40. 140. Horsley V, Taylor J, Colman WS. Remarks on the various surgical procedures devised for the relief or cure of trigeminal neuralgia (tic douloureaux). Brit Med J 1891;2:1139-43. 141. Cushing H. A method for total extirpation of the Gasserian ganglion for trigeminal neuralgia. JAMA 1900;34:1035-41. 142. Cushing H. The major trigeminal neuralgias and their surgical treatment based on experiences with 332 Gasserian operations. I. Varieties of facial neuralgia. Am J Med Sci 1920;160:157-84. 143. Ramonede L. Exerese de trijumeau. Presse Med 1903;11:789-91. 144. Dandy WE. Section of the sensory root of the trigeminal nerve at the pons. Bull Johns Hopkins Hosp 1925;36:105-6. 145. Kirschner M. Zur Electrochirurgie. Arch Klin Chir 1931;161:761-8. 146. Kirschner M. Die Punktionstechnik u. Elektrokoagulation des Ganglion Gasseri. Arch Klin Chir 1933;176:581-620. 147. Murtagh F, Jr, Wycis HT, Spiegel EA. Relief of thalamic pain by mesencephalotomy. AMA Arch Neurol Psychiatry 1951;65:255-7. 148. Frazier CH. Section of the anterolateral columns of the spinal cord for the relief of pain. Arch Neurol Psychiatry 1920;4:137-47. 149. French LA. Cordotomy in the high cervical region for intractable pain. J Lancet 1953;73:283-7. 150. Spiller WG, Martin E. The treatment of persistent pain of organic origin in the lower part of the body by division of the anterolateral column of the spinal cord. J Am Med Assoc 1912;58:1489-90. 151. Gildenberg PL. Mesencephalotomy. In: Burchiel KJ, editor. Surgical management of pain. New York: Thieme; 2002. p. 786-94. 152. Iskandar BJ, Nashold BSJ. History of functional neurosurgery. Neurosurg Clin N Am 1995;6:1-25. 153. White JC, Sweet WH. Pain, its mechanism and neurosurgical control. Springfield, IL: Charles C Thomas; 1955. 154. Spiegel EA, Wycis HT, Szekely EG, Gildenberg PL. Medial and basal thalamotomy in so-called intractable pain. In: Knighton RS, Dumke PR, editors. Pain. Boston, MA: Little Brown; 1966. p. 503-17. 155. He´caen H, Talairach T, David M, Dell MB. Me´moires originaux. Coagulations limite´es du thalamus dans les algies du syndrome thalamique. Rev Neurol 1949;81:917-31. 156. Spiegel EA, Wycis HT. Mesencephalotomy in treatment of intractable facial pain. AMA Arch Neurol Psychiatry 1953;69:1-13.
29
30
1
History of stereotactic surgery
157. Spiegel EA, Wycis HT, Szekely EG, Gildenberg P, Zanes C. Combined dorsomedial, intralaminar and basal thalamotomy for relief of so-called intractable pain. J Int Coll Surg 1964;42:160-8. 158. Nashold BSJ, Wilson WP, Slaughter DG. Stereotaxic midbrain lesions for central dysesthesia and phantom pain. Preliminary report. J Neurosurg 1969; 30:116-26. 159. Nashold BS, Jr., Wilson WP, Slaughter DG. Sensations evoked by stimulation in the midbrain of man. J Neurosurg 1969;30:14-24. 160. Nashold BS, Jr. Extensive cephalic and oral pain relieved by midbrain tractotomy. Conf Neurol 1972;34:382-8. 161. Foltz EL, White LE, Jr. Pain ‘‘relief ’’ by frontal cingulumotomy. J Neurosurg 1962;19:89-100. 162. Mullan S, Moseley R, Harper PV. The creation of deep cerebral lesions by small beta ray sources implanted under guidance of fluoroscopic image intensifiers (as used in the treatment of Parkinson’s disease). Am J Roentgen 1959;82:613. 163. Mullan S, Harper PV, Hekmatpanah J, Torres H, Dobben G. Percutaneous interruption of spinal pain tracts by means of a strontium-90 needle. J Neurosurg 1963;20:931-9. 164. Rosomoff HL, Carrol F, Brown J, Sheptak P. Percutaneous radiofrequency cervical cordotomy: technique. J Neurosurg 1965;23:639-44. 165. Tenicela R, Rosomoff HL, Feist J, Safar P. Pulmonary function following percutaneous cervical cordotomy. Anesthesiology 1968;29:7-16. 166. Lin PM, Gildenberg PL, Polakoff PP. An anterior approach to percutaneous lower cervical cordotomy. J Neurosurg 1966;25:553-60. 167. Sjo¨qvist O. Eine neue Operationsmethode bei Trigeminusneuralgie: durchschneidung des Tractus spinalis trigemini. Zentralbl f Neurochir 1937;2:274-81. 168. White JC, Sweet WH. Pain and the neurosurgeon. A forty year experience. Springfield, IL: Charles C Thomas; 1969. 169. Hitchcock E. Stereotactic trigeminal tractotomy. Farmatsiia 1970;19:131-5. 170. Hitchcock E. Stereotactic cervical myelotomy. J Neurol Neurosurg Psychiatry 1970;33:224-30. 171. Crue BL, Todd EM, Carregal EJA. Posterior approach for high cervical percutaneous radiofrequency cordotomy. Confinia Neurol 1968;30:41-52. 172. Gildenberg PL. Evolution of spinal cord surgery for pain. Clin Neurosurg 2006;53:11-17. 173. Hitchcock E. Stereotactic myelotomy. Proc R Soc Med 1974;67:771-2. 174. Schvarcz JR. Stereotactic extralemniscal myelotomy. J Neurol Neurosurg Psychiatry 1976;39:53-7. 175. Gildenberg PL, Hirshberg RM. Limited myelotomy for the treatment of intractable cancer pain. J Neurol Neurosurg Psychiatry 1984;47:94-6.
176. Stillings D. The first use of electricity for pain treatment. Medtronic Archive on Electro-Stimulation 1971. 177. Isaacson W. Benjamin Franklin. An American life. New York: Simon & Schuster; 2003. 178. Pruel MC. A history of neuroscience from Galen to Gall. In: Greenblatt SH, Dagi TF, Epstein MH, editors. A history of neurosurgery. Park Ridge, IL: American Association of Neurological Surgeons; 1997. p. 99-130. 179. Shelly M. Frankenstein. Oxford: Oxford Press; 1818. ¨ ber die elektrische Erregbarkeit 180. Fritsch G, Hitzig E. U des Grosshirns. Arch Anat Physiol wiss Med 1870;37:300-32. 181. Morgan JP. The first reported case of electrical stimulation of the human brain. J Hist Med Allied Sci 1982;37:51-64. 182. Vilensky JA, Gilman S. Horsley was the first to use electrical stimulation of the human cerebral cortex intraoperatively. Surg Neurol 2002;58:425-6. 183. Hassler R, Riechert T, Mundinger F, Umbach W, Ganglberger JA. Physiological observations in stereotaxic operations in extrapyramidal motor disturbances. Brain 1960;83:337-50. 184. Tasker RR, Organ LW, Hawrylshyn PA. The thalamus and midbrain of man. A physiological atlas using electrical stimulation. Springfield, IL: Charles C Thomas; 1982. 185. Woolsey CN. Cortical localization as defined by evoked potential and electrical stimulation studies. In: Schaltenbrand G, Woolsey CN, editors. Cerebral localization and organization. Madison, WI: University of Wisconsin Press; 1964. p. 17-26. 186. Pool JL. Psychosurgery in older people. J Am Geriatr Soc 1954;2:456-65. 187. Heath RG. Exploring the mind-brain relationship. Baton Rouge, LA: Moran Printing, Inc.; 1996. 188. Olds J, Milner P. Positive reinforcement produced by electrical stimulation of the septal area and other regions of the rat brain. J Comp Physiol Psychol 1954;47:419-27. 189. Heath RG. Psychiatry. Annu Rev Med 1954;5:223-36. 190. Gol A. Relief of pain by electrical stimulation of the septal area. J Neurol Sci 1967;5:115-20. 191. Bechtereva NP, Bondarchuk AN, Smirnov VM. Therapeutic electrostimulations of the deep brain structures. Vopr Neirokhir 1972;1:7-12. 192. Melzack R, Wall PD. Pain mechanisms: a new theory. Science 1965;150:971-9. 193. Laitinen L. Placement of electrodes in transcutaneous stimulation for chronic pain. Neurochirurgie 1976; 22:517-26. 194. Wall PD, Sweet WH. Temporary abolition of pain in man. Science 1967;155:108-9. 195. Sweet WH, Wepsic JG. Treatment of chronic pain by stimulation of fibers of primary afferent neuron. Trans Am Neurol Assoc 1968;93:103-7.
History of stereotactic surgery
196. Shealy CN, Mortimer JT, Reswick JB. Electrical inhibition of pain by stimulation of the dorsal columns. Preliminary clinical report. Anesth Analg (Cleve) 1967;46:489-91. 197. Gildenberg PL. Treatment of spasmodic torticollis with dorsal column stimulation. Acta Neurochir (Wien) 1977;24:Suppl65-6. 198. Cook AW. Electrical stimulation in multiple sclerosis. Hosp Pract 1976;11:51-8. 199. Dooley DM. Spinal cord stimulation. AORN J 1976;23:1209-12. 200. Dooley DM, Kasprak M. Modification of blood flow to the extremities by electrical stimulation of the nervous system. South Med J 1976;69:1309-11. 201. Hosobuchi Y, Adams JE, Rutkins B. Chronic thalamic stimulation for the control of facial anesthesia dolorosas. Arch Neurol 1973;29:158-61. 202. Mundinger F. New stereotactic treatment of spasmodic torticollis with a brain stimulation system (author’s transl). Med Klin 1977;72:1982-6. 203. Reynolds DV. Surgery in the rat during electrical analgesia induced by focal brain stimulation. Science 1969;164:444-5. 204. Richardson DE, Akil H. Pain reduction by electrical brain stimulation in man. Part 1: Acute administration in periaqueductal and periventricular sites. J Neurosurg 1977;47:178-83. 205. Gildenberg PL. Symposium on the safety and clinical efficacy of implanted neuroaugmentive devices. Appl Neurophysiol 1977;40:69-240. 206. Tsubokawa T, Katayama Y, Yamamoto T, Hirayama T, Koyama S. Chronic motor cortex stimulation for the treatment of central pain. Acta Neurochir Suppl (Wien) 1991;52:137-9. 207. Migita K, Uozumi T, Arita K, Monden S. Transcranial magnetic coil stimulation of motor cortex in patients with central pain. Neurosurgery 1995;36:1037-9. 208. Gildenberg PL. History of neuroaugmentative procedures for chronic pain. Neurosurg Clin N Am 2003;14:327-37. 209. Gildenberg PL. History of electrical neuromodulation for chronic pain. Pain Med 2006;7Suppl 1:S7-13. 210. Gildenberg PL. The use of pacemakers (electrical stimulation) in functional neurological disorders. In: Rasmussen T, Marino R, editors. Functional neurosurgery. New York: Raven Press; 1979. p. 59-74. 211. Laitinen LV, Bergenheim AT, Hariz MI. Ventroposterolateral pallidotomy can abolish all parkinsonian symptoms. Stereotact Funct Neurosurg 1992;58:14-21. 212. Spiegel EA, Wycis HT. Ansotomy in paralysis agitans. AMA Arch Neurol Psychiatry 1953;69:652-3. 213. Goetz CG, Diederich NJ. There is a renaissance of interest in pallidotomy for Parkinson’s disease. Nat Med 1996;2:510-4. 214. Iacono RP, Shima F, Lonser RR, Kuniyoshi S, Maeda G, Yamada S. The results, indications, and physiology of
215.
216.
217.
218.
219.
220.
221.
222.
223.
224.
225.
226.
1
posteroventral pallidotomy for patients with Parkinson’s disease. Neurosurgery 1995;36:1118-27. Favre J, Taha JM, Nguyen TT, Gildenberg PL, Burchiel KJ. Pallidotomy: a survey of current practice in North America. Neurosurgery 1996;39:883-92. Lozano A, Hutchison W, Kiss Z, Tasker R, Davis K, Dostrovsky J. Methods for microelectrode-guided posteroventral pallidotomy. J Neurosurg 1996;84: 194-202. Krauss JK, Grossman RG. Historical review of pallidal surgery for treatment of parkinsonism and other movement disorders. In: Krauss JK, Grossman RG, Jankovic J, editors. Pallidal surgery for the treatment of Parkinson’s disease and movement disorders. Philadelphia, PA: Lippincott-Raven; 1998. p. 1-23. Vitek JL, Bakay RA. The role of pallidotomy in Parkinson’s disease and dystonia. Curr Opin Neurol 1997;10:332-9. Nandi D, Chir M, Liu X, Bain P, Parkin S, Joint C, Winter J, Stein J, Scott R, Gregory R, Aziz T. Electrophysiological confirmation of the zona incerta as a target for surgical treatment of disabling involuntary arm movements in multiple sclerosis: use of local field potentials. J Clin Neurosci 2002;9:64-8. Meier MJ, Story J, French LA, Chou SN. Quantitative assessment of behavioral changes following subthalamotomy in the treatment of Parkinson’s disease. Confin Neurol 1966;27:154-61. Brice J, McLellan L. Suppression of intention tremor by contingent deep-brain stimulation. Lancet 1980; 1:1221-2. Benabid AL, Pollak P, Gervason C, Hoffmann D, Gao DM, Hommel M, Perret JE, de Rougemont J. Longterm suppression of tremor by chronic stimulation of the ventral intermediate thalamic nucleus. Lancet 1991;337:403-6. Siegfried J, Lippitz B. Chronic electrical stimulation of the VL-VPL complex and of the pallidum in the treatment of movement disorders: personal experience since 1982. Stereotact Funct Neurosurg 1994; 62:71-5. Benabid AL, Pollak P, Gross C, Hoffmann D, Benazzouz A, Gao DM, Laurent A, Gentil M, Perret J. Acute and long-term effects of subthalamic nucleus stimulation in Parkinson’s disease. Stereotact Funct Neurosurg 1994;62:76-84. Alvarez L, Macias R, Guridi J, Lopez G, Alvarez E, Maragoto C, Teijeiro J, Torres A, Pavon N, Rodriguez-Oroz MC, Ochoa L, Hetherington H, Juncos J, Delong MR, Obeso JA. Dorsal subthalamotomy for Parkinson’s disease. Mov Disord 2001;16:72-8. Jimenez F, Velasco F, Velasco M, Brito F, Morel C, Marquez I, Perez ML. Subthalamic prelemniscal radiation stimulation for the treatment of Parkinson’s disease: electrophysiological characterization of the area. Arch Med Res 2000;31:270-81.
31
32
1
History of stereotactic surgery
227. Spiegel EA, Wycis HT, Szekely EG, Adams DJ, Flanagan M, Baird HW, III. Campotomy in various extrapyramidal disorders. J Neurosurg 1963;20:871-84. 228. Gildenberg PL. Evolution of neuromodulation. Stereotact Funct Neurosurg 2005;83:71-9. 229. Krauss JK, Pohle T, Weber S, Ozdoba C, Burgunder JM. Bilateral stimulation of globus pallidus internus for treatment of cervical dystonia. Lancet 1999; 354:837-8. 230. Coubes P, Roubertie A, Vayssiere N, Hemm S, Echenne B. Treatment of DYT1-generalised dystonia by stimulation of the internal globus pallidus. Lancet 2000;355:2220-1. 231. Kumar R, Dagher A, Hutchison WD, Lang AE, Lozano AM. Globus pallidus deep brain stimulation for generalized dystonia: clinical and PET investigation. Neurology 1999;53:871-4. ¨ ber einen Fall von doppelseitiger 232. Hassler R, Riechert T. U Fornicotomie bei sogenannter temporaler Epilepsie. Acta Neurochir (Wien) 1957;5:330-40. 233. Pagni CA. Stereo-electroencephalographic observations on the psychomotor seizure. Confin Neurol 1966; 27:137-43. 234. Umbach W, Riechert T. The stereotactic coagulation of the fornix as treatment of temporal lobe epilepsy. Proc First Int Congress Neurol Sci 1961;3:565-78. 235. Jinnai D, Nishimoto A. Stereotaxic destruction of Forel H for treatment of epilepsy. Neurochirurgia(Stuttgart) 1963;6:164-76. 236. Laitinen L. Thalmotomy in progressive myoclonus epilepsy. Acta Neurol Scand 1967;43:31-47. 237. Umbach W. Long-term results of fornicotomy for temporal epilepsy. Confin Neurol 1966;27:121-3. 238. Munari C, Hoffmann D, Francione S, Kahane P, Tassi L, Lo Russo G, Benabid AL. Stereo-electroencephalography methodology: advantages and limits. Acta Neurol Scand Suppl 1968;152:56-67. 239. Mundinger F, Becker P, Groebner E, Bachschmid G. Late results of stereotactic surgery of epilepsy predominantly temporal lobetype. Acta Neurochir (Wien) 1976;23Suppl:177-82. 240. Nashold BS, Jr, Flanigin H, Jr, Wilson WP, Stewart B. Stereotactic evaluation of bitemporal epilepsy with electrodes and lesions. Confin Neurol 1973; 35:94-100. 241. Vaernet K. Stereotaxic amygdalotomy in temporal lobe epilepsy. Confin Neurol 1972;34:176-80. 242. Olivier A, Gloor P, Andermann F, Quesney LF. The place of stereotactic depth electrode recording in epilepsy. Appl Neurophysiol 1985;48:395-9. 243. Gildenberg PL. Unpublished data. 1990. 244. Fins JJ. From psychosurgery to neuromodulation and palliation: history’s lessons for the ethical conduct and regulation of neuropsychiatric research. Neurosurg Clin N Am 2003;14:303-6. 245. Spiegel EA, Wycis HT. The central mechanism of emotions. Am J Psychiatry 1951;109:426-31.
246. Orchinik C, Koch R, Wycis HT, Freed H, Spiegel EA. The effect of thalamic lesions upon the emotional reactivity (Rorschach and behavior studies). Res Publ Assoc Res Nerv Ment Dis 1949;29:172-207. 247. Sano K. Sedative neurosurgery with special reference to posteromedial hypothalamotomy. Neurol Med Chir (Tokyo) 1962;4:112. 248. Spiegel EA, Wycis HT, Freed H, Orchinik C. The central mechanism of the emotions; (experiences with circumscribed thalamic lesions). Am J Psychiatry 1951;108:426-32. 249. Gildenberg PL (ed.). Functional neurosurgery. Neurosurg Clin N Am 1995;6:1-181. 250. Fodstad H, Strandman E, Karlsson B, West KA. Treatment of chronic obsessive compulsive states with stereotactic anterior capsulotomy or cingulotomy. Acta Neurochir (Wien) 1982;62:1-23. 251. Martuza RL, Chiocca EA, Jenike MA, Giriunas IE, Ballantine HT. Stereotactic radiofrequency thermal cingulotomy for obsessive compulsive disorder. J Neuropsychiatry Clin Neurosci 1990;2:331-6. 252. Andy OJ, Jurko F. Thalamic stimulation effects on reactive depression. Appl Neurophysiol 1987;50:324-9. 253. Diering SL, Bell WO. Functional neurosurgery for psychiatric disorders: a historical perspective. Stereotact Funct Neurosurg 1991;57:175-94. 254. Benabid AL, Wallace B, Mitrofanis J, Xia R, Piallat B, Chabardes S, Berger F. A putative generalized model of the effects and mechanism of action of high frequency electrical stimulation of the central nervous system. Acta Neurol Belg 2005;105:149-57. 255. Kopell BH, Greenberg B, Rezai AR. Deep brain stimulation for psychiatric disorders. J Clin Neurophysiol 2004;21:51-67. 256. Lienhard JH. Inventing modern. Oxford Press; Oxford: 257. Murtagh F, Wycis HT, Robbins R, Spiegel-Adolph M, Spiegel EA. Visualization and treatment of cystic brain tumors by stereoencephalotomy. Acta Radiol 1956;46:407-14. 258. Gildenberg PL, Kaufman HH. Direct calculation of stereotactic coordinates from CT scans. Appl Neurophysiol 1982;45:347-51. 259. Gildenberg PL, Kaufman HH, Murthy KS. Calculation of stereotactic coordinates from the computed tomographic scan. Neurosurgery 1982;10:580-6. 260. Gildenberg PL. Computerized tomography and stereotactic surgery. In: Spiegel EA, editor. Guided brain operations. Basel: Karger; 1982. p. 24-34. 261. Leksell L, Jernberg B. Stereotaxis and tomography: a technical note with six figures. Acta Neurochir 1980;52:1-7. 262. Barcia-Salorio JL, Broseta J, Hernandez G, Roldan P, Bordes V. A new approach for directed CT localization in stereotaxis. Appl Neurophysiol 1982;45:383-6. 263. Colombo F, Angrilli F, Zanardo A, Pinna V, Alexandre A, Benedetti A. A universal method to employ
History of stereotactic surgery
264.
265. 266.
267. 268.
269.
270.
271.
272.
273.
274.
275. 276. 277.
278.
279.
280.
CT scanner spatial information in stereotactic surgery. Appl Neurophysiol 1982;45:352-65. Colombo F, Angrilli F, Zanardo A, Pinna V, Benedetti A. A new method for utilizing CT data in stereotactic surgery: measurement and transformation technique. Acta Neurochir (Wien) 1981;57: 195-203. Patil AA. Compute tomography-oriented stereotactic system. Neurosurgery 1982;10:370-4. Koslow M, Abele MG. A fully interfaced computerized tomographic-stereotactic surgical system. Appl Neurophysiol 1980;43:174-5. Lunsford LD. A dedicated CT system for the stereotactic operating room. Appl Neurophysiol 1982;45:374-8. Brown RA, Roberts TS, Osborn AG. Stereotaxic frame and computer software for CT-directed neurosurgical localization. Invest Radiol 1980;15:308-12. Kelly PJ, Kall B, Goerss S, Alker GJ, Jr. Precision resection of intra-axial CNS lesions by CT-based stereotactic craniotomy and computer monitored CO2 laser. Acta Neurochir (Wien) 1983;68:1-9. Leksell L, Leksell D, Schwebel J. Stereotaxis and nuclear magnetic resonance. J Neurol Neurosurg Psychiatry 1985;48:14-18. Broggi G, Franzini A, Peluchetti D, Servello D. Treatment of deep brain abscesses by stereotactic implantation of an intracavitary device for evacuation and local application of antibiotics. Acta Neurochir (Wien) 1985;76:94-8. Backlund EO, von Holst H. Controlled subtotal evacuation of intracerebral haematomas by stereotactic technique. Surg Neurol 1978;9:99-101. Tanikawa T, Amano K, Kawamura H, Kawabatake H, Notani M, Iseki H, Shiwaku T, Nagao T, Iwata Y, Taira T. CT-guided stereotactic surgery for evacuation of hypertensive intracerebral hematoma. Appl Neurophysiol 1985;48:431-9. Zamorano L, Chavantes C, Dujovny M, Malik G, Ausman J. Stereotactic endoscopic interventions in cystic and intraventricular brain lesions. Acta Neurochir Suppl (Wien) 1992;54:69-76. Gildenberg PL, Labuz J. Use of a volumetric target for image-guided surgery. Neurosurgery 2006;59:651-9. Bernstein M, Gutin PH. Interstitial irradiation of brain tumors: a review. Neurosurgery 1981;9:741-50. Gildenberg PL, Woo S, Butler B, Grant W, Berner B. Fractionated brachytherapy: catheter insertion and dosimetry. Stereotact Funct Neurosurg 1994;63:246-9. Kelly PJ, Alker GJ, Jr. A method for stereotactic laser microsurgery in the treatment of deep seated CNS neoplasms. Appl Neurophysiol 1980;43:210-5. Kall BA, Kelly PJ, Goerss SJ. Interactive stereotactic surgical system for the removal of intracranial tumors utilizing the CO2 laser and CT-derived database. IEEE Trans Biomed Eng 1985;32:112-6. Kelly PJ. Tumor stereotaxis. Philadelphia, PA: Saunders; 1991.
1
281. Kelly PJ, Kall BA, Goerss S. Transposition of volumetric information derived from computed tomography scanning into stereotactic space. Surg Neurol 1984; 21:465-71. 282. Kosugi Y, Watanabe E, Goto J, Watanabe T, Yoshimoto S, Takakura K, Ikebe J. An articulated neurosurgical navigation system using MRI and CT images. IEEE Trans Biomed Eng 1988;35:147-52. 283. Roberts DW, Strohbehn JW, Hatch JF, Murray W, Kettenberger H. A frameless stereotaxic integration of computerized tomographic imaging and the operating microscope. J Neurosurg 1986;65:545-9. 284. Rousu JS, Kohls PE, Kall B, Kelly PJ. Computer-assisted image-guided surgery using the regulus navigator. Stud Health Technol Inform 1998;50:103-9. 285. Heilbrun MP, McDonald P, Wiker C, Koehler S, Peters W. Stereotactic localization and guidance using a machine vision technique. Stereotact Funct Neurosurg 1992;58:94-8. 286. Day R, Heilbrun MP, Koehler S, McDonald P, Peters W, Siemionow V. Three-point transformation for integration of multiple coordinate systems: applications to tumor, functional, and fractionated radiosurgery stereotactic planning. Stereotact Funct Neurosurg 1994;63:76-9. 287. Heilbrun MP, Koehler S, MacDonald P, Siemionow V, Peters W. Preliminary experience using an optimized three-point transformation algorithm for spatial registration of coordinate systems: a method of noninvasive localization using frame-based stereotactic guidance systems. J Neurosurg 1994;81:676-82. 288. Smith KR, Frank KJ, Bucholz RD. The NeuroStation – a highly accurate, minimally invasive solution to frameless stereotactic neurosurgery. Comput Med Imag Graphics 1994;18:247-56. 289. Gildenberg PL. The Exoscope system for three-dimensional craniotomy. In: Gildenberg PL, Tasker RR, editors. Textbook of stereotactic and functional neurosurgery. New York: Mc-Graw-Hill; 1997. p. 485-90. 290. Gildenberg PL, Ledoux R, Cosman E, Labuz J. The Exoscope – a frame-based video/graphics system for intraoperative guidance of surgical resection. Stereotact Funct Neurosurg 1994;63:23-5. 291. Zamorano L, Dujovny M, Chavantes C, Malik G, Ausman J. Image-guided stereotactic centered craniotomy and laser resection of solid intracranial lesions. Stereotact Funct Neurosurg 1990;54/55:398-403. 292. Gildenberg PL, Labuz J. Stereotactic craniotomy with the exoscope. Stereotact Funct Neurosurg 1997;68:64-71. 293. Pillay PK, Gildenberg PL, Labuz J. Computer workstation applications: stereoplan, virtual reality, and the Gildenberg exoscope. In: Pell MF, Thomas DGT, editors. Handbook of stereotaxy using the CRW apparatus. Baltimore, MD: Williams & Wilkins; 1994. p. 219-33. 294. Gildenberg PL. Confinia neurologica/applied neurophysiology. 50 years. Appl Neurophysiol 1988;51:1-5.
33
11 History of Stereotactic Surgery in China F.-C. Lee . B. Sun . J. Zhang . K. Zhang . F.-G. Meng
Prehistoric and Feudal China China is unique among the world civilizations in that it has a well-established indigenous medical treatment philosophy dating back thousands of years. This was epitomized by the well known Huang Ti Nei Jing (Yellow Emperor’s Inner Classic), a seminal medical text of ancient China where a legendary physician Yu Fu was alleged to possess the skill for surgical exposure of the brain. The Neijing, Shennong Ben Cao Jing (Divine Husbandman’s Classic of the Materia Medica), and a few other ancient texts laid the foundation of the so-called Traditional Chinese Medicine (TCM), an alternative treatment being actively practiced even today in mainland China and among overseas Chinese alongside modern Western medical and surgical therapies. A good depiction appeared in the popular History of the Three Kingdoms where ‘‘cranial surgery’’ was proposed by Hua Tuo (Hua Lun), a famous physician, on one of the three reigning kings Cao Cao who suffered persistent headaches, presumably due to battle trauma–related intracranial hematoma (or perhaps even intracranial tumor) during the turbulent warring period of Eastern Han and Three Kingdoms (222–280 A.D.). Hua Tuo (> Figure 11‐1) had also reportedly performed surgeries with anesthesia, 1,600 years ahead of similar endeavors in Western Civilization, using wine and a herbal concoction of cannabis boil powder. Incidentally, Hua Tuo came from the ancient Qiao City of Pei State in the modern day Anhui Province, where China’s first
#
Springer-Verlag Berlin/Heidelberg 2009
stereotactic neurosurgery institute and training center was established in 1983 and the first National Stereotactic and Functional Neurosurgical Conference was held on June 8–14, 1987. The oldest prehistoric evidence of human trephination in European Civilization was probably the Ensisheim Stone Age skull unearthed from a burial site in France which was dated to 5100 B.C. [1]. Archaeological evidence also surfaced in China in 1995 with the excavation in Shandong Province of an adult male skull, aged 35–45 years, bearing a 31 25 mm round parietal calvarial defect with smooth border. The perimeter showed evidence of scrapping and bone regeneration. This belonged to an ancient inhabitant of the Dawen Kou Culture (third to fifth millennium B.C.) and was 14C-dated to about 5000 B.C. (> Figure 11‐2). The opening appeared to have been made with tools resembling a trephine. This lends credence to the conclusion that trephinations were probably performed in Neolithic China. Similar finds were discovered at other archaeological sites in the Qinghai, Heilongjiang, and Henan Provinces of China, variously radiocarbon-dated to between 2000 and 4000 B.C. [2]. The motivation for these skull openings remained varied and speculative, just like in other European archaeological discoveries of the Paleolithic and Neolithic period. Whether for curative or ritual purposes to heal and alter behavior, these could have been early men’s attempt to relate structure to functions and behavior, perhaps a primitive version of modern day functional neurosurgery.
126
11
History of stereotactic surgery in china
Birth of Neurosurgery in Modern China An understanding of the development of stereotactic and functional neurosurgery in China will . Figure 11‐1 Hua Tuo (Hua Lun), born 208 B.C. in Qiao (present Anhui Province, China)
be incomplete without due reference to the history of general neurosurgery, the foundation on which the former has evolved. Modern neurosurgery did not develop much until the last century of Chinese history. Like the rest of the Chinese civilization, it was shrouded in mystery behind the ‘‘bamboo curtain’’ as a result of the cultural and diplomatic isolation of this monolithic, inward-looking nation. The country was plagued by the ineffective, feudalistic Imperial Manchurian rule, subsequently overthrown in 1912 in a popular revolution led by Dr Sun Yat-sen, the father and founder of modern China. Due credit should be given to Dr Song-Tao Guan (ST Kuan) and Dr Cha-Li Zhang (Charles Chang) from Beijing and Shenyang, respectively, for their pioneering works in neurosurgery in China [3]. Guan completed his residency program at the Peking Union Medical College Hospital (PUMC) in 1926. PUMC was jointly founded in 1906 by the American Board of Commissioner for Foreign Missions, the Presbyterian Church in the USA, the Methodist Episcopal Church, and the London Missionary Society, among others. From 1926 to1930 Guan received neurosurgery training at the University of
. Figure 11‐2 Archaeological find at Dawen Kou 5000 B.C. (picture courtesy of Cheng-Yuan Wu)
History of stereotactic surgery in china
Pennsylvania under Dr Charles H. Frazier, who was among the first group of American neurosurgeons who established neurosurgery as a new discipline from general surgery. In 1938 Yi-Cheng Zhao, another graduate from the same medical college, was sent by the Rockefeller Foundation (established 1913), through its subsidiary, the China Medical Board, for formal neurosurgical training under Dr Wilder Penfield at the Montreal Neurological Institute at McGill University. Upon his return to China in 1940, Zhao practiced neurosurgery at the PUMC Hospital with Guan. In 1952, three years after the Chinese civil war ended with the establishment of the People’s Republic of China, Zhao founded the first Brain Department comprising neurology and neurosurgery in the port city of Tianjin, one of the very few cities with early Western influence in the ethnophobic China of the nineteenth century. It was reluctantly opened by the Qing Dynasty of China as a treaty port accessible to France, Britain, and others in 1860. Formal concessions with foreign settlements were ceded to Western powers and Japan in 1903 after the Imperial Qing Dynasty’s defeat in the war during the Boxer Rebellion of 1899. These three neurosurgeons, along with Tong-He Zhang of Xian City, performed the first neurosurgical procedures in modern China. Zhang performed prefrontal lobotomy for a psychiatric disorder [4]. From 1932 to 1949, only 16 neurosurgical articles were published in the indigenous Chinese Medical Journal (CMJ), which was initially established in 1887 in Shanghai as China Medical Missionary Journal for publications mainly by Western medical missionaries serving in China. The name was changed to China Medical Journal in 1907, thereafter adopted its current name of CMJ in 1932, when it merged with the English language section of National Medical Journal of China. Only about 60 brain tumors were reported in the Chinese literature. With his training under Dr Frazier who pioneered subtemporal retrogasserian neurotomy in
11
1910 as a permanent cure for tic douloureux, Guan reported his experience with Frazier’s operation for trigeminal neuralgia in 1932 in the CMJ [5]. Charles Chang published on similar procedure and alcohol block three years after Guan. He also drew attention to the occurrence of anesthesia dolorosa and abducens nerve palsy and reported on neurofibroma of the Gasserian ganglion [6]. However, further progress in the development of neurosurgery was repeatedly halted when the country was plunged into two devastating civil wars between the ruling Nationalist Army (Kuomintang) and the Communist People’s Liberation Army (PLA) from 1927 to 1936 and 1946 to 1949, with an intervening period of brutal invasion of China by the Japanese Imperial Army that lasted from 1937 to 1945 when World War II finally ended (> Figure 11‐3). Before the formation of the communist People’s Republic of China (PRC) in 1949, international assistance in the development of neurosurgery as in many other fields was predominantly rendered by the United States and European nations mainly through the China Medical Board. After 1949, however, international collaborations were confined to fellow Communist Bloc countries, in particular the USSR. A 6 month neurosurgery training course was conducted in Beijing by Dr A.E. Arutiunov from the Kiev Neurosurgical Institute of the former Soviet Union in October 1954. Participants of this program included Chung-Cheng Wang (Zhong-Cheng Wang), Ya-Du Zhao, and Da-Jie Jiang, while others were sent to the prestigious Moscow Burdenko Neurosurgical Institute. Among them, Tong-Jin Tu set up a neurosurgery service for the People’s Liberation Army in the Fourth Military Medical University in Xian City (ancient capital of Qin dynasty 221–206 B.C.) upon his return to China in 1956. The T J Tu Award for excellence in neurosurgery was established in honor of his invaluable contributions to military neurosurgery. DrYi-Cheng Zhao (> Figure 11‐4) founded the first Brain Department in China, comprising
127
128
11
History of stereotactic surgery in china
. Figure 11‐3 Some early Chinese textbooks, journals, and monographs on stereotactic and functional neurosurgery
. Figure 11‐4 Dr Yi-Cheng Zhao (1908–1974), Father of Neurosurgery of China
neurology and neurosurgery services at the Tianjin Municipal General Hospital in 1952 aided by his students Chung-Cheng Wang and Qing-Cheng Xu. The Neurosurgery Department
at the Beijing Tong-Ren Hospital was established two years later. It was later shifted to the XuanWu Hospital, the predecessor of the Beijing Neurosurgical Institute, the largest neurological center in China. A one-year intensive formal postgraduate neurosurgery training program was started in Tianjin in 1953. The graduates from these programs were to form the core group and foundation of modern neurosurgery in China. Dr Zhao established the Beijing Neurosurgical Institute in March 1960, holding the position of director until his untimely demise in1974, whereupon his student Chung-Cheng Wang took over. The Beijing Neurosurgical Institute, which is now housed within the Beijing Tiantan Hospital, is currently the national clinical, research, and training center with over 300 neurosurgical beds divided into craniocerebral injury, cerebrovascular disease, spinal injury, skull base surgery, intracranial tumors, pediatric neurosurgery, stereotactic and functional neurosurgery, and neurointensive care units in addition to ten research departments. Its basic neuroscience research departments cover neuroanatomy, neurotransmitter, neuropathology, neuropharmacology, neuroepidemiology,
History of stereotactic surgery in china
cytobiology, neurophysiology, immunology, neurochemistry, and electron microscopy. The faculty is led by prominent neurosurgeons like Chung-Cheng Wang, Ya-Do Zhao, Shi-Qi Luo, and Ji-Zong Zhao. More than 1000 neurosurgeons have benefited from its training programs till date. Further south in Shanghai, yet another city subjected to early Western influence after the Opium War with Britain in 1840–1843, neurosurgery service was established in 1953 in the Shanghai Red Cross Society Hospital, the predecessor of the Shanghai Huashan Hospital, by Yu-Quan Shi and Zhen-Qin Zhu. This prestigious institution has been credited with the first Chinese-made stereotactic frame and the first indigenously designed operating microscope and pioneered hemispherectomy for infantile hemiplegia in 1959. Postgraduate neurosurgery training programs have been offered since 1958. Shi retired from the Chair of the Department of Neurosurgery in 1989. The Shanghai Huashan Hospital was incorporated into the Shanghai Huashan Institute of Neurological Surgery (SHIN) in March 2000, with a total annual operative statistics exceeding 4,000. After the inception of the PRC in 1949, there were periodic sociopolitical upheavals, in particular the notorious Chinese Cultural Revolution from 1966 to 1976, causing widespread social, political, and economic chaos throughout the country. Scholastic aptitude was considered bourgeois and decadent and could constitute the basis for severe persecution. Not surprisingly, there was a striking absence of neurosurgical publications during this ‘‘Dark Age’’ of modern Chinese history. Until the beginning of the long anticipated economic reform in 1978 by leader Deng XiaoPing, the country endured diplomatic and cultural isolation, compounded further by economic embargo from Western nations due to ideological difference, a legacy of the Cold War. Language barrier and paucity in resources meant that innovative Chinese neurosurgeons had to develop
11
skills de novo and design diagnostic and surgical equipment, often under conditions harsh by Western standards. Their works have been documented in monographs and native Chinese journals such as the Chinese Medical Journal, Chinese Journal of Neurosurgery, Chinese Clinical Neurosurgery, and subspecialty journals, e.g., Chinese Journal of Stereotactic and Functional Neurosurgery, and Chinese Journal of Minimally Invasive Neurosurgery. It is only perhaps in the last decade or so that more and more Chinese neurosurgeons armed with better command of foreign languages have made their works known to the rest of the world through international journals such as Journal of Neurosurgery, Neurosurgery, British Journal of Neurosurgery, Stereotactic and Functional Neurosurgery, etc. (> Figure 11‐5). . Figure 11‐5 The first issue of the Chinese Journal of Stereotactic and Functional Neurosurgery (published in Anhui Province, China, 1986)
129
130
11
History of stereotactic surgery in china
Stereotactic and Functioning Neurosurgery Historical Developments in China In the 1940s, psychosurgery was reportedly being carried out by Dr T.H. Zhang. Stereotactic neurosurgery has advanced with strides in North America and Europe since 1946 when SpiegelWycis performed the first human stereotactic pallidotomy [7,8] and Prof. Lars Leksell invented of the Leksell stereotactic frame using polar coordinates in 1949 [9]. Nevertheless, the first surgical treatment for Parkinson’s disease in China was a freehand transorbital pallidotomy using Novocain and Iodipamide by Chung-Cheng Wang in 1959 [10]. Chung-Cheng Wang and fellow contemporaries Jian-Pin Xu, Mao-Shan Wang, and Da-Jie Jiang were the forerunners of stereotactic and functional neurosurgery in China.
In 1963, Dr Jian-Ping Xu, currently in the Guangdong Hospital of Traditional Chinese Medicine, completed a two-year training at the Moscow Neurosurgical Institute under the tutelage of Dr Edvard I. Kandel, who was the first Soviet neurosurgeon trained in stereotactic surgery. In the same year, Xu embarked on stereotactic surgery on patients with Parkinson’s disease using his self-designed Cartesian coordinates–based stereotactic frame in Anhui [11]. Sheer necessity and lack of recourse to expensive Western products motivated many Chinese surgeons, notably MaoShan Wang and Da-Jie Jiang of the Shanghai Medical University, in the early 1960s to design stereotactic devices to be used in, e.g., pallidotomy and thalamotomy for extrapyramidal disease [12–14]. Da-Jie Jiang’s frame (> Figure 11‐6 left) was an attempt to incorporate features of the two main groups of stereotactic devices available in the West then. Devices in the first category were
. Figure 11‐6 Stereotactic head frame designed by Da-Jie Jiang, Shanghai, 1964 (left) and with Li Pan 1989 (right) (with permission)
History of stereotactic surgery in china
structurally and mathematically complex, bulky, and difficult to operate. Specialized X-rays were needed. However, they offered a high degree of accuracy. These were exemplified by the SpiegelWycis, Leksell, Riechert, and Talairach equipment [7,9,15,16]. The second was represented by the Cooper and Austin devices [17,18]. These were structurally simple and easy to use but were compromised by a lower degree of accuracy even with repeated intraoperative adjustments with X-rays. Complications were also more prevalent. Limitations in target precision, control of lesion size, and higher degree of complications resulted in the waning of enthusiasm for these early stereotactic procedures (> Figure 11‐7). During the first decade of Western stereotactic neurosurgery, surgeons designed and custom-made their own stereotactic apparatus, as commercially produced frames were not available. The Spiegel-Wycis stereotactic frame was in fact built by Davis, an English machinist, in the workshop of the Temple University Medical Center in Philadelphia. There was also a whole array of Chinese-designed stereotactic equipment, the most prominent among these being
11
the XZ-I to XZ-V brain stereotactic apparatus by Jian-Pin Xu of the Anhui Provincial Hospital (1964–1977). This was not unlike the Guiot– Gillingham frame, small and simple in design, being anchored to the calvarium, and not requiring sophisticated supportive equipment [11,19,20]. However, only procedures within the small confines of one cerebral hemisphere were possible and a lower degree of precision was afforded. The FY85II, designed by the Xian Fourth Military Hospital in 1985, was quite similar in principle to the Todd-Wells and Leksell frames. The patient’s head was secured within the base ring. It was based on the arc principle and had a wider range of maneuverability, permitting bilateral hemispheric procedures with a higher degree of accuracy of within 2 mm. Other stereotactic equipment invented included the DZY-A from Nanjing and a frame by Wu SL from the Guangdong Minimally Invasive Neurosurgery Medical Center, Guangzhou, during the same period (> Figure 11‐8). In 1989, Da-Jie Jiang and Li Pan designed a stereotactic instrument incorporating computerized tomography(CT)andmagneticresonanceimaging(MRI) guidance (> Figure 11‐6 right). The HB-set
. Figure 11‐7 Dr Jian-Ping Xu with his stereotactic device in 1986 (left) and operating with his computer-assisted CT-guided equipment (right)
131
132
11
History of stereotactic surgery in china
. Figure 11‐8 Stereotactic device used by Jiang (left) and Wu SL in 1983 (right)
stereotactic equipment, the fruit of a collaboration between the Beijing Science and Engineering University and the Beijing Naval General Hospital, was launched in 1995 [21]. This later phase of intense activity in research and development of instruments was made possible by the modern economic and sociopolitical reform of the late Chinese leader Deng Xiao-Ping in 1978. Rapid acquisition of wealth, prosperity, and affluence in China after 1978 resulted in enhanced purchasing power, allowing practically unhindered access to Western technology. Accelerated growth in a very technology-dependent field such as neurosurgery inevitably ensued. Many hospitals have since acquired CTand high-resolution MRI instruments. Positron emission tomography (PET), single photon emission computerized tomography (SPECT) and magnetoencephalography (MEG) machines were also installed in some of the leading centers. Computer-assisted frame-based and frameless neuronavigational surgical procedures were performed. Some hospitals employed homemade systems. In the West, computer-assisted stereotactic neurosurgery was available as early as
1969. Digital CT- and MRI-image-guided stereotactic surgery was introduced in China in 1973 and 1980, respectively. Following Schaltenbrand-Wahren’s publication of the Atlas for Stereotaxy of Human Brain in 1977 [22] and Ronald Tasker’s release of the Physiological Atlas of the Thalamus and Midbrain Using Electrical Stimulation [23], Jia-Qing Yao et al. produced the Stereotactic Anatomy of Brain Gray Matter (in Chinese) for use by Chinese neurosurgeons in 1983. In 1992, the same author published the Applied Stereotactic Anatomy of the Human Brain with specific reference to the Chinese population [24,25].
Training Programs, Professional Organizations, and National Conferences The earliest generation of Chinese neurosurgeons received their stereotactic neurosurgery training abroad. Jiang-Ping Xu was trained in Moscow under the distinguished Russian
History of stereotactic surgery in china
pioneer stereotactic surgeon Dr Edvard I. Kandel from 1960 to1963. Cheng-Yuan Wu was a visiting scholar in stereotactic surgery for one year at the University of Utah, USA, with Dr M. Peter Heilbrun. Others like LZ Cheng, XM Fu, and ZP Ling were subsequently trained at Karolinska Hospital, Sweden, and Henri Mondor Hospital, Paris. Many of the leading stereotactic and functional neurosurgeons currently holding senior positions in major Chinese neurosurgical centers have received training in well-known international centers including UCLA, John Hopkins Hospital, Karolinska Hospital, and Hannover Hospital, among many others. After the first Chinese Stereotactic Neurosurgery Institute was established in Anhui Province in 1983 (> Figure 11‐9) by Jian-Ping Xu and Ye-Han Wang, indigenous stereotactic neurosurgery training program became available to native Chinese neurosurgeons with Dr Xu as the program director. A Computer-assisted Stereotactic Training Course was organized the next year. The Annual National Workshop on Stereotactic and Functional Neurosurgery was hosted by the
11
Anhui Provincial Stereotaxic Neurosurgical Institute in 1987. By 1986, the total number of Chinese-made stereotactic devices in use had reached 300, and total number of stereotactic surgeries exceeded 1,000. It is estimated that as many as 80% of the current practicing stereotactic and functional neurosurgeons have benefited from the Anhui workshops. At various intervals, other leading medical centers, e.g., Beijing Neurosurgical Institute, Beijing General Hospital of Armed Police Force, Tianjin General Hospital, Shanghai Huashan Hospital, and hospitals in Guangzhou and Harbin, offer training in stereotactic neurosurgical techniques. The Anhui Provincial Stereotaxic Neurosurgical Institute, in particular, has trained more than 400 neurosurgeons in stereotactic and functional neurosurgery to date. The Neurosurgery Specialty Society of the Chinese Medical Association (CMA) was established in March 1986 with Chung-Cheng Wang as the Chairman. Wang, born 1925, played a very pivotal role in modern neurosurgery in the PRC, having mentored more than one-third of the
. Figure 11‐9 The first Stereotactic Neurosurgery Institute in Anhui Province, China
133
134
11
History of stereotactic surgery in china
current generation of Chinese neurosurgeons. Apart from a very illustrious neurosurgical career, being the pioneer in brain stem microneurosurgery and the first to perform cerebral angiography in China during the impoverished postliberation and Korean War era of 1950s, he served in the Medical Corps in the Korean War against the US-led Allied Forces in 1952 and has been elected as representative to the People’s Congress of China (the Chinese equivalent of the Parliament). The Medal of Honor was bestowed upon him during the XII World Congress of Neurosurgery held in September 15–20, 2001 in Sydney, Australia. The Chinese Society of Stereotactic and Functional Neurosurgery was formed in July 1987 under the umbrella of the CMA. The first National Stereotactic and Functional Neurosurgery Conference in June 6–8, 1987 at Hefei, Anhui Province, was organized by the Neurosurgery Specialty Society of the CMA and attracted 155 participants with 85 papers presented (> Figure 11‐10). It has since been held seven times: May 16–20, 1990, in Chendu; May 9–13, 1993, in Dalian; May 12–16, 1997, in Beijing; August 1–4, 2001, in Harbin; June 6–10, 2004, in Yinchuan; June 26–30, 2006, in Huangshan.
In November 16–19, 2007, the First National Congress of Functional Neurosurgery was held in Xiamen. This was organized by the Stereotactic and Functional Neurosurgery Specialty Committee, Neurosurgery Association of the Chinese Medical Doctors Association (CMDA), a parallel organization to CMA. It was attended by more than 400 participants. The first National Congress for Brain Tissue and Neural Cell Transplantation was held in Kunming in January 1990. The Chinese Association of Epilepsy Surgery was established in 1990 and organized its first National Epilepsy Surgery Work-shop in the following year. In December 1998, the Guangdong Minimally Invasive Neurosurgery Medical Center organized the first National Minimally Invasive Neurosurgery Conference in Guangzhou. The biannual Shanghai International Symposium on Functional Neurosurgery is hosted by the Center for Functional Neurosurgery of the Shanghai Medical University. The Shanghai Medical University, founded in 1927 as the National Fourth Zhongshan University Medical College, was incorporated into the Shanghai Huashan Institute of Neurological Surgery (SHIN) in March 2000. Invited, distinguished international speakers to these national conferences and other meetings
. Figure 11‐10 The historic First National Stereotactic and Functional Neurosurgery Conference. Hefei City, Anhui Province
History of stereotactic surgery in china
have included Bjorn Meyerson, Bengt Linderoth (Sweden); Andres Lozano (Canada); Philip L. Gildenberg, Weiser, Philip Starr, Antonio.A.F.De Salles (USA); and J.P. Nguyen, Pr Pierre Cesaro (France). In recent years, Chinese neurosurgeons have increasingly made their presence felt in regional and international conferences such as the Congress of the Asian Society for Stereotactic, Functional and Computer-assisted Neurosurgery (ASSFCN), Congress of European Society for Stereotactic and Functional Neurosurgery (ESSFN), World Congress of the World Society for Stereotactic and Functional Neurosurgery (WSSFN), WFN World congress on Parkinson’s Disease and Related Disorders, Congress of Neurological Surgeons (CNS), and the American Association of Neurological Surgeons (AANS) meetings. Frequent scholastic exchange with overseas counterparts has further enhanced international collaboration.
Modern Chapter of Chinese Stereotactic and Functional Neurosurgery Within the half a century since stereotactic and functional neurosurgery began in China, remarkable progress has been made. However, there is relatively less emphasis on research, be it laboratory or clinical, no doubt due to the immense workload and the difficulty of gathering follow-up information from patients spread over such a vast countryside. Leading neurosurgery centers can now offer stereotactic and functional neurosurgical treatments with cutting edge technology comparable to the West, by neurosurgeons with impeccable credentials. The Beijing Tiantan Hospital with its Beijing Neurosurgical Institute is one of the world’s top three neurosurgical institutes and Asia’s largest center for neurosurgical treatment, training, and research. It was designated by the World Health
11
Organization (WHO) in 1982, together with the Shanghai Huashan Hospital, as the WHO Collaborating Centers for Research and Training in Neurosciences in China. The Shanghai Huashan Hospital was incorporated into the Shanghai Huashan Institute of Neurological Surgery (SHIN) with the Gamma Knife Hospital and two other affiliated hospitals in March 2000. The large number of practicing stereotactic and functional neurosurgeons became self-evident when close to 500 registered participants attended the First National Congress of Functional Neurosurgery in November 2007 in the southern city of Xiamen which overlooks the Taiwan Strait. Sophisticated modern neurosurgical treatments however remain largely the exclusive and privileged domain of the elite urban rich, as health insurance has yet to gain a foothold in this part of the world. Huge disparity still exists in wealth, education, and social welfare between the affluent cities of the eastern coastal provinces, and the relatively more deprived rural areas in the interior. The paradox exists that while urban China can boast of world-class hospitals and treatment, the rural poor do not have easy access to sometimes even basic medical facilities. It is commendable that the problem is being aggressively addressed by the Chinese central government through its emphasis on preferential developmental programs for the western regions. Providing adequate healthcare to one-fifth of the world’s population spread over such a vast country is a daunting task for any government indeed. Nevertheless, the huge population with seemingly unstoppable economic progress offers a golden opportunity and a rich substrate for a quantum leap of development of this technologydriven division of medical science. Acquisition of abundant clinical data permits derivation of useful observations and conclusions on treatment modalitiesfrom huge pools of patients. J.N. Zhang published 580 case reports on stereotactic intracranial procedures for Parkinson’s disease in 2000 [26], and S.B.Yuan reported on 1,431 cases
135
136
11
History of stereotactic surgery in china
of intracranial lesions treated with the Chinesedesigned rotating gamma system (OUR-XGD) in the Chinese Journal of Stereotactic and Functional Neurosurgery in 2001 [27], while 3,094 patients were treated with the Leksell Gamma Knife by Liang JC in Guangzhou up to 1999 [28]. The extreme degree of subspecialization within many of the large metropolitan or university units, the envy of many foreign neurosurgical centers, permits rapid development of subspecialty neurosurgery in China. Unencumbered by legislative regulatory constraint and working within a much less litigious environment compared to the West, Chinese neurosurgeons are quick to embrace new modalities of treatments, admittedly some still controversial, ahead of their European or American counterparts. The Beijing-based General Hospital of Armed Police Force, for example, established its Department of Neural Stem Cells Transplantation in March 2004 and has since treated close to 200 patients with strokes, traumatic brain injury, spinal cord injury, motor neuron disease, and cerebral palsy [29]. The Xishan Hospital in Beijing has an ongoing program on the implantation of olfactory ensheathing cells (OEC) from the olfactory bulb of aborted fetuses on patients with diagnoses ranging from amyotrophic lateral sclerosis and spinal cord injury to Parkinson’s disease. It has reportedly treated over 600 cases since 2001 [30,31], while in Europe, for example, the laboratory research by Geoffrey Raisman at the Institute of Neurology, Queen Square, London, on adult autotransplanted OEC on rats is currently still awaiting human clinical trials [32]. These nascent therapies have not yielded the same consistent and reproducible results of, for example, deep brain stimulation (DBS) in Parkinson’s disease. Psychiatric neurosurgery has largely fallen into disrepute under public and political scrutiny and clouded by controversy in the West for the past two decades. This was due partly to the risk of litigation, ironically sometimes from patients
rendered well enough by psychosurgery to initiate legal proceedings, and refusal of reimbursement by insurance companies. The 1986, the U.S. Office of Health Technology Assessment report which stated that ‘‘psychosurgery should be considered experimental as it has never been studied in a scientific manner’’ further discouraged its more widespread acceptance. However, we have witnessed its unabated development in China in the same period. For example, psychosurgery for the alleviation of drug dependence gained popularity in China from 2000 to 2004.
Frame-Based Surgery, Neuronavigation, and Robotics This has seen perhaps the most impressive growth since digital imaging, and the advent of stereotactic instrumentation has brought stereotactic surgery into the realms of many Chinese neurosurgeons, who like their Western counterpart have hitherto confined their works to general neurosurgery [33,34]. Whereas previously stereotactic imaging had to rely on angiography and ventriculography, identification of intracranial landmarks can now be made by noninvasive CT or MRI. Diagnostic tissue biopsies, evacuations of hematomas and abscesses, intracavitatory instillations of radioisotopes for tumors [35,36], e.g., combined 32P and Methotrexate chemotherapy for deep-seated gliomas, interstitial 192Ir brachytherapy for glioblastomas [37,38], and transplantations of neural tissues with minimally invasive techniques have been widely carried out in China. CT-guided neuroendoscopies have also been performed. In 1998, Tian ZM from the Beijing Navy General Hospital of People’s Liberation Army reported on 1,300 cases of CT-guided operations under local anesthesia using both the Leksell stereotactic frame and the indigenous HB-III instrument [39]. Intratumoral instillation of radionuclide and chemotherapeutic agent (BCNU) accounted
History of stereotactic surgery in china
for 780 and 24 cases, respectively. Lesioning of functional targets (101 cases), brain tissue biopsy (171 cases) and evacuation of intracerebral hematomas (155 cases) constituted the remainder. The Chinese-made NDY stereotactic frame has also been employed by some neurosurgeons. Frameless stereotaxy or neuronavigation permits neurosurgeons to perform stereotacticassisted volumetric excision of intracranial space occupying lesions accurately, greatly minimizing procedure-related risks [40–42]. It is now being used with intraoperative MRI at the Beijing Tiantan Hospital and Shanghai Huashan Hospital. Neuronavigation with functional MRI (fMRI) has been used in surgery near eloquent areas [43]. Neuronavigation-assisted neuroendoscopies have also been carried out with a high degree of success [44]. It is has become a standard inventory in neurosurgery operating suites in major neurosurgical units in Beijing, Shanghai, Guangzhou, Tianjin, Xian, and other cities. Neuronavigational systems available in China include the SurgiScope (Elekta), StealthStation (Medtronics), VectorVision (BrainLab), and the China-made ASA-610V and ASA-630V (Shenzhen Anke High Tech Co.) [45]. Imageguided, robot-assisted stereotactic system (CRAS HB1) developed by the Beijing Aeronautical and Astronautical Institute was used by Tian ZM of the Beijing Naval General Hospital in 1997 on 32 patients comprising 23 cases of intratumoral brachytherapy, 3 cases of tissue biopsy, and 2 cases each of evacuation of abscess and hematoma, achieving an instrument efficacy of 86% [46].
Surgery for Movement Disorders The first pallidotomy for Parkinson’s disease was done freehand in1959 by Zhong-Cheng Wang, who is the president of the Chinese Neurosurgical Society and the Beijing Neurological Institute. Wang designed the transorbital approach using trocar puncture of orbital roof, contralateral to
11
the side of tremor or rigidity, 4 cm from midsagittal plane aiming 10 superiorly and medially. An FG 20–22 needle was then advanced to the frontal horn, replacing 30–40 ml of cerebrospinal fluid (CSF) with filtered air. The same needle after being withdrawn, was then reinserted a few degrees vertically inclined from Reid’s base line (inferior orbital to superior external auditory meatus) and 10 deviated towards contralateral side to a depth of 9 cm. The pneumoventriculogram thus created confirmed placement of needle tip 2 cm posterior and 1 cm inferior to foramen of Munro, 2 cm lateral to mid-sagittal plane. After 0.5 ml of 1% Novocain confirmed favorable response, 0.5 ml of 40% Iodipamide was introduced. This simplified procedure was intended to bring pallidotomy into the hands of many doctors who may have to practice under relatively Spartan circumstances [10] (> Figure 11‐11). Two years later, Mao-Shan Wang’s stereotactic device was put to fruition and he carried out stereotactic alcohol ablation of globus pallidus, quite akin to the original chemopallidectomy that I.S. Cooper published in Science in 1955. (Cooper subsequently switched the target to the thalamus and reportedly performed over 3,000 cases of chemo-thalamectomies.) [47,48]. MS Wang published his stereotactic procedure ‘‘Surgical Treatment for Parkinsonian Syndrome’’ in the Chinese Journal of Surgery in 1961 [12]. Similar endeavors using self-designed stereotactic apparatus were undertaken by the pioneer surgeon Da-Jie Jiang, who published his twoyear preliminary reports on 20 cases of extrapyramidal system disease in the Chinese Journal of Neurology and Psychiatry in 1964. Jiang’s targets included globus pallidus, ventro-lateral nucleus, centro-median nucleus, nucleus reticularis, and internal capsule depending on the predominance of tremors, rigidity, and upper or lower limb involvement [14]. Localization was based on the Spiegel–Wycis atlas and coordinates [49]. Intraoperative confirmation of electrode placement was
137
138
11
History of stereotactic surgery in china
. Figure 11‐11 Dr Wang’s patient feeding with previously tremulous left hand. Handwriting in dots before (above) and after procedure of another patient, 1959 (with permission from CC Wang)
carried out in three stages: using 0.5 ml 1% Novocain instilled into target through electrode needle, electrical stimulation, and depth electrode recording. Parkinson’s disease constituted the largest category of conditions treated, and the early procedures were predominantly ablative in nature. A large series of papers was published by Guo-Dong Gao from the Xian Fourth Military University (1,478 cases) and by Zhang JN (580 cases) [26,50]. Following Backlund’s first clinical trials in 1982 and Madrazo’s 1987 report on open microsurgical adrenal medullary tissue transplantation to the striatum for parkinsonism demonstrating therapeutic benefit [51,52], Wa-Cheng Zhang et al. of the Beijing Xuanwu Hospital embarked on transplantation of adrenal medulla to the head of caudate nucleus for six patients with Parkinsonian tremors using a XZ-III stereotactic device in 1987, using postoperative CT scan to confirm placement of transplant tissue carriage in the head of caudate nucleus. CSF dopamine and normetanephrine levels were elevated to 1–2 times of presurgical assay, and clinical symptomatic improvements of tremors were documented [53]. In 1994, Wu CYet al. published their results of combined fetal substantia nigra
tissue transplantation and stereotactic thalamotomy in the British Journal of Neurosurgery [54]. Gamma knife thalamotomy (Vim, VO) using a 4-mm collimator delivering 130 Gy for the treatment of 58 cases Parkinson’s disease and other movement disorders was reported by Zhang Jie et al., claiming 80% good response over a follow-up period of up to nine years [55]. The first human microelectrode recording (MER) during functional brain surgery was reported by Albe-Fessard in 1961. MER was introduced in China in 1998 by Yong-Jie Li of the Beijing Institute of Functional Neurosurgery, Xuanwu Hospital. More than 2,000 cases of pallidotomies with CT or MR guidance have been carried out annually in China [50,56,57]. A significant proportion of those employed intraoperative MER for physiologic confirmation of electrode placement. Li published a review of 1,135 cases of surgical treatment of movement disorders in 2001 [58]. In 1972, Bechtereva N.P. of the former Soviet Union carried out therapeutic stimulation of the deep brain structures using intermittent external stimulation [59]. Siegfried and Benebid started DBS surgery for Parkinson’s disease in 1985 after the former unexpectedly noticed
History of stereotactic surgery in china
amelioration of Parkinsonian tremors from the implanted thalamic deep brain stimulator meant for chronic pain [60,61]. The U.S. FDA has approved DBS as a treatment for essential tremors, Parkinson’s disease, and dystonia in1997, 2002, and 2003, respectively. Led by Chung-Cheng Wang, the Beijing Tiantan Hospital performed the first DBS surgery for Parkinson’s disease in China in September 1998. Visits by Dr Philip A. Starr, University of California San Francisco (UCSF), and Dr Andres Lozano, Toronto Western Hospital, to China in 1999 helped DBS surgery, which gained further popularity. By 2003, more than 300 personal cases of DBS of subthalamic nucleus of Luys were performed by leading functional neurosurgeons like Jian-Guo Zhang, Yong-Jie Li (Beijing), and Bomin Sun (Shanghai). Other functional neurosurgeons like Kang-Yong Liu, Xiao-Wu Hu (Shanghai), Zhi-Pei Ling (Anhui), Shi-Zhong Zhang (Guangzhou), and Guo-Dong Gao (Xian) have also accumulated considerable
11
personal experience in DBS surgery. Their works have been variously published in Chinese Journal of Neurosurgery in 2002 [50,58,62,63]. There are more than 30 hospitals in China where DBS for Parkinson’s disease are regularly performed (> Figures 11‐12 and > 11‐13). The total number of cases done since 1998 has exceeded 1,000. In the Beijing Tiantan Hospital, Xuanwu Hospital, and Shanghai Ruijin Hospital, DBS treatment has also been extended to dystonia, Tourette syndrome, Hallervorden–Spatz disease, Meige’s syndrome, and chorea using subthalamic nucleus as target [26,64]. In 2002, Bo-min Sun used subthalamic nucleus of Luys for primary and tardive dystonia and achieved 90% 3 months to 3 years postoperative improvement based on Unified Parkinson Disease Rating Scale (UPDRS) and the Burke–Fahn–Marsden Scale [65]. Microelectrode recording were generally used in DBS surgery for intraoperative neurophysiologic targeting like in other overseas centers [66,67].
. Figure 11‐12 Deep brain stimulation surgery at the Shanghai Medical University (Fudan University) Huashan Hospital
139
140
11
History of stereotactic surgery in china
. Figure 11‐13 DBS electrodes in subthalamic nucleus. (Shanghai Huashan Hospital)
Stereotactic Radiosurgery Stereotactic radiosurgery was first introduced into China in 1993 when the Leksell gamma knife was installed in Wanjie Hospital in the northeastern port city of Qingtao. This was soon followed by the Beijing Tiantan and Shanghai Huashan Hospitals. There are more than 50 stereotactic radiosurgery centers in China. The Chinese-designed rotating gamma system OURXGD was launched in 1994 and relied on only 30 cobalt-60 sources instead of 201 in the Leksell system. The SGS-I stereospecific gamma ray whole body treatment system (> Figure 11‐14) has since been launched. It is estimated that more than 8,000 cases of stereotactic radiosurgery and radiotherapy are treated annually, with Beijing Tiantan Hospital alone accounting for more than 10,000 cases till December 2007. Currently, there are 17 Leksell gamma knife radiosurgery installations (compared to 35 in the USA), almost an equal number of LINAC X-Knife units, over 20 rotating gamma system (OUR-XGD) units, 1 Novalis (Brain Lab) ststem, and 4 CyberKnife (Accuray Inc.) systems in mainland China. The OUR-XGD rotating gamma
system and the SGS-I have been exported to countries such as Egypt, India, and Hungary. In 1999, Liang JC and Wu HX of the Guangzhou General Hospital of PLA reported on the gamma knife treatment of 3,094 cases of intracranial lesions including brain tumors (with or without cytoreductive surgery), arteriovenous malformations, and even functional targets [28,68]. From July 1995 to May 1998, they had treated 280 cases of arterio-venous malformations, of which 42 were Spetzler–Martin grade I, 68 grade II, 95 grade III, and 7 grade IV and grade V 4, with a high obliteration rate for grade I and II. Another large clinical series of 1,431 cases was published by Yuan SB of the Sichuan Gamma Knife Center, Chengdu Army Hospital, using the OUR-XGD rotating gamma system over a four-year period from January 1997 to January 2001 [27]. The results of the LINAC X-Knife treatment on 510 cases were documented by Wang LG in 2000 [69]. Gamma knife radiosurgery have been utilized for thalamotomy in Parkinson’s disease and other functional disorders [55]. It is worth noting that Lars Leksell actually conceived stereotactic radiosurgery in 1951 for the treatment of the functional disorder
History of stereotactic surgery in china
11
. Figure 11‐14 SGS-I: Chinese-made whole-body rotating gamma ray treatment system
of obsessive compulsive neurosis apart from tic douloreaux [9]. X-Knife radiosurgery for intractable epilepsy had been carried out by Qi ST on focal functional areas after preoperative electroencephalography (EEG), MRI, and PET evaluations [70].
Epilepsy Surgery Treatment of epilepsy in China can be traced to ancient times. Apart from Neijing (Yellow Emperor’s Inner Classic) 2000 years ago, there have been many medical classics dating back to Tang Dynasty (Qian Jin Yao Fang – The Invaluable Medical Script 652 A.D.), Sung Dynasty (Ji Sheng Fang Medical Script 1253 A.D.), Yuan Dynasty (Danxi Miscellaneous Comprehensive Medical Text 1481 A.D.) and Ming Dynasty (Treatment Yardstick 1600 A.D.), where traditional (TCM) treatments of epilepsy were described. Victor Horsley introduced surgery for medically intractable epilepsy in 1886. It is estimated that about one-quarter of epilepsy patients
become medically intractable, rendering them candidates for surgical intervention. Epilepsy surgery was accepted as an alternative treatment by the U.S. National Institute of Health Consensus Development Conference on Surgery for Epilepsy, March 19–21, 1990. In China, with a population of 1.3 billion, the total number of surgeries performed for medically refractory epilepsy had quadrupled from 600 before the year 2000 to 2500 in the year 2005 with a quarter from Beijing alone [71]. Review of the medical literature and hospital records indicate that the earliest documented works on epilepsy in China were in the 1950s, when Guo-Sheng Duan reported on lesionectomy for post-traumatic epilepsy with the aid of electrocorticography [72]. Krynauw RA first reported the removal of one cerebral hemisphere as the treatment for infantile hemiplegia [73]. Yu-Quan Shi of Shanghai First Medical College performed hemispherectomy under general anesthesia on four cases of severe infantile hemiplegia aged 3 years 5 months to 9 years in 1956. Pneumoencephalography demonstrated
141
142
11
History of stereotactic surgery in china
ipsilateral ventricular and cisternal enlargement, ventricular shift to contralateral side, as well as porencephaly. A normal contralateral ventricle was a prerequisite. EEG was used in cooperative patients. Hemispherectomy was done en bloc as opposed to Krynauw’s cruciate excision of the hemisphere in four blocks, sparing the caudate nucleus and thalamus (> Figure 11‐15). Shi observed improved seizure control, alleviation in aggressive behavior, and slight improvement in developmental IQ, but persistence of hemiplegia. Spasticity and hypertonia, though reduced, did not translate into improvement in dexterity of hand movements [74]. Ya-Du Zhao of Beijing Xuanwu Hospital also reported his experience on 12 cases of hemispherectomy and anterior temporal lobectomy from 1959 to 1963, employing intraoperative electrocortical recording and bipolar direct electrical cortical stimulation [75] (> Figure 11‐16). Shi and Zhao were considered the pioneers who laid the foundation of modern epilepsy surgery in China. (Ya-Du Zhao was the younger son of Yi-Cheng Zhao, the ‘‘founder of neurosurgery in China’’). Subsequent works were done by
Chen-Ji Liu of the Nanjing Military General Hospital and Zhi-Xun Wu of the Kunming Medical College in 1963. Jian-Ping Xu and Ye-Han Wang of the Anhui Provincial Hospital popularized stereotactic ablative epileptic surgery using homedesigned stereotactic equipment in the 1970s. In 1978, surgical treatment for epilepsy was finally made available in the western interior province of Sichuan by Li-Da Gao, Ge Wu, and ChangGui Zhou. After a quiescence of 10 years during the unfortunate Chinese Cultural Revolution (1966–1976), resurgence of epileptic surgery was made possible by the rapid socioeconomic development commencing 1978. Many procedures were undertaken, including temporal lobectomy, corpus callosotomy, neuroaugmentation (vagal nerve and cerebellar stimulation), and multiple vertical subpial transaction (MST) (Frank Morell, Rush-Presbyterian-St Luke’s Medical Center) [71,76–78]. Techniques in stereotactic corpus callosotomy and hippocampal resection were published by Jian-Ping Xu and Ye-Han Wang of the Anhui Provincial Stereotaxic Neurosurgical Institute in 1984 [79], while in 1994 Xiao AP et al. from
. Figure 11‐15 Pneumoencephalography showing left cerebral hemispheric atrophy, compensatory ipsilateral ventriculomegaly (left picture), and the medial surface of the excised left cerebral hemisphere (right picture). (DrYu-Quan Shi, 1956, Shanghai, with permission)
History of stereotactic surgery in china
11
. Figure 11‐16 Electrocorticography – unipolar lead tracing. Site A showed positive spike waves, electrical stimulation reproduced a subdued form of the original seizure. (Dr YD Zhao, Beijing, 1959, with permission)
the Nanjing Medical University reported 51 cases of stereotactic ablation of Forel H. Field (Mukawa J 1966) [80,81] and unilateral or bilateral amygdalotomy for 51 patients, aged 7–54 years, over a seven-year period since 1987 [82]. Bilateral cingulotomy and anterior capsulotomy were supplemented on 16 patients who had mental disorder. Procedures were done under local anesthesia except for four children. Intraoperative target localization was aided by pulsed electrical stimulation and acoustic impedance. Through similar works by C.C. Jiang and DJ Jiang of the Shanghai Huashan Hospital (1989) [83], Tang YL, and Zhang ZX, 82 cases (1990) [84] have been published. Gui-Sing Wang in 2004 used depth electrode for hippocampal EEG monitoring and stereotactic guidance in temporal lobe epilepsy surgery [85]. Neuronavigation and electrocorticography have been used in the resection of tumors causing secondary epilepsy [86,87]. Notable contributions were made by Qi-Fu Tan from the Hospital of Nanjing Military Area Command, who proposed guidelines on patient selection, preoperative assessment, operative results, and evaluation protocols, and also improved on the techniques of temporal lobectomy.
He performed corpus callosotomy in 1983 and Rasmussen’s functional hemispherectomy (Theodore Rasmussen, Montreal Neurological Institute) [88]. His results were published in Stereotactic and Functional Neurosurgery [89]. The spectrum of operated cases currently ranges from focal or partial epilepsy due to neoplasm to focal cortical dysplasia, hippocampal hemiatrophy, epileptogenic foci in temporal or frontal lobes, and multifocal epileptic zone located at or near eloquent cortex. Standard presurgical workup includes comprehensive neurosurgical, neurological, and neuropsychiatric assessment, electrophysiological studies including video EEG monitoring, digital imaging with MRI, and fMRI and MR spectroscopy where indicated [90,91]. Alterations in cerebral metabolism in the ictal and interictal period are documented with PET or SPECT [92]. MEG and implanted electrodes have also been employed [85,93]. Zhang G.J. of Beijing Xuan Wu Hospital [94] reported his experience on the application of long-term intracranial EEG monitoring using rectangular grids, linear strips, and stereotactically implanted flexible depth electrodes in epilepsy surgery in the Chinese Journal of Neurosurgery in
143
144
11
History of stereotactic surgery in china
2005 (> Figure 11‐17). Luan G.M. from the Beijing San-Bo Brain Science Institute pioneered low-power bipolar electrocoagulation on layer III of cerebral cortex for epileptogenic foci on or near eloquent cortex for patients in whom MST was otherwise indicated, and reported good results [95]. This has been increasingly adopted by other epilepsy surgeons. However, its longterm efficacy and delayed sequelae have yet to be fully elucidated. He and his colleague Yunlin Li carried out various combinations of procedures from lobectomy, hemispherectomy, corpus callosotomy, and vagus nerve stimulation to low-power bipolar electrocoagulation on the eloquent cortex on 77 patients with multiple epileptogenic foci. Intraoperative electrocorticography was employed. Fifty-three patients achieved class I and II seizure control, based on Dr Jerome Engel Jr’s classification on surgical outcomes [96]. Stereotactic radiosurgery in epilepsy surgery was reported by Qi-Fu Tan in 2001 [97]. S.T. Qi used the X-Knife (BrainLab) with EEG, MRI, and PET for intractable seizures from focal functional areas [70]. Treatment on 38 patients diagnosed with medial temporal lobe epilepsy
from January 1998 to December 2004 using the Chinese-made rotating gamma system (OURXGD) was reported by Yuan SB, with 73.68% overall efficacy (Engel I–III) [98]. The Beijing Tiantan Hospital took the lead by setting up the first independent epilepsy surgery unit in China, with a current annual surgical census exceeding 200. The number of hospitals with an epilepsy surgery unit in Beijing City alone has tripled from three in 1990 to nine in 2005. There are between 150 and 200 practicing epilepsy surgeons in China responsible for over 2,500 cases per year. Epilepsy surgery is now available in Nanjing, Guangzhou, Harbin, Chongqing, Chendu, Kunming Anhui, Wuhan, Shandong, Xinjiang, and Shijiazhuang, apart from Beijing and Shanghai. The National Epilepsy Surgery Society was formed in 1990, with the first National Epilepsy Surgery Workshop held in the next year in Qufu (birth place of Confucius). It has since been held four times on annual or biannual basis. The Chinese Newsletter of Epilepsy Surgery was published in 1992 with Qi-Fu Tan as the editor. The China Association Against Epilepsy (CAAE) was formed in June 2005. It has become a member
. Figure 11‐17 Electrocorticography in epilepsy surgery in twenty-first century China
History of stereotactic surgery in china
of the International League against Epilepsy (ILAE). Epilepsy surgery has featured prominently during six of the National Stereotactic and Functional Neurosurgery Conferences since 1987. Works by Chinese epilepsy surgeons are published in the Journal of Epilepsy Surgery, Journal of Asian Epilepsy in addition to the Chinese Journal of Stereotactic and Functional Neurosurgery. The second edition of the Textbook of Epilepsy Surgery edited by Qi-Fu Tan was released in 2006 [99].
Psychosurgery Although psychosurgery was performed in China by Tong-He Zhang of Xi’an in the 1940s, barely five years after the Portuguese Nobel Prize winner neurologist Egas Moniz conceived and performed the first operation for psychiatric disorder with neurosurgeon Almeida Lima on November 12, 1935 [100], this was followed by a lull of almost four decades. This was undoubtedly influenced by the events in neighboring fellow communist Soviet Union. Bekhterev and the father of Russian Neurosurgery Puusepp performed frontal leucotomy for maniac depressive psychosis and psychic equivalents of epileptics in 1906–1910, while classical leucotomies of Moniz and Lima for schizophrenia and severe pain were done from 1930s to late 1940s. Psychosurgery was banned in Soviet Union in1950 for ideological reasons [101]. Interests in psychiatric neurosurgery were rekindled in 1980s in a handful of Chinese hospitals in tandem with developments in the West [102]. Wu SL reported on 23 cases of schizophrenia treated with anterior cingulotomy in 1988 [103]. The first National Psychosurgery Conference was held in Nanjing in 1988, during which 542 cases of neurosurgical treatment for psychiatric disorders were presented. Procedures performed were anterior cingulotomy and amygdalectomies for various intractable mental illnesses including chronic schizophrenia and epileptosis. During the conference, the National
11
Clinical Guidelines for Psychosurgery were formulated, spelling out criteria for patient selection, informed consent, procedure types, and perioperative psychological assessments, among other issues. Since schizophrenia constituted the main bulk of operative cases, surgery expectedly did not result in amelioration of symptoms, just as Ballantine concluded in his retrospective report in 1989 on 198 patients who underwent anterior cingulotomy that schizophrenia and personality disorder responded least well to surgical treatment. A similar conclusion was made by Moniz [104,105]. Surgical option was not warmly received by psychiatrists who were never strong advocates of surgical intervention for their patients. The number of psychosurgeries dwindled progressively, and by the later half of 1990s many hospital had discontinued psychosurgery altogether. The revival of psychosurgery in China in the twenty-first century coincided with the widespread application of CT- and MRI-guided stereotactic surgery in the field of neurosurgery, allowing components of the limbic system, e.g., cingulate gyrus, amygdale, etc., to be accurately and safely approached [106]. Stereotactic limbic surgery, a more appropriate term than psychosurgery, was applied to the more responsive affective, anxiety, and obsessive compulsive disorders. A better understanding of the pathophysiological substrate of psychiatric diseases resulted in more widespread acceptance of psychosurgery. The French neurosurgeon Jean Talairach and colleagues first performed anterior capsulotomy for psychiatric disorders in 1949. Radiofrequency (RF) thermocoagulation stereotactic capsulotomy was subsequently developed by Lars Leksell. In 1999, Bomin Sun of Shanghai introduced MRIguided bilateral capsulotomy (> Figure 11‐18) for refractory obsessive-compulsive disorder, which he reported in the Chinese Journal of Neuropsychiatric Disorder in 2003. Capsulotomy-induced localized orbitofrontal subcortical metabolic changes in obsessive compulsive disorder were documented with PET (> Figure 11‐19) [107].
145
146
11
History of stereotactic surgery in china
. Figure 11‐18 Postoperative MRI. Capsulotomy in obsessive compulsive disorder. (Shanghai Huashan Hospital 1999)
A few cases of DBS for obsessive compulsive disorder have been carried out by Sun, targeting the anterior capsulum. Unilateral (right) lesion combined with DBS of the dominant (left) side appeared to yield better surgical outcomes. Clinical analyses of 138 cases of multitarget stereotactic thermocoagulation of amygdala, anterior cingulum, and anterior internal capsule for intractable psychosis have been reported by Xiao-Feng Wang and Ke-Ming Jiang in the Chinese Journal of Stereotactic and Functional Neurosurgery in 2003 [106]. Since 2000, psychosurgery for opiate addiction was carried out in some Chinese hospitals. For the few years till 2004, when it was finally banned for general clinical applications by the Chinese Ministry of Health because of public outcry from its socioethical controversy, more than 500 cases have been performed. From July 2000 to November 2004, 272 cases of stereotactic
ablation of nucleus accumbens for opiate addiction were performed at the Tangdu Hospital, Xian Fourth Military Medical University [108]. (Note: The Russians had chosen cingulate gyrus as the target.) Guo-dong Gao published the clinical study on the success and relapse in the initial 28 cases, claiming 65% good to excellent results over a mean follow-up period of 15 months [109]. Other workers have also targeted the medial septal diagonal band complex [110]. One case of successful bilateral DBS on nucleus accumbens was included in Xu Ji-Wen’s 27 case reports of neurosurgical treatment for drug addiction in 2005 [111].
Functional Neurosurgery for Pain Surgical treatment for pain had progressed from interruption of pain pathways to stimulation of
History of stereotactic surgery in china
11
. Figure 11‐19 From left: Preoperative and post-PET scan of a Chinese patient with obsessive compulsive disorder
pain-inhibiting pathways. Initial reports in the 1960s by Spiegel-Wycis, Ballantine, and, more recently, Wilkinson consisted of thalamotomy and anterior cingulotomy for chronic, intractable pain [8,112,113]. DBS of periaqueductal-periventricular gray and somatosensory thalamus for the relief of cancer pain and deafferentiation pain was reported by Richardson and Akil in 1977 [114]. Meta-analysis of 13 seriescomprising 1,114patients by Young RF et al. indicated a modest average long-term success rate of 60% [115]. It is currently performed in the United States as an ‘‘off label’’ (physician discretion) procedure. Neuroaugmentation for pain with the exception of Tsubokawa’s motor cortex stimulation (MCS) [116] is rarely performed in China [117]. Instead, the procedures were predominantly ablative in nature. Jiang-Ping Xu and Chung-Cheng Wang pioneered the percutaneous RF thermocoagulation for trigeminal neuralgia in China. A large series
of 1,860 cases was reported by Cheng-Yuan Wu et al. from the Qilu Hospital of Shandong University in 2004 [118]. A stereotactic basal frame for the 3D CT localization of foramen ovale and trigeminal stereoguide was designed. Excellent or good response was noted after selective percutaneous RF thermocoagulation in 96.3% cases, with a two-year recurrence of 24.8%. X-ray guidance, 3D CT, and neuronavigation enhanced the safety and accuracy of the procedure, overcoming the shortcomings of the freehand procedure. Microvascular decompressions for trigeminal neuralgia and hemifacial spasm are regularly carried out. Gamma knife radiosurgery for trigeminal neuralgia was reported by Mian-Shun Pan et al. of the Hefei Gamma Knife Hospital, Anhui Province, on 67 patients from 1997 to 2002, achieving 87.8% relief after an average three weeks [119]. Stereotactic thermocoagulation for central pain has been done since the early 1990s. Yang
147
148
11
History of stereotactic surgery in china
FM from Harbin reported on this procedure in 1992 [120].Ventriculogram and, in more recent cases, CT- or MR-guided RF lesion of centromedian nucleus and amygdala for chronic cancer pain on 46 patients have been reported by Jing GJ in 2002 [121,122]. Hu YS and Li YJ concluded in 2005 that a combination of bilateral anterior cingulotomy with mesencephalotomy resulted in more gratifying postoperative results based on the Visual Analog Scale and the McGill Pain Questionnaire assessment of patients [123]. Long-term response is still pending. Gamma knife stereotactic radiosurgery for the treatment of cancer pain has also been reported in a series of 322 patients by Jin Hu and Li-Jie Yi [124]. Frequent psychiatric overlay of chronic pain patients and the exceptional tolerance for pain unique to a race that has endured centuries of tremendous hardship can make objective verification of pain alleviation difficult in China.
Clinical and Laboratory Research Laboratory research on low-power bipolar coagulation of epileptogenic foci has resulted in its clinical application for intractable epilepsy from eloquent cortex [125]. This was first pioneered by Luan GM of the Beijing Brain Science Institute as an alternative to multiple subpial transaction in 2002 [95]. This has proven effective and is being adopted by many epilepsy surgeons especially in China. Neural transplantation research on the survival, migration, and differentiation, and the effects of Nerve Growth Factor (NGF) or Glial cell line derived neurotrophic factor (GDNF) on transplanted embryonal or bone marrow–derived stem cells and olfactory ensheathing cells in laboratory models of cerebral ischemia, spinal cord injury, and substantia nigra are being carried out [31,126,127], hopefully to be translated to clinical application. Effects of limbic leucotomy on quinpirole-induced obsessive compulsive behavior of rats drew parallel conclusions to human clinical results [128].
Further improvement on the robotic system for stereotactic neurosurgery previously launched in 1998 and 2002 is being carried out by the Robotic Institute of Beijing University of Aeronautics and Astronautics [46], while design and upgradation of software for targeting in stereotactic surgery continues [129]. The effect of controlled-release polymers of BCNU-PLA in experimental glioma in vivo and intratumoral radioimmunotherapy by radio-iodinated monoclonal antibodies form part of the cerebral tumor stereotactic brachytherapy research and experimental treatment by Li AB at the Beijing 304 Military Hospital [130].
Literature and Journals As the numerous clinical and scientific research papers published in the local journals each year are in the Chinese language mainly for local consumption, they are not accessible to foreigners outside China. However, with a better command of foreign languages, Chinese medical professionals have begun to publish in international journals over the past decade. The Chinese Journal of Stereotactic and Functional Neurosurgery was first published by the Anhui Provincial Stereotaxic Neurosurgery Institute in 1986, initially irregularly. It is now released bimonthly (> Figure 11‐5). The Chinese Journal of Minimally Invasive Neurosurgery is a monthly publication of the Guangzhou General Hospital of Guangzhou Military Region. The Journal of Chinese Neurosurgery, the official journal of the Chinese Neurosurgical Society, is published quarterly since 1985. Other related journals include the Journal of Epilepsy Surgery, Chinese Journal of Contemporary Neurology and Neurosurgery, Chinese Journal of Clinical Neurosurgery, Chinese Journal of Neurosurgical Disease Research, and Chinese Journal of Neuromedicine. Stereotactic and functional neurosurgery books and monographs (in Chinese) published include Stereotactic Anatomy of Brain Gray
History of stereotactic surgery in china
Matter by Yao JQ and Chen YM in 1980, Functional and Stereotactic Neurosurgery by Chen BH [131], Applied Stereotactic Anatomy of the Human Brain by Yao JQ and Dai HR in 1992, Brain Transplantation by WU CY in 1993, Stereotactic Radiosurgery by Chen BH [132], Textbook of Epilepsy Surgery edited by Tan QF in 1995 (2nd edition 2006), Minimally Invasive Neurosurgery by Ma LT in 1999, The Temporal Lobe Epilepsy Surgery by Lin Li in 2003, Modern Stereotactic Neurosurgery by Tian ZM in 1997, and the Chinese translation of Epilepsy Surgery edited by H.O. Luders and Y.G.Gomair (Lippincott, Williams and Wilkins, Baltimore, 2000).
Conclusion Napoleon Bonaparte of France (1804) once said: ‘‘Let China sleep. For when China wakes, it will shake the world.’’ In the last half century, we have witnessed the spectacular rise of China from obscurity to world prominence. As China integrates herself into the world community, Chinese medical professionals have redefined their roles in the international medical fraternity [133–144].
9. 10. 11. 12. 13.
14.
15.
16.
17. 18. 19. 20.
21.
References 1. Alt Kw, Jeunesse C, Buitrago-Tellez Ch, et al. Evidence for stone age cranial surgery. Nature 1997;387:360. 2. Han KX, Chen XC. The archaeological evidence of trephination in early China. Archaeology 1999;7:63-8. 3. Jia ZF. A brief history of the development of neurosurgery in China. Chin Med J (English) 1987;100:503-8. 4. Zhao YD, Zhao YL. Neurosurgery in the People’s Republic of China: a century’s review. Neurosurgery 2002;51:468-77. 5. Kwan (Guan) ST. Trigeminal neuralgia: diagnosis and treatment. Chin Med J (English) 1932;46:260-76. 6. Chang C. Neurofibroma of the gasserian ganglion. Chin Med J 1935;49:412. 7. Spiegel EA, Wycis HT, Marks M, Lee AS. Stereotaxic apparatus for operations on the human. Science 1947;106:349-50. 8. Spiegel EA, Wycis HT, Szekely EG, Gildenberg PL. Combined dorsomedial, intralaminar and basal thalamotomy
22. 23.
24.
25. 26.
27.
28.
11
for relief of so called intractable pain. J Int Coll Surg 1964;42:160-8. Leksell L. The stereotactic method and radiosurgery of the brain. Acta Chir Scand 1951;102:316. Wang CC, Huang KQ. A new operation for parkinsonian syndrome. Chin J Neurol Psychiatr 1960;6(2):80-1. Xu JP. Stereotactic application in neurosurgery. Acta Anhui Medical College 1963;6:228-31. Wang MS. Surgical treatment for parkinsonian syndrome. Chin J Surg 1961;9(7):524-5. Jiang DJ. Investigation of stereotactic brain surgery: design and application of a stereotactic instrument. Chin J Neurol Psychiatr 1964;8:238-43. Jiang DJ. Stereotactic operation for extrapyramidal system disease: a preliminary report of 20 cases. Chin J Neurol Psychiatr 1964;8:370-3. Riechert T, Mundinger E. Beschreibung und Anwendung eines Ziegerates fur sterotaktische Hirnoperationen. Acta Neurochirurg Suppl 1954;3:309. Talairach J, Hecaen H, David M. Lobotomie prefrontal limitee par electrocoagulation des fibres thalamo-frontales a leur emergence du bras anterieur de la capsule interne. In: Congress Neurologique International. Paris: Masson; 1949. p. 1412. Parera C, Cooper IS. A modification of chemopallidectomy guide. J Neurosurg 1960;17:547. Austin A, Lee A. A plastic ball-and-socket type of stereotaxic director. J Neurosurg 1958;15:264. Gillingham FJ. Advances in stereotactic and functional neurosurgery. New York: Springer; 1980. Xu JP. The beginning and development of stereotactic and functional neurosurgery in China in the last 25 years. Chin J Stereotac Funct Neurosurg 1990;3:3-5. Tian ZM. Modern stereotactic neurosurgery (Chinese). Beijing Chinese Science and Technology Press; 1997. Schaltenbrand G, Wahren W. Atlas for stereotaxy of the human brain. Stuttgart: Thieme; 1977. Tasker RR, Hawrylshyn P, Rowe IH, Organ LW. Computerized graphic display of results of subcortical stimulation during stereotactic surgery. Acta Neurochir (Wien) 1969;21:173-180. Yao JQ, Dai HR. Applied stereotactic anatomy of the human brain (Chinese). Shanghai Science and Technology Press; 1992. Yao JQ, Chen YM. Stereotactic anatomy of brain gray matter (Chinese). Beijing Science Press; 1983. Zhang JN, Wu SL, Zhang X, et al. Brain stereotactic procedures for Parkinson’s disease: 580 cases report. Mod Rehabil 2000;4:1212-13. Yuan SB, Lei J, Tang Y, Liao SC, Shun SQ, Deng ZP, Zhang XH, Pang XX. Preliminary reporting of rotating gamma system in 1431 cases of intracranial lesions. Chin J Stereotactic Funct Neurosurg 2001;14:209-12. Liang JC, WU HX, Jun ZH, Li L, Fu XP. Preliminary report of 3094 cases of intracranial diseases treated by
149
150
11 29. 30.
31.
32.
33. 34. 35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
History of stereotactic surgery in china
gamma knife. Chin J Minimally Invasive Neurosurg 1999;4:9-12. General Hospital of Armed Police Force Beijing. http:// www.neuralstemcell.com.cn. Huang HY, Liu K. Neural regeneration and functional recovery of spinal cord injury after olfactory ensheathing cell transplantation. J Naval General Hospital (Beijing) 2001;4:65-7. Huang HY, Wang HM, Li BC, et al. Preliminary report of clinical trial for olfactory ensheathing cell transplantation in chronic spinal cord injury. Chin J Stereotac Funct Neurosurg 2004;6:348-50. Raisman J. Repair of spinal cord injury by auto transplantation of adult olfactory mucosal ensheating cells: translation from rat to clinical trials. Queen Square London: Spinal Repair Unit, Institute of Neurology; 2007. Hardy TL, Koch J. Computer assisted stereotactic surgery. Stereotac Funct Neurosurg 1982;45:396-8. Liu ZH. Latest development in stereotactic neurosurgical techniques. Chin J Neurosurg Dis Res 2003;2(1):1-3. Zha WG, Li AM, Fu XP. Stereotactic brachytherapy for cystic glioma in deep or eloquent brain areas. Chin J Minim Invasive Neurosurg 2006;11(10):474-7. Zhu FQ, LI AM, Xu F, Zhou YX, et al. CT guided stereotactic intracavitatory irradiation for intracranial cystic craniopharyngioma. Chin J Neurosurg 1998;14:142-4. Lin YC, Chen WJ, Zhu Ge QX. Combined CT- guided stereotactic interstitial radiotherapy of 32P and methotrexate chemotherapy for treatment of deep brain gliomas. Chin J Neurosurg 1996;12:216-8. Liu ZH, YU X, Guo Y, Cui YH, et al. The experimental and clinical preliminary studies of interstitial brachytherapy using iridium-192 on Glioblastomas. Chin J Neurosurg 1998;14:134-8. Tian ZM, Liu ZH, Li SY. Analysis of 1300 cases with CTguided stereotactic operations. Stereotac Funct Neurosurg 1998;70:147-8. Zhang X, Zhang J, Fei Z, et al. Neuronavigation-guided microsurgery for resection of brain tumors. Natl Med J China 2002;82(4):219-21. Zhao JZ, Cao Y, Lu Z, et al. Clinical evaluation of frameless stereotaxy in minimally invasive neurosurgery. Natl Med J China 2001;81:1042-5. Zhao YL, Wang CC, and Zhao JZ. Clinical use of navigation system in neurosurgical operations with 55 cases reports. Chin J Neurosurg 1998;14:274-6. Li GQ, Niu CS, Wang YH, et al. Functional MRI assisted neuronavigation guided surgery of intracranial tumors located in or near the motor cortex. Chin J Minim Invasive Neurosurg 2003;8(12):532-5. Jiang XF, Wang YH, et al. Neuronavigation assisted endoscopic procedures in neurosurgery. Chin J Stereotac Funct Neurosurg 2004;17(3):160-2.
45. Li GQ, Niu CS, et al. Application of ASA-610V neuronavigation system in minimally invasive neurosurgery. Chin J Contemp Neurol Neurosurg 2002;2(1):17-20. 46. Tian ZM, Wang TM, LI W, Chen MD. Location and mapping in robotic system for stereotactic neurosurgery. Chin J Biomed Engineering 2001;20(5):385-93. 47. Cooper IS. Chemopallidectomy. Science 1955;121:217. 48. Cooper IS, Bravo G, Riklan M, Davidson N, Gorek E. Chemopallidectomy and chemothalamectomy for parkinsonism. Geriatrics 1958;13:127-47. 49. Spiegel EA, Wycis HT: Stereoencephalotomy: thalamotomy and related procedures, vol. I–II. New York: Grune & Strattun; 1952-62. 50. Gao GD, Zhang H, Wang XQ, Liang QC. Indications of stereotactic and functional neurosurgery for Parkinson’s disease. Chin J Neurosurg 2002;18:12-4. 51. Backlund EO, Granberg PO, et al. Transplantation of adrenal medullary tissue to striatum in parkinsonism. First clinical trials. J Neurosurg 1985;62:169-73. 52. Madrazo I, Cuevas C, et al. Open microsurgical autograft of adrenal medulla to the right caudate nucleus in two patients with Parkinson’s disease. N Engl J Med 1987;316:831-4. 53. Zhang WC, Cao JK, et al. Neural transplantation with adrenal medullary tissue for tremors. Chin J Surg 1987;25 (11):650-2. 54. Wu CY, Zhou MD, Bao XF, et al. The combined method of transplantation of foetal substantia nigra and stereotactic thalamotomy for Parkinson’s disease. Br J Neurosurg 1994;8:709-16. 55. Jie Z, Ohye C, Shibazaki T. Gamma knife thalamotomy for the treatment of Parkinson’s disease and other movement disorders. Chin J Stereotac Funct Neurosurg 2005;18(6):354-7. 56. Wu SL, et al. Stereotactic neurosurgery for extrapyramidal disease. Chin J Nerv Ment Dis 1983;16(2):73. 57. Wu SL. Stereotactic surgery for tremors. Report of 507 cases. Chin J Stereotac Funct Neurosurg 1995;8(2):7-9. 58. Li YJ. Surgical treatment for movement disorders: a review of 1135 cases. Chin J Neurosurg 2001;6:350-3. 59. Bechtereva NP, Bondarchuk AN, Smirnov VM. Therapeutic electrostimulation of the deep brain structures. Vopr Neirokhir 1972;1:7-12. 60. Siegfried J, Lippitz B. Bilateral chronic electrostimulation of ventroposterolateral pallidum: a new therapeutic approach for alleviating all parkinsonian symptoms. Neurosurgery 1994;35:1126-9. 61. Benebid AL, Pollak P, Louveau A, Henry S, de Rougemont J. Combined (thalamotomy and stimulation) stereotactic surgery of the VIM thalamic nucleus for bilateral Parkinson’s disease. Appl Neurophysiol 1987;50:344-6. 62. Zhang JG, Wang CC, Zhang XY. Deep brain stimulation in idiopathic Parkinson’s disease. Chin J Neurosurg 2002;18(1):4-7.
History of stereotactic surgery in china
63. Sun BM, Liu KL, Lang LQ. Bilateral Deep brain stimulation of subthalamic nucleus in advanced Parkinson’s disease. Chin J Neurosurg 2002;18(1):8-11. 64. Zhang XH, Li YJ. Immediate and short term outcome after pallidotomy for intractable Tourette’s syndrome. Chin J Stereotactic Funct Neurosurg 2004;2:88-90. 65. Sun BM, Chen S, Zhan S. Subthalamic nucleus stimulation for primary dystonia and tardive dystonia. Acta Neurochir Suppl 2007;97(2):207-14. 66. Wang LX, Zhou XP, Hu XW, et al. MRI combining with micro electrode recording technology guided stereotactic surgery of Parkinson’s disease (270 cases). Chin J Stereotac Funct Neurosurg 2000;13(3):137-9. 67. Lee MK, Lee FC, Chee CP. The sunmed deep brain stimulation program for Parkinson’s disease: review after four years. XVII WFN World Congress on Parkinson’s disease and related disorders, Amsterdam. (Suppl); 2007. 68. Xu BT, Liang JC, Wang WM. Long term clinical efficacy of gamma knife radiosurgery for pituitary adenomas: a report of 487 cases. Chin J Minim Invasive Neurosurg 2006;11(6):244-6. 69. Wang LG, Guo Y, Zhang X. Linac X-knife stereotactic radio neurosurgery: 510 cases report. J Fourth Mil Med Univ 2000;21:1121-3. 70. Qi ST, Qiu BH, Yang KJ, Liu CY, Wang KW. Stereotactic radiosurgery treatment on focal area for intractable epilepsy. Chin J Stereotac Funct Neurosurg 2002;15:213-6. 71. Luan GM, Li YL. Epilepsy surgery in China: the history and current development. Neurology Asia 2007;12 Suppl 2:1-3. 72. Tan QF. Current status of epilepsy surgery in China. Chin J Stereotac Funct Neurosurg 2005;18(1):60-2. 73. Krynauw RA. Infantile hemiplegia treated by removing one cerebral hemisphere. J Neurol Neurosurg Psychiat 1950;13:243. 74. Shi YQ. Hemispherectomy for the treatment of infantile hemiplegia. Chin J Nerv Ment Dis 1959;1:48-51. 75. Zhao YD, Yang ZL, Tan YL. Surgical treatment of epilepsy. Chin J Neurol Psychiat 1965;9(4):325-9. 76. Kuang YQ, GU JW, et al. Combination of lesionectomy with multiple subpial transaction (MST) for intractable Epilepsy: 52 cases (Translation). Chin J Stereotac Funct Neurosurg 2004;02. 77. Morrell F, Whisler WW, Bleck TP. Multiple subpial transaction: a new approach to the surgical treatment of focal epilepsy. J Neurosurg 1989;70:231-9. 78. Yuan SB, Zhang J, Chen MZ, Zeng FJ, Chen LG. Multiple subpial transaction for intractable functional cortex epilepsy. Stereotac Funct Neurosurg 1998;70:92. 79. Xu JP, Wu HX. Amygdalotomy in temporal epilepsy: 10 case report. Chin J Nerv Ment Dis 1984;10:26-8. 80. Jinnai D, Mukawa J. Forel-H-tomy for the treatment of epilepsy. Confin Neurol 1970;32(2):307-15. 81. Mukawa J, Iwata Y, Kobayashi K. Forel-H field lesion effect on subcorticogenic and limbicogenic
82.
83.
84.
85.
86.
87.
88. 89.
90.
91.
92.
93.
94.
95.
96.
97. 98.
11
seizures- experimental studies. Neurologia Medicochirogica 1966;8:282-3. Xiao AP, Chang Y, Chen YY, et al. Stereotactic therapy for epilepsy. Chin J Stereotac Funct Neurosurg 1994;7: 21-22. Jiang CC, Jiang DJ, Yin SJ. Preliminary report of stereotactic callosotomy and/or amygdale nucleus for intractable epilepsy. Chin J Neurosurg 1989;5:260-2. Tang YL, Zhang ZX. The therapy of intractable epilepsy with stereotactic methods: a report of 82 cases. Chin J Neurosurg 1990;6:305-7. Wang GS, Xu JW, et al. The use of depth electrode in hippocampal EEG monitoring for localizing the epileptogenic zone in temporal lobe surgery. Chin J Stereotac Funct Neurosurg 2004;17(5):271-3. Shi YW, Mu J, Shi JJ, Gu F. Role of intraoperative electrocorticography in epilepsy surgery (Translation). Chin J Stereotac Funct Neurosurg 2007;20(3):165-6. Li TD, Bai HM, Jiang XX. Resection of brain tumors with epilepsy by electrocorticogram. Chin J Stereotac Funct Neurosurg 2004;17(2):85-7. Villemure JG, Rasmussen T. Functional hemispherectomy: methodology. J Epilepsy 1990;3 Suppl:177-82. Tan QF, et al. Long term results of functional hemispherectomy for intractable seizure. Chin J Stereotac Funct Neurosurg 2000;75:90-5. Jin LR, Wu LW, Gao W, et al. Ictal video-EEG monitoring in presurgical evaluation for intractable temporal epilepsy in 15 cases. Chin J Neurol 2002;35(3):132-4. Liang SL, LI AM, et al. Usefulness of 128 channels preoperative prolonged video-EEG in locating epileptogenic zone. Chin J Stereotac Funct Neurosurg 2003;16 (4):206-9. Wang HM, Zhang GP, LI L, et al. Clinical application of interictal SPECT in epilepsy surgery. Chin J Stereotac Funct Neurosurg 2004;17(5):268-70. Zhu D, Li L, Hua G, Guo J. Neuronavigation with magnetoencephalography for epilepsy surgery. Chin J Stereotac Funct Neurosurg 2005;18(3):155-7. Zhang GJ, LI YJ, et al. Application of long term intracranial EEG monitoring in epilepsy surgery. Chin J Neurosurg 2005;8:8-11. Li YL, Luan GM. A new method to treat functional intractable epilepsy: Introduction of bipolar electrocoagulation technique. Chin J Stereotac Funct Neurosurg 2002;3:265-8. Engel J, Jr, Van Ness PC, Rasmussen TB, Ojemann LM. Outcome with respect to epileptic seizures. In: Engel J, Jr, Surgical treatment of the epilepsies. New York: Raven Press; 1993. p. 609-621. Tan QF. The stereotactic radiosurgery in epilepsy. Modern Rehabilitation 2001;1:105. Yuan SB, Wu W, Liang X, et al. The rotating gamma knife for medial temporal lobe epilepsy (report of 38 cases). Chin J Stereotactic Funct Neurosurg 2007;20(1):10-14.
151
152
11
History of stereotactic surgery in china
99. Tan QF. Epilepsy surgery (Chinese). Nanjing University Press; 1995. 100. Moniz E. Prefrontal leucotomy in the treatment of mental disorders. Am J Psychiatry 1937;93:1379-85. 101. Lichterman BL. History of psychosurgery in Russia. Acta Neurochirugica 1993;125:1-4. 102. Binder DK, Iskandar BJ. Modern neurosurgery for psychiatric disorders. Neurosurgery 2000;47:9-23. 103. Wu SL. Bilateral stereotactic anterior cingulotomy for schizophrenia: a report of 23 cases. Chin J Neurosurg 1988;4:83-6. 104. Ballantine HT, Jr. Historical overview of psychosurgery and its problems. Acta Neurochir Suppl (Wien) 1988;44:125-12. 105. Feldman RP, Goodrich JT. Psychosurgery: a history overview [legacies]. Neurosurgery 2001;48(3):647-59. 106. Wang XF, et al. The clinical analysis of intractable psychosis treated with stereotactic techniques: report of 138 cases. Chin J Stereotac Funct Neurosurg 2003;16(4): 199-202. 107. Sun BM, Guan Y, Lang L, et al. Capsulotomy induces localized orbitofrontal and subcortical metabolic changes in obsessive compulsive disorder (abstract). Am Assoc Neurol Surg 2001;4 Toronto. 108. Wang XL, He SM, Li J, et al. Comparative study of reasons and influential factors on relapse after abstinence. Chin J Min Invas Neurosurg 2006;8:11. 109. Gao GD, Wang XL, He SM, et al. Clinical study for alleviating opiate drug dependence by a method of ablating the nucleus accumbens with stereotactic surgery. Stereotac Funct Neurosurg 2003;81:96-104. 110. Wang KW, Qi ST, Yang KJ. Medial septal diagonal band complex: A new target point of stereotaxic surgery for psychic dependence to opiods (report of 34 cases). Chin J Minim Invasive Neurosurg 2006;11(3):112-5. 111. Xu JW, Wang GS, Zhou HY. Neurosurgical treatment on alleviating heroin psychological dependence. Chin J Neurosurg 2005;21(10):19-22. 112. Ballantine HT, Jr, Cassidy WL, Flanagan NB, et al. Stereotaxic anterior cingulotomy for neuropsychiatric illness and intractable pain. J Neurosurg 1967; 26:488-95. 113. Wilkinson HA, Davidson KM, Davidson RI. Bilateral anterior cingulotomy for chronic non cancer pain. Neurosurgery 1999;45:1129-36. 114. Richardson DE, Akil H. Pain reduction by electrical brain stimulation in man.1. Acute administration in periaqueductal and periventricular sites. J Neurosurg 1977;47:178-83. 115. Young RF, Rinaldi PC. Brain stimulation in pain. In Levy RM, North RB, editors. The neurosurgery of chronic pain. New York: Springer 2003. 116. Tsubokawa T, Katayama Y, Yamamoto T, Hirayama T, Koyama S. Chronic motor cortex stimulation for the
117.
118.
119.
120.
121.
122.
123. 124.
125.
126.
127.
128.
129.
130.
131. 132.
treatment of central pain. Acta Neurochir Suppl (Wien) 1991;52:137-9. Zhao CS, Luo H, Sun BM. Chronic electrical motor cortex stimulation for intractable central pain. Chin J Stereotac Funct Neurosurg 2004;17(4):345-7. Wu CY, Meng FG, Wang HW, et al. Selective percutaneous radiofrequency thermocoagulation for treatment of trigeminal neuralgia: 1860 cases report. Chin J Neurosurg 2004;20:55-8. Pan MS, Ling ZP, Wang P, Zhen HM. The clinical study on gamma knife radio surgery for trigeminal neuralgia. Chin J Stereotac Funct Neurosurg 2004;17 (6):358-61. Yang FM, Yang SC, Lian YZ, Shu J, Zhao B. Treatment of central pain with stereotactic radiofrequency thermocoagulation. Acta Harbin Med Univ 1992;26:188-9. Jing GJ, Yang KY, et al. CT-guided brain stereotactic therapy of chronic cancer pain: report of 35 cases. Mod Rehab 2001;5:97-98. Jing GJ, Yang KY. Brain stereotaxic therapy for chronic cancer pain by thalamotomy (Chinese). J Sun Yat Sen Univ (Med Sci) 2002;23(2):13-7. Hu YS, LI YJ. Stereotactic neurosurgery for cancer pain. Chin J Pain Medicine 2005;11(4):197-200. Hu J, Yi LJ. The treatment of cancer pain with stereotactic radiotherapy. Chin J Pain Medicine 2002;8 (4): 195-7. Yang ZH, Luan GM, Zhang Y. The modal building and research of temporal lobe epilepsy. Chin J Neurosurg 2005;8:8-11. Wang Y, Duan DY, Xu QY. Survival, migration and differentiation of neural stem cells after transplantation into the brain. Chin J Biotechnology 2004;11: 13-17. Yu LH, Kang DZ, Lin JH. Orient migration of intravenous administration of bone marrow stromal cells after spinal cord injury in adult rats. Chin J Stereotac Funct Neurosueg 2006;19(1):17-20. Guo CL, Zhao AJ, Song J, Liu SJ. The effect of stereotactic limbic leucotomy lesion on obsessive compulsive disorder in rats. Chin J Stereotac Funct Neurosurg 2006;19(1):9-12. Hu ZY, Zhao XZ, et al. Design and application of soft ware for targeting thalamus and basal ganglia in stereotactic surgery (Chinese). Acta Academiae Medicinae Militaris Tertiae 2003;25(11):1019-20. Xu XN, Wang JH, Lu G, et al. Controlled release polymers of BCNU-PLA against experimental glioma in vivo. Chin J Stereotac Funct Neurosurg 2004;17(5): 280-3. Chen BH. Functional and stereotactic neurosurgery (Chinese). Inner Mongolia People’s Press; 1988. Chen BH. Stereotactic Radiosurgery (Chinese). Beijing Press; 1994.
History of stereotactic surgery in china
133. Anderson WS, Lenz FA. Surgery insight: DBS for movement’s disorders. Nature Clinical/Practice Neurol 2006;2:310-20. 134. Chou SN. Perspectives in international neurosurgery: a glimpse of neurosurgery in China. Neurosurgery 1978;3:120-2. 135. Gildenberg PL. Stereotactic surgery- the past and the future. Stereotac Funct Neurosurg 1998;70: 57-70. 136. Li P, Wu BJ, Nan Z, et al. Treatment of functional disorders with gamma knife radiosurgery. Stereotac Funct Neurosurg 1998;70:102-3. 137. Ma LT. Minimally Invasive Neurosurgery (Chinese). Wuhan: Huazhong Science Publisher; 1999. 138. Tan QF, Sun KH, Sun KJ. Results of surgical treatment in 76 patients with temporal lobe epilepsy. Stereotac Funct Neurosurg 1998;70:91.
11
139. Wang YH, Dong YJ, and XU JP. Stereotactic neurosurgery: past and present. Chin J Stereotac Funct Neurosurg 1986;1(1):46-52. 140. Wu CY, Wang YH, Meng FG, Liu YG. Development of stereotactic neurosurgery in China. Neurosurgery 2005;56:851-60. 141. Yi SY, Wu CY. Brain tissue transplantation (Chinese). People’s Hygiene Press; Beijing: 142. Zhang K, Zhang JG, Ma Y, et al. Subthalamic deep brain stimulation in the treatment of secondary dystonia. Chin Med J 2006;119(1):2069-74. 143. Zhang YQ, Li YJ, et al. Surgical treatment for adult onset dystonia: analysis of 16 cases. Chin J Min Invas Neurosurg 2006;11(2):121-5. 144. Tian ZM, Lu WS, Wang TM, et al. Application of a robotic telemanipulation system in stereotactic surgery. Stereotact Funct Neurosurg 2008;86:54-61.
153
9 History of Stereotactic Surgery in France A. L. Benabid . S. Chabardes . E. Seigneuret
Introduction The history of stereotaxy is part of a larger perspective of a methodological approach that is therapeutic and focuses on the search for precision. The search for precision implies the recognition and marking of targets (taxonomic version of the etymology of the word Stereotaxy, from taxis: order) as well as to the tactic act, which is the achievement, at least in human stereotaxy, of this approach (tactic version of the etymology of the word stereotaxy). The common denominator of these two definitions is the space (from the Greek Stereos: solid, volume), which by itself would define the methodology, based on the spatial coordinates of a point that will be called the target. This etymological duality corresponds in fact to a historical perspewctive, as for the first time, in 1917, the need to precisely recognize and localize the various spatial structures of the brain led Horsley and Clark [1] to design an instrumental method aimed at quantifying cerebral space, in order to attribute precise coordinates to the different structures that the anatomo physiologists were studying at that time. This taxonomy approach led to the development of an instrument and a method, and then to the elaboration of the concept of the surgical act based on exact location. This corresponded in fact to the tactic version of the etymological definition of stereotaxy, even if it has been, historically, only secondary. Having designed an apparatus and developed a method, one tended to build on these two concrete elements a philosophy, or at least a state of mind. This constitutes the most interesting, maybe the #
Springer-Verlag Berlin/Heidelberg 2009
most noble, part of the history of stereotaxy. This history has been the reflection of the complex interaction, depending upon circumstances, between the technical means of the moment and the therapeutic needs, as well as on pharmacological alternatives. This process is not specific to stereotaxy. It characterizes every approach of Homo Faber, which tries to solve his current problems, on the bases of the know-how which is available. He often stumbles on technological bottlenecks, which impede the development of methods, and even make it transitorily disappear, until the advance of knowledge in other domains provides the key, which will open the lock. A new momentum of development is therefore observed until the time when a new obstacle stops the process again, or when the need disappears, often because advancements of knowledge in other domains have brought more satisfactory solutions. Stereotaxy is at the crossroads of industrial technology, of surgical technology and therapeutic needs, themselves strongly enclosed in the domain of neurosciences, which, as we know, is undergoing rapid evolution. It is therefore not surprising to see that its history has been chaotic, and it can be foreseen that it will become even more complicated.
Innovation Comes from Necessity: History of the Development of the Stereotactic Frame In technological evolution, as in any kind of method, development is possible only if the need exists. The evolution of the need depends
98
9
History of stereotactic surgery in france
on the evolution of sciences and related disciplines. The need to discover the origin of an idea and identify its conceptual father depends on the value of this idea. There is a rich literature devoted to determining the paternity of stereotactic apparatuses, quite often in a chauvinist perspective. This need witnesses the importance of stereotaxy in the world of neurosciences, as it is easy to recognize its impact on our thinking and practice, in experimental neurophysiology as well as in clinics. This chapter could be included in this international competition, thanks to Professor Grellier in Nice who retrieved a document and was kind enough to give it to us. It proves that the stereotactic frame, and therefore the stereotactic method and practice, was invented in France at the end of the nineteenth century. In a paper published on 27 November 1897, in the review L’ Illustration [2], there is a very detailed description of it, in a scientific style which is not often used in public magazines (but it is probable that the culture of the public at that time was more scholarly than nowadays). The procedure was performed in a laboratory of the faculty of medicine in Paris, where the inventor had set up a system that could be anchored by pins to the skull of the patient presenting with projectiles (bullets or shrapnel pieces). This system featured a topological radiology modality oriented in space, consisting of two Crooke’s tubes and two photographic film supports mounted orthogonally. The radiographic antero-posterior and lateral projections of the head of the patients that were obtained made it possible to localize, as on an analytical geometry diagram, the presence of opaque foreign bodies, generally projectiles, even if they were not in teleradiological conditions. The figure, > Figure 9‐1 presented in this article of L’ Illustration is of remarkable precision and establishes without any doubt the feasibility of this approach, and of its precision, under the appearance of a frame, which could be very well compared to the frame of Horsley and Clark, or even to the Leksell frame, which is wellknown nowadays. The article reports that two interventions were successfully
performed, and reported in the Academy of Medicine by Marey (the physiologist). One cannot say whether this invention inaugurated either stereotaxy or interventional neuroradiology, but it is witnesses that the need to navigate within the brain was manifested a long time ago, whenever it was necessary to perform an intervention inside it. It seems that the encephalometer of Zernov (1889) [3], used in surgery by Altukhov 2 years later (1891) [4], aimed at the same goal of reaching intracranial structures, but without introducing in the approach a precise methodology of localization. It is more classical to quote at the origin of stereotaxy the publication in 1908 of Horsley, neurophysiologist and neurosurgeon, and Clark, mathematician, probably because their publication has given, in a scientific review known and read by their peers, the description of the apparatus, a stereotactic atlas of the monkey brain, and the method of electrolytic lesion by constant current. Ten years later, Mussen, the engineer who had worked on this frame, manufactured a version that he had designed, and then stored in his attic, wrapped in a journal, the headline of which helped to date this invention [5,6]. It was only in 1947 that Spiegel and Wycis [7] reported the use of an apparatus that allowed intraoperative X-rays, and coined the term stereoencephalotomy. This new approach allowed for the first time the mortality after operations to drop to about 2%. Ever since, the evolution of the stereotactic frames has not stopped, following trends and fashions, which accelerated its developments, and its adaptations to new indications. Two main categories of frames were developed, each favoring a modality of approach to the target. The Goniometric frames, putting the target at the pre-established center of an arch, are headed by the Leksell frame, followed by those of Richiert and Mundinger, and Todd and Wells, which finally produced the frames of Brent Robert Wells (BRW), and Cosman Robert Wells (CRW). The Cartesian frames on the contrary have developed orthogonal approaches from the lateral and the anteroposterior sides of the head, such as
History of stereotactic surgery in france
9
. Figure 9‐1 Upper right drawing: corresponds to the upper left photograph. The system, is made of:A frame fixed on the head by a cast. Three rods are firmly pressed on the forehead and on the cheek bones–Two photographic films will be successively placed on one side–On the other side two Crookes tubes (X-ray) which can be oriented–X-ray pictures are taken with one of the Crookes tubes. This is repeated with the other tube and another film–The system is removed from the head, the extremities of the three rods are tattooed on the skin, and the spatial position of the bullet is obtained at the intersection of two threads used to join the focus of the tubes to the images on the films, with respect to the three skin fiducials. Lower right drawing: corresponds to the lower left photographThe extremities A, B, C and D of four sliding rods which are adjusted such as to touch the skin fiducials (or face markers) for A, B, and C, and to the target/bullet for D. The length of the D rod represents the depth of the target/bullet and will guide the surgeon during the intracerebral procedure–This “Operating Compass” can be transported to the operating theatre, in the hospital or at the patient’s house. When it is applied on the face markers, the rod D points the direction and shows the distance to the target
99
100
9
History of stereotactic surgery in france
the Talairach frame, as well as those of OlivierBertrand, and of Hitchcock. Designs to adapt the goniometric approaches to the Talairach system have been made for individual users [8].
Introduction in France of the State of the Art of the Stereotactic Method After the reports of Remy and Contremoulins in 1897, the next mention of stereotaxy in France is found in the early 1950s about movement disorders (treated by ansotomy, pallidotomy and thalamotomy), [9–13], psychosurgery by lobotomy [14–16] and pain [17–24], using the conventional frames, methods and targets available in the western world. The French neurosurgeons practiced functional neurosurgery as a part of their general activity, showing special interest in these methods.. Some were especially attracted by the field and developed their own methods, such as ansotomy and campotomy for movement disorders [25–26], their frames [27–29] and their anatomical charts and atlases [30–34]. Among them would emerge, in academic institutions and practices, some who would provide such important contributions that they would make a profound mark on French stereotaxy.
Visionary Inputs of Two Giants Against this background of random practice and approaches, two main schools emerged and set the rules which have not only changed French stereotaxy but have had a real worldwide impact. They were inspired and led by two giants: Gerard Guiot at Hospital Foch, in the Parisian suburb of Suresnes, and Jean Talairach at Hospital SanteAnne, in Paris, who have educated and trained most of the French functional neurosurgeons and influenced many of the foreign masters in this field.
The Recognition of the Benefits of Multidisciplinarity: G. Guiot and D. Albe-Fessard Besides his tremendous input in neurosurgery at many levels, including pioneering work in pituitary surgery, Gerard Guiot (1912–1988) extended to all aspects of his practice an obsessional search for elements to improve his practice and results. In his functional neurosurgical practice, from pain to movement disorders, he needed to use every means to improve the precision of localization. He introduced the electrophysiological method into stereotactic procedures with the help of neurophysiologists such as Denise AlbeFessard and G Arfel. The use of intraoperative micro-electrophysiological recording has incomparable value, and was strongly stressed by the French stereotactic school in the third quarter of the twentieth century [11–13,27,35–38]. Its use was validated when it was clear that there was a correlation between efficacy, lack of complications, and the precision of the placement of the lesion, which was then used as the therapeutic tool. However, the importance of intraoperative neurophysiology has further decreased, for several reasons. The interest in stereotaxy has decreased because of the advent of levodopa, the expertise required an equipment not being available for neurosurgeons. Electrophysiology also appeared to be obsolete, or useless, because the target became more visible with the new imaging methods. This wrongly suggested that it was sufficient to aim at the structure to reach its function. This is true for the Pallidum and for the Subthalamic Nucleus, but it is not the case for the subdivisions of the Thalamus and for new targets such as the Pedunculopontine Nucleus (PPN), which are not visible and where the functional site has to be discovered through electrophysiology. Despite the advances in modern imaging, there are distortions in MR images, and displacements of cerebral structures during surgery, particularly during the progression of the
History of stereotactic surgery in france
electrodes. The functional target which is related to the efficacy of chronic stimulation cannot be reached without additional investigations. The recent development and the new interest in functional neurosurgery, particularly for movement disorders, the need for precision and safety, and the possibility to mimic the therapeutic effects of chronic stimulation during surgery, have given back intraoperative electrophysiology its double role. As an exploratory method, it provides the characteristic signature of the neuronal set through the morphology of the action potentials, their discharge pattern, and the activity evoked by external stimuli (proprioceptive, auditory, visual). As a therapeutic method, it reproduces accurately the effects which would be observed over a long term and allows the functional recognition of the target. It has become part of the classical panel of methodology in neurosciences. The discovery during those explorations of the differential effects of stimulation at high and low frequency has led to the development of methods of chronic deep brain stimulation at high frequency [39–47]. This development would not have been possible, or at least been long delayed, if implantable stimulators did not already exist, having been developed for the treatment of pain [23,48,49]. There was a long and disappointing inefficacy during which it had slowed down. The development of new applications to the treatment of movement disorders has re-motivated the industry to redevelop flexible electrodes and programmable stimulators.
The Reconsideration of an Unsatisfactory Situation: Jean Talairach Jean Talairach (1911–2007) could be described as a psychiatrist believing in psychosurgery and feeling that surgery was too serious to be left to neurosurgeons. He redesigned a stereotactic frame and raised it to the level of a method, if not a philosophy;
9
he built his own stereotactic atlas. When psychosurgery went out of favor, he had the spirit to refocus his stereotactic doctrine, his skills and his tools, to new applications, thereby opening new avenues such as a totally new vision of the surgical treatment of epilepsy. When movement disorders also went out of the field of stereotaxy, after levodopa was introduced as a treatment, he shifted gears and used his methodology to create a new field.
Stereotactic Biopsies The fate of a method depends on the need for it. When an outside phenomenon, such as a pharmacological newcomer, occurs, which brings, at least for a given period, a satisfactory solution, the need for the method disappears and sends it to oblivion. At the same time, the understanding that some pathologies, such as brain tumors, are not adequately taken care of by classical surgery and could eventually respond correctly to general treatments, such as chemotherapy or radiotherapy, induces the application of this forgotten, or nearlyabandoned, method to a totally new field: this is what Jean Talairach and his associates did when they introduced stereotactic biopsy, at a time where movement disorders needed less surgery as levodopa had come and provided a very attractive treatment. The school led by Jean Talairach at Sainte Anne must take credit for the development of stereotactic biopsies, for multiple reasons [50–54]. The growing awareness in the middle of the twentieth century that brain tumor resection was not always efficient or useful and that it was sometimes sufficient and less invasive to treat brain tumors by radiotherapy or chemotherapy had stressed the necessity to obtain the pathological diagnosis before applying those nonsurgical treatments. In addition to providing the right diagnosis, and sometimes to errors being corrected, multiple staged stereotactic biopsies allowed determining the extent of the lesion, particularly at the time when
101
102
9
History of stereotactic surgery in france
computerized tomography did not exist. Currently, using modern imaging, this need for the determination of the extent has diminished, allowing a decrease in the number of samples to the minimal necessity to obtain the histological signature, for instance, taking advantage of the recent demonstration that oligodendroglioma are chemosensitive when they bear the genetic mutation in 1p19q.
Brachytherapy [55–61] The use of ionizing radiations, in all their forms, has always played an important role in the therapeutic function of stereotaxy. The Parisian school of Jean Talairach had very early taken advantage of the radionecrotic properties of isotopes to create lesions with a reasonably foreseeable size: using Yttrium 90Y, mainly beta emitter, it was possible, because of its reduced propagation, to create a lesion of about 5 mm in diameter, in the place where the seed had been deposited under radiological control. This was used to replace electrolytic lesions for functional neurosurgery. This had also allowed Jean Talairach to develop methods of pituitary freination by Gold 198Au in the treatment of diabetic retinopathy by pituitary stereotaxy, based on the empirical observation, following the occurrence of a Sheehan syndrome, of the amelioration of this severe complication of diabetes. These indications of major hormonal surgery were also responsible for the concept, design and realization of the pituitary stereotactic frame. These methods, which led to very satisfactory results, later disappeared, following the advent of various forms of hormonal suppressive chemotherapy, similarly efficient and easier to manipulate. In the 1970s appeared the need to have a more focal treatment of brain tumors. Jean Talairach, along with Gabor Szikla, using Iridium 192Ir, introduced stereotactic Curitherapy, (named in
France from the multiple Nobel Price winner Marie Curie), in parallel with the German school, particularly in Freibourg with Riechert and Mundinger. In the literature in English, particulary in the United States, the term brachytherapy has been coined to better express the short distance effects of this therapeutic approach. This whole field of interstitial irradiation was undertaken along with experimental approaches and careful postoperative control and follow-up, as well as extensive histological studies of the postmortem specimens, to understand the mechanisms and improve the methodology.
Radiosurgery with Osvaldo Betti [62] Similarly, the Swedish school, led by Lars Leksell, combined stereotaxy and external radiotherapy to deliver high ionizing radiation with the precise doze in a well-defined spot. They developed the Gamma knife, made of several cobalt 60 sources, with convergent collimated beams on a determined point in space within the stereotactic frame. Initially designed to be applied in neurosurgery for movement disorders, this method has reached its full development in the treatment of arterio-venous malformations, improving the surgical outcome and reducing morbidity. More recently, this radiosurgery (which creates a lesion by radionecrosis) has been applied to tumors of small diameters and particularly to multiple brain metastases. In order to make this radiosurgery more accessible without being obliged to purchase the expensive Gamma knife system, the Talairach school, through its Argentinian pupil, Osvaldo Betti, has designed another instrument for radiosurgery using a linear accelerator, which he named LINAC. The stereotactic principles are satisfied by a physical connection between the linear accelerator and a chair rocking in two dimensions and to which the stereotactic frame is
History of stereotactic surgery in france
attached. The system has been installed at Hospital Tenon in Paris and became the first radiosurgical set-up used in France.
Harvesting the Crop When the Seeds Have Been Spread: The Development of Epilepsy Surgery Through Deep EEG Recording, Extended to Brain Tumor Surgery Jean Talairach, Jean Bancaud, Gabor Szikla, and Claudio Munari created a huge momentum in Sainte Anne, a school like which there are few in the world, and which could be compared to what happened at McGill around Penfield and Jasper. Most of the French stereotactic and functional neurosurgeons came here to be trained, and returned to Lille, Marseille, Grenoble, Rennes, Toulouse and Bordeaux, and later to other academic centers. Those who were to become famous in other parts of the world also visited Talairach’s school, learnt from it and also shared their knowledge, expertise and ideas. They came from and returned to Japan, Spain, Argentina, USA and Canada. Talairach had been trained in the Neurosurgery Department of Sainte Anne Hospital, under the supervision of his mentor, M. David. Since 1946, he had been developing stereotactic approaches to functional neurosurgery for chronic pain, movement disorders, pituitary diseases, and unoperable tumors. His interest was especially focused on the stereotactic definition of deep brain structures and he built up a coordinate system based on the anterior commissure-posterior commissure axis, which allowed him to “normalize” anatomical data from several brains. This led to the publication, in 1957, of his first stereotactic atlas on deep brain nuclei. His second stereotactic atlas, devoted to telencephalic structures, would be published 10 years later while he was already working with J. Bancaud. J. Bancaud, a neurologist
9
and electroencephalographer, was a student of H. Fishgold, a radiologist who was working at La Salpe´trie`re Hospital in Paris. At the beginning of the 1950s, Fishgold’s group joined David’s group in Sainte-Anne Hospital, and this undoubtedly represented a turning point for both Talairach and Bancaud’s careers. They were then gaining experience at the same institution, where epilepsy surgery had began, and they benefited from the high scientific experience of P. Buser and M.D. Dell in neurophysiology, and of H. Hecaen and J. de Ajurriaguera in neuropsychology. Bancaud saw the potential of Talairach’s method to localize the sources of EEG discharges in three dimensional space, and he began working with him to develop applications of stereotactic functional exploration for presurgical evaluation of medically intractable epilepsies. An operating room dedicated to stereotactic neurosurgery was opened at Sainte-Anne Hospital in 1959, and the project to record epileptic seizures by means of stereotactically implanted intracerebral electrodes became a reality. The term StereoElectroEncephaloGraphy (SEEG) [63–68] would be coined in 1962 to describe this new method, which, for many reasons, was revolutionary in the field of epilepsy surgery: SEEG recordings allowed the study of spatiotemporal dynamics of seizure discharges with respect to their clinical features, with a high degree of anatomical precision. The technique of SEEG offered the possibility to dissociate the presurgical investigation phase from the surgical therapeutic act: SEEG recordings and electrical stimulations were carried out in the operating room under “acute” conditions; after removal of the intracerebral electrodes, resection surgery was planned, and performed as a second step. SEEG was directly derived from a new conceptual approach for studying partial epileptic seizures, which was based on the assumption that the chronological occurrence of ictal clinical signs reflects the spatio-temporal organization of the epileptic
103
104
9
History of stereotactic surgery in france
discharge within the brain. The same year, Talairach’s team was joined by G. Szikla (1928–1983), an emigrant from Hungary where he had worked with Prof Zoltan, and who devoted his career to anatomy and stereotaxy, and played a major role in the development of stereotactic angiography. More particularly, carefully looking at the close relationships of cortical vessels to the gyri and sulci, Szikla showed that a meticulous analysis of vascular trajectories allowed the extraction of gyral form and dimension, and provided constant landmarks for the interpretation of anatomical variablity. Thus, stereotactic angiography was not only a routine procedure for safe implantation of intracerebral electrodes, but also a valuable tool for deducing the actual location of electrode contacts, particularly before the availability of modern imaging methods. The selection of the structures to be explored was based on a very careful analysis of all the data – notably clinical – collected during the non-invasive presurgical investigations. Such an analysis led to one or several hypotheses concerning the site(s) of seizure onset and the pathways of preferential ictal spread. Electrodes were then placed accordingly, in a way that enabled interpolation of intracerebral EEG activity within the interelectrode space. Proceeding this way, Bancaud and Talairach could precisely study how symptoms accumulated as the epileptic discharge propagated into different cortical structures and, in turn, could confirm or inform their initial hypotheses. The high spatial and temporal resolution, coupled with a high power of localization that the SEEG procedures offered, rapidly provided a large amount of data that had immediate repercussions on the practice of epilepsy surgery. This was mainly oriented towards tailored brain resection, but other avenues were not neglected. Deep brain stimulation for epilepsy in the thalamus (unpublished data) was initiated, following Mondragon and Lamarche’s work [69] on monkeys at Sainte Anne in Talairach’s INSERM unit U97, and this was the beginning of the application of STN stimulation to intractable nonoperable epilepsies
[70,71], in the wake of deep brain stimulation for movement disorders.
The Evolution of the Stereotactic Treatment of Movement Disorders The relation between stereotaxy and the treatment of movement disorders is the most exemplary as their mutual developments have been constantly linked during the last 50 years, in North America as well as in Europe, and particularly in France. In 1950, pallidotomies and ansotomies were performed in France by Fenelon [25,26], Guiot [27] and Gros [28] and by Mazars [72–80] through coagulation of the deep brain target through procedures guided on intraoperative visual landmarks. However, this procedure lacked precision and reliability, and the stereotactic approach was the logical next step. This step was taken by Guiot in France in 1952 [11], and quickly followed by Hassler and Riechert in Germany [81], Leksell in Sweden [82], Gillingham in Scotland [83], and Cooper [84,85] in the US as well as in Japan [86–88]. The lack of medical treatments for Parkinson’s disease, and even more for essential tremor, did no’t provide an alternative to the surgical treatment, which at that time was essentially ablative (microcoagulation, thermolesion, alcohol injection, or isotope insertion). Despite recurrences and complications, and because of its overall efficiency, the treatment of movement disorder remained the main condition for stereotaxy, until the advent of the levodopa era in the late 1960s. Since then, in France as well as in most of the Western countries, the surgical treatment of movement disorders, and therefore stereotaxy, underwent a dramatic decline, a few thalamotomies and pallidotomies still being performed in a limited number of expert centers. When, on the contrary, it appeared that some forms of Parkinson’s disease did not respond to the drugs or were plagued by motor
History of stereotactic surgery in france
fluctuations and dyskinesias, there was a new need for surgery. However, complications were unacceptable at that time and the method was required to evolve towards new directions, where efficacy would result in negligible morbidity. This led eventually to new developments such as brain grafting or, more recently, high frequency deep brain stimulation. Stereotaxy, like other methods, has experienced this type of fluctuation in its history and has on each occasion progressed and acquired new skills and possibilities, far beyond what it was initially designed for modern stereotactic surgery. The observation that after the honeymoon, levodopa had side effects such as levodopa-induced dyskinesias, and dopamine agonists induced hallucinatory and psychotic side effects, re-created the need for surgery and for new approaches. The saga of the brain grafts [89–92] did not involve French stereotaxy until the late 1990s when Marc Peschanski’s team [93–110], under a consensual but unspoken nationwide agreement, was considered as the only French team to engage in this difficult and highly skilled adventure. Since that period, he has contributed significantly to the advancement of the field of neural grafts, necessarily associated with the gene therapy domain [93,95,107,110]. He has applied this approach to movement disorders such as Parkinson’s disease, but mainly to Huntington’s disease [93– 96,100,101,103,106], which currently has no treatment. This is a clear example that some very specific fields of research and their applications should be totally assigned to a group, avoiding a sterile spread of financial and technical support. It was the reintroduction of Leksell’s pallidotomy by his two pupils Laitinen and Hariz in 1992 [111], and the introduction in 1987 of high frequency stimulation (HFS) [39–41,44,112– 114], that reinstalled in France, stereotaxy as an armamentarium of movement disorders. Its flexibility, adaptability, reversibility, and low morbidity, which help in performing HFS bilaterally,
9
have justified the quick and large development of this method in France, where the functional neurosurgery community and the neurologists, forming multidisciplinary teams, convinced the government to provide full reimbursement to this method. This allowed the development of several academic centers (up to 17 in 2007), and the extension to other movement disorders such as essential tremor and dystonia, to other targets than the Vim thalamus, such as the Pallidum and the subthalamic nucleus.
Stereotactic Surgery for Pain and the Development of Deep Brain and Cortical Stimulations Cancer pain was the most frequent indication of stereotaxy, particularly in the 1950s [115]. In the 1960s, Gabriel Mazars, also at Sainte Anne hospital in Paris, elaborated on the neurophysiological bases which were different from the gate control theory, a model of sensory perception and pain generation, on the basis of which he aimed at the thalamus and implanted electrodes [17–24], which were sometimes connected to an implantable stimulator, which was made on his ideas and then commercialized by the ELA company. At the same time in France several teams, sometimes in cooperation with other European countries [116], performed thalamic stimulation or periacqueductal gray (PAG) stimulation. Following the reports of the gate control theory by Wall and Melzack (1960), of the positive effects of thalamic stimulation for deafferentation pain and of the PAG for nociceptive pain, pain surgery also took advantage of the fundamental work of the INSERM team of Jean Marie Besson and Gise`le Guilbaud, which was close to Sainte Anne hospital. Quite often the practitioners of stereotactic surgery were very active in the field of functional neurosurgery as a whole and did not dissociate their activity from other methods such as spinal cord stimulation, for instance.
105
106
9
History of stereotactic surgery in france
This practice of deep brain stimulation for pain occurred in parallel to the large practice of spinal cord stimulation, which is not truly stereotactic, but part of functional neurosurgery. One must also mention Sedan in Marseille, Lazorthes in Toulouse, and Blond in Lille, who were very active in spinal cord stimulation as well as in intrathecal morphine, in strong correlation with the neurophysiologists. This created a real pain culture, which had a strong influence on the French stereotactic community, by calling for a more scientific approach, a critical evaluation of the results, and an ethical look at the overall field. In this stream of activity to try to treat pain, the work of Tsubokawa and his first reports on the efficiency of motor cortex stimulation for various types of pain, such as postherpetic facial pain or post-stroke (mainly thalamic) pain [117], attracted the attention of the teams led by Yves Keravel in Creteil Mondor Hospital [118], Marc Sindou in Lyon [119] and Yves Lazorthes in Toulouse [120].
The Introduction of Robotics and the Emergence of Neuronavigation in Stereotactic Surgery The quest for precision, reliability and safety never ends and takes advantage of all the available means. The logical extention of stereotaxy, increasingly based on numerical data provided by modern imaging, is to use these digital data to visualize targets on the MRI and CTscan modalities, to provide the possibility to draw trajectories and to plan procedures: this is what neuronavigation has progressively achieved in parallel to the increasing power of computers. The existence in Grenoble of industrial developments in the field of robotized multipurpose tools has naturally led to the concept, design and construction of a robotized stereotactic arm (Neuromate1 Schaerer-Mayfield, see chapter 37 in this
Handbook) [121] following the approaches of Kelly (a former fellow of J Talairach) [122] and of Ronald Young [123], aimed at replacing the grids, arch and goniometers of the stereotactic frame (ref Robots). In the same line of thought, a robotized microscope holder (Surgiscope1, ISIS) has been designed and built. These systems have been installed in several centers and are being currently routinely used and improved upon.
The Development of New Applications of Deep Brain Stimulation at High Frequency and the Rebirth of Psychosurgery in France New Applications of HFS Since the success of HFS in the treatment of Parkinson’s diseases, it has been extended to other targets (Vim, Pallidum, STN, CM-Pf) and also to other movement disorders (essential tremor, dystonia, tardive dystonia), based on the example of STN, which was derived from basic science. Other teams, in Italy, Belgium, and the Netherlands have successfully extended it to cluster headaches, obsessive compulsive disorders (OCD), and Tics de Gilles de la Tourette. In France, the fortuitous observation of the improvement by STN HFS of OCD symptoms in parkinsonian patients [124], led to the initiation of a multicenter clinical trial which has been successful in establishing the limbic portion of STN as a specific target for OCD treatments. This has been possible after the approval of the National Ethics Committee, an appeal to which was filed in December 2000 and which gave conditional approval in June 2002, for a carefully controlled clinical multicentric trial, similar to what happened in movement disorders, due to the growing interest in new indications. Psychosurgery had been almost stopped since the early 1970s in France, as a consequence of
History of stereotactic surgery in france
international pressure in the face of the abuses and complications, mainly lobotomies, even when stereotactic leucotomies were considered. There was in France very reduced psychosurgical stereotaxic activity in parallel to the development of the biological psychiatry and psychotherapy. There is now a new trend to return to psychosurgery for Tics and for depression, but the French neurosurgical community is very careful about the risks of jeopardizing one more time the possibility to help mental disorders if mistakes are made again.
Coma and Minimally Conscious States Franc¸ois Cohadon (1993) in Bordeaux, took a very careful and cautious approach to the problem of persisting vegetative post-traumatic states [125,126], at the same time as Tsubokawa [127,128] in Japan. In 25 cases of post traumatic vegetative state persisting 3 months after the initial injury, deep brain stimulation at low frequency has been used to activate the cortex with the hope of producing some degree of functional recovery. Electrodes were stereotactically implanted in the centrum medianum-parafascicularis complex. Bipolar stimulation was provided daily from 8 a.m. to 8 p.m. In 12 cases no changes occurred in the clinical features and overall behavior. DBS was given up after 2 months. All these patients with a follow-up of 1–10 years, remained in a permanent vegetative state, four of them eventually died. In 13 cases, following 1–3 weeks of DBS, a definite improvement was obtained with the recovery of some degree of consciousness and interpersonal relationship. Although some degree of long-term spontaneous recovery was documented, it was concluded that there was no significant effect. The reason why Cohadon, as well as Tubokawa, did not observe the results reported by Schiff et al. might be due to the fact that their patients were truly in vegetative states and not minimally conscious.
9
Conclusion French stereotaxy has reached a very internationally competitive level. Although one might consider that the first stereotactic frame was invented and used in 1897, stereotaxy seemed to start in the 1950s. Through two giant pioneers, it introduced multidisciplinarity as a philosophy. The group of Sante Anne created a whole monument of stereotactic discipline, which was wisely applied to several fields and created several approaches and strategies, developed atlases and frames, and inspired many young surgeons and students. French stereotaxy has remained very innovative since then. This was possible because stereotactic functional neurosurgery was and remained almost exclusively academic, was multidisciplinary with the primary goal of excellence, strong ethical concern and patient’s care, free from financial goals and economic considerations, as the French social security system did not demand economic criteria (length of stay, cost-benefit).
References 1. Horsley V, Clarke RH. The structure and functions of the cerebellum examined by a new method. Brain 1908;31:45–124. 2. Remy M, Contremoulins M. Le chercheur de projectiles, L’Illustration1897;55423 3. Zernov DN. Encephalometer: device for estimation of parts of brain in human. Proc Soc Physicomed Moscow Univ 1889;2:70-80. 4. Altukhov NV. Encephalometric investigations of the brain relative to the age, sex and skull indexes. Moscow; 1891. 5. Picard C, Olivier A, Bertrand G. The first human stereotaxic apparatus: The contribution of Aubrey Mussen to the field of stereotaxis. J Neurosurg 1983;59:673-6. 6. Olivier A, Bertrand G, Picard C. Discovery of the first human stereotactic instrument. Appl Neurophysiol 1983;46:84-91. 7. Spiegel EA, Wycis HT, Marks M, et al. Stereotactic apparatus for operations on the human brain. Science 1947;106:349-50.
107
108
9
History of stereotactic surgery in france
8. Sedan R, Duparet R. Stere´ometre adaptable au cadre ste´re´otaxique de J Talairach. Neurochirurgie 1968;14: 577-82. 9. Fenelon F, Thiebaut F. Re´sultats du traitement neurochirurgical d’une rigidite´ parkinsonienne par intervention striopallidale unilate´rale. Rev Neurol 1950;83:280. 10. Fenelon F. Essai de traitement neurochirurgical du syndrome parkinsonien par intervention directe sur les voies extrapyramidales imme´diatement sous-striopallidales (anses lenticulaires). Rev Neurol 1950;83:580-5. 11. Guiot G Brion S.“[Neurosurgery of choreoathetosic and Parkinsonian syndromes].” Sem Hop 1952;28 (49):2095-9. 12. Guiot G, Hardy J, Albe-Fessard D. De´limitation pre´cise des structures sous-corticales et identification de noyaux thalamiques chez l’homme par l’e´lectrophysiologie ste´re´otaxique. Neurochirurgia 1962;5:1-18 13. Guiot G, Derome P, Trigo JC. Le tremblement d’attitude: indication la meilleure de la chirurgie ste´reotaxique. Presse Med 1967;75:2513-8 14. Petit-Dutaillis, D., MessimyR, et al. “[Effect of prefrontal leucotomy on the behaviour of a maniac].” Rev Neurol (Paris) 1950;82(1):57-60. 15. Gros C,Vlahovitch B.“[Hemispherectomies; neurophysiological considerations].” Rev Neurol (Paris) 1951;85 (6):482-4. 16. Gros C, Vlahovitch B, et al. “[Effects of prefrontal interventions (leukotomy, topectomy) on psychic and kinetic manifestations of Parkinson syndrome].” Sem Hop 1955;31(32):1831-2. 17. Mazars G, Roge´ R, Mazars Y. Re´sultats de la stimulation du faisceau spino-thalamique et leur incidence sur la physiopathologie de la douleur. Revue Neurol 1960;20:136-8. 18. Mazars G, Merienne L, Cioloca G. Stimulations thalamiques intermittentes antalgiques. Revue Neurol 1973;128(4):273-9 19. Mazars G, Merienne L, et al. “[Use of thalamic stimulators in the treatment of various types of pain].” Ann Med Interne (Paris) 1975;126(12):869-71. 20. Mazars G. Intermittent stimulation of nucleus ventralis posterolateralis for intractable pain. Surg Neurol 1975;4:93-5. 21. Mazars G, Merienne L, et al. “[Present state of pain surgery].” Neurochirurgie 1976;22 Suppl 1:5-164. 22. Mazars G, Merienne L, et al. “[Effects of so-called periaqueductal gray substance stimulation].” Neurochirurgie 1979;25(2):96-100. 23. Mazars G, Merienne L, Cioloca G. Control of dyskinesias due to sensory deafferentation by means of thalamic stimulation. Acta Neurochirurgica Suppl (Wien) 1980;30:239-43. 24. Merienne L, Mazars G. Modification du sche´ma corporel sous stimulation thalamique. A propos des stimulations thalamiques pour membre fantoˆmes douloureux. Neurochirurgie 1981;27:121-3.
25. Fenelon F. “[Neurosurgery of parkinsonian syndrome by direct intervention on the extrapyramidal tracts immediately below the lenticular nucleus. Communication followed by film showing patient before and after intervention].” Rev Neurol (Paris) 1950;83(5):437-40. 26. Fenelon F. Bilan de quatre anne´es de pratique d’une intervention personnelle pour maladie de Parkinson. Rev Neurol 1953;89:580-5. 27. Guiot G, Brion S. Traitement des mouvements anormaux pour la coagulation pallidale. Technique et re´sultats. Rev Neurol 1953;89:578-80. 28. Gros C, Vlahovitch B, et al. “[the “Target” Zone in the Surgery of Parkinson’s Disease (Study of a Series of 407 Stereotaxic Operations)].” Neurochirurgie 1964;10:413-26. 29. Talairach J, Tournoux P.“[Apparatus for hypophysial stereotaxis by nasal approach].” Neurochirurgie 1955;1(1):127-31. 30. Szikla G, Talairach J.“Coordinates of the rolandic sulcus and topography of cortical and subcortical motor responses to low frequency stimulation in a proportional stereotaxic system.” Confin Neurol 1965;26(3):471-5. 31. Szikla G, Bouvicr G, Hori T, Petrov V. Angiography of the human brain cortex. New York: Springer; 1977. 32. Talairach J, David M, Tournoux P, et al. Atlas d’anatomie stereotaxique des noyaux gris centraux. Paris: Masson; 1957. 33. Talairach J, Szikla G, Tournoux P, et al. Atlas d’anatomie stereotaxique du telencephale. Paris: Masson; 1967. 34. Talairach J, Tournoux P. Coplanar stereotaxic atlas of the human brain: 3-dimensional proportional system: an approach to cerebral imaging. Stuttgart: Thieme Medical Publishers; 1988. 35. Albe-Fessard D, Arfel G, Guiot G, et al. De´rivation d’activite´s spontane´es et e´voque´es dans les structures ce´re´brales profondes de l’homme. Rev Neurol 1962;2:89-105. 36. Albe-Fessard D, Arfel G, Guiot G. Activite´s ´electriques caracte´ristiques de quelques structures ce´re´brales chez l’homme. Ann Chir 1963;17:1185-214. 37. Talairach J, Hecaen H, David M, et al. Recherches sur la coagulation the´rapeutique des structures sous-corticales chez l’homme. Rev Neurol 1949;81:4-24. 38. Talairach J, Tournoux P, Szikla G, et al. Traitement de la maladie de Parkinson: remarques me´thodologiques et technique chirurgicale. Gaz Med 1961;2:76-91. 39. Benabid A, Pollak P, Louveau A, et al. Combined (thalamotomy and stimulation) stereotactic surgery of the VIM thalamic nucleus for bilateral Parkinson’s disease. Appl Neurophysiol 1987;50:344-6. 40. Benabid A, Pollak P, Gervason C, et al. Long term suppression of tremor by chronic stimulation of the ventral intermediate thalamic nucleus. Lancet 1991;337:403-6. 41. Benabid A, Pollak P, Gao D, et al. Long term suppression of tremor by chronic stimulation of the ventral intermediate thalamic nucleus. J Neurosurg 1996;84:203-14. 42. Blond S, Caparros-Lefebvre D, Parker F. Control of tremor and involuntary movement disorders by chronic
History of stereotactic surgery in france
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54. 55.
56.
57. 58.
stereotactic stimulation of the ventral intermediate thalamic nucleus. J Neurosurg 1992;77:62-8. Caparros-Lefebvre D, Blond S, Vermersch P, et al. Chronic thalamic stimulation improves tremor and levodopa induceddyskinesias in Parkinson’s disease. J Neurol Neurosurg Psych 1993;56:268-73. Limousin P, Pollak P, Benazzouz A, et al. Effect on Parkinsonian signs and symptoms of bilateral subthalamic nucleus stimulation.Lancet 1995;345(8942):91-5. Siegfried J. Effets de la stimulation du noyau sensitif du thalamus sur les dyskine´sies et la spasticite´. Rev Neurol (Paris) 1986;142:380-3. Siegfried J, Lippitz B. Chronic electrical stimulation of the globus pallidus in Parkinson’s disease. Acta Neurochir 1993;124:14-18. Tasker RR. Effets sensitifs et moteurs de la stimulation thalamique chez l’homme. Applications cliniques. Rev Neurol (Paris) 1986;142:316-26. Adams J, Hosobuchi Y, Fields H. Stimulation of internal capsule for relief of chronic pain. J Neurosurg 1974;41:740-4. Hosobuchi Y, Adams J, Rutkins B. Chronic thalamic stimulation for the control of facial anesthesia dolorosa. Arch Neurol 1973;29:158-61. Talairach J, deAJ, et al. “[Therapeutic sub-cortical coagulations; topographic study of the ventricular system in relation to the central gray nuclei].” Presse Med 1950;58 (38):697-701. Talairach J, De Aljuriaguerra J, et al. “[A stereotaxic study of the deep encephalic structures in man; technic; physiopathologic and therapeutic significance.].” Presse Med 1952;60(28):605-9. Talairach J. “[Destruction of the anterior ventral thalamic nucleus in the treatment of mental diseases].” Rev Neurol (Paris) 1952;87(4):352-7. Talairach J, Tournoux P.“[Stereotaxic localization of central gray nuclei].” Neurochirurgia (Stuttg) 1958;1(1): 88-93. Talairach J, Tournoux P, et al. “[Parietal surgery of pain].” Acta Neurochir (Wien) 1960;8:153-250. Talairach J, Ruggiero G, et al. “A new method of treatment of inoperable brain tumours by stereotaxic implantation of radioactive gold; a preliminary report.” Br J Radiol 1955;28(326):62-74. Talairach J, Szikla G, et al. “[Stereotaxic destruction of the nontumoral hypophysis by radioactive isotopes. (Hypophysectomy for hormone-dependent cancer. Interstitial radiotherapy for malignant exphthalmos and Cushing’s syndrome of high origin)].” Presse Med 1962;70:1399-402. Talairach J, Szikla G, et al. “[Pituitary sterotaxic surgery].” Confin Neurol 1962;22:204-16. Talairach J, Bonis A, et al. “[Technic and results of interstitial radiotherapy of the pituitary in various endocrine syndromes: chromophobe adenoma, acromegaly and Cushing’s disease].” Acta Isot (Padova) 1964;4(4):355-85.
9
59. Talairach J, Szikla G.“[Amygdalo-hippocampal partial destruction by yttrium-90 in the treatment of certain epilepsies of rhinencephalic manifestation].” Neurochirurgie 1965;11(3):233-40. 60. Talairach J Szikla G.“[Intrasellar application of radioactive isotopes from a functional standpoint (mammary carcinoma, malignant, exophthalmos, Cushing’s syndrome, diabetes, malignant arterial hypertension)].” Nucl Med (Stuttg) 1965;:2:Suppl 371+. 61. Talairach J, Szikla G, et al. “Therapeutic utilization of radioactive isotopes in pituitary surgery.” Int J Neurol 1965;5(1):78-93. 62. Betti O, Derechinsky V.“[Multiple-beam stereotaxic irradiation].” Neurochirurgie 1983;29(4):295-8. 63. Bancaud J, Talairach J.“[Epilepsy of the supplementary motor area: a particularly difficult diagnosis in children].” Rev Neuropsychiatr Infant 1965;13(6):483-99. 64. Bancaud J, Talairach J.“[Methodology of stereo EEG exploration and surgical intervention in epilepsy].” Rev Otoneuroophtalmol 1973;45(4):315-28. 65. Talairach J, Bancaud J, et al. “[New approach to the neurosurgery of epilepsy. Stereotaxic methodology and therapeutic results. 1. Introduction and history].” Neurochirurgie; 19741:20 Suppl 1-240. 66. Talairach J, Szikla G.“Application of stereotactic concepts to the surgery of epilepsy.” Acta Neurochir Suppl (Wien) 1980;30:35-54. 67. Talairach J, Tournoux P, et al. “Stereotaxic exploration in frontal epilepsy.” Adv Neurol 1992;57:651-88. 68. Talairach J, Bancaud J, et al. “Surgical therapy for frontal epilepsies.” Adv Neurol 1992;57:707-32. 69. Mondragon S, Lamarche M.“Suppression of motor seizures after specific thalamotomy in chronic epileptic monkeys.” Epilepsy Res 1990;5(2):137-45. 70. Benabid AL, Minotti L, Koudsie A, de Saint Martin A, Hirsch E. Antiepileptic effect of high-frequency stimulation of the subthalamic nucleus (corpus luysi) in a case of medically intractable epilepsy caused by focal dysplasia: a 30-month follow-up: technical case report. Neurosurgery 2002;50(6):1385-91; discussion 1391-92. 71. Chabardes S, Kahane P, Minotti L, Koudsie A, Hirsch E, Benabid AL. Deep brain stimulation in epilepsy with particular reference to the subthalamic nucleus. Epileptic Disord 2002;4 Suppl 3:83-93. 72. MazarsMazars G. “[Surgical treatment of Parkinson’s disease].” Gaz Med Fr 1959;66(12):1063-4. 73. Mazars G, Droguet P, et al. “[Volitional dyskinesia of posture. Disappearance of the dyskinesia after thalamic coagulation].” Rev Prat 1960;103:73-4. 74. Mazars G, Pansini A, et al. “[Homolateral responses caused by stimulation of the radiating crown and of the internal capsule].” Rev Prat 1960;103:134-6. 75. Mazars G, Pansini A, et al. “[Contribution to the knowledge of the relations between the lesions produced by surgery of dyskinesias and the principal motor tract].” Rev Prat 1961;104:433-6.
109
110
9
History of stereotactic surgery in france
76. Mazars G, Chodkiewicz JP, et al. “[The cephalogyric contingent of the cortico-capsular fibers].” Rev Neurol (Paris) 1965;112(6):553-7. 77. Mazars G, Chodkiewicz JP, et al. “[Nosology and therapeutic indications in cervico-cephalic dyskinesias].” Rev Neurol (Paris) 1967;116(5):441-3. 78. Mazars G, Merienne L, et al. “[Surgery of cephalic orientation dyskinesia].” Neurochirurgie 1968;14(6):745-52. 79. Mazars G, Merienne L, et al. “[Dyskinesias of the upper limb associated with dyskinesias of cephalic orientation. Study of 45 cases and presentation of a film].” Rev Neurol (Paris) 1970;122(4):275-7. 80. Merienne L, Mazars G. Traitement de certaines dyskine´sies par stimulation thalamique intermittente. Neurochirurgie 1982;28:201-6. 81. Hassler R, Riechert T, Mundinger F, Umbach W, Ganglberger J. Physiological observations in stereotaxic operations inextrapyramidal motor disturbances. Brain 1960;83:337-51. 82. Svennilson E, Torvik A, Lowe R, et al. Treatment of parkinsonism by stereotactic thermolesions in the pallidal region. Acta Psychiatr Neurol Scand 1960;35:359-77. 83. Kelly PJ, Gillingham FJ. The long-term results of stereotacticsurgery and L-dopa therapy in patients with Parkinson’s disease. A 10-year follow-up study. J Neurosurg 1980;53:332-7. 84. Cooper IS, Bravo GI. Chemopallidectomy, and chemothalamectomy. J Neurosurg 1958;15:244-56. 85. Cooper IS. Surgical treatment of parkinsonism. Ann Rev Med 1965;16:309-30. 86. Narabayashi H, Okuma T. Procaine oil blocking of the globuspallidus for treatment of rigidity and tremor of parkinsonism:preliminary report. Proc Jpn Acad 1953;29:134. 87. Narabayashi H, Yokoshi F, Nakajiina Y. Levodopa induced dyskinesia and thalamotomy. J Neurol Neurosurg Psychiatr 1984;47:831-9. 88. Ohye C, Maeda T, Narabayashi H. Physiologically defined VIM nucleus: its special reference to control of tremor. AppliedNeurophysiol 1976;39:285-95. 89. Backlund E, Granberg P, Hamberger B, et al. Transplantation of adrenal medullary tissue to striatum in parkinsonism. First clinical trials. J Neurosurg 1985;62: 169-73. 90. Bjo¨rklund A, Stenevi U. Reconstruction of the nigrostriatal dopamine pathway by intracerebral nigral transplants. Brain Res 1979;177:555-60. 91. Bjo¨rklund A, Stenevi U, Dunnett S. Functional reactivation the deafferented neostriatum by nigral transplant. Nature 1981;289:497-9. 92. Bjorklund A, Dunnett SB, et al. “Neural transplantation for the treatment of Parkinson’s disease.” Lancet Neurol 2003;2(7):437-45. 93. Peschanski’s team Bachoud-Levi AC, Hantraye P, et al. “Prospectives for cell and gene therapy in Huntington’s disease.” Prog Brain Res 1998;117:511-24.
94. Bachoud-Levi AC, Remy P, et al. “Motor and cognitive improvements in patients with Huntington’s disease after neural transplantation.” Lancet 2000;356(9246):1975-9. 95. Bachoud-Levi AC, DeglonN, et al. “Neuroprotective gene therapy for Huntington’s disease using a polymer encapsulated BHK cell line engineered to secrete human CNTF.” Hum Gene Ther 2000;11(12):1723-9. 96. Bachoud-Levi AC, Gaura V, et al. “Effect of fetal neural transplants in patients with Huntington’s disease 6 years after surgery: a long-term follow-up study.” Lancet Neurol 2006;5(4):303-9. 97. Peschanski M, Rudin M, et al. “Magnetic resonance imaging of intracerebral neural grafts.” Prog Brain Res 1988;78:619-24. 98. Peschanski M, Defer G, et al. “Bilateral motor improvement and alteration of L-dopa effect in two patients with Parkinson’s disease following intrastriatal transplantation of foetal ventral mesencephalon.” Brain 1994;117 (Pt 3): 487-99. 99. Peschanski M. “The breaking of embargoes.” Lancet 2001;357(9260):963. 100. Peschanski M, Dunnett SB.“Cell therapy for Huntington’s disease, the next step forward.” Lancet Neurol 2002;1(2):81. 101. Peschanski M, Bachoud-Levi AC, et al. “Integrating fetal neural transplants into a therapeutic strategy: the example of Huntington’s disease.” Brain 2004;127 (Pt 6):1219-28. 102. Peschanski M. “[Stem cells, time for scale-up].” Med Sci (Paris) 2008;24(4):335-7. 103. Lefaucheur JP, Menard-Lefaucheur I, et al. “Electrophysiological deterioration over time in patients with Huntington’s disease.” Mov Disord 2006;21(9):1350-4. 104. Levivier M, Dethy S, et al. “Intracerebral transplantation of fetal ventral mesencephalon for patients with advanced Parkinson’s disease. Methodology and 6-month to 1-year follow-up in 3 patients.” Stereotact Funct Neurosurg 1997;69(1–4 Pt 2):99-111. 105. Mitjavila-Garcia MT, Simonin C, et al. “Embryonic stem cells: meeting the needs for cell therapy.” Adv Drug Deliv Rev 2005;57(13):1935-43. 106. Palfi S, Ferrante RJ, et al. “Chronic 3-nitropropionic acid treatment in baboons replicates the cognitive and motor deficits of Huntington’s disease.” J Neurosci 1996;16 (9):3019-25. 107. Peltekian E, Parrish E, et al. “Adenovirus-mediated gene transfer to the brain: methodological assessment.” J Neurosci Methods 1997;71(1):77-84. 108. Cesaro P, Peschanski M, et al. “Treatment of Parkinson’s disease by cell transplantation.” Funct Neurol 2001;16 (1):21-7. 109. Cochen V, Ribeiro MJ, et al. “Transplantation in Parkinson’s disease: PET changes correlate with the amount of grafted tissue.” Mov Disord 2003;18(8):928-32. 110. Bloch J, Bachoud-Levi AC, et al. “Neuroprotective gene therapy for Huntington’s disease, using
History of stereotactic surgery in france
111.
112.
113.
114.
115.
116.
117.
118.
polymer-encapsulated cells engineered to secrete human ciliary neurotrophic factor: results of a phase I study.” Hum Gene Ther 2004;15(10):968-75. Laitinen LV, Bergenheim AT, Hariz MI. Leksell’s posteroventral pallidotomy in the treatment of Parkinson’s disease. J Neurosurg 1992;76:53-61. Benabid AL, Pollak P, Gervason C, Hoffmann D, Gao DM, Hommel M, et al. Long-term suppression of tremor by chronic stimulation of the ventral intermediate thalamic nucleus. Lancet 1991;337(8738): 403-6. Limousin P, Krack P, Pollak P, Benazzouz A, Ardouin C, Hoffmann D, et al. Electrical stimulation of the subthalamic nucleus in advanced Parkinson’s disease. N Engl J Med;1998;339(16):1105-11. Krack P, Batir A, Van Blercom N, Chabardes S, Fraix V, Ardouin C, et al. Five years follow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinson’s disease. N Engl J Med 2003;349:1925-34. Gros C, Vlahovitch B, et al. “[Personal experience with surgery of pain in cancer].” J Radiol Electrol Arch Electr Medicale 1955;36(7–8):618-20. Blond S, Siegfried J.“Thalamic stimulation for the treatment of tremor and other movement disorders.” Acta Neurochir Suppl (Wien) 1991;52:109-11. Tsubokawa T, Katayama Y, et al. “Chronic motor cortex stimulation for the treatment of central pain.” Acta Neurochir Suppl (Wien) 1991;52:137-9. Nguyen JP, Lefaucher JP, Le Guerinel C, Eizenbaum JF, Nakano N, Carpentier A, et al. “Motor cortex stimulation in the treatment of central and neuropathic pain.” Arch Med Res 2000;31(3):263-5.
9
119. Garcia-Larrea L, Peyron R. Motor cortex stimulation for neuropathic pain: From phenomenology to mechanisms.” Neuroimage 2007;37 Suppl 1:S71-9. 120. Lazorthes Y, Sol JC, Fowo S, Roux FE, Verdie JC. “Motor cortex stimulation for neuropathic pain.” Acta Neurochir Suppl 2007;97(Pt 2):37-44. 121. Benabid AL, Hoffmann D, Lavallee S, Cinquin P, Demongeot J, Lebas JF, et al. Is there any future for robots in neurosurgery? In: SymonL, editors. Advances and Technical Standards in Neurosurgery, vol. 18. Springer: Wien New York; p. 3-45 122. Kelly PJ, Alker GJ. A method for stereotactic laser microsurgery in the treatment of deep seated CNS neoplasms. Appl Neurophysiol 1980;43:210-15. 123. Young RJ. Application of robotics to stereotactic neurosurgery. Neurol Res 1887;9:123-8. 124. Mallet L, Mesnage V, et al. “Compulsions, Parkinson’s disease, and stimulation.” Lancet 2002;360(9342):1302-4. 125. Cohadon F,Richer E. “[Post-traumatic vegetative states].” Neurochirurgie 1993;39(5):269-80. 126. Cohadon F, Richer E. “[Deep cerebral stimulation in patients with post-traumatic vegetative state. 25 cases].” Neurochirurgie 1993;39(5):281-92. 127. Tsubokawa T, Yamamoto T, et al. “Deep-brain stimulation in a persistent vegetative state: follow-up results and criteria for selection of candidates.” Brain Inj 1990;4 (4):315-27. 128. Yamamoto T, Katayama Y, et al. “Deep brain stimulation therapy for a persistent vegetative state.” Acta Neurochir Suppl 2002;79:79-82.
111
4 History of Stereotactic Surgery in Germany J. K. Krauss
Stereotactic and functional neurosurgery has a long and rich history in Germany [1,2]. At the time functional and stereotactic surgery was introduced in clinical practice, medical progress had slowed down considerably during the post-World War II depression era. In this article, a brief overview on the pre-war development of neurosurgery in Germany with a special emphasis on issues relevant to functional neurosurgery, followed by a more thorough review on the birth of functional neurosurgery in Freiburg and its further achievements are discussed. The account is based on original historical documents, previous historical reviews and a personal interview, and the reader is referred to those documents for a more complete bibliography on this topic [1,3–6].
Development of Neurosurgery in Germany Ernst von Bergmann was the first German surgeon who devoted his career to surgery of the nervous system [7]. He became professor of surgery in 1878, in Wu¨rzburg, and he moved to the Charite´ in Berlin, in 1882. He published several books focussing on neurosurgical topics. When he died in 1907, several of his pupils continued to carry on his interests. The two major driving forces for the further development ultimately were Fedor Krause and Otfrid Fo¨rster [8,9]. Fedor Krause developed his career first in Hamburg, and later in Berlin. In February, 1892, he explored the trigeminal nerve via an extradural approach for treatment of trigeminal #
Springer-Verlag Berlin/Heidelberg 2009
pain. He was active in many other aspects of neurosurgery and was highly esteemed when he died in 1937. Otfrid Fo¨rster, who was 18 years younger than Krause, was the first to pioneer the concept that neurosurgery should be an independent discipline in Germany. Fo¨rster was trained as a neurologist and a neurophysiologist, but he was rather an autodidact in surgery. He became well known during the 1920s and had contacts with Bailey, Bucy, Penfield, and Cushing. Fo¨rster was invited to Moscow after Lenin suffered from an apoplectic insult. When Lenin died, it was he who signed the protocol of the autopsy reporting hemorrhage from an aneurysm of the medial cerebral artery. Fo¨rster furthered many innovations in neurosurgery at his institute in Breslau. He was the first to pioneer dorsal rhizotomy for treatment of spasticity, and cordotomy for treatment of pain syndromes. In 1937, when the sky was clouded already by the reign of the Nazi regime and the international reputation of Germany became more and more overshadowed, he still managed to organize a German-British meeting of neurosurgeons in Berlin and became a honorary member of the British neurosurgical society [9]. He died in 1940 at the age of 67. It was Wilhelm To¨nnis who became the first chair of an ‘‘extraordinariate’’ in 1937, in Berlin [10]. Although he is mostly known for his work on trauma surgery, he established fruitful contacts with other neurodisciplines including new concepts such as neuropathology. Spatz and Hallervorden were the first to describe more extensively the disorder which carried their names for decades and was renamed recently in neurodegeneration with brain iron accumulation [11].
54
4
History of stereotactic surgery in germany
The development of neurosurgery in Germany came to a sudden halt when World War II began [10]. When the war was over, neurosurgery soon became an advanced discipline, during the reorganization of healthcare, not devoted mainly to trauma anymore.
Freiburg: the Early Epicentre of German Functional and Stereotactic Neurosurgery Traugott Riechert, born in 1905, was a pupil of To¨nnis who had worked on brain injury during the Second World War. He was the second in Germany to become professor of neurosurgery on June 27, 1946, in Freiburg. At that time, psychiatry and neurology in Freiburg were united under the leadership of the psychiatrist Behringer. Richard Jung, who later became the head of the department of neurology, and Rolf Hassler, who established a laboratory for neuropathology, were members of the faculty. Hassler had come to Freiburg from Berlin after the war was over. He was a pupil of the internationally renowned scientists, Cecile and Oscar Vogt, who had to leave Berlin when the nazis had taken over, to be replaced by Spatz. The Vogts were actually the first to forge the term ‘‘extrapyramidal system.’’ When they left Berlin to move to the Black Forest, they brought with them a large collection of brains (among them was Lenin’s brain) and brain slices which was one of the reason that Rolf Hassler was also attracted to come to Freiburg. When Riechert accepted the call to Freiburg in 1946, he had no rooms available in his department [6]. Therefore, he was housed initially in an attic of the psychiatry department. In the late 1947, the department of neurosurgery assumed its operative activities after Watts had visited Freiburg to demonstrate the technique of Freeman and Watts to perform transorbital leucotomy. This procedure was performed on a few patients in Freiburg and interestingly treated intractable pain also, but it
never became popular. Riechert made rounds in the psychiatry ward, and every Saturday there was a big conference attended by the neurologists and psychiatrists. The first stereotactic frame in Freiburg was developed during this period by Riechert together with the physicist, Wolff [12]. It was quite inaccurate, but it was the prototype for the second model which became to be known as the Riechert-Mundiger frame [13]. When Richard Jung became the interim chair of the department of psychiatry after the death of Behringer, he asked the residents whether one of them would like to join the neurosurgeons [9]. Fritz Mundinger volunteered, and he started his career in neurosurgery in 1951. At that time Hassler had become the head of the neuropathology laboratory. Mundinger first concentrated on the improvement of the accuracy of the stereotactic frame and every afternoon, if possible, he met with the technician F. L. Fischer of the Fischer Company to develop the device further into the shape it later became known and still exists today [13]. Functional stereotactic neurosurgery in Freiburg started with pallidotomies [14]. While it became recognized that rigidity responded quite well to pallidotomy, it was also noted that tremor was ameliorated to a much lesser extent. Also, with the early lesioning systems, hemorrhages were not infrequent and they were thought to be partially related to the vascular supply of the pallidum. By that time Hassler was already known as ‘‘the expert of the thalamus’’ worldwide. He had managed to establish his own nomenclature dividing the thalamus into 34 subnuclei based on their cytoarchitectonic structures and their cortical and subcortical connections [3]. His classification was already published in his habilitation thesis in 1948. It was he who suggested to target the thalamus based on his neuroanatomic concepts, and also to stay out of trouble with the branches of the anterior choroidal artery during pallidotomy. He argued that it would be useful not only to target the pallidothalamic system, but also the dentatothalamic system and their projections to
History of stereotactic surgery in germany
the cortex. He thought that the ventralis oralis anterior nucleus would be a good target for rigidity, and hence he suggested the ventralis oralis posterior nucleus as an ideal target for tremor. The basic problem, however, was to identify and find the nuclei. The time had come to introduce thalamotomy for treatment of movement disorders. The first thalamotomy ever was actually performed on November 14, 1952, in Freiburg [15]. The patient was a 38-year-old man with postencephalitic parkinsonism with prominent tremor. The target coordinates were determined by comparing the pneumencephalography X-rays, which were made several days prior to the operation with 5 mm brain slices that Hassler had prepared. The algorithm for target calculation taking into account cranio-cerebral correlations and X-ray distortion was quite complicated. The operation was performed by Mundinger in the presence of Riechert, Hassler, and von Baumgarten. The immediate result was that the tremor stopped and the response of both, the patient and the physicians, was overwhelming. The results were published two years later in 1954, by Hassler and Riechert, in German, in ‘‘Der Nervenarzt’’ [15]. Over the next few years, pallidotomy was replaced step-by-step by thalamotomy in Freiburg. Subsequently, Freiburg became the first major center for stereotactic and functional neurosurgery in Germany. In 1960, thousands of functional stereotactic procedures had been performed in the department [16,17]. Over the next few years, the department became extremely busy and up to five thalamotomies for Parkinson’s disease were performed each day. Also, the indications were expanded and other movement disorders such as dystonia, myoclonus, hemiballism, and tics were successfully treated by thalamotomies [18–22]. In the early 1960s, the zona incerta was introduced as a target for ablative surgery [23]. With the development of temperature control and radiofrequency lesioning, lesions could now be made much smaller and side effects were significantly
4
reduced [24,25]. The effects of both low frequency and high frequency stimulation with electric current via the radiofrequency probe were studied systematically as early as in the late 1950s [26,27]. Another early development in Freiburg was the use of radionuclides to treat brain tumors [6]. In 1951, the first radionuclide used was phosphor 32. Later the most frequently used radionuclides were iodine 125 and iridium 192. During the 1950s, stereotactic implantation of radionuclides was used in the treatment of movement disorders in 20 patients. In 1963, Mundinger developed the ‘‘Gammamed’’ for brachyradiotherapy [4]. This machine was constructed for high-dose radiation and it was loaded with iridium 192. Radiation therapy was delivered by stereotactic techniques.
Spread Over Germany While the department of neurosurgery in Freiburg grew and became larger, many visitors and several guests from Germany and other countries (> Figure 4-1) were trained there in functional stereotactic neurosurgery. Many of them spread the technique to other cities, were they worked in the frame of general neurosurgery or in specialized subunits. Some of them were: Dieckmann in Homburg, Umbach in Berlin, Spuler in Wu¨rzburg, Mu¨ller in Hamburg, Thomalske in Frankfurt, Nittner in Cologne, and Yasargil in Zurich [28]. Independent departments for stereotactic and functional neurosurgery were built in Freiburg, Homburg, Frankfurt, and Wu¨rzburg. Many of these neurosurgeons modified the stereotactic techniques they had learned in Freiburg and expanded the indications. Struppler, who became active over decades in Munich, actually was trained as a neurologist. When Germany was divided into two countries, there was little cross-border exchange due to the restrictions of the cold war. Nevertheless, Goldhahn managed to establish functional stereotactic neurosurgery in Leipzig,
55
56
4
History of stereotactic surgery in germany
. Figure 4-1 Mundlinger (second from left) and Riechert (third from left) at the entrance of the building which housed the neurosurgical department in Freiburg in the 1960s together with guests from Russia including Ugrumov (third from right)
East Germany (formerly communist ‘‘Deutsche Demokratische Republik’’) [29]. The first major atlas for stereotactic surgery of the human brain ‘‘Stereotaktische Operationen/ Stereotaxis’’ was edited by Georges Schaltenbrand from Wu¨rzburg and by Percival Bailey from Chicago [30]. It was published in three volumes by Georg Thieme Verlag, Stuttgart, in 1959. This atlas had a major impact on the further development of functional stereotactic surgery. Strong input into the atlas was supplied by Hassler, Riechert, and Mundinger, and its development was also supported by Leksell. Twenty years later its second edition ‘‘The Atlas for Stereotaxy of the Human Brain’’ was published, this time edited by Georges Schaltenbrand and Waldemar Wahren
[31]. While the first edition ‘‘The Schaltenbrand/ Bailey Atlas’’ was bilingual – German and English, the second edition ‘‘The Schaltenbrand/Wahren Atlas’’ was only in English. Tragically, after writing the foreword to the accompanying textbook edited by Schaltenbrand and Walker [32], Schaltenbrand suffered a stroke and died before the second edition of his monumental work went to press. While there was a strong focus on functional stereotactic neurosurgery for treatment of movement disorders in Germany, other classical indications like pain, psychoaffective disorders, and epilepsy were developed further. As indicated above, classical leucotomy did not have a wide distribution in Germany. Also, after the introduction of stereotactic techniques to target subcortical
History of stereotactic surgery in germany
4
structures for psychoaffective disorders, psychosurgery was rather limited and such procedures were confined to six centers [33]. As in other countries, psychosurgery came to a complete halt in the 1970s, when it came to the focus of public attention and provoked overwhelmingly unwanted responses in the media.
Charite in ‘‘East’’ Berlin [36]. The revival of movement disorders surgery in Germany was closely related to the introduction of deep brain stimulation [37]. The base for treatment of movement disorders is now widely distributed all over Germany with major centers in Freiburg, Cologne, Berlin, Magdeburg, Kiel and Hannover.
Silent Years and Revival
References
Birkmayer and Hornykiewicz had already published their observations on levodopa in parkinsonism in 1961, in a German journal [34]. After the general introduction of levodopa in clinical practice in the late 1960s, the number of functional stereotactic procedures for treatment of Parkinson’s disease sharply dropped, and over the years many centers completely stopped their activities. Nevertheless, some institutions still remained active. Even when thalamotomy had become completely out of public focus, about 100 thalamotomies were performed each year for treatment of tremor or Parkinson’s disease. When I joined Mundinger in the department of functional and stereotactic neurosurgery as a resident, in Freiburg, in 1988, I had the unique opportunity to learn this procedure. The academic interest in functional and stereotactic neurosurgery was always strong in Germany. During the 1960s and the 1970s networks with colleagues from other European countries became more and more apparent. As detailed in chapter A2 the founding meeting of the European Society of Stereotactic and Functional Neurosurgery was in Freiburg in 1970. While pallidotomy became a major procedure for treatment of movement disorders in the United States and in several European countries after Laitinen had raised it from oblivion [35], it became never really popular in Germany. Also, partially for legal reasons neurotransplantation never had a more widespread application in the 1980s and 1990s with the exception of some patients having been operated on by Vogel in the
1. Krauss JK, Grossmann RG. Historical review of pallidal surgery for treatment of parkinsonism and other movement disorders. In: Krauss JK, Grossmann RG, Jankovic J, editors. Pallidal surgery for the treatment of Parkinson’s disease and movement disorders. Philadelphia: LippincottRaven; 1998. p. 1-23. 2. Sturm V. Stereotaxie. In: Arnold H, et al., editors. Neurochirurgie in Deutschland: Geschichte und Gegenwart. Berlin, Wien: Blackwell; 2001. p. 254-7. 3. Hassler R, Mundinger F, Riechert T. Stereotaxis in Parkinson syndrome. Berlin, Heidelberg, New York: Springer; 1979. 4. Mundinger F. Stereotaktische Operationen am Gehirn. Stuttgart: Hippokrates; 1975. 5. Riechert T. Stereotactic brain operations. Bern, Stuttgart, Vienna: Huber; 1980. 6. Krauss JK. Interview with Professor Mundinger. AANS Archives Video Interview series: Leaders in Neuroscience; 1998. 7. Bushe KA, Collmann H. Neurochirurgie von den Anfa¨ngen bis zum spa¨ten 19. Jahrhundert. In: Arnold H, et al., editors. Neurochirurgie in Deutschland: Geschichte und Gegenwart. Berlin, Wien: Blackwell; 2001. p. 1-23. 8. Collmann H, Halves E, Arnold H. Neurochirurgie in Deutschland von 1880 bis 1932. In: Arnold H, et al., editors. Neurochirurgie in Deutschland: Geschichte und Gegenwart. Berlin, Wien: Blackwell; 2001. p. 25-77. 9. Frowein RA, Dietz H, Rosenow DE, Vitzthum HE. Neurochirurgie in Deutschland von 1932 bis 1945. In: Arnold, H, et al., editors. Neurochirurgie in Deutschland: Geschichte und Gegenwart. Berlin, Wien: Blackwell; 2001. p. 79-95. 10. Frowein RA, Dietz H, Franz K. Neurochirurgie in Deutschland von 1945 bis 1970, In: Arnold H, et al., editors. Neurochirurgie in Deutschland: Geschichte und Gegenwart. Berlin, Wien: Blackwell; 2001. p. 97-119. 11. Thomas M, Hayflick SJ, Jankovic J. Clinical heterogeneity of neurodegeneration with brain iron accumulation (Hallervorden-Spatz syndrome) and pantothenate kinase-associated neurodegeneration. Mov Disord 2004;19:36-42.
57
58
4
History of stereotactic surgery in germany
12. Riechert T, Wolff M. Die technische Durchfu¨hrung von gezielten Hirnoperationen. Arch Psychiat Zeitschr Neurol 1953;190:297-316. 13. Riechert T, Mundinger F. Beschreibung und Anwendung eines Zielgera¨tes fu¨r stereotaktische Hirnoperationen (II. Modell). Acta Neurochir 1956;3:308-37. 14. Mundinger F, Riechert T. Ergebnisse der stereotaktischen Hirnoperationen bei extrapyramidalen Bewegungssto¨rungen auf Grund postoperativer und Langzeituntersuchungen. Dtsch Ztschr Nervenheilk 1961;182:542-76. 15. Hassler R, Riechert T. Indikationen und Lokalisationsmethode der gezielten Hirnoperationen. Nervenarzt 1954;25:441-7. 16. Mundinger F, Riechert T. Die stereotaktischen Hirnoperationen zur Behandlung extrapyramidaler Bewegungssto¨rungen (Parkinsonismus und Hyperkinesen) und ihre Resultate. Fortschr Neurol Psych 1963;31:1-66, 69-120. 17. Mundinger F, Riechert T. Indikationen und Langzeitergebnisse von 1400 uni- und bilateralen stereotaktischen Eingriffen beim Parkinsonsyndrom. Wien Zschr Nervenhk 1966;23:147-77. 18. Riechert T. The stereotactic technique and its application in extrapyramidal hyperkinesia. Confin Neurol 1972;34:325-30. 19. Mundinger F, Riechert T, Disselhoff J. Long-term results of stereotactic treatment of spasmodic torticollis. Confin Neurol 1972;34:41-6. 20. Mundinger F, Riechert T, Disselhoff J. Long term results of stereotaxic operations on extrapyramidal hyperkinesia (excluding parkinsonism). Confin Neurol 1970;32:71-8. 21. Krauss JK, Mundinger F. Functional stereotactic surgery for hemiballism. J Neurosurg 1996;85:278-86. 22. Loher TJ, Pohle T, Krauss JK. Functional stereotactic neurosurgery for treatment of cervical dystonia: review of the experience from the lesional era. Stereotact Funct Neurosurg 2004;82:1-13. 23. Mundinger F. 30 Jahre stereotaktische Hirnoperationen beim Parkinsonismus (Ergebnisse im Vergleich pallidosubthalamischer Ausschaltungen und Indikationen). In: Ga¨nshirt H, Berlit P, Haack G, editors. Pathophysiologie, Klinik und Therapie des Parkinsonismus. Basel: Editiones Roche; 1983. p. 331-57.
24. Gross I. Klinische Vergleichsuntersuchungen zwischen der konventionellen und der frequenzreinen, temperaturkontrollierten Hochfrequenzkoagulation zur stereotaktischen subkortikalen Ausschaltung. Inaugural Dissertation. Albert-Ludwigs-Universita¨t, Freiburg, 1964. 25. Hassler R, Mundinger F, Riechert T. Correlations between clinical and autoptic findings in stereotactic operations of parkinsonism. Confin Neurol 1965;26:282-90. 26. Hassler R, Riechert T. Wirkungen der Reizungen und Koagulationen in den Stammganglien bei stereotaktischen Hirnoperationen. Nervenarzt 1961;32:97-109. 27. Hassler R, Riechert T, Mundinger F, Umbach W, Ganglberger JA. Physiological observations in stereotaxic operations in extrapyramidal motor disturbances. Brain 1960;83:337-54. 28. Krayenbu¨hl H, Wyss OAM, Yasargil MG. Bilateral thalamotomy and pallidotomy as treatment for bilateral parkinsonism. J Neurosurg 1961;18:429-44. 29. Goldhahn G, Goldhahn WE. Experience with stereotaxic brain surgery for spasmodic torticollis. Zentralbl Neurochir 1977;38:87-96. 30. Schaltenbrand G, Bailey P. Introduction to stereotaxis with an atlas of the human brain. Stuttgart: Thieme; 1959. 31. Schaltenbrand G, Wahren W. Atlas for stereotaxy of the human brain. Stuttgart: Thieme; 1977. 32. Schaltenbrand G, Walker EA. Stereotaxy of the human brain. Stuttgart, New York: Thieme; 1982. 33. Mu¨ller D. Psychiatrische Chirurgie (sog. Psychochirurgie). In: Arnold, H, et al., editors. Neurochirurgie in Deutschland: Geschichte und Gegenwart. Berlin, Wien: Blackwell; 2001. p. 258-66. 34. Birkmayer W, Hornykiewicz O. Der L-Dioxyphenylalanin (L-Dopa)-Effekt bei der Parkinson-Akinese. Wien Klin Wschr 1961;73:787-8. 35. Laitinen LV, Bergenheim AT, Hariz MI. Leksell’s posteroventral pallidotomy in the treatment of Parkinson’s disease. J Neurosurg 1992;76:53-61. 36. Vogel S. Mo¨glichkeiten und Grenzen der Neurotransplantation. In: Poewe W, Madeja DU, editors. Aktuelle Aspekte der Therapie des Parkinson-Syndroms. Karlsruhe: Braun; 1991. p. 73-81. 37. Krauss JK, Volkmann J: Tiefe Hirnstimulation. Darmstadt: Steinkopff; 2004.
8 History of Stereotactic Surgery in Great Britain E. A. C. Pereira . A. L. Green . D. Nandi . T. Z. Aziz
The Horsley–Clarke Apparatus Stereotaxis was the name given to a procedure and apparatus invented in London by the Englishmen, Horsley and Clarke. Before then, the German Dittmar had used a guided probe to transect rodent medulla in 1873 [1], and Zernov in Russia had described an encephalometer enabling brain surface localization in 1889 [2]. Neither technique enabled targeting with respect to a fixed threedimensional Cartesian coordinate system. Sir Victor Alexander Haden Horsley (1857–1916; > Figure 8-1a) was the first neurophysiologist who was also a neurosurgeon, pioneering a Great British tradition of such hybrid scholars particularly prevalent in the stereotactic comamunity. His uniquely deft approach to neurosurgery – derived from experiments upon over 100 primates – made a reputation such that ‘‘the Staff intended to have Horsley and nobody else’’ when tenure became available at the National Hospital for the Paralyzed and Epileptic in Queen Square in 1886 [3,4]. Robert Henry Clarke (1850–1926; > Figure 8-1b) studied medicine at St. George’s Hospital in London before surgical training in Glasgow. He worked with Horsley in London in the late 1880s [5]. Throughout the following decades, they became intellectually consumed by the experimental challenges of cerebral localization of motor function following the seminal work of Hughlings Jackson and others. Clarke traveled to Egypt in the early 1890s to convalesce after developing aspiration pneumonia from aspirin inhalation. While there, gazing up at the stars, he conceived an apparatus through #
Springer-Verlag Berlin/Heidelberg 2009
which probing intracranial instruments could be inserted that could be clamped to an animal’s head fixing it to a Cartesian co-ordinate system by skull pins placed laterally, bars attached to plugs inserted into the external auditory meatus and further bars resting upon the nose and orbital margins. The idea was presented to Horsley on return to England [6]. A decade later in 1905 James Swift, a machinist at Palmer & Company in London, was commissioned to construct the first machine from brass, ‘‘Clarke’s stereoscopic instrument employed for excitation and electrolysis,’’ comprising frame, carrier and needle holder and costing £300 (> Figure 8-2) [7]. Results were published in 1906 from experimental use of the first instrument for targeting electrolytic lesions in the deep cerebellar nuclei of non-human primates [8]. In 1908 they described the apparatus and its use in greater depth, coining the term ‘‘stereotaxic’’ from the Greek ‘‘stereos’’ meaning ‘‘solid’’ and ‘‘taxis’’ meaning ‘‘arrangement,’’ commenting that ‘‘by this means every cubic millimeter of the brain could be studied and recorded’’ [9]. Although Clarke suggested that the apparatus might be useful in humans, neither he nor Horsley pursued the idea and shortly afterwards they ceased collaboration. Yet Clarke patented the apparatus including its proposed use in humans in 1914 and devoted much time to its improvement. By 1920, a rectilinear modification enabling needle inclination at different angles in an equatorial frame enabling 360 of movement were described (> Figure 8-2) [10]. Three others used the original apparatus in London for experimental work; first the visiting
78
8
History of stereotactic surgery in great britain
. Figure 8-1 (a) Sir Victor Alexander Haden Horsley; (b) Robert Henry Clarke (courtesy of the Wellcome Library, London)
American surgeon Ernest Sachs who studied the optic thalamus [11], then the neurologist S.A. Kinnier Wilson who studied the basal ganglia of 25 monkeys using the ‘‘Clarke-Horsley machine’’ [12]. Clarke’s original instrument was last utilized by Barrington, a London urologist who used it to study the effect of brain lesions upon micturition in cats [13]. Barrington died suddenly in 1956. Among the contents of his laboratory, ‘‘in true British fashion, was a biscuit tin’’ that contained parts of the original apparatus. After several inquiries, a technician in the Royal Veterinary College where Barrington had once worked produced a mahogany box containing the original model and it was returned to University College London in 1970. It now resides in the Science Museum in London, having been promoted from closed storage to prominent display by Tipu Aziz in 2000 (> Figure 8-3). Two further apparatus designs were made for Clarke by machinists Goodwin and Velacott also of Palmer & Company in London and exported to the United States for animal research soon after the First World War, the latter to Johns Hopkins
after agreement that the Baltimore institution would publish Clarke’s stereotaxic atlas [14]. Another of Horsley’s students, Aubrey Mussen also contemplated translation of Horsley-Clarke stereotaxis to humans. Mussen purchased one of the original Horsley-Clarke apparatus secondhand for £100 while working at the National Hospital in London from 1905 to 1906 and returned to McGill University with it, subsequently publishing results from studies of the hypoglossal nuclei that Horsley traversed with his deep cerebellar lesions [15]. Mussen designed further stereotactic instruments with Clarke, including a ‘‘cyclotome,’’ a probe used to make disk-shaped incisions along its axis and a ‘‘spherotome’’ used to cut spherical volumes bearing much resemblance to Anto´nio Egas Moniz’ leucotome [10]. Mussen designed and had commissioned a modification of the Horsley-Clarke apparatus for use in humans in 1914 on his return to London. It was completed around 1918, again in brass, and most likely again by Palmer & Company (> Figure 8-4). In the frame, electrode holders slid along horizontal graduated bars or vertical corner
History of stereotactic surgery in great britain
8
. Figure 8-2 Clarke and Horsley’s 1905 primate stereotaxic apparatus, showing also Clarke’s 1920 equatorial modification (top left), (courtesy of the Wellcome Library, London)
posts enabling orthogonal approaches to intracranial structures in anteroposterior and lateral directions. The apparatus required a human brain atlas and Mussen envisaged its use to thermocoagulate tumors using ‘‘Galvanic current. . .through a 5 mm trephine in the skull and puncturing the dura without exposing the brain at all’’ [17]. In the two decades that followed, Mussen neither completed the human atlas nor convinced neurosurgical colleagues to take up use of his frame [18]. He wrapped the unused British made apparatus in newspaper dating from the 1940s and placed it in his loft [19].
After Horsley’s original experiments, Sachs, Wilson and Barrington all had loan of the original Horsley-Clarke frame. Sachs and Mussen utilized Clarke’s second and third frames respectively for animal experiments in North America throughout the 1920s [20,21], as did others during the following decade. However, at least two key challenges remained in translating experimental animal stereotaxis into a clinical tool. Firstly there was great variability between human skull landmarks and cerebral structures and secondly humans could not be sacrificed as animals were to enable confirmatory histology of accurate targeting - and thus experimental validity. Three decades on from
79
80
8
History of stereotactic surgery in great britain
. Figure 8-3 The senior author with one of the original HorsleyClarke frames in 2000
major advances were firstly to create a frame tailored to the individual skull by means of a plaster skull cap and secondly to align their so-called ‘‘stereoencephalatome’’ not just to skull but to brain landmarks like the calcified pineal gland and foramen of Monro by means of intraoperative pneumoencephalography.
Post-War Innovation
. Figure 8-4 Mussen’s circa 1918 human stereotactic instrument (after Picard et al. [16])
Horsley and Clarke’s work, Spiegel and Wycis devised an apparatus for stereotactic neurosurgery in humans, publishing their achievement in 1947 [22]. The North Americans established ‘‘stereotactic’’ as the preferred term, fusing Greek with the Latin word ‘‘tactis,’’ the pluperfect passive form of the verb ‘‘tangere’’ meaning ‘‘to touch.’’ Their
With two World Wars, British stereotactic surgery remained fallow for the half century after Horsley’s discovery, the discipline only reaching the clinic once word had spread of Spiegel and Wycis’ invention. At first, primary applications were for treating psychiatric disorders and later clinical usage diffused to movement disorders in the 1950s and chronic pain soon after. Ahead of the rest, an English crusader and two Scottish pioneers emerged, each a clinical polymath but with an academic focus. In London, Geoffrey Knight developed stereotactic subcaudate tractotomy for psychiatric disorders, treating hundreds of patients while his eminent contemporary Sir Wylie McKissock continued freehand approaches. In Edinburgh, John Gillingham established stereotactic surgery for multiple clinical indications, designing his own stereotactic frame. In turn, his talented associate Ted Hitchcock was inspired first at Edinburgh and then in Birmingham to pioneer stereotactic approaches to the high cervical spinal cord and brainstem. Francis John Gillingham (b.1916; > Figure 8-5) trained in St. Bartholomew’s Hospital in London before entering the neurosurgical faculty at Edinburgh in 1950. Gillingham spent 12 years as first assistant to Norman McOmish Dott, one of the great triumvirate alongside Sir Hugh Cairns in Oxford and Sir Geoffrey Jefferson in Manchester, the Cushing apostles who definitively established neurosurgery as a specialty in Great Britain [24–26]. Gillingham was a brilliant and pioneering aneurysm surgeon
History of stereotactic surgery in great britain
8
. Figure 8-5 Francis John Gillingham (left) preparing for a stereotactic thalamotomy in 1968 (after Housepian 2004 [23])
like his mentor Dott [27,28]. A passionate educator, he also introduced the concept of subspecialty fellowships to British neurosurgical training [29], but stereotactic surgery received his greatest contributions. The Parisian neurosurgeon Gerard Guiot. introduced Gillingham to stereotactic surgery. They had become friends after Guiot visited Edinburgh to learn aneurysmal surgery from Dott and Gillingham. Guiot’s 1953 telegram to Gillingham read ‘‘I have something interesting to show you – come over.’’ Four days were spent performing freehand pallidotomies to treat parkinsonism under local anesthesia using a subfrontal approach to the anterior perforated substance interrupting the ansa lenticularis as described by Fenelon and Thiebaut following the seminal discoveries of Cooper [30–32]. Gillingham returned to Edinburgh to treat two patients in 1955 and 1957, reporting improvements in tremor, rigidity and quality of life. Wishing to reduce risks from the demanding subfrontal approach, he adapted Guiot’s stereotactic method [33]. In 1960 he published results from stereotactic ‘‘thermal electrocoagulation lesions of the globus pallidus, internal
capsule and thalamus either separately or in combination’’ in a further 58 patients operated upon from 1957 to 1959 [34]. In addition to globus pallidus and internal capsule, he began targeting the ventrolateral thalamus for refractory tremor based on work by Hassler [35]. ‘‘Of these (60) patients 53, or 88%, had tremor and/or rigidity abolished or significantly reduced without complications’’ [34]. From his early clinical results, Gillingham drew several conclusions. On targeting he wrote that ‘‘The best type of lesion. . .would seem to be the double one, made at the same time in the ventro-lateral nucleus of the thalamus and in the globus pallidus 16mm from the mid-line, both lesions bordering on the internal capsule. . .. Bilateral lesions in the treatment of bilateral Parkinsonism, provided they are small and strategically placed, would seem to be eminently practicable. . . usually with an interval of 3–6 months between the two operations.’’ On his modification of Guiot’s stereotactic apparatus he felt ‘‘that the merits of this method lie in the relatively short operative procedure and in its accuracy and simplicity. Its principles are based
81
82
8
History of stereotactic surgery in great britain
on the fact that the globus pallidus and thalamus bear a reasonably constant anatomical relationship to the anterior and posterior commissures, the intercommissural line, and the mid-sagittal plane of the head. . .. The method used has evolved progressively, and is unique, in allowing the creation of lesions in the globus pallidus, internal capsule, or thalamus with one electrode track at different depths’’ [34]. In their stereotactic apparatus design, Guiot and Gillingham favored operative principles to prioritize patient comfort, not restricting their movements by clamping their head, and to reduce laborious calculations. Guiot planned a parasagittal approach using intraoperative encephalography to delineate the midline and intercommissural point. Gillingham favored an occipitoparietal entry to avoid striate arteries and horizontal patient positioning to reduce putative brain shift. Thus the Guiot-Gillingham stereotactic apparatus was devised (> Figure 8-6). Radioopaque midline markers were used for the procedure and a 1 mm steel ball placed in each 5 mm lesion for subsequent charting. Over the . Figure 8-6 The Guiot-Gillingham stereotactic apparatus using a posterior rather than a coronal approach (after Gillingham et al. [34])
post-operative weeks, the ball was seen to fall through the necrotic lesion on skull radiographs, elegantly providing an estimate of lesion size. The frame’s conception preceded Hassler’s targeting of the thalamus for tremor and hence Gillingham attributed to serendipity that his posterior approach enabled multiple targets to be lesioned in a single pass [36]. Despite impressive clinical outcomes, Gillingham noted some inaccuracy to his lesions in the context of Brierley and Beck’s demonstration that relationships between basal ganglia structures and commissural landmarks were highly variable [37,38]. David Whitteridge, his neurophysiologist colleague at Edinburgh, had demonstrated to him in 1961 how microelectrode recording could distinguish between grey and white matter and thus delineate the lateral geniculate nuclei in the cat [39]. He immediately saw its utility for distinguishing functionally between deep brain structures and, with his colleague Michael Gaze, developed the technique for humans as did Guiot [40]. Fundamental physiological insights were gained in the quest to improve lesion accuracy and clinical efficacy, including spontaneous rhythmical discharge in the thalamus found to be synchronous with tremor [41]. Using microelectrode recording, target localization could be done accurately with a margin of error less than 1mm. Gillingham evolved the Guiot-Gillingham apparatus throughout the 1960s and 1970s, He added a phantom to allow an oblique track to more medial brain targets for epilepsy and psychiatric disorders, then an inferior extension to the posterior limb of the frame for targeting the cerebellum, brain stem and cervical spine in chronic pain and dystonias [42]. In 1977 he added a motor to automatically drive an electrode in at a slow and measured rate for microelectrode recording. Stereotactic surgery for deep hematomas and tumor biopsies was also performed [43]. Ten year follow-up in the post levodopa era of a second 60 patient parkinsonian cohort of Gillingham’s operated upon between
History of stereotactic surgery in great britain
1965 and 1967 showed decline in efficacy for bradykinesia, but consistent relief of tremor and rigidity [44]. Gillingham remained engaged in academic neurosurgery well into his ninth decade [45]. As Gillingham was Dott’s prote´ge´, so Ted Hitchcock was Gillingham’s. Edward Robert Hitchcock (1929–1993) studied medicine at Birmingham then neurosurgery at Oxford before joining the Edinburgh staff at the then recently opened Western General Hospital in 1965. While there, he received unique exposure to Gillingham’s stereotactic surgery which attracted international renown [23]. Hitchcock’s developed an interest in chronic pain and in particular the concept of percutaneous high spinal stereotactic commissural myelotomy. This procedure aimed to divide the decussating spinothalamic tracts through a targeted lesion and reduce the risks of respiratory paralysis conferred by open cordotomy. It was aimed at chronic cancer pain. It required access below the plane of a versatile frame, thus he invented his own target-centered arc system secured on a hollow square aluminum base ring secured to the skull by three-point fixation (> Figure 8-7). Vertical and horizontal bars determined probe length and laterality. The system was first used both for surgery and for microelectrode recording in the spinal cord by
. Figure 8-7 Hitchock’s stereotactic apparatus for brainstem and cervical spine surgery (after Hitchcock [46])
8
percutaneous approach using portable radiographs [46–48]. Hitchcock reported initial results of good or complete pain relief in 13 out of 19 patients at follow-up ranging from 1 week to 4 years [49]. A stereotactic pontine approach to spinothalamic tractotomy and to the trigeminal nucleus for anesthesia dolorosa was also applied [50–54], as were approaches to the thalamus and dentate nucleus to treat dystonia and in particular the spasticity of cerebral palsy [55]. The rationale behind the stereotactic pontine spinothalamic approach was to provide good analgesia with minimal risks to respiration, micturition and upward gaze [54]. Hitchcock wrote of his apparatus in the early 1970s that ‘‘the design and construction make this one of the most accurate, adaptable and simplest of modern stereotactic instruments’’[56]. Hitchcock became Professor of Neurosurgery at Birmingham in 1978, succeeding Brodie Hughes (1913–1989) who was also an established stereotactic surgeon [57,58]. Hitchcock put his stereotactic frame to many further clinical uses including biopsy of supratentorial, infratentorial and high spinal tumors and intraventricular masses [59–61], foreign body removal [62], realtimeclippingofotherwiseinoperablearteriovenous malformations [56], image-guided craniotomies [63], and in the 1980s in the planning and treatmentstagesofradiosurgery[64].Thesemanyvaried clinical indications in brain and spine earned him the nickname ‘‘Columbus of the brain’’ in the local clinical neuroscience community. At the MidlandHospitalforNeurologyandNeurosurgery, he used his stereotactic expertise to established a programme at first for adrenal medullary and then in the late 1980s for fetal mesencephalic transplantation in Parkinson’s disease, performing the procedure on 55 patients [65–68]. Neurosurgery for psychiatric disorders in Great Britain echoed its popularization in the United States following Freeman and Watts’ simplification of Moniz’ procedure in the 1940s [69,70]. Its foremost British proponent was the
83
84
8
History of stereotactic surgery in great britain
London neurosurgeon Sir Wylie McKissock (1906–1994), founder of the neurosurgical department at Atkinson Morley’s Hospital in Wimbledon [71]. McKissock favored a freehand approach to the frontal lobe from above [72]. He described the rostral leucotomy in 1951 as a rejoinder to Freeman and Watt’s transorbital ‘‘icepick’’ leucotomy which he considered to contravene ‘‘established aseptic surgical principles’’ [73,74]. McKissock’s immense South England practice and his reputation for extraordinary surgical speed inculcated a peripatetic service visiting other hospitals in his car with his surgical instrument set in the boot, drawing parallels with Freeman [75,76]. It is suggested that McKissock alone may have performed one quarter of the 10,365 procedures performed in the United Kingdom from 1942 to 1954 [77]. Geoffrey Cureton Knight (1906–1994) of Hammersmith and Brook Hospitals in London and Woolwich saw more readily than McKissock the merits of stereotactic over freehand approaches in reducing the morbidity and mortality of neurosurgery for psychiatric disorders [74,78]. After his freehand experience [79], Knight created the procedure of stereotactic subcaudate tractotomy in 1961 using a modified stereotactic device that his London colleague the Scottish neurosurgeon Ian Reay McCaul (1916–1989) reported in 1959 [58,80]. His first few hundred orbital undercuttings led him to conclude that lesions extending posteriorly under the caudate were most efficacious and that the last 2 cm was key [81,82]. Knight used bony landmarks on lateral radiographs, and later air encephalography to guide him. In addition, he employed brachytherapy as an ablative tool, implanting radioactive Yttrium (Y90) to create flat lesion approximately 20 by 20 by 7 mm (> Figure 8-8) [83–85]. The treatment proved effective and endured four decades amid decline in use of other psychosurgical treatments. The group described treatment of 1,300 patients with ‘‘non-schizophrenic affective disorders,’’ 40–60% going on to live normal or near normal lives with a reduction in
. Figure 8-8 Anteroposterior radiograph of Geoffrey Knight’s stereotactic subcaudate tractotomy showing yttrium seeds in situ for brachytherapy (after Knight [83])
suicide rates from 15 to 1% [86,87]. Long-term outcomes were published by the psychiatrist Bridges and the London neurosurgeon John Bartlett, Knight’s successor from 1972. Following the retirement of Knight, the unit was named the Geoffrey Knight Unit for Affective Disorders to emphasize Knight’s appreciation of the fundamental importance of psychiatric evaluation both in diagnosis and in full consideration of medical treatments prior to offering surgery. In 1996 the unit moved to the Maudsley Hospital and Y90 production ceased. Bartlett adapted a Leksell frame arc compatible with modern neuroimaging using concepts that underlay the McCaul device. Radiofrequency lesioning replaced radioisotope implantation [88,89]. It is a tribute to Knight that McKissock’s colleague at Atkinson Morley’s Hospital, Alan Richardson (1926–1998), adopted a stereotactic approach for his psychiatric procedures [72]. Together with his psychiatrist colleague Desmond Kelly, he combined Knight’s subcaudate tractotomy with a cingulotomy to invent the procedure of limbic leucotomy in the early
History of stereotactic surgery in great britain
1970s [90,91]. It is interesting to note that cingulectomy for psychiatric disorders was first performed in 1948 in Oxford by Sir Hugh Cairns, albeit freehand [92]. Both Knight’s subcaudate tractotomy and Richardson’s limbic leucotomy continue to be performed in carefully selected cases refractory to medical treatment worldwide including in Great Britain and elswehere [93–95]. Stereotactic neurosurgery was embraced by Dott’s unit in Edinburgh and, at Oxford after Cairns, Pennybacker appointed Watkins to establish a service, but other regions were also keen to commence it. In Manchester, Jefferson’s successor Richard Johnson appointed John Dutton to undertake a high volume of ablations for parkinsonism and other stereotactic procedures throughout the 1960s using a Leksell frame. Soon after, John Gleave (1925–2006) established a stereotactic service in Cambridge also using a Leksell frame, treating parkinsonism with cryosurgery and developing a side-cutting stereotactic biopsy cannula [96,97]. Other British neurosurgeons like McCaul made and modified stereotactic frames. Of particular note was the frame of Alfred Michael Bennett (1920–1996) and his use of a sphere inserted into a burr hole to aid targeting [98,99]. Bennett’s apparatus were popular locally, used by Sid Watkins and later David Thomas in London amongst others [100,101]. Most designs were less radical and therefore perhaps less memorable than those of Gillingham and Hitchcock.
Stereotactic Atlases Horsley and Clarke produced the first stereotaxic atlas, a monkey version appearing in their 1908 publication. Later publications were by Clarke at first for the cat in collaboration with the British ophthalmic surgeon E. Erskine Henderson in 1912 and later by Clarke alone for the monkey in 1920 [10,102,103]. Both atlases comprised brain slices 2 mm thick. The latter atlas showed sections of monkey brain at calibrated intervals
8
with a scale giving slice thickness and height from the base of the apparatus. Sections were registered by a Cartesian coordinate system to the skull landmarks of inferior orbital rim and both external auditory canals to which the frame was fixed. Zero axes were the plane between these structures axially, the mid-sagittal plane and the coronal plane between both external auditory meatus orthogonal to both other planes. The human brain atlases produced outside Great Britain, in particular by Spiegel and Wycis, Schaltenbrand and Bailey and Talairach transformed stereotactic functional neurosurgery. Schaltenbrand detailed anatomical nuclei with an emphasis on the thalamus and adjacent deep brain structures now used in deep brain stimulation for movement disorders and Talairach revealed vasculature relevant to epilepsy surgery [104–106]. The British neurosurgeon Sid Watkins also made rigorous contributions. Eric Sidney Watkins (b.1932) was given the task of starting stereotactic neurosurgery at the Radcliffe Infirmary in Oxford in the 1950s by Joe Pennybacker. Dissatisfied with the variability of basal ganglia structures with respect to the frequently uncalcified and thus radiolucent pineal gland using the Spiegel and Wycis atlases, he used the Schaltenbrand and Talairach atlases that appeared shortly after. Initially globus pallidus and ansa lenticularis were targeted for Parkinsonian rigidity, tremor and dystonia then later the lateral thalamus in the 1960s. The desire to create his own atlas arose in the early 1960s from a wish to commence thalamotomy for pain and a concern at the adequacy of available atlases to accurately enable targeting based upon anatomy alone in the absence of subjective or physiological guidance. Encouragement came from London neurosurgical colleagues John Andrew and Valentine Logue who were also keen to begin such therapies. Another atlas that became available was created by Brierley and Beck who sectioned 40 brains in 3–5 mm slices, relating them in a proportional hypothesis for thalamic nuclear determination to
85
86
8
History of stereotactic surgery in great britain
anterior and posterior thalamic limits and the midthalamic point and describing great individual variations [38]. Watkins found the atlas to be limited clinically as the use of simultaneous positive and air ventriculography using air in the ambient cisterns to outline the pulvinar nuclei and thus the thalamic limits was not consistently reproducible. In the 1960s at the National Hospital for Nervous and Mental Diseases together with Watkins then at the Royal London Hospital, John Andrew produced a greatly enlarged atlas with drawings defining in detail deep brain nuclei including the thalamus and its relations [107]. The atlas was based upon 38 formalin fixed brains. It measured the position of the thalamic centromedian nucleus using 1 mm coronal slices with reference planes between the posteroinferior margin of the foramen of Monro and posterior commissure and the midpoint between the ventricular surfaces of the anterior and posterior commissures. Its utility lies in the presentation of statistical data in a graphic form together with stereotactic coordinates superimposed on simple line drawings of the thalamus. In 1978 at the London Hospital, Fari Afshar detailed brain stem and cerebellar nuclei, again under Sid Watkins’ supervision [108]. The impetus for the Afshar atlas came from an interest in attempting to ameliorate spasticity in cerebral palsy by ablation of the cerebellar dentate nucleus. Approximately 30 brains were prepared using positive-contrast ventriculography with skull and brain mounted in a stereotactic frame in order to accurately correlate structures with coordinates. Again, formalin fixation of 1mm slices was performed. Modified Mulligan stain was used and each section magnified, with drawings made using a camera lucida. Reference plains were the fourth ventricular floor and fastigium and the midsagittal plane. As before, variability profiles were quantified and standard deviations presented. Both Watkins atlases are resources used today to help ratify localization and indeed the
Afshar atlas continues to augment heated debates regarding the targeting of the novel functional neurosurgical treatment of deep brain stimulation of the pedunculopontine region for Parkinson’s disease [109,110]. Watkins has commented upon the major difficulties in measurement due to distortion related to fixation and shrinkage, past atlases suffering from approximately 10% shrinkage. To reduce shrinkage to 2–5%, he utilized Corsellis’ technique – a ten day formalin suspension after removing brain and skull en bloc minus the frontal and facial bones [111]. To appease undertakers’ concern at cosmetic consequences, each cadaver’s scalp was replaced over a plaster of Paris prosthesis fixed to a broom handle on a nail inserted into the cervical canal. However, cremation of the augmented cadavers became suboptimal, precipitating a strike among undertakers serving the London Hospital by the time Afshar’s posterior fossa atlas had reached its completion [112].
Computed Tomography By the late 1970s, British stereotactic functional surgery was a decade on from its first successes, having declined with the advent of neuropsychopharmacology. Levodopa was introduced to relieve Parkinson’s disease [113], chlorpromazine and monoamine oxidase inhibitors to ameliorate schizophrenia and depression respectively, and the case series showing good relief after lesional surgery for chronic pain paled against a background of new analgesics and peripheral neuromodulatory therapies. Stereotactic approaches to tumors had been established but most neurosurgeons did not train in stereotactic neurosurgery. Hitchcock continued psychiatric procedures for medically refractory depression, obsessivecompulsive disorders and anxiety alongside Bartlett and Richardson, but the developing subspecialty sought advances in other domains. Enter the British engineer.
History of stereotactic surgery in great britain
Sir Godfrey Newbold Hounsfield (1919– 2004; > Figure 8-9) joined EMI in Hayes, Middlesex in 1951, having been first a mechanic first of radios then later radars in the Royal Air Force and obtained a diploma from Faraday House Electrical Engineering College in London. At EMI he worked first on radars and guided weapons, then the first all-transistor computers. During a weekend ramble in 1967 he conceived what later became the first EMI-scanner and the technique of computed tomography (CT), which he recounted as ‘‘a realization that you could determine what was in a box by taking readings at all angles through it.’’ Recording multiple pictures from a rotating X-ray source, a series of slices could be photographed and a three-dimensional image reconstructed from the slices. After initial successful experiments with a cylindrical phantom containing radio-opaque objects in his Hayes laboratory using X-rays, Hounsfield forged a collaboration . Figure 8-9 Sir Godfrey Hounsfield at the controls of the EMI scanner in Atkinson Morley’s Hospital in London (after Petrik et al. [114])
8
with James Ambrose, radiologist at Atkinson Morley’s Hospital, to translate the device’s utility to humans. McKissock gave the endeavor his blessing [75]. Hounsfield set to work on bullock’s heads obtained from a kosher slaughterhouse in East London to obviate the traumatic intracranial hemorrhage seen after conventional slaughter [114]. Ambrose interpreted the early scans and suggested the use of sodium iothalamate contrast to highlight tumors [115]. His early interpretations and predictions formed the basis of contemporary diagnostic neuroradiology [116,117]. The first patient was scanned in 1971, revealing a cyst [118]. In 1979 Hounsfield was awarded the Nobel Prize for Physiology or Medicine together with Allan Cormack, the Cape Town physicist whose mathematical theories Hounsfield had realized [119]. The advent of CT breathed new life into stereotactic surgery in Britain. Several neurosurgeons began to experiment with CT compatible apparatus, both imported, usually Leksell, and those of Gillingham and Hitchcock [36,120]. Magnetic resonance imaging (MRI) followed shortly after, again with frames being modified as required. After a quiescent decade, the late 1980s augured for a renaissance. Alongside emerging limitations of drug therapies for movement disorders resurrecting clinical indications for functional neurosurgery, great advances came from increasing computer power enabling the fusion of more spatially robust CT information with the greater soft tissue detail of MRI and comparison to computerized brain atlas images. In Britain, the current generation of senior stereotactic neurosurgeons were gaining their clinical training and conducting their first research, some in classical stereotactic methods abroad, some by subspecialty fellowships in Britain thanks to Gillingham’s enduring influence upon neurosurgical programmes, and others by animal experiments true to Horsley’s hybrid scientist-surgeon mould. The stage was set for two decades of rapid advances in British stereotactic neurosurgery.
87
88
8
History of stereotactic surgery in great britain
Radiosurgery Lars Leksell’s brilliance showed not only in his frame design [121], employing the novel arcquadrant principle, but also in his insight that focused radiation could be used as the tool. Many intersecting radiation beams focused towards a target would result in a high cumulative radiation dose, with radiation intensity declining rapidly with distance from the ‘‘isocenter.’’ Thus, a deep brain structure could be lesioned noninvasively by focused radiation. The technology of ‘‘radiosurgery’’ could be applied to acoustic neuromas, arteriovenous malformations and other discrete pathologies [122]. Britain acquired one of the first gamma knives in 1985 at the same time as Argentina just 3 years on from the Falklands War. David Forster achieved incredible feats in leading the campaign to fund the device on the British National Health Service, building the infrastructure over 2 years to host it and ultimately purchasing the first custom-made unit in 1985. The National Centre for Stereotactic Radiosurgery was refurbished in 1991 and its cobalt sources renewed. The unit receives approximately one half to two thirds of all radiosurgery referrals from all over the United Kingdom. Approximately 500 cases a year are performed, a third of cases treated being ateriovenous malformations, with small to medium sized acoustic neuroma treatment having increased from 10% to one third over the period 1994–2001 and approximately 100 meningiomas and other skull base or recurrent tumors treated per year. Other indications treated have included trigeminal neuralgia, pituitary tumors and metastases, although the latter two indications are treated in smaller proportions than outside the United Kingdom, reflecting more conservative referral patterns [123]. For similar reasons, few epilepsy and functional cases have been performed. Several neurosurgical centers use linear accelerators to perform radiosurgery, each performing up to 40 cases per year. A gamma knife was also
acquired by the Cromwell Hospital in London in 1998, run privately by Christer Lindquist. It has treated 1,000 patients since installation and has recently been refurbished.
Functional Surgery European factors driving British stereotactic surgery included Benabid’s application of thalamic deep brain stimulation to Parkinson’s disease in 1987 and Laitinen’s reapplication of Leksell’s pallidotomy in 1992 [124,125]. Functional neurosurgery was resurrected at the Radcliffe Infirmary in Oxford four decades after Watkins’ departure under the headship of Mr. Christopher Adams [126]. We had already established with Alan Crossman in Manchester by the early 1990s at the same time as DeLong’s team across the Atlantic that lesions made to the subthalamic nucleus in primates reversed the motor symptoms of 1-methyl-4-phenyl-1,2,3,6tetrahydropyridine (MPTP) induced parkinsonism [127–129], At Oxford and Charing Cross Hospital in London, we undertook stereotactic surgery of this target and others [130–135]. At the same time we continued non-human primate research into establishing the pedunculopontine nucleus as a potential target for gait freezing and postural instability [136–139]. The former target is now the target of choice for Parkinson’s disease surgery and initial clinical results of the latter show great promise [140–143]. Other translational research at the University of Oxford and Imperial College London included invasive deep brain electrophysiological insights into tremor and dystonia [144–146], use of single photon emission tomography (SPET) [147], magnetoencephalography (MEG) [148], and diffusion tensor imaging (DTI)[149,150] to study deep brain stimulation and research into deep brain stimulation for pain and blood pressure control and brainstem control of exercise [151,152]. We have used deep brain stimulation to treat 70 patients
History of stereotactic surgery in great britain
with dystonia [153], 60 with chronic pain [154], and we perform one fifth of Britain’s movement disorders surgery. In Bristol, Professor Steven Gill has continued the Great British tradition of innovation, creating a stereotactic frame convenient for radiosurgery under Professor David Thomas’ supervision in London [155,156], and performing several clinical firsts including glial-cell derived neurotrophic factor infusion and pedunculopontine nucleus stimulation for Parkinson’s disease [157–160]. With Mr. Nikunj Patel, he continues to drive the field forward. After the retirement last decade of Mr. John Miles in Liverpool whose tremendous pain practice still left time for several innovations [161– 163], Professor David Thomas also recently
8
retired as the Gough-Cooper Professor of Neurosurgery at the National Hospital of Neurology and Neurosurgery at Queen Square. He had devoted three decades to the improvement of stereotactic surgical techniques with and without frames [164–168]. Britain welcomed Professor Marwan Hariz at Queen Square as the first Edmond J. Safra Chair of Functional Neurosurgery, ‘‘solving’’ the functional neurosurgery service there [169], and establishing a biennial international workshop like its host, unsurpassed for its conviviality and candour (> Figure 8-10). Almost all of the 34 hospitals conducting neurosurgery in Great Britain and Northern Ireland have consultants able to offer stereotactic surgery a century on from Horsley’s first experiments. A third of these hospitals have
. Figure 8-10 The faculty of the International Workshop on Functional Neurosurgery for Movement Disorders and Mental Illness & Commemoration of the 150th Anniversary of the Birth of Sir Victor Horsley, London, 2007 (courtesy of Professor Marwan Hariz). Back Row from left to Right: Lazaro Alvarez La Habana, Peter Brown, Gun Marie Hariz, Paul Krack, Pierre Pollak, Bart Nuttin, Roger Melvill, Steven Gill, Patricia Limousin-Dowsey, Niall Quinn, John Rothwell, Veerle Visser-Vandewalle, Roger Lemon, Rees Cosgrove, Andres Lozano, Laura Cif, Ludvic Zrinzo, Marjan Jahanshahi, Stephen Tisch, Hans Speelman, Philippe Coubes, Pat Forsdick. Front Row Left to Right: Takaomi Taira, Jean-Luc Houeto, Alim-Louis Benabid, Tipu Aziz, Boulos-Paul Bejjani Byblos, Alan Crockard, Marwan Hariz, Carmelo Sturiale
89
90
8
History of stereotactic surgery in great britain
subspecialty trained stereotactic surgeons offering functional procedures, the majority of them affiliated to universities and conducting clinical or translational research. Far from being the reserve of the eccentric scientist-surgeons looked upon with suspicion by the rest of the neurosurgical fraternity, stereotactic surgery has become an established clinical subspecialty and academic discipline in its own right. The Society of British Neurological Surgeons formed in 1929 has recently begun devoting specific sections to stereotactic and functional neurosurgery at its meetings. A further society with a focus upon pain that welcomes stereotactic neurosurgeons, the Neuromodulation Society of the United Kingdom and Ireland (NSUKI) was founded in 2001. Such factors prove good indicators of the definitive establishment of the subspecialty in the United Kingdom. Phil Gildenberg has commented that there are four central tenets of the field of stereotactic surgery [170]. The need to be innovative – that a better way to do something may be more apparent to the most junior member of the team rather than the most senior. That stereotactic surgeons work as a community not in isolation. That stereotactic surgery is, to a large extent, a basic science. Although not appealing to the neurosurgeon interested only in a better way to cut, it is exciting to one appreciating the associated basic science. Finally, there is awed appreciation for the insight and courage of the true pioneers in the field. The British school exemplifies such values. It is hoped that its practitioners will continue to uphold tradition as we look to the future with excitement.
Acknowledgments We thank Mr. John Bartlett for helpful comments, Professor David G. T. Thomas, Professor Anthony J. Strong, Professor John D. Pickard, Mr. Robert Macfarlane and Mr. Colin Watts for advice on historical sources and Professor Marwan I. Hariz
for helpful comments and advice on > Figure 8-10. The authors receive financial support for research from the UK Medical Research Council, Normal Collisson Foundation, Charles Wolfson Charitable Trust and Oxford Partnership Comprehensive Biomedical Research Centre.
References 1. Dittmar C. Ueber die Lage des sogenannten Gefaesszentrums in der Medulla oblongata. Bersaechs Ges Wiss Leipzig (Math Phys) 1873;25:449-69. 2. Zernov DN. Encephalometer: device for estimation of parts of brain in human. Proc Soc Physicomed Moscow Univ 1889;2:70-80. 3. Paget S. Sir Victor Horsley: a study of his life and work. New York: Harcourt, Brace and Howe; 1920. 4. Northfield WC. Sir Victor Horsley, his contributions to neurological surgery. Surg Neurol 1973;1:131-4. 5. Davis RA. Victorian physician-scholar and pioneer physiologist (Robert Henry Clarke). Surg Gynecol Obstet 1964;119:1333-40. 6. Sachs E. Victor Horsley. J Neurosurg 1958;15:240-4. 7. Schurr PH, Merrington WR. The Horsley-Clarke stereotaxic apparatus. Br J Surg 1978;65:33-6. 8. Clarke RH, Horsley V. On a method of investigating the deep ganglia and tracts of the central nervous system (cerebellum). Br Med J 1906;2:1799-800. 9. Horsley V, Clarke RH. The structure and functions of the cerebellum examined by a new method. Brain 1908;31:45-124. 10. Clarke RH. Part I: Investigation of the central nervous system, methods and instruments. Baltimore, MD: Johns Hopkins; 1920. 11. Sachs E. On the structure and functional relations of the optic thalamus. Brain 1909;32:95-186. 12. Wilson SAK. An experimental research into the anatomy and physiology of the corpus striatum. Brain 1914;36:427-92. 13. Barrington FJF. The effect of lesions of the hind- and midbrain on micturition in the cat. Q J Exp Physiol 1925;15:81-102. 14. Fodstad H, Hariz M, Ljunggren B. History of Clarke’s stereotactic instrument. Stereotact Funct Neurosurg 1991;57:130-40. 15. Mussen AT. Notes on the movements of the tongue from stimulation of the twelfth nucleus, root and nerve. Brain 1909;32:206. 16. Picard C, Olivier A, Bertrand G. The first human stereotaxic apparatus. The contribution of Aubrey Mussen to the field of stereotaxis. J Neurosurg 1983;59:673-6. 17. Bertrand G. Stereotactic surgery at McGill: the early years. Neurosurgery 2004;54:1244-51; discussion 1251-2.
History of stereotactic surgery in great britain
18. Jensen RL, Stone JL, Hayne R. Use of the Horsley-Clarke stereotactic frame in humans. Stereotact Funct Neurosurg 1995;65:194-7. 19. Gildenberg PL. The history of stereotactic neurosurgery. Neurosurg Clin N Am 1990;1:765-80. 20. Sachs E, Fincher EF. Anatomical and physiological observations on lesions in the cerebellar nuclei in Macacus rhesus (preliminary report). Brain 1926;50:350-6. 21. Mussen AT. Experimental investigations on the cerebellum. Brain 1927;50:313-49. 22. Spiegel EA, Wycis HT, Marks M, Lee AJ. Stereotaxic apparatus for operations on human brain. Science 1947;106:349-50. 23. Housepian EM. Stereotactic surgery: the early years. Neurosurgery 2004;55:1210-4. 24. Fraenkel GJ. Hugh Cairns: first Nuffield professor of surgery, University of Oxford. Oxford: Oxford University Press; 1991. 25. Rush C, Shaw JF. With sharp compassion: Norman Dott Freeman surgeon of Edinburgh. Aberdeen: Aberdeen University Press; 1990. 26. Schurr PH, Royal Society of Medicine (Great Britain). So that was life: a biography of Sir Geoffrey Jefferson, Kt CBE FRS MS FRCS (1886–1961): master of the neurosciences and man of letters. London: Royal Society of Medicine Press; 1997. 27. Todd NV, Howie JE, Miller JD. Norman Dott’s contribution to aneurysm surgery. J Neurol Neurosurg Psychiatry 1990;53:455-8. 28. Gillingham J. Proceedings: twenty-five years’ experience with middle cerebral aneurysms. J Neurol Neurosurg Psychiatry 1975;38:405. 29. Gillingham FJ. Surgical training in the EEC – the training of a specialist. Acta Neurochir (Wien) 1982;61:17-24. 30. Cooper IS. Ligation of the anterior choroidal artery for involuntary movements: parkinsonism. Psychiatr Q 1953;27:317-9. 31. Fenelon F, Thiebaut F. Resultats du traitment neurochirugical du syndrome parkinsonien par intervention direct sur les voies etrapyramidales immediatement sous-striopallidales (anse lenticulaire). Rev Neurol (Paris) 1950;83:437-40. 32. Guiot G, Brion S. Traitement des mouvements anormaux par la coagulation pallidale; technique et resultats. Rev Neurol (Paris) 1953;89:578-80. 33. Guiot G. Le traitement des syndromes parkinsoniens par la destruction du pallidum interne. Neurochirurgia (Stuttg) 1958;1:94-8. 34. Gillingham FJ, Watson WS, Donaldson AA, Naughton JA. The surgical treatment of parkinsonism. Br Med J 1960;2:1395-402. 35. Hassler R, Riechert T. Indikationen und lokalisations: methode de gezeilten hirnoperationen. Nervenarzt 1954;25:441-7.
8
36. Gillingham J. The Guiot-Gillingham Apparatus. In: Gildenberg PL, Tasker RR, editors. Textbook of stereotactic and functional neurosurgery. New York: McGraw Hill; 1998. p. 139-44. 37. Gillingham J. Introduction to the scientific sessions. Accidents in stereotaxy – side-effects or bonus? Acta Neurochir (Wien) 1974;Suppl 21:5–12. 38. Brierley JB, Beck E. The significance in human stereotactic brain surgery of individual variation in the diencephalon and globus pallidus. J Neurol Neurosurg Psychiatry 1959;22:287-98. 39. Gillingham FJ. Neurosurgery. Br J Surg 1966;53:833-6. 40. Guiot G, Hardy J, Albe-Fessard D. Delimination precise des structures sous-corticales et identification de noyaux thalamiques chez l’homme par l’electrophysiologie stereotaxique. Neurochirurgia (Stuttg) 1962;5:1-18. 41. Gaze RM, Gillingham FJ, Kalyanaraman S, Porter RW, Donaldson AA, Donaldson IM. Microelectrode recordings from the human thalamus. Brain 1964;87:691-706. 42. Gillingham FJ, Campbell D. Surgical interruption of the conduction pathways for the control of intractable epilepsy. Acta Neurochir Suppl (Wien) 1980;30:67-74. 43. Kalyanaraman S, Gillingham FJ. Stereotaxic biopsy. J Neurosurg 1964;21:854-8. 44. Kelly PJ, Gillingham FJ. The long-term results of stereotaxic surgery and L-dopa therapy in patients with Parkinson’s disease. A 10-year follow-up study. J Neurosurg 1980;53:332-7. 45. Gillingham J. Forty-five years of stereotactic surgery for Parkinson’s disease: a review. Stereotact Funct Neurosurg 2000;74:95-8. 46. Hitchcock E, Lewin M. Stereotactic recording from the spinal cord of man. Br Med J 1969;4:44-5. 47. Hitchcock E. Stereotaxic spinal surgery. A preliminary report. J Neurosurg 1969;31:386-92. 48. Hitchcock E. An apparatus for stereotactic spinal surgery. Lancet 1969;1:705-6. 49. Hitchcock E. Stereotactic myelotomy. Proc R Soc Med 1974;67:771-2. 50. Hitchcock E. Stereotactic trigeminal tractotomy. Farmatsiia 1970;19:131-5. 51. Hitchcock E, Kim MC, Sotelo M. Further experience in stereotactic pontine tractotomy. Appl Neurophysiol 1985;48:242-6. 52. Hitchcock E, Teixeira MJ. Pontine stereotactic surgery and facial nociception. Neurol Res 1987;9:113-7. 53. Hitchcock E, Sotelo MG, Kim MC. Analgesic levels and technical method in stereotactic pontine spinothalamic tractotomy. Acta Neurochir (Wien) 1985;77:29-36. 54. Hitchcock ER. Stereotaxic pontine spinothalamic tractotomy. J Neurosurg 1973;39:746-52. 55. Gornall P, Hitchcock E, Kirkland IS. Stereotaxic neurosurgery in the management of cerebral palsy. Dev Med Child Neurol 1975;17:279-86.
91
92
8
History of stereotactic surgery in great britain
56. Shieff C. The Hitchcock apparatus. In: Gildenberg PL, Tasker RR, editors. Textbook of stereotactic and functional neurosurgery. New York: McGraw Hill; 1998. p. 101-4. 57. Hughes B. Involuntary movements following stereotactic operations for parkinsonism with special reference to Hemi-Chorea (Ballismus). J Neurol Neurosurg Psychiatry 1965;28:291-303. 58. Lyle I, Taylor S. Lives of the fellows of the Royal College of Surgeons of England: 1983–1990. London: Royal College of Surgeons of England; 1995. 59. Hitchcock ER, Issa AM, Sotelo MG. Stereotactic excision of deeply seated intracranial mass lesions. Br J Neurosurg 1989;3:313-20. 60. Hitchcock E, Kenny BG. Stereotaxy of third ventricular masses. Acta Neurochir Suppl (Wien) 1989;46:82-5. 61. Hitchcock E, Morris CS. Immunocytochemical techniques in stereotactic biopsy. Stereotact Funct Neurosurg 1989;53:21-8. 62. Hitchcock E, Cowie R. Stereotactic removal of intracranial foreign bodies: review and case report. Injury 1983;14:471-5. 63. Hitchcock E. Open stereotactic surgery. Acta Neurochir Suppl (Wien) 1991;52:9-12. 64. Hitchcock E, Kitchen G, Dalton E, Pope B. Stereotactic LINAC radiosurgery. Br J Neurosurg 1989;3:305-12. 65. Hitchcock E. Stereotactic neural transplantation. Stereotact Funct Neurosurg 1994;62:120-33. 66. Hitchcock ER, Kenny BG, Clough CG, Hughes RC, Henderson BT, Detta A. Stereotactic implantation of foetal mesencephalon (STIM): the UK experience. Prog Brain Res 1990;82:723-8. 67. Hitchcock ER, Clough CG, Hughes RC, Kenny BG. Transplantation in Parkinson’s disease: stereotactic implantation of adrenal medulla and foetal mesencephalon. Acta Neurochir Suppl (Wien) 1989;46:48-50. 68. Kudoh C, Detta A, Meyer C, Byrne P, Hitchcock E. Bilateral intracaudate cografting of fetal ventral mesencephalon and striatum in advanced Parkinson’s disease. Transplant Proc 1999;31:3397-402. 69. Freeman W, Watts JW. Psychosurgery in the treatment of mental disorders and intractable pain. 2nd ed. Springfield, IL: C. C. Thomas; 1950. 70. Moniz E. Tentatives Ope´ratoires dans le Traitement de Certaines Psychoses. Paris: Masson; 1936. 71. Richardson AE. Sir Wylie McKissock. Surg Neurol 1988;30:173-4. 72. Kitchen N. Neurosurgery for affective disorders at Atkinson Morley’s Hospital 1948–1994. Acta Neurochir Suppl 1995;64:64-8. 73. McKissock W. Rostral leucotomy. Lancet 1951;2:91-4. 74. McKissock W. Discussion on psychosurgery. Proc R Soc Med 1959;52:206-9. 75. Bell BA. Wylie McKissock – reminiscences of a commanding figure in British neurosurgery. Br J Neurosurg 1996;10:9-18.
76. El-Hai J. The lobotomist: a maverick medical genius and his tragic quest to rid the world of mental illness. Hoboken, NJ: Wiley; 2005. 77. Tooth J, Newton M. Leucotomy in England and Wales 1942–1954: report on public health and medical subjects no. 104. London: Her Majesty’s Stationery Office; 1961. 78. Knight GC. Discussion on psychosurgery. Proc R Soc Med 1959;52:209-10. 79. Knight GC, Tredgold RF. Orbital leucotomy; a review of 52 cases. Lancet 1955;268:981-6. 80. McCaul IR. Stereotaxic surgery and involuntary movement. Proc R Soc Med 1961;54:378-80. 81. Knight G. The orbital cortex as an objective in the surgical treatment of mental illness. The results of 450 cases of open operation and the development of the stereotactic approach. Br J Surg 1964;51:114-24. 82. Knight G. Stereotactic tractotomy in the surgical treatment of mental illness. J Neurol Neurosurg Psychiatry 1965;28:304-10. 83. Knight GC. Bi-frontal stereotactic tractotomy: an atraumatic operation of value in the treatment of intractable psychoneurosis. Br J Psychiatry 1969;115:257-66. 84. Knight G. Neurosurgical aspects of psychosurgery. Proc R Soc Med 1972;65:1099-104. 85. Knight G. Further observations from an experience of 660 cases of stereotactic tractotomy. Postgrad Med J 1973;49:845-54. 86. Bridges PK, Bartlett JR, Hale AS, Poynton AM, Malizia AL, Hodgkiss AD. Psychosurgery: stereotactic subcaudate tractomy. An indispensable treatment. Br J Psychiatry 1994;165:599-611; discussion 612-3. 87. Hodgkiss AD, Malizia AL, Bartlett JR, Bridges PK. Outcome after the psychosurgical operation of stereotactic subcaudate tractotomy, 1979–1991. J Neuropsychiatry Clin Neurosci 1995;7:230-4. 88. Malhi GS, Bartlett JR. A new lesion for the psychosurgical operation of stereotactic subcaudate tractotomy (SST). Br J Neurosurg 1998;12:335-9. 89. Malhi GS, Bartlett JR. Depression: a role for neurosurgery? Br J Neurosurg 2000;14:415-22; discussion 423. 90. Kelly D, Richardson A, Mitchell-Heggs N. Stereotactic limbic leucotomy: neurophysiological aspects and operative technique. Br J Psychiatry 1973;123:133-40. 91. Kelly D, Richardson A, Mitchell-Heggs N, Greenup J, Chen C, Hafner RJ. Stereotactic limbic leucotomy: a preliminary report on forty patients. Br J Psychiatry 1973;123:141-8. 92. Williams LM. Arousal dissociates amygdala and hippocampal fear responses: evidence from simultaneous fMRI and skin conductance recording. Neuroimage 2001;14:1070-9. 93. Christmas D, Matthews K, Eljamel MS. Neurosurgery for mental disorder. Br J Psychiatry 2004;185:173-4; author reply 174. 94. Feldman RP, Alterman RL, Goodrich JT. Contemporary psychosurgery and a look to the future. J Neurosurg 2001;95:944-56.
History of stereotactic surgery in great britain
95. Kopell BH, Machado AG, Rezai AR. Not your father’s lobotomy: psychiatric surgery revisited. Clin Neurosurg 2005;52:315-30. 96. Marks PV, Wild AM, Gleave JR. The ‘Cambridge’ stereotactic biopsy instrument. Br J Neurosurg 1990;4:127-8. 97. Wild AM, Xuereb JH, Marks PV, Gleave JR. Computerized tomographic stereotaxy in the management of 200 consecutive intracranial mass lesions. Analysis of indications, benefits and outcome. Br J Neurosurg 1990;4:407-15. 98. Bennett. A stereotaxic apparatus for use in cerebral surgery. Br J Radiol 1960;33:343–51. 99. Blandy JP, Greig CM, Gillam SJ, Williams DI. Lives of the fellows of the Royal College of Surgeons of England: 1997–2002: vol. 9 with a cumulated index to vol. 1–8. London: Royal College of Surgeons of England; 2005. 100. Eljamel MS, Forster A, Tulley M, Matthews K. Staged functional neurosurgery using image fusion: electronic atlas and microelectrode recording at Dundee. Stereotact Funct Neurosurg 1999;73:140-2. 101. Pell MF, Thomas DG. The initial experience with the Cosman-Roberts-Wells stereotactic system. Br J Neurosurg 1991;5:123-8. 102. Clarke RH. Part II: Atlas of the frontal sections of the cranium and brain of the rhesus monkey. Baltimore, MD: Johns Hopkins; 1920. 103. Clarke RH, Henderson EE. Atlas of photographs of sections of the frozen cranium and brain of the cat (Felis Domestica). J Psychol Neurol 1912;18:391-409. 104. Spiegel EA, Wycis HT, Baird HW. Studies in stereoencephalotomy. I. Topical relationships of subcortical structures to the posterior commissure. Confin Neurol 1952;12:121-33. 105. Schaltenbrand G, Bailey P. Introduction to stereotaxis with an atlas of the human brain. Stuttgart: Thienne; 1959. 106. Talairach J, David M, Tournoux P, Corredor H, Krasina T. Atlas d’Anatomie Stereotaxique. Paris: Masson; 1957. 107. Andrew J, Watkins ES. Stereotactic surgery. A stereotaxic atlas of the human thalamus. Baltimore, MD: Williams & Wilkins; 1969. 108. Afshar F, Watkins ES, Yap JC. Stereotaxic atlas of the human brainstem and cerebellar nuclei: a variability study. New York: Raven Press; 1978. 109. Mazzone P, Insola A, Lozano A, Galati S, Scarnati E, Peppe A, Stanzione P, Stefani A. Peripeduncular and pedunculopontine nuclei: a dispute on a clinically relevant target. Neuroreport 2007;18:1407-8. 110. Zrinzo L, Zrinzo LV, Hariz M. The pedunculopontine and peripeduncular nuclei: a tale of two structures. Brain 2007;130:e73; author reply e74. 111. Corsellis JA. Individual variation in the size of the tentorial opening. J Neurol Neurosurg Psychiatry 1958;21:279-83.
8
112. Watkins ES. On making stereotactic atlases. In: Gildenberg PL, Tasker RR, editors. Textbook of stereotactic and functional neurosurgery. New York: McGraw Hill; 1998. p. 235-6. 113. Cotzias GC, Van WM, Schiffer LM. Aromatic amino acids and modification of parkinsonism. N Engl J Med 1967;276:374-9. 114. Petrik V, Apok V, Britton JA, Bell BA, Papadopoulos MC. Godfrey Hounsfield and the dawn of computed tomography. Neurosurgery 2006;58:780-7; discussion 780-7. 115. Ambrose J, Gooding MR, Richardson AE. The Lancet – saturday II october 1975. Sodium iothalamate as an aid to diagnosis of intracranial lesions by computerised transverse axial scanning. Lancet 1975;2:669-74. 116. Ambrose J. CT scanning: a backward look. Semin Roentgenol 1977;12:7-11. 117. Ambrose J, Hounsfield G. Computerized transverse axial tomography. Br J Radiol 1973;46:148-9. 118. Hounsfield GN. Historical notes on computerized axial tomography. J Can Assoc Radiol 1976;27:135-42. 119. Hounsfield GN. Computed medical imaging. Science 1980;210:22-8. 120. Hitchcock E. Stereotactic-computerized tomography interface device. Appl Neurophysiol 1987;50:63-7. 121. Leksell L. A stereotaxic apparatus for intracerebral surgery. Acta Chir Scand 1949;99:229-33. 122. Leksell L. The stereotaxic method and radiosurgery of the brain. Acta Chir Scand 1951;102:316-9. 123. Rowe JG, Radatz MW, Walton L, Kemeny AA. Changing utilization of stereotactic radiosurgery in the UK: the Sheffield experience. Br J Neurosurg 2002;16:477-82. 124. Benabid AL, Pollak P, Louveau A, Henry S, de Rougement J. Combined (thalamotomy and stimulation) stereotactic surgery of the VIM thalamic nucleus for bilateral Parkinson disease. Appl Neurophysiol 1987;50:344-6. 125. Laitinen LV, Bergenheim AT, Hariz MI. Leksell’s posteroventral pallidotomy in the treatment of Parkinson’s disease. J Neurosurg 1992;76:53-61. 126. Aziz TZ, Adams CB. Neurosurgery at the radcliffe infirmary, Oxford: a history. Neurosurgery 1995;37:505-10. 127. Bergman H, Wichmann T, DeLong MR. Reversal of experimental parkinsonism by lesions of the subthalamic nucleus. Science 1990;249:1436-8. 128. Aziz TZ, Peggs D, Sambrook MA, Crossman AR. Lesion of the subthalamic nucleus for the alleviation of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)induced parkinsonism in the primate. Mov Disord 1991;6:288-92. 129. Aziz TZ, Peggs D, Agarwal E, Sambrook MA, Crossman AR. Subthalamic nucleotomy alleviates parkinsonism in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-exposed primate. Br J Neurosurg 1992;6:575-82.
93
94
8
History of stereotactic surgery in great britain
130. Parkin S, Nandi D, Giladi N, Joint C, Gregory R, Bain P, Scott R, Aziz TZ. Lesioning the subthalamic nucleus in the treatment of Parkinson’s disease. Stereotact Funct Neurosurg 2002;77:68-72. 131. Parkin SG, Gregory RP, Scott R, Bain P, Silburn P, Hall B, Boyle R, Joint C, Aziz TZ. Unilateral and bilateral pallidotomy for idiopathic Parkinson’s disease: a case series of 115 patients. Mov Disord 2002;17:682-92. 132. Liu X, Rowe J, Nandi D, Hayward G, Parkin S, Stein J, Aziz T. Localisation of the subthalamic nucleus using radionics image Fusion? and Stereoplan? combined with field potential recording: a technical note. Stereotact Funct Neurosurg 2001;76:63-73. 133. Aziz TZ, Nandi D, Parkin S, Liu X, Giladi N, Bain P, Gregory RG, Joint C, Scott RB, Stein JF. Targeting the subthalamic nucleus. Stereotact Funct Neurosurg 2002;77:87-90. 134. Aziz T, Torrens M. CT-guided thalamotomy in the treatment of movement disorders. Br J Neurosurg 1989;3:333-6. 135. Pereira EA, Green AL, Nandi D, Aziz TZ. Deep brain stimulation: indications and evidence. Expert Rev Med Devices 2007;4:591-603. 136. Munro-Davies LE, Winter J, Aziz TZ, Stein JF. The role of the pedunculopontine region in basal-ganglia mechanisms of akinesia. Exp Brain Res 1999;129:511-7. 137. Nandi D, Aziz TZ, Giladi N, Winter J, Stein JF. Reversal of akinesia in experimental parkinsonism by GABA antagonist microinjections in the pedunculopontine nucleus. Brain 2002;125:2418-30. 138. Jenkinson N, Nandi D, Miall RC, Stein JF, Aziz TZ. Pedunculopontine nucleus stimulation improves akinesia in a parkinsonian monkey. NeuroReport 2004;15:2621-4. 139. Nandi D, Jenkinson E, Miall C, Stein JF, Aziz TZ. Pedunculopontine nucleus. J Neurosurg 2004;100:978-9; author reply 979. 140. Kleiner-Fisman G, Herzog J, Fisman DN, Tamma F, Lyons KE, Pahwa R, Lang AE, Deuschl G. Subthalamic nucleus deep brain stimulation: summary and metaanalysis of outcomes. Mov Disord 2006;21 Suppl 14: S290-304. 141. Stefani A, Lozano AM, Peppe A, Stanzione P, Galati S, Tropepi D, Pierantozzi M, Brusa L, Scarnati E, Mazzone P. Bilateral deep brain stimulation of the pedunculopontine and subthalamic nuclei in severe Parkinson’s disease. Brain 2007;130(6):1596-1607. 142. Schupbach WM, Chastan N, Welter ML, Houeto JL, Mesnage V, Bonnet AM, Czernecki V, Maltete D, Hartmann A, Mallet L, Pidoux B, Dormont D, Navarro S, Cornu P, Mallet A, Agid Y. Stimulation of the subthalamic nucleus in Parkinson’s disease: a 5 year follow up. J Neurol Neurosurg Psychiatry 2005;76:1640-4. 143. Rodriguez-Oroz MC, Obeso JA, Lang AE, Houeto JL, Pollak P, Rehncrona S, Kulisevsky J, Albanese A,
144.
145.
146.
147.
148.
149.
150.
151.
152.
153.
154.
155.
Volkmann J, Hariz MI, Quinn NP, Speelman JD, Guridi J, Zamarbide I, Gironell A, Molet J, Pascual-Sedano B, Pidoux B, Bonnet AM, Agid Y, Xie J, Benabid AL, Lozano AM, Saint-Cyr J, Romito L, Contarino MF, Scerrati M, Fraix V, Van Blercom N. Bilateral deep brain stimulation in Parkinson’s disease: a multicentre study with 4 years follow-up. Brain 2005; 128:2240-9. Liu X, Miall RC, Aziz TZ, Palace JA, Stein JF. Distal versus proximal arm tremor in multiple sclerosis assessed by visually guided tracking tasks. J Neurol Neurosurg Psychiatry 1999;66:43-7. Liu X, Yianni J, Wang S, Bain PG, Stein JF, Aziz TZ. Different mechanisms may generate sustained hypertonic and rhythmic bursting muscle activity in idiopathic dystonia. Exp Neurol 2006;198:204-13. Wang S, Liu X, Yianni J, Green AL, Joint C, Stein JF, Bain PG, Gregory R, Aziz TZ. Use of surface electromyography to assess and select patients with idiopathic dystonia for bilateral pallidal stimulation. J Neurosurg 2006;105:21-5. Pereira EA, Green AL, Bradley KM, Soper N, Moir L, Stein JF, Aziz TZ. Regional cerebral perfusion differences between periventricular grey, thalamic and dual target deep brain stimulation for chronic neuropathic pain. Stereotact Funct Neurosurg 2007;85:175-83. Kringelbach ML, Jenkinson N, Green AL, Owen SL, Hansen PC, Cornelissen PL, Holliday IE, Stein J, Aziz TZ. Deep brain stimulation for chronic pain investigated with magnetoencephalography. Neuroreport 2007;18:223-8. Muthusamy KA, Aravamuthan BR, Kringelbach ML, Jenkinson N, Voets NL, Johansen-Berg H, Stein JF, Aziz TZ. Connectivity of the human pedunculopontine nucleus region and diffusion tensor imaging in surgical targeting. J Neurosurg 2007;107:814-20. Aravamuthan BR, Muthusamy KA, Stein JF, Aziz TZ, Johansen-Berg H. Topography of cortical and subcortical connections of the human pedunculopontine and subthalamic nuclei. Neuroimage 2007;37:694-705. Green AL, Wang S, Owen SL, Paterson DJ, Stein JF, Aziz TZ. Controlling the heart via the brain: a potential new therapy for orthostatic hypotension. Neurosurgery 2006;58:1176-83; discussion 1176‐83. Green AL, Wang S, Owen SL, Xie K, Bittar RG, Stein JF, Paterson DJ, Aziz TZ. Stimulating the human midbrain to reveal the link between pain and blood pressure. Pain 2006;124:349-59. Yianni J, Bain P, Giladi N, Auca M, Gregory R, Joint C, Nandi D, Stein J, Scott R, Aziz T. Globus pallidus internus deep brain stimulation for dystonic conditions: a prospective audit. Mov Disord 2003;18:436-42. Owen SL, Green AL, Nandi DD, Bittar RG, Wang S, Aziz TZ. Deep brain stimulation for neuropathic pain. Acta Neurochir Suppl 2007;97:111-16. Kitchen ND, Thomas DG. Minimally invasive stereotaxy: clinical use of the Gill-Thomas-Cosman (GTC)
History of stereotactic surgery in great britain
156.
157.
158.
159.
160.
161.
162.
repeat stereotactic localiser. Minim Invasive Neurosurg 1994;37:61-3. Sofat A, Kratimenos G, Thomas DG. Early experience with the Gill Thomas locator for computed tomographydirected stereotactic biopsy of intracranial lesions. Neurosurgery 1992;31:972-4. Gill SS, Patel NK, Hotton GR, O’Sullivan K, McCarter R, Bunnage M, Brooks DJ, Svendsen CN, Heywood P. Direct brain infusion of glial cell line-derived neurotrophic factor in Parkinson disease. Nat Med 2003;9:589-95. Love S, Plaha P, Patel NK, Hotton GR, Brooks DJ, Gill SS. Glial cell line-derived neurotrophic factor induces neuronal sprouting in human brain. Nat Med 2005;11:703-4. Patel NK, Bunnage M, Plaha P, Svendsen CN, Heywood P, Gill SS. Intraputamenal infusion of glial cell line-derived neurotrophic factor in PD: a two-year outcome study. Ann Neurol 2005;57:298-302. Plaha P, Gill SS. Bilateral deep brain stimulation of the pedunculopontine nucleus for Parkinson’s disease. Neuroreport 2005;16:1883-7. Page RD, Miles JB. Validation of CT targeting for functional stereotaxis with postoperative magnetic resonance imaging. Br J Neurosurg 1994;8:461-7. Dervin JE, Miles J. A simple system for image directed stereotaxis. Br J Neurosurg 1989;3:569-74.
8
163. Dervin JE, Miles JB. Development of an analogue method to link stereotactic surgery to computed tomography. Neurochirurgia (Stuttg) 1984;27:162-5. 164. Dorward NL, Alberti O, Palmer JD, Kitchen ND, Thomas DG. Accuracy of true frameless stereotaxy: in vivo measurement and laboratory phantom studies. Technical note. J Neurosurg 1999;90:160-8. 165. Thomas DG, Kitchen ND. Stereotactic techniques for brain biopsies. Arch Dis Child 1993;69:621-2. 166. Kitchen ND, Lemieux L, Thomas DG. Accuracy in frame-based and frameless stereotaxy. Stereotact Funct Neurosurg 1993;61:195-206. 167. Thomas DG, Nouby RM. Experience in 300 cases of CTdirected stereotactic surgery for lesion biopsy and aspiration of haematoma. Br J Neurosurg 1989;3:321-5. 168. Thomas DG, Anderson RE, du Boulay GH. CT-guided stereotactic neurosurgery: experience in 24 cases with a new stereotactic system. J Neurol Neurosurg Psychiatry 1984;47:9-16. 169. Powell M, Kitchen N. The development of neurosurgery at the National Hospital for Neurology and Neurosurgery, Queen Square, London, England. Neurosurgery 2007;61:1077-90; discussion 1090. 170. Gildenberg PL. Stereotactic surgery: the early years (comment). Neurosurgery 2004;55:1214.
95
12 History of Stereotactic Surgery in India P. K. Doshi
Ancient Indian Literature The earliest reference to any functional neurosurgery in world could be found in Indian mythology script Shiva Purana (http://is1.mum.edu/vedicreserve/puran.htm). It reflects a transplantation of elephant head on a human being (Ganesha). Ganesha – the elephant-deity riding a mouse – has become one of the most common mnemonics for anything associated with Hinduism. The son of Shiva and Parvati, Ganesha has an elephantine countenance with a curved trunk and big ears, and a huge pot-bellied body of a human being. (> Figure 12-1) He is worshipped as Lord of success and destroyer of evils and obstacles. He is also worshipped as the god of education, knowledge, wisdom, and wealth. The story of the birth of this zoomorphic deity, as depicted in the Shiva Purana, goes like this: Once goddess Parvati, while bathing, created a boy out of the dirt of her body and assigned him the task of guarding the entrance to her bathroom. When Shiva, her husband returned, he was surprised to find a stranger denying him access, and struck off the boy’s head in rage. Parvati broke down in utter grief and to soothe her, Shiva sent out his squad (gana) to fetch the head of any sleeping being who was facing the north. The company found a sleeping elephant and brought back its severed head, which was then attached to the body of the boy. Shiva restored its life and made him the leader (pati) of his troop, hence his name ‘‘Ganapati.’’ Shiva also bestowed a boon that people would worship him and invoke his name before undertaking any venture. #
Springer-Verlag Berlin/Heidelberg 2009
This narration highlights two important aspects of functional neurosurgery. One is the reproduction of a human being from dermis derived stem cell (‘‘created a boy out of the dirt of her body’’) and second, the ultimate aim/ achievement that could ever happen – ‘‘whole head transplant.’’ In another epic, it is documented that in 1800 BC, Jivaka (physician to the Lord Buddha) removed intracranial mass lesion through trephination.
Development of Stereotactic Surgery Neurosurgery in India is a post World War II development, resulting from the keen desire of the new rulers of independent India, that the country should keep up with all the modern advances in every field of Neurosurgery [1]. It is interesting to note that many of the pioneers of Indian neurosurgery were exposed to stereotactic and functional neurosurgery, during their training abroad, which helped them to develop stereotactic neurosurgical specialty parallel with the international advances. Stereotactic surgery developed in parallel across several centers in India. In this chapter, the development of each subset of functional neurosurgery is described separately along with the contributions of each center, rather than following the chronology of date. Most important references have been included but these are not necessarily all encompassing.
156
12
History of stereotactic surgery in india
. Figure 12-1 Lord Ganesha (Elephant God)
In the 1940s bold pioneers like Chintan Nambiar, a surgeon at Stanley medical college, Madras, used to perform freehand stereotactic lesions by using a template in the temporal region. While mentioning this in an oration, the pioneering stereotactic surgeon B Ramamurthi quotes ‘‘The surgeon was bold and the patients bolder.’’ He performed 74 cases of chemopallidectomy using this free hand technique, out of which 27 had excellent results [2]. The first neurosurgical set up was established at the Christian Medical College (CMC), Vellore, in Tamilnadu, by Jacob Chandy. Chandy had obtained 2 years of training under Wilder Penfield at the Montreal Neurological Institute (MNI). In January 1949, equipped with neurosurgical equipments brought from Canada and an EEG machine, Chandy started the neurosurgical unit at CMC. He started with ten beds spread across medical and surgical wards. In 1950, he
was joined by Baldev Singh, a neurologist. In 1962, CMC acquired the Bertrand stereotactic guide, with the help of which they performed surgeries for Parkinson’s disease and epilepsies. Later in 1987, KV Mathai procured BRW frame. CMC became an established stereotactic unit, where till date around 1,800 stereotactic biopsies, 400 stereotactic craniotomies, 100 functional neurosurgical procedures, and 700 radiosurgical procedures have been performed. Vedantam Rajashekar, the present Head of the Department of Neurosurgery, and past president of the Indian Society of Stereotactic and Functional Neurosurgery, has been conducting stereotactic workshop to train young neurosurgeons [1]. Following on the heels of the department at Vellore was that at the Madras Medical College and Government Hospital; with the joining of Ramamurthi, in October 1950. Ramamurthi started his neurosurgical training with Rowbotham, in Newcastle upon Tyne, England. Subsequently he visited numerous centers across Europe and USA to gain wider experience. He also visited MNI and observed Rasmussen’s and Penfield’s work. Ramamurthi started neurosurgical unit with four beds which were increased to ten after 18 months [1]. Inspired by Irving Cooper’s use of an inflatable balloon to make lesions in pallidum for the treatment of Parkinson’s diseases, V Balasubramaniam and Ramamurthi performed surgeries using Cooper’s balloon (1962) [3]. Under radiological guidance a balloon was introduced and left in place for 48 h. This was followed by alcohol ablation. After performing surgeries on 12 patients of movement disorders, they were discouraged with the results. Import restrictions and bureaucratic hurdles made further imports of these balloons difficult and hence they gave up this method. In 1960 Ramamurthi received an invitation for dinner with the Governor of Hyderabad, General Shrinagesh (> Figure 12-2). It happened that the Governor was suffering from Parkinson’s disease and had undergone unilateral lesion in London by Lawrence Walsh at Atkinson Morley Hospital.
History of stereotactic surgery in india
12
. Figure 12-2 Ramamurthi having dinner with, General Shrinagesh (Black Suit), Governor of Andhra Pradesh
He asked Ramamurthi if such facilities are available in India as he had started experiencing symptoms on the opposite side. Ramamurthi explained that though he had the necessary expertise, the equipments were not available. In those days it was very difficult to import any equipment or obtain foreign exchange to travel abroad. General Shrinagesh immediately called up the Prime Minister of India, Pandit Jawaharlal Nehru; it was 11 PM. The Prime Minister immediately agreed to Shrinagesh’s suggestion to bring in Lawrence Walsh and Denis Williams (Neurologist) from England to conduct a workshop and training for movement disorders surgery. They also obtained permission for Walsh to leave his equipment behind after the workshop. Walsh and Williams came and stayed in Madras for 3 weeks during which they performed 40 surgeries and 30 neuroscientists took advantage of their expertise (> Figure 12-3). On completion of the program as decided, they left the Leksell stereotactic apparatus and the lesion generator back in Madras [4]. This was a major impetus and from there on Madras became a leading stereotactic center where more than 1,700 procedures were performed between 1959 and 1975 (> Table 12-1). Presently very little work is being
done at this center after Ramamurthi and his team retired from the Madras Medical College. In 1970, S Kalyanaraman, a young neurosurgeon in Ramamurthi’s team from the Madras Medical College, reported simultaneous use of two stereotactic apparatus on the same patient. He used the Leksell stereotactic equipment in combination with the Sehgal stereotactic equipment to perform simultaneous targeting of intracranial structures. Sehgal’s stereotactic equipment is a compact, burr hole based stereotactic device designed by Arjun Sehgal in India. The Leksell frame was used to align Sehgal’s apparatus to the target on one side and on the other it was used to approach the target. The purpose was to obtain simultaneous recording from the thalamus, thus reducing the operative time. They also observed that the number of X-rays required to localize the targets were reduced [5,6]. RM Varma started a neurosurgical unit in the (then) All India Institute of Medical Health, Bangalore, in 1958. Varma’s efforts led to the formation of the National Institute of Mental Health and Neurosciences (NIMHANS) out of this unit [7]. Varma was trained in Bristol and started performing lesions for Parkinson’s disease using a unique free hand technique, through
157
158
12
History of stereotactic surgery in india
. Figure 12-3 Lawrence Walsh, Chief Nurse, Ramamurthi and Dennis Williams
. Table 12-1 Stereotactic Surgery at Madras Medical College (1959– 1975) Thalamotomy Amygdalotomy Hypothalamotomy Cingulumotomy Basofrontal tractotomy Dentatectomy Leucotomy Thalamolaminotomy Capsulotomy Pulvinotomy Mesencephalc reticulotomy Hypophysectomy
858 480 122 143 56 73 5 11 16 4 2 2
foramen ovale. Later on in 1980s, they obtained the Leksell stereotactic equipment and recently a gamma knife unit. HM Dastur, started stereotactic surgery at King Edward Memorial (KEM) Hospital, Bombay (Mumbai) in 1959. The initial surgeries were performed using Oliver’s guide. Narabayashi visited KEM hospital in 1962 and lent the design of his stereotactic frame for fabricating a local frame on similar lines. Later on in 1975, Dastur joined Jaslok Hospital, Bombay (Mumbai), where he continued to perform stereotactic surgery using
a Reichert-Mundinger frame. In another center at Bombay Hospital, SN Bhagwati started stereotactic surgery with Mckinney’s apparatus from 1962 and this was replaced by Leksell’s frame in 1964. Multiple centers started practicing stereotactic surgery in the 1970s. These include Post Graduate Institute of Medical Education and Research (PGIMER), Chandigarh (1974), with Mckinney frame, All India Institute of Medical Sciences (AIIMS), New Delhi (1977), with Leksell frame and Shri Chitra Tirunal Institute for Medical Sciences and Technology (SCTIMST) with Leksell frame [2,8]. In 1991 Mr. Bose, an engineer and Apte, a neurosurgeon from Pune, developed an indigenous arc centered stereotactic apparatus. Three revisions have been made to this initial model, which is now compatible for CT and MRI guided procedures. They followed this by manufacturing a radiofrequency lesion generator in 1996. The prices of these equipments are considerably lower than standard stereotactic frames and hence have become popular for performing biopsies and other stereotactic procedures. Presently 40 neurosurgical units have stereotactic equipment and
History of stereotactic surgery in india
they perform various levels of stereotactic procedures from biopsy to functional neurosurgery. The Indian Society of Stereotactic and Functional Neurosurgery was formed in 1997. Its first meeting was held in Delhi with Balasubramaniam as president and Rajashekar as Secretary. In 2007 the tenth meeting of the society was held in Kolkata. This is a rapidly expanding society with the current membership of more than 100 members. Stereotactic radiosurgery was first introduced in India at the Apollo Hospitals, Chennai, using Linac based X-knife system. Linac based radiosurgery is offered at multiple centers including Bombay Hospital and Jaslok Hospital, Mumbai. Gamma Kinfe was introduced at Hinduja Hospital, Mumbai, in 1997 [9]. This was soon followed by similar units at AIIMS and VIMHANS in New Delhi, and later on at other centers including Vellore, NIMHANS, R&R (Army Hospital, New Delhi), PGIMER (Chandigarh) etc.
Epilepsy Surgery In the early days of stereotactic and functional neurosurgery the enthusiasm of the neuroscientists could not be contained. Virtually lesions were placed in every part of the brain for varied disorders ranging from epilepsy to psychiatric illness [10]. To further understand the development drivers of these surgeries, we need to look into India’s social and economical background of that period. A recent (2006) survey (http://www. prb.org/Articles/2006/CommunityBasedHealth InsuranceShowsPromiseinIndia.aspx) showed that only 11% of Indian population had some kind of health insurance, this could not have been more than 2–4% in 1960s. Thus the medical treatment had to be funded by the patient themselves or they had to use government or municipal hospitals to treat them. Though these hospitals were supposed to be providing free medical treatment, the medicines (especially the expensive ones)
12
had to be purchased by the patient. Paradoxically the cost of surgeries in these hospitals was far less than prolonged medical treatment. In a report on epilepsy research, Mathai [11] mentions that follow-up studies indicated that where the person was receiving more than two drugs the cost of the therapy would come >2 $/month and only 10–15% of the patients had the resources to afford this. Hence in developing countries financial insufficiency formed an added indication for surgical therapy in the control of seizures. Chandy, who had trained at the MNI, started epilepsy surgery in CMC, Vellore in 1949. He was joined by Baldev Singh, a neurologist, who had interest in epilepsy and was equipped with an EEG machine to diagnose and manage epilepsy. Singh and Chandy [12] in an exhaustive study of delta waves in 800 EEG records at CMC, commented on their characteristics and localizing value. Those were the days when EEG was the only non-invasive investigation for brain disorders (including tumors). During 1949–1990, 141 epilepsy surgeries were performed at CMC [13]. They performed topectomy and lobectomy for suprasylvian epilepsy; for temporal lobe epilepsy (TLE) the surgical procedures done were topectomy, temporal lobectomy with amygdalectomy, temporal lobectomy with amygdalohippocampectomy, and only amygdalectomy. Hemispherectomy was done for cases with multilobar epilepsy. Temporal lobe resections were done based on the scalp EEG, sphenoidal studies, neuropsychological assessment and the intraoperative ECoG and depth electrode studies. Total or near total seizure control was obtained in 53% patients and a satisfactory outcome in 20% patients. They found that mental retardation, pre operative scalp EEG and post excision electrocorticography were predictors of outcome. In Madras, Ramamurthi developed an excellent team. He along with Balasubramaniam, Kalyanaraman and Kanaka as neurosurgical colleagues; Arjundas, Jaganathan and later on Sayeed as neurological colleagues, Vriddhagirinathan the
159
160
12
History of stereotactic surgery in india
psychologist and Valmikinathan, neurochemist developed a comprehensive epilepsy surgical program. They reviewed the literature and noted that Falconer had obtained good results by standard temporal lobectomies in three fourth of the patients [14]. Narabayashi and Chitanondh had reported 50% seizure control and improvement in behavioral disorders in patients undergoing amygdalectomy. Ramamurthi felt that massive temporal lobectomy, only to ablate these medial structures, is a mutilating procedure. They treated complex partial seizures with proved medial temporal focus with stereotactic lesions rather than by a full temporal lobectomy [15–19]. They established the details of localization of epileptic focus by careful pre and postoperative observations of their neurological colleagues. In a paper published in 1979 [20], Ramamurthi shares his experience of 56 cases that he operated. The patients included, suffered from purely TLE, TLE with secondary generalization or TLE with focal seizures. EEG studies including sphenoidal and depth electrode recordings were performed prior to surgery. Based on these findings, stereotactic lesions of 600–800 m3, were made in the area of maximum abnormality. He found that 26 of the 56 patients became seizure free and eight cases required reoperation. He commented that this procedure was useful in patients who had bilateral TLE or contralateral previous temporal lobectomy, as it preserved the hippocampus and thus the memory. Mathai, in CMC Vellore, also performed amygdalectomy through a craniotomy for similar reasons. They performed intraoperative corticography and depth studies from amygdala to plan their surgical resections. They concluded that when the seizure discharges are from the cortex rather than amygdala, excision of the amygdala only reduces the intensity of cortical activity. However, when the epileptiform discharges are primarily from the amygdala, amygdalectomy alone might suffice; especially when an electrocorticographic seizure can be produced by stimulating the
amygdala [11]. In Bombay (Mumbai), at KEM hospital Dastur, assisted by neurologist Anil Desai, performed epilepsy surgery including temporal lobectomy and hemispherectomy. He continued his work after joining Jaslok Hospital, where he was assisted by Mrs. PN Wadia, neurophysiologist in performing epilepsy surgery. Corticography and depth recordings during the surgery were also performed. During 1960s depth electrodes and corticography were used routinely for epilepsy surgeries. Arjundas observed that the depth electrode supplements information obtained from scalp EEG in identifying the epileptogenic focus and also reveals foci not evident in scalp EEG records [21]. Kanaka and Balasubramaniam found that depth studies were useful in understanding the propagation of epileptic discharges. They found that depth electrode study along with electrocorticography also provided precise localization of the epileptic focus. They used this information for planning appropriate surgery [22]. Kalyanaraman had obtained PhD degree from Edinburgh, University on ‘‘Anatomical and Physiological studies on the internal capsule and adjacent diencephalic structures during human stereotaxy.’’ Based on his work and the observations of Gillingham [23], Kalyanaraman postulated that in an area in the posterior limb of the internal capsule, medial to the pyramidal tract where lie the corticospinal fibers that form the common pathway for the epileptic seizure. A bilateral internal capsulotomy should control clinical seizures without causing pyramidal signs and without producing electroencephalographic improvement. To test this hypothesis, they performed bilateral internal capsulotomy on seven patients with intractable grand mal epilepsy. They had good outcome (comparable to Engel grade I and II) in three patients, considerable reduction in seizure frequency in one and poor outcome in two. One patient died of pneumonia postoperatively. None of these patients had significant permanent morbidity [24,25].
History of stereotactic surgery in india
Mathai postulated that in patients suffering from generalized seizures or focal seizures arising from significant (eloquent) cortical areas, which cannot be excised without producing serious neurological deficit, interruption of propagation pathways may modify the frequency and pattern of the seizures. They performed stereotactic lesioning of ansa and fasciculus lenticularis. Bertrand stereotactic guide was used and the target chosen was 1 cm behind AC and 1.5 cm lateral. A leucotome was used to make a lesion of 0.5–0.8 cm. Scalp and depth EEG recordings were obtained during this stage. They had variable results, ranging from good seizure control to reduction in seizure frequency and severity. They concluded that because of the multiple propagation pathways in epilepsy these procedures may only temporarily modify the clinical seizure pattern. However, in seizures where the frequency is once a day or more such procedures are of value. Destruction of ansa and fasciculus lenticularis bilaterally for generalized cerebral seizures and unilaterally for focal cortical seizures seems to alleviate, although not fully, the intensity and frequency of seizures [26]. Few epilepsy surgeries were performed during 1970–1990. Radhakrishnan, epileptologist, started an epilepsy surgery program at the SCTIMST. This was a complete program with facilities for invasive recording, neuropsychology, video EEG and nuclear medicine. Their initial focus was on temporal lobe epilepsy. From March 1995 through February 2002, they performed 394 epilepsy surgeries, 370 of them were anterior temporal lobectomy with amygdalohippocampectomy for refractory temporal lobe epilepsy. They reported 78% seizure freedom at 2 years follow-up [27]. Epilepsy surgery was started in AIIMS, by VP Singh [28], at Jaslok Hospital by P Doshi and at Hinduja Hospital by CE Deopujari and BK Misra in the late 1990s. All these centers have facilities of state of the art neuroradiology, neurophysiology
12
(including prolonged video EEG), nuclear medicine departments for SPECT and PET scans, neuropsychologist, epileptologist and functional neurosurgeon.
Movement Disorders Surgery Movement disorders surgery has had the most colorful history in any account of functional neurosurgery. This is true for India as well. Enterprising neurosurgeons used varied techniques from free hand technique to frame based systems, unilateral and bilateral lesions, even simultaneous bilateral lesions; alcohol to radiofrequency lesioning and from pallidum to field of Forel, to ameliorate movement disorders. Balasubramaniam and Ramamurthi started performing chemopallidectomy using Cooper’s balloon in 1962 as mentioned earlier [3]. From 1964 they used the Leksell’s apparatus to perform thermal lesions for movement disorders [29]. Kalyanaraman, performed bilateral simultaneous thalamotomies in patients with various movement disorders. They observed that the complication rate was acceptable and no greater than staged procedures in patients with advanced Parkinson’s disease. In patients with bilateral intention tremors this formed a good surgical option as it avoided double hospitalization [30]. Using Sehgal’s stereotactic apparatus he used to perform bilateral simultaneous recordings as described earlier. During surgery an opaque marker was introduced into the site of the lesion. Immediate postoperative X-rays were taken and the exact location of the lesion was charted with the help of the atlas prepared by Schaltendrand and Bailey. This served not only to assess the accuracy of lesion placement but also to correlate the result of surgery with the site of the lesion. They noted minor differences in the anatomical calculations of deep brain structures in different races and groups [31].
161
162
12
History of stereotactic surgery in india
Though in 1970s, the number of surgeries for Parkinson’s disease started decreasing, due to the cost constraints of prolonged Levodopa therapy excellent surgical benefits (especially for unilateral disease) Madras group continued to perform surgeries for Parkinson’s disease. Another unique area of interests was the treatment for cerebral palsy. Following the work of Narabayashi [32], Balasubramaniam and his colleagues operated on a large number of cerebral palsy patients. As their experience evolved they chose different targets depending on the predominant symptom complex. For Rigidity they made a lesion in the area below the ventrolateral nucleus (VL); for dyskinesias they used variety of targets including the ventralis intermedius nucleus (VIM), the centromedian nucleus (CM) and the dentate nucleus of the cerebellum. As most of these surgeries were done under general anesthesia verification of the electrode placement by stimulation, as done in Parkinson’s disease was not possible. Stimulation was still done to exclude electrode placement in the corticospinal tract [33]. They later on introduced stereotactic dentatectomy for patients with predominant spasticity. They found that VL and sub VL lesions were effective for rigidity, whereas for patients with a mix of rigidity and spasticity these lesions had to be supplemented by dentatectomy. Patients with sensory induced involuntary movements benefited from centromedian thalamotomy [34]. For severe hyperkinetic disorders, Kanaka found hypothalamotomy to play a distinct role in their management. She observed that it works because the area destroyed forms part of the limbic system. It seemed to be more on the ‘‘effector’’ side. It does not cause any morbidity. In the management hyperkinetic behavior disorders the first target to be destroyed must be the amygdaloid nucleus. If this operation fails, then hypothalamotomy may be done as the next operation [35]. Varma (1964) in Bangalore developed a free hand technique of lesioning the thalamus for
Parkinson’s disease. He modified the technique of Arthur Ecker and Theoder Perl [36] for this surgery. The technique involved use of two needles, outer 19G, to cannulate the foramen ovale; and the inner, 26G to perform chemothalamotomy. The outer needle had a slight curve at the end to help direct the inner needle to the desired position (> Figure 12-4). Varma used external landmarks to do away with ventriculography. He used lead pellets placed on the external canthus of the eye and in the internal auditory meatus to serve as markers. Lateral and AP radiographs were obtained after the introduction of the needle. A topograph was created outlining the above landmarks in relation to the venterolateral thalamus (based on an atlas). This was then overlain on the lateral x-ray and the distance between the needle and the target obtained. Appropriate radiographic and anthropometric corrections were applied to calculate the relationship of the needle to the target and the final position of needle adjusted. Varma notes that in many patients the tremor used to get arrested when the needle reached the target. This was then followed by chemothalamotomy with absolute alcohol [37]. His work was later on reviewed with MRI imaging of the patients operated, by . Figure 12-4 Varma’s Foramen Ovale ‘‘Thalamotomy’’ for PD. Outer and inner needle seen in situ, with cranial landmarks outlined
History of stereotactic surgery in india
Uday Muthane [38,39]. Muthane analyzed the site of the lesion by MRI and, in one case, postmortem examination. He noted that the lesion was actually placed 1.5 cm below the thalamus and it coincided with the subthalamic nucleus. Stereotactic surgery was performed initially at CMC, Vellore by the free hand technique and later from 1961 onwards using Bertrand’s frame. Parkinson’s disease and dystonia were the main indications. The initial target used was globus pallidum, which was, later on changed to the thalamus. The free hand technique involved localizing the foramen of Monro using pneumoenchephalogram and pallidal target was calculated based on stereotactic atlas derived coordinates. AP and lateral radiographs were used for localization. The lesions were made using absolute alcohol in incremental methods whilst checking for neurological deficits (Mathai KV, personal communication). In Bombay (Mumbai), Dastur started performing lesions in thalamus, pallidum and field of Forel for dystonia at the KEM hospital in 1959 (Dastur HM, personal communication). Desai, neurologist associated with KEM neurosurgical department, visited Narabayashi in 1962, to assist Dastur in performing movement disorders surgery. The program received further impetus following Narabayashi’s visit. Dastur recounts a very interesting experience. They planned to perform a Cooper’s lesion for a patient with severe ‘‘hyperkinesis.’’ Following surgery, the patient had a remarkable improvement. However, on postoperative x-ray analysis they found that their lesions were not in the intended area but the lesions were more in the ventrolateral thalamus above the CA-CP plane. Gajendra Sinh, neurosurgeon at the Jaslok Hospital, used to call this a KEM lesion. The lesions were made using myodil and wax (> Figure 12-5). In 1974 Sinh invited Laitinen to visit Jaslok Hospital. Laitinen demonstrated his pallidotomy operations, which were then followed up by Sinh. After the 1980s movement disorders surgery
12
. Figure 12-5 Notings of ‘‘KEM thalamotomy’’ by Dastur
almost stopped following remarkable improvement in medical management. In another unit in Bombay (Mumbai), Bhagwati and his colleagues performed thalamotomies for Parkinson’s disease in 110 cases. They were also convinced about the efficacy of bilateral lesions and 30 of these patients had undergone bilateral lesions. They reported an interesting observation of reactivation of successfully abolished tremors on thalamic stimulation whilst performing surgery for the second side in Parkinson’s disease. In five of the thirty patients who underwent staged (minimum interval 6 months) bilateral thalamotomy, tremors recurred on the side of the second thalamotomy, i.e., on the ipsilateral side. In three patients, they performed a repeat lesion for the control of these tremors, but the patients had a rather stormy convalescence and developed drowsiness, confusion,
163
164
12
History of stereotactic surgery in india
dysarthria with one of them developing pseudobulbar palsy, in the next two patients simultaneous bilateral lesions were not made. The recurrent tremors persisted in these two patients [40]. Kalyanaraman also had similar experience in patients undergoing bilateral thalamotomy. Bhagwati earlier used diathermy to make lesions but later on switched over to cryo lesioning [41]. Once again, like other places in the world, interest in Parkinson’s disease surgery waned after the introduction of Levodopa. Surgery for movement disorders surgery was revived in 1997. Doshi after his training in Europe started performing pallidotomy. However, within a short period (1998) he switched over to deep brain stimulation surgeries [42]. He and his colleagues observed that STN DBS can produce depression in an occasional patient [43], which was later on accepted as an important side effect in almost 8–18% of STN DBS patients [44,45]. SCTIMST also started performing pallidotomy and deep brain stimulation around the same time. Presently movement disorder surgery is being performed in Mumbai, Hyderabad, Banglore, Trivandrum and Delhi.
Surgery for Chronic Pain Roy offered stereotactic cingulotomy for intractable terminal cancer pain. He used Oliver’s apparatus. The needle was positioned to produce a lesion 2–4 cm posterior to anterior tips of the ventricles and 2 cm in vertical height from above the ventricles. 0.1 ml of carbolic acid with myodil was used in making the lesion. The maximum relief was noted approximately for 2 months after the operation. After this period, the relief of pain was not complete and analgesics were again required. They advocated this surgery for terminally ill cancer patients [46]. Dorsal cordotomy was the preferred procedure in patients suffering from pain of incurable
malignancy in the lower half of the body. However, when the upper half of the body was involved intracranial targets were chosen. Ramamurthi [47] notes that intractable pain due to lesions other than malignancy was not often seen in Indian neurosurgical practice. The incidence of oropharyngeal cancer was very common in South India and advanced cases reported with intractable pain. In such cases section of the trigeminal or the glossopharyngeal nerve was fraught with grave risks especially due to deglutition difficulties with resultant risk of aspiration. Kalyanramana and Ramamurthi studied the neurophysiology of the sensory relay nucleus [48]. From the atlas of Schaltenbrand and Bailey [49] they calculated that the facial area of the sensory relay nucleus of the thalamus was centered on a point 4 mm in front of the center of the posterior commissure, 4 mm above the intercommissural line and 13 mm lateral to the midsagittal plane. Whereas, the termination of the quintothalamic tract into the sensory relay nucleus was calculated to be centered around a point 3 mm in front of the posterior commissure, on the intercommissural line and 10 mm lateral to the midsagittal plane. They used neurophysiological guidance to further refine their target localization. They noted that though sensory responses could be obtained from the internal capsule or other parts of thalamus, the threshold of this response was lowest when the electrode was in the sensory relay nucleus. They also found that microelectrode recordings showed evoked potentials from peripheral stimulation. They initially made large lesions of 8 mm, five times, in the first few patients. However, later on they made only one lesion of 8 mm in the sensory thalamus and an additional lesion in the quintothalamic tract region if pain relief was not adequate. As most of these patients suffered from terminal cancer, they died after a few months and had adequate pain relief till they lived. Two patients who had post herpetic neuralgia
History of stereotactic surgery in india
continued to have pain relief at a follow-up of 6 months [50]. They also performed cingulotomy and hypothalamotomy for pain relief [51].
Psychiatric Disorders Surgery There was a great interest and enthusiasm amongst Indian neurosurgeons in the field of psychiatric disorders surgery as early as 1940. Balkrishna Rao in Bangalore, performed prefrontal leucotomies on patients selected by Govindaswami [52]. BK Anand, a neurophysiologist, participated in the study of the role of the hypothalamus and limbic system in the regulation of feeding behavior conducted by John Fulton and JR Brobeck [53,54]. They discovered, what is since known as, the hypothalamic feeding center. On returning to India, he worked at the Lady Hardinge Medical College, New Delhi and later, at the Department of Physiology, AIIMS, New Delhi [55]. He introduced new experimental techniques and approaches for the study of brain and behavior in India. These include the methods of stereotactic placement of electrodes for making local electrolytic lesions, electrical stimulations, recording of depth EEG, evoked potentials and single unit potentials with microelectrodes and the usage of unanesthetized and free moving animals for behavioral experiments. Manchanda et al. [56] observed that electrical stimulation of perifornical regions of hypothalamus in the carnivore cat evoked different varieties of aggressive behavior: flight, defense, attack. These led other researchers and neurosurgeons to explore the vast field of psychiatric disorders surgery. During his training with Rowbotham, Ramamurthi used to visit St. Lukes Hospital at Middlesborough, where Rowbotham used to perform prefrontal leucotomies. He found that it was successful in more than 60% of the patients. He decided to undertake this upon return to India.
12
Ramamurthi found that the psychiatrists in South India were forward looking and readily referred cases for surgery [10]. For severe depression, Ramamurthi performed a stamp size lesion, extending from 0.7 to 3.0 cm from the midline and 0.8 to 2.8 cm in front of the tip of the anterior clinoid process in the subfrontal region using the diathermy [57]. There was an instantaneous improvement, with some patients describing that ‘‘a great load has been lifted off my chest.’’ Most patients benefitted [58]. In Obsessive Compulsive neurosis, lesions were made in the cingulum. The results were good and long lasting. In some patients, where the results were not as good as expected another lesion was made in the subfrontal region or the cingulum [59–61]. Another interesting indication for psychosurgery was drug addiction. Thirty two cases suffering from drug addictions (alcohol, morphine and pethidine) were operated by Balasubramaniam during 1970–1972. Surgery was done under general anesthesia. Stereotactic localization was performed using pneumoencephalography and carotid angiography. These investigations were essential to determine the thickness of the corpus callosum for making precise lesion in the cingulum, clearing the corpus callosum. The target was selected in line with the foramen of Monro, midway between the pericallosal and callosomarginal arteries. In the coronal plane, the center of the target was 7 mm from the midline. Destruction was done in all cases by injection of myodil, oil, and wax mixture prepared according to the formula of Narabayashi [62]. Both sides were done at one sitting. Postoperative x-rays were taken to confirm the accuracy of the lesion. Of the twenty eight cases followed up for more than 6 months, 22 had been addiction free [63]. Recounting his experience Ravi Ramamurthi says that he found that this was most effective for pethidine addicts. He also mentioned that it was only offered to the patients who were inclined to
165
166
12
History of stereotactic surgery in india
be weaned off their addiction but had failed. The cingulotomy would take away the affective part of their withdrawal symptoms. Balasubramaniam and his colleagues also performed stereotactic amygdalotomy for aggressive behavior in children and adults. Such behavior ranged from continuous severe violent acts towards others, pyromania, destructive tendencies, to episodic attacks of behavior disorders or severe degrees of restlessness. The aim of the operation was to destroy the amygdaloid nucleus or its connections so as to make the patient more manageable either with or without drugs. During 1964–1967, they performed 50 operations on 44 patients; most of them bilateral. The center of the amygdaloid nucleus was considered to be 4 mm in front of the apex of the temporal horn. Two reference points were taken. The first 18 mm below the CACP plane, 6 mm anterior to midcommisural plane and 22 mm from the midsagittal plane. The second reference point was 4–5 mm anterior to the apex of the temporal horn. The author’s preferred the second reference point. In case there was significant cortical atrophy, the first reference point was used. The lesion was made either by diathermy coagulation or Bertrand loop. Diathermy coagulation was done with an 8 mm. electrode, the area of destruction being approximately 200 m3. For the amygdaloid nucleus nine lesions were made. The total volume would be about 1,800 m3 which was slightly greater than the volume of the amygdaloid nucleus which is about 1,200 m3. The lesions made with Bertrand loop were much smaller measuring 500 m3. In 30 operations diathermy was employed and in 19 the Bertrand loop. In one operation both were used. They found that patients having aggressive behavior associated with epilepsy had a better outcome as compared to those suffering from post encephalitic illness aggression [64]. Another target of interest was the posterior hypothalamus. Stereotactic intervention into the posterior hypothalamus was noted to give
satisfactory results for controlling both aggressive, violent behavioral disorders and intractable pain. From the endocrinological point of view, this procedure activates the hypothalamic-hypophyseal axis only temporarily, without causing any serious dysfunctions [65]. Based on these observations Balasubramaniam and Kanaka performed hypothalamotomy on patients who failed to improve after amygdalotomy [66–68]. During the subsequent years, 522 surgeries were performed for aggressive behavior disorder, 402 were bilateral amygdalotomies, and 120 posteromedian hypothalamotomy [69].
Neural Transplantation Stimulated by the reports of A Bjorklund and GD Das at the first congress of the International Brain research organization held at Laussane, Switzerland in 1982, Gopinath (neuroanatomist), Tandon and Mahapatra (neurosurgeons) with Nayar and Mohan Kumar (neurophysiologist) set up a unit to study neural transplantation, with the help of the Department of Science and Technology. They studied neural transplants in rat and primates. Transplantation of embryonic neocortex was performed into cerebellum, lateral ventricle, third ventricle, striatum, hippocampus, tectum and the anterior chamber of the eye in rat. Behavioral and electrophysiological studies were carried out before and after transplantation. Transplantations were also performed in rhesus monkey’s striatum, neocortex and cerebellum for standardization. Parkinson’s disease model was produced in rhesus monkeys using MPTP. As it was difficult to maintain a bilateral Parkinson’s disease model, a unilateral disease model was created. After stabilizing the signs and symptoms for 4–12 weeks, they were grafted with fetal substantia nigra into the striatum using stereotactic techniques. Four monkeys were transplanted. Gopinath [70] notes that three of the four monkeys improved sufficiently to handle food
History of stereotactic surgery in india
while one had to be sacrificed due to complications [71]. Similar transplant program was also initiated at NIMHANS [72] in 1989 and at the post graduate institute of basic medical sciences, Madras [73,74]. Presently, various centers in Bangalore and Delhi are performing studies on Mesenchymal stem cell transplantation for Parkinson’s disease and spinal cord injury. Currently, India is experiencing a renaissance of stereotactic and functional neurosurgery. Imports of equipment has become easier and the present generation of neurosurgeons have the best of both the worlds; extensive clinical experience in India as well as advanced training in specific fields from various centers around the world. There is an increase in understanding and interest in developing India into a global health care provider. Realizing this, a large number of private hospitals have begun to invest in state-of-the-art equipment, thus providing an important platform for the development of this subspecialty.
Acknowledgment I would like to acknowledge the help of NH Wadia and Ravi Ramamurthi for critically reviewing the manuscript. Ravi Ramamurthi, RM Varma, Uday Muthane, KV Mathai, V Rajashekar, and HM Dastur for providing information and personal inputs.
References 1. Sridhar K. Prof. Ramamurthi: the legend and his legacy. Neurol India 2003;52:27-31. 2. Kak V. Neurosurgery. In: Pandya SK, editor. Neurosciences in India retrospect and prospect. New Delhi: The Neurological Society of India; 1989. p. 577-641. 3. Cooper IS. Chemopallidectomy: investigative technique in geriatric parkinsonism. Science 1955;121:217.
12
4. Ramamurthi B. The Governor and the Prime Minister. In: Ramamurthi B, editor. Uphill all the way – an autobiography. Chennai, India: Guardian Press; 2000. p. 206-16. 5. Kalyanraman S, Ramamurthi B. Two machine stereotaxy. Neurol India 1970;18(Suppl 1):53-5. 6. Kalyanaraman S, Ramamurthi B. Simultaneous bilateral stereotaxic lesions in the diencephalon. In: Second international symposium stereoencephalotomy, Copenhagen. Confin Neurol 1965;26:310-14. 7. Nadkarni TD, Goel A, Pandya SK. Neurosurgery in India. Neurol India 2002;48:332-5. 8. Bhatia R, Tandon PN, Banerji AK. Brain abscess – an analysis of 55 cases. Int Surg 1973;58(8):565-8. 9. Kannan V, Deopujari CV, Misra BK, Shetty P, Shroff MM, Pendse AM. Gamma-knife radiosurgery for trigeminal neuralgia. Australasian Radiol 1999;43(3): 339-41. 10. Ramamurthi B. Stereotactic surgery in India: the past, present and the future. Neurol India 2000;48:1-7. 11. Mathai K. Investigations into methods for the rehabilitation of persons disabled by convulsive disorders. Project no. 19-P-5113-F-OI. Division of International activities, Office of research demonstrations and training, social and rehabilitation service. Department of Health, Education and Welfare. Washington D.C: 20201. 12. Singh B, Chandy J. Electroencephalographic study of delta waves. Neurol India 1955;3:5-9. 13. Daniel RT, Chandy MJ. Epilepsy surgery: overview of 40 years of experience. Neurol India 1999;47(2):98-103. 14. Falconer F, Serafetinides FA. A follow up study of surgery in temporal lobe epilepsy. J. Neurol Neurosurg Psychiat 1963;26:300-7. 15. Balasubramaniam V, Kanaka TS. Stereotaxic surgery of the limbic system in epilepsy. Acta Neurochir suppl 1975;23:225-34. 16. Ramamurthi B. Stereotaxic surgery in temporal lobe epilepsy. Neurol India 1970;18:42-5. 17. Balasubramaniam V, Kalyanaraman S, Arjundas G, et al. Stereotaxic ablation of the irritable focus in temporal lobe epilepsy Confin Neurol 1970;32:316-21. 18. Kalyanaraman S, Sayeed ZA, Dharmapal N. Surgical treatment of epilepsy. In: Harris P, Mawdsley C, editor. Epilepsy, Proceedings of the Hans Berger centenary symposium. Edinburgh: Churchill Livingstone; 1974. p. 203-8. 19. Balasubramaniam V, Kanaka TS. Why hemispherectomy? Appl Neurophysiol 1975;38:197-205. 20. Ramamurthi B, Sayeed ZA, Rao DB. Stereotactic surgery of medial temporal structures in temporal lobe epilepsy. Neurol India 1979;27(4):192-5. 21. Arjundas G. Usefulness of depth electrode studies in epilepsies. In: Mani KS, Walker AE, Tandon PN, editors. Proceedings of national seminar on epilepsy. Bangalore:
167
168
12 22.
23.
24. 25.
26.
27.
28.
29. 30. 31.
32.
33.
34.
35.
36. 37.
38.
History of stereotactic surgery in india
Indian epilepsy association, Bangalore Chapter; 1976. p. 75-7. Kanaka TS, Balasubramaniam V. Cortical and depth electrode studies. Neurol India 1980;28:150-4; Mathai KV. Epilepsy – some epidemiological experimental and surgical aspects. Neurol India 1986;14:299-314. Gillingham FJ, Watson WS, Donaldson AA, Naughton JAL. The surgical treatment of parkinsonism. Brit Med J 1960;2:1395. Kalyanaraman S, Ramamurthi B. Stereotaxic surgery for generalized epilepsy. Neurol India 1970;18:34-41. Kalyanaraman S, Ramamurthi B. Stereotaxic capsulotomy for epilepsy. Proc Australian Assoc Neurol 1968;5: 311-12. Mathai KV, Taori GM. Stereotaxic destruction of ansa and fasciculus lenticularis in the control of seizures. Neurol India 1972;20(Suppl 2):169-74. Rao MB, Radhakrishnan K. Is epilepsy surgery possible in countries with limited resources? Epilepsia 2000;41: S31-4. Bhatia M, Singh VP, Jain S, et al. Epilepsy surgery in India: AIIMS experience. J Assoc Phys India 1999;47: 492-5. Balasubramaniam V, Ramamurthi B. Stereotactic surgery. Neurol India 1965;13(3):93-6. Kalyanaraman S, Ramamurthi B. Simultaneous bilateral stereotaxic lesions. Neurol India 1966;14(3):151-3. Kalyanaraman S, Ramamurthi B. The size of the third ventricle in British and Indian subjects. Neurol India Suppl 1:60-4. Narabayashi H. Physiological analysis of ventrolateral thalamotomy for rigidity and tremor. Confin Neurol 1965;26:264-8. Kanaka TS, Balasubramaniam V, Ramanujam PB, Ramamurthi B. Stereotaxic surgery in cerebral palsy – a study of 57 cases. Neurol India 1970;18(1):56-9. Balasubramaniam V, Kanaka TS, Ramanujam PB. Sterotaxic surgery for cerebral palsy. J Neurosurg 1974;40:577-82. Balasubramaniam V, Kanaka TS, Ramanujam, PB, Ramamurthi B. Stereotaxic hypothalamotomy. Stereotact Funct Neurosurg 1973;35:138-143. Ecker A, Perl T. Percutaneous injection of the thalamus in parkinsonism. Arch Neurol 1960;8:271-8. Varma RM. Percutaneous chemothalamectomy for parkinsonism. Neurol India 1964;12(2):54-60; Varma RM. Percutaneous technique of chemothalamotomy for parkinsonism. Paper presented at third world congress of neurological surgeons, Brussels, 1965. Muthane UB, Varma RM, Sundarajan P, et al. Percutaneous trans-foramen ovale approach to subthalamic nucleus (Varma’s Technique): a shorter route and economical technique in surgical treatment of Parkinson’s disease. Mov Disord 1998;13:71.
39. Muthane UB, Bhatt MH, Wadia NH. Parkinson’s disease and other akinetic disorders. In: Wadia NH, editor. Neurological practice an Indian perspective. Elsevier: Amsterdam; 2005. p. 353-66. 40. Bhagwati SN, Singhal BS. Reactivation of successfully abolished tremors on thalamic stimulation whilst performing surgery for the second side in Parkinson’s disease. Neurol India 1972;20(Suppl 2):166-8. 41. Bhagwati SN. Stereotaxic surgery for Parkinsons’s disease. Bombay Hosp J 1969;2:39-42. 42. Doshi PK, Chhaya NA, Bhatt MH. Bilateral subthalamic nucleus stimulation for Parkinson’s disease. Neurol India 2003;51:43-8. 43. Doshi PK, Chhaya NA, Bhatt MH. Depression leading to attempted suicide following bilateral subthalamic nucleus stimulation for Parkinson’s disease. Mov Disord 2002;17 (5):1084-5. 44. Tir M, Devos D, Blond S, et al. Exhaustive, one-year follow-up of subthalamic nucleus deep brain stimulation in a large, single-center cohort of parkinsonian patients. Neurosurgery 2007;61(2):297-304. 45. Temel Y, Kessels A, Tan S, et al. Behavioural changes after bilateral subthalamic stimulation in advanced Parkinson disease: a systematic review. Parkinsonism Relat Disord 2006;12(5):265-72. 46. Roy TK. Stereotactic cingulate lesions for intractable pain in malignant conditions – a report of 5 cases. Neurol India 1983;31(1):55-8. 47. Ramamurthi B, Davidson A. Progress in stereotactic surgery in Madras. Proc Inst Neurol, Madras 1975;5:81-97. 48. Kalyanaraman S, Ramamurthi B. Studies on the sensory relay nucleus of the thalamus during stereotaxic surgery. Neurol India 1972;20(Suppl 2):155-7. 49. Schaltenbrand G, Bailey P, editors. Introduction of Stereotaxis with an atlas of the human brain. Stuttgart: George Thieme; 1959, p. 2. 50. Kalyanaraman S, Ramamurthi B. Stereotaxic surgery for intractable pain neurology India September 1969;17 (3):109-115. 51. Kalyanaraman S, Logamuthukrishnan S. Stereotaxic posteromedial hypothalamotomy for intractable pain. Proc Inst Neurol Madras 1974;4:95. 52. Govindaswamy MV, Rao B. Bilateral prefontal leucotomy in Indian patients. Lancet 1944;1:466. 53. Anand BK, Brobeck JR. Hypothalamic control of food intake in rats and cats. Yale J Biol Med 1951;24:123-40. 54. Anand BK, Brobeck JR. Loacalization of a feeding center in the hypothalamus of rat. Proc Soc Exp Biol Med 1951;77:323-4. 55. Desiraju T. Fundamental neurophysiology. In: Pandya SK, editor. Neurosciences in India: Retrospect and Prospect, Neurological society of India, 1989, p. 113-52.
History of stereotactic surgery in india
56. Manchanda SK, McBrooks C. Homeo-static controls in the regulation of autonomic nervous system function. In: McBrooks C, Koizumi K, Sato A, editors. Integrative Function of the Autonomic Nervous System. University of Tokyo; 1979. p. 427-430. 57. Kalyanaraman S, Ramamurthi B. Stereotaxic basofrontal tractotomy. Neurol India 1973;21:113-18. 58. Ramamurthi B, Ravi R, Narayanan R. Long term follow-up of functional neurosurgery in psychiatric disorders. experience of 30 cases. The World Conference of Stereotactic and Functional Neurosurgery. Zurich: 1981. 59. Balasubramanium V, Kanaka TS. Stereotaxic surgery of the limbic system in epilepsy. Acta Neurochir 1975;23: 225-34. 60. Balasubramaniam V, Ramanujam PB, Kanaka TS, et al. Stereotaxic surgery for behaviour disorders. In: Hitchcock E, Laitinen L, Vaernet KC, editors. Psychosurgery. Springfield, IL: Charles C. Thomas; 1972. p. 156-63; Ramamurthi B, Ravi R, Narayanan. Functional neurosurgery in psychiatric illness. Indian J Psychiatry 1980;22:261-4. 61. Ramamurthi B, Ravi R, Narayanan. Functional neurosurgery in psychiatric illness. Indian J Psychiatry 1980;22:261-4. 62. Balasubramaniam V. Stereotaxic amygdaloid lesions in behaviour disorders. Ph D. Thesis, Madras University; 1969. 63. Balasubramaniam V, Kanaka TS, Ramanujam PB. Stereotaxic cingulumotomy for drug addiction. Neurol India 1973;21(2):63-6. 64. Balasubramaniam V, Ramamurthi B, Jagannathan K, Kalyanaraman S. Stereotaxic amygdalotomy. Neurol India 1967;15(3):119-22.
12
65. Mayanagi Y, Hori T, Sano K. The posteromedial hypothalamus and pain, behavior, with special reference to endocrinological findings. Appl Neurophysiol 1978;41: 223-31. 66. Arjundas G, Balasubramaniam V, Reddy EC, Ramamurthi B. Hypothalamus and its effects on the viscera. J Assoc Physicians India 1971;19:477-81. 67. Balasubramaniam V, Kanaka TS, Ramanujam PB, Ramamurthi B. Stereotactic hypothalamotomy. Confin Neurologica 1973;35:138-43. 68. Balasubramaniam V, Ramamurthi B, Jagannathan K, Kalyanaraman S. Stereotaxic amygdalotomy. Neurol India 1967;25:119-21. 69. Ramamurthi B. Surgery for aggressive behavior disorders. International Congress Series No 433. Neurological surgery. Proceedings of the Sixth Congress of Neurological Surgery. Sao Paulo; 1977. 70. Gopinath G. Neural transplantation. In: Pandya SK, editor. Neurosciences in India retrospect and prospect. Neurological society of India. 1989. p. 219-30. 71. Mahapatra AK, Gopinath G, Tandon PN. Neural transplantation. Prog Clin Neurosci 1987;1:45-8. 72. Murthy SK, Desiraju T. Quantitative assessment of dendritic branching and spine densities of neurons of hippocampal embryonic tissue transplanted into juvenile neocortex. Dev Brain Res 1989;46:33-46. 73. Muthuswamy R, Sheeladevi A, Namasivayam A, et al. Transplantation of monkey embryonic cortical tissue into the brain of bonet monkey (Macaca radiata). J Anat Soc India 1988;37:27. 74. Muthuswamy R, Sheeladevi A, Namasivayam A, et al. Heterospecific transplantation of human embryonic cortical tissue into the cerebellum of bonnet monkey (Macaca radiata). Neuroscience 1987;22:S765.
169
5 History of Stereotactic Surgery in Japan C. Ohye
Production of the First Stereotactic Instrument and its Application in Human It was post-World War II, when a young bud of stereotactic surgery sprouted in Japan as in other countries. Hirotaro Narabayashi, a postgraduate psychiatrist, inspired by Prof. Ogawa’s suggestion, explored the destroyed and almost burned out Tokyo with a drawing of a stereotactic instrument, looking for a factory that had survived the war. Finally, he found one small factory still working, and with the help of a kind engineer there, he produced the first model of a stereotactic instrument in 1949 [1]. His medical student days during the War were miserable; they were shortened because of heavy air strikes and academic school life almost stopped. In fact, as many Professors and young doctors were summoned for military service, lectures were often cancelled. So Narabayashi spent all his free time in the University Library, where he found an interesting book ‘‘Die Extrapyramidales Erkrankungen’’ by the German pathologist Alfons Jacob [2]. The book proved to be a source of great inspiration and perhaps his idea of using the stereotactic instrument to invade the basal ganglia was born around this time. After graduating from the University of Tokyo in 1945, he joined Professor Uchimura’s Department of Psychiatry in 1946. At that time, there was no separate department of Neurology, so patients with Parkinson’s disease and other involuntary movements were treated in the department of Psychiatryand he noticed that there was no effective treatment for several cases. When Professor #
Springer-Verlag Berlin/Heidelberg 2009
Teizo Ogawa, Professor of Neuroanatomy talked to him about a stereotactic instrument used in animal experiments in Ranson’s Lab in the United States, young Narabayashi was excited and immediately took steps to see if it could be applied to the human brain. However, at that time, with Tokyo almost completely destroyed by bombardment, he had difficulty finding a factory to make such an instrument, as mentioned earlier. Being a passionate young man he had the first stereotactic instrument made in 1949 [3,4]. In 1947, Spiegel and Wycis, Philadelphia, United States, reported the first case of stereotactic surgery on thalamic dorsomedial nucleus for a patient with a neurological problem and movement disorder [5,6]. At that time in Japan, communication with other countries was very limited, and Narabayashi was not aware of the exciting news. Unaware of the work of Spiegel and Wycis, Narabayashi continued to improve on his stereotactic instrument and simultaneously worked on a map of the brain(not published). In 1951, he operated on a patient with cerebral palsy with athetoid movement [1]. But his most exciting success was on June 4, 1952 [7,8], when the first pallidotomy on Parkinson’s disease was successfully completed. After the injection of a small amount of oil wax (procaine-oil and honey wax) the tremors and rigidity disappeared almost immediately and completely. Seeing this miraculous result, Professor Uchimura, chairman of the Department, recognizing the scientific worth in the result encouraged Narabayashi to take this up as his life’s work. Following the Professor’s advice faithfully Narbayashi devoted his whole life to this field of stereotaxy. There
60
5
History of stereotactic surgery in Japan
is a moving anecdote reported in the story of his first pallidotomy. The patient was a relatively young telecommunist suffering from tremors and rigidity. The patient was a very generous man and agreed to undergo the new brain operation without any guarantee. The first and the second operations to caudate nucleus and putamen were in vain. When Narabayashi pleaded for a third attempt, the patient willingly accepted and Narabayashi’s enthusiasm was rewarded with triumph. Another important and unforgettable episode that centered around his first pallidotomy was, the now well known but controversial at that time, the bitter criticism of Spiegel and Wycis and the firm response from Narabayashi. However, Spiegel and Wycis soon recognized the hardwork of Narabayashi and a deep understanding of each other’s scientific attitude fostered a long lasting friendship between them. The letters they exchanged were published in the Archives of Neurology and Psychiatry in 1956 and were later reproduced in a tribute to Narabayashi [9,10]. For several years after the war, the Japanese people had a very difficult time with shortage of food, bare necessities, housing etc. Nevertheless, the Japanese people worked hard, doctors with poor medical facilities were no exception. In the Japanese University Clinic, Neurosurgery including stereotactic surgery was started in earnest. Stereotactic surgery attracted many young neurosrgeons and as a natural consequence Narabayashi became the central figure. He extended his stereotactic research working with young researchers of the Brain Research Institute, University of Tokyo, where Professor Toshihiko Tokizane chaired the Neurophysiology section. Narabayashi recognized from the very beginning, that cooperation with researchers in the basic sciences was essential for the development of stereotactic surgery; the author is one of those who worked with him. In 1956 he moved to Juntendo University as Associate Professor of Psychiatry, but in 1957 he founded his own private clinic devoted mainly to
stereotactic surgery; this became a world famous ‘‘Neurological Clinic’’ later on. As Narabayashi was very frank and open minded, many young doctors gathered in his clinic to see new operations and discuss freely with him. His clinic became a salon for discussions on not only stereotactic surgery, but all areas of Neuroscience. From this salon, in fact, many fledgling doctors of Neuroscience – neurologists, neurosurgeons, neurophysiologists, neurochemists, neuroanatomists etc. – have emerged as leaders in their fields. Narabayashi’s contribution in this sphere also has been remarkable.
Birth of the Japanese Society of Functional and Stereotactic Surgery Many Professors of Neurosurgery [11–13] gathered around Narabayashi, and the Japanese Society for Research in Stereoencephalotomy was founded in 1963, following the International Society, which was founded in 1961 The first Meeting was held in Kyoto, February, 1963, hosted by Professor Araki of Kyoto University. The second Meeting was held in June of the same year organized by Professor Takebayashi of Wakayama Medical School. After the third meeting in the following year, the Japanese Society has been meeting regularly every year, so the 47th Meeting will be held in 2008, organized by Professor Namba of Hamamatsu Medical School. We are very proud that the annual meetings have been conducted regularly without interruption. At one point of time there was an effort to establish the Japanese Society (a research group) but the research fund provided by the Japanese Ministry of Education for the purpose of ‘‘Stereotaxic deep brain surgery’’ was accepted by our Society. This governmental support was, in fact, the successor to the previous research fund for the ‘‘Study on the extrapyramidal movement disorder.’’ The name of the society was changed, following the name of the World Society to
History of stereotactic surgery in Japan
‘‘the Japanese Society for Stereotactic and Functional Neurosurgery’’ in 1973. At the first meeting, some 50 interested doctors came from several centers and 13 papers from different University clinics were presented including my special lecture on neurophysiology of the basal ganglia symptoms. There were many lively discussions on this new operation. At present the society has 500 odd members and some 100 papers were presented during the two days of the meeting with lively discussions. We pride ourselves on the organization of these meetings and the free and active discussions. Narabayashi attended the first International Meeting of Neurology and Neurosurgery in Brussels in 1959. In the stereotactic section he was the first Japanese to attend an international meeting. He presented his work on pallidotomy, met most of the pioneering stereotactic surgeons from all over the world who shared their initial experiences of stereotactic surgery. Narabayashi’s work on pallidotomy was appreciated by many. It was very important that the Japanese development of stereotaxy was recognized even in the early stages. Sterotactic surgery was once prevalent all over Japan.Asin othercountries almost every Neurosurgical Department has a stereotactic unit. Following are the names the first generation of stereotactic surgeons in Japan with their areas of work: Ueki (Niigata) – Ultrasonic device, Sano (Tokyo) – hypothalamotomy for behavior disorder, Narabayashi,Nagao(Tokyo)–pallidotomy,Takebayashi, Komai (Wakayama) – Superior colliculus for Nystagmus, Araki, Handa (Kyoto) – Torticollis, Ozawa, Hori (Osaka) – Epilepsy, Jinnai, Nishimoto (Okayama) – Forel H, Hoshino (Hiroshima) – Radioisotope. Many young stereotactic neurosurgeons went abroad (USA, Canada, Germany, France, Swiss, Sweden, Italy, England etc.), to learn about the different aspects of stereotactic surgery. Sterotactic surgery has progressed rapidly in Japan and consequently, nowadays, several young stereotactic doctors of second and third generation attend
5
international meetings related to modern stereotactic research and treatment in the United States and in Europe.
The Ups and Downs of the Society Around the 1970s stereotactic surgery went through several ups and downs. One of the unhappy incidents was the world wide student violence toward the end of the 1960s, known as the Zengakuren in Japan. Some professors were attacked by the students, professors of Psychosurgery being their main focus. Unfortunately the hypothalamotomy of Sano and the amygdalotomy of Narabayashi had to be stopped. As a consequence, the stereotactic team in Tokyo University was almost disbanded, which resulted in a great loss for the progress of stereotactic surgery in Japan. Another critical event for the world of stereotactic surgery was the progress in the L-dopa drug therapy for Parkinson’s disease around the 1970s that sidelined the surgical option. The next 20 years were difficult times for stereotactic surgeons all over the world [14]. In Japan the surgeries were reduced and many Neurosurgical Departments closed their stereotactic section. But Narabayashi’s group continued with surgeries because of his reputation, and the microrecording technique learnt from the French group (Albe-Fessard and Guiot) by Ohye was established during this time. Thereafter, microrecording during stereotactic surgery became one of the prestigious features in Japan. It also gave us an opportunity to study the functional organization of the human thalamus (see ‘‘Thalamotomy’’ in this book). On the other hand, the use of L-Dopa for more than 10 years brought to light an unexpected new hazard in the treatment of Parkinson’s disease. As every one is aware, L-Dopa is quite effective for a few years initially but it has complicated side effects. This inevitably suggests revival of the surgical treatment.
61
62
5
History of stereotactic surgery in Japan
Nevertheless, during this period, the Japanese Society continued to have its annual meetings as before. One aspect of the scientific activity of the Japanese Society, the number of presentations in each year’s meeting, is shown in > Figure 5-1 [15]. It is clear from the figure that the number of papers presented at these meetings has grown steadily from 20 papers in the earlier years to nearly 100 papers in recent times. As a result of this increase, since the 39th meeting (in Fukuoka, in 2000), the Society meets for two days. In addition, the Society, which usually held its meetings just before the Annual Meeting of the Japanese Society of Neurosurgery (usually one day earlier), has since the 43rd meeting in Nara become independent of the general neurosurgery meeting. Moreover we have two local functional neurosurgery study groups namely, the Kanto (East Japan) Functional Neurosurgeon Conference twice a year and the Yamaguchi and Kyusyu
(Western) Stereotactic Neurosurgery Seminar once a year. The former had its 25th conference in 2007, and the latter its 15th seminar in 2007. In effect, our Japanese stereotactic group continues to be very active.
New Age of Stereotactic Surgery During the ‘‘cold age’’ of stereotactic surgery, we had a few pleasant events. Narabayashi hosted the World Society Meeting in Tokyo in 1973, and Ohye the 10th Meeting in Maebashi in 1989. They both served as the President of the World Society following each World Society Meeting they hosted. Further, Narabayashi was awarded the Spiegel Wycis Medal, the highest honor of the Society, in 1989, and Ohye received it in 2001. The Japanese Society is very proud of them. The Japanese Society continued its activity into the new age. We now have several advanced
. Figure 5-1 Time sequential change of number of papers (ordinate) presented at the Annual Meeting of Japanese Society from the beginning (1st) to the present time (47th) (Abscissa). Results of 21st–35th meeting were omitted. Note that since 39th meeting, we had two days meeting, and the submitted pages increased from the next year of 2001. The 15th meeting was somewhat different from others. It was invited lectures by foreign guests and related discussions
History of stereotactic surgery in Japan
techniques in surgery and fresh knowledge of the basal ganglia and hence, Parkinson’s disease. Computerized imaging system is now close to neurosurgeon’s hand to visualize deep brain structures in 3D fashion. Understanding Parkinson’s disease opens several new avenues that are still growing. On the other hand, advanced new surgical technology has given us new tools – chronic stimulation by implanted electrode initiated by A. Benabid in France, stereotactic radiosurgery using Gamma knife by L. Leksell, brain graft and etc. We are now in a new era of stereotaxy with many new possibilities. In fact, more than ten stereotactic centers are actively involved in stereotactic selective thalamotomy with microrecording, chronic stimulation of different targets including the thalamus, GPi, subthalamic nucleus, or stimulation of cerebral cortex (initiated by Tsubokawa), Although we do not yet know the optimal target for the treatment of different symptoms, we are quite optimistic about the future of stereotactic surgery, because the idea of stereotaxy is now widely applied in the field of general neurosurgery and therefore it stands at the center of Neurosurgery. Without stereotaxy, precise minimally invasive neurosurgery cannot be performed.
References 1. Kanazawa I. Interview with Professor Narabayashi. Tracing the idea and footstep of a neurologist. Prog Neurol 2001;45(3):512-22 (Japanese).
5
2. Jacob A. Die extrapyranuidale Erkrankungen. Berlin: Springer-Verlag; 1923. 3. Narabayashi H. (Stereotaxic apparatus (Model II) (Japanese). Psychiat Neurol Jpn 1953;54:669. 4. Uchimura Y, Narabayashi H. Stereotaxic apparatus. Psychiat et neurol 1951;52:265-70. 5. Spiegel EA, Wycis HT. Pallido-thalamotomy in Chorea. Arch Neurol Psychiat 1950;64:495-6. 6. Spiegel EA, Wycis HT, Marks M, Lee AJ. Stereotaxic apparatus for operations on the human brain. Science 1947;106:349-50. 7. Narabayashi H, Okuma R. Procaine oil blocking of the globus pallidus for the treatment of rigidity and tremor of parkinsonism (priminary report). Proc Jpn Acad 1953;29:134-7. 8. Narabayashi H, Okuma T, Shikiba S. Procaine oil blocking the globus pallidus. Arch Neurol Psychiatry 1956;75:36-48. 9. Ohye C, Fodstad H. Forty years with professor Narabayashi. Neurosrgery 2004;55:222-7. 10. Spiegel EA, Wycis HT. Procaine oil blocking the globus pallidus (letter). Arch Neurol Psychiatry 1956;76:263. 11. Narabayashi H. Stereo encephalotomy in Japan. Conf Neurol 1964;24:314-20. 12. Narabayashi H. Begging and development of sterotaxic surgery in Tokyo. Conf Neurol 1975;37:364-73. 13. Ohye C, Fodstad H. Prof. Hiro Narabayashi: in memoriam. Stereotact Funct Neurosurg 2001;76:125-8. 14. Ohye C. Stereotactic surgery of Parkinson’s disease over 30 years. In: Mizuno Y, Fisher A, Hanin I, editors. Mapping the progress of Alzeimer’s and Parkinson’s disease. New York: Kluwer Academic/Plenum Publishers, (2002) p. 429-34. 15. Matumoto K., Abstracts of the papers presented at the 1st–20th meeting of the Japanese society of stereotactic and functional neurosurgery. Neuron Tokushima, Tokyo; 1892.
63
13 History of Stereotactic Surgery in Korea S. S. Chung
In Korean history, the end of the nineteenth century is considered as the period of the Chosun Dynasty’s transition from medieval to modern. This period was marked by wide-ranging social and economic changes on the domestic front, and, externally, by the threat of foreign domination. It was during this period when modern medicine began, perhaps personified historically when Dr. Horace N. Allen (> Figure 13‐1), a Protestant missionary from the Presbyterian Missions in New York, stepped onto Korean soil. In December 1884, Allen was given the opportunity of saving the life of Queen Min’s nephew. In gratitude, at the behest of King Kojong, the Royal Hospital Kwanghyewon, ‘‘House of Extended Grace,’’ was founded on 10 April 1885, and on April 23rd the name was changed to Chejungwon (> Figure 13‐2), which means ‘‘Universal Helpfulness,’’ Kwanghyewon was the first modern hospital in Korea, which later became the Severance Hospital, an affiliate of the Yonsei University Medical School [1,2]. The hospital provided a legitimate venue for the first protestant missionaries to pursue their religious activities while conducting systematic research on diseases endemic to Korea. In March 1886, a year after the founding of Chejungwon, the hospital launched its Medical Department to educate future medical practitioners and professionals in modern medicine, aiming to treat the ailments of the Korean people. Since Allen had been appointed as head of the hospital, J. E. Heron, Charles C. Vinton, and Oliver R. Avison (> Figure 13‐3) successfully carried on his work as a director of the hospital. #
Springer-Verlag Berlin/Heidelberg 2009
As the result of advances made possible by Avison’s fund-raising initiatives in the United States, in 1899 the Medical Department of Chejungwon was accredited as a full-fledged medical educational institution. An American entrepreneur, Louis H. Severance, deeply moved by Avison’s speech about his missionary activities in Chosun, decided to donate $10,000, a very large sum of money in 1900. With funding provided by Severance, a new hospital building was completed on 23 September 1904. It was named Severance Hospital (> Figure 13‐4) to commemorate the man who had made it possible. It was Korea’s first modern Western style hospital building. In 1908, the Chejungwon Medical School (later Yonsei University Medical School) celebrated its first graduation. While the Chejungwon Medical School was founded and established, several medical schools were also founded in sequence. In the early 1940s, a few pioneering general surgeons started to practice neurosurgical procedures in medical school accredited hospitals. The neurosurgical procedures included neurotraumas, epilepsy surgeries, and surgeries for mental disorders. Actually all the early stage neurosurgeons were functional neurosurgeons. In 1943, Dr. Chu Kul Lee performed corticectomy on a patient with posttraumatic epilepsy. Dr. Lee graduated from Daegu Medical College (later Kyungbook National University) in 1937 and continued to study neurosurgery in Nagoya University, Japan. He returned to Korea in 1942 and proceeded to perform cortical resection for epilepsy patients in Seoul’s Women’s Medical College (later Korea University Medical College).
172
13
History of stereotactic surgery in korea
Later on he conducted cortical coagulation to interrupt the epileptic impulse in patients with epilepsy. He was interested in epilepsy surgery throughout his neurosurgical career and . Figure 13‐1 Dr. Horace N. Allen; Presbyterian missionary from New York who founded the first Western style hospital, Chejungwon on 10 April 1885
published his experience in the first issue of the official Journal of the Korean Neurosurgical Society in 1972 [3]. He was a leading figure in the neurosurgical field and trained many neurosurgeons in 1960s. He was elected as the first president of the Korean Neurosurgical Society in 1961. Another pioneer was Dr. Ki Sup Lee. He graduated from Severance Medical College (later Yonsei University) and studied neurosurgery in Kyoto University, Japan. He performed frontal lobotomy for mental disorder, corticectomy for epilepsy, and sympathectomy for pain in Severance hospital in 1943 (Lee KS, 1996, personal communication). Dr. Si Chang Kim graduated from Kyungsung University (later Seoul National University) in 1936. At one stage, he worked in Seoul Women’s Medical College and returned to Kyungsung University Hospital in 1948 where he performed corticectomy and cortical vessel ligation for epilepsy sufferers. He was one of the handful number of active surgeons practicing neurosurgery at that time. Unfortunately, he was
. Figure 13‐2 Chejungwon in Seoul (1885); the first Western style hospital in Korea
History of stereotactic surgery in korea
abducted and taken to North Korea during the Korean War (Moon TJ, Forty years History of the Korean Neurosurgical Society (19612001), 2002, personal communication).
. Figure 13‐3 Oliver R. Avison (served 1893–1935); Canadian missionary who was director of Chejungwon hospital and later Dean of Severance Medical College
13
The Korean War broke out on the 25 June 1950. War-related trauma injuries provided great momentum to the development of neurosurgery in Korea. During the Korean War, Scandinavian countries sent the Danish hospital ship Jutlandia to Busan to care of casualties. The chief neurosurgeon on this ship was Professor Edward A. V. Busch from Copenhagen University, Denmark, and his assistant was Dr. K. Vaernet. They not only treated patients but also taught neurology and neurosurgery to Korean military surgeons. Many American military surgeons also came to Korea during the war, among them colonel Arnold M. Meirowsky and George J. Hayes contributed enormously to the development of neurosurgery in Korea. They served in the mobile army surgical hospitals (MASH) and also taught neurosurgery to Korean military surgeons [4]. Some Korean surgeons had several months of continuous neurosurgical training, which served them well in organizing neurosurgical teams. Casualties from the war included patients with cranio-spinal trauma, peripheral nerve injury, and causalgia. The neurosurgeons in the army
. Figure 13‐4 Severance Hospital (1904), the first Western style hospital building in Korea. Louis H. Severance, businessman from Cleveland, donated for the new hospital building and Chejungwon Hospital was renamed as Severance Hospital
173
174
13
History of stereotactic surgery in korea
performed many thoracic sympathectomy for patients with causalgia. It was a great opportunity for the Korean neurosurgeons to observe Western medicine and neurosurgical practices. After the war some of them continued to practice neurosurgery while others went abroad to study neurosurgery in Western countries [4]. In the late 1950s, neurosurgeons who went abroad came back to Korea after completing neurosurgical training. Dr. Tae Joon Moon had residency training at Thomas Jefferson University (USA), and became a qualified neurosurgeon by the time he came back in 1957. Upon his return, he practiced neurosurgery in Yonsei University and performed trigeminal ganglion block with hot saline or absolute alcohol in the treatment of trigeminal neuralgia and open thoracic cordotomy for intractable cancer pain. Dr. Bo Sung Sim graduated from Seoul National University in 1949. He served as an army neurosurgeon during the Korean War and studied neurosurgery in Minnesota University (USA). He performed hemispherectomy in patients with intractable epilepsy due to cerebral paragonimiasis in 1958. He also performed retrogasserian rhizotomy in the middle cranial fossa for trigeminal neuralgia [4]. Dr. Hun Jae Lee graduated from Severance Medical College in 1944 and finished his residency program in Michigan University Hospital (USA) to become a qualified neurosurgeon. He came to Seoul Women’s Medical College in 1959, after which he performed open cordotomy for intractable cancer pain and retrogasserian rhizotomy for trigeminal neuralgia, while also applying alcohol in gasserian ganglion percutaneously for trigeminal neuralgia [5]. Dr. Kon Huh graduated from Severance Medical College in 1948, and continued studying neurosurgery in Wisconsin University (USA). He came back to Severance Medical College in 1962 where he practiced pain surgery such as open cordotomy or retrogasserian rhizotomy in middle cranial fossa [4]. Dr. Jeong Wha Chu graduated from
Seoul National University in 1956, and studied neurosurgery in Minnesota University (USA). After coming back from Minnesota University in 1961, he practiced some ablative pain surgeries such as open cordotomy and retrogasserian rhizotomy in middle cranial fossa [6]. Stereotactic surgery in Korea began in the early 1960s. Dr. Hun Jae Lee performed chemothalamotomy in patients with Parkinson’s disease using Cooper’s frame in 1960. It was the first stereotactic surgery using stereotactic apparatus in Korea. He presented the results of chemothalamotomy in seven cases of Parkinson’s disease patients and four cases of dystonia patients. He reported on follow up results of those patients in 1963 (Chu JW, 2006, personal communication). Dr. Tae Joon Moon also performed thalamotomy using simple burr hole mounted Mackinie apparatus. Dr. Chul Woo Lee of Kyungbook University, who trained at the Saint Vincent Hospital in Wooster city near Boston (USA), made his own stereotactic frame in 1960. He performed thalamotomy for patients with dystonia and Parkinson’s disease. Dr. Jeong Wha Chu also performed thalamotomy for patients with Parkinson’s disease [6]. However, there were very few neurosurgeons who carried out stereotactic surgery for movement disorders at that time. In 1971, the radiofrequency lesion generator was introduced in to Korea. Sang Sup Chung of Yonsei University performed percutaneous radiofrequency cervical cordotomy, radiofrequency trigeminal thermocoagulation, and radiofrequency ventrolateral thalamotomy for Parkinson’s disease sufferers in 1972 [7,8]. In 1973, radiofrequency facial nerve neurotomy was performed for hemifacial spasm at the stylomastoid foramen [9]. It was the beginning of the subspecialty of stereotactic and functional neurosurgery in Korea. Introduction of radiofrequency lesion generator made it possible for many procedures to be conducted percutaneously, allowing for a more accurate,
History of stereotactic surgery in korea
safer, and simpler operation. In 1976, Computed tomography (CT) scanner was introduced in to Korea. The introduction of radiofrequency lesion generator and CT scanners were substantial moments for the development of stereotactic and functional neurosurgery in Korea. In 1975, Sang Sup Chung went to Edinburgh University, Britain, for further study where he practiced stereotactic and functional neurosurgery under professors F. John Gillingham and Edward R. Hitchcock. After returning to Yonsei University he became a fulltime stereotactic and functional neurosurgeon. In 1976, Chang Rak Choi studied functional neurosurgery under professor Umbach in Berlin, Germany, and came back to Catholic University where he continued to practice stereotactic and functional neurosurgery. In 1978, Dr. Kil Soo Choi and Dr. Hun Jae Lee performed microvascular decompression for hemifacial spasm and trigeminal neuralgia. Soon after, Sang Sung Chung followed the procedures [10]. In 1979, depth recording was performed using semi-micro electrode during thalamotomy. In the 1980s there was rapid progress in research activities and surgical techniques in Korea. Various stereotactic and functional surgeries were performed. In 1980, Sang Sup Chung performed stereotactic chemical or radiofrequency hypophysectomy [11], percutaneous spinal rhizotomy and percutaneous medullary trigeminal tractotomy in treating cancer pain. Dorsal root entry zone lesioning and facet denervation were performed for chronic intractable pain and for chronic low back pain. Centrum medianum and parafacicularis nucleus lesioning were done for chronic central pain. Hypothalamotomy was performed for aggressive psychosis and glycerol injection was introduced as a treatment modality for trigeminal neuralgia. Also in 1980, percutaneous spinal cord stimulation for chronic pain and intrathecal infusion pump were introduced for cancer pain. However, these devices were not covered by insurances and it was difficult to treat many patients in the early stages of diagnosis
13
because of economic problems. During this period, various newly developed functional neurosurgical procedures were attempted while a number of stereotactic apparatus such as ToddWells, Riechert-Mundinger, Guiot-Gillingham, BRW frame were introduced in to Korea. In the 1980s, CT compatible CRW, Hitchcock, and Leksell frame, which were introduced, enabled surgeons to perform image guided surgery. In 1984, stereotactic evacuation of intracerebral hematoma or brain biopsy was carried out by CT image guided surgery. In 1982, Moon Chan Kim came back to Catholic Medical College after studying in Birmingham under Professor Hitchcock. As a full time neurosurgeon, he practiced surgery for movement disorder, pain, and psychiatric illness. In 1988, linear accelerated based radiosurgery were performed by Moon Chan Kim and Sang Sup Chung. In 1988, Sang Sup Chung performed adrenal gland transplantation for Parkinson’s disease sufferers and reported on the 5 year follow-up results in 1993 [12]. Epilepsy surgery was conducted sporadically from its inception. However, a comprehensive epilepsy protocol (Yonsei Epilepsy Protocol) was established in 1989, after which epilepsy surgery became more standardized and several centers were built. During the 1990s, stereotactic and functional neurosurgery achieved substantial development owing to the great technological advancements in computer science, surgical softwares, engineering, neurophysiology, and various diagnostic tools. The resolution of MRI improved immensely and accurate MRI guided surgery became possible from 1995. In 1999, vagal nerve stimulation was done for intractable epilepsy and microdepth recording started during thalamotomy. Deep brain stimulation for Parkinson’s disease was introduced in 2000 by Jin Woo Chang and Sang Sup Chung. From late 1980s through to the 1990s, many young neurosurgeons were getting interested
175
176
13
History of stereotactic surgery in korea
in stereotactic and functional neurosurgery. Uhn Lee of Ghil Hospital, who graduated from Hanyang University, is an active neurosurgeon who treats patients with movement disorders. He conducted thalamotomy and pallidotomy in the late 1980s and now continues to practice DBS. Jae Hyoo Kim of Chonnam University is also an active functional neurosurgeon, who is carrying out pain surgery, DBS for movement disorders and sympathectomy for hyperhidrosis. Jin Woo Chang graduated from Yonsei University and studied at the University of Chicago (USA). He is also an active functional neurosurgeon doing DBS surgery for movement disorders, pain surgery and psychosurgery. Others are Sung Nam Hwang of Chungang University, Young Soo Kim of Hanyang University, Yong Tae Chung of Busan Baik Hospital, Kyung Jin Lee of Catholic University, Jung Yul Park of Korea University, Seong
Ho Kim of Yeungnam University, Young Hwan Ahn of Ajou University, Jeong Il Lee of Samsung Medical Center, Moo Seong Kim of Busan Baik Hospital and Ryoong Huh of Pochon Cha University. They are all active in the treatment of movement disorders or pain surgery. The Korean stereotactic and functional neurosurgery society was founded on February 24th 1990, and the first president and secretary were Sang Sup Chung and Chang Rok Choi, respectively (> Figure 13‐5). Approximately 150 members were registered with the society in 2006. In 1996, Sang Sup Chung was elected as the president of the Asian Society of Stereotactic, Functional and Computer assisted Neurosurgery. In 1999, he held the third Asian Society meeting in Seoul successfully (> Figure 13‐6). The first Gamma Knife unit was installed in 1990, and now there are 11 gamma knife units
. Figure 13‐5 The first meeting of the Korean Sterotactic and Functional Neurosurgery Society; Seoul, February 24, 1990
History of stereotactic surgery in korea
13
. Figure 13‐6 The third meeting of Asian Society of Sterotactic, Functional and Computer assisted Neurosurgery; Seoul, June 13, 1999
in Korea. In 2006, 2700 gamma knife radiosurgeries were conducted in Korea. The gamma knife radiosurgery society meeting was held on 15th of November 2002 and there have been also annual meetings thereafter. The neurosurgeons involved in radiosurgery are Yong Gou Park of Yonsei University, Dong Gyu Kim of Seoul National University, Young Jin Lim of Kyunghee University, Do Hoon Kwon of Seoul Asan Hospital, and Chang Wha Choi of Busan National University. In 2006, Dong Gyu Kim organized the 13th international meeting of Leksell gammma knife society in Seoul successfully. In the beginning of functional neurosurgery, our pioneers conducted epilepsy surgery, while epilepsy surgery according to a comprehensive proocol began in 1989. The Korean epilepsy
society was founded in 1996 and Sang Sup Chung was elected as the first president. Many active and functional neurosurgeons are participating members of the epilepsy society. Active epilepsy surgeons are Jung Kyo Lee of Seoul Asan Hospital of Ulsan University, Hyung IL Kim of Chunju Presbyterian Hospital, Eun Ik Son of Kyemyung University, Seung Chyul Hong of Samsung Hospital, Chun Kee Chung of Seoul National University, Ha Young Choi of Chonbuk National University and Jong Hee Chang of Yonsei University. Many of the institutes use Leksell stereotactic apparatus while 54 hospitals have Leksell apparatus. Many of the hospitals are using the apparatus simply to conduct brain biopsy or evacuation of hematoma. Nineteen hospitals performed 271 DBSs in 2006; 190 Parkinson’s
177
178
13
History of stereotactic surgery in korea
disease, 34 Essential tremors & 36 Dystonia. There were also clinical experiment of DBS for epilepsy and psychiatric disorders. With more than 60 years of achievement, stereotactic and functional neurosurgery has evolved and become one of important fields of neurosurgery in Korea.
References 1. Commemorating symposium of 122nd anniversary of Chejungwon. Evolution of modern medicine in Korea. Yonsei University Medical Center. 2007. 2. Photographs of 120 years of modern medicine in Korea. Yonsei University Medical Center. 2007. 3. Lee CK. Surgical Treatment of Epilepsy, Preoccipital Coagulation. J Kor Neurosurg. 1972;1:1-14. 4. Forty years History of the Korean Neurosurgical Society (1961~2001). The Korean Neurosurgical Society. 2002. 5. Woo CH, Chang NS. Anterolateral Cordotomy for Relief of Various Intractable Pain. Kor J Soci. 1963;5:121-5.
6. Woo CH. Stereoencephalotomy for Extra Pyramidal System Disorders. J Kor Medi Assoc. 1963;6:248-58. 7. Lee CW. A new method of stereotactic Encephalotomy for Dystonia and Dyskinesia. Modern Med. 1960;3:69-78. 8. Chung SS, Park TS, Kim CS, et al. Percutaneous radiofrequency rhizotomy for Trigeminal neuralgia. J Kor Neurosurg. 1975;4:323-9. 9. Kim SH, Lee KH, Chung SS, et al. Percutaneous cervical radiofrequency Cordotomy for Intractable pain. Yonsei Med J 1975;16:72-82. 10. Doh JW, Park JU, Chung SS, et al. Percutaneous Neurotomy for Clonic Facial Spasm, a case report. J Kor Neurosurg. 1975;4:331-4. 11. Kim SH, Chung SS, Lee HJ, et al. Neurovascular decompression in posterior fossa for Trigeminal Neuralgia. J Kor Neurosurg. 1981;10:469-75. 12. Chung SS, Lee HJ, Lee Ks, et al. Stereotatic radiofrequency hypophysectomy, for disseminated breast and prostate cancer: Transseptal trans sphenoidal approach. Yonsei Med J. 1981;22:53-7. 13. Chung SS, Park YG, Chang JW, Cho J. Long term follow-up results of Stereotactic adrenal medullary transplantation in Parkinson’s diseases. Stereotact Funct Neurosurg. 1994;62:141-7.
14 History of Stereotactic Surgery in Spain J. Guridi . M. Manrique
The history of stereotaxy in Spain dates back to the mid-1950s. Its early history was constrained by the negative impact of the Spanish Civil War on the development of medicine, as on other areas of Spanish society. However, during the 1960s, when various neurosurgeons returned to Spain, the speciality grew. Today it is difficult to find reports of surgical interventions performed by neurosurgeons during that time; in some cases, oral transmission provided the only way of finding out about the development of functional stereotactic surgery in Spain during the 1950s and 1960s. Historical concepts concerning movement disorders developed during the early years of the twentieth century. As a result of early pathological studies on Huntington’s or Wilson’s disease, encephalitis letargica and Parkinson’s disease (PD), which affected subcortical circuits such as the striatum, pallidum, subthalamic nucleus (STN), and substantia nigra, it was discovered that the basal ganglia were involved in motor function. The understanding of the extrapyramidal motor system and movement disorders constituted a turning point in the neurosurgical field. Meyers operated on a patient with postencephalitic parkinsonism in 1939, resecting two-thirds of the caudate head using a transcortical-transventricular approach, and achieving complete arrest of tremor [1]. This pioneering procedure was followed by other operations approaching the caudate nucleus, internal capsula and globus pallidus (GP) with the ansa lenticularis. At that time, the GP was the target of choice for PD and for hyperkinetic disorders; early surgery was performed by open #
Springer-Verlag Berlin/Heidelberg 2009
procedures, and later with the stereotactic armamentarium. The first stereotactic operation in humans was performed in 1947 by Spiegel (a neurologist) and Wycis (a neurosurgeon) in Philadelphia, on a patient with a psychiatric disorder. They called stereoencephalotome to the first frame, and an account of this significant technical advance and its benefits for patients was published in Science [2]. During the first decades of the twentieth century, general surgeons performed brain operations in different hospitals in Spain. Initially, some of them were in close contact, and worked under the direction of a neurologist, because early neurosurgery was not well developed in the country. After the Spanish Civil War (1936–1939) and the Second World War (1939– 1945), the creation of the National Health System facilitated the development of specialties, which contributed to medical advances in the field of the neurosurgery in the 1950s. Dr Obrador was a pioneering neurosurgeon, who spearheaded and worked in specific areas in the field of surgical neurology. On one of their several visits to different neurosurgical departments, Obrador went to Oxford with Sherrington, where Liddel was using stereotactic methods in monkeys. Later, he visited Dr. Fulton in New Haven and Bailey in Chicago, where he came to know and study techniques relating to the implantation of electrodes producing localized deep cerebral lesions. The first Spanish experience in the functional field was with electrostimulation and the coagulation of the thalamus in a patient with localized myoclonic epilepsy and in whom a previous cortical excision had failed [3].
180
14
History of stereotactic surgery in Spain
1955–60 Dr Obrador performed GP lesions in parkinsonian patients in 1955; he described a simplified and open technique, approaching the pallidal region from a temporal burr hole and performing a mechanical lesion with a pallidotome without stereotactic armamentarium [4,5]. The instrument described was a canula with a wire loop at the lower end; the rotation of the instrument (the loop) severs the tissue. However, the wire loop may tear tissue and blood vessels resulting in an irregular and non-predictable lesion. Obrador, along with Dierssen, wrote a groundbreaking report describing 100 lesions, (95 in the pallidum and 5 in the thalamus) in 69 patients (55 parkinsonian and 14 hyperkinetic) [6]. The authors reported that 67 surgical procedures had been carried out, involving mechanical lesions. Nine lesions were placed using the mechanical-chemical procedure; five were made using the Cooper balloon technique; a further nine chemical lesions, with alcohol; and ten through stereotactic electrocoagulation. The results showed that, whatever the surgical technique used, more than half of the lesions (56%) led to a dramatic alleviation of rigidity, tremor and general patient mobility [6]. However, during the follow-up period, from months to 3 years, clinical signs returned in a high percentage of cases. The report’s authors concluded that only 32% of the parkinsonian patients who received surgical treatment could be regarded as improved after that time. The complications arising after open surgical procedures were also detailed: facial paresis appeared in 20 of the 79 lesions; 17 patients showed contralateral hemiparesia; there were speech disorders in 7 cases; and cognitive changes in 8 patients. One patient died after mechanical pallidotomy [6]. This interesting report on early neurosurgery in Spain showed a high degree of failure in the clinical signs of parkinsonian patients in lengthy follow-up periods, along with a high
percentage of complications caused by the open surgical procedure. Consequently, the open approach to deep structures such as basal ganglia without stereotactic apparatus was practically abandoned, because it had become obvious that lesion placement was only well controlled using stereotactic armamentarium. There was no stereotactic frame in the country at that time; thus, a number of Spanish neurosurgeons went abroad to learn how to use different stereotactic instruments. In those days, Irving Cooper was a renowned and charismatic neurosurgeon working in the field of movement disorders and, in particular, PD. G Bravo (1957–60), G Dierssen (1958–61) and F Isamat (1960–61) – all from Spain – worked with him at St Barnabas Hospital in New York. This period encompasses a significant moment in the history of neurosurgery, when the surgical target in parkinsonian patients was shifting from the pallidum to the ventrolateral (VL) of the thalamus. At St Barnabas, Cooper and Bravo developed a technique for lesioning in the GP and thalamus. They introduced a balloon on the target, inflating it with air or a liquid and inserting it into the lumen as a compressive test. If the patient improved, the lesion was later produced with alcohol [7–10]. On the basis of this early technique, in conjunction with chemopallidectomy or chemothalamectomy, rigidity was alleviated by 80% and tremor by 75% in parkinsonian candidates after a 4-year follow-up period [9,10]. Cooper and Bravo showed that if the lesion was performed accurately on the target, the improvement induced as a result held for a long follow-up period. Their surgical procedure produced some adverse effects, such as a mortality of 2.4% and 2% hemiplegia in patients treated [9]. Alcohol injections in the brain’s deep structures also had irregular diffusion in the tissue, lesioning cellular and fiber elements without affecting vessels. Cooper et al reintroduced alcohol through the chemopallidectomy approach,
History of stereotactic surgery in Spain
preparing a mixture of ethyl cellulose and ethanol (etopalin) to prevent diffusion in the target. They reported, as Hassler had in Germany, that the interruption of fibers between GP and thalamus was very important for clinical alleviation in parkinsonian and hyperkinetic disorders [10,11]. The lesions in the GP provided good relief of rigidity but less consistent improvement of tremor [10]; they later reported that the alleviation of tremor and rigidity in parkinsonian patients was obtained after a lesion was placed in the VL of the thalamus. Dr Cooper and Bravo made a striking discovery. Cooper planned to perform a lesion in the posterior part of the GP (15 mm behind Monro foramen) in a tremoric patient; after surgery, the tremor disappeared. However, the patient committed suicide 6 months after surgery, and the brain was analyzed in the hospital. The autopsy, which was carried out by Bravo, revealed that the lesion was placed in the VL of the thalamus. Based on this information, Cooper and Bravo changed the surgical approach from the GP to the VL of the thalamus as “the most effective lesion for producing complete and enduring relief of parkinsonian tremor and rigidity” [7,8]. The surgical results with the GP lesion (pallidotomy) for the alleviation of rigidity and tremor in parkinsonian patients were disappointing because large lesions had to be made in the nucleus if improvement was to be obtained. This prompted Hassler to study the pallidal projections to the thalamus; he showed that most of the pallidal efferences projected in the ventral lateral nuclear region: in his own classification, ventralis oralis anterior (Voa). As a consequence, he proposed performing a lesion on the new target; nevertheless, despite the shift to the thalamus, the surgery did not effect total abolition of parkinsonian tremor [11,12]. Whether the shift was based on scientific discovery through an anatomic rationale or on serendipity, the VL of the thalamus replaced the GP as the target of interest [13]. Thus,
14
thalamotomy became the primary surgical procedure for the treatment of PD, starting in the mid-1950s and reaching a peak during the 1960s. Hyperkinetic patients were also operated on in the most anterior part of the VL thalamic nucleus with good results; thus, in practice, pallidotomy was discarded. Ventro lateral-thalamotomy was carried out on patients with PD or other movement disorders throughout the world. However, the accuracy of the surgical procedure at that time fell far short of what is currently being achieved, mainly because imaging analysis was not yet available [14,15]. A common complication arising in surgery was the induction of a movement disorder, hemichorea or hemiballism (Hc/Hb), especially after thalamotomy. Cooper and Bravo analyzed this complication in their patients: 21 patients developed involuntary movements in 850 consecutive BG operations [16]. The patients affected showed a lesion involving the corpus Luysii (subthalamic nucleus STN). After his return to Spain in 1966, Bravo reported that the incidence of Hc/Hb in patients after surgery could be attributed to the placement of the lesion below the intercomissural line. The duration of the disease and the age of patients also became significant factors determining the appearance of complications [17]. He found that no patient over 50 years of age with a disease-duration of less than 10 years developed Hc/Hb. In contrast, patients under 40 years old and with a diseaseduration of more than 10 years developed Hc/Hb after VL thalamotomy, even if the lesion had not reached the subthalamic area [17]. Dr G Dierssen, another Spanish neurosurgeon who trained with Cooper, carried out different studies in the same area, which showed that not all Hc/Hb complications after surgical procedures were due to STN involvement (> Figure 14-1). Dierssen et al first published an account of the case of a 35-year-old parkinsonian patient with severe tremor and rigidity, on whom a right thalamotomy was performed. The patient
181
182
14
History of stereotactic surgery in Spain
. Figure 14-1 Dr Dierssen (front row left in the photograph) with Dr Cooper (second from left) in Brussels in 1958
developed dyskinesia (Hb) hours after surgery, which lasted for days. A second procedure was undertaken to undo the movement induced but the surgical intervention failed and the movement disorder persisted. The patient died of a pulmonary embolism 5 months after the second surgical procedure; an anatomopathological study showed that no lesions were placed in the STN. The first lesion destroyed the VL nucleus, involving the thalamic fasciculus; and the second was placed in the red nucleus and rubro tegmental fibers [18]. This case was also reported by Cooper in his book “The Vital Probe” as case number 1, and by Gioino as case number 6 [19,20].
1960–70 After his return to Spain, Dierssen continued to analyze Hc/Hb in patients with PD who had undergone surgical procedures, developing the diskynesia ipsilateral in thalamotomy [21] and an hemiballism as the primary manifestation of a chiasm-optic nerve glioma in a child [22]. However, the most interesting study, which is not well-known because it was written in Spanish, was the report on 116 Hb with anatomopathological study [23]. In the conclusion, he held that in a high percentage of cases (65%) with Hc/ Hb—not only surgical patients—multiple anatomical structures were affected and the movement
History of stereotactic surgery in Spain
disorder was not always induced by STN involvement; he pointed out that a number of patients with subthalamic lesion showed no clinical sign. A lesion in the striatum or GP may also induce a movement disorder similar to the subthalamic affect [23]. He reported that the percentage of patients developing diskynesia after thalamotomy increased in cases of postencephalitic parkinsonism and after bilateral surgeries [23]. Dierssen returned to the Hospital La Paz with S Obrador and later (1975) moved to Santander, but during that time the number of surgical procedures carried out on patients with PD decreased through the introduction of levodopa as a treatment for the disease. Hundreds of patients with PD were operated on in two different centers in Madrid at that time—the early 1960s—and in Barcelona (Dr Isamat). The procedure was performed with the stereotactic frame (Cooper frame) and ventriculography with lypiodol. The intercommisural line (ICL) was measured and the target was selected in the medial third of the ICL and 11–14 mm from the midline. A leucotome (McKenney) was introduced and opened near the target; patient tremor and rigidity were evaluated intraoperatively. If tremor was arrested, a mixture of lypiodol and hot wax was injected into the target; if tremor persisted, the wire of the leucotome was opened more. Very few papers and reports about Spanish patients operated on during that period of time exist, but a doctoral thesis recorded 200 patients with bilateral thalamotomies (in a group of 1700 patients operated on by Bravo et al) during a 5-year or longer follow-up period in the 1960s and 1970s [24]. The study showed that tremor, rigidity and hypocinesia were stabilized after bilateral surgery, and the benefit persisted for years. Some of the patients presented speech disorders that correlated with the second lesion volume [24]. Stereotactic surgery spread to other Spanish regions when the neurosurgeons returned after their study-periods abroad. In 1969, V Arjona
14
returned from Newcastle (1962–69) (Dr Hankinson) and began to treat parkinsonian patients in Seville with other movement disorders and pain disorders using stereotactic procedures. He had previously spent time with Denise Albe-Fesard and Guiot in Paris, learning about microrecording for target localization. Recording during surgery was a control procedure to place the lesions more accurately and to obtain better results for a long follow-up period. They used a Leksell frame and performed two burr holes: one frontal, for ventriculography and lesioning; the other occipital, for recording. Dr Arjona favored the criogenic lesion because it was reversible, and temperature control was more accurate with this apparatus than in thermolitic procedures. At that time, the target was the ventralis intermedius (Vim) of the thalamus in parkinsonian tremor; the Voa-Vop (Ventralis oralis posterior) was also lesioned in patients with choreo-athetosis or other dyskinesias. Despite the fact that larger lesions were placed, patients with dystonia responded less satisfactorily to thalamic surgery [25]. The surgical target for pain was the Cm-Pf (Centromedianus-Parafascicular) of the thalamus, and the hypothalamus was the target in surgical candidates with aggressivity and oligophrenia erethica [26,27]. Dr JG Martin Rodriguez also performed surgical stereotactic formation with Hankinson in Newcastle (1967–71). He moved to Madrid in 1971, where he developed functional surgery at La Paz Hospital. He acquired extensive experience in thalamic criogenic lesions in patients with PD or other movement disorders. During surgical recording from the occipital cortex, the electrode crossed the internal capsule reaching the GP or VL of the thalamus. Surgical results showed that thalamo-capsular lesions improved tremor, and GP-capsular lesions rigidity [28,29]. During the 1970s, Dr Martin Rodriguez reported on a number of papers about the electrophysiology of the pulvinar nucleus and the role of this anatomical structure in the mechanism of pain
183
184
14
History of stereotactic surgery in Spain
appreciation. He performed pulvinar lesions, combined with other thalamic nuclei (Vop, Vim), in patients with hypertonic-hyperkinetic syndromes; however, the surgical conclusion after patient evolution was not consistent for clinical alleviation [30–32]. In 1972, he and Dr Delgado placed the first electrode with transdermic stimulation in the septal nuclei of the thalamus of a patient with phantom limb pain, by braquial plexus avulsion (see below) [33]. In 1977, Dr Martin Rodriguez moved to the Hospital Santiago Ramo´n y Cajal, but as a result of the positive response to levodopa in patients with PD the number of thalamic surgeries decreased and all stereotactic activity declined. The neurophysiologist, Dr J.M.R. Delgado (et al), developed the technique of implanting multilead electrodes in the brain for several days (> Figure 14-2). The electrodes were made of enameled stainless steel wires or of coated silver wires, which had the advantage of flexibility. Their diameter (0.1–0.2 mm) permitted the insertion of the active portion without isolate for a long period of time [34]. J.M.R. Delgado was . Figure 14-2 Dr Delgado circa 1965
a physiologist who moved to Yale University in 1950; he was to work with Dr J.F. Fulton there for 20 years. After his return to Madrid, he worked with Dr Obrador and Dr J.G. Martin Rodriguez; during the 1960s, they developed a device connected to the electrodes and controlled by radio waves which was able to record neuronal activity and could also enable the stimulation of internal structures. The device was called the “stinoceiver,” and it was tested in different animal species, including a bull. The device was finally implanted in a patient with a phantom limb in 1968 [35]. Electrodes were implanted in the Caudate head and septal nuclei (near accumbens nucleus), bilaterally, through two frontal burr holes. The transdermic stimulation was “on” for a number of hours daily, alleviating patient pain. This may have been the first implantation of a deep brain stimulation device in Europe.
1970s In the early 1970s, Dr G Bravo (> Figure 14-3) and Dr J Miravet, a neurophysiologist in the Clinica Puerta de Hierro in Madrid, designed a operating room (OR) for stereotactic surgery. This OR was built using the Talairach technique for epilepsy study, and their surgical treatment permitted the implantation of multicontact electrodes in stereotactic conditions to record electrical activity during the onset and propagation of seizures (estereoelectroencephalography: SEEG). The operating table could be rotated through 360 to perform ventriculographies with teleradiography, without magnification or distortion in the studies. The Talairach frame permitted their replacement in patients; this approach had the advantage of supplanting different radiological studies, such as angiography and ventriculography. A room with a glass wall next door to the OR was the neurophysiological laboratory, which had 41 electroencephalography
History of stereotactic surgery in Spain
14
. Figure 14-3 Dr Bravo one of the spanish pioneers in the surgical treatment of Parkinson’s disease in his office in Madrid
channels for recording. The room as a whole was electrically insulated. During that period Dr M Manrique spent 2 years in the Delgado lab and later moved to Paris with Dr J Talairach to study electrode placement in patients with epilepsy [36–38]. He and Dr R Garcı´a de Sola went on to develop the surgical treatment of refractory epilepsy in Madrid. These multiple radiological studies enabled the implantation of subdural electrodes, to define the epileptic zone, and their resection with the advantage of superimposed angiography. Later, M Manrique applied stereotactic procedures for epilepsy (stereoelectrocorticography), with reference to subdural electrodes in MRI (magnetic resonance image) in the Talairach gride, and after identification of the anterior and posterior commissures [36,39,40]. The stereotactic study referenced the position of the electrodes and the epileptogenic zone, the different cortical areas and vascular information in the frame space. The information was used during surgery to facilitate accurate resection. In 1977, in the Hospital Santiago Ramo´n y Cajal, a second OR for stereotactic surgery, similar to the one described above, was designed using a Leksell
frame. Dr JG Martin Rodriguez with JC Bustos, F Figueiras in neurosurgery, and E Garcia Austt and W Bun˜o in neurophysiology treated hundreds of patients during that time. Dr Eiras carried out his stereotactic surgical activity in Zaragoza. In 1966, he went to Freiburg in Germany—the Hassler school, with Riechert and Mundinger. These authors described other targets for tremoric parkinsonian patients, such as the zona incerta placed dorsal to the STN, near the fasciculus lenticularis (campotomies). After his return, using the Riechert-Mundinger frame, Dr Eiras performed campotomies in patients with PD, and also targeted the CM-Pf bilateral with asymmetric lesions in patients with pain disorders and Voa-Vop for dystonia. He described a post-traumatic case with action myoclonus that was alleviated through Vop-Vim thalamotomy [41]. Following the introduction of CT (computer tomography), he and his team pioneered the implantation of stereotactic biopsies with frame and the new image tool. They performed tumoral resections with the help of the stereotactic coordinates, using the Robert Wells (Radionics) equipment for these surgical interventions.
185
186
14
History of stereotactic surgery in Spain
The significant clinical impact of the introduction of levodopa therapy for PD during the 1960s brought about a decline in the surgical treatment of movement disorders. In a large number of surgical departments, such surgery practically disappeared. A few neurosurgeons continued to perform the procedure on patients with levodopa intolerance. Sixteen neurosurgeons were questioned by Laitinen in 1985 regarding their preferred target for PD [42]. The responses reveal a variation in the chosen target (VL, zona incerta, Vim, Forel’s field); however, none of the neurosurgeons used GP as the target for PD during the 1980s [42]. During these years in Spain, only one or two teams carried out operations on parkinsonian patients who did not respond to treatment with levodopa. Nowadays, and over the last 15 years, surgery for advanced PD has been revived. The major reasons for this renaissance are the increased understanding of the pathophysiology of the
basal ganglia and the demonstration of surgical alleviation of experimental parkinsonism. In addition, a large proportion of patients suffer severe complications after years of chronic treatment. Moreover, the significant advances in neurosurgery brought about by the introduction of computers and image development, and deep brain stimulation (DBS) that enables low-risk bilateral surgery, are further factors in this regard. However, this new approach in PD surgery centers on a target that had never before been considered: the STN [43].
1970–80 JL Barcia-Salorio, another pioneer in Spanish neurosurgery, worked and carried out stereotaxy in Valencia (> Figure 14-4). He went to Freiburg in 1960, a renowned school at that time, given the participation of Riechert and Toennis;
. Figure 14-4 Dr J L Barcia-Salorio placing the stereotactic frame for a radiosurgical treatment
History of stereotactic surgery in Spain
however, a few months later, he moved to Sweden to work with Prof Leksell in Stockholm and Dr Larsson in Uppsala. His primary intention was to learn about pallidal lesions in patients with movement disorders, but he was very impressed at the Karolinska Institute by the initial steps taken in radiosurgery, using the gamma-knife in different neurosurgical procedures. After Barcia returned to Valencia, he began to carry out stereotactic surgery on parkinsonian patients, above all, pallidotomies; however, he also inaugurated the use of radiosurgical treatment in Spain. Basing his development on the Leksell frame, he designed and built the Barcia frame, capable of supporting the heavy collimators; cobalt was used as the source of radiation in his groundbreaking approach to radiosurgical treatment. He calculated the dose due to a specific target mathematically, and developed a system based on the tomography in AP and lateral to localize a brain lesion using cartesian coordinates, before the technical innovation of CT and MRI. He localized the coordinate position and lesion size through this early procedure, creating a target volume to be treated [44]. In 1975, he performed the first radiosurgical treatment in Spain; this was reported to the ESSFN (European Society of Stereotactic and Functional Neurosurgery) in 1977. In 1979, he described the radiosurgical treatment of large acoustic schwanomas; later (1982), he performed the first treatment of carotid cavernous fistulas (ccf) using radiosurgery [45–48]. He also developed the introduction of low dosages of radiotherapy in patients with resistant epilepsy, who had previously been treated unsuccessfully with a temporal lobectomy [49]. Prof. Barcia’s most significant contribution to neurosurgery was his pioneering study of radiosurgery for ccf with low flow and the treatment of epilepsy. He established the Valencia school that lives on into the present in the work of Dr J Barbera´, Dr J Broseta and Dr P Rolda´n.
14
In 1990, the first LINAC for radiosurgery was installed in the Sanatorio San Francisco de Asis in Madrid, under the direction of Dr Sambla´s and Dr JC Bustos. Today, there are more than 20 LINAC units in Spain; there is also one gamma knife, in the Clinica Ruber in Madrid. As a young Spanish neurosurgeon, J. A. Burzaco (1932–2006) went to Cheshire (England) (1956–57), and later worked in the Walton Hospital in Liverpool as a registrar (1957–58); his contact with L. Leksell in the Serafirmerlassarettat in Stockholm (1961–62) had a decisive impact on his future development in the stereotactic field (> Figure 14-5). He returned to Spain in 1962, practiced surgery in a number of different hospitals in Madrid; he always had a great interest in the neurosurgical treatment of psychiatric diseases. On the basis of Egas Moniz’s description of such surgery in psychiatric patients,
. Figure 14-5 Dr Burzaco, a pioneer figure in psychiatric surgery in Spain also worked with the gamma knife
187
188
14
History of stereotactic surgery in Spain
lobotomies were regularly carried out in Spain at that time. Burzaco combined both approaches: stereotactic surgery and the selection of psychiatric candidates [50]. He reported his experience with thermolesions in the anterior arm of internal capsula (capsulotomies), using the Leksell’s target in patients with obsessive neurosis and severe autoaggressive compulsions, and describing 108 capsulotomies in patients with obsessive compulsive disorders (OCD) [51,52]. Lesions were placed bilaterally and symmetrically, and the follow-up period was either one or 2 years. His results showed a 73% rate of great alleviation (symptom-free or an improvement in normal life); that is, in 62 of the 85 patients on whom he operated. His analysis revealed that improvement was greater among patients who were 30 years old or younger, as compared with patients who were 40 years old or older (83.3% vs. 67%) [51,52]. Dr Burzaco’s most significant contribution to the field of psychiatric surgery was thermolesion in the stria terminalis (fascicle from amygdala to septal nuclei in the hypothalamus) in complex patients experiencing severe character, aggressivity and behavior changes. He first described the technique in a young patient after an encephalitis, but the surgery was later performed on a selection of patients with violent aggressive behavior [53]. The target was placed 3 mm anterior to the anterior commissure, 5–6 mm lateral to the wall of the third ventricle, and at the level of the intercommissural line. However, the complexity of the cases treated led to the formation of combined lesions. The fundus of the stria terminalis lesion was placed in 90% of 116 patients with aggressivity, and a posteromedial hipothalamotomy, a capsulotomy or a cinculotomy was also performed as a complementary target [54]. A second operation was carried out on 27% of the patients, which he himself evaluated as showing a dramatic reduction in aggressive behavior, with no mortality and very few complications. Dr Burzaco was a
pioneer in the field of the surgical treatment of complex cases involving psychiatric disorders, and was renowned for his extensive experience throughout the country. His description of the pathophysiology of aggressivity showed that components of the limbic system (hippocampus and amygdala), as well as cortical areas, such as the orbital and prefrontal cortex, were implicated. His experience led him to conclude that major deafferentization should be carried out to alleviate the clinical condition of chronic psychiatric diseases. A bilateral striae terminalis lesion and—in a high percentage of cases—a bilateral cingulotomy were required to disrupt cortical and deep limbic anatomical pathways in patients with aggressive behavior [54,55].
Final Remarks Stereotactic localization within the brain, using CTor MRI, and the introduction of computers in neurosurgery have marked significant advances in this field. The target may now be visualized in tumors, and may also be recorded in the case of a functional anatomical deep structure. Following target selection, computers automatically introduce the coordinate information of the frame, reducing surgical time with a high degree of accuracy. During the 1970s and 1980s, different neurosurgical departments implemented these techniques for brain biopsies, and later for craneotomies with frame. Tumoral pathology advanced with the help of stereotaxy for years, until the introduction of neuronavigation in neurosurgery in current times. A survey carried out by the Stereotactic and Functional group of the Spanish Society of Neurosurgery (2002) in Spain showed that 50% of departments used stereotaxy and/or neuronavigation for tumoral pathologies; 15 of 34 used stereotaxy alone as a technique; only two groups used neuronavigation without stereotactic frame [56]. At that time, there were 16 Radionics
History of stereotactic surgery in Spain
frames, 14 Leksell, two Barcia, and only one Laitinen and Riecher-Mundinger frame. In 1 year, 656 stereotactic biopsies were performed in thirty-four surgical departments, 125 craneotomies guided by stereotaxy, and 95 cases of cyst or abscess evacuation using frame help. Eleven departments performed functional stereotactic procedures on 115 previously operated PD patients and 10 patients with other movement disorder pathologies. In target planning, nine groups used only TAC, five used MRI, and ten groups fusion image; only one group performed ventriculography for intercommissural reference [56]. Stereotactic and functional neurosurgery is currently one of the major areas in neurosurgery, in Spain, as in other developed countries: its future is wide and open. Current applications and procedures include DBS techniques, ablative procedures, along with drug delivery, gene therapy and cell transplantation in degenerative disorders. The patients involved suffer from PD, essential tremor, dystonia, chorea, multiple sclerosis, pain, epilepsy and psychiatric disorders, such as depression and OCD. Surgical techniques have evolved over the course of many years, but patients continue to suffer the burden of their diseases in ways similar to patients in the past.
Acknowledgments We would like to acknowledge the information provided and opinions shared by M Arra´zola, F Aguilera, V, Arjona, K Buus, M Dierssen, J Eiras, R Figueiras, F Isamat, R Martinez, J Molet, and JG Martı´n Rodriguez.
References 1. Meyers R. Surgical procedure for postencephalitic tremor, with notes on the physiology of the premotor fibers. Arch Neurol Psychiatry 1940;44:455-9.
14
2. Spiegel EA, Wycis HT, Marks M, Lee AJ. Stereotaxic apparatus on the human brain. Science 1947;106:349-50. 3. Obrador S. Personal recollections of the development of human stereotactic neurosurgery. Confinia Neurol 1975;37:378-83. 4. Obrador S, Dierssen G. Cirugı´a de la regio´n palidal en el sı´ndrome de Parkinson. Te´cnica personal y resultados inmediatos en los seis primeros casos operados. Revista Clı´nica Espan˜ola 1956;61:229-37. 5. Obrador S. A simplified neurosurgical technique for approaching and damaging the region of the globus pallidus in Parkinson’s disease. J Neurol Neurosurg Psychiatr 1957;20:47-9. 6. Obrador S, Dierssen G. Results and complications following one hundred subcortical lesions performed in Parkinson’s disease and other hyperkinesias. Acta Neurochi 1959;7:206-13. 7. Cooper IS. Surgical treatment of parkinsonism. Ann Rev Med 1965;16:309-30. 8. Cooper IS, Bravo G. Chemopallidectomy and chemothalamectomy. J Neurosurg 1958;15:244-56. 9. Bravo GJ, Cooper IS. Chemopallidectomy: two recent technical additions. J Am Geriatr Soc 1957;5:651-5. 10. Bravo GJ, Cooper IS. A clinical and radiological correlation of the lesions produced by chemopallidectomy and chemothalamectomy. J Neurol Neurosurg Psychiatr 1959;22:1-10. 11. Hassler R. The pathological and pathophysiological basis of tremor and parkinsonism. Second International Congress of Neuropathology, Excerpta Medical Foundation 1955;1:29-40. 12. Hassler R, Riechert T. Indikationen und Lokalisationsmethode der gezielten Hirnoperationen. Nervenarzt 1954;25:441-7. 13. Guridi J, Lozano AM. A brief history of Pallidotomy. Neurosurgery 1997;41:1169-83. 14. Smith MC. Location of stereotactic lesions confirmed at necropsy. Br Med J 1962;1:900-6. 15. Cooper IS, Bergmann LL, Caracalos A. Anatomic verification of the lesion which abolishes parkinsonian tremor and rigidity. Neurology 1963;13:293-315. 16. Schachter JM, Bravo G, Cooper IS. Involuntary movement disorders following basal ganglia surgery in man. J Neuropathol Exp Neurol 1960;19:228-37. 17. Bravo G, Parera C, Seiquer G. Neurological side-effects in a series of operations of the basal ganglia. J Neurosurg 1966;24:640-7. 18. Dierssen G, Bergmann L, Gioino L, Cooper IS. Hemiballism following surgery for Parkinson’s disease. Arch Neurol 1961;5:627-37. 19. Cooper IS. The vital probe: my life as a brain surgeon. New York: Norton; 1981. p. 293-315. 20. Gioino GG, Dierssen G, Cooper IS. The effect of subcortical lesions on production and alleviation of hemiballic and hemichoreic movements. J Neurol Sci 1966;3:10-36.
189
190
14
History of stereotactic surgery in Spain
21. Dierssen G, Gioino G, Cooper IS. Participation of ipsilateral hemisphere lesions in the pathology of hemichorea and hemiballismus. Neurology 1961;11:894-8. 22. Dierssen G. Sobre un caso de hemibalismo producido por un glioma de ta´lamo o´ptico. Rev Clinica Espan˜ola 1962;84:339-42. 23. Dierssen G, Gioino G. Correlacio´n anatomica del hemibalismo. Rev Clin Esp 1961;82:283-305. 24. Herrero J. Resultados tardı´os de la cirugı´a en la enfermedad de Parkinson. Situacio´n clı´nica de 200 enfermos despue´s de cinco o ma´s an˜os de haber sido operados bilateralmente. Doctoral Thesis, Universidad Autonoma de Madrid; 1980. 25. Burzaco J. Stereotactic pallidotomy in extrapyramidal disorders. Appl Neurophysiol 1985;48:283-7. 26. Arjona V, Hankinson J. Cirugı´a tala´mica del dolor. Rev Clin Espa 1969;113:313-22. 27. Arjona VE, Martin-Rodriguez JG. Cryohypothalamotomy for the treatment of aggressive behavior in oligophrenia erethica. Neurological surgery. Amsterdam: Excerpta Medica; 1978. p. 320-4. 28. Gillingham FJ, Watson WS, Donaldson AA. The surgical treatment of parkinsonism. Brit Med J 1960;12 Nov:1395-402. 29. Gillingham FJ, Kalyanaraman S, Donaldson AA. Bilateral stereotaxic lesions in the management of parkinsonism and the dyskinesias. Brit Med J 1964;2:656-9. 30. McComas AJ, Wilson P, Martin-Rodriguez JG, Wallace AC, Hankinson J. Properties of somatosensory neurons in the human thalamus. J Neurol Neurosurg Psychiatr 1970;33:716-7. 31. Martin-Rodriguez JG, Bun˜o W, Jr, Garcia-Austt E. Human pulvinar units, spontaneous activity and sensorymotor influences. Electroenceph Clin Neuropath 1982;54:388-98. 32. Martin-Rodriguez JG, Obrador S. Evaluations of stereotaxic pulvinar lesions. Conf Neurol 1975;37:56-62. 33. Obrador S, Delgado JMR, Martin-Rodriguez JG, SantoDomingo, Alonso A. Estimulacio´n cerebral transdermica en miembro fantasma doloroso. Revista espan˜ola de OtoNeuro-Oftalmologı´a y Neurocirugı´a 1972;30:269-72. 34. Delgado JMR. Permanent implantation of multilead electrodes in the brain. Yale J Biol Med 1952;24:351-8. 35. Delgado JMR, Mark V, Sweet W, Erwin F, Weiss G, et al. Intracerebral radio stimulation and recording in completely free patients. J Nerv Ment Dis 1968;147:329-40. 36. Talairach J, Peragut JC, Farnarier P, Manrique M. The role of the stereotactic radiographic exploration in neurosurgical interventions. In: Salomon GG, editor. Advances in cerebral angiography. Berlin: Springer; 1976. p. 272-3. 37. Manrique M, Alborch E and Delgado JMR. Cerebral blood flow and behavior during brain stimulation in the goat. Am J Physiol 1977;232:495-9.
38. Bancaud J, Talairach J, Geier S, Bonis A, Trorrier S, Manrique M. Manifestations comportamentales induites par la stimulation electrique du gyrus cingulaire anterior chez l’homme. Rev Neurol (Paris) 1976;10:705-24. 39. Manrique M, Rodriguez Albarin˜o M, De la Torre M, Blazquez MG. Stereocorticography in the surgical treatment of epilepsy. Stereot and Funct Neurosurg 1994;63 (1–4):101. 40. Manrique M, Parera C, Nombela L, Miravet J, Bravo G. Stereotaxic radiographic and computerized axial tomographic explorations in the surgical treatment of epilepsy. Presented at the third meeting of European Society for Stereotactic and Functional Neurosurgery (ESSFN). Freiburg 1977. 41. Eiras J. Sı´ndrome mioclo´nico postrauma´tico. Efectividad de las lesiones tala´micas sobre las mioclonı´as de accio´n. Arch Neurobiol 1980;43:17-28. 42. Laitinen LV. Brain targets in surgery for Parkinson’s disease. Results of a survey of neurosurgeons. J Neurosurg 1985;62:349-51. 43. Guridi J, Luquin MR, Herrero MT, Obeso JA. The subthalamic nucleus: a possible target for stereotaxic surgery in Parkinson’s disease. Mov Disord 1993;8:421-9. 44. Barcia-Salorio JL, Barbera´ J, Broseta S, Soler F. Tomography in stereotaxis. A new stereoencephalotome designed for this propose. Acta Neurochir 1977;Suppl 24:77-83. 45. Barcia Salorio JL. Radiosurgical treatment of huge acoustic neurinomas. In: Szikla, editor. Proceedings of the inserm symposium No. 12, 1979. Oxford, UK: Elsevier; p. 245-9. 46. Barcia-Salorio JL, Hernandez G. Stereotactic radiosurgery in acoustic neuninomas. Acta Neurochir 1984;Suppl 33:373-6. 47. Barcia Salorio JL, Hernandez G, Broseta J. Radiosurgical treatment of carotid-cavernous fistula. Appl Neurophysiol 1982;45:520-2. 48. Barcia-Salorio JL, Roldan P, Hernandez G. Radiosurgical treatment of epilepsy. Appl Neurophysiol 1985;48:400-3. 49. Barcia Salorio JL, Soler F, Hernandez G. Radiosurgical treatment of low flow carotid-cavernous fistulae. Acta Neurochir 1991;Suppl 52:93-5. 50. Burzaco JA, Gutierrez Gomez D. Trastorno de conducta postencefalitico (su tratamiento por cirugia estereotaxica). Archivos de Neurobiologia 1968;31:69-77. 51. Lopez-Ibor JJ, Burzaco J. Stereotaxic anterior limb capsulotomy in selected psychiatric patients. In proceedings of the second international conference in psychosurgery, Copenhagen, Denmark, Springfield, IL: Charles Thomas Publishers; 1970. p. 391-9. 52. Lopez-Ibor JJ, Lopez-Ibor JA, Burzaco JJ, Duque del Rio M. Capsulotomia esterotaxica. Indicaciones y Resultados. Actas Luso Espan˜olas de Neurologı´a Psiquiatrica y Ciencias Afines 1974;3:219-24. 53. Burzaco JA. Fundus Striae terminalis, an optional target in sedative stereotactic surgery. Surgical Approaches in Psychiatry. In: Laitinen L, editor. Third international
History of stereotactic surgery in Spain
congress of psychosurgery. Chapter 20, Lancaster: MTP; 1973. p. 135-7. 54. Burzaco J. The role of some limbic structures in aggressive behavior. In: Gris P, Struwe G, Jansson B, editors. Elsevier. North-Holland. Biomedical Press. Biological psychiatry; 1981. p. 1223–1126.
14
55. Burzaco JA. Stereotactic surgery in severe unresponsive aggressive syndromes. Biological Psychiatry Vol. 2 Ed Racagni. Elsevier Science Publishers BV; 1991. p. 241–244. 56. Guridi J. Stereotactic neurosurgery in Spain. Neurocirugı´a 2002;13:406.
191
3 History of Stereotactic Surgery in US P. L. Gildenberg
The early use of functional neurosurgery was begun by Horsley [1] in England at the end of the nineteenth century, and the invention of stereotactic surgery in animals was begun by Horsley and Clarke [2] at the beginning of the twentieth century. Human stereotactic surgery began almost at mid-century, i.e., 1947, in the United States by Ernest A. Spiegel and Henry T. Wycis [3] at Temple Medical School in Philadelphia. Stereotactic surgery was accepted at so many institutions so quickly that it is difficult to identify individual ‘‘schools’’ of stereotactic surgery, as it is in other countries. Indeed, in the early 1960s many neurosurgeons, who did not publish, were practicing functional neurosurgery, but most quickly left the field when L-dopa was introduced near the end of that decade [4] and have not returned to the field [5]. In the first half of the twentieth century, prior to the introduction of human stereotaxis, several American functional neurosurgeons were dominant in the use of non-stereotactic techniques for movement disorders. Paul Bucy, in Chicago, recommended motor cortex extirpation or corticospinal tractotomy for the treatment of athetosis [6] and Parkinson’s disease [7]. Bucy [7,8] had declared well into the 1950s that it was necessary to damage the corticospinal system in order to achieve relief from involuntary movements. Dandy [9], based on observation of several patients with intracerebral hemorrhage, had averred that damaging the extrapyramidal system would result in permanent intractable coma. Earl Walker, of Johns Hopkins, reported a pedunculotomy wherein he sectioned the lateral #
Springer-Verlag Berlin/Heidelberg 2009
two-thirds of the peduncle for relief of hemiballismus [10] or parkinsonian tremor [11]. It was in attempting the Walker procedure that Irving Cooper had his famous ‘‘surgical accident,’’ where he accidentally cut the anterior choroidal artery and found the patient much improved, which led him to advocate ligation of that vessel for treatment of movement disorders [12], as discussed below. The most important but often overlooked American functional neurosurgeon of the prestereotactic era is Russell Meyers of the University of Iowa. It was his pioneering work that proved both Paul Bucy and Walter Dandy wrong. In 1939, Meyers [13,14] performed a craniotomy and transventricular approach to resect the head of the caudate nucleus for successful treatment of Parkinsonian tremor. His observations were presented at a meeting of the Research Association for Nervous and Mental Disease the following year [14], where a number of senior members encouraged him to pursue these observations, which led to the publication of results in 39 patients operated with various open surgical procedures of the basal ganglia [15]. The results were good, but the mortality rate was so high that even Meyers recommended against such surgery. His observations, however, were critical to defining the first extrapyramidal targets, significant information that led directly to the first stereotactic surgery for a movement disorder, Huntington’s chorea, with the pallidum as the target [3]. Stereotactic surgery produced even better results and brought the mortality rate down to 1% within 3 years of its introduction [16]. In 1963, Meyers resigned his appointment
46
3
History of stereotactic surgery in US
at the University of Iowa, reportedly for personal reasons, and became Chief of Neurosurgery of the Appalachian Regional Hospital at Williamson, West Virginia. It was not uncommon thereafter for the question to be raised at each Stereotactic Society Meeting, ‘‘Whatever happened to Russell Meyers?’’ Ernest A. Spiegel (> Figure 3-1) graduated from the University of Vienna, where he remained on the faculty both as a neurologist and laboratory scientist. As anti-Semitism advanced in Austria during the 1930s, he found himself with fewer students and restricted access to his laboratory. A businessman who observed this on a trip to Vienna related Spiegel’s plight to Dr. William Parkinson, the Dean of Temple Medical School, who invited not only Ernest Spiegel but also his wife, Mona Spiegel-Adolph, PhD, Professor of Colloid Chemistry, to join the faculty in Temple, where they both worked until retirement. Spiegel was small in stature, with wirerimmed glasses, a somewhat unkempt look, and a thick accent which identified him as a professor
steeped in the European academic philosophy. He remembered everything, and as he wrote his manuscripts (usually with a pencil stub on the back of used EEG paper), he inserted the citations without the need to consult his library. His laboratory consisted of two small and cramped rooms – one lined with shelves that contained an uncounted number of cat brains in baby food jars and the other containing a Faraday cage for recording. Inside the cage was an animal stereotaxic apparatus on a small table. Although many considered Spiegel severe because of his professorial look, he had a wry sense of humor, which was a necessary trait to work with Henry T. Wycis, who was a practical joker. Henry Wycis (> Figure 3-2) was incorrigible. He came from a middle class family, and helped supplement his living expenses by playing semiprofessional baseball during college, which seemed remarkable, because he weighed over 300 pounds by the time I met him. He began to work with Spiegel when he was a medical student – although his academic record was unsurpassed for many years, he also held the record
. Figure 3-1 Ernest A. Spiegel
. Figure 3-2 Henry T. Wycis
History of stereotactic surgery in US
for having missed the most classes, which may have accounted for his time in Spiegel’s laboratory. They continued to work together while Wycis was a neurosurgical resident at Temple and after he had joined the faculty. Just the opposite of Spiegel, Henry Wycis loved to be with a diverse group of people. He enjoyed doing card tricks for his friends. It was rumored that he had earned his way through Medical School by successfully playing poker, which may be because of his photographic memory or may explain his love of card tricks. After stereotactic techniques were introduced in 1947 [3], many neurosurgeons visited Spiegel and Wycis at Temple Medical School in Philadelphia to learn of this new field and return home to make their own apparatus, (since none were commercially available for approximately the next decade,) and become practitioners of this new discipline. Although many of the visitors were from Europe, many were also US neurosurgeons. Even though the field was born in the US, the stereotactic community was and remains truly an international community. Those most active in the field during the 1950s, probably fewer than 30 surgeons, met irregularly at various institutions. The information exchange was informal, open, and enthusiastic, since those scientists shared the excitement of participating in a new science. The first patient reported by Spiegel and Wycis in 1947 [3] had Huntington’s chorea. His involuntary movements became more severe when he was stressed. Consequently, two lesions were made, one in the globus pallidus for the involuntary movement and one in the dorsomedial nucleus of the thalamus to lessen the stress reaction. Lesions were made with alcohol injection, in hopes of sparing the fibers en passage. The next patients had lesions made with a direct current, which had been described in detail in Horsley and Clarke’s original animal stereotaxis paper [2]. Spiegel and Wycis concluded their paper by commenting, ‘‘the apparatus is being
3
used for psychosurgery. . . Lesions have been placed in the region of the medial thalamus. . .Further applications of the stereotaxic technique are under study, e.g., interruption of the spinothalamic tract in certain types of pain or phantom limb; production of pallidal lesions in involuntary movements; electrocoagulation of the Gasserian ganglion in trigeminal neuralgia; and withdrawal of fluids from pathological cavities, cystic tumors.’’ Spiegel was somewhat secretive about future plans, and I am certain that they had already done those procedures prior to the first publication. I first met Spiegel and Wycis during my freshman year at Temple Medical School. A new summer research program had been announced, and I was looking for someone in the neurological sciences to be my sponsor and guide my research. I was referred by the head of physiology, since there was no neurophysiologist in that department I walked into Spiegel’s laboratory to find him and Wycis reviewing pre- and postoperative 8-mm motion picture films of patients. They invited me to sit and watch, and 2 hours later I was accepted as Spiegel’s graduate student. This was in the spring of 1956, just 9 years since the field of stereotactic surgery had begun, when it was still in its infancy. Spiegel and Wycis then performed surgery on only one patient per week. Targeting required a pneumoencephalogram, which was performed with the apparatus in place on Tuesday. The patient was so sick from the study that the surgical part was delayed until Thursday, when the apparatus was re-applied and the procedure done. Every time an electrode was inserted into the brain of an awake patient was an opportunity to study human neurophysiology, which both helped our understanding of the human extrapyramidal nervous system and provided physiologic confirmation that the electrode was at the intended target. In 1968, the International Society for Research in Stereoencephalotomy was formed, along with the American and European
47
48
3
History of stereotactic surgery in US
Branches. The officers of the American Branch, which included both Canada and the United States, included Ernest Spiegel as President, Lyle French as Vice President, and Henry Wycis as Secretary-Treasurer. Claude Bertrand of Montreal was the sole Canadian member of the Board, and served along with Blaine S. Nashold, Jr., Vernon Mark, Robert W. Rand, and Orlando Andy. At the next official meeting in 1970, those same neurosurgeons remained officers, an indication of how small the membership was, with French as the President, and the non-neurosurgeon Spiegel resigned from his office. At the next meeting in 1972, John Alksne was a new member of the Board, in place of Lyle French. The organization was revamped in 1973 as the World Society for Stereotactic and Functional Neurosurgery, and the American Society for Stereotactic and Functional Neurosurgery (as well as the European Society) were formed. It was by that time 25 years since the introduction of stereotactic surgery, and each of the founders had taught a number of trainees, so the field had been expanding rapidly. The officers of the American Society retained Blaine Nashold as President and Claude Bertrand as Vice-President. I took over as Secretary shortly after Henry Wycis’ death in 1972, and held that position in both the American Society and the World Society for the next 28 years, except for the years I was President of the American Society from 1977 to 1980 and President of the World Society for Stereotactic and Functional Neurosurgery from 1993 to 1997. The evolution of stereotactic and functional neurosurgery in the United States took a somewhat different course than in most of the rest of the world. Up until the last decade, most stereotactic surgeons performed such surgery as part of a general neurosurgical practice, or one which also featured other subspecialties. There were few ‘‘stereotactic centers,’’ even in universities, as was the norm in other countries where medicine
was directed by government or academic agencies. Both medical school and private neurosurgical practitioners devoted part of their practice to stereotaxis, but also performed other neurosurgery, as well. This is in contrast to the last decade, where more sophisticated technology requires a team, so that multispecialty stereotactic services are more common, usually centered around teaching programs, such as Pat Kelly at NYU, Ali Rezai at the Cleveland Clinic, and others. In the pre-dopa days, the largest functional neurosurgical practice in the US was that of Irving Cooper, who devoted his practice to functional neurosurgery after the mid-1950s [17]. He embarked on functional neurosurgery because of a ‘‘surgical accident’’ [12]. He was performing a Walker [11] pedunculotomy for Parkinson’s disease, when he accidentally cut the anterior choroidal artery. He aborted the procedure, but the patient awoke with marked improvement. That led Cooper to advocate ligation of that vessel for the treatment of Parkinson’s disease [12]. The distribution of that artery, however, is very variable, and so were the results of its ligation. By that time, the pallidum had become a common target for stereotactic treatment of parkinsonism [18], so Cooper advocated injecting alcohol into that structure in so-called chemopallidectomy [19]. Again the results were variable, since the insertion of the needle was free-hand and the alcohol spread in an uncontrolled fashion along the adjacent tracts. He tried using a thicker solution, Etopalin, and a cannula with a balloon at the end in hopes of making a cavity that would contain the injected solution [20]. Neither of these maneuvers produced a more predictable lesion [21,22]. Finally, he recruited a freelance engineer, Arnold StJ. Lee, who designed the cryoprobe, which used a controlled release of liquid nitrogen through a probe to freeze the tissue at the tip [23]. Although it had a large blunt tip that injured tissue on insertion, it produced a predictable lesion at the tip. It was purely coincidence
History of stereotactic surgery in US
that the engineer who built the Cryoprobe was the same Lee who is the fourth author on the first stereotactic paper by Spiegel and Wycis [3] in 1947. Cooper used an aiming device that was not truly stereotactic, in that it was not based on a Cartesian system. He approached the pallidum by inserting a cannula ventral and medial through the temporal lobe. One of his patients, who had an excellent result, was killed in an accident. An autopsy involving the brain showed that the lesion was in the ventrolateral thalamus, a target that had already been described as preferable for tremor. Cooper changed his target to that structure and reported another large series of chemopallidectomy. Cooper brought functional neurosurgery, including stereotactic surgery, to the public through the mass media. One of his patients was Margaret Bourke-White, the famous Life magazine photographer who suffered from Parkinson’s disease. She insisted that her surgery be photographed by one of her colleagues, Alfred Eisenstadt, who was equally famous. Her procedure and result were excellent, and the pictures and article in a mass circulation magazine brought considerable attention to Dr. Cooper and to stereotactic and functional neurosurgery. There have been persistent stories about clashes between Cooper and Wycis at stereotactic meetings. Some of the more colorful stories had them coming to physical blows. Not only is that not true, as far as I can document, but that competition would have gone to Wycis. He was a bear of a man, who weighed more than 300 pounds, and was a semi-professional athlete in his college years. There certainly were verbal assaults, however, and a sense of one-upmanship when they disagreed. I was present at a stereotactic meeting at Temple Medical School in 1958, when such a competition occurred. Cooper brought one his successful patients to the meeting to show how well he could write on a blackboard after surgery. Not to be outdone,
3
Henry Wycis brought a patient to the meeting the following day to demonstrate how well he played the piano after a pallidotomy. We have lost much of the color of those early meetings. There has been a recent shift toward implanting deep brain stimulators, but story actually began in the late 1960s. The development of the majority of neuromodulation devices presently used occurred in the United States, mainly through Medtronic1, that remains the dominant supplier of implantable stimulators [24]. The introduction of the gate theory by Melzack and Wall [25] in 1965 led Norm Shealy, a neurosurgeon at Western Reserve Medical School (now Case Western Reserve) in Cleveland, to consider stimulating the dorsal columns of the spinal cord to ‘‘close the gate’’ for relief of chronic pain [26]. He worked with Tom Mortimer, who previously had spent time working at an American company, Medtronic1, that at time had several implantable stimulators on the market. Their Barostat stimulator that was used for stimulation of the carotid nerve for management of hypertension in 1963 and their Angiostat in 1965 to treat angina, were adapted to electrodes designed by Mortimer to stimulate the spinal cord, and the field of neuromodulation using commercially available implantable stimulators was born [24]. Shealy left neurosurgery soon thereafter to become a horse rancher in Wisconsin and write mystery books. It was in 1973 that Hosobuchi [27], then at the University of California at San Francisco, inserted an electrode into the somatosensory thalamus to treat denervation pain, and deep brain stimulation became a reality. Shortly after that, Don Richardson [28,29], of Tulane in New Orleans, stimulated the periaqueductal area for management of somatic pain. At around the end of the 1970s, the use of deep brain stimulation required approval by the FDA, but only one of the three companies manufacturing implantable stimulators performed the necessary studies to document its benefit in pain; the third company
49
50
3
History of stereotactic surgery in US
was Avery, who obtained approval, but their founder, Roger Avery, retired just at that time, so the use of deep brain stimulators was deapproved. The use of deep brain stimulation ceased in the US. It was not until the use of implantable deep brain stimulators for motor control was documented by Alim Benabid [30] and later by Jean Siegfried [31], colleagues in Europe, that there has been a re-ignition of DBS activity in US centers. This led to the reapproval of DBS for movement disorders by the FDA in early 2002. Lars Leksell [32,33] introduced the Gamma Knife for stereotactic radiosurgery in Stockholm. The first unit installed in the United Staates was guided through the vast regulatory bureaucracy by Dade Lunsford [34] at the University of Pittsburgh, which became one of the most active Gamma Knife centers anywhere. Once the regulatory problems had been addressed by Lunsford, many Gamma Knives were imported into the US, sometimes directed by neurosurgeons who had previously worked in Sweden, such as Ladislau Steiner at the University of Virginia and Georg Noren at Rhode Island Hospital. Although linear accelerator stereotactic radiosurgery was invented by Leksell [32] in Europe, he used primarily his Gamma Knife. Later Betti [35] and Colombo [36] reintroduced the linac for radiosurgery. Several US neurosurgeons also became active in that time when the benefits of radiosurgery were becoming apparent, but there were few Gamma Knives in this country. Linac systems were developed at the University of Florida at Gainesville by William Friedman and Frank Bova [37]. Several neurosurgeons including Peter Heilbrun were involved in development of the Radionics XKnife, which was first used at the Joint Center in Boston by Jay Loeffler [38]. The use of proton beam therapy was reported by Ray Kjellberg [39] as early as 1962. During the past few years, a Proton Beam Center was opened at Loma Linda and more recently at the MD Anderson Cancer Center in Houston.
Pain management has always been a major interest of US neurosurgeons. An appreciation for the evolving philosophy of neurosurgical management of pain can be obtained by perusing the three volumes co-authored by William Sweet, who was the epitome of professorship. The first was by James White and Sweet [40] in 1955, which emphasized the interruption of the primary pain pathways. The second by those same authors appeared in 1969 [41], and provided a somewhat more conservative approach that included the extralemniscal pathways, as well. The third by Jan Gybles of Belgium and Sweet [42] in 1989 emphasized the complexities of pain perception as a guide to management. Anterolateral cordotomy was simplified by Sean Mullan [43], who in 1963 reported a technique of lesioning that part of the spinal cord by percutaneous insertion of a strontium needle at the C1–2 level for a measured duration. The technique was further modified in 1965 by Hu Rosomoff [44], who used a radiofrequency lesion to interrupt the lateral spinothalamic tract, making it accessible to most neurosurgeons. Paul Lin and I [45,46] introduced a technique that introduce the needle through a lower cervical disk, thus avoiding fibers concerned with respiration. It is more than coincidence that the two neurosurgeons with the largest series of percutaneous cervical cordotomies, Rosomoff and I, came to recommend a very conservative approach to surgery and favored comprehensive multidisciplinary management of chronic pain [47,48]. Both of our programs resembled the comprehensive pain management program pioneered at the University of Washington in Seattle, which was led by the anesthesiologist John Bonica [49], and the neurosurgeon John Loeser [50,51]. Interest in stereotactic and functional neurosurgery has been increasing in the United States, especially since the use of deep brain stimulation for motor disorders became available in 2002. The field is of interest to neurologists
History of stereotactic surgery in US
and neurophysiologists, as well as neurosurgeons, which fostered a team approach to management of Parkinson’s disease and other movement disorders. This has developed to the point where the use of intraoperative microelectrode recording has become the norm. Not only has the field become more active, but the scientific basis of the diseases and techniques are being studied with increased intensity in order to assure further progress. Because of the increased complexity and sophistication of electrode implantation, many multidisciplinary centers have developed in the US, such as David Roberts at the DartmouthHitchcosk Medical Center, Michael Kaplitt at Weill Cornell Medical College in New York, Pat Kelly at New York University, Ali Rezai at the Cleveland Clinic, Ray Bakay in Atlanta and then Chicago, then Robert Gross in Atlanta, Michael Schulder in Manhasset, NY, Philip Starr and Nicholas Barbaro at the University of California in San Francisco, Jamie Henderson at Stanford, Tony DeSalles and Mike Apuzzo in Los Angeles, and Kim Burchiel in Portland Oregon, to name but a few. In 1987 I asked, ‘‘Whatever happened to stereotactic surgery?’’ [5] The answer is, ‘‘It is doing well and advancing at an unprecedented rate.’’
References 1. Horsley V, Taylor J, Colman WS. Remarks on the various surgical procedures devised for the relief or cure of trigeminal neuralgia (tic douloureaux). BMJ 1891;2:1139-43. 2. Horsley V, Clarke RH. The structure and functions of the cerebellum examined by a new method. Brain 1908;31:45-124. 3. Spiegel EA, Wycis HT, Marks M, Lee AS. Stereotaxic apparatus for operations on the human brain. Science 1947;106:349-50. 4. Cotzias GC, VanWoert MH, Schiffer LM. Aromatic amino acids and modification of parkinsonism. N Eng J Med 1967;276:374-9. 5. Gildenberg PL. Whatever happened to stereotactic surgery? Neurosurgery 1987;20:983-7.
3
6. Bucy PC, Buchanan DN. Athetosis. Brain 1932;55:479-92. 7. Bucy PC, Case TJ. Tremor. Physiologic mechanism and abolition by surgical means. Arch Neurol Psychiat 1939;41:721-46. 8. Bucy PC. Cortical extirpation in the treatment of involuntary movements. In: Putnam TJ, editor. The Diseases of the Basal Ganglia. New York: Hafner Publishing Co.; 1966. p. 551-95. 9. Meyers R. Dandy’s striatal theory of the ‘‘center of consciousness’’; Surgical evidence and logical analysis indicating its improbability. Arch Neurol Psychiat 1951;65:659-71. 10. Walker AE. Cerebral pedunculotomy for the relief of involuntary movements. I. Hemiballismus. Acta Psychiatr Neurol Scand 1949;24:712-29. 11. Walker AE. Cerebral pedunculotomy for the relief of involuntary movements. Parkinsonian tremor. J Nerv Ment Dis 1952;116:766-75. 12. Cooper IS. Ligation of the anterior choroidal artery for involuntary movements of parkinsonism. Psychiat Quart 1953;27:317-9. 13. Meyers R. Surgical procedure for postencephalitic tremor, with notes on the physiology of premotor fibers. Arch Neurol Psychiat 1940;44:455-9. 14. Meyers R. The modification of alternating tremors, rigidity and festination by surgery of the basal ganglia. Res Publ Ass Res Nerv Ment Dis 1942;21:602-65. 15. Meyers R. Historical background and personal experiences in the surgical relief of hyperkinesia and hypertonus. In: Fields W, editor. Pathogenesis and treatment of parkinsonism. Springfield, IL: Charles C Thomas; 1958. p. 229-70. 16. Gildenberg PL. Neurosurgical treatment of movement disorders: history and update. In: Germano I, editor. Neurosurgical treatment of movement disorders. New York: Thieme; 2002. p. 139-47. 17. Cooper IS. Results of 1000 consecutive basal ganglia operations for parkinsonism. Ann Intern Med 1960;52:483-99. 18. Hassler R, Riechert T. Indikationen und Lokalisationsmethode der gezielten Hirnoperationen. Nervenarzt 1954;25:441-7. 19. Cooper IS. Chemopallidectomy. Science 1955;121:217. 20. Cooper IS, Bravo G, Riklan M, Davidson N, Gorek E. Chemopallidectomy and chemothalamectomy for parkinsonism. Geriatrics 1958;13:127-47. 21. Gildenberg PL. Studies in stereoencephalotomy. VIII. Comparison of the variability of subcortical lesions produced by various procedures (radio-frequency coagulation, electrolysis, alcohol injection). Confin Neurol 1957;17:299-309. 22. Gildenberg PL. Study of methods of producing subcortical lesions and of evaluating their effect upon the tremor of parkinsonism. MS Thesis (Experimental Neurology), Temple University, Philadelphia, PA; 1959. 23. Cooper IS, Lee A St.J. Cryostatic congelation. J Nerv Ment Dis 1961;133:259-63.
51
52
3
History of stereotactic surgery in US
24. Gildenberg PL. Evolution of neuromodulation. Stereotact Funct Neurosurg 2005;83:71-9. 25. Melzack R, Wall PD. Pain mechanisms: a new theory. Science 1965;150:971-9. 26. Shealy CN, Mortimer JT, Reswick JB. Electrical inhibition of pain by stimulation of the dorsal columns. Preliminary clinical report. Anesth Analg (Cleve) 1967;46:489-91. 27. Hosobuchi Y, Adams JE, Rutkins B. Chronic thalamic stimulation for the control of facial anesthesia dolorosas. Arch Neurol 1973;29:158-61. 28. Richardson DE, Akil H. Pain reduction by electrical brain stimulation in man. Part 1: acute administration in periaqueductal and periventricular sites. J Neurosurg 1977;47:178-83. 29. Richardson DE, Akil H. Pain reduction by electrical brain stimulation in man. Part 2: chronic self-administration in the periventricular gray matter. J Neurosurg 1977;47:184-94. 30. Benabid AL, Pollak P, Hommel M, Gaio JM, de Rougemont J, Perret J. Treatment of Parkinson tremor by chronic stimulation of the ventral intermediate nucleus of the thalamus. Rev Neurol (Paris) 1989;145:320-3. 31. Siegfried J. Effect of stimulation of the sensory nucleus of the thalamus on dyskinesia and spasticity. Rev Neurol (Paris) 1986;142:380-3. 32. Leksell L. The stereotaxic method and radiosurgery of the brain. Acta chir scand 1951;102:316-9. 33. Leksell L. Stereotaxis and radiosurgery: an operative system. Springfield, IL: Charles C Thomas; 1971. 34. Lunsford LD, Maitz A, Lindner G. First United States 201 source cobalt-60 gamma unit for radiosurgery. Appl Neurophysiol 1987;50:253-6. 35. Betti O, Derechinsky V. Hyperselective encephalic irradiation with a linear accelerator. Acta Neurochir Suppl 1984;33:385-90. 36. Avanzo RC, Chierego G, Marchetti C, Pozza F, Colombo F, Benedetti A, Zanardo A. Stereotaxic irradiation with a linear accelerator. Radiol Med (Torino) 1984; 70:124-9. 37. Friedman WA, Bova F. The University of Florida radiosurgery system. Surg Neurol 1989;32:334-42.
38. Loeffler JS, Alexander E, III, Siddon RL, Saunders WM, Coleman CN, Winston KR. Stereotactic radiosurgery for intracranial arteriovenous malformations using a standard linear accelerator. Int J Radiat Oncol Biol Phys 1989;17:673-7. 39. Kjellberg RN, Koehler AM, Preston WM, Sweet WH. Stereotaxic instrument for use with the Bragg peak of a proton beam. Conf Neurol 1962;22:183-9. 40. White JC, Sweet WH. Pain, its mechanism and neurosurgical control. Springfield, IL: Charles C Thomas; 1955. 41. White JC, Sweet WH. Pain and the neurosurgeon. A forty year experience. Springfield, IL: Charles C Thomas; 1969. 42. Gybels JM, Sweet WH. Neurosurgical treatment of persistent pain. Physiological and pathological mechanisms of human pain. Pain Headache 1989;11:1-402. 43. Mullan S, Harper PV, Hekmatpanah J, Torres H, and Dobben G. Percutaneous interruption of spinal pain tracts by means of a strontium-90 needle. J Neurosurg 1963;20:931-9. 44. Rosomoff HL, Carrol F, Brown J, Sheptak P. Percutaneous radiofrequency cervical cordotomy: technique. J Neurosurg 1965;23:639-44. 45. Gildenberg PL, Lin PM, Polakoff PP II, Flitter MA. Treating intractable pain with percutaneous cervical cordotomy. GP 1968;37:96-7. 46. Lin PM, Gildenberg PL, Polakoff PP. An anterior approach to percutaneous lower cervical cordotomy. J Neurosurg 1966;25:553-60. 47. Gildenberg PL, DeVaul RA. The chronic pain patient. Evaluation and management. Basel: Karger; 1985. 48. Rosomoff HL, Green C, Silbret M, Steele R. Pain and low back rehabilitation program at the University of Miami School of Medicine. NIDA Res Monogr 1981;36:92-111. 49. Bonica JJ. Organization and function of a pain clinic. In: Bonica JJ, editor. International symposium on pain. New York: Raven Press; 1974. 50. Bonica JJ, Loeser J. Management of pain. Philadelphia, PA: Lea & Febiger; 1990. 51. Loeser JD. Disability, pain, and suffering. Clin Neurosurg 1989;35:398-408.
2 History of the Stereotactic Societies P. L. Gildenberg . J. K. Krauss
The field of human stereotactic neurosurgery was born in 1947 when Spiegel and Wycis (> Figure 2-1) published their groundbreaking manuscript in Science [1]. In this article they describe their device, which was actually a animal apparatus suspended on the patient’s head with a plaster cap. The original Horsley-Clarke animal stereotaxic (which was the original spelling) apparatus that had been used in the laboratory was aligned with the skull by means of earplugs and orbital tabs, and the target was defined by its relationship to those bony landmarks [2]. However, there was too much variability between the boney landmarks and the human brain to use this system clinically. It was only possible when intraoperative encephalographic x-rays became available in the 1940s, that it became possible to align the human apparatus with internal cerebral landmarks. Because their human system was based on encephalographic landmarks, Spiegel and Wycis termed their procedure ‘‘stereoencephalotomy.’’ During the next decade, neurosurgeons who were interested in learning this new technique visited Spiegel and Wycis in Philadelphia, and later at each other’s institutions. Since there were no commercially available stereotactic apparatus, they first had to design and build their own device when they returned home before they could use this new procedure. The pioneer stereotactic neurosurgeons exchanged information informally and personally. They met in small groups at irregular times, usually at each other’s institutions.
#
Springer-Verlag Berlin/Heidelberg 2009
The International Society for Research in Stereoencephalotomy Periodically, the growing number of stereotactic neurosurgeons met more formally. Many of the early meetings were hosted by Ernest A. Spiegel and Henry T. Wycis at Temple Medical School in Philadelphia. In 1961, at one of their meetings, they attempted to provide a more formal venue for the exchange of stereotactic information and experience and founded the first stereotactic society, which was called the International Society for Research in Stereoencephalotomy, a term that never really caught on. When only the American members met, the meeting was designated as the American Branch of the International Society for Research in Stereoencephalotomy, and when European members met as the European Branch. They met as groups at approximately two year intervals (reference). Consequently, when the International Society for Research in Stereoencephalotomy met in 1963 in Philadelphia at what was designated the First International Symposium on Stereoencephalotomy, the society was formally chartered (> Table 2-1). The Second International Symposium on Stereoencephalotomy was held in 1965, partly in Copenhagen and partly in Vienna, the latter coincident with the Meeting of the World Federation of Neurosurgical Societies (WFNS) (> Table 2-2). The Third Symposium on Stereoencephalotomy took place in 1967 in Madrid, hosted by Sixto Obrador.
36
2
History of the stereotactic societies
. Figure 2‐1 Spiegel and Wycis
. Table 2-2 Meetings of the WSSFN 1961 1965
1967 1968
1969
1970
1973
1977 . Table 2‐1 History of the Stereotactic Societies 1961 1963 1968
1970
1973
1980
International Society for Research in Stereoencephalotomy founded, Philadelphia International Society for Research in Stereoencephalotomy chartered, Philadelphia American Branch of the International Society for Research in Stereoencephalotomy founded, Atlantic City, NJ European Society for Stereotactic and Functional Neurosurgery founded, Freiburg, Germany World Society for Stereotactic and Functional Neurosurgery founded, Tokyo, Japan American Society for Stereotactic and Functional Neurosurgery founded as society affiliated with the World Society for Stereotactic and Functional Neurosurgery First Meeting of the American Society for Stereotactic and Functional Neurosurgery, Houston, TX
1981
1985
1989
1993
1997
2001
2005
2009
The first formal independent meeting of the American Branch of the International Society for Research in Stereoencephalotomy met in Atlantic City, New Jersey, in 1968. It focused mainly on Parkinson’s disease, and was the last meeting to do so, since the field became
First International Symposium on Stereoencephalotomy, Philadelphia, PA Second International Symposium on Stereoencephalotomy, Copenhagen and Vienna Third International Symposium on Stereoencephalotomy, Madrid, Spain First Meeting of the American Branch of the International Society for Research in Stereoencephalotomy, Atlantic City, NJ Fourth Symposium of the International Society for Research in Stereoencephalotomy, New York, NY Fifth Symposium of the International Society for Research in Stereoencephalotomy, Freiburg, Germany Sixth Symposium of the International Society for Research in Stereoencephalotomy Meeting of the World Society for Stereotactic and Functional Neurosurgery, Tokyo, Japan Seventh Meeting of the World Society for Stereotactic and Functional Neurosurgery, Sao˜ Paulo, Brazil Eighth Meeting of the World Society for Stereotactic and Functional Neurosurgery, Zurich, Switzerland Ninth Meeting of the World Society for Stereotactic and Functional Neurosurgery, Toronto, Canada Tenth Meeting of the World Society for Stereotactic and Functional Neurosurgery, Maebashi, Japan Eleventh Meeting of the World Society for Stereotactic and Functional Neurosurgery, Ixtapa, Mexico Twelfth Meeting of the World Society for Stereotactic and Functional Neurosurgery, Lyon, France Thirteenth Meeting of the World Society for Stereotactic and Functional Neurosurgery, Adelaide, Australia Fourteenth Meeting of the World Society for Stereotactic and Functional Neurosurgery, Rome, Italy Fifteenth Meeting of the World Society for Stereotactic and Functional Neurosurgery, Toronto, Canada
quiescent after L-dopa was introduced in 1968 [3]. Although the meetings were relatively small, advances in pain management, epilepsy, and other movement disorders were exchanged.
History of the stereotactic societies
The core of dedicated stereotactic neurosurgeons remained interested and active in the field, which led to the Fourth International Symposium on Stereoencephalotomy in 1969 in New York, along with the meeting of the World Federation of Neurosurgical Societies (WFNS). In 1970 the Fifth International Symposium on Stereoencephalotomy was held in Freiburg, Germany, hosted by Traugott Riechert, which is also recognized as the founding and the first meeting of the ESSFN. At that time [4], both the American and European branches were considered components of the International Society.
. Figure 2-2 Logo of both the WSSFN and the ASSFN
The World Society for Stereotactic and Functional Neurosurgery It was well into the early 1970s before activity began to increase once again, which led also to organizational changes. At the Sixth Symposium of the International Society for Research in Stereoencephalotomy in Tokyo in 1973, hosted by Hirotaro Narabayashi, the name of the international organization was changed to the World Society for Stereotactic and Functional Neurosurgery (WSSFN) (> Figure 2‐2). The American branch became the American Society for Stereotactic and Functional Neurosurgery (ASSFN), and the European branch became the European Society for Stereotactic and Functional Neurosurgery (ESSFN) (> Figure 2‐3). This caused an interesting dilemma. Horsley and Clarke designated their new technique as ‘‘stereotaxic,’’ as did many of the neurosurgeons involved in this new field. However, some neurosurgeons, particularly in Europe, spelled the term ‘‘stereotactic’’ [4]. There needed to be consensus on how to spell the names of these new societies.
. Figure 2-3 Logo of the ESSFN
2
37
38
2
History of the stereotactic societies
A vote was taken, and ‘‘stereotactic’’ won, even though it was a mongrel term with ‘‘stereo,’’ meaning three dimensional, derived from Greek and ‘‘tactic,’’ meaning to touch, derived from Latin. The original term ‘‘stereotaxic’’ had been derived from two Greek words, ‘‘stereo’’ and ‘‘taxic,’’ meaning an arrangement. It was felt to reflect what stereotactic surgery did to use the term ‘‘to touch’’ rather than just to observe. The first meeting of the WSSFN was actually the reorganizational meeting held in Tokyo in 1973 and hosted by Hirotoro Narabayashi. At that meeting it was decided that the WSSFN would meet each time just prior to the meeting of the World Federation of Neurological Societies (WFNS) that met every 4 years. Since the WFNS alternated the location of the meetings from one continent to another, it became customary for the WSSFN to meet somewhere near the WFNS so members might attend both meetings with a minimum of travel, but yet in a separate venue. The WSSFN met every 4 years, with the next meeting held in 1977 in Sao˜ Paulo, Brazil, hosted by Raul Marino [5]. However, advances in the field occurred ever more rapidly, so each of the continental societies met during the intervening years. Each four year cycle began with the WSSFN meeting, then the ESSFN the following year, the ASSFN the next year, and the ESSFN the next year again. With that schedule, the ESSFN Meetings were held at 2 year intervals the first and third year of the cycle, whereas the WSSFN and ASSFN each met once every 4 years. Because there was more new information than could be provided to the members every 4 years, during the past 4 years the ASSFN also began to meet at 2 year intervals coincident with the years the ESSFN meets. The atmosphere at each of the meetings has been fraternal and informal, and the gathering of international stereotacticians maintained a feeling of collegiality. A perusal of the topics at each of the meetings of the WSSFN and the ASSFN has
served as a good indicator of what basic and clinical advances were made and what procedures and technology in stereotactic and functional neurosurgery were developed. The Proceedings of each of the meetings was published as complete issues of volumes of the official journal, Stereotactic and Functional Neurosurgery from 1968 through 2001. When the Eighth Meeting of the WSSFN in Zurich in 1981 was hosted by Jean Siegfried [6], the ESSFN declared that it was a joint meeting between the ESSFN and the WSSFN, and the ESSFN was not a component society of the WSSFN, but rather an independent society. The Ninth Meeting of the WSSFN was held in Toronto in 1985, with Ronald Tasker [7] as host. The Tenth Meeting of the WSSFN was held in Maebashi, Japan, in 1989 with Chihiro Ohye [8] as host. The Eleventh Meeting of the WSSFN was held in Ixtapa, Mexico, in 1993, with Philip L. Gildenberg [9] as the host (> Figure 2‐4). The 1997 the Twelfth Meeting of the WSSFN was held in Lyon, France, with Marc Sindou [10] as the host. It was by far the largest meeting and included many European neurologists and many local neurosurgeons. Deep brain stimulation for movement disorders was advancing rapidly, as was reflected in the program. The Thirteenth Meeting of the WSSFN was held in Adelaide, Australia, in 2001, hosted by Brian Brophy, and became a memorable meeting for an unrelated event. It was during that meeting that the terrorist attack on the World Trade Center occurred in New York. In addition to concerns about being far from family, all flights to the United States were canceled, both stranding attendees from returning to the US and preventing US neurosurgeons from attending the meeting of the World Federation of Neurosurgical Societies in Sydney immediately following the WSSFN Meeting. Many of the WSSFN speakers found themselves filling in for
History of the stereotactic societies
2
. Figure 2‐4 Presidents of the WSSFN prior to 1993, when this picture was taken. Gildenberg, Nashold, Tasker, Narabayashi, Siegfried, Ohye, and Gybels
neurosurgeons who were unable to travel to speak at the WFNS in Sydney. The most recent WSSFN Meeting was held in Rome in 2005 and hosted by Mario Meglio. The 2009 meeting will be hosted by the current President, Andres Lozano, in Toronto. Additional updated information may be obtained on the WSSFN web site at www.wssfn.org. At each international symposium from 1977 to 1985, (see > Table 2‐3) the WSSFN has presented an award to an outstanding stereotactic neurosurgeon. At the presentation of the first award, it was named the Spiegel-Wycis Award, and a gold medal was made by Prof. Manuel Velasco-Suarez to be presented to Lars Leksell (> Table 2‐4). In 1981 the award was presented to Traugott Riechert. It was felt that there were too many potential recipients for only one award to be given every 4 years, so in 1985 a gold award was presented to Jean Talairach and a silver award to Manuel Velasco Suarez. Since there was little distinction between the
requirements for the gold and silver awards, two identical gold awards have been presented since 1989. In July, 2006, the WSSFN sponsored and organized for the first time an interim meeting in Shanghai, China, which was hosted by Bomin Sun from the Shanghai Jiao Tong University, Rui Jin Hospital. The meeting covered surgical treatment for movement disorders, pain, psychiatric disease, and epilepsy. Andres M. Lozano was the meeting chairman, and Drs. Benabid, Kaplit, Krauss, Schulder, Taira, Delong, Chang, Aziz, Broggi, Sun and Zhang were among those who gave the 36 plenary lectures. It was the first time that a WSSFN meeting was held on mainland Asia. The meeting coincided with a satellite symposium on functional neurosurgery in China. Overall 250 attendees from mainland China, Taiwan and Hong Kong joined WSSFN members from 15 other countries throughout the world. The event was considered a tremendous success by those who attended.
39
40
2
History of the stereotactic societies
The American Society for Stereotactic and Functional Neurosurgery
. Table 2‐3 Meetings of the ASSFN 1980
1983
1987
1991
1995
1999
2003
2006
2008
Meeting of the American Society for Stereotactic and Functional Neurosurgery, Houston, Texas Meeting of the American Society for Stereotactic and Functional Neurosurgery, Durham, North Carolina Meeting of the American Society for Stereotactic and Functional Neurosurgery Montreal, Canada Meeting of the American Society for Stereotactic and Functional Neurosurgery, Pittsburgh, PA Meeting of the American Society for Stereotactic and Functional Neurosurgery, Los Angeles, CA Meeting of the American Society for Stereotactic and Functional Neurosurgery, Snowbird, Utah Meeting of the American Society for Stereotactic and Functional Neurosurgery, New York, NY Meeting of the American Society for Stereotactic and Functional Neurosurgery, Boston, MA Meeting of the American Society for Stereotactic and Functional Neurosurgery, Vancouver, Canada
The first separate meeting of the ASSFN was held in Houston in 1980 and hosted by Philip L. Gildenberg [11]. The 27 papers included basic neurophysiology, movement disorders, epilepsy and pain. The meeting attracted an international audience, which has been the hallmark of all of both the continental and the world stereotactic society meetings. It is interesting that even at that early date, 11 papers concerned the use of the newly developing use of computers in neurosurgical guidance, both for addressing functional targets and in the new field of image-guided neurosurgery. The next independent ASSFN meeting took place in Durham, North Carolina, in 1983, with Blaine Nashold [12] as the host, as the field showed signs of increased activity and reawakening. New technology was of greatest interest, and half of the papers concerned image guidance as it impacted brain tumor management.
. Table 2‐4 Past Recipients of the Spiegel-Wycis Award Recipient
Location
Year
Lars Leksell Traugott Riechert Jean Talairach Manuel Velasco Suarez Hirataro Narabayashi Denise Albe-Fessard Ronald R. Tasker Blaine S. Nashold, Jr. Bjorn Meyerson Philip L Gildenberg Lauri V. Laitinen Chihiro Ohye Patrick Kelly Alim-Louis Benabid
Sao˜ Paolo Zurich, Switzerland Toronto, Canada Toronto, Canada Maebayashi, Japan Maebayashi, Japan Ixtapa, Mexico Ixtapa, Mexico Lyon, France Lyon, France Adelaide, Australia Adelaide, Australia Rome, Italy Rome, Italy
July, 1977 July, 1981 July, 1985 July, 1985 October, 1989 October, 1989 October, 1993 October, 1993 July, 1997 July, 1997 September, 2001 September, 2001 June, 2005 June, 2005
History of the stereotactic societies
The 1987 meeting in Montreal, which was hosted by Andre´ Olivier, was the largest ASSFN meeting up to that time. Out of the 103 papers, 14 concerned the rapidly developing field of stereotactic radiosurgery and 41 concerned the use of computers in neurosurgery. Functional neurosurgery showed signs of significant awakening, as represented by 36 presentations. The 1991 meeting in Pittsburgh was hostedby L. Dade Lunsford [13], and represented another milestone. The stereotactic surgery meeting was followed by a two day meeting on stereotactic radiosurgery, which attracted nonneurosurgical colleagues in radiation physics and radiotherapy, as well as those stereotactic neurosurgeons who were using the Gamma Knife, the only stereotactic radiosurgical system available at that time. It was at that satellite meeting that the International Society for Stereotactic Radiosurgery was formed. In addition, this large meeting involved such new fields as tissue transplantation. It was also at that meeting that Lauri Laitinen [14,15] first presented his observations on ventroposterior pallidotomy, which signaled the return of Parkinson’s disease surgery to the functional neurosurgery arena. The ASSFN meeting was held in Los Angeles and hosted by Michael Apuzzo and David Roberts in 1995 [16,17]. Half of the papers were about classical functional neurosurgical topics of movement disorders, epilepsy and pain. There were eight papers about stereotactic radiosurgery and the rest about image guided surgery and brain tumor management. The 1999 ASSFN meeting was held in Snowbird, Utah, and hosted by Peter Heilbrun and Douglas Kondziolka. Two-thirds of the 55 papers were in functional neurosurgery, with emphasis on movement disorders, epilepsy, and pain. The rest were divided among imaging and computer guidance, stereotactic radiosurgery, and brain tumors. In 2003, the ASSFN met in New York, and was hosted by Patrick Kelly. The program
2
demonstrated the increasing use of computers in stereotactic and image guided surgery. Prior to that meeting, the FDA had provided approval of chronic deep brain stimulation, and much of the data related to that was presented. The meeting in 2006 was held in Boston and hosted by G. Rees Cosgrove. There was a significant shift toward interest in functional neurosurgery, involving both DBS and extending to epilepsy surgery. The current administrative structure includes officers elected for a term of 4 years, a board of directors and five Continental Vice Presidents. During the period from 2005 – 2009 the ESSFN leadership is represented by Andres M. Lozano (Toronto, Canada) as President, Joachim K. Krauss (Hannover, Germany) as Vice-President, Takaomi Taira (Tokyo, Japan) as Secretary-Treasurer, Michael Schulder (Newark, USA) as Assistant Secretary-Treasurer, and Philip L. Gildenberg (Houston, USA) as Historian. Honorary members include Blaine Nashold, Jean Siegfried, Ronald Tasker, Chihiro Ohye and Philip L. Gildenberg. Prior to 1975, there was no formal arrangement between the ASSFN and the American Association of Neurological Surgeons (AANS), although it was customary to have a representative of stereotactic surgery on the AANS program committee. In 1975, due to the efforts of then ASSFN President John Van Buren, the ASSFN amalgamated with the AANS and became the Section for Stereotactic and Functional Neurosurgery. This arrangement put the responsibility for the program of one and later two afternoon sessions in the hands of the ASSFN officers. In addition, members of the ASSFN automatically became members of the AANS Section of Stereotactic and Functional Neurosurgery. The ASSFN treasury, however, remained independent of the AANS, which provided the opportunity for the ASSFN to hold independent meetings every 4 years. The business meeting of the ASSFN is held at the conclusion of one of the days when the
41
42
2
History of the stereotactic societies
stereotactic scientific program is held. Since 2004, the annual residents’ research award, which has been named the Gildenberg Award, has been presented at one of the Stereotactic Sections programs. In 1984, a similar administrative arrangement was made between the ASSFN and the Congress of Neurological Surgeons (CNS). The amalgamation was facilitated by the presence of the ASSFN President, George Sypert, on the CNS Board of Directors at that time. Current details of the ASSFN can be found on the web site at www.assfn.org.
The European Society for Stereotactic and Functional Neurosurgery The formation of the European Society for Stereotactic and Functional Neurosurgery (ESSFN) dates back as far as 1970 when the International Society for Research in Stereoencephalotomy met in Freiburg, Germany. The founding fathers at that time were Fritz Mundinger, Traugott Riechert and John Gillingham, and this meeting now is also considered as the founding meeting of the ESSFN. The society was registered as a taxexempted society in Freiburg, Germany, in 1971. It was founded primarily in order to represent the interests of European functional neurosurgeons and also to enhance communication and exchange between European countries. At that time, the European idea was blossoming, but it took still several years until traveling from one European country to the other was as comfortable as it is nowadays. The judicial seat of the society was moved to Toulouse, France, in 2002. Initially, the ESSFN had a relatively isolated position in the general neurosurgical community. With the rapid development of new imaging techniques and computer technology, however, stereotactic concepts and methods had a profound impact on the progress of European neurosurgery, in general. The ESSFN over the
decades thus served both the interests of those who were subspecialized in functional and stereotactic neurosurgery and those whose activities were embedded in general neurosurgery. The ESSFN has stated its principal objectives in a mission which is displayed on the ESSFN web site (www.essfn.org). Also, its Constitution and By-Laws are displayed on the web site. The logo of the ESSFN became popular during the 1990s (> Figure 2‐3). The board of officers is elected for a term of 4 years by the General Assembly which comes together during the congresses. It is supported by an Executive Committee which has representatives from 20 different European countries. During the period from 2006 – 2010 the ESSFN leadership is represented by Yves Lazorthes (Toulouse, France) as President, Giovanni Broggi (Milano, Italy) as Vice-President, Joachim K. Krauss (Hannover, Germany) as Secretary, Damianos Sakas (Athens, Greece) as Second Secretary, and Bart Nuttin (Leuven, Belgium) as Treasurer. The honorary presidents are the former ESSFN presidents F. John Gillingham, Fritz Mundinger, Gian Franco Rossi, Bjo¨rn Meyerson, Christoph B. Ostertag, David G. T. Thomas, and Andries Bosch. There are five different categories of membership: active, associate, resident, honorary and benefactor. Membership has increased steadily over the years, and in early 2008 there were more than 230 members. The majority of members come from France, Germany, Italy, Spain, The Netherlands, and the UK. There are members, however, from almost all European countries with increasing numbers in particular from Eastern European countries and Russia. Overall, the society has about 30 non-European members coming mainly from South Korea, Japan, Mexico and the United States. Since its inception in 1970, the ESSFN organized congresses at regular intervals in various locations all over Europe (> Table 2‐5). While the first meeting was still under the umbrella of the International Society for Research in
History of the stereotactic societies
2
. Table 2‐5 Congresses of the ESSFN 1970 1972 1975 1977 1979 1981 1983 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008
Founding Meeting of the European Society for Stereotactic and Functional Neurosurgery, Freiburg, Germany First Congress of the European Society for Stereotactic and Functional Neurosurgery, Edinburgh, United Kingdom Second Congress of the European Society for Stereotactic and Functional Neurosurgery, Madrid, Spain Third Congress of the European Society for Stereotactic and Functional Neurosurgery, Freiburg, Germany Fourth Congress of the European Society for Stereotactic and Functional Neurosurgery, Paris, France Fifth Congress of the European Society for Stereotactic and Functional Neurosurgery, Zurich, Switzerland Sixth Congress of the European Society for Stereotactic and Functional Neurosurgery, Rome, Italy Seventh Congress of the European Society for Stereotactic and Functional Neurosurgery, Birmingham, United Kingdom Eighth Congress of the European Society for Stereotactic and Functional Neurosurgery, Budapest, Hungary Ninth Congress of the European Society for Stereotactic and Functional Neurosurgery, Marbella, Spain Tenth Congress of the European Society for Stereotactic and Functional Neurosurgery, Stockholm, Sweden Eleventh Congress of the European Society for Stereotactic and Functional Neurosurgery, Antalya, Turkey Twelfth Congress of the European Society for Stereotactic and Functional Neurosurgery, Milano, Italy Thirteenth Congress of the European Society for Stereotactic and Functional Neurosurgery, Freiburg, Germany Fourteenth Congress of the European Society for Stereotactic and Functional Neurosurgery, London, United Kingdom Fifteenth Congress of the European Society for Stereotactic and Functional Neurosurgery, Toulouse, France Sixteenth Congress of the European Society for Stereotactic and Functional Neurosurgery, Vienna, Austria Seventeenth Congress of the European Society for Stereotactic and Functional Neurosurgery, Monterux, Switzerland Eighteenth Congress of the European Society for Stereotactic and Functional Neurosurgery, Rimini, Italy
Stereoencephalotomy, the subsequent congresses were proper ESSFN congresses. The 1981 congress in Zu¨rich was considered a joint meeting of both the WSSFN and the ESSFN. While attendance was limited in the early years to those who practiced in a subspeciality frame, the scope of topics in recent years became much broader attracting also many neurosurgeons not devoted exclusively to functional neurosurgery. During the congress, awards are available for best oral presentations by young neurosurgeons and also prizes for best posters. The ESSFN also provides grants for research in stereotactic and functional neurosurgery considering both basic and clinical research. The grant recipients will report at the congresses about their results. In the years between the congresses the ESSFN administers hands-on courses which are open to ESSFN members or those who wish to apply for membership. The purpose of these courses is to provide education and training in
functional and stereotactic techniques. The first course was given in 2003 (Tolochenaz, Switzerland) on movement disorders surgery, followed by courses on pain surgery (Toulouse, France) in 2005, and on radiosurgery (Marseilles, France) in 2007. The topics for future courses will be adapted to include new horizons, but also to provide in-depth teaching about common standards. The web site has become an important medium to keep the membership posted with updated information. Application forms for membership can be downloaded from the web site. Furthermore, members are supplied with a Newsletter that is distributed about yearly detailing new developments and announcements for grants and courses. There are many close institutional and personal links between the ESSFN and the other societies for stereotactic and functional neurosurgery, in particular to the WSSFN. Nevertheless, there is no such a formal affiliation between the
43
44
2
History of the stereotactic societies
ESSFN and the WSSFN as it has been established between the ASSFN and the WSSFN. The ESSFN also communicates with the national functional neurosurgery societies which have been established in almost all European countries. In addition, a subcommittee of the ESSFN cooperates with the UEMS (Union Europeene des Medecines Specialistes) to establish more uniform guidelines and standards for stereotactic and functional neurosurgery for all European countries.
Other International Stereotactic Surgical Societies In addition to the time-honored European and American functional neurosurgery societies, there are other international societies that organize meetings to enhance regional information exchange. The Sociedad Latinoamericana de Neurocirugia Funcional y Estereotaxia (SLANFE) is active since 1998 and covers the interests of functional neurosurgeons in Latin America. The Asian Society for Stereotactic, Functional and Computer Assisted Neurosurgery (ASSFCN) convenes meetings in Asia. In addition, several countries maintain their own Society for Stereotactic and Functional Neurosurgery, in order to accommodate their junior neurosurgeons who may not speak English or have funds to travel to the international meetings, such as Japan, Korea, Argentina, and China.
References 1. Spiegel EA, Wycis HT, Marks M, et al. Stereotaxic apparatus for operations on the human brain. Science 1947;106:349-50. 2. Horsley V, Clarke RH. The structure and functions of the cerebellum examined by a new method. Brain 1908;31:45-124.
3. Cotzias GC, VanWoert MH, Schiffer LM. Aromatic amino acids and modification of parkinsonism. New Engl J Med 1967;276:374-9. 4. Gildenberg PL. ‘‘Stereotaxic’’ versus ‘‘stereotactic’’. Neurosurgery 1993;32:965-6. 5. Gildenberg PL, Marino RJ. Seventh symposium of the international society for research in stereoencephalotomy. Conf Neurol 1978;41:1-250. 6. Gildenberg PL, Siegfried J, Gybels J, et al. Eighth meeting of the world society and the fifth meeting of the european society for stereotactic and functional neurosurgery. Appl Neurophysiol 1982;45:1-554. 7. Tasker RR, Turnbull IM, Gildenberg PL, Franklin PO, editors. Proceedings of the ninth meeting of the world society for stereotactic and functional neurosurgery. Appl Neurophysiol 1985;48(1–6):1‐498. 8. Ohye C, Gildenberg PL, Franklin PO. Proceedings of the tenth meeting of the world society for stereotactic and functional neurosurgery. Stereotac Funct Neurosurg 1990;54–55:1-564. 9. Gildenberg PL, Franklin PO, Escobedo FR, et al. Proceedings of the eleventh meeting of the world society for stereotactic and functional neurosurgery. Stereotac Funct Neurosurg 1994;63:1-301. 10. Sindou M, Martens F, Gildenberg PL, et al. Proceedings of the twelfth meeting of the world society for stereotactic and functional neurosurgery. Stereotac Funct Neurosurg 1997;68:1-318. 11. Gildenberg PL. Proceedings of the meeting of the American society for stereotactic and functional neurosurgery. Appl Neurophysiol 1980;43:89-266. 12. Nashold BS Jr, Gildenberg PL, Franklin PO. Proceedings of the American society for stereotactic and functional neurosurgery. Appl Neurophysiol 1983;46:1-252. 13. Lunsford LD, Gildenberg PL, Franklin PO. Proceedings of the meeting of the American society for stereotactic and functional neurosurgery, Part I. Stereotac Funct Neurosurg 1991;58:1-208. 14. Laitinen LV, Bergenheim AT, Hariz MI. Leksell’s posteroventral pallidotomy in the treatment of Parkinson’s disease. J Neurosurg 1992;76:53-61. 15. Laitinen LV, Bergenheim AT, Hariz MI. Ventroposterolateral pallidotomy can abolish all parkinsonian symptoms. Stereotact Funct Neurosurg 1992;58:14-21. 16. Roberts DW, Apuzzo MLJ, Gildenberg PL, et al. Proceedings of the meeting of the American society for stereotactic and functional neurosurgery, Part I. Stereotac Funct Neurosurg 1995;65:1-207. 17. Roberts DW, Apuzzo MLJ, Gildenberg PL, et al. Proceedings of the meeting of the American society for stereotactic and functional neurosurgery, Part II. Stereotac Funct Neurosurg 1996;66:1-156.
21 Angiography, MRA in Image Guided Neurosurgery P. Jabbour . S. Tjoumakaris . R. Rosenwasser
Introduction
Indication
Image-guided neurosurgery has become a standard practice in the last decade. This technology can assist the surgeon in different steps, including planning and executing the surgical procedure. So far advances in image guided surgery had little influence on vascular neurosurgery because of technical difficulties transferring 3-D vessels imaging to the neuronavigation system. The introduction of the 3-D rotational angiography has revolutionized the way we visualize and treat brain aneurysms. Sometimes it can be challenging to try to compare these views with the intraoperative microscopic view that the surgeon has. Incorporating the images of 3-D angiography in the image guided system during real time surgery is a recent technological innovation that enables the surgeon to navigate using 3-D angiogram pictures. Another innovation is fusing MRA or CTA images to the neuronavigation technique. These new techniques can be used for aneurysm, AVM and tumor surgery. One of the concerns for the neurosurgeon during tumor surgery is to avoid injury of blood vessels surrounding the tumor or encased by the tumor. Tumor surgery may be performed more safely using the intraopeartive navigated 3-D ultrasound angiography. The classical technique involves using an intraoperative ultrasound to correct the shift during surgery to keep the neuronavigation accurate. The same equipment has also Doppler capability with power Doppler to visualize vessels.
Angiography, CTA or MRA coupled to the image guided techniques is a new addition to the armamentarium for the surgical treatment of vascular lesions and tumors in the proximity of big vessels. It is essential to define exactly the 3-D relationship of the neck of an aneurysm with the surrounding vascular branches and perforating vessels in aneurysm surgery, or the exact drainage and feeding pattern of a complex arteriovenous malformation, or the exact relationship of a tumor with the surrounding vessels and its pattern of vascular invasion. All this makes this new technique essential in the surgical treatment of complex brain aneurysms, as well as distal aneurysms like pericallosal aneurysms or distal middle cerebral artery (MCA) aneurysm because it is sometimes challenging to anatomically locate these aneurysms due to the lack of big parent vessels close by. It is also used in the treatment of complex AVMs and it helps in identifying feeding and draining vessels and provides a better understanding of the AVM architecture and facilitates its resection. And finally it is also a useful technique in the surgical resection of skull base tumors located close to big vessels.
#
Springer-Verlag Berlin/Heidelberg 2009
Techniques CTA and Neuronavigation A multislice thin cuts CT is performed with superficial fiducials on the skin with intravenous
300
21
Angiography, MRA in image guided neurosurgery
contrast injection. The data obtained is transferred directly or with a compact disc to the navigation station. A pre surgical planning can start at the navigation station on the 3-D model of the CT angiogram with special attention to the lesion and the surrounding vessels and structure. During surgery the 3-D vessel model is displayed on the navigation screen, and the screen display is matched with the intraoperative microscopic view of the surgeon to try to optimize the microdissection, for skull base pathology the bone can be substracted from the image displayed by the navigation system (> Figure 21-1). Chibbaro et al. [1] reported their experience using this technique, and they concluded that this technique can provide many advantages like a quick processing time of the raw data from the CT, a better understanding of the real position of the aneurysm with respect to the surgical position of the patient, it is also helpful in tailoring the craniotomy in case of small AVMs, sparing eloquent brain and fusing functional MRI with it, and they concluded that this technique avoids major problems, such as premature bleeding and or ischemic complications which often occur due to insufficient anatomical understanding of complex vascular lesions. Kim et al. [2], described their experience using the non formatted CTA coupled to neuronavigation in the clipping of 12 distal Anterior cerebral artery aneurysms (ACA) aneurysms, they concluded that this technique enables neurosurgeons to identify a distal ACA aneurysm and facilitates the safe exposure and clipping of the aneurysm. Rohde et al. [3,4] used the term ‘‘advanced neuronavigation.’’ They coupled the 3-D CTA images, MRA images and tractography to the neuronavigation system, they reported their series of 16 patients where they used this technique during surgery. Nine skull base meningiomas, one craniopharyngioma, one epidermoid, one giant carotid artery aneurysm, two basilar aneurysms, two brain stem cavernomas. This technique
facilitated the approach in four cases contributed to a tailored approach in two cases, helped to identify hidden vessels in nine cases. Rohde [4] later on described his series of 42 aneurysms in a prospective study. They used a 3-D reformatted CTA coupled to the neuronavigation system. Twelve aneurysms of the anterior communicating artery ACOM, 17 MCA aneurysms, six internal carotid artery aneurysms (ICA), three pericallosal artery aneurysms, two vertebro basilar aneurysms, and two superior cerebellar artery (SCA) aneurysms, all ruptured with subarachnoid hemorrhage. In 24 cases the neurosurgeon used this technique to localize the aneurysm, in 18 cases it was used to understand the branching anatomy, in 8 cases to visualize hidden structures, 5 to evaluate the projection of the dome, in 2 cases to tailor the approach, the author concluded that this technique has the potential to improve the operative results by reduction of the surgical trauma and avoidance of intraoperative complications. Coenen et al. [5], used the 3-D CTA coupled to neuronavigation for the surgical treatment of small AVMs with large hematomas in four patients, this technique allows feeding arteries to be distinguished from draining veins thereby allowing the nidus of the AVM to be identified despite the presence of intracerebral blood, and decreasing surgical morbidity.
MRA and Neuronavigation Coupled to Microvascular Doppler Sonography A multislice thin cuts MRI and MRA is performed with fiducials on the skin. The images are archived on a compact disc or transferred directly to the image guidance station at the operating room where the surgery is being performed. The 3-D ultrasound volumes are acquired within the same coordinate system as navigation is performed, when the craniotomy is performed a 3-D ultrasound scan is
Angiography, MRA in image guided neurosurgery
21
. Figure 21-1 Digital subtraction angiography (upper left) and CT angiography (CTA) show a giant aneurysm of the right internal carotid artery (ICA) in a 71-year-old female patient, who was admitted after subarachnoid hemorrhage (Hunt/Hess grade IV). A 3-D reconstruction of the arterial vessel anatomy was created, rotated according to the surgeon’s perspective and displayed on the screen of the neuronavigational system during surgery (upper right). Additionally, the 2-D CT angiographic images were screen-displayed (lower left). For the surgeon, neuronavigation with CTA was especially helpful to tailor the extradural anterior clinoid process resection. Aneurysm clipping was successfully performed (lower right) and the patient made a good recovery despite her poor Hunt/Hess grade (with permission from Rohde et al. Advanced neuronavigation in skull base tumors and vascular lesions. Minim Invasive Neurosurg 2005;48(1):13-8)
performed before the dura is opened, the surgeon will be able to view reconstructed 2-D cross sectional images. Another 3-D ultrasound scan is obtained at the end of the resection and a comparison of the preoperative and postoperative 3-D ultrasounds
images is performed, this is mainly used in AVM surgery (> Figure 21-2). Mathiesen et al. [6] reported their experience using MRA with 3-D ultrasound angiography coupled to neuronavigation in the surgical
301
302
21
Angiography, MRA in image guided neurosurgery
. Figure 21-2 Preoperative angiograms (a–d) and MRI scans (e and f), intraoperative ultrasound navigational 3-D (g, posterior view; h, frontolateral view) and postoperative angiograms (i–l) (with permission from Mathiesen et al. Neuronavigation for arteriovenous malformation surgery by intraoperative 3-D ultrasound angiography. Neurosurgery 2007;60(4 Suppl 2):345-50; discussion 341-50
treatment of AVMs. Nine patients with AVMs were treated, five had underwent pre operative embolization, six patients had pre operative MRIs, in six patients the stereoscopic display technique was used, it is a volume rendering technique that provides a true sense of 3-D vision when using special glasses the skin signal is subtracted and the surgeon is able to look at the MRA images of the AVM, allowing him to plan his approach and could navigate using the pointer and trying to identify feeders and draining vessels. The AVMs were totally removed in all nine patients. The ultrasound images corresponded to the intraoperative findings. This technique led to a
minimum exploration into the nidus, in two cases when the surgeon thought that he had a complete resection of the AVM, the 3-D ultrasound was able to demonstrate a residual nidus, allowing the surgeon to remove the residual AVM. The author concluded that AVM surgery was facilitated by navigation based on preoperative MRA coupled to intraoperative 3-D ultrasound and 3-D reconstruction. The AVMs were well outlined with better understanding of the feeders and draining vessels (> Figure 21-3). Unsgaard et al. [7], used also this technique in nine patients, seven patients had AVMs in eloquent brain with Spetzler-Martin grade II in
Angiography, MRA in image guided neurosurgery
21
. Figure 21-3 Intraoperative photograph (a), biplanar angiogram (b), and 3-D-RA view (c) demonstrating a ruptured right middle cerebral artery aneurysm during the early phase of microsurgical preparation. The 3-D-RA view helps to identify the origin of the branching vessel and the localization of a large daughter bleb that was the site of rupture and was at high risk for intraoperative rerupture. The 3-D-RA model can also be viewed from behind for better identification of the temporal M2 branch that was covered by other temporal branches arising from the M3 and M4 segments (with permission from Raabe et al. 3-D rotational angiography guidance for aneurysm surgery. J Neurosurg. 2006;105 (3):406-11)
four patients, III in four patients and IV in one patient. The 9 AVMs had 28 feeders seen on the preoperative angiograms. Twenty-five of those feeders were clipped at the beginning of surgery based on the stereoscopic information from both 3-D MRA and ultrasound angiography. The other feeders were clipped in a later phase of the surgery, in one patient intraoperative ultrasound angiography revealed residual nidus that was immediately removed. Four of the seven patients with AVMs in an eloquent area had a temporary worsening of their neurological status, one patient had a permanent neurological deficit. One patient with a grade IV AVM had to be reoperated for a post operative hematoma due to a residual nidus. The authors concluded that navigated stereoscopic display of angiography offers a technology that can be used successfully to identify and clip AVM feeders in the initial phase of the operation.
Rotational Angiography and Neuronavigation 3-D rotational angiography is increasingly used to diagnose aneurysms and to better evaluate the branching vessels close to the aneurysm neck. A rotational 3-D angiography is performed. 3-D data is reconstructed, the images are archived on a compact disc or transferred directly through a network to the image guidance station. Conventional registration techniques can not be used without any CT or MRI, because the 3-D data contains only the region of interest of the angiogram, and no data representing surface matching with the patient. The registration is done with a technique that correlates the 3-D angiogram with the head positioning of the patient and this by using the angulation and the rotational coordinates from the angiogram images. A special reference head frame is used just before
303
304
21
Angiography, MRA in image guided neurosurgery
the rotational angiogram is performed to record the position of the head with respect to the coordinates of the angiographic system. After reconstructing the rotational angiography images and optimizing the quality of the image, 3-D data is exported and loaded in the navigation system. The same three point headframe that was used during angiography is repositioned on the patient’s head. The position of the three point headframe in relation to the 3-D angiography data is already known from the fluoroscopic reference position image obtained before the rotational angiography procedure. Raabe et al. [8], used the previous technique on a series of 16 aneurysms, the procedure was successful in all cases. Using this technique added an average of 19 min to the procedure. There was a good correspondence with the intraopeartive vascular anatomy. It was determined by the surgeon, that this technique was helpful in eight cases in predicting the exact location of the aneurysm and the branching vessels covered by clots or brain parenchyma. The authors concluded that this technique provides useful information about the anatomical relationship between the aneurysm and the parent or branching vessel it helps minimizing the exposure and improves the quality of aneurysm surgery in selected cases. Willems et al. [9], used a different registration technique, not involving a headframe. Following the 3-D rotational angiogram acquisition, a navigated pointer is used to determine the position of the fiducials relative to the 3-D rotational angiography data set. The angiography room in this case is provided with navigation capabilities. The rotational angiography is coupled to an image guidance system by using a new software module, that enables the determination of the positional relation between the imaged volume and the tracker plate on the image intensifier, the positional relationship between the patient and the tracker plate is determined after the patient undergoes imaging, the data is transferred to
the operating room to the navigation system, the fiducials position are localized with a pointer and the patient to image registration is completed. They tried this technique on a phantom and on two cases and they concluded that there is a small error using this technique, but this does not prevent them from preserving the orientation of the vascular tree, and the goal of offering a 3-D rotational angiograms mimicking the surgical view was also achieved with technique.
Limitations One of the limitations is the failure to depict small arteries, another limitation would be the accuracy of the navigation system, and the brain shift after the bone is removed, the dura is open, and the cerebrospinal fluid CSF is drained. The registration process is still experimental especially in the image guided rotational 3-D angiography.
Conclusion Image guided vascular surgery using MRA, CTA or 3-D rotational angiography is a new technique developed to help surgeons identifying the topographical relationship between the vascular lesion and the related parent and branching vessels it can help to tailor the exposure and the dissection of the vascular tree, and therefore minimize the risk of intraoperative complications. It is still a new technique and it can benefit from a lot of improvement mainly in the registration step. Further studies and technical development are needed to try to make this new technique a standardized one.
References 1. Chibbaro S, Tacconi L. Image-guided microneurosurgical management of vascular lesions using navigated computed tomography angiography. An advanced IGS technology application. Int J Med Robot 2006;2(2):161-7.
Angiography, MRA in image guided neurosurgery
2. Kim TS, Joo SP, Lee JK, et al. Neuronavigation-assisted surgery for distal anterior cerebral artery aneurysm. Minim Invasive Neurosurg 2007;50(3):140-4. 3. Rohde V, Spangenberg P, Mayfrank L, Reinges M, Gilsbach JM, Coenen VA. Advanced neuronavigation in skull base tumors and vascular lesions. Minim Invasive Neurosurg 2005;48(1):13-8. 4. Rohde V, Hans FJ, Mayfrank L, Dammert S, Gilsbach JM, Coenen VA. How useful is the 3-dimensional, surgeon’s perspective-adjusted visualisation of the vessel anatomy during aneurysm surgery? A prospective clinical trial. Neurosurg Rev 2007;30(3):209-16; discussion, 207-16. 5. Coenen VA, Dammert S, Reinges MH, Mull M, Gilsbach JM, Rohde V. Image-guided microneurosurgical management of small cerebral arteriovenous malformations: the value of navigated computed tomographic angiography. Neuroradiology 2005;47(1):66-72.
21
6. Mathiesen T, Peredo I, Edner G, et al. Neuronavigation for arteriovenous malformation surgery by intraoperative three-dimensional ultrasound angiography. Neurosurgery. 2007;60(4): Suppl 2:345-50; discussion 341-50. 7. Unsgaard G, Ommedal S, Rygh OM, Lindseth F. Operation of arteriovenous malformations assisted by stereoscopic navigation-controlled display of preoperative magnetic resonance angiography and intraoperative ultrasound angiography. Neurosurgery 2005;56 Suppl 2:281-90; discussion 281-90. 8. Raabe A, Beck J, Rohde S, Berkefeld J, Seifert V. Threedimensional rotational angiography guidance for aneurysm surgery. J Neurosurg 2006;105(3):406-11. 9. Willems PW, van Walsum T, Woerdeman PA, et al. Image-guided vascular neurosurgery based on threedimensional rotational angiography. Technical note. J Neurosurg 2007;106(3):501-6.
305
19 CT/MRI Safety in Functional Neurosurgery M. Schulder . A. Oubre´
Introduction Unique and ever-changing applications of imaging technologies have played an integral role in transforming the landscape of stereotactic and functional neurosurgery [1]. Both computer tomography (CT) and magnetic resonance imagery (MRI) continue to push forward the boundaries of image-guided neurological surgery. Despite the great benefits offered by CT and MRI, however, each technique poses certain risks to the safety of patients and, in some cases, of healthcare workers [2–4]. This chapter addresses pivotal considerations of the safe use of CT and MRI. The primary risks of CT scanning are associated with ionizing radiation and reactions to iodinated contrast media (ICM) [2,3,5] While MRI is often considered safer than CT because of the absence of ionizing radiation, MRI has raised its own set of safety issues. The use of gadolinium-based MR contrast agents (GBMCAs) has been linked with various types of adverse reactions, especially contrast-induced nephropathy in patients in advanced stages of renal disease [2]. In addition, MRI must be used with caution in patients with implanted devices [1,2].
DNA and generate free radicals. This is a safety concern because CT scanners usually deliver radiation doses that are often 100 times greater than those of conventional radiographic examinations, including chest X-rays or mammograms [3]. Scanner-based CT radiation carries a small but serious risk of causing cancer. Ionizing radiation can injure biologic material through several mechanisms, including formation of hydroxyl radicals that damage or break DNA doublestrands bases Concern over the risk of ionizing radiation for health problems, including malignancy, has reached a critical level in the current medical climate [5].
CT Scanning and Radiation Parameters The radiation dose for a specific study is determined by several scan parameters. These include the number of scans, tube current and scanning time in milliampseconds (mAs), size of the patient, axial scan range, scan pitch (the amount of overlap between adjacent CT slices), tube voltage in the kilovolt peaks (kVp), and the design features of the scanner used to deliver the dose [5].
Computed Tomography Safety of Ionizing Radiation
Estimations of CT Radiation-Related Cancer Risk
Background Computed tomography uses ionizing radiation: high-energy photons that are known to damage #
Springer-Verlag Berlin/Heidelberg 2009
Estimates of radiation-induced cancer risk are based on epidemiologic follow-up studies of atomic-bomb survivors in Japan and other large
280
19
CT/MRI safety in functional neurosurgery
population investigations. One large-scale study examined 400,000 radiation workers in the nuclear industry receiving about 20 milliSieverts (mSv) (a representative organ dose from a single CT scan for an adult). These investigations yielded a significant association between radiation dose and cancer-related mortality in persons exposed to doses between 5 and 150 mSv [3]. The estimated lifetime risk of death from cancer resulting from a single CT scan of the head is determined by adding the estimated organ-specific cancer risks. Again, these risks are derived from estimations of organ-specific data for cancer incidence or mortality in atomic-bomb survivors. A CT study with 2 or 3 series produces a radiation dose in the range of 30–90 mSv. Findings from some investigations suggest that this dose range is associated with a statistically significant increase in the risk of cancer in adults. Approximately 0.4% of all cancers in the United States may be associated with CT scanner-based radiation. Current estimates of excess radiationrelated cancer rates range between 1.5 and 2.0%. The evidence for increased risk is most compelling for children. Compared to adults, they are more radiosensitive and have a longer remaining life span during which time a radiation-induced cancer could form. The lifetime cancer mortality risk associated with a single head CT protocol in a 1-year-old child was 0.07% [3]. The methodology for estimating the longrange cancer risk from CT radiation is in dispute over bias. Some investigators claim that the linear no-threshold model in this dose range may overestimate the risk. Excess cancer rates have not been reported in humans for doses below 100 mSv, One possible reason for this is that defense mechanisms that inhibit radiocarcinogenesis may be much more effective at low doses [6,7]. Yet, other evidence reveals that exposure to CT-related radiation exceeds low-level radiation doses, instead falling within the range of medium-level exposure. This is noteworthy
because increased cancer risk is related to midlevel radiation doses [5].
Strategies for Reducing Radiation Dose Strategies for radiation dose reduction include inplane bismuth shielding, minimizing multiphase scanning, and decreasing or eliminating multiple scans with contrast material. CT settings can be optimized by decreasing tube current (often via automatic tube currentmodulation(ATCM)), using a larger pitch, and limiting the range of coverage. The automatic exposure-control option on new scanners can be adjusted to decrease the radiation dose. However, there is almost always a tradeoff between lowering the level of radiation dose and producing the highest quality images. The cost of reducing radiation dose by, for example, decreasing gantry rotation time, is an increase in image noise [5,8]. Minimizing patient exposure to radiation remains a priority for healthcare workers in radiology. In some cases, magnetic resonance imaging (MRI) may be a preferred option to CT scans [9]. In the absence of updated CT protocols that reflect current scientific thought, neuroradiologists and neurosurgeons must collaborate to identify optimal techniques for radiation dose reduction during a CT diagnostic or interventional procedure. Since current radiation risk estimates remain ambiguous, CT scans should be performed in accordance with the ‘‘ALARA’’ principle: ‘‘As low as reasonably achievable’’ [5,9]. Nowhere is this more crucial than for the pediatric population. Guidelines established for CT imaging in children recommend adjusting scan parameters for smaller size in order to achieve lower-dose scanning for specific applications. Following CT guidelines protocols for using age-adjusted, relatively lower tube currents may help to reduce the radiation dose for pediatric CT of the brain [3,5].
CT/MRI safety in functional neurosurgery
Iodinated Contrast Media (ICM) Used in Enhanced CT Scans
19
Non-ionic ICM are the preferred agents in CT enhanced scans of the head [2].
Background Indications for the use of iodinated contrast media (ICM) in stereotactic neurosurgery mainly include targeting of mass lesions for stereotactic biopsy and radiosurgery [10]. While in much of the developed world contrast– enhanced MRI has supplanted CT for these uses, CT contrast may still be needed [11]. Patients with pacemakers in general cannot safely undergo MRI, and others who are obese or claustrophobic may not tolerate the small bore and prolonged acquisition times [4]. MRI may not always be available, and in much of the world CT is by far the more common imaging modality. It therefore behooves neurosurgeons to understand potential problems associated with the use of ICM.
Types of Radiographic Iodinated Contrast Media (ICM) Contrast materials consist of ionic (high osmolar), and organic non-ionic (low osmolar) water soluble agents. The higher osmolarity (600–2,100 mOsm/kg) in solution for ionic contrast agents accounts for some of their adverse effects. By contrast, nonionic agents have approximately half the osmolality of ionic substances, making them less likely to affect the blood-brain barrier. These materials exhibit fewer side effects because they do not ionize in solution. Yet, nonionic agents possess the same degree of radiopacity as ionic contrast materials. Both high and low osmolar iodinated contrast agents are used in current medicine, although nonionic ICM are more common [2]. In clinical practice, ICM are typically classified by osmolality. Low-osmolality ICM may be subcategorized further into (1) nonionic monomers, (2) ionic dimers, and (3) nonionic dimers.
Safety Studies of Nonionic Versus Ionic Iodinated Contrast Media Large population studies have demonstrated the relatively lower risk of nonionic ICM compared with ionic ICM. Comparative data from two older large-scale studies suggested that the incidence of mild adverse reactions to contrast media was 2.5% for ionic ICM, but only 0.58% for nonionic ICM. Severe reactions were reported in 0.4% of patients administered ionic ICM and 0% for severe reactions after administration of nonionic ICM [12,13]. Katayama et al. reported that in a series of 337,647 cases, the overall risk of an adverse drug reaction associated with ICM was 12.66% for ionic ICM and 3.13% for nonionic ICM. The risk of a very severe adverse drug reaction was 0.04% for ionic ICM and 0.004% for nonionic ICM [14]. In a meta-analysis of studies published during the 1980s, Caro, et al. documented risks of mortality and severe nonfatal reactions in high-osmolality ICM compared to nonionic ICM. These investigators calculated a rate of severe adverse drug reaction of 0.157% for high-osmolality ICM and 0.031% for nonionic ICM. The rate of a fatal adverse reaction was one death in 100,000 patients for both types of ICM [15].
Adverse Reactions to Iodinated Contrast Media Background Despite their poorer safety record, high-osmolality ICM are still used in current medicine, primarily because of their lower cost. These media should be used selectively. High-osmolality ICM have an increased risk for adverse contrast reactions, and a significantly higher risk for contrast-related severe
281
282
19
CT/MRI safety in functional neurosurgery
adverse events [2]. The overall safety of lowerosmolality nonionic ICM has been well established since the 1990s, but adverse events have been reported. Mild and moderate adverse reactions are generally uncommon. Severe and even fatal adverse effects are quite rare, but may occur unpredictably in some patients. Serious reactions may be preceded by a mild or moderate prodromal phase. A ‘‘test injection’’ administered before a contrast-enhanced CT may increase the risk for severe adverse events [2]. ICM may also impair kidney function in certain patients or exacerbate pre-existing renal insufficiency in persons with compromised kidney function. Although contrast-induced nephropathy is not an adverse allergic reaction, it is a serious adverse event that may have debilitating consequences for high-risk persons undergoing iodinated contrast enhanced CT [2].
Safety Issues: Magnetic Resonance Imaging MRI-Related Management of Metal Implants and Foreign Bodies Background Increasing use of technologically advanced MR systems during the past 20 years has introduced growing safety concerns over the MRI environment itself [16]. Compared to older machines, new MRI scanners have stronger static magnetic fields, faster and stronger gradient magnetic fields, and more powerful radiofrequency transmission coils. While there is no evidence that magnetic fields produce irreversible biologic effects, under certain conditions several features of high-field MRI equipment pose serious hazards for the body and for implanted metal devices [17]. Expanding clinical applications of deep brain stimulation
(DBS), in particular, require a new set of safety measures for performing MRI examinations in patients with implanted neurostimulation devices [16–18]. Metal implants warrant special consideration because they are typically located near, or contiguous with, brain structures or cerebral vasculature. As recent descriptions of several MR-related injuries and at least two fatalities illustrate, strict adherence to updated evidence-based safety guidelines on MRI technology is essential. Failure to follow the manufacturer’s guidelines when performing MRI on patients with a specific neuromodulation or other metal implant may have devastating consequences. In one reported case, the DBS electrode was heated during an MRI scan of the lumbar spine on a patient with Parkinson’s disease. The heating produced a radiofrequency lesion that led to permanent neurological damage [19]. This single case study further underscores the importance of literally complying with safety guidelines for performing MRI in persons with metallic implants. Patients may be subjected to severe injury if healthcare workers attempt to generalize about various conditions, positioning schemes, or other scanning scenarios stipulated for one neurostimulation system during MRI scanning, and then inadvertently apply these generalizations to the operation of other systems [16–18]. The primary hazards associated with MRI equipment in conjunction with implanted devices are categorized as follows [4,16,17,20].
Risks associated with the static magnetic field (Bo), including complications such as movement of ferromagnetic objects, twisting, heating, artifacts, and device malfunction produced by the static magnetic field. Risks associated with radiofrequency field (RF) effects, including complications arising from body coils and specific absorption rate (SAR).
CT/MRI safety in functional neurosurgery
Risks Associated with Static Magnetic Field Strength Projectile Effect The projectile effect, or the disturbing movement of ferromagnetic material, is a primary complication of metallic implants that may occur during an MRI. Also known as the missile effect, it is caused by interactions between the static magnetic field and MRI systems [21]. Magnetic translational and rotational forces that are exerted on a ferromagnetic object can move or dislodge the object from its implanted position. A magnet of high field strength can rapidly pull different types of ferromagnetic objects into the MRI scanner. The patient is subsequently at risk for injury by any number of objects, ranging from internal aneurysm clips and pins in joints to oxygen canisters and wheelchairs [22,23].
Heating The greatest risk for MRI scanning in a patient with a DBS implant is MRI-related heating of metallic objects, especially DBS leads. Heating is poorly tolerated in the central nervous system. When electrically conductive materials are introduced within the magnet and touch the bore of the MRI scanner, these materials may overheat. If a conductive object comes in contact with the patient’s tissue, it may burn his skin, possibly resulting in irreversible lesions [16,18,20]. In addition, conductive loops that come in contact with tattoos and eye-liners containing iron-oxides may cause burns [20]. A neurostimulation system used for DBS can generate variable levels of heating. Which levels are most likely to occur, and what factors are most likely to cause overheating depend upon the specific type of implanted device as well as various parameters used for a given MRI procedure [16,18,23].
19
Intrinsic factors that influence heating include the static magnetic field strength of the MRI system (which determines the transmitted RF used for the device operation); the electrical characteristics and configuration of the individual system (electrode, extension, length, orientation of the IPG); lengths and routing of the extension and leads; the impedance of the wires; and wire breakage [16]. Extrinsic parameters in the heating equation include the type of RF coil used (transmit/ receive body vs. transmit/receive head RF coils); the landmarking site; geometry of the RF coil and the quantity of the DBS lead present within this coil; SAR (amount of RF energy delivered); method for calculating the SAR based on a particular MR system; and quantity of RF energy (whole-body averaged SAR) required for imaging [16]. RF burns may occur if currents are induced into electrocardiographic leads, or into monitoring cables and coils that are placed on the patient’s skin surface [16]. The safety of MRI in patients with DBS may be increased by placing concentric loops of DBS electrode around the burr hole cap, by using a headonly receive coil, and by adhering to the vendor recommendations re the maximum SAR that can be tolerated [16]. For certain implants that have undergone empirical testing, clinically significant thermal changes may occur at 3.0-T but not at 1.5-T. Yet other data indicate that in some cases a particular implant may exhibit clinically significant levels of heating in seconds at 1.5-T but not at 3.0-T. The greatest risk, therefore, appears to be linked with the rate of temperature increase rather than the thermal change per se. According to one report, most heating occurred within the first minute of the MRI procedure and reached a steady-state within 15 min [16]. As noted previously, in order to mitigate risks of excessive heating, established product safety guidelines for MR scanning in patients with metallic implants must be diligently followed.
283
284
19
CT/MRI safety in functional neurosurgery
Even then, the guidelines should be used only insofar as they apply to the magnetic field strengths that have been evaluated and specified in the guidelines. MR scanning at either stronger and/or weaker magnetic field strengths than those indicated in the manufacturer’s guidelines may cause substantial heating. Unless extreme precautionary measures are taken, inadvertent heating may arise and ultimately produce severe injury in the patient [16–18].
Contrast Administration in MRI Scans: Safety Issues Indications for the use of gadolinium based magnetic contrast agents (GBCMAs) are similar to those noted above for CT scanning.
Allergic Reactions Allergic reactions have been linked to the use of GBMCAs in persons with impaired kidney function. Gadopentetate dimeglumine and gadoteridol have elicited adverse reactions such as anaphylaxis in patients with impaired renal function. The package insert for gadopentetate dimeglumine warns that a history of asthma or other allergic respiratory condition may increase the possibility of a reaction, including serious, fatal, life-threatening, anaphylactoid, cardiovascular reactions, or other idiosyncratic reactions [4].
Gadolinium-based ContrastInduced Nephrogenic Systemic Fibrosis (NSF) More disconcerting than the risk for allergic reactions is the mounting evidence for a linkage between GBMCAs administered to kidney disease patients and an emerging disease called
nephrogenic systemic fibrosis (NSF). Nephrogenic systemic fibrosis is a rare, progressive, and potentially fatal fibrosing disorder that affects patients with pre-existing renal dysfunction. It is closely associated with the use of GBMCAs [24]. The disorder was originally called nephrogenic fibrosing dermopathy (NFD) because of its involvement with the skin [18]. NSF is characterized as a systemic disease of connective tissue that targets skeletal muscle, skin, and tendons. The condition is definitively diagnosed by clinical evaluation and a deep skin biopsy of the dermis, subcutaneous fat, and fascia. The pathology involves increased deposits of collagen in connective tissues, resulting in a thickening and hardening of the skin of the extremities [24,25]. In severe illness, joints may become immobile or deformed. In extreme cases of limited motion, some patients may be confined to a wheelchair. NSF also may cause injury to the diaphragm, esophagus, heart, lung, pulmonary vasculature, and skeletal muscles [25]. The disease tends to develop slowly, but advances rapidly in about 5% of patients. At present, there is no consistently efficacious therapy [18]. The American College of Radiology (ACR) recommends that patients at risk for NSF from dialysis or chronic kidney disease be screened before receiving GBMCAs. Glomerular Filtration Rate (GFR) should be measured in patients older than 60 with a history of renal disease, hypertension, diabetes, and/or severe hepatic disease/liver transplant/pending liver transplant. Patients with hepatic dysfunction should undergo a GFR assessment as close as possible to the time at which the GBMCA is to be administered for the MRI examination [26]. GBMCAs should be avoided in patients with GFRs less than 30 mL/min/ 1.73m2 unless absolutely necessary. Persons unaware that they have kidney dysfunction may be identified through medical history. If a definitive diagnosis of kidney status is not known, immediate serum creatinine testing may be warranted in addition to a GFR assessment [26].
CT/MRI safety in functional neurosurgery
Conclusions The evidence presented in this chapter clearly reinforces current expert consensus that safety remains a paramount issue for patients undergoing either CT or MRI scanning. Radiographic safety measures must be properly implemented and followed at the level of the institution, neurosurgical team, and individual healthcare worker. It is incumbent upon clinicians to keep informed of the most recent information generated by the professional organizations that develop practice guidelines and issue advisories, including critical periodic updates.
References 1. Ross PJ, Ashamalla H, Rafla S. Advances in stereotactic radiosurgery and stereotactic radiation therapy. Radiation Therapist 2001;10(1):57-72. 2. Segal AJ, Ellis JH, Baumgartner BR. ACR manual on contrast media. 6th ed. Reston, VA: ACR; 2008. 3. Brenner DJ, Hall EJ. Computed tomography – an increasing source of radiation exposure. N Engl J Med 2007;357 (22):2277-84. 4. Chung SM. Safety issues in magnetic resonance imaging. State of the art. J Neuro-Ophthalmol 2002;22(1):35-9. 5. Coursey CA, Frush DP. CT and radiation: what radiologists should know. Appl Radiol 2008;37(3):22-9. 6. Nagataki S. Comment on: computed tomography and radiation exposure. N Engl J Med 2007;357 (22):2277-84. N Engl J Med 2008;358(8):850‐1. 7. Tubiana M. Computed tomography and radiation exposure. N Engl J Med 2008;358(8):850; author reply 852–3. 8. Fricke BL, Donnelly LF, Frush DP, Yoshizumi T, Varchena V, Poe SA, Lucaya J. In-plane bismuth breast shields for pediatric CT: effects on radiation dose and image quality using experimental and clinical data. Am J Roentgenol 2003;180(2):407-11. 9. Semelka RC, Armao DM, Elias J, Jr, Huda W. Imaging strategies to reduce the risk of radiation in CT studies, including selective substitution with MRI. J Magn Reson Imaging 2007;25(5):900-9. 10. Bauman G, Wong E, McDermott M. Fractionated radiotherapy techniques. Neurosurg Clin N Am 2006;17 (2):99-110. 11. Valk J. The role of CT and NMRI in neurosurgical diagnosis. Neurosurg Rev 1986;9(1–2):43-7.
19
12. Wolf GL, Arenson RL, Cross AP. A prospective trial of ionic vs. nonionic contrast agents in routine clinical practice: comparison of adverse effect. Am J Roentgenol 1989;152:939-44. 13. Lasser EC, Berry CC, Talner LB, Santini LC, Lang EK, Gerber FH, Stolberg HO. Pretreatment with corticosteroids to alleviate reactions to intravenous contrast material. N Engl J Med 1987;317(14):845-9. 14. Katayama H, Yamaguchi K, Kozuka T, Takashima T, Seez P, Matsuura K. Adverse reactions to ionic and nonionic contrast media. A report from the Japanese committee on the safety of contrast media. Radiology 1990;175 (3):621-8. 15. Caro JJ, Trindade E, McGregor M. The risks of death and of severe nonfatal reactions with high- vs lowosmolality contrast media: a meta-analysis. Am J Roentgenol 1991;156(4):825-32. 16. Rezai AR, Baker K, Tkach J, Phillips M, Hrdlicka G, Sharan A, Nyenhuis J, Ruggieri P, Henderson J, Shellock FG. Is magnetic resonance imaging safe for patients with neurostimulation systems used for deep brain stimulation (DBS)? Neurosurgery 2005;57:1056-62. 17. Shellock FG, Crues JV. MR procedures: biologic effects, safety, and patient care. Radiology 2004;232(3):635-52. 18. Kanal E, Barkovich AJ, Bell C, Borgstede JP, Bradley WG, Jr, Froelich JW, Gilk T, Gimbel JR, Gosbee J, Kuhni-Kaminski E, Lester JW, Jr, Nyenhuis J, Parag Y, Schaefer DJ, Sebek-Scoumis EA, Weinreb J, Zaremba LA, Wilcox P, Lucey L, Sass N. ACR Blue Ribbon Panel on MR safety. ACR guidance document for safe MR practices: Am J Roentgenol 2007;188 (6):1447-74. 19. Henderson JM, Tkach J, Phillips M, Baker K, Shellock FG, Rezai AR. Permanent neurological deficit related to magnetic resonance imaging in a patient with implanted deep brain stimulation electrodes for Parkinson’s disease: case report. Neurosurgery 2005;57 (5):E1063. 20. Stecco A, Saponaro A, Carriero A. Patient safety issues in magnetic resonance imaging: state of the art [Article in English, Italian]. Radiol Med (Torino) 2007;112 (4):491-508. 21. Shellock F. Metallic neurosurgical implants: evaluation of magnetic field interactions, heating, and artifacts at 1.5-Tesla. J Magn Reson Imaging 2001;14 (3):295-9. 22. Joint Commission. Sentinel event MRI safety alert. Preventing accidents and injuries in the MRI suite. Issue 38, February 14, 2008. http://www.jointcommission.org/SentinelEvents/SentinelEventAlert/sea_38.htm Accessed 20 May 08. 23. Shellock FG, Crues JV. Commentary: MR safety and the American college of radiology white paper. Am J Roentgenol 2002;178:1349-52.
285
286
19
CT/MRI safety in functional neurosurgery
24. Chewning RH, Murphy KJ. Gadolinium-based contrast media and the development of nephrogenic systemic fibrosis in patients with renal insufficiency. J Vasc Interv Radiol 2007;18(3):331-3. 25. Broome DR, Girguis MS, Baron PW, Cottrell AC, Kjellin I, Kirk GA. Gadodiamide-associated nephrogenic systemic
fibrosis: why radiologists should be concerned. AJR Am J Roentgenol 2007;188(2):586-92. 26. Weinreb JC. Improving gadolinium-based contrast safety. Imaging Biz.com 2008;3(2):1-2.
18 CT/MRI Technology: Basic Principles M. I. Hariz . L. Zrinzo
Introduction Ever since the beginning of human stereotactic neurosurgery in 1947 [1], the radiological study has been an absolute prerequisite for the very existence of this surgery. Furthermore, the radiological study has always constituted an integral part of the surgical procedure itself. It was indeed the limitations of conventional radiology (plain X-ray, pneumoencephalography, ventriculography, arteriography), which for a long time did put the limits for what could be achieved with stereotactic neurosurgery. In 1973, Hounsfield published the method of computerized transverse axial scanning [2], which rendered him a Nobel Price in 1979. This method, which came to be known as computed tomography (CT), has literally revolutionized all neurosurgery including stereotactic neurosurgery. To begin with, and due to the axial views provided by this new imaging technique, CT introduced a new coordinate language, which had to be adopted by stereotactic neurosurgeons [3]: the anteroposterior direction became Y instead of the previous X, the dorso-ventral became Z instead of Y, and the lateral direction became X instead of Z. The advent of MRI further stimulated developments in functional stereotaxis by providing much improved soft tissue contrast, and possibility to image the brain in any desired plan. MRI forced the development of new stereotactic frames, compatible with this new imaging modality. At the same time, concerns were raised about possible distortions in the images due to magnetic susceptibilities and magnetic field inhomogeneities. Lately, #
Springer-Verlag Berlin/Heidelberg 2009
concern arose also about safety of MRI scanning in patients who are implanted with deep brain stimulation (DBS) devices [4,5]. A prerequisite for accurate CT/MRI-guided functional stereotactic neurosurgery is the adaptation of the scanning method and the stereotactic frame to each other. Additionally, for stereotactic MRI scanning, accounting for the eventual geometric distortions and the adaptation of the imaging sequence to the functional brain target aimed at, are of paramount importance. In this chapter the general principles for performing stereotactic CT/MRI scanning in the practice of functional neurosurgery are detailed. A brief description of various imaging protocols in relation to brain structure being targeted is presented.
General Principles Although routine CT/MRI examinations are performed for diagnostic purposes, the stereotactic CT/MRIstudyisalocalizationprocedureandrepresents a crucial step of the functional neurosurgical procedure. The localization procedure delineates and defines a deep-seated structure (nucleus, part of a nucleus, or pathway) in relation to a coordinate system such that the structure may be surgically targeted. This intracranial structure may be readily ‘‘visualized’’ on MRI, provided a suitable imaging sequence, as in the case of the subthalamic nucleus (STN) and the globus pallidus internus (GPi). The structure may also be a particular area of the brain, which has to be defined in relation to visualized ventricular landmarks as in the case of the
270
18
CT/MRI technology: basic principles
ventral intermediate nucleus (Vim) of the thalamus, which cannot be visualized as such on either the CT or the MRI scans. Hence, it is the geometrical accuracy in performing the scanning and the accuracy of definition of brain targets and/ or ventricular landmarks that are the hallmarks of a stereotactic CT/MRI study. Therefore, all possible sources of errors that may interfere with the requirements of accuracy of a stereotactic CT/ MRI study, should be accounted for, or preferably eliminated. The first issue in that respect is that of the geometry and image-compatibility of the stereotactic frame.
Stereotactic Frame The general requirement of any head-containing stereotactic system is to provide compatible, wellvisualized, artifact-free, fiducials or landmarks on every relevant CT/MRI slice of the stereotactic imaging study. Besides, the relationship of these external landmarks to the head must be constant during the whole scanning and also between imaging study and the subsequent surgery. The number of visible landmarks on each CT/MRI slice varies according to which frame system is used; however, a minimum of three is required to define a zero origin for the anteroposterior Y and lateral X coordinates on the slice. The accuracy of measuring the height Z coordinate is much increased if the stereotactic frame contains a fiducial permitting the assessment of the dorso-ventral position of the axial slice containing the brain target, independently from the accuracy of movement of the CT couch (if a CT study is performed, i.e.), and hence, without relying on the accuracy of the Scoutview. In case MRI is used, it is an advantage for minimizing the measurement error, and minimizing the distortion at the periphery of the MR image, where the fiducials of the frame are as close as possible to the head, that is, as close as possible to the intracranial structure, the coordinates of which are to be measured.
Immobilization of the Patient Unlike conventional radiography, a CT study, and even more so, an MRI study lasts for a certain amount of time. In routine CT/MRI scanning, the immobility of the head is important to avoid movement artifacts on the picture. In stereotactic CT/MRI scanning, the immobility of the head in relation to the frame, and to the gantry or head coil is absolutely mandatory because the scanning is performed usually with very thin slices and because at least one of the target coordinates, that is, the height coordinate Z, is sometimes based on the primary position of the head at the beginning of the scanning. To achieve a strict immobility of the head of the nonanesthetized patient during the stereotactic CT/MRI scanning, there is only one solution: it is to secure the head to a frame with skull screws and to fix rigidly the frame to the supporting couch. The discomfort for the patient should be kept to a minimum and the patient should be able to tolerate the frame. Nonrigid fixation using noninvasive interfaces [6–8] or fiducials [9] have also been used, but they assume comprehensive cooperation of the patient during the imaging study, unless general anesthesia is used. A careful explanation of the stereotactic imaging procedure, combined if needed with slight sedation, will contribute to motivate most patients to achieve an acceptable immobility of the head. Otherwise, and especially if the imaging study requires long acquisition time, the patient, who in most cases suffers from a movement disorder, must have general anesthesia during the scanning [10,11].
Scanning Plane and Alignment Conventional diagnostic axial CT images are most often acquired with a scanning plane parallel to Reid’s baseline, the orbito-metal line, or the skull base in general [12,13]. In stereotactic CT/MRI scanning, it is in many stereotactic
CT/MRI technology: basic principles
systems the geometry of the stereotactic frame that dictates the axial scanning plane. Since in many centers, most functional brain targets are still related to the atlas, and thus, to the anterior commissure-posterior commissure (AC–PC) line of the third ventricle, the stereotactic frame is mounted to the head in such a way that its base ring will be parallel to the average orientation of the AC–PC line. [14–16]. Some stereotactic systems that employ various N-shaped, Z-shaped, or V-shaped localizers do not require any specific alignment of the head and frame in relation to the CT gantry [17–23]. However, a dedicated software is then required to enable coordinate calculation. Frameless stereotactic techniques do not require any specific orientation of the head within the CT gantry or the MRI head coil. These methods rely completely on dedicated software that allows reformatting of the images according to AC–PC orientation and fusion between different imaging modalities.
Scan Thickness Even though an axial stereotactic CT/MRI scan containing the target does generally also provide the X and Y coordinates of that target in relation to any external reference fiducials, the dorsoventral Z coordinate is far from being easy to obtain accurately: on axial scans; the Z coordinate depends mainly on the thickness of the CT/ MRI slice. The volume of the voxel, that is, the impact of the partial volume effect on the boundaries of the brain structure, can be reduced by examining thinner slices. However, the spatial resolution of the image is known to decrease with thinner scans because thinner sections have a more unfavorable signal-to-noise ratio than thicker slices. This can be improved by increasing the degree of contrast on the image. The contrast resolution on the image can also be improved by longer scan time, which unfortunately prolongs the duration of the scanning and, in case of CT,
18
increases the radiation dose to the patient. Typically, for functional stereotactic purposes, it is recommended to perform the CT/MRI scanning with contiguous scans not thicker than 2 mm. Furthermore, should a sagittal or coronal reformatting of the axial CT image be desirable, the image resolution of such a reformatted image is greatly enhanced if the axial scanning has been performed with thin contiguous slices. However, it must be stressed that a reformatted or reconstructed CT image almost never reaches the degree of resolution of the axial one, and it does always carry an increased risk of measurement error. MRI readily allows for coronal and sagittal scans, although, in analogy with CT scanning, it is typically the axial scans that are used for target calculation, and most dedicated functional stereotactic softwares are based on evaluation of, and targeting on, axial scans. The distortion on the image is less on axial than on coronal and sagittal views. In all other aspects, the same rules apply for stereotactic MRI scanning (thickness of slices, parallelity with AC-PC plan, immobilization of head and frame, etc.) as for stereotactic CT scanning. The advantage of an MRI coronal scan is that it allows more readily to assess the depth of the target in relation to other visible structures, such as the relation of GPi to optic tract or the relation of STN to Substantia Nigra. Furthermore, MRI is exquisite, when used together with a dedicated software, to rehears the trajectory of the probe from cortex to target, and to allow a virtual ‘‘dry-run’’ through the structures traversed by the probe on its way to the target. Finally, on postoperative stereotactic images, MRI allows very precise evaluation of the location of DBS electrode contacts in relation to targeted structure and its surroundings, provided thin scans are obtained. It must be kept in mind that eventual geometrical distortions on MRI, which are due to the ‘‘potato ship’’ effect at the periphery of the image, are less—all other parameters equal—when the fiducials of the stereotactic frame are as close as possible to the head such as
271
272
18
CT/MRI technology: basic principles
is always the case when using the Laitinen system [24] and quite often the case—depending on the size of the head—with the Leksell system.
Measurements and Calculations For details concerning the various methods to calculate target coordinates proper to each stereotactic system, it is referred to the chapters of this book describing the respective systems. Generally speaking, there are three ways to perform measurements and calculations of the CT/ MRI coordinates of a brain target: One may use either manual measurements on enlarged hard copies, eventually with the help of matching mm-grids, or the inbuilt software of the scanner, or a specifically inbuilt or separately provided software dedicated to stereotactic measurements of a specific system. In some stereotactic systems, all three ways may be available in conjunction. However, it must be kept in mind that although a given CT/MRI machine is proven reliable as far as calibrations, accuracy, and software properties are concerned, and although a given stereotactic frame is proven mechanically accurate and CT/MRI-compatible with minimal image artifacts, the combination of the two during a stereotactic scanning on a ‘‘real’’ patient may result in measurement errors greater than what would be attributable to either of them separately. In 1992, Maciunas published an extensive application accuracy evaluation of the four most commonly used stereotactic frame systems [25]. The study was done on test phantoms scanned stereotactically with a modern CT machine with slice thicknesses from 8 mm down to 1 mm. Even with the 1-mm thick scanning, a mean target error of between 1 and 1.9 mm and a maximal error between 3.1 and 5.0 mm, dependent on the frame used, were obtained! The errors were considered greater than the mechanical accuracy of the frames and greater than the error attributable to the imaging procedure.
Clinical Applications in Functional Stereotaxis A substantial proportion of stereotactic CT/MRI studies is performed in view of a functional stereotactic procedure, that is, a stereotactic surgery for pain, movement disorder, or psychiatric disorder, where an ablative lesion or an implantation of a chronic electrode will be made in an anatomically ‘‘normal’’ structure of the brain. In these cases, the specific sub-nuclei to be targeted may not be readily seen on the CT/MRI image, and their position has to be determined in relation to visible internal reference structures, most commonly the AC-PC line.
CT Scanning for Functional Stereotaxis Computed tomography is both user- and patientfriendly. In CT-guided functional stereotaxis, the brain target cannot be visualized as such even by the most sophisticated CT scanner. Instead, the anatomical position of these structures is always defined in relation to ventricular landmarks, as has been the case during conventional ventriculography. Whereas ventriculography provides lateral and anteroposterior views of the third ventricle, CT provides an axial view. Therefore, the stereotactic CT scanning should be done not only with thin scans, but also the scanning plane should be as parallel as possible to the AC-PC line. One main difficulty has been that the ACPC line does not readily ‘‘show up’’ on the CT Scoutview. Both its exact position within the brain and its inclination in relation to any bony landmarks are subject to individual variations [26]. It has been suggested that the scanning plane which is the most ‘‘parallel’’ to the AC-PC line would be the Glabella-inion plane [27,28]. The Twining line, i.e., a line between the tuberculum sellae and the protuberantia occipitalis interna has also been advocated [29,30]. Using these bony landmarks,
CT/MRI technology: basic principles
an initial scanning of the area of the third ventricle is performed, and if it appears that the AC and PC are visualized on different slices, the gantry of the CT machine is re-angulated accordingly, and the scanning partly repeated. This technique may be time consuming and is only possible when gantryindependent CT-localizers are used. However, this technique has become rather obsolete with the advent of independent dedicated imaging softwares enabling reformatting of CT scans into axial images parallel to the AC-PC line. In other methods in which the parallelity of the scanning plane and AC-PC line can be averaged thanks to a special design of the used frame, a superimposition of the CT slice containing one of the commissures on the CT slice containing the other commissure may be sufficient to measure the length and define the level of the AC–PC line, and to assess its inclination in relation to the stereotactic frame [24,31]. This presupposes that the nonparallelity between the plane of scanning and the plane of the AC-PC line is not too exaggerated. One of the major pitfalls of stereotactic CT scanning for functional stereotactic procedures is the uncertainty in determining the height Z coordinate of the brain target. This uncertainty is not only due to the thickness of the CT slice, to eventual movement inaccuracies of the CT couch, to the geometry of the actual stereotactic frame, and/or to the nonparallelity between scanning plane and plane of AC-PC line, as has been discussed previously. The source of error in defining a functional brain target coordinate may also lie in the uncertainty in defining on CT the ventricular commissures in relation to which this very brain target is to be defined. Spiegel et al. had shown already in 1952 that the dorso-ventral thickness of the AC ranged from 1.5 to 4 mm with a mean of 3 mm [32]. Therefore, even if the CT scanning is performed with contiguous 1.5-mm thin slices, there may still be a risk of error in determining the dorso-ventral position of the AC. This risk of error is theoretically the same for the determination of the PC. Since the
18
height level of the brain target is defined in relation to the height level of the line joining the AC and PC, a repercussion of these errors on the target coordinates in the sagittal plane would ensue. Therefore some authors still prefer to perform ventriculography in addition to a stereotactic CT study (and even in addition to MRI) for the determination of a functional brain target [33–36]. Others do indeed rely solely on the stereotactic CT study and define consequently and always the AC on the CT slice lying 4 mm below the one depicting the ventral-most part of the foramen of Monro, and the PC on the slice immediately above the one showing the beginning of the aqueduct [6,24,37].
MRI Scanning for Functional Stereotaxis General Requirements Two aspects have to be considered for an MR image to be suitable as a source for direct determination of coordinates in functional neurosurgery: First, its ability to provide a good discrimination of the brain structures to be targeted, thus allowing the images to act as the patient’s individualized atlas; and second, the geometrical accuracy of the MR images has to be validated. General requirements to perform stereotactic imaging may include the following: (1) an MRI compatible stereotactic frame than can be attached to the head support of the head coil MRI; (2) a localizer with fiducials as close as possible to the head; (3) an MRI sequence adapted to the brain structure being targeted; (4) a field of view that fits the stereotactic frame; (5) a slice thickness allowing good compromise between resolution and signal-noise-ratio: with current technology, 1.5–2.0 mm slice thickness is a good compromise; (6) the pixel size in the axial plane (x, y) has to be isometric, i.e., based on a square matrix; (7) enough slices to cover with good margins the whole brain target
273
274
18
CT/MRI technology: basic principles
area, without gap between slices, and without overlapping slices; (8) the use of a transmitter– receiver head coil to ensure maximal signal, and to ensure safe postoperative imaging in patients with DBS implants; (9) regular checks of the gradient fields to minimize distortions; (10) avoiding patient movements during scanning. With few exceptions [34–36], most functional stereotactic surgeons today do use a stereotactic MRI study to determine target coordinates; however, it seems that many workers still relate the target to its atlas-defined position in relation to visualized ventricular landmarks on the MRI study. This approach ignores the main advantage of MRI: unlike ventriculography and CTscanning, MRI is not a homogeneous imaging method. Depending on the parameters of imaging, MRI studies can visualize differently and unequally well, various structures in the brain.
Thalamic Subnuclei The only commonly used brain targets in functional stereotactic neurosurgery, which still cannot be visualized as such on stereotactic thin slice MRI, are the subnulei of the thalamus. Stereotactic MRI cannot convincingly visualize several thalamic targets used in the treatment of a range of functional brain disorders: the ventroinetermediate (Vim) nucleus, targeted for treatment of tremor, the ventral oral posterior (Vop) and ventral oral anterior (Voa) nuclei, targeted, although rarely, in the treatment of dystonia and Parkinson’s, the ventroposterolateral (Vpl) and ventroposteromedial (Vpm) targeted for treatment of chronic pain, the centre median, -parafascicular (CMpf) nuclei, and ventral oral internus (Voi) targeted for the treatment of Tourette disorder. All these thalamic targets still have to be defined based on atlas coordinates in relation to AC– PC. Nonetheless, MRI, especially a T2-weighted MRI sequence can delineate the thalamo-capsular border, making it easier to determine at least the laterality of the Vim/Vop/Voa targets.
Posteroventral Pallidum The subdivisions of the globus pallidus (Globus pallidus pars interna, GPi, laminae medullaris interna and externa, Globus pallidus pars externa, GPe) and their surrounding structures (putamen, internal capsule, optic tract) can be visualized stereotactically on thin slice axial and coronal MRI using various sequences. One such sequence [38] is a nonvolumetric proton density sequence (TR/ TE 4000/15, echo-train 7, field of view 250 mm, slice thickness 2 mm, gap 0, matrix 210 256, excitations 3, imaging time 6 min and 5 s), which depicts exquisitely the details of the pallidal area (> Figure 18-1). Another sequence that has been validated by Vayssie`re et al. in Montpellier use volumetric T1 sequences [39,40]. Other MRI scanning methods, based on inversion recovery sequences, have been described by Starr et al. [41]. In all these cases, since the target itself is readily visualized, there is no need to refer to the landmarks of the third ventricle and to an atlas to obtain the location of the target in the individual patient, especially since it has been shown that the individual target location may vary substantially between patients, and also between the two hemispheres in the same patient [38,41,42].
Subthalamic Nucleus (STN) In surgery on STN, most workers who rely on MRI for targeting this structure determine its position on T1-weighted images, in relation to third ventricle landmarks and brain atlases. The few publications reporting on the use of stereotactic MRI for direct visualization and targeting of STN describe volumetric T2 weighted sequences with a rather long acquisition time [10,11,43–45], sometimes requiring reformatting of images and/or additional T1-weighted sequences that are used for targeting, and often necessitating general anesthesia during imaging. These sequences do allow exquisite visualization
CT/MRI technology: basic principles
18
. Figure 18-1 Preoperative, 2-mm thin, axial proton density stereotactic MRI scans, in two patients: (a) patient with atrophic pallidii and enlarged putaminae and (b) patient with large pallidii
of STN. The present authors, together with others [46], implemented a nonvolumetric T2-weighted MRI sequence (TR 3000–4000, TE 80–100) allowing individual visualization of the STN with fast acquisition sequences that allow imaging without general anesthesia in most patients (between 3 min 5 s and 7 min 48 s, depending on the MRI machine) (> Figure 18-2). Here also, direct visualization will make it possible to target the center of the visualized STN at surgery, without need for an indirect localization based on the atlas and the AC–PC landmarks. In all these imaging procedures, the present authors have used the Laitinen stereotactic apparatus [24,46] or the Leksell stereotactic system [47], together with MRI machines of various makings (Siemens, Philips, General Electric), regularly calibrated and assessed for field inhomogeneities and other sources of distortion.
Postoperative Stereotactic Imaging In functional stereotactic neurosurgery, stereotactic postoperative MRI is mandatory: It will unequivocally demonstrate the exact location of the stereotactic lesion or implanted DBS electrode.
. Figure 18-2 Preoperative, 2-mm thin, axial T2 weighted stereotactic MRI scan, at a 4 mm level below the AC–PC line, showing the subthalamic nuclei
This imaging should ideally be obtained with the same sequence parameters as the preoperative one. Concern about the risks of performing an MRI study on patients with implanted DBS systems has been raised [5,48]. However, a number of safety measures can be adopted to address these concerns such as use of a transmitter–receiver head coil, and ensuring the average specific absorption rate (SAR) is no more than 0.4 W/kg [4]. Recently the group of Philip Starr in San Francisco reviewed their experience over 7 years: 405 patients with 746 implanted DBS systems were imaged using a
275
276
18
CT/MRI technology: basic principles
variety of scanning techniques and on various 1.5-Tesla MRI machines, for a total of 1,071 imaging studies. They reported no adverse event, even with a SAR of up to 3 W/kg in few patients [49]. In patients with DBS, it is an advantage if postoperative imaging is performed immediately after surgery, while the frame is still on the head. If the neuro-pacemaker has already been implanted, MR images can still be obtained provided the voltage of the pacemaker is set to zero, and the output is switched off, before the patient enters the MRI room. One should be able to assess the location of the lead within the visible target and if necessary, return the patient to the operating room to relocate the lead should it become apparent that it is misplaced. The present authors consider that a DBS implant procedure is not complete until electrode localization in the intended target is verified by means of a stereotactic MRI (> Figures 18-3– > 18-4), or a stereotactic CT with image fusion to the preoperative stereotactic images.
Conclusions Notwithstanding the imaging method used, the stereotactic CT or MRI study constitutes an integral part of the stereotactic surgical procedure. It is sometimes the most difficult part of the surgery. Taking into consideration its limits and potential errors, and the fact that a measurement or calculation error at this stage may have harmful repercussions on the results of surgery and on the patient, responsibility for the stereotactic CT/MRI study lies with the neurosurgeon. Therefore, the surgeon must be well acquainted with the scanning technique, its potential pitfalls, and with target coordinate calculation, as well as being acquainted with the stereotactic frame being used. It might be wise to keep in mind the following statement made in 1985 by Lars Leksell in a paper entitled ‘‘Stereotaxis and nuclear magnetic resonance’’ [50]: ‘‘In clinical practice brain imaging can now be divided in two parts: the diagnostic
. Figure 18-3 Postoperative, 2-mm thin, axial proton density stereotactic MRI scan, at the level of AC–PC, showing artifacts of DBS electrodes in the posteroventral pallidum
. Figure 18-4 Postoperative, 2-mm thin, axial T2 weighted stereotactic MRI scan, at a 4 mm level below the AC–PC line, showing artifacts of DBS electrodes in the subthalamic nuclei
neuroradiology and the preoperative stereotactic localization procedure. The latter is part of the therapeutic procedure. It is the surgeon’s responsibility and should be closely integrated with the operation.’’
CT/MRI technology: basic principles
References 1. Spiegel EA, Wycis HT, Marks M, Lee AJ. Stereotaxic apparatus for operations on the human brain. Science 1947;106:349-50. 2. Hounsfield GN.Computerized transverse axial scanning (tomography): Part I. Description of system. Br J Radiol 1973;46:1016-22. 3. Goerss S, Kelly PJ, Kall B, Alker GJ. A computed tomographic stereotactic adaptation system. Neurosurgery 1982;10:375-9. 4. Rezai AR, Phillips M, Baker KB, Sharan AD, Nyenhuis J, Tkach J, et al. Neurostimulation system used for deep brain stimulation (DBS): MR safety issues and implications of failing to follow safety recommendations. Invest Radiol 2004;39:300-3. 5. Henderson JM, Tkach J, Phillips M, Baker K, Shellock FG, Rezai AR. Permanent neurological deficit related to magnetic resonance imaging in a patient with implanted deep brain stimulation electrodes for Parkinson’s disease: case report. Neurosurgery 2005;57:E1063. 6. Laitinen LV, Liliequist B, Fagerlund M, Eriksson AT. An adapter for computed tomography-guided stereotaxis. Surg Neurol 1985;23:559-66. 7. Hitchcock E. Stereotactic-computerized tomography interface device. Appl Neurophysiol 1987;50:63-7. 8. Bergstro¨m M, Greitz T. Stereotaxic computed tomography. A J Roentgenol 1976;127:167-70. 9. Bjartmarz H, Rehncrona S. Comparison of accuracy and precision between frame-based and frameless stereotactic navigation for deep brain stimulation electrode implantation. Stereotact Funct Neurosurg. 2007;85:235-42. 10. Bejjani BP, Dormont D, Pidoux B, Yelnik J, Damier P, Arnulf I, Bonnet AM, et al. Bilateral subthalamic stimulation for Parkinson’s disease by using three-dimensional stereotactic magnetic resonance imaging and electrophysiological guidance. J Neurosurg 2000;92:615-25. 11. Patel NK, Heywood P, O’Sullivan K, Love S, Gill SS. MRI-directed subthalamic nucleus surgery for Parkinson’s disease. Stereotact Funct Neurosurg 2002;78:132-45. 12. Lang J, Schlehahn F, Jensen HP, Lemke J, Klinge H, Muhtaroglu U. Cranio-cerebral Topography as a Basis for Interpreting Computed Topograms. In Lanksch W, Kazner E, editors. Cranial computerized tomography. Berlin: Springer, 1976. p. 24-36. 13. Lange S, Golde G. Resolution characteristics of computerized tomography and their impact on quantitative brain diagnosis. In: Lanksch W, Kazner E, editors. Cranial computerized tomography. Berlin: Springer; 1976. p. 52-5. 14. Laitinen LV. The Laitinen system. In: Lunsford LD, editor. Modern stereotactic neurosurgery. Boston: Nijhoff M; 1988. p. 99-116. 15. Lunsford LD, Leksell D. The Leksell system. In: Lunsford LD, editor. Modern stereotactic neurosurgery. Boston: Nijhoff M; 1988. p. 27-46
18
16. Mundinger F, Birg W. The imaging-compatible riechertmundinger system. In: Lunsford LD, editor. Modern stereotactic neurosurgery. Boston: Nijhoff M; 1988. p. 13-25. 17. Goerss S, Kelly PJ, Kall B, Alker GJ. A computed tomographic stereotactic adaptation system. Neurosurgery. 1982;10:375-9. 18. Kelly PJ. Contempory stereotactic ventralis lateral thalamotomy in the treatment of parkinsonian tremor and other movement disorders. In Heilbrun MP, editor. Stereotactic neurosurgery, Vol 2: Concepts in neurosurgery. Baltimore: Williams and Wilkins; 1988. p. 133-47. 19. Heilbrun MP. Computed tomography-guided stereotactic systems. Clin Neurosurg. 1982;31:564-81. 20. Kelly PJ, Goerss SJ, Kall BA. Evolution of contemporary instrumentation for computer-assisted stereotactic surgery. Surg Neurol. 1988;30:204-15. 21. Rosenfeld JV, Barnett GH, Palmer J. Computed tomography guided stereotactic thalamotomy using the BrownRoberts-Wells system for non-Parkinsonian movement disorders. Technical note. Stereotact Funct Neurosurg. 1991;56:184-92. 22. Brown RA. A computerized tomography-computer graphics approach to stereotaxic localization. J Neurosurg. 1979;50:715-20. 23. Couldwell WT, Apuzzo MLJ. Initial experience related to the use of the Cosman-Roberts-Wells stereotactic instrument. Technical note. J Neurosurg. 1990;72:145-8. 24. Hariz MI, Laitinen LV. The Laitinen apparatus. In: Gildenberg PL and Tasker RR, editors. Textbook of stereotactic and functional neurosurgery. New York: Mc Graw Hill; 1997. p. 87-94. 25. Maciunas RJ, Galloway RL Jr, Latimer J, Cobb C, Zaccharias E, Moore A, et al. An independant application accuracy evaluation of stereotactic frame systems. Stereotact Funct Neurosurg. 1992;58:103-7. 26. Talairach J, David M, Tournoux P, Corredor H, Kvasina T. Atlas d? Anatomie Ste´re´otaxique. Paris, Masson, 1957. 27. Tokunaga A, Takase M, Otani K. The glabella-inion line as a baseline for CT scanning of the brain. Neuroradiology. 1977;14:67-71. 28. Takase M, Tokunaga A, Otani K, Hori T. Atlas of the human brain for computed tomography based on the glabella-inion line. Neuroradiology. 1977;14:73-9. 29. Ohye C, Kawashima Y, Hirato M, Wada H, Nakajima H. Stereotactic CT scan applied to stereotactic thalamotomy and biopsy. Acta Neurochir. 1984;71:55-68. 30. Spiegelmann R, Friedman WA. Rapid determination of Thalamic CT-stereotactic coordinates: A method. Acta Neurochir (Wien). 1991;110:77-81. 31. Gouda KI, Freidberg SR, Larsen CR, Baker RA, Silverman ML. Modification of the Gouda frame to allow stereotactic biopsy of the brain using the GE 8800 computed tomographic scanner. Neurosurgery. 1983;13:176-81.
277
278
18
CT/MRI technology: basic principles
32. Spiegel EA, Wycis HT, Baird HW. Studies in Stereoencephalotomy. I. Topical relationships of subcortical structures to the posterior commissure. Confin Neurol. 1952;12:121-33. 33. Fox MW, Ahlskog JE, Kelly PJ. Stereotactic ventrolateralis thalamotomy for medically refractory tremor in postlevodopa era Parkinson’s disease patients. J Neurosurg. 1991;75:723-30. 34. Pinto S, Le Bas JF, Castana L, Krack P, Pollak P, Benabid AL.Comparison of two techniques to postoperatively localize the electrode contacts used for subthalamic nucleus stimulation. Neurosurgery. 2007; Suppl 2:285-92. 35. Breit S, LeBas JF, Koudsie A, Schulz J, Benazzouz A, Pollak P, et al. Pretargeting for the implantation of stimulation electrodes into the subthalamic nucleus: a comparative study of magnetic resonance imaging and ventriculography. Neurosurgery. 2006;58 Suppl 1: ONS83-95. 36. Merello M, Cammarota A, Cerquetti D, Leiguarda RC. Mismatch between electrophysiologically defined and ventriculography based theoretical targets for posteroventral pallidotomy in Parkinson’s Disease. J Neurol Neurosurg Psychiatry. 2000;69:787-91. 37. Hariz MI, Bergenheim AT. A comparative study on ventriculographic and computed tomography-guided determinations of brain targets in functional stereotaxis. J Neurosurg. 1990;73:565-71. 38. Hirabayashi H, Tengvar M, Hariz MI: Imaging of the pallidal target. Mov Disord. 2002;17 Suppl 3: S162-6. 39. Vayssiere N, Hemm S, Zanca M, Picot MC, Bonafe A, Cif L, et al. Magnetic resonance imaging stereotactic target localization for deep brain stimulation in dystonic children. J Neurosurg. 2000;93:784-90. 40. Vayssiere N, Hemm S, Cif L, Picot MC, Diakonova N, El Fertit H, et al. Comparison of atlas- and magnetic resonance imaging–based stereotactic targeting of the globus pallidus internus in the performance of deep brain stimulation for treatment of dystonia. J Neurosurg. 2002;96:673-9.
41. Starr PA, Vitek JL, DeLong M, Bakay RAE. Magnetic resonance imaging-based stereotactic localization of the globus pallidus and subthalamic nucleus. Neurosurgery. 1999;44:303-14. 42. Vayssiere N, Gaag van der N, Cif L, Hemm S, Verdier R, Frerebeau P, et al. Deep brain stimulation for dystonia confirming a somatotopic organization in the globus pallidus internus. J Neurosurg. 2004;101:181-8. 43. Zhu XL, Hamel W, Schrader B, Weinert D, Hedderich J, Herzog J, et al. Magnetic Resonance Imaging-Based Morphometry and Landmark Correlation of Basal Ganglia Nuclei. Acta Neurochir. 2002;144 959-69. 44. Schrader B, Hamel W, Weinert D, Mehdorn HM. Documentation of electrode localization. Mov Disord. 2002;17 Suppl 3:S167-74. 45. Starr PA, Christine CW, Theodosopoulos PV, Lindsey N, Byrd D, Mosley A, et al. Implantation of deep brain stimulators into the subthalamic nucleus: technical approach and magnetic resonance imaging–verified lead locations. J Neurosurg. 2002;97:370-87. 46. Hariz MI, Krack P, Melvill R, Jorgensen JV, Hamel W, Hirabayashi H, et al. A quick, and universal method for stereotactic visualization of the subthalamic nucleus before and after implantation of deep brain stimulation electrodes. Stereotact Funct Neurosurg. 2003;80:96-101. 47. Chen CC, Pogosyan A, Zrinzo LU, Tisch S, Limousin P, Ashkan K, et al. Intra-operative recordings of local field potentials can help localize the subthalamic nucleus in Parkinson’s disease surgery. Exp Neurol. 2006;198:214-21. 48. Baker KB, Tkach JA, Phillips MD, Rezai AR. Variability in RF-induced heating of a deep brain stimulation implant across MR systems. J Magn Reson Imaging. 2006;24:1236-42. 49. Larson PS, Richardson RM, Starr PA, Martin AJ. Magnetic resonance imaging of implanted deep brain stimulators: experience in a large series. Stereotact Funct Neurosurg. 2008;86:92-100. 50. Leksell L, Leksell D, Schwebel J. Stereotaxis and nuclear magnetic resonance. J Neurol Neurosurg Psychiatry. 1985;48:14-18.
22 Diagnostic PET in Image Guided Neurosurgery B. Ballanger . T. van Eimeren . A. P. Strafella
Positron emission tomography (PET) is one of the most popular imaging techniques in current neuroscience research. Due to ongoing innovations, PET continues to flourish as an interesting diagnostic tool in clinical neurology. While some imaging techniques, such as computed tomography (CT) and magnetic resonance imaging (MRI), are able to identify structural changes in the body, PET imaging is capable of detecting areas of functional changes even prior anatomical abnormalities are observed. The PET scanner does this via the use of radiolabeled molecular probes that have different rates of uptake, depending on the type and function of tissue involved. The changing of regional blood flow in various anatomical structures (as a measure of the injected positron emitter) can be visualized and relatively quantified with a PET scan. The first section of this chapter will trace the origin of PET and describe the instruments and radiopharmaceuticals. The second section will highlight studies which have application to stereotactic and/or functional neurosurgery. Whenever applicable, the current role of functional magnetic resonance imaging (fMRI) in preoperative assessment of discrete brain functions will be covered as well.
Background What is PET? PET is a nuclear medical imaging technique first developed by Michel Ter-Pogossian, Michael E. Phelps and others (1975) at the Washington #
Springer-Verlag Berlin/Heidelberg 2009
University School of Medicine [1,2], which produces a three-dimensional image of functional processes in the body with excellent sensitivity and moderate anatomic resolution (4–7 mm). PET may be used to investigate subjects in the resting state or in relationship to an event such as the occurrence of a seizure or the induction of pain. PET enables to study cerebral energy metabolism and receptor function. PET provides in vivo measurements of injected biologically active substances that have been radioactively labeled (radioligands or tracers). These radionuclides are incorporated into compounds normally used by the body such as glucose or water and then injected into the body to trace where they become distributed. Such labeled compounds are known as radiotracers. To conduct the scan, a short-lived radioactive tracer isotope is injected into the body. The radioligand decays by emitting a positron. Then the positron encounters and annihilates with an electron, producing two 511 keV gamma photons radiating at 180 from each other. PET scanners have detectors placed on opposite sides of the region from where the photons are emitted (within the patient), and the detectors register an event only if both detectors record the photon emission at the same time. As the photons are always emitted 180 from each other, this serves to both localize and quantify these events, and hence register the amount of metabolic activity. The technique depends on simultaneous or coincident detection of the pair of photons; photons which do not arrive in pairs (i.e., within a few nanoseconds) are ignored. This provides PET with a unique ability to detect and quantify
308
22
Diagnostic PET in image guided neurosurgery
physiologic and receptor processes in the body, especially in the cancer cells, that is not possible by any other imaging technique. The data collected by the PET scanner is mathematically reconstructed to produce tomographic images of tissue radioactivity concentration. Glucose is the main substrate for energy supply of the brain by oxidation. In fact, brain cells use glucose as fuel, and the more active the brain cells are, the more they will consume radioactive glucose. Using an imaging tracer that is like glucose, the PET scan is able to quantify the activity of brain tissue. Areas of less activity will use less energy, and areas with increased activity will use more energy.
Radiopharmaceuticals The key to molecular imaging in nuclear medicine is radiotracers. Radiotracers allow in vivo evaluation of different functions in the brain, namely cerebral blood flow, glucose metabolism, protein synthesis, neurotransmission, and neuroreceptor density. The application mainly depends on their chemical properties.
The standard tracer for measurement of the cerebral metabolic rate of glucose is 18F-2-fluoro-2deoxy-D-glucose (18F-FDG) [6], a sugar, for which the acquisition period is typically an hour. FDG is transported into tissue and phosphorylated to FDG phosphate, like glucose, but does not undergo significant further metabolism. Thus, it accumulates in brain in proportion to local cerebral metabolic rate of glucose. Although FDG-PET is a very powerful diagnostic tool because it gives a comprehensive image of synaptic function, it lacks specificity with regard to individual transmitter systems. There are neurodegenerative diseases involving specific neurotransmitters that do not have a distinct appearance on FDG-PET scans, probably because the cells synthesizing and releasing these transmitters are too few or too dispersed to have a local impact on energy consumption. The most evident example is Parkinson’s disease (PD), where the substantia nigra pars compacta with its profound degeneration of dopaminergic neurons is too small and metabolically too similar to the rest of the midbrain to be easily recognized in FDG-PET scans.
Specific Neurotransmitter System Unspecific Brain Activity Water
H2O15 radioactive water is administrated by bolus intravenous injection to obtain blood flow images, with a data acquisition time of about 2 min. Injections and data acquisition can be repeated at intervals equals or superior to five times the radioactive half-live of 15O (10 min). This technique is well adapted for comparing a ‘‘resting’’ condition with a condition of sensory, motor or cognitive activation. In activated brain regions, the increase in regional cerebral blood flow (rCBF) leads to a local increase in the tissue radioactive water content detected [3–5]. Glucose
In these instances, we need tracers that image specifically that particular neurotransmitter system. Several radiotracers have been developed for PET that are ligands for specific neuroreceptor subtypes (e.g., dopamine D2, serotonin 5-HT1A, etc.), transporters (e.g., [11C]McN5652, [11C] DASB, [11C]MP), or enzyme substrates (e.g., 6-FDopa). The development of PET radiolabeled receptor ligands for brain imaging holds great promise for improved specificity and sensitivity in cerebral functional imaging. Dopaminergic System
To investigate the function and integrity of presynaptic dopaminergic terminals, the most widely used tracer is the 18F-Fluorodopa (FDopa). This tracer is a substrate to DOPA decarboxylase
Diagnostic PET in image guided neurosurgery
which is expressed in abundance by dopaminergic neurons. The product, 18F-fluorodopamine, accumulates in proportion to decarboxylase activity which in turn reflects the amount of viable dopaminergic cells. Therefore, the amount of radioactivity resulting from 18F-FDopa in a region-of-interest (ROI) will reflect the number of functionally intact dopaminergic neurons within this particular region, as well as presynaptic dopamine uptake, decarboxylation to dopamine and storage [7]. The adequate modeling and interpretation of data from FDopa PET studies are far from simple and different techniques have been developed, e.g., the calculation of the influx constant Ki according to Patlak for the quantification of FDopa accumulation in the striatum [8,9]. Postsynaptically, dopamine exerts actions through several subtypes of the dopamine receptor. The dopamine receptor family consists of 5 subtypes D1-D5. In order to investigate the role of each receptor subtype, selective and high-affinity PET radioligands are required. For the dopamine D1-subtype the most commonly used ligand is [11C]-Schering 23390 ([11C]SCH) or [11C]-NNC 112, whereas for the D2/D3subtype [11C]-raclopride is a common tracer [10,11]. [18F]-fallypride is a suitable PET tracer for the investigation of extrastriatal D2 receptors [12]. For the other subtypes no suitable radioligands have been developed yet. Other tracers like [11C]-methylphenidate ([11C]MP), [11C]-dihydrotetrabenazine ([11C] DTBZ), provide complementary information on the integrity of the dopaminergic system. Methylphenidate inhibits dopamine reuptake and enhances synaptic dopamine levels. One of its isomers, dl-threo-methylphenidate has been labeled with Carbon-11 for PET [13]. Its binding in the human brain is reversible, highly reproducible and saturable and thus [11C]MP is deemed an appropriate PET ligand to measure dopamine transporter (DAT) availability [14]. The dopamine transporter (DAT) is
22
located on the plasma membrane of nerve terminals in a small number of neurons in the brain especially in the striatum and nucleus accumbens, but also in the globus pallidus, cingulate cortex, olfactory tubercle, amygdala and midbrain [15]. DAT regulates the dopamine concentration in the synaptic cleft through reuptake of dopamine into presynaptic neurons; it plays a central role in the spatial and temporal buffering of the released dopamine. Accordingly, DAT provides a good site for monitoring the function and the integrity of the dopaminergic neurons. Several DAT agents have been developed for diagnosis PD and monitoring the treatment of PD patients, based on DAT antagonists such as [11C]MP [14,16], [11C]cocaine [17,18] and 18 F-2-b-carbomethoxy-3-b-(4-fluorophenyl)tropane, ([18F]CFT) [19]. [11C]DTBZ is utilized as a tracer for in vivo imaging of the vesicular monoamine transporter (VMAT2) system [20,21]. Serotoninergic System
5-Hydroxy-tryptamine (5-HT) or serotonin is a monoamine transmitter produced in brainstem raphe nuclei and released at cortical level through widely distributed ascending pathway. Serotonin function seems to be altered in many neurologic and psychiatric disorders, in particular in depression, obsessive compulsive disorders, Alzheimer’s disease (AD) and schizophrenia. Currently at least seven major serotonin receptor classes have been identified, some of them consisting of different subtypes [22]. However, the available radioligands permit the investigation of 5-HT1A and 5-HT2A receptors only. Recently, two antagonists ligands of 5-HT1A receptors have been developed for PET studies. The first one is 18F-trans-4-fluoro-N-2-[4-(2methoxyphenyl)-1-piperazinyl]ethyl]-N-(2pirydyl)cyclohexanecarboxamide known as [18F] FCWAY which presents a much higher affinity than endogenous serotonin for 5-HT1A receptors, comparable to that of the original WAY-100635 labeled with 11C [23]. The second one is the
309
310
22
Diagnostic PET in image guided neurosurgery
4-(20 -methoxyphenyl)-1-(20 -(N-20 -pirydynyl)p-18F-fluoro-benzamido)ethylpiperazine known as [18F]MPPF which has an affinity for the 5HT1A receptor close to that of endogenous serotonin [24]. Using these two ligands, a high level of tracer uptake has been observed in high density in the hippocampus, amygdale, parahippocampal gyrus, hypothalamus, temporal pole, insula, anterior and posterior cingulated gyri [25]. Conversely, 5-HT2A receptors are present in all neocortical regions, with lower densities in hippocampus, basal ganglia and thalamus. The cerebellum and striatum are virtually devoid of 5-HT2A receptors. A quantization of 5-HT2A receptors has been possible with 18F-altanserin and 18F-setoperone [26,27]. These ligands are selective 5-HT2A antagonists and they show good reproducibility. The ligand [11C]MDL 100907 has previously been introduced to image the 5-HT2A receptor in human brain [28]. a-11C-methyl-L-tryptophan ([11C]AMT) is used as a PET marker of brain serotonin synthesis [29]. The a-methyl-L-tryptophan is converted to a-methylserotonin, which is not a substrate for monoamine oxidase and therefore accumulates in the brain. Cholinergic System
Acetylcholine (ACh) is synthesized in certain neurons by the enzyme choline acetyltransferase from the compounds choline and acetyl-CoA. In the synaptic cleft, the enzyme acetylcholinesterase (AChE) converts acetylcholine into the inactive metabolites choline and acetate. Nicotinic ACh receptors have been implicated in many psychiatric and neurologic diseases, including depression and cognitive and memory disorders, such as Alzheimer’s and Parkinson’s disease. Thus 11C-labeled nicotine was used to visualize and quantify nicotinic receptors in the brain. In recent years, the piperidine analogs C-11labeled N-methyl-4-piperidyl-acetate (MP4A) [30] and N-methyl-4-piperidyl-propionate [31] have been developed for in vivo imaging of cere-
bral AChE with PET [32,33]. As a substrate of AChE these tracers are hydrolyzed and accumulate depending on enzyme activity. AChE in human cortex is mainly expressed in the cholinergic axons, and to a lesser extent also in some cholinoceptive neurons. With impaired function and neurodegeneration of these cholinergic axons, the amount of cortical AChE is reduced. GABAergic and Glutamatergic System
GABA (Gamma Aminobutyric acid) is the most important inhibitory neurotransmitter. Its transmission is altered in epilepsy and other psychiatric disorders. Because the GABA receptor is abundant in the cortex and is very sensitive to damage, it represents a reliable marker of neuronal integrity. The tracer most widely used for central GABA binding sites is flumazenil (FMZ) labeled with [11C] [34]. The highest degree of binding is observed in the medial occipital cortex, followed by other cortical areas, the cerebellum, thalamus, striatum and pons with very low binding in the white matter. FMZ is a biochemical marker of epileptogenicity and neuronal loss; benzodiazepine receptor-density changes are more sensitive than 18F-FDG in detecting hippocampal sclerosis and benzodiazepine receptor studies were useful in the selection of patients for targeted surgery and for predicting outcome of these procedures [35,36]. As an antagonist of GABA, glutamate is the main excitatory neurotransmitter in the cortex, and alterations of glutamatergic neurotransmission are associated with many neurologic diseases. In the post-synaptic cell, glutamate receptors, such as the NMDA (N-methyl-D-aspartate) receptor, bind glutamate and are activated. Because of its role in synaptic plasticity, it is believed that glutamic acid is involved in cognitive functions like learning and memory in the brain. The tracers used to study this system ([11C]MK 801, [11C]-ketamine, [18F]-fluoroethyl-TCP and [18F]-memantine) have only
Diagnostic PET in image guided neurosurgery
a low specificity to the NMDA receptor, and relevance for clinical studies has not been established [37].
Limits of PET Radionuclides used in PET scanning are typically isotopes with short half lives such as 11C (20 min), 13N (10 min), 15O (2 min), and 18 F (110 min). Due to their short half lives, the use of PET is limited by the need for an on-site cyclotron. Therefore, PET differs from other imaging techniques in requiring more expensive equipment and highly specialized personnel, not only for scanning but also for production of the radiotracers. Unlike CT or MRI, few hospitals and universities are capable of maintaining such imaging systems.
Functional MRI fMRI is the use of MRI to measure the hemodynamic response related to neural activity in the brain. It is one of the most recently developed forms of neuroimaging. It can noninvasively record brain signals without risks of radiation inherent in other scanning methods, such as PET scans. It can record on a spatial resolution in the region of 3–6 mm with a relatively good temporal resolution (in the order of seconds) compared with techniques such as PET. Since neurosurgery relies on a precise delineation of the structural and functional aspects of brain, the role for fMRI in neurosurgical planning can be significant, especially when the presence of a tumor alters the expected location of a function, or when the location of the tumor is in an area with an uncertain function such as association cortices or language-related processes. An emerging group of investigators have reported fMRI results that are consistent with
22
electrophysiology, PET, cortical stimulation, and magneto-encephalography and serve to document that fMRI does provide a source of precise functional and structural information for neurosurgery [38–45]. Further, the potential role of fMRI in directing decisions about surgical and diagnostic procedures has also been demonstrated [46,47]. fMRI can be useful in the selection of patients for whom a surgical resection is attempted and could aid in the decision-making whether to operate on a patient who has been previously considered inoperable. fMRI is a useful tool in the decisional scheme of treatment of low-grade astrocytomas or arteriovenous malformations (AVM) in the rolandic area in intact or slightly impaired patients. fMRI can be repeated in selected patients with slow growing brain tumors or congenital lesions such as AVM to study cortical reorganization phenomena. In many neurosurgical centers, the Wada-test (barbiturate injection into one of the internal carotid arteries) was substituted by fMRI to determine the hemispheric dominance of language. After the dura had been opened and/or part of the tumor had been removed, the functional tissue that the surgeon wants to preserve might have shifted. As a future potential, real-time fMRI could identify this functional tissue intraoperatively, comparable to an inverse intraoperative frozen section diagnosis.
PET as a Differential Diagnostic Tool Since the development of PET, in vivo imaging has become the method of preference to assess a variety of neurodegenerative and neuropsychiatric disorders [48,49].
311
312
22
Diagnostic PET in image guided neurosurgery
Parkinson’s Disease Advances in Understanding Parkinson’s Disease In the last two decades, measurements of rCBF (with methods like PET or fMRI) as an indicator of neuronal activity have greatly advanced our understanding of the neuronal processes underlying motor and cognitive deficits in patients with idiopathic Parkinson’s disease (IPD). In initial H2O15 PET activation studies on motor planning and execution in IPD, the hypothesis of an excessive cortical inhibition from corticostriatal loops had been strengthened [50,51]. The separate contributions of dopamine loss in the striatum and direct cortical involvement in cognitive impairment in IPD had been the subject of several studies in recent years (see [52,53] for reviews). rCBF PET and fMRI have also been successfully used to investigate the functional impact of pharmacological or surgical interventions in IPD (see [54,55] for reviews). Currently, the identification of genetically defined at-risk populations yields great potential to identify adaptive neuronal mechanisms in a preclinical state of parkinsonism with neuroimaging techniques (see [55] for review).
Differentiate Between Idiopathic Parkinson’s Disease and Atypical Parkinsonism The clinical diagnosis of IPD seems fairly straightforward in most cases. Conversely, clinical pathology studies demonstrated that up to 25% of the patients clinically diagnosed with IPD had the post mortem diagnosis of progressive supranuclear palsy (PSP), multiple system atrophy (MSA), vascular parkinsonism or AD [56]. Yet, establishment of an early and accurate diagnosis impacts on management, helps to avoid inappropriate treatment and assists in the
evaluation of novel drugs. A wide range of objective neuroimaging methods currently contribute to establish diagnosis. Especially functional imaging techniques like PET and single photon emission computed tomography (SPECT) can be used with various radioligands to provide quantitative assessment of dopamine functioning in- and outside of the brain. In terms of a differential diagnostic tool, the degeneration of dopaminergic neurons with reduction of their respective PET markers has been described in all diseases that cause parkinsonism, but the relative involvement of the pre- and postsynaptic metabolism in the striatum and the rostral and caudal parts of the striatum provides some distinction between diseases. PET can demonstrate the disturbance of dopamine synthesis that is the hallmark of PD. The most widely used tracer in PD is FDopa. In Parkinson’s disease, the PET scan shows a characteristic pattern of reduced striatal uptake for FDopa, and it begins to appear very early in the course of the disease [57] most strongly on the side opposite to the major motor signs [58,59] and predominantly in the posterior part of the putamen, indicating loss of more than 50% of dopaminergic neurons projecting to this part of the striatum [60]. This typical differential intrastriatal distribution of reduced uptake is often referred to as the ‘‘rostrocaudal gradient’’ [61]. Although the ability to demonstrate reductions in FDopa uptake in the putamen can help in the diagnosis of IPD, the differentiation between IPD, PSP, MSA, or corticobasal degeneration (CBD) is more difficult. It had been indicated that in PSP and MSA, nigral projections to the caudate nucleus become involved earlier in the course of disease resulting in more equally decreased FDopa uptake in terms of rostrocaudal distribution [62]. However, due to a significant overlap, FDopa PET alone seems not to be able to distinguish between different forms of parkinsonism [63]. An alternative measure of presynaptic dopamine terminal integrity is PET
Diagnostic PET in image guided neurosurgery
imaging with radiotracers that bind either to DAT ([11C]MP) or the VMAT2 ([11C]DTBZ). These methods show similar findings in IPD to those seen with FDopa PET and the ability to differentiate different forms of parkinsonism is similarly low [64]. The distinction is much better (80–90%, according to Brooks 2002 [65]) for indicators of striatal postsynaptic D2 receptors ([11C]raclopride) or glucose metabolism ([18F]FDG) [66] (see > Figure 22-1). This distinction between PSP and MSA on the one side and IPD on the other side rests mainly on the fact that in PSP and MSA, dopaminergic neurodegeneration affects both pre- and postsynaptic nerve fibers, whereas in IPD there is a presynaptic dopaminergic deficit in the striatum while postsynaptic striatal neurons are fairly intact and their D2 receptors binding and glucose metabolism have been shown to be normal or even increased in
22
untreated patients [67,68]. In contrast, striatal glucose metabolism is reduced in 80–100% of probable atypical parkinsonism patients [65]. With FDG, CBD differs from PD by a metabolic decrease in premotor, primary motor, supplementary motor, primary sensory and parietal associative cortices but also in caudate and thalamus [69]. Based on FDG or D2 receptors studies, however, distinction among MSA, PSP and CBD is barely possible, limiting the sensitivity of this investigation in the differential diagnosis of parkinsonism [65,70,71]. In many neuroimaging facilities only SPECT might be available. Generally speaking, in terms of differential diagnosis of parkinsonism, the same strengths and limitations apply to SPECT imaging of the presynaptic dopamine transporter (123I-b-CIT or 123 FP-CIT) and postsynaptic dopamine receptor (123I-IBZM).
. Figure 22-1 Representative PET images of FDOPA, RACLO and FDG at the mid-striatal level from one patient with Parkinson’s disease (PD) (top row) and one patient with multiple system atrophy (MSA) (bottom row). Each FDOPA, RACLO and FDG image is scaled relative to common maximum and background levels. At the putamen level, note the marked decrease of dopamine D2 receptor binding (RACLO) and glucose consumption (FDG) in MSA patient which cannot be found in PD patient. Reduction of putaminal DOPA influx constants are similarly visible in PD and MSA (from Antonini et al., Brain 1997;120(12):2187–95)
313
314
22
Diagnostic PET in image guided neurosurgery
Dementias Dementia is a clinical syndrome most commonly characterized by impaired short- and long-term memory and associated with deficits in many cortical functions which interfere significantly with the activities of daily life. Dementia is a common disease and prevalence continues to grow with increased human life expectancy. The most frequent cause is Alzheimer’s disease, accounting for 60–70% of all dementias in the elderly, followed by Lewy body dementia (LBD) (10–25%), vascular dementia (10–15%) and frontotemporal dementia (FTD) (5–10%). Synaptic dysfunction or loss – a hallmark of different types of dementia – entails at the molecular level several mechanisms that finally result in decreased energy demand. Therefore, assessment of glucose metabolism with FDG-PET is a valid tool for imaging the energy metabolism of the brain and its typical changes in dementia [72,73] as it can differentiate Alzheimer’s from other confounding types of dementia [74–76]. A consistent finding that has been noted since the earliest PET studies in AD is the hypometabolism affecting the temporal and parietal association cortex [77,78]. Recently, some studies using voxel-based comparison against normal reference data show that the posterior cingulate gyrus and the precuneus are also impaired early in the course [79]. In the parieto-temporal association cortices of AD patients, the reduction in glucose metabolism is greater than the reductions in blood flow and oxygen metabolism [78]. In contrast to other dementia types, glucose metabolism in the basal ganglia, primary motor and visual cortex, and cerebellum is usually well preserved [72–75]. Indeed FDG-PET in dementia with Lewy bodies reveals changes similar to those seen in AD plus additional hypometabolism in primary and associative visual cortices [80–83]. Occipital hypometabolism is the feature of DLB that discriminates it from AD. FDG-PET
scanning in FTD is associated with hypometabolism in the frontal, anterior and medial temporal cortices [84,85]. Low striatal DAT activity (e.g., indexed by [11C]MP) occurs in DLB but is normal in AD, making DAT scanning particularly useful in distinguishing between the two disorders [86]. FTD is a syndrome that can be clinically difficult to distinguish from AD, but in FTD amyloid deposition is not a characteristic pathological finding. Recently, it has been shown that N-methyl[11C]2-(40 methylaminophenyl)6-hydroxy-benzothiazole (PIB, a PET tracer with amyloid binding properties) could potentially aid in differentiating between FTD and AD [87,88] (see > Figure 22-2). The assessment of specific neurotransmitter systems with PET is likely to contribute substantially to clinical distinction between different neurodegenerative diseases that may lead to dementia. Receptor ligands for the cholinergic, dopaminergic, and serotoninergic system and newly developed tracers that label amyloid plaques are likely to play an important role. The full clinical relevance of these developments probably will turn up when more specific and also coursemodifying drugs for dementia treatment become available.
PET as a Preoperative Assessment Tool Brain Tumors Diagnosis and Staging PET has emerged as a powerful diagnostic tool in differentiating malignant from benign tumors. Indeed, the uncontrolled cellular proliferation is the hallmark of malignant transformation and offers the perfect target for diagnosis and evaluation with functional imaging. To this end, a
Diagnostic PET in image guided neurosurgery
22
. Figure 22-2 PIB standardized uptake values (SUV) of four representative subjects. FTD PIB negative: a 65-year-old patient with a clinical diagnosis of frontotemporal dementia (FTD; MMSE = 29, CT scan normal) with a PIB retention similar to healthy control. FTD PIB positive: a 75-year-old patient with a clinical diagnosis of FTD (MMSE = 27, CT scan normal) with a PIB retention similar to that found in AD patients. AD: scan of a typical patient with Alzheimer disease. SUV were obtained using the time interval 40–60 min. The results of this study indicate that the majority of FTD patients displayed no PIB retention (FTD negative) in line with the assumption that amyloid deposition is a characteristic neuropathological feature of AD, but not FTD. However, two of the ten patients with clinical diagnosis of FTD tested in this study (FTD positive) had a PIB retention similar to the AD patients suggesting that their true diagnosis might be AD. Therefore, PIB could potentially aid to differentiate between FTD and AD (from Engler et al., Eur J Nucl Med Mol Imaging 2008;35:100–6)
number of radiotracers – targeting changes in glucose metabolism (FDG), protein (e.g., [11C]methionine) or DNA (e.g., [11C]-thymidine) synthesis – have been developed to exploit these differences between malignant and normal cells. FDG-PET
Currently, the evaluation of brain tumors with FDG-PET is widely used in clinical oncology as the rate of glucose utilization is directly proportional to the degree of malignancy [89]. FDG uptake in low-grade gliomas (which are mostly grade II in adults) is usually similar to that of normal white matter, whereas most grade III anaplastic gliomas have a FDG uptake similar to or exceeding that of normal gray matter. Untreated glioblastomas, the most malignant
gliomas (grade IV), also have increased uptake of FDG, which might be heterogeneous throughout the tumor owing to the microscopic and macroscopic necroses that are typical in this tumor type [90]. Accordingly, FDG-PET is a valuable tool for accurate differentiation of lowand high-grade gliomas (> Figure 22-3). Across oncologic applications, the sensitivity and specificity of FDG-PET ranged from 84 to 87% and 88 to 93%, respectively. However, the main limitation of FDG for clinical studies of brain tumors is the high glucose consumption of normal gray matter (45 mmol/100 g/min) that may be in the same range as malignant tumors. Thus, even malignant tumors may be missed if surrounded by intact gray matter. Ac-
315
316
22
Diagnostic PET in image guided neurosurgery
. Figure 22-3 Co-registered MRI and FDG and 11C-methioinine PET images in glioblastoma. The contour delineates the areas of contrast enhancement (top row, middle) and is projected onto all other images. On the FDG images, only part of the contrast-enhancing area shows uptake comparable to normal gray matter, indicating the most aggressive part of the tumor, whereas methionine uptake exceeds the contrast enhancing area, which includes low-grade and tumor infiltration zones (from Herholz et al., Lancet Neurol 2007;6(8):711–24)
cordingly, evaluation of glucose consumption in brain tumors can only be done reliably if the location of the tumor is accurately known, best by digital image coregistration with MRI or CT. Amino Acid Uptake
Most brain tumors show an increased uptake of amino acids that is probably due to increased carrier-mediated transport at the blood–brain barrier (BBB). Increased uptake is also seen in most low-grade gliomas in the absence of BBB damage, which is a substantial advantage over CT, MRI, and FDG-PET [91–93]. Other 18 F-tagged radioisotopes (e.g., tyrosine) have also been found useful to improve diagnostic assessment of cerebral gliomas [94]. The most commonly used radiolabeled amino acid is 11C-L-Methionine (MET). Many studies have reported the use of MET to investigate malignancy, extent of tumor spread, effec-
tiveness of therapy, and prognosis of brain tumors [95–99]. Although MET PET has a limited ability to grade gliomas, it will provide reliable information about the extent of tumor infiltration.
Prognosis PET helps to assess tumor localization, extension and degree of malignancy, but histopathological examination of the tissue still is indispensable for definitive diagnosis and prognosis. Stereotactic biopsy is the least invasive way to obtain a specimen for histopathological classification of brain lesions. To this end, FDG-PET is used as stereotactic PET for directing biopsies accurately in the abnormal foci of brain tumors [100–103]. Many malignant gliomas are heterogeneous, but grad-
Diagnostic PET in image guided neurosurgery
ing needs to be done on the most malignant parts, which are commonly difficult or impossible to identify with standard structural imaging. The most metabolically active tumor part on PET (FDG or amino acid) indicates the most informative location for taking a biopsy [103–105]. Furthermore, in biopsy-proven low-grade gliomas, tumoral FDG uptake correlates well with the risk of malignant transformation [106,107].
Differentiation Between Recurrent Tumor and Radiation Necrosis Detection of recurrent tumor is an important issue because growth of recurrent tumor will lead to increase of symptoms and ultimate death of the patient. FDG-PET has been used successfully for that purpose in high-grade tumors [108], for detection of malignant progression in low-grade gliomas [109] and to distinguish radiation necrosis from recurrent tumor. Classically recurrent tumor is ‘‘hot’’ on FDG-PET studies and radiation necrosis is ‘‘cold’’; however, these two processes are often interleaved, so the reported sensitivity and specificity is quite low [110]. But there is increasing evidence that amino acid tracers can provide this discrimination [111].
Epilepsy Epilepsy is one of the most common neurological conditions. Almost 60% of patients respond to the first tried antiepileptic drug. However, 20% of people with epilepsy continue to have seizures despite adequate anti-epileptic drug treatment. This failure of drug treatment has led to an increasing interest in neurosurgery for epilepsy, particularly surgical approaches which aim to remove the ‘epileptic focus’ and, therefore offer the opportunity for these patients to be-
22
come free of seizures. One of the main objectives of presurgical investigation in patients with medically refractory epilepsy is to define the boundaries of the epileptogenic region to be resected. The epileptogenic zone refers to the region of cerebral cortex that is both necessary and sufficient to generate epileptic seizures, hence its entire removal is required for a successful outcome. Toward this goal, chronic intracranial EEG evaluation remains the gold standard. However, because this method carries a risk of morbidity and possible mortality, it is appropriate only when reliable conclusions cannot be obtained by less invasive methods. In parallel, a wide range of imaging techniques is valuable for imaging the epileptogenic zone, including highresolution T1 MRI, T2 signal quantitation, MR spectroscopy, diffusion imaging, PET, SPECT and simultaneous EEG-fMRI. MRI is often able to identify the source of seizure in patients with focal epilepsy. However, 20 to 30% of potential surgical candidates with focal epilepsy have normal MRI [112]. Interictal FDG-PET has been shown to be more sensitive than MRI in the identification of seizure of temporal lobe origin [113] and a valuable tool in patients with intractable epilepsy without a structural lesion [114–116]. The characteristic finding is a regional reduction in glucose uptake during the interictal state [117–119]. However, when investigating partial epilepsy, the benzodiazepine ligand [11C]FMZ may also be used which is a selective antagonist of GABAA– BZD receptors. The reduction of FMZ binding is much more focally restricted than reductions of FDG uptake [120] and the area of focal reduction of FMZ binding is probably also a better indicator of the epileptogenic zone that needs to be resected to become seizure free [121,122]. Therefore FMZ-PET can be more useful in preoperative planning than [18F]FDG alone. Although interictal PET provides useful information in temporal and extratemporal lobe epilepsy, its
317
318
22
Diagnostic PET in image guided neurosurgery
role is more important in patients with normal structural imaging.
Functional Brain Mapping with PET and fMRI The neurosurgeon often has to balance the benefits of brain tissue removal (e.g., in focal epilepsy or mass lesions) and the potentially devastating iatrogenic disruption of brain functions. Therefore a precise delineation of the structural and functional aspects of brain can prove very helpful in the preoperative decision making process, especially when the presence of a tumor alters the expected location of a function. Two of the most widely used approaches include PET and fMRI. Advantages of fMRI are: higher spatial and temporal resolution, more and different functional runs, shorter examination time, wider availability, longitudinal examinations, non-invasiveness and cost-effectiveness, easy registration to anatomical images. Advantages of PET are: higher signalto-noise ratio, lesser susceptibility to artifacts (motion, draining veins). Moreover, in deep brain stimulation (DBS), functional imaging has been used to investigate the impact of the procedure, but also identified potential stimulation areas.
Motor Mapping H2O15 PET and fMRI both are able to delineate brain regions activated by volitional movements, namely the primary sensorimotor cortex, the supplementary motor area, the lateral premotor cortex and the superior parietal lobule. Preoperative motor mapping with fMRI has been described [123–125] and validated against (the more invasive) electric cortical stimulation (ECS). Majos and colleagues studied 33 patients with brain tumors in the rolandic area and found an agreement of both methods of 98%
for combined motor and sensory representations [124]. Using 3T fMRI and ECS, Roessler et al. studied 22 patients with gliomas involving the primary motor cortex [125]. Motor foci were successfully detected with fMRI in all patients, but a successful intraoperative stimulation of the primary motor cortex was possible only in 77% of the patients. Moreover, in this subgroup, the motor focus in ECS and fMRI was identical within 1 cm. These results point to the direction, that motor mapping with fMRI or PET are safe and reliable techniques to assess the risk of a motor deficit following surgical procedure.
Language Mapping The lateralization of language is of special interest in patients with medial temporal lobe epilepsy and with tumors of the ventral frontal and temporal lobe. To this end, the selective intracarotid amobarbital application (Wada test) had been the gold standard for almost 50 years. Recently however, the Wada test gets more and more replaced by less invasive procedures such as fMRI. Klo¨ppel and Bu¨chel reviewed four studies comparing the Wada test with fMRI based language lateralization [126]. The authors summarize an agreement of about 90% of the two methods and note the advantage of fMRI to additionally provide a precise localization of language functions. They predicted that fMRI will be most widely used to assess language lateralization. But there might be an essential advantage of PET on the clinical application of language activation studies that challenges a predominant role of fMRI. Active speaking during language production tasks does not induce technical artifacts (as it is common with fMRI), and therefore, direct monitoring of task performance is possible even in functionally impaired subjects [127–129].
Diagnostic PET in image guided neurosurgery
Deep Brain Stimulation Pre- and post-operative imaging of H2O15 PET has been used to define the effects of deep brain stimulation in cases of Parkinson’s disease, depression and pain syndromes. For instance, concerning subthalamic nucleus (STN) stimulation in PD, three H2O15 PET studies found an enhanced movement-related activity in the dorsolateral and mesial prefrontal cortex, two areas known to be underactive in unmedicated PD [130–132]. For the treatment chronic pain syndromes, Davis and associates report on the changes on pre- and postoperative PET scanning in patients undergoing placement of thalamic stimulators. The investigators found that DBS caused activation of the contralateral anterior cingulate cortex, a known centre involved in pain and analgesia [133]. In parallel there has been some evidence that chronic electrical stimulation of the primary motor cortex (MCS) may relieve motor symptoms of PD. This surgical technique has been proposed as an alternative for selected PD patients who are considered poor candidates for DBS of the STN. However, to date the MCS technique remains a controversial procedure. Recently, using PET, our group [134] suggest that while unilateral MCS is probably a simpler and safer surgical procedure than DBS of STN, it did not improve motor performance nor significantly modify the activation pattern of movement-related rCBF in patients with advanced PD. These observations along with recent negative clinical experiences in PD patients [135] raise some controversy about the efficacy of this therapeutic modality in PD. Sometimes, functional imaging studies even identify new DBS targets. Based on the converging PET findings that the subgenual cingulate region (Brodmann area, BA, 25) is overactive in treatment resistant depression, Mayberg and colleagues tested DBS of adjacent white matter tracts to modulate BA25. They found a sustained
22
remission of depression in four of six patients. This clinical improvement was associated with a marked reduction in local cerebral blood flow as well as changes in downstream limbic and cortical sites [136]. Using PET, a possible cerebral origin of cluster headache has been visualized in the hypothalamic gray matter [137] prompting the successful use of DBS in the inferior posterior hypothalamus in intractable chronic cluster headache patients [138–140].
References 1. Phelps ME, et al. Application of annihilation coincidence detection to transaxial reconstruction tomography. J Nucl Med 1975;16(3):210-24. 2. Ter-Pogossian MM, et al. A positron-emission transaxial tomograph for nuclear imaging (PETT). Radiology 1975;114(1):89-98. 3. Posner MI, et al. Localization of cognitive operations in the human brain. Science 1988;240(4859):1627-31. 4. Fox PT, et al. Mapping human visual cortex with positron emission tomography. Nature 1986;323(6091):806-9. 5. Raichle ME, et al. Practice-related changes in human brain functional anatomy during nonmotor learning. Cereb Cortex 1994;4(1):8-26. 6. Reivich M, et al. The [18F]fluorodeoxyglucose method for the measurement of local cerebral glucose utilization in man. Circ Res 1979;44(1):127-37. 7. Snow BJ, et al. Human positron emission tomographic [18F]fluorodopa studies correlate with dopamine cell counts and levels. Ann Neurol 1993;34(3):324-30. 8. Firnau G, et al. Cerebral metabolism of 6-[18F]fluoro-L-3, 4-dihydroxyphenylalanine in the primate. J Neurochem 1987;48(4):1077-82. 9. Martin WR, et al. Nigrostriatal function in humans studied with positron emission tomography. Ann Neurol 1989;26(4):535-42. 10. Farde L, et al. Stereoselective binding of 11C-raclopride in living human brain – a search for extrastriatal central D2-dopamine receptors by PET. Psychopharmacology (Berl) 1988;94(4):471-8. 11. Rinne JO, et al. Increased density of dopamine D2 receptors in the putamen, but not in the caudate nucleus in early Parkinson’s disease: a PET study with [11C]raclopride. J Neurol Sci 1995;132(2):156-61. 12. Mukherjee J, et al. Brain imaging of 18F-fallypride in normal volunteers: blood analysis, distribution, testretest studies, and preliminary assessment of sensitivity to aging effects on dopamine D-2/D-3 receptors. Synapse 2002;46(3):170-88.
319
320
22
Diagnostic PET in image guided neurosurgery
13. Ding YS, et al. Pharmacokinetics and in vivo specificity of [11C]dl-threo-methylphenidate for the presynaptic dopaminergic neuron. Synapse 1994;18(2):152-60. 14. Volkow ND, et al. A new PET ligand for the dopamine transporter: studies in the human brain. J Nucl Med 1995;36(12):2162-8. 15. Ciliax BJ, et al. The dopamine transporter: immunochemical characterization and localization in brain. J Neurosci 1995;15(3 Pt 1):1714-23. 16. Lee CS, et al. In vivo positron emission tomographic evidence for compensatory changes in presynaptic dopaminergic nerve terminals in Parkinson’s disease. Ann Neurol 2000;47(4):493-503. 17. Logan J, et al. Graphical analysis of reversible radioligand binding from time-activity measurements applied to [N-11C-methyl]-(-)-cocaine PET studies in human subjects. J Cereb Blood Flow Metab 1990;10(5):740-7. 18. Telang FW, et al. Distribution of tracer levels of cocaine in the human brain as assessed with averaged [11C]cocaine images. Synapse 1999;31(4):290-6. 19. Haaparanta, M., et al. [18F]CFT ([18F]WIN 35,428), a radioligand to study the dopamine transporter with PET: biodistribution in rats. Synapse 1996;23(4):321-7. 20. Koeppe RA, et al. Kinetic evaluation of [11C]dihydrotetrabenazine by dynamic PET: measurement of vesicular monoamine transporter. J Cereb Blood Flow Metab 1996;16(6):1288-99. 21. Chan GL, et al. Reproducibility studies with 11C-DTBZ, a monoamine vesicular transporter inhibitor in healthy human subjects. J Nucl Med 1999;40(2):283-9. 22. Mansour A, et al. Biochemical anatomy: Insights into the cell biology and pharmacology of dopamine and serotonin systems in the brain. In Schatzberg AF, Nemeroff CB, editors. Textbook of psychopharmacology. Washington, DC: American Psychiatric Press; 1998. p. 55-73. 23. Gunn RN, et al. Tracer kinetic modeling of the 5-HT1A receptor ligand [carbonyl-11C]WAY-100635 for PET. Neuroimage 1998;8(4):426-40. 24. Costes, N., et al. A 18F-MPPF PET normative database of 5-HT1A receptor binding in men and women over aging. J Nucl Med 2005;46(12):1980-9. 25. Costes, N., et al. Modeling [18F]MPPF positron emission tomography kinetics for the determination of 5-hydroxytryptamine(1A) receptor concentration with multiinjection. J Cereb Blood Flow Metab 2002;22 (6):753-65. 26. Blin, J., et al. [18F]setoperone: a new high-affinity ligand for positron emission tomography study of the serotonin-2 receptors in baboon brain in vivo. Eur J Pharmacol 1988;147(1):73-82. 27. Kapur, S., et al. Reliability of a simple non-invasive method for the evaluation of 5-HT2 receptors using [18F]-setoperone PET imaging. Nucl Med Commun 1997;18(5):395-9.
28. Hinz, R., et al. Validation of a tracer kinetic model for the quantification of 5-HT(2A) receptors in human brain with [11C]MDL 100,907. J Cereb Blood Flow Metab 2007;27(1):161-72. 29. Rosa-Neto, P., et al. Stability of a-[11C]methyl-L-tryptophan brain trapping in healthy male volunteers. Eur J Nucl Med Mol Imaging 2005;32(10):199-204. 30. Irie, T., et al. , Brain acetylcholinesterase activity: validation of a PET tracer in a rat model of Alzheimer’s disease. J Nucl Med 1996;37(4):649-55. 31. Kilbourn MR, et al. In vivo studies of acetylcholinesterase activity using a labeled substrate, N-[11C]methylpiperdin-4-yl propionate ([11C]PMP). Synapse 1996;22 (2):123-31. 32. Namba H, et al. In vivo measurement of acetylcholinesterase activity in the brain with a radioactive acetylcholine analog. Brain Res 1994;667(2):278-82. 33. Snyder SE, et al. Synthesis of 1-[11C]methylpiperidin-4-yl propionate ([11C]PMP) for in vivo measurements of acetylcholinesterase activity. Nucl Med Biol 1998;25 (8):751-4. 34. Savic I, et al. In-vivo demonstration of reduced benzodiazepine receptor binding in human epileptic foci. Lancet 1988;2(8616):863-6. 35. Richardson MP, et al. Benzodiazepine receptors in focal epilepsy with cortical dysgenesis: an 11C-flumazenil PET study. Ann Neurol 1996;40(2):188-98. 36. Juhasz C, et al. Neuroradiological assessment of brain structure and function and its implication in the pathogenesis of West syndrome. Brain Dev 2001;23(7):488-95. 37. Bressan RA, Pilowsky, LS. Imaging the glutamatergic system in vivo – relevance to schizophrenia. Eur J Nucl Med 2000;27(11):1723-31. 38. Burgess RC. Functional localization by neurophysiologic and neuroimaging techniques. J Clin Neurophysiol 1995;12(5):405. 39. George JS, et al. Mapping function in the human brain with magnetoencephalography, anatomical magnetic resonance imaging, and functional magnetic resonance imaging. J Clin Neurophysiol 1995;12(5):406-31. 40. Simpson GV, et al. Dynamic neuroimaging of brain function. J Clin Neurophysiol 1995;12(5):432-49. 41. Puce A. Comparative assessment of sensorimotor function using functional magnetic resonance imaging and electrophysiological methods. J Clin Neurophysiol 1995;12(5):450-9. 42. Puce A, et al. Functional magnetic resonance imaging of sensory and motor cortex: comparison with electrophysiological localization. J Neurosurg 1995;83(2):262-70. 43. Fried I, et al. Functional MR and PET imaging of rolandic and visual cortices for neurosurgical planning. J Neurosurg 1995;83(5):854-61. 44. Detre JA, et al. Localization of subclinical ictal activity by functional magnetic resonance imaging: correlation with invasive monitoring. Ann Neurol 1995;38(4):618-24.
Diagnostic PET in image guided neurosurgery
45. Ives J.R, et al. Monitoring the patient’s EEG during echo planar MRI. Electroencephalogr Clin Neurophysiol 1993;87(6):417-20. 46. Atlas S.W, et al. Functional magnetic resonance imaging of regional brain activity in patients with intracerebral gliomas: findings and implications for clinical management. Neurosurgery 1996;38(2):329-38. 47. Archip N, et al. Non-rigid alignment of pre-operative MRI, fMRI, and DT-MRI with intra-operative MRI for enhanced visualization and navigation in image-guided neurosurgery. Neuroimage 2007;35(2):609-24. 48. Jones T. The role of positron emission tomography within the spectrum of medical imaging. Eur J Nucl Med 1996;23(2):207-11. 49. Jones T. The imaging science of positron emission tomography. Eur J Nucl Med 1996;23(7):807-13. 50. Jahanshahi M, et al. Self-initiated versus externally triggered movements. I. An investigation using measurement of regional cerebral blood flow with PET and movementrelated potentials in normal and Parkinson’s disease subjects. Brain 1995;118 (Pt 4):913-33. 51. Jenkins IH, et al. Impaired activation of the supplementary motor area in Parkinson’s disease is reversed when akinesia is treated with apomorphine. Ann Neurol 1992;32(6):749-57. 52. Dagher A, Nagano-Saito A. Functional and anatomical magnetic resonance imaging in Parkinson’s disease. Mol Imaging Biol 2007;9(4):234-42. 53. Grafton ST. Contributions of functional imaging to understanding parkinsonian symptoms. Curr Opin Neurobiol 2004;14(6):715-9. 54. Ceballos-Baumann AO. Functional imaging in Parkinson’s disease: activation studies with PET, fMRI and SPECT. J Neurol 2003;250 Suppl 1:I15-I23. 55. van Eimeren T, Siebner HR. An update on functional neuroimaging of parkinsonism and dystonia. Curr Opin Neurol 2006;19(4):412-9. 56. Hughes AJ, et al. Accuracy of clinical diagnosis of idiopathic Parkinson’s disease: a clinico-pathological study of 100 cases. J Neurol Neurosurg Psychiatry 1992;55 (3):181-4. 57. Heiss WD, Hilker R. The sensitivity of 18-fluorodopa positron emission tomography and magnetic resonance imaging in Parkinson’s disease. Eur J Neurol 2004;11(1):5-12. 58. Leenders KL, et al. Brain dopamine metabolism in patients with Parkinson’s disease measured with positron emission tomography. J Neurol Neurosurg Psychiatry 1986;49(8):853-60. 59. Nahmias C, et al. Striatal dopamine distribution in parkinsonian patients during life. J Neurol Sci 1985;69(3):223-30. 60. Damier P, et al. The substantia nigra of the human brain. II. Patterns of loss of dopamine-containing neurons in Parkinson’s disease. Brain 1999;122 (Pt 8):1437-48.
22
61. Garnett ES, et al. A rostrocaudal gradient for aromatic acid decarboxylase in the human striatum. Can J Neurol Sci 1987;14 Suppl 3:444-7. 62. Daniel SE. de Bruin VM, Lees AJ. The clinical and pathological spectrum of Steele-Richardson-Olszewski syndrome (progressive supranuclear palsy): a reappraisal. Brain 1995;118(Pt 3):759-70. 63. Ghaemi, M, et al. Differentiating multiple system atrophy from Parkinson’s disease: contribution of striatal and midbrain MRI volumetry and multi-tracer PET imaging. J Neurol Neurosurg Psychiatry 2002;73 (5):517-23. 64. Frost JJ, et al. Positron emission tomographic imaging of the dopamine transporter with 11C-WIN 35,428 reveals marked declines in mild Parkinson’s disease. Ann Neurol 1993;34(3):423-31. 65. Brooks DJ. Diagnosis and management of atypical parkinsonian syndromes. J Neurol Neurosurg Psychiatry 2002;72(90001):10i-16i. 66. Antonini A, et al. Complementary PET studies of striatal neuronal function in the differential diagnosis between multiple system atrophy and Parkinson’s disease. Brain 1997;120(Pt 12):2187-95. 67. Otsuka M, et al. Differentiating between multiple system atrophy and Parkinson’s disease by positron emission tomography with 18F-dopa and 18F-FDG. Ann Nucl Med 1997;11(3):251-7. 68. Antonini A, et al. Differential diagnosis of parkinsonism with [18F]fluorodeoxyglucose and PET. Mov Disord 1998;13(2):268-74. 69. Laureys S, et al. Fluorodopa uptake and glucose metabolism in early stages of corticobasal degeneration. J Neurol 1999;246(12):1151-8. 70. Booij J, et al. Imaging of the dopaminergic neurotransmission system using single-photon emission tomography and positron emission tomography in patients with parkinsonism. Eur J Nucl Med 1999;26(2):171-82. 71. Piccini P, Whone A. Functional brain imaging in the differential diagnosis of Parkinson’s disease. Lancet Neurol 2004;3(5):284-90. 72. Herholz K. PET studies in dementia. Ann Nucl Med 2003;17(2):79-89. 73. Silverman DH. Brain 18F-FDG PET in the diagnosis of neurodegenerative dementias: comparison with perfusion SPECT and with clinical evaluations lacking nuclear imaging. J Nucl Med 2004;45(4):594-607. 74. Devous MD Sr. Functional brain imaging in the dementias: role in early detection, differential diagnosis, and longitudinal studies. Eur J Nucl Med Mol Imaging 2002;29(12):1685-96. 75. Van Heertum RL, Tikofsky RS. Positron emission tomography and single-photon emission computed tomography brain imaging in the evaluation of dementia. Semin Nucl Med 2003;33(1):77-85.
321
322
22
Diagnostic PET in image guided neurosurgery
76. Ishii K, Minoshima S. PET is better than perfusion SPECT for early diagnosis of Alzheimer’s disease. Eur J Nucl Med Mol Imaging 2005;32(12):1463-5. 77. Jagust WJ, et al. The cortical topography of temporal lobe hypometabolism in early Alzheimer’s disease. Brain Res 1993;629(2):189-98. 78. Fukuyama H, et al. Altered cerebral energy metabolism in Alzheimer’s disease: a PET study. J Nucl Med 1994;35(1):1-6. 79. Minoshima S, et al. Metabolic reduction in the posterior cingulate cortex in very early Alzheimer’s disease. Ann Neurol 1997;42(1):85-94. 80. Albin RL, et al. Fluoro-deoxyglucose positron emission tomography in diffuse Lewy body disease. Neurology 1996;47(2):462-6. 81. Imamura T, et al. Regional cerebral glucose metabolism in dementia with Lewy bodies and Alzheimer’s disease: a comparative study using positron emission tomography. Neurosci Lett 1997;235(1–2):49-52. 82. Ishii K, et al. Regional cerebral glucose metabolism in dementia with Lewy bodies and Alzheimer’s disease. Neurology 1998;51(1):125-30. 83. Higuchi M, et al. Glucose hypometabolism and neuropathological correlates in brains of dementia with Lewy bodies. Exp Neurol 2000;162(2):247-56. 84. Friedland RP, et al. Functional imaging, the frontal lobes, and dementia. Dementia 1993;4(3–4):192-203. 85. Herholz K. FDG PET and differential diagnosis of dementia. Alzheimer Dis Assoc Disord 1995;9(1):6-16. 86. McKeith IG, et al. Diagnosis and management of dementia with Lewy bodies: third report of the DLB Consortium. Neurology 2005;65(12):1863-72. 87. Rowe CC, et al. Imaging b-amyloid burden in aging and dementia. Neurology 2007;68(20):1718-25. 88. Engler H, et al. In vivo amyloid imaging with PET in frontotemporal dementia. Eur J Nucl Med Mol Imaging 2008;35(1):100-6. 89. Di Chiro G, Brooks RA. PET-FDG of untreated and treated cerebral gliomas. J Nucl Med 1988;29(3):421-3. 90. Delbeke D, et al. Optimal cutoff levels of F-18 fluorodeoxyglucose uptake in the differentiation of low-grade from high-grade brain tumors with PET. Radiology 1995;195(1):47-52. 91. Ogawa T, et al. Quantitative evaluation of neutral amino acid transport in cerebral gliomas using positron emission tomography and fluorine-18 fluorophenylalanine. Eur J Nucl Med 1996;23(8):889-95. 92. Kaschten B, et al. Preoperative evaluation of 54 gliomas by PET with fluorine-18-fluorodeoxyglucose and/or carbon-11-methionine. J Nucl Med 1998;39(5):778-85. 93. Chung JK, et al. Usefulness of 11C-methionine PET in the evaluation of brain lesions that are hypo- or isometabolic on 18F-FDG PET. Eur J Nucl Med Mol Imaging 2002;29(2):176-82.
94. Pauleit D, et al. O-(2-[18F]fluoroethyl)-L-tyrosine PET combined with MRI improves the diagnostic assessment of cerebral gliomas. Brain 2005;128(Pt 3):678-87. 95. Derlon JM, et al. [11C]L-methionine uptake in gliomas. Neurosurgery 1989;25(5):720-8. 96. Ogawa T, et al. Cerebral glioma: evaluation with methionine PET. Radiology 1993;186(1):45-53. 97. Nuutinen, J., et al. Radiotherapy treatment planning and long-term follow-up with [11C]methionine PET in patients with low-grade astrocytoma. Int J Radiat Oncol Biol Phys 2000;48(1):43-52. 98. De Witte O, et al. Positron emission tomography with injection of methionine as a prognostic factor in glioma. J Neurosurg 2001;95(5):746-50. 99. Ribom D, et al. Positron emission tomography 11C-methionine and survival in patients with low-grade gliomas. Cancer 2001;92(6):1541-9. 100. Levivier M, et al. Positron emission tomographyguided stereotactic brain biopsy. Neurosurgery 1992; 31(4):792-7; 101. Pirotte B, et al. PET in stereotactic conditions increases the diagnostic yield of brain biopsy. Stereotact Funct Neurosurg 1994;63(1–4):144-9. 102. Pirotte B, et al. Use of positron emission tomography (PET) in stereotactic conditions for brain biopsy. Acta Neurochir (Wien) 1995;134(1–2):79-82. 103. Levivier M, et al. Diagnostic yield of stereotactic brain biopsy guided by positron emission tomography with [18F]fluorodeoxyglucose. J Neurosurg 1995; 82(3):445-52. 104. Goldman S, et al. Regional methionine and glucose uptake in high-grade gliomas: a comparative study on PET-guided stereotactic biopsy. J Nucl Med 1997; 38(9):1459-62. 105. Pirotte B, et al. Combined positron emission tomography and magnetic resonance imaging for the planning of stereotactic brain biopsies in children: experience in 9 cases. Pediatr Neurosurg 2003;38(3):146-55. 106. Spence AM, Mankoff DA, Muzi M. Positron emission tomography imaging of brain tumors. Neuroimaging Clin N Am 2003;13(4):717-39. 107. Schaller B. Usefulness of positron emission tomography in diagnosis and treatment follow-up of brain tumors. Neurobiol Dis 2004;15(3):437-48. 108. Holzer T, et al. FDG-PET as a prognostic indicator in radiochemotherapy of glioblastoma. J Comput Assist Tomogr 1993;17(5):681-7. 109. Francavilla TL, et al. Positron emission tomography in the detection of malignant degeneration of low-grade gliomas. Neurosurgery 1989;24(1):1-5. 110. Ricci PE, et al. Differentiating recurrent tumor from radiation necrosis: time for re-evaluation of positron emission tomography? AJNR Am J Neuroradiol 1998;19(3):407-13.
Diagnostic PET in image guided neurosurgery
111. Tsuyuguchi N, et al. Methionine positron emission tomography for differentiation of recurrent brain tumor and radiation necrosis after stereotactic radiosurgery in malignant glioma. Ann Nucl Med 2004; 18(4):291-6. 112. Carne RP, et al. MRI-negative PET-positive temporal lobe epilepsy: a distinct surgically remediable syndrome. Brain 2004;127(10):2276-2285. 113. Casse R, et al. Positron emission tomography and epilepsy. Mol Imaging Biol 2002;4(5):338-51. 114. Juhasz C, et al. Glucose and [11C]flumazenil positron emission tomography abnormalities of thalamic nuclei in temporal lobe epilepsy. Neurology 1999; 53(9):2037-45. 115. Koepp M.J, et al. 11C-flumazenil PET in patients with refractory temporal lobe epilepsy and normal MRI. Neurology 2000;54(2):332-9. 116. O’Brien TJ, et al. The utility of a 3-dimensional, largefield-of-view, sodium iodide crystal-based PET scanner in the presurgical evaluation of partial epilepsy. J Nucl Med 2001;42(8):1158-65. 117. Kuhl DE, et al. Epileptic patterns of local cerebral metabolism and perfusion in humans determined by emission computed tomography of 18FDG and 13NH3. Ann Neurol 1980;8(4):348-60. 118. Ollenberger GP, et al. Assessment of the role of FDG PET in the diagnosis and management of children with refractory epilepsy. Eur J Nucl Med Mol Imaging 2005;32(11):1311-6. 119. Koepp MJ, Woermann FG. Imaging structure and function in refractory focal epilepsy. Lancet Neurol 2005;4(1):42-53. 120. Savic I, Ingvar M, Stone-Elander S. Comparison of [11C]flumazenil and [18F]FDG as PET markers of epileptic foci. J Neurol Neurosurg Psychiatry 1993;56 (6):615-21. 121. Arnold S, et al. Reduction of benzodiazepine receptor binding is related to the seizure onset zone in extratemporal focal cortical dysplasia. Epilepsia 2000; 41(7):818-24. 122. Muzik O, et al. Intracranial EEG versus flumazenil and glucose PET in children with extratemporal lobe epilepsy. Neurology 2000;54(1):171-9. 123. Yousry TA, et al. Topography of the cortical motor hand area: prospective study with functional MR imaging and direct motor mapping at surgery. Radiology 1995; 195(1):23-9. 124. Majos, A, et al. Cortical mapping by functional magnetic resonance imaging in patients with brain tumors. Eur Radiol 2005;15(6):1148-58. 125. Roessler, K, et al. Evaluation of preoperative high magnetic field motor functional MRI (3 Tesla) in glioma patients by navigated electrocortical stimulation and postoperative outcome. J Neurol Neurosurg Psychiatry 2005;76(8):1152-7.
22
126. Kloppel S, Buchel C. Alternatives to the Wada test: a critical view of functional magnetic resonance imaging in preoperative use. Curr Opin Neurol 2005;18(4):418-23. 127. Kaplan AM, et al. Functional brain mapping using positron emission tomography scanning in preoperative neurosurgical planning for pediatric brain tumors. J Neurosurg 1999;91(5):797-803. 128. Meyer PT, et al. Preoperative mapping of cortical language areas in adult brain tumour patients using PET and individual non-normalised SPM analyses. Eur J Nucl Med Mol Imaging 2003;30(7):951-60. 129. Vanlancker-Sidtis D. McIntosh AR, Grafton S. PET activation studies comparing two speech tasks widely used in surgical mapping. Brain Lang 2003;85 (2):245-61. 130. Limousin P, et al. Changes in cerebral activity pattern due to subthalamic nucleus or internal pallidum stimulation in Parkinson’s disease. Ann Neurol 1997;42 (3):283-91. 131. Ceballos-Baumann AO, et al. A positron emission tomographic study of subthalamic nucleus stimulation in Parkinson disease: enhanced movement-related activity of motor-association cortex and decreased motor cortex resting activity. Arch Neurol 1999;56(8):997-1003. 132. Strafella AP, Dagher A, Sadikot AF. Cerebral blood flow changes induced by subthalamic stimulation in Parkinson’s disease. Neurology 2003;60(6):1039-42. 133. Davis KD, et al. Activation of the anterior cingulate cortex by thalamic stimulation in patients with chronic pain: a positron emission tomography study. J Neurosurg 2000;92(1):64-9. 134. Strafella AP, et al. Subdural motor cortex stimulation in Parkinson’s disease does not modify movement-related rCBF pattern. Mov Disord 2007;22(14):2113-6. 135. Roberto Cilia ALFVESGPAA. Extradural motor cortex stimulation in Parkinson’s disease. Mov Disord 2007; 22(1):111-114. 136. Mayberg HS, et al. Deep brain stimulation for treatment-resistant depression. Neuron 2005;45(5):651-60. 137. May A, et al. Hypothalamic activation in cluster headache attacks. Lancet 1998;352(9124):275-8. 138. Leone M, Franzini A, Bussone G. Stereotactic stimulation of posterior hypothalamic gray matter in a patient with intractable cluster headache. N Engl J Med 2001;345(19):1428-9. 139. Leone M, et al. Hypothalamic deep brain stimulation for intractable chronic cluster headache: a 3-year follow-up. Neurol Sci 2003;24 Suppl 2:S143-S145. 140. Franzini A, et al. Stimulation of the posterior hypothalamus for treatment of chronic intractable cluster headaches: first reported series. Neurosurgery 2003;52 (5):1095-9; 141. Herholz K, Coope D, Jackson A. Metabolic and molecular imaging in neuro-oncology. Lancet Neurol 2007; 6(8):711-24.
323
20 Functional MRI in Image Guided Neurosurgery T. Sankar . G. R. Cosgrove
Introduction The principal goal of neurosurgical procedures on the cerebral cortex is to accurately localize and maximally resect lesions or abnormal tissue. An intimate knowledge of eloquent cortical regions which may be damaged during surgery is therefore a prerequisite for safe cortical neurosurgery. Identifying functionally important areas of the cortex during surgery can be aided by a priori knowledge of cortical functional organization, but is subject to inaccuracy caused by individual variations in anatomy and displacement of normal structures by pathological lesions. Consequently, several cortical mapping techniques have been established to facilitate safe and effective cortical resections. The gold standard for cortical mapping is direct electrocortical stimulation (ECS) of the exposed cortex at surgery [1–3]. ECS is limited, however, because it is an invasive technique which can usually only be performed in an awake patient [3]. Awake craniotomy is not always feasible and presents unique anesthetic and analgesic challenges. In addition, as a surface mapping technique, ECS cannot provide information about function buried within the sulcal depths which represent a substantial proportion of the cortical volume [4]. Perhaps most importantly, ECS is a purely intra-operative technique which cannot be used preoperatively to guide surgical planning [3]. The past two decades have seen the development and implementation of several noninvasive cortical mapping techniques, including positron emission tomography (PET), #
Springer-Verlag Berlin/Heidelberg 2009
magnetoencephalography (MEG), and transcranial magnetic stimulation (TMS) [3]. Functional magnetic resonance imaging (fMRI), which assesses regional variations in cerebral blood flow during various tasks of cortical activation, is one such technique. fMRI makes use of readily available conventional MRI hardware to produce maps of functional cortical activation, which can be superimposed onto anatomical images of the brain. The resulting structural and functional model of the brain – which is obtained prior to surgery – can then be used both for preoperative surgical planning and intra-operative decisionmaking. Recent clinical experience in patients with lesions involving eloquent cortex has demonstrated the utility of fMRI in the neurosurgical armamentarium.
Principles of fMRI Acquisition Physiological Basis of fMRI Basic neurophysiologic studies have long shown that cerebral activity is tightly coupled to cerebral metabolism, which is in turn related to cerebral blood flow [5]. As a result, regionalhemodynamic measurements can serve as a surrogate marker of neuronal activity. This principle forms the basis of metabolic/hemodynamic mapping techniques such as PET [6] and fMRI [7]. In 1991, Belliveau et al. [8] first reported on the use of this principle in MR imaging. After intravenous bolus injection of a paramagnetic contrast agent, they rapidly acquired MR images and demonstrated signal change in the visual
288
20
Functional MRI in image guided neurosurgery
cortex due to activation with photic stimulation. This signal change was presumed to be due to increased concentration of contrast agent accompanying increased regional blood flow to the visual cortex, and became known as the ‘‘contrast bolus tracking’’ technique. fMRI based on contrast bolus tracking is seldom used today. The most commonly used fMRI acquisition method measures blood oxygen level-dependent (BOLD) changes in the magnetic resonance (MR) signal, first described by Ogawa et al. [9,10]. BOLD imaging has become the fMRI acquisition technique of choice because it does not require the administration of an exogenous contrast agent. An increase in neuronal activity in a particular region of the brain – caused by, say, performing a particular task – initially leads to a transient increase in oxygen extraction from the blood supply to that region. This is quickly followed by compensatory vasodilation, which produces a net increase in the overall concentration of oxy-hemoglobin (oxy-Hb) relative to deoxy-hemoglobin (deoxy-Hb) [7,11]. The physiologic mechanisms underpinning this vasodilatory response are not yet fully understood, but various locally-generated chemical mediators have been implicated, including serotonin, acetylcholine, neuroactive peptides, and nitric oxide produced by cortical neurons [12]. Since the iron in deoxy-Hb is paramagnetic, it reduces T2 MR signal through spin-dephasing effects. The relative decrease in deoxy-Hb concentration in a region of increased neuronal activity, then, produces a corresponding increase in T2 signal, forming the basis of BOLD imaging [9]. This T2 signal change is primarily thought to come from the microvasculature consisting of the capillary system and small venules, and is of low magnitude (on the order of 0.5–5%) [13]. Studies in both humans and primates have demonstrated that the BOLD signal is proportional to the neuronal firing rate, though the latency of the observed change is on the order of several seconds [14,15].
Technological Requirements for fMRI Acquisition Given the small magnitude of changes in the BOLD signal accompanying neuronal activation, signal-to-noise ratio is of critical importance during fMRI acquisition. Initial fMRI studies were carried out using high field strength magnets (2–4 T) using standard gradient-echo acquisitions. Subsequent improvements in gradient coil shielding and surface coil design allowed conventional 1.5 T scanners to be used, but at the expense of prolonged and frequently impractical acquisition times [16]. The recent development and application of echo-planar imaging gradients has revolutionized fMRI acquisition, allowing entire volumes to be imaged in seconds, with improved spatial resolution and signal-tonoise [17]. Most current fMRI studies are carried out on conventional 1.5 T MRI scanners with modifications to allow for echo-planar imaging, and involve multiple iterations of image acquisition and task repetition [7]. Following acquisition, fMRI data must be processed and displayed in a format suitable for clinical use. Typically, a conventional desktop computer workstation equipped with imageprocessing software is used to average multiple fMRI acquisitions. Next, voxel-by-voxel analysis is performed to identify those cortical regions which have an increased BOLD signal relative to baseline during performed tasks. The threshold used to classify increased signal as true cortical activation is arbitrary; normally, statistical methods are used to select a threshold coefficient 2–3 standard deviations from baseline. fMRI images are then co-registered, fused, and volume rendered with anatomical MRI images using both surface- and volume-based matching techniques [18]. Images of regions of cortical activation can then be superimposed onto anatomical images for display purposes or incorporated into neuronavigation systems for intra operative surgical guidance [19] (> Figure 20‐1).
Functional MRI in image guided neurosurgery
20
. Figure 20‐1 fMRI of sensorimotor hand and foot activation superimposed on axial T2 weighted images demonstrating a tumor in the left posterior superior frontal gyrus just anterior to the motor strip. Note that the right foot activation is less intense and pushed posteriorly by the tumor
Task Design and Selection For neurosurgical mapping purposes, fMRI data are generated using a block design paradigm. Simply, block design involves multiple trials of a task or stimulus presentation – a so-called ‘‘task block’’ – alternating with multiple trials of a control task (‘‘control block’’) [3,20,21]. The resulting BOLD signals from task and control blocks are compared, allowing for the detection of subtle changes accompanying task-related cerebral activation. There is no standardization of the length of each task block or the number of alternating cycles between task and control blocks. Typically, task blocks last from 15 to 30 s; this ensures that they are of sufficient duration to allow the occurrence of hemodynamic changes underlying the fMRI signal [7,22,23]. Cycling multiple trials of a task block with control blocks improves signal-tonoise ratio and the overall statistical accuracy of inferred cortical activations [24] (> Figure 20‐2). Several task variants have been developed to assess cortical activation in different functional regions of the brain, including the sensorimotor cortex, the speech areas of Broca and Wernicke, and the primary visual cortex. The classical motor
activation task is finger-thumb tapping, though repetitive clenching of the fingers or sponge squeezing have also been employed. It is important to realize that this motor task includes stimulation of both cutaneous and proprioceptive sensory inputs and therefore is more accurately a combined sensorimotor task than a pure motor task. Sensory stimulation is frequently achieved by brushing of the palmar surface of the hand [7]. Language is assessed by picture naming, verb generation, or having the patient classify related nouns into categories [25–27]. Because of the motion artifact created by speech, most centers favor of using silent instead of vocalized speech for language mapping [27]. Visual tasks frequently involve intermittent photic stimulation with patterned stimuli presented via binoculars or on a screen [22].
Application of fMRI to Cortical Resections Illustrative Cases Case # 1 – This 17-year-old male had experienced intractable seizures for nearly 10 years which
289
290
20
Functional MRI in image guided neurosurgery
. Figure 20‐2 Demonstration of the signal changes on fMRI during activation task versus resting state cycle
were characterized by the sudden onset of dystonic posturing and elevation of his right upper extremity followed by rapid secondary generalization. Video EEG recordings suggested a left frontal or bifrontal onset and MRI demonstrated a non-enhancing lesion in the left frontal parasaggital cortex. Functional MRI was carried out for sensorimotor testing of the right hand and the results mapped to the patients structural MRI. At surgery, direct ECS of the exposed cortex confirmed the location of primary sensorimotor cortex of the hand region in the pre and post central gyri. A 4 contact depth electrode was then placed in the region of the presumed supplementary motor area (SMA) and stimulation through the deepest and middle contacts elicited tonic contraction and elevation of the entire right upper extremity consistent with a SMA type seizure (> Figure 20‐3). Careful resection of the cortex just anterior to the SMA revealed a ganglioglioma and the patient has remained seizure free for nearly a decade. Case # 2 – This 24-year-old right-handed man had noticed episodes of sudden onset of a ‘‘tickling’’ in his throat followed by reflexive coughing and some slight difficulty speaking within the past year. These episodes would last for just a few minutes and occur irregularly. After
one such episode, he experienced a generalized tonic conic convulsion and he was taken to a local emergency room where an enhancing lesion was discovered in his dominant subcentral gyrus. A functional MRI was performed for sensorimotor tongue activation and for language mapping using a visual verb generation task (> Figure 20‐4a). At surgery under local anesthesia, cortical mapping of tongue sensorimotor and language areas corresponded closely to that predicted by fMRI. The location of the tumor could be easily determined by the gyral anatomy and cortical vessels and was resected without deficit (> Figure 20‐4b). Pathology demonstrated an xanthoastrocytoma and the patient has been seizure free since surgery. Case #3 – This 47-year-old bilingual, righthanded Greek woman had previously undergone resection of an oligodendroglioma 5 years before presenting with a recurrent tumor. Functional MRI was performed using visual and auditory verb generation tasks in both English and Greek (> Figure 20‐5a). These results were mapped to a structural MRI of her brain and at surgery under local anesthesia correlated closely with language mapping in both languages with ECS enabling safe resection of the recurrent tumor (> Figure 20‐5b).
Functional MRI in image guided neurosurgery
20
. Figure 20‐3 (a) Axial and (b) surface rendered MRI images with superimposed fMRI activations demonstrating a tumor in the superior left frontal gyrus and both primary and supplementary motor area activations during hand movement (c) cropped close up view of cortical surface as predicted by fMRI as compared to (d) intraoperative view of cortical surface with green tags indicating motor responses in the hand, white tags motor responses in the proximal upper extremity and red tag the area where depth electrode stimulation elicited SMA type seizure. Asterisks are for localization purposes but gyral anatomy and cortical veins are also useful
Discussion Sensitivity and Interpretation of fMRI Results According to Kim and Singh [7], the tasks used for fMRI data generation can be considered either simple or complex, based on the area of cortex they consistently activate. For example, finger-thumb tapping is considered a simple motor task, which typically activates a welldefined focal area of motor cortex. However, even so called simple motor tasks are actually a combination of both motor and sensory activations because of the activation of cutaneous and proprioceptive input. Complex tasks such as
language function require simultaneous activation of several regions of the brain, each with multiple sensory inputs and functional outputs. However, even simple tasks may require supplementary areas whose interaction with the primary cortical activation area is poorly understood. Furthermore, in order to increase the overall sensitivity of fMRI, some centers assess cortical activation during tasks completed in both active and passive modes (e.g., spoken and silent speech), because redundancy is thought to exist in functional areas of cortical activation for related tasks [7]. As a consequence of these factors, preoperative fMRI imaging maps frequently demonstrate a greater spread and area of activation than do
291
292
20
Functional MRI in image guided neurosurgery
. Figure 20‐4 (a) Axial MRI images of enhancing nodule in left central region and the surface rendered fMRI of the whole brain with tumor in yellow, tongue activation in pink and verb generation task activation in green (b) images of the cortical surface at surgery and that predicted by fMRI cropped for comparison. Note the excellent correlation between ECS and fMRI. The subcortical location of the tumor is easily deduced from the surface anatomy
invasive intraoperative mapping techniques such as ECS, and it is occasionally unclear which cortical regions are essential to the completion of a given task. Optimum interpretation of preoperative fMRI is predicated, then, on a reasoned consideration of the balance between increased sensitivity generated by task complexity or redundancy on the one hand, and a priori knowledge of functional neuroanatomy on the other. Keeping these principles in mind, the overall sensitivity of fMRI is usually reported as excellent. In a widely cited report, Hirsch et al. [28] were able to demonstrate 100% sensitivity for identifying language cortex in the superior temporal gyrus, motor
function in the precentral gyrus, and visual function in occipital cortex. They also showed 93% sensitivity for Broca’s area.
Comparison of Functional Cortical Localization Between fMRI and ECS Several studies have attempted to validate fMRI data by comparing it against standard cortical localization by ECS. The definition of adequate correlation between the two techniques varies from study to study, but most authors consider
Functional MRI in image guided neurosurgery
20
. Figure 20‐5 (a) fMRI images of cortical activation during language tasks in both English and Greek. Note that the auditory verb generation task activates primary auditory cortex in addition to anterior language areas (b) images of the cortical surface cropped for comparison to that predicted by fMRI. Both the structural and functional correlation is excellent
correlation to be successful when both techniques demonstrate functional localization to within some arbitrary distance of one another, usually 10 or 20 mm. Most existing studies have focused on sensorimotor mapping, and have almost uniformly reported excellent correlation between the two techniques [28–39]. In particular, fMRI appears to be particularly successful at correctly identifying the central sulcus (CS): Majos et al. [35] reported a 98% success rate in mapping
the CS, and Lehericy et al. [32] showed a similar concordance rate of 92% in a large study of 60 patients. Furthermore, brain distortion as a result of edema and deformation due to tumors or other pathological entities does not seem to impact this concordance. Hirsch et al. [28] reported that overall localization of the CS was possible in 97% of patients being surgically treated for lesions of the central region. Excellent sensorimotor functional correlation has also been observed in the hand
293
294
20
Functional MRI in image guided neurosurgery
area located within the pli de passage moyen (PPM) of the precentral gyrus [37], even in the presence of infiltrative glioma [36]. Interestingly, Boling et al. [40] recently used fMRI to demonstrate that there may be a distinct whole hand sensory and motor area within the PPM. To date, agreement between fMRI and ECS has been less robust for language localization. Some studies have indeed reported excellent concordance with word-generation tasks for mapping Broca’s area [28,29,41], as well as for language function in general using a battery of tasks [42]. The most recent and systematic study directly comparing language localization by both methods was published by Roux et al. [43], who examined a series of 14 right-handed patients – all with tumors in the left hemisphere – with preoperative fMRI obtained during naming and verb generation tasks. During operative tumor resection, they also assessed 426 distinct cortical sites by ECS across the patient cohort. In total, 22 sites were found to be ‘‘positive’’ during naming or verb generation (i.e., elicited speech arrest, anomia, hesitations, paraphasic errors, or delayed responses when stimulated). Of these, only 13 (59%) were concordant (i.e., within 1 cm of) with fMRI signals. The authors concluded from these data that fMRI itself is insufficient to make critical surgical decisions in essential cortical language areas. However, the study suffered from some methodological limitations which weakened its dismissal of fMRI for language mapping. Specifically, the study did not use a silent speech generation task during fMRI acquisition, thereby ignoring the principle of task redundancy and increasing susceptibility to motion artifact. In addition, as noted by McKhann II and Hirsch [44] in their comment on the study, language disruption sites within dominant face motor cortex were included among the 22 positive stimulation sites. Normally, fMRI obtained with silent speech tasks does not activate motor cortex, suggesting that the methods may have been inherently biased against the accuracy of fMRI for language. That being
said, recent work has suggested that the weakness of silent speech tasks may lie in their inability to activate contributory cortical sites within the precentral gyrus [45]. Clearly, further work needs to be done to elucidate the optimal testing paradigms which generate the most accurate preoperative fMRI maps of cortical language localization across a varied population of patients. Such work may solidify the value of fMRI in guiding cortical resections in language-specific regions. Currently, all fMRI results for language should be interpreted with caution and must still be validated intra-operatively with ECS in the awake patient.
Advantages of fMRI as a Cortical Mapping Technique in Neurosurgery fMRI has as its principal advantage the fact that it is non-invasive, which significantly increases its safety and repeatability compared to direct ECS. At best, intraoperative testing requires a larger craniotomy, increases operative time, and is associated with significant patient stress and discomfort accompanying an awake procedure under local anesthetic. At worst, ECS may be associated with intraoperative seizures which may force premature termination of mapping or require the procedure to be aborted altogether [3,46]. Furthermore, awake craniotomy generally requires dedicated neuroanesthesia support and sufficient experience which may preclude its use in smaller centers. fMRI has good spatial resolution, estimated to be on the order of 2–5 mm [47]; this is well within the 10 mm spread of electrical current to adjacent cortical areas when ECS is used [7]. fMRI also has the advantage of generating data over the entire extent of the brain, producing global maps of cortical activation. This data can be important even if we concede that the accuracy of fMRI for precise functional localization may
Functional MRI in image guided neurosurgery
be uncertain [22]. Invasive mapping is typically performed if a proposed resection is believed, preoperatively, to place eloquent cortex at risk. fMRI can hence be used prior to surgery to identify patients in whom invasive intraoperative mapping will be required, avoiding in others the significant risks and increased operative time required by such mapping [48]. This complementary application of fMRI confirms its worth even as work continues to establish it as a cortical mapping technique independent of other imaging or stimulatory modalities. Perhaps the most appealing feature of fMRI is that it can demonstrate cortical activation within the sulcal depths, which have been reported to represent as much as two-thirds of functionally eloquent cortex [4]. This is a clear advantage over ECS, which can only assess surface cortex. The ability of fMRI to guide preservation of cortical grey matter in the deep sulci during surgery may be of particular relevance to avoiding damage to cognitive and memory functions, whose underlying neuroanatomical basis is poorly understood [3,7].
Disadvantages of fMRI in Neurosurgical Applications While fMRI can localize distinct cortical activity, it does not localize underlying white matter tracts or the important connections between functional areas. The risks of incurring neurological deficits are often greatest during subcortical resections in the white matter and therefore careful consideration of the location of these pathways must always be paramount. Only surgery performed under local anesthesia with constant testing of the patient’s function during resection can prevent deficits and therefore fMRI will never completely replace standard ECS. Most shortcomings of fMRI are technical. As with any preoperative imaging modality, fMRI is susceptible to inaccuracies because of the brain
20
shift accompanying craniotomy, dural opening, and CSF drainage [49]. Sensitivity to motion artifact, which is principally caused by a patient’s inability to hold the head still, or occasionally by involuntary movements related to breathing or heartbeat, can also be problematic [3,22]. As mentioned previously, during tasks involving overt speech, word vocalization may alone produce sufficient motion to degrade image acquisition and quality. The rapid image acquisition achieved with echoplanar technology partially reduces the impact of motion on fMRI studies. Firmly securing the head and providing a visual fixation crosshair may also help [28,42], but occasionally – particularly in cognitively impaired or uncooperative patients – an fMRI examination simply cannot be completed. Venous effects may also confuse fMRI interpretation. While fMRI is based on the assumption that the BOLD signal is confined to the microvasculature, in reality there is signal spread to veins draining blood away from activated cortical tissue and which may be located several millimeters from the actual focus of neuronal activation. This signal in turn can produce images depicting false-positive activation in these adjacent regions [50]. Obtaining highquality three dimensional anatomic MRI images showing the positions of draining veins can assist the neurosurgeon in anticipating areas of falsepositive signal [4]. In addition, venous signal appears to be less of an issue at higher magnetic field strength [50]. Related to aberrant venous signal is the theoretical possibility that various neurosurgical pathologies may alter normal blood flow in such a way as to disrupt fMRI interpretation. One such example of this phenomenon is the effect of tumor angiogenesis, whose impact on the BOLD signal is as yet unknown. Vascular steal related to arteriovenous malformations near eloquent cortex has similarly been cited as another potential pitfall [7]. Even atherosclerotic disease may potentially impact regional variations
295
296
20
Functional MRI in image guided neurosurgery
in blood flow from individual to individual, and within the same individual at different times. Further study is necessary to assess the BOLD signal in these clinical situations to appropriately interpret fMRI data prior to surgical resections. As mentioned previously, perhaps the greatest current limitation of fMRI for guiding cortical resections relates to the – as yet – uncertain neuroscientific implications of the data it generates. At present, we cannot use fMRI to accurately distinguish between essential functional cortical areas and those which are participatory but nonessential to function [7]. This is certainly true for language, but even more so for memory, whose neuroanatomic basis is still incompletely understood [51]. Fortunately, fMRI data are being analyzed in the setting of several neuropsychological and cognitive functions, and the conclusions reached from this basic work will likely make their way into neurosurgical applications. Additionally, the ongoing development of increasingly standardized fMRI task protocols should help to improve the interpretation and reproducibility of cortical mapping data between different patients and different centers.
Conclusions fMRI is a powerful non-invasive neuroimaging technique that can create a unique structural and functional model of an individual patient’s brain. The widespread availability of MRI scanners allows fMRI exams to be acquired in appropriate patients at most modern neurosurgical centers. fMRI data are generated preoperatively and can be used to determine the feasibility of open cortical resections by delineating those areas of eloquent cortex which may be at risk in a planned surgical procedure. Such data can be of critical importance in assessing surgical risk, optimizing surgical exposure, as well as guiding surgical approaches to and the resection of cortical
lesions. fMRI has several advantages over ECS, including its non-invasiveness, its higher spatial resolution, and its ability to assess function within the sulcal depths. fMRI acquisitions can be impaired, however, by technical factors such as motion artifact, infiltrative tumors and venous effects, and fMRI data are not perfectly correlated with ECS, particularly for language localization. Future investigations will likely establish the appropriate task paradigms required to maximize the accuracy of functional cortical maps generated by fMRI. Meanwhile, fMRI has already come into widespread use in neurosurgery, either independently or as an adjunct to ECS. The future of cortical neurosurgery is one in which all patients will routinely undergo fMRI as part of their presurgical evaluation.
References 1. Penfield W. The cerebral cortex and consciousness, in the Harvey lectures. Baltimore: Williams & Wilkins Co.; 1937. p. 35-69. 2. Ojemann G, Ojemann J, Lettich E, Berger M. Cortical language localization in left, dominant hemisphere. An electrical stimulation mapping investigation in 117 patients. J Neurosurg 1989;71:31-326. 3. Tharin S, Golby A. Functional brain mapping and its applications to neurosurgery. Neurosurgery 2007;60(4 Suppl 2):185-201. 4. Cosgrove GR, Buchbinder BR, Jiang H. Functional magnetic resonance imaging for intracranial navigation. Neurosurg Clin N Am 1996;7:313-22. 5. Colebatch JG, Deiber M-P, Passingham RE, Friston KJ, Frackowiak KS. Regional cerebral blood flow during voluntary arm and hand movements in human subjects. J Neurophysiol 1991;65:1392-401. 6. Fox PT, Burton H, Raichle M. Mapping human somatosensory cortex with positron emission tomography. J Neurosurg 1997;67:34-43. 7. Kim PE, Singh M. Functional magnetic resonance imaging for brain mapping in neurosurgery. Neurosurg Focus 2003;15:1-7. 8. Belliveau JW, Kennedy DN, Jr, McKinstry RC, Buchbinder BR, Weisskoff RM, Cohen MS, et al. Functional mapping of the human visual cortex by magnetic resonance imaging. Science 1991;254:716-19. 9. Ogawa S, Lee TM, Kay AR, Tank DW. Brain magnetic resonance imaging with contrast dependent on blood oxygenation. Proc Natl Acad Sci USA 1990;87:9868-72.
Functional MRI in image guided neurosurgery
10. Ogawa S, Tank DW, Menon R, Ellermann JM, Kim SG, Merkle H, et al. Intrinsic signal changes accompanying sensory stimulation: functional brain mapping with magnetic resonance imaging. Proc Natl Acad Sci USA 1992;89:5951-5. 11. Fox PT, Raichle ME. Focal physiological uncoupling of cerebral blood flow and oxidative metabolism during somatosensory stimulation in human subjects. Proc Natl Acad Sci USA 1986;83:1140-4. 12. Iadecola C. Regulation of the cerebral microcirculation during neural activity: is nitric oxide the missing link? Trends Neurosci 1993;16:206-14. 13. Bandettini PA, Wong C, Hinks RS, Tifosky RS, Hyde JS. Time course EPI of human brain function during task activation. Magn Reson Med 1992;25:390-7. 14. Rees G, Friston K, Koch C. A direct quantitative relationship between the functional properties of human and macaque V5. Nat Neurosci 2000;3:716-23. 15. Heeger DJ, Huk AC, Geisler WS, Albrecht DG. Spikes versus BOLD: what does neuroimaging tell us about neuronal activity? Nat Neurosci 2000;3:631-3. 16. Connelly A, Jackson GD, Frackowiak RS, Belliveau JW, Vargha-Khadem F, Gadian DG. Functional mapping of activated human primary cortex with a clinical MR imaging system. Radiology 1993;188:125-30. 17. Stehling MK, Turner R, Mansfield P. Echo-planar imaging: magnetic resonance imaging in a fraction of a second. Science 1991;254:43-50. 18. Cosgrove GR, Buchbinder BR. Functional MRI for planning cortical resections. In: Alexander E, Maciunas R, editors. Advanced neurosurgical navigation. New York: Thieme; 1997. 19. Roux FE, Ibarrrola D, Tremoulet M, Lazorthes Y, Henry P, Sol JC, et al. Technical and methodological issues for integrating functional MRI in a neuronavigational system. Neurosurgery 2001;49:1145-57. 20. Gore JC. Principles and practice of functional MRI of the human brain. J Clin Invest 2003;112:4-9. 21. Detre JA. fMRI: applications in epilepsy. Epilepsia2004;45 Suppl 4:26-31. 22. Lee CC, Ward HA, Sharbrough FW, Meyer FB, Marsh WR, Raffel C, et al. Assessment of functional MR imaging in neurosurgical planning. Am J Neuroradiol 1999;20: 1511-19. 23. Wilkinson ID, Romanowski CA, Jellinek DA, Morris J, Griffiths PD. Motor functional MRI for pre-operative and intraoperative neurosurgical guidance. Br J Radiol 2003;76:98-103. 24. Elster AD, Burdette JH. Questions and answers in magnetic resonance imaging. 2nd ed. St. Louis: Mosby; 2001. p. 273-80. 25. Binder JR, Swanson SJ, Hammeke TA, Morris GL, Mueller WM, Fischer M, et al. Determination of language dominance using functional MRI: a comparison with the Wada test. Neurology 1996;46:978-84.
20
26. Herholz K, Reulen HJ, von Stockhausen HM, Thiel A, Ilmberger J, Kessler J, et al. Preoperative activation and intraoperative stimulation of language-related areas in patients with glioma. Neurosurgery 1997;41:1253-60. 27. Hinke RM, Yu X, Stillman AE, Kim SG, Merkle H, Salmi R, et al. Functional magnetic resonance imaging of Broca’s area during internal speech. Neuroreport 1993;4:675-8. 28. Hirsch J, Ruge MI, Kim KH, Correa DD, Victor JD, Relkin NR, et al. An integrated functional magnetic resonance imaging procedure for preoperative mapping of cortical areas associated with tactile, motor, language, and visual functions. Neurosurgery 2000;47:711-21. 29. Ruge MI, Victor J, Hosain S, Correa DD, Relkin NR, Tabar V, et al. Concordance between functional magnetic resonance imaging and intraoperative language mapping. Stereotact Funct Neurosurg 1999;72:95-102. 30. Fandino J, Kollias SS, Wieser HG, Valavanis A, Yonekawa Y. Intraoperative validation of functional magnetic resonance imaging and cortical reorganization patterns in patients with brain tumors involving the primary motor cortex. J Neurosurg 1999;91:238-50. 31. Jack CR, Jr, Thompson RM, Butts RK, Sharbrough FW, Kelly PJ, Hanson DP, et al. Sensory motor cortex: correlation of presurgical mapping with functional MR imaging and invasive cortical mapping. Radiology 1994;190:85-92. 32. Lehe´ricy S, Duffau H, Cornu P, Capelle L, Pidoux B, Carpentier A, et al. Correspondence between functional magnetic resonance imaging somatotopy and individual brain anatomy of the central region: comparison with intraoperative stimulation in patients with brain tumors. J Neurosurg 2000;92:589-98. 33. Puce A, Constable RT, Luby ML, McCarthy G, Nobre AC, Spencer DD, et al. Functional magnetic resonance imaging of sensory and motor cortex: comparison with electrophysiological localization. J Neurosurg 1995;83: 262-70. 34. Yetkin FZ, Mueller WM, Morris GL, McAuliffe TL, Ulmer JL, Cox RW, et al. Functional MR activation correlated with intraoperative cortical mapping. Am J Neuroradiol 1997;18:1311-15. 35. Majos A, Tybor K, Stefan´czyk L, Go´raj B. Cortical mapping by functional magnetic resonance imaging in patients with brain tumors. Eur Radiol 2005;15:1148-58. 36. Roux FE, Boulanouar K, Ranjeva JP, Tremoulet M, Henry P, Manelfe C, et al. Usefulness of motor functional MRI correlated to cortical mapping in Rolandic low-grade astrocytomas. Acta Neurochir (Wien) 1999;141:71-9. 37. Yousry TA, Schmid UD, Jassoy AG, Schmidt D, Eisner WE, Reulen HJ, et al. Topography of the cortical motor hand area: prospective study with functional MR imaging and direct motor mapping at surgery. Radiology 1995;195:23-9. 38. Mueller WM, Yetkin FZ, Hammeke TA, Morris GL, III, Swanson SJ, Reichert K, et al. Functional magnetic
297
298
20 39.
40.
41.
42.
43.
44. 45.
46.
47.
48.
49.
50.
51.
Functional MRI in image guided neurosurgery
resonance imaging mapping of the motor cortex in patients with cerebral tumors. Neurosurgery 1996;39:515-20. Schudler M, Maldijian JA, Liu WC. Functional imagedguided surgery of intracranial tumors located in or near the sensorimotor cortex. J Neurosurg 1998;83:412-18. Boling W, Parsons M, Kraszpulski M, Cantrell C, Puce A. Whole-hand sensorimotor area: cortical stimulation localization and correlation with functional magnetic resonance imaging. J Neurosurg 2008;108:491-500. Brannen JH, Badie B, Moritz CH, Quigley M, Meyerand ME, Haughton VM. Reliability of functional MR imaging with word-generation tasks for mapping Broca’s area. Am J Neuroradiol. 2001;22:1711-18. FitzGerald DB, Cosgrove GR, Ronner S, Jiang H, Buchbinder BR, Belliveau JW, et al. Location of language in the cortex: a comparison between functional MR imaging and electrocortical stimulation. Am J Neuroradiol 1997;18:1529-39. Roux FE, Boulanouar K, Lotterie J-A, Mejdoubi M, LeSage JP, Berry I. Language functional magnetic resonance imagine in preoperative assessment of language areas: correlation with direct cortical stimulation. Neurosurgery 2003;52:1335-45. McKhann GM, II, Hirsch J. Comment on reference 43. Neurosurgery 2003;52:1346. Petrovich N, Holodny AI, Tabar V, Correa DD, Hirsch J, Gutin PH, et al. Discordance between functional magnetic resonance imaging during silent speech tasks and intraoperative speech arrest. J Neurosurg 2005;103:267-74. Kombos T, Suess O, Kern BC, Funk T, Hoell T, Kopetsch O, et al. Comparison between monopolar and bipolar electrical stimulation of the motor cortex. Acta Neurochir (Wien) 1999;141:1295-301. Yoo SS, Talos IF, Golby AJ, Black PM, Panych LP. Evaluating requirements for spatial resolution of fMRI for neurosurgical planning. Hum Brain Mapp 2004;21:34-43. Sabbah P, Foehrenbach H, Dutertre G, Nioche C, DeDreuille O, Bellegou N, et al. Multimodal anatomic, functional, and metabolic brain imaging for tumor resection. Clin Imaging 2002;26:6-12. Archip N, Clatz O, Whalen S, Kacher D, Fedorov A, Kot A, et al. Non-rigid alignment of pre-operative MRI, fMRI, and DT-MRI with intra-operative MRI for enhanced visualization and navigation in image-guided neurosurgery. Neuroimage 2007;35:609-24. Gati JS, Menon RS, Ugurbil K, Rutt BK. Experimental determination of the BOLD field strength dependence in vessels and tissue. Magn Reson Med 1997;38: 296-302. Rugg MD, Johnson JD, Park H, Uncapher MR. Encoding-retrieval overlap in human episodic memory: a functional neuroimaging perspective. Prog Brain Res 2008;169:339-52.
Imaging in Stereotactic Surgery
17 General Imaging Modalities: Basic Principles A. A. Gorgulho . W. Ishida . A. A. F. De Salles
Historical Landmarks and Principles Although not based on imaging, the basic principle of stereotactic surgery started with the work of Zernov in 1890 [1]. A Russian anatomist, he developed a map of the brain cortex depicted in a hemisphere that, when attached to the human head, would keep a constant relationship with corresponding functional areas of the cortex. This instrument allowed placement of the craniotomy guided by the patient’s symptoms [2]. Further studies of the function of the central nervous system and symptoms the diseases required a more precise approach than the one devised by Zernov. Precise placement of recording and stimulating electrodes in specific areas of the brain to unveil function of deep structures called for a mathematical approach, Horsley and Clark devised and reported it in 1906 [3]. The Cartesian coordinate system, X (lateral), Y (anterior-posterior), Z (cranial-caudal), was born and remains the basis of stereotactic surgery (> Figure 17-1). If one reads the original work of Horsley and Clark, a striking finding is seen to the modern eyes; no imaging is mentioned for targeting the deep structures of the brain of the experimental animal [4]. As imaging was not readily available at the time, the skull landmarks were used as stereotactic reference. X-rays had just been described in Germany by Roentgen in 1895, only ten years before the seminal description of the stereotactic technique through the collaboration of the two English scientists, a surgeon and a physicist [3,4]. As the information age was not #
Springer-Verlag Berlin/Heidelberg 2009
as fast pace as it is today, years were necessary for the incorporation of scientific accomplishments to surgery. Although stereotactic surgery continued to be largely employed in the laboratory, using the Horsley and Clark’s interaural line and midline as reference [5], these skull-based reference points were too variable to allow a safe determination of a target in the depth of the human brain [6]. Moreover, little was known about the function of the deep structures of the brain to allow intervention in humans. The natural path of animal experimentation was necessary for confirmation of the effects of lesioning of brain structures before one would propose interventions in humans. It was approximately 20 years after the initial studies of Horsley and Clark on the functional anatomy of the deep brain structures that the theory of basal ganglia motor integration was put forward by Spatz [7]. This theory spearheaded the first attempts of surgery in the extrapyramidal system to control movement disorder [8].
Ventriculography and the Stereotactic Landmarks Parallel to these animal laboratory experimentations, imaging of the brain was being developed. Plain skull x-rays were followed by the description of ventriculography by Dandy in 1918 [9] and by angiography in 1927 by Egas Muniz [10]. These two monumental imaging modalities would dominate the landscape of stereotactic surgery for the following 50 years, firmly
252
17
General imaging modalities: basic principles
. Figure 17-1 The Horsley and Clark stereotactic apparatus. Although the X, Y, Z coordinates’ convention changed over the years, modern convention is as follows: the ‘‘X’’, ‘‘Y’’, and ‘‘Z’’ are right-left, antero-posterior and cranio-caudal displacement from the stereotactic space center respectively. (These pictures are a courtesy of the historical collection at UCLA Medical Library. This is the second Horsley and Clark apparatus assembled in history)
establishing the association of imaging and stereotactic localization. Since then stereotactic surgery has developed in parallel with imaging techniques. Early in their work, Spiegel and Wycis realized the importance of the stereotactic technique for morphological and functional neurosurgical interventions [6]. They described the need for improved imaging for visualization of deep brain structures, and actually developed methods of determination of stereotactic coordinates based on the calcification of the pineal gland, lately based on pneumoencephalography. The posterior commissure-pons line served as reference for their measurements. These measurements were used mainly for functional stereotactic surgery. While the development of functional stereotactic surgery was rapid with the perfection of their stereotactic frame (> Figure 17-2), the morphological applications evolved slowly because the visualization of lesions in the brain became available only with the incorporation of angiography to the stereotactic technique. Although few neurosurgeons still use ventriculography for functional neurosurgery, its use is practically a historical legacy.
. Figure 17-2 Spiegel and Wycis stereotactic device constructed in 1954, available at University of California in Los Angeles. As in Figure 19-1 notice the Cartesian coordinates, X, Y, and Z applied to human stereotactic surgery
General imaging modalities: basic principles
The main legacy of ventriculography and pneumoencephalography in stereotoactic surgery is the anterior-commissure (AC) and posteriorcommissure (PC) line. AC is seen approximately 2 mm below the posterior border of the foramen of Monro, while PC is seen just cranial to the entrance of the aqueduct of Sylvius and just caudal to the pineal calcification. These two anatomical landmarks, now readily visualized on all plans of the MRI, specifically seen in the sagittal and axial plan [> Figure 17-3], supported the development of the main atlases of the human brain, Shaltelbrand and Wahren [11] and Talairach’s proportional atlas of the human brain [12]. Specific targets in the brain are described based on the distance of the midcommissural plane which is the intersection of the
17
Cartesian coordinates dividing the brain into eight quadrants. Talairach used the length of the AC-PC line to develop the proportional atlas of the human brain widely used in epilepsy surgery.
Angiography Angiography paralleled the developments in stereotactic surgery. Stereotactic angiography was introduced by Talairach’s group [13]. Those workers dedicated several years of research to developing safe methods of inserting recording electrodes and radioisotope-loaded catheters and mapping the cortical anatomy by means of angiography. Their pioneer work using orthogonal
. Figure 17-3 T1 MRI axial and sagittal sections passing through AC-PC planes. (a) and (b) are axial AC-PC planes, in (a) without correction for AC-PC plane angle, notice the arrows showing AC (upper arrow) and PC (lower arrow). (b) shows the MRI precisely at AC-PC axial plane as reconstructed by the stereotactic software. (c) and (d) are sagittal AC-PC planes, in (c) without correction to the horizontal plane, notice the arrows AC (left arrow) and PC (right arrow). (d) shows AC-PC aligned to the horizontal plane by the stereotactic software
253
254
17
General imaging modalities: basic principles
approach to avoid the cerebral vasculature in functional and tumor stereotaxis established the grounds for several groups to use stereotactic angiography [14]. Specially trained neurosurgeons dedicated to epilepsy surgery followed Talairach’s work. Techniques of angiography mapping of the brain promptly brought to computerized stereotaxis, initially through the use of superimposition techniques [15,16], later by digitization or scanning of angiographic films [17], and more recently in DICOM format, even with three dimensional angiography [18]. Talairach’s group also concentrated on the understanding of the three-dimensional (3D) characteristics of the cerebral vasculature and its relationship with cerebral tumors aiming to develop diagnostic and therapeutic approaches, either with precisely placed craniotomy or by the use of stereotactic-guided placement of isotopes [14,19]. Because of the inherent two-dimensional nature of angiography, they relied on stereoscopic techniques to obtain the 3D information. The knowledge developed with stereotactic angiography led to the treatment of arteriovenous malformation (AVMs) with single dose radiation [20]. Angiography was not widely married with the stereotactic technique to approach intracranial lesions until the application of stereotactic radiosurgery for AVMs was described in 1972 [20]. Talairach described the implantation of isotopes for treatment of subcortical tumors using the blush of the angiogram [14], and Leksell described the use of external beam radiation directed stereotactically to obliterate intracranial targets and coined the term ‘‘Radiosurgery’’ [21]. Diagnostic procedures using the stereotactic technique were initiated only a decade later, despite the poor and only indirect visualization of structural brain lesions [22,23]. The number of morphological procedures surpassed the functional applications of stereotactic surgery with the advent of computed-imaging techniques which allowed direct visualization of
the target [24]. Angiography sponsored the fast development of radiosurgery and approaches for determination of the seizure foci in epilepsy surgery.
Computerized Era When computed tomography (CT) scan became available to stereotactic surgery, approximately 30 years after the first human stereotactic proedure, stereotactic surgery had a second revolution [24]. Now lesions could be visualized and the risk of approaching highly vascularized lesions became measurable. Biopsies of brain tumors, brachytherapy and especially radiosurgery dominated the time of stereotactic surgeons during the 1980s and 1990s [25]. CT scan also brought back the interest of neurosurgeons to lesioning the depth of the brain for symptom control in neurodegenerative disease such as Parkinson’s disease, since the precision and the safety of the stereotactic method improved [26]. The functional stereotactic landmarks well seen in Dandy’s ventriculography, which served the bases for all the electrophysiological studies of the specialty, were now well seen with the CT scan. The 1980s saw the resurrection of lesioning in the brain as a therapeutic option. The ventriculographic approach was compared with the computerized approach and the computerized era for functional neurosurgery was established [27]. However it was the morphological application of the stereotactic method that spearheaded this revolution and extension of the technique to common place in the regular operating room of the general neurosurgeon [28]. This came with the progressive abandonment of the stereotactic frame for image-guided surgery using triangulation methods and explosion of stereotactic radiosurgery as a minimally invasive technique for treatment of brain tumors and arteriovenous malformations [29,30].
General imaging modalities: basic principles
Computer Tomography Stereotactic Principles Hounsfield rightly received the Nobel Prize for medicine in 1979 for his description of X-Ray computed tomography (CT) in 1973 [31]. The imaging modality revolutionized neuroscience and the knowledge of brain pathology, function, and the ability of the stereotactic neurosurgeon to approach the brain safely. The technique was developed for visualization and not for precise calculation of intracranial targets. Therefore, years of work of stereotactic surgeons was necessary to bring this image safely into stereotactic surgery. As CT provided axial images only, stereotactic surgeons could determine only two coordinates from the slice of interest, i.e., where the target was located, either the pathology or the functional site to be targeted. By convention, the ‘‘X’’ became the lateral coordinate and the ‘‘Y’’ the antero-posterior. The vertical coordinate was not seen in the chosen slice, and the stereotactic surgeon had to devise methods of ‘‘Z’’ determination. Initially stereotactic surgeons relied upon the movement of the CT scan table to calculate the ‘‘Z’’, however the manufacturers of the CT scans were not worried about the precision of movement of the table, since the scanners were built for diagnosis. Frames were developed to overcome this imprecise movement of the table. The best example of such strategies is the Laitinen’s device which had transverse bars calibrated to correct the imprecision of the table movements [32]. It was not until the clever oblique bar introduced to a localizing box attached to the stereotactic frame by a graduate student at the University of Utah that the problem of the ‘‘Z’’ coordinate could be solved (> Figure 17-4) [33]. The stereotactic frame with the localization box became standard for all stereotactic procedures, including functional and morphological, from fine lesioning of pathways in the brain, to
17
. Figure 17-4 Axial representation of the fiduciary system with explanation of the oblique fiducial of the stereotactic localizing box. The Brown-Roberts-Wells (BRW) localizing box allows for three-dimensional definition of a point in any imaging slice (insert). The 9 points fiduciary system became widely used because of the possibility to correct for frame misaligment. The X and Y can be directly extracted from the axial slices, while the Z is calculated using the distance between the oblique bar to the reference bar in each slice
implanting electrodes to biopsies and craniotomy [16,34]. All commercially available stereotactic frames were adapted for the use of the oblique fiducials for determination of the ‘‘Z’’ (vertical) coordinate. The accuracy of the method was compared to the most used frames and shown to be submillimetric [35]. CT is considered up to now the most precise method for determination of stereotactic coordinates. The nature of X-rays with its rectilinear path avoids the introduction of distortions in the calculations. Distortions are introduced when using magnetic resonance imaging (MRI).
MRI Principles As the MRI scan became available and its geometric distortions were controlled [36–38], this imaging technique was preferred by stereotactic
255
256
17
General imaging modalities: basic principles
surgeons [39,40]. Because it presents exquisite visualization of the nuances of the brain pathways and nuclei and the consequences of the surgery [41–43], it has revolutionized the approach of functional stereotactic surgery, no longer depending so much on ventricular landmarks, but relying on direct targeting of the structure needing functional modification [44]. Brain-function visualization is the next frontier on the development of stereotactic targeting. The incorporation of the chronic electrical stimulation as a therapeutic approach, initially for treatment of behavioral disorders, then for refractory chronic pain and movement disorders and more recently again for psychiatric disorders, has decreased the serious complications of the approaches in the depth of the brain. Progressively lesions of nuclei and pathways are being replaced by the ability of electrical stimulation to modify function by focally modulating neurons and brain networks. Functional imaging becomes more important for modulation of functional diseases of the brain, such as the neurodegenerative disease, genetic pathologies and brain damage by ischemic or traumatic injuries. Functional MRI and the ability to operate inside the magnet using frame [39,41] or frameless techniques brought new opportunities for functional neurosurgery [45–47]. In the arena of morphologic stereotaxis the revolution was on imaging localization of tumors and malformations in the brain needing intervention. Initially stereotactic surgery was used for simple needle biopsy [48], then to aid resections and guidance [16]. Here also functional imaging and the exclusive visualization of fibers related to lesions are revolutionizing surgical resections and targeting functional stereotactic surgery (> Figure 17-5) [49,50]. These important imaging developments are readily applicable to stereotactic radiosurgery (SRS), currently representing substantial, if not the major application of the stereotactic technique [51]. Initially SRS
was dependent on the stereotactic frame [52] and now, similar to surgical resection, it is becoming independent of the frame approach [53].
Positron Emission Tomography and Stereotactic Procedures CT and MRI scans sometimes do not adequately demonstrate the regions of interest for the stereotactic procedures. Molecular imaging, capable of demonstrating pathologies not seen in morphologic imaging can complement the needs of stereotactic surgery. Positron emission tomography (PET) adds this important metabolic information, and when incorporated by fusion with CT and MRI, may allow more accurate targeting and treatment planning in stereotactic radiosurgery, tumor resection, and biopsy.
PET in Morphological Stereotactic Surgery Most PET scans use a radiotracer made up of a common metabolite, such as glucose or an amino acid, attached to a radioisotope such as 18F (Fluorine) or 11C (Carbon). The 18F-FDG (fluoro-deoxy-glucose) PET is most widely used, and when combined with CTor MRI, will demonstrate with exquisite anatomic precision regions of increased glucose metabolism. Since neoplasms and inflammatory lesions often have high glucose uptake matching that of the brain, differentiation of a lesion from normal surrounding brain may be limited. The amino acids, however, are selectively more utilized by neoplasms than normal brain. The 18F-DOPA (fluoro-phenylalanine) and 11C (Carbon) methionine PET scans utilize amino acid molecules, and have demonstrated increased radioactivity in neoplasms, when compared to normal brain [54,55]. Extent of surgical resection or radiosurgery targeting will sometimes
General imaging modalities: basic principles
17
. Figure 17-5 3-D frame and fiber tracking of the pyramidal system used for subthalamic nucleus targeting (traced arrow). Notice the distortions that can happen in fiducial system of the stereotactic localizing box (full arrows), reconstruction with iPlan software (BrainLab, Germany). This is a Leksell frame with copper sulfide liquid in the fiducial system (Elekta, Sweden)
be modified significantly by incorporation of PET on CT and MRI imaging [56–58]. Fluorodopa PET, C-methionine PET, and other amino acid-based PET scans have proven to be more effective than FDG-PET for imaging of neoplasms [54,59]. Fluorodopa and C-methionine PETscans demonstrated sensitivity to low grade as well as high grade tumors, and may help to differentiate areas of radiation necrosis (> Figure 17-6) [59,60]. PET scans sometimes demonstrate evidence of tumor recurrence before CT or MRI. They have proven to be most helpful in the management of gliomas [56–58], but also useful in treatment of other malignancies. It can be of help with pituitary adenomas [61], meningiomas, and parasellar lesions, where proximity to the cavernous sinus makes differentiation of tumor
from normally enhancing structures difficult. PET has proven to be helpful with spinal cord tumors [62,63], particularly in the presence of instrumentation or in the patient not able to tolerate MRI (pacemaker or electrical stimulator) in preparation for radiosurgery.
PET Scans in Functional Neurosurgery The PET characteristics of brain anatomy under normal and abnormal conditions have provided clues to a better understanding of brain anatomy and physiology. Changes on PET scanning related to deep brain stimulation have added to the still rudimentary body of information relating
257
258
17
General imaging modalities: basic principles
. Figure 17-6 Fluorodopa-PET to differentiate tumor recurrence from radiation necrosis. Nonsmall cell carcinoma brain metastasis treated with radiosurgery using 16 Gy prescribed to the 90% isodose line. (a) T1 MRI with gadolinium showing lesion growth with central hypo-intensity, possibly radiation necrosis. (b) Fluorodopa PET performed one year after the treatment and at the same time of the MRI in (a). (c) Coronal PET showing higher uptake of Fluorodopa in the lesion (full arrow) than in the basal ganglia (traced arrow), consistent with tumor recurrence. Histology of the resected specimen showed nonsmall cell carcinoma with focal necrosis
to pain, movement disorders, and behavioral problems. The noninvasiveness of PET makes it a valuable research modality. Striatal as well as extra-striatal dopaminergic activity in neurological and psychiatric disorders have been studied using PET biomarkers [64]. The PET tracers are very important in advanced research on problems of early and more specific diagnosis of major movement and psychiatric disorders, physiology of the dopaminergic system, evolution of disease processes, and response to medications and surgical interventions [65–68]. FDG-PET studies in patients with major depression have demonstrated increased glucose metabolism in the left amygdala and frontal limbic pathways, with evidence of decreased amygdala metabolism under antidepressant drug treatment [69] and/or vagus nerve stimulation [70]. Similar PET responses have been reported with deep brain stimulation of the anterior limbic system, such as the subcallosal cingulate gyrus [71]. These findings are consistent with the findings of PET blood flow studies in depressed patients by Mayberg et al. [72,73]. Mayberg et al. used the findings of increased blood flow in the sub-genual cingulate cortex, area AcG25, to realize a target
for deep brain stimulation to control the symptoms of medically refractory major depression. Mayberg et al’s early work demonstrated the integral role played by the subgenual cingulate cortex in both, normal, and pathological shifts in mood [73]. Increases in limbic and paralimbic blood flow (as measured using PET) occur in the subgenual cingulate cortex and anterior insula during sadness. There is a significant inverse correlation between blood flow in the subgenual cingulate cortex and right dorsolateral prefrontal cortex [74]. A clinical response to antidepressants is associated with limbic and striatal (subgenual cingulate cortex, hippocampus, insula, and pallidum) decreases in metabolism and dorsal cortical (prefrontal, parietal, anterior, and posterior cingulate cortex) increases in metabolism [69,72]. In 2005, DBS electrodes were bilaterally implanted in the subgenual cingulate cortex [75] of 6 patients with medically refractory major depression. When stimulation was on, patients reported positive emotional phenomena. In the acute postoperative period the patients experienced reproducible increases in activity and mood scores, changes that failed to occur during sham stimulation. Chronic stimulation at high
General imaging modalities: basic principles
frequency, probably leading to suppression of function in the site, resulted in significant response and remission of depression in 4 of the 6 patients at 6 months. These well conducted studies showed the effectiveness of PET findings to enhance the knowledge of brain function leading to diagnosis and therapeutic measures. Anterior capsule deep brain stimulation has resulted in decreased (18) FDG-PET activity or decreased glucose uptake in the subgenual anterior cingulate gyrus and ventral striatum in a group of patients with refractory obsessive compulsive disorder [76]. OCD patients with hoarding behavior showed different patterns of cortical PET- FDG activity, compared to those with nonhoarding behavior [77]. Evidence is accumulating to support the use of PET in routine target determination and follow up of patients undergoing neurosurgical interventions for mental illness. Advances in radioligand technology [78] have provided radioisotope labeled molecules that enable study of neurotransmitters (serotonin, norepinepherine, dopamine and glutamate) and their receptors with PET [79–82]. Patterns and intensity of uptake have added to our understanding of movement and psychological disorders. As an example fluorodopa-PET uptake in the putamen would be decreased in idiopathic Parkinson’s Disease as well as in a Parkinson’s-Plus condition. However, greater loss of striatal D2 receptors in Parkinson’s-Plus on a (11)C-raclopride PET scan might help to identify the patient as being unsuitable for surgical treatment with stereotactic implant of DBS, pallidal or thalamic lesion. Radioligands and PET have advanced the study of neuroreceptors remaining a most valuable tool for ongoing studies and treatment of patients with motor and psychological disorders [83].
Image Fusion The ability to bring multimodality imaging to plan stereotactic procedures started with the work of Talairach. He obtaining information
17
from angiography to avoid vasculature for seizure placement of electroencephalographic electrodes for seizure focus determination [13]. The attempt to bring MRI and CT into the stereotactic space determined by plain X-ray was started for stereotactic radiosurgery with fusion of imaging by photographic magnification manipulation [15]. These efforts have culminated with computerized fusion with software algorithms developed based on contours, image intensity, and voxel matching. These precise approaches have allowed gathering detailed information prior to the procedure, facilitating the surgery planning. Before these techniques were available, the patient had imaging with the stereotactic frame in place and multimodality approaches were unyielding. Now a portable CT scan in the operating room has obviated the need of elaborate stereotactic operating rooms. The fusion techniques offer also the opportunity of atlas information integration to the patient’s image (> Figure 17-7). Moreover, real time information on brain shift and possible complication during the operation are obtained while operating inside the magnet [39,41]. Fusion of multimodality images is very important for correction of distortions of PET, digital angiography and MRI scans. The portable CT in the operating room can offer this correction [38,84].
Image-Guided Surgery The integration of multimodality imaging is possible without the stereotactic frame [85–87]. This capability has revolutionized not only stereotactic surgery but also general neurosurgery. It is now impacting in other specialties such as radiation oncology, orthopedic surgery, head and neck surgery and general surgery. Modern imaging technology brings presurgical information to the surgeon that obviates unknowns. Computer technology, using this information, provides that surgery can be performed virtually on a screen before the patient is even touched. In addition, surgery has advanced to a level
259
260
17
General imaging modalities: basic principles
. Figure 17-7 (A) Shaltenbrand-Wahren coronal plate. (B) Superimposition of the anatomy in (A) onto the corresponding MRI coronal slice by iPlan stereotactic software (BrainLab, Germany). Notice the adjusting sliding scale under (A) capable of matching the atlas with the MRI
where minimal invasion and maximal effectiveness is routine. The term ‘guided surgery’ in the modern sense, should be viewed as ‘modern surgery’. Guided surgery’, however, is still seen by many as the use of computerized imaging, or traditional X-ray-based stereotactic techniques described above to bring the surgeon precisely to the pathology being operated on. The pressures of competition and multimillion dollar malpractice law suits have driven the modern medical centers to invest heavily in technology. This in turn has driven the price of medical procedures to almost unacceptable levels. The
hope is that applied technology can decrease the costs of each patient treatment. Image guided surgery is an area that may lead to substantial savings in medical dollars. The scope of the approaches and the realistic surgical undertaking may lead to shorter hospital stays due to fewer complications related to extensive surgeries, less need for long convalescent and rehabilitation periods, and, consequently, a faster return of the patient to the workforce. Ultimately, this results in decreasing the overall price of medical care. This concept has been exemplified with complex skull base disease.
General imaging modalities: basic principles
These difficult tumors are treated now with transnasal procedures for skull base tumor resections [88] under real time imaging in the operating room, and followed by radiosurgery, reducing patient recovery time, decreasing morbidity, and offering the patient complete control of their disease [89]. The stereotactic developments throughout the 20th century as described in this chapter, spearheaded by the computerized imaging, provided remarkable noninvasive imaging techniques developed primarily for diagnostic studies. These techniques were adapted for surgery guidance with navigation using triangulation techniques [85–87]. Now infrared reflectors or magnetic field are used for real-time localization [90]. Fast computers and smart software packages permitted the introduction of these images to the operating field to guide the surgeon. Digital fluoros copy, ultrasound, computer tomography (CT), magnetic resonance imaging (MRI) and positron emission tomography (PET) are now brought to the operating room and combined with merging data set techniques, allowing the surgeon to take advantage of a wealth of information that was previously unavailable. The surgeons of the past relied upon the principle of ‘‘exposure, exposure, exposure’’, and their individual knowledge of gross anatomy to perform surgery. The surgeons of the present rely on their knowledge of anatomy, anatomical imaging, and functional anatomy to perform minimally invasive procedures and solve previously unapproachable problems [91].
Spine Stereotactic Surgery Invasive stereotactic fixation for radiosurgery of the spine was previously tried without acceptance from the stereotactic community [96]. The procedure proved to be too invasive and impractical to be largely applied. The development of imageguided surgery, as described above, provided the
17
base for the development of the spine stereotactic technique. Image fusion and computerized image are now applied to stereotactic radiosurgery of spine lesions [91,94,95] and for placement of hardware. Completely noninvasive, fiducial systems use infrared triangulation and online image fusion of oblique X-rays and CT reconstruction to provide real-time movement tracking and targeting of lesions in the spine and surrounding regions [97,98]. The technology has reached precision to treat intramedullary lesions (> Figure 17-8).
Future Directions We are on the verge of perfecting real-time imaging in surgery [39,46]. During the past decade, the information brought during surgery by plain X-rays, fluoroscopy, and ultrasound was maximized and their limitations were established. Surgeons have now turned their eyes to the wealth of possibilities brought by portable CT scans and operating rooms equipped with interventional MRIs. MRI offers the possibility of not only exquisite anatomical information during surgery, but also dynamic changes of this anatomy associated with real-time changes in function. It also carries the advantage of not being harmful to the medical personnel, as are techniques dependent on isotopes or X-rays. The operating room with multimodality imaging, also known as operating room of the future, is a focus of studies in major medical centers. The logistics and real advantages of bringing a complex technology such as MRI to the operating room, or bringing the operating room to the complex MRI environment, has become a subject of symposiums on modern surgery [41]. The evolving field of functional MRI has brought the possibility of deciding before surgery the location of a fine function in the brain in relation to pathology (> Figure 17-9). It has also allowed relating the complex wiring of the brain to the location of ‘brain pace makers’ (> Figure 17-5
261
262
17
General imaging modalities: basic principles
. Figure 17-8 Medullary AVM in a 22-year-old woman who bled, developed tetraplegia and recovered after a C2–C7 laminectomy. (a) Sagittal contrasted T1 MRI before radiosurgery. (b) Sagittal CT with the radiosurgery plan (12 Gy, 90% isodose line, 1.60 cc lesion). (c) Sagittal contrasted T1 MRI 24-months post-radiosurgery. (d) Anteroposterior angiogram before treatment, notice the angiographic stereotactic fiducials (traced arrow). (e) Coronal CT showing the radiosurgery plan. (f) Anteroposterior angiogram post-treatment. Full arrows point the initial nidus (a), residual at 24 months (c) and residual at 26 months (f)
and > Figure 17-10). All this information is readily related to imaging during surgery. Products that integrate information from multiple imaging sources with diffusion of fluids through tissues, such as brain parenchyma for delivery of drugs after resection, have started to appeart on the market. This is achieved with stereotactic precision. Similar information is being generated by therapeutic thermal application, electrical current, and radiation. Laser or radiofrequency ablation, electrical stimulation with smart pacemakers capable of receiving and analyzing physiological clues, and radiation delivery with
modulating capabilities are all novel approaches being developed [92,93]. The operating room of the future for ‘guided surgery’ and modern stereotactic surgery, requires real-time anatomical imaging technology related to function of the tissue under therapy at the moment of resection, lesioning or modulation [42,93]. This allows maximization of resection, drug infusion, electrical tissue influence, biopsy and the optimal use of radiation strategies to manipulate biological systems [92,101]. The patient should be least invaded and most helped by the modern stereotactic techniques.
General imaging modalities: basic principles
17
. Figure 17-9 (a) Functional MRI showing Broca’s area (traced arrows) just posterior to an Arteriovenous Malformation (full arrows). The same AVM is shown on anteroposterior (b) and oblique (c) angiogram. The AVM removed through a craniotomy guided by stereotactic triangulation with the patient awake for complete speech preservation, lateral angiogram (d)
. Figure 17-10 MRI fiber tracking from the subgeneal area (AcG25) recognized as the stereotactic target for implantation of deep brain stimulation electrodes for treatment of medically refractory depression, Notice the virtual electrode placement (open arrows). Notice the fibers going to prefrontal and orbitofrontal cortex and cingulate fasciculus (full arrows) [49,50]. (a) shows 3-D, sagittal and coronal reconstructions with axial MRI passing through posterior commissure, (b) same reconstruction passing through anterior commissure (traced arrows)
263
264
17
General imaging modalities: basic principles
References 1. Zernov DN. L’ence´phalome´tre. Rev Gen Clin Ther 1890;19:302. 2. Kandel EI, Schavinsky YV. Stereotaxic apparatus and operations in Russia in the 19th century. J Neurosurg 1972;37:407-411. 3. Clark RH, Horsley V. On a method of investigating the deep ganglia and tracts of the central nervous system (cerebellum). Brit Med Journal 1906;2:1799-1800. 4. Horsley V, Clark RH. The structure and function of the cerebellum examined by a new method. Brain 1908;31:45-125. 5. Sun B, De Salles AAF, Medin P, Solberg T, Hoebel B, Felder-Allen M, Wie C-W, Ackerman R. Reduction of hippocampal-kindled seizure activity in rats by stereotactic radiosurgery.Experimental Neurology. 1998;154:691-695. 6. Spiegel EA, Wycis HT. Stereoencephalotomy, Part I. New York: Grune & Stratton; 1952. 7. Pia HW. In memory of Hugo Spatz 1888–1969. Acta Neurochir (Wien) 1970;23(2):183-6. 8. Gildenberg PL. Evolution of Basal Ganglia Surgery for Movement Disorders. Stereotact Funct Neurosurg 2006;84:131-135. 9. Dandy WE. Ventnculography following the injection of air into the cerebral ventricles. Ann Surg 1918;68:5-11. 10. Moniz E, de Carvalho L, Almeida Lima P. Angiopneumographie. Presse Med 1931;39:996-999. 11. Schaltenbrand G, Wahren AE. Stereotaxy of the human brain. Stuttgart: Thieme; 1982. 12. Talairach J, David M, Tournoux P, Corredor H, Kvasina T. Atlas d’anatomie stereotaxique. Paris: Masson; 1957. 13. Szikla G, Bouvier G, Hori T, et al. Angiography of the human brain cortex. Berlin: Springer-Verlag; 1977. 14. Talairach J, Aboulker P, Ruggiero G, et al. Utilization del methode radiostereotaxic pour le traitement radioactive in situ des tumeurs cerebrales. Rev Neurol 1953;90:656-658. 15. De Salles AAF, Asfora WT, Abe M, Kjellberg RN. Transposition of target information from the magnetic and computed tomography scan images to conventional X-ray stereotactic space. Appl Neurophyisol 1987;50:23-32. 16. Kelly PJ. Tumor Stereotaxis. Philadelphia: Sounders; 1991. p. 108-113. 17. Zhang J, Levesque MF, Wilson CL, et al. Multimodality imaging of brain structures for stereotactic radiosurgery. Radiology 1990;175:435-441. 18. Pedroso AG, De Salles AA, Tajik K, Golish R, Smith Z, Frighetto L, Solberg T, Cabatan-Awang C, Selch MT. Novalis Shaped Beam Radiosurgery of arteriovenous malformations. J Neurosurg 2004;101 Suppl 3:425-34. 19. Bancoud J, Morel P, Talairach J, et al. Interet de l’exploration functionelle stereotaxique dans la localization des lesions expansives. Rev Neurol (Paris) 1961;105: 219-220.
20. Steiner L, Leksell L, Greitz T, et al. Stereotactic radiosugery for cerebral arteriovenous malforations: Report of a case. Acta Chir Scand 1972;138:459-462. 21. Leksell L. The stereotactic method of radiosurgery of the brain. Acta Chir Scand. 1951;102:316-319. 22. Housepian EM, Pool JL. Application of stereotaxic methods to histochemical, electromicroscopic and electrophysiological studies of bran subcortical structures. Confin Neurol 1962;22:173-177. 23. Kalynaraman S, Gillingham FG. Stereotaxic biopsy. J Neurosurg 1964;21:854-858. 24. Gildenberg PL. The history of stereotactic neurosurgery. Neurosurg Clin N Am 1990;1(4):765-80. 25. De Salles AAF, Goetsch SJ (eds.). Stereotactic Surgery and Radiosurgery. Madison, Wisconsin: Medical Physics Publishing Corporation; 1993. p. 1-16. 26. Laitinen LV, Bergenheim AT, Hariz MI. Leksell’s posteroventral pallidotomhy in the treatment of Parkinson’s disease. J Neurosurg 1992;76(1):53-61. 27. Hariz MI, Bergenheim AT. A comparative study on ventriculographic and computerized tomography-guided determinations of brain target in functional stereotaxis. J Neurosurg 1990;73:565-571. 28. Lunsford LD. The genesis of neurosurgery and the evolution of the neurosurgical operative environment: Part 1–Prehistory to 2003. Neurosurgery 2003;52(6): 1512. 29. De Salles AAF, Lufkin R. Minimally Invasive Therapy of the Brain. Thieme Medical Publishers; pp 1–6, 1997. 30. Lunsford LD. Radiosurgery as a future part of neurosurgery. Mayo Clin Proc. 1999;74(1):101-3. 31. Hounsfield GN. Computerized transverse axial scanning (tomography). I. Description of system. Br J Radiol 1973;46:1016-1022. 32. Laitinen LV, Hariz MI. Multi-purpose stereoadapter. Appl Neurophysiol 1987;50(1–6):68-76. 33. Heilbrun MP, Roberts TS, Apuzzo ML, Wells TH Jr, Sabshin JK. Preliminary experience with BrownRoberts-Wells (BRW) computerized tomography stereotaxic guidance system. J Neurosurg 1983;59(2):217-22. 34. Hariz MI. Correlation between clinical outcome and size and site of the lesion in CT-guided thalamotomy and Pallidotomy. Stereotactic and Funct Neurosurg 1990;54 (55):172-185. 35. Maciunas RJ, Galloway RL Jr, Latimer J, et al. An independent application accuracy evaluation of stereotactic frame systems. Stereotactic and Funct Neurosurg 1992;58:103-107. 36. Maciunas RJ, Fitzpatrick JM, Gadamsetty S, Maurer CR Jr. A universal method for geometric correction of magnetic resonance images for stereotactic neurosurgery. Stereotact Funct Neurosurg 1996;66(1–3):137-40. 37. Burchiel KJ, Nguyen TT, Coombs BD, Szumoski J. MRI distortion and stereotactic neurosurgery using the Cosman-Roberts-Wells and Leksell frames. Stereotact Funct Neurosurg 1996;66(1–3):123-36.
General imaging modalities: basic principles
38. Alexander E 3rd, Kooy HM, van Herk M, Schwartz M, Barnes PD, Tarbell N, Mulkern RV, Holupka EJ, Loeffler JS. Magnetic resonance image-directed stereotactic neurosurgery: use of image fusion with computerized tomography to enhance spatial accuracy. J Neurosurg 1995;83(2):271-6. 39. De Salles AAF, Frighetto L, Behnke E, Sinha S, Bronstein J, Torres R, Subramanian I, Cabatan-Awang C, Frysinger R. Functional Neurosurgery in the MRI Environment. Minim Invasive Neurosurg. 2004;47 (5):284-9. 40. Starr PA, Christine CW, Theodosopoulos PV, Lindsey N, Byrd D, Mosley A, Marks WJ Jr. Implantation of deep brain stimulators into the subthalamic nucleus: technical approach and magnetic resonance imaging-verified lead locations. J Neurosurg 2002;97(2):370-87. 41. Lee WYM, De Salles AAF, Frighetto L, Torres R, Behnke E, Bronstein J. Imaging techniques and electrode fixation methods for deep brain stimulation: Intraoperative MRI 0.2 T, 1.5 T and fluoroscopy. Minimally Invasive Neurosurgery 2005;48:1-6. 42. De Salles AAF, Brekhus SD, De Souza EC, Behnke EJ, Farahani K, Anzai Y, Lufkin R. Early postoperative appearance of radiofrequency lesions on magnetic resonance imaging. Neurosurgery 1995;36 (May):932-936. 43. De Salles AAF. Role of stereotaxis in the treatment of cerebral palsy. J of Child Neurology 1996;11:S43-S50. 44. De Salles AAF, Hariz M. MRI Guided Pallidotomy. In: Rengachary SS, editor. Neurosurgical Operative Atlas, Volume 7, Williams & Wilkins, AANS; 1998. pp 141-148. 45. Martin AJ, Hall WA, Roark C, Starr PA, Larson PS, Truwit CL. Minimally invasive precision brain access using prospective stereotaxy and a trajectory guide. J Magn Reson Imaging 2008;27(4):737-43. 46. Larson PS, Richardson RM, Starr PA, Martin AJ. Magnetic resonance imaging of implanted deep brain stimulators: experience in a large series. Stereotact Funct Neurosurg 2008;86(2):92-100. 47. Holloway KL, Gaede SE, Starr PA, Rosenow JM, Ramakrishnan V, Henderson JM. Frameless stereotaxy using bone fiducial markers for deep brain stimulation. J Neurosurg 2005;103(3):404-13. 48. Hariz MI, Bergenheim AT, De Salles AAF, Rabow L, Trojanowski T. Percutaneous stereotactic brain tumor biopsy and cyst aspiration with a non-invasive frame. British Journal of Neurosurgery 1990;4:397-406. 49. Sedrak M, Gorgulho A, De Salles AF, Frew A, Behnke E, Ishida W, Klochkov T, Malkasian D. The role of modern imaging modalities on deep brain stimulation targeting for mental illness. Acta Neurochir Suppl 2008;101:3-7. 50. Hauptman JS, DeSalles AA, Espinoza R, Sedrak M, Ishida W. Potential surgical targets for deep brain stimulation in treatment-resistant depression. Neurosurg Focus. 2008;25(1):E3. 51. De Salles AA, Gorgulho AA, Selch M, De Marco J, Agazaryan N. Radiosurgery from the brain to the
52. 53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65. 66.
67.
17
spine: 20 years experience. Acta Neurochir Suppl 2008;101:163-8. Leksell L. A stereotaxic apparatus for intracerebral surgery. Acta Chir Scand 1951;102:316-319. Agazaryan N, Tenn SE, DeSalles AA, Selch MT. Imageguided radiosurgery for spinal tumors: methods, accuracy and patient intrafraction motion. Phys Med Biol 2008;21;53(6):1715-27. Becherer A, Karanikas G, Szabo M, et al. Brain tumour imaging with PET: a comparison between [18F] fluorodopa and [11C] methionine. Eur J Nucl Med Mol Imaging 2003;30:1561-1567. Chen W, Cloughesy T, Kamdar N, et al. Imaging Proliferation in Brain Tumors with 18F-FLT PET: Comparison with 18F-FDG. J Nucl Med 2005;46:945-952. Grosu A, Weber W, Astner S, et al. 11C-Methionine PET Improves the Target Volume Delineation of Meningiomas Treated with Stereotactic Fractionated Radiotherapy. Int J Radiat Oncol Biol Phys 2006;66:339-344. Pirotte B, Goldman S, Dewitte O, et al. Integrated positron emission and magnetic resonance imagingguided resection of brain tumors: a report of 103 consecutive procedures. J Neurosurg 2006;104:238-253. Singhal T, Narayanan T, Jain V, et al. 11C-L-Methionine Positron Emission Tomography in the Clinical Management of Cerebral Gliomas. Mol Imaging Biol. 2008;10 (1):1-18. Chen W, Silverman D, Delaloye S, et al. 18 F-DOPA PET Imaging of Brain Tumors: Comparison Study with 18FFDG PET and Evaluation of Diagnostic Accuracy. J Nucl Med 2006;47:904-911. Terakawa Y, Tsuyuguchi N, Iwai Y, et al. Diagnostic accuracy of 11C-Methionine PET for differentiation of recurrent brain tumors from radiation necrosis after radiotherapy. J Nucl Med 2008;49(5):694-9. Tang B, Levivier M, Heureux M, et al. 11C-Methionine PET for the diagnosis and management of recurrent pituitary adenomas. Eur J Nucl Med Mol Imaging 2006;33:169-178. Shimizu T, Saito N, Aihara M, et al. Primary Spinal Oligoastrocytoma: A Case Report. Surg Neurol 2004;61:77-81. Wilmshurst JM, Barrington SF, Pritchard D, et al. Positron emission tomography in imaging spinal cord tumors. J Child Neurol 2000;15(7):465-72. Elsinga PH, Hatano K, Ishiwata K. PET tracers for imaging of the dopaminergic system. Curr Med Chem 2006;13(18):2139-53. Berg D. Biomarkers for the early detection of Parkinson’s and Alzheimer’sdisease.Neurodegener Dis 2008;5(3–4):133-6. Broussolle E, Dentresangle C, Landais P, et al. The relation of putamen and caudate nucleus 18F-Dopa uptake to motor and cognitive performances in Parkinson’s disease. J Neurol Sci 1999;166(2):141-51. Koerts J, Leenders KL, Koning M, et al. Striatal dopaminergic activity (FDOPA-PET) associated with cognitive
265
266
17 68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
General imaging modalities: basic principles
items of a depression scale (MADRS) in Parkinson’s disease. Eur J Neurosci 2007;25(10):3132-6. Moore RY, Whone AL, Brooks DJ. Extrastriatal monoamine neuron function in Parkinson’s disease: an 18Fdopa PET study. Neurobiol Dis 2008;29(3):381-90. Drevets WC, Bogers W, Raichle ME. Functional anatomical correlates of antidepressant drug treatment assessed using PET measures of regional glucose metabolism. Eur Neuropsychopharmacol 2002;12(6):527-44. Pardo JV, Sheikh SA, Schwindt GC, et al. Chronic vagus nerve stimulation for treatment-resistant depression decreases resting ventromedial prefrontal glucose metabolism. Neuroimage 2008;42(2):879-89. Lozano AM, Mayberg HS, Giacobbe P, et al. Subcallosal cingulate gyrus deep brain stimulation for treatment of resistant depression. Biol Psychiatry 2008;15:64(6): 461-7. Mayberg HS, Brannan SK, Tekell JL, Silva JA, Mahurin RK, McGinnis S, et al. Regional metabolic effects of fluoxetine in major depression: serial changes and relationship to clinical response. Biol Psychiatry 2000;48:830-843. Mayberg HS, Liotti M, Brannan SK, McGinnis S, Mahurin RK, Jerabek PA, et al. Reciprocal limbic-cortical function and negative mood: converging PET findings in depression and normal sadness. Am J Psychiatry 1999;156:675-682. Liotti M, Mayberg HS, Brannan SK, McGinnis S, Jerabek P, Fox PT. Differential limbic—cortical correlates of sadness and anxiety in healthy subjects: implications for affective disorders. Biol Psychiatry 2000;48:30-42. Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, et al. Deep brain stimulation for treatment-resistant depression. Neuron 2005;45:651-660. Van Laere K, Nuttin B, Gabriels L, et al. Metabolic imaging of anterior capsular stimulation in refractory obsessive-compulsive disorder: a key role for the subgenual anterior cingulated and ventral striatum. J Nucl Med 2006;47(5):740-7. Saxena S, Brody A, Maidment K, et al. Cerebral Glucose Metabolism in Obsessive-Compulsive Hoarding. Am J Psychiatry 2004;161:1038-1048. Schou M, Pike V, Halldin C. Development of radioligands for imaging of brain norepinephrine transporters in vivo with positron emission tomography. Curr Top Med Chem 2007;7(18):1806-16. Cannon DM, Ichise M, Rollis D, et al. Elevated serotonin transporter binding in major depressive disorder assessed using positron emission tomography and (11C)DASB; comparision with bipolar disorder. Biol Psychiatry 2007;62(8):870-7. Meyer JH. Imaging the serotonin transporter during major depressive disorder and antidepressant treatment. J Psychiatry Neurosci 2007;32(2):86-102.
81. Rauch SL, Dougherty DD, Malone D, et al. A functional neuroimaging investigation of deep brain stimulation in patients with obsessive-compulsive disorder. J Neurosurg 2006;104(4):558-65. 82. Yu M. Recent developments of the PET imaging agents for metabotropic glutamate receptor subtype 5. Curr Top Med Chem 2007;7(18):1800-5. 83. Heiss W and Herlolz K. Brain Receptor Imaging. J Nucl Med 2006;47:302-312. 84. Bezrukiy NV, De Salles AAF, Dahlbom M, DeMarco J, Selch M, Smathers J. ‘‘Multimodality Image Fusion for Stereotactic Radiosurgery Planning and Follow-Up’’ Scientific paper exhibit - 87th RSNA annual meeting, Chicago, IL. Radiology 221(P):223, 2001. 85. Maciunas RJ, Galloway RL Jr, Fitzpatrick JM, et al. A universal system for interactive image-directed neurosurgery. Stereotactic Funct Neurosurg 1992;58:108-113. 86. Roberts DW, Strohbehn JW, Hatch JF, et al. A frameless stereotaxic integration of computerized tomographyic imaging and the operating microscope. J Neurosurg 1986;65:545-549. 87. Barnett GH, Kormos DW, Steiner CP, et al. Use of frameless, armless stereotactic wand for brain tumor localization with 20d and 3-D neuroimaging. Neurosurgery 1993;33(4):674-678. 88. Dusick JR, Esposito F, Kelly DF, et al. The extended direct endonasal transsphenoidal approach for nonadenomatous suprasellar tumors. J Neurosurg 2005;102 (5):832-841. 89. Selch MT, Ahn E, Laskari A, et al. Stereotactic radiotherapy for treatment of cavernous sinus meningioma. Int J Radiation Oncology, Biol Phys 2004;59:101-111. 90. Lionberger DR, Weise J, Ho DM, Haddad JL. How does electromagnetic navigation stack up against infrared navigation in minimally invasive total knee arthroplasties? J Arthroplasty 2008;23(4):573-80. 91. De Salles AAF, Pedroso AG, Medin P, Agazaryan N, Solberg T, Cabatan-Awang C, Espinosa DM, Ford J, Selch MT. Novalis Shaped Beam and Intensity Modulated Radiosurgery and Stereotactic Radiotherapy for Spine Lesions. J Neurosurg 2004;101Suppl 3:435-40. 92. De Salles AAF, Melega WP, Lacan G, et al. Radiosurgery with a 3 mm collimator in the subthalamic nucleus and substantia nigra of the vervet monkey. J Neurosurg 2001;95:990-997. 93. Anzai Y, Lufkin R, De Salles AAF, et al. Radiofrequency ablation of brain tumors using MR guidance. Min Invas Ther & Allied Technol 1996;5:232-242. 94. Kim CW, Lee YP, Taylor W, Oygar A, Kim WK. Use of navigation-assisted fluoroscopy to decrease radiation exposure during minimally invasive spine surgery. Spine J 2008;8(4):584-90. 95. Villavicencio AT, Burneikiene S, Bulsara KR, Thramann JJ. Intraoperative three-dimensional fluoroscopy-based computerized tomography guidance for
General imaging modalities: basic principles
percutaneous kyphoplasty. Neurosurg Focus. 2005;15: 18(3):e3. 96. Hamilton AJ, Lulu BA, Fosmire H, Stea B, Cassady JR. Preliminary clinical experience with linear acceleratorbased spinal stereotactic radiosurgery. Neurosurgery 1995;36(2):311-9. 97. Adler JR Jr. Image-guided frameless stereotactic radiosurgery. In: Maciunas RJ editor. Interactive Image-guided neurosurgery. Park Ridge, IL: American Association of Neurological Surgeons; 1993. pp. 81-89. 98. Ryu S, Jin R, Jin JY, Chen Q, Rock J, Anderson J, Movsas B. Pain control by image-guided radiosurgery for solitary spinal metastasis. J Pain Symptom Manage. 2008;35(3):292-8.
17
99. Gorgulho A, De Salles AA, Frighetto L, Behnke E. Incidence of hemorrhage associated with electrophysiological studies performed using macroelectrodes and microelectrodes in functional neurosurgery. J Neurosurgery 2005;102(5):888-96. 100. Gorgulho A, Juillard C, Uslan DZ, Pegues D, Tajik K, Aurasteh P, Behnke E, De Salles AAF. Infection following deep brain stimulator implantation performed in the conventional versus in the MRI operating room. J Neurosurgery, in press. 101. De Salles AAF, Bittar G. Thalamic pain syndrome: anatomical and metabolic correlation. Surg Neurol 1994;41:147-51.
267
24 Image Reconstruction and Fusion B. A. Kall
A wide variety of digital radiological images may be used in stereotactic and functional surgery. These data are generally computed tomography (CT) and variations of magnetic resonance imaging (MRI, MRA, MRV, fMRI etc). Stereotactic data may also be incorporated from other imaging sources like positron emission tomography (PET), single photon emission computed tomography (SPECT) and ultrasound. The variety of applicable imaging sources will continue to grow. Images are created on an image scanning device and transferred to an image-guided system or other off-line computer system for image processing and procedure planning. These systems decode the images, offer a variety of image processing techniques including image reconstruction of two-dimensional slices into a threedimensional volumes, multimodality (CT and MR) or monomodality (T1 MR to T2 MR) image registration and image fusion. Single imaging volumes and/or fused imaging volumes may then registered to the coordinate system of an image-guided device for the surgical procedure or intervention or used postoperatively to confirm an outcome. Registration and Fusion are two image processing techniques that are often used interchangeably, but are not the same. Image registration is a prerequisite for image fusion. Registration is the process of determining a spatial transformation between two coordinate systems. Each imaging scanner produces images in its own three-dimensional coordinate system. Spatial transformations may be calculated between two or more imaging volumes (image registration), between imaging volumes and a stereotactic
#
Springer-Verlag Berlin/Heidelberg 2009
device (stereotactic registration) or between two or more individual imaging volumes and then to a stereotactic device (image registration followed by stereotactic registration). Image fusion is the process of merging or overlaying two or more registered imaging volumes into a hybrid image. The main purpose of nearly all stereotactic planning and intraoperative image-guidance systems is to decode and reconstruct a set of two-dimensional slices into a three-dimensional volume, transform coordinate spaces between one (or more) imaging database(s) and a stereotactic device to precisely aid the stereotactician in the planning and performance of the surgical procedure or treatment. This chapter will review various image reconstruction, registration and fusion techniques utilized in stereotactic and functional surgery.
Introduction to Three-dimensional Radiological Imaging A typical image-guided radiological image dataset is a collection of (contiguous) two-dimensional slices. A medical image scanner defines the location, size and orientation of each slice in its own coordinate system (S). Each slice is typically 512 columns (X) by 512 rows (Y), denoted as 512 512. Some scanners produce slices with a smaller or larger number of rows and columns like 256 256, 128 128 or 1024 1024. Each dot in a slice at a particular row, column intersection is known as a pixel (picture element) and has a two-dimensional size in X and Y measured in millimeters (mm).
336
24
Image reconstruction and fusion
Pixels do not have to be square (same size in X and Y), but usually are collected in that manner for stereotactic procedures. Each pixel is represented internally as a number which represents a gray scale intensity or components of RGB (red, green blue) color so it may be displayed on a computer graphic display. The intensity or color value in some manner is a function of the type of scanner on which the image was produced (e.g., CT intensities are represented in Houndsfield units which are a measure of the attenuation of X-rays). Each image slice (and therefore each pixel) has a location (Z) and thickness in the coordinate system of the scanning device (both usually represented in millimeters). Each component in a medical image is a threedimensional cube known as a voxel (volume element). Each voxel has an X, Y, and Z coordinate and three-dimensional size. The anatomical orientation of a slice is also defined by the scanner (e.g., transverse, sagittal, coronal, oblique). Most image-guided systems and image processing systems for image registration and fusion prefer an imaging series to be scanned as a contiguous set of slices with a homogenous voxel size. In order to facilitate the transfer of images from a scanner to an external device like an image-guided system, the DICOM (Digital Imaging and Communications in Medicine) standard is utilized (http://medical.nema.org). This transfer may be accomplished with tapes, disks or over a network. DICOM adds a header (collection of tags) onto a computer file containing each image which, in addition to containing information like demographics also contains all the relevant information describing how the images were acquired and spatial information to reconstruct them into three-dimensional volumes. An image-guided or image processing system extracts information from the DICOM headers to detect the orientation, coordinate system and voxel dimensions of the imaging series. The individual slices may then be stacked into a three-dimensional volume matrix. In most
circumstances, the volume is reconstructed into homogeneous three-dimensional voxels by interpolation.
Registration Registration is the process of finding a spatial transformation that maps points from one coordinate system (C1) to those in another coordinate system (C2). Image registration is the process of finding a spatial transformation (T) that maps all points in one image/volume to homologous points in another image/volume (> Figure 24-1).
Methods of Registration One image volume is referred to as the reference (or fixed) volume and a second image volume to be registered to the reference volume is generally referred to as the moving (or working) volume when performing image registration. The second volume may be resampled volumetrically to match the reference volume and overlaid or fused with the reference volume and in some instances resliced to match the original slices in the reference volume following registration. . Figure 24-1 Image registration involves determining a spatial transformation (T) that maps points in the coordinate system from image volume 1 (C1) to a matching point in the coordinate system of image volume 2 (C2)
Image reconstruction and fusion
Image registration algorithms are categorized by the type of transformation method employed. Most registration techniques used in stereotactic surgery are rigid (linear) transformations. Rigid registration assumes the images are isotropic in that the original images are not warped, skewed or distorted. A series of rotations and translations are used to calculate the overall mathematical transformation to spatially register two image datasets. Deformable (or elastic) transformations apply additional warping components that may be applied either globally or locally to subsets of the image volumes. Deformable image registration techniques are used more often for intrasubject image registrations as well as to study disease progression (e.g., tumor volume growth) or to register anatomy that moves (heart). Deformable transformations are a useful method to correct geometric distortions in the original imaging data [1]. This chapter will focus primarily on rigid transformation methods while a deformable transformation component may be introduced if the image data is warped or distorted.
Stereotactic Frame Registration The earliest form of multimodality image registration utilized stereotactic frames [2]. The mathematical properties of registering medical images to a stereotactic frame are covered in much detail elsewhere in this textbook. Briefly, a head frame is rigidly attached to a patients head and a localization device with know geometric properties is attached to the head frame when a radiological scan is acquired. The localizer deposits marks in the imaging data. A computer can then transpose every voxel in stereotactically collected image data into the coordinate system of the particular frame using the geometric properties of the localizer. Multiple scans from multiple modalities may be stereotactically collected.
24
In some systems, the headframe can be removed and precisely reapplied to collect data on different days [3]. Once more than one image volume is registered to the coordinate system of the frame, points, volumes and entire scans may be related to each other using the coordinate system of the stereotactic frame [4–7].
Point Matching Registration Point matching registration, also known as iterative closest point registration, is the simplest method of registering two image volumes without a stereotactic frame [8,9]. Neurosurgeons will recognize point matching registration as the manner by which matching points are registered between locations selected on images (anatomical, stickon or screwed in fiducials) and corresponding locations touched on the patient with an imageguided probe in the operating room. A number of N corresponding anatomical points are identified by the user in each imaging set. The point matching algorithm iteratively alters a spatial transformation by applying a stepwise pattern of translations and rotations. The updated transformation is applied to transform coordinate points from the second set and the sum of squares distance between matching points in the two data sets is recalculated. The algorithm iterates (repeats) for new combinations of rotations and translations until the sum of squares measure is minimized usually measured as a root mean square difference (RMS).
Surface/Edge Matching Registration Surface matching is similar in nature to point matching except that the sets of points are extracted from the surfaces or edges of structures in the image data. These surfaces and edges are commonly determined by thresholding segmentation techniques. Surfaces or edges that are
337
338
24
Image reconstruction and fusion
commonly utilized are the skin surface, cortical surface or edges of the ventricular system for cranial procedures or the surface of vertebra for spinal imaging. This method does not require a set of homologous points (> Figure 24-2).
Surface/edge matching may be performed by variations of point matching as well as algorithms generally knows as the ‘‘hat and head’’ method originally developed for registration of CT, MR, and PET Images [10–14]. Surfaces and edges of
. Figure 24-2 Surface-based image registration: (a) determination of surface in Ictal and Interictal SPECT, (b) determination of MRI cortical surface, (c) SPECT hyperperfusion fused to MRI
Image reconstruction and fusion
matching structures are extracted from the imaging data. One set is represented as a cloud of threedimensional points (‘‘hat’’) and the other as a stack of slices (‘‘head’’). A spatial transformation is iteratively calculated by transforming the hat points onto the head surface until the closest fit of the hat on the head is determined. The square of the distances from points on the hat and its corresponding closest point on the head drawn toward the centroid of the head model is the metric that is minimized. This type of surface registration has also been used previously in registering imaging data to an image guided system [14].
Voxel-based Similarity and Mutual Information Registration One drawback to point and surface matching registration is that manual interaction is necessary to identify either matching anatomical points for point matching registration or surfaces/edges using segmentation methods for surface-based registration methods. These can be both time consuming and are often not precisely reproducible. One user may select slightly different matching anatomical points or the threshold parameters utilized may alter the surfaces or edges extracted using segmentation techniques. Voxel-based registration methods involve calculating a spatial transformation by optimizing some particular measure that may be determined directly from the voxel values rather than geometric relationship in point matching and surface matching registration. Woods et al. [15,16] first proposed voxel similarity registration measures based on the premise that the relative grey values of similar tissues in different images would correspond. Voxel similarity measures were further refined by Hill et al. [17] to define a joint histogram approach that shows changes as the alignment converges so that when anatomical structures overlap the histogram shows clusters around
24
the corresponding grey levels of those structures. Dispersions in the clustering decrease as the alignment approaches convergence. Unfortunately these voxel based-techniques assume that the intensities of corresponding anatomy in corresponding image volumes are linearly related, which is not generally true for multimodality registration. Measuresofdispersioninthetwo-dimensional histograms were then proposed. Hill et al. [18] proposed the third order moment and is a measure of the skewness of the distribution while Collignon [19] and Strudeholme [20] proposed entropy, a measure from information theory [21] as a metric of registration. Entropy, in general, is a measure of dispersion of a probability dispersion. Mutual information registration methods, or relative entropy, were then proposed by Viola and Wells [22–25] as well as Collignon et al. [26–28] at about the same time. Briefly, mutual information (MI) is measure of ‘‘the statistical dependence between two random variables or the amount of information that one variable contains about the other’’ [27]. In medical imaging, one random variable is a voxel intensity in one image volume and the second random variable is a voxel intensity in a second image volume. When two image volumes are geometrically aligned, the mutual information of the image intensity values of corresponding voxels is maximized. This image registration method can be performed without human manual interaction and it does not require any linear relationship of the grey level intensities of matching anatomical structures. Mutual information registration is also implemented as an iterative algorithm. An initial transformation is usually calculated to automatically align the centroids of the imaging volumes. Rotations and translations are applied during each iteration (for rigid transformation), the measure of mutual information is recalculated and the iterations continue in a stepwise optimized manner until the mutual information measure is maximized.
339
340
24
Image reconstruction and fusion
Discussion A number of monomodality or multimodality images may be acquired for a stereotactic or functional procedure. These imaging modalities primarily include CT and variations of MR imaging, but may include a variety of other modalities. The number of image sources considered when planning and performing a stereotactic procedure will continue to grow as will the volume of information from each imaging source. Registering mono or multimodality image volumes to correlate points, volumes or entire imaging volumes and in some instances fusing multiple imaging modalities into a hybrid image volume will continue to be a valuable tool for the stereotactician when planning or performing a stereotactic procedure or confirming the outcome. Rigid spatial registration methods are typically used to perform mono and multimodality imaging registration in stereotactic and functional surgery. Deformable transformation parameters may be introduced if geometric distortion or warping of the imaging data is suspected. Stereotactic frame image registration may be the most accurate and failsafe, but cannot be applied retrospectively. Manual registration methods include point matching and surface alignment registration. Point matching registration is the quickest way to register imaging volumes, but is highly dependent on the selection of matching landmarks. Some imaging systems prefer matching points around the area of interest and some work better when the matching points are dispersed around a larger area of the imaging volume. A small change in a threshold value used for extracting surfaces and edges for surface matching registration may produce slightly differing results. More recently developed voxel-based, and specifically mutual information registration techniques, offer an automated method of image volume registration that requires no assumptions about the underlying characteristics of
the imaging intensity information of a specific imaging acquisition. It is imperative that the result of any image registration is validated and verified. All systems that offer image registration and fusion options offer various mechanism for validation and verification and should be used carefully and comprehensively. Registration algorithms may provide a numeric root mean square (RMS) error estimation or graphically depicted ‘‘zones of confidence’’ to estimate how well the algorithm believes the registration aligns. These numeric and graphical confidence methods should be considered secondary to a comprehensive visual review of the registration accuracy over the entire image volume. The moving image volume is usually resampled to match the orientation of the fixed image volume for validation and verification. Typical visual methods for image registration verification and verification include synchronized comparison of internal and external anatomical locations (> Figure 24-3a) or graphical overlay/fusion techniques such as blending (> Figure 24-3b). The two registered volumes may also be validated and verified by displaying one volume in grayscale and one in color and fusing the registered volumes (> Figure 24-3c). Small angular misregistrations may only become apparent by verifying a large number of both internal and external locations throughout the entire image volume (> Figure 24-4). Errors in image registration and fusion may also occur because of underlying problems in the original imaging data. Image registration methods work best with contiguous, thin slice imaging acquisitions. Patient movement during the scan may result in ghost artifacts in the images. Even with high quality control, medical image scanners, and in particular MR scanners may produce imaging containing distortions because of gradient field nonlinearities [29] that may acceptable for diagnostic purposes, but not for quantitative purposes like image registration and
Image reconstruction and fusion
24
. Figure 24-3 Validation and verification techniques: (a) reviewing corresponding points, (b) blending of CT with stereotactic frame and MR frameless, (c) fusing one volume in grayscale (MRI) and one in color (SPECT)
stereotactic surgery. Image registration methods used in stereotactic and functional surgery generally assume that the images are not warped, skewed or distorted.
A routine diagnostic scan performed at a typical institution may not be a good candidate for an image registration because of the relative thickness of each image. Radiological technicians
341
342
24
Image reconstruction and fusion
. Figure 24-4 Small angular misregistration apparent by comparing matching points on CT and MRI
may also not follow appropriately prescribed imaging protocols and by the time the surgeon may want to do an image registration, it may be too late to repeat the imaging study over again. Automated methods of image registration may occasionally not converge on the correct result because the algorithm was not optimized correctly or because the orientation the two original image volumes are too far apart from one another (e.g., neck flexion significantly different between two clinical imaging volumes which is analogous to a straight CT gantry tilt in one image volume and 10–20 gantry tilt in another). Image reconstruction, registration and fusion techniques perform best with high quality images as inputs.
Summary A variety of images from multiple sources may be considered by the stereotactic and functional surgeon for planning a procedure or confirming results. Images from monomodality or multimodalities may be spatially registered and in some instances fused to provide useful information to improve the ability for a positive outcome. Comprehensive verification and validation of manually
or automatically registered or fused volumes is a mandatory step in performing these techniques. When used appropriately, image reconstruction, registration and fusion techniques may be a powerful tool for the stereotactic and functional surgeon.
References 1. Archip N, Clatz O, Whalen S, DiMaio SP, Black PM, Jolesz FA, Golby A, Warfield SK. Compensation of geometric distortion effects on intraoperative magnetic resonance imaging for enhanced visualization. Oper Neurosurg 2008;62(3):209-16. 2. Goerss S, Kelly PJ, Kall B, Alker GJ, Jr. A computed tomographic stereotactic adaptation system. Neurosurgery 1982;10:375-9. 3. KellyPJ, Earnest R, IV, Kall BA, Goerss SJ, Scheithauer B. Surgical options for patients with deep-seated brain tumors: computer-assisted stereotactic biopsy. Mayo Clin Proc. 1985;60:223-9. 4. Kall BA, Kelly PJ, Goerss SJ, Earnest, F, IV. Crossregistration of points and lesion volumes from MR and CT. In: Proceedings of the seventh annual meeting of frontiers of engineering and computing in health care, Chicago; 1985, p. 692–5. 5. Kelly PJ, Kall B, Goerss S, Alker GJ, Jr. A method of CT-based stereotactic biopsy with arteriographic control. Neurosurgery 1984;14:172-7. 6. Kelly PJ, Kall BA, Goerss S. Computer-assisted stereotactic biopsy utilizing CT and digitized arteriographic data. Acta Neurochirurgica 1984; Suppl 33:233–5.
Image reconstruction and fusion
7. Kall, BA, Kelly PJ, Goerss SJ. Comprehensive Computerassisted data collection, treatment planning and interactive surgery. Proc SPIE Med Imaging 1987;767:509-14. 8. Besl PL, McKay ND. A method for registration of 3D shapes. IEEE Trans Pattern Anal Mach Intell 1992;14 (2):239-56. 9. Zhang ZY. Iterative point matching for registration of free-form curves and surfaces. Int J Comput Vis 1994;13(2):119-52. 10. Pelizzari CA, Chen GTY. Registration of multiple diagnostic imaging scans using surface fitting. In: The use of computers in radiation therapy. North-Holland: Elsevier; 1987, p. 437-40. 11. Pelizzari CA, Chen GTY, Spelbring DR, Weichselbaum RR, Chen CT. Accurate three-dimensional registration of CT, PET and MRI images of the brain. J Comput Assist Tomogr 1989;13:1, 20-26. 12. Pelizzari CA, Levin DN, Chen GTY, Chen CT. Image registration based on anatomical surface matching. In: Interactive image-guided neurosurgery. American Association of Neurological Surgeons, New York: Theime; 1993. p. 47-62. 13. Levin DL, Pelizzari CA, Chen GTY, Chen CT, Cooper MD. Retrospective geometric correlation of MR, CT and Pet images. Radiology 1988;169:817-23. 14. Kall BA. Computerized visualization techniques in stereotactic neurosurgery. In: Textbook of stereotactic and functional neurosurgery. New York: McGraw Hill; 1998. p. 351-5. 15. Woods. RP, Cherry SR, Mazziotta JC. Rapid automated algorithm for aligning and reslicing PET images. J Comput Assist Tomogr 1992;16(4):620-33. 16. Woods RP, Mazziotta JC, Cherry SR. MRI-PET registration with automated algorithm, J Comput Assist Tomogr 1993;17(4):536-46. 17. Hill DLG, Hawkes NA, Harrison NA, Ruff CF. A strategy for automated multimodality image registration incorporating anatomical knowledge and imager characteristics. In: Proceedings of the 13th international conference on information processing in medical imaging. LNCS, vol. 687, Berlin: Springer; 1993. p. 182-96. 18. Hill DLG, Studholme C, Hawkes DJ. Voxel similarity measures for automated image registration, visualization in biomedical computing. Proc SPIE 1994;2359:205-16.
24
19. Collignon A, Vandermeulen D, Suetens P, Marchial G. 3D multimodality medical image registration using feature space clustering. In: Proceedings of the first international conference on computer vision, virtual reality and robotics in medicine. LNCS, vol. 905, Berlin: Springer; 1995. p. 195-204. 20. Studholme S, Hill DLG, Hawkes DJ. Multiresolution voxel similarity measures for MR-PET registration. In: Information processing in medical imaging. Dordrecht: Kluwer; 1995. p. 287-98. 21. Cover TM, Thomas JA. Elements of information theory. New York: Wiley; 1991. 22. Wells WM, III, Viola P, Kikinis R. Multi-modal volume registration by maximization of mutual information. In: Medical robotics and computer assisted surgery. New York: Wiley; 1995. p. 55-62. 23. Viola P, Wells WM, III. Alignment by maximization of mutual information. In: Proceedings of the fifth international conference on computer vision (ICCV 95), Boston, MA. IEEE Computer Society Press; 1995. p. 16-23. 24. Viola P. Alignment by maximization of mutual information. Ph.D. Thesis, Massachusetts Institute of Technology, Boston, MA; 1995. 25. Wells WM, Viola P, Atsumi H, Nakajima S, Kikinis R. Multi-modal volume registration by maximization of mutual information. Med Image Anal 1996;1(1):35-51. 26. Collignon A. Multi-modality medical image registration by maximization of mutual information. Ph.D. Thesis, Catholic University of Leuven, Leuven, Belgium; 1998. 27. Collignon A, Maes F, Delaere D, Vandermeulen D, Suetens P, Marchial D. Automated multimodality image registration based on information theory. In: Information processing in medical imaging. Dordecht: Kluwer; 1995. p. 263-74. 28. Maes F, Collignon A, Vandermuelen D, Marchal G, Suetens P. Multimodality image registration by maximization of mutual information, IEEE Trans Med Imaging 1997;16(2):187-98. 29. Wang D, Strungell W, Cowin G, Doddrell DM, Slaughter R. Geometric distortion in clinical MRI systems: part I: evaluation using a 3D phantom. Magn Reson Imaging 2004;22(9):1211-21.
343
23 Neurophysiologic Mapping for Glioma Surgery: Preservation of Functional Areas R. M. Richardson . M. S. Berger
The resection of tumors located within or adjacent to eloquent cortical regions is often necessary to alleviate focal neurological deficits secondary to mass effect and increased intracranial pressure. In addition, there is growing evidence that greater extent of tumor resection correlates with increased time to tumor progression and overall survival for glioma patients [1,2–4]. The surgical aim is to achieve maximal tumor removal without producing permanent morbidity. Neurophysiological mapping of functional areas is critical for minimizing the morbidity associated with removing abnormal tissue from eloquent cortex. Here we present techniques for intraoperative cortical and subcortical stimulation mapping to maximize safe removal of tumors located in motor and language cortex. Cortical stimulation techniques have been adapted from the pioneering methods of Penfield and Boldrey [5], while localization of subcortical motor and sensory tracts was first described by Berger et al. [6].
Asleep Craniotomy with Motor Function Mapping
pathways, it is important to perform both cortical and subcortical stimulation mapping. Regardless of the degree of tumor infiltration, swelling, apparent necrosis, and gross distortion by the tumor mass, functional cortex and subcortical white matter may be located within the tumor itself or the adjacent infiltrated brain [7]. When using stimulation mapping methods to identify subcortical pathways, the surgeon is able to achieve an acceptable risk of permanent motor deficits in patients with gliomas that are within or adjacent to motor tracts.
Preoperative Neurological Evaluation Although motor mapping will often not be useful in patients with severe hemiparesis, if antigravity movement is present preoperatively it is usually possible to stimulate both cortical and subcortical motor pathways intraoperatively. In children younger than 6 years of age, who may have cortical electrical inexcitability, somatosensory evoked potentials must be available and used to identify the central sulcus via phase reversal.
Indications Preoperative Functional Imaging Hemispheric tumors located within or adjacent to rolandic cortex, supplementary motor area (SMA), corona radiata, internal capsule and uncinate fasciculus constitute the major indications for intraoperative motor function mapping. Due to the risk of damaging descending motor #
Springer-Verlag Berlin/Heidelberg 2009
A volumetric MRI scan is obtained preoperatively for use with an intraoperative navigational system. The relation of the tumor to primary motor cortex is assessed by identifying the central or Rolandic sulcus on the rostral cuts of the axial T2-weighted
326
23
Neurophysiologic mapping for glioma surgery: preservation of functional areas
MRI scan. This landmark is always present, regardless of mass effect, and is a reliable marker for the motor strip located within the gyrus directly anterior to the sulcus. The motor cortex can be found on midsagittal cuts by following the cingulate sulcus posteriorly and superiorly to its termination, at which point the motor cortex is directly anterior to this sulcus. On far lateral images, the inferior to mid portion of the motor cortex is localized to a region bisected by a perpendicular line drawn from the posterior corner of the insular triangle. Each of these MRI landmarks is useful for determining the proximity of the lesion to the motor cortex preoperatively. Preoperative functional imaging of primary motor cortex is achieved by magnetic source imaging (MSI), in which the source localization of functional cortical areas by magnetoencephalography (MEG) is coregistered with an anatomic MRI scan [8,9]. MEG is used to detect the magnetic field associated with neuronal activity itself, rather than relying on an indirect correlate such as the hemodynamic response upon which fMRI is based. MEG is generated by dipole currents associated with dendritic excitatory and inhibitory postsynaptic potentials, which produce frequency specific oscillations whose rhythms change upon brain activation. In this way, the somatosensory cortex and motor cortex are reliably localized preoperatively (> Figure 23-1). Resecting brain tumors involves the risk of damaging the descending subcortical motor pathways. Diffusion tensor image (DTI) fiber tracking is a noninvasive MRI technique that can delineate the subcortical course of the motor pathway by modeling three-dimensional local water diffusion along axonal membranes. DTI is used to visualize descending motor pathways starting from a functional cortical site and extending through the corona radiata, posterior limb of the internal capsule and cerebral peduncle [10] (> Figure 23-2). Fiber tracts delineated using DTI can be used to identify the motor tract in deep white matter and define a safety margin around
. Figure 23-1 Magnetic source imaging (MSI) shows the primary motor cortex involving the right second digit. The white tract mesial to the motor peak depicts the subcortical pathway subserving the motor cortex
. Figure 23-2 A diffusion tensor image (DTI) shows the subcortical motor pathway within the cerebral peduncle and upper pons (corticospinal tract) in relationship to a cystic pilocytic tumor involving the upper brain stem. The DTI tract is depicted in white
the tract during tumor resection, a method that has been validated using intraoperative subcortical stimulation mapping of the motor tract and magnetic source imaging [11]. DTI tractography, however, may be limited by tract termination or deviation in regions of peritumoral vasogenic edema and therefore should be used in combination with intraoperative stimulation mapping.
Neurophysiologic mapping for glioma surgery: preservation of functional areas
Surgical Technique After the dura is opened, and the contralateral arm, leg, and face are uncovered to observe for movement, stimulation mapping begins with identification of the motor cortex. A bipolar electrode (5 mm spacing, 60 Hz, 1ms phase duration) is placed on the surface of the brain for 2–3 sec with a current amplitude between 2 and 16 mA. The motor strip is stimulated in the asleep patient with a starting current of 4 mA and increased in 2 mA increments until a motor response is visually identified. A current above 16 mA has never been necessary to evoke sensory or motor responses and should be avoided [12]. The current is reduced to 2 mA when stimulating the awake patient and is raised in 1 mA increments for eliciting responses from both the motor and sensory cortex. At this point, cold Ringer’s lactate solution should be available for immediate irrigation of the stimulated cortex should a focal motor seizure develop. This will abruptly stop the seizure activity originating from the irritated cortex, without using short-acting barbiturates. Multichannel electromyography recording may be used for greater sensitivity in detecting muscle movements, allowing the use of a lower stimulation current and decreasing the risk of stimulation-induced seizure activity. Stimulated brain sites are marked with sterile numbered tickets (> Figures 23-3a and b). First, the inferior aspect of the rolandic cortex is identified by eliciting responses in the face and hand. For leg motor cortex, a strip electrode may be inserted along the falx and stimulated using the same current applied to the lateral cortical surface to evoke leg motor movements. This maneuver is safe because of a lack of bridging veins between the falx and the leg motor cortex. Similarly, a subdural strip electrode may be used to locate the motor strip when the craniotomy is near but not overlying this cortex. Once the motor cortex is defined, the resection proceeds with identification of the descending
23
tracts using similar stimulation parameters. Functional white matter axons are depolarized using the same current parameters applied to the cortex (> Figures 23-4a and b). When movements or paresthesias are evoked, the resection should cease because of the close proximity of intact functional pathways (current spread with bipolar stimulation is 2–3 mm from the electrode contacts). Following completion of tumor resection, a final stimulation of previously identified cortical motor sites confirms that underlying functional tracts have been preserved, which is equally valuable in cases where subcortical responses are not obtained. The presence of intact cortical and subcortical motor pathways implies that any surgery-related deficit would likely be transient, with resolution in days to weeks. In the senior author’s experience with surgery for hemispheric gliomas within or adjacent to the rolandic cortex, patients whose subcortical pathways were identified with stimulation mapping were more prone to develop an additional (temporary or permanent) motor deficit than those in whom subcortical pathways could not be identified (27.5% vs. 13.1%) [23]. Although motor deficits that lasted more than 3 months occurred in 7.4% of the patients whose subcortical pathways were identified, compared to 2.1% of those without subcortical responses, very few patients have been left with a dense paresis using this method.
Awake Craniotomy with Language Function Mapping Indications Due to the discrete localization of essential language areas in the individual patient and great variation in their location across the population, an awake craniotomy with cortical stimulation mapping is indicated for any lesion involving the dominant temporal, mid- to posterior frontal, and mid- to anterior parietal lobes. Stimulation
327
328
23
Neurophysiologic mapping for glioma surgery: preservation of functional areas
. Figure 23-3 Intraoperative mapping of the motor cortex involving the hand and shoulder is shown (a). The tumor is seen on the sagittal MRI scan to be located within the supplementary motor area (b)
mapping has shown that multiple discrete areas in perisylvian cortex of the dominant hemisphere as essential for language functions, with separation of areas for different aspects of language, including naming in two languages, different semantic classes, naming compared to reading, and language from verbal memory [12]. Sites where stimulation repeatedly interferes with naming have been localized to focal areas of approximately 1 cm2 in the dominant hemisphere cortex, frequently with one such site in perisylvian inferior frontal cortex and several others in temporoparietal cortex [13]. The exact location of these sites in the language dominant hemisphere varies substantially across the patient
population, particularly in temporoparietal cortex, with random distribution within the temporal lobe and in the inferior posterior frontal and anterior inferior parietal lobes. Additionally, no specific region of the temporal lobe cortex may be found to be essential for language function in many patients (55% in the study referenced above). With regard to bilingual patients, cortical stimulation studies have demonstrated the existence of both shared and distinct languagespecific cortical centers. In one study of 25 bilingual patients, primary and secondary language representations were similar in total cortical extent, but differed in anatomical distribution [14].
Neurophysiologic mapping for glioma surgery: preservation of functional areas
23
. Figure 23-4 Enhancing tumor is seen within the thalamus distorting the internal capsule laterally (a). The intraoperative map is shown following resection of the thalamic tumor and identification of the subcortical motor pathways within the internal capsule involving the upper extremity (b)
Secondary sites were located exclusively in the posterior temporal and parietal regions, while sites for the primary language were found throughout the mapped region. Other studies in bilingual patients have demonstrated both common and separate cortical anomia areas for both languages in temporoparietal and frontal areas [15], anomia sites for the second acquired language that were always colocalized with anomia sites in the first acquired language [16], and anomia sites for each language that were
always distinct and separate [17]. For these reasons, it is advisable to individually speech map each language in which the patient is fluent.
Preoperative Neurological Evaluation Patients with dominant hemisphere tumors in close proximity to language sites are ideal candidates for an awake craniotomy. Note that patients
329
330
23
Neurophysiologic mapping for glioma surgery: preservation of functional areas
with significant vasogenic edema and mass effect from their tumor may not be candidates for an awake craniotomy because of the potential for cerebral herniation out of the dural opening. Despite the use of osmotic diuretics, awake patients are at risk for developing alterations in arterial CO2 which may compromise the safety of the planned craniotomy and tumor resection. Swelling, herniation, and contusion may occur, resulting in termination of the procedure. The left hemisphere is dominant for language in 85% of the population, with rightsided dominance present in 6% and bilateral representation present in 9%. Ninety-eight to ninety-nine percent of right-handed individuals have left-sided language dominance. Thus, a Wada (intracarotid amytal) test is used almost entirely for verifying cerebral dominance in lefthanded patients. Those patients who undergo intraoperative mapping for language sites should be preoperatively tested for language errors by presenting the individual with a series of visual slides with common objects and words to be named and read, respectively. After confirming that the face motor cortex and Broca’s area are functional by asking the patient to protrude the tongue and count to ten, the slides of common objects and words are shown. Patients must be able to name common objects with a baseline error rate lower than 25%, with each slide presented at least three times. In patients who have moderate to severe dysphasia in either comprehension or expression, successful language mapping will not be possible. Therefore, these patients may either be asleep during surgery, without any attempt to do more than an internal decompression, or be challenged with steroids and diuretics for 7–10 days and reevaluated regarding their baseline naming error rate. An alternative approach is to biopsy the tumor, confirm histopathology, and radiate the lesion to reduce its size or stabilize its growth, in hopes of producing functional recovery sufficient to allow for intraoperative mapping.
Preoperative Functional Imaging Literature describing the correlation between cortical sites identified with language fMRI and those found by direct cortical stimulation at the time of surgery show variable agreement in localization [18], and therefore fMRI is not routinely used for preoperative planning. DTI of the superior longitudinal, inferior fronto-occipital and uncinatus fasciculi, reconstructed from anatomical landmarks, have depicted tracts occurring mostly at the periphery in high-grade gliomas, but frequently located inside the tumor mass in low grade gliomas [20]. As we move into an era of functional imaging to preoperatively map descending language pathways, it becomes important to keep in mind that the pathways identified are purely anatomical, and may not reflect the true functionality of the axonal bundles identified. In the above study, there was high correlation (97%) between DTI fiber tracking and intraoperative subcortical stimulation, the combined use of which may decrease the duration of surgery, patient fatigue, and intraoperative seizures.
Surgical Technique The initial primary goal during an awake craniotomy is to have a cooperative patient for speech mapping purposes. It is imperative that the patient be kept comfortably sedated when mapping is not being performed. A propofol and remifentanil infusion is titrated for patient sedation during the incision and craniotomy. Once the bone flap is removed, the dura is infiltrated with local anesthetic along the middle meningeal artery. The dura should remain closed until the patient is awake and alert; otherwise, coughing and straining during emergence from propofol may cause the brain to herniate outward, especially if tumor edema and mass effect are present. All sedatives are then discontinued to restore the
Neurophysiologic mapping for glioma surgery: preservation of functional areas
patient to an awake, cooperative state. During cortical mapping of language function, no sedatives are administered. If seizures occur during cortical mapping and are not controlled with cold Ringer’s lactate solution, propofol can be given for seizure suppression. After the motor pathways have been identified, the electrocorticography (ECoG) equipment is placed on the field and attached to the cranium. ECoG is used to monitor for after discharges induced by bipolar electrode stimulation of the cortex (> Figure 23-5). The presence of after discharge potentials indicates that the stimulation current is too high and must be decreased by 1–2 mA until no after discharge potential is present following stimulation. Using the ideal stimulation current, object-naming slides are presented and changed every 4 sec, and the patient is expected to correctly name the object during stimulation mapping. The answers are carefully recorded, and each cortical site is checked three times to ensure that there is no anomic or dysnomic stimulation-induced error. All cortical sites essential for naming are marked on the surface of the brain with sterile numbered tickets. Additionally, the patient is asked to count from 1 to 50 while the stimulation probe is placed near the inferior aspect of the motor strip . Figure 23-5 Electrocorticography equipment is shown in place, and electrodes are recording during stimulation of the cortex for after-discharge potentials
23
to identify Broca’s area. Interruption of counting (complete speech arrest), without oropharyngeal movement, localizes Broca’s area. Speech arrest is usually localized to the area directly anterior to the face motor cortex within a few centimeters. On occasion, however, stimulation-induced speech arrest can be found anteriorly in the pars opercularis or above the face motor cortex in the inferior frontal gyrus. Throughout language mapping, ECoG is continuously monitored for after discharge spikes to alleviate the possibility that naming errors are caused by the propagated effects of current spread or ongoing cortical depolarization (> Figure 23-6a–d). Subcortical stimulation may also be used for detection of eloquent white matter bundles which are essential for language function [19]. Routine use of subcortical language site identification has been reported to result in the identification of language related cortical sites in 59% of patients [20]. In the group of patients in whom a subcortical language site was identified during resection, the likelihood of developing a permanent deficit was 3.8% (7% in patients with a preexisting language deficit), independent of histology and location. When no subcortical sites were found at the time of surgery, no permanent deficits were noted, indicating that when a subcortical response is reliably detected, resection must stop. Cortical stimulation studies have shown that the distance of the resection margin from the nearest language site is likely the most important variable in predicting improvement in preoperative language deficits, duration of postoperative language deficits, and permanence of postoperative language deficits [21]. If the distance of the resection margin from the nearest language site is >1 cm, significantly fewer permanent language deficits occur (> Figure 23-7).
Summary Identification of functional cortical areas in patients with brain tumors provides the neurosurgeon with
331
332
23
Neurophysiologic mapping for glioma surgery: preservation of functional areas
. Figure 23-6 Preoperative MRI scan showing a diffuse infiltrative lesion involving the suprasylvian region of the dominant hemisphere (a). Intraoperative map demonstrating the face motor cortex (#1 and #2) (b). Stimulation-induced speech arrest, i.e., Broca’s area, was not identified anywhere within this region. The post-resection intraoperative map shows preservation of the face motor cortex and resection of the inferior aspect of the somatosensory cortex and the frontal operculum, with sites marked for stimulation on the underlying insular cortex (c). Stimulationinduced anomia was seen at site #30 in the superior temporal gyrus. Postoperative MRI scan showing the resection cavity with the face motor cortex isolated within the resection cavity
. Figure 23-7 1 cm rule: Resection within 1 cm of an essential language site results in permanent deficits in a small percentage of cases. Resecting adjacent brain greater than 1 cm from an essential site will result in temporary deficits; however, no new deficits are seen past 3 months [21]
Neurophysiologic mapping for glioma surgery: preservation of functional areas
the ability to achieve aggressive resections while preserving neurological function. The localization of intracerebral tumors via mapping of functional cortical and subcortical tracts has become an important tool in the preoperative assessment of patients with intrinsic cerebral tumors. The advantages of combining functional imaging information with a surgical navigation system are optimized when combined with intraoperative cortical and subcortical mapping. In the future, based on recent ECoG data showing task-specific changes in the spatial pattern of neuronal oscillations [22], it may be possible to localize function by correlating alterations in these rhythms with various behavioral tasks administered at the time of surgery, rather than by directly stimulating cortical tissue.
References 1. Berger MS, Deliganis AV, Dobbins J, Keles GE. The effect of extent of resection on recurrence in patients with low grade cerebral hemisphere gliomas. Cancer 1994;74:1784-91. 2. Keles GE, Anderson B, Berger MS. The effect of extent of resection on time to tumor progression and survival in patients with glioblastoma multiforme of the cerebral hemisphere. Surg Neurol 1999;52:371-9. 3. Keles GE, Lamborn KR, Berger MS. Low-grade hemispheric gliomas in adults: a critical review of extent of resection as a factor influencing outcome. J Neurosurg 2001;95:735-45. 4. Lacroix M, Abi-Said D, Fourney DR, Gokaslan ZL, Shi W, DeMonte F, Lang FF, McCutcheon IE, Hassenbusch SJ, Holland E, Hess K, Michael C, Miller D, Sawaya RA. multivariate analysis of 416 patients with glioblastoma multiforme: prognosis, extent of resection, and survival. J Neurosurg 2001;95:190-8. 5. Penfield W, Boldrey E. Somatic motor and sensory representation in the cerebral cortex of man as studied by electrical stimulation. Brain 1937;60:389-443. 6. Berger MS, Kincaid J, Ojemann GA, Lettich E. Brain mapping techniques to maximize resection, safety, and seizure control in children with brain tumors. Neurosurgery 1989;25:786-92. 7. Skirboll SS, Ojemann GA, Berger MS, Lettich E, Winn HR. Functional cortex and subcortical white matter located within gliomas. Neurosurgery 1996;38:678-84; discussion 675-84. 8. Benzel EC, Lewine JD, Bucholz RD, Orrison WW Jr. Magnetic source imaging: a review of the Magnes system
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
23
of biomagnetic technologies incorporated. Neurosurgery 1993;33:252-9. Gallen CC, Sobel DF, Waltz T, Aung M, Copeland B, Schwartz BJ, Hirschkoff EC, Bloom FE. Noninvasive presurgical neuromagnetic mapping of somatosensory cortex. Neurosurgery 1993;33:260-8; discussion 268. Berman JI, Berger MS, Mukherjee P, Henry RG. Diffusion-tensor imaging-guided tracking of fibers of the pyramidal tract combined with intraoperative cortical stimulation mapping in patients with gliomas. J Neurosurg 2004;101:66-72. Berman JI, Berger MS, Chung SW, Nagarajan SS, Henry RG. Accuracy of diffusion tensor magnetic resonance imaging tractography assessed using intraoperative subcortical stimulation mapping and magnetic source imaging. J Neurosurg 2007;107:488-94. Ojemann GA. The neurobiology of language and verbal memory: observations from awake neurosurgery. Int J Psychophysiol 2003;48:141-6. Ojemann G, Ojemann J, Lettich E, Berger M. Cortical language localization in left, dominant hemisphere. An electrical stimulation mapping investigation in 117 patients. J Neurosurg 1989;71:316-26. Lucas TH II, McKhann GM II, Ojemann GA. Functional separation of languages in the bilingual brain: a comparison of electrical stimulation language mapping in 25 bilingual patients and 117 monolingual control patients. J Neurosurg 2004;101:449-57. Roux FE, Tremoulet M. Organization of language areas in bilingual patients: a cortical stimulation study. J Neurosurg 2002;97:857-64. Walker JA, Quinones-Hinojosa A, Berger MS: Intraoperative speech mapping in 17 bilingual patients undergoing resection of a mass lesion. Neurosurgery 2004;54:113-17; discussion 118. Bello L, Acerbi F, Giussani C, Baratta P, Taccone P, Songa V, Fava M, Stocchetti N, Papagno C, Gaini SM. Intraoperative language localization in multilingual patients with gliomas. Neurosurgery 2006;59:115-25; discussion 115-25. Roux FE, Boulanouar K, Lotterie JA, Mejdoubi M, LeSage JP, Berry I. Language functional magnetic resonance imaging in preoperative assessment of language areas: correlation with direct cortical stimulation. Neurosurgery 2003;52:1335-45; discussion 1337-45. Duffau H, Capelle L, Sichez N, Denvil D, Lopes M, Sichez JP, Bitar A, Fohanno D. Intraoperative mapping of the subcortical language pathways using direct stimulations. An anatomo-functional study. Brain 2002; 125:199-214. Bello L, Gallucci M, Fava M, Carrabba G, Giussani C, Acerbi F, Baratta P, Songa V, Conte V, Branca V, Stocchetti N, Papagno C, Gaini SM. Intraoperative subcortical language tract mapping guides surgical removal of gliomas involving speech areas. Neurosurgery 2007;60:67-80; discussion 62-80.
333
334
23
Neurophysiologic mapping for glioma surgery: preservation of functional areas
21. Haglund MM, Berger MS, Shamseldin M, Lettich E, Ojemann GA. Cortical localization of temporal lobe language sites in patients with gliomas. Neurosurgery 1994;34:567-76; discussion 576. 22. Canolty RT, Edwards E, Dalal SS, Soltani M, Nagarajan SS, Kirsch HE, Berger MS, Barbaro NM, Knight RT. High gamma power is phase-locked to theta oscillations in human neocortex. Science 2006;313:1626-8.
23. Keles GE, Lundin DA, Lamborn KR, Chang EF, Ojemann G, Berger MS. Intraoperative subcortical stimulation mapping for hemispherical perirolandic gliomas located within or adjacent to the descending motor pathways: evaluation of morbidity and assessment of functional outcome in 294 patients. Journal of neurosurgery 2004;100(3):369-75.
28 Accuracy in Stereotactic and Image Guidance A. Hartov . D. W. Roberts
Stereotactic surgery has traditionally been defined by its unique mechanical instruments, by a subset of neurological and neurosurgical conditions, and by an expectation of accuracy and precision, but at its most fundamental level there is the underlying concept of coregistration. It is this ability to correlate an atlas or imaging study with the surgical field that enables the highly accurate, reliable, and safe procedures of the field today. This chapter will review the mathematical methodology underlying this process and the type of accuracy that can be achieved.
Methods of Stereotactic Registration A Cartesian coordinate system consists of a reference point, the origin, and orthogonal reference directions. One can express the location of any points in such a system in reference to the origin and along the reference directions by using the distances one travels in each to reach the point in question. The same point while fixed in absolute space can be expressed in any number of coordinate systems or frames of reference (frame for short). The coordinates of that point will be a function of the chosen frame. It is sometime useful to be able to express the coordinates of a point in a given frame F1 in those of a different frame F2. This is known as transforming the point from frame F1 to F2 and can be done by applying a translation and rotation to the given point: #
Springer-Verlag Berlin/Heidelberg 2009
2
xF2
3
2
r11 r12 r13
32
xF1
3
2
Dx
3
7 6 76 7 6 7 6 4 yF2 5 ¼ 4 r21 r22 r23 54 yF1 5 þ 4 Dy 5 Dz zF2 r31 r32 r33 zF1 or PF 2 ¼ R PF 1 þ T
ð1Þ
For computational efficiency it is desirable to express points in a homogeneous coordinate system as 4-tuples with the last vector component 1 and simplify the combined rotation and translation operations into one matrix multiplication: 3 2 3 2 3 2 xF 1 r11 r12 r13 Dx xF 2 6 y 7 6 r r r Dy 7 6 y 7 7 6 F1 7 6 F 2 7 6 21 22 23 76 7¼6 7 6 4 zF 2 5 4 r31 r32 r33 Dz 5 4 zF 1 5 1
0
0
0
1
or PF 2 ¼ F 2 TF 1 PF1
1 ð2Þ
The two methods are exactly equivalent. Here we use a notational convention in which one can express the coordinates of a point P which is given in F1, hence PF1, in F2 using a transformation T which takes a point from F1 to F2, hence the notation F2 TF1 . This combined operation for rotation and translation is known as an affine transformation. A transformation consisting only of pure rotation and pure translation is said to be rigid because it maintains the relative position and orientation of points or objects transformed. The principal advantage of using affine transformations is that they can be combined by successive multiplications of the transformation matrices. There are 16 elements to a transformation matrix, but only 6independent parameters: ðDx; Dy; DzÞ and yx ; yy ; yz .
444
28
Accuracy in stereotactic and image guidance
Translation is expressed in the last the rotation terms have the form: 2 1 0 0 6 0 cosðy Þ sinðy Þ x x 6 Rx ¼ 6 4 0 sinðyx Þ cosðyx Þ 2
0 0 cosðyy Þ 0
6 0 6 Ry ¼ 6 4 sinðyy Þ 2
0 sinðyy Þ
1
0
0
cosðyy Þ
column and 0
3
07 7 7; 05
1 3 0 07 7 7; 05
0 0 0 cosðyz Þ sinðyz Þ
0
cosðyz Þ
0
0
1
1 3 0 07 7 7; 05
0
0
1
6 sinðy Þ z 6 Rz ¼ 6 4 0 0
ð3Þ
where yx ; yy ; yz are rotation angles about the respective axes. Any arbitrary rigid transformation can thus be expressed as a series of rotations about the z, x and y axes, followed by a translation: 3 2 1 0 0 tx 60 1 0 t 7 y7 6 tr TUS ¼6 7 4 0 0 1 tz 5 2
0 1
60 6 6 40 2
0 0 1 0 cosðyx Þ sinðyx Þ
0 0 cosðyy Þ 0
6 0 6 6 4 sinðyy Þ 2
1 0
0
0
3
07 7 7 cosðyx Þ 0 5 sinðyx Þ
0 1 3 sinðyy Þ 0 0 07 7 7 cosðyy Þ 0 5
0 0 0 cosðyz Þ sinðyz Þ
6 sinðy Þ z 6 6 4 0
0
0
1 3 0 0 0 07 7 7 1 05
0
0 1
cosðyz Þ
One can say of two frames of reference for which a transformation mapping points from one to the other is known that they are registered. In the following we present in greater detail methods used to compute the transformation between frames of reference, in the context of neuronavigation. We call such procedures co-registration. In the situations of interest to this discussion, we have a patient’s head attached rigidly to a frame which defines a coordinate system. A second coordinate system is defined by a preoperative imaging study, MRI or CT, which also represents the patient’s head. Although the two do not co-exist in a common real space, it is possible to co-register these two spaces using various techniques we will explore in the following.
Point Based Co-registration We assume that we have two sets of homologous points defined in two frames of references. In practice this can be done by using anatomical references, fiducial markers attached to the skin or bone implanted markers. The markers are identified in MRI/CT coordinates visually, and their position in operating room (OR) space can be recorded with a 3D tracker. This set of homologous points can be paired and the transformation matrix that best maps one set onto the other can be computed using several techniques. Given N point pairs, we can state the problem as follows: 2
xF2;1 xF 2;2 ::: xF 2;N
3
2
r11 r12 r13 Dx
3
7 7 6 6 6 yF2;1 yF 2;2 ::: yF 2;N 7 6 r21 r22 r23 Dy 7 7 7¼6 6 7 7 6 6z 4 F2;1 zF 2;2 ::: zF 2;N 5 4 r31 r32 r33 Dz 5 ð4Þ
Based on this definition, it is easy to see that to reverse a transformation one needs only take its inverse.
2
1
1 :::
1
3
0
0
0
1
xF1;1 xF 1;2 ::: xF 1;N 7 6 6 yF1;1 yF 1;2 ::: yF 1;N 7 7; or Pa ¼ F2 TF1 Pa 6 7 6z F2 F1 4 F1;1 zF 1;2 ::: zF 1;N 5 1
1 :::
1
ð5Þ
Accuracy in stereotactic and image guidance
in which we use the a superscript to indicate the augmented matrices of points. There are 12 unknowns in this system of equations, the elements of the rotation sub matrix and the translation components. With 4 points, we could write 12 equations (the last row of this system of equations provides no information) and in principle have enough information to solve for all the unknowns. A simple method to solve this set of equations would be to solve the least squares problem using the pseudoinverse: 1 PaF2 ðPaF 1 ÞT PaF1 ðPaF 1 ÞT ¼ F2 TF 1 ð6Þ While this computation could work in well behaved cases, it often fails to produce a rigid transformation matrix, i.e., a matrix of pure rotation and translation, as described above. Formally, this solution does not enforce the orthonormality of the transformation matrix. The reason for this is that all the elements in the rotation sub matrix are treated as independent variables, while we have seen in equation (4) that they are related and that there are only 3 independent variables. A more reliable method of calculation is available based on the singular value decomposition of the de-meaned point coordinate matrices [1]. This approach minimizes the fiducial registration error (FRE) [2], the RMS distance between point pairs and it enforces orthonormality. The method can be summarized as follows, starting with N points expressed in 3x1 vectors: 3 2 xF1;1 .. . xF1;N 7 6 PF 1 ¼4 yF1;1 .. . yF1;N 5; zF1;1 .. . zF1;N 2 3 xF2;1 .. . xF2;N 6 7 PF 2 ¼4 yF2;1 .. . yF2;N 5; homologous points; zF2;1 .. . zF2;N
28
2
3 2 3 xF 1 xF2 mF 1 ¼ 4 yF 1 5; mF 2 ¼ 4 yF2 5; coordinate means zF 1 zF 2
Compute QF 1 ¼ PF 1 1N mTF1 ; QF 2 ¼ PF2 1N mTF 2 , the de-meaned coordinate vectors obtained with the use of the outer products of a Nx1 ‘‘1’’ vector and the means. Then compute H ¼ QF1 QTF 2 , the outer product of the de-meaned coordinate vectors, from which the 3x3 rotation matrix R is obtained: ½U S V ¼ svdðHÞ; R ¼ V U T using the singular values decomposition of the H matrix. At this stage, it is necessary to check whether the determinant of R is positive, if not the last column of V should be negated and R recomputed. The translation vector corresponds to the difference in the means, expressed in the same frame of reference: t ¼ mF 2 R mF1 , from this we can form the complete transformation matrix: 3 2 F2
6 TF1 ¼ 6 4
Rð33Þ 0
0
tð31Þ 7 7 5 0 1
This basic algorithm produces the best results and is the most reliable, in large part due to the robustness of the singular value decomposition algorithm. A different formulation of this algorithm, albeit exactly equivalent, and based on quaternions is given in [3]. A small but practical improvement can be made to this algorithm, which has to do with the need to correctly pair points in the two frames of reference. In practice, during surgery for example, it is often the case that a larger number of fiducial markers will be placed on the patient prior to preoperative imaging, knowing that not all may be reachable once in the operating room. This leads to the need for some labeling of the markers and some record keeping during the registration procedure, which can be a source of mistakes and result in a botched co-registration procedure.
445
446
28
Accuracy in stereotactic and image guidance
We have devised a method based on the genetic algorithm which will compute the transformation based on randomly selected and randomly paired points, until a ‘‘good’’ result is obtained. This method relies to some extent on a non-symmetric arrangement of the markers, which otherwise could results in registrations seeming correct while rotated by a large angle [4]. Another approach that bypasses the need to match point pairs is to use the iterative closest point algorithm [3], but one must contend with the fact that it can settle on local minima in its optimization procedure. In order to discuss the accuracy of registration methods, it is useful to introduce specific measures of error. An extensive discussion of errors in co-registration can be found in [5]. We have alluded to one already, the fiducial registration error, or FRE. It is defined as the RMS distance between corresponding registration fiducials once the two sets have been merged into the same frame of reference. This is a useful measure in that it is easily obtainable anytime one computes the co-registration. It is also readily verifiable in laboratory tests. The fiducial location error (FLE) is a measure of the quality of the instrumentation, rather than that of coregistration. It is the error distance in the actual location of a fiducial and its location as reported by either the 3D tracking system or as it is obtained from an imaging study. This error will have an impact on the accuracy of the coregistration, although it is not possible to keep track of its effect outside of carefully conducted laboratory experiments. From the surgeon’s point of view the most useful error metric would be the target registration error (TRE), however it is not readily measured under practical OR conditions. The TRE is the error distance between the actual location of a point of interest, the target, which could be the tumor center for example, and its location following co-registration. This is again a difficult quantity to evaluate since a priori it is not possible to point to the tumor center prior to surgery and see how far off it is from its predicted location. One can compute an estimate for it,
based on one’s knowledge of the relative location of the fiducial marker in relation to the point of interest. It is in fact possible to produce an expected distribution of the TRE, given a setup and use this to gauge the quality of a coregistration. An extensive discussion of TRE and associated quantities is given in [5]. As we suggested in the preceding discussion, the error in point-based registration (FRE or TRE) will be a function of how accurately we can locate the points, in other words a function of FLE. The different methods that have been presented over time are known to produce more or less error, based on how well one can identify the registration points. The use of anatomical landmarks has proved to be the least accurate method because of the ambiguity inherent in identifying them on patients. The ambiguity translates into large FLEs in both frames of reference, that is on the patient when in the OR and when identifying the landmarks on the MRI or CT studies. Landmarks that have been used consisted of the medial and lateral canthi of the eyes, both tragii, the bridge of the nose and the tip of the ear projected on the scalp. From the description of these references it is apparent that these landmarks are not points but small regions on the anatomy; there may easily be a few mm of ambiguity in defining their location precisely. This is a large number compared to the 0.35 mm accuracy in locating a single point for a typical 3D tracking system [6]. A more accurate set of fiducial markers can be produced by using specially designed markers that are visible in MR or CT images and which are readily located in space. They are made of a toroid-shaped object soaked with contrast and encapsulated in an adhesive plastic capsule which can be attached anywhere on the skin. These markers define a precise location on the skin and their location can be identified with a ball-end stylus that fits exactly in the toroid center. This generally constitutes a significant improvement over the landmark-based registration but it still suffers from FLE due to the motion
Accuracy in stereotactic and image guidance
of the skin in relation to the cranium. This is true during the MRI acquisition in which the patient is resting on his back and applying some tension of the skin, and in the OR where the head clamp attachment points may be near a marker and pull it slightly. It is also possible for an inexperienced user to apply some force with the stylus on the markers while conducting the co-registration measurements. In carefully conducted in vivo experiments in a CT suite, Lunn et al. were able to demonstrate a TRE of 1.5 mm [7]. Although early on this was probably a best case result, typical results in the operating room today are approaching this level. The method recognized as the most accurate consists of implanting specially designed markers in the bone so that they are rigidly coupled. The markers consist of a threaded stem which is screwed into the bone through a small incision in the skin and a concave spherical receptacle on the exposed end which is designed to receive a stylus tip. The center of the spherical shape defines uniquely and very precisely the registration point. The marker are designed to be visible in MR or CT and their appearance in imaging studies is large enough that an accurate estimate of the spherical center can be obtained by analyzing the curvature of the receptacle end, provided the images are of a sufficient voxel resolution. This method produced the best results, but it is not widely accepted in practice, due to the discomfort and risks associated with it. Using this technique, Maurer et al. have demonstrated clinical TREs of 0.74 0.44 mm with CT, and 1.25 0.45 mm with MRI [8], the difference in accuracy between the two being accounted for the higher spatial resolution CTs.
Surface Based Co-registration It is possible, rather than using discrete reference points (fiducials) to extract the surface of an object of interest such as a patient’s head in two frames of reference and to compute a
28
transformation matrix that best ‘‘matches’’ spatially these two surfaces. This approach implies means of representing and capturing surfaces. It also implies a measure of the spatial match between two surfaces. Furthermore the nature of the computation will be very different since we cannot base it on co-locating discrete points in two spaces and use well tested direct solutions to the least squares problem. The reason being that while we may have many points representing a surface, we do not have a correspondence between these points, nor does such a correspondence necessarily exist since surfaces can be interpolated differently between the points that represent them. Surfaces can be described parametrically as a smooth continuous function, or they can be discretized as a set of points. To produce a surface from a set of points one can view them as forming the vertices of triangular facets, which together form a surface. One sometimes refers to this construct as a surface mesh. In computer applications surface meshes are in widespread use, from computer graphics to engineering CAD systems and a large body of algorithms exists to process surfaces thus described. Surface digitization of a physical object can be achieved by several means. One can simply move the tip of a tracked stylus back and forth on the surface of interest and record points in an irregular pattern until enough points have been gathered [9]. This method lacks predictability and may result in unevenly and incompletely digitized surfaces if one is not attentive. A better method that has been used consists of using a separately tracked laser scanner. The surface is digitized in the coordinate system of the scanner, the scanner’s position in space is itself known and it is therefore possible to register the digitized surface points with the desired frame of reference. Such scanners come with very good resolution and produce very high point densities and thus describe the surface very well [10]. Another method that has been used to digitize surfaces is stereopsis. It is possible with two images of the
447
448
28
Accuracy in stereotactic and image guidance
same scene taken from two separate points of view and with the same features identifiable in both to locate in relation to the cameras the 3D coordinates of these features in the scene. Most operating microscopes are binocular devices and allow one to attach video cameras on both optical paths. Using such an arrangement, it has been possible to digitize cortical surfaces, for example [11]. This approach has the distinct advantage that it is readily integrated physically into an operating microscope, which is normally used in neurosurgery. Extracting surface information from a preoperative imaging study, MRI or CT, is an easier problem, generally, in that segmenting the boundary between skin and air is a relatively simple programming task, usually solved by using thresholding, with some additional processing to smooth the resulting surface. One can follow this step with the marching cubes algorithm [12] which will produce the mesh describing the surface from the list of pixels corresponding to the skin/air interface. It is usually also desirable to decimate such a surface since it tends to have a very high point density. Given two surface descriptions, in the form of points, triangular patches or parametric formulation, the iterative closest point algorithm (ICP [3]) can produce a transformation matrix that minimizes a distance metric between the surfaces. Its operation is elegant and its convergence fast, compared to other optimization algorithms. We summarize the algorithm here. Given a set of points P ¼ fpi g; i ¼ 1:::N, measured from a physical object (e.g., a patient’s head or an object of interest digitized in a machine shop), and given a computer representation of the same object denoted X consisting of a 2nd set of points, it is possible to compute the shortest distance of any point pi in the set P from the surface in question by defining it as the smallest distance between that point and any point on the surface X denoted dmin ðpi ; X Þ. The specific details of the computation of this distance depends
on the internal representation of the surface X, it could be a parametric surface or a triangulated set of points, for example, nevertheless the distance can be computed. As a byproduct of computing the distance dmin ðpi ; X Þ, one can also record the set of points Q ¼ fqi g; i ¼ 1:::N corresponding to the closest points in X to the points pi in P. With these matched point pairs, one can compute the optimal transformation that will co-register points in P with their closest point matches in Q using the same algorithm we presented in the point based co-registration discussion. At this stage we can transform the initial set of points P0 to a new set P1 (here the index indicates the iteration number) and repeat the process until no improvements are obtained. This algorithm can be summarized as follows: 0) 1) 2) 3) 4)
Starting with K=0, given pi;k 2 Pk and X Compute dmin ðpi;k ; XÞ; i ¼ 1:::N and qi;k 2 X Compute Q TP which merges best pi;k qi;k Q Compute Pkþ1 ¼ TP Pk If RMS dmin ðpi;kþ1 ; X Þ > threshold go to 1)
One should note that this algorithm will work on two sets of points as well as surfaces. This makes it useful in point based registration as well as surface based registration. However, the biggest drawback of this algorithm is that it will deterministically drive the optimization procedure towards a local minimum corresponding to the starting point. Because of this it is necessary to start with different initial transformations Q TP which will reach different local minima, and increase the probability of finding the global optimum. While it is rare that the algorithm fails to converge properly, it can happen. The reported accuracy obtained in surgery for patient registration based on the digitization of the head surface is on the order of 2.4 1.7 mm typically [13].
Accuracy in stereotactic and image guidance
Volume Based Co-registration Volumetric data is becoming increasingly available from many sources. MRI and CT are well known sources, but improved technology has made 3D ultrasound more widespread and it is becoming a valuable source of intraoperative data. Given the significantly lower cost of 3D ultrasound, compared to MRI or CT, and given its simple and fast use in the OR environment, its integration in neuronavigation systems is seen as a potential replacement for intraoperative MRI, for example. Given different sources of 3D imaging data of a same patient, it is sometime desirable to co-register these volumes, so as to overlap corresponding regions. To achieve this, one could use the methods we have outlined above, point based or surface based co-registration and probably with acceptable results when the methods are realizable. This is particularly likely to work well if the identified features are correctly recognizable in both modalities and, most importantly, if a rigid transformation is suitable to achieve the match. An example of such a situation would be trying to co-register two CT studies of a patient’s head at different times. There are some situations in which this is not possible, while the need for co-registering remains. A few examples of this would be that two MRI or CT studies of the same patient do not capture the same regions and only have a partial overlap. One can have, for example a MRI that covers a patient’s head completely and a series of a few high resolution CT slices that were selected to cover a tumor and omit the rest of the head. The CT in such a case represents a small slab of the volume encompassed by the MRI. Another situation that may occur is when trying to coregister 3D ultrasound data with MRI. Here the ultrasound acquired through a craniotomy will not have any of the surface features of the patient.
28
In practice, a physician can usually identify features in both imaging series, orient them properly and produce an overlap of the two volumes manually, if necessary. To produce this same result automatically represents some difficulties. The main reason for this difficulty is that generally, the numerical values of voxels corresponding to the same anatomical features are not the same in both imaging series. This is true to a small extent when working with the same type of images (e.g., CT with CT) in both studies, but it is much worse when dealing with different imaging modalities. The physical properties that are behind the various imaging technologies (e.g., density, proton density, acoustic impedance) result in different distributions of pixel values for different tissue types. When looking at images from the same patient in two modalities (e.g., MR and CT), even though corresponding features have different pixel values, identifiable regions (e.g., gray matter, white matter, bone, ventricles) tend to have uniform or at least consistent pixel values (mean variance) within modalities even if these differ between modalities. One can in fact establish statistically this correspondence between pixel values and use this knowledge to align image pairs or volumes. A statistical quantity called the mutual information between two pixel or voxel sets represents the degree of congruence between the two given a transformation that aligns them. For an understanding of mutual information based algorithms, one needs to start with the concept of information in an image. The concept is closely related to that of entropy, as defined by Shannon [14] in the context of communication theory: H¼
N X i¼1
pi log
N X 1 ¼ pi log pi pi i¼1
The probabilities pi in this equation represent the likelihood of an event (e.g., a pixel of a given value) occurring and H is the entropy or
449
450
28
Accuracy in stereotactic and image guidance
information content of a given set of pixel (e.g., image). If an image is made of all pixels having the same value then its information content is 0 since the probability of every pixel value pi is N/N = 1 (using the frequency of occurrence as the measure of probability). The mutual information between two images is a measure of how much we can predict about image B given information from image A. Clearly this is maximized if images A and B are identical. One can predict the value of all the pixels in B if one knows A. One choice (there are several possible choices, see [15] for a detailed presentation) of a formal definition of mutual information is the Kullback-Leibler distance: IðA; BÞ ¼
X a2A b2B
pða; bÞ log
pða; bÞ pðaÞpðbÞ
where p(a,b) is the joint distribution of pixel values, the probability of pixel values a and b occurring simultaneously at the a same image location in images A and B, while p(a) and p(b) are the probabilities of a occurring in A and b occurring in B. Note that if A and B have independent probability distributions, e.g., knowing A has no predictive value regarding B, then p(a,b) = p(a) p(b). One can align two images or volumes by using mutual information as the objective function to maximize while trying different transformation matrices. Although conceptually this is a simple notion, there are many ways to achieve this and the methods that have been presented tend to be specific for certain types of images. A number of problems need to be worked out which will have an impact on the operation of the algorithm. These are: (1) the nature of interpolation when matching pixels or voxels, (2) the optimization method, (3) remaining within the range of convergence. When rotating slightly one image in relation to another, the pixel centers do not correspond exactly and one is left with a choice of how to establish the pixel value from the rotated
corresponding to the unrotated reference image. Another situation in which a direct correspondence between pixels is not readily defined is when the imaging modalities have different resolutions or pixel sizes. One can chose the nearest neighbor method, or one can interpolate the pixel values using any number of methods, with higher methods giving smoother results. The method selected will have an impact on the convergence of the optimization and more importantly on the type of optimization technique that can be used. Interpolation with higher order methods (splines) results in very smooth changes of the objective function given small changes in the transformation matrix, while the nearest neighbor method often results in a very ‘‘jumpy’’ objective function. The tradeoff here is the computational effort. The methods of optimization that have been used in conjunction with mutual information include most of the well known techniques. Powell’s method and derivative (gradient) based methods work in many implementations that have been presented. With more difficult cases such as aligning ultrasound data with MRI data, for example which have very dissimilar appearances, these techniques seem to readily find local minima from which they cannot escape. Other types of searches can be used in such situations, which are known to avoid this problem, such as the genetic algorithm (GA) and the method of simulated annealing (SA). A serious problem with the mutual information approach, which is present with all algorithms but to which the GA and SA are particularly prone is that the mutual information will be as large and sometime larger when overlapping regions are very small and thus tend to be much more uniform. This happens when shift and rotation are such that only a part of the two images overlaps (e.g., a corner outside of the region of interest, representing air) which results in a seemingly very good agreement between the two images. To avoid this, it is necessary to integrate some a priori information regarding the allowable range of translation and rotation to search during the optimization.
Accuracy in stereotactic and image guidance
Other refinements have been presented which consist of preprocessing the two data sets to improve the performance. Preprocessing generally consists of applying some spatial filtering operation to the two images to smooth them. Wu et al. [16] have reported re-registration accuracy of 1.919 mm and 3.742 mm in two cases when aligning multiple tracked ultrasound images to a preoperative MRI. In one of the early paper to introduce the concept of mutual information for registration, Viola and Wells [17] conducted some experiments in which they obtained errors defined as the variance of registered points with respect to the reference in all three directions. Their results were 1.87 mm in X, 2.22 mm in Y, 14.19 mm in Z, and a combined angular variance 3.05 . These results were based on benchtop experiments and do not necessarily reflect what could be achieved in a clinical setting.
Non-rigid Transformations All the discussion that proceeded assumed that a rigid transformation was to be obtained. This is usually sufficient when aligning images of a patient’s head or co-registering a patient’s cranium with the corresponding preoperative MRI. When dealing with soft tissue data this may not be the case. In such situations, methods collectively classified as non-rigid registration may be required. We will not discuss them at length here, except to briefly outline the two broad categories of methods that are being developed for this purpose. In the first, one can alter the affine matrix so that it includes different scaling factors along the axes. The matrix 2
sx 60 T ¼6 40 0
0 sy 0 0
0 0 sz 0
3 0 07 7 05 1
will applying different scaling factors (sx, sy, sz) on each of the coordinates, resulting in a distorted
28
mapping. It is also possible to include shear parameters in the upper left 3x3 matrix so as to produce transformations in which the coordinates in the destination frame of reference are function of all three coordinates in the source frame: 2
1 4 syx 0
sxy 1 0
32 3 2 3 0 x x þ sxy y 0 54 y 5 ¼ 4 y þ syx x 5 1 1 1
In the example above, which we limited to the 2D case for simplification, we can see that the coordinate x is transformed to x + sxyy in the destination frame due to the off diagonal terms (sxy, syx) in the upper left 2x2 submatrix. Note that this is distinct from the translation operation in that the value of y in the source frame will affect the value of x in the destination frame. It is possible to combine all these terms and obtain a global non-rigid transformation incorporating rotation, translation, scaling and shear terms that will better match the deformation that exists between two sets of voxels. This is generally not a sufficient approach to obtain good matching between the source and destination frames. To better capture deformation, it is necessary to define ‘‘local’’ deformation parameters. One technique that has been presented [18] consists of defining a grid (u, v, w) which can have any desired resolution and which is used to parameterize the elements of the affine transformation matrix. In other words, the rotation, translation scaling and shear parameters can be adjusted locally using the (u, v, w) values and by interpolating them using a cubic spline, for example, as was done in [18].
Discussion and Conclusion The evolution of stereotactic neurosurgery from a frame-based methodology reliant upon frontal and lateral radiographs to the widespread utilization of its capabilities throughout neurosurgery
451
452
28
Accuracy in stereotactic and image guidance
and beyond has been the result of computational resources becoming available in the operating room environment. A direct consequence of having direct access to sufficiently powerful and affordable computing ability has been the growth of transformational stereotaxy, utilizing quantitative methods to accomplish co-registration between high content three-dimensional data-sets. Clinical application today has largely employed rigid transformations based upon pairs of corresponding points identifiable in the datasets to be linked. Such methods are straightforward, reliable, and implementable using existing digitizers. Sufficient efficiency and accuracy have been achieved in this manner to make image-guided systems useful and practical in the operating room. As increasing computational resources become available and the algorithms are developed to support other registration methodologies, these more sophisticated strategies will bring only greater efficiency, lower costs, and improved capability. Different registration methods have their respective strengths and weaknesses, dependent upon the requirements of the digitizing technology employed, the nature of the preoperative datasets, the characteristics of the surgical field, the surgical procedure itself, and the needs of the surgeon.
References 1. Arun KS, Huang TS, Blostein SD. Least-squares fitting of two 3-D point sets. IEEE Trans Pattern Anal Mach Intell 1987;9:698-700. 2. Fitzpatrick JM, West J, Maurer CR, Jr. Predicting error in rigid body, point-based registration. IEEE Trans Med Imaging 1998;17:694-702. 3. Besl PJ, McKay ND. ‘‘A method for registration of 3D shapes.’’ IEEE Trans Pattern Anal Mach Intell 1992;14(2):239-56. 4. Hartov A, Roberts DW, Paulsen KD. ‘‘A comparative analysis of coregistered ultrasound and magnetic resonance imaging in neurosurgery.’’ Neurosurgery 2008;62 3 Suppl 1:91-9; discussion: 99–101. 5. West JB, Fitzpatrick JM. ‘‘The distribution of target registration error in rigi-body point-based registration.’’
6.
7.
8.
9.
10.
11.
12.
13.
14. 15.
16.
17.
18.
In: Kuba A editor. IPMI’99: information processing in medical imaging. Lecture notes in computer science, vol 1613. Berlin: Springer; 1999. p. 460-5. Polaris Optical Tracking System. ‘‘Application programmer’s interface guide.’’ Waterloo ON: Northern Digital Inc.; 1999. Lunn KE, Paulsen KD, Roberts DW, Kennedy FE, Hartov A, West JD. Displacement estimation with coregistered ultrasound for image guided neurosurgery: a quantitative in vivo porcine study. IEEE Trans Med Imaging 2003;22(11):1358-68. Maurer CR, Fitzpatrick JM, Galloway RL, Wang MY, Maciunas RJ, Allen GS. The accuracy of imageguided neurosurgery using implantable fiducial markers. In: Lemke HU, Inamura K, Jaffe CC, Vannier MW, editors. Computer assisted radiology. Berlin: Springer; 1995. p. 1197-202. (http://citeseer.ist.psu.edu/maurer95accu racy.html) Friets E. ‘‘The frameless stereotaxic operating microscope: system analysis, enhancements, and nonfiducial registration.’’ PhD Thesis, Dartmouth College, Hanover NH; 1993. Miga MI, Sinha TK, Cash DM, Galloway RL, Weil RJ. Cortical surface registration for image-guided neurosurgery using laser-range scanning. IEEE Trans Med Imaging 2003;22(8):973-85. Sun H, Lunn KE, Farid H, Wu Z, Roberts DW, Hartov A, Paulsen KD. Stereopsis-guided brain shift compensation. IEEE Trans Med Imaging 2005;24(8): 1039-52. Lorensen WE, Cline HE. Marching cubes: a high resolution 3D surface construction algorithm. Comput Graphics 1987;21:163-9. Raabe A, Krishnan R, Wolff R, Hermann E, Zimmermann M, Seifert V. Laser surface scanning for patient registration in intracranial image-guided surgery. Technique Assessments. Neurosurgery 2002;50(4): 797-803. Shannon CE. A mathematical theory of communication. Bell Syst Tech J 1948;27:379-423, 623-56. Pluim JPW, Maintz JBA, Viergever MA. Mutualinformation-based registration of medical images: a survey. IEEE Trans Med Imaging 2003;22(8):986-1004. Wu Z, Hartov A, Paulsen KD, Roberts DW. Multimodal image re-registration via mutual information to account for initial tissue motion during image-guided neurosurgery. Conf Proc IEEE Eng Med Biol Soc 2004;3:1675-8 (IEEE Cat. No.04CH37558). William PV, Wells M, III. Alignment by maximization of mutual information. Int J Comput Vis 1997;24(2): 137-54. Rueckert D, Sonoda LI, Hayes C, Hill DLG, Leach MO, Hawkes DJ. Nonrigid registration using free-form deformations: application to breast MR images. IEEE Trans Med Imaging 1999;18(8):712-21.
27 Anatomical and Probabilistic Functional Atlases in Stereotactic and Functional Neurosurgery W. L. Nowinski
Introduction Early stereotactic human brain atlases were constructed to support human stereotactic instruments [1]. The advent of computed tomography (CT) and magnetic resonance imaging (MRI) enabled imaging of patients’ brains. At present, direct visualization of stereotactic target structures is feasible [2–7] and their depiction, particularly on 3 Tesla systems, is of high quality [8,9]. MRI acquisitions, though superior to CT scans, result in unpredictable and non-reproducible deformations [10]. There are also certain controversies regarding a mismatch between imaging and electrophysiology as well as a target structure incompleteness in the scans [11–13]. Therefore, despite tremendous progress in diagnostic imaging, the stereotactic atlas, particularly in electronic (computerized) format, is still considered an important aid [8,14–17]. It took about two decades from the publication of the first print stereotactic atlas by Speigel and Wycis in 1952 [18] to have a computerized brain atlas available in a clinical setting in 1974 [19]. After the next two decades, at the end of the 1990s, the computerized brain atlases have become prevalent in neurosurgical workstations [20]. The content, role, and use of the atlas in stereotactic and functional neurosurgery (SFN) have been evolving over time in various aspects. This evolution spans: (1) from print atlases to electronic atlases to atlas databases with deformable atlases to population-based multiple complementary atlases with distributions of anatomy, #
Springer-Verlag Berlin/Heidelberg 2009
function, connectivity, and vasculature as well as with clinical results, (2) from sparse anatomical plates to high resolution volumetric multi-modal atlases at various scales, (3) from atlas-assisted targeting to pre-, intra- and postoperative atlas support, (4) from atlas-assisted identification of a few deep brain targets to that in the whole brain, (5) from a single user, single site/resource tool to web-enabled applications to community-centric atlas building and sharing, and (6) from personal handwritten paper records to softcopy spreadsheets to globally shared databases. This chapter features anatomical and probabilistic functional human brain atlases in SFN. It contains three sections covering: (1) atlases including their comprehensive and up-to-date overview, construction, features, and limitations, (2) atlas-based applications along with atlas use and benefits, and (3) future directions. Through these sections we attempt to present a continuous evolution of the stereotactic human brain atlases and their growing potential, both present and future.
Brain Atlases Numerous human anatomical and functional brain atlases have been created for SFN. We review the print atlases as well as describe computerized atlases and their features. The construction of the Cerefy brain atlases, including the probabilistic functional atlas, is described in more detail along with their features, potential, and limitations.
396
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
Anatomical Brain Atlases Print Atlases A number of stereotactic human brain atlases in print format have been created since the first stereotactic atlas was published by Speigel and Wycis in 1952 [18]. They include: Talairach et al. atlas of deep gray nuclei in 1957 [21], Schaltenbrand and Bailey atlas of the brain in 1959 [22], Andrew and Watkins atlas of the thalamus and adjacent structures with a probability study in 1969 [23], Van Buren and Borke atlas of variations and connections of the thalamus in 1972 [24], Schaltenbrand and Wahren stereotactic atlas of the brain in 1977 [25], Szikla et al. atlas of vascular patterns and stereotactic localization in 1977 [26], Afshar et al. atlas of the brainstem and cerebellar nuclei in 1978 [27], Talairach and Tournoux co-planar stereotactic atlas of the brain in 1988 [28], Ono et al. atlas of the cerebral sulci in 1990 [29], Talairach and Tournoux atlas of stereotactic anatomical correlations for gray and white matter in 1993 [30], and Morel et al. multiarchitectonic and stereotactic atlas of the thalamus in 1997 [31]. The content of print atlases is static, non expandable, and non transferable. These atlases are typically not fully segmented and not completely labeled. A major limitation to their use in clinical practice is the difficulty in mapping the print plates into an individual brain.
Computerized Atlases and their Features Though the print stereotactic brain atlases have been available for almost six decades, these are the electronic atlases that have been integrated with neurosurgical workstations and adopted worldwide by the neurosurgical community. Computerized atlases overcome certain shortcomings of the print atlases, can be processed algorithmically,
and offer new features, such as fully segmented and labeled images, atlas to scan registration resulting in automatic segmentation and interactive labeling of patient’s images, cross-correlated presentation in two- (2D) and three-dimensions (3D), defining regions of interest for analysis, structure searching, and quantification. They also constitute reference frameworks enabling integration of information from multiple sources. Features of Computerized Atlases
The main features of computerized atlases are described and illustrated below. Atlas segmentation. Atlas segmentation refers to parcellation of atlas images or volume into individual structures. Each atlas structure is defined either by specifying its region by contouring it (> Figure 27-1d), color-coding its pixels/ voxels (> Figure 27-1b), or both (> Figure 27-1c). As a result, every location in the segmented atlas belongs uniquely to a certain structure (or to the background). Various parcellations are in use; e.g., the thalamus can be parcellated by Hassler’s [32] (> Figure 27-4b) or Walker’s parcellation [33] (> Figure 27-1b), or simultaneously by both (> Figure 27-15b). Atlas labeling. Atlas labeling refers to assignment of names (or generally some classes) to atlas structures (> Figures 27-1b and > 27-4b). Each structure is assigned a unique name (with synonyms, if necessary) which identifies it. All these names form an index. The existing stereotactic atlases, such as Schaltenbrand and Wahren [25] (> Figure 27-4) and Talairach and Tournoux [28] (> Figures 27-1– > 27-3) use their own nomenclatures. Other nomenclatures, such as Terminologia Anatomica [34], are also applicable for labeling. The labels may be full or abbreviated (to facilitate multiple labeling) placed in 2D or 3D (> Figures 27-1 and > 27-4). Atlas deformation (registration, warping). A deformable atlas can be warped (individualized) to match a patient’s scan. Once individualized, the segmentation and labeling content in the atlas is transferred to the patient-specific data
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
(> Figures 27-1f and > 27-4f). Atlas to data registration is discussed in Sections ‘‘Atlas to scan registration’’ and ‘‘Atlas to data registration.’’ Atlas representation. The segmented and labeled atlases can have various representations including bitmaps (images) with colored structures, bitmaps with delineated (contoured) structures, geometric contours of structures (e.g., in form of splines), 3D geometric surface (polygonal) models, and volumetric models (> Figures 27-1, > 27-3– > 27-5, and > 27-7d). These representations are useful in various atlas-assisted applications. For instance, an atlas in contour representation does not eclipse a scan when superimposed on it (> Figures 27-1f and > 27-4f ) in contrast to that in bitmap representation (> Figure 27-16a); geometric contours can be zoomed in without changing their line thickness which provides a more accurate separation of neighboring structures (> Figure 27-13); surface models are useful for 3D visualization and spatial exploration (> Figures 27-1e, > 27-3, > 27-4e, and > 27-5b); and volumetric models are suitable for non-rigid warping and reformatting in any plane (> Figures 27-7d and > 27-8c). Structure searching. Any structure in the index is searchable to localize it in the original or individualized atlas. The search can be performed in the current atlas image, across entire atlas or multiple atlases. Atlas-assisted quantification. The individualized stereotactic atlas placed in a coordinate system enables reading of stereotactic coordinates (of targets and other locations) and calculating distances. Two main coordinate systems in use are: the Schaltenbrand system [25] with the origin at the midcommissural point (> Figure 27-5a) and the Talairach system [28] with the origin at the anterior commissure (> Figure 27-1). Spatial cross-correlation. The atlas or, generally, a multiple atlas database facilitates spatial correlation across orientations, atlases, and dimensions. A location (e.g., a target point)
27
in a single atlas available in multiple orientations can be correlated across orthogonal planes (> Figure 27-14). Multiple atlases spatially coregistered and cross-correlated enable to examine the same structure or region in various atlases and explore multi-atlas complementarity (> Figures 27-5 and > 27-15a). Correlation can be across all three orientations (on the triplanar) (> Figures 27-2, > 27-5a, > 27-6a, and > 27-15a) or between 2D and 3D (enabling 3D surface or volumetric models to be correlated with the original or reformatted images) (> Figures 27-3a, > 27-15a, and > 27-16a). Computerized Atlases
To benefit from the electronic format, most of the print atlases have been converted into it, including: Schaltenbrand and Bailey atlas [35–37]; Schaltenbrand and Wahren atlas [35,38–40]; Afshar et al. atlas [41]; Van Buren and Borke atlas [35]; Talairach and Tournoux atlas [35]; and Morel atlas [42]. Moreover, electronic versions of the classic Thieme stereotactic print atlases [25,28–30] are included into the Cerefy brain atlas database (Section ‘‘Cerefy atlases’’). Computerized versions of the print atlases vary substantially ranging from a simple, direct digitization of the original printed material to a sophisticated, fully segmented, labeled, enhanced, and 3D extended deformable atlas (see also Section ‘‘Atlas-based applications’’). The existing stereotactic atlases are usually sparse with a variable inter-plate distance, varying from 0.5 to 4.0 mm for the Schaltenbrand and Wahren atlas, and from 2 to 5 mm for the Talairach and Tournoux atlas. However, to enable nonrigid warping or reformatting in an oblique plane, the atlas must be volumetric and of high resolution. Technically, image or contour interpolation of sparse atlases is feasible, enabling reconstruction of 3D models and generation of intermediate sections. For instance, the Schaltenbrand and Bailey atlas [22] was interpolated with 0.5 mm step in [37]. However, interpolation or 3D modeling does not compen-
397
398
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
sate for the intrinsic shortcomings of the original print material (as discussed in Section ‘‘Limitations of anatomical atlases’’). An analysis of the main target structures in the Schaltenbrand and Wahren atlas: the subthalamic nucleus (STN) [43], globus pallidus internus (GPi) [44] and ventrointermediate nucleus of the thalamus (VIM) [45], reconstructed in 3D shows that their shapes are not realistic (see e.g., > Figure 27-6c). To facilitate non-rigid warping and/or reformatting in any plane, some existing atlases have been extended and new stereotactic atlases in 3D geometric and high resolution volumetric representations have been developed [46–49]. A 3D atlas of the STN and its adjacent structures is described in [48]. A deformable atlas of the thalamus, called the Talairach 2000 atlas [47], is constructed by applying 3D geometric modeling to the Talairach and Tournoux atlas. Moreover, the Cerefy atlases contain 3D geometric and volumetric models of cerebral structures (Section ‘‘Cerefy atlases’’). A 3D deformable atlas of the basal ganglia is created from histological images [49]. The left hemisphere, after removal of the frontal and occipital lobes, was cut into 1.5 cm blocks. Each block was then cut into 800 coronal sections of a 70 mm thickness divided into two series. Every tenth section was stained. One series was Nissl-stained with cresyl violet and the other was immuno-stained for calbindin. Eighty structures, including basal ganglia nuclei, fiber bundles and ventricles, were traced on the histological sections. Another histology-based atlas of the thalamus and basal ganglia is built from 86 pairs of coronal sections cut with 0.7 mm intervals [46]. For each pair of sections, one section was stained with Luxol Blue for myelin and the other with a Nissl stain for cell bodies. The sections were manually segmented and labeled with 105 structures.
Cerefy Atlases The most prevalent computerized atlases in SFN are the Cerefy atlases (www.cerefy.com) [20,50– 53]. Their acceptance and adoption by the community resulted probably not only thanks to providing a high quality content, but also by proposing novel atlas-based solutions, and developing sophisticated yet user-friendly tools for planning, intraoperative support, and postoperative analysis (Section ‘‘Atlas-based applications’’). The Cerefy anatomical atlases have been derived from the classic print brain atlases edited by Thieme [25,28–30]. They are available in neurosurgical workstations [20] (as add-on libraries) and also distributed on CD-ROM [54–57] (by the publisher of the original print atlases), see Section ‘‘Atlasbased applications.’’ The Cerefy anatomical brain atlas database contains, among others, the electronic versions of: Atlas for Stereotaxy of the Human Brain by Schaltenbrand and Wahren (SW) [25] and CoPlanar Stereotactic Atlas of the Human Brain by Talairach and Tournoux (TT) [28]. To create these computerized atlases, the original print materials were intensely processed, enhanced, and extended. This processing involved: (1) scanning of the print images and compiling textual materials, (2) full segmentation (contouring or color coding) of all atlas structures, (3) complete labeling (naming) of all atlas structures, (4) arrangement of the atlas images into volumes, (5) atlas checking, correcting, enhancing, and extending, (6) constructing 3D versions, (7) developing various representations in 2D and 3D (bitmap, contour, polygonal, and volumetric), (8) mutual coregistration of all 2D and 3D atlases, and (9) studying atlas properties. Atlas construction required the development of several dedicated and sophisticated tools. To enable integration of the Cerefy atlases with the existing neurosurgical workstations, two add-on atlas libraries with the viewers are
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
developed: Cerefy Electronic Brain Atlas Library and Cerefy Brain Atlas Geometrical Models (www. cerefy.com) [20]. The Cerefy Electronic Brain Atlas Library contains the atlases in image representation and the Cerefy Brain Atlas Geometrical Models comprise the atlases in geometric contour and 3D polygonal representations. Cerefy Talairach and Tournoux Atlas
The TT atlas contains axial, coronal, and sagittal images of gross neuroanatomy derived from a single normal brain specimen. The brain had been sectioned and photographed sagittally, and the axial and coronal orientations were interpolated manually. To create the Cerefy TT (C-TT) atlas, the original print images were digitized with 0.2 mm resolution, and extensively processed, enhanced, and extended as follows: (1) the original grids, rulers, and annotations were removed, (2) each atlas structure was assigned a unique color-coded representation, in contrast to a mixture of contour, color-coded, and texture representations in the print atlas, (3) the left thalamic nuclei and the basal ganglia, not available in the print atlas on the axial and coronal plates, were outlined and color-coded, (4) the right hemisphere cortex on the axial plates was added by mirroring the left hemisphere cortex, and (5) Brodmann’s areas and gyri, which are labeled but not segmented in the print atlas, were segmented and color-coded on the axial plates [50]. The C-TT atlas has continuously been improving in terms of its quality, image [52] and textual content [56], representation, and spatial consistency [53]. Examples of the C-TT atlas in multiple orientations and representations are shown in > Figure 27-1. The GPi correlated across all three C-TT orthogonal planes is illustrated in > Figure 27-2. An example of 2D-3D cross-correlation in the C-TT atlas is presented in > Figure 27-3a. > Figure 27-3b shows an enhanced version of the 3D C-TT atlas suitable for interpolation [58]. Cerefy Schaltenbrand and Wahren Atlas
27
The SW atlas, based on 111 brains, contains photographic plates of macroscopic and microscopic sections through the hemispheres and brainstem. The macroscopic plates with gross anatomy provide the extent of variation of cerebral structures. The microscopic myelinstained sections show the deep brain structures in great detail. To create the Cerefy SW (C-SW) atlas, the microseries along with the overlays were digitized with 0.1 mm resolution. They were processed, enhanced, and extended as follows: (1) the digitized images were rotated (typically within 0.2–0.5 ) to ensure their proper alignment, (2) all structures on the overlays were contoured manually under a high magnification (of up to 20 times), (3) numerous contours open in the original SW atlas were closed, enabling interactive labeling and structure searching (the detailed list of the processed structures is enclosed in the User Guide of [56]), (4) geometric contour and 3D polygonal models were constructed, (5) each contour and polygonal model was assigned a label (or labels) consistently with the SW overlays, (6) the microseries and the contours with the corresponding labels were extended to cover both hemispheres: the axial (Brain LXXVIII, right hemisphere) and coronal (Brain LXVIII, right hemisphere) plates with the corresponding contours were mirrored along the midsagittal (interhemispheric) plane, and the sagittal plates (Brain LXXVIII, left hemisphere) with the contours were replicated for the right hemisphere; the 3D models were also replicated for the other hemisphere, and (7) corresponding sections were aligned and stacked to form a brain volume; the processed material was initially organized into 18 atlas volumes [50], but practically the axial, coronal, and sagittal microseries have been in use. The C-SW atlas has continuously been enhancing (see the User Guide of [56] and Section ‘‘PFAcentric combined atlas’’). Examples of the C-SW atlas in multiple orientations and representations are shown in > Figure 27-4.
399
400
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
Cross-correlation of and complementarity in multiple atlases are shown in > Figure 27-5.
Limitations of Anatomical Atlases Creation of stereotactic print atlases required monumental efforts, particularly that their creators
. Figure 27-1 C-TT atlas in multiple orientations and representations labeled in 2D and 3D: (a) digitized original axial image (note that the Talairach grid (Section ‘‘Talairach transformation’’) encompassing the brain is not exactly rectangular), (b) color-coded axial image with gyri, Brodmann’s areas, and subcortical structures labeled with full and abbreviated names, (c) coronal image in color-coded and contour representations along with the Talairach grid, (d) sagittal image in contour representation, (e) 3D C-TT atlas labeled in 3D, (f) sagittal scan segmented and labeled by the atlas in contour representation
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
27
. Figure 27-1 (Continued)
. Figure 27-2 The GPi correlated across all three orthogonal planes in the C-TT atlas. The label and coordinates at the intersection of the reference lines (the pointed location) are displayed. The description of the GPi is also shown
401
402
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
. Figure 27-3 3D C-TT atlas: (a) subcortical structures of the 3D C-TT atlas cross-correlated with the coronal images of the 2D C-TT atlas; the subthalamic nucleus and its surrounding structures are labeled in 3D, (b) enhanced version of the 3D C-TT atlas with subcortical structures and some white matter tracts (the thalamus is rendered semi-transparent to show its component nuclei)
did not have access to information technology. By today’s standards and requirements, however, some features of these atlases are regarded as limitations. The shortcomings of stereotactic atlases have been well studied [39,43–45,51,59–62]. These studies have been possible thanks to the advances in computers enabling a qualitative and quantitative atlas analysis. This analysis is facilitated by constructing electronic versions of print atlases, building 3D models of structures, developing tools for atlas quantification on the orthogonal planes and in 3D (> Figure 27-6), and formulating and computing measures characterizing the location, size, shape, and mutual relationships of the studied structures [43]. The SWaxial, coronal, and sagittal microseries derived from three different hemispheres are not spatially consistent, meaning that a given point in 3D may belong to more than one anatomical structure. The original printed material is not fully consistent in terms of plates (neuroanatomy),
grid, image-overlay correspondence (see e.g., > Figure 27-4d), and landmarks. Inaccuracies existing across all three SW microseries are clearly visible on the triplanar (> Figure 27-6a) or in 3D (> Figure 27-6c), and those within a given orientation in 3D (> Figure 27-6b). The SW axial plates were not acquired in the intercommissural (but in Reid’s) plane and are rotated 7 clockwise. The resulting inaccuracy for the P.m.i (GPi) 2.0 mm in front of the midcommissural point (i.e., on plate Fa 2.0) is +0.25 mm, where ‘‘+’’ means dorsal and ‘‘ ’’ ventral offset. The atlas inaccuracy for the V.im.e ranges from 0.49 mm (on plate Fp 4.0) to 0.86 mm on plate Fp 7.0; that for the Sth (STN) spans from +0.25 mm on plate Fa 2.0 to 0.86 mm on plate Fp 7.0. Some quantitative verification of the SW atlas was done in [39,59]. It demonstrates that the sagittally sectioned thalamus is 10% larger than the coronally sectioned thalamus and 40% larger than that sectioned axially. Another group
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
of studies (summarized below) quantifies the main stereotactic target structures in the C-SW atlas: STN [43], GPi [44], and VIM [45]. It partly concurs qualitatively though not quantitatively with [59].
27
The 3D models of the STN: 3D-A, 3D-C and 3D-S, reconstructed from the C-SW axial, coronal and sagittal microseries, respectively, are placed in the Schaltenbrand coordinate system, and compared quantitatively in terms of location
. Figure 27-4 C-SW atlas in multiple orientations and representations labeled in 2D and 3D: (a) digitized original axial image, (b) axial image in contour representation labeled with full and abbreviated names, (c) color-coded coronal image, (d) sagittal image in contour and image representations labeled (note a mismatch between the SW image and its corresponding overlay in the brachium colliculi superioris), (e) 3D C-SW atlas labeled in 3D, (f) axial scan segmented and labeled by the atlas in contour representation
403
404
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
. Figure 27-4 (Continued)
(centroids), size (volumes), shape (normalized eigen values), orientation (eigen vectors), and mutual spatial relationships (overlaps and inclusions) [43]. These 3D STN models differ in location, size, shape, orientation, overlap size, and inclusion rate as follows. The 3D-S volume of 207.1 mm3 is 1.27 times larger than that of 3D-A and 1.38 times larger than that of 3D-C. The highest overlap size is between 3D-A
and 3D-S. The highest inclusion rates of 52.5 and 66.6% are for 3D-A and 3D-S. 3D-C has the lowest overlap size and lowest inclusion rates (about 20–30%), meaning that 3D-C is considerably displaced in comparison to 3D-A and 3D-S. The lateral centroid coordinate of 3D-C is 9.18 mm and that of 3D-S is 12.17 mm. Each model has some limitation: 3D-A in orientation, 3D-C in location, and 3D-S in shape realism.
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
27
. Figure 27-5 Cross-correlation of and complementarity in multiple atlases: (a) subthalamic nucleus in the C-SW and C-TT atlases mutually correlated (the axes of the Schaltenbrand reference system along with the coordinates are also shown), (b) complementarity of a 3D highly parcellated thalamus in the SW atlas with a gross neuroanatomy in the TT atlas
. Figure 27-6 Inconsistency in the C-SW atlas: (a) structure mismatch on the triplanar, (b) unrealistic shape of the 3D thalami reconstructed from the coronal contours, (c) unrealistic and inconsistent 3D VIM models reconstructed from axial (in blue), coronal (in red), and sagittal (in green) microseries (dorsal view)
405
406
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
The reconstructed 3D GPi models substantially vary in location, size, and inclusion rate [44]. The centroid of 3D-C is located more medially (15.6 mm) than these of 3D-A (17.5 mm) and 3D-S (18.2 mm), and that of 3D-A is more ventrally ( 2.3 mm) than those of 3D-C ( 0.1 mm) and 3D-S ( 0.4 mm). 3D-S has the smallest volume of 347.3 mm3, 3D-A is 1.18 and 3D-C 1.85 times larger. The highest inclusion rate is for 3D-S (54.3 and 56.3%) and the lowest for 3D-C (28.8 and 30.6%). The shape, orientation, and overlap size are less variable. The reconstructed 3D models of the VIM (> Figure 27-6c), VIM externum (VIMe), and VIM internum (VIMi) also differ significantly in location, size, shape, and inclusion rate [45]. The centroid of 3D-A/VIM differs considerably from those of 3D-C/VIM and 3D-S/VIM. The difference between the centroids of 3D-C/VIM and 3D-S/VIM is in laterality only: this of 3D-C/ VIM is located more medially (11.85 mm) than that of 3D-S/VIM (14.62 mm). 3D-A/VIM has the smallest volume (of 69.00 mm3), 3D-C/VIM is 3.71 and 3D-S/VIM 3.89 times larger. The overlap is also highly variable: 104.88 mm3 for 3D-C/ VIM with 3D-S/VIM, and very low (3.22 and 7.45 mm3) when 3D-A/VIM is involved. The highest inclusion rate is for 3D-C/VIM with 3D-S/VIM (39.10 and 40.97%) and the lowest for 3D-A/VIM with 3D-C/VIM (1.26 and 4.66%). The centroid of 3D-A/VIMe differs noticeably from those of 3D-C/VIMe and 3D-S/VIMe. The difference between the centroids of 3D-C/VIMe and 3D-S/VIMe is mainly in laterality: this of 3D-C/VIMe is located more medially (12.91 mm) than that of 3D-S/ VIMe (16.65 mm). 3D-A/VIMe has the smallest volume (of 49.87 mm3), 3D-S/VIMe is 3.24 and 3D-C/VIMe 3.36 times larger. The overlap sizes are low: 32.72 mm3 for 3D-C/VIMe with 3D-S/ VIMe, and very low (1.32 and 2.01 mm3) when 3D-A/VIMe is involved. The inclusion rates are also low: the highest is for 3D-C/VIMe with 3DS/VIMe (19.53 and 20.29%) and the lowest for 3D-A/VIMe with 3D-C/VIMe (1.19 and 4.01%).
There are substantial differences among the centroids of 3D-A/VIMi, 3D-C/VIMi and 3D-S/ VIMi. This of 3D-A/VIMi is located more anteriorly ( 1.92 mm) than that of 3D-C/VIMi ( 5.02 mm). The centroid of 3D-A/VIMi is located more ventrally (2.88 mm) than those of 3D-C/VIMi and 3D-S/VIMi (each at 5.34 mm). 3D-A/VIMi has the smallest volume (of 19.75 mm3), 3D-S/VIMi is 3.23 and 3D-C/VIMi 4.30 times larger. 3D-A/VIMi practically does not overlap with 3D-C/VIMi and 3D-S/VIMi. The inclusion rates for 3D-C/VIMi with 3D-S/VIMi are medium (32.63 and 43.43%). The overall conclusion from these three studies is that the SW atlas shows inter- and intraorientation spatial inaccuracies. Quantification of these inaccuracies may help enhancing the SW atlas and is useful in atlas registration (see Section ‘‘Combined anatomical-functional atlases’’). The original TT atlas, though constructed from a single brain specimen, is also not spatially consistent. This consistency was defined as uniformity of labeling across all three orientations at the common points and calculated for the entire C-TT atlas, majority of its structures, and cortical areas [62]. It was also analyzed in function of discrepancy measuring the spatial offset in labeling. The C-TT atlas has 27% consistency (three labels are common), 62% partial consistency (two labels are common), and 38% inconsistency. The thalamus with 86% consistency is the most consistent structure. The basal ganglia have a good consistency. The inconsistency of major subcortical gray matter structures is very low for 3 mm discrepancy. The inconsistency of all subcortical structures is higher (17%), caused mainly by a very high inconsistency of the white matter tracts. The entire atlas consistency increases by 20% for 1 mm discrepancy, then constantly grows by 10% for each of 2, 3, and 4 mm discrepancy, and finally slows down to 3% for each of 5 and 6 mm discrepancy. The consistency increase for the cortical areas is higher than that for the subcortical structures.
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
In addition to the above limitations, each of the SW and TT atlases exhibits shortcomings in terms of the landmarks defining its reference system. In the print SW atlas, the anterior commissure (AC) and posterior commissure (PC) landmarks are inconsistent. This atlas provides three sources of information about their location and distance: (1) guide attached to the atlas, (2) microseries plates, and (3) plates showing the unsectioned hemispheres. These sources give three different AC-PC distances [61]. Moreover, on the SW sagittal microseries (used most frequently for planning) the external outlines (semi-circles) of the AC and PC structures are marked only 2.5 mm away from the midsagittal plane (on plate Sl 1.5 the AC and PC are not marked at all, whereas the midsagittal plane, where the AC-PC distance should be measured, is missing). In addition, the intercommissural distance between the marked AC and PC outlines varies, e.g, 20.5 mm on plate Sl 2.5, 20.0 mm on plate Sl 3.5, and 20.5 mm on plate Sl 5.0. From a computerized atlas perspective, there are at least four problems with the original Talairach landmarks [63] (see Section ‘‘Talairach transformation’’ for their definitions). First, not all of them are available in the original atlas. Second, locations of some landmarks contradict their definitions. In particular, on the axial plates, the L and P landmarks are beyond the Talairach grid (which also is not exactly rectangular, see > Figure 27-1a), and the R landmark is not present at all. The atlas plates do not cover the entire Talairach space and, consequently, the AC, PC, S and I landmarks are not available on the axial plates, A and P landmarks are missing on the coronal plates, and L and R landmarks are not present on the sagittal plates. Third, the cortical landmarks are not defined constructively. Fourth, the intercommissural landmarks are located outside their own structures and, despite being defined precisely, their accurate constructions are not easily doable on a scanner console or neurosurgical workstation. To cope with these
27
problems, the modified landmarks were introduced [63] (Section ‘‘Atlas use’’). As a consequence of all these studies, an absolute and direct reliance on the original SW and TT atlases is unsafe and these atlases must be used with a great care and understanding of their features and limitations.
Probabilistic Functional Brain Atlases Despite their great usefulness in SFN, the current computerized anatomical atlases have two major limitations (besides the abovementioned shortcomings). First, they are constructed from a few brain specimens only. Second, these atlases are anatomical, while the actual stereotactic targets are functional. Anatomical variability studies were already included in several stereotactic print atlases and some of them were based on multiple brain specimens. The first stereotactic atlas by Speigel and Wycis provides data on anatomical and radiographic variability [18]. A statistical analysis of variations in skull-brain relationships, endocranical reference system, and intercommissural line is given in the Schaltenbrand and Bailey atlas [22]. The Andrew and Watkins atlas [23] includes a probabilistic study of variability of the thalamic nuclei and neighboring structures. The SW atlas, constructed from 111 brains, presents macroscopic variation of selected structures and provides a possible extent of variation. The outlines on the overlays demarcate both the least common region and overlapping parts in several structures of interest including the STN and GPi. The accompanying tables give measurements made on coronal sections of eight brains with the mean distances from the structure’s center to the midline at different levels. In addition, contours for several ventricles and skulls along with their average outlines are provided. Despite the wealth of material used to construct the
407
408
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
entire SW atlas, the microseries, which are the most often employed, are derived from two brain specimens only. There are several ongoing efforts aimed at developing electrophysiology databases and atlases. A database of subcortical electrophysiology [64] stores stimulation-induced responses and microelectrode recording data coded and plotted along stereotactic trajectories from 106 procedures (88 patients). This database contains only intraoperative data and does not include the best targets. The data points are displayed in 3D as clusters of color-coded spheres. The approach applies nonrigid warping to normalize brains [65] and does not rely on the AC and PC landmarks. It also does not take into account the geometry of electrode. An initial atlas of target points based on 18 patients [66] exploits nonrigid registration for brain normalization [67]. This effort also aims to create a database for storing all pre-, intra-, and postoperative information for the treated patients. The CASS system contains a well-established database of about 2,500 electrophysiological response points from microelectrode diencephalic stimulation and recording collected over a few decades [35]. To overcome both limitations of the current computerized anatomical atlases as well as to avoid a cumbersome dealing with a huge number of electrophysiology numerical values (including point coordinates and various responses), the probabilistic functional atlas (PFA) was introduced conceptually in [20] and algorithmically in [68] as a new addition to the Cerefy family of brain atlases. The PFA algorithm is able to convert thousands of numbers with the best contacts, their location, orientation and size as well as coordinates of the patient-specific landmarks into simple to use 2D maps and 3D volumetric models. The PFA opens new directions not only in planning but also in providing community-centric solutions in SFN.
PFA: Concept and Algorithm The PFA is calculated from intraoperative neuroelectrophysiology, pre- and intraoperative neuroimaging, and postoperative neurological assessment. The PFA algorithm converts the coordinates of the neurologically most effective contacts into probabilistic functional maps taking into account the geometry of a stimulating electrode and patient’s anatomy. This atlas provides the distribution of the best stereotactic targets in a normalized atlas space, enables determining the accuracy of targeting, and facilitates studying properties of functional structures [20,57,68–73] (see also Section ‘‘Atlas use and benefits’’). The PFA algorithm calculates the atlas from the selected best contacts in the following steps: (1) 3D reconstruction of contact coordinates, (2) contract normalization, (3) voxelization of the normalized contacts, (4) calculation of the atlas function, and (5) probability computing [68]. Each best contact is normalized by applying the corresponding patient-specific normalization parameters and placed in the common atlas space. The normalized best contacts are voxelized by a rapid and optimal algorithm for voxelization of a deformed cylinder [68]. The atlas function at a given point is defined as the number of the best contacts residing at this point. In a given voxel, the (discretized) atlas function is calculated by counting the number of the best contacts containing this voxel. The atlas probability is computed as a linear function of the atlas function and four probability definitions are given in [68]. The probability distribution is then presented as color-coded (> Figures 27-7 and > 27-8) or gray scale (> Figures 27-9, > 27-10, and > 27-19) maps. The PFA is dynamic and can rapidly be recalculated for arbitrary (userdefined) resolution and extended by adding new patient’s data. The atlas can easily be reformatted and warped to match patient-specific data. The detailed processing steps depend on a data acquisition process and stereotactic procedure. For
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
the acquired data, the PFAs were calculated for the STN (PFA-STN) and VIM (PFA-VIM) with a 0.25 mm3 spatial resolution and 0.25 mm accuracy.
Data Acquisition Multi-modal data were acquired pre-, intra-, and postoperatively during surgical treatment of Parkinson’s disease (PD) patients [74–76]. Preoperatively, lateral and antero-posterior X-ray ventriculography projections were acquired. The X-rays imaged the AC, PC, thalamus, and third ventricle, see > Figure 27-19b. The AC-PC distance, the height of the thalamus (HT), and the width of the third ventricle (V3) were measured on these X-rays for each patient, and subsequently used to normalize the best contact(s). Intraoperatively, the best contact was identified neuroelectrophysiologically and its position was imaged on two orthogonal X-rays. Its coordinates were measured on them and reconstructed in 3D. Two types of quadripolar electrodes were used: DBS 3387 with a 1.5 mm inter-contact gap and DBS 3389 with a 0.5 mm gap. For thalamic stimulation, the monopolar electrode was also employed. Postoperatively, each best contact was verified neurologically during a 3 month patient assessment follow-up and, if necessary, updated and re-measured on the X-rays. The selection of the best contacts aimed at improvements in akinesia, rigidity, and tremor [75,76].
PFA-STN Two versions of the PFA-STN were constructed. The first (main) version was built from all available 366 best contacts of 184 PD patients [70]. This version is available in [57] for clinical use (Section ‘‘Cerefy Clinical Brain Atlas: Enhanced Edition with Surgical Planning and Intraoperative Support (CCBA-Plan)’’). The second (bilateral) version was developed from 168 bilateral cases in two situations: with and without lateral compensation
27
against the V3 [71]. It is useful in studying STN properties and comparing the left and right functional STN [71] (Section ‘‘Atlas use and benefits’’). Contact normalization employs the landmarks identified during the data acquisition and used in neurosurgery planning: AC, PC, HT, and V3. In the atlas (normalized) space, the distance between AC-PC = 24 mm, HT = 16 mm, and V3average = 6 mm, where V3average is the average value of V3. Consequently for each case, the coordinates of the best contact(s) were scaled antero-posteriorly to match the atlas AC-PC distance and dorso-ventrally to match the atlas HT. Lateral compensation against the width of the third ventricle was performed by shifting its lateral coordinate by ((V3average V3)/2). Points of interest on the X-rays, measured with accuracy of about 0.25 mm, were reconstructed in 3D from two orthogonal projections. Under a simplified assumption that the patient’s midsagittal plane is ideally parallel to the lateral plane and perpendicular to the anteroposterior plane, the AC-PC distance and HT could be measured directly on the lateral projection, V3 on the antero-posterior projection, and the 3D electrode coordinates calculated easily from its two 2D ideally orthogonal projections. The 3D reconstruction problem, however, must be considered taking into account the stereotactic environment, data acquisition process, and potential mis-positioning of the patient’s head. Two 3D reconstruction methods taking into account these factors are formulated and their errors estimated in [70]. The validation of the PFA-STN in terms of the voxelization procedure by applying the Monte Carlo method and the correctness of the contact data is addressed in [70]. In general, the more contacts are used for PFA calculation, the better is the atlas from a statistical standpoint. However, the contact data may be inaccurate or even incorrect. Therefore, a suitable balance between the size of contact population and atlas quality has
409
410
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
to be maintained. Two quality criteria were applied in the process of contact selection and checking: (1) contact cluster formation (by visual inspection), and (2) reconstructed contact height preservation (by automatic calculation) [70]. The best contacts, normalized and put together in the atlas space, form typically a cluster. The outliers located outside this cluster required additional checking (these contacts sometimes were assigned a wrong electrode type resulting, due to a different inter-contact gap, in a wrong contact positioning). The second criterion is more reliable. The physical height of the contact is 1.5 mm, and its accurate reconstruction from two perpendicular X-ray projections (before spatial normalization) should result in the same height. Besides a finite accuracy of calculations, there are several other factors causing the reconstructed contact height differs from its physical one. First, the length of the electrode may not be measured accurately enough on the projections. Second, a wrong type may be assigned to the electrode resulting in a misplaced contact location. Third, 3D reconstructions from 2D projections may be distorted because of patient’s mispositioning, particularly when the patient’s midsagittal plane is not parallel to the plane of the lateral projection. Finally, 3D reconstructions may not be accurate because of incorrect input parameters of the stereotactic frame (frame size, size of angiographic localizers, or projections of angiographic localizers) and positions of the Xray sources and film plates. The height of each contact was reconstructed and checked against a given range of accuracy taken as 0.25 mm. If the reconstructed height was outside this range, the contact was re-examined by checking and correcting, if necessary, the electrode type, and re-measuring the electrode length on the X-rays. If after these operations the contact height was still outside the given accuracy range, this contact was rejected. Consequently, the contacts with the reconstructed height lower than 1.25 mm or higher than 1.75 mm were rejected in the process of PFA-STN construction. > Figure 27-7 shows
the PFA-STN including the normalized contacts, normalized best contacts, voxelized best contacts, and axial, coronal and sagittal color-coded maps.
PFA-VIM The same algorithm was applied to construct the PFA-VIM. It was built from 107 best contacts in two situations: with and without lateral compensation against the V3 [72]. This compensation slightly changes the laterality of the PFA-VIM mean value location from 13.99 to 13.83 mm for the left and from 14.13 to 13.84 mm for the right hemisphere. It also reduces the lateral coordinate of the standard deviation by 22% for the left and 15% for the right hemisphere. > Figure 27-8 illustrates the PFA-VIM. The algorithm for PFA calculation is fast. The complete PFA-VIM was calculated in 2 s for 0.5 mm resolution and in 14 s for 0.25 mm resolution on a standard PC [72]. An atlas update with new cases is feasible in a fraction of second without re-calculation of the entire atlas as the atlas function is linear [68].
Advantages of the PFA The PFA has a number of advantages. It is a truly probabilistic functional atlas calculated from intraoperative neuroelectrophysiology, preand intraoperative neuroimaging, and postoperative neurological assessment by converting the coordinates of the neurologically most effective contacts into probabilistic functional maps taking into account the patient-specific anatomy and geometry of a stimulating electrode. The PFA provides a quantitative spatial distribution of the best stereotactic targets in a normalized atlas space, enables determining the accuracy of targeting, and facilitates studying functional properties of structures (see Section ‘‘Atlas use and benefits’’). The PFA aggregates knowledge from the previously operated cases. This knowledge can
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
27
. Figure 27-7 PFA-STN (the origin of the coordinate system is at the PC, the AC-PC distance is divided into 12 units of 2 mm each, HT is divided into eight units of 2 mm each, and lateral units are in mm): (a) normalized contacts, (b) normalized best contacts, (c) voxelized best contacts, (d) axial (A), coronal (C), and sagittal (S) color-coded maps as well as the voxelized atlas (3D) along with the coordinate system and the locations of the orthogonal planes, (e) probability color bar
be aggregated individually by the neurosurgeon, within a group of users, or over the entire neurosurgical community by means of e.g., a PFAbased portal for SFN [69] (see Section ‘‘Research prototypes’’). This portal facilitates data sharing among functional neurosurgeons, calculates rapidly PFAs from local and/or global (shared) databases, facilitates comparison of data collected at various centers, and enables creation of the
PFA for various structures over the Internet by the neurosurgical community. This shifts a paradigm in atlas construction and extension from manufacturer-centric to community-centric. The PFA shows a distribution of the best contacts in image and volumetric representations calculated with a user-specified resolution. The PFAs were constructed for the STN and VIM with a high spatial resolution of 0.25 mm3 and
411
412
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
. Figure 27-8 PFA-VIM (the reference system and denotations as in > Figure 27-7): (a) normalized contacts, (b) normalized best contacts, (c) axial, coronal, sagittal color-coded maps and the voxelized atlas
accuracy of 0.25 mm. Consequently, the PFA is a volumetric atlas, consistent in 3D, and can easily be reformatted in any orientation. The PFA is dynamic and scalable. Any new cases can be added to the current PFA or various PFAs can easily be merged as the atlas function is linear (additive). Moreover, these calculations are fast. This enables a near real-time update of the PFA and its use in web-based applications. Finally, the PFA can easily be used for planning, see Section ‘‘Atlas use,’’ particularly in combination with the anatomical atlas (Sections ‘‘Combined anatomical-functional atlases’’ and ‘‘Atlas use’’).
Limitations of the PFA Although the PFA is a novel concept yielding a new type of atlas, its current implementation has several limitations in terms of data used, contact modeling, case normalization, and content.
The PFA does not contain the entire pre-, intra-, and postoperative data; neither does it include the complete electrophysiological findings from microrecording and stimulation. Therefore, attempts such as [66] aimed at storing all these data are of importance. However, to make this vast amount of information beneficial within neurosurgeon’s time constraints, it must be aggregated, knowledge extracted, and presented in a useful and easy way. The PFA, whose current version takes into account three symptoms only (akinesia, rigidity, and tremor), must generally be able to provide distribution of any individual symptom as well as any combinations of them. The atlas should be queried not only against a single value (probability at present) but also against any outcome, feature, and scoring system, such as UPDRS. In addition, it must be possible to mask the atlas with the regions that are positive (with improvement), negative (without improvement), and unexplored (not studied yet).
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
The current PFA algorithm assumes that the tissue activation region is limited to the shape of the normalized contact. Advances in imaging enable to visualize brain conductivity [77] and consequently predict the volume of tissue activated during deep brain stimulation (DBS) [78] on a patient-specific basis [79]. Subsequently this volume can be visualized in 3D along with the electrode, scan(s) and atlas(es), as e.g., in the Cicerone system [80] (Section ‘‘Research prototypes’’). Incorporation of any new geometric model of activated tissue into the PFA requires only substituting the algorithm for model voxelization, while the other calculations remain the same [68]. The brain normalization applied is limited to antero-posterior and dorso-ventral scalings, and lateral translation. It results from the anatomical features (AC, PC, HT, and V3) present in the acquired neuroimages and neurosurgery planning performed. Besides the speed in PFA normalization and fast mapping of the PFA onto patients’ scans, another advantage of this approach is that the optimal voxelization algorithm for the deformed (normalized) contact holds for this transformation [68], making the calculation of the entire PFA or its update fast. In principle, non-rigid normalization is feasible if a 3D scan is available, however, unpredictable and non-reproducible deformations in MRI [10] will limit the PFA accuracy (of 0.25 mm at present) and the current voxelization algorithm may not be valid any longer (whereas a direct voxelization of an arbitrarily deformed contact will take much more time). The PFA-GPi is not available yet. In addition, construction of PFAs for emerging targets, such as Brodmann’s area 25 [81] and pedunculopontine nucleus [82] has to be considered.
Combined Anatomical-functional Atlases Anatomical and functional atlases are complementary, so their combination is potentially useful. In
27
particular, the PFA and the SW atlas are highly complementary as the PFA, which is probabilistic, of a high spatial resolution (of 0.25 mm3), dynamic, composed from theoretically unlimited number of brain specimens, and consistent in 3D, complements a deterministic, sparse, static, based on two brains, and 3D inconsistent SW atlas. Conversely, a low parcellated PFA is enhanced by a highly parcellated SW atlas. Two approaches are feasible to combine the PFA with the SW atlas: (1) SW-centric by warping the PFA against the SW atlas [61], and (2) PFA-centric by warping the SW atlas against the PFA [83]. Both approaches were applied to the SW coronal and sagittal microseries; the SW axial plates were not considered as they are inclined by 7 (if they are needed, the corresponding PFA axial images suitably tilted must be generated first by interpolation). The TT atlas is (by construction) in spatial correspondence with the PFA created by normalizing its component cases to the AC-PC distance of 24 mm and HT of 16 mm (i.e., the same as in the TT atlas). Another effort towards building a combined anatomical-functional atlas is presented in [84]. The results for 28 patients with the implanted DBS electrodes in the STN were normalized nonrigidly [65] to form the postoperative maps of improvement in terms of the UPDRS increase. The maps were created for two groups of cases: less favorably (30% or higher improvement) and more favorably (50% or higher improvement).
SW-centric combined atlas The registration of the PFA to the SW atlas enhances the latter while not eliminating its limitations (Section ‘‘Limitations of anatomical atlases’’). The PFA was warped against the C-SW atlas by applying the same normalization transformation used for its construction to each brain separately (i.e., Brain LXVIII and Brain LXXVIII) for the normalization parameters determined in
413
414
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
[61]. The PFA was registered with the coronal microseries by scaling it dorso-ventrally by 19.4/ 16.0; the other two orientations remained unchanged. This is equivalent to stretching the size of the PFA voxel dorso-ventrally by 1.2125. The PFA was registered with the sagittal microseries by scaling its voxel dorso-ventrally by 19/16 and left-right by 23/24; the third orientation remained unchanged.
PFA-centric combined atlas The registration of the SW atlas to the PFA preserves spatial consistency of the PFA enabling planning in any orientation(s). It also reduces a SW inter-orientation spatial inconsistency. Moreover, its target structure dependence makes it more accurate. To register the C-SW atlas to the PFA, two criteria were applied: (1) normalization of the C-SW atlas, and (2) matching of the target structure’s centroid to the best target [83]. The C-SW atlas was treated as a component case of the PFA. Normalization of the C-SW atlas required its scaling in the antero-posterior and dorso-ventral directions to ensure the same AC-PC distance and HT in each atlas. The normalization parameters had been determined earlier [61]. The goal of the centroid-best target matching was to align the centroids of the target structures in the C-SW atlas with the locations of the best targets in the PFA. The scaling parameters were determined in the studies that quantified the main target structures STN [43], GPi [44], and VIM [45] in the C-SW atlas (see also Section ‘‘Limitations of anatomical atlases’’). The major difference between the C-SW coronal and sagittal microseries is in laterality, while the maximum difference between the posterior/anterior and ventral/dorsal centroid coordinates is small: 0.11 mm for the STN and 0.85 mm for the VIM. Hence, the centroid-best target matching was done laterally only and the other two
orientations were used for AC-PC and HT alignment. The lateral scaling factors differed considerably for the STN and VIM: the STN had to be stretched by 18% more than the VIM on the coronal microseries, and the VIM had to be compressed by 13% less than the STN on the sagittal microseries [83]. Therefore, the C-SW lateral scaling has to be target structure dependent (the other two scaling factors remain the same across target structures). The lateral scaling also matches (approximately due to the abovementioned small posterior/anterior and ventral/ dorsal mismatches) the target structure centroids for the C-SW coronal and sagittal microseries improving their mutual spatial consistency. The PFA-STN combined with the C-SW atlas is shown in > Figure 27-9. The PFA-VIM combined with the C-SW atlas is presented in > Figure 27-10.
Atlas-assisted Stereotactic and Functional Neurosurgery The usefulness of a ‘‘raw’’ atlas in SFN is limited. These are the atlas to scan registration and a battery of tools which make the atlas a valuable aid. Moreover, the deformable atlas along with these tools must be available to the neurosurgeon, preferable in a neurosurgical workstation.
Atlas to Scan Registration Overview Registration is an essential operation transferring the segmentation and labeling information from the atlas to the patient-specific data. Despite the existence of numerous techniques for brain warping, overviewed e.g., in [85,86], there is no nonrigid solution acceptable in clinical practice yet.
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
27
. Figure 27-9 PFA-STN combined with the C-SW atlas: (a) coronal view, (b) sagittal view. The probability is proportional to the gray scale and the locations of the images are shown in the top left corner
. Figure 27-10 PFA-VIM combined with the C-SW atlas: (a) coronal view, (b) sagittal view. The probability is proportional to the gray scale and the locations of the images are shown in the top left corner
The current practice in SFN it to use the Talairach transformation (Section ‘‘Talairach transformation’’) normalizing brains piecewise linearly [28] with a manual setting of the landmarks. This transformation is automated by developing the Fast Talairach Transformation which
maps the C-TT atlas onto a scan in 5 s [87] (Section ‘‘Fast Talairach transformation’’). Nonrigid brain warping methods are theoretically more accurate than piecewise linear warping and some of these methods are already employed in SFN [65,67,88], mainly for atlas construction.
415
416
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
Nonetheless, a landmark-based study [89] indicates that in the region between the AC and PC, where the main stereotactic targets are located, the difference between the linear and nonlinear (nonrigid) warping is practically negligible. The existing methods typically disregard the AC and PC, though the error due to their inaccurate localization is up to 0.5 cm [90]. The use of nonrigid methods is obviously advantageous in situations when the computational time is practically irrelevant, such as atlas construction, atlas-to-atlas pre-registration, or research applications. Their long computational time, however, is not acceptable in routine clinical procedures. Moreover, despite the increasing computing speed and development of more efficient brain warping methods, their ‘‘black box’’ nature limits their practical use in a clinical setting. The use of complex mathematical models and laws of physics with a little or no relevance to anatomy is barely to be understood and accepted. Unrealistically warped images do not help in trusting these methods. Finally, the need of setting multiple parameters prior to applying a method often restricts its use to its developers.
Talairach Transformation The Talairach transformation [28] normalizes brains piecewise linearly. It is based on the Talairach landmarks: two internal landmarks located on the midsagittal plane and six external landmarks lying on the smallest bounding box encompassing the cortex. The original Talairach point landmarks are: AC – the anterior commissure (point) is the intersection of the lines passing through the superior edge and the posterior edge of the anterior commissure (structure), PC – the posterior commissure (point) is the intersection of the lines passing through the inferior edge and the anterior edge of the posterior commissure (structure), L/R – most lateral point of the parietotemporal cortex for the
left/right hemisphere, A – most anterior point of the frontal cortex, P – most posterior point of the occipital cortex, S – most superior (dorsal) point of the parietal cortex, I – most inferior (ventral) point of the temporal cortex. The Talairach bounding box and the reference planes (i.e., the intercommissural plane, midsagittal plane, and coronal planes passing through the AC and PC) divide the brain into 12 regions (> Figure 27-12b). The Talairach transformation normalizes the brain by warping its scan within each region linearly to match the corresponding landmarks, resulting in an overall piecewise linear warping.
Fast Talairach Transformation To automate the Talairach transformation, the Fast Talairach Transformation (FTT), is developed which warps the C-TT atlas against a T1-weighted scan in 5 s [87]. The FTT exploits the modified Talairach landmarks [63] (see Section ‘‘Atlas use’’) and calculates them automatically in three steps: extraction of the midsagittal plane [91], identification of the AC and PC [92], and localization of the cortical landmarks [93]. Furthermore, the original Talairach transformation is extended by introducing two additional landmarks: the top of the corpus callosum and the most ventral point of the orbito-frontal cortex, and formulating an automated method for their calculation [94]. This extension, dividing the brain into 24 regions, enhances the quality of the FTT [94] and enables its use even when a complete brain axial scan is not available.
Atlas Use and Benefits The most often use of atlas is for targeting. The computerized atlases may provide additional benefits when employed properly and armed with powerful tools.
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
Atlas-derived best target points
27
PFA-determined accuracy of targeting
The computerized brain atlas enables determining the best stereotactic targets. An atlas-derived best target can be the mean value for a population of target points, the target structure’s centroid (or a centroid-related location), or the highest probability point or region for probabilistic atlases. The axial, coronal, and sagittal scatter plots of 54 DBS electrodes (of 29 patients) implanted into the STN are presented in [95]. The best contacts were those of being most effective during at least a 3 month postoperative follow-up. These plots allow for calculation of the populationbased (mean) best target which is (11.72, 1.62, 2.47), where the (right/left, posterior/anterior, and ventral/dorsal) coordinates (in mm) are in the Schaltenbrand reference system. The centroid coordinates of the 3D STN [43], 3D GPi [44], and 3D VIMe [45] in the CSW atlas are summarized in > Table 27-1. The PFAs versions constructed so far vary in terms of the number of contacts used and lateral compensation (applied or not). The mean best target points across these versions for the PFASTN and PFA-VIM determined in [83] are given in > Table 27-2.
As the atlas function is computable at any location, it is possible to calculate it in the neighborhood of the best target and study its behavior. The more level the function is (i.e., the wider ‘‘plateau’’ exists), the lower spatial accuracy of targeting is acceptable. A probability histogram (> Figure 27-11) shows that the decrease in probability around the best target voxel(s) is very high and, practically, the plateau does not exist [71]. Subsequently, a probability threshold was determined by comparing the size of the left and right PFA-STN [71]. Their ratio exhibits two behaviors: for low and medium probabilities it equals to one, and above probability of 0.77 it grows rapidly (up to 43 without lateral compensation against the V3 and up to 11 for lateral compensation), indicating a differentiation between the left and right STN in the regions, where the highest number of contacts were implanted (compare also > Figure 27-9a). This probability (of 0.77) divides the STN into the cold and hot STN, and the hot STN is taken as the target region. The size of the hot STN to that of the entire STN is between 1–2% indicating that targeting has to be done with a high spatial accuracy.
. Table 27-1 Centroid (right/left, posterior/anterior, ventral/dorsal) coordinates (in mm in the Schaltenbrand reference system) of the 3D STN, 3D GPi, and 3D VIMe in the C-SW atlas Orientation C-SW axial C-SW coronal C-SW sagittal
STN centroid (11.21, 0.64, 3.83) (9.18, 2.03, 3.95) (12.17, 1.92, 3.84)
GPi centroid (17.50, 5.44, 2.35) (15.58, 4.80, 0.11) (18.25, 4.06, 0.43)
VIMe centroid (13.77, 2.99, 2.35) (12.91, 5.66, 4.80) (16.65, 6.51, 4.91)
. Table 27-2 STN and VIM mean best target point coordinates (denotations as in > Table 27-1) PFA-STN mean best target Left
Right
(11.75,
( 11.75,
2.00,
2.75)
0.75,
2.75)
PFA-VIMe mean best target Left
Right
( 14.00,
( 14.00,
6.00, 1.75)
6.00, 1.00)
417
418
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
. Figure 27-11 Probability histogram of the PFA-STN calculated without lateral compensation against the V3. The length of each horizontal bar is proportional to the atlas volume for the probability determined by its color. Note that the high probability volumes (at the top) are very small in comparison to the low probability volumes (at the bottom). The probability color bar is on the right
Atlas use An early practical use of the stereotactic atlas for targeting and trajectory planning was to employ a single, usually sagittal, orientation scaled by means of a projector. Multiple complementary computerized brain atlases available in multiple orientations enable novel and beneficial ways of atlas use. Landmarks Stereotaxy requires a precise coordinate reference system. Therefore, the landmarks defining the coordinate system of a stereotactic atlas are critical. The error due to inaccurate localization of the AC and PC is up to 5 mm [90] and their lack may result even in a higher inaccuracy. In general, the landmarks must be defined uniquely, be easily identifiable in a scan, and their automatic identification should be computationally efficient. There are, however, several problems with the original landmarks in the SW and TT atlases as discussed in Section ‘‘Limitations of anatomical atlases.’’ To overcome these problems, a new, equivalent set of the Talairach landmarks was introduced (> Figure 27-12) and the resulting errors analyzed [63]. The new landmarks are defined in a more constructive way facilitating their efficient and automatic calculation. Two intercommis-
sural lines are defined on the midsagittal plane: central and tangential. The central intercommissural line is passing through the centers of the AC and PC structures, each approximated by a circle. The tangential intercommissural line is tangential dorso-posteriorly to the AC structure and ventro-anteriorily to the PC structure. In addition to the original Talairach AC and PC landmarks, three other AC and PC point landmarks (and the corresponding AC-PC distances) were introduced as follows: AC – is a point within the intersection of the anterior commissure (structure) with the midsagittal plane which can be: (1) central (the center of the anterior commissure structure), (2) internal (the most posterior point on the central intercommissural line), and (3) tangential (the tangent point of the tangential intercommissural line with the anterior commissure). PC – is a point within the intersection of the posterior commissure (structure) with the midsagittal plane which can be: (1) central (the center of the posterior commissure structure), (2) internal (the most anterior point on the central intercommissural
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
27
. Figure 27-12 Modified Talairach landmarks: (a) locations of the landmarks on three orthogonal planes: intercommissural, coronal passing thought the AC, and coronal passing through the PC, (b) reference planes and grid superimposed on a scan in 2D and 3D
line), and (3) tangential (the tangent point of the tangential intercommissural line with the posterior commissure). The locations of all modified Talairach landmarks are show in > Figure 27-12a. The modified landmarks can efficiently be calculated automatically (Section ‘‘Fast Talairach transformation’’). There are two factors speeding up their identification. First, two coordinates of the cortical landmarks are already fixed by definition so the third coordinate has to be calculated only. Second, in contrast to processing the entire brain, the identification of the modified landmarks is limited to three predefined planes only: intercommissural plane, and two coronal planes passing through the AC and PC landmarks, > Figure 27-12. This approximation is verified in [87]. The choice of the intercommissural line and distance is application-dependent. For SFN, the central intercommissural line and internal intercommissural distance are recommended [63].
The simultaneous use of axial, coronal, and sagittal orientations is advantageous and a computerized atlas is suitable for this purpose. The atlas structures can be presented on three cross-correlated images displayed individually (> Figure 27-2) or jointly (> Figure 27-6a) as a triplanar. This potentially enhances the accuracy of targeting by identifying and setting the landmarks on the orthogonal planes as well as increases neurosurgeon’s confidence. The detailed targeting steps for pallidotomy/ pallidal stimulation, thalamotomy/thalamic stimulation, and subthalamotomy/subthalamic stimulation performed simultaneously on all three orthogonal planes supported by global and local landmark-based registrations along with the selection of suitable local landmarks are given in [90]. The concept of global and local registrations is illustrated in > Figure 27-13. An example of a subthalamic stimulation targeted simultaneously on axial, coronal, and sagittal orientations by employing the C-SW Targeting on multiple orientations
419
420
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
. Figure 27-13 Global and local registrations of the C-SW coronal microseries in contour representation with an MRI scan. (a) Global (Talairach transformation-based) registration providing a correspondence between the atlas and the scan. The Talairach grid is attached to the contours and any corner of it is movable whose displacement causes a piecewise linear deformation of the contours in real-time. The globally registered head of the caudate nucleus and the optic tract in the atlas do not fit well the data. (b) Local registration enhancing the atlas-to-scan fit in the region of the GPi (highlighted) performed with the following landmarks: the head of the caudate nucleus to scale dorsally, and the optic tract to scale ventrally and laterally. This approach (illustrated on a single orientation) is applicable to all three planes simultaneously
atlas in contour representation is illustrated in > Figure 27-14. The target structure STN is delineated on all three orthogonal planes by the contours corresponding to atlas plates Hv 3.5, Fp 3.0, and S 12.0, and the target point is set manually to lie within all three contours. Typically a single stereotactic atlas is employed for planning. The use of multiple atlases with various complementary contents is beneficial by potentially improving the accuracy of targeting, increasing neurosurgeon’s confidence, and compensating for some atlas shortcomings [51,96]. Any orientation of a multi-atlas triplanar may come from any atlas. For instance, the triplanar may be formed from the SW axial and sagittal orientations and TT coronal orientation (as the SW coronal microseries differ significantly from the SW axial and sagittal microseries, Section ‘‘Limitations of anatomical atlases’’), > Figures 27-5a and > 27-15a. This approach also allows various parcellations and nomenclatures to be employed jointly for labeling, such as Hassler’s parcellation [32] in the SW Planning with Multiple Atlases
atlas and Walker’s parcellation [33] in the TT atlas, > Figure 27-15b. Atlas-assisted Preoperative Planning, Intraoperative Support, and Postoperative Assessment The most typical use of the stereo-
tactic atlas is for preoperative planning, mainly targeting. Intraoperatively, the atlas can also serve as a global positioning system and provide the neuroanatomy surrounding the target structure, list of structures along the stereotactic trajectory, and spatial relationships between the tip of electrode and critical structures, such as the optic tract. It also aids in data archival [57]. Postoperatively, the atlas assists in a spatial assessment of DBS/lesion placement [96]. The use of multiple complementary anatomical atlases in multiple orientations for preoperative planning, intraoperative support, and postoperative assessment is addressed in [51] and illustrated in > Figure 27-16. An integration of web technology with the PFA concept
Community-centric Atlas Construction
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
27
. Figure 27-14 Subthalamic stimulation targeted simultaneously in multiple orientations by means of the C-SW atlas in contour representation: (a) axial (plate Hv 3.5), (b) coronal (plate Fp 3.0), (c) sagittal (plate S 12.0). The target structure is highlighted, the reference axes are drawn, and the target point within the target structure is marked by a zoomable cursor
shifts a paradigm in atlas construction and extension from manufacturer-centric to communitycentric. This is enabled and illustrated by a community-centric PFA-based portal [69] providing an infrastructure for data collection and sharing as well as calculation of the PFA over the Internet, see also Section ‘‘Research prototypes.’’ The PFA is easily individualized to a patient’s scan by performing a transformation inverse to that applied for contact normalization as follows: (1) postero-anterior linear scaling to match the patient’s AC-PC distance,
PFA-based Planning
(2) ventro-dorsal linear scaling to match the patient’s HT, and (3) lateral translation matching the patient’s V3 (if a lateral compensation against the V3 was applied to create the atlas). The best target point is taken as the location with the highest probability (or, alternatively, for the STN as a point within the hot STN [71] (Section ‘‘PFAdetermined accuracy of targeting’’). The PFA may be superimposed on the patient’s scan such that the region of zero probability is rendered transparent and the non-zero probability region is displayed in gray scale proportionally to probability, see > Figure 27-19.
421
422
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
. Figure 27-15 Planning with multiple atlases. (a) Pallidotomy targeting with multiple atlases: C-SW atlas in axial (plate Hv 3.5) and sagittal (plate S 20.0) orientations, C-TT atlas in coronal orientation (plate TT 8.0), and 3D C-TT atlas with scan triplanar. The target point is marked by a zoomable cursor . (b) Multiple labeling in the thalamic region with two nomenclatures by Walker’s and Hassler’s
In the individualized anatomical-functional atlas, the SW atlas delineates the target structure and its surrounding neuroanatomy, while the PFA determines the target point within the target structure as the location with the highest probability (i.e., the highest density of the normalized best contacts for the previously operated cases). Consequently, the PFA enables precise targeting for the current and next tracks while the SWatlas serves as a global positioning system providing the neuroanatomy surrounding the target structure, list of structures traversed along the trajectory, and spatial relationships to critical structures. The anatomical-functional atlas created by the SW-centric approach (Section ‘‘SW-centric combined atlas’’) is probably easier acceptable as a well-known SW atlas is enhanced by the PFA. Its limitation is that planning on each orthogonal orientation has to be done separately. The PFA-centric approach (Section ‘‘PFA-centric combined atlas’’) produces a superior combined atlas allowing for planning with all orientations
Combined Anatomical-functional Atlas
simultaneously. In addition, it reduces spatial inconsistency among the SW microseries. The spatial correlation between the anatomical STN (from the SW atlas) and functional STN (PFA-STN) was studied quantitatively [73]. For probability p 0.3, more than 95% of the functional STN is inside the anatomical STN and for p 0.5 the complete functional STN is inside the anatomical STN. Therefore, the PFA-STN for p = 0.5 after registration to a scan can potentially be used for 3D identification of the STN in neuroimages substituting the SW atlas for this purpose.
Potential Benefits of Atlas Use The use of the stereotactic atlases as discussed above has several potential benefits, including reduced cost and time, reduced invasiveness, increased accuracy of targeting, increased neurosurgeon’s confidence, facilitated rapid planning, support for new procedures, facilitated
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
27
. Figure 27-16 Atlas-assisted preoperative planning, intraoperative support, and postoperative assessment: (a) subthalamic stimulation planning with multiple 2D and 3D atlases, (b) intraoperative support with the GPi (pallidum mediale internum) delineated and labeled, stereotactic trajectory (thin line) with the current position of the microelectrode (thick line) shown, and the structures traversed along this trajectory listed (left); and a 3D view of the microelectrode, target structure and atlas-scan triplanar (right), (c) postoperative assessment of a DBS placement in the GPi, (d) postoperative analysis of a thalamic lesion in all three planes
inter-clinician communication, and enabled building of community-centric solutions. The atlas notably lowers the surgical cost by reducing the duration of surgery by decreasing the number of stereotactic tracts. An initial evaluation suggested that the atlas could potentially reduce
the number of tracks from five to one per hemisphere [96] resulting in cost and time savings. This cost is further reduced by decreasing the number of microelectrodes inserted. This decrease also lowers the invasiveness of the surgical procedure, decreasing potential surgical complications by
423
424
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
reducing the risk of hemorrhage as well as surgically induced capsular and visual deficits. The atlas increases the accuracy of targeting by employing multiple, complementary atlases in multiple orientations, global and local registrations with any clearly visible landmarks set on three orthogonal planes, and sub-pixel accuracy measurement. This is further enhanced by the combined anatomical-functional atlas, such as the PFA co-registered with the SWatlas. Then, the PFA enables precise targeting (with a spatial resolution of 0.25 mm3) for the current and next tracks while the SW atlas serves as a global positioning system providing the delineated neuroanatomy surrounding the target structure, list of structures traversed along the track, and spatial relationships to critical structures. The individualized atlas increases neurosurgeon’s confidence pre-, intra-, and postoperatively by employing multiple atlases in multiple orientations, providing scan labeling, and allowing for an intraoperative measurement of distances to critical structures. This confidence is anatomical, functional, and spatial. Anatomically, the atlas delineates the structure of interest that, depending on acquisition, may be indiscernible or incompletely visible on the scan. The anatomical atlas also provides a detailed neuroanatomy labeled on all three orthogonal planes with a higher parcellation than that of the scan itself (> Figure 27-4f). Functionally, the atlas determines the target point within the target structure and provides the distribution of probabilities. Spatially, the atlas provides the triplanar, 2D-3D cross-correlation, and 3D relationships. The atlases and the approaches described above facilitate rapid planning, suitable for surgical procedures where the time between the scanning and the operation is short, so that the scanning, planning, and surgery can be done during the same session without removing the stereotactic frame. This also allows the neurosurgeon to plan more sophisticated trajectories by displaying the track on all three planes and in
3D, providing the list of structures traversed along it, and measuring distances to critical structures. The availability of the segmented and labeled Brodmann’s areas (BAs) in the C-TT atlas (> Figure 27-1b) facilitates localizing cortical areas, such as BA25 [81], or exploring them, for instance, BA4, BA6, BA24, BA32 [97]. While the understanding of an underlying neuroanatomy in the scan is easy for a neurosurgeon, communicating the individualized neuroanatomy to other clinicians may be tedious and time-consuming. A scan annotated by a deformable atlas is able to transfer this information to other clinicians including neurologists and neuroradiologists as well as to residents and medical students (> Figures 27-1f and > 27-4f). The concepts and initial solutions addressed here enable atlas construction in a communitycentric manner by enhancing and extending the PFAs for the existing and creating PFAs for new target structures (as discussed in Section ‘‘Limitations of the PFA’’). Finally, though the initial evaluation suggested that the atlas could potentially reduce the number of stereotactic tracks, further analysis and studies are required to quantify atlas benefits.
Atlas-based Applications Numerous atlas-based applications for SFN have been developed as research prototypes and commercial products. The use of atlas in these applications varies in numerous terms including:
Atlas(es) employed (2D SW, 2D TT, 3D SW, 3D TT, PFA, and/or others) Construction of the computerized atlas (from a directly digitized print plates to an enhanced, expanded, and 3D deformable atlas) Availability of all orthogonal orientations (available, not available)
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
Availability of multiple atlases (single atlas, multiple isolated atlases, multiple mutually co-registered atlases) Atlas representation (image, contour, polygonal, volumetric) Atlas-to-scan registration approach (interactive, automatic, mixed) Atlas-to-scan warping transformation (linear scaling, 3D piecewise linear scaling, nonrigid warping against an anatomically normal scan, nonrigid warping against a pathological scan) Structure labeling (not available, available at the border, available for the entire structure) Atlas display (atlas alone, atlas and scan sideby-side, atlas images overlaid onto a scan, atlas images overlaid onto a scan with usercontrolled blending, atlas contours overlaid onto a scan) Other supportive tools.
Research Prototypes Since the development of the first atlas-based tool in 1974 [19], numerous research atlas-assisted prototypes have been developed [69,80,96,98–103]. We briefly feature some of them. NeuroPlanner is an atlas-based software system for SFN that supports preoperative planning and training, intraoperative procedures, and postoperative analysis [96] (see > Figures 27-12b, > 27-13–> 27-16). NeuroPlanner contains multiple 2D and 3D Cerefy atlases mutually coregistered [50] with about 1,000 structures and 400 sulcal patterns. Numerous tools provide four groups of functions: data-related (data reading, interpolation, reformatting, image processing), atlas-related (real-time interactive atlas-to-data warping, multiple labeling in 2D and 3D), atlasdata exploration-related (interaction in three orthogonal and one 3D views, continuous dataatlas exploration), neurosurgery-related (targeting, path planning, measurements, simulating
27
electrode insertion, simulating therapeutic lesioning). NeuroPlanner along with the Cerefy atlases, trial licensed to several commercial, clinical, and research sites, has been playing an important educative role and influencing the design of commercial systems that incorporated the Cerefy brain atlases (Section ‘‘Neurosurgical workstations’’). BrainBench resulted from integration of the brain atlas with virtual reality [102]. It contains a suite of tools for SFN along with the 3D C-TT atlas. It employs a virtual workbench called the Dextroscope (see also Section ‘‘Future stereotactic environments’’) where the user reaches with both hands behind a mirror into a computergenerated 3D stereoscopic object, and moves and manipulates it in real time with natural hand movements. BrainBench facilitates preparing faster plans, provides a better and more accurate choice of target points, improves the avoidance of sensitive structures, has fewer sub-optimal frame attachments, and enables faster and more effective planning and training. The Cerefy Neuroradiology Atlas (CNA) [100] is a general purpose, public domain tool for rapid labeling and exploration of scans by means of the C-TT atlas warped by applying the Talairach transformation. The CNA provides interactive scan labeling, navigation on all three planes, zoomable triplanar, data-atlas blending, reading Talairach coordinates, mensuration, putting annotations, and drawing regions of interest. It saves an atlas-labeled and annotated scan in a Dicom or XML file for subsequent use by other clinicians or for presentations. A community-centric PFA-based portal for SFN [69] provides an infrastructure for data collection, sharing, and calculation of the PFA over the Internet. This portal links two (not necessarily exclusive) groups of neurosurgeons: these who are willing to share their data with those who would like to use data from others. A neurosurgeon is able to generate a customized PFA in three ways, as: (1) local PFA from the
425
426
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
neurosurgeon’s own data, (2) globally combined PFA from the data of others (shared in the global database), and (3) globally and locally combined PFA from the neurosurgeon’s own data combined with selected data of others. The portal provides tools for: (1) data input, transfer, and editing, (2) data selection for PFA creation, (3) PFA calculation, (4) PFA display and interactive manipulation, (5) target planning and probability histogram generation, and (6) parameter setting. This portal facilitates data sharing among functional neurosurgeons, calculates rapidly PFAs in three ways as stated above, facilitates comparison of data collected at various centers, and enables creation of the PFA for different structures over the Internet by the neurosurgical community (a server with this public domain portal was maintained for a few years; about 200 users registered, but no data was deposited into the global database for public use). An atlas-assisted software system described in [98] is dedicated to assist neurosurgical planning. It contains a 3D version of the TT atlas, is able to segment the cortex and ventricles in less than 5 min, and provides a nonrigid registration of the atlas to a patient’s MRI scan in a few minutes. This application is able to deform the atlas against brain tumors and is interfaced with a navigation system allowing for an intraoperative use of the atlas. CASMIL is a comprehensive, augmented reality software/toolkit with the Cerefy atlas for image-guided neurosurgeries [99]. It integrates a variety of modules and provides multiple tools for rigid and nonrigid registration (imageimage, image-atlas, and image-patient), automated 3D segmentation, brain shift prediction, knowledge-based querying, and intelligent planning. Brain shift is predicted by applying a patient-specific finite element model. CASMIL provides near real-time interaction with intraoperative MRI. It also has been securely webenabled and optimized for remote web and PDA access.
Cicerone is a software tool for stereotactic neuroelectrophysiological recording and DBS electrode placement [80]. It enables interactive 3D visualization of the co-registered MRI and CT scans, 3D brain atlases, neuroelectrophysiological microelectrode recordings, and DBS electrode(s) with the volume of tissue activated (VTA) as a function of the stimulation parameters. Preoperatively, for the intended anatomical target, Cicerone assists in selecting the optimal position on the skull for burr hole to maximize the likelihood of complete microelectrode and DBS coverage. Intraoperatively, it allows visualization of the electrode location in 3D relative to the surrounding neuroanatomy and neurophysiology. Moreover, Cicerone enables prediction of the VTA generated by DBS for a range of electrode trajectories and tip locations.
Commercial Products Two groups of atlas-assisted commercial products are featured below: neurosurgical workstations and CD-ROMs. Neurosurgical Workstations
Computerized brain atlases are commonly available in neurosurgical workstations. The Cerefy Electronic Brain Atlas Library and/or Cerefy Brain Atlas Geometrical Models are available in the StealthStation (Medtronic Surgical Navigation Technologies, Louisville, CO), Target and iPlan (BrainLAB AG, Feldkirchen, Germany), SurgiPlan (Elekta Instrument, Stockholm, Sweden), and SNN 3 Image-Guided Surgery System (Surgical Navigation Specialists, Mississauga, Ontario, Canada) as well as in the neurosurgical robot NeuroMate (Integrated Surgical Systems, Davis, CA). The Cerefy brain atlas libraries are also in a process of evaluation by Prosurgics (UK), Renishaw (UK), Cedara Software (Canada), and Z-KAT (USA). Other companies have developed their own digital versions of the SW and TT print atlases, including Tyco/Radionics
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
(Burlington, MA) and Stryker/Leibinger (Kalamazoo, MI). Electronic atlases are also available in the COMPASS System of Stereotactic Medical Systems [38] and in the CASS system of MIDCO [35]. Illustrations of the Cerefy atlases in the StealthStation and iPlan are shown in > Figure 27-17. CD-ROM Applications
Several atlas-assisted applications are available on CD-ROM. We feature some of them below. The ECBA [54] is a reference tool with the C-SW and C-TT atlases. It provides many features not available in the original atlases, including co-registered atlases; flexible display, manipulation, and printing of atlases in multi-atlas and triplanar modes; and about 17,000 structures pre-labeled on 1,500 atlas images. The foreword to this atlas was written by Dr. Jean Talairach (> Figure 27-18). The ECBA allows the individualized atlas to be generated without loading a scan and provides a simple targeting procedure [104]. The C-SW atlas is fit to the scan by means of 2D local deformation. First a rectangular region of interest (ROI), set among any clearly visible landmarks, is measured on the film or scanner console. The corresponding atlas image with the target structure is then deformed in real-time for the same landmarks such that the dimensions of the atlas and film ROIs are the same. The target point is then set on the individualized atlas image and its coordinates are read. The individualized image printed on a transparent foil can be overlaid on the film or, alternatively, this superimposition can be done electronically. The Electronic Clinical Brain Atlas (ECBA)
Functional imaging is an established technique in neurosurgery for studying the brain in health and disease. Identifying multiple activation loci on numerous functional images, determining their underlying cortical and subcortical anatomy, and reading their coordinates along with Brain Atlas for Functional Imaging (BAFI)
27
anatomical and functional values is a tedious, time-consuming, and error-prone task. The BAFI [55] was developed to facilitate this task by providing rapid atlas-assisted analysis of functional images [60] along with a locus-driven mechanism [105]. This CD-ROM allows the user to load anatomical and functional datasets, co-register them and place in the Talairach space, identify activation loci and label them automatically with Brodmann’s areas, gyri, and subcortical structures by using the C-TT atlas. Numerous tools are available for identification of activation loci, placing marks on them, editing and labeling of marks, and saving the results in electronic format. Cerefy Clinical Brain Atlas: Enhanced Edition with Surgical Planning and Intraoperative Support (CCBA-Plan) The CCBA-Plan [57] contains the
C-SW atlas, C-TT atlas, and PFA-STN. It provides standard atlas-related operations (> Figures 27-2 and > 27-5a) as well as is equipped with numerous tools for planning and intra-operative support by means of the combined anatomicalfunctional atlas. For a loaded patient-specific image (MRI, CT, or X-ray ventriculography), the CCBA-Plan allows the neurosurgeon to: warp the C-SW atlas in 1D, 2D, or 3D as well as translate, rotate and flip the image; plan the target point on an individualized C-SW atlas; plan the entry point on an individualized C-SW atlas by providing either its coordinates or the angles and distance to the target point; plan up to five electrode tracks; display trajectories in two orthogonal views; simulate electrode insertion with a label and PFA-STN probability readout at the electrode tip as well as provide an atlas display guided by tip movement; annotate the electrode track with recording and stimulation findings, label, and distances to the target and the intercommissural plane; and save the annotated electrode to an external file. These features allow the neurosurgeon to collect findings intraoperatively in electronic format and use the CCBA-Plan as his/her own local archive. Tra-
427
428
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
. Figure 27-17 Cerefy atlases in commercial neurosurgical workstations: (a) StealthStation: intraoperative use of the C-SW atlas in all three orientations (image courtesy of Dr. J. Henderson), (b) iPlan: trajectory verification based on side-by-side viewing of the scaled C-SW images superimposed on the corresponding reconstructions of a patient’s scan (image courtesy of T. Schwan)
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
27
. Figure 27-18 Demonstration of the ECBA (right bottom corner) by the author to Drs. Talairach (center) and Tournoux (right) at Saint Anne Hospital in Paris in 1996
jectory planning and simulation of electrode insertion are shown in > Figure 27-19.
new environments providing paradigm shifts in SFN.
Future Directions Tremendous technological advances impacting neurosurgery [106], particularly in information and communication technology, nanotechnology and molecular imaging, will drive the progress in SFN. This progress will also impact atlasassisted SFN. As the role of the atlas in SFN has been evolving, we predict that its importance and usefulness will be growing. We believe that the future efforts in atlas-assisted SFN should be carried out in three following directions: 1.
2.
3.
Construction of more accurate, more detailed, high resolution volumetric, multimodal probabilistic atlases at different scales Development of faster, more accurate, reliable, and automatic methods for atlas to data registration along with their validation Development of novel therapeutic procedures, more powerful applications, and
Future Stereotactic Atlases We believe that the future atlas for SFN must: be population based; contain anatomy, function (including neuroelectrophysiology), connectivity, and vasculature along with their variability; cover the entire brain and represent it at different scales (gross neuroanatomy, cytoarchitecture, myelination patterns, neurochemistry, neuroreceptors, gene expressions); provide distributions of the best targets in function of the symptoms treated, neurological findings, and clinical outcomes; differentiate regions into positive (with quantitative improvement), negative (without improvement), and unexplored (not studied yet); be self-updated dynamically and continuously with the new cases processed; be fully segmented and labeled, and highly parcellated; be consistent in 3D; be volumetric of high resolution; be correct and
429
430
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
. Figure 27-19 CCBA-Plan. (a) Trajectory planning by the combined anatomical-functional atlas. Three trajectories are set. The probability of the target point is displayed in the bottom left corner. (b) Simulation of electrode insertion on a sagittal ventriculogram. The distance to the target as well as the probability and structure name at the tip of electrode are displayed. The combined atlas is fit to the image by performing translation, rotation, and 3D scaling. The PFA images are rendered transparent for zero probability
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
validated; and provide powerful tools supporting its use and dynamic growth. The future atlas should be updated dynamically and continuously with the clinical cases, ideally globally without geographic, political, and/or vendor-imposed constraints (atlas without borders). The efforts on the PFA [68] and the community-centric PFA-based portal [69] partly illustrate this direction. The architecture of the future stereotactic multiatlas is presented in > Figure 27-20 (modified after [83]). To enable atlas potential in clinical practice, it must be equipped with numerous tools supporting a broad variety of operations including: fast and accurate registration (for atlas-to-data, datato-atlas, data-to-data, and atlas-to-atlas mappings), labeling with multiple features (reading and placing of meta labels); planar and curved reformatting (in the orthogonal planes, electrode planes (the ‘‘probe eye’s view’’), and arbitrary planes); rendering (triplanar display, and surface and volume rendering); exploration (of atlases, data, and other associated materials); incorporation and processing of clinical and imaging results and studies; readout (of stereotactic coordinates, probabilities, and scoring systems); quantification (of distances, angles, areas, and volumes); selection of any atlas subset (volume
. Figure 27-20 Architecture of the future stereotactic multi-atlas
27
of interests) according to given features/criteria; representation conversion (for image, contour, polygonal, and volumetric representations); and handling specific applications and/or devices (e.g., stereotactic frames, navigation devices). Towards constructing this type of atlas, a multi-atlas was proposed and its first version presented composed of the PFA, interpolated C-TT atlas, and enhanced C-SW atlas, mutually co-registered [83], see > Figures 27-3b, > 27-9, and > 27-10. Recently, new atlases are being constructed which eventually may find some future applications in SFN, including a probabilistic atlas of the basal ganglia and (yet unparcellated) thalamus [107], Cerefy atlas of cerebral vasculature with 3D arterial and venous systems and their variability [108], population-based atlas of white matter tracts [109], atlases of blood supply territories [110,111], and 3D probabilistic atlas of the cortical structures [112]. Though this review is limited to stereotactic human brain atlases, development of new, 3D, high resolution, stereotactic animal atlases for experimental surgical procedures is also important [113].
431
432
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
Atlas to Data Registration Registration is a key technique enabling the growth of atlas use. We believe that a clinically usefully registration must be rapid, reliable, accurate, automatic, and validated in a clinical setting. Warping techniques should handle anatomically normal and pathological cases, including brain tumors causing significant mass effect. Some initial solutions able to warp the atlas against a tumor are already proposed [98,99,114], providing the deformed neuroanatomy surrounding the tumor. Moreover, this tumor deformed neuroanatomy has to be correlated with tractography [115]. Validation of registration techniques in SFN is critical and efforts such as [116] are of importance.
Future Stereotactic Environments The future atlases with clinically accepted warping techniques will enable novel procedures and open new avenues provided that suitable applications and environments will be developed. This development should tackle data explosion, novel therapeutic procedures, and conceptually new stereotactic environments enabling paradigm shifts in SFN. To manage a rapidly growing amount of patient’s and atlas’ data, new ways of visualization of and interaction with the data are necessary. A potentially useful solution may be the Dextroscope originating from BrainBench (Section ‘‘Research prototypes’’) as a research prototype for SFN, extended subsequently to tumor stereotaxy and vascular malformations in [117], and becoming at present a product applied in various neurosurgical procedures [118,119]. Development of new procedures, particularly targeted therapy delivery (of drugs, antibodies, and biological agents) including gene/cell therapy (with a new wave of compounds under development), will require more detailed and parcellated atlases at different scales and very accurate warping techniques.
In a longer term, fundamental paradigm shifts must be proposed and new neurosurgical environments developed. We believe that these paradigm shifts should be along two directions: technology-related and patient-related. A technology-related paradigm shift is addressed in [106] by proposing a futuristic environment called DOTELL. The current surgical environment is both device- and information-centric. With a technological progress, the neurosurgeon will (sooner or later) be overwhelmed by data and equipment explosions, unless this burden is taken over by some intelligent assisting environment capable of acquiring, integrating, and processing the entire data as well as controlling and handling all instrumentation. This assistant should isolate the neurosurgeon from the instrumentation and information, and be able to perform two main groups of functions: TELL or show me and DO it (hence DOTELL). The architecture of DOTELL and its construction via technology integration are presented in [106]. DOTELL is an intelligent atlas-assisted device with robotic capabilities able to integrate hospital infrastructure, imaging systems, knowledgebased decision support, and therapeutic modalities. An example of DOTELL assisting a neurosurgeon in a bilateral subthalamic stimulation is given in [106]. It performs a broad spectrum of actions ranging from preoperative patient scheduling, scanning and surgery planning to intraoperative supervised robotic-based execution and visualization to postoperative ICU management and scanning. The second major change concerns the patient. We believe that it is the patient who should make all major decisions regarding his/her health, life, past, and future. The neurosurgeon armed with his/her knowledge, experience, and technology should assist the patient in his/her choices. This patient-centric concept of do-ityourself-neurosurgery is illustrated in the appendix as a futuristic neurosurgical procedure. Though it may still sound as science fiction (as
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
a head-mounted multi-modal scanner, knowledge and skill mapping, and memory preservation do not exist yet), nonetheless many other components are available today. In either technology-related or patient-related paradigm shift, the role of the atlas is tremendous: to equip a device with domain knowledge and ability to educate the patient on his/her specific demand. To continue the atlas growth and to keep exploiting its potential, the clinicians, researchers, and engineers have to work closely together.
Acknowledgments I am deeply grateful to Drs. J Talairach and P Tournoux for insightful discussions on their atlases. This work might not have been advanced without their initial enthusiasm about our atlases. I am truly indebted to Drs. AL Benabid, Grenoble, TT Yeo, Singapore, and AM Lozano, Toronto for the stimulating discussions and opportunity to observe their procedures. I was also inspired by the presentations and discussions at the 1997, 2001, and 2005 meetings of the World Society for Stereotactic and Functional Neurosurgery. The creation of the PFA and the PFA-portal for SFN was a joint effort with Dr. AL Benabid. The NeuroPlanner was developed within a joint project with Dr. TT Yeo of Tan Tock Seng Hospital, Singapore. The construction of the first electronic version of the TT atlas and development of the ECBA was a joint project with Dr RN Bryan of Johns Hopkins Hospital. The BAFI was developed in consultation with Dr. DN Kennedy of Massachusetts General Hospital. > Figure 27-17a of StealthStation is courtesy of Dr. J. Henderson of St. Louis University Health Sciences Center (now at Stanford University Medical Center). > Figure 27-17b of iPlan is
27
courtesy of Thomas Schwan of BrainLAB; iPlan is a registered trademark of BrainLAB AG in Germany and the US. Many ideas, solutions, atlases, and applications proposed would not materialize without their efficient implementation. Numerous persons from our Biomedical Imaging Lab, A*STAR, Singapore contributed to the development of tools for atlas construction and atlas-assisted applications, including A Thirunavuukarasuu (brain atlas CD-ROMs), D Belov (PFA, PFA-based portal for SFN, CNA), A Fang (atlas tools, 3D TT, initial version of NeuroPlanner), BT Nguyen (atlas tools), J Liu (interpolation, modeling, 3D TT), GL Yang (NeuroPlanner), L Serra (BrainBench), KN Bhanu Prakash (FTT), QM Hu (FTT), and GY Qian (FTT). I thank Aminah Bivi for her editorial assistance. This work has been funded by A*STAR, Singapore.
Appendix: Unlocked Brain: Do-it-yourself Neurosurgery Mr. Green, a successful venture capitalist in wireless communication, entered the famous Brainsterium. He had developed a malignant brain tumor as confirmed by a non-excisional optical biopsy. His feelings were mixed. He had expected to undergo a gene therapy, but he was offered a classic tumor resection. His colleagues in life sciences invested heavily in gene therapy, so he expected to benefit out of it. He realized that none of them invested in image-guided surgery. Thus, he felt that an obsolete technology would decide about his life. On the other hand, the Brainsterium had a brand name and an impeccable reputation to be the best. His first impression upon entering the Brainsterium was surprisingly positive. He rather expected a terrible hospital smell, miserable patients being moved around, and a noisy crowd of visitors, as he had experienced while
433
434
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
visiting a hospital the last time when his father had got a stroke. What he saw instead looked like a quiet high-tech lab. He headed towards the office of the chief neurosurgeon, Dr. Noki, who would explain to him the procedure and supervise the surgery. Supervise, not operate; this was the major difference that the Brainsterium offered to its patients. ‘‘It is you who decides what to keep in and what to remove from your brain. We just provide the right environment to do it’’ – started Dr. Noki. Mr. Green looked surprised. Do-it-yourself neurosurgery? He had no idea about this market segment. ‘‘Your brain will be unlocked and you will be able to view its content, such as knowledge, skills and memories, and examine how they are being invaded by the tumor’’ – continued Dr. Noki. ‘‘This surgery has two contradictory goals. One is to destroy all tumorous cells completely and the other is to maximally preserve the functions of your brain. I will provide you with two extreme brain resection regions: the conservative region with the core tumor only, and the aggressive region that contains the core tumor along with all tumorous cells that have migrated away from it. You have to balance between them to plan your postoperative life. When all tumorous cells are completely removed, the chances of physical survival are higher. This is the best for your body, but not necessarily for your mind and career. Preserving maximally your brain functions sounds more attractive but it puts your life at a higher risk.’’ Mr. Green suddenly visualized his brain as a financial asset and things became clear. As the exclusive owner of his brain and its content, he himself wanted to have a full control over this asset and decide about the associated risk. ‘‘Someone’s brain is much more valuable than his bank account’’ – he thought, ‘‘so why for decades has this been working differently?’’ He was more and more eager to understand this worth-investing technology. He once had
run an R&D department before becoming a venture capitalist. Dr. Noki provided more operational details. ‘‘You will be given access to our patients’ database and have permission to communicate with anyone who underwent this type of surgery. If you decide to proceed and accept our terms and conditions, you will be allowed to access all tutorials and simulators, and you can play back any previous surgery with, of course, no access to the brain contents of our patients.’’ ‘‘In the next step’’– continued Dr. Noki – ‘‘your brain will be unlocked by measuring its magnetic, electrical, chemical, and optical properties using a battery of techniques. They will produce the images of anatomy, vasculature, connectivity, function, pathology, and knowledge in your brain. If you are interested in technical details, refer to our tutorials. The extent of the tumor will be defined, and the conservative and aggressive resection regions prepared for you. You will be trained to understand the images employed to plan the resection. These images show tissue at the micron’s scale and at this resolution it is easy to distinguish normal from tumorous cells. The content of the resected brain region can be partly recovered. Your memories will be retrieved and saved on a disk. Play it later at home so that your brain will restore these memories in new locations. Remapping of the skills and knowledge is still at an experimental stage. At present, the content of the resected brain will be recorded and stored. We will acquire the knowledge distribution map of your brain later to find suitable locations for placing back the recovered content. There is a chance that some day, with the advancements in knowledge remapping and brain reconnection technologies, your skills and knowledge will be fully recovered. Finally, the tumor will be ablated with the collimated ultrasonic scalpel and removed without opening your skull. It will be dissolved and sucked out through the vasculature. All actions and operations will be
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
controlled by our revolutionary surgical environment DOTELL.’’ The introduction was over. Mr. Green was ready for this fascinating journey. He began it from the Resource Center ushered by a lion-robot. He was requested to put his fingers on the scanning plate and look into a camera to capture his biometrics. Mr. Green logged into the patients’ database and entered his personal wavelength. He had quite a broad band, which substantially accelerated operations. First, he registered with the system and displayed the list of patients who underwent a similar procedure. There were several thousands of them. No mortality, no technical failure during surgery; neurological deficits were quite variable, however. Whom to ask? He entered his year and place of birth. There was a familiar name, Jack Case. They had been schoolmates in grade six before his family moved to the West Coast. Mr. Green requested a videoconferencing session. He was lucky. Jack was in his garden and Mr. Green recognized his old friend. Jack had chosen the complete tumor removal 2 years back. He quit his job and was spending his days tendering his oceanside garden. Today Jack would opt differently. Mr. Green terminated the session and was led to the Brain Unlocking Center. A pretty nurse with an east European accent welcomed him. He was asked to provide a detailed list of his skills and related knowledge. The questionnaire was quite boring but Mr. Green realized that it was critical for an accurate planning of his brain stimulation and knowledge mapping. Next, the nurse put a bulky helmet on his head. ‘‘This must be that famous BCC, Brainsterium’s collector and collimator – one of the key unfair advantages of the Brainsterium. How did they manage to design this three-in-one gadget able to acquire multi-modal data and to collimate myriad of energy sources dynamically providing a non-invasive access to any location in the brain at micron’s accuracy for stimulation and excision?’’ – wondered Mr. Green. While
27
stimulated, he experienced unusual sensations. He saw some strange visual effects, heard funny voices, smelled oriental plants, and had an impression he was flying while the angels were singing. ‘‘No, not yet’’ – he said. ‘‘This is just a knowledge mapping procedure.’’ He looked at his body tightly attached to the chair. His excitement rocketed when he entered the Brain Exploration Center. It resembled a cyber cafe he had used to visit with his son a long time ago. Patients with helmets sitting in cubicles appeared playing games and navigating through some mazes. But everyone played seriously, as he could win or lose his past and future life. Mr. Green entered a cubicle and touched the start button on the screen. A welcome message with his name appeared and a colorful image of his brain showed up. ‘‘It recognizes my biometrics’’ – he thought. Three available functions were displayed, ‘‘explore your brain,’’ ‘‘plan your surgery,’’ and ‘‘preserve your memories.’’ Mr. Green started with the first one. He was astonished with the ease he could navigate his brain and how the Cerefy Atlas was able to give him the name of any tiny structure along with description of its function. Mr. Green began to appreciate its potential. It was time to start doing the job. He touched the second button. The nurse appeared and demonstrated how to distinguish on the images the normal from tumorous cells, and how to edit the resected region. It was quite easy with the Dextroscope stereoscopic display and a 3D reach-in, tactile user interface differentiating normal from pathological tissues. His hands reached into the brain space and worked as the tools reshaping the resected region. His future was really in his hands. He did not realize that he was the only person in the whole Brainsterium authorized to change the resection plan, as the system was monitoring the user’s biometrics. He started the inventory of his brain in the area of difference between the conservative and
435
436
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
aggressive resection regions. Some skills were there that he might lose, such as climbing, driving, and playing the piano. He had given up his dream to climb K2 a long time ago. His personal driver was doing well so he would keep him. Playing the piano – no compromise. He kept reshaping the region to be resected. The good-bye part of his brain was finally defined. Its knowledge and skills would be attempted to be recovered in a postoperative process. Now to memories and Mr. Green touched the last button. The button-called nurse appeared, put a BCC helmet on his head, and activated an array of transcranial magnetic stimulators. ‘‘I am lucky’’ – he thought – ‘‘she might have been a robot.’’ He projected his brain’s image, positioned the pointer within the region to be resected, and pressed the stimulation button. Nothing happened. He changed location and pressed the button again. Now he was watching the Titanic movie with his first love. He kept on pressing. It worked as a time machine moving him back to distant events and places. He could hardly believe that there were so many memories in such a small piece of tissue. Every memory he evoked was recorded. After the surgery, he would just play back any recorded piece to re-enter it into his brain. The surgery plan was completed. He touched the submit button and was asked to confirm the plan and accept the legal statement. The session was terminated and he was invited for tea. The surgery would start in half an hour’s time. He loved this stuff. ‘‘I have got to subscribe to the Brainsterium Club, so I can come here every weekend for some brain surfing and unlocking’’ – he thought. His new friend, the lion-robot ushered him to the Operating Rooms area. In this high-tech environment it looked so classic and trustworthy. Dr. Noki and the pretty nurse were already there. Mr. Green laid down on the operating table, a BCC helmet was put on his head, and some
monitoring probes attached to his body. ‘‘Are you ready?’’ – asked Dr. Noki and added – ‘‘Do not be afraid. Though the whole procedure is fully automatic, I will be controlling its every step.’’ Mr. Green pressed the start button initiating his own neurosurgery. The stereoscopic image of his brain was projected directly into his retinas and the resection plan prepared by him appeared. Initially, numerous sparkles surrounding the core tumor were visible. Later, he saw blood vessels feeding the core tumor being closed and the tumor separated from its surrounding tissues. Mr. Green felt an injection. He noticed a wire going towards the tumor through the biggest blood vessel which remained still open. ‘‘This has to be a catheter’’ – he recalled. A magnetic system guided its tip automatically towards the tumor. The tip reached the tumor. A small balloon was inflated closing the vessel. The dissolved tumor tissues started disappearing fast. The space previously occupied by the core tumor kept on shrinking. Finally, the last blood vessel was closed. The surgery was over. It was a long, eye-opening day for Mr. Green as a patient and investor. He had to stay overnight at the Brainsterium under monitoring. Scanning was being performed automatically on continuous basis. Everything was normal as expected. In the morning Mr. Green was discharged. He went into Dr. Noki’s office and looked at him in a way that only a few multi-billionaires and CEOs deserved so far. ‘‘So far, so good’’ – Dr. Noki welcomed him. ‘‘We still have some work to do to recover the content of the resected part of your brain. A disk was handed to Mr. Green. ‘‘My preserved memories’’ – he thought. Mr. Green was requested to keep on monitoring at home. He was given a wearable monitor capable of transmitting his scans from his home to the Brainsterium wirelessly. Finally he found some of his contribution.
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
Mr. Green’s driver brought him back home. He realized that despite many urgent messages, he made one call only asking his secretary to donate anonymously to the Brainsterium’s R&D Center. He sat at his old grand piano and started playing his favorite pieces. He was quite happy with his technical performance and got an impression that he played even with a greater passion than before. Now he knew what he was going to invest in.
References 1. Alexander E, III, Nashold BS, Jr. A history of neurosurgical navigation. In: Alexander E, III, Maciunas RJ, editors. Advanced neurosurgical navigation. New York: Thieme; 1999. p. 3-14. 2. Ashkan K, Blomstedt P, Zrinzo L, Tisch S, Yousry T, Limousin-Dowsey P, Hariz MI. Variability of the subthalamic nucleus: the case for direct MRI guided targeting. Br J Neurosurg 2007;21(2):197-200. 3. Kosta P, Argyropoulou MI, Markoula S, Konitsiotis S. MRI evaluation of the basal ganglia size and iron content in patients with Parkinson’s disease. J Neurol 2006;253 (1):26-32. 4. Reich CA, Hudgins PA, Sheppard SK, Starr PA, Bakay RA. A high-resolution fast spin-echo inversionrecovery sequence for preoperative localization of the internal globus pallidus. Am J Neuroradiol 2000;21(5): 928-31. 5. Sather MD, Patil AA. Direct anatomical localization of the subthalamic nucleus on CT with comparison to Schaltenbrand-Wahren atlas. Stereotact Funct Neurosurg 2007;85(1):1-5. 6. Starr PA, Vitek JL, DeLong M, Bakay RA. Magnetic resonance imaging-based stereotactic localization of the globus pallidus and subthalamic nucleus. Neurosurg 1999;44(2):303-13. 7. Zhu XL, Hamel W, Schrader B, Weinert D, Hedderich J, Herzog J, Volkmann J, Deuschl G, Mu¨ller D, Mehdorn HM. Magnetic resonance imaging-based morphometry and landmark correlation of basal ganglia nuclei. Acta Neurochir 2002;144(10):959-69; discussion 968-9. 8. Slavin, KV, Thulborn, KR, Wess C, Nersesyan, H. Direct visualization of the human subthalamic nucleus with 3T MR imaging. Am J Neuroradiol 2006;27(1):80-4. 9. Spiegelmann R, Nissim O, Daniels D, Ocherashvilli A, Mardor Y. Stereotactic targeting of the ventrointermediate nucleus of the thalamus by direct visualization with high-field MRI. Stereotact Funct Neurosurg 2006;84(1): 19-23.
27
10. Benabid AL, Koudsie A, Benazzouz A, Le Bas JF, Pollak P. Imaging of subthalamic nucleus and ventralis intermedius of the thalamus. Mov Disord 2002;17 Suppl 3:S123-9. 11. Cuny E, Guehl D, Burbaud P, Gross C, Dousset V, Rougier A. Lack of agreement between direct magnetic resonance imaging and statistical determination of a subthalamic target: the role of electrophysiological guidance. J Neurosurg 2002;97(3):591-7. 12. Dormont D, Ricciardi KG, Tande´ D, Parain K, Menuel C, Galanaud D, Navarro S, Cornu P, Agid Y, Yelnik J. Is the subthalamic nucleus hypointense on T2-weighted images? A correlation study using MR imaging and stereotactic atlas data. Am J Neuroradiol 2004;25(9):1516-23. 13. Richter EO, Hoque T, Halliday W, Lozano AM, SaintCyr JA. Determining the position and size of the subthalamic nucleus based on magnetic resonance imaging results in patients with advanced Parkinson disease. J Neurosurg 2004;100(3):541-6. 14. Coffey RJ. Stereotactic atlases in printed formats. In: Gildenberg PL, Tasker RR, editors. Textbook of stereotactic and functional neurosurgery. New York: McGraw-Hill; 1998. p. 237-48. 15. Germano IM, Weisz DJ. Computer technology as an adjuvant for target location and validation: In: Israel Z, Burchiel KJ, editors. Microelectrode recording in movement disorder surgery. New York – Stuttgart: Thieme; 2004. p. 152-63. 16. Stancanello J, Romanelli P, Modugno N, Cerveri P, Ferrigno G, Uggeri F, Cantore G. Atlas-based identification of targets for functional radiosurgery. Med Phys 2006;33(6):1603-11. 17. Yelnik J, Damier P, Demeret S, Gervais D, Bardinet E, Bejjani BP, Francois C, Houeto JL, Arnule I, Dormont D, Galanaud D, Pidoux B, Cornu P, Agid Y. Localization of stimulating electrodes in patients with Parkinson disease by using a three-dimensional atlas-magnetic resonance imaging co-registration method. J Neurosurg 2003;99(1):89-99. 18. Speigel EA, Wycis HT. Stereoencephalotomy: part I. Methods and stereotactic atlas of the human brain. New York: Grune and Stratton; 1952. 19. Bertrand G, Olivier A, Thompson CJ. Computer display of stereotaxic brain maps and probe tracts. Acta Neurochir Suppl 1974;21:235-43. 20. Nowinski WL, Benabid AL. New directions in atlasassisted stereotactic functional neurosurgery. In: Germano IM, editor. Advanced techniques in image-guided brain and spine surgery. New York: Thieme; 2002. p. 162-74. 21. Talairach J, David M, Tournoux P. Atlas d’anatomie stereotaxique des noyaux gris centraux. Paris: Masson; 1957. 22. Schaltenbrand G, Bailey W. Introduction to stereotaxis with an atlas of the human brain. Stuttgart: Thieme; 1959.
437
438
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
23. Andrew J, Watkins ES. A stereotaxic atlas of the human thalamus and adjacent structures. A variability study. Baltimore: Williams and Wilkins; 1969. 24. Van Buren JM, Borke RC. Variations and connections of the human thalamus. Berlin: Springer; 1972. 25. Schaltenbrand G, Wahren W. Atlas for stereotaxy of the human brain. Stuttgart: Thieme; 1977. 26. Szikla G, Bouvier G, Hori T. Angiography of the human brain cortex: atlas of vascular patterns and stereotactic localization. Berlin: Springer; 1977. 27. Afshar E, Watkins ES, Yap JC. Stereotactic atlas of the human brainstem and cerebellar nuclei. New York: Raven Press; 1978. 28. Talairach J, Tournoux P. Co-planar stereotactic atlas of the human brain. Stuttgart – New York: Thieme; 1988. 29. Ono M, Kubik S, Abernathey CD. Atlas of the cerebral sulci. Stuttgart: Thieme; 1990. 30. Talairach J, Tournoux P. Referentially oriented cerebral MRI anatomy: atlas of stereotaxic anatomical correlations for gray and white matter. Stuttgart: Thieme; 1993. 31. Morel A, Magnin M, Jeanmonod D. Multiarchitectonic and stereotactic atlas of the human thalamus. J Comp Neurol 1997;387:588-630. 32. Hassler R. Anatomy of the thalamus In: Schaltenbrand G, Bailey W, editors. Introduction to stereotaxis with an atlas of the human brain. Stuttgart: Thieme; 1959. p. 230-90. 33. Walker AE. The primate thalamus. Chicago: University of Chicago Press; 1938. 34. Federative Committee on Anatomical Terminology (FCAT). Terminologia anatomica. Stuttgart – New York: Thieme; 1999. 35. Hardy TL, Deming LR, Harris-Collazo R. Computerized stereotactic atlases. In: Alexander E, III, Maciunas RJ, editors. Advanced neurosurgical navigation. New York: Thieme; 1999. p. 115-24. 36. Kazarnovskaya MI, Borodkin SM, Shabalov VA. 3-D computer model of subcortical structures of human brain. Comput Biol Med 1991;21:451-7. 37. Yoshida M. Three-dimensional maps by interpolation from the Schaltenbrand and Bailey atlas. In: Kelly PJ, Kall BA, editors. Computers in stereotactic neurosurgery. Boston: Blackwell; 1992. p. 143-52. 38. Kall BA. Computer-assisted stereotactic functional neurosurgery. In: Kelly PJ, Kall BA, editors. Computers in stereotactic neurosurgery. Boston: Blackwell; 1992. p. 134-42. 39. Niemann K, Naujokat C, Pohl G, Wollner C, von Keyserlingk D. Verification of the Schaltenbrand and Wahren stereotactic atlas. Acta Neurochir 1994;129 (1–2):72-81. 40. Sramka M, Ruzicky E, Novotny M. Computerized brain atlas in functional neurosurgery. Stereotact Funct Neurosurg 1997;69:93-8. 41. Niemann K, van den Boom R, Haeselbarth K, Afshar F. A brainstem stereotactic atlas in a three-dimensional
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
magnetic resonance imaging navigation system: first experiences with atlas-to-patient registration. J Neurosurg 1999;90(5):891-901. Niemann K, Mennicken VR, Jeanmonod D, Morel A. The Morel stereotactic atlas of the human thalamus: atlas-to-MR registration of internally consistent canonical model. NeuroImage 2000;12(6):601-16. Nowinski WL, Liu J, Thirunavuukarasuu A. Quantification of three-dimensional inconsistency of the subthalamic nucleus in the Schaltenbrand-Wahren brain atlas. Stereotact Funct Neurosurg 2006;84(1):46-55. Nowinski WL, Liu J, Thirunavuukarasuu A. Quantification and visualization of three-dimensional inconsistency of the globus pallidus internus in the SchaltenbrandWahren brain atlas. Stereotact Funct Neurosurg 2006;84(5–6):236-42. Nowinski WL, Liu J, Thirunavuukarasuu A. Quantification and visualization of three-dimensional inconsistency of the ventrointermediate nucleus of the thalamus in the Schaltenbrand-Wahren brain atlas. Acta Neurochir; 2008;150(7): 647-53. Chakravarty MM, Bertrand G, Hodge CP, Sadikot AF, Collins DL. The creation of a brain atlas for image guided neurosurgery using serial histological data. NeuroImage 2006;30(2):359-76. Henderson JM. Integration of stereotactic position with MER. In: Israel Z, Burchiel KJ, editors. Microelectrode recording in movement disorder surgery. New York – Stuttgart: Thieme; 2004. p. 130-7. Nakano N, Taneda M. Three-dimensional atlas of subthalamic nucleus and its adjacent structures. No Shinkei Geka 2005;33(7):683-92. Yelnik J, Bardinet E, Dormont D, Malandain G, Ourselin S, Tande´ D, Karachi C, Ayache N, Cornu P, Agid Y. A three-dimensional, histological and deformable atlas of the human basal ganglia. I. Atlas construction based on immunohistochemical and MRI data. NeuroImage 2007;34(2):618-38. Nowinski WL, Fang A, Nguyen BT, Raphel JK, Jagannathan L, Raghavan R, Bryan RN, Miller G. Multiple brain atlas database and atlas-based neuroimaging system. Comput Aided Surg 1997;2(1):42-66. Nowinski WL. Computerized brain atlases for surgery of movement disorders. Semin Neurosurg 2001;12 (2):183-94. Nowinski WL. Electronic brain atlases: features and applications. In: Caramella D, Bartolozzi C, editors. 3D image processing: techniques and clinical applications. Medical Radiology series, Heidelberg: Springer; 2002. p. 79-93. Nowinski WL. The Cerefy Brain Atlases: continuous enhancement of the electronic Talairach-Tournoux brain atlas. Neuroinformatics 2005;3(4):293-300. Nowinski WL, Bryan RN, Raghavan R. The electronic clinical brain atlas: multiplanar navigation of the human brain. New York: Thieme; 1997.
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
55. Nowinski WL, Thirunavuukarasuu A, Kennedy DN. Brain atlas for functional imaging: clinical and research applications. New York: Thieme; 2000. 56. Nowinski WL, Thirunavuukarasuu A. The cerefy clinical brain atlas on CD-ROM. New York: Thieme; 2004. 57. Nowinski WL, Thirunavuukarasuu A, Benabid AL. The cerefy clinical brain atlas: enhanced edition with surgical planning and intraoperative support. New York: Thieme; 2005. 58. Liu J, Nowinski WL. A hybrid approach to shape-based interpolation of stereotactic atlases of the human brain. Neuroinformatics 2006;4(2):177-98. 59. Niemann K, van Nieuwenhofen I. One atlas – three anatomies: relationships of the Schaltenbrand and Wahren microscopic data. Acta Neurochir 1999;141: 1025-103. 60. Nowinski WL, Thirunavuukarasuu A. Atlas-assisted localization analysis of functional images. Med Image Anal 2001;5(3):207-20. 61. Nowinski WL. Co-registration of the SchaltenbrandWahren microseries with the probabilistic functional atlas. Stereotact Funct Neurosurg 2004;82:142-6. 62. Nowinski WL, Thirunavuukarasuu A. Quantification of spatial consistency in the Talairach and Tournoux stereotactic atlas. Acta Neurochir; 2009 (in press), DOI: 10.1007/s00701-009-0364-8. 63. Nowinski WL. Modified Talairach landmarks. Acta Neurochir 2001;143(10):1045-57. 64. Finnis K, Starreveld Y, Parrent A, Sadikot A, Peters T. Three-dimensional database of subcortical electrophysiology for image-guided stereotactic functional neurosurgery. IEEE Trans Med Imag 2003;22(1):93-104. 65. Collins DL, Evans AC. ANIMAL: validation and application of non-linear registration-based segmentation. Int J Pattern Recogn Artif Intell 11(8):1271–94. 66. D’Haese P, Cetinkaya E, Kao C, Fitzpatrick J, Konrad P, Dawant B. Toward the creation of an electrophysiological atlas for the pre-operative planning and intra-operative guidance of deep brain stimulators (DBS) implantation. In: Proceedings of the international conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI 2004), Saint-Malo, France. LNCS, vol. 3216, Berlin: Springer; 2004. p. 729-36. 67. Rohde GK, Aldroubi A, Dawant BM. The adaptive bases algorithm for intensity based non-rigid image registration. IEEE Trans Med Imag 2003;22(11):1470-9. 68. Nowinski WL, Belov D, Benabid AL. An algorithm for rapid calculation of a probabilistic functional atlas of subcortical structures from electrophysiological data collected during functional neurosurgery procedures. NeuroImage 2003;18(1):143-55. 69. Nowinski WL, Belov D, Benabid AL. A communitycentric internet portal for stereotactic and functional neurosurgery with a probabilistic functional atlas. Stereotact Funct Neurosurg 2002;79:1-12.
27
70. Nowinski WL, Belov D, Pollak P, Benabid AL. A probabilistic functional atlas of the human subthalamic nucleus. Neuroinformatics 2004;2(4):381-98. 71. Nowinski WL, Belov D, Pollack P, Benabid AL. Statistical analysis of 168 bilateral subthalamic nucleus implantations by means of the probabilistic functional atlas. Neurosurg 2005;57 Suppl 4:319-30. 72. Nowinski WL, Belov D, Thirunavuukarasuu, A. Benabid AL. A probabilistic functional atlas of the VIM nucleus constructed from pre-, intra- and post-operative electrophysiological and neuroimaging data acquired during the surgical treatment of Parkinson’s disease patients. Stereotact Funct Neurosurg 2006;83(5–6):190-6. 73. Nowinski WL, Thirunavuukarasuu A, Liu J, Benabid AL. Correlation between the anatomical and functional human subthalamic nucleus. Stereotact Funct Neurosurg 2006;85:88-93. 74. Benabid AL, Pollak P, Gross C, Hoffmann D, Benazzouz A, Gao DM, Laurent A, Gentil M, Perret J. Acute and long-term effects of subthalamic nucleus stimulation in Parkinson’s disease. Stereotact Funct Neurosurg 1994;62(1–4):76-84. 75. Benabid AL, Koudsie A, Benazzouz A, Fraix V, Ashraf A, Le Bas JF, Chabardes S, Pollak P. Subthalamic stimulation for Parkinson’s disease. Arch Med Res 2000;31 (3):282-9. 76. Krack P, Fraix V, Mendes A, Benabid AL, Pollak P. Postoperative management of subthalamic nucleus stimulation for Parkinson’s disease. Mov Disord 2002;17 Suppl 3:S188-97. 77. Tuch DS, Wedeen VJ, Dale AM, George JS, Belliveau JW. Conductivity tensor mapping of the human brain using diffusion tensor MRI. Proc Natl Acad Sci USA 2001;98 (20):11697-701. 78. McIntyre CC, Mori S, Sherman DL, Thakor NV, Vitek JL. Electric field and stimulating influence generated by deep brain stimulation of the subthalamic nucleus. Clin Neurophysiol 2004;115(3):589-95. 79. Butson CR, Cooper SE, Henderson JM, McIntyre CC. Patient-specific analysis of the volume of tissue activated during deep brain stimulation. NeuroImage 2007;34 (2):661-70. 80. Miocinovic S, Noecker AM, Maks CB, Butson CR, McIntyre CC. Cicerone: stereotactic neurophysiological recording and deep brain stimulation electrode placement software system. Acta Neurochir Suppl 2007;972: 561-7. 81. Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, Schwalb JM, Kennedy SH. Deep brain stimulation for treatment-resistant depression. Neuron 3 2005;45(5):651-60. 82. Mazzone P, Lozano A, Stanzione P, Galati S, Scarnati E, Peppe A, Stefani A. Implantation of human pedunculopontine nucleus: a safe and clinically relevant target in Parkinson’s disease. Neuroreport 2005;16(17):1877-81. 83. Nowinski, WL. Towards construction of an ideal stereotactic brain atlas. Acta Neurochir 2008;150(1):1-14.
439
440
27
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
84. Chakravarty MM, Sadikot AF, Mongia S, Collin DL. Towards a multi-modal atlas for neurosurgical planning. In: Proceedings of the ninth international conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI 2006), Copenhagen, Denmark. LNCS, vol. 4191, Berlin: Springer; 2006. p. 389-96. 85. Toga AW, editor. Brain warping. San Diego: Academic Press; 1998. 86. Warfield SK, Guimond A, Roche A, et al. Advanced nonrigid registration algorithms for image fusion. In: Toga AW, Mazziotta JC, editors. Brain mapping. The methods. 2nd ed. San Diego: Academic Press; 2002. p. 661-90. 87. Nowinski WL, Qian G, Bhanu Prakash KN, Hu Q, Aziz A. Fast talairach transformation for magnetic resonance neuroimages. J Comput Assist Tomogr 2006;30 (4):629-41. 88. Christensen GE, Joshi SC, Miller MI. Volumetric transformation of brain anatomy. IEEE Trans Med Imaging 1997;16(6):864-77. 89. Grachev D, Berdichevsky D, Rauch SL, Heckers S, Kennedy DN, Caviness VS, Alpert NM. A method for assessing the accuracy of intersubject registration of the human brain using anatomic landmarks. NeuroImage 1999;9(2):250-68. 90. Nowinski WL. Anatomical targeting in functional neurosurgery by the simultaneous use of multiple Schaltenbrand-Wahren brain atlas microseries. Stereotact Funct Neurosurg 1998;71:103-16. 91. Hu Q, Nowinski WL. A rapid algorithm for robust and automatic extraction of the midsagittal plane of the human cerebrum from neuroimages based on local symmetry and outlier removal. NeuroImage 2003;20(4): 2154-66. 92. Bhanu Prakash KN, Hu Q, Aziz A, Nowinski WL. Rapid and automatic localization of the anterior and posterior commissure point landmarks in MR volumetric neuroimages. Acad Radiol 2006;13(1):36-54. 93. Hu Q, Qian G, Nowinski WL. Fast, accurate and automatic extraction of the modified Talairach cortical landmarks from MR images. Magn Reson Med 2005;53: 970-6. 94. Nowinski WL, Bhanuprakash KN. Dorso-ventral extension of the Talairach transformation and its automatic calculation for MR neuroimages. J Comp Assist Tomogr 2005;29(6):863-79. 95. Saint-Cyr JA, Hoque T, Pereira LC, Dostrovsky JO, Hutchison WD, Mikulis DJ, Abosch A, Sime E, Lang AE, Lozano AM. Localization of clinically effective stimulating electrodes in the human subthalamic nucleus on magnetic resonance imaging. J Neurosurg 2002;97(5): 1152-66. 96. Nowinski WL, Yang GL, Yeo TT. Computer-aided stereotactic functional neurosurgery enhanced by the use of the multiple brain atlas database. IEEE Trans Med Imaging 2000;19(1):62-9.
97. Fukuda M, Mentis M, Ghilardi MF, Dhawan V, Antonini A, Hammerstad J, Lozano AM, Lang A, Lyons K, Koller W, Ghez C, Eidelberg D. Functional correlates of pallidal stimulation for Parkinson’s disease. Ann Neurol 2001;49(2):155-64. 98. Ganser KA, Dickhaus H, Metzner R, Wirtz CR. A deformable digital brain atlas system according to Talairach and Tournoux. Med Image Anal 2004;8(1):3-22. 99. Kaur G, Tan J, Alam M, Chaudhary V, Chen D, Dong M, Eltahawy H, Fotouhi F, Gammage C, Gong J, Grosky W, Guthikonda M, Hu J, Jeyaraj D, Jin X, King A, Landman J, Lee J, Li QH, Lufei H, Morse M, Patel J, Sethi I, Shi W, Yang K, Zhang Z. CASMIL: a comprehensive software/toolkit for image-guided neurosurgeries. Int J Med Robot 2006;2(2):123-38. 100. Nowinski WL, Belov D. The cerefy neuroradiology atlas: a Talairach-Tournoux atlas-based tool for analysis of neuroimages available over the internet. NeuroImage 2003;20(1):50-7. 101. Shabalov VA, Kazarnovskaya MI, Borodkin SM, Kadin AL, Krivosheina VY, Golanov AV. Functional neurosurgery using 3-D computer stereotactic atlas. Acta Neurochir Suppl (Wien) 1993;8:65-7. 102. Serra L, Nowinski WL, Poston T, Ng H, Lee CM, Chua GG, Pillay PK. The brain bench: virtual tools for stereotactic frame neurosurgery. Med Image Anal 1997;1(4): 317-29. 103. St-Jean P, Sadikot AF, Collins L, Clonda D, Kasrai R, Evans AC, Peters TM. Automated atlas integration and interactive three-dimensional visualization tools for planning and guidance in functional neurosurgery. IEEE Trans Med Imaging 1998;17(5):673-80. 104. Nowinski WL, Yeo TT, Thirunavuukarasuu A. Microelectrode-guided functional neurosurgery assisted by electronic clinical brain atlas CD-ROM. Comput Aided Surg 1998;3(3):115-22. 105. Nowinski WL, Thirunavuukarasuu A. A locus-driven mechanism for rapid and automated atlas-assisted analysis of functional images by using the brain atlas for functional imaging. Neurosurg Focus 2003;15(1): Article 3. 106. Benabid AL, Nowinski WL. Intraoperative robotics for the practice of neurosurgery: a surgeon’s perspective. In: Apuzzo ML, editor. The operating room for the 21st century. Rolling Meadows: American Association of Neurological Surgeons; 2003. p. 103-18. 107. Ahsan RL, Allom R, Gousias IS, Habib H, Turkheimer FE, Free S, Lemieux L, Myers R, Duncan JS, Brooks DJ, Koepp MJ, Hammers A. Volumes, spatial extents and a probabilistic atlas of the human basal ganglia and thalamus. NeuroImage 2007;38(2):261-70. 108. Nowinski WL, Thirunavuukarasuu A, Volkau, Marchenko Y, Runge VM. The cerefy atlas of cerebral vasculature. New York: Thieme; 2009. 109. Lawes IN, Barrick TR, Murugam V, Spierings N, Evans DR, Song M, Clark CA. Atlas-based segmentation of
Anatomical and probabilistic functional atlases in stereotactic and functional neurosurgery
110.
111.
112.
113.
114.
115.
white matter tracts of the human brain using diffusion tensor tractography and comparison with classical dissection. NeuroImage 2008;39(1):62-79. Kim SJ, Kim IJ, Kim YK, Lee TH, Lee JS, Jun S, Nam HY, Lee JS, Kim YK, Lee DS. Probabilistic anatomic mapping of cerebral blood flow distribution of the middle cerebral artery. J Nucl Med 2008;49(1):39-43. Nowinski WL, Qian G, Bhanu Prakash KN, Thirunavuukarasuu A, Hu QM, Ivanov N, Parimal AS, Runge VL, Beauchamp NJ. Analysis of ischemic stroke MR images by means of brain atlases of anatomy and blood supply territories. Acad Radiol 2006;13(8):1025-34. Shattuck DW, Mirza M, Adisetiyo V, Hojatkashani C, Salamon G, Narr KL, Poldrack RA, Bilder RM, Toga AW. Construction of a 3D probabilistic atlas of human cortical structures. NeuroImage 2007;39(3): 1064-80. Chan E, Kovacevı´c N, Ho SK, Henkelman RM, Henderson JT. Development of a high resolution threedimensional surgical atlas of the murine head for strains 129S1/SvImJ and C57Bl/6J using magnetic resonance imaging and micro-computed tomography. Neuroscience 2007;144(2):604-15. Nowinski WL, Belov D. Towards atlas-assisted automatic interpretation of MRI morphological brain scans in the presence of tumor. Acad Radiol 12:1049-57. Nimsky C, Ganslandt O, Hastreiter P, Wang R, Benner T, Sorensen AG, Fahlbusch R. Preoperative and intraoperative diffusion tensor imaging-based fiber
116.
117.
118.
119.
27
tracking in glioma surgery. Neurosurg 2005;56(1): 130-7; discussion 138. Castro FJ, Pollo C, Meuli R, Maeder P, Cuisenaire O, Cuadra MB, Villemure JG, Thiran JP. A cross validation study of deep brain stimulation targeting: from experts to atlas-based, segmentation-based and automatic registration algorithms. IEEE Trans Med Imaging 2006;25(11):1440-50. Kockro RA, Serra L, Yeo TT, Chan C, Sitoh YY, Chua GG, Ng H, Lee E, Lee YH, Nowinski WL. Planning and simulation of neurosurgery in a virtual reality environment. Neurosurg 2000;46(1): 118-37. Anil SM, Kato Y, Hayakawa M, Yoshida K, Nagahisha S, Kanno T. Virtual 3-dimensional preoperative planning with the dextroscope for excision of a 4th ventricular ependymoma. Minim Invasive Neurosurg 2007;50(2): 65-70. Wong GK, Zhu CX, Ahuja AT, Poon WS. Craniotomy and clipping of intracranial aneurysm in a stereoscopic virtual reality environment. Neurosurg 2007;61(3): 564-8; discussion 568-9.
441
29 Development of a Classic: The Todd-Wells Apparatus, the BRW, and the CRW Stereotactic Frames J. Arle
In many ways, the history of refined stereotactic surgery, particularly in the US, is intimately related to the concurrent progression in development of three devices, from the Todd-Wells apparatus, through the BRW frame, to ultimately the CRW frame system. It would involve the important use and feedback of many pioneers in stereotactic and function neurosurgery including Apuzzo (who used both BRW frame #1 and CRW frame #1) (Cosman E, 2008, personal communication), Gildenberg, Heilbrun, Nashold, and others. A catalogue of devices was created over the years for human stereotaxy, from the Zernov encephalometer in 1889 [1] to the multitude of instruments that arose from the publication in 1947 by Spiegel et al. [2], extending the ability of progressive and innovative neurosurgeons to attempt reliable intracranial targeting throughout the 1950s and 1960s, performing tens of thousands of stereotactic procedures [3]. However, these attempts were inevitably hampered by rudimentary imaging of intracranial anatomy. The best technique remained ventriculography, developed by Walter Dandy between 1916 and 1919 [4], though these were relatively uncomfortable procedures by most accounts, and invasive, and limited in their reliability by patient positioning at the time of the stereotactic procedure relative to when imaging was performed, and potentially by the anatomical consistency of locating intracranial structures with reference to #
Springer-Verlag Berlin/Heidelberg 2009
ventricular widths and margins [5]. Despite these limitations, and combined with sophistication in human brain atlas detail, an emergence and, in breadth and sheer numbers, dominance of stereotactic procedures occurred over two decades in the 1950s and 1960s, pushing the field beyond the open and unpredictable procedures attempted earlier (e.g., anterior choroidal artery ligation by Cooper [6] or open ansotomy by Meyers [7]).
The Todd-Wells Apparatus Trent Wells, Jr. (> Figure 29-1a) had begun an engineering firm in that same seminal year of 1947 and he became involved in developing instruments for stereotaxy in animals at UCLA with Horace Magoun and Jack French [9]. Fate would eventually lead him to be involved in the design solutions and manufacturing of many important stereotactic devices, in addition to developing Gardner-Wells tongs for spinal traction. Early on, he found himself collaborating with a young neurosurgeon named Edwin Todd (> Figure 29-1b), who had become interested in surgery for movement disorders during his neurosurgical training at the Cleveland Clinic. As expressed by Wells and Todd themselves [9], ‘‘it was at this time, in the early 1960s, that the fortuitous meeting of a prospective stereotactic surgeon and a highly motivated stereotactic
454
29
Development of a classic: the Todd-Wells apparatus, the BRW, and the CRW stereotactic frames
. Figure 29-1 (a) Trent Wells, Jr, holding the BRW frame. Few were ever as involved in so many critical design innovations within the history of stereotaxy. He provided creative and durable engineering and machining for the development of at least five different stereotactic systems with three different neurosurgeons (Rand, Todd, and Roberts) [8] (b) Edwin M. Todd, neurosurgeon and instrumental motivation in the design and refinement of the Todd-Wells stereotactic apparatus at UCLA. Developed at the height of the era using ventriculography, this frame became one of the most widely used in the world [8]
apparatus inventor occurred; the former knew what he wanted, but not how to get it and the latter had the experience and the engineering ingenuity to turn seminal idea into reality.’’ All stereotaxy must develop a reliable coordinate system, whether Cartesian or not, that allows one to link such a system to particular anatomical landmarks, creating invariance that survives the vagaries of human anatomy and asymmetries of device placement. Ultimately, it also became clear that such invariance needed to achieve approximately 1 mm of spatial accuracy with every use. Left with only X-ray technology at the time as an imaging modality, Wells realized nonetheless that accuracy using biplanar X-rays depended not only on the refinement of the instrument per se, but on the ability to eliminate variation in X-ray magnification by placing the film cassettes at the same distance from the source in both planes every time. This could
be achieved by fixing the X-ray sources within the operating room, and placing the target at the precise focal point of both film planes each time by moving the head with the head frame. Wells had already begun part of his intracranial targeting education in the 1950s when he worked with neurosurgeon Robert Rand to develop and manufacture a transverse arc system for performing pallidectomies and then a second device for hypophysectomies [10]. Modifying this from a bur hole mounted system with an arc, to using a head mounted frame, and using Lucite reticules similar to gunsights on fighter aircraft (Wells had been an accomplished fighter pilot in WWII) in both lateral and AP directions, the desired target could be positioned along the arc axis and then accessed by any number of approaches on that side of the hemisphere. This device was the first modular human stereotactic instrument, using
Development of a classic: the Todd-Wells apparatus, the BRW, and the CRW stereotactic frames
29
. Figure 29-2 Conceptual schematic of the Todd-Wells apparatus, showing the focal point, defined by the intersection of precalibrated AP and lateral imaging with ventriculography. The head is then moved in the base ring to the make the desired target meet the focal point. Reliability can be ensured by fixing the X-ray sources and the frame mount in the OR [9]
two different arcs and allowing access to a target from virtually any direction. Ideally, however, fixed X-ray sources in the OR could be set up, adjusted appropriately by collimators, and with the multiaxial movement provided by the head holder, create a very versatile and reliable stereotactic device (see > Figures 29-2–29-4). It was technically similar, it turns out, to the device developed by Schaltenbrand in the Wurzburg operating suite in 1959, whereby X-ray sources were fixed 4m away and the target was moved to the focal point by moving the base ring [11]. The Todd-Wells device was subsequently used by several others in developing stereotactic foreign body removal, placement of depth electrodes,
electrothrombosis and metallothrombosis of intracranial aneurysms, radiofrequency lesioning of multiple targets, and even percutaneous cordotomy [9]. Given the imaging constraints of that time, this device was likely one of the most reliably accurate means of achieving accurate targeting for standard biopsies as well as the other procedures developed using the device, functional and otherwise. Although reliance on X-rays was widely practiced for many years, particularly in conjunction with ventriculography, a consistent and reliable means of ensuring accurate intracranial access was brought to bear in the Todd-Wells device, first presented to the neurosurgical community at large during the
455
456
29
Development of a classic: the Todd-Wells apparatus, the BRW, and the CRW stereotactic frames
. Figure 29-3 The Todd-Wells apparatus. Rigid, and accurate when X-ray sources are appropriately calibrated, and reliable [9]
International Neurological Surgery Meeting in Copenhagen in 1965 [9]. Commercialization of the device was initially procured by Codman [12] but eventually transferred to Radionics, helping broaden use of the device by multiple centers. Freidman and Coffey referred to it as one of the most widely used traditional stereotactic devices in the world [13].
The BRW But the nature of imaging changed radically in the 1970s with the development of the CT scanner, allowing for the first time, the potential of obtaining digital information for all three axes. The key was redesigning stereotactic devices to take advantage of this new informational largess, and Wells was perhaps the best – positioned engineer in the world for this task. Many were considering how to exploit this new development, realizing that one obvious advantage was that if one could extract relative coordinates from one axial scan slice to another, then the necessarily fixed referentiality between frame,
target and imaging source could be decoupled, and the inaccuracies and asymmetries of frame placement, human anatomy, and imaging calibrations could impact accuracy less overall, and make the process of obtaining stereotactic spatial coordinates easier in general. Through the collaborative work of Dr. Ted Roberts, chairman of neurosurgery at the University of Utah and Russell Brown, then a 3rd year medical student at the university, several key new concepts were developed that culminated in a new CT-compatible stereotactic frame, presented initially in 1979. Roberts and Brown introduced their conceptualizations to Trent Wells in 1978, and given his expertise in design translation, machining, and manufacturing of stereotactic devices for animal research over 30 years, Wells was able to distill their key innovations into a real device, capable of exceptional accuracy, consistency, sterilizability, and sturdiness. This became known as the BRW (BrownRoberts-Wells) frame and CT localizer system. Roberts was motivated by the need for more reliable and improved accuracy in stereotactic procedures, particularly in tumor biopsies [14], and the promise that CT brought was the ability better than ever to see lesions, with the additional potential to provide a coordinate structure to access them. Initially, Brown worked on using a 3-D multiplanar imaging system developed by Evans and Sutherland Computer Corporation in Salt Lake City, Utah, to explore methods of improving biopsy accuracy and yield [14]. He created software to use outlines of CT images of the brain and lesions, graphically stacking them together for manipulation [14]. When the Varian V-360-CT body scanner became available in Utah, he instead shifted to developing a prototype frame that could use the CT planar information as a potential coordinate reference frame [14]. This initial prototype was made by Brown in Lucite (> Figure 29-5) and incorporated versions of several innovations described below. The first
Development of a classic: the Todd-Wells apparatus, the BRW, and the CRW stereotactic frames
29
. Figure 29-4 Drawings showing the versatility of the Todd-Wells system. The wide array of procedures enabled by this frame helped create its widespread use. Note, however, the difficulty involved in taking a low lateral trajectory to a target. This limitation would prove to be one of the motivations in development of both the BRW and CRW later [9]
innovation was in finding a means to relate each axial planar slice uniquely to the frame. This was accomplished by Brown’s idea to use two rings suspended apart by three sets of rods arranged in ‘‘N’’ patterns. The ‘‘localizer ring’’ as it was called attached to the frame base in only one way with unique ball clamps, minimizing human error. As a separate unit, the localizer continued the modularity of the system, the sturdiness, and the relative ease of sterilization of system components. The initial Lucite prototype used four vertical rods with three diagonal rods between
them, running contiguously along the ring margin. The eventual metal commercial version would use six vertical epoxy-graphite rods with the three pairs of them separated by a diagonal rod and spread in these 3-rod sets equidistantly around the localizer ring. Such a configuration, widely adapted in modified ways by others later, allowed each axial slice to also have a unique set of nine nearly-circular cross-sections of the rods around the edge of the scan image (> Figure 29-6). The distances between the outer two rods and the center of the diagonal
457
458
29
Development of a classic: the Todd-Wells apparatus, the BRW, and the CRW stereotactic frames
rod, for each of the three sets, allowed for precise determination of the axial plane orientation to the frame, thus allowing any point in space within the localizer ring to have a unique x, y, and z . Figure 29-5 Prototype of the BRW frame, created by Russell Brown. Made of Lucite, it was composed of interlocking arcs and incorporated the innovative ‘‘N’’ shaped fiducial rods along the edge. Note that in this prototype the configuration of fiducials uses four contiguous rods with three diagonals interspersed. This was ultimately changed in the final localizer ring design [14]
set of coordinates. The transformation matrix from the two-dimensional axial image to full 3-D coordinates is given in > Figure 29-7. Importantly, the ‘‘z’’ or vertical height from the frame base was precisely related to the relative x and y inter-rod distances, freeing users from both the requirement that the patient be fixed to the scanner table and from, as Roberts put it, ‘‘the unreliable scanner table movements required by earlier CT equipment’’ [14]. Brown ultimately went on to patent this aspect of the design and other aspects as well [16]. The second main innovation derived from the motivation to allow targeting to all locations in the head from virtually any access point and trajectory through the hemispheres, a limited characteristic in most frames until then. This was accomplished primarily by allowing the . Figure 29-7 Transform matrix to convert from 2-D to 3-D coordinates. The vertical ‘‘Z’’ value is intrinsically related to the relative distances between the center diagonal fiducial and the two outer fiducials [15]
. Figure 29-6 Drawing showing the BRW and localizer ring in place, denoting to the right axial images at three different vertical reference lines with fiducial configurations. The interfiducial distances allow unique coordinates to be computed for any point within that localizer volume [14]
. Figure 29-8 Schematic showing the four angles that need to be used in computing the trajectory with the BRW frame. The fifth degree of freedom is the probe length. Despite the cumbersome nature of this solution, it allowed a much wider variation of target access [15]
Development of a classic: the Todd-Wells apparatus, the BRW, and the CRW stereotactic frames
rotation of the probe holder along a rotational base ring and including several other adjustable angulations and a probe holder pivot, in order to reach the desired target. Calculations were required involving four unique angles (alpha, beta, gamma, and delta), but once learned, the method was trustworthy and consistent. This was not a target-centered device and calculations of each of the necessary four angles to finalize the trajectory to the target was at first cumbersome, and required running the computations for several minutes on an adjunct DEC
29
machine, the equivalent of one of today’s desktop computers (Cosman E, 2008, personal communication). The device uses a polar coordinate system with 5 degrees of freedom (> Figure 29-8). Alpha refers to the base-ring angle of rotation. Beta refers to an up to 30 pivot in the same plane as Alpha. Gamma refers to a measure of angle perpendicular to the base ring, up to 180 , and Delta refers to an up to 90 pivot of the probe holder in the same plane as Gamma. The fifth degree of freedom is the probe length. Values can be read off the engraved vernier scales
. Figure 29-9 (a) BRW frame mounted on the base ring [14] (b) BRW base ring with the BRW localizer ring attached [14] (c) BRW phantom base allowed users to check trajectories and target alignment before the procedure as an internal safety check and calibration [14]
459
460
29
Development of a classic: the Todd-Wells apparatus, the BRW, and the CRW stereotactic frames
. Figure 29-10 (a) Original letter written by Dr. Eric Cosman to Trent Wells Jr in 1980 before the Houston stereotactic and functional meeting explaining that he believed he had formulated the solutions to transform the BRW to an arc-radius type of device and giving the derivation for a simplified determination of the x, y, and z coordinates of the target (courtesy of Dr. Eric Cosman) (b-d) Original handwritten calculations from ‘‘27 Sept. 1980’’ within which Dr. Cosman derives the CT-to-frame coordinate transformations, which ultimately showed the feasibility of the CRW design (courtesy of Dr. Eric Cosman)
Development of a classic: the Todd-Wells apparatus, the BRW, and the CRW stereotactic frames
on the arc system itself. Several efficiencies were made in short order to ease calculations, however, with first the introduction of the laptop-sized Epson Hx-20, followed by code written for the HP-41cv handheld calculator by Dr. Eric Cosman (Cosman E, 2008, personal communication), securing relative ease of use by many more surgeons within the field of stereotaxy. Wells had superbly transformed Brown’s prototype in Lucite into a robust reliable instrument, relatively free of X-ray artifact. Between 1981 (when the BRW was first commercially offered) and 1988,
29
the system became the most popular stereotactic frame system in the world.
The CRW As the liberal use of brain imaging progressed in quality and clinical importance, market forces also began to play a role in improvements from one manufacturer to another. While the BRW enjoyed relative success worldwide during the 1980s, two other frame systems also made a
. Figure 29-11 (a) Original hand drawn schematics of the CRW arc system, emphasizing several of its key features – the ability to translate the arc itself, (b and d) the ability of the trunion and arc composite to be relocated to a lateral aspect of the base thereby allowing direct lateral or otherwise difficult trajectories (c) and the ability to move the probe holder along the arc maintaining the same target (courtesy of Radionics, Inc)
461
462
29
Development of a classic: the Todd-Wells apparatus, the BRW, and the CRW stereotactic frames
. Figure 29-12 Photograph showing (left to right) Cosman, Roberts, and Wells behind the CRW system at the Radionics booth at the Toronto AANS meeting in 1988, the year the CRW was introduced commercially (courtesy of Dr. Eric Cosman)
relatively successful transition to CT, and then MRI – the Leksell frame and the ReichertMundinger frame. Both of these systems contributed important innovations to frame technology that were subsequently incorporated into modifications to the BRW frame. Leksell had already created what came to be called a ‘‘target centered’’ arc-radius system in 1949 [17], whereby an arc is mounted on the moveable frame which has a semicircular configuration. The probe, or other device, is mounted on the arc and can be moved along the arc into almost any desired location – the innovation being that the probe is made to be equal in length to the radius of the arc, thereby making every location for its trajectory terminate at the desired target. This greatly simplifies the preoperative calculations necessary for planning the setup of the system. Several others successfully incorporated this concept into frame designs, and then modified them later as well for use with CT [18,19], but they had limitations in certain trajectories, setup, or were simply never commercialized to any significant degree. The innovation of Mundinger, working with Reichert and Wolff, was the use of a phantom
. Figure 29-13 (a, b) Photographs of a similar target trajectory using the CRW (left) and BRW (right) frames, but highlighting the difference between the arc-radius type of system (CRW) and the original polar coordinate-based BRW [20]
Development of a classic: the Todd-Wells apparatus, the BRW, and the CRW stereotactic frames
29
. Figure 29-14 Four different examples showing the versatility of the CRW system, allowing a wide variety of trajectories to target: (a) arc translation, (b) straight lateral, (c) posterior translation, and (d) below the base ring transnasal [20]
base. This was also made available in the BRW system and served as a convenience for evaluating the trajectory and arc setup beforehand but also, more importantly perhaps, created an internal check making sure the surgeon had set up coordinates correctly (> Figure 29-9). Cosman understood, perhaps better than anyone, that neurosurgeons in general may have been willing to use the more complicated angular calculations involved with the BRW, but if faced with an easier to use and equivalent device, they would inevitably switch allegiances. Moreover, some trajectories, such as straight lateral or a rising transnasal approach, were not possible with
earlier versions or other frames. He tried to convince Wells to consider a target-centered arc system. Wells was reluctant at first (Cosman E, 2008, personal communication), but Cosman calculated out the transformation from the BRW semi-polar coordinate system to the target-centered semicircular arc-radius device he envisioned, and sent Wells a putative drawing to consider. > Figure 29-10 shows the handwritten calculations and the original letter Cosman wrote to Wells after solving the coordinate transforms. > Figure 29-11a–d shows his original drawings of the CRW and how it solved these remaining access problems that had eluded other frames
463
464
29
Development of a classic: the Todd-Wells apparatus, the BRW, and the CRW stereotactic frames
and even the BRW to some degree. Wells came around, and agreed he could build such a device (Cosman E, 2008, personal communication), producing what then replaced the BRW system in 1988 commercially – the Cosman-Roberts-Wells (CRW) stereotactic frame (> Figure 29-12). While the nature of the arc system and initial planning calculations changed, the localizer ring remained essentially the same (though with only two sets of vertical rods and three diagonals), as this innovation was proven to be a streamlined means of extracting the CT, and MRI, axial fiducial information. The CRW localizer ring was a modest simplification of the BRW localizer ring, allowing one (if the frame base is aligned nearly parallel with scanner gantry angle) simply to read off the vertical value by noting the distance between the diagonal middle rod and the closest vertical rod. The CRW may still be used with the BRW localizer ring, however, without any change in technique. The base of the CRW system, and this may be one of its most significant innovations, is a square with slots to hold the trunion apparatus equivalently in either the AP direction or the lateral direction, and the ring slides were reproduced on both sides (a benefit by averaging out potential error that might occur if only one were used). Even the BRW could not go to a horizontal trajectory or below. The arc can slide from one laterality to the other, and the probe-holder may be positioned from one end of the arc to the other, a combination that allows the widest possible number of trajectories of any frame, all 160 mm from the target with the standard probe holder. This distance was chosen by Cosman based on using anthropomorphic model heads he had in his lab (Cosman E, 2008, personal communication), adjusting the arc size to allow enough distance to mount certain accessories if needed, but not so much distance that the arc becomes unwieldy, especially in an OR environment where an efficient design pays dividends. > Figure 29-13a,b show the essential difference in
the geometrical solutions to solving the stereotactic targeting and trajectory problem between the BRW and the simpler, more versatile arcradius CRW [20]. Since its inception in the late 1980s, the CRW system has been updated in materials and accessories several times. The versatility of the system for accessing intracranial targets is unparalleled. > Figure 29-14 shows only four such variations possible with the standard system [20]. The current version of the CRW frame (shown in > Figure 29-15), has become lighter, but retains the same degree of broad neurosurgical applicability. Typically, the current system is used with a software interface provided by Radionics (Stereocalc and OmniSight Excel, copyright, Radionics Corporation) that not only calculates the appropriate coordinate transform from the fiducials on the scan with a fiducial automatic search and algorithm, but also graphically shows the frame itself in an interactive window, allows . Figure 29-15 Photograph of the current CRW system on the phantom base (courtesy of Radionics, Inc)
Development of a classic: the Todd-Wells apparatus, the BRW, and the CRW stereotactic frames
29
. Figure 29-16 Software included now with the CRW system, example screens shown in (a) and (b), allow the surgeon to automatically locate the centroid of the fiducials and localize the scan in the system, graphically pick targets, trajectories, fuse imaging modalities, visualize in 3-D graphics, overlay anatomical atlases, and plan many aspects of surgery with a user interface that underscores the enormous leaps technology has made in the last 30 years (courtesy of Radionics, Inc)
465
466
29
Development of a classic: the Todd-Wells apparatus, the BRW, and the CRW stereotactic frames
for a virtually error-free image fusion with other imaging modalities (fusing a CT in the frame, for example, with a prior MRI series done previously without the frame applied), and gives the user access to a digitized brain atlas, measurement techniques and 3-D brain surface renderings for sophisticated operative planning (> Figure 29-16).
Presently, the CRW system, with its clean design, robust manufacturing and quality control, wide array of adjunctive instruments (from angiographic localizers, repeat fixation kits, laser holders, interstitial radiotherapy holders, brain retractor attachments, depth electrode placement kits, and wide array of biopsy needles, microdrives, and offset probe holders), accuracy, and
. Figure 29-17 (a) Photographs showing the typical setup of the frame, (b) frame application, (c) alignment to intended target using the phantom base in the OR, and (in only one of many uses for the CRW frame) (d) mounting of a microdrive for microelectrode recording for placement of a DBS lead in the STN of a Parkinsons Disease patient – a modern incarnation of what has been taking place in one form or another for more than 60 years
Development of a classic: the Todd-Wells apparatus, the BRW, and the CRW stereotactic frames
sterilizability, has risen to pre-eminence as a stereotactic device. For example, a recent study of centers performing deep brain stimulation (DBS) for Parkinsons Disease found that almost half of the centers responding to questionnaires regarding their stereotactic technique for DBS used the CRW frame (> Figure 29-17). These 36 centers had performed over 4,500 DBS implantations [21]. The system continues to sell well on a worldwide basis. There will be further refinements and advancements but the story of progress, from the Todd-Wells apparatus to the CRW system, remains a fantastic example of brilliant collaboration between neurosurgery, engineering, technology, and commercialization.
References 1. Speigel EA, Wycis HT, Marks M, Lee A. Stereotaxis apparatus for operations on the human brain. Science 1947;106:349-50. 2. Gildenberg PL. Stereotactic surgery: present and past. In: Heilbrun MP, editor. Stereotactic neurosurgery. Baltimore, MD: Williams & Wilkins; 1988. p. 1-15. 3. Dandy WE. Ventriculography following the injection of air into the cerebral ventricles. Ann Surg 1918;70:378-84. 4. Harris MI, Bergenheim AT. A comparative study on ventricular graphic and computerized tomography-guided determination of brain targets and functional stereotaxis. J Neurosurg 1990;73:565-71. 5. Cooper IS. Ligation of the anterior choroidal artery for involuntary movements of parkinsonism. Psychiatr Q 1953;27:317-19. 6. Meyers R. Surgical experiments in the therapy of certain ‘extrapyramidal diseases’. Acta Psychiat Neurol 1951;26:1-42.
29
7. Apuzzo MLJ. A fantastic voyage: a personal perspective on involvement in the development of modern stereotactic and functional neurosurgery (1974–2004). Neurosurgery 2005;56(5):1115-33. 8. Wells TH, Todd EM. The Todd-Wells apparatus. In: Gildenberg PL, Tasker RR, editors. Textbook of stereotactic and functional surgery. New York: McGraw-Hill; 1998. p. 95-9. 9. Rand RW. A stereotactic instrument for pallidothalamectomy. J Neurosurg 1961;18:258-60. 10. Schaltenbrand G. Personal observations on the development of stereotaxy. Conf Neurol 1975;37:410-16. 11. Freidman WA, Coffey RJ. Stereotactic surgical instrumentation. In: Heilbrun MP, editor. Stereotactic neurosurgery. Baltimore, MD: Williams & Wilkins; 1988. p. 55-72. 12. Kandel EI. Stereotaxic apparatus and operations in Russia in the 19th century. J Neurosurg 1972;37:407-11. 13. Roberts TS. The brw/crw stereotactic apparatus. In: Gildenberg PL, Tasker RR, editors. Textbook of stereotactic and functional surgery. New York: McGraw-Hill, 1998. p. 65-71. 14. Brown RA, Roberts TS, Osborn AG. Stereotaxic frame and computer software for CT-directed neurosurgical localization. Ivest Radiol 1980;15:308-12. 15. Brown RA. US patent #4608977, issued 1986. 16. Leksell LA. Stereotactic apparatus for intracerebral surgery. Acta Chir Scand 1949;99:229-33. 17. Van Buren JM. A stereotaxic instrument for man. Electroencephalogr Clin Neurophysiol 1965;19:398-403. 18. Gouda K, Gibson RM. New frame for stereotaxic surgery: technical note. J Neurosurg 1980;3:256-9. 19. Cosman ER. Development and technical features of the Cosman-Roberts-Wells stereotactic system. In: Pell MF, Thomas DGT, editors. Handbook of stereotaxy using the CRW apparatus. Baltimore, MD: Williams & Wilkins; 1994. p. 1-52. 20. Ondo W, the DBS Study Group. The North American survey of placement and adjustment strategies for deep brain stimulation. Stereotact Funct Neurosurg 2005;83:142-7.
467
26 Electronic Stereotactic Atlases J. Yelnik . E. Bardinet . D. Dormont
Localizing structures and functions in the brain is a quest which has concerned human being since its early history. Although it is known that trepanations have been performed in primitive societies such as in the Neolithic period (7000 year BC), in Egypt (3000 years BC) or in Peru (2000 year BC), these practices were motivated by religious reasons rather than by medical reasons, as far as we know. The first rational approaches of brain anatomy and physiology were those of Hippocrates (400 years BC) although the knowledge of brain anatomy was still rudimentary at this period. Surgical interventions in the brain have remained very empirical during the middle-ages with the notable exception of Ambroise Pare´, who was able to treat fractures of the vertebral column and to make trepanations of the skull. Neurosurgery in fact began with Victor Horsley (1857–1924), while in an overlapping period Claude Bernard (1813–1878) developed his fundamental approach of experimental physiology, which opened an access to an understanding of brain functions. Localization of brain functions had for a while lost its way in the concepts of ‘‘Phre´nologie’’ or ‘‘organology’’ of Franz Joseph Gall (1758–1828) and Johann Spurzheim (1776–1832) who proposed that the most complex human functions such as compassion, moral sense, vanity, feeling of property, kindness, and benevolence could be localized in specific parts of the brain, and even identified as a prominent development of the external shape of the skull in individual subjects with specific development of such functions. Brilliant neurologists, such as Paul Broca (1824–1880), Jean-Martin Charcot (1825–1893), John Hughling Jackson (1835–1911), and Joseph Babinski
#
Springer-Verlag Berlin/Heidelberg 2009
(1857–1932) have finally provided the bases of a really scientific exploration of the localization of brain functions. This scientific approach has developed rapidly in the domain of neurology based on the anatomo-clinical method which consists of a systematic comparison between the neurological symptoms observed in a living patient and the anatomic lesions discovered in this brain after death. In the domain of neurosurgery, brain localization is a central issue. The technique by which a given region is localized in the brain of a living patient is called stereotaxy, which comes form the Greek stereo for three-dimensional and taxis for disposition, hence ‘‘localize in space.’’ The first stereotactic operations were performed on the basis of bony landmarks as proposed with the system of Horsley-Clarke [1]. This bony system is still in use for all experimental studies in the rat with the most commonly used atlas of Paxinos [2]. In the human, the bony landmarks have been shown in the 1950s by Spiegel and Wycis to be insufficiently accurate for human stereotactic surgery. They developed a method based on the Horsley-Clarke system [3], which relied on their own atlas of the human brain (> Figure 26-1) and on intracerebral landmarks visible on ventriculography. This pioneer neuro-imaging modality, introduced in 1918 by the American neurosurgeon Walter Dandy [4] and still in use in some neurosurgical centers, consists in taking X-ray images of the ventricular system after injection of filtered air directly into one or both lateral ventricles of the brain via one or more small trephine holes drilled in the skull under local anesthesia. The landmarks they selected were
374
26
Electronic stereotactic atlases
. Figure 26-1 A frontal unstained section of the first stereotactic atlas of the human brain, that of Spiegel and Wycis (Reference 5). The three-dimensional localization of cortical areas or deep brain structures is given by the medio-lateral and dorso-ventral millimetric grids and by the position of the section in the series
the epiphysis and the foramen of Monro [5]. Afterwards Jean Talairach at the Sainte-Anne Hospital in Paris, France, developed a system based on two different and more reliable ventricular landmarks, the anterior and posterior commissural points (AC-PC) [6–9], a coordinate system which has become the gold standard in the field of human stereotactic neurosurgery, even when ventriculography was later replaced by magnetic resonance imaging (MRI). The ACPC system of coordinates has been widely used since the mid-twentieth century for neurosurgical interventions for movement or psychiatric disorders [10–13].
In the 1980s, the development of MRI completely transformed the concept of in vivo brain structure localization. MRI uses magnetic fields and radio frequencies to produce high quality two- or three-dimensional images of brain structures. When the main magnetic field is imposed, each point in space has a unique radio frequency at which the signal is received and transmitted. Sensors read the frequencies, which are then used to build an image. With MRI, it has become possible to image both surface and deep brain structures with a high degree of detail. This technical progress has completely transformed the concept of brain anatomy, targeting, or localization, and consequently, the notion of brain atlases itself. Another important development in the domain of neurosurgery had a major influence on the issue of brain localization when AlimLouis Benabid in Grenoble, France, observed in 1987 that brain lesioning (e.g., thalamotomy) could be replaced by chronic high frequency electrical stimulation of the same region for the treatment of tremor [14]. Chronic stimulation had already been used for treating pain in the early 1970s [15], but Benabid first proposed to combine the implantable pacemaker technology (developed for cardiac pacemaker by 1960) with chronically implanted deep brain electrodes [16]. The advantages of stimulation over lesion were that the method did not produce a definitive lesion, the electrode could be removed or displaced if adverse events occurred, the electrical parameters could be adjusted to obtain the best possible clinical result. In stereotactic functional deep brain stimulation (DBS) neurosurgery, three successive stages that are linked to each other work toward a precise identification of deep brain structures. The preoperative step is that of targeting, which consists of identifying a given deep brain structure, previously chosen on the basis of theoretical and physiopathological arguments, in the brain of a given patient. The peroperative step is that of exploration during which the particular characteristics of the target in the concerned patient
Electronic stereotactic atlases
are investigated using electrophysiological and clinical testing. The postoperative step is that of localization which determines the exact position of each implanted electrode within the deep brain structures of the operated patient. The first DBS target was the thalamic nucleus Vim for tremor, either essential tremor or parkinsonian tremor [17]. It was proposed by Alim-Louis Benabid following the peroperative observation that stimulation of the Vim nucleus at high frequency (130 Hz) resulted in the immediate suppression of tremor, thus mimicking the classical thalamotomy of the same nucleus in a reversible way. He then had the great idea of a permanent stimulation of the Vim through a stimulator implanted in the subclavicular region. Then, on the basis of previous results of pallidotomy for movement disorders [10], the internal globus pallidus was proposed for the treatment of the other parkinsonian symptoms, rigidity, akinesia, and Ldopa-induced dyskinesias [18]. Finally, following experimental researches in non human primates [19,20], the subthalamic nucleus (STN) came forward as the best target for the treatment of the whole symptomatology of Parkinson’s disease [21]. These successive targets rapidly raised problems of localization. The Vim thalamic target was first identified on the basis of the AC-PC system of coordinates, the underlying assumption being that all brains, at least the deep brain structures, can be considered similar after spatial normalization (i.e., deformation) based on the proportional Talairach system or other similar consideration. The pallidal target raised particular problem due to its lateral localization, far from the midline. Indeed, the AC-PC distance did not provide any information about a precise mediolateral localization of the target. Finally, the STN raised difficult problems because of its small size (12 5 3 mm3) [22], its complex oblique orientation, and the numerous axonal bundles by which it is surrounded. At the present time, DBS has greatly developed. The number of patients operated has
26
increased dramatically over the past 20 years, with DBS centers being now present in a huge number of different countries. The large series of patients operated have provided an invaluable source of anatomo-clinical data from which the relationship between electrode localization and the clinical outcome could be studied in great details. The varying effects that were obtained, which depend on both the precise localization of the electrodes and the electrical settings that were applied, have been studied thoroughly, which has increased at the same time the fine knowledge of DBS mechanisms and the need for more refined electrode localization. Today, electrodes are implanted not only within a given target (a nucleus as the STN or the globus pallidus) but within a given anatomofunctional subdivision of the nucleus. In addition, the volume of tissue activated (VTA) can now be analyzed by taking into account the electrical parameters that are applied and the fine anatomical structure of the region stimulated [23]. Neuro-imaging during the DBS procedure is required at two stages: preoperatively to determine the position of the target in a given patient, which is done using ventriculography and/or MRI, and postoperatively to determine the precise localization of the electrodes and their four contacts in the patient’s brain, which is done using MRI or CT scan which has been proposed as an alternative following a 2002 FDA alert about DBS and postoperative MRI. This latter possibility has become necessary in the particular context of DBS studies, due to accidents that occurred for patients submitted to MRI acquisition with the electrodes and the stimulator activated. In addition, peroperative local electrophysiological exploration (in the neighborhood of the planned target) and estimation of the VTA [23] provide a better volumetric vision of the neighborhood of the targets, including gray matter nuclei and white matter bundles. These different data contribute to improvement of current knowledge of the mechanisms of DBS and of anatomo-clinical correlations.
375
376
26
Electronic stereotactic atlases
At the same time, these data raise a problem of scale. Indeed, the resolution of MRI or scanner are roughly of the same magnitude (about 1 mm) but are all insufficient with regard to the required definition of the DBS targets whose definition becomes progressively more and more accurate and restricted to subportions of deep brain nuclei. Peroperative micro-electrode recordings are close to this level of resolution, while the accuracy of definitive stimulating macro-electrodes is lower. Taking into consideration these various scales of resolution is a challenging issue for any system intending to localize precise targets for stereotactic functional neurosurgery, and electronic stereotactic atlases are in a central position in this process.
method stains both the cell bodies (pink to violet) and the myelin sheaths (blue to green). Immunohistochemical methods stain specific compounds of the membrane or cytoplasm, as for example, the enzyme tyrosine hydroxylase in dopaminergic neurons, the calcium-binding proteins parvalbumin, calretinin, and calbindin D28K particularly present in the basal ganglia. Finally, histology is the best method because it reveals the real cellular structure of the brain. However, it cannot be applied to living brains and therefore require other methods.
Histology: An Accurate Method to Identify Deep Brain Structures
The resolution and contrast provided by a MRI machine depends largely on the strength of its magnetic field. In this domain, clinical practice and advanced MRI research must be distinguished. The most common MRI machines encountered in neuroradiology departments are 1.5 T MRI machines. Today, 3 T machines have begun to replace 1.5 T machines in most research imaging centers, and they will progressively be installed almost everywhere for use in clinical practice. Some advanced research imaging centers are already equipped with 7 T machines. MR images are in fact created by using different sequences that provide different contrasts. T1 and T2 being relaxation times of proton spins after radio-frequency excitation, different combinations of T1 and T2 weights can be applied to obtain different contrasts. The T1-weighted MRI is commonly used to reveal the 3D anatomy of the entire brain. At 1.5 T, it clearly reveals the caudate nucleus and putamen. Optimized T2-weighted sequences at 1.5 T have been proposed that reveal the STN and substantia nigra as zones of hyposignal, whereas they are invisible in T1-weigthed sequences. This property is used to target the STN at the preoperative step of neurosurgery for Parkinson’s disease [37]. Other
Histology is the most accurate method to identify deep brain structures because it makes it possible to identify the regions in which cell bodies of neurons group together to form nuclei. This is the cytoarchitectonic method, that has been first used for establishing a parcellation of the cerebral cortex [24] and then of thalamic nuclei [25–27] or basal ganglia structures [28]. Histology can also reveal axon fascicles by staining the myelin sheaths, which has been used to construct brain atlases [29,30]. In addition, functional subdivisions can be revealed by using immunohistochemical methods such as the revelation of calcium-binding proteins [31–36]. Histology is applied to thin slices of brain tissue obtained from a dead specimen and most often submitted to a formalin fixation. There are a vast number of methods that can reveal different components of the nervous tissue. The Nissl method, with cresyl violet, thionin, or toluidin blue stains the acid components, DNA and RNA, and therefore of the cell bodies in which they are localized. The Weigert or Weil methods stain the myelin sheaths hence the axon bundles, while the Kluver–Barrera
MRI: Can Deep Brain Structures be Identified in MRI?
Electronic stereotactic atlases
T2-weighted sequences have been proposed to target the globus pallidus [38]. Deep brain structures can be identified on such brain images using a number of different methods proposed by the medical image processing community, mostly on 1.5 T MR images [39–45]. These methods are referred to as direct methods as they allow one to identify structures directly in the patient’s brain MRI. On the contrary, indirect methods, referred to as atlas-based methods, require an atlas previously constructed and its adaptation to patient’s MRI. Obviously, direct methods are intrinsically limited as only the structures that are visible or partially visible in these clinical practice MR images (e.g., lateral ventricles, caudate nucleus, putamen, thalamus, optic tract) can be actually identified with some certainty. Diffusion tensor imaging (DTI) has recently appeared as a highly promising imaging technique. Based upon the diffusion properties of water molecules submitted to a magnetic field, it measures the anisotropy coefficient that depends upon the more or less privileged direction of the nervous tissue. In grey matter, nervous tissue has no privileged direction whereas in white matter, axon bundles impose a strong orientation. It is therefore possible to detect this direction as a fraction of anisotropy (FA) and to reconstruct the entire trajectory of individual axon fascicles by determining a privileged direction common to adjacent voxels in a DTI acquisition. This property has been used to identify subnuclei in the thalamus [46]. Whatever the sequence used, MRI has two limitations. First, only some deep brain structures are visible on MRI. Many brain structures remain invisible. This is particularly obvious for 1.5 T acquisitions. Development of more powerful magnetic fields provides more MR signal and allows increasing the resolution. 3 T machines, which represent the future of clinical practice, clearly provide better anatomical (T1-weighted) images, but differentiation of deep brain structures such as external and internal parts of the globus pallidus or thalamic nuclei still remains difficult. 7 T
26
research machines provide really fascinating images but many technological as well as safety issues need to be resolved before 7 T machines can be used in clinical practice. A second limitation of MRI is that the exact relationship between the nature of the MRI signal and the real nature of the cytological architecture of the nervous tissue has never been demonstrated. It must be underlined that a 1 cubic millimeter of MRI is characterized by a single parameter, namely the grey level corresponding to a combination of relaxation times of all the protons present in this cubic volume. In the real nervous tissue, the same cubic millimeter consists of hundreds of neurons with their cell bodies, local axonal arborizations and dendritic arborizations, of glial cells and of afferent axonal arborizations, which are all inter-connected in highly specific networks of synaptic contacts (> Figure 26-2). Although it has not yet fully demonstrated that MRI contrasts actually reflect the exact histological structure of the nervous tissue, it is widely accepted in the neurological and neurosurgical communities that MRI provides a reliable image of the brain of individual living subjects or patients. Researches are in progress to better understand this MRI/histology relationships in particular with the T2-weighted sequences [47,48]. Nowadays, MRI provides an irreplaceable tool to visualize the anatomy of the deep brain structures of individual patients, but MRI has nevertheless intrinsic limitations in terms of resolution. This is the main reason for which brain atlases are still being developed, together with methods to adapt the atlas to the particular configuration of the brain of a given patient. Different types of brain atlases have been developed that are presented here, and different adaptation (or deformation) methods will be described and discussed later.
2D Printed Atlases The first published atlas was that of Spiegel and Wycis [5], which was to be used in coordination
377
378
26
Electronic stereotactic atlases
. Figure 26-2 Upper left box: one cubic millimeter of the globus pallidus consists of hundreds of neurons with their cells body (here revealed in blue with the Nissl method) and dendritic arborization (redrawn from Yelnik J, Percheron G, Franc¸ois C, A Golgi analysis of the primate globus pallidus. II. Quantitative morphology and spatial orientation of dendritic arborizations, J Comp Neurol., 227:200-213, 1984). Lower left box: the localization of this one cubic millimeter is shown in a frontal section of a MRI acquisition. Lower right box: the region of the globus pallidus is enlarged to show the one millimeter resolution of the MRI. Upper right box: the same one cubic millimeter of pallidal tissue is represented by a uniform grey level in the MRI
with the stereotactic apparatus they developed from that of Horsley and Clarke [3]. The atlas consisted of photographs of regularly-spaced sections in the sagittal, frontal, and horizontal planes. The slices were unstained and allowed identification of only gross subdivisions of grey and white matter (> Figure 26-1)Their stereotactic system was based on two intracerebral landmarks: the calcified pineal gland and the foramen of Monro by which the third and the lateral ventricles communicate. This atlas was soon followed by the atlas of Talairach [7] who first proposed the AC-PC system as the most reliable ventricular landmarks coordinate system. The first version of the atlas of Schaltenbrand appeared in 1959 [49]. It was the first atlas with
three series of histological sections with myelin sheath staining in the frontal, sagittal, and horizontal planes. The atlas provided photographs of the sections and transparent pages with a delineation of the nuclei and fiber fascicles, which made it possible to compare histology and tracings. This atlas has rapidly become a gold standard in the field of stereotactic neurosurgery, although its three-dimensional coherence was very approximate [50–53]. A second version of the same atlas was published later [29]. The atlases of Guiot of the globus pallidus and thalamus in 1961 [54], Van Buren of the basal ganglia in 1962 [55], Andrew and Watkins of the thalamus in 1969 [56], Afshar of the brain stem and cerebellum in 1978 [57], and Hassler for surgery of Parkinson’s disease in
Electronic stereotactic atlases
1979 [58] followed. More recently (in 1997), an atlas based on myelin sheath staining of the human brain has been published by Paxinos and coworkers [30]. In these atlases, the number of sections and the interval between sections varied greatly (> Table 26-1), which makes the anatomical accuracy and the sampling of the 3D structure of the brain highly variable from one atlas to another.
3D Electronic Atlases The major limitation of printed atlases is that they consist of 2D information, whereas the brain is a 3D structure. Four solutions have been proposed to overcome this problem.
Construction of 3D Atlases from 2D Printed Histological Atlases In 1987, Yoshida at Kurume, Japan, was probably the first who proposed the creation of a 3D atlas [59] by interpolation of the Schaltenbrand and Bailey’s atlas [49]. In 1997, Nowinski started to work on an ideal digital stereotactic atlas, which fuses the Talairach and Schaltenbrand atlases into a common navigation software tool (> Figure 26-3) . Table 26-1 Inter section interval in different histological atlases Existing printed atlases
Inter section interval
Spiegel and Wycis [5] Talairach et al. [7] Schaltenbrand and Bailey [49] Guiot et al. [54] Van Buren and MacCubbin [55] Andrew and Watkins [56] Schaltenbrand and Wahren [29] Afshar et al. [57] Hassler et al. [58] Talairach and Tournoux [9] Mai et al. [30] Morel et al. [31]
5 mm 3–6 mm 1–4 mm 0.5–1 mm 5 mm 1 mm 1–4 mm 1 mm 2 mm 4 mm 0.7–2.5 mm 0.9–1.8 mm
26
[60,61]. In 1998 the group of Collins and Peters at the Montreal Neurological Institute (MNI) of McGill University in Montreal, Canada [62], constructed an atlas from the Schaltenbrand atlas [29] by first aligning the successive sections on thebasis ofthegridpresentin theatlas,thencreating a 3D volume and submitting it to an interpolation operation using spline parameterization. The group of Ganser in Heidelberg, Germany, in 2003, constructeda3DdigitalversionoftheTalairachatlas [9]. They scanned the series of coronal sections, segmented manually the different structures, interpolated additional cross-sections, and created shell surfaces of each structure. The problem with such printed atlases is that their 3D accuracy and coherency is primarily dependant on the quality of the 2D individual sections of the original atlas but above all on the carefulness with which successive sections have been aligned one onto the other. Unfortunately, the two most widely used atlases, those of Schaltenbrand and Talairach, do not fulfill these requirements. The atlas of Schaltenbrand is well-known for its 3D inconsistency [50,51], and the atlas of Talairach is based on photographed sagittal sections of the brain of a 60year-old woman from which coronal and axial sections were obtained by manual interpolation.
Construction of 3D Atlases from MRI In order to build fully 3D brain atlases, several groups have proposed MRI-based atlases. These types of atlases are useful because they allow easy three-dimensional navigation in the brain. Also, as MRI is nowadays the reference in vivo brain imaging, it is very useful to be able to study correspondence between the MR signal and subcortical structure delineation. Started in 1990, the project of the Surgical Planning Laboratory (MGH, Harvard Medical School, Mass., USA) aimed at proposing a detailed
379
380
26
Electronic stereotactic atlases
. Figure 26-3 The electronic atlas developed by Wieslaw Nowinski (References 60,61) consists of a digitized version of the atlas of Schaltenbrand and Wahren (Reference 29). a shows a frontal section the original atlas, b shows the electronic contours in both hemispheres, c shows color-coded images and d shows a 3D view of the thalamic nuclei
morphological brain atlas built from a T1-weighted (i.e., anatomical) MRI. The first goal of this project was to develop a tool for education. It is today referred to as the SPL anatomy browser and it is freely available on the Internet (www.spl. harvard.edu). It consists of a browser that permits navigation in a T1-weighted MRI volume in which numerous cortical and subcortical structures are included (> Figure 26-4). This atlas has also been used for presurgical planning and segmentation tasks. In 2002, the Neurofonctional Imaging Group (GIN, UMR6095, CYCERON, Caen, France)
proposed a macroscopic anatomical parcellation of the MNI MRI single subject brain used by the functional brain imaging community. This atlas, referred to as the AAL (Anatomical Automatic Labelling), has been included in the Statistical Parametric Mapping (SPM, Wellcome Department of Imaging Neuroscience, UCL, London, UK) software as a toolbox. Although this atlas is devoted to fMRI studies and cortical mapping, it includes some subcortical structures (caudate nucleus, putamen, pallidum, thalamus). More generally, many medical image processing groups have developed homemade
Electronic stereotactic atlases
26
. Figure 26-4 The Surgical Planning Laboratory (SPL) browser developed at the Harvard Medical School (www.spl.harvard.edu) consists of a T1-weighted MRI volume with cortical and subcortical structures digitized and labeled
MRI-based 3D brain atlases that they have included in brain segmentation methods. These methods can be grouped under the naming ‘‘pseudo-direct methods,’’ as they all work directly in the living patient’s brain MRI, but use a priori information provided by these homemade atlases. Among these methods, Pitiot et al. proposed in 2004 an expert knowledge-guided system that allowed to robustly identify the corpus callosum, lateral ventricles, hippocampus, and caudate nucleus on in vivo MRI [39]. In 2005, Zhou et al. proposed a feature-based method using fuzzy templates built from a training set, which allowed to segment five subcortical structures, thalamus, putamen, caudate, hippocampus, and amygdala [63]. Fischl et al. in 2002
have presented a very nice and powerful whole brain segmentation algorithm [43]. The method is defined in a probabilistic framework and assigns 1 of 37 labels to each voxel of the brain volume. Subcortical structures include the caudate, putamen, pallidum, thalamus, hippocampus, and amygdala (> Figure 26-5). This algorithm is available as it has been included in the FreeSurfer software (MGH, Harvard, Mass., USA). In a radiotherapy context, Bondiau et al. [64] have also designed a whole brain atlas that comprises organs at risk, including thalamus, caudate, putamen, and pallidum, and is embedded in a segmentation pipeline [64]. In 2005, Lemaire and coworkers of Clermont Ferrand, France, proposed another solution based
381
382
26
Electronic stereotactic atlases
. Figure 26-5 An automated algorithm for the probabilistic segmentation of a MRI volume has been developed by Fischer and coworkers (reference 43, http://surfer.nmr.mgh.harvard.edu/). Each voxel is assigned to one of the 37 structures available
on a MRI obtained from a post mortem specimen submitted to both a prolonged high resolution 4.7 T acquisition and a standard 1.5 T acquisition. The MRIs were studied by comparison with available printed histological atlases and imaging criteria to identify deep brain structures directly in the 1.5 T MRIs of individual patients were described [65]. However, in the context of DBS clinical practice, MRI is limited in terms of spatial resolution and contrast, which makes these approaches less accurate. As mentioned earlier, histology thus remains the best tool to reveal the anatomy of the basal ganglia, but histology can provide only bidimensional images of the brain. The challenge is therefore to obtain a ‘‘three-dimensional histology’’ by any possible tool.
Construction of 3D Atlases from Histology The elaboration of a 3D atlas from 2D histological sections requires first a good histology (choice of staining techniques, careful processing of individual sections, and reliable tracing by experts in brain anatomy), and second a method that would allow to align the whole series of individual sections into an anatomically
consistent 3D block. Indeed, sectioning of an anatomical specimen for histology provides a series of disconnected 2D slices, whose original 3D shape is lost. Several methods have been proposed to solve this problem, one of which being the use of an accurate system of landmarks, another one being coregistration of successive 2D slices [66]. Histological 3D atlases have been built by three different teams. An atlas of the human thalamus was constructed in 1997 in Zurich, Switzerland, on the basis of cyto- and myeloarchitectonic criteria and on the use of calcium-binding proteins (parvalbumin, calretinin, calbindin D-28K) as functional markers (> Figure 26-6). [25]. Three series of successive sections were traced (sagittal, horizontal, and frontal) from three anatomical specimens and were subsequently normalized with each other (frontal and horizontal to sagittal) to obtain an internally consistent Canonical model of the atlas [67]. Two atlases of the basal ganglia and thalamus have been published successively; one by the McGill Hospital in Montreal, Canada in 2006 [68], and the other one by the authors of this review in the Salpeˆtrie`re Hospital, Paris, France in 2007 [28]. The Canadian atlas was developed by Louis Collins as a continuation of his previous work on the Schaltenbrand atlas [62]. It consisted
Electronic stereotactic atlases
of sections stained alternatively with Luxol Blue for myelin and with a Nissl stain for cell bodies (n = 86 pairs of sections) and were aligned through a slice-to-slice nonlinear registration, . Figure 26-6 A 3D histological atlas of the thalamus constructed by Morel and co-workers (reference 25) and transformed into a digitized 3D stereotactic atlas (Reference 67) is used to define a trajectory that pierces the central lateral nucleus of the thalamus (indicated by the arrow)
26
which was optimized by minimizing the mean distance between the segmented contours in adjacent pairs of slices. Then, 3D geometric objects were created by tessellation to represent different anatomic regions (> Figure 26-7). An additional feature of the French atlas was a MRI acquisition (T1- and T2-weigthed sequences) performed previous to brain extraction, which provided a reliable anatomical reference for both the construction of the atlas from stained histological sections and the adaptation of the atlas to the MRIs of individual patients. Histological sections were stained alternatively with Nissl stain and the calcium binding protein calbindin D-28K (n = 80 pairs of sections), and were aligned by piecewise linear coregistration with the MRI and a cryo block constructed from photographs taken during cryosectioning. Contours of basal ganglia and thalamus were traced from histological sections
. Figure 26-7 The atlas developed by Louis Collins and co-workers (Reference 68) consists of anatomical regions digitized from histological data and that can be transferred to the MRI volumes of different patients
383
384
26
Electronic stereotactic atlases
and digitized. A specific procedure combining a multimodal optimization (MRI, cryoblock, nissl, calbindin) and a 3D optimization was implemented to assure an optimal 3D coherency [28]. Anatomically and geometrically consistent 3D
surfaces of each traced region were constructed by shape-based interpolation (> Figure 26-8). The specific qualities of the French atlas are a histological level of resolution, the inclusion of functional information based on calbindin
. Figure 26-8 The atlas developed by the authors of this review (Reference 28) consists of anatomical regions digitized from histological sections of a post-mortem specimen. Registration between atlas and patients is calculated from the MRI acquisitions of the same specimen obtained before histological sectioning. The structures transferred to a patient’s MRI are shown in 3D (upper line), in a T1- (middle line) and a T2-weighted acquisition (bottom line)
Electronic stereotactic atlases
staining, a large number of sections (160 sections with 0.35 mm interval), and MRI acquisitions of the same specimen that allowed the construction of truly continuous 3D surfaces. This differs from the Talairach atlas in which the contours of cerebral structures were traced from one specimen sectioned in the sagittal plane (section interval section 4 mm) and were extrapolated in the coronal and axial planes by point-to-point projection (interval section 5 mm). In the Schaltenbrand and Wahren atlas, three series of sections are available (sagittal, coronal, and axial) but the number of sections is low (18, 20, 20), the section interval is high and variable (1–4 mm), and the 3D coherency is very low. The atlas of Collins and coworkers was derived from a set of serial histological data (0.7 mm section interval, 86 pairs of slices). It closely resembles the French atlas except that it lacks a MRI of the brain specimen. To overcome this problem, a pseudoMRI was created from the reconstructed voxellabeled atlas volume and used for adapting the atlas to patients through nonrigid registration.
Functional Population-based Atlases Another strategy consists in building an atlas from functional data collected in a population of subjects or patients. The data can be peroperative electrophysiological recordings or clinical exploration data (e.g., points that provoke arm dyskinesias or somesthetic perceptions on the hand) or postoperative tuning data (e.g., a contact that provokes heat sensation or diplopia). Once these data have been collected, they are placed in a common reference space that defines the atlas. Terry Peters and colleagues, who originally worked at the MNI and participated to the histology-based atlas [62], moved to London, Ontario, and started to develop a functional population-based strategy [69,70]. Electrophysiological and clinical peroperative data obtained from a
26
series of 88 patients operated for DBS at the London Health Science Centre were used to define this atlas which is, in turn, integrated into a computeraided system for DBS targeting (> Figure 26-9). A somewhat similar system has been developed by the group of Benoit Dawant at Vanderbilt University, Nashville, Tennessee. Pierre-Franc¸ois D’Haese has developed during his PhD thesis an atlas-based method for the automatic determination of DBS targets [71,72]. Their atlas is based on micro electrode recordings, stimulation parameters and final implant positions. In addition, a population-based electrophysiological map has been created by analyzing automatically the peroperative micro electrode recordings with signal processing techniques (wavelets-based de-noising, spike detection). Nowinski and Benabid have developed an atlas that they have called the PFA (probabilistic and functional atlas), which combines pre-, per-, and postoperative neuro-imaging data with peroperative electrophysiological data from 274 parkinsonian patients operated at the Joseph Fourier University School of Medicine in Grenoble, France [73–75]. As mentioned earlier, the strategy for building such population-based atlases is to place data coming from different patients in a common reference space that is to say to ‘‘normalize.’’ This is done by applying spatial normalization algorithms on the patients’ images (most often MR images), which allows one to compensate for inter-individual brain shape variability. Spatial normalization consists in computing a deformation between the patient’s image and a reference image chosen arbitrarily. More generally, computing a deformation between a reference and the patient is the necessary remaining step, whatever the atlas retained. Indeed, by definition an atlas is a template, thus a unique set of data which must be adapted, or deformed, to fit the particular brain geometry of each subject or patient. The crucial issue is to make the best possible choice of a reliable deformation procedure to adapt one of the available atlases to the
385
386
26
Electronic stereotactic atlases
. Figure 26-9 The atlas developed by Tony Peters and co-workers (Reference 70) consists of electrophysiological and clinical preoperative data obtained in a series of 88 patients, here indicated as small spheres localized within a 3D MRI volume
individual brain anatomy of each patient, and to validate this choice.
From Atlases to the Brain of Living Patients An atlas can consist of series of anatomical 2D maps, 3D regions segmented in a reference MRI, 3D surfaces or functional data localized in a reference space. The brain geometry of subjects or patients is generally available in the form of a 3D MRI or a scanner. Deformation of an atlas can be performed using different strategies but all strategies are not appropriate for all types of atlases. There is also a ‘‘cultural’’ aspect in choosing among the variety of deformation strategies available nowadays. Indeed, adapting a brain atlas to the brain of a patient in the DBS context appears to be at the crossroad of two, rather different scientific communities: on the one side, the neurosurgical and neuroanatomical communities who traditionally have used landmark-based deformation methods, and on the other side the medical image processing community who is particularly expert in the development of automatic deformation, or registration algorithms.
Visual Deformation This is the most ancient way of using the information contained in an atlas. The user has a printed version of the atlas in one hand, generally a series of 2D anatomical plates with labeled structures, and a series of anatomical sections of the brain subject to be labeled in the form of MRI sections (or histological sections for a dead specimen) in the other hand. He must decide visually what atlas section best corresponds to a given subject section and then places mentally or using a tracing paper the atlas section on the subject section. Then he can attribute to a given point of the subject section an anatomical label coming from the atlas section. Such a procedure is in fact an atlas adaptation, but a totally manual one with mental/visual identification of the required landmarks. A somewhat original alternative relies on the direct analysis of MRI acquisitions. Deep brain structures are directly identified in the MRI acquisition of the subject by comparison with a previous study of a dead brain specimen MRI [41,65]. This procedure requires a skilled expertise of the user and above all, raises the question of the actual anatomic significance and reliability of the MRI signal.
Electronic stereotactic atlases
Manual or Semiautomatic Linear Deformations The most widely used system in the DBS community to adapt a brain atlas to the individual anatomy of a living subject is the proportional system of Talairach [9]. It relies primarily on the AC-PC distance, i.e., the length between the anterior and posterior commissural points, well-identifiable on a ventriculography or a mid-sagittal section of a MRI acquisition. As the Talairach atlas has been constructed in reference to the AC-PC line, the user has just to measure the AC-PC distance in the living brain and to adapt the antero-posterior length of the atlas to that of the brain. In addition, the proportional system comprises an adaptation along the medio-lateral and infero-superior dimensions, which is based on the overall size of the brain, including the entire cerebral cortex, along these two axes. The three adaptations are independent of one another, which takes into account the size and the particular shape of each studied brain (e.g., dolichocephalic or brachycephalic brains). A grid system is proposed that makes the practical application of the proportional system easily applicable even without any computer. The system has been largely applied all over the world as for example in the context of lesional surgery for the determination of the targets for thalamotomy or pallidotomy [54]. The proportional system of Talairach is therefore a reliable system, although it is inhomogeneous since the adaptation along the antero-posterior dimension is based upon two deep brain ventricular landmarks, whereas adaptation along the medio-lateral and inferosuperior dimensions depends on the overall size of the cerebral cortex. This is due to the fact that with ventriculography, internal landmarks are less clear along these dimensions (the height of the thalamus and width of the third ventricle are the best possible landmarks). Another difficulty of the AC-PC system is the definition of the two commissural points, which can vary slightly but
26
significantly from one atlas to another and from one user to another. AC-PC length and angle can be defined from the center of each commissural point or from the most anterior and posterior points, i.e., minimizing the AC-PC length. In the Schaltenbrand [29] atlas, for example, the ACPC line of the horizontal series passes by the upper border of AC and the lower border of PC, whereas it passes through the center of the two points in the two other series (frontal and sagittal). This makes a 7 angle difference and a significantly different topological and metrical aspect of cerebral regions in the 2D atlas sections. Other linear semiautomated methods have been proposed. In our studies of the localization of the DBS electrodes in the GPi [76] and STN [77], we proposed a method of tri-linear deformation of the Schaltenbrand atlas [29] based on internal brain landmarks, namely the AC-PC landmarks along the antero-posterior dimension, the superior limit of the putamen and the individualization of the cerebral peduncles along the supero-inferior dimension, the lateral limit of the putamen, the anterior column of the fornix, and the mamillo-thalamic tract along the medio-lateral dimension [77]. These landmarks, also used by others [67], are more appropriate than the outer limits of the cerebral cortex which in fact has not strong relationships with the deep brain nuclei. In the functional atlas of Nowinski and Benabid [73,74], the landmarks are the ACPC distance, the height of the thalamus, and the width of the third ventricle and the deformation of atlas data also follows a tri-linear procedure. Besides the aforementioned atlas-based localization studies, it is worth noting that several groups have conducted localization studies that aimed at defining the optimal DBS target from a population of patients, but without any reference to an atlas. These studies included spatial normalization transformations, because for comparing the optimal contacts in a group of patients, it is necessary to place all the data available in a common or template space (just the same as for
387
388
26
Electronic stereotactic atlases
the construction of population-based atlases). These works can therefore be viewed as population-based studies. Lanotte et al. [78] studied a series of 14 consecutive patients and expressed mean positions and distances (e.g., position of the central point of the most effective contact) with respect to the midpoint AC-PC line. No spatial normalization was applied to the individual data before their transfer in the reference space. Saint-Cyr et al. [79] studied a series of 29 patients (54 postoperative macro-electrodes). Spatial normalization of electrode contact positions was done along the antero-posterior axis using the AC-PC length. In Hamel et al. [80], 25 patients were included, spatial coordinates expressed with respect to the mid-commissural point, and spatial normalization was performed along the antero-posterior axis.
Automatic Methods of Deformation Several groups of the medical image processing community have developed brain atlases, some of which are dedicated to DBS [62,71,81–83]. These groups have a high level of expertise in the development of automatic deformation (or registration) algorithms. Therefore, it was almost natural for them to apply these algorithms to the atlas-to-patient adaptation problem. An automatic registration algorithm is based on the comparison of features (grey-level values, points, lines, graphs, etc.) present in the two images to be registered. The algorithm is defined by three main characteristics: the similarity measure, the space of allowed deformations (the number of degrees-of-freedom (DoF) of the deformation, e.g., 6 DoF for a 3D rigid transform), and the optimization method that is used. Deformations can be very constrained (limited number of DoF), e.g., linear scaling (7 DoF) or not, like elastic, fluid, or even free-form deformations. These last types of deformations are
often referred to as morphing or warping transforms. The choice of the most adequate deformation type is important, as it directly influences the quality of the atlas-to-patient result. Kikinis, Dengler and coworkers have proposed to deform their MRI-based brain atlas (the core of the SPL anatomy browser) by an elastic registration procedure consisting of a warping of the atlas image onto the patient image [84]. Ganser et al. have developed a nonrigid registration method to deform their 3D digital version of the Talairach atlas, consisting of the automated establishment of point correspondences between atlas and patient (these points being defined on the skull and the ventricles) and then interpolation of the corresponding displacement vectors using radial basis functions (> Figure 26-10) [85]. Louis Collins, at the McGill University Montre´al, Canada, developed in 1994 the ANIMAL (Automated Nonlinear Image Matching and Anatomical Labelling) algorithm [86]. The algorithm is twofold: first a constrained affine transformation (translations, rotations, scales), followed by the calculation of a 3D nonlinear deformation field . Figure 26-10 The atlas developed by Ganser and coworkers (Reference 85) proposes a nonrigid registration method with which the atlas of Talairach (Reference 9) is deformed to fit the brain MRI geometry of a given patient
Electronic stereotactic atlases
in a piece-wise fashion, fitting cubical neighborhoods in sequence. This algorithm has been applied to register the patient and atlas volumes of St-Jean [62]. It has also been used by Peters and colleagues for deforming their populationbased atlas on patients [70]. The same task was solved by a nonrigid registration algorithm based on radial basis functions in Dawant et al. [71]. These methods have in common a characteristic which is to register the images of the entire brains. On the contrary, Bardinet et al. propose a method that, after a first constrained alignment computed on the whole head, allows registering only the region of interest centered on the atlas of the basal ganglia [83].
Validation of the Atlas-to-patient Deformation Image processing algorithms, whether they are segmentation or registration, must be validated, which is always a difficult task. People of the medical image processing community have developed validation procedures adapted to a given type of algorithm. Image registration, which is crucial in atlas-to-patient deformation, can be validated by geometrical criteria (see for example, Hellier et al. [87]). But the key issue in validation is to clearly identify the context, the application and the question which is addressed, and to adapt the validation procedure to the problem to be solved. In the context of functional neurosurgery, the application is DBS. We have described different types of atlases and different tools to adapt a given atlas to the brain of a patient. The deformed atlas is supposed to give a precise and detailed description of the anatomy of the patient’s brain. The question that must be answered is therefore how to insure that the deformed atlas gives a reliable and anatomically plausible representation of the individual anatomy of the patient? This obviously depends on the type of the atlas used. If 2D printed atlases are used to identify brain regions in the MRI of a patient, no validation can
26
be performed and the expert’s opinion is the only reference available. If 3D electronic atlases are used, quantitative validation becomes possible as the atlas can be superimposed on the patient’s brain MR image. As the deformed atlas provides a segmentation of the brain, validation can be performed separately for each segmented structure. Segmentation of brain structures that are at least partially visible on MRI (such as the caudate or putamen) can be validated by comparing manually outlined brain structures with the same structures automatically segmented by the deformed atlas. Overlap indices, e.g. the Dice similarity coefficient, can be computed. For structures that are not visible on MRI, given by histological or functional population-based atlases, validation has to be done differently. How to insure that histological contours have been mapped accurately to an MRI with a mean voxel size of 1 mm? One way is to validate the MRI-based registration method that has been proposed [71,81]. But the ideal validation would be based on data that are comparable in scale to the level of definition of the histology. In DBS, these data are available and consist in the peroperative micro-electrode recordings. Indeed, these recordings, consisting of stereotactic positions and anatomical labels noted by the electrophysiologist provide a local representation of the brain structural organization that can be confronted to the information given by a histological atlas. Chakravarty et al. [82] have used intraoperative recordings of the sensory thalamus (during thalamotomy for Parkinson’s disease) to locally validate the deformation of their atlas. Bardinet et al. [83] confront their atlas with reconstructed electrophysiological recordings of the STN area collected during STN DBS of Parkinson’s disease patients.
Conclusion The perspective in the domain of brain localization remains completely open, with the goal of filling
389
390
26
Electronic stereotactic atlases
in the gap that still exists between histology and imagery. (1) To date, histology remains the sole method that can reveal not only the precise anatomical structure of the nervous tissue from the cellular to the regional level, but also the functional aspect of the basal ganglia, for example, its sensorimotor, associative and limbic subdivisions that immunohistochemical methods can reveal. Progress in this domain should concern individual variations of deep brain nuclei size and shape and particularly at different ages of life and in different pathological conditions. (2) Imagery has become the irreplaceable method for visualizing the brain anatomy of a given living patient and particularly its characteristic three-dimensional architecture. Progress in this domain is brought into play before our eyes and concerns both the level of resolution of MRI, which will probably increase tremendously in the next few years, and the nature of imaging with the spectacular development of DTI-based tractography. (3) The decisive progress is in the hands of the image processing community, which should now develop more and more powerful techniques of image deformation to adapt a more and more sophisticated histological and immunohistological knowledge to individual imagery at the finest level of resolution. A sort of 3D histology could thus be obtained for individual living patients, from which a probabilistic population-based brain anatomy could emerge.
References 1. Horsley V, Clarke RH. The structure and function of the cerebellum examined by a new method. Brain 1908;31:54-124. 2. Paxinos G, Watson C. The rat brain in stereotaxic coordinates. 4th ed. London: Academic press; 1998. 3. Spiegel EA, Wycis HT, Marks M, et al. Stereotaxic apparatus for operations on the human brain. Science 1947;106:349-50. 4. Dandy WE. Ventriculography following the injection of air into the cerebral ventricles. Ann Surg 1918;68(1):5-11. 5. Spiegel EA, Wycis HT. Stereoencephalotomy (thalamotomy and related procedures). New York: Grune and Stratton; 1952.
6. Talairach J, de Ajuriaguerra J, Hecaen H. Etude topographique des images ventriculaires en fonction des noyaux gris centraux. Annales de M,decine 1950;51: 83-100. 7. Talairach J, David M, Tournoux P, Corredor H, Kvasina T. Atlas d’anatomie ste´re´otaxique. Paris: Masson; 1957. 8. Talairach J, Szikla G. Atlas d’anatomie ste´re´otaxique du te´lence´phale. Paris: Masson; 1967. 9. Talairach J, Tournoux P. Co-planar stereotaxic atlas of the human brain. New York: Thieme; 1998. 10. Hassler R, Riechert T, Mundinger F, Umbach W, Ganglberger JA. Physiological observations in stereotaxic operations in extrapyramidal motor disturbances. Brain 1960;83:337-50. 11. Hassler R, Dieckmann G. Stereotaxic treatment of compulsive and obsessive symptoms. Confin Neurol 1967;29 (2):153-8. 12. Hassler R, Mundinger F, Riechert T, Gillingham FJ, Donaldson IML. Target reliability of the stereotaxic method as compared with autopsie findings. In the Third Symposium on Parkinson’s Disease. Livingstone; 1969:206–10. 13. Hassler R, Dieckmann G. Stereotactic treatment of different kinds of spasmodic torticollis. Confin Neurol 1970;32:135-43. 14. Benabid AL, Pollak P, Louveau A, Henry S, De Rougemont J. Combined (thalamotomy and stimulation) stereotactic surgery of the VIM thalamic nucleus for bilateral Parkinson disease. Appl Neurophysiol 1987;50:344-6. 15. Hosobuchi Y, Adams JE, Rutkin B. Chronic thalamic stimulation for the control of facial anesthesia dolorosa. Arch Neurol 1973;29(3):158-61. 16. Perlmutter JS, Mink JW. Deep brain stimulation. Annu Rev Neurosci 2006;29:229-57. 17. Benabid AL, Pollak P, Gervason C, et al. Long-term suppression of tremor by chronic stimulation of the ventral intermediate thalamic nucleus. Lancet 1991;337 (8738):403-6. 18. Kumar R, Lozano AM, Montgomery E, Lang AE. Pallidotomy and deep brain stimulation of the pallidum and subthalamic nucleus in advanced Parkinson’s disease. Mov Disord 1998;13 Suppl 1:73-82. 19. Bergman H, Wichmann T, DeLong MR. Reversal of experimental parkinsonism by lesions of the subthalamic nucleus. Science 1990;249:1436-8. 20. Benazzouz A, Gross C. F.ger J, Boraud T, Bioulac B. Reversal of rigidity and improvement of motor performance by subthalamic high-frequency stimulation in MPTP-treated monkeys. Eur J Neurosci 1993;5:382-9. 21. Limousin P, Pollak P, Benazzouz A, et al. Effect of parkinsonian signs and symptoms of bilateral subthalamic nucleus stimulation. Lancet 1995;345(8942):91-5. 22. Yelnik J. Functional anatomy of the basal ganglia. Mov Disord 2002;17 Suppl 3:S15-S21.
Electronic stereotactic atlases
23. Butson CR, Cooper SE, Henderson JM, McIntyre CC. Patient-specific analysis of the volume of tissue activated during deep brain stimulation. NeuroImage 2007;34 (2):661-70. 24. Brodmann K. Vergleichende Lokalisationslehre der Grosshirnrinde: in ihren Principien dargestellt auf Grund des Zellenbaues. Leipzig: Barth; 1909. 25. Morel A, Magnin M, Jeanmonod D. Multiarchitectonic and stereotactic atlas of the human thalamus. J Comp Neurol 1997;387(4):588-630. 26. Jones EG, Burton H. Cytoarchitecture and somatic sensory connectivity of thalamic nuclei other than the ventrobasal complex in the cat. J Comp Neurol 1974;154:395-432. 27. Percheron G. The motor thalamus. J Neurosurg 1997;87 (6):981-2. 28. Yelnik J, Bardinet E, Dormont D, et al. A three-dimensional, histological and deformable atlas of the human basal ganglia. I. Atlas construction based on immunohistochemical and MRI data. NeuroImage 2007;34 (2):618-38. 29. Schaltenbrand G, Wahren W. Atlas for stereotaxy of the human brain. Stuttgart: Georg thieme Verlag, 1977. 30. Mai JK, Assheuer J, Paxinos G. Atlas of the human brain. San Diego: Academic Press; 1997. 31. Morel A, Loup F, Magnin M, Jeanmonod D. Neurochemical organization of the human basal ganglia: anatomofunctional territories defined by the distributions of calcium-binding proteins and SMI-32. J Comp Neurol 2002;443(1):86-103. 32. Franc¸ois C, Yelnik J, Percheron G, Tande´, D. Calbindin D-28k as a marker for the associative cortical territory of the striatum in macaque. Brain Res 1994;633:331-6. 33. Prensa L, Richard S, Parent A. Chemical anatomy of the human ventral striatum and adjacent basal forebrain structures. J Comp Neurol 2003;460(3):345-67. 34. Karachi C, Francois C, Parain K, et al. Threedimensional cartography of functional territories in the human striatopallidal complex by using calbindin immunoreactivity. J Comp Neurol 2002;450(2):122-34. 35. Augood SJ, Waldvogel HJ, Mnkle MC, Faull RLM, Emson PC. Localization of calcium-binding proteins and GABA transporter (GAT-1) messenger RNA in the human subthalamic nucleus. Neuroscience 1999;88 (2):521-34. 36. Hontanilla B, Parent A, Gim,nez-Amaya JM. Parvalbumin and calbindin D-28k in the entopeduncular nucleus, subthalamic nucleus, and substantia nigra of the rat as revealed by double-immunohistochemical methods. Synapse 1997;25(4):359-67. 37. Bejjani BP, Dormont D, Pidoux B, et al. Bilateral subthalamic stimulation for Parkinson’s disease by using three-dimensional stereotactic magnetic resonance imaging and electrophysiological guidance. J Neurosurg 2000;92(4):615-25.
26
38. Hariz MI, Krack P, Melvill R, et al. A quick and universal method for stereotactic visualization of the subthalamic nucleus before and after implantation of deep brain stimulation electrodes. Stereotact Funct Neurosurg 2003;80 (1–4):96-101. 39. Pitiot A, Delingette H, Thompson PM, Ayache N. Expert knowledge-guided segmentation system for brain MRI. NeuroImage 2004;23 Suppl 1:S85-S96. 40. Shen D, Herskovits EH, Davatzikos C. An adaptive-focus statistical shape model for segmentation and shape modeling of 3-D brain structures. IEEE Trans Med Imaging 2001;20(4):257-70. 41. Barra V, Boire JY. Automatic segmentation of subcortical brain structures in MR images using information fusion. IEEE Trans Med Imaging 2001;20(7):549-58. 42. Bueno G, Musse O, Heitz F, Armspach JP. Threedimensional segmentation of anatomical structures in MR images on large data bases. Magn Reson Imaging 2001;19(1):73-88. 43. Fischl B, Salat DH, Busa E, et al. Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain. Neuron 2002;33(3):341-55. 44. Gonzalez Ballester MA, Zisserman A, Brady M. Segmentation and measurement of brain structures in MRI including confidence bounds. Med Image Anal 2000;4 (3):189-200. 45. Mangin JF, Poupon F, Duchesnay E, et al. Brain morphometry using 3D moment invariants. Med Image Anal 2004;8(3):187-96. 46. Behrens TE, Johansen-Berg H, Woolrich MW, et al. Non-invasive mapping of connections between human thalamus and cortex using diffusion imaging. Nat Neurosci 2003;6(7):750-7. 47. Dormont D, Ricciardi KG, Tande D, et al. Is the subthalamic nucleus hypointense on T2-weighted images? A correlation study using MR imaging and stereotactic atlas data. AJNR Am J Neuroradiol 2004;25(9):1516-23. 48. Meadowcroft MD, Zhang S, Liu W, et al. Direct magnetic resonance imaging of histological tissue samples at 3.0 T. Magn Reson Med 2007;57(5):835-41. 49. Schaltenbrand G, Bailey P. Introduction to stereotaxis with an atlas of the human brain. Stuttgart: Georg Thieme Verlag; 1959. 50. Nowinski WL, Liu J, Arumugam T. Quantification and visualization of three-dimensional inconsistency of the globus pallidus internus in the Schaltenbrand-Wahren brain atlas. Stereotact Funct Neurosurg 2006;84(5– 6):236-42. 51. Nowinski WL, Liu J, Thirunavuukarasuu A. Quantification and visualization of the three-dimensional inconsistency of the subthalamic nucleus in the Schaltenbrand-Wahren brain atlas. Stereotact Funct Neurosurg 2006;84 (1):46-55. 52. Niemann K, van Nieuwenhofen I. One atlas – three anatomies: relationships of the Schaltenbrand and
391
392
26 53.
54.
55.
56.
57.
58. 59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
Electronic stereotactic atlases
Wahren microscopic data. Acta Neurochir (Wien) 1999;141(10):1025-38. Niemann K, Naujokat C, Pohl G, Wollner C, von Keyserlingk D. Verification of the Schaltenbrand and Wahren stereotactic atlas. Acta Neurochir (Wien) 1994;129(1–2):72-81. Guiot G, Brion S, Akerman M. Anatomie ste´re´otaxique du pallidum interne du thalamus et de la capsule interne. Etude des variations individuelles. Ann Chir 1961;15:557-86. Van Buren JM, MacCubbin DA. An outline atlas of the human basal ganglia with estimation of anatomical variants. J Neurosurg 1962;19:811-39. Andrew J, Watkins ES. A stereotaxic atlas of the human thalamus and adjacent structures. Baltimore: Williams and Wilkins; 1969. Afshar F, Watkins ES, Yap JC. Stereotaxic atlas of the human brain-stem and cerebellar nuclei. A variability study. New York: Raven Press; 1978. Hassler R, Mundinger F, Riechert T. Stereotaxis in Parkinson syndrome. Berlin: Springer; 1979. Yoshida M. Creation of a three-dimensional atlas by interpolation from Schaltenbrand-Bailey’s atlas. Appl Neurophysiol 1987;50(1–6):45-8. Nowinski WL. Towards construction of an ideal stereotactic brain atlas. Acta Neurochir (Wien) 2008;150 (1):1-13. Nowinski WL, Fang A, Nguyen BT, et al. Multiple brain atlas database and atlas-based neuroimaging system. Comput Aided Surg 1997;2(1):42-66. St-Jean P, Sadikot AF, Collins L, et al. Automated atlas integration and interactive three-dimensional visualization tools for planning and guidance in functional neurosurgery. IEEE Trans Med Imaging 1998;17 (5):672-80. Zhou J, Rajapakse JC. Segmentation of subcortical brain structures using fuzzy templates. NeuroImage 2005;28 (4):915-24. Bondiau PY, Malandain G, Chanalet S, et al. Atlas-based automatic segmentation of MR images: validation study on the brainstem in radiotherapy context. Int J Radiat Oncol Biol Phys 2005;61(1):289-98. Villeger A, Lemaire JJ, Boire JY. Localization of target structures through data fusion applied to neurostimulation. Conf Proc IEEE Eng Med Biol Soc 2005;2:1508-11. Malandain G, Bardinet E, Nelissen K, Vanduffel W. Fusion of autoradiographs with an MR volume using 2-D and 3-D linear transformations. NeuroImage 2004;23(1):111-27. Niemann K, Mennicken VR, Jeanmonod D, Morel A. The Morel stereotactic atlas of the human thalamus: atlasto-MR registration of internally consistent canonical model. NeuroImage 2000;12(6):601-16. Chakravarty MM, Bertrand G, Hodge CP, Sadikot AF, Collins DL. The creation of a brain atlas for image guided
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
neurosurgery using serial histological data. NeuroImage 2006;30(2):359-76. Guo T, Finnis KW, Parrent AG, Peters TM. Development and application of functional databases for planning deep-brain neurosurgical procedures. Med Image Comput Comput Assist Interv Int Conf Med Image Comput Comput Assist Interv 2005;8(Pt 1):835-42. Finnis KW, Starreveld YP, Parrent AG, Sadikot AF, Peters TM. Three-dimensional database of subcortical electrophysiology for image-guided stereotactic functional neurosurgery. IEEE Trans Med Imaging 2003;22 (1):93-104. D’Haese PF, Cetinkaya E, Konrad PE, Kao C, Dawant BM. Computer-aided placement of deep brain stimulators: from planning to intraoperative guidance. IEEE Trans Med Imaging 2005;24(11):1469-78. D’Haese PF, Pallavaram S, Niermann K, et al. Automatic selection of DBS target points using multiple electrophysiological atlases. Med Image Comput Comput Assist Interv Int Conf Med Image Comput Comput Assist Interv 2005;8(Pt 2):427-34. Nowinski WL, Belov D, Thirunavuukarasuu A, Benabid AL. A probabilistic functional atlas of the VIM nucleus constructed from pre-, intra- and postoperative electrophysiological and neuroimaging data acquired during the surgical treatment of Parkinson’s disease patients. Stereotact Funct Neurosurg 2005;83(5–6):190-6. Nowinski WL, Belov D, Pollak P, Benabid AL. A probabilistic functional atlas of the human subthalamic nucleus. Neuroinformatics 2004;2(4):381-98. Nowinski WL, Belov D, Benabid AL. An algorithm for rapid calculation of a probabilistic functional atlas of subcortical structures from electrophysiological data collected during functional neurosurgery procedures. NeuroImage 2003;18(1):143-55. Yelnik J, Damier P, Bejjani BP, et al. Functional mapping of the globus pallidus. Contrasting effect of stimulation in the internal and external pallidal nuclei in Parkinson’s disease. Neuroscience 2000;101(1):77-87. Yelnik J, Damier P, Demeret S, et al. Localization of stimulating electrodes in patients with Parkinson disease by using a three-dimensional atlas-magnetic resonance imaging coregistration method. J Neurosurg 2003;99 (1):89-99. Lanotte MM, Rizzone M, Bergamasco B, Faccani G, Melcarne A, Lopiano L. Deep brain stimulation of the subthalamic nucleus: anatomical, neurophysiological, and outcome correlations with the effects of stimulation. J Neurol Neurosurg Psychiatry 2002;72(1):53-8. Saint-Cyr JA, Hoque T, Pereira LC, et al. Localization of clinically effective stimulating electrodes in the human subthalamic nucleus on magnetic resonance imaging. J Neurosurg 2002;97(5):1152-66. Hamel W, Fietzek U, Morsnowski A, et al. Deep brain stimulation of the subthalamic nucleus in Parkinson’s
Electronic stereotactic atlases
disease: evaluation of active electrode contacts. J Neurol Neurosurg Psychiatry 2003;74(8):1036-46. 81. Guo T, Finnis KW, Parrent AG, Peters TM. Visualization and navigation system development and application for stereotactic deep-brain neurosurgeries. Comput Aided Surg 2006;11(5):231-9. 82. Chakravarty MM, Sadikot AF, Germann J, Bertrand G, Collins DL. Anatomical and electrophysiological validation of an atlas for neurosurgical planning. Med Image Comput Comput Assist Interv Int Conf Med Image Comput Comput Assist Interv 2005;8(Pt 2):394-401. 83. Bardinet E, Bhattacharjee M, Dormont D, et al. A threedimensional, histological and deformable atlas of the human basal ganglia. II. Atlas deformation strategy and evaluation on retrospective series of parkinsonian
84.
85.
86.
87.
26
patients treated by deep brain stimulation. J Neurosurg, In press. Iosifescu DV, Shenton ME, Warfield SK, et al. An automated registration algorithm for measuring MRI subcortical brain structures. NeuroImage 1997;6(1):13-25. Ganser KA, Dickhaus H, Metzner R, Wirtz CR. A deformable digital brain atlas system according to Talairach and Tournoux. Med Image Anal 2004;8(1):3-22. Collins DL, Neelin P, Peters TM, Evans AC. Automatic 3D intersubject registration of MR volumetric data in standardized Talairach space. J Comput Assist Tomogr 1994;18(2):192-205. Hellier P, Barillot C, Corouge I, et al. Retrospective evaluation of intersubject brain registration. IEEE Trans Med Imaging 2003;22(9):1120-30.
393
33 Laitinen Stereotactic Apparatus M. I. Hariz . L. V. Laitinen
The Laitinen stereotactic system consists of the Stereoadapter and Stereoguide with auxiliaries for various radiological and surgical uses. The Stereoadapter is mainly used for stereotactic imaging with computed tomography (CT), magnetic resonance imaging (MRI), and stereotactic angiography. Together with the biopsy kit, the Stereoadapter is used for localization, biopsy, or stereotactic resection of brain lesions. It has also been used with the linear accelerator for fractionated stereotactic irradiation of brain tumors and arteriovenous malformations. The Stereoguide is used together with the Stereoadapter in functional neurosurgery such as pallidotomy, thalamotomy, and deep brain stimulation for movement disorders and pain as well as anterior capsulotomy and hypothalamotomy for psychiatric disorders.
Noninvasive Multipurpose Stereoadapter The stereoadapter was developed in 1982–1983 [1]. The original aim was to design a noninvasive, imaging-compatible, relocatable instrument for biopsy of brain tumors. It soon became evident that the Stereoadapter was accurate enough for functional neurosurgery, and since 1987 it has been used in all functional neurosurgery without ventriculography or plain radiology. The key for the high localizing accuracy is that the three reference points of the skull – i.e., the external auditory meatus and the bridge of the nose – give a good relocating stability. The second reason for the high accuracy is that the
#
Springer-Verlag Berlin/Heidelberg 2009
imaging fiducials lie immediately on the scalp and not at a far distance from it, as is the case with other stereotactic frames. The Stereoadapter is mounted to the patient’s head by means of a nasion support, two earplugs, and a strapping band at the occiput. Neither general nor local anesthesia is needed. Repeated mountings of the Stereoadapter have shown a high degree of reproducibility and tolerability [2–4]. The Stereoadapter is made of an aluminum alloy and reinforced plastic. It consists of two lateral triangular components with four transverse bars each, a connector plate, a nasion support component, and frontal laterality indicator pins (> Figure 33-1). The transverse bars are 2 mm thick in a dorsoventral direction and lie 25 mm apart from each other. They connect the anterior and posterior ear arms and are perpendicular to the posterior ear arm. Cogwheel cases join the lateral triangle components to the nasion support arms, which have millimeter scales. By winding the cogwheel screws, the nasion support is pressed against the bridge of the nose. The earplugs lie at the posterior ear arms and are pressed against the external auditory meati by means of a threaded lever in front of the nasion support. The connector plate at the vertex joins the lateral triangle components together and serves to press the triangle components against the scalp. A frontal pin mounted between the vertex connector plate and the nasion support serves as the laterality reference structure (> Figure 33-1). When the Stereoadapter has been mounted on the head, its position is recorded by the
512
33
Laitinen stereotactic apparatus
. Figure 33-1 The Stereoadaptermounted on the head. Left: the patient’s head is lying on a plastic cushion, and the Stereoadapter is immobilized by a multijoint mechanism prior to the imaging study. The star indicates the posterior ear arm of the right triangular component. TB = transverse bar. Right: frontal view. CP = connector plate, holding the lateral triangular components pressed to the scalp
symmetrical millimeter scales on both sides of the nasion support arms and on the connector plate of the Stereoadapter. The Cartesian reference structures of the Stereoadapter are the sagittal midplane passing through the frontal pin for the laterality x coordinate, the frontal plane between the anterior borders of the right and the left posterior ear arms for the antero-posterior (AP) y coordinate, and the transverse plane between a pair of transverse bars for the dorsoventral z coordinate. With the Stereoadapter mounted to the head, the dorsal most transverse bar level corresponds to the average height level of the cingulum, the second dorsal most bars to the average height level of the intercommissural line, and the third bar pair to the average level of the amygdala. For CT, MRI, or angiography study as well as for stereotactic irradiation, the Stereoadapter is mounted to the head and then immobilized to a plastic plate with a multijoint mechanism (> Figure 33-1). For MRI studies, thin tubes containing 2 mmol copper sulfate or olive oil are attached to the reference structures of the Stereoadapter. These fiducials give an artifactfree sharp image. For single photon emission
computed tomography (SPECT) or positron emission tomography (PET) studies, the tubes mentioned above can be filled with an appropriate solution of isotope, thus providing visible reference marks on the respective pictures. For angiography, either a conventional or digital angiography, the ordinary earplugs of the Stereoadapter are replaced by similar earplugs containing a 1-cm lead pin, and, a 1-cm lead pin is placed on the forehead of the patient. These lead pins provide the magnification factor on the AP view. On the side view, the magnification is given by the already known distance of 25 mm between two sets of the transverse bars [5]. For stereotactic irradiation, plastic plates are attached to the triangular components of the Stereopadapter (> Figure 33-2). These plates have lines indicating the position of the transverse bars and posterior ear arms, respectively, and serve to align the patient’s head according to the lateral laser beams of the linear accelerator. Instead of the ordinary frontal pin, a ruler with sliding millimeter scales and a cone are used (> Figure 33-2). This device serves to align the brain target in a lateral direction according to the frontal laser beam. The laser beam must coincide
Laitinen stereotactic apparatus
. Figure 33-2 The stereoadapter mounted to the head prior to stereotactic irradiation. Plastic plates to indicate the brain target’s y and z coordinates are attached to the lateral triangular components. The laser cross lines are aligned with the reference markings of the AP and height coordinates on the plate. The cone on the forehead is attached to a laterality ruler. Both can slide according to the laterality of the target. The couch is moved until the frontal laser line hits the tip of the cone
33
cannulas. The probe carrier is rigidly attached to the connector plate of the Stereoadapter for either an anterior (frontal) or posterior (parietooccipital) approach (> Figure 33-3). The ordinary rod of the probe carrier can be replaced by a long curved one should a lateral or posterior fossa approach be needed. The Sedan biopsy probe consists of an outer cannula 250 mm long and 2 mm thick, with a 5-mm side opening at its distal end. An inner cannula with a sharpened end functions as a guillotine when it is advanced through the outer cannula during aspiration of a brain tumor specimen. After setting the phantom base according to the CT/MRI coordinates of the target, the tip of the biopsy cannula is directed against the phantom target, after which the hinge clamps are tightened (> Figure 33-3).
Stereoguide
with the tip of the cone. Thus, the cone allows for a visual verification of the proper alignment of the target when the couch is rotated for multiplanar irradiation.
Tumor Biopsy Kit The biopsy kit consists of a phantom base, probe carrier, twist drill, diathermy probe, and biopsy cannula. The phantom base is mounted between the right and left transverse bars of the Stereoadapter at a desired height level (> Figure 33-3). It has two slide components with millimeter scales for the x and the y coordinates, respectively. The slide component for the y coordinate has a millimeter-scaled vertical rod for the z coordinate. The probe carrier has two steel rods, two hinge clamps, and two concentric probe-guiding
The Laitinen Stereoguide is a stereotactic frame that functions according to the arc-radius principle [6]. It consists of an oval base ring fixed to the head with four steel pins (> Figure 33-4). Cylindrical components with millimeter scales are mounted on the lateral sides of the base ring. Cogwheel mechanisms permit the cylinder components to slide in such a way that their common axis coincides with the y and z positions of the intracranial target. A vernier scale ensures an accuracy of 0.25 mm. A semicircular arc carrying the electrode, endoscope, or other probe is mounted to the cylindrical components. It can slide in a lateral direction to bring the probe tip to the lateral x position of the target (> Figure 33-4). Thus the target lies at the center of the spherical system of the Stereoguide and can be reached from any suitable direction. If needed, the base ring of the Stereoguide, mounted on the patient, may be attached rigidly to a floor stand for surgery. Then, if intraoperative
513
514
33
Laitinen stereotactic apparatus
. Figure 33-3 The biopsy procedure. Left: the biopsy kit mounted to the Stereoadapter. The phantom base is mounted at the level of the second dorsal most transverse bars and shows the x, y, and z positions of the brain target. The probe carrier is mounted to the connector plate of the Stereoadapter for a parietal approach. The biopsy needle points to the phantom target and the hinge clamps are tightened. Right: the Stereoadapter mounted to a dummy. The probe carrier is attached and the guiding cannula is directed toward the intracranial target
radiology is needed, the central beams of suitably adjusted lateral and AP x-ray tubes will pass through the y, z, and x origins of the frame, respectively [7].
Stereotactic Radiological and Surgical Applications Nonfunctional Stereotaxis The stereotactic management of a brain tumor, abscess, cyst, deep brain hematoma, etc., begins by mounting the Stereoadapter to the head and performing a stereotactic CT (or MRI) study [8–10]. The scanning is performed throughout the tumor area with 2- or 3-mmthick contiguous slices parallel to the transverse bars of the Stereoadapter (> Figure 33-5). The Stereoadapter is then detached from the head. The calculation of target coordinates may be done manually or using the software of the CT
or MRI machine. The y coordinate of the target is its distance from the interaural plane, the x coordinate is the distance from the medial border of the right posterior ear arm, and the z is the distance from the CT scan containing the target point and to that showing the nearest pair of transverse bars (> Figure 33-5) [8]. Surgery may take place at any suitable time after the CT/MRI study. The phantom base and probe carrier are set according to the CT/MRI target coordinates, after which the phantom base is detached (> Figure 33-3). The Stereoadapter with the probe carrier locked to it is remounted to the head. The inner probe-guiding cannula is replaced by a similar sterile one. Using local anesthesia, the skull is trephined with a 2.15mm-thick twist drill introduced through the guiding cannula. Diathermy coagulation is applied to the dura and cortex. The biopsy needle is then introduced to the target. Tumor specimens are obtained from various depths along the track of the needle. The procedure usually
Laitinen stereotactic apparatus
. Figure 33-4 The Stereoadapter and Stereoguide mounted to a patient during a functional stereotactic procedure. A. The base ring of the Stereoguide is fixed to the head with screws. A steel pin introduced through the cylindrical component points at the y and z origins of the Stereoadapter (arrow) during setting of the surgical y and z coordinates. B. The probe carrier arc is attached to the cylindrical components. The electrode points to the frontal pin of the Stereoadapter (arrow) during setting of the surgical x coordinate (see text for details)
33
. Figure 33-5 A 2-mm-thick stereotactic CT scan at the level of the second dorsal most transverse bars of the Stereoadapter. The dotted line xb indicates the laterality x of the biopsy target, measured from the right transverse bar (see text). The AP coordinate of the target is indicated by y. The target’s laterality measured from the sagittal midplane of the Stereoadapter is indicated by x
the Stereoadapter can be drawn on the scalp using the same technique. The Stereoadapter is then removed and a small centered craniotomy is done [9]. takes 25–35 min [8]. The same instrumentation and technique can be used for the stereotactic placement of a drainage catheter in cases of cyst or abscess. Similarly, for a deep-seated tumor scheduled for stereotactic craniotomy and resection, a guiding catheter can be placed stereotactically at the edge of the tumor, after which the Stereoadapter is detached. The catheter is cut along the surface of the skin, a small centered flap is created, and a craniotomy is done. The catheter is then followed toward the tumor using small spatulas and routine microsurgical techniques. For resection of small superficial brain tumors, the location of the tumor in relation to
Fractionated Stereotactic Irradiation When a brain lesion is scheduled for stereotactic irradiation, the stereotactic CT scanning ought to include not only the area of the brain pathology but also the whole calvarium. With an arteriovenous malformation (AVM), stereotactic angiography with the Stereoadapter is done. The target coordinates are indicated on special side plates attached to the triangular components of the Stereoadapter. The lateral x coordinate is measured on the CT scan in relation to the midsagittal plane of the Stereoadapter (> Figure 33-5).
515
516
33
Laitinen stereotactic apparatus
The Stereoadapter is remounted to the head with the patient lying on the couch of the linear accelerator (> Figure 33-2). The couch is moved so that the two laser cross lines from each side and the vertical laser beam from the ceiling indicate that the isocenter of the accelerator coincides with the Stereoadapter markings of the brain target – that is, the markings on the side plates and the frontal cone (> Figure 33-2). The Stereoadapter is then locked to the couch with a multijoint mechanism similar to that used for CT and MRI. During irradiation, the patient is monitored using a video camera. Each irradiation session, comprising five or six multiplanar fixed beams with variable collimation, lasts for about 30 min. The procedure may be repeated according to the schedule of fractionation [5,11,12].
Basically, MRI and CT scanning with the Stereoadapter are very similar. The positioning of the patient into the coil is, however, not as crucial as positioning into the CT gantry as far as the right-left alignment is concerned. The sagittal survey image of the triangular components should be obtained first. On this image, the tubes filled with copper sulfate or olive oil, indicating the references for the AP and height coordinates, are visualized (> Figure 33-6). The axial MRI scanning should also be done parallel to the transverse bars, and coronal scanning should be parallel to the posterior ear arms. In our experience, a stereotactic MRI study with axial thin slices takes about 15 min.
Functional Stereotaxis
Enlarged film copies of the CT/MRI scans are obtained from the area between the foramina of Monro (FM) and the proximal aqueduct, including the second pair of transverse bars. On the CT scan, the anterior commissure (AC) is localized according to the method of Laitinen and coworkers [1] on a slice lying 4 mm ventral to the ventral most margin of the foramina of Monro. If the scanning plane is parallel to the intercommissural line (ICL) – i.e., the transverse bars of the Stereoadapter are parallel to the ICL – the posterior commissure (PC) is seen on the same slice. If the scanning plane is not parallel to the ICL, adjacent CT/MRI slices of the area are studied in order to visualize the beginning of the aqueduct; the last slice before the appearance of the aqueduct is chosen to represent the level of the PC. This film copy is superimposed on that where the AC had been marked, and the position of the PC is transferred to the latter. Thus, the level of the ICL is determined in relation to the scanning level [3]. The mean angulation between the transverse bars and intercommissural line is 0.75 with a range of 7 [7].
The CT and MRI Studies After mounting the Stereoadapter, the head, resting on a plastic cushion, is aligned so that the connector plate of the Stereoadapter is parallel to the transverse laser beam of the gantry, after which the Stereoadapter is locked to a plastic plate on the CT table and a lateral Scoutview of the head is obtained. The cursor line is brought to the level of and parallel to the dorsal most transverse bars of the Stereoadapter, so that the scanning plane is parallel to the transverse bars and perpendicular to the posterior ear arms of the Stereoadapter. The visualization of the transverse bars is checked on the first CT slice to assure parallel alignment. Beginning from the level of the dorsal most transverse bars, 1.5- or 2-mm thick slices are scanned in 2-mm steps until the second transverse bars and the proximal part of the aqueduct are visualized, following which the Stereoadapter is detached. The CT study lasts for 10–15 min.
Determination of the Ventricular Landmarks and Target Coordinates
Laitinen stereotactic apparatus
33
. Figure 33-6 CT and MRI scans of one patient performed on consecutive days. Both scans are 2 mm thick at a dorsoventral level 4 mm ventral to that of the anterior commissure-posterior commissure line and at the level of the second dorsal most transverse bars of the Stereoadapter. The left posteroventral pallidal target is indicated by an encircled dot
The anatomic position of the brain target – be it the ventrolateral thalamus, the posteroventral pallidum, or the anterior internal capsule, etc. – can now be plotted on the appropriate CT/MRI slice in relation to the AC and PC. Then, the coordinates of the target point are measured in relation to the reference structures of the Stereoadapter (> Figure 33-6). The y coordinate of the target is its distance from the interaural plane of the Stereo-adapter. The x coordinate is measured in relation to the sagittal midplane of the Stereoadapter, formed by projecting the frontal laterality indicator pin perpendicularly onto the interaural line. The z coordinate is the distance between the target level and the level of the second pair of transverse bars. In routine procedures, calculation of the functional brain target coordinates on either CT or MRI scans lasts for 5–10 min.
Surgery The surgery may be performed at convenience after the CT/MRI study. The Stereoadapter is
remounted on the head. The base ring of Laitinen’s Stereoguide is mounted around the Stereoadapter fairly parallel to its transverse bars. By means of two adjustable lateral support components, the base ring is so positioned that it lies as symmetrically as possible in relation to the Stereoadapter. Under local anesthesia, the base ring is rigidly fixed to the skull by means of four percutaneous pins. The patient is placed on the surgical table. The cylinder components of the Stereoguide are mounted on the left and the right sides of the base ring. Plastic cylinder blocks with axial steel pins are introduced through the cylinders, which are moved into such a position that the steel pins point to the y and z origins of the Stereoadapter – i.e., to the intersection of the second transverse bar and the anterior margin of the posterior ear arm. In this way, the y and z origins of the Stereoadapter are transferred to the cylinder components, and recorded on the corresponding millimeter scales of the Stereoguide (> Figure 33-4). The CT coordinates y and z of the surgical target are then added to the y and z readings of the Stereoadapter’s origin, after which the cylinder components of the
517
518
33
Laitinen stereotactic apparatus
Stereoguide are positioned according to the sum. Thus the steel pins of the cylinder blocks point to the brain target, the y and the z coordinates of which are read on the millimeter-scales of the Stereoguide. The semicircular arc of the Stereoguide is mounted to the cylinder components. The electrode carrier on the arc is moved into a 90 position. The surgical probe is directed toward the frontal pin (> Figure 33-4). The position of the pin is recorded on the lateral millimeter scale of the cylinder components. Then the CT x value of the target is added to the recorded position of the frontal pin; the sum is the final stereotactic x coordinate. The arc is moved into this position. The Stereoadapter is then detached, and the surgery may proceed as usual [3].
Remote Postoperative Imaging Studies The noninvasive Stereoadapter permits a stereotactic imaging study to be performed months
after surgery for control of the final size and site of the stereotactic lesion [13]. This unique feature is of outmost importance, since a remote postoperative imaging study may show the final shape of the lesion after the complete resolution of postoperative edema (> Figure 33-7). The stereotactic imaging study is performed in a manner similar to the preoperative study. In this way the lesion can be accurately assessed in relation to the preoperative target point and in relation to the reference structures of the third ventricle (> Figure 33-7). Since the Stereoadapter’s position on the head will be the same postoperatively as it was pre- and intraoperatively, an exact radiological correlation between pre- and postoperative scanning can be made [13]. In cases where a permanent electrode for chronic electrical stimulation had been implanted in the brain, a plain x-ray performed in a stereotactic manner with the Stereoadapter remounted to the head provides a stereotactic control of the exact position and coordinates of the electrode tip at any time after the surgery.
. Figure 33-7 Preoperative and remote postoperative stereotactic CT scans of one patient. On the left scan, which is 2 mm thick, the preoperative pallidal target is indicated by a dot (arrow). The right scan, performed 4 months after surgery, represents the superposition of two contiguous, 2-mm-thick CT slices of the same area. The pallidal radiofrequency lesion is thus enhanced
Laitinen stereotactic apparatus
Conclusions The Laitinen system is based on stereotactic imaging using the noninvasive Stereoadapter. The Stereoadapter is not individualized and fits most heads [2]. The reference structures of the Stereoadapter lie extremely close to the head and therefore to the target, which is a unique feature of this system. Calculation of target coordinates can be done easily and quickly by using the inherent software of the CT/MRI machine or manually with a ruler, an ink pen, and a mini calculator. The noninvasive design permits flexible and rational planning of different diagnostic and therapeutic stereotactic procedures. The technique obviates the need for surgery inside the CT machine. Patients can be operated on or irradiated when suitable for them, the surgeon, the radiotherapists, and the involved staffs. For brain biopsy, there is no need for an additional frame, since the Stereoadapter as such also functions as a probe carrier. This simplifies the procedure and markedly reduces the duration and costs of surgery. For stereotactic open resection of small brain tumors, the Stereoadapter does not interfere with the craniotomy, since it is removed from the head once the position of the tumor has been indicated on the scalp or by a catheter. The reproducibility of results from the noninvasive Stereoadapter permits an accurate repositioning of the brain target into the isocenter of the linear accelerator for fractionation of stereotactic irradiation. Furthermore, the use of the Stereoadapter for target localization can be combined with a neurosurgical navigation system by providing relocatable Cartesian references [14]. A great advantage to the patient in functional stereotaxy is the avoidance of ventriculography [15,16]. The accuracy of the method permits a functional stereotactic procedure to be carried on with minimal side effects and short hospital stays [16,17]. The noninvasive Stereoadapter makes possible a remote post-operative stereotactic
33
CT study for checking the site and the size of the final radiofrequency lesions and also for assessing the accuracy of the whole stereotactic procedure [2,3,13]. It is important to keep in mind that the noninvasive fixation of the Stereoadapter to the head requires good cooperation of the patient unless sedation is used [2]. The pressure exerted by the earplugs on the external auditory meatus and by the nasion support on the bridge of the nose, although generally well tolerated, may be uncomfortable to some patients. Besides, the noninvasivity calls for great care on the part of the surgeon in mounting the frame, positioning the patient, and closely supervising the scanning or irradiation procedure. Through 10 years of intensive use on more than 1,000 patients for all stereotactic imaging, surgical, and radiotherapeutic procedures, it is felt that the Laitinen system has been versatile, reliable, easy to use, time-saving, and inexpensive. The lack of sophistication and simplicity of the system have been experienced as an advantage rather than an inconvenience. However, as in any system, the surgeon should be well acquainted with it and learn to profit from its advantages while avoiding its pitfalls.
References 1. Laitinen LV, Liliequist B, Fagerlund M, Eriksson AT. An adapter for computed tomography-guided stereotaxis. Surg Neurol 1985;23:559-66. 2. Hariz MI. A non-invasive adaptation system for computed tomography-guided stereotactic neurosurgery. Thesis, Umea˚ University Medical Dissertations, New series no 269, ISSN 0346–6612. Umea˚, Sweden: Umea˚ University Printing Office; 1990. 3. Hariz MI. Clinical study on the accuracy of the Laitinen’s non-invasive CT-guidance system in functional stereotaxis. Stereotact Funct Neurosurg 1991;56:109-28. 4. Hariz MI, Eriksson AT. Reproducibility of repeated mountings of a noninvasive CT/MRI stereoadapter. Appl Neurophysiol 1986;49:336-47. 5. Bergenheim AT, Hariz MI, Henriksson R, Lo¨froth P-O. Fractionated stereotactic irradiation of brain tumors and
519
520
33 6.
7.
8.
9.
10.
11.
Laitinen stereotactic apparatus
arteriovenous malformations using the linear accelerator and a non-invasive frame. In: Lunsford LD, editor. Stereotactic radiosurgery update. Elsevier; New York: p. 73-5. Laitinen LV. The Laitinen system. In: Lunsford LD, editor. Modern stereotactic neurosurgery. Boston: Martinus Nijhoff; 1998. p. 99-116. Hariz MI, Bergenheim AT. A comparative study on ventriculographic and computed tomography-guided determinations of brain targets in functional stereotaxis. J Neurosurg 1990;73:565-71. Hariz MI, Bergenheim AT, DeSalles AAF, et al. Percutaneous stereotactic brain tumor biopsy and cyst aspiration using a non-invasive frame. Br J Neurosurg 1990;4:397-406. Hariz MI, Fodstad H. Stereotactic localization of small subcortical brain tumors for open surgery. Surg Neurol 1987;25:345-50. Nguyen J-P, Decq P, Brugie`res P, et al. A technique for stereotactic aspiration of deep intracerebral hematomas under computed tomographic control using a new device. Neurosurgery 1992;31:330-5. Delannes M, Daly NJ, Bonnet J, et al. Fractionated radiotherapy of small inoperable lesions of the brain
12.
13.
14. 15.
16.
17.
using a non-invasive stereotactic frame. Int J Radiat Oncol Biol Phys 1991;21:749-55. Hariz MI, Henriksson R, Lo¨froth P-O, et al. A noninvasive method for fractionated stereotactic irradiation of brain tumors with linear accelerator. Radiother Oncol 1990;17:57-72. Hariz MI. Correlation between clinical outcome and size and site of the lesion in CT-guided thalamotomy and pallidotomy. Stereotact Funct Neurosurg 1990;54:172-85. Takizawa T. Neurosurgical navigation using a noninvasive stereoadapter. Surg Neurol 1993;40:299-305. Hariz MI, Bergenheim AT. Clinical evaluation of CT-guided versus ventriculography-guided thalamotomy for movement disorders. Acta Neurochir Suppl 1993;58:53-5. Laitinen LV, Bergenheim AT, Hariz MI. Leksell’s posteroventral pallidotomy in the treatment of Parkinson’s disease. J Neurosurg 1992;76:53-61. Hariz MI, Bergenheim AT, Fodstad H. Airventriculography provokes an anterior displacement of the third ventricle during functional stereotactic surgery. Acta Neurochir 1993;123:147-52.
30 Leksell Stereotactic Apparatus L. D. Lunsford . D. Kondziolka . D. Leksell
Background Any description of the Leksell stereotactic system must begin with a historical vignette that describes the genesis of its creation. Lars Leksell was a brilliant, innovative, and persistent pioneer in the emerging field of neurological surgery. His creative genius eventually led him to become professor of neurological surgery at the Karolinska Institute in Stockholm, Sweden. His career, however, began under the direction of Ragnar Granit, a Nobel Prize-winning neurophysiologist with whom Leksell collaborated. Leksell’s doctoral thesis presented the first description of the spinal cord gamma motor neuron system. Following this work, Leksell’s neurosurgical training began in the department of Herbert Olivecrona. At this time, in the early 1940s, the alarming mortality rate for routine neurosurgical procedures approximated 40%. Poor outcomes, difficulty with anesthesia (neurosurgical patients were allowed to ventilate spontaneously) and severe blood loss during surgery were features that made a lasting impression on Leksell. He was struck by the paradox of crude instrumentation that was poorly designed for the delicate central nervous system. His subsequent career was committed to the development of less invasive surgical techniques that facilitated management of a wide variety of intracranial problems. During the subsequent four decades, he became one of the most original contributors to the field of neurological surgery. In 1947 Leksell studied in Philadelphia with Spiegel and Wycis, who developed the first human stereotactic apparatus [1]. When he returned to Stockholm a year later, he developed his
#
Springer-Verlag Berlin/Heidelberg 2009
first-generation stereotactic guiding device, which was ‘‘easy to handle and practical in routine clinical work’’ (> Figure 30-1). His first device was reported in 1949 [2]. Ease of use and practicality were concepts that remained preeminent principles of all subsequent Leksell systems as they evolved. The designs have focused on utility, accuracy, and versatility. The continued development of stereotactic instruments responded to the challenge of new surgical needs presented by deep brain surgery, diagnostic biopsy, radiosurgery, and functional neurosurgery. In addition, constantly improving neurodiagnostic imaging tools such as ultrasound, encephalography, angiography, computed tomography (CT), positron emission tomography (PET), magnetic resonance imaging (MRI), magnetic resonance angiography (MRA) [3], and magnetic source imaging required repeated revisions in the system design.
The Arc-Centered Principle Lars Leksell’s original description is itself an eloquent and simple analysis of the arc centered concept (> Figure 30-2): ‘‘Essentially, it consists of a semicircular arc with a movable electrode (probe) carrier. The arc is fixed to the patient’s head in such a manner that its center corresponds with a selected cerebral target. The electrodes (probes) are always directed towards the center and hence to the target. Rotation of the arc around the axis rods in association with lateral adjustment of the electrode (probe) carrier enables any convenient point of entrance of the
470
30
Leksell stereotactic apparatus
. Figure 30-1 The original Leksell stereotactic coordinate frame first reported by Lars Leksell in ‘‘The principle of the semicircular arc is identified in this first model stereotactic instrument’’
. Figure 30-2 The Leksell arc-centered design comprises a movable instrument carrier that slides on a semicircular arc. The carrier can be swung along the left-right ‘‘arc’’ angle and in an anterior-posterior ‘‘ring’’ angle
electrodes (probes) to be chosen independent of the site of the target [4].’’ First, an imaging-compatible base ring is attached by pins to the patient’s head. Next, the base ring is coupled with a fiducial system to enable target localization using a wide variety of radiological techniques. The frame has been modified
steadily over the past 50 years in order to be compatible with each advance in neurodiagnostic imaging. Versions of the frame over the years (currently model G) have included improvements in instrument carriers and coordinate fixation attachments as well as the use of non-ferromagnetic materials enabling use with MRI, PET, and CT.
Leksell stereotactic apparatus
As of early 2008 the Leksell stereotactic system was used at more than 1,300 neurosurgical centers in more than 50 countries worldwide. A semicircular arc is attached to the base ring of the frame such that the arc can be swung around its rings in the anteroposterior direction, or left and right if mounted on the anterior and posterior parts of the frame base. The instrument carrier slides along the arc. The radius of the arc is 190 mm. In essence, an intracranial target can be approached from any entry point. The most important principle is to select a safe trajectory to reach the defined target. Within the Leksell stereotactic family of instruments, the only exception to this is the Leksell Gamma Knife, first used by Leksell in 1967 [4]. Because the locations of the multiple cobalt-60 sources are fixed, the patient is moved into the point of beam intersection at the center of the collimation system. This position is determined by the stereotactic X, Y, and Z coordinates of the target. The same base ring is used for open image-guided surgery, functional neurosurgical procedures and stereotactic radiosurgical operations.
The Model G Instrument The currently available Leksell stereotactic instrument can be used interchangeably with all currently available imaging techniques. The frame consists of a rectangular base ring 190 210 mm (> Figure 30-3). The X (left-right) coordinate is set on the semicircular arc, which is attached to the frame at the chosen Y (anteroposterior) and Z (superoinferior) coordinates. Either a straight front piece or a curved front piece that accommodates the nose is used (> Figure 30-4). The X, Y, and Z axes of the coordinate system conforms to the X, Y, and Z geometry of CT, MRI, and PET scans. An imaginary frame origin (X, Y, and Z = 0) is in the upper posterior right side of the coordinate frame (> Figure 30-5). A fiducial system is attached to the base ring during the
30
. Figure 30-3 The current base ring for the model G Leksell stereotactic instrument
. Figure 30-4 The base ring has an front plate (upper) to which the semicircular arc can be attached in an antero-posterior position. Usually the curved front piece is used and the semicircular arc attached to the right and left sides
471
472
30
Leksell stereotactic apparatus
. Figure 30-5 The coordinate system of the Leksell model G stereotactic base ring. At the center of the frame the X, Y, and Z coordinates are 100. An imaginary ‘‘zero’’ exists superior, posterior and on the right side of the fiducial system
. Figure 30-6 The patient is positioned for CT localization. The fiducial system can be seen attached to the base ring
. Figure 30-7 The MRI fiducial system with frontal, posterior, lateral, and superior plates for determination of the X, Y, and X coordinates in multiplanar MRI imaging. The fiducial consists of a plastic tube filled with diluted copper sulfate solution
neurodiagnostic imaging component of the procedure (> Figure 30-6). A variety of fiducial systems are available for different imaging modalities, including systems for multiplanar MRI studies (> Figure 30-7). A radio opaque scale is available for conventional radiographic examination during the now rarely performed contrast encephalography for functional procedures (> Figure 30-8). A fiducial system compatible with computer-derived coordinate techniques is also available for digital subtraction angiography, which is often used during Gamma Knife radiosurgical procedures for arteriovenous malformations (> Figure 30-9). A variety of surgical instruments can be attached to the semicircular arc (> Figure 30-10). The instrument carrier can be moved to any arc angle and the arc rotates to any ring angle, both of which can be determined using a computer simulation technique. The working length of all probes is 190 mm, which corresponds to the radius of the semicircular arc. When the probe stop on
Leksell stereotactic apparatus
the instrument carrier of the arc is set at zero, the operative end of the probe will reach the target with an accuracy of 0.7 mm. The accuracy of reproducibility and precision of stereotactic . Figure 30-8 Angiographic localizers for computer based determination of the coordinates (left) or conventional coordinate determination (right) with a radio opaque system that is attached to the base ring
30
targeting using the Leksell system meets the guidelines established by the American Society for Testing and Manufacturing. Ultimately, accuracy is dependent on the imaging resolution, which is 0.5 mm for 512 512 matrices used during digital imaging acquisitions such as CT.
Application of the Coordinate Frame The stereotactic coordinate base ring is attached to the patient’s head using four pins (either aluminum screws with hard metal tips, titanium screws or disposable aluminum screws of various lengths (> Figure 30-11). The location of the pins is adjusted by changing the position of the posts on the base ring. (> Figure 30-12). The position of the base ring on the patient’s head is adjusted using ear bars that are temporarily placed in the external auditory canal. Although symmetrical placement is not critical,
. Figure 30-9 Antero-posterior and lateral x-rays visualize the coordinate x-ray indicators on the x-ray film. The X, Y, and Z coordinates are determined by simple calculation or a graphic method
473
474
30
Leksell stereotactic apparatus
. Figure 30-10 The Leksell model G expanded stereotactic arc can be attached to the base ring in the lateral or anteroposterior direction
. Figure 30-11 Application of the stereotactic frame requires that pins be inserted through the skull and into the outer table of the skull under local anesthesia. After application, gently pulling on the frame ensures that the frame and head move together
these ear bars allow for appropriate symmetrical placement of the frame on the patient’s head. Cushioning the ear bars with a small amount of foam in the external auditory canal helps to reduce discomfort from frame torsion during pin placement. Surgeons should view the preoperative imaging in order to make decisions relative to frame placement. Although appropriate shifting of the frame relative to the patient’s head is of little consequence for most open stereotactic or functional neurosurgical procedures, it is very
. Figure 30-12 The position of the pins is adjusted by changing the position of the posts on the base ring
important for procedures done with earlier Gamma Knife models. During such procedures, the surgeon attempts to place the lesion as close as possible to the center of the frame (X, Y, and Z = 100). Such frame application issues are not as relevant for procedures performed with the latest generation Gamma Knife Perfexion, since there is a greatly expanded working space with little or no chance of frame or pin interference with the collimating system. In the vast majority of patients, the stereotactic head frame is applied within approximately 5 min. Most patients benefit from mild sedation (either oral or intravenous) in order to reduce mild discomfort and anxiety. Some surgeons mix the local anesthetic with sodium bicarbonate to reduce the initial burning sensation associated with injection of the mixture of local anesthetic agents. For those patients who express significant anxiety (especially young men who seem to have a propensity for vasovagal symptoms), both additional sedation and an intravenous anticholinergic medication may be beneficial. We routinely prep the entire head only with isopropyl alcohol. No hair shave is performed.
Leksell stereotactic apparatus
The base ring and all pins are sterilized in advance of placement. Screw-type pins reduce frame application time to a matter of minutes. The pins can be used satisfactorily in children over the age of two without difficulty but should be used carefully in patients under that age, since the skull may still be compressible. We have used blood Vacutainer stoppers against the scalp (to diffuse pressure from the pin) in children as young as 5 months. In addition, the pins must be inserted in such a way that excessive distraction of the support bars does not occur. We tighten the pins diagonally, two and two at the same time, alternating between the pairs. No torque instrument is necessary. Once the pins are firmly secured, subsequent movement of the head will not result in frame dislodgment. Proper application can be assured by gently lifting the frame base ring and making sure that the patient’s head and the frame move together. In general, we apply the frame with the patient on a stretcher in the semi-sitting position with the legs slightly flexed at the hips, which enhances comfort and allows immediate transfer to the imaging site.
Conventional Radiographic Target Localization The x-ray-compatible indicator box is attached to the base ring. Anterior-posterior, and lateral images during planar radiography, encephalography, or conventional, single, and bi-plane digital subtraction angiography will result in recognition of proximal and distal X, Y and Z coordinates. Leksell’s original ingenious construction of a geometric spiral diagram to determine the target coordinates was never met with great acceptance. It may even have helped to perpetuate the myth that stereotactic surgery was excessively complex. The planar film localization technique has been replaced by scaled graphs, tabletop stereotactic calculators or computer based planning systems
30
(Leksell SurgiPlan for open Stereotactic procedures and Leksell GammaPlan for Gamma Knife surgery). It is critical to understand the position of the x-ray source referable to the frame so that the surgeon can distinguish between proximal and distal frame values. Magnification is irrelevant, and any distance between x-ray source, frame, and film can be chosen, provided that all fiducials are visualized on the resultant images. Selection of the central beam at an X, Y, and Z of 100 helps to facilitate easy visualization of the frame fiducials. Rotation is not a problem as long as all fiducials are recognized in the films. Film subtraction techniques are especially valuable during angiography.
Computed Tomography and Magnetic Resonance Imaging The Leksell system was one of the first instruments to be compatible with the development of computed tomography. In 1977, Lars Leksell together with Bengt Jernberg and Hans Sundquist, his engineering colleagues, began to redesign the frame to make it CT-compatible [5]. This required reduction in the amount of metal within the frame then in use (Model B). They developed imaging-compatible pins (plastic or carbon fiber) and an appropriate fiducial system that would allow conversion of CT scanner pixel information into accurate twodimensional stereotactic imaging. The model D was the first MRI-compatible frame [6,7]. In the current model G instrumentation, a CT or MRI-compatible fiducial system is anchored to the base ring using snap-on clips. Four slots in the base of the fiducial system prevent incorrect positioning of the fiducial system. Lateral side plates are critical and an anterior or posterior plate may be used to provide stability during the initial CT or MRI. During the imaging examination, the responsible surgeon takes time to confirm that the frame is not significantly rotated
475
476
30
Leksell stereotactic apparatus
referable to the scanner plane. All scans are performed parallel (or rarely perpendicular) to the base ring, as can be seen from the initial CT scout film or the MRI localizer scan. Because of the rapid acquisition time of current CT 64 slice scanners (entire head with 2 mm slices in 4 s), the scan time is very brief. Similarly for MRI, a volume acquisition of the head at 1.5–2 mm slices can be performed in 8–10 min depending on the number of excitations. Most imaging can be performed with a 25-cm field of view in order to increase target resolution. The frame-center (X, Y = 100) central pixel is defined and the target coordinates are determined referable to the frame center. Most brain lesions undergoing stereotactic surgery can be visualized with 2 mm-thick images. Narrower images (e.g., 1 mm) can be used but, while they improve spatial resolution, they do so at the expense of contrast resolution. Enhancement of the brain or the target by intravenous contrast sometimes helps gray-white differentiation and facilitates targeting of contrast enhancing lesions. In essence, CT is very satisfactory for lesions that can be seen well on preoperative CTs, while MRI is the preferable imaging tool for lesions that are best identified by MRI scan and for imaging during functional neurosurgery. Determination of the superoinferior stereotactic plane (Z coordinate) is done by the measuring of the fiducials and requires simple distance measurement functions available on the computer. The Leksell system is very stable in the field of 1.5 Tesla imaging, so that in most cases additional CT imaging is not necessary. If both MR and CT imaging are used, fusion of the two data sets is easy in the current planning systems. Direct calculation of the target can be performed on routinely available imaging software at the imaging site, using measure distance functions that allow determination of the target coordinates referable to the frame center. Alternatively, the images can be transferred by CD or Ethernet to the planning system [8].
SurgiPlan permits rapid calculation of stereotactic targets; however, in general, the scanner software itself can be used to accurately and rapidly make target coordinate determinations during the actual image acquisition interval. As noted modern-generation CT scanners (spiral scanners and high-resolution head techniques) often allow CT scan data acquisition in a matter of seconds. When using CT, effort should be made to place the pins outside of the imaging area of interest in order to reduce artifacts on the images. Magnetic resonance imaging stereotactic localization requires a consistent commitment to quality assurance, daily evaluation of image accuracy, and recognition of potential sources for distortion [3,9–11]. This modality has proven to be an excellent and reliable neurodiagnostic imaging tool for stereotactic surgery with the Leksell system. MRI-guided stereotactic surgery has several clear advantages, including significant reduction in image artifacts (since the pins and frame give no MR signal). In addition, the ability to do multiplanar imaging, to verify target coordinates between both coronal and axial locations, and to visualize certain targets by variations in imaging techniques related to T1 and T2 signal or contrast agents, clearly establish MRI as superior to CT in most instances. During CT imaging, the head frame is fixed in a stereotactic adapter compatible with the various commercially available scanners [12]. The patient must be leveled in advance and must be screened to make sure that there are no significant imaging artifacts, which would reduce image quality. One of the advantages of the Leksell MRI compatible stereotactic system is that the fiducial system for coordinate calculation is close to the patient’s head and to the center of the magnetic field. As both magnetic susceptibility and other distortions appear greater at the periphery of the field, other systems with fiducials located far from the center of the field may be associated with a greater risk of distortion. Even with axial
Leksell stereotactic apparatus
imaging, warping of the fiducials can be seen when the target is very close to the base ring. Keeping the base ring low is generally advisable. The magnet itself must be properly maintained and checked. Residual distortions of the field left by hairpins, surgical clips, and so on can result in image degradation. Both scientific investigations and accumulated clinical experience with hundreds of thousands of patients worldwide have validated the role of MRI stereotactic surgery. Multiplanar direct and reformatted imaging facilitates simulation of probe pathways in advance of the procedure itself. These virtual reality techniques have been incorporated into advanced surgical planning systems.
Versatility of the System Since the semicircular arc in general describes a sphere where the surgical target is at the center, any probe approach, regardless of entry point, will reach the target. The main principle is that the trajectory should be safe and should minimize the number of pial, ependymal, and critical brain structures that the probe traverses. Convexity, vertex, full lateral, suboccipital, and transsphenoidal routes are readily possible using the system. With the model G system, the expanded Leksell stereotactic arc can be placed in both the conventional left-right and in the antero-posterior positions. The arc can be reversed for supraorbital approaches so that the coordinates as well as the arc and ring angles can be read easily. Patients undergoing posterior fossa biopsy procedures usually do so in the semi-sitting position. We normally recommend a transcerebellar approach to intra-axial lesions at the level of the middle cerebellar peduncle or transcortical coronal approaches to midline brainstem targets [13]. Twist drill craniotomy is used almost exclusively for most functional, diagnostic, and therapeutic stereotactic surgery. Such an approach embodies the concept of minimal invasiveness [14–24].
30
Accessories A tremendously diverse variety of instruments are available for diagnostic, therapeutic, and functional brain surgery. In addition to the twist-drill craniotomy set (> Figure 30-13), electrodes of various sizes and shapes [25], microelectrodes [26], and a wide variety of lesion-generating devices are available [27]. For therapeutic and diagnostic surgery, the Backlund biopsy set (> Figure 30-14), a forceps system and a Sedan-type needle aspiration biopsy system are available. It is important that the user review and understand each instrument in advance. The distance from the stop to the operative end of the probe, i.e., the working length, is always 190 mm. However, most biopsy probes extend some distance beyond the probe operative end and this must be taken in consideration when deciding on probe placement. Percutaneous small needles are available for craniopharyngioma aspiration or puncture [15– 17,28–30]. The Backlund hematoma evacuation kit can be used for percutaneous evacuation of deep intracerebral hematomas (> Figure 30-15). Special probes for aspiration of colloid cysts and brain abscesses are available [20,31]. Imaging . Figure 30-13 The percutaneous twist-drill craniotomy set for the majority of conventional open stereotactic procedures. Burr holes are rarely used
477
478
30
Leksell stereotactic apparatus
. Figure 30-14 The Backlund biopsy set includes both cyst puncture needles and the ‘‘spiral’’ biopsy system
. Figure 30-15 The Backlund hematoma evacuator includes an ‘‘Archimedes screw’’ to remove deep seated intracerebral hematomas
compatible probe holders are available and permit intraoperative imaging with a probe at the target position [32]. Accessories to the operating room table lock the frame to the standard Mayfield adapter.
Stereotactic Microsurgery The availability of new diagnostic imaging tools has promoted less invasive strategies to
reach both deep and superficial targets of the brain. The issues relative to localization of the target within the brain have largely been resolved by preoperative planning. The development of refined tools with reduced risk to patients has been exemplified by the combination of stereotactic technology and microsurgery. Deep targets in the brain can be reached through limited-exposure craniotomies. Microsurgical or endoscopic evacuation of colloid cysts, excision of cavernous malformations, and removal of brain tumors are among the examples of the usage of current stereotactic techniques. At the University of Pittsburgh, more than 2,500 brain operations have been done using intraoperative CT guidance. Currently we use a General Electric 64 slice CT scanner that is located in the operating room suite itself. Imaging is done immediately prior to the surgical intervention, occasionally during surgery itself, and always afterward in order to detect potential complications [33–36]. Since the risk for complications in most stereotactic procedures is less than 1%, the concept of microsurgical intervention coupled with highresolution imaging has opened the doors to a wide variety of successful surgical techniques. > Table 30-1 illustrates the experience using the Leksell Stereotactic system during a 28 year interval at the University of Pittsburgh Medical Center. > Table 30-2 shows the observed complications after frame based stereotactic procedures during the same time interval. During the past 15 years we have not had a patient with intraoperative bleeding requiring evacuation. We have also not seen a scalp or bone infection in the 15 year interval since we switched to using twist drill craniotomies. > Table 30-3 is a comparison between frame based and frameless image-guided biopsy techniques. Various other accessories are available for the Leksell stereotactic system. These include retractors, small trephine craniotomy systems, a laser guide (> Figure 30-16), endoscopic adapters
30
Leksell stereotactic apparatus
and probe holders for catheter arrays. These types of instruments have largely supported the concept that present and future neurosurgery will be almost exclusively image-guided. . Table 30-1 Image-Guided Frame-Based Stereotactic Procedures (UPMC-Presbyterian, 1979–2007) Diagnostic
No. of procedures
Brain tumor biopsy Infection Degenerative disease Hematoma Stroke Subtotal Therapeutic Cyst aspiration Abscess drainage Isotope implantation Movement disorder Lesion DBS* Pain DBS Epilepsy depth electrodes Miscellaneous Subtotal Total
1,584 58 6 12 4 1,664 197 97 145 219 148 24 95 62 987 2,651
The Leksell Gamma Knife The Leksell Gamma Knife was developed through the fertile collaboration between Borje Larson, a physicist and radiation biologist, and Lars Leksell in the 1960s. First working with cross-firing of proton beams, Larson and Leksell eventually decided to create a dedicated radiosurgical instrument which could be used with ease in the standard hospital environment. The initial 179 cobalt-60 source prototype Gamma Knife was used primarily to create radio necrotic lesions within the deep gray and white matter tracts of the brain. Primary indications were intractable obsessive compulsive and anxiety neuroses, chronic pain from cancer, and movement disorders such as Parkinson’s disease. The second-generation device had 179 beams but a spherical dose profile (Leksell Gamma Unit, model U). A dramatic increase in the development of radiosurgical technology resulted in the 1980s and 1990s, when the redesigned 201 source Models B and C Gamma Knives became available (> Figure 30-17). In 2007 the completely redesigned and completely
. Table 30-2 Complications Related to Frame Based Stereotactic procedures: UPMC-Presbyterian; 1979–2007 Procedure type
Total
Hemorrhagea
Diagnostic biopsy Cyst aspiration Radiation implant Brain abscess Catheter and cyst reservoir insertion Hematoma aspiration Frame based craniotomy Pallidotomy Thalamotomy Deep brain stimulation (movement disorders) Depth electrodes for seizures Deep brain stimulation(chronic pain) Cell transplantation Mesencephalotomy/Capsulotomy Total
1,664 197 145 97 19 9 10 147 72 148
43 (2.58%) 5 (2.53%) 2 (1.37%) 1 (1.03%) 2 (1.36%) 1 (1.38%) 1 (0.67%)
95 24 20 4 2,651
55 (2.07%)
a
Six patients (0.36%) required Craniotomy and hematoma evacuation Includes cerebritis, meningitis
b
Seizure
Infectionb
6 (0.36%) 3 (1.52%) 1 (0.68%) 1 (1.38%) 11 (0.41%)
2 (0.12%) 2 (1.01%) 1 (0.68%) 4 (4.12%)
Total complications
1 (0.67%)
51 (3.06%) 10 (5.07%) 4 (2.75%) 5 (5.15%) 0 2 (1.36%) 2 (2.76%) 2 (1.34%)
1 (1.05%) 11 (0.41%)
1 (1.05%) 77 (2.90%)
-
479
480
30
Leksell stereotactic apparatus
. Table 30-3 Comparison of Image-Guided Brain Biopsy Techniques
Image compatibility Image integration Fixation of head Precise trajectory pre-plotting Target calculation method Software calculation Cranial target access Anesthesia Accuracy Ease of performing craniotomy Hemorrhage risk requiring craniotomy Accessible targets Lobar Deep Neuroimaging/neuropathy correlation . Figure 30-16 The stereotactic hemiarc attached to the base ring can be used to guide a laser beam or an endoscope to a target
Frame based
Frameless
MRI, CT, PET, MEG Intraoperative 4 point (frame) Yes Frame fiducials Yes Twist drill Local 0.7–1 mm Low Low (<1%)
MRI, CT, PET, MEG Preoperative 3 point (Mayfield) No Scalp or mask fiducials Yes Burr hole General 1–3 mm High ?
Yes Yes Yes
Yes No Yes
worldwide, and more than 500,000 patients had undergone Gamma Knife surgery. The centers perform between 150 and 1,000 radiosurgical procedures per year. As an alternative to more invasive conventional brain surgery or skull-base surgery, radiosurgery has become an important part of the field of image-guided neurosurgery. The Gamma Knife itself has become virtually a ‘‘gold standard’’ against which the accuracy, precision, cost-effectiveness, and outcomes of both conventional surgery and other stereotactic radiosurgical techniques are compared.
Leksell SurgiPlan
robotic Gamma Knife Perfexion was introduced worldwide (> Figure 30-18). As of early 2008, close to 300 Leksell Gamma Knives were in use in more than 50 countries
SurgiPlan is an image-integrated surgical planning computer program that converts stereotactically acquired images (e.g., MRI, CT) to actual Leksell stereotactic frame space. After imaging is complete, the data sets are transferred by CD or Ethernet to the surgical workstation. Conventional angiographic films can be digitized into the planning system as well. SurgiPlan allows preoperative viewing of images either in original orientation or in any reformatted plane. Three-dimensional representations of the whole brain can be rotated and the surgical target reviewed from a variety of
Leksell stereotactic apparatus
30
. Figure 30-17 Left: the 201 cobalt-60 source Leksell model B Gamma Knife. Right: the outer collimator helmet slides into the inner collimator containing the cobalt sources. The redesign of the unit facilitate reloading of the cobalt sources between 5 and 10 years after initial source placement
directions. The surgical image interaction is virtually in real time, since the image handling by the computers is extremely rapid. SurgiPlan increases the accuracy of target recognition and allows preplotting of simulated probe trajectories (> Figure 30-19). Using SurgiPlan, an optimal trajectory can be pre-plotted prior to actual placement of the probe. These ‘‘virtual reality’’ probe trajectories can be varied in order to minimize the risks. SurgiPlan is also valuable during functional neurosurgical procedures, especially
with MRI-acquired data sets. With such information, the anterior and posterior commissures, inter-commissural plane and functional targets can be determined and visualized in multiple or even probe’s-eye-view ‘‘planes’’ (> Figure 30-20).
Summary In a prior publication, we posed eight questions that should be answered referable to any stereotactic
481
482
30
Leksell stereotactic apparatus
. Figure 30-18 Left: the completely robotic 192 cobalt-60 source Leksell Gamma Knife Perfexion. Right: the sources are fixed in eight independently movable sectors mounted on the collimator body, containing collimators of 4, 8, and 16 mm
system [37]. Although time has passed, these eight questions remain pertinent: (1) Is it a complete system?, (2) Is it simple?, (3) Is it dependable?, (4) Is it versatile?, (5) Is it accurate?, (6) Is it compatible with multiple current imaging modalities?, (7) Is it both computer-compatible and independent?, and (8) Is its development keeping pace with developments in other related technologies?
The evolutionary Leksell stereotactic system provides affirmative answers to all these questions.
Acknowledgment The authors are indebted to Elekta Instrument AB for help in preparation of the figures.
Leksell stereotactic apparatus
30
. Figure 30-19 SurgiPlan is a versatile surgical planning system that determines the X, Y, Z coordinates and preplots probe trajectories
. Figure 30-20 Functional targets can be localized with Leksell SurgiPlan using AC-PC based formulas which generate the corresponding stereotactic coordinates
483
484
30
Leksell stereotactic apparatus
References 1. Spiegel EA, Wycis HT, Marks M, Lee ASJ. Stereotaxic apparatus for operations on the human brain. Science 1947;106:349-50. 2. Leksell L. A stereotaxic apparatus for intracerebral surgery. Acta Chir Scand 1949;99:229-33. 3. Kondziolka D, Lunsford LD, Kanal E, et al. Stereotactic magnetic resonance angiography for targeting in AVM radiosurgery. Neurosurgery 1994;35:585-91. 4. Leksell L. Stereotaxis and radiosurgery: an operative system. Springfield, IL: Charles C. Thomas; 1971. 5. Leksell L, Jernberg B. Stereotaxis and tomography: a technical note. Acta Neurochir (Wien) 1980;52:1-7. 6. Leksell L, Leksell D, Schwebel J. Stereotaxis and nuclear magnetic resonance. J Neurol Neurosurg Psychiatry 1985;48:14-18. 7. Leksell L, Lindquist C, Adler JR, et al. A new fixation device for the Leksell stereotactic system. J Neurosurg 1987;66:626-9. 8. Peters TM, Clark JA, Pike GB, et al. Stereotactic neurosurgery planning on a personal-computer-based workstation. J Digital Imaging 1989;2:75-81. 9. Kondziolka D, Dempsey PK, Lunsford LD, et al. A comparison between magnetic resonance imaging and computed tomography for stereotactic coordinate determination. Neurosurgery 1992;30:402-7. 10. Lunsford LD. MRI stereotactic thalamotomy: report of a case with comparison to CT. Neurosurgery 1988;23:363-7. 11. Lunsford LD, Martinez AJ, Latchaw RE. Stereotaxic surgery with a magnetic resonance and computerized tomography compatible system. J Neurosurg 1986; 64:872-8. 12. Latchaw RE, Lunsford LD, Kennedy WH. Reformatted imaging to define the intercommissural line for CT-guided stereotactic functional neurosurgery. Am J Neuroradiol 1985;6:429-33. 13. Kondziolka D, Lunsford LD. Stereotactic biopsy for intrinsic lesions of the medulla through the long axis of the brainstem. Acta Neurochir 1994;129:89-91. 14. Backlund EO. A new instrument for stereotaxic brain tumor biopsy. Acta Chir Scand 1971;137:825-7. 15. Backlund EO. Stereotactic treatment of craniopharyngiomas: a 15 year material. In: Proceedings of the 32nd annual meeting of the scandinavian neurosurgical society, Linkoping, Sweden, 1979. 16. Backlund EO. Stereotaxic evacuation of hematomas (letter). J Neurosurg 1985;62:460-1. 17. Backlund EO. Studies on craniopharyngiomas: III. Stereotactic treatment with intracysticyttrium-90. Acta Chir Scand 1973;139:237-47. 18. Duma CM, Kondziolka D, Lunsford LD. Image-guided stereotactic management of non-AIDS related cerebral infection. Neurosurg Clin N Am 1992;3:291-302.
19. Engle D, Lunsford LD. Brain tumor resection guided by intraoperative computed tomography. J Oncol 1987; l4:361-70. 20. Hall WA, Lunsford LD. Changing concepts in the treatment of colloid cysts. An 11-year experience in the CT era. J Neurosurg 1987;66:186-91. 21. Lunsford LD, Coffey RJ, Cojocaru T, Leksell D. Image guided stereotactic surgery: a ten year evolutionary experience. Stereotact Funct Neurosurg 1990;54–55: 375-86. 22. Lunsford LD, Deutsch M, Yoder V. Stereotactic interstitial brachytherapy – current concepts and concerns in twenty patients. Appl Neurophysiol 1985;48:117-20. 23. Lunsford LD, Gumerman LW, Levine G. Stereotactic intracavitary irradiation of cystic neoplasms of the brain. Appl Neurophysiol 1985;48:146-50. 24. Lunsford LD, Somaza S, Kondziolka D, Flickinger JC. Brain astrocytomas: biopsy then irradiate. Clin Neurosurg 1995;42:464-79. 25. Roberts DW. Epilepsy: Deep brain electrodes. In: Lunsford LD, editor. Modern stereotactic neurosurgery. Boston: Martinus Nijhoff; 1988. p. 413-23. 26. Ohye C. Selective thalamotomy for movement disorders. Microrecording stimulation techniques and results. In: Lunsford LD, editor. Modern stereotactic neurosurgery. Boston: Martinus Nijhoff; 1988. p. 315-40. 27. Svennilson E, Torvik A, Lowe R, Leksell L. Treatment of parkinsonism by stereotaxic thermolesions in the pallidal region: a clinical evaluation of 81 cases. Acta Psychiatry Neurol Scand 1960;35:358-77. 28. Lunsford LD. Stereotactic treatment of craniopharyngioma: intracavitary irradiation and radiosurgery. Contemp Neurosurg 1989;11:1-6. 29. Lunsford LD, Pollock BE, Kondziolka DS, et al. Stereotactic options in the management of craniopharyngioma. Pediatr Neurosurg 1994;21:90-7. 30. Pollack IF, Lunsford LD, Slamovitz T, et al. Stereotaxic intracavitary irradiation for cystic cranipharyngiomas. J Neurosurg 1988;68:227-33. 31. Lunsford LD, Nelson PB, Rosenbaum AB. Stereotactic aspiration of a brain abscess using the therapeutic CT scanner: case report. Acta Neuro Chir (Wien) 1982;62:25-9. 32. Lunsford LD, Leksell L, Jernberg B. Probe holder for stereotactic surgery in the CT scanner: a technical note. Acta Neurochir 1983;69:297-304. 33. Lunsford LD. A dedicated CT system for the stereotactic operating room. Appl Neurophysiol 1982;45:374-8. 34. Lunsford LD. Advanced intraoperative imaging for stereotaxis: the surgical CT scanner. Acta Neurochir (Wien) 1984;33:573-5. 35. Lunsford LD, Parrish R, Albright L. Intraoperative imaging with a therapeutic CT scanner: technical note. Neurosurgery 1984;15:559-61.
Leksell stereotactic apparatus
36. Lunsford LD, Rosenbaum AE, Perry JP. Stereotactic surgery using the ‘‘therapeutic’’ CT scanner. Surg Neurol 1982;18:116-22. 37. Lunsford LD, Leksell D. The Leksell stereotactic system In: Lunsford LD, editor. Modern stereotactic neurosurgery. Boston: Martinus Nijhoff; 1988. p. 27-46. 38. Kondziolka D, Lunsford LD. Guided neurosurgery using the ISG viewing wand. Contemp Neurosurg 1995;17:1-6.
30
485
34 Miniframe Stereotactic Apparatus M. A. Madera* . W. D. Tobler*
Introduction As stereotactic localization is increasingly used in cranial neurosurgery, its applications now include biopsy of mass lesions, aspiration of hematomas/cystic lesions, catheter placement, and planning small, strategically placed craniotomy flaps. Another important, but less often used, stereotactic application includes placement of electrodes for stimulation of deep structures. A more rarely used application is lesion creation in the treatment of movement disorders. Options for stereotactic targeting have increased in recent years with the introduction of image-guided systems, commonly known as frameless stereotaxis. There has been a rapid evolution from framebased, mechanical stereotactic devices to imageguided stereotactic systems. Although large, frame-based devices may still be preferred in some centers, image guidance has overall displaced mechanically-based systems in popularity. Our review of the evolution of miniframe stereotactic devices begins with the earliest freehand stereotactic-guided procedures in which small devices attached to the skull served as probe holders. During the frame-based stereotactic era, miniframe stereotactic devices were largely overlooked as large frame-based devices were promoted as the gold standard. With the introduction of image-guided techniques, the smallest miniframe devices have become important tools for stereotactic surgery.
*The authors have no financial relationships to any of the companies mentioned. #
Springer-Verlag Berlin/Heidelberg 2009
Becoming increasingly popular, miniframe devices not only generate a stereotactic trajectory but also provide the rigidity requirements of stereotactic procedures. The concept of a small, ball-and-socket device fixed to the skull through a burr hole for stereotactic targeting was one of the earliest stereotactic devices. The trajectory was generated by rotating the ball and its axis through the lesion, which was chosen on standard X-rays, and then by passing the probe to the calculated target depth. With the wide proliferation and availability of image-guided systems, the use of the ball-andsocket device seemed to provide a simple solution for firm fixation and trajectory generation for stereotactic surgery. This type of miniframe device is perhaps the most intuitive, compact, yet simple stereotactic device available for even the most complex stereotactic procedures. The history of technological evolution and increasing supportive literature points to an emerging era of image-guided surgery using miniframe-like devices for frameless stereotaxis.
Burr-Hole Mounted Ball-andSocket Probe Holders In 1956, Austin and Lee presented their three-piece, plastic, burr-hole-mounted, stereotactic device (> Figure 34-1) that was threaded into a trephine craniotomy [1]. Adjustments were made using plain X-rays to create a desired trajectory to a chosen entry point. Once the trajectory was determined, the probe was passed to the target. There are no additional publications of experience with
522
34
Miniframe stereotactic apparatus
. Figure 34-1 Schematic drawing of the Austin and Lee three-piece ball-and-socket stereotaxic instrument. The device is threaded and tapped into a trephined hole in the skull. The plastic ball-and-socket joint is mounted within the shell and a needle is passed to the desired depth [1]
this technique, which leads to the conclusion that this device never became widely used. However, there has been a series of reports, published from 1955 to 1975, of similar devices for stereotactic trajectory generation fixed to the skull [2–5]. Most of theses authors used ventriculography for target localization; applications included lesion creation for Parkinson’s disease and cryosurgery for lesion creation and tumor treatment. In a 1982 study on a series of 26 CT-guided biopsies, Greenblatt et al. reported the use of a skull-mounted, ball-and-socket holding device for one biopsy [6]. This device was useful for multidirectional aiming of a biopsy needle, especially for superficial lesions, and thus prevented needle wobbling or dislodging during the procedure. In 1983, a skull-mounted ball-and-socket device was used as a needle holder for CT-guided biopsy in ten cases [7]. When compared with the contemporary frames, it was described as a less
cumbersome stereotactic device but criticized for its lack of true stereotactic functionality. The trajectory was chosen based on a CTscan obtained before needle insertion into the brain while the biopsy probe remained in the ball-and-socket holder. The trajectory, aimed at the lesion, was adjusted on subsequent scans until it was advanced into the target. In contrast with stereotactic frames in which the procedure was simulated in a threedimensional (3D) phantom-coordinate system derived from the CT dataset, the ball-and-socket device served as a holder for the probe, which was then advanced by hand. Although the ball-andsocket device was deemed better than using a freehand biopsy technique, it fell short of recognition as a stereotactic device. In a series of reports in the early 1980s, Patil described a succession of devices that were smaller and less complex than traditional frames. Patil placed the Z coordinate of the lesion in line with the head clamp, or in later versions, with the larger Patil frame itself. The earliest version was a head clamp that was fixed to the patient in line with the laser beam of the CT scanner; since the head clamp and probe holder were in the same plane, one CT image showing these and the lesion allowed planning of the trajectory without lengthy calculations [8]. Although it was deemed useful for large lesions, its accuracy was criticized [9]. Subsequent versions of the device improved upon the degree of freedom of the probe holder at the expense of somewhat more cumbersome mechanics. With this device and thinner CT slices, Patil achieved greater accuracy and reported advantages over other frames including less artifact and simpler calculations for trajectory planning [10,11].
First Burr-Hole Mounted (miniframe) Mechanical Stereotactic Device The Pelorus system (Schaerer Mayfield USA, Cincinnati, Ohio) is a skull-mounted, true
Miniframe stereotactic apparatus
stereotactic device conceived as an early alternative to traditional frame-based devices. Like traditional frame-based stereotactic devices, translation of the target coordinates from CT scan to a biopsy frame space was achieved mechanically on a special phantom frame and did not require any proprietary computer software (> Figure 34-2). The Pelorus system was based in part on the ball-and-socket device introduced by Austin and Lee in 1956 [1]. In the phantom frame, the trajectory through the balland-socket device was determined by pointing and advancing the probe to the target, which was set in the frame by a simple arithmetic calculation. The depth to the target was measured directly in the phantom. Unlike the large framebased devices, the small ball-and-socket was attached with a special fitting to a ring, measuring less than 4 cm in diameter; this ring was fixed to
. Figure 34-2 The Pelorus stereotactic system components: double transfer plate (a), reference ball and socket (b), phantom frame with x, y and z coordinates (c), target point (d), arc carrier post (e), arc (f) and adjustable biopsy ring (g) (printed with permission from Mayfield Clinic)
34
the skull with self-securing, bi-cortical screws that were placed with the patient under local anesthesia (> Figure 34-3a). When first reported on by Carol et al. in 1985, advantages of the Pelorus stereotactic system were unobstructed trajectories to the temporal lobe or posterior fossa, greater ease of application, and better patient tolerance [12]. In addition to the ball-and-socket burr-hole mounted device, an arc adaptor could be attached to the skull mount to permit center-ofarc targeting. Such targeting allowed for multiple entry points that could be changed at the time of surgery and was especially useful for stereotactic craniotomies. In our experience with the Pelorus stereotactic system at the Mayfield Clinic/University of Cincinnati Department of Neurosurgery between 1989 and 1995, the device was used in more than 400 stereotactic procedures for stereotactic biopsy, hematoma aspiration, placement of depth electrodes, and numerous stereotacticguided craniotomies (> Figure 34-3b). Our experience confirmed the Pelorus stereotactic system offered advantages over the frame-based systems, such as simplicity of application and ease of accessibility to cranial targets, especially in the temporal lobe and posterior fossa since no frame impeded the trajectory. As an alternative to frame-based systems that have limited working space and trajectory options, we reported our experience with the Pelorus apparatus with the arc adapter for implantation of depth electrodes for seizure monitoring [13]. In confirming target accuracy during surgery or postoperative CT evaluation for 96 stereotactic biopsy and/or aspiration procedures, we obtained nondiagnostic tissue (indicating a possible targeting error) in three patients; that is, diagnostic rate was 96.9% [14]. While the Pelorus stereotactic system never gained the popularity or mainstream acceptance of the large frame-based stereotactic devices, it served our stereotactic requirements in a most satisfactory manner.
523
524
34
Miniframe stereotactic apparatus
. Figure 34-3 Illustration of the Pelorus stereotactic system. (a) Single-transfer plate is attached to the patient’s head with bicortical screws. (b) Ball-and-socket is mounted to the transfer plate with the arc carrier post locked at the exact coordinates. The surgeon can swivel the arc adapter along the post to determine the optimal trajectory and entry point to the targeted lesion (printed with permission from Mayfield Clinic)
Miniframe Devices in the Image-Guided Era Image-Guide Surgery In the mid-1990s, image-guided surgery was introduced into the operating room. Computing power enabled the creation of 3D image sets of the brain by using MRI or CT that could be viewed and manipulated in all three orthogonal planes, then allowing surgeons to navigate on these images. In contrast with the mechanical phantoms of the frame-based systems, these image-guided computers created a virtual phantom; the surgical workstation used a 3D phantom for simulation of the stereotactic procedure using the patient’s own anatomy. Using digitizing
instruments, surgeons could match the fiducial markers placed on the patient with the same markers identified on the 3D image dataset (CT or MRI). Specifically, the surgeon could move a pointer around the outside of the registered target on the patient’s skull and compare that with the position of the pointer tip on the workstation image of the patient’s brain. This new surgical technique, quickly termed frameless stereotaxis, enabled the surgeon to navigate in the operating room during the selection of a precise entry point for a small stereotactic craniotomy flap. The first digitizers were attached to mechanical arms that had spatial sensors embedded in their joints. Although these devices were effective, they were cumbersome and were soon replaced by active optical digitizers, which use infra-red
Miniframe stereotactic apparatus
light to track LEDs placed on the surgical probes. Later replacement of the LEDs with reflective spheres permitted passive optical tracking, thereby eliminating the power cord to the probe so as to fashion a wireless navigational tool. Electromagnetic digitizers were developed to help eliminate the optical interference that can be problematic with optical digitizers, but they are not yet widely used [15]. These frameless stereotactic systems were first used to navigate outside the patient’s skull. Soon after their introduction, true stereotactic procedures using image-guided techniques were developed. The digitizing probe was easily positioned on the surface of the skull, an entry point was selected, and the computer then created a virtual trajectory to an intracranial target, including the depth calculated to that target. Using a fixed ball-and-socket device (miniframe apparatus) to generate a trajectory and to firmly hold the probe was an intuitive solution for live navigation on a 3D image of the brain for true stereotactic procedures.
Miniframe Frameless Stereotaxis At the University of Cincinnati, we modified the Pelorus ball-and-socket device so that it could be used with an optical tracking system. For a frameless stereotactic procedure, the Pelorus ring would be positioned over the entry point chosen during the actual navigation. Then the ring would be fixed to the skull at this location with self-securing screws. Using a simple computer program and an adaptor to hold the optical probe, we could place the probe in the balland-socket device. While navigating live and moving the probe in the ball-and-socket device, we then aligned the virtual trajectory from the tip of the probe to intersect the target. Fine adjustments could be made and the ball would be locked in place. The computer then would calculate the depth from the probe tip to the target. After opening the scalp and drilling a
34
small hole, the biopsy probe was passed to the target and the procedure performed. Modification of the Pelorus system to create a noninvasive stereotactic option resulted in a more versatile design called the AccuPoint (Schaerer Mayfield USA, Cincinnati, Ohio) ball-and-socket device. It was fitted with an adaptor that could be attached to the Mayfield Headrest System or the Budde Halo retractor system (Integra LifeSciences Corporation, Plainsboro, New Jersey). The AccuPoint device could be moved around the outside of the skull and positioned over the entry point, immediately against or close to the scalp, without placing screws into the scalp [16] (> Figure 34-4). Our cumulative experience with the Mayfield ACCISS (Schaerer Mayfield USA, Cincinnati, Ohio) image guided-system was reported for 300 cranial stereotactic procedures [16].
Stereotactic Biopsy with Frameless Stereotaxis Adaptation of the AccuPoint sphere for stereotactic procedures using image-guided surgery raised the question, ‘‘Could image-guided, frameless stereotactic techniques replace frame-based stereotactic procedures with the safety and accuracy profiles established for frame-based stereotaxis?’’ In our 2000 report of 79 patients who underwent frameless stereotactic biopsy using these techniques, three (4.4%) patients had biopsies that were nondiagnostic or a 95.6% diagnostic rate [17]. Our findings demonstrated nondiagnostic rates equivalent to those reported for frame-based systems and were confirmed in the ensuing years from other centers. Multiple other series validating the efficacy of stereotactic surgery have appeared recently in the literature. In 2001, Paleologos et al. reported a 97.6% diagnostic rate in 125 cases using a similar frameless technique [18]. Woodworth et al. reported similar a 89% diagnostic yield that was comparable for both frameless (110 cases) and frame-based (160 cases) techniques [19]; the SNN-Olivier FreeGuide (Philips Medical Systems,
525
526
34
Miniframe stereotactic apparatus
. Figure 34-4 Mayfield ACCISS stereotactic system. (a) AccuPoint Targeting Sphere attached to the Budde Halo Retractor System by a mechanical arm. (b) With rotation of the mechanical arm, the surgeon can view the trajectory and target displayed on the computer workstation. (c) Once the optimal trajectory is determined and locked, the mechanical arm is removed and a biopsy needle can be passed to the target through the sphere (printed with permission from Mayfield Clinic)
Best, The Netherlands), used in the frameless cases, is essentially a probe holder (not a ball-and-socket device) that attaches to the Mayfield head holder and secures to the calvarium with a metal pin. In a retrospective review of their 10-year experience with stereotactic biopsies in 391 patients, Dammers et al. reported a combined diagnostic rate of 89.4%, with no differences between 227 frame-based and 164 frameless procedures [19]; their experience mirrored the findings of Woodworth et al. [20]. Interestingly, in a 9-year retrospective review of
213 consecutive stereotactic biopsies, Smith et al. reported equivalent diagnostic rates and complications between 139 frame-based and 74 frameless procedures [21]. At their institution, hospital stays were longer for the patients who underwent frameless procedures, which were performed under general anesthesia, than for patients who underwent frame-based procedures, which were mostly under local anesthesia. Noting the cost differential between the two groups, the authors recommended that frame-based techniques should be the first-line
Miniframe stereotactic apparatus
consideration in stereotactic biopsy. In contrast, in their comparative study of 76 frameless and 79 frame-based biopsy procedures, Dorward et al. reported shorter operative times and significantly fewer complications in the frameless group [22]. They opined that superior imaging, better target visualization, and improved flexibility of frameless techniques translate into tangible benefits. Frameless stereotaxic biopsy seems to have come of age. The combination of the computerbased, image-guided, frameless technique with mainframe devices is now a more intuitive, user-friendly method than mechanical stereotactic frames. Our experience and mounting clinical evidence reported by others supports the safety and efficacy of frameless stereotaxic biopsy.
34
. Figure 34-5 Schematic drawing of a MRI-compatible SnapperStereoguide device. The base of the device is inserted into a 15-mm burr hole and secured with the swivel (a). Fixation wheel (b) secures the instrument guide, with a moveable angle of 35 degrees being provided by a spherical joint (c). Optical tracking system (d) is coupled to the instrument guide. Biopsy cannula (e), composed of carbon fiber-reinforced composite, is inserted into the instrument guide (with permission from [23])
Miniframe Devices with Intraoperative Imaging Intraoperative imaging is a solution for stereotactic updating in real time. Image-guided systems rely on historical data, that is, images acquired before entering the operating room. Innovation in MRI technologies now provide a sophisticated way to acquire updated information during the course of the surgical procedure. This updated anatomical information can then be downloaded into the image-guided system. To accomplish this, Bernays et al. devised the Snapper-Stereoguide, a multi-component assembly composed of an MRI-compatible plastic [23]. After its insertion into a burr hole, LEDs for image-guidance are attached to the assembly for intraoperative scans (> Figures 34-5 and > 34-6). Their 2000 report noted no adverse outcomes in 20 patients. In the authors’ experience, this system combined the advantages of intraoperative imaging with a small, MRI-compatible mounted system to improve operative time and maintain accuracy. A less expensive, less complex alternative to intraoperative MRI is CT whose advantages are no need for room shielding, lower cost, and less
. Figure 34-6 Photograph of the Snapper-Stereoguide (b) coupled to the optical tracking system (a) (with permission from [23])
527
528
34
Miniframe stereotactic apparatus
logistical difficulty. At Mayfield Clinic, we have intraoperatively used the MobileSCAN CT (distributed by Schaerer Mayfield USA, Cincinnati, Ohio). Since the CT scanner is in the operating room, the patient, surgeon, anesthesiologist, staff, and hospital benefit from the efficiency of being able to immediately obtain images for verification of biopsy site, hematoma
aspiration, or tumor resection (> Figure 34-7). In our experience, accuracy has been excellent, and information gained from intraoperative scanning has helped guide the progress of the operation (‘‘Use of Mobile Intraoperative Computed Tomography Scanner for Intracranial and Spinal Procedures,’’ poster presentation, American Association of Neurological Surgeons, 2003).
. Figure 34-7 Intraoperative CT (iCT) in the operating room. (a) MobileSCAN iCT and image-guided system. (b) Position of the MobileScan CT for a biopsy. (c) Preoperative MRI of a 12-mm thalamic glioma. Six contiguous slices from an intraoperative scan (d) from the MobileSCAN CT shows the biopsy cannula in position. Contrast was not given for this biopsy (printed with permission from Mayfield Clinic)
Miniframe stereotactic apparatus
Other Miniframe Devices The Navigus trajectory guide (Medtronic, Minneapolis, MN) is currently widely used for frameless stereotactic applications. Initially designed for use in stereotactic biopsy, the guide rigidly attaches directly to the skull and has attachments for different image-guided systems (> Figure 34-8). The Navigus is also conceptually identical to the Austin-Lee device. There are internal (fixed inside a 14-mm burr hole) and external (fixed to the surface of the skull) versions of the skullmounted part of the trajectory guide. Guides may be used for biopsy, shunt placement, endoscope insertion, or functional stimulation procedures. Quinones-Hinjosa et al. validated
34
the accuracy of this device for stereotoactic targeting [24]. The Nexframe (Medtronic, Minneapolis, MN) is a skull-mounted miniframe apparatus for accurate stereotactic targeting for functional stereotactic procedures (> Figure 34-9). In an important validation study, Henderson et al. demonstrated equivalent accuracy with the Nexframe device to the published results of accuracy for stereotactic frames in the laboratory setting [25]. Holloway et al. further showed equivalent accuracy to frame-based stereotaxis for functional procedures using the Nexframe device in a multicenter evaluation of 38 patients who underwent deep brain stimulation [26]. These devices are commercially available.
. Figure 34-8 Coronal (a) and along-the-probe (b) views during a biopsy procedure are shown using the Navigus trajectory guide (c) with inserted stereotactic probe (d) (reprinted with permission of Medtronic, Inc. ß 2008)
529
530
34
Miniframe stereotactic apparatus
. Figure 34-9 Photo of the Nexframe stereotactic device (a), which is attached to the skull with the rigidly attached reference device (b) and microelectrode driver (c) (Nexdrive, Medtronic, Minneapolis, MN) (reprinted with permission of Medtronic, Inc. ß 2008)
Summary The history of stereotactic surgery is rich with technology and innovation and continues to evolve at a rapid pace. Beginning with those early, large, and often awkward looking devices for stereotactic localization, frames have always occupied center stage. From the outset, however, the ball-and-socket trajectory device, the original precursor miniframe apparatus, has been an enduring, though not mainstream, concept. These various, small devices that attach to the skull are now finding increasing popularity in the area of image-guided surgery and are likely to permanently displace stereotactic frames.
References
Other interesting and custom solutions for the challenges of functional neurosurgery have been reported. Eljamel et al. described the use of a small polyethylene cube containing five small parallel channels implanted into a burr hole for micro-electrode placement using frameless stereotaxis [27]. Winkler et al. reported their experience with this polyethylene cube in one patient who underwent deep brain stimulation [28]. Fiducial-like anchors were implanted into the skull before image acquisition. A custommade Micro Targeting platform is created for the patient to establish an individualized trajectory based on image-fusion data from CT, MRI, and target selection. In surgery, the platform is attached to the patient, and the stereotactic procedure is performed using the preplanned trajectory built into this custom made, one-time use stereotactic device.
1. Austin GM, Lee AS, Grant FC. A new type of locally applied stereotaxic instrument. J Am Med Assoc 1956;161:147-8. 2. Cooper IS. Chemopallidectomy: an investigative technique in geriatric parkinsonians. Science 1955;121: 217-18. 3. McCaul IR. A method for the localization and production of discreet destructive lesions in brain. J Neurol Neurosurg Psychiatr 1959;22:109-12. 4. Rand RW. A stereotaxic instrument for pallidothalamectomy in Parkinson’s disease. J Neurosurg 1961;18:258-60. 5. Kandel EI. New stereotactic apparatus and cryogenic device for stereotactic surgery. Confin Neurol 1975; 37:128-32. 6. Greenblatt SH, Rayport M, Savolaine ER, et al. Computed tomography-guided intracranial biopsy and cyst aspiration. Neurosurgery 1982;11:589-98. 7. Levy WJ. Simple plastic stereotactic unit for use in the computed tomographic scanner. Neurosurgery 1983; 13:182-5. 8. Patil AA. Computed tomography stereotactic head clamp. Acta Neurochirurgica 1982;60:125-9. 9. Patil AA. Computed tomography-oriented stereotactic system. Neurosurgery 1982;10:370-4 (comment). 10. Patil AA. Computed tomography (CT) oriented rotary stereotactic system. Acta Neurochirurgica 1983;68:19-26. 11. Patil AA. Computed tomography plane of the target approach in computed tomographic stereotaxis. Neurosurgery 1984;15:410-4. 12. Carol M. A true ‘‘advanced imaging assisted’’ skullmounted stereotactic system. Appl Neurophysiol 1985;48:69-72.
Miniframe stereotactic apparatus
13. Yeh HS, Taha JM, Tobler WD. Implantation of intracerebral depth electrodes for monitoring seizures using the Pelorus stereotactic system guided by magnetic resonance imaging. Technical note. J Neurosurg 1993;78:138-41. 14. Tobler WD. The Pelorus apparatus. In: Gildenberg, editors. Textbook of stereotactic and functional neurosurgery. New York: McGraw-Hill, Health Professions; 1998. p. 119–26. 15. Mascott CR. Comparison of a magnetic tracking and optical tracking by simultaneouos use of two independent framelesss stereotactic systems. Neurosurgery 2005;57 (4 suppl):295-301. 16. Tobler WD. Surgical Navigation with the OMI System. In: Schulder M, Gandhi CD, editors. Handbook of stereotactic and functional neurosurgery. New York: M. Dekker; 2003. p. 91-101. 17. Tobler WD. To demonstrate that image-guided frameless stereotactic biopsy can be performed with precision equivalent to traditional frame-based stereotaxy. Presented at the XIV Congress: European society for stereotactic and functional neurosurgery, London; 25–27 October 2000. 18. Paleologos TS, Dorward NL, Wadley JP, et al. Clinical validation of true frameless stereotactic biopsy: analysis of the first 125 cases. Neurosurgery 2001;49:830-8. 19. Dammers R, Haitsma IK, Schouten JW, et al. Safety and efficacy of frameless and frame-based intracranial biopsy techniques. Acta Neurochir (Wien) 2005;150:23-9. 20. Woodworth GF, McGirt MJ, Samdani A, et al. Frameless image-guided stereotactic brain biopsy procedure: diagnostic yield, surgical morbidity, and comparison
21.
22.
23.
24.
25.
26.
27.
28.
34
with the frame-based technique. J Neurosurg 2006;104: 233-7. Smith JS, Quinones-Hinojosa A, Barbaro NM, et al. Frame-based stereotactic biopsy remains an important diagnostic tool with distinct advantages over frameless stereotactic biopsy. J Neurooncol 2005;73:173-9. Dorward NL, Paleologos TS, Alberti O, et al. The advantages of frameless stereotactic biopsy over frame-based biopsy. Br J Neurosurg 2002;16:110-18. Bernays RL, Kollias SS, Khan N, Romanowski B, Yonekawa Y. A new artifact-free device for frameless magnetic-resonance imaging-guided stereotactic procedures. Neurosurgery 2006;46(1):112-7. Quinones-Hinojosa A, Ware ML, Sanai N, et al. Assessment of image guided accuracy in a skull model: comparison of frameless stereotaxy techniques vs. frame-based localization. J Neurooncol 2006;76:65-70. Henderson JM, Holloway KL, Gaede SE, et al. The application accuracy of a skull-mounted trajectory guide system for image-guided functional neurosurgery. Comput Aided Surg 2004;9:155-60. Holloway KL, Gaede SE, Starr PA, et al. Frameless stereotaxy using bone fiducial markers for deep brain stimulation. J Neurosurg 2005;103:404-13. Eljamel MS, Tulley M, Spillane K. A simple stereotactic method for frameless deep brain stimulation. Stereotact Funct Neurosurg 2007;85:6-10. Winkler D, Strauss G, Helm J, et al. MicroTargeting1 platform: An individual stereotaxic device in functional neurosurgery. Int J Comput Assist Radiol Surg 2007;1: 295-9.
531
Stereotactic Targeting
25 Printed Stereotactic Atlases, Review R. J. Coffey
Introduction There is something so satisfying and practical about having a book that one can carry from place to place, open to and flip back and forth between whatever pages one pleases, that it is hard to imagine a world without stereotactic atlases in print format. The continued use of printed stereotactic atlases to guide functional neurosurgical operations still represents the legacy of nineteenth century neurological localizationalists and their phrenolological forebears – after more than 100 years of advances in neuroimaging and electrophysiological technology. Even apparently simple neurological processes involve the interaction of integrated functional systems that have nuclei in different geographical parts of the central nervous system, and that are connected by complex pathways. Although the concept of discrete anatomic ‘‘centers’’ for specific functions or behaviors is now outmoded, several well-defined intracerebral targets have retained their therapeutic utility since the advent of modern human stereotaxis in the mid-twentieth century. Unlike the pioneers of stereotactic surgery, who depended on the capricious appearance of normally calcified midline landmarks (pineal gland or habenular commissure) to navigate the brain, and unlike their immediate successors, who depended on the positive or negative shadows cast by air- or contrast-filled ventricles on x-ray film, contemporary neurosurgeons work directly from computed tomographic (CT) scans and magnetic resonance images (MRI) of the brain itself. Some – although the list is diminishing as imaging technology improves – important functional stereotactic #
Springer-Verlag Berlin/Heidelberg 2009
targets are still indistinguishable from surrounding structures on CTor MRI; they remain invisible, or at least well camouflaged. Functional neurosurgeons solve this dilemma by referring to one of the excellent contemporary or historic stereotactic atlases or to a computerized atlas – the subject of Chapters 26 and 27. Experience with stereotactic atlases conjures up images of ponderous and expensive folio-sized volumes of high-quality photographic plates of unstained whole brain sections or, more often, magnified and stained sections of the thalamus, basal ganglia, and upper brain stem. Useful stereotactic atlas sections are cut at regular intervals in two or more planes. Of necessity, more than one brain is required if sagittal, frontal, and horizontal photographic sections are presented; over 100 brains were used to produce some atlases. Each section or photographic plate has a twodimensional scale in the margins, on a clear overlay, or on an accompanying line drawing. The numerical coordinate of the section plane and the two-dimensional coordinates of the target structure within the plane determine the set of three-dimensional coordinates required to reach the target with a stereotactic instrument. The fact that dimensions of individual brains vary from each other and from the idealized or standard brains depicted in atlas sections raises a contradiction for functional stereotactic atlases: For all the effort and expense required to produce a brain atlas, the results are not perfectly accurate (within less than a few millimeters) for any particular patient. In most instances, a purely atlasguided operation is not sufficiently accurate to justify the placement of a permanent therapeutic
348
25
Printed stereotactic atlases, review
lesion or stimulating electrode. Modern MR imaging of the individual patient comes to the rescue in most contemporary cases. However, the neurosurgeon still must undertake the sometimes tedious task of obtaining physiological corroboration of the probe’s position. This usually involves the use of intraoperative electrical stimulation, evoked potential recording, or other electroanatomical techniques on an awake, cooperative patient. The correlation of intraoperative electroanatomic phenomena with a stereotactic atlas, in light of the surgeon’s knowledge and the published results of previous surgeons’ experiences, is the essence of functional stereotaxis. Based on intraoperative observations during each probe trajectory, the surgeon must decide in which direction and by what distance to adjust the target for each subsequent trajectory. In this manner, the stereotactic atlas and a compendium of prior electroanatomic observations (memorized by the surgeon) are complementary tools that guide the successful performance of the surgical procedure. With few exceptions, judging from their intraoperative responses, most patients appear to have studied a stereotactic atlas as well.
A Chronology of Selected Stereotactic Atlases in Print Format Historic and Out-of-print Stereotactic Atlases 1952, 1962: Spiegel and Wycis Stereoencephalotomy (thalamotomy and related procedures), Part I: Methods and stereotaxic atlas of the human brain (Spiegel and Wycis, 1952)
In the spring of 1947 E.A. Spiegel, a clinical and research neurologist who had considerable experience in experimental animal stereotaxis, and H.T. Wycis, a neurosurgeon who had worked in Spiegel’s laboratory at Temple University in Philadelphia, performed the first modern stereotactic operation on a human patient; they had operated more
than 100 additional patients by 1952. By that time Spiegel and Wycis realized that functional stereotactic surgery ‘‘requires an exact preoperative calculation of the electrode position, and such a calculation depends on two conditions: (1) determination of a reference point by means of an X-ray picture taken under definite standard conditions, and (2) an exact knowledge of the position of the area to be destroyed in relation to the reference point. Thus . . . a stereotaxic atlas of the human brain is presented’’ [1]. Their landmark publication of the first human stereotactic atlas became the foundation for their own future work and for all who followed. Brains destined for inclusion in an atlas had to be fixed in situ as soon as possible after death. Before opening the cranium, the authors applied a stereotactic frame in order to pass metal rods completely through the skull and cerebrum at known distances from each other and at known stereotactic reference points in one or more planes. In this manner, shrinkage after fixation, freezing, or other processing could be quantified precisely and corrected by photographic enlargement or manipulation during preparation of the photographic atlas plates. Another major contribution was their demonstration of considerable variability in the contours and dimensions of the thalamus and other brain regions independent of variations in skull morphology. Even the cerebral midline deviated unpredictably from the midline of the skull. Brain atlases based on external cranial landmarks were suitable for small and medium-sized laboratory animals; that was not the case for humans. Thus, Spiegel and Wycis began a systematic search for reliable, radiographically demonstrable reference points on which to base stereotactic atlas planes of section and surgical procedures. Their initial reference point – the center of the pineal gland calcification on plain x-ray films – varied by 12 mm or more in the anteroposterior dimension and by as much as 16 mm relative to the interaural plane. They also
Printed stereotactic atlases, review
utilized the habenular calcification in some operations. In others, by means of lumbar pneumography, they visualized the posterior commissure (CP or PC), the foramen of Monro (FM), and rarely, the anterior commissure (AC). Spiegel and Wycis employed a line connecting the center of the PC with the pontomedullary sulcus at the posterior border of the pons (PO), to define the CP-PO line – an imaginary baseline – to construct their first atlas (> Figure 25-1). They cut their 5-mm-thick frontal (coronal) unstained sections and their 2–4-mm-thick myelin-stained frontal sections parallel to their so-called average cerebral directional line (inclined 4 degrees behind the CP-PO line). Oblique unstained sections 5 mm thick were cut through the brainstem at an angle of 30 degrees anterior to the CP-PO line and centered at the PC. A series of myelin-stained oblique sections 0.5 mm thick were cut parallel to the same . Figure 25-1 Landmarks for the intracerebral coordinate system used in Spiegel and Wycis’s first stereotactic atlas. Ch, commissura habenularum; Cp, posterior commissure; Cp-Po, posterior commissure – pons line; cran. 1, cran. 2, cranial direction lines; h1, horizontal line perpendicular cran. 1; ho, horizontal line perpendicular to Cp-Po; i + i, angles of inclination; Ob, medulla oblongata; Po, pons; Th, taenia habenulae (from [1], reproduced with permission)
25
plane. Unstained 5-mm-thick sagittal sections were cut parallel to the median plane. Unstained horizontal sections were cut perpendicular to both the median plane and the CP-PO line. This was as complicated as it sounds – and within 10 years Spiegel and Wycis and others eventually settled upon a simpler and more useful coordinate system. The final two chapters of Stereoencephalotomy (Part I) consist of anatomic and radiographic variability studies in 30 normal brains, plus postmortem examples of accurately and incorrectly placed lesions. In the variability studies, the authors catalogued the range of coordinates at which the borders of various nuclei, tracts, and other selected structures could be found in relation to both the center of the pineal gland and the posterior commissure. The 15 anatomic structures studied in this manner included the head of the caudate nucleus; the putamen; the globus pallidus; the anterior, dorsomedial, and ventrolateral thalamic nuclei; the pulvinar, tuber cinereum, mammillary bodies, corpus Luysii (subthalamic nucleus), substantia nigra, red nucleus, and medial and lateral geniculate bodies; and the mesencephalic spinothalamic tract. Interest in some of these targets would wane (dorsomedial nucleus of the thalamus, for example) and other more refined targets would emerge (the thalamic nucleus ventralis intermedius and the periaqueductal-periventricular gray matter, to cite a few). However, Spiegel and Wycis established a valuable precedent. Variability studies by other investigators – a process that continues into the present – have provided valuable insights into how stereotactic operations can go awry. The postmortem studies, especially the offtarget ‘‘misses,’’ illustrated the shortcomings of radiographic localization based on a single point such as the pineal gland, habenular calcification, or CP-PO line. By 1962, when Spiegel and Wycis published their second volume, Stereoencephalotomy (Part II) [2], general acceptance of the anterior commissure-posterior commissure line (AC-PC line (intercommissural line, IC line)) as the standard stereotactic reference system
349
350
25
Printed stereotactic atlases, review
had overcome the limitations inherent in the earlier method. Nevertheless, Stereoencephalotomy (Part I) probably did more to stimulate widespread interest and advancement in the field of stereotaxis than any other single publication. Early successes of Spiegel and Wycis were all the more remarkable given the nearly complete lack of previous experience in human stereotaxis between 1947 and 1952. Stereoencephalotomy, Part II: Clinical and physiological applications (Spiegel and Wycis, 1962)
Stereoencephalotomy (Part II), which was published 10 years after Part I, was primarily a textbook and only secondarily a revised and updated brain atlas [2]. The series of myelin-stained sections had appeared two years earlier in Confinia Neurologica [3,4]. In the interval, Spiegel and Wycis refined their stereotactic instrument (Stereoencephalotome Model V), improved and simplified their radiographic localization technique, and most importantly, embraced the Franco-German (Talairach and Schaltenbrand and Bailey) use of the intercommissural line as a stereotactic reference [5–10]. Most of the book was devoted to stereotactic techniques and clinical results, including honest morbidity and mortality figures for commonly performed stereotactic operations. Indications included psychosurgery, pain, involuntary movement disorders, epilepsy, and subcortical tumors. Perhaps the most valuable feature was presentation of post-mortem findings correlated with a patient’s radiographic, clinical, and surgical findings.
1957: Talairach and Colleagues Atlas D’Anatomie Stereotaxique: Reperage Radiologique Indirect des Noyaux Gris Centraux des Re´gions Mesencephalo-sou-Optique et Hypothalamique de L’Homme (Talairach, David, Tournoux, Corredor, and Kvasina, 1957)
Talairach’s first stereotactic atlas in book format appeared in 1957 (a magnificently produced and bound folio-sized volume), however, his pub-
lished scientific work on the subject dated back at least to 1949 [8,9]. Among the most important of Talairach’s contributions to stereotaxis were the introduction and popularization of combined positive-contrast and air ventriculography to demonstrate the AC and PC reliably, the invention of an accurately relocatable stereotactic instrument that utilized teleradiographic techniques and a ‘‘double grid’’ localization system, and the integration of angiography and ventriculography to create the most advanced stereotactic system in the pre-CT era. His technical developments shaped Talairach’s first and subsequent stereotactic atlases and radiographic-anatomic research over more than four decades [10]. One cannot overemphasize the importance of Talairach’s elegant demonstration that the deep structures of interest to stereotactic neurosurgeons bear a generally consistent relationship to the intercommissural line and its derivative planes [midsagittal plane, horizontal intercommissural plane, and the two vertical planes passing through the AC (VCA) and PC (VCP), respectively]. Later investigators would abandon Talairach’s two vertical planes in favor of a single intercommissural plane. Still, because of variation in length of the IC line between individuals (range, 23–28 mm from the center of AC to the center of PC; mean, 25.5 mm in Talairach’s work), one usually finds stereotactic coordinates listed as a distance anterior or posterior to PC (less often AC) as well as in relation to the mid-IC plane. Thus, Talairach’s system has exerted a lasting influence – even on workers who believe they have abandoned it for a more modern one. > Figure 25-2 reproduces Talairach’s illustrations of how his intracerebral reference system (the IC line and the VCA and VCP planes) and the locations of deep cerebral structures could bear a firm anatomic relationship to each other, yet vary considerably from the antiquated Horsley-Clarke reference system based on external landmarks. Even in a small sample of human ventriculograms, the axis of
Printed stereotactic atlases, review
25
. Figure 25-2 (a) Talairach’s illustration of the three early human stereotactic reference planes based on the same osseous cranial landmarks as used in animal stereotaxis. These were the infraorbitomeatal plane (Reid’s base line or the Frankfort line), the interaural plane (zero plane of Spiegel and Wycis), and the midline plane of the cranium. (b) Talairach’s demonstration of the variability in the major intracerebral axis (intercommissural line) from the osseous (Reid’s, or Frankfort) base line (from [9], reproduced with permission)
the IC line varied between 11.5 and 18.5 degrees from the infraorbitomeatal line (Frankfort line, Reid’s base line) (> Figure 25-2b). Talairach used his double-grid stereotactic instrument to create perforations in craniocerebral specimens at known locations and distances in the frontal and lateral planes. After this procedure, accurate coordinate measurements and profiles were derived for deep cerebral nuclei, subnuclei, and tracts, and the stereotactically marked brains were cut in either parasagittal or frontal sections along Talairach’s standard planes. He then mapped the three-dimensional profiles of the thalamic nuclei and other structures on millimeter-ruled diagrams. By presenting each sectional drawing next to the corresponding unstained or myelin-stained photographic plate at the same magnification, Talairach set the standard for all future stereotactic atlases. In this sense, even the apparently novel utilization of transparent overlays by Schaltenbrand and Bailey, and later by Schaltenbrand and Wahren [5,6], is derivative of Talairach. Talairach also invented a method to localize the ventral tier of thalamic nuclei in any patient. Using Talairach’s rules to proportionately subdivide the simple geometric
forms outlined by the IC line and the roof of the thalamus seen on lateral ventriculogram films, a neurosurgeon could draw or scratch Talairach’s diagram directly on the film, thereby recreating a properly scaled atlas template from which to derive stereotactic coordinates. Those who do not read French may be intimidated by Talairach’s atlas. However, the illustrations, captions, and labels are so clear that most of the essential data require no translation.
1959: Schaltenbrand and Bailey Introduction to stereotaxis with an atlas of the human brain (3 volumes) (edited by Schaltenbrand and Bailey, 1959)
The Schaltenbrand and Bailey atlas probably was the world’s most widely used compendium of brain maps during the nostalgic epoch of stereotactic surgery for involuntary movement disorders in the pre-L-dopa era (early 1960s to mid-1970s) [5]. Two oversized loose-leaf folio volumes contained the highest-quality myelinstained and unstained photographic atlas plates available at the time. The accompanying text, in both German and English, contained scholarly
351
352
25
Printed stereotactic atlases, review
treatises on neuroanatomy, physiology, and stereotactic techniques by the editors and 24 other contributors from the United States, Germany, and Italy. Despite having one editor and a plurality of contributors from the United States, the Schaltenbrand and Bailey atlas is decidedly Teutonic in style and content. Schaltenbrand’s stereotactic suite in Wurzburg contained an elaborate bidirectional optical projection system to superimpose atlas plates, anatomic outlines, adjustable magnification scales, and the patient’s ventriculogram images on the same translucent screen. In the decades before computer graphics, the atlas sections and scales could be optically modeled to match an individual patient’s intercommissural distance and other anatomic features. Volume II, the major portion of the atlas, will never become obsolete. Although the first three series of maps in this volume (in each of three orthogonal planes) contain unstained sections that are of limited usefulness to stereotactic surgeons, the next three series of maps contain the most magnificent myelin-stained atlas plates of
their day. The frontal series, cut orthogonal to both the mid-sagittal plane and the intercommissural line (and parallel to the midcommissural plane) (> Figure 25-3), begins with plate 36. The sections are presented four per page at 4 magnification, with a scaled and labeled transparent overlay attached to each page. The 16 sections, each 1–4 mm thick and all cut from the same brain, span the region from 16.5 mm anterior to 16.5 mm posterior to the midcommissural plane. The sagittal series, beginning with plate 42, is presented in the same manner, except that one or two sections appear on each page. The 18 sections are cut at 0.5–2.5-mm intervals, spanning the region between 2.0 and 27.5 mm lateral to the midline. Schaltenbrand and Bailey’s myelin-stained sagittal series has been a stereotactic bible of sorts for the past 50 years because the majority of functional stereotactic operations involve a transfrontal (precoronal) approach to the thalamus or upper midbrain through a parasagittal entry point. Among the 18 sections in this series,
. Figure 25-3 Schaltenbrand and Bailey’s three cardinal reference planes and their relation to the anterior and posterior commissures – a system that appears derivative of Talairach’s. With few exceptions (e.g., Afshar et al. [11], Andrew et al. [12]) every English language stereotactic atlas published since 1959 employed this same reference system. The illustration also shows the size of the central block of the brain (in millimeters) used to prepare the myelin-stained photographic atlas plates (from [5], reproduced with permission)
Printed stereotactic atlases, review
there is something almost magical about plate 47 (brain LXXVIII), which depicts the 13.5-mm and 15.0-mm planes. In most individuals, the hand area (median nerve territory) of the thalamic somatosensory relay nucleus resides in one of these planes (usually at 13.5 mm) and corresponds to the laterality at which the therapeutic lesion most often should be inflicted to relive parkinsonian tremor or essential tremor of the upper extremities (> Figure 25-4). Given the widespread use of the Schaltenbrand and Bailey atlas in both the printed format and as the basis for computerized software, the original owner of brain LXXVIII made an immense contribution to functional stereotactic surgery.
25
The myelin-stained horizontal series begins with plate 52 and, like the frontal series, is presented at four planes per page at 4 magnification. The 20 sections, all cut from a single brain, span the region from 16 mm above to 9.5 mm below the midcommissural point. Unfortunately, although the sections are all parallel to each other, they deviate from the axis of the IC line by approximately +7 anteriorly. Therefore, the +0.5-mm plane crosses the intercommissural plane within 0.5 mm of the midpoint of the IC line, but crosses the anterior commissure approximately 2.0 mm above its midpoint. Volume III contains 10-cresyl-violet-stained frontal sections and eight sagittal sections prepared
. Figure 25-4 Plate 47, brain LXXVIII, myelin-stained sagittal section 13.5 mm from the midline, with ruled and labeled acetate overlay. In the pre-CT stereotactic era (between 1959-late 1970s) this single atlas section probably guided more stereotactic operations – specifically, thalamotomies for tremor – than all of the other brain maps in the world combined (from [5], reproduced with permission)
353
354
25
Printed stereotactic atlases, review
in the same manner, each magnified 20 times. Aside from occasional use by the odd neurophysiologist or comparative anatomist, most copies of volume III spend a lifetime undisturbed on a library shelf.
1969, 1978: Andrew, Afshar, and Watkins A stereotaxic atlas of the human thalamus and adjacent structures; a variability study (by Andrew and Watkins, with the collaboration and histological assistance of Tomlinson, 1969)
A. Earl Walker wrote in his Foreword to this book that ‘‘In spite of thousands of stereotactic operations, the maps of target areas have been based upon very few anatomical preparations so that variations in size, shape, and position of the subcortical nuclei are inadequately established. It has been noted that the variations are so great that, using stereotactic coordinates alone, the chances of placing an electrode in a given small nucleus are indeed slight’’ [12]. Likewise, the authors noted that as of 1961 – when work on this book commenced as a project to delineate borders of the centromedian nucleus of the thalamus for surgery to relieve pain – even the recently published Schaltenbrand and Bailey atlas [5] contained relatively few thick-section variability diagrams based on only seven human specimens. The authors’ initially modest aims expanded into a more comprehensive textbook-sized study. The Methods section details how the authors – in a manner similar to the work of previous investigators – prepared, preserved insitu, indexed, and cut their specimens to permit calculation of- and compensation for shrinkage after formalin fixation. They ended up with 19 brains (38 hemispheres) after discarding specimens that were imperfectly cut, or were otherwise unsuitable for study. Because of their focus on thalamic nuclear variability, and because measurements based on dimensions of the thalamus were associated with less variability than measurements based upon the intercommissural (AC – PC) line,
the authors recorded AC-PC in all of their specimens, but preferred to base their variability measurements differently from classical Talairach space. In addition to AC – PC they measured: 1) the distance between the postero-inferior margin of the foramen of Monro to the midpoint of the ventricular (anterior) surface of the posterior commissure – the so-called FM – PC line or distance; and 2) the distance between FM and the tip of the pulvinar to determine the total thalamic length, or T.ThL. Readers should note that the postero-inferior margin of FM corresponds to the rostral limit of the thalamus (anterior nucleus). Most of the book consists of tabular and statistical analyses of the size, shape, location, and borders of thalamic nuclei and nearby structures (e.g., basal ganglia, optic tract). Probability tables and planar representations of each structure estimate the likelihood of finding a structure of interest at reasonable neighborhood coordinates relative to ventriculographically demonstrable landmarks – namely, FM, PC, the midcommissural plane, and the authors’ mid-thalamic plane (> Figure 25-5). From our perspective in the present high field strength MRI era, when one can see exquisite images of each patient’s deep brain structures immediately before (or even during) stereotactic operations, it is difficult to imagine what an eye opener the Andrew and Watkins atlas was for neurosurgeons of the 1960s. The actual atlas portion appears in the final two chapters – approximately 112 of the more than 250 pages – which contain full page 2.5 magnified line drawings and on the facing page, a corresponding stained photographic section (Nissl and or myelin-stained, from a representative specimen). This compendium of 21 coronal line drawings and macro photographs at 1.0 mm intervals from 1 to 21 mm behind FM, the anterior tip of the thalamus, and 17 sagittal line drawings and macro photographs at 1.0 mm intervals from 3–20 mm from the midline spans the entire thalamus, adjacent basal ganglia, and medial temporal lobes (> Figure 25-6). Only the principal nuclei and fiber tracts that were the
Printed stereotactic atlases, review
25
. Figure 25-5 Coronal and sagittal variability profiles of the subthalamic nucleus in the coronal plane 12 mm posterior to the foramen of Monro (FM, upper illustration) and in the parasagittal plane (lower illustration). White stippled shading represents 1 standard deviation (S.D.) and gray stippled shading represents 2-S.D., a range into which 66% of the population will fall relative to the FM-PC line and the mid-thalamic plane. (from [12] (Andrew, Tomlinson and Watkins), reproduced with permission)
subject of statistical analysis in the first portion of the book are illustrated. The outlines of the various structures are solid (based upon data from 70% or more of the specimens), light dashes (based on 40–70% of specimens), heavy dashes (based on <40% of specimens), or lightly dotted (histological features with no statistical analysis). In this manner, the authors graphically summarize and communicate the contents of nearly 130 pages of tables, graphs, charts, and statistical analyses onto 38 atlas plates. Andrew and Watkins’ comprehensive imaging and anatomical analysis (and similar analyses by Van Buren and Borke [13]), when combined with Emmers and Tasker’s physiological mapping techniques (see below), would represent the epitome of ste-
reotactic surgical knowledge and practices in the pre-CT era [13–15]. Stereotaxic atlas of the human brainstem and cerebellar nuclei: a variability study (Afshar Watkins and Yap, 1978)
‘‘This work began in 1971, a time when general interest was aroused by the possibility of treating spasticity of cerebral palsy and other disturbances of muscle tone and posture by ablation of the dentate nucleus of the cerebellum. . . . The technique devised to measure the cerebellar nuclei was then extended to measure within the brainstem (1) the tracts involved in the treatment of pain and (2) the other important nuclear structures and tracts [11].’’ This quote summarizes the purpose and historical context of this atlas.
355
356
25
Printed stereotactic atlases, review
. Figure 25-6 Coronal section 12 mm posterior to FM. The principal nuclei and fiber tracts that were the subject of statistical analysis are illustrated in this particular tracing using solid outlines (based on data from 70% or more of the specimens) or heavy dashes (based on <40% of specimens). Anatomical abbreviations are standard (from [12], reproduced with permission)
. Figure 25-7 The hemibrainstem, with reference points and planes defined by Afshar et al. F, fastigium of fourth ventricle; FB, line from fastigium to fourth ventricle floor; HBG, fourth ventricle floor plane, at right angle to the FB line; YFX, line passing through fastigial point and parallel to the fourth ventricle floor plane (from [11] (Andrews et al.), reproduced with permission)
Dentatotomy, stereotactic medullary trigeminal tractotomy, pontine or medullary spinothalamic tractotomy, and cerebellar stimulation are no longer performed. However, the Afshar atlas remains a valuable anatomic reference for the performance of other, more recently introduced open surgical procedures. These include ablation of the trigeminal nucleus caudalis for facial deafferentation pain and ablation of the nucleus solitarius for visceralbranchial pain associated with malignancy. The atlas was based on a study of 30 brains, using positive-contrast ventriculography and stereotactic marking of the specimens in situ. The authors defined the three planes useful for stereotactic localization of structures near the rhombencephalon (> Figure 25-7). The ventricular floor plane defined the floor of the fourth ventricle, the
fastigium-floor line (and plane) formed a right angle between the fastigium (tent shaped roof) of the fourth ventricle and the ventricular floor plane, and the plane through the fastigial point parallel to the ventricular floor plane defined the roof of the fourth ventricle. Most of the book consists of computer-generated variability profiles that graphically illustrate the probability (from >90 to 0%) of finding major brainstem nuclei and tracts at particular coordinates in any sectional plane. The atlas proper contains 54 myelin-stained brain stem sections photographed at 5 magnification and presented one per page. All sections are cut parallel to the fastigium-floor line (FFL) and orthogonal to the ventricular floor plane. An outline diagram with stereotactic coordinates and anatomic labels accompanies each photographic plate (in the style of Talairach, but organized more clearly). The 1-mm-thick sections extend from 23 mm rostral to the FFL (corresponding to the level of the red nucleus) to 30 mm caudal to the FFL (corresponding to the spinal tract of the trigeminal nerve, caudal to the gracile and cuneate nuclei) (> Figure 25-8). A similar treatment of the four
Printed stereotactic atlases, review
deep cerebellar nuclei appears immediately after the brain stem variability study. The variability studies and statistical profiles are illustrated in parasagittal graphic illustrations as well as in the transverse planes parallel to the FFL. The 12 myelin-stained cerebellar atlas plates are presented in exactly the same manner and at the same magnification as the brain stem photographs. They cover the range of transverse coordinates from 1 mm rostral to 10 mm caudal to the FFL. The anatomic detail presented in this atlas makes it the most valuable resource for brain stem stereotactic coordinates available in book format.
1972: Van Buren and Borke Variations and connections of the human thalamus, volumes 1 and 2 (Van Buren and Borke, 1972)
Van Buren and Borke produced this fine text and atlas while working at the National Institutes of Health in Bethesda, Maryland, and the folio-sized volumes were printed in Germany by SpringerVerlag [13]. Their work joined that of Schaltenbrand and Bailey [5], Schaltenbrand and Wahren [6], and Talairach and coworkers [9] to round out the ‘‘big four’’ stereotactic atlases of the pre-CT era, all printed in Europe, three in Germany.
25
Volume I includes a comprehensive textbook and a cytoarchitectonic study of the thalamus. Hundreds of high-quality photomicrographs of cresyl violet, myelin, and Golgi preparations are presented. In addition, extensive postmortem material from 54 patients shows both the original (therapeutic) cerebral lesions and the site and extent of secondary thalamic degeneration. Volume 2, Variations of the Human Diencephalon, continues the stereotactic atlas in the three orthogonal planes relative to the intercommissural line: sagittal, horizontal, and transverse (‘‘frontal’’ in the terminology of Schaltenbrand and Bailey). The atlas catalogues thalamic nuclei and their stereotactic coordinates relative to the IC line and the sagittal plane. Each series of cresyl-violet-stained plates is reproduced, one per page, at relatively high (8–10) magnification. The sagittal series consists of ten slices at 10 magnification, and spans the region between 2 and 25 mm lateral to the midline at 0.5–4 mm intervals. Outlines of nuclear groups and tracts are printed directly on the photographic plates, along with anatomic labels, coordinate index marks, and a magnification scale. Although the cell-stained sections at first appear unusual to surgeons accustomed to working from myelin-stained atlas plates, the stereotactic coordinates derived from this atlas correspond
. Figure 25-8 (a) Reference planes for measuring structural borders rostral to the 15-mm level (15 mm caudal to FB in > Figure 25-7). (b) Reference planes for structures at 15 mm or more caudal to the fastigial level are measured from the midline and the posterior medullary or spinal cord surface. a, anterior; l, lateral; m, medial; p, posterior (from [11], reproduced with permission)
357
358
25
Printed stereotactic atlases, review
closely to those obtained from other references. The cresyl-violet-stained horizontal series, presented at 8 magnification, consists of eight sections cut precisely parallel to the intercommissural plane. Sections approximately 3.5 mm thick span the region from 17 mm above to 8.1 mm below the IC line. The ten transverse (frontal) sections are photographed at 10 magnification, the same as the sagittal series. The plates cover a distance of 28.1 mm, from 23.4 mm anterior to PC to 4.7 mm posterior to PC. Individual nuclear profiles from five or six brains (depending on the plane of section) are mapped in the next chapters on minified (0.7) grids in the three orthogonal planes at 5-mm intervals on one or two pages each. The final chapter presents similar data for the gross anatomic structural outlines of 25 hemispheres normalized to either the AC or the PC. Simplified diagrams showing the region of densest overlap (median values) and extreme ranges also are presented. This feature allows a surgeon to visualize potential variations in the dimensions and locations of many structures conveniently (> Figure 25-9). But caution is in order; differences among individual brains tend to correct themselves when recalculated as fractional proportions of the IC distance (and thalamic height). For example, a point 6 mm anterior to PC in a brain with a 24-mm IC distance could be mapped in exactly the same relative (proportional) position if it were located 6.5 mm anterior to PC in another brain with a 26-mm IC distance. Both points are exactly one-fourth of the intercommissural distance anterior to PC. Thus, if one bears in mind the lessons of Talairach and others, the Van Buren and Borke atlas is a valuable tool.
1975, 1982: Tasker and Colleagues The human somesthetic thalamus, with maps for physiological target localization during stereotactic neurosurgery (Emmers and Tasker, 1975)
Owing to difficult accessibility during preparation of this chapter for the first edition, the present author omitted a discussion of Emmer’s and Tasker’s masterpiece of photographic, microscopic, physiological, and stereotactic spatial correlation of anatomy and function within the human somesthetic thalamus. This lavish folio edition presents the ten most useful (circa mid-to-late 1960s) somesthetic maps in the context of preCT, pre-computer – and of course, pre MRI – stereotactic surgery for Parkinson’s disease, other involuntary movement disorders, and intractable pain [14]. The book opens with a detailed description of Tasker’s stereotactic localization techniques using Leksell’s instrument, plus his own innovative attachments and accessories. Physiological localization is then explained as an iterative process during which each awake patient contributed new data to the aggregate responses obtained from electrical stimulation along similar (and nearby) trajectories in previous individuals. Tasker and others of his generation achieved extraordinary accuracy, confirmed by autopsy in cases with lesions inflicted for cancer pain, in adjusting the placement of thalamotomy lesions to account for variations in the dimensions and shape of the thalamus and adjacent structures. The particular planes selected for detailed presentation by the authors reflect the most commonly used stereotactic targets at the time. Portrait-quality black and white whole brain photographs – displayed larger than life size, one to a page – of a single brain with a 25 mm intercommissural distance and divided in the midsagittal plane included parasagittal plates at 9.0, 11.0, 13.5, 16.0, and 18.0 mm lateral to the midline cut from one hemisphere and coronal plates cut from the other hemisphere at 8.5, 10.0, 11.0, 12.5, and 15.0 mm posterior to the midcommissural point (at an angle 40 degrees frontal to the vertical plane perpendicular to the intercommissural line) (> Figure 25-10). Magnified and cropped microscopic sections from another 25-mm brain also were presented as full-page photographs to
Printed stereotactic atlases, review
25
. Figure 25-9 a and b. Sagittal variability maps in the 15-mm plane from Van Buren and Borke, normalized to AC and PC, respectively. Note the index mark (+) centered on AC in 9A and on PC in 9B. The solid outlines represent the regions of densest overlap of the labeled structures in specimens from 25 hemispheres. Broken outlines represent the extreme range of anatomical variability encountered among those same 25 hemisphere specimens (from [13], reproduced with permission)
identify the target sites and trajectories in each sample case corresponding to all ten sagittal and coronal whole brain sections. Finally, Tasker prepared a set of Woolsey-style [14,15] figurine diagrams to accompany each whole brain photograph and microphotograph. The site of electrical stimulation at 2-mm intervals along each trajectory was illustrated along with the type and distribu-
tion of somesthetic responses elicited during surgery (> Figure 25-11). Each three-figure suite graphically presented what a surgeon needed to know about the anatomy and physiology of the somesthetic thalamus and upper brainstem along commonly employed stereotactic trajectories.
359
360
25
Printed stereotactic atlases, review
. Figure 25-10 Parasagittal section 13.5 mm to the right of midline in a brain specimen having a 25 mm intercommissural distance. Short vertical hatch marks along the horizontal intercommissural line indicate AC and PC; the long vertical line indicates the midcommissural plane; the two horizontal lines above and below the IC line and the hatch marks at AC and PC approximate the boundaries of an accompanying photomicrograph that is not reproduced here. Amy, amygdala; CA, commissura anterior (AC); CC, corpus callosum; Cd, nucleus caudatus; CI, capsula interna; D, nucleus dentatus cerebelli; FH, fimbria hippocampi; GM, corpus geniculatum mediale; GP, globus pallidus; P, pes pedunculi; PCM, pedunculus cerebelli medius; TO, tractus opticus; II, ventriculus lateralis (from [16] (Tasker et al.) reproduced with permission)
Near the end of the volume, the pictures of nuclear models and homunculi representing the somesthetic thalamus, sculpted out of Styrofoam and clay, and photographed from different perspectives, appear quaint compared to the nearly instantaneous, high resolution computerassisted, three-dimensional graphic images now on display in any radiology suite or operating room (> Figure 25-12). Still, it is instructive to see and study an actual (in contrast to virtual) physiological model of the somesthetic thalamus in its totality. The value of this atlas transcends historical interest; it is the foundation for Tasker’s follow-on work, ‘‘The Thalamus and Midbrain of Man,’’ discussed in the next paragraphs. The thalamus and midbrain of man: a physiological atlas using electrical stimulation (Tasker, Organ, and Hawrylyshyn, 1982)
This 505-page 6- by 9-inch book by Tasker and colleagues does not contain a photographic stereotactic atlas like the other works described in this chapter [15]. Instead, and in light of his previous work, Tasker’s physiological atlas provides the most lucid English-language analysis of electroanatomic observations (based on 9,383 stimulation sites during 198 operations) relevant to stereotactic surgery on the thalamus and upper brain stem available to date. The 90-page miniatlas near the end of the volume depicts the results of stimulation mapping from the author’s clinical material in graphic form. The sites at which specific subjective experiences or observable phenomena were elicited are displayed on outline maps in the 2–20-mm sagittal planes based on the Schaltenbrand and Bailey atlas. Tasker’s elegant technique of anatomically and physiologically normalizing coordinates from different-sized
Printed stereotactic atlases, review
25
. Figure 25-11 Figurine maps of the peripheral field(s) for stimulation trajectories in six different patients superimposed on the same 13.5 mm lateral diagram. Each patient is identified in the coordinate diagram in the upper right-hand corner of the illustration by initials at the bottom of the trajectory ‘‘run’’ (Ao, Tim, Obe, F, Har, Mac). The vertical hatch mark at the left-hand side of the horizontal (intercommissural) line indicates the midpoint of the IC line. Numbers along the IC line, divided by four, equal the distance in millimeters behind the midpoint of the IC line where each trajectory crosses the IC line. The need to divide by 4 arises from the trajectories originally having been plotted on 4 life-size diagrams. ‘‘The limits of the projection field given by the patient with this initial localization of paresthesias at a relatively low stimulus intensity are outlined in solid black . . . (usually) 0.5–0.8 V with a 3 msec pulse duration at 100 Hz. After this, the stimulation intensity was increased in small steps . . . A change in the localized area was usually reported at a stimulus intensity which was twice its threshold. The area of the changed projection field is indicated by shading it with closely spaced lines’’ (Emphasis in original) (from [16] (Tasker et al.), reproduced with permission)
brains made the pooling of data possible from different patients. To perform the actual surgical procedure, he constructed sagittal brain maps for each patient, using a computer graphics program. The computer could expand or shrink selected atlas diagrams to match the patient’s intercommissural distance as determined by stereotactic
ventriculography. This was a refinement of Talairach’s system of proportionate coordinates, and a simplification of Schaltenbrand’s system of optical modeling. Later, once data from many patients (in the form of observations during electrode trajectory ‘‘runs’’ in a given sagittal plane) were available, Tasker superimposed the results on
361
362
25
Printed stereotactic atlases, review
. Figure 25-12 Homuncular composite sculpture of SI and SII. The open letter ‘‘B’’ indicates Plate B (from Figure 24) in the original text. Directional orienting symbols A, P, S, and I indicate anterior, posterior, superior, and inferior, respectively. Note ‘‘that the hand occupied an area that was approximately 1/4 of the total body area of the SI homunculus; whereas the shoulder of this homunculus occupied only 1/45 of the entire body . . . The relative size of the body parts of the SII homunculus was not as disproportionate as that of SI. . . . it resembled a reasonably realistic human figure’’ (from [16] (Tasker et al.), reproduced with permission)
composite maps (> Figure 25-13). In addition, during surgery or between steps in a multistage procedure, he transposed the anatomic boundaries of thalamic nuclei on the individual patient’s map according to the results of intraoperative stimulation. Consequently, important insights and generalizations emerged, greatly enhancing the surgeon’s ability to make rational decisions about what to do next during functional stereotactic procedures. Most important, Tasker emphasized that physiologically defined anatomy rather than blind obedience to atlas coordinates should determine the conduct of functional stereotactic operations. The preceding 80% of the book provides a detailed account of the remarkably stereotypical experiences that patients report and phenomena that neurosurgeons observe during stimulation mapping of the thalamus and midbrain. The reader learns how to identify responses that arise from stimulation of structures belonging to the dorsal column-lemniscus system, the spinothalamic pathway, the pyramidal and extrapyramidal systems, the auditory and vestibular systems,
. Figure 25-13 (a) Pooled data from Tasker’s early thalamotomy series, showing sites in the 13.5-mm sagittal plane at which electrical stimulation at any current threshold arrested the tremor of Parkinsonism, essential tremor, or cerebellar disease. (b) Data from Tasker’s series, showing the sites of thalamotomy lesions in the 13.5-mm sagittal plane made to relieve tremor in Parkinson’s disease and essential tremor (from [15], reproduced with permission)
Printed stereotactic atlases, review
the visual and oculomotor systems, and other structures. The last section provides postmortem anatomic correlations of electrode trajectories, stimulation sites, and lesion sites with records of intraoperative physiological observations in six patients with involuntary movement disorders or intractable pain. One might compare the process of functional stereotaxis to the navigation of previously charted but personally unfamiliar territory. In this sense, brain atlases are the maps, but the work of Tasker and colleagues is like the Michelin Guide. Both help the traveler to complete the journey successfully.
Traditional Stereotactic Atlases Currently in Print 1977: Schaltenbrand and Wahren Atlas for stereotaxy of the human brain with an accompanying guide (Schaltenbrand and Wahren, 1977)
This single-volume, oversized loose-leaf edition published in Germany more than 30 years ago is the last of the great stereotactic atlases of the twentieth century [6]. The ‘‘second, revised and enlarged’’ version of the Schaltenbrand and Bailey atlas represented, according to the authors, an effort to expand on the most clinically useful portions of the original work, debride the impractical or irrelevant material, and fit the finished product into a single volume. The authors were successful. They drastically reduced the number of unstained macrosections to 34 and eliminated the entire set of quadruple-foldout 20 Nissl-stained plates that occupied volume III of the first edition. Furthermore, the revised companion text that occupied volume I of the original atlas was delayed in publication until 1982, when it was released separately as a textbook [7]. The 1977 atlas reflects some of the stereotactic procedures that were in vogue at that time (analogous to the atlas by Afshar et al. [11]). For example, the myelin-stained transverse brain
25
stem and cerebellar series included 21 planes from the pontomedullary junction to the medulla (a span of 46 mm). Like all the other myelinstained microseries in this volume, each photographic section was presented at 4 magnification with a transparent overlay bearing anatomic legends and stereotactic coordinates. Surgeons interested in dentatotomy (for movement disorders), pontine spinothalamic tractotomy, mesencephalic tractotomy, and medullary trigeminal tractotomy or nucleotomy would have found this series of plates more helpful than those in the earlier edition – but still imperfect. Interest by the authors in ablative hypothalamic operations to control deviant sexual behavior led to carryover of hypothalamic Nissl-stained sections from the first edition. The ten 8 magnified sections, plus two anatomic key sections, occupy only two pages of the atlas. Other additions to the 1977 atlas include 25 color diagrams (on six pages) that summarize the radiographic and electroanatomic observations during stereotactic surgery on more than 300 patients. The Schaltenbrand and Wahren atlas contains only 34 macroseries photographs, all at 2 magnification and divided into three series as follows: 19 frontal planes from 57 mm anterior to 44 mm posterior to AC, five sagittal planes from the midline (0 plane) to 22 mm lateral to the midline, six horizontal planes from 18 mm above to 20 mm below the IC line from one brain, and four additional horizontal planes from 5 mm to 28 mm below the intercommissural line from another brain. The expanded interest in the horizontal unstained macroseries and the myelin-stained microseries was stimulated by the advent of CT and the authors’ foresight in recognizing the important role that axial imaging would play in the future. This time around, the authors cut all horizontal sections parallel to the intercommissural plane. In addition to the transverse myelin-stained brainstem series (21 planes) mentioned above, the three standard planes also were well represented, for a
363
364
25
Printed stereotactic atlases, review
total of 78 myelin-stained atlas photographs: 20 frontal planes from 16.5 mm anterior to 16.5 mm posterior to the midcommissural plane, 17 sagittal planes from 1.5 to 27.5 mm lateral to the midline, and 20 horizontal planes from 16 mm above to 9.5 mm below the IC line. Every neurosurgeon who performs functional stereotactic operations should have unlimited access to a stereotactic atlas. > Figure 25-14 reproduces Plate 43, brain LXXVIII, a myelinstained sagittal section 12.0 mm from the midline. This section probably holds the current world’s record – surpassing the 13.5 mm section from the 1959 edition – for having guided the most stereotactic operations (subthalamic nucleus region deep brain stimulation) for movement disorders. Although many older atlases are long out
of print, owing to constant demand, Schaltenbrand and Wahren’s 1977 volume is still available.
1988: Talairach and Tournoux Co-planar stereotaxic atlas of the human brain: three-dimensional proportional system: an approach to cerebral imaging (Talairach and Tournoux, 1988)
This volume, the next to last of Talairach’s stereotactic atlases, departs from the usual focus of such works. ‘‘In contrast to the majority of stereotaxic atlases that are primarily intended for the localization of the deep central nuclei, this atlas emphasized the interpretation of the vast cortical and subcortical spaces’’ [10]. The advances in high-resolution CT and MR imaging over the past 20–30 years and the accompanying resur-
. Figure 25-14 Plate 43, brain LXXVIII, myelin-stained sagittal section 12.0 mm from the midline, with ruled and labeled acetate overlay from the Schaltenbrand and Wahren atlas (10). In the modern era of MR image-guided deep brain stimulation of the subthalamic nucleus region (S th) for Parkinson’s disease, this atlas section probably holds the current world’s record – surpassing the 13.5 mm thalamic map from the 1959 edition – for having guided the most stereotactic operations for movement disorders (from [6], reproduced with permission)
Printed stereotactic atlases, review
gence of functional and anatomic stereotaxis (dealing with tumors and other structural lesions, and of course, deep brain stimulation instead of lesion creation) make such a guide through the borderland between functional and structural neurosurgery timely and informative. Mark Rayport observed in the translator’s foreword that the ‘‘images produced by these instruments [MRI and CT] have been utilized principally in the traditional manner of radiological interpretation: verbal description, identification of lesions.’’ For Talairach and coworkers, the vague, imprecise language, and frequent anatomy errors in MRI reports did not convey sufficient data for adequate neurosurgical planning and decision making. The potential pitfalls of traditional functional stereotactic operations are well known to experienced practitioners. However, nowadays many neurosurgeons with no training or experience in functional stereotaxis routinely perform anatomic stereotactic procedures such as biopsy, tumor resection, or radiosurgery. The planning and execution of such anatomic stereotactic procedures ideally should take into account possible immediate or delayed functional consequences. While some structures that surgeons wish to avoid, such as the optic chiasm and the midbrain tectum are obvious on imaging studies, other important structures, including the subcortical course of the pyramidal tract, the optic radiations, Forel’s fields, and the hypothalamic nuclei, are invisible or at least not obvious even on excellent-quality MRI. Talairach’s 1988 atlas provides neurosurgeons with a tool to help them navigate around and through such regions. In addition, by applying the lessons of this atlas to the interpretation of routine diagnostic imaging studies, one can achieve a high degree of accuracy in anatomic localization and clinical correlation. The authors begin with an exposition of Talairach’s proportional grid system in three dimensions, based on the length, height, and width of the whole brain. The orthogonal refer-
25
ence planes are based on the midline, the intercommissural plane, and the two verticofrontal planes that intersect the anterior and posterior commissures. As the authors explain, direct distance coordinates (in millimeters) vary widely from one brain to another. This is especially true the farther the point of interest lies from the IC line. Thus, Talairach and Tournoux divided the entire brain into cuboidal and rectangular prism-shaped parcels called ‘‘orthogonal parallelograms.’’ Each hemisphere is 9 major parcels in length (A-I along the axis of the IC line), 4 parcels wide (a-d along the transverse plane orthogonal to the midline and IC plane), and 12 parcels high (1–12 in vertical planes parallel to those defined by the commissures). The dimensions of each parcel are determined as follows: One-eighth of the distance between the IC line and the highest point of the parietal cortex and one-fourth of the distance between the IC line and the lowest point of the temporal cortex (parcels 1 through 12 in height); one-fourth of the distance from AC to the frontal pole, one-fourth of the distance from PC to the occipital pole, and the whole distance (subdivided into thirds) between AC and PC (nine parcels, A-I, in length); and one-fourth of the distance from the midline to the most lateral point of the parietotemporal cortex (4 parcels wide in each hemisphere). Even though this sounds complex when expressed verbally, things become clearer when one studies the diagrams (> Figures 25-15 and > 25-16) and remembers that each voxel represents a fixed proportion – not a distance – within the brain. The authors are careful to point out the limitations of this atlas, noting that ‘‘the millimetric values are valid for the brain presented here [only].’’ While this atlas is not among those one would consult before performing a procedure such as a thalamotomy for tremor, it is a valuable adjunct in planning or analyzing the effects of radiosurgery, or in planning surgical approaches to the deep hemispheric structures using stereotactic techniques. This work laid the foundation
365
366
25
Printed stereotactic atlases, review
. Figure 25-15 Illustration of Talairach’s space showing the brain divided into orthogonal rectangular prisms (Talairach calls them ‘‘parallelograms’’), the dimensions of which vary according to the principal axes of the brain. Each mini-volume is identified by three dimensions – indicated by a capital letter, a lowercase letter, and a number, e.g., A-d-1 for the shaded area in upper right-hand front corner) (from [10], reproduced with permission)
. Figure 25-16 Coronal (verticofrontal) section 20 mm behind AC, corresponding to plane E-3 in Talairach space. The original atlas illustration is in color. The boldface numbers refer to Brodmann’s cortical areas. GPrc, precentral gyrus; Ra, auditory radiation; Ro, optic radiation; other anatomic abbreviations are standard (from [10], reproduced with permission)
Printed stereotactic atlases, review
for Talairach’s final MRI and functional anatomy atlas published in 1993 [17], and inspired the present author to begin work in 1994 on an anatomical stereotactic imaging atlas of radiosurgical case studies [16].
Contemporary Stereotactic Atlases with Accompanying Digital Media 1998–2004: Mai and Colleagues Atlas of the human brain, second edition (Mai, Paxinos, and Assheuer, 2003)
The current version of the Mai atlas consists of two principal sections (in print) plus an accompanying CD and software [18]. A spiral-bound format with 9.5 13 inch (24 33 cm) pages makes this book especially convenient to carry to the operating room or radiology suite where it lies flat on the table, and folds easily to show one or two pages, and thereby occupies less space than any of the Schaltenbrand or Talairach volumes. Modest price, easy availability, portability, and the inclusion of a CD that contains digitized images and software all have contributed to the popularity of this atlas over the past 10 years. An introductory Preface and Materials and Methods section describes and illustrates in detail how the authors obtained, processed, and oriented the specimens for the anatomical and MRI preparations. Material included 17 heads – of which 11 were excluded owing to unexpected pathology, image artifacts, or other technical reasons. A ‘‘healthy, 25-year-old volunteer’’ contributed in-vivo MR images to the macroscopic imaging section of the atlas. Talairach’s method of spatial orientation and proportional boxes (voxels) was used and acknowledged. The first principal section, the Macroscopic Atlas, contains seventeen horizontal, fifteen coronal, and eight sagittal sections – each presented as a full page multi-modality plate consisting of several elements. Three pages of orienting diagrams introduce each series of macroscopic plates in the horizontal, coronal, and sagittal planes, respectively. Then, on each individual plate, small orienting
25
drawings of the whole-head (and in some cases, of the whole brain) indicate the plane and location of each individual section. These are accompanied by two small T-2 weighted MR images, and in some cases, by a small bone-windowed CT image – all in the same plane and at the same anatomic level as the anatomical section on the same page. Whole head specimens underwent post-mortem MR imaging followed by precisely oriented sectioning into approximately 1.0-cm-thick slices. Some of the CT images in the horizontal and coronal planes contain superimposed outlines of the vascular territories irrigated by major branches of the circle of Willis. Most of each page is occupied by an approximately 80% life-size (0.8) photograph of the whole-head and brain slice in-situ. On the facing page there is a comprehensively annotated artist’s tracing at the same scale. This mode of presentation helps the reader to become oriented when viewing only a single page or atlas plate at a time. Detailed identification of the extracerebral structures (muscles, orbital contents, paranasal sinuses, etc.) is a welcome addition. The largest section of the book, the ‘‘Microscopic Atlas’’ represents approximately threequarters of a century of detailed anatomical and radiographic study of one brain specimen from the Vogt collection – that of a 24-year-old man who died in 1929. This is where the Mai Atlas shines. Readers should bear in mind, however, that the Microscopic Atlas illustrates coronal sections from a single brain. The authors provide a complete bibliography and a summary of the previous and pertinent anatomical studies performed on this specimen. Four introductory pages diagram the orientation and precise location (in Talairach space) of the 69 coronal brain sections that span a distance of 60 mm anterior to the anterior commissure to 100 mm posterior to the anterior commissure – from the frontal pole to the occipital pole. Sections vary in thickness according to the anatomical complexity of each region. For example, the polar regions are sliced at 5.0–6.0 mm intervals, and the sections containing the hypothalamus are sliced at 0.6–0.7
367
368
25
Printed stereotactic atlases, review
mm intervals. Myelin-stained whole-mount sections are presented one per page, approximately 2.5 larger than life size. The facing page contains a richly labeled and annotated artist’s tracing of the section in Talairach space, and with ruled margins (> Figure 25-17). Each
cortical gyrus is labeled, as are deep gray matter nuclei, subnuclei, and white matter tracts. The detail presented here equals or surpasses that of any comparable work.
. Figure 25-17 Plate from the CD that accompanies the Mai et al. atlas [18]. This coronal section, in color on the CD, is at the level of the anterior commissure as indicated on the thumbnail inset in the upper right-hand corner. The horizontal and vertical 0 lines represent the intercommissural and midsagittal planes, respectively. Each numbered index line in the margins represents 1 cm; a millimeter scale is in the lower left-hand corner. Anatomical abbreviations are standard (from [18], reproduced with permission)
Printed stereotactic atlases, review
2007: Morel Stereotactic atlas of the human thalamus and basal ganglia (Morel, 2007)
This small (<150 pages; 6.5 10 inch) book represents 15 years of anatomical and functional anatomical studies by the author and various multidisciplinary collaborators and predecessors [19]. Introductory sections describe the methods used to obtain, preserve, orient, section, and stain the brains of seven older adults used to construct the atlas and neurochemical functional brain maps. The subjects had ages of 46–70 years, and died of cancer (n = 4) or cardiac disease (n = 3). Rather than reproduce previous investigators’ morphologic cytoarchitectonic work, Morel used protein and neurotransmitter immunohistochemical staining to infer functional relationships among and between thalamic nuclei and neighboring structures. Apart from the neurochemical findings, the regional brain anatomy illustrated in this atlas falls well within the normal range established by previous work. Although the considerable amount of neurochemical details and analyses interspersed throughout the book is interesting and important from a scientific perspective, it is not information that most stereotactic and functional neurosurgeons would find important while planning or executing individual operative procedures. On the other hand, this is exactly the kind of information that helps to generate ideas for new kinds of functional stereotactic surgical procedures. The book is organized by anatomical regions (thalamus, basal ganglia, subthalamic fiber tracts), with the pertinent atlas drawings embedded within each chapter. Eight pages of color illustrations from three different chapters (3, 6, and 7) are inserted between pages 86–87 – apparently a measure to allow color plates in an affordable print version. The full-color subject matter includes immunohistochemical findings by nuclear group in selected horizontal and sagittal sections (Chapter 3), two-dimensional variability diagrams of selected multi-planar sections (Chapter 6), and
25
superimposed two- and three-dimensional renderings of intraoperative electrode tracts in nine patients with their physiological responses to electrical stimulation (Chapter 7). The thalamus chapter contains introductory text, 13 pages of half-page horizontal section drawings (approximately 3.5; 26 sections; millimeter ruled margins) from 14 mm superior to 8.1 mm inferior to the intercommissural line in Talairach space, and 12 pages (24 half-page sections) of sagittal plates from 4.6 to 25.8 mm from the midline (1.8 – 23.0 mm lateral to the thalamo-ventricular border). Both planes of section used to construct the drawings were cut from opposite hemispheres of the same brain – as were the post mortem MR images, and myelin-, Nissl-, and immunohistochemical-stained multi-planar sections presented and analyzed in the final pages of this chapter. The basal ganglia chapter is organized much the same way, with a few introductory pages, a series of frontal sections arranged from posterior to anterior from 3.0 mm posterior to PC to 41.5 mm anterior to PC in Talairach space, and followed by 25 more (full-page) sagittal sections from 3.0 to 27 mm lateral to the midline (1.0–25.0 mm lateral to the thalamo-ventricular border). Again, the findings of multiplanar immunohistochemical staining in the basal ganglia are presented and analyzed according to current concepts of parallel neural processing within distinct anatomical and functional domains. For Morel these are sensorimotor, associative, limbic, and paralimbic. A brief chapter on Subthalamic Fiber Tracts is not part of the surgical atlas, but consists of wellorganizedsetsofmyelin-,Nissl-,andartistsdrawing sections to illustrate the anatomical details of subthalamic connections to the thalamus and basal ganglia. These are especially useful as investigators refine their understanding of the mechanism of action of current surgical procedures in this region, e.g., deep brain stimulation for Parkinson’s disease. Other plates compare the anatomical features visible on high resolution post-mortem MR images to corresponding myelin stained sections. The
369
370
25
Printed stereotactic atlases, review
. Figure 25-18 Drawing from Morel [19] showing a comparison of thalamic and basal ganglia subdivisions, and of cerebellothalamic tract (fct), in frontal maps of two different brains (red and white contours). Selected structures are in color in the original text (one of eight color pages) to better illustrate the variability. The larger brain (Cd closest to the top of the page) has an AC-PC distance = 30 mm; the other brain has an AC-PC distance = 26 mm. The scale bar in the lower left hand corner = 5 mm. DV 0, intercommissural plane; L 0, midsagittal plane; other anatomical abbreviations are standard (from Morel [19], reproduced with permission)
chapter on Interindividual Anatomical Variability accompanies the color plates that appear earlier in the book. It compares pairs of specimens with respect to anatomical similarities and overlap versus considerable differences in the location and borders of important structures (> Figure 25-18). The discussion of methods used to validate comparisons of nuclei and tracts between in vivo (or post mortem) MR imaging and measurements on preserved cut brain sections is even more valuable. This atlas-textbook provides important information to neurosurgeons regarding the limitations of other conventional atlases – and how to avoid pitfalls when ‘‘going by the book’’.
Conclusion The indications for guided brain operations have evolved over the past half-century (fewer lesions, more stimulation); and contemporary radiological techniques now provide magnificently detailed images of the brain itself, and its functional state – in contrast to the sometimes vague shadows used to navigate stereotactic procedures during the preMRI epoch. Elsewhere in this volume digital and computer-based stereotactic atlases are reviewed. Those products have long surpassed the tedious (in retrospect) printer-plotter based output of earlier generations in terms of speed, accuracy, and
Printed stereotactic atlases, review
utility. But for surgeons of each generation there is still something reassuring about opening one’s favorite atlas to the pages pertinent to the case at hand – and seeing the familiar myelin-stained sections and anatomical outlines right there on paper. Brain atlas books become friends – a relationship that is difficult to imagine with the computerresident atlases. This is not to diminish the value of computer-based or digital atlases, but perhaps it helps to explain the continued demand for new stereotactic atlases in print format, and for repeated press runs of the classic atlases. One barely notices the loss of computer atlas version 1.0 once version 2.0 is installed, but surgeons carefully pass their stereotactic atlas books from one generation to the next.
References 1. Spiegel EA, Wycis HT. Stereoencephalotomy, thalamotomy and related procedures. I. Methods and stereotaxic atlas of the human brain. New York: Grune and Stratton; 1952. 2. Spiegel EA, Wycis HT. Stereoencephalotomy. II. Clinical and physiological applications. New York: Grune and Stratton; 1962. 3. Baird RA, Spiegel EA, Wycis HT. Studies in stereoencephalotomy. IX. The variability in the extent and position of the amygdala. Confin Neurol 1960;20:26-36. 4. Benz RA, Wycis HT, Spiegel EA. Studies in stereoencephalotomy. XI. Variability studies of the nuclei ventralis lateralis thalami. Confin Neurol 1960;20: 366-374. 5. Schaltenbrand G, Bailey P. Introduction to Stereotaxis with an Atlas of the Human Brain. Stuttgart: Thieme; 1959.
25
6. Schaltenbrand G, Wahren W. Atlas for Stereotaxy of the Human Brain. Stuttgart: Thieme; 1977. 7. Schaltenbrand G, Walker AE. Stereotaxy of the Human Brain: Anatomical, Physiological and Clinical Applications. Stuttgart: Thieme; 1982. 8. Talairach J, Hecaen H, David M, et al. Recherche´s sur la coagulation the´rapeutique de structures sous-corticales chez L’Homme. Rev Neurol (Paris) 1949;81:1-24. 9. Talairach J, David M, Tournoux P, et al. Atlas d’Anatomie Stereotaxique. Paris: Masson; 1957. 10. Talairach J, Tournoux P. Co-planar stereotaxic atlas of the human brain: 3-dimensional proportional system: an approach to cerebral imaging. Stuttgart: Thieme; 1988. 11. Afshar F, Watkins ES, Yap JC. Stereotaxic atlas of the human brainstem and cerebellar nuclei. New York: Raven Press; 1978. 12. Andrew J, Tomlinson, JDW, Watkins ES. A stereotaxic atlas of the human thalamus and adjacent structures; a variability study by J. Andrew and E. S. Watkins, with the collaboration and histological assistance of J. D. W. Tomlinson. Baltimore: Williams and Wilkins; 1969. 13. Van Buren JM, Borke RC. Variations and connections of the human thalamus. Berlin, Heidelberg: Springer; 1972. 14. Emmers R, Tasker RR. The human somesthetic thalamus, with maps for physiological target localization during stereotactic neurosurgery. New York: Raven Press; 1975 15. Tasker RR, Organ LW, Hawrylyshyn PA. The thalamus and midbrain of man: a physiological atlas using electrical stimulation. Springfield, IL: Thomas; 1982. 16. Coffey RJ, Nichols DA. A neuroimaging atlas for surgery of the brain: including radiosurgery and stereotaxis. Philadelphia: Lippincott-Raven; 1998. 17. Talairach J, Tournoux P. Referentially oriented cerebral MRI anatomy. An atlas of stereotaxic anatomical correlations for gray and white matter. Stuttgart: Thieme; 1993. 18. Mai J, Paxinos G, Assheuer J. Atlas of the human brain. 2nd ed. New York: Academic Press; 2004. 19. Morel A. Stereotactic atlas of the human thalamus and basal ganglia. New York: Informa Healthcare; 2007.
371
31 The Riechert/Mundinger Stereotactic Apparatus J. K. Krauss
The Riechert–Mundinger (RM) stereotactic system was one of the first stereotactic devices which became widely accepted and which had a major distribution, in particular in Europe. The original system was developed by Traugott Riechert in Freiburg in the late 1940s together with the physicist Wolff [1,2]. With this first device, however, the accuracy was rather limited. Soon thereafter Fritz Mundinger modified and improved the original system [3,4] which then became known worldwide as the Riechert–Mundinger system (> Figure 31-1). Over the decades Mundinger continuously modified the stereotactic device [5–11] which was manufactured and distributed until the 1980s by the Fischer Company. Incorporating always emerging technical developments such as the introduction of computer tomography or magnetic resonance imaging, considering the needs for expanded uses of stereotactic frames such as image-guided craniotomy and tumor resection, and managing the challenge with new indications, the RM system has maintained its place over the years until nowadays in contemporary stereotactic and functional neurosurgery [12–14]. One important step forward was the introduction of the Zamorano–Dujovny (ZD) arc development in the early 1990s [15,16]. This application transformed the RM system which is a translational system into a center-of-arc system using the Zamorano–Dujovny 3/8 arc. This transition implies also that the target calculation which is based primarily on a polar coordinate system with the RM system moves to target calculation primarily based on Cartesian coordinates with #
Springer-Verlag Berlin/Heidelberg 2009
the ZD development (> Figure 31-2). For both systems the same base ring can be used. Neurosurgeons who worked for years with the frame therefore have the option to use both the advantages of the original RM system or the ZD development [14]. The RM system as well as its newer variant the ZD system are universal stereotactic systems which allow their use basically for all classical and new stereotactic procedures including tumor biopsy and interstitial curietherapy, drainage of cysts and catheter implantation, functional stereotactic surgery with radiofrequency lesioning or deep brain stimulation, image-guided craniotomy and stereotactically controlled tumor resection, stereotactic image fusion technology and stereotactic radiosurgery or radiotherapy. For decades the history and development of the RM system was firmly coupled with the F. L. Fischer company from Freiburg. Later the Leibinger company managed the development and distribution of the systems and the instruments. During the subsequent period when the RM and the ZD systems were managed by the Stryker company the major impetus was on the integration of the systems for radiosurgical and radiotherapeutic purposes. Nowadays, both systems are maintained and distributed by the Inomed company located in Teningen, Germany, a small town which is just a few kilometers away from Freiburg, the site where the original system was developed more than 50 years earlier. With the dedicated expertise of Inomed both systems have been expanded once more as exquisite tools for the performance of functional and stereotactic neurosurgery.
488
31
The riechert/mundinger stereotactic apparatus
. Figure 31-1 The classical Riechert–Mundinger system mounted on the phantom ring
. Figure 31-2 The translational principle on which the Riechert– Mundinger system is based (a) and the center-of-arc principle which is realized with the Zamorano–Dujovny system (b)
The Riechert–Mundinger Apparatus: A Translational System The Riechert–Mundinger apparatus basically consists of a circular base ring, an aiming bow and an instrument holder which sits on the aiming bow (> Figures 31-3 –31-5). The base ring has been made from different materials over the decades: steel, titanium, carbon fibre and ceramics. The most commonly used variant nowadays is made out of titanium. It is fixed to the patient’s head with four screws which are available in various lengths to accommodate for different head sizes (> Figure 31-6). Four fiducial plates are mounted on the base ring which are adapted according to the imaging technology. Using
current technology the system produces only very little artefacts with any imaging modality. As with any other stereotactic system it is important, of course, to have rectilinear alignment of the frame on the patient’s head for the purpose of stereotactic functional neurosurgery based on the anterior commissure/posterior commissure line [17], although deviations of pitch, yaw and roll can be corrected with modern planning software. The RM aiming bow is fixed along three points on the base ring which gives the system a high mechanical stability and target accuracy. The coordinates for the target point which have been defined according to polar coordinates can be checked with the help of a phantom system. This phantom system allows to detect miscalculations, however, it also allows easily to note errors in laterality and to detect even slight bendings of the guiding-cannulas or any other stereotactic instruments. The instrument holder accommodates any instruments used for stereotactic and functional stereotactic neurosurgery allowing easy readings within the submillimeter range. The main advantages of the RM systems are both its stability and its versatility, basically, any lesion in the brain can be reached from any entrance point.
The Zamorano–Dujovny Development: A Center-of-Arc System The Zamorano–Dujovny system uses the same base ring than the RM apparatus. The main difference between the two systems is the aiming bow, which along with the center-of-arc principle allows more flexibility in choosing the entry point during stereotactic surgery (> Figures 31-7 and > 31-8). Also, the instrument carrier has been modified in order to its altered functionality. Similar than it is the case with other
The riechert/mundinger stereotactic apparatus
31
. Figure 31-3 The Riechert–Mundinger system mounted on a skull: anterior view
. Figure 31-4 The Riechert–Mundinger system mounted on a skull: lateral view
center-of-arc systems the distance from the arc to its center is 19 cm. The aiming bow is not a complete semicircle, but a 3/8 circle. With that regard, of course, it cannot be fixed at three points on the base ring as
it is the case with the RM system. Instead, the ZD aiming bow is fixed at one point either on the lateral aspect of the base ring or on its frontal aspect incorporating all three othogonal Cartesian axes. The fixation of the aiming bow also
489
490
31
The riechert/mundinger stereotactic apparatus
allows to set and modify the coordinates in all three axes (x, y and z) during the operation which makes multifocal targeting quite comfortable. With this special construction care has to be taken to consider on which side the arc is fixed to the base ring in a given patient according to its laterality when bilateral surgery is performed. While any distance between 0 and 70 mm at an . Figure 31-5 The Riechert–Mundinger system with a contemporary instrument holder
angle between 0 and 70 can be chosen on the side where the aiming bow is fixed, the range for the x coordinate on the other side is limited to 20 mm, and the entry angle is between 0 and 20 . The instrument holder has been modified to carry any tools for functional stereotactic surgery including both electrical and mechanical microdrives. The CD development offers elegance and simplicity in its use coupled with high accuracy but with somewhat less stability than the original RM apparatus. It has expanded the scope of the RM system mainly by allowing its use as a centerof-arc system.
Tools for Stereotactic Surgery Both the RM and the ZD systems are supplied with a wide variability of hardware, targeting instruments and software. In particular, for use with magnetic resonance scanners an open stereotactic system has been developed (> Figure 31-9). The small diameter of the frame allows that it can be used even with small MR head coils. This particular frame is made out . Figure 31-6 Fixation of the base ring of the RM system to a patient’s head
The riechert/mundinger stereotactic apparatus
31
. Figure 31-7 The Zamorano–Dujovny development demonstrating the fixation of the aiming bow to the base ring
. Figure 31-8 Set-up of the Zamorano–Dujovny development on a model patient’s head
of ceramics. A closing bow can be supplemented to provide additional stability when needed. Special interfaces are available to connect the base ring to the head rests of all CT and MR scanners. Both systems are delivered with a wide range of biopsy sets including Sedan type instruments or a mini-forceps as well as dedicated tools
for functional stereotactic neurosurgery. Microelectrode recording is possible with a microdrive allowing either exploration of a single trajectory through the central channel or exploration of five trajectories simultaneously (> Figure 31-10). It is not advisable to use heavy electric microelectrode manipulators with the ZD aiming bow
491
492
31
The riechert/mundinger stereotactic apparatus
. Figure 31-9 The Zamorano–Dujovny system with an open base ring
. Figure 31-10 Dedicated mechanical microdrive for the RM and the ZD systems
since their weight might induce minor bending of the arc resulting in small yet relevant targeting inaccuracies. The classical radiofrequency electrodes have been developed further including monopolar and bipolar electrodes, and the unique chord
. Figure 31-11 The chord electrode which allows radiofrequency coagulation at a given distance from the target
The riechert/mundinger stereotactic apparatus
31
. Figure 31-12 CT stereotactic control of electrode positioning with fiducial plates mounted on the base ring after implantation of four electrodes in a patient with dystonia
electrode which allows to modify and shape the target with radiofrequency lesioning without the need to reinsert the guiding-cannula and alter the target coordinates (> Figure 31-11). Radiofrequency lesions can be made with the Neuro N50 under thermal control. Both the RM and the ZD systems can be used with any planning software with minor modifications. Dedicated software which has been developed specifically for these systems is also available, the IPS software which can handle all Dicom material such as CT, MR, PET and X-ray data.
Conclusions Both the RM and the ZD system have been developed as ideal instruments for stereotactic and for functional stereotactic neurosurgery. I have used the ZD system regularly for deep brain stimulation for treatment of movement disorders and pain syndromes over the years. Some modification of instrument technology on the instrument holder has made it possible to use the same guiding-cannula for both microelectrode recording and implantation of the electrodes exactly at the sites which have been determined by the microelectrode recordings. Stereotactic control of the positioning of the electrodes is straightforward (> Figure 31-12).
Acknowledgments The support of Mrs Mattmu¨ller and Dengler from Inomed providing several of the figures is greatly appreciated.
References ¨ ber ein neues Zielgera¨t zur intrak1. Riechert T, Wolff M. U raniellen elektrischen Ableitung und Ausschaltung. Arch Psychiat Z Neurol 1951;186:225-30. 2. Spiegel EA. In memoriam, Traugott Riechert (1905– 1983). Appl Neurophysiol 1983;46:320-322. 3. Riechert T, Mundinger F. Beschreibung und Anwendung eines Zielgera¨tes fu¨r stereotaktische Hirnoperationen (II. Modell). Acta Neurochir 1955; Suppl 3:308–37. 4. Riechert T. Development of human stereotactic surgery. Confin Neurol 1975;37:399-409. 5. Birg W, Mundinger F. Computer calculations of target parameters for a stereotactic apparatus. Acta Neurochir 1973;29:123-9. 6. Mundinger F, Birg W. CT-aided stereotaxy for functional neurosurgery and deep brain implants. Acta Neurochir 1981;56:245. 7. Birg W, Mundinger F. Direct target point determination for stereotactic brain operations from CT data and the calculation of setting parameters for polar-coordinate stereotactic devices. Appl Neurophysiol 1982;34:387-95. 8. Birg W, Mundinger F. CT-guided stereotaxy with the Riechert-Mundinger apparatus for biopsy and interstitial curietherapy of intracranial processes. J Neurooncol 1984;2:280. 9. Birg W, Mundinger F, Mohadjer M, et al. X-ray and magnetic resonance stereotaxy for functional
493
494
31 10.
11.
12.
13.
The riechert/mundinger stereotactic apparatus
and non-functional neurosurgery. Appl Neurophysiol 1985;48:22-9. Mundinger F, Birg W. The image-compatible RiechertMundinger system. In: Lunsford LD, editor. Modern stereotactic neurosurgery. Boston: Martinus Nijhoff; 1988. p. 13-25. Mundinger F, Braus DF, Krauss JK, Birg W. Longterm outcome of 89 low-grade brain-stem gliomas after interstitial radiation therapy. J Neurosurg 1991;75:740-6. Mundinger F, Boesecke R. The Riechert/Mundinger apparatus. In: Gildenberg PL, Tasker RR, editors. Textbook of stereotactic and functional neurosurgery. New York: McGraw-Hill; 1998. p. 73-8. Schrader B, Mehdorn HM. Operative Technik der tiefen Hirnstimulation. In: Krauss JK, Volkmann J, editors. Tiefe Hirnstimulation. Darmstadt: Steinkopff; 2004. p. 108-24.
14. Krauss JK, Grossman RG. Principles and techniques of movement disorders surgery. In: Krauss JK, Jankovic J, Grossman RG, editors. Surgery of Parkinson’s disease and movement disorders. Philadelphia: Lippincott Williams & Wilkins; 2001. p. 74-109. 15. Zamorano L, Kadi M, Jiang Z, Diaz F. ZamoranoDujovny multipurpose neurosurgical image-guided localizing unit: experience in 866 consecutive cases of ‘‘open stereotaxis’’. Stereotact Funct Neurosurg 1994;63:45-51. 16. Zamorano L, Martinez-Coll A, Dujovny M. Transposition of image-defined trajectories into arc-quadrant centered stereotactic systems. Acta Neurochi Suppl 1989;46:109-11. 17. Krauss JK, King DE, Grossman RG. Alignment correction algorithm for transformation of stereotactic anterior commissure/posterior commissure-based coordinates into frame coordinates in image-guided functional neurosurgery. Neurosurgery 1998;42:806-12.
32 The Talairach Stereotactic System A. L. Benabid . S. Chabardes . E. Seigneuret . D. Hoffmann . J. F. LeBas
History of the Concept It is generally agreed that stereotaxy was invented in 1905 by Horsley and Clarke [1] who needed an accurate tool for electrode insertion and lesion making in laboratory animals. This laboratory equipment was later adapted for neurosurgical purposes by Spiegel and Wycis in 1947 [2], although the first human stereotactic apparatus was probably built in London around 1919 by Aubrey Mussen, who had worked with Clarke [3]. There is however a report in the French magazine L’Illustration in 1897 [4] that two operations were performed to remove head projectiles using a system very similar to the current ones, including a biorthogonal X-ray set-up of Crookes tubes and X-ray films. This surgery, reported by Marey in 1897 in front of the French Academy of Medicine could have been the first ever performed frame-based stereotaxy Since that time, a large variety of stereotactic frames have become commercially available. However, the differences between frames are more related to the industrial features than to the methodological principles, since a frame plays the role of ‘‘sugar tongs’’ holding the skull, and its content the brain, in a fixed position. All stereotactic frames share a common goal, which is to establish rigid relationships between the patients’ head and brain and the outer space, which contains surgical tools, such as cannulas, probes, electrodes, or larger systems such as X-ray tubes used either for diagnosis or for therapy, To achieve this goal, the frames are firmly anchored to the patient’s skull by several pins. Making frame positioning reproducible is easy to achieve with every type
#
Springer-Verlag Berlin/Heidelberg 2009
of frame, at the cost of very few changes. Every frame has advantages and drawbacks. All of the specific features of each frame (such as goniometers) can be easily redesigned and adapted to the others, and. finally, all stereotactic strategies can be universal. Therefore, architectural differences between stereotactic frames, evolving with stereotaxy [5], express the main preoccupations of their designers and simultaneously, represent the solutions they considered best adapted to their needs. Most stereotactic systems are based on a polar or spherical (center-of-arch) approach, the purpose of which is solely to reach a point-like target (mainly in the basal ganglia for functional surgery), without endangering the structures encountered along the track. The first applications were in the study of abnormal movements (in order to confirm physiopathological hypotheses, either biochemical, electrical, or surgical) of pain and of epilepsy (in search of the putative pacemaker of centrencephalic epilepsies). One of Talairach’s original contributions was to try to determine the location of structures from radiological data, not only directly (such as Ammon’s horn, which makes a specific pattern in the ventricular occipital horn) but also indirectly (such as the amygdala, which is situated anterior to the temporal horn). In this perspective, during the 1950s Talairach at Sainte-Anne Hospital, in Paris, evaluated several coordinate systems, including that based on the corpus callosum, which was later readapted by Olivier [6] on the basis of MRI images. He finally chose the intercommissural line, drawn from the anterior commissure (AC) to the posterior commissure (PC).
496
32
The talairach stereotactic system
The second original contribution is undoubtedly the proportional grid system, which takes skull size into account. This provides an orthogonal system of grids, parallel or perpendicular to the AC-PC line, which respects the anatomic planes perpendicular to the X-rays used during the radiological examination. This is the rationale for the rectangular frame and the perpendicular approaches, which make possible the simultaneous exploration with a single electrode of the T2 temporal cortex and the amygdala or of the F2 frontal cortex and the cingular cortex. The theory behind this concept already appears in Talairach and Tournoux’s 1957 book on temporal epilepsy [7] which is very often quoted by Crandall [8]. In 1967, Talairach and coworkers extended these principles to the whole human cortex [9] in foresight of tridimensional imaging, long before radiologists even started to think about it. The atlas published in 1988 with Tournoux [10] is simply a colored and translated edition of the 1967 atlas, which is now commonly used as a reference by epileptologists as well as specialists in positron emission tomography (PET) and single photon emission computed tomography (SPECT). The identification of vascular patterns in 1977 by Szikla and colleagues [11] ‘‘was made possible because the proportional grid system enabled localization of the various cortical sulci. Talairach was also the first to perform stereotactic biopsies, taking a sample before introducing isotopes into lesions, in order to confirm the diagnosis. At that time, he considered biopsies of secondary interest and was essentially concerned with a tridimensional view of the brain. He devoted only two lines to biopsies at the end of his paper [12], which could explain why he is never quoted on this subject. The Talairach frame is the result of a rational attempt [9,10,13–15] to design a universal yet simple system, fulfilling these specific prerequisites: Patient placement in the frame should be reproducible in order to make it possible to divide a procedure into different stages while
still being able to take advantage of the data gathered during the previous stages Placement of the frame with respect to the X-ray system should also be reproducible for the same reasons. The X-ray system should provide a set of accurate two-dimensional (2D) projections of the cranioencephalic spatial object on films, with minimal magnification or parallax distortions, and with coherence of the Cartesian coordinates of a point in space on the two projections (which means that the vertical elevation Z of the point above the zero plane of the frame should be the same on both projections). The system should provide simple and safe means for the introduction of a tool into a designated target, without extensive calculations and with easy means of checking the safety of the chosen penetrating track, through a minimal opening of the skull limited to the size of the probes. Obviously, the recent development of computerized images and stereotactic softwares make these specificities less mandatory, as reformatting the 3D data has become easy and often transparent to the user. But the Talairach’s frame is more than a tool, it is the substrate of a methodology, if not a philosophy it lays the basis of a strategically thinking in 3D, based on 2D projections of the cranio-cerebral complex.
Description of the Talairach System The Talairach system comprises the frame itself (with the fixation pins), the double grids (which are used for calibration as well as tool introduction), and a long-distance X-Ray system. It is designed to fulfill the previously described requirements (> Figure 32-1): 1.
Reproducible placement of the patient is achieved by the use of a heavy frame base,
The talairach stereotactic system
32
. Figure 32-1 The Talairach stereotactic system. The frame is mounted on a rotating circular support in the original Grenoble setup. X-ray images are provided by a fluoroscopic amplifier. Penetration of tools (electrodes, biopsy cannula) is made (A), using a robotized stereotactic arm (B). The various components evolve according to technological developments: the DIXI frame (C) replicates the main structure of the Talairach frame, the X-Ray amplifiers are replaced by flat digital detectors (D), and the prototype of the robot has evolved to the third generation of the Neuromate stereotactic robot (E)
which cannot be distorted by the mechanical stresses required by the procedure, and by four strong pins inserted into 2.5-mm holes twist-drilled through the full depth of the skull, held by verniers, the graduations of which arc recorded and saved, and can be replicated. These pins can be replaced in the same holes weeks or months later as long as these holes have not been obliterated by bone regrowth. It is usually easy, under local anesthesia, to make a small skin incision above the holes and to reinsert the pins. Their exact repositioning is achieved by reproducing the same vernier readings as during the initial placement. The visibility of these pins allows exact superposition of subsequent X-rays, showing a ventriculogram, angiograms, and
2.
the grids through which probes will be inserted. It is sometimes necessary to replace the patient in the same stereotactic position several times within a 2-week period. Semi permanent screws have been designed (Stevis, Sofamor, France), which are made of titanium – compatible with magnetic resonance imaging (MRI). They are inserted into a 4-mm drill hole and can receive a connecting piece with the frame vernier. Replacement is then easy, painless, and can be done in non sterile conditions, allowing the use of a compatible MRI frame. The accuracy of radiological measurements is based on perfect orthogonality of the X-ray beams in the anteroposterior and
497
498
32
4.
The talairach stereotactic system
lateral directions, achieved by a specific installation of the frame at the center of a longdistance (3.5-m) X-ray machine set up in the operating room, using X-ray controls of the image of double grids placed on both sides of the patient’s head or using laser beam reflection on mirrors attached to the sides of the frame. The Talairach frame is positioned at the focus of a bidirectional X-ray system made of two tubes, the beams of which are orthogonal. A first vertical tube, located at the ceiling of the operating room, is used for anteroposterior views of the patient’s head in the supine position. The second horizontal tube is located along a wall of the room, for lateral X-ray views. Simple and safe introduction of tools is achieved by orthogonal approaches through the grids, placed on the sides of the frame base and perpendicular two by two. The tracks are therefore parallel to the X-ray beams used for the X-Ray examination (angiography and ventriculography) performed as a first step in the stereotactic procedure. In the absence of 3D computerized neuronavigation, this makes it possible to avoid vessels and enables high-precision positioning. The grids and grid-hole diameter (2.3 mm) are designed for specifically adapted screws, tubes for afterloading brachytherapy [16,17] or electrodes for stereoelec-troencephalography (SEEG) [18–21].
Like many other frames, the Talairach system provides the possibility of trajectory simulation and also of a double oblique track, using a multijoint probe or oblique grids. These accessories, however, were not well suited to polar approaches, which are achieved in a much more practical manner using the center-of-arch systems. Sedan’s [22] and Scerrati’s [23] goniometers were conceived and designed to provide the Talairach system with the advantages and flexibility of the center-of-arch systems, allowing easy and
precise access via oblique approaches to targets near the midline [22]. Sedan’s goniometer is made of a carrier moving back and forth and mounted on two lateral poles of the frame. This carrier holds a rod, which can rotate with a sagittal angle b and move laterally. On the medial end of this rod is mounted a sector on which the probe holder can be set up with a frontal (coronal) angle a. Correspondence between the x, y, and z cartesian coordinates of a point P; the a and y angles of the r, a, and y spherical (or polar) coordinates: the b and g angles read on the X-rays, and the b and e angles set up on the goniometer are given by the following simple equations (> Figure 32-2), where r = OP: x = r.cosa.siny y = r.cosa.cosy z = r.sina
b = arctan(tana/cosy) g = arctan(siny/tana) e = arctan [x/(x2 + z2)1/2]
Scerrati’s goniometer is a center-of-arch system [23]. Similarly, all calculations used for the center-of-arch system are applicable to this goniometer [24].
Talairach Frame Setups In all cases, long-distance X-rays contribute to the basic concept of the Talairach system in order to avoid image distortion. Computer correction of the image distortion is achieved within the neuronavigation softwares currently available, which are based on digital images, and allow using X-ray setups, with a shorter distance. The frame by itself can be mounted, as can other frames, in several different manners, as below: 1.
The Sainte Anne setup is an all-mobile system, with a motorized operating table, two X-ray tubes mounted on a motorized ceiling arch, and on a vertical lateral pole with a laser beam centering system. All patient
The talairach stereotactic system
. Figure 32-2 Spherical and Cartesian coordinates of a point P. A point P may be represented by a set of three projections on the axes X, Y.Z (Cartesian coordinates) or by a vector OP and two polar angles (b and g)
4.
32
Like any other type of frame, the Talairach stereotactic frame can be the basis of any robotized system [25–29], which requires the head to be firmly held in a fixed position. This setup is currently used in Grenoble [25], even though the various compounds have evolved and benefited from technological progresses.
Stereotactic Practice with the Talairach System Routine Procedure
2.
3.
positions are possible, making this system especially useful for ventriculography. The Rennes setup includes an isocentric mobile seat (CGR Isocentrix), which is an ideal system for permanent centering of the frame and head whatever the position of the patient. The Grenoble setup of the frame is inexpensive and practical. The frame is mounted on a rotating holder (> Figure 32-7) coaxial to the patient, fixed on a solid-state base screwed onto the floor of the operating room at the focus point of a permanent biorthogonal X-ray system. The sitting position is not possible and complete examination of the ventricular system is obtained by rotating the patient around his or her longitudinal axis.
The patient can be placed on the frame under general or local anesthesia. Standard X-rays are taken with the double grids, which localize the central beam and display the various grid holes through which the penetrating track will be made. Angiograms and ventriculograms are performed and the ‘‘synthetic’’ diagram is then made. Computed tomography (CT) and MR1 data can be reported on this diagram, from which the final target may be determined. The penetrating track can be chosen through the grid hole that best corresponds to the target, or coordinates can be measured to direct either a Sedan’s or Scerrati’s arch or to drive a robotized system. During the penetration step, X-rays are taken that check the position of the inserted probe or of the recording or coagulating electrode, biopsy cannula, brachytherapy tubes., depth electroencephalographic (SEEG) semipermanent electrodes, ventriculoscope, and so on. These positions can, in turn, be reported on the diagram. A document summarizing all data specific to the case is progressively built up and may be used for further therapeutic steps, such as radiotherapy or cortectomy planning. Location of functional structures can be added to this diagram, using the proportional grid system.
499
500
32
The talairach stereotactic system
Overlay of Stereotactic Neuroradiological Modalities
Data Processing in the Talairach System
The pins inserted into the skull are in a fixed and reproducible position and their appearance is the same on every X-ray picture, taken during the same or different stereotactic sessions provided that the pin-holding verniers are set at identical values each time. Therefore, by matching the pin projections on the X-ray pictures, the Talairach system has the major advantage of providing the possibility of superimposing different X-Ray modalities, such as angiograms, ventriculograms, or any other kind of image acquired in stereotactic conditions (> Figure 32-3).
Proportional Grid System Talairach has developed a proportional anamorphosis procedure [9,12,13], based on the thirdventricular landmarks (anterior and posterior commissures) and on the inner skull contours in order to normalize every individual brain on a standard diagram and eliminate individual anatomic variability (> Figure 32-4). Recognition of various structures (cortical sulci and lobes, white matter bundles, as well as basal ganglia substructures) can be done using this proportional
. Figure 32-3 X-Ray images acquired in stereotactic conditions. All the steps of the stereotactic procedure yield X-rays, showing the ventricles, vessels, and position of the probes (during biopsies or during implantation of electrodes or of cannulas). On each X-ray, certain features are always visible: pins, fiducials and even the skull bone. These can be used to match X-rays and to draw on a common diagram all the relevant data of every modality. (a) Position of the deep brain EEG electrodes. (b) Drawing of arterial vessels, superficial (continuous lines), and deep (dotted lines). (c) Superposition of vessels and sulci from MRI. (d) Planning of temporal resection
The talairach stereotactic system
32
. Figure 32-4 Proportional grid system: the borders or the rectangle are tangent to the inner table of the skull, parallel and perpendicular to the intercommissural AC-PC line. Various structures, such as sulci, may be located using the grid parcellation. AC, PC, anterior and posterior commissures; SS, sylvian sulcus; STS, superior temporal sulcus; LFS, lower frontal sulcus; RS, rolandic sulcus; CS, calcarine sulcus; MF, motor fibers; PT, pyramidal tract; upper diagram, lateral; lower diagram, anteroposterior
anamorphosis [9,12,13]. The predicted location of cortical sulci as compared with their actual position as shown by MRI has recently been validated [30].
Data Computation in the Talairach System The Talairach system is particularly adapted to calculations developed to correct spatial distortions and provide corrected co- ordinates for accurate probe placements. X-ray images in stereotactic conditions provide spatial localization of points within the cerebral space, but this localization is inaccurate because of two phenomena: 1.
Magnification. Magnification depends on the respective distances to the film, of the point of the cerebral space (d) and of the X-ray tube (D). The magnification coefficient that enlarges every measured distance between two points in the space is: G = D/(Dd). Our setup, in which tubes are 3.5 m away
2.
from the center of the frame, achieves a low magnification ratio of 1.05. Correction of parallax errors. Parallax depends on the distance of a given point in the cerebral space to the axis of the X-ray beam, which is the zero point. Every point that has, with respect to this central X-ray, coordinates x, y, will actually have coordinates X, Y on the film. These are X = G x and Y = G y
This must be taken into account in calculating the setup parameters of the frame. Precise data regarding the central X-ray beam, perpendicular to the frame faces, can be determined on the doublegrid image where the central beam passes through similar holes on both grids (> Figure 32-5). The simplest situation corresponds to a central beam placed on the area of interest and centered on the target. When this central beam is centered at distance from the target, its actual position is used for exact correction of the parallax distortion for any point in the brain [30].
501
502
32
The talairach stereotactic system
. Figure 32-5 (A) lateral X-ray of a grid with an array of 27 31 holes. (B) interference pattern of the two grids. One may recognize the central beam (cross). The next interference corresponds to a shift of one hole spacing (3 mm) from the grid on one side to the other one
Detection of Vascular Injury Along a Double Oblique Biopsy Track With any penetrating trajectory into the brain there is the risk of encountering a vascular structure; the greatest risk occurs during biopsies [12]. The Talairach grid system is mainly set up for orthogonal, frontal, or lateral approaches. Biopsy tracks performed through the grids are aligned along the X-ray axes and provide the safest procedures, since it is possible to check on the corresponding X-ray images that the projected track, which appears as a point, does not correspond to any projection of a vessel. In the case of double-oblique approaches, the problem of detecting vascular collisions is not as easy to solve, despite the fact that CT-guided stereotactic biopsies without angiographic control are popular [8,22,24,31,32]. Effective solutions must be found, and some have already been designed and used. The intrinsic features of the Talairach system has led Szikla et al. [11,17] to develop a routine procedure proven to be effective and easy to perform without any computation, but which can be easily computerized and automated. Provided that the two X-ray beams are orthogonal, a given point in the brain appears on X-rays as two pairs of coordinates (x, z) on the frontal view and (y, z) on the lateral view, z being the same in both pairs (> Figure 32-6). Therefore,
projections of the intersection of a putative track with a vessel must have the same z value (as measured on X-ray films from the base plate of the frame or from any other reference plane) on both lateral and frontal planes (> Figure 32-7). Obviously, the reciprocal is not true, and it may happen that lateral and frontal intersections having a same z value do not correspond to the same vessel: In these cases of false collision. The decision between true and false collision is made by the surgeon’s expertise. This method of collisiondetection, which can be computerized on digitized angiograms. is much easier and faster to achieve but less elegant than true collision detection without false-positive points provided by real 3D reconstruction of the vascular network. Another approach is derived from the ‘‘floating line’’ concept [33]. A specially built stereocomparator features two movable lines on transparent grids, applied onto two stereoscopic angiograms and representing the projections on these tilted angiograms of a theoretical line in brain space. Observation of this line through the stereocomparator allows the surgeon to check for eventual collision of the line with vessels and eventually to change it. The Talairach system provides another approach that has been used profitably in routine practice [17,34] to recognize the in-depth position of the vessels using small-angle double-incidence
The talairach stereotactic system
. Figure 32-6 X-ray setup. Any point of the brain with xi, yi, zi coordinates will have Xi, Yi radiological coordinates with a magnification coefficient G(xi) that depends on the geometry of the system. When the frame (and the head) is rotated by 5 around the patient’s axis (lower figure), the spatial coordinates of point P become x0 i, y0 i, and zi remains unchanged. Its radiological coordinates become X0 i, Y0 i, and Zi
32
. Figure 32-7 Vascular collision between a vessel and a track must have the same Z altitudes on frontal and lateral X-rays. In this case B and E correspond to a probable vascular collision
. Figure 32-8 Superposition of two X-ray angiograms taken with a 5 tilt angle of the lateral X-ray beam (obtained by a 5 tilt of the frame-head ensemble). Coincidence of vessels can be obtained only for limited segments situated in the same plane. On this figure, only the pericallosal artery (midline plane) is matched, while more superficial branches of the sylvian artery are shifted. A different shift of the films relative to each other would lead to the superposition of the images of vessels situated in a plane at a different depth
angiograms (SADIA) taken under a 5 tilt angle, which corresponds to the natural binocular vision angle. One may use a stereocomparator or, with some training, it is possible to squint and obtain a 3D perception of the vascular network. One may also superimpose the two angiograms and try to make the vessels correspond. Coincidence of the two images of the vessels is possible only for those that are in the same plane perpendicular to the X-ray axis (> Figure 32-8). Slightly sliding the films one over the other will change this ‘‘coincidence plane’’, and display another array of vessels
situated within it. This technique is easily used in daily routine to evaluate the depth of vessels projecting on a proposed trajectory. Obviously, the approach described above can be formally demonstrated and could be used as a possible basis for 3D reconstruction [35]. Consider lateral views taken as SADIAs. Every point P
503
504
32
The talairach stereotactic system
of the brain is assigned a triplet of coordinates (x. y, and z) in brain space, a pair of coordinates (Y, Z) on the regular lateral view film, and (Y, Z0 ) on the lateral view film of the 5 tilted head. Therefore, x corresponds to the ‘‘depth’’ of a point along an axis Ox perpendicular to the film plane. When the two films are superimposed with a given shift d with respect to an arbitrary reference (Y + d = Y0 ), two sets of points belonging to the two films are placed in coincidence. One may easily demonstrate [35] that there is a relationship between d and x, which is dependent on y (> Figure 32-6). When superimposition of the films is achieved, some structures, such as vessels, can be matched on both films when there is a shift equal to d. Therefore: x ¼ 11:9d 0:04Y The depth x can therefore be calculated for all points of the film that are situated at the coordinate Y and coincident to their homologous projection on the tilted film when the shift is equal to d. A paradigm can be derived from this procedure. A complete set of coordinates is therefore generated and, when displayed, provides a 3D reconstruction of the vascular network. It is clear that these methods of computations were extremely precise and useful at the time of the conception of the Talairach’s system, and are nowadays obsolete: current methods of modern imaging resident of most of the neuronavigation softwares associated to commercial stereotactic frames provide an efficient solution to these problems and allow interactive safe navigation on the MRI or CT images
Connection of a Stereotactic Frame to a Computerized Imaging System The original Talairach frame does not have localizers designed for MRI or CT examinations in stereotactic conditions. Moreover, its metallic composition makes it incompatible with MRI, and the pins verniers are too clumsy to fit easily
within the MRI gantry. Several solutions have been proposed to overcome this problem. Computed Tomography
For CT data obtained in stereotactic conditions using other MRI-compatible frames, specific adapters can be designed. Sedan has adapted the Leksell frame system in which the patient is initially set up and CT examination is performed using the localizers and software developed for this system. While still on the Leksell frame, the patient is then transferred to the Talairach frame, using a specifically designed adapter. We have adapted the Fischer-Lcibinger and then later the CRW frames and we simply transfer the vernier values from one frame to the other. When CT examination has been done under regular circumstances, several methods help in reporting the shape of the lesion in terms of the stereotactic diagram, making possible the use of this information for stereotactic procedures [35,36]. Magnetic Resonance Imaging
It has been stated that MRI cannot provide a precise spatial localization because of its nonlinearity [37,38]. This is actually partly wrong: precision is a matter of tuning the system correctly. The easiest way is to enlarge the images at the scale of the stereotactic pictures and to match them to similarly visible features and anatomic structures. Sedan et al. [30] had designed a television-based system that can pick up MRI parasagittal views and redisplay them, using a variable gain along the X and Y axes. The recent MRI systems can actually display hard copies at any desired magnification. Provided that MRI gradients are properly checked and adjusted if needed, MRI images are used by superimposition of a calibration grid positioned identically on each picture. This provides a composite picture featuring all relevant data, such as the inner contour of the skull, the coronal suture, depression of the torcular, the ventricular system and essentially the third ventricle, the aqueduct of Sylvius and the fourth ventricle, the rostrum and sple-
The talairach stereotactic system
nium of the corpus callosum, and sometimes the siphon of the carotid artery. All these structures are visible on the stereotactic ventriculogram and angiograms, providing a stereotactic diagram that can therefore be matched to the MRI data, particularly since all these image modalities are digitized and can be therefore numerically processed and the image fusion and 3D reconstruction fully computerized and automated. Digital Radiology
Digital subtraction angiography (DSA) and ventriculography are replacing conventional X-ray films. These digital radiological images are easily processed and matched with other image modalities. Target coordinates are then precisely and quickly obtained and may be used to drive a
32
computerized or robotized system. Two flat digital detectors (Pixray, Bioscan, Geneva, Switzerland) are mounted on supports, solitary to the rotating stand holding the frame, and can be move away using revolving hinges, which allow them to give more working space around the patient’s head, particularly during SEEG procedures where implantation of the depth electrodes require a large free access on both sides of the head at the time of electrode insertion. X-ray images can be taken easily both on lateral and antero-posterior views, and immediately observable on the flat digital screen disposed close to the operating field, to be observed by the surgeon, during the introduction of probes, as well as during contrast angiography or ventriculography. Moreover, this
. Figure 32-9 The new version of a Talairach derived stereotactic frame. (on the right). (a) X-ray angiolocalizer with four plates bearing four opaque fiducials each. (b) The modified Talairach frame mounted on the rotating plate of the Grenoble set-up. (c) The MRI localizer including N shaped fiducials (being filled with copper sulphate. (d) Frame mounted on the rotating stand and equipped with the two orthogonal flat digital X-ray detectors (one for lateral X-ray pictures, and two for antero-posterior X-ray pictures.) (e) X-ray command console (1), computer screen (2) and central power unit (3) for control of the digital data acquisition and processing system
505
506
32
The talairach stereotactic system
strongly decreases the cost of the investigation (X-ray films are no more needed) and its duration (film processing is also suppressed). The visualization of the fiducials on the angiolocalizer allows precise matching of subsequent series of images taken even at several days intervals, and is used by the robot computer to match these data with other types of digital data, such as MRI, but probably in the future, SPECT, PET and MEG data. Design of a New Talairach Frame
To circumvent the non compatibility of the original Talairach’s frame, and to make it still compatible with the basic principles of the Talairach system, we have redesigned it (Universal Talairach frame, DIXI, Besancon, France), keeping the structural design (heavy rectangular base, non deformable, able to receive grids and other mechanical systems or arches fixed on the four corner
posts), but adding new features allowing multimodal digitized imaging (CT and MRI localizers, fiducials for neuronavigation), and coregistration with the navigation software of robotized systems such as the last version of the Neuromate (Schaerer-Mayfield, Lyon, France). This frame is currently available and can be adapted to various stereotactic tool holders (from the robotized arm Neuromate to center-of-arch systems and goniometers from the industry). The fixation using transcranial pins has been kept, as well as the possibility to reposition them in bone screws previously used, allowing the repositioning of the patients along a sequence of subsequent steps (for instance ventriculography under general anesthesia, stereotactic MRI and CT in awake condition in the same evening or the day after, the implantation of deep brain stimulation electrodes 2 or 3 days
. Figure 32-10 Neuronavigation windows of the IVS Voxim software exhibiting the two orthogonal X-ray views of the target determination on the ventrieulogram (Subthalamic Nucleus), the coronal MRI image at the level of the target, with superimposition of the implantation tracks, and the 3D head reconstruction showing the image of the N shaped fiducials
The talairach stereotactic system
32
. Figure 32-11 Pre-implantation control procedure of the accuracy of the placement by the robot of a ‘‘phantom’’ probe: (a) and (d) Ventriculographic determination of the target (coronal and lateral views). (b) and (e) Images of the ‘‘phantom’’ probe in position according to the planning. (c) and (f) Superimposition of the two sets of digital images showing the exact projection of the tip of the probe on the target
later, a second, postoperative, MRI in stereotactic conditions 2 days later, after which the screws can be removed, a few days before the implantation of the programmable stimulator under general anesthesia). One may take advantage of this easy and precise repositioning to check the accuracy of the targeting done during the preplanning session: when the targets and trajectories have been chosen and set, using the neuronavigation software, which had fed the robotized arm controller with the appropriate data of the robot to move and reach the desired position, one can perform a simulation or shame procedure just before setting the patient on the frame using the replaceable pins with same readings of the verniers as during the ventriculographic step: The robot is launched and set to reach its position, determined by the target coordinates; a ‘‘phantom tool’’, made of a rod is set in place instead of the electrode guidance system, the length of which is set a the same length than the implanted electrode; digitized X-rays images are taken,
which can be digitally superimposed to the digital scheme of the target: the correspondence must be complete, and any shift is easily detected, avoiding errors or allowing making corrections. The phantom rod is then withdrawn; the robot is retracted to a standby temporary position, the patient is reinstalled on the frame. The robot can therefore be sent again in working position and the stereotactic procedure of electrode implantation can be safely performed with minimal risks. (> Figures 32-9– > 32-11).
Conclusion The main characteristic of the Talairach system is that it renders compatible all procedures (diagnostic and therapeutic) performed on the frame during the same or during subsequent sessions, which may be separated from each other by weeks or even months. It is also designed to provide minimally distorted numerical spatial data and to allow corrections of these distortions. The
507
508
32
The talairach stereotactic system
Talairach system has been the basis of a rational approach, taking advantage of the orthogonality of X-ray incidence to define precisely the position of the targets and vascular structures within the brain with respect to the coordinates of the frame system. These features have been used to develop methods of computation accessible to surgical teams with little or no computational means but also applicable to automated software. The stereotactic Talairach frame is suited for connection with spatially guided and computer-assisted robots, as it provides a basic spatial reference that is easy to integrate into a routine for driving a robot toward a spatially defined target.
References 1. Horsley VA, Clarke RH. On the intrinsic libels of the cerebellum, its nuclei and its effect tracts. Brain 1905;28: 12-29. 2. Spiegel EA. Wycis HT. Marks M. Lee A. stereotactic apparatus for operations on the human brain. Science 1947;57:164-7. 3. Picard C. Olivier A. Bertrand G. The first human stereotaxic apparatus: the contribution of Aubrey Mussen to the field of stereotaxis. J Neurosurg 59:67-36. 4. Remy and Contremoulins G. Le chercheur de projectiles. L’Illustration 1897;55:423. 5. Gildenberg PL. Whatever happened to stereotactic surgery? Neurosurgery 1987;20:983-7. 6. Olivier A. Extratemporal resections. In: Engel J, editor. Surgical treatment of the epilepsies. 2nd ed. New York: Raven Press; 1993. p. 489-500. 7. Talairach J, David M, Tournoux P. Exploration chirurgicale stereotaxique du lobe temporal. Paris: Manon et Cie; 1958. p. 123. 8. Crandall PH. Cortical resections. In: Engel J, editors. Surgical treatment of the epilepsies. New York: Raven Press; 1987. p. 377-404. 9. Talairach J, Ajuriaguerra JD, David M. A propos des coagulations therapeutiques sous-corticales: etude topographique du systeme ventriculaire en fonction des noyaux gris centraux. Presse Medicale 1950;58:697-701. 10. Talairach J, Tournoux P. Coplanar stereotaxic atlas of the human brain: 3-dimensional proportional system: an approach to cerebral imaging. Stuttgart: Thieme Medical Publishers; 1988. 11. Szikla G, Bouvicr G, Hori T, Petrov V. Angiography of the human brain cortex. New York: Springer-Verlag; 1977.
12. Talairach J, Ruggiero G, Aboulker J, David M. A new method of treatment of inoperable brain tumors by stereotaxic implantation of radioactive gold: a preliminary report. Br J Radiol 1955;28:62-74. 13. Talairach J, Ajuriaguerra J de, David M. Etudes stereotaxiques des structures encephaliques profondes chez I’Homme Technique, interet physiologique et therapeutique. Presse Med 1952;28:605-9. 14. Talairach J, David M, Tournoux P, et al. Atlas d’anatomie stereotaxique des noyaux gris centraux. Paris: Masson; 1957. 15. Talairach J, Szikla G, Tournoux P, et al. Atlas d’anatomie stereotaxique du telencephale. Paris: Masson; 1967. 16. Benabid AL, Chirossel JP, Mcrcier C, et al. Removable, adjustable and reusable implants for stereotactic interstitial radiosurgery of brain tumors. Appl Neurophysiol 1987;50:278-80. 17. Szikla G, Peragut JC. Irradiation interstitielle des gliomes. In: Constans JP, Schliengcr M, editors. Radiotherapie des tumeurs du systeme nerveux central. Neurochirurgie (Suppl) 1975;21:187-228. 18. Bancaud J, Talairach J, Bonis A, et al. Stereo-electroencephalographie dans l’epilepsie. Paris: Masson; 1965. 19. Bouvier G, Saint Hilaire JM, Giard N, et al. Depth electrode implantation at Notre-Dame Hospital, Montreal. In: Engel J Jr, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1987. p. 589-94. 20. Olivier A. Double-headed stereotaxic carrier apparatus for insertion of depth electrodes. J Neurosurg 1986;65: 258-9. 21. Peters TM, Clark JA, Olivier A, et al. Integrated stereotaxic imaging with CT. MR imaging, and digital subtraction angiography. Radiology 1986;161:821-6. 22. Sedan R, Duparet R. Stere´ometre adaptable au cadre ste´re´otaxique de J Talairach. Neurochirurgie 1968;14: 577-82. 23. Scerrati M, Fiorentino A, Fiorentino M, Pola P. Stereotaxic device for polar approaches in orthogonal systems (technical note). J Neurosurg 1984;61:1146-7. 24. Colombo F, Angrilli F, Zanardo A, et al. A universal method to employ CT scanner spatial information in stereotactic surgery. Appl Neurophysiol 1982;45:352-4. 25. Benabid AL, I.avallee S, Hoffmann D, et al. Computer driven robot for stereotactic neurosurgery, In: Kelly P, Kail A, editors. Computers in stereotactic neurosurgery. Cambridge. MA: Blackwell; 1992. p. 330-42. 26. Goerss SJ, Kelly PJ, Kail BA, Alker GJ. A computed tomographic stereotactic adaptation system. Neurosurgery 1982;10:375-9. 27. Kail BA, Kelly PJ, Goerss SJ, Earnest F IV. Crossregistration of points and lesion volumes from MR and CT. Proceedings of the Seventh Annual Meeting of Frontiers of Engineering and Computing in Health Care. 1985; 935-42.
The talairach stereotactic system
28. Kelly PJ, Kail BA, Goerss SJ. Transposition of volumetric information derived from computed tomography scanning into stereotactic space. Surg Neurol 1984;21:465-71. 29. Lavallee S. Gestes medico-chirurgicaux assiste´s par ordinateur. The`se sciences mathe´matiques. Grenoble. France: University Joseph Fourier; 1989. 30. Steinmetz H, Fiirst G, Freund HJ. Cerebral cortical localization: application and validation of the proportional grid system in MR imaging. J Comput Assist Tomogr 1989;13:10-19. 31. Brown RA. A computerized tomography-computer graphics approach to stereotactic localization. J Neurosurg 1979;50:715-20. 32. Mundinger F, Birg W, Klar M. Computer-assisted stereotactic brain operations by means including computerized axial tomography. Appl Neurophysiol 1978;41:169-82. 33. Cloutier L, Nguyen DN, Ghosh S, et al. Simulator allowing spatial viewing of cerebral probes by using a floating
34.
35.
36.
37.
38.
32
line concept. Symposium on Optical and Electro-Optical Applied Science and Engineering. Cannes, France. 1985. Szikla G, Bouvier G, Hori T. Localization of brain sulci and convolutions by arteriography: a stereotactic anatomo-radiological study. Brain Res 1975;95:497-502. Benabid AL, Lavallee S, Hoffmann D, et al. Computer support for the Talairach system. In: Kelly P, Kall A, editors. Computers in stereotactic neurosurgery. Cambridge, MA: Blackwell; 1992. p. 230-45. Nguyen JP, Van Effentere R, Fohanno D, et al. Methode pratique de reperage spatial des petites neoformations intracraniennes a partir des donnees de la tomodensitometrie. Neurochirurgie 1980;26:333-9. Schad L, Loll S, Schmitt F, et al. Correction of spatial distortion in MR imaging: a prerequisite for accurate stereotaxy. J Comput Assist Tomogr 1987;11:499-505. Wyper DJ. Turner JW. Patterson J, et al. Accuracy of stereotactic localisation using MRI and CT. J Neural Neurosurg Psychiatry 1986;49:1445-8.
509
37 BrainLab Image Guided System J. F. Fraser . T. H. Schwartz . M. G. Kaplitt
Stereotaxic neurosurgery has always been somewhat dependent upon new technologies. The ability to safely navigate to and alter the physiology of deep brain structures that are inaccessible by open surgery invariably benefits from technologies which combine maximal information regarding functional neuroanatomy with advanced three dimensional navigational tools. From the development of the first modern stereotaxic frame through incorporation of detailed anatomical targeting with CT and MRI, stereotaxic neurosurgery has benefited from early adoption of new technologies. In recent years, this has resulted in widespread use of computer-based image analysis and navigational guidance systems among even more seasoned practitioners who had long relied on homegrown methods. Although several commercial packages are currently available, and an academic manuscript is not intended to promote a particular vendor, nonetheless detailed evaluations of the major systems can provide valuable information to investigators who are considering entering this field as well as to those experienced surgeons who may be less familiar with the details of each system. Here we will review the history and current applications of the technology offered by one of the popular current vendors, BrainLab.
Neuronavigation: History, Principles, and Practice Although details of the history of stereotaxic surgery are likely to be reviewed elsewhere in this volume, understanding the tenets that guide the use and development of modern
#
Springer-Verlag Berlin/Heidelberg 2009
neuronavigational tools such as BrainLab can benefit from a brief review of historical context. In the mid-twentieth century, stereotaxy in neurosurgery was focused upon attention to the detail of anatomical relationships among deep brain structures. For example, such anatomical references as the Schaltenbrand-Bailey stereotactic atlas provided a platform for overlaying an early neuronavigational atlas on planning of stereotactic procedures [1,2]. Use of atlases relied upon relative preservation of anatomical relationships from patient to patient, and could, at best, estimate the position of deep brain structures by applying the atlas to standard outer landmarks. However, atlas-based stereotaxy underscores the importance of mastering anatomical relationships. Stereotactic neurosurgery, therefore, requires not only a knowledge of the exact location of a target in the brain, but a thorough appreciation for important structures that surround that target in threedimensional space. To improve the accuracy of stereotactic targeting based upon more individual patient data, the first real-time image-guidance systems in stereotactic neurosurgery developed based upon ventriculography. Injection of air or contrast into the ventricular system provides an outline of the borders of the ventricles on plain radiographs and fluoroscopy, which can yield a real-time image of deep brain anatomical relationships. Prior to computed tomography and magnetic resonance, combining this imaging with atlasbased relationships of targets to standard intracranial landmarks could be used to derive the relative locations of anatomical structures [3]. More importantly, this development represented
568
37
Brainlab image guided system
a method for obtaining real-time intracranial anatomical information. Neurosurgeons could use imaging of each patient’s intracranial anatomy to guide stereotactic procedures, establishing the principle of precision in image-guidance. Although this technique was somewhat morbid, there are those who still believe that this has a role in certain procedures as a direct, real-time assessment of intracranial anatomy. An example of an ongoing application in some centers is for accurate placement of Ommaya reservoir intraventricular catheters, although rare practitioners continue to use this for functional procedures as well [4]. Computed tomography and magnetic resonance provided obvious advantages for modern neuronavigation. Rather than extrapolating the location of targeted structures from invasive radiographic studies, relevant neuroanatomy can be directly visualized. The progressively improving resolution of CT and MRI, combined with incorporation of fine-cut preoperative framebased imaging protocols, has provided a direct targeting methodology for stereotactic neurosurgery. In directly selecting the target on an MRI image, a neurosurgeon can plan trajectories, approaches, and compute relative distances and angles. Indeed, there have been multiple studies attempting to evaluate the relative accuracy of MRI-based direct targeting and ventriculography or frame-based indirect targeting [5–7]. Interestingly, much of the information obtained from earlier studies using fluoroscopy has been retained and refined in the era of CT and MRI, such as the location of functional targets in relation to landmarks such as the anterior and posterior commissures. This highlights one encouraging feature of stereotaxic neurosurgery, which is that useful methodologies are often retained in some form even as technology advances. Against this background, intraoperative neuronavigational guidance tools have developed. There have been two major advances over the years in computer-based imaging technology which now
offer novel tools for surgeons using both framebased and frameless navigational technology. For frame-based stereotaxic surgery, the use of computer planning has several advantages. There is a significant time-saving when frames can be automatically identified and registered from source images as compared with manual calculations. The ability to simultaneously visualize various image data sets as well as reconstructed probe views can facilitate optimized trajectory planning. Accurate fusion of different data sets, such as CT and MRI, can permit advanced creation of stereotactic plans and also can combine unique advantages of different modalities in a single plan to improve accuracy. Of course, the advent of frameless stereotaxy is entirely novel and dependent upon high-level computer reconstruction technology. This provides important information for general neurosurgery, radiosurgery and offers new opportunities and challenges for stereotaxic neurosurgery. In considering these roles, we present applications for each using the BrainLab neuronavigation technology. While several corporations produce neuronavigational equipment, the history and development of BrainLab as a company mirrors the advances of neuronavigational devices and principles in the last two decades. Specific features of the BrainLab system as currently configured for both frame-based and frameless procedures will also be outlined for the benefit of those will little or no experience with this particular technology.
History and Technology of BrainLab BrainLab was founded in Munich, Germany in 1989, BrainLab primarily as a software company dedicated to generating neuronavigational systems which would be more accurate and user-friendly for general neurosurgical practice. In the company timeline, 1990 saw the first use of BrainLab software at the University of Vienna, while in 1994
Brainlab image guided system
BrainLab USA was initiated. The first VectorVision navigation system was cleared for use by the FDA in 1997, while the first BrainSUITE was sold in 2004. Over the last 18 years, BrainLab has rapidly become a multi-national company providing neuronavigational software to 15 countries. Currently over 2,000 hospitals are using one or more BrainLab systems throughout the world. From a developmental perspective, BrainLab started by producing a mouse-controlled, menudriven software application for neurosurgical planning (> Figure 37-1). By 1998, BrainLab was also producing the hardware needed for several applications in neurosurgery; it had introduced an integrated radiosurgery system, as well as the mobile cranial planning and navigation unit (> Figure 37-1). Since the release of optical tracking mobile units, neuronavigation has . Figure 37-1 The VectorVision2 mobile optical tracking unit
37
allowed direct integration of preoperative radiographic information into the real-time operative environment [8,9]. By 2003, BrainLab had begun integrating the hardware into operating room design, introducing a comprehensive BrainSUITE, which features the cranial planning/ navigation software and hardware, as well as an incorporated intraoperative MRI. BrainLab has also helped enable new imaging applications for neurosurgical intervention. Starting in 2005, BrainLab’s systems incorporated fiber tracking into their software packages, allowing this technology to become a more user-friendly tool for preoperative and intraoperative technical planning without the need for highly specialized physicists and technicians (> Figure 37-2). Since that time, tracking of important neurological pathways preoperatively has been popularized
569
570
37
Brainlab image guided system
. Figure 37-2 An example of fiber tracking with DTI (diffusion tensor imaging) incorporated into the neuronavigational plan on the BrainLab screen
in the resection of intra-axial tumors [10,11]. Finally, in 2006 iPlan Net was introduced, which is a network-based system that facilitates immediate image transfer between different sites, remote preoperative planning, and direct feeds of imaging and planning information into the operative setting. In addition to preoperative image planning, BrainLab also provides hardware that allows intraoperative registration of tools to permit realtime navigation during surgery. The optical tracking system, combined with hardware attachments, allows instruments to be converted into cable-free, battery-free probes that can be seen on the intraoperative displays. The software package also allows integration and image fusion of multiple imaging modalities [12]. These hardware and software features are essential for frameless stereotactic
work; it allows the surgeon to see the trajectory and depth of an instrument relative to its anatomical target. The system also allows tracking using neurosurgical microscopes from most major manufacturers, with optional heads up displays to place navigational information directly into one of the microscope oculars if this is desired. This even uses the focal length of the microscope to identify the location of the area being visualized at any given time. BrainLab provides various modular systems which can be used for particular applications, and they can also be combined into a more comprehensive system which provides an integrative approach to planning and navigation in stereotactic and functional neurosurgery. Although certain hardware is needed for these various procedures, software development has been the major focus of
Brainlab image guided system
BrainLab over the years, reflecting the importance of superior software to the ultimate value of computer-based imaging technologies. Below we present several BrainLab technologies which are currently commercially available along with examples highlighting how they can be applied to the advantage of the practicing neurosurgeon.
Applications Frame-based Targeting Frame-based targeting is not a novel concept in neurosurgery; its efficacy and precision for deep brain structures has been well-established. However, pre- and peri-operative trajectory planning was previously based on hand-calculation of coordinates from either known landmarks (anterior and posterior commissures) or standardized atlases. While these techniques are still very applicable today, and can often serve as methods of verification for a selected plan, image fusion and computer-based trajectory design have added a novel dimension to functional neurosurgical preparation. iPlan stereotaxy was designed specifically for stereotactic targeting and trajectory planning for both functional neurosurgical procedures and stereotactic biopsies. Information on all major stereotactic frames is contained within the package, so registration is almost immediate once the images are loaded and the type of frame and localizer (CTor MRI) is identified for the system. After the frame is registered, the anterior and posterior commissures and midline are all identified, although for direct targeting for a biopsy or if this information is not necessary for some other reason, this step can be bypassed. The intercommisural distance is provided, which can help to evaluate the validity of the data. Multiple data sets can then be merged simultaneously, such as multiple MRI sequences as well as CT scans. Even functional MRI or diffusion tensor imaging (DTI) data can now be input for both functional as well
37
as anatomical planning. With this feature, thin cut slabs in multiple orientations can be merged with larger data sets, thereby reducing the MRI time for the patient. Small deep nuclei such as the subthalamic nucleus can now be visualized and targeted using MR technology. With the BrainLab iPlan stereotaxy planning software, a patient may have an MRI prior to the day of surgery without a frame. On the day of surgery, a frame is applied to the patient, who subsequently undergoes a thin-slice CT scan. The CT and MRI images are fused using the BrainLab planning software, and both sets of images can be used to perform the surgery. Alternatively, the coordinates for the target(s) can be obtained from the planning software, negating the need for an active neuronavigation system in the operating room. The iPlan Net can be of particular utility as an adjunct to frame-based stereotactic planning. DICOM images from any modality (MRI, CT and even PET) can be immediately ‘‘pushed’’ to the iPlan Net server by the radiology technician, in a manner identical to sending images to either a PACS or other network server. The server is simply a computer which can accept networked images, and which runs the various software packages, so that it can be placed at any remote site which is accessible to a particular network. The system is both HIPAA compliant and can be configured as a ‘‘one-way’’ server, such that images can be pushed to the iPlan Net server directly and not through the PACS system, but they cannot be sent back to the radiology PACS system. This feature can help allay any concerns by radiology information technology personnel regarding possible unauthorized entry into the PACS clinical imaging system. Since the iPlan net server can run the iPlan stereotaxy software package, image analysis and all planning can be performed online from any computer networked to the iPlan Net server. With the iPlan Net system, the surgeon is freed from the obligation to plan only at a single workstation, and planning can be completed from
571
572
37
Brainlab image guided system
any office or even from home (if this is in fact desirable). If planning from a preoperative MRI is completed in advance, then the image obtained after the frame is placed (MRI or CT) can be sent to the server and is available as soon as the surgeon reaches any networked computer. The images can then simply be merged, and, within a few minutes after obtaining the post-frame image, final frame coordinates are obtained to accelerate initiation of the surgical procedure. The same plan can also be projected from a networked computer in the operating room to a large screen monitor, so that the images and stereotactic information can be both displayed and modified in the operating room without the need for specific workstations or equipment. We find this feature of the iPlan Net to be a particularly useful teaching tool, as anatomical relationships and various considerations when planning surgery can be displayed anywhere that a computer can be networked to the server, such as in teaching or case conferences or in small groups within the neurosurgical offices.
Deep Brain Stimulation for Parkinson’s Disease The iPlan stereotaxy package was designed for use in deep brain stimulation (DBS) surgery, although it would be equally and readily applicable to emerging technologies such as infusion of biological agents (growth factors, gene therapy and cell-based therapies). We routinely utilize the iPlan stereotaxy with the iPlan net server for such procedures, for example targeting the subthalamic nucleus (STN) for DBS in Parkinson’s disease. In advance of surgery, we obtain an outpatient 3 Tesla MRI for pre-surgical planning. Fine cut T2-weighted axial sections are obtained in a single acquisition from just below the STN to the top of the head (usually the maximum possible in a single acquisition is just over 60 slices). Multiple acquisitions to obtain more images should be avoided, as they often are not perfectly
aligned and reconstruction results in staggered, jagged images. A second set of fine-cut, coronal T2-weighted images are obtained as a slab just around the level of the STN. Although we do not obtain contrast images, a third set of double-dose T1-weighted contrast images could be obtained and incorporated as another data set, as is the custom in some practices. These images are immediately pushed to the iPlan net server, where they are available at any time to perform target and trajectory planning. We routinely complete this from our personal office computer in advance of surgery. After merging the axial and coronal images and identifying AC, PC and midline, the STN target is then chosen. This can be determined by direct visualization of the axial and/or coronal images, indirect targeting based upon the distance in all three coordinates from the mid-commissural point (MCP) (or from AC or PC, if preferred) or based upon the position on the SchaltenbrandWharen atlas, which can be overlaid and manually adjusted to the MRI images. We routinely target initially based upon a distance from the MCP (> Figure 37-3a), then adjust that position after visually determining the location relative to nearby structures such as the red nucleus, substantia nigra and internal capsule. In general we use the atlas as a teaching tool, although occasionally an adjustment is made based on this information as well (> Figure 37-3b). The entry point can then be determined based upon visual targeting or based upon angles relative to midline and to the intercommissural line (> Figure 37-3c). This can then be adjusted manually, and two obliqued images are provided which are cut along the axis of the trajectory to visualize the entire tract in one image (> Figure 37-3d). Scrolling along the trajectory can also be performed to demonstrate the location on true axial or coronal images at each point along the tract. The trajectory can be further adjusted from this view to immediately avoid an undesirable structure without having to guess as to a change and then re-check the
Brainlab image guided system
37
. Figure 37-3 (a). The AC-PC View allows identification of the anterior commissure, posterior commissure, and mid-sagittal planes. From these, the mid-commissural point (MCP) may be derived. (b) The Atlas View overlays the SchaltenbrandWahren atlas, which can be adjusted to the actual anatomy of the patient, onto the MRI, and permits its use as a reference. (c) The Trajectories Overview is a 3-orientation visualization at a particular position (in this case, the target is the STN). (d) The Probe View provides two different obliqued views in the plane of the trajectory to see what is being traversed in a single image. By scrolling through the images, the user can understand the structures that the trajectory will traverse. In this case, the solid pink line represents the selected trajectory for scrolling, while the dotted green line represents another trajectory in the plan
573
574
37
Brainlab image guided system
. Figure 37-3 (Continued)
plan. The final plan is saved on the server for future access. On the day of surgery, a volumetric, fine cut non-contrast CT is obtained after placement of
the stereotactic frame. While the patient is brought to the OR, the images are sent directly to the server, where they can be uploaded into the folder with the prior plan from any networked
Brainlab image guided system
computer either in or out of the operating room. The frame is identified and the fiducials automatically register (> Figure 37-4a). A threedimensional reconstruction of the localizing box with the fiducial bars provides added confirmation as to the accuracy of the registration. Although we prefer a CT due to the additional accuracy and minimal scan time, an MRI performed with the frame in place can also be used as the software package contains information on MRI as well as CT localizers for most popular frames. The frame-based image is then merged with the preoperative plan, and stereotaxic coordinates are generated for the particular frame (> Figure 37-4b). We utilize the Leksell frame, and the software generates coordinates for all four possible orientations of the frame, along with a picture of the frame in the orientation for the selected set of coordinates. This can be a useful feature if different orientations are routinely used, but for centers which use primarily one orientation, care should be taken to ensure that the correct set of coordinates are used since there is currently no option available to remove the irrelevant coordinates. Entry angles are also provided to facilitate passage along the exact trajectory planned to target. We project this data to a screen in the operating room from a networked computer. The data is also output into a formatted report as a PDF file for easy printing. This report also provides coordinates for all four orientations, which can create confusion, so we blacken the three irrelevant sets. The report also includes other useful information, such as the intercommissural distance and location of the target relative to MCP. This report can be printed for use in the OR and also is an excellent source for documentation in the patient chart. Intraoperatively, notes can be generated at various points along the trajectory, documenting such things as patient responses or electrophysiological data. There is the option to output data from various electrophysiological recording systems into the BrainLab system to document
37
actual recording data at different points along the trajectory. Another useful application of this technology is for post-operative analysis of electrode placement. We routinely perform a post-surgical MRI in the axial and coronal planes, however in some patients only a CT scan is possible (if there is a pacemaker in place, for example). Either of these can be merged with the preoperative data sets, and the planned trajectory and target localization can be compared with the actual electrode location. This can be particularly useful when analyzing images which have not been symmetrically cut. Electrodes which may not appear to be ideal on an image which actually obliqued often are clearly well-placed when analyzed in this manner. It should be noted that the iPlan Net server is not obligatory to utilize all of the features outlined above. Stand alone workstations are also available which run this software, and even if the server is utilized, at least one such workstation should still be obtained as a backup in case of a server failure. Images are then loaded onto the workstation using standard tapes or USB drives.
Frameless Targeting While using a fixed frame allows instruments to be steadily oriented to a selected target, hardware innovation in neuronavigation has created an alternative to frame-based stereotactic surgery. Optical tracking systems such as the VectorVision2 (BrainLab), can track any surgical instrument that is registered with attached fiducials. For example, a suction catheter or bipolar cautery instrument may be tracked by attaching a ‘‘star,’’ that maintains three fiducial markers in a fixed relationship. The instrument is established in the neuronavigation virtual environment using a mobile registration unit, commonly called ‘‘the elephant’’(> Figure 37-5). The instrument then becomes a de facto probe, allowing
575
576
37
Brainlab image guided system
. Figure 37-4 The planning software can accept many different frames based on both CT and MRI. (a) The Frame Localizer shows how the frame is localized and identified to the machine. (b) A CT and noncontrast T2-weighted MRI undergo image fusion. In the upper left panel, the auto-fusion of images may be checked by manipulating the window shown to verify that anatomical structures match between the two imaging modalities
Brainlab image guided system
. Figure 37-5 Registration of a biopsy instrument using the mobile registration unit ("the elephant")
simultaneous neuronavigation and progression of the operative plan. Through such operational flexibility, frameless stereotaxy has been successfully used for both functional (deep brain stimulation) and oncologic (tumor biopsy) neurosurgery [13–15]. With regard to frameless functional neurosurgical cases, the planning and targeting software described above is used in the same fashion. However, rather than generating frame-based coordinates, frameless navigation is used to orient either the microelectrode or permanent DBS electrode guide tubes to match the planned trajectory and then pass to the desired target. For these cases, to achieve accuracies which approach frame-based targeting, fiducials which are fixed to the skull (so-called ‘‘bone fiducials’’) are used, as compared with skinbased fiducials which usually suffice for oncological and other procedures which don’t require the same level of accuracy. However, not all instruments require ‘‘star’’ attachments and intraoperative registration. BrainLab produces particular hardware items that are pre-registered with already-attached fiducials. Such instruments are easily integrated into the neuronavigational environment, where their standardization and fixed fiducial relationships help to ensure accuracy and precision. Of particular
37
note, BrainLab produces a frameless biopsy system, consisting of a pre-calibrated biopsy needle, precalibrated biopsy alignment sheath, and a frameless biopsy arm that attaches to a standard Mayfield head holder (> Figure 37-6a). Once assembled, the system allows the surgeon to alter trajectory and entry point in real-time from virtually any angle (> Figure 37-6b). In addition to tumor biopsy, frameless stereotaxy can also be utilized for placement of intracranial monitoring electrodes in cases of intractable epilepsy. Typically, a patient will undergo a craniotomy, and strips of electrodes will be layed onto the brain in the subdural space. In addition, depth electrodes will be placed in deep temporal structures: the hippocampus and deep temporal gyri. Although the surface gyri are directly visible within the craniotomy window, the depth electrodes must be accurately placed to avoid injury to the brainstem, thalamus, and other deep brain structures. An example of this technology is demonstrated in > Figure 37-7. In this setting, frameless stereotactic guidance can be helpful in depth electrode placement [16,17].
Intraoperative MRI Intraoperative MRI technology represents a fusion of preoperative neurosurgical imaging with intraoperative planning and trajectory adjustment based. Desired as a tool for intraoperative assessment of surgical success and changes in intracranial anatomy, intraoperative MRI’s were introduced in the last decade of the twentieth century. Initial attempts to incorporate MR imaging into the operative environment were mixed, due to the low-field magnets used (0.5 Tesla) and the long transition times required. With the incorporation of neuronavigational units, the advancement of magnet strength (to 1.5 and 3.0 Tesla), and the establishment of ergonomic operative suites (set up specifically for this purpose), intraoperative MR imaging is rapidly
577
578
37
Brainlab image guided system
. Figure 37-6 The BrainLab stereotactic biopsy components (a) are integrated to allow intraoperative adjustment of the entry point and trajectory (b)
Brainlab image guided system
37
. Figure 37-7 Intraoperative screenshots of neuronavigation software demonstrating proposed trajectories for depth electrode placement for analysis of seizure foci
gaining popularity [18–21]. In the BrainSUITE setting, the intraoperatively obtained images are easily integrated into the preoperative neuronavigational plan. Logistically, either the operating table or the MR scanner move into and out of the scanning position in an ergonomicallydesigned fashion. The neuronavigation software allows fusion of intraoperative and preoperative data, which provides the advantage of real-time adjustments based upon contemporaneous intraoperative imaging data with the speed of navigation based upon previously acquired data. This
also provides feedback as to the current location relative to the original and presumably desirable plan. Currently, this technology is being applied to the resection of intra-axial brain tumors, transsphenoidal resection of pituitary tumors, placement of deep brain stimulation electrodes, and epilepsy surgery [22–26]. Objective data on the relative utility of intraoperative MRI for each of these applications is being collected, and its role in neurosurgical procedures has yet to be fully elucidated. Clearly this is a major undertaking for any center, both in terms of cost, space and logistics to
579
580
37
Brainlab image guided system
manage an MRI environment within an operating room setting. Nevertheless, the BrainSUITE platform provides an efficient method for obtaining and integrating intraoperative imaging data into the surgical plan. The features of the BrainSUITE that are designed to optimize neurosurgical efficiency and ease of use may be attractive to those centers with the means and space to consider intraoperative MRI imaging. In the modern neurosurgical era, neuronavigation plays a vital role in oncologic, stereotactic, and functional neurosurgery. It allows the surgeon to preoperatively plan complex operative approaches, to target deep brain nuclei for biopsy, stimulation, or lesioning, and to individualize the operative plan to each patient’s specific intracranial anatomy. Successful neuronavigation depends upon the principles of accurate and relevant imaging, precise correspondence between images and patient anatomy, and appropriate incorporation of navigational tools into the operative protocol. Neuronavigational devices will never substitute for a vital three-dimensional understanding of neuroanatomy, but can serve to aid the neurosurgeon in accurate targeting and planning. Continued advances in both software and hardware by companies such as Brainlab, with a particular focus upon the needs and constraints of the neurosurgical environment, will continue to facilitate otherwise difficult functional and stereotactic neurosurgical procedures.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
References 15. 1. Kazarnovskaya MI, Borodkin SM, Shabalov VA, Krivosheina VY, Golanov AV. 3-D computer model of subcortical structures of human brain. Comput Biol Med 1991;21(6):451-7. 2. Housepian EM. Stereotactic surgery: the early years. Neurosurgery 2004;55(5):1210-4. 3. Slaughter DG, Nashold BS, Jr. Intracranial measurements for stereotactic surgery. Confin Neurol 1970;32(2):250-4. 4. Sandberg DI, Bilsky MH, Souweidane MM, Bzdil J, Gutin PH. Ommaya reservoirs for the treatment of
16.
17.
leptomeningeal metastases. Neurosurgery 2000;47 (1):49-54; Breit S, LeBas JF, Koudsie A, et al. Pretargeting for the implantation of stimulation electrodes into the subthalamic nucleus: a comparative study of magnetic resonance imaging and ventriculography. Neurosurgery 2006;58 Suppl 1:ONS83-95. Slavin KV, Thulborn KR, Wess C, Nersesyan H. Direct visualization of the human subthalamic nucleus with 3T MR imaging. Am J Neuroradiol 2006;27(1):80-4. Schlaier J, Schoedel P, Lange M, et al. Reliability of atlas-derived coordinates in deep brain stimulation. Acta Neurochir (Wien) 2005;147(11):1175-80; discussion 1180. Gumprecht HK, Widenka DC, Lumenta CB. BrainLab vectorvision neuronavigation system: technology and clinical experiences in 131 cases. Neurosurgery 1999;44 (1):97-104; discussion 104–105. Mascott CR. In vivo accuracy of image guidance performed using optical tracking and optimized registration. J Neurosurg 2006;105(4):561-7. Berman JI, Berger MS, Chung SW, Nagarajan SS, Henry RG. Accuracy of diffusion tensor magnetic resonance imaging tractography assessed using intraoperative subcortical stimulation mapping and magnetic source imaging. J Neurosurg 2007;107(3):488-94. Bello L, Gambini A, Castellano A, et al. Motor and language DTI fiber tracking combined with intraoperative subcortical mapping for surgical removal of gliomas. Neuroimage 2008;39(1):369-82. Schlaier JR, Warnat J, Dorenbeck U, Proescholdt M, Schebesch KM, Brawanski A. Image fusion of MR images and real-time ultrasonography: evaluation of fusion accuracy combining two commercial instruments, a neuronavigation system and a ultrasound system. Acta Neurochir (Wien) 2004;146(3):271-6; discussion 276–277. Eljamel MS, Tulley M, Spillane K. A simple stereotactic method for frameless deep brain stimulation. Stereotact Funct Neurosurg 2007;85(1):6-10. Pan HC, Wang YC, Lee SD, Chen NF, Chang CS, Yang DY. A modified method to perform the frameless biopsy. J Clin Neurosci 2003;10(5):602-5. Holloway KL, Gaede SE, Starr PA, Rosenow JM, Ramakrishnan V, Henderson JM. Frameless stereotaxy using bone fiducial markers for deep brain stimulation. J Neurosurg 2005;103(3):404-13. Mehta AD, Labar D, Dean A, et al. Frameless stereotactic placement of depth electrodes in epilepsy surgery. J Neurosurg 2005;102(6):1040-5. Murphy MA, O’Brien TJ, Cook MJ. Insertion of depth electrodes with or without subdural grids using frameless stereotactic guidance systems–technique and outcome. Br J Neurosurg 2002;16(2):119-25.
Brainlab image guided system
18. Maciunas RJ, Dean D, Lewin J, Selman WR, Ratcheson RA. Integration of neurosurgical image guidance and an intraoperative magnetic resonance scanner. The University Hospitals of Cleveland experience. Stereotact Funct Neurosurg 2003;80(1–4):136-9. 19. Samdani A, Jallo GI. Intraoperative MRI: technology, systems, and application to pediatric brain tumors. Surg Technol Int 2007;16:236-43. 20. Yrjana SK, Tuominen J, Koivukangas J. Intraoperative magnetic resonance imaging in neurosurgery. Acta Radiol 2007;48(5):540-9. 21. Jones J, Ruge J. Intraoperative magnetic resonance imaging in pituitary macroadenoma surgery: an assessment of visual outcome. Neurosurg Focus 2007;23(5):E12. 22. Lee MW, De Salles AA, Frighetto L, Torres R, Behnke E, Bronstein JM. Deep brain stimulation in intraoperative MRI environment – comparison of imaging techniques
23.
24.
25.
26.
37
and electrode fixation methods. Minim Invasive Neurosurg 2005;48(1):1-6. Nimsky C, Ganslandt O, von Keller B, Fahlbusch R. Intraoperative high-field MRI: anatomical and functional imaging. Acta Neurochir Suppl 2006;98:87-95. Nimsky C, von Keller B, Ganslandt O, Fahlbusch R. Intraoperative high-field magnetic resonance imaging in transsphenoidal surgery of hormonally inactive pituitary macroadenomas. Neurosurgery 2006;59(1):105-14; discussion 105–114. Nimsky C, Fujita A, Ganslandt O, von Keller B, Kohmura E, Fahlbusch R. Frameless stereotactic surgery using intraoperative high-field magnetic resonance imaging. Neurol Med Chir (Tokyo) 2004;44(10):522-33; discussion 534. Hall WA, Truwit CL. Intraoperative MR-guided neurosurgery. J Magn Reson Imaging 2008;27(2):368-75.
581
47 Comprehensive Brain Tumor Management M. Tamber . M. Bernstein
Introduction
Provision of a Histologic Diagnosis
The contemporary management of patients with primary intra-axial brain tumors is multidisciplinary. Surgery followed by radiotherapy was until fairly recently the principal treatment, but provides only palliative benefit. Historically, chemotherapy has been relegated to an adjuvant role, but recent developments have validated certain chemotherapeutic agents as important tools in our therapeutic armamentarium, providing a meaningful, albeit modest, improvement in survival. Perhaps the most exciting areas of advance lie in the genetic targeting of conventional treatments and in the expanding use of genomics and proteomics to develop novel agents specifically directed at critical molecular targets within the cell [1]. In this chapter, we summarize the current status and future prospects for each of the treatment modalities for brain tumors. We tailor our discussions in a relatively generic sense around gliomas, without in-depth discussion of more controversial and nuanced issues such as the dilemma over management of presumed lowgrade gliomas.
With relatively few exceptions, histologic diagnosis is imperative in most patients with an intracranial mass lesion, especially if a malignant intra-axial neoplasm is considered in the preoperative differential diagnosis. An accurate tissue diagnosis not only excludes the possibility of another diagnosis which may require a different treatment paradigm altogether (e.g., cerebral abscess), but offers critical information regarding the histological type and grade of the lesion which will influence decisions on further management and provide important guidance regarding prognosis. A histological diagnosis of the lesion may be achieved via craniotomy and open or ‘‘excisional’’ biopsy, or by a minimally-invasive, imageguided stereotactic procedure. The latter may be frame-based or frameless. The diagnostic accuracy of various biopsy procedures is proportional to the amount of tissue obtained and the accurate targeting of areas within the lesion of potentially high diagnostic yield. Although it provides less tissue for examination than an open excisional biopsy (with the attendant increased risk of sampling error), stereotactic/image-guided biopsy can largely replace craniotomy for the purposes of histological diagnosis with similar efficacy and comparable morbidity. In other words, craniotomy is not mandatory for the histological diagnosis of brain tumors. In fact, one can envision several circumstances where an image-guided biopsy would be the preferred means of obtaining tissue, such as
Role of Surgery Surgery has the distinct but interrelated aims of providing a histological diagnosis to guide further therapy, the relief of mass effect, cytoreduction, and facilitation of the local delivery of adjuvant therapy. Each of these roles is discussed in turn. #
Springer-Verlag Berlin/Heidelberg 2009
736
47
Comprehensive brain tumor management
when the therapeutic benefit from craniotomy and excisional biopsy is too small to justify the risks, or, conversely, if the risk of craniotomy, regardless of the potential benefit, is felt to be prohibitive. When deciding to perform a stereotactic biopsy, one must consider both the lesion and the patient. The ideal lesion for stereotactic biopsy (versus craniotomy) is a small deep lesion or one located in eloquent cortex for which the risk of craniotomy and excisional biopsy is felt to be prohibitively high. Implicit in this is that cytoreduction is not required and/or could not be done effectively or safely due to the diffuse and indistinct nature of thee tumor (> Figure 47-1). Other characteristics of lesions suited for stereotactic biopsy are diffuseness (e.g., across corpus callosum) and multiplicity (i.e., more than one lesion in disparate parts of the brain). Concerning the patient, sometimes an individual who, under normal circumstances, would benefit from an aggressive resection is better served by a stereotactic biopsy because of significant medical illness or very advanced age. Conversely, if the patient is young and neurologically intact, the risk of neurological
morbidity from craniotomy can be avoided either at the surgeon’s or the patient’s discretion in specific cases. The important issues to examine regarding stereotactic biopsy, after patient selection, are the success rate and the complication rate. The success rate (i.e., the probability of obtaining a positive and definitive diagnosis from the procedure) varies in the literature. Based on a meta-analysis of over 4,000 stereotactic biopsies reported in the literature, one can conclude that a positive diagnosis should be attainable in excess of 90% of cases [2]. In the case of neoplastic disease, it is unusual to miss the diagnosis entirely, although it is not uncommon to miss the area of worst grade within a particular tumor due to sampling error (e.g., sampling only the grade II portion of an anaplastic astrocytoma). The reported complication rates of stereotactic biopsy vary considerably in the literature. The accepted complication rate appears to be up to 5%. The major complication is intracerebral hemorrhage resulting in neurological deterioration. In a recent large single-surgeon series from
. Figure 47-1 Diffuse glioma in eloquent cortex in a 40-year-old woman presenting with sensory seizures. A = T2. B = T1 with gadolinium enhancement. Awake image-guided biopsy revealed anaplastic oligodendroglioma
Comprehensive brain tumor management
the University of Toronto, comprising almost 750 biopsies, the incidence of complications (mainly intracerebral hemorrhage) is about 5%. Two percent of these have resulted in death or neurological morbidity while the other 3% resulted in minor and/or transient deficits only. An important risk factor for hemorrhage appears to be a malignant histology [3–6]. The differential survival of patients with malignant intra-axial brain tumors diagnosed with stereotactic biopsy and then irradiated, as opposed to undergoing resection plus radiation, has been the subject of a few papers [7,8]. In general, retrospective data suggests that stereotactic biopsy plus radiation was as effective as craniotomy plus radiation for selected patients with malignant tumors in treacherous locations. However, these studies suffer from critical flaws related to selection bias, as patients were not randomly assigned to the two surgical arms. Given the present state of the evidence, one cannot conclude that for any given malignant cerebral tumor, stereotactic biopsy plus radiation is better, worse, or equal to resection plus radiation in terms of length of survival, with the assumption that debulking was not required for palliation of symptoms. At present, the treatment of brain tumor patients depends principally upon the histological signature of the lesion; in the future, it is likely to depend increasingly on tumor genotyping and the molecular phenotype of the tumor, features that can only be assessed using tissue samples. As a result, it is postulated that tumor biopsy will continue to play an important role in the future management of brain tumor patients.
Surgery and Symptom Relief Tumors exert their clinical effects by two groups of mechanisms: direct invasion of nervous tissue, and various distant actions, including high intracranial pressure, seizures, hydrocephalus, and
47
compression and distortion of adjacent nervous tissue. Clearly, no surgical intervention can reverse the consequence of direct neoplastic invasion of adjacent neural structures. Surgery, however, can relieve global symptoms caused by raised intracranial pressure, such as headache, vomiting, and a general feeling of malaise that is not recognized as abnormal until after the tumor has been removed (when the patient realizes how unwell they felt before surgery). Epilepsy secondary to intra-axial brain tumors is amenable to surgical palliation in those circumstances where seizures are multiple, intrusive, and refractory to medical management alone. In addition, some focal deficits seem to respond to surgical palliation. The mechanism of such deficits is likely some combination of distortion and local compression of adjacent nervous tissue; tumor debulking effects decompression of these distorted yet viable tissues, thereby allowing partial, if not total, recovery of function [9]. Surgically remediable symptoms typically respond to steroids. Major tumor decompression can facilitate prompt reduction of steroid medication, with the avoidance of the side effects produced by their long term use at high doses. The relative ease with which intracranial pressure can be reduced and neurological symptoms alleviated in each case is balanced against the associated risks of surgery. Drainage of a cystic lesion or resection of a lesion in a non-dominant frontal or temporal lobe, for example, are relatively low risk procedures, which will produce rapid decompression and potentially some degree of neurologic recovery. Surgical decision making in this instance is clearly more straightforward than contemplating the removal of a diffuse, ill defined lesion close to an eloquent brain area. Many tumors in more eloquent cortex are still better served by resection than biopsy, especially if they are producing neurological deficit (which might be helped by surgery) and appear relatively discrete on imaging (> Figure 47-2). In short, any surgical intervention should be tailored to the patient’s symptoms,
737
738
47
Comprehensive brain tumor management
. Figure 47-2 Relatively discrete glioblastoma in eloquent cortex in a 70-year-old man. His left hemiparesis and hemisensory dysfunction resolved after awake image-guided craniotomy for gross total resection of the tumor
clinical condition and needs, prognosis, as well as to the requirements of any adjuvant treatments. In regards to this latter point, relief of mass effect related to a tumor may also serve to enhance the safety and efficacy of adjuvant treatments such as radiation, although clear evidence in support of this is lacking [10]. Although it seems intuitive that the relief of mass effect following surgery is associated with improved functional outcome (not necessarily survival), data in support of this thesis is scant. The reason behind this relative dearth of data is straightforward – it is far more methodologically challenging to measure the impact of an intervention on quality of life than it is to measure the impact of the same intervention on the same patients in terms of a ‘‘hard’’ endpoint such as prolongation of survival. Nevertheless, the observational data that does exist on the subject of palliation of symptoms following surgery seems to imply that some functional improvement after surgery may be seen in certain patients [11].
Cytoreductive Surgery and Survival Issues regarding quality of life aside, the really difficult and unresolved issue regarding the surgical treatment of malignant intra-axial brain tumors relates to the inherent value of aggressive resection in prolonging the life of the patient. Several reports have assessed the effect of surgery on survival, with most concluding that more extensive resection is associated with longer survival [12–15]. Once again, most studies are retrospective and interpretation is complicated because of selection bias. Patients presumed to have a favorable prognosis, on the basis of young age at diagnosis, absence of neurological deficits, lower grade tumor, and location remote from eloquent areas (which likely is a surrogate for the feasibility of extensive surgical resection), tend to be selected for more aggressive surgery. In studies such as these, it is difficult, if not impossible, to disentangle the effect of these important confounding variables from the effect of aggressive surgery on patient survival. Studies that use statistical modeling to account for the effect of these confounding variables tend to document more modest survival advantages than those studies that do not adjust for the effect of other important prognostic variables, with overall prolongation of survival in those patients treated with aggressive surgery ranging from none [16] to about four months [17] in adjusted analyses. Importantly, these estimates are likely optimistic, as no statistical model is able to compensate fully for other unmeasured variables that are associated with the outcome of interest. Interpretation of evidence relating the degree of resection to patient survival is further complicated by the difficulty of defining the extent of resection. The surgeon’s impression of the extent of tumor removal is unreliable beyond the gross distinction between biopsy and resection. More objective reports have sought to use
Comprehensive brain tumor management
postoperative imaging to quantitatively assess resection, but even in these studies, the results depend on multiple factors, including the imaging modality used, the degree of inter-rater reliability, and the time interval between surgery and post-operative imaging [18]. On balance, more papers report improved survival than do not; this seems to influence much of current surgical practice. However, the axiom that a large quantity of poor quality data does not necessarily replace a small amount of methodologically sound information should not be forgotten. Systematic reviews of the observational literature have repeatedly found no convincing evidence for an independent benefit from surgery [19,20]. Randomized trials have been difficult to do and are conspicuously lacking. Only one randomized study has been executed to specifically examine the role of surgery for these tumors [8]; this represented an outstanding and noble effort but, in the end, was a flawed and inconclusive study. It is quite likely that no large, properly executed randomized controlled trial will ever be successfully conducted. Neurosurgeons will have to base their judgments regarding surgery for malignant brain tumor patients on available biased nonrandomized data, personal ‘‘feelings’’ based on experience, patient preference, and their individual concept of the disease.
Surgery to Facilitate Delivery of Local Therapy The blood-brain-barrier (BBB) imposes a limiting and often impervious barrier to the delivery of conventional antineoplastic agents to the local tumor microenvironment. Over the past several years, numerous attempts to circumvent this difficulty have relied on the loco-regional administration of conventional as well as novel therapeutic agents at the time of surgical treatment. Most have met with limited success, but the delivery
47
methods, the context within which they are utilized (e.g., along with cranial radiotherapy and/or systemic chemotherapy), as well as the agents themselves, are in a process of continual evolution.
Photodynamic Therapy This technique involves intravenous injection of a photosensitizing porphyrin-based dye approximately 24 h before surgery. This dye is preferentially taken up by tumor cells. At the time of surgery, the tumor bed is illuminated with light of the appropriate wavelength; this activates the dye and kills the tumor cells that have taken up the dye, but not the surrounding brain cells. Some studies of this treatment have been published, but no good evidence of benefit has been documented [21].
Gene Therapy The Herpes Simplex Virus-Thymidine Kinase (HSV-TK) viral oncolytic system relies upon the application of mouse producer cells carrying replication-incompetent retroviruses transfected with the herpes simplex thymidine kinase gene to the resection margin at the time of surgery. Once the virus is locally administered in this way, the premise is that local infection of tumor cells occurs, such that that these cells now harbor the HSV-TK gene in their DNA complement. After systemic administration of ganciclovir, the HSV-TK gene metabolizes ganciclovir to a cytotoxic nucleotide, which induces apoptosis in rapidly dividing cells. Several phase I and phase II trials have been done documenting the feasibility of the approach and some toxicity [22]. A randomized trial of 248 patients found no benefit [23], but a small randomized trial of 36 patients has since found a survival advantage with treatment (70 weeks vs. 40 weeks survival,
739
740
47
Comprehensive brain tumor management
p = 0·001) [24]. New research with replication competent viruses and viruses of different types may produce more encouraging results. Other biologicals used intratumorally include interferon [25].
Brachytherapy Brachytherapy involves the direct implantation of sources of radiation into the tumor bed at the time of surgery or by stereotactic means shortly after resection. It is really a form of focused radiation therapy but is included here as it involves neurosurgical intervention. This treatment has the theoretical advantage of delivering high-dose radiotherapy to the tumor margin with concurrent minimization of radiation exposure to more distant tissue. Reports on the use of brachytherapy to treat gliomas first appeared in the late 1960s. The most contemporaneous evidence includes two randomized studies which suggest no significant benefit of brachytherapy in the setting of multimodality treatment of brain tumor patients with surgery, conventional radiation, and chemotherapy [26].
Chemotherapy Wafers Chemotherapy wafers are perhaps the best known surgical adjuvant treatment. Biodegradable wafers (Gliadel), which are placed in the tumor bed at the time of surgery have been developed. They are fabricated to release carmustine, a nitrosurea, in such a manner as to deliver high concentrations to the resection margin, where most recurrences occur, for several weeks. A large phase III trial with 240 patients with newly diagnosed malignant glioma randomized to receive either carmustine or placebo wafers at the time of primary surgical resection followed by radiation therapy demonstrated a modest survival benefit with median survival of 13.9 months
for the carmustine wafer-treated group and 11.6 months for the placebo treated group [27]. Similarly, in patients undergoing resection for recurrent GBM, placement of carmustine wafers only provided a modest prolongation of survival [28].
Enhancing Local Delivery A very promising approach for the delivery of drugs and other macromolecules to the brain explores the feasibility of using bulk flow within the brain extracellular fluid (ECF) space for the intracerebral distribution of agents [29]. This method has been called convection enhanced delivery (CED) and involves the placement within the brain parenchyma or tumor substance of one or more catheters that are subsequently connected to continuous infusion pumps. The pumps must be able to deliver the very low infusion rates that are critical for successful fluid ‘‘convection’’ within the brain; rates of infusion greater than a few ml/min will produce backflow along the catheter and loss of pressure, while too low a pressure will lead to failure of delivery altogether. An increasing body of animal and, more recently, human data shows that a much larger volume of distribution is achieved compared to previous delivery methods. CED is now the method of choice in a variety of phase I through phase III clinical trials investigating novel agents in CNS malignancy.
Surgical Adjuncts Although the clinical benefit of pursuing the goal of maximal tumor resection is unproven, the technology for achieving it is developing rapidly. Conventional surgery debulks tumors from within; resection is stopped when visually and palpably normal brain is reached, though this method of identifying the edge of a tumor is difficult and notoriously unreliable. Several technologies can
Comprehensive brain tumor management
assist in extending the resection margin to its maximum. Image guided surgery involves the use of preoperative imaging data to identify the extent of tumor to be resected [30]. The equipment for this technique has become standard in neurosurgical units and consequently it is the most commonly used aid. Functional MRI data and subcortical tract location data from diffusiontensor MRI can be used with image guidance to assist in the definition of eloquent areas during surgery. The advantage of frameless stereotaxis is that the position of the pointer can be continually updated during the procedure to show the position of the pointer relative to the tumor and brain based on archival information (i.e., the MRI or CT done the morning of the procedure). The disadvantage of frameless stereotaxis is that the position is updated relative to archived (i.e., old) data and is not real-time; therefore, as the position of the tumor changes with loss of CSF, tumor resection, etc., the position of the pointer does not accurately reflect position. This makes the technique inaccurate, particularly at the end of a resection when it would potentially be of most utility in locating any residual tumor. Intraoperative imaging, with ultrasound, CT, or MRI, circumvents this problem [30]. New directions in frameless stereotaxis include pointer systems which are updated to reflect real-time position based on ultrasound and other inputted data sets. Despite relatively widespread use of technologies such as these, real data quantifying the incremental value of these imaging-based surgical adjuncts, vis-a`-vis improved patient outcomes, over the expense and other drawbacks of these systems, is still forthcoming. Awake craniotomy with intraoperative cortical mapping is an excellent treatment option as an alternative to craniotomy under general anesthesia for the routine surgical management of patients with supratentorial intra-axial tumors in whom maximal safe tumor resection is desirable. The advantages of awake craniotomy
47
with cortical mapping over standard craniotomy performed under general anesthesia include the opportunity for brain mapping to identify functional cortical areas and descending subcortical motor, sensory, and language white matter tracts, in an attempt to decrease neurological complications and maximize the extent of safe tumor resection. Awake craniotomy also avoids the use of general anesthesia with its attendant morbidity. The latter is often related to invasive monitoring techniques such as arterial lines, central venous lines, and urinary catheters as well as prolonged immobilization. A recent large series shows the benefits of awake craniotomy used non-selectively for intra-axial tumors and delineates the complication profile as a function of whether positive mapping was or was not encountered [31].
Role of Radiation Therapy Radiotherapy plays a vital role in the treatment of central nervous system malignancy, secondary only to surgery. Radiation therapy has been shown to extend survival time and improve quality of life in patients with primary intra-axial brain tumors. There have been major technological advances in both the delivery of radiotherapy and in diagnostic imaging in the last 5–10 years which have refined the delivery of radiation and have opened up new possibilities for the targeted delivery of a lethal dose of irradiation to the tumor site whilst sparing normal surrounding brain.
Conventional Radiation Post-operative external beam radiotherapy is well supported by randomized studies and remains standard therapy in the comprehensive care of brain tumor patients. Before the computed tomography (CT) and magnetic resonance imaging (MRI) era, many reports on the management of malignant
741
742
47
Comprehensive brain tumor management
intra-axial brain tumors employed whole brain irradiation. However, the last 20 years have seen a definite shift away from utilizing whole brain fields to the use of regional fields with margins around enhancing disease of the order of 2 cm. Numerous factors are responsible for this shift in treatment philosophy-included amongst them are the better tumor localization afforded by CT and MR imaging, the many reports documenting that the primary cause of treatment failure was related to tumor recurrence at the original site in over 90% of cases [32], and the wish to reduce radiation related morbidity associated with whole brain irradiation. Based upon data from two randomized controlled trials [33,34], most centers have now eliminated the use of whole brain radiation for the adjuvant treatment of malignant intra-axial brain tumors following surgery in favor of local radiation fields (i.e., the high dose volume being the enhancing primary plus 2 cm margins) for the whole course of treatment, with no apparent difference in survival. The use of conformal radiotherapy techniques at most centers nowadays allows for accurate targeting of the residual tumor volume and/or the tumor bed with sparing of surrounding normal tissue. The pattern of local recurrence observed in malignant intra-axial brain tumors has led to an investigation of ways of intensifying the dose of radiation delivered by conventional means in an effort to improve local control rates. An increase in radiation dose through conventional fractionation and hyperfractionation regimens have been investigated. With respect to the results of dose escalation with conventional fractionation, no randomized studies of such protocols have shown any convincing benefit of radiation doses as high as 70 Gy compared with conventional doses in the range of 50–60 Gy [35,36]. Accordingly, the evidence would support the use of post-operative radiotherapy to a total dose in the range of 50–60 Gy utilizing conventional fractionation (60 Gy in 30 fractions over 6
weeks), particularly in view of the fact that higher doses are likely associated with higher toxicity. Hyperfractionation involves the use of a larger number of smaller sized fractions to a total dose which is higher than with conventionally administered irradiation in the same overall treatment time. Normal glial and vascular cells limit the total amount of irradiation that can be administered. These cells divide very slowly, and are better able to repair sub-lethal damage than neoplastic cells. Consequently, there might be an advantage to administering multiple smaller sized fractions to a higher total dose, the theory being that the improved repair of sub-lethal damage at lower sized fractions might allow a higher total dose to be associated with the same degree of late sequelae. Neoplastic cells are relatively rapidly dividing cells, and the increased number of daily fractions would increase the chance of radiating them at a more sensitive phase of their cell cycle. At smaller radiation doses per fraction, cell killing is less dependant on oxygen, which might be advantageous given the known areas of hypoxia in these tumors. Although investigators were able to safely escalate the dose to 72 Gy utilizing hyperfractionation, randomized studies did not demonstrate any advantage over conventionally fractionated doses in the range of 50–60 Gy [37,38]. No other modifications of the fractionation schedule, including accelerated fractionation, hypofractionation, or any combination of these, have produced any meaningful survival advantage [39].
Stereotactic Radiosurgery Stereotactic radiosurgery has similarities to surgery – a single large dose of radiation that is destructive to all tissue is given to a localized and accurately defined area. Unlike surgery, the effect of stereotactic radiosurgery is not immediate, taking weeks to months to develop. Although it has not been rigorously applied as
Comprehensive brain tumor management
an alternative to surgery, several studies have investigated the role of stereotactic radiosurgery as an adjuvant to surgery plus fractionated radiotherapy. Although nonrandomized data suggested this treatment may improve survival [40], the results of a randomized trial examining the use of a radiosurgery boost to the tumor bed followed by external beam radiotherapy and BCNU chemotherapy as compared with external beam radiotherapy and BCNU chemotherapy alone found that it did not confer benefit in terms of overall survival, quality of life, or patterns of failure, and may in fact be associated with an increased incidence of radiation toxicity [41]. These results might be expected from the two older randomized brachytherapy studies – gliomas are simply not focal diseases lending themselves to cure by focal treatments. Presently, the role of stereotactic radiosurgery in the setting of malignant intra-axial brain tumors remains experimental. In order to better clarify its role in the overall management of these patients, with both newly diagnosed and recurrent tumors, several multi-center randomized controlled trials are currently underway [42].
47
unclear. By common consent, the combination of procarbazine, CCNU, and vincristine (PCV) was adopted as the ‘‘gold standard’’ in spite of there being relatively little firm evidence in support of this. The most frequent study design incorporated nitrosourea based chemotherapy as adjuvant to standard surgery and radiotherapy in comparison to standard treatment alone. Most studies were underpowered and failed to show any survival advantage [43,44]. In 2002, the glioma metaanalysis trialist group [45] carried out a systematic review of individual patient data from 12 such trials whose data were sufficiently homogeneous to allow such an analysis to be performed. Radiotherapy doses ranged from 40–60 Gy and volumes varied from whole brain to tumor only with a margin. All patients received a nitrosourea, some as single agent, and others as combination therapy. There was a statistically significant increase in median survival from 10 to 12 months, equivalent to a 6% absolute improvement in survival at 1 year (from 40 to 46%). This was just sustained at 2 years but had effectively vanished at 3 years.
Tailored Chemotherapy Chemotherapy A significant role for adjuvant chemotherapy in the majority of malignant intra-axial brain tumors is difficult to demonstrate. Until relatively recently, its major role in the treatment of most high grade intra-axial brain tumors has been as a palliative option for recurrent disease following conventional treatment, although many centers have routinely included chemotherapy as an integral part of the initial adjuvant therapy for anaplastic gliomas. The nitrosoureas (BCNU, CCNU) have historically played a central role in the chemotherapy of gliomas. However, there are very few studies in which either single agents or combinations were compared and optimum treatment is therefore
It has been intuitively recognized for some years that there exists a differential chemosensitivity among different histologic classes of malignant intra-axial brain tumors, and even differential chemosensitivity between tumors that appear histologically indistinguishable. Recently, the interrogation of brain tumors at a molecular level has been able to extract objective molecular differences between histologically similar tumors, giving a foundation to the burgeoning line of inquiry involving molecular predictors of response. Two seminal examples of this are the stratification of PCV chemotherapy response in anaplastic oligodendroglioma based upon loss of heterozygosity at chromosomes 1p and 19q, and the differential response of patients with glioblastoma multiforme
743
744
47
Comprehensive brain tumor management
(GBM) to temozolamide chemotherapy predicated upon the presence or absence of epigenetic silencing of the MGMT gene.
Anaplastic Oligodendroglioma Cairncross and colleagues were the first to show that in patients with relapsed anaplastic oligodendroglioma, the deletion of the short arm of chromosome 1 and the long arm of chromosome 19 (1p-, 19q-) was a common finding, and, when present, was associated with pronounced chemosensitivity to procarbazine, lomustine (CCNU) and vincristine (PCV). Indeed, patients with loss of heterozygosity at 1p and 19q enjoyed prolonged remissions, sometimes lasting several years [46]. Based upon this initial observation in relapsed anaplastic oligodendroglioma, a large multi-center trial investigating the addition of PCV chemotherapy to radiotherapy in newly diagnosed anaplastic oligodendroglioma has been undertaken [47]. Although an increase in progression free survival in the PCV arm was observed, this did not translate into an increase in overall survival. However, patients with 1p/ 19q loss demonstrated a clearly better outcome, with median survival over 6–7 years as compared to 2–3 years in patients without 1p/19q loss. Because of the exquisite chemosensitivity of the variant of this tumor in which deletions of 1p and 19q are present, many clinicians today advocate upfront chemotherapy with either PCV or temozolamide as first-line adjuvant treatment in 1p/19q loss oligodendroglioma, whilst reserving radiotherapy for relapse, despite strong evidence in favor of this approach from a well designed and conducted randomized controlled trial.
Glioblastoma Multiforme Temozolamide first demonstrated its activity in patients with recurrent high-grade glioma. Three
pivotal phase II studies with identical entry criteria were conducted for patients with GBM and with anaplastic astrocytoma (AA). Despite disappointingly low objective response rates in GBM of 5 and 7%, respectively, but with an interesting response rate of 35% in the AA trial, these studies suggested an increase in the fraction of patients being progression-free at 6 months compared to a historical database [48]. Based on the activity in recurrent glioma, as well as in vitro data suggesting additive or supraadditive activity when temozolamide was administered concomitantly with radiotherapy, the European Organization for Research and Treatment of Cancer (EORTC) and the National Cancer Institute of Canada (NCIC) Clinical Trials Group compared the combination of temozolamide plus radiotherapy to standard radiotherapy alone in a large prospective randomized phase III trial on 573 patients in which the primary outcome was overall survival [49]. This study unequivocally demonstrated that the combination of temozolamide and radiotherapy followed by up to six cycles of adjuvant temozolamide improved survival. With combined modality treatment, the 2-year survival increased from 10 to 26%. Subgroup analyses suggested that patients of all age groups benefited from this treatment. Overall, the combined treatment was well tolerated, and the main reason for early discontinuation was disease progression. The clinical relevance of the mechanistic implication of the functional status of the DNA repair enzyme MGMT in alkylating chemotherapy was investigated within the context of this randomized EORTC/NCIC trial. Samples from 206 patients were analyzed to ascertain the methylation status of the MGMT gene promoter using methylation-specific PCR [50]. In 45% of tumor samples, the MGMT gene promoter was methylated, and the gene was silenced; this epigenetic silencing of a critical DNA repair gene greatly diminished the tumor cell’s ability to repair the DNA damage induced by the alkylating
Comprehensive brain tumor management
chemotherapy. Patients with a silenced MGMT gene, which was, in essence, a predictor of chemosensitivity, indeed had longer survival. Breakdown of the data by treatment strongly suggested that the MGMT methylation status was a predictive marker for benefit from temozolamide chemotherapy. For patients who received combined radiotherapy and temozolamide, the 2-year survival rate was 46% when their tumor had a methylated MGMT promoter, in contrast to only 14% in patients with an unmethylated MGMT gene promoter. Thus, in this molecularly defined subgroup, temozolamide was found to be even more effective, increasing the median survival from 9 months to 21.7 months, while patients with an unmethylated MGMT gene promoter had little, if any, temozolamidederived benefit, with a median survival of 12.7 months. These data suggest that only patients whose tumors harbor a methylated MGMT promoter (and therefore, a silenced MGMT gene) should receive temozolamide, while for other patients, alternative strategies should be considered. Strategies aiming at overcoming MGMT-mediated treatment resistance are currently under development. Continuous administration of temozolamide, for instance, has been shown to deplete intracellular MGMT, and novel dose-dense schedules are currently being explored as adjuvant treatments in large randomized trials.
New Directions Despite significant gaps which persist in our understanding, genomic and proteomic technologies have provided a wealth of information regarding the clinical and biological behavior of malignant intra-axial brain tumors, the genetic pathways involved in their genesis, and the nature and role of prototypic alterations in these pathways. The challenge now is to integrate all of this knowledge in an interdisciplinary way,
47
involving the expertise of clinicians, epidemiologists, and basic scientists, in order to fully understand this complex disease. One of the most perplexing issues remains to comprehend, and to hopefully take advantage of, the signature molecular heterogeneity of these tumors, a feature which is in large part responsible for their resistance to therapy. On the one hand, malignant gliomas harbor a wide array of molecular defects, yielding a large number of potentially exploitable, therapeutic targets. On the other hand, given the diversity and number of molecular defects in these tumors, as well as the known redundancy and cross-talk between aberrant signal transduction pathways, one would predict that inhibition of a single target is unlikely to have a major, durable antitumor effect in most instances. Indeed, results of first generation clinical trials, conducted to evaluate a wide array of molecular targeted agents as monotherapeutics, support this prediction. Molecular profiling of malignant intra-axial tumors has uncovered some key molecules and pathways that, if perturbed, could disrupt the hallmark characteristics of a neoplastic cell, namely the propensity for uncontrolled replication, the capacity for angiogenesis, and the ability to migrate and invade adjacent normal tissue. Although a detailed review of the multiplicity of pathways and potential targets is beyond the scope of this chapter, the interested reader is referred to several review articles on this topic [1,51]. What follows is a brief discussion of a representative example of a small molecule that upsets, in a targeted way, one of the key attributes of a neoplastic cell, for which some clinical data regarding its potential efficacy is available.
Inhibition of Cellular Proliferation In malignant gliomas, the most prominent altered tyrosine kinase receptor is the epidermal growth factor receptor (EGFR), which is amplified and
745
746
47
Comprehensive brain tumor management
overexpressed in approximately 50% of GBMs. The EGFR tyrosine kinase inhibitor gefitinib has been investigated in several studies in malignant glioma. The combination of gefitinib and radiotherapy was examined in a phase I/II trial with a total of 147 patients with newly diagnosed GBM [52]. Median survival for the whole population was only 11 months, comparable to that of historical controls receiving radiation treatment alone. Despite these rather disappointing preliminary results, it is possible that such a treatment strategy will work only in selected patients who have tumors that are strictly dependent on EGFR signaling for growth and survival. Alternatively, because of the redundancy of the pathway, additional targets may need to be considered in a combination therapeutic strategy. An alternative approach targeting the EGFR pathway employs a vaccination strategy. Promising results have been reported from a phase II trial of patients with newly diagnosed GBM vaccinated with the anti-EGFRvIII peptide CDX-110 in the temozolamide maintenance phase of a temozolamide/radiotherapy combination protocol [53]. Complete absence of EGFRvIII in the tumor tissues at recurrence suggested elimination of such cells by an activated immune system.
Inhibition of Angiogenesis Malignant gliomas are highly vascularized and infiltrative tumors strongly dependent on endothelial cell proliferation regulated by proangiogenic cytokines (e.g., vascular endothelial growth factor [VEGF]). Preclinical and clinical evidence suggests that antiangiogenic treatment may work best in combination with chemotherapy and radiotherapy, and a number of early phase clinical trials have been initiated. A recent clinical study with correlative imaging and biologic end points suggests that VEGF receptor inhibition by cediranib leads to normalization of tumor vasculature and restoration of the blood-brain barrier, thus reducing
contrast enhancement and edema [54]. In a recent trial, the VEGF receptor tyrosine kinase inhibitor vatalanib (which also inhibits the PDGF receptor [PDGFR]) is being evaluated in combination with radiotherapy and temozolamide chemotherapy [55]. Preclinical observations of potential increase in hypoxia and radiation resistance raise concern as to the optimal timing of antiangiogenic therapy. A further phase II trial therefore is designed to evaluate the safety and efficacy of the addition of vatalanib either concurrent with temozolamide and radiotherapy or by adding the VEGF receptor inhibitor only after completion of concomitant chemoradiotherapy [56].
Inhibition of Cell Migration Integrins are heterodimer transmembrane receptors for the extracellular matrix, regulating cell adhesion and migration. In preclinical models, inhibition of integrin function efficiently suppressed tumor cell migration and inhibited tumor progression. Cilengitide, a synthetic peptide, competitively inhibits integrin binding to extracellular matrix proteins. In a phase I study in recurrent glioma, single-agent activity of cilengitide was observed, with objective radiographic responses noted [57]. A subsequent phase II trial of cilengitide added to a standard temozolamide/ radiotherapy regimen (temozolamide with concomitant radiotherapy, followed by temozolamide with or without cilengitide) has recently been completed in patients with newly diagnosed GBM. Addition of the novel agent was associated with little or no additional toxicity, and initial results suggest efficacy in a subgroup of patients.
References 1. Furnari FB, Fenton T, Bachoo RM, Mukasa A, Stommel JM, Stegh A, Hahn WC, Ligon KL, Louis DN, Brennan C, Chin L, DePinho RA, Cavenee WK. Malignant astrocytic glioma: genetics, biology, and paths to treatment. Genes Dev 2007;21:2683-710.
Comprehensive brain tumor management
2. Greene GM, Hitchon PW, Schelper RL, Yuh W, Dyste GN. Diagnostic yield in CT-guided stereotactic biopsy of gliomas. J Neurosurg 1989;71:494-7. 3. Bernstein M, Parrent AG. Complications of CT-guided stereotactic biopsy of intra-axial brain lesions. J Neurosurg 1994;81:165-8. 4. Kulkarni AV, Guha A, Lozano A, Bernstein M. Incidence of silent hemorrhage and delayed deterioration after stereotactic brain biopsy. J Neurosurg 1998;89:31-5. 5. Soo TM, Bernstein M, Provias J, et al. Failed stereotactic biopsy ina series of 518 cases. Stereotact Funct Neurosurg 1995;64:183-96. 6. Kongkham PN, Knifed E, Tamber MS, Bernstein M. Complications in 622 cases of frame-based stereotactic brain biopsy – a decreasing procedure. Can J Neurol Sci 2008;35(1):79-84. 7. Coffey RJ, Lundsford LD, Taylor FH. Survival after stereotactic biopsy of malignant gliomas. Neurosurgery 1988;22:465-473. 8. Vuorinen V, Hinkka S, Farkkila M, Jaaskelainen J. Debulking or biopsy of malignant glioma in elderly people – a randomized study. Acta Neurochir 2003;145:5-10. 9. Whittle IR, Pringle AM, Taylor R. Effects of resective surgery for left sided intracranial tumours on language function: a prospective study. Lancet 1998;351:1014-18 10. Castro MG, Cowen R, Williamson IK, et al. Current and future strategies for the treatment of malignant brain tumors. Pharmacol Ther 2003;98:71-108. 11. Muacevic A, Kreth FW. Quality-adjusted survival after tumor esection and/or radiation therapy for elderly patients with glioblastoma multiforme. J Neurol 2003; 250:561-8. 12. Keles GE, Lamborn KR, Chang SM, Prados MD, Berger MS. Volume of residual disease as a predictor of outcome in adult patients with recurrent supratentorial glioblastomas multiforme who are undergoing chemotherapy. J Neurosurg 2004;100:41-6. 13. Keles GE, Anderson B, Berger MS. The effect of extent of resection on time to tumor progression and survival in patients with glioblastoma multiforme of the cerebral hemisphere. Surg Neurol 1999;52:371-9. 14. Laws ER, Parney IF, Huang W, et al. Survival following surgery and prognostic factors for recently diagnosed malignant glioma: data from the Glioma Outcomes Project. J Neurosurg 2003;99:467-73. 15. Lacroix M, Abi-Said D, Fourney DR, et al. A multivariate analysis of 416 patients with glioblastoma multiforme: prognosis, extent of resection, and survival. J Neurosurg 2001;95:190-8. 16. Curran WJ Jr, Scott CB, Horton J, et al. Does extent of surgery influence outcome for astrocytoma with atypical or anaplastic foci (AAF)? A report from three Radiation Therapy Oncology Group (RTOG) trials. J Neurooncol 1992;12:219-27. 17. Kreth FW, Berlis A, Spiropoulou V, et al. The role of tumor resection in the treatment of glioblastoma multiforme in adults. Cancer 1999;86:2117-21.
47
18. Albert FK, Forsting M, Sartor K, Adams HP, Kunze S. Early postoperative magnetic resonance imaging after resection of malignant glioma: objective evaluation of residual tumor and its influence on regrowth and prognosis. Neurosurgery 1994;34:45-60. 19. Grant R, Metcalfe SE. Biopsy versus resection for malignant glioma. Cochrane Database Syst Rev 2005; 2:CD002034. 20. Hess KR. Extent of resection as a prognostic variable in the treatment of gliomas. J Neurooncol 1999; 42: 227-31. 21. Rosenthal MA, Kavar B, Hill JS, et al. Phase I and pharmacokinetic study of photodynamic therapy for high-grade gliomas using a novel boronated porphyrin. J Clin Oncol 2001;19:519-24. 22. Germano IM, Fable J, Gultekin SH, Silvers A. Adenovirus/herpes simplex-thymidine kinase/ganciclovir complex: preliminary results of a phase I trial in patients with recurrent malignant gliomas. J Neurooncol 2003; 65: 279-89. 23. Rainov NG. A phase III clinical evaluation of herpes simplex virus type 1 thymidine kinase and ganciclovir gene therapy as an adjuvant to surgical resection and radiation in adults with previously untreated glioblastoma multiforme. Hum Gene Ther 2000;11: 2389-401. 24. Immonen A, Vapalahti M, Tyynela K, et al. AdvHSV-tk gene therapy with intravenous ganciclovir improves survival in human malignant glioma: a randomised, controlled study. Mol Ther 2004;10:967-72. 25. Farkkila M, Jaaskelainen J, Kallio M, et al. Randomised controlled study of intratumoral recombinant gammainterferon treatment in newly diagnosed glioblastoma. Br J Cancer 1994;70:138-41. 26. Laperriere NJ, Leung PM, McKenzie S, et al. Randomized study of brachytherapy in the initial management of patients with malignant astrocytoma. Int J Radiat Oncol Biol Phys 1998;41:1005-11. 27. Westphal M, Hilt DC, Bortey E, Delavault P, Olivares R, Warnke PC, et al. A phase 3 trial of local chemotherapy with biodegradable carmustine (BCNU) wafers (Gliadel wafers) in patients with primary malignant glioma. Neurooncol 2003;5:79-88. 28. Brem H, Piantadosi S, Burger PC, Walker M, Selker R, Vick NA, et al. Placebo-controlled trial of safety and efficacy of intraoperative controlled delivery by biodegradable polymers of chemotherapy for recurrent malignant gliomas. The Polymer-Brain Tumor Treatment Group. Lancet 1995;345:1008-12. 29. Kioi M, Husain SR, Croteau D, Kunwar S, Puri RK. Convection-enhanced delivery of interleukin-13 receptordirected cytotoxin for malignant glioma therapy. Technol Cancer Res Treat 2006;5:239-50. 30. McDermott MW, Bernstein M. Image-guided Surgery. In: Bernstein M, Berger MS, editors. Neuro oncology. The essentials, 2nd ed. New York: Thieme; 2008. p. 112-23.
747
748
47
Comprehensive brain tumor management
31. Serletis D, Bernstein M. Prospective study of awake craniotomy used routinely and nonselectively for supratentorial tumors. J Neurosurg 2007;107:1-6. 32. Hochberg FH, Pruitt A. Assumptions in the radiotherapy of glioblastoma. Neurology 1980;30:907-11. 33. Shapiro WR, Green SB, Burger PC, et al. Randomized trial of three chemotherapy regimens and two radiotherapy regimens in postoperative treatment of malignant glioma. Brain Tumor Cooperative Group Trial 8001. J Neurosurg 1989;71:1-9. 34. Kita M, Okawa T, Tanaka M, et al. Radiotherapy of malignant glioma – prospective randomized clinical study of whole brain versus local irradiation (in Japanese). Gan No Rinsho 1989;35:1289-94. 35. Bleehen NM, Stenning SP. A Medical Research Council trial of two radiotherapy doses in the treatment of grades 3 and 4 astrocytoma. The Medical Research Council Brain Tumour Working Party. Br J Cancer 1991;64:769-74. 36. Nelson DF, Diener-West M, Horton J, et al. Combined modality approach to treatment of malignant gliomas – re-evaluation of RTOG 7401/ECOG 1374 with long-term follow-up: a joint study of the Radiation Therapy Oncology Group and the Eastern Cooperative Oncology Group. NCI Monogr 1988;6:279-84. 37. Scott CB, Curran WJ, Yung WKA. Long term results of RTOG 9006: a randomized study of hyperfractionated radiotherapy (RT) to 72.0 Gy and carmustine versus standard RT and carmustine for malignant glioma patients with emphasis on anaplastic astrocytoma (AA) patients (abstract). Proc Annu Meet Am Soc Clin Oncol 1998;17:401a. 38. Shin KH, Urtasun RC, Fulton D, et al. Multiple daily fractionated radiation therapy and misonidazole in the management of malignant astrocytoma. A preliminary report. Cancer 1985;56:758-60. 39. Glinski B. Postoperative hypofractionated radiotherapy versus conventionally fractionated radiotherapy in malignant gliomas. A preliminary report on a randomized trial. J Neurooncol 1993;16:167-72. 40. Nwokedi EC, DiBiase SJ, Jabbour S, Herman J, Amin P, Chin LS. Gamma knife stereotactic radiosurgery for patients with glioblastoma multiforme. Neurosurgery 2002;50:41-6. 41. Souhami L, Seiferheld W, Brachman D, et al. Randomized comparison of stereotactic radiosurgery followed by conventional radiotherapy with carmustine to conventional radiotherapy with carmustine for patients with glioblastoma multiforme: report of Radiation Therapy Oncology Group 93-05 protocol. Int J Radiat Oncol Biol Phys 2004;60:853-60. 42. Tsao MN, Mehta MP, Whelan TJ, Morris DE, Hayman JA, Flickinger JC, et al. The American Society for Therapeutic Radiology and Oncology (ASTRO) evidence-based review of the role of radiosurgery for malignant glioma. Int J Radiat Oncol Biol Phys 2005;63:47-55. 43. Walker MD, Alexander E, Hunt WE, et al. Evaluation of BCNU and/or radiotherapy in the treatment of anaplastic
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
gliomas. A cooperative clinical trial. J Neurosurg 1978; 49:333-43. Walker MD, Green SB, Byar DP, et al. Randomized comparisons of radiotherapy and nitrosoureas for the treatment of malignant glioma after surgery. N Engl J Med 1980;303:1323-29. Glioma Meta-Analysis Group. Chemotherapy in adult high-grade glioma: a systematic review and metaanalysis of individual patient data from 12 randomised trials. Lancet 2002;359:1011–18. Cairncross JG, Ueki K, Zlatescu MC, et al. Specific genetic predictors of chemotherapeutic response and survival in patients with anaplastic oligodendrogliomas. J Natl Cancer Inst 1998;90:1473-79. Cairncross JG, Seiferheld W, Shaw E, et al. An intergroup randomized controlled clinical trial (RCT) of chemotherapy plus radiation (RT) versus RT alone for pure and mixed anaplastic oligodendrogliomas: initial report of RTOG 94– 02. Proc Am Soc Clin Oncol 2004;23:107 (abstract 1500). Brada M, Hoang-Xuan K, Rampling R, et al. Multicenter phase II trial of temozolomide in patients with glioblastoma multiforme at first relapse. Ann Oncol 2000; 12:259-66. Stupp R, Mason WP, van den Bent MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005;352:987-96. Hegi ME, Diserens AC, Gorlia T, et al. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med 2005; 352:997-1003. Sathomsumetee S, Reardon DA, Desjardins A, Quinn JA, Vredenbrugh JJ, Rich JN. Molecularly targeted therapy for malignant glioma. Cancer 2007; 110:13-24. Chakravarti A, Berkey B, Robins H, et al. An update of phase II results from RTOG 0211: a phase I/II study of gefitinib with radiotherapy in newly diagnosed glioblastoma. J Clin Oncol 2006;24:64s (suppl; abstract 1527). Heimberger A, Hussain S, Aldape K, et al. Tumor-specific peptide vaccination in newly diagnosed patients with GBM. J Clin Oncol 2006;24:107s (suppl; abstract 2529). Batchelor TT, Sorensen AG, di Tomaso E, et al. AZD2171, a pan-VEGF receptor tyrosine kinase inhibitor, normalizes tumor vasculature and alleviates edema in glioblastoma patients. Cancer Cell 2007;11:83-95. Brandes A, Stupp R, Hau P, et al. EORTC Study 26041– 22041:Phase I/II study on concomitant and adjuvant zemozolomide (TMZ) and radiotherapy with or without PTK787/ZK222584 (PTK/ZK) in newly diagnosed glioblastoma: results of Phase I. J Clin Oncol 2007; 25:81s, (suppl; abstract 2026). Nabors B, Mikkelsen T, Rosenfeld S, et al. A phase I and correlative biology study of cilengitide in patients with recurrent malignant glioma. J Clin Oncol 2007; 25:1651-7. Stupp R, Goldbrunner R, Neyns B, et al. Phase I/IIa trial of cilengitide (EMD121974) and temozolomide with concomitant radiotherapy, followed by temozolomide and cilengitide maintenance therapy in patients with newly diagnosed glioblastoma. J Clin Oncol 2007;25:75s, (suppl; abstract 2000).
40 CT in Image Guided Surgery D. Kondziolka . L. D. Lunsford
The last three decades in neurosurgery have been the era of image-guided surgery. Neurosurgeons who were trained to perform surgery without sophisticated parenchymal imaging, were used to large craniotomies, free-hand biopsies, palpation of the brain surface, and interpretation of angiograms, x-rays, or air studies. Computed tomography, as the first high-resolution imaging tool to study brain tissue, ushered in an era of tremendous possibility. Neurosurgeons are now trained in stereotactic technology, image-guided navigation, more precise lesion resection, accurate and reliable deep brain surgery, and are accustomed to seeing an imaging device within an operating room. In this chapter, we review the current use of intraoperative CT imaging. At our center, the first CT compatible stereotactic head frame, in collaboration with industry, was constructed in 1978 and utilized in 13 patients [1,2]. Virtually no brain operation in today’s era is done without image guidance. This includes craniotomy for most non trauma indications, conventional stereotactic surgery, functional surgery, and radiosurgery [3]. During this interval, the newly redesigned Leksell CT compatible stereotactic head frame [4] was used for dedicated brain biopsies under the direction of its inventor, Lars Leksell. Several groups were working on devices to allow accurate CT-based stereotactic surgery [5]. Upon returning to Pittsburgh in 1981, Dr. Lunsford began a program to obtain a dedicated CT scanner equipped operating room at a time when the regional health systems agency had allocated two diagnostic CT scanners to the entire city of Pittsburgh (> Figure 40‐1) We built a new operating room and began our first series of stereotactic #
Springer-Verlag Berlin/Heidelberg 2009
procedures [6–15]. The scanner was updated to a GE 9800 in 1991, and a new spiral scanner in 2007. Neurosurgical technologies should be simple and practical. It must assist the surgeon to perform the operation, it should be efficient, it must promote better outcomes, and it must reduce morbidity. Intraoperative imaging using CT and then MRI technologies fulfilled these goals, although MRI has posed unique challenges to surgical efficiency.
Frame Based Stereotactic Surgery Our experience from 1977 to 2007 using frame based stereotactic surgery is shown in > Table 40‐1. Many procedures used CT imaging for ‘‘open’’ (when a surgical opening was made through the skull) stereotactic surgery. We added MRI based target localization in selected cases beginning in 1991 [16,17]. As the resolution of MR imaging improved, we increasingly relied on MRI for functional neurosurgery in patients eligible for MRI [18,19]. We also used MRI for selected biopsies, especially those in high risk locations or when (> Figure 40‐2) their imaging characteristics were best defined by MRI [19–22]. However, for brain tumor biopsy, cyst management or intracavitary irradiation, CT remains the imaging modality of choice. All of our own patients undergo stereotactic frame placement in an operating room environment followed by imaging either in our dedicated CT operating room or a nearby diagnostic 1.5 Tesla MRI unit. Calculation of coordinates is performed either using the standard software available on the scanners,
620
40
CT in image guided surgery
. Figure 40‐1 The first dedicated therapeutic CT scanner for brain surgery was installed at the University of Pittsburgh Medical Center in 1982. The room had a dedicated GE 8800 CT scanner and a Phillips C-Arm image intensifier
. Table 40‐1 Morphologic stereotactic Presbyterian; 1979–2007
brain
surgery:
UPMC-
No. patients Diagnostic biopsy Cyst aspiration Radiation implant Brain abscess Catheter and cyst reservoir insertion Hematoma aspiration Frame based craniotomy Ventriculostomy placement Total
1,664 197 145 97 19 9 10 6 2,147
or more recently using an image integrated surgical planning system (SPS Elekta, Inc., Norcross, GA). We precisely pre-plot probe trajectories to reach the target and choose the route designed to avoid as many pial or ependymal surfaces as possible (> Figure 40‐3). Brain biopsies are performed in all areas of the brain including the brain stem [21–23]. Many lesions in the deep locations of the brain
can be approached from a single trans-frontal intra-axial trajectory. In selected cases for lesions adjacent to the fourth ventricle or within the cerebellar hemisphere, a transcerebellar trajectory is performed with the patient moved to a semi-sitting position on the operating room table. For virtually all patients older than 12 years the procedures are performed under monitor assisted conscious sedation. We have performed CT stereotactic biopsy in a child as young as 5 months [24]. CT imaging is always performed with contrast-enhancement, even with non-enhancing lesions, in order to identify blood vessels along the planned trajectory. Axial images are obtained at slice intervals of 1.5–3 mm depending on the size and location of the target. The image thickness is typically 3–5 mm to provide adequate signal. For frame-based biopsy, it is important not to place the stereotactic frame pins at the plane of the target, or metallic artifact can obscure the image.
CT in image guided surgery
40
. Figure 40‐2 Stereotactic frame-based biopsy for precision and reliability. Certain lesions are best seen with MRI. CT imaging (left) showed poor contrast enhancement and pin artifacts compared to long relaxation time MRI (right) for this thalamic astrocytoma
Although ‘‘frameless’’ biopsy has increased in popularity, we think that the concept of general anesthesia, three point (Mayfield) rather than four point fixation, frequent use of a burrhole, a larger incision rather than a twist drill craniostomy puncture site, and other often more complex tools to hold the probe, seem counterintuitive. Reaching a final diagnosis involves a collaborative effort with the neuropathologic team, including review of the pre and postoperative images. A confirmatory intraoperative CT scan is always done, which shows a small air density at the target site. This image is used to demonstrate to the neuropathologist that sampling was accurate. In addition, immediate imaging helps rule out hemorrhage. In our experience, the current risk of a brain biopsy hemorrhage requiring evacuation using frame based techniques is less than 0.5% [25]. Frame based systems provide extreme accuracy to less than 1 mm as mandated by standards of the American Society of Testing and Manufacturing. Rigid probe fixation is critical. In addition, we are able to precisely reduplicate the
pathway of the needle previously plotted using the surgical planning system.
Brain Cyst Management CT imaging is valuable in cyst drainage. > Table 40‐2 shows our experience with stereotactic cyst aspirations. CT is used to calculate the cyst volume using available imaging software, and to choose the safest trajectory into the cyst. For cyst drainage, the target should be in the dependent portion of the cyst according to supine head position so that gravity facilitates maximum drainage. We typically allow cyst fluid to egress on its own and always collect the volume to compare to the pre-surgical calculation. For more viscous fluid, we aim to evacuate 70–80% of the contents. The alleviation of brain cysts by simple aspiration has been done in a large variety of patients, including those with glial cysts, neuroepithelial cysts, and in the management of cystic craniopharyngioma [6,26–30]. Simple aspiration of craniopharyngoma cysts
621
622
40
CT in image guided surgery
. Figure 40‐3 Surgical planning with SurgiPlan software assists the use of CT imaging to define the optimal probe trajectory, with angles and coordinates provided for the stereotactic frame and arc
. Table 40‐2 Stereotactic management of brain cysts: UPMC Presbyterian;1981–2007 No. patients Tumor cyst aspiration Colloid cyst aspiration Craniopharyngioma cyst aspiration Total
119 47 31 197
may be periodically needed, and we preferentially prefer this as opposed to placement of an Ommaya reservoirs drainage system. Using the fine needle (0.9 mm) puncture technique advocated by Backlund, we are able to introduce
colloidal chromic phosphate P32 into a craniopharyngioma to result in involution of the cyst over the course of time [31]. Cyst reservoir systems in our experience seem to irritate a craniopharyngioma leading to a progressively more frequent need for cyst aspiration. In contrast, A P32 injection with a dose of approximately 20,000 cGy offers an effective management for primarily monocystic craniopharyngomas >5 cc. in volume. As a pure beta emitter, the dose falloff is quite rapid, thereby keeping low the risk to surrounding critical structures such as the optic apparatus. > Table 40‐3 shows our experience with the stereotactic isotope strategies.
CT in image guided surgery
. Table 40‐3 Open stereotactic irradiation surgery: UPMC Presbyterian; 1981–2007 No. patients Intracavitary radiation (P-32): Craniopharyngioma 73 Glioma 13 Arachnoid cyst 4 Subtotal 90 Stereotactic interstitial radiation (I-125 Brachytherapy): Glioma 51 Ependymoma 2 Malignant meningioma 1 CNS lymphoma 1 Subtotal 55 Total 145
Image-Guided Craniotomy CT-guided, stereotactic frame-based craniotomies have been performed for more than 25 years. In the early 1980s we began to perform image guided resections using the Leksell frame in our intraoperative CT suite. Although certainly useful in finding the lesion, our initial experience failed to disclose a major benefit from intraoperative CT assisted cytoreductive glioma surgery [20,32–34]. We found little evidence that taking out the ‘‘majority’’ of a glial tumor yielded superior results. It is possible that patients eligible for extensive cytoreductive surgery, primarily those with lobar or polar tumors, may benefit from such an approach if their tumors have less infiltrative borders and are located in less critical locations. Such surgery must then be combined with additional adjuvant management strategies including radiation therapy, chemotherapy, and potentially radiosurgery. While we are confident that centers employing intraoperative MRI are making headway in terms of their ability to better resect glial neoplasms, we are, nonetheless, struck by the fact that a 99% (‘‘gross total’’) removal, still contains an enormous tumor cell population. Although subsequent adjuvant management may be better able to control such
40
tumors, this remains to be shown. In order to define such a benefit a large randomized trial would be needed to show a meaningful survival increase for gliomas.
Surgical Navigation for Resection While frame based surgical resection procedures including craniotomy were performed on our operating room CT scanner table, the introduction of frameless systems changed the paradigm to using preoperative imaging to guide most brain and many spine operations [3]. A variety of products were used successively including the early generation ISG wand and products sold by Elekta, Medtronic, and Stryker. The Surgiscope proved to be the most challenging of the image guidance products. Now almost all intracranial surgery requires precise CT or MRI guidance technologies. More recently, magnetoencephalography (MEG) and positron emission tomography (PET) imaging have been fused with MRI or CT for certain frame based procedures. We continue to believe that intraoperative CT is a practical imaging tool with high reliability. It adds to but does not modify the goal or flow of the surgical procedure. In addition, as a shared resource, intraoperative CT can be used by other surgical services. For neurosurgical procedures we fuse preoperative MRI scans with intraoperative CT to enhance further diagnostic, functional and endoscopic surgical procedures.
Abscess Management We have performed stereotactic aspiration for the diagnosis and simple or catheter assisted drainage of 97 patients with brain abscesses. Frame based stereotactic surgery is an excellent way to manage brain abscesses [9,12,35]. As in all surgical pyogenic infections, the principals include early recognition, drainage, and identification of the
623
624
40
CT in image guided surgery
appropriate organism where possible. A variety of techniques are available to manage brain abscesses including frame based and frameless approaches. We have treated most brain abscesses using a twist drill craniostomy, stereotactic puncture, followed by drainage of the abscess cavity. The bacteriological studies in most cases are able to define the causative organism(s). At the time of drainage, appropriate broad spectrum antibiotics are given in the operating room. We perform a full assessment of the potential sources for a pyogenic brain abscess, including heart, sinus or other septicemia sources. Rarely brain abscess cases require re-drainage if slow resolution is not forthcoming. The treatment regimen consists of appropriate antibiotic management, placement of a catheter in the abscess (for larger volumes), and gradual removal of the drainage catheter after several days. This approach has a high success rate, and virtual elimination of mortality from bacteriological brain abscesses. For patients with fungal abscesses in the context of chronic immune suppression, especially those in the transplant arena, biopsy and identification of the organism allows the potential for aggressive anti-fungal management. In our community, the need for brain biopsy in the diagnosis of HIV related conditions has declined. Patients with a progressive mass lesion within the brain in the context of acquired immunodeficiency syndrome (AIDS), normally undergo empiric antitoxoplasmosis treatment first. Biopsy is performed if the lesion progresses. On occasion, lymphoma requires brain biopsy for histological diagnosis, or for the ability to provide prognostic information to family members and patients who may have progressive multi-focal leukoencephalopathy (PML) associated with AIDS.
Functional Neurosurgery Functional neurosurgery is dependent on high resolution multiplanar imaging. In the 1980s
pain management using deep brain stimulation techniques (prior to the advent of MRI), CT based imaging proved to be quite accurate for identification of targets in the ventromedial thalamus or periaqueductal gray. We now prefer high resolution intraoperative MRI for recognition of both pallidal, thalamic, and subthalamic targets [16,18,36,37]. > Table 40‐4 provides a summary of our usage of intraoperative imaging for functional neurosurgery over the last 25 years. MRI provides superior anatomic resolution to clearly delineate white matter tracts and gray matter nuclei. Currently, stereotactic MRI is used primarily for electrode placement during movement disorder surgery, or for ablative surgery [37]. CT is used in patients who already have stimulation hardware or have a contraindication to MRI. Using fusion software we can merge preoperative MRIs with intraoperative CT for functional targeting if needed. We also have used CT-based targeting for cellular transplantation research in patients with stroke [38].
CT-based Stereotactic Endoscopic Surgery One of the most gratifying combinations of current technology has been the role of combined CT stereotactic frame-based assisted craniotomy coupled with endoscopic removal of selected . Table 40‐4 Functional stereotactic brain surgery: UPMC-Presbyterian; 1979–2007 No. patients Pallidotomy Thalamotomy Deep brain stimulation (movement disorders) Depth electrodes for seizures Deep brain stimulation (chronic pain) Cell transplantation Mesencephalotomy/Capsulotomy Total
147 72 98 95 24 20 4 460
CT in image guided surgery
deep seated colloid cysts and brain tumors [39,40]. Over the last 5 years, we switched to performing almost all colloid cyst removal or other intraventricular lesion removals using a small trephine craniotomy at the coronal suture region, followed by stereotactic placement of an ‘‘endoport’’ (> Figure 40‐4) [39]. Through this 11 mm. conduit, the visualization endoscope can be placed. Standard or specially constructed microsurgical instruments can be used to remove colloid cysts or brain tumors in the third or the lateral ventricles. The ability to do both intraoperative pre-planning CT imaging as well as intraoperative post-procedure CT has assisted us in our determination of adequate or complete resection of such lesions. Surgical morbidity remains low due to the limited transcortical dissection necessary, and the lack of a transcallosal section or retraction of the medial hemisphere.
Frameless CT- based Stereotactic Surgery Beginning in the early 1990s, we began to evaluate the first generation of frameless stereotactic systems to assist localization and resection of mass lesions of the brain. First using the ISG
40
wand system, a mechanically assisted arm, with image integration, we recognized early the value for lesion localization and surgical planning. Although we believe that the precision of frameless stereotactic surgery is less than frame based surgery, for many tumor craniotomies, this is sufficient. Using image-guidance, ideally placed and smaller craniotomies, using trephine or standard technique are the norm. We have used and evaluated almost every commercially available system. While each has merit, we continue to see new improvements such as image fusion (CT, MRI, fMRI, MEG) to define targets. We use imaging compatible fiducials attached to the patient’s scalp or new mask systems if the head is not rotated. At our university teaching hospital thousands of patients have undergone imageguided intracranial and skull base procedures. Various manufacturers have redesigned freehand or probe holders during image-guided MRI and CT and frameless procedures. These products are reminiscent of some of the old burr hole mounted systems first developed in the 1960s. Such devices are widely marketed for routine diagnostic or therapeutic procedures. These include brain tumor biopsies, cyst management, and now even deep brain stimulation electrode placement. Such procedures offer benefit over old
. Figure 40‐4 Stereotactic endoscopic resection of a colloid cyst is performed via an 11 mm conduit and a 22 mm trephine craniotomy
625
626
40
CT in image guided surgery
free-hand methods. However, in the present era, the benefit of frameless as opposed to frame-based stereotactic surgery is hard to assess. Frame based systems allow you to pre-plot precise probe placement to identify the exact target to be biopsied, and assure rigid fixation and delivery of the probe or device. Burr hole or otherwise mounted stereotactic frameless systems with trajectory recalculations during the actual probe passage may be associated with a higher risk of morbidity. The learning curve for frameless stereotactic systems may seem rapid, whereas the training and educational needs for a neurosurgeon to become proficient in frame-based stereotactic surgery may be greater. We believe that patients continue to benefit from the proper use of frame based systems, where exact planned trajectories are used and exact brain targets are manipulated. If one performs frameless stereotactic surgery, outcomes should be compared to established benchmarks using rigid frame fixation. This is especially true for deep seated lesions in the basal ganglia, thalamus, pons and medulla, and for lesions in other high risk locations of the brain such as the perisylvian area.
Development and Evolution of a Dedicated Stereotactic Operating Room In the late 1970s, our planning process for a new stereotactic operating room debated the value of bi-plane angiography versus conventional radiological imaging techniques. Fortunately, the rapid development of CT imaging clearly pointed to the need for a marriage between neurosurgical techniques and neuroimaging devices, elegantly proposed by Professor Erik-Olof Backlund. We recognized early that a dedicated operating room for stereotactic procedures was highly desirable. We were able to secure final approval for a certificate of need for a dedicated therapeutic CT scanner housed in our operating room in 1981.
A GE 8800 CT scanner was placed in a newly designed operating room. The scanner was inverted from its usual position, so that the patient and the head came through the back part of the scanner into the main area of the operating room [8,41,42]. We also integrated this with a ceiling mounted fluoroscopic system, which could be used to provide real time fluoroscopic imaging, which was done to assist certain CT guided stereotactic procedures including transsphenoidal approaches. Similarly, the C-arm fluoroscope was also used in over 1,000 patients who underwent percutaneous trigeminal neuralgia management (mostly glycerol rhizotomy) during the next 20 years. In 1991, the scanner was updated to a GE 9800. During this interval, several thousand patients underwent CT assisted stereotactic procedures. A routine stereotactic biopsy could be completed in 60 min. The entire procedure time was greatly improved by our lack of need to move the patient between surgical and diagnostic imaging sites. The life span of such an advanced technology operating room suite is only about 10 years, because of changes in the field and advances in surgical and imaging technologies. In particular, we felt that this was true related to the emerging and increasing role of endoscopic surgical management especially for intraventricular and skull base lesions. Accordingly, we set about re-designing a new operating room using new CT scan imaging technologies, eventually placing a 64 slice GE CT scanner with fluoroscopic capabilities (> Figure 40‐5). This expanded operating room suite also provides video-assisted surgical techniques for endoscopic surgery, permitting simultaneous minimally invasive procedures with almost real time imaging. Because we do not have any of the issues associated with high magnetic fields of MRI, the standard operation paradigm remained in place. No tools needed to be redesigned, no special instruments needed to be created, and anesthesia services remained the same. Again, our goal was
CT in image guided surgery
40
. Figure 40‐5 The new dedicated CT stereotactic operating room at UPMC Presbyterian (Courtesy of Burt Hill Architects)
to enhance the operation, not to create a difficult working environment foreign to the best current neurosurgical procedures. We evaluated the role of intraoperative MRI scan over these years, but were concerned about low resolution of the resistive 0.1 and 0.3 Tesla versions. Most of the prototype units had poor image quality, and limited field of view. Certainly new systems have been developed based on pioneering efforts done in Calgary, Minneapolis, Newark, New Jersey, and in Erlangen, Germany. Numerous investigators have shown the potential abilities to better resect certain tumors, although conclusive data is lacking. As noted above, we think the primary thrust of intraoperative MRI to facilitate aggressive resection of glial tumors is problematic. Perhaps a small minority of clinically recognized gliomas, those with circumscribed tumors in polar or lobar locations, may have better outcomes. Overall, we continue to use high resolution imaging with CT. The newest device has dramatic improvement in acquisition times, reformatting in multiple planes, fusion techniques, and
real time CT fluoroscopy. Our new operating room allows us to fuse preoperative or intraoperative MRI’s and intraoperative and postoperative CT, especially valuable in functional neurosurgical cases where electrodes are being implanted. This is facilitated by using the Surgical Planning System SPS (Elekta, Inc.) which allows image fusion of sterereotactic, non-stereotactic, and other imaging output such as MEG and PET. Image-guided neurosurgery requires technologies that assist our abilities to deal with a wide variety of brain and spine problems and to reduce patient morbidity. In today’s era, the practicing surgeon or surgeon in training should never have to ask the question, ‘‘Where is it?,’’ that was so common years ago.
References 1. Perry JH, Rosenbaum AE, Lunsford LD, Swink CA, Zorub DS. Computed tomography-guided stereotactic surgery: conception and development of a new stereotactic methodology. Neurosurgery 1980;7:376-81.
627
628
40
CT in image guided surgery
2. Rosenbaum AE, Lunsford LD, Perry JH. Computerized tomography guided stereotaxis: a new approach. Appl Neurophysiol 1980;43:172-3. 3. Kondziolka D, Lunsford LD. Intraoperative navigation during resection of brain metastases. Neurosurg Clin N Am 1996;7:267-77. 4. Leksell L, Jernberg B. Stereotaxis and tomography: a technical note. Acta Neurochir (Wien) 1980;52:1-7. 5. Roberts TS, Brown R. Technical and clinical aspects of CT-directed stereotaxis. Appl Neurophysiol 1980;43:170-1. 6. Coffey RJ, Lunsford LD. The role of stereotactic techniques in the management of craniopharyngiomas. Neurosurg Clin N Am 1990;1:161-72. 7. Coffey RJ, Lunsford LD. Stereotactic surgery for mass lesions of the midbrain and pons. Neurosurgery 1985;17:12-18. 8. Lunsford LD. A dedicated CT system for the stereotactic operating room. Appl Neurophysiol 1982;45:374-8. 9. Lunsford LD. Stereotactic drainage of brain abscesses. Neurol Res 1987;9:270-4. 10. Lunsford LD, Latchaw RE, Vries JK. Stereotactic implantation of deep brain electrodes using computed tomography. Neurosurgery 1983;13:280-6. 11. Lunsford LD, Leksell L, Jernberg B. Probe holder for stereotactic surgery in the CT scanner: a technical note. Acta Neurochir (Wien) 1983;69:297-304. 12. Lunsford LD, Nelson PB. Stereotactic aspiration of a brain abscess using a ‘‘therapeutic’’ CT scanner. A case report. Acta Neurochir (Wien) 1982;62:25-9. 13. Lunsford LD, Parrish R, Albright L. Intraoperative imaging with a therapeutic computed tomographic scanner. Neurosurgery 1984;15:559-61. 14. Lunsford LD, Rosenbaum AE, Perry J. Stereotactic surgery using the ‘‘therapeutic’’ CT scanner. Surg Neurol 1982;18:116-22. 15. Lunsford LD, Woodford J, Drayer BP. Cranial computed tomographic demonstration of intracranial penetration by an orbital foreign body. Neurosurgery 1977;1:57-9. 16. Kondziolka D, Dempsey PK, Lunsford LD, Kestle J, Dolan E, Kanal E, Tasker RR. A comparison between magnetic resonance imaging and computed tomography for stereotactic coordinate determination. Neurosurgery 1992;30:402-7. 17. Lunsford LD. Magnetic resonance imaging stereotactic thalamotomy: report of a case with comparison to computed tomography. Neurosurgery 1988;23:363-7. 18. Kondziolka D, Flickinger JC. Current use of magnetic resonance imaging in stereotactic surgery: an ASSFN member study. Stereotact Funct Neurosurg 1997;66:193-7. 19. Lunsford LD, Martinez AJ, Latchaw RE. Stereotaxic surgery with a magnetic resonance – and computerized tomography-compatible system. J Neurosurg 1986;64:872-8.
20. Kondziolka D, Lunsford LD. The role of stereotactic biopsy in the management of gliomas. J Neurooncol 1999;42:205-13. 21. Kondziolka D, Lunsford LD. Stereotactic biopsy for intrinsic lesions of the medulla through the long-axis of the brainstem: technical considerations. Acta Neurochir (Wien) 1994;129:89-91. 22. Kondziolka D, Lunsford LD. Results and expectations with image-integrated brainstem stereotactic biopsy. Surg Neurol 1995;43:558-62. 23. Coffey RJ, Lunsford LD. Diagnosis and treatment of brainstem mass lesions by CT-guided stereotactic surgery. Appl Neurophysiol 1985;48:467-71. 24. Kondziolka D, Adelson PD. Technique of stereotactic biopsy in a 5-month old child. Childs Nervous Syst 1996;12:615-18. 25. Field M, Witham TF, Flickinger JC, Kondziolka D, Lunsford LD. Comprehensive assessment of hemorrhage risks and outcomes after stereotactic brain biopsy. J Neurosurg 2001;94:545-51. 26. Kondziolka D, Lunsford LD. Stereotactic aspiration of colloid cysts. J Neurosurg 1993;79:965-6. 27. Kondziolka D, Lunsford LD. Stereotactic techniques for colloid cysts: roles of aspiration, endoscopy, and microsurgery. Acta Neurochir Suppl 1994;61:76-8. 28. Kondziolka D. Intracavitary irradiation with colloidal phosphorus-32 for management of arachnoid cysts. Minim Invasive Neurosurg 1997;40:55-8. 29. Lunsford LD, Pollock BE, Kondziolka DS, Levine G, Flickinger JC. Stereotactic options in the management of craniopharyngioma. Pediatr Neurosurg 1994;21 Suppl 1:90-7. 30. Niranjan A, Witham T, Kondziolka D, Lunsford LD. The role of stereotactic cyst aspiration for glial and metastatic brain tumors. Can J Neurol Sci 2000;27:229-35. 31. Pollack IF, Lunsford LD, Slamovits TL, Gumerman LW, Levine G, Robinson AG. Stereotaxic intracavitary irradiation for cystic craniopharyngiomas. J Neurosurg 1988;68:227-33. 32. Coffey RJ, Lunsford LD. Factors determining survival of patients with malignant gliomas diagnosed by stereotactic biopsy. Appl Neurophysiol 1987;50:183-7. 33. Coffey RJ, Lunsford LD, Taylor FH. Survival after stereotactic biopsy of malignant gliomas. Neurosurgery 1988;22:465-73. 34. Lunsford LD, Somaza S, Kondziolka D, Flickinger JC. Brain astrocytomas: biopsy, then irradiation. Clin Neurosurg 1995;42:464-79. 35. Lunsford LD. Stereotactic drainage of brain abscesses. J Neurosurg 1989;71:154. 36. Kondziolka D, Bonaroti E, Baser S, Brandt F, Kim YS, Lunsford LD. Outcomes after stereotactically guided pallidotomy for advanced Parkinson’s disease. J Neurosurg 1999;90:197-202.
CT in image guided surgery
37. Lee JYK, Kondziolka D. Thalamic deep brain stimulation for management of essential tremor. J Neurosurg 2005;103:400-3. 38. Kondziolka D, Steinberg G, Wechsler L, et al. Neurotransplantation for patients with subcortical motor stroke: a phase 2 randomized trial. J Neurosurg 2005;103:38-45. 39. Harris AE, Hadjipanayis CG, Lunsford LD, Lunsford AK, Kassam AB. Microsurgical removal of intraventricular lesions using endoscopic visualization and stereotactic guidance. Neurosurgery 2005;56:125-32.
40
40. Kondziolka D, Lunsford LD. Microsurgical resection of colloid cysts using a stereotactic transventricular approach. Surg Neurol 1996;46:485-92. 41. Lunsford LD, Kondziolka D, Bissonette DJ. Intraoperative imaging of the brain. Stereotact Funct Neurosurg 1996;66:58-64. 42. Lunsford LD, Martinez AJ. Stereotactic exploration of the brain in the era of computed tomography. Surg Neurol 1984;22:222-30.
629
Image Guided Neurosurgery
35 Engineering Aspects of Electromagnetic Localization in Image Guided Surgery E. C. Parker . P. J. Kelly
Introduction
General Principles
As the concepts of stereotactic surgery have evolved over the past few decades, so has the equipment developed and adapted for its use. Head frame systems that were initially used for point in space stereotaxis were modified to allow for volumetric procedures allowing the visualization of tumors without the need for making translations of the frame coordinates to define the borders of target structures. Over the past decade, however, use of frameless stereotactic systems has become the dominant means of performing open image guided brain surgery. Although many frameless systems allow definition of tumor volumes, the common use of these devices again implements point in space stereotaxis, but with the convenience of real time feedback and multiplanar image reconstruction. All such frameless systems make use of a digitizer system to correlate the position of the patient’s head and a pointer of some type with a stereotactic coordinate system that has been applied to an imaging database. These systems have included articulated arms with angle sensors at each joint, as well as sonic, optical, and electromagnetic digitizers. Of these, only optical and electromagnetic systems remain in widespread use, with optic devices being the more prevalent. Each of these systems has unique characteristics that determine its applicability to a given circumstance. This chapter with focus on the unique aspects of electromagnetic digitizers in their application to image guided surgery.
Electromagnetic digitizers allow identification of the position and angular orientation of a movable receiver with respect to a transmitter that is generating an electromagnetic field. Such systems were described as early as the 1970s and are now readily available [1,2]. They have been adapted to a wide range of uses including multiple neurosurgical applications [3–7]. A commercially available electromagnetic digitizer, the Flock of Birds (Ascension Technology, Burlington, VT) [8] was first adapted by Kelly et al. for use in a stereotactic system [9]. This application evolved into the Cygnus frameless stereotactic system (Compass International, Rochester, MN), a commercially available neurosurgical stereotactic system that exclusively utilizes an electromagnetic digitizer. Other navigational systems using electromagnetic digitizer systems are now available as well. All frameless stereotactic systems for neurosurgical applications consist of several key components. Unique to magnetic systems are the electromagnetic transmitter and receiver. As with any type of stereotactic system, a computer software package that relates the position and orientation data to the patient’s preoperative imaging database is necessary to transform this digital information into a graphical form that can be readily utilized by the surgeon to facilitate preoperative and intraoperative decision making. The conventional setup for a magnetic system requires that the transmitter be fixed relative
#
Springer-Verlag Berlin/Heidelberg 2009
536
35
Engineering aspects of electromagnetic localization in image guided surgery
to the position of the patient’s head. This is usually accomplished by the use of an adaptor that rigidly secures the transmitter to a multi-pin head fixation system. This transmitter generates a magnetic field that encompasses the region of anatomical interest. A sensor can then be manipulated within this magnetic field and its relative position in space conveyed to the stereotactic computer system. Although such sensors are relatively small (2.5 cm or smaller) they can be coupled to a probe, suction tip, or other instrument of fixed size and orientation for intraoperative use, allowing the surgeon to utilize the instrument in an area too small for the magnetic sensor. Such a system is illustrated in > Figures 35-1–> 35-3. An arrangement of this type can be used in a variety of ways. It is possible to define the contours of a tumor on individual imaging slices, thereby creating a tumor volume that can be visualized along the axis of the localizing probe. More commonly, however, the position or trajectory of the probe tip will be utilized to plan the position of a scalp, bone, and dural opening and to help confirm spatial orientation during tumor resection. Electromagnetic stereotaxis can also lend itself to situations where it is desirable not to place the patient’s head in rigid fixation during surgery. This is often the case during sinus procedures performed by otolaryngologists who are not accustomed to placing the patient in pins. For this type of arrangement, a second electromagnetic sensor may be attached to the patient’s head, which can then be positioned and moved as desired. Through the use of two electromagnetic receivers, the position of both the patient’s head and the localizing probe are tracked simultaneously in relation to the transmitter, which is rigidly secured to the operating table. Alternatively, a small transmitter may be secured directly to the patient’s head, achieving the same result.
. Figure 35-1 Examples of an (a) electromagnetic transmitter, and (b) receiver
Accuracy Of prime importance in any frameless stereotactic system is accuracy. Ascension Technology, manufacturer of the Flock of Birds electromagnetic digitizer, claims positional accuracy of 1.8 mm and orientation accuracy of 0.5 with the sensor in a range of between 20.3 and 76.2 cm from the transmitter. Testing has confirmed equal or better accuracy with mean position error of 0.5 mm and maximum error of
Engineering aspects of electromagnetic localization in image guided surgery
35
. Figure 35-2 Electromagnetic transmitter secured to the head holder in preparation for surgery
. Figure 35-3 Electromagnetic receiver being used to confirm location during surgery
1 mm within an optimal operational zone of 22.5–64.0 cm [10]. The accuracy of magnetic localization when applied to neurosurgical stereotaxis has been well documented. During development of what would become the Cygnus system a magnetic digitizer was adapted for use with the COMPASS head frame (Compass International, Rochester, MN) and testing revealed average three dimensional errors of approximately 3 mm [5,9]. This testing was done using known frame reference points to minimize inaccuracy introduced during image registration. Further testing has been done under more clinically relevant circumstances. This testing has included comparison to various other digitizer types, including articulated arm and optical based systems. Measurement of spatial accuracy of the Cygnus system revealed a mean accuracy of 1.90 0.7 when tested on a stereotactic phantom and registered with standard surface fiducial markers that had been attached to the surface of the phantom [11]. Such testing takes into account not only the positional accuracy of the
537
538
35
Engineering aspects of electromagnetic localization in image guided surgery
digitizer system, but also the process of registering the position of the frame to the imaging database. This concept was taken a step further and performed intraoperatively on a series of 70 patients. Measurements of error from intracranial anatomical landmarks was 1.4 0.6 mm, again using the Cygnus system [12]. In each of these studies the accuracy of the magnetic based system compared very favorably to an optical tracking system tested in the same manner.
Magnetic Interference Metallic objects in the operating room environment may produce distortion of the magnetic field used for localization of the electromagnetic sensor. This is a chief concern regarding use of magnetic stereotactic systems. Such distortion can be quite significant. Relative sensitivity of magnetic tracking systems to various metallic objects is dependent on the type of current employed by the transmitting device. Transmitters based on alternating or direct current will react quite differently in the presence of ferromagnetic and electrically conductive material. Tracking reliability of a direct current based system is relatively unaffected by nonferromagnetic materials including aluminum. The presence of an aluminum object within a few centimeters of the sensor can cause significant deviation in both position and orientation in an alternating current system, on the other hand [13]. These deviations can be greater than 2.5 mm and 2 in position and orientation respectively when the aluminum object is in close proximity to the electromagnetic sensor. When tested in the presence of metal cylinders composed of titanium, stainless steel, cobalt chrome, aluminum, and mild steel, a pulsed direct current system was significantly affected only by the mild steel sample [10]. Therefore, given the relative prevalence of these conductive but nonferromagnetic metals in the typical operating environment, direct current magnetic tracking provides a more stable platform for stereotaxis.
Although the effects of nonferromagnetic but highly conductive materials on direct current tracking systems are minor, large aluminum objects, for example, can cause the introduction of some error. This effect can be minimized by altering the sampling rate of the electromagnetic sensor system. Aluminum’s high electrical conductivity causes the creation of eddy currents when exposed to a magnetic field. With a pulsed magnetic field, these currents form and then rapidly decay. A relatively low pulse frequency will allow more time for these currents to resolve before sampling occurs. Eddy currents in highly ferromagnetic material, on the other hand, do not decay as quickly and generate secondary magnetic fields that increase interference. Therefore, a lower sampling rate (20 Hz) will reduce error associated with aluminum, while a higher sampling rate (140 Hz) 0has been demonstrated to moderate the effects of steel [14]. It should be noted that even at this high sampling rate, the presence of ferromagnetic steel in the operating environment will still introduce significant error with the use of a direct current system and it should be avoided. Another source for potential interference in the function of magnetic tracking systems is electromagnetic background noise in the operating room. This interference may result from wires, lights, electrical equipment, etc. Comparison of background noise between a shielded room and a typical operating room, however, has demonstrated only very minimal effect from this ambient interference. Significant interference and increase in noise can be seen when electrical or electronic equipment is used in close proximity to the electromagnetic receiver. This phenomenon is diminished dramatically by positioning these devices further from the receiver. Distances as small as 30 cm can be effective in eliminating this problem [15]. Generally speaking, however, any nonessential metallic or electronic devices should be positioned away from the surgical field. While the interference resulting from metallic objects can be minimized, it cannot be
Engineering aspects of electromagnetic localization in image guided surgery
eliminated altogether. This is a result of the presence of large conducting objects (the operating table, Mayfield or Gardner head holder, retractors, etc) that will create some distortion in the magnetic field and introduce some error into the system. One method for dealing with this problem is to utilize a calibrated distortion phantom. Such a device allows the registration of a three dimensional matrix of points of know spatial relationship within the surgical field, prior to registration of the patient’s head position. Any distortion of the magnetic field will cause unexpected discrepancies in the relative positions of these points. The stereotactic computer can then use these data to calculate correction factors for all points within this calibrated area. Unfortunately, use of this method requires the registration of many points prior to surgery and is quite cumbersome. Additionally, any metallic objects brought into the field after this calibration has taken place (retractors for example) will alter the distortion, possibly reducing the accuracy of the correction factors. A more accessible approach to minimizing error associated with metallic interference is to monitor the presence of metallic objects within the magnetic field. The most recent version of the
35
Flock of Birds system tracks the error introduced by the proximity of highly electrically conductive (aluminum) or magnetically permeable (steel) to either the transmitter or sensor. Although the system cannot correct for the distortion introduced by these metals, it can report their relative presence and the likelihood of interference to the user. The latest version of the Cygnus software relays this information in a graphical form. A meter in the lower left hand corner of the display reports the relative electromagnetic interference being observed by the system (> Figure 35-4). The number and height of the meter rise as more distortion is noted within the magnetic field. A higher number serves as a warning to the user that there is an increased relative risk of error in reporting of position and orientation of the sensor and that metal objects such as retractors should be moved or taken out of the field for optimal performance. A final issue involving the magnetic interference and electromagnetic stereotaxy is interference caused by the transmitter itself. It has been our experience that the magnetic transmitter creates noise in neurophysiological monitoring equipment. This prevents the monitoring of evoked potentials while the magnet is active.
. Figure 35-4 Interference warning indicator reporting potential distortion within the magnetic field secondary to the presence of metallic objects
539
540
35
Engineering aspects of electromagnetic localization in image guided surgery
This problem is easily solved by placing the transmitter in standby mode when not in use. Since the frameless system is generally used only intermittently to confirm position or orientation, this issue has not been viewed as a serious limitation of the technology.
can be mounted to a tower in a more conventional, non-mobile, setup, it can also be stored in a small carrying case. This allows easy mobility from hospital to hospital if a surgeon covers more than one facility and dedicated systems are not available at each location.
Advantages of Electromagnetic Stereotactic Systems
Future Directions
Electromagnetic tracking systems hold some significant advantages over the more widely used optical systems. The primary advantage of magnetic systems is their unobtrusiveness during surgery [16]. In the typical arrangement, the magnetic transmitter is placed outside the sterile field under the drapes. The most convenient location for securing this transmitter is to utilize the extra starburst attachment at the base of the Gardner or Mayfield head holder. This is preferable to attaching the transmitter directly to the operating table as adjustments can be made in the position of the arm securing the head clamp to the table, and thus ˜ s head position, without altering the the patientO spatial relationship between the transmitter and patient. This arrangement requires use of an extension bracket approximately 20 cm in length. In addition to allowing room for securing the transmitter away from the head and pin apparatus, this bracket provides optimal spacing between the transmitter and the operative field, minimizing inaccuracy. The only part of the frameless system present on the sterile field is the sensor, usually attached to a blunt tipped probe. This sensor is gas sterilized and is connected to the receiver system by a thin cable allowing easy manipulation during surgery. No line of sight is necessary with this arrangement as it is with optical systems. Another advantage to electromagnetic frameless technology is its relatively small size and transportable nature. Although such a system
As with most technology, electromagnetic sensors have been introduced that are much smaller that older models. Sensors capable of providing six degrees of freedom (three positional and three rotational) are available in sizes under 2 mm in diameter, and those providing five degrees of freedom (three positional and two rotational) are as small as 0.55 mm in diameter. Sensors of this type have been found to be accurate enough for surgical applications [17]. Such small sizes allow placement of the sensor at the tip of a flexible device as opposed to the usual attachment of frameless stereotactic localizers to the handle of a rigid instrument. This has already been successfully utilized in bronchoscopy applications, allowing navigation to lesions that would otherwise be unreachable by conventional visual navigation [18,19]. Although flexible endoscopy has seen limited use within neurosurgery, these types of sensors could certainly be adapted to this application.
References 1. Kuipers JB. SPASYN – an electromagnetic relative position and orientation tracking system. IEEE Trans Instrum Meas 1980;29(4):462-6. 2. Raab FH, Blood EB, Steiner TO, Jones HR. Magnetic positioin and orientation tracking system. IEEE Trans Aerosp Electron Syst 1979;AES-15(5):709-17. 3. Fried MP, Kleefield J, Gopal H, Reardon E, Ho BT, Kuhn FA. Image-guided endoscopic surgery: results of accuracy and performance in a multicenter clinical study using an electromagnetic tracking system. Laryngoscope 1997;107:594-601.
Engineering aspects of electromagnetic localization in image guided surgery
4. Fried MP, Kleefield J, Taylor R. New armless imageguidance system for endoscopic sinus surgery. Otolaryngol Head Neck Surg 1998;119:528-32. 5. Rousu JS, Kohls PE, Kall B, Kelly PJ. Computer-assisted image-guided surgery using the regulus navigator. Stud Health Technol Inform 1998;50:103-9. 6. Sagi HC, Manos R, Benz R, Ordway NR, Connolly PJ. Electromagnetic field-based image-guided spine surgery part one: results of a cadaveric study evaluating lumbar pedicle screw placement. Spine 2003;28:2013-18. 7. Sagi HC, Manos R, Park SC, Von Jako R, Ordway NR, Connolly PJ. Electromagnetic field-based image-guided spine surgery part two: results of a cadaveric study evaluating thoracic pedicle screw placement. Spine 2003;28: E351-4. 8. Ascension Technology Corporation, http://www.ascension-tech.com/. 9. Goerss SJ, Kelly PJ, Kall B, Stiving S. A stereotactic magnetic field digitizer. Stereotact Funct Neurosurg 1994;63:89-92. 10. Milne AD, Chess DG, Johnson JA, King GJ. Accuracy of an electromagnetic tracking device: a study of the optimal range and metal interference. J Biomech 1996;29:791-3. 11. Benardete EA, Leonard MA, Weiner HL. Comparison of frameless stereotactic systems: accuracy, precision, and applications. Neurosurgery 2001;49:1409-15; discussion 1415-6.
35
12. Mascott CR. Comparison of magnetic tracking and optical tracking by simultaneous use of two independent frameless stereotactic systems. Neurosurgery 2005;57: 295-301; discussion 295-301. 13. Birkfellner W, Watzinger F, Wanschitz F, Enislidis G, Kollmann C, Rafolt D, Nowotny R, Ewers R, Bergmann H. Systematic distortions in magnetic position digitizers. Med Phys 1998;25:2242-8. 14. LaScalza S, Arico J, Hughes R. Effect of metal and sampling rate on accuracy of Flock of Birds electromagnetic tracking system. J Biomech 2003;36:141-4. 15. Poulin F, Amiot LP. Interference during the use of an electromagnetic tracking system under OR conditions. J Biomech 2002;35:733-7. 16. Mascott CR. The Cygnus PFS image-guided system. Neurosurgery 2000;46:235-8. 17. Hummel J, Figl M, Kollmann C, Bergmann H, Birkfellner W. Evaluation of a miniature electromagnetic position tracker. Med Phys 2002;29:2205-12. 18. Hautmann H,Schneider A, Pinkau T, Peltz F, Feussner H. Electromagnetic catheter navigation during bronchoscopy: validation of a novel method by conventional fluoroscopy. Chest 2005;128:382-7. 19. Schwarz Y, Greif J, Becker HD, Ernst A, Mehta A. Realtime electromagnetic navigation bronchoscopy to peripheral lung lesions using overlaid CT images: the first human study. Chest 2006;129:988-94.
541
45 Image Guided Craniotomy for Brain Tumor I. E. McCutcheon
The application of technology to the surgical resection of intracranial tumors has involved a gradual accretion of instruments and instrument systems ever since such tumors were first successfully removed in the late nineteenth century. This background of gradual improvement has been punctuated by explosions of activity in specific areas that have greatly expanded the boundaries of neurosurgical practice. These include the invention of the bipolar cautery, the introduction of angiography and ventriculography for diagnosis, the advent of the intraoperative microscope, and the development in the 1970s and 1980s of computerized imaging systems such as computed tomography (CT) and magnetic resonance imaging (MRI). The intracranial images that these modalities now provide offer a wealth of anatomical detail and delineate tumor location and tumor extent with precision. The gradual fusion of improved imaging with stereotactic localizers has yielded the latest such revolution, that of neuronavigation. This generic term covers a variety of systems designed to provide technical solutions to the basic problems of (1) Correlating a lesion seen on scan with the anatomical reality of the patient, (2) Obtaining a maximal resection of that lesion, while (3) Breaching normal brain tissue to the smallest degree compatible with an adequate resection. The original mode of image guidance during tumor surgery was through a CT- or MRIcompatible frame-based system of stereotactic localization. This has largely given way to frameless systems that permit navigation by comparison of patient anatomy during surgery with images obtained before surgery. Such systems are now #
Springer-Verlag Berlin/Heidelberg 2009
widely available and significant experience has accrued with their use in tumor resection. Recent efforts shift the time of imaging from preoperative to intraoperative stem from the increasing availability of systems of intraoperative CT and MRI. This chapter will discuss the available technology, its application in intracranial tumor surgery, and the advantages and pitfalls attendant upon its use.
History and Methodology The stereotactic approach to cerebral localization was originally described by Horsley and Clarke in 1906 [1]. Their fairly complicated apparatus and its successors (notably Leksell’s innovative frame system designed for use with ventriculography) had relatively limited utility in tumor resection until the advent of CT technology, which allowed a shift from localizing single points within a space (i.e., the brain) to describing volumes within that larger space [2]. Such description of volume was necessary for stereotactic resection as opposed to stereotactic biopsy. Although CT scans can be rendered in a multiplanar fashion through computerized reconstruction, the advent of MRI scanning has greatly expanded the ability to display tumors in coronal, sagittal, and axial planes. With the concomitant development of relatively powerful mini-computers, such threedimensional data sets could be rendered in ways useful to a surgeon seeking guidance in tumor resection. The first practical ‘‘volumetric’’ surgery rendered a localization target as a volume rather than a point, and placed images of that volume
700
45
Image guided craniotomy for brain tumor
into a convenient and accessible visual display for use by the surgeon during surgery. The pioneer of this approach was Patrick Kelly, who developed a frame-based system called the COMPASS stereotactic system (Stereotactic Medical Systems), which he designed specifically for volumetric tumor resection [3–5]. This device allows projection of scaled planar images onto a heads up display within the operating microscope. This display projects the images onto a real-time image of the brain surface, and allows the surgeon to define the radiological abnormality precisely during surgery irrespective of any grossly visible (or invisible) abnormality. Given that the margins of a glioma are usually indistinct due to the infiltrative quality of such tumors, such technology enhanced the surgeon’s ability to achieve more complete resections of the region identified as tumor on scan. Foreshadowing the trajectory of development of more recent systems for surgical navigation, Kelly later incorporated MRI, digital subtraction angiography, and magnetoencephalography into his system [6].
The next step in the evolution of neuronavigation was the advent of the frameless stereotactic method initiated by David Roberts in a seminal paper published in 1986 [7]. He described integration and display of CT images within the operating microscope, images that were specially registered by determining the position of the microscope as its focal point was placed on three fiducial markers placed on the scalp and evident in the CT images (> Figure 45-1). This paper marked the true beginning of the avalanche of activity in surgical navigation that continues to this day. All subsequent work can be traced back to this paper which established the concept of fiducial markers on the scalp, rather than suspended in space at various points around the head as occurs with frame-based systems. In essence, stereotactic methodology and neuronavigation are used to map image space onto physical space. Many surgeons using such technology now prefer frameless systems, which are less cumbersome to the patient and more versatile, especially when volumetric resection (rather than point
. Figure 45-1 The stereotactic system of Roberts, which initiated modern frameless neuronavigation. The nonlinear array of three spark gaps, providing an acoustic localizing system, is seen protruding from the operating microscope. The optical projection system is attached to the side of the microscope (arrow), and the microphone array on the ceiling is not shown here [7]
Image guided craniotomy for brain tumor
biopsy) is being done. Such systems have four basic elements: (1) A method for co-registering images with physical space, (2) A device for intraoperative localization (such as a pointer), (3) A computer monitor that displays the images in multiplanar fashion, (4) Intraoperative feedback in real time (e.g., display of location of the localization device on the displayed images). Registration. Although displayed as slices, standard digital images of the brain obtained through CT or MRI are actually databases of three-dimensional volume. Mapping these image sets onto the physical reality of a patient’s anatomy allows them to be used for navigation. Such mapping, otherwise called registration, can use data acquired before or during surgery. The most common method of registration to date has been point-based, which defines linkage between points in the image versus points in the physical space to allow geometric linkage between the respective volumes. Typically, fiducial markers (detachable temporary tags of uniform size) placed on the surface of the patient’s face and scalp have been used. However, intrinsic anatomic landmarks can be used instead of, or as a supplement to, such registration to enhance its accuracy. Surface-based registration can also be done. This takes a number of points (usually
45
40 or more) chosen randomly on the contours of the patient’s surface anatomy, and fits them to similar contours within the previously acquired images. Accuracy is lower in surface-based methods than in point-based registration, and for that reason most systems depend upon the latter as their primary method of registration, and use the former as a fallback option should fiducial markers be displaced or absent. Intraoperative Localizers. A variety of localizing devices have been put into use for navigation in tumor resection. These have each had their proponents and each manufacturer claims superiority over its competitors’ systems. The truth is that no single system for surgical navigation has proven more accurate or more helpful than any other. Error can be introduced at any number of steps from preoperative imaging through registration, and can be further introduced by intraoperative events. However, when care is applied to the acquisition of the images, the placement and then localization of the fiducial markers, and the registration just prior to surgery, localization accuracy of 1–2 mm can generally be obtained. This accuracy degrades during surgery, largely due to the phenomenon of brain shift (> Figure 45-2). Systems that depend upon images acquired prior to surgery cannot be
. Figure 45-2 Brain shift during surgery, shown by intraoperative MRI captured (a) prior to skin incision (b) after pterional/ subfrontal craniotomy to resect a recurrent, invasive gonadotropin-secreting pituitary macroadenoma
701
702
45
Image guided craniotomy for brain tumor
corrected for such displacements without intraoperative updating of the imaging data set. Such displacements are predictably more profound at the brain surface and less evident internally, but they can still be significant at deep locations, particularly in the vicinity of critical structures like the brainstem or internal capsule. Either a surgeon must gain a sense of the degree of shift by repeated comparison of localization on the image display with specific and identifiable anatomic landmarks, or some degree of correction must be imposed by the acquisition of new realtime data. In an analysis of brain shift by Roberts et al. they found a mean displacement of 10.7 mm at the cortical surface, a deformation caused more by the effect of gravity than by other possible causes of such shift (including hyperventilation, loss of cerebrospinal fluid (CSF) and direct tissue loss as resection proceeds) [8]. Articulated Arms. The first description of a passive articulated arm for intracerebral localization came from Watanabe et al. [9,10]. They described a ‘‘neuronavigator’’ which included an
articulated arm with six joints and registration accomplished with scalp-based fiducial markers as Roberts et al. had proposed. This system gave way to the ISG viewing wand system (Elekta), which also has a six-jointed arm and electrogoniometers to detect the angle at each joint (> Figure 45-3). The arm is thus subject to electromagnetic interference, an important impediment in the operating room environment, and the physical limitations of the arm impose constraint on patient placement relative to the device. In particular, its use in conjunction with a microscope can be difficult. These systems are nonetheless quite accurate, with Golfinos et al. reporting accuracy within 2 mm in 92% of cases when localization was based on MRI [11]. While articulated arms need not be positioned in the line of site of an optical detector, they are physically limited in their reach and have given way in most centers to free-hand devices. However, they remain in use in some centers [12]. Sonic Digitizers. The system originally reported by Roberts et al. utilized spark gap generators
. Figure 45-3 ISG viewing wand. The articulated arm of this device is limited in its reach but has no line-of-sight constraint as is true for currently popular neuronavigational systems dependent on reflectance of infrared light
Image guided craniotomy for brain tumor
emitting an ultrasonic signal and attached to the operating microscope in a fixed configuration, with sound detectors placed at several locations in the operating room [7]. Similar ultrasonic emitters have since been placed on free-hand devices (probes or other surgical instruments) the location of which is determined by a nearby microphone array. Although generally reliable, such systems can be affected by noise or by physical obstruction within the line of site from the emitter to the detector. In addition, temperature fluctuation affects the speed of sound in air and temperatures are typically not constant within an operating room. For these reasons, most systems now use light emitting diode (LED) technology. Light Emitting Diode Systems. In such systems the light source can be a camera flashing a pulsed infrared light, which is reflected off coated spheres attached to the pointing probe. The reflected light is then detected by a charge coupled device. Examples of this include the Stealth (Medtronic) and Vector Vision (BrainLab) systems (> Figure 45-4). The spheres must not be fouled with blood and must be in the line of sight of the light source. An alternative concept places the LED on the probe itself which must then be attached to a power source, usually through a flexible cable. Such devices are less easy to manipulate then those using passive reflectors as described above. The LED can also be placed on the microscope, the lens of which has a known focal length and whose focal point represents the tip of virtual pointer, but such systems (e.g., the Viewscope or Zeiss MKM System) are hampered by line-of-sight issues particularly when used in operations on patients in the lateral position or when the scrub nurse’s instrument tray sits above the patient’s torso [13]. The Easy-Guide Neurosystem (Philips) is one example of a system with the LED on the pointing device. The SurgiScope (ISIS) is a combination of neuronavigation with robotics, in which the surgeon can specify a target or trajectory during preoperative planning,
45
. Figure 45-4 BrainLab vectorvision system for neuronavigation. The localizing probe reflects infrared light back to two light sources, and the resulting stereotactic data are processed in a mobile workstation with screen for displaying multiplanar images
and thereby program the microscope to act as a slave pointer. It can also be aimed actively at targets by the surgeon, converting its point of focus into a virtual probe as described above. Magnetic Field Systems. Instead of using reflected light as the mode of triangulation, some groups have used magnetic field guidance [14]. The field emanates from a transmitter near the patient’s head and is detected by a receiver on the surgical localizing probe. Although relatively simple, this technique is less accurate than other
703
704
45
Image guided craniotomy for brain tumor
methods as it is susceptible to field distortion within the surgical environment because of the abundance of metals there, and also because of the presence of electromagnetic fields from monopolar cautery and other devices. Although software algorithms have been applied to compensate for field inhomogeneity and therefore enhance the technique’s accuracy, the abundance of ferromagnetic instruments in cranial surgery (including typically the head clamp) make adoption of magnetic field localization somewhat problematic. Machine Vision (Passive Stereoscopic Video). This method requires no emitters or reflectors as it localizes the pointer’s position by determining positional differences on video images acquired by cameras placed at different angles. This technique was proposed by Heilbrun et al. and utilizes two video cameras one meter apart and eight fiducial markers in a box-like configuration of known dimensions [15]. This system has been reported to have a mean error of 2.1 mm in a series of 21 patients in whom its accuracy of localization was compared with that obtained by a BrownRoberts-Wells frame-based system (Radionics) [16]. However, it has not been brought to market, and line-of-sight considerations apply. Computer Displays. Each neurosurgical navigation system comes with an image-processing workstation preloaded with software specific for the system (> Figure 45-5). Peripherals are usually system-specific as well. Software upgrades are ongoing as clinicians detect previously unknown kinks in the various systems and manufacturers work to correct those flaws. It is typically possible for each surgeon to select from a menu of display options that allow standard multiplanar views (axial, coronal, sagittal) as well as trajectory and in-line views created from the three-dimensional dataset. Pre-contrast T1-or T2-weighted images can be loaded, as is appropriate in the case of low grade gliomas or non-enhancing tumors (> Table 45-1). Additional imaging modalities such MR spectroscopy, fMRI, and positron emission tomography (PET) can also be loaded into
the workstation but require additional software to permit cover overlay on standard images. Placement of the video display can be difficult as ideally the surgeon should not have to turn away from the operative field in order to see the monitor while holding the probe in place. With the presence of an anesthetic delivery system, instrument tables, ultrasound machine, and other bulky equipment it may be necessary to place the monitor in an inconvenient location. To get around this limitation, transfer of the images into the microscope to create a heads-up display has been helpful [17]. However, not all tumor cases are done through a microscope, thus headmounted displays have been proposed and will likely be increasingly available in the future [18]. Feedback During Surgery (Real-Time). Standard systems currently in place provide a navigational map based upon preoperative acquisition, a map which is not updated during the procedure. Because of the brain shift alluded to above, such maps can become increasingly inaccurate as the operation proceeds. Efforts at updating the preoperative map with new intraoperative data are underway and rely upon ultrasonography or upon intraoperative acquisition of new scans [19,20]. Cortical tracking by cameras mounted on a stereotactic microscope has also been used to predict deformation of the brain surface [21]. Strategies for applying intraoperative ultrasound to the problem of correction of tissue shift have been presented by several groups [8,22–25]. Algorithms for such compensation continue to be refined and there is no standard technique to achieve this. In one study, the mean displacement of cortical landmarks ranged from 0.8 to 14.3 mm, about half of which was due to reduction in tumor volume [26,27]. Shifting of superficial landmarks exceeded that of subcortical structures in all patients. Brain shift has not thus far been predictable prior to surgery, so intraoperative updating of images must be done to solidify this fourth component of surgical navigation [28].
Image guided craniotomy for brain tumor
45
. Figure 45-5 Screen view from intraoperative MRI environment (BrainSuite). Patient has an oligodendroglioma in the right frontal lobe undergoing craniotomy. Images were taken during surgery after partial resection was accomplished. Note that left-right conventions are reversed in this system. The probe is touching the lateral edge of the right lateral ventricle, medial to which significant tumor remains. The top row contains an intraoperative real-time ultrasonographic image on the left, and overlay of that image on intraoperative MRI on the right
. Table 45-1 Imaging for different tumor types Tumor type suspected
Image
Metastasis Meningioma with hyperostosis Meningioma with no hyperostosis Glioma, non-enhancing (low grade or occasional high-grade) Enhancing (high-grade) Tumors of skull base
MRI (T1, post-contrast) CT (pre- and post-contrast) MRI (T1, post-contrast) MRI (fast spin echo, no fat suppression; T2) MRI (T1, post-contrast, T2) MRI (T1, post-contrast, fat suppressed; T2) and CT (pre- and post-contrast)
705
706
45
Image guided craniotomy for brain tumor
In the subset of patients who undergo an awake craniotomy for intraoperative mapping of speech function, this issue takes on some importance. Although such operations can be done with the head fixed in a rigid clamp if a complete scalp block is performed, the technique is easier if the head is not fixed. A mobile head, however, requires updating of the registration with each movement. A reference sensor affixed to the retroauricular region has been described which can provide updating of the patient’s head position and recalculation of registration, making neuronavigation possible for an awake craniotomy without head fixation [29]. Infants and young children with deformable skulls form another subset of patients in whom rigid cranial fixation is not possible. For them a similar solution has been reported which involves screwing a reference arc to the outer table of the skull through a separate incision within the operative field [30]. Equally, innate anatomical fiducials might be used, although their localization would be subject to minor errors due to the same deformability that prevents application of a head clamp for immobilization in this age group. Ultrasonography for arterial and venous display using a color doppler is easy to acquire and can be imported into a navigational work station [31]. This adds information about hidden vessels and thereby increases safety and allows tailored surgical approaches to deep tumors near critical arteries. It also allows the vascular tree to be used as the focus for updating and provides yet another way of correcting for brain shift. Updating through intraoperative MRI scan (or in some cases CT) is the obvious solution to the dilemma. However, this poses problems of its own and discussion of this modality will be deferred to the end of the chapter.
Practical Issues Prior to Surgery The use of neuronavigation is best reserved for tumor patients undergoing surgery on an elective
basis. MRI scan provides more accuracy than CT, particularly for maximizing resection of gliomas where T2-weighted changes may reveal tumor that CT cannot show. Also, patients with pacemakers or other metal implants may not be able to undergo MRI, and CT will have to be used. Regardless of the type of scan, it is ideally performed either the day before surgery or the morning of the procedure to minimize the chance for dislodgment of fiducials. Such markers with a central indentation are affixed immediately prior to scanning and are placed abundantly so that registration can still be performed if one or two of the markers drop away in the interim. The markers are best placed on skin sites that are relatively immobile (> Table 45-2). For that reason, several are placed on the forehead, two at the vertex on the midline (in spots requiring a small amount of hair shave), two in the retroauricular area over the mastoids, and two on the right and left parietal areas (again, requiring a small shave). The total number is therefore at least nine, and two or three additional can be placed in the lateral supraorbital area or in shaved areas on the frontal scalp to enhance accuracy. The occipital area is avoided due to the folding of skin that occurs in this area with neck movement. If a patient is already bald, more freedom in placing fiducials is automatically obtained. In fixing the head with the threepoint clamp, pin placement should be carefully planned to avoid disturbing fiducials and in particular to avoid pulling the skin and shifting
. Table 45-2 Position of fiducial skin markers Position
Number
Root of mastoid (both) Midline at vertex (anterior, posterior) Forehead (lower left, lower right, middle, superior) Zygoma (both) Parietal bossa
2 2 4
a
2 2
Optional, depending on location of lesion (anterior vs. posterior within head)
Image guided craniotomy for brain tumor
them from their position on the images. A reference system, typically an array of reflective balls in the LED systems, is placed as far to the side of the patient as possible to avoid interference with the surgeon’s movements and those of the scrub nurse. Additionally, it must be placed within the line of sight of the infrared light generator. Registration then proceeds by touching the probe to the fiducial markers in sequence and importing the probe’s position into the system. The position of the reference array must similarly be imported by touching the probe to a divot placed on it for that purpose. Once registration has been completed to acceptable accuracy, the fiducial markers can be removed and shaving can commence to the degree appropriate for the operation.
Selection of Imaging The choice of imaging depends on the suspected identity of the tumor, and of course on the imaging features revealed in MRI scans done prior
45
to the surgical planning. Recommendations for imaging based on tumor type are given in > Table 45-1. Segmentation can be performed to bring out vascular structures, a maneuver that may be helpful in preserving those structures when they are displaced or encased by a bulky tumor (> Figure 45-6). In addition, overlay of image sets from fMRI or from MRI spectroscopy can also be performed, and can be very useful in extending the limits of the resection in non-eloquent areas and in identifying occult tumor [32–35]. PET has also been integrated into neuronavigation to provide metabolic information on tumor heterogeneity and extent. In one series a final target volume different from that obtained with MRI imaging alone was found in 80% of patients by FDG-PET and 88% by MET-PET [36]. The caveat here is that although it is logical to assume that application of functional or metabolic imaging maximizes tumor resection and extends survival, no proof has yet been published that this is indeed the case. In addition, fMRI is most secure in assigning hemispheric dominance rather than precise
. Figure 45-6 Segmentation of anatomic structures. (a) Venous anatomy. By subtracting the overlying layers of scalp, bone, and dura, and performing automated highlighting of slow-flow structures, display of veins can show their relation to tumor, a technique particularly helpful in resection of meningioma (b) Ventricles. Each is assigned a different color, and even the aqueduct of Sylvius can be seen. Simultaneous segmentation and coloring of neural elements can be done at the same time, as is the case here (the basal ganglia are noted just lateral to the lateral ventricles)
707
708
45
Image guided craniotomy for brain tumor
localization of cortical speech function. Intraoperative electrophysiological mapping remains the gold standard and cannot at this time be replaced by fMRI.
Surgical Planning It takes 10–20 min to register a patient at the beginning of an operation. Combining surface with fiducial registration is more time consuming, and may enhance accuracy, but in most cases simple fiducial registration is sufficient. One cadaver study has shown that fiducials give the smallest error in localizing targets of the various methods tested [37]. In checking error prior to the final step of accepting registration, it is important to test localization accuracy not just by touching fiducials but by touching skin in the proposed craniotomy zone. If this testing is off by more than a millimeter, than a ‘‘surface merge’’ is one option to enhance accuracy; another is simply to reject the registration and start again from scratch while taking care not to shift any fiducials when the probe touches them. Planning includes delineation of the incision, the underlying craniotomy opening, and the trajectories through the brain or skull base to the lesion that will least disrupt eloquent areas and vital vascular structures (> Figure 45-7). For deep lesions or tumors of the skull base, trajectory views can be very helpful in avoiding critical structures (> Figure 45-8). The surgeon should also decide whether proximity to eloquent cortical areas or important white matter tracts prevents complete resection, and if so which areas of tumor should be left in place. If multiple craniotomies are to be performed (e.g., when two or more metastases are being removed), planning includes consideration of how to create several competing incisions without compromising the neurovascular supply of the various scalp flaps. In reoperative cases, planning includes decisions on whether previous incisions and craniotomy
. Figure 45-7 Neuronavigation increases precision of localization of tumors. This patient had melanoma and three metastatic tumors in the left posterior frontal lobe. Circles have been drawn where the surgeon predicted the tumors’ locations, and additional circles after localization with StealthSystem neuronavigation (which correlated perfectly with true location once surgery proceeded). The anterior lesion was predicted correctly; however, the two posterior lesions were not, and one was placed by the surgeon across the midline, an error corrected by the stereotactic localization
openings need to be extended, and how to do that in a way that does not compromise scalp healing.
Surgical Nuances Metastasis Although radiosurgery has supplemented open craniotomy in the treatment of many smaller brain metastases, larger tumors or those that have failed radiosurgery remain candidates for surgical resection. As such tumors have sharp margins both anatomically and radiographically, they are amenable to resection through small craniotomies with the expectation of a complete removal, an outcome that is enhanced by en bloc resection. In the only report focusing on image guidance for brain metastasis, degree of resection and rates of complication were similar to those seen in series done without image guidance
Image guided craniotomy for brain tumor
45
. Figure 45-8 Transsphenoidal surgical targeting and trajectory planning. These images were acquired in the Viewscope, an earlier version of the StealthStation that married frameless surgical navigation to the operating microscope. The point of focus of the microscope was the point of a virtual localizing probe, and the trajectory shown represents the projection of the center of the field of view within the microscope
[38,39]. Neuronavigational techniques facilitate the approach to deep lesions and are particularly important when multiple lesions are present as they allow the surgeon to devise the least disruptive and most efficient trajectory to each lesion. When a metastasis is associated with a cyst, the cyst should ideally not be punctured if probe accuracy is to be maintained. In theory, decompressing the cyst facilitates tumor removal, but it also promotes brain shift, thereby making localization of margins less accurate, and it may foster seeding of tumor cells leading to later recurrence. Certain tumor types are notoriously friable due to necrosis (adenocarcinoma of lung) or intratumoral hemorrhage (melanoma) and en bloc resection including a small margin of adjacent normal brain should be pursued if the tumor’s location does not prohibit this. Image-guided placement of Ommaya reservoirs for leptomeningeal metastasis has been reported, but this is a low-risk procedure in which traditional free-hand ventricular cannulation is quite successful. Thus, the benefit of neuronavigation in such patients is evident mainly in those with slit ventricles [40].
Meningioma The main utility of neuronavigation in meningioma resection comes in planning a minimally invasive incision and in tailoring the craniotomy opening. In creating that opening, care must be taken to incorporate the full extent of the dural tail on all imaging planes, as a margin of normal dura must be achieved circumferentially around the entire tumor for an ideal Simpson grade I resection. In operations done to remove convexity meningiomas and in particular parasagittal tumors, superimposition of the venous anatomy may prove very helpful. Parasagittal meningiomas adjacent to the middle third of the superior sagittal sinus have a tendency to nestle against one or more important veins which drain into the sinus and demand preservation. Neuronavigation systems can assist in mapping out those veins ahead of time. Meningiomas are usually isolated from the brain with cottonoids and debulked internally, collapsing the tumor upon itself and therefore allowing some brain shift to occur. However, the margin is usually easy enough to detect by tactile and visual clues,
709
710
45
Image guided craniotomy for brain tumor
thus brain shift is not as important a factor in meningioma resection as it is in intraaxial tumor removal. Paleologos et al. have reported a series of 100 patients treated with image-guided surgery for meningioma [41]. These patients were compared to 170 patients treated without neuronavigation. They found shorter surgical times for the image-guided group and a shorter hospital stay likely due to the significantly lower complication rate observed when image guidance was applied. As a result of these differences, the mean cost per patient was 20% higher without image guidance, in spite of the extra costs associated with application of the technology.
Glioma The application of neuronavigation to glioma surgery is largely directed at bringing the degree of resection of the radiographic lesion as close to 100% as possible. This effort makes sense if, and only if, such incremental improvements in resection make a difference in patient outcome. This has been a matter of some controversy over the years. However, a recent detailed meta-analysis by Sanai and Berger supports the notion that improving resection improves survival [42]. This has been supported for glioblastoma by several other rigorous studies using MRI assessment of volumetric resection in all patients (as opposed to more subjective measures of resection), studies which showed the strongest benefit comes from resection exceeding 98% of the preoperative volume of contrast-enhancing tumor [43,44]. The survival advantage for low grade gliomas was even more striking with resection based on the tumor boundaries delineated on T2-weighted (not post-contrast) images, a necessary difference given that most low grade gliomas do not enhance [42]. The most difficult area of the brain for maintaining accuracy of glioma resection under neuronavigation is the occipital region. Whether the patient is placed prone or in the lateral
position, the occipital portion of the brain tends to bulge significantly once the dura is opened. Thus, a relatively immediate brain shift occurs and it is difficult to prevent or avoid that shift. As a general rule, tumors in the posterior third of the brain pose greater difficulties for accurate localization than those in the anterior third. Another compounding factor leading to lower accuracy in the posterior cerebrum (and indeed in the posterior fossa) stems from the need to concentrate fiducial markers anteriorly to take advantage of the richer variety of surfaces there. Thus, registration accuracy declines when the tumor is located farther away from that anterior cluster. Smaller tumors can be resected with better accuracy using neuronavigation, and in deep tumors error accumulates to a greater degree than in superficial tumors [45]. Retrospective analysis of 76 patients with glioblastomas undergoing resection with or without neuronavigation has been reported by Kurimoto et al. [46]. They found a greater likelihood of gross total removal in the neuronavigation group (64% vs. 38%), which correlated with increases in survival time. However, given the low incidence of aggressive removal overall, it is possible that the advantage of neuronavigation might diminish were greater effort made to achieve complete resection independent of the adjunctive technologies used. Reoperation for gliomas is encountered more frequently now given the multiplicity of approaches to tumor suppression by medical neurooncologists and the resulting longer survival times achieved. Patients with recurrent gliomas are likely to show less brain shift than those who have not previously undergone surgery. The cortical surface usually adheres to the dura around the prior craniotomy site and therefore retracts less with hyperventilation and CSF release. Indeed, surgical navigation is very helpful in such patients because the gliosis and treatment effect from prior radiation and chemotherapy tend to make the tumor’s margins less distinct and more difficult to follow. When radiation necrosis is superimposed on a
Image guided craniotomy for brain tumor
tumor, the typical hyperechogenicity seen by ultrasound converts to a bland and indistinct look that makes complete resection more difficult to achieve. Navigation is therefore of great utility in such cases. Resection of the entire extent of enhancement, including those areas of enhancement that might be interpreted as post-irradiation effect, is important because such regions often consist of a mix of tumor and treatment effect, which provides a seed for further growth if left in place. The greatest advantage of all is seen when image-guided surgery is applied to low-grade gliomas (> Figure 45-9). These tumors generally have fairly distinct margins on T2-weighted images, but a relatively bland appearance during surgery with borders that can be quite difficult to define. A highly skilled and experienced surgeon may achieve good resection of such tumors without neuronavigation, but we have found that even those with a wealth of experience still leave behind small bits of tumor that they catch only when neuronavigation is used. To minimize the effects of brain shift, the interface between tumor and adjacent brain (as defined by the navigation system) is opened by resecting along that edge and placing cottonoids to maintain the distinction. In this fashion, any brain shift that occurs as internal debulking of the tumor proceeds will be minimized. Just as is true for metastases, cysts associated with gliomas should be maintained for as long as possible in the planning and delineating stages, so that tumor outlines can be obtained while registration is still accurate. Those areas in which neuronavigation is absolutely vital in glioma surgery include tumors within the insula, those which cross the midline, those affecting the thalamus, and those brainstem tumors which by virtue of their exophytic quality are amenable at least to partial resection.
45
complications. Shirane et al. have shown that after including neuronavigation in the occipital pranstentorial approach to pineal tumors, their complication rate fell from 27% to zero [47]. Since such tumors often induce hydrocephalus by blockage of the aqueduct of Sylvius, endoscopic third ventriculostomy guided by computerassisted navigation is a useful adjunct to frameless stereotactic biopsy of the mass [48]. In a similar vein, neuronavigational endoscopy (with the endoscope enrolled as the localizing probe) can effectively be applied for septal fenestration to relieve ventricular trapping, for removal of colloid cysts, and for biopsy of tumors within the ventricle or arising from the ventricular wall [49]. Pineal endoscopy has also been proposed based on cadaver study, but it has not yet been reported clinically [50].
Stereotactic Biopsy For many years surgeons have relied on framebased methods of localization to achieve the precision required in stereotactic biopsy of deep tumors. Similar precision is now available from frameless techniques, as shown by Dammers et al. in their analysis of 227 frame-based and 164 frameless biopsies [51]. Both groups were identical in complication rates and diagnostic yield (12 and 89% respectively). This confirms the results of an earlier series whose authors made similar comparisons while achieving a lower rate (6%) of permanent morbidity and concluded that frameless biopsy was equally effective as, but more efficient than, biopsy using a frame [52].
Intraoperative Imaging Pineal Tumors and Endoscopy In the tricky, anatomically rich area of the pineal recess, image guidance can help avoid
Neuronavigation relying on images obtained during surgery might be deemed ‘‘imaginginteractive’’ as opposed to the ‘‘image-guided’’ techniques thus far described. Intraoperative
711
712
45
Image guided craniotomy for brain tumor
. Figure 45-9 Tumor resection in the BrainSuite intraoperative MRI. (a) This low grade astrocytoma in the right frontal lobe is well delineated on FLAIR images taken prior to resection (b) After resection of all visible tumor, a new image shows that removal was in fact incomplete at the deepest and most posterior part of the resulting cavity (c) Further resection eliminates the residual abnormality. Without intraoperative MRI, this patient would have had a less complete resection than was obtained using that technology
scans offer clear advantages in that they can give up-to-the-minute preoperative information if done after induction of anesthesia but before the incision has been made. The stages of the resection can be imaged in real time, extent of resection can be quantified, residual tumor identified, and reimaging to update registration
can minimize the issue of brain shift. The initial foray into intraoperative scanning was made in Pittsburgh in the early 1980s. A fixed CT scanner was placed in an operating room used for stereotactic biopsy and for craniotomies, and was used to confirm target localization for biopsies, to rule out post-resection hemorrhage, and to show
Image guided craniotomy for brain tumor
the extent of resection [53]. In the years since, mobile CT scanners have been developed for intraoperative application and produce images of sufficient quality to be usable for determining the limits of resection (> Figure 45-10). Series have been reported by two groups in Germany who were favorably impressed with the utility of the CT scan in the operative environment [54,55]. The most recent descriptions of intraoperative CT have come in pituitary surgery, specifically the use of the Arcadis Orbic which allows both conventional fluoroscopic views and multiplanar reconstructions to be acquired during surgery [56]. Fluoroscopic images match or exceed the quality of those images from standard C-arms and multiplanar reconstructions gave images equal in quality to those provided by preoperative stealth CT. These authors felt that this system provided more reliability than did registration of CT images acquired prior to surgery. The advantages of CT are its speed and lower cost. However, it sacrifices anatomic detail and defines most intracranial tumors less precisely than does MRI. Gliomas can be mapped
45
more precisely on MRI and areas of subtle involvement can be shown that CT scans overlook. In addition, extent of resection is more complete with MRI due to its better appreciation of subtle areas of residual tumor along the resection margin. For these reasons, tumor surgeons have embraced intraoperative MRI as a superior technology, particularly when resecting gliomas. The disadvantages of intraoperative MRI scanners are their lack of widespread availability and high cost. Placing an intraoperative MRI into use demands a major commitment of resources by the hospital installing it, and often pre-existing space must retrofitted with magnetic shielding and reinforced to account for the weight of the scanner. Ferromagnetic instruments and equipment cannot be used within the magnetic field, and therefore certain instruments (like the specula used in transsphenoidal surgery) must be replaced with expensive MRIcompatible equipment. For these reasons, it is likely that intraoperative CT will retain a role for many years to come, particularly in cases of contrast-enhancing gliomas or complex metastases. The ideal cases for the intraoperative MRI
. Figure 45-10 CereTom mobile CT unit which can provide intraoperative images. It is battery-powered, has a 25-cm field of view, and generates up to eight slices per revolution using a modular multi-row detector
713
714
45
Image guided craniotomy for brain tumor
are non-enhancing cerebral tumors, typically of low grade, in which a surgeon may have difficulty identifying margins precisely, and complex pituitary macroadenomas, in which lateral or superior extension of tumor may persist despite the best efforts of a skilful surgeon. Neither of those tumor types will show well on CT. Intraoperative MRI dates to the mid 1990s and was first applied to intracranial tumor surgery by Peter Black at the Brigham & Women’s Hospital [57,58]. Black’s group used a fixed machine of 0.5 T strength into which patients were placed for surgery performed within the center of the magnetic field (> Figure 45-11). Thus all instruments had to be MRI-compatible and patients did not move in and out of the machine for scanning. In the first 200 patients from this series, hyperacute hemorrhage was noted in two patients from whom the clot was then removed. Other groups reporting the use of 0.5 T MRI during surgery included Zimmerman et al. [59]. The use of lower field strengths down to 0.12 T (as in
the vertical gap systems produced by Fonar or Hitachi) allows use of standard surgical instruments within the magnetic field, but yields reduced temporal resolution and spatial resolution per unit time, as well as a higher signal-to-noise to ratio even in newer generation models [60]. For this reason, high field (1.5 T) systems have been utilized to improve image quality, which is particularly important in glioma surgery. All intraoperative MRI systems provide basic imaging capacity for T1- and T2-weighted views, but high field systems can also acquire other MRI subsets including angiography, functional MR, diffusion-weighted imaging, chemical shift imaging, and MR spectroscopy. Direct comparison by Bergsneider et al. of results obtained from surgery using 0.2 T versus 1.5 T magnets showed that a greater mean extent of resection was achieved for gliomas using intraoperative MRI versus standard imaged-guided frameless neuronavigation (91% vs. 79% respectively). Interestingly enough, there was little difference in the degree
. Figure 45-11 Intraoperative MRI at the Brigham and Women’s Hospital. In this Signa SP ‘‘double doughnut’’ unit (GE Medical Systems) with a field strength of 0.5 T, the patient remains inside the core magnetic field during surgery
Image guided craniotomy for brain tumor
of resection achieved using the 0.2 T versus the 1.5 T intraoperative MRI without updated neuronavigation; with updated registration, the mean percent resection increased from 92 to 98% [61]. Another study using the 0.5 T magnet showed no difference between two matched sets of 32 patients with high-grade gliomas, one of which underwent operation with neuronavigation and one without [62]. Accessible high-grade gliomas can be resected without neuronavigation in many patients to a near-complete level by surgeons experienced in such procedures. However, lower grade tumors pose particular challenges even to those well versed in their nuances, and we routinely perform such operations within the BrainSuite environment, an integrated operating room with a fixed magnet and rotating operating room table that
45
swings out of the MRI machine and allows performance of the procedure at a distance far enough from the magnet to allow the use of normal surgical instruments (> Figure 45-12). A detachable head coil permits the surgeon access to the patient’s head when the patient is outside the scanner and ready for the procedure to continue (> Figure 45-13). This particular system does provide updating of registration with each new scan performed and we find that feature to be particularly valuable for increasing the accuracy of localization. Nimsky et al. have published a series of 182 procedures performed using a 1.5 T magnet, and derived results favoring the use of such equipment in glioma surgery [63]. They found that the intraoperative MRI influenced the procedure in 36% of patients, in whom surgery would have
. Figure 45-12 The BrainSuite intraoperative MRI. This system places the patient on a rotationally mobile operating table which can move into the MRI for scanning, then out for the operation. This allows the use of ferromagnetic instruments during surgery, but materials that enter the magnet (e.g., intravenous catheters, endotracheal tubes, or patient implants) must still be MRI-compatible. The system includes an operating microscope and frameless neuronavigation with automatic re-registration, along with wall display of intraoperative images. Differently colored zones on the floor indicate sectors of higher or lower magnetic field strength
715
716
45
Image guided craniotomy for brain tumor
otherwise stopped but in whom instead it continued for removal of residual tumor [63]. The percent of final tumor volume was significantly reduced by intraoperative scanning in both low-grade and high-grade tumors with complete resection achieved in 57% of the lowgrade cases and in 27% of the high-grade cases. This series may be criticized for not achieving a higher degree of resection in both groups, and one might argue that more aggressive resection might have resulted in less difference between first and subsequent scans without compromising safety. Functional MRI can also be performed intraoperatively in high field machines although no one to date has performed a rigorous comparison between the results of functional MRI and those of direct cortical stimulation (for speech and motor mapping) or somatosensory evoked potentials (for identification of the central sulcus) [32,64]. It might be argued that functional MRI scans acquired prior to surgery and combined with intraoperative images by an overlay technique provide greater reliability due to their
acquisition in a more controlled environment (> Figure 45-14). This question also has yet to be studied. Precise alignment of previously obtained images with those acquired during surgery may require correction of geometric distortion inherent in intraoperative MRI [65]. The impact of neuronavigation incorporating functional MRI increases when direct subcortical stimulation is added to it, with positive stimulation indicating proximity to the corticospinal tracts within 10 mm [34]. Additional methods of identifying motor functioning include direct display by diffusion tensor imaging of motor fiber tracts overlaid on standard anatomical images (> Figure 45-15) [67]. Such imaging can be used to delineate glioma margins more crisply [68]. Tractography in intraoperative MRI has been used by Mikuni et al. to define the accuracy of motor evoked potentials [69]. These authors found that such potentials consistently provoked motor activity at distances <7 mm from the center of the tract and consistently failed to do so at distances >13 mm. When stimulation was applied in the region of the corona radiata (as
. Figure 45-13 The head-holder for the BrainSuite intraoperative MRI. The magnetic coil sits above the patient’s head during scan acquisition, and is removed to allow access to the surgical site during the operative phase
Image guided craniotomy for brain tumor
45
. Figure 45-14 Functional MRI (fMRI) showing activation of speech centers with a receptive task. This patient’s tumor was an anaplastic astrocytoma with low grade surround. Wernicke’s area had been diffused and anteriorly displaced in relation to the tumor, and was immediately adjacent to the tumor margin. This dataset was imported into the intraoperative MRI (BrainSuite) and successfully applied to neuronavigation to preserve speech function during tumor removal
717
718
45
Image guided craniotomy for brain tumor
. Figure 45-15 Diffusion tensor imaging in the intraoperative MRI. The white matter tracts in the vicinity of a cavernous hemangioma are shown in right frontal oblique (left) and left dorsal oblique (center) views, with infiltration of the left medial lemniscus shown by a segmented view showing only the tracts and the lesion (right) [66]
opposed to the corticospinal tract itself), movements were elicited at distances of 8–12 mm. Use of intraoperative MRI in this fashion can validate older techniques and provide valuable input that limits or extends resection [70]. Tractography has also been applied to patients with brainstem lesions around which the patterns of tract alteration include deviation, deformation, infiltration, and apparent tract interruption. That information can alter the ratio of benefit to risk to a favorable enough degree to make operation on a brainstem tumor actually possible with reasonable safety (> Figure 45-15) [66]. The brainstem is less subject to brain shift than are the contents of the supratentorial compartment, but the margin for error during surgery is almost nonexistent due to the anatomical density of that region. Both diffusion tensor fiber tracking and MR spectroscopy appear to be valid even in the immediate vicinity of a glioma margin. Given that such tumors are inherently invasive, it is possible that changes in neuronal connectivity at the interface between brain and tumor might impair these radiographic techniques. However, Stadlbauer et al. have found that although the number of neuronal fibers per voxel and the fractional anisotropy are both lower and that mean diffusivity is increased, these changes do
not prevent creation of useful images [33,35]. These alterations were magnified in patients with sensorimotor deficits but even in that group, they still did not prevent image creation. These authors found that overlay of proton MR spectroscopy helps in deciding whether diffusivity changes in nearby fiber tracts are due to tumor infiltration or peritumoral edema, a difference relevant for surgical decision-making. The availability over the past several years of even higher field magnets for MRI has led some groups to apply these during surgery in an effort to enhance further the image quality obtained from the intraoperative MRI. Hall et al. have reported a similar surgical environment in the 3 T magnet to that seen at 1.5 T [71]. Such minimally ferromagnetic items as surgical needles, staples, and disposable scalpels were safely controlled by the surgeon in their version of the intraoperative MRI, which involves performance of the entire procedure within the magnet. However, image quality is not sufficiently different to allow the conclusion that increasing magnet strength beyond 1.5 T further improves the extent of tumor resection, or that it reduces complications from impingement on functional fiber tracts. Intraoperative high field MRI may provide particular benefit in transsphenoidal surgery and
Image guided craniotomy for brain tumor
by extension, in craniotomy for pituitary tumor (> Figure 45-16). Although it has little role in the resection of microadenomas, many of which can be relatively occult on imaging, its ability to enhance resection of macroadenomas is relatively strong. Nimsky et al. have reported the results of intraoperative MRI in 106 patients with clinically nonfunctional pituitary macroadenomas, in whom resection continued if interim intraoperative imaging showed an accessible tumor remnant [72]. Of the 106 patients, 85 were operated with intent to achieve complete removal of tumor. In those 85, imaging showed a tumor remnant in 36 (42%), 29 (34%) of whom went on to further resection, which was then achieved completely in 21. Thus, the rate of complete tumor removal increased from 58 to 82% because of the use of intraoperative MRI. Even in the group with an intended partial removal, resection could be extended in 38% when intraoperative imaging showed accessible fragments not hitherto suspected. When combined with endoscopy, intraoperative MRI can enhance the degree of resection achieved in these often anatomically complex lesions. The difficulty with
45
such studies, whether for pituitary tumors, gliomas, or any other tumor subset, is that surgical experience may allow some surgeons to achieve similar results without intraoperative MRI; thus the degree of success of the technology may reflect the lack of experience of its user. For pituitary tumors as for gliomas and other tumors, intraoperative MRI will likely be most effectively applied to the more difficult cases, in which its benefit will be most profoundly felt. These would include patients undergoing reoperation, those whose tumors who have indistinct boundaries or which blend with adjacent structures, and those with extensions carrying them close to critical anatomical structures the integrity of which must be maintained. The great advantage of the intraoperative MRI is its ability to provide reimaging and reregistration, and thereby it reduces directly the primary impediment to accurate neuronavigation, namely brain shift. The BrainSuite system in use at our institution contains an automatic registration function that requires no external fiducial markers and no touching of a probe to the skin surface. As this feature eliminates several
. Figure 45-16 Intraoperative MRI done in the BrainSuite on the same patient with pituitary tumor shown in Figure 2. (a) Preoperative images (coronal, post-contrast) show the tumor invading the cavernous sinus, filling the sella, and compressing the optic apparatus (b) Post-resection images show nice clearance of the tumor with the exception of a small remnant adherent to the lateral wall of the right cavernous sinus, which was subsequently partially resected to preserve the function of cranial nerves to which it was adherent
719
720
45
Image guided craniotomy for brain tumor
of the potential sources of inaccuracy and cuts down on the time the surgeon spends on the registration process, automatic re-registration has been one of the particularly attractive features of this otherwise somewhat cumbersome system. Practical issues in the intraoperative MRI environment include not only the need to use non-ferromagnetic endotracheal tube components and intravenous catheter hubs, but also the absence of a simple system for maintenance of the sterile field in systems such as ours in which surgery proceeds outside the magnet with intermittent forays into the heart of the magnetic field for scanning. The drapes must be tucked up carefully around the patient before the couch is shifted into the MRI; and then recovered with an additional sterile layer when surgery recommences after each scan. Manipulation of the drapes adds to the risk of infection, and indeed we have found that our infection rate is somewhat higher in intraoperative MRI cases than in those performed in a standard operating room with simple neuronavigation. This excess infection risk has not been a factor, however, in the systems in which the operation proceeds within the magnet and in which no such draping changes are necessary.
Caveats The full utility of neuronavigation has not yet been realized, and its value has been difficult to assess objectively. Although the aficionados of technology in surgery suggest that frameless stereotactic systems are mandatory in craniotomy for tumor, it is better to conclude that they can be immensely helpful in difficult or sensitive areas of the brain, but that careful selection will maximize benefit. Bearing in mind the tendency of human beings to fall in love with their toys (and the unfortunate truth that to the man with a bat, all round objects may resemble a ball), we should take the more nuanced view that resection of
some tumors will be greatly enhanced by neuronavigation, whereas others may be well removed without image guidance. The degree to which neuronavigation is applied in any given surgeon’s practice will depend on many factors, including the nature of the cases operated, the experience of that surgeon with more traditional methods of resection, and the availability of equipment of proven accuracy. Such technology will certainly augment the reach of a skilled surgeon, but cannot substitute for the fundamentals of surgical skill and good intraoperative judgment. Efforts at proving the value of neuronavigation have been confounded by selection bias: in the Glioma Outcomes Project, it was used more for younger patients and for those with smaller low grade tumors, and multivariate analysis to eliminate those variables showed no survival advantage [73]. Willems et al. randomized 45 patients with solitary contrast-enhancing tumors to surgery with or without neuronavigation [74]. They found no difference between the two groups in either survival, extent of resection, or induction of postoperative neurological deficit, and concluded that there was no rationale for routine use of the technology. However, this study’s numbers were small, and it is possible (although certainly not proven) that greater statistical power achieved by a larger sample size would allow detection of a difference that a smaller sample size cannot show. The ongoing evolution of the available navigation systems has advanced this field tremendously, but rigorous analysis of outcomes is difficult, as no one system is accepted as ‘‘best’’ and large series of patients cannot be accrued quickly enough to prevent one or more changes in methodology during the accrual.
Future Directions Practical advances in neuronavigation will likely come from three areas. The first is imaging, which is undergoing a revolution of its own
Image guided craniotomy for brain tumor
with the advent of methods for creating visual maps of metabolic and even gene activity. Current systems already have multimodality capability including fMRI, tractography, and MR spectroscopy in addition to standard images. As newer imaging based on molecular events becomes more robust and allows detection of tumor at earlier stages than now possible, or even permits detection of pretumoral tissue or of stem cells, such images will certainly be converted to use for intraoperative image guidance. The degree of advantage conferred by surgical navigation depends on the quality of the images that drive it. The second area is virtual reality. Gildenberg and Labuz have used and described a system that superimposes a computer-generated rendering of the target volume on a real-time video image, updated during surgery to show only the part of the tumor actually being resected [75]. Kockro et al. described preoperative planning in a stereoscopic virtual reality environment for tumors and arteriovenous malformations in areas with difficult access, and found that it allowed efficient assembly of surgically relevant spatial information from multiple modalities, and predictably that its utility depends on accurate co-registration of the imaging datasets and on its rapidity of real-time interaction [76]. Such systems have been described for training residents in ventriculostomy and for planning aneurysm clipping, but they have not yet been much applied to tumor work [77–79]. The third predictable area of future activity is robotics. Already in use in laparoscopic procedures, its applications in neurosurgery have barely been touched but show clear logic for a marriage between robotic and neuronavigational technology. Navigated robotic endoscopic ventriculostomy has been reported [80]. In addition, instrument tracking during surgery has established criteria for useable neurosurgical robotic systems including translational speeds of up to 12.7 cm/s and rotational speeds of up to 40 /s [81]. Telemanipulation within the context of frameless
45
stereotactic localization will undoubtedly be applied to tumor resection as robotic technology develops further, and it represents the next frontier for stereotactic surgery.
References 1. Horsley V, Clarke RH. ‘‘The structure and function of the cerebellum examined by a new method.’’ Brain 1906;31:45-124. 2. Leksell L. ‘‘A stereotactic apparatus for intracerebral surgery.’’ Acta Chir Scan 1949;99:229-33. 3. Kelly PJ, Kall BA, et al. ‘‘Computer-assisted stereotaxic laser resection of intra-axial brain neoplasms.’’ J Neurosurg 1986;64(3):427-39. 4. Morita A, Kelly PJ. ‘‘Resection of intraventricular tumors via a computer-assisted volumetric stereotactic approach.’’ Neurosurgery 1993;32:920-6. 5. Moshel YA, Link MJ, et al. ‘‘Stereotactic volumetric resection of thalamic pilocytic astrocytomas.’’ Neurosurgery 2007;61:66-75. 6. Rezai AR. ‘‘Integration of functional brain mapping in image-guided neurosurgery.’’ Acta Neurochir Suppl 1997;68:85-9. 7. Roberts DW, Strohbehn JW, et al. ‘‘A frameless stereotaxic integration of computerized tomographic imaging and the operating microscope.’’ J Neurosurg 1986;65:545-9. 8. Roberts DW, Hartov A, et al. ‘‘Intraoperative brain shift and deformation: a quantitative analysis of cortical displacement in 28 cases.’’ Neurosurgery 1998;43:749-58. 9. Watanabe E, Watanabe T, et al. ‘‘Three-dimensional digitizer (neuronavigator): new equipment for computed tomography-guided stereotaxic surgery.’’ Surg Neurol 1987;27:543-7. 10. Watanabe E, Mayanagi Y, et al. ‘‘Open surgery assisted by the neuronavigator, a stereotactic, articulated, sensitive arm.’’ Neurosurgery 1991;28:792-9. 11. Golfinos JG, Fitzpatrick BC, et al. ‘‘Clinical use of a frameless stereotactic arm: results of 325 cases.’’ J Neurosurg 1995;83:197-205. 12. Kleinpeter G, Lothaller C. ‘‘Frameless neuronavigation using the ISG-system in practice: from craniotomy to delineation of lesion.’’ Minim Invasive Neurosurg 2003;46:257-64. 13. Roessler K, Ungersboeck K, et al. ‘‘Frameless stereotactic lesion contour-guided surgery using a computer-navigated microscope.’’ Surg Neurol 1998;49:282-8. 14. Kato A, Yoshimoto T, et al. ‘‘A frameless, armless navigational system for computer-assisted neurosurgery. Technical note.’’ J Neurosurg 1991;74(5):845-9. 15. Heilbrun MP, McDonald P, et al. ‘‘Stereotactic localization and guidance using a machine vision technique.’’ Stereotact Funct Neurosurg 1992;58:94-8.
721
722
45
Image guided craniotomy for brain tumor
16. Heilbrun MP, Koehler S, et al. ‘‘Preliminary experience using an optimized three-point transformation algorithm for spatial registration of coordinate systems: a method of noninvasive localization using framebased stereotactic guidance systems.’’ J Neurosurg 1994; 81(5):676-82. 17. Roberts DW, Strohbehn JW, et al. ‘‘The stereotactic operating microscope: accuracy refinement and clinical experience.’’ Acta Neurochir Suppl 1989;46:112-4. 18. Chen JCT, Moffit K, et al. ‘‘Head-mounted display system for microneurosurgery.’’ Stereotact Funct Neurosurg 1997;68:25-32. 19. Ray WZ, Barua M, et al. ‘‘Anatomic visualization with ultrasound-assisted intracranial image guidance in neurosurgery: a report of 30 patients.’’ J Am Coll Surg 2004;199(2):338-43. 20. Wu Z, Hartov A, et al. ‘‘Multimodal image re-registration via mutual information to account for initial tissue motion during image-guided neurosurgery.’’ Conf Proc IEEE Eng Med Biol Soc 2004;3:1675-8. 21. Sun H, Roberts DW, et al. ‘‘Cortical surface tracking using a stereotactic operating microscope.’’ Neurosurgery 2005;56:86-97. 22. Miga MI, Paulsen KD, et al. ‘‘Model-updated image guidance: initial clinical experiences with gravity-induced brain deformation.’’ IEEE Trans Med Imaging 1999;18:866-74. 23. Comeau RM, Sadikot AF, et al. ‘‘Intraoperative ultrasound for guidance and tissue shift correction in imageguided neurosurgery.’’ Med Phys 2000;27:787-800. 24. Roberts DW, Lunn K, et al. ‘‘Intra-operative image updating.’’ Stereotact Funct Neurosurg 2001;76:148-50. 25. Roth J, Biyani N, et al. ‘‘Real-time neuronavigation with high-quality 3D ultrasound sonowand in pediatric neurosurgery.’’ Pediatr Neurosurg 2007;43:185-91. 26. Nimsky C, Ganslandt O, et al. ‘‘Quantification of, visualization of, and compensation for brain shift using intraoperative magnetic resonance imaging.’’ Neurosurgery 2000;47:1070-9. 27. Reinges MH, Nguyen HH, et al. ‘‘Course of brain shift during microsurgical resection of supratentorial cerebral lesions: limits of conventional neuronavigation.’’ Acta Neurochir 2004;146:369-77. 28. Wittek A, Miller K, et al. ‘‘Patient-specific model of brain deformation: application to medical image registration.’’ J Biomech 2007;40:919-29. 29. Suess O, Picht T, et al. ‘‘Neuronavigation without rigid pin fixation of the head in left frontotemporal tumor surgery with intraoperative speech mapping.’’ Neurosurgery 2007;60:330-8. 30. Mangano FT, Limbrick DDJ, et al. ‘‘Simultaneous imageguided and endoscopic navigation without rigid cranial fixation: application in infants: technical case report.’’ Neurosurgery 2006;58 4 Suppl 2:ONS-E377. 31. Rygh OM, Nagelhus Hernes TA, et al. ‘‘Intraoperative navigated 3-dimensional ultrasound angiography in tumor surgery.’’ Surg Neurol 2006;66:581-92.
32. Liu H, Hall WA, et al. ‘‘The roles of functional MRI in MR-guided neurosurgery in a combined 1.5 Tesla MRoperating room.’’ Acta Neurochir Suppl 2003;85:127-35. 33. Stadlbauer A, Moser E, et al. ‘‘Integration of biochemical images of a tumor into frameless stereotaxy achieved using a magnetic resonance imaging/magnetic resonance spectroscopy hybrid data set.’’ J Neurosurg 2004;101: 287-94. 34. Mikuni N, Okada T, et al. ‘‘Clinical impact of integrated functional neuronavigation and subcortical electrical stimulation to preserve motor function during resection of brain tumors.’’ J Neurosurg 2007;106(4):593-8. 35. Stadlbauer A, Nimsky C, et al. ‘‘Changes in fiber integrity, diffusivity, and metabolism of the pyramidal tract adjacent to gliomas: a quantitative diffusion tensor fiber tracking and MR spectroscopic imaging study.’’ AJNR Am J Neuroradiol 2007;28:462-9. 36. Pirotte B, Goldman S, et al. ‘‘Integrated positron emission tomography and magnetic resonance imaging-guided resection of brain tumors: a report of 103 consecutive procedures.’’ J Neurosurg 2006;104:238-53. 37. Helm PA, Eckel TS. ‘‘Accuracy of registration methods in frameless stereotaxis.’’ Comput Aided Surg 1998;3 (2):51-6. 38. Sawaya R, Hammoud M, et al. ‘‘Neurosurgical outcomes in a modern series of 400 craniotomies for treatment of parenchymal tumors.’’ Neurosurgery 1998; 42(5):1044-55. 39. Tan TC, McL Black P. ‘‘Image-guided craniotomy for cerebral metastases: techniques and outcomes.’’ Neurosurgery 2003;53(1):82-9. 40. Takahashi M, Yamada R, et al. ‘‘Navigation-guided ommaya reservoir placement: implications for the treatment of leptomeningeal metastases.’’ Minim Invasive Neurosurg 2007;50(6):340-5. 41. Paleologos TS, Wadley JP, et al. ‘‘Clinical utility and costeffectiveness of interactive image-guided craniotomy: clinical comparison between conventional and image-guided meningioma surgery.’’ Neurosurgery 2000;47:40-7. 42. Sanai N, Berger MS. ‘‘Glioma extent of resection and its impact on patient outcome.’’ Neurosurgery 2008;62:753-64. 43. Lacroix M, Abi-Said D, et al. ‘‘A multivariate analysis of 416 patients with glioblastoma multiforme: prognosis, extent of resection, and survival.’’ J Neurosurg 2001;95:190-8. 44. Stummer W, Reulen HJ, et al. ‘‘Extent of resection and survival in glioblastoma multiforme: identification of and adjustment for bias.’’ Neurosurgery 2008;62:564-76. 45. Benveniste R, Germano IM. ‘‘Evaluation of factors predicting accurate resection of high-grade gliomas by using frameless image-guided stereotactic guidance’’. Neurosurg Focus 2003;14(2):E5. 46. Kurimoto M, Hayashi N, et al. ‘‘Impact of neuronavigation and image-guided extensive resection for adult patients with supratentorial malignant astrocytomas: a
Image guided craniotomy for brain tumor
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
single-institution retrospective study.’’ Minim Invasive Neurosurg 2004;47:278-83. Shirane R, Shamoto H, et al. ‘‘Surgical treatment of pineal region tumours through the occipital transtentorial approach: evaluation of the effectiveness of intra-operative micro-endoscopy combined with neuronavigation.’’ Acta Neurochir 1999;141:801-8. Kim IY, Jung S, et al. ‘‘Neuronavigation-guided endoscopic surgery for pineal tumors with hydrocephalus.’’ Minim Invasive Neurosurg 2004;47(6):365-8. Schroeder HW, Wagner W, et al. ‘‘Frameless neuronavigation in intracranial endoscopic neurosurgery.’’ J Neurosurg 2001;94(1):72-9. Youssef AS, Keller JT, et al. ‘‘Novel application of computer-assisted cisternal endoscopy for the biopsy of pineal region tumors: cadaveric study.’’ Acta Neurochir 2007;149(4):399-406. Dammers R, Haitsma IK, et al. ‘‘Safety and efficacy of frameless and frame-based intracranial biopsy techniques.’’ Acta Neurochir 2008;150(1):23-9. Woodworth GF, McGirt MJ, et al. ‘‘Frameless imageguided stereotactic brain biopsy procedure: diagnostic yield, surgical morbidity, and comparison with the frame-based technique.’’ J Neurosurg 2006;104 (2):233-7. Engle DJ, Lunsford LD. ‘‘Brain tumor resection guided by intraoperative computed tomography.’’ J Neuro-Oncol 1987;4:361-70. Grunert P, Muller-Forell W, et al. ‘‘Basic principles and clinical applications of neuronavigation and intraoperative computed tomography.’’ Comput Aided Surg 1998;3:166-73. Gumprecht H, Lumenta CB. ‘‘Intraoperative imaging using a mobile computed tomography scanner.’’ Minim Invasive Neurosurg 2003;46:317-22. Fox WC, Wawrzyniak S, et al. ‘‘Intraoperative acquisition of three-dimensional imaging for frameless stereotactic guidance during transsphenoidal pituitary surgery using the Arcadis Orbic System.’’ J Neurosurg 2008;108:746-50. Martin C, Alexander EI, et al. ‘‘Surgical treatment of lowgrade gliomas in the intraoperative magnetic resonance imager.’’ Neurosurg Focus 1998;4(4):E8. Schwartz R, Shamoto H, et al. ‘‘Intraoperative MR imaging guidance for intracranial neurosurgery: experience with the first 200 cases.’’ Radiology 1999; 211(2):477-88. Zimmermann M, Seifert V, et al. ‘‘Open MRI-guided microsurgery of intracranial tumours in or near eloquent brain areas.’’ Acta Neurochir 2001;143:327-37. Schulder M, Salas S, et al. ‘‘Cranial surgery with an expanded compact intraoperative magnetic resonance images.’’ J Neurosurg 2006;104:611-7. Bergsneider M, Sehati N, et al. ‘‘Extent of glioma resection using low-field (0.2 T) versus high-field (1.5 T) intraoperative MRI and image-guided frameless neuronavigation.’’ Clin Neurosurg 2005;52:389-99.
45
62. Hirschberg H, Samset E, et al. ‘‘Impact of intraoperative MRI on the surgical results for high-grade gliomas.’’ Minim Invasive Neurosurg 2005;48:77-84. 63. Nimsky C, Fujita A, et al. ‘‘Volumetric assessment of glioma removal by intraoperative high-field magnetic resonance imaging.’’ Neurosurgery 2004;55:358-70. 64. Nimsky C, Ganslandt O, et al. ‘‘Intraoperative visualization of the pyramidal tract by diffusion-tensor-imagingbased fiber tracking.’’ Neuroimage 2006; 30(4):1219-29. 65. Archip N, Clatz O, et al. ‘‘Compensation of geometric distortion effects on intraoperative magnetic resonance imaging for enhanced visualization in image-guided neurosurgery.’’ Neurosurgery 2008; 62 3 Suppl 1:209-15. 66. Chen X, Weigel D, et al. ‘‘Diffusion tensor imaging and white matter tractography in patients with brainstem lesions.’’ Acta Neurochir 2007;149(11):1117-31. 67. Nimsky C, Ganslandt O, et al. ‘‘Intraoperative high-field MRI: anatomical and functional imaging.’’ Acta Neurochir Suppl 2006;98:87-95. 68. Price SJ, Jena R, et al. ‘‘Improved delineation of glioma margins and regions of infiltration with the use of diffusion tensor imaging: an image-guided biopsy study.’’ AJNR Am J Neuroradiol 2006;27(9):1969-74. 69. Mikuni N, Okada T, et al. ‘‘Clinical significance of preoperative fibre-tracking to preserve the affected pyramidal tracts during resection of brain tumours in patients with preoperative motor weakness.’’ J Neurol Neurosurg Psychiatry 2007;78(7):716-21. 70. Mikuni N, Okada T, et al. ‘‘Comparison between motor evoked potential recording and fiber tracking for estimating pyramidal tracts near brain tumors.’’ J Neurosurg 2007;106(1):128-33. 71. Hall WA, Galicich W, et al. ‘‘3-Tesla intraoperative MR imaging for neurosurgery.’’ J Neuro-Oncol 2006;77: 297-303. 72. Nimsky C, von Keller B, et al. ‘‘Intraoperative high-field magnetic resonance imaging in transsphenoidal surgery of hormonally inactive pituitary macroadenomas.’’ Neurosurgery 2006;59(1):105-14. 73. Litofsky NS, Bauer AM, et al. ‘‘Image-guided resection of high-grade glioma: patient selection factors and outcome.’’ Neurosurg Focus 2006;20(4):E16. 74. Willems PW, Taphoorn MJ, et al. ‘‘Effectiveness of neuronavigation in resecting solitary intracerebral contrast-enhancing tumors: a randomized controlled trial.’’ J Neurosurg 2006;104:360-8. 75. Gildenberg PL, Labuz J. ‘‘Use of a volumetric target for image-guided surgery.’’ Neurosurgery 2006;59:651-9. 76. Kockro RA, Serra L, et al. ‘‘Planning and simulation of neurosurgery in a virtual reality environment.’’ Neurosurgery 2000;46:118-35. 77. Anil SM, Kato Y, et al. ‘‘Virtual 3-dimensional preoperative planning with the dextroscope for excision of a 4th ventricular ependymoma.’’ Minim Invasive Neurosurg 2007;50(2):65-70.
723
724
45
Image guided craniotomy for brain tumor
78. Banerjee PP, Luciano CJ, et al. ‘‘Accuracy of ventriculostomy catheter placement using a head- and hand-tracked high-resolution virtual reality simulator with haptic feedback.’’ J Neurosurg 2007;107(3):515-21. 79. Wong GK, Zhu CX, et al. ‘‘Craniotomy and clipping of intracranial aneurysm in a stereotactic virtual reality environment.’’ Neurosurgery 2007;61(3):564-8.
80. Zimmermann M, Krishnan R, et al. ‘‘Robot-assisted navigated endoscopic ventriculostomy: implementation of a new technology and first clinical results.’’ Acta Neurochir 2004;146(7):697-704. 81. Woerdeman PA, Willems PW, et al. ‘‘The analysis of intraoperative neurosurgical instrument movement using a navigation log-file.’’ Int J Med Robot 2006; 2(2):139-45.
55 Image Guided Management of Cerebral Metastases P. Kongkham . M. Bernstein
Introduction Cerebral metastases represent the most common type of brain tumor seen in adults in clinical practice, with an annual incidence outnumbering primary brain tumors by approximately 10:1 [1]. Over 1 million people per year in the United States are diagnosed with cancer, and of these, up to 170,000 will develop brain metastases [1–3]. In fact, between 10 and 15% of cancer patients are ultimately diagnosed with metastatic brain disease during their lifetime, making cerebral metastasis the most common neurologic complication of systemic malignancy [1,4]. Up to 20–40% of patients with metastatic disease will have evidence of cerebral involvement at autopsy [1]. The most common sources of metastases to the brain include tumors from lung (40–60% of metastases), followed by breast, melanoma, and less often colon or kidney primary sites [1,5]. The site of the primary tumor may be unknown in up to 10% of cases [1,5]. The distribution of metastatic disease follows that of the cerebral volume and blood flow, with 80% occurring in the cerebral hemispheres, 15% in the cerebellum, and 5% in the brainstem [1,6]. Up to 75% of patients with cerebral involvement will harbor multiple lesions, based on modern imaging series using contrast-enhanced MRI [7–9]. Specific tumor histologies appear to be more likely to result in multiple metastatic lesions, including melanoma, colon, breast, and lung primaries. In contrast, renal cell carcinoma metastases are more likely to be single lesions. #
Springer-Verlag Berlin/Heidelberg 2009
In recent years the incidence of brain metastasis appears to have been on the rise, attributable at least in part to an aging population, improved therapy for systemic malignancy resulting in longer survival of cancer patients, an increasing incidence of lung cancer and melanoma, and improvements in neuroimaging offering the ability to detect smaller lesions [4,10]. Due to this rising incidence, the management of patients with cerebral metastatic disease represents an increasingly significant clinical as well as economic challenge. Fortunately, along with improvements in our ability to detect cerebral metastases, innovations in neuroimaging and stereotaxy have expanded the therapeutic armamentarium available to the clinician to treat this growing patient population. This chapter will summarize the evidence in the medical literature regarding current treatment options available for patients with cerebral metastatic disease.
Diagnostic Work-up and Patient Selection for Therapy A contrast-enhanced MRI scan of the brain remains the most sensitive diagnostic tool for the detection and follow-up surveillance of cerebral metastatic disease [11]. On MRI, cerebral metastases typically appear as contrast-enhancing lesions located at the grey-white matter junction, with abundant peritumoral edema. In addition to disclosing the number of lesions present, T1-weighted contrast-enhanced MRI of the brain may identify leptomeningeal involvement, as
832
55
Image guided management of cerebral metastases
evidenced by an irregular brightly enhancing pial surface, with or without involvement of the arachnoid, dura, or ependymal surfaces. Furthermore, tumor histology may be suggested by MRI, with metastatic melanoma often appearing bright on non-contrast T1-weighted images due to the presence of blood and melanin within the lesions. Finally, associated peritumoral edema is demonstrated clearly on either T2-weighted or fluid-attenuated inversion recovery (FLAIR) sequences. From a practical standpoint however, many physicians typically perform a CT scan of the brain initially to confirm the diagnosis of suspected intracranial lesions including cerebral metastases. CT studies demonstrate single brain lesions in up to 50% of patients with cerebral metastases, while MRI examination reveal a single metastasis in up to 33%, with the remaining patients harboring multiple lesions [6,7]. Therefore, patients with single lesions identified by CT should have this finding confirmed by contrastenhanced MRI studies prior to initiating focal therapy targeting a single lesion, such as surgical resection or radiosurgery. For patients presenting with a cerebral mass suspected of being a metastasis, without a history of a known systemic primary lesion, it is important to identify the site of origin of the brain lesion. Lung cancer should be suspected as the most likely primary source, due to the high incidence with which it metastasizes to the brain. The diagnostic work-up should include chest X-ray or CTscan. In addition, an abdominopelvic CTscan may identify gastrointestinal or renal primary tumors. Mammography may be performed to rule out breast cancer as the primary lesion, although brain metastases from breast cancer prior to the primary disease declaring itself are rare. Finally, a positive radionuclide bone scan may suggest primary tumors that have a tendency to spread to bone. If an extracranial lesion is identified, it is typically biopsied first to confirm the diagnosis due to lower biopsy-associated risk in comparison to the brain lesion. If no extracranial site is identified, one
should consider either biopsy or open resection of the cerebral lesion for tissue diagnosis in order to direct further management. In patients with a cerebral lesion which appears typical for a cerebral metastasis, without a history of malignancy or identifiable systemic lesion on examination, the brain lesion will ultimately prove to be a metastasis in as little as 15% of cases [12,13]. Furthermore, some argue for the importance of tissue diagnosis for the cerebral lesion itself, as even brain lesions in patients with a known history of systemic malignancy may not be metastatic in up to 11% of cases [14]. Several factors play a significant role in the treatment decision-making process for patients with cerebral metastases. Among these are the number, size, and location of the cerebral lesions, as well as the presence of leptomeningeal disease [15–18]. Additional factors include the primary tumor histology, status of systemic disease activity, and presence or absence of extracranial metastases [19]. The patient’s neurologic status or Karnofsky Performance Status (KPS) and the disease-free interval prior to the diagnosis of cerebral metastasis are also prognostically important [5,20–22]. Among these factors, the most important appear to be the status of the primary cancer, and the patient’s neurologic status, with patients suffering from uncontrolled systemic malignancy and demonstrating significant neurologic dysfunction due to their intracranial involvement carrying a poor prognosis despite neurosurgical intervention [19]. Tumor histology factors significantly in the decision-making process for patients with cerebral metastases. Patients with brain metastases from renal cell carcinoma or malignant melanoma often exhibit poor survival, in comparison to patients with cerebral breast metastases [16,17]. Melanoma, renal cell carcinoma, and non-small cell lung cancer (NSCLC) are traditionally regarded as being chemoresistant, while melanoma, renal cell carcinoma, and sarcoma are considered resistant to standard fractionated
Image guided management of cerebral metastases
radiotherapy. Conversely, small cell lung cancer typically responds dramatically to radiotherapy. Similarly, tumors such as testicular tumors, choriocarcinoma, and secondary lymphoma typically respond to fractionated radiotherapy, focal radiation, or systemic chemotherapy. When the diagnosis of a ‘‘non-surgical’’ tumor type is made, a potentially unnecessary craniotomy may be avoided. In addition, studies have shown renal cell carcinoma and melanoma to be more susceptible to radiosurgery in comparison to fractionated radiotherapy [18]. In order to identify subgroups of patients with cerebral metastases whose overall prognosis warranted more aggressive therapy, the Radiation Therapy Oncology Group (RTOG) devised a recursive partitioning analysis (RPA) method of classifying patients with cerebral metastases into three subgroups according to their KPS score, age, and the status and extent of extracranial disease [23]. RPA class I consists of patients age 65 years or less, with a KPS of 70 or greater, with good control of their systemic disease and absence of any extracranial metastases. Patients in this subgroup have the best prognosis, and are considered optimal candidates for aggressive treatment of their disease. RPA class III includes all patients with a KPS of <70. This subgroup of patients is typically not appropriate for surgical resection or aggressive management of their disease. RPA class II includes all patients not in either class I or III. Careful consideration of the expected survival and risks associated with therapy is needed for this subgroup of patients.
Surgical Resection for the Treatment of Cerebral Metastases Advances in microsurgical technique, and imageguided technology based on stereotactic principles, combined with improved therapy for systemic malignancies, have increased the number
55
of patients considered eligible for surgical resection of cerebral metastases, and have reduced the morbidity and mortality associated with surgical resection to approximately 6–8% and 2–5%, respectively [12,24,25]. Factors favoring surgical resection for brain metastases include patients with RPA class I, tumor size greater than 3 cm in diameter, less than four cerebral lesions, symptomatic lesions (mass effect, hydrocephalus, or uncontrollable edema or seizures), lesions with unknown primary tumor histology, and those located in areas amenable to surgical resection [26,27]. Surgical removal leads to immediate relief from symptomatic mass effect or obstruction of CSF flow, and removal of the focus of peritumoral edema. It may allow for more rapid cessation of corticosteroid or antiepileptic drug use, and thereby reduce complications associated with medication-related side effects.
Surgery in Addition to WBRT for the Treatment of Cerebral Metastases To date, three prospective randomized controlled clinical trials (RCT) have assessed the benefit of microsurgical removal of single brain metastases. Patchell et al. published the first randomized trial examining the role of surgical resection plus whole brain radiotherapy (WBRT) in comparison with a strategy of needle biopsy followed by WBRT [14]. They randomized 25 patients to the surgery plus WBRT arm, and 23 to the needle biopsy plus WBRT arm. The WBRT regimen consisted of providing a total dose of 3,600 cGy in 12 daily fractions of 300 cGy. Results of this trial demonstrated that the addition of surgical resection lead to a reduction in local recurrence. In addition, median overall survival was significantly increased in the surgical group (40 weeks) compared with the needle biopsy group (15 weeks). Finally, surgical resection plus WBRT resulted in the maintenance of functional independence for a significantly
833
834
55
Image guided management of cerebral metastases
longer duration (38 weeks) in comparison to needle biopsy plus WBRT (8 weeks). A second RCT published by Vecht et al. also offered support for the role of surgical resection for the treatment of cerebral metastatic disease [28]. This multi-institutional trial randomized a group of 63 patients to receive either complete surgical resection of their cerebral metastasis followed by WBRT, or WBRT alone. A total dose of 4,000 cGy was given to both groups in twice daily fractions of 200 cGy. The results of this study demonstrated a significant increase in median overall survival in the surgery plus WBRT arm (10 months) compared with WBRT alone (6 months). A non-significant trend towards increased functional independence was also seen. The third RCT looking at the role of surgery in the treatment of cerebral metastases failed to demonstrate a clinical benefit [29]. This study randomized 84 patients to WBRT alone (3,000 cGy) versus surgical resection plus WBRT. In comparison to the previous trials by Patchell et al. and Vecht et al., no difference in median overall survival was seen (6.3 months in the WBRT group versus 5.6 months in the surgery plus WBRT group). Possible explanations for the discrepancy in results between this trial and the former RCTs include the fact that this trial included a significantly greater proportion of patients with poorly controlled extracranial disease and lower performance status. In addition, it has been criticized for having an unequal distribution of primary tumor types between the two study arms, with a greater proportion of radioresistant colon cancer in the surgery arm and radiosensitive breast cancer in the WBRT arm.
WBRT Following Surgical Resection of Cerebral Metastases The benefit of adjuvant WBRT following surgical resection of brain metastases has been shown in both retrospective case series as well as one RCT.
Smalley et al. published results from a retrospective series from the Mayo Clinic in which they studied patients with a single brain metastasis who underwent surgery followed by either observation alone or WBRT [30]. In the observation group, 85% of patients experienced local or distant recurrence, compared with only 21% of the surgery plus WBRT group. Median survival was also longer in the group that received postoperative adjuvant WBRT (21 months versus 11.5 months for surgery plus observation alone). Patchell et al. published results from the only RCT addressing the issue of whether postoperative adjuvant WBRT improves outcomes following initial surgical resection of an intracerebral metastasis [31]. Their study randomized a group of 95 patients with completely resected single brain metastases to either post-operative WBRT (50.4 Gy) or observation alone. They observed a reduction in local or distant intracerebral tumor recurrence following WBRT (18%) compared with observation alone (70%). Recurrence at the site of the original tumor resection was reduced to 10%, versus 46% in the observation alone group. In addition, patients in the WBRT arm were less likely to die from progression of their neurologic disease (14%) compared to those managed with post-operative observation alone (37%). Despite these benefits seen with post-operative WBRT, this study failed to demonstrate any difference in duration of functional independence or overall survival. One criticism of this study is the fact that 61% of the surgery plus post-operative observation group crossed over to receive delayed WBRT at the time of tumor recurrence, effectively making this a trial comparing surgery plus immediate post-operative WBRT versus surgery plus delayed WBRT. The results of this study have been widely interpreted. Some argue that the lack of a difference in overall survival supports the decision to withhold adjuvant WBRT following surgery. The primary endpoint of this study was to detect a reduction in tumor recurrence, and as a result the study
Image guided management of cerebral metastases
was not powered to detect small improvements in overall survival. In contrast, others claim that the results support the use of adjuvant WBRT following surgery due to the reduction in local and distant tumor recurrence associated with its use. More recently, questions regarding the utility of adding an adjuvant radiosurgical boost to the surgical resection cavity, in place of postoperative adjuvant WBRT, have arisen [32]. This may be an attractive therapeutic option in a select subgroup of patients who are neurologically well, and prefer to avoid adjuvant WBRT. This strategy has not yet been proven in any RCTs to date, however. Should this strategy be employed, one must consider following the patient closely with serial imaging surveillance in order to detect subsequent local or distant tumor recurrence.
Surgical Resection in the Treatment of Multiple Metastases Traditionally, surgery has been reserved for patients with single brain metastases. Patients diagnosed with multiple cerebral metastases were typically treated with WBRT alone [33]. In recent years, an increasing number of patients with multiple cerebral lesions are being offered surgery as a treatment option. The role for surgical resection in the patient with multiple cerebral metastases, however, is less clear than that for single or solitary brain metastases. To date, no RCTs have addressed the issue of whether surgery offers clinical benefit for this patient population [19]. Some evidence from retrospective case series exists which addresses this question however. Bindal et al. reported on a series of 56 patients treated surgically for multiple brain metastases and found that survival was similar to matched controls treated for single lesions [34]. The beneficial effect of surgery was only
55
seen if all lesions were resected. An additional finding from this study was the fact that the surgical morbidity associated with resection of multiple lesions is not significantly different from the risk associated with resection of a single metastasis. Similarly, Wronski et al. found no significant difference in overall survival in patients treated surgically for single or multiple brain metastases [15]. Often, the decision whether or not to offer surgical therapy for any or all of the lesions in a patient with multiple cerebral metastases relies on the patient’s unique clinical scenario and the experience of the surgeon. Occasionally, patients harboring multiple lesions may demonstrate one lesion that poses a considerable immediate threat due to its location in a critical anatomic compartment (medial temporal lobe, posterior fossa), due to significant mass effect, or due to uncontrollable seizures or edema with an associated significant reduction in quality of life. Despite the presence of additional metastases, one might elect in these cases to resect the lesion posing the particular threat, while treating the remaining lesions with non-surgical strategies. Surgery may also be indicated in the face of multiple brain lesions when the diagnosis of metastatic disease is in question and a tissue diagnosis is sought. The most important factor to consider when selecting patients with multiple metastases for surgical excision is the extent and activity of their systemic disease. In the studies by both Bindal et al. and Iwadate et al., multivariate analysis found extensive systemic disease to be a critical predictor of poor outcome [34,35]. In addition, one must ensure that all lesions are surgically accessible in order to achieve complete resection of all lesions and thereby derive benefit from surgery. Typically, microsurgical treatment would not be recommended up front for patients with four or more brain metastases, as they generally have a poor prognosis and are not thought to benefit from resection [34,36].
835
836
55
Image guided management of cerebral metastases
The Role of Surgery in the Treatment of Recurrent Disease Surgery remains a treatment option in patients with local or distant intracerebral tumor recurrence following initial surgical or radiosurgical management. It has been shown to improve both survival and quality of life in this subgroup of patients [37,38].
Surgical Adjuncts Based on Image Guidance for the Treatment of Cerebral Metastases The use of numerous modern surgical adjuncts is now commonplace in the neurosurgical operating room. The purpose of these adjuncts is to reduce the surgical morbidity and mortality associated with resection of cerebral lesions including metastases. Several of these tools have resulted from progress made in the fields of neuroimaging and stereotaxy, including frame-based and frameless stereotaxy, functional neuronavigation, and intraoperative imaging modalities including MRI and 3-D ultrasound. These image-guidance modalities, along with additional surgical adjuncts such as cortical mapping and awake craniotomy, have made possible the excision of cerebral metastases previously considered inoperable due to their location in critical, eloquent, or inaccessible regions. Frame-based image-guided stereotactic biopsy has served as a vital tool in the neurosurgical management of cerebral metastases for over the last three decades. The rationale for its use has relied on the need to obtain a histologic tissue diagnosis to help guide further therapy. This technique provided the ability to access almost any point in the intracranial space, including lesions that were small, deeply-located, multifocal, located in or adjacent to eloquent cortex, or in patients unable to tolerate craniotomy. Despite many advances in anatomic, functional and physiologic
imaging modalities, accurate non-invasive diagnosis is not yet feasible based on imaging studies alone, and as such needle biopsy for intracranial lesions remains a vital tool for the practicing neurosurgeon. With respect to the management of cerebral metastatic disease, this has particular relevance for patients in which no systemic primary tumor can be identified, and in whom a tissue diagnosis is needed to guide further therapy. Some also advocate for a tissue diagnosis in cases where the primary systemic malignancy has been in remission for a significant length of time, casting suspicion on the diagnosis of cerebral metastasis. In recent years, CT and MRI-guided frameless stereotaxy has largely supplanted frame-based systems. Approximately 80% of biopsy cases previously performed using frame-based techniques may currently be done with equivalent safety and accuracy using frameless stereotaxy [39]. As a result, frame-based biopsy has become increasingly reserved for cases involving lesions located in perceived high-risk areas, such as the brainstem or pineal region. Frameless image-guidance systems typically use optical, electromagnetic, or ultrasonic sensors to track the position of a surgical tool within the operative space, and to map its position onto image space based on either CT or MRI images obtained preoperatively. Images acquired preoperatively are loaded onto a computer workstation, and subsequently coregistered to the patient’s head either using radiographically identifiable fiducial markers or surface matching and a reference device attached to the surgical head holder. The surgeon may then employ the system to identify tumor and normal anatomy, with a high degree of precision. Frameless systems offer several advantages over frame-based systems. Among these are improved patient comfort and physician comfort due to the absence of a need to apply and operate around a bulky stereotactic head frame. In addition, these systems allow for interactive guidance intraoperatively, making it feasible for the surgeon to explore multiple targets or
Image guided management of cerebral metastases
trajectories during the procedure. Image-guidance allows for more accurately placed, smaller craniotomies, accurate localization of subcortical targets, facilitates identification of the brain-tumor interface at surgery, and may improve clinical outcomes [12,27,40–42]. This technology also has the potential to facilitate reduced post-operative duration of hospital admission and associated cost of treatment [43]. More recently, functional neuroimaging modalities such as functional MRI (fMRI) and magnetoencephalography (MEG) have been incorporated into frameless neuronavigation platforms – a strategy referred to as ‘‘Functional Neuronavigation’’ [44]. This strategy has become routine in surgery for lesions adjacent to eloquent brain such as the primary motor strip or cortical language areas, and has been associated with reduced morbidity associated with the resection of such lesions [45]. The incorporation of novel imaging modalities such as Diffusion Tensor Tractography (DTT) also has the potential to increase the margin of safety associated with surgical neuronavigation [46]. DTT offers the potential to image major subcortical white matter tracts such as the corticospinal tract (CST) or optic radiations, and assess their relationship with respect to adjacent intraparenchymal lesions [47,48]. Case reports exist describing its use in conjunction with intraoperative neuronavigation platforms [49,50]. Incorporation of tractography into intraoperative navigation systems may provide a better alternative for identifying important subcortical tracts, as subcortical white matter stimulation has not proven to be as reliable or safe in comparison to cortical mapping for localization purposes [48,51]. A limitation associated with traditional frameless stereotaxy has been its reliance on images obtained pre-operatively. It is well established that the accuracy of neuronavigation systems decreases during the course of the surgical procedure, due to the phenomenon of ‘‘brain shift’’ [40, 52]. This shift results from the combined effects of
55
gravity, head positioning, brain retraction, cerebrospinal fluid removal, and tissue resection that occur during the surgical procedure. Brain shifts in excess of 5 mm have been reported [53]. Furthermore, in up to 50% of cases, brain shift may occur even prior to the start of the surgical procedure [54]. In addition, intraoperative events such as hemorrhage cannot be detected. To overcome the problem of brain shift, strategies for obtaining updated images intraoperatively have been pursued, including intraoperative MRI and ultrasound. To address these issues, our institution and others have developed intraoperative open MRI systems to provide real-time imaging throughout the course of the surgical procedure [55–60]. Early systems typically employed magnets with low field strengths. The University of Toronto intraoperative MRI (IGMIT, image-guided minimally invasive therapy unit) consisted of a 0.2 T vertical gap MRI system [55]. More recently, others have experimented with systems of higher field strength [59]. The use of intraoperative MRI as opposed the earlier intraoperative CT platforms offered the benefits of superior image quality and soft tissue discrimination, the ability to obtain non-reformatted multiplanar images, and the avoidance of exposure of the patient and surgical team to ionizing radiation [58]. Two strategies for incorporating MRI into the operative procedure have typically been followed. The first involves the incorporation of the magnet directly within the operating theatre. This offered the advantage of not requiring patient transport to and from the magnet when images were required. Disadvantages of this strategy included limited working space for the surgical team secondary to physical constraints of working in or around the magnet, the need for non-ferromagnetic surgical and anesthetic equipment, and the generally restricted clinical use of these MRI units for operative cases [55,56]. The second strategy was that of a ‘‘twin operating theater’’ system, in which the surgical procedure
837
838
55
Image guided management of cerebral metastases
is performed in a standard operating room, with the MRI scanner located in an adjacent room. This strategy required a mechanism for sterile transport of the patient to and from the scanner during the surgical procedure. An advantage, however, is the ability to use the MRI scanner for non-surgical clinical applications, making it potentially more cost-effective. Other advantages of this strategy include the ability to use standard surgical and anesthetic equipment, and the performance of the surgical procedure in a familiar, standard operating room environment [57,58,60]. In general, intraoperative MRI offers the advantage of real-time imaging during the surgical procedure. This facilitates surgical planning, lesion localization, assessment of extent of resection, and detection of unexpected intraoperative complications. Images obtained intraoperatively can be utilized with frameless stereotactic neuronavigation systems, thereby minimizing the problem associated with brain shift through intermittently updating the images used for navigation purposes. This surgical adjunct is not applicable to all patients, however. Contraindications include the presence of a cardiac pacemaker or other metallic implant, severe claustrophobia (for awake surgical procedures), body habitus precluding the patient from fitting into the magnet, and surgical targets difficult to access within the confines of operating within the magnet (e.g., posterior fossa lesions) [55]. In addition, some question the added value of using intraoperative MRI for relatively discrete lesions such as metastatic cerebral tumors, where the gross tumor margins are typically not difficult for the surgeon to discern intraoperatively [55]. Finally, the questions regarding the cost-effectiveness of using this technology, and whether or not its use impacts patient outcomes such as overall survival, surgical morbidity, or quality of life remain to be answered. Intraoperative MRI requires special equipment and modifications to the OR environment, and remains prohibitively expensive for
widespread adoption in neurosurgical services worldwide. As a result, there has been a renewed interest in 3-D ultrasound (3-D U/S) as an alternative strategy for real-time intraoperative imaging in neurosurgery [61]. Similar to intraoperative MRI, intraoperative U/S enables the surgeon to acquire real-time multiplanar images during the course of the surgical procedure. Integration of U/S images acquired intraoperatively with navigation platforms allow for the images to be displayed in a similar fashion as preoperatively acquired MRI. This allows for easier interpretation by the neurosurgeon. In addition, any US plane can be displayed together with similar MRI image planes. Part of the renewed interest in intraoperative U/S may be accounted for by the fact that U/S image quality has improved significantly in recent years, and may even rival that of intraoperative MRI [62,63]. Regarding U/S image quality, higher frequency typically produces better resolution, but this occurs at the expense of reduced tissue penetration [61]. Recent improvements in image quality are in part due to ability to electronically tune the U/S probe for a range of frequencies, allowing optimal resolutions to be obtained at various tissue depths [61]. For most brain operations, a 4–8 MHz frequency gives the optimal image quality for depths of 2.5–6 cm from the probe tip [61]. For more superficially located lesions, a 10 MHz probe would be better, with optimal image quality in the 0.5–4 cm range [61]. A ‘‘phased array’’ probe provides an image which fans out from the probe tip. A linear probe provides an image limited to the width of the probe itself, and therefore is better suited to superficial lesions. New 3-D-U/S datasets are acquired intraoperatively using a pre-calibrated, tracked U/S probe, and a 3-D dataset is constructed from 100–200 2D images collected by moving or tilting the probe over the area of surgical interest [61]. With this method, the probe does not remain in the surgical field except during image acquisition. Two different approaches for using intraoperative 3-D U/S data are typically employed.
Image guided management of cerebral metastases
The first approach involves using intraoperativelyacquired U/S data to ‘‘update’’ preoperativelyacquired MR images to accommodate for brain shift, with the modified MR images being used for navigation purposes [54,64–67]. When this strategy of using intraoperative U/S to ‘‘correct’’ preoperative MR data is employed, navigation accuracy can be improved, however this is technically difficult as automatic MRI to U/S registration is challenging and may introduce another source of error. This approach does carry distinct advantages, however. In particular, it allows for continued navigation based on imaging modalities that currently can only be acquired preoperatively (fMRI, DTT) [68]. The second approach involves constructing a 3-D U/S image volume, and using this image volume directly for navigation purposes [63,69–71]. Using this approach, clinical accuracies of under 2 mm have been reported [72]. Advantages of using intraoperative 3-D U/S include the ability to correct for brain shift during the procedure, the benefit of avoiding stereotactic frame placement, no need for preoperative imaging, and reduced cost in comparison to intraoperative MRI strategies. An additional advantage is the ability of this modality to image vessels based on Doppler signals from the blood stream [61]. Limitations include the inability to obtain images prior to performing the craniotomy, thereby limiting its utility in planning the craniotomy site, and the limited field of view in comparison with MRI.
Stereotactic Radiosurgery for the Treatment of Cerebral Metastases It has been estimated that less than one third of patients with cerebral metastases are surgical candidates, due to the presence of multiple metastases, metastases located in surgically inaccessible regions, or medical comorbidities precluding surgery [73]. For select patients deemed
55
inoperable, stereotactic radiosurgery (SRS) provides a viable alternative to open surgical resection. Improvements in the areas of neuroimaging, stereotaxy, computer planning and robotics have facilitated the widespread adoption of SRS as a treatment modality for intracranial pathology, including cerebral metastases. SRS refers to the process of image-guided delivery of high-dose single-fraction high-energy radiation to a focal target, while minimizing exposure to surrounding normal tissue. SRS does not replace conventional radiotherapy, but rather offers an alternative to microsurgical resection for intracranial lesions. Cerebral metastases are ideal radiosurgery targets, due to their typically small size, discrete imaging appearance, and minimal invasion into surrounding normal brain. The two most common delivery systems for SRS are the gamma knife (GK) and linear accelerator (Linac). GK radiosurgical devices employ 201 concentrically aligned cobalt-60 sources that emit gamma radiation. Different size collimators are placed into spherical helmets, and aligned with the cobalt-60 sources during treatment. The use of multiple sources spreads the dose delivery over a large area, while converging within the center of the helmet to deliver a high focal dose on the target lesion. The patient’s head position is stereotactically positioned within the helmet to ensure the defined target is at this central position. Linac based systems generate high-energy photons which are delivered to the defined target along variable arcs while the patient’s head remains immobilized. In addition to intracranial pathology, Linac-based systems can be used to treat other body sites, and to deliver fractionated radiotherapy. Recent modifications to Linac radiosurgery include the advent of the Cyberknife and intensity-modulated radiation therapy (IMRT). The Cyberknife is a commercially available Linacbased system that combines a robotic arm in conjunction with an image-guidance system to track patient position, obviating the need for applying a
839
840
55
Image guided management of cerebral metastases
stereotactic frame. IMRTemploys a system of rapidly opening and closing collimators to shape the beam of radiation delivered, producing higher conformality with the defined target volume, thereby minimizing radiation delivery to normal adjacent tissues. Recent data from the RTOG 95-08 trial found no survival difference attributable to using either GK versus Linac for SRS dose delivery [74].
SRS in Addition to WBRT for the Treatment of Cerebral Metastases Several retrospective studies have demonstrated a survival advantage resulting from the addition of adjuvant SRS to WBRT in comparison with WBRT alone for the treatment of cerebral metastatic disease [75–81]. Median overall survival in the SRS plus WBRT groups range from 8 to 13.5 months – comparable with results seen for microsurgical resection plus WBRT [75–81]. A number of non-randomized, prospective studies have reported similar results [82–86]. To date, two prospective RCTs have examined the benefit of adjuvant SRS following WBRT for cerebral metastatic disease. The first of these studies compared WBRT alone (30 Gy given in 12 fractions) with WBRT plus a SRS boost (with a marginal dose prescription of 16 Gy) [87]. The primary end-point of the study was control of cerebral disease as demonstrated by imaging follow-up. In total, 27 patients with 2–4 cerebral metastases were randomized, with 14 patients in the WBRT arm and 13 patients in the WBRT plus SRS boost arm. The study was stopped at only 60% accrual following an interim analysis as a significant result with respect to the primary end-point had been observed. Use of a SRS boost significantly improved tumor control rates, with 1-year local failure rates of 100% in the WBRT alone arm compared to only 8% in the WBRT plus SRS boost arm. The median time to
local failure was 6 months versus 36 months, respectively (p = 0.0005). No significant difference was seen in median overall survival; however, the trial was not significantly powered to detect differences in overall survival, and the majority of patients died as a result of systemic disease progression. The RTOG subsequently completed the first multi-center prospective RCT (RTOG 95–08) to assess whether a SRS boost following WBRT improves survival in patients with newly diagnosed cerebral metastases [74]. The study included patients diagnosed with 1–3 cerebral metastases, with the largest being up to 4 cm diameter, and any remaining lesions up to 3 cm diameter. No RPA class III patients or patients with active extracranial disease were included. In total, 331 patients were randomized to receive WBRT plus a SRS boost (167 patients) or WBRT alone (164 patients) between January 1996 and June 2001. Thirty-one patients from the WBRT plus SRS group did not ultimately receive SRS, however they were included in their original group for the intention to treat analysis. The most significant finding of this study was that of an overall survival benefit in patients with a single brain metastasis treated with WBRT plus a SRS boost in comparison with WBRT alone. Median survival in this group increased from 4.9 months to 6.5 months (p = 0.0393). The data also supported the use of a SRS boost following WBRT in patients with up to three brain metastases, to improve performance status and minimize the need for corticosteroid use. However, no survival advantage was seen in the subgroup of patients with multiple metastases following a SRS boost. The study authors concluded that WBRT plus a SRS boost should be considered standard therapy for patients with single unresectable brain metastasis due to the survival advantage this regimen confers over WBRT alone. In addition, a SRS boost should be considered for patients with 2–3 metastases in order to maintain or improve performance status.
Image guided management of cerebral metastases
WBRT Following SRS of Cerebral Metastases Given the increased tumor control rates and overall survival demonstrated resulting from adding a SRS boost following WBRT for the treatment of cerebral metastases, the question has arisen regarding whether or not SRS treatment alone may offer similar benefits. The primary rationale for using SRS alone is to limit the theoretical neurocognitive side effects associated with WBRT. To the best of our knowledge, to date there have been five retrospective series and one completed prospective RCT documenting results obtained using SRS alone or in combination with WBRT for the treatment of cerebral metastases. A retrospective study by Pirzkall et al. examining a series of 236 patients, found no difference in overall survival between those treated with SRS plus WBRT versus SRS alone [83]. Sneed et al. reviewed their experience with 105 patients with newly diagnosed brain metastases treated with SRS alone (62 patients) or in combination with WBRT (43 patients) [88]. No differences in overall or local progression-free survival were seen. Recurrence at distant cerebral sites was more frequent in the SRS alone group. Following the addition of salvage therapy in patients who recurred, disease control did not differ between groups at the 1-year mark. These authors concluded that omitting WBRT up front in patients treated with SRS for up to four lesions does not reduce survival, or control of cerebral disease if appropriate salvage therapy is instituted at the time of recurrence. This same group subsequently conducted a multi-institutional review to assess the utility of adding up-front WBRT to SRS compared with SRS alone for the treatment of cerebral metastatic disease [89]. Data from ten institutions were reviewed. Of 569 evaluable patients, 268 had SRS alone and 301 had SRS with up-front WBRT. No differences in median overall survival times for
55
patients in RPA class I, II, or III were seen. Of note, however, is the fact that 24% of the SRS alone group ultimately received salvage WBRT. Chidel et al. retrospectively reviewed 135 patients that underwent SRS for the treatment of brain metastases with either gamma knife or Linacbased systems [90]. Similar to others, they found that omission of up-front WBRT did not impact overall survival. Progressive cerebral disease did occur in over 50% of patients, however. Lastly, in a review of 69 patients with cerebral metastases from renal cell carcinoma, local control was achieved in 96% of patients treated with SRS, and overall survival increased to 15 months [91]. Survival did not correlate with adjuvant use of WBRT. Importantly, renal cell carcinoma has traditionally been regarded as radioresistant to conventional WBRT, which may explain the lack of added benefit from supplementing SRS with WBRT in this series. To date, only one prospective RCT has examined the added benefit of WBRT following SRS for cerebral metastases [92]. This study randomized a total of 132 patients with 1–4 brain metastases (each less than 3 cm in diameter) to receive either SRS alone (67 patients) versus SRS plus WBRT (65 patients). Patients were treated during the period from October 1999 to December 2003. Of note, the prescribed radiosurgical dose in the SRS plus WBRT group was reduced by 30% compared to the SRS alone group. The results of this trial demonstrated no difference in overall median survival. However, a significant increase in the recurrence rate at 12 months was seen in the SRS alone group (76.4%) versus the SRS plus WBRT group (46.8%) (p = < 0.001), requiring more frequent salvage therapy in the SRS alone group. Nonetheless, following salvage therapy, there was no difference in the death rate attributable to neurologic disease. These results suggested that if WBRT is omitted from therapy following initial SRS, early detection and salvage therapy for recurrent disease might prevent
841
842
55
Image guided management of cerebral metastases
neurologic deterioration and death from neurologic disease progression. Recurrence of brain metastases, however, may have a negative overall impact on neurocognitive function. Regine et al. found that 71% of patients suffering from recurrent brain metastasis following SRS treatment were symptomatic from the recurrent lesions, and that in 59% of these patients deficits remained following salvage therapy [93]. The European Organization for Research and Treatment of Cancer (EORTC) has recently completed enrolment in a prospective RCT (EORTC 22952) which will provide additional insight into the importance of adjuvant WBRT following initial SRS for brain metastases [94].
SRS in the Treatment of Multiple Metastases No study has prospectively examined the role that SRS alone plays in the management of cerebral metastases. The recent RTOG 95-08 trial did examine a subgroup of patients with oligometastases (2–3 lesions) treated with either WBRT alone or in conjunction with an adjuvant SRS boost [74]. This trial failed to demonstrate a survival advantage following the SRS boost in patients with 2–3 lesions. It did, however, demonstrate benefits in terms of maintained performance status and a reduction in requirement for corticosteroid usage – both of which may potentially translate into improved patient quality of life. The apparent equivalence in terms of tumor control following microsurgical excision versus SRS, and the survival benefit seen following complete excision for up to three brain metastases, suggests that SRS for multiple brain lesions, either alone or in combination with surgery, would have similar potential to prolong survival in carefully selected patients. Whether or not this is the case awaits further study [15,34,35]. In everyday clinical practice, SRS with our without surgery, provides an excellent
option for salvage of patients with multiple metastases who have failed WBRT (> Figure 55-1).
Surgery Versus Radiosurgery for the Treatment of Cerebral Metastases The available evidence comparing microsurgical resection versus SRS for the treatment of brain metastases is currently limited to retrospective case series. Two such series suggest a clinical superiority for surgical resection in comparison with SRS. Bindal et al. reported their results for a matched case-control study, examining the efficacy of SRS versus open surgery [95]. The SRS group was comprised of 13 patients, who received a median dose of 20 Gy, while the surgical group consisted of 62 patients. Groups were matched based on their primary tumor type, extent of systemic disease, KPS, time to brain metastasis, number of brain metastases, age, and sex. This study identified a significant difference in median overall survival between the SRS group (7.5 months) and the surgical group (16.4 months), and this difference in survival was attributed to progression of the treated lesion in the SRS group. These authors concluded that indications for radiosurgery should be restricted to surgically inaccessible lesions or for patients unfit to undergo surgical resection. A criticism of this study is that the dosing regimen used in the SRS group resulted in a lower prescribed dose to the tumor margins than is considered standard. Therefore, underdosing may have hindered the efficacy of SRS at effecting tumor control in comparison with microsurgery. Also, a selection bias may have favored the surgical group, as only lesions amenable to surgery were included in this study. Another study suggesting improved outcome following surgery for cerebral metastasis in comparison with SRS was reported by Shinoura et al. [96]. This group compared recurrence rates of metastatic brain tumors following Linac SRS
Image guided management of cerebral metastases
55
. Figure 55-1 MRI of young woman with four metastatic tumors from rectal carcinoma, all growing 3 months following WBRT. The large left frontal tumor (a) was removed using awake image-guided craniotomy, sparing speech which was found overlying the tumor. General anesthesia was then immediately induced and the large cerebellar metastasis (b) was removed with the aid of surgical navigation based on a fresh registration. Five days later she underwent Gamma Knife radiosurgery for the right frontal tumor (c) and the right parietal tumor (d)
versus surgery plus WBRT. They found that the time to recurrence was 25 months in the surgical group but only 7.2 months in SRS group (p = 0.0199). Baseline patient characteristics describing extent and activity of extracranial disease was not provided, however. In addition, the SRS group had a greater number of patients with multiple metastases than the surgical group, which may have biased the study in favor of the surgical arm. Two retrospective studies support the conclusion that surgical resection and SRS offer equivalent clinical benefit for the treatment of cerebral metastases amenable to both therapies.
Auchter et al. reported the results from a multiinstitutional retrospective series looking at a group of patients with newly diagnosed brain metastasis treated with SRS plus WBRT, and also met the study inclusion criteria used by Patchell et al. previously [14,97]. In total, 122 patients from 4 institutions were identified. All patients except five received WBRT following SRS treatment. An overall local tumor control rate of 86% was observed in this group of patients. In addition, a median overall survival of 56 weeks and a median duration of functional independence (KPS > 70) of 44 weeks were observed. These authors concluded that SRS with WBRT
843
844
55
Image guided management of cerebral metastases
for the treatment of a single brain metastasis resulted in functional survival comparable with surgical resection followed by WBRT. Muacevic et al. retrospectively examined their experience comparing their results of surgery plus adjuvant WBRT with gamma knife SRS alone for the treatment of solitary brain metastases deemed suitable for radiosurgery [25]. All tumors were 3.5 cm in diameter or smaller. In total, 52 patients in the microsurgery plus WBRT group were compared with 56 patients in the GK SRS group. No statistically significant difference in median overall survival was identified. Local tumor control rates were 75 and 83% for the surgery versus radiosurgery groups, respectively (p = 0.49). There was also no difference observed in neurologic death rates at 1 year. The authors concluded that SRS alone results in local tumor control rates equal to surgery plus WBRT in selected patients, and therefore adjuvant WBRT need not be combined with SRS in this population. The remaining two studies provide data that supports a conclusion that SRS provides better local tumor control rates in comparison with microsurgery. Schoggl et al. reported results from their retrospective case-control study comparing SRS (67 patients) with microsurgical resection (66 patients) for a single cerebral metastasis [98]. Patients were treated between August 1992 and October 1996. All patients received adjuvant WBRT. Groups differed in their baseline characteristics, with the SRS group having on average smaller lesions (median size 7,800 ml) compared with surgically treated lesions (median size 12,500 ml). No significant difference in overall survival was identified. The SRS group, however, demonstrated a lower rate of local recurrence (5% vs. 17%), attributed to a better response rate of metastases traditionally considered ‘‘radioresistant’’ to WBRT. Based on their results, these authors advocate for the use of SRS as a first line therapy for the treatment of single cerebral metastases, unless the lesion is greater than 3 cm in diameter, or requires debulking due to symptomatic mass effect.
O’Neil et al. retrospectively reviewed their experience with newly diagnosed solitary brain metastases treated between 1991 and 1999 [99]. All patients had surgically accessible lesions, less than 3.5 cm diameter, and without evidence of obstructive hydrocephalus. A total of 74 patients made up the surgical arm, with 23 patients in the GK SRS arm. This study found no difference in overall survival between the two treatment arms, but did identify better local tumor control rates associated with GK SRS (p = 0.020). To date, no prospective RCT directly comparing surgery with radiosurgery has been completed. An international phase III trial (EORTC 22952) comparing surgery and radiosurgery with or without adjuvant WBRT for patients with 1–3 brain metastases has been accruing patients since 1996 [94]. Enrolment in this trial has been completed. The results from this study are eagerly anticipated, as they will shed important light on the roles each of the currently available therapeutic modalities has to play in the management of patients with cerebral metastases. Until the results of EORTC 22952 are available, the clinician must take into account several important considerations when choosing between surgical resection, SRS, or a combination of the two for the treatment of cerebral metastases. Among these considerations are the location, size and number of lesions, the presence of significant mass effect or edema, and the need for tissue diagnosis. In addition, one must consider the extent of extracranial disease and the patient’s performance status. Small, superficial tumors that are resectable and associated with minimal edema or mass effect are candidates for either modality of treatment. In this scenario, the decision regarding surgery versus SRS depends on factors such as the expertise of the treating physician, access to care (e.g., radiosurgical facilities), and patient preference. Deeply located, surgically inaccessible lesions may be better treated with SRS. Large lesions greater than approximately 3–3.5 cm may preclude SRS. Lesions causing significant symptomatology or posing an immediate threat to
Image guided management of cerebral metastases
life due to mass effect or exuberant edema may necessitate surgical resection. Patients with poor performance status and uncontrolled systemic disease are unlikely to benefit from aggressive management of their intracranial disease, even if they possess only a single cerebral lesion. In these cases, minimally invasive approaches may be more appropriate, whether this be with WBRT, fractionated stereotactic radiotherapy, or SRS. One must be careful to discern, however, whether the patient’s poor KPS is due to tumorrelated mass effect or edema, which may be improved with aggressive treatment of the cerebral disease.
55
Treatment Algorithm for Patients with Cerebral Metastatic Disease Choosing among the numerous treatment options available to the patient with cerebral metastatic disease requires that the treating clinician consider a multitude of patient and tumor-related variables in order to prescribe the most appropriate therapeutic plan. > Figure 55-2 provides a simplified outline of a treatment algorithm for this patient population. The most important prognostic factors appear to be the presence or absence of active systemic malignancy, and the patient’s neurologic and performance (KPS) status.
. Figure 55-2 A simplified treatment algorithm for the management of patients with cerebral metastatic disease. SRS, stereotactic radiosurgery; FSR, fractionated stereotactic radiotherapy; WBRT, whole-brain radiotherapy; Sx, surgical resection
845
846
55
Image guided management of cerebral metastases
Patients with aggressive systemic cancer or a poor KPS in general have few appropriate treatment options available to them. For this group, in patients with 1–3 cerebral lesions, supportive care plus WBRT, with or without SRS to all lesions is appropriate. For patients with four or more lesions, WBRT is typically offered, unless the primary histology is considered ‘‘radioresistant’’ (renal cell carcinoma, sarcoma, melanoma) in which circumstance one may elect to add SRS to the treatment plan. For patients with absent or controlled systemic disease and good performance status, one must then consider the number of cerebral lesions identified on contrast-enhanced MRI. For patients with a single lesion, surgical resection or SRS, followed by adjuvant WBRT is appropriate if the lesion size is under 3 cm in diameter. If the lesion is greater than 3 cm, or causing significant symptomatology or mass effect, surgical resection should be considered up front. For poor surgical candidates, fractionated stereotactic radiotherapy (FSR) may be used, followed by SRS if the lesion’s size reduces in response to FSR. For patients with 2–3 lesions, one may offer surgical resection of all lesions, SRS to all lesions, or a combination of surgery plus SRS, followed by adjuvant WBRT. Again, for lesions greater than 3 cm or highly symptomatic, one should consider surgical resection of this lesion up front. Patients with four or more lesions are typically offered WBRT and supportive care. In those with good systemic control and performance status one may consider offering combinations of surgical resection and SRS for progressive disease following WBRT.
Conclusions Cerebral metastasis represents in increasingly common complication in the cancer patient, owing in part to improved therapy for and survival associated with systemic cancers. Advances
in neuroimaging, microsurgical technique, stereotactic-based surgical adjuncts and noninvasive treatment modalities such as stereotactic radiosurgery have increased our ability to treat brain metastases. At the same time, the availability of numerous treatment options raises many questions regarding which strategy is most appropriate for the individual patient with metastatic brain disease. Ongoing and future studies will provide some insight into these management decisions.
References 1. Posner JB. Management of brain metastases. Rev Neurol (Paris) 1992;148(6–7):477-87. 2. Landis SH, Murray T, Bolden S, Wingo PA. Cancer statistics, 1999. CA Cancer J Clin 1999;49(1):1,8‐31. 3. Chang SD, Lee E, Sakamoto GT, Brown NP, Adler JR Jr. Stereotactic radiosurgery in patients with multiple brain metastases. Neurosurg Focus 2000;9(2):e3. 4. Wen PY, Loeffler JS. Management of brain metastases. Oncology (Williston Park), 1999. 13(7):941-54, 957‐61; discussion 961‐2, 969. 5. Lagerwaard FJ, Levendag PC, Nowak PJ, Eijkenboom WM, Hanssens PE, Schmitz PI. Identification of prognostic factors in patients with brain metastases: a review of 1292 patients. Int J Radiat Oncol Biol Phys 1999; 43(4):795-803. 6. Delattre JY, Krol G, Thaler HT, Posner JB. Distribution of brain metastases. Arch Neurol 1988;45(7): 741-4. 7. Sze G, Milano E, Johnson C, Heier L. Detection of brain metastases: comparison of contrast-enhanced MR with unenhanced MR and enhanced CT. AJNR Am J Neuroradiol 1990;11(4):785-91. 8. Mintz AP, Cairncross JG. Treatment of a single brain metastasis: the role of radiation following surgical resection. JAMA 1998;280(17):1527-9. 9. Davis PC, Hudgins PA, Peterman SB, Hoffman JC Jr. Diagnosis of cerebral metastases: double-dose delayed CT vs contrast-enhanced MR imaging. AJNR Am J Neuroradiol 1991;12(2):293-300. 10. Patchell RA. The management of brain metastases. Cancer Treat Rev 2003;29(6):533-40. 11. Healy ME, Hesselink JR, Press GA, Middleton MS. Increased detection of intracranial metastases with intravenous Gd-DTPA. Radiology 1987;165(3): 619-24. 12. Lang FF, Sawaya R. Surgical management of cerebral metastases. Neurosurg Clin N Am 1996;7(3):459-84.
Image guided management of cerebral metastases
13. Lang FF, Sawaya R. Surgical treatment of metastatic brain tumors. Semin Surg Oncol 1998;14(1):53-63. 14. Patchell RA, Tibbs PA, Walsh JW, Dempsey RJ, Maruyama Y, Kryscio RJ, Markesbery WR, Macdonald JS, Young B. A randomized trial of surgery in the treatment of single metastases to the brain. N Engl J Med 1990; 322(8):494-500. 15. Wronski M, Arbit E, McCormick B. Surgical treatment of 70 patients with brain metastases from breast carcinoma. Cancer 1997;80(9):1746-54. 16. Korinth MC, Delonge C, Hutter BO, Gilsbach JM. Prognostic factors for patients with microsurgically resected brain metastases. Onkologie 2002;25(5):420-5. 17. Pollock BE, Brown PD, Foote RL, Stafford SL, Schomberg PJ. Properly selected patients with multiple brain metastases may benefit from aggressive treatment of their intracranial disease. J Neurooncol 2003; 61(1):73-80. 18. Brown PD, Brown CA, Pollock BE, Gorman DA, Foote RL. Stereotactic radiosurgery for patients with ‘‘radioresistant’’ brain metastases. Neurosurgery 2002;51(3): 656-65; discussion 665‐7. 19. Vogelbaum MA, Suh JH. Resectable brain metastases. J Clin Oncol 2006;24(8):1289-94. 20. Pieper DR, Hess KR, Sawaya RE. Role of surgery in the treatment of brain metastases in patients with breast cancer. Ann Surg Oncol 1997;4(6):481-90. 21. Saitoh Y, Fujisawa T, Shiba M, Yoshida S, Sekine Y, Baba M, Iizasa T, Kubota M. Prognostic factors in surgical treatment of solitary brain metastasis after resection of non-small-cell lung cancer. Lung Cancer 1999;24(2): 99-106. 22. Salvati M, Capoccia G, Orlando ER, Fiorenza F, Gagliardi FM. Single brain metastases from breast cancer: remarks on clinical pattern and treatment. Tumori 1992;78(2):115-7. 23. Gaspar L, Scott C, Rotman M, Asbell S, Phillips T, Wasserman T, McKenna WG, Byhardt R. Recursive partitioning analysis (RPA) of prognostic factors in three Radiation Therapy Oncology Group (RTOG) brain metastases trials. Int J Radiat Oncol Biol Phys 1997;37(4): 745-51. 24. Sawaya R. Surgical treatment of brain metastases. Clin Neurosurg 1999;45:41-7. 25. Muacevic A, Kreth FW, Horstmann GA, SchmidElsaesser R, Wowra B, Steiger HJ, Reulen HJ. Surgery and radiotherapy compared with gamma knife radiosurgery in the treatment of solitary cerebral metastases of small diameter. J Neurosurg 1999;91(1):35-43. 26. Black PM, Johnson MD. Surgical resection for patients with solid brain metastases: current status. J Neurooncol 2004;69(1–3):119-24. 27. Tan TC, Mc LBP. Image-guided craniotomy for cerebral metastases: techniques and outcomes. Neurosurgery 2003;53(1):82-9; discussion 89‐90.
55
28. Vecht CJ, Haaxma-Reiche H, Noordijk EM, Padberg GW, Voormolen JH, Hoekstra FH, Tans JT, Lambooij N, Metsaars JA, Wattendorff AR, et al. Treatment of single brain metastasis: radiotherapy alone or combined with neurosurgery? Ann Neurol 1993;33(6):583-90. 29. Mintz AH, Kestle J, Rathbone MP, Gaspar L, Hugenholtz H, Fisher B, Duncan G, Skingley P, Foster G, Levine M. A randomized trial to assess the efficacy of surgery in addition to radiotherapy in patients with a single cerebral metastasis. Cancer 1996;78(7): 1470-6. 30. Smalley SR, Schray MF, Laws Jr, ER O’Fallon JR. Adjuvant radiation therapy after surgical resection of solitary brain metastasis: association with pattern of failure and survival. Int J Radiat Oncol Biol Phys 1987;13(11): 1611-16. 31. Patchell RA, Tibbs PA, Regine WF, Dempsey RJ, Mohiuddin M, Kryscio RJ, Markesbery WR, Foon KA, Young B. Postoperative radiotherapy in the treatment of single metastases to the brain: a randomized trial. JAMA 1998;280(17):1485-9. 32. Simpson JR, Mendenhall WM, Schupak KD, Larson D, Bloomer WD, Buckley JA, Gaspar LE, Gibbs FA, Lewin AA, Loeffler JS, Malcolm AW, Schneider JF, Shaw EG, Wharam MD Jr, Gutin PH, Rogers L, Leibel S. Follow-up and retreatment of brain metastasis. ACR appropriateness criteria. Radiology 2000;215 Suppl:1129-35. 33. Ransohoff J. Surgical management of metastatic tumors. Semin Oncol 1975;2(1):21-7. 34. Bindal RK, Sawaya R, Leavens ME, Lee JJ. Surgical treatment of multiple brain metastases. J Neurosurg 1993;79(2):210-16. 35. Iwadate Y, Namba H, Yamaura A. Significance of surgical resection for the treatment of multiple brain metastases. Anticancer Res 2000;20(1B):573-7. 36. Nieder C, Andratschke N, Grosu AL, Molls M. Recursive partitioning analysis (RPA) class does not predict survival in patients with four or more brain metastases. Strahlenther Onkol 2003;179(1):16-20. 37. Bindal RK, Sawaya R, Leavens ME, Hess KR, Taylor SH. Reoperation for recurrent metastatic brain tumors. J Neurosurg 1995;83(4):600-4. 38. Arbit E, Wronski M, Burt M, Galicich JH. The treatment of patients with recurrent brain metastases. A retrospective analysis of 109 patients with nonsmall cell lung cancer. Cancer 1995;76(5):765-73. 39. Linskey ME. The changing role of stereotaxis in surgical neuro-oncology. J Neurooncol 2004;69(1–3):35-54. 40. Golfinos JG, Fitzpatrick BC, Smith LR, Spetzler RF. Clinical use of a frameless stereotactic arm: results of 325 cases. J Neurosurg 1995;83(2):197-205. 41. Wadley J, Dorward N, Kitchen N, Thomas D. Preoperative planning and intra-operative guidance in modern neurosurgery: a review of 300 cases. Ann R Coll Surg Engl 1999;81(4):217-25.
847
848
55
Image guided management of cerebral metastases
42. Maciunas RJ. Intraoperative cranial navigation. Clin Neurosurg 1996;43:353-81. 43. Bernstein M. Outpatient craniotomy for brain tumor: a pilot feasibility study in 46 patients. Can J Neurol Sci 2001;28(2):120-4. 44. Nimsky C, Ganslandt O, Fahlbusch R. Implementation of fiber tract navigation. Neurosurgery 2006;58 4 Suppl 2:ONS-292-303; discussion ONS-303‐4. 45. Ganslandt O, Fahlbusch R, Nimsky C, Kober H, Moller M, Steinmeier R, Romstock J, Vieth J. Functional neuronavigation with magnetoencephalography: outcome in 50 patients with lesions around the motor cortex. J Neurosurg 1999;91(1):73-9. 46. Wu JS, Zhou LF, Hong XN, Mao Y, Du GH. Role of diffusion tensor imaging in neuronavigation surgery of brain tumors involving pyramidal tracts. Zhonghua Wai Ke Za Zhi 2003;41(9):662-6. 47. Clark CA, Barrick TR, Murphy MM, Bell BA. White matter fiber tracking in patients with space-occupying lesions of the brain: a new technique for neurosurgical planning? Neuroimage 2003;20(3):1601-8. 48. Holodny AI, Schwartz TH, Ollenschleger M, Liu WC, Schulder M. Tumor involvement of the corticospinal tract: diffusion magnetic resonance tractography with intraoperative correlation. J Neurosurg 2001;95(6): 1082. 49. Hlatky R, Jackson EF, Weinberg JS, McCutcheon IE. Intraoperative neuronavigation using diffusion tensor MR tractography for the resection of a deep tumor adjacent to the corticospinal tract. Stereotact Funct Neurosurg 2005;83(5–6):228-32. 50. Smits M, Vernooij MW, Wielopolski PA, Vincent AJ, Houston GC, van der Lugt A. Incorporating functional MR imaging into diffusion tensor tractography in the preoperative assessment of the corticospinal tract in patients with brain tumors. AJNR Am J Neuroradiol 2007;28(7):1354-61. 51. Lang FF, Olansen NE, DeMonte F, Gokaslan ZL, Holland EC, Kalhorn C, Sawaya R. Surgical resection of intrinsic insular tumors: complication avoidance. J Neurosurg 2001;95(4):638-50. 52. Nimsky C, Ganslandt O, Hastreiter P, Fahlbusch R. Intraoperative compensation for brain shift. Surg Neurol 2001;56(6):357-64; discussion 364‐5. 53. Nauta HJ. Error assessment during ‘‘image guided’’ and ‘‘imaging interactive’’ stereotactic surgery. Comput Med Imaging Graph 1994;18(4):279-87. 54. Lindseth F, Kaspersen JH, Ommedal S, Lango T, Bang J, Hokland J, Unsgaard G, Hernes TA. Multimodal image fusion in ultrasound-based neuronavigation: improving overview and interpretation by integrating preoperative MRI with intraoperative 3D ultrasound. Comput Aided Surg 2003;8(2):49-69. 55. Bernstein M, Al-Anazi AR, Kucharczyk W, Manninen P, Bronskill M, Henkelman M. Brain tumor surgery with the Toronto open magnetic resonance imaging system: preliminary results for 36 patients and analysis of
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
advantages, disadvantages, and future prospects. Neurosurgery 2000;46(4):900-7; discussion 907‐9. Black PM, Moriarty T, Alexander E III, Stieg P, Woodard EJ, Gleason PL, Martin CH, Kikinis R, Schwartz RB, Jolesz FA. Development and implementation of intraoperative magnetic resonance imaging and its neurosurgical applications. Neurosurgery 1997;41(4):831-42; discussion 842‐5. Steinmeier R, Fahlbusch R, Ganslandt O, Nimsky C, Buchfelder M, Kaus M, Heigl T, Lenz G, Kuth R, Huk W. Intraoperative magnetic resonance imaging with the magnetom open scanner: concepts, neurosurgical indications, and procedures: a preliminary report. Neurosurgery 1998;43(4):739-47; discussion 747‐8. Tronnier VM, Wirtz CR, Knauth M, Lenz G, Pastyr O, Bonsanto MM, Albert FK, Kuth R, Staubert A, Schlegel W, Sartor K, Kunze S. Intraoperative diagnostic and interventional magnetic resonance imaging in neurosurgery. Neurosurgery 1997;40(5):891-900; discussion 900‐2. Pamir MN, Peker S, Ozek MM, Dincer A. Intraoperative MR imaging: preliminary results with 3 tesla MR system. Acta Neurochir Suppl 2006;98:97-100. Yrjana SK, Katisko JP, Ojala RO, Tervonen O, Schiffbauer H, Koivukangas J. Versatile intraoperative MRI in neurosurgery and radiology. Acta Neurochir (Wien) 2002;144(3):271-8; discussion 278. Unsgaard G, Rygh OM, Selbekk T, Muller TB, Kolstad F, Lindseth F, Hernes TA. Intra-operative 3D ultrasound in neurosurgery. Acta Neurochir (Wien) 2006;148(3): 235-53; discussion 253. Unsgaard G, Gronningsaeter A, Ommedal S, Nagelhus Hernes TA. Brain operations guided by real-time twodimensional ultrasound: new possibilities as a result of improved image quality. Neurosurgery 2002;51(2): 402-11; discussion 411‐2. Unsgaard G, Ommedal S, Muller T, Gronningsaeter A, Nagelhus Hernes TA. Neuronavigation by intraoperative three-dimensional ultrasound: initial experience during brain tumor resection. Neurosurgery 2002;50(4): 804-12; discussion 812. Roberts DW, Miga MI, Hartov A, Eisner S, Lemery JM, Kennedy FE, Paulsen KD. Intraoperatively updated neuroimaging using brain modeling and sparse data. Neurosurgery 1999;45(5):1199-206; discussion 1206‐7. Lindner D, Trantakis C, Renner C, Arnold S, Schmitgen A, Schneider J, Meixensberger J. Application of intraoperative 3D ultrasound during navigated tumor resection. Minim Invasive Neurosurg 2006;49(4):197-202. Pallatroni H, Hartov A, McInerney J, Platenik LA, Miga MI, Kennedy FE, Paulsen KD, Roberts DW. Coregistered ultrasound as a neurosurgical guide. Stereotact Funct Neurosurg 1999;73(1–4):143-7. Rasmussen IA Jr, Lindseth F, Rygh OM, Berntsen EM, Selbekk T, Xu J, Nagelhus Hernes TA, Harg E, Haberg A,
Image guided management of cerebral metastases
68.
69.
70.
71.
72.
73. 74.
75.
76.
77.
78.
Unsgaard G. Functional neuronavigation combined with intra-operative 3D ultrasound: initial experiences during surgical resections close to eloquent brain areas and future directions in automatic brain shift compensation of preoperative data. Acta Neurochir (Wien) 2007;149(4):365-78. Nagelhus Hernes TA, Lindseth F, Selbekk T, Wollf A, Solberg OV, Harg E, Rygh OM, Tangen GA, Rasmussen I, Augdal S, Couweleers F, Unsgaard G. Computer-assisted 3D ultrasound-guided neurosurgery: technological contributions, including multimodal registration and advanced display, demonstrating future perspectives. Int J Med Robot 2006;2(1):45-59. Bonsanto MM, Staubert A, Wirtz CR, Tronnier V, Kunze S. Initial experience with an ultrasound-integrated single-RACK neuronavigation system. Acta Neurochir (Wien) 2001;143(11):1127-32. Gronningsaeter A, Kleven A, Ommedal S, Aarseth TE, Lie T, Lindseth F, Lango T, Unsgard G. SonoWand, an ultrasound-based neuronavigation system. Neurosurgery 2000;47(6):1373-9; discussion 1379‐80. Tronnier VM, Bonsanto MM, Staubert A, Knauth M, Kunze S, Wirtz CR. Comparison of intraoperative MR imaging and 3D-navigated ultrasonography in the detection and resection control of lesions. Neurosurg Focus 2001;10(2):E3. Lindseth F, Lango T, Bang J, Nagelhus Hernes TA. Accuracy evaluation of a 3D ultrasound-based neuronavigation system. Comput Aided Surg 2002;7(4):197-222. Bradley KA, Mehta MP. Management of brain metastases. Semin Oncol 2004;31(5):693-701. Andrews DW, Scott CB, Sperduto PW, Flanders AE, Gaspar LE, Schell MC, Werner-Wasik M, Demas W, Ryu J, Bahary JP, Souhami L, Rotman M, Mehta MP, Curran WJ Jr. Whole brain radiation therapy with or without stereotactic radiosurgery boost for patients with one to three brain metastases: phase III results of the RTOG 9508 randomised trial. Lancet 2004;363(9422): 1665-72. Flickinger JC, Kondziolka D, Lunsford LD, Coffey RJ, Goodman ML, Shaw EG, Hudgins WR, Weiner R, Harsh GRT, Sneed PK, et al. A multi-institutional experience with stereotactic radiosurgery for solitary brain metastasis. Int J Radiat Oncol Biol Phys 1994;28 (4):797-802. Alexander E, 3rd, Moriarty TM, Davis RB, Wen PY, Fine HA, Black PM, Kooy HM, Loeffler JS. Stereotactic radiosurgery for the definitive, noninvasive treatment of brain metastases. J Natl Cancer Inst 1995;87(1):34-40. Joseph J, Adler JR, Cox RS, Hancock SL. Linear accelerator-based stereotaxic radiosurgery for brain metastases:the influence of number of lesions on survival. J Clin Oncol 1996;14(4):1085-92. Chen JC, Petrovich Z, O’Day S, Morton D, Essner R, Giannotta SL, Yu C, Apuzzo ML. Stereotactic radiosurgery in the treatment of metastatic disease to the brain. Neurosurgery 2000;47(2):268-79; discussion 279‐81.
55
79. Hoffman R, Sneed PK, McDermott MW, Chang S, Lamborn KR, Park E, Wara WM, Larson DA. Radiosurgery for brain metastases from primary lung carcinoma. Cancer J 2001;7(2):121-31. 80. Gerosa M, Nicolato A, Foroni R, Zanotti B, Tomazzoli L, Miscusi M, Alessandrini F, Bricolo A. Gamma knife radiosurgery for brain metastases: a primary therapeutic option. J Neurosurg 2002;97 Suppl 5:515-24. 81. Petrovich Z, Yu C, Giannotta SL, O’Day S, Apuzzo ML. Survival and pattern of failure in brain metastasis treated with stereotactic gamma knife radiosurgery. J Neurosurg 2002;97 Suppl 5:499-506. 82. Gerosa M, Nicolato A, Severi F, Ferraresi P, Masotto B, Barone G, Foroni R, Piovan E, Pasoli A, Bricolo A. Gamma Knife radiosurgery for intracranial metastases: from local tumor control to increased survival. Stereotact Funct Neurosurg 1996;66 Suppl 1:184-92. 83. Pirzkall A, Debus J, Lohr F, Fuss M, Rhein B, EngenhartCabillic R, Wannenmacher M. Radiosurgery alone or in combination with whole-brain radiotherapy for brain metastases. J Clin Oncol 1998;16(11):3563-9. 84. Lutterbach J, Cyron D, Henne K, Ostertag CB. Radiosurgery followed by planned observation in patients with one to three brain metastases. Neurosurgery 2003; 52(5):1066-73; discussion 1073‐4. 85. Hasegawa T, Kondziolka D, Flickinger JC, Germanwala A, Lunsford LD. Brain metastases treated with radiosurgery alone: an alternative to whole brain radiotherapy? Neurosurgery 2003;52(6):1318-26; discussion 1326. 86. Muacevic A, Kreth FW, Tonn JC, Wowra B. Stereotactic radiosurgery for multiple brain metastases from breast carcinoma. Cancer 2004;100(8):1705-11. 87. Kondziolka D, Patel A, Lunsford LD, Kassam A, Flickinger JC. Stereotactic radiosurgery plus whole brain radiotherapy versus radiotherapy alone for patients with multiple brain metastases. Int J Radiat Oncol Biol Phys 1999;45(2):427-34. 88. Sneed PK, Lamborn KR, Forstner JM, McDermott MW, Chang S, Park E, Gutin PH, Phillips TL, Wara WM, Larson DA. Radiosurgery for brain metastases: is whole brain radiotherapy necessary? Int J Radiat Oncol Biol Phys 1999;43(3):549-58. 89. Sneed PK, Suh JH, Goetsch SJ, Sanghavi SN, Chappell R, Buatti JM, Regine WF, Weltman E, King VJ, Breneman JC, Sperduto PW, Mehta MP. A multiinstitutional review of radiosurgery alone vs. radiosurgery with whole brain radiotherapy as the initial management of brain metastases. Int J Radiat Oncol Biol Phys 2002; 53(3): 519-26. 90. Chidel MA, Suh JH, Reddy CA, Chao ST, Lundbeck MF, Barnett GH. Application of recursive partitioning analysis and evaluation of the use of whole brain radiation among patients treated with stereotactic radiosurgery for newly diagnosed brain metastases. Int J Radiat Oncol Biol Phys 2000;47(4):993-9.
849
850
55
Image guided management of cerebral metastases
91. Sheehan JP, Sun MH, Kondziolka D, Flickinger J, Lunsford LD. Radiosurgery in patients with renal cell carcinoma metastasis to the brain: long-term outcomes and prognostic factors influencing survival and local tumor control. J Neurosurg 2003;98(2):342-9. 92. Aoyama H, Shirato H, Tago M, Nakagawa K, Toyoda T, Hatano K, Kenjyo M, Oya N, Hirota S, Shioura H, Kunieda E, Inomata T, Hayakawa K, Katoh N, Kobashi G. Stereotactic radiosurgery plus whole-brain radiation therapy vs stereotactic radiosurgery alone for treatment of brain metastases: a randomized controlled trial. JAMA 2006;295(21):2483-91. 93. Regine WF, Huhn JL, Patchell RA, St Clair WH, Strottmann J, Meigooni A, Sanders M, Young AB. Risk of symptomatic brain tumor recurrence and neurologic deficit after radiosurgery alone in patients with newly diagnosed brain metastases: results and implications. Int J Radiat Oncol Biol Phys 2002;52(2):333-8. 94. Mueller RP. European Organisation for Research and Treatment of Cancer: EORTC protocol 22952: No radiotherapy versus whole brain radiotherapy for 1 to 3 brain metastases from solid tumor after surgical resection or radiosurgery – A randomized phase III trial. http://www.eortc.be/protoc/details.asp?protocol=22952.
95. Bindal AK, Bindal RK, Hess KR, Shiu A, Hassenbusch SJ, Shi WM, Sawaya R. Surgery versus radiosurgery in the treatment of brain metastasis. J Neurosurg 1996;84(5):748-54. 96. Shinoura N, Yamada R, Okamoto K, Nakamura O, Shitara N. Local recurrence of metastatic brain tumor after stereotactic radiosurgery or surgery plus radiation. J Neurooncol 2002;60(1):71-7. 97. Auchter RM, Lamond JP, Alexander E, Buatti JM, Chappell R, Friedman WA, Kinsella TJ, Levin AB, Noyes WR, Schultz CJ, Loeffler JS, Mehta MP. A multiinstitutional outcome and prognostic factor analysis of radiosurgery for resectable single brain metastasis. Int J Radiat Oncol Biol Phys 1996;35(1):27-35. 98. Schoggl A, Kitz K, Reddy M, Wolfsberger S, Schneider B, Dieckmann K, Ungersbock K. Defining the role of stereotactic radiosurgery versus microsurgery in the treatment of single brain metastases. Acta Neurochir (Wien) 2000;142(6):621-6. 99. O’Neill BP, Iturria NJ, Link MJ, Pollock BE, Ballman KV, O’Fallon JR. A comparison of surgical resection and stereotactic radiosurgery in the treatment of solitary brain metastases. Int J Radiat Oncol Biol Phys 2003;55(5): 1169-76.
52 Image Guided Management of Intracerebral Hematoma A. Losiniecki . G. Mandybur
Introduction Spontaneous intracerebral hemorrhage (ICH) remains a major cause of hemorrhagic strokes. Mortality from ICH far outpaces that from subarachnoid hemorrhage or cerebral infarction. Incidence of spontaneous ICH ranges from 13 to 35 per population of 100,000 [1–4]. As the prevalence of mortality and severity of morbidity associated with spontaneous ICH is much higher in comparison with other types of stroke, identifying potential surgical candidates is of utmost importance. To address patient selection, randomized studies comparing best medical management to open surgical evacuation might reduce the risk of death without improving functional outcome. With open surgical techniques providing less efficacious outcomes than best medical management, stereotactic evacuation of ICH may provide a less-is-more scenario. Widespread use of computed tomography (CT) scanning has made the identification of spontaneous ICH quick and simple and allows for identification of location and quantification of hemorrhage volumes. In the CT era, early diagnosis also makes earlier treatment possible. Management of spontaneous ICH to this point has had two main pathways – medical and surgical. Surgical intervention has been undertaken only when patients were no longer able to be managed medically as evidenced by either progression of symptoms or increased intracranial pressure (ICP). During these circumstances, patient outcomes after surgery are usually poor, characterized by significant mortalities and morbidities [5,6]. However, stereotactic #
Springer-Verlag Berlin/Heidelberg 2009
evacuation of spontaneous ICH provides a quick, minimally invasive approach to this difficult problem. Although some practitioners believe that less collateral injury during stereotactic aspirations may improve patient outcomes, guidelines for management of spontaneous ICH continue to evolve and Class I data are minimal [7]. A landmark study by McKissock et al. in 1961 is often cited to argue against the routine use of surgical evacuation of intracranial hemorrhages [8]. Yet the study was completed in an era before CT scans and without the advantages of many contemporary microsurgical techniques and practices of postoperative intensive care, all of which can potentially improve surgical outcomes.
Etiology of ICH Hypertension is an important risk factor for spontaneous ICH that is found in 40–60% of affected patients. Other identified causes include aneurysms, vascular malformations, coagulopathies, tumors, conversion of ischemic to hemorrhagic infarctions, amyloid angiopathy, post-traumatic reactions, and reactions to drugs, both legal and illegal. Given this long list, identification of patient-specific causes is important because treatment can vary depending on the cause. CT, which provides the diagnosis in the vast majority of ICH patients, is the acute study of choice. It provides quick identification of the location, determination of the amount of midline shift, detection of hydrocephalus, and volume estimation of the clot (> Figure 52-1). Once the patient is stabilized, additional studies
798
52
Image guided management of intracerebral hematoma
. Figure 52-1 CT is the initial study of choice in the identification of ICH and its location, extent of midline shift, hydrocephalus and volume. a, Right-sided lobar hemorrhage with extension to the cortex and accompanying midline shift b, Midline cerebellar ICH. c, Left-sided hemorrhage of the basal ganglia with extension into the ventricle and associated with hydrocephalus (with permission from Mayfield Clinic)
can be attempted to identify the root cause of the hemorrhage; such studies that may be helpful include magnetic resonance imaging (MRI) and cerebral angiography. MRI is effective in characterizing the age of the hemorrhage and, with addition of contrast, may help to identify underlying tumor as the source. Cerebral angiography can be preformed if a vascular lesion is highly suspected based on hemorrhage characteristics and location. Cardiac ultrasound can be performed to evaluate for possible valvupopathies if an embolic or infectious cause is suspected. Without completion of these studies during the evaluation of spontaneous ICH, eliminating a structural, vascular, or distant lesion as a cause can be difficult. One major issue with ICH is that, after hemorrhage, much brain damage may already have occurred. Evacuation of the hematoma may only prevent further brain injury, either by physical compression or cytotoxic by-products. Thus initial presentation and progression of neurologic symptoms are paramount in the evaluation of a patient for any kind of surgical evacuation.
Medical Management Any discussion of surgical management of spontaneous ICH must also include some basics of medical management. Initial management should be directed at the basics of airway, breathing, and circulation. The acute management of patients with spontaneous ICH involves admittance to an intensive care unit (ICU) setting, followed by frequent neurologic and medical monitoring. Neurologic exams (i.e., Glasgow Coma Score, type quick assessments) are performed often soon after stabilization. Intubation is not routinely used in all patients with spontaneous ICH but is used in those who show signs of insufficient ventilation (i.e., pO2 > 60 mm Hg or pCO2 > 50 mm Hg) or those who are obtunded and unable to protect their airway. In the setting of suspected elevated ICP, hyperventilation is appropriate yet is only a temporary measure; its use should not be prolonged. Prevention of increase in size of spontaneous ICH and maintenance of the surrounding unaffected brain is the goal of most if not all of the available medical interventions. ICH expansion
Image guided management of intracerebral hematoma
52
. Table 52-1 Medical management goals and study benefits related to spontaneous intracerebral hemorrhage (with permission from the Mayfield Clinic) Intervention
Goal range
Potential benefit
Study results
Control of blood pressure
Undefined, suggestions toward normotensive N/A
Decrease hemorrhage expansion, maintain cerebral perfusion
SBP <210 and >140 suggest improvement, yet recent studies show no correlation [10,11]
Reduction of cerebral edema
None, suggests worse outcome [12]
CPP >70 mm Hg, ICP <20 mmHg
Improve cerebral perfusion
Tend to improve outcome with CPP >70 [13,14]
N/A
Prevention of status epilepticus and associated increase in cerebral metabolism Normoglycemia provides for ideal cerebral metabolism
High incidence of seizures with lobar hemorrhages, seizures should always be treated [15,16] Elevated blood glucose is associated with increased mortality [17]
Administration of corticosteroids Monitoring of intracranial pressure Use of antiseizure medications Control of hyperglycemia
Unknown
after initial presentation has been shown to be as high as 28% within the first 24 h [9]. Although the optimal blood pressure in management of spontaneous ICH has not yet been determined, studies do exist that suggest mean arterial pressure (MAP) and cerebral perfusion pressure (CPP) should be kept within a nominal range [10,11]. Spontaneous ICH often occurs outside the setting of large-vessel vasculopathy. Therefore, risk of hemorrhagic expansion with mild blood pressure elevation may actually be less than anticipated. A clear target blood pressure in the setting of spontaneous ICH does not exist – rather physicians must balance the risk of hemorrhage expansion with maintenance of CPP. Many controversies exist regarding the management of blood pressure when spontaneous ICH presents and as such no definitive guidelines are available. Underlying coagulopathies should be addressed with correction of prothrombin time (PT) and partial thromboplastin time (PTT) and use of appropriate blood products as necessary. The routine use of steroids has not been shown to improve outcome in spontaneous ICH in comparison to placebo [12] and thus is not included as a first-line
management. Some of the medical management parameters discussed in treating spontaneous ICH are shown in > Table 52-1. The use of intraventricular catheters (IVCs) is preferred because of their effectiveness to monitor ICP and provide cerebrospinal fluid (CSF) drainage. Treatment regimens used for normalization of ICP include elevation of the head of bed, CSF drainage, pain medication and sedation, and osmotic therapy (3% saline and mannitol). When most of these treatments fail, barbiturate-induced coma can be used. These strategies can be used alone or in combination, and may provide sufficient effects to avoid surgical decompression. Much of the data that exist about the treatment of elevated cerebral hypertension come from the traumatic brain injury (TBI) literature. In TBI studies that specifically assess ICP control for spontaneous ICH patients [13,14], normalization of ICP and maintenance of CPP are indicated. Seizure activity should be monitored closely because EEG criteria show seizure activity in nearly 30% of patients; however, only 5% of these patients exhibit seizures clinically [15,16]. Any evidence of seizure activity should be aggressively
799
800
52
Image guided management of intracerebral hematoma
treated: Prophylactic administration of antiepileptic medications may reduce the risk of developing seizures, especially in patients with lobar hemorrhages [15]. In addition, blood glucose should be tightly controlled; use of an insulin drip in refractory cases is appropriate. Although the exact parameters for blood glucose do not exist, high-admission levels of blood glucose correlate with poor outcomes [17]. The medical management paradigm for spontaneous ICH may change over the next few years. Preliminary studies have examined with varying degrees of success the utility of hypothermia, recombinant activated factor VII, and hyperosmolar therapy. However, further studies are required before their universal acceptance.
Surgical Evacuation of Intracerebral Hemorrhage Questions regarding timing and which patients are most likely to benefit from surgery have yet to be scientifically answered. Some of the following randomized studies still leave questions, thus providing motivation to further evaluate new minimally invasive techniques.
Techniques for Evacuation Surgical techniques for evacuation of intracranial hemorrhage are numerous, ranging from small, minimally invasive procedures to large, decompressive evacuation techniques. The location and size of hemorrhage and medical comorbidities dictate which and when surgical procedures should be performed. A cerebellar location is a site where reasonable evidence exists for the early evacuation of hemorrhages >3 cm. Studies have shown that surgical evacuation can be superior to medical management alone, especially in patients with hemorrhage to the cortical surface
and a GCS >5 [7]. The goals of surgery must be defined before the selection of the ultimate surgical techniques. Minimally invasive/stereotactic techniques for evacuation of spontaneous ICH carry some theoretical benefits. These include reduced time under anesthesia, decreased anesthetic load with less violation of uninjured brain tissue, and the ability to evacuate deep-seated lesions (e.g., in the thalamus or pons) [18–20]. This chapter culminates with a review of the pros and cons of both craniotomy and stereotactic aspiration techniques for the treatment of spontaneous ICH.
Craniotomy The traditional craniotomy approach for evacuation of ICH involves the creation of bony window that must be sufficiently sized over the hemorrhagic site to allow for its evacuation and enough space for visualization for hemostasis. At the University of Cincinnati, a keyhole craniotomy is performed in the region of noneloquent cortex to access hematomas that lie close to the cortex (> Figure 52-2). After opening the dura mater, a small cortical tunnel is typically made to enter the clot by using bipolar forceps and suction. Once sufficiently evacuated, the ICH cavity is explored. Using bipolar forceps, the surgeon obtains hemostasis with or without use of additional synthetic hemostatic agents. Intraventricular catheters are only inserted in the setting of symptomatic hydrocephalus or large intraventricular clot burden. Some surgeons have advocated creation of large craniotomies (e.g., as used for TBIs) to relieve ICP and allow surgical hematoma evacuation. However, the true benefits for such a procedure still require study. A multicenter study, the International Surgical Trial in Intracerebral Hemorrhage, which randomized more than 1,000 patients to undergo best medical management or surgical evacuation,
Image guided management of intracerebral hematoma
52
. Figure 52-2 Surgical technique for removal of an intracerebral hemorrhage (ICH). (a) Operative set-up for frameless imageguided stereotactic trajectory to a right-sided frontal lesion. (b) Top figure, dura is opened and a cortical entry point is selected. (c) Bottom figure, blunt dissection is performed with bipolar until the clot is entered and evacuated with suction (with permission from Mayfield Clinic)
showed that the patients most likely to benefit from surgical evacuation had a moderate GCS and a clot within 1 cm of the cortical surface [5]. However, this study also suggested that some patients may actually worsen after surgery. Although the above-mentioned group was suggested to benefit from surgical evacuation, no significant differences were obtained between the two major groups. Therefore further trials are required.
Timing of Surgery Early diagnosis of an ICH that can potentially undergo surgical treatment is soon followed by a question of optimal timing. Surgery performed within 12 h has been shown to be only modestly
effective [6]. At 1-month evaluation in this study, Morgenstern et al. reported a nearly 15% mortality and poor initial functional outcomes. As only one patient had undergone surgery within 4 h of symptom onset, these findings beg the question – Is this early enough? [6]. In a retrospective review of 100 patients who underwent surgery within 7 h of symptom onset, Kaneko et al. reported good functional outcome in 35% of patients and mortality only slightly higher than 5%; these data suggest that early surgical treatment results in improved outcomes [21]. Studies have shown that enlargement of ICH occurs most dramatically within the first few hours after initial symptom onset [22,23]. Angiographically, contrast extravasation into the ICH cavity appears to slow after 6 h from onset [24]. How this relates to ideal surgical timing has not been convincingly shown.
801
802
52
Image guided management of intracerebral hematoma
Minimally Invasive/Stereotactic Evacuation of ICH The idea of using a minimally invasive technique to aide in evacuation of ICH has been studied since the late 1970s. One of the numerous techniques evaluated involved the insertion of a cannula and an Archimedes screw aspirator to aid in the break-up of blood clot and evacuation of ICH. Initial results were disappointing with a mortality rate exceeding 80% [20]. Other advances included insertion of a modified nucleotome [25], a double-track aspiration system [26], and ultrasonic aspirator [27], a waterjet irrigation system [28], and others. Although the results were not particularly encouraging, the future provides hope for improvements in imaging capabilities and surgical techniques. On the basis of the STICH trial findings, the authors concluded that all non-craniotomy techniques had worse outcome when compared with conservative management. Yet the data were not overwhelming. Many patients who were selected to undergo minimally invasive evacuations had deep-seated lesions seemingly destined to do poorly [5]. This highlights one of many questions regarding the optimal patient selection that has yet to be completely understood. Other factors that affect patient selection involve lesion size, location, and ideal surgical timing. Minimally invasive techniques offer significant theoretical advantages over traditional craniotomy techniques: use of local anesthesia, shorter operating times, and less damage to normal surrounding brain. One key to success of minimally invasive techniques lies in the accuracy of localization. Techniques for localization first included ultrasound, then CT scans, and finally MRI, all of which evolved with potential intraoperative application that could improve the precision of localization. After accurate localization of the ICH, the next step involves its evacuation, for which various techniques are available or remain under study. The ideal technique can be performed
quickly, effect minimal trauma to surrounding tissues, allow complete hematoma removal, and provide means for improved hemostasis. Endoscopic techniques described for ICH evacuation consist of insertion of an endoscope through a burr hole directly into the ICH cavity and evacuation of contents under direct visualization (> Figure 52-3). The trajectory and planning of the burr-hole placement can be simplified with stereotactic guidance. Placement of the entry point and planned trajectory is dependent on ICH location, with all attempts made to avoid eloquent tissue and known vascular structures. Endoscopic techniques using lavage followed by suction of the ICH have been reported to evacuate nearly 90% of clot burden [18]. In a randomized study of 100 patients who either underwent endoscopic techniques or were managed medically, the endoscopic group had prolonged survival, especially those with large ICHS of >50 cc. Patients younger than 60 years with lobar hemorrhages showed a significant benefit in quality of life assessments after endoscopic management versus the best medical management [18]. Another purported advantage of endoscopic evacuation is speed of surgery. In fact, some evacuations have been completed in less than 60 min [19]. The concept of stereotactic aspiration of ICH is similar to that of endoscopic evacuation. Instead of direct endoscopic visualization, an aspirating needle is placed into the middle of the hemorrhage cavity. After the evacuation of the ICH clot, a drain placed directly into the cavity then encourages further drainage. If a CT scan of the patient’s head still shows residual clot, injection of a thrombolytic agent can provide additional evacuation [29]. Although injection of a thrombolytic agent into an ICH would seem to be counterintuitive to maintenance of hemostasis, studies have shown that rebleeding rates are in the range of those encountered in open craniotomy [30]. Urokinase, which is no longer available in the U.S., was used in initial studies that involved injection of thrombolytic
Image guided management of intracerebral hematoma
52
. Figure 52-3 Endoscopic technique for removal of an intracerebral hemorrhage (ICH). A large burr hole is made and the dura is opened. In one pass, a 14 French peel-away introducer is placed into the central core of the hematoma to a depth at least two thirds of the overall hematoma diameter. The inner stylet is carefully removed while the cannula remains within the clot. Hematoma is aspirated using a syringe until there is no longer a fluid component of the clot. The endoscope is inserted and the hematoma cavity is inspected. Additional techniques, such as lavage and piece-meal removal with a nucleotome (inset) or forceps, may be used to remove the solid components of the clot (with permission from Mayfield Clinic)
agents. More recent studies have shown tPA to be a viable alternative [31,32]. Currently the MISTIE study funded by the National Institute of Health is underway to better answer this question: This study hypothesizes that tPA injection is safe, reduces clot size, and improves clinical outcomes when compared with best medical management. > Figure 52-4 shows one case example by imaging studies in which tPA was administered in conjunction with clot evacuation.
There are differences and similarities between complications of the minimally invasive techniques and those of traditional craniotomy evacuation. Rates of rebleeding are about 3% after both procedures [19,33–35]. As the use of indwelling catheter would seem to increase the risk of infection, the prophylactic use of antibiotics needs to be researched. Use of tPA introduces a rebleeding risk that is not inherently present for traditional craniotomy. In studies in
803
804
52
Image guided management of intracerebral hematoma
. Figure 52-4 CT imaging studies showing ICH before and after evacuation with instillation of tissue plasmogen activator (t-PA). (a) initial scan. (b) after one instillation of t-PA at a rate of 1 mg/10cm3 of clot. (c) after 30 days (with permission from Mayfield Clinic)
which tPA was administered, patients experienced an increase in rebleeding when aspiration was performed within 6 h, with rebleeding rates of nearly 40% if performed within the first 4 h; of note, these studies did not include a thrombolytic agent [6,36]. However, maximal benefit appears to occur no later than 12–24 h after hemorrhage [6,37], seemingly providing an ideal window for evacuation in the 6–24 h range. For example, should a structural lesion (i.e., aneurysm or arteriovenous malformation) be the underlying cause of hemorrhage, a minimally invasive operation may provide inadequate exposure to control bleeding and the result would be conversion to an open craniotomy. Although not statistically significant, Cho et al. suggested that minimally invasive techniques (especially endoscopic techniques) are more cost effective and entail shorter ICU stays than traditional open craniotomy techniques [19].
Conclusions In patients with spontaneous ICH for whom medical management is no longer tolerated and who experience a progression of neurological
deficits and/or increasing intracranial pressures, surgical evacuation of the hemorrhage can be performed. The surgical goal is to remove the blood clot related to the hemorrhagic stroke, with the aim to reduce further brain damage and thus improve the patient’s chances of survival and return to independent living. Stereotactic surgical techniques for evacuation of ICH provide an alternative to traditional open surgical evacuation. Craniotomy (in its current form) appears to be relatively ineffective or maybe even worse than medical treatment in certain patients. Less invasive techniques, including stereotactic evacuation, may improve the chances of making a meaningful recovery. The numbers of available trials that compare medical management to minimally invasive techniques are limited. Current trials have looked at routine use rather than selective use of surgery in patients who would most likely benefit. As imaging techniques continue to advance and become more available in the community setting and understanding of the pathophysiology of spontaneous ICH improves, minimally invasive evacuation may become available earlier and eventually offer long-term benefits over that of traditional open craniotomy.
Image guided management of intracerebral hematoma
References 1. Fogelholm R, Nuutila M, Vuorela AL. Primary intracerebral hemorrhage in the Jyva¨skyla¨ region, central Finland, 1985‐89: incidence, case fatality rate and functional outcome. J Neurol Neurosurg Psychiatry 1992;55:546-52. 2. Giroud M, Gras P, Chadan N, et al. Cerebral hemorrhage in a French prospective population study. J Neurol Neurosurg Psychiatry 1991;54:595-8. 3. Nilsson OG, Lindgren A, Stohl N. Incidence of intracerebral and subarachnoid hemorrhage in Southern Sweden. J Neurol Neurosurg Psychiatry 2000;69:601-7. 4. Ojemann RG, Heros RC. Spontaneous brain hemorrhage. Stroke 1983;14:468-75. 5. Mendelow AD, Gregson BA, Fernandes HM, et al. Early surgery versus initial conservative treatment in patients with spontaneous supratentorial intracerebral haematomas in the International Surgical Trial in Intracerebral Haemorrhage (STICH): a randomised trial. Lancet 2005;365:387-97. 6. Morgenstern LB, Frankowski RF, Shedden P, et al. Surgical treatment for intracerebral hemorrhage (STICH): a single center, randomized clinical trial. Neurology 1998;51:1359-63. 7. Broderick JP, Adams HP, Jr, Barsan W, et al. Guidelines for the management of spontaneous intracerebral hemorrhage: a statement for healthcare professionals from a special writing group of the Stroke Council, American Heart Association. Stroke 1999;30:905-15. 8. McKissock W, Richardson A, Taylor J. Primary intracerebral hemorrhage: a controlled trial of surgical and conservative treatment. Lancet 1961;2:221-6. 9. Jauch EC, Lindsell CJ, Adeoye O, et al. Lack of evidence for an association between hemodynamic variables and hematoma growth in spontaneous intracerebral hemorrhage. Stroke 2006;37:2061-5. 10. Flaherty ML, Woo D, Haverbusch M, et al. Racial variations in location and risk of intracerebral hemorrhage. Stroke 2005;36:934-7. 11. Flaherty ML, Haverbusch M, Sekar P, et al. Long-term mortality after intracerebral hemorrhage. Neurology 2006;66:1182-6. 12. Italian Acute Stroke Study Group. Haemodilution in acute stroke: results of the Italian haemodilution trial. Lancet 1988;1:318–21. 13. Chambers IR, Banister K, Mendelow AD. Intracranial pressure within a developing intracerebral haemorrhage. Br J Neurosurg 2001;15:140-1. 14. Fernandes HM, Siddique S, Banister K, et al. Continuous monitoring of ICP and CPP following ICH and its relationship to clinical, radiological and surgical parameters. Acta Neurochir Suppl 2000;76:463-6. 15. Passero S, Rocchi R, Rossi S, et al. Seizures after spontaneous supratentorial intracerebral hemorrhage. Epilepsia 2002;43:1175-80.
52
16. Vespa PM, O’Phelan K, Shah M, et al. Acute seizures after intracerebral hemorrhage: a factor in progressive midline shift and outcome. Neurology 2003;60:1441-6. 17. Fogelholm R, Murros K, Rissanen A, et al. Admission blood glucose and short term survival in primary intracerebral haemorrhage: a population based study. J Neurol Neurosurg Psychiatry 2005;76:349-53. 18. Auer LM, Deinsberger W, Niederkorn K, et al. Endoscopic surgery versus medical treatment for spontaneous intracerebral hematoma: a randomized study. J Neurosurg 1989;70:530-5. 19. Cho DY, Chen CC, Cheng CS, et al. Endoscopic surgery for spontaneous basal ganglia hemorrhage: comparing endoscopic surgery, sterotactic aspiration, and craniotomy in noncomatose patients. Surg Neurol 2006;65:547-56. 20. Broseta J, Gonzalez-Darder J, Barcia-Salorio JL. Stereotactic evacuation of intracerebral hematomas. Appl Neurophysiol 1982;45:443-8. 21. Kaneko M, Tanaka K, Shimada T, et al. Long-term evaluation of ultra-early operation for hypertensive intracerebral hemorrhage in 100 cases. J Neurosurg 1983;58:838-42. 22. Brott T, Broderick J, Kothari R, et al. Early hemorrhage growth in patients with intracerebral hemorrhage. Stroke 1997;28:1-5. 23. Kazui S, Naritomi H, Yamamoto H, et al. Enlargement of spontaneous intracerebral hemorrhage. Stroke 1996;27:1783-7. 24. Takada I. On the phenomena of extravasation of contrast media in cerebral angiogram of the case of hypertensive intracerebral hematoma and their clinical significance-analysis of 14 cases 1976;4:471‐8. 25. Nguyen JP, Decq P, Brugieres P, et al. A technique for stereotactic aspiration of deep intracerebral hematomas under computed tomographic control using a new device. Neurosurgery 1992;31:330-4. 26. Tanikawa T, Amano K, Kawamura H, et al. CT-guided stereotactic surgery for evacuation of hypertensive intracerebral hematoma. Appl Neurophysiol 1985;48:431-9. 27. Donauer E, Faubert C. Management of spontaneous intracerebral and cerebellar hemorrhage. In: Kaufman HH, editor. Intracerebral hematomas. New York: Raven Press; 1992. p. 211-27. 28. Mukai H, Yamashita J, Kitamura A . et al. Stereotactic aqua-stream and aspirator in the treatment of intracerebral hematoma. An experimental study. Stereotact Funct Neurosurg 1991;57(4):221-7. 29. Zuccarello M, Brott T, Derex L, et al. Early surgical treatment for supratentorial intracerebral hemorrhage: a randomized feasibility study. Stroke 1999;30:1833-9. 30. Fujii Y, Tanaka R, Takeuchi S, et al. Hematoma enlargement in spontaneous intracerebral hemorrhage. J Neurosurg 1994;80:51-7. 31. Lippitz BE, Mayfrank L, Spetzger U, et al. Lysis of basal ganglia haematoma with recombinant tissue
805
806
52 32.
33.
34.
35.
36.
37.
Image guided management of intracerebral hematoma
plasminogen activator (rtPA) after stereotactic aspiration: initial results. Acta Neurochir (Wien) 1994; 127:157-60. Schaller C, Rohde V, Meyer B, et al. Stereotactic puncture and lysis of spontaneous intracerebral hemorrhage using recombinant tissue plasminogen activator. Neurosurgery 1995;36:328-33. Broderick JP, Brott T, Zuccarello M. Management of intracerebral hemorrhage. In: Batjer HH, editor. Cerebrovascular disease. Philadelphia, PA: LippincottRaven; 1997. p. 611-27. Kanaya H, Kuroda K. Development in neurosurgical approaches to hypertensive intracerebral hemorrhage in Japan. In: Kaufman HH, editor. Intracerebral hematomas. New York: Raven Press; 1997. p. 197-210. Kaufman HH. Stereotactic aspiration with fibrinolytic and mechanical assistance. In: Kaufman HH, editor. Intracerebral hematoma. New York: Raven Press; 1992. p. 182-5. Niizuma H, Shimizu Y, Yonemitsu T, et al. Results of stereotactic aspiration in 175 cases of putaminal hemorrhage. Neurosurgery 1989;24:814-9. Lee JI, Nam do H, Kim JS, et al. Stereotactic aspiration of intracerebral hematoma: significance of surgical timing and hematoma volume reduction. J Clin Neurosci 2003;10:439-43.
49 Image-Guided Management of Brain Abscess E. Taub . A. M. Lozano
"
In uncomplicated abscess of the brain, operated on at a fairly early period, recovery ought to be the rule. —William Macewen, 1893 [1]
Incidence The incidence of brain abscesses has been estimated at 3 per 1 million people per year in Northern Ireland [2] and 0.8 per 1 million people per year in the region around Lund, Sweden [3]. The incidence in other developed areas of the world is probably similar. These figures include only cases treated by neurosurgeons. Brain abscesses are thus much less common than intracranial neoplasms but are encountered occasionally at every major neurosurgical center. This entity poses both the diagnostic challenge of early recognition to achieve the best possible outcome [2–5], and the therapeutic challenge of permanent eradication of infection while doing the least harm to the surrounding brain. In recent decades, a major decrease in mortality has been achieved through the use of computed tomography (CT), magnetic resonance imaging (MRI), image-guided stereotaxy, and improved antimicrobial agents [4].
Etiology and Pathogenesis Many organisms can give rise to brain abscesses. Streptococcus and Staphylococcus species are the most common pathogens, although anaerobes, #
Springer-Verlag Berlin/Heidelberg 2009
especially Bacteroides, are also common [3–7]. Anaerobic abscesses occur particularly frequently in children [8]. Multiple organisms are cultured in about 20% of cases, and no organisms are cultured in up to 25% [3,4,6,9]. It has been observed that even when a presumptive extracerebral source of infection can be identified, the brain abscess may nevertheless be due to a different organism or organisms [10]; the implications for treatment are clear. Pathogenic microorganisms can reach the brain parenchyma in three ways: by extension from contiguous structures, by hematogenous spread, and by direct inoculation. Some of the more common sources of infection are listed in > Table 49-1. Infection can extend contiguously from the paranasal sinuses, middle ear, mastoid air cells, or teeth. Hematogenous spread can arise from infections of the heart, lung, or other organs; intravenous drug abuse and congenital cyanotic heart disease are predisposing factors. Infectious material can also be introduced directly into the brain by trauma or by neurosurgical procedures. In civilian life, post-traumatic brain abscesses are often the result of open, depressed skull fractures, especially when these are unrecognized or inadequately treated [11]. Brain abscesses arising in military settings are often due to metallic splinters lodged in the brain and may not become clinically apparent till decades after the initial injury [12,13]. The source of brain abscesses is undetermined (presumably hematogenous) in 20–50% of cases [3–5]. Immunosuppressed persons are susceptible to brain abscesses, especially of fungal and mycobacterial origin. The acquired immune deficiency
770
49
Image-guided management of brain abscess
. Table 49-1 Sources of brain abscesses Extension of infection from contiguous structures Paranasal sinusitis Otitis media Mastoiditis Dental abscess Hematogenous spread Endocarditis Pulmonary infection Intravenous drug use Congenital cyanotic heart disease Direct inoculation Trauma (civilian, military) Neurosurgical procedures Unknown sources
syndrome (AIDS) is a common cause of immunosuppression. Stereotactic procedures of particular relevance to the care of AIDS patients are discussed elsewhere in this book.
Clinical Presentation and Differential Diagnosis Brain abscesses may come to medical attention because of a focal neurological deficit, symptoms and signs of intracranial hypertension (headache, nausea, vomiting, papilledema, impairment of consciousness), seizures, fever and other systemic signs of infection, or none of these manifestations. The diagnosis is sometimes suspected from the presence of one or more contributing historical factors (> Table 49‐1). Unfortunately, none of the typical clinical and laboratory findings are highly specific. The characteristic ring enhancement with central clearing seen on CT and MRI may be seen with primary and metastatic neoplasms as well, and neoplasms are much more common than brain abscesses. Thirty-five percent of patients with brain abscesses have a normal erythrocyte sedimentation rate, more than half have a normal leukocyte count, and more than 60% are afebrile [3]. Intraventricular rupture of a brain abscess can have rapid and
devastating effects; only by the early recognition and prompt treatment of brain abscesses can this complication be prevented [14]. In the pre-CT era, patients with suspected brain abscess sometimes were subjected to lumbar puncture to obtain cerebrospinal fluid for culture. This practice frequently fails to yield an organism [3] and has been found to lead to clinical deterioration in as many as 25% of patients [5]. A brain abscess may be a surprise finding at surgery when there is a different presumptive diagnosis. When 54 patients with known systemic cancer underwent resection or biopsy of presumed brain metastases, 2 proved to have brain abscesses [15]. In other reported series, 11 of 67 cases of brain abscess were initially misdiagnosed [9], as were a majority of 12 cases of fungal brain abscess [16]. It is prudent neurosurgical practice to send portions of brain biopsy specimens for bacterial, fungal, and mycobacterial culture whenever an abscess is considered in the differential diagnosis. Advances in diagnostic imaging in the last 10 years have made it easier to identify a brain abscess correctly as such before the diagnosis is confirmed by biopsy. Brain abscesses have been shown to have markedly different signal characteristics from cystic and/or necrotic brain tumors on diffusion-weighted MRI: the signal intensity and apparent diffusion coefficient of the diseased tissue provide highly indicative, though not absolutely reliable, clues to the underlying pathology [17–20]. When proton magnetic resonance spectroscopy is performed in addition to diffusion-weighted MRI, it seems that even higher diagnostic specificity and selectivity can be achieved, although only a small number of cases have been reported to date [21].
Nonstereotactic Methods of Treatment The currently practiced methods of treatment for brain abscesses are listed in > Table 49‐2.
Image-guided management of brain abscess
. Table 49‐2 Current methods of treatment for brain abscesses Non-stereotactic methods Nonsurgical treatment (antimicrobial agents alone) Excision Open evacuation of pus Freehand aspiration Stereotactic methods Conventional stereotactic aspiration Frameless stereotactic aspiration Endoscopic stereotactic aspiration Stereotactic aspiration with real-time imaging: Ultrasound-guided stereotactic aspiration Interactive MRI-guided stereotactic aspiration
Although this chapter deals primarily with stereotactic treatment, a review of other methods will provide the proper context for rational clinical decision making.
49
In one series, lesions that responded to antimicrobial agents alone had a mean diameter of 1.7 cm, while lesions that did not respond had a mean diameter of 4.2 cm [25]. Yang and Zhao [5] obtained good results by restricting nonsurgical treatment to lesions smaller than 2 cm in diameter. We consider nonsurgical treatment a reasonable option if the diagnosis of abscess is nearly certain on clinical grounds, the presumed extracerebral source of the infection is known and drug-sensitive organisms have been cultured from it, and the lesion is less than 2.5 cm in diameter. In such cases, broad-spectrum antimicrobial coverage should be combined with specific treatment of the presumed etiologic organism.
Excision Nonsurgical Treatment (Antimicrobial Agents Alone) Six clinical series from 1971 to 1993 reported on the use of antimicrobial agents alone in a total of 50 patients [5,9,22–25]. Five patients (10%) died, and the rest recovered. Most of the mortality was encountered in only one of the six series (4 of 10 patients) [23]. The small number of patients in each of these series precludes any definitive conclusions, and the good outcomes may be at least partly a result of ‘‘publication bias.’’ It does seem that, for carefully selected patients, the results of nonsurgical treatment may be as good as those of stereotactic aspiration, as is discussed below. Nonetheless, nonsurgical treatment generally should be avoided. All modes of surgical treatment accomplish at least three purposes: confirmation of the diagnosis of abscess, reduction of the infective load, and acquisition of tissue for culture. None of these things are possible without surgery. The size of the lesion is an important determinant of the success of nonsurgical treatment.
Large series of primary excisions of brain abscesses via craniotomy have generally had excellent results. Some representative mortality figures are 1 of 16 patients (6%) [26], 3 of 50 patients (6%) [27], 5 of 56 patients (9%) [5], and 7 of 36 patients (19%) [9]. Less invasive methods of treatment obviously can do no better than excision with regard to reduction of the infective load and acquisition of tissue for culture. Excision is more likely than stereotactic aspiration to require a general anesthetic and involves more extensive disruption of the normal brain tissue surrounding the abscess, although these factors seem to add little morbidity. The outcome data for excision have been matched in recent years by those for stereotactic aspiration. Thus, stereotactic aspiration is generally preferred as the initial treatment, with excision reserved for its occasional failures. There remain a number of specific situations in which excision is indicated as the initial treatment. An abscess containing a foreign body should be excised, not aspirated, so that the foreign body can be removed [28]. It has been suggested that
771
772
49
Image-guided management of brain abscess
posttraumatic abscesses secondary to contaminated wounds or to a communication with the paranasal sinuses almost always require excision [29]. The presence of gas in an abscess cavity demonstrates the presence of an extracranial communication or gas-forming bacteria; excision has been recommended in this situation [30]. Nocardial brain abscesses have high mortality, are usually multiloculated, and may more suitable for excision than for aspiration [31,32], although the mortality in a recent series of patients with nocardial abscesses treated by aspiration alone was zero [33]. Some neurosurgeons prefer to treat large abscesses in the posterior fossa by excision because of the greater risk associated with failed aspiration [34,35]. Finally, multi-locularity of an abscess may be an indication for excision if the neurosurgeon thinks that aspirating one or more loculi probably will not achieve an adequate removal of infective material.
only when a lesion is so large that it is unlikely to be missed on the first pass of the aspiration probe. To date, there have been at least three published series of brain abscesses treated by freehand CT-guided aspiration [37–39].
Stereotactic Methods of Treatment Conventional Stereotactic Aspiration Except in the particular situations listed above, excision has given way to stereotactic aspiration of brain abscesses as the initial treatment of choice. Typical mortality figures in recent large series of aspiration (at least 20 patients) are 8% [5], 7% [40], 5% [41], 4% [42], and 0% [43]. These results compare favorably with those of excision.
Operative Technique
Open Evacuation of PUS Maurice-Williams [36] originated a variation on excision of brain abscesses in which pus is thoroughly removed from the interior of the abscess via craniotomy but the capsule is left intact. The theoretical advantage is that the surrounding brain is disturbed less than it would be by excision; the theoretical disadvantage is that some infective material is likely to be left behind. Excellent results were obtained in MauriceWilliams’s hands: Only 1 of 27 patients (4%) died, and 24 (89%) recovered free of disability. To our knowledge, there are no reports of others using this technique.
Freehand Aspiration Freehand technique offers no advantage over stereotaxy in terms of safety or efficacy and probably should be discarded. It is a reasonable option
Our experience and that reported in the literature suggest that the safety and efficacy of stereotactic aspiration are maximized by adhering to a few technical principles. The procedure is performed under CT or MRI guidance with any of several commercially available stereotactic systems. We have used the Brown-Roberts-Wells, Leksell, and Fischer-Leibinger systems. Local anesthesia alone is sufficient in all patients except those too anxious or agitated to stay still for the procedure. Antimicrobial agents should be withheld until after surgery if possible to preserve the best chance of obtaining a positive culture. The stereotactic target should be chosen on minimally thin (1.5 mm) CT or MRI slices to minimize error caused by the partial volume effect. If the abscess contains a large volume of fluid, one should select a target that will be in the lowest part of the abscess when the patient is positioned on the operating table. In this way, a maximum volume of aspirate will be obtained. On the other
Image-guided management of brain abscess
hand, if the abscess is expected to be solid (i.e., in the cerebritis stage) or if an abscess is only one of the diagnostic possibilities, the probability of a diagnostic specimen should be maximized by targeting both the center of the lesion and an area near the contrast-enhancing rim. The trajectory to the target should be as short as possible and should avoid ventricles, cisterns, sulci, major vessels, and vital brain areas such as the primary motor cortex. The surgeon should advance the aspiration probe slowly toward the target while being aware of the mechanical resistance of the tissue. An increase and then a decrease in resistance are generally felt as the probe traverses the abscess capsule. Electrical impedance monitoring is an optional technique for further intraoperative confirmation of the position of the probe [44]. After positioning, the probe should be connected to a 10-ml syringe partially filled with normal saline, and the plunger should be pulled out 1 ml to exert negative pressure on the abscess. Excessive suction should be avoided, as a hemorrhage may result. Whatever volume of fluid emerges should be sent for pathological examination, Gram staining, aerobic and anaerobic bacterial cultures, and fungal and mycobacterial cultures. The volume of the aspirate should be compared to the estimated volume of the abscess as calculated from the preoperative imaging study. For a spherical unilocular abscess, the relevant formula is V = 4/3 pr3, where r is the radius of the sphere.
Antimicrobial Agents Aspiration of abscesses is intended to remove most, but not all, of the infective organisms and must be combined with appropriate antimicrobial chemotherapy to be effective. As was stated above, we prefer to withhold antimicrobial agents before surgery if possible. Broadspectrum coverage is begun immediately after the procedure and is tailored to the causative
49
organism or organisms once the species and drug sensitivities are known or is given for a full course if the cultures are negative. Antimicrobial agents generally are continued for 4–6 weeks after surgery. The abscess is monitored with follow-up imaging studies until it has disappeared. The optimal choice of antimicrobial agents for broad-spectrum coverage has not, to our knowledge, been defined in a controlled study. Probably no single regimen can be recommended as being the best, as the prevalence of organisms resistant to particular agents varies among communities and in general increases over time. However, a few principles of agent selection can be derived from the published data. Grampositive aerobes may be covered with penicillin [6] or a related agent. In many hospitals, nafcillin or vancomycin may be the preferred agent because of the likely presence of resistant organisms. An increasing incidence of methicillinresistant Staphylococcus aureus (MRSA) brain abscesses has been documented in recent years [45]. Gram-negative aerobes may be covered with an aminoglycoside or a third-generation cephalosporin [4,5,34]. Anaerobes may be sensitive to penicillin but may require another agent, such as metronidazole [6,8,34]. In the absence of universally valid recommendations, it is best for each neurosurgical service to determine an appropriate regimen in consultation with specialists in infectious disease. Some neurosurgeons have instilled antimicrobial agents directly into the abscess cavity either at the time of aspiration or afterward via indwelling catheters [9,42,46,47]. To our knowledge, there are no clinical data regarding the possible additional benefit of this practice.
Other Medications Dexamethasone
The well-known efficacy of dexamethasone in reducing peritumoral brain edema has led
773
774
49
Image-guided management of brain abscess
neurosurgeons to ask whether it has a role to play in the treatment of brain abscesses. The consensus in experimental studies of dexamethasone in brain abscess is that it reduces perilesional edema but also slows the process of abscess encapsulation [48–51]. The clinical significance of the latter finding is unclear. Thus, dexamethasone should be used in all patients with brain abscess in whom perilesional edema is severe enough to contribute to neurological impairment. Its use in other patients is optional. Anticonvulsants
Twenty-five to forty-five percent of patients with brain abscesses have seizures as a component of the presentation [27,52]. Other patients develop a seizure disorder weeks, months, or even years after eradication of the lesion or lesions. We give anticonvulsants to all patients before surgery. We do not recommend the long-term use of anticonvulsants as prophylaxis in patients who have never had a seizure.
Possible Need for Reaspiration Yang and Zhao [5] in Tienjin, China, reported the largest series of brain abscesses treated by aspiration in the CT era. In the 69 patients so treated, ‘‘two or three procedures were usually sufficient,’’ and 12 patients went on to require excision when repeated aspiration failed. Five of sixteen patients in another series required excision when repeated aspiration failed; these were unusually large abscesses that were aspirated with the freehand technique [37]. In contrast, only 1 of 20 patients in the Toronto Hospital series required more than a single aspiration procedure, and none required excision [43].
Drainage Catheters In a number of series, external drainage catheters were placed in some [9] or all [42,47]
brain abscesses that were treated by aspiration. Kondziolka and coworkers [41] placed external drainage catheters only in large abscesses (>3 cm). Broggi and associates [46] implanted an intracavitary catheter connected to a subcutaneous Rickham reservoir, which could then be tapped to gain access to the abscess cavity. Indwelling catheters allow continued drainage of pus in the days after surgery and thus may reduce the need for repeated aspiration (which, however, is rare in some series; see above). They also may be used for repeated intracavitary instillation of antimicrobial agents; the possible benefit of this has not been determined. The outcome data for these techniques do not appear to differ significantly from those for simple aspiration, although the numbers are small.
Other Techniques Frameless Stereotactic Aspiration Frameless stereotaxy is an alternative method of directing the aspiration probe to a target chosen on preoperative CT or MRI. The results obtainable with this technique would be expected a priori to be the same as those of conventional stereotactic aspiration, as long as the frameless targeting is accurate. Laborde and associates [53] reported good results in two patients.
Endoscopic Stereotactic Aspiration Hellwig and colleagues [54] reported on the stereotactic aspiration of brain abscesses through an endoscope in seven patients, with good results. There have been at least two further small series [55,56]. Endoscopy adds to the stereotactic technique a direct visual confirmation of the adequate evacuation of pus, as well as an opportunity to see and electrocoagulate bleeding points on the inside of the abscess
Image-guided management of brain abscess
capsule. Whether these small advantages confer a better outcome is unknown.
Ultrasound-Guided Stereotactic Aspiration Berger [57] described a skull-mounted apparatus for ultrasound-guided stereotactic biopsy of brain lesions through a burr hole in awake or anesthetized patients. Five patients underwent stereotactic aspiration of brain abscesses with this device in the original study [57] and a subsequent report [58]. A further nine cases were reported by a third group of neurosurgeons [59], with good results. This technique offers a real-time intraoperative view of the advancing needle as it enters the abscess cavity and of the abscess before and after aspiration. These features are unavailable with conventional or frameless stereotaxy.
Interactive MRI-guided Stereotactic Aspiration Kollias and Bernays [60] recently described the stereotactic aspiration of brain abscesses under intraoperative MRI guidance, with good results. Like intraoperative ultrasound, this technique provides real-time images of the procedure as it is being performed.
Multiple Brain Abscesses Multiple brain abscesses are seen in 10–50% of patients with brain abscesses, depending on the series [61]. There may be as few as 2 abscesses or as many as 20 or more. The general principles of treatment are the same as those for solitary brain abscesses: tissue diagnosis, reduction of the infective load, and directed antimicrobial
49
chemotherapy. It is obviously logistically impractical as well as unsafe to operate on each of many brain abscesses in a single patient. Fortunately, it is also unnecessary, as multiple abscesses generally arise by hematogenous spread and are likely to be due to a single organism or a single mix of organisms. Basit and coauthors [52] in 1989 and Kratimenos and Crockard [62] in 1991 aspirated or excised only the largest of multiple abscesses and then gave tailored antimicrobial therapy. The mortality figures were 5 of 21 patients (24%) and 2 of 11 patients (18%), respectively. Mamelak and coauthors [61] operated on all abscesses that were larger than 2.5 cm in diameter, were situated in critical areas of the brain, or caused significant mass effect. This more aggressive policy resulted in the performance of 43 procedures on 13 patients, while 3 further patients were treated with antimicrobial agents alone. The results were excellent: 15 of 16 patients recovered and 1 (6%) died. Rousseaux et al. [63] used antimicrobial agents alone in 10 of 12 patients with multiple brain abscesses. One patient (10%) died, and the rest made a good recovery. These authors recommended operating on all abscesses larger than 3 cm in diameter.
Conclusion Brain abscesses are curable lesions that should be diagnosed and treated rapidly to prevent progression and permanent neurological sequelae. The mainstay of treatment for brain abscesses is stereotactic aspiration. This should be followed by antimicrobial chemotherapy directed at the causative organism or organisms, which is continued for 4–6 weeks. For multiple brain abscesses, good results have been obtained by aspirating or excising only those greater than 2.5 cm in diameter, those situated in critical brain areas, and those causing significant mass effect.
775
776
49
Image-guided management of brain abscess
References 1. Macewen W. Pyogenic infective diseases of the brain and spinal cord. Glasgow: James Maclehose and Sons; 1893. 2. McClelland CJ, Craig EF, Crockard HA. Brain abscesses in Northern Ireland: a 30 year community review. J Neurol Neurosurg Psychiatry 1978;41:1043. 3. Svanteson B, Nordstrom CH, Rausing A. Non-traumatic brain abscess: epidemiology, clinical symptoms and therapeutic results. Acta Neurochir (Wien) 1988;94:57. 4. Mampalam TJ, Rosenblum ML. Trends in the management of bacterial brain abscesses: a review of 102 cases over 17 years. Neurosurgery 1988;23:451. 5. Yang S, Zhao C. Review of 140 patients with brain abscess. Surg Neurol 1993;39:290. 6. Gortvai P, De Louvois J, Hurley R. The bacteriology and chemotherapy of acute pyogenic brain abscess. Br J Neurosurg 1987;1:189. 7. Stapleton SR, Bell BA, Uttley D. Stereotactic aspiration of brain abscesses: is this the treatment of choice? Acta Neurochir (Wien) 1993;121:15. 8. Brook I. Aerobic and anaerobic bacteriology of intracranial abscesses. Pediatr Neurol 1992;8:210. 9. Bidzifiski J, Koszewski W. The value of different methods of treatment of brain abscess in the CT era. Acta Neurochir (Wien) 1990;105:117. 10. Loftus CM, Osenbach RK, Biller J. Diagnosis and management of brain abscess. In: Wilkins RH, Rengachary SS, editors. Neurosurgery. 2nd ed. New York: McGraw-Hill; 1996. p. 3285-98. 11. Stephanov S. Brain abscesses from neglected open head injuries: experience with 17 cases over 20 years. Swiss Surg 1999;5(6):288-92. 12. Wegner-Kempf L, Tornow K, Schmiedek P. Intrazerebraler Abszess 48 Jahre nach Granatsplitterverletzung [Intracerebral abscess 48 years after grenade splinter injury]. Der Radiologe 1994;34(11):671-3. 13. Lee JH, Kim DG. Brain abscess related to metal fragments 47 years after head injury. Case report. J Neurosurg 2000;93(3):477-9. 14. Takeshita M, Kagawa M, Yato S, Izawa M, Onda H, Takakura K, Momma K. Current treatment of brain abscess in patients with congenital cyanotic heart disease. Neurosurgery 1997;41(6):1270-8. 15. Patchell RA, Tibbs PA, Walsh JW, et al. A randomized trial of surgery in the treatment of single metastases to the brain. N Engl J Med 1990;322:494. 16. Young RF, Gade G, Grinnell V. Surgical treatment for fungal infections in the central nervous system. J Neurosurg 1985;63:371. 17. Noguchi K, Watanabe N, Nagayoshi T, Kanazawa T, Toyoshima S, Shimizu M, Seto H. Role of diffusionweighted echo-planar MRI in distinguishing between brain abscess and tumour: a preliminary report. Neuroradiology 1999;41(3):171-4.
18. Chang SC, Lai PH, Chen WL, Weng HH, Ho JT, Wang JS, Chang CY, Pan HB, Yang CF. Diffusion-weighted MRI features of brain abscess and cystic or necrotic brain tumors: comparison with conventional MRI. Clin Imaging 2002;26(4):227-36. 19. Guzman R, Barth A, Lo¨vblad KO, El-Koussy M, Weis J, Schroth G, Seiler RW. Use of diffusion-weighted magnetic resonance imaging in differentiating purulent brain processes from cystic brain tumors. J Neurosurg 2002;97(5):1101-7. 20. Fertikh D, Krejza J, Cunqueiro A, Danish S, Alokaili R, Melhem ER. Discrimination of capsular stage brain abscesses from necrotic or cystic neoplasms using diffusionweighted magnetic resonance imaging. J Neurosurg 2007;106(1):76-81. 21. Lai PH, Hsu SS, Ding SW, Ko CW, Fu JH, Weng MJ, Yeh LR, Wu MT, Liang HL, Chen CK, Pan HB. Proton magnetic resonance spectroscopy and diffusion-weighted imaging in intracranial cystic mass lesions. Surg Neurol 2007;68 Suppl 1:S25-36. 22. Berg B, Franklin G, Cuneo R, et al. Nonsurgical cure of brain abscess: early diagnosis and follow-up with computerized tomography. Ann Neurol 1978;3:474. 23. Chun CH, Johnson JD, Hofstetter M, Raff MJ. Brain abscess: a study of 45 consecutive cases. Medicine (Baltimore) 1986;65:415. 24. Heineman HS, Braude AL, Osterholm JI. Intracranial suppurative disease: early presumptive diagnosis and successful treatment without surgery. JAMA 1971; 218:1542. 25. Rosenblum ML, Hoff JT, Norman D, et al. Nonoperative treatment of brain abscesses in selected high-risk patients. J Neurosurg 1980;52:217. 26. Choudhury AR, Taylor JC, Whitaker R. Primary excision of brain abscess. BMJ 1977;2:1119 27. Taylor JC. The case for excision in the treatment of brain abscess. Br J Neurosurg 1987;1:173. 28. Emery E, Redondo A, Berthelot JL, Bouali I, Ouahes O, Rey A. Abce`s et empye`mes intracraˆniens. prise en charge neurochirurgicale [Intracranial abscess and empyema. neurosurgical management]. Ann Fr Anesth Reanim 1999;18(5):567-73. 29. Patir R, Sood S, Bhatia R. Post-traumatic brain abscess: experience of 36 patients. Br J Neurosurg 1995;9:29. 30. Young RF, Frazee J. Gas within intracranial abscess cavities: an indication for surgical excision. Ann Neurol 1984;16:35. 31. Hall WA, Martinez AJ, Dummer JS, Lunsford LD. Nocardial brain abscess: diagnostic and therapeutic use of stereotactic aspiration. Surg Neurol 1987;28:114. 32. Mamelak AN, Obana WG, Flaherty JF, Rosenblum ML. Nocardial brain abscess: treatment strategies and factors influencing outcome. Neurosurgery 1994;35:622. 33. Lee GY, Daniel RT, Brophy BP, Reilly PL. Surgical treatment of nocardial brain abscesses. Neurosurgery 2002;51(3):668-71.
Image-guided management of brain abscess
34. Gormley WB, del Busto R, Saravolatz LD, Rosenblum ML. Cranial and intracranial bacterial infections. In: Youmans JR, editor. Neurological surgery. 4th ed. Philadelphia: Saunders; 1996. p. 3191-220. 35. Agrawal D, Suri A, Mahapatra AK. Primary excision of pediatric posterior fossa abscesses—towards zero mortality? A series of nine cases and review. Pediatr Neurosurg 2003; 38(2):63-7. 36. Maurice-Williams RS. Experience with ‘‘open evacuation of pus’’ in the treatment of intracerebral abscess. Br J Neurosurg 1987;1:343. 37. Stroobandt G, Zech F, Thauvoy C, et al. Treatment by aspiration of brain abscesses. Acta Neurochir (Wien) 1987;85:138. 38. Savitz MH. CT-guided needle procedures for brain lesions: 20 years’ experience. Mt Sinai J Med 2000;67 (4):318-21. 39. Seliem RM, Assaad MW, Gorombey SJ, Moral LA, Kirkwood JR, Otis CN. Fine-needle aspiration biopsy of the central nervous system performed freehand under computed tomography guidance without stereotactic instrumentation. Cancer 2003;99(5):277-84. 40. Hsieh PC, Pan HC, Chung WY, Lee LS. Computerized tomography-guided stereotactic aspiration of brain abscesses: experience with 28 cases. Zhonghua Yi Xue Za Zhi (Taipei) 1999;62(6):341-9. 41. Kondziolka D, Duma CM, Lunsford LD. Factors that enhance the likelihood of successful stereotactic treatment of brain abscesses. Acta Neurochir (Wien) 1996;127:85. 42. Hasdemir MG, Ebeling U. CT-guided stereotactic aspiration and treatment of brain abscesses: an experience with 24 cases. Acta Neurochir (Wien) 1993;125:58. 43. Shahzadi S, Lozano AM, Bernstein M, et al. Stereotactic management of bacterial brain abscesses. Can J Neurol Sci 1996;23:34. 44. Organ LW, Tasker RR, Moody NF. Brain tumor localization using an electrical impedance technique. J Neurosurg 1968;28:35. 45. Roche M, Humphreys H, Smyth E, Phillips J, Cunney R, McNamara E, O’Brien D, McArdle O. A twelve-year review of central nervous system bacterial abscesses; presentation and aetiology. Clin Microbiol Infect 2003;9(8):803-9. 46. Broggi G, Franzini A, Peluchetti D, Servello D. Treatment of deep brain abscesses by stereotactic implantation of an intracavitary device for evacuation and local application of antibiotics. Acta Neurochir (Wien) 1985;76:94. 47. Itakura T, Yokote H, Ozaki F, et al. Stereotactic operation for brain abscess. Surg Neurol 1987;28:196.
49
48. Bohl I, Wallenfang T, Bothe H, Schiumami K. The effect of glucocorticoids in the combined treatment of experimental brain abscess in rats. Adv Neurosurg 1981;9:125. 49. Quartey GRC, Johnston JA, Rozdilsky B. Decadron in the treatment of cerebral abscess: an experimental study. J Neurosurg 1976;45:301. 50. Schroeder K, McKeever PE, Schaberg D, Hoff JT. Effect of dexamethasone on experimental brain abscess. J Neurosurg 1987;66:264. 51. Yildizhan A, Pas¸aog˘lu A, Kandemir B. Effect of dexamethasone on various stages of experimental brain abscess. Acta Neurochir (Wien) 1989;96:141. 52. Basit AS, Ravi B, Banerji AK, Tandon PN. Multiple pyogenic brain abscesses: an analysis of 21 patients. J Neurol Neurosurg Psychiatry 1989;52:591. 53. Laborde G, Klimek L, Harders A, Gilsbach J. Frameless stereotactic drainage of intracranial abscesses. Surg Neurol 1993;40:16. 54. Hellwig D, Bauer BL, Dauch WA. Endoscopic stereotactic treatment of brain abscesses. Acta Neurochir Suppl (Wien) 1994;61:102. 55. Fritsch M, Manwaring KH. Endoscopic treatment of brain abscess in children. Minim Invasive Neurosurg 1997;40(3):103-6. 56. Longatti P, Perin A, Ettorre F, Fiorindi A, Baratto V. Endoscopic treatment of brain abscesses. Childs Nerv Syst 2006;22(11):1447-50. 57. Berger MS. Ultrasound-guided stereotaxic biopsy using a new apparatus. J Neurosurg 1986;65:550. 58. Borgstein RL, Moxon RA, Hately W, et al. Preliminary experience with the Berger neurobiopsy device for ultrasound guided aspiration and biopsy of intracranial lesions. Clin Radiol 1991;44:98. 59. Strowitzki M, Moringlane JR, Steudel W. Ultrasoundbased navigation during intracranial burr hole procedures: experience in a series of 100 cases. Surg Neurol 2000;54(2):134-44 60. Kollias SS, Bernays RL. Interactive magnetic resonance imaging-guided management of intracranial cystic lesions by using an open magnetic resonance imaging system. J Neurosurg 2001;95(1):15-23 61. Mamelak AN, Mampalam TJ, Obana WG, Rosenblum ML. Improved management of multiple brain abscesses: a combined surgical and medical approach. Neurosurgery 1995;36:76. 62. Kratimenos G, Crockard HA. Multiple brain abscesses: a review of fourteen cases. Br J Neurosurg 1991;5:153. 63. Rousseaux M, Lesoin F, Destee A, et al. Developments in the treatment and prognosis of multiple cerebral abscesses. Neurosurgery 1985;16:304.
777
50 Image-Guided Management of Brain Stem Lesions M. Levivier
The brain stem, which comprises the mesencephalon, pons, and medulla, is a highly complex and functional structure. It is packed with cranial nerves nuclei, many neuronal fascicules and pathways, as well as reticular formations. Its surface has intimate relationships with afferent and efferent cranial nerves, and is surrounded by delicate vasculature. The brain stem is embedded deep in the cranium, located almost entirely in the posterior fossa, protected by the clivus and the petrous bone anteriorly and laterally, and covered by the cerebellum posteriorly. Because of the complex neurosurgical access and the functional importance of the structure, the management of mass lesions of the brain stem remains difficult and controversial. The advances in imaging modalities have improved many aspects of the diagnostic and therapeutic techniques currently available for patients with intracranial pathology. Investigations such as computed tomography, magnetic resonance (MR), and angiography, have become more sensitive and provide finer details. New imaging techniques based on MR technology, such as spectroscopy and diffusion tensor imaging (DTI), or based on the use of radioactive tracers, such as positron emission tomography (PET), can provide additional information on both the nature of the lesion and the functional aspects of the surrounding neural structures. Current imaging modalities, combined with improved computational speed and advanced display technology, allow optimized management of mass lesions of the brain stem, not only at the diagnostic level, but also in using image-guided neurosurgery for their treatment. Indeed, an #
Springer-Verlag Berlin/Heidelberg 2009
increasing number of imaging techniques can be incorporated in software used for framebased stereotactic biopsy, conventional neuronavigation, or radiosurgery, and are all applied for the management of brain stem lesions.
Imaging of Brain Stem Tumors and Management Strategy MR represents the current imaging of choice for the diagnostic work-up of mass lesions of the brain stem. MR has high-contrast resolution, is devoid of artifacts, and provides multiplanar, multimodal (e.g., T1-weighted, T2-weighted, FLAIR, contrast-enhancement) information that are invaluable on the topography, anatomical relationships, and characteristics of lesions involving the brain stem. It allows evaluating if the tumor is focal, diffuse, or infiltrating. In many instances, it will determine if the origin of the tumor is intraaxial or extra-axial. These characteristics are of importance to better approach the diagnosis and the therapeutic approach. In some cases, however, it is difficult to establish the origin of the tumor, as intra-axial tumors may be associated with an exophytic extension, and extra-axial tumors may present with secondary infiltration of the brain stem through one of the foramina of Luschka. MR is therefore the current single most important imaging technique to decide on which neurosurgical approach can be offered to patients with brain stem lesions. A typical intrinsic lesion, which is suggestive of a brain stem glioma, will need further evaluation on its operability. Based on its MR characteristics, the status of the patient,
780
50
Image-guided management of brain stem lesions
and the therapies available, stereotactic biopsy is often indicated; in some selected cases, direct resection may be performed. Pilocytic astrocytoma may be suggested by its typical marked contrast enhancement. It may be managed by stereotactic biopsy first, or by direct approach for which, neuronavigation is of great help. Residual or recurrent pilocytic astrocytoma can be treated by radiosurgery. Extrinsic tumors, such as ependymoma or medulloblastoma, may invade the brain stem secondarily. A careful imaged-based planning and preparation of the surgery allows optimal resection of such tumors under neuronavigation. Metastasis spread into the brainstem is relatively rare. However, typical images of brain stem metastases can be found, especially in patients with melanoma, lung and breast carcinoma. Such lesions can benefit from radiosurgical therapy. Bleeding in the brain stem deserve detailed MR work-up to evaluate the exact location of the hematoma with respect to the brain structure, especially the surface of the brain stem, as well as to identify an underlying cause. Cavernoma is the most common cause of bleeding in the brain stem and its surgical excision, when possible, is the treatment of choice. Radiosurgery represent an alternative when surgery in contraindicated. These image-guided approaches can also be used in symptomatic cavernomas with progressive brain stem dysfunction due to growing of the lesion without bleeding. Other vascular lesions that may bleed in the brain stem include arteriovenous malformations, for which radiosurgery is indicated. In some instances, other image modalities are useful for the diagnosis and management of lesions of the brain stem. Typically, when MR cannot identify a cause of bleeding, or when an arteriovenous malformation is suspected, angiography is mandatory. This will be useful not only to establish the diagnosis, but also to determine the treatment approach, which may include radiosurgery. In tumors, PET is useful to reveal areas of hypermetabolism, which correlates with tumors of higher grade. Also, the PET characteristics,
especially when heterogeneous, are helpful to guide a stereotactic biopsy, or to focus a partial resection under neuronavigation towards the most aggressive part of the tumor.
Stereotactic Biopsy for Brain Stem Lesions Although the use of image-guided stereotactic brain biopsy is regarded as a safe and reliable approach in supratentorial lesions, its application to the management of infratentorial lesions, especially those involving the brain stem, has remained limited. Moreover, since the direct surgical approach of brainstem lesions for their removal is often associated with a high morbidity, it has favored a nonsurgical management of brain stem tumors in many centers, leading to the prescription of empiric adjuvant treatment, with radiation therapy and/or chemotherapy. Modern neuroimaging techniques, especially MR, have increased our knowledge on brain stem tumors. The more precise assessment of their location and extension, and more specific characteristics of their nature has also predispose to base their management of a tentative diagnosis based on the clinical history and image characteristics, especially in the pediatric population [1]. However, it is well established that the tumor diagnosis and grading is a significant prognostic factor, including in adult brain stem gliomas [2,3]. Moreover, tumor sampling helps in establishing more refined prognostic factors based on biological and genomic characteristics. More information on the characteristics of the tumor give access to more treatment possibilities, especially regarding new approaches with targeted therapies. Similarly, in the pediatric population, the use of systematic biopsy of brain stem tumors allow to include those young patients in new therapeutic protocols [4]. Actually, in many centers, stereotactic biopsy of lesions of the brain stem is performed whenever possible, with an
Image-guided management of brain stem lesions
acceptable low surgical morbidity and mortality, which is no greater than that associated with lesions in other brain locations, including in the pediatric population [5–10]. Stereotactic biopsy of lesions of the brain stem can be performed using either a transfrontal approach or a transcerebellar approach. Location of the lesion, as defined on the preoperative MR work-up or during the analysis of the stereotactic images in the planning software, will determine the best approach. Ideally however, the approach should be defined before the intervention is scheduled, as it will influence frame placement position, as well as the anesthesia setting and the position of the patient on the operating table. MR images, whether obtained in stereotactic conditions, or co-registered with a stereotactic CT, are mandatory for optimal planning of stereotactic biopsy in the brain stem. It allows to visualize the anatomy of brain stem and surrounding structures accurately, as well as to define precisely the target for biopsy within the
50
brain stem lesion. However, the diagnostic yield of this procedure culminates at 95–96%. Other imaging modalities may help to improve the accuracy of targeting lesions of the brain stem. The author has integrated PET data information in the planning of stereotactic biopsy, routinely since 1990 [11,12]. Its systematic prospective use in brain stem lesions suggest that PET guidance improves the diagnostic yield allowing to reduce the sampling procedure, including in the pediatric population [13,14] (> Figure 50-1). The transfrontal approach provides a direct route to all parts of the brain stem, as it follows its longitudinal axis. It is the only available option for the biopsy of tegmental lesions of the midbrain. Lesions located close to the midline and lower in the brain stem, such as in the pons or medulla, can also be approached using the same route. Actually, this stereotactic approach is a well-known route used under local anesthesia in functional procedures, such as the stereotactic electrode implantation for deep brain stimulation,
. Figure 50-1 Examples of combined MR- and PET-guided targeting for stereotactic biopsy in gliomas of the brain stem, illustrating the two approaches. (a) The lesion is located lower in the pons, and the PET-defined target is lateral in the lesion; a transcerebellar approach has been planned. (b) The lesion is located high in the midbrain and the PET-defined target is on the midline; a transfrontal approach has been planned
781
782
50
Image-guided management of brain stem lesions
with a long record of safety. Usually, a parasagittal, anterior coronal, entry point is used. After traversing the frontal cortex, the trajectory will run either through the frontal horn of the ventricle or lateral to it, before entering the anterior thalamus, the cerebral peduncle, and then the target in the brain stem. The planning software, which allows to visualize simultaneously the three orthogonal planes (axial, coronal, and sagittal) and/or reformatted images, is used to calculate an optimal trajectory and to adjust the entry point, based on the target definition and analysis of the safety of the route through the cerebrum. This will lessen the risk of hemorrhage from the pial vessels, especially at the level of the midbrain and cerebellum, by staying within the brain and not crossing the subarachnoid space. Typically, this leads to a burr-hole located about 2 cm from the midline, avoiding the sagittal sinus, entering the brain through a midpoint in a gyrus, and compatible with a trajectory close to the midline. Technically, the procedure is conducted as for stereotactic biopsies in other brain regions. In most frame systems, the coordinate system requires that the plane of the base ring be positioned below the level of the target. Thus, careful frame placement is needed in order to position the base ring low enough, while ensuring that this will not limit the fixation of the fiducial systems for image acquisition, or limit the proper placement of the arc system during surgery. In any case, the base ring must be placed far enough from the target, to ensure the acquisition of artifact-free stereotactic images. The biopsy procedure is performed with the patient in prone position, under local or general anesthesia, depending on the individual practice and setting. A side-cutting canulla (Sedan biopsy needle) is the most commonly used tool for tissue sampling; most manufacturers provide smaller diameter canullas (2.1 mm or less), which should be used for stereotactic biopsy in that area. Staged biopsies along the trajectory are not recommended when performing biopsy in the brain stem; limited suction-aspiration
(usually, with a maximum of 1-ml suction) is obtained at the level of the target. The suboccipital trancerebellar approach is used to reach lesions located laterally in the pons. The trajectory is located entirely in the posterior fossa; it traverses the homolateral cerebellar hemisphere and middle cerebellar peduncle. As such, this approach is also indicated for lesions of the cerebellar peduncles and of the cerebellar hemispheres. Based on the target definition, the planning software is used to define the shortest and safest route through the cerebellar peduncle and hemisphere, while avoiding the tentorium. The planning software then allows to define the entry point and burr-hole position in the lateral occipital bone. The technical aspects described above for the transfrontal approach also apply here. Frame placement may differ however, and there may be some physical limitation to access the entry point and to allow proper setting of the coordinates on the stereotactic apparatus. Frames, such as those used in the BRW/CRW or Leksell systems, have to be placed low, below the level of the lesion, even when planning a trancerebellar approach. Care must taken to anticipate correct access to the occipital bone for the time of the surgical procedures. In some instances, if the biopsy access is rendered difficult by the position of the homolateral posterior post, the latter may be removed at the time of surgery. Some other systems, such as the Fischer–Leibinger ZD, or the Laitinen system, use of an inverted frame placement (i.e., high base ring, with low inverted posts) for posterior fossa lesions, with the plane of the base ring placed above the level of the lesion; the entry point and biopsy trajectory are entirely located below the base ring. In most instances, the biopsy is performed under general anesthesia with the patient in prone position. However, local anesthesia with the patient in a sitting position can be considered if there is contraindication to the general anesthesia. The approach to lesions of the tectum of the midbrain differs however. They should be
Image-guided management of brain stem lesions
biopsied similarly to lesions of the pineal region. The major theoretical risk is bleeding from injury of the internal cerebral veins situated above the tumor, the basilar veins of Rosenthal laterally, as well as the precentral cerebellar vein, the great vein of Galen, and the posterior medial choroidal arteries posteriorly. Thus, tectal lesions should be approached from a more anterior and lateral entry point, in order to pass under the internal cerebral veins, above the basilar vein of Rosenthal, and beneath the precentral cerebellar vein. Current planning software allow to simulate accurately such trajectories on MR. Actually, stereotactic biopsy in the pineal region has mortality, morbidity, and diagnostic rates that are not different from stereotactic biopsy in other regions of the brain [15].
Other Image-guided Procedures for Brain Stem Lesions The evolution of imaging technology allows an earlier diagnosis and a better anatomical understanding of lesions of the brain stem. Parallely, the advent of image-guidance has modified our neurosurgical approaches to the brain stem and open new therapeutic avenues. Some of them are illustrated below, as examples of the developments in image-guided neurosurgery for lesions of the brain stem.
Neuronavigation for Surgery of the Brain Stem The evolution of skull base surgery and a better understanding of the surgical anatomy of the brain stem have allowed neurosurgeons to develop new avenues in brain stem surgery. These advances reflect technological improvements, which include the use of frameless neuronavigation systems, both for the planning of the approach to the brain stem and for the surgical guidance during the procedure. This is especially
50
true for discrete lesions, for which more aggressive approaches are now recommended. The reports of successful resection of cavernomas of the brain stem during the recent years represent typical examples of the evolution of the management of brain stem lesions, with that respect [16–19]. The selection of an optimal and safe approach for the resection of discrete lesions of the brain stem remains a critical issue, however. Using the two-point method described by the group of R. Spetzler at the Barrow Neurological Institute [20,21] different surgical approaches can be defined. A point is placed in the center of the lesion and a second point is placed where the lesion is the closest to the surface of the brain stem. A straight line connecting these two points dictates the optimal trajectory and surgical approach. This method helps to define the choice between the different surgical approaches, which allow access to a corresponding area of the brain stem. These approaches usually represent major surgical procedures, and include the orbitozygomatic, subtemporal, petrosal, retrosigmoid, supracerebellar infratentorial, retrosigmoid suboccipital, and far lateral approaches. Simulating the two-point method in a navigation planning software using high quality MR is of great help to determine the angle of attack and to analyze the surgical approaches that are possible (> Figure 50-2). The integration of this information in the surgical navigation system is also helpful during surgery, including during the skull base approach [22]. Moreover, because surgery for brain stem lesions is at high risk the analysis of tensor imaging–based fiber tracking adds essential information to the preoperative planning. Hence, the integration of the tractography into the neuronavigation system will make possible to anticipate the location of the major fiber tracts during surgery. Thus, this image-guided method may also increase the likelihood of total resection of tumors adjacent to, or involving, eloquent fiber tracts in the brain stem, in order to avoid new neurological deficits after surgery. This strategy
783
784
50
Image-guided management of brain stem lesions
. Figure 50-2 Composite picture of two different planning performed to study the possibility to perform open surgical approaches in an hemorrhagic cavernoma of the brain stem, comparing a retrosigmoid approach (a), and a supracerebellar infratentorial approach (b), which are both simulated in 3-D (c)
has already been applied to the resection of a deeply located brain stem cavernous angioma [23] and will certainly benefit to other surgical approaches to the brain stem. Although the management of intrinsic lesions of the brain stem remains more controversial, the surgical management of intrinsic gliomas has evolved towards a more aggressive surgical treatment during the two last decades. Indeed, direct resection can be performed with relatively low morbidity in subgroups of lesions previously managed conservatively [24–26]. Based on MR imaging, brain stem gliomas can be categorized as focal, cervicomedullary, dorsally exophytic, or diffuse. A focal brain stem glioma grows as an expanding
mass, which usually dislocates the neighboring nervous structures without invading them. Also, as it grows, the tumor tends to move towards the surface of the brain stem, which will influence the feasibility of the resection and choice of surgical approach. It is the experience of many neurosurgeons that brain stem glioma with a focal growth pattern are often benign and totally respectable without neurological worsening, which suggest that these tumors do not directly invade the surrounding neurological structures [24,26– 28]. As such, the image-guided neurosurgical techniques should be used to optimize their approach, as described above for discrete lesions of the brain stem.
Image-guided management of brain stem lesions
Frameless Image-Guidance in the Brain Stem Although frameless navigation systems are now routinely used for the biopsy of supratentorial tumors, most neurosurgeons recognize that frame-based procedures are still mandatory for lesions of the brain stem. As frameless stereotaxy has already been used for deep brain stimulation
50
comparable accuracy than frame-based systems [29], it is not far fetched that frameless systems may be used in the future to biopsy brain stem lesions. Along the same line, the placement of an infusion cannula for the convectionenhanced delivery of therapeutic agents in the brain stem via a transfrontal approach, as already been performed using a frameless navigation system [30].
. Figure 50-3 Radiosurgery using combined MR- and PET-target definition in a patient with a recurrent pilocytic astrocytoma of the brain stem. The hypermetabolic area that was visible in the tumor at the time of radiosugery (LGK Leksell Gamma Knife) disappeared 9 months after radiosurgery; at that time the tumor was still visible but had started to disclose a necrotic center. The latter was even more visible between 12 and 24 months after radiosurgery. At 36 months, the lesion has disappeared on MR. During that time, there was recurrence of increased metabolism in the lesion area
785
786
50
Image-guided management of brain stem lesions
Radiosurgery for Brain Stem Lesions Radiosurgery is used an alternative or as a combined treatment for numerous lesions of the brain. It represents one the least invasive image-guided procedures in neurosurgery, as the stereotactic delivery of convergent beams is performed through the intact skull with minimal constraints for the patient. Radiosurgery is often used to treat tumors in the brain stem. One of the most common indications is metastases of the brain stem. Our group and several others have recently shown that radiosurgery for brain stem metastases provide a high tumoral control (up to 95%) and prolonged survival, with a relatively low complication rate related to the treatment [31–35]. This confirms that, owing to the high risk of surgical resection of brain stem metastases and low efficacy of medical treatment, radiosurgery should be proposed upfront when the size of the lesion is compatible with this approach. Other tumors of the brain stem, especially focal gliomas, may also benefit from radiosurgery [36]. Pilocytic astrocytomas have a high tumoral control with radiosurgery, and this technique is very useful when a multimodality approach is necessary [37]. In order to better define the target volume on stereotactic images, we have integrated the metabolic information of PET in the planning procedure for radiosurgery of brain tumors [38,39]. When combined with PET metabolic data during the follow-up after radiosurgery, this approach provides valuable information on the response and prognosis of intrinsic tumors of the brain stem treated with radiosurgery (> Figure 50-3). Vascular lesions of the brain stem are also treated with radiosurgery. Arteriovenous malformation (AVM) of the brain stem are difficult to treat. Radiosurgery can be performed as a single treatment in small lesions, or as part of a multimodality approach in larger AVM. However, the rate of obliteration is lower, and the risk of new neurological deficit is higher than for AVM of
other locations [40–42]. This emphasizes the difficulty in treating patients with deeply located AVM, the majority of whom are also poor candidates for resection or embolization. In symptomatic cavernomas, with imaging-confirmed hemorrhages, and for which resection is considered to be of too high risk, radiosurgery represents an interesting alternative, as it confers a reduction in the risk of new hemorrhage [43].
References 1. Albright AL, Packer RJ, Zimmerman R, Rorke LB, Boyett J, Hammond GD. Magnetic resonance scans should replace biopsies for the diagnosis of diffuse brain stem gliomas: a report from the Children’s Cancer Group. Neurosurgery 1993;33:1026-9. 2. Kesari S, Kim RS, Markos V, Drappatz J, Wen PY, Pruitt AA. Prognostic factors in adult brainstem gliomas: a multicenter, retrospective analysis of 101 cases. J Neurooncol 2008;88:175-83. 3. Rosenthal MA, Ashley DM, Drummond KJ, Dally M, Murphy M, Cher L, Thursfield V, Giles GG. Brain stem gliomas: patterns of care in Victoria from 1998–2000. J Clin Neurosci 2008;15:237-40. 4. Roujeau T, Machado G, Garnett MR, Miquel C, Puget S, Geoerger B, Grill J, Boddaert N, Di Rocco F, Zerah M, Sainte-Rose C. Stereotactic biopsy of diffuse pontine lesions in children. J Neurosurg 2007;107:1 Suppl Pediatrics:1-4. 5. Kondziolka D and Lunsford LD. Results and expectations with image-integrated brainstem stereotactic biopsy. Surg Neurol 1995;43:558-62. 6. Kratimenos GP, Nouby RM, Bradford R, Pell MF, Thomas DG. Image directed stereotactic surgery for brain stem lesions. Acta Neurochir (Wien) 1992;116: 164-70. 7. Hall WA. The safety and efficacy of stereotactic biopsy for intracranial lesions. Cancer 1998;82:1749-55. 8. Pincus DW, Richter EO, Yachnis AT, Bennett J, Bhatti MT, Smith A. Brainstem stereotactic biopsy sampling in children. J Neurosurg 2006;104: 2 Suppl Pediatrics:108–14. 9. Samadani U, Stein S, Moonis G, Sonnad SS, Bonura P, Judy KD. Stereotactic biopsy of brain stem masses: decision analysis and literature review. Surg Neurol. 2006;66:484-9. 10. St George EJ, Walsh AR, Sgouros S. Stereotactic biopsy of brain tumours in the paediatric population. Childs Nerv Syst 2004;20:163-7. 11. Levivier M, Goldman S, Bidaut LM, Luxen A, Stanus E, Przedborski S, Bale´riaux D, Hildebrand J, Brotchi J.
Image-guided management of brain stem lesions
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
Positron emission tomography-guided stereotaxic brain biopsy. Neurosurgery 1992;31:792-7. Levivier M, Goldman S, Pirotte B, Brucher J-M, Bale´riaux D, Luxen A, Hildebrand J, Brotchi J. Diagnostic yield of stereotactic brain biopsy guided by positron emission tomography with [18F]fluorodeoxyglucose. J Neurosurg 1995;82:445-52. Massager N, David P, Goldman S, Pirotte B, Wikler D, Salmon I, Nagy N, Brotchi J, Levivier M. Combined MR and PET imaging in brain mass lesions: diagnostic yield in a series of 30 stereotactically biopsied patients. J Neurosurg 2000;93:951-7. Pirotte B, Lubansu A, Massager N, Wikler D, Goldman S, Levivier M. Results of positron emission tomography guidance and reassessment of the utility of and indications for stereotactic biopsy in children with infiltrative brainstem tumors. J Neurosurg 2007;107:5 Suppl Pediatrics: 392-9. Regis J, Bouillot P, Rouby-Volot F, Figarella-Branger D, Dufour H, Peragut JC. Pineal region tumors and the role of stereotactic biopsy: review of the mortality, morbidity, and diagnostic rates in 370 cases. Neurosurgery 1996;39: 907-12. Bruneau M, Bijlenga P, Reverdin A, Rilliet B, Regli L, Villemure JG, Porchet F, de Tribolet N. Early surgery for brainstem cavernomas. Acta Neurochir (Wien) 2006;148:405-14. Ferroli P, Sinisi M, Franzini A, Giombini S, Solero CL, Broggi G. Brainstem cavernomas: long-term results of microsurgical resection in 52 patients. Neurosurgery 2005;56:1203-12. Fritschi JA, Reulen HJ, Spetzler RF, Zabramski JM. Cavernous malformations of the brain stem. A review of 139 cases. Acta Neurochir (Wien) 1994;130:35-46. Samii M, Eghbal R, Carvalho GA, Matthies C. Surgical management of brainstem cavernomas. J Neurosurg 2001;95:825-32. Brown AP, Thompson BG, Spetzler RF. The two-point method: evaluating brain stem lesions. BNI Q 1996;12:20-4. Porter RW, Detwiler PW, Spetzler RF. Surgical approaches to the brain stem. Op Tech Neurosurg 2000;3:114-30. Brinker T, Arango G, Kaminsky J, Samii A, Thorns U, Vorkapic P, Samii M. An experimental approach to image guided skull base surgery employing a microscope-based neuronavigation system. Acta Neurochir (Wien) 1998;140:883-9. Chen X, Weigel D, Ganslandt O, Fahlbusch R, Buchfelder M, Nimsky C. Diffusion tensor-based fiber tracking and intraoperative neuronavigation for the resection of a brainstem cavernous angioma. Surg Neurol 2007;68:285-91. Bricolo A, Turazzi S. Surgery for gliomas and other mass lesions of the brainstem. Adv Tech Stand Neurosurg 1995;22:261-341.
50
25. Le´vesque MF, Parker F. MKM-guided resection of diffuse brainstem neoplasms. Stereotact Funct Neurosurg 1999;73:15-8. 26. Teo C, Siu TL. Radical resection of focal brainstem gliomas: is it worth doing? Childs Nerv Syst 2008;24: 1307-14. 27. Constantini S, Epstein F. Surgical indication and technical considerations in the management of benign brain stem gliomas. J Neurooncol 1996;28:193-205. 28. Hoffman HJ, Becker L, Craven MA. A clinically and pathologically distinct group of benign brain stem gliomas. Neurosurgery 1980;7:243-8. 29. Holloway KL, Gaede SE, Starr PA, Rosenow JM, Ramakrishnan V, Henderson JM. Frameless stereotaxy using bone fiducial markers for deep brain stimulation. J Neurosurg 2005;103:404-13. 30. Lonser RR, Warren KE, Butman JA, Quezado Z, Robison RA, Walbridge S, Schiffman R, Merrill M, Walker ML, Park DM, Croteau D, Brady RO, Oldfield EH. Real-time image-guided direct convective perfusion of intrinsic brainstem lesions. Technical note. J Neurosurg 2007;107:190-7. 31. Fuentes S, Delsanti C, Metellus P, Peragut JC, Grisoli F, Regis J. Brainstem metastases: management using gamma knife radiosurgery. Neurosurgery 2006;58:37-42. 32. Hussain A, Brown PD, Stafford SL, Pollock BE. Stereotactic radiosurgery for brainstem metastases: survival, tumor control, and patient outcomes. Int J Radiat Oncol Biol Phys 2007;67:521-4. 33. Kased N, Huang K, Nakamura JL, Sahgal A, Larson DA, McDermott MW, Sneed PK. Gamma knife radiosurgery for brainstem metastases: the UCSF experience. J Neurooncol 2008;86:195-205. 34. Lorenzoni JG, Devriendt D, Massager N, Desmedt F, Simon S, Van Houtte P, Brotchi J, Levivier M. Brain stem metastases treated with radiosurgery: prognostic factors of survival and life expectancy estimation. Surg Neurol 2008;(in press). 35. Yen CP, Sheehan J, Patterson G, Steiner L. Gamma knife surgery for metastatic brainstem tumors. J Neurosurg 2006;105:213-9. 36. Yen CP, Sheehan J, Steiner M, Patterson G, Steiner L. Gamma knife surgery for focal brainstem gliomas. J Neurosurg 2007;106:8-17. 37. Hadjipanayis CG, Kondziolka D, Gardner P, Niranjan A, Dagam S, Flickinger JC, Lunsford LD. Stereotactic radiosurgery for pilocytic astrocytomas when multimodal therapy is necessary. J Neurosurg 2002;97:56-64. 38. Levivier M, Wikler D Jr, Massager N, David P, Devriendt D, Lorenzoni J, Pirotte B, Desmedt F, Simon S Jr, Goldman S, Van Houtte P, Brotchi J. The integration of metabolic imaging in stereotactic procedures including radiosurgery: a review. J Neurosurg 2002;97 5 Suppl:542-50. 39. Levivier M, Massager N, Wikler D, Lorenzoni J, Ruiz S, Devriendt D, David P, Desmedt F, Simon S, Van Houtte P,
787
788
50
Image-guided management of brain stem lesions
Brotchi J, Goldman S. Use of stereotactic PET images in dosimetry planning of radiosurgery for brain tumors: clinical experience and proposed classification. J Nucl Med 2004;45:1146-54. 40. Massager N, Re´gis J, Kondziolka D, Njee T, Levivier M. Gamma knife radiosurgery for brainstem arteriovenous malformations: preliminary results. J Neurosurg 2000; 93:102-3. 41. Maruyama K, Kondziolka D, Niranjan A, Flickinger JC, Lunsford LD. Stereotactic radiosurgery for brainstem
arteriovenous malformations: factors affecting outcome. J Neurosurg 2004;100:407-13. 42. Pollock BE, Gorman DA, Brown PD. Radiosurgery for arteriovenous malformations of the basal ganglia, thalamus, and brainstem. J Neurosurg 2004;100: 210-4. 43. Hasegawa T, McInerney J, Kondziolka D, Lee JY, Flickinger JC, Lunsford LD. Long-term results after stereotactic radiosurgery for patients with cavernous malformations. Neurosurgery 2002;50:1190-7.
41 Impedance Recording in Central Nervous System Surgery R. J. Andrews . J. Li . S. A. Kuhn . J. Walter . R. Reichart
Introduction and Historical Aspects Introduction This chapter will first describe electrical bioimpedance and its relevance to neurosurgeons, and provide a brief historical review. Current and near-future techniques for central nervous system (CNS) tissue identification (e.g., brain vs. tumor) are then considered. A brief review of impedance ‘‘imaging’’ – electrical impedance tomography EIT) – is then presented, given EIT may assume a significant clinical role in the future. The final section presents the new field of charge transfer at the neuronal/subneuronal level, made possible in the past decade by advances in nanoelectrode techniques, and contrasts such a neural-electrical interface (NEI) with the traditional macro- and micro-electrodes for neuromodulation or deep brain stimulation (DBS).
Electrical Bioimpedance The conduction of electric current through biological tissues depends on the tissue’s composition. This is important not only in the use of impedance for localization within the CNS (e.g., gray vs. white matter, specific nuclei), but also for tissue identification (e.g., tumor vs. brain) and tissue status (e.g., normal vs. edematous brain). Although the clinical use of impedance monitoring in neurosurgery was greater prior #
Springer-Verlag Berlin/Heidelberg 2009
to the advent of computed tomography (CT) and magnetic resonance imaging (MRI), there continue to be important applications of the principles of brain impedance in neuromodulation (deep brain stimulation – DBS – in particular) and also the evolving field of brain imaging based on impedance (electrical impedance tomography – EIT). However, as technological advances allow the development of a NEI at the neuronal or sub-neuronal level (e.g., axon, dendrite, intracellular), the need to understand the properties of charge monitoring and transfer at the neuronal level become paramount. Ohm’s Law describes the relationship between voltage, current, and resistance to current flow in a direct current (DC) situation: Rðor Oresistance; in ohmsÞ ¼ Vðvoltage; in voltsÞ=Iðcurrent; in amperesÞ In an alternating current (AC) situation, one must add another term – reactance – in addition to resistance, because of the phase changes in voltage and current in the AC situation: Z ðimpedance; in ohmsÞ ¼ R ðresistance; in ohmsÞ þ X ðreactance; in ohmsÞ
If X = 0, the impedance is purely resistive. If X < 0, the reactance is capacitive; if X > 0, the reactance is inductive. A capacitor can be used to store electric charge; an inductor can be used – in the form of electromagnetic coils – as a transformer. Conductance or conductivity is the reciprocal of resistance.
632
41
Impedance recording in central nervous system surgery
Bioimpedance in brain depends only in part on characteristics of the brain or spinal cord tissue: gray versus white matter, cell composition (percentage of neurons vs. glial cells), extent of myelination, extracellular tissue fluid composition, etc. Characteristics of the electrode are relevant: composition (e.g., noble metal vs. ceramic), contact area with CNS tissue, orientation within CNS tissue, proximity to capillaries or other blood vessels, corrosion, etc. Also relevant is monopolar versus bipolar recording. Perhaps the major source of impedance in deep brain stimulation is the layer of gliosis or scar tissue which forms around the electrode [1,2]. The importance of impedance is clear to neurosurgeons performing DBS surgery: the greatly reduced ‘‘life’’ of a pulse generator driving a deep brain electrode with low impedance (<500 Ω), versus one driving an electrode with high impedance (e.g., >2 KΩ) is clear – one pulse generator may function for more than 5 years, the other for only a year or less (not to mention differences in clinical efficacy and volume of tissue activated (VTA) [1]).
Historical Aspects Bioimpedance was first studied by Hoeber nearly 100 years ago, who investigated the conductivity of the erythrocyte membrane and the cell interior [3]. It was not until the middle of the century, however, that Hodgkin and Huxley conducted their Nobel prize-winning research that marked the era of membrane biophysics. The phenomenon of spreading depression was also described in the early 1950s, and noted to be accompained by a significant increase in corticial impedance. The conductance (reciprocal of impedance) of the rabbit cerebal cortex was studied in a series of experiments in the mid-1950s, with various factors affecting impedance being noted: brain cooling, cerebrospinal fluid (CSF) drainage, exsanguinations, and cirulatory arrest all resulted in increased
brain impedance-but intrestingly ether anesthesia did not have a consistent effect on impedance [4]. A brief review of bioimpedance research has been written by one of the pioneers in the field [5].
Impedance Monitoring for CNS Tissue Identification CNS Impedance Prior to CT/MRI Prior to CT and MRI imaging for tissue identification and localization in the CNS, impedance monitoring played a significant role in several neurosurgical procedures. These will be reviewed briefly to place the current interest in impedance measurement in perspective. One early study of impedance differences between normal brain and malignant gliomas studied formalin-fixed autopsy specimens [6]. Malignant gliomas typically had impedance measurements less than one-half that of normal brain. A clinical study of 14 patients with intracranial tumors reported shortly prior to the CT era found that brain tumors softer than normal brain tended to have lower impedance than normal brain, whereas firmer tumors (meningiomas) had higher impedance; penetration of a tumor capsule with the recording electrode demonstrated a transient increase in impedance [7]. Impedance monitoring has also been used in conjunction with image-guided stereotactic brain biopsies [8,9]. General observations from such studies include: 1.
2.
Tissue impedance tends to vary with tissue density (i.e., increased impedance in cyst walls, tumor capsules, and firm tumors such as certain meningiomas). Cerebrospinal fluid (CSF), edematous brain, and necrotic tissue all have lower impedance than normal brain.
Impedance recording in central nervous system surgery
3.
Given variations in electrode characteristics and monitoring systems, changes in impedance are probably of more value than absolute impedance values (in ohms).
Apart from brain tumor and cyst localization, the other major use of impedance monitoring has been in percutaneous cordotomies. Localization is determined by the drop in impedance as the neck tissues are penetrated and the CSF entered, followed by a rise in impedance as the spinal cord is penetrated once the CSF space has been traversed [10,11].
Impedance Monitoring During CT/MRI Image Guidance Brain Biopsy – Current Status The advent of DBS for movement disorders over the past 20 years has brought focus on electrical stimulation of the brain as a substitute for, and extension of, ablative procedures. The importance of impedance in the efficacy of DBS has become increasingly apparent, as noted in A. II. above and in the literature [1]. Additionally, microrecording of spontaneous electrical activity from specific brain nuclei (e.g., the subthalamic nucleus (STN) in DBS for Parkinson’s disease) has become an important aspect of precise localization – an adjunct to high-resolution MRI localization for functional neurosurgery. Thus a platform has been established that offers an opportunity for refinements in intraoperative impedance monitoring over techniques of previous decades. The group at Friedrich-Schiller-University in Jena, Germany, has recently begun state-ofthe-art impedance measurements during intracranial procedures with CT/MRI guidance, in particular stereotactic brain biopsy. The technique is similar to that customarily used for DBS electrode placement: following high-resolution CT and MRI scans with a stereotactic frame attached
41
to the patient’s head, a burr hole is made and the dura cauterized and opened. A standard microdrive is used to introduce the microelectrode with a platin tip (1 mm thickness) into the brain along the trajectory to the tumor (> Figure 41-1). Continuous photo documentation is made of the electrode depth and recording data to facilitate correlation with histopathology and molecular biology at each point along the trajectory. Following the recording of extracellular potentials along the trajectory, serial biopsies are taken for standard neurohistopathology as well as mRNA extraction. Assays include ion channels (sodium, potassium, chloride, calcium) and neurotransmitter receptors. This multimodality technique promises to yield correlations between imaging, electrical patterns, histopathology, and molecular biology (> Figure 41-2).
Electrical Impedance Tomography Background Electrical impedance tomography (EIT) is the term adopted over 20 years ago for the technique of stimulating, electrode by electrode, a volume of tissue with multiple electrodes applied (e.g., the head) in order to obtain data on the impedance of the tissue at any point within the volume queried (> Figure 41-3). In practice, for EIT of the brain EEG-type electrodes are affixed to the scalp and recordings made (> Figure 41-4). The concept of mapping the impedance of a volume of material was first utilized more than 75 years ago in geological studies, and more recently in industry to detect air bubbles, etc [12]. The term ‘‘tomography’’ – in comparison with computed tomography (CT) – is a misnomer since, unlike the x-rays used in CT, electrical current injected into a volume of tissue cannot be confined to a plane. The resulting EIT ‘‘images’’ are values for
633
634
41
Impedance recording in central nervous system surgery
. Figure 41-1 Intraoperative deep brain microrecording in a brain tumor patient (prior to collection of biopsy specimens). The monitor shows the recorded potentials. The platin electrode is clamped into the microdrive device, which in turn is fixed to the Leksell® stereotactic frame. Total electromagnetic quiescence is required during microrecording [Reichart R, et al., unpublished results]
the impedance at various points (more accurately volumes or voxels) within the volume investigated. Because of variations in the contact impedances of the electrodes, absolute values are problematic; thus relative impedances at the various points or voxels are usually recorded (so-called ‘‘difference imaging,’’ in contrast to ‘‘absolute imaging’’). Additional data are obtained by varying the frequency of the stimulation (usually from the kHz to MHz range), multi-frequency electrical impedance tomography (MFEIT). Several recent reviews detail the issues involved in the development of EIT for clinical use [12,13].
Applications of EIT Several clinical applications of EIT are currently being pursued, although none has achieved clinical status: (1) breast cancer detection;
(2) pulmonary emboli detection; (3) gastric emptying time (in the assessment of gastrointestinal disorders); (4) brain disorders [12]. The use of EIT in brain disorders has concentrated on epilepsy focus localization and the early detection of ischemic versus hemorrhagic stroke. The latter application, stroke identification, is seen as valuable in emergency departments and clinics where CT is not readily available; EIT could provide rapid information that a stroke is ischemic and not hemorrhage – thus increasing the number of patients who might benefit from early hemolytic therapy in acute stroke (where the treatment must be instituted within a few hours of stroke onset). The most extensive experience to date with MFEIT for brain lesions (stroke, brain tumors, and arteriovenous malformations) has yet to document sufficient specificity to allow MFEIT to assume a clinical role at present [14].
Impedance recording in central nervous system surgery
41
. Figure 41-2 Correlation between intraoperative photo documentation, brain microrecording, and postoperative histopathology in a brain tumor patient [Reichart R, et al., unpublished results]
EIT in the Future Advances in computational processing have made EIT and MFEIT feasible over the past
10–15 years, but the issue of contact (electrode) impedance variability in particular has made clinical usefulness elusive. Thus recent techniques have sought to develop noncontact methods for
635
636
41 . Figure 41-2
Impedance recording in central nervous system surgery
(Continued)
inducing the current and measuring the resulting electrical field (impedance), i.e., the use of coils for induction and recording [12]. One such technique is magnetic induction tomography (MIT), which has the possibility of resolution referred to as MIT-spectroscopy (MITS).
Another technique is combining EITwith MRI, i.e., magnetic resonance electrical impedance tomography (MREIT). One variation of MREIT obtains the MRI with a current injection through surface electrodes to obtain a current density/conductivity image of the tissue under investigation [12,15].
Impedance recording in central nervous system surgery
. Figure 41-3 Schematic of the application of a current around the surface electrodes on a biological tissue (e.g., the human scalp) to obtain an impedance tomogram of tissue conductivity
Electrical Charge Monitoring and Transfer at the Neuronal Level – Contrasts Between Macro/MicroElectrodes and Nano-Electrodes
spatial complexities introduced by the extracellular matrix and the non-neuronal supporting cells (e.g., astrocytes). As CNS recording and stimulation become more fine tuned – potentially down to the sub-neuronal level – the need to construct a neural-electrical interface (NEI) that mimics the communication within the CNS becomes evident. As we move from simple ‘‘brain stimulation’’ to true neuroprosthetics, the NEI must become bidirectional and multifunctional: electrical and neurochemical (neurotransmitter) information must be monitored by the neuroprosthetic device, and in turn the device must be able to ‘‘sculpt’’ the local CNS landscape toward normal functioning (e.g., for epilepsy, movement disorders, spinal cord injury). Until recently, the major advances in neuromonitoring and neurostimulation have been refinements in metal microelectrodes – mostly miniaturization. Limitations of such electrodes for NEI include the following: 1.
Limitations of Current CNS Electrodes Neuromodulation of brain tissue (deep brain stimulation – DBS) with macro- or microelectrodes functionally appears to be a form of ‘‘reversible ablation,’’ i.e., the effects on the CNS function as a whole is quite similar to irreversibly ablating a similar volume of tissue by, e.g., thermocoagulation. Our understanding of the mechanisms of action of DBS is increasing rapidly, with computer modeling being a powerful technique [16]. Even a microelectrode, being tens of microns or larger in diameter, does not interface intimately with individual neurons in the way that axons and dendrites interact. The 2-D nature of the microelectrode – nervous tissue contact surface does not resemble the 3-D nature of axo-dendritic interactions with neuron al cell bodies or processes, not to mention the
41
2.
3.
The electrode surface is 2-D, interacting with a 3-D complex of neurons, neuronal processes, supporting cells such as astrocytes, etc. Impedance is greatly increased in noble metal microelectrodes in comparison with alternatives such as composite materials and carbon nanotube arrays [17,18]. The stiffness (Young’s modulus) of neural tissue is roughly six orders of magnitude less than metal microelectrodes (~2.5 kPa vs. >10 GPa, respectively).
The problems with noble metals for CNS microelectrodes have been discussed recently [19]. Considerable effort has been spent recently in modifying the surface of traditional metal microelectrodes to lower impedance and thus improve electrical charge transfer for neurostimulation, most notably through the use of electrically conductive polymer (ECP) coatings [20]. However,
637
638
41
Impedance recording in central nervous system surgery
. Figure 41-4 Multi-frequency electrical impedance tomography (MFEIT). Thirty-one EEG scalp electrodes are placed using a modified 10–20 EEG scheme, and contact impedance checked using two-terminal impedance measurements between electrode 1 and the other electrodes. Each image data set is made from 258 impedance measurements obtained from combinations of the 31 EEG electrodes
carbon nanotubes (CNTs), appropriately configured, have been shown to possess properties highly desirable for precise recording and stimulation of neural tissue: impedance can be greatly reduced, and capatinance greatly increased, in comparison with platinum or other noble metal electrodes, especially when coated with an ECP such as polypyrrole (PPy) – as measured by electrochemical impedance spectroscopy [17–19]. In addition to the three issues noted above, neurons must ‘‘cohabit’’ successfully with their ‘‘electrical’’ counterparts in the NEI, i.e., long-term toxicity must not be a problem. This has been demonstrated so far for periods of days to weeks [18,19].
Fabrication and Advantages of Carbon Nanotube Electrodes As an example of some of the issues involved in creating a 3-D NEI, we here summarize recently
reported findings with CNT (or carbon nanofiber, CNF) arrays [18]. > Figure 41-5 is a schematic of the steps needed to create a CNF or CNT nanoelectrode array that will support the growth of PC12 cell networks. PC12 (rat pheochromocytoma) cells are neuron-like cells that under appropriate conditions can release dopamine, and thus are of interest in models for movement disorders such as Parkinson’s disease and mood disorders such as severe refractory depression (both Parkinson’s disease and severe depression involve abnormalities of dopamine). Not only does the polypyrrole (PPy) coating greatly improve the electrical properties of the nanoelectrode arrays (as noted above), but it also prevents the clumping of the vertically aligned CNFs when immersed in physiologic solutions. Additional steps required for successful PC12 cell neural network growth include coating the PPy-treated CNF array with a thin (~3 nm) layer of type IV collagen to promote PC12 cell adhesion, and addition of nerve growth
Impedance recording in central nervous system surgery
41
. Figure 41-5 Schematic of the sample preparation procedure of CNF arrays for PC12 cell culture. CNFs collapse into microbundles when dried after submersion in solution. With a thin conformal coating of PPy, the vertical alignment of the CNFs is preserved in biological solutions. Coating with type IV collagen (an ECM protein) improves adhesion of the PC12 cells. NGF facilitates the extension of neurites to form a neural network that interfaces directly with the CNFs
factor (NGF) to the cell culture medium to promote the formation of mature neurites (over 100 microns in length). > Figure 41-6 illustrates the effect of the PPy coating on the neural network of PC12 cells that grow on the CNF array. > Figure 41-6a shows the PC12 cells grown on CNF array without PPy coating – note the abundance of neural nanofibrils (~100 nm diameter, similar to the diameter of the CNFs) bridging between the neurite branches over the clumped CNFs. These neural nanofibrils are likely a local stress response resulting from adhesion to a stiff substrate (the clumped CNFs); similar nanoscale filopodia have been demonstrated extending from human corneal epithelial cells grown on parallel nanoridges of silican substrates [21]. > Figure 41-6b shows the PC12 cells grown on the CNF array with PPy coating – note the absence of the neural nanofibrils, presumably because of the bending of the individual CNFs supporting the PC12 cells and neurites (which is not possible with the clumped CNFs). > Figure 41-6c is a highmagnification image which illustrates the flexibility
of the individual CNFs, as well as occasional penetration of a CNF through the cell membrane. The flexibility of the CNF array can be modified quite readily by changing the outer diameter of the individual CNFs – a property which is likely to be important for NEIs in different regions of the CNS with differing tissue characteristics (e.g., gray vs. white matter). An additional benefit of coating electrode arrays with an ECP such as PPy (beyond improved electric charge transfer and the anti-clumping of CNFs noted above) is that such ECPs allow the controlled release of drugs that were preloaded in the ECP at the time of nanoarray manufacture [22,23]. Examples of drugs likely to be of benefit in neuromodulation include anti-inflammatory drugs such as dexamethasone, NGF, and various neurotrophic factors (e.g., brain derived neurotrophic factor, BDNF) [22]. By fine-tuning the depth of ECP film deposition on the CNF nanoelectrode array, one can optimize the array’s electrochemical characteristics for the specific purpose (recording vs. stimulation vs. drug release) and tissue location (e.g., gray vs. white matter) [18].
639
640
41
Impedance recording in central nervous system surgery
. Figure 41-6 Scanning electron microscopy images of PC12 neurons in a network on a CNF array. (a) CNFs without PPy coating are collapsed into microbundles, and the PC12 neurons demonstrate bridging neural nanofibrils, likely a stress response. (b) PPy-coated CNFs remain vertically aligned, supporting the PC12 neural network so that bridging neural nanofibrils are not seen. (c) PPy-coated CNFs are sufficiently flexible to bend with the weight of the cell body
Micro- and Nano-level Neurotransmitter Monitoring/ Modulating: CNS Electrochemistry Neurons in the CNS communicate with each other through a combination of electrical and chemical means. Brain electrical activity and electrical characteristics such as impedance have been studied with increasing precision over many
decades, but the study of neurotransmitters in vivo is much more recent. Microdialysis has been used to determine neurotransmitter levels in specific regions of the brain, but due to its large size (typically >100 mm) and slow response time (typically >1 min) the microdialysis probe is of limited value for study dynamic brain activity. The most important technique for in vivo neurotransmitter monitoring currently is fast-scan
Impedance recording in central nervous system surgery
cyclic voltammetry (FSCV). Although first described about 20 years ago, it was not until a decade later that FSCV was used to follow changes in dopamine levels during behavior [24]. Although not small enough to permit neurotransmitter monitoring within the synaptic cleft (~100 nm), the 5 mm diameter carbon-fiber microelectrodes used for FSCV make very localized monitoring of neurotransmitters, e.g., dopamine, possible. FSCV is illustrated in > Figure 41-7: dopamine is rapidly oxidized to dopamine-oquinone, then reduced back to dopamine by ramping the electrode potential from 0.4 to +1.0 V, and back again, at 300 V/s – typically at 10 times per second (10 Hz) repetition rate. This technique detects subsecond dopamine concentration changes [25]. By changing the
41
parameters to 0.6 to +1.4 V and 450 V/s, sensitivity can be increased 10-fold, but the response time lengthens by >0.5 s. As shown in > Figure 41-7, the cyclic voltammogram for dopamine is found by subtracting the large background charging current from the current obtained when dopamine is present. FSCV can be quite accurate at detecting changes in dopamine levels with subsecond resolution, but the subtraction process makes basal level determination questionable [25]. A recent report documents dopamine release within the nucleus accumbens of the rat in reward-seeking behavior that is spatially and temporally heterogeneous [26]. CNF nanoelectrode arrays can improve significantly on the neurochemistry monitoring with carbon-fiber microelectrodes. An individual
. Figure 41-7 Fast-Scan Cyclic Voltammetry. (a) Electrode potential is scanned from 0.4 to +1.0 V and back every 100 ms at 300 V/s. (b) Dopamine is oxidized to dopamine-o-quinone and then reduced back to dopamine. (c) Black line: large background charging current of the electrode. Red line: small changes in the presence of dopamine. (d) Subtracting the black line form the red line in (c) produces the cyclic voltammogram “fingerprint’’ for dopamine. (e) The current at the oxidation potential is converted to concentration (using an in vitro calibration value) which is plotted versus time to monitor dopamine concentration changes after brief electrical stimulation (4 pulses at 100 Hz, indicated by hash marks) Reprinted with permission from Analytical Chemistry 2003, 75, 414A–417A. Copyright 2003 American Chemical Society
641
642
41
Impedance recording in central nervous system surgery
CNF tip is <100 nm diameter (less than the width of the synaptic cleft), whereas the carbon-fiber microelectrode is 5 mm diameter. Because the number of recording electrodes is greatly increased with CNF nanoarrays versus a carbon-fiber microelectrode or a single CNF nanoelectrode (say N times as great), the resulting signal amplitude is magnified by a factor of N. Preliminary FSCV studies with CNF nanoelectrode arrays have shown one to two orders of magnitude increase in both sensitivity (signal to noise ratio), spatial, and temporal resolution in dopamine detection – in comparison with carbon-fiber microelectrodes [27,28]. Since tens of thousands of CNF nanoelectrodes are connected to the same microelectrode pad of 100 mm diameter, there is a high probability that some of them are very close to the synaptic cleft where a high concentration of neurotransmitter is present, it is possible to measure the neurotransmitter release directly at the synapse (rather than measure the small portion that is diffused into the extracellular space and diluted). An array-in-array design is of particular interest for the development along this direction [28,29]. > Figure 41-8 demonstrates the measurements of the electrochemically active neurotransmitter dopamine in phosphate buffered saline with an inlaid CNF nanoelectrode array. The detection limit can readily reach down to 60 nM. With further optimization, this technique may provide the capability to detect dopamine at the 10 nM level (Nguyen-Vu TD, Mandikian D, Cassell AM, Andrews RJ, Meyyappan M, Kawagoi K, and Li J – unpublished results 2008).
Nanoelectrode Arrays for Neuromonitoring and Neuromodulation As described above, and detailed in the references in the previous section, nanoelectrode arrays are already a laboratory research tool for monitoring and modulating both neurotransmitter and electrical
. Figure 41-8 Dopamine detection (in phosphate buffered saline solution) by FSCV using CNF nanoarrays. (a) A small dopamine peak is detectable at 64 nM. (b) Calibration curve of measured signal versus dopamine concentration over 5 orders of magnitude (13 nM–1.0 mM)
activity in neuronal systems. Much work remains, however, on optimizing the fabrication details, miniaturization of the connectors between the nanoarrays and the external recording/stimulating device, and biocompatibility issues. Although trials of nanoelectrodes arrays in small animal models (e.g., rodent models of epilepsy or Parkinson’s disease) may be expected within the next year or two, it will likely be 5 years or more before nanoelectrode arrays are in clinical practice in humans. An important step will be comparing nanoelectrode arrays with the current ‘‘gold standard’’ DBS macroelectrode array in a large animal model (e.g., a primate model of Parkinson’s disease). > Figure 41-9 illustrates a configuration of CNF
Impedance recording in central nervous system surgery
41
. Figure 41-9 Macro-size nanoarray for DBS (comparison with macroelectrode). (a) Schematic of 1.5 mm diameter electrode. (b) 3 3 CNF arrays, each with independent lead. (c) Uninsulated CNF array for electrical stimulation/recording (large contact area). (d) Insulated (e.g., silicon dioxide) CNF array for electrochemical (neurotransmitter) recording. Scale bars: (b) 200 mm; (c) 1 mm; (d) 2 mm
nanoarrays for both electrical and chemical (neurotransmitter) recording and stimulating that can be compared with the current DBS macroelectrode. The CNF nanoarray would be much larger than necessary, but would allow for physical similarity with the current DBS macroelectrode: the left hemisphere might be implanted with the standard DBS macroelectrode, the right with the CNF nanoarray (of similar size/configuration). Such a ‘‘head-tohead’’ comparison would not take advantage of most of the benefits of the CNF nanoelectrode array (noted above), but would permit baseline comparisons of both baseline recording sensitivity and energy requirements for stimulation.
References 1. Butson CR, Maks CB, McIntyre CC. Sources and effects of electrode impedance during deep brain stimulation. Clin Neurophysiol 2006;117:447-54.
2. Moss J, Ryder T, Aziz TZ, Graeber MB, Bain PG. Electron microscopy of tissue adherent to explanted electrodes in dystoniz and Parkinson’s disease. Brain 2004;127:2755-63. 3. Hoeber R. Eine methode die elektrische leitfaehigkeit im inner von zellen zu messen. Arch Ges Physiol 1910;133:237-59. 4. Van Harreveld A, Ochs S. Cerebral impedance changes after circulatory arrest. Am J Physiol 1956;187:180-92. 5. Schwan HP. The practical success of impedance techniquest from an historical perspective. In: Riu PJ, Rosell J, Bragos R, Casa O, editors. Electrical bioimpedance methods: applications to medicine and biotechnology. Ann NY Acad Sci 1999;873:1-12. 6. Grant FC. Localization of brain tumors by determination of the electrical resistance of the growth. JAMA 1923;81:2169-71. 7. Organ LW, Tasker RR, Moody NF. Brain tumor localization using an electrical impedance technique. J Neurosurg 1968;28:35-44. 8. Bullard DE, Makachinas TT. Measurement of tissue impedance in conjunction with computed tomographyguided stereotaxic biopsies. J Neurol Neurosurg Psychiatry 1987;50:397-401. 9. Rajshekhar V. Continuous impedance monitoring during CT-guided stereotactic surgery: relative value in cystic and solid lesions. Br J Neurosurg 1992;6:439-44.
643
644
41
Impedance recording in central nervous system surgery
10. Gildenberg PL, Zanes C, Flitter M, et al. Impedance measuring device for detection of penetration of the spinal cord in anterior percutaneous cervical cordotomy. J Neurosurg 1969;30:87-92. 11. Taren JA, Davis R, Crosby EC. Target physiologic corroboration in stereotaxic cervical cordotomy. J Neurosurg 1969;30:569-84. 12. Bayford RH. Bioimpedance tomography (electrical impedance tomography). Ann Rev Biomed Eng 2006;8:63-91. 13. McEwan A, Cusick G, Holder DS. A review of errors in multi-frequency EIT instrumentation. Physiol Meas 2007;28:S197-215. 14. Romsauerova A, McEwan A, Horesh L, Yerworth R, Bayford RH, Holder DS. Multi-frequency electrical impedance tomography (EIT) of the adult head: initial findings in brain tumours, arteriovenous malformations and chronic stroke, development of an analysis method and calibration. Physiol Meas 2006;27:S147-61. 15. Oh SH, Lee BI, Woo EJ, et al. Electrical conductivity images of biological tissue phantoms in MREIT. Physiol Meas 2005;26:S279-88. 16. McIntyre CC, Miocinovic S, Butson CR. Computational analysis of deep brain stimulation. Expert Rev Med Devices 2007;4:615-22. 17. Nguyen-Vu TD, Chen H, Cassell AM, Andrews R, Meyyappan M, Li J. Vertically aligned carbon nanofiber arrays: an advance toward electrical-neural interfaces. Small 2006;2:89-94. 18. Nguyen-Vu TD, Chen H, Cassell AM, Andrews RJ, Meyyappan M, Li J. Vertically aligned carbon nanofiber architecture as a multifunctional 3-D neural electrical interface. IEEE Trans Biomed Eng 2007;54:1121-9. 19. Wang K, Fishman HA, Dai H, Harris JS. Neural stimulation with a carbon nanotube microelectrode array. Nano Lett 2006;6:2043-8.
20. Llinas RR, Walton KD, Nakao M. Hunter I, Anquetil PA. Neuro - vascular central neruous recording/stimulating system: Using nonotechnology probes. J Nanopart Res 2005;7:111-127. 21. Karuri NW, Liliensiek S, Teixeira AI. Biological length scale topography enhances cell-substratum adhesion of human corneal epithelial cells. J Cell Sci 2004;117:3153-64. 22. Wadhwa R, Lagenaur CF, Cui XT. Electrochemically controlled release of dexamethasone from conducting polymer polypyrrole coated electrode. J Control Release 2005;110:531-41. 23. Abidian M, Kim DH, Martin DC. Conducting polymer nanotubes for controlled drug release. Adv Mater 2006;18:405-9. 24. Rebec GV, Christensen JR, Guerra C, Bardo MT. Regional and temporal differences in real-time dopamine efflux in the nucleus accumbens during free-choice novelty. Brain Res 1997;776:61-7. 25. Venton BJ, Wightman RM. Psychoanalytical electrochemistry: dopamine and behavior. Anal Chem 2003;75:414A-421A. 26. Wightman RM, Heien MLAV, Wassum KM, et al. Dopamine release is heterogeneous within microenvironments of the rat nucleus accumbens. Eur J Neurosci 2007;26:2046-54. 27. Li J, Ng HT, Cassell A, et al. Carbon nanotube nanoelectrode array for ultrasensitive DNA detection. Nano Lett 2003;3:597-602. 28. Li J, Koehne JE, Cassell AM, et al. Inlaid multi-walled carbon nanotube nanoelectrode arrays for electroanalysis. Electroanalysis 2005;17:15-27. 29. Koehne JE, Chen H, Cassell AM, et al. Miniaturized multiplex label-free electronic chip for rapid nucleic acid analysis based on carbon nanotube nanoelectrode arrays. Clin Chem 2004;50:1886-93.
54 Intraoperative Image Guidance in Skull Base Tumors D. Omahen . F. Doglietto . D. Mukherjee . F. Gentili
Introduction With the advent of modern brain imaging techniques, neurosurgeons came into possession of powerful diagnostic tools. Intraoperative image guidance systems (IGS) attempt to apply this technology to therapeutic surgical interventions, creating the field of image-guided therapy (IGT). It is hoped that widespread utilization of these techniques will translate into improved patient outcomes. Examples of ways in which preoperative or realtime image guidance can aid the surgeon abound. Image guidance systems can help to optimally place skin incisions and bone flaps, and can help minimize their size. Their ability to precisely localize vital anatomic structures can help to safely guide the approach to surgical targets. The extent of lesions can also be gauged, which is invaluable in the resection of lesions with grossly indistinct borders. Image guidance systems fall into two broad categories: those that use images acquired preoperatively, and those in which imaging is updated during the procedure. Each system has its own advantages and disadvantages. They may utilize many different imaging modalities including fluoroscopy, ultrasound, computerized tomography (CT), magnetic resonance imaging (MRI), positron emission tomography (PET), angiography, single photon emission CT (SPECT) [1,2]. Images obtained using these varied techniques may be fused together, taking advantage of the slightly different information captured by these disparate methods. Additionally, some systems allow these images to be superimposed upon real time images obtained with the operating microscope or endoscope [3]. #
Springer-Verlag Berlin/Heidelberg 2009
In this chapter we will survey common methods in current use, stressing their advantages and disadvantages. An overview of specific applications in skull base surgery will then be provided. Technology is changing at a rapid pace, and it is anticipated that many new advances and improvements will be forthcoming in short order.
Methods Based on Preoperative Imaging At the most basic level, the majority of modern neurosurgery is guided by some form of imaging. Preoperative imaging is studied by the neurosurgeon and the information gleaned is used in formulating a surgical plan. An advantage of using preoperative imaging is that optimal image quality is possible under controlled conditions [1]. A giant leap forward was provided by the advent of stereotaxy. The Greek words for ‘‘three dimensional’’ (stereo) and ‘‘arrangement’’ (taxis) were combined to create the term ‘‘stereotaxic’’. Use of the Latin term tactus (‘‘to touch’’) gave rise to the synonymous term stereotactic [4]. The Russian anatomist Zernov [5] was the first to use a rudimentary frame, but it was the team of Horsley and Clarke [6] who created the first true stereotactic system designed to create precise lesions in the cerebellum of monkeys. The first stereotactic surgery performed on human beings was reported by Spiegel and Wycis [7], who used X rays and ventriculograms to image the brain. The use of imaging as opposed
816
54
Intraoperative image guidance in skull base tumors
to atlas-derived coordinates ushered in the era of image-guided surgery. Localization procedures fall into two broad categories [4]. The first is a linked system, in which angle-sensing devices are employed to determine the position of a point localized by a mechanical linkage arm. This is the method used by frame-based systems. The second method, that of unlinked systems, uses devices such as cameras, magnetic sensors, or microphones which receive positional information from a pointer, which is not contiguous with the pointing device.
Frame-based, or Linked Systems Frame-based systems utilize a mechanical arm, which is affixed to the patient’s head prior to the procedure. Most arms are moved passively by the surgeon, although robotic systems with active control of arm movement have also been devised [8]. This process involves the localization of target tissue and anatomical structures relative to each other in space utilizing imaging technology. Pioneering stereotactic techniques utilized frames, which were rigidly fixed to the patient’s head prior to imaging. From the images thus obtained, the position of a target relative to the frame could be fixed in space using a Cartesian or polar coordinate system [4]. Since the relationship between the head and frame was rigidly maintained, this allowed for vary precise spatial accuracy, on the order of 1–3 mm of error [9,10]. Such systems are still widely used for performing biopsies, for electrode insertion or lesion localization in functional neurosurgical procedures, and in targeting for radiosurgical procedures. Examples of well-known stereotactic frames which use interlocking arcs, include the Brown-RobertsWells (BRW) frame [11], and the target-centered Cosman-Roberts-Wells (CRW) frame [4]. Advantages of frame-based systems include a proven track record, reliability and documented accuracy.
Disadvantages include the fact that frame application can be uncomfortable and time consuming [4]. Since the frame covers most of the patient’s skull and the frame must be left in place to be used, applications in skull base surgery are limited.
Frameless Stereotaxy In an effort to overcome the limitations of framebased stereotactic systems, frameless approaches have been devised. They are categorized as unlinked systems, since position sensing devices are not in physical continuity with pointing devices. They are easier to use, provide full access to the patient’s head, and more comfortable for the patient. Studies have shown a similar degree of reliability and spatial accuracy as frame-based systems [12,13], especially using a ‘‘probe’s eye view’’ for targeting [12] (> Figure 54-2). The basic principles of frameless stereotaxy are based on those of frame-based stereotaxy. Simplistically, an image of the patient is taken, sometimes with fiducial markers in place to define set points. The patient’s head is placed in a pin-based surgical head holder, which is affixed to the operating table. A reference array, which defines points on a spatial plane in physical space, is used. It is paramount that this array is held in a fixed spatial relationship to the patient’s head. Newer systems do not necessitate a rigid head fixation and are based on systems which are able to keep a fixed spatial relationship with the head even when this is moved. Several methods have been employed for localization. Some groups have experimented with magnetic field sensor-based localization [14] and initial attempts were abandoned due to problems with magnetic interference; newer systems are now becoming available with markedly increased results [15]: theoretically problems hich are present in optical based system, such as the line-of-sight problem and missing
Intraoperative image guidance in skull base tumors
tracking of the tip of flexible instruments, should be solved by the new magnetic field sensor-based tracking systems [15]. An early system used a spark gap-generated ultrasonic emission, which was used by an array of detecting microphones to triangulate the position of the emitter [16]. This method suffered from susceptibility to echoes and non-linear variations with changes in temperature [4], and has largely been supplanted by optical methods of localization. Most optical systems use two infrared cameras, which are set a fixed distance apart and used to triangulate the distance to markers on the reference array, allowing the position of the array relative to the cameras to be calculated by the computer. To use IGS, two main processes must be carried out: Segmentation
The first step is known as segmentation [17]. In this step the tissue of anatomic or pathological interest is delineated and selected. This is often done on a computer workstation prior to initia-
54
tion of the procedure (> Figure 54-1). This step is not essential if the lesion is easily recognizable in the imaging that is used and the simple images in the three planes (sagittal, axial, coronal) can be used (> Figure 54-2). Registration
A registration process must be performed to relate coordinates in physical space to coordinates in the virtual image space [4] (i.e., matching points on the patient with corresponding points in the imaging data set(s)) [18]. There are two main methods of carrying this out. The first, known as point-based transformation, uses intrinsic anatomic landmarks, or extrinsically applied fiducial markers (> Figure 54-1). The positions of the markers are identified on the preoperative imaging. The pointing device is then used to identify these points on the patient using the infraredsensing camera system. Calculations show that there is an inverse relationship between the targeting error and the square root of the number of fiducials. Thus, to double the accuracy of such
. Figure 54-1 Frameless stereotactic optic navigation system. (a) The computer work station and video monitor are visible and are together with the optic reading system (arrow). (b) The head is rigidly fixed. The reference frame (encircled in blue) is attached to the head pin fixation system. The fiducials are visible on the patient’s forehead (StealthStation TREON – Medtronic)
817
818
54
Intraoperative image guidance in skull base tumors
. Figure 54-2 Neuronavigation and repeat surgery. Neuronavigation during an extended transsphenoidal approach for a recurrent suprasellar craniopharyngioma (segmentation has not been performed as the lesion is easily recognized in the MRI). In this case neuronavigation was essential to determine the midline and optimize the sellar and suprasellar opening in a patient with an intercarotid distance in the suprasellar area of only 10 mm (coronal post-contrast MRI in the upper left corner). In the lower right corner the so called ‘‘trajectory view’’ shows to the surgeon the direction of the probe, which is depicted in blue. The coronal, sagittal and axial images show where the tip of the pointer has been positioned by the surgeon (at the level of the sella, in the midline) (StealthStation TREON – Medtronic)
a system, four times as many fiducial points must be utilized [4]. A second method, known as surface matching, uses points obtained by scanning over the patient’s facial features. A surface rendering of the patient’s face and scalp is created from the imaging data set by the computer. The set of points acquired during scanning is then fitted to the corresponding set of points obtained from the image data surface rendering using a least-squares-based transformation [4]. The location of the surgical target can then be identified from the area of interest selected from the imaging set (via the process of segmentation). The position of the target can be measured relative to the fiducial points and/or patient’s fixed
anatomical points identified on the imaging set. Since the target position is known relative to the fiducial points (on the imaging data set), and the location of the fiducial points are known relative to the reference array (in physical space), it is a straightforward calculation to determine the position of the target relative to the reference frame. Basic Components of a Frameless IGS
There are currently many commercial frameless tereotaxy systems available, however they share in common the same basic components (> Figure 54-1). Computer work station and video monitor
Most systems employ a trolley-mounted com-
Intraoperative image guidance in skull base tumors
54
Reference frame An array containing a variable
puter system onto which the image set is loaded prior to the procedure. In the process of segmentation anatomic areas with pathological or physiological importance are selected [1]. After the registration process, information from the camera system is compared with information contained in the image data set, and data is displayed interactively during the course of the procedure.
number of fixed points marked by reflective spheres is used to create a reference plane. Mathematically, three points are needed to define a plane, but most systems use addition points to increase accuracy and reliability. The position of this reference array is maintained in a fixed spatial relationship to the patient’s head, usually by attaching it to the Mayfield or Sugita head clamp.
Several different localization systems have been devised, as outlined above. At present, most systems use optical methods. An infrared beam is produced by light-emitting diodes (LEDs). After it is emitted, it bounces off reflective markers and is picked up by two infrared-sensing CCD cameras located a fixed distance from each other. An alternative method is to place LEDs on the reference frames and probes [19]. Infrared radiation emitted from the probe and reference frame is then picked up by the cameras and used to triangulate their positions in physical space.
A variety of pointing devices are available. Commonly, a blunt tipped pointer with an attached array of reflective spheres is used. The array can be detected by the camera and is located a predefined distance from the tip of the pointer. Hence, the position of the pointer is known relative to its array. Additionally, most systems are equipped with mobile arrays, which can be attached to surgical instruments. Through a registration process in which the position of the instrument’s tip is measured in relation to the attached array, almost any instrument can be used as a pointing device.
Camera system
Many systems use small markers, which are affixed to the skin of the patient’s face and scalp prior to imaging. These serve to provide readily identifiable points to aid in the registration process. Fiducials, which screw into bone, have also been developed to diminish the possibility of them changing position in the time between imaging and registration; they are used much less frequently, due to their invasiveness.
Pointing device
Skin fiducials
Infrared radiation emitted from the localization system bounces off reflective markers placed on components of the localizing system and is detected by the camera system. The markers are usually passive spherical glass beads impregnated with aluminium [4]. These beads can be gas sterilized, and can be fitted into sterile adaptors on pointing devices and reference arrays.
Reflective markers
Sources of Registration and Navigation Errors There are a multitude of potential errors that can degrade navigational accuracy [4]. Scanner slice thickness is one variable that can be manipulated. Voxel size places a limit on the ultimate accuracy obtainable, as localization is only as good as the data set used. The position of skin fiducials can shift considerable, especially when placed on mobile areas of skin. Bone-anchored fiducials help overcome this problem, but are invasive [20]. Data sets from CT scans may suffer from error due to shear, which can be compensated for if the gantry angle is known. MRI scans suffer from image distortions due to magnetic field inhomogeneity. Several methods designed to correct for such problems exist, and aim to create a more accurate image known as the
819
820
54
Intraoperative image guidance in skull base tumors
rectified image. Unfortunately, automated registration algorithms can find a transformation solution which meets criteria for a global acceptable fit, but which has wide degrees of local mismatch error. Another potential source of error is inadvertent movement of the patient’s head relative to the reference array.
Real-time Intraoperative Imaging A major disadvantage of using imaging obtained preoperatively is that the position of intracranial structures can shift during the course of an operation [21]. Removing a bone flap, draining cerebrospinal fluid, administering mannitol, and manipulating or resecting tissue can change the relationship between cranial structures and reference points, resulting in erroneous localization. Even prior to tumor resection, surface deformation greater than one centimeter has been documented after dural opening in over half of patients [22]. Fortunately for skull base surgeons, the rigid bony base of the skull is less susceptible to this problem of brain shift [23]. In an attempt to overcome this limitation, methods of updating the imaging set during the course of the procedure have been devised, producing real-time or near real-time images. Intraoperative imaging provides not only surgical guidance, but also an ability to monitor the progress of an operation through repeated updating of images. The goal is 3D imaging in real time — a concept which has been labeled ‘‘4D-imaging’’ [1]. Unlike diagnostic imaging, where a greater emphasis is placed on specificity, for therapeutic applications, sensitivity is of primary importance [1]. Although intraoperative MRI has been the most widely publicized, several other imaging modalities have also been used. The main advantage of intraoperative imaging is the ability to update images in essentially real time. Since tissue shift and deformation of anatomic and pathologic structures during the
course of surgery is inevitable, this poses clear advantages for the neurosurgeon. This also allows for early detection and treatment of operative complications, such as hemorrhage, or unsuspected residual tumor [1]. The disadvantages of this approach are mainly technical in nature, related to adapting imaging technology to the constraints imposed by the operating room environment. Ready availability of technical support and input from experienced radiologists is also required.
Ultrasound The use of intraoperative ultrasound in neurosurgery is not recent [24,25]. Its ability to differentiate tissues of different densities has long been recognized. It has been used for tumor biopsy, catheter placement, and cyst aspiration [26]. Experience is required to interpret ultrasound images, and blood and air must be scrupulously cleared from the operative site to improve image quality [27]. Structures such as the falx or ventricles aid in image interpretation. Positioning the patient in such a way that tumor resection cavities can be filled with water aids in visualization. Ultrasound provides an excellent method of localizing major vascular structures [28]. Recently, systems which can update preoperative MRI data sets using intraoperative ultrasound, have been developed [27,29]. An ultrasound probe with a range of frequencies from 5 to 7.5 MHz allows tissue penetration to 120 mm. This allows correction for tissue shift and distortion, which ranges up to 1.5 cm. A recent study has documented that adequate fusion with MRI data can be obtained in 95% of cases [29]. Imaging proved better with metastases and meningiomas compared to gliomas [29]. As visualized with ultrasound, high-grade tumors appeared larger than expected based on CT/MRI imaging [27]. Difficulty differentiating tumor mass from edematous brain appears to be
Intraoperative image guidance in skull base tumors
the cause. In half of cases, the size of the craniotomy flap limited visualization. Over 15% of the time additional resection was carried out as a result of ultrasound imaging [29]. Error as low as 2 mm has been described [30].
Intraoperative MRI One of the most eagerly anticipated advances in image-guided surgery was the development of intraoperative magnetic resonance imaging (iMRI) in the early 1990s [31]. At present, the field has yet to reach its full potential, yet its role in providing immediate quality control appeals to many [32]. The main application to date has been in glioma surgery, and more complete resection has been reported as a result, but an improvement in clinical outcomes has yet to be definitively proven [1,33]. One group has reported that unsuspected residual tumor was discovered in approximately one third of cases in which iMRI was utilized [34]. Others report numbers as high as 57% [35]. Studies on low and high grade gliomas seem to indicate potential for a greater degree of resection using iMRI [35]: this may be dependent upon advances in imaging that allow better differentiation of the interface between invasive glioma tissue and normal brain. Working in the MR environment requires physical adaptation of the operating room. Installation, shielding and maintenance can be costly. Use of these devices prohibits the presence of ferromagnetic instruments in the vicinity due to the magnetic fields utilized. Specially designed head coils must also be used, specific for the iMRI system being used [35]. Using iMRI also increases the length of surgical procedures, although a deleterious effect on patient outcome has not been demonstrated [36]. Design of iMRI requires a tradeoff between field strength and image quality on one hand, and convenience, cost, and the surgeon’s access to the patient on the other hand [1,37].
54
Three main categories of iMRI exist: ‘‘Double donut’’ Configuration
General Electric introduced one of the first designs, the so-called ‘‘double donut’’ configuration, in 1991. This device used two magnets with a 54 cm gap between them to allow surgical access to the patient. The magnetic fields of the two magnets overlap and work in concert, producing an overall field strength of 0.5 Tesla. Advantages of this type of design include almost continual patient access by the surgeon and assistant, and the ability to integrate an operating microscope [1]. Surgeon and patient positioning options are physically limited by this design, but the patient does not leave the imaging space during the procedure facilitating frequent imaging updates [1]. An optical 3D frameless stereotactic system has been designed to be used in the bore of this iMRI [38]. In addition, since the patient remains in the imaging field, the need to reregister guidance systems after each scan is obviated. The design does limit field strength, and suffers from increased magnetic field inhomogeneites [39]. Biplanar Magnet Design
A second iMRI design paradigm, known as biplanar design, allows virtually unlimited access to the patient at the expense of extra time and effort required to update imaging. In this type of iMRI, mobile horizontally or vertically oriented magnets can be moved in or out of the operative field. When the magnet is out of the operative field surgeon access, comfort and freedom of movement is maximized. The cost of such freedom of movement is a relatively lower field strength [39]. Cylindrical Superconducting Magnets
A third approach uses cylindrical superconducting magnets in an effort to maximize field strength and homogeneity [39]. The patient and/or magnet is moved in and out of the field when imaging is required. There are fewer restrictions on ferromagnetic instruments used for the surgery, as long as they are removed prior
821
822
54
Intraoperative image guidance in skull base tumors
to imaging. When not in use for an operation, some designs allow for use of the iMRI unit for diagnostic purposes by movement along a ceiling track-mounted magnet to an imaging suite adjacent to the operating room [40]. The disadvantage is that each time updated imaging is required the magnet must be moved into position, causing delays in the surgery. Complex local RF shielding, such as covering the patient with a copper-impregnated Plexiglas tent may be required [40]. During scanning access to the patient is quite limited [39]. Recent work with higher field magnets has focused on the application of diffusion tensor imaging, MR spectroscopy, MR angiography, and functional MRI to aid in operative neurosurgical procedures [1,32]. Magnets with field strengths up to 3 Tesla are now in use [41]. At the moment, the main drawback to widespread use of higher field iMRI technology is cost [1]. As with frameless stereotactic systems, the position of instruments or pointers in space can be determined, but with real-time updating of tissue deformation [1]. Real-time 3-dimensional reformatting capability is provided by technologies such as the 3D Slicer [42]. Methods of compensating for image degradation caused by radiofrequency pulses from bipolar cautery or surgeon hand motion have been developed [1]. Several authors have reported on the use of temperature-sensitive iMRI protocols (e.g., chemical shift, diffusion imaging) to guide the creation of lesions by thermal ablation, focused ultrasound treatment, cryoablation, or interstitial laser therapy [1]. The temporal resolution of iMRI is currently around one second with contemporary units [1]. The ability to fuse MRI images with other modalities such as CT, PET, SPECT, and MEG is also under intense investigation, but generally requires field strengths of 1 Tesla or higher. Field strength up to 3 T may be required for optimal neurovascular imaging [1]. One day routine use of diffusion imaging may
aid in the early detection of ischemic changes during the course of the operation [1]. Fusion with high field MRI images obtained preoperatively is one way to augment the inferior image quality of low field iMRI without incurring the higher financial costs of upgrading to more powerful magnets. Algorithms are in development, which allow the higher quality preoperative images to be warped to match the tissue deformation resulting from surgical manipulations. This process is known as ‘‘single modality image augmented fusion’’ [1]. Combination of iMRI and surgical endoscopy to guide minimally invasive surgical approaches has sparked some interest as of late. Work on optimal head coil design for iMRI is also in progress [32]. Finally, further integration with new robotic designs may one day allow for real-time image-guided robotic assisted surgery [1].
Fluoroscopy The utility of intraoperative fluoroscopic guidance for spinal surgery applications such as pedicle screw insertion and kyphoplasty is well-documented [19,43–45]. As described below, this was a standard intraoperative imaging technique for pituitary surgery for many years [46]. Intraoperative angiography has expanded the role of fluoroscopy in the operating room. Disadvantages of fluoroscopy include being limited to viewing in a single plane, and radiation exposure [19]. The combination of C-arm fluoroscopy and image guidance has been referred to as ‘‘virtual fluoroscopy’’ [19]. This allows for updating of preoperatively obtained image sets and reduction of radiation exposure.
CT Scanning Limitations of CT-based intraoperative imaging include exposure to ionizing radiation, poor multiplanar imaging capability, and poor tissue detail and resolution compared with MRI [1].
Intraoperative image guidance in skull base tumors
Advantages include superior bony visualization and affordability, portability, and ‘‘near real time’’ imaging [47]. Interest in this technique have led to the development of nascent spinal and CT angiographic [48] intraoperative techniques. Using a mobile cone beam CT unit, Rafferty et al. [47] were able to achieve sub-millimeter spatial resolution with about one tenth the dose of ionizing radiation of a conventional CT scan. Cone beam CT scanners differ from conventional scanners in that they acquire all needed images in a single rotation or less about the patient, without translation of the patient [47]. Three-dimensional reconstruction is performed using a back projection modified Feldkamp algorithm. Imaging radiation dose ranges from 0.5 to 3 mGy [49].
Angiography Intraoperative angiography has been utilized as an adjunct to vascular neurosurgery, allowing assessment of aneurysm or AVM obliteration and vessel patency [50,51]. Using road-mapping techniques, intraoperative angiograms can be used for lesion localization [52].
Application of Image Guidance to Skull Base Surgery Skull base surgeons have been quick to adopt image guidance methods. The bony structures of the skull base do not suffer the same degree of shift and deformation that brain and other soft tissues do. Thus, the drawbacks of guidance systems that use static preoperative imaging are minimized. Image guidance systems provide important information regarding the position and extent of lesions, as well as about the location of vital vascular and nervous structures.
54
Sellar Region and Anterior Skull Base An early application of image guidance to skull base procedures was the adoption of fluoroscopy to aid in transsphenoidal pituitary surgery. Guiot was a pioneer in this application [23]; Jules Hardy [53] introduced its use in North America. Its modern application has been recently described by Jane et al. [72]. Imaging is limited to a single plane at one time, however. Experience with intraoperative CT scanning has been described [54], but not widely adopted. Similarly, ultrasound applications to pituitary surgery have also been limited [18,23,28]: ecographic probes are still too big or have low specificity for application in transsphenoidal surgery; Atkinson et al. [55] described a method to ultrasonically monitor pituitary surgery via a burr hole placed at the coronal suture. On the other hand, doppler ultrasound is an excellent way to locate vascular structures [28], such as the internal carotid artery in the cavernous sinus (> Figure 54-5). The introduction of the pure endoscopic transsphenoidal approach has led to a decrease in the use of fluoroscopy, due mostly to the wider view and the use of a natural pathway, which allows a prompt recognition of anatomical landmarks during surgery (choana, ostium sphenoidale, sella, clivus, carotid protuberances, optic-carotid recess). Some Authors actually recommend the use of the endoscope, in microscopic surgery, as an ‘‘intraoperative navigator’’ due to its wider visualization. Anatomical landmarks are not though always easily recognized, as in the case of a conchal sphenoid sinus: the incomplete pneumatization does not allow the recognition of anatomical landmarks, such as the clivus, the carotid prominence and the optic carotid recess; anatomical landmarks might not be recognizable in repeat surgery [23] (> Figure 54-2): a navigation system is mandatory in these cases.
823
824
54
Intraoperative image guidance in skull base tumors
The use of preoperative images to guide pituitary surgery has indeed gained widespread popularity, even as an adjunct to endoscopic approaches [56]. During the approach to the sellar region, landmarks along the bony skull base are of interest to the surgeon, and this is best imaged using CT scanning. Once the sella is reached however, soft tissue detail provided by MRI imaging has greater utility. Early work on CT–MRI data set fusion shows this is a safe and effective method to maximize navigational safety [20] (> Figure 54-4). The utility of intraoperative stereotacic navigation is also obvious in complex approaches, as in extended endonasal approaches to the suprasellar area (> Figure 54-2), to the anterior cranial fossa (> Figure 54-3) and to the cavernous sinus (> Figure 54-5), as it provides assurance to the
surgeon on the complex anatomy which needs to be mastered for these approaches. iMRI has met with much interest in anterior skull base surgery, as it may have the capability to identify residual tumor not initially identified by the surgeon. Identification of bony landmarks is difficult [23]. Darakchiev et al. [57] provide a comprehensive review of this topic and elaborate on details of their own iMRI setup for pituitary surgery . Others have suggested that MR-angiography may aid in localizing vascular structures such as the carotid or sphenopalatine arteries [20]. Operating in the pre-endoscopic era, Walker and Black have reported that in 7 of 19 patients with macroadenomas additional unsuspected, but ultimately resectable tumor was identified [46]. SPGR (spoiled gradient recalled sequence) and sagittal T1-weighted images were used for
. Figure 54-3 Neuronavigation and extended endoscopic anterior skull base surgery. Extended transsphenoidal approach for an olfactory groove meningioma. The neuronavigation probe is positioned over the anterior skull base, after an ethmoidectomy has been performed, to confirm the exposure of the right anterolateral margin of the olfactory groove meningioma. The three MRI images document the position of the probe in respect to the meningioma; the endoscopic intraoperative picture of the probe position (lower right quadrant) can be transferred to the neuronavigation system and recorded (iNtellect Cranial Navigation System – Stryker)
Intraoperative image guidance in skull base tumors
54
. Figure 54-4 CT–MRI fusion technique. Endoscopic approach for a giant pituitary adenoma. Intraoperative determination of the medial optic carotid recess before the dural opening: both CT (a) and MRI (b) reconstructions can be used for intraoperative image guidance, as well as fusion images that incorporate variable percentages of data images from the two image sets. The position of the probe (asterisk) over the medial OCR is confirmed, both on CT (a) and MRI (b) images (iNtellect Cranial Navigation System – Stryker)
. Figure 54-5 Stereotactic neuronavigation and doppler in cavernous sinus surgery. Endoscopic transsphenoidal surgery for a pituitary adenoma extending in the cavernous sinus. (a) Neuronavigation is used to confirm the position of the probe inside the right cavernous sinus, at the level of the internal carotid artery (ICA – visible in all its intracavernous portion in the sagittal post-contrast MRI reconstructions – upper right quadrant). (iNtellect Cranial Navigation System – Stryker). (b) After further pituitary adenoma removal, a doppler ultrasound probe (asterisk) is positioned over the posterior portion of the right ICA to confirm its exposure in the cavernous sinus. A suction tube is positioned over the anterior portion of the cavernous ICA
navigation purposes. A SPGR sequence takes about 1.5 min to perform [23]. They describe a protocol for intraoperative dynamic contrast imaging to take advantage of the delayed contrast
uptake exhibited by pituitary adenomas. The use of specialized heme-sensitive gradient echo [46] sequences have made differentiation between blood and residual tumor easier [23,58].
825
826
54
Intraoperative image guidance in skull base tumors
Contrast-enhanced scans are less useful than initially anticipated due to blood pooling in the sella [24], and enhancing areas created by electrocautery and surgical manipulation [35,59]. More recently, an iMRI ‘‘road mapping’’ technique to augment endoscopic sinus surgery has been described [60]. In general, it is felt that field strength of at least 0.2 T is required for pituitary surgery [37]. This allows for detecting tumor residual, hemorrhage, and cavernous sinus invasion. A group from the University of Cincinnati has reported on a series of 115 patients undergoing resection of pituitary macroadenomas using 0.3 T iMRI [57]. They utilized 3 mm coronal slices (TR = 450 ms, TE = 20 ms). At the very end of the procedure, these images are followed by standard dose postcontrast scans. iMRI revealed unsuspected, but ultimately resectable tumor in 56% of patients, after what was initially felt to be gross total resection. Intraoperative images were felt to be of superior quality to postoperative images due to the absence of artifact from fat grafting and sellar reconstruction. Operating on acromegalic patients in a 1.5 T magnet, Fahlbusch et al. [61], reported that iMRI increased the rate of total tumor resection and normalization of postoperative serum growth hormone from 33% to 44%, with another 17% experiencing ‘‘near-normalization’’. Not all reports of iMRI for sellar lesions are favorable, however. Nimsky et al. [62] reported disappointing results in a series of 21 procedures for craniopharyngioma using a 0.2 T iMRI unit. Complete resection, as indicated by iMRI did not preclude tumor recurrence. This probably stresses the importance of the biological behavior of the tumor and the intrinsic limitations of surgery.
Surgery of the Temporal Bone CT scanning may be the ideal technique to image the temporal bone intraoperatiavely [47]. Studies of the use of cone-beam CT in temporal bone surgery have demonstrated sub-millimeter accuracy with low radiation dose, as described
above. Special utility was found in demonstrating the degree of bony coverage during skeletonization of the facial nerve [47]. Stereotactic neuronavigation has proved useful in lateral skull base surgery, providing greater assurance in avoiding neurovascular damage during complex skull base procedures [63,64].
Posterior Fossa Surgery Recent papers have proved the utility of stereotactic neuronavigation in the retrosigmoid approach (> Figure 54-6): classic anatomical landmarks (for example: the asterion for localizing the transverse-sigmoid sinus transition (TST) complex) can be individualized to the patient (the asterion was located from 2 mm medial to 7 mm lateral and from 10 mm inferior to 17 mm superior to the TST, respectively) [65]. A recent retrospective study analyzed the impact of image guidance on complication rates (venous sinus injury, venous air embolism, postoperative morbidity caused by venous air embolism) and operation times for the lateral suboccipital craniotomies performed with the patient in the semi-sitting position: a significant increased speed and safety in lateral suboccipital approaches was documented when stereotactic neuronavigation was used [66]. Doppler sonography is useful in confirming the position of the vertebral artery in the farlateral approach. Recent applications of stereotactic neurosurgery in skull base surgery include also the quantitative evaluation of new surgical approaches: with the aid of neuronavigation quantitative data about surgical exposure, depth and trajectory can be obtained, allowing an objective comparison of different surgical approaches [67].
Future Advances and Applications Technological advances will continue to drive progress in the field of intraoperative image
Intraoperative image guidance in skull base tumors
54
. Figure 54-6 Stereotactic neuronavigation in posterior fossa surgery. Definition of the transverse and sigmoid sinus during a right retrosigmoid approach: (a) The head is fixed and the reference frame (encircled in blue) attached to the Mayfield; the pointer is used to define the margins of the transverse and sigmoid sinus. (b) The superior and inferior margins of the sinus is defined as well as the projection of the tumor on the skin. (c) After bone exposure the position of the sinuses is checked on the bony surface before its drilling
guidance. As computing power increases, more automation, less computational time, and more user-friendly formats should be expected. Digital deformation models [68] will allow updating of preoperative data sets using inexpensive, user friendly technology [27,29]. Improvements in coil design and field strength will make iMRI a more powerful technique. In an attempt to overcome problems with non tumor-specific intraoperative contrast enhancement, work is being done on monocrystalline iron oxide nanoparticles (MIONs), which appear to have selective uptake by malignant glioma cells [35]. Coupling of intraoperative imaging advancements with robotic technology may help bring the promise of robotic surgery to fruition. iMRIguided robotic devices have already been developed [69,70]. Finally, work is already in progress to create stereoscopic 3-dimensional images to facilitate visualization of complex pathological and/or anatomical structures [71]. These might one day be integrated with images from an
endoscope or operating microscope in a ‘‘heads up display’’. In conclusion, image-guided surgery has been applied to skull base surgery since Guiot introduced the fluoroscope in transsphenoidal surgery in 1950s. Since then, tremendous technological advances have provided tools that aid skull base neurosurgeons. Further technological development are awaited in image-guided skull base surgery in the near future, possibly making it even safer and more effective.
References 1. Jolesz FA. Future perspectives for intraoperative MRI. Neurosurg Clin N Am 2005;16:201-13. 2. Peters TM, Henri CJ, Munger P, Takahashi AM, Evans AC, Davey B, Oliverier A. Integration of stereoscopic DSA and 3D MRI for image-guided neurosurgery. Comput Med Imaging Graph 1994;18:289-99. 3. Gleason PL, Kikinis R, Altobelli D, et al. Video registration virtual reality for nonlinkage stereotactic surgery. Stereotact Funct Neurosurg 1994;63:139-43. 4. Bucholz RD LA, Marzouk S. Stereotactic and imageguided techniques. In: Kaye AH, Black PMcLB, editor.
827
828
54 5.
6. 7.
8.
9.
10.
11. 12.
13.
14.
15.
16.
17.
18.
19.
Intraoperative image guidance in skull base tumors
Operative neurosurgery. Toronto: Harcourt Publishers; 2000. p. 99-116. Kandel EI, Schavinsky YV. Stereotaxic apparatus and operations in Russia in the 19th century. J Neurosurg 1972;37:407-11. Horsley V CR. The structure and function of the cerebellum examined by a new method. Brain 1908;31:45-124. Spiegel EA, Wycis HT, Thur C. The stereoencephalotome (model III of our stereotaxic apparatus for operations on the human brain). J Neurosurg 1951;8:452-3. Kwoh Y. Special stereotactic techniques: robotic methods applied to stereotactic surgery. In: Heilbrun MP, editor. Stereotactic neurosurgery. Baltimore: Williams & Wilkins; 1988. p. 219-32. Maciunas RJ, Galloway RL, Jr, Latimer JW. The application accuracy of stereotactic frames. Neurosurgery 1994;35:682-94; discussion 94-5. Galloway RL, Jr, Maciunas RJ, Latimer JW. The accuracies of four stereotactic frame systems: an independent assessment. Biomed Instrum Technol 1991;25: 457-60. Brown RA. A stereotactic head frame for use with CT body scanners. Invest Radiol 1979;14:300-4. Quinones-Hinojosa A, Ware ML, Sanai N, McDermott MW. Assessment of image guided accuracy in a skull model: comparison of frameless stereotaxy techniques vs. frame-based localization. J Neurooncol 2006;76:65-70. Dorward NL, Alberti O, Palmer JD, Kitchen ND, Thomas DG. Accuracy of true frameless stereotaxy: in vivo measurement and laboratory phantom studies. Technical note. J Neurosurg 1999;90:160-8. Kato A, Yoshimine T, Hayakawa T, et al. A frameless, armless navigational system for computer-assisted neurosurgery. Technical note. J Neurosurg 1991;74:845-9. Schichor C, Witte J, Scho¨ller K, Tanner P, Uhl E, Goldbrunner R, Tonn JC. Magnetically guided neuronavigation of flexible instruments in shunt placement, transsphenoidal procedures, and craniotomies. Neurosurgery 2008;63 1 Suppl 1:ONS121-7. Roberts DW FE, Strohbehn JW. The sonic digitizing microscope. In: Maciunas RJ, editor. Interactive imageguided neurosurgery. Chicago: AANS Publications Committee; 1993. p. 179-200. Cline HE, Lorensen WE, Kikinis R, Jolesz F. Threedimensional segmentation of MR images of the head using probability and connectivity. J Comput Assist Tomogr 1990;14:1037-45. Ram Z, Shawker TH, Bradford MH, Doppman JL, Oldfield EH. Intraoperative ultrasound-directed resection of pituitary tumors. J Neurosurg 1995;83: 225-30. Nakagawa H, Kamimura M, Uchiyama S, Takahara K, Itsubo T, Miyasaka T. The accuracy and safety of image-guidance system using intraoperative fluoroscopic images: an in vitro feasibility study. J Clin Neurosci 2003;10:226-30.
20. Sindwani R, Bucholz RD. The next generation of navigational technology. Otolaryngol Clin North Am 2005;38: 551-62. 21. Kelly PJ, Kall BA, Goerss SJ. Results of computed tomography-based computer-assisted stereotactic resection of metastatic intracranial tumors. Neurosurgery 1988;22: 7-17. 22. Hill DL, Maurer CR, Jr, Maciunas RJ, Barwise JA, Fitzpatrick JM, Wang MY. Measurement of intraoperative brain surface deformation under a craniotomy. Neurosurgery 1998;43:514-26; discussion 27-8. 23. Walker DG, Kaye AH. Image guidance and transsphenoidal surgery: past, present and future. J Clin Neurosci 2003;10:289-92. 24. Hwang KP, Lim J, Wendt M, Merkle E, Lewin JS, Duerk JL. Improved device definition in interventional magnetic resonance imaging using a rotated stripes keyhole acquisition. Magn Reson Med 1999;42:554-60. 25. Chandler WF, Knake JE, McGillicuddy JE, Lillehei KO, Silver TM. Intraoperative use of real-time ultrasonography in neurosurgery. J Neurosurg 1982;57:157-63. 26. Sutcliffe JC. The value of intraoperative ultrasound in neurosurgery. Br J Neurosurg 1991;5:169-78. 27. Miller D, Heinze S, Tirakotai W, et al. Is the image guidance of ultrasonography beneficial for neurosurgical routine? Surg Neurol 2007;67:579-87; discussion 87-8. 28. Arita K, Kurisu K, Tominaga A, et al. Trans-sellar color Doppler ultrasonography during transsphenoidal surgery. Neurosurgery 1998;42:81-5; discussion 6. 29. LIndner D, Trantakis TC, Arnold S, Schmitgen A, Schneider J, Meixensberger J. Neuronavigation based on intraoperative 3D-ultrasound during tumor resection. International Congress Series 2005;1281:81-82. 30. Comeau RM, Sadikot AF, Fenster A, Peters TM. Intraoperative ultrasound for guidance and tissue shift correction in image-guided neurosurgery. Med Phys 2000;27: 787-800. 31. Schenck JF, Jolesz FA, Roemer PB, et al. Superconducting open-configuration MR imaging system for image-guided therapy. Radiology 1995;195:805-14. 32. Fahlbusch R, Nimsky C. Intraoperative MRI developments. Neurosurg Clin N Am 2005;16:xi-xiii. 33. Schneider JP, Schulz T, Schmidt F, et al. Gross-total surgery of supratentorial low-grade gliomas under intraoperative MR guidance. AJNR Am J Neuroradiol 2001;22: 89-98. 34. Black PM, Alexander E, 3rd, Martin C, et al. Craniotomy for tumor treatment in an intraoperative magnetic resonance imaging unit. Neurosurgery 1999;45:423-31; discussion 31-3. 35. Yrjana SK, Tuominen J, Koivukangas J. Intraoperative magnetic resonance imaging in neurosurgery. Acta Radiol 2007;48:540-9. 36. Schulder M, Carmel PW. Intraoperative magnetic resonance imaging: impact on brain tumor surgery. Cancer Control 2003;10:115-24.
Intraoperative image guidance in skull base tumors
37. Lewin JS, Metzger A, Selman WR. Intraoperative magnetic resonance image guidance in neurosurgery. J Magn Reson Imaging 2000;12:512-24. 38. Moriarty TM, Quinones-Hinojosa A, Larson PS, et al. Frameless stereotactic neurosurgery using intraoperative magnetic resonance imaging: stereotactic brain biopsy. Neurosurgery 2000;47:1138-45; discussion 45-6. 39. Hinks RS, Bronskill MJ, Kucharczyk W, Bernstein M, Collick BD, Henkelman RM. MR systems for imageguided therapy. J Magn Reson Imaging 1998;8:19-25. 40. Sutherland GR, KaibaraT, LouwD, HoultDI, TomanekB, Saunders J. A mobile high-field magnetic resonance system for neurosurgery. J Neurosurg 1999;91:804-13. 41. Hall WA, Galicich W, Bergman T, Truwit CL. 3-Tesla intraoperative MR imaging for neurosurgery. J Neurooncol 2006;77:297-303. 42. Nabavi A GD, Kacher DF, Talos IF, Wells WM, Kikinis R, et al. Surgical navigation in the open MRI. Acta Neurochir Suppl (Wein) 2003;85:121-5. 43. Acosta FL, Jr, Thompson TL, Campbell S, Weinstein PR, Ames CP. Use of intraoperative isocentric C-arm 3D fluoroscopy for sextant percutaneous pedicle screw placement: case report and review of the literature. Spine J 2005;5:339-43. 44. Rajasekaran S, Vidyadhara S, Shetty AP. Iso-C3D fluoroscopy-based navigation in direct pedicle screw fixation of Hangman fracture: a case report. J Spinal Disord Tech 2007;20:616-9. 45. Villavicencio AT, Burneikiene S, Bulsara KR, Thramann JJ. Intraoperative three-dimensional fluoroscopy-based computerized tomography guidance for percutaneous kyphoplasty. Neurosurg Focus 2005;18:e3. 46. Walker DG BP. Use of intraoperative MRI in pituitary surgery. Operative techniques in neurosurgery 2002;5: 231-8. 47. Rafferty MA, Siewerdsen JH, Chan Y, et al. Intraoperative cone-beam CT for guidance of temporal bone surgery. Otolaryngol Head Neck Surg 2006;134:801-8. 48. Chibbaro S TL. Image-guided microneurosurgical management of vascular lesions using navigated computed tomography angiography. An advanced IGS technology application. The International Journal of Medical Robotics and Computer Assisted Surgery: MRCAS 2006;2. 49. Rafferty MA, Siewerdsen JH, Chan Y, et al. Investigation of C-arm cone-beam CT-guided surgery of the frontal recess. Laryngoscope 2005;115:2138-43. 50. Klopfenstein JD, Spetzler RF, Kim LJ, et al. Comparison of routine and selective use of intraoperative angiography during aneurysm surgery: a prospective assessment. J Neurosurg 2004;100:230-5. 51. Schievink WI, Vishteh AG, McDougall CG, Spetzler RF. Intraoperative spinal angiography. J Neurosurg 1999;90: 48-51. 52. Ayad M, Ulm AJ, Yao T, Eskioglu E, Mericle RA. Realtime image guidance for open vascular neurosurgery
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
54
using digital angiographic roadmapping. Neurosurgery 2007;61:55-61; discussion 2. Hardy J, Wigser SM. Trans-sphenoidal surgery of pituitary fossa tumors with televised radiofluoroscopic control. J Neurosurg 1965;23:612-9. Okudera H, Takemae T, Kobayashi S. Intraoperative computed tomographic scanning during transsphenoidal surgery: technical note. Neurosurgery 1993;32:1041-3. Atkinson JL, Kasperbauer JL, James EM, Lane JI, Nippoldt TB. Transcranial-transdural real-time ultrasonography during transsphenoidal resection of a large pituitary tumor. Case report. J Neurosurg 2000;93:129-31. Gong J, Mohr G, Vezina JL. Endoscopic pituitary surgery with and without image guidance: an experimental comparison. Surg Neurol 2007;67:572-8; discussion 8. Darakchiev BJ, Tew JM, Jr, Bohinski RJ, Warnick RE. Adaptation of a standard low-field (0.3-T) system to the operating room: focus on pituitary adenomas. Neurosurg Clin N Am 2005;16:155-64. Pergolizzi RS, Jr, Nabavi A, Schwartz RB, et al. Intraoperative MR guidance during trans-sphenoidal pituitary resection: preliminary results. J Magn Reson Imaging 2001;13:136-41. Knauth M, Aras N, Wirtz CR, Dorfler A, Engelhorn T, Sartor K. Surgically induced intracranial contrast enhancement: potential source of diagnostic error in intraoperative MR imaging. AJNR Am J Neuroradiol 1999;20:1547-53. Fried MP, Topulos G, Hsu L, et al. Endoscopic sinus surgery with magnetic resonance imaging guidance: initial patient experience. Otolaryngol Head Neck Surg 1998;119:374-80. Fahlbusch R, Keller B, Ganslandt O, Kreutzer J, Nimsky C. Transsphenoidal surgery in acromegaly investigated by intraoperative high-field magnetic resonance imaging. Eur J Endocrinol 2005;153:239-48. Nimsky C, Ganslandt O, Hofmann B, Fahlbusch R. Limited benefit of intraoperative low-field magnetic resonance imaging in craniopharyngioma surgery. Neurosurgery 2003;53:72-80; discussion 1. Staecker H, O’malley BW, Eisenberg H, Yoder BE. Use of the LandmarXtrade mark surgical navigation system in lateral skull base and temporal bone surgery. Skull Base 2001;11:245-55. Van Havenbergh T, Koekelkoren E, De Ridder D, Van De Heyning P, Verlooy J. Image guided surgery for petrous apex lesions. Acta Neurochir 2003; 145:737-42. Gharabaghi A, Rosahl SK, Feigl GC, Liebig T, Mirzayan JM, Heckl S, Shahidi R, Tatagiba M, Samii M. Imageguided lateral suboccipital approach: part 1-individualized landmarks for surgical planning. Neurosurgery 2008;62 3 Suppl 1:18-22. Gharabaghi A, Rosahl SK, Feigl GC, Safavi-Abbasi S, Mirzayan JM, Heckl S, Shahidi R, Tatagiba M, Samii M.
829
830
54
Intraoperative image guidance in skull base tumors
Image-guided lateral suboccipital approach: part 2-impact on complication rates and operation times. Neurosurgery. 2008;62 3 Suppl 1:24-9. 67. Chatrath P, Nouraei SA, De Cordova J, Patel M, Saleh HA. Endonasal endoscopic approach to the petrous apex: an image-guided quantitative anatomical study. Clin Otolaryngol 2007 Aug;32(4):255-60. 68. Roberts DW, Miga MI, Hartov A, et al. Intraoperatively updated neuroimaging using brain modeling and sparse data. Neurosurgery 1999;45:1199-206; discussion 206-7. 69. Chinzei K, Miller K. Towards MRI guided surgical manipulator. Med Sci Monit 2001;7:153-63.
70. Bumm K, Wurm J, Rachinger J, et al. An automated robotic approach with redundant navigation for minimal invasive extended transsphenoidal skull base surgery. Minim Invasive Neurosurg 2005;48:159-64. 71. Hernes TA, Ommedal S, Lie T, Lindseth F, Lango T, Unsgaard G. Stereoscopic navigation-controlled display of preoperative MRI and intraoperative 3D ultrasound in planning and guidance of neurosurgery: new technology for minimally invasive image-guided surgery approaches. Minim Invasive Neurosurg 2003;46:129-37. 72. Jane JA, Jr, Thapar K, Alden TD, Laws ER, Jr. Fluoroscopic frameless stereotaxy for transsphenoidal surgery. Neurosurgery 2001;48:1302-7; discussion 7-8.
39 MRI in Image Guided Surgery M. Schulder . L. Jarchin
Introduction Magnetic resonance imaging (MRI) has transformed the field of neurosurgery by improving the accuracy of diagnostic imaging, integration into presurgical planning, and moving into the operating room (OR) itself. For surgical applications, most innovations have occurred within the realm of stereotactic neurosurgery. There has been some controversy regarding credit for the discovery that the physical properties of nuclear magnetic resonance could be used to image living tissue. In 2003 the Nobel Prize in Medicine was awarded to Drs. Paul Lauterbur and Peter Mansfield for this work. Dr. Raymond Damadian publicly objected to his exclusion despite his pioneering efforts in the field [1,2]. Be that as it may, by the mid 1980s MRI had become a commercially available technology whose benefits were obvious. These included much better soft tissue imaging of the central nervous system, multiplanar views, and the avoidance of ionizing radiation. Within short order MRI was adapted to image guided surgery and is now the mainstay of this approach in the developed world.
Stereotactic Biopsy with MRI The initial use of MRI for image guided surgery was in frame-based stereotactic brain biopsy. The imaging advantages of MRI over CT were readily apparent, especially improved target definition [3]. Several authors described their experience in patients whose lesions could be seen only on MRI [4,5]. This was particularly so for stereotactic targeting of lesions in the brainstem [6]. It is #
Springer-Verlag Berlin/Heidelberg 2009
worth noting that as early as 1985 Kelly described the integration of MRI in his pioneering volumetric stereotactic system [7]. For most neurosurgeons that sort of technology belonged to the future, and ‘‘conventional’’ frames were used, including the Leksell (Elekta, Norcross GA) [8] and Brown/Cosman-Roberts-Wells (Radionics, Burlington MA) [9] systems (> Figure 39-1). Much effort was expended so CT-compatible frames would be made suitable for MRI. The most obvious difference was the need to avoid ferromagnetic materials in the frame [9]. Besides the frame construct itself, the rods on the localizer box needed to be visible on MRI. Some of the early solutions to this problem, such as filling empty tubes with copper sulfate solution, were quite cumbersome, leading the neurosurgeon to choose CT based guidance when possible. The accuracy of stereotactic frames was well established, and quantified in 1992 by Maciunas et al [10]. Using CT scans with a variety of commercially available frames, they demonstrated that an error of less than 2 mm could be reasonably expected. This error increased with greater imaging slice thickness, scanner gantry angle, and arc reapplication. Early clinical use of MRI made plain the presence of image distortions, from chemical shifts and magnetic susceptibility artifacts. Kondziolka et al compared the accuracy of stereotactic targeting using MRI and CT [11]. They found discrepancies of about 2 mm overall on average, increasing with distance from the center of the patient’s head. They concluded that MRI targeting for frame-based stereotaxy was sufficiently accurate for most procedures (with obvious implications for radiosurgery and for functional procedures).
600
39
MRI in image guided surgery
. Figure 39-1 Cosman-Roberts-Wells frame on phantom base in OR
latter technique eliminating the need to subject patients to prolonged MRI scans with a frame applied. Still, most patients needing biopsies today will have them done using a ‘‘frameless’’ technique or with intraoperative MRI (see below).
Frameless Stereotactic Surgery and MRI
The safety and accuracy of stereotactic biopsy had been established in the 1980s with the emergence of CT-guided techniques [12,13]. There remained, then as now, a risk of nondiagnostic biopsy of about 8% [14]. This remains a risk with frame-based stereotaxis, although it can be lowered to about 3% with the use of intraoperative pathological examination [15]. MRI-guided frame-based stereotactic biopsies remain an important part of operative neurosurgery, but there are limitations with this technology. Conceptually, frames provide registration of single points in space. Multiple points can indeed be targeted but this requires a separate calculation for trajectory and distance, with resetting of the frame coordinates each time. While stereotactic craniotomy was described early in the era of CT guided stereotaxis, the ergonomic difficulty of doing an open surgical procedure with a frame in place have relegated this approach to historical interest [12]. The need to image with the frame in place also prolongs the procedure and requires patient transport through the hospital. The use of frame-based stereotaxis has greatly improved with the incorporation of volumetric navigation [16] and image fusion [17], with the
The benefits of MRI in image guided surgery were perhaps best realized with the advent of frameless stereotaxy, also termed surgical navigation or image guided surgery (IGS) [18]. Kelly described the concept of stereotactic volumetric resection with 3-dimensional computer modeling in 1981 [18]. This method used a modified Todd-Wells stereotactic frame as its platform. The concept of a ‘‘frameless’’ approach was proposed by Roberts et al. who envisioned using the operating microscope as a pointer [19]. This technique was marketed as part of the Zeiss microscope system [20]. However, it proved more practical to uncouple the operating microscope from the IGS technology. Various systems became commercially available in the early to mid 1990s, employing different technologies. The first device, the IGS viewing wand, tracked an articulated arm in space [21]. Due to the bulk and ergonomic interference of this articulated arm, other authors created a digitizing platform with ultrasound [22], magnetic vectors [23,24] or infrared light [25,26]. The latter technology for now has become the one most commonly used in IGS. The most obvious advantage of IGS is that the entire volumes of interest are registered, for instance, the patient’s head and his entire MRI scan. Typically this is done by scanning the patient with fiducial markers applied to the scalp. The image may be obtained as far as in advance of surgery as is convenient. As alluded to above, this uncoupling of imaging and surgery allows for prolonged scanning sessions, in particular anatomical MRI and functional imaging as well.
MRI in image guided surgery
Multiple datasets may be acquired for registration in the OR, and loaded onto the planning workstation before surgery, resulting in further saving of time. These different images can be viewed separately or as a blended view after registration. The other images may include CT [27], PET [28], and MR angiography [29]. Particularly useful has been the incorporation of functional MRI in IGS, allowing for navigation based on more than anatomical information alone [30,31]. McDonald et al. found that fMRI, registered for use in IGS, was as accurate as magnetoencephalography for localizing the primary cortex [32]. This may be combined additionally with tractography derived from diffusion tensor imaging, to truly provide a comprehensive anatomical and functional map of the brain [33] (> Figure 39-2). In the OR, after general anesthesia is induced and the patient’s head secured in a pin fixation clamp, a probe or ‘‘wand’’ with active infrared light emitting diodes (IRLEDs) or passive infrared reflecting spheres is touched to at least four fiducials that have been marked on the preoperative image. When enough spatial information has been gathered to allow for a transformation that matches image and surgical spaces, an accurate registration has been achieved. At this time, unlike with a frame-based procedure,
39
the surgeon has theoretically unlimited entry points, targets, and trajectories from which to choose. In practice, of course, the surgical plan will be constrained by different considerations such as cosmetics, eloquent brain areas, previous surgery, etc. At first, IGS was mainly used to guide craniotomies and tumor resections. The technology was useful for planning incisions and craniotomies [23–26]. These benefits may be lost after dural opening and tumor resection is begun, because of brain shift [34,35]. This phenomenon, in part, spurred the development of intraoperative MRI (see below). In any event, intracranial surgery with IGS using preoperative anatomical MRI has become a routine part of contemporary neurosurgery; for surgery at the skull base, concerns regarding brain shift matter relatively little (> Figure 39-3). Application of IGS to such ‘‘purely stereotactic’’ applications as biopsy and functional stereotaxis required evidence that the accuracy of frameless approaches equaled that of framebased surgery. Dorward et al demonstrated that this was the case [36]. They found that CT-based navigation was slightly more accurate than that based on MRI, with a mean error of 1.1 mm for CT and 1.4 mm for MRI; however, this did not have clinical implications in their series of
. Figure 39-2 fMRI and DTI in a patient with right frontal low grade glioma. Images confirm that the right primary motor cortex is posterior to the tumor (a) and that the corticospinal tract is also behind and inferior to the lesion (b)
601
602
39
MRI in image guided surgery
. Figure 39-3 Image guided surgery: transsphenoidal surgery with StealthStation images. (a), navigation with CT. The pituitary gland and tumor are poorly seen. (b), navigation with MRI, clearly showing the gland and microadenoma
MRI in image guided surgery
21 stereotactic biopsies using IGS. Several years later, the same group demonstrated advantages of IGS guidance over frame-based stereotaxis for biopsy [37]. OR time was shorter, the complication rate was lower, and as a result hospital stays and costs were reduced. It is worth noting that other authors have found that for the localization of very small cavernous angiomas a stereotactic frame was more accurate than a surgical navigation system [38]. As the accuracy of ‘‘frameless’’ localization for functional indications and/or for radiosurgery has been shown elsewhere to be comparable to that of stereotactic frames (see below), the significance of this finding is unclear. Another group examined the effect of different MRI sequences on stereotactic accuracy. They assumed that CT was the standard against which accuracy should be measured, and that infrared IGS technology was inherently accurate. Although the findings were that CT was indeed more accurate, T1 weighted MRI was only 23% less so, and T2 weighted images 37% less accurate. These results indicated that even lesions seen only on T2W scans could be approached with adequate precision using IGS [39]. With this knowledge in hand, the obvious question relates to the desirability of using IGS to perform stereotactic biopsy. IGS biopsies have been done with freehand guidance or with an articulated and rigidly fixed arm holding the biopsy instrument. A skull-mounted device (Navigus, Medtronic Navigation, Louisville CO), invented as a guide for intraoperative MRI based surgery, has been adapted for biopsies and functional stereotaxis using ‘‘conventional’’ IGS. One group found that targeting with the Navigus system was slightly more accurate than with a stereotactic frame [40]. The mean error from target was 0.33 mm +/ 0.16 mm with the probe’s eye planning method using the Navigus and StealthStation software, versus 1.03 mm +/ 0.19 mm with the Cosman-Roberts-Wells frame.
39
The conceptual advantages of IGS for stereotactic biopsy are in part similar to those achieved with craniotomy. These include uncoupling of imaging from surgery, the ability to preplan the procedure, and registration of multiple datasets. For instance, the use of PET has been described as a way to maximize the diagnostic yield of biopsy by targeting the metabolically most active area [41]. IGS also brings to stereotactic biopsy certain specific advantages. Perhaps the most important is the ability to pick multiple targets with ease, and without a separate registration for each, as is needed for ‘‘traditional’’ frame-based biopsies. Of course, IGS computers are compatible with the most commonly used stereotactic frames (Leksell and Radionics), in which case the planning software can be used to choose biopsy sites; however, coordinates will still need to be set separately for each target, with the attendant error and possibility of missing the area of interest. Image-guided biopsy with a frame allows for surgery to be done easily under local anesthesia. However, if the patient’s head is secured in three-point fixation (such as the Mayfield clamp, Integra Radionics), positioning may be easier than with a frame regardless of anesthetic technique. This is so for lateral and especially prone positions, which may be desirable based on lesion location and the safest trajectory to the target. Several authors employing different technologies have established the safety and reliability of IGS guided biopsy [42–45]. These studies confirmed that the diagnostic yield and complication rate were similar whether biopsies were done using a frame-based or ‘‘frameless’’ technique. The complication rate was on the order of 2–3%, and the chance of a nondiagnostic biopsy about 8% overall. As in the case of frame-based biopsies, the odds of making a diagnosis are improved by confirmation, via intraoperative pathological examination, that lesional tissue has been obtained [37,43].
603
604
39
MRI in image guided surgery
Stereotactic Functional Surgery with MRI The use of IGS to target lesions visible on MRI may be termed ‘‘morphological stereotaxis’’, as opposed to ‘‘functional stereotaxis’’ aimed at nuclei and/or tracts that are not necessarily seen on contemporary imaging. The small volumes involved, and the great precision required to achieve a good clinical result, allow for very little error in IGS. The history of clinically applied functional stereotaxis dates to the late 1940s, when Spiegel and Wycis applied the concept of Cartesian coordinates put forward by Horsley and Clarke in 1908 (but used only for physiological animal experimentation by Horsley) [46]. Air or positive contrast ventriculography were the standard method of functional target localization for nearly 40 years, before the advent of CT-guided stereotaxis [47,48]. Measurements derived from the coordinates of the anterior and posterior commisures were used to identify target locations in the basal ganglia, and thalamus according to the method described by Talairach [49,50]. The advent of CT held out the possibility of functional targeting without the discomfort, morbidity, and time needed for ventriculography-guided surgery. CT has the advantage of high spatial accuracy, with little distortion. However, it is mainly useful in functional stereotaxis as a source of indirect targeting derived from ventricular landmarks. MRI, on the other hand, holds out the potential of detailed anatomic visualization and accurate direct targeting of deep brain structures (> Figure 39-4). As in morphological stereotaxis, the accuracy of digital, sectional imaging compared to the proven technique of ventriculograpy was required. This was done first with CT [51]. Hariz showed how distortion from air injection caused third ventricular widening and anterior displacement of the midbrain; physiologically determined targets were found equally well with both techniques [52]. Not long afterwards, similar studies were done comparing the accuracy of
MRI-guided surgery to ventriculography, a more sensitive task given the possibility of spatial inaccuracies in MRI; in fact, one study found that ventriculography was significantly more accurate, with errors of up to 4.6 mm with the MRI derived coordinates [53]. However, the consensus emerging from various studies comparing the two methods was that MRI accuracy was adequate for functional neurosurgery [49,54,55]. As a result, ventriculography has largely become a technique of historical interest. In fact, a relatively recent study showed that even this time-tested technique was inadequate for accurate target definition in the globus pallidus, and that intraoperative electrophysiology found differences as high as 10 mm compared to the target predicted by imaging alone. There is little disagreement now that even with the best MRI-derived direct targeting, intraoperative electrophysiology with some combination of stimulation and microelectrode recording will help to achieve the best results in functional stereotactic surgery, as is discussed in detail elsewhere in this book [56,57]. There is a reasonable consensus now regarding current technique for functional stereotaxis. Most patients will undergo a high resolution stereotactic MRI in advance of the surgery or as part of the procedure, with a stereotactic frame applied. Acquisition with 3T MRI may provide more anatomical detail without loss of spatial accuracy compared to 1.5 T scanners [58,59] On various sequences the anterior and posterior commisures are easily identified for indirect targeting. On coronal T2W images the red nucleus can be identified as a landmark for the STN; ideally the STN, globus pallidus or other targets can be seen, allowing for true direct targeting. Imaging data is transferred to the IGS computer in the OR. If a stereotactic CT is done as part of the procedure, image fusion is performed to register this scan to the MRI. A combination of targeting methods is done to define the target – indirect off the commisures, indirect off the red nucleus
MRI in image guided surgery
39
. Figure 39-4 IGS for functional neurosurgery. (a), CT image for targeting (on imaging console) of left Vim nucleus in patient with essential tremor. (b), MRI on StealthStation with targeting of left Vim
605
606
39
MRI in image guided surgery
(if the STN is the target), direct based on the MRI, and possibly an atlas that is included with the IGS software. The ultimate site for DBS electrode placement or lesioning is refined using intraoperative electrophysiology [60–62]. Zonenshayn et al. found that no one imaging technique clearly was the single best choice but that in fact a ‘‘combination’ correlated best with electrophysiology [55]. It is worth noting that direct targeting of the STN was the least reliable method in their series. The reemergence of surgery for patients with refractory psychiatric conditions including obsessive-compulsive disorder and depression has been supported in large part by fMRI studies that have demonstrated a measurable organic substrate in these disorders. MRI-targeted lesioning, or more frequently DBS, has been applied based on this new physiologic information with promising results [63,64]. While DBS and in some cases lesioning are at this point the most common techniques of modulating the brain via functional stereotaxis, the future holds the prospect of biological therapies for movement disorders and other conditions [65,66]. These treatments still require precise insertion of catheters, implanted cells, etc. into specific and small areas of the brain. MRI guided IGS will remain a critical part of functional stereotactic neurosurgery. Technical advances in image guidance may make the stereotactic frame obsolete, but these new methods likewise will rely on MRI as the basis for targeting [67]. MRI is an important part of IGS in functional neurosurgery beyond targeting the diencephalon, basal ganglia or brainstem with ‘‘purely stereotactic’’ technique. Epidural motor cortex stimulation is a method proposed as a treatment for neuropathic pain [68] or movement disorders [69]. Placement of the stimulating electrode strip over the primary motor cortex is greatly facilitated by using fMRI and IGS [70]. Ablative surgery for patients with refractory seizure disorders is facilitated by the use of
stereotactic MRI for evaluation and definitive surgery. This includes accurate insertion of depth electrodes [71] and stereotactic lesioning of the medial temporal lobe [72].
Intraoperative MRI All of the above uses of MRI refer to preoperative scans used for image guidance. The next stage in the evolution of MRI in IGS has been the introduction of intraoperative MRI (iMRI), which has been applied to practically all of the indications described above. The impetus behind this development derived from the desire for resection control [73,74] and compensation for brain shift [34,75], primarily in tumor surgery. Inevitably, as the possibilities of this new technology became apparent, different groups have used it for other types of stereotactic neurosurgery. The first iMRI was conceived and built in the early to mid 1990s at the Brigham and Women’s Hospital in Boston, as a collaboration between the Departments of Neurosurgery and Radiology and General Electric Medical Systems (GEMS, Milwaukee, WI) [76,77]. Based on a 0.5 T magnet, it was named the GE Signa. This remains the most ‘‘purely stereotactic’’ system to date, as the Cartesian coordinate space and the surgical space are identical, with surgery performed via a gap between the two vertical magnet poles. Scans can be done as often as needed, even without interrupting surgery, and an integrated infrared navigation device made this a truly IGS system. However, certain factors limited the acceptance of the GE Signa, including the narrow (56 cm) gap between the magnet poles, difficulties in patient positioning, the location of the magnet outside of the main OR area, and the need for all instruments to be completely MRI compatible. As other iMRI solutions have gained popularity, GEMS no longer markets the Signa. Conceptually, however, it remains the iMRI against which other devices should be measured.
MRI in image guided surgery
The search for the ideal iMRI has involved an attempt to reach various compromises with the two main conflicting goals of intraoperative imaging: image quality and overall functionality versus user-friendliness. iMRIs have been classified by the way that patients or images are moved for scanning [78]. Most neurosurgeons, though, refer to systems as being either ‘‘high field’’ or ‘‘low field’’, referring to the magnet strength and hence to how close the iMRI approaches the utility of a diagnostic system [79]. Not longer after the advent of the GE Signa, several authors described their experience using a 0.2 T–0.3 T iMRI with a 25 cm horizontal gap [80–82]. Practically this meant that surgery had to be done outside the ‘‘fringe fields’’ of the magnet, with the patient then moved into the iMRI for imaging [83]. The desire to achieve diagnostic image quality, with the ability to acquire functional imaging and MR spectroscopy, led the Erlangen group to develop a 1.5 T iMRI [84]. This system is implemented in a regular OR and includes integrated navigation. The rapid falloff the magnetic field allows for the use of routine instrumentation. When imaging is done, the patient is rotated on the specially designed table into the magnet. Nimsky et al have described true intraoperative, updated diffusion tensor imaging (DTI) tractograpy with this system [85]. Another method of acquiring high field iMRI in an otherwise routine OR environment was developed in Calgary, and uses a 1.5 T magnet that sits in a shielded alcove between imaging sessions. When needed, the imager moves on a track into scanning position [86]. Further work with this system has allowed for the incorporation of tractography (G Sutherland, personal communication). At the University of Minnesota a 1.5 T diagnostic magnet was used for surgery in the radiology department [87]. Doing so meant revisiting many of the ergonomic obstacles of the Signa system, but full functional imaging capability was available including fMRI. This group was the first to describe surgery in a 3 T
39
environment [88]. Two main benefits have come from their work. First is the feasibility and utility of a shared resource iMRI that is used for diagnostic imaging when (as is true most of the time) it is not being used for surgery. Second, Hall et al. developed the abovementioned Navigus guide, an ingenious skull-mounted tool that can be used for any procedure requiring stereotactic guidance [89]. More on 3 T iMRI later on. The above projects are efforts to bring diagnostic MRI – the more the better – to the neurosurgical OR. To some extent they require adapting the OR to the MRI. There has been one iMRI system that has taken the opposite approach, i.e., to adapt the MRI to the OR. In 2000 Hadani et al. reported the development of a 0.12 T iMRI, meant to be used in a regular neurosurgical OR [90]. Other groups also reported their early experience with this system, named the PoleStar N-10 (Medtronic Navigation, Louisville CO) [78,79,91]. This compact iMRI is stored in a shielded compartment and rolled out as needed for surgery, and is parked under a regular OR table. The low field strength allows for the use of regular instrumentation (except when operating directly through the magnet poles). The major limitations of this system were the limited field of view (FOV), 25 cm vertical magnet pole gap, and the variable image quality. This spurred an upgrade that enlarged the device somewhat but increased the magnet field strength to 0.15 T and the gap to 27 cm, increased the FOV, and in general improved the image quality [92]. fMRI was acquired with this system, although not during surgery [93]. The above and other publications, unsurprisingly, have mostly focused on the use of iMRI for tumor resection control. Across all platforms there were fairly consistent conclusions: that iMRI is particularly useful for surgery in patients with gliomas (especially low grade tumors) [94,95] and pituitary macroadenomas [96,97]. Overall, intraoperative imaging has led to the resection of otherwise unvisualized tumor
607
608
39
MRI in image guided surgery
in about 1/3 of cases. Compared to ‘‘conventional’’ IGS, surgery with iMRI was more likely to achieve a 90% tumor resection in patients with low grade gliomas (M Schulder, unpublished data). A similar trend in patients with high grade tumors did not reach statistical significance. Of course, surgical navigation updated with new imaging has avoided the problem of brain shift [98]. iMRI has been used for indications other than tumor resection. Perhaps the most obvious is for stereotactic biopsy. The theoretical advantages of doing so are that imaging can be used to confirm that specimens are indeed being taken from the desired target, thereby improving diagnostic yield. The other main benefit is to rule out a procedurerelated hematoma or other surgical complication. Bernays et al described their experience with stereotactic biopsy in a GE Signa iMRI [99] in 114 patients. Pathological tissue was obtained from all. One patient required a delayed emergency craniotomy because of edema. Stereotactic biopsy in the PoleStar iMRI has been done by this chapter’s primay author. All imaging was done in the OR, with a Navigus guide used to direct an MRI compatible cannula to the target. Biopsies were taken after imaging with the cannula in place. In 30 patients, a diagnosis was obtained in all. Intraoperative imaging led to a repositioning of the cannula to ensure ideal placement in the lesion in 8/30 patients (> Figure 39-5). There were no complications in this series. In patients with refractory epilepsy iMRI use has been similar to that used in tumor resection – namely, for navigation and resection control of nonlesional epileptic foci [100,101]. Applications that have used iMRI for more ‘‘stereotactic uses’’ have included management of complex hydrocephalus in children, when intraoperative verification of catheter placement is desired [102]. iMRI for DBS electrode placement has been described, and with the increasing indications for this procedure we may see greater of iMRI use in the coming decade. The functional neurosurgery group at UCSF has shown that DBS can be done
in a high field MRI without increased risk, with the ability to target the STN without intraoperative electrophysiology, and with excellent placement confirmed using near real-time imaging [103]. Of course, more work will need to be done to ensure that clinical results are as good or better using this method as with frame-based or ‘‘frameless stereotaxis’’. Recently, the move to 3 T iMRI has begun, with reports of new systems appearing [104]. These and other iMRI implementations have been proposed as part of a comprehensive imaging suite, to include CT, digital angiography, and possibly PET [105]. The advantages of these projects have been touted as being able to serve multiple operating rooms and services, and to facilitate shared-resource imaging that make intraoperative imaging a profit center and not a financial drain. These are worthy goals, of course. But the question is raised, at what point do these systems cease to provide actual intraoperative imaging? If the patient has to be moved from the OR to another site for imaging, even if it is next door, is that different in concept than moving to a completely separate radiology department? At a minimum, such moves will not encourage routine use of iMRI and will allow for enough ongoing change, in the time between imaging and return to the OR, that the benefits of updated navigation also will be unclear. The jury will remain out on this concept of iMRI. A move back towards the original Signa concept, where stereotactic, image, and surgical space are one, is the best possible alternative.
Stereotactic Radiosurgery with MRI A field that has benefited greatly from the use of MRI has been stereotactic radiosurgery (SRS). Invented and named by Leksell in 1951, this minimally invasive surgical technique almost by definition relies on image guidance to find its target and ensure a safe and effective result [106].
MRI in image guided surgery
39
. Figure 39-5 Stereotactic biopsy with iMRI. (a), patient positioned prone with MRI receive coil in place. (b), preoperative iMRI with entry and target point chosen for biopsy. Field of view does not encompass entire cranium. (c), Compare function shows (left) initial cannula pass and (right) improved placement in the lesion
Until the mid 1980s SRS was done with predigital imaging – angiography, ventriculography, or skull X-rays (or a version thereof such as polytomography) [107]. The availability of CT-guided
stereotaxis began to open up SRS to treat a greater variety of lesions, and smaller ones, than had been possible before. This in large part encouraged the development of linear accelerator (Linac) based
609
610
39
MRI in image guided surgery
treatments [108,109] and the wider use of the gamma knife (GK) SRS [110]. This was true as well for the use of heavy particle irradiation, but the number of such centers is limited by the cost and complexity of the systems [111]. CT not only allowed for the targeting of directly visualized lesions, but also for treatment planning that would show the dose to be delivered to adjacent normal structures. The advent of MRI-guided SRS has greatly improved the efficacy and safety of SRS. Radiobiology suggests that SRS is most potent for treating benign lesions, such that patients with schwannomas and meningiomas of the skull base are often ideal candidates for SRS [112]. These tumors are seen far better on MRI than CT thanks to the elimination on MRI of the bonehardening artifact (> Figure 39-6). The same is true for patients with lesions in close proximity to the skull elsewhere, such as parasagittal meningiomas [113]. Furthermore, the proximity of critical structures such as the optic chiasm that are poorly seen on CT makes the risk of visual loss from SRS potentially high. Using MRI, even individual cranial nerves can be seen and the
treatment plan adjusted accordingly [114]. This is so whether direct MRI targeting is used [115] or image fusion between MRI and CT [116]. The Cyberknife (Accuray, Sunnyvale CA), an innovative, robotic, Linac SRS tool (discussed in more detail elsewhere in this volume) has been used to treat patients with medically refractory trigeminal neuralgia (TN). As recently as several years ago, CT cisternography was recommended as the imaging technique of choice to target the trigeminal nerve [117]. This reflected the vendor’s (reasonable) insistence on CT as the reference image, for maximum spatial accuracy. Fortunately, excellent image fusion software now allows for the use of MRI to treat TN patients with the Cyberknife [118], as is done with other Linacs [119] and the GK [120]. Functional neuroimaging has been evaluated as an adjunct to SRS. The tolerances to irradiation of certain cranial nerves and the brainstem, which can be imaged with standard morphological MRI, are reasonably well understood [121,122]. fMRI can be registered for use in SRS planning software, and identification of ‘‘eloquent’’ cortical areas can be used to redirect
. Figure 39-6 Cyberknife SRS plan in patient with a right vestibular schwannomas. (a), CT based plan shows only tumor in cerebellopontine angle. (b), MRI (fused to CT) shows intracanalicular component and edge of tumor with greater clarity
MRI in image guided surgery
dose curves in a way that will limit the chance of a radiation-induced deficit (> Figure 39-7) [123,124]. Maruyama et al. have incorporated DTI tractography into GKRS planning, and have reported data suggesting that the dose of single session SRS to the optic radiation should be limited to eight Gy or less [125] whereas the corticospinal tract can tolerate up to 23 Gy [126]. More work will need to be done to confirm these results and to determine to what extent functional neuroimaging can affect SRS planning.
MRI for Functional Stereotactic Radiosurgery with MRI An irony of SRS is that while Leksell envisioned and invented the technique primarily as a tool for functional neurosurgery, the explosive interest in SRS that began some 20 years ago has been almost all directed at treating lesions – . Figure 39-7 fMRI for SRS. Treatment plan in patient with parasagittal meningiomas. Isodose lines confirm low dose is delivered to the left primary motor cortex (activation area colored blue)
39
mostly patients with neoplasms, and some with arteriovenous malformations. More recently, however, neurosurgeons have again begun to employ SRS to treat patients with functional disorders, besides TN. The latter condition has the advantage of a target that can usually be seen with relative ease on MRI [120]. This is not the same for functional targets in the diencephalon, basal ganglia, or brainstem. For instance, defining thalamic, pallidal, or subthalamic targets in patients with tremor or Parkinson’s disease who are to undergo DBS implantation requires a combination of imaging and physiological methods [127]. For decades, before making a surgical lesion for movement disorders, macrostimulation at a minimum was done to ensure the safety of this irreversible step. The main skepticism regarding functional SRS, therefore, has been related to the lack of such feedback in the non-invasive SRS [128]. Several authors have countered this argument [129], in particular Sato et al. who performed stereotactic radiofrequency thalamotomy in four patients using MRI-derived coordinates that would be used for radiosurgical lesioning [130]. The safety and efficacy of radiosurgical thalamotomy and pallidotomy, for patients with essential tremor, Parkinson’s disease, and multiple sclerosis tremor has been reported over the last decade [131–133]. Even the STN has been proposed as a radiosurgical target [134]. Needless to say, these procedures could not be considered without the most accurate MR imaging possible. Further evaluation of clinical and imaging results after functional SRS will still need to be done before this technique can be considered as an equivalent alternative treatment for patients with movement disorders.
Conclusions Contemporary stereotactic and functional neurosurgery is barely conceivable without image guided surgery based on MRI. The near future will see refinements based on magnets with 3 T or even
611
612
39
MRI in image guided surgery
higher field strengths [135]. The role of neurosurgeons in confirming the spatial accuracy of these devices will be critical in allowing them to be used as the imaging source for open and radiosurgical treatments of functional disorders. Stereotactic neurosurgeons will also be at the forefront of refinements in IGS and iMRI that will bring these methods even more into the neurosurgical mainstream. The next edition of this textbook will have many exciting new developments to report.
References 1. Macchia RJ, Termine JE, Buchen CD, Raymond V. Damadian, MD.: magnetic resonance imaging and the controversy of the 2003 Nobel Prize in Physiology or Medicine. J Urol 2007;178(3 Pt 1):783-5. 2. Damadian R, Minkoff L, Goldsmith M, Stanford M, Koutcher J. Tumor imaging in a live animal by focusing NMR (FONAR). Physiol Chem Phys 1976;8(1):61-5. 3. Lee JY, Lunsford LD, Subach BR, Jho HD, Bissonette DJ, Kondziolka D. Brain surgery with image guidance: current recommendations based on a 20-year assessment. Stereotact Funct Neurosurg 2000;75(1):35-48. 4. Bradford R, Thomas DG, Bydder GM. MRI-directed stereotactic biopsy of cerebral lesions. Acta Neurochir Suppl (Wien) 1987;39:25-7. 5. Mercier C, Le Bas JF, Benabid AL, Pasquier B. MRI contribution to the stereotactic management of cerebral tumours. Appl Neurophysiol 1987;50(1–6):155-8. 6. Frank F, Fabrizi AP, Frank-Ricci R, Gaist G, Sedan R, Peragut JC. Stereotactic biopsy and treatment of brain stem lesions: combined study of 33 cases (Bologna-Marseille). Acta Neurochir Suppl (Wien) 1988;42:177-81. 7. Kelly PJ, Kall BA, Goerss S, Earnest F. Present and future developments of stereotactic technology. Appl Neurophysiol 1985;48(1–6):1-6. 8. Ferry M, Cabanis E, Cosnard G, Bocquet M, Desgeorges M. Stereotactic biopsy of the brain with the Leksell frame in the context of modern radiology. J Neuroradiol 1988;15(1):63-75. 9. Heilbrun MP, Sunderland PM, McDonald PR, Wells TH, Jr., Cosman E, Ganz E. Brown-Roberts-Wells stereotactic frame modifications to accomplish magnetic resonance imaging guidance in three planes. Appl Neurophysiol 1987;50(1–6):143-52. 10. Maciunas R, Galloway R, Lattimer J. The application accuracy of stereotactic frames. Neurosurgery 1994;35: 682-95. 11. Altschuler E, Lunsford LD, Kondziolka D, Wu A, Maitz AH, Sclabassi R, Martinez AJ, Flickinger JC.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
Radiobiologic models for radiosurgery. Neurosurg Clin N Am 1992;3(1):61-77. Apuzzo ML, Sabshin JK. Computed tomographic guidance stereotaxis in the management of intracranial mass lesions. Neurosurgery 1983;12(3):277-85. Couldwell WT, Apuzzo ML. Initial experience related to the use of the Cosman-Roberts-Wells stereotactic instrument. Technical note. J neurosurg 1990;72(1): 145-8. Soo T, Bernstein M, Provias J, Tasker R, Lozano A, Guha A. Failed stereotactic biopsy in a series of 518 cases. Stereotact Funct Neurosurg 1995;64(4):183-96. Chandrasoma PT, Smith MM, Apuzzo ML. Stereotactic biopsy in the diagnosis of brain masses: comparison of results of biopsy and resected surgical specimen. Neurosurgery 1989;24(2):160-5. Winkler D, Trantakis C, Lindner D, Richter A, Schober J, Meixensberger J. Improving planning procedure in brain biopsy: coupling frame-based stereotaxy with navigational device STP 4.0. Minim Invasive Neurosurg 2003;46(1):37-40. Alexander EI, Kooy M, van Herk M, Shwartz M, Barnes D, Tarbell N, Mulkern R, Holupka E, Loeffler J. Magnetic resonance image-directed sterotactic neurosurgery: use of image fusion with computerized tomography to enhance spatial accuracy. J Neurosurg 1995;83 (August,1995):271-6. Kelly PJ, Alker GJ, Jr. A stereotactic approach to deepseated central nervous system neoplasms using the carbon dioxide laser. Surg Neurol 1981;15(5):331-4. Roberts DW, Strohbehn JW, Hatch JF, Murray W, Kettenberger H. A frameless stereotaxic integration of computerized tomographic imaging and the operating microscope. J Neurosurg 1986;65(4):545-9. Roessler K, Ungersboeck K, Aichholzer M, Dietrich W, Goerzer H, Matula C, Czech T, Koos WT. Frameless stereotactic lesion contour-guided surgery using a computernavigated microscope. Surg Neurol 1998;49(3):282-8; discussion 8-9. Golfinos J, Fitzpatrick B, Smith L, Spetzler R. Clinical use of a frameless stereotactic arm: results of 325 cases. J Neurosurg 1995;83:197-205. Barnett G, Kormos D, Steiner C, Weisenberger J. Intraoperative localization using an armless, frameless stereotactic wand. J Neurosurg 1993;78:510-4. Benardete EA, Leonard MA, Weiner HL. Comparison of frameless stereotactic systems: accuracy, precision, and applications. Neurosurgery 2001;49(6):1409-15. Mascott CR. Comparison of magnetic tracking and optical tracking by simultaneous use of two independent frameless stereotactic systems. Neurosurgery 2005; 57(4 Suppl):295-301; discussion 295-301. Smith KR, Frank KJ, Bucholz RD. The Neurostation – a highly accurate, minimally invasive solution to frameless stereotactic neurosurgery. Comput Med Imaging Graph 1994;18:247-56.
MRI in image guided surgery
26. Germano I. The NeuroStation system for image-guided frameless stereotaxy. Neurosurgery 1995;37:348-50. 27. Chen JC, Petrovich Z, Giannotta SL, Yu C, Apuzzo ML. Radiosurgical salvage therapy for patients presenting with recurrence of metastatic disease to the brain. Neurosurgery 2000;46(4):860-6; discussion 6-7. 28. Pirotte B, Goldman S, Dewitte O, Massager N, Wikler D, Lefranc F, Ben Taib NO, Rorive S, David P, Brotchi J, Levivier M. Integrated positron emission tomography and magnetic resonance imaging-guided resection of brain tumors: a report of 103 consecutive procedures. J Neurosurg 2006;104(2):238-53. 29. Vougioukas VI, Coulin CJ, Shah M, Berlis A, Hubbe U, Van Velthoven V. Benefits and limitations of image guidance in the surgical treatment of intracranial dural arteriovenous fistulas. Acta Neurochir (Wien) 2006; 148(2):145-53; discussion 53. 30. Schulder M, Maldjian JA, Liu WC, Holodny AI, Kalnin AT, Mun IK, Carmel PW. Functional imageguided surgery of intracranial tumors located in or near the sensorimotor cortex. J Neurosurg 1998;89(3): 412-8. 31. Atlas S, Howard RI, Maldjian J, Alsop D, Detre J, Listerud J, D’Esposito M, Judy K, Zager E, Stecker M. Functional MRI of regional brain activity in patientws with intracerebral gliomas: findings and implications for clinical management. Neurosurgery 1996;38:329-38. 32. McDonald J, Chong B, Lewine J, Jones G, Burr R, McDonald P, Koehler S, Tsuruda J, Orrison W, Heilbrun M. Integration of preoperative and intraoperative functional brain mapping in a frameless stereotactic environment for lesions near eloquent cortex. J Neurosurg 1999;90:591-8. 33. Holodny AI, Schwartz TH, Ollenschleger M, Liu WC, Schulder M. Tumor involvement of the corticospinal tract: diffusion magnetic resonance tractography with intraoperative correlation. J Neurosurg 2001;95(6):1082. 34. Dorward N, Alberti O, Velani B, Gerritsen F, Harkness W, Kitchen N, Thomas D. Postimaging brain distortion: magnitude, correlates, and impact on neuronavigation. J Neurosurg 1998;88(4):656-62. 35. Roberts D, Hartov A, Kennedy F, Miga M, Paulsen K. Intraoperative brain shift and deformation: a quantitative analysis of cortical displacement in 28 cases. Neurosurgery 1998;43(4):749-58; discussion 58-60. 36. Dorward NL, Alberti O, Palmer JD, Kitchen ND, Thomas DG. Accuracy of true frameless stereotaxy: in vivo measurement and laboratory phantom studies. Technical note. J Neurosurg 1999;90(1):160-8. 37. Dorward NL, Paleologos TS, Alberti O, Thomas DG. The advantages of frameless stereotactic biopsy over frame-based biopsy. Br J Neurosurg 2002;16(2):110-8. 38. Grunert P, Charalampaki K, Kassem M, BoecherSchwarz H, Filippi R, Grunert P, Jr. Frame-based and frameless stereotaxy in the localization of cavernous angiomas. Neurosurg Rev 2003;26(1):53-61.
39
39. Schulder M, Fontana P, Lavenhar MA, Carmel PW. The relationship of imaging techniques to the accuracy of frameless stereotaxy [In Process Citation]. Stereotact Funct Neurosurg 1999;72(2–4):136-41. 40. Quinones-Hinojosa A, Ware ML, Sanai N, McDermott MW. Assessment of image guided accuracy in a skull model: comparison of frameless stereotaxy techniques vs. frame-based localization. J Neurooncol 2006; 76(1):65-70. 41. Levivier M, Goldman S, Pirotte B, Brucher J-M, Baleriaux D, Brotchi J. Diagnostic yield of stereotactic brain biopsy guided by positron emission tomography with [18-F] fluorodeoxyglucose. J Neurosurg 1995;83: 445-52. 42. Paleologos TS, Dorward NL, Wadley JP, Thomas DG. Clinical validation of true frameless stereotactic biopsy: analysis of the first 125 consecutive cases. Neurosurgery 2001;49(4):830-5; discussion 5-7. 43. Barnett GH, Miller DW, Weisenberger J. Frameless stereotaxy with scalp-applied fiducial markers for brain biopsy procedures: experience in 218 cases. J Neurosurg 1999;91(4):569-76. 44. Woodworth GF, McGirt MJ, Samdani A, Garonzik I, Olivi A, Weingart JD. Frameless image-guided stereotactic brain biopsy procedure: diagnostic yield, surgical morbidity, and comparison with the frame-based technique. J Neurosurg 2006;104(2):233-7. 45. Dammers R, Haitsma IK, Schouten JW, Kros JM, Avezaat CJ, Vincent AJ. Safety and efficacy of frameless and frame-based intracranial biopsy techniques. Acta Neurochir (Wien) 2008;150(1):23-9. 46. Gildenberg P. The history of stereotactic neurosurgery. Neurosurg Clin No Am 1990;1:765-81. 47. David M, Ruggiero G, Talairach J. Comparison between encephalography and ventriculography. Acta radiol 1954;42(1):37-42. 48. Tasker RR. One man’s recollection of 50 years of functional and stereotactic neurosurgery. Neurosurgery 2004; 55(4):968-74; discussion 74-6. 49. Levesque MF, Wilson CL, Behnke EJ, Zhang JX. Accuracy of MR-guided stereotactic electrode implantation. 1. Preimplantation correlation with the Talairach system. Stereotact Funct Neurosurg 1990;54–55:51-5. 50. Hawrylyshyn PA, Tasker RR, Organ LW. Third ventricular width and the thalamocapsular border. Appl Neurophysiol 1976;39(1):34-42. 51. Narabayashi H. Recent status of stereotaxic surgery. Surg Neurol 1983;19(6):493-6. 52. Hariz MI, Bergenheim AT. A comparative study on ventriculographic and computerized tomography-guided determinations of brain targets in functional stereotaxis. J Neurosurg 1990;73(4):565-71. 53. Mandybur G, Morenski J, Kuniyoshi S, Iacono RP. Comparison of MRI and ventriculographic target acquisition for posteroventral pallidotomy. Stereotact Funct Neurosurg 1995;65(1–4):54-9.
613
614
39
MRI in image guided surgery
54. Schuurman PR, de Bie RM, Majoie CB, Speelman JD, Bosch DA. A prospective comparison between threedimensional magnetic resonance imaging and ventriculography for target-coordinate determination in frame-based functional stereotactic neurosurgery. J Neurosurg 1999;91(6):911-4. 55. Zonenshayn M, Rezai AR, Mogilner AY, Beric A, Sterio D, Kelly PJ. Comparison of anatomic and neurophysiological methods for subthalamic nucleus targeting. Neurosurgery 2000;47(2):282-92; discussion 92-4. 56. Hutchison WD, Allan RJ, Opitz H, Levy R, Dostrovsky JO, Lang AE, Lozano AM. Neurophysiological identification of the subthalamic nucleus in surgery for Parkinson’s disease. Ann Neurol 1998;44(4):622-8. 57. Weinberger M, Hamani C, Hutchison WD, Moro E, Lozano AM, Dostrovsky JO. Pedunculopontine nucleus microelectrode recordings in movement disorder patients. Exp Brain Res 2008;188(2):165-74. 58. Spiegelmann R, Nissim O, Daniels D, Ocherashvilli A, Mardor Y. Stereotactic targeting of the ventrointermediate nucleus of the thalamus by direct visualization with highfield MRI. Stereotact funct Neurosurg 2006;84(1):19-23. 59. Azmi H, Schulder M. Stereotactic accuracy of a 3-tesla magnetic resonance unit. Stereotact funct Neurosurg 2003;80(1–4):140-5. 60. Starr PA, Christine CW, Theodosopoulos PV, Lindsey N, Byrd D, Mosley A, Marks WJ, Jr. Implantation of deep brain stimulators into the subthalamic nucleus: technical approach and magnetic resonance imagingverified lead locations. J Neurosurg 2002;97(2):370-87. 61. Machado A, Rezai AR, Kopell BH, Gross RE, Sharan AD, Benabid AL. Deep brain stimulation for Parkinson’s disease: surgical technique and perioperative management. Mov Disord 2006;21 Suppl 14:S247-58. 62. diPierro CG, Francel PC, Jackson TR, Kamiryo T, Laws ER, Jr. Optimizing accuracy in magnetic resonance imaging-guided stereotaxis: a technique with validation based on the anterior commissure-posterior commissure line. J Neurosurg 1999;90(1):94-100. 63. Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, Schwalb JM, Kennedy SH. Deep brain stimulation for treatment-resistant depression. Neuron 2005;45(5):651-60. 64. Greenberg BD, Malone DA, Friehs GM, Rezai AR, Kubu CS, Malloy PF, Salloway SP, Okun MS, Goodman WK, Rasmussen SA. Three-year outcomes in deep brain stimulation for highly resistant obsessivecompulsive disorder. Neuropsychopharmacology 2006; 31(11):2384-93. 65. Feigin A, Kaplitt MG, Tang C, Lin T, Mattis P, Dhawan V, During MJ, Eidelberg D. Modulation of metabolic brain networks after subthalamic gene therapy for Parkinson’s disease. Proc Natl Acad Sci USA 2007;104(49): 19559-64. 66. Marks WJ, Jr., Ostrem JL, Verhagen L, Starr PA, Larson PS, Bakay RA, Taylor R, Cahn-Weiner DA,
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
Stoessl AJ, Olanow CW, Bartus RT. Safety and tolerability of intraputaminal delivery of CERE-120 (adenoassociated virus serotype 2-neurturin) to patients with idiopathic Parkinson’s disease: an open-label, phase I trial. Lancet Neurol 2008;7(5):400-8. Holloway KL, Gaede SE, Starr PA, Rosenow JM, Ramakrishnan V, Henderson JM. Frameless stereotaxy using bone fiducial markers for deep brain stimulation. J Neurosurg 2005;103(3):404-13. Tsubokawa T, Katayama Y, Yamamoto T, Hirayama T, Koyama S. Chronic motor cortex stimulation for the treatment of central pain. Acta Neurochir 1991;52: 137-9. Arle JE, Apetauerova D, Zani J, Deletis DV, Penney DL, Hoit D, Gould C, Shils JL. Motor cortex stimulation in patients with Parkinson disease: 12-month follow-up in 4 patients. J Neurosurg 2008;109(1):133-9. Mogilner AY, Rezai AR. Epidural motor cortex stimulation with functional imaging guidance. Neurosurg Focus 2001;11(3):E4. Pillay PK, Barnett G, Awad I. MRI-guided stereotactic placement of depth electrodes in temporal lobe epilepsy. Br J Neurosurg 1992;6(1):47-53. Parrent AG, Blume WT. Stereotactic amygdalohippocampotomy for the treatment of medial temporal lobe epilepsy. Epilepsia 1999;40(10):1408-16. Berger MS, Deliganis AV, Dobbins J, Keles GE. The effect of extent of resection on recurrence in patients with low grade cerebral hemisphere gliomas. Cancer 1994; 74(6):1784-91. Laws ER, Parney IF, Huang W, Anderson F, Morris AM, Asher A, Lillehei KO, Bernstein M, Brem H, Sloan A, Berger MS, Chang S. Survival following surgery and prognostic factors for recently diagnosed malignant glioma: data from the Glioma Outcomes Project. J Neurosurg 2003;99(3):467-73. Hill D, Maurer C, Jr., Maciunas R, Barwise J, Fitzpatrick J, Wang M. Measurement of intraoperative brain surface deformation under a craniotomy. Neurosurgery 1998; 43(3):514-26; discussion 27-8. Black P, Moriarty T, Alexander E, 3rd, Stieg P, Woodard E, Gleason P, Martin C, Kikinis R, Schwartz R, Jolesz F. Development and implementation of intraoperative magnetic resonance imaging and its neurosurgical applications. Neurosurgery 1997;41(4): 831-42; discussion 42-5. Black PM, Alexander E, 3rd, Martin C, Moriarty T, Nabavi A, Wong TZ, Schwartz RB, Jolesz F. Craniotomy for tumor treatment in an intraoperative magnetic resonance imaging unit. Neurosurgery 1999;45(3): 423-31; discussion 31-3. Kanner AA, Vogelbaum MA, Mayberg MR, Weisenberger JP, Barnett GH. Intracranial navigation by using low-field intraoperative magnetic resonance imaging: preliminary experience. J Neurosurg 2002;97 (5): 1115-24.
MRI in image guided surgery
79. Schulder M, Liang D, Carmel PW. Cranial surgery navigation aided by a compact intraoperative magnetic resonance imager. J Neurosurg 2001;94(6):936-45. 80. Fahlbusch R, Ganslandt O, Nimsky C. Intraoperative imaging with open magnetic resonance imaging and neuronavigation. Childs Nerv Syst 2000;16(10–11): 829-31. 81. Bohinski RJ, Kokkino AK, Warnick RE, Gaskill-Shipley MF, Kormos DW, Lukin RR, Tew JM, Jr. Glioma resection in a shared-resource magnetic resonance operating room after optimal image-guided frameless stereotactic resection. Neurosurgery 2001;48(4):731-42; discussion 42-4. 82. Wirtz CR, Bonsanto MM, Knauth M, Tronnier VM, Albert FK, Staubert A, Kunze S. Intraoperative magnetic resonance imaging to update interactive navigation in neurosurgery: method and preliminary experience. Comput Aided Surg 1997;2(3–4):172-9. 83. Rubino G, Farahani K, McGill D, Van De Wiele B, Villablanca J, Wang-Mathieson A. Magnetic resonance imaging-guided neurosurgery in the magnetic fringe fields: the next step in neuronavigation. Neurosurgery 2000;46(3):643-53; discussion 53-4. 84. Nimsky C, Ganslandt O, Fahlbusch R. Functional neuronavigation and intraoperative MRI. Adv Tech Stand Neurosurg 2004;29:229-63. 85. Nimsky C, Ganslandt O, Fahlbusch R. Implementation of fiber tract navigation. Neurosurgery 2006;58(4 Suppl 2): ONS-292-303; discussion ONS-4. 86. Sutherland G, Kaibara T, Louw D, Hoult D, Tomanek B, Saunders J. A mobile high-field magnetic resonance system for neurosurgery. J Neurosurg 1999;91(5):804-13. 87. Hall W, Liu H, Martin A, Pozza C, Maxwell R, Truwit C. Safety, efficacy, and functionality of high-field strength interventional magnetic resonance imaging for neurosurgery. Neurosurgery 2000;46(3):632-41; discussion 41-2. 88. Hall WA, Galicich W, Bergman T, Truwit CL. 3-Tesla intraoperative MR imaging for neurosurgery. J Neurooncol 2006;77(3):297-303. 89. Hall WA, Liu H, Truwit CL. Navigus trajectory guide. Neurosurgery 2000;46(2):502-4. 90. Hadani M, Katznelson E, Zuk Y, Sinai R. A novel, realtime intraoperative MR imaging-guided system: a wideopen, compact, and displaceable IMR imaging system for conventional neurosurgical operating rooms. J Neurosurg 2000;92:560. 91. Levivier M, Wikler D, De Witte O, Van de Steene A, Baleriaux D, Brotchi J. PoleStar N-10 low-field compact intraoperative magnetic resonance imaging system with mobile radiofrequency shielding. Neurosurgery 2003;53 (4):1001-6; discussion 7. 92. Schulder M, Salas S, Brimacombe M, Fine P, Catrambone J, Maniker AH, Carmel PW. Cranial surgery with an expanded compact intraoperative magnetic resonance imager. Technical note. J Neurosurg 2006;104 (4):611-7. 93. Azmi H, Biswal B, Salas S, Schulder M. Functional imaging in a low-field, mobile intraoperative magnetic
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
105.
39
resonance scanner: expanded paradigms. Neurosurgery 2007;60(1):143-8; discussion 8-9. Nimsky C, Ganslandt O, Buchfelder M, Fahlbusch R. Glioma surgery evaluated by intraoperative low-field magnetic resonance imaging. Acta Neurochir Suppl 2003;85:55-63. Wirtz C, Knauth M, Staubert A, Bonsanto M, Sartor K, Kunze S, Tronnier V. Clinical evaluation and follow-up results for intraoperative magnetic resonance imaging in neurosurgery [In Process Citation]. Neurosurgery 2000;46(5):1112-20; discussion 20-2. Bohinski RJ, Warnick RE, Gaskill-Shipley MF, Zuccarello M, van Loveren HR, Kormos DW, Tew JM, Jr. Intraoperative magnetic resonance imaging to determine the extent of resection of pituitary macroadenomas during transsphenoidal microsurgery. Neurosurgery 2001;49(5): 1133-43; discussion 43-4. Fahlbusch R, Ganslandt O, Buchfelder M, Schott W, Nimsky C. Intraoperative magnetic resonance imaging during transsphenoidal surgery. J Neurosurg 2001; 95(3):381-90. Nabavi A, Black PM, Gering DT, Westin CF, Mehta V, Pergolizzi RS, Jr., Ferrant M, Warfield SK, Hata N, Schwartz RB, Wells WM, 3rd, Kikinis R, Jolesz FA. Serial intraoperative magnetic resonance imaging of brain shift. Neurosurgery 2001;48(4):787-97; discussion 97-8. Bernays RL, Kollias SS, Khan N, Brandner S, Meier S, Yonekawa Y. Histological yield, complications, and technological considerations in 114 consecutive frameless stereotactic biopsy procedures aided by open intraoperative magnetic resonance imaging. J Neurosurg 2002.97: 354-62. Schwartz TH, Marks D, Pak J, Hill J, Mandelbaum DE, Holodny AI, Schulder M. Standardization of amygdalohippocampectomy with intraoperative magnetic resonance imaging: preliminary experience. Epilepsia 2002;43(4):430-6. Buchfelder M, Ganslandt O, Fahlbusch R, Nimsky C. Intraoperative magnetic resonance imaging in epilepsy surgery. J Magn Reson Imaging 2000;12(4):547-55. Samdani AF, Schulder M, Catrambone JE, Carmel PW. Use of a compact intraoperative low-field magnetic imager in pediatric neurosurgery. Childs Nerv Syst 2005; 21(2):108-13; discussion 14. Martin AJ, Larson PS, Ostrem JL, Keith Sootsman W, Talke P, Weber OM, Levesque N, Myers J, Starr PA. Placement of deep brain stimulator electrodes using realtime high-field interventional magnetic resonance imaging. Magn Reson Med 2005;54(5):1107-14. Jankovski A, Raftopoulos C, Vaz G, Hermoye L, Cosnard G, Francotte F, Duprez T. Intra-operative MRI at 3T: short report. Jbr-Btr 2007;90(4):249-51. Matsumae M, Koizumi J, Fukuyama H, Ishizaka H, Mizokami Y, Baba T, Atsumi H, Tsugu A, Oda S, Tanaka Y, Osada T, Imai M, Ishiguro T, Yamamoto M, Tominaga J, Shimoda M, Imai Y. World’s first
615
616
39 106. 107. 108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
MRI in image guided surgery
magnetic resonance imaging/x-ray/operating room suite: a significant milestone in the improvement of neurosurgical diagnosis and treatment. J Neurosurg 2007; 107(2):266-73. Leksell L. The stereotaxic method and radiosurgery of the brain. Acta Chir Scand 1951;102:316-9. Leksell L. Stereotactic radiosurgery. J Neurol Neurosurg Psyc 1983;46:797-803. Betti O, Derechinsky V. Hyperselective encephalic irradiation with a linear accelerator. Acta Neurochir Suppl 1984;33:385-90. Winston KR, Lutz W. Linear accelerator as a neurosurgical tool for stereotactic radiosurgery. Neurosurgery 1988;22:454-64. Lunsford LD, Flickinger JC, Linder G, Maitz A. Stereotactic radiosurgery of the brain using the first United States 210 cobalt-60 source gamma knife. Neurosurg 1989;24:151-9. Hoh DJ, Liu CY, Pagnini PG, Yu C, Wang MY, Apuzzo ML. Chained lightning, part I: Exploitation of energy and radiobiological principles for therapeutic purposes. Neurosurgery 2007;61(1):14-27; discussion-8. Larson DA, Flickinger JC, Loeffler JS. The radiobiology of radiosurgery. Int J Radiat Oncol Biol Phys 1993; 25:557-61. Kondziolka D, Flickinger JC, Perez B. Judicious resection and/or radiosurgery for parasagittal meningiomas: outcomes from a multicenter review. Gamma Knife Meningioma Study Group. Neurosurgery 1998;43(3): 405-13; discussion 13-4. Lunsford LD, Witt TC, Kondziolka D, Flickinger JC. Stereotactic radiosurgery of anterior skull base tumors. Clin Neurosurg 1995;42:99-118. Wowra B, Muacevic A, Jess-Hempen A, Tonn JC. Safety and efficacy of outpatient gamma knife radiosurgery for multiple cerebral metastases. Expert Rev Neurother 2004;4(4):673-9. Mehta VK, Lee QT, Chang SD, Cherney S, Adler JR, Jr. Image guided stereotactic radiosurgery for lesions in proximity to the anterior visual pathways: a preliminary report. Technol Cancer Res Treat 2002;1(3):173-80. Romanelli P, Heit G, Chang SD, Martin D, Pham C, Adler J. Cyberknife radiosurgery for trigeminal neuralgia. Stereotact funct Neurosurg 2003;81(1–4): 105-9. Villavicencio AT, Lim M, Burneikiene S, Romanelli P, Adler JR, McNeely L, Chang SD, Fariselli L, McIntyre M, Bower R, Broggi G, Thramann JJ. Cyberknife radiosurgery for trigeminal neuralgia treatment: a preliminary multicenter experience. Neurosurgery 2008;62(3): 647-55; discussion-55. Chen JC, Greathouse HE, Girvigian MR, Miller MJ, Liu A, Rahimian J. Prognostic factors for radiosurgery treatment of trigeminal neuralgia. Neurosurgery 2008;62(5 Suppl):A53-60; discussion A-1.
120. Jawahar A, Kondziolka D, Kanal E, Bissonette DJ, Lunsford LD. Imaging the trigeminal nerve and pons before and after surgical intervention for trigeminal neuralgia. Neurosurgery 2001;48(1):101-6; discussion 6-7. 121. Tishler RB, Loeffler JS, Lunsford LD, Duma C, Alexander E, Kooy H, Flickinger J. Tolerance of cranial nerves of the cavernous sinus to radiosurgery. Int J Radiat Oncol Biol Phys 1993;27:215-21. 122. Sims E, Doughty D, Macaulay E, Royle N, Wraith C, Darlison R, Plowman PN. Stereotactically delivered cranial radiation therapy: a ten-year experience of linac-based radiosurgery in the UK. Clin Oncol (R Coll Radiol) 1999;11(5):303-20. 123. Witt TC, Kondziolka D, Baumann SB, Noll DC, Small SL, Lunsford LD. Preoperative cortical localization with functional MRI for use in stereotactic radiosurgery. Stereotact funct Neurosurg 1996; 66(1–3):24-9. 124. Liu WC, Schulder M, Narra V, Kalnin AJ, Cathcart C, Jacobs A, Lange G, Holodny AI. Functional magnetic resonance imaging aided radiation treatment planning. Med Phys 2000;27(7):1563-72. 125. Maruyama K, Kamada K, Shin M, Itoh D, Masutani Y, Ino K, Tago M, Saito N. Optic radiation tractography integrated into simulated treatment planning for Gamma Knife surgery. J Neurosurg 2007;107(4): 721-6. 126. Maruyama K, Kamada K, Ota T, Koga T, Itoh D, Ino K, Aoki S, Tago M, Masutani Y, Shin M, Saito N. Tolerance of pyramidal tract to gamma knife radiosurgery based on diffusion-tensor tractography. Int J Radiat Oncol Biol Phys 2008;70(5):1330-5. 127. Deogaonkar M, Walter BL, Boulis N, Starr P. Clinical problem solving: finding the target. Neurosurgery 2007;61(4):815-24; discussion 24-5. 128. Okun MS, Vitek JL, DeLong MR. Not our knife: gamma knife surgery for Parkinson disease. Arch Neurol 2002;59(8):1334-5; author reply 5. 129. De Salles AA, Frighetto L, Lacan G, Melega W. Radiosurgery can achieve precision needed for functional neurosurgery. Arch Neurol 2003;60(10):1494-6; author reply 6. 130. Sato S, Ohye C, Shibazaki T, Zama A, Cai X. Neurophysiological evaluation of the optimum target in gamma thalamotomy: indirect evidence. Stereotact Funct Neurosurg 2005;83(2–3):108-12; discussion 13-4. 131. Young RF, Shumway-Cook A, Vermeulen SS, Grimm P, Blasko J, Posewitz A, Burkhart WA, Goiney RC. Gamma knife radiosurgery as a lesioning technique in movement disorder surgery. J Neurosurg 1998; 89(2):183-93. 132. Kondziolka D, Ong JG, Lee JY, Moore RY, Flickinger JC, Lunsford LD. Gamma Knife thalamotomy for essential tremor. J Neurosurg 2008;108(1):111-7.
MRI in image guided surgery
133. Mathieu D, Kondziolka D, Niranjan A, Flickinger J, Lunsford LD. Gamma knife thalamotomy for multiple sclerosis tremor. Surg Neurol 2007;68(4):394-9. 134. Keep MF, Mastrofrancesco L, Erdman D, Murphy B, Ashby LS. Gamma knife subthalamotomy for Parkinson disease: the subthalamic nucleus as a new
39
radiosurgical target. Case report. J Neurosurg 2002; 97(5 Suppl):592-9. 135. Anschel DJ, Romanelli P, Benveniste H, Foerster B, Kalef-Ezra J, Zhong Z, Dilmanian FA. Evolution of a focal brain lesion produced by interlaced microplanar X-rays. Minim Invasive Neurosurg 2007;50(1):43-6.
617
48 Novel Therapies for Brain Tumors G. Al-Shamy . R. Sawaya
Introduction Approximately 18,000 new cases of primary brain tumors are diagnosed annually in the United States [1]. Despite significant advancements in neurosurgical techniques, radiotherapy, imaging modalities, and molecular neuro-oncology, there remains little improvement in clinical outcome for most patients. In fact, the median survival time of patients with malignant gliomas, particularly glioblastoma multiforme (GBM), the most common primary brain tumor, remains less than 1 year, with more than 90% of patients succumbing within 5 years of diagnosis [2]. Several factors are thought to underlie the lack of progress in developing effective treatments for malignant brain tumors. First, the central nervous system (CNS) presents a unique environment with limited capacity for self repair. The presence of the blood-brain barrier (BBB) further complicates systemic delivery of chemotherapeutic agents to CNS lesions. In addition, malignant brain tumor cells have unique innate properties that pose additional problems. These tumors are inherently aggressive, as highlighted by their remarkable degree of resistance to conventional therapies. Their widely infiltrating nature also makes them suboptimal candidates for surgical intervention. Finally, the lack of predictive preclinical models, coupled with our relatively poor understanding of glioma pathogenesis, has impeded the development of novel treatment options. The current ‘‘gold standard’’ for the management of malignant gliomas involves an attempt at complete or maximal safe surgical resection #
Springer-Verlag Berlin/Heidelberg 2009
(as specified in the National Comprehensive Cancer Network CNS guidelines [3]) in conjunction with radiotherapy and temozolomide (TMZ) chemotherapy, followed by six monthly cycles of TMZ. Although the above multimodality treatment regimen provides an additional improvement above the median survival time seen in patients treated without TMZ (14.6 and 12 months, respectively), clinical recurrence or progression is nearly universal [4,5]. Conventional treatment modalities have largely failed to yield significant progress in the treatment of malignant gliomas, despite extensive research efforts over the past two decades. This has led to increasing interest in exploring alternative treatment strategies, including tumor-specific immunotherapy, gene therapy, molecularly targeted therapeutics and tumor oncolysis. Additionally, recent advances in the knowledge of basic tumor biology have led to a paradigm shift in the way brain tumors are being studied, allowing for novel approaches to understanding gliomagenesis and glioma therapeutics. Strategies to overcome the therapy-limiting properties of the CNS, including attempts at improving delivery of therapeutic agents via targeted delivery systems, have now become a major focus of brain tumor research. The aim of this chapter is to consolidate the various recently developed novel glioma therapies highlighted above, along with several others. We will not only examine our current understanding of the molecular events underlying malignant glioma pathogenesis but also focus on the history, evolution, and clinical implementation of new treatment modalities. The purpose of this review
750
48
Novel therapies for brain tumors
is to provide the reader with a broad perspective on the various applications of novel brain tumor therapeutics.
Direct Delivery of Chemotherapeutic Agents In light of the limitations posed by the BBB in systemic drug delivery, direct intracranial delivery of chemotherapeutic agents presents a unique approach for treating malignant gliomas by obviating the need for a drug to cross the BBB and limiting the toxicity of systemic anticancer agents. This technology makes it possible to achieve very high local concentrations of chemotherapeutic agents at the tumor site, while avoiding the side effects associated with systemic administration of extremely high doses. The two main approaches that have yielded the most promising results are polymerically-controlled release and convectionenhanced delivery. Each system has its own advantages and disadvantages, as detailed below.
Polymerically-Controlled Release In an effort to improve local control after brain tumor resection, researchers have developed controlled-release polymers that are implanted directly at the resection site to allow for restricted slow release of chemotherapeutic agents into the tumor bed. This system is particularly useful because it can provide reliable sustained drug release for periods of days, or even years [6]. Constructed from either biodegradable or nonbiodegradable polymers, the polymerically-controlled release system depends on diffusion of the drug through the polymer matrix. Biodegradable polymer systems have been considered more attractive for clinical application because they do not require removal at a later date. Importantly, the diffusion kinetics of these systems have been well characterized [7], and the use of these systems has allowed the
emergence of a number of clinical studies exploring alternative treatment strategies, including gene therapy and tumor-specific immunotherapy, with further applications under investigation. The most common type of polymer currently used intracranially is poly[bis(p-carboxyphenoxy) propane-sebacic acid]. This polymer has achieved the best success when loaded with nitrosourea (BCNU) in the form of Gliadel®. A randomized, placebo-controlled, double blind, prospective, phase III clinical trial demonstrated an additional median survival benefit of 8 weeks with the use of Gliadel® polymers in patients with recurrent GBM [8]. This study failed to identify any local or systemic adverse effects attributed directly to Gliadel®. Several additional studies have since evaluated the effectiveness and safety of Gliadel® in the initial treatment of malignant gliomas. A prospective, randomized double blind clinical trial in Europe [9] reported a statistically significant benefit in survival with Gliadel® relative to placebo in patients with GBM (53 and 40 weeks, respectively; p = 0.0083). Therapy with Gliadel®, even when combined with radiotherapy, has been shown to be well tolerated, with no significant increase in toxicity, infection, or inflammation [10]. This method of delivery has proved to be particularly attractive clinically in that it adds minimal complexity, requires no additional surgical procedures, and minimizes solubility limitations. A main limitation of polymerically-controlled release, however, is that local penetration of the drug is frequently hindered by diffusion [11]. In addition to the local environment of a tumor, the extracellular matrix of the brain parenchyma also imposes its own limits on diffusive transport. Together these factors confine high drug concentrations to within approximately 3 mm of the delivery site [12]. The poor drug penetration and the consequent dependence of the drug dosage on implant size has therefore hindered the therapeutic benefits of polymerically-controlled release, despite all of the advantages of the system.
Novel therapies for brain tumors
Convection-Enhanced Delivery Convection-Enhanced Delivery (CED) is an alternative method of controlled local drug release that has been developed to deliver compounds throughout the brain so as to overcome the diffusion barrier seen with polymerically-controlled release systems [13]. CED utilizes an applied external positive pressure infusion to induce fluid convection in the brain and thus force chemotherapeutic drugs throughout the parenchyma via interstitial spaces. Fluid is typically administered via a small catheter using a pump [14]. The main benefit of administering drugs by CED is a greater distribution volume and a longer infusion time, allowing for continued drug exposure. An additional benefit arises from the flexibility of the technique and its ability to be used with gene therapy as well as with immunotherapeutics. Multiple therapeutic agents have been evaluated in clinical trials for glioma treatment using this method [15]. Perhaps the most promising agent investigated thus far is IL13-PE38QQR (cintredekin besudotox). This synthetic drug is derived from a human protein, interleukin 13 (IL13), linked to a bacterial toxin, Pseudomonas exotoxin (PE). The IL13 portion of the drug is able to bind to specific receptors on the tumor cells, allowing for a form of targeted therapy. A recent phase III double blind, randomized controlled trial compared the clinical outcome between implantation of Gliadel® wafers containing BCNU (carmustine) and CED using IL13-PE38QQR in the treatment of patients with recurrent GBMs. The median survival times were comparable between the two modalities, though there was some suggestion of additional benefit with CED using IL13-PE38QQR. An additional advantage of the use of CED is its ability to permit the monitoring of the distribution of chemotherapeutic agents and to thus provide some measure of controlled delivery. Furthermore, through the use of gadolinium liposomes, the infusion of particles can be monitored
48
directly using magnetic resonance imaging (MRI). More recently, this imaging method has been validated in a primate model [16]. Just as with polymerically-controlled release systems, CED has its own shortcomings and complications. Perhaps most importantly, it is an invasive procedure that has the potential for inducing high intracranial pressures. It also has the disadvantage of having an unpredictable drug distribution [17]. Current research is aimed at developing better methods to track infused agents and to optimize the efficacy of CED for brain tumor therapeutics. Interstitial drug delivery of chemotherapeutic agents does provide an effective means of bypassing the BBB, minimizing systemic toxicity, and producing high concentrations of the drug at the site of interest. Nontheless, clinical trials have shown only a modest improvement in survival time with this method when compared with conventional therapy, owing to inherent limitations of the technology. Further development may allow this unique approach to increase the efficacy of novel antitumor therapeutics and conventional multimodality therapies and to thereby potentially improve the survival benefits in the clinic.
Molecularly Targeted Therapy The recent advances in our understanding of the molecular biology of brain tumors and the discovery of various dysregulated cell signaling pathways found in glioma cells (> Figure 48-1) have set forward a number of unconventional targets that could be used in the treatment of malignant gliomas. A number of approaches aimed at inhibiting these signaling pathways, ranging from upstream growth factor ligands and their receptors to downstream intracellular effectors, have been explored. To date, the targets identified include vascular endothelial growth factor (vEGF), platelet-derived growth factor (PDGF), agents targeting components of the Ras- and AKT (proto-oncogene
751
752
48
Novel therapies for brain tumors
. Figure 48-1 A schematic overview of the major signaling pathways involved in glioma progression, including those involving the epidermal growth factor receptor/phosphatidylinositol-3’ kinase (EGFR/PI3K), platelet-derived growth factor (PDGFR), protein kinase C (PKC), and vascular endothelial growth factor receptor (vEGFR). Inhibitors currently in clinical trials are highlighted in red at the various stages of the pathways on which they act. (Abbreviations: mTOR, mammalian target of rapamycin; PTEN, phosphatase and tensin homolog)
protein)-mediated pathways, and the human epidermal growth factor receptor (HER2/neu). Each of these molecules has been studied extensively and found to play a critical role in tumorigenesis and disease progression [18].
Human Epidermal Growth Factor Receptor The epidermal growth factor receptor (EGFR), a member of the tyrosine kinase receptor family, is
Novel therapies for brain tumors
a cell-surface transmembrane protein that has been implicated in cell growth and proliferation, as well as in cell survival, motility, and resistance to chemotherapy and radiation therapy [19]. The EGFR gene is located on chromosome seven and encodes a 170 kDa transmembrane glycoprotein with intrinsic tyrosine kinase activity [20]. In GBMs, the EGFR is both overexpressed (approximately 50% of tumors have chromosome seven EGFR locus amplification) and mutated (by deletion of exons 2–7), giving rise to a ligand-independent, constitutively activated form called EGFRvIII [20,21]. EGFRvIII strongly and persistently activates the phosphotidylinositol-3’ kinase (PI3K) survival and antiapoptotic pathway, a signaling cascade that is essential in providing radioresistance to glioma cells [22,23]. More recently, a study by Pelloski and colleagues [24] has also suggested that having an EGFRvIII-positive GBM is an independent prognostic factor for poor survival outcome. Recently, three EGFR inhibitors (> Figure 48-1) have been approved by the FDA for use in clinical trials. The results with gefitinib (Iressa), the first of these inhibitors, were largely disappointing. A phase II study using gefitinib as the sole agent for treating recurrent tumors showed no radiographic tumor regression [25]. In a second study, 98 newly diagnosed GBM patients receiving gefitinib in conjunction with radiation therapy failed to show any clinical improvement or survival benefit [26]. Importantly, in both of these studies, there was no relation between clinical outcome and degree of EGFR expression. The results with erlotinib (Tarceva), a related drug, have been more variable. A number of phase II trials showed tumor regression, with 6-month progression-free survival (PFS) rates ranging from 17 to 27% [27,28] in patients treated with Tarceva. Promising results have also come from a phase I trial combining Tarceva with TMZ [29]. However, these results failed to be duplicated when Tarceva was added to the standard protocols for patients newly diagnosed with GBM [30]. Cetuximab, the
48
third EGFR inhibitor agent (a monoclonal antibody), is currently being studied as a possible treatment in patients with recurrent GBM [31]. Two possibilities have been suggested for the relative ineffectiveness of EGFR inhibitors. One of these has to do with the complexity of cellular signaling and the interactions of various downstream molecules and pathways. That is, inhibition of the receptor alone may not be sufficient to mitigate downstream signaling, and additional inhibition of downstream effectors may be necessary. In fact, the protein kinase B (PKB)/AKT signaling pathway has been of interest recently, as its inhibition significantly increases the effectiveness of EGFR inhibitors. Studies combining EGFR inhibitors with rapamycin, an inhibitor of the PKB/AKT signaling pathway, have yielded promising results [32,33]. The second potential reason for the lack of efficacy of EGFR inhibitors stems from the fact that 40% of the patients with EGFR overexpression have the mutated EGFRvIII. This mutation bypasses the need for ligand binding and yields a constitutively active receptor even in the absence of growth promoting signals and factors. Consequently, recent work has focused on the development of a vaccine specific to EGFRvIII.
Vascular Endothelial Growth Factor (vEGF) and Receptor (vEGFR) A characteristic feature of GBM is exuberant angiogenesis, a key event in tumor growth and progression. One possible mechanism underlying tumor neovascularization may involve the recruitment of native blood vessels to the tumor through expression of hypoxia-inducible factor (HIF)-1 alpha and vascular endothelial growth factor (vEGF) in perinecrotic pseudopalisading glioma cells. VEGF is a secreted peptide that acts through its receptors, FLT1 and FLK-1-KDR (vEGFR2), to stimulate endothelial cell division and the formation of new blood vessels [34].
753
754
48
Novel therapies for brain tumors
Recently, Avastin (bevacizumab), a recombinant human antibody specifically designed to block vEGF, was first used for patients with recurrent brain tumors in combination with CPT 11 (irinotecan), a camptothecin-derived anticancer agent with DNA topoisomerase one inhibitory activity [35]. Although the initial results were very encouraging (after the initial course of treatment, significant tumor regression was observed), long-term survival data proved to be disappointing. A number of recent trials studying the effect of Avastin and CPT 11 on malignant gliomas suggest that patients treated with this combination show a rapid initial response followed by a rapid regrowth of the tumor thereafter [36]. Taken together, the results of these studies demonstrated a median survival interval similar to that seen with conventional therapy, but with the added complications of increased risk of internal bleeding and blood clots. Vatalanib, a vEGFR inhibitor, has also been evaluated recently for treatment of patients with gliomas in a multicenter phase I/II trial, both alone and in combination with chemotherapy. The results in both cases were equivalent, with 66% of patients showing stable disease after treatment [37]. More recently, a phase II trial of AZD2171 (Astrazeneca Pharmaceuticals), a vEGFR2 inhibitor, has been initiated in the treatment of recurrent GBM [38,39].
tumor growth by indirect mechanisms pertaining to the tumor stroma or vasculature [45–47]. Gleevec (STI-571, imatinib mesylate), an inhibitor of the bcr-abl tyrosine kinase involved in the growth of chronic myelogenous leukemia, has recently been shown to inhibit glioma growth in animal studies [48]. The reason lies in the large sequence homology and biochemical similarity between the tyrosine kinase domain of the bcrabl protein and the PDGFR. Yet, these results could not be repeated in clinical studies. As a monotherapy, Gleevec had minimal success, failed to show any significant clinical benefit, and was associated with increased risk of intracranial hemorrhage [49]. Additional preclinical studies have suggested that combining Gleevec with hydroxyurea may provide additional cytotoxic effects. In vitro studies have demonstrated that Gleevec increases the chemo- or radiosensitivity of GBM cells in culture [50–52], suggesting that it may act to enhance the activity of chemotherapeutic agents currently used to treat GBM. Hydroxyurea, a cytotoxic agent that inhibits DNA synthesis, is widely used in cancer therapy and penetrates the BBB [53,54]. More recently, a study combining Gleevec and hydroxyurea demonstrated encouraging antitumor efficacy, with a reported 6-month PFS rate of 32% [55]. This result was later confirmed in a phase II study [56].
Protein Kinase C Pathway Platelet-Derived Growth Factor Receptor A number of preclinical studies have provided considerable support for the hypothesis that autocrine signaling by platelet-derived growth factor (PDGF) plays a key role in gliomagenesis [40–42]. These findings suggest that inhibition of PDGF receptors (PDGFRs) may be able to arrest GBM progression by disrupting autocrine signaling and, consequently, the glioma cell cycle [43,44]. Alternatively, inhibition of PDGFRs may affect GBM
Protein kinase C (PKC) is a serine/threonine kinase that has been shown to regulate tumor cell proliferation, migration/invasion, and angiogenesis [39]. Tamoxifen, an antiestrogen medication historically used in the treatment of breast cancer, plays a role in inhibiting the enzymatic reaction of PKC. One stage II clinical trial reported a considerable survival benefit associated with the administration of oral tamoxifen to patients with recurrent gliomas [57]. Tumor regression was reported in 25% of patients and
Novel therapies for brain tumors
stabilization of tumor growth for an additional 20%. Despite the reported benefits, however, this modality of therapy did pose some potential problems. The dosage of tamoxifen used in this study was significantly higher than the dose typically used in breast cancer patients, which poses an increased risk of blood clots, weight gain, and uterine cancer in women and decreased libido in men as common side effects. Nevertheless, the capacity of tamoxifen to inhibit the chemoresistance of brain tumor cells made it a suitable agent to use in combination with traditional chemotherapeutic approaches. Tamoxifen has thus been studied in combination with carboplatin as well as BCNU for the treatment of gliomas. Mastronardi and colleagues studied the role of tamoxifen and carboplatin for patients with newly diagnosed brain tumors. This group reported a 1- and 2-year survival rate of 52 and 32%, respectively [58]. Unfortunately, this result could not be replicated in later studies. Tamoxifen has also been evaluated in combination with BCNU as the initial treatment after radiation in a number of phase II and phase III clinical studies [59,60], with variable results. Some studies reported long-term survival effects with tamoxifen, and others reported no significant increase in overall median survival. This controversy was recently explained by a Canadian study using magnetic resonance spectroscopy. The authors demonstrated that tamoxifen appears to work on only a minority of patients and that one can predict which patients will respond to it based on the presence of different metabolites [61].
48
of the second messenger molecule PIP3. Consequently, PTEN inactivation results in accumulation of PIP3 in cells and renders the PI3K pathway constitutively active via activity of the serine/threonine kinase AKT/PKB cascade downstream of PI3K. Interestingly, PTEN loss has also been shown to promote resistance to EGFR kinase inhibitors by dissociating EGFR/ EGFRvIII inhibition from downstream inhibition of PI3K signaling cascades. Overall, loss of PTEN and constitutive activation of PI3K signaling results in promotion of antiapoptotic effects, cell growth and proliferation. In glioma patients, activation of PI3K is associated with poor outcome [63]. Consistent with this, increased levels of AKT and PKB phosphorylation downstream of PI3K have been clinically associated with resistance to Tarceva in malignant glioma patients. None of the tumors expressing high levels of PKB/AKT responded to Tarceva administration, whereas 8 of 18 tumors with low PKB/AKT levels responded to treatment. The mammalian target of rapamycin (mTOR) is another serine/threonine kinase found downstream of PI3K. Rapamycin and its synthesized analog, temsirolimus, inhibit mTOR signaling and have been evaluated for the treatment of recurrent gliomas. Phase II studies with temsirolimus have failed to show any survival benefit [64,65]. Perifosine, an oral AKT inhibitor, is currently undergoing clinical evaluation in patients with malignant gliomas. Despite all of these negative results, molecularly targeted therapy offers great promise.
The Role of Immunotherapy PTEN Signaling Pathway The phosphatase and tensin homolog (PTEN) is a phosphatase that inhibits PI3K and is lost in 50% of patients with GBM [62]. PTEN has been shown to negatively regulate the phosphatidylinositol-3-kinase (PI3K) pathway by removing a phosphate from the inositol ring
Brain tumors have long been considered to arise at a site that is privileged with respect to immune surveillance. A number of studies have demonstrated that gliomas express immunosuppressive characteristics (> Figure 48-2), both locally [66] and systemically [67]. Gliomas have been shown to be associated with overall lymphopenia,
755
756
48
Novel therapies for brain tumors
. Figure 48-2 A schematic representation of the interaction of glioma cells with the native immune system. By employing various signaling mechanisms, glioma cells are able to evade the immune response and proliferate within the brain. A number of strategies currently being investigated are aimed at enhancing the immune response as an effective approach to cancer treatment. (Abbreviations: GM-CSF, granulocyte/macrophage-colony stimulating factor; IL-2, interleukin-2; TGF-ß, transforming growth factor beta)
depressed antibody production, and impaired antigen-presenting cell (APC) function [68–70]. Additionally, patients with gliomas present with depressed peripheral T-cell responsiveness as well as depressed T-cell receptor-mediated signaling [71]. Taken together, these studies strongly suggest that weakening the immune system is a fundamental characteristic of the malignancy of GBM. A natural corollary to this idea would be that strengthening one’s immune system may be an effective approach to cancer treatment. This concept has been further validated in experimental rodent model systems of intracranial gliomas, where boosting otherwise impaired tumor-specific immune responses can successfully eradicate wellformed neoplasms [72,73]. Overexpression of transforming growth factor (TGF)-ß2 is thought to be a primary contributing factor to the immunosuppressed state seen in patients with
gliomas and, consequently, a significant cause of the failure of current immunotherapeutic strategies [74]. To combat this, Gorelik and colleagues [75] recently demonstrated that with T-cell-specific blockade of TGF signaling, an effective immune response could be generated that was capable of eradicating tumors in mice. A similar strategy was employed by Ruffini and colleagues, who demonstrated that by effectively antagonizing TGF-ß2 secretion with a neutralizing antibody, the in vitro proliferative capacity and antitumor cytotoxicity of adherent lymphokine-activated killer (LAK) cells could be enhanced [76]. More recently, treatments that involve tumorspecific immune reactions have received great attention because of their high benefit-to-risk potential. Both passive and active immunotherapeutic approaches have been attempted, both alone and in combination, in patients with gliomas.
Novel therapies for brain tumors
Passive Immunotherapeutic Approach Receptor/Antigen Targeting Recent advances in the understanding of the underlying molecular biology of brain tumors have led to the discovery of cell-surface epitopes specific to cancer cells. Developing antibodies against these epitopes offers a potential way to selectively target and eradicate neoplastic cells dispersed within normal brain tissue and is a very attractive means of eliminating all intracranial neoplastic foci left behind after surgical resection of the primary tumor mass. Tenascin, an extracellular matrix protein with unknown function, is one of the recently discovered targets that has been studied in clinical trials. It has been detected in almost all highgrade gliomas [77], and monoclonal antibodies specific to it have been generated successfully. The BC2 form of the antibody has been conjugated to radioactive iodine-131 (I-131) and infused directly into the tumor cavity after resection of both newly diagnosed and recurrent high-grade gliomas [78–80]. Radioimmunotherapy was used in the above trials in order to provide an additional radiation boost to the tumor bed without incurring the known toxicity effects of external beam irradiation. More recently, a phase II study using an I-131-labeled antitenascin monoclonal antibody in patients with recurrent brain tumors reported a significantly better median survival time with this method compared with conventional therapy [81]. This encouraging result has been reproduced in a different study but with a higher incidence of hematological and neurological toxicity [82]. A variation on this approach involves the use of interleukin-13, which is conjugated to Pseudomonas endotoxin, a bacterial toxin that has been shown to be lethal to glioma cells in previous studies. A phase I/II study employing this agent
48
to treat recurrent brain tumors recorded a median survival interval of 46 weeks [83]. Another variation on this approach involves transferrin-CRM, which is conjugated to a modified diphtheria toxin. Using this conjugate, a phase I study reported an encouraging result, with significant tumor regression in more than 50% of patients treated [84]. A subsequent phase II study was able to reproduce this positive result, with 35% of patients reported to have tumor regression and a median survival time of 37 weeks [85].
Cytokine/Modulator Therapy Cytokines play a critical role in the induction, stimulation, and promotion of the immune response. Consequently, they have received significant attention in research as a possible tool to potentiate the immune response and induce cell-mediated antitumor immunity. A number of different studies have used different approaches for the treatment of gliomas, including systemic, intrathecal, and intratumoral administration of cytokines. One of the first cytokines to be studied in gliomas was interleukin-2 (IL-2), an important T-cell growth factor that has been implicated in the proliferation of CD8 + T cells and the modulation of their cytotoxic activity. A number of early in vitro studies supported a rationale for the use of IL-2 in patients with malignant brain tumors. This idea was largely based on the ability of IL-2 to block the T-cell depressing activity of TGF-ß, a key mediator of the suppressed immune response observed in glioma patients [86,87]. However, the use of IL-2 proved to have several important limitations, including not only being poor at blocking the effects of TGF-ß but also producing increased vascular permeability leading to brain edema. Several clinical trials have also studied the role of interferons alpha and beta for treating patients with recurrent gliomas. The primary
757
758
48
Novel therapies for brain tumors
constraints in these studies were the lack of reproducibility as well as poor study design [88,89]. Nevertheless, more recent studies have described novel approaches for the delivery of these cytokines. These include techniques that are cell based, gene therapy based, and neural stem cell based. These methods allow for the robust delivery of tumoricidal cytokines directly to neoplastic regions, thereby circumventing the limitations of systemic therapy.
Immunological Cell Transfer In the normal immune system, immune surveillance is regulated by natural killer (NK) cells and to a lesser degree by cytotoxic T cells. In a neoplastic setting, this same mechanism is employed to eliminate newly formed neoplastic cells, with the goal of preventing their propagation and consequently hindering the formation of clinically detectable masses. Recently, a new approach called adoptive cell transfer has been developed to try to harness the cytotoxic abilities of T-cell populations by transplanting potentially tumoricidal T cells into patients with malignant brain tumors. Delivery methods can vary from systemic infusion to local intracerebral inoculation. To date, use of LAK cells has produced the most promising results. LAK cells are peripherally circulating populations of lymphocytes that have been shown to be capable of lysing NK cellresistant tumor cells in vitro after exogenous stimulation with IL-2 [90]. For adult patients with recurrent malignant gliomas, Hayes and colleagues have reported an improved long-term survival time following IL-2 and LAK cell infusion into the tumor cavity after resection [91]. Unfortunately, as with many of the other methods detailed thus far, these results have not been reproducible and are complicated further by side effects resulting from administration of IL-2. Additional significant limitations to treatment
using T-cell populations that have been expanded and sensitized in vitro are their lack of specificity and their inability to retain memory (unlike normal immune cells in the body).
The Role of APCs/Dendritic Cells Perhaps the most critical step in inducing tumor antigen-specific immunity involves devising ways to effectively deliver tumor antigens to T cells and developing methods to enhance detection of tumor antigens. Currently, these are both areas of intense research in tumor biology. The basic understanding of normal immunological principles has greatly underscored the function of these cells in a neoplastic setting. The presentation of antigens to naı¨ve T cells triggers a cascade of events comprising an active immune response that includes clonal expansion, the formation of memory cells, and finally, cytolytic action. To date, the most potent APC has been recognized to be the dendritic cell. Dendritic cells (DCs) are bone marrow-derived cells that can be isolated in cultures of peripheral blood mononuclear cells. Briefly, large numbers of DCs can be obtained by stimulating bone marrow cultures with granulocyte/macrophage-colony stimulating factor (GM-CSF). These DCs are capable of presenting both endogenous and exogenous antigens to naı¨ve T cells in an HLA-restricted manner [92]. Findings from a number of preclinical animal model systems have suggested that immunizing mice or rats with DCs pulsed with tumor antigens is sufficient to prime a cytotoxic lymphocytic response that is specific to tumor antigens and consequently can provide protective immunity against CNS tumors [93–95]. The first study to use a DC-based vaccine to target malignant gliomas was performed by Heimberger and colleagues [93]. In this work, bone marrow-derived DCs were pulsed with a specific subunit of the constitutively active EGFRvIII mutation. Immunized DCs were then administered to
Novel therapies for brain tumors
mice intraperitoneally once a week for 3 weeks. Importantly, even after being challenged with tumor antigens more than once, the group receiving DC injections showed increased survival, demonstrating the development and induction of immunologic memory [93]. A follow-up study employed a tumor model system that mirrored the antigenic properties of spontaneous human gliomas. After being pulsed with tumor-specific antigens, DCs were administered to mice intraperitoneally each week, as in the above experiment, for a 4-week series. Animals were later challenged with intracranial injections of tumor cells. According to this study, mice immunized with DC injections exhibited a 160% increase in survival time; moreover, more than 50% of the animals showed long-term survival. Finally, mice rechallenged with tumor also exhibited enhanced survival, once again demonstrating the induction of immunological memory [94]. Based on the apparent successes observed in the animal studies above, Yu and colleagues [96] embarked on a phase I clinical trial administering a DC-based vaccine to glioma patients. In this trial, patients received autologous peripheral DCs prepared with IL-4 and GM-CSF stimulation after a brief exposure to peptides eluted from the surface of autologous glioma cells. Despite the small sample size, this study reported that patients receiving DC vaccinations showed a prolonged survival time compared with age- and sex-matched controls, indicating that DC vaccination may confer some survival benefit. This study conclusively demonstrated the safety, feasibility, and biological activity of the procedure, with evidence of an increased intratumoral cytotoxic CD8 + and memory T-cell (CD45RO+) infiltration. Another phase I trial investigating the feasibility of this procedure was conducted by Liau and colleagues at UCLA [97]. Twelve patients with histologically confirmed GBM received three biweekly intradermal injections of 1, 5, or 10 million autologous DCs, pulsed with the same amount (100 mg per injection) of acid-eluted
48
autologous tumor peptides. In this study, DC vaccinations were not associated with any evidence of dose-limiting toxicity or serious adverse effects. Additionally, one patient exhibited an objective clinical response documented by magnetic resonance imaging (MRI), and six patients developed significant systemic antitumoral cytotoxic T-lymphocyte responses, but these did not consistently translate into objective clinical responses or correlate with increased survival times. Based on the results of these studies, a phase II clinical trial is currently under way.
Gene Technique-based Immunotherapy More recently, advances in recombinant gene technology have opened the doors for the development of novel tumor-specific therapies. Genetic constructs can routinely be modified by viral vectors (or, alternatively, plasmid DNA) to express a variety of genes encoding tumor antigens, cytokines, or accessory molecules. In this context, genetically modifying tumor cells could increase their immunogenicity and potentially enhance the systemic immune response generated against an intracranial tumor. In principle, genetic modification can also lead to tumor cell death via induction of immunological cascades. Viruses, especially adenoviruses, have been pivotal in the application and development of these recombinant approaches, as these organisms can efficiently deliver their genome into eukaryotic cells and then transcribe and translate it within them. By replacing a part of their genome, scientists can readily convert adenoviruses from lytic pathogens to lysogenic vectors for gene delivery. Among the different genes whose delivery has been attempted by this approach, only a handful have been studied in clinical trials. These include p53, the herpes simplex virus-thymidine kinase (HSV-tk) gene, and the genes for interferons alpha and beta.
759
760
48
Novel therapies for brain tumors
Adenovirus-Mediated p53 Gene Therapy The gene for the p53 tumor suppressor is located on the short arm of chromosome 17 and is one of the most frequently mutated genes in human gliomas [98]. p53 plays a critical role in cell cycle arrest and apoptosis, and its loss is a key factor in the development of glial neoplasms [99]. The North American Brain Tumor Consortium (NATBC) recently completed a phase I trial of adenovirus-mediated p53 gene therapy in 12 patients with recurrent malignant gliomas [100]. Patients first underwent stereotactic surgical implantation of a catheter into the center of tumor mass. Ad-p53, an adenovirus (type 5) in which the E1 coding region is replaced with p53 cDNA, whose expression is driven by a cytomegalovirus promoter, was delivered via this catheter. Three days after infusion of the virus, patients underwent open craniotomy, and the tumor and catheter were resected en bloc. After tumor resection, Ad-p53 was directly injected into the post-resection tumor bed, and the craniotomy was subsequently closed. Despite the serious limitation of having a narrow distribution of the therapeutic gene in the brain parenchyma, the results of this study demonstrated the feasibility and safety of such a procedure. Clinical toxicity was minimal, and time to recurrence within the treated group was 7 months. The problem of poor tissue penetration could potentially be circumvented, while also increasing the efficacy of adenoviral vectors, by using replication-competent vectors. This technique can not only exert a cytolytic effect on infected cells but also allow for subsequent propagation of the virus to neighboring cells, resulting in additional tumor/ tissue penetration. The main limitation of these vectors, however, is that the host’s normal brain cells must be spared. One potential strategy to avoid this problem is to produce mutant E1 genes, leading to the development of conditionally replicative adenoviruses [101]. An example of
this is ONYX-015, a conditionally replicative adenovirus capable of specifically affecting tumor cells. ONYX-015 contains a mutation in the p53binding protein E1B-55K, thus rendering it unable to transfect normal cells, which have normal p53 proteins. A recent phase I study of 24 glioma patients demonstrated the safety of this technique but failed to conclusively prove the specificity of the drug for glioma cells [102]. Recently, a new oncolytic adenovirus called delta 24 has been developed in an attempt to increase the specificity of these vectors for tumor cells. Delta 24, an oncolytic virus with a 24-base-pair deletion in the viral E1A gene, results in selective replication in cells harboring a mutation in the retinoblastoma (Rb) protein or its regulatory pathway [103]. In animal studies, delta 24-RGD has been shown to be effective against gliomas [104]. Importantly, the pathways involved in adenovirus-mediated cell death remain unclear. Delta 24 appears to induce the formation of acidic vesicular organelles, a process termed autophagy, in both in vitro and in vivo models [105]. More recently, two independent studies reported that the combined use of oncolytic adenovirus delta 24-RGD and chemotherapeutic agents results in an enhanced antiglioma effect in vivo [106,107].
Herpes Simplex Virus-based Gene Therapy A recently developed gene therapy model that has received much attention over the past decade is so-called suicide gene therapy. This technique is based on the transduction of tumor cells with herpes simplex virus-thymidine kinase (HSV-tk) in the presence of gancyclovir as a prodrug. Over the past few years, the method of specific gene delivery as well as the understanding of the biology of the suicide effect has evolved tremendously. Initial studies on the transfer of HSV-tk depended largely on the use
Novel therapies for brain tumors
of recombinant-deficient retroviruses (RVs). RVs were generated and surgically released within the tumor bed by injection of genetically modified mouse fibroblasts (virus producer cells, or VPCs) [108,109]. In vitro experiments and in vivo animal tumor model systems of glioma have demonstrated the feasibility of RV-mediated gene transduction and the ability of this method to kill glioma cells by insertion of toxicity-generating transgenes. In spite of these encouraging reports of effective RV-mediated gene therapy in experimental animal model systems, clinical trials in glioma patients have been largely disappointing. Phase I and II clinical studies in patients with recurrent malignant gliomas have demonstrated some efficacy of RVmediated gene therapy in addition to a favorable safety profile [110,111]. However, a phase III study of adjuvant gene therapy in 248 GBM patients failed to report any survival benefit with administration of HSV-tk retrovirus VPCs [112]. The failure of RV gene therapy systems prompted researchers to investigate the development of alternative therapeutic viral vector systems to increase efficacy and yield better transduction efficiency. One of these methods utilizes replication-competent vectors [113]. Such vectors permit higher transduction rates in addition to enhanced cytolytic effects on transduced cells. Animal studies using replication-competent vectors have yielded promising results [114]. Additionally, our better understanding of the bystander cell death theory has begun to shed further light on the limitations of the RV gene therapy system. After treatment with the antiviral drug gancyclovir, untransfected tumor cells adjacent to transfected cells are killed as well. This bystander effect is thought to be mediated by gap junction intercellular transfer of toxic phosphorylated gancyclovir molecules by connexin protein family members [115,116]. In fact, several groups have suggested that the combined activity of HSV-tk and connexin 43 could enhance gene therapy [117].
48
The Role of Antisense Therapy The last decade has fostered the use of a novel and extremely potent method of experimentally manipulating gene expression known as RNA interference (RNAi and shRNA) [118,119]. RNAi is a process whereby double-stranded RNA (dsRNA) sequences target messenger RNA (mRNA) for destruction in a sequence-dependent manner. These small dsRNAs are delivered into target cells as short hairpin precursors (shRNA) by expression plasmids. DNA expression plasmids are encapsulated inside liposomes, which serve as a safeguard to protect DNA from ubiquitous degradation in vivo. The surface of the liposome has specific molecules conjugated to it that will bind endogenous receptors, inducing receptor-mediated transcytosis through the BBB, as well as transport by endocytosis to the nuclear compartment of brain cells [120,121]. Consequently, this method enables the efficient expression of plasmid DNA in brain tumor cells after intravenous administration of the gene. Once inside the cell, shRNA is processed by an enzyme called Dicer into active 21-nucleotide RNA fragments that can recognize target mRNA sequences via base-pairing interactions. The end result of this process is the suppression of specific genes at the posttranslational level. Several genes implicated in the malignant growth of gliomas have been studied as potential targets for gene therapy using RNA interference. One of these genes is TGF-b2, known to play a pivotal role in tumor progression by regulating key mechanisms such as proliferation, metastasis, and angiogenesis. One group in Germany recently developed a specific antisense oligonucleotide against TGF-b2 (AP12009). After demonstrating the safety and efficacy of this fragment in vitro, the researchers set out to compare the effect of AP12009, as administered by CED, with that of chemotherapeutic agents (TMZ or PCV, combined procarbazine, lomustine, and vincristine) in a phase II trial of patients with malignant
761
762
48
Novel therapies for brain tumors
recurrent gliomas. They reported improved survival times in the former group, with some patients exhibiting long-term total remission. Thus, both the preclinical and the clinical results implicate targeted TGF-b2 suppression as a promising therapeutic approach for malignant tumor therapy [122,123]. The EGFR, another key regulator of the pathogenesis of brain tumors, has also been studied extensively as a possible target for silencing using RNA interference. In an experimental human brain tumor model system in mice with severe combined immunodeficiency, weekly intravenous injections of RNAi directed against the EGFR led to reduced expression of EGFRs within the tumor and an 88% increase in survival time of these mice [124]. The effectiveness and safety of this technology demonstrated in preclinical and clinical trials suggests it to be a promising method for further study in human clinical trials. One particular avenue of interest might involve designing an RNAi-based method to suppress both wild type and mutant EGFRvIII mRNAs.
Summary The past two decades have seen tremendous strides in cancer biology research. The discovery of signaling cascades mediating tumor cell growth and propagation has allowed for a deeper understanding of the basic biology of tumors and revealed a series of mutations that may directly or indirectly be involved in the pathogenesis of brain tumors. New studies aimed at dissecting the complex delicate interactions across a number of signaling pathways in the pathogenesis of gliomas has led to the rational development of different targeted therapies against one or a number of these pathways. Some of these therapies include the use of novel chemotherapeutic agents, immunotherapy, or viral-mediated gene therapy.
In general, preclinical studies have shown promising results with a number of these modalities. The great challenge, however, has been to translate these results into significant improvement of the clinical outcome in patients. A major problem lies in the fact that most clinical trials use these novel agents as monotherapies as opposed to adding them to existing protocols as adjuvant therapies. Consequently, most of these agents have failed to demonstrate survival benefit in unselected patient populations. The greatest limitation, however, remains the striking hetereogeneity of human gliomas, both in terms of cellular phenotype and behavior and with respect to the vast number of genetic aberrations seen, which can vary widely among individuals and sometimes even within the same tumor. Coupled with the confined delivery of drugs to the tumor bed, this is the single most significant problem underlying therapeutic failure. Despite all the advances highlighted in this chapter, prognosis for GBM remains poor and survival outcomes dismal. Future success in achieving improved targeted therapies depends on the proper segregation of patients based on the major mutations and molecular profile of their respective tumors. If scientists can better identify correlative biomarkers, one can envision ‘‘individualized’’ targeted therapy based on predictive molecular or genetic signatures of individual tumors from each patient. Consequently, this will allow therapy to be targeted against an individual mutation or, more broadly, tailored to individual patients. Strategies that might be employed in order to achieve this include genomic analysis to identify new targets and promising treatment combinations, in addition to methods to improve therapeutic delivery systems. The challenge lies in being able to unify our understanding of tumor biology with a keen grasp of the interactions between numerous therapeutic modalities and signaling pathways. Our success in combating these tumors will rely on our ability to effectively implement a multimodality paradigm including surgery, radiotherapy, and a form of targeted therapy.
Novel therapies for brain tumors
Acknowledgments We thank David M. Wildrick, Ph.D., for editorial assistance with the manuscript.
References 1. Ries LAG, Eisner MP, Kosary CL, et al. (eds.). SEER cancer statistics review, 1975–2002 (http://seer.cancer. gov/csr/1975_2002/). Bethesda, MD: National Cancer Institute; 2005. 2. Deorah S, Lynch CF, Sibenaller ZA, Ryken TC. Trends in brain cancer incidence and survival in the United States: surveillance, epidemiology, and end results program, 1973 to 2001. Neurosurg Focus 2006;20:E1. 3. Central nervous system cancers. National comprehensive cancer network clinical practice guidelines in oncology, vol. 1. http://www.nccn.org/professionals/ physician_gls/PDF/cns.pdf, 2008 (accessed on March 27, 2008). 4. Stupp R, Mason WP, van den Bent MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005;352:987-96. 5. Stupp R, Dietrich PY, Ostermann Kraljevic S, et al. Promising survival for patients with newly diagnosed glioblastoma multiforme treated with concomitant radiation plus temozolomide followed by adjuvant temozolomide. J Clin Oncol 2002;20:1375-82. 6. Sampath P, Brem H. Implantable slow-release chemotherapeutic polymers for the treatment of malignant brain tumors. Cancer Control 1998;5:130-7. 7. Mak M, Fung L, Strasser JF, Saltzman WM. Distribution of drugs following controlled delivery to the brain interstitium. J Neurooncol 1995;26:91-102. 8. Brem H, Piantadosi S, Burger PC, et al. Placebocontrolled trial of safety and efficacy of intraoperative controlled delivery by biodegradable polymers of chemotherapy for recurrent gliomas. The polymer-brain tumor treatment group. Lancet 1995;345:1008-12. 9. Valtonen S, Timonen U, Toivanen P, et al. Interstitial chemotherapy with carmustine-loaded polymers for highgrade gliomas: a randomized double-blind study. Neurosurgery 1997; 41:44-8; discussion 48-9. 10. Brem H, Ewend MG, Piantadosi S, et al. The safety of interstitial chemotherapy with BCNU-loaded polymer followed by radiation therapy in the treatment of newly diagnosed malignant gliomas: phase I trial. J Neurooncol 1995;26:111-23. 11. Zamecnik J. The extracellular space and matrix of gliomas. Acta Neuropathol 2005;110:435-42. 12. Fleming AB, Saltzman WM. Pharmacokinetics of the carmustine implant. Clin Pharmacokinet 2002;41:403-19.
48
13. Bobo RH, Laske DW, Akbasak A, et al. Convectionenhanced delivery of macromolecules in the brain. Proc Natl Acad Sci USA 1994;91:2076-80. 14. Morrison PF, Chen MY, Chadwick RS, et al. Focal delivery during direct infusion to brain: role of flow rate, catheter diameter, and tissue mechanics. Am J Physiol 1999;277:R1218-29. 15. Lopez KA, Waziri AE, Canoll PD, Bruce JN. Convectionenhanced delivery in the treatment of malignant glioma. Neurol Res 2006;28:542-8. 16. Saito R, Krauze MT, Bringas JR, et al. Gadoliniumloaded liposomes allow for real-time magnetic resonance imaging of convection-enhanced delivery in the primate brain. Exp Neurol 2005;196:381-9. 17. Sawyer AJ, Piepmeier JM, Saltzman WM. New methods for direct delivery of chemotherapy for treating brain tumors. Yale J Biol Med 2006;79:141-52. 18. Arteaga CL. The epidermal growth factor receptor: from mutant oncogene in nonhuman cancers to therapeutic target in human neoplasia. J Clin Oncol 2001;19:32S-40S. 19. Chamberlain MC. Treatment options for glioblastoma. Neurosurg Focus 2006;20:E2. 20. Kuan CT, Wikstrand CJ, Bigner DD. EGF mutant receptor vIII as a molecular target in cancer therapy. Endocr Relat Cancer 2001;8:83-96. 21. Mellinghoff IK, Wang MY, Vivanco I, et al. Molecular determinants of the response of glioblastomas to EGFR kinase inhibitors. N Engl J Med 2005;353:2012-24. 22. Li B, Yuan M, Kim IA, et al. Mutant epidermal growth factor receptor displays increased signaling through the phosphatidylinositol-3 kinase/AKT pathway and promotes radioresistance in cells of astrocytic origin. Oncogene 2004;23:4594-602. 23. Choe G, Horvath S, Cloughesy TF, et al. Analysis of the phosphatidylinositol 3 0 -kinase signaling pathway in glioblastoma patients in vivo. Cancer Res 2003;63:2742-6. 24. Pelloski CE, Ballman KV, Furth AF, et al. Epidermal growth factor receptor variant III status defines clinically distinct subtypes of glioblastoma. J Clin Oncol 2007;25:2288-94. 25. Rich JN, Reardon DA, Peery T, et al. Phase II trial of gefitinib in recurrent glioblastoma. J Clin Oncol 2004;22:133-42. 26. Uhm JH, Ballman KV, Giannini C, et al. Phase II study of ZD1839 in patients with newly diagnosed grade 4 astrocytoma [abstract]. J Clin Oncol 2004;22:1505. 27. Cloughesy T, Yung A, Vrendenberg J, et al. Phase II study of erlotinib in recurrent GBM: molecular predictors of outcome [abstract]. J Clin Oncol 2005;23:1507. 28. Vogelbaum MA, Peereboom D, Stevens G, et al. Phase II trial of the EGFR tyrosine kinase inhibitor erlotinib for single agent therapy of recurrent glioblastoma multiforme: interim results [abstract]. J Clin Oncol 2004;22:1558. 29. Prados M, Chang S, Burton E, et al. Phase I study of OSI774 alone or with temozolomide in patients with malignant glioma [abstract]. J Clin Oncol 2003;22:394.
763
764
48
Novel therapies for brain tumors
30. Peereboom DM, Brewer CJ, Suh JH, et al. Phase II trial of erlotinib with temozolomide and concurrent radiation therapy in patients with newly diagnosed glioblastoma multiforme: final results [abstract]. Neuro-oncol 2006; 8:448. 31. Sadones J, Chaskis C, Joosens EJ, et al. A stratified phase II study of cetuximab for the treatment of recurrent glioblastoma multiforme: preliminary results [abstract]. J Clin Oncol 2006;24:1558. 32. Reardon DA, Quinn JA, Vredenburgh JJ, et al. Phase 1 trial of gefitinib plus sirolimus in adults with recurrent malignant glioma. Clin Cancer Res 2006;12:860-8. 33. Doherty L, Gigas DC, Kesari S, et al. Pilot study of the combination of EGFR and mTOR inhibitors in recurrent malignant gliomas. Neurology 2006;67:156-8. 34. Fournier E, Birnbaum D, Borg JP. [Receptors for factors of the VEGF (vascular endothelial growth family)]. Bull Cancer 1997;84:397-405. 35. Stark-Vance V. Bevacizumab and CPT-11 in the treatment of relapsed malignant glioma [abstract]. Neurooncol 2005;7:369. 36. Goli KJ, Desjardins A, Herndon JE, et al. Phase II trial of bevacizumab and irinotecan in the treatment of malignant gliomas [abstract]. J Clin Oncol 2007; 25:2003. 37. Conrad C, Friedman H, Reardon D, et al. A phase I/II trial of single-agent PTK787/ZK 222584 (PTK/ZK), a novel, oral angiogenesis inhibitor, in patients with recurrent glioblastoma multiforme (GBM) [abstract]. J Clin Oncol 2004;22:1512. 38. Wong ML, Kaye AH, Hovens CM. Targeting malignant glioma survival signalling to improve clinical outcomes. J Clin Neurosci 2007;14:301-8. 39. Sathornsumetee S, Reardon DA, Desjardins A, et al. Molecularly targeted therapy for malignant glioma. Cancer 2007;110:13-24. 40. Newton HB. Molecular neuro-oncology and development of targeted therapeutic strategies for brain tumors. Part 1: Growth factor and Ras signaling pathways. Expert Rev Anticancer Ther 2003;3:595-614. 41. Fleming TP, Saxena A, Clark WC, et al. Amplification and/or overexpression of platelet-derived growth factor receptors and epidermal growth factor receptor in human glial tumors. Cancer Res 1992;52:4550-3. 42. Lokker NA, Sullivan CM, Hollenbach SJ, et al. Plateletderived growth factor (PDGF) autocrine signaling regulates survival and mitogenic pathways in glioblastoma cells: evidence that the novel PDGF-C and PDGF-D ligands may play a role in the development of brain tumors. Cancer Res 2002;62:3729-35. 43. George D. Platelet-derived growth factor receptors: a therapeutic target in solid tumors. Semin Oncol 2001; 28:27-33. 44. Strawn LM, Mann E, Elliger SS, et al. Inhibition of glioma cell growth by a truncated platelet-derived growth factor-beta receptor. J Biol Chem 1994;269:21215-22.
45. Pietras K, Rubin K, Sjoblom T, et al. Inhibition of PDGF receptor signaling in tumor stroma enhances antitumor effect of chemotherapy. Cancer Res 2002;62:5476-84. 46. Plate KH, Breier G, Farrell CL, Risau W. Platelet-derived growth factor receptor-beta is induced during tumor development and upregulated during tumor progression in endothelial cells in human gliomas. Lab Invest 1992;67:529-34. 47. Pietras K. Increasing tumor uptake of anticancer drugs with imatinib. Semin Oncol 2004;31:18-23. 48. Geng L, Shinohara ET, Kim D, et al. STI571 (Gleevec) improves tumor growth delay and survival in irradiated mouse models of glioblastoma. Int J Radiat Oncol Biol Phys 2006;64:263-71. 49. Wen PY, Yung WK, Hess K, et al. Phase I study of STI571 (Gleevec) for patients with recurrent malignant gliomas and meningiomas (NABTC 99–08) [abstract]. J Clin Oncol 2002;21:73. 50. Holdhoff M, Kreuzer KA, Appelt C, et al. Imatinib mesylate radiosensitizes human glioblastoma cells through inhibition of platelet-derived growth factor receptor. Blood Cells Mol Dis 2005;34:181-5. 51. O’Reilly T, Wartmann M, Maira SM, et al. Patupilone (epothilone B, EPO906) and imatinib (STI571, Glivec) in combination display enhanced antitumour activity in vivo against experimental rat C6 glioma. Cancer Chemother Pharmacol 2005;55:307-17. 52. Katayama R, Huelsmeyer MK, Marr AK, et al. Imatinib mesylate inhibits platelet-derived growth factor activity and increases chemosensitivity in feline vaccineassociated sarcoma. Cancer Chemother Pharmacol 2004;54:25-33. 53. Dogruel M, Gibbs JE, Thomas SA. Hydroxyurea transport across the blood-brain and blood-cerebrospinal fluid barriers of the guinea-pig. J Neurochem 2003;87:76-84. 54. Gwilt PR, Manouilov KK, McNabb J, Swindells SS. Pharmacokinetics of hydroxyurea in plasma and cerebrospinal fluid of HIV-1-infected patients. J Clin Pharmacol 2003;43:1003-7. 55. Dresemann G. Imatinib and hydroxyurea in pretreated progressive glioblastoma multiforme: a patient series. Ann Oncol 2005;16:1702-8. 56. Reardon DA, Egorin MJ, Quinn JA, et al. Phase II study of imatinib mesylate plus hydroxyurea in adults with recurrent glioblastoma multiforme. J Clin Oncol 2005; 23:9359-68. 57. Couldwell WT, Hinton DR, Surnock AA, et al. Treatment of recurrent malignant gliomas with chronic oral high-dose tamoxifen. Clin Cancer Res 1996;2:619-22. 58. Mastronardi L, Puzzilli F, Couldwell WT, et al. Tamoxifen and carboplatin combinational treatment of high-grade gliomas. Results of a clinical trial on newly diagnosed patients. J Neurooncol 1998;38:59-68. 59. Napolitano M, Keime-Guibert F, Monjour A, et al. Treatment of supratentorial glioblastoma multiforme with radiotherapy and a combination of BCNU and
Novel therapies for brain tumors
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
tamoxifen: a phase II study. J Neurooncol 1999; 45:229-35. Vertosick FT, Selker RG. The treatment of newly diagnosed glioblastoma multiforme using high dose tamoxifen (TMX), radiotherapy and conventional chemotherapy [abstract # 2887]. Proc Am Assoc Cancer Res 1997; 2887. Preul MC, Caramanos Z, Villemure JG, et al. Using proton magnetic resonance spectroscopic imaging to predict in vivo the response of recurrent malignant gliomas to tamoxifen chemotherapy. Neurosurgery 2000;46:306-18. Bianco R, Shin I, Ritter CA, et al. Loss of PTEN/ MMAC1/TEP in EGF receptor-expressing tumor cells counteracts the antitumor action of EGFR tyrosine kinase inhibitors. Oncogene 2003;22:2812-22. Chakravarti A, Zhai G, Suzuki Y, et al. The prognostic significance of phosphatidylinositol 3-kinase pathway activation in human gliomas. J Clin Oncol 2004;22:1926-33. Galanis E, Buckner JC, Maurer MJ, et al. Phase II trial of temsirolimus (CCI-779) in recurrent glioblastoma multiforme: a North Central Cancer Treatment Group Study. J Clin Oncol 2005;23:5294-304. Chang SM, Wen P, Cloughesy T, et al. Phase II study of CCI-779 in patients with recurrent glioblastoma multiforme. Invest New Drugs 2005;23:357-61. Black KL, Chen K, Becker DP, Merrill JE. Inflammatory leukocytes associated with increased immunosuppression by glioblastoma. J Neurosurg 1992;77:120-6. Elliott LH, Brooks WH, Roszman TL. Inability of mitogen-activated lymphocytes obtained from patients with malignant primary intracranial tumors to express high affinity interleukin 2 receptors. J Clin Invest 1990;86:80-6. Bhondeley MK, Mehra RD, Mehra NK, et al. Imbalances in T cell subpopulations in human gliomas. J Neurosurg 1988;68:589-93. Kikuchi T, Abe T, Ohno T. Effects of glioma cells on maturation of dendritic cells. J Neurooncol 2002;58:125-30. Yang T, Witham TF, Villa L, et al. Glioma-associated hyaluronan induces apoptosis in dendritic cells via inducible nitric oxide synthase: implications for the use of dendritic cells for therapy of gliomas. Cancer Res 2002;62:2583-91. Morford LA, Elliott LH, Carlson SL, et al. T cell receptormediated signaling is defective in T cells obtained from patients with primary intracranial tumors. J Immunol 1997;159:4415-25. Ehtesham M, Samoto K, Kabos P, et al. Treatment of intracranial glioma with in situ interferon-gamma and tumor necrosis factor-alpha gene transfer. Cancer Gene Ther 2002;9:925-34. Plautz GE, Touhalisky JE, Shu S. Treatment of murine gliomas by adoptive transfer of ex vivo activated tumordraining lymph node cells. Cell Immunol 1997;178:101-7. Bodmer S, Strommer K, Frei K, et al. Immunosuppression and transforming growth factor-beta in
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
48
glioblastoma. Preferential production of transforming growth factor-beta 2. J Immunol 1989;143:3222-9. Gorelik L, Flavell RA. Immune-mediated eradication of tumors through the blockade of transforming growth factor-beta signaling in T cells. Nat Med 2001;7:1118-22. Ruffini PA, Rivoltini L, Silvani A, et al. Factors, including transforming growth factor beta, released in the glioblastoma residual cavity, impair activity of adherent lymphokine-activated killer cells. Cancer Immunol Immunother 1993;36:409-16. Wikstrand CJ, Fung KM, Trojanowski JQ, et al. Immunohistochemistry and antigens of diagnostic significance. In: Bigner DD, McLendon RE, Bruner JM, editors. Russell and Rubinstein’s pathology of the nervous system. New York, NY: Oxford University Press; 1998. p. 251-304. Riva P, Arista A, Tison V, et al. Intralesional radioimmunotherapy of malignant gliomas. An effective treatment in recurrent tumors. Cancer 1994;73:1076-82. Riva P, Arista A, Sturiale C, et al. Glioblastoma therapy by direct intralesional administration of I-131 radioiodine labeled antitenascin antibodies. Cell Biophys 1997; 24-25:37-43. Riva P, Franceschi G, Arista A, et al. Local application of radiolabeled monoclonal antibodies in the treatment of high grade malignant gliomas: a six-year clinical experience. Cancer 1997;80:2733-42. Cokgor I, Akabani G, Kuan CT, et al. Phase I trial results of iodine-131-labeled antitenascin monoclonal antibody 81C6 treatment of patients with newly diagnosed malignant gliomas. J Clin Oncol 2000;18:3862-72. Reardon DA, Akabani G, Coleman RE, et al. Phase II trial of murine (131)I-labeled antitenascin monoclonal antibody 81C6 administered into surgically created resection cavities of patients with newly diagnosed malignant gliomas. J Clin Oncol 2002;20:1389-97. Prados M, Kunwar S, Lang FF, et al. Final results of phase I/II studies of IL13-pE38QQR administered intratumorally (IT) and/or peritumorally (PT) via convectionenhanced delivery (CED) in patients undergoing tumor resection for recurrent malignant glioma [abstract]. J Clin Oncol 2005;23:1506. Laske DW, Youle RJ, Oldfield EH. Tumor regression with regional distribution of the targeted toxin TF-CRM107 in patients with malignant brain tumors. Nat Med 1997;3:1362-8. Weaver M, Laske DW. Transferrin receptor ligandtargeted toxin conjugate (Tf-CRM107) for therapy of malignant gliomas. J Neurooncol 2003;65:3-13. Kuppner MC, Hamou MF, Sawamura Y, et al. Inhibition of lymphocyte function by glioblastoma-derived transforming growth factor beta 2. J Neurosurg 1989;71:211-7. Siepl C, Bodmer S, Frei K, et al. The glioblastomaderived T cell suppressor factor/transforming growth factor-beta 2 inhibits T cell growth without affecting the interaction of interleukin 2 with its receptor. Eur J Immunol 1988;18:593-600.
765
766
48
Novel therapies for brain tumors
88. Brandes AA, Scelzi E, Zampieri P, et al. Phase II trial with BCNU plus alpha-interferon in patients with recurrent high-grade gliomas. Am J Clin Oncol 1997;20:364-7. 89. Yung WK, Prados M, Levin VA, et al. Intravenous recombinant interferon beta in patients with recurrent malignant gliomas: a phase I/II study. J Clin Oncol 1991;9:1945-9. 90. Hiserodt JC, Vujanovic NL, Reynolds CW, et al. Studies on lymphokine activated killer cells in the rat: analysis of precursor and effector cell phenotype and relationship to natural killer cells. Prog Clin Biol Res 1987; 244:137-46. 91. Hayes RL, Koslow M, Hiesiger EM, et al. Improved long term survival after intracavitary interleukin-2 and lymphokine-activated killer cells for adults with recurrent malignant glioma. Cancer 1995;76:840-52. 92. Porgador A, Gilboa E. Bone marrow-generated dendritic cells pulsed with a class I-restricted peptide are potent inducers of cytotoxic T lymphocytes. J Exp Med 1995;182:255-60. 93. Heimberger AB, Archer GE, Crotty LE, et al. Dendritic cells pulsed with a tumor-specific peptide induce longlasting immunity and are effective against murine intracerebral melanoma. Neurosurgery 2002; 50:158-64; discussion 164-6. 94. Heimberger AB, Crotty LE, Archer GE, et al. Bone marrow-derived dendritic cells pulsed with tumor homogenate induce immunity against syngeneic intracerebral glioma. J Neuroimmunol 2000;103:16-25. 95. Ni HT, Spellman SR, Jean WC, et al. Immunization with dendritic cells pulsed with tumor extract increases survival of mice bearing intracranial gliomas. J Neurooncol 2001;51:1-9. 96. Yu JS, Wheeler CJ, Zeltzer PM, et al. Vaccination of malignant glioma patients with peptide-pulsed dendritic cells elicits systemic cytotoxicity and intracranial T-cell infiltration. Cancer Res 2001;61:842-7. 97. Liau LM, Prins RM, Kiertscher SM, et al. Dendritic cell vaccination in glioblastoma patients induces systemic and intracranial T-cell responses modulated by the local central nervous system tumor microenvironment. Clin Cancer Res 2005;11:5515-25. 98. Bogler O, Huang HJ, Kleihues P, Cavenee WK. The p53 gene and its role in human brain tumors. Glia 1995; 15:308-27. 99. Prives C, Hall PA. The p53 pathway. J Pathol 1999; 187:112-26. 100. Lang FF, Bruner JM, Fuller GN, et al. Phase I trial of adenovirus-mediated p53 gene therapy for recurrent glioma: biological and clinical results. J Clin Oncol 2003;21:2508-18. 101. Curiel DT. The development of conditionally replicative adenoviruses for cancer therapy. Clin Cancer Res 2000;6:3395-9.
102. Chiocca EA, Abbed KM, Tatter S, et al. A phase I open-label, dose-escalation, multi-institutional trial of injection with an E1B-attenuated adenovirus, ONYX-015, into the peritumoral region of recurrent malignant gliomas, in the adjuvant setting. Mol Ther 2004; 10:958-66. 103. Fueyo J, Gomez-Manzano C, Alemany R, et al. A mutant oncolytic adenovirus targeting the Rb pathway produces anti-glioma effect in vivo. Oncogene 2000; 19:2-12. 104. Jiang H, Gomez-Manzano C, Alemany R, et al. Comparative effect of oncolytic adenoviruses with E1A55 kDa or E1B-55 kDa deletions in malignant gliomas. Neoplasia 2005;7:48-56. 105. Jiang H, Gomez-Manzano C, Aoki H, et al. Examination of the therapeutic potential of Delta-24-RGD in brain tumor stem cells: role of autophagic cell death. J Natl Cancer Inst 2007;99:1410-4. 106. Alonso MM, Jiang H, Yokoyama T, et al. Delta-24RGD in combination with RAD001 induces enhanced anti-glioma effect via autophagic cell death. Mol Ther 2008;16:487-93. 107. Alonso MM, Gomez-Manzano C, Bekele BN, et al. Adenovirus-based strategies overcome temozolomide resistance by silencing the O6-methylguanine-DNA methyltransferase promoter. Cancer Res 2007;67:11499-504. 108. Culver KW, Ram Z, Wallbridge S, et al. In vivo gene transfer with retroviral vector-producer cells for treatment of experimental brain tumors. Science 1992; 256:1550-2. 109. Oldfield EH, Ram Z, Culver KW, et al. Gene therapy for the treatment of brain tumors using intra-tumoral transduction with the thymidine kinase gene and intravenous ganciclovir. Hum Gene Ther 1993;4:39-69. 110. Trask TW, Trask RP, Aguilar-Cordova E, et al. Phase I study of adenoviral delivery of the HSV-tk gene and ganciclovir administration in patients with current malignant brain tumors. Mol Ther 2000;1:195-203. 111. Prados MD, McDermott M, Chang SM, et al. Treatment of progressive or recurrent glioblastoma multiforme in adults with herpes simplex virus thymidine kinase gene vector-producer cells followed by intravenous ganciclovir administration: a phase I/II multi-institutional trial. J Neurooncol 2003;65:269-78. 112. Rainov NG. A phase III clinical evaluation of herpes simplex virus type 1 thymidine kinase and ganciclovir gene therapy as an adjuvant to surgical resection and radiation in adults with previously untreated glioblastoma multiforme. Hum Gene Ther 2000; 11:2389-401. 113. Wildner O, Morris JC, Vahanian NN, et al. Adenoviral vectors capable of replication improve the efficacy of HSVtk/GCV suicide gene therapy of cancer. Gene Ther 1999;6:57-62. 114. Nanda D, Vogels R, Havenga M, et al. Treatment of malignant gliomas with a replicating adenoviral vector
Novel therapies for brain tumors
115.
116.
117.
118. 119.
120.
expressing herpes simplex virus-thymidine kinase. Cancer Res 2001;61:8743-50. Mesnil M, Piccoli C, Tiraby G, et al. Bystander killing of cancer cells by herpes simplex virus thymidine kinase gene is mediated by connexins. Proc Natl Acad Sci USA 1996;93:1831-5. Warn-Cramer BJ, Cottrell GT, Burt JM, Lau AF. Regulation of connexin-43 gap junctional intercellular communication by mitogen-activated protein kinase. J Biol Chem 1998;273:9188-96. Marconi P, Tamura M, Moriuchi S, et al. Connexin 43enhanced suicide gene therapy using herpesviral vectors. Mol Ther 2000;1:71-81. Hannon GJ. RNA interference. Nature 2002;418:244-51. Martinez J, Patkaniowska A, Urlaub H, et al. Singlestranded antisense siRNAs guide target RNA cleavage in RNAi. Cell 2002;110:563-74. Shi N, Zhang Y, Zhu C, et al. Brain-specific expression of an exogenous gene after i.v. administration. Proc Natl Acad Sci USA 2001;98:12754-9.
48
121. Shi N, Boado RJ, Pardridge WM. Receptor-mediated gene targeting to tissues in vivo following intravenous administration of pegylated immunoliposomes. Pharm Res 2001;18:1091-5. 122. Schlingensiepen KH, Fischer-Blass B, Schmaus S, Ludwig S. Antisense therapeutics for tumor treatment: the TGF-beta2 inhibitor AP 12009 in clinical development against malignant tumors. Recent Results Cancer Res 2008;177:137-50. 123. Hau P, Jachimczak P, Schlingensiepen R, et al. Inhibition of TGF-beta2 with AP 12009 in recurrent malignant gliomas: from preclinical to phase I/II studies. Oligonucleotides 2007;17:201-12. 124. Zhang Y, Zhang YF, Bryant J, et al. Intravenous RNA interference gene therapy targeting the human epidermal growth factor receptor prolongs survival in intracranial brain cancer. Clin Cancer Res 2004;10:3667-77.
767
43 Pathology Techniques in Stereotactic and Image Guided Biopsy P. T. Chandrasoma . N. E. Klipfel
Stereotactic brain biopsy has become a standard and widely available technique in the past two decades for obtaining tissue from intracranial lesions. The success of the procedure is dependent on (1) the neurosurgeon obtaining a representative sample from the lesion and (2) the ability of the pathologist to make an accurate diagnosis. It has been well documented that stereotactic brain biopsy is a highly effective diagnostic procedure in major academic centers where neurosurgeons specialize in stereotactic procedures and special expertise in neuropathology is readily available. The reported non-diagnostic rate in major centers varies between 4 and 7.2% [1–6]. The incidence of diagnostic failure does not appear to depend on the computed tomographic (CT) morphology of the lesion or the experience of the surgeon [7]. The high level of accuracy is maintained even when results of biopsy are compared with subsequent resection [8]. With the increasing use of the procedure in community hospitals, the success of stereotactic biopsy at the present time depends, among other factors, on the ability of the community hospital’s pathologist to make a diagnosis [9]. The pathologist’s expertise in many such settings is limited by lack of specific training in stereotactic biopsies and the lack of experience working with the very small volumes of tissue available from stereotactic biopsies. It is critical to develop a technique for pathological processing of stereotactic biopsies that will utilize the strengths that community pathologists already have, so as to optimize diagnosis [4]. #
Springer-Verlag Berlin/Heidelberg 2009
When the first stereotactic biopsies were received in our laboratory in the early 1980s, it was clear that they represented a new problem different from that of handling biopsies taken at open craniotomy. First, the specimens were far more representative of the lesion because the location of the biopsy was within a selected target in the CT lesion; in an open biopsy, on the other hand, the biopsy was commonly taken from the periphery of a grossly abnormal area. Second, the specimen was of extremely small size, and our first attempts at using frozen section were not very successful. Third, the ability to obtain additional material from the lesion was more limited than in an open craniotomy, so that requests for additional tissue were less acceptable. This led, in our laboratory, to a trial with the smear technique. Our experience over the past decades has led to the conviction that the best way to handle a stereotactic biopsy is with the initial performance of a smear [10]. This technique provides excellent cytological detail without the artifact caused by freezing the tissue. This utilizes the application of cytopathological expertise, which is usually developed far more in the community pathologist than is neuropathology, making this technique especially appropriate for use in the community setting [4,11].
Types of Biopsies and Smear Technique The type of biopsy obtained varies with different neurosurgeons. At the University of Southern
664
43
Pathology techniques in stereotactic and image guided biopsy
California, the biopsy is taken from a single point target within the lesion with the equivalent of a pediatric bronchoscopic forceps. The specimen consists of one to three samples of tissue that are 1–2 mm in greatest dimension. These are extremely small samples, but the diagnostic success achieved from this sample, coupled with the exceedingly low morbidity and mortality reported with this type of biopsy from our institution, has confirmed the effectiveness of this method [1]. In other centers, larger numbers of specimens might be obtained, often from the entire transit zone within the lesion or from multiple targets. Alternatively, the use of sidecutting biopsy needles provides a core of tissue from the lesion 1+ cm in length. Some authorities report increased diagnostic success with larger specimens [12]. The technique for making a smear is as follows [10]: A small piece from each of the biopsy samples is cut with a scalpel blade and placed on a glass slide. Any necessary orientation can be maintained by labeling the different tissue pieces on the slide. The tissue is smeared on the slide with a second glass slide, using pressure while drawing the slides apart. The amount of pressure required for smearing the tissue varies with the type of specimen. Normal brain and most glial neoplasms smear easily. Reactive gliosis, some fibrillary astrocytomas, and schwannomas smear with difficulty. In specimens that resist smearing, the tissue should be broken up by to-and-fro movements of the slide to effect maximum smearing. While this produces considerable crush artifact, it is preferable to having a thick uncrushed tissue fragment that is impossible to interpret. The way in which tissue smears on the slide is unpredictable, and it is important that both slides used for smearing are stained; they can vary considerably in appearance and cellularity. The smears are immediately fixed in methyl alcohol and stained by the rapid hematoxylin and eosin stain. The method used
for staining is identical to that which is used for frozen sections and requires no special materials. Selection of the tissue used for smearing is simple with the specimens usually received at our institution. When multiple samples are received from different target points or when the tissue specimen is a long core sample, selection of which part of the specimen to use for the smear requires clear understanding of the relationship of the different specimens or areas to the target point. In these cases, it is frequently necessary to process several different pieces of tissue. In evaluating the smear, the ease or difficulty of smearing must be taken into consideration. Tissue that spreads out easily will appear less cellular on the smear than that which is cohesive. In cohesive areas of the smear, it is more difficult to evaluate cytological features because of stratification of cells and tissue opacity. Tissue that has required crushing by repeated movements of the slides will be affected by drying artifact, which can produce considerable nuclear enlargement. The pathologist intending to use the smear technique should be familiar with the normal appearance of different areas of the brain on smear. This can be done by (1) reference to a variety of cytopathological texts and (2) keeping a reference set of smears made from autopsy brain tissue. It is of particular importance to recognize the normally cellular layers of the cerebellum. Also, the temporal lobe is normally moderately hypercellular and contains numerous neurons which can be confused with neoplastic astrocytes. If a diagnosis cannot be rendered through evaluation of the smear, whether from a thick preparation or insufficient diagnostic features (e.g., architectural features), the remaining tissue should be processed for frozen section. If the tissue is nondiagnostic (due to sampling or necrosis), additional tissue should be requested.
Pathology techniques in stereotactic and image guided biopsy
Need for Correlation with Computed Tomography and Clinical Features The two most important factors determining the success of stereotactic biopsy is the communication between the pathologist and the neurosurgeon and collection of the ideal tissue sample. The pathologist must be aware of the patient’s clinical history, the radiological appearance of the lesion, the differential diagnosis, and the indication for the biopsy. The pathologist must develop basic skills in CT radiology and clearly understand how the target point selected for the biopsy relates to the lesion as a whole. The pre-biopsy clinical discussion can be done in one of two ways: (1) the pathologist being present in the operating room as the biopsy is done and examining the CT scan there or (2) the neurosurgeon coming to the frozen section room with the specimen and CT scan and discussing the case during the processing of the specimen. It is highly recommended that one of these schemes is followed.
43
and extraparenchymal. These distinctions are not always precise. We have encountered cases of secretory meningioma associated with severe cerebral edema that were mistakenly thought to be intraparenchymal by radiological features. Modern radiological imaging modalities, such as magnetic resonance imaging and positron emission tomography, can indicate the histological type and grade of cerebral lesions. While these indications are not always accurate, the pathologist should be aware of these radiological opinions. In general, radiological appearances will divide cerebral hemispheric lesions into (1) suspected high grade neoplasms, which are characterized by irregular lesions with irregular contrast enhancement, surrounding edema and mass effect; (2) suspected low grade gliomas, which are typically large, irregular, nonenhancing lesions; and (3) lesions with features that point to specific entities such as pilocytic astrocytoma, pleomorphic xanthoastrocytoma, and ganglioglioma. In all these categories, the possibility of a nonneoplastic condition exists. Selection of the target point within the lesion is of critical importance in obtaining a diagnostic sample and is the most important part of the procedure [17].
Biopsy of Mass Lesions in the Cerebral Hemisphere Mass lesions in the cerebral hemispheres represent the commonest indication for stereotactic brain biopsy. Within this category, it is helpful to separate pediatric patients and those who are positive for human immunodeficiency virus (HIV), in whom diagnostic considerations are frequently different (vide infra) [13–16]. In populations with a high prevalence of HIV positive patients, malignant lymphoma and toxoplasmosis should be kept in the differential even in those who are not known to be at risk. Also important to distinguish from lesions of the cerebral hemispheres are mass lesions that are recognizable as intraventricular, suprasellar, pineal,
Biopsy in Clinically Suspected High Grade Neoplasms Biopsies in suspected high grade lesions are usually obtained from the enhancing edge of the lesion because of the possibility that the central area is necrotic. While this is the best approach, this principle commonly results in the undergrading of high grade lesions due to sampling and might cause diagnostic difficulty if the biopsy is from the infiltrative edge of the neoplasm. In most high grade neoplasms, there are (1) areas composed entirely of neoplasm where the brain has been completely replaced, (2) areas of necrosis without viable tumor cells, and
665
666
43
Pathology techniques in stereotactic and image guided biopsy
(3) areas of neoplastic cell infiltration of the adjacent brain. On radiological imaging, these areas are not precisely distinguishable. Also, adjacent brain showing secondary changes, such as reactive gliosis and edema, might be perceived as neoplasm radiographically. It must be accepted that target selection in these lesions is difficult and does not always produce an ideal specimen. An ideal specimen from a pathologist’s standpoint is one composed entirely of viable neoplasm where it has completely replaced normal brain. With such a specimen, there is no question as to the diagnosis of a malignant neoplasm, the pathologist must then identify the histological type. The vast majority of biopsied high grade lesions are high grade astrocytomas, either anaplastic astrocytoma or glioblastoma multiforme. These are characterized by increased cellularity, pleomorphism, mitoses and cytological atypia in cells recognizable as astrocytes (> Figure 43-1). Glioblastoma multiforme differs from anaplastic astrocytoma by the presence of necrosis (> Figure 43-2) or microvascular proliferation (> Figure 43-3). In stereotactic biopsies, because selection of the target point attempts to avoid necrotic areas, there is a tendency to under-grade glioblastoma multiforme. If there is a clinical
need to differentiate anaplastic astrocytoma from glioblastoma multiforme, biopsies from multiple target points within the lesion should be done. Accurate grading of astrocytomas is based on the presence of atypia, mitotic activity, vascular proliferation and necrosis. While the small size of the sample may lead to under-grading, the accuracy is generally adequate for clinical management. Astrocytes are recognized by the presence of thin cytoplasmic fibrils that form a background network between the neoplastic cells (> Figure 43-1). The presence of gemistocytes, which are large cells with abundant eosinophilic cytoplasm, are seen in some astrocytic neoplasms. In rare astrocytomas where the cells are so poorly differentiated as to lack prominent fibrils or show astrocytic features, immunoperoxidase staining of formalin fixed, paraffin embedded tissue for gliofibrillary acidic protein might be helpful. In cases where a diagnosis of high grade astrocytoma is uncertain, immunoperoxidase staining with the MIB1 epitope (Ki-67 antigen) will show the proliferative neoplastic cells staining >5% in anaplastic astrocytomas and provides a good correlation with prognosis in this group of neoplasms [18]. Other neoplasms that might be encountered in a suspected high-grade hemispheric lesion are
. Figure 43-1 Smear of anaplastic astrocytoma showing cellular proliferation of astrocytes with enlarged atypical nuclei (arrow) and cytoplasmic fibrils (H&E, 400)
. Figure 43-2 Smear of glioblastoma showing necrotic debris (H&E, 200)
Pathology techniques in stereotactic and image guided biopsy
43
. Figure 43-3 Smear of glioblastoma showing microvascular proliferation (arrow) (H&E, 400)
. Figure 43-4 Smear of metastatic adenocarcinoma showing cohesive groups of epithelial cells with intracytoplasmic vacuoles (long arrow) and glandular lumens (short arrow) (H&E, 400)
anaplastic oligodendrogliomas, mixed gliomas, metastatic carcinoma (> Figure 43-4), metastatic melanoma (> Figure 43-5), high-grade malignant lymphoma (> Figure 43-6), and a variety of very rare lesions. These neoplasms have
distinctive cytological features that permit diagnosis on smear. In some cases, recognition of cells on the smear is not absolute but permits processing of the tissue in such a manner as to perform special studies to characterize the cells.
667
668
43
Pathology techniques in stereotactic and image guided biopsy
. Figure 43-5 Smear of metastatic melanoma showing loosely cohesive cells with cytoplasmic pigment (arrow) (H&E, 400)
. Figure 43-6 Smear of malignant lymphoma showing discohesive round cells with typical chromatin pattern of transformed lymphocytes and lymphoglandular bodies (arrow) (H&E, 400)
The two special procedures commonly utilized for stereotactic biopsies are (1) immunoperoxidase staining for antigens that characterize the derivation of neoplastic cells such as CD45 (lymphoid), GFAP (astrocytic); keratin (epithelial), and HMB45 (melanocytic) and (2) electron microscopy, which is usually done only in academic institutions. The combination of smear, permanent
section, immunoperoxidase staining and rarely EM will provide an accurate diagnosis in all cases of high grade lesions in which a representative sample is provided to the pathologist. At this time, molecular genetics is not used for diagnosis, but testing can provide treatment indications, such as fluorescent in situ hybridization (paraffin tissue) for 1p and 19q losses in oligodendroglioma [19]. In cases where the sample is less than ideal, the diagnostic accuracy is less, and the likelihood of diagnostic error increases as the sample quality decreases. A less than ideal sample is one taken from an area where a high-grade astrocytoma infiltrates normal brain, containing elements of both neoplasm and brain. In such cases, a diagnosis of high grade astrocytoma can be made when the neoplastic cells are recognized as such by virtue of their anaplasia. In such biopsies, however, there are two possible sources of error: (1) under-grading of the astrocyctoma, which is not a serious error because the diagnosis that will be given will be an anaplastic astrocytoma [20], and (2) misdiagnosis of reactive gliosis as astrocytoma. Reactive gliosis is usually characterized by moderate cellularity, cellular polymorphism and minimal cytological atypia without the anaplasia that characterizes anaplastic astrocytoma. However, experience is needed in these cases, particularly when the neurosurgeon reports that there is a high clinical likelihood of a high grade glioma and insists that the sample is representative. In clinically suspected high-grade lesions, a pathological diagnosis of low grade (welldifferentiated) astrocytoma should never be rendered by the pathologist or accepted by the neurosurgeon. In such cases, biopsies showing increased cellularity without anaplasia represent areas of reaction around the lesion, and therefore samples that are not representative of the pathological lesion. When the lesion is a high grade astrocytoma, the pathological diagnosis of low grade astrocytoma represents clinically significant under-grading. When the lesion turns
Pathology techniques in stereotactic and image guided biopsy
out to be anything other than an astrocytoma, the incorrect diagnosis of low grade astrocytoma has the potential to be devastating. Cases of cerebral abscesses rarely have had an initial diagnosis rendered on stereotactic biopsy as low grade astrocytoma.
Biopsy in Clinically Suspected Low Grade Glioma Low grade astrocytomas of the cerebral hemispheres tend to be large, ill-defined, non-enhancing lesions on CT scan without much associated edema or mass effect. Normal brain tissue is replaced by the neoplasm in the central area, with extensive infiltration at the periphery. Low grade astrocytoma is characterized by low to moderate cellularity and slight cytological atypia (> Figure 43-7). In comparison, normal brain shows low cellularity, capillaries, a polymorphic population of astrocytic, microglial, oligodendroglial and neural cells and granular background with few fibrils (> Figure 43-8). Anaplasia, mitotic figures, microvascular proliferation, and necrosis are absent. Rarely, these neoplasms have a cystic component. . Figure 43-7 Smear of low grade astrocytoma showing low cellularity, minimal cytological atypia (arrow) and a fibrillary background. The astrocytic nuclei are enlarged (H&E, 400)
43
In solid lesions, the target point selected is usually in the central part of the lesion, where the tissue consists entirely of neoplastic cells. In such samples, the smear shows slightly increased cellularity, a monomorphous population of fibrillary astrocytes with prominent cytoplasmic fibrillary processes forming the background of the smear, and mild cytological atypia (> Figure 43-7). The pathological diagnosis rests on the differentiation of the neoplastic astrocytic proliferation from normal brain. This is easier in the smear preparation, where the abnormal fibrillary background of the neoplasm varies greatly from the finely granular eosinophilic background of normal brain. A diagnosis of low grade astrocytoma can be made with much greater confidence on a smear preparation than in sections (either frozen or paraffin) for this reason. In sections, the background is less distinctive and the diagnosis rests on identification of cellularity and cytological atypia, which might deviate only slightly from normal or reactive. MIB-l immunoperoxidase staining in low grade astrocytomas in most cases will show a very low proliferative index with <2% of cells showing positive nuclear staining. When a higher percentage of cells are positive, a more aggressive lesion is possible. . Figure 43-8 Smear of normal cerebrum with typical vessel (arrow) and paucicellular polymorphous neuroglial cell population in a granular and minimally fibrillary background (H&E, 400)
669
670
43
Pathology techniques in stereotactic and image guided biopsy
Distinguishing between low grade astrocytoma and reactive gliosis is an age-old problem that has improved with the advent of stereotactic biopsy. In most cases, reactive gliosis is not even a diagnostic consideration due to the targeted tissue sampling. Reactive gliosis is usually characterized by low cellularity with several different cell types, commonly including chronic inflammatory cells. The astrocytic proliferation in reactive gliosis is usually dominated by reactive gemistocytes. The presence of numerous gemistocytic astrocytes in a smear of low cellularity is the best single criterion for a diagnosis of reactive gliosis. In rare types of reactive gliosis, the smear is uniform and composed of fibrillary astrocytes in a densely fibrillary stroma, often containing Rosenthal fibers (> Figure 43-9). These cases are impossible to distinguish from fibrillary astrocytomas by the smear alone. This type of gliosis occurs adjacent to craniopharyngioma and syringomyelia; clinical and CT data are essential for diagnosis in such cases. Cystic low grade astrocytomas represent one of the most difficult diagnostic problems on stereotactic biopsy, because nonneoplastic cystic lesions are surrounded by reactive gliosis. Cystic astrocytomas usually contain protein-rich clear yellow fluid, which commonly represents the
first sample. Biopsy of either the wall or a mural nodule shows cellular material that can make it very difficult to distinguish between low grade astrocytoma from reactive gliosis. In such cases, a good plan is to follow the patient with repeat biopsy if the cyst refills or the lesion progresses, clinical events that make astrocytoma more likely. Some low grade glial neoplasms are composed either predominantly of oligodendroglial cells (> Figure 43-10) or of mixtures of oligodendroglia and astrocytes. These cases are commonly associated with microcalcification. In such cases, the smear shows the presence of the typical oligodendroglial cells in varying numbers. These are round cells with uniform round nuclei which have a finely granular chromatin pattern. They appear either as naked nuclei or have abundant and faintly eosinophilic cytoplasm. When oligodendroglial cells are few in number, it is usually not necessary to change the diagnosis of low grade astrocytoma. In cases where oligodendroglial cells are present in substantial numbers, a diagnosis of low grade mixed glioma should be made. In cases where oligodendroglial cells dominate the smear and sections and astrocytes are few in number, the differential diagnosis is a mixed oligoastrocytoma or the infiltrating edge of a pure oligodendroglioma with associated
. Figure 43-9 Smear of tissue adjacent to a craniopharyngioma showing reactive gliosis with Rosenthal fibers (arrow) (H&E, 400)
. Figure 43-10 Smear of oligodendroglioma showing a cellular smear with uniform round cells and calcification (arrow) (H&E, 400)
Pathology techniques in stereotactic and image guided biopsy
reactive gliosis. When the astrocytic cells are mainly gemistocytic, reactive gliosis is favored. MIB1 immunoperoxidase staining is useful in these well-differentiated mixed gliomas to detect the possibility of a more aggressive biological behavior.
43
. Figure 43-11 Smear of subependymal giant-cell astrocytoma showing enlarged cells with atypical nuclei and abundant eosinophilic cytoplasm (H&E, 400)
Biopsy in Clinically Distinctive Neoplasms Pediatric or young adult neoplastic lesions in the cerebral hemisphere can have clinically distinctive features. These include pilocytic astrocytoma (young age, circumscription, cystic component), pleomorphic xanthoastrocytoma (young age, surface location) and ganglioglioma (young age, temporal lobe location, calcification). Smears and sections from these neoplasms frequently have higher cellularity and pleomorphism than typical low grade astrocytomas, and these features might lead to an erroneous diagnosis of high grade astrocytoma. Good communication between the pathologist and the neurosurgeon is essential for accurate diagnosis.
Biopsy of Mass Lesions in Distinctive Locations The pathologist must be aware of the exact location of the lesion being biopsied. This will permit an appropriate differential diagnosis to be developed. Mass lesions related to the ventricles include ependymoma, choroid plexus neoplasms, colloid cyst of the third ventricle, craniopharyngioma, central neurocytoma, intraventricular meningioma, and subependymal giant cell astrocytoma (> Figure 43-11). In some of these lesions, the pathological features on smear and sections might resemble other lesions, leading to potential misdiagnosis in the absence of adequate clinical and radiographic information. Examples
include subependymal giant-cell astrocytoma, which might mimic an anaplastic gemistocytic astrocytoma, and a central neurocytoma, which can closely resemble oligodendroglioma. The differential diagnosis of mass lesions in the pineal region includes pinealocyte neoplasms and germ cell neoplasms. Germinoma is the commonest of these and is characterized on smear by the presence of a triple cell population consisting of large malignant germ cells which have large nuclei with prominent nucleoli, small lymphocytes, and epithelioid histiocytes. Germinoma can present difficulties in diagnosis by stereotactic biopsy when there is extensive granulomatous inflammation within the neoplasm. Nongerminomatous germ cell neoplasms include teratoma, choriocarcinoma, embryonal carcinoma and yolk sac carcinoma. The presence of elevated levels in serum and cerebrospinal fluid of beta-hCG (choriocarcinoma) and alpha-fetoprotein (yolk sac carcinoma and embryonal carcinoma) is useful from a diagnostic standpoint. It should be noted that these neoplasms, particularly choriocarcinoma, tend to bleed profusely at biopsy, and therefore, caution is necessary.
671
672
43
Pathology techniques in stereotactic and image guided biopsy
The differential diagnosis of suprasellar lesions includes craniopharyngioma, pituitary adenoma, hypothalamic glioma, meningioma, germinoma and chordoma. Many of these lesions have distinctive radiological features such as the presence of calcification in craniopharyngioma and the origin from the skull base in chordoma. One major difficulty in diagnosis is the biopsy from the wall of a suprasellar cystic lesion of a young patient that shows a population of fibrillary astrocytes, sometimes with Rosenthal fibers. Care must be taken before this is mistakenly interpreted as a hypothalamic glioma. In such cases, obtaining a sample of cyst fluid is very helpful; craniopharyngioma has the typical brown, oily fluid, which can be shown to contain cholesterol crystals on a direct smear, whereas hypothalamic astrocytoma has a thick, yellow, protein-rich fluid devoid of cholesterol crystals. Lesions of the cerebellum are extremely varied. In the elderly, the commonest cerebellar mass lesion by far is metastatic carcinoma. In the young, juvenile pilocytic astrocytoma, hemangioblastoma and dysplastic gangliocytoma are the main considerations in hemispheric lesions and medulloblastoma (> Figure 43-12) and ependymoma (> Figure 43-13) in midline lesions. The diagnosis of hemangioblastoma is rarely made by stereotactic biopsy which usually produces a sample with little recognizable cellular material. Dysplastic gangliocytoma has a distinctive radiographic appearance which should be recognized. Extra-axial lesions are rarely biopsied stereotactically because they are amenable to resection. However, meningiomas (> Figure 43-14 and > 43-15) might sometimes be biopsied stereotactically when there are unusual features. Secretory meningiomas tend to produce marked cerebral edema, which might lead to misinterpretation of radiographic images as intra-axial neoplasms. Secretory meningiomas show typical meningothelial cells, which are spindle cells with ovoid nuclei, but they contain prominent cytoplasmic inclusions, which stain immunohistochemically
. Figure 43-12 Smear of medulloblastoma showing hypercellularity, uniform oval nuclei, indistinct cytoplasm and rosettes (arrow) (H&E, 400)
. Figure 43-13 Smear of ependymoma showing columnar cells with oval nuclei arranged around small vessels (perivascular pseudorosettes, arrow) (H&E, 400)
with carcinoembryonic antigen and might be misdiagnosed as adenocarcinoma.
Biopsies of Mass Lesions in HIV-Positive Patients The main indications for stereotatic biopsy in mass lesions of the brain in HIV-positive patients in our institution are (1) to make a diagnosis of
Pathology techniques in stereotactic and image guided biopsy
. Figure 43-14 Smear of meningioma showing whorl of cohesive cells with oval nuclei, one of which contains an intranuclear cytoplasmic pseudoinclusion (arrow) (H&E, 400)
. Figure 43-15 Smear showing psammoma bodies (laminated calcifications) in a meningioma (H&E, 400)
malignant lymphoma in a patient whose mass lesion has not responded to empiric treatment against Toxoplasma and (2) to confirm the diagnosis of progressive multifocal leukoencephalopathy [13–16]. In this setting, cerebral toxoplasmosis is rarely encountered, unlike previously, when this was the commonest diagnosis
43
. Figure 43-16 Section of cerebral toxoplasmosis showing a pseudocyst, scattered tachyzoites (short arrow) and (long arrow) chronic inflammation (H&E, 400)
in HIV-positive patients undergoing stereotactic brain biopsy. We still encounter toxoplasmosis in patients who are either not known to be at risk for HIV or in those presenting emergently in whom stereotactic biopsy is performed before HIV status is known. Toxoplasmosis is characterized by the presence of necrosis, chronic inflammation, reactive gliosis, and Toxoplasma organisms, either as pseudocysts or small crescentic tachyzoites (> Figure 43-16). The diagnosis can be confirmed by the use of immunoperoxidase staining for Toxoplasma antigens. Malignant lymphoma of the brain has become a common diagnosis at stereotactic biopsy in our institution. In addition to the epidemic of this neoplasm in HIV-positive patients, we have encountered an increase in the frequency of central nervous system (CNS) lymphoma in the HIV-negative population, both over and under 65 years of age [21–22]. Malignant lymphomas of the CNS tend to be high grade lymphomas and are commonly associated with extensive necrosis. Necrosis presents a problem in diagnosis when the biopsies show only nonviable tissue. Repeat biopsies are frequently needed to demonstrate the malignant lymphoid cells,
673
674
43
Pathology techniques in stereotactic and image guided biopsy
which are distinctive on smear as large, round cells with large nuclei, prominent nucleoli and lymphoglandular bodies (detached spherical fragments of cytoplasm) (> Figure 43-6). These cells are present as diffusely infiltrating cells but might have a perivascular distribution, especially in the periphery of the lesion. The diagnosis of lymphoma can be confirmed with immunoperoxidase staining for CD45 (common leukocyte antigen). Most CNS lymphomas are B-cell lymphomas. Progressive multifocal leukoencephalopathy (PML) usually produces a distinctive radiological appearance with multifocal white matter lesions. Histological diagnosis might be necessary in cases that are being considered for high dosage antiviral drug therapy. In these cases, stereotactic biopsy represents the optimal method of obtaining tissue. This condition (PML) is characterized by demyelination (associated with numerous foamy macrophages), necrosis, and a reactive glial proliferation, which shows the presence of infected cells. The gliosis is commonly very cellular with numerous highly atypical, giant astrocytes, and occasional mitotic figures. In the absence of the appropriate clinical background, the smear features can mimic an anaplastic astrocytoma. The diagnostic feature is the presence of enlarged oligodendroglial cells with large round nuclei that contain viral inclusions. In smears and sections, these distinctive nuclei have ground-glass basophilia and clumped chromatin is frequently visible (> Figure 43-17). When inclusions are rare, as in a biopsy from the edge of a lesion, immunoperoxidase staining for the JC papovavirus is helpful in confirming the diagnosis of PML (> Figure 43-18). In many HIV-positive patients, the brain biopsy tissue appears abnormal, even when no specific lesion is identified. These abnormalities include edema, chronic inflammatory cell infiltration and reactive gliosis with astrocytic atypia. The relationship of such abnormalities, which can broadly be classified as ‘‘subacute HIV encephalitis,’’ to AIDS dementia and microglial nodules is uncertain. However, it almost certainly
. Figure 43-17 Section of progressive multifocal leukoencephalopathy showing infected enlarged nuclei with a groundglass appearance (arrow) (H&E, 400)
. Figure 43-18 Immunoperoxidase stain for JC virus antigen showing positive (dark-stained) nuclei in PML (400)
reflects infection of neural elements with HIV. HIV infection of the tissue must be assumed in all biopsies of HIV-positive patients, and stringent precautions must be taken to ensure that those handling the tissue are protected during all stages of the procedure. In our institution, the incidence of HIV positivity is high enough to recommend that all tissue at stereotactic brain biopsy be handled with extreme precautions. It is of great importance that the pathologist be informed as to the HIV-positive status of a patient when this information is available to the neurosurgeon.
Pathology techniques in stereotactic and image guided biopsy
Biopsy of a Recurrent Lesion after Radiation for Glioma A common and difficult problem encountered at stereotactic biopsy is in a patient who develops a mass lesion after radiation therapy for a glioma. In these cases, the differential diagnosis is between recurrent astrocytoma and changes induced by radiation. Radiation changes are characterized by necrosis; vascular changes, including neovascularization, hyalinization and endothelial atypia; chronic inflammation; and reactive gliosis. The reactive glial cells often show cytological atypia, which can sometimes be severe. The reactive cells that show radiation-induced cytologic atypia can be impossible to differentiate from neoplastic residual astrocytes. Examination of radiated brain for lesions other than astrocytomas (e.g., metastatic carcinoma) demonstrates that radiation-induced cytological atypia in astrocytes can be very severe. The diagnosis of recurrent astrocytoma should be made in these cases only when there is evidence of neoplastic astrocytic proliferation. This includes hypercellularity of the astrocytic proliferation and areas where the tissue is overrun by the proliferating neoplastic astrocytes (> Figure 43-19). In biopsies that show necrotic
. Figure 43-19 Smear of recurrent astrocytoma after radiation showing high cellularity and marked cytological atypia (arrow) (H&E, 400)
43
and gliotic tissue with scattered atypical astrocytes, the diagnosis is radiation change with a descriptive comment about the atypical astrocytes. This comment may include a statement that it is not possible to distinguish reactive astrocytes with radiation change from residual single neoplastic astrocytes. The critical information provided in such a diagnosis should be the fact that the biopsy contains no evidence that the clinical lesion is the result of recurrent neoplastic astrocytic proliferation. The pathologist cannot reliably confirm the sterilization of a glial neoplasm by radiation.
Handling of a Biopsy that Shows Inflammation Stereotactic biopsies from inflammatory lesions of the brain present the greatest amount of difficulty [7]. The only time a definitive diagnosis can be made in such cases is when the etiology of the inflammatory lesion is identified, usually by recognizing an infectious agent. Diagnosis of PML, herpes and cytomegalovirus encephalitis, tuberculoma (where acid-fast stains show mycobacteria), fungal granulomas such as those caused by Cryptococcus and Coccidioides, cysticercosis, toxoplasmosis, and cerebritis caused by Mucor are examples of specific inflammatory lesions. Tissue for infectious work-up (culture and/or PCR) should be submitted from the sterile operating room environment. When the biopsy shows inflammation without a recognizable etiological agent (> Figure 43-20), diagnosis becomes very uncertain. With a small sample being obtained from a lesion, it is difficult to know whether the observed findings are representative of the whole lesion or whether the changes represent inflammation and reactive gliosis at the edge of a neoplastic lesion that has not been sampled. If the target selected was in the peripheral part of the lesion, the finding of inflammation and reactive gliosis is an indication
675
676
43
Pathology techniques in stereotactic and image guided biopsy
. Figure 43-20 Section showing nonspecific inflammation composed of perivascular and scattered infiltrate of reactive lymphoid cells (H&E, 400)
for taking a biopsy from a second, more central intralesional target point at some distance from the first. If this also shows similar features, a diagnosis of nonspecific chronic inflammation and reactive gliosis is made. This assumes that there is communication between neurosurgeon and pathologist and a high level of trust in the ability of the procedure to procure tissue accurately from the defined target. In cases where the final diagnosis is nonspecific inflammation and fibrosis, patient follow-up must be ensured. In those cases where the lesion progresses on followup imaging, repeat biopsy is necessary. We have encountered a few cases where a neoplasm was diagnosed on a follow-up biopsy in a lesion that progressed. When the smear shows nonspecific inflammation and reactive gliosis, the tissue must be processed to optimize diagnosis. The pathologist must ask the neurosurgeon to provide the maximum amount of tissue possible. The available tissue must be triaged depending on the features observed in the smear. In rare cases, the dominant inflammatory cell is a neutrophil; in these cases, a bacterial etiology is likely and routine and anaerobic bacterial cultures are necessary. In one case,
we isolated Nocardia species from such a specimen. Usually, however, the inflammatory cells consist of a mixture of lymphocytes and plasma cells. In such cases, a small sample of tissue can be taken for electron microscopy and the rest processed for permanent sections. Immunoperoxidase staining and, if necessary, polymerase chain reaction studies can be performed on paraffin sections for herpes simplex virus, cytomegalovirus, JC virus (PML), and Toxoplasma. In a few cases, granulomatous inflammation is identified in smears and sections. This is characterized by the presence of epithelioid histiocytes in aggregates, a feature that is more easily recognizable in sections than smears. When nonnecrotizing granulomatous inflammation is present in a biopsy of a mass in the pineal or suprasellar region, the possibility of germinoma must be considered and additional tissue requested. In other locations, and where necrosis is present, tissue must be sent for mycobacterial and fungal cultures and sections with granulomas stained with acid-fast and fungal stains in an attempt to establish a diagnosis. Cerebral sarcoidosis is also a rare possibility in cases with nonnecrotizing granulomas.
References 1. Apuzzo MLJ, Chandrasoma PT, Cohen D, et al. Computed imaging stereotaxy: experience and perspective related to 500 procedures applied to brain masses. Neurosurgery 1987;20:930-7. 2. Bleggi-Torres LF, de Noronha L, Gugelmin ES, et al. Accuracy of the smear technique in the cytological diagnosis of 650 lesions of the central nervous system. Diagn Cytol 2000;24:293-5. 3. Firlik KS, Martinez AJ, Lunsford LD. Use of cytological preparations for the intraoperative diagnosis of stereotactically obtained brain biopsies: a 19-year experience and survey of neuropathologists. J Neursurg 1999;91:454-8. 4. O’Neill KS, Dyer PV, Bell BA, et al. Is preoperative smear cytology necessary for CT-guided stereotactic biopsy? Br J Neurosurg 1992;6: 421-7. 5. Yu X, Liu Z, Tian Z, et al. Stereotactic biopsy for intracranial space-occupying lesions: clinical analysis of 550 cases. Stereotact funct neurosurg 2000;75:103-8.
Pathology techniques in stereotactic and image guided biopsy
6. Ferreira MP, Ferreira NP, Pereira AA, et al. Stereotactic computed tomography-guided brain biopsy: diagnostic yield based on a series of 170 patients. Surg Neurol 2006;65:27-32. 7. Ranjan A, Rajshekhar V, Joseph T, et al. Nondiagnostic CT-guided stereotactic biopsies in a series of 407 cases: influence of CT morphology and operator experience. J Neurosurg 1993;79:839-44. 8. Chandrasoma PT, Smith MM, Apuzzo MLJ. Stereotactic brain biopsy in brain masses: comparison of results at biopsy versus resected surgical specimen. Neurosurgery 1989;24:160-5. 9. Abernathey CD, Ramsey M, Knight K. Utilization of image-derived computer-assisted stereotaxis in a community-based practice setting. Stereotact Funct Neurosurg 1992;58:99-102. 10. Chandrasoma PT, Apuzzo MLJ. Stereotactic Brain Biopsy. New York: Igaku-Shoin, 1989. 11. Burger PC. Use of cytological preparations in the frozen section diagnosis of central nervous system neoplasms. Am J Surg Pathol 1985;9:344-9. 12. Kleihues P, Volk B, Anagnostopoulos J, et al. Morphologic evaluation of stereotactic brain tumour biopsies. Acta Neurochir Suppl 1984;33:171-81. 13. Chappell ET, Guthrie BL, Orenstein J. The role of stereotactic biopsy in the management of HIV related focal brain lesions. Neurosurgery 1992;30:825-9. 14. Cajulis, RS, Hayden R, Frias-Hidveqi D, et al. Role of cytology in the intraoperative diagnosis of HIV-positive patients undergoing stereotactic brain biopsy. Acta Cytol 1997;41:481-6.
43
15. Skolasky RL, Dal Pan GJ, Olivi A, et al. HIV-associated primary CNS morbidity and utility of brain biopsy. J Neurol Sci 1999;163:32-8. 16. Vallat-Decouvelaere AV, Chre´tien F, et al. The neuropathology of HIV infection in the era of highly active antiretroviral therapy. Ann Pathol 2003;23:408-23. 17. Chandrasoma PT: Problems relating to pathological interpretation in stereotactic biopsy procedures. In: Apuzzo MLJ, editor. Brain surgery: complication avoidance and management. New York: Churchill Livingstone, 1993. p. 425-31. 18. Wakimoto H, Aoyagi M, Nakayama T, et al. Prognostic significance of Ki-67 labeling indices obtained using MIB-1 monoclonal antibody in patients with supratentorial astrocytomas. Cancer 1996;77:373-80. 19. Kouwenhoven MC, Kros JM, French PF, et al. 1p/19q loss within oligodendroglioma is predictive for response to first line temozolomide but not to salvage treatment. Eur J Cancer 2006;42:2499-503. 20. Apuzzo MLJ, Hinton DR. Clinically relevant issues attendant to pathology. In: Apuzzo MLJ, editor. Malignant cerebral glioma. Park Ridge, IL: American Association of Neurological Surgeons, 1990. p. 19-21. 21. Namiki TS, Nichols P, Young T, et al. Stereotactic biopsy diagnosis of central nervous system lymphoma. Am J Clin Pathol 1988;90:40-5. 22. Haldorsen IS, Krossnes BK, Aarseth JH, et al. Increasing incidence and continued dismal outcome of primary central nervous system lymphoma in Norway 1989–2003. Cancer 2007;110:1803-14.
677
38 Robotic Neurosurgery P. L. Gildenberg
What is a Robot? In order to discuss robotic surgery, it is first necessary to ask, ‘‘What is a robot?’’ The term robot is assigned to many devices that have little in common. As a minimum, to be called a robot, the device should have a program or series of steps that are initiated by a command. Some items do not meet even that simple definition, but are called robotic as a marketing tool, rather than a technological descriptor. To begin our discussion, let us look at some of the earliest examples. The earliest use of the term robot referred to the mechanical men and women in Karel Capek 1923 play ‘‘R.U.R.’’ or ‘‘Rossum’s Universal Robots.’’ The term stems from the Czech word robota, meaning to work or compulsory labor, and is related to the Greek arbeit, also meaning to work. The ‘‘robotniks’’ were peasants who owed such labor. The primary dictionary definition is ‘‘a machine that looks like a human and performs various complex human tasks, such as walking or talking,’’ although this describes an animatron, which may or may not be a robot. Subsequent definitions include ‘‘any machine or mechanical device that operates automatically with humanlike skill’’ (Random House) or ‘‘an automatic apparatus or device that performs functions ordinarily ascribed to human beings or operates with what appears to be almost human intelligence’’ (Webster). Surprisingly, NASA defines a robot as simply ‘‘a mechanical device that operates automatically.’’ The first ‘‘robots’’ were mechanical devices that obtained power from such things as falling liquid, such as water. In 1,000 BCE, an ancient #
Springer-Verlag Berlin/Heidelberg 2009
Egyptian water clock depended on the rate water drained through holes in a container. As early as 300 BCE, Heron of Alexandria described, in a series of five volumes that still survive, a group of hydraulic devices that were robotic in nature [1]. Placing a cup on a platform (the command) would fill the cup with wine and then stop (the program). Overfilling a wine ‘‘greedy cup’’ would cause the wine to drain from an opening on the bottom of the cup to deposit the wine on the lap of the greedy guest. Heron also used pneumatic pressure by running water through a pipe into an air filled container so the escaping air made several silver birds chirp as they moved their wings. He also invented the cam, which he used to program a robotic puppet theater powered by ropes, wires and levers to cause several figures to move in such a way that they seemed to interact within a group. The program could be changed by shortening or lengthening the ropes or levers, which may constitute the earliest computer programming. About that same time, a coin operated (the command – insertion of a coin) dispenser (the program) of a measured amount of water and a piece of soap (actually a ball of pumice) was invented in Byzantium to be used for worshippers to wash their hands as they entered a temple. None of those robots resembled a person. A robot that is made to look like a person or animal is called an animatron. If speech or sounds supposedly emanating from the animatron are added, the result is a resulting robot is an audio-animatron. Leonardo da Vinci is considered by many to be the inventor of the first animatronic robot. He made a robot that looked like a knight in armor
584
38
Robotic neurosurgery
who extended his arms or returned them to his sides (the program) or sat down (another independent program) or arose (another program) on command, a working replica of which is in the da Vinci Museum in Florence [1]. Among a multitude of his inventions, he also made a walking lion that could be programmed to perform acts which could be varied through a series of gear ratios, which could be selected. Some time later, he also used such gears in machines designed to move men and weapons, a precursor to the tank. The ancient Greek water clock, which relied on power from falling water to implement the routine or program, would qualify, as would a present day alarm clock ‘‘programmed’’ to sound a bell at a predetermined time. Other early examples include an animatron whose voice was supplied by a recording of a human voice, long before computerized speech simulation was developed. Elektro, was designed to look somewhat like a person. He (or it) was introduced at the 1939 World’s Fair in New York, and probably represents the first robot built in the US. As people approached, he would greet them or offer other pre-recorded messages. A person commanded each program to run, which was necessary in that pre-computer stage. As direct descendants, the best examples of present day audio-anamatronics may be seen at Disneyland. Human and animal figures go through a series of actions synchronized with pre-recorded speech or animal sounds. There may be a multitude of programmed gestures and facial expressions. The computer operated program may sometimes be extremely complicated, so that one audio-animatron seems to interact with another. You can see somewhat less sophisticated animatrons at your local Chuck E. Cheese. In its simplest form, the program is predetermined by the operator, and the robot has no decision-making capabilities, but the command may initiate a series of steps. A simple robot may be under direct moment-to-moment
control of the operator, for instance a wireless remote-controlled toy car. A program might initiate a series of operations, as in a robotic coffee maker that may at the programmed time grind the beans, put a measured amount into the basket, and let a measured amount of water drip through the coffee grounds to fill the cup below. Robots may be electronic, electrical, or mechanical programmable self-controlled devices. Typically, a modern robot performs a task by following a set of instructions stored in an onboard computer or robot controller that specifies exactly what it must do to complete the job. The computer sends commands to each of the robot’s motorized joints, which function much like human joints to move various parts of the robot. The operator input is channeled through a computer interface that is programmed by the operator and then instructs the robot controller to accomplish the programmed task. The commands that constitute the program may be a series of steps in a routine or subroutine that is initiated by the command. Alternatively, the robot can also be used to denote a computer interfaced technique wherein the operator controls each step of the program by a series of stepby-step manual inputs, which constitutes a master-slave electronic manipulator, as in the da Vinci surgical robot. The use of feed-back that is recognized by the computer and influences subsequent steps requires sensory input, which may mimic the human in that it can be visual (video), tactile (a sensory device such as a pressure transducer), thermal (a thermostat), or proprioceptive (to let the robot know the position of its arm or arms, from which position further actions may ensue). Such systems may also be used to improve spacial accuracy of a robot, for instance with continuous video or other visual registration feedback to correct the trajectory or the tool held at the end of the robot arm and make it increasingly accurate as the target is approached.
Robotic neurosurgery
Additional feed-back controls and multistep programs may be superimposed on a simple design. At the given command, a series of activities can occur, with the complexity increasing significantly, especially if the feedback from the environment is used to determine which steps of the program the robot will follow. There may be a number of environmental variables, with significant increase in the complexity of the robot program. When this interaction with the environment becomes so complex that it suggests the robot is thinking for itself, such artificial intelligence (AI) may provide the robot with a complex multitude of options for responding to the environment. An annual competition involves designing a vehicle that can navigate on the open road, following a specific route, with no input except the sensors in the robotically controlled vehicle. Another example of complex feedback being integrated into the robotic program is a robotic floor sweeper, which may have a feed-back system so it can decide on a path around the room, while avoiding objects on the floor or change directions when a wall is contacted. If a piece of furniture is moved, the robot can correct its route and remember it during future excursions. An electro-mechanical suturing robot is presently being developed in Houston. It is essentially a hand-held device that passes a suture through tissue on command of the surgeon. Because the feed-back involves the surgeon who positions the robot while looking at the surgical field, initiating a series of steps by pressing a control button and moving on to the next step, the surgeon is part of the feed-back loop, so we refer to the device as a ‘‘semi-robotic’’ suturing device. The robot must have a customized tool, an end effector, mounted at the end of the arm to perform the specific tasks for which the robot has been programmed. It may be welding tips for spot welding on an auto assembly line, a paint sprayer to paint a vehicle part or a wall,
38
or suction cups to pick up a part from an assembly line. A robot designed for surgery may have an end effector designed to hold an endoscope or retractor, an electrode or biopsy needle holder, a small forceps or needle holder, an electrocautery, or a dissecting tool. A robot for use in orthopedic surgery may have a drill or machining assembly mounted at the end of the arm.
When does a Device Become a Robot? There is no exact line separating non-robotic mechanical devices from a robot, and the opinion whether some mechanical device is a robot depends on whether you ask the inventor, the engineer, or the investor. (In that regards, the word ‘‘robot’’ is in the same category as the word ‘‘nano,’’ which Richard Smalley, who won the Nobel Prize for nanotechnology, said only half joking, ‘‘Nanotechnology is anything that bears the name ‘nano’ and makes a lot of money.’’) A surgical or industrial robot usually does not have arms and legs like a person, or even two arms. In fact the most commonly used industrial robot consists mainly of only one arm with a tool at the end (> Figure 38‐1). The tool mounted at the end of the arm is selected depending on the robot’s task, such as welding, painting, picking up or moving a part on an assembly line, or fitting together components of the article being manufactured. The arm of a robot may move by hydraulic or pneumatic pressure that can be finely regulated. Alternatively, a robot may move by electromechanical motors or actuators, which have become more available and precise and controllable as computers have become more sophisticated. The robot joint may move by a fly-by-wire system that that has either a computer control system or the mechanical drive near the part that is to be moved, and such systems operate flight control surfaces in airplanes.
585
586
38
Robotic neurosurgery
. Figure 38‐1 (a) and (b): a model of a Kuka industrial robot, typical of those used in automobile manufacturing and the type used in the CyberKnife
The industrial robot arm (> Figure 38‐1) has multiple joints, usually six or seven, corresponding to degrees of freedom. Often, the joints can be compared with the shoulder, elbow, wrist, or fingers, which can extend, flex, or even rotate. In order for such a robot to move the ‘‘hand’’ smoothly from point A to point B, it may be necessary for all the joints to move sequentially or simultaneously. The path between the points can be defined, and the computer that serves as a robot controller then calculates how much to move each joint to accomplish the action. In order for the robot to interact with its environment, the robot must know precisely where a target or task is in relation to the robot. In surgery, this may take the same form as registering the patient’s head and target to any image guidance device. Registration may be accomplished by the use of surface or inserted fiducials, again like image guided surgery. On a production line, registration may involve determination of the position of the target in relation to the robot body by accurately placing the target at a specific spot relative to the robot, or by providing the robot accurate feedback from the tool at the end of the robot arm so the robot can adjust the localization of the tool accordingly. For instance, in the hair transplantation robot, described
below, both registration of the target on the scalp to the location of the robot arm and identification of the target are done by stereoscopic video cameras mounted on the tool at the end of the robot arm. In neurosurgery, registration may be done by any technique that is used for image-guided or stereotactic surgery to determine the precise position of the patient or organ relative to the robot. What jobs are best done by a robot? Those that are predictable, repetitive, stressful, intense, tedious, and/or spatially oriented. Those that require a repetitive series of steps, even though complex, may qualify. Is short, those characteristics that make a task boring or tedious for a human may make it ideal for a robot. A passive robot is one that attains a position that is to be maintained for a longer period of time than would be comfortable for a surgical assistant. Examples are robots that serve not as the surgeon, but as the surgical assistant, holding retractors or positioning endoscopes. Such robots are easier to develop, and have been the first robots developed for clinical use. An active robot, on the other hand, is programmed to do things (the program) and may have a number of tasks to do during surgery. One approach in the development of robotic surgery is to take stock of how
Robotic neurosurgery
many individual steps are done in a particular type of surgery. Then concentrate on those individual tasks that are repetitive, spatially oriented and tedious and consider whether a robot could do the job better than the surgeon or assistant. When is a robot (or other technologic advance) of use to the neurosurgeon, or to anyone else, for that matter? A scale that has recently been called ‘‘the Gildenberg technology scale’’ has four phases.
Phase 1 – The device can do what the surgeon can do manually, but not as efficiently or not as fast. This is of interest only to the inventor. Phase 2 – The device can do what the surgeon does, just as fast and just as efficiently. This is of interest to the developer and possibly to potential or actual investors. It may be more useful for marketing than for surgery. Phase 3 – The device can do what the surgeon does, but faster or better. This is of interest to other surgeons and to investors, may be cost effective, and may be used at many institutions. Phase 4 – The device can do something the surgeon cannot do without the robot. This is disruptive or revolutionary technology, of interest to all.
There are very few techniques that reach phase 4 (See chapters E-5 and E-14 on the CyberKnife). At present, the CyberKnife may be the only neurosurgical device that reaches that level. It consists of an x-ray tube mounted at the end of an arm of an industrial robot. The targeting and distribution of the radiation cannot be achieved without the robot and its programming, since the non-isocentric program is too complex to be calculated manually. Because of the flexibility of the programming, the same device can be used for stereotactic radiosurgery, hypofractionated radiotherapy as well as conventional radiation therapy throughout the body. One particularly
38
impressive feature is the ability to radiate pulmonary lesions during respiration, since the robot can follow the movement of the tumor on a moment-to-moment basis. The other outstanding example of a phase 4 robotic device is Lasik eye surgery that reconfigures the shape of the cornea in order to correct vision. The device determines the pre-operative error in refraction, calculates the way the cornea should be recontoured, and contours the cornea. Although the surgeon has ultimate control, the procedure cannot be done manually. On the bottom line, however, is that most surgical robots are planned to permit the surgeon to do a better job, and not to replace the surgeon.
Surgical Robots Although there has been great optimism since the early 1990s about the incorporation of robotics into surgery, including neurosurgery [2], progress has been very slow, despite major advances in computer science. Such development is expensive and consequently requires either an investor not afraid of significant risk or an industrial collaboration. It requires a developmental team with a variety of interdigitating specializations, takes a great deal of time to evolve, and confronts many regulatory issues. In addition, there are concerns about intellectual property and patents that make development difficult in the need to avoid infringement on existing patents or to know when infringement may occur [3]. Regulatory concerns are many, since the safety of the robot performing all or part of a procedure is paramount. Advances in other less expensive image guidance technology may supply many of the benefits of robotics at less expense and be sooner to market. Ideally, a surgical robot should be proven to be cost effective before being introduced into the marketplace, by making the surgery more efficient. Even if not cost justified, the term ‘‘robotic’’ is attractive and an excellent
587
588
38
Robotic neurosurgery
marketing tool, so it may find buyers even in Phase 2. Some robots designed for surgery are adaptations of existing industrial robots, which may restrict their capabilities, but make it less expensive to develop. Other surgical robots are designed ‘‘from the ground up,’’ which involves much more laborious and expensive development, but may result in a more functional system. The task of some robots is to attain a certain position and then remain at that position, such as a robot designed as a retractor holder. The position may be modified to another position, which is then held until the robot is commanded to change positions again. Other robots are programmed to perform tasks that require frequent or continuous movements, such as auto industry robots that may apply paint to the outside surface of an automobile. Most robots of interest to neurosurgeons are not configured at all like a person. Most have but a single arm, but a few may have two. The tool at the end of the arm may be a retractor holder, a holder of an endoscope, a holder for insertion of a biopsy cannula or electrode, or even a surgical microscope. Thus, the combination of a robot and an image guidance system may accurately place an electrode or aim an operating microscope at predetermined stereotactic coordinates or at a point-in-space determined from a preoperative CT scan or MRI using targeting technology similar to image guided surgery. The use of a robot to insert an electrode stereotactically may not provide a clinical advantage, even if less than a millimeter improvement of the accuracy of a stereotactic frame is achieved. The robotic procedure may take longer than using a conventional stereotactic frame or image guidance, most of which have a computer program with appropriate planning software. In competition with such computerized but manually operated guidance systems, the surgical robot may not reach a level of cost effectiveness. In evaluating cost-effectiveness, many interacting processes must be considered. Operating room
workflow may be optimized by use of a robot which may decrease operating room time by taking advantage of potentially more efficient technology [4]. Conversely, the set-up time for a robot may disrupt operating room workflow. One of the first surgical uses of a robot was by Sakaguchi [5] in 1985, whose robot performed percutaneous nephrostomy. The localization was performed by establishing an entry point on the skin and a target point in the dilated renal pelvis. Then the trajectory was calculated and the robot programmed to follow that course. A more recent contemporary technique to aspirate or biopsy with a needle involves registration of the target by three-dimensional ultrasound [6,7], with robotic placement of the needle. The surgical robot that presently has dominated the field is the da Vinci robot, made by Intuitive Surgical. A perusal of its design and capabilities may provide a basis for discussing other robots that are being developed for neurosurgical use. Each motion is controlled directly by the surgeon, so one may argue that it is an electro-mechanical or master-slave device rather than a robot per se. It has very few if any internal program sequences but the controls are sophisticated but yet intuitive. The main claim to being a robot is that it allows scaling of movements, so a surgeon may, for instance, move his or her finger one centimeter, but the robot only moves the instrument 1 mm. It may also dampen physiologic tremor to hold the instruments with a steady hand. Surgeons generally feel that the ability to scale movements provides more of a benefit than of dampening physiologic tremor. The presently available commercial version does not have haptic feed-back, although attempts to provide that have been reported [8] and recently have had considerable success [9]. It has, however, what is arguably one of the best stereoscopic endoscopic views of the operative field available. A study demonstrated that suturing with the three-dimensional vision of the da Vinci robot was 65% faster
Robotic neurosurgery
than suturing when the surgeon had a twodimensional view [10]. By using a master-slave control, the surgeon may command the da Vinci system at each movement, which does not require preoperative planning of steps prior to surgery, so the surgeon has the ability to respond to contingencies as they arise. The surgeon sits across the room at a console and looks through two openings at two monitors that together provide exquisite stereoscopic vision. Two or three working channel arms plus the endoscope are manually inserted through access ports. The tools for each working channel are selected and secured to the intended arm. A variety of tools is available and can be changed during the surgery without withdrawing the working channel, ordinarily by a surgeon who is stationed next to the patient. The da Vinci system includes tools for various functions, such as cutting, coagulating, and suturing. That, in combination with direct real-time visualization of the operating field at the business end of the endoscopic tools provides continuous feedback to the surgeon, who exerts moment-to-moment control. Despite minimal if any haptic sensory feed-back, the combination of micromanipulator control over the tools and excellent stereoscopic visualization provides the surgeon with improved capabilities for a number of types of surgery [11]. The da Vinci Surgical Robot was invented by Philip S. Green, of the Stanford Research Institute, who also invented clinical useful ultrasound in the late 1960s [12]. The idea of having a surgeon sit at a console to perform the surgery, which could be miles away from the operating room was named the Green Telepresence System. It was initially called Mona, after da Vinci’s Mona Lisa, but the name was changed to da Vinci in 1999. In 2000 it became the first robotic surgical system to be approved by the FDA. The da Vinci robot is presently in common use in pelvic surgery, especially such procedures as radical prostatectomy [13–15]. The view of that surgical field in that procedure has always
38
been problematic because of the prostate’s position behind the pubic rim, and yet great care must be taken to avoid damage to small nerves and blood vessels overlying it. The Da Vinci provides excellent visualization of the tissue and tools throughout the procedure. Gynecologic surgery has been the next most frequent indication for the use of the Da Vinci [16]. Certain abdominal surgery procedures are now being done robotically [17–19]. Although it was originally intended primarily for thoracic surgery, only recently have techniques been developed for the da Vinci system to allow mitral valve replacement and coronary artery bypass on the beating heart [20,21]. The da Vinci has not to date made significant roads into neurosurgery. There seem to be several reasons for that. First, those surgical exposures which require excellent visualization of critical structures are already done in conventional minimally invasive surgery, often with an operating microscope. Second, the da Vinci arm bearing the endoscopes prevents direct access of the surgeon to the surgical field, which may be critical if an emergency should arise. Third and perhaps most important, neurosurgeons already have a superb stereoscopic view with depth perception deep within the surgical field by means of the stereoscopic operating microscope, which can be aimed with image guidance if necessary, and the target outlined by a heads-up display. Since the surgeon uses the tools held in his or her hand, there is haptic feedback that is not available through the robot. On the other hand, microscopic neurosurgery may be reaching the limits of dexterity of the surgeon’s hand [20], so having the option of scaling the movements may be significant in future procedures. Consequently, neurosurgeons already have excellent stereoscopic microscopic visualization, although the availability of scaling (movement reduction) with instruments free from physiologic tremor may entice neurosurgeons to begin to use the da Vinci system.
589
590
38
Robotic neurosurgery
Robotics is beginning to make inroads into orthopedic surgery. There are three types of orthopedic related robots, one directing a cutting guide block or drilling sleeve, another to constrain the range of movement of a surgical instrument, and one which directs a milling device automatically in order to secure a custom made prosthesis according to the preoperative plan [22]. One of the earliest surgical robotic systems is the RoboDoc, that fits into the third category, When it was introduced in 1992, it was the first robot to assist the orthopedic surgeon in a total hip arthroplasty [23]. It integrates planning and production of both the acetabular cup and the femoral stem, providing a significant advantage over conventional artificial hip implantation. It is used in conjunction with the OrthoDoc Preoperative Planning Workstation to determine the optimal shape of each component of the implantable hip and femur stem to achieve maximal 95% contact with the bone for optimal stability of the prosthesis. A similar system incorporating tremor filter and motion scaling was reported that same year from Johns Hopkins and was called the Steady Hand System. In many respects, this system resembled the da Vinci system [24], which became their model, in that it allowed scaling of movements and filtering of tremor. According to a 2002 report from the company, the intellectual property was assigned to ImageGuide, that began collaboration with GE at that time. There is one robotic surgical technique that I introduced that is still under development that used a smaller version of the kuka industrial robot as the CyberKnife – robotic hair transplantation [25] (> Figure 38‐2). It is based on experience with image guided and stereotactic surgery. It is equivalent to robotically inserting an electrode into a brain target, but you stop at the skull and may do it 2,000 times. A normal non-robotic hair transplantation procedure involves removing a strip of hair-bearing scalp from the occiput. Perhaps four to six technicians with desktop
. Figure 38‐2 Robotic hair transplantation involves planning the procedure in advance, retrieval of individual hair follicles or follicular units, and inserting them into preprogrammed retrieval sites in the bald area
operating microscopes and razor blades trim the tissue until just one follicle or follicular unit is in each piece, which may number 1,000–2,000. The surgeon makes that same number of punctate incisions in the bald part of the scalp, and then inserts each follicular unit graft bearing one or several hair shafts into each incision. The process ordinarily takes 6 h or more. It is tedious, time consuming, stressful, repetitive, and spatially oriented, which makes it a perfect job for a robot. The robot control and programming have several similarities to the CyberKnife program (see chapter E-5), in that it uses an industrial robot with a specifically designed tool at the end of the arm. However, since all the donor follicular units and scalp recipient targets are on or near the surface, registration of the target can be done with stereoscopic video cameras mounted on the tool at the end of the robot arm. The video can provide both the scanning of the head for preoperative planning, registration of the head to the robot, and spacial accuracy of less than one millimeter for follicular unit retrieval. Both the retrieval of grafts and the insertion sites are planned on the same computer. Although
Robotic neurosurgery
the computer holds all the information for the entire procedure, the surgeon still has the ultimate control [26].
Neurosurgical Robots There have been many preliminary reports of the use of robotics in neurosurgery. It is often difficult to know whether those devices were developed into a commercially available system under a different name. Existing companies may acquire a robotic system during pre-clinical development in order to merge some of its technology into their own technology. Further it is somewhat frustrating to review the development of new technology, such as the use of robotics in surgery, since much of the information is proprietary and involves industrial development, so never appears in the surgical literature. These considerations make it particularly difficult to follow each robotic system throughout its development. I have tried to review those robotic systems that were demonstrated to be clinically useful, but some systems merit mention may not be included, and some that are discussed may not attain commercial success. In order for a robotic system to be attractive to neurosurgery and to achieve phase 4, several criteria would have to be met. Foremost, the surgery would have to be both more accurate and faster than a neurosurgeon can do with an operating microscope. The system would be image guided, since it would be in direct competition with manual image guided techniques. It would have to be under the direct control of the surgeon, but add programming of some or all the steps of the surgery so that it would enhance image guidance, rather than interfere with it. It should be adaptable to telerobotic surgery, especially since there are an increasing number of places where there are few neurosurgeons available. It must be cost-effective, since we are headed toward a future of tightening of health
38
care budgets. Several neurosurgical robots under development are still in Phase 1 or Phase 2 of development. Reviews of existing neurosurgical robots have recently been presented by Louw [27] and McBeth [28]. Benabid [29,30] was the first neurosurgeon to report the use of a robot to carry out stereotactic targeting in 1987, when he reported preliminary results with a six-axis robot linked to a stereotactic frame. The probe holder was positioned robotically to reach a target that had been calculated from x-rays and angiograms. By 1992, he had experience in 140 cases, and hailed its use particularly in image guided robotic endoscopy [30]. The NeuroMate (Integrated Surgical Systems, ISS) was the first neurosurgical robot approved by the FDA and consequently the first commercially available. The navigation is based on pre-operative 3-D imaging, and it can be used in either a frame based or frameless configuration. The system can be either ceiling or floor mounted, with accuracy comparable to other image guided or stereotactic techniques [31]. It has been used most frequently to guide an operating microscope [32] or to implant DBS electrodes [33,34]. Its reported use for electrode insertion is similar to using a frame in the manual mode, which would probably put it into phase 2 of the Gildenberg technology scale. The Zeiss MKM stereotactically guided microscope, the Mehrkoordinaten Manipulator (MKM) robotic navigation system for frameless stereotactic procedures, which was introduced in the late 1990s sand was considered by them to be a robotic device in that it is registered to the patient as in other image guided surgery, after which computerized techniques maintain its optical orientation to the surgical field. The accuracy compared favorably to the BRW stereotactic frame [35]. It was one of the first image guided surgery devices and generated a great deal of interest [36]. It provides not only stereotactic optics of a neurosurgical surgical microscope, but superimposes on the view of the surgical field a definition of the target that is not dissimilar
591
592
38
Robotic neurosurgery
to the system that Kelly [37] had introduced as early as 1980. Kelly’s apparatus was a frame-based stereotactically surgical microscope that was manually positioned, and provided the target visible in a heads-up display through the microscope. Kassell [38] linked it to a one of the first remote telepresence system, which he named SuMIT, for Surgical Manipulator Interface Technology. In 1987, Young [39] introduced the Unimation PUMA (Programmable Universal Machine for Assembly) industrial robot modified for neurosurgery, which is presumably the same PUMA 560 that Erich Mu¨he used in 1985 to perform the first robotic-assisted laparoscopic cholecystectomy [40]. A urologic version of that robot was successfully programmed to perform transurethral resection of the prostate (TURP), which, in turn, led to a version specifically related to urologic procedures such as prostatectomies [41]. In neurosurgical use of the PUMA robot, frame-based coordinates were calculated from pre-operative imaging with a dedicated CTscanner. A biopsy was the first robotic image guided procedure using this system [42]. Target coordinates and trajectory were calculated and fed into the robot
control computer, similar to non-robotic image guidance programs. Experience with its use in a series of children for brain tumor resection was reported by Drake (> Figure 38‐3) [43]. The Minerva robot, developed at the University of Lausanne, Switzerland, was described in 1990, and offered the neurosurgeon frameless guidance. The robotic configuration included an intraoperative CTscanner [44]. It has been used clinically for both biopsy and electrode insertion [45–47]. A Scandanavian neurosurgical robot from Finland was introduced in 1986. the Oulu Neuronavigator System/Leksell Index System. A clinical trial was done from 1980–1984 involving 93 successful operations in 77 patients [48]. A version that involved ultrasonic localization of the target was also introduced about the same time [49]. Louw [27] and his group from Canada, which also included Rizun and Sutherland, who is the project director, presented their new neurosurgical robot, the NeuroArm, in 2004 (> Figure 38‐4). It is one of the most sophisticated robots to be used primarily for neurosurgery. The system was designed from the ground up, rather than attempting to convert
. Figure 38‐3 The use of a PUMA robot in pediatric brain tumor resection [43]
Robotic neurosurgery
. Figure 38‐4 The NeuroArm robot neurosurgery [27]
designed
specifically
for
a commercial system to the operating room, which has allowed the developers much more flexibility. Their system would include haptic feedback, the sensation that is conveyed when the surgeon grasps tissue or an instrument [50], in addition to motion scaling and physiologic tremor suppression [51]. It can be used either with one arm in an MRI compatible mode, or as an ambidextrous or two-arm configuration. The surgeon sits at a workstation in an adjacent control room, where he or she has access to several monitors, including a stereoscopic viewer. They use servo control of individual joints of the robot arm, in a master-slave system, similar to the da Vinci. Each arm has eight degrees of freedom, which provides particularly good dexterity. The two-arm configuration is designed for fine
38
microsurgical movements, depending on the tools that are mounted on each arm. The MRI configuration permits only the use of one arm. An area where robotics is of considerable interest at present is telesurgery. If a surgeon can perform surgery when sitting across the operating room, why not provide that capability to a hospital miles away where they have no neurosurgeon? There has been considerable interest in such a system by the military and space programs. As computer communication becomes commonplace, the restrictions of offering such a services to remote locations are becoming less daunting. The NeuRobot was introduced in 2002 by Hongo and Goto in Shinshu University School of Medicine, Matsumoto, Japan [52]. It consisted of a master and a slave micromanipulator and three-dimensional display. By 2003, a telecontrolled manipulator was incorporated [53,54]. In 2003, the use of an AESOP 3000 robot was used for telesurgery between Baltimore and Sao˜ Paulo, Brazil, in one case to hold and direct a laparoscope, and in another case to use a robot specifically designed for access to the kidney [55]. A recent report from China by Tian and his group [56], using their CAS-BH5 robotic system, demonstrated the usefulness of telemanipulation in 10 patients via a digital data network between Beijing and Yan’an, 1,300 km away. Lum, from a group from the University of Washington in Seattle led by Hannaford [57], have recently introduced a light weight robot designed for military mobility, sponsored by the US Defense Advanced Research Projects Agency (DARPA). Its mobility was of primary importance, so it could be transported to the field of battle, for which it underwent a simulated trial in the California desert [58]. Computer Motion, Inc. has developed several surgical robotic systems. One of its premier devices was the voice-controlled Automatic Endoscopic System for Optimal Positioning (AESOP),
593
594
38
Robotic neurosurgery
which allowed the surgeon to direct or move an endoscope or working channel with either a manual mode or with voice commands. A study in 1997 with medical students as subjects showed that there was a steep learning curve at the beginning, but once facility with vocal commanding the robot was achieved, the stability and movement of the endoscope image provided a significant benefit [59]. A report 2 years later concluded that the use of the voice-controlled AESOP robot for placing the endoscope allowed a single surgeon to perform endoscopic mitral valve surgery, a distinct cost advantage [60]. The AESOP was also found to be an advantage in endoscopic pituitary surgery [61]. A recent report added a capability for eye gaze tracking to aim the AESOP robotically positioned endoscope [62]. The AESOP has been used for urology [63], general surgery [64], laparoscopy [65], laryngoscopy [66], but the main use that may affect neurosurgery is the use in the endoscopic approach to the sella [61]. Computer Motion also produced the Zeus system which provided voice control of the endoscope and manipulation through the working channels, which was introduced into the US through Medtronic. Both of these systems were successfully used for endoscopic cholecystectomy [67], coronary artery bypass surgery [26] and gynecologic surgery [68], but it is uncertain whether they were used in neurosurgery. A direct comparison between the AESOP, Zeus, and da Vinci demonstrated that the da Vinci required less operating room time than the other two systems [69]. In 1995, Giorgi [70] presented a preliminary report of a robotically directed operating microscope, developed while he was at the University of Maryland. After he had returned to Italy, in 2000, he presented clinical experience in 14 children with a more sophisticated robotically driven operating microscope. The microscope was registered to the patient’s head with three progressive scan-synchronized infrared cameras mounted around the lens of the microscope,
which could be directed with a six-axis joystick that was used as the microscope handle [71]. Also in 1995, a Robot-Assisted Microsurgery System (RAMS) was developed with NASA collaboration. MRI was integrated into the system. There was a master-slave or micromanipulator control system with six degrees of freedom for three-dimensional manipulation. Outstanding features were an adjustable filter of physiological tremor and adjustable motion scaling which enhanced surgical precision [72]. A hexapod-based robot system was introduced in 2004, the Evolution 1, (Universal Robot Systems, Schwerin, Germany) [73]. It was tracked by the Stealth Neuronavigation System (Medtronic), and was used for endoscopic transsphenoidal surgery with good accuracy and control. It had the advantage of being able to track two instruments simultaneously. An advanced design was introduced in 2007, which also combined frameless stereotaxy, endoscopy, and robotics. The user interface was simplified and additional safety features were introduced [74]. Several robotic systems involving spine surgery are of interest to neurosurgeons. Perhaps the first such system, an ingenious small robot that is attached directly to the spine was introduced for the accurate placement of pedicle and translaminar screws [75]. One of the difficulties in the advance of robotics in endoscopic surgery is the need to miniaturize instruments that can be inserted through working channels [76]. Such devices are becoming more realistic with improvements in miniaturization and computer design. One device that may be useful in neurosurgery, as well as other vascular procedures, is a hand-held robot that streamlines the task of suturing. It is called a semi-robotic suturing device, since the surgeon remains part of the feed-back loop. It is anticipated that the microversion will make vascular suturing more efficient and less stressful to tissues for the expert surgeon. A larger design will be used to provide
Robotic neurosurgery
expertise to those less skilled in surgical techniques who are occasionally called upon in the early management of wounds, such as in the field or on-board ship for the military, where a skilled surgeon may not be immediately available. It is still under development at Suture Robotics in Houston. Where are we in regard to neurosurgical robotics becoming a reality? Although robotics is making inroads into certain surgical procedures, most progress is in visualization of minimally invasive or microsurgical targets in regions that are difficult to visualize or gain access to. The site where this is a particular concern is in the pelvis, and the Da Vinci robot has become a cost effective tool for surgery in that region. It has more recently become useful in thoracic surgery, both because it minimizes the exposure and because it has been successfully adapted to coronary artery and mitral valve procedures. In all these cases, surgery can be done more effectively and efficiently with the Da Vinci or similar system than without. Attempts to introduce robotics into neurosurgery have been less successful than in other specialties. The superb stereoscopic visualization of the surgical field through a small incision is already available with image guided neurosurgery and an operating microscope. To date, there has arguably not been a neurosurgical procedure that can be done more efficiently or more effectively with a robot than without (Phase 3), except the CyberKnife in stereotactic radiosurgery, although non-robotic image guidance provides much of the function usually assigned to a robot. As robotic surgery becomes more sophisticated, as computer programming introduces subroutines that can be done efficiently by robotics, as we reach the limits of manual microsurgical techniques, the ability to scale movements and eliminate physiological tremor will invite robotic participation. The problem of access to limited surgical exposure because of the presence of a robot will recede as smaller
38
and smaller robots with more capability are developed, perhaps with flexible optics that would permit free access to the surgical field and still remain out of the surgeon’s way.
References 1. History of Robots. The History Channel. 4–17–2007. 2. Preising B, Hsia TC, Mittelstadt B. A literature review: robots in medicine. IEEE Eng Med Biol Mag 1991;10:13-22. 3. McLean TR, Torrance AW. Are the Brookhill-Wilk patents impediments to market growth in cybersurgery? Int J Med Robot 2008;4:3-9. 4. Sutherland JV, Van den Heuvel WJ, Ganous T, Burton MM, Kumar A. Towards an intelligent hospital environment: OR of the future. Stud Health Technol Inform 2005;118:278-312. 5. Sakaguchi T. Percutaneous puncture with a robot. Hinyokika Kiyo 1985;31:1265-8. 6. Boctor EM, Choti MA, Burdette EC, Webster Iii RJ. Three-dimensional ultrasound-guided robotic needle placement: an experimental evaluation. Int J Med Robot 2008;4:180-91. 7. Boctor EM, Webster RJ III, Mathieu H, Okamura AM, Fichtinger G. Virtual remote center of motion control for needle placement robots. Comput Aided Surg 2004;9:175-83. 8. Shimachi S, Kameyama F, Hakozaki Y, Fujiwara Y. Contact force measurement of instruments for forcefeedback on a surgical robot: acceleration force cancellations based on acceleration sensor readings. Med Image Comput Comput Assist Interv Int Conf Med Image Comput Comput Assist Interv 2005;8: 97-104. 9. Shimachi S, Hirunyanitiwatna S, Fujiwara Y, Hashimoto A, Hakozaki Y. Adapter for contact force sensing of the da Vinci robot. Int J Med Robot 2008;4:121-30. 10. Badani KK, Bhandari A, Tewari A, Menon M. Comparison of two-dimensional and three-dimensional suturing: is there a difference in a robotic surgery setting? J Endourol 2005;19:1212-5. 11. Bhayani SB, Link RE, Varkarakis JM, Kavoussi LR. Complete davinci versus laparoscopic pyeloplasty: cost analysis. J Endourol 2005;19:327-32. 12. Da Vinci’s Hands. European Patent Office, 2008. 13. Rassweiler J, Hruza M, Teber D, Su LM. Laparoscopic and robotic assisted radical prostatectomy – critical analysis of the results. Eur Urol 2006;49:612-24. 14. Takenaka A, Leung RA, Fujisawa M, Tewari AK. Anatomy of autonomic nerve component in the male pelvis: the new concept from a perspective for robotic
595
596
38 15.
16.
17.
18.
19.
20.
21.
22. 23.
24.
25.
26.
27.
28. 29.
Robotic neurosurgery
nerve sparing radical prostatectomy. World J Urol 2006;24:136-43. Walsh PC. A randomized trial comparing radical prostatectomy with watchful waiting in early prostate cancer. J Urol 2003;169:1588-9. Elliott DS, Chow GK, Gettman M. Current status of robotics in female urology and gynecology. World J Urol 2006;24:188-92. Di Marco DS, Chow GK, Gettman MT, Elliott DS. Robotic-assisted laparoscopic sacrocolpopexy for treatment of vaginal vault prolapse. Urology 2004;63:373-6. Heemskerk J, Van Gemert WG, Greve JW, Bouvy ND. Robot-assisted versus conventional laparoscopic Nissen fundoplication: a comparative retrospective study on costs and time consumption. Surg Laparosc Endosc Percutan Tech 2007;17:1-4. Sanchez BR, Mohr CJ, Morton JM, Safadi BY, Alami RS, Curet MJ. Comparison of totally robotic laparoscopic Roux-en-Y gastric bypass and traditional laparoscopic Roux-en-Y gastric bypass. Surg Obes Relat Dis 2005;1:549-54. Chitwood WR Jr, Nifong LW. Minimally invasive videoscopic mitral valve surgery: the current role of surgical robotics. J Card Surg 2000;15:61-75. Dogan S, Aybek T, Westphal K, Mierdl S, Moritz A, Wimmer-Greinecker G. Computer-enhanced totally endoscopic sequential arterial coronary artery bypass. Ann Thorac Surg 2001;72:610-1. Sugano N. Computer-assisted orthopedic surgery. J Orthop Sci 2003;8:442-8. Cain P, Kazanzides P, Zuhars J, Mittelstadt B, Paul H. Safety considerations in a surgical robot. Biomed Sci Instrum 1993;29:291-94. Leven J, Burschka D, Kumar R, Zhang G, Blumenkranz S, Dai XD, Awad M, Hager GD, Marohn M, Choti M, Hasser C, Taylor RH. DaVinci canvas: a telerobotic surgical system with integrated, robot-assisted, laparoscopic ultrasound capability. Med Image Comput Comput Assist Interv Int Conf Med Image Comput Comput Assist Interv 2005;8:811-8. Riordan T. Patents: implanting hair is tedious, exacting work – the perfect work for a robot. New York Times, C4. 9–15–2003. New York Times, NY, USA. Gildenberg PL. Method and apparatus for stereotactic robotic hair transplantation. USPTO Patent 6, 585,746, 1999. Louw DF, Fielding T, McBeth PB, Gregoris D, Newhook P, Sutherland GR. Surgical robotics: a review and neurosurgical prototype development. Neurosurgery 2004;54:525-36. McBeth PB, Louw DF, Rizun PR, Sutherland GR. Robotics in neurosurgery. Am J Surg 2004;188:68S-75S. Benabid AL, Cinquin P, Lavalle S, Le Bas JF, Demongeot J, De Rougemont J. Computer-driven robot for stereotactic surgery connected to CT scan and magnetic
30.
31.
32. 33.
34.
35.
36. 37.
38.
39. 40. 41.
42.
43.
44.
45.
resonance imaging. Technological design and preliminary results. Appl Neurophysiol 1987;50:153-4. Benabid AL, Lavallee S, Hoffmann D, Cinquin P, Demongeot J, Danel F. Potential use of robots in endoscopic neurosurgery. Acta Neurochir Suppl (Wien) 1992;54:93-7. Li QH, Zamorano L, Pandya A, Perez R, Gong J, Diaz F. The application accuracy of the Neuromate robot–a quantitative comparison with frameless and frame-based surgical localization systems. Comput Aided Surg 2002;7:90-8. Heilbrun MP, McDonald JD. The future of image guided surgery. Clin Neurosurg 2000;46:89-101. Varma TR, Eldridge P. Use of the Neuromate stereotactic robot in a frameless mode for functional neurosurgery. Int J Med Robot 2006;2:107-13. Varma TR, Eldridge PR, Forster A, Fox S, Fletcher N, Steiger M, Littlechild P, Byrne P, Sinnott A, Tyler K, Flintham S. Use of the Neuromate stereotactic robot in a frameless mode for movement disorder surgery. Stereotact Funct Neurosurg 2003;80:132-5. Willems PW, Noordmans HJ, Berkelbach van der Sprenkel JW, Viergever MA, Tulleken CA. An MKMmounted instrument holder for frameless pointstereotactic procedures: a phantom-based accuracy evaluation. J Neurosurg 2001;95:1067-74. Jordan A. Miracle microscope offers hope. Mich Health Hosp 1997;33:30. Kelly PJ, Alker GJ Jr. A method for stereotactic laser microsurgery in the treatment of deep seated CNS neoplasms. Appl Neurophysiol 1980;43:210-5. Kassell NF, Downs JH III, Graves BS. Telepresence in neurosurgery: the integrated remote neurosurgical system. Stud Health Technol Inform 1997;39:411-9. Young RF. Application of robotics to stereotactic neurosurgery. Neurol Res 1987;9:123-8. The story behind robotic-assisted endoscopic surgery. European Patent Office, 2008 www.epo.prg. Davies BL, Hibberd RD, Ng WS, Timoney AG, Wickham JE. The development of a surgeon robot for prostatectomies. Proc Inst Mech Eng 1991;205:35-8. Kwoh YS, Hou J, Jonckheere EA, Hayati S. A robot with improved absolute positioning accuracy for CT guided stereotactic brain surgery. IEEE Trans Biomed Eng 1988;35:153-60. Drake JM, Joy M, Goldenberg A, Kreindler D. Computer- and robot-assisted resection of thalamic astrocytomas in children. Neurosurgery 1991;29:27-33. Glauser D, Flury P, Durr P, Funakubo H, Burckhardt CW, Favre J, Schnyder P, Fankhauser H. Configuration of a robot dedicated to stereotactic surgery. Stereotact Funct Neurosurg 1990;54–55:468-70. Fankhauser H, Glauser D, Flury P, Piguet Y, Epitaux M, Favre J, Meuli RA. Robot for CT-guided stereotactic neurosurgery. Stereotact Funct Neurosurg 1994;63:93-8.
Robotic neurosurgery
46. Glauser D, Fankhauser H, Epitaux M, Hefti JL, Jaccottet A. Neurosurgical robot Minerva: first results and current developments. J Image Guid Surg 1995;1:266-72. 47. Hefti JL, Epitaux M, Glauser D, Fankhauser H. Robotic three-dimensional positioning of a stimulation electrode in the brain. Comput Aided Surg 1998;3:1-10. 48. Heikkinen ER. Stereotactic neurosurgery: new aspects of an old method. Ann Clin Res 1986;18 Suppl 47:73-83. 49. Koivukangas J, Ylitalo J, Alasaarela E, Tauriainen A. Three-dimensional ultrasound imaging of brain for neurosurgery. Ann Clin Res 1986;18 Suppl 47:65-72. 50. Rizun P, Gunn D, Cox B, Sutherland G. Mechatronic design of haptic forceps for robotic surgery. Int J Med Robot 2006;2:341-9. 51. Rizun PR, McBeth PB, Louw DF, Sutherland GR. Robot-assisted neurosurgery. Semin Laparosc Surg 2004;11:99-106. 52. Hongo K, Kobayashi S, Kakizawa Y, Koyama J, Goto T, Okudera H, Kan K, Fujie MG, Iseki H, Takakura K. Neurobot: telecontrolled micromanipulator system for minimally invasive microneurosurgery-preliminary results. Neurosurgery 2002;51:985-8. 53. Goto T, Hongo K, Kakizawa Y, Muraoka H, Miyairi Y, Tanaka Y, Kobayashi S. Clinical application of robotic telemanipulation system in neurosurgery. Case report. J Neurosurg 2003;99:1082-4. 54. Hongo K, Goto T, Miyahara T, Kakizawa Y, Koyama J, Tanaka Y. Telecontrolled micromanipulator system (neurobot) for minimally invasive neurosurgery. Acta Neurochir Suppl 2006;98:63-6. 55. Rodrigues NN Jr, Mitre AI, Lima SV, Fugita OE, Lima ML, Stoianovici D, Patriciu A, Kavoussi LR. Telementoring between Brazil and the United States: initial experience. J Endourol 2003;17:217-20. 56. Tian Z, Lu W, Wang T, Ma B, Zhao Q, Zhang G. Application of a robotic telemanipulation system in stereotactic surgery. Stereotact Funct Neurosurg 2008;86:54-61. 57. Lum MJ, Rosen J, King H, Friedman DC, Donlin G, Sankaranarayanan G, Harnett B, Huffman L, Doarn C, Broderick T, Hannaford B. Telesurgery via Unmanned Aerial Vehicle (UAV) with a field deployable surgical robot. Stud Health Technol Inform 2007;125:313-5. 58. Rosen J, Lum M, Trimble D, Hannaford B, Sinanan M. Spherical mechanism analysis of a surgical robot for minimally invasive surgery–analytical and experimental approaches. Stud Health Technol Inform 2005;111:422-8. 59. Jacobs LK, Shayani V, Sackier JM. Determination of the learning curve of the AESOP robot. Surg Endosc 1997;11:54-5. 60. Mohr FW, Onnasch JF, Falk V, Walther T, Diegeler A, Krakor R, Schneider F, Autschbach R. The evolution of minimally invasive valve surgery–2 year experience. Eur J Cardiothorac Surg 1999;15:233-238. 61. Nathan CO, Chakradeo V, Malhotra K, D’Agostino H, Patwardhan R. The voice-controlled robotic assist scope
62.
63. 64. 65.
66.
67.
68. 69.
70.
71.
72.
73.
74.
75.
76.
38
holder AESOP for the endoscopic approach to the sella. Skull Base 2006;16:123-31. Ali SM, Reisner LA, King B, Cao A, Auner G, Klein M, Pandya AK. Eye gaze tracking for endoscopic camera positioning: an application of a hardware/software interface developed to automate Aesop. Stud Health Technol Inform 2008;132:4-7. Moran ME. Robotic surgery: urologic implications. J Endourol 2003;17:695-708. Hazey JW, Melvin WS. Robot-assisted general surgery. Semin Laparosc Surg 2004;11:107-12. Proske JM, Dagher I, Franco D. Comparative study of human and robotic camera control in laparoscopic biliary and colon surgery. J Laparoendosc Adv Surg Tech A 2004;14:345-8. Alessandrini M, De Padova A, Napolitano B, Camillo A, Bruno E. The AESOP robot system for video-assisted rigid endoscopic laryngosurgery. Eur Arch Otorhinolaryngol 2008;265(9):1121-3. Lunca S, Bouras G, Stanescu AC. Gastrointestinal robotassisted surgery. A current perspective. Rom J Gastroenterol 2005;14:385-91. Advincula AP, Falcone T. Laparoscopic robotic gynecologic surgery. Obstet Gynecol Clin North Am 2004;31:599. Nguan C, Kwan K, Al Omar M, Beasley KA, Luke PP. Robotic pyeloplasty: experience with three robotic platforms. Can J Urol 2007;14:3571-6. Giorgi C, Eisenberg H, Costi G, Gallo E, Garibotto G, Casolino DS. Robot-assisted microscope for neurosurgery. J Image Guid Surg 1995;1:158-63. Giorgi C, Sala R, Riva D, Cossu A, Eisenberg H. Robotics in child neurosurgery. Childs Nerv Syst 2000;16:832-4. Siemionow M, Ozer K, Siemionow W, Lister G. Robotic assistance in microsurgery. J Reconstr Microsurg 2000;16:643-9. Nimsky C, Rachinger J, Iro H, Fahlbusch R. Adaptation of a hexapod-based robotic system for extended endoscopeassisted transsphenoidal skull base surgery. Minim Invasive Neurosurg 2004;47:41-6. Rachinger J, Bumm K, Wurm J, Bohr C, Nissen U, Dannenmann T, Buchfelder M, Iro H, Nimsky C. A new mechatronic assistance system for the neurosurgical operating theatre: implementation, assessment of accuracy and application concepts. Stereotact Funct Neurosurg 2007;85:249-55. Togawa D, Kayanja MM, Reinhardt MK, Shoham M, Balter A, Friedlander A, Knoller N, Benzel EC, Lieberman IH. Bone-mounted miniature robotic guidance for pedicle screw and translaminar facet screw placement: part 2–Evaluluation of system accuracy. Neurosurgery 2007;60:129-39. Menciassi A, Quirini M, Dario P. Microrobotics for future gastrointestinal endoscopy. Minim Invasive Ther Allied Technol 2007;16:91-100.
597
44 Stereotactic and Image Guided Craniotomy E. C. Parker . P. J. Kelly
Introduction In 1906, Robert Henry Clarke and Victor Horsley described the stereotactic method, which consisted of systems for defining the brain in Cartesian coordinates and accessing subcortical structures by means of a three-dimensional positioning device based on that coordinate system [1–2]. Although Clarke had suggested to Horsley that stereotaxis could be used to treat brain tumors, Horsley could not see applications beyond laboratory investigations in animals. Subcortical stereotaxis was not employed in humans until Spiegel and Wycis [3] introduced it in 1947 for functional neurosurgery. Nonetheless, Spiegel and Wycis also predicted the application of stereotaxis for the management of human brain neoplasms. The general availability of L-dopa in the late 1960s brought about a precipitous decline in the number of functional stereotactic procedures performed worldwide until the advent of computed tomography (CT in the 1970s led to renewed interest in stereotactic surgery and prompted many neurosurgeons to rethink their approaches to common intracranial tumors [4–7]. In contrast to projection radiography and ventriculography, which was employed in functional stereotactic procedures, CT was a natural three dimensional data source for tumor stereotaxis that could easily be incorporated into a stereotactic coordinate system. In addition, with CT scanning surgeons could actually see the intracranial tumor target volume directly and
#
Springer-Verlag Berlin/Heidelberg 2009
define its margins. Later, magnetic resonance imaging (MRI) was rapidly incorporated into stereotactic database acquisition techniques. CT-based, and later MRI-based, stereotactic biopsy procedures for the diagnosis of intracranial tumors became commonplace [4,8–10]. These imaging-based biopsy procedures usually target a point whose coordinates are derived from stereotaxic CT scanning or MRI. In addition to biopsy, point stereotasis can be used to center a craniotomy over a superficial lesion or to find a deep lesion. However, the use of point-in-space stereotactic techniques to identify tumor margins is cumbersome and inconvenient. Volumetric stereotactic methods were developed to facilitate the intra operative identification of CT-and MRI-defined tumor borders within a stereotactically defined surgical field and to maintain spatial orientation within irregularly shaped neoplasms [11–14]. In contrast to point-in-space stereotaxis, which requires mathematics no more complicated than addition, subtraction, multiplication, and simple trigonometry, volumetric stereotaxis is mathematically complex and requires interpolations, integrations, elaborate image processing, and reconstructions. Our first volumetric stereotactic procedures were performed utilizing inexpensive manual methods [7]. However, those methods proved to be cumbersome, timeconsuming, and impractical on a day-to-day basis. The incorporation of an operating room computer system rendered these procedures practical and time- and cost-efficient [11,13–15].
680
44
Stereotactic and image guided craniotomy
This chapter will describe the instrumentation and the current methodology for and results of computer-assisted stereotactic extirpation of intra- axial lesions.
Methods Volume in Space Patients undergo stereotactic imaging studies with their heads fixed in CT/MRI-compatible stereotactic head frames. The frames are applied under local anesthesia and are secured to the patient’s skull by means of four flanged carbon fiber pins that are inserted through drill holes made in the out table of the skull into the diploe. Detachable micrometers are used to measure the distance from the end of the carbon fiber pins to the vertical supports of the stereotactic head holder. This provides a fixed reference so that the head holder can be removed after data acquisition and replaced for surgery when convenient. Stereotactic CT scanning is performed as follows: the base ring of the stereotactic head holder fits into a CT table adaptation plate, and a localization system indexes into the base ring. The localization system consists of carbon fiber rods arranged in the shape of the letter N that are located on each side of the patient’s head and anteriorly. Each CT slice then exhibits nine reference marks from which stereotactic coordinates for any pixel on that slice may be calculated. Similarly, MRI examinations are performed utilizing a localizing system that in principle resembles the one developed for CT, except that the localizing system for MRI contains N-shaped localizing devices bilaterally, anteriorly, posteriorly, and superiorly to take advantage of the multiplanar imaging capability of MRI. Stereotactic angiography is performed using a biplane digital angiography (DA) table. Again, an adaptation plate is used for fixation of the stereotactic head holder. The DA localization
system consists of Lucite plates, each containing nine reference markers located bilaterally, anteriorly, and posteriorly, which create 18 reference marks on each anteroposterior and lateral DA image. By using the known location of these reference markers in space, one can determine magnification and stereotactic coordinates for any point in space and cross-correlate CT- and MRI-defined stereotactic target points and interpolated lesional volumes in the correct position on each DA image. In practice, orthogonal and 6-degree stereoscopic pairs are obtained for each stereotactic angiogram. Deep vascular segments identified on the stereotactic angiogram delineate the stereotactic position of all the major sulci and fissures of the cerebral hemispheres. The computer-generated CT, MRI, and DA data are transferred to the operating room computer system (COMPASS configured dual display Linux workstation, Compass International, Rochester, MN). By means of a menu-based program with an intuitive graphic interface (Admiral; Compass International, Rochester, MN) and a mouse subsystem, the surgeon simply traces around the contours of the tumor on serial CT slices and MRI images. The computer program then suspends those slices within a threedimensional image matrix that corresponds directly to the stereotactic coordinate system of the stereotactic frame. An interpolation program creates intermediate slices between the digitized slices and then fills them in with 1-mm cubic voxels, thus creating, for the CT- and MRI-defined limits of the lesion, separate volumes in abstract stereotactic space. This volume can be sliced perpendicular to the intended surgical viewline to present to the surgeon the appearance of the lesion as it will be encountered at surgery. Alternatively, it can be represented by a shaded graphics algorithm, rotating in space to provide the surgeon with a conceptualization of the lesion as a threedimensional volume. The lesional volume scaled to the proper image size also can be displayed in
Stereotactic and image guided craniotomy
the correct location in the stereoscopic angiogram or on a map of the brain sulci and fissures derived from that angiogram. Finally, all these data can be displayed in a shaded graphics rendition of the patient’s skull that has been extracted from the stereotactic CT scan. Such displays are used to plan the stereotactic trajectory to an intracranial lesion to approach and extract a tumor in the safest manner. The best and safest surgical approach to the lesion is one that: (1) preserves important vascular and neuroanatomic structures, (2) traverses nonessential brain parenchyma in a direction parallel to major white matter fiber tracts and, (3) approaches the tumor along its longest axis.
44
. Figure 44-1 Schematic of a simplified COMPASS stereotactic frame, which includes a circular stereotactic holder (a), a 160mm-radius arc quadrant (b), and a three-axes slide mechanism (c) that attaches to a semi-permanent floor stand or a standard operating table. The hand cranks shown in the drawing provide mechanical backups to three-axes computer-controlled stepper motors (not shown)
Surgical Procedures These procedures employ a COMPASS stereotactic frame (> Figure 44-1), occasionally a headsup video display terminal that is attached to the operating microscope (> Figure 44-2), a carbon dioxide laser system (for deep tumors only), and, for the resection of deep tumors, various custom extra-long microsurgical instruments that have been adapted for these procedures. The COMPASS stereotactic frame is basically a Cartesian robotics system in which the patient’s head, while fixed in the stereotactic head holder, is moved in X, Y, and Z space by a stepper motorcontrolled three- dimensional slide system to position the intracranial target volume in the isocenter of a fixed arc quadrant [16]. Surgical trajectories are expressed in terms of setting on the arc quadrant: the Collar (angle from the horizontal plane) and Arc (angle from the vertical plane) angles. Stereotactic coordinates automatically detected by linear optical encoders on each axis of the three-dimensional slide system are relayed to the host computer system in the operating room. Stereotactic coordinates calculated by the computer are executed by means of the stepper motors that are controlled from a
remote location. Electronic and manual backup systems are provided for each automated feature in this system. Computer-generated images of the imagingdefined tumor volume sliced perpendicular to the surgical approach trajectory are displayed on video monitors in the operating room. The slice images can also be projected into a heads-up display unit mounted on the operating microscope, scaled to the exact size of the surgical field viewed through the operating microscope as they are superimposed on the surgical field. Thus, during these procedures, the surgeon views not only the surgical field but also a computergenerated rendition of the CT and MRI defined
681
682
44
Stereotactic and image guided craniotomy
. Figure 44-2 Heads-up display device mounts on a standard operating microscope. This projects computergenerated slice images into one of the oculars of the microscope and allows translation and scaling for precise superimposition of CT/MRI-derived computergenerated tumor slice images onto the surgical field
tumor and its boundaries at any given level in the stereotactic surgical field. The carbon dioxide surgical laser is only useful in vaporizing tissue from deep-seated tumors that are approached and removed through a stereotactically directed 140-cm-long cylindrically shaped retractors 2 cm in diameter (> Figure 44-3) Computer-assisted stereotactic resections can be performed in superficial and deep-seated lesions. In superficial lesions, a circular trephine is turned on a stereotactically placed cranial pilot hole centered over the lesion. The boundaries of the trephine craniotomy with a known configuration and size serves as a reference structure for the indexing of the scaled image within the heads-up display of the operating microscope (> Figure 44-4) or displayed on video monitors in the operating room. The computer-generated
. Figure 44-3 Cylindrical 2-cm-diameter stereotactic retractor mounted on a stereotactic arc quadrant. This provides not only a means of maintaining surgical exposure to deep-seated lesions but also a fixed reference structure whose configuration, size, and location in the stereotactic surgical field are known and to which the computer-generated slice images can be related. Also shown is the dilator device that is inserted into the retractor and used to dilate a subcortical incision
image of the trephine with respect to the CT/ MRI-defined tumor volume is superimposed over the actual trephine in the surgical field. Thus, the computer-generated tumor slice images serve as a template that guides the dissection around subcortical lesions and facilitates identification of the plane between the lesion and the surrounding brain parenchyma. Deep-seated lesions are resected by means of a stereotactically directed cylindrical retractor that is inserted through a dilated cortical and subcortical white matter incision. The incision is made with the carbon dioxide laser and is dilated by means of the retractor-dilator system (> Figure 44-5). The configuration of the deep end of the cylindrical retractor is represented as a circle in the computer-generated slice images (> Figure 44-6) so that it can be superimposed over the actual surgical field by means of the heads-up display unit on the operating microscope. In practice, a plane is developed between the tumor and the surrounding brain tissue before any tumor is removed. Lesions that are
Stereotactic and image guided craniotomy
44
. Figure 44-4 Method employed for the stereotactic resection of a superficial tumor. A trephine opening of the skull is performed, centered on a pilot hole drilled by means of the stereotactic frame. The computer displays the position of the tumor slice in the proper position with respect to the location of the trephine at a specified distance along the viewline on the display monitor and into the heads-up display unit of the operating microscope (A). The image is scaled in the heads-up display, and the microscope is moved until the configuration of the trephine in the image display is exactly the same size as the actual trephine in the surgical field and aligns to the trephine. The surgeon then uses the tumor slice image as a template, to aid in the identification of the surgical plane between CT-and MRI-defined tumor and surrounding brain tissue. This facilitates isolation of the tumor from surrounding brain tissue. (From: Kelly, with permission [16].)
much larger than the retractor can be removed by multiple images translations on the display screen, which result in the calculation of new stereotactic coordinates. Once executed on the stereotactic slide system, these translations position a new part of the tumor under the stereotactic retractor (> Figure 44-7). In practice, a plane is established entirely around the lesion before removing it. Once this plane has been established, the tumor can be partially debulked to allow completion of the deepest aspect of the dissection on the far side of the tumor and removal of the remaining specimen through the retractor tube.
Avoidance of ‘‘Brain Shift’’ Much has been written on the problem of ‘‘brain shift’’ during stereotactically directed
craniotomies. Early on our group was also so concerned about this theoretical problem that we used to place a series stainless steel reference balls though a twist drill cranial opening and a stereotactically directed probe along the stereotactic viewline within a tumoral target volume. AP and lateral stereotactic teleradiographs document their position prior to the craniotomy [14]. With experience, we found this step unnecessary and discontined this practice; the reference balls didn’t shift in their position on serial radiographs in the vast majority of cases. We initially thought that this was because the position of deep tumors didn’t change because of the tethering effect of parenchymal blood vessels or the spatial stabilization resulting from the stereotactic cylindrical retractor. However, there was probably a much more valid reason that ‘‘brain shift’’ has not posed a significant problem in our experience. From our
683
684
44
Stereotactic and image guided craniotomy
. Figure 44-5 Method for extending and dilating a subcortical incision by using the stereotactic retractor and dilator. (a) A sulcus opened microsurgically. (b) An incision made in deep cortex with the retractor cylinder advanced. (c) A subcortical white matter incision deepened with a carbon dioxide laser. (d) An incision dilated with a dilator and a retractor advanced over the dilator. (e) A. retractor cylinder at depths of incision. (f) An incision deepened farther. (g) Retractor advanced, with the incision made to the superficial level of the tumor. (h) Superficial extent of tumor exposed. (i) Retractor advanced to superficial aspects of the tumor
first cases we have always employed the following methodological principles: 1.
The trephine craniotomy is in the least dependent position in the surgical field. The patient’s head (in the stereotactic frame) is rotated so that the vertical approach (arc) angle is close to zero. The
2.
patient is placed in reverse Trendelenberg position until the vertical approach angle is also zero. The dura is not opened until the intracranial pressure was controlled by hypocarbia, reverse Trendelenburg position, barbituates and only a small dose of hyperosmotic agents (rarely required in our experience)
Stereotactic and image guided craniotomy
44
. Figure 44-6 The stereotactic cylindrical retractor is employed during the resection of deep-seated lesions. The computer displays the configuration of a cross section of the retractor (circle) with respect to a selected slice through the tumor volume cut perpendicular to the surgical viewline. This information is displayed on a computer monitor in the operating room as well as in the heads-up display unit of the operating microscope (A). (From: Kelly, with permission [18].)
. Figure 44-7 Method of accessing the extent of a tumor by stereotactic translation. At any set of stereotactic coordinates, the position of the tumor edge with respect to the end of the stereotactic retractor is shown in the computer-generated image. (a) Accessing the posterior margin of the tumor. Slice distance = 10, x = 14, y = 45, z = 20. (b) Accessing the lateral aspect of the tumor. In the COMPASS system, the patient’s head is moved to place a different part of the tumor under the aperture of the stereotactic retractor, which is always directed at the center of the stereotactic arc quadrant. Slice distance = 10, x = 4, y = 35, z = 20
685
686
44 3.
4.
Stereotactic and image guided craniotomy
The ventricular system and tumor cysts are not entered until after the peritumoral plane had been defined surgically. In cases having intraventricular tumors, the cylindrical stereotactic retractor that is used to resect these lesions is, after incising the ependyma, rapidly advanced to the superficial aspect of the tumor in order to minimize the loss of ventricular fluid. Once placed, the retractor prevents the brain from ‘‘sagging’’ or shifting. Tumors are not ‘‘debulked’’ until a plane is developed entirely around the lesion. Here the stereotactic slice images are used to identify that plane within the surgical field. Developing the plane before debulking the tumor mass may, in fact, be contrary to the method that most tumor neurosurgeons use; traditionally surgeons tended to work from the center of the lesion to the periphery. However, if this is done, the tumor will collapse upon itself and ‘‘brain shift’’ WILL occur. Our principle is to first separate the tumor volume from surrounding brain, then remove it. This process may require a few steps in especially large tumors. Here one works in 10–15 mm ‘‘layers’’: first developing the plane down to a specific depth, then removing the tumor tissue in that layer down to that specific depth before developing the plane between tumor and surrounding brain in the next layer. The surgeon thus removes the tumor in defined layers progressing from the most superficial to the deepest.
After removing the tumor the brain does, indeed shift and the cortical surface may, indeed, have dropped a centimeter or so below the inner surface of the skull. But after the tumor is out what difference does this make? In addition, if the craniotomy is in the least dependent position (i.e., the top) of the surgical field, the ‘‘brain shift’’ will usually be along the stereotactically
defined viewline and tumor depth can be monitored by a measured (and from the stereotactic viewline calculations, known) distance from the cortical surface to the superficial slice (aspect) of the tumor.
Patient Selection Selective and accurate resection of any CTor MRI-defined intracranial volume can be performed by employing imaging-based computer-assisted volumetric stereotactic methods. Although the target volume can be an intracranial lesion, volumetric resection techniques were most frequently applied to the most common intraaxial lesions: glial neoplasms in eloquent brain regions. Stereotactic serial biopsy studies have shown that glial neoplasms frequently have two elements: tumor tissue and isolated tumor cells that infiltrate brain parenchyma. The tumor tissue component of high-grade gliomas is most accurately defined by the volume that exhibits contrast enhancement. However, tumor tissue is low-grade (non-pilocytic) gliomas usually is indistinguishable from infiltrated parenchyma on CT and MRI; both are hypodense on CT and do not usually exhibit contrast enhancement. Stereotactic serial biopsy is the only reliable method by which tumor tissue that is hypodense on CT or has prolonged T2 signal characteristics on MRI can be differentiated from infiltrated parenchyma in low-grade (nonpilocytic) astrocytomas, oligodendrogliomas, and mixed gliomas. Stereotactic volumetric resection of infiltrated parenchyma defined by CT/MRI is advisable only in nonessential brain regions. In eloquent brain areas, stereotactic resection is appropriate for the glial tumor tissue component of high-grade glial neoplasms, pilocytic astrocytomas, and low-grade CT-hypodense gliomas in which a stereotactic serial biopsy procedure has confirmed tumor tissue only.
Stereotactic and image guided craniotomy
Specific Lesion Types On the basis of this overall experience, we have been able to draw certain conclusions about surgical approaches for specific anatomic lesions and the selection of appropriate patients for computer-assisted volumetric stereotactic resection. We have developed certain technical maneuvers that are useful in the stereotactic removal of various lesions, depending on their histology and anatomic location.
High-Grade Glial Tumors Computer-assisted stereotactic resection can be used to remove all CT or MRI defined contrastenhancing portions of high-grade glial neoplasms from neurologically important subcortical areas with acceptable levels of mortality and morbidity [12,14,17,18]. Postoperative studies usually demonstrate an absence of contrast enhancement around the surgical defect (> Figure 44-8). Nevertheless, mean postoperative survival of our patients harboring grade 4 astrocytomas treated with postoperative external-beam radiation therapy (50–65 Gy) was 50.6 weeks. This compares favorably to a consecutive series of patients with grade 4 gliomas who underwent radiation therapy after biopsy alone (mean survival, 33 weeks). However, after resection, new areas of contrast enhancement on CT scanning developed in lowdensity areas surrounding the surgical defect within 6–9 months of the procedure. Death in the majority of these cases was therefore due to tumor recurrence and progression. However, quality of survival is better than resection of the lesion and radiation therapy than it is after biopsy alone and radiation. The mean survival time of 50.6 weeks in our patients with grade 4 astrocytomas in central and deep-seated locations (historically associated with poor survival and high surgical morbidity and mortality) is slightly better than the survival
44
times of 37 weeks quoted in other series of the literature [19–22]. Nonetheless, there are two major reasons why our results cannot be compared to historical controls. First, a high percentage of the patients in the other series had lesions in the frontal and temporal lobes, which are more amenable to radical surgical resection by lobectomy [17,22]. By contrast, many of the lesions we treated were centrally located and deep-seated. Second, the survival statistics quoted above after stereotactic resection involve grade 4 astrocytomas only. Readers should exercise caution in comparing these survival statistics to those reported in other series for ‘‘glioblastomas,’’ which in some series include a significant number of grade 3 (Kernohan) neoplasms with a better prognosis. In most historical grading schemes for astrocytomas, the presence or absence of necrosis separates glioblastomas from anaplastic astrocytomas, respectively, but in the Kernohan classification scheme; necrosis can be found in grade 3 and grade 4 astrocytomas. Endothelial proliferation and frequency of mitoses are used to differentiate between grade 3 and grade 4 tumors. Therefore, in most histological classification schemes, glioblastomas can contain some Kernohan grade 3 tumors mixed in with grade 4 neoplasms. Therefore, mean survival in a series of ‘‘glioblastomas’’ (which include grades 3 and 4 astrocytomas) should be longer than mean survival in a series of pure grade 4 astrocytomas. Computer-assisted volumetric stereotactic resection allows safe and complete resection of the contrast-enhancing mass lesion in high-grade gliomas. However, preoperative stereotactic MRI (especially the T2-weighted image) in high-grade glial tumors always demonstrates much larger areas of abnormality than are indicated by contrast enhancement on CT scanning [9,10]. Examination of stereotactic serial biopsy specimens obtained in patients with high-grade gliomas from these MRI-defined abnormalities outside the contrast-enhancing tumor mass reveals a
687
688
44
Stereotactic and image guided craniotomy
. Figure 44-8 Preoperative (a and b) and postoperative (c and d) contrast-enhanced MRI in a patient with deep parietal GBM. Patient was neurologically stable postoperatively
larger area of intact edematous brain parenchyma infiltrated by aggressive isolated tumor cells [8–10,22–24]. In fact, this parenchyma usually extends as far as, and in some cases beyond, the area of signal prolongation abnormality on T2-weighted MRI [9,10]. It would be technically possible, using volumetric stereotaxis, to resect the volume defined by the MRI abnormality, and this in theory would substantially prolong postoperative survival [25,26]. However,
unacceptable neurological deficits would result from removal of the intact, albeit infiltrated, parenchyma. Therefore, in grade 4 astrocytomas, resection of the volume of tissue defined by contrast enhancement permits the most aggressive reduction of tumor burden that allows preservation of neurological function. Radiotherapy and chemotherapy can be used to their best advantage in a patient with the least possible tumor burden.
Stereotactic and image guided craniotomy
These modalities are given to kill the isolated tumor cells that reside in the intact parenchyma that is left behind. However, until we have a means for selectively and safely targeting these isolated tumor cells, we can expect few significant changes in the survival of these patients. A similar problem exists for patients with grade 3 astrocytomas, mixed gliomas, and oligodendrogliomas. Although the cellular elements in these tumors are not as mitotically active as are those in grade 4 tumors, isolated tumor cells also infiltrate intact and surrounding edematous parenchyma [9,10] and defy surgical attempts to cure them. In addition, grade 3 gliomas, particularly astrocytomas, tend to have larger infiltrative components with respect to tumor tissue components than do grade 4 lesions. Therefore, the benefit of resecting a relatively small tumor tissue mass in the face of a large volume of infiltrated parenchyma is questionable; thus, for the most grade 3 lesions, surgeons tend to recommend biopsy over stereotactic resection. In stereotactically resected grade 3 gliomas, tumor recurrences are seen later in the patient’s postoperative course than in the case with grade 4 astrocytomas but recur in the same spatial pattern as that described for grade 4 tumors. Nevertheless, as with grade 4 astrocytomas, the procedure can remove all the solid tumor tissue in selected neoplasms with significant contrastenhancing volume on stereotactic CT scanning.
Low-Grade Astrocytomas The resectability of low-grade astrocytomas depends on the degree of histological circumstances. In adults, these tumors usually are manifest by an area of low density on CT and prolongation of signal on T2 weighted MRI [9]. Stereotactic serial biopsy studies of these socalled fibrillary astrocytomas reveal that the tumor is composed almost entirely of infiltrated intact parenchyma with little tumor tissue
44
proper [9,10]. Therefore, resection of the tumor by stereotactic craniotomy involves resection of intact but infiltrated parenchyma defined by the low-density areas on CT and T2 weighted hyperintensity on MRI. However, in important brain areas, this results in a postoperative neurological deficit, and stereotactic resection is therefore rejected as an option [14,18]. In some cases, the lesion is confined to expendable brain tissue, such as the posterior portion of the superior frontal convolution (> Figure 44-9). These lesions can be resected in their entirety. Pontine astrocytomas are usually fibrillary and are not well circumscribed. They are best biopsied using stereotactic techniques and then treated with radiation therapy. However, radiation therapy can demarcate some of these lesions from pontine parenchyma, creating a zone of neovascularity between necrotic tumor centrally and edematous (usually infiltrated) parenchyma peripherally. The area of central necrosis can be resected stereotactically without inherent neurological risk if the zone of contrast enhancement extends to the floor of the fourth ventricle or far laterally into the middle cerebellar peduncle, facilitating the approach.
Pilocytic Astrocytomas Pilocytic astrocytomas, which tend to occur in children and young adults, are histologically circumscribed. Despite the fact that many are located in the thalamus and other important subcortical locations, they can be resected completely by computer-assisted stereotactic technique with excellent postoperative results [14,18,27,33] (> Figure 44-10). These lesions exhibit prominent enhancement on CT or MR imaging with gadolinium, and the histological borders are defined accurately by the contrast enhancement. As with higher grade gliomas that demonstrate contrast enhancement, this enhancing region contains no functional tissue
689
690
44
Stereotactic and image guided craniotomy
. Figure 44-9 Preoperative (top) and postoperative (bottom) gadolinium-enhanced MRI studies in a patient with a grade 2 mixed glioma involving the posterior portion of the left superior frontal convolution. The patient had no neurological deficit preoperatively or postoperatively. Most of the tumor consisted of intact edematous parenchyma infiltrated by isolated tumor cells
and can be removed without neurological deficit aside from that associated with damage to any normal brain that must be traversed to reach the tumor.
Metastatic Tumors It is generally thought that metastatic tumors are histologically circumscribed and that there is usually no problem identifying the plane between tumor and edematous brain. In some cases, however, this interface may not be clear, especially when bleeding is encountered. In fact, reported surgical series of metastatic tumors removed at nonstereotactic craniotomy report a certain percentage of patients with incomplete resections [28–31].
Many surgeons have had the unsettling experience of trying unsuccessfully to locate deep subcortical metastatic lesions during a conventional craniotomy. Metastatic tumors usually are located at the gray-white junction subcortically. They can be located superficially near the crown of a gyrus. They also can be located at the gray-white junction in the depths of a deep sulcus and can be difficult to find with conventional cramotomy. Other tumors may be deep to the insular cortex, deep to the mesial occipital cortex, or under the cortex of the interhemispheric fissure. Stereotactic techniques can be advantageous in the resection of superficial metastases as well as deeply situated lesions [32]. First, stereotactic point localization helps center small cranial trephines directly over superficial lesions
Stereotactic and image guided craniotomy
44
. Figure 44-10 Thalamic pilocytic astrocytomas in two patients. Anterior thalamic tumors are approached through the anterior limb of the internal capsule (a and b). Dorsal posterior tumors may be reached via the superior parietal lobule (c and d)
(the trephine needs be no larger than the crosssectional area of the neoplasm). The approach is therefore selective and direct, and no more brain than absolutely necessary need be exposed. With volumetric stereotaxis and intraoperative image
displays, identification of the plane between tumor and brain is straightforward and simple. These lesions are readily resectable by either frameless or frame based stereotactic techniques (> Figure 44-11).
691
692
44
Stereotactic and image guided craniotomy
. Figure 44-11 Preoperative (a) and postoperative (b) MRI in a 62-year old man with a metastatic tumor (non-small cell lung cancer) located deep to the motor strip. In this case the tumor was approached by splitting the posterior aspect of the superior frontal sulcus. For metastatic tumors, frameless stereotaxis provides efficient intraoperative navigation, aiding in placement of scalp and bone openings as well as optimizing the trajectory of approach
Our postoperative morbidity for stereotactic resection of centrally located and deep-seated metastatic tumors (mortality, 0; morbidity, 4.3%) compares favorably with that association with conventional craniotomy for these lesions in the past (mortality, 11%) [27,28,30–32]. External-beam radiation therapy has followed surgery in most of our patients to treat possible microscopic metastatic lesions that are not visible on CT. In a 5.5-year experience with computer-assisted stereotactic resection of intracranial metastases at the Mayo Clinic, we have had no known local recurrences of the tumor as determined on serial postoperative CT scans.
Meningioma Meningiomas generally have a very obvious and discrete tumor/brain interface and are often superficially located. Stereotactic techniques can still be quite useful in their removal, however. The general principles of optimizing bony and dural openings
to minimize exposure of normal brain are often especially important when operating on extraaxial masses. Accurate intraoperative navigation often allows opening the scalp with a small linear incision and placing the borders of a craniotomy flap immediately outside the superficial limits of the tumor. When the brain is ‘‘tight’’ because of edema or simply from tumor related mass effect, any exposed brain may quickly herniate out of the dural opening and be strangulated. A properly placed dural incision will be directly over this tumor/brain interface. This serves the purpose of protecting the surrounding brain tissue as well as aiding in the resection. Intracranial pressure will often help force the tumor out through a properly sized dural opening, allowing more rapid dissection of the tumor from the surrounding arachnoid. Although similar spatial information can be obtained by using ultrasound to guide the dural incision, stereotactic techniques provide more precise and usable feedback. The usefulness of stereotaxis to meningiomas and other tumors located at the skull base is of less
Stereotactic and image guided craniotomy
clear benefit. Anatomical landmarks remain the primary cues in locating such tumors as well as important nearby neural and vascular structures. Frameless stereotaxis is often employed for these cases as it is occasionally helpful to retain orientation in the face of highly distorted anatomy.
44
. Figure 44-12 Preoperative (left) and postoperative (right) CT scans in a 29-year old woman with a mesencephalic arteriovenous malformation (AVM) with a hemorrhage. The hematoma and AVM were resected with the 2-cmdiameter stereotactic retractor. The patient was lethargic and had a syndrome of Weber preoperatively. She made an excellent neurological recovery after evacuation of the hematoma and resection of the AVM
Vascular Malformation Closed stereotactic needle biopsy of superficial or deep-seated circumscribed lesions that demonstrate intense contrast enhancement on CT and no perilesional ‘‘edema’’ can be dangerous. These lesions could represent an occult vascular malformation even though arteriography may fail to demonstrate vascularity consistent with an AVM. A postoperative hemorrhage can occur after biopsy of a cryptic AVM or cavernous hemangioma. Computer-assisted stereotactic microsurgical resection provides an alternative to closed stereotactic biopsy and observation. Cryptic AVMs and cavernous hemangiomas are well-circumscribed lesions that can be completely removed stereotactically with relatively low risk. A by-product of establishing the histology is that a cessation or significant reduction of seizures, when present, usually results. Small deep-seated active AVMs also may be resected with similar techniques (> Figure 44-12). The position of the feeding vessels is established in the three-dimensional surgical-planning matrix and is approached and clipped or coagulated before the remainder of the lesion is dissected away from the surrounding parenchyma.
Intraventricular Lesions Intraventricular landmarks can be used to maintain surgical orientation to locate lesions during conventional craniotomies for intraventricular tumors. There usually are few problems with
spatial orientation in patients with large lateral ventricles. However, more difficulty can be encountered staying oriented in small- or normalsized ventricles or, in some case, even finding the ventricle. A more limited but direct approach to intraventricular lesions can be made stereotactically. Brain and ventricular incisions need to be large enough only to remove the lesion. Thus, intraventricular lesions are removed through a 1.5-in trephine and a 2-cm cylindrical retractor (> Figure 44-13 and Video). Large third ventricular lesions usually are approached through the right lateral ventricle. One fornix can be incised to extend the stereotactic retractor into the third ventricular lesion, where an internal decompression of the lesion is performed with a carbon dioxide laser until only a thin rim of the capsule remains. The computer display of the cross sections of the digitized tumor volume are extremely useful in this step, as a surgeon, knowing where the tumor stops and the third ventricular wall begins, can be aggressive within the tumor with no risk of extending through the capsule and damaging the walls of the third ventricle. After this internal
693
694
44
Stereotactic and image guided craniotomy
. Figure 44-13 Preoperative (a and b) and postoperative (c and d) MRI in a woman with a solid intraventricular tumor. The lesion was totally resected utilizing a 2-cm stereotactic retractor and multiple translations. A gross total excision of the lesion was accomplished. The patient did not require a shunt. A video of this surgery is provided on the accompanying DVD
decompression, the retractor is withdrawn to the level of the roof of the third ventricle and the capsule is carefully dissected from the walls of the third ventricle. The tumor capsule can be contracted by using the defocused laser, which facilitates the dissection of the capsule from the wall of the third ventricle.
Colloid cysts are approached through the lateral ventricle and the foramen of Monro in which the cyst shows the greatest extension. The approach features an anterior trephine craniotomy (about at the frontal hairline), splitting of the superior frontal sulcus, and exposure of the cyst in the foramen of Monro by
Stereotactic and image guided craniotomy
means of the 2-cm diameter stereotactic retractor (> Figure 44-14). This approach does not violate eloquent brain tissue and provides an anterior vantage point for dealing with the attachment of the cyst. We have resected 28 colloid cysts in the manner without permanent complication.
44
the representation of a tumor volume in stereotactic space but also demonstrate the relationships of normal vascular and neuroanatomic structures to that volume. In general, the technique allows more thorough removal of intracranial lesions and has less morbidity for patients with centrally located and deep-seated lesions. Without volumetric stereotaxis three things are possible:
Discussion 1. Intracranial mass lesions (tumors) are volumes in space that are readily apparent on review of contiguous CT and MR slice images. However, translation of this three-dimensional information from the imaging studies (CT and MRI) to the three-dimensional surgical operating space in the patient’s head during conventional surgical procedures is difficult and imprecise. In addition, computer reconstructions of stereotactically acquired CT and MRI data not only allow
2.
A surgeon can get lost attempting to find the tumor, and brain tissue is damaged unnecessarily. This can result in a neurological deficit and prolonged and expensive rehabilitation efforts. A surgeon cannot tell where tumor ends and normal brain tissue begins. Thus, there is some risk that the surgeon will resect normal brain tissue along with the tumor. In important brain areas, this will result in a neurological deficit.
. Figure 44-14 A giant colloid cyst in a 56-year old man presenting with obstructive hydrocephalus memory disturbance. Preoperative (a) and postoperative (b). The lesion had a solid interior and was partially calcified. Marked improvement in his recent memory over the preoperative level was noted at the 3-month postoperative examination
695
696
44 3.
Stereotactic and image guided craniotomy
A surgeon performs a subtotal removal of the lesion. Much tumor remains behind and will recur sooner and require another operation later on.
1.
Volumetric stereotaxis provides the following specific advantages: 1. 2. 3.
4.
It allows the surgeon to find the lesion. It imparts a concept of the three-dimensional shape of the lesion that is to be removed. It allows preoperative surgical simulation and surgical approach or trajectory planning with respect to the configuration of the lesion and normal brain and vascular anatomy that must be preserved. Thus, the safest and most effective surgical approach may be selected. It indicates by means of a scaled real-time display, interactive software, and stereotactic instrument where tumor ends and normal brain begins.
2.
3.
Volumetric stereotaxis has major advantages for the patient: 1.
2.
3.
The smallest possible skin incision, craniotomy, and brain incision can be made. This minimizes injury to normal brain tissue. Since the surgeon knows exactly where tumor ends and normal brain begins, a more complete tumor removal can be accomplished with much less risk to surrounding brain tissue. The postoperative neurological results are better than those associated with conventional (nonstereotactic, nonvolumetric) surgical techniques.
Finally, volumetric stereotaxis costs less than standard neurosurgical procedures, since patients get out of the hospital faster, do better neurologically, and return to work earlier. In a practical sense, volumetric stereotaxis saves third-party payers money because:
4.
Less money is spent on intensive care unit charges and postoperative hospital days. Total hospital charges, including surgical fees for patients with astrocytic brain tumors undergoing computer-assisted stereotactic volumetric resection procedures, accounted for approximately 67% of the total hospital charges for conventional surgical procedures in similar patients (Kelly, unpublished data). Volumetric stereotactic procedures require less time in the operating room (2–3 h less in most cases) than do conventional neurosurgical procedures for brain tumors, because the procedures are simulated on a computer system beforehand and can proceed efficiently as planned. This saves money on operating room charges. ‘‘Inoperable’’ tumors (by conventional surgical techniques) can be resected with volumetric stereotactic resection procedures. Frequently, these are deep-seated relatively benign tumors in children and young adults. Many of these tumors can be cured with volumetric stereotaxis, which saves money wasted on radiotherapy and chemotherapy that is not effective for these lesions and on rehabilitation and terminal care as the tumor progressively disables and kills the patient. Neurological results are better, fewer patients require rehabilitation programs, and patients return to work sooner.
Future Directions for Computer-Assisted and Volumetric Stereotactic Surgery In our experience in the development of volumetric and computer-assisted stereotactic surgery, we have seen a tremendous potential for computer assistance of surgery in general. First, it is clear that computers can be used to monitor
Stereotactic and image guided craniotomy
and display to a surgeon the position of surgical instruments in a stereotactically defined work envelope. This will allow more minimally invasive and endoscopic surgery not only in subarachnoid and intraventricular spaces but also for intraaxial target volumes. However, many other techniques will affect the future of neurosurgery in general and surgery for brain tumors and other lesions in particular. These technologies may include electronics, robotics, lasers, and other technologies adapted from industry and the military. They will have the greatest impact in stereotactic neurosurgery, a three-dimensional and mathematically precise surgical discipline that can best exploit techniques (such as endoscopy) borrowed from other fields. Most important, surgical computer systems coupled to surgical instrumentation will make significant advances possible. Since this chapter was originally published, frameless stereotactic equipment and procedures have evolved and matured, and computational power available in workstation and laptop computers has increased dramatically. Most stereotactic brain tumor operations performed today utilize some form of frameless stereotaxy instead of a frame-based system. Modern frameless systems clearly possess sufficient accuracy to allow the safe and effective resection of both intra- and extra-axial tumors. The volumetric techniques described here have been adapted to an electromagnetic tracking based frameless system, the Cygnus PFS (Compass International, Rochester, MN). As with the COMPASS, the target lesion can be digitized by outlining the contours on individual imaging slices. Cross-sectional lesion contours can then be viewed along the axis of the stereotactic probe. For most types of cases, the benefits of volumetric stereotaxis should be realized whether the system utilizes a headframe or some frameless technology. In practice, however, most tumor resections done with the assistance of a frameless system are probably performed using only point in space stereotaxis. In truth, this is easily sufficient for
44
planning the scalp, bone, and dural opening. Once a tumor with an obvious border, such as a meningioma, metastasis, or even a high grade glial tumor, is found, it is not usually difficult to follow the tumor plane and resect the tumor. The stereotactic system can then be used to confirm orientation or proximity to some anatomic structure of interest during tumor resection. For primary, low grade tumors, however, the authors continue to feel that frame based volumetric techniques provide the most efficient and reliable means of surgical navigation. The chance of a poor registration is virtually eliminated, and the visual representation of the tumor contours superimposed on the craniotomy outline provides an intuitive reference for locating and following the tumor border. Additionally, the ability to define the target and surgical trajectory and therefore the location of the craniotomy prior to surgery makes the actual procedure very efficient. It eliminates the need to frequently check the position of a probe or some other registered instrument and compare this with the desired location, a process that can become tedious. Furthermore, deep-seated lesions such as thalamic tumors and intraventricular lesions are most amenable to resection with a cylindrical retractor system like the one described here. After more than 3,000 computer-assisted stereotactic procedures, the methods for computerassisted stereotactic and volumetric surgery described in this chapter are well-developed, and the instrumentation and software are quite mature. Technical support personnel are not required for the day-to-day performance of these procedures. However, continued development of computer-assisted surgery can best be accomplished with full-time technical people. Computer programming is a tedious, painstaking endeavor. Few neurosurgeons will ever have the patience, much less the ability, to write and debug a complex computer program (even though various object-oriented development programs may allow us to manipulate existing programs and customize them for our own purposes).
697
698
44
Stereotactic and image guided craniotomy
Dedicated engineers and computer scientists should become part of academic neurosurgical departments. They will provide the tools for surgery in the future. Surgeons will provide the needed guidance by telling them what we need and what is useful.
References 1. Clark RH, Horsley V. One a method of investigating the deep ganglia and tracts of the central nervous system (cerebellum). Br Med J. 1906;2:1799-1800. 2. Horsley V, Clark RH. The structure and function of the cerebellum examined by a new method. Brain 1908;31:45-124. 3. Spiegel EA, Wycis HT, Marks M, et al. Stereotaxic apparatus for operation on the human brain. Science 1947;106:349-50. 4. Apuzzo MLJ, Savshin JK. Computed tomographic guidance stereotaxis in the management of intracranial mass lesions. Neurosurgery 1983;12:277-85. 5. Brown RA. A computerized tomography-computer graphics approach to stereotaxic localization. Neurosurgery 1979;50:715-20. 6. Goerss S, Kelly PJ, Kall B, et al. A computer tomographic stereotactic adaptation system. Neurosurgery 1982;10: 375-9. 7. Kelly PJ, Alker GJ Jr. A stereotactic approach to deep seated CNS neoplasms using the carbon dioxide laser. Surg Neurol. 1981;15:331-4. 8. Daumas-Duport C, Monsaigngeon V, Szenthe L, et al. Serial stereotactic biopsies: a double histological code of gliomas according to malignancy and 3-D configuration, as an aid to therapeutic decision and assessment of results. Appl Neurophysicol. 1982;45:431-7. 9. Kelly PJ, Daumas-Duport C, Kispert DB, et al. Imagingbased stereotactic serial biopsies in untreated intracranial glial neoplasms. J Neurosurg 1987;66:865-74. 10. Kelly PJ, Daumas-Duport C, Scheithauer BW, et al. Stereotactic histologic correlations of computed tomography and magnetic resonance imaging defined abnormalities in patients with glial neoplasms. Mayo Clin Proc. 1987;62:450-9. 11. Kelly PJ, Alker GJ Jr, Georgss S. Computer assisted stereotactic laser microsurgery for the treatment of intracranial neoplasms. Neurosurgery 1982;10:324-31. 12. Kelly PJ, Alker GJ Jr, Kall B, et al. Precision resection of intra-axial CNS lesions by CT-based stereotactic craniotomy and computer monitored CO2 laser. Acta Neurochir (Wien) 1983;68:1-9. 13. Kelly PJ, Kall BA, Goerss SJ. Transposition of volumetric information derived from computed tomography scanning into stereotactic space. Surg Neurol. 1984;21:465-71.
14. Kelly PJ, Kall B, Goerss S, et al. Computer-assisted stereotaxic resection of intra-axial brain neoplasms. J Neurosurg 1986;64:427-39. 15. Kelly PJ, Kall BA, Goerss SJ. The results of CT based computer assisted stereotactic resection of metastatic intracanial tumors. Neurosurgery 1988;22:7-17. 16. Kelly PJ, Georss SJ, Kall BA. Evolution of contemporary instrumentation for computer-assisted stereotactic surgery. Surg Neurol. 1988;30:204-15. 17. Devaux BC, O’Fallon JR, Kelly PJ. Resection, biopsy and survival in malignant glial neoplasms: a retrospective study of clinical parameters, therapy, and outcome. J Neurosurg. 1993;78:767-75. 18. Kelly P. Volumetric stereotactic surgical resection of intra-axial brain mass lesions. Mayo Clin Proc. 1988; 63:1186-98. 19. Frankel SA, German WJ. Glioblastoma multiforme: Review of 219 cases with regard to natural history, diagnostic methods, and treatment. Neurosurgery 1958;15:489-503. 20. Gehan EA, Walker MD. Prognostic factors for patients with brain tumors. NCR Monogr. 1977;46:189-95. 21. Hitchcock E, Sato F. Treatment of malignant gliomata. J Neurosurg. 1964;21:497-505. 22. Jelsma R, Bucy PC. The treatment of glioblastoma multiforme of the brain. Neurosurgery 1967;27:38-400. 23. Burger, PC, Dubois PJ, Schold SC Jr, et al. Computerized tomographic and pathologic studies of the untreated, quiescent, and recurrent glioblastoma multiforme. Neurosurgery 1983;59:159-68. 24. Daumas-Duport C, Scheithauer BW, Kelly PJ. A histologic and cytologic method for the spatial definition of gliomas. Mayo Clin Proc. 1987;62:435-49. 25. Hoshino T. A commentary on the biology and growth kinetics of low grade and high grade gliomas. J Neurosurg. 1984;27:388-400. 26. Hushino T, Barker M, Wilson CB, et al. Cell kinetics of human gliomas. Neurosurgery 1972;37:15-26. 27. McGirr SJ, Kelly PJ, Scheithauer BW. Stereotactic resection of juvenile pilocytic astrocytomas of the thalamus and basal ganglia. Neurosurgery 1987;20:447-52. 28. Haar F, Patterson RH Jr. Surgery for metastatic intracranial neoplasms. Cancer 1972;30:1241-5. 29. MacGee EE. Surgical treatment of cerebral metastases from lung cancer. The effect on quality and duration of survival. J Neurosurg. 1971;35:416-20. 30. Raskind R, Weiss SE, Manning JJ, et al. Survival after surgical excision of single metastatic brain tumors. AJR 1971;111:323-8. 31. Yardeni D, Reichenthal E, Zucker G, et al. Neurosurgical management of single brain metastasis. Surg Neurol. 1984;21:377-84. 32. Van Eck JHM, Go KG, Ebels EJ. Metastatic tumors of the brain. Psychiatr Neurol Neurochir. 1965;68:443-62. 33. Moshel Y, Link M, Kelly P. Stereotactic volumetric resection of thalamic pilocytic astrocytomas. Neurosurgery 2007;61:66-75.
42 Stereotactic and Image-Guided Biopsy J. B. Elder . A. P. Amar . M. L. J. Apuzzo
Introduction and History Image-guided stereotactic biopsy for histopathologic diagnosis of cranial lesions has become a standard component of the neurosurgical armamentarium [1–3]. The word ‘‘stereotactic’’ derives from the Greek word ‘‘stereos’’ for ‘‘three dimensions,’’ and the Latin word ‘‘tactus’’ for ‘‘to touch’’ [4,5]. In 1908, Horsley and Clarke reported the first stereotactic device in the English literature, which was used to access the dentate nucleus in the cerebellum of monkeys [6]. Nearly 40 years later, in 1947, stereotactic techniques were introduced in humans by Spiegel and Wycis, who used their system for ablative neurosurgical procedures [7,8]. Nearly simultaneously, Leksell developed a separate stereotactic system in 1949 based on the concept of the arcquadrant. By combining the use of stereotactic equipment, a quantitative anatomy atlas and radiographic imaging, these innovators established the concepts on which current techniques and methods are based. Prior to stereotactic techniques, biopsies were conducted during open craniotomies or via free-handed needle aspiration using guidance from indirect radiographic images such as ventriculography and angiography. Over the last 50 years, achievements in mathematics, physics and computing technology have rendered phrenology virtually obsolete, and allow today’s neurosurgeon to visualize and potentially sample the brain and its pathologies with sub-millimeter resolution [9]. With continued advances, stereotactic image-guided biopsies have progressed from subjective techniques whose success #
Springer-Verlag Berlin/Heidelberg 2009
depended largely on the skill and experience of the neurosurgeon, towards an increasingly precise scientific exercise more reliant on measurable factors such as patient- and lesion-specific variables. Early human stereotactic techniques were primarily used for functional neurosurgical procedures such as the treatment of Parkinsonian tremor. Although mortality was greatly reduced compared to open procedures (16% vs. 2%), subsequent advances in medical treatments largely replaced most surgical interventions [9,10]. Reports regarding stereotactic methods for tissue sampling became more common in the 1960s [11,12]. However, stereotactic techniques struggled to gain broad acceptance during this time due to difficulties with image-localization provided by angiography and ventriculography. The advent of the computed tomography (CT) era in the 1970s heralded the renaissance of stereotactic neurosurgery. This technology provided improved visualization of intracranial lesions and allowed the use of patient-specific anatomic data, thus avoiding the problems associated with standardized atlases [13]. Subsequent coordination of the data provided by CT imaging with known reference points in three-dimensional space allowed for the target localization and coordination with the stereotactic apparatus necessary for successful neuronavigation [14,15]. This major development led to increased clinical applications of stereotactic image-guided techniques to intracranial biopsy procedures, such as for neoplastic lesions. Clinical reports, beginning in 1973 with a series of 31 patients who underwent stereotactic biopsy for deep-seated intracranial neoplasms, helped CT-guided stereotactic biopsy gain wider
646
42
Stereotactic and image-guided biopsy
acceptance among neurosurgeons [16]. Subsequently, the late 1970s and early 1980s saw the use of and indications for stereotaxy for lesion biopsy increase, and this technique has now become a standard neurosurgical procedure [17,18]. Since gaining wider acceptance, stereotactic methods have been refined, instruments and devices have improved, and radiographic techniques offer a wider range information and greater detail. This chapter discusses current techniques, tools and concepts governing stereotactic and image-guided biopsy of intracranial lesions. The indications, methods and complications associated with this procedure are reviewed. Specific stereotactic systems are discussed in other chapters, but a brief, general discussion of framebased and frameless systems is offered. Unique considerations based on location or pathology are discussed where they may differ from standard practice. Historical techniques, methods and instruments are also mentioned where their knowledge is felt to be an important element of understanding current stereotactic image-guided practices. Finally, some ideas regarding future tools, directions and concepts in stereotactic brain biopsy are presented.
Indications For patients with an intracranial lesion, the first goal in medical management is to establish a diagnosis. Over the last 30 years, the implementation of new cranial imaging modalities has increased the diagnostic power of imaging alone. However, there continues to be a need for procedures that allow safe acquisition of tissue samples for histopathologic characterization. This is partly because some lesions cannot be definitively characterized on imaging. Previous reports have indicated that a significant percentage of histologic diagnoses made using tissue obtained with stereotactic biopsy were unsuspected based
on radiographic imaging prior to surgery [19]. Some authors report having to alter therapy due to unexpected histopathologic findings obtained via stereotactic biopsy [1,20]. Even if the radiographic studies correctly diagnose the lesion, characterization using traditional histologic classification methods can today be supplemented with information from electron microscopy, as well as molecular and genetic assays [21]. Each of these techniques requires tissue, which indicates a continued role for stereotactic biopsy in the foreseeable future. Histologic characterization using tissue obtained with stereotactic biopsy may have other significant advantages beyond lesion diagnosis. If operative resection is warranted, information from histopathologic examination may help guide surgical planning. For infectious pathologies, antibiotic sensitivities can be assayed using the biopsy tissue to optimize medical management. Patients requiring stereotactic biopsy may also benefit from other types of stereotactic intervention that can occur during the same procedure. For example, patients with liquid components to their lesion, such as pus or blood, may benefit from stereotactic drainage after the biopsy has been performed. In other cases, placing a catheter into a cytic cavity at the time of surgery may allow post-operative, percutaneous drainage for cysts expected to require chronic drainage. These types of disease-specific decisions and actions would be difficult without surgical sampling of the lesion. Once the need for a tissue specimen has been determined, the next decision is whether to obtain the material via open craniotomy or stereotactic image-guided biopsy. The majority of brain lesions revealed using current imaging modalities can be safely sampled and classified histopathologically using standard stereotactic biopsy procedures with a high diagnostic yield [18,22]. In general, stereotactic biopsy, rather than open craniotomy, is indicated if the lesion is surgically inaccessible, deep-seated or in a
Stereotactic and image-guided biopsy
functionally critical area. Examples include lesions in the motor cortex, basal ganglia, corpus callosum or brainstem. Open surgery for lesions in these areas could possibly result in an unacceptable neurologic deficit. Multifocal lesions are likely more appropriate for biopsy alone due to the presumed morbidity associated with attempted resection of multiple lesions. Also, patients with a lesion that would be better treated using noninvasive methods such as chemotherapy or radiation, such as lymphoma or germ cell tumor, should undergo stereotactic biopsy rather than craniotomy to histologically confirm the suspected diagnosis. Most other types of neoplastic brain lesions should be maximally resected [23]. Patient-related factors such as significant medical illness and extreme age may contribute to selection of stereotactic biopsy instead of craniotomy as the operative intervention (> Table 42‐1). In commonly cited large case series reporting stereotactic biopsies of a brain lesions, the most common indications based on postoperative diagnosis were neoplastic diseases [18,24,25]. This included primary central nervous system (CNS) neoplasms such as glioma, as well as metastatic tumors. Other reported etiologies included a variety of intracranial pathologies including infectious processes such as abscess and encephalitis,
42
vascular pathologies such as hemorrhage and cavernous malformation, demyelinating conditions such as multiple sclerosis, and inflammatory conditions such as sarcoidosis [26]. Recommended contraindications to performing a stereotactic biopsy vary based on the literature source, and should be considered on a patient-specific basis. Factors such as HIV status, lesion location, suspected diagnosis and patientrelated factors such as age, comorbidities and treatment preference may play a role in determining the best treatment. In general, lesions associated with significant mass effect should not undergo stereotactic biopsy due to the likelihood of swelling and postoperative neurologic worsening after biopsy. Lesions that are possibly vascular pathologies, such as arteriovenous malformations, should not undergo stereotactic biopsy due to the risk of hemorrhage. For some lesions, radiographic diagnoses can be confirmed by serum or cerebrospinal fluid analysis, which may obviate the need for confirmatory tissue diagnosis prior to planning definitive therapy. Some contraindications may not be specific to stereotactic biopsy. For example, patients with hematologic disorders or difficulties with bleeding may need temporary correction of their disorders prior to undergoing the procedure. Certain medication regimens, such as
. Table 42-1 Indications and contraindications for performing stereotactic image-guided brain biopsy versus open craniotomy
Location
Number of lesions Size Suspected diagnosis
Patient
Possible Indications
Possible Contraindications
Surgically inaccessible Deep-seated (basal ganglia, thalamus, brainstem) Eloquent area Multifocal
Superficial Amidst vascular structures
Very small Infection Lymphoma Germ cell tumor Inflammatory pathology Any for which histologic, molecular or genetic information will affect management Too ill for open resection Extreme age
Very large Gliomatosis Vascular pathology Radiographic diagnosis can be confirmed by less invasive method Multiple lesions in setting of metastatic disease Too ill for any surgical intervention Coagulopathy
647
648
42
Stereotactic and image-guided biopsy
aspirin and plavix, may need to be discontinued temporarily prior to surgery. Other contraindications within specific scenarios are discussed later in the chapter.
Technique and Instruments A number of frame-based and frameless stereotaxy systems appropriate for image-guided stereotactic biopsy are available. Each is described and discussed in a separate chapter in this book. Therefore, only a brief overview of our protocol for stereotactic brain biopsy is presented. Continued development of stereotactic techniques has been conducted at our institution since 1981 with the prototype arc-quadrant Brown-Roberts-Wells (BRW) system and then the Cosman-Roberts-Wells (CRW) system (Radionics, Burlington, MA). Our institution employs the Cosman-Roberts-Wells (CRW) stereotactic system (Radionics, Inc., Burlington, MA, USA) for frame-based CT-guided stereotactic biopsies, and the Leksell G-Frame stereotactic system (AB Elekta Instruments, Stockholm, Sweden) for functional neurosurgery procedures and radiosurgery procedures. Another popular frame-based system is the Brown-Roberts-Wells system (Radionics, Inc., Burlington, MA, USA), which was the predecessor of the CRW system.
Frame-based Stereotactic Biopsy The first step for patients undergoing stereotactic brain biopsy is application of the CRW base frame. Pin sites are selected so that they are outside the imaging plane of the biopsy target, and not in a position to interfere with the anticipated trajectory of the biopsy probe. After the patient receives intravenous sedation, the scalp at the intended pin sites is infiltrated with local anesthetic. Pins are then applied through the scalp and to the skull until ‘‘finger tight.’’ The pins should not be over-tightened and
should be the only part of the frame touching the head. Compression of soft tissue structures such as the cheeks or nose by other parts of the frame may result in pressure necrosis. The second step is image acquisition. Our preference for most frame-based stereotactic brain-biopsies is a contrast enhanced CT. This is acquired after attaching a localizer with N-shaped fiducial bars to the base frame. Although CT resolution is lower than that of MRI, it is typically adequate for most biopsy procedures. Also, CT acquisition time is significantly faster than for MRI. Disadvantages may include artifact associated with the pins, hypersensitivity to the contrast dye, and difficulty targeting lesions that are poorly contrast enhancing. The biopsy target is selected based on multiple factors. The target should be the location most likely to yield diagnostic tissue. Our practice is to select a contrast-enhancing region well inside the borders of the lesion. The trajectory of the biopsy probe should avoid eloquent neural anatomy and vascular structures. Data points are read from the scanner display and documented by both the attending and resident neurosurgeons. This data is then entered into a stereotactic computer in a similar manner. This computer generates the final stereotactic coordinates, which are applied to the stereotactic arc on a phantom base. A stereotactic cannula placed on the arc should coincide with the target coordinates set onto the base. This phantom targeting process is a method of ensuring the accuracy of the coordinates and components of the stereotactic frame. The stereotactic frame is then secured to the base frame. After sterilizing the operative field, the scalp overlying the entry point is identified using a probe fixed to the stereotactic frame and then infiltrated with local anesthetic. A 1 cm incision is made in the scalp. The arc is then rotated into position and a probe passed via the arc to the skull. The probe is then removed and a burr hole is created at the site indicated by the probe using
Stereotactic and image-guided biopsy
a drill guide attached to the stereotactic frame. The dura is punctured and a blunt obturator is inserted into the target site. The desired biopsy instrument is then inserted for tissue sampling. We typically use cupped forceps, but a variety of biopsy instruments may be used, including screw aspirators, needle core instruments, and the side cutting needle. Several passes with the needle may be required to achieve diagnosis [27]. Indeed, some work indicates that the diagnostic accuracy of the procedure improves as the number of biopsy samples taken increases [28]. For lesions that appear heterogeneous on imaging, the needle should be directed to the highest grade portion of the lesion (high contrast enhancement on CT or MRI, hypermetabolism on PET) [4]. Techniques for sampling the lesion may vary depending on the suspected pathology. For example, a tract biopsy can be planned if the lesion is suspected of having areas of different histologic grades. A tract biopsy involves taking multiple biopsies along a tract through the lesion. This technique may optimize histologic grading of a lesion by decreasing the influence of heterogeneity within a tumor [29,30]. Also, for peripherally enhancing lesions suspicious for high grade neoplasms such as glioblastomas, sampling from the margin reduces the incidence of a histologic diagnosis of ‘‘necrosis.’’ Alternatively, tissue at the periphery of the lesion may diagnosed histologically as ‘‘gliosis.’’ After confirmation from the pathologist that adequate diagnostic tissue has been taken, the cannula is removed. The wound is irrigated, and the scalp reapproximated. After removal of the stereotactic frame and base frame, a head CT without contrast is obtained. This allows verification of the biopsy target, typically by demonstrating a small amount of air at the previously identified target site. The CT also evaluates for hemorrhage. Typically, patients remain in the recovery room until they return to their neurologic baseline. They are then transferred to the ward, and discharged the next morning if there are no new problems. Patients with preexisting
42
morbidity or who undergo a complicated biopsy are admitted to the intensive care unit. In general, the vast majority of patients are discharged within 24 h of their procedure. As will be discussed later, postoperative imaging and hospital admission are two areas of debate in the postoperative management of patients undergoing elective stereotactic biopsy [31,32].
Frameless Stereotactic Biopsy Frameless stereotactic techniques initially developed in the 1980s represent an alternative to frame-based systems. The development of this technique has depended largely on advances in computing technology and processing power. Improved cranial imaging permitted the necessary accuracy of three-dimensional volumetric rendering of the brain required for reliable surgical planning. Advances in computing power enabled real-time visualization of cranial anatomy and orientation of a pointing device in three dimensions. Over the last 20 years, decreasing costs of computing equipment and improvements in the underlying technologies have resulted in more wide-spread use of this technology [3,33,34]. The techniques for performing a frameless stereotactic brain biopsy are discussed in greater detail in other chapters. In general, though, the goals of each step are similar. Rather than applying a stereotactic frame to the patient’s skull, fiducial markers or anatomic landmarks serve as reference points. If fiducial markers are used, they are affixed to the skin in a pattern optimal for the anatomical location of the underlying pathology. The word ‘‘fiducial’’ stems from the Latin word meaning ‘‘trust,’’ and implies the faith placed on these markers to accurately maintain their anatomic identity. The patient then undergoes cranial imaging and this data is entered into a computer. The use of anatomic landmarks as reference points allows the imaging to occur a longer time before surgery.
649
650
42
Stereotactic and image-guided biopsy
The reference points are registered as fixed points in space using a pointing device, an external passive or active infrared camera, and the computing interface of the three-dimensional digitizer. Most modern frameless neuronavigation systems are nonlinked, meaning the positional probe is not directly linked to the positional detector. The output from this data entry is the ability to use the pointing device to identify the optimal entry point, trajectory and depth for performing the biopsy. Compared to frame-based systems, fiducials or anatomic landmarks are used in lieu of the localizer attached to the stereotactic frame. Data entry of fiducial bars on the frame is replaced with registration of reference points using the pointing device and infrared optical imaging camera. Thus, the preparation is conceptually quite similar. Fixed points in space with known location relative to cranial anatomy (including the target lesion) are reported as three-dimensional coordinates to a software package that permits surgical planning of the biopsy in terms of entry point, trajectory and depth. One distinction is that the biopsy entry point and trajectory can be visually evaluated intraoperatively after registering the reference points. This allows some degrees of freedom in planning the surgical approach based on the anatomic location of the lesion. Typically, the frameless surgical navigation system is used to position an instrument holder, which, after being fixed in the desired location, is used to stabilize the biopsy instrument. Early reports of frameless neuronavigation systems reported accuracy approaching 4 mm [35,36], whereas recent reports indicate accuracy of approximately 1 mm [37]. The frameless stereotaxy systems used at our institution are the StealthStation-TREON (Medtronic, Louisville, CO) and BrainLAB (BrainLAB, Feldkirchen, Germany) neuronavigation systems. We most commonly use them during functional neurosurgery procedures and open craniotomies, rather than for stereotactic biopsy.
Although frame-based systems represent the ‘‘gold standard’’ in terms of stereotactic brain biopsy, frameless systems have been increasingly utilized and investigated since their inception [30,38]. Some authors maintain that the frameless techniques are gradually replacing frame-based procedures in the same way that freehand brain biopsies were replaced by frame-based stereotactic techniques [3]. Others assert that the three-dimensional accuracy of frameless techniques using skull-applied fiducials can exceed that of stereotactic frames [3]. Initially, frameless systems were used for planning and guidance in open craniotomies. More recently, frameless biopsy techniques and instruments have been refined making this option more practical. Recent work has compared frameless stereotactic biopsy to frame-based techniques, and evaluated for differences in diagnostic yield and complications. A large series in 1999 reported rates of diagnostic biopsy (96.3%), neurologic morbidity (1.4%) and death (1.0%) that were comparable to those reported for frame-based procedures [2]. However, results for posterior fossa biopsy were significantly worse. Other work has also reported that frameless procedures showed no significant differences in terms of diagnostic yield or permanent morbidity when compared to frame-based biopsies. The authors noted that frameless techniques were potentially advantageous for larger or cortical lesions, whereas frame-based stereotaxy was possibly more effective for smaller or deep-seated lesions [38]. For lesions at least 2-cm in diameter, frameless biopsy techniques are likely equal to frame-based systems in terms of targeting accuracy, diagnostic results and complications [39]. The time in the operating room is typically longer for frameless biopsy procedures compared to frame-based biopsies, although the time spent with the head immobilized is usually shorter (> Table 42‐2) [27,39]. Various adjuncts to frame-based and frameless stereotaxy have been developed and incorporated
Stereotactic and image-guided biopsy
. Table 42‐2 Comparison of frame-based versus frameless stereotactic biopsy
Region imaged Image reconstruction Length of cranial immobilization Trajectory Intraoperative time
Frame-based
Frameless
Focal at level of lesion None
Entire head
Longer
Threedimensional Shorter
Fixed Shorter
Free Longer
(Adapted from Bernstein Berger [27,39])
into neurosurgical procedures including biopsy. This includes coordination of multiple imaging modalities, such as ultrasound and endoscopy, into the procedure for real-time verification of target localization. MR images can also be coregistered with CT images prior to biopsy using cranial landmarks, which is reported to increase accuracy [40]. Robot-assisted frameless stereotactic biopsy has also been investigated as another alternative to frame-based systems [41].
Types of Image-Guidance As imaging modalities have improved in terms of sensitivity and specificity, their incorporation into stereotactic procedures has improved the diagnostic accuracy and safety of biopsies, and decreased the need for repeat biopsies. This includes CT technology capable of finer detail, as well as other imaging modalities such as magnetic resonance imaging and positron emission tomography. In addition, advancements in the instruments used for stereotactic biopsy has minimized the technical difficulty and decreased the morbidity of these procedures. Over the last 30 years, computed tomography (CT) has been established as a mainstay in
42
image-guided stereotactic biopsy. Prior to CT, neurosurgeons had to infer the location of intracranial lesions based on displacement of blood vessels using angiography, or distortion of ventricular anatomy using ventriculography [3]. CT technology was initially used to direct freehanded intracranial procedures, thus providing image guidance without stereotactic localization [42,43]. In fact, some authors reported that the diagnostic yield and complications were comparable to those for CT-guided stereotactic procedures [44]. Free-hand CT-guided biopsies were often performed in the CT suite. After placement of a burr hole, a biopsy instrument was advanced and repeated CT scans were taken to determine the proximity of the instrument to the lesion. Naturally, this technique had greater success in large, superficial lesions for which a larger margin for error existed in choosing the trajectory. Despite numerous case series regarding stereotactic CT-guided biopsy, arguments for image-guided free-hand techniques continued to appear in the literature [45]. Ultimately, however, literature regarding stereotactic image-guided brain biopsies demonstrated this technique to be the new standard in terms of diagnostic yield, morbidity and mortality [1,22,46,47]. Eventual acceptance of stereotactic imageguided biopsy as a superior tool for obtaining tissue for histopathologic diagnosis of a brain lesion occurred concurrently with widespread implementation and evaluation of CT technology. Researchers realized that mathematical data from CT images could allow localization of a lesion in three-dimensions relative to predetermined reference points [14,48]. This led to the development of concepts behind the head frame and fiducials. The locations of the fiducials were fixed relative to the frame and their locations relative to the lesion could be decoded on each image slice into mathematical coordinates (e.g., Cartesian coordinates). Further refinements in stereotactic techniques, such as target and trajectory selection, have been aided
651
652
42
Stereotactic and image-guided biopsy
by improvements in CT imaging [3]. For large lesions with distinct target areas, a CT with contrast using standard scan thickness and interscan spacing is adequate for selecting a biopsy site. Smaller lesions may require a decrease in the inter-scan spacing and alteration of the scan thickness so that the target area of the lesion can be identified. Magnetic resonance imaging technology, introduced in the 1980s, enhanced the diagnostic power of non-invasive cranial imaging. Some lesions difficult to visualize using CT can be seen in great detail using standard MRI techniques. The introduction of magnetic resonance (MR) technology fostered interest in evaluating MR-guided stereotactic biopsy as a possibly more accurate alternative to CT-guided procedures [49]. In general, advantages of MR over CT typically include finer detail of brain anatomy and the ability to identify lesions not visualized using standard contrast enhanced CT. Also, the risks of gadolinium are much less than those of CT contrast agents. These properties have resulted in MR as the study of choice for open procedures in which intraoperative neuronavigation is used [37]. For stereotactic biopsies, however, acquiring an MR requires significantly more time than acquiring a CT. Also, previous reports indicated concern due to magnetic susceptibility artifacts causing anatomic distortion, and found a discrepancy of 2 mm when comparing CT and MR coordinates [49]. Other work has shown increased accuracy with coregistration of CT and MR images [40]. Ultimately, the role of MR in stereotactic biopsy depends partly on surgeon preference and partly on how well the target lesion is visualized using both studies. Brain lesions may have significant heterogeneity on imaging that complicates selection of a target. In these cases, accurate target sampling is often crucial for making the correct histologic diagnosis. For example, radiation necrosis can appear radiographically similar to recurrence of a neoplasm. Advanced imaging techniques such
as MR spectroscopy and positron emission tomography (PET) may help differentiate these lesions radiographically. Alternatively, these adjuvant imaging techniques can be coordinated with CT and/or MR and used in target selection for stereotactic biopsy. Currently, the majority of image-guided stereotactic brain biopsies are performed with CT- or MR-guidance. However, these techniques fail to provide a tissue diagnosis in approximately 5% of cases [1]. Some authors feel that inaccuracies associated with CT- or MR-guidance could result in misdiagnosis or incorrect histologic grading [9]. To address these issues, supplemental imaging modalities have been investigated for their usefulness in improving diagnostic accuracy and yield. One example is positron emission tomography, which evaluates the metabolic profile of the target tissue. Data from this imaging modality can be used to identify a target for stereotactic brain biopsy by identifying the abnormal metabolic regions in brain tumors. One study performed stereotactic biopsies using combined CT- and PET-guidance. The authors reported that targets defined using PETwere always diagnostic, whereas 17% of the targets defined using CT were nondiagnostic. Furthermore, some regions that appeared normal on CT yielded diagnostic tissue with PET-guided stereotactic biopsy [50,51]. These results suggest that PET-guidance may increase the diagnostic yield of stereotactic biopsy with no increase in procedure-related morbidity [52]. Similar improvements in accuracy and diagnostic yield have been observed when combining MR- and PET-guidance for stereotactic brain biopsies in pediatric patients [53]. For glioma biopsies, a recent report recommended using 11C-methionine (MET) as the tracer rather than 18F-fluorodeoxyglucose (FDG) because MET provides a more sensitive signal [54,55]. PET is most useful in high-grade lesions because the study depends on increased metabolism to show differential uptake, and lower grade lesions may not demonstrate increased uptake.
Stereotactic and image-guided biopsy
Real-time visualization of the target lesion cannot be achieved using standard image-guided stereotactic techniques unless the operating room is equipped with imaging equipment. This creates concern regarding the effects that ‘‘brain shift’’ may have on specific target localization. Slight alteration of the location of the target lesion relative to the reference point used for stereotaxy may occur for a number of reasons. After imaging, surgical creation of the entry point for the biopsy probe may result in brain relaxation and alteration in the underlying anatomy. Alternatively, administration of anesthetic agents, diuretics or steroids may affect the relative location of the lesion. If the imaging occurs a longer time before surgery, changes in the lesion itself may shift the three-dimensional coordinates of the target. These concerns could be addressed by real-time imaging offered by intra-operative CT- or MR-guided stereotactic biopsy. Alternatively direct visualization using endoscopy or ultrasound may confirm target location. Each technique has been employed for intracranial biopsy, but each also has limitations that prevent universal application in the same fashion as CT- or MR-guided stereotactic biopsy. Early experience with the use of intraoperative CT-guidance for stereotactic biopsies was reported in 1982 [39]. Advantages of intraoperative imaging included the potential for real-time verification that a lesion was accurately targeted and that the biopsy caused no immediate hemorrhage. Intraoperative MRI offers the potential for enhanced anatomic detail, but presents additional difficulties in terms of compatibility of operative equipment. Endoscopy is another category of imaging technology that can be used to assist with intracranial biopsy, and has been described for a variety of intracranial pathologies. Larger series typically discuss both biopsy and resection of intra- or paraventricular lesions such as pineal region tumors [56]. However, some studies suggest that although the rate of serious complications is lower with
42
endoscopic biopsy, this technique is less accurate in terms of diagnosis than frame-based stereotactic biopsy procedures [57]. More recently, simultaneous frameless image-guided and endoscopic neuronavigation was applied to infants with intraventricular pathology [58]. The results indicated that such combined methods could prove advantageous in specific situations. The ability to perform the procedures using frameless neuronavigation was important for avoiding unnecessary procedure-related morbidity. Although most commonly used for intraventricular lesions, endoscopes may also be helpful in verifying complete evacuation of cystic lesions such as abscesses [59]. Additional measure such as third ventriculostomy may take place simultaneously with endoscopic biopsy procedures [57]. Ultrasound has been used as image-guidance in a variety of neurosurgical settings, including ultrasound-guided free-hand biopsy of cortical lesions [45]. The ultrasound probe can be fixed in a specific position and images obtained intraoperatively can be superimposed on preoperatively acquired MR or CT images in order to track brain shift [60]. Comparisons between different imaging techniques used as guidance for lesion biopsy are inconclusive, but some work suggests that frameless MR-guidance may have higher diagnostic accuracy than ultrasound-guided techniques [61].
Complications Although stereotactic biopsy is minimally invasive, the procedure has distinct risks that must be weighed for each patient against the benefits of obtaining histologic diagnosis. The incidence of morbidity due to stereotactic biopsy ranges from 1.0–6.5%, and mortality rates range from 0 to 1.7% in commonly referenced large series [1,25,62]. Patient-specific factors may place varying importance on each risk, and must be considered accordingly. Additionally, a preoperative
653
654
42
Stereotactic and image-guided biopsy
discussion of the risks and advanced directive with the patient and family is important in the unlikely event that the patient is neurologically devastated after the procedure. One potentially severe complication of stereotactic biopsy is hemorrhage. Up to 60% have asymptomatic small hemorrhages [63]. In general, though, the risk of morbidity due to hemorrhage is approximately 1%. However, this risk is higher in malignant pathologies, possibly due to neovascularization, and approaches 6% in patients with GBM [25]. Minimization of the number of biopsy specimens taken and passes with the biopsy probe will help reduce the risk of hemorrhage. Preoperative risk assessment in each patient should help identify patients with higher risk of hemorrhage, such as those in whom a higher grade lesion is suspected. If brisk bleeding is encountered from the biopsy probe, the bleeding may be allowed to continue until it stops spontaneously [64]. If brisk bleeding continues for more than a few minutes, intravenous thrombin can be administered [65]. If the postoperative CT shows a large hemorrhage and the patients has deteriorated neurologically, an immediate craniotomy for hematoma evacuation is likely warranted. Another possible complication of stereotactic biopsy is neurologic worsening not associated with hemorrhage. This neurologic deterioration may be associated with focal neurologic symptoms, such as motor weakness, or an altered mental status without focal signs. There may not be an obvious clinical or radiographic cause for this deterioration, and literature regarding this observation is largely anecdotal. In patients with large tumors, postoperative brain swelling after biopsy may worsen neurologic condition [25]. Most patients recover over several days to a few weeks, but some do not. In such cases, the neurosurgeon must rule out treatable causes of the neurologic worsening [4]. A non-diagnostic biopsy is another negative outcome that must be considered prior to
stereotactic biopsy. Failure to achieve histological or microbiological diagnosis based on the obtained tissue can occur in 8.1% of cases according to a large series from 1995 [27]. Another report showed that the diagnosis based on the first biopsy sample was inaccurate in 33% of cases, but this decreased to 11% by the fourth biopsy specimen [66]. Factors associated with a non-diagnostic biopsy may include immunocompromised patients, non-neoplastic lesions and non-enhancing lesions [66]. If the initial biopsy specimens are non-diagnostic, the instruments and coordinates must be rechecked for accuracy. Based on the planned target and histopathology results from the initial specimens, the biopsy depth may be slightly adjusted for further sampling. For example, if the histology results show necrosis, the depth of the biopsy probe could be reduced to obtain specimens closer to the margin of the lesion. The opposite corrective measure could be taken if the initial histologic examination shows gliosis or normal brain parenchyma.
Pathology The purpose of performing a stereotactic imageguided biopsy is to achieve histopathologic diagnosis. Only a small amount of tissue is obtained with such techniques. Using the cup forceps described above, biopsy specimens are typically 1–2 mm3. Therefore, collaboration with an experienced neuropathologist is critical. In addition, all relevant clinical and radiographic information must be communicated to the pathologist in order to make best use of the small amount of tissue. The initial evaluation of a biopsy specimen commonly involves a smear, except when the specimen is especially firm and will not smear on a slide [22]. The smear involves placing one or more specimens on a glass slide and using a second slide to crush the specimen. The specimen
Stereotactic and image-guided biopsy
is then stained with hematoxylin and eosin (H&E) and examined using light microscopy. If examination fails to provide a diagnosis, a frozen section is performed. This technique involves examination of thin slices of the specimen stained with H&E. Either method of quick section examination, smear or frozen section, can predict the final histologic diagnosis in up to 90% of cases, but the primary goal of such techniques is to ensure that the specimens contain diagnostic tissue, even if a definitive diagnosis cannot be made using quick section [67]. If both the smear and frozen section are non-diagnostic, additional biopsy specimens may be required. At some point, if continued samples are non-diagnostic using quick section techniques, the surgeon will have to re-evaluate the risks and benefits of continued sampling. Ultimately, after conclusion of the biopsy procedure, the remaining tissue samples are fixed in formalin and undergo more definitive evaluation depending on the diagnosis made during quick section. For example, tumors will undergo immunohistochemistry staining for specific antigens such as glial fibrillary acid protein (GFAP) and epithelial membrane antigen (EMA). Infectious lesions may undergo a variety of culturing techniques to isolate the infectious organism and assess antimicrobial efficacy and resistances. Current stereotactic image-guided biopsy techniques in conjunction with an experienced neuropathologist provide the correct diagnosis in up to 95% of cases. In our experience with specimens that were non-diagnostic, necrosis was seen in 45% and inflammation was observed in 41% [1]. Other non-diagnostic results can include gliosis and granuloma. In addition to failure of diagnosis, the histologic results can fail to accurately diagnose the histologic grade of a neoplasm. An example would be a histologic diagnosis of an anaplastic astrocytoma after stereotactic biopsy that is upgraded to glioblastoma based on pathology results after subsequent open resection of the lesion. This type of nonperfect
42
correlation between stereotactic biopsy diagnosis and final diagnosis of the resected tumor would only be proven if the patient underwent a second surgical procedure after their initial stereotactic biopsy. Thus the incidence of this error is difficult to assess. In our series of 30 patients who underwent craniotomy for resection of a lesion after having stereotactic biopsy, the histologic diagnosis achieved with stereotactic biopsy directed clinical management appropriately in 28 patients. However, an exact correlation between histopathology results was observed in only 19 of the 30 patients [22].
Locations Stereotactic image-guided biopsy as described above can be conducted for lesions in nearly any intracranial location [68]. However, the safety and efficacy of the procedure in some anatomic locations has been debated. A few of these are discussed here.
Pineal Region Open surgical exploration of the pineal region can be associated with significant morbidity. However, the diverse collection of possible diagnoses mandates histo pathologic evaluation to guide the treatment plan [69]. Stereotactic approaches to the pineal region are considered by some to carry greater risk of hemorrhage than stereotactic biopsy in other parts of brain due to the proximity of deep draining veins and propensity of some pineal region tumors to bleed. Early work in the 1970s established stereotactic biopsy as a an option in the management of lesions of the pineal/third ventricle region [16,20]. Later work confirmed that stereotactic biopsy of pineal and para-third ventricle lesions could safely guide further medical management
655
656
42
Stereotactic and image-guided biopsy
[1,70]. A large series involving stereotactic biopsy of pineal region lesions demonstrated successful diagnosis in 94% of 370 patients, with a mortality rate of 1.3% and neurologic morbidity of 7.8%. These percentages are comparable to those for other regions of the brain and show that stereotactic image-guided biopsy is a safe and effective method for achieving histologic diagnosis of a pineal region lesion [71]. To avoid injuring veins in the pineal region, the entry point of the biopsy probe should be at least 3 cm anterior to the coronal suture and 3 cm lateral to midline.
Posterior Fossa and Brainstem Lesions located within the brainstem and deep cerebellum represent a broad spectrum of pathologies, and are often a challenge in terms of diagnosis and management. Most cannot be surgically resected, and open biopsy carries a high risk of morbidity and mortality [72]. As with most pathologies, ideal treatment would be guided in part by histologic diagnosis [73]. Stereotactic biopsy of brain stem lesions avoids complications associated with open surgical techniques and is associated with minimal morbidity and mortality rates [74]. Comparatively, it is a minimally invasive and safe technique that provides adequate tissue for histopathologic diagnosis in adults and pediatric patients [75,76]. Furthermore, by providing histologic diagnosis, the technique allows optimization of the patient’s future medical management [77]. Stereotactic biopsy of brainstem lesions can be performed via supratentorial approach or an infratentorial, transcerebellar approach under local anesthesia and intravenous sedation [77]. The infratentorial approach proceeds through a sub-occipital burr hole with the patient in a semi-sitting or prone position. A technique that may reduce the risk of venous air embolism is to place the burr hole while the patient is prone,
and then perform the biopsy with the patient in a sitting position [75]. The supratentorial approach may be performed either through a precoronal frontal burr hole for brain stem lesions, or through a parieto-occipital, transtentorial trajectory for cerebellar lesions. For brain stem biopsies, as the biopsy needle penetrates the brainstem, vital signs such as heart rate, respiratory rate and blood pressure should be carefully monitored. Alterations in the vital signs could indicate compression of brain stem nuclei and the instrument should be withdrawn until the vital signs return to baseline. Lesions in the brain stem and posterior fossa may not be as amenable to CT-guided stereotactic biopsy due to bone artifact, and MR-guided procedures may improve diagnostic yield. Other work showed that combined MR and PET imaging improved the radiologic evaluation of brain stem lesions, but could not replace the clinical guidance provided by histologic diagnosis, which can only be achieved through surgical sampling. This work also demonstrated that combined MR- and PET-guidance improved the accuracy and diagnostic yield of stereotactic biopsies of brain stem lesions [78].
Skull Base Skull base lesions most commonly undergo open resection. Minimally invasive procedures for tissue diagnosis alone are uncommon. However, in some situations patients who are in poor medical condition and cannot tolerate open craniotomy may benefit from stereotactic biopsy. Although little literature exists, reports indicate that stereotactic biopsy can be safely and accurately performed for the diagnosis of skull base lesions [79,80]. This has been demonstrated for pathologies of the cavernous sinus, jugular foramen, and clivus. A transoral stereotactic approach to the second cervical vertebral body has also been described [81].
Stereotactic and image-guided biopsy
Other Considerations Pediatric Histologic diagnosis of an intracranial lesion is especially important in children given the higher sensitivity of their immature brain to radiation and certain types of chemotherapy. Although dedicated pediatric studies regarding stereotactic imageguided biopsy are less numerous than those for adult or mixed patient populations, reports indicate that the technique is a safe and efficacious method for obtaining tissue, even in eloquent areas such as the brain stem [82]. In addition, the possibility of preventing unnecessary adjunctive therapy is worth the risks associated with the procedure if radiographic studies are not definitively diagnostic [83].
HIV/AIDS HIV is associated with several opportunistic intracranial pathologies that can occur in the setting of immunosuppression. Many of these are treatable if diagnosed accurately. Infectious etiologies such as toxoplasmosis are much more likely in this subset of patients. Toxoplasmosis is so common that some authors have recommended an empirical antitoxoplasmosis trial as the first step prior to considering stereotactic biopsy if the cranial imaging is suspicious for toxoplasmosis [96]. A number of other infectious possibilities have been reported in various case series and reports. Because optimal management varies for each type of infection, and because patients are often in poor condition medically, obtaining tissue is critically important to optimizing treatment. Diagnoses such as progressive multifocal leukoencephalopathy are rare in the general population, but are a common etiology of brain lesions in HIV patients. Lymphoma is much more common in HIV patients, and is a common diagnosis
42
after stereotactic biopsy [84]. In one series, lymphoma accounted for 46% of all lesions, followed by progressive multifocal leukoencephalopathy (23%) and toxoplasmosis (15%) [85]. However, the same series reported a high rate (10%) of morbidity and mortality for stereotactic biopsy in HIV patients. Another study reported 2.9% mortality and 8.4% morbidity associated with stereotactic biopsy procedures in HIV patients [86]. Thus, although cerebral lesions are seen in a significant percentage of patients with AIDS, the high complication rate is an important consideration prior to performing a stereotactic biopsy in an HIV patient.
Cystic Lesions Some cases will warrant not only stereotactic biopsy, but also drainage of an associated cystic lesion. In these instances, the target should be selected such that after biopsy of the enhancing border, the probe can be advanced using the same trajectory into the cyst for drainage. Biopsy of the cyst wall should occur first so as not to distort the three-dimensional orientation of the lesion. After diagnostic biopsy is confirmed by the pathologist, the probe can be advanced the appropriate distance for cyst drainage [1].
Postoperative Management Stereotactic biopsies at our institution are inpatient procedures, though the vast majority of patients admitted electively are discharged within 24 h. At some centers, stereotactic biopsy is an outpatient procedure. This practice is reportedly safe for patients with supratentorial lesions if there was no notable intraoperative bleeding or evidence of such on postoperative cranial imaging, and if the postoperative neurologic status of the patient is unchanged [87]. These authors recommended non-enhanced brain CT 2 h after
657
658
42
Stereotactic and image-guided biopsy
the biopsy to verify the absence of hemorrhage larger than 1 cm. Other work has debated the necessity of postoperative imaging to rule out hematoma. In one report, postoperative CT scans did not affect management in the majority (90%) of patients. The authors recommended imaging only for patients with a new postoperative neurologic deficit, or in whom significant bleeding was encountered during surgery [32].
Spine Just as with cranial lesions, certain clinical or radiographic considerations may argue for biopsy rather than surgical resection or non-invasive management for patients with spinal lesions. In these cases, spine lesions can often be successfully sampled with image guidance alone, such as with fluoroscopy or CT-guidance. However, application of stereotactic principles may improve the diagnostic yield from these procedures. Increasingly, modern neuronavigation techniques are being used in spine procedures such as the insertion of instrumentation. Recent reports demonstrate the efficacy of stereotactic devices in spine biopsy procedures [88]. Continued refinement of these techniques may further the development of stereotactic image-guided biopsy of spine lesions.
Non-neoplastic Lesions The majority of intracranial lesions that undergo stereotactic biopsy are tumors [18]. In larger series, non-neoplastic lesions typically comprise up to 20% of diagnoses [89]. These lesions are often infectious, demyelinating or inflammatory disorders. However, the reported diagnostic accuracy of stereotactic biopsy for non-neoplastic lesions is typically lower than that for tumors [90]. This is
possibly due to a greater variability in the pathologic processes underlying non-neoplastic lesions. Despite an increased risk of diagnostic inaccuracy, stereotactic biopsy is a safe and likely underused technique in the management of non-neoplastic cranial lesions that has clear potential benefits in guiding treatment [91,92].
Future Much like the development of stereotactic techniques to this point, continued advances will be influenced to a large degree by technological advances in a broad spectrum of sciences that includes physics, mathematics, computing, materials science and molecular biology. Future innovations will continue progress towards minimizing invasiveness while maximizing efficiency, accuracy and the amount of information gleaned from the biopsy specimens. Advances in intraoperative image guidance may help address concerns regarding brain shift, and lead to increased accuracy and safety of intracranial biopsy procedures. Many large academic centers have operating rooms equipped with MR or CT technology [93]. However, the resulting anatomic detail is typically inferior to scanners in the radiology suites and may not be suitable for real-time image guidance for small or non-enhancing lesions. Intraoperative imaging combined with compatible surgical navigation instruments can allow real-time visualization of the biopsy target. This technique could be useful for very small lesions or those located within or deep to eloquent tissue or vascular structures [94]. Future improvements could include refinements that render these machines less cumbersome while offering higher resolution that allows intraoperative targeting of very small lesions. Another option for real-time target verification is the incorporation of endoscopy and ultrasound technology into the stereotactic biopsy procedure. Combining endoscopy or ultrasound
Stereotactic and image-guided biopsy
devices with stereotactic equipment has been described as a useful imaging adjunct for some biopsy procedures. Continued minimalization of each technology may yield further benefits. One target for modification could be the biopsy probe itself. Integrating ultrasound capabilities into the tip of the probe could enhance the diagnostic accuracy and confidence of stereotactic biopsies by providing intraoperative visual confirmation of target localization. Smaller endoscopes, and flexible versions of both equipped with actuators, could allow for steering of the tip if the real-time feedback indicates that the trajectory is inaccurate or if initial histologic results are non-diagnostic. Linking such real-time images to three-dimensional reconstructions of preoperative images could occur with heads-up or holographic display technology, thus maximizing the benefit of each technology. The biopsy probe could also be ‘‘functionalized’’ in other ways. Advanced tumor therapies will be partly dependent on genetic and molecular subtyping of the lesion, and may be implemented at the time of biopsy using the same surgical access. Instantaneous biochemical feedback could provide real-time information about the type of tissue being encountered by the probe [95]. Biologic assays could detect molecular abnormalities such as high lactate or genetic defects such as a 19q deletion. The probe could also sense the presence of intravenous contrast agents such as gadolinium or alterations in local pressure. Analysis of these data could yield guidance regarding optimal biopsy location, and offer early insight into the histologic and biomolecular profile of the lesion. As imaging modalities continue to improve in terms of their resolution and diagnostic capability, the role for stereotactic tissue biopsy may change. Traditional histopathology and immunohistochemistry techniques might not be necessary if future imaging technologies can non-invasively diagnose the lesion with equal or greater specificity. However, new biomolecular and genetic assays important for treatment
42
of intracranial lesions continue to be developed. Thus, for the foreseeable future, there will likely continue to be a need to acquire tissue from intracranial locations via the least invasive method possible.
Conclusion Stereotactic image-guided biopsy is the least invasive technique for obtaining issue from an intracranial lesion. Lesions of nearly all cranial regions can be sampled with minimal trauma to surrounding neural structures. Frame-based and frameless systems available today are easy to use and provide high diagnostic accuracy with minimal morbidity. Although the diagnostic power of non-invasive imaging techniques is likely to increase in the future, advances in histopathologic and genotyping techniques will perpetuate the need for neurosurgeons to be familiar with stereotactic principles and methods.
References 1. Apuzzo ML, Chandrasoma PT, Cohen D, Zee CS, Zelman V. Computed imaging stereotaxy: experience and perspective related to 500 procedures applied to brain masses. Neurosurgery 1987;20:930-7. 2. Barnett GH, Miller DW, Weisenberger J. Frameless stereotaxy with scalp-applied fiducial markers for brain biopsy procedures: experience in 218 cases. J Neurosurg 1999;91:569-76. 3. Germano IM. Advanced techniques in image-guided brain and spine surgery. New York: Thieme; 2002. 4. DeAngelis L, Gutin P, Leibel S, Posner J. Intracranial tumors: diagnosis and treatment. London: Martin Dunitz; 2002;411. 5. Gildenberg PL. Stereotactic versus stereotaxic. Neurosurgery 1993;32:965-6. 6. Horsley V, Clarke R. The structure and function of the cerebellum examined by a new method. Brain 1908;31:45-124. 7. Gildenberg PL. Spiegel and Wycis – the early years. Stereotact Funct Neurosurg 2001;77:11-16. 8. Spiegel EA, Wycis HT, Marks M, Lee AJ. Stereotaxic apparatus for operations on the human brain. Science 1947;106:349-50.
659
660
42
Stereotactic and image-guided biopsy
9. Black PM, Loeffler JS. Cancer of the nervous system. Philadelphia, PA: Lippincott Williams & Wilkins; 2005. 10. Kelly PJ. Applications and methodology for contemporary stereotactic surgery. Neurol Res 1986;8:2-12. 11. Housepian EM, Pool JL. The accuracy of human stereoencephalotomy as judged by histological confirmation of roentgenographic localization. J Nerv Ment Dis 1960;130:520-5. 12. Kalyanaraman S, Gillingham FJ. Stereotaxic biopsy. J Neurosurg 1964;21:854-8. 13. Iskandar BJ, Nashold BS Jr. History of functional neurosurgery. Neurosurg Clin N Am 1995;6:1-25. 14. Bergstrom M, Greitz T. Stereotaxic computed tomography. AJR Am J Roentgenol 1976;127:167-70. 15. Chen TC, Rabb C, Apuzzo ML. Complex technical methodologies and their applications in the surgery of intracranial meningiomas. Neurosurg Clin N Am 1994;5:261-81. 16. Conway LW. Stereotaxic diagnosis and treatment of intracranial tumors including an initial experience with cryosurgery for pinealomas. J Neurosurg 1973;38:453-60. 17. Lunsford LD, Coffey RJ, Cojocaru T, Leksell D. Imageguided stereotactic surgery: a 10-year evolutionary experience. Stereotact Funct Neurosurg 1990;54–55:375-87. 18. Yu X, Liu Z, Tian Z, Li S, Huang H, Xiu B, Zhao Q, Liu L, Jing W. Stereotactic biopsy for intracranial spaceoccupying lesions: clinical analysis of 550 cases. Stereotact Funct Neurosurg 2000;75:103-8. 19. Friedman WA, Sceats DJ Jr, Nestok BR, Ballinger WE Jr. The incidence of unexpected pathological findings in an image-guided biopsy series: a review of 100 consecutive cases. Neurosurgery 1989;25:180-4. 20. Pecker J, Scarabin JM, Vallee B, Brucher JM. Treatment in tumours of the pineal region: value of sterotaxic biopsy. Surg Neurol 1979;12:341-8. 21. Bouvier C, Roll P, Quilichini B, Metellus P, Calisti A, Gilles S, Chinot O, Fina F, Martin PM, Figarella-Branger D. Deletions of chromosomes 1p and 19q are detectable on frozen smears of gliomas by FISH: usefulness for stereotactic biopsies. J Neurooncol 2004;68:141-9. 22. Chandrasoma PT, Smith MM, Apuzzo ML. Stereotactic biopsy in the diagnosis of brain masses: comparison of results of biopsy and resected surgical specimen. Neurosurgery 1989;24:160-5. 23. Kondziolka D, Lunsford LD. The role of stereotactic biopsy in the management of gliomas. J Neurooncol 1999;42:205-13. 24. Apuzzo ML, Sabshin JK. Computed tomographic guidance stereotaxis in the management of intracranial mass lesions. Neurosurgery 1983;12:277-85. 25. Bernstein M, Parrent AG. Complications of CT-guided stereotactic biopsy of intra-axial brain lesions. J Neurosurg 1994;81:165-8. 26. Grand S, Hoffmann D, Bost F, Francois-Joubert A, Pasquier B, Le Bas JF. Case report: pseudotumoral brain lesion as the presenting feature of sarcoidosis. Br J Radiol 1996;69:272-5.
27. Soo TM, Bernstein M, Provias J, Tasker R, Lozano A, Guha A. Failed stereotactic biopsy in a series of 518 cases. Stereotact Funct Neurosurg 1995;64:183-96. 28. Jain D, Sharma MC, Sarkar C, Deb P, Gupta D, Mahapatra AK. Correlation of diagnostic yield of stereotactic brain biopsy with number of biopsy bits and site of the lesion. Brain Tumor Pathol 2006;23:71-5. 29. Daumas-Duport C, Monsaingeon V, Szenthe L, Szikla G. Serial stereotactic biopsies: a double histological code of gliomas according to malignancy and 3-D configuration, as an aid to therapeutic decision and assessment of results. Appl Neurophysiol 1982;45:431-7. 30. Woodworth G, McGirt MJ, Samdani A, Garonzik I, Olivi A, Weingart JD. Accuracy of frameless and framebased image-guided stereotactic brain biopsy in the diagnosis of glioma: comparison of biopsy and open resection specimen. Neurol Res 2005;27:358-62. 31. Boulton M, Bernstein M. Outpatient brain tumor surgery: innovation in surgical neurooncology. J Neurosurg 2008;108:649-54. 32. Warnick RE, Longmore LM, Paul CA, Bode LA. Postoperative management of patients after stereotactic biopsy: results of a survey of the AANS/CNS section on tumors and a single institution study. J Neurooncol 2003;62:289-96. 33. Barnett GH, Kormos DW, Steiner CP, Weisenberger J. Intraoperative localization using an armless, frameless stereotactic wand. Technical note. J Neurosurg 1993;78:510-4. 34. Watanabe E, Watanabe T, Manaka S, Mayanagi Y, Takakura K. Three-dimensional digitizer (neuronavigator): new equipment for computed tomography-guided stereotaxic surgery. Surg Neurol 1987;27:543-7. 35. Gumprecht HK, Widenka DC, Lumenta CB. BrainLab vectorvision neuronavigation system: technology and clinical experiences in 131 cases. Neurosurgery 1999;44:97-104; discussion 104-5. 36. Muacevic A, Uhl E, Steiger HJ, Reulen HJ. Accuracy and clinical applicability of a passive marker based frameless neuronavigation system. J Clin Neurosci 2000;7:414-8. 37. Jung TY, Jung S, Kim IY, Park SJ, Kang SS, Kim SH, Lim SC. Application of neuronavigation system to brain tumor surgery with clinical experience of 420 cases. Minim Invasive Neurosurg 2006;49:210-5. 38. Woodworth GF, McGirt MJ, Samdani A, Garonzik I, Olivi A, Weingart JD. Frameless image-guided stereotactic brain biopsy procedure: diagnostic yield, surgical morbidity, and comparison with the frame-based technique. J Neurosurg 2006;104:233-7. 39. Bernstein M, Berger MS. Neuro-oncology: the essentials. New York: Thieme; 2000. 40. Cohen DS, Lustgarten JH, Miller E, Khandji AG, Goodman RR. Effects of coregistration of MR to CT images on MR stereotactic accuracy. J Neurosurg 1995;82:772-9. 41. Willems PW, Noordmans HJ, Ramos LM, Taphoorn MJ, Berkelbach van der Sprenkel JW, Viergever MA,
Stereotactic and image-guided biopsy
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
Tulleken CA. Clinical evaluation of stereotactic brain biopsies with an MKM-mounted instrument holder. Acta Neurochir (Wien) 2003;145:889-97; discussion 897. Hahn JF, Levy WJ, Weinstein MJ. Needle biopsy of intrancranial lesions guided by computerized tomography. Neurosurgery 1979;5:11-15. Levy WJ, Hahn JF. Inexpensive plastic brain biopsy needle for use in the computed tomographic scanner. Neurosurgery 1980;7:606-7. Goldstein S, Gumerlock MK, Neuwelt EA. Comparison of CT-guided and stereotaxic cranial diagnostic needle biopsies. J Neurosurg 1987;67:341-8. Di Lorenzo N, Esposito V, Lunardi P, Delfini R, Fortuna A, Cantore G. A comparison of computerized tomography-guided stereotactic and ultrasound-guided techniques for brain biopsy. J Neurosurg 1991;75:763-5. Gomez H, Barnett GH, Estes ML, Palmer J, Magdinec M. Stereotactic and computer-assisted neurosurgery at the Cleveland clinic: review of 501 consecutive cases. Cleve Clin J Med 1993;60:399-410. Lee T, Kenny BG, Hitchock ER, Teddy PJ, Palividas H, Harkness W, Meyer CH. Supratentorial masses: stereotactic or freehand biopsy? Br J Neurosurg 1991;5:331-8. Brown RA. A computerized tomography-computer graphics approach to stereotaxic localization. J Neurosurg 1979;50:715-20. Kondziolka D, Dempsey PK, Lunsford LD, Kestle JR, Dolan EJ, Kanal E, Tasker RR. A comparison between magnetic resonance imaging and computed tomography for stereotactic coordinate determination. Neurosurgery 1992;30:402-6; discussion 406-7. Levivier M, Goldman S, Pirotte B, Brucher JM, Baleriaux D, Luxen A, Hildebrand J, Brotchi J. Diagnostic yield of stereotactic brain biopsy guided by positron emission tomography with [18F]fluorodeoxyglucose. J Neurosurg 1995;82:445-52. Pirotte B, Goldman S, Brucher JM, Zomosa G, Baleriaux D, Brotchi J, Levivier M. PET in stereotactic conditions increases the diagnostic yield of brain biopsy. Stereotact Funct Neurosurg 1994;63:144-9. Pirotte B, Goldman S, Bidaut LM, Luxen A, Stanus E, Brucher JM, Baleriaux D, Brotchi J, Levivier M. Use of positron emission tomography (PET) in stereotactic conditions for brain biopsy. Acta Neurochir (Wien) 1995;134:79-82. Pirotte B, Goldman S, Salzberg S, Wikler D, David P, Vandesteene A, Van Bogaert P, Salmon I, Brotchi J, Levivier M. Combined positron emission tomography and magnetic resonance imaging for the planning of stereotactic brain biopsies in children: experience in 9 cases. Pediatr Neurosurg 2003;38:146-55. Pirotte B, Goldman S, Massager N, David P, Wikler D, Lipszyc M, Salmon I, Brotchi J, Levivier M. Combined use of 18F-fluorodeoxyglucose and 11C-methionine in 45 positron emission tomography-guided stereotactic brain biopsies. J Neurosurg 2004;101:476-83.
42
55. Pirotte B, Goldman S, Massager N, David P, Wikler D, Vandesteene A, Salmon I, Brotchi J, Levivier M. Comparison of 18F-FDG and 11C-methionine for PET-guided stereotactic brain biopsy of gliomas. J Nucl Med 2004;45:1293-8. 56. Oi S, Shibata M, Tominaga J, Honda Y, Shinoda M, Takei F, Tsugane R, Matsuzawa K, Sato O. Efficacy of neuroendoscopic procedures in minimally invasive preferential management of pineal region tumors: a prospective study. J Neurosurg 2000;93:245-53. 57. O’Brien DF, Hayhurst C, Pizer B, Mallucci CL. Outcomes in patients undergoing single-trajectory endoscopic third ventriculostomy and endoscopic biopsy for midline tumors presenting with obstructive hydrocephalus. J Neurosurg 2006;105:219-26. 58. Mangano FT, Limbrick DD Jr, Leonard JR, Park TS, Smyth MD. Simultaneous image-guided and endoscopic navigation without rigid cranial fixation: application in infants: technical case report. Neurosurgery 2006;58: ONS-E377; discussion ONS-E377. 59. Longatti P, Perin A, Ettorre F, Fiorindi A, Baratto V. Endoscopic treatment of brain abscesses. Childs Nerv Syst 2006;22:1447-50. 60. Giorgi C, Casolino DS. Preliminary clinical experience with intraoperative stereotactic ultrasound imaging. Stereotact Funct Neurosurg 1997;68:54-8. 61. Jain D, Sharma MC, Sarkar C, Gupta D, Singh M, Mahapatra AK. Comparative analysis of diagnostic accuracy of different brain biopsy procedures. Neurol India 2006;54:394-8. 62. Lunsford LD, Martinez AJ. Stereotactic exploration of the brain in the era of computed tomography. Surg Neurol 1984;22:222-30. 63. Kulkarni AV, Guha A, Lozano A, Bernstein M. Incidence of silent hemorrhage and delayed deterioration after stereotactic brain biopsy. J Neurosurg 1998;89:31-5. 64. Kelly PJ. Tumor stereotaxis. Saunders; Philadelphia, PA: 65. Chimowitz MI, Barnett GH, Palmer J. Treatment of intractable arterial hemorrhage during stereotactic brain biopsy with thrombin. Report of three patients. J Neurosurg 1991;74:301-3. 66. Brainard JA, Prayson RA, Barnett GH. Frozen section evaluation of stereotactic brain biopsies: diagnostic yield at the stereotactic target position in 188 cases. Arch Pathol Lab Med 1997;121:481-4. 67. Colbassani HJ, Nishio S, Sweeney KM, Bakay RA, Takei Y. CT-assisted stereotactic brain biopsy: value of intraoperative frozen section diagnosis. J Neurol Neurosurg Psychiatry 1988;51:332-41. 68. Chandrasoma P, Apuzzo MLJ. Stereotactic brain biopsy. New York: Igaku-Shoin; 1989. 69. Blakeley JO, Grossman SA. Management of pineal region tumors. Curr Treat Options Oncol 2006;7:505-16. 70. Apuzzo ML, Chandrasoma PT, Zelman V, Giannotta SL, Weiss MH. Computed tomographic guidance stereotaxis
661
662
42 71.
72. 73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
Stereotactic and image-guided biopsy
in the management of lesions of the third ventricular region. Neurosurgery 1984;15:502-8. Regis J, Bouillot P, Rouby-Volot F, Figarella-Branger D, Dufour H, Peragut JC. Pineal region tumors and the role of stereotactic biopsy: review of the mortality, morbidity, and diagnostic rates in 370 cases. Neurosurgery 1996;39:907-12; discussion 912-4. Villani R, Gaini SM, Tomei G. Follow-up study of brain stem tumors in children. Childs Brain 1975;1:126-35. Boviatsis EJ, Voumvourakis K, Goutas N, Kazdaglis K, Kittas C, Kelekis DA. Stereotactic biopsy of brain stem lesions. Minim Invasive Neurosurg 2001;44:226-9. Rajshekhar V, Chandy MJ. Computerized tomographyguided stereotactic surgery for brainstem masses: a riskbenefit analysis in 71 patients. J Neurosurg 1995;82: 976-81. Mathisen JR, Giunta F, Marini G, Backlund EO. Transcerebellar biopsy in the posterior fossa: 12 years experience. Surg Neurol 1987;28:100-4. Valdes-Gorcia J, Espinoza-Diaz DM, Paredes-Diaz E. Stereotactic biopsy of brain stem and posterior fossa lesions in children. Acta Neurochir (Wien) 1998;140:899-903. Boviatsis EJ, Kouyialis AT, Stranjalis G, Korfias S, Sakas DE. CT-guided stereotactic biopsies of brain stem lesions: personal experience and literature review. Neurol Sci 2003;24:97-102. Massager N, David P, Goldman S, Pirotte B, Wikler D, Salmon I, Nagy N, Brotchi J, Levivier M. Combined magnetic resonance imaging- and positron emission tomography-guided stereotactic biopsy in brainstem mass lesions: diagnostic yield in a series of 30 patients. J Neurosurg 2000;93:951-7. Patil AA, Leibrock LG, Kumar PP, Aarabi B. Stereotactic approach to skull-base lesions. Skull Base Surg 1991;1: 235-9. Rosenfeld JV, Wallace D, Klug GL, Danks A. Transnasal stereotactic biopsy of a clivus tumor. Technical note. J Neurosurg 1992;76:878-9. Patil AA. Transoral stereotactic biopsy of the second cervical vertebral body: case report with technical note. Neurosurgery 1989;25:999-1001; discussion 1001-2. Roujeau T, Machado G, Garnett MR, Miquel C, Puget S, Geoerger B, Grill J, Boddaert N, Di Rocco F, Zerah M, Sainte-Rose C. Stereotactic biopsy of diffuse pontine lesions in children. J Neurosurg 2007;107:1-4. Pincus DW, Richter EO, Yachnis AT, Bennett J, Bhatti MT, Smith A. Brainstem stereotactic biopsy sampling in children. J Neurosurg 2006;104:108-14.
84. Chappell ET, Guthrie BL, Orenstein J. The role of stereotactic biopsy in the management of HIV-related focal brain lesions. Neurosurgery 1992;30:825-9. 85. Luzzati R, Ferrari S, Nicolato A, Piovan E, Malena M, Merighi M, Morbin M, Gerosa M, Rizzuto N, Concia E. Stereotactic brain biopsy in human immunodeficiency virus-infected patients. Arch Intern Med 1996;156:565-8. 86. Skolasky RL, Dal Pan GJ, Olivi A, Lenz FA, Abrams RA, McArthur JC. HIV-associated primary CNS lymorbidity and utility of brain biopsy. J Neurol Sci 1999;163:32-8. 87. Kaakaji W, Barnett GH, Bernhard D, Warbel A, Valaitis K, Stamp S. Clinical and economic consequences of early discharge of patients following supratentorial stereotactic brain biopsy. J Neurosurg 2001;94:892-8. 88. Chakeres D, Slone W, Christoforidis G, Bourekas E. Real-time CT-guided spinal biopsy with a disposable stereotactic device: a technical note. AJNR Am J Neuroradiol 2002;23:605-8. 89. Brucher JM. Neuropathological diagnosis with stereotactic biopsies. Possibilities, difficulties and requirements. Acta Neurochir (Wien) 1993;124:37-9. 90. Whiting DM, Barnett GH, Estes ML, Sila CA, Rudick RA, Hassenbusch SJ, Lanzieri CF. Stereotactic biopsy of non-neoplastic lesions in adults. Cleve Clin J Med 1992;59:48-55. 91. Bhatia R, Tandon P, Misra NK. Inflammatory lesions of the basal ganglia and thalamus: review of twenty-one cases. Neurosurgery 1986;19:983-8. 92. Waltregny A, Maula AA, Brucher JM. Contribution of stereotactic brain biopsies to the diagnosis of presenile dementia. Stereotact Funct Neurosurg 1990;54‐55:409-12. 93. Truwit CL, Hall WA. Intraoperative magnetic resonance imaging-guided neurosurgery at 3-T. Neurosurgery 2006;58:ONS-338–45; discussion ONS-345–36. 94. Hall WA, Liu H, Truwit CL. Navigus trajectory guide. Neurosurgery 2000;46:502-4. 95. Reisner LA, King BW, Klein MD, Auner GW, Pandya AK. A prototype biosensor-integrated image-guided surgery system. Int J Med Robot 2007;3:82-8. 96. Scheld M, Whitely RJ, Marra CM. Infections of the Central Nervous System 3rd edition. Published by Lippincott Williams & Wilkins; 2004.
51 Stereotactic Approaches to the Brain Stem L. U. Zrinzo . D. G. T. Thomas
Background The stereotactic concept was first developed by Horsley and Clarke in 1905 to allow precise surgical navigation within the monkey cranium [1]. Applying these principles to clinical practice in 1947, Spiegel and Wycis ushered in a new era where surgical access to deep seated and highly eloquent brain areas could be achieved with minimal morbidity and mortality [2]. These stereotactic techniques were swiftly applied to pathologies and procedures involving the brain stem [3]. In the decades that followed, the introduction of cross-sectional imaging led to the first imagedirected stereotactic brain stem procedures [4,5]. With judicious application, stereotactic brainstem approaches remain a powerful clinical tool. Biopsy of brainstem masses provide a safe and reliable method of obtaining a histological diagnosis, aspiration of cysts, blood clots and abscesses provide therapeutic relief, and brainstem targets are increasingly being considered in functional procedures for pain control and movement disorders.
Role in the Management of Brainstem Lesions Providing a Histological Diagnosis Dorsally exophytic brainstem lesions are generally subjected to debulking by an open procedure [6]. However, stereotactic approaches should be considered for ventral midline exophytic lesions,
#
Springer-Verlag Berlin/Heidelberg 2009
focal lesions within the substance of the brainstem and diffuse intrinsic lesions (> Figure 51-1). Image directed stereotactic brainstem biopsy was first reported by Gleason et al. in 1978 [5]. Since this first report, some authors have documented an above average complication rate from brainstem biopsy when compared to that from supratentorial sites [7–9]. However, proponents of this technique refute this and claim that stereotactic biopsy is safe and provides a histological diagnosis on which to base treatment protocols and provide prognostic information [9–11]. Indeed, a recent meta-analysis of stereotactic biopsy in 293 consecutive brainstem masses revealed one mortality (0.3%), 1% permanent and 4% temporary morbidity (including hemiplegia, cranial neuropathies, diplopia and lethargy) with a 94% diagnostic rate on first attempt rising to 96% with repeated sampling [12]. This high diagnostic yield and low morbidity rival those at other sites within the brain [10,13–17]. CT and MRI directed stereotactic targeting are both used in brainstem biopsies. However, MRI provides better anatomical definition and may allow subtle advantages in distinguishing areas of contrast enhancement that may minimize sampling errors. Initial concerns of image distortion affecting MRI targeting accuracy were initially resolved by using CT/MRI fusion protocols. However, this issue can now be addressed with strict attention to MRI quality control and correction of MR distortion [18]. The concerns over sampling error and possible erroneous diagnosis have led some authors to combined MRI and PET-guided
790
51
Stereotactic approaches to the brain stem
. Figure 51-1 Coronal T1 with contrast and axial T2 MR images of brainstem lesion. Biopsy revealed low grade glioma
stereotactic biopsy to improve the diagnostic yield of the representative sample [19,20]. After the introduction of MRI, it was suggested that the higher resolution afforded by this modality could reliably establish the diagnosis of high grade glioma when a diffuse pontine lesion was visualized, particularly in the pediatric age group [21–24]. This led a number of authors to recommend that such patients should receive radiotherapy empirically, without a tissue diagnosis and that stereotactic biopsy should be reserved for patients with focal brainstem lesions. However, a recent study has shown that sensitivity of MR imaging for diagnosis of pediatric brainstem tumors was high (0.94) but specificity was low (0.43) [25]. Another study reports on 24 children with a suspected diagnosis of high grade glioma on MR imaging; the diagnosis was overturned in two by stereotactic biopsy with a significant alteration in the management plan [26]. In summary, in the absence of a tissue diagnosis, patients with ‘‘false positive’’ MRI findings would receive inappropriate therapy directed against high grade lesions and appropriate treatment would be omitted. In many pediatric centers current practice is to treat diffuse pontine lesions
empirically as presumed high grade gliomas. However, tissue diagnosis is likely to play an increasingly important confirmatory role in the management of all pediatric brainstem tumors as new treatment protocols are developed [15]. It has been amply shown that in the adult population, histological diagnosis does not always agree with the preoperative assessment based on clinical presentation and radiological findings. Unexpected findings can occur in over 15% of cases [10,17]. Even groups supplementing diagnostic imaging with PET find discordance between the predictive value of imaging in the detection of malignancy and histological diagnosis and document a concordance of 63% for MRI and 79% for combined MRI and PET [20]. The literature contains a long list of brainstem pathologies where the correct diagnosis was confirmed only after stereotactic sampling. These include: low grade glioma, metastasis, lymphoma, ependymoma, gangliocytoma, pineoblastoma, epidermoid cyst, epidermal cyst, gangliosidoses, angioma, granuloma, vasculitis, leucoencephalopathy, radionecrosis, demyelination, encephalitis, amoebiasis, tuberculoma and pyogenic abscess [10,16,17,20,27–32]. In a number of these reports
Stereotactic approaches to the brain stem
51
. Figure 51-2 Coronal FLAIR and axial T1 with contrast MR images of brainstem lesion. Biopsy excluded tumor and showed granulomatous inflammation
high grade glioma was the top radiological differential and biopsy made a significant impact on prognosis and patient management. Given the small but not insignificant risks of stereotactic biopsy versus the risk of an incorrect diagnosis and management without biopsy, it is perhaps not surprising that, even in the adult literature, there is debate as to the best course of action when presented with a brainstem lesion. Samadani et al. propose a decision analysis method based upon the probability of an incorrect radiological diagnosis leading to suboptimal empirical treatment, weighed against the probability of stereotactic biopsy leading to a suboptimal outcome (defined as a surgical complication, non-diagnostic sample or sampling error) [13]. The complex interplay of probabilities ensures that one cannot be dogmatic about the role of stereotactic biopsy in brainstem. However, the wide variety of adult brainstem pathologies, the relative rarity of primary gliomas in this age group and our personal experience of a low complication rate, leads the present authors to advise stereotactic biopsy of brainstem lesions unless there are surgical contraindications (> Figure 51-2).
Tumor Cyst Aspiration Aspiration of cysts associated with brainstem tumors can temporarily alleviate neurological symptoms [16,33–35]. Placement of an Ommaya reservoir may allow repeated aspiration of recurring cysts without further surgical intervention. Additionally, it allows for instillation of intracavitary instillation of radioisotopes should repeated aspirations fail to achieve cyst control [35].
Brachytherapy Interstitial brachytherapy of brainstem lesions has been reported in the literature and is held by some authors to be an alternative or an adjunct to external beam radiotherapy [36–39]. Stereotactic interstitial therapy can be performed will minimal surgical complications and it is thought that I-125 low-dose rate implants may limit the risk of radiation-induced necrosis. Although radiological response to this therapy has been documented, the role of interstitial brachytherapy for these lesions remains undefined and its use in the neurosurgical community remains limited [38].
791
792
51
Stereotactic approaches to the brain stem
Stereotactic Aspiration of Brainstem Abscess Pyogenic abscess of the brainstem is rare and prior to 1974 was invariably fatal. The pons is the most frequently affected region. Haematogenous and direct spread from adjacent structures are the commonest causes. However, in over a third of cases no source is identified [40–42]. Stereotactic aspiration allows confirmation of the diagnosis since this is not always obvious by means of CT or MR imaging alone [28]. Drainage of pus, reduction in mass effect and the possible identification of the offending organism together with its antibiotic sensitivity, are additional advantages and may provide an informed choice of antimicrobial agents. Indeed, there is a suggestion that stereotactic aspiration of brainstem abscess in combination with antimicrobial treatment may be associated with a superior functional outcome that medical treatment with or without open surgical drainage [28]. Both transfrontal and suboccipital transcerebellar approaches have been described in treating brain abscesses [28,43–49]. Avoiding ventricular penetration when draining an abscess assumes even greater importance as this may not only lower the risk of hemorrhage but may also be important in avoiding ventriculitis.
Role in Functional Neurosurgery Brainstem structures have been targeted in the management of chronic pain with procedures including stereotactic mesencephalic tractotomy and deep brain stimulation of the periaqueductal gray [50–52]. The brainstem is also a potential target in the treatment of movement disorders with the pedunculopontine nucleus (PPN) holding promise as a potential target for deep brain stimulation (DBS) in parkinsonian patients with gait disturbance and postural instability refractory to other treatment modalities [53–55]. This elongated
neuronal collection in the lateral pontine and mesencephalic tegmental reticular zones is unfamiliar territory to most functional neurosurgeons [56]. MRI guided stereotactic targeting is an important technique in identification and targeting of appropriate anatomical structures in Functional Neurosurgery [57–59]. MRI protocols that allow localization of the PPN will be relevant to groups evaluating the clinical role of PPN DBS (> Figure 51-3) [60].
Technical Considerations Surgery can be performed under local or general anesthesia. Numerous frames have been used in stereotactic procedures on the brainstem including the Leksell, Riechert and BRW/CRW frames. CT and MRI are the main modalities used in image directed targeting with PET also being used in an attempt to improve sampling accuracy of diagnostic biopsies. In general MRI provides superior resolution in the posterior fossa and, when possible, is the preferred imaging modality for the authors. Modern MRI equipment, strict quality assurance and the use of frames with fiducials close to the skull (e.g., Leksell frame) reduce the effect of geometric distortion [62]. Three-dimensional planning software now allows an accurate assessment of both target and trajectory prior to surgery. Transgression of multiple pial or ependymal surfaces places the associated vascular structures at risk. Therefore, a trajectory confined to brain parenchyma would, theoretically, reduce the risk of hemorrhage. CSF loss from a penetrated ventricular system may result in increased brain shift and our own research has suggested that targeting accuracy may be reduced with trajectories that pass close to or through the ventricle walls (as yet unpublished results). A blunt round-tipped probe advanced slowly to the target without precession allows minimal disruption of brain tissues as they are displaced around the probe rather than being transected.
Stereotactic approaches to the brain stem
51
. Figure 51-3 Axial MRI images through the inferior colliculi reformatted in a plane perpendicular to the midline of the fourth ventricular floor. The T1 image on the left shows excellent contrast between brain and CSF but little internal tissue contrast. The middle image is a Proton Density MRI where gray matter appears hyperintense (bright); white matter appears hypointense (dark). Identifiable anatomical structures are labeled on the right. CTT, central tegmental tract; DSCP, decussation of the superior cerebellar peduncles; LL, lateral lemniscus; ML, medial lemniscus; PAG, periaqueductal gray; PPN, pedunculopontine nucleus; PT, pyramidal tract; SN, substantia nigra; STT, spinothalamic tract. Reproduce with kind permission from Oxford University Press [61]
An ipsilateral transfrontal entry point provides access to the mesencephalon and midline regions of the pons [17]. A contralateral transfrontal entry point has also been described that allows access to more laterally placed pontine lesions without having to traverse the ventricular system [63]. Both approaches allow the patient to remain supine during surgery, in a similar position to that in which images are traditionally acquired thus preventing error due to positional brain shift. In this region, a burrhole can be placed without painful muscle dissection and twist drill holes can be planned to avoid sulci. The transtentorial route has been virtually abandoned and is not employed by the authors because of the increased risk of hemorrhage and trajectory deviation by traversing the tentorium and several other anatomical surfaces. The suboccipital transcerebellar approach is often used to access brainstem lesions [4,26,27,64–67]. Care must be taken to ensure that the frame is placed low enough to allow the lesion to be visualized and to physically allow the required trajectory with a particular frame [66]. Semi-recumbent, lateral and prone positions have been described to provide access, some of which may limit the possibility of awake
surgery. This approach provides the shortest distance to the desired target [66,67]. Twist drill holes provide the convenience of minimal tissue disruption and at this site minimize pain associated with more excessive muscle stripping. However, one is then committed to that particular trajectory and access for haemostatic control of the dura and pia is restricted. With supratentorial entry points, planning software can minimize this risk by avoiding sulci. However, such planning over the folia of the cerebellar surface may not be possible. Nonetheless, there is no evidence that the risk of hemorrhage depends upon whether entry is via a burrhole or a twist drill hole [8].
Concluding Remarks Stereotactic procedures provide surgical access to deep seated eloquent brain areas with minimum morbidity and mortality and are thus ideally suited to surgery of the brainstem. Judicious clinical application and meticulous surgical planning and technique will exploit the benefits and minimize the risks of this powerful surgical tool.
793
794
51
Stereotactic approaches to the brain stem
References 1. Horsley V, Clarke RH. The structure and functions of the cerebellum examined by a new method. Brain 1905;28 (1):1-98. 2. Spiegel EA, et al. Stereotaxic apparatus for operations on the human brain. Science 1947;106(2754):349-50. 3. Ward AA, Jr. Trends in the application of stereotaxy to the brain stem. Clin Neurosurg 1958;6:223-39. 4. Mathisen JR, et al. Transcerebellar biopsy in the posterior fossa: 12 years experience. Surg Neurol 1987;28(2):100-4. 5. Gleason CA, Wise BL, Feinstein B. Stereotactic localization (with computerized tomographic scanning), biopsy, and radiofrequency treatment of deep brain lesions. Neurosurgery 1978;2(3):217-22. 6. Epstein F, Wisoff JH. Surgical management of brain stem tumors of childhood and adolescence. Neurosurg Clin N Am 1990;1(1):111-21. 7. Frank F, et al. Stereotactic biopsy and treatment of brain stem lesions: combined study of 33 cases (BolognaMarseille). Acta Neurochir Suppl (Wien) 1988;42: 177-81. 8. Grossman R, et al. Haemorrhagic complications and the incidence of asymptomatic bleeding associated with stereotactic brain biopsies. Acta Neurochir (Wien) 2005;147 (6):627-31; discussion 631. 9. Kondziolka D, Lunsford LD. Results and expectations with image-integrated brainstem stereotactic biopsy. Surg Neurol 1995;43(6):558-62. 10. Kratimenos GP, Thomas DG. The role of image-directed biopsy in the diagnosis and management of brainstem lesions. Br J Neurosurg 1993;7(2):155-64. 11. Thomas DG, et al. Computer-directed stereotactic biopsy of intrinsic brain stem lesions. Br J Neurosurg 1988; 2(2):235-40. 12. Samadani U, Judy KD. Stereotactic brainstem biopsy is indicated for the diagnosis of a vast array of brainstem pathology. Stereotact Funct Neurosurg 2003;81(1–4):5-9. 13. Samadani U, et al. Stereotactic biopsy of brain stem masses: decision analysis and literature review. Surg Neurol 2006;66(5):484-90; discussion 491. 14. Chico-Ponce de Leon F, et al. Stereotactically-guided biopsies of brainstem tumors. Childs Nerv Syst 2003;19 (5–6):305-10. 15. Pincus DW, et al. Brainstem stereotactic biopsy sampling in children. J Neurosurg 2006;104(2 Suppl):108-14. 16. Rajshekhar V, Chandy MJ. Computerized tomographyguided stereotactic surgery for brainstem masses: a riskbenefit analysis in 71 patients. J Neurosurg 1995;82 (6):976-81. 17. Kratimenos GP, et al. Image directed stereotactic surgery for brain stem lesions. Acta Neurochir (Wien) 1992;116 (2–4):164-70.
18. Menuel C, et al. Characterization and correction of distortions in stereotactic magnetic resonance imaging for bilateral subthalamic stimulation in Parkinson disease. J Neurosurg 2005;103(2):256-66. 19. Aker FV, et al. Accuracy and diagnostic yield of stereotactic biopsy in the diagnosis of brain masses: comparison of results of biopsy and resected surgical specimens. Neuropathology 2005;25(3):207-13. 20. Massager N, et al. Combined magnetic resonance imaging- and positron emission tomography-guided stereotactic biopsy in brainstem mass lesions: diagnostic yield in a series of 30 patients. J Neurosurg 2000;93(6):951-7. 21. Albright AL, et al. Magnetic resonance scans should replace biopsies for the diagnosis of diffuse brain stem gliomas: a report from the Children’s Cancer Group. Neurosurgery 1993;33(6):1026-9; discussion 1029–30. 22. Albright AL. Tumors of the pons. Neurosurg Clin N Am 1993;4(3):529-36. 23. Epstein F, McCleary EL. Intrinsic brain-stem tumors of childhood: surgical indications. J Neurosurg 1986;64 (1):11-15. 24. Cartmill M, Punt J. Diffuse brain stem glioma. A review of stereotactic biopsies. Childs Nerv Syst 1999;15 (5):235-7; discussion 238. 25. Schumacher M, et al. Magnetic resonance imaging compared with biopsy in the diagnosis of brainstem diseases of childhood: a multicenter review. J Neurosurg 2007; 106(2 Suppl):111-19. 26. Roujeau T, et al. Stereotactic biopsy of diffuse pontine lesions in children. J Neurosurg 2007;107(1 Suppl):1-4. 27. Coffey RJ, Lunsford LD. Stereotactic surgery for mass lesions of the midbrain and pons. Neurosurgery 1985;17 (1):12-18. 28. Bavetta S, et al. Brainstem abscess: preoperative MRI appearance and survival following stereotactic aspiration. J Neurosurg Sci 1996;40(2):139-43. 29. Nassogne MC, et al. Unusual presentation of GM2 gangliosidosis mimicking a brain stem tumor in a 3-year-old girl. AJNR Am J Neuroradiol 2003;24(5):840-2. 30. Rajshekhar V, Chandy MJ. Tuberculomas presenting as isolated intrinsic brain stem masses. Br J Neurosurg 1997;11(2):127-33. 31. Valdes-Gorcia J, Espinoza-Diaz DM, Paredes-Diaz E. Stereotactic biopsy of brain stem and posterior fossa lesions in children. Acta Neurochir (Wien) 1998;140 (9):899-903. 32. Lowichik A, et al. Leptomyxid amebic meningoencephalitis mimicking brain stem glioma. AJNR Am J Neuroradiol 1995;16(4 Suppl):926-9. 33. Akhtar N, Lewis TT. Management of midbrain cyst with repeated CT guided aspiration. Neuroradiology 1994;36 (8):642-3. 34. Hood TW, McKeever PE. Stereotactic management of cystic gliomas of the brain stem. Neurosurgery 1989;24 (3):373-8.
Stereotactic approaches to the brain stem
35. Giovanini MA, Mickle JP. Long-term access to cystic brain stem lesions using the Ommaya reservoir: technical case report. Neurosurgery 1996;39(2):404-7. 36. Mundinger F, et al. Long-term outcome of 89 low-grade brain-stem gliomas after interstitial radiation therapy. J Neurosurg 1991;75(5):740-6. 37. Matsumoto K, et al. Stereotactic brachytherapy for a cystic metastatic brain tumor in the midbrain. Case report. J Neurosurg 1998;88(1):141-4. 38. Chuba PJ, et al. Permanent I-125 brain stem implants in children. Childs Nerv Syst 1998;14(10):570-7. 39. Julow J, et al. Iodine-125 brachytherapy of brain stem tumors. Strahlenther Onkol 2004;180(7):449-54. 40. Hall WA. Infectious lesions of the brain stem. Neurosurg Clin N Am 1993;4(3):543-51. 41. VanGilder JC, Allen WE, III, Lesser RA. Pontine abscess: survival following surgical drainage. Case report. J Neurosurg 1974;40(3):386-90. 42. Jamjoom ZA. Solitary brainstem abscess successfully treated by microsurgical aspiration. Br J Neurosurg 1992; 6(3):249-53. 43. Nakajima H, et al. Successful treatment of brainstem abscess with stereotactic aspiration. Surg Neurol 1999;52 (5):445-8. 44. Nauta HJ, et al. Brain stem abscess managed with computed tomography-guided stereotactic aspiration. Neurosurgery 1987;20(3):476-80. 45. Fuentes S, et al. Management of brain stem abscess. Br J Neurosurg 2001;15(1):57-62. 46. Fujino H, et al. Cure of a man with solitary abscess of the brain-stem. J Neurol 1990;237(4):265-6. 47. Kalarostaghi AH, et al. Polymicrobial brain stem abscess due to Streptococcus anginosus and Actinomyces species. J Clin Neurosci 1999;6(5):415-18. 48. Rajshekhar V, Chandy MJ. Successful stereotactic management of a large cardiogenic brain stem abscess. Neurosurgery 1994;34(2):368-71; discussion 371. 49. Rossitch E, Jr, et al. The use of computed tomographyguided stereotactic techniques in the treatment of brain stem abscesses. Clin Neurol Neurosurg 1988;90 (4):365-8. 50. Amano K, et al. Stereotactic mesencephalotomy for pain relief. A plea for stereotactic surgery. Stereotact Funct Neurosurg 1992;59(1–4):25-32. 51. Fountas KN, et al. MR-based stereotactic mesencephalic tractotomy. Stereotact Funct Neurosurg 2004; 82(5–6):230-4. 52. Young RF, Chambi VI. Pain relief by electrical stimulation of the periaqueductal and periventricular gray matter. Evidence for a non-opioid mechanism. J Neurosurg 1987;66(3):364-71. 53. Stefani A, et al. Bilateral deep brain stimulation of the pedunculopontine and subthalamic nuclei in severe Parkinson’s disease. Brain 2007;130(Pt 6):1596-607.
51
54. Nandi D, et al. Deep brain stimulation of the pedunculopontine region in the normal non-human primate. J Clin Neurosci 2002;9(2):170-4. 55. Plaha P, Gill SS. Bilateral deep brain stimulation of the pedunculopontine nucleus for Parkinson’s disease. Neuroreport 2005;16(17):1883-7. 56. Zrinzo L, Zrinzo LV, Hariz M. The pedunculopontine and peripeduncular nuclei: a tale of two structures. Brain 2007;130(Pt 6):e73; author reply e74. 57. Hirabayashi H, Tengvar M, Hariz MI. Stereotactic imaging of the pallidal target. Mov Disord 2002;17 Suppl 3: S130-4. 58. Hariz MI, et al. A quick and universal method for stereotactic visualization of the subthalamic nucleus before and after implantation of deep brain stimulation electrodes. Stereotact Funct Neurosurg 2003;80 (1–4):96-101. 59. Vayssiere N, et al. Comparison of atlas- and magnetic resonance imaging-based stereotactic targeting of the globus pallidus internus in the performance of deep brain stimulation for treatment of dystonia. J Neurosurg 2002;96(4):673-9. 60. Muthusamy KA, et al. Connectivity of the human pedunculopontine nucleus region and diffusion tensor imaging in surgical targeting. J Neurosurg 2007;107 (4):814-20. 61. Zrinzo L, et al. Stereotactic localization of the human pedunculopontine nucleus: atlas-based coordinates and validation of a magnetic resonance imaging protocol for direct localization. Brain 2008;131:1588-98. 62. Yu C, et al. A phantom study of the geometric accuracy of computed tomographic and magnetic resonance imaging stereotactic localization with the Leksell stereotactic system. Neurosurgery 2001;48(5):1092-8; discussion 1098–9. 63. Amundson EW, McGirt MJ, Olivi A. A contralateral, transfrontal, extraventricular approach to stereotactic brainstem biopsy procedures. Technical note. J Neurosurg 2005;102(3):565-70. 64. Abernathey CD, Camacho A, Kelly PJ. Stereotaxic suboccipital transcerebellar biopsy of pontine mass lesions. J Neurosurg 1989;70(2):195-200. 65. Guthrie BL, Steinberg GK, Adler JR. Posterior fossa stereotaxic biopsy using the Brown-Roberts-Wells stereotaxic system. Technical note. J Neurosurg 1989;70 (4):649-52. 66. Neal JH, Van Norman AS. Transcerebellar biopsy of posterior fossa lesions using the Leksell gamma model stereotactic frame. Neurosurgery 1993;32(3):473-4; discussion 474–5. 67. Spiegelmann R, Friedman WA. Stereotactic suboccipital transcerebellar biopsy under local anesthesia using the Cosman-Roberts-Wells frame. Technical note. J Neurosurg 1991;75(3):486-8.
795
53 Technical Aspects of Image-Guided Neuroendoscopy J. D. Caird . J. M. Drake
Neuroendoscopy has become an established neurosurgical technique permitting neurosurgeons to address deep-seated intracranial pathology under direct vision. The advent of image-guided surgery has enhanced the potential of neuroendoscopy beyond the boundary of direct vision, thus minimizing trauma to the brain. Image-guided neuroendoscopy does not replace frame-based stereotactic surgery where access to deep parenchymal structures for the purposes of biopsy or electrode placement are concerned. The experience of image-guided neuroendoscopy at our institution has been intracranial using a rigid, zero-degree Aesculap neuroendoscope in combination with Brainlab Neuronavigation.
Since Horsley and Clarke reported their use of a stereotactic frame device to study the cerebellum in monkeys in 1906 there have been innumerable advances in stereotactic surgery. The earliest frame-based devices utilized boney reference points; with the advent of air ventriculography and cerebral angiography, image guided stereotaxy using intracranial anatomic reference points was feasible. Amongst the many contributors to the field, Leksell is notable for linking stereotactic surgery with modern imaging modalities [3,4]. Frameless stereotaxy developed as a concept during the 1980s but has only become widespread in neurosurgical practice since the mid 1990s. Its application to neuroendoscopy is very recent and continues to progress [5–7].
Background Dandy pioneered the field of neuroendoscopy in the early 1920s and coined the term ‘‘ventriculoscope’’ in a brief article in which he described the use of an endoscope to inspect the lateral ventricles of two children with hydrocephalus [1]. Limitations at that time with regard to compatible endoscopic instrumentation restricted the potential use of the endoscope as a working surgical instrument. Present-day neuroendoscopy has been made possible through successful attempts at miniaturization of instrumentation. Fukushima is credited with pioneering the technique of neurendoscopic biopsy of intraventricular tumor in 1978 [2]. Since that time, many refinements in instrumentation have taken place in an effort to optimize the working neuroendoscope. #
Springer-Verlag Berlin/Heidelberg 2009
Surgical Technique of ImageGuided Neuroendoscopy Preoperative volumetric MRI or CT sequences are a pre-requisite for image-guided surgery with CT being useful to define sphenoid bone and air sinus anatomy. Whilst we have had some experience with transphenoidal neuroendoscopic surgery, the majority of our cases are concerned with the management of hydrocephalus, intraventricular tumors and cysts. The surgical approach for neuroendoscopy is clearly dictated by the location of the intraventricular pathology. A frontal entry point for ventricular access placed slightly anterior to the coronal suture and approximately 3 cm lateral to the midline is considered standard practice. Where two
808
53
Technical aspects of image-guided neuroendoscopy
or more procedures such as endoscopic third ventriculostomy (ETV), septostomy, tumor biopsy, or shunt insertion are considered, two or more trajectories may be required; we have found it practical in this setting to fashion a small craniotomy through a curvilinear skin incision, with two separate dural incisions to accommodate the endoscope for each approach. Simultaneous use of two trajectories using both rigid and steerable neuroendoscopes has been described in the ‘‘multi-axial’’ approach to the pineal region and floor of the third ventricle [8]. As with shunt surgery, neuroendoscopy demands meticulous attention to detail, and should be carried out in a skilled and expeditious fashion. Body wash and shampoo the night before and again before surgery with an antiseptic solution such as Chlorhexidine has been the practice at our institution, but is not mandatory. In the operating room the patient is positioned under general endotracheal anesthesia. The head is generally positioned in the neutral position in a pinned head clamp with some neck flexion or table tilt ‘‘head up’’ to prevent excessive CSF losses unless shunt insertion during the same procedure is anticipated, in which case the head needs to be rotated to the appropriate side. For babies and children with a patent fontanelle or fragile cranium, including those with osteogenesis imperfecta, a pinned head clamp may be undesirable in which case pre- and intraoperative ultrasound may well suffice. Prophylactic antibiotic administration is highly recommended; we use 30 mg kg 1 of cefazolin at induction. The image-guidance reference arc is applied to the head clamp and surface registration obtained with either the pointer wand or with laser registration. Using the pointer and off-set application, the surgical trajectories required may be planned. The midline and coronal suture landmarks, and skin incisions are marked on the patient, bearing in mind that a shunt valve should not be directly under a wound if possible. The hair is then clipped, not shaved,
to assist with wound closure and dressing application. The eyes are taped closed after registration and the skin is meticulously prepped with an iodine or chlorhexidine solution. Disposable, adhesive drapes are used to cover the patient and the operating table entirely except for a small area of skin required for incision. An iodineimpregnated transparent adhesive is applied to the prepped skin. At this juncture, it is prudent to establish that the irrigation solution, either Ringer’s lactate or normal saline, is warmed to 37 C and that the endoscope and light-source are fully operational. The video camera is passed through a sterile plastic sheath and attached to the endoscope. With the light-source connected, white-balancing can be performed. It is important to check that the scope is correctly orientated and focused, for which purpose a suture packet can be used. It is useful at this point to ensure that the endoscopic instruments (monopolar/bipolar diathermy, biopsy forceps, alligator forceps, balloon catheters) are functional prior to ventricular access. Both the endoscope and the image-guidance screens should be positioned for easy viewing to avoid having to turn one’s head. For navigation purposes, a reference arc is clamped to the endoscope side irrigation port (to prevent crimping of the working channel) and registered with the image-guidance device (> Figure 53-1). The skin incision and burrhole or minicraniotomy are fashioned and the dura coagulated at the optimal entry point as determined with image-guidance. A cruciform durotomy sufficient to permit passage of the endoscope outer casing, generally 6 mm in diameter, is made and the pia and cortex are coagulated and incised with a size 15 or 11 scalpel blade. Once the ventricle has been entered, the obturator can be withdrawn and the endoscope introduced by the surgeon, with the operative assistant holding the neuroendoscope motionless. It is important to recheck the orientation
Technical aspects of image-guided neuroendoscopy
. Figure 53-1 Instrument calibration device with Aesculap endoscope barrel at registration. Note reference arc clamped to endoscope irrigation port to prevent crimping of working channel
of the scope at this stage as it may have altered in the setting up process. Prior to entering the ventricle, the anticipated depth of cortex to be traversed can be measured with the navigation ‘‘off-set’’ function and the endoscope trajectory can be followed with ultrasound (> Figure 53-2). A sensation of slight resistance is usually encountered at the ependyma followed by a slight ‘‘give.’’ Once the endoscope tip has been confirmed visually to be intraventricular, the irrigation entry port should be opened, ensuring the irrigation egress port is patent to avoid a net increase in CSF volume. The operative assistant can now ‘‘drive’’ the scope within the ventricle, the first step being to identify a constant landmark such as the foramen of Monro by following choroid plexus antero-medially and to reference the foramen with the image-guidance. The surgeon is now in a position to perform the desired image-guided endoscopic procedure; the use of a rigid endoscope holding device permits the operative assistant to partake in the endoscopic instrumentation. Over-head monitors displaying the endoscopic image and image-guidance sideby-side are strongly advocated to avoid having to turn one’s head to see different screens at this delicate stage (> Figure 53-3 and > Figure 53-4).
53
. Figure 53-2 Trajectory confirmed after drilling of burr-hole. Depth of cortical mantle to be traversed can be calculated with ‘‘off-set’’ mode and confirmed with ultrasound
Where ETV is contemplated, it should be done prior to any further procedure such as tumor biopsy to avoid bloody CSF obscuring the view of the floor of the third ventricle. It has been our practice when performing ETV, to very carefully perforate the floor of the third ventricle using the metal stylet from a 35 cm ventriculostomy catheter. Once perforated, the stoma can be widened by passing a closed alligator forceps into the stoma and gently opening it, or by passing a Fogarty balloon catheter and slowly dilating the balloon within the stoma. Where tumor biopsy is concerned, consideration must be given as to whether the lesion in truly intraventricular or subependymal; the latter may be occult on ventriculoscopy and it is here that image-guidance can be indispensable. It is preferable to avoid diathermy at the biopsy site which may cause thermal artifact in the pathology specimen. Furthermore, where a choroid plexus tumor is suspected on preoperative imaging, craniotomy may be more prudent to avoid uncontrollable intraventricular hemorrhage. On completion of the neuroendoscopy, the scope is carefully withdrawn, irrigating
809
810
53
Technical aspects of image-guided neuroendoscopy
. Figure 53-3 Single split screen showing multi-axial neuronavigation and endoscope view for septostomy
throughout, until free of the cortex. The corticotomy site may be plugged with a piece of hemostatic foam to prevent further CSF loss or oozing from the cortex. In doing so, one must be mindful that such material may migrate and has been known to cause failure of an ETV by plugging the stoma [9]. If sufficient intraventricular hemorrhage has occurred, the scope tract and ventricle may be examined with ultrasound prior to wound closure. If, following hemostasis, the CSF is bloodstained at the end of the procedure it may be preferable to leave an external ventricular drain in situ. The scalp is closed in two layers, using absorbable sutures for galea and an absorbable monofilament suture for skin.
Applications of Image-Guided Neuroendoscopy Neuroendoscopy has become an effective tool in the treatment of obstructive hydrocephalus,
intracranial cysts, and a variety of intraventricular and periventricular mass lesions [10,11]. The exponential improvement in the technology of frameless and wireless image-guidance systems has all but superceded the requirement for frame-based neuroendoscopy. Even in the case of small ventricles, intraoperative ultrasound permits real-time definition of the ventricular system and image-guidance can facilitate ventricular cannulation [12]. Image-guidance excels in the planning stage of the surgery as once the patient is anaesthetized, the neuroendoscope trajectory can be delineated and the burrhole position marked out on the cranium. This is particularly crucial for cases of intraventricular tumors where the ventricles may be distorted and perhaps both septum pellucidotomy and tumor biopsy are envisaged; optimal burrhole position is frequently more anterior and lateral than one would have anticipated. Image-guided neuroendoscopy has been utilized for a variety of intracranial conditions,
Technical aspects of image-guided neuroendoscopy
53
. Figure 53-4 Ceiling-mounted display allows surgeons to focus on their instrumentation at critical moments without having to turn their heads
including pathology of the lateral, third, and fourth ventricles, transnasal approaches to the sellar and parasellar lesions, arachnoid cysts of the anterior, middle and posterior fossa, and vascular lesions including cavernous malformations and cerebral aneurysms. At our institution, we utilize neuroendoscopy predominantly for endoscopic third ventriculostomy, intraventricular tumor biopsy, cyst fenestration and the management of complex hydrocephalus where septum pellucidotomy and/or shunt placement is indicated. Furthermore, it has proven useful in the conversion of long-term shunted hydrocephalus to ETV where extraction of an old ventricular catheter may be imprudent without direct vision and diathermy. We do not routinely employ image-guidance where ETV alone is considered. The accuracy of neuroendoscopic biopsy of pineal region tumors and intraventricular germinomas has been reported as high as 89 and 100% respectively [13,14]. It has been suggested that the higher yield from germinomas compared
with pineal tumors of various sub-types has been due to the fact that germinomas are generally exophytic within the ventricle, whereas gliomas are frequently subependymal; we postulate that the concomitant use of image-guidance for these cases could improve the histologic yield from subependymal lesions where obscuration of direct vision by bloody CSF following the first biopsy sample does not preclude further image-guided tissue sampling. A retrospective analysis of endoscopic tumor biopsy at our institution suggests an acceptable degree of histologic certainty in diagnosis is achieved in only 69% of cases, with 31% of biopsies yielding problematic or uninterpretable samples [15].
Pitfalls and Complications of Neuroendoscopy Rigid skull fixation during image-guided surgery is preferable but not an absolute necessity; there
811
812
53
Technical aspects of image-guided neuroendoscopy
may be occasions where intraoperative adjustment of head position is desirable such as during shunt placement. Pinned head clamps are undesirable with very young children due to the risk of skull fracture or distortion and epidural hematoma formation; trans-fontanelle or trans-cortical ultrasonography is invaluable in infants and young children undergoing neuroendoscopy. Alternatively, non-pinned head holders are available in these circumstances. Where the patient’s head is placed in a nonneutral position including for septostomy and shunt placement, particular care must be taken to ensure that the neuroendoscope is correctly orientated (‘‘up is up’’) as the usual intraventricular landmarks are not in the neutral anatomical position. Furthermore, attachment of the imageguidance reference arc to the barrel of the neuroendoscope may distort the barrel making passage of instruments through the working channel difficult. At our institution, we have found that the reference arc can be clamped to one of the irrigation ports of the neuroendoscope barrel (Aesculap) without hindering passage of either the camera or the working instruments. Neuroendoscopy incurs a risk of intraventicular hemorrhage, particularly where biopsy of a mass is undertaken. Similarly, the dissemination of tumor cells along the surgical tract is a rare but recognized complication of neuroendoscopic biopsy [16]. Where hemorrhage in the ventricle or from the cortical tract is encountered, copious saline irrigation or the endoscopic bipolar device may provide effective hemostasis. Alternatively, local tamponade with an inflated Fogarty balloon catheter may be employed. Where ETV is anticipated with or without fenestration or biopsy procedure, the likelihood of successful CSF drainage diminishes with younger age, with failure in infants younger than 1 month of age reaching 75%. One and 5 year success rates (i.e., not requiring further CSF diversion) are reported as 65 and 52% respectively.
Complications following ETV including CSF leak, meningitis, hemorrhage, seizure, injury to cerebrum/cranial nerves has collectively been reported at 13.6%, with isolated reports of precipitous decline and death from ETV failure [17]. A mathematical model to determine the likelihood of ETV success has been developed and is being prepared for publication [18].
Ultrasound Directed Neuroendoscopy Intraoperative ultrasound is considered invaluable in our experience and is employed routinely in our neuroendoscopic cases. It allows real-time inspection of the ventricular system and the endoscope tract following withdrawal of the endoscope where bleeding may have been a concern. It also permits reliable placement of a ventricular catheter, even through the stoma of a septum pellucidotomy. It is employed where image-guidance is unavailable or fails, or where rigid head fixation is undesirable. We also use it in conjunction with neuronavigation if it is felt that excess CSF egress may render image registration inaccurate. In infants, ultrasound may be used to image the ventricular system via the anterior fontanelle whereas for older children and adults, its use generally requires a larger than standard 16 mm burrhole. This is generally fashioned with a highspeed drill to allow placement of the burrhole probe flush against the dura.
Future Dimensions and Conclusion Although image-guided neuroendoscopy is currently limited torigidneuroendoscopes,technologic advances will, no doubt, permit image-guided tracking of flexible neuroendoscopes. Intraoperative MRI or CT may permit real-time updating
Technical aspects of image-guided neuroendoscopy
of registration data to allow for brain-shift during surgery; 3-D ultrasound has already been developed with this purpose in mind. Further developments in integrated monitor display units may decrease the volume of equipment required in the operating room and permit the surgeons to concentrate only on one screen displaying both navigation and endoscope views. Image-guided neuronavigation and intraoperative ultrasonography have exponentially expanded the horizon of neuroendoscopic surgery, offering surgeons a greater sense of security in the pre and intraoperative decision making processes in the treatment of complex neurosurgical pathology. Useful company websites for neuroendoscopic, neuronavitgation and ultrasound devices. www.medtronic.com www.brainlab.com www.stryker.com www.codman.com www.aesculapusa.com www.karlstorz.com www.aloka.com
References 1. Dandy WE. An operative procedure for hydrocephalus. Bull Johns Hopkins Hosp. 1922;33:189-90. 2. Fukushima T. Endoscopic biopsy of intraventricular tumors with the use of a ventriculofiberscope. Neurosurg 1978;2:110-3. 3. Clarke RH, Horsley V. On a method of investigating the deep ganglia and tracts of the central nervous system (cerebellum). Br Med J. 1906;1799–1800. 4. Leksell L, Jernberg B. Stereotaxis and tomography. A technical note. Acta Neurochir Wien. 1980;52:1-7.
53
5. Watanabe E, Watanabe T, Manaka S, Mayanaqi Y, Takakura K. Three dimensional digitizer (neuronavigator): new equipment for computed tomographyguided stereotaxic surgery. Surg Neurol. 1987;27:543-7. 6. Golfinos J, Fitzpatrick B, Smith L, Spetzler R. Clinical use of a frameless stereotactic arm: results in 325 cases. J Neurosurg. 1995;83:197-205. 7. Mayberg M, La Presto E, Cunningham E. Image-guided endoscopy: description of technique and potential applications. Neurosurg Focus. 2005;19(1):E10. 8. Oi S, Kamio M, Joki T, Abe T. Neuroendoscopic anatomy and surgery in pineal region tumors. J Neuro-Onc. 2001;54:227-86. 9. Edwards R, Dirks P. Gelfoam obstruction of endoscopic third ventriculostomy. Case illustration. J Neurosurg. 2006;105:154. 10. Abbott R. The endoscopic management of arachnoidal cysts. Neurosurg Clin N Am. 2004;15:9-17. 11. Hellwig D, Grotenhuis J, Tirakotai W, Riegel T, Schulte D, Bauer B, Bertalanffy H. Endoscopic third ventriculostomy for obstructive hydrocephalus. Neurosurg Rev. 2005;28:1-34. 12. Souweidane M. Endoscopic surgery for intraventricular brain tumors in patients without hydrocephalus. Neurosurgery 2005;57:312-8. 13. Yamini B, Refai D, Rubin G, Frim D. Initial endoscopic management of pineal region tumors and associated hydrocephalus: clinical series and literature review. J Neurosurg. 2004;100:437-41. 14. Shono T, Natori Y, Morioka T, et al. Results of long-term follow-up after neuroendoscopic biopsy and third ventriculostomy in patients with intracranial germinomas. J Neurosurg. 2007;107:193-8. 15. Depreitere B, Dasi N, Rutka J, Dirks P, Drake J. Endoscopic biopsy for intraventricular tumors in children. J Neurosurg:Pediatrics 2007;106:340-6. 16. Haw C, Steinbok P. Ventriculoscope tract recurrence after endoscopic biopsy of pineal germinoma. Paediatric Neurosurg. 2001;34:215-7. 17. Drake JM. Endoscopic third ventriculostomy in pediatric patients: the Canadian experience. Neurosurgery 2007;60:881-6. 18. Kulkarni AV. Personal communication. 2007.
813
36 The History, Current Status, and Future of the StealthStation Treatment Guidance System R. Bucholz . L. McDurmont
History of the StealthStation Treatment Guidance System The StealthStation neuronavigation system was specifically developed to facilitate broad adoption of stereotactic techniques across the entire spectrum of neurosurgical procedures. It has evolved into a system that can improve the accuracy and efficacy of a broad spectrum of procedures performed by a variety of surgical specialists. Stereotactic surgery was first developed by Sir Victor Horsley and RH Clarke [1] in 1908 to improve the accuracy of functional procedures as performed in laboratory animal investigations. In 1945, Spiegel and Wycis developed the first human application, and approximately a year later performed the first stereotactic procedure on a patient with Huntington’s chorea [2]. In 1953, Cooper found that small lesions within the basal ganglion could ameliorate the symptoms of Parkinson’s disease [3], and soon thereafter the most frequent application of stereotactic technique was the production of lesions to enhance function of patients with movement disorders, a field called functional stereotactic neurosurgery. With the advent of medical treatment for Parkinson’s disease, the frequency of use of stereotactic technique declined rapidly during the latter half of the twentieth century [4], and in most centers it was employed for biopsy procedures only. Even though the accuracy of stereotaxis was widely accepted, the difficulty of performing stereotactic interventions resulted in the technique #
Springer-Verlag Berlin/Heidelberg 2009
being used only in select procedures in which accuracy was of paramount importance. Therefore, the vast majority of neurosurgical procedures were not performed stereotactically, a situation that persisted even after the advent of threedimensional imaging techniques such as computed tomographic (CT) scanning and, subsequently, magnetic resonance imaging (MRI). These imaging modalities were capable of providing detailed anatomical information to the surgeon intraoperatively, which could be highly useful in the performance of routine cranial procedures. Even though stereotactic procedures could couple this information to such operations, they were not employed due to the difficulty in using such techniques.
Integrating Imaging into the Intra-operative Space The goal of our development team at Saint Louis University was to integrate advances in imaging seamlessly into our neurosurgical technique. For that to occur, two issues had to be addressed. First, the coordinate system in which the threedimensional images were obtained preoperatively had to be registered to a reference system established during surgery, a process termed registration. Second, instrumentation had to be tracked within the surgical space during performance of the procedure. Both of these issues required the appropriate use of computational power. Images obtained from CT scanners were saved as large
544
36
The history, current status, and future of the StealthStation treatment guidance system
data files reflecting the detailed information obtained by even early units. A high degree of computational power was needed to manipulate these files, and the development of a solution to these issues had to wait until computers with sufficient power to perform these tasks became sufficiently inexpensive to be available to our team, and subsequently, the operating surgeon. Further, although computer networks were slowly becoming available, for our effort, and for the vast majority of operating rooms, there was also a requirement that the computers used for the system be sufficiently small to enable the system to be brought into the operating room during the procedure, and then removed to allow other procedures to be performed in the same room. It became evident that, for our effort, the only viable solution consisted of a computer on a cart with sufficient power to manipulate the large threedimensional image data sets, coupled with a three-dimensional digitizer of sufficient accuracy and robustness to be used in a sterile operative field. It became possible to satisfy both of these requirements in the late 1980s, leading to the development of our system and first application of a generalized solution in 1990. In the early to mid 1980s, computers capable of the manipulation and display of diagnostic images were large, requiring multiple equipment racks, and had significant power and cooling requirements. Propelled by the continuing miniaturization of computer power, and fueled by the rapid advancement of gaming and graphic applications, personal computers (PCs) were available by 1987 that met these requirements while being small enough in be placed in a cart. The first PC specifically altered for surgical use was the Heilbrun Stereotactic Model One, using an IBM PC with a graphics card capable of the intraoperative display of diagnostic images [5]. An added benefit of this system was its use of a common stereotactic apparatus, the BRW system, to establish a surgical coordinate system. We acquired the first production model of this
computer with the specific intent to employ it within a navigational system that would use an alternative means of tracking surgical instruments rather than relying on the arc-based BRW system.
The First Surgical Navigation Prototype The Model One was originally developed to replace the small laptop-style calculator that came with the BRW system, and to allow the required stereotactic calculations to be performed in the operating room rather than using the computer associated with the scanning device. Our team used the display capabilities of the system to serve as the display device for our navigational system. The Model One was modified by the group in Utah to display on the pre-operative scans the position of an instrument whose position was known within the BRW coordinate system. For our first prototype the registration of the preoperative images, and the establishment of an intraoperative coordinate system, would be handled by the BRW frame, as this stereotactic system had already established its accuracy and was clinically accepted and approved. The remaining requirement was for the tracking of instruments within the coordinate system established by the BRW frame. This required the use of a three-dimensional (3D) digitizer to produce these coordinates, and the development of instrumentation which could be tracked by the digitizer system employed. Further, additional software was needed to transfer these coordinates from the digitizer software into the visualization software from the Utah group. A major benefit of the Model One was the use of an International Business Machines (IBM) PC, running a standard Microsoft operating system (MS-DOS). Even at that early stage in the use of PCs, we were able to leverage our knowledge of this operating system, coupled with software routines available
The history, current status, and future of the StealthStation treatment guidance system
for the platform, to rapidly program the first prototype. An integral aspect of our development was that all hardware and software components of the system had to be modular in nature to allow us to upgrade or change a specific component without interfering with the operation of the remainder of the system. This design constraint was based on our appreciation of the speed of development of both computer capability and digitization technology, and the recognition that the system should be capable of benefiting from the replacement of any component that was rendered obsolete by the rapid pace of technological evolution. Therefore, the digitization hardware and software was separate from the visualization hardware and software, and prototypes could be updated quickly to take advantage of new technology.
Digitization in Navigation At the time of initial development of the first prototype, the only digitizers that were sufficiently accurate for generation of instrument coordinates were mechanical, electromagnetic or acoustic in nature. Mechanical digitizers, using angle detectors mounted in jointed arms to determine the position of a surgical probe, had already been employed in surgical navigational systems by Watanabe [6]. Although accurate, these mechanical arms had serious ergonomic drawbacks, and their intraoperative use was similar to operating with a device that resembled, and handled like, a dentist’s drill. It was our impression that manipulating such a system within the confines of the deep and narrow exposures routinely employed during neurosurgery would be problematic, and it was therefore not the system of choice for general application. Electromagnetic digitizers determine position using a magnetically sensitive probe placed within a generated magnetic field. In the late 1980s these systems used magnetic fields which were easily distorted by the presence of ferromagnetic
36
substances. Given that the vast majority of surgical instruments are made of such substances, it was our impression that such a system would never be sufficiently accurate for use in neurosurgery. Therefore, at the time of our initial development, acoustic digitizers appeared to be the best devices for surgical use. Acoustic digitizers measure the time of flight of a sound pulse generated by an emitter (such as spark gap emitter) to a microphone. Given the speed of sound, the distance of the emitter from the microphone can be determined. Using an array of microphones, the position of the emitter can be triangulated and determined within a three-dimensional space. As these digitizers require an unobstructed path from the emitter to the microphone, a simple one-emitter solution could not be used for surgery as the instrument being localized would be within the body of the patient. The solution consisted of positioning at least two emitters along the handle of the instrument, with each emitter being outside the body of the patient with an unobstructed view of the microphone array. By designing and fabricating an instrument with emitters located at known distances from the tip of the instrument, and locating the position of each emitter, the position of the probe tip could be calculated. Given these advantages, acoustic digitization technology appeared to be ideal for neurosurgical applications. The only issue that remained was the location of the microphone array; and, given the geometry of the surgical field, it became apparent that locating the array within the surgical lights would maximize the chances of the emitters being seen by the array. Once the decision had been made as to the digitalization technique, it was necessary to decide what to track in the surgical field. Prior navigation devices had tracked the eye of the surgeon by connecting digitalization technology to the microscope. An example of an eye-tracking system was Kelly’s Compass system [7], consisting of a mechanical digitizer built into the microscope
545
546
36
The history, current status, and future of the StealthStation treatment guidance system
support to display position using a framed stereotactic system. The same concept was used by Roberts to track sound emitters attached to a surgical microscope using an acoustic digitizer [8]. Roberts described using the system to perform spinal surgery which precluded the use of a stereotactic frame. The system therefore employed a frameless spinal registration technique using markers (called fiducials) applied to the skin over the area of interest prior to obtaining pre-operative images. This first implementation of a frameless stereotactic system was hindered by the inaccuracy in registration caused by the deformation of the spine, and movement of the fiducials relative to the spine, between imaging and surgery. This inaccuracy resulted in infrequent use of frameless registration for spine surgery; however, the frameless solution appeared to be highly applicable to cranial surgery, given the thinness of the soft tissue overlying the skull, as opposed to that over the spine. With less soft tissue, the inaccuracy of the registration process could be minimized by using multiple fiducials and averaging out inaccuracies caused by local distortions. In order to be applicable across a broad spectrum of surgeries, including those that did not involve the use of the microscope, we felt that . Figure 36-1 Sonic base ring for BRW frame
the concept of tracking the surgical instrument, as suggested by Watanabe [9], should be coupled to the use of a sonic digitizer, as employed by Roberts. We proceeded with the modification of a surgical instrument by attaching sound emitters. Given that a bipolar coagulating forceps was one of the most commonly used instruments during a cranial procedure, and that this instrument already had a cable connected to it, the first tracked instrument so modified was a forceps with two emitters attached in a known geometry, produced by Karl Storz in 1990. It was also necessary to track the position of the body part undergoing surgery (> Figure 36-1). To speed development and to insure accuracy of the first prototype, we decided to register the images using the BRW frame and N-bar fiducials employed by the Heilbrun system. The BRW stereotactic frame attached to the patient prior to imaging has three attachment points for placement of the N-bar fiducial cage. This cage is attached during imaging and then removed, and in traditional stereotactic surgery an arc system is attached to the frame to allow biopsy and creation of lesions. Rather than using this arc system, we created a reference system of three sound emitters attached to a ring with three
The history, current status, and future of the StealthStation treatment guidance system
frame attachment balls. This was constructed in the Saint Louis University machine shop. This reference system attached to the same points as the usual arc system, and to simplify calculations the emitters were positioned in the base plane of the BRW coordinate system (z ¼ 0). Software provided by PixSys of Boulder, Colorado, generated the coordinates of the tip of the instrument within the BRW coordinate system, coordinates that could then be transferred to the Heilbrun routines to display position using their modified PC. Using an early version of a multitasking operating system called DesqView (Quarterdeck), the routine that generated positional coordinates was operated in one block of memory and was then transmitted, via a scripting macro, to the localization routine in another block of memory. The first prototype indicated the forceps position by selecting the closest axial CT image and placing a blue dot over the location of the tip of the surgical probe. The system required 5 s to display position after activating the system using a floor switch. The graphics card employed in the PC was not capable of generating coronal or sagittal projections, and could not produce interpolated axial images between the axial CT images. Therefore, it was
36
imperative to obtain a pre-operative CT scan with thin slices to minimize error in z localization; the system employed generated slices 1.5 mm thick to address this issue. The accuracy of the system was extensively investigated prior to clinical application by using the phantom base supplied with the BRW system. The sonic localizing ring was attached to the phantom base, and the forceps were rigidly attached to the pointer of the base, with the tip of the forceps touching the tip of the pointer. This technique essentially converted the phantom base into a calibration jig. For these experiments, the microphone array was suspended above the phantom base at a distance consistent with that achievable in the operating room. The pointer was moved through a variety of specific coordinates using the Vernier scales in the base, and we verified that the sonic localizer reported the same coordinates throughout a volume which would normally contain the head of a patient (> Figure 36-2). We then wanted to check the accuracy of the system with a patient within the ring to make certain that the presence of the head did not cause a distortion in the system; and we wanted to make certain that ambient noise did
. Figure 36-2 BRW phantom base with sonic base ring and sonic forceps attached
547
548
36
The history, current status, and future of the StealthStation treatment guidance system
not render the system inoperative. These concerns were addressed by conducting experiments on patients with attached frames who were undergoing stereotactic radiosurgery using a linear accelerator at Saint Louis University. For these early cases using radiosurgery, a stereotactic floor stand was attached to the BRW patient ring and used to support the patient’s head. The target to be treated was entered into the Vernier scales of the floor stand, which would bring the target into the center of the LINAC beam. Prior to treating the patient, the sonic reference arc would be attached to the ring, the floor stand coordinates would be set to an anatomical landmark on the surface of the patient, and the room’s lasers were used to verify the accuracy of the floor stand in identifying the anatomical landmark. The landmarks were then touched to make certain that the forceps identified the landmark on the display screen of the system. As the LINAC produced a great deal of noise, this was a robust test of the system which indicated, prior to any use of the system in the operating room, that the sonic system was relatively immune to ambient noise and was accurate in the presence of human anatomy attached to the ring. This was demonstrated with both
CT and MRI scans being used for the imaging technique.
Clinical Use, Testing, and Refinement of Model One From July 1991 through January 1992, we used the first prototype to perform nine sonically navigated cranial procedures under a research protocol approved by the Institutional Review Board (IRB) of Saint Louis University. Procedures were chosen that could be completed in a normal fashion should the system fail. For each of these procedures, a BRW frame was attached prior to placing the patient under anesthesia, and imaging was performed using either CT or MRI with the appropriate localizer attached to the ring. In the operating room, the microphone array was attached to a stanchion supporting the operating room lights and centered above the patient’s head (> Figure 36-3). The patient was then brought into the operating room, placed under anesthesia, and prepped. Prior to making the incision, the emitter-equipped base ring was attached to the frame. This allowed the arc system to be used to
. Figure 36-3 Microphone array over operative field and sonic workstation
The history, current status, and future of the StealthStation treatment guidance system
confirm the location produced by the sonic system. Points were marked on the scalp of the patient after draping, and the coordinates of each point determined by attaching the arc system, pointing to the spot using the arc, and then placing the arc on the phantom base. By moving the pointer of the phantom base to the tip of the arc, the coordinates of each point on the scalp could be determined. We could then determine the accuracy of the prototype throughout the course of the procedure by pointing to these three points with the forceps. The error of the system could be determined by measuring the distance of the coordinates of the forceps when pointing to these calibration spots from the actual coordinates as determined by the arc system; we found the accuracy of the system to be within 1.5 mm, and felt that the utility of the system had been validated. However, during use, the system would occasionally display a position for the forceps that was outside the patient’s head or far removed from where the procedure was occurring. Upon reviewing the triangulation data coming from the timeof-flight calculations generated by the sonic digitizer routines, it became apparent that, during erroneous localization, the time of flight from an emitter to the microphone array was much longer than usual for correct localizations. We were obviously dealing with echo issues that were inherent to acoustic digitizers. Sound coming from the emitter could bounce off a reflective surface and be picked up by the microphone, causing the emitter to be localized far from its actual location. We attempted to mitigate echoes during surgery by using blankets on the walls, positioning equipment so that flat sides would not be perpendicular to the surgical field, and positioning personnel strategically to absorb echoes. In spite of these measures, it became apparent that an acoustic solution would not be practical for a wide variety of operating rooms which had highly reflective surfaces such as tiled walls. Fortunately, at this pivotal juncture in the development process, optical digitizers became commercially available.
36
Optical digitizers employed either reflected or emitted points of light focused upon an array of optical cameras usually consisting of charged coupled device (CCDs). In a manner analogous to human stereoscopic localization, the cameras, separated by a known fixed distance, could determine the position of the source of light by comparing the pixels on the CCDs illuminated by the light coming from the object (> Figure 36-4). To avoid becoming confused by light emanating from the surgical field, these devices generally used light of a highly controlled wavelength in the infrared spectrum, and filters were placed over the cameras to allow only this wavelength to reach the CCDs. The first optical digitizers to become available employed light emitted from infrared light-emitting diodes (IR LEDs). To change our prototype to an optically based system, all that was needed was to mount LEDs in place of the sound emitters on the forceps, to replace the emitters on the reference ring with LEDs, and to suspend the camera array where the microphone array had been positioned for the first prototype. As with the sonic system, determination of the position of the tip of a probe within the body of the patient required at least two IR LEDs to be mounted on the handle of a surgical instrument, the geometry of the instrument being programmed into the system. Given this similarity, many of the same software routines from the acoustic prototype were used for this new optically based system, and the optically based navigational system was ready for clinical trials by the end of 1991. Aesculap created the first commercially available LED forceps for the second prototype in 1991 (> Figure 36-5). It also became apparent after clinical use of the sonic system that indication of surgical position using a single axial image, although useful, did not convey a complete appreciation of location in the superior/inferior dimension. Therefore, the surgeon was forced to create an appreciation for location in all three dimensions. Although other systems in use during this period
549
550
36
The history, current status, and future of the StealthStation treatment guidance system
. Figure 36-4 Original optical cameras
. Figure 36-5 Operative picture showing closeup of forceps and base ring with LED
were capable of producing sagittal and coronal images by reformatting the axial images, the computation power needed to render such images in a reasonable time-frame far surpassed that available when using an IBM PC-based
system. Therefore, we made the decision to create a new system that was capable of reformatting images continuously, based on a system by the leader in computer graphics at that time, Silicon Graphics Incorporated (SGI).
The history, current status, and future of the StealthStation treatment guidance system
Integrating Frameless Localization One of the goals of the initial design of the first prototype was frameless localization, as pioneered by Roberts in his use of skin-mounted fiducials. However, due to the inaccuracy of skin-based fiducials, as witnessed by Roberts in spine applications, and the computational requirements for such a registration, this goal was not realized in the initial prototype. It was thought that the inaccuracy of skin-based fiducials would be minimized in cranial applications due to the proximity of scalp-based markers to the rigid body of the skull, which would limit the amount of movement of fiducials relative to the intracranial anatomy between imaging and surgery. However, there remained the possibility of sliding or slipping of any fiducial attached to the scalp due to the natural movement plane of the subgaleal space. One way to address this source of error was presented by Allen [10] in his proposal for skullbased fiducials. These markers were screwed directly into the skull, thereby eliminating movement between the marker as seen on imaging and the anatomy within the skull. The markers were designed to hold a variety of contrast agents that could be seen by the imaging technologies employed, such as CT, MRI, or even positron emission tomography (PET) [11]. These fiducials would eliminate error due to scalp movement, but would not correct for movement in the position of the brain relative to the skull between imaging and surgery. Although it was believed that the brain did not move significantly within an intact skull, the movement of the brain following opening of the skull became grounds for the future development of intraoperative imaging. Skull-based fiducials were demonstrated to be highly accurate and potentially employable by our system; however, this solution, requiring a small but potentially painful procedure prior to imaging, was felt to be necessary only in those applications requiring the highest accuracy.
36
For the much wider variety of intracranial procedures commonly performed, we felt that a multiplicity of scalp-based markers would be sufficient, as long as the registration routine employed by the navigation system compared distances between markers to determine which markers had moved between imaging and surgery.
Second Generation Navigational System After the initial clinical cases with the first sonicbased prototype, it became clear that the PCbased Model One was not sufficiently powerful to generate the visualization necessary or to implement ‘‘frameless’’ registration based on point fiducials. The Model One required approximately 5 s to display the pre-operative CT slice with overlying cross-hairs, a time-frame that was unacceptable in the intra-operative situation. This was true for both the optical and sonicbased versions of the initial prototype. Therefore, even though the optical system was available for clinical trials late in 1991, we had reached the conclusion that considerable programming and engineering support was needed to develop a completely new second-generation prototype, along with experience in programming SGI computers. Dr. Kenneth R. Smith, MD, who at the time was the Director of Neurosurgery at Saint Louis University School of Medicine, referred us to a development team at Southern Illinois University at Edwardsville (SIUE) headed by his nephew, Kurt Smith, PhD, an assistant professor at that institution. Dr. Smith’s team was developing a digital audio recording and editing device called the StealthStation. The team consisted of students working at the lab and included Kevin Frank, Kurt Noltensmeyer and Paul Kessman, all of whom had considerable experience working with the Silicon Graphics processors that were at the heart of the StealthStation. After
551
552
36
The history, current status, and future of the StealthStation treatment guidance system
an initial meeting and discussions, the focus of his group was expanded to include the development of a navigational system. After a year of development work, the audio device was abandoned, but the name of that recording device was salvaged and applied to the new navigational system. It should be noted that the name of the system was initially selected to appeal to the potential users of the audio device, such as rock and pop musicians, but was subsequently found to appeal to surgeons as well. Given the intensity of the development work, SIUE granted credit to the students working on this project at an off-site location, and a separate corporate entity, Stealth Technologies, was formed. A suitable Silicon Graphics Incorporated processor with sufficient capability to handle the more complex frameless registration algorithms and visualization demands was chosen, taking advantage of the then unparalleled computer graphics capabilities of this system. The initial software was actually patterned after software used for visualization of oil fields to allow more efficient placement of oil wells (> Figure 36-6).
In early 1992, the IRB protocol was modified to allow the use of the new optical system, and initial procedures were performed with a framed registration solution, the sound emitters on the ring being retrofitted with LEDs. After extensive lab-based testing, the first frameless optical procedure was performed at Saint Louis University Hospital on 21 Feb 1992. After the development and refinement of multiple prototypes, an optical ring was produced by Stealth Technologies, and the Neurostation, the second-generation navigational system, was available for clinical testing.
Early Additions to the Surgical Navigation Device In October 1993, collaboration was begun with Moeller-Wedel and the SIUE group, now Surgical Navigation Technologies, on the development of a computer-tracked microscope. In February 1994, the first microscope-tracked cranial procedure was performed at St. Louis University Hospital using
. Figure 36-6 Workstation showing hybrid system of a PC and SGI computer working together
The history, current status, and future of the StealthStation treatment guidance system
LEDs mounted on a bracket attached to the microscope, essentially using the microscope as an additional instrument, or pointer replacement, for the Neurostation. The robotic features of the microscope, consisting of auto focus and auto position, as well as video input into the Neurostation and calibration verification, were all co-developed by Moeller-Wedel and Surgical Navigation Technologies in 1995 (> Figure 36-7). Kevin Foley, MD, of the University of Tennessee in Memphis, approached the development team in 1994 to explore the potential for a spinal application of the system. With the frameless cranial solution fully developed, Dr. Foley suggested using anatomic spinal landmarks exposed within the surgical field to serve as fiducials for the registration process. Registration would therefore not occur at the beginning of the procedure, as with the cranial application, but after the exposure of these points during surgery. This called for a modification of the LED reference array, and the first spinal array was created by mounting LEDs onto a locking pliers which was clamped onto the spinous
. Figure 36-7 Operative microscope with LED tracking array
36
process of the vertebra undergoing surgery. The first frameless stereotactic procedure on the spine was performed by Dr. Foley using the system on 20 July 1994. With the success of this procedure, it was apparent that the Neurostation had applications in both cranial and spine procedures, and as there was concern over complications associated with pedicle screw placement [12] during this period, financial support for further development was obtained from Sofamor Danek, a leading manufacturer of pedicle screws. With this funding, a third prototype was developed that was easier to use by surgical teams outside Saint Louis University and Memphis. Sixteen units of this first prototype, marketed under the name StealthStation, were sold throughout the world, and those systems were used to support a Food and Drug Administration (FDA) application for the device, which was granted in 1996. Sofamor Danek wholly acquired Surgical Navigation Technologies in 1996, and they actively marketed the StealthStation system until 1999, when Sofamor Danek was acquired by Medtronic.
553
554
36
The history, current status, and future of the StealthStation treatment guidance system
Current Status of the StealthStation Treatment Guidance System From 1999 to 2008, the StealthStation system has realized continuous advances due to faster computers and graphics processors, new software algorithms, increased focus on user-centric design, and incorporation of intra-operative imaging. Usage has expanded to include a number of additional cranial, spinal and ENT procedures. As a framework for discussion of both the current and future status of the StealthStation system, we have chosen the following essential elements of contemporary navigation: 1. 2. 3. 4. 5. 6. 7.
Preoperative imaging Patient and image registration Intra-operative tracking Visualization System control Effectors Intra-operative imaging
Preoperative Imaging The process of transferring preoperative images into the navigational system was viable yet cumbersome with the early prototypes and commercial systems. The initial protocol for importing images involved the use of unwieldy tapes or optical disks to transfer raw scanner media into the system. As Digital Imaging and Communications in Medicine (DICOM) became the standard mode for distribution of images, importation became more facile. While DICOM remains a preferred transfer medium, the StealthStation system has been evolved to import many different data types via a variety of media, comprising the Ethernet, compact disk (CD), medical imaging archiving systems (PACS) and, in the near
future, flash drives attached via the universal serial bus (USB). A broad spectrum of imaging modalities from radiological scanners can be input into the StealthStation system. The most common imaging modalities used in navigation are MRI and CT. Other anatomic modalities include computed tomographic angiography (CTA) and magnetic resonance angiography (MRA). At the start of the twenty-first century, functional imaging modalities such as functional MRI (fMRI), positron emission tomography (PET), and single photon emission computed tomography (SPECT) became prominent, and commercial neuronavigation products, including the StealthStation system, began to allow importation and manipulation of these modalities as well.
Patient and Image Registration Once image data is imported into the StealthStation system, the user has the option of coregistering two or more volumetric medical images into a common coordinate system so that they can be fused and visualized together. Advanced image-processing algorithms made StealthMerge possible in 1998. This multimodality image fusion software permitted precise correlation and merging of multiple image data sets (e.g., CT fusion with MR, PET fusion with CT, etc). This software allows surgeons a combined view of complementary image information (e.g., bony anatomy from CT with soft tissue from MR, or functional information from PETwith anatomy from MR). This fusion and visualization provides additional information to surgeons, allowing greater confidence during stereotactic procedures. The ability to create surgical plans has been available from the earliest incarnation of the Model One. These plans are often used to represent a desired outcome during surgery, such as the path to be traveled by a biopsy needle or the implanted location of a pedicle screw. Over
The history, current status, and future of the StealthStation treatment guidance system
the past several years, surgical planning tools have begun to make increasing use of information about the therapies to be delivered or the procedures to be performed. For example, planning capabilities in the StealthStation system’s spinal software incorporate three-dimensional models of each of the pedicle screws sold by Medtronic Spine and Biologics. These models allow preplanning of screw placement and confirmation of correct sizing and screw type selection. Software used for performing biopsies incorporates information about the location of the cutting window on the biopsy needle to ensure accurate placement of the needle and resulting tissue sample.
Intra-operative Tracking Intra-operative tracking is the position measurement technology that provides the basis for real-time feedback of surgical instrument locations during a procedure. For many years, intra-operative tracking in the StealthStation system has been built upon commercially available optical measurement systems from Northern Digital, Inc. (NDI). There are two widely adopted methods of optical measurement in navigation: active and passive, both of which have been employed within the StealthStation system. The infrared LEDs that were used at the inception of navigation are still in use today and are activated by an electrical signal. Passive tracking involves markers made of a retro-reflective material, which reflect infrared light emitted by the optical tracker. In both cases, the optical tracker measures the positions of individual markers attached to an instrument and, based on a model of the instrument stored within the system, can infer the location of the entire instrument, usually with sub-millimetric accuracy. Optical tracking has at least two weaknesses: the need for a line of sight between the instrument and the tracker; and the requirement for
36
rigidity between the markers and the portions of the instrument to be tracked. Within the past 5 years, the StealthStation has incorporated a second form of tracking which uses electromagnetic energy rather than infrared light. Electromagnetic tracking has become a viable and valuable method of position measurement, especially as the applications of navigation have expanded and the number of less invasive procedures being performed has increased. Electromagnetic energy passes through the body and is therefore not susceptible to line-of-sight issues. The electromagnetic tracker can be much smaller than a functionally equivalent optical tracker, thus opening navigation to a host of new applications. Electromagnetic trackers have been manufactured that are approximately 1 mm in diameter and approximately 5 mm in length, whereas a similar optical device may be over 100 mm in length. In addition, electromagnetic tracking provides the ability to track the distal end of a flexible instrument, which expands the applicability of navigation into the realm of various soft tissue procedures and more reliable placement of ventricular shunts. These systems differ from the initial electromagnetic tracking systems available early on in the development of the system in that they have improved preservation of accuracy in the presence of ferromagnetic substances.
Visualization As discussed previously, the preoperative images must be mapped to the physical anatomy of the patient via a registration process. Registration allows for real-time, intra-operative visualization of preoperative imaging data based on the position and orientation of one or more tracked surgical instruments relative to the actual patient anatomy. Registration requires the establishment of a spatial transformation between the respective coordinate spaces of the intra-operative
555
556
36
The history, current status, and future of the StealthStation treatment guidance system
tracker and the image. The most common and longest-standing method for patient registration for the system as described above is paired-point matching, called PointMerge. This approach can use either fiducial markers that are attached to the patient prior to the preoperative scan or anatomical landmarks to establish corresponding landmarks in the system’s coordinate systems for the image and the patient. Based on this data, computer algorithms can compute the spatial transformation between the two coordinate systems. Recent advances in PointMerge include the automatic identification of fiducial markers from image data, as well as the automatic establishment of correspondence between points in the image and patient space. This latter enhancement allows the user to touch fiducial points in any order without the need to manually specify which patient point corresponds to which image point. Another popular method of registration in cranial procedures uses a surface matching technique. This approach extracts the surface of the patient’s face from preoperative images, and requires the user to collect corresponding surface data on the actual patient anatomy using a specially designed surgical instrument. This Tracer registration algorithm allows for accurate registration without the use of fiducial markers, even in situations when the patient’s face is not directly visible to the optical localizer. The registration process for preoperative CT images in spinal procedures is a clinically challenging task that has been simplified in the StealthStation system through the use of fluoroscopy. The FluoroMerge registration algorithm uses ‘‘virtual fluoroscopy’’ techniques to perform patient registration. Using these techniques, a synthetic fluoroscopic image can be created from the preoperative CT images (often referred to as a digitally reconstructed radiograph or DRR). The concept is to acquire two actual fluoroscopic images of the spine from different directions. Using a mathematical model of the fluoroscope’s image formation process, the computer can then determine where
a ‘‘virtual’’ fluoroscope would need to be positioned in order to generate DRRs that match the acquired fluoroscopic images. Since the fluoroscope is being tracked by a localizer, the mathematical solution to this problem results in the computation of the transformation between the CT images and the patient. The result is an automated registration.
System Control A core requirement of neuronavigation is the ability to display clinically relevant data from which a surgeon can make decisions relevant to the surgical procedure. Such decisions often result in minimizing damage to healthy tissue, reducing surgical procedure times by selecting optimized paths to a target and making smaller incisions, or increasing the chances of a more complete tumor resection. Early iterations of the StealthStation provided simple grey-scale display of CT and MR images in coronal, sagittal and axial cross sections, along with a 3D surface reconstruction. As graphics and processors improved, the ability to display more complex ‘‘volume rendered’’ 3D images was developed. More recently, there has been a focus on designing visualization capabilities that are best suited to the clinical task at hand. For example, when evaluating the trajectory of a biopsy needle, it is possible to display a ‘‘look-ahead’’ view, which shows the anatomy that the needle will pass through as it continues along the trajectory. A similar view shows the remaining distance to the target, as well as any deviation from the planned trajectory. The ability of the surgeon to visualize anatomy from a myriad of sources, such as multimodality anatomic or functional images, real-time intra-operative fluoroscopy, ultrasound, MR or CT, allows the surgeon to minimize the risk of trauma to surrounding tissue, minimize incision size and displacement, and, ideally, reduce operative times.
The history, current status, and future of the StealthStation treatment guidance system
Over the 18 years since the stereotactic surgery revival, navigation has become an integral part of most neurosurgical practices and has been integrated into the curricula of internships, residencies and medical schools. In order to maintain and increase relevance during this time of rapid proliferation, the StealthStation system needed to evolve with the advances in other areas of medicine, such as imaging, communications and operating room control. In 2005, Medtronic Navigation embarked on an extensive study of navigation system usability and workflow and the application of user-centric design, resulting in several ergonomic and workflow enhancements to the StealthStation system. The result of this study was the creation of the ‘‘Synergy Experience’’ for the StealthStation system, which incorporated innovations in display technologies, user interface devices, and software user interfaces. In addition to designing products for the surgeon, much consideration was given to the operating room support staff and their needs. The resulting products are greatly simplified and intuitive compared to their predecessors, and incorporate procedure specific workflows, wireless mice and built-in surgeon customization to help minimize operating room time and workflow, and maximize efficiency. For example, it is possible to pre-define operative room layout, instrument usage, visualization options and software workflow on a procedure-by-procedure basis for each surgeon that uses the system. In addition, features such as peripheral connectivity dashboards allow support staff to ensure that connections to other devices in the operating room (e.g., surgical microscopes, fluoroscopes, etc.) are operating properly before the surgeon even enters the room.
Effectors As intra-operative tracking technology progressed, the need to track all manner of surgical instru-
36
ments, from simple to complex became imperative, in order to offer surgeons an unprecedented visual window into the surgical space. As discussed previously, the first instruments to be tracked consisted of a coagulating forceps followed by several simple instruments for bone cutting and suction. Further developments enabled the tracking of many spinal surgery tools, such as drills, screw systems and delivery instruments for inter-vertebral devices. Using a tracked surgical microscope, an ultrasound probe, or even an intra-operative imaging system such as the PoleStar1 iMRI System or the O-Arm1 Intra-operative Imaging System provides real-time guidance to accommodate changes in anatomy that occur during the course of a procedure (> Figure 36-8). As surgeons became more proficient with neuronavigation and refined the art, a potential drawback to using frameless approaches for biopsies and functional neurosurgery was the lack of a rigid frame to provide support. The StealthStation Navigus device uses a disposable trajectory guide that is mounted directly to the skull with titanium screws. This device allows the surgeon to insert a probe into the guide and pivot it via a joystick-like mechanism to achieve the proper trajectory based on images of the brain as displayed on the workstation. The trajectory is then fixed through a locking ring. Another way to address this issue is via the attachment of the Vertek arm directly to the operating room table. This flexible arm can be locked in position once the target trajectory has been selected. As surgeries and techniques evolve, applications of such rigid fixation devices will expand beyond biopsies and shunt placement into drug delivery, tumor removal, biotechnology and neurostimulation.
557
558
36
The history, current status, and future of the StealthStation treatment guidance system
. Figure 36-8 Intraoperative MRI system
Intra-operative Imaging The overarching goals of twenty-first-century navigation include reduced invasiveness, improved outcomes, increased efficiency and productivity, and more cost-effective clinical solutions. The incorporation of intra-operative imaging into the practice of surgery provides immediate and dynamic images to support clinical decisionmaking in the operating room. In addition, when coupled with navigation, intra-operative imaging facilitates the radical simplification of the patient registration process, allowing for automatic registration of images to the patient with no action required by the user. As the surgery is performed, surgeons have the capability to visualize changes in the patient’s anatomy and react immediately to those changes as they acquire and review updated, patient-specific medical images during the procedure. The first intra-operative imaging device incorporated into the StealthStation system was based on ultrasound. The SonoNav application allowed for side-by-side viewing of real-time ultrasound with preoperative CT or MR. Several years
later, virtual fluoroscopy was introduced in the form of the FluoroNav application. This innovation allowed the acquisition of multiple fluoroscopic images and subsequent navigation based on these images, thus enabling multi-planar fluoroscopic capabilities. In early 2003, the PoleStar1 iMRI System was introduced. This MR imaging technology was designed specifically for cranial neurosurgery to help surgeons monitor tumor tissue throughout the course of a resection, thereby increasing the likelihood of more complete resections. The O-arm1 intra-operative 2D and 3D fluoroscope was designed to provide 3D X-ray-based imaging for bony anatomy. This system, which incorporates robotic positioning, a unique gantry that opens for positioning around operating room beds, a flat panel detector, automatic registration and many other innovations, has recently gained popularity in spine surgery. This intra-operative imaging device has tremendous promise for future applications (> Figure 36-9). The StealthStation system has always been a platform for research. With the addition of the StealthLink networking capabilities, a con-
The history, current status, and future of the StealthStation treatment guidance system
. Figure 36-9 Intraoperative O-arm1
duit between external research tools and the StealthStation is possible. Many exciting projects are currently in development using this collaborative tool. Enhancing surgical performance and enabling new patient treatment options are the hallmark of the all-encompassing navigation systems of the modern-day operating room. As the future unfolds, the StealthStation system is becoming a node on the information highway, offering an integrated digital infrastructure for surgeons as well as hospital managers and support staff. Hospitals and surgeons have access to an integrated navigation and intra-operative imaging hub that empowers them to make data-driven decisions in the OR as well as to maximize cost efficiencies in today’s cost-conscious healthcare marketplace.
Future Development The system should be considered as being still in its adolescence. The StealthStation is being actively developed to address new procedures being performed in neurosurgery and to take advantage of advances in new technologies which will improve the functionality of the device (> Figure 36-10).
36
. Figure 36-10 Current S7 system
The remaining portion of this chapter will focus on neurosurgical applications only, but some of the most exciting developments with the system are occurring in specialties other than neurosurgery. To reiterate our framework for discussion, these are the essential elements by which we will discuss contemporary navigation: 1. 2. 3. 4. 5. 6. 7.
Preoperative imaging Patient and image registration Intra-operative tracking Visualization System control Effectors Intra-operative imaging
Preoperative Imaging Imaging modalities available and useful to neurosurgery continue to develop, as has been the case
559
560
36
The history, current status, and future of the StealthStation treatment guidance system
since the revival of stereotactic surgery was initiated in the late 1980s. Most recent advances have focused on the imaging of brain function, via magnetoencephalography (MEG), and functional MRI (fMRI). Although these modalities have been partially implemented in the StealthStation system, further development is in progress to fully integrate them and allow their use on a routine basis, and to make such use practical for physicians without requiring extensive imaging support staff. As the use of these imaging technologies becomes more commonplace and review of the appropriate images prior to a surgical intervention becomes necessary, the concept of a planning station becomes more appealing. Ideally, it would be best to not only analyze the images needed to plan the procedure, but also to actually perform and merge the actions required for the procedure using the preoperative images. This would require not only fusion of different imaging modalities, but also segmentation of key structures on these images and implementation of specific surgical plans which, once perfected, would have to be registered to the surgical act. Further, it would be optimal to plan, and practice, the procedure with three-dimensional vision, or stereoscopy, in order to produce the most realistic environment for generating a valid plan. Such a comprehensive integration of a surgical planning/ practice system into the intra-operative navigation system is being developed in cooperation with Volume Interactions of Singapore. This stereoscopic planning system uses a magnetic threedimensional digitizer to track the movements of the surgeon as tissue is removed while displaying any form of medical imaging in a co-registered three-dimensional virtual body. The final surgical plan is then transmitted to the StealthStation using Stealth Link, described in the preceding section. The combined system enables the visualization of all key structures and targets for the procedure prior to encountering them in surgery, and the stereoscopic projection allows precise
surgical paths to be generated and followed by the operator.
Patient and Image Registration The registration techniques described above employ either artificial or anatomical landmarks seen on preoperative imaging to register those images to the surgical workspace. With the increased availability and use of intra-operative imaging, an alternative approach could be the use of images acquired in the operating room to perform the registration process. For example, an intra-operative CT scanner, rendering a detailed representation of the skull, could be employed in a process identical to the aforementioned image fusion process with the intraoperative MRI scanner to perform an image fusion between the intra-operative images and the preoperative images. Critical to such an application would be a means by which to correlate the intra-operative image to the coordinate system of the procedure. This could be accomplished by rigidly attaching the skull to the CT scanner or, to allow the CT scanner to be moved out of the way during the surgical intervention, a tracker device could be placed on the skull prior to imaging to allow the anatomy to be tracked after the CT scanner has been retracted. This technique would allow any preoperative diagnostic image of sufficient resolution for navigation to be employed intra-operatively without the application of fiducials prior to imaging. The accuracy of the registration process should be equivalent to, or surpass, that of using either a stereotactic frame or skull-based fiducials. This technique would thereby eliminate the discomfort of skull-based fiducials, and lower the barrier to use of image guided techniques for a variety of surgical interventions, as any diagnostic image could be used for navigation.
The history, current status, and future of the StealthStation treatment guidance system
36
Intra-operative Tracking
Visualization
The vast majority of navigational devices currently employ two relatively mature tracking technologies, electromagnetic and optical. It does not appear that either technology will eliminate the other in the near future, as each has benefits and drawbacks that render them complementary to each other rather than essentially equivalent. For example, the ability of electromagnetic devices to track effectors within the body of the patient that are not rigidly attached to any tool outside the body renders electromagnetic localization the technique of choice for endovascular procedures. This benefit is so critical in these procedures that the need and expense of re-engineering the surgical workspace to eliminate ferromagnetic instruments is easily justified. However, in open procedures in which a variety of instruments must be employed, the relative robustness of optical localization will probably remain the technique of choice unless electromagnetic localization can be made immune to the presence of ferromagnetic substances within the surgical field. One possible alternative, or adjunct, to these tracking technologies is the use of intra-operative imaging to track effectors. For example, during placement of a stimulator electrode, intraoperative CT or ultrasound could be used to visualize the tip of the electrode as it is inserted into the region of interest. However, even with this simple example, the use of continuous imaging to navigate may expose both patient and surgeon to excessive amounts of ionizing radiation, and intense use of ultrasound frequently results in procedure delays. Therefore, even in this example, some sort of simple tracking of the effectors would seem to be preferable to a continuous imaging solution. Verification of position using intra-operative imaging would seem to be best used to confirm the correct position of an effector after it is navigated into position using standard tracking technologies.
The sheer amount of information becoming available about the structure and function of a specific patient’s brain will soon overload the ability of the clinician to visualize the entirety of the information at a glance. What is needed is method within the StealthStation system by which to selectively de-emphasize certain information when not applicable to the particular task at hand. For example, MEG information is not particularly germane during dissection of the deeper portion of a tumor, whereas DTI information becomes far more relevant. To avoid ‘‘information overload’’ of the surgeon during the procedure, the system would have to be ‘‘situationally aware’’ about the task at hand and suppress information not of use to the surgeon at that point in the procedure. This awareness of the system could be part of the planning procedure; in addition to practicing the procedure using the preoperative images, the system could learn, based on the information selected by the surgeon during the process, what information should be displayed at specific points during the operation.
System Control As surgeons become more reliant on navigational systems to perform surgery, and as the capabilities and complexity of these systems increase, control of the system will become increasingly demanding. The more the surgeon becomes dependent upon the system, the more interactions will occur between the surgeon and the system. The issue of system control has not been adequately addressed by any current navigational device. Use of a touch screen display beneath a sterile transparent drape, while it can increase usability, is problematic in terms of concerns about sterility as well as the decreased contrast
561
562
36
The history, current status, and future of the StealthStation treatment guidance system
in images resulting from the touch screen overlay. Indeed, this problem of contrast has led to the use in the operating room of liquid crystal display (LCD) panels without touch screen capabilities, dedicated to providing the best image for the surgeon and leaving control of the system to another device or person. Computer mice have also been employed in the operating room, but sterilization of these electronic devices can be problematic, and there are few flat surfaces in the operating room that can be used as a base for the mouse. Computer mice designed to be used in free air have also been tried, however these tools leave much to be desired from an ergonomic perspective. Given these restraints, most busy neurosurgical facilities rely on an individual outside the sterile field to control the system, which adds both cost and complexity to routine use of a navigational system. In looking for a solution to this issue of system control, it is best to reflect on what has worked well in the past in operating rooms, and that is the use of voice commands. Almost every communication that occurs in a conventional operating room is verbal in nature, and it would seem logical to employ the same mechanism to control the navigational system. Voice control systems have been experimented with in the operative environment and have been generally rejected on the basis of their lack of reliability, probably due to the high background noise in the operating room, and the large vocabulary needed to respond effectively to an operating surgeon. Voice control systems usually fall into two broad categories: those that require training to a specific voice, and those that do not, literally working ‘‘out of the box’’ and responding to a speaker without prior exposure to that voice. In order to maintain reasonable accuracy and speed of response, non-educable systems would usually have a limited vocabulary, forcing the user to remember specific words. With the advent of increased computer power in the
operating room, and the proliferation of electronic devices, a system that learns the surgeon’s words and adapts to the surgeon, rather than the other way around, would seem to be the ultimate answer to making voice system control a viable solution.
Effectors In spite of considerable developments in visualization, as delineated in this chapter, the effectors used by surgeons, consisting of instruments and devices that manipulate tissue to achieve a desired effect, have remained relatively unchanged over the past decade. The forceps, scalpels, biopsy instruments, and drills that have been used for decades are still being used with more knowledge of the patient’s anatomy. Although some new effectors are in common use, such as ultrasonic aspirators [13] and bipolar coagulators [14], it is safe to say that a surgeon from 30 years ago would be comfortable using the effectors found in the current neurosurgical operating room. This static situation is not the case across all surgical specialties. Very complex and expensive robots are currently and routinely employed in urological procedures. Modern-day abdominal surgery has experienced a renaissance with the employment of new endoscopic surgery techniques which has in turn required the development of a complete set of effectors capable of working through small surgical channels [15]. The application of these new effectors has required significant retraining on the part of surgeons in these fields, an educational effort justified by the benefits afforded by using these technologies. These benefits include fewer complications in robotic urological surgery [16] and less invasiveness coupled with shorter recovery times in endoscopic versus open abdominal surgery. Given these advances in other surgical specialties, it is surprising that similar technological advances in
The history, current status, and future of the StealthStation treatment guidance system
effector design have not been apparent in neurosurgery, with its demand for the ultimate accuracy and precision in movement. It is of interest that the application of these new effector technologies has generally occurred in those fields in which patients routinely pick and choose their surgeon due to the elective nature of the surgical intervention. Therefore, the adoption of such interventions such as endoscopic cholecystectomy has been driven by patients who can and do exert considerable pressure on surgeons to use techniques that maximize effectiveness and minimize impact on the patient. One could imagine that if neurosurgical procedures were more common, and more elective, patients would exert similar pressure on neurosurgeons to adopt new effector technologies. Perhaps the most obvious effector technology that could be used in neurosurgery is the use of robotic devices. Although the term robot implies a device that moves on its own without necessarily having human control, the term has been broadened to describe a device that can assist a human in performing a task that is difficult to perform without assistance. An area of intense interest is the concept of a scaling robot [17], in which the movements of the surgeon in traditional dimensions can be scaled to a smaller, even microscopic level. For example, the very small movements necessary to clip an aneurysm could be performed by the surgeon on a human scale and translated to the microscopic level by a scaling robot. This type of application has already been employed in the urological robotic application previously mentioned. A significant factor holding back neurosurgical robotic applications is the very small corridors employed to reach neurosurgical targets, which are not compatible with the size of robotic effectors currently employed. At many engineering centers, this limitation is being overcome by miniaturization of robotic devices that have been specifically designed for neurosurgical
36
applications rather than being simply borrowed from some other field. The term effector could also be employed for non-mechanical agents that achieve a desired therapeutic effect. Recently, there has been considerable attention in neuro-oncology to the use of intra-tumoral injection of chemotherapeutic agents to address the infiltrative nature of malignant glial neoplasms [18]. This area of medicine, termed convection-enhanced delivery, has proposed the use of a variety of agents for this purpose, many of which have molecules with an attachment point specific to antigens seen on the cell membrane of malignant cells, and a killing moiety which once inserted into the cell through the attachment point results in the killing of the cell. Although the protocols investigating these agents have yet to demonstrate efficacy, there is already a commercially approved chemotherapeutic wafer which has been shown to have some effect on the progression of glioblastoma multiforme [19], and it is to be expected that additional efficacious agents will be found in the future. It is also to be anticipated that the benefit accrued with these agents will be maximized if the agent is placed in the optimal location, and that navigation will have a role in this process. Another area of effector development is neural stimulation. Significant clinical efficacy has been found with localized stimulation for pain of spinal origin [20], Parkinson’s disease [21], and essential tremor [22]. Studies are currently planned on the use of stimulation for the treatment of depression [23]. Stimulation techniques are still in their infancy, and the electrode arrays currently employed are orders of magnitude larger than the structures they stimulate; indeed, as currently employed, stimulation simply drives the implanted structure to the point where the neurotransmitters are depleted, and the net effect is akin to a sort of reversible functional lesion [24]. As our knowledge of the nervous system evolves, and specific targets are found to exert a
563
564
36
The history, current status, and future of the StealthStation treatment guidance system
highly specific effect, it can be imagined that microscopic stimulation devices that can already be manufactured using integrated chip technology could be inserted to achieve a desired effect. The rate limiting factor is the knowledge of exactly where to place the device, and how to get the device to that point; both answers will be found in the further development of navigational devices like the StealthStation system.
Intra-operative Imaging As mentioned in the previous sections of this chapter, intra-operative imaging has become routinely employed in more complex procedures to improve the overall accuracy of surgical interventions. The implementation of these imaging technologies introduces additional complexities and delays into procedures already known for their complexity and length. For example, intraoperative MRI, particularly using high-field magnetic devices, imposes significant issues upon the surgeon in conducting a typical cranial procedure. It is to be anticipated that with further development the difficulty in using current imaging technologies will be reduced, and, as the threshold to use is reduced, the application of intra-operative imaging will increase. Of greater interest is the advent of new ways to ‘‘image’’ or ‘‘visualize’’ the surgical environment to further improve the surgical act. These techniques may or may not fall into what is normally considered imaging; however, as they do impact the overall situational knowledge of the surgical environment, they can be considered as a form of imaging insofar as that term is used to imply a device that improves knowledge of the surgical situation. One technique that is already being used is the application of fluorescence. By injecting a contrast agent that attaches to specific tissues within the surgical field, and exposing the agent to specific wavelengths of illumination, the tissue
so labeled will exhibit fluorescence to differentiate itself from surrounding, non-labeled tissue. A major issue with this technique is that surgery cannot proceed during specific wavelength illumination; therefore, in order to make maximal use of this technology, the illuminated field needs to be back-projected into the surgeon’s view of the field under normal illumination conditions. The StealthStation system could implement this process by tracking a device viewing the surgical field, such as a microscope, taking an image of field, and then superimposing the illuminated field onto the field visualized under normal illumination. As agents evolve to label specific components of the brain (normal or abnormal), it can be anticipated that surgical interventions will be improved by this additional knowledge gained during surgical exposure. Another ‘‘imaging’’ technique could employ sampling and analysis of tissue with respect to the specific properties of the tissue. For example, if the target tissue had specific chemical characteristics, then analysis with a miniature mass spectrometer could allow resection based upon the presence of that chemical characteristic. Alternatively, analysis of tissue DNA, perhaps made possible by DNA chip technology, could allow resection based upon the elaboration of specific genetic traits, which could prove quite useful particularly in the resection of malignant glial neoplasms. Finally, navigation through the brain could be performed using the specific signals present within the brain. Current neurophysiologic techniques, using large macro electrodes, must undergo tremendous development to be scaled to the environment at which the brain routinely operates. By using microscopic electrodes, on the scale of a neuron, signals could be recorded that when analyzed (probably requiring computational power rivaling that of the brain) would ‘‘show the way’’ to implant the microscopic effectors discussed in the prior section.
The history, current status, and future of the StealthStation treatment guidance system
Conclusions The StealthStation system has seen tremendous improvement and development in the nearly two decades of its existence. The improvements in surgical technique made possible by the device will only be exceeded by further development of the device. It can be foreseen that, in the near future, almost every neurosurgical device will employ some form of navigation to improve its efficacy and minimize the complications associated with the surgery. This is perhaps the most exciting time to be a surgeon in the field, as these improvements in surgery will redefine what a surgeon can accomplish, and in the process, literally redefine what it means to be a neurosurgeon. Any views, ideas, opinions or historical matter expressed in this paper are solely attributed to the authors. None of the material in this paper should be construed as the views or representations of any third party, including in particular Medtronic, Inc. or any of its divisions or subsidiaries, and such third parties assume no legal liability for the accuracy, completeness, or usefulness of any information contained herein.
7.
8.
9.
10.
11.
12.
13.
14. 15.
References 16. 1. Horsley V, Clark RH. The structure and functions of the cerebellum examined by a new method. Brain 1908;31:45-124. 2. Spiegel EA, Wycis HT, Marks M, Lee AS. Stereotaxic apparatus for operations on the human brain. Science 1947;106:349-50. 3. Gildenberg PL. Whatever happened to stereotactic surgery? Neurosurgery 1987;20:983-7. 4. Heilbrun MP, McDonald P, Wiker C, Koehler S, Peters W. Stereotactic localization and guidance using a machine vision technique. Stereotact Funct Neurosurg 1992;58(1–4):94-8. 5. Heilbrun MP, Brown RA, McDonald P. Real-time threedimensional graphic reconstructions using Brown-RobertsWells frame coordinates in a microcomputer environment. Stereotact Funct Neurosurg 1985;48:7-10. 6. Watanabe E, Watanabe T, Manaka S, Mayanagi Y, Takakura K. Three-dimensional digitizer (neuronaviga-
17.
18. 19.
20.
21. 22.
36
tor): new equipment for computed tomography-guided stereotaxic surgery. Surg Neurol 1987;27(6):543-7. Kelly PJ. Computer-assisted stereotaxis: new approaches for the management of intracranial intra-axial tumors. Neurology 1986;36(4):535-41. Friets EM, Strohbehn JW, Roberts DW Improved accuracy for the frameless stereotaxic operating microscope. In Proceedings of the 14th Annual Northeast Bioengineering Conference, Durham, New Hampshire; 1988. p. 47-9. Watanabe E, Watanabe T, Manaka S, Mayanagi Y, Takakura K. Three-dimensional digitizer (neuronavigator): new equipment for computed tomography-guided stereotaxic surgery. Surg Neurol 1987;27(6):543-7. Maciunas RJ, Fitzpatrick JM, Galloway RL, Allen GS. Beyond stereotaxy: extreme levels of application accuracy are provided by implantable fiducial markers for interactive image-guided neurosurgery. In: Maciunas RJ, editor. Interactive image-guided neurosurgery. Park Ridge, IL: AANS; 1993. p. 259-70. Maurer CR, Jr, Fitzpatrick JM, Wang MY, alloway RL, Jr, Maciunas RJ, Allen GS. Registration of head volume images using implantable fiducial markers. IEEE Trans Med Imaging 1997;16(4):447-62. Laine T, Lund T, Ylikoski M, Lohikoski J, Schlenzka D. Accuracy of pedicle screw insertion with and without computer assistance: a randomised controlled clinical study in 100 consecutive patients. Eur Spine J 2000; 9(3):235-40. Brock M, Ingwersen I, Roggendorf W. Ultrasonic aspiration in neurosurgery: Neurosurg Rev 1984;7(2–3): 173-7. Bulsara K, Sukhla S, Nimjee S. History of bipolar coagulation. Neurosurg Rev 2006;29(2):93-6. Fork F. Endoscopy of the upper gastrointestinal tract: radiology of the stomach and duodenum. Berlin: Springer; 2008. p. 29-72. Shah N, Hemal A, Menon M. Robot-assisted radical cystectomy and urinary diversion. Current Urol Rep 2005;6(2):122-5. Hockstein N, Gourin C, Faust R, Terris D. A history of robots: from science fiction to surgical robots. J Robotic Surg 2007;1(2):113-18. Joshi S, Ornstein E, Bruce J. Targeting the brain. Neurocrit Care 2007;6(3):200-12. Lawson H, Sampath P, Bohan E, et al. Interstitial chemotherapy for malignant gliomas: the Johns Hopkins experience. J Neuro-Oncol 2007;83(1):61-70. Deer T. Current and future trends in spinal cord stimulation for chronic pain. Curr Pain Headache Rep 2001; 5(6):503-9. Watts C. Deep brain stimulation for Parkinson’s disease. Acta Neurochirurgica 2008;150(1):97. Pilitsis JG, Bakay RAE. The future of deep brain stimulation. In: Tarsy D, Vitek JL, Starr PA, Okun MS, editors. Deep brain stimulation in neurological and
565
psychiatric disorders. Clifton, UK: Humana Press; 2008. p. 571-91. 23. Greenberg B. Deep brain stimulation in depression: background, progress, and key issues. In: Tarsy D, Vitek JL, Starr PA, Okun MS, editors. Deep brain stimulation in
neurological and psychiatric disorders. Clifton, UK: Humana Press; 2008. p. 511-29. 24. Stojanovic M. Stimulation methods for neuropathic pain control. Curr Pain Headache Rep 2001;5(2):130-7.
46 Virtual Reality in the Operating Room P. L. Gildenberg
Modern imaging techniques have made the computer reconstruction of three-dimensional volumetric anatomy and pathological targets, such as tumors routine. They can be derived from preoperative CT or MRI scans, or the merging of both, sometimes with the addition of other modalities. The problem still to be solved is What is the most efficient and intuitive way to present this valuable three-dimensional information to the surgeon during an operation? It is virtual reality image guided surgery. In virtual reality surgery, the target tumor as well as significant surrounding or overlying structures are visualized on a monitor as a localized opaque image of the structure, superimposed on a real time video of the surgical field. The images displayed are those of significance to the immediate surgical need, under control of the surgeon by use of the surgical instrument presently in hand. The display of necessary information and none other can be selected. The depth of the image under resection can be determined by the position of the resection instrument and that image can be frozen or updated by means of a foot pedal. In addition, auditory signals can be added to inform the surgeon about the actual position of the instrument in hand in relation to the desired position. The technology to achieve virtual reality surgery is straight forward. A video camera is localized in space, using the same image guidance system as localizing the anatomy or any Electronic supplementary material Supplementary material is available in the online version of this chapter at http:// dx.doi.org/10.1007/978-3-540-69960-6_46 and is accessible for authorized users. #
Springer-Verlag Berlin/Heidelberg 2009
other instrument. The optical characteristics of the camera are known, so the view that the camera sees can readily be registered to the surgical field. The monitor simultaneously displays both the image from the camera and a virtual reality computer generated image of the intended target or other anatomy accurately superimposed. The same registration can be used with an endoscope, which is essentially a video camera with specialized optics attached. In order to be useful for intraoperative image guidance, the three-dimensional anatomy and target volume (1) is reconstructed from a series of two-dimensional imaging slices into a three-dimensional volume, (2) is registered to the patient’s anatomy and image guidance system for localization, (3) is under direct control of the surgeon, (4) is interactive with the surgical instruments, which can be used to control the imaging, (5) is updated to represent the remaining tumor and the part of the tumor being resected at the time, and (6) is compatible with techniques under development to account for shift of the brain or other tissue. The question becomes how can one present such three-dimensional information to the surgeon while limited by two-dimensional displays without making it necessary for the surgeon to change views repeatedly between the surgical field and the monitor or to wear stereoscopic glasses or other modification of vision? Current image guidance techniques identify from preoperative imaging studies not only intracranial pathology but landmarks that may be used to register the patient’s head to the image guidance apparatus and guide the surgeon to the target.
726
46
Virtual reality in the operating room
Landmarks may be based on surface contours or may be based on the localization of several fiducial marks which may be taped to the head prior to the preoperative CT or MR scan. The image guidance computer constructs a volumetric virtual image of the head and face, the fiducials taped to the head, the internal anatomy, and whatever mass is identified as the target. The volumetric image is registered to the patient’s head, which is registered to the same stereotactic space as the image guidance system, which allows the surgeon to identify where a probe or instrument is in relation to the patient’s head and its contents. Presently used monitors present twodimensional pictures. Stereoscopic or 3-D images displayed on a 2-D monitor may use an artificial perspective to provide a simulation of a threedimensional world. Even with stereoscopic viewing techniques, depth perception may not be accurate enough for surgical guidance, and wearing stereoscopic glasses makes it difficult to see the surgical field. The configuration of the image display in presently commercially available image guidance systems is very similar and has not changed appreciably since image guided surgery was introduced almost two decades ago. In the present generation of image guidance systems, the images that are displayed on the monitor generally show three planes which intersect at right angles at the tip of the pointer the surgeon holds in his or her hand. The surgeon must hold the pointer at the point of reference on or within the head, look away from the operative field to the monitor, and then, while looking away from the surgical field, reconstruct in his or her mind the three-dimensional virtual concept of the brain to determine the location of the tip of the pointer in relation to the head, and then look back to the surgical field. A fourth image on the monitor may be a rendered image of the head or slices or wedges through the head, which has the appearance of being three-dimensional, but is actually a two-dimensional display. How can a system be organized so the surgeon can see both the localizing information on the
monitor and the surgical field at the same time? There are two possible ways to show the localizing information and the surgical field on the same monitor – one can move the localizing information to a view of the surgical field (as with the Compass system or heads-up display through an operating microscope [1]), or move a real-time picture of the surgical field to the monitor, which is done routinely in endoscopic surgery, which does not have image guided localization of the surgical field. Present day endoscopes contain a video camera to display the surgical field on a monitor, which the surgeons looks at to guide the surgery. The surgeon can perform image guided surgery while looking at a similar monitor, which contains the needed localization information in addition to a real-time view of the surgical field. Just the video camera display of the endoscope is used in virtual reality surgery, and that is localized in space by the same image guided surgery presently used. A localized picture of a large surgical field is displayed, along with a virtual reality image of, for instance, the tumor that lies beneath the surface of the tissue, the blood vessels that may be of concern to the surgeon, other eloquent avoidance areas or structures, or anatomical structures that may serve as landmarks to the surgeon. Alternatively, the surgical field can be visualized on a video monitor that has the localizing information displayed in virtual reality superimposed on the real time view of the operative field. The surgeon sees a picture of the localizing information superimposed on a picture of the real anatomy, along with a real-time image of the surgical tools and tissue.
How does Virtual Reality Image Guidance Compare with Conventional Image Guidance? In conventional image guided brain tumor surgery, the images obtained by preoperative CT
Virtual reality in the operating room
or MRI scanning are reconstructed into a threedimensional image of the skull and its contents. The surgery ordinarily begins with the registration ritual, to indicate the location of the head in three-dimensional space, even before the surgical field is sterilized and draped. A system of applying sterile field localizing fiducials is mounted so localization reference information remains available within the sterile field. The entry point is localized with the pointer, using the three-dimensional localization information derived from the preoperative scan. The pointer is then set aside and the scalp is incised, the bone flap opened and the approach is designed. Often the pointer and all the localization information within the computer are not used again until final confirmation of the extent of resection, so that valuable and available localization information is underutilized or not used at all during the resection. At other times, the surgeon may use localization repeatedly as resection is done. Each time, the surgeon interrupts the surgery to set down the resection instrument, pick up the pointer, place the poinert at a point of interest, look back and forth between the surgical field and the monitor to determine how accurately he or she may have approximated the position of the pointer, decide how the surgery should proceed, put down the pointer, pick up the resection instrument, look back to the surgical field, and continue the surgery. The inefficiency of this exercise discourages the surgeon from using image guidance frequently and repeatedly throughout tumor resection. At the conclusion of resection, the surgeon may apply the pointer to the base of the resection cavity to estimate how complete the resection may have been. In virtual reality image guided brain tumor surgery, preoperative planning is done on previously obtained merged images, just as in image guided surgery. Any scanning modality that can be input is acceptable, and many times multiple modalities are merged to obtain the maximum
46
targeting information possible. CT and/or MRI may be used, depending on which shows the target best. Functional MRI can be used to detect eloquent areas to be avoided. Threedimensional angiography can be incorporated, or, if that is not available, major vessels can be detected on enhanced MRI studies. To use tumor resection as an example, the intended resection line around the tumor can be indicated automatically as the tumor surface, and then edited by the surgeon, or the surgeon can map the line of resection manually on each slice. The various structures can be color coded – for instance, green for the tumor, red for arteries, blue for venous structures, etc. The volume of the ventricle can be displayed, either for additional orientation, to plan the surgery to avoid entering the ventricle, or an approach to a mass within the ventricle. Eloquent areas and other avoidance structures can also be indicated. In virtual reality tumor surgery, the entire tumor volumein-space is addressed, rather than just the pointin-space that is used in conventional image guided surgery. In virtual reality surgery, the video camera has fiducials so the position and direction can be localized, so the video image is also registered to the head. The resection instrument also can be localized with fiducials mounted on the handle out of the grasp of the surgeon, so the computer can determine where the tip of the instrument is in relation to the tumor. The video camera is positioned stereotactically just out of the line of sight of the surgeon, either with a stereotactic head frame (> Figure 46-1) or with a frameless system, so the computer knows the exact location of the video camera in space. Consequently, it can accurately determine the location of the virtual image, which then matches with the real-time video image. The video camera shows a real time image of the surgical field. Repositioning the video camera allows the surgeon to see an alternate approach to deep structures, so superficial vascular structures can be avoided. Structures can be visualized as either
727
728
46
Virtual reality in the operating room
. Figure 46-1 The video camera is mounted on the CRW1 frame aimed at the surgical field. The camera can be raised or lowered to show just the size of the field on the monitor
a solid mass or just an outline of the surface of the mass, which for a tumor generally represents the line of resection. The scalp is visualized in the monitor. An image of the tumor within the brain is projected on the surface of the scalp, with perspective adjusted so the actual size is seen, along with other selected structures (> Figures 46-2 and > Figures 46-3a). Last minute adjustments for the approach can be made, for instance, to avoid vascular structures, which can be displayed along with the tumor, to use the best area of the scalp, or to obtain an ideal opening in the skull. With a sterile pen, the surgeon outlines the tumor on the scalp while looking at the monitor (Video A). This allows him or her to design the smallest scalp incision that will provide sufficient access, so brain shift is lessened and the surrounding cortex remains protected, so there is less chance for a neurological deficit from damage to the surrounding cortex, which has led to a perceptibly faster post-operative course. A smaller than conventional opening may be used, so most often a lazy-S incision rather than a flap is sufficient. The scalp is incised and retracted. The tumor is then projected on the bone flap, and
. Figure 46-2 The tumor is projected onto the scalp and outlined with a sterile pen just prior to planning the scalp incision. The small window in the corner shows a lateral projection of the tumor, and the size of the tumor at the slice indicated is shown on the scalp in the monitor
the bone opening is verified. The bone and dura are opened, just enough to facilitate the intended surgical approach, leaving the surrounding brain protected outside the exposure. After the brain is exposed, the monitor shows a live image of the surgical field and resection instruments. The surgeon operates by
Virtual reality in the operating room
46
. Figure 46-3 Selected slices during tumor resection. (a) The tumor projected on the scalp. (b) The tumor lies just beneath the cortex. (c) The tumor during resection, about half way through. (d) The tumor is gone from the display. The outline is the ventricle, which was purposefully avoided so intracavitary chemotherapy could be administered at the conclusion of resection. Video A: The tumor is projected on the scalp and outlined with a sterile pen prior to planning the incision. Video B: The cortex is incised to approach the tumor just beneath. Video C: At a deeper level, the edge of the tumor lies just lateral to the edge of the craniotomy opening
looking primarily at the monitor mounted just above the surgical field, not unlike endoscopic surgery, although the actual surgical field is available for direct vision at any time. Since the resection instrument is also localized stereotactically, it can be used just like the pointer in conventional image guidance. Thus, the relationship between the tip of the resection instrument and the tumor volume is always apparent to the computer, which is necessary if the surgeon wishes to have an audio signal showing where the tip is, as described below. The point where the tumor will first be met is projected on the image of the surgical field, and a tract to the tumor is made. If the tumor is on the surface, good correspondence between the virtual and actual image should be noted. If the tumor is deep to the surface, the entry point on the cortex is selected as the point where the tumor is closest to the surface (> Figure 46-3b, Video B). If an eloquent area or blood vessel overlies the tumor in that
approach, the best entry point can be visualized by moving the video camera so there is no obstruction to the approach to the tumor, although that is usually anticipated and corrected during the pre-surgical planning. During surgery, planes bearing the outline of slices of the tumor at right angles to the surgeon’s eye view can be displayed. The location of the displayed slice in relation to the depth of the tumor is shown graphically in a small window at the corner (> Figures 46-2 and > 46-3). As the tumor is approached, just the closest edge of the tumor is displayed to guide the surgeon most efficiently to the tumor beneath the surface (> Figure 46-3b). Resection can be done by resecting along the line that indicates the surface of the identifiable tumor at the level being resected or the pre-planned line of resection. As the surgery proceeds deeper, the resection instrument is used to select the outline at which depth to be displayed. Since the resection instrument is localized by image guidance, an audible
729
730
46
Virtual reality in the operating room
signal can also be used to indicate whether the tip is within or outside of the tumor or at the proper line of resection (> Figures 46-4 and > 46-5). The tract to the tumor is developed and two opposing self-retaining retractors used to maintain access. The depth of the tumor beneath the surface can be seen in the lateral projection window. The size of the tract can be smaller than the tumor diameter, because it is possible to work more laterally at deeper levels. When the tract reaches the tumor, the self-retaining retractor . Figure 46-4 The tumor monitor shows the outline of the tumor at the level being resected and the resection instrument as held in the field
. Figure 46-5 Tumor tissue is resected beneath the edge of the craniotomy opening
blades can be bent to afford a larger access at the depth of the tumor. When a centimeter or more depth of the resection line is mobilized completely around the tumor, the debulking can be carried down to that level (> Figure 46-3c, Video C). The virtual image is upgraded to show the next centimeter or more of the tumor and the process is repeated. Each time the virtual tumor outline is readjusted to the level being dissected. When the virtual outline of the tumor disappears from the monitor, the resection is complete (> Figure 46-3d). The video camera that captures the live image can localized also with conventional image guidance, which provided more flexibility in the localization of the camera and the ability to have more than one instrument or device localized, such as the resection instrument or ultrasound transducer, which make it possible to use an audio signal for additional guidance. The development of virtual reality image guidance began with the use of the Radionics X-Knife software and the CRW1 frame [2,3]. The volumetric target was constructed with the same protocol as used for a radiosurgery target. I originally termed the system the ‘‘Exoscope,’’ since it involved working with a video image, similar to endoscopic surgery, but with the endoscope mounted outside the body. When the endoscope was omitted, it became ‘‘Videotactic surgery.’’ When the stereotactic frame was omitted, it became ‘‘virtual reality image guided surgery.’’ The first artificial targets used to prove both the concept and the accuracy were five volumes, a sphere, two pyramids, and two cubes, each 2.5–3.0 cm maximum dimension. They were secured within a skull, which was secured into a CRW head ring. The N-shaped fiducial array was attached, and a CT scan was done. The identifiable targets were the tip of the pyramid, the closest point on the surface of the sphere, and each accessible corner of the cube. The volumetric display of each of the test objects were produced with the XKnife software that was
Virtual reality in the operating room
designed to produce volumes that were targets for stereotactic radiosurgery. The CRW arc was secured to the head ring, and the pointer was directed to each of the target points. The endoscope was attached to the probe holder on the arc and coordinates were adjusted to show each of the test objects in turn. The real time image was turned off so only the virtual image was seen on the monitor. The surgeon touched each of the target points guided by the image on the monitor, and the observer measured any discrepancy between the hand-held pointer and each target point. Measurements were taken with the arc at a variety of angles, simulating various surgical approaches. In all cases, the target points were approached with accuracy within two mm in all cases and a mean of 1.5 mm. Once the accuracy of Videotactic surgery was demonstrated in the laboratory, it was used in surgery, at first with targets that were clearly visible even without stereotactic guidance, and later to guide resection of gliomas with very irregular shapes. I have performed 74 craniotomies for brain tumors using this system of virtual reality guided surgery [3]. The first generation software required a CRW frame. In all cases, the postoperative scan revealed that the intended resection had been accomplished. There were no complications related to the use of virtual reality image guidance. The patients had a shorter post-operative recovery than in my prior craniotomies, and the length of hospital stay averaged one day shorter. There are significant benefits of using virtual reality image guidance to obtain optimal tumor resection. There is reason to believe from this pilot study that addressing the tumor as its entire mass with virtual reality image guidance, rather than a series of points-in-space, may lead to a better controlled and more complete resection. Dissecting along the pre-planned resection line at the first encounter of the tumor conceivably allows the
46
surgeon to remove more of a glioma, which, may provide an improvement in length of survival if more than 95% of tissue identified on MR scan has been resected [4–6]. Excision of solid tumors en bloc with virtual reality image guided surgery may prevent dural seeding of tumor that may cause a recurrence of the tumor [7]. Many neurosurgeons fail to take advantage of the full use of image guidance because of potential brain shift. Some brain shift invariably occurs when the dura is opened and resection and retraction are done. Even so, localization reference to the pre-operative study can often provide a better determination of the location of the tumor and anatomical avoidance structures than the surgeon can estimate without image guidance, and the surgeon can minimize this inaccuracy and accommodate for it. Several means of minimizing brain shift can be used, as outlined in Chapter D-15 [8]. (1) The opening should be at the highest point to minimize loss of cerebrospinal fluid and consequent settling of the brain within the cranial cavity. (2) Aspiration of cerebrospinal fluid should be minimized for the same reason. (3) Retraction should be applied symmetrically, so as not to push the brain in one direction or another. (4) When the tumor is encountered, the plane at the edge of the tumor should be developed before gutting or decompressing the tumor itself, so that the plane of dissection is established while the brain or tumor shift is at the minimum, just the opposite as indicated in most non-guided surgery. A number of techniques are available to identify brain shift so the pre-operative scan can be updated during surgery. Most involve taking a CT or MRI during surgery. This is helpful not only to localize the tumor, but also to estimate whether the resection has been complete or whether additional resection is indicated. However, most operating rooms do not have that capability. Intraoperative ultrasound has been used to identify the target mass prior to dissecting the
731
732
46
Virtual reality in the operating room
approach to it [9]. Most often the ultrasound image is used without integrating the imaging data with pre-operative studies, the same as it would be in non-imaged procedures. There have been attempts to localize the ultrasound transducer stereotactically with the same type of fiducials used on the image-guided pointer, and this possibility lends itself readily to the virtual reality system described herein, so that the ultrasound image is also localized to the same stereotactic space and may be merged with pre-operative images, that may be morphed into the dimensions of the intraoperative ultrasound. The quality and information content of three-dimensional intraoperative ultrasound is improving rapidly, and may one day soon become the modality of choice to correct for brain shift. Endoscopic surgery: Other present systems rely on the surgeon operating from a video image. Endoscopic surgery could be improved significantly by registering the endoscope in space, using the same technology as described herein. Once the endoscope were localized, the field of view of the endoscope could be determined, and the surface that first lies in the field of view could be used with the virtual reality targeting information. In the configuration that shows the view of the endoscope on the monitor, localizing information can be superimposed on the view of the surgery, which would provide three-dimensional localization to the endoscope. DaVinci robotic surgery is performed while looking at a stereoscopic high definition image of the surgical field as seen by two video cameras attached to an endoscope. The surgeon looks with each eye at a stereo monitor at a location within the operating room away from the patient. The surgeon not only has an excellent stereoscopic view of the surgical field, but has access to hand and foot controls to guide two or three tools introduced through access ports. This is somewhat akin to operating with a surgical microscope, which provides excellent threedimensional images of the surgical field. If the
microscope is image guided, the target volume can be localized, and a localizing view of the circumference of the target can be superimposed on the view through the microscope with a heads up display. However, the microscope field of view may not encompass the entire circumference of the target tumor, or even a portion of the surface, which makes a microscopic display of a large tumor problematic. Since much tumor surgery does not require a microscope, the use of image guided microscopy may not be appropriate for many tumors. Audiotactic surgery: One advantage to the virtual images containing both the localization of the anatomy and the location of the surgical resection instrument is that auditory signals can be used to indicate the position of the tip of the instrument in relation to the intended resection line. For instance, a steady tone can be used if the instrument is within the tumor boundary and an interrupted tone if the instrument lies outside the tumor. The pitch of the tone can increase as the resection line is approached, so the highest pitched continuous tone indicates that the instrument is along the intended line of resection. In addition to tones, computer generated words can be used, just as in GPS automobile navigation, so the computer can warn of a ‘‘blood vessel on the right in 5 mm.’’ For training purposes, the attending surgeon can program all the admonitions he or she would tell the resident as the surgery proceeds. The pedicle of a vertebra can be drilled, with the surgeon looking at the operative field, rather than the guidance system, all the time directed by an audible signal. Thus, tumor surgery can be facilitated by adding a video camera, localized with image guidance and looking at the surgical field. Virtual images of the tumor and relevant anatomy can be graphically integrated with this video image, mostly using software already present in most image guided systems. By localizing the resection instrument to the same image guided space, the relationship of the instrument to the tumor
Virtual reality in the operating room
provides information that can be transmitted to the surgeon with an audible signal. Disclosure: The author holds patents on Videotactic, Audiotactic, and image guided endoscopic technology.
References 1. Rousu JS, Kohls PE, Kall B, Kelly PJ. Computer-assisted image-guided surgery using the Regulus Navigator. Stud Health Technol Inform 1998;50:103-9. 2. Gildenberg PL, Labuz J. Stereotactic craniotomy with the exoscope. Stereotact Funct Neurosurg 1997;68:64-71. 3. Gildenberg PL, Labuz J. Use of a volumetric target for image-guided surgery. Neurosurgery 2006;59:651-9. 4. Hentschel SJ, Sawaya R. Optimizing outcomes with maximal surgical resection of malignant gliomas. Cancer Control 2003;10:109-14.
46
5. Sawaya R. Neurosurgery issues in oncology. Curr Opin Oncol 1991;3:459-66. 6. Sawaya R, Hammoud M, Schoppa D, Hess KR, Wu SZ, Shi WM, Wildrick DM. Neurosurgical outcomes in a modern series of 400 craniotomies for treatment of parenchymal tumors. Neurosurgery 1998;42:1044-55. 7. Suki D, Abouassi H, Patel AJ, Sawaya R, Weinberg JS, Groves MD. Comparative risk of leptomeningeal disease after resection or stereotactic radiosurgery for solid tumor metastasis to the posterior fossa. J Neurosurg 2008;108:248-57. 8. Kelly PJ. Tumor stereotaxis. Philadelphia, PA: Saunders; 1991. 9. Hernes TA, Ommedal S, Lie T, Lindseth F, Lango T, Unsgaard G. Stereoscopic navigation-controlled display of preoperative MRI and intraoperative 3D ultrasound in planning and guidance of neurosurgery: new technology for minimally invasive image-guided surgery approaches. Minim Invasive Neurosurg 2003;46:129-37.
733
68 Cyberknife: Clinical Aspects F. C. Henderson Sr . W. Jean . N. Nasr . G. Gagnon
CyberKnife Stereotactic Radiosurgery Treatment of neoplasms of the spinal column and spinal cord can be most challenging. Spinal tumors are usually large, and surrounded by radio-sensitive structures – the spinal cord, heart, kidneys, airway and gastrointestinal system – the increased radio-sensitivities of which allows for a very small margin of error. Furthermore, the mobility and potential instability of the spine can introduce motion and potential difficulties with positioning during treatment. Malignancies of the spine are common. Of the 560,000 cancer related deaths each year in the USA, most result from metastatic disease [1]. The bone is the third most common metastatic site, after lung and liver [2], of which the spine is the most common target. The estimate of 100,000 new spinal metastases in the United States each year is probably low; the number of cases is far larger if one considers multiple lesions in the same patient. Bone pain is the most common pain syndrome requiring treatment in cancer patients, and is typically more severe than that due to visceral metastases. Patients with bone metastases survive longer, and become symptomatic earlier than those with visceral metastases. Furthermore, complications arise in 1/3 patients with bone metastases, and are usually associated with significant morbidity [1]. Treatment options for the spine are somewhat limited, because of its functional requirements as well as the significant internal or adjoining critical structures. The spine has several functions which it must perform in the healthy state – structural support, weight distribution, a basis for attachment of the appendicular skeleton, #
Springer-Verlag Berlin/Heidelberg 2009
flexibility, protection of the spinal cord and nerve roots, hematopoeisis and mineral storage. In the diseased state, these functions are compromised, resulting in significant pain, functional impairment and deteriorating quality of life. Therapy is based upon the length of anticipated survival. Survival depends primarily on histology. Median survivals are 29.3 months for metastatic prostate cancer, 22.6 months for breast, 11.8 months for renal cell, and 3.6 months for lung cancer [3]. Survival is also a function of the number and location of other sites [4,5,29], performance status and a host of other parameters. Surgery is indicated primarily for decompression of neurological structures, stabilization of the spine and for severe, pain resulting from ‘‘micro-instability’’ – weakened structures that cause severe pain with the normal activities of daily living. Secondary indications for surgery include need for biopsy, the occasions in which resection of a solitary lesion may effect cure, or in those cases of recurrent lesion when all other treatment modalities have been exhausted. Stereotactic irradiation may be an effective adjunct to surgery. When severe pain, instability or neurological compromise is not present, the goals of stereotactic irradiation are to maximize tumor control, and where possible to achieve tumor ablation, while maintaining quality of life. This is accomplished by maximizing radiobiological efficacy and minimizing complications due to radiation toxicity.
What is the CyberKnife? The CyberKnife is a stereotactic radiosurgery system with a 6-MV X-band linear accelerator
1112
68
Cyberknife: clinical aspects
(LINAC) capable of delivering 600 monitor units (MU)/min. The robot delivers collimated beams 5–60 mm in diameter at target. The LINAC is mounted on a fully articulated robotic arm capable of six degrees of freedom allowing full rotational and translational movements (> Figure 68-1). Real time image-guidance is provided by two ceilingmounted diagnostic X-ray tubes with corresponding orthogonal floor-mounted amorphous silicon detectors which periodically acquire images during treatment to track bony skull, spinal anatomy or implanted metal fiducials. The beam is dynamically brought into alignment during treatment to account for patient/ target movement. Non-isocentric treatment allows the delivery of highly conformal and homogenous radiation doses to complex target volumes. The Gamma Knife uses isocentric planning, wherein the radiation beams intersect in a sphere, creating spherical dose distributions which are stacked adjacently to cover the lesion. The CyberKnife uses a linear algorithm to ‘‘paint out’’ any complex shape (> Figure 68-2). Steep dose gradients limit the radiation dose to surrounding normal
structures. Multiple treatment paths are available using approximately 110 nodes in space (robot positions) and 12 different pointing directions from each node, providing approximately 1,320 beam directions for selection. Flexibility to move with six degrees of freedom, end-to-end accuracy of less than 1 mm, non-isocentric targeting, and real-time image guidance allow the robot to accommodate moment-to-moment changes in patient position. The CyberKnife is particularly well-suited for treatment of spinal tumors [6–10].
Radiobiological Considerations Direct and Indirect Effects of X-irradiation The primary site of injury following irradiation appears to be the nucleus, and there is overwhelming evidence that DNA is the principle target of irradiation. X-irradiation causes breaks in chromosomes. The broken ends of the chromosomes may then rejoin in their normal position, may join
. Figure 68-1 The CyberKnife is a miniature linear accelerator (a) attached to a robot that moves with 6 degrees of freedom around the patient. The X-irradiation is collimated to beams ranging from 6 to 50 mm at the target. The patient position is adjusted in real-time to accommodate patient movement
Cyberknife: clinical aspects
68
. Figure 68-2 The difference between CyberKnife and Gamma Knife. (a) Gamma Knife uses converging beams to create spherical dose or distribution at selected isocenters. This ‘‘isocentric’’ planning may result in hot spots in regions of overlap and cold spots between the edges of the isocenters. (b) CSRS uses a linear algorithm that selects any of 1,300 beams to ‘‘paint out’’ the lesion; CyberKnife also uses an inverse treatment planning to avoid critical structures in a forward planning mode. While hot spots also occur in CK/SRS, the linear algorithm is more useful for complex and non spherical shapes
to other broken chromosomes, or may fail to rejoin; the results are chromatid aberrations (before telophase) or chromosomal aberrations (after telophase). Chromosomal injury may be sublethal, potentially lethal or lethal. Sub-lethal cell injury is potentially reparable, but may accumulate with other sub-lethal injuries with further irradiation to become lethal injury. Repair of sub-lethal chromosomal injury is the repair of double strand breaks, and usually occurs within 5 h of irradiation, between radiation fractions [1]. Sub-lethal chromosomal repair is dependent upon cell metabolism, oxygen and nutrients. Potentially lethal injury can be modified in the post-irradiation environment: inhibition of mitosis can allow chromosomal repair. Lethal injury is unredeemable. Most single stranded DNA injuries produced by irradiation are repaired, the opposite DNA strand serving as the template. Double strand breaks which are separated along the chromosome can also be repaired in the same manner. When the double strand breaks occur close together, a ‘‘mitotic cell death’’ usually ensues. Whilst many aberrations of the DNA stand are possible, there are three which are lethal combinations: the ring, the dicentric and the anaphase bridge [11]. Tumor cells have a broad range of radiosensitivity – the more radiosensitive cells
manifesting relatively more apoptotic changes, and the more radio-resistant tumors almost exclusively mitotic cell death. While cell death may be the result of induced apoptosis, most cell death in tumors is the result of mitotic cell death. Cells die when they attempt mitosis with injured chromosomes. Cell kill is the result of the abnormal recombination of chromosomes resulting from two double-strand breaks and the formation of lethal chromosomal aberrations. An important factor in fractionated irradiation is tumor repopulation. Tumor doubling time is approximately 28 days for prostate cancer, 14 days for breast cancer, and 4 days for head and neck cancers. Tumor cell repopulation, though an important factor when considering multifraction regimens with rapidly growing tumors, is not an important factor when considering stereotactic irradiation in one to five fractions.
Early and Late Effects of Irradiation The goal with staged radiosurgery is to maximize lethal cell kill and to minimize lethal injury to normal tissues. The effects of irradiation should be considered in terms of early and late effects. Early effects include injury to most neoplastic
1113
1114
68
Cyberknife: clinical aspects
cells and to normal tissues like skin, hair follicles and esophagus; they occur within a month of treatment. Late effects include peritumoral brain necrosis and transverse myelitis, and are due to loss of functional parenchymal cells; the delay, due to slow parenchymal cell turnover, arises 6 months to 5 years after treatment. The magnitude of early effects is related to the number of fractions and to the overall time over which they are administered. The magnitude of Late effects on the other hand is determined by the size of the individual fractions. In general, therefore, multi-session treatments will enhances early effects (tumor cell kill), and decreases late effects(CNS injury). Radiation in general is complicated by the presence of significant dose-limiting structures within and surrounding the spine: the spinal cord, spinal nerve roots, peripheral nerves, esophagus, bowel, kidneys, lung, and heart. The dose tolerances of central nervous critical structures have been determined through bitter experience: brain necrosis occurs with >4,000 cGy in 10 fx; optic myelitis with a single fraction >800 cGy; radiation myelitis in thoracic cord with 3,500 cGy in 10 fx administered over 12 cm of spinal cord length. With standard fractionation, there is a <1% risk of late effects (myelitis) with <5,040 cGy, 5% risk with 5,400 cGy and 50% risk with 70 Gy to the cord. Late effects are due micro-vascular thrombosis, endothelial injury, necrosis, progressive glial atrophy, oligodendrocytic injury and demyelination [12]. Late CNS effects are increased in certain conditions: hyper-fractionation (where more than one fraction is administered in 1 day), irradiation of children, certain chemotherapeutic regimens including (cytoxan, methotrexate, vincristin or araC), or intrathecal drug administration of drugs of methotrexate. Lymphoma and alcohol are also thought to increase the risk of radiation [13]. The dose tolerance of the esophagus is approximately 50 Gy in standard fractionation, or 25 Gy for a single dose administered to over
6 cm length of esophagus. The dose tolerance for the larynx is 60 Gy, or 30 Gy if given as a single dose; for liver, kidney the dose tolerance is 45 Gy where the irradiation covers 50% of volume of those structures. Prior radiation limits the dose that can be given. Although several tissues have the capacity for recovery with time, this recovery is not complete and is variable among tissue types. For example, spinal cord and lung have some capacity for injury recovery, but mesenchymal tissue in general recovers poorly from prior radiation therapy, and some tissue sites, such as the intestine, bladder and kidney have no capacity for long-term recovery [14]. Radiation risk is balanced against tumor control probability (TCP). TCP follows a sigmoidal response to total dose; unfortunately, normal tissue complication probability (NTCP) follows a similar sigmoidal response to total dose, limiting the ability to deliver an effective dose safely. For instance, the dose administered in regimen B (> Figure 68-3) is associated with an almost zero normal tissue complication rate, but is only 75% effective in achieving tumor control; on the other hand, the 95% effectiveness of treatment B comes with a risk to normal tissue of 10%, or TD10. A major difficulty in re-irradiation is the determination of the normal tissue tolerance and the avoidance of toxicity by its excess. Recurrent tumors are relatively more radio-resistant, if only by a process of selection; more sensitive tumors are controlled by the previous irradiation. Cell lines from recurrent squamous cell carcinomas from human head and neck tumors are more resistant to invitro radiation than those from tumors not previously irradiated. Furthermore, the process of radiation may stimulate increased mitosis in some tumors [5]. The authors argue that the use of extremely conformal irradiation and large doses will achieve a therapeutic gain on spinal recurrences. Careful contouring and fractionation, or ‘‘staging,’’ will usually allow a sufficient dose to the tumor.
Cyberknife: clinical aspects
. Figure 68-3 Radiation risk is balanced against tumor control probability (TCP). TCP follows a sigmoidal response to total dose; normal tissue complication probability (NTCP) follows a similar sigmoidal response to total dose, limiting the ability to deliver an effective dose safely. For instance, the dose administered in regimen B is associated with an almost zero normal tissue complication rate, but is only 75 % effective in achieving tumor control; on the other hand, the dose administered in treatment C is 95% effective in tumor control, but comes with a risk to normal tissue of 10%, or TD10
The Alpha/Beta Ratio The radiation beam induces a lethal injury to tumor cell DNA through direct injury by photons, the ‘‘direct effect,’’ and a substantially larger DNA injury through the ionizing effects of radiation on oxygen atoms, the ‘‘indirect effect.’’ Cell kill occurs in a linear relationship to dose, and also related to the quadratic function of the dose. It is thought that the linear portion of the mitotic death is related to double stranded DNA injury due to a single electron, and is referred to as the a cell kill. Where two separate charged particles are involved in the DNA injury, the statistical probability relates to square (quadratic function) of dose, and is referred to as the b cell kill. The ratio of the linear cell kill to the quadratic cell kill is called the Alpha / Beta ratio (> Figure 68-4). The a/b ratio, is that dose in Gray, where linear (a) cell kill equals quadratic (b) cell kill. The a/b ratio reflects
68
. Figure 68-4 The negative logarithmic curve of tumor cell survival, showing the upper linear portion – the alpha cell kill – and the lower curved portion of the curve representing the quadratic beta cell kill. The a/b ratio is that dose in Gray where the a cell kill is equal to the b component of cell kill. The a/b ratio is low for the CNS and tumors such as chordoma, and high for most adenocarcinomas
the susceptibility of a given tissue to injury from irradiation to a hypofractionated regimen versus standard fractionation: neuroblastoma, most lung cancers and skin have a high a cell kill, and therefore, high a/b ratios, approximating 10 Gy, whereas meningioma, chordoma, prostate cancer, and normal central nervous system are considered ‘‘slow reacting,’’ have a low a cell kill, and therefore a low a/b ratio, less than 3. The CNS is thought to have an a/b ratio of 2.2 Gy. In general, radiation sensitive tumors, such as neuroblastoma are sensitive to irradiation, and rapidly respond. Slow reacting cells, such as the CNS, respond more slowly. The a/b ratio offers a method to compare fractions of different size. The Linear quadratic Formalism, described by Fowler, allows comparison of the relative biological equivalent dose (BED) of one fractionation regimen with another [3].
The Linear Quadratic Formulation Biological effect E of a given dose D is given by the cell kill due to the linear component a D + cell kill due the quadratic component bD2.
1115
1116
68
Cyberknife: clinical aspects
Therefore, E ¼ aD þ bD2 This rearranges to BED ¼ ndð1 þ d=a=bÞ [3]:
[TV] (> Figure 68-5). The New conformity index takes into account the location of the prescription volume with respect to the target volume by dividing the ratio of [PV] to [TVPV] by the coverage. New conormity INDEX
Mathematically, the formulation begins to break down for hypo-fractionation (<5 fractions). Nevertheless, in the absence of an alternative scheme, we have used the Linear Quadratic formulation successfully to design tumor treatments to avoid ‘‘late effects’’ from damaged CNS. Accordingly, the linear Quadratic calculations will be presented with each example. It must be remembered that the calculated BED is not a real number describing the dose of irradiation to the CNS or tumor, but rather a relative dose for comparing one regimen to another. Thus, from the literature we know that a dose of 54 Gy in standard fractionation to the spinal cord carries a 1% risk of ‘‘delayed effects,’’ or injury to the spinal cord. Using the Linear Quadratic formulation: BED ¼ ndð1 þ dX =a=bÞ; therefore for n ¼ 27 fractions; d ¼ 2Gy; ðtotal 54GyÞ; ¼ 54 Gyð1 þ 2Gy=2:2GyÞ ¼ 103 Gy which would represent the BED at which you begin to expect complications from irradiation to the spinal cord. Thus, any hypofractionation regimen can be compared to this relative quantity, to ensure a low risk to the spinal cord. These calculations will be used in the illustrative examples. Conformality
Three parameters have been reported to evaluate treatment plans: coverage, conformity index, and new conformity index. Coverage of the tumor can be defined by obtaining the ratio of the target volume within the prescribed isodose surface [TVPV] to the total target volume [TV]. Conformity index is calculated as the ratio of the prescription volume [PV], to the target volume,
¼
PV=TVPV TV PV ¼ TVPV2 TVPV =TV
A perfect plan would have [TVPV] = [TV] = [PV], yielding a New Conformity Index = 1.0 as well as a conformity index of 1.0 [15,16].
Rationale for Multi-Sessioned Treatment (Staging, or Fractionation) Clinical and invitro studies have clearly demonstrated the efficacy of single fraction irradiation [2] Gamma Knife [17] and Linac [18,19] have employed single dose treatments in the brain with efficacy and low morbidity. Furthermore, it is well established in tissue cultures, that high doses of radiation administered over one or a few . Figure 68-5 Conformality (the Conformity Index) is the ratio of prescription volume (i.e., the isodose volume) (PV) to target volume, or PV/TV. When the prescribed isodose volume equals the target volume, the conformality is 1.0
Cyberknife: clinical aspects
staged treatments are the most effective in achieving cell kill [2]. Single fraction treatments may be appropriate for lesions with low a a/b ratio (AVMs), and small spherical lesions. However, to at least some extent, single fractionation is driven by the necessity of placing a rigid frame on the patient; staging treatments (fractionating) radiosurgery with a rigid frame is excessively burdensome to both patient and treating physician. The CyberKnife is not constrained by the presence of an external frame. The CyberKnife utilizes real-time radiographs and the X-sight1 algorithm to register precise location of the lesion based upon the radiographic anatomy, minute by minute. The radiographic images are compared to the digitally reconstructed images stored within the treatment plan. Comparison of these images allows detection of any patient movement. In turn, this movement is registered, and automatically accommodated by the ‘‘adaptive beam pointing’’ algorithm; the latter makes corrections for beam angle to accommodate up to 1 cm of patient movement (>1 cm triggers an e-stop). Exact repositioning of the patient with each session is thus not required, and multisession treatments pose no difficulty [10]. The beam to target accuracy has been defined at this institution and others to be 0.5–0.7 mm.
Rationale for Multi-Session Treatment Recognizing that the radiobiology of high dose, single fraction irradiation remains unclear and falls outside of the domain of present radiobiological theory, we have chosen use multi-session treatments (fractionation) to achieve greater conformity and safety through the temporal partitioning of dose and exploitation of radiobiological differences between target and normal tissue. Our treatment strategy, to minimize late effects, and maximize tumor control probability, is based on seven principles. First, the low a/b ratio of normal brain and spinal cord predicts that for a given dose to
68
tumor, the adjacent CNS tissue will see a higher biological equivalent dose (BED); conversely, administering the total dose in 3–5 fractions, will decrease the BED to the CNS. Thus, the inherent difference of a/b ratio between tumor and CNS can theoretically be exploited with staging of treatment (administering in multi-sessions). Second, normal CNS tissue more faithfully undergoes repair of DNA injury than does tumor. Third, fractionated irradiation theoretically results in improved cell kill through ‘‘cell reassortment’’: cells passing through the cell cycle tend to hold up in the late G2 and M phase as a result of DNA injury, and the resulting expression of check-point genes which prevent spindle formation until DNA repair is completed; it is in these stages – M and late G2 – that tumor cells are most sensitive to irradiation. Fourth, to some extent, irradiation results in opening up of arterioles and increased oxygenation of the tumor, which imparts greater radio-sensitivity after the first dose. Fifth, small errors introduced into the treatment will tend to average out with fractionation. Sixth, treatment times are faster; and seventh, early side effects are less. The general dose strategies at GUH are shown > in Table 68-1. Generally, we treat the clinical treatment volume (CTV), which includes the gross tumor volume (GTV) plus the surrounding margin of tissue at risk for microscopic disease. The average volume of tissue irradiated is 60 cm3, with a range of 1–850 cm3. The tumors occupy all levels of the spine with a proclivity to the cervicothoracic and thoracolumbar junctions.
Neuro-Surgical Considerations The neurosurgeon is an integral member of the radio-surgical team. The surgeon is responsible for determination of which patients are reasonable candidates for surgery as opposed to radiation, assessment of neurological and spinal stability, and availability if the patient deteriorates during Radiosurgical treatment. The neurosurgeon probably best understands the urgency of the spinal
1117
1118
68
Cyberknife: clinical aspects
lesion, and should have input into the planning and timing of radiosurgery. The contouring should be performed by the neurosurgeon, who best understands the anatomy, likely regions of spread, and location of critical structures such as the spinal cord within the spinal canal. As the surgeon steps into the realm of radiation delivery, he assumes a new responsibility for following the patient after irradiation. Though initially engaged in treatment of a central nervous system cancer, the neurosurgeon, en passant, will inevitably encounter other lesions and become involved in treatment of non-CNS, medical problems that arise.
Illustrative Cases Case #1 Eighty-eight-year-old Caucasian female presented with a 1 year history of increased difficulty walking. . Table 68-1 Dose/fraction (cGy)
# fractions
800
3
800
5
700
3
500
5
Examples of use Untreated spine: (met. breast, thyroid, colon, renal, bladder, melanoma, non small cell lung, small cell lung) Chordoma, chondrosarcoma, osteogenic sarcoma Retreatment of spinal lesions after external beam (met. renal, nsc lung, sc lung, breast, giant cell, melanoma, colon, cervix, bladder) Benign spinal lesions (neuroma, schwannoma, meningioma, hemangioma, ependymoma). Gross disease of less resistant histology (leukemia, lymphoma)
She developed progressive bilateral lower extremity weakness, foot drop, and sensory loss and eventually became wheel-chair bound. She also reported mid-lower back pain. At that time, she also suffered an acute myocardial infarction. MRI imaging of the spine on 12/03 revealed a 1.5 1.3 1.4 cm diffusely enhancing intradural, extramedullary mass at the level of T10–11. There was evidence of adjacent dural enhancement, and imaging characteristics were consistent with spinal meningioma (> Figure 68-6). The lesion was located posteriorly in the spinal canal with evidence of significant, but chronic, cord compression. On examination, she reported poor quality of life and a pain level of 90/100. She was unable to stand; strength was 2/5 in both lower extremities and an incomplete sensory level to pinprick was evident at L1. Steroids were initiated. Given her history of severe cardiac disease s/p CABG, recent MI, and severe peripheral artery disease, she was not felt to be an acceptable candidate for surgical resection. Tracking fiducials were placed percutaneously. . Figure 68-6 Sagittal MRI of the thoracic spine (T2 Weighted, providing a myelogram effect) shows the spinal cord severely compressed by an intradural, extra-medullary dural based tumor at the T11 level. The patient was in severe pain and densely paraparetic
Cyberknife: clinical aspects
The T10 meningioma was treated with the CyberKnife with a dose regimen of 2,500 cGy (5 sessions of 500 cGy) prescribed to the 80% isodose line (> Figure 68-7). Treatment was completed over the course of 7 days, and was extremely
68
well-tolerated without acute toxicity. Her VAS pain score improved dramatically from 90 prior to treatment to 10/100 at 1 month after treatment, and she was able to stand and take a few steps. She remains pain free at 3-year follow up.
. Figure 68-7 Treatment plan for T11 meningioma in the Illustrative Case #1.Clockwise from top left: the axial view through T11 shows that the tumor (red line with yellow dots) is almost completely covered by the 80% IDL (orange line); the sagittal view of the reconstructed CT shows the contour in this case occupying most of the spinal canal; the coronal view of the reconstructed CT showing that the tumor contour occupies most of the spinal canal: the various beam trajectories of the CyberKnife
1119
1120
68
Cyberknife: clinical aspects
Her SF-12 scores improved from 20 (MCS) and 22 (PCS) before treatment to 37 (MCS) and 31 (PCS) at 1 month after treatment and 57 (MCS) and 37 (PCS) at 2-year follow up. She is now neurologically intact and her ability to walk has returned to normal.
Case #2 A 27-year-old woman was diagnosed with Ewing’s sarcoma of the right T7 rib in the paraspinal region. She underwent conventional, external beam irradiation encompassing the rib and right side of the spine at the T6, 7, 8 levels in a foreign country. The patient did well for 2 years before suffering a recurrence in the spine at T7 (> Figure 68-8). She was retreated with external beam from T2 to T10 with 10 300 cGy. Approximately 1 year later, she presented to the emergency room at GU with severe pain (10/10) and progressive spastic paraparesis. As she had already undergone irradiation . Figure 68-8 Sagittal MRI (T1 weighted, contrast enhanced), showing recurrent Ewing’s sarcoma, previously irradiated. There is spinal cord compression, maximal at T8, but extending within the posterior longitudinal ligament from T7 to T9
beyond the tolerance dose for a 5% risk of myelitis, further irradiation was considered to be unsafe. Therefore, a surgical plan was entertained. Subsequent spinal angiography identified the Artery of Adamkiewicz at the T7 level, raising concerns about the safety of vertebrectomy to remove the recurrent tumor. Therefore, the patient underwent a posterior decompression, bilateral resection of the posterior and lateral structures (the facets and pedicles) and stabilization. Fiducials were placed at the same time, to allow tumor localization for CK/SRS. Even though the patient had been irradiated twice, and had received well over what was considered to be a safe radiation dose, her severe neurological condition militated for SRS. The surgeon performed the contouring of the tumor and critical structures, including the spinal cord (> Figure 68-9b). The treatment plan (> Figure 68-9a) was generated. An ideal plan would be 4 8 Gy to the sarcoma and margin. In this case, however, a lower dose was elected to lessen the risk of spinal cord necrosis. CyberKnife SRS was administered over 5 days (regimen of 5 450 cGy for a total dose of 22.5 Gy to the 70% IDL). Treatment was completed without complication. At 2 months there was no neurological deficit, no pain, and the patient had returned to work as a computer engineer. At 6 months, there was no evidence of the spinal sarcoma on MRI (> Figure 68-10) and PET. Unfortunately, the tumor recurred several months later, at 9 months, and was associated with multiple metastases. She died 16 months after CK/SRS treatment.
Results of CK/SRS Treatment at GUH Patient Selection Georgetown University Hospital acquired the CyberKnife system in 2002 and immediately applied the technology to treatment of spinal tumors. Each patient was prospectively evaluated by a multidisciplinary team. Two hundred patients with a
Cyberknife: clinical aspects
68
. Figure 68-9 (a) The CyberKnife treatment plan demonstrating the surgeon’s contour of the recurrent sarcoma tumor wrapped around the spinal cord, extending from the T6 toT8 level. The spinal cord (25% of the volume) received less than 1,600 cGy. (b) The radio-surgical plan of Illustrative Case#2. CT scan at the T8 level showing the X-ray beams wrapping around the spinal cord to minimize cord toxicity. Clockwise from top left: the treatment plan in axial view, sagittal view, beam entry pattern with tumor shown in red and critical structure (spinal cord) in green; the coronal view (lower left) of CT. Approximately 500 beams were used in this very complex plan to maximize conformity and minimize cord toxicity to the CNS
1121
1122
68
Cyberknife: clinical aspects
. Figure 68-10 Sagittal MRI (T2 weighted view) of the thoracic spine showing remission of the sarcoma 6 months later. The high signal of the vertebrae indicates previous irradiation as well as the recent CyberKnife SRS; the high signal reflects increased fatty replacement of the hematopoietic region (cancellous portion) of the vertebrae
total of 274 spinal tumor sites with both primary and metastatic tumors of the spine who met eligibility criteria for CyberKnife treatment were consecutively enrolled, and treated with the CyberKnife. Forty nine patients presented with primary tumors of the spine and 151 patients with metastatic disease. All spinal levels were treated, although thoracic levels predominated. No patients were excluded on the basis of pathologic subtype. The most common tumor types treated were breast cancer, non-small cell lung cancer, and sarcoma. Of the 274 tumor sites, 125 had received prior conventional irradiation to the site that was to undergo SRS. CyberKnife SRS was used in the initial management in 118, or following surgery in 19.
Pain Seventy six precent of patients reported some degree of pain prior to treatment, the mean
pain score being 40/100 with medication. Most patients were managed with narcotic analgesics and NSAIDS. A paired t-test was used to analyze change in pain. Statistically significant overall pain relief was seen at 1 month (p < 0.001), 12 months (p = 0.003), 24 months: (p = 0.001), 36 months (p = 0.0003) (> Figure 68-11).
Quality of Life Patients were followed with quality of life surveys. The mean physical component of quality of life remained stable, regardless of whether patients were treated for benign disease or malignant disease. Scores for the mental component of quality of life improved significantly (p.01). Mean PCS and MCS scores of patients treated with CyberKnife as initial management were not significantly different to those undergoing retreatment after previous radiotherapy (> Figure 68-12a,b).
Cyberknife: clinical aspects
68
. Figure 68-11 A durable and statistically significant overall pain relief was seen at 1 month (p < 0.001), 12 months (p = 0.003), 24 months: (p = 0.001), 36 months (p = 0.0003)
. Figure 68-12 (a) The mean physical component of quality of life showed a statistically significant improvement (p. 01). The improvement occurred and was similar in magnitude in both benign and malignant disease trended upward, but did not reach statistical significance. (b) Scores for the mental component of quality of life improved significantly (p. 01)
1123
1124
68
Cyberknife: clinical aspects
Survival Overall, the Kaplan Meier curve for those patients undergoing CK/SRS as a first modality enjoyed a survival of 15 months (> Figure 68-13). Those treated with CK/SRS as a salvage modality survived a mean of 10 additional months.
the center of the treatment field, there was no evidence of late effects or treatment related myelitis.
Treatment of Intracranial Lesions with CyberKnife Brain Metastases
Complications Most patients experience minimal or no side effects. Most acute complications were selflimited and mild. The most commonly reports toxicities were fatigue, nausea, esophagitis, dysphagia, and transient diarrhea. One patient developed breakdown at a surgical site requiring surgery for debridement and re-closure of the wound, one patient developed an intraperitoneal infection, and two patients who developed vertebral fractures in the irradiated spine. One of these had been previously irradiated with external beam. With exception of minor radicular symptoms from nerve roots passing through . Figure 68-13 Median survival for all metastatic lesions was 15 months
Although the development of brain metastases is often viewed as an end-stage event of cancer, modern aggressive management of brain metastases has significantly improved survival and functional outcome compared to statistics oftcited in the literature. The majority of patients with controlled intracranial metastases will die from systemic disease rather than from recurrence and progression of these metastases. Surgical resection was considered gold standard treatment of single metastasis to non-eloquent brain areas. However, physicians and patients alike are now choosing more often to treat these tumors with radiosurgery. Patients with cancer have usually gone through many physically taxing treatments by the time they develop brain metastasis, and the non-invasive nature of radiosurgery provides significant comfort and reassurance. Radiosurgery is the first-line treatment of brain metastases smaller than 4 cm in diameter in eloquent areas, especially when edema is minimal. The addition of whole-brain radiotherapy (WBRT) decreases tumor recurrence rate, but has not been correlated with better survival. Radiosurgery seems to be effective in controlling tumors that are so-called ‘‘radioresistant,’’ such as melanoma, renal call carcinoma and sarcoma, for which WBRT is rather ineffective. However, radiosurgery is seldom effective against cystic brain metastasis, and surgical treatment must be considered for these if they progress in size. Radiosurgery utilized as a boost to the tumor region after WBRT has also been shown to reduce the risk of tumor recurrence or progression.
Cyberknife: clinical aspects
Nishizaki et al. [20] recently reported results from CyberKnife treatment of multiple or large brain metastases. At 44 weeks median follow up, the local tumor control rate was 83%. No patients died from their intracranial disease, even though 21 patients were treated with additional CyberKnife sessions for new metastases during their follow up. The authors concluded that fractionated treatment with the CyberKnife is highly useful for controlling large brain metastases in patients with advanced cancer. Our dose-regimens are as follows: 1. 2. 3.
4.
Boost after WBRT: 14–18 Gy in single session For a single met (<3 cm): 18–24 Gy in a single session Larger metastases (>3 cm), or lesions considered complex by virtue of their shape, size or location: 2,400 cGy in three treatments Retreatment of complex lesions: 1,600–2,500 cGy in five sessions
Vestibular Schwannoma Although surgical resection remains the gold standard treatment of vestibular schwannomas, stereotactic radiosurgery is a reasonable alternative for most patients with small to moderate tumors, not exceeding 3 cm in the largest dimension and which do not compress the brainstem. Because it is less invasive than surgical therapy, it is also quickly becoming most popular choice of treatment for the internet-savvy patients, who first seek webbased information, and then hi-tech medical care. Although the longevity of its effect is still questionable beyond 15–20 years, the efficacy of radiosurgery on vestibular schwannoma is clear. Most reports in the literature cite a 5–10-year tumor control rate between 95 and 98%. The risk of brainstem, trigeminal or facial nerve injury has been reduced to 1% or less with modern treatment techniques. The possibility of hearing-preservation
68
is much more variable, and is dependent on the patient’s pre-treatment hearing status, and the technique of radiosurgical treatment. In a recent report, fractionated treatment achieved a hearing preservation rate 2.5-fold higher than singlesession radiosurgery (Andrews [21]). One of the best hearing-preservation rates reported in the literature comes from Chang et al. [22] who utilized the CyberKnife to treat vestibular schwannomas with marginal doses between 18 and 21 Gy given in three sessions. The tumor control rate in that study was similar to previous reports, but the hearing preservation rate of 74% in patients with serviceable hearing is noteworthy.
Cavernous Sinus Meningioma Given the cerebrovascular and cranial nerve morbidity of surgical resection of meningiomas in the cavernous sinus, radiosurgery is now the first-line treatment of small-to-moderate size tumors in this region. These tumors usually have a characteristic appearance on MRI images: brightly enhancing, almost always homogenous appearing tumors originating from the cavernous sinus; thus, surgical biopsy is rarely necessary. In addition to its efficacy in halting tumor progression, radiosurgery is also highly effective in reversing cranial nerve deficits when given promptly after the appearance of symptoms. The risk of injury to the optic apparatus can be kept reasonably low if there is a 2–3 mm gap between tumor and the optic chiasm, and if the chiasmal dose is 8 Gy or less for single-session treatment. Hasegawa et al. [23] recently reported their long-term outcomes for gamma knife treatment for these tumors. The actuarial 5- and 10-year tumor control rate is 94 and 92% respectively. Forty six percent of patients experienced improvement in their neurological function, whereas only 12% experienced any cranial nerve morbidity. We recently reported the experience from our own institution, treating meningiomas of the
1125
1126
68
Cyberknife: clinical aspects
skull base with the CyberKnife. Forty four percent of the patients had meningiomas in the sphenocavernous region, and all patients were treated with 25 Gy to the tumor margin in five treatment sessions. With this regimen, the short-term tumor control rate was 97% and only 6% of patients experience deterioration of any neurological function. Even with large tumors directly in contact with the optic apparatus, the risk of visual complication was acceptably low with the CyberKnife regimen of 25 Gy in five sessions. Schwannomas of the cavernous sinus are also treated with the same dose (> Figure 68-14a, b, c).
In this case, a neuroma recurred 6 year s/p gross total resection (> Figure 68-14a). The formal visual field showed a field cut superiorly. The patient was treated with 450 cGy 5 to 80% IDL; the Rt Optic nerve dose was <1,600 cGy. Subsequent testing revealed that the visual deficit had substantially decreased (> Figure 68-14c). There was no alteration of vision over the follow-up period of 4.5 years. There have been no neurological deficits as a result of CyberKnife irradiation of benign lesions with multi-session treatments in the cavernous sinus.
. Figure 68-14 (a) CT, axial view through the orbits, contrasted. An irregular shaped lesion (Schwannoma) is contrast enhanced, and occupying the right optic neural foramen. (b) The formal visual field testing at the time of presentation shows a superior hemianopsia (black). (c) The visual field testing 6 months after irradiation, showing a minimal visual field deficit
Cyberknife: clinical aspects
Non-Neurosurgical Tumors The authors note that the CyberKnife has been very effective in the treatment of non-neurosurgical tumors. At GUH, CyberKnife is routinely used with great efficacy in the treatment of cancers of the head and neck, urogenital organs, pancreas, lung, and sarcomas of the extremities. A case is presented (> Figure 68-15a,b,c) that is emblematic of lung cases performed at GUH. The patient, a woman with poor lung function, underwent a Bx, yielding a diagnoses
68
of NSCLC. She was intermittently O2 dependent, and clearly not a surgical candidate. Fiducials were placed into posterior chest wall. She was contoured (> Figure 68-15b), and treated with CyberKnife (800 cGy 3). At 6 months, there was no evidence of tumor (> Figure 68-15c). Nasopharyngeal tumors are excellent candidate lesions for CyberKnife. At GUH, we have treated 10 patients treated to date with external beam irradiation (5,040 cGy in standard fractionation) followed by CyberKnife boost
. Figure 68-15 (a) Bx+ NSCLC in a woman with poor lung function, intermittently O2 dependent. Not a surgical candidate. Fiducials were placed into posterior chest wall. (b) Treatment plan of the lesion, encompassing the gross tumor volume and a margin. (c) No evidence of the primary lung cancer 6 months after irradiation
1127
1128
68
Cyberknife: clinical aspects
. Figure 68-16 Contouring of a nasopharyngeal tumor in preparation for CyberKnife irradiation. Large series have demonstrated excellent results using conformal external beam with a CyberKnife boost to the tumor and margin
structures, and of the physicist, who creates the treatment plan. A weekly planning session to discuss cases and treatment plans is a useful adjunct to daily discussions.
References
(200 cGy 10). There has been no evidence of disease in six at 3 years; three were lost to follow up, and one had a local recurrence and distant metastases. Contouring of a nasopharyngeal tumor in preparation for CyberKnife irradiation is shown (> Figure 68-16). Stanford University has noted excellent results: their patients enjoyed a 100% local control of nasopharyngeal carcinoma in 23 patients treated with a CyberKnife radiosurgical boost [24].
Conclusion Stereotactic radiosurgery is an interdisciplinary endeavor. The best treatment results from active input from the surgeon, who provides perhaps the best understanding of the anatomy and of course surgical management of the tumor, of the radiation oncologist who provides the greatest insight into dosing strategies and understanding of critical dose to previously radiated critical
1. Ang AK, Thames HD, Van der Kogel AJ, Van der Schueren E. Is the rate of repair of radiation induced sublethal damage in rat spinal cord dependent on the size per fraction? Int J Radiat Oncol Biol Phys 1987;13 (4):557-62. 2. Brenner DJ, Martel MK, Hall EJ. Fractionated regimens for stereotactic radiotherapy of recurrent tumors in the brain. Int J Radiat Oncol Biol Phys 1991;21(3):819-24. 3. Fowler JF, Joiner MC, Williams MV. Br J Radiol 1983;56 (668):599-601. 4. Walther HE. Krebsmetastasen. Basel: Bens Schwabe Verlag; 1948. 5. Weichselbaum RR, Beckett MA, Vijayakumar S, et al. Radiobiological characterization of head and neck and sarcoma cells derived from patients prior to radiotherapy. Int J Radiat Oncol Biol Phys 1990;19:313-9. 6. Chang SD, Main W, Martin DP, Gibbs IC, Heilbrun MP. An analysis of the accuracy of the cyberknife: a robotic frameless stereotactic radiosurgery system. Neurosurgery 2003;52(1):140-7. 7. Degen JW, Gagnon GJ, Voyadzis JM, McRae DA, Lunsden M, Dieterich S, Molzahn I, Henderson FC. CyberKnife stereotactic radiosurgical treatment of spinal tumors for pain control and quality of life. J Neurosurg Spine 2005;2:540-9. 8. Gerszten PC, Ozhasoglu C, Burton SA, Vogel WJ, Atkins BA, Kalnicki S, Welch WC. CyberKnife frameless stereotactic radiosurgery for spinal lesions: clinical experience in 125 cases. Neurosurgery 2004;55(1):89-98. 9. Gerszten PC, Ozhasoglu C, Burton SA, Vogel WJ, Atkins BA, Kalnicki S, Welch WC. CyberKnife frameless singlefraction stereotactic radiosurgery for benign tumors of the spine. Neurosurg Focus 2003;14(5):e16. 10. Ryu SI, Chang SD, Kim DH, Murphy MJ, Le QT, Martin DP, Adler JR, Jr. Image-guided hypo-fractionated stereotactic radiosurgery to spinal lesions. Neurosurgery 2001;49:838-46. 11. Hall EJ. Radiobiology for the radiologist: DNA breaks and aberrations, pp. 22–229. 12. Wigg DR, Koschel K, Hodgson GS. Tolerance of the mature human nervous system to photon irradiation. Br J Radiol 1981;54:787-98. 13. Schultheiss TE, Kun LE, Ang KK, Stephens LC. Radiation response of the central nervous system. Int J Radiat Oncol Biol Phys 1995;31(5):1093-112. 14. Nieder C,MilasL, Ang KK. Tissue tolerance to reirradiation. Sem Radiat Oncol 2000;10:200-9.
Cyberknife: clinical aspects
15. Nakamura JL, Verhey LJ, Smith V, Petti PL, Lamborn KR, Larson DA, Ware WM, McDermott MW, Sneed PK. Dose conformity of gamma knife radiosurgery and risk factors for complications. Int J Radiat Oncol Biol Phys 2001;51(5):1313-9. 16. Paddick I. A simple scoring ratio to index the conformity of radiosurgical treatment plans. Technical note. J Neurosurg 2000;93 Suppl 3:219-22. 17. Stafford SL, Pollock BE, Leavitt JA, Foote RL, Brown PD, Link MJ, Gorman DA, Schomberg PJ. A study on the radiation tolerance of the optic nerves and chiasm after stereotactic radiosurgery. Int J Radiat Oncol Biol Phys 2003;55(5):1177-81. 18. Friedman WA, Foote KD. Linear accelerator radiosurgery in the management of brain tumours. Ann Med 2000; 32(1):64-80. 19. Naoi Y, Cho N, Miyauchi T, Iizuka Y, Maehara T, Katayama H. Usefulness and problems of stereotactic radiosurgery using a linear accelerator. Radiat Med 1996;14(4):215-9. 20. Nishizaki T, Saito K, Jimi Y, et al. The role of CyberKnife radiosurgery/radiotherapy for brain metastases of multiple
21.
22.
23.
24. 25. 26. 27. 28.
68
or large size tumors. Minim Invasive Neurosurg. 2006;49 (4):203-9. Andrews DW, Suarez O, Goldman HW, et al. Stereotactic radiosurgery and fractionated stereotactic radiotherapy for the treatment of acoustic schwannomas: comparative observations of 125 patients treated at one institution. Int J Radiot Onc Biol Phy 2001;50:1265-78. Chang SD, Gibbs IC, Sakamoto GT et al. Staged stereotactic irradiation of acoustic neuroma. Neurosurgery 2005;56(6):1254-61. Hasegawa T, Kida Y, Yoshimoto M et al. Long-term outcomes of Gamma Knife surgery for cavernous sinus meningioma. J Neurosurg 2007;107(4):745-51. Tate DJ, et al. IJROBP 1999;45:915-21. Harrington KD. Orthopaedic management of metastatic bone disease. pp. 283–307. Herbert D. Prediction of response in radiation therapy 1988:400–513. Landis SH, Murray T, Bolden S, Wingo PA. Cancer statistics 1999. CA Cancer J Clin 1999;49(1):8-31. Markman M. Early recognition of spinal cord compression in cancer patients. Cleve Clin J Med 1999;66(10): 629-31.
1129
60 CyberKnife: Technical Aspects J. R. Adler . D. W. Schaal . A. Muacevic
Robotic Radiosurgery – Overview The design of the CyberKnife System (Accuray Incorporated, Sunnyvale, CA, USA) is based on principles of radiosurgery that have been in clinical practice for over 30 years. The practice of radiosurgery involves the precise application of an ablative dose of radiation to a defined target volume while protecting the surrounding healthy tissue. During radiosurgery, many radiation beams from different directions intersect in the tumor region, where they accumulate to a total dose, while the surrounding healthy tissue receives only a small fraction of the radiation. Until recently, the Gamma Knife system (Elekta AG, Stockholm, Sweden) was considered the standard instrument for neuro-radiosurgical applications. In addition, some centers use linear accelerators (LINAC) for radiosurgical procedures that are most commonly used for conventional radiation therapy but can be altered for radiosurgical treatment. Detailed physics testing is part of this alteration process because radiosurgical applications demand significantly higher quality and precision than conventional radiotherapy indications. Gamma Knife and conventional LINACs both require the application of an invasive stereotactic frame onto the patient’s head to achieve the desired accuracy of +/ 1 mm. The revolutionary development of the frameless CyberKnife technology, which combines integrated image guidance and robotic technology, has led to a paradigm shift in radiosurgery. During CyberKnife radiosurgery, instead of a stereotactic frame, real-time intraoperative imaging is used to establish the tumor position with
#
Springer-Verlag Berlin/Heidelberg 2009
reference to skeletal anatomy or implanted fiducial markers. In many clinical circumstances, the combination of image guidance and robotics has numerous advantages over conventional approaches to radiosurgery. For the first time in history, a truly non-invasive and pain-free radiosurgical treatment is available. Moreover, whenever deemed necessary, the treatment can be delivered in several fractions or sessions. This has the potential to be safer for the treatment of lesions in highly sensitive areas, i.e., meningiomas of the optic system, acoustic neuromas, or larger lesions [1–3]. In this chapter, we describe the technological background of CyberKnife robotic radiosurgery and its component parts.
Registration: Patient Alignment and Target Localization The primary control point of the CyberKnife operational system is a graphical user interface delivery system that initiates and monitors operations of the different components. During treatment delivery, the software monitors the system status and safety controls, reports errors, manages the patient database and records treatment data log files for post-treatment assessment and analysis. The target localization system (TLS) is composed of two orthogonally positioned diagnostic X-ray sources and two corresponding amorphous silicon plates. They provide near real-time digital X-ray images of the patient in the treatment position. During patient setup, the target position is determined (relative to nearby bony anatomy
950
60
Cyberknife: technical aspects
or, in extracranial treatments, implanted fiducials) by comparison of the left anterior oblique (LAO) and right anterior oblique (RAO) X-rays, with digitally reconstructed radiographs (DRRs) in the same LAO and RAO positions derived from the preoperative CT scans. The system uses a movable treatment table to position the patient before treatment. This operating table can be moved automatically along five axes (three translations and two rotations). The final rotational movement (yaw) is applied manually. The newly developed robotic couch (RoboCouch Patient Positioning System, Accuray Incorporated) can be adjusted entirely robotically (i.e., based on image information and without intervention by the user) in all 6 axes. > Figure 60-1 illustrates the overall system configuration.
The TLS is activated repeatedly throughout the treatment to verify the position of the target. Again, intraoperative X-rays are registered in real-time to a library of pre-operative DRRs that samples the full range of motion of the target center in 6 degrees of freedom (see > Figure 60-2). The registration process verifies that the real-time images represent an acceptable position within the range sampled by the reference images. It interpolates the actual position and orientation of the real-time image with respect to the reference, and sends positional correction data to the robot. Patient movements within certain limits (10 mm in x, y or z; 1 pitch and roll, 3 yaw) can be compensated for automatically by the robot. If a movement exceeds these limits, an error warning is triggered
. Figure 60-1 A recent configuration of the CyberKnife System. Orthogonal X-ray sources project to in-floor amorphous silicon detectors. A compact 6-MV LINAC is manipulated by an industrial 6-axis robot (KUKA Robotics Corp, Augsburg, Germany). The RoboCouch is a robotic patient positioning system. The Synchrony camera collects positional information from light-emitting diodes attached to a patient vest for use in tracking of moving extracranial tumors
Cyberknife: technical aspects
60
. Figure 60-2 A library of digitally reconstructed radiographs (DRRs) is generated from the treatment planning CT. Each of these images, which approximate an oblique projection, emulates a unique pose of the patient’s anatomy
causing the system to pause for patient realignment. By verifying the position of the target frequently throughout the treatment in this manner, the CyberKnife System ensures the accurate delivery of radiation to the target without framebased targeting and immobilization. The clinically relevant accuracy of the entire system has been determined at several centers using thin-slice CT to be sub-millimetric for intracranial treatments [4,5]. The same measure of accuracy can also be achieved when ablating spinal lesions, whether tracking using implanted markers [6] or without these markers, but using the Xsight Spine Tracking System (Accuray Incorporated) [7,8]. Despite being frameless, the CyberKnife emulates an important feature of the Gamma Knife, its ability to deliver many beams from many, noncoplanar orientations. Unlike the Gamma Knife however, the CyberKnife is not restricted to treating isocenters; instead, at each position of the robot, the beam can be directed toward a different area of the target region. This design feature enables the treating surgeon to select from a large array of non-isocentric, non-coplanar beams during treatment planning, thereby creating dose distributions that conform especially well to irregularly shaped lesion volumes (> Figure 60-3) [9]. In contrast, conventional radiosurgical devices
are only capable of constructing spherical dose distributions around a discrete isocenter.
Surgical Guidance: 6D Skull Tracking and 6D Fiducial Tracking The CyberKnife enables the full, 6D range of motion of the target center to be tracked within the reference frame of the treatment planning CT. The original version of the CyberKnife System’s 6D skull-tracking algorithm was able to locate targets relative to known skeletal landmarks, such as the mastoid process [10]. Later versions extract this information from bony anatomy automatically, without relying on named landmarks [11]. The current algorithm employs a multi-phase registration strategy to achieve submillimetric targeting and tracking accuracy in near real-time [5]. The total clinical accuracy of such image-guided localization has been shown to be sub-millimetric in phantom studies using thin-slice planning CTs [4,5]. Arguably the greatest advantage of frameless, image-guided targeting and tracking is that this makes radiosurgery outside the brain possible. Extracranial application of the CyberKnife began with the groundbreaking work for lesions in the
951
952
60
Cyberknife: technical aspects
. Figure 60-3 Panel a: In standard radiosurgery, dispersed isocentrical beams all intersect at a common region (light gray). Because multiple spherical volumes are needed to cover non-spherical lesions, the resulting dose distribution tends to be inhomogeneous. Panel b: Non-isocentric beams from various directions do not all intersect a single point and the dose is more homogeneous
spine by Ryu and co-workers [12]. Spinal targets were initially treated using radio-opaque fiducial markers (stainless steel screws) implanted in vertebrae adjacent to the lesion prior to preoperative imaging. In this technique, which is still necessary in certain circumstances, the fiducial tracking algorithm searches the pair of X-ray images, detects the fiducial markers, and calculates translations and rotations of the target. While only one fiducial is required to track anatomical translations, three or more are necessary to compute both translational and rotational motions [13]. CyberKnife’s total clinical accuracy for fiducial tracking in the spine is also sub-millimetric [6]. In recent years, system enhancements have facilitated fiducialbased tracking of targets throughout the body, including lesions that move with respiration, such as lung and abdominal tumors [13–17].
Spinal Tracking Without Fiducials Spinal radiosurgery is a new class of procedures designed for primary or adjuvant treatment of certain spinal disorders [18,19]. Because such large doses of radiation are administered, spinal radiosurgery, similar to its intracranial predecessor, requires extremely accurate targeting.
In contrast, the lack of precision inherent in conventional external beam radiation therapy, and the limitations of target immobilization techniques, generally preclude large single-fraction irradiation near radiosensitive structures such as the spinal cord. The frameless CyberKnife radiosurgery system has overcome these problems by using real-time image guidance which allows the paraspinous target to be tracked even in the presence of occasional patient movement. Continuous tracking and correction for motion of the spine throughout treatment is a requirement for spinal radiosurgery, because patients do move after set-up is complete [20]. Until recently, clinicians surgically implanted fiducials into the spine to track the movement of the lesion during treatment [12,21]. However, this step introduces the added surgical risks associated with an invasive surgery (albeit minor), lengthens treatment time, and reduces patient comfort. It would be ideal if it were possible to track spinal lesions using bony landmarks (similar to tracking intracranial lesions based on skull anatomy) instead of fiducials. Recently, such a fiducial-free spinal tracking system has been introduced (XsightSpine Tracking System, Accuray Incorporated). The Xsight fiducial-free localization process is performed in several stages, beginning with image
Cyberknife: technical aspects
enhancement, in which DRRs and intra-treatment radiographs undergo processing to improve the visualization of skeletal structures. Prior to treatment, a region of interest (ROI) surrounding the target volume is selected based on an initial userdefined position, which is refined automatically by an algorithm that seeks to maximize the image entropy within the ROI. The resulting optimal ROI typically includes one to two vertebral bodies which form the basis of patient-motion tracking and alignment. 2D-3D image registration uses similarity measures to compare the X-ray images and DRRs, and a spatial transformation parameter search method to determine changes in patient position. A mesh is overlaid in the ROI, and local displacements in the mesh nodes are
60
estimated individually, constrained by displacement smoothness. Nodal displacements in the two images within the mesh form two, 2D displacement fields. 3D displacements of the targets and global rotations of spinal structures within the ROI can then be calculated from the two 2D displacement fields by interpolation. A screen shot from an Xsight treatment session shows that the overlaid-mesh technique is successful even in the presence of spinal instrumentation (> Figure 60-4). The main advantages of the fiducial-free system are: (1) the ability to account for nonrigid deformation, thereby improving the targeting accuracy in the situation that a patient-pose change occurs subsequent to the CT scan, and
. Figure 60-4 Xsight spine tumor tracking after dorsal transpedicular stabilization surgery. Bone tracking of the vertebra was possible even though the tumor area was overshadowed by the metal implants
953
954
60
Cyberknife: technical aspects
(2) no risk of complication and increased convenience for both the patient and the clinician. The fiducial-free tracking system of the CyberKnife has been proven in end-to-end phantom tests and simulations, using existing CT-image sets of the spine, to be accurate to within about 0.5 mm [7,8].
system has been shown under laboratory conditions to result in highly precise radiation delivery without patient restraints [17,22,23]. Its utility in treating patients has also been documented in multiple institutions [13,15–17,24].
CyberKnife Team Robotic Motion Compensation The Synchrony Respiratory Tracking System (Accuray Incorporated) is one of the most advanced components of the CyberKnife System’s product configuration. It is only briefly described here as it is mainly used for non-neurosurgical, extracranial indications (i.e., lung, liver and pancreatic tumors). With Synchrony, tumor movement is detected by means of implanted fiducials, usually small gold seeds that are introduced into or around the tumor under CT fluoroscopy or via a bronchoscope. Waiting a few days after placement allows the fiducials to settle and edema to subside, after which a preoperative planning CT scan is made so as to define the 3D operative volume. During patient setup, these fiducials are used to align the DRRs generated from the planning CT with the X-ray images obtained in the treatment room. In the meantime, multiple light emitting diodes (LEDs) positioned on the patient’s chest or abdomen are monitored in real time (32 frames per second) by cameras mounted on the ceiling. A correlation model between the internal and external markers is created prior to treatment and is updated and verified throughout the treatment. The correlation between the positions of the external LEDs and internal fiducials enables the robot to move the LINAC synchronously with the patient’s breathing as it relates to the motion of the tumor inside. Displacement errors higher than a user-prescribed value based on an analysis of tumor motion triggers the system to pause so that adjustments can be made before continuing with the treatment. This
High quality radiosurgical applications are becoming increasingly more complex. Different treatment applications require a range of medical professionals, such as surgical specialist, radiation oncologists, radiologists, and specially trained medical physicists, to choose, plan, and conduct the most effective and safe procedure. Intracranial and spinal treatments should be reviewed by experienced neurosurgeons capable of understanding the complex topographical relationships of cranial and spinal anatomy and pathology. Orthopedic surgeons may also be helpful to support spinal treatments. The contribution of experienced imaging experts for optimal selection and interpretation of radiologic studies will often significantly enhance the quality of a radiosurgical procedure. Non-invasive robotic radiosurgery is a rapidly emerging interdisciplinary field that is opening new horizons in the area of cancer treatment and perhaps beyond, into select non-neoplastic disorders.
References 1. Adler JR Jr, Gibbs IC, Puataweepong P, et al. Visual field preservation after multisession cyberknife radiosurgery for perioptic lesions. Neurosurgery 2006;59:244-54; discussion 244‐54. 2. Chang SD, Gibbs IC, Sakamoto GT, et al. Staged stereotactic irradiation for acoustic neuroma. Neurosurgery 2005;56:1254-61; discussion 1253‐61. 3. Giller CA, Berger BD, Fink K, et al. A volumetric study of CyberKnife hypofractionated stereotactic radiotherapy as salvage for progressive malignant brain tumors: initial experience. Neurol Res 2007;29(6):563-8. 4. Chang SD, Main W, Martin DP, et al. An analysis of the accuracy of the CyberKnife: a robotic frameless
Cyberknife: technical aspects
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
stereotactic radiosurgical system. Neurosurgery 2003;52: 140-6; discussion 146‐7. Fu D, Kuduvalli G, Mitrovic V, et al. Automated skull tracking for the CyberKnife image-guided radiosurgery system. In: Reinhardt JM, Pluim JP, editors. SPIE medical imaging: image processing. vol 5744. San Diego, CA: SPIE; 2005. p. 366-77. Yu C, Main W, Taylor D, et al. An anthropomorphic phantom study of the accuracy of Cyberknife spinal radiosurgery. Neurosurgery 2004;55:1138-49. Ho AK, Fu D, Cotrutz C, et al. A study of the accuracy of Cyberknife spinal radiosurgery using skeletal structure tracking. Neurosurgery 2007;60:147-56. Muacevic A, Staehler M, Drexler C, et al. Technical description, phantom accuracy, and clinical feasibility for fiducial-free frameless real-time image-guided spinal radiosurgery. J Neurosurg Spine 2006;5:303-12. Yu C, Jozsef G, Apuzzo ML, et al. Dosimetric comparison of CyberKnife with other radiosurgical modalities for an ellipsoidal target. Neurosurgery 2003;53:1155-62; discussion 1153‐62. Adler JR. Frameless radiosurgery. In: Goetsch SJ, De Salles AAF, editors. Stereotactic surgery and radiosurgery. vol 17. Madison, Wisconsin: Medical Physics Publishing; 1993. p. 237-48. Murphy MJ. An automatic six-degree-of-freedom image registration algorithm for image-guided frameless stereotaxic radiosurgery. Med Phys 1997;24:857-66. Ryu SI, Chang SD, Kim DH, et al. Image-guided hypofractionated stereotactic radiosurgery to spinal lesions. Neurosurgery 2001;49:838-46. Collins BT, Erickson K, Reichner CA, et al. Radical stereotactic radiosurgery with real-time tumor motion tracking in the treatment of small peripheral lung tumors. Radiat Oncol 2007;2:39. Brown WT, Wu X, Fayad F, et al. CyberKnife radiosurgery for stage I lung cancer: results at 36 months. Clin Lung Cancer 2007;8:488-92.
60
15. Le QT, Loo BW, Ho A, et al. Results of a phase I doseescalation study using single-fraction stereotactic radiotherapy for lung tumors. J Thorac Oncol 2006;1:802-9. 16. Pennathur A, Luketich JD, Burton S, et al. Stereotactic radiosurgery for the treatment of lung neoplasm: initial experience. Ann Thorac Surg 2007;83:1820-4; discussion 1824‐5. 17. Muacevic A, Drexler C, Wowra B, et al. Technical description, phantom accuracy, and clinical feasibility for single-session lung radiosurgery using robotic imageguided real-time respiratory tumor tracking. Technol Cancer Res Treat 2007;6:321-8. 18. Degen JW, Gagnon GJ, Voyadzis JM, et al. CyberKnife stereotactic radiosurgical treatment of spinal tumors for pain control and quality of life. J Neurosurg Spine 2005; 2:540-9. 19. Gerszten PC, Burton SA, Ozhasoglu C, et al. Radiosurgery for spinal metastases: clinical experience in 500 cases from a single institution. Spine 2007;32:193-9. 20. Murphy MJ, Chang SD, Gibbs IC, et al. Patterns of patient movement during frameless image-guided radiosurgery. Int J Radiat Oncol Biol Phys 2003;55: 1400-8. 21. Gerszten PC, Ozhasoglu C, Burton SA, et al. Evaluation of cyberknife frameless real-time image-guided stereotactic radiosurgery for spinal lesions. Eur J Cancer Suppl 2003;1:S151. 22. Seppenwoolde Y, Berbeco RI, Nishioka S, et al. Accuracy of tumor motion compensation algorithm from a robotic respiratory tracking system: a simulation study. Med Phys 2007;34:2774-84. 23. Wong KH, Dieterich S, Tang J, et al. Quantitative Measurement of CyberKnife Robotic Arm Steering. Technol Cancer Res Treat 2007;6:589-94. 24. Nuyttens JJ, Prevost JB, Praag J, et al. Lung tumor tracking during stereotactic radiotherapy treatment with the CyberKnife: Marker placement and early results. Acta Oncol 2006;45:961-5.
955
71 Focused and Conventional Radiation for Acoustic Nerve Tumors R. Den . S. H. Paek . D. W. Andrews
Introduction Acoustic neuromas, also referred to as vestibular schwannomas, are benign tumors commonly arising from the transition zone between central oligodendroglial cells and peripheral schwann cells within the vestibular portion of cranial nerve VIII. They account for approximately 80–90% of all cerebellopontine angle tumors in adults and have a median age of diagnosis of approximately 50 years [1]. The management of vestibular schwannomas has evolved over the past half century from a microneurosurgical modality to a non-invasive approach using focused radiation either as a single treatment stereotactic radiosurgery or as a fractionated course fractionated stereotactic radiotherapy. Stereotactic radiosurgery or SRS has been widely regarded as the standard of care for small to intermediate-sized tumors with the current standard dose of 12 Gy prescribed to the 50% isodose line. This current dose represents two decades of dose reductions that have resulted in decreased incidence of cranial neuropathies and improved hearing preservation rates while maintaining a high tumor control rate. A parallel development has occurred using fractionated stereotactic radiotherapy or FSR. FSR treatment courses may be as short as a week or as long as 5–6 weeks depending on the daily dose schedule. Fractionation has been thought to provide several radiobiological advantages over single fraction treatment resulting in higher rates of hearing preservation. This chapter is a #
Springer-Verlag Berlin/Heidelberg 2009
review of acoustic neuromas and their treatment with the use of FSR.
Epidemiology In a review of 1,400 temporal bones, Schuknecht found an incidence of occult acoustic neuromas in 0.57%. According to the 1991 National Institute of Health Consensus Statement, an estimated 2,000–3,000 new cases of unilateral acoustic neuromas are diagnosed in the United States each year – an incidence of about 1 per 100,000 per year [2]. There has been an increasing incidence of asymptomatic lesions detected, which has been posited to be secondary to the increasing use of intracranial imaging modalities such as MRI or high resolution CT [3]. This indicates that the vast majority of acoustic neuromas that exist never become clinically evident, because of very slow or arrested growth. Five percent of all diagnosed tumors are associated with neurofibromatosis type 2 [4], an autosomal dominant disorder which results in bilateral acoustic neuromas and multiple meningiomas. This disorder is characterized by mutations in the gene product merlin, located on chromosome 22q12 [5].
Pathology Acoustic neuromas are benign tumors that arise from neural crest-derived Schwann cells. They
1152
71
Focused and conventional radiation for acoustic nerve tumors
are generally well circumscribed, encapsulated masses which on microscopic examination have two growth patterns: a high cellularity zone and low cellularity zone classified as Antoni A, compact tissue with spindle cells in palisades, and Antoni B, loose tissue with cyst formation [6]. Since tumors do not arise from the neurons, nerves are generally displaced by the tumor; however, they may become entrapped in the capsule. Malignant degeneration to fibrosarcoma is quite rare, with only case reports in the literature.
suggested that tumors smaller than 1 cm could be watched, especially in elderly patients. Bederson et al. compared patients who required surgery to those who did not require intervention and found that a larger initial size (2.7 cm vs. 2.1 cm) and a faster initial growth rate (7.9mm/year vs. 1.3 mm/year) were significantly prognostic [7]. Fucci et al. in a review of 119 patients found that tumors greater than 2 cm at presentation were more likely to grow than tumors less than 2 cm [19]. Currently, the underlying mechanisms, which explain the differential tumor growth seen in various patients has not been elucidated.
Natural History There are several reports, which detail the natural history of tumor growth in patients with sporadic acoustic neuromas [7–15]. Silverstein et al. demonstrated a growth rate of 0.2 cm per year in seven patients under surveillance [16]. Mirz et al. reported on 64 patients with acoustic neuromas, and over the median follow-up between 5 months and 15 years, 14 tumors (22%) regressed, 35 tumors did not grow or had only minimal growth (growth rate up to 1 mm/year), whereas 15 tumors grew > 1 mm/year [17]. Walsh et al. investigated the natural history of 72 patients with a radiological diagnosis of unilateral acoustic neuromas [15]. Over the median follow-up period of 37.8 months, they reported that significant tumor growth (total growth >1 mm) in 36.4%, no or insignificant growth (0–1 mm) in 50%, and negative growth (<0 mm) in 13.6% of tumors. Furthermore, they noted that the growth rate of cerebellopontine angle tumors (1.4 mm/year) was significantly greater than that of tumors limited to the internal auditory canal (0.2 mm/year) [15]. Thomsen and Tos found in a study of 21 patients that after a mean followup of 4 years, 14% of tumors grew to a size requiring surgery. In a larger study of 35 patients, Valvassori and Guzman showed that 43% of tumors did not grow and 6% demonstrated growth greater than 50% of the original tumor size [18]. They
Clinical Presentation Cochlear nerve involvement occurs in 95% of cases of acoustic neuromas generally with symptoms of hearing loss or tinnitus. The incipient nature of this disease is that hearing loss is typically chronic and often undetected for years [16]. However, an acute hearing loss may occur, regardless of tumor size. Two proposed mechanisms include direct injury to the cochlear nerve by the tumor or interruption of cochlear blood supply. Tinnitus is thought to arise in 63% of patients with an average duration of approximately 3 years. While deaf patients have a lower incidence of tinnitus, this symptom may still arise. Involvement of the vestibular nerve occurs in approximately 60% of patients with common symptoms of mild unsteadiness with walking. However, true vertigo symptoms are quite rare given the slow growth of these tumors. Trigeminal nerve involvement manifests as facial paresthesias, hyperesthesia and facial pain. Facial nerve disturbance results in facial paresis and occasional taste disturbance. Facial nerve function is evaluated using the House-Brackmann Grading System [17], which divides dysfunction in both motion as well as gross appearance into 5 categories ranging from normal function to complete paralysis.
Focused and conventional radiation for acoustic nerve tumors
Diagnostic Testing The diagnosis is usually made with imaging, either MRI or high resolution CT. Schwannomas appear as homogenous soft tissue masses with intense enhancement on CT, while on MRI they are generally more heterogenous due to tumor vascularity. Audiometry is the gold standard by which hearing loss is evaluated. Hearing loss is evaluated using two parameters: tone audiometry and speech audiometry. Tone audiometry measures the ability to hear pure tones of various frequencies as a function of intensity measured in decibels. The pure tone frequencies evaluated in a complete audiogram are 250 Hz, 500 Hz, 1000 Hz, 2000 Hz, 3000 Hz, 4000 Hz, and 8000 Hz. Speech audiometry consists of two components: the speech threshold (also called the speech reception threshold) and the word discrimination score. The speech threshold determines the patient’s threshold for two-syllable words, the faintest intensity that the words can be correctly repeated at least 50% of the time. Word discrimination score determines the subject’s ability to correctly repeat single syllable words presented at a comfortable intensity. Tone and speech audiometry are the two components that compromise the Gardner-Robertson Classification [18]. This classification system divides patients into those with serviceable and non serviceable hearing. Serviceable hearing is classified as patients with a speech discrimination of greater than 50% and a pure tone average of <50 dB. This is an important parameter when comparing FSR to SRS or surgery.
Treatment Goals and Options There are three primary goals for the management of acoustic neuromas: the achievement of local control, the preservation of hearing, and the preservation of facial and trigeminal function. Currently, there are three possible treatment
71
modalities for vestibular schwannomas: observation, surgery, or radiation. Often the choice of treatment is based on the patient’s age, symptoms, tumor size, tumor growth, hearing level, patient and clinician preference. However, 20–40% of patients observed will require intervention. Sakimoto et al. demonstrated a direct relationship of tumor size to hearing loss [19]. They followed 31 patients with acoustic neuromas who underwent MRI and audiometry. They found a mean annual growth rate of 2.4 mm/year and average annual hearing loss of 2.3 dB/year, thus indicating that as tumor increase in size, the risk of deafness increases. Advances in microsurgical resection and intraoperative cranial nerve monitoring have improved the surgical outcome of the treatment of acoustic neuromas. Most modern surgical reviews show hearing preservation ranging from 30 to 80% [20–22]. Tumor size is directly related to cranial nerve dysfunction. Gormley et al. showed that the risk of functional hearing loss and facial nerve palsy was 52 and 4% for lesions less than 2 cm in size, while 100 and 62% for those greater than 4 cm [23]. The adaptation of the microscope to the operating theatre, development of facial and auditory nerve monitoring techniques, improved anesthesia, and improved perioperative management has decreased the risk of facial and trigeminal nerve dysfunction dramatically. Facial nerve function is preserved in over 90% of patients with tumors <1 cm and 56% in tumors greater than 2.5 cm [21],[24]. There is a significant role for microsurgery in the management of large tumors. Forster and colleagues showed that for large tumors ( >3 cm), tumor control rate with SRS was only 33% [25]. Further, Chan et al. showed a direct relationship between tumor volume and the necessity of surgical intervention [26]. However, given the significant impact of hearing and facial nerve function on patient’s quality of life, radiation has become the standard of care of tumors smaller than 3 cm.
1153
1154
71
Focused and conventional radiation for acoustic nerve tumors
A study by Regis et al. compared microsurgery to Gamma Knife using a patient questionnaire to evaluate the functional outcome and quality of life in a series of 224 consecutive patients [27]. They found preserved hearing function, facial nerve function and importantly a higher percentage of patients who were able to maintain the same professional activity after less invasive Gamma Knife treatment.
Radiation Biology Radiation damage induces disruption of the cell machinery that results in cessation of tumor proliferation. Ionizing radiation initiates the excitation of free electrons within tissue leading to the production of free radicals that disrupt cell division. Double strand breaks in DNA compromise the integrity of the genome and the ability of the cell to undergo multiple rounds of replication. The arrest of mitosis leads to the induction of apotosis or programmed cell death. Further radiation damage to cell phospholipid membranes triggers the apoptotic pathway [28]. There are radiobiologic advantages and disadvantages to the use of a fractionationated course versus a single large dose. Fractionation achieves similar tumor lysis while allowing adequate time for normal tissue repair between doses. Thus, this technique should limit associated toxicities but requires a greater total dose than a single fraction approach to overcome any repair by the tumor.
Radiation Physics Both photons and particles have been used in the treatment of acoustic neuromas through ionization, a process where electrons are ejected from a target atom. While photons are indirectly ionizing, charged particles are directly ionizing, thus photons themselves are not causing tissue damage, rather the electrons that are produced from the absorption of photons cause radiation injury. Photons are generated either through
the acceleration of electrons into a metal target (X-rays) or by the decay of certain elements (gamma rays). Linear accelerators use X rays to deliver radiation, while the Gamma Knife uses the decay of Cobalt-60 into its daughter element. The Gamma Knife stereotactic radiosurgery unit contains 201 cobalt radiation sources. The target lesion is placed at the center position of the helmet using a stereotactic frame affixed to the patient’s skull. The radiation sources are directed to the lesions using shuttered channels directed toward the center of the helmet. For irregularly shaped tumors, multiple isocenters are required to shape a conformal plan. Linear Accelerators or LINACs, are able to rotate the treatment beam around the target. Thus a single isocenter may be employed with this technique. There are two methods employed to achieve a conformal dose prescription: cone based collimators and multileaf collimators. Cone based collimators require multiple isocenters similar to gamma knife. Conformality is achieved with noncoplanar arc beam shaping and differential beam weighting. Multileaf collimators use a single isocenter and conformal dynamic arcs or IMRT. 6 MV photons are the energy used in this treatment modality. Dose prescriptions are distinct between Gamma Knife and Linac. Gamma Knife doses are defined as 50% of the peak dose delivered to the tumor periphery resulting in a steeper dose gradient. For irregularly shaped tumors, multiple isocenters are used to achieve conformality at the expense of dose homogeneity. Together, these variables can result in surrounding critical structures (cochlea, cochlear nerve) receiving higher doses. Linac based plans usually prescribe the peripheral dose to the 90–95% isodose line, resulting in a shallower dose gradient. Particularly with single isocenter dynamic arc technique, less dose is delivered to critical structures.
Patient Fixation The most critical aspect of the treatment of any tumor with radiation, particularly with
Focused and conventional radiation for acoustic nerve tumors
radiosurgery or radiotherapy, is the immobilization of the patient. This ensures that the target is treated at the correct prescription and minimizes toxicity from radiation exporsure to other organs at risk. For radiosurgery, with either a linear accelerator, LINAC, or a Cobalt-60 Gamma Knife, a metal frame is screwed into the skull. MRI and CT scans may be obtained with the frame in place or prior to placement depending on the composition of the metal. With a fractionated course, thermoplastic molded masks or an upper arch dental mold are employed for fixation.
71
. Figure 71-1 Axial T1W gadolinium enhanced image of right acoustic neuroma with Novalis treatment plan overlay
Fractionated Stereotactic Treatment Technique > Figure 71-1 features a typical FSR treatment plan for an acoustic neuroma. Fractionated courses are designed using LINAC stereotactic systems. This isodose plan is achieved using a single isocenter treatment. Five non-coplanar arcs are utilized to deliver an isodose prescription typically to the 90% isodose line. This technique is feasible due to the capability of conforming dose with mini-leaf dynamic collimation. As a tertiary collimating system, mini-leaf technology involves the design of 26 opposing pairs of 1 mm tungsten leaves, which move in a coordinated fashion to conform to the three-dimensional structure of an irregularly shaped target such as an acoustic neuroma. While achieving a high degree of conformality, this also obviates the need for more than one isocenter, thus increasing the homogeneity of the plan. The peripheral dose encompassing the tumor contains a 2 mm margin to account for the small interfractional daily movement.
Brief History of SRS In the initial series of acoustic neuromas treated with sterotactic radiosurgery, tumor control was
obtained in 8 of 9 patients, however hearing loss was reported in the majority of patients [29]. The U.S. experience from Pittsburgh demonstrated tumor control rates of 95% and greater with doses of 16–20 Gy; however, there was an unaccepatably high risk of fifth, seventh, as well as eighth nerve palsies as shown in > Table 71-1. The literature reports of acoustic neuromas published from 1988 to 1998, document the average rates of trigeminal, facial, and cochlear neuropathies as 34, 33, and 40% respectively. This led to a reduction in marginal SRS dose prescription to the current standard of 12 Gy. Tumor control rates were not compromised until the prescribed dose was reduced to below 10 Gy [32]. While, rates of trigeminal and facial neuropathies were dramatically reduced, hearing preservation was suboptimal with rates ranging from 33 to 83%. This led to the investigation of fractionated radiation therapy as opposed to single dose treatment regimens. The hypothesis as mentioned above is that optimal dose prescription balances tumor kill and normal tissue survival. Multiple smaller doses of radiation can achieve an equivalent tumor effect while limiting toxicity. Fractionation allows for the cellular repair mechanisms within normal tissue time to correct any damage to DNA
1155
1156
71
Focused and conventional radiation for acoustic nerve tumors
. Table 71-1 Summary of published experience with reliable reports of audiometry
Reference
N
Isodose prescription
Gamma knife radiosurgery Hirsch [30] 126 18–25 Gy Flickinger [31] 85 14–20 Gy (18 median) Foote [32] 36 16–20 Gy Kondziolka [33] 162 12–20 Gy (16.6 mean) 1988–1998 mean results Andrews [34] 69 12 Gy Karpinos [35] 96 10–24 Gy (14.5 mean) Regis [27] 104 12–14 Gy Iwai, Y [36] 51 8–12 (12 median) Flickinger [37] 313 12–13 Gy Van Eck [38] 78 13 Gy Hasegawa [39] 74 13 Gy 1999–2005 mean results Fractionated stereotactic radiotherapy Kagei [40] 39 36–44 Gy 2 Gy/fraction + 4 Gy boost Andrews [34] 56 50 Gy 2 Gy/fraction Chung [41] 72 45 Gy 1.8 Gy/fraction 1999–2005 mean results
Rate of cranial neuropathy
Tumor control rate
V
VII
VIII (m/s/yrs*)
91% 97% 100% 98% 96 2
21% 29% 59% 27% 34 8
15% 30% 67% 21% 33 12
24% (m,4.7)a 46% (s,2)b 42% (s,2)b 47% (s, 5–10)a 40 11
100% 91% 100% 96% 99% 98% NS 97 1
2% 11% 4% 4% 4% 2.5% NS 5 31
1% 4% 2% 0% 0% 1.2% NS 1 12
33% (s, 0.8)b 44% (s, 4)a 50% (s, > 3)a 56% (s, 5)a 79% (s, 6)b 83% 68% (s, 7) 59 18
97%
16%
8%
78% (s, 2)b,d
97%
7%
2%
100%
7%
4%
98 2
10 5
53
81% (s, 1.2)a 71% (s, 5)a,c 85% (s, 1)b 57% (s, 2) b 69 113
*m/s/yrs reflects measurable or serviceable hearing preservation rates at follow up intervals in years a raw hearing preservation rate; bactuarial hearing preservation rate; c current follow-up for this cohort (unpublished observation); d unclear impact of 4 Gy end-boost on hearing preservation 1 p = 0.0028 vs. early Gamma Knife cohort, trigeminal neuropathy; 2 p = 0.0087 vs. early Gamma Knife cohort, facial neuropathy; 3 p = 0.0166, FSR longest follow-up hearing vs. early Gamma Knife cohort hearing
to maintain the integrity of the genome prior to a mitotic event. Further, LINAC based fractionated courses is more widely available than Gamma Knife and may be applicable to more patients.
Summary of Literature for FSR Treatment of Acoustic Neuromas Tumor control rates reported in fractionated stereotactic radiosurgery series are well above 90% irrespective of fractionation schedule [34],
[42–48]. In modern series, facial nerve preservation rates have been reported from 95 to 100% [34],[42–44],[46],[48]. There seems to be little or no difference in treatment-related facial nerve toxicity between the gamma knife series, with a possibly more conformal dose distribution, and the linac series. Trigeminal nerve preservation rates in modern series have been reported from 92 to 100%, both in LINAC and in gamma knife series and also both in single-fraction and in fractionated series [34],[42–44],[46],[48]. There are, however, no randomized studies on a comparison of single and fractionated treatments. For the
Focused and conventional radiation for acoustic nerve tumors
purpose of this chapter, conventional fractionation is defined as single treatments of 1.5–2.5 Gy, while hypofractionation refers to radiation fraction sizes of 3–5 Gy.
Hypofractionation Kalapurkal et al. reported on 19 patients treated with either 36 or 30 Gy in 6 fractions [49]. This patient group had large tumors with a mean pons to petrous diameter of 28 mm. The total dose was reduced after the first six patients secondary to ataxia experienced by two patients. The follow up was a median 54 months either using CT or MRI. This regimen resulted in tumor regression in ten patients and stabilization of size in nine patients. Further, of the nine patients who had hearing prior to treatment, eight had preservation of hearing and one had improvement. There was no incident of facial or trigeminal dysfunction. Poen and colleagues [44] described the FSR experience of 33 patients, of whom ten had neurofibromatosis type two and seven had prior surgery. In this cohort, the median tumor diameter was 20 mm; three patients had tumor diameters greater than 3 cm. Multiple isocenters (1–4) were used for treatment planning and radiation dose was 21 Gy administered in 3 fractions. Patient follow up was a median of 24 months and 34% of patients had tumor regression, 63% had stable disease, and one patient had tumor enlargement. For this patient, although growth was noted at 2 years after the treatment, subsequent follow up revealed tumor regression. Useful hearing preservation, GardnerRobertson class I & II, overall was 77%, 92% in sporadic cases and 67% in patients with NF2. Trigeminal neuropathy was noted in five patients and 3% of patients had facial nerve injury (HouseBrackmann Grade III). Williams [50] described the treatment of 125 patients 111 of whom received 25 Gy in 5 fractions for tumors <3 cm and 14 being treated with 30 Gy in 10 fractions for tumors 3 cm. This report
71
demonstrated no new tumor growth as documented by MRI after a median follow up of 21 months. Of the 56 patients with audiometric follow-up, 26 patients maintained GardnerRobertson classification, 20 had worsening of hearing and ten experienced improvement. There were no permanent CN V or VII dysfunction. Meijer et al. [48] report on 80 patients treated with either 20 or 25 Gy in 5 fractions. After a mean of 35 months, the 5 year probability of tumor control as defined as increase in size of 2 mm or more was 94%. In all cases of tumor growth, this occurred within the first 3 years post treatment. The actuarial 5-year facial and trigeminal nerve preservation rates were 98 and 97% respectively. The 5-year hearing preservation probability was 61%, though this was measured subjectively. One patient developed gait instability 6 months after treatment.
Conventional Fractionation Early results of conventional fractionation were reported by Varlotto et al. [47]. They examined 12 patients, eight treated with primary FSR and four patients treated after primary surgery (3 had recurrence and one had persistent disease). The tumors received 1.8 Gy/day normalized to the 95% isodose line to a total dose of 54 Gy. After a median follow up of 26.5 months, there was 100% tumor control with three cases of reduction in tumor volume. No new cranial neuropathies occurred. Furthermore, hearing preservation occurred in 100% of patients. Fuss et al. reported on 51 patients treated with FSR to a mean total dose of 57.6 Gy in 1.8–2 Gy fractions [42]. With a mean follow up of 42 months, the local control rate was 95% and the actuarial 2- and 5-year tumor control rates were 100 and 97.7%, respectively. The actuarial useful hearing preservation rate was 85% at 2 and 5 years, with new hearing loss being diagnosed in four patients with NF2. There were no facial nerve
1157
1158
71
Focused and conventional radiation for acoustic nerve tumors
palsy and two cases of trigeminal nerve dysesthesia requiring medication. Sawamura and colleagues reported on 106 patients treated with a range of 40–50 Gy in 20– 25 fractions [51]. 5 patients had NF2 and 12 had prior surgery. After a median follow up of 45 months by MRI, neurologic as well as otologic examination, the 5 year actuarial tumor control rate was 91.4%. The 5 year hearing preservation rate and Gardner-Robertson class preservation was 72 and 65% respectively. Twelve patients developed hydrocephalus and required VP shunt placement. Mean tumor size was noted to be prognostic for hydrocephalus (25 mm vs. 18 mm). Other transient complications included 4% rate of facial nerve palsy, 14% rate of trigeminal palsy, and 17% rate of disquilibrium. However, preexisting tinnitus improved in 11 of 37 patients and vertigo improved in 11 of 20 cases. Chung et al. [41] documented the results of 72 vestibular schwannoma patients treated with either SRS or FSR. All deaf patients were treated with SRS and 27 patients were treated with FSR. 25 patients were treated to a total dose of 45 Gy in 25 fractions to the 90% isodose line, one patient received the same total dose in 28 fractions and one patient was treated to a dose of 47.5 Gy in 25 fractions prescribed to the 50% isodose line. The median tumor diameter was 16 mm. There was no tumor progression noted and hearing preservation was 85% at 1 year and 57% after 2 years. Two patients had transient facial numbness and one patient experienced temporary facial paresis. One patient required a VP shunt for hydrocephalus. Selch and colleagues [52] treated 48 patients with a dose of 54 Gy in 1.8 Gy/fraction prescribed to the 90% isodose line. The range of tumor diameters was 0.6–4 cm and tumor shrinkage of 1–14 mm was noted in 12 patients at a median of 6 months. MRI scan revealed an increase in tumor diameter by 1–2 mm in 12 patients at a median of 6 months follow up. Yet, 4 of these tumors decreased to the original size and
2 shrank to a size smaller than originally. Useful hearing, as defined subjectively as the inability to use a telephone, was preserved in 93% of evaluable patients, with a 5 year actuarial probability of preservation of function of 91.4%. The actuarial 5 year probability of facial nerve and trigeminal nerve preservation was 97.2 and 96.2% respectively. Tinnitus improved in six patients and worsened in two. In another study, Combs et al. [53] reported on 106 patients treated with radiation prescription of 57.6 Gy administered in 32 fractions. 13 patients had tumor diameters 1 cm, 48 had between 1 and 2 cm, 30 patients had between 2 and 3 cm in size, 13 had 3–4 cm in diameter and 2 had tumors larger than 4 cm. At a median of 48.5 months follow up, tumor control rate was 94.3% at 3 years and 93% at 5 years. Hearing preservation was 94% at 5 years for patients with Gardner-Robertson class I and II. Irreversible trigeminal neuropathy occurred in three patients and two patients experienced permanent facial nerve damage. Chan and colleagues [26] reported on 70 patients, of whom 11 had NF2 and 15 had prior surgery, who received a total dose of 54 Gy in 30 fractions to the 95% isodose line. The patients were followed for a median of 45.3 months. Radiologic data was available on 68 patients, with actuarial tumor shrinkage of 36 and 62% at 3 and 5 years respectively. Actuarial tumor control rates were 100 and 98% at 3 and 5 years. Facial nerve preservation was 99% and trigeminal nerve function was maintained in 96% of patients. This study documented that prior resection increased the risk of trigeminal neuropathy. Andrews et al. [34] treated 56 patients with FSR with median follow up of 27 months. Tumor control was 97%, trigeminal nerve toxicity was 7% and facial nerve dysfunction was 2%. They found that 85% of patients treated with SRT maintained their Gardner-Robertson grade, in comparison to 60% treated with Gamma Knife SRS. Further, in examining those patients with
Focused and conventional radiation for acoustic nerve tumors
serviceable hearing, 81% versus 33% maintained serviceable hearing in SRT versus SRS. Sakamoto et al. demonstrated that FSR reduced the average annual hearing loss after radiation [54]. In their cohort of 72 patients, 21 had audiometric data prior to radiation. When comparing the mean annual rates of hearing loss before irradiation and after therapy, they found the rates diminished with time from 18.6 prior to treatment, to 11.2, 6.2, 5.1, and 5.0 dB/year in the first, second, third, and fourth year of follow up respectively. Thomas et al. [55] further reduced the total fractionated dose to 45 Gy in 25 fractions prescribed to the 90% isodose line in a cohort of 34 patients with serviceable hearing. They documented a 2 year and 4 year actuarial control rates of 100 and 95.7%. Permanent trigeminal and facial nerve complications were 0 and 6%, respectively. The actuarial 2 and 3 year serviceable hearing preservation rates were 63%, with a median loss of speech reception threshold of 15 dB. They analyzed various factors and found that the prescription dose to the cochlea was a significant prognostic factor.
Reduction in Cumulative Dose For FSR treatments, although dose conformality probably plays a role, the most important variable affecting hearing is cumulative dose. As shown above, reduction in total dose to acoustic tumors has resulted in higher hearing preservation rates. However, a direct prospective comparison of dose has not been published thus far. At Jefferson Hospital for Neuroscience, a study was conducted with dose reduction from 50.4 to 46.8 Gy. Tumor control rates and hearing preservation rates in these two cohorts were analyzed in 101 patients with serviceable (Gardner-Robertson Grade I or II level) treated with FSR between 1994 and 2007. Eighty-nine patients had serial audiometric data available for analysis.
71
The higher dose cohort (HDC) included 43 patients treated to 50.4 Gy with a median follow-up (latest audiogram) of 53 weeks and the lower dose cohort (LDC) included 46 patients treated to 46.8 Gy with a median follow-up of 65 weeks, both in daily fractions with a 1.8 Gy isosurface prescription. Pure tone average was significantly improved in the LDC compared to the HDC (32 dB vs. 40 dB, p = 0.023, chi square) and the difference in PTA after treatment was smaller in the LDC (10 dB vs. 15 dB change after FSR, p = 0.0577). PTA remained significantly improved in the LDC when assessing the G-R 1 pre-treatment hearing cohort (27 dB vs. 36 dB, p = 0.028, chi square) and the difference in PTA after treatment was significantly smaller in the LDC (10 dB vs. 17 dB change after FSR, p = 0.044). PTA approached a significant improvement for the G-R 2 cohort (43 dB vs. 50 dB, p = 0.09, chi square). When patients with serviceable hearing were analyzed at comparable 3 year follow-up, the raw hearing preservation rate was better in the LDC group (79% vs. 68% in the HDC) and the actuarial hearing preservation rate was significantly longer for the LDC compared to the HDC (165 weeks vs. 79 weeks, p = 0.04, logrank and p = 0.012, Wilcoxon, > Figure 71-2a). When analyzing outcomes for patients within each serviceable hearing cohort, a greater likelihood of maintaining G-R level 1 hearing was noted in the LDC with a trend approaching significance (p = 0.059, logrank and p = 0.007, Wilcoxon, > Figure 71-2b) and a significantly greater likelihood of maintaining G-R level 2 hearing was noted in the LDC (p <0.001, logrank and p = 0.014, Wilcoxon, > Figure 71-2c). Multivariate analysis revealed higher dose class and pre-treatment G-R Grade to be highly significant factors contributing to the likelihood of hearing preservation while neither age nor tumor size had any significance. All audiometric outcomes are reflected in tabular form in > Table 71-2.
1159
1160
71
Focused and conventional radiation for acoustic nerve tumors
. Figure 71-2 (a): Kaplan Meier analysis of hearing outcome in patients with pre-treatment serviceable hearing, corrected for follow-up ( 165 weeks) with significant improvement in LDC (p =0.04, logrank test). Green is low dose cohort, N = 42; red is high dose cohort, N = 31 (b): Kaplan Meier analysis of hearing outcome in patients with pre-treatment Gardner-Robertson level 1 hearing, corrected for follow-up ( 165 weeks) with a trend favoring the LDC that approached significance (p = 0.059, logrank test). Green is low dose cohort (N = 28); red is high dose cohort (N = 20). (c): Kaplan Meier analysis of hearing outcome in patients with pre-treatment Gardner-Robertson level 2 hearing, corrected for follow-up ( 165 weeks) with significant improvement in LDC (p <0.001, logrank test). Green is low dose cohort (N = 14); red is high dose cohort (N = 11)
Radiobiological Principals of FSR for Late Responding Tissues The precise mechanism of hearing loss from radiation has not been established. However, Paek et al. demonstrated that the mean dose to the cochlea was predictive of useful hearing. They found that when the mean dose exceeded 11 Gy, hearing declined [59]. Further, Massager et al. found a significant relationship with intracanalicular tumor volume (<100 vs. 100 mm3) as well
as intracanilicular integrated dose as determinants of hearing function [60]. They postulated that hearing damage resulted from increased intracanalicular pressure from inflammatory edema due to radiation. However, this has not been fully investigated. Fractionated radiotherapy has been used to treat patients with skull base tumors over the last half-century. Modern imaging and more robust radiation treatment planning software has made the treatment of skull base tumors far more
Focused and conventional radiation for acoustic nerve tumors
71
. Table 71-2 Summary of audiometric results form two FSR dose cohorts Pre-treatment Dose cohort 50.4 Gy Audiometric data (N) Serviceable [43] Serviceable G-R 1 [31] Serviceable G-R 2 [12] Corrected for follow-up (165 weeks) [30] G-R I 165 weeks [20] G-R II 165 weeks [10] 46.8 Gy Audiometric data (N) Serviceable [46] Serviceable G-R 1 [30] Serviceable G-R 2 [16] Corrected for follow- up (165 weeks) [44] G-R I 165 weeks [29] G-R II 165 weeks [15]
Post-treatment
PTA (dB)
SDS (%)
PTA
DPTA
SDS
DSDS
25 19 40 26 20 40
89 93 80 90 95 80
40 36 50 40 35 50
15 17 10 14 15 10
67 72 54 70 80 51
22 21 26 20 15 29
23 17 34 23 16 35
90 94 81 90 94 82
33a 27c 43 33 30 44
10b 10d 9 10 14 9
75 82 62 74 82 60
13 12 19 16 12 16
a p = 0.023 vs. 50.4 dose cohort, Wilcoxon, 1-way ChiSquare; bp = 0.0577 vs. 50.4 dose cohort, Wilcoxon, 1-way ChiSquare; cp = 0.028 vs. 50.4 dose cohort, Wilcoxon, 1-way ChiSquare; dp = 0.044 vs. 50.4 dose cohort, Wilcoxon, 1-way ChiSquare
precise with the ability to increase dose to the target lesion while minimizing dose to contiguous normal structures. Two exceptions are optic nerve sheath meningiomas and acoustic neuromas where these cranial nerves are intrinsic to the target volume. As special sensory cranial nerves, injury to sensory function occurs much more frequently than either sensory or motor function in mixed cranial nerves, reflecting a lower threshold for injury. Low daily doses of radiation and a cumulative dose below a threshold value, however, have proven to be safe for the optic nerves [58] and more recently for the cochlear nerve [61]. Radiation toxicity must take into account the dose delivered per fraction as well as the volume of tissue treated. The length of nerve radiated may be a function of vascular damage or of repopulation threshold. Goldsmith and colleagues, drawing from published data relating to radiation-induced optic neuropathy, were able to derive a formula in which daily dose and fraction number always
yielded a safe and equivalent dose to the optic apparatus [62]. The authors promulgated the term optic ret as a newly derived unit of nominal standard dose that incorporated total dose and the number of fractions into a single quantitative term. This optic ret formula provided a practical model and a roadmap which minimized injury to vision when the optic apparatus was in a field of radiation. Based on the published literature, as shown in > Table 71-1, high rates of tumor control have been achieved for both a single dose of 12–13 Gy or 45–50 Gy in 1.8–2 Gy daily fractions, suggesting a dose equivalence for treatment of acoustic neuromas. Unlike the optic apparatus, however, acceptably low rates of hearing loss have not yet been recognized nor has a hearing ret formula yet been derived. In order to compare different radiation schemes, a biologic equivalent dose must be defined. Utilizing a dose equivalence of 12 in a single fraction and 50.4 Gy in 1.8 Gy daily fractions,
1161
1162
71
Focused and conventional radiation for acoustic nerve tumors
and assuming BED1 = BED2 for tissue of an unknown a/b we have previously published a derivation of a biologically equivalent dose [1]: D1 ½1 þ d1=ða=bÞ ¼ D2 ½1 þ d2 =ða=bÞ or a=b ¼ ðD1 d1 D2 d2 Þ=D2 D1 For acoustic neuromas using the assumption of equivalent BED for 12 Gy in a single fraction (d1 = 12 Gy) and 46.8 Gy in 26 fractions (d2 = 1.8 Gy): a=b ¼ ð12 12Þ ð46:8 1:8Þ=ð46:8 12Þ ¼ 1:72 Gy and the corresponding BED is calculated as: BED ¼ 12ð1 þ 12=1:72Þ ¼ 46:8ð1 þ 1:8=1:72Þ ¼ 95:7 Gy Although this model could be used for other dose-fractionation schemes for treatment of acoustic tumor, its greatest utility would be its application as a predictive model of serviceable hearing loss [63]. A similar predictive model to Goldsmith [62] for hearing preservation referred to as the hearing ret model has been derived. Pan et al. proposed a model relating radiation dose to sensorineural hearing loss when various doses per fraction were prescribed, and built into the model other covariates including differences in hearing threshold between the normal and affected ear at baseline and age [61]. Within this study, the authors observed that for almost all cases in which significant hearing loss occurred in the affected ear receiving radiation, the dose was 45 Gy. Although fewer published data are available, a plot of the log of dose versus the log of the number of fractions for data with the highest hearing preservation rates, drawing from Goldsmith’s model [62], was generated [63]. The linear regression provided a formula for dose/fraction size regimens with reasonably high hearing preservation rates with a high correlation
coefficient. Drawing from similar assumptions in the optic ret model, this model postulated that dose/fraction schemes with very high hearing preservation rate could be represented by a parallel line on the same plot, intersecting a point corresponding to the dosage/fraction scheme known to be safe. Data from Pan et al. suggest that 45 Gy in 25 1.8 Gy fractions (or less) is such a point. By substitution to the linear regression equation with a lower y-intercept value, the following formula is derived: D ¼ 1180 * N0:42 or 1180 hearing ret = D * N 0.42 for dose/fraction schemes with high hearing preservation rates. For one fraction: D = 11.8 Gy. This implies that the radiosurgical dose should not exceed a threshold around 11.8 Gy to obtain high rates of hearing preservation. Based on the hearing ret formula, single fraction doses above 11.8 Gy or a cumulative FSR dose above 45 Gy could result in higher rates of hearing loss (> Figure 71-3). Although the mechanism for hearing loss remains unclear, it has been assumed that hearing loss could be related to the dose to the cochlear nerve, either directly or due to vascular injury from radiation, or to the cochlea, both discussed below. The mean value for published doses of 12 Gy have, in fact, resulted in serviceable hearing losses ranging from 17 to 67%, and cumulative FSR doses of 45–50 Gy have resulted in serviceable hearing losses ranging from 29 to 43% (> Table 71-1).
The Hearing Ret Formula and the Dose to the Cochlear Nerve None of the papers featured in > Table 71-1 provide uniform data about mean number of isocenters stratified by tumor sizes, and these data might build a relationship between such variables as tumor size, serviceable hearing loss, mean number isodose centers as a measure of conformality, and
Focused and conventional radiation for acoustic nerve tumors
. Figure 71-3 (a): Logarithmic plot of dose v. number of fractions based on published rates of hearing preservation according to the Gardner-Robertson criteria (black line); parallel line (blue) through the threshold 45 Gy point (arrowhead) established by Pan et al. [61] to establish high hearing preservation ret formula. (b): Plots of biologically equivalent doses utilizing the linear quadratic formula with an a/b ratio of 1.72; comparison with a hearing ret plot. The yellow bar signifies the threshold range of dose and fraction number yielding a high probability of hearing preservation and tumor control
dose to the cochlea. As an example, for less conformal Gamma Knife radiosurgery treatment plans, it is possible that segments of the VIII nerve are included inside the standard 50% isodose prescription volume. Therefore portions of the nerve may fall within higher dose gradient regions resulting in higher dose exposure and increased risk of serviceable hearing loss (> Figure 71-4). An important additional variable
71
at play may be the segment or length of nerve exposed, or the inclusion of the cochlea above threshold dose, discussed below.
The Hearing Ret Formula and the Dose to the Cochlea The hearing ret formula was designed around a threshold dose of 45 Gy which Pan et al. prospectively identified as the radiation dose to the cochlea associated with hearing loss [61]. In another recent analysis, Thomas and colleagues corroborated the importance of dose to the cochlea following FSR of acoustic neuroma [55]. The radiotherapy doses received by the cochlea were significantly different between the deteriorated and preserved hearing group for all cochlear dosimetric parameters measured, namely cochlear V90, V80, V50%, representing the percentage cochlear volume receiving, respectively, at least 90, 80, or 50% of the prescribed dose. This study showed that high radiation doses to the cochlea resulted in a significantly larger loss in speech reception threshold. If the percentage volume of the cochlea exposed to the V90% of the prescription dose (45 Gy) was less than 73.3%, then the median hearing loss was 10 dB. However, if the cochlear V90% was greater than or equal to 73.3%, the median hearing loss was 25 dB. We have corroborated this by measuring radiation dose delivered to the cochlea in 52 patients treated with FSR with treatment plans in which dose to the cochlea could be assessed (> Table 71-3). A mean dose of 47.3 8.7 Gy (range, 32.1– 56.8 Gy) was delivered to the cochlea in patients who lost serviceable hearing compared with a mean dose of 39.5 13.9 Gy (range, 10.7–60.8 Gy) to the cochlea in the patients who had preservation of serviceable hearing during the follow-up (p = 0.0223). Other structures at risk, including the vestibulocochlear nerve in either the internal auditory canal or the cisternal space and the cochlear nucleus
1163
1164
71
Focused and conventional radiation for acoustic nerve tumors
. Figure 71-4 Comparison of dose distribution to the cochlear nerve with Gamma Knife and FSR treatments (a): axial T-1 gadolinium enhanced MRI scan of right acoustic neuroma; (b): artists rendering of translucent acoustic tumor with cranial nerves VII and VIII cranial nerves adherent to the anterior and caudal surface of the tumor, coursing to internal auditory canal (c): eight shot Gamma Knife radiosurgery treatment plan with a 12 Gy prescription to the 50% isodose line (yellow) for right acoustic neuroma (d): magnified sagittal view of actual treatment plan in the distal porous acousticus (yellow line is 50% isodose prescription line; green line is 60% isodose line; magenta line is tumor surface). Assuming cochlear nerve is in 7:00 position, the nerve is within a 10% dose gradient above isodose prescription. (e): single isocenter Novalis FSR treatment plan with a 1.8 Gy prescription to the 90% isodose line for a right acoustic neuroma (tumor is pink; tiel line is 90% isodose prescription line); (f): magnified sagittal view of actual treatment plan in the distal porous acousticus (tumor is pink; tiel line is 90% isodose prescription line; yellow line is 95% isodose line). Assuming cochlear nerve is in 7:00 position, the nerve is within a 5% dose gradient above isodose prescription. (g): artists rendering of magnified sagittal cross section of intracanalicular portion of right acoustic tumor and contiguous VIII nerve at 7:00 position. The VIII nerve is within the prescribed isodose line and exposed to higher dose gradients. (h): plot of a focused radiation dose distribution with typical isodose prescriptions at 50% (Gamma Knife) and 90% (FSR). Yellow bar represents potential actual dose range within the spatial location of the cochlear nerve with 50% isodose prescription which is steeper; tiel bar represents potential actual dose range within the spatial location of the cochlear nerve with 90% isodose prescription which is shallower
were not significantly different in the two outcome groups.
Proposed Treatment Guidelines Since acoustic neuromas grow slowly, they often go undiagnosed for many years. Rosenberg
reported in his series the mean delay in diagnosis of acoustic neuroma was 7.3 years with one patient presenting with a unilateral hearing loss over a 50-year period [10] Thomsen and Tos in a review of 233 patients with acoustic neuromas found a mean delay in diagnosis of 7.1 years from the patient’s initial symptoms [12]. Charabi et al also found that in a review of 94 patients the
Focused and conventional radiation for acoustic nerve tumors
71
. Table 71-3 Analysis of radiation dose delivered to cochlear apparatus after FSR (n = 52)
Cochlea Dmax Dmin VII nerve Dmax Dmin Cochlear nucleus Dmax Dmin
Serviceable hearing Preserved (N = 29)
Serviceable hearing lost (N = 23)
Total
39.5 13.9 27.2 11.9
47.3 8.7* 32.3 11.0
42.9 12.4 29.4 11.7
59.3 7.3 37.6 13.6
58.2 5.2 38.1 13.0
58.8 6.5 37.8 13.2
49.6 15.2 22.3 12.0
43.0 17.8 19.2 9.8
46.7 16.5 20.9 11.1
*p-value = 0.0223
mean duration of symptoms before diagnosis was 5.3 years [64]. The typical initial symptom, gradual hearing loss, may not interfere enough with the patient’s daily living to warrant medical attention. The delay in diagnosis may also be partially due to the inability of imaging studies in the past to identify small tumors. In a patient who presents with a history of long-standing asymmetric hearing loss who has had a negative CT scan in the past, the possibility of the patient’s having an acoustic neuroma should not be discounted [10]. With the development of more sophisticated MRI images, the incidence of asymptomatic patients being diagnosed with small acoustic neuromas, especially those located within the internal auditory canal will increase. Incidental asymptomatic acoustic neuromas may be managed with observation alone. Once patients experience hearing deterioration or increase in tumor size, either radiosurgery or radiotherapy is warranted. Pretreatment audiometry is critical in establishing a baseline of overall hearing. For those patients with serviceable hearing defined as Gardner-Robertson grade I or II, fractionated radiotherapy has been associated with higher hearing preservation rates than has single fraction radiosurgery while maintaining identical tumor control rates [34]. Further, treatment with FSR can maintain
hearing within the pretreatment GardnerRobertson grade at a greater rate compared to SRS [34]. For those patients with non-serviceable hearing, the decision for SRS versus FSR is either clinician or patient dependent. Microsurgery is generally not recommended unless the tumor size is greater than 3 cm or is compressing the brainstem or after failure to maintain tumor control with radiation. The rationale, as stated above, is that microsurgery is associated with higher rates of facial and trigeminal neuropathy immediately postoperatively as well with longer term follow up. In addition, patient satisfaction and quality of life favor radiation over surgery. MRI scans are obtained at regular intervals following therapy, generally 3 months after completion of radiation. Loss of central enhancement is a common radiographic finding seen within the 6 months after treatment representing enlargement and capsular thickening. This may be transient and unlike surgery, radiation generally does not result in tumor elimination but rather shrinkage or lack of growth for slow growing tumors (> Figure 71-5). Audiometry should be monitored as well to determine precise improvement and preservation of hearing rates. Hearing improvement is the exception and some audiometric decay after treatment is the norm (> Figure 71-6). If there is no documentation of change in hearing
1165
1166
71
Focused and conventional radiation for acoustic nerve tumors
. Figure 71-5 Typical radiographic post-treatment response after FSR: (a): axial T1W gadolinium-enhanced image of right acoustic neuroma; (b): 6 months after treatment with central necrosis; (c): at 1 year with shrinkage; (d): at 3 years with further shrinkage
within the first 3 months of follow up, repeat audiograms should only be performed for new symptoms prior to the first annual check. If there is hearing deterioration of more than 15 dB in pure tone average or 30% in speech discrimination score, steroid therapy is recommended for 3 weeks (prednisone 60 mg daily)
until next follow-up visit at 1 month. Steroids have an anti-inflammatory effect as well as membrane stabilization effect, which has been employed in the transient management of peritumoral brain edema or transient tumor swelling in various brain tumors including both intra-axial and extraaxial tumors after radiosurgery [56,57,65].
Focused and conventional radiation for acoustic nerve tumors
. Figure 71-6 Composite plots of pure tone average (PTA) over time. (a): with observation; red lines denote drop in PTA below serviceable hearing during a 10 year follow-up. (b): after FSR to a total dose of 50.4 Gy cohort of patients with stable or improved within the same G-R grade were hearing improved or remained in the same G-R grade with less than 15 dB loss in PTA during the follow-up: (c); same dose cohort with documented hearing loss of greater than 20 dB with drop below serviceable hearing. Note that most cases of serviceable hearing occur within the first year
71
the tumor or the surrounding normal neurovascular structures, preventive use of steroid over the specific time period, i.e., 3–6 months after treatment, when the transient swelling or adverse effects start to appear, might reduce hearing deterioration in the patient treated with SRS or FSR. However, due to the serious side effects, their use should be very cautious and limited in short-term duration with close monitoring. Thus if there is no further progression of the hearing deterioration or tumor growth on the next follow up, steroid may be tapered off. Most patients tolerate treatment without incident, however, there are some significant though minor possible sequelae of treatment. Hydrocephalus in the absence of tumor progression has been reported in 3–11% of patients treated with either radiosurgery or FSR [34,35]. Hydrocephalus occurs at a median of 1 year, is associated with treatment of larger tumors, may resolve spontaneously or require shunting and is suspected to be secondary to proteinaceous debris blocking the flow of CSF. Tinnitus and vertigo may also become worsened after treatment.
Conclusions The treatment of vestibular schwannomas by FSR is effective, and based on literature review, is associated with decreased rates of cochlear neuropathy compared to either microsurgery or SRS. For patients with serviceable hearing, therefore, FSR should be recommended. Since both Gamma Knife and FSR treatments are widely practiced, a prospective randomized trial will be necessary to establish a standard of care. There have been several reports that document that hearing has improved after the application of steroids in patients who had experienced hearing deterioration during the followup after SRS [66,67]. Since steroids can attenuate the adverse effects caused by the radiation to
References 1. Propp JM, McCarthy BJ, Davis FG, Preston-Martin S. Descriptive epidemiology of vestibular schwannomas. Neuro Oncol 2006;8:1-11.
1167
1168
71
Focused and conventional radiation for acoustic nerve tumors
2. Cohen RJ. Acoustic neuroma: summary of the NIH consensus. Md Med J 1992;41:1128-30. 3. Lin D, Hegarty JL, Fischbein NJ, Jackler RK. The prevalence of ‘‘incidental’’ acoustic neuroma. Arch Otolaryngol Head Neck Surg 2005;131:241-4. 4. Eldridge R, Parry D. Vestibular schwannoma (acoustic neuroma). Consensus development conference. Neurosurgery 1992;30:962-4. 5. Trofatter JA, MacCollin MM, Rutter JL, Murrell JR, Duyao MP, Parry DM, Eldridge R, Kley N, Menon AG, Pulaski K, et al. A novel moesin-, ezrin-, radixin-like gene is a candidate for the neurofibromatosis 2 tumor suppressor. Cell 1993;72:791-800. 6. Woodruff J, Kourea H, Louis D, Scheithauer B. Schwannoma. In: Kleihues P, Cavenee W, editors. Pathology and Genetics: tumours of the Nervous system. Lyon: IARC Press; 2000. p. 164-6. 7. Bederson JB, von Ammon K, Wichmann WW, Yasargil MG. Conservative treatment of patients with acoustic tumors. Neurosurgery 1991;28:646-51. 8. Fucci MJ, Buchman CA, Brackmann DE, Berliner KI. Acoustic tumor growth: implications for treatment choices. Am J Otol 1999;20:495-9. 9. Mirz F, Pedersen CB, Fiirgaard B, Lundorf E. Incidence and growth pattern of vestibular schwannomas in a Danish county. Acta Otolaryngol Suppl 2000;543:30-3. 10. Rosenberg SI. Natural history of acoustic neuromas. Laryngoscope 2000;110:497-508. 11. Silverstein H, McDaniel A, Norrell H, Wazen J. Conservative management of acoustic neuroma in the elderly patient. Laryngoscope 1985;95:766-70. 12. Thomsen J, Tos M. Acoustic neuromas. Diagnostic delay, growth rate and possible non-surgical treatment. Acta Otolaryngol Suppl 1988;452:26-33. 13. Valvassori GE, Guzman M. Growth rate of acoustic neuromas. Am J Otol 1989;10:174-6. 14. Valvassori GE, Shannon M. Natural history of acoustic neuromas. Skull Base Surg 1991;1:165-7. 15. Walsh RM, Bath AP, Bance ML, Keller A, Tator CH, Rutka JA. The natural history of untreated vestibular schwannomas. Is there a role for conservative management? Rev Laryngol Otol Rhinol(Bord) 2000;121:21-6. 16. Matthies C, Samii M. Management of 1000 vestibular schwannomas (acoustic neuromas): clinical presentation. Neurosurgery 1997;40:1-9; discussion 9–10. 17. House JW, Brackmann DE. Facial nerve grading system. Otolaryngol Head Neck Surg 1985;93:146-7. 18. Gardner G, Robertson JH. Hearing preservation in unilateral acoustic neuroma surgery. Ann Otol Rhinol Laryngol 1988;97:55-66. 19. Sakamoto T, Fukuda S, Inuyama Y. Hearing loss and growth rate of acoustic neuromas in follow-up observation policy. Auris Nasus Larynx 2001;28 Suppl:S23-27. 20. Betchen SA, Walsh J, Post KD. Long-term hearing preservation after surgery for vestibular schwannoma. J Neurosurg 2005;102:6-9.
21. Danner C, Mastrodimos B, Cueva RA. A comparison of direct eighth nerve monitoring and auditory brainstem response in hearing preservation surgery for vestibular schwannoma. Otol Neurotol 2004;25:826-32. 22. Mohr G, Sade B, Dufour JJ, Rappaport JM. Preservation of hearing in patients undergoing microsurgery for vestibular schwannoma: degree of meatal filling. J Neurosurg 2005;102:1-5. 23. Gormley WB, Sekhar LN, Wright DC, Kamerer D, Schessel D. Acoustic neuromas: results of current surgical management. Neurosurgery 1997;41:50-8; discussion 58–60. 24. Somers T, Offeciers FE, Schatteman I. Results of 100 vestibular schwannoma operations. Acta Otorhinolaryngol Belg 2003;57:155-66. 25. Forster DM, Kemeny AA, Pathak A, Walton L. Radiosurgery: a minimally interventional alternative to microsurgery in the management of acoustic neuroma. Br J Neurosurg 1996;10:169-74. 26. Chan AW, Black P, Ojemann RG, Barker FG, II, Kooy HM, Lopes VV, McKenna MJ, Shrieve DC, Martuza RL, Loeffler JS. Stereotactic radiotherapy for vestibular schwannomas: favorable outcome with minimal toxicity. Neurosurgery 2005;57:60-70; discussion 60–70. 27. Regis J, Pellet W, Delsanti C, Dufour H, Roche PH, Thomassin JM, Zanaret M, Peragut JC. Functional outcome after gamma knife surgery or microsurgery for vestibular schwannomas. J Neurosurg 2002;97:1091-100. 28. Elshaikh M, Ljungman M, Ten Haken R, Lichter AS. Advances in radiation oncology. Annu Rev Med 2006;57:19-31. 29. Hirsch A, Noren G, Anderson H. Audiologic findings after stereotactic radiosurgery in nine cases of acoustic neurinomas. Acta Otolaryngol 1979;88:155-60. 30. Hirsch A, Noren G. Audiological findings after stereotactic radiosurgery in acoustic neurinomas. Acta Otolaryngol 1988;106:244-51. 31. Flickinger JCLunsford LD, Coffey RJ, Linskey ME, Bissonette DJ, Maitz AH, Kondziolka D. Radiosurgery of acoustic neurinomas. Cancer 1991;67:345-353. 32. Foote KD, Friedman WA, Buatti JM, Meeks SL, Bova FJ, Kubilis PS. Analysis of risk factors associated with radiosurgery for vestibular schwannoma. J Neurosurg 2001;95:440-49. 33. Kondziolka D, Lunsford LD, McLaughlin MR, Flickinger JC. Long-term outcomes after radiosurgery for acoustic neuromas. N Engl J Med 1998;339:1426-33. 34. Andrews DW, Suarez O, Goldman HW, Downes MB, Bednarz G, Com BW, Werner-Wasik M, Rosenstock J, Curran WJ, Jr. Stereotactic radiosurgery and fractionated stereotactic radiotherapy for the treatment of acoustic schwannomas: comparative observations of 125 patients treated at one institution. Int J Radiat Oncol Biol Phys 2001;50:1265-78. 35. Karpinos M, Teh BS, Zeck O, Carpenter LS, Phan C, Mai WY, Lu HH, Chiu JK, Butler EB, Gormley WB, Woo SY.
Focused and conventional radiation for acoustic nerve tumors
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
Treatment of acoustic neuroma: stereotactic radiosurgery vs. microsurgery. Int J Radiat Oncol Biol Phys 2002;54:1410-21. Iwai YYK, Shiotani M, Uyama T. Radiosurgery for acoustic neuromas: results of low-dose treatment. Neurosurgery 2003;53:282-8. Flickinger JC, Kondziolka D, Niranjan A, Maitz A, Voynov G, Lunsford LD. Acoustic neuroma radiosurgery with marginal tumor doses of 12 to 13 Gy. Int J Radiat Oncol Biol Phys 2004;60:225-30. van Eck AT, Horstmann GA. Increased preservation of functional hearing after gamma knife surgery for vestibular schwannoma. J Neurosurg 2005;102 Suppl:204-6. Hasegawa T, Kida Y, Kobayashi T, Yoshimoto M, Mori Y, Yoshida J. Long-term outcomes in patients with vestibular schwannomas treated using gamma knife surgery: 10-year follow up. J Neurosurg 2005;102:10-16. Kagei KSH, Suzuki K, et al. Small-field fractionated radiotherapy with or without stereotactic boost for vestibular schwannoma. Radiother Oncol 1999;50:341-7. Chung HT, Ma R, Toyota B, Clark B, Robar J, McKenzie M. Audiologic and treatment outcomes after linear accelerator-based stereotactic irradiation for acoustic neuroma. Int J Radiat Oncol Biol Phys 2004;59:1116-21. Fuss M, Debus J, Lohr F, Huber P, Rhein B, EngenhartCabillic R, Wannenmacher M. Conventionally fractionated stereotactic radiotherapy (FSRT) for acoustic neuromas. Int J Radiat Oncol Biol Phys 2000;48:1381-7. Lederman GLJ, Wertheim S, Fine M, Lombardi E, Wronski M, Arbit E. Acoustic neuroma: potential benefits of fractionated stereotactic radiosurgery. Stereotact Funct Neurosurg Suppl 1997;69:175-82. Poen JC, Golby AJ, Forster KM, Martin DP, Chinn DM, Hancock SL, Adler JR, Jr. Fractionated stereotactic radiosurgery and preservation of hearing in patients with vestibular schwannoma: a preliminary report. Neurosurgery 1999;45:1299-1305; discussion 1305–1297. Song DY, Williams JA. Fractionated stereotactic radiosurgery for treatment of acoustic neuromas. Stereotact Funct Neurosurg Suppl 1999;73:45-9. Szumacher E, Schwartz ML, Tsao M, Jaywant S, Franssen E, Wong CS, Ramasesham R, Lightstone AW, Michaels H, Hayter C, Laperviere NJ. Fractionated stereotactic radiotherapy for the treatment of vestibular schwannomas: combined experience of the Toronto-Sunnybrook Regional Cancer Centre and the Princess Margaret Hospital. Int J Radiat Oncol Biol Phys 2002;53:987-91. Varlotto JM, Shrieve DC, Alexander E, III, Kooy HM, Black PM, Loeffler JS. Fractionated stereotactic radiotherapy for the treatment of acoustic neuromas: preliminary results. Int J Radiat Oncol Biol Phys 1996;36: 141-5. Van Leeuwen JP, Cremers CW, Thewissen NP, Harhangi BS, Meijer E. Acoustic neuroma:correlation among tumor size, symptoms, and patient age. Laryngoscope 1995;105: 701-7.
71
49. Kalapurakal JA, Silverman CL, Akhtar N, Andrews DW, Downes B, Thomas PR. Improved trigeminal and facial nerve tolerance following fractionated stereotactic radiotherapy for large acoustic neuromas. Br J Radiol 1999;72:1202-7. 50. Williams JA. Fractionated stereotactic radiotherapy for acoustic neuromas. Int J Radiat Oncol Biol Phys 2002;54:500-4. 51. Sawamura Y, Shirato H, Sakamoto T, Aoyama H, Suzuki K, Onimaru R, Isu T, Fukuda S, Miyasaka K. Management of vestibular schwannoma by fractionated stereotactic radiotherapy and associated cerebrospinal fluid malabsorption. J Neurosurg 2003;99:685-92. 52. Selch MT, Pedroso A, Lee SP, Solberg TD, Agazaryan N, Cabatan-Awang C, DeSalles AA. Stereotactic radiotherapy for the treatment of acoustic neuromas. J Neurosurg 2004;101 Suppl 3:362-72. 53. Combs SE, Volk S, Schulz-Ertner D, Huber PE, Thilmann C, Debus J. Management of acoustic neuromas with fractionated stereotactic radiotherapy (FSRT): longterm results in 106 patients treated in a single institution. Int J Radiat Oncol Biol Phys 2005;63:75-81. 54. Sakamoto T, Shirato H, Takeichi N, Aoyama H, Fukuda S, Miyasaka K. Annual rate of hearing loss falls after fractionated stereotactic irradiation for vestibular schwannoma. Radiother Oncol 2001;60:45-8. 55. Thomas C, Di Maio S, Ma R, Vollans E, Chu C, Clark B, Lee R, McKenzie M, Martin M, Toyota B. Hearing preservation following fractionated stereotactic radiotherapy for vestibular schwannomas: prognostic implications of cochlear dose. J Neurosurg 2007;107:917-26. 56. Pan DH, Guo WY, Chung WY, Shiau CY, Liu RS, Lee LS. Early effects of Gamma Knife surgery on malignant and benign intracranial tumors. Stereotact Funct Neurosurg 1995;64:19-31. 57. Kalapurakal JASC, Akhtar N, Laske DW, Braitman LE, Boyko OB, Thomas PR. Intracranial meningiomas: factors that influence the development of cerebral edema after stereotactic radiosurgery and radiation therapy. Radiology 1997;204:461-5. 58. Parsons JT, Bova FJ, Fitzgerald CR, Mendenhall WM, Million RR. Radiation optic neuropathy after megavoltage external beam irradiation: analysis of time-dose factors. Int J Radiat Oncol Biol Phys 1994;30:755-63. 59. Paek SH, Chung HT, Jeong SS, Park CK, Kim CY, Kim JE, Kim DG, Jung HW. Hearing preservation after gamma knife stereotactic radiosurgery of vestibular schwannoma. Cancer 2005;104:580-90. 60. Massager N, Nissim O, Delbrouck C, Devriendt D, David P, Desmedt F, Wikler D, Hassid S, Brotchi J, Levivier M. Role of intracanalicular volumetric and dosimetric parameters on hearing preservation after vestibular schwannoma radiosurgery. Int J Radiat Oncol Biol Phys 2006;64:1331-40. 61. Pan CC, Eisbruch A, Lee JS, Snorrason RM, Ten Haken RK, Kileny PR. Prospective study of inner ear radiation
1169
1170
71
Focused and conventional radiation for acoustic nerve tumors
dose and hearing loss in head-and-neck cancer patients. Int J Radiat Oncol Biol Phys 2005;61:1393-402. 62. Goldsmith BJ, Rosenthal SA, Wara WM, Larson DA. Optic neuropathy after irradiation of meningioma. Radiology 1992;185:71-6. 63. Andrews D, Bednarz G, Downes B, Werner-Wasik M. Fractionated stereotactic radiotherapy: acoustic neuromas and other benign tumors. In: Chin L. and Regine W. editors. Principles and practice of stereotatic radiosurgery. New York: Springer Science and Business Media, LLC; 2008. p. 289-98. 64. Charabi S, Thomsen J, Mantoni M, Charabi B, Jorgensen B, Borgesen SE, Gyldensted C, Tos M. Acoustic neuroma (vestibular schwannoma): growth and surgical
and nonsurgical consequences of the wait-and-see policy. Otolaryngol Head Neck Surg 1995;113:5-14. 65. Souhami L, Olivier A, Podgorsak EB, Villemure JG, Pla M, Sadikot AF. Fractionated stereotactic radiation therapy for intracranial tumors. Cancer 1991;68: 2101-8. 66. Aronzon ARM, Bigelow DC. The efficacy of corticosteroids in restoring hearing in patients undergoing conservative management of acoustic neuromas. Otol Neurotol 2003;24:465-8. 67. Nedzelski JMCR, Kassel EE, Rowed DW, Tator CH. Is no treatment good treatment in the management of acoustic neuromas in the elderly? Laryngoscope 1986; 96:825-9.
66 Gamma Knife: Clinical Aspects L. Steiner . C. P. Yen . J. Jagannathan . D. Schlesinger . M. Steiner
Radiosurgery was defined by Lars Leksell as the technique of destroying intracranial targets through the intact skull with the use of highly focused ionizing beams. As a pupil of Olivecrona, Leksell witnessed both the successes and the failures of the pioneering endeavors of his teacher and concluded that neurosurgery should be made less traumatic. He felt that the stereotactic method introduced by Clarke and Horsley [1] in laboratory research and applied clinically by Spiegel and associates [2] was an avenue for function-preserving and less-invasive surgery. He then developed his own stereotactic system for open intervention and adapted his elegant concept of the arc device in building ‘‘Gamma Knife’’ as a tool for neurosurgeons (> Figure 66-1). Originally, Leksell’s aim was to use Gamma Knife for the treatment of functional disorders by producing necrotic lesions in specific nuclei or pathways of the brain. Subsequent development proved that Gamma Knife could be used in the management of brain tumors and cerebral vascular malformations. Subnecrotic doses were found to trigger cellular reactions in tumor cells and vasculature, leading to tumor shrinkage or control and obliteration of vascular malformations. Leksell’s original definition of radiosurgery was thus modified to include the destruction of intracranial targets and induction of desired biological effects in target tissue by the use of a single high dose of focused ionizing beams through the intact skull. The terms radiosurgery and Gamma Knife often are considered misnomers, however Leksell deliberately settled on the ambiguous terms because they articulated his concept of a tool that would be available to neurosurgeons as an #
Springer-Verlag Berlin/Heidelberg 2009
alternative to the scalpel for neurosurgical intervention. In his autobiography, Leksell wrote: ‘‘I have in my hand a new type of brain surgery, an operative system, a more sophisticated and less risky surgical procedure based on progressively improving imaging of the brain and on mechanical accuracy and modern physics, a necessary addition to classical bloody surgery’’ [3]. As a neurosurgeon with a mathematical mind, Leksell was convinced that the advancement of neurosurgery depended on the adoption of advances in other technical fields. In using ionizing beams in the radiosurgical apparatus he merely supplemented the ‘‘physical agents’’ used by neurosurgeons in their various surgical tools. Well aware of the resistance the neurosurgical establishment would mount against the new idea, he was determined to emphasize that the term Gamma Knife symbolized the neurosurgical nature of the tool. The current scope of Gamma Knife radiosurgery has been expanded to include arteriovenous malformations (AVMs), dural arteriovenous fistulas, pituitary adenomas, craniopharyngiomas, meningiomas, vestibular schwannomas, gliomas, metastatic tumors, and indications in functional neurosurgery such as intractable pain, trigeminal neuralgia, movement disorders, and intractable epilepsy. There is a growing body of data with longterm follow-up on large series of patients that are helping to define the role of Gamma Knife in the neurosurgical management of these conditions. Radiosurgery remains an integral part of neurosurgery, and with the increasing number of centers with radiosurgical facilities Leksell’s intention that a neurosurgeon should be able to choose between scalpel and Gamma Knife for a given case is being fulfilled. However, the increased availability of
1038
66
Gamma knife: clinical aspects
. Figure 66-1 Professor Lars Leksell and the first patient with vestibular schwannoma treated with Gamma Knife (1970)
in a higher incidence of lethal damage to cells, enhancing the biological efficacy of the radiation. Differential responses are based on the rate of proliferation of cells, resulting in increased sensitivity of endothelial, glial and subependymal plate cells. Vascular obliteration also seems to play a role in the death of tumor cells as well. Several of these effects are governed by volumetric considerations within the treated volume.
Effects at Tissue Level
the technique raises the issue of training and standards to ensure that this neurosurgical tool is not abused.
Radiobiology Despite the extensive clinical use of Gamma Knife over the last four decades, information about the radiobiological aspects of its use, such as the effects of a single high dose of radiation on tumor tissue, normal brain, and vascular structures remained incomplete. The current state of knowledge is summarized below.
Effects at Cellular Level Fractionated radiation results in a distribution of lethal and sublethal effects across a broad radiation field. In contrast, the small field size and sharp dose fall-off of radiosurgery permits the delivery of a high volume of radiation to a precisely defined area. The relative abundance of replicating nuclear materials in tumor tissue and the higher potential for normal brain cells to repair their DNA after sublethal injury are the keys to the differential sensitivity to radiation. In radiosurgery, the use of a single dose results
The radiation doses prescribed for classical radiotherapy have been developed from decades of clinical experience. Early in the era of clinical radiotherapy, it was observed that multiple treatments (fractions) with reduced doses per fraction improved the therapeutic ratio when treating both benign and malignant tumors. Radiotherapy uses a per-fraction prescription dose which is below the lethal dose threshold of normal tissue in the volume treated so as to purposely include a ‘‘margin’’ of normal tissue around the target lesion to account for daily setup error or subclinical disease. The time between fractions then allows this normal tissue to repair sublethal damage. The radiobiological principles which govern the design of multifraction treatments do not necessarily apply to the high-dose ionizing beams as used in radiosurgery. In contrast to radiotherapy, radiosurgery specifies a precise delivery of a high single-fraction dose of ionizing beams to a defined target volume. Normal tissue is excluded from the target volume as much as possible. The steep dose gradient at the margin of the target volume assures that normal tissue receives minimal dose and tissue inside the target periphery receives a higher dose (> Figure 66-2). Thus, repair of normal tissue during the treatment is of little concern in radiosurgery. The delivery of an inhomogeneous dose to the treatment field with a higher dose at the center of a tumor (the so-called hot spot) may be desirable for
Gamma knife: clinical aspects
. Figure 66-2 Section of brain showing a lesion in the nucleus centrum medianum 14 months after gammathalamotomy in a patient with intractable pain from terminal cancer. The sharply delineated lesion is limited to the target area. The patient was pain-free until death
several reasons. First, it offsets the relative protection offered by the poor oxygenation of the tumor core; second, it increases the cell kill in the tumor cells adjacent to those in the hot spots due to the fact that the effect of a given dose to a population of cells is more damaging if the neighboring cells receive a high dose [4].
Cranial Nerve Sensitivity The mechanisms of radiation injury to the cranial nerves are most probably secondary to damage of small vessels and protective Schwann cells
66
or oligodendroglia. There is a difference in the tolerance of different cranial nerves with sensory nerves (optic and acoustic) tolerating the least radiation and the nerves in the parasellar region, the facial nerves and the lower cranial nerves tolerating higher doses. This may be due to the fact that both the optic and acoustic nerves are actually fiber tracts of the central nervous system and carry more complex data. Clinical experience suggests that these specialized sensory nerves do not show a capacity to recover from injury. The radiosensitivity of the cranial nerves often necessitates limits on the doses given to tumors and vascular lesions in close relation to these structures. Although the precise dose tolerance of the cranial nerves is unclear, the anterior visual pathways seem to be the least radio-resistant to single doses above 8 Gy [5–7]. Hence in this situation, the distance between nerve and the lesion being treated should be carefully assessed. It appears that the risk is related to the volume of the optic apparatus receiving the dose [8–11]. A distance of 5 mm between the tumor and the optic apparatus is desirable to achieve an optimal dose fall-off, but occasionally a distance of as little as 2 mm may be acceptable due to shielding capabilities of Gamma Knife. The tolerable distance is a function of the degree to which a dose plan can be designed to deliver a suitable radiation dose to the tumor yet spare the optic apparatus. The largest experience on the radiation tolerance of cranial nerves is available for the trigeminal and facial nerves [12]. In our series of 151 patients who underwent radiosurgery for trigeminal neuralgia, 12 patients (9%) had new-onset facial numbness after treatment [13]. Nore´n and associates [12] analyzed risk factors for facial and trigeminal neuropathy in tumors receiving 12–20 Gy and concluded that the most significant factor is the length of the nerve irradiated, not the volume of tumor or dose. In treating patients with AVMs, meningiomas and secretory pituitary adenomas we have given doses of between 20–25 Gy to the cranial
1039
1040
66
Gamma knife: clinical aspects
nerves in the parasellar region without complications. Tishler et al. [14] noted that the maximum dose delivered to cranial nerves were related to neurologic deficits in 29 patients after LINAC radiosurgery and 33 patients after Gamma Knife surgery (GKS). Twelve new neuropathies were observed that were related to the nerves in the parasellar region, but they were all unrelated to a maximum dose in the range of 10–40 Gy. The conclusion of this study was that doses up to 40 Gy are relatively safe for nerves in the parasellar region. In our recently published paper on radiosurgery for Cushing’s disease, four of ten patients developed visual acuity reduction (two of whom also developed oculomotor nerve palsies) after repeat radiosurgery. Two additional patients who had previous fractionated radiation also developed oculomotor neuropathies [5]. Although three of these six patients had subsequent improvements in their visual deficits, these findings have convinced us to stop retreating pituitary adenoma patients with Gamma Knife until a safe cumulative dose can be determined.
or middle cerebral arteries 2–24 months after Gamma Knife irradiation with an 8-mm collimator and doses of 10–100 Gy [15]. In another experimental study, irradiation of the basilar arteries of cats by a stereotactic technique was performed with doses varying from 100 to 300 Gy in a gamma unit [16]. Histologically, vascular lesions such as vacuolization, degeneration, and desquamation of the endothelium and necrosis of the muscular coat predominated; reparatory reactions were relatively sparse and thrombosis was completely absent. Kamiryo et al. [17] irradiated the anterior cerebral arteries at the circle of Willis in rats with Gamma Knife. The maximum doses varied from 25 to 100 Gy. Occlusion of the anterior cerebral artery was observed in one rat 20 months after irradiation with 100 Gy. The changes included arterial wall thickening with fibrosis, splitting of the internal elastic membrane, and formation of a luminal organized thrombus. The differences between rat and human AVM studies were that more thrombus formation was observed in rat vessels and a higher dose was required to occlude the normal anterior cerebral artery in rats.
Effects on Normal Brain Vasculature
Effects on AVMs
The observations that the injury repair paradigm described above does not seem to extend to the normal vasculature in the region of radiosurgery and that the incidence of stenosis in major brain vasculature remains less than one percent are largely unexplained. There were only two patients in our experience who had clinical effects from such changes; one had a quadrantanopia, and the other had transient headaches secondary to vasogenic edema as a result of straight sinus occlusion. The incidence of such changes is low despite the use of radiation doses of 15–25 Gy at the periphery of the treated targets. In an experimental study on hypercholesterolemic rabbits, no changes were found in the basilar
The proliferation of intimal cells after the irradiation of AVMs was described by Cushing in 1928 [18]. Andersson and associates [19] observed intimal changes associated with thickened collagenous vessel walls and perivascular cuffing after they irradiated goat brains with proton beams. We examined nine AVMs specimens obtained 10–60 months after GKS [20]. The irradiated vessels displayed progressive changes that led to narrowing and obliteration of the lumen (> Figure 66-3). The earliest or least severe changes were endothelial damage and endothelial-intimal separation. These were followed by subendothelial and intimomedial proliferation of smooth muscle cells with elaboration of
Gamma knife: clinical aspects
66
. Figure 66-3 AVM vessel after GKS. Masson trichrome-stained section of a vessel from an AVM 6 months after radiosurgery (a). Note the intimal separation and the subendothelial and intimomedial proliferation of the smooth muscles. Hematoxylin and eosin-stained section of an AVM vessel after complete obliteration showing the hyalinized acellular matrix that occludes the lumen (b)
extracellular matrix components including type IV collagen; then cellular degeneration and hyaline transformation of vessel walls; and finally end-stage appearance showing complete obliteration of the vessels. The above-mentioned histopathological changes correlated with time after radiosurgery. Thus, the mechanism of obliteration of AVMs after radiosurgery is endothelial damage followed by progressive sclerosis resulting from smooth muscle proliferation and extracellular matrix deposition. The long-term effect of the process is obliteration of the vessel lumen.
Instrumentation and Technique Treatment Protocol Patients are admitted to the hospital the night before or the day of treatment. Laboratory studies and electrocardiography are performed as required. Patients who harbor AVMs require angiograms, and should have adequate renal function tests. Conditions that would preclude performing a magnetic resonance image (MRI) (i.e. pacemaker or foreign bodies that are incompatible with MRI) should be noted, so that
computed tomography (CT) scans can be used instead. For the technical aspects of GKS, please refer to Chapter 70.
Follow-up The need for adequate and thorough clinical and imaging follow-up cannot be overemphasized. A follow-up protocol should be complemented with a database of follow-up information to provide a means for analysis and conclusions at later dates. As a rule of thumb, we ask for MRI and clinical follow-up every 6 months for benign lesions and every 3 months for malignant lesions, although exceptions to this rule exist and are noted in the following sections. For AVMs, a follow-up angiogram should be performed when the AVM nidus is no longer visible on MRI. The follow-up imaging studies should be reviewed by the treating neurosurgeon and neuroradiologist to ensure the unanimity of the observation. A variety of algorithms have been implemented in software to allow the estimation of lesion volume based on MRI and CT images [21]. This can be a useful quantitative metric of treatment efficacy. However, since the margin of
1041
1042
66
Gamma knife: clinical aspects
error of volumetric imaging can be significant, we corroborate the volume measurements with the diameters of the lesion on follow-up imaging.
Indications, Specific Aspects of Technique, Imaging and Clinical Outcomes Benign Tumors Pituitary Adenomas The goals of radiosurgery in the treatment of pituitary tumors are: (1) to control tumor growth and (2) to normalize hormone overproduction in the case of secretory tumors. All patients suspected of harboring a pituitary tumor should undergo a complete neurologic, ophthalmologic, endocrinologic, and imaging workup. Each facet of the hypothalamic-pituitary-end organ axis should be assessed. Mild elevations in serum prolactin commonly result from a stalk effect while levels greater than 200 ng/mL suggest a prolactin-secreting adenoma. Thyroid function should be evaluated by measuring free thyroxine and thyroid-stimulating hormone. Adrenal function should be assessed by a morning serum cortisol and adrenocorticotropic hormone (ACTH) level. In cases of suspected Cushing’s syndrome, a 24-h urine free cortisol and a dexamethasone suppression test should be performed. Serum growth hormone (GH) and insulin-like growth factor (IGF)-1 levels should be measured to evaluate acromegaly. Imaging evaluation is achieved with thin sliced pre- and post-contrast MRI of the sellar and parasellar regions. CT may be useful to assess degree of sinus aeration and bony destruction. If a patient has a neurological deficit attributable to an adenoma, surgery is the initial treatment of choice for all tumors except a prolactinoma. Transsphenoidal surgery (endoscopic or microscopic) allows for the most rapid relief of mass effect and reduction in excessive
hormone levels in patients with Cushing’s disease and acromegaly [22–27]. This approach is associated with a low rate of complications in the hands of an experienced neurosurgeon [28]. In our observation of over 400 pituitary adenomas treated by experienced microsurgeons, macroadenomas that require radiosurgery remain macroadenomas even following the initial transsphenoidal resection. In 2000, Landolt et al. reported a significantly lower hormone normalization rate in acromegalic patients who were receiving antisecretory medications at the time of radiosurgery [29]. The precise mechanism by which antisecretory medications lower hormonal normalization rates is unknown, but it is thought to be related to changes in cell cycling caused by these drugs, potentially decreasing tumor cell radiosensitivity [29,30]. Our experience in treating 90 patients with acromegaly and 23 patients with prolactinomas found a relationship between withholding suppressive medications and endocrine remission in both secretory subtypes. As a result, we advise acromegaly and prolactinoma patients to hold suppressive medications 8 weeks before treatment, and resume 6–8 weeks after radiosurgery. However, as is true for much of neurosurgical practice, class I evidence is still unavailable to support this treatment approach. It is possible that these findings are confounded by the fact that the majority of patients with more aggressive tumors were on anti-secretory medications from the onset. Since a tumor may rarely enlarge quickly in the absence of suppressive medications, it is important that the decision to hold suppressive medications be performed on an individualized basis. Nonsecretory Adenomas
The primary goal of GKS in the treatment of nonsecretory tumors is to stabilize or reduce adenoma volume, especially for tumors in the parasellar region. In our material of 90 patients treated by GKS for nonsecretory tumors, tumor volume decreased in 59 patients (65.6%) (> Figure 66-4), remained
Gamma knife: clinical aspects
unchanged in 24 (26.7%), and increased in seven (7.8%), at a mean follow-up of 44.9 months [31]. The mean prescription dose was 18.5 Gy (Range 5–25 Gy). We found that the minimal effective prescription dose for nonsecretory tumors was 12 Gy, and doses of greater than 20 Gy did not confer an additional benefit. A total of 12 patients (19.7%) suffered a new endocrinopathy after GKS, and 25% of patients followed for more than 2 years had some hormone deficits (usually thyroid or growth hormone). Eight patients were treated with Gamma Knife as a primary therapy for medical reasons or patient preference, and in these patients, a decrease in tumor volume occurred in three (42%); five patients had stabilization of tumor volume (58%) at 34 months follow-up. In the remaining 82 patients who were treated for recurrent or residual tumors after microsurgery, a reduction in tumor volume occurred in 56 (68%), no change was detected in 19 (23%), and an increase occurred in seven (8%). The median time to tumor shrinkage on MRI was 9 months (range, 6–48 months) following GKS.
66
Tumors involving the parasellar space require special consideration. Of the 61 tumors that involved the parasellar space treated at the Lars Leksell Center, Department of Neurosurgery, University of Virginia, 39 decreased in volume (63%) and 17 remained unchanged (27%). Secretory Adenomas
Reported rates of endocrine remission following radiosurgery for functional pituitary adenomas vary greatly. The variance is likely a function of methodology, endocrine criteria used to define remission, study population, and length of followup. Most series report a higher prescription dose to patients with secreting adenomas, with a range between 20–25 Gy in most reports [5,32–34]. Because hormone normalization may be followed by relapse, we prefer the term ‘‘remission’’ over ‘‘cure.’’ > Table 66-1 summarizes the imaging and endocrinologic outcomes of 342 pituitary tumor patients treated with GKS at Lars Leksell Center. Cushing’s disease is a devastating pituitary disorder and is associated with significant morbidity and premature death. Even
Cushing’s Disease
. Figure 66-4 Sagittal T1-weighted MR image obtained before GKS demonstrates a residual nonsecretory pituitary macroadenoma following three microsurgical removals in a 34-year-old man (a). The tumor decreased significantly 30 months after GKS (b)
1043
1044
66
Gamma knife: clinical aspects
. Table 66-1 Gamma Knife surgery for pituitary adenomas Pathology (n=) Nonsecretory (90) Cushing’s (107)* Nelson’s (22) Prolactinoma (28) Acromegaly (95)
Remission rate (%)
MRI volume decrease (%)
MRI volume no change (%)
MRI volume increase (%)
Endocrinopathy following GKS (%)
18.5
NA
66
27
7
25
23
53
80
14
6
22
20 19
45 26
54 81
36 9
10 10
40** 28
22
53
72
15
3
34
Mean prescription dose (Gy)
*Tumor visible on MRI in 49 patients **Only ten patients with post-operative endocrine evaluation
after transsphenoidal surgery, up to 30% of patients may have persistent disease [26,28,35]. Most centers define an endocrine remission as a urine-free cortisol in the normal range associated with the resolution of clinical stigmata or a series of normal post-operative serum cortisol levels obtained throughout the day [10,36]. Reported endocrine remission rates following transsphenoidal surgery vary from 10 to 100%, with higher remission rates when radiosurgery follows surgical debulking [5,11,37–43]. In series with at least ten patients and a median follow-up of 2 years, endocrine remission rates range from 17 to 83% [41–44]. Ra¨hn and associates reported their experience at Karolinska Institute involving 59 patients with Cushing’s disease who were treated using Gamma Knife and followed for 2–15 years. The efficacy rate of the initial treatment was 50%, with retreatment eventually providing normalization of cortisol production in 76% of patients [45]. Our first published paper on Cushing’s disease evaluated 44 patients with a mean follow-up of 39 months [10]. In this study, the remission rate was observed to be 73% with only three late recurrences using a mean prescription dose of 22 Gy. More recently, we have updated these results with 107 patients treated at the Lars Leksell Center for Cushing’s disease. With more patients,
and a longer follow-up of 44 months, we observed a lower remission rate of 53%. In this latter series, the rate of remission was statistically correlated with tumor volume, but not tumor invasion into the cavernous sinus or suprasellar region [5]. The mean prescription dose in this series was 23 Gy, and was lower in patients who had prior radiation. Although the rate of endocrinopathy (22%) was similar to the previous series, this follow-up series was notable for 10 patients who experienced a relapse of Cushing’s disease, with a mean time to recurrence of 27 months. The differences in remission rates and recurrences between these two series demonstrate the importance of long-term follow-up in judging the true effectiveness of radiosurgery in secretory adenomas. As reported by others, the rate of hormone normalization after radiosurgery for Cushing’s disease appears to be difficult to predict, with remission occurring as early as 2 months and as late as 8 years [46,47]. About 70% of patients who have hormonal normalization do so within the first 2 years after radiosurgery. Patients with persistent disease should consider alternative treatments such as adrenalectomy, or repeat radiosurgery (although this may be associated with a higher rate of cranial nerve damage) [5].
Gamma knife: clinical aspects
A subset of Cushing’s patients does not achieve hormone normalization following microsurgery and radiosurgery, and require adrenalectomy as a ‘‘salvage’’ treatment for their disease. Although adrenalectomy is the definitive treatment for cortisol overproduction, a small subset of patients may develop Nelson’s syndrome, characterized by rapid adenoma growth, hyperpigmentation and tumor invasion into the parasellar structures [48]. There are relatively few studies detailing the results of radiosurgery for Nelson’s syndrome [25,41,49–54]. These studies report a mean prescription dose ranging from 12 to 28.7 Gy, and an endocrine remission rate ranging from 0 to 36%, however only a minority of these studies defined what was meant by endocrine remission. Our experience involves 23 patients with at least 6 months follow-up. Thirty percent of patients had a reduction in tumor size, and 60% had no change in size. A decrease in ACTH levels occurred in 67% of patients with elevated level before GKS, but normalization only occurred in four patients [52]. Pollock and Young reported on 11 patients who underwent GKS for Nelson’s syndrome. They reported control of tumor growth in 9 of 11 patients, with ACTH normalization in 4 patients (36%) [53]. Nelson’s syndrome
Just as the endocrine criteria for remission in Cushing’s disease remain the subject of debate, the criteria for remission in acromegaly have also been inconsistent. The most widely accepted guidelines for a remission in acromegaly consist of a GH level less than 1 ng/ml in response to an oral glucose challenge and a normal serum IGF-1 when matched for age and gender [55,56]. In a large series, Jezkova et al. reported a remission rate of 50% at 42 months follow-up in 96 patients with acromegaly who received radiosurgery [57]. Nearly one-third of these patients had radiosurgery as a primary treatment without previous surgical extirpation of the adenoma. Pollock et al. demonstrated a remission rate of 50% in 46 patients who were treated with Acromegaly
66
GKS for recurrent and residual tumors, with a higher remission rate in patients who were off of suppressive medications at the time of radiosurgery [58]. Our experience(accepted for publication in ‘‘Neurosurgery’’) with 135 patients with mean follow-up of 57 months demonstrates a remission rate of 59% in patients off of suppressive medications compared with 37% in patients receiving a suppressive medication (most commonly octreotide) (> Figure 66-5). In both groups, the mean prescription dose was 22 Gy (range 12–28 Gy). As a result of this finding, University of Virginia endocrinologists currently recommend a cessation of somatostatin analog medication 8 weeks before and for 8 weeks after GKS. Symptomatology of remission was noted in 85% of patients who were followed for more than 48 months, but in only 33% of patients with less than 12-months follow-up, indicating that clinical remission may take significantly longer than normalization of laboratory values. In patients with prolactinomas, the criteria to define endocrine remission are generally consistent, with most studies defining a remission as a patient who has a normal serum prolactin level while off of antisecretory medications. We use GKS as a treatment for prolactinomas after failure of medical and/or surgical treatment. Of the 23 patients treated at our institution, normalization of prolactin levels occurred in 26%, at an average time of 24.5 months, with a prescription dose of 19 Gy [34]. Consistent with the work of Landolt and Lomax, we also found that the remission rate was lower in patients receiving an anti-secretory medication at the time of GKS [30,34]. In published studies of radiosurgery for prolactinomas, the mean prescription dose has varied from 13.3 to 33 Gy, and remission rates varied from 0 to 84% [25,30,32–34,59,60]. Variations in success rate are likely related to the dose delivered to the tumor. Witt et al. noted no remission with a prescription dose of 19 Gy [11,61]. Pan et al. [62] reported a 52% endocrine Prolactinomas
1045
1046
66
Gamma knife: clinical aspects
. Figure 66-5 Coronal T1-weighted MR image obtained before GKS demonstrates a residual pituitary macroadenoma following a transsphenoidal resection in a 50-year-old woman with acromegaly (a) The latest follow-up MR image performed 4 years after GKS shows the tumor decreased in size (b) Clinically, the patient had hormone remission 2 years following GKS
‘‘cure’’ rate in a retrospective study of 128 patients in whom GKS was used as a first-line treatment for prolactinomas with a prescription dose of 30 Gy. This study is on a large sample size, and is interesting because GKS was used as a first-line treatment before medical therapy [63]. It remains to be seen if the favorable results experienced by this group can be reproduced.
Craniopharyngiomas Surgery has remained the mainstay in the treatment of craniopharyngiomas. However, a partial resection is associated with a high recurrence rate and gross total excision (a goal which can be achieved only in selected cases) carries an average recurrence rate of 16% [64]. Even in cases where in the surgeons’ assessment was that the tumor was totally resected, open-ended MRI follow-up will detect a high percentage of residuals/recurrences. Reducing the significant endocrine and visual morbidity that often accompanies a radical surgical approach necessitates the quest
for other modalities of treatment. Combined approaches have included partial resection with radiotherapy that results in a marked decrease in the recurrence rate. The results in 61 children who were treated for craniopharyngiomas at Children’s Hospital and the Joint Center for Radiation Therapy in Boston from 1970 to 1990 [65] provide valuable insights into the role of radiation in the management of craniopharyngiomas. The l0-year actuarial overall survival was 91% for all patients. The 10-year actuarial freedom from progression for the surgery group was 31% compared with 100% for patients treated with radiation therapy only and 86% for patients treated with surgery plus radiotherapy. There were two treatment-related deaths, both in the surgery plus radiotherapy group. A higher incidence of visual loss and diabetes insipidus was associated with the use of aggressive surgery. Five of six patients with tumors more than 5 cm experienced recurrences, while only six of thirty had a recurrence when the tumor was less than 5 cm.
Gamma knife: clinical aspects
A study from the Royal Marsden Hospital for a series of 173 patients with craniopharyngiomas treated between 1950 and 1986 with external beam radiotherapy either alone or after surgery [66] showed that the 10- and 20-year progression-free survival (PFS) rates were 83 and 79% and the 10- and 20-year survival rate were 77 and 66% at a median follow-up of 12 years. Survival and PFS were not found to be influenced by the extent of surgical excision. Visual field defects improved after radiotherapy in 36% of patients (38 of 106), and visual acuity improved in 30% (27 of 91). No patient developed radiation optic neuropathy. The authors concluded that limited surgery and radiotherapy achieve excellent long-term tumor control and survival with low morbidity. Personal Data
We treated 37 craniopharyngiomas in 35 patients. Of these, 3 patients had biopsies, and 22 had had prior transcranial or transsphenoidal surgeries (ranging from 1 to 6 procedures). In seven cases, intracavitary 32P or 90Yt instillation was combined with microsurgery. In one case, intracavitary instillation of 32P was performed alone following cyst aspiration. When GKS was used, the prescription doses ranged from 6 to 25 Gy (mean, 13.3 Gy). The follow-up ranged between
66
8 and 212 months with a mean of 62.5 months. Four tumors increased in size. A decrease in the solid component of the tumor was seen in 29 (> Figure 66-6) and no change was seen in 4. However, of the patients whose solid tumors decreased or remained unchanged, ten developed new or enlarged cystic component with four of them requiring further surgical resection and four receiving intracavitary 32P instillation. In total, 23 patients improved or remained stable clinically. Twelve deteriorated with ten of them died from complications of disease or surgeries. The mean 5-year survival was 71%. Review of Literature
Leksell and Liden [67] first instilled radioactive 32 P in a cystic craniopharyngioma. This patient did well initially but died 2 years later of progressive enlargement of the solid part of the tumor. Since then, intracavitary irradiation of craniopharyngioma cysts has been performed by several authors [68–77] with successful reduction in the cyst volume in 80% of patients followed for a long observation period. In our material, a total of 11 patients received intracavitary isotope instillation before or after Gamma Knife treatment. Eight of these patients had cyst shrinkage. The other three had continuous enlargement of the cyst requiring drainage.
. Figure 66-6 Residual craniopharyngioma following microsurgery on contrast enhanced T1-weighted images before (a) and 4 months after GKS (b) shows a marked decrease in the tumor size
1047
1048
66
Gamma knife: clinical aspects
In view of the fact that intracavitary irradiation could successfully deal with the cystic part, GKS of the solid part of craniopharyngiomas seemed to be an attractive option. Backlund’s first patient with craniopharyngioma treated with GKS died of a shunt malfunction 4 months after the procedure. The postmortem histology revealed that except for a small island of cells, no evidence of tumor was found. Subsequently, Backlund treated four other patients with a combined radiosurgery- intracavitary radiation approach, all of whom did well on follow-up to 3.5 years. Backlund [78] reported one case of visual loss in his material, and although it is not possible to pinpoint the exact cause of the damage of optic nerve, it has to be assumed that radiation was the cause of it. Coffey and Lunsford [79] also reported visual impairment after GKS in two patients.
Meningiomas The treatment of choice for meningiomas is microsurgical extirpation. This should be achieved with morbidities corresponding to the expectations of the patients. GKS should be considered for meningiomas involving locations where it is difficult to control neuronal or vascular structures, for residuals of skull base tumors following microsurgery, for tumors where complete microsurgical extirpation including the involved meninges was not achieved, for patients not fit for major surgery and for patients who refused surgery. Personal Material
From 1976 to 1987, Steiner and Lindquist treated thirty-one meningiomas with GKS at Karolinska Institute with long term follow-up between 10 and 21 years. Two-thirds of these tumors have either shrunk or remained stable. Among these cases were a few where only the vascular supply was targeted. (> Figures 66-7 and > 66-8) This approach has resulted in significant tumor shrinkage in the long term.
From 1989 to 2007, 750 meningiomas have been treated at Lars Leksell Center with GKS. The number of meningiomas treated until 2005 is 690. In a series of 206 patients with 1–6 years follow-up, a mean prescription dose of 14 Gy was used (range 10–20 Gy). There were 142 postmicrosurgical residuals and 64 primarily treated with Gamma Knife. Imaging follow-up was available in 151 patients (> Figures 66-9). Ninety four (63%) of the tumors shrank, forty (26%) remained unchanged and seventeen (11%) increased in size. Other centers report similar results [80–82]. Of 112 meningiomas involving the parasellar space, 68% shrank; 30% remained unchanged; and 2% increased in size. No adverse effects were experienced in treating meningiomas. However, in one case a tumor without histological diagnosis and with equivocal imaging characteristics in the pineal region was treated as a presumed meningioma and edema of bilateral basal ganglia occurred in this patient with associated cognitive disturbance. The patient had an incomplete recovery.
Vestibular Schwannomas In 1910, Henschen proposed to change the misnomer ‘‘acoustic neuroma’’ to ‘‘vestibular schwannoma’’ that correctly indicates the anatomic and histologic origin of this tumor [83]. Henschen also predicted that ENT surgeons would be interested in the management of this tumor. Today, both neurosurgeons and otolaryngologists are operating on vestibular schwannomas. The neurosurgeons prefer a suboccipital removal whereas otolaryngologists often prefer a translabyrinthine approach. Leksell and Steiner first applied Gamma Knife radiosurgery for vestibular schwannomas [84]. Steiner, at that time having doubts about the method, refused to be coauthor of the first report and was only mentioned in the acknowledgement. Leksell’s idea to treat vestibular schwannomas with
Gamma knife: clinical aspects
66
. Figure 66-7 A right parasellar meningioma treated with Gamma Knife radiosurgery. Only the nutrient vessels were targeted as defined by CT (a) and angiogram (b). The tumor remained decreased in size 18 years after the treatment (c and d)
radiosurgery turns out today to be an alternative to microsurgery. We usually use a prescription dose of 11 Gy to the isodose configuration which includes the facial nerve. Significant volumes of the tumor may in this way receive 13–15 Gy. Since March 1989–2000, we treated with Gamma Knife 200 vestibular schwannomas and from 2000 to date an additional 268 were treated.
Because we believe on principal that only series of tumors with long follow-up should be published, we published the clinical and imaging outcomes of the series we treated until 2000 [85]. In that series, follow-up periods ranging from 1 to 10 years were available in 153 patients. Followup images were analyzed using computer software to calculate lesion volumes and the clinical
1049
1050
66
Gamma knife: clinical aspects
. Figure 66-8 T1-weighted contrast enhanced MRI shows a left clinoidal meningioma pushing the left optic nerve medially (a). The GKS targeted a segment of the internal carotid artery with efferent nutrient vessels that supply the tumor. The carotid artery can be better defined with the source image from the MR angiography (b)
condition of the patients was assessed using questionnaires. GKS was the primary treatment modality in 96 cases and followed microsurgery in 57 cases. The volume of the tumors ranged from 0.02 to 18.3 cm3. In the group in which the Gamma Knife was the primary treatment, a decrease in volume was observed in 78 cases (81%) (> Figure 66-10), no change in 12 (12%) and an increase in volume in six cases (6%). The decrease was more than 75% in seven cases. In the group treated following microsurgery, a decrease in volume was observed in 37 cases (55%), no change in 14 (25%) and an increase in volume in six (11%). The decrease was more than 75% in eight cases. Five patients experienced trigeminal dysfunction; in three cases this was transient and in the other two it was persistent although there has been improvement. Three patients have facial paresthesias; in one it was transient lasting only six weeks; in one case, there was 80% recovery at 18 months
post-treatment. In the third case, the patient had surgery when the facial palsy occurred and the nerve was cut. During a 6 year period, hearing deteriorated in 60% of the patients. Three patients saw an improvement in hearing. No hearing deterioration was observed during the first 2 years of follow-up review. When the hearing was useful, it was preserved in 58% of the patients. Brackman [86] contends that following radiosurgery, the results of microsurgery are unfavorable. In the five cases treated by Slater and Brackman, one tumor had undergone prior microsurgery which would be as likely a cause of the intraoperative difficulties as a previous radiosurgery. Other neurosurgeons like Donlin Long and Pitts contend that they have no problem in operating cases following radiosurgery. It is clear that these discussions are around anecdotal observations only. For definitive conclusions, observations of large series will be necessary.
Gamma knife: clinical aspects
66
. Figure 66-9 Large left parasellar meningioma residual following microsurgery visualized on postcontrast axial and coronal MRI (a and b). Images obtained 6 months later showed that the tumor had disappeared. Repeated control axial and coronal MRI over a 5-year period showed no recurrence of the tumor (c and d)
In commenting on our results, Samii wrote that ‘‘the study is a milestone in the treatment of vestibular schwannomas and in contemporary neurosurgery,’’ however, he considered microsurgery the first choice and only in selected cases advises Gamma Knife radiosurgery [87]. Malis wrote that Gamma Knife radiosurgery ‘‘has forced reluctant neurosurgeons to consider major changes in classic thinking about the proper care of vascular malformations, cavernous sinus meningiomas and acoustic neuromas’’ [88]. He continued ‘‘I now have come to believe that over the next generation Gamma Knife ra-
diosurgery will be the mainstay of vestibular neuroma care with surgical resection being reserved for those needing urgent decompression or the very young patient’’ [88]. We contended in our report [85], ‘‘The emergence of newer approaches in the management of these and other tumors, such as gene therapy, may someday make surgery performed with the microscope or Gamma Knife obsolete. It is humbling to imagine that, in the future, the act of physically cutting out a disease process and disentangling it from normal structures, or, for that matter, treating it in situ with a burst of high
1051
1052
66
Gamma knife: clinical aspects
. Figure 66-10 Contrast enhanced T1-weighted MRI of a left vestibular schwannoma before (a) and 24 months after GKS (b). The tumor measured 11 cm3 at the time of GKS and decreased to a volume of 4.4 cm3
energy, will seem primitive, no matter how refined and glamorous we make it look today.’’ Long term outcomes in patients treated with radiosurgery or microsurgery should be thoroughly evaluated to determine the place of the different methods in the management of this challenging lesion.
and worsening of the disease in three (12%). On imaging, the schwannomas shrank in 12 patients (48%), remained stable in 10 patients (40%), and increased in size in three patients (12%). These results were the same in the group with upfront GKS and the group with microsurgery plus GKS. No tumor growth following GKS was observed in patients with neurofibromatosis.
Trigeminal Scwannomas Neurocytomas Trigeminal schwannomas are rare intracranial tumors. In the past, resection and radiation therapy were the mainstays of their treatment. More recently, neurosurgeons have begun to use radiosurgery in the treatment of trigeminal schwannomas because of its successful use in the treatment of vestibular schwannomas. We treated 26 patients with trigeminal schwannomas at Lars Leksell Center between 1989 and 2005 [89]. Five of these patients had neurofibromatosis. The median tumor volume was 3.96 cm3. The median prescription radiation dose was 15 Gy (range 10.2– 17 Gy). The mean follow-up period was 48.5 months. There was clinical improvement in 18 patients (72%), stable symptoms in four (16%),
Although considered benign tumors, neurocytomas have various biological behavior, histological patterns, and clinical courses. Microsurgical extirpation is the widely accepted upfront treatment. Rades and Fehlauer [90] compared 108 and 74 patients who underwent complete or incomplete resection without adjuvant radiotherapy; at 5 years they reported a tumor control rate of 85 and 46% in the former and latter groups, respectively. Although adjuvant therapy is generally not indicated if a total removal can be accomplished, a high recurrence rate after gross-total resection as assessed by surgeons has been reported [91,92].
Gamma knife: clinical aspects
Therefore, postoperative fractionated radiotherapy has been suggested. In the same study by Rades and Fehlauer [90], fractionated radiotherapy did not seem to provide additional benefit after complete resection. The 5-year tumor control rate with adjuvant radiotherapy increased from 46 to 83% after incomplete resection, although the treatment did not seem to improve survival rate. Given the possible adverse effect of radiotherapy on cognitive function, radiosurgery has been tried as the postoperative adjunct for neurocytoma residuals or recurrences. Between 1989 and 2004, we performed GKS in seven patients with a total of nine neurocytomas [93]. Three patients harbored five recurrent tumors after gross-total resection, three had progression of previous partially resected tumors, and one had undergone a tumor biopsy only. The mean tumor volume at the time of GKS ranged from 1.4 to 19.8 cm3 (mean 6.0 cm3). A mean prescription dose of 16 Gy (range 13–20 Gy) was prescribed to the tumor margin. After a mean follow-up period of 60 months, four of the nine tumors disappeared and four shrank significantly (> Figure 66-11). Because of secondary hemor-
66
rhage, in the remaining tumor an accurate volume could not be determined. Four patients were asymptomatic during the follow-up period, and the condition of a single patient who had residual hemiparesis from a previous transcortical resection of the tumor was stable. The patient with an intratumoral hemorrhage and rupture into ventricles required a shunt revision, and another patient died of sepsis due to a shunt infection.
Hemangioblastomas The accepted treatment for hemangioblastomas is the surgical resection of the solid component of the tumor, although radiosurgery may be a reasonable alternative for tumors in the pituitary stalk and brainstem. We have treated a total of 16 hemangioblastoma patients, 5 of whom had von Hippel Lindau disease. As with surgery, the solid portion was targeted in all of these cases. These patients were treated with a mean prescription dose of 15 Gy to the tumor margin and followed for an
. Figure 66-11 T1-weighted contrast-enhanced MR image reveals a moderately enhancing neurocytoma spanning both lateral ventricles (a). The previous surgical track can also be seen. The last follow-up image obtained 14 years later shows that the tumor has decreased in size significantly with only some residual tissue in the septum pellucidum (b)
1053
1054
66
Gamma knife: clinical aspects
average of 20.9 months. Twelve patients (75%) had a decrease in the solid component of the tumor, while four patients had no change (25%). It is not unusual for the cystic component of the tumor to grow regardless of the behavior of the solid portion. Six of sixteen patients (42%) in our series required surgery for expanding cysts. Some have hypothesized that the molecular basis for hemangioblastomas in patients von Hippel Lindau disease may make them even less sensitive to radiation, but we do not have enough patients to support or refute this observation.
although it should be noted that the value of the study is reduced by the fact that less than half of the patients had confirmatory pathology. We have follow-up of more than 1-year for 20 grade II astrocytoma patients (10 with biopsy confirmation). Three (15%) disappeared, ten (50%) shrank, two (10%) remained unchanged, and five (25%) increased in size. One patient died from progression of his disease at 46 months after treatment [95].
Brainstem Gliomas
Malignant Brain Tumors Low Grade Astrocytomas Heppner et al. reported 60 low-grade (I and II) astrocytomas treated at the Lars Leksell Center. The general indications for Gamma Knife radiosurgery were deep-seated tumors not easily approached surgically, or cases in which the patients insisted on GKS. Of 15 patients with grade I astrocytomas (6 with biopsy confirmation) with greater than 1-year follow-up, nine shrank after radiosurgery (60%). Five patients were pediatric and ten were adults. Tumor size was found to be a significant prognostic factor, with the best results found in patients with a tumor volume of less than 3 cm3, and in patients who had previous craniotomy and debulking. Failure of radiosurgery occurred in six cases (40%), and five of these patients were adults. Two patients had subsequent surgery, one for an increase in tumor size and one for hemorrhage and radiation-induced changes. In two patients, a cyst associated with the tumor enlarged in spite of the solid portion becoming smaller. One of these patients was the only patient to have a decline in neurologic function following GKS [94]. These overall findings support the fact that grade I astrocytomas in adult patients tend to behave like grade II tumors,
Brainstem gliomas deserve special mention. They have an indolent clinical course. Management in the past involved monitoring with open-ended imaging studies and shunt placement if cerebrospinal fluid diversion becomes required. However, more recently the taboo against treating well-defined brainstem gliomas with microsurgery and radiosurgery has been eliminated. We treated 22 patients with brainstem gliomas; 17 tumors were located in the midbrain, four in the pons, and one in the medulla oblongata. The selection criteria were a well-defined tumor and progression on imaging and/or deterioration in clinical condition. The mean tumor volume at the time of GKS was 2.5 cm3. A tissue diagnosis was available in only 11 cases (50%), and the remaining patients were treated based upon an appearance on imaging as well-defined small tumors. Mean prescription dose was 15 Gy (Range 10–18 Gy). Of 20 patients with greater than 12-months follow-up, the tumors disappeared in four patients (20%) (> Figure 66-12) and shrank in 12 patients (60%). Of these patients, one experienced transitory extrapyramidal symptoms and fluctuating impairment of consciousness (from somnolence to coma) for 6 months. Tumor progression occurred in four patients; of these four, one patient developed hydrocephalus requiring a ventriculoperitoneal shunt, two
Gamma knife: clinical aspects
66
showed neurological deterioration, and one patient died of tumor progression [96].
efficacy, if any, of radiosurgery in treating high grade gliomas.
High Grade Astrocytomas
Metastatic Tumors
It is difficult to accept the rationale to treat invasive and diffuse high grade gliomas with the highly focused Gamma Knife. Nevertheless, due to pressure from patients, families and referring physicians, we have treated 56 patients with malignant (grade III and IV) gliomas, with the majority of patients showing an initial decrease in tumor size. However, recurrence and progression was the general rule, with a median survival of 14 months in these cases. Because of the differences in histology and the variety of therapies and protocols available for these tumors, it is difficult to judge the benefit of GKS. However, our group, along with others [97], have observed a statistically significant prolongation of life expectancy in the group of patients undergoing aggressive multimodality treatment (e.g. including some or all of the following: radical tumor debulking, radiation therapy, chemotherapy and GKS). A carefully conducted randomized control study will ultimately be required to evaluate the
Except for solitary lesions causing mass effect, the treatment of brain metastases is primarily palliative. In the case of solitary metastases, the occurrence of long-term survival is not unheard of; but in general, the guiding philosophy is palliation, reversal of neurologic deficits and maintenance of quality of life and functional status. There has been disagreement regarding the total number and volume of tumors which can be treated using Gamma Knife radiosurgery in the instance of multiple metastases. Published data from other groups have suggested that more than three lesions should be treated using whole brain radiotherapy. However, we have experience successfully treating more than this number of tumors using radiosurgery alone in one or multiple sessions. Surgical extirpation of a solitary brain metastasis has been shown to significantly prolong survival if the primary disease is controlled. Likewise, whole-brain irradiation has been shown to be of benefit for certain tumor subtypes. These conclusions and the well-defined
. Figure 66-12 (a) Sagittal enhanced T1-weighted MRI 6 months after GKS shows a ring form enhanced tectal tumor. (b) MRI obtained 7 years after GKS shows complete tumor disappearance
1055
1056
66
Gamma knife: clinical aspects
margins of most metastatic lesions on neuroimaging studies make them amenable to GKS. Because of the high incidence of these lesions, the treatment of metastatic tumors has become one of the most common indications for GKS worldwide. Eight hundred and ninety-five patients have been treated at the Lars Leksell Center for metastatic brain tumors. Evaluation of our overall series demonstrates an overall control rate of 84%, with 10% of tumors completely disappearing on imaging follow-up, and 64% shrinking in size (> Figure 66-13). The median survival was 7.5 months, with most patients dying from systemic disease progression (compared with 4–6 months without treatment). In cases where adequate follow-up exists, we have evaluated the effectiveness of radiosurgery in treating various tumor subtypes (> Table 66-2). We have treated a total of 40 patients with 65 renal cell carcinoma brain deposits [98]. The average survival was 9.2 months after radiosurgery. A total of 41 tumors decreased in volume, 6 tumors disappeared and 16 tumors remained unchanged in size. Only two tumors increased in size following treatment. By contrast, in an unmatched control group of 119 patients that received whole brain radiation ther-
apy, patients had an average survival of 4.4 months [99]. Factors associated with longer survival included a higher presenting Karnofsky performance status, absence of extracranial metastases, adjuvant whole brain radiation therapy, and prior surgical resection. The size and the number of metastases did not have a significant effect on survival, although in cases of single metastases with controlled local disease, long survival could be achieved. We have treated a total of 90 melanoma patients, with a total of 133 tumors [100]. Forty tumors (30%) disappeared, 45 tumors (34%) shrank, 23 tumors (17%) remained unchanged in size, and 25 tumors (19%) grew. Mean prescription dose to the tumor margin was 19 Gy (Range, 12–23 Gy). The median survival was 10.4 months, with patients harboring a single lesion without visceral metastases having a better prognosis. The most common pathology that we have experience treating is lung carcinoma, with 903 metastases in 262 patients treated over the past 15 years. The median survival in patients treated with Gamma Knife was 15.4 months, compared with 14 months in patients treated with combined GKS and whole brain irradiation. Tumor control rates varied considerably with tumor size with an 84% control rate in tumors less than 0.5 mL; 95%
. Figure 66-13 Axial enhanced T1-weighted MR images demonstrating a metastatic deposit (21 cm3) from renal cell carcinoma (a). The lesion involved the midbrain, thalamus, and pineal gland. The tumor started to decrease 3 months after GKS (b) and disappeared 18 months after treatment (c)
Patients (nx)
86 40 262 90
Primary Pathology
Breast Renal Lung Melanoma
166 65 903 133
Tumors (n) 18 20 22 21
Mean prescription dose (Gy)
. Table 66-2 Gamma Knife surgery for brain metastases
10 (3–54) 7 (3–60) 12 (1–150) 9 (3–78)
MRI follow-up (mean/ range) (months) 12 9 21 17
% With disappearance 51 68 59 47
% With decrease
19 3 10 19
% With increase
18 20 10 17
% With no change
13 (3–60) 8 (3–64) 15 (1–160) 10.4 (1–82)
Survival (mean/ range) (months)
Gamma knife: clinical aspects
66 1057
1058
66
Gamma knife: clinical aspects
in tumors between 0.5 and 2mL; 89% in tumors between 2 and 4 mL; 93% in tumors between 4 and 8 mL; 86% in tumors between 8 and 14 mL and 87% in tumors greater than 14 mL. Age of less than 65 years, a Karnofsky performance score of greater than 70, controlled extracranial disease, and more than one GKS for multiple tumors were associated with increased survival. Eighty-six breast cancer patients with a total of 166 lesions had a median survival of 14 months after radiosurgery. High Karnofsky performance score and the presence of a single lesion correlated with prolonged survival [101]. In our series of 53 metastatic deposits in the brainstem [102], a mean tumor volume of 2.8 mL was treated with a mean prescription dose of 18 Gy. Follow-up was obtained in 37 patients (mean follow-up, 10 months). The tumors disappeared in seven patients, shrank in 22 and remained stable in three. Five patients experienced tumor growth. Of 35 patients with neurologic symptoms, amelioration was experienced by 21 patients, stabilization was seen in 11, while in three patients their neurological status deteriorated. The absence of extracranial disease was the only favorable prognostic factor. Previous whole brain radiation and the number of intracranial tumors were unrelated to survival length.
Tumors in the Pineal Region Tumors in the region of pineal gland and quadrigeminal plate can be treated by radiosurgery as long as they are relatively small and well demarcated. Backlund et al. [103] reported three pineocytomas, two ependymomas, three astrocytomas, one medulloblastoma, and three tumors of unknown histology in the pineal region treated with Gamma Knife. The average tumor diameter varied between 1 and 3 cm, and target doses of between 20 and 75 Gy were delivered to the lesions. The average duration of follow-up was 5 years. In three pineocytomas and two cases in which the biopsy had failed to provide the histological diagnosis, the
therapeutic results were excellent. In one ependymoma and two astrocytomas, the results were also good 1–3 years after the treatment. One ependymoma and one astrocytoma increased in size after treatment. A patient with a medulloblastoma and a patient with a tumor erroneously classified as a pineocytoma died 2 and 3 years, respectively, after treatment. We treated six pineocytomas, two pineoblastomas, two astrocytomas, two germinomas, one hemangioblastoma, and three tumors with unknown histology located in pineal region. Prescription doses ranging from 12 to 20 Gy were used in 12 patients treated with Gamma Knife alone. Prescription doses ranging from 6 to 15 Gy were used for four cases where GKS was used as a booster treatment. Thirteen tumors decreased in size. No side effects have been observed. Malignant tumors of the pineal region should be treated with conventional radiotherapy. However, if the size and shape of the malignant tumor permits the use of radiosurgery, this is likely advantageous because the results are comparable with those of radiotherapy without systemic side effects.
Chordomas We have treated a total of 19 patients with chordomas located in the clivus. Twelve of these tumors (63%) had cranial nerve involvement. The median prescription dose to the tumor margin was 16 Gy (Range, 12–23 Gy). We have follow-up longer than 2-years on 12 patients (mean follow-up 77 months). Five tumors (42%) shrank, four remained stable (33%), and three (25%) increased in size. With longer follow-up, however, the efficacy of radiosurgery in controlling tumor growth and/or recurrence is less convincing. Of four patients with longer than 5 years follow-up, only one had decreased in size, while the remaining three patients had local tumor recurrence. None of the patients had resolution of their cranial neuropathy.
Gamma knife: clinical aspects
Chondromas and Chondrosarcomas Skull base chondromas and chondrosarcomas are rare. We treated with GKS four chondromas and eight chondrosarcomas. More than 50% reduction in size was seen in two cases (both chondrosarcomas), and three cases shrank 25–50%. None progressed at follow-ups ranging from 1 to 5 years (median 3.5 years). Muthukumar et al. treated 15 patients (nine chordomas and six chondrosarcomas) with GKS. After 4 years, four of their patients had died; only two deaths were related to progression of disease, and both of these had progression outside of the treated area. Only one of the 11 surviving patients had tumor progression, and 5 had shrunk [104].
66
and proton beam therapy. The first uveal melanoma treated with Gamma Knife was in Buenos Ares. The use of GKS and its stereotactic technique requires that the eyeball is fixed relative to the stereotactic frame. This is accomplished using retrobulbar blocks and external fixative sutures which are attached to the frame. At the University of Vienna, a specially designed suture device attached to the frame is utilized [106]. We have treated five patients with uveal melanomas. At 18 months follow-up, slight shrinkage or unchanged size of the tumors were observed. When required, we place a spacer to elevate the eyelid to prevent radiation injury.
Vascular Malformations Hemangiopericytomas
Arteriovenous Malformations
Hemangiopericytomas are richly vascularized and aggressive neoplasms of mesenchymal origin. They have a predilection for both local and distant central nervous system recurrence, and a tendency to metastasize. The high recurrence rate of hemangiopericytomas during their history usually requires a combination of open surgery, radiosurgery and radiotherapy depending on the tumor’s size at the time of the recurrence. We have treated a total of 20 hemangiopericytomas in 17 patients. Mean prescription dose was 19 Gy (Range, 11–23 Gy). At a mean follow-up of 17 months post-radiosurgery, 12 tumors decreased in size, while 3 tumors remained stable. However, of the tumors that were followed for more than 36 months, 6 of 9 increased in size (67%).
After Roentgen’s discovery of X-rays, there was intense interest in the use of radiation for AVMs during the period 1914–1950, but the results were not encouraging [107–110]. This led to an almost unanimous consensus in the assessment of radiation as being worthless in the management of AVMs. With the introduction of Gamma Knife, the potential value of irradiation in vascular malformations was reassessed. Contributory factors included an increasing body of evidence that the cells constituting the vessel wall were responsive to ionizing radiation. Long-term angiographic follow-up of a small series of AVMs treated with fractionated conventional radiation by Johnson [111] in the 1950s revealed that the AVMs were obliterated in 45% of cases, although the result was never reproduced [112]. It seemed only logical that Gamma Knife be tested in the treatment of vascular malformations. In Apri1 1970, the first radiosurgical treatment for an AVM was performed by Steiner et al. at Karolinska Institute in Stockholm [113]. Since then, thousands of patients have been treated with this technique which has proven to be safe and effective.
Uveal Melanomas The most common surgical treatment for uveal melanomas is enucleation, but several centers have relatively large series in the treatment of these tumors with GKS [105,106]. Other therapeutic options include radium plaque therapy
1059
1060
66
Gamma knife: clinical aspects
Decision making of GKS for AVMs
The decision to treat an AVM radiosurgically should be based upon: 1. 2.
3. 4.
5. 6. 7. 8.
The natural history of the disease The presenting clinical symptoms: hemorrhage, epilepsy, headache, or the malformation being diagnosed incidentally The patient’s overall condition, including age, medical and neurological status The characteristics of the AVM, including its location, size, the number and pattern of feeding and draining vessels The presence or absence of a concurrent intracerebral hematoma The anticipated risk-benefit ratio of the available therapeutic alternatives The results of cerebral blood flow studies The relationship of the malformation to surrounding brain structures as determined by MRI or CT
In an AVM that has not hemorrhaged by the time of diagnosis, the risk of hemorrhage in the interval between the treatment and the subsequent cure of the patient is not high. In these cases radiosurgery is recommended, especially if the AVM is in a location that makes surgical excision hazardous to the patient. Patients with a prior history of hemorrhage are at a higher risk to rebleed at least in the first 2 years after the hemorrhage. For these cases microsurgery should not be discarded as an option since it provides immediate cure. Radiosurgery is also indicated in patients in whom other medical conditions preclude surgery or increase the risk of anesthesia or surgery. An issue that is sometimes discussed in the context of AVMs is the management of associated aneurysms. These aneurysms may be present on vessels that are unrelated to the malformation, and in these cases they should be managed as independent pathologies. However, they are sometimes present on a major feeding
vessel and may decrease in size as the AVM flow diminishes in response to radiosurgery. In this case, it is perhaps wiser to wait and observe this group of aneurysms until the AVM has been obliterated. A third group of aneurysms are intranidal aneurysms. These aneurysms disappear with the obliteration of the AVMs. There are some contentions that intranidal aneurysms are thin-walled structures that are potential sites of hemorrhage in the AVM and therefore should be embolized before radiosurgery [114]. This seems reasonable; however we observe cases where both the nidi and the peri- and intranidal aneurysms are obliterated by radiosurgery. Definition of Total, Subtotal and Partial Obliteration
Angiography following radiosurgery reveals that hemodynamic changes occur before changes in the size and shape of an AVM. The flow rate decreases progressively. Sometimes the sizes of the feeding arteries and outflow veins decrease as well. Total obliteration (> Figure 66-14) of the AVM after radiosurgery was defined by Lindquist and Steiner as ‘‘complete absence of former nidus, normalization of afferent and efferent vessels, and a normal circulation time on high-quality rapid serial subtracted angiography’’ [115]. Any remaining nidus, regardless of its size, represents partial obliteration (> Figure 66-15). Subtotal obliteration of an AVM (> Figure 66-16) means the angiographic persistence of an early filling draining vein but no demonstrable nidus. The early filling venous drainage suggests that the malformation has not been completely obliterated. Personal Material
We treated the first AVM with Gamma Knife in April 1970 [113]. In 1977, we reported 30 AVMs treated with Gamma Knife [116]; and up to date, 2,650 AVMs have been treated by the senior author. The presenting symptoms were hemorrhage (70%), seizure (16%), headache (5%), neurologi-
Gamma knife: clinical aspects
66
. Figure 66-14 Total obliteration of a midbrain AVM. Vertebral angiograms with AP and lateral views before (a and b) and 2 years after GKS (c and d). The AVM obliterated completely and patient had no neurological deficits
cal deficits (4%), and other symptoms (2%). Seventy-three percent of the malformations treated were located in deep or eloquent areas of the brain. Results
Up to March 1991, 880 patients were considered to have been optimally treated with prescription doses of 23–25 Gy. Four hundred and sixty-one of these patients had appropriate angiographic follow-up. This relatively low percentage (52%) of follow-up is explained by the fact that accord-
ing to the follow-up protocol, the majority of patients treated from March 1989 to March 1991 had not yet angiographic follow-up at the time of analysis. Complete obliteration of the AVM within 1 year after treatment occurred in 230 of 306 patients (75%). Of the 461 AVMs with satisfactory angiographic follow-up at 2 years, 369 (80%) were totally obliterated. Karlsson confirmed that the obliteration rate is dose-related. With doses less than 5 Gy, obliteration was seen
1061
1062
66
Gamma knife: clinical aspects
. Figure 66-15 Partial obliteration of an AVM. A left Sylvian fissure AVM is shown in AP and lateral views of left carotid angiograms before GKS (a and b). Same views 4 years after GKS show a decrease in the size of the nidus but persistent shunting through the partially obliterated malformation (arrowheads) (c and d)
in only 3% of cases. With a prescription dose of 5–14 Gy, 47% were obliterated. Increasing the dose up to 24 Gy resulted in an increase in the obliteration rate to 69%. With prescription doses of 25 Gy or more, obliteration rates of 88% have been achieved. The increasing benefit in terms of the obliteration rate of malformation levels out to a plateau at a prescription dose beyond 25 Gy (> Figure 66-17).
It is important to point out a concern that has been expressed by some authors about the alleged bias in reporting the results of radiosurgery. It is correct that angiographic follow-up of radiosurgically treated patient series is unsatisfactory. If the loss of follow-up angiography were to occur randomly, the remaining population would be a representative sample of the entire case material. However, if angiography is recom-
Gamma knife: clinical aspects
66
. Figure 66-16 Subtotal obliteration of an AVM. Stereotactic angiograms performed in 1991 shows a superior vermian AVM fed by bilateral cerebellar arteries and draining through the vein of Rosenthal into the straight sinus and superior vermian vein (a and b). Angiograms obtained in 1993 demonstrate the disappearance of the nidus but a persistent filling of the superior vermian vein at late arterial phase (arrowheads) (c and d)
mended only when there are no flow voids on the MRI, this introduces an inherent bias into the case material. To identify the extent of bias introduced by this fallacy, we assumed the worst-case scenario that the AVMs with MRI evidence of flow voids were still patent and added them to the unobliterated group. In the subgroup of patients who were treated with 25–30 Gy to the periphery of the malformation, this made no difference in regard to the obliteration rate. In the group of patients receiving a prescription dose of 20–24 Gy, the obliteration rate dropped
from 73.9 to 73% (statistically insignificant) [117]. Thus, although the ideal situation demands that uniform follow-up be available in all cases with an angiographic control, at least in our material, the current method provides a reliable estimate of the actual outcomes. We have published long-term neurological outcomes for 239 patients with AVMs treated with Gamma Knife between April 1970 and December 1983. Headache resolved in 65 (66%) of the 98 patients presenting with this symptom and improved in an additional 9 (9%).
1063
1064
66
Gamma knife: clinical aspects
. Figure 66-17 Dose-response curve of GKS for arteriovenous malformations. The curve illustrates the dose dependence of obliteration rate
Preexisting neurological deficits improved or disappeared completely after radiosurgery in 57% of the affected patients. The overall figure for residual neurological deficits was 3% [118]. Among the 247 patients analyzed and reported by us in 1992 [119], 59 had seizures in addition to their AVMs and 41 (69.4%) became seizurefree or improved markedly following GKS. An updated analysis of the outcome of epilepsy associated with AVM after GKS has revealed that of 178 patients with seizures associated with their AVMs before radiosurgery, 149 (84%) had improvement in their seizures. Of these, 110 (62%) were seizure-free, with 57 patients on no anticonvulsant therapy. Interestingly, 49 of the 110 patients who were cured of their seizures still had patent malformations at the last follow-up. Among the 317 pediatric patients, 106 had 2-year follow-up angiograms. Eighty-seven percent had total obliteration, 6% had subtotal obliteration, and 7% had only partial obliteration of the malformation. Embolization and Radiosurgery
Embolization has been used to shrink the nidus to a size suitable for radiosurgery. Howev-
er, it is important that the procedure do not fragment the nidus because doing so it increases the difficulty of localizing and targeting the multiple remnants and decreases the rate of total obliteration. Sometimes, the existence of embolic material also impedes a clear visualization of the nidus. Literature on the relative merits of various embolic materials is sparse and no single author has a large enough experience with different materials to draw valid conclusions. The two pertinent choices are particle materials and acrylic. Particles are believed to result in a higher rate of recanalization as compared to acrylic. However, exact differences in the rates of recanalization are not known. A new liquid embolic agent (Onyx) has been introduced recently. The agent is less adhesive and polymerizes slowly. This allows a better control of intranidal injection of the embolization material. However, the impact of this new agent upon radiosurgery awaits further evaluation. Microsurgery and Radiosurgery
In our material, GKS was used for residual nidus following microsurgery in 218 patients; 182 of them underwent follow-up angiography at 2 years and, 153 (84%) were found to be cured. Hemorrhage of Unobliterated AVMs
The issue of possible protection against hemorrhage in irradiated but still patent AVMs is highly controversial. We contend that patients, whether treated with microsurgery, radiosurgery or endovascular techniques, remain at risk for bleeding as long as the malformation is still patent. To assess the rate of hemorrhage, we calculated a probability estimate using both the personyears method and the Kaplan-Meier life table [118,119]. With the person-years method, the actual hemorrhage rate is similar to the value observed in the natural history. Analyzed using the Kaplan-Meier method, we found a risk of 3.7% per year up to 60 months post radiosurgery. Five years following the treatment, the life table ended in a plateau which could be
Gamma knife: clinical aspects
interpreted as indication of decreased risk of hemorrhage. However, this does not imply that the real risk of bleeding is negligible unless a large number of patients have been followed well into or beyond the plateau region. Karlsson studied our material to evaluate the incidence of hemorrhage during the first 2 years after radiosurgery in 1,565 patients treated before May 1992 [120]. Forty-three patients experienced hemorrhage before obliteration of their AVMs in this group. This amounts to an interval bleeding rate of 2.8% over 22 years. This risk of hemorrhage is similar to the natural history of the disease. The hemorrhage was fatal in 14 (0.9%) cases and left residual neurological deficits in 9 (0.6%). In our series of AVMs treated with GKS, we observed no recurrent hemorrhage after angiographically confirmed obliteration of a vascular malformation. In one case reported by Guo, a rebleed occurred after angiographic documentation of nidus obliteration [121]. The MRI findings suggested that hemorrhage possibly resulted from radiation-induced tissue damage. Furthermore, the histological examination of the suspected recanalized AVM revealed channels that were one-fiftieth the size of the smallest vascular channels in AVMs, making it unlikely that these were vessels with significant blood flow. This view has been further confirmed by the fact that a repeat angiogram revealed no evidence of residual malformation or recanalization. Rebleeding, in spite of posttreatment angiograms interpreted as normal, may be explained by unsatisfactory quality of the neuroimaging studies or inadequate interpretation leading to the misdiagnosis of angiographic cure [122,123]. A small residual nidus may have been missed as well because the nidus did not fill due to hemodynamic condition at the time of followup angiography. Subtotal Obliteration of AVMs (SOAVMs)
66
Subtotal obliteration of an AVM implies a complete disappearance of AVM nidus but persistence of early filling drainage veins. We reported a series of 159 patients with SOAVMs [124]. The incidence of SOAVMs was 7.6% from a total of 2,093 AVM patients who were treated with GKS and had angiographic follow-up available. The volume of the AVM nidi before GKS ranged from 0.1 to 11.5 cm3 (mean, 2.5 cm3). Sixteen patients underwent two GKS before the development of SOAVMs. The diagnosis was made after a mean of 29.4 months (range 4–178 months) following the initial GKS. Of the 175 treatments in 159 patients, the mean prescription dose was 22.5 Gy (range 15–31 Gy).
Patients and AVMs Parameters
Four patients were lost to follow-up. Of the remaining 155 patients, the clinical follow-up ranged from 5 to 185 months (mean 59.4 months). During the cumulative period of 767 patient-years (a mean of 4.9 years per patient) no SOAVM had ruptured. Follow-up angiography was performed in 90 of 136 patients in whom SOAVMs had no further treatment. These studies showed a total obliteration of the AVM as well as disappearance of the early filling vein in 66 patients (73%). Twentyfour patients (27%) had persistent SOAVMs. In patients with deeply located nidi or deep draining veins, the incidence of subsequent obliteration of SOAVMs is higher. Twenty-three patients with SOAVMs were treated with Gamma Knife targeting the proximal end of the early filling veins. In this group, follow-up angiography was performed in 19 patients, confirming disappearance of the early filling vein in 15 patients (79%) and persistent SOAVMs in four patients (21%). Compared with patients who received no further treatment, patients treated with repeat GKS had a slightly higher incidence of subsequent disappearance of the draining vein (79% compared with 73%), but the difference is not statistically significant. None of the 155 patients suffered a rupture of
Imaging and Clinical Outcomes
1065
1066
66
Gamma knife: clinical aspects
the lesion. This suggests that the protection from rebleeding at the stage of subtotal obliteration is significant. Our series shows that subtotal obliteration of the AVMs did not necessarily prove to be a premature stage of an ongoing obliteration, and instead might be the end point of the obliteration process. Earlier in our series, we repeated GKS for SOAVMs, targeting the proximal segment of the early filling vein. After repeat treatment, 79% SOAVMs were obliterated. However, the necessity of retreatment remains to be determined given the fact that in the whole group no hemorrhage occurred and that 73% of SOAVMs obliterated spontaneously. Repeat GKS for AVMs
The rationale for repeat GKS in AVMs is the persistence of risk of hemorrhage as long as the nidus remains patent. The alternatives for the management of a still patent AVM are embolization, microsurgery, or radiosurgery. Of these alternatives, surgical extirpation or radiosurgery proved to be more efficient than endovascular techniques. In location of difficult approach with microsurgery, the risk of morbidity and mortality may be high. In the following, we present our results of repeat GKS when the first treatment did not totally obliterate the malformations. One-hundred twenty patients underwent repeat GKS for a still-patent AVM at Lars Leksell Center between 1989 and 2003. The mean ages at initial and repeat GKS were 28.1 (range 4–70 years) and 32.4 years (range 8–74 years), respectively. The mean interval between the first and second GKS was 4.4 years (range 2–12.8 years). Thirty-seven patients underwent one or more embolization procedures (range 1–5). Incomplete surgical resection was carried out in 14 and a combined approach with embolization and microsurgery was performed in three. Fourteen patients had a hemorrhage following the initial Gamma Knife
Patients and AVMs Parameters
procedure over 539 risk years. The annual incidence of hemorrhage was 2.5%. The locations of the AVMs were in the cerebral hemispheres in 64, thalamus or basal ganglion in 34, corpus callosum in 6, brainstem in 12 and cerebellum in 4 patients. Ninety-four (78.3%) AVMs were located in eloquent area. Thirty-three (27.5%) AVMs only had superficial venous drainage, 87 (72.5%) had deep venous drainage. The causes of initial treatment failure were (1) inaccurate nidus definition at initial GKS in nine patients (7.5%); (2) failure to visualize some segments of nidus on angiography due to hemodynamic factors or still-existing hematoma in 11 patients (9.2%); (3) recanalization of prior embolized nidal compartment in six patients (5%); (4) suboptimal radiosurgical dose (less than 23 Gy) required because of a large targeted volume or a critical location of the nidi in 65 cases (54.2%); (5) subtotal obliteration of AVMs in 13 (10.8%); (6) unknown causes in 16 (13.3%). At the initial GKS, the maximum diameter of the nidus ranged from 8 to 50 mm (mean 26.6 mm), and the volume ranged from 0.1 to 24 cc (mean 4.0 cc). Mean prescription dose was 20.5 Gy (range 5–30 Gy). The maximum diameter of the retreated nidus ranged from 3 to 47 mm (mean 16.2 mm), and the volume ranged from 0.1 to 12.7 cc (mean 1.4 cc). Mean prescription dose of 19.6 Gy (range 4–27 Gy) was used at repeat treatment.
Treatment Parameters
Imaging and Clinical Outcome After Repeat GKS
Follow-up angiography was carried out after a mean of 45.9 months (range 12–189 months). The MRI or angiograms visualized residual nidus in 41 (34.2%) patients. Twelve (10%) patients refused a follow-up angiography in spite of absence of flow voids on the MRI. Angiography confirmed total obliteration of the nidus in 60 (50%) and subtotal obliteration in 7 (5.8%). Nine patients experienced 13 episodes of hemorrhage in 553 risk years after repeat
Gamma knife: clinical aspects
GKS (1 patient had 4 hemorrhages, 1 had 2, and 7 had 1), yielding an annual incidence of 2.4%. A lower rate of nidus obliteration was related to poor response to initial GKS, lower prescription dose, larger nidus size, and previous embolization. The clinical follow-up ranged from 18 to 237 months (mean 80.2 months). Twenty patients remained asymptomatic since their repeat GKS and forty-five improved to be symptom-free. An additional 44 improved significantly but still had residual neurological deficits. Eleven patients deteriorated; eight related to a bleed, two caused by persistent arteriovenous shunting, and one related to radiation induced changes. No GKS-related mortality has occurred in this group. One patient developed a meningioma 12 years following the initial and 7 years following the repeat GKS. Three patients developed an asymptomatic cyst at the site of the treated AVMs at 5, 7, and 7 years, respectively following the repeat treatment. GKS in Large AVMs
While satisfactory results in small and moderately sized AVMs following radiosurgery are well documented, reports on the imaging and clinical outcomes in large AVMs are sparse. The explanation presumably is that few neurosurgeons ventured the challenge or that few treated large enough series with appropriate follow-up periods. In addition, less enthusiasm to report meager results may explain the paucity of data in a field often characterized more by a deluge than scarcity of publications. The main problem with larger AVMs is due to the dependence of the obliteration response on dose and volume; this dependency requires a delicate balance in deciding an efficient dose but low enough to avoid adverse neurological deficits. The following strategies are currently available to treat large AVMs (Spetzler-Martin grade IV and V) with radiosurgery. First, one can embolize of the AVM then perform radiosurgery if the nidus shrinks to a size manageable with radiosurgery. However, embolization frequently fragments the nidus into a number of
66
segments making the radiosurgical planning difficult and increasing the probability of radiosurgery failure. Another strategy involves radiosurgery using staged dose delivery to the entire AVM volume or serial staged radiosurgery to selected volumes of the AVM. Sirin et al. used staged volumetric radiosurgery in 28 large AVMs [125]. Out of the 21 patients, seven underwent repeat radiosurgery and were eliminated from outcome analysis. Of the remaining 14 patients, 3 had total obliteration on angiograms, and 4 had no flow voids on MRI but had no follow-up angiography. Four patients had hemorrhages after radiosurgery resulting in two deaths. Worsened neurological deficits occurred in one patient. Seizure control improved in three patients and deteriorated in one. Twenty-one patients out of 28 had follow-up longer than 36 months. Sirin’s report, while flawed due to selection bias (eliminating seven cases where AVMs failed to respond to the initial treatment in other 7 patients only 3 had angiographic confirmation of nidus obliteration) still provides some pertinent information. Pan et al. [126] reported an obliteration rate of 25% for AVMs with volume larger than 15 cm3 treated with a single GKS. The obliteration rate increased to 50% at 50 months follow-up. The morbidity was 3.3%. Post-treatment hemorrhage occurred in 9.2% of cases. Miyawaki et al. observed 22 cases of radiation necrosis in a series of AVMs of comparable size to that of Pan’s series [127]. We evaluated a protocol using combined radiosurgery and microsurgery for the management of large AVMs. Radiosurgery was performed for the deep medullary portion of the AVM as a first step. The second step was planned as microsurgical extirpation of the superficial segment if the goal of the first step, obliteration of the deep segment of the AVM, was achieved. However, in no case was this goal achieved. The management of large AVMs demonstrates that every treatment has its limits. In an effort to solve the problems of the management
1067
1068
66
Gamma knife: clinical aspects
of large AVMs, a cautious approach is warranted pending the development of new techniques and agents for embolization, the development of new energy source for the focus beam therapy and the development of brain protectors. In very large AVMs perhaps ‘‘wait and see’’ may occasionally be the best management.
Dural Malformations Vascular malformations arising wholly or partially from the dura are amenable to treatment by GKS. We have treated 53 patients with dural fistulas, of whom 9 also had an associated AVM. Among 19 fistulas under 15 mm, 10 had adequate follow-up. There were seven (70%) cures and two subtotal obliterations (> Figure 66-18). In patients with malformations larger than 15 mm but smaller than 25 mm, 6 of 20 patients had adequate follow-up. Three were cured and one had subtotal obliteration. Of 14 patients with fistulas larger than 25 mm, follow-up was available in 7 patients. Five of these were cured, and one fistula was subtotally obliterated. There were three rebleeds. One of the malformations subsequently was obliterated and the other two were still patent. Radiation-induced changes appeared in three patients 8–12 months after radiosurgery but disappeared in all three by 18 months. No neurological deficits from these untoward effects were observed. We recommend radiosurgery for dural malformations that extend over a short distance or have few ‘‘holes.’’ Malformations over a long stretch with ‘‘multiple holes’’ should be managed by radiosurgery preceded by surgical and/or embolization.
Cavernous Malformations Cavernous malformations constitute the major portion of so called angiographically occult vascular malformations. They tend to have a benign course and the reported hemorrhage rates have been low in various natural history series, varying from 0.25 to 0.7% per year [128–130]. The main indications for aggressive treatment include repeated hemorrhages, progressive neurological symptoms or medically intractable seizures. Karlsson reported 23 cavernous malformations treated with Gamma Knife at Karolinska Institute, 16 by Steiner between 1985 and 1987 and 7 by Lindquist and Karlsson between 1988 and 1996 [131]. One was lost to follow-up. Nine of the twenty-two patients suffered a postradiosurgical hemorrhage and six developed a radiation-induced complication. MRI revealed a decrease in the size in three and no size change in the rest. The annual post-radiosurgical hemorrhage rate was 8%. There was a trend in the hemorrhage rate to decrease 4 years postradiosurgery. Higher prescription doses seemed to result in a lower risk of posttreatment hemorrhage. However, it could not be concluded whether radiosurgery changes the natural course of a cavernous malformation and the incidence of radiation-induced complications was approximately seven times higher than that expected if the same number of patients had been treated by GKS with the same dose for AVMs. Based on these disappointing results, we stopped treating cavernous malformations. In a number of publications, Kondziolka et al. and Hasegawa et al. studied the natural history of cerebral cavernous malformations and reduction of hemorrhage risk as well as long term results after radiosurgery for patients with cavernous malformations managed in the Department of Neurosurgery, Presbyterian Hospital, University of Pittsburgh [129,132,133].
Gamma knife: clinical aspects
66
. Figure 66-18 Total obliteration of a dural AVM following GKS. Left common carotid injection reveals the dural AVM in the area of left transverse sinus on AP and lateral projections (a and b). The AVM obliterated 2 years following GKS (c and d)
In a series of 47 patients, they observed a postradiosurgery annual hemorrhage rate of 8.8% [133] as compared to 0.6% prospective annual rate of hemorrhage in patients without a prior bleed and 4.5% annual bleed rate in patients with prior hemorrhage by studying the rate of symptomatic hemorrhage in a series of 122 nontreated patients [129]. In the group of 47 patients, they compared the post-radiosurgery hemorrhage rates with the pre-radiosurgery rate assuming that the rate could be based on an ‘‘epoch’’ starting from the first hemorrhage. This is fallacious since the malformation might have been present since birth.
Recomputed on this basis, the pre-radiosurgery annual bleeding rate comes to 5.9% which is more congruent with the expected natural history. The incidence of hemorrhage after radiosurgery appears to be higher but it is unlikely that radiosurgery increases the risk for hemorrhage. However, it seems to offer no protection or provide some protection at the high cost in terms of side effects. Hua et al., Connolly et al., and Friedman [132] while mentioning the thoughtful presentation of a controversial issue emphasized some flaws in the publications of Kondziolka et al. and Hasegawa et al. like the difference in
1069
1070
66
Gamma knife: clinical aspects
hemorrhage rate between the patients from Pittsburgh and series of Mariority et al. [134] or when using the study group as its own control. It is difficult to quantify the effects of radiosurgery without a true control population studied in parallel not in serial. Hua et al. mentioned that without an accurate quantification of the benefits of radiosurgery, a risk-to-benefit analysis in consideration of the complications of radiosurgery is not possible [132]. Karlsson assuming a less than 1% decrease in the annual risk for hemorrhage and a 20–30% complication rate states that it will take significantly more than 20 years before the beneficial effect from the treatment with radiosurgery exceeds the complication rate (personal communication). Pollock publishing the results of Gamma Knife radiosurgery in cavernous malformations treated in Mayo Clinic could not confirm the findings in publications he coauthored in Pittsburgh. He concludes that considering the complications of radiosurgery, its benefits may not be sufficiently great to warrant its use [135].
Venous Angiomas Based on our published results [136], we concluded that radiosurgery for venous anomalies (venous angiomas) does not fulfill the rigid criteria of minimal risk that must be set for the treatment of a lesion with a benign natural history. Thirteen patients with venous angiomas were treated between 1977 and 1987. In two cases, the venous angioma shared venous drainage with the associated AVMs. Cavernous malformations coexisted in two cases. Imaging follow-up was available in all but two cases. After treatment, complete obliteration of the venous angioma was observed in one patient, partial obliteration was observed in four, and no effect was found in four. In two patients where the AVMs used venous angiomas as venous drainage, only the AVMs were treated. Both
AVMs obliterated but the venous angiomas remained unchanged. Undue effects of radiation occurred in four patients; one patient had focal edema, and three had radionecrosis. Extirpation of the radionecrotic tissue 6 months after radiosurgery was necessary in one case. Literature elucidating the natural history of the venous angiomas became available only after the treatments included in the study were completed and was reviewed when we were drawing our conclusions [137]. We quoted Garner and associates [138], highlighting the statement that ‘‘surgical resection of venous anomalies is rarely indicated.’’ This fact was far from universally accepted at the time when these patients were treated.
Vein of Galen Malformations We have treated a series of nine Vein of Galen malformations (VGMs) including eight children aged 4–14 years and one adult with GKS [139]. Obliteration was achieved in four patients (three following a single treatment and one after two treatments) and partial obliteration in another three. Another patient with a significant initial response underwent repeat GKS but refused to undergo follow-up angiography after the second procedure in spite of the fact that the MRI could no longer visualize the malformation. In one case GKS and multiple embolization sessions had no effect on the size of the VGM. One patient experienced a transient neurological deficit. Another patient had evidence of a radiation-induced change on MRI, but the change was clinically silent.
Carotid-Cavernous Fistulas The ideal treatment of a spontaneous carotidcavernous fistula would consist of obliteration of the fistula with maintenance of the patency of the carotid artery. This goal is not achieved in practice by many of the current methods.
Gamma knife: clinical aspects
We have treated eight patients with carotidcavernous fistulas with Gamma Knife. The ages of these patients ranged from 27 to 70 years. Follow-up is available on six, and five of the fistulas were cured. No patient experienced any undue side effects. Since these patients usually have acute complaints that require immediate intervention, endovascular procedures are the first choice. If these procedures do not occlude the fistula, radiosurgery should be used.
Arterial Aneurysms Leksell insisted to try the GKS in arterial aneurysm. A 61-year-old lady who sustained a subarachnoid hemorrhage from a left posterior communicating artery aneurysm was treated with Gamma Knife [140]. The patient refused microsurgery, and the aneurysm was treated with two isocenters and received a prescription dose of 25 Gy. Over the next 11 months, there was a progressive decrease in size and finally obliteration of the aneurysm. The posterior communicating artery adjacent to the aneurysm also progressively narrowed in caliber and ultimately was obliterated without a neurological deficit. The patient refused a vertebral angiography, hence we cannot know whether the aneurysm would fill or not from posterior circulation. An additional 15 cases of arterial aneurysms were treated by Forster at Karolinska institute. All except one died of a hemorrhage a few weeks to months after the Gamma Knife treatment. For cases of perinidal and intranidal aneurysms associated with AVMs, Gamma Knife quite often occluded these lesions following obliteration of the AVM nidus. However, these aneurysms may increase the risk of hemorrhage during the latency period from the time of radiosurgical procedure to the resulting obliteration of the aneurysms. Therefore, they should be embolized before radiosurgery. If embolization is not feasible, radiosurgery can often obliterate these small arterial aneurysms.
66
Functional Disorders At the time when Leksell introduced radiosurgery, the term was synonymous with functional neurosurgery. Gammathalamotomies were used not only for tremor but also for intractable pain [141]. The Gasserian ganglion was irradiated for trigeminal neuralgia and gamma-capsulotomies were performed to interrupt fronto-limbic connections in the treatment of intractable anxiety and obsessive-compulsive disorders [142,143]. The introduction of better drugs, the emergence of non-ablative methods, the lack of imaging to provide the precision required by functional neurosurgery, and the fact that the placement of the lesion with Gamma Knife could not be corroborated by physiological methods, caused a decline in the use of Gamma Knife for functional disorders. However, recent developments in imaging techniques have led to a reassessment of the possibilities of functional radiosurgery.
Cancer Pain Gammathalamotomy for intractable pain was one of the first procedures performed with the Gamma Knife. Steiner et al. reported the outcome in 52 patients with terminal cancer treated with Gamma Knife for pain control [141]. Since CT scan or MRI was at that time not available, pneumoencephalography was used to target the thalamic centranum medianum (CM-Pf complex). Three by five and three by seven millimeter collimators were used. Lesions occurred in 21 of 36 patients that had postmortem examination. No lesion was observed with a maximum dose less than 140 Gy. The most effective lesions were more medially located near the wall of the third ventricle. Better results were observed for face and arm pain. Pain relief of variable degree was obtained in 26 patients and the pain relief was only temporary lasting not longer than 2–4 days. In 5 of 8 patients, a relatively satisfactory pain
1071
1072
66
Gamma knife: clinical aspects
relief lasted until the patient’s death 13, 10, 7, 4, and 1 months after the procedure, respectively. In three cases, the pain relief lasted for 9, 6, and 13 months, respectively and then returned with preoperative intensity. Hayashi et al. reported pain alleviation in patients with cancer pain and post-stroke thalamic pain following radiation of hypophysis with Gamma knife [144]. Young et al. obtained pain relief in chronic intractable pain using gammathalamotomy [145].
Trigeminal Neuralgia During his neurosurgical carrier, Leksell has been obsessed with the idea to alleviate pain. Twenty years before building the Gamma Knife, he used orthovoltage stereotactic technique to treat patients with trigeminal neuralgia (TN) and achieved long term relief of symptoms. From 1996–2003, we treated 151 cases of TN with Gamma Knife at Lars Leksell Center [13]. Radiosurgery was performed once in 136 patients, twice in 14 patients, and three times in 1 patient. One hundred twenty-two patients had typical TN, three with atypical TN, four with multiple sclerosisassociated TN, and seven with TN and a history of a cavernous sinus tumor. In each case, the chosen radiosurgical target was located 2–4 mm anterior to the entry of the trigeminal nerve into the pons. The maximum radiation doses ranged from 50 to 90 Gy. The mean time to relief of pain was 24 days (range 1–180 days). Forty-seven, Forty-five, and thirty-four percent of patients were pain free without medication at the 1-, 2-, and 3-year follow-ups, respectively. Ninety, seventy-seven, and seventy percent of patients experienced some improvement in pain at the 1-, 2-, and 3-year follow-ups, respectively. Twelve patients (9%) suffered the onset of new facial numbness after treatment. Although less effective than microvascular decompression, GKS remains a reasonable treatment option for those unwilling or unable to undergo more invasive surgical approaches.
Movement Disorders Purely out of historical interest, we mention that between 1968 and 1970, Leksell used the prototype of Gamma Knife for the production of thalamic lesions in five cases of tremor. The target was indirectly determined by using derived coordinates relative to anterior and posterior commissures as visualized by pneumoencephalography. The results were unsatisfactory. Following the introduction of stereotactic MRI, two patients with Parkinsonian tremor were treated [146]. Hirai and colleagues clarified the position, anatomic organization, and physiologic significance of the thalamus as it pertains to tremor, rigidity, and dyskinesia [147]. The correlations between neuroanatomic and electrophysiological findings in the human ventrolateral thalamic nuclei (e.g., VLa, VLp, VPLa, and VPLc) are now better understood. For GKS, the difficulty arises in identifying the VLp and VLa nuclei in the human thalamus purely by imaging. Rand observed improvement of tremor in four of seven patients with nucleus ventralis lateralis (NVL) lesion and improvement of rigidity in two (personal communication). In 4 of 8 patients treated with gamma-pallidotomy, the symptoms improved significantly. Two of three patients treated with an NVL lesion for intention tremor showed improvement. Duma et al. produced thalamic lesions in 34 patients and observed improved tremor in 63% of the patients [148]. Young et al. obtained similar results [149]. Ohye (personal communications) has used a single four millimeter isocenter and 130 Gy for gamma-thalamotomy in 27 patients with Parkinson’s disease. Tremor and/or rigidity improved in 85% of patients. Hirai (personal communications) has treated 14 patients with GKS for involuntary movement disorders. Of these 14 patients, 8 had tremor dominant Parkinson’s disease, 4 had rigidity and dyskinesia-dominant Parkinson’s disease, and 2 had essential tremors. Hirai’s target points were the VLp nucleus for
Gamma knife: clinical aspects
control of tremor and the VLa nucleus for control of rigidity and dyskinesia. Thirteen out of fourteen patients had subsequent improvement in symptoms. In nine of these patients, symptomatic improvement occurred by 50–80% in the patients’ Unified Parkinson’s Disease Rating Scale for tremor, rigidity and dyskinesia scores. Young et al. reported improvements in Unified Parkinson’s Disease Rating Scale tremor and rigidity scores in 74 out of 102 patients (73%). Of the patients with essential tremor, 88.2% became tremor-free [149]. An important change in the surgical management of movement disorders was the introduction of deep brain stimulation. Deep brain stimulation achieves amelioration of symptoms without a destructive lesion and has supplanted destructive lesions as the surgical procedure of choice.
Psychosurgery For obsessive-compulsive disorders Leksell used his open stereotactic system to target the frontolimbic connections in both anterior internal capsules and coined the term capsulotomy [150]. With the advent of Gamma Knife, he used it for psychosurgery instead of the open stereotactic tool. Mindus reported the series of patients with obsessive-compulsive disorders treated at Karolinska Institute; one group of 24 patients being treated with capsulotomy via a conventional thermocoagulation technique and followed for 1 year, another group of seven patients being treated by Gamma Knife and followed for 7 years. The clinical effects of these treatments were evaluated subjectively by two independent observers and were also rated on the Comprehensive Psychopathological Rating Scale. Ratings were performed 10 days before and 2, 6, and 12 months after surgery. The effects on the personality were evaluated by the Karolinska Scales of Personality. These scales have been developed to measure traits related to frontal lobe dysfunction and to reflect
66
different dimensions of anxiety proneness. At the 12-month follow-up, statistically and clinically significant improvement was noted in all assessments of symptomatic and psychosocial function. Freedom from symptoms or considerable improvement was noted in 79% of patients, and none were worse after the operation. A number of patients were preoperatively unable to work or function socially. Postoperatively, these patients could return to their previous occupation and to a normal social function. The results of gammacapsulotomy were found to be comparable to those of capsulotomy performed by the thermocoagulation technique.
Epilepsy Seizure was the presenting symptom in 59 of the 247 patients with AVMs of the brain treated by the senior author with Gamma Knife between 1970 and 1984. The treatment resulted in significant reduction of frequency or total content of seizures in 52 of these patients. Eleven were successfully taken off anticonvulsant medication. In three patients the seizure disorder stopped before the obliteration of AVMs. These observations prompted the idea of testing focal irradiation as a treatment modality for focal epilepsy. At University of Virginia, basic science research was done on changes in neuroexcitability after irradiation. The hippocampal slices from rats treated with the Gamma Knife were found to have a higher seizure threshold than those of controls when placed in solutions of varying concentrations of penicillin. This effect was lost at high concentrations (Henson and colleagues, personal communication, 2001). Using single doses of either 20 or 40 Gy to the hippocampus in a rat model of chronic spontaneous limbic epilepsy, a reduction in both the frequency and duration of spontaneous seizures was observed [151]. Histological evaluation of the targeted region revealed no signs of necrosis,
1073
1074
66
Gamma knife: clinical aspects
and hippocampal slice recordings revealed intact synaptically driven neuronal firing. Biochemical analysis of changes in rats brains after GKS, performed by Re´gis and colleagues, showed changes in the concentrations of excitatory and inhibitory amino acids (particularly gammaaminobutyric acid) [152]. Epilepsy has been treated with radiosurgery at many centers, but there have been few published long-term results. Barcia-Salorio and colleagues treated 11 patients with idiopathic epilepsy [153]. Complete relief from seizures was obtained in four patients, and significant reduction in seizure activity was seen in five patients. Re´gis and colleagues reported a case of mesial temporal lobe epilepsy treated with Gamma Knife [154]. They used 25 Gy given to the 50% isodose line. The patient was seizure-free after the treatment. Further results of 25 patients with medically intractable mesial temporal lobe epilepsy showed that of the 16 patients with more than 2 years follow-up, 13 were seizure-free and 2 were improved. There were three cases of nonsymptomatic visual field deficits. There was no mortality associated with the treatment [155]. The potential of a less invasive, nondestructive therapy to treat epilepsy prompted the creation of prospective European and a National Institutes of Health (NIH) sponsored multicenter studies of GKS for temporal lobe epilepsy. In the European study, three centers enrolled 21 patients with mesial temporal lobe epilepsy. The anterior parahippocampal cortex, the basal and lateral portions of the amygdala, and the anterior hippocampus were targeted, and patients received a mean dose of 24 Gy. At 2 years postradiosurgery, 65% of the patients were seizurefree. However, nine patients developed visual field deficits, and five suffered transient side effects including depression, headache, nausea, vomiting, and imbalance [155,156]. The NIH sponsored study randomized 40 patients into two dosage groups and monitored several clinical and imaging characteristics over 3 years
following radiosurgical treatment. These evaluation points include the effects on seizure frequency and severity, MR imaging, MR spectroscopy, and neuropsychological outcomes. Of four patients in the NIH study treated at Lars Leksell Center, two received a high dose (24 Gy) and became seizure free. Two cases received low dose (20 Gy). One became seizure free with auras; one had significant decrease of seizure attacks. All patients had headache and two had exacerbation of auras 9–12 months following Gamma Knife treatment. All patients had significant radiation induced changes on MRI at this period of time. One of the patients treated at another center as part of the NIH study experienced a serious adverse event that included persistent headache, visual changes, and cerebral edema, and these consequences necessitated a standard anterior temporal resection. Gamma Knife’s long-term effectiveness for epilepsy needs to be demonstrated. Also, it is unclear what underlying mechanisms are responsible for amelioration of seizures following radiosurgery. Some have suggested a ‘‘neuromodulation’’ phenomenon following GKS with accompanying glial cell reduction, stem cell migration, neuronal plasticity and sprouting, and biochemical changes [157]. Rigorous scientific studies evaluating the cellular and subcellular mechanisms responsible for improvements in epilepsy after GKS are thus far lacking.
Undue Effects The radiosurgical procedure is not associated with any immediate or short-term side effects per se. Patients sometime experience nausea. Rarely, seizures occurred in the posttreatment period, usually in patients with supratentorial lesions and who already had a history of seizure disorders. We recommend patients with seizure history maintain pre- and post-operative antiepileptic medications.
Gamma knife: clinical aspects
Radiation Induced Changes The imaging of radiation-induced changes is characterized by a bright signal on T2-weighted images on MRI (> Figure 66-19). In cases where this is associated with contrast enhancement on T1-weighted MRI, it presumably represents radiation-induced injury with an associated breakdown of the blood brain barrier. Guo reported that the radiation induced changes can be observed in 47% of cases following GKS for AVMs [158]. The onset of these changes occurred 3–15 months after the treatment in a majority of cases (92%) and more than 26 months after treatment in 8%. Progressive resolution of the radiation-induced effects is the usual course. The resolution is observed 1–17 months (mean 5 months) after the detection of changes. The clinical manifestations included headache, symptoms of raised intracranial pressure, and focal neurological deficits. In a small percentage of patients, this is associated with focal damage to neural tissue. Neurological deficits were still present at the time of the last follow-up in 3% of our patients. The nature of radiation-induced changes remains to be elucidated. These changes presumably represent a whole gamut of pathological processes, ranging from gliosis to true necrosis. It is important to emphasize that the signal changes on MRI associated with clinical deterioration are too frequently interpreted as radionecrosis despite the fact that the changes are usually transitory.
Delayed Cyst Formation A delayed cyst is defined as a collection of fluid at the site of the treated AVMs. A fluid cavity corresponding to previous hematoma or encephalomalacia should not be considered as a complication of radiosurgery. Up to date, small series of delayed cyst formation following GKS were
66
published by several authors [159,160]. Our large series of AVMs patients treated with GKS with long term follow-up provided more pertinent information about incidence and timing of cyst occurrence as well as its clinical symptoms [161]. In a total of 1,203 patients with long term MRI follow-up, 20 cysts were identified; 10 developed between 10 and 23 years, 9 between 5 and 10 years, and 1 in less than 5 years following the treatment (> Figures 66-20 and > 66-21). The incidence of cyst formation in the entire patient population was 1.6%, and 3.6% in those undergoing follow-up examination for more than 5 years. Six patients were symptomatic, including three with seizures and three with new neurological deficits. Two patients underwent craniotomy and drainage of the symptomatic cyst. In another patient, a cystoperitoneal shunt was implanted. Cyst wall specimens were obtained in two cases showing no evidence of neoplasia in either case. There are a number of hypothesis concerning the pathogenesis of the delayed cyst formation [160]. Certainly, hypotheses are important; they are like fishermen’s nets ‘‘only he who casts will catch’’ (Novalis). Nevertheless, without critical scrutiny, they are simply speculations. The validity of the causal agents in delayed cyst formation remains to be proven.
Radiosurgery Induced Neoplasia Kahan et al. [162] defined the criteria for a tumor to be considered as a result of irradiation: (1) the tumor must occur in the irradiation field; (2) it cannot be present prior to irradiation; (3) any primary tumor must differ histologically from the induced tumor; and (4) there must be no genetic predisposition for occurrence of a secondary malignancy or tumor progression. Based on the literature, the incidence of radiosurgery-induced neoplasia ranges between zero and three per 200,000 patients [49]. However, the
1075
1076
66
Gamma knife: clinical aspects
. Figure 66-19 Onset and resolution of radiation-induced changes of brain tissue surrounding an AVM nidus. Radiation induced changes appeared 6 months following GKS of a left basal ganglion AVM with a prescription dose of 20 Gy (a and b). These changes showed progressive regression (c) and a complete disappearance at 2 years following the onset (d). Angiography documented total obliteration of the AVM (not shown)
true incidence is likely to be higher because few of these 200,000 patients treated with radiosurgery were followed over a long period. Rowe presented the results of a study in which he cross-referenced patients treated with radiosurgery in Sheffield against England’s national mortality and cancer
databases [163]. With a group of 4,896 patients and more than 30,000 patient-years of data, 2 patients were found to have new malignant brain tumors. However, data covering at least 10 years was only available in 1,048 of these patients.
Gamma knife: clinical aspects
66
. Figure 66-20 Cyst formation following GKS for an AVM. The right temporal AVM visualized on the lateral carotid arteriogram (a) was cured as shown on a control angiogram obtained 2 years following GKS (b). The development of headache and personality changes prompted a MRI examination 7 years after GKS (c). The cyst was surgically decompressed (d)
We reviewed 1,333 patients with AVMs treated with Gamma Knife and followed with sequential MRI. A subset of 288 patients in this group underwent neuroimaging and participated in clinical follow up for at least 10 years [164]. In two cases, radiosurgically induced neoplasia were identified (> Figure 66-22). Each of the patients was found to have an incidental, uniformly enhancing, dura-based mass lesion; one precisely at the site of previous nidus, another one 5 cm from previous AVM. These lesions displayed
the imaging characteristics of a meningioma. Patients have been asymptomatic and refused any treatment. From our series, if we conservatively estimate that radiosurgery-induced lesions would be evident within a 10-year time interval, our incidence of radiosurgery-induced neoplasia is 2 in 2,880 person-years or 69 in 100,000 person-years. Thus, there is a 0.7% chance that a radiation-induced tumor may develop within 10 years following GKS. This is less than the 1.3%
1077
1078
66
Gamma knife: clinical aspects
. Figure 66-21 In a total of 1,203 AVM patients with long term MRI follow-up, 20 cysts were identified; one between 15 and 23 years, nine between 10 and 23 years, nine between 5 and 10 years, and one in less than 5 years following the GKS
risk over the first 10 years or 1.9% over 20 years detailed by Brada et al. in radiotherapy series [165]. However, our results encompass a follow-up period of only 10 years. It is our contention that radiation-induced neoplasia must be considered in broad terms when evaluating patients who have undergone GKS. The radiation passes through the head along as many as 201 (or 192 in Perfexion) different trajectories and distant areas of the brain are exposed to low doses of radiation. Therefore, even though the risk of radiosurgery-induced secondary tumor is low, it must be weighed in the treatment of pediatric patients and in patients with benign tumors and a long life expectancy.
Conclusion The legitimacy of radiosurgery is no longer in discussion. When Madjid Samii commenting a report on radiosurgery for vestibular schwannoma states ‘‘the study is a milestone in the treatment of vestibular schwannoma and in contemporary neurosurgery’’ [87] and when Leonard Malis writes that ‘‘the procedure has forced reluctant neurosurgeons to consider major changes in classic thinking about the proper care of many illnesses, including vascular malformations,
cavernous sinus meningiomas and acoustic neuromas’’ [88], it is time for neurosurgeons using radiosurgical tools to correct exaggerations as well as overstatements and to temper undue euphoria and unrealistic expectations. The Gamma Knife is a radiosurgical tool. Therefore, the ideal would be that only neurosurgeons who are able to remove a vestibular schwannoma or a meningioma of difficult approach, or who can extirpate a difficult AVM would have a Gamma Knife. They are the ones who would be able to make an unbiased decision on when to use the Gamma Knife instead of other neurosurgical instruments for best treatment of the given patient. Gamma Knife surgery should be used for AVMs located in the basal ganglia, brainstem and other locations of difficult approach. Microsurgery should not be easily discarded for vestibular schwannoma or skull base meningioma if the neurosurgeon is confident that he or she can perform it without damaging the patient’s quality of life. The radiosurgical and microsurgical tools are complementary and should be combined if required. When the ionizing focused beams are applied to treat lesions in children, the risk of secondary tumors should be kept in mind. To be sure, the rare risk of secondary tumor is of minor concern in adults aged 50 years or more. However a rate of 1.2% within 10 years and 1.9% within 20 years in children is not negligible. Therefore, it has to be taken into consideration at the decision making and radiosurgery should be used only in lack of other alternatives. In the past four decades, improved results with radiosurgery parallel a development in neuroimaging. This may result in more accurate definition of thalamic nuclei, in case research would win new insight in the function of the nuclei. The improved anatomic and physiological information could open new windows of opportunity for functional diseases. In the heated atmosphere of today’s ‘‘turf wars,’’ it should be remembered that the
Gamma knife: clinical aspects
66
. Figure 66-22 Antero-posterior and lateral views of vertebral angiograms demonstrate a right temporal AVM before GKS (a and b). Two years after GKS the nidus obliterated completely (c and d). Axial and coronal contrast-enhanced T1-weighted MR images (e and f) obtained 10 years postradiosurgery show a meningioma adjacent to the superior surface of tentorium. It is located in the area where the previous AVM was situated
1079
1080
66
Gamma knife: clinical aspects
radiosurgical tool can be used not only for brain lesions - currently indications are evolving for tumors in different organs. When it is used for a brain lesion, it is a neurosurgical tool and should be used by a neurosurgeon. When Walter Dandy used a pediatric cytoscope to penetrate the ventricles, he was performing a neurosurgical and not a urological intervention. If the Gamma Knife is used for a uveal melanoma, the ophthalmologist should be responsible for the procedure, and when the narrow beams are used for targeting the liver or pancreas it is not logical for the neurosurgeon to be responsible for the procedure. Leksell’s idea of radiosurgery additionally to novelty featured other ingredients characteristic of creativity, among them the ability to stimulate the minds of others for work of excellence. Over the past four decades, neurosurgeons, medical physicists, and radiation oncologists improved the Gamma Knife and adjusted the linear accelerators for radiosurgery. Currently, there are a number of sophisticated tools using focused narrow ionizing beams to treat lesions not only in the brain but also in the whole body. The litmus test of these medical tools is whether they will significantly improve the outcomes of the techniques. An in-depth investigation of this problem is overdue. The findings will suggest the line of research where spending brain and money would be most cost-effective. Imaging, software planning, pharmacology of radiation sensitizers and protective drugs for the normal tissue, ‘‘second factors’’ to enhance the effect of radiation on AVMs, and last but not least, to find a better physical agent as a source to supplant the ionizing beams of current radiosurgery are a few lines of investigation that may be considered. Following the observation of Pierre and Jacque Curie in 1880 that sound energy resulted when electric energy was applied to the surface of a quartz crystal, experiments with ultrasound energy culminated in clinical applications. Leksell, considering heat as the best physical
agent to destroy intracranial neuronal or pathological structures, by the 1950s had already investigated the possibility of using ultrasound in neurosurgery. He performed capsulotomy for obsessive-compulsive disorder using ultrasound, but erratic results and the need for craniotomy made him opt for ionizing beams as the physical agent for intracranial lesions. An integrated focused ultrasound-magnetic resonance imaging system with closed loop control of energy delivery and online tumor control has been recently developed and tested. The heat energy generated at the focus of the ultrasound beams successfully destroyed target tissue. The problem of craniotomy was addressed by Hynynen et al. [166], who were able to use ultrasound phased arrays to achieve delivery of focused ultrasound through the intact skull. Nevertheless, there remain many problems that require solutions. A detailed discussion of these challenges is beyond the scope of this conclusion, however they include heat buildup in bony structures and the protection of cranial nerves in close relation to tumors of the skull base, clinoid, parasellar space, auditory canal, foramen magnum, and foramen jugularis, which cannot tolerate the significant heat induced in the bone by the ultrasound. As long as these problems exist, ultrasound will not supplant the ionizing beams and the two physical agents will coexist and complement each other in the management of brain lesion with focused beams. One disgraceful aspect in radiosurgery and medicine in general is advertising. The widely touted statements of ‘‘noninvasive radiosurgery’’ and radiosurgery being ‘‘first choice of management for all neurosurgical indications’’ do not bear scrutiny. Ambition – while a most efficient and necessary quality for achieving work of excellence occasionally when temptation exists may result in acts of dishonesty. Like publication of inflated results, incorrect information of the patient or instead of considering the best option for a
Gamma knife: clinical aspects
given patient – performing radiosurgery always when the patient is referred specifically for radiosurgery. Until one is able to harness ambition and to quote Charles Drake, ‘‘be honest so that it hurts’’; until one is able to critically review possibilities and limitations of one’s activity; and until one avoids ‘‘business thinking’’; radiosurgery risks under-achievement and patients will be at risks.
References 1. Clarke RH, Horsley V. One method of investigating the deep ganglia and tracts of the central nervous system (cerebellum). Br Med J 1906;2:1799-800. 2. Spiegel E, Wycis H, Marks M, Lee A. Stereotaxic apparatus for operations on the human brain. Science 1947;106:349-50. 3. Leksell L. Brain fragments in radiosurgery. In: Steiner L, editor. Radiosurgery: baseline and trends. New York: Raven Press; 1992. p. 623-92. 4. Hopewell JW, Wright EA. The nature of latent cerebral irradiation damage and its modification by hypertension. Br J Radiol 1970;43:161-7. 5. Jagannathan J, Sheehan JP, Pouratian N, Laws ER, Steiner L, Vance ML. Gamma Knife surgery for Cushing’s disease. J Neurosurg 2007;106:980-7. 6. Leber KA, Bergloff J, Langmann G, Mokry M, Schrottner O, Pendl G. Radiation sensitivity of visual and oculomotor pathways. Stereotact Funct Neurosurg 1995;64 Suppl 1:233-8. 7. Leber KA, Bergloff J, Pendl G. Dose-response tolerance of the visual pathways and cranial nerves of the cavernous sinus to stereotactic radiosurgery. J Neurosurg 1998;88:43-50. 8. Chen JC, Giannotta SL, Yu C, Petrovich Z, Levy ML, Apuzzo ML. Radiosurgical management of benign cavernous sinus tumors: dose profiles and acute complications. Neurosurgery 2001;48:1022-30. 9. Lim YL, Leem W, Kim TS, Rhee BA, Kim GK. Four years’ experiences in the treatment of pituitary adenomas with gamma knife radiosurgery. Stereotact Funct Neurosurg 1998;70 Suppl 1:95-109. 10. Sheehan JM, Vance ML, Sheehan JP, Ellegala DB, Laws ER Jr. Radiosurgery for Cushing’s disease after failed transsphenoidal surgery. J Neurosurg 2000;93:738-42. 11. Witt TC, Kondziolka D, Flickinger JC, Lunsford LD. Gamma knife radiosurgery for pituitary tumors. In: Lunsford LD, Kondziolka D, Flickinger J, editors. Gamma knife brain surgery. Progress in neurological surgery, vol 14. Basel: Karger; 1998. p. 114-27.
66
12. Noren G, Greitz D, Hirsch A, Lax I. Gamma knife surgery in acoustic tumours. Acta Neurochir Suppl (Wien) 1993;58:104-107. 13. Sheehan J, Pan HC, Stroila M, Steiner L. Gamma knife surgery for trigeminal neuralgia: outcomes and prognostic factors. J Neurosurg 2005;102:434-441. 14. Tishler RB, Loeffler JS, Lunsford LD, Duma C, Alexander E III, Kooy HM, et al. Tolerance of cranial nerves of the cavernous sinus to radiosurgery. Int J Radiat Oncol Biol Phys 1993;27:215-21. 15. Kihlstrom L, Lindquist C, Adler J. Histological studies of gamma knife lesions in normal and hypercholesterolemic rabbits. New York: Raven Press; 1992. 16. Nilsson A, Wennerstrand J, Leksell D, Backlund EO. Stereotactic gamma irradiation of basilar artery in cat. Preliminary experiences. Acta Radiol Oncol Radiat Phys Biol 1978;17:150-60. 17. Kamiryo T, Lopes MB, Berr SS, Lee KS, Kassell NF, Steiner L. Occlusion of the anterior cerebral artery after Gamma Knife irradiation in a rat. Acta Neurochir (Wien) 1996;138:983-90. 18. Cushing H, Bailey P. Tumors arising from the blood vessels of the brain. Springfield, IL: Charles C Thomas; 1928. 19. Andersson B, Larsson B, Leksell L, Mair W, Rexed B, Sourander P, et al. Histopathology of late local radiolesions in the goat brain. Acta Radiol Ther Phys Biol 1970;9:385-94. 20. Schneider BF, Eberhard DA, Steiner LE. Histopathology of arteriovenous malformations after gamma knife radiosurgery. J Neurosurg 1997;87:352-7. 21. Snell JW, Sheehan J, Stroila M, Steiner L. Assessment of imaging studies used with radiosurgery: a volumetric algorithm and an estimation of its error. Technical note. J Neurosurg 2006;104:157-62. 22. Laws ER Jr, Ebersold M, Piepgras DG. The results of transsphenoidal surgery in specific clinical entities. In: Laws ER Jr, Randall R, Kern EB, editors. Management of pituitary adenomas and related lesions with emphasis on transsphenoidal microsurgery. New York: AppletonCentury-Crofts; 1982. p. 277-305. 23. Laws ER Jr, Fode NC, Redmond MJ. Transsphenoidal surgery following unsuccessful prior therapy. An assessment of benefits and risks in 158 patients. J Neurosurg 1985;63:823-9. 24. Laws ER Jr, Thapar K. Recurrent pituitary adenomas. In: Landolt AM, Vance ML, Reilly P, editors. Pituitary adenomas. Edinburgh: Churchill-Livingtone; 1996. p. 385-94. 25. Laws ER Jr, Vance ML. Radiosurgery for pituitary tumors and craniopharyngiomas. Neurosurg Clin N Am 1999;10:327-36. 26. Laws ER, Sheehan JP. Pituitary surgery: a modern approach. Basel; New York: Karger; 2006. 27. Laws ER, Vance ML, Thapar K. Pituitary surgery for the management of acromegaly. Horm Res 2000;53 3:71-5.
1081
1082
66
Gamma knife: clinical aspects
28. Ciric I, Ragin A, Baumgartner C, Pierce D. Complications of transsphenoidal surgery: results of a national survey, review of the literature, and personal experience. Neurosurgery 1997;40:225-36. 29. Landolt AM, Haller D, Lomax N, Scheib S, Schubiger O, Siegfried J, et al. Octreotide may act as a radioprotective agent in acromegaly. J Clin Endocrinol Metab 2000;85:1287-9. 30. Landolt AM, Lomax N. Gamma knife radiosurgery for prolactinomas. J Neurosurg 2000;93 Suppl 3:14-18. 31. Mingione V, Yen CP, Vance ML, Steiner M, Sheehan J, Laws ER, et al. Gamma surgery in the treatment of nonsecretory pituitary macroadenoma. J Neurosurg 2006;104:876-83. 32. Kim MS, Lee SI, Sim JH. Gamma Knife radiosurgery for functioning pituitary microadenoma. Stereotact Funct Neurosurg 1999;72 Suppl 1:119-24. 33. Kim SH, Huh R, Chang JW, Park YG, Chung SS. Gamma Knife radiosurgery for functioning pituitary adenomas. Stereotact Funct Neurosurg 1999;72 Suppl 1:101-10. 34. Pouratian N, Sheehan J, Jagannathan J, Laws ER Jr, Steiner L, Vance ML. Gamma knife radiosurgery for medically and surgically refractory prolactinomas. Neurosurgery 2006;59:255-66. 35. Mampalam TJ, Tyrrell JB, Wilson CB. Transsphenoidal microsurgery for Cushing disease. A report of 216 cases. Ann Intern Med 1988;109:487-93. 36. Nieman LK. Medical therapy of Cushing’s disease. Pituitary 2002;5:77-82. 37. Arnaldi G, Angeli A, Atkinson AB, Bertagna X, Cavagnini F, Chrousos GP, et al. Diagnosis and complications of Cushing’s syndrome: a consensus statement. J Clin Endocrinol Metab 2003;88:5593-602. 38. Chu JW, Matthias DF, Belanoff J, Schatzberg A, Hoffman AR, Feldman D. Successful long-term treatment of refractory Cushing’s disease with high-dose mifepristone (RU 486). J Clin Endocrinol Metab 2001;86:3568-73. 39. Izawa M, Hayashi M, Nakaya K, Satoh H, Ochiai T, Hori T, et al. Gamma knife radiosurgery for pituitary adenomas. J Neurosurg 2000;93 Suppl 3:19-22. 40. Jackson IM, Noren G. Gamma knife radiosurgery for pituitary tumours. Baillieres Best Pract Res Clin Endocrinol Metab 1999;13:461-9. 41. Kobayashi T, Kida Y, Mori Y. Gamma knife radiosurgery in the treatment of Cushing disease: long-term results. J Neurosurg 2002;97:422-8. 42. Morange-Ramos I, Regis J, Dufour H, Andrieu JM, Grisoli F, Jaquet P, et al. Gamma-knife surgery for secreting pituitary adenomas. Acta Neurochir (Wien) 1998;140:437-43. 43. Petrovich Z, Yu C, Giannotta SL, Zee CS, Apuzzo ML. Gamma knife radiosurgery for pituitary adenoma: early results. Neurosurgery 2003;53:51-9.
44. Mahmoud-Ahmed AS, Suh JH. Radiation therapy for Cushing’s disease: a review. Pituitary 2002;5:175-80. 45. Rahn T, Thoren M, Hall K, Backlund EO. Stereotactic radiosurgery in Cushing’s syndrome: acute radiation effects. Surg Neurol 1980;14:85-92. 46. Hayashi M, Taira T, Ochiai T, Chernov M, Takasu Y, Izawa M, et al. Gamma knife surgery of the pituitary: new treatment for thalamic pain syndrome. J Neurosurg 2005;102 38-41. 47. Hoybye C, Grenback E, Rahn T, Degerblad M, Thoren M, Hulting AL. Adrenocorticotropic hormoneproducing pituitary tumors: 12- to 22-year follow-up after treatment with stereotactic radiosurgery. Neurosurgery 2001;49:284-91; discussion 282–291. 48. Nagesser SK, van Seters AP, Kievit J, Hermans J, Krans HM, van de Velde CJ. Long-term results of total adrenalectomy for Cushing’s disease. World J Surg 2000;24:108-13. 49. Ganz JC. Gamma knife radiosurgery and its possible relationship to malignancy: a review. J Neurosurg 2002;97:644-52. 50. Ganz JC, Backlund EO, Thorsen FA. The effects of Gamma Knife surgery of pituitary adenomas on tumor growth and endocrinopathies. Stereotact Funct Neurosurg 1993;61 Suppl 1:30-7. 51. Levy RP, Fabrikant JI, Frankel KA, Phillips MH, Lyman JT, Lawrence JH, et al. Heavy-chargedparticle radiosurgery of the pituitary gland: clinical results of 840 patients. Stereotact Funct Neurosurg 1991;57:22-35. 52. Mauermann WJ, Sheehan JP, Chernavvsky DR, Laws ER, Steiner L, Vance ML. Gamma Knife surgery for adrenocorticotropic hormone-producing pituitary adenomas after bilateral adrenalectomy. J Neurosurg 2007;106:988-93. 53. Pollock BE, Young WF Jr. Stereotactic radiosurgery for patients with ACTH-producing pituitary adenomas after prior adrenalectomy. Int J Radiat Oncol Biol Phys 2002;54:839-41. 54. Wolffenbuttel BH, Kitz K, Beuls EM. Beneficial gammaknife radiosurgery in a patient with Nelson’s syndrome. Clin Neurol Neurosurg 1998;100:60-3. 55. Giustina A, Barkan A, Casanueva FF, Cavagnini F, Frohman L, Ho K, et al. Criteria for cure of acromegaly: a consensus statement. J Clin Endocrinol Metab 2000;85:526-9. 56. Vance ML. Endocrinological evaluation of acromegaly. J Neurosurg 1998;89:499-500. 57. Jezkova J, Marek J, Hana V, Krsek M, Weiss V, Vladyka V, et al. Gamma knife radiosurgery for acromegaly – longterm experience. Clin Endocrinol (Oxf) 2006;64:588-95. 58. Pollock BE, Jacob JT, Brown PD, Nippoldt TB. Radiosurgery of growth hormone-producing pituitary adenomas: factors associated with biochemical remission. J Neurosurg 2007;106:833-8.
Gamma knife: clinical aspects
59. Post KD, Habas JE. Comparison of long term results between prolactin secreting adenomas and ACTH secreting adenomas. Can J Neurol Sci 1990;17:74-7. 60. Yildiz F, Zorlu F, Erbas T, Atahan L. Radiotherapy in the management of giant pituitary adenomas. Radiother Oncol 1999;52:233-7. 61. Witt TC. Stereotactic radiosurgery for pituitary tumors. Neurosurg Focus 2003;14:e10. 62. Pan L, Zhang N, Wang EM, Wang BJ, Dai JZ, Cai PW. Gamma knife radiosurgery as a primary treatment for prolactinomas. J Neurosurg 2000;93 Suppl 3:10-13. 63. Molitch ME. Pathologic hyperprolactinemia. Endocrinol Metab Clin North Am 1992;21:877-901. 64. Ohmori K, Collins J, Fukushima T. Craniopharyngiomas in children. Pediatr Neurosurg 2007;43:265-78. 65. Hetelekidis S, Barnes PD, Tao ML, Fischer EG, Schneider L, Scott RM, et al. 20-year experience in childhood craniopharyngioma. Int J Radiat Oncol Biol Phys 1993;27:189-95. 66. Rajan B, Ashley S, Gorman C, Jose CC, Horwich A, Bloom HJ, et al. Craniopharyngioma – a long-term results following limited surgery and radiotherapy. Radiother Oncol 1993;26:1-10. 67. Leksell L, Liden K. A therapeutic trial with radioactive isotopes in cystic brain tumors. London: H. M. Stationary Office; 1951. 68. Backlund EO, Axelsson B, Bergstrand CG, Eriksson AL, Noren G, Ribbesjo E, et al. Treatment of craniopharyngiomas – the stereotactic approach in a ten to twentythree years’ perspective. I. Surgical, radiological and ophthalmological aspects. Acta Neurochir (Wien) 1989;99:11-19. 69. Bond WH, Richards D, Turner E. Experiences with radioactive gold in the treatment of craniopharyngioma. J Neurol Neurosurg Psychiatry 1965;28:30-8. 70. Guevara JA, Bunge HJ, Heinrich JJ, Weller G, Villasante A, Chinela AB. Cystic craniopharyngioma treated by 90yttrium silicate colloid. Acta Neurochir Suppl (Wien) 1988;42:109-12. 71. Julow J, Lanyi F, Hajda M, Szeifert G, Simkovics M, Toth S, et al. Further experiences in the treatment of cystic craniopharyngeomas with yttrium 90 silicate colloid. Acta Neurochir Suppl (Wien) 1988;42:113-19. 72. Kobayashi T, Kageyama N, Ohara K. Internal irradiation for cystic craniopharyngioma. J Neurosurg 1981; 55:896-903. 73. Leksell L, Backlund EO, Johansson L. Treatment of craniopharyngiomas. Acta Chir Scand 1967; 133:345-50. 74. Lindgren E, Westberg G. Radioactive bismuth phosphate for the treatment of craniopharyngioma. Acta Radiol Ther Phys Biol 1964;2:113-20. 75. Munari C, Landre E, Musolino A, Turak B, Habert MO, Chodkiewicz JP. Long term results of stereotactic
76.
77.
78.
79.
80.
81.
82.
83.
84. 85. 86.
87. 88. 89.
90. 91.
92.
66
endocavitary beta irradiation of craniopharyngioma cysts. J Neurosurg Sci 1989;33:99-105. Strauss L, Sturm V, Georgi P, Schlegel W, Ostertag H, Clorius JH, et al. Radioisotope therapy of cystic craniopharyngomas. Int J Radiat Oncol Biol Phys 1982;8:1581-5. Sturm V, Scheer KE, Schlegel W, Strauss L, Penzholz H, Georgi P. Intracavitary contact radiation of cystic craniopharyngiomas by stereotaxic Y-90 application. Strahlentherapie (Sonderb) 1981;76:113-18. Backlund EO. Colloidal radioisotopes as part of a multimodality treatment of craniopharyngiomas. J Neurosurg Sci 1989;33:95-7. Coffey RJ, Lunsford LD. The role of stereotactic techniques in the management of craniopharyngiomas. Neurosurg Clin N Am 1990;1:161-72. Duma CM, Lunsford LD, Kondziolka D, Harsh GRt, Flickinger JC. Stereotactic radiosurgery of cavernous sinus meningiomas as an addition or alternative to microsurgery. Neurosurgery 1993;32:699-704. Muthukumar N, Kondziolka D, Lunsford LD, Flickinger JC. Stereotactic radiosurgery for tentorial meningiomas. Acta Neurochir (Wien) 1998;140:315-20. Steiner L, Lindquist C, Steiner M. Meningiomas and Gamma Knife radiosurgery. In: Al-Mefty, editor. Meningiomas. New York: Raven Press; 1991. p. 263-72. Henschen F. Zur Histologie und Pathogenese der Kleinhirnbruchenwinkeltumoren. Arch Psychiatr 1916; 56:21-122. Leksell L. A note on the treatment of acoustic tumours. Acta Chir Scand 1971;137:763-5. Prasad D, Steiner M, Steiner L. Gamma surgery for vestibular schwannoma. J Neurosurg 2000;92:745-59. Slattery WH III, Brackmann DE. Results of surgery following stereotactic irradiation for acoustic neuromas. Am J Otol 1995;16:315-19. Samii M, Matthies C. Gamma surgery for vestibular schwannoma. J Neurosurg 2000;92:892-6. Malis L. Gamma surgery for vestibular schwannoma. J Neurosurg 2000;92:894-6. Sheehan J, Yen CP, Arkha Y, Schlesinger D, Steiner L. Gamma knife surgery for trigeminal schwannoma. J Neurosurg 2007;106:839-45. Rades D, Fehlauer F. Treatment options for central neurocytoma. Neurology 2002;59:1268-70. Kim DG, Chi JG, Park SH, Chang KH, Lee SH, Jung HW, et al. Intraventricular neurocytoma: clinicopathological analysis of seven cases. J Neurosurg 1992; 76:759-65. Yasargil MG, von Ammon K, von Deimling A, Valavanis A, Wichmann W, Wiestler OD. Central neurocytoma: histopathological variants and therapeutic approaches. J Neurosurg 1992;76:32-7.
1083
1084
66
Gamma knife: clinical aspects
93. Yen CP, Sheehan J, Patterson G, Steiner L. Gamma knife surgery for neurocytoma. J Neurosurg 2007;107:7-12. 94. Szeifert GT PD, Kamiryo T. The role of gamma knife in the therapy of intracranial astrocytomas. In Second Annual Meeting of the Japanese society for Brain Tumor Surgery. Nagasaki, Japan, 1997. 95. Szeifert GT, Prasad D, Kamyrio T, Steiner M, Steiner LE. The role of the Gamma Knife in the management of cerebral astrocytomas. Prog Neurol Surg 2007;20:150-63. 96. Yen CP, Sheehan J, Steiner M, Patterson G, Steiner L. Gamma knife surgery for focal brainstem gliomas. J Neurosurg 2007;106:8-17. 97. Nwokedi EC, DiBiase SJ, Jabbour S, Herman J, Amin P, Chin LS. Gamma knife stereotactic radiosurgery for patients with glioblastoma multiforme. Neurosurgery 2002;50:41-6. 98. Payne BR, Prasad D, Szeifert G, Steiner M, Steiner L. Gamma surgery for intracranial metastases from renal cell carcinoma. J Neurosurg 2000;92:760-5. 99. Wronski M, Maor MH, Davis BJ, Sawaya R, Levin VA. External radiation of brain metastases from renal carcinoma: a retrospective study of 119 patients from the M. D. Anderson Cancer Center. Int J Radiat Oncol Biol Phys 1997;37:753-9. 100. Mingione V, Oliveira M, Prasad D, Steiner M, Steiner L. Gamma surgery for melanoma metastases in the brain. J Neurosurg 2002;96:544-51. 101. Goyal S, Prasad D, Harrell F Jr, Matsumoto J, Rich T, Steiner L. Gamma knife surgery for the treatment of intracranial metastases from breast cancer. J Neurosurg 2005;103:218-23. 102. Yen CP, Sheehan J, Patterson G, Steiner L. Gamma knife surgery for metastatic brainstem tumors. J Neurosurg 2006;105:213-19. 103. Backlund EO, Rahn T, Sarby B. Treatment of pinealomas by stereotaxic radiation surgery. Acta Radiol Ther Phys Biol 1974;13:368-76. 104. Muthukumar N, Kondziolka D, Lunsford LD, Flickinger JC. Stereotactic radiosurgery for chordoma and chondrosarcoma: further experiences. Int J Radiat Oncol Biol Phys 1998;41:387-92. 105. Marchini G, Gerosa M, Piovan E, Pasoli A, Babighian S, Rigotti M, et al. Gamma Knife stereotactic radiosurgery for uveal melanoma: clinical results after 2 years. Stereotact Funct Neurosurg 1996;66 Suppl 1:208-13. 106. Zehetmayer M, Menapace R, Kitz K, Ertl A, Strenn K, Ruhswurm I. Stereotactic irradiation of uveal melanoma with the Leksell gamma unit. Front Radiat Ther Oncol 1997;30:47-55. 107. Krayenbiihl H. Discussion eds rapports sur ules angiomes supratentoriels. Bruxelles: Congres International de Neurochirurgie; 1957.
108. McKissock W, Hankinson J. The surgical treatment of supratentorial angiomas. Premier Congress International de Neurochirurgie, Rapports et Discussion Bruxelles; 1957. 109. Olivecrona H, Ladenheim J. Congenital arteriovenous aneurysms of the carotid and vertebral systems. Berin: Springer; 1957. 110. Wegemann HJ. Das Schicksal operierter Patienten mit arteriovenosem Aneurysma des Gehrins. Zurich: Juris Druck Verlag; 1969. 111. Johnson R. Radiotherapy of cerebral angiomas with a note on some problems in diagnosis. Berlin: Springer-Verlag; 1965. 112. Lindqvist M, Steiner L, Blomgren H, Arndt J, Berggren BM. Stereotactic radiation therapy of intracranial arteriovenous malformations. Acta Radiol Suppl 1986;369:610-13. 113. Steiner L, Leksell L, Greitz T, Forster DM, Backlund EO. Stereotaxic radiosurgery for cerebral arteriovenous malformations. Report of a case. Acta Chir Scand 1972;138:459-64. 114. Marks MP, Lane B, Steinberg GK, Snipes GJ. Intranidal aneurysms in cerebral arteriovenous malformations: evaluation and endovascular treatment. Radiology 1992;183:355-60. 115. Steiner L, Lindquist C, Steiner M. Radiosurgery. Adv Tech Stand Neurosurg 1992;19:19-102. 116. Steiner L, Greitz T, Leksell L. Radiosurgery in intracranial arteriovenous malformation. In sixth International Congress of Neurological Surgeons. Amsterdam: Excerpta Medica; 1977. 117. Karlsson B, Lindquist C, Steiner L. Prediction of obliteration after gamma knife surgery for cerebral arteriovenous malformations. Neurosurgery 1997;40: 425-30. 118. Steiner L, Lindquist C, Adler JR, Torner JC, Alves W, Steiner M. Clinical outcome of radiosurgery for cerebral arteriovenous malformations. J Neurosurg 1992;77:1-8. 119. Steiner L, Lindquist C, Adler JR, Torner JC, Alves W, Steiner M. Outcome of radiosurgery for cerebral AVM. J Neurosurg 1992;77:823. 120. Karlsson B, Lindquist C, Steiner L. Effect of Gamma Knife surgery on the risk of rupture prior to AVM obliteration. Minim Invasive Neurosurg 1996;39:21-7. 121. Guo WY, Karlsson B, Ericson K, Lindqvist M. Even the smallest remnant of an AVM constitutes a risk of further bleeding. Case report. Acta Neurochir (Wien) 1993;121:212-15. 122. Shin M, Kawahara N, Maruyama K, Tago M, Ueki K, Kirino T. Risk of hemorrhage from an arteriovenous malformation confirmed to have been obliterated on angiography after stereotactic radiosurgery. J Neurosurg 2005;102:842-6.
Gamma knife: clinical aspects
123. Yamamoto M, Jimbo M, Kobayashi M, Toyoda C, Ide M, Tanaka N, et al. Long-term results of radiosurgery for arteriovenous malformation: neurodiagnostic imaging and histological studies of angiographically confirmed nidus obliteration. Surg Neurol 1992;37:219-30. 124. Yen CP, Varady P, Sheehan J, Steiner M, Steiner L. Subtotal obliteration of cerebral arteriovenous malformations after gamma knife surgery. J Neurosurg 2007;106:361-9. 125. Sirin S, Kondziolka D, Niranjan A, Flickinger JC, Maitz AH, Lunsford LD. Prospective staged volume radiosurgery for large arteriovenous malformations: indications and outcomes in otherwise untreatable patients. Neurosurgery 2006;58:17-27. 126. Pan DH, Guo WY, Chung WY, Shiau CY, Chang YC, Wang LW. Gamma knife radiosurgery as a single treatment modality for large cerebral arteriovenous malformations. J Neurosurg 2000;93 Suppl 3:113-19. 127. Miyawaki L, Dowd C, Wara W, Goldsmith B, Albright N, Gutin P, et al. Five year results of LINAC radiosurgery for arteriovenous malformations: outcome for large AVMS. Int J Radiat Oncol Biol Phys 1999;44:1089-106. 128. Del Curling O Jr, Kelly DL Jr, Elster AD, Craven TE. An analysis of the natural history of cavernous angiomas. J Neurosurg 1991;75:702-8. 129. Kondziolka D, Lunsford LD, Kestle JR. The natural history of cerebral cavernous malformations. J Neurosurg 1995;83:820-4. 130. Robinson JR, Awad IA, Little JR. Natural history of the cavernous angioma. J Neurosurg 1991;75:709-14. 131. Karlsson B, Kihlstrom L, Lindquist C, Ericson K, Steiner L. Radiosurgery for cavernous malformations. J Neurosurg 1998;88:293-7. 132. Hasegawa T, McInerney J, Kondziolka D, Lee JY, Flickinger JC, Lunsford LD. Long-term results after stereotactic radiosurgery for patients with cavernous malformations. Neurosurgery 2002;50:1190-7. 133. Kondziolka D, Lunsford LD, Flickinger JC, Kestle JR. Reduction of hemorrhage risk after stereotactic radiosurgery for cavernous malformations. J Neurosurg 1995;83:825-31. 134. Moriarity JL, Wetzel M, Clatterbuck RE, Javedan S, Sheppard JM, Hoenig-Rigamonti K, et al. The natural history of cavernous malformations: a prospective study of 68 patients. Neurosurgery 1999;44:1166-71. 135. Pollock BE, Garces YI, Stafford SL, Foote RL, Schomberg PJ, Link MJ. Stereotactic radiosurgery for cavernous malformations. J Neurosurg 2000; 93:987-91. 136. Lindquist C, Guo WY, Karlsson B, Steiner L. Radiosurgery for venous angiomas. J Neurosurg 1993; 78:531-6.
66
137. Stein BM, Sisti MB, Mohr JP, Pile-Spellman J. Radiosurgery and venous malformations. J Neurosurg 1994;80:175-7. 138. Garner TB, Del Curling O Jr, Kelly DL Jr, Laster DW. The natural history of intracranial venous angiomas. J Neurosurg 1991;75:715-22. 139. Payne BR, Prasad D, Steiner M, Bunge H, Steiner L. Gamma surgery for vein of Galen malformations. J Neurosurg 2000;93:229-36. 140. Steiner L. Radiosurgery in cerebral arteriovenous malformations. In: Flamm E, editor. Cerebrovascular surgery, vol 4. New York: Springer-Verlag; 1985. p. 1161-215. 141. Steiner L, Forster D, Leksell L, Meyerson BA, Boethius J. Gammathalamotomy in intractable pain. Acta Neurochir (Wien) 1980;52:173-84. 142. Mindus P. Capsulotomy in anxiety disorders, a multidiscipline study. In: Neurosurgery. Stockholm: Karolinska Institute; 1991. 143. Rylander G. Stereotactic radiosurgery in anxiety and obsessive-compulsive neurosis. Amsterdam: Elsevier/ North-Holland Biomedical Press; 1979. 144. Hayashi M, Taira T, Chernov M, Izawa M, Liscak R, Yu CP, et al. Role of pituitary radiosurgery for the management of intractable pain and potential future applications. Stereotact Funct Neurosurg 2003;81:75-83. 145. Young RF, Jacques DS, Rand RW, Copcutt BC, Vermeulen SS, Posewitz AE. Technique of stereotactic medial thalamotomy with the Leksell Gamma Knife for treatment of chronic pain. Neurol Res 1995;17:59-65. 146. Lindquist C, Steiner L, Hindmarsh T. Gamma Knife thalamotomy for tremor: report of two cases. In: Steiner L, editor. Radiosurgery. Baseline and trends. New York: Raven Press; 1992. 147. Hirai T, Ohye C, Nagaseki Y, Matsumura M. Cytometric analysis of the thalamic ventralis intermedius nucleus in humans. J Neurophysiol 1989;61:478-87. 148. Duma CM, Jacques DB, Kopyov OV, Mark RJ, Copcutt B, Farokhi HK. Gamma knife radiosurgery for thalamotomy in parkinsonian tremor: a five-year experience. J Neurosurg 1998;88:1044-9. 149. Young RF, Vermeulen SS, Grimm P, Posewitz AE, Jacques DB, Rand RW, et al. Gamma Knife thalamotomy for the treatment of persistent pain. Stereotact Funct Neurosurg 1995;64 Suppl 1:172-81. 150. Bingley T, Leksel L, Meyerson BA. Long term results of stereotactic anterior capsulotomy in chronic obsessivecompulsive neurosis. In: Sweet W, editor. Neurosurgical treatment in psychiatry, pain and epilepsy. Baltimore: University Park Press; 1977. p. 287-99. 151. Chen ZF, Kamiryo T, Henson SL, Yamamoto H, Bertram EH, Schottler F, et al. Anticonvulsant effects of gamma surgery in a model of chronic spontaneous limbic epilepsy in rats. J Neurosurg 2001;94:270-80.
1085
1086
66
Gamma knife: clinical aspects
152. Regis J, Kerkerian-Legoff L, Rey M, Vial M, Porcheron D, Nieoullon A, et al. First biochemical evidence of differential functional effects following Gamma Knife surgery. Stereotact Funct Neurosurg 1996;66 Suppl 1:29-38. 153. Barcia-Salorio JL, Barcia JA, Hernandez G, Lopez-Gomez L. Radiosurgery of epilepsy. Long-term results. Acta Neurochir Suppl 1994;62:111-13. 154. Regis J, Peragui JC, Rey M, Samson Y, Levrier O, Porcheron D, et al. First selective amygdalohippocampal radiosurgery for ‘mesial temporal lobe epilepsy’. Stereotact Funct Neurosurg 1995;64 Suppl 1:193-201. 155. Regis J, Bartolomei F, Rey M, Genton P, Dravet C, Semah F, et al. Gamma knife surgery for mesial temporal lobe epilepsy. Epilepsia 1999;40:1551-6. 156. Regis J, Bartolomei F, Rey M, Hayashi M, Chauvel P, Peragut JC. Gamma knife surgery for mesial temporal lobe epilepsy. J Neurosurg 2000;93 Suppl 3:141-6. 157. Regis J, Bartolomei F, Hayashi M, Chauvel P. Gamma Knife surgery, a neuromodulation therapy in epilepsy surgery. Acta Neurochir Suppl 2002;84:37-47. 158. Guo WY. Radiological aspects of gamma knife radiosurgery for arteriovenous malformations and other nontumoural disorders of the brain. Acta Radiol Suppl 1993;388:1-34. 159. Kihlstrom L, Guo WY, Karlsson B, Lindquist C, Lindqvist M. Magnetic resonance imaging of obliterated
160.
161.
162.
163.
164.
165.
166.
arteriovenous malformations up to 23 years after radiosurgery. J Neurosurg 1997;86:589-93. Yamamoto M, Ide M, Jimbo M, Hamazaki M, Ban S. Late cyst convolution after gamma knife radiosurgery for cerebral arteriovenous malformations. Stereotact Funct Neurosurg 1998;70 Suppl 1:166-78. Pan HC, Sheehan J, Stroila M, Steiner M, Steiner L. Late cyst formation following gamma knife surgery of arteriovenous malformations. J Neurosurg 2005;102 Suppl:124-7. Cahan WG, Woodard HQ, Higinbotham NL, Stewart FW, Coley BL. Sarcoma arising in irradiated bone: report of eleven cases. 1948. Cancer 1998;82:8-34. Rowe J, Grainger A, Walton L, Silcocks P, Radatz M, Kemeny A. Risk of malignancy after gamma knife stereotactic radiosurgery. Neurosurgery 2007;60:60-65. Sheehan J, Yen CP, Steiner L. Gamma knife surgeryinduced meningioma. Report of two cases and review of the literature. J Neurosurg 2006;105:325-9. Brada M, Ford D, Ashley S, Bliss JM, Crowley S, Mason M, et al. Risk of second brain tumour after conservative surgery and radiotherapy for pituitary adenoma. BMJ 1992;304:1343-6. White J, Clement GT, Hynynen K. Transcranial ultrasound focus reconstruction with phase and amplitude correction. IEEE Trans Ultrason Ferroelectr Freq Control 2005;52:1518-22.
65 Gamma Knife: Clinical Experience A. Niranjan . L. D. Lunsford . J. C. Flickinger . J. Novotny . J. Bhatnagar . D. Kondziolka
Historical Review Professor Lars Leksell selected Cobalt-60 as the ideal photon radiation source for radiosurgery after investigating protons and cross fired photons form early generation linear accelerator [1,2]. The first Gamma Knife (179 Co-60 sources) created a discoid-shaped lesion suitable for movement disorder and intractable pain surgery. Clinical experience with the Gamma Knife began in 1967 with the treatment of a patient with craniopharyngioma. Lunsford and colleagues introduced the first clinical 201-source Gamma Knife unit (model U) to North America (the fifth gamma unit worldwide) which was installed in August 1987 at University of Pittsburgh Medical Center. To eliminate challenging reloading issues associated with model U design, the Gamma Knife was redesigned with sources arranged in a circular (O-ring) configuration. The second generation unit (Model B) was installed at the University of Pittsburgh in 1996. Later the robotic automated positioning system (APS) transformed this unit into the third generation technology model C in March 2000. In January 2005 the fourth generation Leksell Gamma Knife model 4-C was installed. The model 4-C was equipped with hardware and software enhancements designed to improve workflow and provide integrated imaging capabilities, especially image fusion. A completely redesigned version of the Gamma knife was introduced in year 2007. This unit is called the ‘‘Perfexion’’ model (> Figure 65-1). It allows irradiation of wide range of anatomic targets, and increases efficiency and patient flow. The dose
#
Springer-Verlag Berlin/Heidelberg 2009
profile and dose delivery are improved. The beam size (4, 8, and 16 mm) is changed internally by robotic devices by moving radiation sources within a central collimator body. Sectors of different beam sizes can be mixed (> Figure 65-2). Changes in the dose profile are achieved by sector blocking of the Cobalt-60 sources. The treatable volume is greatly expanded to include targets anywhere in the cranial vault and upper neck. It uses robotics extensively, and has made the positioning of the patient even easier. The Perfexion Unit is especially valuable in the treatment of multiple brain metastases, since patients do not need to be re-positioned, and the risk of collision for lateral, inferior or posterior lesions is resolved using the expanded aperture of the gamma knife. Initial clinical trials with LGK Perfexion were completed at Marseille, France where the first clinical Perfexion was placed. Additional units are now operational in the UK, USA, and Canada. LGK Perfexion became operational at our institution on 28 September 2007. Between 1987 and 2007 8826 patients underwent gamma knife radiosurgery at our center (> Table 65-1).
Clinical Experience Vascular Malformation Radiosurgery Arteriovenous Malformations In the first 20 years of experience (1987–2007) in Pittsburgh, 1,164 patients with AVMs underwent single or multiple staged radiosurgery
1008
65
Gamma knife: clinical experience
. Figure 65-1 LGK Perfexion: A completely redesigned version of the Gamma knife which allows irradiation of wide range of anatomic targets, and increases efficiency and patient flow
. Figure 65-2 Collimator body of LGK Perfexion. The beam size (4 mm, 8 mm, and 16 mm) is changed internally by robotic devices by moving radiation sources within a central collimator body. Sectors of different beam sizes can be mixed. Changes in the dose profile are achieved by sector blocking of the Cobalt-60 sources
. Table 65-1 Indications treated with Gamma Knife radiosurgery at University of Pittsburgh (1987–2007) Brain disorder Vascular Disorders Benign Tumors Glial neoplasms Metastatic Tumor Functional Targets Miscellaneous Total
Number of patients treated 1329 2812 682 2587 869 347 8626
procedures (> Table 65-2). The goals of AVM radiosurgery are to achieve complete AVM obliteration, to improve symptoms, and to preserve existing neurological function. The chief benefit of radiosurgery is to eliminate the threat of spontaneous intracranial hemorrhage by gradual obliteration of the AVM nidus over 2–3 years [3,4]. Obliteration is a process resulting from endothelial proliferation within the AVM blood
Gamma knife: clinical experience
. Table 65-2 Gamma Knife radiosurgery for Vascular Brain Disorders Brain disorder
Indications
Vascular Disorders
AVM Cavernous Malformation A V Fistula
Total
Number of patients treated 1164 132 34 1329
vessel walls, supplemented by myofibroblast proliferation. This leads to contraction and eventual obliteration of the AVM blood vessel lumens. We evaluated 906 patients who were eligible for 3 year follow-up. The median nidus volume was 3.4 cc (range, 0.065–57.7 cc) and the median margin dose was 20 Gy (range, 13–32). A single procedure was performed in 865 (95.5%) patients. Prospective volume-staged radiosurgery was performed in 41 (4.5%) patients. Repeat radiosurgery for incomplete nidus obliteration after 3 years was needed in 113 (12.5%) patients. At a median follow-up of 38 months (1–204) complete nidus obliteration was achieved in 78% (angiographic confirmation in 67%, and MRI in 33%) (> Figure 65-3). In addition 20.8% of patients had achieved partial nidus obliteration. A total of 38 hemorrhages (4.1%) occurred after radiosurgery. Seizure control improved in 51% of those who presented with seizures. Adverse radiation effects included new neurological deficits in 24 patients (2.6%) and peri-AVM MRI T2 signal increase in 108 patients (12%). Long-term complications included cyst formation or encephalomalacia in 16 patients (1.7%). No radiation induced tumors were detected.
Risk of Hemorrhage after AVM Radiosurgery In our previous reports we analyzed the risk of hemorrhage during the latency interval from radiosurgery until complete AVM obliteration and
65
studied the clinical and angiographic outcomes of 312 patients who had a mean follow-up of 47 months [5]. The actuarial hemorrhage rate from a patent AVM (before complete obliteration) was 4.8% per year during the first 2 years after radiosurgery and totaled 5.0% per year for the third to fifth years after radiosurgery. Multivariate analysis of clinical and angiographic factors correlated the presence of an unsecured proximal aneurysm with an increased risk of postradiosurgical hemorrhage. Aneurysms, immediately proximal (flow related) to the AVM, usually close as the AVM obliterates. No AVM hemorrhages were observed after radiosurgery in seven patients with intranidal aneurysms. We recommend that aneurysms more than one arterial branch division proximal to the AVM be secured by endovascular or microsurgical approaches prior to (if the aneurysm bled) or shortly after radiosurgery. Inoue et al. identified a single draining vein, deep drainage, AVMs with a varix, AVMs with venous obstruction, high-flow (shunt- and mixed-type) AVMs, and large AVMs with a volume of more than 10 cc as risk factors for hemorrhage [6]. No patient in our study suffered a hemorrhage after angiography had confirmed complete obliteration (n = 140) or suffered from an early draining vein without residual nidus (n = 19). In this study no clear hemorrhage reduction benefit was conferred on patients who had incomplete nidus obliteration in early (<60 months) follow-up after radiosurgery. Karlsson et al. in a study of post radiosurgery hemorrhage noted that that the risk for hemorrhage decreased during the latency period [7]. In addition, these authors contended that the risk for having a hemorrhage in the latency period after gamma knife radiosurgery was dependent on minimum dose delivered to the AVM nidus. Maruyama et al. in a retrospective analysis involving 500 patients who had undergone AVM radiosurgery found that the risk of hemorrhage decreased by 54% during the latency period and by 88% after obliteration [8]. These authors
1009
1010
65
Gamma knife: clinical experience
. Figure 65-3 Posterior-Anterior (a) and Lateral carotid angiograms (b) showing large left parietal arteriovenous malformation at the time of gamma knife radiosurgery. Follow-up posterior-anterior (c) and Lateral carotid angiograms (d) showing complete nidus obliteration after 3 years
concluded that radiosurgery may decrease the risk of hemorrhage in patients with cerebral arteriovenous malformations, even before there is angiographic evidence of obliteration. This is an intriguing hypothesis that to date has defied widespread verification. The risk of hemorrhage is further reduced, although not eliminated, after obliteration (estimated lifetime risk of a bleed is <1%).
Probability of AVM Obliteration with Radiosurgery We studied the rate of AVM obliteration after Gamma knife radiosurgery at the University of
Pittsburgh in 351 patients with 3–11 years of follow-up imaging [9]. The median marginal dose was 20 Gy (range, 12–30) and median treatment volume was 5.7 cc (range, 0.26–24). AVM obliteration was documented by angiography in 193/264 (73%) and by MR alone in 75/87 (86%) patients who refused further angiography. Assuming a 96% accuracy for MR-detected obliteration, the corrected obliteration rate for all patients was 75% [10]. In some AVM patients treated by radiosurgery, follow-up angiography showed evidence of an early-draining vein but no discernable nidus. To our knowledge no patient has bled with this finding, and therefore we consider those patients obliterated or cured as well. Repeat radiosurgery is
Gamma knife: clinical experience
the preferred option for most patients with residual nidus remaining 3 years or more after initial radiosurgery [11].
Staged Volume Radiosurgery for Large AVMs Large AVMs pose a challenge for surgical resection, embolization, and radiosurgery. Some may be treated using multimodality management but a population of patients with large AVMs remains ‘‘untreatable.’’ Although AVM embolization prior to radiosurgery has been used for patients with large AVMs, recanalization was observed in 14–15% of patients [9,12]. Single-stage radiosurgery of large volume AVM either results in unacceptable radiation-related risks due to large volumes of normal surrounding tissue or low obliteration efficacy. Using single radiosurgery strategy in a subgroup of 48 patients with AVMs larger than 15 ml, Pan et al. found an obliteration rate of 25% after 40 months [13]. In their follow-up examinations, they observed 37% moderate and 12% severe adverse radiation effects in patients with AVMs larger than 10 ml. Miyawaki et al. reported that the obliteration rate in patients with AVMs larger than 14 ml treated using Linear accelerator radiosurgery was 22% [14]. Inoue et al. reported an obliteration rate of 36.4% and hemorrhage rate of 35.7% in the subgroup of AVMs larger than 10 ml treated by radiosurgery [6]. It became clear to us that in the narrow corridor between dose response and complication, the chances of achieving a high obliteration rate with a low complication rate for large AVM radiosurgery were slim. For this reason, radiosurgical volume staging was developed as an option to manage large AVMs [15]. We planned to prospectively divide the AVM nidus into two parts if the total volume was more than 15 cc. Each stage was defined at the first procedure, and then recreated at subsequent stages using internal anatomic landmarks.
65
The second stage radiosurgery procedure was performed 3–6 months after the first procedure. We reported an obliteration rate of 50% (7 of 14) after 36 months without new deficits, with an additional 29% showing near total obliteration [16]. Other reports have also documented the potential role of staged radiosurgery for large AVMs [17]. An increased neurological deficit was detected in only one patient and imaging showed peri-AVM changes in four (14%) patients. In this series, hemorrhage was observed in four (14%) patients. The concept of volume staging with margin dose selection at a minimum of 16 Gy seems reasonably safe and effective.
Adverse Radiation Effects of AVM Radiosurgery A multi-institutional study analyzed 102 of 1,255 AVM patients who developed neurological sequelae after radiosurgery [18]. The median marginal dose was 19 Gy (range, 10–35) and the median treatment volume was 5.7 cc (range, 0.26–143). The median follow-up after the onset of complications was 34 months (range, 9–140). Complications consisted of 80 patients with evidence of radiation related changes in the brain parenchyma. Seven also had with cranial nerve deficits, 12 developed seizures, and five had delayed cyst formation. Symptom severity was classified as minimal in 39 patients, mild in 40, disabling in 21, and fatal in two patients. Symptoms resolved completely in 42/105 patients with an actuarial complete resolution rate of 547% at 3 years post-onset. Delayed complications of radiosurgery include the risk of hemorrhage despite angiographically documented completely obliteration AVMs, the risk of temporary or permanent radiation injury to the brain such as persistent edema, radiation necrosis, cyst formation, and the risk of radiation-induced tumors. Cyst formation after AVM radiosurgery was first reported by Japanese
1011
1012
65
Gamma knife: clinical experience
investigators who reviewed the outcomes of patients initially treated in Sweden [19]. Delayed cyst formation has been reported in other recent long-term follow-up studies [20,21]. In our own twenty-year experience we have detected 16 patients (1.7%) with delayed cyst formation. Patients who developed delayed cyst formation were more likely to have had prior bleeds. Various surgical approaches ranging from surgical fenestration to cyst shunting were needed to manage these patients. Patients with T2 signal change without additional neurological problems generally do not need any active intervention.
Dural Arteriovenous Fistulas We have treated 34 DAVF patients using radiosurgery. In the 1990s, stereotactic radiosurgery followed by transarterial particulate embolization of accessible external carotid artery feeding vessels became a primary mode of treatment for dural arteriovenous fistulas (DAVFs) at our institution. Radiosurgery results in obliteration of DAVFs between 1 and 3 years after treatment, analogous to the experience with parenchymal AVMs [22–26]. Transarterial embolization, usually performed the same day and a few hours after radiosurgery, provides early palliative relief of intractable tinnitus, orbital venous congestion, and symptoms such as diplopia. In addition, it substantially reduces cortical venous drainage which may reduce the risk of hemorrhage during the latency period after radiosurgery. Even if recanalization of the embolized fistula occurs, the DAVF undergoes evantual radiosurgery-induced obliteration. Embolization is performed after radiosurgery to avoid the pitfall of having embolization temporarily obscure portions of the nidus that would then not be targeted during the radiosurgical procedure. Thus, the combination of radiosurgery and transarterial embolization, when possible, provides both rapid symptom relief and long-term cure of DAVFs.
We prefer to perform radiosurgery first and then embolization.
Cavernous Malformations Cavernous malformation radiosurgery has provided a therapeutic option for patients with symptomatic, hemorrhagic malformations in high-risk brain locations not amenable to resection [27,28]. Radiosurgery is performed for patients with symptomatic, imaging-confirmed hemorrhages for which resection was believed to be associated with high risk. At the University of Pittsburgh 132 patients with deep seated cavernous malformation have undergone radiosurgery. Most patients had multiple hemorrhages from brainstem or diencephalic cavernous malformations. More than 2-year follow-up was available on 108 patients. During an average observation period of 3.98 years per patient (for a total of 528 patientyears) prior to radiosurgery, 292 hemorrhages were noted, for an annual hemorrhage rate of 34.2%, excluding the first hemorrhage. After radiosurgery 36 hemorrhages were identified during an average of 5.6 years per patient (for a total of 625 patient-years). The annual hemorrhage rate was 13.96% per year for the first 2 years after radiosurgery, followed by 0.8% per year after 2 years. A significant decrease in the symptomatic hemorrhage rate after stereotactic radiosurgery of cerebral cavernous malformations indicates that radiosurgery is an effective management strategy for patients with hemorrhagic malformations in high-risk brain locations.
Tumor Radiosurgery Vestibular Schwannoma The goals of vestibular schwannoma radiosurgery are to prevent further tumor growth, preserve cochlear and other cranial nerve function where
Gamma knife: clinical experience
possible, to maintain or to improve the patient’s neurological status, and to avoid the risks associated with open surgical resection. Conformal dose planning is key to preserving hearing function (> Figure 65-4). We have managed 1,290 vestibular schwannoma patients using radiosurgery (> Table 65-3). Long-term results have established radiosurgery as an important minimally invasive alternative to microsurgery. Tumor Growth Control
Long-term results of Gamma Knife radiosurgery for vestibular schwannomas have been documented [22,29–32]. Recent reports suggest tumor control rates of 93–100% after radiosurgery [22,29–48]. Kondziolka et al. studied 5- to 10-year outcomes in 162 vestibular schwannoma patients who had radiosurgery at the University of Pittsburgh [42]. In this study a long-term 98% tumor control rate was reported (> Figure 65-5). Sixty-two percent of tumors became smaller, 33% remained unchanged, and 6% became slightly larger. Some tumors initially enlarged 1–2 mm during the first 6–12 months after radiosurgery as they lost their central contrast enhancement. Such tumors generally regressed in volume compared to their pre-radiosurgery size. Only 2% of patients required tumor resection after radiosurgery. Nore´n, in his 28-year experience with vestibular schwannoma radiosurgery, reported a 95% long-term tumor control rate. Litvack et al. reported a 98% tumor control rate at a mean follow-up of 31 months after radiosurgery using a 12 Gy margin dose [49]. Niranjan et al. analyzed the outcome of intracanalicular tumor radiosurgery performed at the University of Pittsburgh [50]. All patients (100%) had imaging-documented tumor growth control. Flickinger et al. performed an outcome analysis of acoustic neuroma patients treated between August 1992 and August 1997 at the University of Pittsburgh. The actuarial 5-year clinical tumor control rate (no requirement for surgical intervention) was 99.4 0.6% (> Figure 65-3) [32,34]. The long-term (10–15 year) outcome of benign tumor radiosurgery also has been evaluated. In
65
a study which included 157 patients with vestibular schwannomas, the median follow-up for the patients still living at the time of the study (n = 136) was 10.2 years. Serial imaging studies after radiosurgery (n = 157) showed a decrease in tumor size in 114 patients (73%), (> Figure 65-4 and 65-5) no change in 40 patients (25.5%), and an increase in three patients who later had resection (1.9%) [31]. No patient developed a radiation-associated malignant or benign tumor (defined as a histologically confirmed and distinct neoplasm arising in the initial radiation field after at least two years have passed). Hearing Preservation
Pre-radiosurgery hearing can now be preserved in 60–70% of patients, (> Figure 65-6) with higher preservation rates found for smaller tumors. In a long-term (5–10 year follow-up) study conducted at the University of Pittsburgh, 51% of patients had no change in hearing ability [34,42]. All patients (100%) who were treated with a margin dose of 14 Gy or less maintained a serviceable level of hearing after intracanalicular tumor radiosurgery [50]. Among patients treated after 1992, the 5-year actuarial rates of hearing level preservation and speech preservation were 75.2% and 89.2%, respectively, for patients (n = 89) treated with a 13 Gy tumor margin dose (> Figure 65-7a, b). The 5-year actuarial rates of hearing level preservation and speech preservation were 68.8 and 86.3%, respectively, for patients (n = 103) treated with >14 Gy as the tumor margin dose [32]. Unlike microsurgery, immediate hearing loss is uncommon after radiosurgery. If hearing impairment is noted, it occurs gradually over 6–24 months. Early hearing loss after radiosurgery (within 3 months) is rare and may result from cranial nerve edema or demyelination. Facial Nerve and Trigeminal Nerve Preservation
Facial and trigeminal nerve function can now be preserved in the majority of patients (>95%). In the early experience at University of Pittsburgh normal facial function was preserved in 79% of patients after 5 years and normal trigeminal nerve
1013
. Figure 65-4 Conformal gamma knife radiosurgery dose plan for acoustic neuroma. SPGR contrast enhanced MRI showing conformal dose plan in axial (a), coronal (b) and sagittal (c) plane. A margin dose of 12.5 Gy was prescribed to 50% isodose line (Yellow line marked with white arrow in a). The isodose lines are projected on 3D T2 weighted images (c, d, and e). The cochlea which is seen in 3D T2 weighted (single arrow in d) images receives less than 5 Gy (20%) (Green line marked by double arrows) of central dose
1014
65 Gamma knife: clinical experience
Gamma knife: clinical experience
function was preserved in 73%. These facial and trigeminal nerve preservation rates reflected the higher tumor margin dose of 18–20 Gy used during the CT based planning era before 1991. In a later study using MR based dose planning, a 13 Gy tumor margin dose was associated with 0% risk of new facial weakness and 3.1% risk of facial numbness (5-year actuarial rates). A margin dose of >14 Gy was associated with a 2.5% risk of new onset facial weakness and a 3.9% risk of facial
. Table 65-3 Gamma Knife radiosurgery for Benign Brain Neoplasms Brain disorder
Indications
Benign Tumors
Vestibular Schwannoma Meningioma Pituitary Adenoma Non-Vestibular Schwannoma
Total
Number of patients treated 1290 1170 266 86 2812
65
numbness (5-year actuarial rates) [32]. Similar 10-year facial and trigeminal neuropathy rates have been documented [12]. None of the patients who had radiosurgery for intracanalicular tumors developed new facial or trigeminal neuropathies. Neurofibromatosis 2
Patients with vestibular schwannomas associated with neurofibromatosis 2 represent a special challenge because of the risk of complete deafness. Unlike the solitary sporadic tumors that tend to displace the cochlear nerve, tumors associated with NF2 tend to form nodular clusters that engulf or even infiltrate the cochlear nerve. Complete resection may not always be possible. Radiosurgery has been performed for patients with NF2. Subach et al. studied our first 40 patients (with 45 tumors) who were treated with radiosurgery for NF2. Serviceable hearing was preserved in 6 of 14 patients (43%), and this rate improved to 67% after modifications made to the technique in 1992. The tumor control rate was 98% [51] only one patient showed imaging documented growth. Normal facial nerve function and
. Figure 65-5 Axial contrast enhanced MRI showing of a 49 year old man showing left sided acoustic tumor at radiosurgery (a). Long-term (11 years) MRI follow-up shows significant tumor shrinkage (b)
1015
1016
65
Gamma knife: clinical experience
. Figure 65-6 Axial contrast enhanced T1 weighted MR image (a and b) showing left frontal and right parietal brain metastases with significant surrounding edema. Three month follow-up MRI (right) showing tumor shrinkage and significant reduction in surrounding edema (c and d)
trigeminal nerve function was preserved in 81 and 94% of patients, respectively. In two recent series, [52,53] serviceable hearing was preserved in only 30% [52] and 40% [53] of cases, respectively. The tumor control rate was respectively 71% [52] and 79%. [53] Mathieu et al. updated outcomes of our NF2 series in 2007 [54]. The tumor control rate was 87.5%. The rate of serviceable hearing preservation using current technique was 52.6%. It now appears that preservation of serviceable hearing in patients with NF2 is an attainable goal using gamma knife radiosurgery. Early radiosurgery when the hearing level is still excellent may become an appropriate strategy in the future. At present we
generally delay radiosurgery in NF2 patients until we see hearing deterioration or tumor growth.
Meningioma The optimal treatment for meningioma when feasible is complete resection of tumor with its dural base. However, when meningiomas are attached to skull base cranial nerves or vascular structures, complete resection may not be possible. Recurrence rates are higher for meningiomas in critical locations where only subtotal resections are possible due to limited access and involve-
Gamma knife: clinical experience
65
. Figure 65-7 (a) Axial contrast enhanced T1 weighted MR image shows a recurrent skull base at radiosurgery. (b) Long-term (12 years) follow-up MRI showing tumor regression
ment of critical structures. Radiosurgery offers an attractive option for patients with residual or recurrent meningioma as well as for patients in whom complete resection of tumor is considered attainable but only with unacceptable morbidity. Meningiomas attached to or within major venous sinuses, can be successfully treated by radiosurgery. Tumor regression may occur slowly over several years after radiosurgery. Radiosurgery provides long-term tumor control associated with high rate of neurological preservation and patient satisfaction [55]. Surgical excision is the preferred first line approach for most symptomatic convexity, anterior fossa, or lateral sphenoid ridge meningiomas. Even for these radiosurgery can be offered as a first management approach unless the tumor needs debulking because of mass effect. Malignant meningiomas require multimodality management that includes resection, radiosurgery, and radiation therapy. Gamma-knife radiosurgery (GKRS) has proven an effective strategy for many patients with recurrent meningiomas [56]. In general, the biological nature of the meningioma is the main factor that determines how effectively radiosurgery will control tumor growth. Recently, stereotactic radiosurgery has been performed for an increasing number of patients with smallto moderate-size meningiomas as an alternative
to surgical excision [55,57–67]. Gamma-knife radiosurgery is used as a first-line treatment and/or postoperative adjuvant therapy for suitable patients with meningiomas of the skull base, posterior fossa or cavernous sinus region [68]. Small, sharply demarcated tumors are the best candidates for radiosurgery. Gamma-knife procedures can be performed even after surgery and radiation therapy have failed [69]. Several centers have reported tumor control rates between 84 and 98% after meningioma radiosurgery with a range of follow-up intervals [55,57,59–61,64,65]. Significantly, most prior studies included patients with atypical or malignant tumors who had lower control rates after radiosurgery [63,66]. At our center 1,170 meningioma patients have undergone radiosurgery. Long-term follow-up (up to 20 years) now confirms the high tumor control rate and low morbidity of radiosurgery [70]. At many centers radiosurgery has become the preferred treatment for patients with small- to moderate-size meningiomas without symptomatic mass effect. Results of meningioma radiosurgery supports the concept that radiosurgery should be considered as primary management option for patients with tumors involving the skull base, where a Simpson Grade 1 resection often cannot be accomplished with acceptable risk [71–73].
1017
1018
65
Gamma knife: clinical experience
The morbidity of radiosurgery for meningiomas of the cavernous sinus and petroclival regions has been analyzed separately and is considered overall to be <10% [58,61,62,64,67]. Patients with meningiomas of the falx and tentorium are another group who may benefit from radiosurgery compared with surgical resection if the tumor is not too large [60]. Younger age and smaller tumor size are associated with better tumor control after GKRS [74,75]. For cavernous sinus meningiomas, GKRS may be a valuable alternative to surgical removal [74]. In addition to being an attractive adjuvant treatment for some meningiomas, high tumor control rates and low morbidity enable GKRS to even replace microsurgery in many cases [76]. In order to reduce morbidity, some authors recommend the combination of nonradical resection and subsequent radiosurgery in selected patients with unresectable meningiomas [75].
Pituitary Adenoma Multimodality management often is needed for patients with pituitary tumors. The primary aim of treatment for clinically non-functioning pituitary macroadenomas is tumor removal with preservation of visual function. Transphenoidal surgery is the preferred approach for managing most pituitary adenomas. Radiosurgery is often indicated as an adjuvant management after partial resection or later recurrence of non-functioning pituitary adenomas. However, radiosurgery can be performed as the primary management of non-functioning adenomas in carefully selected patients such as those who have major surgical risks or for patients who decline microsurgery. Cavernous sinus invasion can occur de novo in patients with large pituitary macroadenomas, but is more commonly seen as a residual tumor after attempted microsurgical resection. The cranial nerve complication risks and cerebrovascular risks of cavernous sinus microsurgery warrant consideration of radio-
surgery. In many cases, the cavernous sinus mass can be treated while selectively sparing not only the optic apparatus, but also the pituitary stalk and residual pituitary gland within the sella. Tumor growth control rates of 90–100% have now been confirmed by multiple centers after pituitary adenoma radiosurgery [77]. The antiproliferative effect of radiosurgery has been reported in nearly all patients who underwent gamma knife radiosurgery. Relatively few patients (who usually had received lower margin doses) eventually required additional treatment. At our institution 266 pituitary adenoma patients have undergone radiosurgery. For secretory adenomas, medical management is extremely useful as either first line therapy or as an adjunct in a combined multimodality approach to overall patient management. Tumor resection is the preferred management strategy when medical management fails to normalize pituitary function. Radiosurgery is often indicated as an adjunct to control residual or recurrent secretory adenoma. The initial first stage extracavernous microsurgery is often optimal in order to reduce the subsequent tumor volume, create space between the tumor and the optic apparatus, and facilitates safe delivery of the highest dose of radiosurgery possible. The goal of radiosurgery for functional adenomas is pituitary hormone normalization. Biochemical remission for Growth Hormone (GH) secreting adenomas is defined as GH level suppressed to below 1 mg/l on oral glucose tolerance test (OGTT) and normal age-related serum insulin like growth factor-1 (IGF I) levels. OGTT remains the gold standard for defining a cure of acromegaly. The IGF 1, however, is far more practical. Decrease of random GH to less than 2.5 mg/l is achieved more frequently than the normalisation of IGF 1 but it is necessary to fulfil both criteria to define complete remission. Hormonal normalization after radiosurgery varies between 23–82% of cases in published series [78–82]. The suppression of hormonal hyperac-
Gamma knife: clinical experience
tivity is more effective when higher doses of radiation are used. In a study at the University of Pittsburgh 38% of recurrent tumor patients were cured (GH 1 mg/l) and overall 66% had growth hormone levels 5 mg/l, 3–5 years after radiosurgery [83]. The effect of radiosurgery requires a latency of about 20–28 months [82,84]. During this interval hormone suppressive medications may be beneficial. Because hormone-suppressive medication during radiosurgery may act as a radioprotective agent, this medication should be discontinued several weeks prior to radiosurgery but may be resumed after a week. Patients with Cushing’s disease (ACTH Secreting Adenomas) respond to radiosurgery but more than one procedure may be needed if the tumor cannot be well defined during the initial imaging. In addition there is a latency of about 14–18 months for maximal therapeutic response [82,84]. In various published series 38–83% patients achieve hormone normalization after radiosurgery [81,85–87]. Most prolactinomas are managed successfully by dopaminergic suppressive therapy. Surgery is indicated for cases of intolerance to medical treatment, in cases where women desire to have children, or when patients are dopamine agonist resistant (5–10% of patients). Some patients prefer microsurgery or radiosurgery to the need for highly expensive life long medical treatment. In published studies of patients treated with radiosurgery, 25–29% showed normalization [81,82]. The possible radioprotective effect of dopaminergic drugs should be taken into account. Because patients treated on dopamine agonist during radiosurgery had lower remission rate, it is recommended that prolactinoma radiosurgery be performed during a period of drug withdrawal. New pituitary hormone deficiency has been reported in 20–30% patients after radiosurgery for functional pituitary adenomas [81,82]. The most important factor influencing post radiosurgery hypopituitarism seems to be the mean dose to
65
the hypophysis (pituitary stalk). Vladyka et al. observed some worsening of gonadotropic, corticotropic or thyrotropic functions 12–87 months after radiosurgery, usually within 4–5 years after radiosurgery [88]. Deterioration in pituitary functions was observed when pituitary stalk received higher doses (>15 Gy). The risk for hypopituitarism after stereotactic radiosurgery thus becomes a primary function of the anatomy of the tumor and the dose prescribed. For recurrent tumors primarily where the pituitary stalk (and even at times the residual pituitary gland) is separate from the tumor, is easily visualized, when the stalk and gland are excluded from higher dose, the risk of hypopituitarism is reduced. For adenomas that cannot be visually separated from the normal gland, particularly if they extend upward to involve or compress the pituitary stalk, the risk is predominantly related to the dose necessary to effectively achieve all outcome goals for the functional status of the tumor (higher for secretory than nonsecretory adenomas). Gamma-Knife radiosurgery is superior to radiation therapy because there is a faster response and fewer adverse radiation effects. Response to radiosurgery is best with ACTH producing tumors, followed by GH producing tumors, prolactinoma having the poorest response usually because they have failed prior medical management due to their invasive nature. Hypopituitarism can be expected to occur in up to 30% within 4–5 years, but can be avoided by minimizing radiation to pituitary stalk and hypothalamus. Somatostatin analogs and dopamine agonists may have a radioprotective effect [82,89]. Although the radioprotective effect of these drugs was not confirmed in subsequent studies [78–80] it is advisable to stop these drugs prior to radiosurgery. Short-acting form of somatostatin analogs can be given until 2 weeks prior to GK. Long-acting somatostatin analogs should be discontinued as much as 4 months prior to GK. Dopamine agonists should be discontinued 2 months prior to radiosurgery.
1019
1020
65
Gamma knife: clinical experience
After radiosurgery, once hormone levels are normal on medical therapy, somatostatin analogs should be stopped for 4 months each year to assess for biochemical cure. Similarly, dopamine agonists should be stopped for 2 months. A panel of tests to detect hypopituitarism should be done at 6 month intervals for the first 5 years and then yearly.
Craniopharyngioma Multi-modality therapy is often necessary for such patients because of the development of refractory cystic components of their tumors. Radiosurgery is usually part of a multi-modality management when prior therapies have failed [90,91]. Sixty-eight patients have undergone gamma knife radiosurgery as part of a primary or adjuvant management strategy for craniopharyngioma. Long-term follow-up in our patient series was available in 29 patients. The median tumor volume was 0.4 (range, 0.12–6.36) cm3. One to nine isocenters of different beam diameters were used. The median dose to the tumor margin was 12.5 Gy (range, 9–20), and the maximum dose was 25 Gy (range, 21.8–40). The dose to the optic apparatus was limited to less than 8 Gy. Clinical and imaging follow-up data were obtained at a median of 24 months (range, 13–150) from radiosurgery. Overall, 14 of 29 tumors regressed or vanished, and ten remained stable after radiosurgery. Further tumor growth was noted in five patients, three of whom underwent surgical resection and one had repeat radiosurgery. Two additional patients needed management for cyst enlargement. One patient with prior visual defect had further vision deterioration 9 months after radiosurgery. No patient developed new-onset diabetes insipidus. We found that stereotactic radiosurgery was a reasonable option in selected patients with small recurrent or residual craniopharyngiomas. Adverse radiation risks related to adjacent cranial nerve structures or the development of new
extraocular movement deficits are rare, providing that the optic nerve and tract dose is kept lower than 8 Gy or less in a single procedure. In general, we prefer the use of multi-modality management including microsurgery, radiosurgery, and intracavitary radiation rather than stereotactic or fractionated radiation therapy. The goal has been to maintain endocrinological function whenever possible, reduce the risks of visual dysfunction, and subsequently control tumor growth. There are other reports that have similar results in the management of craniopharyngiomas using gamma knife [86,92,93].
Glial Tumors Malignant Gliomas
Malignant gliomas continue to represent one of the most serious challenges in neurosurgery. Radiation therapy has become the mainstay of the treatment. The observation that local control and median survival can be improved through the dose escalation is the basis for the application of radiosurgery to malignant gliomas. Radiosurgery is used for boost irradiation of patients with malignant glial tumors in addition to conventional wide-margin fractionated radiotherapy. It has been used mainly for patients with tumors <3.5 cm in diameter as part of multimodality approach to malignant gliomas. At our institution 682 patients with glial tumors have been treated with radiosurgery (> Table 65-4). Early radiosurgery reports widely varied in the outcomes for malignant gliomas with a median survival for GBM patients ranging from 9.5 to 26 months. These variations could result from patient selection biases and other prognostic factors [94]. We performed a retrospective study to evaluate the result of radiosurgery on 64 GBM and 43 anaplastic astrocytoma patients. The median survival for the GBM was 16 months after radiosurgery and 26 months after diagnosis. The 2-year survival rate was 51%. For patients with anaplas-
Gamma knife: clinical experience
. Table 65-4 Gamma Knife radiosurgery for Glial Neoplasms Number of patients treated
Indications Astrocytoma Pilocytic Fibrillary Anaplactic GBM
81 40 94 308
Mixed
overall survival. No acute Grade 3 or Grade 4 toxicity was encountered. There appears to be a significant survival advantage using radiosurgery boost in patients with malignant glioma, especially if appropriately used with surgery and other adjuvant therapies. However, a carefully designed prospective randomized trial is needed to reliably establish any survival benefit from a radiosurgical boost for patients with malignant gliomas. Lower Grade Gliomas
Astro-Ologo Anaplastic Astro-Ologo GBM-Oligo Undifferentiated cell type-oligo Ependymoma Meduloblastoma Total
65
9 31 7 11 64 23 682
tic astrocytomas median survival after radiosurgery was 21 months and after diagnosis was 32 months. The 2-year survival rate after diagnosis was 67%. Other centers have recently reported survival rates that seem significantly improved as compared to 9-month median survival and 10% 2-year actuarial rate reported for standard therapy. Nwokedi et al. compared survival between 33 patients treated with external beam radiation therapy (EBRT) alone (Group 1), and 31 patients managed with EBRT plus a gamma knife radiosurgery (GK-SRS) boost (Group 2) [95]. GK-SRS was administered to most patients within 6 weeks of the completion of EBRT. The median EBRT dose was 59.7 Gy (range, 28–70.2 Gy), and the median GK-SRS dose to the prescription volume was 17.1 Gy (range, 10–28 Gy). Both groups were comparable in age, Karnofsky performance status, extent of resection, and tumor volume. The median survival was significantly better in patients treated with EBRT plus GK-SRS (13 months in EBRT alone versus 25 months in EBRT plus GK-SRS). Age, Karnofsky performance status, and the addition of GK-SRS were all found to be significant predictors of
Low-grade gliomas have been treated with radiosurgery. Simonova et al. treated 68 low-grade gliomas patients using gamma knife surgery [96]. The median patient age was 17 years and median target volume was 4,200 mm3. The median marginal prescription dose was 25 Gy. Ninety-five percent of patients were treated in five daily stages. These authors reported 83% rate of partial or complete tumor regression with a median time to response of 18 months. In this series the progression-free survival was 92% at 3 years and 88% at 5 years. Moderate acute or late toxicity was noted in 5% of patients. Kida et al. treated 51 patients harboring lowgrade gliomas (12 Grade I astrocytomas, 39 Grade II astrocytomas) using gamma knife [97]. The mean margin dose was 12.5 Gy for Grade I and 15.7 Gy for Grade II tumors. In the mean followup of 27.6 months grade I astrocytomas had a response rate of 50% and a control rate of 91.7%. Grade II astrocytomas had a 46.2% response rate and an 87.2% control rate. Despite the favorable histological characteristics and prognosis of pilocytic astrocytomas, some patients may not be cured after microsurgery because of an adverse location, recurrence, or tumor progression. Radiosurgery is an effective alternative therapeutic approach for these patients. Hadjipanayis et al. reported outcome in 38 patients harboring unresectable pilocytic astrocytomas who were treated with radiosurgery [98] at our center. The median radiosurgical dose to the tumor margin was 15 Gy (range, 9.6–22.5 Gy). After radiosurgery, serial imaging demonstrated complete tumor resolution in
1021
1022
65
Gamma knife: clinical experience
ten patients, reduced tumor volume in eight, stable tumor volume in seven, and delayed tumor progression in 12. Three patients died of local tumor progression. Stereotactic radiosurgery is a valuable adjunctive strategy in the management of recurrent or unresectable pilocytic astrocytomas especially small volume, sharp bordered tumors. Radiosurgery can play an important role in the treatment of low-grade astrocytomas, and complete cure of these tumors has been achieved in at least some of the cases [98]. The acute complications following radiosurgery are unusual and limited to exacerbation of existing symptoms. The most frequently seen delayed complication of radiosurgical boost is tumor swelling radiation reaction in the tumor or surrounding brain swelling. Symptoms are usually controllable by steroid therapy. The reported incidence of radiation necrosis ranges from 2 to 22%. Re-operation rates ranging from 21 to 33% have been reported after radiosurgery. Neither radiation necrosis nor reoperation is associated with diminished length of survival.
Brain Metastases Radiosurgery (SRS) as the sole initial management or as a boost before or after whole brain radiation therapy (WBRT) has emerged as a widely practiced treatment modality for brain metastases. The goal of radiosurgery without WBRT is to achieve brain control without the possible long term neurotoxic or cognitive side effects of WBRT [99]. The rationale for radiosurgery, when used as a boost after WBRT, is to achieve improved local brain tumor control. Radiosurgery boost improves survival in selected patients in whom the predominant problem is brain disease rather than extracranial disease. Radiosurgery is also used as salvage treatment for progressive intracranial disease after surgery or WBRT. Traditionally radio-insensitive histologies tend to be more responsive to SRS than to conventional fractionated radiation treat-
ment. In addition, SRS causes indirect vascular injury and subsequent sclerosis of blood vessels, and eventual compromise of the blood supply and circulation within the tumor [100]. At our institution 2,587 patients have undergone radiosurgery for brain metastases (> Table 65-5). Retrospective series have consistently revealed local control of the target lesions in the range of 80–85% or even higher with a very acceptable side effect profile[101–107]. Prospective randomized trials have demonstrated that the one-year local control rate of target lesions with radiosurgery is 73%, which increases to 82–89% with the addition of WBRT [108,109]. Several studies have reported excellent local control (70–80% at 1 year) following radiosurgery for brain metastases [110,111] (> Figure 65-6). Other published series of patients treated with SRS have demonstrated a risk of distant brain failure at 1 year, ranging from 43 to 57% [112–115]. In general, the risk of new metastasis in patients with solitary tumors is approximately 37% (crude), but the actuarial risk is 50% at 1 year [116,117]. The histologic features or tumor type may play a role, with melanoma being more likely to be associated with multiple metastases than some other tumor types [118]. Despite a relatively high risk of new metastases outside the radiosurgery volume in patients who have SRS alone, retrospective studies have not . Table 65-5 Gamma Knife radiosurgery for Metastatic Neoplasms Primary Tumors Breast Sarcoma Gestro-intestinal Kidney Lung Melanoma Nasopharynx Thyroid Others Unknown primary Total indications
Number of patients treated 463 17 132 204 1168 392 29 12 106 64 2587
Gamma knife: clinical experience
confirmed a survival benefit to adjuvant WBRT [113,119,120]. Freedom from local progression in the brain at 1 year was significantly superior in patients who received both SRS and WBRT compared with SRS alone (28% vs. 69%), although the overall survival rate was not significantly different [114]. A retrospective, multi-institutional study in which patients were treated with SRS alone (n = 268) or SRS + WBRT (n = 301) also reported no significant difference in the overall survival rate [161]. Despite the higher rate of new lesions developing in patients treated with SRS alone, the overall survival appears to be equivalent to SRS + WBRT since salvage therapies are fairly effective and patients’ extracranial disease is frequently the cause of death [113]. Only 24% of patients managed initially with radiosurgery alone required salvage WBRT. Pirzkall et al. reported that there was no survival benefit for an overall group of 236 patients with adjuvant WBRT but these authors noted a trend toward improved survival in a subset of patients with no extracranial tumor (15.4 vs. 8.3 months, p = 0.08) [119]. Chidel et al. reported 78 patients managed initially with SRS alone and 57 patients treated with SRS and adjuvant WBRT [67]. Whole-brain radiation therapy did not improve the overall survival rate but was useful in preventing both the local progression and the development of new brain metastases (74% vs. 48%, p = 0.06). These retrospective studies suggest that WBRT will improve local and distant control in the brain, but do not clearly demonstrate a survival advantage [113]. A multicenter retrospective analysis was performed with 502 patients treated at ten institutions in which all of the patients were treated with WBRT and SRS. The patients were stratified by the recursive partitioning analysis and compared with similar patients from the RTOG database who had been treated with WBRT alone. [121]. The study revealed that patients with higher KPS, controlled primary tumor, absence of extracranial metastases and lower RPA class
65
had statistically superior survival. The addition of an SRS boost resulted in a median survival of 16.1, 10.3, and 8.7 months, respectively, for RPA classes I, II, and III. This is in comparison to 7.1, 4.2, and 2.3 months for similar RPA class patients from the RTOG database. This improvement in overall survival, stratified by RPA class with an SRS boost, was statistically significant [121]. In a recent study SRS alone was found to be as effective as resection plus WBRT in the treatment of one or two brain metastases for patients in RPA classes I and II [122]. Stereotactic radiosurgery is an effective treatment for patients with multiple brain metastases. A substantial amount of published literature now supports use of radiosurgery in the treatment of multiple brain metastases. Stereotactic radiosurgery offers a very high control rate with a low risk of serious side effects.
Pineal Region Tumors Management of pineal region tumors remains a significant challenge because of the anatomic complexity of the area and the presence of critical brain and vascular structures. Microsurgical techniques are often successful in obtaining a tissue diagnosis; however, the likelihood of curative resection remains low. There are only few published reports on radiosurgery for pineal tumors [123,124]. At the authors’ institution 29 patients with parenchymal pineal tumors were treated between 1989 and 2007. Outcome analysis on 14 patients with longterm follow-up showed that local tumor control was achieved in 13 patients while one died of tumor progressions despite chemotherapy and craniospinal irradiation prior to radiosurgery. Neuroimaging follow-up showed complete disappearance of tumor in three, decrease in tumor size in seven, no change in tumor size in three, and tumor growth in one patient.
1023
1024
65
Gamma knife: clinical experience
Skull Base Tumors Radiosurgery is a primary and adjuvant management for tumors of skull base [81,125–130]. From September 1987 through December 2004, 238 patients with a variety of skull base tumors were treated with gamma knife radiosurgery at the University of Pittsburgh Medical Center. These tumors and their subsequent management are described below in more detail. Non-vestibular Schwannomas
Eighty-six patients underwent radiosurgery for non-Vestibular nerve sheath tumors defined by clinical examination, high-resolution intraoperative imaging, and in selected cases prior surgery. Our results of trigeminal schwannoma have been recently published [127]. The records of 23 patients were reviewed with a median follow-up of 40 months. Twenty of twenty-three patients (91%) had tumor growth control, with regression noted in 15 and no further tumor growth in five. Patients who had subsequent tumor enlargement underwent a second radiosurgical procedure. Twelve of twenty-three trigeminal nerve sheath tumor patients (52%) reported symtomatic improvement. Nine (39%) had no change in their symptoms. Only two patients noted new neurological complaints such as facial weakness (one patient) or worsening of the pre-radiosurgical facial numbness (one additional patient). Of interest, trigeminal nerve sheath tumors have a much higher likelihood of developing transient, but occasionally impressive short-term swelling of their tumor in the first year after radiosurgery. This is quite distinct from those patients who have undergone acoustic tumor radiosurgery. In the majority of trigeminal neuroma patients, transient swelling is followed by delayed shrinkage, often of profound degree. Therefore, it is critical that patients and referring doctors do not despair during this transient tumor enlargement phase identified by imaging and sometimes associated with temporary neurological symptoms. Most such symptoms
will resolve as the tumors regress during the next 3–6 months. Radiosurgery using the gamma knife proved to be an effective management strategy for those patients who had undergone both primary as well as adjuvant (post microsurgery) radiosurgery [128]. Three patients underwent gamma knife radiosurgery for facial schwannomas, all identified at the time of prior microsurgery and associated with recurrence or subtotal prior resection. Tumors of the ninth and tenth cranial nerve pose special challenges. Twenty-six patients with jugular bulb schwannomas underwent radiosurgery for between August 1987 and September 2004. Most such patients present with symptoms related to imbalance, incoordination, dysphasia, or hearing loss. A total of 12 patients had previously undergone gross total resection with tumor recurrence, and four had undergone prior partial resection. Results to date show a high likelihood of long-term tumor growth control for such tumors. In an earlier report including 17 patients, we reported a tumor control rate of 94% (eight decreased and eight was stable in size) after Jugular foramen schwannoma radiosurgery [131]. Zhang et al. reported 96% (26/27) tumor growth control with a follow-up period of 38.7 months [130]. In the series of non vestibular schwannomas Pollock et al., reported 96% (22/23) tumor growth control after Gamma Knife radiosurgery [81]. Glomus Tumors
Radiosurgery using the gamma knife has been performed in 17 patients in a 20-year interval. The sparse number of patients (of 8,600 who had gamma knife radiosurgery) is accounted for by the tendency of such tumors to extend well below the skull base. When surgical resection is not feasible, we consider staged radiosurgery technologies such as LINAC based radiosurgery for the extracranial component and the gamma knife for the intracranial portion. Some patients also have undergone elective embolization for shrinkage of their tumor or subtotal microsurgical resection. Only one patient in our series had a glomus tympanicum
Gamma knife: clinical experience
tumor. The new Perfexion model will facilitate extracranial radiosurgery to the level of C 4. Gamma knife radiosurgery appears to have a long-term high tumor control rate of glomus tumors, paralleling the benefit provided by fractionated radiation therapy. However, the gamma knife provides a superior biological effective tumor dose, with better dose sparing of the adjacent brainstem and cranial nerve structures. Pollock et al. in a series including 42 patients reported 98% tumor control after glomus jugulare radiosurgery at a mean follow-up of 44-month [132]. Neurological improvement or stability was observed in the majority of patients in published series. Centers using LINAC-based radiosurery continue to support radiosurgery as an effective and safe method of treatment for glomus jugulare tumors that results in low rates of morbidity. Hemangiomas
Radiosurgery for hemangioma was performed in eight patients. Hemangiomas of neurosurgical interest are histologically benign vascular epithelial cell origin tumors that most often occur in the orbit or cavernous sinus or both. These patients tend to present with ocular symptoms or signs such as orbital pain, ophthalmoplegia, proptosis or impaired visual acuity. They can, in fact, be diagnosed by their characteristic imaging appearance by MRI. Since they may hemorrhage dramatically at the time of attempted removal, it is prudent for surgeons considering biopsy or resection of such tumors to get the appropriate imaging in advance. Asymptomatic lesions do not require intervention but are often approached surgically in pursuit of a diagnosis. Symptomatic lesions require treatment. Options include en bloc resection, embolization, or radiation. Radiosurgery is a better option. In our relatively limited experience, some patients have had incomplete resection because of excessive blood loss, and one patient had undergone unsuccessful embolization. We recently reported the outcomes of four patients treated with radiosurgery with tumor doses ranging from 14–19 Gy at the margin
65
[133]. All patients had symptomatic improvement, and all had shown a dramatic reduction in the overall volume of their tumor. One patient had persistent diplopia. In our early experience, stereotactic radiosurgery proved to be a very effective management strategy, which avoided potentially serious complications associated with skull base microsurgery or embolization. The other reports including 3–5 cases in each, also achieved reduction in tumor volume after radiosurgery [123,134,135]. Hemangioblastoma
Over a 20-year interval 44 patients with intracranial hemangioblastomas, usually in conjunction with the syndrome of von Hippel-Lindau disease, have been treated by radiosurgery at our center. We studied outcome in 28 patients with 30 hemangioblastomas who had one year or longer follow-up. The mean patient age was 48 years (range, 28–83). The median tumor volume was 5.5 cc (range, 0.26–16.6). A median dose of 16 Gy (range, 11–20) was prescribed to tumor margins. Clinical and neuroimaging follow-up was obtained for all patients between 12–156 months (mean 41 months) after radiosurgery. Local tumor control was achieved in 28 of 30 tumors. The mean volume of lesions that were controlled by radiosurgery was 5 cc whereas median volume of tumors that failed radiosurgery was 11.5 cc. The lesions that were controlled by radiosurgery had received a median tumor margin dose of 16.4 Gy (range, 11–20) compared to 13.5 Gy (range, 13–14) prescribed to tumors that ultimately failed radiosurgery. The margin dose (16 Gy or more) was a significant predictor of tumor control after radiosurgery. At the last assessment, 20 patients (71%) were alive and eight (29%) had died. The mean survival after radiosurgery was 6.7 years. Early experience from several centers indicated that radiosurgery could lead to tumor control or regression [136,137] (> Figure 65-7). For the most part, we have treated tumors with documented tumor growth, which are usually solid and
1025
1026
65
Gamma knife: clinical experience
almost exclusively located in the posterior fossa, cerebellum, and brainstem. Such tumors are generally treated when they have shown evidence of objective growth and neurological symptoms develop. Prophylactic radiosurgery for hemangioblastomas in the case of von Hippel-Lindau disease (VHL) is not performed unless tumor growth or new symptoms are documented. For those patients with cystic hemangioblastomas, we have less optimism related to the overall role of radiosurgery at least as a single option. However, surgical removal of a large cystic component of a tumor producing mass effect symptoms is usually appropriate followed by radiosurgery for any residual solid component. In selected cases stereotactic aspiration of the cyst followed by subsequent radiosurgery is feasible. Repeat radiosurgery may be required over many years when other tumors show additional growth [138]. Chordoma and Chondrosarcoma
We continue to regard these tumors as difficult tumors to manage. Almost invariably they require multi-modality management over the course of many years. These invasive tumors provide a management challenge because of their critical location and their tendency to aggressively recur locally despite multi-modality treatment. Radiosurgery has been used both as a primary and adjuvant management strategy [126,139]. During our 20-year experience, 28 patients with chordoma and 19 patients with chondrosarcomas have undergone management with radiosurgery (> Table 65-6). We recently analyzed outcome of 33 eligible patients (chordoma 14 and chondrosarcomas 19). Five-year actuarial tumor control rates for chondrosarcoma and Chordomas after single procedure were 78 and 57% respectively [140] (> Figure 65-7). Our recent analysis indicated both the promise and difficulties of multi-modality management for chordomas and chondrosarcomas of the skull base. We found that chondrosarcomas generally respond well to radiosurgery. At present, radiosurgery for chordomas is best considered as
an adjuvant management except for very small biopsy proven chordomas wherein high dose, highly conformal and highly selective radiosurgery can be given. The addition of radiosurgery after surgical resection and radiation therapy or proton beam radiation has shown to increase both local tumor control and survival. For tumors without compression of the brainstem or local mass effect, SRS is an alternative option to surgical resection. In our experience, none of our patients developed adverse radiation effects. Radiosurgery appears to be a safe and effective management for small volume tumors, but over the course of many years, especially from 5 to 10 years after initial surgery and radiosurgery, recurrence rates continue to increase. In such cases, repeat radiosurgery or perhaps fractionated radiation therapy and repeat radiosurgery should be considered. Most such patients require one or more microsurgical approaches . Table 65-6 Gamma Knife radiosurgery for miscellaneous Neoplasms Indications Pineal Region Tumor Craniopharyngioma Hemangioblastoma Chondrosarcoma Chordoma Hemangiopericytoma Glomus Tumor Hemabngioma Myoepithelioma Rhabdomyosarcoma Esthesioneuroblastoma Choroid Plexux papilloma Colloid cust Hamartoma Lymphoma Neurofibrosarcoma Dysembryoplastic Neuroepithelial tumors (DNET) Fibrohistiocytoma Invasive skullbase tumors Others Total indications
Number of patients treated 29 68 44 19 28 35 19 8 1 3 4 11 1 6 11 1 2 5 31 21 347
Gamma knife: clinical experience
for tumor cytoreduction. More recently, we have embarked on the usage of endoscopic transsphenoidal resection followed by radiosurgery. Invasive Skull Base Cancers
After combined otolaryngological and neurosurgical procedures, we have used adjuvant radiosurgery for invasive skull base cancers (31 patients). Sixteen patients had adenocarcinomas, 14 had squamous cell carcinomas, and one patient had a metastatic neuroendocrine tumor. In such cases, radiosurgery has been used as an adjuvant or in combination with external beam fractionated radiation therapy. Many reports have documented the role of radiosurgery as salvage procedure for malignant tumors involving the skull base [139,141,142].
Radiosurgery for Functional Brain Disorders Trigeminal Neuralgia Radiosurgery At our institution 775 patients have undergone radiosurgery for management of trigeminal neuralgia. Our last detailed review studied 220 consecutive radiosurgery procedures for typical trigeminal neuralgia, all performed between 1992 and 1998 [143]. All 220 patients had trigeminal neuralgia that was idiopathic, long standing, and refractory to medication therapy. Most of the patients had a long history of medical treatment with the median symptom duration of 96 months (range, 3–564 months). Pain was predominantly distributed in the V2 and V3 distributions of the trigeminal nerve (29.5%), followed by V2 alone (22.3%) and V3 alone (13.2%). Prior surgery was performed in 135 patients (61.4%), including microvascular decompression, glycerol rhizotomy, radiofrequency rhizotomy, balloon microcompression, peripheral neurectomy, or ethanol injections. In the remaining 85 patients (38.6%), radiosurgery was the first surgical pro-
65
cedure performed. The median central dose at trigeminal nerve was 80 Gy. The pain relief after radiosurgery was graded into four categories: excellent, good, fair, and poor. Complete pain relief without the use of any medication was defined as an excellent outcome. Patients with complete pain relief with some medication were considered as good outcomes. Patients with partial pain relief (more than 50% pain relief) were considered to have a fair outcome [57]. No pain relief or less than 50% pain relief were considered as poor. Of the 220 patients, 47 (25.1%) required further additional surgical procedures because of insufficient pain control. These patients were considered as treatment failures (poor outcome), and the results after the additional procedure were excluded from this analysis. Most of the patients started experiencing pain relief within 6 months of radiosurgery (median, 2 months). At the initial followup assessment performed at 6 months, excellent results were obtained in 105 patients (47.7%), and excellent plus good results were found in 139 patients (63.2%). More than 50% pain relief (excellent, good, or fair) was noted in 181 patients (82.3%). At the last follow-up evaluation, 88 patients (40%) had excellent outcomes, 121 patients (55.9%) had excellent plus good outcomes, and 152 were fair or better (69.1%). Thirty patients (13.6%) had recurrence of pain after the initial achievement of pain relief (25 patients after complete relief, five patients after more than 50% relief) between 2 and 58 months after radiosurgery. Recurrences occurred at a mean of 15.4 months from irradiation. The median time to achieving complete pain relief (good or excellent) was also 2 months (2.05.1). Complete pain relief (good or excellent) was achieved in 64.93.2% of the patients at 6 months, 70.33.16% by 1 year, and in 75.43.49% of patients by 33 months. Complete pain relief (excellent or good) was achieved and maintained in 63.63.3% of patients at 1 year, 59.23.5% of patients at 2 years, and 56.63.8% of patients at 3 years. A history of no prior surgery
1027
1028
65
Gamma knife: clinical experience
was the only factor significantly associated (p = 0.01) with achieving and maintaining complete pain relief. No patient sustained an early complication after any radiosurgery procedure. Seventeen patients (7.7%) developed increased facial paresthesia and/or facial numbness that lasted more than 6 months. Others have noted a dry eye, without significant facial numbness. No patient developed a mastication deficit after radiosurgery or noted problems in facial motor function. One patient (0.4%) developed deafferentation pain after radiosurgery. The low incidence of complications is the greatest advantage of stereotactic radiosurgery compared to all other surgical options. Radiosurgery can be repeated if pain returns after initial relief. We advocate repeat radiosurgery only if complete pain relief had been achieved with subsequent recurrence [144]. We advocate a maximum dose of 60–70 Gy at a second procedure.
Movement Disorder Radiosurgery Stereotactic radiosurgery is an option to manage movement disorders for patients who are high risk for surgery and anesthesia. Gamma Knife is the preferred radiosurgical tool for treatment of movement disorders. We have treated 80 movement disorders patients with radiosurgery (> Table 65-7). Gamma knife thalamotomy
The VIM nucleus of the thalamus is the target for a tremor patient. We have previously reported our experience with the treatment of essential tremor (ET) and Parkinson’s disease (PD) tremor using gamma Knife radiosurgery. Niranjan et al. evaluated 11 patients managed with gamma knife thalamotomy for essential and Multiple sclerosis (MS) related tremor [145]. All patients noted improvement in action tremor. Six of eight ET patients had complete tremor arrest, and the violent action tremor in all three patients
. Table 65-7 Gamma Knife radiosurgery for Functional Brain Disorders Number of patients treated
Indication Trigeminal neuralgia Cancer Pain Movement Disorder
OCD Total
775
Essential tremor Parkinson’s Disease Others
2 40 30 10 3 869
with MS was improved. One patient developed transient arm weakness. Duma et al. treated 42 patients with tremor from PD or ET with VIM thalamotomy using Gamma Knife. Median time of onset of improvement was 2 months [146]. No change in tremor occurred in four gamma knife thalamotomies (8.6%), ‘‘mild’’ improvement was seen in four (8.6%), ‘‘good’’ improvement was seen in 13 (28%), and ‘‘excellent’’ improvement in 13 (28%). In 12 thalamotomies (26%), the tremor was eliminated completely. One patient, after bilateral treatment, suffered a mild acute dysarthria 1 week after GK thalamotomy. Ohye, et al. reported 36 Gamma thalamotomies in 31 patients. Maximum dose was 150 Gy in the first six cases, which was subsequently reduced to 130 Gy [6,147]. In two patients undergoing repeat procedures the dose was decreased to 120 Gy. In all cases except one, a single 4-mm isocenter was used. In their 15 cases with more than 2 years follow-up, a clinically good result was seen in 87%, with no noticeable side effects. In a more recent report, these authors have compared the results of 51 patients who had thalamotomy after reloading of gamma Knife with that of previous patients. Young, et al., in a large series of patients reviewed their use of GK thalamotomy for the treatment of tremor [148]. Their series included 102 patients
Gamma knife: clinical experience
with parkinsonian tremor, 52 patients with essential tremor, and four patients with tremor of other etiology. The single 4-mm collimator was used with doses varying from 110–160 Gy. At a median follow up of 52.5 months (11–93 months) 76% were tremor free, and 12% were ‘‘nearly free of tremor.’’ Thus there was failure in 12%. In 52 patients who had radiosurgery for disabling ET 92% were completely at 1 year and 88% were completely free after 4 years. Gamma Knife Pallidotomy
Duma et al. performed Gamma Knife pallidotomy on 18 patients with medically recalcitrant and disabling symptoms of PD. Only six patients (33%) showed improvement in rigidity and dyskinesia. Three patients (17%) were unchanged, and nine patients (50%) worsened. Of the six patients with improvement, two exhibited visual field deficits. Overall, four (22%) patients had a visual field deficit, three patients had speech and/or swallowing difficulties, three had worsening of their gait, and one had numbness in the contralateral hemibody. Nine patients (50%) had one or more complications related to treatment. Friedman, et al. had similar experience [149]. They described their results in four patients using Gamma Knife pallidotomy in advanced disease. No patient improved in a significant manner within the followup interval of 18 months. One patient experienced an improvement in his dyskinesia, but also became transiently psychotic and demented. The other three patients suffered no adverse effects. We do not recommend this procedure because of the high rate of morbidity noted. Radiosurgery for Epilepsy
Radiosurgery is an emerging therapeutic approach for the treatment of medically intractable seizures. Most experience of the treatment of medically refractory epilepsy is with gamma-knife radiosurgery for lesional epilepsy associated with arteriovenous malformations, cavernomas, and tumors. There have been some studies of the treatment of epilepsy associated with hypothalamic hamarto-
65
mas and mesial-temporal sclerosis as described below. There is evidence that gamma-knife radiosurgery improves seizure in patients affected by medically refractory epilepsy associated with hypothalamic hamartomas [150–153]. Overall, the results of these studies suggest that radiosurgery is a safe and effective option for the treatment of seizures associated with hypothalamic hamartomas. Marginal doses of 17 Gy or higher seem to be required when using gamma-knife radiosurgery. Regis et al. reported outcome of radiosurgery as a treatment for mesial-temporal-lobe epilepsy in a long-term multicentre trial [154]. Patients were selected for gamma-knife radiosurgery by use of the same criteria as for microsurgical amygdalohippocampectomy, including the presence of hippocampal sclerosis and the absence of space-occupying lesions. The target volume was approximately 7 cm3. The target included the head and body of the hippocampus, anterior part of the parahippocampal gyrus, and the basolateral region of the amygdaloid complex (sparing the upper and mesial part). A margin dose of 25 Gy was delivered to the 50% isodoseline. Analysis of seizure control after a 2-year follow-up showed a reduction of the median number of seizures from 6.2 to 0.3 per month. This study suggests that radiosurgery can be effective for medically refractory mesial-temporallobe epilepsy. The exact mechanism of seizure abolition after radiosurgery (or conventional irradiation) is not well understood. Depending on the dose and the target volume, radiosurgery can induce necrosis and consequent destruction of the epileptic focus and its pathways of spread. Alternatively, suppression of epileptic activity by a neuromodulatory effect at non-necrotising doses has been proposed as a possible mechanism of action [155]. In patients with seizures associated with arteriovenous malformations, seizure outcome seems to be independent of nidus occlusion [97,156,157], suggesting that radiosurgery can reduce seizure frequency through
1029
1030
65
Gamma knife: clinical experience
an intrinsic effect on the epileptogenic cortex surrounding the nidus. Radiosurgery of the rat hippocampus induces neuromodulatory effects such as reduction of the cholinergic and excitatory aminoacid concentrations with preservation of the GABAergic system, suggesting that it might be possible to modify an epileptogenic cortex sufficiently for it to become non-epileptic, while preserving its functional role [155]. The experimental evidence suggests that, at least in rats, it is possible to achieve seizure control with subnecrotic doses of radiation [158–160].
11.
12.
13.
14.
References 1. Ammar A. Lars Leksell’s vision–radiosurgery. Acta Neurochir Suppl 1994;62:1-4. 2. Lunsford LD. Lars Leksell: notes at the side of a raconteur. Stereotact Funct Neurosurg 1996;67:153-68. 3. Liscak R, Vladyka V, Simonova G, Urgosik D, Novotny J, Jr, Janouskova L, Vymazal J. Arteriovenous malformations after Leksell gamma knife radiosurgery: rate of obliteration and complications. Neurosurgery 2007;60:1005-14; discussion 1015-6. 4. Pollock BE, Gorman DA, Coffey RJ. Patient outcomes after arteriovenous malformation radiosurgical management: results based on a 5- to 14-year follow-up study. Neurosurgery 2003;52:1291-96; discussion 1296-7. 5. Pollock BE, Flickinger JC, Lunsford LD, Bissonette DJ, Kondziolka D. Hemorrhage risk after stereotactic radiosurgery of cerebral arteriovenous malformations. Neurosurgery 1996;38:652-9; discussion 659-61. 6. Inoue HK, Ohye C. Hemorrhage risks and obliteration rates of arteriovenous malformations after gamma knife radiosurgery. J Neurosurg 2002;97:474-6. 7. Karlsson B, Lax I, Soderman M. Risk for hemorrhage during the 2-year latency period following gamma knife radiosurgery for arteriovenous malformations. Int J Radiat Oncol Biol Phys 2001;49:1045-51. 8. Maruyama K, Kawahara N, Shin M, Tago M, Kishimoto J, Kurita H, Kawamoto S, Morita A, Kirino T. The risk of hemorrhage after radiosurgery for cerebral arteriovenous malformations. N Engl J Med 2005;352:146-53. 9. Flickinger JC, Kondziolka D, Maitz AH, Lunsford LD. An analysis of the dose-response for arteriovenous malformation radiosurgery and other factors affecting obliteration. Radiother Oncol 2002;63:347-54. 10. Pollock BE, Kondziolka D, Flickinger JC, Patel AK, Bissonette DJ, Lunsford LD. Magnetic resonance imaging: an accurate method to evaluate arteriovenous mal-
15.
16.
17.
18.
19.
20.
21.
22.
formations after stereotactic radiosurgery. J Neurosurg 1996;85:1044-9. Maesawa S, Flickinger JC, Kondziolka D, Lunsford LD. Repeated radiosurgery for incompletely obliterated arteriovenous malformations. J Neurosurg 2000;92:961-70. Chopra R, Kondziolka D, Niranjan A, Lunsford LD, Flickinger JC. Long-term follow-up of acoustic schwannoma radiosurgery with marginal tumor doses of 12 to 13 Gy. Int J Radiat Oncol Biol Phys 2007;68:845-51. Pan DH, Guo WY, Chung WY, Shiau CY, Chang YC, Wang LW. Gamma knife radiosurgery as a single treatment modality for large cerebral arteriovenous malformations. J Neurosurg 2000;93 Suppl 3:113-9. Miyawaki L, Dowd C, Wara W, Goldsmith B, Albright N, Gutin P, Halbach V, Hieshima G, Higashida R, Lulu B, Pitts L, Schell M, Smith V, Weaver K, Wilson C, Larson D. Five year results of LINAC radiosurgery for arteriovenous malformations: outcome for large AVMS. Int J Radiat Oncol Biol Phys 1999;44:1089-106. Pollock BE, Kline RW, Stafford SL, Foote RL, Schomberg PJ. The rationale and technique of stagedvolume arteriovenous malformation radiosurgery. Int J Radiat Oncol Biol Phys 2000;48:817-24. Sirin S, Kondziolka D, Niranjan A, Flickinger JC, Maitz AH, Lunsford LD. Prospective staged volume radiosurgery for large arteriovenous malformations: indications and outcomes in otherwise untreatable patients. Neurosurgery 2006;58:17-27; discussion 17-27. Raza SM, Jabbour S, Thai QA, Pradilla G, Kleinberg LR, Wharam M, Rigamonti D. Repeat stereotactic radiosurgery for high-grade and large intracranial arteriovenous malformations. Surg Neurol 2007;68:24-34; discussion 34. Flickinger JC, Kondziolka D, Lunsford LD, Kassam A, Phuong LK, Liscak R, Pollock B. Development of a model to predict permanent symptomatic postradiosurgery injury for arteriovenous malformation patients. Arteriovenous Malformation Radiosurgery Study Group. Int J Radiat Oncol Biol Phys 2000;46:1143-8. Hara M, Nakamura M, Shiokawa Y, Sawa H, Sato E, Koyasu H, Saito I. Delayed cyst formation after radiosurgery for cerebral arteriovenous malformation: two case reports. Minim Invasive Neurosurg 1998;41: 40-5. Izawa M, Hayashi M, Chernov M, Nakaya K, Ochiai T, Murata N, Takasu Y, Kubo O, Hori T, Takakura K. Long-term complications after gamma knife surgery for arteriovenous malformations. J Neurosurg 2005;102 Suppl:34-7. Pan HC, Sheehan J, Stroila M, Steiner M, Steiner L. Late cyst formation following gamma knife surgery of arteriovenous malformations. J Neurosurg 2005;102 Suppl:124-7. Koebbe CJ, Singhal D, Sheehan J, Flickinger JC, Horowitz M, Kondziolka D, Lunsford LD. Radiosurgery for dural arteriovenous fistulas. Surg Neurol 2005;64:392-8; discussion 398-399.
Gamma knife: clinical experience
23. Linskey ME, Johnstone PA, O’Leary M, Goetsch S. Radiation exposure of normal temporal bone structures during stereotactically guided gamma knife surgery for vestibular schwannomas.[see comment]. J Neurosurg 2003;98:800-6. 24. Pan DH, Chung WY, Guo WY, Wu HM, Liu KD, Shiau CY, Wang LW. Stereotactic radiosurgery for the treatment of dural arteriovenous fistulas involving the transverse-sigmoid sinus. J Neurosurg 2002;96: 823-9. 25. Pollock BE, Nichols DA, Garrity JA, Gorman DA, Stafford SL. Stereotactic radiosurgery and particulate embolization for cavernous sinus dural arteriovenous fistulae. Neurosurgery 1999;45:459-66; discussion 466-7. 26. Shin M, Kurita H, Tago M, Kirino T. Stereotactic radiosurgery for tentorial dural arteriovenous fistulae draining into the vein of Galen: report of two cases. Neurosurgery 2000;46:730-3; discussion 733-4. 27. Hasegawa T, McInerney J, Kondziolka D, Lee JY, Flickinger JC, Lunsford LD. Long-term results after stereotactic radiosurgery for patients with cavernous malformations. Neurosurgery 2002;50:1190-97; discussion 1197-8. 28. Liscak R, Vladyka V, Simonova G, Vymazal J, Novotny J, Jr. Gamma knife surgery of brain cavernous hemangiomas. J Neurosurg 2005;102 Suppl:207-13. 29. Chung WY, Liu KD, Shiau CY, Wu HM, Wang LW, Guo WY, Ho DM, Pan DH. Gamma knife surgery for vestibular schwannoma: 10-year experience of 195 cases. J Neurosurg 2005;102 Suppl:87-96. 30. Hasegawa T, Kida Y, Kobayashi T, Yoshimoto M, Mori Y, Yoshida J. Long-term outcomes in patients with vestibular schwannomas treated using gamma knife surgery: 10-year follow up. J Neurosurg 2005;102:10-6. 31. Kondziolka D, Nathoo N, Flickinger JC, Niranjan A, Maitz AH, Lunsford LD. Long-term results after radiosurgery for benign intracranial tumors.[see comment]. Neurosurgery 2003;53:815-21; discussion 821-2. 32. Maruyama K, Kondziolka D, Niranjan A, Flickinger JC, Lunsford LD. Stereotactic radiosurgery for brainstem arteriovenous malformations: factors affecting outcome. J Neurosurg 2004;100:407-13. 33. Delbrouck C, Hassid S, Massager N, Choufani G, David P, Devriendt D, Levivier M. Preservation of hearing in vestibular schwannomas treated by radiosurgery using Leksell Gamma Knife: preliminary report of a prospective Belgian clinical study. Acta Oto-RhinoLaryngologica Belgica 2003;57:197-204. 34. Flickinger JC, Kondziolka D, Niranjan A, Lunsford LD. Results of acoustic neuroma radiosurgery: an analysis of 5 years’ experience using current methods.[see comment]. J Neurosurg 2001;94:1-6. 35. Flickinger JC, Kondziolka D, Pollock BE, Lunsford LD. Evolution in technique for vestibular schwannoma radio-
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
65
surgery and effect on outcome. Int J Radiat Oncol Biol Phys 1996;36:275-80. Foote KD, Friedman WA, Buatti JM, Meeks SL, Bova FJ, Kubilis PS. Analysis of risk factors associated with radiosurgery for vestibular schwannoma. J Neurosurg 2001;95:440-9. Harsh GR, Thornton AF, Chapman PH, Bussiere MR, Rabinov JD, Loeffler JS. Proton beam stereotactic radiosurgery of vestibular schwannomas. Int J Radiat Oncol Biol Phys 2002;54:35-44. Horstmann GA, Van Eck AT. Gamma knife model C with the automatic positioning system and its impact on the treatment of vestibular schwannomas. J Neurosurg 2002;97:450-5. Inoue HK. Low-dose radiosurgery for large vestibular schwannomas: long-term results of functional preservation. J Neurosurg 2005;102 Suppl:111-3. Ishihara H, Saito K, Nishizaki T, Kajiwara K, Nomura S, Yoshikawa K, Harada K, Suzuki M. CyberKnife radiosurgery for vestibular schwannoma. Minim Invasive Neurosurg 2004;47:290-3. Kondziolka D, Lunsford LD, Flickinger JC. Gamma knife radiosurgery for vestibular schwannomas. Neurosurg Clin N Am 2000;11:651-8. Kondziolka D, Lunsford LD, McLaughlin MR, Flickinger JC. Long-term outcomes after radiosurgery for acoustic neuromas.[see comment]. N Engl J Med 1998;339:1426-33. Linskey ME. Stereotactic radiosurgery versus stereotactic radiotherapy for patients with vestibular schwannoma: a Leksell Gamma Knife Society 2000 debate. J Neurosurg 2000;93 Suppl 3:90-5. Linskey ME, Johnstone PA. Radiation tolerance of normal temporal bone structures: implications for gamma knife stereotactic radiosurgery. Int J Radiat Oncol Biol Phys 2003;57:196-200. Linskey ME, Lunsford LD, Flickinger JC. Tumor control after stereotactic radiosurgery in neurofibromatosis patients with bilateral acoustic tumors. Neurosurgery 1992;31:829-38; discussion 838-9. Meijer OW, Wolbers JG, Vandertop WP, Slotman BJ. Stereotactische bestraling van het vestibulair schwannoom (acusticusneurinoom). Nederlands Tijdschrift voor Geneeskunde 2000;144:2088-93. Petit JH, Hudes RS, Chen TT, Eisenberg HM, Simard JM, Chin LS. Reduced-dose radiosurgery for vestibular schwannomas. Neurosurgery 2001;49:1299-306; discussion 1306-7. Shamisa A, Bance M, Nag S, Tator C, Wong S, Noren G, Guha A. Glioblastoma multiforme occurring in a patient treated with gamma knife surgery. Case report and review of the literature. J Neurosurg 2001;94:816-21. Litvack ZN, Noren G, Chougule PB, Zheng Z. Preservation of functional hearing after gamma knife surgery for vestibular schwannoma. Neurosurg Focus 2003;14:e3.
1031
1032
65
Gamma knife: clinical experience
50. Niranjan A, Lunsford LD, Flickinger JC, Maitz A, Kondziolka D. Dose reduction improves hearing preservation rates after intracanalicular acoustic tumor radiosurgery. Neurosurgery 1999;45:753-62; discussion 762-5. 51. Subach BR, Kondziolka D, Lunsford LD, Bissonette DJ, Flickinger JC, Maitz AH. Stereotactic radiosurgery in the management of acoustic neuromas associated with neurofibromatosis type 2.[see comment]. J Neurosurg 1999;90:815-22. 52. Roche PH, Robitail S, Thomassin JM, Pellet W, Regis J. Radiochirurgie gamma knife des schwannomes vestibulaires associes a une neurofibromatose de type 2. Neuro Chir 2004;50:367-76. 53. Rowe J, Grainger A, Walton L, Silcocks P, Radatz M, Kemeny A. Risk of malignancy after gamma knife stereotactic radiosurgery. Neurosurgery 2007;60:60-5; discussion 65-6. 54. Mathieu D, Kondziolka D, Flickinger JC, Niranjan A, Williamson R, Martin JJ, Lunsford LD. Stereotactic radiosurgery for vestibular schwannomas in patients with neurofibromatosis type 2: an analysis of tumor control, complications, and hearing preservation rates. Neurosurgery 2007;60:460-8; discussion 468-70. 55. Kondziolka D, Levy EI, Niranjan A, Flickinger JC, Lunsford LD. Long-term outcomes after meningioma radiosurgery: physician and patient perspectives. J Neurosurg 1999;91:44-50. 56. Pollock BE, Stafford SL, Link MJ. Gamma knife radiosurgery for skull base meningiomas. Neurosurg Clin N Am 2000;11:659-66. 57. Chang SD, Adler JR, Jr. Treatment of cranial base meningiomas with linear accelerator radiosurgery. Neurosurgery 1997;41:1019-25; discussion 1025-7. 58. Duma CM, Lunsford LD, Kondziolka D, Harsh GR, Flickinger JC. Stereotactic radiosurgery of cavernous sinus meningiomas as an addition or alternative to microsurgery. Neurosurgery 1993;32:699-704; discussion 704-5. 59. Hakim R, Alexander E, III, Loeffler JS, Shrieve DC, Wen P, Fallon MP, Stieg PE, Black PM. Results of linear accelerator-based radiosurgery for intracranial meningiomas. Neurosurgery 1998;42:446-53; discussion 453-4. 60. Kondziolka D, Flickinger JC, Perez B. Judicious resection and/or radiosurgery for parasagittal meningiomas: outcomes from a multicenter review. Gamma Knife Meningioma Study Group. Neurosurgery 1998;43:405-13; discussion 413-4. 61. Liscak R, Simonova G, Vymazal J, Janouskova L, Vladyka V. Gamma knife radiosurgery of meningiomas in the cavernous sinus region. Acta Neurochir (Wien) 1999;141:473-80. 62. Morita A, Coffey RJ, Foote RL, Schiff D, Gorman D. Risk of injury to cranial nerves after gamma knife radiosurgery
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
for skull base meningiomas: experience in 88 patients. J Neurosurg 1999;90:42-9. Ojemann RG. Management of cranial and spinal meningiomas (honored guest presentation). Clin Neurosurg 1993;40:321-83. Roche PH, Regis J, Dufour H, Fournier HD, Delsanti C, Pellet W, Grisoli F, Peragut JC. Gamma knife radiosurgery in the management of cavernous sinus meningiomas. J Neurosurg 2000;93 Suppl 3:68-73. Shafron DH, Friedman WA, Buatti JM, Bova FJ, Mendenhall WM. Linac radiosurgery for benign meningiomas. Int J Radiat Oncol Biol Phys 1999;43:321-7. Stafford SL, Pollock BE, Foote RL, Link MJ, Gorman DA, Schomberg PJ, Leavitt JA. Meningioma radiosurgery: tumor control, outcomes, and complications among 190 consecutive patients. Neurosurgery 2001;49:1029-37; discussion 1037-8. Subach BR, Lunsford LD, Kondziolka D, Maitz AH, Flickinger JC. Management of petroclival meningiomas by stereotactic radiosurgery. Neurosurgery 1998;42: 437-43; discussion 443-5. Nicolato A, Foroni R, Alessandrini F, Bricolo A, Gerosa M. Radiosurgical treatment of cavernous sinus meningiomas: experience with 122 treated patients. Neurosurgery 2002;51:1153-59; discussion 1159-61. Aichholzer M, Bertalanffy A, Dietrich W, Roessler K, Pfisterer W, Ungersboeck K, Heimberger K, Kitz K. Gamma knife radiosurgery of skull base meningiomas. Acta Neurochir (Wien) 2000;142:647-52; discussion 652-3. Kondziolka D, Mathieu D, Lunsford LD, Martin JJ, Madhok R, Niranjan A, Flickinger JC. Radiosurgery As Definitive Management Of Meningiomas. Neurosurgery 2008;62(1):53-8. Couldwell WT, Fukushima T, Giannotta SL, Weiss MH. Petroclival meningiomas: surgical experience in 109 cases. J Neurosurg 1996;84:20-8. De Jesus O, Sekhar LN, Parikh HK, Wright DC, Wagner DP. Long-term follow-up of patients with meningiomas involving the cavernous sinus: recurrence, progression, and quality of life. Neurosurgery 1996;39:915-9; discussion 919-20. DeMonte F, Smith HK, al-Mefty O. Outcome of aggressive removal of cavernous sinus meningiomas. J Neurosurg 1994;81:245-51. Iwai Y, Yamanaka K, Ishiguro T. Gamma knife radiosurgery for the treatment of cavernous sinus meningiomas. Neurosurgery 2003;52:517-24; discussion 523-4. Kurita H, Sasaki T, Kawamoto S, Taniguchi M, Terahara A, Tago M, Kirino T. Role of radiosurgery in the management of cavernous sinus meningiomas. Acta Neurol Scand 1997;96:297-304. Pollock BE, Stafford SL, Utter A, Giannini C, Schreiner SA. Stereotactic radiosurgery provides equivalent tumor control to Simpson Grade 1 resection for patients with
Gamma knife: clinical experience
small- to medium-size meningiomas. Int J Radiat Oncol Biol Phys 2003;55:1000-5. 77. Sheehan JP, Kondziolka D, Flickinger J, Lunsford LD. Radiosurgery for nonfunctioning pituitary adenoma. Neurosurg Focus 2003;14:e9. 78. Attanasio R, Epaminonda P, Motti E, Giugni E, Ventrella L, Cozzi R, Farabola M, Loli P, Beck-Peccoz P, Arosio M. Gamma-knife radiosurgery in acromegaly: a 4-year follow-up study. J Clin Endocrinol Metab 2003;88:3105-12. 79. Castinetti F, Taieb D, Kuhn JM, Chanson P, Tamura M, Jaquet P, Conte-Devolx B, Regis J, Dufour H, Brue T. Outcome of gamma knife radiosurgery in 82 patients with acromegaly: correlation with initial hypersecretion. J Clin Endocrinol Metab 2005;90:4483-88. 80. Gutt B, Wowra B, Alexandrov R, Uhl E, Schaaf L, Stalla GK, Schopohl J. Gamma-knife surgery is effective in normalising plasma insulin-like growth factor I in patients with acromegaly. Exp Clin Endocrinol Diabetes 2005;113:219-24. 81. Pollock BE, Flickinger JC. A proposed radiosurgery-based grading system for arteriovenous malformations. J Neurosurg 2002;96:79-85. 82. Sheehan JP, Niranjan A, Sheehan JM, Jane JA, Jr, Laws ER, Kondziolka D, Flickinger J, Landolt AM, Loeffler JS, Lunsford LD. Stereotactic radiosurgery for pituitary adenomas: an intermediate review of its safety, efficacy, and role in the neurosurgical treatment armamentarium. J Neurosurg 2005;102:678-91. 83. Kondziolka D, Maitz AH, Niranjan A, Flickinger JC, Lunsford LD. An evaluation of the Model C gamma knife with automatic patient positioning. Neurosurgery 2002;50:429-31; discussion 431-2. 84. Laws ER, Jr, Vance ML. Radiosurgery for pituitary tumors and craniopharyngiomas. Neurosurg Clin N Am 1999;10:327-36. 85. Kobayashi T, Kida Y, Mori Y. Gamma knife radiosurgery in the treatment of Cushing disease: long-term results. J Neurosurg 2002;97:422-28. 86. Lippitz BE, Kraepelien T, Hautanen K, Ritzling M, Rahn T, Ulfarsson E, Boethius J. Gamma knife radiosurgery for patients with multiple cerebral metastases. Acta Neurochir Suppl 2004;91:79-87. 87. Sheehan JM, Vance ML, Sheehan JP, Ellegala DB, Laws ER Jr. Radiosurgery for Cushing’s disease after failed transsphenoidal surgery. J Neurosurg 2000;93:738-742. 88. Vladyka V, Liscak R, Novotny J, Jr, Marek J, Jezkova J. Radiation tolerance of functioning pituitary tissue in gamma knife surgery for pituitary adenomas. Neurosurgery 2003;52:309-16; discussion 316-7. 89. Pollock BE, Nippoldt TB, Stafford SL, Foote RL, Abboud CF. Results of stereotactic radiosurgery in patients with hormone-producing pituitary adenomas: factors associated with endocrine normalization. J Neurosurg 2002;97:525-30.
65
90. Chiou SM, Lunsford LD, Niranjan A, Kondziolka D, Flickinger JC. Stereotactic radiosurgery of residual or recurrent craniopharyngioma, after surgery, with or without radiation therapy. Neuro Oncol 2001;3:159-66. 91. Hasegawa T, Kondziolka D, Hadjipanayis CG, Lunsford LD. Management of cystic craniopharyngiomas with phosphorus-32 intracavitary irradiation. Neurosurgery 2004;54:813-20; discussion 820-2. 92. Amendola BE, Wolf AL, Coy SR, Amendola M, Bloch L. Gamma knife radiosurgery in the treatment of patients with single and multiple brain metastases from carcinoma of the breast. Cancer J 2000;6:88-92. 93. Chung WY, Pan DH, Shiau CY, Guo WY, Wang LW. Gamma knife radiosurgery for craniopharyngiomas. J Neurosurg 2000;93 Suppl 3:47-56. 94. McDermott MW, Cosgrove GR, Larson DA, Sneed PK, Gutin PH. Interstitial brachytherapy for intracranial metastases. Neurosurg Clin N Am 1996;7:485-95. 95. Nwokedi EC, DiBiase SJ, Jabbour S, Herman J, Amin P, Chin LS. Gamma knife stereotactic radiosurgery for patients with glioblastoma multiforme. Neurosurgery 2002;50:41-6; discussion 46-7. 96. Simonova G, Novotny J, Jr, Liscak R. Low-grade gliomas treated by fractionated gamma knife surgery. J Neurosurg 2005,102 Suppl:19-24. 97. Kida Y, Kobayashi T, Mori Y. Gamma knife radiosurgery for low-grade astrocytomas: results of long-term follow up. J Neurosurg 2000;93 Suppl 3:42-6. 98. Hadjipanayis CG, Levy EI, Niranjan A, Firlik AD, Kondziolka D, Flickinger JC, Lunsford LD. Stereotactic radiosurgery for motor cortex region arteriovenous malformations. Neurosurgery 2001;48:70-6; discussion 76-7. 99. Chang EL, Wefel JS, Maor MH, Hassenbusch SJ, III, Mahajan A, Lang FF, Woo SY, Mathews LA, Allen PK, Shiu AS, Meyers CA. A pilot study of neurocognitive function in patients with one to three new brain metastases initially treated with stereotactic radiosurgery alone. Neurosurgery 2007;60:277-83; discussion 283-4. 100. Szeifert GT, Massager N, DeVriendt D, David P, De Smedt F, Rorive S, Salmon I, Brotchi J, Levivier M. Observations of intracranial neoplasms treated with gamma knife radiosurgery. J Neurosurg 2002;97:623-6. 101. Auchter RM, Lamond JP, Alexander E, Buatti JM, Chappell R, Friedman WA, Kinsella TJ, Levin AB, Noyes WR, Schultz CJ, Loeffler JS, Mehta MP. A multiinstitutional outcome and prognostic factor analysis of radiosurgery for resectable single brain metastasis. Int J Radiat Oncol Biol Phys 1996;35:27-35. 102. Bhatnagar AK, Kondziolka D, Lunsford LD, Flickinger JC. Recursive partitioning analysis of prognostic factors for patients with four or more intracranial metastases treated with radiosurgery. Technol Cancer Res Treat 2007;6:153-60. 103. Chernov MF, Nakaya K, Izawa M, Hayashi M, Usuba Y, Kato K, Muragaki Y, Iseki H, Hori T, Takakura K.
1033
1034
65 104.
105.
106.
107.
108.
109.
110.
111.
112.
Gamma knife: clinical experience
Outcome after radiosurgery for brain metastases in patients with low Karnofsky performance scale (KPS) scores. Int J Radiat Oncol Biol Phys 2007;67:1492-8. DiLuna ML, King JT, Jr, Knisely JP, Chiang VL. Prognostic factors for survival after stereotactic radiosurgery vary with the number of cerebral metastases. Cancer 2007;109:135-45. Flickinger JC, Kondziolka D, Lunsford LD, Coffey RJ, Goodman ML, Shaw EG, Hudgins WR, Weiner R, Harsh GR, Sneed PK, et al. A multi-institutional experience with stereotactic radiosurgery for solitary brain metastasis. Int J Radiat Oncol Biol Phys 1994;28:797-802. Hussain A, Brown PD, Stafford SL, Pollock BE. Stereotactic radiosurgery for brainstem metastases: Survival, tumor control, and patient outcomes. Int J Radiat Oncol Biol Phys 2007;67:521-4. Mathieu D, Kondziolka D, Cooper PB, Flickinger JC, Niranjan A, Agarwala S, Kirkwood J, Lunsford LD. Gamma knife radiosurgery in the management of malignant melanoma brain metastases. Neurosurgery 2007;60:471-81; discussion 481-2. Andrews DW, Scott CB, Sperduto PW, Flanders AE, Gaspar LE, Schell MC, Werner-Wasik M, Demas W, Ryu J, Bahary JP, Souhami L, Rotman M, Mehta MP, Curran WJ, Jr. Whole brain radiation therapy with or without stereotactic radiosurgery boost for patients with one to three brain metastases: phase III results of the RTOG 9508 randomised trial. Lancet 2004;363:1665-72. Aoyama H, Shirato H, Tago M, Nakagawa K, Toyoda T, Hatano K, Kenjyo M, Oya N, Hirota S, Shioura H, Kunieda E, Inomata T, Hayakawa K, Katoh N, Kobashi G. Stereotactic radiosurgery plus whole-brain radiation therapy vs stereotactic radiosurgery alone for treatment of brain metastases: a randomized controlled trial. JAMA 2006;295:2483-91. Chidel MA, Suh JH, Reddy CA, Chao ST, Lundbeck MF, Barnett GH. Application of recursive partitioning analysis and evaluation of the use of whole brain radiation among patients treated with stereotactic radiosurgery for newly diagnosed brain metastases. Int J Radiat Oncol Biol Phys 2000;47:993-9. Nieder C, Nestle U, Motaref B, Walter K, Niewald M, Schnabel K. Prognostic factors in brain metastases: should patients be selected for aggressive treatment according to recursive partitioning analysis (RPA) classes? Int J Radiat Oncol Biol Phys 2000;46:297-302. Chitapanarux I, Goss B, Vongtama R, Frighetto L, De Salles A, Selch M, Duick M, Solberg T, Wallace R, Cabatan-Awang C, Ford J. Prospective study of stereotactic radiosurgery without whole brain radiotherapy in patients with four or less brain metastases: incidence of intracranial progression and salvage radiotherapy. J Neuro Oncol 2003;61:143-9.
113. Flickinger JC, Kondziolka D, Lunsford LD, Pollock BE, Yamamoto M, Gorman DA, Schomberg PJ, Sneed P, Larson D, Smith V, McDermott MW, Miyawaki L, Chilton J, Morantz RA, Young B, Jokura H, Liscak R. A multi-institutional analysis of complication outcomes after arteriovenous malformation radiosurgery. Int J Radiat Oncol Biol Phys 1999;44:67-74. 114. Hasegawa T, Kondziolka D, Flickinger JC, Germanwala A, Lunsford LD. Brain metastases treated with radiosurgery alone: an alternative to whole brain radiotherapy? Neurosurgery 2003;52:1318-26; discussion 1326. 115. Lutterbach J, Cyron D, Henne K, Ostertag CB. Radiosurgery followed by planned observation in patients with one to three brain metastases. Neurosurgery 2003;52:1066-73; discussion 1073-4. 116. Lavine SD, Petrovich Z, Cohen-Gadol AA, Masri LS, Morton DL, O’Day SJ, Essner R, Zelman V, Yu C, Luxton G, Apuzzo ML. Gamma knife radiosurgery for metastatic melanoma: an analysis of survival, outcome, and complications. Neurosurgery 1999;44:59-64; discussion 64-6. 117. Patchell RA, Tibbs PA, Regine WF, Dempsey RJ, Mohiuddin M, Kryscio RJ, Markesbery WR, Foon KA, Young B. Postoperative radiotherapy in the treatment of single metastases to the brain: a randomized trial. JAMA 1998;280:1485-9. 118. Posner JB. Brain metastases: 1995. A brief review. J Neurooncol 1996.27:287-93. 119. Pirzkall A, Debus J, Lohr F, Fuss M, Rhein B, Engenhart-Cabillic R, Wannenmacher M. Radiosurgery alone or in combination with whole-brain radiotherapy for brain metastases. J Clin Oncol 1998;16:3563-9. 120. Sneed PK, Suh JH, Goetsch SJ, Sanghavi SN, Chappell R, Buatti JM, Regine WF, Weltman E, King VJ, Breneman JC, Sperduto PW, Mehta MP. A multi-institutional review of radiosurgery alone vs. radiosurgery with whole brain radiotherapy as the initial management of brain metastases. Int J Radiat Oncol Biol Phys 2002;53:519-26. 121. Sanghavi SN, Miranpuri SS, Chappell R, Buatti JM, Sneed PK, Suh JH, Regine WF, Weltman E, King VJ, Goetsch SJ, Breneman JC, Sperduto PW, Scott C, Mabanta S, Mehta MP. Radiosurgery for patients with brain metastases: a multi-institutional analysis, stratified by the RTOG recursive partitioning analysis method. Int J Radiat Oncol Biol Phys 2001;51:426-34. 122. Rades D, Bohlen G, Pluemer A, Veninga T, Hanssens P, Dunst J, Schild SE. Stereotactic radiosurgery alone versus resection plus whole-brain radiotherapy for 1 or 2 brain metastases in recursive partitioning analysis class 1 and 2 patients. Cancer 2007;109(12):2515-21. 123. Kobayashi T, Kida Y, Mori Y. Long-term results of stereotactic gamma radiosurgery of meningiomas. Surg Neurol 2001;55:325-31.
Gamma knife: clinical experience
124. Reyns N, Blond S, Gauvrit JY, Touzet G, Coche B, Pruvo JP, Dhellemmes P. Role of radiosurgery in the management of cerebral arteriovenous malformations in the pediatric age group: data from a 100-patient series. Neurosurgery 2007;60:268-76; discussion 276. 125. Miller RC, Foote RL, Coffey RJ, Gorman DA, Earle JD, Schomberg PJ, Kline RW. The role of stereotactic radiosurgery in the treatment of malignant skull base tumors. Int J Radiat Oncol Biol Phys 1997;39:977-81. 126. Muthukumar N, Kondziolka D, Lunsford LD, Flickinger JC. Stereotactic radiosurgery for tentorial meningiomas. Acta Neurochir (Wien) 1998;140: 315-20; discussion 320-1. 127. Nettel B, Niranjan A, Martin JJ, Koebbe CJ, Kondziolka D, Flickinger JC, Lunsford LD. Gamma knife radiosurgery for trigeminal schwannomas. Surg Neurol 2004;62:435-44; discussion 444-6. 128. Pan L, Wang EM, Zhang N, Zhou LF, Wang BJ, Dong YF, Dai JZ, Cai PW. Long-term results of Leksell gamma knife surgery for trigeminal schwannomas. J Neurosurg 2005;102 Suppl:220-4. 129. Saringer W, Khayal H, Ertl A, Schoeggl A, Kitz K. Efficiency of gamma knife radiosurgery in the treatment of glomus jugulare tumors. Minim Invasive Neurosurg 2001;44:141-6. 130. Wang LG, Guo Y, Zhang X, Song SJ, Xia JL, Fan FY, Shi M, Wei LC. Brain metastasis: experience of the Xi-Jing hospital. Stereotact Funct Neurosurg 2002;78:70-83. 131. Muthukumar N, Kondziolka D, Lunsford LD, Flickinger JC. Stereotactic radiosurgery for anterior foramen magnum meningiomas. Surg Neurol 1999;51: 268-73. 132. Pollock BE. stereotactic radiosurgery in patients with glomus jugulare tumors. Neurosurg Focus 2004;17:E 10. 133. Levy EI, Niranjan A, Thompson TP, Scarrow AM, Kondziolka D, Flickinger JC, Lunsford LD. Radiosurgery for childhood intracranial arteriovenous malformations. Neurosurgery 2000;47:834-41; discussion 841-2. 134. Nakamura N, Shin M, Tago M, Terahara A, Kurita H, Nakagawa K, Ohtomo K. Gamma knife radiosurgery for cavernous hemangiomas in the cavernous sinus. Report of three cases. J Neurosurg 2002;97:477-80. 135. Peker S, Ozduman K, Kilic T, Pamir MN. Relief of hemifacial spasm after radiosurgery for intracanalicular vestibular schwannoma. Minim Invasive Neurosurg 2004;47:235-7. 136. Georg AE, Lunsford LD, Kondziolka D, Flickinger JC, Maitz A. Hemangioblastoma of the posterior fossa. The role of multimodality treatment. Arquivos de NeuroPsiquiatria 1997;55:278-86. 137. Pan L, Wang EM, Wang BJ, Zhou LF, Zhang N, Cai PW, Da JZ. Gamma knife radiosurgery for hemangioblastomas. Stereotactic Funct Neurosurg 1998;70 Suppl 1:179-86.
65
138. Jawahar A, Shaya M, Campbell P, Ampil F, Willis BK, Smith D, Nanda A. Role of stereotactic radiosurgery as a primary treatment option in the management of newly diagnosed multiple (3–6) intracranial metastases. Surg Neurol 2005;64:207-12. 139. Lunsford LD, Kondziolka D, Flickinger JC, Bissonette DJ, Jungreis CA, Maitz AH, Horton JA, Coffey RJ. Stereotactic radiosurgery for arteriovenous malformations of the brain. J Neurosurg 1991;75:512-24. 140. Martin JJ, Niranjan A, Kondziolka D, Flickinger JC, Lozanne KA, Lunsford LD. Radiosurgery for chordomas and chondrosarcomas of the skull base. J Neurosurg 2007;107:758-64. 141. Firlik KS, Kondziolka D, Lunsford LD, Janecka IP, Flickinger JC. Radiosurgery for recurrent cranial base cancer arising from the head and neck. Head Neck 1996;18:160-5; discussion 166. 142. Lee DJ, Westra WH, Staecker H, Long D, Niparko JK, Slattery WH, III. Clinical and histopathologic features of recurrent vestibular schwannoma (acoustic neuroma) after stereotactic radiosurgery. Otol Neurotol 2003;24:650-60; discussion 660. 143. Maesawa S, Salame C, Flickinger JC, Pirris S, Kondziolka D, Lunsford LD. Clinical outcomes after stereotactic radiosurgery for idiopathic trigeminal neuralgia. J Neurosurg 2001;94:14-20. 144. Hasegawa T, Kondziolka D, Spiro R, Flickinger JC, Lunsford LD. Repeat radiosurgery for refractory trigeminal neuralgia. Neurosurgery 2002;50:494-500; discussion 500-2. 145. Niranjan A, Kondziolka D, Baser S, Heyman R, Lunsford LD. Functional outcomes after gamma knife thalamotomy for essential tremor and MS-related tremor. Neurology 2000;55:443-6. 146. Duma CM. Movement Disorder Radiosurgery. In: Loftus CM, Batjer HH, editors. Techniques in Neurosurgery. Philadelphia: Williams and Wilkins, 2003, pp 181-90. 147. Ohye C, Shibazaki T, Ishihara J, Zhang J. Evaluation of gamma thalamotomy for parkinsonian and other tremors: survival of neurons adjacent to the thalamic lesion after gamma thalamotomy. J Neurosurg 2000;93 Suppl 3:120-7. 148. Regine WF, Huhn JL, Patchell RA, St Clair WH, Strottmann J, Meigooni A, Sanders M, Young AB. Risk of symptomatic brain tumor recurrence and neurologic deficit after radiosurgery alone in patients with newly diagnosed brain metastases: results and implications. Int J Radiat Oncol Biol Phys 2002;52:333-8. 149. Friedman DP, Goldman HW, Flanders AE, Gollomp SM, Curran WJ, Jr. Stereotactic radiosurgical pallidotomy and thalamotomy with the gamma knife: MR imaging findings with clinical correlation – preliminary experience. Radiology 1999;212:143-50.
1035
1036
65
Gamma knife: clinical experience
150. Mathieu D, Kondziolka D, Niranjan A, Flickinger J, Lunsford LD. Gamma knife radiosurgery for refractory epilepsy caused by hypothalamic hamartomas. Stereotact Funct Neurosurg 2006;84:82-7. 151. Regis J, Hayashi M, Eupierre LP, Villeneuve N, Bartolomei F, Brue T, Chauvel P. Gamma knife surgery for epilepsy related to hypothalamic hamartomas. Acta Neurochir Suppl 2004;91:33-50. 152. Regis J, Scavarda D, Tamura M, Nagayi M, Villeneuve N, Bartolomei F, Brue T, Dafonseca D, Chauvel P. Epilepsy related to hypothalamic hamartomas: surgical management with special reference to gamma knife surgery. Childs Nerv Syst 2006;22:881-95. 153. Regis J, Scavarda D, Tamura M, Villeneuve N, Bartolomei F, Brue T, Morange I, Dafonseca D, Chauvel P. Gamma knife surgery for epilepsy related to hypothalamic hamartomas. Semin Pediatr Neurol 2007;14:73-9. 154. Regis J, Rey M, Bartolomei F, Vladyka V, Liscak R, Schrottner O, Pendl G. Gamma knife surgery in mesial temporal lobe epilepsy: a prospective multicenter study. Epilepsia 2004;45:504-15. 155. Regis J, Kerkerian-Legoff L, Rey M, Vial M, Porcheron D, Nieoullon A, Peragut JC. First biochemical evidence of differential functional effects following Gamma Knife
156.
157.
158.
159.
160.
surgery. Stereotact Funct Neurosurg 1996;66 Suppl 1:29-38. Eisenschenk S, Gilmore RL, Friedman WA, Henchey RA. The effect of LINAC stereotactic radiosurgery on epilepsy associated with arteriovenous malformations. Stereotact Funct Neurosurg 1998;71:51-61. Steiner L, Lindquist C, Adler JR, Torner JC, Alves W, Steiner M. Clinical outcome of radiosurgery for cerebral arteriovenous malformations. J Neurosurg 1992;77:1-8. Liscak R, Vladyka V, Novotny J, Jr, Brozek G, Namestkova K, Mares V, Herynek V, Jirak D, Hajek M, Sykova E. Leksell gamma knife lesioning of the rat hippocampus: the relationship between radiation dose and functional and structural damage. J Neurosurg 2002;97:666-73. Maesawa S, Kondziolka D, Dixon CE, Balzer J, Fellows W, Lunsford LD. Subnecrotic stereotactic radiosurgery controlling epilepsy produced by kainic acid injection in rats. J Neurosurg 2000;93:1033-40. Mori Y, Kondziolka D, Balzer J, Fellows W, Flickinger JC, Lunsford LD, Thulborn KR. Effects of stereotactic radiosurgery on an animal model of hippocampal epilepsy. Neurosurgery 2000;46:157-65; discussion 165-8.
58 Gamma Knife: Technical Aspects D. J. Schlesinger . C. P. Yen . C. Lindquist . L. Steiner
Lars Leksell designed the Gamma Knife1 unit (> Figure 58-1) from the ground up to be a tool of the neurosurgeon. The physics and engineering choices inherent in the design of the unit is perhaps best considered from the perspective of a surgeon end-user. Of all of the radiosurgical tools, the Gamma Knife is the oldest and best established. Since its first conception by Lars Leksell and Bo¨rje Larsson, the system has evolved through several generations of improvements; however the basic principles of the instrument remain primarily unchanged. This chapter will describe the physical principles and method of operation of the Gamma Knife, paying particular attention to the new features and changes found in the recently released Leksell Gamma Knife1 Perfexion™.
Introduction and Overview of Basic Concepts The technique Lars Leksell coined as ‘‘stereotactic radiosurgery’’ in 1951 grew out of his prior work in stereotactic neurosurgery and the idea that such surgery could be made ‘‘bloodless’’ using cross-firing beams of ionizing radiation [1,2]. Leksell’s early experiments attempted to harness the superior dosimetric and biological effect of proton beams generated by a cyclotron [3]. However, because of the extreme expense and difficulty associated with proton beams, Leksell abandoned this approach in search of alternative energy sources. Finally, after many prototypes and many failures, Leksell settled on the gamma emissions of cobalt-60 as an energy source that would best fit his goal of a minimallyinvasive form of stereotactic neurosurgery. #
Springer-Verlag Berlin/Heidelberg 2009
60
Co Radioactive Decay
The cobalt-60 decay series begins with the creation 59 of radioactive 60 27 Co from stable 27 Co by bombarding it with neutrons in a reactor. 60 27 Co is therefore neutron-heavy, and lies above the line of stability with a high neutron/proton ratio. To regain a stable configuration, 60 27 Co decays through b-decay (> Figure 58-2). The nucleus first emits (with >99% probability) a b particle (or negatron) with energy 0.32 MeV, and a neutrino, both created when a neutron in the nucleus spontaneously disintegrates. The recoiling nucleus then releases two gamma photons at 1.17 and 1.33 MeV respectively [4]. In a Gamma Knife unit, the b-particles are absorbed by the source itself. It is the high-energy gamma photons which create the clinical effect through indirect ionization of tissue molecules, OH-free radical formation, and resulting chemical reactions with vital cellular targets such as chromosomal DNA [5].
Superposition of Beams Depending on the model of the unit, a Gamma Knife contains an array of 201 (model B and C) or 192 (Perfexion) individual cobalt-60 sources aligned with a collimation system. The collimation system (described in more detail below) directs the individual beams of gamma radiation to a very precise focus point. While an individual beam has a relatively low dose rate and causes minimal biological effect, the superposition of all beams at the focus point have a much higher dose rate. This simultaneous cross-firing of beams is the heart of the Gamma Knife technique
898
58
Gamma knife: technical aspects
. Figure 58-1 The Gamma Knife Perfexion at the Lars Leksell Center for Gamma Surgery at the University of Virginia
. Figure 58-2 The creation and radioactive decay of 60Co
and creates an advantageous target to normal tissue biological effect. The Gamma Knife can very precisely target small areas of tissue with a high radiation dose. However, because the energy
is spread out among the individual beams, the Gamma Knife can also achieve a very large dose gradient outside of the target. Gamma Knife treatments are therefore quite heterogeneous in
Gamma knife: technical aspects
terms of dose distribution inside and outside of the target – a very different dose characteristic than is found in radiotherapy. Gamma Knife surgery is a high-dose, single-fraction procedure. Tumor control and tissue-sparing is achieved via the steep dose gradient at the target periphery rather than utilizing the radiobiological differential between normal and target tissue as in fractional radiotherapy. This suggests that a radiobiological effect different from that considered by traditional radiation oncology may be active during a Gamma Knife treatment [6,7].
58
. Figure 58-3 The Leksell stereotactic G-Frame and the Leksell coordinate system. (Image courtesy of Elekta AB, Stockholm)
Stereotactic Localization The ability to direct an array of beams on a single point does no good by itself unless the intended target can be precisely located at the focus point. The Gamma Knife operates on the principle of stereotaxy to achieve a high level of precision in localization. As with classic stereotactic neurosurgery, a stereotactic frame is mounted to a patient’s head and defines a reference Cartesian coordinate system known as the Leksell Coordinate System (> Figure 58-3). The base ring of Leksell frame defines the inferior limit of the coordinate system, with the origin located at a point to the right lateral, superior, and posterior to the base ring [3,8]. For the Gamma Knife, the arc and probe used for standard stereotactic procedures has been replaced by the cross-firing configuration of gamma rays. Patients are imaged with the frame attached as well as an external fiducial system which is visible on the resultant imagery. These images are used by the treatment planning system to locate targets in terms of this coordinate space. Once the coordinates for a given target are known, the patient can be positioned in the unit in such a way that the target coincides with the radiation focus point. During treatment, the stereotactic frame is locked to the Gamma
Knife unit so there is a direct connection between the internal machine coordinate system and the frame-defined stereotactic coordinate system.
Technical Description of the Gamma Knife The Gamma Knife with its plastic covers removed resembles a large metal sphere. The vast majority of this 20 metric ton ball of cast iron and tungsten is shielding that exists to protect approximately 20 g of 60Co at the heart of the unit. A Gamma Knife system, regardless of model, consists of six primary components as summarized in > Table 58-1.
The Radiation Body The radiation body (> Figure 58-4) makes up the bulk of what one sees when looking at a Gamma Knife unit. The radiation body is a more or less
899
900
58
Gamma knife: technical aspects
. Table 58-1 The primary components of the Gamma Knife Component
Purpose
Radiation body
Primary shielding for the 60Co sources Container for sources and primary collimation system. Container for source sectors in the Perfexion model Mechanism to connect the stereotactic frame to the unit A mechanism to accurately position the patient in the radiation field Allows user control of the Gamma Knife Create appropriate radiation distributions
Central body
Frame docking mechanism Stereotactic positioning system Control panel Treatment planning station(s)
. Figure 58-4 The radiation body of the Gamma Knife model C. The five ports on the side of the radiation body align with reach row of 60Co sources to facilitate loading and unloading. The round control panel on the back of the unit facing the reader is part of the bearing of the central body. During loading and unloading the central body can rotate so a particular set of sources aligns with the loading ports on the side
spherically shaped container that serves as bulk shielding for the 60Co sources and houses the central body. On the front of the radiation body is a steel door which allows passage of the proximal end of the treatment table into the unit. The door has manual controls for opening or closing the entrance door in the unlikely event of a complete loss of system power. For loading and unloading sources, the radiation body has a series of loading channels formed in the right side which align with each row of sources in the central body. source with a corresponding loading channel on the radiation unit.
Central Body and Collimation System The 60Co Sources Inside the radiation unit is the central body, which houses the 60Co sources and the primary collimators (and in the Perfexion model Gamma Knife also houses the collimators, source sectors, and sector drive rods, with the sector motors at the rear of the unit). The central body is mounted on a bearing that is mechanically fixed during clinical use, but can be released and rotated during loading and unloading operations to align each
The 60Co sources are located within the central body of the unit in specially constructed source bushings (> Figure 58-5). Each source is a series of 18 60Co pellets, held inside a series of three welded stainless-steel cylinders, which are themselves held within the steel bushing assembly. Each source has an activity of slightly more than 30 Ci for a total activity at loading of
Gamma knife: technical aspects
58
. Figure 58-5 The 60Co source assembly, showing the outer two steel cylinders and the outer assembly
approximately 6,000 Ci, providing a dose rate of greater than 3 Gy/min at the focal point.
Collimation System The details of the structure and operation of the central body and collimation system are one of the biggest differences between the Gamma Knife Perfexion and earlier B, and C models. In the earlier models, the 201 60Co source assemblies are configured in five concentric rings within the hemispherical central body. The sources are at a constant 400 mm from the focal point. Precisely machined beam channels consisting of a precollimator and a tungsten primary collimator are matched to the source assemblies. Each beam aligns with the focus point at the center of the unit with a tolerance of <0.3 mm3. Once loaded, the source array is mechanically fixed on its bearing and remains stationary (including during treatments). Perhaps the most visually distinctive part of the B and C models is the external helmets which provide final beam collimation(> Figure 58-6a).
Each helmet has beam channels corresponding to the beam channels in the inner radiation body. The channels are fitted with removable tungsten-alloy collimators with circular apertures that result in a 4, 8, 14, or 18 mm field. There is one collimator helmet per collimator size, and while all four helmets are identical and can in theory take any size collimator, in practice each helmet is used with uniform-size collimators. Individual collimators can be replaced with tungsten-alloy ‘‘plugs’’ in order to affect desirable changes in the resulting dose distribution. In the Perfexion, the central body and collimation system are somewhat more complex because the 60Co sources in the Perfexion are not stationary and the collimators are entirely internal to the radiation body; there are no external helmets. In the Perfexion, the 192 60Co sources are grouped into eight independent source sectors (> Figure 58-6b). Each source sector is housed in an aluminum frame which is attached to sector drive motors at the rear of the radiation body via linear graphite bushings (> Figure 58-7). There are 576 collimators machined into 12 cm-thick tungsten in five concentric rings to align with the
901
902
58
Gamma knife: technical aspects
. Figure 58-6 (a) The 4 mm collimator helmet for the model C (b) The internal, sector-based collimator of the Perfexion (image courtesy of Elekta AB, Stockholm)
source assembly configurations. Each source has three available collimators (4, 8, and 16 mm) as well as two shielded positions; ‘‘blocked’’ and ‘‘home.’’ The blocked position is used when repositioning a patient between isocenters or when shielding a critical structure. The home position is a shielded position used when the unit is inactive. To achieve a particular collimation, the sector drive motors move the sources along their bushings to the correct position over the appropriate
collimator opening. The position of each sector is monitored by linear and rotational encoders which have a positioning repeatability of <0.01 mm.
Stereotactic Frame Attachment In order to restrain the patient from any movement during a treatment and to create a correspondence between the stereotactic coordinate system defined by the patient frame and the internal coordinate
Gamma knife: technical aspects
. Figure 58-7 The rear of the Perfexion unit with the covers removed, exposing the sector drives which move the internal collimation system source sectors
58
respect to the treatment unit (known as the gamma angle). The model C Gamma Knife introduced the Automatic Positioning System (APS). The APS is an electromechanical system that mounted to the proximal end of the treatment table, the collimator helmet, and the lateral sides of the stereotactic frame. During treatment, the APS automatically moves the patient’s head from location to location, and provides much greater precision in positioning (0.3 mm). The APS uses four standard gamma angles of (72 , 90 , 110 , and 125 ). One drawback to the APS is that it has a reduced available treatment volume as compared to the trunnion system, sometimes requiring the neurosurgeon to switch to the trunnion system during treatments of lateral or posterior lesions.
Patient Positioning System and Frame Adapter
system of the unit, the patient’s stereotactic frame must be fixed to the treatment unit; a process often termed ‘‘docking.’’ Differences in the way the stereotactic frame docks to the Gamma Knife unit track the evolution of the Gamma Knife through its various models.
Trunnion and APS Systems The earliest method for attaching the stereotactic frame was through a mechanical trunnion system which attached to the collimator helmets. The system was machined with measurement indicators and the neurosurgeon would manually set the isocenter coordinates and the angle of the patient’s head in the sagittal plane with
In the Gamma Knife Perfexion, the treatment table itself acts as a positioning device (the Patient Position System, or PPS). The stereotactic frame is attached to the treatment table via a removable frame adapter (> Figure 58-8) which attaches to the frame and acts as an interface with the treatment table. The PPS has several potential advantages over the older APS system. Because the entire treatment table moves from position to position, the relative positions of the patient’s head and neck do not change during the treatment, increasing patient comfort. The PPS is also a simpler design than the APS, which required two independent sliders to agree on position. Linear and rotational encoders ensure that the PPS is always at a known, correct position with a repeatability of <0.05 mm [9]. The PPS and frame adapter permit gamma angles of 70 , 90 , and 110 . Gamma angle adjustments are the only manual adjustment required during a treatment on the Perfexion.
903
904
58
Gamma knife: technical aspects
. Figure 58-8 Perfexion Patient Positioning System (PPS) with frame adapter mounted to the table. In clinical use, the adapter would connect the patient frame to the PPS
. Figure 58-9 Picture of the GK console
Console The Gamma Knife console (> Figure 58-9) contains functionality to allow an operator to import treatment plans, verify all critical treatment
information, commence treatment, monitor treatment status, pause or end treatments, and in an emergency stop treatment and safely remove a patient. The console consists of a control panel, a computer monitor to display and monitor
Gamma knife: technical aspects
58
treatment status, a video display for patient surveillance, a microphone for patient communication, and a keyboard for entering pertinent information.
distributions, evaluate treatment plans, verify treatment parameters, and export completed plans to the treatment unit. > Table 58-2 summarizes this functionality.
Planning System
Basic Treatment Process
The treatment planning system for the Gamma Knife, known as Leksell GammaPlan1 (Elekta Instruments AB, Stockholm), is housed in a separate computer, or computers from the clinical unit. Historically, treatment planning was accomplished on a Unix workstation with a direct serial connection to the Gamma Knife console computer for exporting treatment plans. With the Perfexion the treatment planning system runs on a Linux/PC platform and a TCP/IP-based communication system (older models are being retrofitted with the new planning system). The new version allows networking of treatment planning systems, communication with multiple gamma knife units, and simplified communication with external data sources such as hospital PACs and neuroimaging units. GammaPlan includes the functionality required to import patient images, create dose
The basic process of the Gamma Knife procedure begins with placement of the Leksell stereotactic frame, proceeds to imaging and treatment planning, pre-treatment checks, and concludes with the treatment itself and the removal of the stereotactic frame (> Figure 58-10).
Frame Placement The technique of Gamma Knife surgery begins with the placement of the Leksell G-Frame on the patient’s head by the neurosurgeon. The details of the procedure vary by center. The majority of institutions place the frame in procedure rooms using local anesthesia. In our center we believe the frame is best placed in an OR setting using controlled sedation, local anesthesia, and giving strict attention to aseptic conditions.
. Table 58-2 Overview of the functionality found in Elekta GammaPlan for used with the Gamma Knife Feature
Purpose
DICOM image import Image registration
Imports image data from imaging consoles or from institutional PACs systems Registers images into Leksell stereotactic space based on fiducial system visible in images Registers non-stereotactic image studies into the coordinate system of existing stereotactic images Creates a blended display of registered image studies from the same or different modalities (e.g., MR and CT to show both soft tissue and bone) To define targets, critical structures, and other structures of interest To define the skull morphology for use in dosimetry and collision calculations To define one or more dose distributions To create metrics by which to evaluate a treatment plan. Tools include dose volume histograms, mean/min/max doses, and point dose sampling Exports treatment plans to the clinical unit Prints treatment plans to be used as a written directive. Plans include all relevant treatment data which allows verification before and during treatment
Image co-registration Image fusion Contouring tools Skull data Treatment planning tools Treatment evaluation tools Treatment export Printing functionality
905
906
58
Gamma knife: technical aspects
. Figure 58-10 Flowchart of basic treatment process
The Leksell G-Frame [10] (> Figure 58-11) consists of a rectangular aluminum base ring to which four aluminum posts are attached. The frame is attached to the patient through the posts using titanium pins which are screwed through the skin to the outer table of the skull. A variety of post shapes and lengths are available
to allow optimal fitting of the frame. In addition, frames are available which use replaceable plastic inserts which electrically isolate the pins from the frame in order to prevent pin-site heating in high-strength MRI units. The quality of the frame placement is absolutely critical to the subsequent steps in of the gamma surgery. Proper frame placement is an acquired skill and necessitates a comprehensive understanding of neuroanatomy as viewed from outside of the head. Because the Leksell frame determines the relative location of stereotactic space, and because the Gamma Knife has a finite treatment volume, it is critical that the surgical target be placed as close as possible to the center of the coordinate space. For patients with lateral lesions this may mean shifting the frame off of the midline of the patient. For patients with multiple lesions scattered across both hemispheres of the brain, this may mean choosing to treat all lesions on one-side of the brain in one session and returning for a second session to treat the other lesions1. In other cases, such as lesions of the parasellar region, it may be advantageous to align the frame with the base ring parallel and below the level of the anterior optic pathways. The resulting orientation of the brain may help shift dose away from these critical and radiosensitive structures, although there is often a tradeoff with increased dose to the lenses of the eye. Incorrect frame placement can result in great difficulty in achieving optimal dose distributions around critical structures, and in the case of lateral or posterior lesions can even make treatment impossible without repositioning the frame and re-imaging the patient.
1
Note that this is not as critical an issue for the Perfexion model Gamma Knife as it has a significantly larger available treatment volume. For the Perfexion, the optimal frame placement is generally neutral, with the frame centered on the midline of the head.
Gamma knife: technical aspects
58
. Figure 58-11 Leksell G-Frame as it would appear affixed to a patient’s head and attached to the unit via the frame adapter
Imaging The accuracy of a Gamma Knife treatment is ultimately dependent on the neurosurgeon’s ability to visualize the intended target. Thus, the technique would be impossible without the advent of technology that allows three-dimensional views of anatomical structures in the brain, including tomographic modalities such as MR and CT, as well as bi-plane angiography. Each modality has advantages and disadvantages in particular situations as discussed below. Regardless of modality, all imaging proceeds using a stereotactic fiducial system. This consists of a box which attaches securely to the Leksell frame during imaging. Fiducial markers built into the box are made to be visible in the resulting image set and allow the Gamma Knife treatment planning system to calculate positions in the images relative to Leksell space. Patient immobilization during imaging is achieved by taking advantage of the rigid Leksell frame. Adapters designed for each modality attach to both the frame and the patient table and
are sufficient to prevent unwanted movement during imaging.
MR In the majority of centers, MR is the most used treatment modality because of its superior visualization of soft tissue structures and solid tumors. Typical MR protocols include T1-weighted preand post-contrast (Gadolinium-enhanced) images through the entire volume of the head. Sequences may be a collection of 2D image slices, or a true 3D acquisition such as the MP-RAGE [11] or its successors. Specialized sequences such as constructive interference in steady state (CISS) [12–14] protocols may be used for circumstances such as visualization of the internal auditory canals, the cerebellopontine angle, and parasellar regions. Current research is also exploring the use of more exotic MR protocols, such as MR fiber tractography using diffusion-tensor imaging, to better visualize critical brains structures not easily identified on standard pulse sequences [15].
907
908
58
Gamma knife: technical aspects
MR images should be specified to have a field of view large enough for the fiducial markers to be clearly visible. Sequences should have a thin (1–3 mm) slice thickness with no gap between slices to ensure complete coverage of the head, increase the probability of detection of small lesions [16,17], and maximize the accuracy of volumetric measurements [18] and dose volume histograms. The MRI fiducial system consists of a plastic box which attaches directly to the stereotactic frame (> Figure 58-12a and b). Machined into
the box are channels with a distinctive ‘‘N’’ shape. The channels can be filled with a cupric sulfate solution which will enhance in commonly used pulse sequences. When resampled as axial and coronal sections, the fiducial marks appear as a series of dots in the image. By calculating the relative distances between the dots the treatment planning system can determine the 3D location of any point with respect to stereotactic space. It should be noted that MRI is susceptible to a variety of linear and nonlinear geometric distortion, primarily from gradient field nonlinearity
. Figure 58-12 (a) The MR fiducial box used with the Gamma Knife. (b) The resulting fiducial indicators on an MRI
Gamma knife: technical aspects
and magnetic field inhomogeneities, including from the patient and frame [19]. Thus, all MR protocols used for radiosurgery should be validated by a qualified physicist to ensure minimal distortion. In cases where a target is close to an airtissue interface or some other area of quickly varying magnetic susceptibility one may consider using CT to help detect any distortion. There exist a variety of commercially available imaging phantoms which can help physicists detect and quantify the amount of distortion for a pulse sequence under a particular set of conditions, and can help minimize any risk from MR distortion.
CT Because of its relatively poor soft tissue visualization, CT imaging is a less-commonly used imaging modality. However in cases where MR is contraindicated or in cases where bony anatomy may provide useful information CT imaging is still quite useful. Gamma Knife CT protocols include pre- and post-contrast imaging with thin (1–3 mm) slice thickness and no gap between slices. Techniques such as CT cisternography can be useful in situations such as trigeminal neuralgia where structures in the CSF must be detected but when MR is not an available option [20]. CT
58
does not suffer from distortion in the same way as MR however the pins attaching the stereotactic frame to the patient cause artifacts which can obscure lesions proximal to the pin sites.
XA For vascular lesions such as arteriovenous malformations, digital subtraction angiography remains the imaging modality of choice. Gadoliniumenhanced MRI is sometimes used to correlate the extent of the AVM nidus with angiography, and there is current research investigating the use of MRA for nidus definition [21]. As with the MR and CT, images are acquired using a fiducial system (> Figure 58-13a and b), however the DSA system is based on projections rather than tomographic information. Images from some DSA systems must be geometricallycorrected to account for the curvature of the imageintensifierscreenbeforeimportingtheimages into the treatment planning system [22].
Skull Measurements The Gamma Knife treatment planning system requires the depth of the target point in each
. Figure 58-13 (a) The fiducial box used for digital subtraction angiography. (b) The resulting fiducial markers on an AP projection. Similar marks are found on the lateral projections
909
910
58
Gamma knife: technical aspects
. Figure 58-14 ‘‘Bubble’’ device used to measure the dimensions of the patient’s skull. Skull measurements are used to calculate beam attenuation through the head and detect possible collisions when positioning the patient
beam direction in order to calculate photon attenuation and avoid collisions between the patient and the unit. GammaPlan accomplishes this using a skull scaling instrument (> Figure 58-14) to collect sampled skull measurement data, which is then interpolated and used to construct the required skull and frame model. A member of the treatment team uses a simple ruler to measure the distance from the scaling ‘‘bubble’’ to the patient’s head and records this information in the treatment planning system.
tolerable for the patient. As treatment planning is such an integral part of the Gamma Knife technique, it is explained in more detail in later sections.
Treatment Following the creation, approval, and export of a valid treatment plan to the Gamma Knife unit, the irradiation portion of the Gamma surgery may commence. > Figure 58-15 illustrates the steps in the procedure.
Planning Treatment planning is the process of creating a dose distribution that conformally treats the intended target. Treatment planning is an iterative technique requiring detailed knowledge of neuroanatomy, neuroradiology, the biological effect of single-fraction radiosurgery, and the compromises required to create a treatment which will be effective and at the same time
Verification of Treatment Parameters The Gamma Knife console receives a treatment planning file from the treatment planning system after the plan is approved and exported. Before treatment commences, the operator should have a printout (written directive) of the treatment which lists the vital treatment parameters such
Gamma knife: technical aspects
58
. Figure 58-15 A flowchart of a Gamma Knife surgery
as patient identification, treatment plan identification, isocenter locations, collimator sizes, isocenter dwell times, etc. The operator should verify that the information on the written directive matches the information displayed on the console and in the treatment planning system before treatment commences.
previously described. If using the trunnion system, it is vital to use the utmost care in assuring that the correct coordinates have been set and all screws have been tightened. In the APS and PPS systems, the correct angle should be set (sensors on the unit will trigger a warning if this is not the case).
Docking the Patient
Clearance Checks
This step involves attaching the patient’s stereotactic frame to the treatment unit using one of the methods (trunnion, APS, frame adapter)
Moving a patient so an isocenter with a location far from the center of the head is at the focus point can cause the opposite side of the head or
911
912
58
Gamma knife: technical aspects
. Figure 58-16 The clearance check tool used with the Perfexion. The arm of the tool rotates around a volume equal to the treatment cavity. When the patient is positioned relative to the clearance tool, a full rotation of the arm of the tool will detect and frame or skull collisions
stereotactic frame to collide with the treatment unit. In most cases the treatment planning system warns the user of possible collisions based on the acquired skull measurement data, and when using the APS the system tests all questionable locations to ensure they will avoid collisions. In cases with many collision checks, this can add significantly to the total treatment time for a patient. In the Perfexion, clearance checks are an infrequent problem; however they still can and do occur. The Perfexion has a specially designed clearance check tool (> Figure 58-16) which simulates the achievable treatment volume and simplifies the process of checking for collisions. The Perfexion also makes use of a ‘‘frame cap’’ (> Figure 58-17) which conservatively models a human scalp and is of dimensions known by the treatment planning system. The frame cap is placed over the stereotactic frame and allows the system to eliminate the possibility of collisions in many instances, reducing the required number of skull and frame measurements. Finally, the Perfexion has a collision sensor in the form of
. Figure 58-17 The Perfexion’s frame cap fits over the patient’s head and frame. It is of a known geometry, and assists the system in detecting collisions
an aluminum cap covering the treatment cavity. Pressure exerted on this cap will trigger the system to retract the sources to their shielded home position to minimize radiation exposure and permit manual removal of the patient from the unit.
Gamma knife: technical aspects
Treatment Execution After the patient is comfortably situated and all checks are complete, the operator may commence treatment. The console displays data to allow the operator to keep track of treatment progress, and video and two-way audio surveillance of the patient allows the operator to react to any problems (medical or technical) that may arise. Following treatment, the surgeon removes the stereotactic frame from the patient’s head (often in the Gamma Knife suite), and the patient is either discharged or delivered to a room for overnight observation.
Treatment Planning GammaPlan (Elekta AB, Stockholm), the Gamma Knife’s treatment planning system, is a proprietary system specific to the Gamma Knife, making it somewhat different from linac-based delivery systems which often make use of generalized planning systems such as ADAC/Pinnacle [3] (Phillips Healthcare, Andover, MA). However, the decision to develop a specialized planning system is driven in part by the unique physics and specialized indications of Gamma Knife surgery, and likely simplifies the overall treatment planning process.
General Philosophy GammaPlan achieves the dual goals of allowing plans to be created in a reasonably short period of time and maintaining the look and feel of ‘‘surgery.’’ Gamma Knife surgery proceeds immediately following frame placement as soon as an approved plan can be generated. This is unlike traditional radiation therapy where days or even weeks can elapse between simulation and treatment. Because the patient is literally waiting for treatment with a frame on his head, short planning
58
times are important. Designing the planning system to have a ‘‘surgical’’ feel allows the neurosurgeon to approach the target from the perspective where he is most comfortable; that of a surgery, albeit a surgery on computer.
Basic Planning Process GammaPlan is expressly a forward-planning system. This means that the surgeon is responsible for defining and optimizing the plan. This is in contrast to inverse planning systems (commonly used with IMRT systems), which begins with a series of user-defined dose-volume constraints and the computer uses one of a variety of mathematical optimization algorithms in attempt to find a solution that best matches those constraints.2 Gamma Knife treatment planning proceeds in a series of steps as illustrated in > Figure 58-18.
Image Registration Image registration is the procedure in which the treatment planning system matches the fiducials visible in each image slice to an internal model of the fiducial system and stereotactic space. The system will report calculated deviations from the internal model so the user can be informed as to whether there are any gross distortions or other problems in the images.
Defining Targets and Critical Structures Following image registration, the next step in treatment planning is to define targets and critical 2
A reader interested in inverse planning for the Gamma knife may find references [23–25] useful. These methods have met little commercial acceptance in the Gamma Knife community.
913
914
58
Gamma knife: technical aspects
. Figure 58-18 Procedures for treatment planning
structures. Using tools built into the treatment planning system, targets are outlined on a sliceby-slice basis. Critical structures are defined in a similar manner and may be marked as avoidance structures. Gamma Knife radiosurgery has traditionally followed a surgical approach to defining targets with the target outline matching as closely as possible to the visible target on the imagery. This is a departure from the approach in radiotherapy, where a gross target volume (GTV) is often expanded in steps to create a clinical target
volume (CTV) and planning target volume (PTV) to account for subclinical disease and setup error during the treatment [26].
Dose Matrices/targets The Gamma Knife treatment planning system makes use of one or more dose matrices (termed ‘‘targets’’ in recent releases of GammaPlan). A dose matrix is a calculation matrix with a fixed number (31 31 31) of sampling points. The
Gamma knife: technical aspects
58
. Figure 58-19 A typical dose distribution (yellow line) is a superposition of a number of individual isocenters, or shots (red circles)
dose matrix defines an a volume over which GammaPlan will calculate dose and display isodose curves and also provides a method for prescribing a dose for a group of isocenters. Because the dose matrix has a fixed number of sampling points, the size of the matrix should be kept to a reasonable minimum that allows good visualization of the dose distribution and minimizes sampling error.
Defining a Treatment Field The goal of Gamma Knife treatment planning is to devise a plan that achieves complete coverage of the target at the desired dose while minimizing
dose to normal tissue. The neurosurgeon creates the dose distribution by defining one or more isocenters (commonly known as ‘‘shots’’) at locations within the volume of the target so that the prescription isosurface conformally matches the target (> Figure 58-19). Each isocenter defines a location to which the Gamma Knife must move the patient so as to place the isocenter in the radiation focus of the unit. Multiple isocenters allow the surgeon to create dose distributions with irregular shapes. In the model B and C gamma knives, the surgeon has a choice of four sizes of isocenters corresponding to
3
Note that some institutions use a superposition method to achieve isocenters with intermediate sizes [27].
915
916
58
Gamma knife: technical aspects
the four collimator helmet sizes3. In the Perfexion model Gamma Knife the surgeon can vary the collimator size of the isocenter by sector. The surgeon may also vary parameters such as gamma angle, the weight of the isocenter relative to all other isocenters, the prescription isodose line, and the prescription dose. Shielding
It is frequently the case that the lesions targeted for Gamma Knife surgery are present close to critical structures such as the optic pathways, cranial nerves, etc. In these cases, the neurosurgeon may choose to protect these structures through the application of beam-blocking patterns that minimize the contribution of beams which intersect those structures. The practical effect of shielding is to compress the dose gradient in a particular direction, and shift the dose to a different direction. In the model B and C models it is possible to block individual beams by replacing collimators in the helmets with solid ‘‘plugs.’’ The ‘‘plugging patterns’’ as they are commonly called, are generated using the planning software using a beam-intersection method and refined by the neurosurgeon (> Figure 58-20a) [28]. With the Perfexion, manual plugging of individual beams has been replaced by the ability to automatically shield any combination of collimator sectors, so all beams in a shielded sector are blocked. Any contoured structure may be defined as an avoidance structure. The system will then compute for any combination of isocenters the beams which intersect the avoidance structure(s). If the number of intersecting beams in any sector exceeds a given threshold, the sector is blocked (> Figure 58-20b). Because the Perfexion’s collimation system is fully automatic, the burden on the treatment team when shielding has been eliminated. However, overuse of shielding can result in extended treatment times due to the reduced dose rate as the number of unblocked sources decreases.
Plan Evaluation Gamma Knife treatment plans are evaluated based on coverage of the target, conformity of the treatment field to the target, dose to normal and critical tissue, and the time and effort involved in executing the treatment. Treatment coverage and conformity are evaluated visually, using dose volume histograms, and often using one of a number of proposed indicies of plan conformity [29–32]. The conformity index attempts to quantitatively measure the degree to which the target is covered at the level of the prescribed dose and the degree to which normal tissue is included in the treatment field. In a perfectly conformal plan, the target would be completely covered by the prescribed radiation dose and there would be no spillover of dose outside of the target. Conformity evaluates a treatment plan at a particular dose. It does not evaluate the dose falloff and the irradiation of normal tissue at low doses. To address this, Paddick et al have devised the Gradient Index as a metric to gauge the dose falloff of a treatment plan below the prescription isodose [33]. Wagner, et al. attempt to combine the conformity index and gradient index ideas into a single metric they term the Confomity/Gradient (CGI) Index [34]. Several authors have also devised dose-volume constraints similar to those used in traditional radiation therapy such as the 12 Gy volume as a predictor for radiation necrosis following gamma surgery [35].
Treatment Planning – Tricks of the Trade Gamma Knife treatment planning is as much an art as a science and one of the drawbacks of forward-planning is that the end result depends greatly on the skill and experience of the individual performing the dose planning. In this section
Gamma knife: technical aspects
58
. Figure 58-20 (a) Plugging. Individual isocenters may replace collimators with ‘‘plugs’’ in order to protect critical structures, in this case the anterior optic pathways in a treatment plan for a pituitary lesion. (b) Plugging with the Perfexion. The optic nerves are defined as avoidance structures. The hatched sector in the sector configuration diagram represents a sector which is to be shielded. The resulting isodose distribution can spare the optic pathways a high dose
917
918
58
Gamma knife: technical aspects
we present some ‘‘tricks of the trade’’ the reader may find useful: 1.
2.
3.
4.
5.
6.
Targets should be filled with larger shots near the center and smaller shots on the periphery in order to maximize dose falloff outside of the target. Some reports suggest that placing shots along the central axis of a target is a useful first approximation [24]. Shots which are located significantly outside the boundary of the target should be avoided. While this technique can in some cases lead to a highly conformal plan at the prescription isodose line, it generally leads to poor dose falloff and a larger volume of normal tissue contained within lower isodose regions. When placing a shot, it is difficult to visualize where the shot fits in a plan without information for all three orthogonal directions. Therefore, it is often useful to plan using a workspace view that includes axial, coronal, and sagittal cuts through the brain. While Gamma Knife treatments are traditionally prescribed at the 50% isodose line, this does not always lead to optimal conformity or dose gradient [33]. Careful consideration should be used when choosing the appropriate isodose line in a given situation. Shots which are clustered together can cause hot spots in the plan and make the isodose distribution contract. This can be used as an advantage in treatment planning. Monitoring the hot spots (for instance, by displaying the 95% isodose line) can make it possible to predict when the prescription isodose line will expand or contract. Placing a small shot directly on a hot spot can provide a ‘‘control’’ shot that can be used intentionally to assist in planning by modulating the weight of the ‘‘control.’’ When planning parasellar tumors, changing the gamma angle to a lower angle may help align the major axis of the treatment plan
7.
8.
9.
to be parallel to the anterior optic pathways. This can help lower the dose to these critical structures and minimize the need for shielding. Because dose matrices in GammaPlan have a fixed number of sampling points, the dose matrix size should be kept to a reasonable minimum. Small-sized shots with large dose matrix sizes can lead to significant calculation error due to inadequate sampling frequency. Combinations of very large and very small shots (such as 18 and 4 mm) should be avoided. The difference in scale between these sizes is such that the smaller shot will have little significant effect on the resulting dose distribution. The B and C models ship with 100 physical plugs, but GammaPlan permits up to 166 plugs in a treatment plan. Taking advantage of this makes it easier to generate complicated plugging patterns. However, this technique should be used with caution so a plan is not approved with more than the physically-present number of plugs.
Gamma Knife Dosimetry From a dosimetric perspective, the basic physics the Gamma Knife treatment planning system needs to account for are the inverse square law, photon attenuation through tissue, reduction in dose rate due to collimation, and dose falloff away from the central axis of the beam. The algorithm used in GammaPlan makes the assumption that the brain is a homogenous, water-equivalent volume of tissue and that there is no buildup region near the surface. This makes the algorithm relatively simple from a physics standpoint, however from a geometric perspective it is quite complex as the dose
Gamma knife: technical aspects
contribution from each of the 60Co sources must be computed. In the earlier Gamma Knife models (models B, C, and 4C), much of the beam data used in the dose calculation model was acquired empirically. While the general algorithm remains the same for all gamma knife models, the design of the Perfexion makes the details of the algorithm somewhat more complex, and in this new machine both empirical and monte-carlo methods were used to create the dose calculation model. Some of these differences created by the new Perfexion will be discussed at the end of this section.
Single-beam Model In the B, C, and 4C models of the Gamma Knife, the sources are within an approximation equal in activity, have identical collimation (for a given collimator helmet), and have identical 400 mm source to focus distances. Therefore for dosimetric purposes, each of the 201 beams may be considered to be identical.
58
To characterize any individual beam, a single source and beam channel was constructed to the specification used with the actual Gamma Knife unit (> Figure 58-21). From this single beam channel, the radial dose distribution and percent depth doses for a single beam were measured for each collimator size. From the percent depth doses an attenuation coefficient can be derived for the brain as well as output factors for each collimator size normalized to the 18 mm collimator.
Superposition Algorithm From the characterization of an individual beam and the assumption that all 201 beams are identical, a superposition of beam data leads to the final dose calculation model. The only difference between beams is the path each beam takes through the head. Thus, the problem becomes one of finding the depth each beam must traverse to reach the focal point so that the contribution of each beam may be appropriately attenuated.
. Figure 58-21 The single-beam measurement system used to characterize an individual beam channel (image courtesy of Elekta AB, Stockholm)
919
920
58
Gamma knife: technical aspects
This problem is solved by interpolating the entrance point of the beam with the earlier described skull measurements via an iterative search algorithm. Once the depth is known, the beam may be adjusted by a constant exponential tissue attenuation factor and the dose rate contribution of the beam may be computed. The sum over all such beams leads to the dose rate at the focus point. To create a three-dimensional dose distribution at locations away from the isocenter, the radial dose distribution of each beam is used. For any point, the radial distribution and the attenuation coefficient allows computation of the difference in dose from the central axis of each beam (> Figure 58-22).
Dose Normalization Dose distributions for the Gamma Knife are traditionally normalized to the maximum dose point in the treatment plan. In treatment plans with multiple targets, isodose curves can be displayed relative to the global maximum dose point or to the local maximum dose point for
. Figure 58-22 The geometry used to determine the dose contribution of a single beam. To calculate the dose contribution at any point P from Beami, the depth and off-axis distance from the central beam must be computed. Note: df <0 if P is upstream from focus and df >0 if P is downstream
each target. Note that in situations where the surgeon intends to prescribe different doses to each target or to prescribe to different isodose lines for each target it is important to be aware of whether one is looking at a local set of isodose curves or curves normalized to the global plan maximum.
Differences with the Gamma Knife Perfexion The Perfexion departs from the idea that every beam may be treated identically. Because the Perfexion’s collimator body is conical, not hemispheric, each row of sources has a different source to focus distance ranging from 374 to 433 mm. In addition, in some rows the sources are placed at an angle to the beam channels, resulting in asymmetric treatment fields (> Figure 58-23a and b). The design changes mean that parameters such as output factors, attenuation factors, and source to focus distances are no longer constant, but instead vary by both collimator size and beam channel location. In addition, because the sources are angled to the beam channels the off-axis distributions must be stored as a two-dimensional table rather than a one dimensional table. The dose calculation model for the Perfexion is therefore more complex than its predecessors; however the design allows more complex isocenter distributions to be created. One example is an isocenter which is approximately ‘‘square’’ in the axial plane (> Figure 58-24).
Commissioning and Quality Assurance A robust quality assurance program is vital to ensure that the Gamma Knife is functioning within all radiological and mechanical tolerances and that all electromechanical and computerized
Gamma knife: technical aspects
58
. Figure 58-23 (a) A cutaway of the beam channels in the Perfexion demonstrating how the sources are angled relative to the beam channels. (b) The resulting asymmetric beam at the focal point makes dose computation for the Perfexion more complex than in previous models (images courtesy of Elekta AB, Stockholm)
systems are functioning properly. > Table 58-3 summarizes some of the primary aspects of the system that are a part of any well-designed quality assurance program for the Gamma Knife. The remainder of this section will highlight some of the tools that ship with the Gamma Knife to assist in the quality assurance process.
Dose Rate (output) Tests Output tests for the Gamma Knife are conducted with a chamber/electrometer setup within an 80-mm spherical polystyrene phantom (> Figure 58-25) that ships with the Gamma Knife unit. The unit is calibrated to the largest
921
922
58
Gamma knife: technical aspects
. Figure 58-24 A ‘‘square’’ dose distribution created using a composite shot with the Gamma Knife Perfexion
available collimator size (18 mm for the B, C, and 4C; 16 mm for the Perfexion). Field-sized dependent output factors are a standard quality assurance item in linearaccelerator systems; however for the Gamma Knife this is difficult due to the focal spot size. In the earlier models it is possible to verify the manufacturer-recommended output factors for the larger collimator sizes [36–38]. However, the small focal spot of the 4 mm collimator has been difficult to verify in a clinical setting. This becomes even more difficult with the Gamma Knife Perfexion as the output factors for individual beams are dependent on collimator position in the central body. However, unlike with the collimator jaws of a linac, the collimator openings of the Gamma Knife do not change size, so periodic output factor verification may be less of an issue.
Radiological Focus Point and Patient Positioning System Calibration In the Perfexion, the movements of the Patient Positioning System (PPS) and the position of the radiological focus point (RFP) are calibrated separately at installation. The movements of the PPS are verified for linearity using a laser interferometer. The mechanical alignment of the collimator beam channels are determined mechanically for the outer 4 mm collimators, and the relative locations of the remaining beam channels are known to a precise mechanical tolerance. Once the PPS and radiation body are mated at installation, it remains to calibrate the PPS versus the RFP. During installation and during interval quality assurance this calibration is accomplished
58
Gamma knife: technical aspects
. Table 58-3 Components of a Gamma Knife QA program (note that individual centers may include more than appears on this list) System functionality Dose rate (output) Radial isocenter profile Focus point coincidence Patient positioning accuracy Timers Safety interlocks Treatment planning system Backup power supply
Model B, C, and 4C
Perfexion
Ionization chamber/ electrometer and water-equivalent phantom Film
Ionization chamber/ electrometer and waterequivalent phantom Film
Film
Film + diode tool
Trunnion centricity tool; APS QA tool Chamber/electrometer and independent timer Manual testing Independent dose calculation for a given test setup Power supply self test
Recommended tolerance
Frequency
3% of predicted dose
Monthly
1 mm of predicted profile at 50% isodose 0.4 mm
Yearly
Diode tool + clearance check tool
0.4 mm
Monthly
Chamber/electrometer and independent timer Manual testing Independent dose calculation for a given test setup Power supply self test
0.1 min or 10%
Monthly
No failures 1%
Daily Weekly
No failures
Monthly
Yearly
. Figure 58-25 Electrometer/ionization chamber and Elekta 16 cm diameter polystyrene phantom used for output tests
through the use of a specially-designed diode tool as well as radiochromic film such as GafChromic1 EBTor MD-55 (International Specialty Products, USA) films [39].
The diode tool consists of a diode mounted on a rigid metal frame engineered to dock to the PPS. The diode tool scans along a predefined volumetric space and searches for the maximum
923
924
58
Gamma knife: technical aspects
. Figure 58-26 (a) The precision diode tool used to test radiological accuracy. (b) The film holder used for verifying beam profiles and positional accuracy
dose point as well as the penumbra regions of the profile for the 4 mm collimator. It compares these measurements against the stored installation/calibration information and reports an error if it is out of range with an accuracy of 0.4 mm and a repeatability of <0.1 mm [9] (> Figure 58-26a). The film tool is a special adapter which can hold a piece of radiochromic film. A pinhole is punched at the expected location of the
RFP. After irradiation, the exposed relative dose distribution can be analyzed and an RMS error determined (> Figure 58-26b).
Isocenter Profiles Isocenter profiles for the Gamma Knife are measured using the same film holder as in the previous tests; however with the film no pinprick
Gamma knife: technical aspects
is required. The film dose distributions can be measured using optical density techniques and compared to the distributions predicted by the treatment planning system.
Future QA Radiosurgery in general may benefit for the ability to capture a true three-dimensional dose distribution for use in quality assurance. One promising area of research uses polymer-gel dosimetry in which a phantom filled with gel is irradiated with a specified dose distribution [40–44]. The gel polymerizes in proportion to the absorbed dose, and can be scanned with MRI or optical CT to obtain a 3D dose distribution. Questions regarding the accuracy and resolution of this technique have so far prevented its widespread adoption.
Summary of Features Specific to the Gamma Knife Perfexion Treatment Volume The elimination of the external collimation helmets required for the previous models opens up a much larger potential treatment volume in the Perfexion Gamma Knife. > Table 58-4 shows the differences in reachable range on the Perfexion
. Table 58-4 The treatment ranges of the trunnion system, model C APS, and Perfexion PPS Dimensions
Perfexion
Trunnion
APS
X
160 mm
100 mm
Y
180 mm
150 mm
Z
260 mm (distance from focus point to inner collimator surface)
125 mm
82 mm 120 mm 153 mm
58
versus the Trunnion and APS systems of the models B and C. This means that many cases which could not be optimally reached in one treatment or that required changes of gamma angle during the treatment in order to avoid collisions will now in most cases be easily treated in the new unit. > Figure 58-27 illustrates an example of a patient with multiple meningiomas widely distributed in the brain. In previous Gamma Knife units multiple treatment sessions would be required to reach all of the targets. With the Perfexion unit, these targets can be reached in a single session with no collision checks and no gamma angle changes, greatly reducing the treatment burden for the patient. In the infrequent situation where a potential collision is detected with the Perfexion, the previously described clearance tool that ships with the Gamma Knife reduces the effort required to control whether or not a collision will occur.
. Figure 58-27 Patient with multiple meningiomas distributed in both cerebral hemispheres. All lesions were successfully treated in a single Gamma Knife session on the Perfexion with no collision checks or angle changes
925
926
58
Gamma knife: technical aspects
Composite Shots and Automated Shielding With the sector-based collimation system, any individual sector can be configured as any of the three available collimator sizes or to a blocked position. This means it is now possible to create composite isocenters which simultaneously make use of different collimator sizes. The promise of composite shots is that each isocenter can be more carefully tailored to match the shape of the target. The potential drawback is that the automation afforded by the Perfexion will make it tempting to increase the use of smaller collimators and shielding even in cases where larger collimators would suffice. The consequential increase in beam time will partly offset the gain in efficiency.
Automated Shielding With the Perfexion unit, the manual plugging of the older units becomes a fully automated process. In addition, the Perfexion make it possible to shield more (168 of 192) of the sources than was possible in the previous Gamma Knife models (100 of 201). This lowers the burden to the use of shielding and makes possible isodose distributions that were not conventionally possible with previous models. However, the granularity to which beam channels can be blocked has been reduced, as it is now possible to block only at the granularity of a sector. The ability to create annular plug patterns has been eliminated, which may limit the ability to shield critical structures superior to the target. Finally, the increased use of shielding may lengthen beam-on treatment time as there are fewer beams available to deliver the prescribed dose.
Reduced Shuttle Time The Perfexion significantly reduces the positioning time required between shots. In the Gamma
Knife models with no APS system, every isocenter requires stopping the treatment, manually setting coordinates, and restarting the treatment; a timeconsuming process. The APS automatically moves the patient from location to location; however in this system the patient is withdrawn from the focus of the Gamma Knife (the defocus position) while the coordinates are changing. The defocus/ reposition/refocus procedure takes 30 s to complete. Thus, with the APS system, there is significant time overhead for every isocenter added. The Perfexion’s PPS moves the patient at speeds up to 10 times (7 mm/s vs. 0.7 mm/s) faster than the APS. No defocusing of the patient is required as the sources move to a blocked position during repositioning. The result is that a much larger proportion of the Gamma Knife treatment is beam-on time with the Perfexion, reducing the overall treatment time for the patient.
Improved Patient Comfort The Gamma Knife Perfexion purports to significantly increase patient comfort. We have previously described the redesigned docking mechanism for the patient’s head and stereotactic frame and its implications for patient comfort. To further patient comfort, the Perfexion ships with a much thicker mattress and a treatment table that better supports the upper back and shoulders.
Lower Extracranial Dose The internal shielding and the beam channel directions of the Perfexion greatly reduce the extracranial dose patients will receive. Lindquist, et al. report doses on anthropomorphic phantoms which are up to 10 times lower than those reported for the B and C models [45]. The manufacturer claims an average dose of 4 mSv/h at a distance of 60 cm from the side surfaces of the unit. This is important for the safety of the device and the shielding required when designing treatment suites.
Gamma knife: technical aspects
58
Future Possibilities
References
The elimination of the external collimation helmets opens up a much larger potential treatment volume in the Perfexion model. This means that many cases which could not be optimally reached in one treatment or that required awkward positional changes during the treatment will now in most cases be easily treated in the new unit. In addition, the Perfexion has the potential ability to treat indications down into the cervical spine (as low as 26 cm caudal from the vertex of the cranium [45], opening up the possibility of treating head and neck carcinomas and other pathologies. While this will require changes in fixation techniques and dose calculation algorithms [46], the new unit promises to significantly expand the pool of potential indications for the Gamma Knife.
1. Leksell L. The stereotaxic method and radiosurgery of the brain. Acta Chir Scand 1951;102(4):316-19. 2. Lindquist C, Kihlstrom L. Department of Neurosurgery, Karolinska Institute: 60 years. Neurosurgery 1996;39(5): 1016-21. 3. Wu A, Lindner G, Maitz AH, et al. Physics of gamma knife approach on convergent beams in stereotactic radiosurgery. Int J Radiat Oncol Biol Phys 1990;18(4):941-9. 4. Khan FM. The physics of radiation therapy. Baltimore: Williams and Wilkins; 1984. 5. Hall EJ. Radiobiology for the radiologist. 3rd ed. London: J. B. Lippincott; 1988. 6. Buatti JM, Friedman WA, Meeks SL, Bova FJ. The radiobiology of radiosurgery and stereotactic radiotherapy. Med Dosim;1998;23(3):201-7. 7. Hall EJ, Brenner DJ. The radiobiology of radiosurgery: rationale for different treatment regimes for AVMs and malignancies. Int J Radiat Oncol Biol Phys 1993;25(2): 381-5. 8. Leksell GammaPlan 8.0 online reference manual. 1003197 Rev. 01 ed. Stockholm: Elekta Instrument AB; 2006. 9. Leksell gamma knife perfexion: system description. Art no 1002703. Stockholm: Elekta, AB; 2006. 10. Leksell L, Lindquist C, Adler JR, Leksell D, Jernberg B, Steiner L. A new fixation device for the Leksell stereotaxic system. Technical note. J Neurosurg 1987;66(4): 626-9. 11. Mugler JP III, Brookeman JR. Three-dimensional magnetization-prepared rapid gradient-echo imaging (3D MP RAGE). Magn Reson Med 1990;15(1):152-7. 12. Casselman JW, Kuhweide R, Deimling M, Ampe W, Dehaene I, Meeus L. Constructive interference in steady state-3DFT MR imaging of the inner ear and cerebellopontine angle. AJNR Am J Neuroradiol 1993;14(1): 47-57. 13. Held P, Fellner C, Fellner F, Seitz J, Strutz J. MRI of inner ear anatomy using 3D MP-RAGE and 3D CISS sequences. Br J Radiol 1997;70(833):465-72. 14. Stuckey SL, Harris AJ, Mannolini SM. Detection of acoustic schwannoma: use of constructive interference in the steady state three-dimensional MR. AJNR Am J Neuroradiol 1996;17(7):1219-25. 15. Hlatky R, Jackson EF, Weinberg JS, McCutcheon IE. Intraoperative neuronavigation using diffusion tensor MR tractography for the resection of a deep tumor adjacent to the corticospinal tract. Stereotact Funct Neurosurg 2005;83(5-6):228-32. 16. Litt AW, Kondo N, Bannon KR, Kricheff II. Role of slice thickness in MR imaging of the internal auditory canal. J Comput Assist Tomogr 1990;14(5):717-20. 17. Johnson CD, Fletcher JG, MacCarty RL, et al. Effect of slice thickness and primary 2D versus 3D virtual dissection on colorectal lesion detection at CT colonography in
Conclusions Lars Leksell conceived ‘‘radiosurgery’’ as a minimally-invasive technique for neurosurgeons. It is only very recently that radiosurgery has become a viable extracranial technique, and it is possible only because of tremendous advances in computing power, imaging technology, and manufacturing precision. Leksell developed his device without these advances, and its basic design remains basically unchanged to this day. It is a tribute to the elegant design of the Gamma Knife and its specificity to the problems of brain surgery that the Gamma Knife remains the ‘‘gold standard’’ by which new devices are compared. Over the coming decades it will be interesting to observe what changes new technology will bring to the field of radiosurgery. The authors hope that the accrued experience of the Gamma Knife will help temper what is often a rush to push new technology as a panacea to every clinical problem and will help clarify the appropriate role of radiosurgery both inside and outside of the brain.
927
928
58 18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
Gamma knife: technical aspects
452 asymptomatic adults. AJR Am J Roentgenol 2007; 189(3):672-80. Snell JW, Sheehan J, Stroila M, Steiner L. Assessment of imaging studies used with radiosurgery: a volumetric algorithm and an estimation of its error. Technical note. J Neurosurg 2006;104(1):157-62. Sumanaweera TS, Adler JR Jr, Napel S, Glover GH. Characterization of spatial distortion in magnetic resonance imaging and its implications for stereotactic surgery. Neurosurgery 1994;35(4):696-703; discussion 694-703. Worthington C, Hutson K, Boulware R, et al. Computerized tomography cisternography of the trigeminal nerve for stereotactic radiosurgery. Case report. J Neurosurg 2000;93 Suppl 3:169-71. Bednarz G, Downes B, Werner-Wasik M, Rosenwasser RH. Combining stereotactic angiography and 3D time-offlight magnetic resonance angiography in treatment planning for arteriovenous malformation radiosurgery. Int J Radiat Oncol Biol Phys 2000;46(5):1149-54. Soderman M, Picard C, Ericson K. An algorithm for correction of distortion in stereotaxic digital subtraction angiography. Neuroradiology 1998;40(5):277-82. Shepard DM, Ferris MC, Ove R, Ma L. Inverse treatment planning for Gamma Knife radiosurgery. Med Phys 2000;27(12):2748-56. Wu QJ, Chankong V, Jitprapaikulsarn S, et al. Real-time inverse planning for Gamma Knife radiosurgery. Med Phys 2003;30(11):2988-95. Zhang P, Wu J, Dean D, et al. Plug pattern optimization for gamma knife radiosurgery treatment planning. Int J Radiat Oncol Biol Phys 2003;55(2):420-7. Morgan-Fletcher SL. Prescribing, recording and reporting photon beam therapy (Supplement to ICRU Report 50), ICRU Report 62. Br J Radiol 2001; 74(879):294. Thorsen FA, Ganz JC. Dose planning with the Leksell Gamma Knife: the effect on dose volume of more than one shot at the same target point. Stereotact Funct Neurosurg 1993;61 Suppl 1:151-63. Schlesinger D, Snell J, Sheehan J. Shielding strategies for Gamma Knife surgery of pituitary adenomas. J Neurosurg 2006;105(7):241-8. Paddick I. A simple scoring ratio to index the conformity of radiosurgical treatment plans. Technical note. J Neurosurg 2000;93 Suppl 3:219-22. Borden JA, Mahajan A, Tsai JS. A quality factor to compare the dosimetry of gamma knife radiosurgery and intensity-modulated radiation therapy quantitatively as a function of target volume and shape. Technical note. J Neurosurg 2000;93 Suppl 3:228-32. Lomax NJ, Scheib SG. Quantifying the degree of conformity in radiosurgery treatment planning. Int J Radiat Oncol Biol Phys 2003;55(5):1409-19.
32. Wu QR, Wessels BW, Einstein DB, Maciunas RJ, Kim EY, Kinsella TJ. Quality of coverage: conformity measures for stereotactic radiosurgery. J Appl Clin Med Phys 2003;4(4):374-81. 33. Paddick I, Lippitz B. A simple dose gradient measurement tool to complement the conformity index. J Neurosurg 2006;105(7):194-201. 34. Wagner TH, Bova FJ, Friedman WA, Buatti JM, Bouchet LG, Meeks SL. A simple and reliable index for scoring rival stereotactic radiosurgery plans. Int J Radiat Oncol Biol Phys 2003;57(4):1141-9. 35. Korytko T, Radivoyevitch T, Colussi V, et al. 12 Gy gamma knife radiosurgical volume is a predictor for radiation necrosis in non-AVM intracranial tumors. Int J Radiat Oncol Biol Phys 2006;64(2):419-24. 36. Ma L, Li XA, Yu CX. An efficient method of measuring the 4 mm helmet output factor for the Gamma knife. Phys Med Biol 2000;45(3):729-33. 37. Bilge H, Osen Z, Senkesen O, Kucucuk H, Cakir A, Sengoz M. Determination of output factors for the Leksell gamma knife using ion chamber, thermoluminescence detectors and films. J BUON 2006;11(2):223-7. 38. Cheung JY, Yu KN, Ho RT, Yu CP. Monte Carlo calculated output factors of a Leksell Gamma Knife unit. Phys Med Biol 1999;44(12):N247-N249. 39. Sanders M, Sayeg J, Coffey C, Patel P, Walsh J. Beam profile analysis using GafChromic films. Stereotact Funct Neurosurg 1993;61 Suppl 1:124-9. 40. Maryanski MJ, Ibbott GS, Eastman P, Schulz RJ, Gore JC. Radiation therapy dosimetry using magnetic resonance imaging of polymer gels. Med Phys 1996;23(5): 699-705. 41. Scheib S, Crescenti R, Vogelsanger W, et al. Application of normoxic polymer gels in 3D-dosimetry for radiosurgery. Z Med Phys 2006;16(3):180-7. 42. Watanabe Y, Akimitsu T, Hirokawa Y, Mooij RB, Perera GM. Evaluation of dose delivery accuracy of Gamma Knife by polymer gel dosimetry. J Appl Clin Med Phys 2005;6(3):133-42. 43. Karaiskos P, Petrokokkinos L, Tatsis E, et al. Dose verification of single shot gamma knife applications using VIPAR polymer gel and MRI. Phys Med Biol 2005; 50(6):1235-50. 44. Sandilos P, Tatsis E, Vlachos L, et al. Mechanical and dose delivery accuracy evaluation in radiosurgery using polymer gels. J Appl Clin Med Phys 2006;7(4):13-21. 45. Lindquist C, Paddick I. The Leksell Gamma Knife Perfexion and comparisons with its predecessors. Neurosurgery 2007;61 Suppl 3:130-40; discussion 131-140. 46. Solberg TD, Holly FE, De Salles AA, Wallace RE, Smathers JB. Implications of tissue heterogeneity for radiosurgery in head and neck tumors. Int J Radiat Oncol Biol Phys 1995;32(1):235-9.
75 Gamma Knife Radiosurgery: Technical Issues D. Kondziolka . A. Niranjan . J. Novotny . J. Bhatanagar . L. D. Lunsford
The Gamma Knife was developed by Lars Leksell and Borje Larsson, to achieve their goal of an efficient, precise, hospital-based stereotactic radiosurgery system [1]. Clinical work with the Gamma Knife began in 1967 and the first patient had a craniopharyngioma. The patient’s head was immobilized using a plaster-molded headpiece. Subsequently, gamma knife surgery was performed in patients with pituitary tumors, vestibular schwannomas, vascular malformations, and functional disorders such as intractable pain. In 1975, a series of surgical pioneers at the Karolinska Hospital, Stockholm began to utilize a new Gamma Knife, redesigned to create a more spheroidal dose-profile better suited for the treatment of intracranial tumors and vascular malformations. Units 3 and 4 were placed in Buenos Aires and Sheffield England in the early 1980s. Lunsford et al. introduced the first clinical 201-source Gamma Knife unit to North America (the fifth gamma unit worldwide) and the first patient was managed in August 1987 at the University of Pittsburgh Medical Center [2]. The encouraging results of radiosurgery for benign tumors and vascular malformations led to an exponential rise of radiosurgery cases and installations of radiosurgical units (> Table 75-1). In recent years metastatic brain tumors have become the most common indication of radiosurgery. Brain metastases now comprise 30–50% of radiosurgery cases at busy centers.
#
Springer-Verlag Berlin/Heidelberg 2009
The Evolution of Gamma Knife Technology Models A, B, and C There have been numerous changes to the Gamma Knife since the original 1967 design. In the first models (Model U or A) 201 Cobalt sources were arranged in a hemispheric configuration. These units presented challenging Cobalt-60 loading and reloading issues. To facilitate reloading, the unit was redesigned so that sources were arranged in a circular configuration (Model B, C, and 4C) (> Figure 75-1). Gamma Knife radiosurgery usually involves the use of single or multiple isocenters of different beam diameters to achieve a treatment plan that conforms to the 3-dimensional volume of the target. The total number of isocenters may vary depending upon the size, shape, and location of the target. Each isocenter has a set of three Cartesian (X, Y, Z) stereotactic coordinates corresponding to its location in three-dimensional space as defined using a rigidly fixed stereotactic frame. When multiple isocenters are used, the stereotactic coordinates will need to be set individually. In 1999, the Model C Gamma Knife was introduced and first installed in the United States at the University of Pittsburgh Medical Center in March 2000 [3]. This technology combined dose planning advances with robotic engineering. The unit incorporated an automatic positioning system
1226
75
Gamma knife radiosurgery: technical issues
. Table 75-1 Brain Disorders treated World-wide Using Gamma Knife Radiosurgery by December 2006
Brain disorder
Indications
Vascular disorders
AVM
Benign tumors
Malignant tumors
Functional targets
Ocular disorders
Aneurysm Cavernous malformation Other vascular Vestibular schwannoma Trigeminal schwannoma Other schwannoma Meningioma Pituitary adenoma Pineal region tumor Craniopharyngioma Hemangioblastoma Hemangiopericytoma Chordoma Glomus tumor Other benign tumors Glial tumors (grade I–II) Glial tumors (grade III–IV) Metastatic tumor Chondrosarcoma Nasopharyngeal carcinoma Other malignant tumors Trigeminal neuralgia Parkinson’s disease Pain Epilepsy Obsessive compulsive disorder Other functional targets Uveal melanoma Glaucoma Other ocular disorders Total indications
Number of patients treated 48,407 270 1,887 3,777 36,843 2,312 1,005 49,558 31,901 3,150 3,397 1,656 946 1,619 1,107 3,490 2,169 23,610 14,1210 520 1,277
(APS) with submillimetric accuracy, used to move the frame to each coordinate. This technology obviates the need to manually adjust each set of coordinates in a multiple-isocenter plan. The robot eliminates the time spent removing the patient from the helmet, setting the new coordinates for each isocenter and repositioning the patient in the helmet. This has significantly reduced the total time spent to complete the procedure and also increased accuracy and safety [4–8]. The other features of the Model C unit include an integral helmet changer, dedicated helmet installation trolleys, and color-coded collimators. In 2005 the fourth generation Leksell Gamma Knife model 4C was introduced. The first unit was installed at the University of Pittsburgh in January of 2005. The model 4C is equipped with enhancements designed to improve workflow and provide integrated imaging capabilities. The imaging enhancements available with the Leksell GammaPlan, offers image fusion capability. These images can also be exported to a CD-ROM, so the referring physician can receive pre- or post-operative images for reference and follow-up. The planning information can be viewed on both sides of the treatment couch. The helmet changer and robotic Automatic Positioning System are faster and reduce total treatment time.
5,609 25,198 1,309 566 2,243 140 867 1,354 210 65 397,672
LGK Perfexion The newest iteration of Gamma Knife technology is the PERFEXION unit. Beginning in 2002, an invited group of neurosurgeons, radiation oncologists and medical physicists was asked by the manufacturer to define specifications for a new Leksell Gamma Knife system. The group agreed on five critical features for a new system: (1) best dosimetry performance, (2) unlimited cranial reach, (3) best radiation protection for patient and stuff, (4) full automation of the treatment process,
Gamma knife radiosurgery: technical issues
75
. Figure 75-1 Schematic diagram of the Leksell Gamma Knife 4C
(5) patient and staff comfort, and (6) similar radiation dose profiling as prior units. The new unit was first installed in 2006, and was first used at our center in September 2007 (> Figure 75-2). The radiation unit was redesigned. A total of 60 192 Co sources were arranged in a cylindrical configuration in five concentric rings. This differs substantially from the previous hemispherical arrangements and results in different source to focus distances for each ring varying from 374 to 433 mm. The primary and secondary collimators have been replaced by a single large 120 mm thick tungsten collimator array ring (> Figure 75-3). Consequently no collimator helmets are needed for the PEREFXION system. Three collimators are available for the PERFEXION system. The 4 and 8 mm collimators remain, and a new 16 mm collimator
replaces the prior 14 and 18 mm collimators. The tungsten collimator array is subdivided into eight identical but independent sectors, each containing 72 collimators (24 collimators for 4 mm, 24 collimators for 8 mm, 24 collimators for 16 mm). The collimator size for each sector is changed automatically by moving 24 sources over the selected collimator set. Each sector with 24 sources can be moved independently into five different positions: (1) sector in home position when system is standby, (2) 4 mm collimator, (3) 8 mm collimator, (4) 16 mm collimator, and (5) sector off position (defined as the position between the 4 and 8 mm collimators providing blocking of all 24 beams for that sector) (> Figure 75-3). Sector movement is performed by servo-controlled motors with linear scales located at the rear of the radiation unit.
1227
1228
75
Gamma knife radiosurgery: technical issues
. Figure 75-2 Leksell Gamma Knife PERFEXION
The radiation cavity has been increased by more than 300% compared to previous models. However, due to an improved collimation system (120 mm tungsten ring), the average distance from source to focus is very close to previous models. This results in similar output for the prior 18 mm and new 16 mm administrations. The increase in the volume of the radiation cavity to more than three times allows for a greater mechanical treatment range in X/Y/Z. It is (160/180/220 mm) for the PERFEXION system compared to (100/120/ 165 mm) for other gamma knife models. This provides virtually unlimited cranial reach, so crucial in the care of patients with multiple brain metastases [9]. To date, we have seen no problems related to potential helmet collisions. The Automatic Positioning System (APS) used in the C units was replaced by the Patient Positioning System (PPS). Rather than just the head, the whole couch moves into pre-selected stereotactic coordinates. This provides better patient comfort and allows complete of the majority of radiosurgeries in one single run. Docking of the patient into the PPS is done by means of an
adaptor that attaches to the standard stereotactic Leksell G frame with three clips. The adapter is then directly docked to the PPS (> Figure 75-4). The patient can be attached in three different positions, with gamma angles of 70, 90 or 110 reflecting neck flexion or extension. The gamma angle is the only treatment parameter that requires manual set up. The PPS has repeatability better than 0.05 mm. The redesigned hardware of the PERFEXION unit has had significant impact on the planning software Leksell GammaPlan PFX (LGP PFX), a new version of the LGP running on a PC platform with the Linux operating system. There are in principle three possible approaches in the treatment planning: (1) use of classical combinations of 4, 8, and 16 mm isocenters (shots), (2) use of composite shots containing combinations of 4, 8, 16 mm or blocked sectors, and (3) dynamic shaping using blocked selected sectors to protect volumes defined as critical structures (> Figure 75-5). The most revolutionary change in the treatment planning is the ability to generate a single isocenter composed of different beam diameters. Such
Gamma knife radiosurgery: technical issues
75
. Figure 75-3 Diagrams of the PERFEXION Gamma Knife radiation unit and collimator system. (a) Cross section of the Leksell Gamma Knife PERFEXION radiation unit. (b) Each sector holds 24 60Co sources and can be moved independently on other sectors to the desired position to define collimator size or block groups of beams. (c) Sector position in 4 mm collimator. (d) Sector position in 8 mm collimator. (e) Sector position in 16 mm collimator
a composite shot design allows an optimized dose distribution shape for each individual shot. The setup of any sectors, combinations of different collimators, or blocking takes only minimal time (a few seconds done automatically). The new PERFEXION system provides further improvements in patient and staff radiation shielding. The sectors are always in the off position (blocked) during patient transportation in the treatment position, transition into new stereotactic coordinates, pause or emergency interrupt. These results in significantly (about 5–10 times) lower extracranial irradiation to the patient compared to modelsBandC.Ourpreliminarycomparisonstudy shows that for a patient with ten brain lesion, the total time saved is about 1.5–2.0 h compared to other systems. The new Leksell Gamma Knife PERFEXION provides excellent dosimetry
performance, unlimitedcranialreach, enhancedradiationprotectionforpatientandstaff,fullautomation of the treatment process and better patient and staff comfort compared to previous models (> Table 75-2). Thus, the PERFEXION unit provides the potential to increase the spectrum of treatable indications including multiple brain metastasis, access to the upper cervical spine, and other pathologies of the head and neck. The use for the lower to mid-cervical spine will require the development of a new fixation device.
The Radiosurgery Procedure In the following sections we discuss the basic technical steps of gamma knife radiosurgery using LGK 4C and PERFEXION.
1229
1230
75
Gamma knife radiosurgery: technical issues
. Figure 75-4 Leksell stereotactic frame is docked by means of frame adapter in the Leksell Gamma Knife PERFEXION
. Figure 75-5 Example of the treatment plan with composite shots for the Leksell Gamma Knife PERFEXION. Multiple 4 mm collimators were used to design dose plan for a vestibular schwannomas. Sector blocking was used in one shot to achieve high conformity and sharp dose fall
Daily Quality Assurance Gamma Knife quality assurance testing is performed by an authorized medical physicist every morning. The purpose of Daily Quality Assurance is to assure proper system function in standard treatment conditions plus verify all safety and
emergency functions. The medical physicist ensures that all system tests required by Nuclear Regulatory Commission (NRC) regulations are performed and functioning properly. These tests include the permanently mounted radiation monitor inside the treatment room and it’s remote indicator, hand held radiation monitor, patient
Gamma knife radiosurgery: technical issues
75
. Table 75-2 Technical Specifications of Different Leksell Gamma Knife Units Comparing parameter Accuracy Radiological accuracy Positioning accuracy Positioning repeatability Radiation safety Room shielding required around back wall, 180 Room shielding required around front wall, 180 Body dose to patient lower than other devices Treatment planning Mechanical treatment range X/Y/Z Shape of accessible volume Effective target dose rate Composite shot Dynamic shaping Workflow Typical treatment time Approximate patient set up time Stereotactic coordinates set up time Collimator size set up time Collimator blocking set up time Composite collimator set up time
Leksell Gamma Knife PERFEXION
Leksell Gamma Knife 4C
Leksell Gamma Knife B
<0.25 mm <0.20 mm <0.05 mm
<0.50 mm <0.30 mm <0.20 mm
<0.50 mm <0.50 mm <0.25 mm
0 cm Standard concrete
10–100 times lower
10 cm Standard concrete <50 cm Standard concrete 2–20 times lower
10 cm Standard concrete <50 cm Standard concrete 2–20 times lower
160/180/220 mm Cylindrical >3.0 Gy/min Yes Yes
100/120/165 mm Spherical >3.0 Gy/min No No
100/120/150 mm Spherical >3.0 Gy/min No No
20 min 2 min 3–5 s <3 s <3 s <3 s
50 min 10 min 30–40 s 7–10 min 10–20 min NA
80 min 10 min 7–10 min 7–10 min 10–20 min NA
<50 cm Standard concrete
viewing and communication systems, door interlock, timer termination of exposure, treatment pause and emergency stops, test of all required beam status indicators and alarm indicators, availability of the release rods for the emergency removal of a patient and function of the helmet hoist used for collimator helmet exchange. A test run simulating the standard treatment procedure is performed and also includes check of automatic positioning system accuracy test. There are other monthly and annual checks, as well as preventive maintenance of the Leksell Gamma Knife. Daily Quality Assurance for PERFEXION system is very similar to 4C system as described above. The test run includes a sector positioning check verifying proper function of automatic collimator set up. The patient docking device function is tested together with overall system
accuracy (including patient positioning system and sector positioning) by a diode test tool and focus precision check. The same monthly and annual tests are performed as for 4C system.
Application of the Stereotactic Frame For Gamma Knife radiosurgery, appropriate stereotactic frame placement is an initial critical part of the procedure. Prior to frame placement, the radiosurgery team should review the preoperative images and discuss optimal frame placement strategy. An effort should be made to keep the lesion as close to the center of the frame as possible. The possibility of collision by the frame base ring, the posts/pins assembly, or the patient’s
1231
1232
75
Gamma knife radiosurgery: technical issues
head with the collimator helmet during treatment should also be considered prior to the frame application. Steps to avoid a possible collision should be taken during frame placement. These collisions should be minimal with the LGK PERFEXION. The frame is shifted lower or higher on the head, to the left or right, or anteriorly or posteriorly, using the ear bars attached to the sides of the base ring. The anterior posts are positioned low along the supraorbital region to avoid collision of frontal post/pin assembly with the collimator helmet. For radiosurgery planning, a higher gamma angle (120–140 ) is used, if a collision is detected at the default angle of 90 . While shifting frame laterally, it is important to make sure that there is enough space on the contralateral side to allow positioning of the fiducial box on the base ring of the frame. The MRI or CT fiducial should be tried on frame prior to sending the patient to MRI unit. If the fiducial box does not fit on to the frame due to excessive shifting of the frame, the frame will have to be repositioned.
Frame Adaptor and Frame Cap Fitting Check The frame adaptor (which attaches the frame to the table) is checked for fit. If the frame is shifted too anteriorly and the back of head ring is too close to the neck adaptor may not fit and consequently treatment can not be carried out. Tight fitting of the adaptor may cause neck discomfort to patient especially during a long treatment. The frame cap check provides information about geometry of all stereotactic frame parts including posts and screws and also information about patient head geometry to the treatment planning system. This information is needed for the prediction of potential collisions or close contact with the gamma knife unit collimator system. If frame cap does not fit then the exact post and screw measurements must be given to the treatment planning system.
Techniques of Stereotactic Imaging Imaging is crucial in radiosurgery. Magnetic resonance imaging (MRI) is the preferred imaging modality. CT is used when MR imaging is not possible. Angiograms are used in conjunction with MRI for arteriovenous malformation radiosurgery. Since 1993, MRI has been used for stereotactic radiosurgery planning in almost all eligible cases using a 1.5 Tesla unit. In addition, arteriovenous malformations are imaged also by biplane angiography. At our institution high-resolution, a gadolinium enhanced 3-D localizer (T2* images) image sequence is used first to localize the tumor in axial, sagittal, and coronal images. Using the axial images, the fiducials can be measured and compared to the opposite side to exclude the possibility of MR artifacts and confirm that there is no angulation or head tilt. The average time for this sequence is approximately 1.5 min. For stereotactic imaging of most lesions, a 3Dvolume acquisition using Fast Spoiled-Gradient Recalled Acquisition in Steady State (GRASS) sequence at 512 256 matrix and 2 NEX (number of excitations) covering the entire lesion and surrounding critical structures is preferred. To define the radiosurgery target, this volume is displayed as 1 or 1.5 mm thick axial slices. The FOV (field of view) is kept at 25 25 cm in order to visualize all fiducials. The approximate imaging time for this sequence is 8 min. We generally prefer 3-D Spoiled-GRASS sequence for most lesions. Additional sequences are performed when more information is needed. Pituitary lesions are particularly difficult to image especially if there has been prior surgery. A half dose of paramagnetic contrast is usually given to image pituitary adenomas. For residual pituitary tumors, after trans-sphenoid resection, a fat suppression SPGR sequence is recommended in order to differentiate tumor from the fat packed in the resection cavity. For cavernous malformations, an additional
Gamma knife radiosurgery: technical issues
VEMP (Variable Echo Multi Planar) imaging is obtained to define the hemosiderin rim. For thalamotomy planning, an additional fast inversion recovery sequence is performed to differentiate thalamus from the internal capsule. Brain metastases patients receive a double dose of contrast agent and the entire brain is imaged by 2 mm slices to identify all of the lesions. Before removing the patients from the MR scanner, the images must be checked for accuracy. When using CT imaging, it is advisable to use short posterior posts to avoid metallic artifacts from the posts and pins. Care should also be taken in deciding the optimal place for the pins since they cause artifacts on CT. With modern CT scanners 1 or 2 mm thick slices (depending upon the size of the lesion) without any gap can be obtained quickly. Angiography is the gold standard for AVM radiosurgery planning. It should be used in conjunction with MR or CT imaging. The orthogonal images (instead of oblique or rotated) are preferred but are not necessary. For AVM nidi that are not properly visualized in orthogonal planes a rotation of up to 10 in two dimensions or aspects can be used without compromising the accuracy of radiation delivery. Before removing the angiography catheter the images should be reviewed to make sure that all the fiducials are seen on the images. Digital subtraction techniques, despite a potential radial distortion error, have proven satisfactory spatial accuracy.
Determination of Target Volumes Target determination is an important step in order to make a conformal plan. Target volume can be outlined using the LGP software (manual or semiautomatic mode). Although experienced surgeons can create conformal dose plans without outlining the target, the target outline allows for a more quantitative assessment of the plan. For new centers especially where physicists
75
assume the initial responsibility for planning, target definition and outlining by the surgeon or oncologist becomes an important step. The surgeon’s input is required to define radiosurgery targets for patients with AVMs, tumors and functional neurosurgery as used by some centers. By defining the target volume and volumes of critical structures better evaluation and quantification of the treatment plan can be done. Various parameters such as dose volume histograms for the target volume and critical structures plus conformity indexes can be obtained.
Techniques of Conformal Dose Planning In the process of treatment planning, several strategies can be used. The Model C allows treatment using robotic automatic patient positioning system (APS mode), manual positioning (trunnion mode) or mixed treatment (some isocenters in APS mode and some in trunnion mode). Most users will select shots and directly place them over the target. Beginners can also use the inverse dose-planning algorithm (Wizard) to create a plan and then optimize it manually. The conformal dose planning is enhanced by the use of multiple small collimators. Three different approaches in the treatment planning can be applied when using LGK PERFEXION. The first is to use the same strategy as described for 4C system above. Since only 4, 8, and 16 mm collimators are available only combination of these three different collimators can be used to cover the entire target volume. The second approach is to use dynamic shaping that is new feature in the treatment planning introduced for the PEREFXION system. This automatic procedure will provide solutions to block selected sectors to protect volumes defined as critical structures. Different levels to reduce dose delivered to critical structures can be selected. The treatment planning system then automatically calculates
1233
1234
75
Gamma knife radiosurgery: technical issues
which sectors should be blocked for each individual shots. One should be aware that each blocking will significantly increase total exposure time. The third approach is to use single isocenters composed of different beam diameters or blocked sectors. Any pattern of sectors including 4, 8, 16 mm collimators and blocks can be generated. This can help significantly for shaping dose distribution especially for irregular volumes.
Radiation Administration During Stereotactic Radiosurgery The Model 4C Gamma knife allows radiation delivery using trunnion mode (manual patient positioning) or APS mode (Robotic positioning) or a combination of the two (mixed treatment). In trunnion treatment, the x, y, and z of each isocenter are set manually and triple-checked to avoid errors. The APS plan is transferred directly from the planning computer to control computer. The operator selects the run (a combination of isocenters of same beam diameter) that matches the collimator helmet on the Gamma Unit. The APS is moved to the dock position and the patient’s head frame is fixed into the APS. The accuracy of the docking position is checked. The system prompts the user to perform clearance checks first for all those planned isocenters in which the pins, posts, frame or patient’s head would be less than 12 mm away from the inner surface of the collimator helmet (even though they may not match with the collimator size which is being used for first run). The clearance check is performed by moving the patient to those positions under APS manual control and by visual check of collision with the collimator helmet. After the clearance check, the system prompts the surgeon to carry out position checks. In the position checks, all the isocenters using the same helmet are checked, one by one, by moving the patient’s head to these positions using APS manual control to make sure patient will handle all head position
changes with sufficient comfort. All personnel then leave the room, and the radiosurgical dose is administered. The APS moves the patient to all planned positions, one by one, until the isocenters using that size collimator helmet are completed. The team monitors the patient and the coordinates of different isocenters on the control computer. If other runs using a different gamma angle but using the same helmet are planned, then the patient is taken out, next run is selected, APS is moved to the dock position and patient’s head is again fixed in the APS using the planned angle (72, 90, 110, or 125 ). Radiosurgery with the LGK PERFEXION is a fully automated process for all aspects of the procedure including set up of the stereotactic coordinates, set up of different sector positions defining collimator size or blocked beams and set up of exposure times. All treatment data are exported to the operating console. The only manual part of the procedure is the positioning of the patient’s head in the docking device and adjustment of the couch height for optimal comfort. After confirmation of the patients’ identity, most PERFEXION radiosurgeries are administered in one single run (95%). Rarely a clearance check is needed. For this, a special test tool simulating the shape and dimensions of the inner collimator is attached and rotated around patient’s head. Once radiosurgery begins, the team monitors the patient and the set up of coordinates, exposure times and sector set up of different isocenters on the control computer of operating console. The system allows audiovisual communication with the patient during irradiation and the process can be interrupted at any time if needed.
Conclusion In the past two decades, we have witnessed dramatic improvements in stereotactic radiosurgery technologies. Gamma knife radiosurgery now
Gamma knife radiosurgery: technical issues
offers better image-handling features, faster and more compact software platforms that make the calculations almost real time, automated and robotic patient positioning thus reducing the potential for human error, inverse treatment planning, and expanded indications.
References 1. Leksell L. The stereotaxic method and radiosurgery of the brain. Acta Chir Scand 1951;102:316-19. 2. Lunsford LD, Flickinger J, Lindner G, Maitz A. Stereotactic radiosurgery of the brain using the first United States 201 cobalt-60 source gamma knife. Neurosurgery 1989;24:151-9. 3. Kondziolka D, Maitz AH, Niranjan A, Flickinger JC, Lunsford LD. An evaluation of the model C gamma knife with automatic patient positioning. Neurosurgery 2002;50:429-31; discussion 431–2.
75
4. Goetsch SJ. Risk analysis of Leksell gamma knife model C with automatic positioning system. Int J Radiat Oncol Biol Phys 2002;52:869-77. 5. Kuo JS, Yu C, Giannotta SL, Petrovich Z, Apuzzo ML. The Leksell gamma knife model U versus model C: a quantitative comparison of radiosurgical treatment parameters. Neurosurgery 2004;55:168-72; discussion 172–3. 6. Regis J, Hayashi M, Porcheron D, Delsanti C, Muracciole X, Peragut JC. Impact of the model C and automatic positioning system on gamma knife radiosurgery: an evaluation in vestibular schwannomas. J Neurosurg 2002;97:588-91. 7. Tlachacova D, Schmitt M, Novotny J Jr, Novotny J, Majali M, Liscak R. A comparison of the gamma knife model C and the automatic positioning system with Leksell model B. J Neurosurg 2005;102 Suppl:25-8. 8. Yu C, Jozsef G, Apuzzo ML, MacPherson DM, Petrovich Z. Fetal radiation doses for model C gamma knife radiosurgery. Neurosurgery 2003;52:687-93; discussion 693. 9. Lindquist C, Paddick I. The Leksell Gamma Knife Perfexion and comparisons with its predecessors. Neurosurgery 2007;61:130-40; discussion 140–1.
1235
62 IMRT: Technical and Clinical Aspects M. P. Carol
Introduction The primary goal of a focused radiation therapy technique such as radiosurgery is the eradication of the target, be it a tumor or a vascular malformation. In general, this is accomplished by delivering as high a dose of radiation to the target as possible. Theoretically, if one could deliver radiation in such a way that only the target, regardless of its shape, received a lethal dose, while surrounding sensitive structures received no dose at all, radiation therapy could significantly alter the outcome in the vast majority of cancers. However, the dose of radiation needed to kill a tumor cannot always be used because it might also be delivered to, and possibly destroy, adjacent normal tissue that must be spared if the patient’s immediate well being is to be preserved. The fact that these two objectives–kill the tumor and preserve normal tissue–often are in conflict has limited the overall effectiveness of radiation in the fight against cancer. Intensity modulated radiation therapy (IMRT) has been used to reduce or eliminate this conflict, in the process creating and delivering some of the most complex dose distributions imaginable. Studies of patients treated for prostate cancer, lung cancer, cancer of the head and neck, and cancers of the brain suggest strongly that the use of IMRT increases the rate of cure and prolongs survival in those patients not cured, while simultaneously decreasing, and in some cases eliminating, complications. Initially commercialized by NOMOS in 1992 and used clinically for the first time at Methodist Hospital in Houston, Texas in 1994, IMRT has #
Springer-Verlag Berlin/Heidelberg 2009
been implemented, or is in the process of being implemented, at more than 30% of all radiation oncology facilities in the U.S. and being used all around the world. It has become the standard of care for patients with prostate cancer, head and neck cancers, tumors involving the spine or the abdomen, tumors of the CNS, and is being explored for use in breast and lung cancer. Beginning in 1996, a specialized version of IMRT, Intensity Modulated RadioSurgery (IMRS), a term coined by NOMOS and promoted by BrainLab, was introduced into the clinic. IMRS differs from IMRT only in the number of fractions (1–5) used to treat the patient and in the range of delivery technology available. Early in its development IMRS incorporated some form of rigid fixation of the target anatomy as did all other forms of RS. However, that ‘‘requirement’’ has been relaxed with the development of highly sophisticated ‘‘image guided radiation therapy’’ (IGRT) techniques that insure that the target volume is correctly positioned relative to the treatment beam throughout the period of radiation delivery (> Figure 62-1). All IM techniques share at least two features; the ability to deliver non-uniform intensity maps of radiation and an inverse planning system used to determine the intensity map. Non-uniform intensity maps or beam modulation involves varying the intensity or strength of the treatment fields, used to treat the patient, across the fields. In general, IMRT systems do this by functionally dividing the fields into a very large number of very small beams each of which can have its intensity varied (imagine using a lot of small spotlights with independent dimmer switches,
966
62
IMRT: technical and clinical aspects
. Figure 62-1 Principle of IMRS treatment plan. (a) By delivering modulated fields from three directions it is possible to create a dose distribution that treats conformally the target (b) Even with the addition of regions with dose limitations, it is still possible to move dose around such that the resultant dose is conformal to the target and limits dose to other regions specifically as required (stippled areas are regions of dose limitation, the severity of the limit proportional to the density of stippling)
rather than a single large light with only an on/off switch, to illuminate a room). By hitting all or portions of the target from a number of directions with these small sub-beams while simultaneously
varying the intensity of the sub-beams, it becomes possible to deliver a high dose of radiation to the target, regardless of its shape or size, while controlling the amount of high dose received by
IMRT: technical and clinical aspects
any specific volume of uninvolved tissue, regardless of that tissue’s location relative to the target. For instance, those portions of a field that pass through dose limiting structures could have their intensities dialed down while those portions that pass though regions that are not dose-limited could have their intensities increased so as to deliver the desired dose to the target. This process of limiting the strength of beams that pass through sensitive tissue is not new to IMRT, but the ability to reduce the intensity of only a portion of a beam, by dividing the beam up into a large number of very small independently controlled pieces, is. IMRS may use as many as fifty thousand or more beams each as small as several millimeters in size. This allows a ‘‘finer grain’’ of dose delivery, and therefore a more conformal plan, then is possible with older ‘‘conventional’’ RS delivery techniques. The effective exclusion by IMRT of normal tissue from exposure to high radiation doses permits increases in the tumor dose to levels beyond those feasible with conventional radiation therapy. At the same time, it can reduce the risk of complications that may occur in normal tissue. IMRT can be viewed as a noninvasive scalpel, cutting out the bad and leaving the good; as such, IMRT can be viewed as the ultimate form of radiation treatment delivery. It also can be viewed as a means for delivering a more ‘‘idealized’’ dose to both target tissue and nontarget tissue alike. In the past dose prescriptions were based more on what could be tolerated by normal tissue than on what was the dose required to destroy target tissue. As a result targets were often underdosed in order to avoid normal tissue complications. With IMRT, because of its ability effectively to shield normal tissue from excessive dose, prescriptions can be tailored to meet target requirements. In addition to the ability to vary physically beam intensity across a field the second essential component of an IM system is an inverse treatment planning system. Every treatment
62
delivered to a patient must have a ‘‘plan,’’ a set of instructions that specifies where and how the radiation dose will be delivered, and what the distribution of radiation (dose) inside the patient will look like if that exact plan is delivered. This plan is created by a computerized treatment planning system.
Treatment Planning The term treatment planning system, as applied to radiation therapy and radiosurgical systems used in the past, probably was somewhat of a misnomer. These are not really planning systems; instead, they are extremely useful tools for dose simulation and plan visualization. Radiosurgical personnel put together a treatment plan (a combination of beams directions and strengths) based on their knowledge of radiation physics, the needs of the patient, their training, and their past clinical experience. The computer, in addition to providing sophisticated graphical tools to assist in the visualization of the best way to deliver the treatment, then calculates and displays the dose that would be achieved from such a physician-designed treatment plan. If the physician is not happy with the result, then changes will be made to the plan and a dose distribution will be created by the computer. This process is repeated until the entire radiosurgical team is happy with the plan. The process is called ‘‘forward planning’’ because a way of delivering the plan is determined first, then the result examined and modified in an iterative fashion until it is correct. The real ‘‘planner’’ in the system is the physician, the computer functioning as a high speed calculator and a graphic display device. Conversely, inverse planning, an essential aspect of IMRT and IMRS, starts with a description by the physician to the planning system of what the dose should be, the planning system determining what combination of beams and beam strengths
967
968
62
IMRT: technical and clinical aspects
are best suited for realizing such a goal. Inverse planning systems were first developed in the early 1990s as part of IMRT systems. When it came to considering the multitude of options available while working with the thousands of small beams required to deliver an IMRT plan, computers were simply better, faster, more complete, and more efficient than human treatment planners. A planning problem that would take a human days or weeks to solve, if it even could be solved, could be cracked by an inverse planning computer is a matter of hours; With improvements in computer technology and planning algorithms, IMRT and IMRS plans can now be created virtually on the fly. The secret behind an inverse planning system – the ‘‘smarts’’ if you will – is an optimization routine, a mathematical process used to determine the best combination of beams, beam angles and beam intensities, from all possible combinations, for achieving a desired result. From a functional point of view, the user enters into the planning system a set of goals: deliver such and such dose of radiation to the target volume while not exceeding a set dose for each organ or region at risk. The optimizer then uses any of a number of algorithms (gradient descent, simplex, Fournier transform, simulated annealing, etc.) to create a set of delivery parameters that will approximate these goals (‘‘approximate’’ rather than ‘‘achieve’’ because in most cases all goals and limits cannot be met in their entirety). Once reviewed and accepted by the user, the parameters are translated into directions that are delivery device specific and are used to deliver the treatment.
Treatment Delivery IMRS treatments can be delivered using the same technology for delivering IMRT, namely a linear accelerator equipped with a beam modulation device, include physical compensators and active
collimation systems, or by technology designed specifically for radiosurgery, namely the Gamma Knife and the Cyberknife. All of these devices can be used to vary the intensity of a field of radiation across the field. For the purposes of this discussion a field of radiation can be defined as a set of beams that share a common factor. The beams can all be delivered at the same time (as in the Gamma Knife) or from given gantry direction (as in the use of a MLC, binary collimator, or variable field shaping device). In theory, conformality is improved as the number of fields is increased; since rotational therapy is equivalent to anywhere from 40 to 60 fixed fields it is usually the most conformal. That said, the clinical advantage of a rotational treatment diminishes greatly as the number of fixed fields exceeds 9–11.
Physical Compensators Manufactured by a company called ‘‘.Decimal,’’ the intensity map is delivered by employing a custom physical compensator created by a milling machine that is placed between the linac and the patient, one for each field that is used to treat the patient. The compensators are manufactured on a ‘‘mail order patient-bypatient’’ basis and are placed in a tray attachment on a standard linear accelerator. Although requiring manual interaction with the accelerator between every treatment field, these devices are potentially able to create the most conformal dose distributions on a field-by-field basis because the grain size of the beamlets used to treat the patient can be smaller than any physical leaf used in a multileaf. They are also the most efficient in terms of treatment time (> Figure 62-2). Physical compensators for intensity modulation were developed to deliver IMRT. However, when used in conjunction with precision
IMRT: technical and clinical aspects
62
. Figure 62-2 Sample dose distributions. (a) Glioblastoma with two target zones treated to different doses; (b) Ethmoidal Meningioma; (c) Metastatic Adeno Ca of Breast; (d) AVM; (e) Spinal metastasis
localization and immobilization techniques, the very fine resolution possible with the, Decimal approach makes it a highly appropriate tool for IMRS. The only limitation is the need to use many fields with IMRS, thereby increasing the time required to change manually the physical compensators (> Figure 62-3).
Multileaf/Multivane Collimator Systems Linac-based systems use two different types of collimating systems, binary or multileaf. Binary collimators employ vanes with very limited travel (1–5 cm) and are used in the MIMiC and in the
969
970
62
IMRT: technical and clinical aspects
. Figure 62-3 Intensity modulating technologies. (a) Multileaf collimator; (b) Binary collimator (MIMiC on left, Tomotherapy Hi-Art on left); (c) Gamma knife perfexion; (d) Physical compensator (.decimal); (e) Cyberknife IRIS
Tomotherapy Hi-Art system. The MIMiC, part of the Peacock System, was the first device to be used to delivery IMRT and true IMRS while the Hi-ART (HA) consists of a second
generation binary collimator designed to treat larger fields with slightly smaller beams (6 mm vs. 7 mm). Multileaf collimators use leaves ranging in width from 2.5 to 5 mm that can
IMRT: technical and clinical aspects
travel as much as 20 cm across a field and are added on or integral to an accelerator.
Binary Collimators All binary collimators consist of tungsten vanes (40 in the MIMiC and 50 in the HA) used to define a single (HA) or two (MIMiC) slices of beams. Each vane is powered by a miniature pneumatic piston controlled by a solenoid valve. Turning the valve on causes air to flow to the front side of the piston, driving the vane out of the field and creating an opening (beamlet) through which radiation can travel. When the valve is turned off constant pressure is applied to the backside of the vane driving it back into the field and turning the beam off. Each vane has associated with it a set of sensors that track movement as well as speed of movement insuring that the device is operating correctly. By varying the amount of time each vane is open (the time an individual beamlet is on), the percentage of total dose from each beamlet can be changed anywhere from 0 to 100%. As a result of very rapid vane movement rotational treatments are possible; Treatments are delivered with the linac rotating around the patient in an arc (MIMiC) or in a helix (HA). A rotation is modeled by the planning system as a series of consecutive fixed ports; every 5 to 10 of rotation are treated as a separate field. As the gantry rotates about the patient with the accelerator turned on, each of the small beams defined by the vanes is turned on/off according to the treatment plan. Varying the on/off period over the five to ten degrees controls the effective weight of that beam relative to all other beams; spatial modulation through temporal variable attenuation of the treatment beam results. This process continues in a single continuous helical arc (HA where the couch translates longitudinally while the accelerator rotates) or in a series of sequential independent slices (MIMIC where the
62
couch is indexed in between rotations of the gantry) until the entire target volume is treated. A typical IMRS treatment using a binary collimator will use the equivalent of 40–60 or more fixed fields of radiation to deliver the desired dose to the target volume. Since the dose is spread over a large region dose to uninvolved tissue can be kept very low with the high dose made to be extremely conformal to the target volume. Although the Peacock system using a MIMiC and the Tomotherapy Hi-Art system are considered technically to be IMRT and not IMRS system, they has the accuracy, resolution, and image guidance to be used in hypofractionated treatment paradigms. They therefore are more than ‘‘qualified’’ to be used to deliver IMRS treatments.
Multileaf Collimators Mini Multileaf Collimators (MMLC, called mini because the leafs define beams that are as small as 2.5 mm compared to a typical first generation multileaf collimator with 10 mm wide beams) are designed to work with a ‘‘conventional’’ linear accelerator and are offered by the large therapy companies such as Varian, Elekta, and Siemens, as well nonlinac companies such as Initia and Brainlab. They consist of as many as 60 pairs of radiation blocking tungsten leaves arranged sideby-side like the mating pairs of upper and lower teeth in the mouth. Unlike a row of teeth, however, that must move as a set, each leaf in a pair of leaves can move individually apart from its mate. As the pair of leaves move across the treatment field, the distance they are apart from each other varies, thus defining a variable size opening. The radiation beam passes through the leaf arrangement: Where the beam encounters a leaf it is blocked and where it encounters no leaf it passes through unhindered. By moving each pair of leaves at the same time, a computer-specified shape consisting of a regular or irregular area ‘‘open’’ to the beam
971
972
62
IMRT: technical and clinical aspects
surrounded by an area ‘‘closed’’ to the beam can be created. The amount of time the leaves remain in a given position determines the amount of radiation that falls on the area described by the opening. By changing the shape of the open area defined by the leaves a given pattern of radiation can be deposited in the patient. Traditionally, rather than using a rotationally delivery technique such as is employed with a binary collimator, MMLC IMRS delivers radiation with the gantry held still. With the accelerator in a fixed orientation relative to the patient and with the beam turned on the leaf pairs cycle through their patterns until the desired intensity
map is delivered. The gantry then rotates to the next field and the process is repeated. Typically 6–12 fields are used to deliver a treatment (> Figure 62-4). Recent developments in accelerator technology now actually allow rotational treatments to be delivered with MLCs and MMLCs using Varian and Elekta technology (based on original work first performed in the 1990s by Cedric Wu). Although the actual means used to achieve a rotational delivery is very different than that used with binary collimators, the effect is essentially the same – delivery of dose from a large number of fields.
. Figure 62-4 Delivering intensity map. For each field an Intensity Map is translated into a Relative Beam Strengths resulting in a dose distribution. (a) With a binary collimator (b) each window is opened for an appropriate percent of total time. With a physical compensator (c) the beam results from passing through a variable amount of attenuation. With the Cyberknife, the robot dwell time for each beam is adjusted. For a multileaf collimator, the dwell time for each aperture defined by the leaf pair is varied
IMRT: technical and clinical aspects
Cyberknife The CyberKnife Robotic Radiosurgery System (CK), manufactured by Accuracy, is not viewed by the majority of users nor by the manufacturer as an IMRS device but in fact it is exactly that. It scans a varying size circular beam of radiation across the target volume, varying the dwell time at each of a large number of dwell positions. Treatments delivered by the CK typically include several hundred radiation beams positioned within a large non-coplanar workspace surrounding the patient. Each of these beams can be directed from a unique location outside the patient to a unique position within the target volume. By varying the amount of time the robot aims the gun at one location, by changing the size of the beam, and by delivering radiation to any given location from a variety of angles, the desired amount of dose can be delivered to the target while limiting the amount of radiation that has to pass through any other tissue. Until recently the size of each radiation beam was determined by a circular collimator fitted manually to the treatment machine head. A set of 12 fixed collimators is provided as part of the system, corresponding to nominal field diameters ranging from 5 to 60 mm (field diameters are specified at a X-ray target to patient distance of 800 mm). The appropriate field size(s) needed to treat each patient is determined during treatment planning based on the target volume size and shape, its location relative to nearby organs at risk, and the desired dose distribution. Smaller collimators tend to maximize treatment conformality and dose gradients around the target volume; larger collimators tend to maximize the dose uniformity within the target volume while also minimizing the number of beams and total Monitor Units (MU). In practice, field size selection for any given case generally involves a compromise based on the relative importance of these clinical goals and usually consists of three collimators sizes intermingled.
62
Since there is only a single small beam that has to be repositioned many hundreds of times in order to deliver the multitude of beams that are used to deliver the IMRS treatment, the creation of very complex and refined dose distribution can be quite time consuming. A number of years ago Accuracy developed a robotic manipulator that made the task of changing the collimator easier to perform. Most recently, a significant reduction in the time required for a delivery has been achieved recently through an increase in the system’s dose rate and the use of an automated collimation system (Iris) that can change the size of the aperture on the fly during the treatment. The Iris Collimator contains twelve triangular collimator segments, oriented so as to define a dodecagon-shaped beam aperture. The twelve segments are divided into two banks of six that are mounted in series, with the two banks rotated by 30 relative to each other. Each segment is mounted on a linear bearing; rotational movement of the collimator mounting plate is thus converted into linear movement of the inner aperture surface of each segment. This design allows all twelve segments to be driven by a single motor. The aperture can be nearly completely closed (it is limited to 0.25 mm) or opened to a maximum size of 70 mm (projected at 800 mm distance). In practice the largest usable opening is constrained by the aperture of the primary collimator. Although the aperture size of the Iris Collimator is essentially continuously variable up to its maximum size 70 mm, its use in the CK is restricted currently to a set of twelve sizes corresponding to the sizes of the set of twelve fixed collimators, which range from 5 to 60 mm. The great precision of the robot, the ability to use beams of different sizes, and the freedom the robot has to deliver beams from virtually any direction, support the creation of very complex IMRS dose distributions with very sharp margins.
973
974
62
IMRT: technical and clinical aspects
Gamma Knife The new Gamma Knife Perfexion (GKP), like its predecessors, uses a large number of small beams of gamma radiation emitted from cobalt sources with all beams coinciding at a fixed focal point within the radiation unit. Each such ‘‘convergence,’’ delivered to a portion of the target volume, is called a shot, with a typical treatment requiring many such shots to completely cover the target volume. Unlike its predecessors, which required manual plugging of individual beams of radiation and manual changing of collimator size, GKP uses a total of 192 cobalt- 60 sources arranged in eight independently movable sectors mounted on the collimator body containing collimators of 4, 8, and 16 mm. Servo controlled motors in each sector mechanism positions the sources so as to align them with one of the collimator sizes for ‘‘Beam On’’ (aligning the sources with the collimators) or between collimators achieving ‘‘Beam Off.’’ Each sector, equivalent to a field of radiation, can have its beam size and dwell time varied independently. The delivered dose is shaped to
the precise contour of the target by combination of shots using different collimator sizes, plugging of sectors, and sector beam-on times. This adds significantly flexibility in beam shaping extending the range of delivery options when protection of surrounding critical structures becomes important (> Figure 62-5). As is the case with all IMRS systems, an inverse planning system is an integral part of Perfexion. The user defines targets and regions at risk and enters dose goals (for targets) and limits (for regions at risk). The optimization system determines the shot positions, the number of shots, and the sector beam size and dwell time for each shot required to meet those goals and limits. The plan is then delivered under computer control with automatic movement of patient for correct shot position as well helmet adjustment for correct sector beam size and dwell times. By virtue of its use of an inverse planning system, computerized control of delivery, and the ability to vary the intensity of various portions of a field independently of other portions of the field, Perfexion can be considered an IMRS
. Figure 62-5 Delivering intensity map with a gamma knife. Rather than individual beams being modulated sectors of beams are modulated
IMRT: technical and clinical aspects
system, although it differs greatly from the other system discussed in the way that it achieves a variable intensity map, in its source of X-rays, and in the scope of anatomical regions for which it can be used.
Conclusion Regardless of the mode of delivery, IMRS brings the best of RS and intensity modulation together to create the ultimate radiation delivery technology. As the field moves more and more toward hypofractionation, and as precision localization through IGRT becomes standard operating procedure, IMRS may become more the norm than the exception when it comes to the delivery of therapeutic radiation.
Selected References 1. Ammirati M, Bernardo A, Ramsinghani N, Yakoob R, Al-Ghazi M, Kuo J, Ammirati G. Stereotactic radiotherapy of central nervous system and head and neck lesions, using a conformal intensity-modulated radiotherapy system (Peacock). Skull Base 2001;11(2):109-19. 2. Carol M, Grant WH 3rd, Pavord D, Eddy P, Targovnik HS, Butler B, Woo S, Figura J, Onufrey V, Grossman R, Selkar R. Initial clinical experience with the Peacock intensity modulation of a 3-D conformal radiation therapy system. Stereotact Funct Neurosurg 1996; 66(1–3):30-4. 3. Carol MP. Integrated 3D conformal planning/multivane intensity modulating delivery system for radiotherapy. In: Purdy JA, Emami B, editors. 3D radiation treatment planning and conformal therapy. Madison, WI: Medical Physics Pyblishing; 1993. p. 435-45. 4. Clark B, McKenzie M, Robar J, Vollans E, Candish C, Toyota B, Lee A, Ma R, Goddard K, Erridge S. Does intensity modulation improve healthy tissue sparing in stereotactic radiosurgery of complex arteriovenous malformations? Med Dosim 2007;32(3):172-80. 5. Convery D, Rosenbloom M. The generation of intensitymodulated fields for conformal radiostherapy by dynamic collimation. Phys Med Biol 1992;37:1359-74. 6. Ernst-Stecken A, Lambrecht U, Ganslandt O, Mueller R, Fahlbusch R, Sauer R, Grabenbauer G. Radiosurgery of small skull-base lesions. No advantage for intensitymodulated stereotactic radiosurgery versus conformal arc technique. Strahlenther Onkol 2005;181(5):336-44.
62
7. Fenwick JD, Riley SW, Scott AJ. Advances in intensitymodulated radiotherapy delivery. Cancer Treat Res 2008;139:193-214. 8. Fenwick JD, Tome´ WA, Soisson ET, Mehta MP, Rock Mackie T. Tomotherapy and other innovative IMRT delivery systems. Semin Radiat Oncol 2006;16 (4):199-208. 9. Fuss M, Shi C, Papanikolaou N. Tomotherapeutic stereotactic body radiation therapy: Techniques and comparison between modalities. Acta Oncol 2006;45(7):953-60. 10. Fuss M, Salter BJ. Intensity-modulated radiosurgery: improving dose gradients and maximum dose using post inverse-optimization interactive dose shaping. Technol Cancer Res Treat 2007;6(3):197-204. 11. Fuss M, Salter BJ, Caron JL, Vollmer DG, Herman TS. Intensity-modulated radiosurgery for childhood arteriovenous malformations. Acta Neurochir (Wien) 2005;147 (11):1141-9; discussion:1149–50. 12. Gibbs IC. Frameless image-guided intracranial and extracranial radiosurgery using the cyberknife robotic system. Cancer Radiother 2006;10(5):283-7. 13. Hara W, Soltys SG, Gibbs IC. CyberKnife robotic radiosurgery system for tumor treatment. Expert Rev Anticancer Ther 2007;7(11):1507-15. 14. Holmes TW, Hudes R, Dziuba S, Kazi A, Hall M, Dawson D. Stereotactic image-guided intensity modulated radiotherapy using the HI-ART II helical tomotherapy system. Med Dosim 2008;33(2):135-48. 15. Jin JY, Yin FF, Ryu S, Ajlouni M, Kim JH. Dosimetric study using different leaf-width MLCs for treatment planning of dynamic conformal arcs and intensity-modulated radiosurgery. Med Phys 2005;32 (2):405-11. 16. Lindquist C, Paddick I. The Leksell Gamma Knife Perfexion and comparisons with its predecessors. Neurosurgery 2007;61 3 Suppl:130-40; discussion:140–1. 17. Meyer JL, Verhey L, Xia P, Wong J. New technologies in the radiotherapy clinic. Front Radiat Ther Oncol 2007;40:1-17. 18. Qi XS, Schultz CJ, Li XA. Possible fractionated regimens for image-guided intensity-modulated radiation therapy of large arteriovenous malformations. Phys Med Biol 2007;52(18):5667-82. 19. Sharma SC, Ott JT, Williams JB, Dickow D. Commissioning and acceptance testing of a cyberKnife linear accelerator. J Appl Clin Med Phys 2007;8(3):2473. 20. Steffey-Stacy EC. Frameless, image-guided stereotactic radiosurgery. Semin Oncol Nurs 2006;22(4):221-32. 21. Tobler M, Leavitt DD, Watson G. Optimization of the primary collimator settings for fractionated IMRT stereotactic radiotherapy. Med Dosim 2004; 29(2):72-9. 22. Wang SJ, Choi M, Fuller CD, Salter BJ, Fuss M. Intensity-modulated radiosurgery for patients with brain metastases: a mature outcomes analysis. Technol Cancer Res Treat 2007;6(3):161-8.
975
976
62
IMRT: technical and clinical aspects
23. Woo SY, Grant WH, III, Bellezza D, Grossman R, Gildenberg P, Carpentar LS, Carol M, Butler EB. A comparison of intensity modulated conformal therapy with a conventional external beam stereotactic radiosurgery system for the treatment of single and multiple intracranial lesions. Int J Radiat Oncol Biol Phys 1996;35(3):593-7. 24. Yu C. Intensity-modulated arc therapy with dynamic multileaf collimation: an alternative to tomotherapy. Phys Med Biol 1995;40:1435-49.
25. Zytkovicz A, Daftari I, Phillips TL, Chuang CF, Verhey L, Petti PL. Peripheral dose in ocular treatments with CyberKnife and Gamma Knife radiosurgery compared to proton radiotherapy. Phys Med Biol 2007;52 (19):5957-71.
59 Linac Radiosurgery W. A. Friedman . F. J. Bova
Introduction Stereotactic radiosurgery (SRS) is a minimally invasive treatment modality that delivers a large single dose of radiation to a specific intracranial target while sparing surrounding tissue. Unlike conventional fractionated radiotherapy, SRS does not maximally exploit the higher radio sensitivity of brain lesions relative to normal brain (therapeutic ratio). Its selective destruction is dependent mainly on sharply focused high-dose radiation and a steep dose gradient away from the defined target. The biological effect is irreparable cellular damage (probably via DNA strand breaks) and delayed vascular occlusion within the high-dose target volume. Because a therapeutic ratio is not required, traditionally radio resistant lesions can be treated. Since destructive doses are used, however, any normal structure included in the target volume is subject to damage. The basis for SRS was conceived over 40 years ago by Lars Leksell [1]. He proposed the technique of focusing multiple beams of external radiation on a stereotactically defined intracranial target. The averaging of these intersecting beams results in very high doses of radiation to the target volume, but innocuously low doses to non-target tissues along the path of any given beam. His team’s implementation of this concept culminated in the development of the Gamma Knife. The modern Gamma Knife employs 201 fixed cobalt radiation sources in a fixed hemispherical array, such that all 201 photon beams are focused on a single point. The patient is stereotactically positioned in the Gamma Knife so that the intracranial target coincides with the isocenter of radiation. Using #
Springer-Verlag Berlin/Heidelberg 2009
variable collimation, beam blocking, and multiple isocenters, the radiation target volume is shaped to conform to the intracranial target. An alternate radiosurgical solution using a linear accelerator (LINAC) was first described in 1984 by Betti et al. [2]. Colombo et al. described such a system in 1985 [3], and LINACs have subsequently been modified in various ways to achieve the precision and accuracy required for radiosurgical applications [4–7]. In 1986, a team composed of neurosurgeons, radiation physicists and computer programmers began development of the University of Florida LINAC-based radiosurgery system [8]. This system has been used to treat over 2,800 patients at the University of Florida since May 1988, and is in use at multiple sites worldwide. Many other commercial versions of radiosurgical systems are currently available, including the Brain Lab system, the Radionics (X-knife system), the Accuray (Cyberknife system), and others. Most LINAC radiosurgical systems rely on the same basic paradigm: A collimated x-ray beam is focused on a stereotactically identified intracranial target. The gantry of the LINAC rotates around the patient, producing an arc of radiation focused on the target (see > Figures 59-1 and > 59-2). The patient couch is then rotated in the horizontal plane and another arc performed. In this manner, multiple non-coplanar arcs of radiation intersect at the target volume and produce a high target dose, with minimal radiation to surrounding brain. This dose concentration method is exactly analogous to the multiple intersecting beams of cobalt radiation in the gamma knife. The target dose distribution can be tailored by varying collimator sizes, eliminating
930
59
Linac radiosurgery
. Figure 59-1 Linear accelerators are the preferred device, worldwide, for conventional radiotherapy. They accelerate electrons to near light speed, then collide them with a heavy metal (like tungsten) in the head of the machine. The collision mainly produces heat but a small percentage of the energy is converted into highly energetic photons. These photons, because they are electronically produced, are called ‘‘x-rays.’’ The x-radiation is collimated and focused on the target
undesirable arcs, manipulating arc angles, using multiple isocenters, and differentially weighting the isocenters [9]. In our center, multiple isocenters are used to achieve highly conformal dose distributions, exactly analogous to the gamma knife technique (> Figure 59-3). Some linear accelerator systems use an alternative approach which relies upon a computer driven multileaf collimator to generate non-spherical beam shapes which are conformal to the beam’s eye view of the tumor. The multileaf collimator can be adjusted statically or dynamically as the linear accelerator rotates. Intensity modulation can be used to achieve dose distributions which are
close to those seen with multiple isocenters and treatment time can be reduced. For a complete discussion of alternative techniques for linear accelerator dose planning, the reader is referred to Frank Bova’s chapter on the Technical Aspects of LINAC Radiosurgery. Achievable dose distributions are similar for LINAC-based and gamma knife systems. With both systems, it is possible to achieve dose distributions that conform closely to the shape of the intracranial target, thus sparing the maximum amount of normal brain. Recent advances in stereotactic imaging and computer technology for dose planning, as well as refinements in radiation delivery systems have
Linac radiosurgery
. Figure 59-2 This diagram shows an add on device, designed to improve the accuracy of the linear accelerator, in place. The LINAC arcs around the patient, with its beam always focused on the stereotactically positioned target. The patient is then moved to a new horizontal (table) position, and another arc performed. The result is multiple, non-coplanar arcs of radiation, all converging on the target point
59
led to improved efficacy, fewer complications and a remarkable amount of interest in the various applications of SRS. Perhaps of equal importance is the fact that increasing amounts of scientific evidence have persuaded the majority of the international neurosurgical community that radiosurgery is a viable treatment option for selected patients suffering from a variety of challenging neurosurgical disorders. This chapter will present a brief description of LINAC radiosurgical technique, followed by a review of the more common applications of stereotactic radiosurgery in the treatment of intracranial disease.
LINAC Radiosurgery Technique
. Figure 59-3 This choroidal fissure arteriovenous malformation required four, 1cm isocenters to produce a conformal plan. The inner line (70% isodose) is the prescription dose line. The outer line (35% isodose) is half of the prescription dose
Although the details of radiosurgical treatment techniques differ somewhat from system to system, the basic paradigm is quite similar everywhere. Below, is a brief description of a typical radiosurgical treatment at the University of Florida. Almost all radiosurgical procedures in adults are performed on an outpatient basis. The patient reports to the neurosurgical clinic the day before treatment for a detailed history and physical, as well as an in depth review of the treatment options. If radiosurgery is deemed appropriate, the patient is sent to the Radiology department for a volumetric MRI scan. A radiosurgical plan can be generated, in advance, using this MRI study. The next morning, the patient arrives at 7:00 AM. A stereotactic head ring is applied under local anesthesia. No skin shaving or preparation is required. Subsequently, stereotactic CT scanning is performed. One millimeter slices are obtained throughout the entire head. The patient is then transported to an outpatient holding area where he and his family have breakfast and relax until the treatment planning process is complete. The stereotactic CT scan and the non-stereotactic volumetric MRI scan are transferred via Ethernet to the treatment-planning computer. The CT images are quickly processed
931
932
59
Linac radiosurgery
so that each pixel has an anteroposterior, lateral, and vertical stereotactic coordinate, all related to the head ring previously applied to the patient’s head. Using image fusion software, the nonstereotactic MRI is fused, pixel for pixel, with the stereotactic CT. The ‘‘pre-plan’’ performed the day before is, likewise, fused to the stereotactic CT. Final dosimetry then begins and continues until the neurosurgeon, radiation therapist, and radiation physicist are satisfied that an optimal dose plan has been developed. A variety of options are available for optimizing the dosimetry. The fundamental goal is to deliver a radiation field that is precisely conformal to the lesion shape (see > Figure 59-3), while delivering a minimal dose of radiation to all surrounding neural structures. When dose planning is complete, the radiosurgical device is attached to the LINAC. The patient then is attached to the device and treated. It takes approximately 10 min to treat each isocenter. The head ring is removed and, after a short observation period, the patient is discharged. The radiosurgical device is disconnected from the LINAC, which is then ready for conventional usage. Close clinical and radiologic follow-up is arranged at appropriate intervals depending on the pathology treated and the condition of the patient.
Radiosurgery for Benign Tumors SRS has proven useful for the treatment of a variety of benign intracranial neoplasms. These tumors commonly arise from the skull base, where their dramatic impact on quality of life belies their benign histology and small size. Despite progressive improvement in microsurgical techniques, outcomes for patients with these difficult tumors continue to be less than optimal [10–12]. A significant amount of experience has been accumulated using SRS in the treatment of schwannomas and meningiomas. We will focus on each of these tumor types in turn.
Vestibular Schwannomas Among benign intracranial tumors, vestibular schwannoma (acoustic neuroma) has to date been the most frequent target for stereotactic radiosurgery. This common tumor (representing approximately 10% of all primary brain tumors) is a benign proliferation of Schwann cells arising from the myelin sheath of the vestibular branches of the eighth cranial nerve. These tumors are slightly more common in women, present at an average age of 50 years, and occur bilaterally in patients with neurofibromatosis type II. Leksell first used stereotactic radiosurgery to treat a vestibular schwannoma in 1969 [13]. SRS is a logical alternative treatment modality for this tumor for several reasons. A vestibular schwannoma is typically well demarcated from surrounding tissues on neuroimaging studies. The sharp borders of this non-invasive tumor make it a convenient match for the characteristically steep dose gradient produced at the boundary of a radiosurgical target. This allows the radiosurgeon to minimize radiation of normal tissue. Excellent spatial resolution on gadolinium-enhanced MRI facilitates radiosurgical dose planning. These tumors typically occur in an older population that may be less fit for microsurgical resection under general anesthesia. Finally, the location of these tumors at the skull base in close proximity to multiple critical neurologic structures (i.e., cranial nerves, brainstem) leads to appreciable surgical morbidity and rare mortality even in expert hands. This makes the concept of an effective, less invasive, less morbid alternative treatment that can be performed in a single day under local anesthesia quite attractive. Certainly, the role of radiosurgery is limited by its inability to expeditiously relieve mass effect in patients for whom this is necessary. The radiobiology of SRS also requires lower, potentially less effective doses for higher target volumes in order to avoid complications. This limits the use of SRS to the treatment of smaller tumors.
Linac radiosurgery
Despite these limitations, there is a growing body of literature that substantiates the claim that radiosurgery is a safe and effective alternative therapy for acoustic schwannomas. The published experience using LINACbased radiosurgery for the treatment of vestibular schwannomas is relatively limited compared to the gamma knife literature. Foote et al. [14] performed an analysis of risk factors associated with radiosurgery for vestibular schwannoma at UF. The aim of this study was to identify factors associated with delayed cranial neuropathy following radiosurgery for vestibular schwannoma (VS) and to determine how such factors may be manipulated to minimize the incidence of radiosurgical complications while maintaining high rates of tumor control. From July 1988 to June 1998, 149 cases of VS were treated using linear accelerator radiosurgery at the University of Florida. In each of these cases, the patient’s tumor and brainstem were contoured in 1-mm slices on the original radiosurgical targeting images. Resulting tumor and brainstem volumes were coupled with the original radiosurgery plans to generate dose-volume histograms. Various tumor dimensions were also measured to estimate the length of cranial nerve that would be irradiated. Patient follow-up data, including evidence of cranial neuropathy and radiographic tumor control, were obtained from a prospectively maintained, computerized database. The authors performed statistical analyses to compare the incidence of post treatment cranial neuropathies or tumor growth between patient strata defined by risk factors of interest. One hundred thirty-nine of the one hundred forty-nine patients were included in the analysis of complications. The median duration of clinical follow-up for this group was 36 months (range 18–94 months). The tumor control analysis included 133 patients. The median duration of radiological follow-up in this group was 34 months (range 6–94 months). The overall 2-year actuarial incidences of facial and trigeminal neuropathies were 11.8 and 9.5%, respectively.
59
In 41 patients treated before 1994, the incidences of facial and trigeminal neuropathies were both 29%, but in the 108 patients treated since January 1994, these rates declined to 5 and 2%, respectively. An evaluation of multiple risk factor models showed that maximum radiation dose to the brainstem, treatment era (pre-1994 compared with 1994 or later), and prior surgical resection were all simultaneously informative predictors of cranial neuropathy risk. The radiation dose prescribed to the tumor margin could be substituted for the maximum dose to the brainstem with a small loss in predictive strength (> Figure 59-3). The overall radiological tumor control rate (> Figure 59-4) was 93% (59% tumors regressed, 34% remained stable, and 7.5% enlarged), and the 5-year actuarial tumor control rate was 87% (95% confidence interval [CI] 76–98%). Based on this study, the authors currently recommend a peripheral dose of 12.5 Gy for almost all acoustics, as that dose most likely to yield long-term tumor control without causing cranial neuropathy. Friedman et al. recently updated the University of Florida experience [15]. Between July 1988 and August 2005, 390 patients with VSs were . Figure 59-4 Pre-treatment MRI scan shows left sided vestibular schwannoma
933
934
59
Linac radiosurgery
treated. One- and two-year actuarial control rates were both 98%, and the 5-year actuarial control rate was 90%. Only four patients (1%) required surgery for tumor growth. Seventeen patients (4.4%) reported facial weakness and 14 patients (3.6%) reported facial numbness after radiosurgery. The risk of these complications rose with increasing tumor volume or increasing radiosurgical dose to the tumor periphery. Since 1994, when doses were deliberately lowered to 1,250 cGy, only two patients (0.7%) have experienced facial weakness and two (0.7%) have experienced facial numbness. Beegle and Friedman [16] specifically studied the effect of treatment plan ‘‘quality’’ on outcome in the same patient population. In this study the authors looked at dosimetry variables: conformity of treatment plan and steepness of dose gradient, in the same group of patients. Over the duration of this study, dosimetry evolved from a single isocenter with marginal conformity to multiple isocenters with high conformity. Multivariate statistics were used to determine the effects of these variables on tumor control and on two types of complications, facial weakness and facial numbness. Dosimetry, surprisingly had no effect on tumor control or complications. The overwhelmingly important variable was dose, not dosimetry quality. Vachani and Friedman [17] also looked at radiosurgery in patients with bilateral vestibular schwannomas. Patients with bilateral ves opportunity to determine the effectiveness of radiosurgery. By using the untreated tumor as an internal control, one can determine whether radiosurgery was able to interrupt the natural history of the treated tumor. From September 1998 to November 2004, 13 patients with neurofibromatosis type II had 14 tumors treated with radiosurgery at the University of Florida. A retrospective analysis was performed on these patients. The average follow-up length was 38 months. One patient failed to send a follow-up MRI. Actuarial local control in the treated tumors was 100% at
1 year and 92% at 2 and 5 years. Only one of the treated tumors continued to grow. In the untreated tumors, actuarial local control was 100% at 1 year, 78% at 2 years, and 21% at 5 years. None of the untreated tumors decreased in size. This study showed that radiosurgery alters the natural history of vestibular schwannomas (> Figure 59-5). Spiegelmann et al. [18,19] have reported their experience. They reviewed the methods and results of linear accelerator (LINAC) radiosurgery in 44 patients with acoustic neuromas who were treated between 1993 and 1997. Computerized tomography scanning was selected as the stereotactic imaging modality for target definition. A single, conformally shaped isocenter was used in the treatment of 40 patients; two or three isocenters were used in four patients who harbored very irregular tumors. The radiation dose directed to the tumor border was the only parameter that changed during the study period: in the first 24 patients who were treated the dose was 15–20 Gy, whereas in the last 20 patients the dose was reduced to 11–14 Gy. After a mean follow-up period of 32 months (range 12–60 . Figure 59-5 Four years post-treatment, the MRI scan shows the schwannoma to be much smaller
Linac radiosurgery
months), 98% of the tumors were controlled. The actuarial hearing preservation rate was 71%. New transient facial neuropathy developed in 24% of the patients and persisted to a mild degree in 8%. Radiation dose correlated significantly with the incidence of cranial neuropathy, particularly in large tumors (4 cm3). Several reports on smaller series of patients treated with linear accelerator-based radiosurgery for vestibular schwannomas have been published in recent years. Martens et al. reported on 14 patients with at least 1 year of follow-up after radiosurgery on the LINAC unit in the University Hospital in Ghent, Belgium [20]. A mean marginal dose of 19.4 Gy (range 16–20 Gy) was delivered to the 70% isodose line with a single isocenter. Mean follow-up duration was 19 months (range 12–24 months.). During this relatively short follow-up interval, 100% radiographic tumor control has been achieved (29% regressed, 71% stable, zero enlarged). Rates of delayed facial and trigeminal neuropathy were 21% and 14%, respectively and two of three facial nerve deficits resolved. Preoperative hearing was preserved 50% of the time. Valentino and Raimondi reported on 23 patients treated with LINAC radiosurgery in undergone previous surgery. Total radiation dose to the tumor margin ranged from 12 to 45 Gy (median 30 Gy) and was delivered in 1–5 sessions. One or two isocenters were used and mean duration of follow-up was 40 months (range 24–46 months.). Results using this less conventional method of multi-session radiosurgery were comparable to other radiosurgical techniques. Tumor control was achieved in 96% of patients (38% regressed, 58% stable, 4% enlarged), facial and trigeminal neuropathies each occurred at a rate of 4%, and ‘‘hearing was preserved at almost the same level as that prior to radiosurgery in all patients.’’ Ishihara [22] discussed 38 patients treated with Cyberknife radiosurgery. The treatment volumes of these two groups were 0.5–24.0 cm3
59
(mean 4.7 cm3), and 0.5–41.6 cm3 (mean 8.2 cm3). Target irradiation was administered in 1–3 fractions (mean 2.5 fractions). The total marginal radiation doses were 15.0–20.5 Gy (mean 17.0 Gy), and 11.9–20.1 Gy (mean 16.9 Gy), respectively. After a mean follow-up period of 31.9 months (range 12–59 months, median 27 months), 94% of the tumors were controlled. Only one patient in the group with non-serviceable hearing underwent additional surgical resection for a presumed increase in tumor size. The hearing preservation rate was 93%. Facial weakness did not develop in any of the patients in the serviceable hearing group. New trigeminal symptoms did not develop in any patients in either group. The use of LINAC radiosurgery for acoustics is briefly discussed in reports by Delaney[23] and Barcia [24]. As of December 2008, the University of Florida experience with vestibular schwannomas comprised 448 patients.
Meningiomas Meningiomas are the most common benign primary brain tumor, with an incidence of approximately 7/100,000 in the general population. Surgery has long been thought to be the treatment of choice for symptomatic lesions and is often curative. Many meningiomas, however, occur in locations where attempted surgical cure may be associated with morbidity or mortality, such as the cavernous sinus or petroclival region [25,26]. In addition, many of these tumors occur in the elderly, where the risks of general anesthesia and surgery are known to be increased. Hence there is interest in alternative treatments, including radiation therapy and radiosurgery, either as a primary or adjuvant approach. Simpson, in a classic paper, described the relationship between completeness of surgical resection and tumor recurrence [27]. A grade I resection, that is complete tumor removal with
935
936
59
Linac radiosurgery
excision of the tumor’s dural attachment and involved bone, has a 10% recurrence rate. A grade II resection, complete resection of the tumor and coagulation of its dural attachment, has up to a 20% recurrence rate. Grade III resection is complete tumor removal without dural resection or coagulation. Grade IV resection is subtotal and Grade V resection is simple decompression. Recurrence rates in the Grade IV and V groups basically reflect the natural history of the tumor, with high rates of recurrence over time. Unfortunately, some common meningioma locations, such as the cavernous sinus or petroclival region are not readily amenable to a complete dural resection or coagulation strategy because of location and the proximity of vital neural and vascular structures. In addition, relatively high complication rates have been described for meningioma surgery in some locations and in the elderly. Pollock and colleagues recently analyzed 198 patients with meningiomas less than 35 mm in diameter, treated with either surgical resection or gamma knife radiosurgery [28]. Tumor recurrence was more frequent in the surgical resection group (12% vs. 2%). No statistically significant difference was detected in the 3 and 7 year actuarial progression-free survival rate between patients with Simpson Grade 1 resections and those who underwent radiosurgery. Progression Complications were lower in the radiosurgery group. Multiple linear accelerator radiosurgical series have been published [29–34]. Hakim and colleagues described the largest such series, and the only one of the group to report actuarial statistics [35]. One hundred twenty-seven patients with one hundred fifty-five meningiomas were treated. Actuarial tumor control for patients with benign tumors was 89.3% at 5 years. Six patients (4.7%) had permanent radiation induced complications. The University of Florida report on linear accelerator radiosurgery treatment of meningiomas is the largest yet published [36]. Two hundred ten patients were treated from May 1989 through December 2001.
All patients had follow-up for a minimum of 2 years and no patients were lost to follow-up. Actuarial local control for benign tumors was 100% at 1 and 2 years, and 96% at 5 years (see > Figures 59-6 and > 59-7). Actuarial local control for atypical tumors was 100% at 1 year, 92% at 2 years, and 77% at 5 years. Actual control for malignant tumors was 100% at 1 and 2 years, but only 19% at 5 years. Permanent radiation induced complications occurred in 3.8%, all of which involved malignant tumors. These tumor control and treatment morbidity rates compare well with all other published series. We found that reliance on imaging characteristics rather than surgical pathology did not yield a high incidence of missed diagnoses. During the time interval of this study, only two patients were treated as presumed meningiomas and later found to have other diagnoses. One had a dural-based metastasis which was surgically excised when it enlarged. The other had a hemangiopericytoma of the lateral cavernous sinus which was surgically excised when it enlarged.
. Figure 59-6 MRI scan shows right cavernous sinus meningioma. The patient presented with a sixth nerve paresis
Linac radiosurgery
. Figure 59-7 Three years later, the meningioma is barely visible. The sixth nerve paresis completely resolved. We believe that radiosurgery is the treatment of choice for many cavernous sinus meningiomas
Radiosurgery for Malignant Tumors Malignant tumors are radiobiologically more amenable to fractionated radiotherapy than benign lesions. Malignancies tend to infiltrate surrounding brain, resulting in poorly definable tumor margins. A priori, these two traits of cerebral malignancies would seem to make radiosurgery an unattractive treatment option. Nevertheless, SRS has proven to be a useful weapon in the armamentarium against malignant brain tumors. The most common applications of SRS to malignant tumors are the treatment of cerebral metastases and the delivery of an adjuvant focal radiation ‘‘boost’’ to malignant gliomas.
Cerebral Metastases Metastatic brain tumors are up to ten times more common than primary brain tumors with an annual incidence of between 80,000 and
59
150,000 new cases each year [37]. Fifteen to forty percent of cancer patients will be diagnosed with a brain metastasis during the course of their illness. Once a brain metastasis has been diagnosed, the median life expectancy is less than 1 year. However, in many patients, aggressive treatment of metastatic disease has been shown to restore neurologic function and prevent further neurological manifestations. Debate exists concerning the optimum treatment for metastatic brain disease. In autopsy series, brain metastases occur in up to 50% of cancer patients [38]. Approximately 30–40% present with a solitary metastasis. Brain metastases frequently cause debilitating symptoms which can seriously impact the patient’s quality of life. With no treatment or steroid therapy alone, survival is very limited (1–2 months). Whole brain radiotherapy extends median survival, but the duration of survival is typically low (3–4 months). Several randomized trials have suggested that, when possible, surgery followed by WBRT is superior to WBRT alone. Patchell et al. reported a randomized clinical trial involving 46 patients with a single metastasis and well-controlled systemic disease [39]. They found a significant improvement in survival (40 vs. 15 weeks) and local recurrences in the CNS (20% vs. 52%) for patients in the surgery plus whole brain radiotherapy arm of the study. Likewise, Noordijk et al. randomized 66 patients and found a significant survival advantage (10 vs. 6 months) for the combination therapy arm [40]. In contrast, Mintz et al. studied a group of 84 patients and did not show an advantage of surgery plus radiotherapy over radiotherapy alone [41]. It has been suggested that the inclusion of a higher percentage of patients with active systemic disease and lower performance scores did not allow the benefit of improved local control to affect survival in this series. Haines points out that survival and quality of life are the most important outcomes measures in evaluating a clinical treatment for cancer [42]. Surrogate
937
938
59
Linac radiosurgery
endpoints, like local control, are inherently unreliable, especially when the definition of local control is changed. This applies to a comparison of radiosurgery with surgery for brain metastasis. In surgical series, local control means no visible tumor on follow-up scans. In radiosurgical series, local control means no growth (or sometime minimal growth) on follow-up scans. These endpoints are unlikely to be equivalent. In addition, comparison of current results to historical controls is fraught with hazard to selection bias. This issue led to erroneous conclusions about the efficacy of brachytherapy for malignant gliomas and to overly optimistic reports regarding the efficacy of intra-arterial chemotherapy. Of equal import is the difficulty and variability of reporting standards for local control. Few series provide actuarial local control. They simply provide a ‘‘raw’’ number at an arbitrary point in time. Less commonly appreciated is the difficulty in documenting local control. Many of these patients die away from the medical center where radiosurgery was performed. It is frequently impossible to determine from family or local physician telephone interview, whether the proximate cause of death was loss of local control, new intracranial disease (loss of regional control), or systemic disease. Most radiosurgical series have assumed that, unless an MRI was performed documenting local loss of local control prior to death, local control was maintained. This assumption may lead to a systematic overestimation of local control rates. Sturm [43–45], Black [46,47], and Adler [48–50] published early reports on linear accelerator radiosurgery for brain metastases. Alexander [47] reported on 248 patients. Median tumor volume was 3 cc and median tumor dose was 15 Gy. Median survival was 9.4 months. Actuarial local control was 85% at 1 year and 65% at 2 years. Auchter et al. reported a multiinstitutional study of 122 patients [51]. Actuarial 1 and 2 year survivals were 53% and 30% respectively. Local control was 86%. Many other LINAC series have been reported [45,52–59].
As radiosurgery has emerged as a treatment option, clinicians have attempted to define prognostic factors, which may help to define patient populations most likely to benefit from radiosurgical treatment [60–62]. Multiple factors have been discerned from retrospective analysis and include: Karnofsky performance scale score, status of systemic disease, histology, number of metastases, volume of metastases, time interval between the diagnosis of the primary lesion and the metastatic lesion, pattern of enhancement [63,64], the RTOG recursive partitioning categories [65], and radiation dose. The University of Florida published their early experience with radiosurgery for brain metastases [66] in 2004. Three hundred eighty three patients were treated. Median survival was 9 months. Melanoma histology and increasing number of metastases predicted poorer survival. Increasing age, somewhat surprisingly, slightly improved survival, possibly because younger patients tended to have more radioresistant histologies. Actuarial local control was 75% (> Figures 59-8 and > 59-9). Increasing dose provided better control and eloquent location was also associated with better control (possibly because eloquent tumors tended to be discovered at a smaller size). Regional control was poorer in melanoma or breast patients and in those with synchronous presentation of brain metastasis and primary tumor. In this retrospective analysis, whole brain radiotherapy did not improve regional control. Andrews et al. [67] reported the phase III results of RTOG protocol 9508, the first prospective, randomized trial of WBRT with or without SRS for the treatment of metastatic brain disease to reach full accrual. Three hundred thirtythree patients with 1–3 newly diagnosed brain metastases, KPS 70, and no history of WBRT were randomly allocated either WBRT alone (167 patients) or WBRT followed by an SRS boost (164 patients). Analysis revealed a significant survival advantage for patients with a single brain
Linac radiosurgery
59
. Figure 59-8 The patient with known breast carcinoma presented with a symptomatic pontine lesions. She was treated with radiosurgery (15 Gy to the 80% isodose line)
. Figure 59-9 Three years later, the site of the lesion was barely visible
metastasis treated with WBRT plus SRS rather than WBRT alone (median survival 6.5 months vs. 4.9 months), but failed to reveal a similar advantage for patients with 2–3 brain metastases.
However, all patients in the WBRT plus SRS group were significantly more likely to have a stable or improved KPS six months post-treatment, and were less likely to be dependent on corticosteroids, than those treated with WBRT alone. Multivariate analysis revealed RPA Class I and non-small cell lung primary to be significantly associated with improved survival. Analysis also revealed significantly better 1-year local control of lesions treated with WBRT plus SRS, compared to those treated with WBRT alone (82% vs. 71%). The authors concluded that the addition of SRS to WBRT should be standard treatment for patients with a single brain metastasis and considered for patients with 2–3 brain metastases. Aoyama et al. [68] recently reported results of the first prospective, multi-institutional, randomized trial of SRS with or without WBRT for the treatment of metastatic brain disease. One hundred thirty-two patients with 1–4 brain metastases 3 cm in diameter and KPS 70 were randomly allocated either SRS plus WBRT (65 patients) or SRS alone (67 patients).
939
940
59
Linac radiosurgery
Adjuvant WBRT did not affect survival; patients who received both SRS and WBRT had a median survival of 7.5 months and a 1-year actuarial survival rate of 38.5%, compared to 8.0 months and 28.4% for those treated with SRS alone. However, patients who did not initially receive WBRT underwent significantly more salvage procedures. Multivariate analysis revealed age <65, controlled primary tumor, stable systemic disease, and KPS 90 to be significantly associated with improved survival. Analysis revealed significantly better local, regional, and total brain control following treatment with SRS plus WBRT rather than SRS alone. Patients treated with SRS plus WBRT had one-year actuarial local control, regional control, and total brain control rates of 88.7, 58.5, and 53.2%, respectively, compared with 72.5, 36.3, and 23.6%, respectively, for patients treated with SRS alone. Multivariate analysis also revealed stable systemic disease and KPS 80 to be significantly associated with improved regional control, and a single brain metastasis approached significance (P = 0.06). Analysis revealed no difference in post-treatment neurologic performance; patients treated with SRS plus WBRT had a 1-year actuarial neurologic preservation rate of 72.1%, compared with 70.3% for patients treated with SRS alone. Swinson and Friedman have updated the UF experience with this disease [69]. We performed a retrospective analysis of 619 patients who underwent linear accelerator-based stereotactic radiosurgery for 1,569 brain metastases between May 1989 and February 2006. Patient characteristics and treatment parameters were obtained prospectively. Patients were followed-up at regular intervals clinically and with imaging studies to document local control, regional control, and survival. Median actuarial survival was 7.9 months. One- and two-year actuarial survival probabilities were 0.36 and 0.14, respectively. Radiation Therapy Oncology Group Recursive Partitioning Analysis (RPA) Class I or II was associated with improved survival, but the difference between the two was insignificant.
Female sex, younger age, higher Karnofsky performance status, controlled primary tumor, absence of systemic metastases, asynchronous presentation of brain metastasis, fewer brain metastases, smaller total volume of brain metastases, surgery prior to radiosurgery, and multiple radiosurgical treatments were also associated with improved survival. Melanoma metastasis was associated with impaired survival. Local control was achieved in 84.3% of all lesions treated. One- and two-year actuarial local control probabilities were 0.82 and 0.72, respectively. Whole brain radiation therapy prior to radiosurgery was associated with improved regional control.
Malignant Gliomas Current conventional treatment for malignant gliomas involves a combination of surgery, radiation and temozolomide chemotherapy. The prognosis in these patients remains poor [38]. The majority of recurrences occur within 2 cm of the enhancing lesion as seen on initial imaging. Gross total excision may be associated with prolonged median survival in patients with malignant gliomas. Some studies have shown that other aggressive local therapies, such as interstitial brachytherapy, may favorably impact survival [70–72]. Radiosurgery is another attempt at forestalling local recurrence by aggressive local therapy. A number of linear accelerator radiosurgery series have been published. Shrieve and colleagues reported on 32 patients receiving interstitial brachytherapy and 86 patients receiving radiosurgical boost [73]. The found similar survival rates between the two groups and recommended radiosurgery because of its outpatient, non-invasive nature. Hall and colleagues reported 35 patients and felt that radiosurgery did confer a survival advantage, with fewer complications than brachytherapy [74]. Buatti et al., at the University of Florida, reported on 11 patients treated with radiosurgical boost [75]. No significant survival advantage was found.
Linac radiosurgery
Likewise, Masciopinto and colleagues [76] reported on 31 patients so treated and found that the ‘‘curative value of radiosurgery is significantly limited by peripheral recurrence.’’ Other studies include those of Regine [77] Prisco [78], and Gannett [79]. A recurring theme in all retrospective studies of brain tumor therapies is the question of selection bias influencing the results of therapy more than the therapy itself. In an attempt to control for selection bias in retrospective treatment trials for malignant gliomas [80], Curran [81] developed the recursive partitioning analysis approach. Sarkaria and colleagues used this methodology to analyze 115 patients from three institutions treated with linear accelerator radiosurgery [82]. They found that patients treated with radiosurgery had a significantly improved 2-year and median survival compared to RTOG historical controls. The improvement was seen predominately in the worst prognostic classes [3–6]. Kondziolka performed a similar analysis on 65 patients who underwent upfront radiosurgery [83]. He also found that patients in RTOG classes 3–5 appeared to benefit. At the University of Florida, we retrospectively reviewed one hundred patients with WHO grade III and IV malignant gliomas who received SRS boost therapy for residual or recurrent enhancing disease [84]. The patients in our study were divided into RPA classifications for comparison to historical controls. Class III and IV patients had median survival times very similar to the historical controls. Class V patients demonstrated an increase in median survival (15.6 vs. 8.9 months) and two year survival rate (12.5% vs. 6%) compared to historical controls. Eloquent location correlated with poorer survival. This may be due to the selection of less aggressive therapies for this group of patients. Recurrence at time of radiosurgery was associated with longer survival. Very probably this reflects the fact that patients judged ‘‘eligible’’ for radiosurgery at time of recurrence are already selected for longer survival than the average patient treated upfront. However, it remains possible that radiosurgery at time of
59
recurrence is truly more effective than upfront radiosurgery. What about drawbacks of the recursive partitioning technique? The RTOG classes used are broad and do not include all known prognostic variables, most notably tumor size. In addition, important linear variables like age, mental status, and KPS are converted into binary ones. This approach, therefore, is flawed, as are all attempts at retrospective analysis. Irish and colleagues, in an analysis of 101 consecutive malignant glioma patients, have shown that those ‘‘eligible’’ for radiosurgery have a median survival of 23.4 months, compared with 8.6 months for ‘‘ineligible’’ patients [85]. Likewise, Curran found a marked survival advantage in radiosurgery ‘‘eligible’’ versus ‘‘ineligible’’ patients [86]. The only complete solution to the issue of selection bias affecting outcome is a prospective randomized study. Such a study has been performed and the results recently published. RTOG Study 93–05 randomized patients with glioblastoma into two treatment arms [87]. One received postoperative radiosurgery, followed by conventional radiotherapy and BCNU chemotherapy. The other arm received radiotherapy and chemotherapy without radiosurgery. At a median follow-up time of 61 months, the median survival in the radiosurgery group was 13.5 months, compared with 13.6 months in the standard treatment arm. There were no significant differences in 2- or 3-year survival, patterns of failure, or quality of life between the two groups. Notably RTOG 93–05 did not address the use of radiosurgery for recurrent malignant gliomas.
Radiosurgery for Arteriovenous Malformations Patient Selection Open surgery is generally favored if an AVM is amenable to low-risk resection (e.g., low Spetzler-Martin grade, young healthy patient) or is felt to be at high risk for hemorrhage during
941
942
59
Linac radiosurgery
the latency period between radiosurgical treatment and AVM obliteration (e.g., associated aneurysm, prior hemorrhage, large AVM with diffuse morphology, venous outflow obstruction). Radiosurgery is favored when the AVM nidus is small (<3 cm) and compact, when surgery is judged to carry a high risk or is refused by the patient, and when the risk of hemorrhage is not felt to be extraordinarily high. Endovascular treatment, although rarely curative alone, may be useful as a preoperative adjunct to either microsurgery or radiosurgery. The history, physical examination and diagnostic imaging of each patient are evaluated and the various factors outlined above are weighed in combination to determine the best treatment approach for a given case. The decision about optimal AVM treatment is best made by a multidisciplinary team composed of experts in operative, endovascular, and radiosurgical treatment.
commonly used for this purpose. Diagnostic (non-stereotactic) angiography is used to characterize the AVM, but because of its inherent inadequacies as a treatment planning database, stereotactic angiography has been largely abandoned at our institution. We use contrast enhanced, stereotactic CT as a targeting image database for the vast majority of AVMs. Our CT technique employs rapid infusion (1cc/s) of contrast while scanning through the AVM nidus with 1-mm slices. The head ring is bolted to a bracket at the head of the CT table, assuring that the head/ring/localizer complex remains immobile during the scan. This technique yields a very clear threedimensional picture of the nidus (> Figure 59-1). Alternative approaches use MRI/MRA, as opposed to CT. Attention to optimal image sequences in both CT and MRI is essential for effective AVM radiosurgical targeting.
Dose Selection Stereotactic Image Acquisition The most problematic aspect of AVM radiosurgery is target identification. In some series targeting error is listed as the most frequent cause of radiosurgical failure [88,89]. The problem lies with imaging. While angiography very effectively defines blood flow (feeding arteries, nidus, and draining veins), it does so in only two dimensions. Using the two-dimensional data from stereotactic angiography to represent the three-dimensional target results in significant errors of both overestimation and underestimation of AVM nidus dimensions [90–92]. Underestimation of the nidus size may result in treatment failure, while overestimation results in the inclusion of normal brain within the treatment volume. This can cause radiation damage to normal brain, which – when affecting an eloquent area – may result in a neurological deficit. To avoid such targeting errors, a true three-dimensional image database is required. Both contrast-enhanced CT and MRI are
Various analyses of AVM radiosurgery outcomes have elucidated an appropriate range of doses for the treatment of AVMs [93–96]. We prefer to deliver a dose of 20 Gy to the periphery of the AVM nidus whenever possible. Larger AVMs, or those in critical locations, may require a lower dose – but this will reduce the chances of complete obliteration.
Follow-up Standard follow-up after AVM radiosurgery typically consists of annual clinic visits with MRI/ MRA to evaluate the effect of the procedure and monitor for neurological complications (> Figures 59-9 and > 59-10). If the patient’s clinical status changes, she/he is followed more closely at clinically appropriate intervals. Each patient is scheduled to undergo cerebral angiography at three years post-radiosurgery, and a
Linac radiosurgery
. Figure 59-10 Pretreatment angiogram shows a left parietal AVM. It was treated with radiosurgery (17.5 Gy to the 70% line using three isocenters)
definitive assessment of the success or failure of treatment is made based on the results of angiography. If no flow is observed through the AVM nidus, the patient is pronounced cured and is discharged from follow-up. If the AVM nidus is incompletely obliterated, appropriate further therapy (most commonly repeat radiosurgery on the day of angiography) is prescribed, and the treatment/follow-up cycle is repeated.
The University of Florida Experience From 18 May 1988 to 6 November 2007, 606 patients with arteriovenous malformations were treated at the University of Florida. The mean age was 40 (4–78 years). The median treatment volume was 6 cc (0.2 – 45.3 cc). Many patients early in the series were treated with single isocenters but in recent years an effort has been made to produce highly conformal plans by employing multiple isocenters. The median radiation dose to the periphery of the AVM was
59
1750 cGy and the mean follow-up duration was 31 months. Angio or MRI cure rates vary according to AVM size: <1 cc = 93%, 1–4 cc = 81%, 4–10 cc = 64%, and >10 cc = 35%. With a second treatment, cure rates substantially increase, even for larger AVMs: <1 cc = 100%, 1–4 cc = 92%, 4–10 cc = 87%, and >10 cc = 78%. Ellis et al. [88] performed a detailed analysis of treatment failures in our series in 1998. He found that 26% of the failures were due to targeting error, at least in part. Statistical predictors of failure were increasing AVM size, decreasing treatment dose, and increasing Spetzler-Martin score. Of particular interest were the ‘‘cutpoints’’ which were identified. There was a dramatic increase in cure rates when the peripheral dose was raised to a least 15 Gy. There was a dramatic decrease in cure rate when AVM size exceeded 10 cc (size D). In a more recent analysis, a study was undertaken to determine which factors were statistically predictive of radiographic and clinical outcomes in the radiosurgical treatment of arteriovenous malformations [96]. The computerized dosimetry and clinical data on 269 patients were reviewed. The AVM nidus was hand contoured on successive enhanced CT slices through the nidus, to allow detailed determination of nidus volume, target miss, normal brain treated, dose conformality and dose gradient. In addition, a number of patient and treatment factors, including Spetzler-Martin score, presenting symptoms, dose, number of isocenters, radiographic outcome, and clinical outcome were subjected to multivariate analysis. None of the analyzed factors were predictive of permanent radiation induced complications or of hemorrhage after radiosurgery in this study. Eloquent AVM location and 12 Gy volume correlated with the occurence of transient radiation induced complications. Better conformality correlated with a reduced incidence of transient complications. Lower SpetzlerMartin scores, higher doses, and steeper dose gradients correlated with radiographic success.
943
944
59
Linac radiosurgery
When AVMs are not cured, current practice frequently involves a ‘‘retreatment,’’ usually 3 years after the original treatment. We reviewed the cases of 52 patients who underwent repeat radiosurgery for residual AVM at our institution between December 1991 and June 1998 [97]. In each case, residual arteriovenous shunting persisted beyond 36 months after the initial treatment. The mean interval between the first and second treatments was 41 months. Each AVM nidus was measured at the time of original treatment and again at the time of retreatment, and dosimetric parameters of the two treatments were compared. After retreatment, patients were followed, and their outcomes evaluated, according to our standard post-AVM radiosurgery protocol. Definitive endpoints included angiographic cure, radiosurgical failure (documented persistence of AVM flow 3 years after retreatment), and death. The mean original lesion volume was 13.8 cc and the mean volume at retreatment was 4.7 cc, for an average volume reduction of 66% after the initial ‘‘failed’’ treatment. As pointed out above, retreatment when studied in our entire patient population has substantially increased overall cure rate, with minimal morbidity. Zipfel and Friedman [98] sought to determine which morphological features of arteriovenous malformations (AVMs) are statistically predictive of preradiosurgical hemorrhage, postradiosurgical hemorrhage, and neuroimagingdefined failure of radiosurgical treatment. Archived CT dosimetry and available angiographic and clinical data for 268 patients in whom AVMs were treated with linear accelerator radiosurgery were retrospectively reviewed (> Figure 59-11). Many of the morphological features of AVMs, including location, volume, compact or diffuse nidus, neovascularity, ease of nidus identification, number of feeding arteries, location (deep or superficial) of feeding arteries, number of draining veins, deep or superficial venous drainage, venous stenoses,
. Figure 59-11 Two years later, the angiogram is normal
venous ectasias, and the presence of intranidal aneurysms, were analyzed. In addition, a number of patient and treatment factors, including patient age, presenting symptoms, radiation dose, repeated treatment, and radiological outcome, were subjected to multivariate analyses. A larger AVM volume (odds ratio [OR] 0.349; p = 0.004) was associated with a decreased rate of pretreatment hemorrhage, whereas periventricular location (OR 6.358; p = 0.000) was associated with an increased rate of pretreatment hemorrhage. None of the analyzed factors was predictive of hemorrhage following radiosurgery. A higher radiosurgical dose was strongly correlated with neuroimaging-defined success (OR 3.743; p = 0.006), whereas a diffuse nidus structure (OR 0.246; p = 0.008) and associated neovascularity (OR 0.428; p = 0.048) were each associated with a lower neuroimaging-defined cure rate. A strong correlation between CT scanning and angiography was noted for both nidus structure (p = 0.000; Fisher exact test) and neovascularity (p = 0.002; Fisher exact test). Clinical follow-up information is available on 472 patients. Twelve (2.5%) have sustained permanent radiation induced neurological deficits.
Linac radiosurgery
59
. Table 59-1 LINAC AVM radiosurgery series with patient numbers greater than 20 Series Betti [99] Colombo [100] Engenhart [101] Friedman et al. [96] Schlienger [102] Young [103]
#Patients
Angiographic success rate
286 180 212 554 169 66
82% 80% 72% 74% 64% 50%
Fifteen (3%) have sustained transient deficits. Most importantly, 41 have had post-treatment hemorrhages (eight fatal). One must remember that the major drawback of radiosurgery for arteriovenous malformations is the continued risk of hemorrhage during the latent period (typically 2 years). The results of other LINAC radiosurgery series are summarized in > Table 59-1.
1 patient 2.2% 4.3% 1.8%
11.
12.
13. 14.
References 1. Leksell L. The stereotaxic method and radiosurgery of the brain. Acta Chir Scand 1951;102:316-9. 2. Betti OO, Derechinsky VE. Hyperselective encephalic irradiation with a linear accelerator. Acta Neurol Chir Suppl 1984;33:385-90. 3. Colombo F, Benedetti A, Pozza F, et al. External stereotactic irradiation by linear accelerator. Neurosurgery 1985;16:154-60. 4. Hartmann GH, Schlegel W, Sturm V, et al. Cerebral radiation surgery using moving field irradiation at a linear accelerator facility. Int J Radiat Oncol Biol Phys 1985;11:1185-92. 5. McGinley PH, Butker EK, Crocker IR, Landry JC. A patient rotator for stereotactic radiosurgery. Phys Med Biol 1990;35:649-57. 6. Podgorsak EB, Olivier A, Pla M, Lefebvre PY, Hazel J. Dynamic stereotactic radiosurgery. Int J Radiat Oncol Biol Phys 1988;14:115-26. 7. Winston KR, Lutz W. Linear accelerator as a neurosurgical tool for stereotactic radiosurgery. Neurosurgery 1988; 22:454-64. 8. Friedman WA, Bova FJ. The University of Florida radiosurgery system. Surg Neurol 1989;32:334-42. 9. Kondziolka D, Lunsford LD, McLaughlin MR, Flickinger JC. Long-term outcomes after radiosurgery for acoustic neuromas. N Engl J Med 1998;339(20):1426-33. 10. Samii M, Matthies C. Management of 1,000 vestibular schwannomas (acoustic neuromas): surgical management
Permanent radiation complications
15.
16.
17.
18.
19.
20.
21.
22.
23.
and results with an emphasis on complications and how to avoid them. Neurosurgery 1997;40:11-23. Samii M, Matthies C. Management of 1,000 vestibular schwannomas (acoustic neuromas): the facial nerve – preservation and restitution of function. Neurosurgery 1997;40:684-95. Samii M, Matthies C. Management of 1,000 vestibular schwannomas (acoustic neuromas): hearing function in 1,000 tumor resections. Neurosurgery 1997;40:248-62. Leksell L. A note on the treatment of acoustic tumors. Acta Chir Scand 1971;137:763-5. Foote KD, Friedman WA, Buatti JM, Meeks SL, Bova FJ, Kubilis PS. Analysis of risk factors associated with radiosurgery for vestibular schwannoma. J Neurosurg 2001; 95(3):440-9. Friedman WA, Bradshaw P, Myers A, Bova FJ. Linear accelerator radiosurgery for vestibular schwannomas. J Neurosurg 2006;105(5):657-61. Beegle RD, Friedman WA, Bova FJ. Effect of treatment plan quality on outcomes after radiosurgery for vestibular schwannoma. J Neurosurg 2007;107(5):913-6. Vachhani JA, Friedman WA. Radiosurgery in patients with bilateral vestibular schwannomas. Stereotact Funct Neurosurg 2007;85(6):273-8. Spiegelmann R, Gofman J, Alezra D, Pfeffer R. Radiosurgery for acoustic neurinomas (vestibular schwannomas). Isr Med Assoc J 1999;1(1):8-13. Spiegelmann R, Lidar Z, Gofman J, Alezra D, Hadani M, Pfeffer R. Linear accelerator radiosurgery for vestibular schwannoma. J Neurosurg 2001;94(1):7-13. Martens F, Verbeke L, Piessens M, Van Vyve M. Stereotactic radiosurgery of vestibular schwannomas with a linear accelerator. Acta Neurochir Suppl 1994; 62:88-92. Valentino V, Raimondi AJ. Tumour response and morphological changes of acoustic neurinomas after radiosurgery. Acta Neurochir 1995;133:157-63. Ishihara H, Saito K, Nishizaki T, et al. CyberKnife radiosurgery for vestibular schwannoma. Minim Invasive Neurosurg 2004;47(5):290-3. Delaney G, Matheson J, Smee R. Stereotactic radiosurgery: an alternative approach to the management of acoustic neuromas. Med J Austral 1992;156:440.
945
946
59
Linac radiosurgery
24. Barcia Salorio JL, Hernandez G, Ciudad J, Bordes V, Broseta J. Stereotactic radiosurgery in acoustic neurinoma. Acta Neurochir Suppl 1984;33:373-6. 25. Sekhar LN, Jannetta PJ, Burkhart LE, Janosky JE. Meningiomas involving the clivus: a six-year experience with 41 patients. Neurosurgery 1990;27:764-81. 26. Sekhar LN, Altschuler EM. Meningiomas of the cavernous sinus. In: Al-Mefty O, editor. Meningiomas. New York: Raven Press; 1991. p. 445-60. 27. Simpson D. The recurrence of intracranial meningiomas after surgical treatment. J Neurol Neurosurg Psychiatry 1957;20:22-39. 28. O’Neill BP, Iturria NJ, Link MJ, Pollock BE, Ballman KV, O’Fallon JR. A comparison of surgical resection and stereotactic radiosurgery in the treatment of solitary brain metastases. Int J Radiat Oncol Biol Phys 2003; 55:1169-76. 29. Valentino V, Schinaia G, Raimondi AJ. The results of radiosurgical management of 72 middle fossa meningiomas. Acta Neurochir 1993;122:60-70. 30. Villavicencio AT, Black PM, Shrieve DC, Fallon MP, Alexander E, Loeffler JS. Linac radiosurgery for skull base meningiomas. Acta Neurochir (Wien) 2001; 143(11):1141-52. 31. Engenhart R, Kimmig BN, Hover KH, et al. Stereotactic single high dose radiation therapy of benign intracranial meningiomas. Int J Radiat Oncol Biol Phys 1990; 19:1021-6. 32. Spiegelmann R, Nissim O, Menhel J, Alezra D, Pfeffer MR. Linear accelerator radiosurgery for meningiomas in and around the cavernous sinus. Neurosurgery 2002;51 (6):1373-79; discussion 1379–80. 33. Chuang CC, Chang CN, Tsang NM, et al. Linear accelerator-based radiosurgery in the management of skull base meningiomas. J Neurooncol 2004;66(1–2): 241-9. 34. Deinsberger R, Tidstrand J, Sabitzer H, Lanner G. LINAC radiosurgery in skull base meningiomas. Minim Invasive Neurosurg 2004;47(6):333-8. 35. Hakim R, Alexander E, III, Loeffler JS, et al. Results of linear accelerator-based radiosurgery for intracranial meningiomas. Neurosurgery 1998;42:446-54. 36. Friedman WA, Murad G, Bradshaw P, et al. Linear accelerator radiosurgery for meningiomas. J Neurosurg 2005;103:206-9. 37. Lohr F, Pirzkall A, Hof H, Fleckenstein K, Debus J. Semin Surg Oncol 2001;20:50-6. 38. DeAngelis LM. Brain tumors. N Engl J Med 2001; 344:114-23. 39. Patchell RA, Tibbs PA, Walsh JW, et al. A randomized trial of surgery in the treatment of single metastases to the brain. N Engl J Med 1990;322:494-500. 40. Noordijk EM, Vecht CJ, Haaxma-Reiche H, et al. The choice of treatment of single brain metastasis should be made based on extracranial tumor activity and age. Int J Radiat Oncol Biol Phys 1994;29:711-7.
41. Mintz AH, Kestle J, Rathbone MP, et al. A randomized trial to assess the efficacy of surgery in addition to radiotherapy in patients with a single cerebral metastasis. Cancer 1996;78(7):1470-6. 42. Haines SJ. Moving targets and ghosts of the past: outcome measurement in brain tumour therapy. J Clin Neurosci 2002;9:109-12. 43. Sturm V, Kober B, Hover KH, et al. Stereotactic percutaneous single dose irradiation of brain metastases with a linear accelerator. Int J Radiat Oncol Biol Phys 1987;13:279-82. 44. Sturm V, Kimmig B, Engenhardt R, et al. Radiosurgical treatment of cerebral metastases. J Stereotactic Funct Neurosurg 1991;57:7-10. 45. Voges J, Treuer H, Erdmann J, et al. LINAC radiosurgery in brain metastases. Acta Neurochir Suppl 1994; 62:72-6. 46. Black PM. Solitary brain metastases: radiation, resection, or radiosurgery? Chest 1993;103:367S-9S. 47. Alexander E, Moriarty TM, Davis RB, et al. Stereotactic radiosurgery for the definitive, noninvasive treatment of brain metastases. J Nat Cancer Inst 1995;87:34-40. 48. Adler JR, Cox RS, Kaplan I, Martin DP. Stereotactic radiosurgical treatment of brain metastases. J Neurosurg 1992;76:444-9. 49. Fuller BG, Kaplan ID, Adler J, Cox RS, Bagshaw MA. Stereotaxic radiosurgery for brain metastases: the importance of adjuvant whole brain irradiation. Int J Radiat Oncol Biol Phys 1992;23:413-8. 50. Joseph J, Adler JR, Cox RS, Hancock SL. Linear accelerator-based stereotaxic radisourgery for brain metastases: the influence of number of lesions on survival. J Clin Oncol 1996;14:1085-92. 51. Auchter RM, Lamond JP, Alexander E, et al. A multiinstitutional outcome and prognostic factor analysis of radiosurgery for resectable single brain metastasis. Int J Radiat Oncol Biol Phys 1996;35:27-35. 52. Becker G, Jeremic B, Engel C, et al. Radiosurgery for brain metastases: the Tuebingen experience. Radiother Oncol 2002;62:233-7. 53. Breneman JC, Warnick RE, Albright RE, et al. Stereotactic radiosurgery for the treatment of brain metastases. Cancer 1997;79:551-7. 54. Buatti JM, Friedman WA, Bova FJ, Mendenhall WM. Treatment selection factors for stereotactic radiosurgery of intracranial metastases. Int J Radiat Oncol Biol Phys 1995;32:1161-6. 55. Caron J-L, Souhami L, Podgordak EB. Dynamic stereotactic radiosurgery in the palliative treatment of cerebral metastatic tumors. J Neuro Oncol 1992;12:173-9. 56. Gutin PH, Wilson CB. Radiosurgery for malignant brain tumors. J Clin Oncol 1990;8:571-3. 57. Mehta MP, Rozental JM, Levin AB, et al. Defining the role of radiosurgery in the management of brain metastases. Int J Radiat Oncol Biol Phys 1992;24:619-25. 58. Mehta M, Noyes W, Craig B, et al. A cost-effectiveness and cost-utility analysis of radiosurgery vs. resection for
Linac radiosurgery
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
single-brain metastases. Int J Radiat Oncol Biol Phys 1997;39:445-54. Valentino V. The results of radiosurgical management of 139 single cerebral metastases. Acta Neurochir Suppl 1995;63:95-100. Cho KH, Hall WA, Gerbi BJ, Higgins PD, Bohen M, Clark HB. Patient selection criteria for the treatment of brain metastases with stereotactic radiosurgery. J Neuro Oncol 1998;40:73-86. Fernandez-Vicioso E, Suh JH, Kupelian PA, Sohn JW, Barnett GH. Analysis of prognostic factors for patients with single brain metastasis treated with stereotactic radiosurgery. Rad Onc Invest 1997;5:31-7. Maor MH, Dubey P, Tucker SL, et al. Stereotactic radiosurgery for brain metastases: results and prognostic factors. Int J Cancer 2000;90:157-62. Goodman KA, Sneed PK, McDermott MW, et al. Relationship between pattern of enhancement and local control of brain metastases after radiosurgery. Int J Radiat Oncol Biol Phys 2001;50:139-46. Shiau C-Y, Sneed PK, Shu H-KG, et al. Radiosurgery for brain metastases: relationship of dose and pattern of enhancement to local control. Int J Radiat Oncol Biol Phys 1997;37:375-83. Gaspar L, Scott C, Rotman M, et al. Recursive partitioning analysis of prognostic factors in three radiation therapy oncology group (RTOG) brain metastases trials. Int J Radiat Oncol Biol Phys 1997;37:745-51. Ulm AJ, Friedman WA, Bova FJ, Bradshaw P, Amdur RJ, Mendenhall WM. Linear accelerator radiosurgery in the treatment of brain metastases. Neurosurgery 2004; 55:1076-85. Andrews DW, Scott CB, Sperduto PW, et al. Whole brain radiation therapy with or without stereotactic radiosurgery boost for patients with one to three brain metastases: phase III results of the RTOG 9508 randomised trial. Lancet 2004;363(9422):1665-72. Aoyama H, Tago M, Kato N, et al. Neurocognitive function of patients with brain metastasis who received either whole brain radiotherapy plus stereotactic radiosurgery or radiosurgery alone. Int J Radiat Oncol Biol Phys 2007;68(5):1388-95. Swinson BM, Friedman WA. Linear accelerator stereotactic radiosurgery for metastatic brain tumors: 17 years of experience at the University of Florida. Neurosurgery 2008;62(5):1018-31; discussion 1031-32. Bernstein M, Laperriere N, Glen J, Leung P, Thomason C, Landon AE. Brachytherapy for recurrent malignant astrocytoma. Int J Radiat Oncol Biol Phys 1994; 30(5):1213-7. Chang CN, Chen WC, Wei KC, et al. High-dose-rate stereotactic brachytherapy for patients with newly diagnosed glioblastoma multiformes. J Neuro Oncol 2003; 61(1):45-55. Prados MD, Gutin PH, Phillips TL, et al. Interstitial brachytherapy for newly diagnosed patients with
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
59
malignant gliomas: the UCSF experience. Int J Radiat Oncol Biol Phys 1992;24(4):593-7. Shrieve DC, Alexander E, III, Wen PY, et al. Comparison of stereotactic radiosurgery and brachytherapy in the treatment of recurrent glioblastoma multiforme. Neurosurgery 1995;36(2):275-82; discussion 82–4. Hall WA, Djalilian HR, Sperduto PW, et al. Stereotactic radiosurgery for recurrent malignant gliomas. J Clin Oncol 1995;13(7):1642-8. Buatti JM, Friedman WA, Bova FJ, Mendenhall WM. Linac radiosurgery for high-grade gliomas: the University of Florida experience. Int J Radiat Oncol Biol Phys 1995; 32(1):205-10. Masciopinto JE, Levin AB, Mehta MP, Rhode BS. Stereotactic radiosurgery for glioblastoma: a final report of 31 patients. J Neurosurg 1995;82(4):530-5. Regine WF, Patchell RA, Strottmann JM, Meigooni A, Sanders M, Young AB. Preliminary report of a phase I study of combined fractionated stereotactic radiosurgery and conventional external beam radiation therapy for unfavorable gliomas. Int J Radiat Oncol Biol Phys 2000;48(2):421-6. Prisco FE, Weltman E, de Hanriot RM, Brandt RA. Radiosurgical boost for primary high-grade gliomas. J Neuro Oncol 2002;57(2):151-60. Gannett D, Stea B, Lulu B, Adair T, Verdi C, Hamilton A. Stereotactic radiosurgery as an adjunct to surgery and external beam radiotherapy in the treatment of patients with malignant gliomas. Int J Radiat Oncol Biol Phys 1995;33(2):461-8. Roberge D, Souhami L. Stereotactic radiosurgery in the management of intracranial gliomas. Technol Cancer Res Treat 2003;2(2):117-25. Curran WJ, Jr., Scott CB, Horton J, et al. Recursive partitioning analysis of prognostic factors in three Radiation Therapy Oncology group malignant glioma trials. J Natl Cancer Inst 1993;85(9):704-10. Sarkaria JN, Mehta MP, Loeffler JS, et al. Radiosurgery in the initial management of malignant gliomas: survival comparison with the RTOG recursive partitioning analysis. Int J Radiat Oncol Biol Phys 1995;32(4): 931-41. Kondziolka D, Flickinger JC, Bissonette DJ, Bozik M, Lunsford LD. Survival benefit of stereotactic radiosurgery for patients with malignant glial neoplasms. Neurosurgery 1997;41(4):776-83; discussion 83–5. Ulm AJ, III, Friedman WA, Bradshaw P, Foote KD, Bova FJ. Radiosurgery in the treatment of malignant gliomas: the University of Florida experience. Neurosurgery 2005;57(3):512-7; discussion 517. Irish WD, Macdonald DR, Cairncross JG. Measuring bias in uncontrolled brain tumor trials – to randomize or not to randomize? Can J Neurol Sci 1997; 24(4): 307-12. Curran WJ, Jr., Scott CB, Weinstein AS, et al. Survival comparison of radiosurgery-eligible and -ineligible
947
948
59
Linac radiosurgery
malignant glioma patients treated with hyperfractionated radiation therapy and carmustine: a report of Radiation Therapy Oncology Group 83–02. J Clin Oncol 1993; 11(5):857-62. 87. Souhami L, Seiferheld W, Brachman D, et al. Randomized comparison of stereotactic radiosurgery followed by conventional radiotherapy with carmustine to conventional radiotherapy with carmustine for patients with glioblastoma multiforme: report of Radiation Therapy Oncology Group 93–05 protocol. Int J Radiat Oncol Biol Phys 2004;60(3):853-60. 88. Ellis TL, Friedman WA, Bova FJ, Kubilis PS, Buatti JM. Analysis of treatment failure after radiosurgery for arteriovenous malformations. J Neurosurg 1998;89(1):104-10. 89. Pollock BE, Flickinger JC, Lunsford LD, Maitz A, Kondziolka D. Factors associated with successful arteriovenous malformation radiosurgery. Neurosurgery 1998; 42(6):1239-44; discussion 44–7. 90. Bova FJ, Friedman WA. Stereotactic angiography: an inadequate database for radiosurgery? Int J Radiat Oncol Biol Phys 1991;20:891-5. 91. Blatt DL, Friedman WA, Bova FJ. Modifications in radiosurgical treatment planning of arteriovenous malformations based on CT imaging. Neurosurgery 1993;33:588-96. 92. Spiegelmann R, Friedman WA, Bova FJ. Limitations of angiographic target localization in planning radiosurgical treatment. Neurosurgery 1992;30:619-24. 93. Flickinger JC, Pollock BE, Kondziolka D, Lunsford LD. A dose-response analysis of arteriovenous malformation obliteration after radiosurgery. Int J Radiat Oncol Biol Phys 1996;36:873-9. 94. Karlsson B, Lindquist C, Steiner L. Prediction of obliteration after gamma knife surgery for cerebral arteriovenous malformations. Neurosurgery 1997;40(3):425-30; discussion 30–1. 95. Pollock BE, Kondziolka D, Lunsford LD, Bissonette D, Flickinger JC. Repeat stereotactic radiosurgery of arteriovenous malformations: factors associated with incomplete obliteration. Neurosurgery 1996;38(2): 318-24. 96. Friedman WA, Bova FJ, Bollampally S, Bradshaw P. Analysis of factors predictive of success or complications in arteriovenous malformation radiosurgery. Neurosurgery 2003;52:296-308. 97. Foote KD, Friedman WA, Ellis TL, Bova FJ, Buatti JB, Meeks SL. Salvage retreatment after failure of radiosurgery in patients with arteriovenous malformations. J Neurosurg 2003;98:337-41. 98. Zipfel GJ, Bradshaw P, Bova FJ, Friedman WA. Do the morphological characteristics of arteriovenous malformations affect the results of radiosurgery? J Neurosurg 2004;101(3):393-401. 99. Betti OO, Munari C, Rosler R. Stereotactic radiosurgery with the linear accelerator: treatment of arteriovenous malformations. Neurosurgery 1989;24:311-21.
100. Colombo F, Pozza F, Chierego G, Casentini L, De Luca G, Francescon P. Linear accelerator radiosurgery of cerebral arteriovenous malformations: an update. Neurosurgery 1994;34:14-21. 101. Engenhart R, Wowra B, Debus J, et al. The role of high-dose single fraction irradiation in small and large intracranial arteriovenous malformations. Int J Radiat Oncol Biol Phys 1994;30:521-9. 102. Schlienger M, Atlan D, Lefkopoulos D, et al. LINAC radiosurgery for cerebral arteriovenous malformations. Int J Radiat Oncol Biol Phys 2000;46:1135-42. 103. Young C, Summerfield R, Schwartz M, O’Brien P, Ramani R. Radiosurgery for arteriovenous malformations: the University of Toronto experience. Can J Neurol Sci 1997;24:99-105.
67 Linac Radiosurgery: Technical Aspects F. J. Bova . W. A. Friedman
Introduction Radiosurgery, a term coined in 1951 by the Swedish neurosurgeon Lars Leksell [1], was first practiced using an orthovoltage x-ray apparatus, then a particle accelerator and finally a Co-60 isotope unit. The majority of Leksell’s clinical work was carried out using the latter device, known as the Gamma Knife. With this dedicated tool Leksell was able to pioneer many radiosurgical procedures. By the mid-1980s only a handful of GammaKnife’s existed, limiting the application of this new technique. By the 1980s the linear accelerators had, for the most part, replaced Cobalt-60 units for routine radiation therapy. In the mid 1980s researchers began to adapt linacs as radiosurgical tools [2–4]. These developments had immediate effects on the field of radiosurgery. Once the successful application of the linac based radiosurgery paradigm, was demonstrated, hundreds of radiation oncologists and neurosurgeons found that their facilities could purchase the necessary hardware and software to upgrade their linac making them radiosurgery capable. Surprisingly, this upgrade could be accomplished for a tenth the cost of the Gamma Knife. In less then a decade the number of radiosurgical units worldwide went from fewer than half a dozen to hundreds. As radiation oncologists became more involved in the day-to-day use of radiosurgical techniques, they realized that stereotactic techniques had application beyond the traditional neurosurgical procedures pioneered by Leksell. These applications were both intra and extra cranial. Many of the new extra cranial applications required #
Springer-Verlag Berlin/Heidelberg 2009
further innovative targeting technologies employing ultrasound, orthogonal x-ray and conebeam. ‘‘Radiosurgery’’ was initially defined as a ‘‘single-fraction’’ technique that uses ‘‘stereotactic principles’’ for targeting and treatment of ‘‘intracranial’’ lesions through the use of ‘‘multiple noncoplanar beams’’. Over the past decade the stereotactic techniques on which radiosurgery are founded have been applied to fractionated treatments, termed stereotactic radiotherapy. Currently the line between single fraction radiosurgery and a limited number of fraction, i.e., usually less then five, radiosurgery and fractionated stereotactic radiosurgery, FRS, is not particularly distinct. For this text the term radiosurgery will be used to designate the single fraction technique. In considering the accuracy and precision necessary in a radiosurgical system, it is important to distinguish between the addition of a margin to account for the inclusion of microscopic disease and the necessity of adding a margin because of system uncertainty. If the lack of system accuracy requires the clinician to prescribe a margin of normal tissue to all stereotactic targets to ensure that the identified target is included in the selected isodose volume, normal tissues will unnecessarily be exposed to radiation. For example, adding a margin of 2 mm to a 24 mm target increases the volume irradiated from 7.2 to 11.5 cc, an volume increase of 60%, all of which is normal tissue. Another aspect of the stereotactic treatment must be considered before one can discuss the requirements of a radiosurgical system. Most radiosurgical targets cannot be imaged through conventional radiotherapy techniques involving
1088
67
Linac radiosurgery: technical aspects
simulation and portal verification. The majority of radiosurgical targets can only be visualized during diagnostic examinations, such as angiography or from a three-dimensional data set assembled from computed tomography (CT) and magnetic resonance imaging (MRI). Recently functional imaging in the form of positron emission tomography, PET, and functional magnetic resonance imaging, f MRI, have begun to be used to help define target as well as normal critical structures. At the time of treatment, i.e., the alignment of the patient to the therapeutic x-ray beam, none of these imaging modalities are available, blinding the clinician and forcing the reliance upon the spatial integrity of the stereotactic targeting system. The quality assurance procedures necessary to guarantee system accuracy under these conditions make up a large part of each radiosurgical treatment. Also, the use of a stereotactic system for radiosurgery is more technically demanding than the use of the same system for other stereotactic techniques. In the case of a stereotactic biopsy the neurosurgeon can target a region and acquire a series of biopsies along a probe track or from multiple placements of the biopsy probe. In the case of deep brain stimulation the neurosurgeon has the stimulation to help verify probe placement. The radiosurgeon has none of these verification tools available. It is this absolute dependence on stereotactic targeting that places more stringent requirements on radiosurgical systems than on other stereotactic neurosurgical procedures. When deciding the accuracy and precision necessary for stereotactic applications, one is inevitably told that the most inaccurate portion of the procedure is the identification of the target. This often comes in the form of the question; ‘‘Since we don’t know precisely where the target begins or ends, why should we worry about system accuracy?’’ This line of logic has two serious flaws. The first is that all inaccuracies are cumulative, starting with diagnostic imaging, through treatment planning and then through treatment
delivery. The fact that the edge of the target volume cannot be accurately defined is little justification for not being able to precisely predict a prescribed dose distribution. The second is that avoidance of critical structures, such as the optic chiasm, is often as important as the inclusion of target tissues. Very often the adjacent critical structures, the location of which are exactly known, place constraints on the prescribed dose. Inaccurate targeting and treatment delivery can as easily include critical structures as it can exclude target tissues. Knowing the position of the highisodose volumes as well as the location of the dose gradient to within a pixel of its true position is an absolute necessity. Over the years many different immobilization techniques have been employed in radiation oncology in attempts to provide a high degree of alignment between diagnostic imaging, that allowed the target and normal tissues to be appreciated, and patient alignment with the therapeutic beam. Radiosurgery initially overcame this obstacle through the use of a rigidly fixed reference system in the form of a stereotactic frame. The first step of each stereotactic procedure is the application of the reference frame to the patient’s skull. Once applied, this frame remains rigidly attached to the patient’s skull throughout the procedure. Several frames have been developed for general stereotactic applications, and many of these have been adapted for radiosurgical use. Each of these systems has unique advantages and disadvantages as well as its own coordinate system. They all, however, have one common feature: once the frame is fixed to the patient’s skull, a rigid relationship between that patient’s intracranial anatomy and the frame’s coordinate system is established. Frameless approaches have been pioneered [5] each system substitutes a somewhat different object or structure for the rigid frame. All require the same assumption of a rigid alignment between the patient’s anatomy and the reference system.
Linac radiosurgery: technical aspects
All of the basic principles, a long with the advantages and disadvantages of a frame based approach, apply to these newer techniques. Initially all diagnostic information necessary for targeting and planning had to occur after ring placement. However, with the advent of image fusion techniques, non-stereotactic exams, exams carried out prior to the application of the headring, can be registered to the frame-based exam, providing significant flexibility in obtaining target images. The three most common diagnostic procedures used for stereotactic localization are angiography, CT, and MR with fMRI, PET and DTI are beginning to find application in the planning process. Angiography provides unique information concerning vascular structures, and CT and MRI provide information about both target tissues and normal anatomy and allow a full threedimensional model of the patient’s intracranial anatomy to be reconstructed. Stereotactic imaging systems can be separated into two categories. The first includes systems that produce images containing all the information necessary to compute the stereotactic coordinates of all tissues contained within that image. Embedded in each image is a description of the stereotactic coordinate system. The fiducial marker on each image allows the trajectory of the x-ray beam relative to the stereotactic coordinate system to be computed. In the case of CT the axial sections are not required to be absolutely parallel to the plane of the stereotactic reference system. The plane of the scan relative to the reference system can be computed with the information provided within each transaxial image. When an independent system is used, the stereotactic application can be designed so that all necessary quality assurance can be carried out on the images used in each procedure. This is usually accomplished through overdefined fiducial systems or by verifying the geometry of fixed fiducial markers.
67
The second type of stereotactic system, a dependent system, does require special knowledge of the imaging conditions. For biplanar imaging, either orthogonality or exact knowledge of the angle between image planes and the x-ray source to detector distance is required. When a dependent system is used, the quality assurance of the stereotactic procedure must be linked to the quality assurance of the diagnostic equipment at the time of image acquisition. Such quality assurance issues may include the precision of the indexing of the CT patient table, the orthogonality of the CT gantry to the axis of the stereotactic system, the alignment of the laser alignment system within the CT gantry, or the orthogonality of plane film images within a biplanar angiographic procedure. While it is feasible to use dependent reference systems, there appears to be little to gain and a great deal to lose in terms of potential for errors and increased quality assurance time. It is recommended that an independent system be used whenever possible. To bring a better focus to the discussion of stereotactic frames and coordinate systems as applied to radiosurgery and stereotactic radiotherapy, one system, the Brown–Robert–Wells (BRW) system [6], will be used throughout this text. The basic stereotactic system is shown in > Figure 67-1. It consists of a frame, angiographic localizer, CT localizer, and a method of fixing the frame to a linac. This system has three orthogonal axes: anteroposterior, lateral, and axial. The origin of the anteroposterior axis as well as of the lateral axis is defined at the center of the circular reference frame with the origin of the axial axis is defined to be 80 mm from the superior surface of the ring. The function of each localization system is to embed a set of fiducial marks in each image (> Figure 67-2). The goal of treatment planning, as is stated in almost every text written about the process, is to devise a plan that concentrates the dose to the target tissues while sparing all normal tissues. A perfect plan would therefore provide the
1089
1090
67
Linac radiosurgery: technical aspects
. Figure 67-1 BRW Headring, patient being CT scanned and patient mounted on stereotactic floorstand aligned for treatment
prescribed dose to the target tissues and no dose to any normal tissue, an impossible task. This would require a dose distribution with an infinitely steep dose gradient, that is, a dose gradient that instantaneously decreases from a target dose to zero dose, i.e., a step function. For a specific target volume the best plan is the plan which most highly conforms to the target volume while providing the minimum dose to all normal, nonetarget, tissues. The multiarc treatment technique used in linac radiosurgery is capable of producing conformal dose volumes that only allow a few tenths of a cc of normal tissue to be included in the prescription volume while providing a dose gradient that
decreases from a specified target dose to a subclinical dose, usually the target isodose level to half that isodose level, within approximately 3 mm. This distribution can be easily achieved using a treatment planning technique know as sphere packing for both circular and non-circular target volumes. The sphere packing technique allows the radiosurgeon to apply a set of non-coplanar beams, usually grouped into several non-coplanar arcs. These arcs effectively allow the planner to apply hundreds of small circular non-coplanar beams, all with unique entrance and exit trajectories to the target volume. The result of such a beam arrangement is a spherical distribution of
Linac radiosurgery: technical aspects
67
. Figure 67-2 BRW CT localizer, without headring, aligned with transaxial images showing fiducial rods in each image
dose with the high dose region being the volume where all beams overlap and a rapid decrease in dose as the beams diverge away from the target tissues. This approach represents a class solution that when applied with accompanying rules for sphere spacing and weighting, can produce a highly conformal plan with steep dose gradients at the edge of the treatment volume. The planning process for non-spherical target volumes involves the repeat application of these spheres of dose with the alignment of the outer target edge with the outer shell of the dose spheres, (> Figure 67-3). This technique provides a somewhat inhomogeneous dose inside the target volume. However, unlike routine fractionated radiation therapy, where the target volume contains a high percentage of normal cells, the radiosurgical target volume is nearly 100% target cell, containing almost no normal cells. It is this high concentration of target cells, small total treatment volume of doses about 10% of the prescription dose, as well as minimal dose to normal tissues, that allow sphere packing to provide excellent clinical results.
In many clinical situations it is desirable to provide a high dose gradient along the entire surface of the target volume. This is generally true because most intra cranial tissue has similar sensitivity to radiation. One exception are tissues involved in optical processing. There are clinical situation where it is desirable to provide a nonuniformly steep gradient. In such cases the aim of planning may be to provide the appropriate dose concentration over the target tissues while orienting the steepest possible dose gradient in the direction of the most important adjacent normal tissue structure. For example, when treating a target involving or near the pituitary, the most important adjacent normal tissue structure is usually the optic chiasm. Treating this target with a uniform dose gradient can produce a relatively high dose to the brainstem (> Figure 67-4). A simple modification of this plan can create a much steeper dose gradient in the direction of the optic chiasm that then results in a less steep dose gradient lateral to the target tissues. In this case these lateral tissues are less important than the tissues superior to the target,
1091
1092
67
Linac radiosurgery: technical aspects
. Figure 67-3 Non-spherical target planned with three isocenters, image showing the location of each of the three isocenters
. Figure 67-4 Two treatments plans for treating a pituitary tumor. The plan on the left uses a symmetric set of nine arcs and the plan on the right uses six arcs selected to maximize the dose gradient in the direction of the optic chiasm
Linac radiosurgery: technical aspects
and this plan modification provides significant clinical benefit. The requirements of a radiosurgical planning system center on providing information necessary to appreciate the target’s size and shape and the relative location of all normal tissues as well as providing the information necessary to optimize dose to the target tissues and dose avoidance to all normal tissues. The parameters available for plan optimization are collimator size, arcing angle, arc placement, isocenter position and in the case of multiple isocenters, relative isocenter weight [7].
Equipment Certification Linac Setup and Isocentric Testing The first procedure in preparing a linac for radiosurgery is to measure the isocentric accuracy of the gantry and patient support systems to allow the combined system’s isocentric accuracy to be evaluated. This procedure does not necessarily parallel the isocentric test performed during acceptance testing or yearly quality assurance testing. In most cases the accepted method of testing the accuracy of a linear accelerator is to individually test the isocentric accuracy of the gantry, collimator and patient support system. Although the customary procedure allows for the accuracy of each subsystem to be evaluated, it does not test the combined accuracy of the gantry and patient support system. If for example, the gantry’s isocentric accuracy is at its high point i.e., outer limit compared to the mean or nominal isocenter, when the gantry is at 0 and the beam is pointed straight down, and if the patient support system is also at one of its extremes at 0 , the combined accuracies of the total gantry plus patient support system will exceed the individual accuracies of either subsystem. It is therefore possible to have both rotational specifications to be within their tolerance
67
of plus or minus 1 mm and have the effective combined accuracy to approach twice the individually certified specification. While these issues are important for all non-coplanar based linac based treatments, they are especially important for older equipment, which is often underused and often relegated to the less frequent procedures such as radiosurgery.
Radiation Testing of Isocenter The testing of radiation isocenter is one of the most important quality assurance procedures in radiosurgery. This procedure provides the certification that the beam delivery system can remain focused on the target coordinate. This test should be executed so that it examines the beamtarget alignment at points throughout the stereotactic space and throughout the extent of beam gantry and patient support motion. For example, if during CT or angiography the stereotactically defined space includes coordinates of – 100 to 100 mm in each of the three orthogonal directions, anteroposterior, lateral and axial, the test should have test points at these extremes. Since more than 95% of the radiosurgical targets are confined to a space defined by a 100 100 100 mm cube, one can make an argument for concentrating on system tests using plus or minus 50-mm points as opposed to plus or minus 100-mm points. However, as will be discuss, pretreatment setup test should always test the accuracy of actual patient specific target settings. It is best to first systematically test each orthogonal axis separately and to then test combinations of axes. It is not sufficient to test the center of the target space or any other single pointing space and to consider that a validation of accuracy across all available stereotactic space. The treatment process requires that tissues throughout the available target space be localized and brought to the linac’s isocenter. The linac QA process must guarantee that errors such as the potential misalignment of axis with the linacs
1093
1094
67
Linac radiosurgery: technical aspects
target alignment system are fully evaluated and eliminated. The exact method of carrying out this test depends on the stereotactic system being used. These systems can be divided into basic types. The first division concerns the patient-isocenter alignment. There are systems that use auxiliary patient support systems, such as a floor stand, to position the patient at the system’s isocenter. These systems usually rely upon mechanical micrometer scales for aligning the target tissue to the linac’s isocenter. There are also systems that use the linac’s native patient support system, i.e., patient table support, and rely upon room laser or optical systems for alignment to the linac’s isocenter. The commercial floor stand systems provide very rigid and robust alignment, requiring little dayto-day adjustment as well as providing stability during patient rotations. The table mount systems often require daily certification of laser alignment or calibration of optical alignment systems prior to a radiosurgical procedure. Plus as the equipment ages, these systems can often require added realignment whenever patient table rotations, new arcing planes, are required. Because the stereotactic process begins with diagnostic imaging and ends with the alignment and application of a therapeutic beam, it is critical that institutional procedures exist to certify the entire localization-treatment chain. Many systems provide stereotactic phantoms with targets that are compatible with stereotactic CT image acquisition as well as with megavoltage x-ray beam alignment testing. Some such phantoms are rigid, providing a set of targets, set at known stereotactic coordinates (> Figure 67-5). Others systems provide variable phantoms, allowing for the testing at user defined coordinates. Each of these systems allow for the errors in localization and treatment to be separate measured separately. This is often helpful when trying to minimize system errors during instillation or in evaluating long term system drifts. If neither of these systems are available it is suggested that a commercial phantom system
. Figure 67-5 Absolute stereotactic target. This target is compatible with both the systems CT localizer as well as the linac stereotactic mounting systems. Allowing end-to-end system testing
capable of detecting sub millimeter alignment errors be employed prior to system certification. As previously mentioned the radiosurgery treatment procedure is fully reliant on the mechanical alignment of the stereotactic coordinate system. The testing of system alignment should therefore be a requirement prior to each and every radiosurgery patient treatment. One technique for such a test was initially designed by Ken Winston and Wendell Lutz and is commonly referred to as the Winston-Lutz test [3]. In this test a variable stereotactic phantom is set to the patient’s stereotactic coordinates. The phantom is then mounted onto the linac patient support system and the linac is fitted with the beam collimation system that will be used for treatment. Films are then taken at a variety of gantry patient support settings. These setting are usually selected to demonstrate the alignment of the target to the radiation beam through the range of movement that will be required to execute the patient’s radiosurgical treatment. Initially these alignments were recorded on x-ray sensitive film
Linac radiosurgery: technical aspects
and then developed. This has now been replaced by radiochromic films [8] that require no developing and by electronic portal imaging. The accuracy of the treatment can be measured by evaluating the accuracy at which the target is aligned with the center of the therapeutic beam. This test and the subsequent image analysis can usually be carried out in less then 10 min. In carrying out the above test it is recommended that the setting of the stereotactic phantom and the alignment setting of the linac be carried out by two different individuals to add a double blind verification of patient specific alignment parameters. If the test fails, which it will from time to time due to human errors in setting of stereotactic coordinates, the participants should again recheck their own work and repeat the test from scratch. It is important to provide independent verification of setup parameters during this highly precise single fraction therapy.
Stereotactic Imaging Radiosurgical procedures require that all targets be precisely imaged in stereotactic space. The three diagnostic examinations most commonly associated with stereotactic techniques are angiography, CT, and MRI. Because of the necessity to compute radiological paths during the doseplanning phase of the procedure and the need to display the estimated dosimetry on the appropriate anatomical structures, CT is a required database for each radiosurgical procedure. For vascular structures such as arteriovenous malformations, planar angiography is the gold standard for diagnoses. As will be discussed, angiography does have substantial limitations that can make precise three-dimensional planning impossible. MRI, while capable of providing a true threedimensional database, is susceptible to spatial nonuniformities. It is best to use an MRI data set only in combination with and registered to
67
a spatially precise CT data set. Each of these imaging modalities requires complete evaluation prior to use in a radiosurgical procedure.
Plane Film Angiography Angiography remains the gold standard for the identification of an arteriovenous malformation. The angiographic examination allows the clinician to follow the flow of contrast through the malformation. Through rapid sequential imaging the flow of contrast can be followed starting with the arterial phase and progressing through the shunting nidus and into the venous return. Through the 1990s plane film angiography using a sequence of images that freeze in time delivered this detailed progression of the contrast. Typically these images are taken in 0.125–0.25 s intervals. Modern systems depend upon electronic imaging, initially image intensifier technology and more recently flat solid state imaging systems, all of which substantially increased the rate of image acquisition. Throughout this technological progression images were first spatially flat, film systems, then warped when image intensifiers were introduced and now again spatially flat with the introduction of solid-state imagers. It is important to understand the technology being used and the QA and potential correction algorithms required to accurately compute stereotactic coordinates from angiographic views. The implication of orthogonal image sequencing is that when two views that best detail the nidus are chosen, they are guaranteed to be taken at different times. Because of the rapid progression of the contrast in most lesions, this shift in time results in the anteroposterior and the lateral images documenting different physical structures. This can result in a substantial discrepancy between the two orthogonal views. For three-dimensional objects, two orthogonal views cannot always provide an accurate threedimensional description. Such an object is shown
1095
1096
67
Linac radiosurgery: technical aspects
in > Figure 67-6. The concave nature of this nidus cannot be appreciated through the two orthogonal views. While it has been suggested that the use of several other noncoplanar views can aid in the reconstruction of nonspherical targets, these solutions solve only a small subset of cases. These tend to be thin, elongated structures that lend themselves to simple orthogonal correlations along their length. In general the nidus of an arteriovenous does not lend itself to such simple deconvolution. This time shift along with the inability to precisely appreciate overlapping structures results I a skew error during data evaluation [9]. . Figure 67-6 Three potential target shapes. While each target can be projected onto an orthogonal plan the resultant orthogonal images cannot be reconstructed to describe the original structure
To relate the nidus to stereotactic space, a set of fiducial markers is attached to the head ring. These markers provide reference points on each angiographic view. From these markers the exact stereotactic coordinate of the center of the spherical object can be obtained. Such a set of films is shown in > Figure 67-7. A true three-dimensional database is required to solve this problem. A contrast-enhanced CT scan can demonstrate the true three-dimensional nature of a nidus. Early CT angiographic sequences required many compromises to avoid CT x-ray tube overheating. It was common for a 1990s circa scan of the head to require 30 min. Modern multi slice scanners have little trouble providing a scan of the entire brain in less than 10 s. These rapid scan sequences coupled with maximum intensity image reconstructions have allowed CT angiography, CTA, to provide high quality true three dimensional information on the shape and location of a nidus. These scan sequences are compatible with routine stereotactic CT image acquisition.
CT Localization Computed tomography (CT) is the most reliable imaging modality for radiosurgery treatment planning. CT numbers correspond directly to electron density. This is important, because the knowledge of electron density within tissue is necessary for correctly calculating the x-ray beam attenuation characteristics. CT also provides an accurate spatial database that introduces little image distortion, and provides an accurate representation of the patient’s external contour and internal anatomy. It is best to obtain stereotactic CT images using a scanner independent fiducial localizer, such as the Brown-RobertsWells (BRW) compatible CT localizer. This localizer attaches to the stereotactic ring using three tooling balls providing a precise and repeatable
Linac radiosurgery: technical aspects
67
. Figure 67-7 Anterior and lateral angiographic view showing the nidus of an AVM along with projections of fiducial markers, allowing exact computation of target within the headings stereotactic space
fit. Since the geometry of the localizer is known relative to the head ring, stereotactic coordinates may be accurately calculated based on the CT scan of the localizer. Analyzing the nine rods provides the plane relative to the headring as well as a map of the image pixels to their ring coordinates. The characteristic N shape allows for a scanner independent calculation of stereotactic coordinates. In other words, the x,y,z coordinates of any point in space can be mathematically determined relative to the head ring rather than relying on the CT coordinates. This method provides more accurate spatial localization, and minimizes the CT scanner quality assurance requirements [10]. If a phantom base is part of the radiosurgical system, it can be used as a primary reference in CT image evaluation. While some phantoms are not configured in the proper orientation for insertion into a CT scanner, this can usually be solved by simply modifying the base of the phantom so that it can be properly mounted onto the CT patient support system. As with the evaluation of linac isocentric accuracy, the full stereotactic space should be evaluated. A table similar to > Table 67-1 should be used to determine the test points.
. Table 67-1 Positions for validating the alignment of stereotactic space and treatment delivery space Anteriorposterior mm
Lateral mm
Axial mm
CSV CSV
CSV CSV
CSV
CSV CSV
CSV
Extreme right setting Extreme left setting CSV
CSV
CSV
Extreme anterior space Extreme posterior space
Extreme right space Extreme left space
CSV Extreme anterior setting Extreme posterior setting CSV CSV
CSV Extreme superior setting Extreme inferior setting Extreme superior space Extreme inferior space
CSV – Center of Stereotactic Volume. For the BRW Systems this would be AP = 0.0, Lat = 0.0, Axial = 0.0
When using a three-dimensional database such as CT or MRI, the user should consider the size of the individual pixels during data acquisition. If, for example, a scan diameter of 35 cm is used and a matrix of 256 256 pixels is chosen,
1097
1098
67
Linac radiosurgery: technical aspects
the in-plane pixel dimension is 1.37 1.37 mm. If for the same scan diameter a 512 512 matrix isused,theindividualpixeldimensiondecreasesbya factor of four to 0.68 0.68 mm. This change doubles the resolution of the in-plane image. Similarly, if one uses thick slices during image acquisition, the out-of-plane resolution is degraded. This point can be seen by looking at the coronal reconstruction of an axially acquired data set. The coronal image shown in > Figure 67-8 has been reformatted from a transaxial data set that contained 0.68 0.68 1 mm pixels. > Figure 67-8 also demonstrates the same reformatting, except this time the data set was sampled from transaxial data
with pixel dimensions of 0.68 0.68 5 mm. As can be seen, the vertical resolution of the second image has been significantly degraded.
Magnetic Resonance Imaging Localization If an MRI-compatible stereotactic system is available, it can be tested in the same manner as the CT system. MRI has capabilities not available on CT, and MRI is susceptible to image perturbations not found in stereotactic CT scans. For example, MRI allows image acquisition in planes
. Figure 67-8 Object and two sets of 3D renderings and sagittal reconstructions. The left set are from 1 mm thick CT scans and the ones on the right are from 5 mm thick CT scans
Linac radiosurgery: technical aspects
other than transaxial. Sagittal as well as coronal images are routinely available. Also, many different MRI sequences can be used. Depending on the target tissues, T1 or T2 weighted images may be required to provide target normal tissue contrast. Each of these parameters can introduce image nonlinearities that must be evaluated. It is also important to appreciate image perturbations induced by the patient being scanned. It is often difficult to appreciate the distortions that MR scanning can introduce if direct MR scanning is used. Most MR fields of view are most uniform towards the center of the scan and less uniform towards the outer edges. When one examines the stereotactic localizers one realizes that the fiducials that are used to map imaged pixels to ring coordinates lie towards the outer edge, the region that is most likely to have spatial non-uniformities. It should also be realized that susceptibility artifacts occur in regions where large local changes in MR response are present. The fiducial markers usually have high signal values and sit in air, which has an exceedingly low MR signal. These two factors combine to produce unreliable MR stereotactic coordinate mapping. It is therefore recommended that stereotactic MRI never be used as a stereotactic database without the correlation to an identical stereotactic CT database. As a replacement to directly obtained stereotactic MRI, many systems have adopted image fusion. In this technique a nonstereotactic MRI is obtained prior to the application of the stereotactic head ring. A stereotactic CT scan is obtained after ring placement. Through one of several techniques the nonstereotactic MRI is rotated, translated, and sized to match the CT scan pixel for pixel. Once this has been accomplished, a new stereotactic MRI database can be prepared for target localization and treatment planning. > Figure 67-9 shows a split-screen view of such a matched data set. On the right side of each view is the MRI data and on the left side is the CT data. As can be seen, the internal anatomy has
67
been matched. In > Figure 67-10 a set of split views is shown at the target level.
Treatment Plan Optimization As with other treatment optimization systems, each radiosurgery planning system is based on a specific dose model. Unlike radiotherapy planning systems, which must accommodate large radiation portals, typically 5 5 cm–40 40 cm, the radiosurgical planning system’s range is approximately 4 cm down to 0.5 cm in diameter. These small fields present some special problems as well as some unique opportunities for radiosurgical beam models. The limitations on field size enable the radiosurgical systems to use relatively simple dose models. For circular fields a model that predicts field shape according to the product of tissue phantom ratios and off-axis ratios is usually sufficient. For irregular fields this model can be expanded to account for the irregular fields and field edge effects. The small fields used allow for the models to remain relatively simple as compared with those of normal external beam planning systems.
Photon Radiosurgery Paradigm Radiosurgery requires large dose deposition within the target volume and a steep dose gradient resulting in very little dose delivered to normal tissue. Clearly, the single static photon beam fails to meet these criteria. Any target that is not located at the exact depth of dmax (the depth at which a dose of radiation reaches maximum intensity) has a tissue–maximum ratio of less than 1, and the point of dmax will receive a higher dose than will the target. Since beam intensity only decreases at approximately 5–6% per centimeter up and down the beam path from isocenter, tissues near the target will receive nearly the full target dose.
1099
1100
67
Linac radiosurgery: technical aspects
. Figure 67-9 Inspection windows for judging proper alignment of CT, left and bottom, and MR, right and top, scans
The basic photon radiosurgery paradigm relies on the use of multiple tightly focused beams that use unique entry paths and converge on a point of interest. This concept is illustrated in > Figure 67-11, in which 16-mm diameter 6-MV x-ray beams converge in the center of a 17.5-cm-diameter spherical phantom. For a static beam the target (isocenter) dose is 64% of the maximum dose. With two beams, the target dose is 2 (64%) = 128% of the dose at either dmax. After normalizing to the point of maximum dose, the dose at entry dmax for either beam is only 78% of the maximum dose. Similarly, the four beams shown in > Figure 67-11c result in a peak dose of 4 (64%) = 256%, and the eight
beams in > Figure 67-11d result in 512% at the isocenter. Normalizing these to the point of maximum dose demonstrates the reduction in relative dose delivered to the entry dmax for each beam. Adding more beams that coincide only at the point of interest results in a dose distribution that is highly peaked at their point of intersection and that has a steep dose gradient outside of the target volume. One may logically surmise that the optimal dose gradient results from an infinite number of beams that irradiate the sphere from all possible entry angles (4p irradiation). Because of the physical constraints of the patient’s geometry in relation to the dose delivery systems, 4p irradiation is
Linac radiosurgery: technical aspects
67
. Figure 67-10 Inspection windows for judging alignment of Ct and MR scans taken through target region
impossible to achieve. All photon radiosurgical paradigms are attempts to irradiate isotopically typically over half of a sphere, or 2p irradiation. Linacs use multiple noncoplanar arcs focused on the target. > Figure 67-12 demonstrates commonly used 2p irradiation schemata. Almost all of the published literature detailing clinical results for radiosurgical treatments have relied on circular collimators that resultant in dose distributions are nominally spherical for the target isodose lines. For most typical plans between 5 and 11 non-coplanar arc are arranged, generally covering the available 2p. These beam arrangements allow for many hundreds of beams to be applied to the target. Each having unique
entrance and exit pathways, only overlapping over the target tissues. When fewer beams are used then less geometric beam concentration is available, each beam must supply a higher percentage of the total target dose along its entrance and exit pathway, resulting in a decrease in dose concentration and higher doses to tissues along the various beam paths. The high geometric concentration of dose not only provides a high dose concentration but the rapid divergence of beam paths provides a very steep dose decrease as one moves away from the edge of the area of geometric overlap. Typically this decrease in dose, referred to as the dose gradient, allows for the dose to drop from
1101
1102
67
Linac radiosurgery: technical aspects
. Figure 67-11 Intensity profiles of one beam, two beams, four beams, eight beams and 36 beams demonstrating the increase in concentration of dose with increasing geometric convergence of beams at the target point and divergence away from the target point
. Figure 67-12 Set of nine arcs evenly spaced
the target or prescription dose to half that dose in approximately 3–4 mm in all directions. In > Figure 67-13 a profile of the radiation dose intensity across the spherical dose distribution. From this curve one can appreciate the concept behind single isocenter radiosurgical doses being prescribed to the 80% isodose line. As mentioned earlier the radiosurgical target volume usually has a very high concentration of
tumor cells. The aim of the therapy is to provide a dose sufficient to eliminate the cells within the target. While the target volume contains almost 100% target cells the non-target volume, i.e., normal tissue volume, contains almost no target cells. Just as one aim of radiosurgery is to provide cell death to all target cells, a parallel and equally important goal is to minimize the dose to normal tissues. To achieve both goals it is advantageous
Linac radiosurgery: technical aspects
. Figure 67-13 A plot of the intensity of a single isocenter using five evenly spaced arcs. The choice of a target isodose that coincides with the steep dose fall off, i.e., the 80% of maximum dose, reduces the volume of the tissue receiving greater then one half the prescribed tumor dose
to prescribe to a point along the surface of the dose distribution where the dose is decreasing very rapidly. If such a prescription point is placed at the edge of the target volume then the shell of normal tissue exposed to the high prescription dose will be minimized. This principle is demonstrated in > Figure 67-14, which contains the above mentioned dose cross plot. If, as is shown in > Figure 67-14, the 95% isodose point is chosen as the prescription point then the dose would decrease from 95 to 47.5%, i.e., half of the prescription dose, in 5.7 mm. If, as in > Figure 67-14 the 90% isodose point was placed at the edge of the target volume then it would require 4.5 mm for the dose to decrease from the prescription level, 90%, to half that intensity, 40%. If we repeat this exercise for the 80% isodose level we see the distance is 3.8 mm and the 70% isodose level we see that 3.9 mm is required. It can therefore be seen that when the 80% isodose point is chosen for the prescription isodose level the thickness of the shell between target dose and half target dose is minimized. When non-spherical targets are encountered then the single isocenter treatment technique is
67
expanded to utilize not a single large sphere to cover the target while including normal tissues but rather the packing of the target volume with multiple spheres covering or packing the volume, with spherical distributions. In general the spheres are arranged such that the outside of the target volume is aligned with the outer edge of various spheres. This allows the high dose gradient to be aligned with the target-normal tissue interface. This provides prescription dose to the target while minimizing the dose to normal tissues. There is only one caveat when using multiple isocenters and that is the shift in prescription isodose level from 80 to 70% of the maximum dose in the target volume. This shift is the result of very small hot spots that result from spheres being pack next to one another. Since these small hot spots are within the target volume and as we have mentioned all target cells are to receive doses that result in cell death no negative clinical results have ever been reported due to these hotspots. For many spherical targets a single isocenter is sufficient to provide 100% target coverage with exceedingly little normal tissue contained within the prescription isodose shell. The next level of complexity is for targets that are elongated. Many acoustic neurinoma have relative circular portions that are adjacent to the brain step and a section that extends down the internal auditory canal. This shape can easily be covered by two isocenters. > Figure 67-15 shows such a target and prescription isodose. In this case the first, a 14 mm isocenter covering the more medial spherical portion of the target volume and a 10 mm centered towards the lateral extent of the target. The combination of these two spherical isocenters results in a very conformal treatment. As mentioned earlier the technique of sphere packing allows the edge of the sphere, which has a very steep dose gradient to be aligned with the edge of the target volume. In this case the first isocenter is placed with this steep dose
1103
1104
67
Linac radiosurgery: technical aspects
. Figure 67-14 A set of graphs demonstrating that selecting the prescription isodose at the 80% of max dose provides a more rapid fall off then selecting the 95, 90 or 70% of the maximum dose points
edge providing the steep gradient in the direction of the brain stem. The region where the twoisocenter plans present the previously mentioned hot spot is along the line that joins the two isocenters. As mentioned this is always within the target volume and has not been shown to result in any treatment complications. Examining this distribution further one finds that the target volume is 0.8 cc and the prescription isodose volume is 0.9 cc providing only 0.1 cc of normal tissue within the prescription treatment volume. One can also determine that the volume of the isodose shell that encompasses one-half the treatment isodose value, in this case the prescription was 1250 cGy to the 70% isodose shell
so the one-half treatment shell would be the 35% isodose shell which would deliver 625 cGy, is only.95 cc. This shell is on average 4.1 mm from the treatment isodose shell. This means that the on average the dose decreases from 1250 to 625 cGy in less then 4.1 mm. This same principle can be applied to significantly more complex targets. > Figure 67-16 demonstrated the application of multiple isocenters to cover a complex target with adjacent critical structures. The important parameters in developing the above plan are the selection of the correct collimators, the correct placement of the isocenters and the relative intensities of the isocenters. The planning system should
Linac radiosurgery: technical aspects
have tools to allow these optimization parameters to be readily manipulated and optimized. > Figure 67-17 shows a planning system screen with the tools required for isocenter placement, spacing and weighting.
. Figure 67-15 Isodose distribution for the treatment of a nonspherical acoustic neurinoma. Demonstrating the typical degree of conformality and dose gradient obtainable using two correctly placed, spaced and weighted isocenters
67
The above examples are can be planned on either a fully merged CT-MR dataset or preplanned on a non-stereotactic MR prior to ring placement. This function allows the clinician to obtain the MR targeting scan days before the planned therapy and to plan the treatment at their leisure. On the day of treatment the stereotactic headring can be rigidly fixed to the patient, the preplan transferred to the fused dataset and rebalanced. This process can easily be automated, resulting in a planning session on the day of therapy that is simply an inspection of the previously-approved plan. Using this process the isodose coverage at the time of preplanning is identical to the coverage after plan transfer and rebalancing. The described arcing delivery paradigm is one method of providing the requisite number of noncoplanar beams that converge precisely at a target point and rapidly diverge while traveling through non tissue tissues. As previously mentioned Leksell [11] had developed a dedicated radiation delivery unit that achieved almost identical spherical dose distributions, with accompanying steep dose gradients. The most widely commercialized design consisted of 201 non-coplanar pencil radiation
. Figure 67-16 Multi-isocenter circular cone plan demonstrating the ability of the technique to not only provide a conformal plan but to also maintain a steep dose gradient
1105
1106
67
Linac radiosurgery: technical aspects
. Figure 67-17 Screen showing the tools used to produce the plan in figure 15. Isocenter placing, weighting and spacing tools are interactively used to optimize the dose distribution
beams. While 201 beam provided almost identical dose distributions as the five non-coplanar arcs spread out over nearly 2p it has been demonstrated that an equivalent distribution can be delivered from a few as 15 beam spread across the same geometry [12]. Alternate approaches for providing the needed high target conformality as well as providing normal tissue sparing have used a technique that modulates the intensity of the radiation beam, IMRT. One of the early explanations of intensity modulation suggested that this could be achieved through the previously described sphere packing technique [13], however, iterative dose
optimizations have been used to implement a majority of the commercial intensity modulation algorithms [14]. The use of IMRT has found wide clinical acceptance in larger field radiation dose delivery. While IMRT has shown to provide equivalent conformal target coverage, it has also been shown to provide overall shallower dose gradients [15,16] when compared to multiple isocenter circular beam techniques. This may in part be due to the tradeoff of fewer beams being utilized. The fewer beams are in part due to the computation time required for each fiend, the QA required per IMRT beam or the simple extrapolation of an adequate number of larger fields to a
Linac radiosurgery: technical aspects
small field technique. The previously reference work of St John [12] indicates that utilizing fewer then 15 non-coplanar beam paths will yield this lower dose gradient and higher doses to normal tissues. A solution to the initial IMRT problem posed by Barth [13] has been demonstrated by St John. This planning and delivery technique provides the radiosurgeon with the conformality and gradient benefits of multiple isocenter planning, combined with the speed of single isocenter delivery. When this approach is combined with the automated multi isocenter planning, as demonstrated by Wagner [17] plans as demonstrated in > Figure 67-16 can be derived in under 1 min of cpu planning time and delivered in under 20 min. Aside from the general desire to reduce dose to normal tissue the 12 Gy dose volume has been shown to correlate with permanent radiation induced complications [18–22]. With the upper dose for many radiosurgical targets being in the 20 to 22.5 Gy range it is critical to provide a very steep dose gradient, one that decreases the target dose to approximately one half the target dose value over a short distance. With the critical parameters being conformality and dose gradient a simple metric for scoring radiosurgical plans has been suggested [23]. This metric is comprised of two scores, one for inclusion of normal tissue in the prescription isodose shell and a second that measures the average dose gradient between the prescription isodose shell and the shell that encloses the one-half the prescription isodose value. This is the gradient previously discussed and shown in > Figure 67-15. When multiple sets of rival stereotactic radiosurgery plans were ranked with respect to this single score index, the resulting plan rankings closely matched the plan rankings according to biologic indices (calculated nontarget brain normal tissue complication probabilities). When applied to radiosurgery treatment outcomes have been shown to have sensitivities
67
to not only conformality and dose gradients but also the time required to deliver the treatment. It has been demonstrated that when treatment times extend beyond 15–20 min that sub lethal damage repair begins to decrease the overall effect of the prescription dose [24,25]. For multiple isocenter planning and treatment delivery this is usually not a problem. Although a multiple isocenter plan may require over an hour to fully deliver each individual isocenter seldom requires more then 20 min of dose delivery time. So although the total treatment delivery to the acoustic neurinoma shown in > Figure 67-15 may require 25 min the cells covered by each of the two isocenters receive their full dose in less then 15 min. Once delivered moving to the next isocenter then treats the next subsection of the target, again in less than 20 min. This may not be the case if complex modulations are required or if extended alignments are required between non-coplanar beam sets.
Treatment The final treatment plan may have several isocenters each requiring a set of arcs and one or more collimators. As previously discussed the setup of the treatment unit and setting of the patient specific target coordinates is best carried and tested using a double-blinded phantom test. It is also very helpful to have a checklist that walks the treatment team through the entire procedure from first step to last. > Figure 67-18 shows several pages of the checklist for the above detailed two-isocenter treatment. It is suggested that at a minimum of three treatment staff be involved in the setup and treatment process: Two staff to setup of the linac and phantom for initial testing plus a third team member whose sole purpose is not to setup the patient or the linac but to observe and provide a blinded check of all settings and procedures.
1107
1108
67
Linac radiosurgery: technical aspects
. Figure 67-18 Printout of single isocenter plan. This printout is used to guide the QA during treatment delivery. Each treatment team member checks that the treatment plan is precisely followed
References 1. Leksell L. The stereotaxic method and radiosurgery of the brain. Acta Chir Scand 1951;102:316-19. 2. Colombo F, Benedetti A, Pozza F, et al. Stereotactic radiosurgery utilizing a linear accelerator. Appl Neurophysiol 1985;48:133-45. 3. Winston KR, Lutz W. Linear accelerator as a neurosurgical tool for stereotactic radiosurgery. Neurosurgery 1988;22:454-64. 4. Friedman WA, Bova FJ. The University of Florida radiosurgery system. Surg Neurol 1989;32:334-42.
5. Bova FJ, Meeks SL, Friedman WA, Buatti JM. Optic-guided stereotactic radiosurgery. Med Dosim 1998;23:221-8. 6. Brown RS. A computerized tomography–computer graphics approach to stereotaxic localization. J Neurosurg 1979;50:715-20. 7. Friedman WA, Buatti JM, Bova FJ, et al. LINAC Radiosurgery – a practical guide. Berlin: Springer; 1998. 8. Mack A, Czempiel H, Kreiner HJ, Du¨rr G, Wowra B. Quality assurance in stereotactic space. A system test for verifying the accuracy of aim in radiosurgery. Med Phys 2002;29(4):561-8. 9. Siddon RL, Barth NH. Stereotaxic localization of intracranial targets. Int J Radiat Oncol Biol Phys 1987; 13:1241-6. 10. Huh SN. Incorporation of magnetic resonance imaging and digital angiography in the application of stereotactic radiosurgery. Gainesville: University of Florida; 1994 (dissertation). 11. Leksell L. Cerebral radiosurgery. I. Gammathalanotomy in two cases of intractable pain. Acta Chir Scand 1968;134(8):585-95. 12. St John TJ, Wagner TH, Bova FJ, Friedman WA, Meeks SL. A geometrically based method of step and shoot stereotactic radiosurgery with a miniature multileaf collimator. Phys Med Biol 2005;50(14):3263-76. 13. Barth, NH. An inverse problem in radiation therapy. Int J Radiat Oncol Biol Phys 1990;18(2):425-31. 14. Webb S. Advances in treatment with intensity-modulated conformal radiotherapy. Tumori 1998;84(2):112-26. 15. Nakamura JL, Pirzkall A, Carol MP, Xia P, Smith V, Wara WM, Petti PL, Verhey LJ, Sneed PK. Comparison of intensity-modulated radiosurgery with gamma knife radiosurgery for challenging skull base lesions. Int J Radiat Oncol Biol Phys 2003;55(1):99-109. 16. Sankaranarayanan V, Ganesan S, Oommen S, Padmanaban TK, Stumpf J, Ayyangar KM. Study on dosimetric parameters for stereotactic radiosurgery and intensity-modulated radiotherapy. Med Dosim 2003;28 (2):85-90. 17. Wagner TH, Yi T, Meeks SL, Bova FJ, Brechner BL, Chen Y, Buatti JM, Friedman WA, Foote KD, Bouchet LG. A geometrically based method for automated radiosurgery planning. Int J Radiat Oncol Biol Phys 2000;48(5):1599-611. 18. Flickinger JC, Kondziolka D, Maitz AH, Lunsford LD. Analysis of neurological sequelae from radiosurgery of arteriovenous malformations: how location affects outcome. Int J Radiat Oncol Biol Phys 1998;40(2):273-8. 19. Flickinger JC, Kondziolka D, Lunsford LD, Pollock BE, Yamamoto M, Gorman DA, Schomberg PJ, Sneed P, Larson D, Smith V, McDermott MW, Miyawaki L, Chilton J, Morantz RA, Young B, Jokura H, Liscak R. A multi-institutional analysis of complication outcomes after arteriovenous malformation radiosurgery. Int J Radiat Oncol Biol Phys 1999;44(1):67-74.
Linac radiosurgery: technical aspects
20. Flickinger JC, Kondziolka D, Lunsford LD, Kassam A, Phuong LK, Liscak R, Pollock B, Arteriovenous Malformation Radiosurgery Study Group. Development of a model to predict permanent symptomatic postradiosurgery injury for arteriovenous malformation patients. Int J Radiat Oncol Biol Phys 2000;46(5):1143-8. 21. Korytko T, Radivoyevitch T, Colussi V, Wessels BW, Pillai K, Maciunas RJ. Einstein DB. 12 Gy gamma knife radiosurgical volume is a predictor for radiation necrosis in non-AVM intracranial tumors. Int J Radiat Oncol Biol Phys 2006;64(2):419-24. 22. Friedman WA, Bova FJ, Bollampally S, Bradshaw P. Analysis of factors predictive of success or complications in arteriovenous malformation radiosurgery. Neurosurgery 2003;52(2):296-307; discussion 307–8.
67
23. Wagner TH, Bova FJ, Friedman WA, Buatti JM, Bouchet LG, Meeks SL. A simple and reliable index for scoring rival stereotactic radiosurgery plans. Int J Radiat Oncol Biol Phys 2003;57(4):1141-9. 24. Shibamoto Y, Ito M, Sugie C, Ogino H, Hara M. Recovery from sublethal damage during intermittent exposures in cultured tumor cells: implications for dose modification in radiosurgery and IMRT. Int J Radiat Oncol Biol Phys 2004;59(5):1484-90. 25. Tomita N, Shibamoto Y, Ito M, Ogino H, Sugie C, Ayakawa S, Iwata H. Biological effect of intermittent radiation exposure in vivo: recovery from sublethal damage versus reoxygenation. Radiother Oncol 2008;86(3):369-74.
1109
57 Overview of Radiosurgery Technology M. Schulder
Introduction Stereotactic radiosurgery (SRS) is still often referred to as a new or ‘‘high tech’’ treatment. And yet it is 57 years since Lars Leksell invented the concept and coined the term radiosurgery [1]. After experimenting with orthovoltage irradiation and considering heavy particle use, Leksell invented the Gamma Knife (GK) in 1967 [2]. In the early 1980s, Betti and Derechinsky, and Colombo, et al. adapted linear accelerators (Linacs) to deliver photon based SRS [3–6], thereby ushering in the second generation of SRS devices. Linacs offered a promising alternative form of external beam radiotherapy to the gamma knife [7–9]. Improvements in imaging and especially in computer power led to ever-increasing sophistication, speed, and clinical use of SRS. And yet, what had not changed for nearly 50 years was the concept that delivering SRS required the use of isocenters and either a Linac or a GK to deliver a conformal treatment with a steep dose gradient. This notion has been termed ‘‘sphere packing’’ or the use of multiple spherical isocenters created by circular collimators for irradiating irregularly shaped lesions [8]. SRS, invented as a single-session technique of minimally invasive neurosurgery utilizing ionizing radiation, was understood to be completely different from standard, fractionated methods of radiation therapy. In the late 1990s this paradigm began to change. Radiation oncologists came to appreciate the value of stereotactic localization,
#
Springer-Verlag Berlin/Heidelberg 2009
now available without the use of rigid frames, and neurosurgeons came to understand the potential benefits of fractionation [10–12]. Though each treatment was distinct, to some degree the line between SRS and stereotactic radiotherapy (SRT) became blurred. Studies confirmed that fractionated dose delivery of SRT offers the combined benefits of stereotactic targeting and minimal radiobiologic harm to normal tissue [13,14]. SRS treatment plans still should be based on a set of quantitative parameters. Dosimetry indices for SRS planning include the coverage index (the isodose surface that covers the entire target volume as a fraction of the prescription isodose); the homogeneity index (the ratio of the maximum dose in the target volume to the prescription isodose); and the conformity index (the ratio of the prescription isodose volume to the target volume) the last two of which should be less than two [15,16]. The conceptual changes noted above, together with great leaps in the technology of radiation delivery, have transformed the radiosurgical landscape. Rapid innovations in Linacbased devices, and to some extent in charged particle SRS technology, are redefining the field [17]. The implementation of multiaxial robotic technology in certain Linac machines has greatly refined the integrative capabilities of SRS devices in all aspects of planning and treatment [3,7,18]. This chapter focuses mainly on the newer, commercially available radiotherapy applications of Linac and GK, the most common forms of SRS worldwide.
868
57
Overview of radiosurgery technology
Linear Accelerator SRS Linacs produce a radiation beam with a magnetron and gantry. Accelerated electrons generated from the magnetron are injected into an evacuated copper tube. There, they collide with a heavy metal target (typically tungsten) to produce photons. The scattered photons are subsequently focused through a portal to create a directed, or collimated, beam of gamma radiation. The X-rays produced by modern Linac electron guns usually have higher energies (4–6 MeV) than those generated through isotope decay [19]. The Linac is mounted to a gantry that rotates through an arc around the patient. This frame-gantry geometry of Linacs is suitable for treating lesions throughout the body. The circular gantry movement lends itself to the development of Linac based SRS, wherein the use of multiple noncoplanar arcs rotated around an isocenter deposit a high dose of radiation at that isocenter in essentially an additive fashion. Treatment volume could be altered by changing collimator diameter. Adjusting the spherical shape of a typical unmodified treatment would be done by adding isocenters, and to a lesser extent by adjusting beam length and weight. Recent technological advances in the dosimetry and physics of SRS have led to a new generation of treatment planning software that has simplified and increased the versatility of Linacbased SRS. The newest generation of Linac-based SRS devices includes X-Knife (Radionics, Burlington, MA), Novalis (Helmstetten, Germany), Trilogy (Varian Medical Systems, Palo Alto, CA), and CyberKnife (CK) (Accuray, Sunnyvale, CA). Compared to the GK, Linac sources offer a wider array of treatment options for both traditional SRS and fractionated stereotactic radiotherapy treatment [17]. Other technical and electronic innovations have further enhanced the accuracy and precision of the newest versions of Linac radiosurgical devices. Nonetheless, the mechanical complexity of these devices demands that strict
quality control measures be implemented and routinely followed [17]. The finely collimated X-ray beams from Linac sources can be effectively targeted to both intracranial and extracranial lesions. Replicable data suggest that planning and treatment of Linac SRS is comparable overall to those of the Gamma Knife [17,20]. The physical characteristics of Linac devices render them capable of irradiating extracranial lesions such as tumors of the spine, prostate and lung [17,21]. Treatment of each isocenter can usually be performed in less than 10 min on a Linac-based SRS device [19]. Linac systems can operate using conventional circular arcs, conformal SRS based on multiple static-shaped beams, dynamic arc SRS employing micromultileaf collimator (MMLC) field shaping, and intensity modulated radiotherapy (IMRT). While all of these techniques can be implemented, MMLCs lie at the core of modern Linac-based applications for SRS [19].
Micromulitleaf Collimators and Inverse Planning Multileaf collimation for beam shaping and delivery originated in Japan in the 1960s [19]. By the 1990s, medical physicists and clinicians had learned to shape focused radiation to conform to irregular concave surfaces while decreasing the dose reaching normal tissues. This approach involved modulating the intensity of the radiation beam within the field of radiation, and it signaled the advent of IMRT. The technique was later refined with the use of a computercontrolled, flexible high resolution beam shaping device called a multileaf collimator (MLC). MLCs are used to shape the radiation beam during IMRT. The micro-multileaf collimator (MMLC) was eventually designed for smaller intracranial lesions that could not be treated with beams transmitted through collimators with wider leaf widths [3,19].
Overview of radiosurgery technology
SRS treatment planning evolved as an iterative, forward planning process. Using a computer model based on stereotactic computed tomography (CT) and/or magnetic resonance imaging (MRI), clinicians and physicists place isocenters in the target and examine such features as conformality, dose gradient, and dosing of adjacent areas. This process can be very time consuming when planning SRS using circular collimators for patients whose lesions have complex shapes and, in particular, are close to such critical structures as the optic chiasm or brainstem. Inverse treatment planning evolved as a numerical optimization strategy for finding optimizing beam directions during SRS and IMRT procedures. In 1996 Carol et al. described their experience with the Linac-based Peacock system, designed to deliver highly conformal radiation using inverse planning and MMLC [22]. The Peacock system was used for fully fractionated radiation therapy (RT), and required implantation of skull screws for repeat fixation. While not truly ‘‘radiosurgical,’’ this invention set the stage for what has become an increasingly common and powerful concept in SRS: the use of the MMLC. The MMLC contains many small vanes that change position as the Linac rotates. Consequently, the need for ‘‘sphere packing’’ is avoided, and even complex shapes can be treated without the need to reposition the patient. Only target volume, the risk for critical volumes, and other clinical considerations need limit the use of SRS. In essence, by using an MMLC and an inverse planning algorithm, clinicians can shape the dose such that it has conformality for the target lesion. When used with either an IMRT or nonisocentric beam, the inverse planning algorithm spares, or ‘‘wraps around,’’ around healthy tissue [21]. Yet, planning these treatments in a ‘‘traditional,’’ forward manner is a formidable task. By taking advantage of modern computer speed and power, however, the treatment team can define key parameters such as the target and the dose
57
tolerance of adjacent structures. Technologically sophisticated Linac-based SRS systems, including some with computer-controlled robotic functions for MMLC performance, generate a plan based on these constraints. The physicians and physicists can adjust the plan by changing various parameters of the treatment plan, depending upon the core features and options available through the particular SRS system software. This section will discuss several systems that bring new technology to bear on SRS, each from a different perspective.
Adapted Linac: Radionics XKnife Description The Radionics XKnife (Integra Radionics, Burlington MA) was the first dedicated Linac for SRS, developed in the 1980s [9] and made commercially available shortly thereafter. It is a 3-dimensional (3-D) treatment planning software and hardware system meant to adapt a conventional Linac for SRS and SRT. Before the advent of the MMLC, radiosurgical plans and delivery were realized using multiple, non-coplanar arcs. The Linac rotates around a fixed isocenter and target shaping was done by changing couch angle and to a lesser extent by adjusting arc length and weighting. This system was used with excellent results [23]. However, as treatment technology evolved, the limitations of the couch mount system spurred certain technological advances. In particular, conformal treatments of nonspherical targets were difficult to achieve with the couch mount, as isocenter changes were time-consuming and required multiple steps to verify. This was an obstacle to overcome, especially for patients needing SRS for benign tumors or functional disorders. The current XKnife version features tighter tolerances for isocentric rotation and fixed primary and collimation units that minimize
869
870
57
Overview of radiosurgery technology
gantry sag. The basic principle of a rotating gantry that irradiates X-rays to the patient lying on a couch during the procedure has remained unchanged. However, the system has evolved with major innovations such as treatment planning algorithms, fractionated SRT treatment protocols, and a relocatable frame. The XKnife is compatible with most Linacs and requires only simple mounting and dismounting. Components of the latest XKnife include stereotactic localizers and immobilizers, SRS collimators, a MMLC, and instrumentation for enhanced quality assurance. The precision and accuracy of the XKnife treatments depend totally on the rotational accuracy of the treatment couch and gantry [24], both of which are computer-controlled by the XKnife software. The device contains advanced algorithms for the localization, planning, setup, and delivery of radiation therapy without the need for separate software. In addition to treatment planning and SRS, conformal MMLC uses are included.
controlled by the XKnife Version 4 IMRT software. Using equipment already on the linear accelerator, the XKnife Version 4 IMRT customizes dose delivery for treating brain tumors and vascular malformations as well as head, neck, spine, and body tumors. With RT real-time planning, the XKnife Version 4 is equipped to deliver collimated beam radiation therapy in frameless stereotaxy [25]. Cranial accessories for the XKnife include the standard BRW head ring, the GTC relocatable head ring, and the TLC pediatric head ring. These devices are used for precise fixation, localization, and repositioning of targets in the head and neck. Target sites in the spine and body are immobilized and localized for fractionated SRT using a vacuum-based stereotactic system. The Radionics HDRT is a comprehensive platform that manages patient repositioning and internal anatomical shifts during extracranial frameless stereotactic treatments. The HDRT integrates Radionics stereotactic localizers, XKnife IMRT, Linac-based CT imaging, and Radionics stereotactic ImageFusion [25].
Performance Features During SRS or SRT, the XKnife MMLC provides greater conformality than do circular collimators or blocks. This hybrid device contains 62 leaves that can be opened to various widths along the rotational arc of the gantry, as dictated by the treatment plan. The uniform leaf width of 4 mm at the isocenter is appropriate for finely shaping small and large field patterns. Through computerized control of the motorized leaves of the MMLC, the clinician can maximize the delivery of dose to the target volume with minimal radiation to nearby normal tissue. The MMLC has a large field size of 10 cm 12 cm for treating irregular tumor volumes, including lesions in the brain, head and neck, and spine. Other advanced optimization techniques providing increased dose conformality are also
Evidence-based Studies on Quality and Efficacy of the XKnife Urie et al. compared the Radionics circular collimators with the MMLC XKnife treatment planning system to determine any differences in tissue maximum ratios (TMRs). Beam data, including TMRs, output factors, penumbrae, and isodose distributions of these fields were measured. The mechanical accuracy of the MMLC compared well with that of the circular collimators, while also producing a smaller leakage dose. Dose distributions were basically equivalent for circular collimators and for simulated circular fields (total, 360) of four arcs of the MMLC. The resulting TMRs were the same as those of circular collimators of equivalent size.
Overview of radiosurgery technology
However, since the MMLC is bulkier than the circular collimators, mechanical collisions may occur [26]. Recent clinical data support the technical efficiency and safety of the Radionics SRS system. Hillard et al. conducted a retrospective study of 10 patients undergoing multiple SRS treatments with the XKnife for multiple brain lesions. Each patient received at least two treatments delivered to at least three isocenters with a minimum follow-up period of 6 months. The average of the maximum doses directed to a point within the whole brain and or a critical brain structure varied widely, ranging from 159 cGy to the left optic nerve to 2,402 cGy within the brain. The authors found no complications that could be attributed to the SRS, except for one patient who developed seizures linked with radiation necrosis. Hillard et al. concluded that multiple SRS treatments at the cumulative doses delivered in this study were safe for patients with multiple brain lesions [27]. Plowman and Doughty compared dose gradient ‘‘fall-off ’’ at the margin of the target (the distance between isodoses) between the Gamma Knife and XKnife. Both techniques produced similar values for isocenter treatment volumes up to 1.5 cm diameter, but these results diverged when planning treatments for larger and more complex targets. While more frequent multiple isocenter ‘‘shots’’ of the gamma unit produced greater conformity in complex target volumes, it did so at the cost of steeper internal dose gradients. For instance, in the treatment of an acoustic neuroma (vestibular schwannoma), the XKnife generated a considerably smaller internal dose gradient (11%) than in the GK (100%) (28). That said, the clinical significance of this dose heterogeneity is uncertain. The authors suggested that this may have contributed to improved hearing preservation in patients with acoustic neuroma who have been treated with XKnife SRT [28]. Other evidence also supports
57
the safety and efficacy of the XKnife for acoustic neuromas. McClelland evaluated 20 patients who received fractionated SRT (54 Gy in 1.8 Gy daily fractions) via the XKnife 4.0 3-D planning system for the treatment of acoustic neuromas. Local tumor control occurred in each patient and there were no complications All nine patients who had preserved hearing prior to treatment had sustained hearing preservation at the last follow-up. Four patients experienced decreased hearing after treatment [29]. > Figure 57‐1 shows a conformal plan for single session SRS with the XKnife. Experience with the XKnife indicates that adapting a regular Linac with inverse planning SRS software and an MMLC allows for the full gamut of single session and fractionated stereotactic treatments to be delivered with safety and efficacy. This may indeed be an attractive option for centers looking to upgrade an older SRS system or to install a new one, in that they may not want to justify the installation of a dedicated radiosurgical device.
Dedicated Isocentric Linac for SRS: Novalis Tx System Description The Novalis SRS system was among the first devices in a new generation of 6-MeV electron Linacs that emerged during the mid 1990s [30]. The Novalis Tx system, a joint venture of a stereotactic neurosurgical vendor (Munich) and a Linac manufacturer (Varian, Palo Alto, CA) utilizes a dedicated Linac, MMLC for cone-based beam shaping, and a treatment planning system, which are all integrated through an information management system. Designed for rapid and high dose delivery, the Novalis Tx is used to treat lesions that range widely in size and shape. As in the case of Linac-based radiotherapy
871
872
57
Overview of radiosurgery technology
. Figure 57‐1 XKnife treatment plan for a 54-year-old man with a left acoustic neuroma. The 1,250 cGy prescription dose is conformal, and the plan has a craniocaudal orientation to minimize irradiation of the brainstem
devices in general, this system is designed for both single, high-dose sessions in SRS and multiple, smaller dose sessions in SRT [31].
Performance Features The Novalis Tx can generate radiation fields from a minimum of 3 mm 3 mm to a maximum of 98 mm 98 mm in size. It can rapidly deliver multiple beam energies of up to 18 MeV for targeting deep-seated lesions. The dynamic, computer-controlled, fine ‘‘beam shaping’’ capabilities of the MMLC are able to precisely match the contour of the target lesion. Full doses of radiation can be delivered to tumors with varying complicated shapes while sparing irradiation to nearby healthy tissue or critical structures such as the brainstem or spinal cord [31]. After the path to frameless SRS was blazed with the Cyberknife (see below), the Novalis system was adapted to allow for SRS in one or more
fractions without the use of a stereotactic frame [31]. The ExacTrac X-Ray 6D imaging system employs infrared tracking to monitor patient movement during treatment. Similar tothe CK registration system, this imaging technology consists of two X-ray units recessed into the Linac room floor and two flat-panel silicon detectors mounted to the ceiling (> Figure 57‐2). Fluoroscopic images are registered to digitally reconstructed radiographs, based on pre-treatment CT scans. Any malalignment is corrected by movement of the couch until precise registration is confirmed, after which treatment is resumed. The adaptive respiratory gating module, another feature of the ExacTrac system, tracks the patient’s breathing in real time. The respiratory gating software activates the Linac beam only when the lesion is at the isocenter position for radiation delivery. Regardless of patient body movement, the Novalis robotics-driven systems are able to direct a maximum, effective dose of radiation to quickly penetrate the entire target
Overview of radiosurgery technology
. Figure 57‐2 Patient being prepared for Novalis SRS
anatomical site from multiple angles. This process allows for accurate spinal SRS to be delivered without the need for fiducial marker insertion. While the Novalis Tx does not constitute a robotics-based Linac SRS device in a strict sense, its robotic functions are pivotal to performing complex set-ups during both stereotactic (SRS and SRT) and IMRT treatments of cranial, head and neck, or spine indications.
Evidence-based Studies on Quality and Efficacy of Novalis SRS Devices In five patients receiving IMRT treatment for either recurrent pituitary tumors or meningiomas, the Novalis-based pretreatment technique helped to streamline and optimize the treatment plan before fitting patients with a head ring [32]. Wurm et al. reported that the Novalis image-guided noninvasive SRS was accurate and viable for treating intracranial
57
benign and malignant lesions [31] (> Figure 57‐3). For 2-mm treatment planning CT slice thickness, the hidden target test (phantom measurements) for the ExacTrac robotics module revealed an overall system accuracy of 1.04+/0.47 mm for the Novalis device after stereoscopic X-ray verification and correction. The overall average translational setup error was 0.31+/0.26 mm and the rotation error was 0.26+/0.23 . These values are comparable with verification and correction of the Cyberknife [31,33,34] Pedroso et al. evaluated Novalis shaped beam and IMRT delivered as a single dose (SRS, hypofractionated radiotherapy (HSRT), or a fractionated regimen (FSRT) in eight patients with chordoma. Treatment resulted in local tumor control in all cases, although followup was only about 26 months overall. Clinical assessments were based on volumes and dose range for optic apparatus and brainstem. In the FSRT group, two lesions disappeared, one diminished in volume, and the other was unchanged. The authors suggest that the biological effects and treatment outcome of Linac-based radiation are equivalent, whether delivered from a gantry or a robot [35]. Other studies also confirm the accuracy and versatility of the BrainLab Novalis system for both frame-based and frameless image-guided, cranial and extracranial SRS, including functional treatments [36]. In framed SRS, a Novalis Linac device attached to a 4-mm collimator was safe and effective for treatment of trigeminal neuralgia in 32 patients treated during a 12-month interval. Clinical outcomes compared well with those of the Gamma unit [37]. In a study of the geometrical accuracy of Novalis SRS, Rahimian, et al. showed that positioning errors of the Novalis system were less than 1 mm in all axes [38]. They stressed the need for careful quality assurance (QA) procedures to ensure that this level of accuracy was reached, especially before treating patients with the high and critically placed doses needed for the relief of trigeminal neuralgia. Novalis-based stereotactic irradiation has also been used to treat spinal or paraspinal lesions
873
874
57
Overview of radiosurgery technology
. Figure 57‐3 Novalis treatment plan beam’s eye view for SRS in a patient with acoustic neuroma, showing planned position of MMLC vanes as gantry rotates around isocenter
such as Ewing sarcoma, metastatic carcinoma and adenocarcinoma, hemangioblastoma, vascular malformations, schwannoma, meningioma, and chordoma [39]. The Novalis system is a powerful dedicated Linac for SRS and SRT that can also be used for IMRT and other conformal treatments in a standard fractionation regimen. Users may enjoy this flexibility but as in any system must be diligent in maintaining strict levels of QA to ensure safe and effective treatments.
Varian Trilogy Image-guided Radiotherapy Platforms Description The Trilogy device (also produced by Varian) epitomizes the incorporation of stereotaxis into an overall RT planning system. Trilogy IGRS consists of a Linac, two gantry/emission heads, beam
collimators, robotic motorized arms mounted on the imaging device, a multi-modal volumetric imaging system, and a remote control operated treatment couch [3,15] (> Figure 57‐4). Setup error and organ motion during treatment are tracked via image-guided motion management techniques for immobilization and positioning. High quality digital image data generated through intraoperative 2D imaging techniques of radiography and fluoroscopy and 3D conebeam CT are used to precisely locate target lesions with submillimetric accuracy. The Trilogy imaging management system also interfaces with CT/PET scanners [15,31,40].
Performance Features Setup and monitoring of the patient and target position are performed in real time for each treatment beam during a Trilogy-based procedure.
Overview of radiosurgery technology
. Figure 57‐4 Trilogy gantry and integrated digital imaging system (courtesy of Varian Medical Systems)
An optically guided frameless device is used for immobilization and relocalization. This consists of an optical fiducial array that is attached to a customized bite block of the patient. The Trilogy optical imaging system then determines the patient’s position and motion by tracking the optical markers in real time throughout treatment. Optical markers may be active infrared light-emitting diodes (IRLEDs) or passive reflective spheres [15]. The efficiency of Trilogy is aided by the OnBoard Imager (OBI), a gantry-mounted imaging system controlled by robotic technology and integrated software control. It consists of a highperformance, low energy kilovoltage (kV) X-ray imaging source, a large-area flat-panel amorphous silicon (aSi) digital X-ray detector, and two robotic arms that independently position the kV source and OBI at right angles to the treatment beam. High resolution X-ray images produced during treatment via the OBI are on par with those used for treatment planning [3,15]. To maximize image-guided performance, Varian has coupled the OBI with the Trilogy accelerator for fluoroscopic, radiographic, and volumetric cone-beam CT imaging and delivery within a single machine. CT images, produced
57
from very fine slices, provide key reference images [15]. To some extent this parallels the technology developed for Tomotherapy (see below), and points to the increasing incorporation of high quality digital imaging in radiation delivery systems. The Trilogy system allows for tracking and stereotactic treatment of extracranial targets without the need for fiducial marker insertion. Optimal immobilization is achieved using both physical techniques and software methods. A complementary or alternative approach utilizes Trilogy Respiratory Gating. This system electronically monitors the beam, controlling beam activity in response to the patient’s own respiratory waveform. Trilogy Respiratory Gating uses an infrared camera that tracks a passive marker block placed on the patient’s chest or abdomen [3,15]. Any changes detected in the patient’s normal breathing pattern trigger a gating mechanism that turns the radiation beam on or off. The beam does not go on unless the lesion falls within the planned treatment target area [3,15]. Besides SRS, Trilogy offers several external beam therapies, including 3-Dimensional conformal radiation therapy (3D CRT), IMRT, IGRT and ‘‘Dynamic Adaptive Radiotherapy’’ (DART). Minute lesions can be stereotactically targeted, due to the tight isocentric alignments on three axes rendered in part by the robotic components. The isocentric accuracy of the beam on the gantry and two collimator axes is 0.5 mm or less. The isocenter radius of the couch rotational axis is 0.75 mm or less [15]. Equivalent dose distributions are attained by creating as many beam directions as possible and by managing beam sizes. Geometrically optimized noncoplanar beam arrangements (the ‘‘bouquet of beams’’) and circular collimators or MLCs with narrow leaf sizes limit the size of beam apertures. The result is a high conformal dose distribution with rapid dose fall-off for healthy surrounding tissues in all three dimensions. The Trilogy system can operate either in dynamic
875
876
57
Overview of radiosurgery technology
mode (‘‘sliding window’’) through an MMLC, or in segmental mode (‘‘step-and-shoot’’). The dynamic mode confers a high degree of spatial fidelity, leading to the most uniform dose in the target, the steepest dose gradients, and the lowest dose delivered to normal tissues [15].
Evidence-based Studies on Quality and Efficacy of Trilogy Radiotherapy Platforms The capability of the Trilogy to deliver high dose rates in a short time is a key feature of this system. Typical dose rates for Trilogy stereotactic mode (6 MV beam, up to 1,000 monitor units (MU /min dose rate, up to 6,000 MU/field total dose, up to a maximum field size of 15 cm 15 cm, and up to 60 MU/deg dose rate for arc-based treatments) range from 600 cGy/min to 1,000 cGy/min. Through its advanced capabilities, the Trilogy accelerator can also produce an 18-mV high-energy beam (40 cm 40 cm, up to 600 cGy/min at 100 cm). Compared to other Linacs, the Trilogy system delivers a stereotactic dose rate 20% higher at a rate of 1,000 MU per min. Fan et al. advise exercising cautious when using the Varian recommended beam data to perform Trilogy SRS [41]. Trilogy-based treatments have shown high dose conformity for various intracranial and extracranial conditions. In a case study, a single fraction 20 Gy of frameless Trilogy SRS was delivered to a single frontal lobe metastasis located in Broca’s area (0.6 cm3/1.0-cm diameter). A single isocenter with a 10-mm cone and five 100 arcs resulted in high dose conformality with sharp (3 mm) dose fall-off [15]. VanderSpek, et al. performed single fraction single isocenter Trilogy IMRS for multiple brain metastases in 10 patients. The median prescribed dose was 16 (range, 14–18) Gy and the median total planning target volume (PTV) was 35.0 cm3. Eight patients either showed an improvement in
their symptoms or did not experience complications with IMRS [42]. In a quantitative study, the automated quality assurance QA procedure for SRS on a Varian Trilogy Linac was found to be quick, precise and an efficient replacement for the conventional film-based quality assurance method [43].
Linac SRS: The Future Linacs will be the most oft-used tools used for SRS in coming years for a variety of reasons: (1) they are the most common devices for administering RT, hence can be adapted for stereotactic treatments: (2) as noted, Linacs are being created as inherently stereotactic devices for use in standard RT; (3) radiation oncologists control their purchase, implementation, and use in almost all cases. We are still at the beginning of this era of allinclusive Linac systems that provide imaging, inverse planning for IMRT and SRS, and routine stereotactic localization. Much work needs to be done to assess the different technical solutions that have been proposed. For example, in an assessment of SRS treatment plans involving dynamic conformal arcs for intracranial lesions, the shift from a 5-mm leaf size in the Novalis MLC to a 3-mm leaf size in the Varian Millennium MMLC was linked with improved target conformity and increased normal tissue sparing. However, ongoing studies are required to determine whether the parameter of size alone applies to all collimators, regardless of total leaf count, and whether smaller leaf size necessarily translates into clinical improvement [44]. Other issues to assess will be the confirmation of frameless localization as a routine method for even high-dose single session SRS (as in for functional SRS for patients with trigeminal neuralgia or movement disorders). The benefits of fluoroscopic versus CT registration should be examined. The cost-effectiveness of dedicated SRS units will be compared with Linacs adapted for
Overview of radiosurgery technology
SRS. Without question, Linac-based SRS will be an ever larger component of clinical neurosurgery and radiation oncology.
Tomotherapy Description Tomotherapy, which literally means ‘‘slice therapy,’’ is a novel platform for administering IMRT. It can deliver higher dose radiation therapy to target sites, including multiple sites simultaneously, in various regions of the body while still limiting the amount of radiation to healthy tissue. Also known as helical tomotherapy (HT), it seamlessly integrates into a single system (1) a customized inverse treatment planning system; (2) onboard, real time, online megavoltage CT image-guided patient positioning; and (3) a spiral pattern of IMRT treatment delivery. The HT device also includes a MLC and a quality assurance module individualized for each patient [21,45–48]. The HT prototype was developed in the late 1980s by the Radiotherapy Research Group, led by Mackie and Reckwerdt at the University of Wisconsin. It was based in part on principles of serial tomotherapy and a modulated multileaf collimator technology, both of which characterized the abovementioned Peacock system [49]. The current HT version, the Tomotherapy Hi-Art Treatment System, is an intensitymodulated, rotational radiation therapy that contains a ring gantry geometry with fan-beam delivery [21,45,47]. HT utilizes a helical CT platform instead of the static treatment arm and couch platform found in traditional Linacbased SRS devices. In the HT unit, a miniaturized 6 MV Linac standing only 18 inches high is mounted to a CT gantry, such that the combined device is used for both imaging and treatment. The accelerator revolves around the patient as he moves slowly through the gantry ring [21,45,47].
57
Since 2003, HT has been used clinically as an alternative SRS technique to GK, CK, and other Linac-based SRS devices for the treatment of brain tumors [21,45,47]. For imaging, the HT unit generates a megavoltage CT (MVCT) 3-D scan that delineates the precise contours of each tumor just prior to irradiation. This is essential for verifying the location of the tumor and adjusting the patient’s position. Accurate patient setup is achieved by coregistering pretreatment daily MVCT scans with the initial planning CT scan [21,45,50,51]. Since daily MVCT localization can be attained with frameless stereotaxis, internal or external fiducial markers for tumor localization may not be needed [52]. In the Tomotherapy Hi-Art System, the Linac uniquely delivers thousands of minute radiation beamlets from every point along a spiral pattern as the photon energy forms multiple circles around the gantry ring. The beamlets are delivered from various angles, intersecting with multiple targets and, at the same time minimizing the amount of radiation that reaches neighboring tissue (> Figure 57‐5). These capabilities enable HT machines to irradiate lesions usually not targeted with conventional Linac SRT and SRS devices [21,45].
Performance Features Daily IMRT treatment with HT is continually modified through an innovative verification method called dose reconstruction [49,51]. By providing a comparison of CT radiation dose images with planned dose images, future treatments are modified to correct for errors in completed treatments. The intensity of the beams can be changed, if necessary, by adjusting the width of the MLC leaves. This process of daily correction during multi-treatment hypofractionation sessions is called adaptive radiation therapy [51]. During HT treatment, the beam
877
878
57
Overview of radiosurgery technology
. Figure 57‐5 Illustration of helical tomotherapy concept. Radiation is delivered from 360 as the patient moves through the gantry (courtesy of TomoTherapy Inc.)
treatment team be highly skilled as both a system user and an expert in managing all technical aspects of verification QA [53].
Evidence-based Studies on Quality and Efficacy of Tomotherapy
path is totally shielded to reduce radiation ‘‘scatter’’ to the patient. In a technologically advanced fashion, the motion of the MV radiation source in the HT unit is synchronized with the movement of both the patient and the MLC. As the Linac moves in a circular pattern delivering the beamlets, the patient slowly advances in and out of the center of the gantry ring. Simultaneously, the shape of the radiation beam, and essentially the angle of the treatment, are modified by the pneumatically driven MLC, which has two sets of interlaced leaves resembling a zipper. As the treatment couch moves, the beamlets can be turned on or off by opening and closing the 64 leaves of the MLC [21,50]. The dynamic delivery of radiation in HT is complex, and carefully orchestrated by multi-level integrated, automated systems that control coordinated patterns of motion in the gantry, treatment couch, and multileaf collimator leaves. Yet, this complex scheme is partly hidden from the end-user. Therefore, it is imperative that the medical physicist on the
Tomotherapy has been used mainly as a means of delivering fractionated RT throughout the body in a way that is more precise than standard conformal treatments. As such, there is relatively little literature analyzing it as a tool for SRS. Yartsev conducted a small scale study (n = 12) comparing HT plans with other radiotherapy modalities in the treatment of small brain tumors. Of five methods evaluated, proton techniques (see below) exhibited the overall best dose distributions. HT resulted in higher target dose uniformity (average SD ¼ 1.3%), and showed a risk of irradiation to critical structures comparable to that of other Linac methods [54]. Shi, et al. found that HT treatment plans generate fairly uniform doses to target treatment sites in the head and neck, brain, lung, prostate, pelvis, and cranio-spinal axis when compared with Linacbased step-and-shoot IMRT planning for special treatment sites. Their findings also confirm that HT produces smaller integral doses to healthy structures when delivered to these target anatomical sites [55]. In eight patients treated for spine tumors (seven with metastatic disease), the Tomotherapy Hi-ART System produced accurate set-up and delivery without the need for fiducial markers. Acute and late toxicity was minimal, and four patients were still alive (median overall survival, 5.1 months) at the completion of the study [52]. The on-board MVCT image system increases the accuracy and precision of patient positioning over time during delivery of multiple fractions. Repeat CT imaging and re-planning is highly advantageous, especially for IMRT treatment of head and neck cancer. Daily setup corrections may be
Overview of radiosurgery technology
needed due to weight changes and subsequent alteration in anatomy near the target lesion. Hansen, et al. reported that the median rate of decline of the clinical target volume was 1.7–1.8% per treatment day, accompanied by a shrinking volume that was often asymmetric in patients with head or neck cancer. In a case study, the modified radiation dose delivered to a patient had a sparing effect on spinal cord and trachea [45,56]. In a comparison of single fraction HT and Gamma Knife SRS used to treat single brain metastasis in 5 patients, these two techniques produced different dosimetric and conformal indices. Compared to Gamma Knife SRS, HT was linked with larger lower isodose line volumes and longer treatment times [57]. HT is not without other limitations. A single slice of MVCT is reconstructed from a 180 rotation in 5 s, but it contains motion artifacts from breathing. Yet, a full 360 cone-beam CT scan usually produced in 45–60 s is also constrained by motion artifacts. Future refinements to HT imaging technology eventually may resolve this problem [45,52,58]. Gutie´rrez, et al. evaluated the feasibility of using composite HT planning to deliver whole brain radiotherapy in 10 patients with metastatic brain tumors treated with an integrated boost of SRS. The planning used HT with the original CT scans and MRI-CT fusion-defined target and normal structure contours. Composite HT planning achieved favorable outcomes of hippocampal avoidance; homogeneous whole brain dose distribution equal to that of conventional whole brain radiotherapy; and radiosurgically equivalent dose distributions to individual metastases [59]. Welsh, et al. reported that HT treatment has yielded excellent results for reducing the risk of alopecia in patients undergoing standard whole brain radiotherapy (WBRT). By planning a scalp-sparing treatment, HT appeared to be superior in hair preservation when compared with both conventional modalities and other IMRT techniques used for WBRT. The superior
57
dose-rate of HT may offer advantages over conventional Linac-based IMRT, particularly as clinical targets become more complicated [60]. Documented clinical benefits of HT include the amount of the time required to create plans, overall tumor target conformality, capacity for conformally avoiding critical structures, and the time required to deliver each fraction [60]. Delivery of the conventional dose of 300 cGy with the HT method may take approximately 20 min for all steps, including the MV CT used for image-guidance and set-up. By contrast, complex IMRT delivery of a 100 cGy dose fraction takes an estimated 30 min with a Varian EX accelerator in dynamic delivery mode. A standard 300 cGy dose delivery from the Varian EX would, therefore, become intolerably long and impractical for clinical use [60,61]. Helical tomotherapy has been an important addition to the SRS/SRT landscape, and has spurred interest throughout the field in the combination of digital sectional imaging as an integral part of stereotactic irradiation. Clinically, its advantages will probably be felt mainly in conformal and tissue sparing treatments of patients with relatively large target volumes (> Figure 57-6).
Cyberknife: A Robotic Linac Description The Cyberknife (CK) (Accuray, Sunnyvale CA) is an innovative tool in SRS. Developed in the early 1990s by John Adler and collaborators at Stanford University, the CK was the first device to allow for frameless SRS for treating targets in the head as well as anywhere in the body (> Figure 57‐7). The latter was the impetus for development of the CK. After a fellowship in Stockholm in the early 1980s studying SRS with Lars Leksell, Adler sought an SRS technology that would not be limited to intracranial targets
879
880
57
Overview of radiosurgery technology
. Figure 57-6 Tomotherapy treatment plan for a patient with brainstem glioma (courtesy Dr. Jed Pollack, Long Island Radiation Therapy, Lake Success, NY)
. Figure 57‐7 Patient about to undergo SRS with Cyberknife
Overview of radiosurgery technology
(J Adler, personal communication). The CK employs an X-band Linac, small and light enough to be mounted on an industrial robot. SRS is delivered by a stop-and-shoot technique that avoids the need for isocentric planning. Instead, the target is ‘‘painted’’ with the prescription dose. Treatment planning is done with an inverse technique, after clinicians define the target, designate the dose and number of fractions, and set dose constraints for adjacent critical structures [3,62,63]. The CK can deliver both single-fraction and multifraction SRS with submillimetric accuracy. The CK frameless image-guidance system includes precisely calibrated, paired orthogonally placed (90-degree offset), diagnostic X-ray tubes that are rigidly attached to the ceiling. Real-time images of the anatomical treatment site are generated using two amorphous silicon X-ray screens, or detectors that produce high-resolution digital images. These are
57
registered to digitally reconstructed radiographs (DRRs) generated from the patient’s pretreatment CT scan [62,64,65] (> Figure 57‐8). CK was the first SRS system to use 3D digital data for enhanced complex treatment planning for SRS [3].
Performance Features of the CK The CK has gained increasing acceptance as a SRS tool, largely because of its high level of accuracy, rapid radiation dose fall-off and precision on a par with that of frame-based SRS systems [62,64–67]. Through an image-guided control loop connecting the imaging system and the robotic manipulator, the CK provides real-time tracking of both patient motion and the exact position of the lesion. An advanced algorithm controls the loop, which in turn prompts the tracking system to automatically
. Figure 57‐8 Screen view of the Cyberknife registration system. Mismatches between the new fluoroscopic images and the DRR from pretreatment CT (left) are corrected by automated table movement until accurate overlay is confirmed
881
882
57
Overview of radiosurgery technology
re-position the robotic manipulator to locate and track the target, and to move the radiation beam into alignment with the target lesions. The miniaturized Linac in the CK is advantageous for rapidly positioning the robotic arm over a wide range of beam orientations [65]. Radiation beams can be aimed almost anywhere in space. In a typical treatment, the target is irradiated with 80–120 stereotactic, precisely cross-fired, non-isocentric radiation beams that produce complex overlapping arrangements. As noted above, patient positioning is corrected via co-registration of projection images that are taken with the silicon detectors positioned on either side of the patient’s anatomy. An algorithm directs the computer to correlate the pair of alignment radiographs of the target from the perspective of the two cameras with the patient’s treatment planning CT scan [68,69]. This mechanism allows for SRS to be delivered using only mask immobilization. Digital imaging and registration can be repeated after each beam, but in practice this would unnecessarily lengthen treatment times. Depending on the degree of patient movement, images can be repeated more or less often, but typically at least after 10 beams are delivered. Movements over several mm, which are unusual, may require manual repositioning by the radiation therapist to bring the couch back into position. When the CK algorithm redetermines the spatial coordinates of the target lesion, and then relays to the robot the information needed to correct the patient positioning, the patient himself is not moved. Instead, the CK image-based computer software accurately directs adjustment of the treatment couch [68]. Treatment plan optimization in the CK system is based on inverse planning algorithms, including a planned dose distribution analysis for selecting beam directions and weights of each node surrounding the target volume [70]. The CK usually is programmed to function with
a non-isocentric technique, using multiple overlapping isocenters to irradiate irregularly shaped lesions. In addition, the CK also has algorithms that can direct a beam to be delivered to a single isocenter for treating spherical targets [3]. During treatment, the robotic arm can be deployed to 1,200 positions for controlling beam direction [68,71]. In practice, between 80 and 120 beams typically are used to keep treatment time to about 30 min per session, on average. Yu, et al. compared treatment plans obtained with the CK with those of other SRS therapies, including GK, Linac multiple arcs, conformally shaped static fields, and IMRT. The evaluations were based on dosimetric indices such as dose-volume histograms and other standard radiosurgical parameters (target coverage, homogeneity index, and conformity index). While all of the tested treatment modalities provided the same level of full target coverage, the treatment plans for the CK yielded more flexibility for a given target size and shape. Conformal dosimetry with the CK and Gamma Knife was similar for lesions of limited volume, regardless of shape. Again, the ability of the CK to irradiate irregularly shaped targets is a function of its nonisocentric treatment option [68]. Giller, et al. suggested that the maximum error reported for the robotically controlled CK system is not just comparable with frame-based SRS systems, but may even be superior to many frameless and frame-based systems. They proposed that the CK may therefore be feasible for SRS treatment of malignant brain tumors in infants [72].
Evidence-based Studies on Quality and Efficacy of the CK The frameless technology of the CK renders it effective in treating not only patients with head and neck cancers and spinal lesions, but also
Overview of radiosurgery technology
abdominal and thoracic tumors, pancreatic and liver cancer, and pelvic tumors such as prostate cancer and select rectal tumors [73–77]. Yet, it is in the realm of intracranial SRS that the CK initially made its mark as a technically feasible and practical treatment device for both malignant disease and functional disorders. In an early study, Chang, et al. evaluated the effects of CK in singlefraction SRS treatment (mean radiation dose, 18.1 Gy; range, 12–36 Gy) on 84 tumors (primarily metastatic carcinoma) in 72 patients. Their findings demonstrated that the overall accuracy of the frameless CK system is comparable to that of other Linac-based systems requiring invasive stereotactic frames for skeletal fixation [78]. By 2001, Chang and Adler reported that the clinical results for nearly 1,900 intracranial tumors and arteriovenous malformations (AVM) treated with the CK closely paralleled the outcomes of other SRS techniques [79]. Shimamoto et al. reviewed clinical outcomes of CK irradiation (dose range dose, 9 to 30 Gy) on 66 metastatic brain tumors in 41 patients. Tumors treated with at least 24 Gy were less likely to progress than lesions treated with 20 Gy (p = 0.0244; log-rank test). No severe side effects occurred. The complete response (CR) rate was significantly higher if the maximum dose to the tumor was at least 24 Gy (p = 0.0045) [80]. More recently, Collins et al. assessed CK treatment planning parameters for fractionated SRS of 82 skull base lesions in 80 patients. The plans showed excellent conformity, homogeneity, and percent target coverage for both simple and complex skull base tumors percent target coverage [70]. In a multicenter retrospective analysis of 95 patients treated for idiopathic trigeminal neuralgia, Villavicencio et al. identified optimal radiosurgical treatment parameters for the treatment of idiopathic trigeminal neuralgia. Of the 95 patients, 64 (67%) initially experienced excellent pain relief, with a median time to pain relief of 14 days. In contrast to previous reports, in this
57
study higher doses were used (median maximal dose, 78 Gy; range, 70–85.4 Gy) and longer segments of the trigeminal nerve were irradiated (median length, 6 mm; range, 5–12 mm), resulting in both better pain relief and a higher incidence of hypesthesia. Post-treatment numbness was linked with improved pain relief. At a mean follow-up of 2 years, 47 patients (50%) had experienced sustained pain relief, and were completely off pain medications at follow-up (mean 2 years). The overall complication rate was 18% [81]. A major impact of the CK has been on SRS for patients with skull base tumors. The ability to fractionate has allowed for treatments in 2–5 sessions to be given to patients with vestibular schwannomas [82] and parasellar tumors [83] (> Figure 57‐9). Results have shown tumor control of 95% or better with few complications. Hearing preservation was maintained in about 70% of patients, comparable to Gamma Knife and other SRS series [82]. Patients with parasellar lesions did not develop visual loss, except in one patient who had prior fractionated irradiation [83].
Spinal SRS with the Cyberknife The first description of spinal SRS was by Hamilton et al. [84]. This groundbreaking innovation did not in the end prove practical, due to the need to fix a registration marker to the patient’s spinous process. Besides the invasive aspect of the procedure, this meant the treatment would be done prone and most likely under general anesthesia. The CK was the first tool to make spinal SRS practical. This has included treatment of patients with metastatic tumors but also with intradural and intramedullary neoplams, and AVMs [79]. Gerszten, et al. analyzed the effects of single-fraction SRS delivered via the CK to cervical spine lesions in 115 consecutive patients. Both benign and metastatic tumors were treated,
883
884
57
Overview of radiosurgery technology
. Figure 57-9 Cyberknife treatment plan for 34-year-old man with recurrent pituitary tumor in left cavernous sinus and nearing the left side of the optic chiasm. He received 2,500 cGy in 5 daily fractions
and lesion volume ranged from 0.3 to 232 ml (mean, 27.8 ml). Tumor radiation dose was held at 12–20 Gy to the 80% isodose line (mean 14 Gy). Skull bony landmarks were used to locate and track cervical spine lesions. Fiducial markers were positioned to pinpoint the lower spinal lesions. Patients showed rapid recovery and good response after a short treatment time in an outpatient setting. No acute radiation toxicity or new neurological deficits were observed during the follow-up period (3–24 months) [65]. Data from an anthropomorphic phantom study confirm the submillimetric accuracy of the CK in spinal SRS. Based on head and torso
phantoms, Yu et al., reported an average treatment delivery precision of 0.3 0.1 mm for CT slice thickness (mean, 0.7 0.3 mm; range, 0.625–1.5 mm). The tracking error for fiducial markers was less than 0.3 mm for radial translations up to 14 mm, and less than 0.7 mm for rotations up to 4.5 [68]. These values are generally comparable with those of previous phantom studies of the targeting of brain lesions using standard SRS head frames [68,85–89]. CK software now supports accurate tracking of spinal targets without the need for fiducial placement. The error of this system was found to be 0.6 mm, hence as accurate as that attained with fiducial implantation [90].
Overview of radiosurgery technology
SRS treatment of spinal conditions using the CK can be completed in a single day, and may significantly improve local control of cancer of the spine [91]. In a prospective evaluation of 18 patients with malignant sacral lesions treated with single-fraction SRS via the CK, pain improved in all 13 patients who had symptoms before treatment. Followup imaging showed no tumor progression. No acute radiation toxicity or new neurological deficits were reported during the followup period (mean, 6 months). In contrast to conventional external-beam radiotherapy, the CK was able to effectively deliver large doses of radiation while sparing nearby radiosensitive structures, particularly the spinal cord and cauda equina. The volume of the cauda equina receiving more than 8 Gy ranged from 0 to 1 ml (mean, 0.1 ml) [92]. Similarly, SRS administered with the CK was tolerated in 15 patients receiving singlefraction radiosurgery for treatment of benign spinal lesions (12 cervical, one thoracic, and two lumbar). Dose plans for the CK did not produce any acute radiation-induced toxicity or new neurological deficits during the follow-up period. No tumor progression was evidenced in follow-up imaging (mean, 12 months) [66]. A recent study further demonstrates the shortterm clinical benefits of single fraction SRS in treating benign extramedullary spinal neoplasms. Gerszten et al. evaluated the effects of SRS (mean intratumoral dose, 2,164 Gy) on the treatment of benign intradural extramedullary spinal tumors, primarily located in the cervical region in 73 patients [120]. The cases included neurofibroma (25), schwannoma (35), and meningioma (13). SRS was used mainly as a primary treatment modality or as a therapy of postsurgical radiographic progression. The median follow-up period was 37 months. Longterm radiographic tumor control was seen in all cases, with long-term pain improvement achieved in 22 (73%) out of 30 cases. New symptoms associated with radiation-induced
57
spinal cord toxicity developed in three patients 5–13 months after treatment. Studies conducted over longer follow-up periods are critical to assessing whether SRS has long-term efficacy for benign spinal tumors [67].
Definition of SRS and the Cyberknife SRS was defined as a single session treatment by Leksell, and this concept went unchallenged for 50 years. The advent of the CK called this into question. The frameless technique, with accuracy shown to be equivalent to that achieved with a stereotactic frame, allowed SRS to be done in more than one fraction. Prior stereotactic fractionation, done with relocatable frames, had meant the use of standard fractionation regimens – typically 25–30 treatments – but with smaller volumes and less irradiation of surrounding tissues than with standard fixed field treatments. The CK began to be used to treat patients with lesions that were too large or critically located for single session SRS, in up to five sessions, depending on the perceived risk of hypofractionation [82,83]. There does not seem to be any logical reason to define SRS as necessarily confined to a single session. With this in mind a committee convened by the leadership of American organized neurosurgery redefined SRS as a stereotactically targeted, image guided treatment of 1–5 sessions [13]. This expanded definition of course applies to all devices used for SRS, including the other Linac based systems discussed above. This has allowed neurosurgeons to recommend for their patients the best individualized treatment, without concern that they will lose their involvement in patient care on the one hand, or prescribe a suboptimal single session treatment on the other.
885
886
57
Overview of radiosurgery technology
Leksell Gamma Knife Perfexion Description The Leksell Gamma Knife Perfexion (Elekta, Stockholm, Sweden), an updated version of previous Gamma Knife (GK) units, features a new design and major improvements in performance capabilities. The Perfexion is a fully robotized GK that can treat patients with a wide variety of lesions in the skull base and neck. Like its predecessors, the Perfexion still uses a multi-isocenter treatment plan, but it incorporates inverse treatment planning and depends less on decreased forward treatment planning [3,19,50,93,94]. Compared to earlier models of the GK, the Perfexion is better able to treat target volumes of up to 300% greater than previously possible, while simultaneously sparing healthy tissue near eloquent areas [19,50,93,94]. Full automation of the dose-delivery process in the Perfexion has decreased manual input and errors. As a result of the smoother workflow of integrated components of the system, both treatment time and discomfort to patient have been significantly reduced [3,93].
Performance Features The radiation unit of the Perfexion has been reconfigured with a new beam geometry of 192 Cobalt-60 sources arranged in a cone section pattern. This modification translates into a higher dose rate for any given source activity. The redesigned cylindrical shape of the collimator helmet of the Perfexion allows anatomic structures in once inaccessible locations to be targeted for SRS treatment. The caudal reach of the Perfexion extends further than in the traditional helmet of the GK Model C unit, thus providing improved peripheral coverage. With the redesigned collimator
. Figure 57‐10 Graphic illustration of the single Perfexion collimator helmet. Different collimator diameters and blocking patterns can be automatically set without the need for manual helmet changes (courtesy of Elekta, Inc.)
helmet, upper cervical lesions as well as extracranial head and neck tumors can now be treated [19,93,94]. The Perfexion is the only instrument in the GK series to contain a single collimator within the housing unit (> Figure 57‐10). The collimator consists of a larger, single, 12-cm thick tungsten array arranged in a series of five concentric rings. Older models of the GK have multiple standard collimators with fixed apertures, but these require manual changing between shots requiring different collimator sizes. In the Perfexion, the multiple, exchangeable collimator helmets have been replaced with an integrated automated adjustable collimator system. The built-in modulating apertures of the new collimator system are automatically adjusted by internal controls, thereby eliminating the time-consuming process of manually switching helmets [19,50,93]. The longer axial diameter of the Perfexion collimator helmet allows peripherally located targets to be positioned at isocenter with less potential for collimator-frame collisions. The ability to automatically adjust the size of the collimator
Overview of radiosurgery technology
openings is essential for protecting critical structures adjacent to the target from excessive irradiation. These performance-related design features enable the Perfexion to create complex shapes of isodose volumes through dynamic shaping of dose distribution [19,93,94]. The robotic-driven automatic positioning system of the Perfexion operates through the couch unit rather than the head frame. Repositioning the patient by moving the couch instead of the head frame is less disruptive for the patient. In addition, the robotic-controlled couch system facilitates a seamless transition between ‘‘shots’’ fired in a multi-isocenter treatment plan [19,93] (> Figure 57-11).
57
the Perfexion relative to the GK 4C. These findings, combined with other documented clinical benefits of the Perfexion, indicate that it meets the highest standards of an SRS instrument [94]. As a frame-based SRS device, the GK does not lend itself readily to fractionated treatments. However, the Perfexion unit has greatly increased the versatility of this instrument, which will remain an important way to deliver SRS for the foreseeable future.
Heavy Particle Radiotherapy Description
Evidence-based Studies on Quality and Efficacy of the Leksell Gamma Knife Perfexion Studies confirm that compared to previous models of the GK, the Perfexion has an increased ability to shape isodose volumes, even for single isocenter treatment plans. As noted, the new collimator arrangement and the automated patient positioning system have led to decreased patient transit times and less exposure during movement between isocenters. The larger radiation cavity of the Perfexion also permits more regions of the head and neck to be targeted for treatment [93]. In a blinded randomized prospective comparative study on various treatment related parameters, including QA, 29 patients were treated with the GK 4C and 30 patients with the GK Perfexion. Patients underwent treatment for various lesions, including multiple metastases. Treatment with the Perfexion unit was associated with greater radiation protection and with collision-free procedures. The time of the SRS treatment, time of clinician and physicist intervention on the machine, and time of the QA procedure were all reduced with
Charged particle, or heavy particle beams have been used since the 1950s to treat intracranial tumors, arteriovenous malformations (AVMs), and subfoveal neovascularization [95]. To some extent this reflects the fact that heavy particle therapy provided the only way, until the invention of the GK in the late 1960s, to deliver highly conformal irradiation with a steep dose gradient. Today, while there are still relatively few facilities offering this method, heavy particle irradiation, particularly with protons and carbons, is gaining momentum as an efficacious treatment. Unlike radiation from gamma or photon sources that decays exponentially, there is no exit dose of proton energy deposition beyond the target volume. When proton beam energy is delivered to the tissue, it first enters the area as a slowly rising dose followed by a rapid rise to a maximum (the Bragg peak), and then a fall to near zero [96]. In practice, an overly sharp radiation deposition is insufficient to treat a volume such as a tumor. Therefore heavy particle beams need to be shaped and flattened with a variety of filters to create a spread-out Bragg peak (SOBP) [97] (> Figure 57‐12). During treatment with heavy particles, beams are delivered via either a passive (scattering) or active (scanning) method. The technique can be selected
887
888
57
Overview of radiosurgery technology
. Figure 57‐11 Gamma Knife Perfexion treatment plan for patient with a right vestibular schwannoma. Note the high number of shots, the different collimator sizes, and the blocking patterns (courtesy of Dr. Jean Re´gis, University Hospital La Timone, Marseilles)
by changing the operating parameters of several pieces of equipment such as the accelerator and collimator. The Bragg peak coincides with the treatment volume, and is determined by the beam energy that can be altered by controlled by adjusting the accelerator. In pencil beam scanning, multiple beams of varying energies are superimposed on each other, thus causing the treatment volume to exhibit an SOPB [97]. Heavier ion beams produce better dose distributions precisely because of this higher biologic property: the ability to generate a higher energy transfer in the Bragg peak as compared with the entrance point of beam transport [98,99]. The physical characteristics of protons and other high energy, heavy particles make them attractive sources of radiation beams for therapeutic radiation. Charged-particle SRS is particularly advantageous for the conformal treatment of large and/or irregularly shaped lesions, as well as for minimizing radiation reaching eloquent brain structures located near the target site. As newgeneration intensity-modulated proton techniques
emerge, proton SRS may yield even more favorable outcomes for treating large target volumes while sparing critical structures such as the spinal cord, eyes, and brain [96,100]. Newer units also are smaller and have beams that exit from relatively compact gantries that can rotate around an isocenter (> Figure 57‐13).
Evidence-based Studies of Heavy Particle Radiotherapy Proton irradiation produces high dose isodose conformality, particularly for tumors with complex geometry, especially concavity [101]. In a clinical comparison assessment of proton-based and proton SRS for the treatment of AVMs, vestibular schwannomas, and pituitary adenomas, protonbased techniques achieved excellent outcomes. The results were at least on par with those of photon-based radiation therapies. Compared to photon irradiation, proton radiation beams produced superior conformality in dose distribution, particularly in larger sized lesions, and showed good sparing effects of normal tissue [96].
Overview of radiosurgery technology
. Figure 57‐12 The spread-out Bragg peak 2000;31:6, with permission)
(Europhysics News
Proton beam SRS improved tumor control for vestibular schwannoma [102,103] and showed reasonable rates of facial and trigeminal nerve functional preservation [102]. Weber, et al. evaluated 88 patients with vestibular schwannomas (mean, 16 mm; range, 2.5–35 mm) who were treated with proton beam SRS that applied two to four convergent fixed beams of 160-MeV protons. Facial nerve function (House-Brackmann Grade 1) and trigeminal nerve function were normal in 79 patients (89.8%). Hearing was good or excellent (Gardner-Robertson (GR) Grade 1 in eight patients (9%) and serviceable in 13 patients (15%) at follow-up (median period, 38.7 months). Seven (33.3%) of the 21 patients%) with functional hearing (GR Grade 1 or 2) retained serviceable hearing ability (GR Grade 2). The actuarial 2- and 5-year tumor control rates were 95.3% (95% confidence interval (CI), and 93.6% (95% CI). The actuarial 5-year cumulative radiological reduction rate was 94.7% (95% CI,). Actuarial 5-year normal facial and trigeminal nerve function preservation rates were 91.1% (95% CI) and 89.4% (95% CI) [102]. Cumulative data indicate that fractionated proton beam radiotherapy improves tumor control in patients with skull base tumors [98,104] including adenoid cystic carcinoma of
57
. Figure 57‐13 A patient about to be treated with proton beam therapy using a compact gantry (Europhysics News 2000;31:6, with permission)
the skull base [105] and pediatric base of skull tumors [106]. Proton radiation therapy offers excellent prospects for durable tumor control in patients with low-grade chondrosarcomas or chordomas, and without causing undue complications in some series [98,107]. However, others have reported an alarming 50% risk of visual loss after proton treatment of clivus chordomas [108]. Similarly, in a comparison of conformal radiation methods, proton therapy was superior to conventional photon irradiation and photon IMRT for craniospinal axis irradiation and posterior fossa boost in a pediatric patient with medulloblastoma [109]. The dose to the cochlea was reduced from 101.2% of the prescribed posterior fossa boost dose to 33.4% with IMRT, and only 2.4% with protons. Dose to 50% of the heart volume was decreased from 72.2% for conventional X-rays to 29.5% for IMRT and 0.5% for protons. The considerably enhanced effect of protons on nontarget tissue sparing may be favorable for long-term toxicity,
889
890
57
Overview of radiosurgery technology
particularly in regard to and endocrine and cardiac function [109]. Carbon ion radiotherapy has been proposed because these heavy ions can provide highly effective dose-localization in the body. Growing evidence reveals that carbon ion radiotherapy is effective in treating patients with tumors of the skull base and head and neck, as well as bone and soft-tissue sarcomas, prostate cancer, non-small-cell lung cancer, and hepatocellular carcinomas [99]. In studies conducted in Japan, carbon ion hypofractionated radiotherapy using larger doses per fraction resulted in local control rates for locally advanced tumors with minimal toxicity. Overall treatment times of carbon ion radiotherapy were less than conventional radiotherapy [110]. This treatment had a sparing effect on critical structures such as the spinal cord, eyes, and brain [100]. Higher doses of carbon ion radiotherapy administered in combination with X-ray radiotherapy and chemotherapy for malignant gliomas increased the survival rate in patients [111]. The physical and biological properties of carbon ion beams appear to have advantages over those of photons. Mizoe, et al., conducted phase I/II dose-escalation studies and clinical phase II studies evaluating the effects of carbon ion radiotherapy on 1,601 patients with various types of malignant tumors. All patients except those with malignant glioma were delivered carbon ion radiotherapy alone. The fraction number and overall treatment time was fixed for each tumor site, and administered to one field per day and 3 or 4 days per week. The total dose was escalated by 5 or 10% increments to guarantee safety. Carbon ion radiotherapy resulted in high local control rates for locally advanced tumors, including advanced head and neck tumors, and pathologically non-squamous cell type of tumors, producing minimized toxicity. Carbon ion radiotherapy has led to improved outcomes in several cancers, including chordoma and chondrosarcoma of
the skull base and cervical spine, locally advanced prostate carcinomas in high-risk patients, and post-operative pelvic recurrence of rectal cancer [112]. However, in this study, severe late complications of the recto-sigmoid colon and esophagus developed in patients receiving high dose levels of carbon ion radiotherapy for prostate, uterine cervix and esophageal cancer. These adverse effects disappeared when patients were administered safe dose levels that were established after improved irradiation methods were implemented. The results showed that carbon ion could be delivered with larger per fraction doses of hypofractionated radiotherapy and in a shorter treatment time compared to conventional radiotherapy [112].
Heavy Particle Therapy and SRS There are about 20 proton bean units in operation around the world, with a handful of carbon ion facilities. For some conditions, especially in patients with chordoma, there is an assumption by many clinicians that proton beam irradiation is the treatment of choice [113]. Compared to conventional fractionated radiotherapy 10-year survival rates are much better, on the order of 65% versus under 20%. On the other hand, the use of stereotactic photon SRT or SRS has shown much better results than with RT alone, with 82% progression free and overall survival of 82% at 8 years [114]. These results exemplify this debate: to what extent does the Bragg peak and high energy deposition of heavy particles confer an advantage over photon SRS? Or does the presumed benefit of protons and other heavy particles mainly reflect the historical fact that this was the only real way of delivering highly focused therapeutic irradiation, until relatively recently? This question is hardly academic. Commercial installations of proton bean units cost on the order of
Overview of radiosurgery technology
$100 million; for heavier ions, with more complex cyclotron requirements, costs will be even higher. Medical centers around the US and elsewhere are considering implementation of heavy particle units. Can these centers, and national health systems around the world, sustain the costs involved? Academia and industry may eventually pave the way for practical proton-based SRS instrumentation. Caporaso, et al. and colleagues at Lawrence Livermore National Laboratory are designing a novel compact CT-guided proton therapy system that uses a dielectric wall accelerator. This device can be housed in a conventional Linac vault, and feasibility tests of an optimization system are underway. The completed proton therapy system will permit rapid, optimized CT-based IMPT delivery for the treatment of larger target volumes as well as for motion management in pediatric patients. It will be capable of generating high electric field gradients by using alternating insulators and conductors and short pulse times. This proton therapy device is based on emerging new technologies such as high gradient, vacuum insulators, solid dielectric materials, SiC photoconductive switches and compact proton sources [115]. Certain vendors, including Varian and Still River Systems Proton Therapy may succeed in creating practical proton beam devices that are less expensive – but still on the order of $15 million per. Perhaps most neurosurgeons interested in radiosurgery should focus their energies on continuing to improve techniques of photon and gamma SRS.
Conclusions Stereotactic radiosurgery was, for the first 45 years of its existence, the provenance of a few neurosurgeons (and allied radiation oncologists and medical physicists) who took a special interest in this then esoteric technique. As a result of
57
the success of the SRS pioneers, this has now become a standard part of stereotactic neurosurgery – so much so that neurosurgical training programs now require SRS to be taught to all residents. As neurosurgeons understood more of the history and radiobiology of radiation therapy, the benefits of fractionation led to the current definition of SRS. At the same time, the obvious advantages of the stereotactic technique became clear to manufacturers of Linacs and to radiation oncologists as a whole. It was inevitable that radiation therapy in general would become increasingly stereotactic – in other words, that SRS would become more like RT, and that RT in turn would become more like SRS. Newer SRS devices and the treatment planning processes used to implement SRS therapy are changing the practice of neurosurgery. While SRS will never totally replace conventional surgical neurology, the distinct advantages that each of the newer SRS modalities offers has broadened the range of options for intervention that are available to neurosurgeons. The fundamental goals for the safe and effective use of SRS remain unchanged: dose homogeneity, target coverage, healthy structure sparing, and dose conformality [21]. The challenge of SRS today lies in determining which competing device is best for delivering the optimal radiation doses for a given target. As the contents of this chapter reveal, the lines separating conventional SRS, hypofractionation, and SRT are thin. In addition, the clinician may have to decide whether the treatment should be delivered using either 3D conformal radiotherapy or IMRT [40]. The answer to the clinician’s dilemma may very well be that SRS can be delivered with equivalent safety and efficacy by all of the systems discussed in this chapter. The evolutionary landmarks in SRS over the past 50 years include advances in computerized, real-time imaging technology, inverse planning techniques, and the ability to localize and treat both intra- and extracranial targets with submillimetric precision [3]. Frameless SRS has led to
891
892
57
Overview of radiosurgery technology
techniques for fractionation, reduced pain and discomfort to patients, and decreased requirements for anesthesia in pediatric patients [30]. The growing clinical use of IMRT for targets in the upper head-and-neck region is a direct consequence of accurate and reproducible thermoplastic mask–based stereotactic head-and-neck fixation [116]. The use of advanced digital imaging has made extracranial targets accessible to SRS [30]. As a result of new technologies in surgical robotics, some SRS devices are now equipped with enhanced computerized features, including robotic-assisted planning, robotic patient positioning, and robotic beam delivery. The OnBoard Imager of Varian Trilogy is integrated with robotic functions [3,30] just as the CK [62,79] and the updated GK Perfexion utilize robotics for pivotal functions [94]. Many of the limitations of earlier versions of the GK have also been overcome with the redesign of the Perfexion, including significant changes in the helmet and collimator [94]. In Trilogy and CK, imaging systems that rapidly and precisely calculate the target volume as well as patient positioning are fully integrated with complex automated programs that factor continuous patient movement into the treatment plan [3,21,79]. In some cases, the novelty of emerging platforms is the hybrid configuration of the instrumentation and software, as in the case of a Trilogy imaging system integrated with the CK for use in treating prostate cancer [21,76]. The clinical applications of SRS have greatly expanded since the concept was first described in 1951, and the number of published studies on clinical indications treatable with SRS has exponentially increased [30]. The advantage of one SRS system over another ultimately may depend on the planning target volume and its proximity to critical structures. Yet, the fact remains that comparisons among a broad array of specific SRS/SRT devices are lacking [117]. Evidence from clinical studies will assist in formulating a generally accepted method for evaluation of the
planning technique. As IMRT gains increased use for SRS and SRT treatment of irregularly shaped tumors, inverse planning techniques to determine dose optimization distributions have become more widely accepted [118]. Another important challenge is ensuring that research on the clinical implementation of SRS keeps pace with the commercial availability of varying technologies. Manufacturers, clinicians, and investigators must engage in discussion of the technological tools and applications of SRS instruments. The key current issues on the table include dosimetric algorithms, methods of identifying targets for treatment, and discussions of field sizes and immobilization. Only through continued clinical research can adequate data be generated to not only resolve the immediate issues at hand, but also to address new concerns that will surely arise [119]. The stereotactic method has moved out of the brain, to the spine, and completely out of the neurosurgical realm and into the rest of the body. Patients with tumors of the chest, pancreas, liver, prostate, and elsewhere are being offered SRS or SRT, a trend that will no doubt only increase. Neurosurgeons will confine their SRS treatments to patients with disorders of the nervous system and their colleagues in other surgical disciplines will learn the principles of SRS appropriate to their patients’ needs. Whether with newer Linac systems, dedicated systems like the Cyberknife, the Perfexion Gamma Knife, heavy particle irradiation, or the descendants of any of these devices, stereotactic radiosurgery will be an increasingly powerful tool in the hands of neurosurgeons. Leksell’s brainchild has become the father to one of the most important and exciting fields of medicine in the twenty-first century.
References 1. Leksell L. The stereotaxic method and radiosurgery of the brain. Acta Chir Scand 1951;102:316-9.
Overview of radiosurgery technology
2. Leksell L. Stereotactic radiosurgery. J Neurol Neurosurg Psyc 1983;46:797-803. 3. Jayarao M, Chin LS. Robotics and its applications in stereotactic radiosurgery. Neurosurg Focus 2007;23(6):E6. 4. Betti O, Derechinsky V. Hyperselective encephalic irradiation with a linear accelerator. Acta Neurochir Suppl 1984;33:385-90. 5. Betti OO, Derechinsky VE. Irradiacion hiperselectiva encefalica con acelerador lineal: Unidad Multihaz Jean Talairach. Prensa Me´dica Argentina 1983;70:102-7. 6. Colombo F, Benedetti A, Pozza F, Avanzo RC, Marchetti C, Chierego G, et al. External stereotactic irradiation by linear accelerator. Neurosurgery 1985;16:154-60. 7. Niranjan A, Gobbel GT, Kondziolka D, Lunsford LD. Heritage of radiosurgical research, current trends and future perspective. Prog Neurol Surg 2007;20:359-74. 8. Hoh DJ, Liu CY, Pagnini PG, Yu C, Wang MY, Apuzzo ML. Chained lightning, part I: Exploitation of energy and radiobiological principles for therapeutic purposes. Neurosurgery 2007;61(1):14-27. 9. Winston KR, Lutz W. Linear accelerator as a neurosurgical tool for stereotactic radiosurgery. Neurosurgery 1988;22(3):454-64. 10. Maciunas RJ. Computer-assisted neurosurgery. Clin Neurosurg 2006;53:267-71. 11. Linskey ME. The changing role of stereotaxis in surgical neuro-oncology. J Neurooncol 2004;69(1–3):35-54. 12. Housepian EM. Stereotactic surgery: the early years. Neurosurgery 2004;55(5):1210-4. 13. Barnett GH, Linskey ME, Adler JR, Cozzens JW, Friedman WA, Heilbrun MP, et al. American Association of Neurological Surgeons; Congress of Neurolofical Surgeons Washington Committee Stereotactic Radiosurgery Task Force. Stereotactic radiosurgery–an organized neurosurgery-sanctioned definition. J Neurosurg 2007;106(1):1-5. 14. Niranjan A, Gobbel G, Kondziolka D, Lunsford D. Radiosurgical research: what has it told us? what do we still need to know?. Tech Neurosurg 2003;9(3):242-50. 15. Huntzinger C, Friedman W, Bova F, Fox T, Bouchet L, Boeh L. Trilogy image-guided stereotactic radiosurgery. Med Dosim 2007;32(2):121-33. 16. Shaw E, Kline R, Gillin M. Radiation therapy oncology group: radiosurgery quality assurance guidelines. Int J Radiat Oncol Biol Phys 1993;27:1231-9. 17. Chen JC, Girvigian MR. Stereotactic radiosurgery: instrumentation and theoretical aspects – part 1. Permanente J 2005;9(4). 18. Schweikard A, Schlaefer A, Adler JR. Resampling: an optimization method for inverse planning in robotic radiosurgery. Med Phys 2006;33(11):4005-11. 19. Hoh DJ, Liu CY, Pagnini PG, Yu C, Wang MY, Apuzzo ML. Chained lightning, part I: Exploitation of energy and radiobiological principles for therapeutic purposes. Neurosurgery 2007;61(1):14-27.
57
20. Suh JH, Saxton JP. Conventional radiation therapy for skull base tumors: an overview. Neurosurg Clin N Am 2000;11(4):575-86. 21. Andrews DW, Bednarz G, Evans JJ, Downes B. A review of 3 current radiosurgery systems. Surg Neurol 2006;66 (6):559-64. 22. Carol M, Grant WH, III, Pavord D, Eddy P, Targovnik HS, Butler B, et al. Initial clinical experience with the Peacock intensity modulation of a 3-D conformal radiation therapy system. Stereotact Funct Neurosurg 1996;66 (1–3):30-4. 23. Alexander E III, Loeffler JS. Radiosurgery using a modified linear accelerator. Neurosurg Clin N Am 1992;3:167-90. 24. Gerbi BJ, Higgins PD, Cho KH, Hall WA. Linac-based stereotactic radiosurgery for treatment of trigeminal neuralgia. J Appl Clin Med Phys 2004;5(3):80-92. 25. Shiu AS, Kooy HM, Ewton JR, Tung SS, Wong J, Antes K, et al. Comparison of miniature multileaf collimation (MMLC) with circular collimation for stereotactic treatment. Int J Radiat Oncol Biol Phys 1997;37 (3):679-88. 26. Urie MM, Lo YC, Litofsky S, FitzGerald TJ. Miniature multileaf collimator as an alternative to traditional circular collimators for stereotactic radiosurgery and stereotactic radiotherapy. Stereotact Funct Neurosurg 2001;76 (1):47-62. 27. Hillard VH, Shih LL, Chin S, Moorthy CR, Benzil DL. Safety of multiple stereotactic radiosurgery treatments for multiple brain lesions. J Neurooncol 2003;63(3):271-8. 28. Plowman PN Doughty D. Stereotactic radiosurgery, X: clinical isodosimetry of gamma knife versus linear accelerator X-knife for pituitary and acoustic tumours. Clin Oncol (R Coll Radiol) 1999;11:(5)321-9. 29. McClelland S III, Gerbi BJ, Higgins PD, Orner JB, Hall WA. Safety and efficacy of fractionated stereotactic radiotherapy for acoustic neuromas. J Neurooncol 2008;86 (2):191-4. 30. Niranjan A, Lunsford LD. Radiosurgery: where we were, are, and may be in the third millennium. Neurosurgery 2000;46(3):531-43. 31. Wurm RE, Erbel S, Schwenkert I, Gum F, Agaoglu D, Schild R, et al. Novalis frameless image-guided noninvasive radiosurgery: initial experience. Neurosurgery 2008;62(5 Suppl):A11-18. 32. Jensen RL, Wendland MM, Chern SS, Shrieve DC. Novalis intensity-modulated radiosurgery: methods for pretreatment planning. Neurosurgery 2008;62 (5 Suppl):A2-9; discussion A9-10. 33. IGRT Adaptive Gating, radiotherapy solutions. 34. Yan H, Yin FF, Kim JH. A phantom study on the positioning accuracy of the Novalis Body system. Med Phys 2003;30(12):3052-60. 35. Pedroso AG, De Salles AA, Tajik K, Golish R, Smith Z, Frighetto L, et al. Novalis Shaped Beam Radiosurgery of
893
894
57 36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46. 47.
48. 49.
50.
Overview of radiosurgery technology
arteriovenous malformations. Neurosurg 2004;101 Suppl 3:425-34. Teh BS, Paulino AC, Lu HH, Chiu JK, Richardson S, Chiang S, et al. Versatility of the Novalis system to deliver image-guided stereotactic body radiation therapy (SBRT) for various anatomical sites. Technol Cancer Res Treat 2007;6(4):347-54. Chen JC, Girvigian M, Greathouse H, Miller M, Rahimian J. Treatment of trigeminal neuralgia with linear accelerator radiosurgery: initial results. J Neurosurg 2004;101 Suppl 3:346-50. Rahimian J, Chen JC, Rao AA, Girvigian MR, Miller MJ, Greathouse HE. Geometrical accuracy of the Novalis stereotactic radiosurgery system for trigeminal neuralgia. J Neurosurg 2004;101 Suppl 3:351-5. Frighetto L, De Salles A, Medin P, Selch M. Shaped beam stereotactic radiosurgery and radiotherapy for the brain and spine. Techniques in Neurosurgery. Radiosurgery 2003;9(3):204-17. Huntzinger C, Friedman W, Bova F, Fox T, Bouchet L, Boeh L. Trilogy image-guided stereotactic radiosurgery. Med Dosim 2007;32(2):121-33. Fan J, Paskalev K, Li J, Wang L, Chen L, Price R, et al. MO-D-AUD-07: determination of output Ffactors for stereotactic radiosurgery beams by Monte Carlo and measurements. Med Phys 34(6):2522. VanderSpek L, Wang J, Alksne J, Murphy K. Single fraction, single isocenter Intensity Modulated Radiosurgery (IMRS) for multiple Brain netastases: dosimetric and early clinical experience. Int J Radiat Oncol Biol Phys 2007;69(3):S265. Mamalui-Hunter M, Drzymala R, Willcut V, Santanam L, Low D. TH-D-AUD-08: Automated stereotactic radiosurgery quality assurance using a portal imager. Med Phys 2007;34(6):2642. Shiu AS, Kooy HM, Ewton JR, Tung SS, Wong J, Antes K, et al. Comparison of miniature multileaf collimation (MMLC) with circular collimation for stereotactic treatment. Int J Radiat Oncol Biol Phys 1997;37 (3):679-88. Yartsev S, Kron T, Van Dyk J. Tomotherapy as a tool in image-guided radiation therapy (IGRT): current clinical experience and outcomes. Biomed Imaging Interv J 2007;3(1):e16. Mackie TR. History of tomotherapy? Phys Med Biol 2006;51(13):R427-53. Bauman G, Yartsev S, Coad T, Fisher B, Kron T. Helical tomotherapy for craniospinal radiation. Br J Radiol 2005; 78(930):548-52. Tomotherapy Product Information. Mackie R, Balog J, Ruchala K, Shepard D, Aldridge S, Fitchard E, et al. Tomotherapy. Semin Radiat Oncol 1999;9:108-17. Baisden JM, Benedict SH, Sheng K, Read PW, Larner JM. Helical TomoTherapy in the treatment of central
51.
52.
53.
54.
55.
56.
57.
58. 59.
60.
61.
62.
63.
64.
nervous system metastasis. Neurosurg Focus 2007; 22(3):E8. Welsh JS, Lock M, Harari PM, Tome´ WA, Fowler J, Mackie TR, et al. Clinical implementation of adaptive helical tomotherapy: a unique approach to image-guided intensity modulated radiotherapy. Technol Cancer Res Treat 2006;5(5):465-79. Kim B, Soisson ET, Duma C, Chen P, Hafer R, Cox C, et al. Image-guided helical Tomotherapy for treatment of spine tumors. Clin Neurol Neurosurg 2008;110 (4):357-62. Balog J, Soisson E. Helical tomotherapy quality assurance. Int J Radiat Oncol Biol Phys 2008;71(1 Suppl): S113-7. Yartsev S, Kron T, Cozzi L, Fogliata A, Bauman G. Tomotherapy planning of small brain tumours. Radiother Oncol 2005;74(1):49-52. Shi C, Pen˜agarı´cano J, Papanikolaou N. Comparison of IMRT treatment plans between linac and helical tomotherapy based on integral dose and inhomogeneity index. Med Dosim 2008;33(3):215-21. Hansen EK, Bucci MK, Quivey JM, et al. Repeat CT imaging and replanning during the course of IMRT for head-and-neck cancerInt J Radiat Oncol Biol Phys 2006;64(2):355-62. Pen˜agarı´cano JA, Yan Y, Shi C, Linskey ME, Ratanatharathorn V. Dosimetric comparison of Helical Tomotherapy and Gamma Knife Stereotactic Radiosurgery for single brain metastasis. Radiat Oncol 2006;1:26. Zhang T. Respiratory gating and 4-D tomotherapy. Med Phys 2004;31:3529. Gutie´rrez AN, Westerly DC, Tome´ WA, Jaradat HA, Mackie TR, Bentzen SM, et al. Whole brain radiotherapy with hippocampal avoidance and simultaneously integrated brain metastases boost: a planning study. Int J Radiat Oncol Biol Phys 2007;69(2):589-97. Welsh JS, Mehta MP, Mackie TR, Orton N, Jaradat H, Khuntia D, et al. Helical tomotherapy as a means of delivering scalp-sparing whole brain radiation therapy. Technol Cancer Res Treat 2005;4(6):661-2. Roberge D, Parker W, Niazi TM, Oliveres M. Treating the contents and not the container: Dosimetric study of hair-sparing Whole Brain Intensity Modulated Radiation Therapy. Technol Cancer Res Treat 2005;4:567-70. Adler JR,Murphy MJ, Chang SD, Hancock SL. Image-guided robotic radiosurgery. Neurosurgery 1999;44:1299-307. Gildenberg PL, Woo SY. Multimodality program involving stereotactic surgery in brain tumor management. Stereotact Funct Neurosurg 2000;75(2–3):147-52. Chang SD, Main W, Martin DP, Gibbs IC, Heilbrun MP. An analysis of the accuracy of the CyberKnife: a robotic frameless stereotactic radiosurgical system. Neurosurgery 2003;52:140-7.
Overview of radiosurgery technology
65. Gerszten PC, Ozhasoglu C, Burton SA, Vogel W, Atkins B, Kalnicki S, Welch WC. Evaluation of CyberKnife frameless real-time image-guided stereotactic radiosurgery for spinal lesions. Stereotact Funct Neurosurg 2003;81(1–4):84-9. 66. Gerszten PC, Ozhasoglu C, Burton SA, Vogel WJ, Atkins BA, Kalnicki S, Welch WC. CyberKnife frameless singlefraction stereotactic radiosurgery for benign tumors of the spine. Neurosurg Focus 2003;14(5):e16. 67. Hara W, Soltys SG, Gibbs IC. CyberKnife robotic radiosurgery system for tumor treatment. Expert Rev Anticancer Ther 2007;7(11):1507-15. 68. Yu C, Gabor J, Apuzzo M, Zbigniew P. Dosimetric comparison of CyberKnife with other radiosurgical modalities for an ellipsoidal target. Neurosurgery 2003;53 (5):1155-63. 69. Romanelli P, Chang SD, Koong A, Adler John R. Extracranial radiosurgery using the CyberKnife. Techniques in neurosurgery. Radiosurgery 2003;9(3):226-31. 70. Collins SP, Coppa ND, Zhang Y, Collins BT, McRae DA, Jean WC. CyberKnife radiosurgery in the treatment of complex skull base tumors: analysis of treatment planning parameters. Radiat Oncol 2006;1:46. 71. Muacevic A, Wowra B. Cyberknife radiosurgery – overview. European neurological disease. Touch briefings; 2007. 72. Giller CA, Berger BD, Gilio JP, Delp JL, Gall KP, Weprin B, et al. Feasibility of radiosurgery for malignant brain tumors in infants by use of image-guided robotic radiosurgery: preliminary report. Neurosurgery 2004;55 (4):916-24. 73. Adler JR, Chang SD, Murphy MJ, Doty J, Geis P, Hancock SL. The Cyberknife: a frameless robotic system for radiosurgery. Stereotact Funct Neurosurg 1997;69:124-8. 74. Gibbs IC. Frameless image-guided intracranial and extracranial radiosurgery using the Cyberknife robotic system. Cancer Radiother 2006;10(5):283-7. 75. Hara W, Soltys SG, Gibbs IC. CyberKnife robotic radiosurgery system for tumor treatment. Expert Rev Anticancer Ther 2007;7(11):1507-15. 76. Pawlicki T, Cotrutz C, King C. Prostate cancer therapy with stereotactic body radiation therapy. Front Radiat Ther Oncol 2007;40:395-406. 77. Romanelli P, Schweikard A, Schlaefer A, Adler J. Computer aided robotic radiosurgery. Comput Aided Surg 2006;11(4):161-74. 78. Chang SD, Poen J, Hancock SL, Martin DP, Adler JR, Jr. Acute hearing loss following fractionated stereotactic radiosurgery for acoustic neuroma: report of two cases. J Neurosurg 1998;89:321-5. 79. Chang SD, Adler JR. Robotics and radiosurgerythe cyberknife. Stereotact Funct Neurosurg 2001;76 (3–4):204-8. 80. Shimamoto S, Inoue T, Shiomi H, Sumida I, Yamada Y, Tanaka E, et al. CyberKnife stereotactic irradiation
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
57
for metastatic brain tumors. Radiat Med 2002;20 (6):299-304. Villavicencio AT, Lim M, Burneikiene S, Romanelli P, Adler JR, McNeely L, et al. Cyberknife radiosurgery for trigeminal neuralgia treatment: a preliminary multicenter experience. Neurosurgery 2008;62(3): 647-55. Chang SD, Gibbs IC, Sakamoto GT, Lee E, Oyelese A, Adler JR Jr. Staged stereotactic irradiation for acoustic neuroma. Neurosurgery 2005;56(6):1254-61. Adler JR Jr, Gibbs IC, Puataweepong P, Chang SD. Visual field preservation after multisession cyberknife radiosurgery for perioptic lesions. Neurosurgery 2006;59 (2):244-54. Hamilton AJ, Lulu BA, Fosmire H, Stea B, Cassady JR. Preliminary clinical experience with linear acceleratorbased spinal stereotactic radiosurgery. Neurosurgery 1995;36(2):311-9. Yu C, Apuzzo MLJ, Zee CS, Petrovich Z. A phantom study of the geometric accuracy of computed tomographic and magnetic resonance imaging svereotactic localization with the Leksell stereotactic system. Neurosurgery 2001;48:1092-9. Drzymala RE, Mutic S. Stereotactic imaging quality assurance using an anthropomorphic phantom. Comput Aided Surg 1999;4:248-55. Maciunas RJ, Galloway RL Jr, Latimer JW. The application accuracy of stereotactic frames. Neurosurgery 1994;35(4):682-94. Serago CF, Lewin AA, Houdek PV, Gonzalez-Arias S, Hartmann GH, Abitbol AA, et al. Stereotactic target point verification of an X ray and CT localizer. Int J Radiat Oncol Biol Phys 1991;20:517-23. Lutz W, Winston KR, Maleki N. A system for stereotactic radiosurgery with a linear accelerator. Int J Radiat Oncol Biol Phys 1988;14:373-81. Ho AK, Fu D, Cotrutz C, Hancock SL, Chang SD, Gibbs IC, et al. A study of the accuracy of cyberknife spinal radiosurgery using skeletal structure tracking. Neurosurgery 2007;60(2 Suppl 1):147-56. Gerszten PC, Welch WC. Cyberknife radiosurgery for metastatic spine tumors. Neurosurg Clin N Am 2004;15 (4):491-501. Gerszten PC, Ozhasoglu C, Burton SA, Welch WC, Vogel WJ, Atkins BA, et al. CyberKnife frameless singlefraction stereotactic radiosurgery for tumors of the sacrum. Neurosurg Focus 2003;15(2):E7. Lindquist C, Paddick I. The Leksell Gamma Knife Perfexion and comparisons with its predecessors. Neurosurgery 2007;61(3 Suppl):130-40. Regis J, Tamura M, Guillot C, Muracciolle X, Nagaje M, Porcheron D. Radiosurgery. Radiosurgery of the head and neck with the world’s first fully robotized 192 Cobalt-60 source Leksell Gamma Knife Perfexion in clinical use. www.elekta.com/healthcare-internationalgamma-knife-surgery.php
895
896
57
Overview of radiosurgery technology
95. Levy P, Schulte RWM, Slater JD, Miller DW, Slater JM. Stereotactic radiosurgery. The role of charged particles. Acta oncologica 1999;38(2):165-9. 96. Chen CC, Chapman P, Petit J, Loeffler J. Proton radiosurgery in neurosurgery. Neurosurg Focus 2007;23 (6):E5. 97. Suit H, Goldberg S, Niemierko A, Trofimov A, Adams J, Paganetti H, et al. Proton beams to replace photon beams in radical dose treatments. Acta Oncol 2003;42 (8):800-8. 98. Schulz-Ertner D, Tsujii H. Particle radiation therapy using proton and heavier ion beams. J Clin Oncol;2007;25(8):953-64. 99. Schulz-Ertner D, Karger CP, Feuerhake A, Nikoghosyan A, Combs SE, Ja¨kel O, et al. Effectiveness of carbon ion radiotherapy in the treatment of skull-base chordomas. Int J Radiat Oncol Biol Phys 2007;68(2):449-57. 100. Levin WP, Kooy H, Loeffler JS, DeLaney TF. Proton beam therapy. Br J Cancer 2005;93(8):849-54. 101. Hug EB. Protons versus photons: a status assessment at the beginning of the 21st Century. Radiother Oncol 2004;73 Suppl 2:S35-7. 102. Weber DC, Chan AW, Bussiere MR, Harsh GR, Ancukiewicz M, Barker FG II, et al. Proton beam radiosurgery for vestibular schwannoma: tumor control and cranial nerve toxicity. Neurosurgery 2003;53: 577-88. 103. Harsh GR IV, Thornton AF, Chapman PH, Bussiere MR, Rabinov JD, Loeffler JS. Proton beam stereotactic radiosurgery of vestibular schwannomas. Int J Radiat Oncol Biol Phys 2002;54:35-44. 104. Hug EB. Review of skull base chordomas: prognostic factors and long-term results of proton-beam radiotherapy. Neurosurg Focus 2001;10(3):E11. 105. Pommier P, Liebsch NJ, Deschler DG, Lin DT, McIntyre JF, Barker FG 2nd, et al. Proton beam radiation therapy for skull base adenoid cystic carcinoma. Arch Otolaryngol Head Neck Surg 2006;132(11):1242-9. 106. Hug EB, Sweeney RA, Nurre PM, Holloway KC, Slater JD, Munzenrider JE. Proton radiotherapy in management of pediatric base of skull tumors. Int J Radiat Oncol Biol Phys 2002;52(4):1017-24. 107. Hug E, Slater J. Proton radiation therapy for chordomas and chondrosarcomas of the skull base. Neurosurg Clin North Am 2000;11:627-38. 108. Bowyer J, Natha S, Marsh I, Foy P. Visual complications of proton beam therapy for clival chordoma. Eye 2003;17 (3):318-23. 109. St Clair WH, Adams JA, Bues M, Fullerton BC, Shell S, Kooy HM, et al. Advantage of protons compared to conventional X-ray or IMRT in the treatment of a
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
pediatric patient with medulloblastoma. Int J Radiat Oncol Biol Phys 2004;58:727-34. Tsujii H, Mizoe J, Kamada T, Baba M, Tsuji H, Kato H, et al. Clinical results of carbon ion radiotherapy at NIRS. J Radiat Res (Tokyo) 2007;48 Suppl A:A1-A13. Mizoe JE, Tsujii H, Hasegawa A, Yanagi T, Takagi R, Kamada T, et al. Organizing Committee of the Central Nervous System Tumor Working Group. Phase I/II clinical trial of carbon ion radiotherapy for malignant gliomas: combined X-ray radiotherapy, chemotherapy, and carbon ion radiotherapy. Int J Radiat Oncol Biol Phys 2007; 69(2):390-6. Tsujii H, Mizoe JE, Kamada T, Baba M, Kato S, Kato H, et al. Overview of clinical experiences on carbon ion radiotherapy at NIRS. Radiother Oncol 2004;73 Suppl 2:S41-9. Munzenrider JE, Liebsch NJ. Proton therapy for tumors of the skull base. Strahlenther Onkol 1999;175 Suppl 2:57-63. Debus J, Schulz-Ertner D, Schad L, Essig M, Rhein B, Thillmann CO, et al. Stereotactic fractionated radiotherapy for chordomas and chondrosarcomas of the skull base. Int J Radiat Oncol Biol Phys 2000;47 (3):591-6. Caporaso GJ, Mackie TR, Sampayan S, Chen YJ, Blackfield D, Harris J, et al. A compact linac for intensity modulated proton therapy based on a dielectric wall accelerator. Phys Med 2008;24(2):98-101. Dietmar G, Bogner J, Dieckmann K, Po¨tter R. Is maskbased stereotactic head-and-neck fixation as precise as stereotactic head fixation for precision radiotherapy? Int J Radiat Oncol Biol Phys 2006;66(4Suppl 1):S61-6. Grabenbauer GG, Ernst-Stecken A, Schneider F, Lambrecht U, Ganslandt O. Radiosurgery of functioning pituitary adenomas: comparison of different treatment techniques including dynamic and conformal arcs, shaped beams, and IMRT. Int J Radiat Oncol Biol Phys 2006;66(4 Suppl 1):S33-9. Grzadziel A, Grosu A, Kneschaurek P. Threedimensional conformal versus intensity-modulated radiotherapy dose planning in stereotactic radiotherapy: application of standard quality parameters for plan evaluation. Int J Radiat Oncol Biol Phys 2006;66 (4 Suppl 1):S87-94. Wurm R, Okunieff P. Intracranial and extracranial stereotactic radiosurgery and radiotherapy. Int J Radiat Oncol Biol Phys 2006;66(4 Suppl 1):S1-2. Gerszten PC, Burton SA, Ozhasoglu C, McCue KJ, Quinn AE. Radiosurgery for benign intradural spinal tumors. Neurosurgery 2008;62(4):887-95; discussion 895-6.
69 Proton Beam Radiosurgery: Clinical Experience H. A. Shih . P. H. Chapman . J. S. Loeffler
Introduction The clinical benefits of proton radiation lie within the ability to deliver radiation therapy with improved conformality. This translates into a reduction of potential treatment-related adverse effects. In some situations, the improved dose distribution also permits for higher doses to be delivered without increasing the normal tissue injury probability. Despite these advantages of proton radiotherapy, injury to normal tissues still occur and include serious morbidities such as cranial neuropathies or neurocognitive dysfunction. Such injuries are often delayed in onset and may arise months to years following treatment. Neurological toxicities are perhaps most devastating in patients with benign lesions as they can significantly compromise long-term quality of life. Potential adverse effects of treatment should be carefully weighed when considering any treatment by proton radiosurgery.
Benign CNS Lesions Arteriovenous Malformations When surgery or embolization is not an option for the management of arteriovenous malformations (AVMs), radiosurgery can be considered. When applied, the purpose of radiation therapy for AVMs is to completely obliterate the nidus. This eliminates the risk of intracranial hemorrhage that otherwise occur at rate of 2–4% per year [1]. Common indications for radiosurgery #
Springer-Verlag Berlin/Heidelberg 2009
include surgical inaccessibility, expected high probability of surgical morbidity due to large AVM size or eloquent location, medically inoperable status, or patient refusal. When treating with radiosurgery, factors to be considered include AVM size, location, volume, venous drainage, and patient age [2,3]. Photon-based stereotactic radiosurgery can effectively obliterate relatively small lesions with low treatment-related toxicity and cure rates of approximately 80% [4–6]. The rate of obliteration is related to dose and increases with time. Re-irradiation of incompletely responding lesions can raise obliteration rates further without adding significant morbidity [7]. Proton radiosurgery can achieve the same success as photon-based techniques for treating small lesions, but the primary benefit of proton radiosurgery is best appreciated when treating larger unresectable AVMs. Proton radiosurgery has been utilized for this purpose often in situations when photon-based methods would be considered unsafe. Proton radiation enables safer dose escalation than photon-based therapy. Even at doses equivalent to photons, proton radiosurgery would be expected to have fewer long-term sequelae as a result of the spared excess normal tissue radiation exposure while maintaining equivalent obliteration rates. AVMs have been treated with proton radiosurgery at the Harvard Cyclotron Laboratory (HCL) since 1965 [8]. In the initial report of 75 patients followed for 2–16 years, symptomatic improvement was achieved in 75% of those who had presented with seizures or severe headaches. Of 62 patients with angiographic follow up, 20% showed complete
1132
69
Proton beam radiosurgery: clinical experience
AVM obliteration, 56% showed AVM size reduction of greater than 50%, and 13% showed no radiographic changes. Seifert et al. [9] reported a limited study of 63 patients referred to a proton treatment facility in the U.S. but who had both pre-radiation and post-radiation follow up locally. Patients were largely selected because of surgically inoperable AVMs or patient refusal of surgery. Among those patients with AVMs 3 cm in diameter, 76% of them had clinical improvement in symptoms, most commonly of seizures, headaches, or other neurological symptoms. Twelve percent of patients with small AVMs had progression of symptoms. As the size of AVM increased, the obliteration rate decreased while the rate of adverse effect increased (33 and 44.5%, respectively, for AVMs >6 cm diameter). Likewise, 86% of patients with SpetzlerMartin grade I or II AVMs had clinical improvement whereas only 54% of grade III and 24% of grade IV patients experienced improvement. The limitation of this study was the lack of treatment details of the proton radiation delivered. The iThemba LABS in the Republic of South Africa reported their proton radiosurgery experience with AVMs [10]. These investigators chose their radiation doses judiciously and frequently hypofractionated in two or four fractions for large volume lesions. At a median follow up of 62 months, they achieved a 67% obliteration rate for AVM treatment volumes of <14 cc and 43% for larger AVMs. These investigators suggest that hypofractionation can optimize normal tissue tolerance without compromising treatment efficacy. They used minimum doses converted to single fraction equivalents of approximately 15 GyE for lesions less than 14 cc and 10.4 GyE for lesions 14 cc. Grade 4 toxicity by RTOG or EORTC scales was only 3%, supporting accurate prediction of normal brain tissue tolerances. A smaller experience from Uppsala, Sweden also reported on the use of hypofractionated
proton radiosurgery in the management of AVMs [11]. Using two or four fractions to total doses of 20–25 Gy, 26 patients were treated and complete obliteration was achieved in 7 patients by 3 years. When stratified by AVM size, this data set is generally consistent with other series, with an approximately 70% obliteration rate for AVMs <25 cc and approximately 30% obliteration of AVMs >25 cc. An 85% or greater AVM size reduction was seen in 13 of the 26 patients. Clinical improvement by resolution of chronic seizures occurred in seven of nine patients. It is likely that further response will occur with longer follow up. Treatment-related morbidity of radiosurgical treatments in a highly curative patient population cannot be over emphasized. Kjellberg and colleagues determined parameters for safe radiosurgical doses based upon their early experiences of proton radiosurgery at the Harvard Cyclotron Laboratory (HCL). They readily recognized the risk of neurological complications increased with dose and AVM size and developed a model to predict complication risks, attempting to select doses based upon 1–3% isoeffective centile [8]. However, on recent retrospective analysis of the long-term complication rates from the HCL experience of 1,250 patients with available follow up, there were far greater toxicities seen than expected [12]. A 1.8% complication rate was documented among those expected to have <1% risk. Yet worse, there was a 4.7% complication rate for the 128 patients treated at a 1–1.8% predictive complication rate, and a 34% rate of complication among 61 patients treated with an expectant 2–2.5% complication risk. Clearly, the efficacy of radiosurgery may come at a price of significant adverse effects of treatment. Barker et al. [12] proposes a new model of proton radiosurgery complication risk based upon dose and volume of AVMs and this is currently applied in clinical practice at the current proton facility at the Massachusetts General Hospital (MGH).
Proton beam radiosurgery: clinical experience
Angiographically Occult Vascular Malformations Angiographically occult vascular malformations (AOVM) represent a heterogeneous group of vascular lesions such as cavernous malformations that are not detectable by cerebral angiogram. As with AVMs, the goal of proton radiosurgery is to avert significant intracranial bleeding when surgery is not a viable management option. The efficacy of radiosurgery is less well defined than for AVMs in part because there is no well-defined radiographic endpoint because these lesions are angiographically occult. There is one report on the application of proton beam radiosurgery for AOVMs [13]. Among 98 lesions with radiographic findings consistent with cavernous malformations based upon imaging and clinical history, the risk of hemorrhage was reduced from 17.3 to 4.5% per year after a two year latency period from treatment. Complication rates were higher than comparative radiosurgery series with AVMs. Sixteen percent of patients suffered a permanent neurological deficit and there was a 3% mortality rate. Whereas proton radiosurgery appears to successfully reduce the hemorrhagic tendency of AOVMs, this reduction is achieved with a significant rate of complication at equivalent radiation doses to those used for AVM treatment.
Vestibular Schwannomas (Acoustic Neuromas) Vestibular schwannomas, also known as acoustic neuromas, are benign tumors that arise from Schwann cells of the vestibular nerve within the cerebellopontine angle. Although less common, schwannomas may arise from other cranial nerve as well, with the exception of the optic and olfactory nerves. Progressive growth of vestibular tumors can lead to symptoms such as hearing loss, tinnitis, dizziness, facial numbness
69
and weakness. When surgery is not a reasonable management option, radiation can be used to control further tumor growth with high success rates. Treatments can be delivered with either fractionated radiation or single fraction radiosurgery. The decision to treat with either fractionated or radiosurgery form depends upon tumor size, neurological function, and patient preference. When treating small target sizes (2 cm), the benefit of proton radiation as compared to photon-based stereotactic treatments is less clear and likely comparative. Fractionated radiotherapy is typically favored when useful hearing exists [14]. When hearing preservation is not a concern, radiosurgery is the preferred treatment method because of equivalent efficacy and better patient convenience. Either proton or photon radiation can be applied successfully with similar control rates of roughly 95% at 5 and 10 years [15–18]. Single fraction treatments are generally prescribed to the minimum target dose of 12 Gy. This relatively well accepted dose is the result of reported neurological injury following treatment to higher doses [19]. Following radiation treatment, tumors may slightly reduce in size but do not disappear. The use of proton radiosurgery in managing vestibular schwannomas is limited. Harsh et al. [15] reported on the proton radiosurgical experience at the HCL for vestibular schwannomas that was subsequently updated by Weber et al. [18]. Among 88 patients treated by proton stereotactic radiosurgery with a median tumor volume of 1.4 cc, the median dose delivered was 12 GyE prescribed to the 70% isodose. Actuarial local control rates at 2 and 5 years were 95.3 and 93.6%, respectively. Five-year actuarial injury to the facial and trigeminal nerves were 9.9 and 10.6%, respectively. In subjects with pretreatment useful hearing, actuarial hearing preservation rate was 79.1% at 2 years but declined to 21.9% by 5 years, reflecting characteristic late effect injury of radiation. Facial nerve injury
1133
1134
69
Proton beam radiosurgery: clinical experience
did not occur with target doses of less than 17 GyE. In other reports, facial and trigeminal neuropathies may occur at rates of approximately 5–10% with radiosurgical methods but the risk is also dependent upon the size of treatment volume. It should not be overlooked that even with proton radiation therapy, fractionation is able to preserve facial and trigeminal nerve function at 99 and 96%, respectively [16]. In summary, published data on the application of proton radiosurgery for vestibular schwannomas have yet to show a clinical advantage over non-proton methods. This is likely due to limited experiences. Larger tumors are inherently associated with greater radiation dose deposition to neighboring normal tissues. Although late effects are low and risk for radiation-induced carcinogenesis is small, these risks should be further reduced or obviated with the use of proton radiation therapy.
Pituitary Adenomas Both functional and nonfunctional pituitary adenomas can benefit from radiation therapy when options for surgical and medical therapy are insufficient or not feasible. The role for radiosurgery is best utilized with tumors or residual tumors that do not pose threat of injury to neighboring important neurological tissues such as the optic chiasm, cranial nerves within the cavernous sinus, and brain parenchyma. Fractionated radiation therapy is the preferred modality when these factors are of concern. Proton radiosurgery has a particularly useful role in the radiation management of pituitary adenomas in the subset of functionally active tumors. Functioning tumors are typically hormonally active with excess secretion of one or more pituitary hormones. Depending on the hormone secreted, the clinical manifestations may be Cushing’s disease, acromegaly, hyperprolactinemia, or hyperthyroidism. Secreting tumors
are often microadenomas (less than 1 cm in diameter). Stereotactic radiosurgery enables a faster hormonal ablative response for secretory tumors than with fractionated radiation treatment [20]. Thus, in patients with hormonally active tumors at least 5 mm from the chiasm, radiosurgery is usually the preferred option. The surface dose to the optic system should be kept under 8 Gy. Local tumor control is comparative to fractionated radiation and expectant similar between proton and photon-based radiation modalities. An early report of the MGH proton experience with treatment of acromegaly established the initial efficacy of proton radiosurgery [21]. Current updates have also been favorable. Of 22 patients with persistent acromegaly treated with single fraction proton radiosurgery to a median dose of 20 GyE, 95% have achieved at least a partial response at 6 years and 50% have had a complete response [22]. Median time to complete response was 30.5 months. Thus far, a third of patients have developed at least one new pituitary deficiency that is correctable with supplementation. Similarly, among 38 patients with persistent Cushing’s disease following surgery treated with a median dose of 20 GyE, a 79% response rate was achieved with a complete response seen in 50% of patients [23]. Median time to response was 14 months. New pituitary deficiency occurred in 36% of patients thus far at a median follow up of 38 months. Regardless of the radiation therapy modality implemented, patients should be counseled regarding the importance of regular endocrinological evaluation. Over a third of patients will develop a new hormonal deficiency and this risk increases with time [22,23]. These deficiencies are easily correctable with pharmacotherapy.
Meningiomas The majority of meningiomas are benign and ideally managed by resection. However, radiation
Proton beam radiosurgery: clinical experience
therapy is a useful modality of adjuvant or primary management when tumors are unresectable or subtotally resected and are either associated with or at risk of causing symptoms. This is frequently the case with skull base tumors in inaccessible locations such as the cavernous sinus. Experience with fractionated or single fraction photon radiotherapy in the primary or adjuvant treatment of benign meningiomas has shown local control rates at 5 years of 89–96% with complication rates of 2–5% when doses are selected with normal tissue tolerances respected [24–28]. As with other tumors, the advantage of protons is greatest in the management of large or irregularly shaped tumors. The role of proton radiosurgery as compared to fractionated protons is best defined for small to intermediate sized tumors, particularly those of irregular contours. These tumors can be treated with a single fraction or hypofractionated in few fractions with variable doses depending upon the size, location, clinical history, and patient performance status. Limited hypofractionated data support this approach. Gudjonsson et al. [29] reported on 19 patients with largely skull base meningiomas treated with hypofractionated proton radiotherapy at the Svedberg Laboratory in Uppsala, Sweden. Treatment was 24 GyE divided into four daily fractions. Minimum follow up has been 36 months for all patients with no local failures or new neurological symptoms to date. Follow up time is relatively short but this study nonetheless suggests that hypofractionated schedules might be an excellent treatment option with reduced risk of neurological sequelae. Results with longer follow up will be useful. Another report of using hypofractionated stereotactic proton radiotherapy comes from the National Accelerator Center [30]. Eighteen patients with skull base meningiomas were treated in three fractions to a mean dose of 20.3 GyE. With a mean follow-up period of 40 months, 16/18 (89%) of patients remained at least clinically stable whereas two patients showed
69
tumor progression. Two patients developed complications related to radiation treatment.
Malignant CNS Lesions Brain metastases have been increasing treated with proton radiosurgery at the MGH and possibly other proton radiation facilities. Unfortunately there is no current published reports that describes the long-term outcomes of this treatment but from our clinical experience, we feel proton radiosurgery provides equivalent therapeutic value with minimal risk of adverse effects when treating patients with small intracranial brain metastases. For patients that have larger lesions or irregularly shaped tumors such as those recurring along a resection bed, proton radiosurgery provides a superior dosimetric therapy than photon-based radiosurgery techniques. Clinical data is expected to be forthcoming within the next few years. Primary gliomas and other primary malignant brain tumors are inherently invasive into the brain parenchyma, making fractionation of radiotherapy critical to minimize potential radiation injury to the normal brain tissue. In addition, these cases typically involve treatment volumes that are far larger than those of brain metastases due to the inclusion of generous margins to encompass microscopic extensions. Thus, there is currently not an accepted clinical application of proton radiosurgery in primary malignant brain tumors.
Treatment-Related Adverse Effects in CNS Tissues Radiation therapy to intracranial targets is in part defined by neighboring normal brain tissue tolerances. Despite technological advancements in radiation planning and delivery that have markedly improved the conformality of radiation
1135
1136
69
Proton beam radiosurgery: clinical experience
delivery such as with the development of proton radiation, there continue to be unavoidable risks of radiation because of the embedded nature of tumor targets in the otherwise normal brain. Risks of radiation adverse effects are defined by probability and do not occur at well defined thresholds that are predictable or uniform across all patients. As neurological side effects can be significantly debilitating if not life threatening, radiation treatment planning and dose selection must account for multiple factors. The clinician must consider the radiosensitivity of the tumor target, volume of target, volume of normal brain being irradiated, proximity to more sensitive or critical cranial structures, and multiple patient factors such as the performance status, prognosis, and co-morbidities that may affect radiation tolerance. Most acute side effects are limited, reasonably well-tolerated, and resolve with time. These include focal alopecia, skin erythema and dryness, fatigue, mild headaches, and nausea. Hypoallergenic topical creams may be applied for irritated or dry skin. Headaches are typically mild and persistent ones are most often remedied with over the counter pain medications. Nausea is treated with standard antiemetics. If headaches or nausea are thought to be related to significant cerebral edema, short course steroid therapy is an appropriate management. Truly debilitating neurological symptoms such as impairment to vision, hearing, motor or sensory function, vestibular function, or neurocognitive status are late effect concerns. These may begin to manifest about 6 months following completion of radiation therapy to many years removed from radiation treatment. Some symptoms are temporary and will resolve either untreated or with steroid therapy over weeks to months. Symptoms persisting beyond 12 months are more likely to be permanent as a result of irreversible injury to the tissues. From photon radiosurgery literature, 8–10 Gy in a single fraction is the generally accepted
dose tolerance to the optic chiasm and nerves. Single fraction dose to the brainstem should ideally not exceed 12 Gy to more than 1 cc of normal tissue. Because these are not absolute safe thresholds, dose to these critical structures should always be minimized well below even these limits whenever possible. Although pituitary dysfunction is correctable with hormonal supplementation, avoidance of unnecessary endocrine dysfunction is worthwhile. This is affected by irradiation to either or both the hypothalamus and pituitary gland [31]. As with any radiation therapy, risks of toxicity should be discussed in detail with patients prior to acceptance for treatment.
References 1. Choi JH, Mohr JP. Brain arteriovenous malformations in adults. Lancet Neurol 2005;4(5):299-308. 2. Spetzler RF, Martin NA. A proposed grading system for arteriovenous malformation. J Neurosurg 1986;65:476-83. 3. Pollock BE, Flickinger JC. A proposed radiosurgerybased grading system for arteriovenous malformations. J Neurosurg 2002;96(1):79-85. 4. Pollock BE, Lunsford LD, Kondziolka D, et al. Patient outcomes after stereotactic radiosurgery for ‘‘operable’’ arteriovenous malformations. Neurosurgery 1994;35(1): 1-7. 5. Pan DH, Guo WY, Chung WY, et al. Gamma knife radiosurgery as a single treatment modality for large cerebral arteriovenous malformations. J Neurosurg 2000;93 3:113-9. 6. Schlienger M, Atlan D, Lefkopoulos D, et al. Linac radiosurgery for cerebral arteriovenous malformations: results in 169 patients. Int J Radiat Oncol Biol Phys 2000;46(5):1135-42. 7. Mirza-Aghazadeh J, Andrade-Souza YM, Zadeh G, et al. Radiosurgical retreatment for brain arteriovenous malformation. Can J Neurol Sci 2006;33(2):189-94. 8. Kjellberg RN, Hanamura T, Davis KR, et al. Bragg-peak proton-beam therapy for arteriovenous malformations of the brain. N Engl J Med 1983;309:269-74. 9. Seifert V, Stolke D, Mehdorn HM, et al. Clinical and radiological evaluation of long-term results of stereotactic proton beam radiosurgery in patients with cerebral arteriovenous malformation. J Neurosurg 1994; 81:683-9. 10. Vernimmen FJAI, Slabbert JP, Wilson JA, et al. Stereotactic proton beam therapy for intracranial arteriovenous
Proton beam radiosurgery: clinical experience
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
malformations. Int J Radiat Oncol Biol Phys 2005; 62(1):44-52. Silander H, Pellettieri L, Enblad P, et al. Fractionated, stereotactic proton beam treatment of cerebral arteriovenous malformations. Acta Neurol Scand 2004;109:85-90. Barker FG II, Butler WE, Lyons S, et al. Dose-volume prediction of radiation-related complications after proton beam radiosurgery for cerebral arteriovenous malformations. J Neurosurg 2003;99:254-63. Amin-Hanjani S, Ogilvy CS, Candia GJ, et al. Stereotactic radiosurgery for cavernous malformations: Kjellberg’s experience with proton beam therapy in 98 cases at the Harvard Cyclotron. Neurosurg 1998;42:1229-36. Andrews DW, Suarez O, Goldman HW, et al. Stereotactic radiosurgery and fractionated stereotactic radiotherapy for the treatment of acoustic schwannomas: comparative observations of 125 patients treated at one institution. Int J Radiat Oncol Biol Phys 2001;50(5):1265-78. Harsh GR, Thornton AF, Chapman PH, et al. Proton beam stereotactic radiosurgery of vestibular schwannomas. Int J Radiat Oncol Biol Phys 2002;54(1):35-44. Chan AW, Black P, Ojemann RG, et al. Stereotactic radiotherapy for vestibular schwannomas: favorable outcome with minimal toxicity. Neurosurg 2005;57(1): 60-70. Lunsford LD, Niranjan A, Flickinger JC, et al. Radiosurgery of vestibular schwannomas: summary of experience in 829 cases. J Neurosurg 2005;102:195-9. Weber DC, Chan AW, Bussiere MR, et al. Proton beam radiosurgery for vestibular schwannoma: tumor control and cranial nerve toxicity. Neurosurgery 2003;53:577-88. Miller RC, Foote RL, Coffey RJ, et al. Decrease in cranial nerve complications after radiosurgery for acoustic neuromas: a prospective study of dose and volume. Int J Radiat Oncol Biol Phys 1999;43(2):305-11. Landolt AM, Haller D, Lomax N, et al. Stereotactic radiosurgery for recurrent surgically treated acromegaly: comparison with fractionated radiotherapy. J Neurosurg 1998;88(6):1002-8.
69
21. Kjellberg RN, Shintani A, Frantz AG, et al. Protonbeam therapy in acromegaly. N Engl J Med 1968; 278(13):689-95. 22. Petit JH, Biller BMK, Swearingen B, et al. Proton stereotactic radiosurgery is effective and safe in the management of persistent acromegaly. The Endocrine Society’s 88th Annual Meeting, Boston, MA, June 24–27, 2006. 23. Petit JH, Biller BMK, Swearingen B, et al. Proton stereotactic radiosurgery is effective and safe in the management of Cushing’s disease. The Endocrine Society’s 88th Annual Meeting, Boston, MA, June 24–27, 2006. 24. Goldsmith BJ, Wara WM, Wilson CB, et al. Postoperative irradiation for subtotally resected meningiomas. J Neurosurg 1994;80:195-201. 25. Friedman WA, Murad GJ, Bradshaw P, et al. Linear accelerator surgery for meningiomas. J Neurosurg 2005;103(2):206-9. 26. Hakim R, Alexander E, Loeffler J, et al. Results of linear accelerator-based radiosurgery for intracranial meningiomas. Neurosurg 1998;42(3):446-54. 27. Flickinger JC, Kondziolka D, Maitz AH, et al. Gamma knife radiosurgery of imaging-diagnosed intracranial meningioma. Int J Radiat Oncol Biol Phys 2003;56(3): 801-6. 28. Debus J, Wuendrich M, Pirzkall A, et al. High efficacy of fractionated stereotactic radiotherapy of large base-ofskull meningiomas: long-term results. J Clin Oncol 2001;19(15):3547-53. 29. Gudjonsson O, Blomquist E, Nyberg G, et al. Stereotactic irradiation of skull base meningiomas with high energy protons. Acta Neurochi (Wien) 1999;141:933-40. 30. Vernimmen FJ, Harris JK, Wilson JA, et al. Stereotactic proton beam therapy of skull base meningiomas. Int J Radiat Oncol Biol Phys 2001;49(1):99-105. 31. Pai HH, Thornton A, Katznelson L, et al. Hypothalamic/ pituitary function following high-dose conformal radiotherapy to the base of skull: demonstration of a dose-effect relationship using dose-volume histogram analysis. Int J Radiat Oncol Biol Phys 2001;49(4):1079-92.
1137
61 Proton Beam Radiotherapy: Technical and Clinical Aspects S. Y. Woo
What is a Proton? Proton is the positively charged particle in the nucleus of a hydrogen atom.
History of Proton Therapy Rutherford first proposed the existence of protons in 1919. In 1946 Robert Wilson proposed the use of protons for the therapy of tumors. Proton therapy on patients was begun at the Lawrence Berkley laboratory in 1955 and at the Harvard Cyclotron in 1961. The first hospital based proton facility was established at Loma Linda University Medical Center in 1991. Currently there are 25–30 facilities worldwide treating patients and over 35,000 patients have been treated with protons.
Physics of Proton Therapy Protons are positively charged particles that, after entering the body, will travel a finite distance depending in part on their initial energy. After entering the body they deposit relatively little energy and produce relatively little ionization along their path until the end of the path where the majority of ionization (Bragg Peak) occurs. Therefore with a proton beam there is less radiation dose to the normal tissue in front of the tumor and hardly any radiation dose to the normal tissue behind it. This is in contrast with photons or X-rays which are commonly used for #
Springer-Verlag Berlin/Heidelberg 2009
radiation therapy. As soon as photons enter the body, they deposit their energy and ionize the cells. In the process they lose their energy exponentially along the path. If a tumor is at a certain depth in the body, a photon beam actually deposits more energy in front of the tumor than in the tumor itself. In addition, energy is being continuously deposited behind the tumor (exit dose). > Figure 61-1 illustrates the difference between a proton beam and a photon beam. The unique property of protons allows proton therapy to spare more normal tissues than photon therapy (> Figure 61-2).
Production of Protons for Therapy When a hydrogen atom is stripped off, using electric fields, its orbiting negatively charged electron, what remains is a proton. In a treatment facility, the protons are usually initially accelerated in a linear accelerator before being injected into a cyclotron or a synchrotron where they are accelerated to the required energy for treatment. A cyclotron will accelerate protons to the highest energy of that particular system. If protons of lower energies are required, an energy degrading process is used to accomplish that. A synchrotron, on the other hand, can produce protons of a specified energy in one pulse and change the energy on every pulse. The exiting narrow beam of high energy protons from either a cyclotron or a synchrotron can then be either spread out by various scattering devices (scattered protons), or deflected by special magnets to scan across the
958
61
Proton beam radiotherapy: technical and clinical aspects
. Figure 61-1 Dosimetric advantange of protons
vertical or horizontal dimensions of a tumor (scanning protons). Currently most treatment facilities use the scattered proton beams for therapy but increasing number of sites are investigating the use of scanning proton beams because of the potential of improved dose distributions over scattered proton beams.
Clinical Experience of Proton Therapy in Brain and Skull-base Tumors Meningioma Approximately 90–95% of intracranial meningiomas are benign, 5% are atypical and 3% malignant. Surgery is the primary treatment. However, incomplete resection can result in high recurrence rate over time (up to 55% at 10 years and 91% at 15 years) [1]. Tumors at certain locations such as
the cavernous sinus, petro-clival region or sphenoid wing are not generally suitable for aggressive surgery because of the significant morbidity associated with it. Several retrospective series have demonstrated a reduction of recurrence after incomplete surgery or good local control with radiation therapy [2–5]. Because benign meningiomas are usually well delineated on MRI, they are well suited to be irradiated by highly conformal radiation therapy techniques. Proton therapy is such a technique. When compared to other photon based conformal techniques, proton therapy could reduce the volume of normal brain receiving a low to moderate dose of radiation because of the physical characteristics of protons as described above. The reported 3–5 year control rate of recurrent, incompletely resected meningiomas(including a few cases of atypical and malignant histologies) with proton therapy(+ or a component of photon therapy) to doses of 50–74 CGE is 80–100% [6–9]. > Table 61-1 summarizes the result. Late
Proton beam radiotherapy: technical and clinical aspects
61
. Figure 61-2 Medulloblastoma
. Table 61-1 Results of proton therapy for meningiomas Investigators
Follow-up (months)
Local control
Survival
Noel et al.a Vernimmen et al. Wenkel et al. Weber et al.
37 (median) 40 (mean) 53 (median) 34
4-year: 87.5% +/ 12% 89% 93% 10-year: 77% 3-year: 91.7%
4-year: 88.9% +/ 11% – 5-year: 93% 3-year: 92.7%
Reference [6] [7] [8] [9]
a
Includes a few patients with atypical/malignant meningiomas
complication rates were reported as 10–25%. For atypical and malignant meningiomas specifically, one series from Boston reported 5-and 8-year control rates of 38%(atypical) and 52%(malignant), and 19%(atypical) and 17%(malignant) respectively [10]. The investigators noted significantly improved local control for proton therapy versus photon radiation therapy (80% vs. 17% at 5 years, p = 0.003).
Craniopharygioma Craniopharyngioma is curable with complete resection. However, in many cases a gross total resection is not possible without a significant morbidity because of the attachment of the tumor to critical structures such as the optic nerve/chiasm and hypothalamus. Radiation therapy after tumor cyst decompression and biopsy
959
960
61
Proton beam radiotherapy: technical and clinical aspects
has been reported to produce an equivalent curerate as gross total resection [11–13]. In one retrospective series, the neuron-cognitive outcome was reported to be better with radiation therapy than with surgery [14]. Proton therapy has been used in a small number of patients with reported 5-year and 10-year local control rates of 93 and 85% respectively [15,16]. The functional status of the living adult patients was reported as unaltered from the pre-radiotherapy status. One of five children showed some learning difficulties [15]. One of twelve patients in another series developed new-onset panhypopituitarism [16]. Craniopharyngioma is currently considered suitable for proton therapy.
Pituitary Adenoma Surgery is the primary treatment for pituitary adenoma. Residual tumor (macroscopic or microscopic) after surgery can be present if the tumor invades the cavernous sinus, has a large supraseller extension, or significantly erodes the bony floor of the sella. Post-operative radiation therapy in doses of 45–50.4 Gy has produced a 10-year relapse-free survival rate higher than 90% [17]. Currently one of the conformal radiation techniques is recommended. The experience with proton therapy is limited. One series with relatively short follow-up reported no progression of tumor in 44 out of 47 patients [18]. The only morbidity was hypopituitarism.
Low-grade Glioma Low-grade glioma is usually subjected to resection. If a gross total resection is achieved, adjuvant radiation therapy is generally not necessary. Even if the resection is incomplete, if the tumor is not in a eloquent area where a recurrence could cause significant functional deficit, radiation therapy can be postponed till evidence of recurrence. This
approach is more relevant in children because the delay in radiation therapy allows for the further development of the young brain. A small experience of conformal proton therapy for 27 children with progressive or recurrent low-grade astrocytoma has been reported [19]. The target doses were between 50.4 and 63 CGE (Cobalt Gray Equivalent). The mean follow-up period was 3.3 years (0.6–6.8 years). The local control and survival was respectively 87 and 93% for central tumors, 71 and 86% for hemispheric tumors and 60 and 60% for brainstem tumors. All children with local control were reported to have maintained their performance status. All children with optic pathway tumors maintained or improved their visual status. One child with Type 1 neurofibromatosis developed Moyamoya syndrome. Because proton therapy, when compared to photon therapy reduces the volume of normal brain receiving low to modest dose of radiation, it has at least a theoretical advantage in children [20].
Medulloblastoma Radiation therapy is a very important modality in any potentially curative treatment program for medulloblastoma. Because medulloblastoma has a propensity to spread in the leptomeningeal space, irradiation of the entire craniospinal axis is usually required. Additional radiation is directed to the tumor bed after craniospinal irradiation. Several dosimetric studies comparing protons with photons showed a clear advantage of proton therapy in its ability to reduce the radiation dose to the eyes, optic chiasm, hypothahamus-pituitary axis, cochleae,thyroid, lung, heart, liver and kidneys [21,22,23]. In addition, one study estimated that proton therapy for medulloblastoma could reduce the risk of second malignancy by ten folds when compared to photon therapy [24]. A recent costeffective evaluation concluded that proton therapy could be cost-effective and cost-saving compared with conventional photon radiation therapy in the
Proton beam radiotherapy: technical and clinical aspects
treatment of children with medulloblastoma [25]. Increasing number of children with medulloblastoma are currently being treated with proton therapy. Prospective studies to evaluate the late morbidity and quality of life of children treated with proton craniospinal irradiation are underway.
Chordoma/Chondrosarcoma Chordoma and chondrosarcoma at the base of skull usually present with therapeutic challenges. Although surgery is the mainstay of treatment, total resection is not always possible because of the adjacent vital structures. Post-operative radiation therapy for subtotally resected tumor is usually indicated. Conventional photon radiation therapy to moderately high doses of radiation (50–55Gy) has produced local recurrence rate of 80–100% [26,27]. Higher doses of radiation (66–80 CGE) delivered by proton therapy have produced 3-year and 5-year local control rates of 91.6–94%, and 75% respectively for chondrosarcoma. For chordoma the 3- to 4-year local control rates have been reported to be 53.8–87.3%, and the 3- to 5-year survival rates have been reported to be 80.5– 93.8% [28,29,30,31] (> Table 61-2). In pediatric patients, the local control rates for chordoma and chondrosarcoma have been reported to be 60 and 100% respectively [32]. Proton therapy has been accepted as the standard treatment for chordoma and chondrosarcoma. Although occasionally chordomas can metastasize to lymph
61
nodes, lungs or bones, or recur along the surgical pathway, the predominant failure pattern for chordomas of the base of skull is local, and savage therapy after a local failure is generally unsuccessful. Thus aggressive upfront treatment is justified. These high doses deliverable with proton therapy nonetheless have been associated with significant complications in a minority of patients. The rate of significant complications such as cranial nerve injury or brain necrosis has been reported to be 7–18% [30,33]. However endocrinopathy has been reported in a significant number of patients [34].Currently 70 CGE is deemed appropriate for chondrosarcoma. The best radiation dose for chordoma is still under study. There are ongoing two prospective randomized studies comparing 70 CGE against 78 CGE, and 72 CGE against 80 CGE.
Conclusions The physical characteristics of protons render them attractive particles for radiation therapy of many tumors in the brain and base of skull, especially in children. The new technologies of proton therapy such as scanning beam and intensitymodulated proton therapy, couple with new images which can more clearly delineate tumors, tumor subpopulations and specific functional areas of the brain will bring about exciting investigations that hopefully will significantly improve the therapeutic ratio of radiation therapy for future patients.
. Table 61-2 Results of proton therapy for chordomas Investigators
Follow-up (months)
Local control
Survival
Reference
Fagundes et al. Hug et al. Noel et al. Weber et al.
54 (median) 33 (mean) 31 (median) 29 (median)
71% 76% 4-year: 53.8% 3-year: 87.3%
– 79% 5-year: 80.5% 3-year: 93.8%
[28] [29] [30] [31]
961
962
61
Proton beam radiotherapy: technical and clinical aspects
References 1. Mirimanoff RO, Dosoretz D, Linggood R, et al. Meningioma: analysis of recurrence and progression following neurosurgical resection. J Neurosurg 1985;62:18-24. 2. Goldsmith BJ, Wara WM, Wilson CB, et al. Postoperative irradiation for subtotally resected meningiomas. A retrospective analysis of 140 patients treated from 1967 to 1990. J Neurosurg 1994;80:195-201. 3. Nutting C, Brada M, Brazil L, et al. Radiotherapy in the treatment of benign meningiomas of the skull base. J Neurosurg 1999;90:823-7. 4. Pourel N, Auque J, Bracard S, et al. Efficacy of external fractionated radiation therapy in the treatment of meningiomas: a 20-year experience. Radiother Oncol 2001;61:65-70. 5. Uy NW, Woo SY, Teh BS, et al. Intensity-modulated radiation therapy (IMRT) for meningiomas. Int J Radiat Oncol Biol Phys 2002;53:1265-70. 6. Noel G, Habrand JL, Mammar H, et al. Highly conformal therapy using proton component in management of meningiomas. Preliminary experience of the Centre de Protontherapie d’Orsay. Srahlenther Onkol 2002;178 (9):480-5. 7. Vernimmen FJ, Harris JK, Wilson JA, et al. Stereotactic proton beam therapy of skull base meningiomas. Int J Radiat Oncol Biol Phys 2001;49(1):99-105. 8. Wenkel E, Thornton AF, Finkelstein D, et al. benigh meningioma: partially resected, biopsies, and recurrent intracranial tumors treated with combined proton and photon radiotherapy. Int J Radiat Oncol Biol Phys 2000;48(5):1363-70. 9. Weber DC, Lomax AJ, Rutz HP, et al. Spot-scanning proton radiation therapy for recurrent, residual or untreated intracranial meningiomas. Radiother Oncol 2004;71(3):251-8. 10. Hug EB, Devries A, Thornton AF, et al. Management of atypical and malignant meningiomas: role of high-dose, 3D-conformal radiation therapy. J Neurosurg 2000;48 (2):151-60. 11. Weiss M, Sutton L, Marcial V, et al. The role of radiation therapy in the management of childhood craniopharyngioma. Int J Radiat Oncol Biol Phys 1989;17:1313-21. 12. Minniti G, Saran F, Traish D, et al. Fractionated stereotactic conformal radiotherapy following conservative surgery in the control of craniopharyngiomas. Radiother Oncol 2007;82(1):90-5. 13. Combs SE, Thilmann C, Huber PE, et al. Achievement of long-term local control in patients with craniopharyngiomas using high precision stereotactic radiotherapy. Cancer 2007;109(11):2308-14. 14. Merchant TE, Kiehna EN, Kun LE, et al. Phase II trial of conformal radiation therapy for pediatric patients with craniopharygioma and correlation of surgical factors and radiation dosimetry with change in cognitive function. J Neurosurg 2006;104(2 Suppl):94-102.
15. Luu QT, Loredo LN, Archambeau JO, et al. Fractionated proton radiation treatment for pediatric craniopharyngioma: preliminary report. Cancer J 2006;12(2):155-9. 16. Fitzek MM, Linggood RM, Adams J, et al. Combined proton and photon irradiation for craniopharyngioma: long-term results of the early cohort of patients treated at Harvard Cyclotron Laboratory and Massachusetts General Hospital. Int J Radiat Oncol Biol Phys 2006;64 (5):1348-54. 17. Brada M, Rajan B, Traish D, et al. The long-term efficacy of conservative surgery and radiotherapy in the control of pituitary adenomas. Clin Endocrinol (Oxf) 1993;38:571-8. 18. Ronson BB, Schulte RW, Han KP, et al. Fractionated proton beam irradiation of pituitary adenomas. Int J Radiat Oncol Biol Phys 2006;64(2):425-34. 19. Hug EB, Muenter MW, Archambeau JO, et al. Conformal proton radiation therapy for pediatric low-grade astrocytomas. Strahlenther Onkol 2002;178(1):10-17. 20. Bolsi A, Fogliata A, Cozzi L. Radiotherapy of small intracranial tumours with different advanced techniques using photon and proton beams: a treatment planning study. Radiother Oncol 2003;68(1):1-14. 21. St Clair WH, Adams JA, Bues M, et al. Advantage of protons compared to conventional X-ray or IMRT in the treatment of a pediatric patient with medulloblastoma. Int J Radiat Oncol Biol Phys 2004;58(3): 727-34. 22. Lee CT, Bilton SD, Famiglietti RM, et al. Treatment planning with protons for pediatric retinoblastoma, medulloblastoma, and pelvic sarcoma: how do protons compare with other conformal techniques? Int J Radiat Oncol Biol Phys 2005;63(2):362-72. 23. Timmermann B, Lomax AJ, Nobile L, et al. Novel technique of craniospinal axis proton therapy with the spotscanning system: avoidance of patching multiple fields and optimized ventral dose distribution. Strahlenther Onkol 2007;183(12):685-8. 24. Miralbell R, Lomax A, Cella L, et al. Potential reduction of the incidence of radiation-induced second cancers by using proton beams in the treatment of pediatric tumors. Int J Radiat Oncol Biol Phys 2002;54(3):824-9. 25. Lundkvist J, Ekman M, Ericsson SR, et al. Costeffectiveness of proton radiation in the treatment of childhood medulloblastoma. Cancer 2005;103(4):793-801. 26. Catton C, O’Sullivan B, Bell R, et al. Chordoma: longterm follow-up after radical photon irradiation. Radiother Oncol 1996;41(1):67-72. 27. Colli B, Al-Mefty O. Chordomas of the craniocervical junction: follow-up review and prognostic factors. J Neurosurg 2001;95(6):933-43. 28. Fagundes MA, Hug EB, Liebsch NJ, et al. Radaition therapy for chordomas of the base of skull and cervical spine: patterns of failure and outcome after relapse. Int J Radiat Oncol Biol Phys 1995;33(3):579-84.
Proton beam radiotherapy: technical and clinical aspects
29. Hug EB, loredo LN, Slater JD, et al. Proton radiation therapy for chordomas and chondrosarcomas of the skull base. J Neurosurg 1999;91(3):432-9. 30. Noel G, Feuvret L, Calugaru, et al. Chordomas of the base of skull and upper cervical spine. One hundred patients irradiated by a 3D conformal technique combining photon and proton beams. Acta Oncol 2005; 4497:700-8. 31. Weber DC, Rutz HP, Pedroni ES, et al. results of spotscanning proton radiation therapy for chordoma and chondrosarcoma of the skull base: the Paul Scherrer Institute experience. Int J Radiat Oncol Biol Phys 2005;63 (2):401-9.
61
32. Hug EB, Sweeney RA, Nurre PM, et al. Proton radiotherapy in management of pediatric base of skull tumors. Int J Radiat Oncol Biol Phys 2002;52(4):1017-24. 33. Santoni R, Liebsch N, Finkelstein DM, et al. Temporal lobe (TL) damage following surgery and high-dose photon and proton irradiation in 96 patients affected by chordomas and chondrosarcomas of the base of skull. Int J Radiat Oncol Biol Phys 1998;41(1):59-68. 34. Pai HH, Thornton A, Katznelson L, et al. Hypothalamic/ pituitary function following high-dose conformal radiotherapy to the base of skull: demonstration of a dose-effect relationship using dose-volume histogram analysis. Int J Radiat Oncol Biol Phys 2001;49(4):1079-92.
963
Stereotactic Radiosurgery
56 Radiobiology of Stereotactic Radiosurgery D. C. Shrieve . J. S. Loeffler
Radiobiology is the study of the effects of radiation on biological systems. Ionizing radiation is known to cause biological effects due to a cascade of events occurring in rapid sequence following the irradiation. An understanding of the biophysical and radiobiological principles governing effects following the administration of therapeutic radiation is essential to the safe and efficacious practice of radiotherapy and radiosurgery. This chapter will discuss these principles in the context of what is commonly referred to as stereotactic radiosurgery, or single fraction radiotherapy. The radiobiology of single fraction treatment is firmly rooted in the radiobiology of fractionated radiotherapy.
Types of Ionizing Radiation Gamma rays and X-rays. Gamma rays and X-rays are electromagnetic radiation with energies ranging from 100 to 2 billion electron volts (eV). X-rays are produced when electrons transition from a higher to lower energy level, usually in the outer shell of heavy atoms and are thus produced outside the nucleus. X-rays may be products of radioactive decay (electron capture) or may be produced from X-ray tubes or linear accelerators, which accelerate electrons onto a heavy metal target producing both a continuous spectrum of photon energies called bremsstrahlung and monoenergetic characteristic X-rays. Gamma rays are photons emitted by radioactive nuclei and have a much narrower range of energies than X-rays, 10 keV to 10 MeV. Once produced, gamma rays and X-rays are #
Springer-Verlag Berlin/Heidelberg 2009
indistinguishable. 60Co, commercially produced from 59Co, is the most common source of gamma rays used in radiotherapy and undergoes betadecay with a half-life of 5.27 years. It is the subsequent gamma ray emission that makes 60Co applicable to radiotherapy and, more commonly, to stereotactic radiosurgery. 60Co decay results in emission of two discreet gamma energies, 1.17 and 1.33 MeV, giving an effective average energy of 1.21 MeV. Protons. The use of protons in radiotherapy and radiosurgery is based on the physical properties of these particles and the related characteristics of dose deposition in irradiated tissues [1]. Dose deposition is characterized by the Bragg peak. Quantitatively, the entrance dose for particle beams is relatively low compared to photons. An unaltered proton beam will deposit more than 50% of its energy over a narrow 2- to 3-cm path at a depth in water that depends on the beam energy. The beam may be altered to spread the Bragg peak to conform to the thickness and depth of the volume to be treated. However, the entrance dose is significantly increased in this case (> Figure 56-1). The biological effectiveness of X-rays, gamma rays and protons are roughly equivalent and each is considered to be low LET (linear energy transfer), sparsely ionizing radiation. Of note, protons have only a slightly higher radiobiological effectiveness (RBE) than 60Co and megavoltage X-rays. In practice this small difference is accounted for by calculating dose for protons in cobalt Gray equivalents (CGE), whether for single or multiple fractions. Proton radiotherapy and radiosurgery are fully discussed elsewhere in this volume.
854
56
Radiobiology of stereotactic radiosurgery
. Figure 56-1 Schematic representation of depth dose curves for a 160 MeV proton beam. Both unmodulated and spread-out Bragg Peak curves are shown. A 10 MVp X-ray curve is shown for comparison
Basic Principles of Radiobiology
Mammalian Cell Survival Curves
Direct Versus Indirect Effects of Radiation
Cell survival following single doses of ionizing radiation is a probability function of absorbed dose measured in the unit Gray (Gy)*. Typical mammalian cell survival curves obtained following single-dose irradiation in culture have a characteristic shape (> Figure 56-2): a low-dose ‘‘shoulder’’ region is followed by a more steeply sloped, or continuously bending, portion at higher doses. Measurement of radiation doseresponse of cultured human tumor cells has largely been limited to malignant tumor cell types. Studies on such cell lines have shown that the apparent radiosensitivity depends heavily on culture conditions and the assay used to assess cell survival [3–5]. The shoulder region is interpreted as accumulation of sublethal damage at low doses with lethality resulting
When cells are irradiated with low LET radiation, the majority of photon interactions are with water molecules, producing a fast electron and an ionized water molecule through Compton Scattering. These fast electrons interact with other water molecules through further ionizing events. The resulting positively charged water molecules dissociate into H+ ions and OH free hydroxyl radicals with an extremely short half-life (1010 s). Hydroxyl radicals are highly reactive and have sufficient energy to break chemical bonds in nearby (within 2 nm) molecules. This indirect effect of radiation, through the free radical intermediary, produces about 70% of radiation damage. The direct effect results from direct interaction of fast electrons with biologically important molecules (DNA) [2].
*1 J/kg
Radiobiology of stereotactic radiosurgery
. Figure 56-2 Curve for mammalian cell survival as a function of single dose of radiation (red line) given as a single fraction. The a/b is 10 Gy, a dose at which the contributions to cell killing by single events (aD, dashed blue line) and the interaction of sublethal events (bD2) are equal
from the interaction of two or more such sublethal events [2,6,7]. It may be considered that DNA is the target molecule for cell killing by ionizing radiation and that a double strand break in the DNA is necessary and sufficient to cause cell death (defined as loss of ability to divide). Double strand breaks may be effectively produced by a single particle track or by the interaction of two single strand breaks caused by separate particle tracks and occurring closely in space and time (> Figure 56-3). Single strand breaks alone may be repaired and therefore represent sublethal damage. Such a model is described by the linear-quadratic formula 2
SF ¼ eðaDþbDÞ
where SF is surviving fraction and D is dose of radiation in Gy [8], a is the coefficient related to single event cell killing and b the coefficient related to cell killing through the interaction of sublethal events. a/b is the ratio of the relative contributions of these two components to overall cell kill. a/b is the single dose at which overall cell killing is equally due to these two components (> Figure 56-2). aD ¼ bD2 or D ¼ a=b
56
. Figure 56-3 Schematic representation of double-strand break production by single events (aD) or interaction of events (bD2)
Most mammalian cell survival curves are well fit to the linear-quadratic model [3,4]. Cell survival following a single dose of radiation in vitro reflects the intrinsic radiosensitivity of a particular cell type to a particular type of radiation [3]. Cell types and tissues may vary in the a/b, resulting in slightly different shaped response curves (> Figure 56-4a). The a component varies little from tissue to tissue and variation in a/b is largely due to variation in the amount of b-type damage. A cell or tissue demonstrating a low a/b will have a relative abundance of b-type damage compared to those demonstrating a high a/b.
Radiobiology of Fractionated Radiotherapy A spectrum of fractionation schedules are used to treat intracranial disease, ranging from single fraction radiosurgery to fully fractionated courses of radiotherapy involving a series of 30 or more daily treatments. For fractionated radiotherapy each dose (fraction) produces similar biological effects, given sufficient interfraction interval (> Figure 56-4b). The linear quadratic formula for fractionated doses becomes 2 n SF ¼ eadbd
855
856
56
Radiobiology of stereotactic radiosurgery
where d is the dose per fraction and n is the total number of fractions. A basic principle of radiobiology and radiotherapy is that dose fractionation ‘‘spares’’ virtually all cell and tissue types. ‘‘Sparing’’ in this context means that, for a given total dose, there will always be less molecular damage and a lower
level of biological effect associated with multiple fractions compared to a single dose. As the number of fractions increases, the total dose (n x d) required to achieve a certain level of biological effect also increases (> Figure 56-5). The magnitude of the sparing effect of dose fractionation varies, however, and depends on a/b.
. Figure 56-4 Comparison of single-dose effect curves (a) and fractionated dose-effect curves (B) for low (blue) and high (red) a/b tissues. The small advantage seen in the low dose region sparing low a/b tissues (A) is amplified through dose fractionation (b)
. Figure 56-5 The effect of dose fractionation on the biological effectiveness of x radiation for low a/b (blue lines) versus high a/b (red lines) tissues. Isoeffect curves show the increase in total dose required to maintain biological effectiveness with increasing number of fractions
Radiobiology of stereotactic radiosurgery
56
The biologically effective dose (BED) is represented by d BED Gya=b ¼ nd 1þ a=b
context of single dose radiosurgery, a time factor is not likely to be important and has not, therefore, been included.
Where BED is expressed in Gya/b to indicate that it should be used only to compare effects in tissues with the same a/b, n is the number of fractions of dose d and nd is, therefore, the total dose (D). BED can be expressed as BED Gya=b ¼ D F
Biologically Effective Dose of Single Dose Radiosurgery
where F is a ‘‘fractionation factor’’ d F ¼ 1þ a=b F increases with increasing dose/fraction d but the effect is greatest for lower a/b and may be negligible for very high a/b, since as a/b increases F approaches 1. The linear-quadratic formulation is a means of estimating the effects of dose fractionation. Other factors, such as a rapid doubling time may be accounted for by additional terms [9]. In the
The biological effectiveness of single dose radiosurgery increases with dose much more precipitously than does the BED of fractionated radiotherapy. This is due to the unavoidable increase in both total dose and dose per fraction in the BED equation, which for radiosurgery becomes D2 BED Gya=b ¼ Dþ a=b So while a doubling in fractionated radiotherapy dose will exactly double the BED (assuming dose per fraction is not changed), doubling a radiosurgery dose will always more than double the BED by an additional increment dependant on a/b (> Figure 56-6)
. Figure 56-6 Biologically Effective Dose as a function of total dose for radiosurgery (solid line) and fractionated radiotherapy (2 Gy per fraction)
857
858
56
Radiobiology of stereotactic radiosurgery
D2 : a=b Various doses of radiosurgery may be compared to the dose of fractionated radiotherapy required to achieve an equivalent BED [10]. This comparison may be made only for tissues having the same a/b. A separation is again noticed between curves for tissues with different a/b (> Figure 56-7).
Tumor and Normal Tissue Radiobiology Radiobiology of CNS tumors. Tumors treated with radiosurgery range from highly malignant brain metastases to ‘‘benign’’ neoplasms, such as vestibular neuroma and meningioma and vascular malformations. Less differentiated, rapidly proliferating tumors, such as brain metastases are generally radiosensitive and radioresponsive. These are often referred to as ‘‘early responding’’ tissues and exhibit a dose response with a high a/b of about 10. More differentiated, slowly proliferating targets and normal CNS are considered to be ‘‘late responding’’ and exhibit a low a/b of 5 or less.
. Figure 56-7 Isoeffective doses for radiosurgery and fractionated radiotherapy (2 Gy per fraction) for low a/b tissues (blue line) and high a/b tissues (red line)
The probability of tumor control (TCP) is a function of the likelihood of inactivating all tumor cells in a given tumor following Poisson statistics. If it is assumed that every tumor cell must be killed to control a tumor, the probability of tumor control (TCP) is given by TPC ¼ eSFN where SF is the surviving fraction and N the total number of cells in the tumor. SF N is then the average number of viable cells remaining in a tumor following a certain treatment. TPC is the probability of no cells remaining viable under these conditions. TPC is a function of total dose and dose per fraction (BED), N (tumor bulk) and radiosensitivity of the tumor. This model leads to a sigmoid dose response curve for TCP (> Figure 56-8). The shape of this curve, indicating no probability of tumor control at low doses, a high probability at high doses and a steep rise in TPC through an intermediate range of doses is characteristic of all radiation dose response data. It should be appreciated that radiosensitivity is not equivalent to radioresponsiveness. Some CNS tumors are very radioresponsive but inevitably recur (e.g., CNS lymphoma), while others may show little or no radiographic evidence of response but are well controlled by modest radiation doses (e.g., meningioma, acoustic neuroma). Benign CNS tumors do not typically grow rapidly and do not shrink rapidly following radiotherapy or radiosurgery. Lack of growth should, therefore, be the goal of such therapy. Benign tumors of the CNS are slowly proliferating, relatively differentiated and have a delayed response to radiotherapy or radiosurgery. Estimates of a/b for meningioma and acoustic neuromas are 2.3–4 Gy [11]. AVMs are extremely slow growing and exhibit a delayed response to radiosurgery. The a/b for AVMs has been estimated to be 2.2 Gy [12]. Brain metastases have relatively rapid rates of growth, are typically derived from highly malignant,
Radiobiology of stereotactic radiosurgery
. Figure 56-8 Curves schematically comparing the probability of tumor control (TCP) with the probability of a normal tissue complication (NTCP). (a) The curves are positioned relatively close to one another. Normal tissue complications may be avoided only by minimizing the dose to the critical normal structure. Such a situation may occur when a normal structure such as optic nerve lies adjacent to a benign tumor being treated with single dose radiosurgery. (b) Dose fractionation separates the TCP and NTCP curves allowing for a higher probability of tumor control without significant risk of normal tissue complication. The ‘‘Uncomplicated Cure’’ curve is TCP-NTCP
undifferentiated primary cancers and respond quickly to radiosurgery, consistent with their having a high a/b of approximately 10 Gy.
Factors Affecting the Radiosensitivity of Tumors Physical factors. The relative biological effectiveness (RBE) of a particular beam of radiation depends upon the type of radiation emitted
56
(photons, protons, heavy particles, neutrons) and the energy of the emitted radiation. In general, the RBE is related to the linear energy transfer (LET) of the beam. LET is a measure of the amount of energy deposited per length along the particle track, usually expressed in keV/mm. Photons and protons are considered to be sparsely ionizing or low LET radiation with BED very nearly equal to 1 (reference radiation for RBE is 250 keV X-rays). Neutrons and heavy particles are considered densely ionizing and high LET radiation. In general LET increases with increasing mass or charge and decreases with increasing energy (velocity). LET varies from 0.2 keV/mM for Co60 to more than 1,000 keV/mM for high energy Fe ions (cosmic radiation). RBE reaches a peak at an LET of about 100 keV/ mM and can be as high as about 10 for some endpoints. Another physical factor affecting radiosensitivity is dose rate. Below a dose rate of about 1 Gy per minute cells can repair sublethal damage more quickly than it is produced. This leads to increased survival for a given dose compared to higher dose rates. The dose rate effect is most important between dose rates of about 1 cGy per minute and 100 cGy (1 Gy) per minute. Therapeutic radiotherapy is usually delivered at up to 10 Gy per minute. It has been suggested that overall treatment time for radiosurgery may also be important, especially for multi-isocenter methods, such as gamma knife, when treatments may be delivered over an hour or more. Biological factors. The major biological factor affecting radiosensitivity is position in the cell cycle. In general cells are most sensitive to radiation in the post replication phases of the cell cycle (G2 and mitosis). Cells are most resistant to radiation during replication (S phase). An intermediate radiosensitivity is found in cells that are in the pre-replicative phase (G1 or G0). The magnitude of the difference seen in radiosensitivity between S-phase cells and late G2/M phase cells is up to a magnitude of 3 in terms of a dose modifying factor [13].
859
860
56
Radiobiology of stereotactic radiosurgery
Chemical factors. Malignant tumors often are characterized by radiographic and pathologic evidence of necrosis. Necrosis develops as tumors cells grow away from the vascular supply faster than new blood vessels are formed. A gradient of nutrients and oxygen develops between the supplying blood vessel and the areas of necrosis where the lack of oxygen and nutrients reach critically low levels. It has clearly been demonstrated that cells irradiated in this ‘‘hypoxic’’ state are resistant to radiation. At oxygen levels below about 2.5% a threefold dose of radiation may be required to achieve the same biological effect as under ‘‘normoxic’’ conditions. The phenomenon is a manifestation of the ‘‘oxygen effect’’ resulting from the radiosensitizing properties of elemental oxygen. Radiobiologic hypoxia is thought to be important only in tumors and not in normal tissues. Fractionated radiotherapy can effectively overcome hypoxia to some extent since irradiated areas of tumors are know to undergo ‘‘reoxygenation’’ in the intervals between individual fractions [2]. There is evidence that hypoxia may affect the response of brain metastases to radiosurgery [14].
Normal Tissue Radiobiology Model Predicting Normal Tissue Complications Treating physicians must be concerned not only with effects of treatment on tumor, but also with normal tissue effects. The normal tissues of particular interest in the treatment of benign brain tumors are spinal cord and brain stem, optic apparatus and other cranial nerves and brain parenchyma. Also of interest are effects on the vasculature within both normal and tumor tissue. The probability of normal tissue complication (NTCP) following radiotherapy is, like tumor control probability, a function of dose and dose per fraction (BED), the tissue at risk (radiosensitivity) and the volume irradiated.
NTCP has been shown to be well represented by the model NTCP ¼ 1 expR where R is the variable related to dose and volume k R ¼ d=d0 with d0 determining the slope of the NTCP versus dose curve and k being a constant accounting for volume effects [15,16]. This is represented graphically as a sigmoid-shaped curve similar to that obtained for tumor cure (> Figure 56-8). Curves for a wide variety of normal tissue endpoints have been generated. Although each has a similar shape the relative placement of these curves along the dose axis may be quite different. In clinical radiotherapy, the relative positions of the curves for tumor cure and normal tissue complication defines what is known as the therapeutic ratio. The therapeutic ratio may be calculated as Probabality of tumor cure : Probabality of complication An ideal therapeutic ratio would be described by curves that allow 100% tumor cure without appreciable probability of normal tissue complication. The opposite extreme would be exemplified by a tumor requiring high dose radiation for cure located within a critical normal structure with a low tolerance to radiation. In practice a regimen that maximizes the probability of an uncomplicated cure is optimal (> Figure 56-8b). For the situation where a/b for tumor is higher than that for critical normal tissue, dose fractionation will always serve to separate the TCP and NTCP curves and increase the therapeutic ratio. Single dose radiosurgery relies on localizing dose in a way that minimizes dose to normal tissue. The tolerance dose for specific tissues is a function of the selected toxicity endpoint, volume irradiated, total dose, dose per fraction used and the level of acceptable risk [17–20]. For example, the total dose to cause cerebral necrosis
Radiobiology of stereotactic radiosurgery
in 5% of patients treated with a single dose of radiosurgery is vastly different than the dose associated with the same risk when conventional fractionation (1.8–2 Gy/day) is employed [10]. Tolerance doses may be expressed as D5/5, or the dose expected to produce complication in 5% of patients within 5 years of treatment [18]. This concept may be useful for effects such as necrosis or pituitary dysfunction, but is not useful for effects such as optic neuropathy or myelitis. Ideally, when dealing with benign tumor treatment, a dose regimen thought to be ‘‘safe’’ to the optic apparatus or spinal cord would seem the better choice. A dose and fractionation scheme effective in achieving tumor control and below the ‘‘threshold’’ for producing clinical neurologic dysfunction would be optimal [21,22]. The tolerance of normal structures in the CNS to single doses of radiosurgery and various fractionation regimens indicates that these structures are typical of late-responding tissues and demonstrate exquisite sparing through dose fractionation consistent with a low a/b. Estimates of a/b for spinal cord (myelopathy), brain parenchyma (cerebral necrosis) and optic nerve and chiasm (neuropathy) are 1–2 Gy [17,22,23].
Models to Compare BED of Different Fractionation Regimens It is important when investigating nonconventional fractionation regimens to have some basis for the choice of fraction size, total dose and interval between fractions. If a/b were well established for all tumors and normal tissues, the linear-quadratic model could provide such a basis. The formula for BED can be used to compare dose regimens of varying total doses and dose per fraction in a particular tissue. The equation may also be used to determine isoeffective total doses D associated with different doses per fraction d D1 =D2 ¼ ða=b þ d2 Þ=ða=b þ d1 Þ
56
Sheline et al. described a model for predicting the risk of brain necrosis as a function of total dose and number of fractions [23]. The model defined an isoeffect line for total dose as a function of fraction number. They defined the neuret, similar to BED, as Neuret ¼ D N0:41 XT0:03 ; where D is the total dose in cGy, N the number of fractions and T the overall time in days. This relationship demonstrated the strong dependence on N, a surrogate for fraction size, and the very weak dependence on overall time, T. These data may also be well fit to the linear –quadratic model using an a/b of 2.0 without a time factor. Although this model was not based on single dose data, it predicts a tolerance dose for brain necrosis of approximately 10 Gy in a single fraction. The literature would indicate that the formula derived by Sheline et al. could approximate isoeffect curves for other CNS effects such as optic neuropathy and spinal cord injury [21,24,25]. Common features of these models are an exponent of N similar to that found by Sheline et al. and an exponent of time (T) that is nearly 0. This emphasizes the importance of the number of fractions, or fraction size, in determining the tolerance dose of normal tissues in the CNS. For normal tissue, and probably most benign brain tumors, the overall time of treatment is relatively unimportant within the range normally encountered in a single radiation course, up to about 8 weeks at which time repopulation may begin in the normal CNS tissue [24]. Attempts to model isoeffective dose regimens for the risk of optic neuropathy following fractionated radiotherapy have lead to a similar model published by Goldsmith et al. [21] who defined the optic ret as Optic ret ¼ D N0:53 : They defined a threshold for optic neuropathy, defining a ‘‘safe’’ regimen as one resulting in no more than 890 optic ret. This corresponds to
861
862
56
Radiobiology of stereotactic radiosurgery
5,400 cGy in 30 fractions, 3,750 cGy in 15 fractions and 3,000 cGy in 10 fractions, all commonly used fractionation schedules. Although not based on single fraction data, the optic ret model predicts that a single radiosurgery dose of 8.9 Gy or less would be safe to the optic apparatus. This agrees very well with single fraction tolerance doses proposed in the literature, which range from 8 to 10 Gy [26,27]. The total dose predicted by the optic ret model to be safe may be calculated as DðGyÞ ¼
8:9Gy N0:53
emphasizing the well established importance of fraction size in determining optic nerve tolerance to radiotherapy [28,29]. Leber et al. also demonstrated a relationship between radiosurgery dose to optic nerve and the risk of developing optic neuropathy, data that conform to the expected sigmoid-shaped curve with a steep slope relative to dose (> Figure 56-9). The tolerance of cranial nerves III-VI appears to be substantially higher than for the optic nerves. Leber et al. studied 210 nerves among 50 patients who received single doses of up to 30 Gy with no patient developing neuropathy [26].
. Figure 56-9 NTCP curve constructed for optic neuropathy following radiosurgery. The data points are from [26]
Volume Effects in CNS Normal Tissue Tolerance In animal models there is a clear effect of volume on spinal cord tolerance to radiation [17,30]. This effect is closely related to length of cord irradiated rather than volume. Experiments done on large animal models, including primates, have shown that the effect of increasing the irradiated volume of spinal cord or brainstem is to lower the threshold and increase the slope of the sigmoidshaped dose response curve [17]. The volume effect is much more important in the high dose region than in the low dose region. For example, increasing the volume receiving dose associated with a very low risk of myelopathy or necrosis will not affect the risk very much. However, in the high dose region, volume reduction may substantially lower the risk of toxicity. The dose-response work of Sheline et al. was based largely on patients treated with whole brain radiotherapy [23]. Recent work has clearly shown volume of brain irradiated to be a factor in development of post radiation toxicity. A study undertaken by the Radiation Therapy and Oncology Group (RTOG) examined the maximum tolerated dose (MTD) of single fraction radiosurgery as a function of irradiated volume [31]. Volumes ranging up to 34 cc (40 mm diameter) were included and all tumors were recurrent following previous radiotherapy. Single dose MTDs were established for volumes 4.2–14 cc and those 14.1–34 cc as 18 Gy and 15 Gy to the target margin, respectively. The MTD for tumors smaller than 4.2 cc was not reached at 24 Gy. 15, 18 and 24 Gy are well in excess of the tolerance dose predicted by the neuret model. The corresponding BEDs are 127.5, 180 and 312 Gy2, respectively. Equivalent doses given in 2 Gy fractions would be 64, 90 and 156 Gy, respectively. Volume effects for optic neuropathy are not well established. It has been proposed that small volumes of optic nerve or chiasm may tolerate
Radiobiology of stereotactic radiosurgery
higher doses of radiation [32,33], but clear dosevolume guidelines do not exist.
Long-term Recovery of Radiation Damage in the CNS Re-irradiation of critical CNS structures presents an all too common dilemma in radiation oncology. Most commonly of concern is the tolerance of the optic pathway or spinal cord to re-irradiation. Long-term repair of radiation damage in the spinal cord has been clearly demonstrated in animal models. Hornsey and colleagues found longterm repair of rat spinal cord re-irradiated 100 days following an initial dose of radiation [34]. The degree of residual damage present at re-irradiation was dependent on the magnitude of the first dose. Wong and Hao reported similar results, also in rat spinal cord, finding up to 50% recovery following sufficient time, about one year for the maximum effect [35]. Ang and colleagues have reported on similar recovery of occult radiationinduced spinal cord injury in rhesus monkeys [17]. Significant recovery occurred at 1, 2 and 3 years following fractionated radiation to the cervical and thoracic spine. Based on a 5% incidence of myelopathy, recovery was quantified as 76, 85 and 101% at the 1, 2 and 3 year intervals, respectively. Histologic analysis revealed a mixture of white matter necrosis and vascular injury in the symptomatic animals, whereas histologically normal spinal cords were found in asymptomatic animals. Nieder et al. reported on clinical experience with spinal cord reirradiation [36]. They found a very low risk of reirradiation in a group of patients with a cumulative BED of 135 Gy2, an interval of at least 6 months between courses and neither course exceeding a dose equivalent to 98 Gy2. There are no good animal models for radiation optic neuropathy. Clinical data support long-term repair of radiation damage in the visual pathway, however. Schoenthaler et al. [37]
56
reported on 15 patients who received a second course of radiotherapy for recurrent pituitary tumors 2–17 years following initial treatment with radiotherapy. With follow-up of 1–30 years after the second course of treatment, no patient experienced optic neuropathy. Total doses ranged from 5,865 cGy to 10,400 cGy (median 7935 cGy). Flickinger et al. [38] reported on ten patients retreated for suprasellar tumors 1–17 years following initial treatment. Total doses ranged from 7,600 cGy to 9,865 cGy (median 8,500 cGy). One patient developed optic neuropathy 1.5 years following the second radiation course. Using these data it was estimated that 40% of the initial radiation damage remained as residual and recommended that this value be used cautiously in retreating tumors near the optic nerves and chiasm. Overall there is compelling evidence that longterm repair of radiation damage occurs in the CNS, in particular the spinal cord and optic apparatus. This phenomenon is likely at least partly due to repopulation of normal cells from surviving stem cell populations or migration of cells from unirradiated tissue [24]. However, extrapolation from the preclinical and retrospective clinical data available to clinical practice should be taken with caution. The potential benefit of reirradiation, treatment alternatives, and the risk of serious permanent sequelae need to be considered.
Radiobiological Considerations in Treatment Planning for Radiosurgery The best approach to avoiding radiation-induced CNS toxicity is always to minimize dose and volume of irradiated normal tissue. Careful treatment planning based on modern CT and/or MR imaging, effective patient positioning and immobilization and accurate treatment delivery all contribute to the ability to minimize the treated volume and to accurately measure doses to
863
864
56
Radiobiology of stereotactic radiosurgery
normal structures. Dose-volume histogram analysis is an essential element in the optimization of treatment planning. Inverse treatment planning and intensity modulated radiotherapy have contributed to reduction in the volume of normal tissue receiving high dose as well as reduction in the dose per fraction and total dose. However, in many situations, when normal structures, especially optic chiasm, brain stem or spinal cord, lie adjacent to the tumor to be irradiated, dose fractionation is the better and perhaps only method to allow safe and efficacious treatment. The high BED associated with radiosurgery at once represents the key to the efficacy of
. Table 56-1 Classification of radiosurgery targets [10] Category
Target
Embedded/ surrounded
Example
I II III IV
Late Late Early Early
Embedded Surrounded Embedded Surrounded
AVM Meningioma Glioma Metastasis
. Figure 56-10 Physical dose (blue) and biologically effective dose (red) as a function of distance from the 50% isodose line for a gamma knife radiosurgery plan. Negative distance is toward target center and positive values are outward from target surface. a/b of 2 Gy is assumed
radiosurgery and the potential for complications associated with its use. Larson et al. described four categories of radiosurgery targets depending upon whether the target is early (high a/b) or late (low a/b) responding and whether it is embedded in or simply surrounded by normal tissue (assumed to always be late responding tissue, > Table 56-1) [10]. The steep dose gradients associated with radiosurgery treatment plans represent an even steeper gradient in BED (> Figure 56-10). This emphasizes the importance of accurate delineation of target and normal structures, precise delivery of treatment as planned and knowledge of the potential error involved in treatment delivery, as even a small error on the order of a millimeter or less can deliver a subtherapeutic or toxic dose.
Conclusion Stereotactic radiosurgery is a powerful tool in the treatment of intracranial tumors. The biological effectiveness of a single dose of radiation provides excellent outcomes in terms of tumor control, when tumor size and location are appropriate. When normal structures such as optic nerve or brain stem are near the target structure, the ability to deliver a therapeutic dose to the target may be hampered by risk to normal tissue. When large volumes are considered for radiosurgery, account must be taken of the large volume of brain parenchyma at risk for necrosis. Understanding of basic radiobiological principles and the accepted tolerance doses for normal tissues are essential for safe and efficacious use of stereotactic radiosurgery.
References 1. Munzenrider JE, Crowell C. Charged Particles. In: Mauch PM, Loeffler JS, editors. Radiation oncology: biology and technology. Philadelphia: W.B. Saunders; 1994. p. 34-55.
Radiobiology of stereotactic radiosurgery
2. Hall EJ, Giaccia AJ. Radiobiology for the radiologist. 6th ed. Philadelphia: Lippincott Williams and Wilkins; 2006. 3. Fertil B, Malaise EP. Intrinsic radiosensitivity of human cell lines correlated with radioresponsiveness of human tumors: analysis of 101 published survival curves. Int J Radiat Oncol Biol Phys 1985;11(9):1699-707. 4. Taghian A, Suit H, Pardo F, et al. In vitro intrinsic radiation sensitivity of glioblastoma multiforme. Int J Radiat Oncol Biol Phys 1992;23:55-62. 5. Weichselbaum RR, Nove J, Little JB. X-ray sensitivity of human tumor cells in vitro. Int J Radiat Oncol Biol Phys 1980;6:437-40. 6. Elkind MM, Sutton H. X-ray damage and recovery in mammalian cells in culture. Nature 1959;184:1293-5. 7. Withers H. Biologic basis for altered fractionation schemes. Cancer 1985;55(9):2086-95. 8. McBride WH, Withers HR. Biological basis of radiation therapy. In: Perez CA, Brady LW, Halperin EC, SchmidtUllrich RK, editors. Principles and practice of radiation oncology. 4th ed. Philadelphia: Lippincott, Williams and Wilkins. 2004. p. 96-136. 9. Travis EL, Tucker SL. Isoeffect models and fractionated radiation therapy. Int J Radiat Oncol Biol Phys 1987;13:283-7. 10. Larson DA, Flickinger JC, Loeffler JS. The radiobiology of radiosurgery. Int J Radiat Oncol Biol Phys 1993; 25(3):557-61. 11. Shrieve DC. Basic principles of radiobiology applied to radiotherapy of benign intracranial tumors. Neurosurg Clin N Am 2006;17:67-78. 12. Qi XS, Schultz CJ, Li XA. Possible fractionated regimens for image-guided intensity-modulated radiation therapy of large arteriovenous malformations. Phys Med Biol 2007;52(18):5667-82. 13. Sinclair WK. Cyclic X-ray responses in mammalian cells in vitro. Radiat Res 1968;33:620-43. 14. Goodman KA, Sneed PK, McDermott MW, et al. Relationship between pattern of enhancement and local control of brain metastases after radiosurgery. Int J Radiat Oncol Biol Phys 2001;50(1):139-46. 15. Alber N, Nusslin F. A representation of an NTCP function for local complication mechanisms. Phys Med Biol 2001;46:439-47. 16. Lyman JT, Wolbarst AB. Optimization of radiation therapy. III. A method of assessing complication probabilities from dose-volume histograms. Int J Radiat Oncol Biol Phys 1987;13:103-9. 17. Ang KK. Radiation injury to the central nervous system: clinical features and prevention. In: Meyer JL, editor. Radiation injury: advances in management and prevention, vol 32. Basil: Karger; 1999. p. 145-54. 18. Emami B, Lyman J, Brown A, et al. Tolerance of normal tissue to therapeutic irradiation. Int J Radiat Oncol Biol Phys 1991;21(1):109-22. 19. Marks JE, Baglan RJ, Prassad SC, Blank WF. Cerebral radionecrosis: incidence and risk in relation to dose, time,
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
56
fractionation and volume. Int J Radiat Oncol Biol Phys 1981;7:243-52. Marks LB, Spencer DP. The influence of volume on the tolerance of the brain to radiosurgery. J Neurosurg 1991; 75:177-80. Goldsmith BJ, Rosenthal SA, Wara WM, Larson DA. Optic neuropathy after irradiation of meningioma. Radiology 1992;185:71-6. Shrieve DC, Hazard L, Boucher K, Jensen RL. Dose fractionation in stereotactic radiotherapy for parasellar meningiomas: radiobiological considerations of efficacy and optic nerve tolerance. J Neurosurg 2004; 101:390-5. Sheline GE, Wara WM, Smith V. Therapeutic irradiation and brain injury. Int J Radiat Oncol Biol Phys 1980;6(9):1215-28. van der Kogel AJ. Central nervous system radiation injury in small animals. In: Gutin PH, Leibel SA, Sheline GE, editors. Radiation injury in the nervous system. New York: Raven Press; 1991. p. 91-111. Wara WM, Phillips TL, Sheline GE, Schwade JG. Radiation tolerance of the spinal cord. Cancer 1975; 35:1558-62. Leber KA, Bergloff J, Pendl G. Dose-response tolerance of the visual pathways and cranial nerves of the cavernous sinus to stereotactic radiosurgery. J Neurosurg 1998;88: 43-50. Tishler RB, Loeffler JS, Lunsford LD, et al. Tolerance of cranial nerves of the cavernous sinus to radiosurgery. Int J Radiat Oncol Biol Phys 1993;27(2):215-21. Harris JR, Levene MB. Visual complications following irradiation for pituitary adenomas and craniopharyngiomas. Ther Radiol 1976;120:167-71. Parsons JT, Bova FJ, Fitzgerald CR, Mendenhall WM, Million RR. Radiation optic neuropathy after megavoltage external-beam irradiation: analysis of time-dose factors. Int J Radiat Oncol Biol Phys 1994; 30(4):755-63. Hopewell JW, Morris JH, Dixon-Brown A. The influence of field size on the late tolerance of the rat spinal cord to single doses of X-rays. Br J Radiol 1987;60:1099-108. Shaw E, Scott C, Souhami L, et al. Single dose radiosurgical treatment of recurrent previously irradiated primary brain tumors and brain metastases: final report of RTOG protocol 90–05. Int J Radiat Oncol Biol Phys 2000;47:291-8. Habrand JL, Austin-Seymour M, Birnbaum S, et al. Neurovisual outcome following proton radiation therapy. Int J Radiat Oncol Biol Phys 1989;16:1601-6. Martel MK, Sandler HM, Cornblath WT, et al. Dosevolume complication analysis for visual pathway structures of patients with advanced paranasal sinus tumors. Int J Radiat Oncol Biol Phys 1997;38:273-84. Hornsey S, Myers R, Warren P. Residual injury in the spinal cord after treatment with X-rays or neutrons. Br J Radiol 1982;55:516-9.
865
866
56
Radiobiology of stereotactic radiosurgery
35. Wong CS, Hao Y. Long-term recovery kinetics of radiation damage in rat spinal cord. Int J Radiat Oncol Biol Phys 1997;37:171-9. 36. Nieder C, Grosu AL, Andratschke NH, Molls M. Proposal of human spinal cord reirradiation dose based on collection of data from 40 patients. Int J Radiat Oncol Biol Phys 2005;61:851-5.
37. Schoenthaler R, Albright NW, Wara WM, Phillips TL, Wilson CB, Larson DA. Reirradiation of pituitary adenoma. Int J Radiat Oncol Biol Phys 1992; 24(2):307-14. 38. Flickinger JC, Deutsch M, Lunsford LD. Repeat megavoltage irradiation of pituitary and suprasellar tumors. Int J Radiat Oncol Biol Phys 1989;17(1):171-5.
64 Radiosensitizers in Neurooncology D. Khuntia . A. Chakravarti . H. I. Robins . K. Palanichamy . M. P. Mehta
Introduction Both brain metastasis and malignant gliomas represent a significant problem in neurooncology with very poor outcomes with current therapies. Brain metastasis, which affects around 200,000 patients a year portends a poor prognosis with median survivals ranging between two and seven months [1,2]. Malignant gliomas are comprised of World Health Organization (WHO) Grade III and IV gliomas. Median survival times for patients with Grade IV tumors, also referred to as glioblastoma (GBM), is noted to be especially poor, remaining just over one year with current therapeutic regimens [3]. The treatment regimen for malignant glioma patients has traditionally involved maximal surgical debulking, followed by radiation +/ chemotherapy. Cooperative group trials over the past several decades have demonstrated that adjuvant radiation significantly prolongs survival compared to surgery alone in malignant glioma patients [4]. The role of cytotoxic chemotherapy with radiation for malignant gliomas and brain metastasis has been less welldefined until more recently. More contemporary cytotoxic chemotherapeutic agents such as temozolomide (TMZ) have demonstrated activity in malignant gliomas, both when used as single agents as well as in combination with radiotherapy. Further, targeted therapies have been developed which inhibit specific molecular pathways required for tumor-specific survival, proliferation, migration, and angiogenesis. The ensuing discussion will focus on a description of ‘‘classic’’ radiosensitizers as well as novel radiosensitizers currently being evaluated. Further, we will describe #
Springer-Verlag Berlin/Heidelberg 2009
the possible biological mechanisms of treatment resistance in malignant gliomas and the roles of chemotherapeutic and biotherapeutic modifiers of radiation response in patients with malignant gliomas.
Traditional Radiation Sensitizers Radiosensitizers concurrently enhance the effect of ionizing radiation on tumor cells while sparing the effects of radiotherapy on normal, healthy cells. Traditional radiation sensitizers are generally classified into the following three categories: hypoxic cell sensitizers, hypoxic cytotoxins, and nonhypoxic cell sensitizers.
Hypoxic Sensitizers and Hypoxic Cytotoxins Hypoxic cell sensitizers are able to increase the radiosensitivity of tumor cells deficient of oxygen without adversely affecting normal cells. Their mechanism of action centers around the ability to induce the formation and stabilization of toxic DNA radicals, mimicking the effects of oxygen [5]. Tumor cells are hypoxic in relation to the surrounding normal tissue secondary to obstruction of blood flow, defective or inadequate angiogenesis, or because of outstripping of capillary blood supply. These include the drugs nitroimadazole, misonidazole, etanidazole, nimorazole, and efaproxaril. A small randomized trial by Urtasun and colleagues initially reported a survival advantage with the use of metronidazole; however,
988
64
Radiosensitizers in neurooncology
. Table 64-1 Study
Therapy
N
Median survival (wks)
Urtasan [6]
RT (3000 rad/9F/3 wk) RT and MNZ RT (5800 rads/30F/6 wk) RT (3900 rads/9F/3 wk) and MISO RT (3900 cGy/9F/3 wk) and MNZ RT (5656 cGy/5.5 wks) RT (4352 cGy/12 F/4 wks) RT and MISO RT (4950 cGy/15F) RT (4950 cGy) and MISO RT (6000 cGy/7 wks) and MISO RT (6650 cGy/31F) RT (6650 cGy/31F) and MISO RT (4500 cGy/20F) and placebo RT (4500 cGy/20F) and MISO
15 16 17 15 17 20 18 17 81 82 54 27 18 195 188
15 26 25a 24a 25a 36 31 39 46.1 44.5 33+ 43 60 36 33
Urtasan [7] (2nd study)
Bleehen [8] (Cambridge)
EORTC [9] and placebo RTOG 78–01 [10,11] Vienna Study Group [11] Medical Research Council [12] a
Number of patients still alive at time data was reported Source: Adapted from [13]
subsequent studies failed to show an advantage (> Table 64-1) [6,13]. Hyperbaric oxygen (HBO) also has been used as a hypoxic cell sensitizer. Initial clinical trials included the use of hyperbaric oxygen to potentially reduce the hypoxia present in radioresistant GBM cells [14]. There is some improvement in local control for certain tumors, such as cervix and head and neck cancers, but the results are not accepted as conclusive. A non-randomized pilot study evaluated survival in 38 glioma patients treated with HBO + RT [15]. Overall survival was not significantly different versus historical controls (median survival of 46 weeks). Given this disappointing finding along with the cumbersome and expensive nature of the treatment, further development of HBO as a sensitizer was largely abandoned. Efaproxiral (also known as RSR-13), is a synthetic allosteric modifier of hemoglobin that functions by noncovalently binding to the hemoglobin tetramer and decreasing the hemoglobinoxygen binding capacity, allowing more oxygen to be available to the tissues [16]. This hypoxic
sensitizers is novel in that the radiation-enhancing effect does not rely on the direct diffusion of the drug into the tumor cells. Initial results in both GBM and brain metastasis were promising [17–19]. A phase II study of RSR-13 with radiation therapy for GBM reported a median survival of 12.3 months with 1-year and 18-month survivals of 54 and 24%, respectively [19]. In a phase III study looking at WBRT with or without RSR13, patients with metastatic breast cancer were found to have improved survival [17]. As a result, a large phase III study including only patients with metastatic breast cancer to the brain was conducted randomizing patients to WBRT with or without RSR-13. Results have shown that the addition of RSR-13 improves median survival (4.47 vs. 9 months, p = 0.001), quality of life (p = 0.019), and quality adjusted survival (p = 0.001) [20]. As a result of this trial, a confirmatory open-label, phase III study, Enhancing Whole-brain Radiation Therapy in Patients With Breast Cancer and Hypoxic Brain Metastases (ENRICH) has recently been completed. This follow up study was negative and has called into the question the value of this drug.
Radiosensitizers in neurooncology
Hypoxic cytotoxins act as radiosensitizers by selectively killing hypoxic cells, which are generally more resistant to radiation. Hypoxic cytotoxins fall into three classes: quinone antibiotics such as mitomycin C, nitroaromatic compounds, and benzotriazine di-N-oxides such as tirapazamine. As with many of the other radiation sensitizers, these bioreductive agents have not been widely used because they are inconvenient and have minimal, if any, clinical benefit [15]. Tirapazamine, arguably the most attractive agents of the hypoxic cytotoxins, is a bioreductive agent with enhanced toxicity of hypoxic tumor cells. The RTOG conducted a phase II trial of two dose levels of tirapazamine (RTOG 94-17) in patient with GBM’s and compared their survivals with RPA matched patients within the RTOG database [21]. Unfortunately, there was no improvement in survival with the use of tirapazamine.
Non-hypoxic Sensitizers Promising non-hypoxic sensitizers include the halogenated pyrimidines such as 5-iododeoxyuridine (IUdR) and 6-bromodeoxyuridine (BUdR). These agents function as DNA base analogues, becoming incorporated in newly synthesized DNA, and rendering tumors more sensitive to damage from ionizing radiation. Halogenated pyrimidines are preferentially incorporated into tumor cells, presumably based on the higher proliferation index of tumors cells compared to normal tissue. Bromodeoxyuridine (BUdR) is one of the earliest halogenated pyrimidines to be studied. Because of concerns about rapid catabolism of halopyrimidines after intravenous infusion, early clinical trials focused on inter-arterial infusion of BUDR. In a nonrandomized Japanese study of intra-arterial BUdR, survival outcomes were encouraging, but at the expense of significant toxicity from carotid artery catheterization [22,23]. Subsequently, it was determined that prolonged
64
intravenous infusion can achieve radiosensitization equivalent to intra-arterial administration [23], with adequate steady-state plasma concentration and acceptable toxicities [24,25]. Other studies of have evaluated the use of continuous infusions of BUdR or iododeoxyuridine (IUdR) concurrently with radiation. Initial studies included a combined series of four phase I studies at the National Cancer Institute reported a median survival of 13 months for GBM patients treated with infusional BUdR or IUDR [26]. Other phase I and II studies evaluating continuous infusion IUdR with hyperfractionated RT reported median survivals of 11–15 months [25,27,28]. In another series, the NCOG (Northern California Oncology Group) treated newly diagnosed GBM patients with continuous infusion BUdR during RT and observed a median survival of 56 weeks. This was followed by adjuvant chemotherapy with PCV (procarbazine, CCNU, and vincristine). A subgroup of patients receiving higher cumulative doses of BUdR, showed improvement in progression-free survival [29,30]. The results of the NCOG study led to a subsequent single-institution trial investigating significantly higher doses of BUdR with hyperfractionated RT. However, median survival was not improved at 50 weeks, and significant toxicities with the escalated dose of BUdR were observed [31]. The RTOG also conducted a phase III study looking at radiation plus procarbazine, lomustine, and vincristine with or without BUdR in the management of anaplastic astrocytomas [32]. In this 286 patient randomized trial, BUdR was shown to have no survival advantage in this patient population, with a 4-year survival rate of 51% in both arms.
Contemporary Radiosensitizers One of the more promising radiation sensitizers under active clinical investigation is motexafin gadolinium (MGd). MGd is in a class of
989
990
64
Radiosensitizers in neurooncology
drugs referred to as redox modulators. It is a metallotexaphyrin that catalyzes the oxidation of intracellular-reducing metabolites and generates reactive oxygen species that selectively concentrate in tumor cells to promote apoptosis. Because it contains the paramagnetic compound gadolinium, this drug allows excellent visualization of the tumor on MRI [33,34]. Original studies with the drug were performed in patients with brain metastasis and have found the drug safe and well tolerated [35–37]. Further, the drug has the ability to pick up additional metastasis as MGd accumulates over time after all of the dose is delivered (11.4% increase after ten doses). A subsequent phase III study (SMART trial) compared WBRT with or without MGd. The primary endpoints of the study were survival and time to neurologic progression. Secondary endpoints included time to loss of functional independence, radiologic response rate, time to radiologic progression, tome to progression of brain specific quality of life (using the Functional Assessment of Cancer Therapy-Brain [FACT-Br]). Unfortunately, the study did not show a benefit for the primary endpoints of the study, however there was an improvement for time to neurologic progression and neurocognitive function for patients with nonsmall cell lung cancer [38]. This was a pioneering study in that it was able to do very thorough neurocognitive testing on a large scale. It was found that over 90% of patients had impairment of at least one neurocognitive domain at presentation. This was highly correlated with baseline volume of disease [39]. As a result of the SMART trial, a follow-up international phase III study was conducted randomizing 554 patients with brain metastasis from NSCLC to WBRT with or without MGd [40]. The primary endpoint of this study was time to neurologic progression (TNP). TNP improved with MGd from 10 to 15.4 months, however this was not statistically significant (p = 0.12). Fewer patients in the MGd arm required salvage
brain surgery or radiosurgery, as well. It has been shown that the reason there was not a statistically significant benefit to neurologic progression was secondary to the fact that patients treated outside of North America, did not start WBRT soon after diagnosis (2.2 weeks vs. 6.5 weeks for Europe and Australia). When the entire cohort was reanalyzed with patients receiving WBRT within 3 weeks of radiotherapy, TNP was significantly prolonged with the use of MGd (p = 0.006). MGd has also been studied in malignant gliomas. Initial studies have shown the drug to be safe and tolerable with the maximum tolerated dose of 22 doses of 5 mg/kg [41]. Subsequently, a phase II study evaluated the safety and efficacy of MGd with 60 Gy radiation in 25 patients [42]. With 6-month follow-up, the Kaplan-Meier estimate of survival was 80% and median survival has not yet been reached. With these promising results, other studies are underway within the RTOG evaluating MGd with temozolomide in patients with GBM. Currently, the drug is being investigated in head and neck cancer, pediatric pontine gliomas, pancreatic carcinoma, and lung cancer [33].
Chemotherapeutic Agents as Radiosensitizers The use of systemic chemotherapy in conjunction with ionizing irradiation is now evolving as a therapeutic modality for both brain metastases and primary brain neoplasms. Common mechanisms for radiosensitization include: inhibition of DNA repair, (e.g., by pyrimidine or purine analogs); perturbation of cell cycling, (e.g., by paclitaxel) to optimize the fraction of G2/M-phase cells; specific effects on hypoxic cells, (e.g., mitomycin). There are multiple complicating factors which require investigative explication for such a multi-modality approach to be successful. Most obvious is the therapeutic index, because there is the potential for normal central nervous system (CNS) injury,
Radiosensitizers in neurooncology
and/or radiation induced necrosis secondary to drug induced radiosensitization. Systemic chemotherapeutic toxicities present another area for potential morbidity. Additionally, the blood-brain barrier (BBB) may represent a significant obstacle to drug delivery. Two recent reports concerning the BBB highlight the significance of drug exclusion from the central nervous system by glycoprotein efflux pumps [43,44] beyond the traditional consideration of lipid solubility. It is of interest to note that such efflux pump drug extrusion may be less efficient in brain metastases than primary gliomas [43]. This point is dramatically illustrated by Gerstner and Fine in the case of paclitaxel [43]. It is notable that the BBB may fluctuate during therapy. Thus, a drug which may initially penetrate a neoplastic focus that has disrupted the BBB, may become less permeable as successful anti-neoplastic therapy allows reconstitution of the BBB [45]. This is illustrated in a report of the use of chemotherapy in newly diagnosed small cell lung cancer (SCLC) in which the relative response rate for CNS and systemic disease was 27 and 73%, respectively [46]. It is predictable to have similar results among other chemotherapy responsive tumors, (e.g., germ cell tumors, lymphoma, medulloblastoma/primitive neuro-ectodermal tumor) with chemotherapy that does not readily cross the BBB, i.e., an initial response, but diminished drug penetration after BBB reconstitution. This concept is particularly relevant to any clinical scenario involving radiotherapy, as a radiation therapeutic effect is highly likely for most neoplasms. Taking the aforementioned considerations collectively, the ideal chemotherapeutic radiosensitizer should have good penetration across the BBB, a demonstrated preclinical radiosensitizer effect, inherent tumor activity, and at least a presumption of therapeutic index. Other caveats relate to the fact that BBB penetration is particularly relevant in any clinical scenario addressing CNS prophylaxis. In spite of the difficulty of accessing long term neurological toxicity,
64
it remains a critical clinical endpoint for future studies. The use of various radiosensitizers, and in particular chemo-therapeutic agents, require long term follow up in any patient population expected to have a survival benefit measured in years. The risk of a potential radiation induced dementia [47] aggravated by a drug effect is a significant concern. A classic example of a deleterious drug/radiation interaction is the experience with high dose methotrexate [48,49]. An excellent example of the type of prospective neurocognitive studies that are possible has been reported by Meyers et al. [50]. Emerging laboratory data suggest that one possible mechanism of relative resistance of brain metastases to chemotherapy might be the florid astrocytic response induced by tumor cells in the brain, and laboratory data suggest that tumor cell-astrocyte cell-to-cell communication dramatically increases chemoresistance.
Specific Agents Temozolomide (TMZ) TMZ is an oral alkylating agent. It is highly bioavailable, crosses the BBB, and achieves significant levels in the cerebrospinal fluid [51]. It dissociates to form the active alkylating agent methyltriazenoimidazole-arboxamide (MTIC) at physiologic pH, which methylates the O-6 position of guanylic acid in DNA. TMZ is taken orally and is absorbed rapidly and completely after oral administration. Compared to other alkylating agents, its doselimiting toxicity, i.e., myelosuppression, is modest [52]. The first speculation that TMZ might have a radiosensitization effect related to the results of a positive phase II study [53] in newly diagnosed glioblastoma multiforme (GBM) patients performed by Stupp and colleagues. This was followed by a confirmatory phase III study [3] (performed by the European Organization for Research and Treatment of Cancer/National Cancer Institute of
991
992
64
Radiosensitizers in neurooncology
Canada) showing a survival advantage for administering daily TMZ (75 mg/m2) during the course of radiotherapy, followed by maintenance TMZ [on a 5/28 day schedule (150–200 mg/m2)] starting one month post radiation. This approach has become standard of care for this patient population. Interestingly, a contemporary phase II study in which TMZ was only delivered with radiation (and not post-radiotherapy) at a somewhat lower dose of 50 mg/m2 per day of radiation (vs. 75 mg/m2 used in the Stupp regimen) had comparable results [54]. Parenthetically, preradiation TMZ tested in the context of another phase II study did not appear efficacious comparing results across studies [55]. The results of these clinical trials taken collectively (as well as results in the setting of brain metastases discussed below) resulted in hypothesis that TMZ was in part acting as a radiation sensitizer. The concept of TMZ radio-sensitization at the time of the aforementioned clinical trials was only weakly supported by a preclinical study which essentially observed an additive effect between these medications [56]. A convincing laboratory demonstration of radiosensitization, however, has recently reported by Chakravarti et al. [57]. Interestingly, these investigators found TMZ improves radiation response most effectively in MGMTnegative GBM cell lines by increasing radiationinduced double-stranded DNA damage. These results are also consistent with clinical results showing patients with a silent MGMT gene have a better clinical outcome [58]. The antitumor activity of TMZ has been attributed primarily to the methylation of DNA, which is highly dependent upon the formation of a reactive methydiazonium cation [59]. Nearly 70% of total DNA methylation by TMZ occurs at the N7-guanine, 9 and 5% of adducts are formed at the N3-adenine and O6-guanine, respectively. The cytotoxicity of TMZ is influenced by three DNA repair activities in particular. The first is O6-alkylguanine-DNA alkyltransferase (AGT). There is accumulating evidence that
the cytotoxicity of TMZ is highly dependent on the formation of O6-methylguanine, despite the fact that this lesion accounts for only a small percentage of the total DNA adducts formed. Adducts produced at the O6-position of guanine have been found to be especially mutagenic and cytotoxic. Methyl adducts at the O6-guanine in DNA are repaired by the cytoprotective DNA repair protein, MGMT, which transfers the methyl group to an internal cysteine acceptor residue. This reaction results in an irreversible inactivation of MGMT, requiring increased de novo protein synthesis to restore repair activity. Depletion of MGMT via pretreatment with substrate analogs such as O6-benzylguanine (O6-BG) has been investigated. It has been demonstrated in preclinical models that O6-BG can increase the cytotoxicity of TMZ by several fold. It has been found that continuous administration of O6-BG is more effective than intermittent dosing, indicating that regeneration of MGMT activity following MGMT inhibitor may be of clinical significance. The second mechanism of resistance to TMZ involves DNA mismatch repair pathways. One mechanism involves binding by a heterodimer complex consisting of hMSH2 and GTBP/p160 proteins and subsequent DNA incision following recruitment of an additional heterodimer consisting of hPMS2 and hMLH1 proteins. A section of DNA is then removed between the incision and the mismatch, and replaced by resynthesis and ligation. When this pathway is targeted to the strand directly opposite O6-MG, its unsuccessful attempt to find a complementary base results in continued excision/insertion which produces persistent damage to the DNA. The resulting interruptions in the daughter strands prevent replication in the subsequent S-phase and may account for two cell divisions being required before the emergence of TMZ toxicity. Since the cytotoxicity of TMZ is dependent upon a functional DNA mismatch repair pathway, resistance may be conferred by a mutation in any of the genes encoding for a
Radiosensitizers in neurooncology
protein involved in mismatch recognition/incision (e.g., germline mutations in hereditary non-polyposis colorectal cancer). Such abnormalities result in a TMZ ‘‘tolerant’’ phenotype which is unaffected by MGMT activity. The third mechanism of TMZ resistance involves base excision repair and poly(ADP-ribose) polymerase. Methyl adducts produced at N7-guanine and N3-adenine by TMZ may also hinder DNA replication, as enzymatic or spontaneous depurination will ultimately result in DNA strand breakage. Preclinical data also suggests that TMZ has at least additive activity with radiation in human glioblastoma cells [60]. It was determined that TMZ and radiation had at least an additive effect. In U373MG GBM cells, it was determined that the addition of 10 mM of TMZ to 1–2 Gy of radiation increased cell kill by 2.5–3.0-fold. However, in a cell line with 100-fold greater MGMT activity, there was actually an antagonistic effect observed. This antagonistic effect was mitigated by co-incubation with O6-BG, revealing a strategy to enhance the additive effect of TMZ and radiation. The combination of TMZ/radiation has also been studied in the context of brain metastases. In a small randomized phase II study Antonadou and associates treated patients with TMZ (75 mg/m2/day) concurrent with 40 Gy of fractionated conventional radiotherapy (2 Gy, 5day/week) for 4 weeks versus radiotherapy alone [61]. Results demonstrated a statistically significant increase in response with TMZ plus whole brain radiotherapy (WBRT) (96%) versus WBRT alone (67%) with a trend toward improved survival. A confirmatory phase III study has also been performed [62] in a series of 108 randomized patients (82% with NSCLC) the difference in response rates for brain metastasis was statistically significant favoring WBRT plus TMZ (response rate, 53%) versus WBRT alone (response rate, 33%), with a trend toward increased survival (8.3 vs. 6.3 months). It is noteworthy that several studies (reviewed by Chang et al. [63]) have
64
demonstrated modest response to TMZ for brain metastases following post-radiotherapy relapse. In part, based on data regarding TMZ summarized above, the Radiation Therapy Oncology Group (RTOG) is now conducting a 3 arm phase III clinical trial in non-small cell lung cancer (NSCLC) patients with one to three brain metastases. This clinical trial compares WBRT and stereotactic-radiosurgery alone versus with radiosurgery/WBRT with TMZ or erlotinib. Details regarding this can be obtained online at the web-site: clinicaltrials.gov.
Camptothecins Topotecan is a topoisomerase 1 inhibitor with demonstrated activity in both NSCLC [64] and SCLC [65]. It has known radiosensitization effects [66,67] and high brain capillary permeability [68,69]. Topotecan has a response rate of 33–63%, as mono-therapy in patients with brain metastases from SCLC [70]. Several phase I/II clinical trials of topotecan and radiation therapy for brain metastases have been performed [71–73] with response rates ranging from 46 to 72%. Toxicities were variable depending on topotecan dosing with grade 3/4 hematological toxicity being predominant. To date a definitive randomized study has not been reported with this agent as an adjunct to radiation in the setting of CNS metastases. It is of interest to note that topotecan has been reported to have some activity in recurrent gliomas [74,75]. It was tested as a radiosensitizer in newly diagnosed GBM by the RTOG. The median survival was 9.3 months, not significantly different from historical controls evaluated with recursive partitioning analysis using the RTOG database [76]. Another camptothecan, irinotecan, may prove more effective for high grade glioma [77]. Recent data regarding the use of irinotecan with bevacizumab has been promising in recurrent
993
994
64
Radiosensitizers in neurooncology
disease [78]. (It should be noted that irinotecan is administered intravenously (IV); topotecan has the advantage of also having an oral formulation, which may be advantageous for daily dosing as a radiosensitizer).
Taxanes Preclinical studies have shown activity for paclitaxel as a radiosensitizing agent in malignant cell lines [79,80]. The RTOG performed a phase II study evaluating the efficacy and feasibility of conventional radiotherapy and concurrent weekly paclitaxel in newly-diagnosed GBM [81]. The results, (i.e., a median survival of 9.7 months) did not represent an improvement when compared to the historical RTOG database. However, the concurrent use of anticonvulsant therapy may have contributed to increased paclitaxel metabolism compromising the results of this study [82]. In a later study in recurrent GBM patients in which there was adjustment for the use of anticonvulsant therapy, paclitaxel was not found to have significant activity [83]. Paclitaxel has also been studied as a radiosensitizing agent in the setting of fractionated stereotactic radiotherapy for recurrent GBM [84]. The results of this study suggested tumor volume was a significant prognosticator. Overall median survival was 7 months; 1- and 2-year actuarial survival rates were 17 and 3.4%, respectively [84].
in the setting of CNS disease. They are, however, excellent candidates for future study.
Platinum Agents Preclinical studies have demonstrated that platinum agents can inhibit repair of radiation induced damage, and exert direct cytotoxic effects on a variety of neoplasms including glioma cell lines [89,90]. These observations were applied to a study of fractionated stereotactic radiotherapy with cis-platinum in patients with recurrent high grade glioma with the possible suggestion of benefit [91]. A phase III intergroup trial of standard radiotherapy with and without continuous infusion cis-platinum in newly diagnosed GBM patients, however, found no significant difference in survival between groups [92]. A randomized phase III study of carboplatin and WBRT versus WBRT was initiated in patients with NSCLC brain metastases [93]. This study closed prematurely do to poor patient accrual (n = 44). The observed median survival in the WBRT alone arm was 4.4 months, and 3.7 months in the combined arm (p = 0.64). There was no difference in response rate between arms.
Molecularly Targeted Agents as Sensitizers
Pyrimidines
Signal Transduction Pathways Involved in Treatment Resistance
It is of passing interest to note the fluropyrimidines including capecitabine, and 5-flurouracil (its parent drug) both have good penetration into the CNS [63]. They are often used for radiosensitization in a variety of tumor types for which there is a firm preclinical rationale [85–88]. We are not aware of definitive controlled clinical studies using either these drugs
It is known that certain genetic events underlie disregulation of key signaling pathways in gliomagenesis. In secondary GBM’s, which arise from lower-grade gliomas, initial events involve p53 loss or mutation combined with Rb and PTEN mutations as tumors progress in grade. Primary GBM’s do not have clear track-records of arising from lower-grade tumors, and have
Radiosensitizers in neurooncology
certain molecular features that distinguish them from secondary GBM. For instance, amplification of the epidermal growth factor receptor (EGFR) and overexpression of the EGFR protein are relatively common events in primary GBM’s, occurring in upwards of 50% of cases. Often, along with enhanced signaling through EGFR, there is concomitant loss of the INK4 gene, which codes for both p16 and p14ARF in primary GBM’s. Many primary GBM’s express the constitutively active EGFRvIII mutant that lacks the extracellular binding domain. The prognostic value of EGFR in GBM has been controversial. Wild-type EGFR expression, as determined by immunohistochemical analysis does not appear to be of prognostic value in GBM’s. There are reports that the constitutively active EGFRvIII mutant has prognostic value in selected series; however, has yet to be rigorously confirmed in a prospective manner. Loss of heterozygosity of chromosome 10 is a common event in GBM’s, occurring in upwards of 90% of cases. PTEN (phosphatase and tensin homology gene), which is a 30 phosphoinositol phosphatase is located on Chromosome 10q23.3 and is commonly lost in both primary and secondary GBM’s. Loss of PTEN results in constitutive activation of downstream mediators of the phosphatidyl 3-inositol kinase (PI3K) pathway, including AKT, which is a potent pro-survival molecule. > Figure 64-1 illustrates this pathway, as well as upstream regulators (e.g., EGFR). PI3K is a lipid kinase that promotes diverse biological functions including cellular proliferation, survival and motility [94]. The PI3K signaling pathway is frequently disregulated in glioblastoma, [95,96] often in combination with the ERK pathway, and mouse genetic studies suggest a causal role of this pattern [97]. Upwards of 40% of GBM’s contain alterations of the PTEN tumor suppressor gene, a negative regulator of PI3K signaling, which results in constitutive activation of the PI3K pathway [96]. Upstream of PI3K, the epidermal growth factor receptor (EGFR)
64
is commonly over-expressed, which may lead to disregulated PI3K and RAS/ERK signaling [98–104]. Other receptor tyrosine kinases such as PDGFR and c-MET are also commonly overexpressed in glioblastomas, and may deregulate these same pathways [98–105]. The PI3K and RAS/ERK pathways connect richly to other signaling cascades, thereby integrating signals associated with other cell surface events, stress activation pathways and extracellular matrix proteins. RAC1 is one such protein that links PI3K and RAS signaling with integrin-linked signaling, potentially playing a key role in promoting glioblastoma growth and survival [106]. Therefore, the PI3K and ERK pathways, provide important therapeutic targets. > Figure 64-1 illustrates various points within this pathway that can be targeted. The connectivity of the PI3K signaling pathway in GBM has recently been demonstrated by Mischel et al. It was demonstrated that PTEN loss was tightly linked to AKT activation. Further, it was determined that in a subset of glioblastoma patients treated by radiation alone, activation of PI3K pathway members was associated with adverse clinical outcome, providing direct clinical evidence of the role of PI3K signaling in radiation resistance in GBM’s.
Targeting EGFR Pathway Signaling in GBM’s – Clinical Data Phase I/II studies on the safety and efficacy of anti-EGFR agents in the setting of GBM suggest modest activity in the recurrent setting [107,108]. Interestingly, EGFR status, either measured by levels of IHC-detected EGFRwt or EGFRvIII or EGFR gene amplification, was not associated with outcome as measured by overall survival. In a separate Phase II study of another EGFR tyrosine kinase inhibitor, erlotinib (OSI-774) for recurrent high-grade glioma, [109] there were 6/25partial responses. Therefore, the emerging clinical data suggests that
995
996
64
Radiosensitizers in neurooncology
. Figure 64-1 Angiogenesis, signaling pathway and inhibitors in clinical trails. Binding of growth factors to its receptors leads to an activation of receptor tyrosine kinase activity and binding of adaptor and Ras activating proteins and other nucleotide exchange factor. Mature Ras-GDP will be converted to Ras-GTP, thereby activating Ras and active form (GTP) can be converted to inactive form (GTP) by guanosine triphosphate activating protein (GAP). Ras-GTP stimulates the downstream effector molecules like Raf, Rac, MEK, PI3K, etc., which promotes cell proliferation and survival. The Inhibitor used in the phaseI-III trials as well as in the clinical setting to inhibit the key signaling molecule which mediates treatment resistance is also shown. Vascular endothelial growth factor (VEGF) functions by activating two receptor tyrosine kinases, Flt-1 (VEGFR-1) and KDR (VEGFR-2), both of which are selectively expressed on the primary vascular endothelium. KDR is responsible for VPF/VEGF-stimulated endothelial cell (EC) proliferation and migration. KDR mediates cell survival through downstream targets such as PLG and CSK leading to proliferation, migration, vasculature and survival
there may be a subset of patients with malignant gliomas who are responsive to anti-EGFR therapies. As response to anti-EGFR therapies when used as single agents in the recurrent setting appears to be independent of EGFR status, the role of downstream signaling pathways must be more carefully scrutinized. In one report, activation of phosphatidylinositol 3-kinase members (PI3K, AKT, p70s6k) was associated with significantly worse survival times in glioma patients as a whole [110]. More specifically, in GBM patients treated with radiation as the primary adjuvant
therapy without chemotherapy, activation of these critical PI3K pathway members was associated with adverse survival, indicating a direct role for this pathway in treatment resistance. As more evidence that activation of downstream pathways may mediate resistance to anti-EGFR agents, in a study of two primary GBM cell lines with equivalent expression of EGFR, the two cell lines were found to have very different sensitivities to the anti-EGFR agent AG1478 [111]. Further investigation revealed that upregulation of the insulin-like growth factor receptor 1 (IGFR1) contributed to resistance to AG1478. Upon dual
Radiosensitizers in neurooncology
inhibition of IGFR1 and EGFR, it was possible to more significantly downregulate activity of PI3K/ AKT and reduce cellular survival as well as sensitivity to radiotherapy. In a correlative study on patients with recurrent GBM’s treated by antiEGFR agents, it was determined that patients most responsive to anti-EGFR therapies are those with expression of EGFRvIII and PTEN. Tumors that were PTEN-deficient were found to be resistant to anti-EGFR agents as a rule. It can be conceptualized that EGFRvIII drives signaling through PI3K in tumors with intact PTEN and would therefore be responsive to anti-EGFR therapies. In contrast, tumors with PTEN deficiency have constitutive signaling through PI3K/AKT. Since neither EGFR nor EGFRvIII is driving PI3K pathway activation in PTEN-deficient tumors, it stands to reason that anti-EGFR agents would have limited efficacy in suppressing pro-survival signaling through PI3K/AKT. The RTOG has investigated the safety and efficacy of Gefitinib, an EGFR tyrosine kinase inhibitor, in combination with radiation for newly diagnosed GBM patients (RTOG 0211) [112]. However, the preliminary results from the phase II portion of the study revealed similar survivals with the use of Gefitinib when compared to historical controls [113].
mTor Pathway Inhibition – CCI-779 Preclinical data suggests that the mammalian target of rapamycin (mTor) pathway, which is downstream of EGFR and PI3K/AKT, represents an attractive therapeutic target. CCI-779 is a small molecule inhibitor (Rapamycin analog) of mTor. In a recently reported NCCTG Phase II trial of CCI-779 in recurrent GBM, 41 patients with recurrent GBM were treated by CCI-779 at a dose of 250 mg i.v. qweek [114]. Assessment of tumor Akt and p70s6k kinase phosphorylation pretreatment demonstrated activation in the majority of
64
patients (14/17 and 11/17 patients, respectively). Post-treatment, an analysis of peripheral blood mononuclear cell phosphorylation of p70s6k showed post-treatment inhibition in 7/10 patients. There were signs of some activity, with a significant decrease in T2 abnormality in five patients and a significant decrease in T1 gadolinium enhancement in three patients, which fell short of partial response. Investigation of mTor inhibitors in combination with radiation is presently ongoing.
Angiogenesis Pathways There is ample evidence that angiogenesis plays an important role in the pathogenesis and treatment resistance in gliomas. One of the key underlying factors of angiogenesis may be the hypoxic environment that results when tumor growth outstrips blood and, hence, nutrient supply. A key transcription factor involved in angiogenesis in gliomas is hypoxia-induced factor 1 (HIF-1). HIF-1 is a heterodimer protein consisting of HIF-1a and a constitutively expressed aryl hydrocarbon receptor nuclear translocator (ARNT, also known as HIF1b). In pathologic specimens taken from GBM patients, HIF-1 expression has been commonly detected in the leading edge of invading tumor cells, as well as in the necrotic core of tumor [115,116]. In hypoxic conditions, HIF-1 binds to hypoxia response elements (HRE’s), thereby inducing expression of hypoxia-responsive gene that are involved in angiogenesis, invasion, and survival. Under normoxic conditions, HIF-1a is rapidly degraded by the proteosome. It is known that HIF-1a interacts with the Von Hippel Landau (VHL) protein, which helps to target HIF-1a for proteosomal degradation. It is also important to realize that hypoxia-independent factors may serve to increase HIF-1a expression, including loss of PTEN [117]. The vascular endothelial growth factor (VEGF) has also been found to play an important
997
998
64
Radiosensitizers in neurooncology
role in angiogenesis in malignant gliomas. VEGF is a highly specific endothelial mitogen. In addition to VEGF (VEGF-A), there are other important family members of this pathway including VEGF-B, VEGF-C, VEGF-D, VEGF-E, and placenta growth factor (PGF). There appears to be an accumulation of VEGF that increases with tumor grade and tumor size in gliomas [118]. Extracellular VEGF is known to bind to its tyrosine kinase receptors, VEGFR1 and VEGFR2, which are strongly expressed on endothelial cells present in the vasculature surrounding high-grade gliomas, but absent in normal brain vasculature. Hypoxia serves to increase VEGF expression levels via the HIF family. It has been found that VEGF-A and VEGF-B mRNA are commonly overexpressed in GBMs [118]. One report suggests that expression of VEGF family members is tightly linked to EGFR expression levels [119]. Further, it has been found that PI3K pathway activation serves to increase VEGF mRNA expression levels independent of hypoxia. Given the potential importance of angiogenic pathways in mediating radiation resistance, these molecules present themselves as being attractive targets in GBM’s.
PTK 787 (Vatalanib): Clinical Data Given the promise of anti-angiogenic strategies in human tumors, studies have been conducted in the setting of recurrent GBM. PTK 787/ZK 222584 is a drug that has been found to inhibit all known VEGF receptors and therefore inhibits signaling by VEGF’s (A-D). In a study of 55 patients with recurrent GBM treated by PTK787, [120] there were 2 (4%) partial responses, 31 (56%) stable disease, and 14 (25%) disease progression. Median duration of stable disease was 12.1 weeks. Dynamic contrast-enhancedenhanced (DCE) and dynamic susceptibility change (DSC) MRI revealed decreases in vascular permeability and cerebral blood volume. In a separate study, PTK-787 was combined with
either TMZ or CCNU [121]. Among 51 patients evaluable for response, 4 had a partial response and 27 patients had stable disease. The median time to progression was 15.7 weeks for the PTK787+TMZ arm and 10.4 weeks for the PTK-787+ CCNU arm. Phase I/II studies are currently planned investigating PTK-787 + TMZ in combination with radiotherapy for newly-diagnosed GBM patients. In an ongoing trial, vatalanib is evaluated in combination with radiotherapy and TMZ chemotherapy [122]. Preclinical observations of potential increase in hypoxia and radiation resistance raise concern as to the optimal timing of anti-angiogenic therapy [123]. The EORTC randomized phase II trial (EORTC 26041-22041) therefore evaluates safety and efficacy of the addition of vatalanib (PTK787) either concurrent with TMZ/RT, or by adding the VEGFR inhibitor only after completion of concomitant chemoradiotherapy [122].
Bevacizumab and AZD2111 Increased anti-tumor activity with the addition of bevacizumab to chemotherapy has been demonstrated for several types of malignancies including colorectal cancer, renal cell carcinoma, breast cancer, and non-small cell lung cancer. A recent study in recurrent malignant glioma reports a relatively high radiographic response rate of 60% with the combination of bevacizumab and irinotecan [78,124]. However, it remains unclear whether this increased response rate translates into a true prolongation of survival. Therefore, a definitive phase III trial is about to be initiated by the RTOG.A clinical trial with correlative imaging and biological endpoints suggests that VEGF receptor inhibition by AZD2171 leads to normalization of tumor vasculature and restoration of the blood-brain barrier, thus reducing contrast enhancement and edema [125]. These findings are consistent with prior observations in rectal cancer, that anti-VEGF therapy leads to
Radiosensitizers in neurooncology
normalization of the vasculature, decreased tumor interstitial pressure, better oxygenation and drug delivery [126]. A trial with the combined VEGFR and EGFR tyrosine kinase inhibitor vandetinib (ZD6474) is about to start accrual.
Cilengitide Integrins are heterodimer transmembrane receptors for the extracellular matrix, regulating cell adhesion and migration. In vessels, integrins interact with the basal membrane, thereby maintaining vascular quiescence; during angiogenesis they are essential for endothelial cell migration, proliferation and survival [127,128]. In preclinical models inhibition of integrin function efficiently suppresses angiogenesis and inhibits tumor progression, the integrins aVb3 and
64
aVb5 were identified as specific for tumor angiogenesis. Cilengitide, a synthetic RGD-motif peptide binds to the aVb3 and aVb5 integrin receptors. In a phase I study in recurrent glioma single agent activity of cilengitide was observed [129]. A phase II trial of cilengitide added to standard TMZ-based chemoradiation has recently been completed. Therapy was associated with little or no additional toxicity, initial results suggest efficacy in a subgroup of patients [128].
Summary Malignant gliomas and brain metastasis remain among the most treatment-refractory tumors. Traditional upfront treatment regimens have incorporated nitrosurea-based chemotherapy. This strategy has evolved to include temozolomide-based approaches. Promising Phase I/II
. Figure 64-2 Future treatment strategy. Over one-quarter of the patients enrolled on the TMZ+RT arm survived beyond 2-years, there appears to be a finite percentage of patients who derive long-term benefit from this treatment regimen. To this end, targeted therapies have emerged as an attractive option. Accumulating evidence suggests that certain molecular pathways are selectively upregulated in tumor vs. normal cells. Some of these pathways have been shown to be instrumental in proliferation, migration, invasion, angiogenesis, and/or survival in preclinical models. These would appear to represent ideal therapeutic targets, as their antagonism may lead to an improvement in the therapeutic ratio of radiation. Emerging data from clinical studies on ‘‘first generation’’ targeted therapies appear to demonstrate benefit for select patients. Further molecular/genetic profiling must be undertaken to identify exactly which patients benefit
999
1000
64
Radiosensitizers in neurooncology
data with TMZ in the recurrent setting prompted a Phase III EORTC study of TMZ in combination with RT for patients with newly-diagnosed GBM. The landmark EORTC 26981-22981/NCIC CE3 study demonstrated a significant improvement in not only median survival, but also in terms of 2-year survival. Given that over one-quarter of the patients enrolled on the TMZ+RT arm survived beyond 2 years, there appears to be a finite percentage of patients who derive long-term benefit from this treatment regimen. Given that the EORTC-based regimen represents an incremental improvement in the standard of care, rather than a truly curative solution for most patients, further efforts must be expended to identify novel therapeutic approaches. To this end, targeted therapies have emerged as an attractive option. > Figure 64-2 illustrates some of the promising targeted therapy strategies in gliomas. Accumulating evidence suggests that certain molecular pathways are selectively upregulated in tumor vs. normal cells. Some of these pathways have been shown to be instrumental in proliferation, migration, invasion, angiogenesis, and/or survival in preclinical models. These would appear to represent ideal therapeutic targets, as their antagonism may lead to an improvement in the therapeutic ratio of radiation. Emerging data from clinical studies on ‘‘first generation’’ targeted therapies appear to demonstrate benefit for select patients. Further molecular/genetic profiling must be undertaken to identify exactly which patients benefit.
References 1. Khuntia D, Brown P, Li J, Mehta MP. Whole-brain radiotherapy in the management of brain metastasis. J Clin Oncol 2006;24(8):1295-304. 2. Gaspar L, Scott C, Rotman M, et al. Recursive partitioning analysis (RPA) of prognostic factors in three Radiation Therapy Oncology Group (RTOG) brain metastases trials. Int J Radiat Oncol Biol Phys 1997;37(4):745-51. 3. Stupp R, Mason WP, van den Bent MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005;352(10):987-96.
4. Walker MD. The contemporary role of chemotherapy in the treatment of malignant brain tumor. Clin Neurosurg 1978;25:388-96. 5. Rowinsky EK. Novel radiation sensitizers targeting tissue hypoxia. Oncology (Williston Park) 1999;13 10 Suppl 5:61-70. 6. Urtasun R, Band P, Chapman JD, Feldstein ML, Mielke B, Fryer C. Radiation and high-dose metronidazole in supratentorial glioblastomas. N Engl J Med 1976; 294(25):1364-7. 7. Chang CH, Housepian EM. Tumors of the central nervous system: modern radiotherapy in multidisciplinary management. New York: Mason Publishing; 1982. 8. Bleehen NM, Wiltshire CR, Plowman PN, et al. A randomized study of misonidazole and radiotherapy for grade 3 and 4 cerebral astrocytoma. Br J Cancer 1981; 43(4):436-42. 9. Misonidazole in radiotherapy of supratentorial malignant brain gliomas in adult patients: a randomized double-blind study. Eur J Cancer Clin Oncol 1983; 19(1):39-42. 10. Carabell SC, Bruno LA, Weinstein AS, et al. Misonidazole and radiotherapy to treat malignant glioma: a phase II trial of the radiation therapy oncology group. Int J Radiat Oncol Biol Phys 1981;7(1):71-7. 11. Stadler B, Karcher KH, Kogelnik HD, Szepesi T. Misonidazole and irradiation in the treatment of high-grade astrocytomas: further report of the Vienna Study Group. Int J Radiat Oncol Biol Phys 1984;10(9): 1713-7. 12. A study of the effect of misonidazole in conjunction with radiotherapy for the treatment of grades 3 and 4 astrocytomas. A report from the MRC Working Party on misonidazole in gliomas. Br J Radiol 1983;56(669): 673-82. 13. Chang JE, Khuntia D, Robins HI, Mehta MP. Radiotherapy and radiosensitizers in the treatment of glioblastoma multiforme. Clin Adv Hematol Oncol 2007;5(11): 894-902, 7-15. 14. Nelson DF, Urtasun RC, Saunders WM, Gutin PH, Sheline GE. Recent and current investigations of radiation therapy of malignant gliomas. Semin Oncol 1986;13(1):46-55. 15. Chang CH. Hyperbaric oxygen and radiation therapy in the management of glioblastoma. Natl Cancer Inst Monogr 1977;46:163-9. 16. Suh JH. A phase 3, randomized, open-label, comparative study of standard whole brain radiation therapy (WBRT) with supplemental oxygen (O2), with or without RSR13, in patients with brain metastases. Neuro Oncol;2003:345-6. 17. Suh J, Stea B, Nabid A. Standard whole brain radiation therapy (WBRT) with supplemental oxygen (O2), with or without RSR-13 (efaproxiral) in patients with brain metastasis: results of the randomzied REACH (RT-009) study. Proc Am Soc Clin Oncol;22 Suppl 14:115s.
Radiosensitizers in neurooncology
18. Shaw E, Scott C, Suh J, et al. RSR13 plus cranial radiation therapy in patients with brain metastases: comparison with the Radiation Therapy Oncology Group Recursive Partitioning Analysis Brain Metastases Database. J Clin Oncol 2003;21(12):2364-71. 19. Kleinberg L, Grossman SA, Carson K, et al. Survival of patients with newly diagnosed glioblastoma multiforme treated with RSR13 and radiotherapy: results of a phase II new approaches to brain tumor therapy CNS consortium safety and efficacy study. J Clin Oncol 2002;20(14): 3149-55. 20. Scott C, Suh J, Stea B, Nabid A, Hackman J. Improved survival, quality of life, and quality-adjusted survival in breast cancer patients treated with efaproxiral (Efaproxyn) plus whole-brain radiation therapy for brain metastases. Am J Clin Oncol 2007;30(6):580-7. 21. Del Rowe J, Scott C, Werner-Wasik M, et al. Single-arm, open-label phase II study of intravenously administered tirapazamine and radiation therapy for glioblastoma multiforme. J Clin Oncol 2000;18(6):1254-9. 22. Sano K, Hoshino T, Nagai M. Radiosensitization of brain tumor cells with a thymidine analogue (bromouridine). J Neurosurg 1968;28(6):530-8. 23. Goffinet DR, Brown JM. Comparison of intravenous and intra-arterial pyrimidine infusion as a means of radiosensitizing tumors in vivo. Radiology 1977;124(3): 819-22. 24. Kinsella TJ, Mitchell JB, Russo A, et al. Continuous intravenous infusions of bromodeoxyuridine as a clinical radiosensitizer. J Clin Oncol 1984;2(10):1144-50. 25. Kinsella TJ, Collins J, Rowland J, et al. Pharmacology and phase I/II study of continuous intravenous infusions of iododeoxyuridine and hyperfractionated radiotherapy in patients with glioblastoma multiforme. J Clin Oncol 1988;6(5):871-9. 26. Jackson D, Kinsella T, Rowland J, et al. Halogenated pyrimidines as radiosensitizers in the treatment of glioblastoma multiforme. Am J Clin Oncol 1987;10(5): 437-43. 27. Freese A, O’Rourke D, Judy K, O’Connor MJ. The application of 5-bromodeoxyuridine in the management of CNS tumors. J Neurooncol 1994;20(1):81-95. 28. Sullivan FJ, Herscher LL, Cook JA, et al. National Cancer Institute (phase II) study of high-grade glioma treated with accelerated hyperfractionated radiation and iododeoxyuridine: results in anaplastic astrocytoma. Int J Radiat Oncol Biol Phys 1994;30(3): 583-90. 29. Phillips TL, Levin VA, Ahn DK, et al. Evaluation of bromodeoxyuridine in glioblastoma multiforme: a Northern California Cancer Center Phase II study. Int J Radiat Oncol Biol Phys 1991;21(3):709-14. 30. Phillips TL, Bodell WJ, Uhl V, Ross GY, Rasmussen J, Mitchell JB. Correlation of exposure time, concentration and incorporation of IdUrd in V-79 cells with radiation response. Int J Radiat Oncol Biol Phys 1989;16(5): 1251-5.
64
31. Groves MD, Maor MH, Meyers C, et al. A phase II trial of high-dose bromodeoxyuridine with accelerated fractionation radiotherapy followed by procarbazine, lomustine, and vincristine for glioblastoma multiforme. Int J Radiat Oncol Biol Phys 1999;45(1):127-35. 32. Prados MD, Seiferheld W, Sandler HM, et al. Phase III randomized study of radiotherapy plus procarbazine, lomustine, and vincristine with or without BUdR for treatment of anaplastic astrocytoma: final report of RTOG 9404. Int J Radiat Oncol Biol Phys 2004;58(4):1147-52. 33. Forouzannia A, Richards GM, Khuntia D, Mehta MP. Motexafin gadolinium: a novel radiosensitizer for brain tumors. Expert Rev Anticancer Ther 2007;7(6):785-94. 34. Khuntia D, Mehta M. Motexafin gadolinium: a clinical review of a novel radioenhancer for brain tumors. Expert Rev Anticancer Ther 2004;4(6):981-9. 35. Carde P, Timmerman R, Mehta MP, et al. Multicenter phase Ib/II trial of the radiation enhancer motexafin gadolinium in patients with brain metastases. J Clin Oncol 2001;19(7):2074-83. 36. Viala J, Vanel D, Meingan P, Lartigau E, Carde P, Renschler M. Phases IB and II multidose trial of gadolinium texaphyrin, a radiation sensitizer detectable at MR imaging: preliminary results in brain metastases. Radiology 1999;212(3):755-9. 37. Rosenthal DI, Nurenberg P, Becerra CR, et al. A phase I single-dose trial of gadolinium texaphyrin (Gd-Tex), a tumor selective radiation sensitizer detectable by magnetic resonance imaging. Clin Cancer Res 1999;5(4):739-45. 38. Mehta MP, Rodrigus P, Terhaard CH, et al. Survival and neurologic outcomes in a randomized trial of motexafin gadolinium and whole-brain radiation therapy in brain metastases. J Clin Oncol 2003;21(13):2529-36. 39. Meyers CA, Smith JA, Bezjak A, et al. Neurocognitive function and progression in patients with brain metastases treated with whole-brain radiation and motexafin gadolinium: results of a randomized phase III trial. J Clin Oncol 2004;22(1):157-65. 40. Mehta M, Gervais R, Chabot P, et al. Motexafin gadolinium (MGd) combined with prompt whole brain radiation therapy (RT) prolongs time to neurologic progression in non-small cell lung cancer (NSCLC) patients with brain metastases: results of a phase III trial. Proc Am Soc Clin Oncol 2006;24(18s):7014. 41. Ford JM, Seiferheld W, Alger JR, et al. Results of the phase I dose-escalating study of motexafin gadolinium with standard radiotherapy in patients with glioblastoma multiforme. Int J Radiat Oncol Biol Phys 2007;69(3): 831-8. 42. Regine WF, Schmitt FA, Scott C, et al. Feasibility of neurocognitive outcome evaluations in patients with brain metastases in a multi-institutional cooperative group setting: results of radiation therapy oncology group (RTOG) trial BR-0018. Int J Radiat Oncol Biol Phys 2004;58(5):1346-52. 43. Gerstner ER, Fine RL. Increased permeability of the blood-brain barrier to chemotherapy in metastatic brain
1001
1002
64 44.
45.
46.
47.
48.
49. 50.
51.
52.
53.
54.
55.
56.
57.
Radiosensitizers in neurooncology
tumors: establishing a treatment paradigm. J Clin Oncol 2007;25(16):2306-12. Muldoon LL, Soussain C, Jahnke K, et al. Chemotherapy delivery issues in central nervous system malignancy: a reality check. J Clin Oncol 2007;25(16): 2295-305. Ott RJ, Brada M, Flower MA, Babich JW, Cherry SR, Deehan BJ. Measurements of blood-brain barrier permeability in patients undergoing radiotherapy and chemotherapy for primary cerebral lymphoma. Eur J Cancer 1991;27(11):1356-61. Seute T, Leffers P, Wilmink JT, ten Velde GP, Twijnstra A. Response of asymptomatic brain metastases from smallcell lung cancer to systemic first-line chemotherapy. J Clin Oncol 2006;24(13):2079-83. DeAngelis LM, Mandell LR, Thaler HT, et al. The role of postoperative radiotherapy after resection of single brain metastases. Neurosurgery 1989;24(6):798-805. Bleyer WA. Neurologic sequelae of methotrexate and ionizing radiation: a new classification. Cancer Treat Rep 1981;65 Suppl 1:89-98. Evans AE. Central nervous system workshop. Cancer Clin Trials 1981;4 Suppl:31-5. Meyers CA, Smith JA, Bezjak A, et al. Neurocognitive function and progression in patients with brain metastases treated with whole-brain radiation and motexafin gadolinium: results of a randomized phase III trial. J Clin Oncol 2004;22(1):157-65. Ostermann S, Csajka C, Buclin T, et al. Plasma and cerebrospinal fluid population pharmacokinetics of temozolomide in malignant glioma patients. Clin Cancer Res 2004;10(11):3728-36. Brada M, Judson I, Beale P, et al. Phase I dose-escalation and pharmacokinetic study of temozolomide (SCH 52365) for refractory or relapsing malignancies. Br J Cancer 1999;81(6):1022-30. Stupp R, Dietrich PY, Ostermann Kraljevic S, et al. Promising survival for patients with newly diagnosed glioblastoma multiforme treated with concomitant radiation plus temozolomide followed by adjuvant temozolomide. J Clin Oncol 2002;20(5):1375-82. Combs SE, Gutwein S, Schulz-Ertner D, et al. Temozolomide combined with irradiation as postoperative treatment of primary glioblastoma multiforme. Phase I/II study. Strahlenther Onkol 2005;181(6):372-7. Gilbert MR, Friedman HS, Kuttesch JF, et al. A phase II study of temozolomide in patients with newly diagnosed supratentorial malignant glioma before radiation therapy. Neuro Oncol 2002;4(4):261-7. Wedge SR, Porteous JK, Glaser MG, Marcus K, Newlands ES. In vitro evaluation of temozolomide combined with X-irradiation. Anticancer Drugs 1997;8 (1):92-7. Chakravarti A, Erkkinen MG, Nestler U, et al. Temozolomide-mediated radiation enhancement in glioblastoma:
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
a report on underlying mechanisms. Clin Cancer Res 2006;12(15):4738-46. Hegi ME, Diserens AC, Gorlia T, et al. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med 2005;352(10):997-1003. Newlands ES, Stevens MFG, Wedge SR, Wheelhouse RT, Brock C. Temozolomide: a review of its discovery, chemical properties, pre-clinical development, and clinical trials. Cancer Treat Rev 1997;23:35-61. Wedge SR, Porteous JK, Glaser MG, Marcus K, Newlands ES. In vitro evaluation of temozolomide combined with X-irradiation. Anticancer Drugs 1997;8(1): 92-7. Antonadou D, Paraskevaidis M, Sarris G, et al. Phase II randomized trial of temozolomide and concurrent radiotherapy in patients with brain metastases. J Clin Oncol 2002;20(17):3644-50. Antonadou D, Coliarakis N, Paraskevaidis M, et al. Whole Brain Radiotherapy Alone or in Combination with Temozolomide for Brain Metastases. A Phase III Study. Int J Radiat Oncol Biol Phys 2002;54 Suppl 1:93-4. Chang JE, Robins HI, Mehta MP. Therapeutic advances in the treatment of brain metastases. Clin Adv Hematol Oncol 2007;5(1):54-64. Stewart DJ. Update on the role of topotecan in the treatment of non-small cell lung cancer. Oncologist 2004;9 Suppl 6:43-52. von Pawel J, Schiller JH, Shepherd FA, et al. Topotecan versus cyclophosphamide, doxorubicin, and vincristine for the treatment of recurrent small-cell lung cancer. J Clin Oncol 1999;17(2):658-67. Kim JH, Kim SH, Kolozsvary A, Khil MS. Potentiation of radiation response in human carcinoma cells in vitro and murine fibrosarcoma in vivo by topotecan, an inhibitor of DNA topoisomerase I. Int J Radiat Oncol Biol Phys 1992;22(3):515-8. Lamond JP, Mehta MP, Boothman DA. The potential of topoisomerase I inhibitors in the treatment of CNS malignancies: report of a synergistic effect between topotecan and radiation. J Neurooncol 1996; 30(1):1-6. Baker SD, Heideman RL, Crom WR, Kuttesch JF, Gajjar A, Stewart CF. Cerebrospinal fluid pharmacokinetics and penetration of continuous infusion topotecan in children with central nervous system tumors. Cancer Chemother Pharmacol 1996;37(3):195-202. Sung C, Blaney SM, Cole DE, Balis FM, Dedrick RL. A pharmacokinetic model of topotecan clearance from plasma and cerebrospinal fluid. Cancer Res 1994;54 (19): 5118-22. Wong ET, Berkenblit A. The role of topotecan in the treatment of brain metastases. Oncologist 2004;9(1): 68-79. Gruschow K, Klautke G, Fietkau R. Phase I/II clinical trial of concurrent radiochemotherapy in combination
Radiosensitizers in neurooncology
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
with topotecan for the treatment of brain metastases. Eur J Cancer 2002;38(3):367-74. Hedde JP, Neuhaus T, Schuller H, et al. A phase I/II trial of topotecan and radiation therapy for brain metastases in patients with solid tumors. Int J Radiat Oncol Biol Phys 2007;68(3):839-44. Kocher M, Eich HT, Semrau R, Guner SA, Muller RP. Phase I/II trial of simultaneous whole-brain irradiation and dose-escalating topotecan for brain metastases. Strahlenther Onkol 2005;181(1):20-5. Kadota RP, Stewart CF, Horn M, et al. Topotecan for the treatment of recurrent or progressive central nervous system tumors – a pediatric oncology group phase II study. J Neurooncol 1999;43(1):43-7. Macdonald D, Cairncross G, Stewart D, et al. Phase II study of topotecan in patients with recurrent malignant glioma. National Clinical Institute of Canada Clinical Trials Group. Ann Oncol 1996;7(2):205-7. Fisher B, Won M, Macdonald D, Johnson DW, Roa W. Phase II study of topotecan plus cranial radiation for glioblastoma multiforme: results of Radiation Therapy Oncology Group 9513. Int J Radiat Oncol Biol Phys 2002;53(4):980-6. Castellino RC, Elion GB, Keir ST, et al. Scheduledependent activity of irinotecan plus BCNU against malignant glioma xenografts. Cancer Chemother Pharmacol 2000;45(4):345-9. Vredenburgh JJ, Desjardins A, Herndon JE II, et al. Bevacizumab plus irinotecan in recurrent glioblastoma multiforme. J Clin Oncol 2007;25(30):4722-9. Tishler RB, Schiff PB, Geard CR, Hall EJ. Taxol: a novel radiation sensitizer. Int J Radiat Oncol Biol Phys 1992; 22(3):613-7. Tishler RB, Geard CR, Hall EJ, Schiff PB. Taxol sensitizes human astrocytoma cells to radiation. Cancer Res 1992;52(12):3495-7. Langer CJ, Ruffer J, Rhodes H, et al. Phase II Radiation therapy oncology group trial of weekly paclitaxel and conventional external beam radiation therapy for supratentorial glioblastoma multiforme. Int J Radiat Oncol Biol Phys 2001;51(1):113-9. Chang SM, Kuhn JG, Rizzo J, et al. Phase I study of paclitaxel in patients with recurrent malignant glioma: a North American Brain Tumor Consortium report. J Clin Oncol 1998;16(6):2188-94. Chang SM, Kuhn JG, Robins HI, et al. A Phase II study of paclitaxel in patients with recurrent malignant glioma using different doses depending upon the concomitant use of anticonvulsants: a North American Brain Tumor Consortium report. Cancer 2001;91(2): 417-22. Lederman G, Wronski M, Arbit E, et al. Treatment of recurrent glioblastoma multiforme using fractionated stereotactic radiosurgery and concurrent paclitaxel. Am J Clin Oncol 2000;23(2):155-9.
64
85. Crane CH, Sargent DJ. Substitution of oral fluoropyrimidines for infusional fluorouracil with radiotherapy: how much data do we need? J Clin Oncol 2004;22(15): 2978-81. 86. Rich T. Capecitabine and radiation therapy for advanced gastrointestinal malignancies. Oncology (Williston Park);2002;16(12 Suppl 14):27-30. 87. Rich TA, Shepard RC, Mosley ST. Four decades of continuing innovation with fluorouracil: current and future approaches to fluorouracil chemoradiation therapy. J Clin Oncol 2004;22(11):2214-32. 88. Shewach DS, Lawrence TS. Antimetabolite radiosensitizers. J Clin Oncol 2007;25(26):4043-50. 89. Coughlin CT, Richmond RC. Biologic and clinical developments of cisplatin combined with radiation: concepts, utility, projections for new trials, and the emergence of carboplatin. Semin Oncol 1989;16 (4 Suppl 6):31-43. 90. Peterson K, Harsh Gt, Fisher PG, Adler J, Le Q. Daily low-dose carboplatin as a radiation sensitizer for newly diagnosed malignant glioma. J Neurooncol 2001;53(1): 27-32. 91. Glass J, Silverman CL, Axelrod R, Corn BW, Andrews DW. Fractionated stereotactic radiotherapy with cisplatinum radiosensitization in the treatment of recurrent, progressive, or persistent malignant astrocytoma. Am J Clin Oncol 1997;20(3):226-9. 92. Grossman SA, O’Neill A, Grunnet M, et al. Phase III study comparing three cycles of infusional carmustine and cisplatin followed by radiation therapy with radiation therapy and concurrent carmustine in patients with newly diagnosed supratentorial glioblastoma multiforme: Eastern Cooperative Oncology Group Trial 2394. J Clin Oncol 2003;21(8):1485-91. 93. Guerrieri M, Wong K, Ryan G, Millward M, Quong G, Ball DL. A randomised phase III study of palliative radiation with concomitant carboplatin for brain metastases from non-small cell carcinoma of the lung. Lung cancer 2004;46(1):107-11. 94. Vivanco I, Sawyers CL. The phosphatidylinositol 3-Kinase AKT pathway in human cancer. Nat Rev Cancer 2002;2(7):489-501. 95. Choe G, Horvath S, Cloughesy TF, et al. Analysis of the phosphatidylinositol 30 -kinase signaling pathway in glioblastoma patients in vivo. Cancer Res 2003;63(11): 2742-6. 96. Ermoian RP, Furniss CS, Lamborn KR, et al. Dysregulation of PTEN and protein kinase B is associated with glioma histology and patient survival. Clin Cancer Res 2002;8(5):1100-6. 97. Holland EC, Celestino JC, Dai C, Schaefer L, Sawaya RE, Fuller GN. Combined activation of Ras and Akt in neural progenitors induce glioblastoma formation in mice. Nat Genet 2000;25(1):55-7. 98. von Deimling A, von Ammon K, Schoenfeld D, Wiestler OD, Seizinger BR, Louis DN. Subsets of
1003
1004
64 99.
100.
101.
102.
103. 104.
105.
106.
107.
108.
109.
110.
111.
112.
Radiosensitizers in neurooncology
glioblastoma multiforme defined by molecular genetic analysis. Brain Pathol 1993;3(1):19-26. Watanabe K, Tachibana O, Sata K, Yonekawa Y, Kleihues P, Ohgaki H. Overexpression of the EGF receptor and p53 mutations are mutually exclusive in the evolution of primary and secondary glioblastomas. Brain Pathol 1996; 6(3):217-23; discussion 23-4. Kleihues P, Ohgaki H. Primary and secondary glioblastomas: from concept to clinical diagnosis. Neuro Oncol 1999;1(1):44-51. Frederick L, Eley G, Wang XY, James CD. Analysis of genomic rearrangements associated with EGRFvIII expression suggests involvement of Alu repeat elements. Neuro Oncol 2000;2(3):159-63. Kleihues P, Louis DN, Scheithauer BW, et al. The WHO classification of tumors of the nervous system. J Neuropathol Exp Neurol 2002;61(3):215-25; discussion 26–9. Mischel PS, Cloughesy TF. Targeted molecular therapy of GBM. Brain Pathol 2003;13(1):52-61. Rao RD, Uhm JH, Krishnan S, James CD. Genetic and signaling pathway alterations in glioblastoma: relevance to novel targeted therapies. Front Biosci 2003;8: e270-e280. Abounader R, Lal B, Luddy C, et al. In vivo targeting of SF/HGF and c-met expression via U1snRNA/ribozymes inhibits glioma growth and angiogenesis and promotes apoptosis. FASEB J 2002;16(1):108-10. Senger DL, Tudan C, Guiot MC, et al. Suppression of Rac activity induces apoptosis of human glioma cells but not normal human astrocytes. Cancer Res 2002;62(7): 2131-40. Prados M, Yung WK, Wen PY, et al. Phase I study of ZD1839 plus temozolomide in patients with malignant glioma. J Clin Oncol 2004;22(14s):1504. Rich JN, Reardon DA, Peery T, et al. Phase II trial of gefitinib in recurrent glioblastoma. J Clin Oncol 2004;22(1):133-42. Prados M, Chang S, Burton EC, et al. Phase I study of OSI-774 alone or with temozolamide in patients with malignant glioma. Proc Am Soc Clin Oncol 2003;22:99. Chakravarti A, Zhai G, Suzuki Y, et al. The prognostic significance of phosphatidylinositol 3-kinase pathway activation in human gliomas. J Clin Oncol 2004; 22(10):1926-33. Chakravarti A, Loeffler JS, Dyson NJ. Insulin-like growth factor receptor I mediates resistance to antiepidermal growth factor receptor therapy in primary human glioblastoma cells through continued activation of physphoinositide 3-kinase signaling. Cancer Res 2002; 62(1):200-7. Chakravarti A, Seiferheld W, Robins IA, et al. An update of phase I data from RTOG 0211: A phase I/II clinical study of gefitinib + radiation for newly-diagnosed GBM patients. J Clin Oncol 2004;22(14S): 1571.
113. Chakravarthy A, Berkey B, Robins HI, et al. An update of phase II results from RTOG 0211: A phase I/II study of gefitinib with radiotheapy in newly diagnosed glioblastoma. J Clin Oncol 2006;24 Suppl 18:1527. 114. Galanis E, Buckner JC, Maurer M, et al. NCCTG phase II trial of CCI-779 in recurrent glioblastoma multiforme (GBM). J Clin Oncol 2004;22(14S):1503. 115. Zagzag D, Capo V. Angiogenesis in the central nervous system: a role for vascular endothelial growth factor/ vascular permeability factor and tenascin-C. Common molecular effectors in cerebral neoplastic and nonneoplastic ‘‘angiogenic diseases’’. Histol Histopathol 2002; 17(1):301-21. 116. Zagzag D, Amirnovin R, Greco MA, et al. Vascular apoptosis and involution in gliomas precede neovascularization: a novel concept for glioma growth and angiogenesis. Lab Invest 2000;80(6):837-49. 117. Zundel W, Schindler C, Haas-Kogan D, et al. Loss of PTEN facilitates HIF-1-mediated gene expression. Genes Dev 2000;14(4):391-6. 118. Gollmer J, Ladoux A, Gioanni J, et al. Expression of vascular endothelial growth factor-b in human astrocytoma. Neuro Oncol 2000;2(2):80-6. 119. Mischel PS, Shai R, Shi T, Horvath S, Lu KV, Choe G, Seligson D, Kremen TJ, Palotie A, Liau LM, Cloughesy TF, Nelson SF. Identification of molecular subtypes of glioblastoma by gene expression profiling. Oncogene 2003;22(15):2361-73. 120. Conrad C, Friedman HS, Reardon DA, et al. A phase I/II trial of single-agent PTK 787/ZK 222584 (PTK/ZK), a novel, oral angiogenesis inhibitor, in patients with recurrent GBM. J Clin Oncol 2004;22(14S):1512. 121. Reardon DA, Friedman HS, Yung WK, et al. A phase I/II trial of PTK-787/ZK 222584 (PTK/ZK), a novel, oral angiogenesis inhibitor, in combination with either temozolomide or lomustine for patients with recurrent GBM. J Clin Oncol 2004;22(14S):1513. 122. Brandes AA, Stupp R, Hau P, et al. EORTC study 26041–22041: phase I/II study on concomitant and adjuvant temozolomide (TMZ) and radiotherapy (RT) with or without PTK787/ZK222584 (PTK/ZK) in newly diagnosed glioblastoma – results of a phase I trial. J Clin Oncol 2007; ASCO Annual Meeting Proceedings 2007; 25(18S):2026. 123. Wachsberger PR, Burd R, Marero N, et al. Effect of the tumor vascular-damaging agent, ZD6126, on the radioresponse of U87 glioblastoma. Clin Cancer Res 2005; 11(2 Pt 1):835-42. 124. Vredenburgh JJ, Desjardins A, Herndon JE II, et al. Phase II trial of bevacizumab and irinotecan in recurrent malignant glioma. Clin Cancer Res 2007;13(4):1253-9. 125. Batchelor TT, Sorensen AG, di Tomaso E, et al. AZD2171, a pan-VEGF receptor tyrosine kinase inhibitor, normalizes tumor vasculature and alleviates edema in glioblastoma patients. Cancer Cell 2007;11(1):83-95.
Radiosensitizers in neurooncology
126. Willett CG, Boucher Y, di Tomaso E, et al. Direct evidence that the VEGF-specific antibody bevacizumab has antivascular effects in human rectal cancer. Nat Med 2004;10(2):145-7. 127. Alghisi GC, Ruegg C. Vascular integrins in tumor angiogenesis: mediators and therapeutic targets. Endothelium 2006;13(2):113-35.
64
128. Stupp R, Ruegg C. Integrin inhibitors reaching the clinic. J Clin Oncol 2007;25(13):1637-8. 129. Nabors LB, Mikkelsen T, Rosenfeld SS, et al. Phase I and correlative biology study of cilengitide in patients with recurrent malignant glioma. J Clin Oncol 2007;25 (13):1651-7.
1005
73 Radiosurgery for Functional Neurosurgery D. Kondziolka
Introduction The history of stereotactic radiosurgery goes back almost as far as the history of functional neurosurgery. Leksell initially conceived the idea of closed-skull, single-session irradiation of a precisely defined intracranial target in 1951 and applied this concept immediately to functional neurosurgery [1]. At a time when functional destruction of normal brain required thermal energy or chemical injection, Leksell crossfired photon or proton radiation beams to achieve a similar goal. The initial radiosurgical concept was to create a small, precisely defined focal lesion, which was defined by image guidance. The procedure would not completely avoid brain penetration since contrast encephalography provided the information for identification of the targets. Whereas the ganglionic portion of the trigeminal nerve could be indirectly located using plain radiographs or cisternograms, deep brain targets required air or positive contrast ventriculography; direct visualization of the target for functional radiosurgery required the later development of computed imaging technology. Thus, radiosurgical techniques were used to create image-guided, physiologic inactivity or focally destructive brain lesions without neurophysiologic guidance. This was controversial, and the lack of neurophysiologic guidance remained the greatest argument against the use of radiosurgery for selected disorders. Nevertheless, the current use of radiosurgery as a ‘‘lesion generator’’ is based on
#
Springer-Verlag Berlin/Heidelberg 2009
extensive animal studies that defined the dose, volume, and temporal response of the irradiated tissue. The utility of radiosurgery has now been compared to microsurgical, percutaneous injection, and electrode-based techniques used for functional neurologic disorders. Current anatomic targets include the trigeminal nerve and other ganglia (facial pain syndromes – discussed in a separate chapter), the thalamus (for tremor or pain), the cingulate gyrus or anterior internal capsule (for pain or psychiatric illness), the hypothalamus (for cancer pain), and the hippocampus or other brain targets (for epilepsy) [2,3]. Leksell first coupled an orthovoltage X-ray tube to his early generation stereotactic frame, a concept used for trigeminal neuralgia but not for intraparenchymal brain targets [4]. Thus, he began work with physicist Borje Larsson to crossfire proton beams [5], and subsequently used a modified linear accelerator. His decision to build and then use the first Gamma Knife in 1967 reflected his frustration with particle beam technology which required travel of the patient to a special cyclotron center. As originally designed, the first Gamma Knife collimator helmets created a discoid volume of focal irradiation that could ‘‘section’’ white matter tracts or brain tissue in a manner similar to a leucotome or other instrument. Later models of the Gamma Knife have provided more flexibility in the creation of lesions or effects of different volumes (more suitable for tumors or vascular malformations), together with precise robotic delivery and efficiency.
1192
73
Radiosurgery for functional neurosurgery
The Beginning Years Before 1978, all uses of radiosurgery remained limited due to the lack of high-resolution, neuroimaging techniques to identify brain lesions or functional brain regions. Angiographic targeting of arteriovenous malformations proved successful, but proved to be limited by the two-dimensional estimates of complex three-dimensional target volumes. Functional radiosurgery was performed for a limited number of patients with intractable pain related to malignancy [6,7], movement disorders [2], psychiatric dysfunction [8,9], and trigeminal neuralgia [4]. On the sidelines, percutaneous retrogasserian glycerol rhizotomy was developed during an observation made during the refinement of the Gamma Knife technique for trigeminal neuralgia. Haka¨nsson and Leksell attempted to localize the trigeminal nerve within its cistern using glycerol mixed with tantalum powder (as a radiopaque marker) placed before radiosurgery. However, after injection of the glycerol, trigeminal neuralgia pain was relieved. For intractable pain related to malignancy, radiosurgery was used both for hypophysectomy as well as for medial thalamotomy. Although the procedure was non-invasive, the latency interval for lesion generation and pain relief was one obvious drawback. Steiner and colleagues presented results from an autopsy study after radiosurgery for cancer pain in 1980 [7]. The ablative dose for tissue volumes had been estimated during animal experiments through the 1960s using protons and photons [5,10,11]. Initial patients who had radiosurgery for tissue ablation received maximum doses of 100–250 Gy. At small volumes, doses in excess of 150 Gy provided consistent tissue necrosis in animal models. Since these first patients were treated for pain from a terminal malignancy, they did not live long enough to sustain a potential complication from such high doses. The clinical use of such doses proved to be the foundation for later use in tremor management.
Dose-Selection for Parenchymal Functional Radiosurgery Early animal experiments showed consistent lesion creation at doses at or above 150 Gy [10,11]. Clinical data showed that pain relief occurred usually within 3 weeks after radiosurgery [7]. In rat experiments at 200 Gy using a single 4 mm isocenter, we found a consistent relationship for lesion generation that substantiated observations from that human study [12]. Doses of 200 Gy were delivered to the rat frontal brain and then the brain was studied at one, 7, 14, 21, 60, and 90 days after irradiation. At 1 and 7 days, we noted that the brain continued to appear normal. By 14 days, the parenchyma appeared slightly edematous within the target volume. However, by 21 days, a complete circumscribed volume of necrosis was identified within the radiation volume (4 mm diameter). This remained consistent thereafter. Thus, the clinical observation of pain relief at 21 days noted by Steiner et al. was correlated with laboratory findings at the 200 Gy dose. The ablative radiosurgery lesion appears as a punched-out, circumscribed volume of complete parenchymal necrosis with cavitation. Within a 1–3 mm rim that characterizes the steep falloff in radiation dose, normalization of the tissue appearance is found. In this zone, blood vessels appear thickened and hyalinized, and often protein extravasation can be identified. The brain is edematous in this region, either from an increase in extracellular fluid, or from the intracellular swelling of gliosis. Acute or chronic inflammatory cells are present. MR imaging demonstrates all of these features after radiosurgical thalamotomy – a sharply defined, contrast-enhanced rim that defines the low signal lesion (on short TR images) surrounded by a zone of high-signal (on long TR images) brain tissue [13]. Friehs and colleagues collected imaging data from four centers who created functional radiosurgery lesions (n = 56). They found that maximum doses in excess of 160 Gy were more
Radiosurgery for functional neurosurgery
likely to produce lesions larger than expected and recommended single 4 mm isocenter lesions at doses below 160 Gy [14]. Studies at the University of Pittsburgh found that in both large and small animal models, doses at or above 100 Gy caused necrosis, but the delay to necrosis was longer [12,15]. To identify the effect of increasing volume, we used an 8 mm collimator in a baboon model and found that half of the animals developed an 8 mm diameter necrotic lesion at doses as low as 50 Gy [16]. Dose, volume and time are the three key factors that determine the nature of the functional ablative lesion. Once created, this lesion remains stable over years [10]. The limitation of radiosurgery technology as a lesion generator stems from the inability to reliably control the effects of dose and volume. When a larger brain target may be desirable, the sharp fall-off in dose outside the target becomes less steep with increasing volume. The risk of an adverse radiation effect outside the target volume becomes problematic [17]. At small volumes (i.e., single 4 mm collimator), the radiosurgerycreated lesion appears more consistent.
Imaging in Functional Surgery Since physiological information is excluded from the targeting component of a functional radiosurgery procedure, high-quality stereotactic neuroimaging must be performed. The imaging must be accurate since small volumes are irradiated. In addition, the imaging must be of sufficient resolution to identify the target structure but also show important regional tissues. Magnetic resonance imaging is the preferred imaging tool for functional radiosurgery [18–20]. Accurate stereotactic MRI-based localization should be confirmed at each institution [21]. The use of fast inversion recovery or other long relaxation time MR sequences helps to separate gray and white matter structures. However,
73
to this day the targeting of physiologically abnormal brain regions such as groups of kinesthetic thalamic tremor cells or epileptic foci using imaging alone remains indirect. We believe that with improvements in subcortical imaging using higher field strength magnets, gamma knife radiosurgery will play an expanded role in movement disorders.
Radiosurgical Thalamotomy Ventrolateral thalamic surgery for the management of tremor related to Parkinson’s Disease remains a proven and time-honored concept within functional neurosurgery. Traditionally, this has involved imaging definition of the thalamic target, placement of an electrode into the thalamus, physiologic recording and stimulation at the target site, and creation of a lesion or providing electrical stimulation. Radiosurgical thalamotomy by definition avoids placement of the electrode and evaluation of the physiologic response. In radiosurgery, imaging definition alone is used to determine lesion placement. Through the use of contrast ventriculography, computed tomographic imaging, and more recently stereotactic MRI, thalamotomy using the Gamma Knife has been performed at centers across the world [18,22–24]. As discussed above, the issues of lesion volume and dose-selection remain important. Although radiosurgery can abolish tremor, many surgeons currently believe although adequate results might be obtained, better results may be possible with deep brain stimulation (DBS). The challenges inherent in choosing the best possible ablative target using imaging alone are significant. Radiosurgical thalamotomy, if performed, should be performed by surgeons experienced in radiofrequency thalamotomy or DBS. Due to the absence of electrophysiological information, the inability to stop the lesion during surgery, and the latency to the clinical
1193
1194
73
Radiosurgery for functional neurosurgery
response, most surgeons use radiosurgery primarily for patients with advanced age or medical disorders where electrode placement would be associated with higher-risk. Ohye began to perform radiosurgical thalamotomy contralateral to a prior radiofrequency lesion, or to enlarge a previously mapped lesion [22]. Duma et al. reported a 5-year experience with 38 thalamotomies using the Gamma Knife and 28 month mean follow-up [18]. Complete tremor abolition was noted in 24%, excellent relief in 26%, good improvement in 29%, and little to no benefit in 21%. The median time to improvement was 2 months, consistent with data from previous animal experiments. They used a dose range of 110–165 Gy with better results at higher doses. Such higher doses may exert effects on a larger surrounding tissue volume of kinesthetic tremor cells (outside the sharply defined necrotic volume) that translates into tremor reduction and overcomes any limitations in target selection. Young et al. reported that 88% of 27 patients who had radiosurgical thalamotomy for tremor (120–160 Gy) became tremor free or ‘‘nearly’’ tremor free [20]. Hirato et al. also found tremor suppression after Gamma Knife thalamotomy in a small patient series [22]. Friehs et al. reported an experience of radiosurgical thalamotomy (n = 3) and caudatotomy (n = 10) with clinical improvement in most patients and no morbidity [25]. The mean age of 77 years in the Pittsburgh radiosurgical series is older than the mean age of 60 years in their DBS series [26]. Gamma Knife radiosurgery proved to be effective in improving medically-refractive essential tremor in a predominantly elderly patient series (> Figure 73-1) [27]. Eighteen patients (69%) improved both action tremor and writing scores, and an additional six (23%) improved their action tremor scores. Thirteen patients (50%) had either no or only slight intermittent tremor in the affected extremity and 90% had some degree of clinically significant tremor improvement. Overall, the
. Figure 73-1 CT scan 4 months after a left gamma knife thalamotomy (140 Gy) in a 90 year-old woman with essential tremor. The radiosurgical lesion is shown at the v.i.m target
mean Fahn-Tolosa-Marin tremor score improved from 3.8 to 1.7 (p < 0.000015). After radiosurgery, MRI usually showed a 4–5 mm round, well-circumscribed lesion with peripheral contrast enhancement surrounding a low signal region (> Figure 73-2). A localized area of high signal (seen on long relaxation time studies) demonstrated the peripheral neuronal effect, manifested as an increase in intra- or extracellular water. Two patients with complications had different MRI findings. Although the enhancing lesion created in one patient was unexpectedly large, complete resolution was seen on subsequent imaging. This indicated that the response was related to temporary blood brain barrier changes, and not to permanent radiation necrosis. As noted earlier, the target volume is crucial. Early results with larger target volumes using an 8 mm collimator were reported by Lindquist et al. [2]. Delayed cerebral edema and regions of radiation necrosis at high doses testified to the
Radiosurgery for functional neurosurgery
73
. Figure 73-2 (a) MRI scan (flair) 6 months after gamma knife radiosurgery in a 56-year-old man with essential tremor who had refused deep brain stimulation. Significant tremor abolition was noted without side effects. (b) the MRI scan with contrast 3 years later is shown depicting a stable lesion
volume effects of radiosurgery [13]. Similar problems have been noted using combinations of 4 mm isocenters to construct a cylindrical rather than spherical target volume [17,20]. Nevertheless, the ability to create a small-volume lesion using radiosurgery without placement of a craniostomy or the invasive placement of an electrode remain attractive considerations. To that end, several surgeons have evaluated the use of radiosurgery for medial thalamotomy and for pallidotomy, procedures where the usefulness of physiologic recording or stimulation initially was less clear.
Radiosurgical Pallidotomy There was a resurgence in the use of radiofrequency-based stereotactic pallidotomy for patients with advanced Parkinson’s disease (PD) beginning in 1992. Some investigators then performed Gamma Knife pallidotomy using imageguidance alone as an alternative to electrode techniques. Rand et al. reported their preliminary results after radiosurgical pallidotomy and
noted relief of contralateral rigidity in 4 of 8 patients [28]. No patient in their series sustained a complication. Friedman et al. reported on four patients after Gamma Knife pallidotomy (180 Gy) with improvement in only one patient [29]. They noted heterogeneity in lesion volumes on MRI, a finding also documented by others. In contrast to the thalamus, where small radiosurgery lesions appeared consistent, pallidotomy lesions may be more variable due to effects on perforating arteries that supply that region of the basal ganglia. The lesion volumes and contrastenhancement patterns seem less consistent [20]. At our center only one radiosurgical pallidotomy has been performed. At present, this technique is performed rarely and DBS remains a much more valuable concept for most patients with an array of PD symptoms [30].
Radiosurgery for Pain The use of radiosurgery as an ablative tool to treat pain has a long history. Unfortunately too few patients have been managed to draw any
1195
1196
73
Radiosurgery for functional neurosurgery
strong conclusions. Since the case report by Leksell in 1968 and the larger series by Steiner et al. in 1980, little has been written [6,7]. In Leksell’s two patients with carcinoma, the centrum medianum target received doses of 250 and 200 Gy. The second patient had bilateral radiosurgery spaced by 2 months and became pain-free. In Steiner’s series, doses as high as 250 Gy were believed unnecessary because of the sharp dosegradient. Young et al. performed medial thalamotomy for the treatment of chronic non-cancer pain in patients who had failed comprehensive medical, surgical, and behavioral therapies [31]. In 1996 they described that two-thirds of their 41 patient series had at least a 50% reduction in pain intensity estimates with improvements in physical and social functioning [3]. As might be expected, patients with deafferentation pain responded poorly, but more encouraging results were identified in patients with nociceptive syndromes. Again they cautioned on the use of larger volumes above that obtained with a single 4 mm isocenter, and on the use of doses above 160 Gy. Hayashi et al. performed pituitary glandstalk ablation by Gamma Knife radiosurgery, targeting the border between the pituitary stalk and gland with a maximum dose of 160 Gy using the 8-mm collimator to control cancer pain. They enrolled nine patients that had bone metastases and pain controlled well by morphine, KPS >40, and no previous radiation therapy [32]. All patients had failed the previous pain treatments except morphine. All patients became pain-free within a few days after radiosurgery, and which maintained as long as they lived. No recurrence of pain occurred. In addition, there was no panhypopituitarism and diabetes insipidus in the patients. This strategy of pituitary gland-stalk ablation for pain control also showed a good initial response (87.5%) of 8 patients with thalamic pain syndrome, however, the majority of patients (71.4%) experienced pain recurrence during the 6 month follow-up [33].
Radiosurgery for Psychiatric Disorders There is renewed interest in radiosurgical lesioning of the anterior internal capsule (anterior capsulotomy) in patients with medically refractory obsessive compulsive disorder (OCD). Radiosurgery for obsessive-compulsive and anxiety neurosis has been performed for over 45 years [8]. The first radiosurgical capsulotomy was performed by Leksell in 1953 using 300 kV X-rays [34]. Initially, pneumoencephalography was used for target definition in the placement of bilateral anterior internal capsule lesions. Five of the initial 7 patients had long-term benefit after 7 years of follow-up [2]. Since 1988, an additional 10 patients were treated at the Karolinska Institute using stereotactic MRI guidance. The initial use of an 8 mm collimator resulted in excessive edema, so these authors recommended the use of only 4 mm isocenters [2]. The results seem to be as efficacious as when conventional radiofrequency lesioning is performed [35]. Kihlstrom et al. described the stable imaging appearance of radiosurgical lesions 15–18 years after capsulotomy [36]. Oval shaped radiosurgical lesions in the anterior internal capsule or cingulate gyrus may impact on affective disorders or anxiety neuroses. Recently, Ruck et al. reported on long-term follow-up in 25 patients, 16 with an electrode and 9 with gamma knife surgery [37]. Response rates did not differ between methods and they concluded that capsulotomy was effective in reducing OCD symptoms. A series of patients from Brown University and the University of Pittsburgh have been presented at national meetings. The radiosurgical capsulotomy is performed only after comprehensive psychiatric evaluation and management leading to a diagnosis of severe OCD, and after failure of non-surgical approaches. In Pittsburgh, we performed Gamma Knife surgery on three patients with severe, medically intractable OCD (> Figure 73-3). According to our protocol,
Radiosurgery for functional neurosurgery
. Figure 73-3 Bilateral anterior capsulotomies are shown on coronal contrast-enhanced MRI, 1 year after gamma knife radiosurgery (140 Gy) in a patient with obsessivecompulsive disorder
all patients were evaluated by at least two psychiatrists who recommended the capsulotomy procedure. The patient had to request the procedure, and have severe OCD according to the Yale Brown Obsessive Compulsive Scale (YBOCS). Patient ages were 37, 55, and 40, and pre-radiosurgery YBOCS scores were 32/40/39/ 40, and 39/40. Bilateral lesions were created with two 4 mm isocenters to create an oval volume in the ventral capsule at the putaminal midpoint. A maximum dose of 140–150 Gy was used. There was no morbidity after the procedure and all returned immediately to baseline function. All three patients had described functional improvements, and reduction in OCD behavior. One patient with compulsive skin picking and an open wound had later healing of their chronic wound and a reduction in the YBOCS score from 39 to 8/40 [38]. We believe this technique should be evaluated further in patients with severe and disabling behavioral disorders.
73
Radiosurgery for Epilepsy There is current interest in the use of radiosurgery for patients with focal epilepsy. The observation that brain irradiation (via radiation therapy or radiosurgery) could lead to cessation of seizures has spurred several groups to work in this field despite the lack of a consistent approach to defining the target volume. In 1985, BarciaSalorio et al. reported on six patients with epilepsy who had low dose radiosurgery. The epileptic focus was localized by means of conventional scalp electroencephalogram (EEG), subarachnoid electrodes, and depth electrodes. Radiosurgery (a 10 mm collimator to deliver an estimated dose of 10 Gy) was performed using a cobalt unit coupled to a stereotactic localizer. They hypothesized that this low radiation dose provided a specific effect on epileptic neurons, without inducing tissue necrosis. In 1994, they provided a long-term analysis in a series of 11 patients using a dose range of 10–20 Gy. Five patients had complete cessation of seizures, and an additional five were improved. Seizures began to decrease gradually after 3–12 months following radiosurgery [39]. Following this work, Lindquist and colleagues at the Karolinska Institute began to perform epilepsy radiosurgery using advanced localization techniques that included magnetoencephalography (MEG) to define interictal activity [2,40]. In some patients, the epileptic dipole activity identified on MEG before radiosurgery later resolved along with seizure cessation. At the same time, radiosurgery was evaluated in animal models of epilepsy. We used the kainic acid model of hippocampal epilepsy in the rat, and were able to stop seizures and improve animal behavior [41,42]. Rats were randomized to control or radiosurgery arms (20, 40, 60, or 100 Gy) and then evaluated with serial EEG, behavioral studies, functional MRI, and histology. More recently, radiosurgery has been of value in patients with gelastic or generalized seizures related to hypothalamic hamartomas
1197
1198
73
Radiosurgery for functional neurosurgery
. Figure 73-4 Gamma knife radiosurgery plan for a patient with gelastic seizures and complex epilepsy related to a hypothalamic hamartoma
(> Figure 73-4) [43]. A larger indication may rest with the use of epilepsy to create an amygdalohippocampal lesion for patients with mesial temporal sclerosis as proposed by Regis et al. [44,45]. In 1993, Re´gis and associates in Marseille performed selective amygdalohippocampal radiosurgery for mesial temporal lobe epilepsy. Gamma Knife radiosurgery was used to create a conformal volume of radiation for the amygdala and hippocampus. This approximate 7 ml volume represented the largest functional target irradiated to that time. They delivered a margin dose of 25 Gy to the 50% isodose line, a dose that later caused target necrosis. The first patient became seizure free immediately and the second
after a latency of almost 1 year. Serial MR scans showed target contrast-enhancement that corresponded to the 50% isodose line [44]. Patients managed at their center have been part of a multidisciplinary prospective evaluation and treatment protocol. A recently published longer-term evaluation with 8 year mean follow-up (margin dose of 24 Gy), found that 9 of 16 patients were seizure free [46]. The first prospective multicenter clinical trial in the United States was recently completed (> Figure 73-5). A number of questions remain to be addressed regarding the role of radiosurgery for mesial temporal sclerosis-related epilepsy. The optimal target may include both
Radiosurgery for functional neurosurgery
73
. Figure 73-5 Coronal MRI images at amygdalohippocampal radiosurgery (a) in a patient with mesial temporal sclerosis and at 15 months (b), showing the left sided lesion. Significant reduction in his seizures began at 10 months, with no adverse effects
amygdala and hippocampus, but the total target volume remains debated. Target volume helps to determine dose selection, including the dose received by regional structures such as the brainstem or optic tract. Finally, investigators need to determine whether the balance between seizure response and morbidity is acceptable, particularly in comparison to surgical resection. A randomized trial comparing radiosurgery to resection is planned. ‘‘Radiosurgery has been evaluated in animal models and in clinical use for epilepsy. The opportunity to use radiosurgery to disrupt an epileptic focus, or to change abnormal neurophysiologic patterns is of interest. We used the kainic acid model of hippocampal epilepsy in the rat, and were able to stop seizures and improve animal behavior (41,42). In clinical use, radiosurgery has been of value in patients with gelastic or generalized seizures related to hypothalamic hamartomas (47). A larger indication may rest with the use of epilepsy to create an amygdalohippocampal lesion for patients with mesial temporal sclerosis. First tested by Regis and colleagues from Marseille, a first prospective
clinical trial in the United States was recently completed.’’ Current issues that remain important for epilepsy radiosurgery include dose-selection (necrotizing vs. non-necrotizing), localization methods for non-lesional epilepsy, the target volume necessary for irradiation, and the expected short and long-term outcomes. It is not known what kind of tissue effect is required to stop the generation or propagation of seizures. Some groups have used low doses (i.e., 10–20 Gy) where few if any histologic changes would be expected. Others have used doses as high as 100 Gy that cause target necrosis and regional brain edema [48]. A typical amygdalohippocampal radiosurgery maximum dose to a volume less than 7.5 ml is 40–50 Gy. If focal hippocampal (or any other brain tissue) irradiation can eliminate seizures without the need for complete tissue destruction, then radiosurgery may become an important therapy for patients with intractable epilepsy. At the same time we await improvements in tools for localization of the seizure focus.
1199
1200
73
Radiosurgery for functional neurosurgery
Functional Imaging and Radiosurgery Improvements in functional imaging eventually will impact on radiosurgery. Functional magnetic resonance imaging to localize cortical function prior to radiosurgery have been evaluated in pilot studies. Localization of hand motor function, leg motor function, and speech areas surrounding arteriovenous malformations and brain tumors before radiosurgery has assisted dose planning [49]. This information can be used to restrict the radiosurgery dose away from functional areas. Advancements in functional imaging may improve the localization of epileptic foci, and perhaps even regions of excitation in the basal ganglia. Magnetoencephalography is an exciting tool to identify functional activation. An ability to identify hyperactivity in deep brain structures would be valuable. With further improvements in neuroimaging and non-invasive physiologic studies, the future will see a significant linkage between functional brain disorders and stereotactic radiosurgery.
References 1. Leksell L. The stereotaxic method and radiosurgery of the brain. Acta Chir Scand 1951;102:316-19. 2. Lindquist C, Kihlstrom L, Hellstrand de. Functional neurosurgery-a future for the gamma knife? Stereotact Funct Neurosurg 1991;57:72-81. 3. Young RF, Vermeulen S, Posewitz A, Grimm P, Blasko J, Jacques D, Rand R, Copcutt B. Functional neurosurgery with the Leksell gamma knife. Radiosurgery 1996;1:218-25. 4. Leksell L. Stereotaxic radiosurgery in trigeminal neuralgia. Acta Chir Scand 1971;137:311-14. 5. Larsson B, Leksell L, Rexed B, Surander P, Mair W, Andersson B. The high-energy proton beam as a neurosurgical tool. Nature 1958;182:1222-3. 6. Leksell L. Cerebral radiosurgery I. Gammathalamotomy in two cases of intractable pain. Acta Chir Scand 1968;134:585-95. 7. Steiner L, Forster D, Leksell L, Meyerson B, Boethius J. Gammathalamotomy in intractable pain. Acta Neurochir 1980;52:173-84.
8. Leksell L, Backlund EO. Stereotactic gammacapsulotomy. In: Hitchcock ER, Ballantine HT, Meyerson BA, editors. Modern concepts in psychiatric surgery. Elsevier; Amsterdam: 1979. p. 213-16. 9. Rylander G. Stereotactic radiosurgery in anxiety and obsessive-compulsive states: Psychiatric aspects. In Hitchcock ER, Ballantine HT, Meyerson BA, editors. Modern concepts in psychiatric surgery. Amsterdam: Elsevier; 1979. p. 235-40. 10. Andersson B, Larsson B, Leksell L, Mair W, Rexed B, Sourander P, Wennerstrand J. Histopathology of late local radiolesions in the goat brain. Acta Radiologica 1970;9:385-94. 11. Rexed B, Mair W, Sourander P, Larsson B, Leksell L. Effect of high energy protons on the brain of the rabbit. Acta Radiologica 1960;53:289-99. 12. Kondziolka D, Lunsford LD, Claassen D, Maitz A, Flickinger J. Radiobiology of radiosurgery. Part I: the normal rat brain model. Neurosurgery 1992;31:271-9. 13. Leksell L, Herner T, Leksell D, Persson B, Lindquist C. Visualization of stereotactic radiolesions by nuclear magnetic resonance. J Neurol Neurosurg Psych 1985;48:19-20. 14. Friehs G, Noren G, Ohye C, Duma C, Mark R, Plombon J, Young RF. Lesion size following Gamma Knife treatment for functional disorders. Stereotact Funct Neurosurg 1996;66 Suppl 1:320-8. 15. Kondziolka D, Linskey ME, Lunsford LD. Animal models in radiosurgery. In: Alexander E, Loeffler JS, Lunsford LD, editors. Stereotactic radiosurgery. New York: McGraw-Hill; 1993. p. 51-64. 16. Lunsford LD, Altschuler EM, Flickinger JC, Wu A, Martinez AJ. In vivo biological effects of stereotactic radiosurgery: a primate model. Neurosurgery 1990; 27:373-82. 17. Kihlstrom L, Guo WY, Lindquist C, Mindus P. Radiobiology of radiosurgery for refractory anxiety disorders. Neurosurgery 1995;36:294-302. 18. Duma C, Jacques D, Kopyov O, Mark R, Copcutt B, Farokhi HK. Gamma knife radiosurgery for thalamotomy in Parkinsonian tremor: a five-year experience. Neurosurg Focus 1997;2(3). 19. Kondziolka D, Lunsford LD, Flickinger JC, Young R, Vermeulen S, Duma C, Jacques D, Rand RW, Regis J, Peragut JC, Manera L, Epstein M, Lindquist C. Stereotactic radiosurgery for trigeminal neuralgia: a multiinstitution study using the gamma unit. J Neurosurg 1996;84:940-5. 20. Young RF, Shumway-Cook A, Vermeulen S, Grimm P, Blasko J, Posewitz A. Gamma knife radiosurgery as a lesioning technique in movement disorder surgery. Neurosurg Focus 1997;2(3):e11. 21. Kondziolka D, Dempsey PK, Lunsford LD, Kestle J, Dolan E, Kanal E, Tasker RR. A comparison between magnetic resonance imaging and computed tomography
Radiosurgery for functional neurosurgery
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
for stereotactic coordinate determination. Neurosurgery 1992;30:402-7. Hirato M, Ohye C, Shibazaki T, Nakamura M, Inoue H, Andou Y. Gamma knife thalamotomy for the treatment of functional disorders. Stereotact Funct Neurosurg 1995;64 Suppl 1:164-71. Otsuki T, Jokura H, Takahashi K, Ishikawa S, Yoshimoto T, Kimura M, Yoshida R, Miyazawa T. Stereotactic gamma-thalamotomy with a computerized brain atlas: Technical case report. Neurosurgery 1994;35:764-8. Pan L, Dai J, Wang BJ, Xu W, Zhou L, Chen XR. Stereotactic gamma thalamotomy for the treatment of Parkinsonism. Stereotact Funct Neurosurg 1996;66 Suppl 1:329-32. Friehs G, Ojakangas CL, Pachatz P, Schrottner O, Ott E, Pendl G. Thalamotomy and caudatomy with the Gamma Knife as a treatment for Parkinsonism with a comment on lesion sizes. Stereotact Funct Neurosurg 1995;64 Suppl 1:209-21. Kondziolka D, Ong J, Lee JYK, Moore R, Flickinger J, Lunsford LD. Gamma knife thalamotomy for essential tremor. J Neurosurg 2008;108:111-17. Kondziolka D, Ong J, Lee JYK, Moore R, Flickinger J, Lunsford LD. Gamma knife thalamotomy for essential tremor. J Neurosurg 2008;108:111-17. Rand RW, Jacques DB, Melbye RW, Copcutt B, Fisher M, Levenick M. Gamma knife thalamotomy and pallidotomy in patients with movement disorders: preliminary results. Stereotact Funct Neurosurg 1993;61 Suppl:65-92. Friedman J, Epstein M, Sanes J, Lieberman P, Cullen K, Lindquist C, Daamen M. Gamma Knife pallidotomy in advanced Parkinson’s disease. Ann Neurol 1996; 39:535-8. Kwon Y, Whang CJ. Stereotactic Gamma knife radiosurgery for the treatment of dystonia.2. Stereotact Funct Neurosurg 1995;64 Suppl 1:222-7. Young RF Jacques DB Rand RW, Copcutt B. Medial thalamotomy with the Leksell gamma knife for treatment of chronic pain. Acta Neurochir (Suppl) 1994;62:105-10. Hayashi M, Taira T, Chernov M, Fukuoka S, Liscak R, Yu CP, Ho RT, Regis J, Katayama Y, Kawakami Y, Hori T. Gamma knife surgery for cancer pain-pituitary glandstalk ablation: a multicenter prospective protocol since 2002. J Neurosurg 2002;97:433-7. Hayashi M, Taira T, Chernov M, Izawa M, Liscak R, Yu CP, Ho RT, Katayama Y, Kouyama N, Kawakami Y, Hori T, Takakura K. Role of pituitary radiosurgery for the management of intractable pain and potential future applications. Stereotact Funct Neurosurg 2003;81:75-83. Leksell L, Herner T, Liden K. Stereotaxic radiosurgery of the brain. Report of a case. Kungl Fysiogr Sallsk Lund Forhandl 1955;25:1-10.
73
35. Alexander E, Lindquist C. Special indications: radiosurgery for functional neurosurgery and epilepsy. In: Alexander E, Loeffler JS, Lunsford LD, editors. Stereotactic radiosurgery. New York: McGraw-Hill; 1993. p. 221-5. 36. Kihlstrom L, Hindmarsh T, Lax I, Lippitz B, Mindus P, Lindquist C. Radiosurgical lesions in the normal human brain 17 years after Gamma knife capsulotomy. Neurosurgery 1997;41:396-402. 37. Ruck C, Karlsson A, Steele JD, Edman G, Meyerson B, Ericson K, Nyman H, Asberg M, Svanborg P. Capsulotomy for obsessive-compulsive disorder. Long-term followup of 25 patients. Arch Gen Psych 2008;65:914-22. 38. Kondziolka D, Hudak R. Management of obsessivecompulsive disorder-related skin picking with gamma knife radiosurgical anterior capsulotomies: a case report. J Clin Psychiatry 2008;69:1337-40. 39. Barcia-Salorio JL, Barcia JA, Hernandez G, Lopez Gomez L. Radiosurgery of epilepsy. Long-term results. Acta Neurochir (Suppl) 1994;62:111-13. 40. Hellstrand DE, Abraham-Fuchs K, Jernberg B, Kihlstrom L, Knutsson E, Lindquist C, Schneider S, Wirth A. MEG localization of interictal epileptic focal activity and concomitant stereotactic radiosurgery. A non-invasive approach for patient with focal epilepsy. Physiol Meas 1993;14:131-6. 41. Maesawa S, Kondziolka D, Dixon E, Balzer J, Fellows W, Lunsford LD. Subnecrotic stereotactic radiosurgery controlling epilepsy produced by kainic acid injection in rats. J Neurosurg 2000;93:1033-40. 42. Mori Y, Kondziolka D, Balzer J, Fellows W, Flickinger JC, Lunsford LD, Thulborn KR. Effects of stereotactic radiosurgery on an animal model of hippocampal epilepsy. Neurosurgery 2000;46:157-68. 43. Mathieu D, Kondziolka D, Niranjan A, Lunsford LD, Flickinger JC. Gamma knife radiosurgery for epilepsy caused by hypothalamic hamartomas. Stereotact Funct Neurosurg 2006;84:82-7. 44. Re´gis J, Peragut JC, Rey M, Samson Y, Levrier O, Porcheron D, Regis H, Sedan R. First selective amygdalohippocampal radiosurgery for mesial temporal lobe epilepsy. Stereotact Funct Neurosurg 1995;64 Suppl 1: 193-201. 45. Regis J, Rey M, Bartolomei F, Vladyka V, Liscak R, Schrottner O, Pendl G. Gamma knife surgery in mesial temporal lobe epilepsy: a prospective multicenter study. Epilepsia 2004;45:504-15. 46. Bartolomei F, Hayashi M, Tamura M, Rey M, Fischer C, Chauvel P, Regis J. Long-term efficacy of gamma knife radiosurgery in mesial temporal lobe epilepsy. Neurology 2008;70:1658-63. 47. Mathieu D, Kondziolka D, Niranjan A, Lunsford LD, Flickinger JC. Gamma knife radiosurgery for epilepsy
1201
1202
73
Radiosurgery for functional neurosurgery
casued by hypothalamic hamartomas. Stereotact Funct Neurosug 2006;84:82-87. 48. Whang CJ, Kim CJ. Short-term follow-up of stereotactic Gamma Knife radiosurgery in epilepsy. Stereotact Funct Neurosurg 1995;64 Suppl 1:202-8.
49. Witt TC, Kondziolka D, Baumann S, Noll D, Small S, Lunsford LD. Pre-operative cortical localization with functional MRI for use in stereotactic radiosurgery. Stereotact Funct Neurosurg 1996;66:24-9.
70 Radiosurgery for Metastases M. Maarouf . C. Bu¨hrle . M. Kocher . V. Sturm
Introduction Brain metastases are the most common intracranial neoplasms in adults. Between 20 and 40% of patients with systemic cancer develop brain metastasis over the course of their disease [1–4]. Certainly, the recognized incidence of brain metastasis has risen over the past 30 years, partly as a result of the advent of MRI, and partly as tumor patients have longer survival owing to more effective systemic therapies. Metastatic brain tumors derived from lung cancer are the most common type, making up 40–60% of the total, followed by those derived from breast cancer (15–20%) and melanoma (10–20%). Colorectal and renal cell carcinomas account for 5–10% each and unknown primary tumor for 15% of patients [5,6]. The prognosis of patients with brain metastases is generally poor. Without any treatment the median survival is 4 weeks [7]. The addition of steroids to reduce the edema induced by brain metastases improves this to 2–3 months [8]. External-beam radiotherapy applied to the whole brain further extends median survival to 4–6 months [8,9]. The choice of the therapeutic strategy for patients with intracranial metastases must be made not only with regard to expected survival times but also in terms of the quality of the remaining life span. The quality of life of these patients is nearly always significantly impaired by progressing focal neurological and neuropsychiatric symptoms, symptoms arising from raised intracranial pressure, symptoms caused by extracranial systemic disease, and/or psychological problems. Prolonged hospitalization #
Springer-Verlag Berlin/Heidelberg 2009
associated with exhaustive treatment (surgery plus radiation therapy) can further impair the quality of life. The prognosis for patients with brain metastases is associated strongly with their recursive partitioning analysis (RPA) class. Using data from three Radiation Therapy Oncology Group (RTOG) brain metastases trials involving more than 1,200 patients treated by WBRT, RPA was performed to define prognostic factors [8]. The Karnofsky performance status (KPS), age, primary tumor status (controlled vs. uncontrolled), and extracranial metastases were the most relevant prognostic factors. On the basis of these factors, 3 prognostic classes were defined: RPA Class 1 (KPS 70%; age <65 years; controlled primary tumor; and no extracranial metastases), RPA Class 2 (KPS 70%; age 65 years, and/or uncontrolled primary tumor, and/or extracranial metastases), and RPA Class 3 (all patients with KPS < 70%). The median survival for patients in RPA Classes 1, 2, and 3 was 7.1, 4.2, and 2.3 months, respectively [8]. This analysis provides a framework that allows grouping patients and analyzing treatment effects in a nonrandomized manner. The appropriate treatment may vary with RPA class. The current principal options for the treatment of cerebral metastasis include whole-brain radiotherapy (WBRT), surgical resection and stereotactic radiosurgery (SRS). Yet, the management of this malignancy is still controversial and quite complex. Whole-brain radiotherapy is the standard treatment in terms of palliation, in particular due to the fact that the majority of patients have multiple brain metastases. Whole-brain
1140
70
Radiosurgery for metastases
radiotherapy is often given after surgery to reduce local recurrence and eliminate micrometastases. The conventional dose of such radiotherapy is 30 Gy in 10 fractions; larger fractions result in an increased risk of neurotoxicity. However, this regimen does not eliminate the longterm cognitive effects of whole brain irradiation. Given the longer survival due to aggressive oncologic treatment of various cancers with a propensity for brain metastasis, this delayed toxicity has been seen more often in recent years, and is difficult or impossible to reverse. Patients with single brain metastasis account for 30–50% and represent a subgroup of patients with a generally better prognosis [3,10]. There are three prospective randomized trials [10–12] which have assessed the value of surgical removal of single brain metastases and have compared surgery plus subsequent radiotherapy with radiotherapy alone (> Table 70-1). In the first prospective randomized trial, published 1990 by Patchell et al. [10], 48 patients with known systemic cancer were treated with either biopsy of the suspected brain metastasis plus WBRT or complete surgical resection of the metastasis plus WBRT. The radiation doses were the same in both groups and consisted of a total dose of 36 Gy given as 12 daily fractions of 300 cGy each. There was a statistically significant increase in survival in the surgical group (9.2 vs.
3.5 months). In addition, the time to recurrence of brain metastases, freedom from death due to neurologic causes and duration of functional independence were significantly longer in the surgical resection group. The one month mortality was 4% in each group, indicating that there was no extra mortality from surgery. While the results obtained by Vecht et al. [12] are similar to those of the group of Patchell and confirm a significant benefit of surgery on overall survival and on quality of life, the results of Mintz et al. [11] were divergent with respect to benefit of surgery on overall survival (> Table 70-1). In the study of Mintz the benefit of surgery may be lost in patients with poor prognostic factors such as advanced extracranial disease or lower performance status. Of the patients in the study by Mintz et al. [11], 45% had extracranial metastases; in the trials by Patchell et al. [10] and by Vecht et al. [12] this number was only 37.5 and 31.7%, respectively. In the trial by Mintz et al. [11], 21% of patients had a KPS of <70, but patients in the trials by Patchell et al. and Vecht et al. had performance scores equivalent to a KPS of 70%. Stereotactically guided brachytherapy is an alternative treatment for good prognosis patients (PRA class I) with single large (>3.5 cm in diameter) and unresectable metastasis as primary treatment modality or salvage therapy for
. Table 70-1 Randomized trials of surgery plus WBRT as compared with WBRT alone Variables
Patchell et al. [10]
Vecht et al. [12]
Mintz et al. [11]
Treatment Patients (n) Eligibility criteria Steroids Median survival (months) Local recurence (%) Median functionally independent survival (months) CNS death (%)
WBRT Surg. + WBRT 23 25 KPS 70, age 18 All 3.5 9.2; p < 0.1 52 20; p < 0.02 1.8 8.8; p < 0.005
WBRT Surg. + WBRT 31 32 WHO PS 2, age 18 Most 6 10; p < 0.04 NR 3.5 7.5; p < 0.06
WBRT Surg. + WBRT 43 41 KPS 50, age < 80 All 6.3 5.6;p < 0.24 NR NR
50
33
63
29
35
46
WBRT = Whole-brain radiation therapy; Surg. = surgery; KPS = Karnofsky performance status; WHO PS = World Health Organisation performance status; NR = not reported
Radiosurgery for metastases
recurrent tumor after failed surgery or fractionated radiotherapy or radiosurgery. Our results (only available in abstract form) and the report of Ostertag and Kreth [13] demonstrate a high rate of local tumor remission with low radiation toxicity. Chemotherapy with newer drugs, such as oral temozolomide, is used occasionally as salvage therapy in those patients with recurrent non-germ cell and non-small cell metastases who, for a variety of reasons, cannot be treated surgically or radiosurgically. In most patients, however, chemotherapy yields little improvement in survival or quality of life and is usually a last resort after other therapies have failed [14].
Radiosurgery Compared to brain gliomas the intracranial metastases are usually more or less spherical, relatively small, and typically have discrete borders that are clearly visualized with cCT and MRI. On the basis of these facts, stereotactic radiosurgery is an appropriate treatment option for patients harboring single or few (up to 3–4) brain metastases. The goal of radiosurgery for the treatment of intracranial metastases is the complete destruction of tumor or at least a local control and improvement or complete remission of the tumor-related symptoms. Since the first published report on the efficacy of radiosurgery for brain metastases by Sturm et al. 1987 [15], ten thousands of patients have been treated worldwide either by Gamma Knife- or LINAC radiosurgery. Currently, radiosurgery is a widely accepted treatment modality for brain metastases. When used as initial therapy, radiosurgery is applied either as a monotherapy or as a boost with whole-brain radiotherapy. Radiosurgery has also been used as salvage treatment for progressive intracranial disease after whole-brain radiotherapy.
70
Numerous retrospective studies [16–21] have shown improved local control rates after radiosurgery (73–94%) when compared to whole brain radiotherapy (WBRT) alone (55– 60%) [9], and most of these studies also reported an increase in median survival from 3–6 months (WBRT) to 7–12 months (radiosurgery) for selected cases [14,17–24]. Two representative cases with complete or partial remission are shown in > Figures 70-1– > 70-2. For single metastases with a maximum diameter of approximately 3 cm, the optimum therapeutic dose is 20–24 Gy (radiosurgery alone) [9,23,25]. The incidence of side effects (symptomatic adjacent edema and/or radiation necrosis) is generally low, ranging between 2 and 5% in most series. A possible reason for the low incidence of radiation-induced side effects, in particular radiation necrosis, is that brain metastases are usually small and spherical, thus being more easily treatable with a conformal field than, e. g., in complex configurated arteriovenous malformation or skull base tumor. Voges et al. [26] have shown that the volume of perilesional brain tissue receiving 10 Gy and more is a major risk factor for radiation necrosis. The incidence of symptomatic radiation necrosis increases exponentially, if the volume of perilesional tissue receiving 10 Gy and more exceeds 10 cc. The better the dose conformation to the target volume, the smaller is the exposure of the surrounding normal brain tissue to radiation and to the risk of radiation necrosis.
Radiosurgery Versus Surgery and Whole Brain Radiotherapy For solitary metastasis, there are one prospective randomized study [27] and three retrospective trials comparing radiosurgery with surgery plus WBRT [28–30]. In the randomized study by Muacevic et al. [27] 64 patients harboring a single, resectable metastasis 3 cm in diameter with a Karnofsky performance score (KPS) 70, and
1141
1142
70
Radiosurgery for metastases
. Figure 70-1 MR images of a 61-year-old female with a single lung cancer (NSCC) metastasis in the insular area left-sided treated with SRS (target surface dose, 20 Gy). The images (left image before, right image 3 years after SRS) illustrate a complete tumor remission
. Figure 70-2 MR images of a 55-year-old patient with 2 cerebellar metastases of mamma carcinoma, left image before SRS (target surface dose, 17 Gy) and right image 21 months thereafter with a significant size reduction of both tumors (partial remission)
stable systemic disease were randomly assigned to microsurgery plus WBRTor Gamma Knife radiosurgery alone. Both treatment groups (31 SRS alone and 33 surgery and WBRT) did not differ in terms of age, KPS, tumor size, tumor location, site of the primary tumor, RPA classification, and
time to metastasis. Length of survival did not differ between the treatment groups: median survival was 9.5 months after surgery plus WBRT and 10.3 months after radiosurgery alone. The 1-year local tumor control rate was statistically not significant: 82% in the surgery group and
Radiosurgery for metastases
96.8% in the radiosurgery group. Patients of the radiosurgery group more often experienced distant recurrences (1-year distant recurrence rate: 25.8% vs. 3%); after adjustment for the effects of salvage radiosurgery this difference was lost. The retrospective study by Muacevic et al. [28] reviewed 108 patients with a single metastasis no larger than 3.5 cm in diameter and stable systemic disease who received SRS alone or surgery plus WBRT. Patients in the SRS group had significantly smaller tumors than did the patients in the surgery plus WBRT group (mean size: 2.07 vs. 2.7 cm). The SRS group also contained a higher proportion of patients with melanoma. Although median survival was 15.7 months in the surgery plus WBRT group and 8.1 months in the SRS group, the survival difference was not statistically significant. No significant differences in local control or complications were observed between the groups, but a higher incidence of distant recurrences was reported in the SRS group. According to these results the authors concluded that radiosurgery provided local tumor control rates as good as resection and WBRT in selected patients. The review by Scho¨ggl et al. [29] retrospectively matched 133 patients who received WBRT and either Gamma Knife SRS or surgery for the treatment of a single brain metastasis less than 3 cm in diameter. Median survival and 1-year overall survival did not differ significantly between the groups; however, the authors reported that SRS was superior in terms of local control and treatment-related morbidity. To be included in the review by O’Neill et al. [30], patients had to be candidates for both SRS and surgical resection. Tumor size had to be no larger than 3.5 cm in diameter, and patients with deep-seated tumors or ventricular obstruction were excluded. These inclusion criteria were met by 23 patients who had received SRS and 74 patients who had received surgery, most of whom had additionally been treated with WBRT. In the SRS group significantly fewer patients had
70
a good performance score. No significant differences in survival or cause of death were detected between the groups, and the authors concluded that neither SRS nor surgical resection was superior in that study.
Radiosurgery as Boost with Whole Brain Radiotherapy (WBRT) Versus WBRT Alone There have been three randomized trials [16,31,32] examining the use of whole-brain radiotherapy plus radiosurgery boost as compared with whole-brain radiotherapy alone in selected patients with brain metastases. Kondziolka et al. [32] reported the first trial on the subject. Patients with two to four brain metastases all 25 mm were randomized to whole-brain radiotherapy alone (30 Gy in 12 fractions) or whole-brain radiotherapy plus radiosurgery. The study was stopped at 60% accrual. Only 27 patients were randomized and the results were reported for 14 patients treated with whole-brain radiotherapy alone and 13 patients treated with whole-brain radiotherapy and additional radiosurgery boost. The two groups were well balanced with respect to age, sex, tumor type, number of tumors, and extent of extracranial disease. The largest randomized study to date has been reported by Andrews et al. [31], which randomized 164 patients to whole-brain radiotherapy and radiosurgery boost versus 167 patients randomized to whole brain radiotherapy alone. Patients with one to three newly diagnosed brain metastases were included. The brain metastases could have a maximum diameter of 4 cm for the largest lesion and the additional lesions could not exceed 3 cm (based on contrast-enhanced magnetic resonance imaging). The arms of the trial were well balanced for baseline characteristics known to affect survival such as age, Karnofsky performance status
1143
1144
70
Radiosurgery for metastases
(KPS) and status of extracranial metastases. The study was powered to establish possible differences in outcome and stratified for number of brain metastases (one vs. two to three) and extent of extracranial disease (none vs. present). The Chougule et al. trial [16] was published in abstract form. Patients with one to three brain metastases with tumor volume 30 cc and minimum life expectancy of 3 months were included. The randomization arms were Gamma Knife radiosurgery alone (30 Gy to the tumor margin), whole brain radiotherapy (30 Gy in 10 fractions) plus gamma knife boost (20 Gy to the tumor margin), and whole-brain radiotherapy alone (30 Gy in 10 fractions). Overall median survival was 7, 5, and 9 months for the arms, respectively (not significant). All three randomized trials [16,31,32] revealed an improvement in local brain tumor control in patients treated with whole-brain radiotherapy and radiosurgery boost as compared with whole-brain radiotherapy alone. But none of the three trials reports a statistically significant improvement in overall survival in the radiosurgery boost arms as compared with whole-brain radiotherapy alone. Median survival for the whole-brain radiotherapy alone arm ranged from 5.7 to 9 months. Median survival for the whole-brain radiotherapy and radiosurgery boost arms ranged from 5 to 11 months. However, the RTOG study, Andrews et al. [31], demonstrated that radiosurgery is beneficial in patients with single metastasis. The rates of local control and overall survival with radiosurgery and WBRT are comparable to the rates achieved in patients treated with conventional surgery plus WBRT. In contrast, other reports have questioned the usefulness of WBRT for patients having radiosurgery [14,33]. It has been argued that if patients develop additional brain tumors following radiosurgery, the procedure can be repeated to minimize the risk for delayed cognitive decline frequently noted after WBRT [34].
Radiosurgery or Whole Brain Radiotherapy The randomized trial by Chougule et al. [16] investigated this subject and was only reported in abstract form. The authors randomized patients to gamma knife radiosurgery alone versus whole-brain radiotherapy and gamma knife radiosurgery versus whole-brain radiotherapy alone (> Table 70-2). The local tumor control rate was higher in the two radiosurgery arms (87% for gamma knife radiosurgery alone and 91% for gamma knife radiosurgery and wholebrain radiotherapy) than in the whole-brain radiotherapy arm (62%). Median survival was 7, 5, and 9 months for gamma knife radiosurgery alone versus whole-brain radiotherapy and Gamma Knife radiosurgery versus whole-brain radiotherapy, respectively. Survival was not different among the three arms. Hasegawa et al. suggested that brain metastases are well controlled with SRS alone and that WBRT, in addition to SRS, may be omitted to reduce the risk of radiation-related toxicity [40]. Those authors suggested that SRS alone was a reasonable alternative to WBRT plus SRS. Sneed et al. suggested in two retrospective studies that the omission of WBRT in the initial management of patients who undergo radiosurgery does not appear to compromise survival [14,38]. However, the studies by Sneed et al. and Hasegawa et al. did not compare SRS alone with WBRT alone. Rades et al. [35] compared SRS alone with WBRT alone. They investigated whether stereotactic radiosurgery alone improved outcomes for patients in recursive partitioning analysis (RPA) Classes 1 and 2 who had 1–3 brain metastases compared with whole-brain radiotherapy. The authors found that SRS alone is associated with improved entire brain control and local control compared with 30–40 Gy of WBRT alone for patients in RPA Classes 1 and 2 who have 1–3 brain metastases, whereas overall survival and distant
Radiosurgery for metastases
control do not differ significantly. They concluded that SRS alone appears to be more effective than WBRT alone. Furthermore, it is less time consuming: SRS takes only 1 day, which means that the patients have to spend only very little time of their limited life span receiving treatment. For our institution [41], the survival times in patients with inoperable brain metastases treated by radiosurgery alone (117 patients, median age 60 years) were compared with those of a historical collective (138 patients, median age 58 years), treated by WBRT alone. Only patients with one to three metastases were included and treated with LINAC radiosurgery. Details of the radiosurgical procedure have been described before [41–43]. The most frequent primary tumor type in the SRS group was non-small-cell lung cancer (30%), followed by malignant melanoma (27%), renal cell carcinoma (13%), breast carcinoma (12%), and other types (18%). Patients with singular metastases in this group had either deep-seated tumors not suitable for resection or were referred for radiosurgery rather than surgery because of the physician’s or patient’s preference. In the WBRT group, non-small-cell lung cancer (28%) was also dominant, but breast cancer (19%) which is thought to have a slightly better prognosis [44] was more frequent (melanoma 6%, renal cell carcinoma 5%, others 42%). Patients with singular metastases in this group were either unresectable because of localization or size of the lesion. All patients in both groups were classified into the three RPA prognostic classes. In RPA class I patients (Karnofsky performance score 70, primary tumor controlled, no other metastases, age <65 years), radiosurgery resulted in a median survival of 25.4 months (n = 23) which was significantly longer than for WBRT (n = 9, 4.7 months). In RPA class III (Karnofsky performance score <70), no significant difference in survival between radiosurgery (n = 20,
70
4.2 months) and WBRT (n = 68, 2.5 months) was found. In RPA class II (all other patients), radiosurgery produced a small, but significant survival advantage (radiosurgery: n = 74, 5.9 months, WBRT: n = 61, 4.1 months). In the radiosurgery group, local control of the irradiated metastases was equally high in all RPA groups. The development of new outfield metastases was, however, significantly more frequent in RPA groups II and III than in group I, probably because of reseeding from active extracranial tumor. Local control and freedom from new intracerebral metastases could not be evaluated in the WBRT group. In the radiosurgery group, 4/117 patients (3%) developed symptomatic radiation necrosis indicated by CT and/or MRI scans 15 (range, 5–18) months after irradiation, which were treated by corticosteroids. In the WBRT group, radiation necrosis was not diagnosed. In both groups, signs or symptoms of dementia or neuropsychological dysfunction were not observed after WBRT. In conclusion, it seems unequivocal that radiosurgery improves local control of the irradiated metastases when compared to WBRT. The reason for failed survival benefit may be the fact that patients with brain metastases have, in addition to local brain recurrence, competing risks, namely systemic progression, distant intracranial failure, and intercurrent death. Therefore, one can expect a survival advantage as a result of an increased local control rate of the irradiated brain metastases only in patients in good clinical condition with a low systemic tumor burden and in whom intracranial distant relapses can be controlled, e.g., by salvage radiosurgery or/and adjuvant WBRT. Only patients in the RPA class I group fulfill the aforementioned criteria and may, therefore, substantially benefit from radiosurgery. RPA class III patients have a poor prognosis per se which cannot be altered by localized irradiation of the brain. Some of the class II patients may benefit as well.
1145
RS
PRS
PRS
PRS
RS
RS
RS RS RS
RS RS RS
Scho¨ggl et al. [29]
WBRT with or without SRS Andrews et al. [31]
Chougule et al. [16]
Kondziolka et al. [32]
SRS or WBRT Rades et al. [35]
Kocher et al. [27]
SRS with or without WBRT Sturm et al. [36] Lutterbach et al. [37] Sneed et al. [38]
Flickinger et al. [23] Alexander et al. [39] Auchter et al. [22]
SRS SRS SRS SRS + WBRT SRS + Some WBRT SRS + WBRT SRS + WBRT
SRS WBRT SRS WBRT
WBRT WBRT + SRS SRS WBRT + SRS WBRT WBRT WBRT + SRS
SRS Surg + WBRT SRS Surg + most WBRT SRS + WBRT Surg + WBRT
Treatment
30 55 168 175 116 171 122
95 91 117 138
164 167 36 37 31 14 13
31 33 23 74 67 66
Patients (n)
– – – –
3.6 cm – –
–
64 26 For SRS: LC equally high in all RPA classes
Inoperable 3 cm –
4 cm, 1–3 3 cm 1–3
2.5 cm, 2–4
4 cm, 1–3 3 cm, 1–3 71 82 87 91 62 0 92
–
<3
<3.5
96 82 –
1-year local control (%)
3 cm
Metastasis: Diameter (cm), Number (n)
6.5 (mean) 7.7 8.3 8.4 11 10.3 12.9
13 7 RPA I: 25.4 vs. 4.2, RPA II: 5.9 vs. 4.1, RPA III: 4.2 vs. 2.5
5.7 6.5 7 5 9 7.5 11
10.3 9.5 13 16 12 9
Median survival (months)
SRS = stereotactic radiation surgery; WBRT = whole brain radiation therapy; PRS = prospective randomized study; RS = retrospective study; Surg = surgery; RPA = recursive partitioning analysis; LC = local control
RS
O’Neill et al. [30]
SRS versus surgery and WBRT Muacevic et al. [27] PRS
Study type
70
Reference
. Table 70-2 Studies addressing the role of stereotactic radiosurgery (SRS) in the treatment of patients with brain metastases
1146 Radiosurgery for metastases
Radiosurgery for metastases
Radiosurgery Versus Whole Brain Radiotherapy and Radiosurgery Boost Two randomized trials have compared SRS plus WBRT to SRS alone [16,45]; however, numerous retrospective reviews addressed the efficacy of SRS with or without WBRT [22,23,36–39] (> Table 70-2). In the Japanese trial by Aoyama et al. [45], patients with KPS of at least 70, four or fewer brain metastases with diameters of 3 cm or less were randomized to radiosurgery alone (67 patients, mean age, 62.1 years) versus wholebrain radiotherapy and radiosurgery (65 patients, mean age, 62.5 years). Radiosensitive histologies were excluded. Patients were stratified by number of lesions (one vs. two to four), primary tumor (lung vs. other), and extracranial disease (present or absent/stable). The radiosurgery dose was lowered by 25% in the whole-brain radiotherapy arm. Actuarial 1-year brain tumor control rate for the lesions treated with radiosurgery was 73% in the radiosurgery alone arm and 89% in the radiosurgery and whole-brain radiation arm. The median survival time and the 1-year actuarial survival rate were 7.5 months and 38.5% in the WBRT plus SRS group and 8.0 months and 28.4% for SRS alone. Chougule et al. [16] compared radiosurgery alone versus whole-brain radiotherapy and radiosurgery. The local brain control rate was comparably high in both arms (87% for radiosurgery alone and 91% for radiosurgery and whole-brain radiotherapy). Survival was reported as not different among the two arms (7 months for radiosurgery alone versus 5 months for whole-brain radiotherapy and radiosurgery). A subgroup analysis of the largest review by Sneed et al. [38] compared 168 patients with single brain metastasis who received SRS alone with 175 patients who received SRS with WBRT as
70
initial treatment. Overall, patients who received SRS alone included a higher percentage of patients over 65 years and patients with a KPS of <70. A number of patients, particularly those who initially received SRS alone, underwent one or more salvage therapies for recurrent or new metastases. No significant survival difference was detected between the groups (8.3 months for SRS alone vs. 8.4 months for WBRT and SRS). Flickinger et al. [23] reviewed 116 patients with single metastasis treated with linear accelerator SRS. Of those patients, 56% also received fractionated radiation therapy. In this study, 45 patients (39%) had tumors that recurred after previous WBRT, and 71 (61%) were treated with SRS as initial management for their metastasis. The median survival was 11 months, with local tumor control in 85% of patients. Recurrence was documented in 15%. In a multivariate analysis, local tumor control was significantly better in patients receiving both fractionated radiation therapy and SRS as compared with SRS alone, but no effect on survival was observed. Two non-comparative retrospective reviews [22,39] investigated the efficacy of SRS plus WBRT. The study by Auchter et al. [22] retrospectively reviewed 122 patients who matched the eligibility criteria for entry into the randomized trial by Patchell et al. [10] and who had been treated with SRS followed by WBRT. None of those patients had received prior surgery or radiation therapy. Median survival was 12.9 months, and the 1- and 2-year survival rates were 53% and 30% respectively. Complete response was observed in 25% of patients, and partial response in 34%. Local control rates at 1 and 2 years were 85% and 77% respectively. Intracranial recurrence outside the SRS volume was experienced by 22% of patients. Median duration of functionally independent survival, defined as a KPS > 70, was 10.2 months. A second retrospective review by Alexander et al.
1147
1148
70
Radiosurgery for metastases
[39] included 171 patients with single brain metastasis. Most of the patients in that review received SRS to treat recurrent lesions. All patients received WBRT, either as part of their initial therapy or in combination with SRS. Median survival for patients with single brain metastasis was 10.3 months. A small prospective case series of 24 patients who received SRS plus WBRT 12 reported a similar median survival of 10 months and tumor shrinkage in 58% of patients for whom data were available. Two single-arm prospective studies [36,37] investigated the efficacy of SRS alone. The case series by Sturm et al. [36] of 30 patients with inoperable single brain metastasis reported mean survival of 6.5 months, improvement of clinical symptoms in 18 of 27 patients, and tumor regression in 13 of 22 patients. A subgroup analysis of the study by Lutterbach et al. [37] reported median survival of 7.7 months for patients with single brain metastasis. Thus, there is evidence, from the two randomized trials as well as from the retrospective studies, that addition of adjuvant WBRT to radiosurgery seems to increase local control and to decrease the frequency of new intracranial failures, but survival remains unchanged in patients with newly diagnosed brain metastases treated with up-front radiosurgery alone [16,38,46]. In summary, radiosurgery is a highly effective and well-tolerated treatment modality for up to three brain metastases with diameter under 30 mm. Radiosurgery can be used either as a sole therapy or as a boost in addition to whole-brain radiotherapy in newly diagnosed patients as well as a salvage therapy for progressive or recurrent intracranial disease after whole-brain radiotherapy. The results of sole radiosurgery are roughly comparable to those achieved with more aggressive intervention consisting of surgery and whole brain radiotherapy. The main advantages of radiosurgery are non-invasiveness, short hospitalization times (1–2 days), and cost effectiveness.
References 1. Nussbaum ES, Djalilian HR, Cho KH, Hall WA. Brain metastases. Histology, multiplicity, surgery, and survival. Cancer 1996;78(8):1781-8. 2. Posner JB. Management of brain metastases. Rev Neurol 1992;148:477-87. 3. Walker AE, Robins M, Weinfeld FD. Epidemiology of brain tumours: the national survey of intracranial neoplasms. Neurology 1985;35:219-26. 4. Zimm S, Wampler GL, Stablein D, Hazra T, Young HF. Intracerebral metastases in solid-tumor patients: natural history and results of treatment. Cancer 1981;48:384-94. 5. Kihlstrom L, Karlsson B, Lindquist C. Gamma knife surgery for cerebral metastases. Implications for survival based on 16 years experience. Stereotact Funct Neurosurg 1993;61Suppl 1:45-50. 6. Posner JB. Diagnosis and treatment of metastases to the brain. Clin Bull 1974;4:47-57. 7. Markesbery WR, Brooks WH, Gupta GD, Young AB. Treatment for patients with cerebral metastases. Arch Neurol 1978;35(11):754-6. 8. Gaspar L, Scott C, Rotman M, et al. Recursive partitioning analysis (RPA) of prognostic factors in three radiation therapy oncology group (RTOG) brain metastases trials. Int J Radiat Oncol Biol Phys 1997;37:745-51. 9. Nieder C, Nestle U, Walter K, et al. Dose/effect relationship for brain metastases. J Cancer Res Clin Oncol 1998;124:346-50. 10. Patchell RA, Tibbs PA, Walsh JW, Dempsey RJ, Maruyama Y, Kryscio RJ, Markesbery WR, Macdonald JS, Young B. A randomized trial of surgery in the treatment of single metastases to the brain. N Engl J Med 1990;322(8):494-500. 11. Mintz AH, Kestle J, Rathbone MP, Gaspar L, Hugenholtz H, Fisher B, Duncan G, Skingley P, Foster G, Levine M. A randomized trial to assess the efficacy of surgery in addition to radiotherapy in patients with a single cerebral metastasis. Cancer 1996;78(7):1470-6. 12. Vecht CJ, Haaxma-Reiche H, Noordijk EM, Padberg GW, Voormolen JH, Hoekstra FH, Tans JT, Lambooij N, Metsaars JA, Wattendorff AR, et al. Treatment of single brain metastasis: radiotherapy alone or combined with neurosurgery? Ann Neurol 1993;33(6):583-90. 13. Ostertag CB, Kreth FW. Interstitial iodine-125 radiosurgery for cerebral metastases. Br J Neurosurg 1995; 9(5):593-603. 14. Sneed PK, Lamborn KR, Forstner JM, McDermott MW, Chang S, Park E, Gutin PH, Phillips TL, Wara WM, Larson DA. Radiosurgery for brain metastases: is whole brain radiotherapy necessary? Int J Radiat Oncol Biol Phys 1999;43(3):549-58. 15. Sturm V, Kober B, Ho¨ver KH, Schlegel W, Boesecke R, Pastyr O, Hartmann GH, Schabbert S, zum Winkel K, Kunze S, et al. Stereotactic percutaneous single dose irra-
Radiosurgery for metastases
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
diation of brain metastases with a linear accelerator. Int J Radiat Oncol Biol Phys 1987;13(2):279-82. Chougule PB, Burton-Williams M, Saris S, et al. Randomized treatment of brain metastasis with gamma knife radiosurgery, whole brain radiotherapy or both. Int J Radiat Oncol Biol Phys 2000(Astro-Abstracts 2000); 48:114. Herfarth KK, Izwekowa O, Thilmann C, et al. Linacbased radiosurgery of cerebral melanoma metastases. Strahlenther Onkol 2003;179:366-71. Joseph J, Adler JR, Cox RS, et al. Linear accelerator-based stereotactic radiosurgery for brain metastases: the influence of number of lesions on survival. J Clin Oncol 1996;14:1085-92. Kocher M, Voges J, Mu¨ller RP, et al. Linac radiosurgery for patients with a limited number of brain metastases. J Radiosurg 1998;1:9-15. Pirzkall A, Debus J, Lohr F, et al. Radiosurgery alone or in combination with whole-brain radiotherapy for brain metastases. J Clin Oncol 1998;16:3563-9. Young RF, Jacques DB, Duma C, et al. Gamma knife radiosurgery for treatment of multiple brain metastases. Radiosurgery 1996;1:92-101. Auchter RM, Lamon JP, Alexander E, III, et al. A multiinstitutional outcome and prognostic factor analysis of radiosurgery for resectable brain metastases. Int J Radiat Oncol Biol Phys 1996;35:27-35. Flickinger JC, Kondziolka D, Lunsford LD, Coffey RJ, Goodman ML, Shaw EG, Hudgins WR, Weiner R, Harsh GR, IV, Sneed PK, et al. A multi-institutional experience with stereotactic radiosurgery for solitary brain metastasis. Int J Radiat Oncol Biol Phys 1994; 28(4):797-802. Moriarty TM, Loeffler JS, Black PM, et al. Long-term follow-up of patients treated with stereotactic radiosurgery for single or multiple brain metastases. Radiosurgery 1996;1:83-91. Vogelbaum MA, Angelov L, Lee SY, Li L, Barnett GH, Suh JH. Local control of brain metastases by stereotactic radiosurgery in relation to dose to the tumor margin. J Neurosurg 2006;104(6):907-12. Voges J, Treuer H, Sturm V, Bu¨chner C, Lehrke R, Kocher M, Staar S, Kuchta J, Mu¨ller RP. Risk analysis of linear accelerator radiosurgery. Int J Radiat Oncol Biol Phys 19961;36(5):1055-63. Muacevic A, Wowra B, Siefert A, Tonn JC, Steiger HJ, Kreth FW. Microsurgery plus whole brain irradiation versus Gamma Knife surgery alone for treatment of single metastases to the brain: a randomized controlled multicentre phase III trial. J. Neurooncol 2008;87(3): 299-307. Muacevic A, Kreth FW, Horstmann GA, et al. Surgery and radiotherapy compared with GammaKnife radiosurgery in the treatment of solitary cerebral metastases of small diameter. J Neurosurg 1999;91:35-43.
70
29. Scho¨ggl A, Kitz K, Reddy M, et al. Defining the role of stereotactic radiosurgery versus microsurgery in the treatment of single brain metastases. Acta Neurochir (Wien) 2000;142:621-6. 30. O’Neill BP, Iturria NJ, Link MJ, Pollock BE, Ballman KV, O’Fallon JR. A comparison of surgical resection and stereotactic radiosurgery in the treatment of solitary brain metastases. Int J Radiat Oncol Biol Phys 2003; 55:1169-76. 31. Andrews DW, Scott CB, Sperduto PW, Flanders AE, Gaspar LE, Schell MC, Werner-Wasik M, Demas W, Ryu J, Bahary JP, Souhami L, Rotman M, Mehta MP, Curran WJ, Jr. Whole brain radiation therapy with or without stereotactic radiosurgery boost for patients with one to three brain metastases: phase III results of the RTOG 9508 randomised trial. Lancet 2004;363 (9422):1665-72. 32. Kondziolka D, Patel A, Lunsford LD, et al. Stereotactic radiosurgery plus whole brain radiotherapy versus radiotherapy alone for patients with multiple brain metastases. Int J Radiat Oncol Biol Phys 1999;45:427-34. 33. Shaw EG. Radiotherapeutic management of multiple brain metastases: ‘‘3000 in 10’’ whole brain radiation is no longer a ‘‘no brainer’’. Int J Radiat Oncol Biol Phys 1999;45(2):253-4. 34. DeAngelis LM, Delattre JY, Posner JB. Radiationinduced dementia in patients cured of brain metastases. Neurology 1989;39(6):789-96. 35. Rades D, Pluemer A, Veninga T, Hanssens P, Dunst J, Schild SE. Whole-brain radiotherapy versus stereotactic radiosurgery for patients in recursive partitioning analysis classes 1 and 2 with 1 to 3 brain metastases. Cancer 2007;110(10):2285-92. 36. Sturm V, Kimmig B, Engenhardt R, Schlegel W, Pastyr O, Treuer H, Schabbert S, Voges J. Radiosurgical treatment of cerebral metastases. Method, indications and results. Stereotact Funct Neurosurg 1991;57(1–2):7-10. 37. Lutterbach J, Cyron D, Henne K, Ostertag CB. Radiosurgery followed by planned observation in patients with one to three brain metastases. Neurosurgery 2003;52: 1066-74. 38. Sneed PK, Suh JH, Goetsch SJ, et al. A multi-institutional review of radiosurgery alone vs. radiosurgery with whole brain radiotherapy as the initial management of brain metastases. Int J Radiat Oncol Biol Phys 2002;53:519-26. 39. Alexander E, Moriarty TM, Davis RB, et al. Stereotactic radiosurgery for the definitive, noninvasive treatment of brain metastases. J Natl Cancer Inst 1995;87:34-40. 40. Hasegawa T, Kondziolka D, Flickinger JC, Germanwala A, Lunsford LD. Brain metastases treated with radiosurgery alone: an alternative to whole brain radiotherapy? Neurosurgery 2003;52:1318-26. 41. Kocher M, Maarouf M, Bendel M, Voges J, Mu¨ller RP, Sturm V. Linac radiosurgery versus whole brain
1149
1150
70
Radiosurgery for metastases
radiotherapy for brain metastases. A survival comparison based on the RTOG recursive partitioning analysis. Strahlenther Onkol 2004;180(5):263-7. 42. Maarouf M, Treuer H, Kocher M, Voges J, Gierich A, Sturm V. Radiation exposure of extracranial organs at risk during stereotactic linac radiosurgery. Strahlenther Onkol 2005;181(7):463-7. 43. Hartmann GH, Schlegel W, Sturm V, Kober B, Pastyr O, Lorenz WJ. Cerebral radiation surgery using moving field irradiation at a linear accelerator facility. Int J Radiat Oncol Biol Phys 1985;11(6):1185-92. 44. Kocher M, Mu¨ller R-P, Staar S, et al. Long-term survival after brain metastases in breast cancer. Strahlenther Onkol 1995;171:290-5.
45. Aoyama H, Shirato H, Tago M, Nakagawa K, Toyoda T, Hatano K, Kenjyo M, Oya N, Hirota S, Shioura H, Kunieda E, Inomata T, Hayakawa K, Katoh N, Kobashi G. Stereotactic radiosurgery plus whole-brain radiation therapy vs. stereotactic radiosurgery alone for treatment of brain metastases: a randomized controlled trial. JAMA 2006;295(21):2483-91. 46. Chidel MA, Suh JH, Reddy CA, et al. Application of recursive partitioning analysis and evaluation of the use of whole brain radiation among patients treated with stereotactic radiosurgery for newly diagnosed brain metastases. Int J Radiat Oncol Biol Phys 2000;47:993-9.
72 Radiosurgery for Pituitary Tumors J. P. Sheehan . J. Jagannathan . W. J. Elias . E. R. Laws
Abstract Objective: Pituitary adenomas are common neoplasms and represent between 10 and 20% of all primary brain tumors. Historically, the treatment armamentarium for pituitary adenomas included medical management, microsurgery, and fractionated radiotherapy. More recently, radiosurgery has emerged as the adjuvant treatment of choice. This review is intended to assess the efficacy, safety, and role of radiosurgery for treatment of pituitary adenomas. Methods: Medical literature databases were searched for articles pertaining to pituitary adenomas and stereotactic radiosurgery. Each study was evaluated for the number of patients, radiosurgical parameters (e.g. maximal dose, margin dose, type of unit), length of follow-up, tumor growth control rate, complications, and rate of hormonal normalization in the case of functioning adenomas. Results: A total of 53 peer reviewed studies including 2,483 patients were evaluated. Radiosurgery resulted in control of tumor size in more than 90% of treated patients. The reported rates of hormonal normalization for functioning adenomas vary substantially; this range is in part due to widespread differences in endocrinological criteria utilized for post-radiosurgical assessment. Fortunately, the incidence of serious complications following radiosurgery is quite low. Post-treatment hypopituitarism remains a concern, along with infrequent relapses and recurrences. Conclusions: Although microsurgery remains the primary treatment modality in most cases, stereotactic radiosurgery offers both safe and #
Springer-Verlag Berlin/Heidelberg 2009
effective treatment for recurrent or residual pituitary adenomas. In rare instances, radiosurgery may be appropriate initial treatment for patients with pituitary adenomas.
Introduction Pituitary adenomas are very common lesions and represent between 10 and 20% of all primary brain tumors [1,2]. Epidemiological studies have demonstrated that nearly 20% of the general population has a pituitary adenoma [1]. Pituitary adenomas are broadly classified into two groups: functioning (endocrinologically active) and nonfunctioning tumors. The first category of tumor is those that secrete excess amounts of normal pituitary hormones and, consequently, present with a variety of clinical syndromes depending upon hormones secreted. These functioning or secretory adenomas include the following: (1) prolactinomas causing amenorrhea, galactorrhea in women and infertility and impotence in men; (2) growth hormone secreting adenomas causing acromegaly and gigantism; (3) corticotrophin secreting tumors causing Cushing’s disease; (4) corticotrophin secreting adenomas in patients with adrenalectomies that result in Nelson’s syndrome [3–5]. The second category of pituitary adenomas is comprised of tumors that do not secrete any known biologically active pituitary hormones, and these represent approximately 30% of all pituitary tumors [6]. These nonfunctioning gonadotropic or null cell pituitary adenomas progressively enlarge in the pituitary fossa and
1172
72
Radiosurgery for pituitary tumors
can extend beyond the confines of the sella turcica. These tumors may cause symptoms related to mass effect whereby the optic nerves and chiasm are compressed, and a bitemporal visual field loss characteristically results. Patients with nonfunctioning adenomas often develop hypopituitarism as a result of compression of the normal functioning pituitary gland. For both types of pituitary adenomas, tumor invasion into surrounding structures (e.g. dural or cavernous sinus invation) or incomplete tumor resection is quite common [7,8]. Long-term tumor remission rates after microsurgery alone vary from 50 to 90% [1,2,9–11]. Radiation therapy or radiosurgery can be administered post-operatively as adjuvant therapy to inhibit recurrent growth or, later, when clinical symptoms or radiographic signs indicate recurrence. They may also be utilized post-operatively to treat known residual tumor following incomplete resection. In 1951, stereotactic radiosurgery was described by Lars Leksell as the ‘‘closed skull destruction of an intracranial target using ionizing radiation’’ [12]. In 1968, Leksell treated the first pituitary adenoma patient with the Gamma knife. Since that time, stereotactic radiosurgery has been utilized in thousands of patients to control tumor growth and normalize hormonal production from pituitary adenomas. At the same time, great attention and effort in the field of stereotactic radiosurgery have been placed on the preservation of surrounding neuronal, vascular, and hormonal structures. In this chapter, we review the role of stereotactic radiosurgery for pituitary adenomas.
Radiosurgical Techniques Radiosurgery is performed using the Gamma knife, a modified linear accelerator based system, or proton beams produced by cyclotrons. Gamma knife radiosurgery usually involves multiple isocenters
of different beam diameter to achieve a dose plan that conforms to the irregular three-dimensional volumes of most mass lesions. The total number of isocenters may vary depending upon the size, shape, and number of lesions. The Gamma Knife Model C and 4C combined advances in dose planning with robotic engineering and obviates the need to set coordinates manually for each isocenter [13]. The Gamma Knife Perfexion utilizes automated positioning and adds the ability for composite shots whereby different beam diameters (0, 4, 8, and 16 mm) can be used simultaneously (> Figure 72-3). In LINAC-based radiosurgery (e.g. Cyberknife, Novalis, Trilogy, Synergy), multiple radiation arcs are utilized to crossfire photon beams at a target defined in stereotactic space [14]. Most of the presently functioning systems use nondynamic techniques in which the patient couch is set at an angle and the arc is moved around its radius to deliver radiation that enters the skull through many different points. Numerous techniques have been developed to enhance conformity of dose planning and delivery using LINAC-based systems. These include beam shaping and intensity modulation. Newer developments include the introduction of jaws, noncircular, and mini- and micro-leaf collimators. The conformal beam can be delivered with the micromultileaf collimator or conformal blocks. Proton beam radiosurgery takes advantage of the intrinsic superior dose distribution of protons versus photons because of the Bragg-peak at treatment depth [15]. These facilities are only available at a limited number of centers due to financial and logistical constraints. In preparation for radiosurgery, many centers have recommended a temporary cessation of antisecretory medications in the perioperative time period. In 2000, Landolt et al. first reported a significantly lower hormone normalization rate in acromegalic patients who were receiving antisecretory medications at the time of radiosurgery [16]. Since then, this same group as well as others
Radiosurgery for pituitary tumors
have documented a counterproductive effect of antisecretory medications on the rate of hormonal normalization following radiosurgery [17,18]. The mechanism by which antisecretory medications lower hormonal normalization rates is unknown, but Landolt et al. have hypothesized that these drugs alter cell cycling and thus potentially decrease tumor cell radiosensitivity [16,17]. Moreover, the optimal time period to hold antisecretory medications in conjunction with stereotactic radiosurgery is not clear. Depending upon the antisecretory agents’ pharmacokinetics, 2–4 months cessation of the agent prior to radiosurgery seems prudent for most cases [16,17,19]. The effective delivery of radiation to a target requires clear and accurate imaging of that target. Over the past 20 years, significant advances have increased the efficacy and safety of radiosurgical treatment of pituitary lesions. Tumor localization for dose planning is better achieved with enhanced coronal MR than with CT imaging [20]. An MRI sequence consisting of post-contrast, thin-slice (e.g. 1 mm) volume acquisition is typically utilized to define the tumor within the sellar region. In patients with previous surgery, fat suppression techniques can prove useful for differentiating tumor from surgical fat grafts. In the pre-MRI era, CT was utilized routinely. However, now it is generally reserved for patients who cannot undergo an MRI (e.g. a patient with a pacemaker). PET imaging may also be used to define the location of a functioning adenoma for radiosurgical targeting [21]. For hormonally active lesions, if the tumor is unable to be localized on imaging studies, radiosurgery may still be successful in achieving hormonal normalization. In these cases, the entire sellar region (including the inferior and cavernous sinus dura) is selected as the radiosurgical target when no definitive tumor can be visualized [5,22,23]. After frame placement or other kinds of target immobilization and stereotactic image acquisition, dose planning is performed. Through the strategic selection of isocenters, gamma
72
angle, prescription dose, beam blocking patterns, and isodose selection, the borders of the tumor can be encompassed and a suitable radiation dose delivered. The radiosurgical team must take into account the radiation fall off characteristics unique to the type of unit utilized. Moreover, if fractionated radiation therapy has previously been administered, the dose and timing of that treatment must be considered when selecting a new dose plan for radiosurgery.
Methods for Review of the Literature PreMedline, Medline, Cochrane, and Pubmed databases were searched for articles pertaining to pituitary adenomas and stereotactic radiosurgery. Studies had to be published in peer-reviewed journals to be included in this review. Each study was evaluated for the number of patients, type of radiosurgical unit, radiosurgical parameters (e.g. maximal dose, tumor margin dose), length of follow-up, tumor growth control rate, complications, and rate of hormonal normalization in the case of functioning adenomas. Results of this review are detailed in > Tables 72-1–72-5. A total of 53 published studies including 2,483 patients were reviewed [2,4,16–18,24–39,46–49,51,52,54,59–63, 71,73,74,76,77].
Radiosurgical Goals and Expectations Based upon the Literature For patients with pituitary adenomas, radiosurgery is designed to inactivate the tumor cells thereby preventing tumor growth and, for functioning adenomas, normalizing hormonal overproduction as well as controlling tumor growth. Ideally, these goals are met without damaging the residual normal pituitary gland and surrounding vascular and neural structures.
1173
1174
72
Radiosurgery for pituitary tumors
Moreover, radiosurgery should be carried out in such a way so as to avoid radiation associated secondary tumor formation.
Radiosurgery for Nonfunctioning (Nonsecretory) Adenomas Control of adenoma growth is typically the goal for radiosurgery in patients with nonsecretory adenomas. Most series define tumor control as either an unchanged or decreased volume on follow-up radiological imaging studies. In nearly all published series, stereotactic radiosurgery afforded excellent control of tumor growth (> Table 72-1) [24–39]. Most studies reported a greater than 90% control of tumor size (range
89.8–100%). A weighted average adenoma control rate for all published nonsecretory adenoma series detailing such findings and encompassing a total of 696 patients was 94.8%. Tumor growth cessation, not amount of volume reduction, is still considered successful treatment. Some series have even demonstrated improvement in visual function following radiosurgery related to shrinkage of the tumor [28,29,32,35,40,78,79]. Most pituitary adenomas are slow growing lesions. As such, it is important to perform longterm neuro-imaging follow-up of radiosurgical patients. Such follow-up will help to differentiate between the natural history of an adenoma and true radiosurgical induced volume control. It has been our experience that the longer the follow-up the more likely a patient’s pituitary
. Table 72-1 Radiosurgery for patients with nonfunctioning pituitary adenomas
Authors (year)
Radiosurgery unit
No. of patients
Mean or median follow-up (months)
Martinez et al. [24] Lim et al. [25] Mitsumori et al. [26] Witt et al. [27] Yoon et al. [28] Hayahsi et al. [29] Inoue et al. [30] Mokry et al. [31] Izawa et al. [32] Shin et al. [33] Feigl et al. [34] Sheehan et al. [35] Wowra and Stummer [36] Petrovich et al. [37] Muramatsu et al. [38] Pollock and Carpenter [39] Losa et al. [40] Muacevic et al. [41] Picozzi et al. [42] Kajiwara et al. [43] Iwai et al. [44] Mingione et al. [45]
GK GK LINAC GK LINAC GK GK GK GK GK GK GK GK GK LINAC GK GK GK GK CK GK GK
14 22 7 24 8 18 18 31 23 3 61 42 30 56 8 33 54 51 51 14 28 100
36 26 47 32 49 16 >24 21 28 19 55 31 58 41 30 43 41 21.7 40.6 35.3 36.4 44.9
Maximal dose (Gy)
Margin dose (Gy)
Growth control (%)
28 48 19 38 21 NR 43 28 NR NR NR 32 29 30 26.9 36 33 NR NR NR NR 41.5
16 25 15 19 17 20 20 14 22 16 15 16 16 15 15 16 17 16.5 16.5 12.6 12.3 18.5
100 92 100 94 96 92 94 98 94 100 94 98 93 100 100 97 96 95 89.8 93 93 92
Note: NR, not reported; GK, Gamma Knife radiosurgery; LINAC, linear accelerator radiosurgery unit; CK, Cyberknife
Radiosurgery for pituitary tumors
72
. Table 72-2 Radiosurgery for patients with Cushing’s disease Mean or median followup (months)
Endocrine remission rate (%)
Endocrinological criteria for remission
Maximal dose (Gy)
Margin dose (Gy)
NR
150
NR
86
4
18
58.25
25
50
GK
3
36
40
24
100
GK LINAC
4 5
26 47
48 19
25 15
25 40
GK
6
20
NR
28
67
GK LINAC
25 1
32 49
38 21
19 17
28 NR
UFC < 90 mg/24 h; normal ACTH and cortisol Normal 24 h UFC NR
GK
10
16
NR
24
10
NR
GK
3
>24
43
20
100
NR
GK
8
27
55
29
62
GK GK
50 5
NR 56
NR 35
NR 17
58 33
UFC < 100 mg/24 h Normal 24 h UFC NR
GK
12
28
NR
22
17
NR
GK
43
44
47
20
63
Normal 24 h UFC
GK GK
7 18
88 204
NR 60–240
32 NR
50 83
UFC < 90 mg/24 h Normal 24 h UFC
GK GK
4 20
55 64
NR 49
15 29
NR 35
GK
9
42
40
20
78
NR ACTH < 50 pg/mL; cortisol < 10 mg/dL UFC < 90 mg/24 h
LINAC
5
38
16–20
14.8– 19.2
80
Authors (year)
Radiosurgery unit
No. of patients
Levy et al. [46]
Proton/helium beam
64
Ganz et al. [47]
GK
Martinez et al. [24]
Lim et al. [25] Mitsumori et al. [26] MorangeRamos et al. [48] Witt et al. [27] Yoon et al. [28] Hayashi et al. [29] Inoue et al. [30] SH Kim et al. [49] Laws et al. [50] Mokry et al. [31] Izawa et al. [32] Sheehan et al. [4] Shin et al. [33] Hoybye et al. [51] Feigl et al. [34] Kobayashi et al. [52] Pollock et al. [18] Wong et al. [53]
Normal basal cortisol and dexamethasone test AM UC < 650 nmol/24 h and PM UC < 250 nmol/ 24 h ACTH < 10 microg/L; UFC < 650 nmol/ 24 h NR NR
Normal urinary cortisol and dexamethasone suppression test
1175
1176
72
Radiosurgery for pituitary tumors
. Table 72-2 (continued) Mean or median followup (months)
Endocrine remission rate (%)
Maximal dose (Gy)
Margin dose (Gy)
41
30
15
50
5 35
42.5 42
54.1 33.7
28.5 14.7
56 49
CK
2
35.3
NR
17.5
50
GK
40
54.7
NR
29.5
42.5
GK
90
45
49
23
54
Authors (year)
Radiosurgery unit
Petrovich et al. [37]
GK
Choi et al. [54] Devin et al. [55] Kajiwara et al. [43] Castinetti et al. [56]
GK LINAC
Jagannathan et al. [57]
No. of patients 4
Endocrinological criteria for remission Normal serum cortisol, ACTH, and 24-h UF cortisol UFC < 90 mg/24 h Normal cortisol Normal ACTH and cortisol Normal UFC and dexamethasone suppression test Normal 24 h UFC
Note: NR, not reported; GK, Gamma Knife radiosurgery, LINAC, linear accelerator radiosurgery unit; CK, Cyberknife
adenoma falls into one of two categories—tumor enlargement or tumor reduction. No growth may simply be a limitation of inadequate follow-up following radiosurgery.
Radiosurgery for Cushing’s Disease Cushing’s disease, perhaps the most famous of pituitary disorders, was described by Harvey Cushing in 1912 as a polyglandular disorder [80]. It was not until 1933 that Cushing first performed neurosurgery to treat a patient with a basophilic pituitary adenoma presumed to be secreting excess ACTH [80]. Over the years, neurosurgeons and endocrinologists have debated the criteria for defining ‘‘cure’’ for Cushings disease [81]. Many advocate the use of a 24-h urine free cortisol (UFC) determination as the ‘‘gold standard.’’ Others measure ACTH or basal serum or salivary cortisol and factor these into the evaluation of endocrinological success or failure in Cushing’s disease. In fact in a recent consensus statement by leading endocrinologists, there was no consensus
regarding the definition of endocrinological cure, and the remission rates vary according to the criteria used and the time assessed [82]. Most centers define an endocrinological remission as a 24-h UFC in the normal range coupled with the resolution of clinical stigmata, or a series of normal post-operative serum cortisol levels obtained throughout the day (range 5.4–10.8 mg/dL or 150–300 nmol/L) [82]. Twenty-seven series have reported the results for 486 patients with Cushing’s disease treated surgically with postoperative radiosurgery (> Table 72-2) [2,5,18,24–26,28–34, 37,46,48,51,52,54,83,84]. Mean radiosurgical margin doses for these series range from 15 to 32 Gy. Fifteen series utilized the urine cortisol collection as part of the criteria for endocrinological evaluation. Unfortunately, another eight of these studies do not report the methodology employed to establish endocrinological remission or failure. The others utilize a combination of the aforementioned endocrinological tests. Endocrinological remission rates vary from 10 to 100%. In those series with at least ten patients and a median follow-up of 2 years, endocrinological
Radiosurgery for pituitary tumors
72
. Table 72-3 Radiosurgery for Patients with Acromegaly
Maximal dose (Gy)
Margin dose (Gy)
Endocrine remission rate (%)
Endocrinological criteria for remission
Authors (year)
Radiosurgery unit
Ganz et al. [47] Martinez et al. [24] Landolt et al. [58]
GK
4
18
54.25
19.5
25
GH < 5 mIU/L
GK
7
36
39
25
71
Normal IGF-1
GK
16
NR
50
25
81
GK LINAC
20 1
26 47
48 19
25 15
38 0
GH < 10 mIU/L and IGF-1 < 50 mIU/L GH < 2 ng/mL NR
GK
15
20
NR
28
20
GH < 5 ng/mL; normal IGF-1
GK LINAC
20 2
32 49
38 21
19 17
20 50
Normal IGF-1 GH < 5 ng/mL
GK
22
16
NR
24
41
NR
GK
12
>24
43
20
58
NR
GK
2
12
36
22
0
NR
GK
11
27
55
29
46
GH < 5 ng/mL
GK
56
NR
NR
NR
25
Mokry et al. [31]
GK
16
46
33
16
31
Izawa et al. [32] Shin et al. [33]
GK
29
28
NR
22
41
Normal IGF-1 for gender and age GH < 7 ng/mL; IGF-1 < 380 IU/ mL NR
GK
6
43
NR
34
67
Zhang et al. [60] Fukuoka et al. [61] Ikeda et al. [62] Feigl et al. [34] Pollock et al. [18]
GK
68
34
NR
31
96
GK
9
42
41
20
50
GK
17
48
NR
25
82
GK GK
9 26
55 42
NR 40
15 20
NR 42
GK
30
46
40
20
37
Lim et al. [25] Mitsumori et al. [26] MorangeRamos et al. [48] Witt et al. [27] Yoon et al. [28] Hayashi et al. [29] Inoue et al. [30] MS Kim et al. [59] SH Kim et al. [49] Laws et al. [50]
Attanasio et al. [63]
No. of patients
Mean or median followup (months)
GH < 10 mIU/L and IGF-1 < 450 ng/mL GH < 12 ng/mL GH < 5 ng/mL and normal IGF-1 Normal IGF-1 for age NR GH < 2 ng/mL and normal IGF-1 for age GH < 2.5 microg/ L
1177
1178
72
Radiosurgery for pituitary tumors
. Table 72-3 (continued)
Maximal dose (Gy)
Margin dose (Gy)
Endocrine remission rate (%)
Endocrinological criteria for remission
Authors (year)
Radiosurgery unit
Petrovich et al. [37] Muramatsu et al. [38] Choi et al. [54] Gutt et al. [64] Kajiwara et al. [43] Castinetti et al. [65]
GK
6
41
30
15
100
LINAC
4
30
24.5
15
50
Normal IGF-1 and GH NR
Koybayashi et al. [66] Jezkova et al. [67] Pollock et al. [68] Roberts et al. [69]
No. of patients
Mean or median followup (months)
GK GK CK
12 44 2
42.5 22.8 35.3
54.1 36 NR
28.5 18 17.5
50 48 0
GH < 5 mIU/L Normal IGF-1 NR
GK
82
49.5
NR
25
17
GK
67
63.3
35.3
18.9
GH < 2 ng/mL and normal IGF-1 for age GH < 1 ng/mL
GK
96
43.2
70
35
50
GK
46
63
43.5
20
60
CK
9
25.4
NR
21
44.4
4.8
GH < 2.5 mg/L and normal IGF-1 GH < 2 ng/mL and normal IGF-1 for age Normal IGF-1
Note: NR, not reported; GK, Gamma Knife radiosurgery; LINAC, linear accelerator radiosurgery unit; CK, Cyberknife
remission rates range from 17 to 83%. The wide range of endocrinological remission rates is likely a function of the testing utilized and the practice of some centers to perform testing while patients remain on steroid synthesis inhibiting agents (e.g. ketoconazole). Rare cases of delayed recurrence following initial radiosurgical remission have been observed at the University of Virginia. This only stands to reason as such delayed recurrences can also occur after microsurgery. Delayed recurrence underscores the need for long-term periodic endocrinologic testing.
Radiosurgery for Acromegaly Just as the endocrinological criteria for Cushing’s disease remain the subject of debate, the criteria for curing acromegaly have also been inconsistent. The most widely accepted guidelines for
remission in acromegaly consist of a GH level less than 1 ng/mL in response to a glucose challenge and a normal serum IGF-1 when matched for age and gender [85–88]. Thirty-two studies detail the results of radiosurgical treatment for 766 patients with acromegaly (> Table 72-3) [2,18,24–34, 37,38,47–49,54,58– 63]. The mean radiosurgery margin doses in these series range from 15 to 35 Gy. Eight studies did not report the criteria utilized to define an endocrinological remission. Of the remaining twenty-four studies, a myriad of criteria are employed to define remission. Remission rates following radiosurgery vary from 0 to 100%. In those series with at least ten patients and a median follow-up of 2 years, endocrinological remission rates still vary with a wide range (4.8–96%). Certainly, some of the variation in endocrinological remission rates with acromegaly may be attributed to the inconsistent criteria defining remission. Another confounding
Radiosurgery for pituitary tumors
72
. Table 72-4 Radiosurgery for patients with prolactinomas Mean or median follow-up (months)
Maximal dose (Gy)
Margin dose (Gy)
Endocrine remission rate (%)
20
12
150
NR
60
3
18
40
13.3
0
Authors (year)
Radiosurgery unit
Levy et al. [46] Ganz et al. [47]
Proton/ helium beam GK
Martinez et al. [24] Lim et al. [25] Mitsumori et al. [26] Witt et al. [27] Yoon et al. [28] Hayashi et al. [29] Inoue et al. [30] MS Kim et al. [59] SH Kim et al. [49] Laws et al. [50] Mokry et al. [31] MorangeRamos et al. [70] Izawa et al. [32] Landolt et al. [16]
GK
5
36
43
33
0
GK
19
26
48
25
56
4
47
19
15
0
NR
GK
12
32
38
19
0
NR
LINAC
11
49
21
17
84
<25 ng/mL
GK
13
16
NR
24
15
NR
GK
2
>24
43
20
50
NR
GK
20
12
36
22
19
NR
GK
18
27
55
29
17
<20 ng/mL
GK
19
NR
NR
NR
7
<20 ng/mL
GK
21
31
30
14
21
<25 ng/mL
GK
4
20
NR
28
0
GK
15
28
NR
22
20
GK
20
29
50
25
25
Pan et al. [71] Feigl et al. [34] Pollock et al. [18] Petrovich et al. [37] Muramatsu et al. [38]
GK
128
33
45
32
15
<19 ng/mL for women and <16 ng/mL for men <30 ng/mL
GK
18
55
NR
15
NR
NR
GK
7
42
40
20
29
<23 ng/mL
GK
12
41
30
15
83
1
30
30
15
0
Normal serum prolactin NR
LINAC
LINAC
No. of patients
Endocrinological criteria for remission (prolactin level) Normal serum prolactin <525 mIU/L for women and <415 mIU/L men/ postmenopausal women <20 ng/mL <25 ng/mL
<25 ng/mL for women and <20 ng/mL for men NR
1179
1180
72
Radiosurgery for pituitary tumors
. Table 72-4 (continued)
No. of patients
Mean or median follow-up (months)
Maximal dose (Gy)
Margin dose (Gy)
Endocrine remission rate (%)
Endocrinological criteria for remission (prolactin level)
Authors (year)
Radiosurgery unit
Choi et al. [54] Kajiwara et al. [43] Pouratian et al. [19] Ma et al. [72]
GK
21
42.5
54.1
28.5
24
<20 ng/mL
CK
3
35.3
NR
17.5
33
GK
23
55
42.2
18.6
26
GK
51
37
50.41
26.1
40
Normal serum prolactin Normal serum prolactin Normal serum prolactin
Note: NR, not reported; GK, Gamma Knife radiosurgery; LINAC, linear accelerator radiosurgery unit; CK, Cyberknife
variable is the degree to which somatostatin analogs may have been utilized during the time of radiosurgery and subsequent endocrinological evaluation in each of the series. It is imperative that patients be taken off antisecretory agents prior to testing for radiosurgical induced endocrinological remission (> Figure 72-1).
Radiosurgery for Prolactinoma Most prolactinomas are treated medically. For those patients who are refractory to or unable to tolerate antisecretory dopamine agonist medications, radiosurgery represents an alternative. Prolactinoma remission criteria are more consistent among endocrinologists. Most studies define remission as a patient who has a normal serum prolactin level for gender. Twenty-five radiosurgical studies report the results for 470 patients with prolactinomas (> Table 72-4) [2,3,17,18,24–30,32,34,37,38,46,48,49,54,59,60, 71,83]. The mean radiosurgical dose to the tumor margin varied from 13.3 to 33 Gy. Although eight of these studies did not report the endocrinological criteria defining remission, the remaining studies utilize relatively similar criteria. Remission rates varied from 0 to 84%. In our series of patients [19], we observed a
significant improvement in prolactinoma remission in patients who were not receiving antisecretory medications at the time of their Gamma Knife radiosurgery. In studies with at least ten patients and a median or mean follow-up of 2 years, the range in remission rates following radiosurgery was just as varied. The largest series by Pan et al. reported a 15% endocrinological remission rate for 128 patients with a median follow-up of 33 months [71]. Although the remission rates for prolactinomas appear to be lower than those for Cushing’s disease or acromegaly, a substantial number of patients have a reduction but not complete normalization of their hyperprolactinemia following radiosurgery. Similar to acromegaly, differences in the use of antisecretory dopamine agonists at some centers may confound the efficacy of radiosurgery and the endocrinological assessment of treated patients with prolactinomas in these series. In addition, Hoybye et al. have demonstrated that radiosurgery may cause an elevation in prolactin levels, possibly through injury or irritation of the infundibulum and impaired transport of dopamine to the anterior pituitary [51]. This elevation can last for several years and may falsely lower the reported remission rates for patients with prolactinomas treated with radiosurgery.
3 1 9 6 11 23
GK GK
GK
GK
17
No. of patients
Proton/helium beam GK GK
Radiosurgery unit
Note: NR, not reported. GK, Gamma Knife radiosurgery
Ganz et al. [47] Wolffenbuttel et al. [73] Laws et al. [50] Kobayashi et al. [52] Pollock and Young [74] Mauermann et al. [75]
Levy et al. [46]
Authors (year)
50
37
NR 63
18 33
NR
Mean or median follow-up (months)
. Table 72-5 Radiosurgery for Patients with Nelson’s syndrome
50
40
NR 49.4
56.7 40
150
Maximal dose (Gy)
25
20
NR 28.7
NR 12
NR
Margin dose (Gy)
17
36
11 33
0 0
NR
Endocrine cure rate (%)
Normal ACTH
NR
NR ACTH< 50 pg/mL
NR NR
Normal ACTH
Endocrinological criteria for cure
91
82
NR 100
100 100
94
Growth control (%)
Radiosurgery for pituitary tumors
72 1181
1182
72
Radiosurgery for pituitary tumors
. Figure 72‐1 The patient had persistent Cushing’s disease. He underwent Gamma Knife radiosurgery to treat his invasive pituitary adenoma (a). Within 6 months post-radiosurgery, the patient achieved endocrine remission of his Cushing’s disease. His pituitary adenoma has markedly shrunk in size (b)
Radiosurgery for Nelson’s Syndrome Compared to nonfunctioning and other functioning pituitary adenomas, much less information is available about the efficacy of stereotactic radiosurgery for the treatment of Nelson’s syndrome. In patients with ACTH secreting tumors who have undergone adrenalectomies, these pituitary adenomas tend to result in more aggressive growth rates. As such, endocrinological remission and growth control are critical for Nelson’s syndrome. Seven studies detailed results of stereotactic radiosurgery for 70 patients with Nelson’s syndrome (> Table 72-5) [2,18,46,47,52,73]. Mean tumor margin dose varied from 12 to 28.7 Gy. Unfortunately, only three of the studies described the endocrinological criteria utilized to define a remission. Endocrinological remission rates ranged from 0 to 36%. However, tumor growth control rates were more favorable and varied from 82 to 100%, but with relatively short follow-up.
Radiosurgical Rates of Endocrine Improvement and Late Recurrence An ideal treatment would lead to a rapid endocrinological normalization. The rate of hormonal
improvement and normalization following radiosurgery is difficult to predict. Some series have reported hormonal improvement in as little as 3 months following radiosurgery whereas others have reported normalization occurring more than 8 years afterwards [5,33]. Generally, if endocrinological normalization is going to occur following radiosurgery, it usually does so within the first 2 years [2,5,60,74,89]. When alternative approaches such as adrenalectomy for Cushing’s patients are available, the benefits of waiting longer than 2 years for a radiosurgical induced endocrinologic normalization must be weighed against the risks of morbidity from hormonal oversecretion. Several instances of late recurrence of hormonal oversecretion have been reported despite earlier symptomatic and endocrinological confirmed remissions [5]. As such, long-term radiological and endocrinological follow-up is recommended for all pituitary patients treated with radiosurgery to detect any possibility of late recurrence and tumor growth. The effects of treatment volume and dose selection on the rate and extent of hormonal normalization remain the subject of debate. Some investigators have found that radiation dose and treatment volume do not affect the rate or extent of hormonal normalization [59,90]. Others have found a correlation between
Radiosurgery for pituitary tumors
hormonal normalization and the following: treatment isodose; maximal dose; margin dose; tumor invasiveness and the absence of hormonesuppressive medications around the time of radiosurgery [16–18,71]. There does not appear to be a correlation between tumor volume response and the endocrinological response following radiosurgery [5,16,34]. Fortunately, as most pituitary adenomas are well within the size of lesion that is suitable for stereotactic radiosurgery, dose-volume considerations are not as much of an issue. The dose is usually limited by the proximity of the adenoma to the optic apparatus, and current shielding techniques can help to facilitate delivery of higher doses. Since the systemic effects of functioning adenomas can be so devastating to patients, it seems intuitive to deliver a reasonably high dose (equal to or greater than 20 Gy to the adenoma margin) to effectuate hormonal normalization and tumor growth control. Nonfunctioning pituitary adenomas appear to require a lower radiosurgery treatment dose (i.e. 15–18 Gy) than functioning adenomas [11,25,32,33,91,92]. The lowest effective dose for a nonfunctioning tumor is not known (> Figure 72-2).
Complications Following Radiosurgery for Pituitary Adenomas Cranial Neuropathies Following Radiosurgery In our review of 53 studies encompassing 2,483 patients, there were 27 cases of damage to the optic apparatus. The post-radiosurgical visual field deficits ranged from quadrantanopsias to complete visual loss. Radiosurgical doses associated with visual field loss varied from 0.7 to 12 Gy. The tolerable level of radiation to the optic apparatus is still a subject of debate. Some believe that the optic apparatus can tolerate doses as high as 12 to 14.1 Gy [79,93,94]. Others recommend an
72
upper limit of 8–10 Gy [95,96]. Small volumes of the optic apparatus exposed to doses of 10 Gy or less may be acceptable in some cases [93,97]. Both the tolerable absolute dose and volume undoubtedly vary from patient to patient. This degree of variability likely depends upon the extent of damage to the optic apparatus by pituitary adenoma compression, ischemic changes, type and timing of previous interventions (e.g. fractionated radiation therapy and surgery), the patient’s age, and the presence or absence of other co-morbidities (e.g. diabetes) [98,99]. The other consideration for limiting damage to the optic apparatus during radiosurgery is the distance between nerve and residual adenoma. A distance of 5 mm between the adenoma and the optic apparatus is desirable, but a distance of as little as 1–2 mm may be acceptable, particularly with shielding capabilities of newer radiosurgical units such as the Perfexion. Dose volume of the optic apparatus may be a better way to determine dose and risk [5,25,27,79]. The tolerable distance is a function of the degree to which a dose plan can be designed to deliver a suitable radiation dose to the adenoma yet spare the optic apparatus. Without achievement of a suitable stereotactic radiosurgery dose plan, alternative treatment modalities (i.e. surgical resection, medical management, or fractionated radiation therapy) should be chosen. Just as visual dysfunction of the optic apparatus has been described following radiosurgery, so too has improvement in visual function. However, such improvement is rare. Improvement in visual acuity and fields has been noted following radiosurgery in some patients with pituitary adenomas and may be a result of tumor shrinkage and optic nerve decompression [28,29,32,35,78,79,100]. The other cranial nerves in the cavernous sinus appear to be much more resistant to injury from stereotactic radiosurgery. In the 53 studies reviewed, 27 patients had other cranial neuropathies in oculomotor, trochlear, trigeminal, or abducent nerves, and nearly half of these were transient.
1183
1184
72
Radiosurgery for pituitary tumors
. Figure 72‐2 This depicts a pituitary adenoma (yellow) in the right cavernous sinus with a highly conformal dose plan. The carotid arteries are outlined in red and the optic apparatus is outlined in blue
Radiosurgical Injury to Adjacent Vascular Structures
Radiosurgical Induced Parenchymal Brain Injury
Injury to the cavernous segment of the carotid artery is rare following radiosurgery. A total of four cases have been reported and only two of these cases were symptomatic from carotid artery stenosis [18,25,38]. Pollock et al. [18] have recommended that the prescription dose should be limited to less than 50% of the intracavernous carotid artery vessel diameter [18]. Shin et al. recommended restricting the dose to the internal carotid artery to less than 30 Gy [33]. The dose restriction advocated to the carotid is not based upon a strong radiobiological foundation.
Thirteen cases of parenchymal changes consistent with radiation effect are noted. These findings were most often associated within the hypothalamic and temporal regions. Clinical manifestations of the parenchymal injury occurred as early as 5 h and as late as approximately 1 year after radiosurgery. Six of the thirteen patients had fractionated radiation therapy prior to radiosurgery [18,26,29,32,46,74]. Clearly, previous radiotherapy represents an added risk factor for the development of necrosis after radiosurgery. Both the timing and total dose of previous radiation
Radiosurgery for pituitary tumors
72
. Figure 72‐3 The Gamma Knife Perfexion is the latest Gamma Knife unit. It combines automatic positioning with moving sources and nearly instantaneous plugging of radioactive sources
should be considered when developing a stereotactic radiosurgery dose plan.
Radiosurgical Induced Hypopituitarism The incidence of hypopituitarism following radiosurgery is uncertain, with a wide discrepancy in reported series. Well respected groups have reported a low incidence (0–36%) of pituitary dysfunction following radiosurgery [4,5,39,46,101]. A long term study from the Karolinska Institute with a mean follow-up of 17 years, detected a 72% incidence of hypopituitarism [51]. However, many of these patients included in that study were targetted with antiquated imaging techniques and received doses much higher than those used today. We have observed hypopituitarism of 20–30% in our post-Gamma Knife pituitary adenoma patients. The difficulty with determining the exact incidence of radiosurgery-induced hypopituitarism stems in part from the fact that many of the patients have already undergone previous surgical resection and some previous fractionated radiotherapy. In addition, pituitary deficiencies may partially result from aging. Thus, it is likely that hypopituitarism
in the post-radiosurgical population is multifactorial in etiology and related to radiosurgery as well as age-related changes and previous treatments (e.g. microsurgery and radiotherapy). The methods of endocrinological follow-up are inconsistent and unreliable; the indications for obtaining hormone levels and the time at which they were obtained vary widely from study to study.
Radiosurgical Associated Secondary Neoplasms The incidence of radiosurgical induced neoplasms is unknown. It appears to be low and is less common than that reported following fractionated radiation therapy. In the 2,483 patients reviewed, there were no reports of radiation induced neoplasms. Despite over 39 years of radiosurgery in over 250,000 patients, the precise incidence of radiosurgery induced neoplasms is difficult to estimate because of the delayed fashion in which such lesions may develop and the apparently low rate of occurrence. For a tumor to be considered a result of radiosurgery, the following criteria must be met: (1) the tumor must occur within the prior radiation field; (2) it cannot be
1185
1186
72
Radiosurgery for pituitary tumors
present prior to radiation; (3) the primary tumor must differ histologically from the induced tumor; (4) there must be no genetic predisposition for occurrence of a second malignancy or tumor progression; and (5) a certain latency period is required from treatment to tumor formation (e.g. this is usually 5 or more years) [102]. In a recent review, Ganz noted three reports published in peer-reviewed journals that suggest development of a radiosurgery induced neoplasm [83,103–105]. In two of these reports, a glioblastoma arose whereas in the other malignant transformation of a vestibular schwannoma occurred [103–105]. At the University of Virginia, we have observed the late formation of meningiomas in two pediatric patients who underwent Gamma Knife radiosurgery [76]. The latency of radiosurgical induced neoplasia underscores the need for long-term radiologic follow-up in patients treated with radiosurgery. Other cases of patients treated with radiosurgery who have later been diagnosed with a malignant brain tumor have been reported but they generally do not meet the aforementioned criteria [106–108]. Based upon the literature, the incidence of stereotactic radiosurgical induced neoplasia is quite low, and the potential risk of radiosurgical associated secondary neoplasia must be weighed against potential benefits. It is generally believed that the risk will be lower than that seen with fractionated therapy because of the smaller irradiation volumes associated with radiosurgery. Alternative treatments if available are not without appreciable risks and sometimes there are no viable options other than stereotactic radiosurgery. Although time will reveal the true incidence of complications associated with radiosurgery, the substantial body of information presently available would suggest that its risk to benefit profile is attractive.
Conclusions Except for prolactinomas, microsurgery remains the primary treatment in surgically fit patients for
sellar lesions, particularly so when the lesion is demonstrating mass effect on the optic apparatus or hormonal overproduction. Nevertheless, 20–50% of patients eventually demonstrate recurrence of their adenomas, and adjuvant treatment is often necessary for these patients. Stereotactic radiosurgery has been demonstrated to be a safe and highly effective treatment for patients with recurrent or residual pituitary adenomas. Radiosurgery affords effective growth control and hormonal normalization for patients with a generally shorter latency period than that of fractionated radiotherapy. This short latency period with radiosurgery can typically be managed with suppressive medications. Furthermore, the complications (e.g. radiation induced neoplasia, cerebral vasculopathy, etc.) associated with radiosurgery are infrequent. Radiosurgery may even serve as a primary treatment for those patients deemed unfit for microsurgical resection as a result of other comoribidities or with demonstrable tumors in a surgically inaccessible location. In the future, radiosurgery will yield even better results. The introduction of technical advances such as the Gamma Knife Perfexion, incorporation of new neuro-imaging technologies into dose planning, and improvements in the shielding techniques of radiosurgical units will provide improved conformity and steeper dose fall-off [109,110]. Long-term periodic neurological, neuro-imaging, and endocrinological follow-up of patients must be performed to assess for delayed complications or tumor recurrence. Finally, physicians caring for pituitary patients should establish uniform endocrinological criteria and diagnostic testing for preand post-radiosurgical evaluations.
References 1. Laws ER Jr, Ebersold M, Piepgras DG. The results of transsphenoidal surgery in specific clinical entities. In: Laws ER Jr, Randall R, Kern EB, editors. Management of pituitary adenomas and related lesions with emphasis on transsphenoidal microsurgery. New York: AppletonCentury-Crofts; 1982. p. 277-305.
Radiosurgery for pituitary tumors
2. Laws ER Jr, Vance ML. Radiosurgery for pituitary tumors and craniopharyngiomas. Neurosurg Clin N Am 1999;10:327-36. 3. Laws ER Jr, Fode NC, Redmond MJ. Transsphenoidal surgery following unsuccessful prior therapy. An assessment of benefits and risks in 158 patients. J Neurosurg 1985;63:823-9. 4. Sheehan JM, Lopes MB, Sheehan JP, Ellegala D, Webb KM, Laws ER Jr. Results of transsphenoidal surgery for Cushing’s disease in patients with no histologically confirmed tumor. Neurosurgery 2000;47:33-6; discussion 37–9. 5. Sheehan JM, Vance ML, Sheehan JP, Ellegala DB, Laws ER Jr. Radiosurgery for Cushing’s disease after failed transsphenoidal surgery. J Neurosurg 2000; 93:738-42. 6. Milker-Zabel S, Debus J, Thilmann C, Schlegel W, Wannenmacher M. Fractionated stereotactically guided radiotherapy and radiosurgery in the treatment of functional and nonfunctional adenomas of the pituitary gland. Int J Radiat Oncol Biol Phys 2001;50:1279-86. 7. Meij BP, Lopes MB, Ellegala DB, Alden TD, Laws ER Jr. The long-term significance of microscopic dural invasion in 354 patients with pituitary adenomas treated with transsphenoidal surgery. J Neurosurg 2002; 96:195-208. 8. Shaffi OM, Wrightson P. Dural invasion by pituitary tumours. N Z Med J 1975;81:386-90. 9. Chandler WF, Schteingart DE, Lloyd RV, McKeever PE, Ibarra-Perez G. Surgical treatment of Cushing’s disease. J Neurosurg 1987;66:204-12. 10. Friedman RB, Oldfield EH, Nieman LK, Chrousos GP, Doppman JL, Cutler GB Jr, Loriaux DL. Repeat transsphenoidal surgery for Cushing’s disease. J Neurosurg 1989;71:520-7. 11. Mampalam TJ, Tyrrell JB, Wilson CB. Transsphenoidal microsurgery for Cushing disease. A report of 216 cases. Ann Intern Med 1988;109:487-93. 12. Leksell L. The stereotaxic method and radiosurgery of the brain. Acta Chir Scand 1951;102:316-19. 13. Kondziolka D, Maitz AH, Niranjan A, Flickinger JC, Lunsford LD. An evaluation of the Model C gamma knife with automatic patient positioning. Neurosurgery 2002;50:429-31; discussion 422–31. 14. Friedman WA, Foote KD. Linear accelerator radiosurgery in the management of brain tumours. Ann Med 2000;32:64-80. 15. Lyman JT, Phillips MH, Frankel KA, Levy RP, Fabrikant JI. Radiation physics for particle beam radiosurgery. Neurosurg Clin N Am 1992;3:1-8. 16. Landolt AM, Haller D, Lomax N, Scheib S, Schubiger O, Siegfried J, Wellis G. Octreotide may act as a radioprotective agent in acromegaly. J Clin Endocrinol Metab 2000;85:1287-9. 17. Landolt AM, Lomax N. Gamma knife radiosurgery for prolactinomas. J Neurosurg 2000;93 Suppl 3:14-18.
72
18. Pollock BE, Nippoldt TB, Stafford SL, Foote RL, Abboud CF. Results of stereotactic radiosurgery in patients with hormone-producing pituitary adenomas: factors associated with endocrine normalization. J Neurosurg 2002;97:525-30. 19. Pouratian N, Sheehan J, Jagannathan J, Laws ER Jr, Steiner L, Vance ML. Gamma knife radiosurgery for medically and surgically refractory prolactinomas. Neurosurgery 2006;59:255-66; discussion 255–66. 20. Kondziolka D, Dempsey PK, Lunsford LD, Kestle JR, Dolan EJ, Kanal E, Tasker RR. A comparison between magnetic resonance imaging and computed tomography for stereotactic coordinate determination. Neurosurgery 1992;30:402-6; discussion 406–7. 21. Tang BN, Levivier M, Heureux M, Wikler D, Massager N, Devriendt D, David P, Dumarey N, Corvilain B, Goldman S. 11C-methionine PET for the diagnosis and management of recurrent pituitary adenomas. Eur J Nucl Med Mol Imaging 2006;33:169-78. 22. Semple PL, Laws ER Jr. Complications in a contemporary series of patients who underwent transsphenoidal surgery for Cushing’s disease. J Neurosurg 1999;91:175-9. 23. Shimon I, Ram Z, Cohen ZR, Hadani M. Transsphenoidal surgery for Cushing’s disease: endocrinological follow-up monitoring of 82 patients. Neurosurgery 2002;51:57-61; discussion 52–61. 24. Martinez R, Bravo G, Burzaco J, Rey G. Pituitary tumors and gamma knife surgery. Clinical experience with more than two years of follow-up. Stereotact Funct Neurosurg 1998;70 Suppl 1:110-18. 25. Lim YL, Leem W, Kim TS, Rhee BA, Kim GK. Four years’ experiences in the treatment of pituitary adenomas with gamma knife radiosurgery. Stereotact Funct Neurosurg 1998;70 1:95-109. 26. Mitsumori M, Shrieve DC, Alexander E III, Kaiser UB, Richardson GE, Black PM, Loeffler JS. Initial clinical results of LINAC-based stereotactic radiosurgery and stereotactic radiotherapy for pituitary adenomas. Int J Radiat Oncol Biol Phys 1998;42:573-80. 27. Witt TC, Kondziolka D, Flickinger JC, Lunsford LD. Gamma Knife radiosurgery for pituitary tumors. In: Lunsford LD, Kondziolka D, Flickinger J, editors. Gamma Knife brain surgery Progress in Neurological Surgery. Basel: Karger; 1998. p. 114-27. 28. Yoon SC, Suh TS, Jang HS, Chung SM, Kim YS, Ryu MR, Choi KH, Son HY, Kim MC, Shinn KS. Clinical results of 24 pituitary macroadenomas with linacbased stereotactic radiosurgery. Int J Radiat Oncol Biol Phys 1998;41:849-53. 29. Hayashi M, Izawa M, Hiyama H, Nakamura S, Atsuchi S, Sato H, Nakaya K, Sasaki K, Ochiai T, Kubo O, Hori T, Takakura K. Gamma Knife radiosurgery for pituitary adenomas. Stereotact Funct Neurosurg 1999;72 Suppl 1:111-18.
1187
1188
72
Radiosurgery for pituitary tumors
30. Inoue HK, Kohga H, Hirato M, Sasaki T, Ishihara J, Shibazaki T, Ohye C, Andou Y. Pituitary adenomas treated by microsurgery with or without Gamma Knife surgery: experience in 122 cases. Stereotact Funct Neurosurg 1999;72 Suppl 1:125-31. 31. Mokry M, Ramschak-Schwarzer S, Simbrunner J, Ganz JC, Pendl G. A six year experience with the postoperative radiosurgical management of pituitary adenomas. Stereotact Funct Neurosurg 1999;72 Suppl 1:88-100. 32. Izawa M, Hayashi M, Nakaya K, Satoh H, Ochiai T, Hori T, Takakura K. Gamma knife radiosurgery for pituitary adenomas. J Neurosurg 2000;93 Suppl 3:19-22. 33. Shin M, Kurita H, Sasaki T, Tago M, Morita A, Ueki K, Kirino T. Stereotactic radiosurgery for pituitary adenoma invading the cavernous sinus. J Neurosurg 2000;93 Suppl 3:2-5. 34. Feigl GC, Bonelli CM, Berghold A, Mokry M. Effects of gamma knife radiosurgery of pituitary adenomas on pituitary function. J Neurosurg 2002;97:415-21. 35. Sheehan JP, Kondziolka D, Flickinger J, Lunsford LD. Radiosurgery for residual or recurrent nonfunctioning pituitary adenoma. J Neurosurg 2002;97:408-14. 36. Wowra B, Stummer W. Efficacy of gamma knife radiosurgery for nonfunctioning pituitary adenomas: a quantitative follow up with magnetic resonance imaging-based volumetric analysis. J Neurosurg 2002; 97:429-32. 37. Petrovich Z, Yu C, Giannotta SL, Zee CS, Apuzzo ML. Gamma knife radiosurgery for pituitary adenoma: early results. Neurosurgery 2003;53:51-9; discussion 59–61. 38. Muramatsu J, Yoshida M, Shioura H, Kawamura Y, Ito H, Takeuchi H, Kubota T, Maruyama I. Clinical results of LINAC-based stereotactic radiosurgery for pituitary adenoma. Nippon Igaku Hoshasen Gakkai Zasshi 2003;63:225-30. 39. Pollock BE, Carpenter PC. Stereotactic radiosurgery as an alternative to fractionated radiotherapy for patients with recurrent or residual nonfunctioning pituitary adenomas. Neurosurgery 2003;53:1086-91; discussion 1084–91. 40. Losa M, Valle M, Mortini P, Franzin A, da Passano CF, Cenzato M, Bianchi S, Picozzi P, Giovanelli M. Gamma knife surgery for treatment of residual nonfunctioning pituitary adenomas after surgical debulking. J Neurosurg 2004;100:438-44. 41. Muacevic A, Uhl E, Wowra B. Gamma knife radiosurgery for nonfunctioning pituitary adenomas. Acta Neurochir Suppl 2004;91:51-4. 42. Picozzi P, Losa M, Mortini P, Valle MA, Franzin A, Attuati L, Ferrari da Passano C, Giovanelli M. Radiosurgery and the prevention of regrowth of incompletely removed nonfunctioning pituitary adenomas. J Neurosurg 2005;102 Supp1:71-4. 43. Kajiwara K, Saito K, Yoshikawa K, Kato S, Akimura T, Nomura S, Ishihara H, Suzuki M. Image-guided stereotactic radiosurgery with the CyberKnife for pituitary adenomas. Minim Invasive Neurosurg 2005;48(2):91-6.
44. Iwai Y, Yamanaka K, Yoshioka K. Radiosurgery for nonfunctioning pituitary adenomas. Neurosurgery 2005;56(4):699-705. 45. Mingione V, Yen CP, Vance ML, Steiner M, Sheehan J, Laws ER, Steiner L. Gamma surgery in the treatment of nonsecretory pituitary macroadenoma. J Neurosurg 2006;104(6):876-83. 46. Levy RP, Fabrikant JI, Frankel KA, Phillips MH, Lyman JT, Lawrence JH, Tobias CA. Heavy-charged-particle radiosurgery of the pituitary gland: clinical results of 840 patients. Stereotact Funct Neurosurg 1991;57:22-35. 47. Ganz JC, Backlund EO, Thorsen FA. The effects of Gamma Knife surgery of pituitary adenomas on tumor growth and endocrinopathies. Stereotact Funct Neurosurg 1993;61 Suppl 1:30-7. 48. Morange-Ramos I, Regis J, Dufour H, Andrieu JM, Grisoli F, Jaquet P, Peragut JC. Gamma-knife surgery for secreting pituitary adenomas. Acta Neurochir (Wien) 1998;140:437-43. 49. Kim SH, Huh R, Chang JW, Park YG, Chung SS. Gamma Knife radiosurgery for functioning pituitary adenomas. Stereotact Funct Neurosurg 1999;72 Suppl 1:101-10. 50. Laws ER Jr, Vance ML. Radiosurgery for pituitary tumors and craniopharyngiomas. Neurosurg Clin N Am 1999;10(2):327-36. 51. Hoybye C, Grenback E, Rahn T, Degerblad M, Thoren M, Hulting AL. Adrenocorticotropic hormoneproducing pituitary tumors: 12- to 22-year follow-up after treatment with stereotactic radiosurgery. Neurosurgery 2001;49:284-91. 52. Kobayashi T, Kida Y, Mori Y. Gamma knife radiosurgery in the treatment of Cushing disease: long-term results. J Neurosurg 2002;97:422-8. 53. Wong GK, Leung CH, Chiu KW, Ma R, Cockram CS, Lam MJ, Poon WS. LINAC radiosurgery in recurrent Cushing’s disease after transsphenoidal surgery: a series of 5 cases. Minim Invasive Neurosurg 2003;46(6): 327-30. 54. Choi JY, Chang JH, Chang JW, Ha Y, Park YG, Chung SS. Radiological and hormonal responses of functioning pituitary adenomas after gamma knife radiosurgery. Yonsei Med J 2003;44:602-7. 55. Devin JK, Allen GS, Cmelak AJ, Duggan DM, Blevins LS. The efficacy of linear accelerator radiosurgery in the management of patients with Cushing’s disease. Stereotact Funct Neurosurg 2004;82(5-6):254-62. 56. Castinetti F, Nagai M, Dufour H, Kuhn JM, Morange I, Jaquet P, Conte-Devolx B, Regis J, Brue T. Gamma knife radiosurgery is a successful adjunctive treatment in Cushing’s disease. Eur J Endocrinol 2007;156(1):91-8. 57. Jagannathan J, Sheehan JP, Pouratian N, Laws ER, Steiner L, Vance ML. Gamma Knife surgery for Cushing’s disease. J Neurosurg 2007;106(6):980-7. 58. Landolt AM, Haller D, Lomax N, Scheib S, Schubiger O, Siegfried J, Wellis G. Stereotactic radiosurgery for recur-
Radiosurgery for pituitary tumors
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
rent surgically treated acromegaly: comparison with fractionated radiotherapy. J Neurosurg 1998; 88:1002-8. Kim MS, Lee SI, Sim JH. Gamma Knife radiosurgery for functioning pituitary microadenoma. Stereotact Funct Neurosurg 1999;72 Suppl 1:119-24. Zhang N, Pan L, Wang EM, Dai JZ, Wang BJ, Cai PW. Radiosurgery for growth hormone-producing pituitary adenomas. J Neurosurg 2000;93 Suppl 3:6-9. Fukuoka S, Ito T, Takanashi M, Hojo A, Nakamura H. Gamma knife radiosurgery for growth hormone-secreting pituitary adenomas invading the cavernous sinus. Stereotact Funct Neurosurg 2001;76:213-17. Ikeda H, Jokura H, Yoshimoto T. Transsphenoidal surgery and adjuvant gamma knife treatment for growth hormone-secreting pituitary adenoma. J Neurosurg 2001;95:285-91. Attanasio R, Epaminonda P, Motti E, Giugni E, Ventrella L, Cozzi R, Farabola M, Loli P, Beck-Peccoz P, Arosio M. Gamma-knife radiosurgery in acromegaly: a 4-year follow-up study. J Clin Endocrinol Metab 2003;88:3105-12. Gutt B, Wowra B, Alexandrov R, Uhl E, Schaaf L, Stalla GK, Schopohl J. Gamma-knife surgery is effective in normalising plasma insulin-like growth factor I in patients with acromegaly. Exp Clin Endocrinol Diabetes 2005;113(4):219-24. Castinetti F, Taieb D, Kuhn JM, Chanson P, Tamura M, Jaquet P, Conte-Devolx B, Re´gis J, Dufour H, Brue T. Outcome of gamma knife radiosurgery in 82 patients with acromegaly: correlation with initial hypersecretion. J Clin Endocrinol Metab 2005;90(8):4483-8. Kobayashi T, Mori Y, Uchiyama Y, Kida Y, Fujitani S. Long-term results of gamma knife surgery for growth hormone-producing pituitary adenoma: is the disease difficult to cure? J Neurosurg 2005;102 Supp 1:19-23 Jezkova´ J, Marek J, Ha´na V, Krsek M, Weiss V, Vladyka V, Lisa´k R, Vymazal J, Pecen L. Gamma knife radiosurgery for acromegaly–long-term experience. Clin Endocrinol (Oxf) 2006;64(5):588-95. Pollock BE, Jacob JT, Brown PD, Nippoldt TB. Radiosurgery of growth hormone-producing pituitary adenomas: factors associated with biochemical remission. J Neurosurg 2007;106(5):833-8. Roberts BK, Ouyang DL, Lad SP, Chang SD, Harsh GR, Adler JR, Soltys SG, Gibbs IC, Katznelson L. Efficacy and safety of CyberKnife radiosurgery for acromegaly. Pituitary 2007;10(1):19-25. Morange-Ramos I, Regis J, Dufour H, Andrieu JM, Grisoli F, Jaquet P, Peragut JC. Gamma-knife surgery for secreting pituitary adenomas. Acta Neurochir (Wien) 1998;140(5):437-43. Pan L, Zhang N, Wang EM, Wang BJ, Dai JZ, Cai PW. Gamma knife radiosurgery as a primary treatment for prolactinomas. J Neurosurg 2000;93 Suppl 3:10-13. Ma ZM, Qiu B, Hou YH, Liu YS. Gamma knife treatment for pituitary prolactinomas. Zhong Nan Da Xue Xue Bao Yi Xue Ban 2006 Oct;31(5):714-6.
72
73. Wolffenbuttel BH, Kitz K, Beuls EM. Beneficial gammaknife radiosurgery in a patient with Nelson’s syndrome. Clin Neurol Neurosurg 1998;100:60-3. 74. Pollock BE, Young WF Jr. Stereotactic radiosurgery for patients with ACTH-producing pituitary adenomas after prior adrenalectomy. Int J Radiat Oncol Biol Phys 2002;54:839-41. 75. Mauermann WJ, Sheehan JP, Chernavvsky DR, Laws ER, Steiner L, Vance ML. Gamma Knife surgery for adrenocorticotropic hormone-producing pituitary adenomas after bilateral adrenalectomy. J Neurosurg 2007;106(6):988-93. 76. Sheehan JP, Jagannathan J, Pouratian N, Steiner L. Stereotactic radiosurgery for pituitary adenomas: a review of the literature and our experience. Front Horm Res 2006;34:185-205. 77. Sheehan JP, Niranjan A, Sheehan JM, Jane JA Jr, Laws ER, Kondziolka D, Flickinger J, Landolt AM, Loeffler JS, Lunsford LD. Stereotactic radiosurgery for pituitary adenomas: an intermediate review of its safety, efficacy, and role in the neurosurgical treatment armamentarium. J Neurosurg 2005;102:678-91. 78. Abe T, Yamamoto M, Taniyama M, Tanioka D, Izumiyama H, Matsumoto K. Early palliation of oculomotor nerve palsy following gamma knife radiosurgery for pituitary adenoma. Eur Neurol 2002;47:61-3. 79. Chen JC, Giannotta SL, Yu C, Petrovich Z, Levy ML, Apuzzo ML. Radiosurgical management of benign cavernous sinus tumors: dose profiles and acute complications. Neurosurgery 2001;48:1022-30; discussion 1022–30. 80. Lindholm J. Cushing’s syndrome: historical aspects. Pituitary 2000;3:97-104. 81. Labeur M, Arzt E, Stalla GK, Paez-Pereda M. New perspectives in the treatment of Cushing’s syndrome. Curr Drug Targets Immune Endocr Metabol Disord 2004;4: 335-42. 82. Arnaldi G, Angeli A, Atkinson AB, Bertagna X, Cavagnini F, Chrousos GP, Fava GA, Findling JW, Gaillard RC, Grossman AB, Kola B, Lacroix A, Mancini T, Mantero F, Newell-Price J, Nieman LK, Sonino N, Vance ML, Giustina A, Boscaro M. Diagnosis and complications of Cushing’s syndrome: a consensus statement. J Clin Endocrinol Metab 2003;88:5593-602. 83. Ganz JC. Gamma knife radiosurgery and its possible relationship to malignancy: a review. J Neurosurg 2002;97:644-52. 84. Witt TC. Stereotactic radiosurgery for pituitary tumors. Neurosurg Focus 2003;14:e10. 85. Biochemical assessment and long-term monitoring in patients with acromegaly: statement from a joint consensus conference of the Growth Hormone Research Society and the Pituitary Society. J Clin Endocrinol Metab 2004;89:3099–102. 86. Giustina A, Barkan A, Casanueva FF, Cavagnini F, Frohman L, Ho K, Veldhuis J, Wass J, Von Werder K, Melmed S. Criteria for cure of acromegaly: a consensus statement. J Clin Endocrinol Metab 2000;85:526-9.
1189
1190
72
Radiosurgery for pituitary tumors
87. Melmed S, Casanueva F, Cavagnini F, Chanson P, Frohman LA, Gaillard R, Ghigo E, Ho K, Jaquet P, Kleinberg D, Lamberts S, Laws E, Lombardi G, Sheppard MC, Thorner M, Vance ML, Wass JA, Giustina A. Consensus statement: medical management of acromegaly. Eur J Endocrinol 2005;153:737-40. 88. Vance ML. Endocrinological evaluation of acromegaly. J Neurosurg 1998;89:499-500. 89. Laws ER, Vance ML, Thapar K. Pituitary surgery for the management of acromegaly. Horm Res 2000;53 Suppl 3:71-5. 90. Estrada J, Boronat M, Mielgo M, Magallon R, Millan I, Diez S, Lucas T, Barcelo B. The long-term outcome of pituitary irradiation after unsuccessful transsphenoidal surgery in Cushing’s disease. N Engl J Med 1997;336:172-7. 91. Jackson IM, Noren G. Gamma knife radiosurgery for pituitary tumours. Baillieres Best Pract Res Clin Endocrinol Metab 1999;13:461-9. 92. Pan L, Zhang N, Wang E, Wang B, Xu W. Pituitary adenomas: the effect of gamma knife radiosurgery on tumor growth and endocrinopathies. Stereotact Funct Neurosurg 1998;70 Suppl 1:119-26. 93. Ove R, Kelman S, Amin PP, Chin LS. Preservation of visual fields after peri-sellar gamma-knife radiosurgery. Int J Cancer 2000;90:343-50. 94. Stafford SL, Pollock BE, Leavitt JA, Foote RL, Brown PD, Link MJ, Gorman DA, Schomberg PJ. A study on the radiation tolerance of the optic nerves and chiasm after stereotactic radiosurgery. Int J Radiat Oncol Biol Phys 2003;55:1177-81. 95. Girkin CA, Comey CH, Lunsford LD, Goodman ML, Kline LB. Radiation optic neuropathy after stereotactic radiosurgery. Ophthalmology 1997;104:1634-43. 96. Tishler RB, Loeffler JS, Lunsford LD, Duma C, Alexander E III, Kooy HM, Flickinger JC. Tolerance of cranial nerves of the cavernous sinus to radiosurgery. Int J Radiat Oncol Biol Phys 1993;27:215-21. 97. Leber KA, Bergloff J, Pendl G. Dose-response tolerance of the visual pathways and cranial nerves of the cavernous sinus to stereotactic radiosurgery. J Neurosurg 1998;88:43-50. 98. Lundstrom M, Frisen L. Atrophy of optic nerve fibres in compression of the chiasm. Degree and distribution of ophthalmoscopic changes. Acta Ophthalmol (Copenh) 1976;54:623-40. 99. Rodriguez O, Mateos B, de la Pedraja R, Villoria R, Hernando JI, Pastor A, Pomposo I, Aurrecoechea J. Postoperative follow-up of pituitary adenomas after trans-sphenoidal resection: MRI and clinical correlation. Neuroradiology 1996;38:747-54. 100. Muron T, Rocher FP, Sentenac I, Marquis I, Romestaing P, Gatignon D, Croisille M, Gerard JP. Stereotactic radiosurgery. Preliminary experience of a team of Lyon. Ann Med Interne (Paris) 1993; 144:9-14.
101. Jane JA Jr, Vance ML, Woodburn CJ, Laws ER Jr. Stereotactic radiosurgery for hypersecreting pituitary tumors: part of a multimodality approach. Neurosurg Focus 2003;14:e12. 102. Cahan WG, Woodard HQ, Higinbotham NL, Stewart FW, Coley BL. Sarcoma arising in irradiated bone: report of eleven cases. 1948. Cancer 1998;82:8-34. 103. Hanabusa K, Morikawa A, Murata T, Taki W. Acoustic neuroma with malignant transformation. Case report. J Neurosurg 2001;95:518-21. 104. Kaido T, Hoshida T, Uranishi R, Akita N, Kotani A, Nishi N, Sakaki T. Radiosurgery-induced brain tumor. J Neurosurg 2001;95:710-13. 105. Shamisa A, Bance M, Nag S, Tator C, Wong S, Noren G, Guha A. Glioblastoma multiforme occurring in a patient treated with gamma knife surgery. Case report and review of the literature. J Neurosurg 2001; 94:816-21. 106. Comey CH, McLaughlin MR, Jho HD, Martinez AJ, Lunsford LD. Death from a malignant cerebellopontine angle triton tumor despite stereotactic radiosurgery. Case report. J Neurosurg 1998;89:653-8. 107. Shin M, Ueki K, Kurita H, Kirino T. Malignant transformation of a vestibular schwannoma after gamma knife radiosurgery. Lancet 2002;360:309-10. 108. Yu JS, Yong WH, Wilson D, Black KL. Glioblastoma induction after radiosurgery for meningioma. Lancet 2000;356:1576-7. 109. Horstmann GA, Van Eck AT. Gamma knife model C with the automatic positioning system and its impact on the treatment of vestibular schwannomas. J Neurosurg 2002;97:450-5. 110. Regis J, Hayashi M, Porcheron D, Delsanti C, Muracciole X, Peragut JC. Impact of the model C and automatic positioning system on gamma knife radiosurgery: an evaluation in vestibular schwannomas. J Neurosurg 2002;97:588-91.
63 What Every Neurosurgeon Should Know About Stereotactic Radiosurgery P. M. Black . F. Tariq
Stereotactic radiosurgery is the delivery of a large, highly focal stereotactically directed dose of radiation to destroy or modify a target. In neurosurgery and radiation oncology, it is one of the most important techniques available for the treatment of many brain lesions, including arteriovenous malformation and brain tumors. It may also have application to trigeminal neuralgia and epilepsy. There are several facts every neurosurgeon should know about radiosurgery: 1. 2. 3.
4. 5.
6.
7. 8. 9.
#
Radiosurgery is an important neurosurgical tool It has several forms, not all of which are equivalent It is best done as a collaborative effort between neurosurgeons and radiation oncologists There is a special body of knowledge associated with it It is effective treatment for many arteriovenous malformations and should be considered in their management routinely It is effective as primary or adjunct treatment for many meningiomas, schwannomas, and pituitary adenomas It is effective palliative treatment for some metastases and recurrent glioblastomas Radiosurgery can now be used for spinal lesions There are several other applications of radiosurgery being assessed today including treatment of functional disorders
Springer-Verlag Berlin/Heidelberg 2009
Radiosurgery is an Important Neurosurgical Tool In the last 30 years, there has been increasing interest in minimally invasive destruction of brain lesions. Radiosurgery is a major example of this interest. The use of radiation as an ablative energy has fundamentally changed neurosurgery. The pioneers in radiosurgery were neurosurgeons, sometimes working with radiation physicists, but always challenging the radiation oncology wisdom of the time. That wisdom proclaimed that only fractionation could be used to achieve a radiation response. Several developments occurred simultaneously to create the radiosurgery we know today. Lars Leksell, a Swedish neurosurgeon, developed the concept of multiple cobalt sources with convergence on a target, creating the gamma knife. His neurosurgical student Dade Lunsford brought this device to North America and helped to change neurosurgical history with it [1]. In Boston, Ray Kjellberg, a neurosurgeon at the Massachusetts General Hospital developed the concept of proton beam destruction of a lesion and actually carried out treatment for arteriovenous malformations and pituitary adenomas with great skill and innovative capacity [2–4]. The problem at this time was the targeting of the lesion. Kjellberg did in his pioneering work using angiograms for avm’s and skull films for pituitary adenomas to demonstrate that single fraction radiation could act as a destructive force, ablating a target as surely as suction or
978
63
What every neurosurgeon should know about stereotactic radiosurgery
dissection. His views were not accepted favorably by radiation oncologists. Also in Boston, a group at the Brigham and Women’s Hospital developed linac radiosurgery in the 1980s [5]. Drs. Ken Winston, Wendell Lutz, and William Saunders developed the concept of linac radiosurgery using non-coplanar arcs. This was also developed by Betti in Argentina, but the first North American treatments were done almost simultaneously at the Brigham [6]. The first stereotactic radiotherapy was also carried out by Drs Jay Loeffler, Eben Alexander, and Peter Black at Brigham and Women’s Hospital. It is extremely important for today’s neurosurgeon to understand and use wisely the potential of radiosurgery. Its power includes destructive possibilities that extend minimally invasive neurosurgical capacities significantly.
Radiosurgery has Several Forms Forms of radiosurgery include: 1.
Multiple gamma radiation emitters to focus on one central target – the Gamma knife [7,8]
. Figure 63-1 The first gamma knife model, 1982
2.
3.
Multiple co-planar arcs delivered with a conventional linear accelerator – ‘‘linac’’ radiosurgery [9,10]. A modification of this is the Cyberknife, which is a linear accelerator on a robotic arm [11]. Use of the Bragg Peak effect of protons to destroy a target with virtually no exit dose – the proton beam [2–4].
These are all important techniques and their relative efficacy and applications are still being developed. The Gamma knife appears to be most important for small lesions that can be well treated with one fraction. The vestibular schwannoma is a prototypical example of this but some skull base meningiomas such as cavernous sinus meningiomas or parasaggital meningiomas, and some metastases are also in this category (> Figure 63-1). The linear accelerator (linac) may possibly be useful for larger lesions than the Gamma knife. Its concept is that, by focusing wedges of radiation it is possible to destroy a target (> Figure 63-2).
What every neurosurgeon should know about stereotactic radiosurgery
. Figure 63-2 A linac radiosurgery system
63
these has an important role and each is critical to the enterprise as a whole. The radiologist uses imaging to identify the target; the neurosurgeon develops the target and understands what should not be radiated because of collateral damage; the radiation oncologist determines the safe dosage; the physicist helps to make the dose delivery accurate; and nurses provide critical patient support. Some recent developments have attempted to limit this collaborative effort, suggesting that neurosurgeons alone or radiation oncologists alone can do radiosurgery. This point of view is significantly limited, as the ideal program includes both groups.
Radiosurgery has a Special Body of Knowledge Associated with it The neurosurgeon who wishes to do radiosurgery should understand principles of radiation physics. These include: 1.
The proton beam has a totally different concept,using a natural property of protons. It has been particularly important for pediatric tumors in which controlling exit dose for the developing brain is extremely critical. It has also been useful for chordomas at the skull base (> Figure 63-3).
2.
3.
Radiosurgery is Best Done as a Collaborative Effort Radiosurgery is one of the great examples of collaboration in neurosurgery. Ideally, it involves joint work by neurosurgeons, radiation oncologists, physicists, radiologists, and nurses. Each of
The difference between early and late responding tissue. In general, tumor tissue is early responding because of dividing cells, and normal brain tissue is late responding Acute toxic effects: Headache, increased deficit if the target is close to brainstem or eloquent cortex, seizures if the target is in cortex or nausea if it is near the area postrema, are all possible short term effects of radiosurgery Planning Factors associated with toxicity: Size of target (Larger size requires smaller dose) Dose of radiation (safe doses have been established by toxicity data) Location of target (a lesion in an eloquent area may only be safely treated with a lower dose than a lesion in a ‘‘silent’’ brain area)
979
980
63
What every neurosurgeon should know about stereotactic radiosurgery
. Figure 63-3 The proton beam unit, which uses the Bragg peak effect of protons to minimize exit dose
4.
5.
Precision of focus of radiation (conformality) – gamma knife and proton are best at this Homogeneity of dose (linac and proton beam are good, gamma knife is less homogeneous) Patient factors associated with toxicity a) Prior treatment including chemotherapy, radiation, or surgery b) Underlying conditions such as diabetes, Nf2, nature of the lesion Late effects, which may include cyst formation as a lesion is destroyed, and new vascular lesions resembling cavernous angiomas; pseudo aneurysm, infarction and hemorrhagic stroke. Malignant conversion of benign lesions can be seen in some cases [4,5]
Radiosurgery is Effective Treatment for Many AVM’s Radiosurgery is an important tool for treatment of avms which are deep seated or located around
eloquent areas such as motor strip and brain stem. Symptomatic relief from seizure control and reduction in the likelihood of hemorrhage are achieved along with image defined volume reduction. Size, marginal dose, prior embolization and flow rate are the factors affecting prognosis [56,57]. Small sized avm’s (<15 ml) can be successfully treated with a higher single dose. Larger avms are best treated with staged volume radiosurgery. An avm with volume greater than 20 ml can be treated in two staged or three staged radiosurgery where separate anatomical volumes are targeted at one stage with doses up to 16 gy [59,60]. Five year survival data have shown marked volumetric reduction and stabilization of avm symptoms including seizures and bleeding rates. Reduction in the rate of bleeding depends on the Spetzler-Martin score, volume, and applied single dose. A volume greater than 4 cc, and score greater than 1.5 are associated with higher bleeding rate [58]. The natural course of most avm’s is a bleeding rate of 3% per year untreated; radiosurgery reduces this to 1.5% until the lesion is obliterated, when it almost reaches zero.
What every neurosurgeon should know about stereotactic radiosurgery
Reduction in seizure rate is also achieved with radiosurgery and is dependent on prior seizure score, size and volume of the avm’s. Prior embolization reduces the rate of obliteration after radiosurgery.
Radiosurgery is Important Treatment for Some Meningiomas, Vestibular Schwannomas, and Pituitary Adenomas Many centers have demonstrated that radiosurgery provides good control of benign meningioma with 10 year control rates of 90% and higher [12]. (Prior surgery, tumor grade, and size of lesion are important variables; if the tumor is larger than 3 cm, fractionated conformal therapy should be considered. Success in radiosurgery for meningiomas is defined by stopping growth rather than by shrinking the tumor. Radiosurgery can produce faster improvement in cranial nerve function than fractionated radiotherapy. Radiosurgery is particularly beneficial for lesions around the cavernous sinus and petroclival and deep seated skull base lesions where morbidity related to surgical intervention hampers complete resection and hence increases chances of recurrence. In these cases radiosurgery is often used as upfront treatment. For many other meningioma locations radiosurgery can add growth control to residual tumor after surgery. This is true for most skull base meningiomas and also parasagittal tumors. Control rate is 90% at 10 years and moribidity is 2– 3%, usually from cranial nerve deficit. Although much of the literature discusses the gamma knife as the radiosurgical technique, similar results are found with linac radiosurgery. There is not a great deal of experience with proton beam radiosurgery for meningiomas yet [12–17]. The change in practice over the last 10 years caused by advocates of radiosurgery in vestibular
63
schwannomas schwannoma management has been quite remarkable. For many patients with schwannomas less than 2 cm in diameter, radiosurgery is the upfront treatment of choice. Control rates are more than 90% over 10 years, and morbidity is less than with surgery (> Figures 63-4 and > 63-5). The risk of facial nerve weakness is now less than 3%; hearing loss for a tumor less than 2 cm is in the 5–10% range, and there is no likelihood of CSF leak. For patients with NF2 and bilateral vestibular schwannomas radiosurgery is also the treatment of choice because of the likelihood of preserving hearing. Finally, patients with recurrence of tumor after surgery often opt for this treatment. Complications increase if there has been prior surgery, if there are pre-existing deficits, or if volume is over 3 cm [18–23]. Radiosurgery is an important tool for the management of residual pituitary adenomas and in some cases for primary treatment of pituitary adenomas if the patient is medically unstable [24–26]. Contol rates in terms of tumor growth are above 90% at 10 years. For pituitary tumors with hormone production, there is a steady diminution in hormone overproduction for several years, with 40–60% control but rare normalization. There is also gradual panhypopituitarism for most patients. Often radiosurgery is used to control residual or recurrent tumor after trans-sphenoidal surgery. Some authors have recommended it as upfront treatment for Cushing’s Disease and aeromegaly. Control of tumor progression is achieved in 91–100% of the cases along with clinical improvement and a very low recurrence rate. Normalization of hormone status is achieved within 6–12 months. Surgery and medical treatment are the first line therapy for patients with acromegaly; the rates of remission for surgery are 60–90% and for medical therapy 30%. Residual, recurrent and refractory cases can be successfully treated with radiosurgery.
981
982
63
What every neurosurgeon should know about stereotactic radiosurgery
. Figure 63-4 Tumor control in vestibular schwanoma after radiosurgery [9]
Radiosurgery is Useful Palliation for Some Metastases and Recurrent Glioblastomas Brain metastases occur in 20–40% of all cancer patients and constitute about 10% of all intracranial tumors. The commonest primaries are the lung, breast, gastrointestinal tract, genitourinary tract and skin. Radiosurgery is a valuable tool for small, multiple, and deep brain metastases with the advantage of being a 1 day outpatient procedure performed under local anesthesia. In general, it provides 90% tumor control for the lifetime of the patient; most patients die of their systemic disease [27–32]. Survival after radiosurgery for metastases depends on 1) 2) 3)
Age Number of metastases Extent and activity of disease outside the brain
4)
5)
6)
Histology (germ cell tumors are more responsive than breast carcinomas, which are more responsive than lung carcinomas, which are more responsive than melanoma) Performance status at the onset of treatment – a higher performance scale at the start of treatment better produces better post treatment results Whether whole brain therapy is added it lessens recurrence and other tumor occurrence.
In choosing radiosurgery versus surgery for metastases, patients who are over 70 years of age and have significant medical problems are better treated with stereotactic radiosurgery. Younger patients with Karnofsky performance scale over 70 and no extracranial disease, also do well with radiosurgery. Studies have shown that radiosurgery is as effective as surgical excision for a single brain metastasis [27–32].
What every neurosurgeon should know about stereotactic radiosurgery
63
. Figure 63-5 MRI imaging appearance of a vestibular schwanoma before (a) and after radiosurgery
For glioblastoma multiforme, the role of radiosurgery is less clear than for metastases. Surgical debulking with adjuvant radiotherapy and chemotherapy is treatment of choice for glioblastomas. Stereotactic radiosurgery may be a useful option for well circumscribed recurrent lesions [32–37]. Post treatment radiation induced necrosis is a limiting factor; it is much more prevalent in glioblastoma than in metastases.
Radiosurgery can now be Used for Some Spinal Lesions Recent developments such as the Cyberknife and the Novalis spinal radiosurgery system have made the spine a possible target for radiosurgery [38– 44]. This is a result of the capacity to do frameless image-guided targeting. Spinal column metastases, intramedullary masses such as avm’s and ependymomas, and some spinal chordomas and chordrosarcomas are now good target lesions for radio surgery. This will likely change spine surgery as substantially as it did cranial radiosurgery. Radiosurgery is an extremely important tool for spinal avm’s, especially in instances where
the angioarchitecture makes surgery difficult. Myelopathy secondary to ischemia can occur but the chances are lessened when the dose is calculated carefully. Image based volume reduction becomes evident between 24 and 36 months with marked improvement in symptomatology. Immediate pain control, improvement in neurological status, and radiological regression can be achieved with minimal morbidity and significant success rates. In patients with metallic spine instrumentation and prior radiation, conventional imaging may not be useful. F-flurodeoxyglucose PET can be used for radiosurgery treatment planning and to assess response.
Radiosurgery is Being Evaluated for a Number of Functional Disorders Radiosurgery has significantly changed tumor and vascular neurosurgery for the past two decades. Recently it is being considered for functional disorders including pain syndromes, epilepsy and behavioral disorders [45–57]. Current
983
984
63
What every neurosurgeon should know about stereotactic radiosurgery
targets include the trigeminal nerve (for trigeminal neuralgia), cingulate gyrus and internal capsule (for pain disorders and obsessive compulsive disorder), globus pallidus (for relief of Parkinson’s disease) and the corpus callosum and medial temporal lobe (for epilepsy). Treatment for trigeminal neuralgia has shown promising results with remission in many cases. Higher dose is recommended for pain control, up to 90 Gy. Higher success rate is associated with patients who are responsive to drugs, who have primary SRS or who have no prior surgery done.
Summary Radiosurgery has completely changed neurosurgery and radiation oncology in the past 30 years. It has been demonstrably effective in controlling many arteriovenous malformations, meningiomas, vestibular schwannomas, pituitary adenomas, cerebral metastases, and malignant gliomas. It is now being applied to the spine and to several functional disorders. Neurosurgeons must be aware of this important modality and guide its further development.
References 1. Leksell L. Stereotactic radiosurgery. J Neurol Neurosurg Psychiatry 1983;46:797-803. 2. Kjellberg RN, Hanamura T, Davis KR, Lyons SL, Adams RD. Bragg-peak proton-beam therapy for arteriovenous malformations of the brain. N Engl J Med 1983;309(5):269-74. 3. Kjellberg RN, Kliman B. Bragg peak proton treatment for pituitary-related conditions. Proc R Soc Med 1974;67 (1):32-3. 4. Kjellberg RN, Shintani A, Frantz AG, Kliman B. Protonbeam therapy in acromegaly. N Engl J Med 1968;278 (13):689-95. 5. Alexander E, III, Loeffler JS. The case for radiosurgery. Clin Neurosurg 1999;45:32-40. 6. Betti OO. History of radiosurgery. Neurochirurgie 1972;18(3):235-65.
7. Kondziolka D, Lunsford LD, Flickinger JC. The application of radiosurgery to disorders of brain. Neurosurgery 2008;62 Suppl 2:SHC 707-SHC7 19; discussion SHC 719–SHC 720. 8. Kondziolka D. Stereotactic radiosurgery: what’s turning people on. Clin Neurosurg 2007;54:23-5. 9. Loeffler JS, Neiemierko A, Chapman PH. Second tumors after radiosurgery: tip of the iceberg or a bump in the road? Neurosurgery 2003;52:1436-40. 10. Friedman WA, Foote KD. Linear accelerator radiosurgery in the management of brain tumors. Ann Med 2000; 32:64-80. 11. Adler JR, Colombo F, Heilbrun MP, Winston K. Toward an expanded view of radiosurgery. Neurosurgery 2004;55:1374-6. 12. Kondziolka D, Mathien D, Lunsford LD, Martin JJ, Madhok R, Niranjan A, Flickinger JC. Radiosurgery as a definitive treatment for management of intracranial meningiomas. Neurosurgery 2008;62(1):53-8; discussion 58–60. 13. Pollock BE, Stafford SL, Utter A, Giannini C, Schreiner SA. Stereotactic radiosurgery provides a equivalent control to Simpson Grade 1 resection for patients with small to medium sized meningiomas. Int Radiat Oncol Biol Phys 2004;55(4):1000-50. 14. Black PM, Villavicencio AT, Rhouddou C, Loeffler JS. Aggressive surgery and focal radiation in the management of meningiomas of the skull base: preservation of function with maintenance of local control. Acta Neurochir (Wien) 2001;143:555-62. 15. Kondziolka D, Flickinger JC, Lunsford D. Principles of skull base radiosurgery. Neurosurg 2008;24(5):E11. 16. Kollova A, Lisca KR, Viadyka V, Simonova G, Janouskova L.Gamma knife radiosurgery for benign meningiomas. J Neurosurg 2007;107(2):325-36. 17. Pamir MN, Peker S, Kilic T, Sangoz M. Efficiency of gamma knife radiosurgery for treatment of meningiomas including superior sagittal sinus. Zentralbl Neurochir 2007;68(2):73-8. 18. Okida Y, Kobayashi T, Tanaka T, Mori Y. Radiosurgery for bilateral neurinomas associated with neurofibromatosis type 2. Surg Neurol 2000;53(4):383-89; discussion 389–90. 19. Pellet W, Regis J, Roche PH, Delsanti C. Relative indication for radiosurgery and microsurgery for acoustic schwanomas. Adv Tech Stand Neurosurg 2003;28:227-82; discussion 282–4. 20. Chan AW, Black P, Ojemann RG, et al. Stereotactic radiotherapy for vestibular schwannomas; favorable outcome with minimal toxicity. Neurosurgery 2005; 57:60-70. 21. Oqunrinde OK, Lunsford DL, Bissonette DJ, Flickinger JC. Cranial nerve preservation with stereotactic radiosurgery of intracanalicular tumors. Stereotact Funct Neurosurg 1995;64 Suppl 1:87-97. 22. Pollock BE. Vestibular schwanoma management: an evidence based comparison of stereotactic radiosurgery
What every neurosurgeon should know about stereotactic radiosurgery
23.
24.
25.
26.
27.
28.
29.
30. 31.
32.
33.
34.
35.
36.
and microsurgical resection. Prog Neurol Surg 2008; 21:222–27. Kondziolka D, Lunsford LD. Future prospective in acoustic neuroma management. Prog Neurol Surg 2006; 21:247-54. Mitsunori M, Shrieve DC, Alexander E. III, Kaiser UB, Richardson GE, Black PM, Loeffler JS. Initial clinical results of LINAC-based stereotactic radiosurgery and stereotactic radiotherapy for pituitary adenomas. Int J Radiat Oncol Biol Phys 1998;42(3):573-80. Pollock BE, Brown PD, Nippoldt TB, Young WF, Jr. Pituitary tumor type affects the chances of biochemical remission after radiosurgery of hormone secreting pituitary adenomas. Neurosurgery 2008;62(6): 1271-6; discussion 1276–8. Devin JK, Alelek GS, Cmelak AJ, Blevins LS. The efficacy of linear accelerator radiosurgery in patients with Cushings Disease. Stereotactic Funct Radiosurg 2004; 82(5–6):254-62. Brown PD, Brown CA, Pollock BE, Gorman DA, Foote RL. Stereotactic radiosurgery for patients with radiosensitive brain metastasis. Neurosurgery 2008;62 Suppl 2:656-65; discussion 665–7. Monaco E, III, Kondziolka D, Mongia S, Niranjan A, Flickinger JC, Lunsford LD. Management of brain metastasis from ovarian and endometrial carcinoma with stereotactic radiosurgery. Cancer 2008;113:2610–4. Alexander EA, III, Alexander E, Moriarty TM, Davis RB, Wen PY, Fine HA, Black PM, Kooy HM, Loeffler JS. Stereotactic radiosurgery for the definitive, noninvasive treatment of brain metastases. J Natl Cancer Inst 1995;87:34-40. Kaal EC, Niel CG, Vecht CJ. Therapaeutic management of brain metastasis. Lancet Neurol 2005;4(5):289-98. Jaggannathan J, Sherman JH, Mehta GU, Chin LS. Radiology of brain metastasis: applications in stereotactic radiosurgery. Neurosurg Focus 2007;22(3):E4. Chernov MF, Nakaya K,Izawa M, Hayashi M, Usuba Y, Kato K, Muragaki Y, Iseka H, Hori T, Takakura X. Outcome after radiosurgery for brain metastasis in patients with low Karnofsky Performance Scale(KPS)scores. Int J Radiat Oncol Bio Phys 2007;67(5). Kong DS, Lee JI, Park K, Kim JH, Lim D, Nam DH. Efficiency of stereotactic radiosurgery as a salvage for treatment for recurrent malignant gliomas. Cancer 2008;112(9):2046-51. Mahajan A, McCutcheon IE, Suki D, Chang EL, Hassenbusch SJ, Weinberg JS, Shiu A, Maro MH, Woo SY. Case control study of stereotactic radiosurgery for recurrent glioblastoma multiforme. J Neurosurg 2005;103(2):210-7. Wang YY, Yang GK, Li SY, Baol XF, Wu CY. Prognostic factors for deep situated malignant gliomas treated with linac radiosurgery. Chin Med Sci J 2004;19(2):105-10. Cho KH, Hall WA, Gerbi BJ, Higgins PD, McGuire WA, Clark HB. Single dose versus fractionated stereotactic
37.
38.
39.
40.
41.
42.
43.
44.
45. 46.
47.
48.
49.
50.
51.
63
radiotherapy for recurrent high grade gliomas. Int Radiat Oncol Biol Phys 1999;45(5):1133-41. Schwer AL, Damek DM, Kavanagh BD, Gaspar LE, Lillehei K, Stuhr K, Chen C. A phase 1 dose – escalation study of fractionated stereotactic radiosurgery in combination with geftinib in patients with recurrent malignant gliomas. Int J Radiat Oncol Biol Phys 2008; 70(4):993-1001. Dodd RL, Ryu MR, Kamnerdsupaphon P,Gibbs IC, Chang SD, Jr, Adler JR, Jr. Cyber knife radiosurgery for benign intradural extramedullary spinal tumors. Neurosurgery 2006;58(4):674-85; discussion 674–85. Gerszten PC, Burton SA, Ozhasglu C, McCune KJ, Quinn AE. Radiosurgery for benign intradural spinal tumors. Neurosurgery 2008;62(4):887-95; discussion 895–6. Ryu S, Jin R, Jin JY, Chen Q, Rock J, Anderson J, Movsa B. Pain control by image guided radiosurgery for solitary spinal matastasis. J Pain Symptom Manage 2008; 35(30):292-8. Sinclair J, Chang SD, Gibbs IC, Adler JR, Jr. Multisession Cyberknife radiosurgery for intramedullary spinal cord arteriovenous malformations. Neurosurgery 2006; 58(6):1081-9; discussion 1081–9. Mcloughlin GS, Sciubba DM, Wolinsky JP. Chondroma/ Chondrosarcoma of the spine. Neurosurg Clin N Am 2008;19(1):57-63. Gwak HS,Yoo HJ,Youn SM, Chan U, Lee DH,Yoo SY, Rhee CH. Hypofractionated stereotactic radiation therapy for skull base and upper cervical chordoma and chordosarcoma: preliminary results. Stereotact Funct Neurosurg 2005;83(5–6):233-43. Gerszten PC, Burton SA. Clinical assessment of stereotactic IGRT: spinal radiosurgery. Med Dosim 2008; 33(2):107-16. Kondziolka D. Functional radiosurgery. Neurosurgery 1999;44(1):12-20; discussion 20–2. Forcadas-Berdusan M. Indications and results of nonpharmacological treatments of epilepsies: vagal stimulation, ketogenic diet and gamma rays. Rev Neurol 2002;35 Suppl 1:S144-50. Friehs GM, Park MC, Goldman MA, Zerris VA, Noren G, Sampath P. Stereotactic radiosurgery for functional disorders. Neurosurg Focus 2007;23(6):E3. Regis J, Bartolomei F, Hayashi M, Chauvel P. What is the role of radiosurgery in mesial temporal lobe epilepsy. Zentralbl Neurochir 2002;63(30):101-5. House PA, Kim JH, Lanerolle ND, Barbaro NM. Radiosurgery in epilepsy-pathological considerations. Prog Neurol Surg 2007;20:279-88. Eder HG. Gamma knife radiosurgery for callosotomy in children with drug resistant epilepsy. Childs Nerv Syst 2006;22(8):1012-7. Gellner V, Kurschel S, Kreil W, Holl E, Ofner-Kopening, Unger F. Recurrent trigeminal neuralgia: long term
985
986
63 52.
53.
54.
55. 56.
What every neurosurgeon should know about stereotactic radiosurgery
outcome of repeat gamma knife radiosurgery. J Neurosurg Pscychiatry 2008. Guo S, Chao ST, Reuther AM, Barnett GH, Suh JH. Review of the treatment of trigeminal neuralgia with gamma knife radiosurgery. Stereotactic Funct Neurosurg 2008;86(3):135-46. Chen JC, Grethouse HE, Girvigian MR, Miller MJ, Liu A, Rahimian J. Prognostic factor for radiosurgery treatment of trigeminal neuralgia. Neurosurgery 2008; 62(5) Suppl:A53-60. Ruck C, Andreewitch S,Flyckt K, Edman G, Nyman H, Meyerson BA, Lippitz BE, Hindmarsh T, Svanborg P, Mindsu P, Asberg M. Capsulotomy for refractory anxiety disorders: long term follow-up of 26 patients. Am J Psychiatry 2003;160(3):513-21. Kim MC, Lee TK. Stereotactic lesioning for mental illness. Acta Neurochir Suppl 2008;101:39-43. Maratuza RL, Chiocca EA, Jenike MA, Giriunas IE, Ballantine HT. Stereotactic radiofrequency thermal
57.
58.
59.
60.
cingulotomy for obsessive compulsive disorder. J Neuropsychiatry Clin Neurosci 1990;2(3):331-6. Kondziolka D, Hudak R. Management of obsessive compulsive disorder related skin picking–gammaknife radiosurgery anterior capsulotomies: a case report. J Clin Pscychiatry 2008;69(8):1337-40. Fukoka S, et al. Radiosurgery for arteriovenous malformations with Gamma knife: a multivariaqte analysis of factors affecting the complete obliteration rate. J Clin Neurosci 1998;5 Suppl:18-71. Pan DH, Guo WY, Chung WY, Shiau CY, Chang YC, Wang LW. Gamma knife as a single treatment modality for large arteriovenous malformations. J Neurosurg 2000;93 Suppl 3:113-9. Sirins S, Kondziolka D, Niranjan A, Lunsford LD, Flickinger JC, Maitz JC. Prospective staged volume radiosurgery for large arteriovenous malformations: indications and outcome in otherwise unpredictable patients. Neurosurgery 2008;62 Suppl 2:17-26; discussion 26–7.
74 Whole Body and Spinal Radiosurgery P. C. Gerszten
Introduction Over the past decade, minimally invasive surgical techniques have evolved in the field of spine surgery [1]. Such techniques follow a natural trend in surgery to minimize injury to normal tissue while obtaining the same or better surgical outcome. In a similar fashion, there has been substantial interest in applying minimally invasive techniques to the field of spine oncology. Malignancy involving the spine is an important clinical problem in oncology. Although primary tumors of the spine are relatively rare, they are typically very symptomatic and difficult to treat [2]. Secondary malignancy of the spinal column is extremely common. In a study of 2,000 patients with bony metastases, nearly 70% were found to have vertebral body metastases [3]. There are over 180,000 new cases of spinal metastases diagnosed in North America each year, with 20,000 clinical cases of spinal cord compression [2,4,5]. The incidence and prevalence of spine tumors is expected to rise in the future. Standard treatment options for spinal metastases include radiotherapy alone, radionuclide therapy, radiotherapy plus systemic therapy or surgical decompression and/or stabilization followed by radiotherapy [6]. The role of radiation therapy in the treatment of metastatic tumors of the spine is well established and is often the initial treatment modality [6–12]. The goals of local radiation therapy in the treatment of spinal tumors have been palliation of pain, prevention of local disease progression and subsequent pathologic fractures, and halting progression of or #
Springer-Verlag Berlin/Heidelberg 2009
reversing neurological compromise [13]. Surgery is usually reserved for spinal instability or subluxation, neurological deficits despite other forms of therapy and intractable pain attributable to an isolated lesion. During the past decade, surgical spinal oncology has focused on new surgical approaches to the spine, the application of new instrumentation for spinal reconstruction, various forms of radiation delivery systems, and most importantly, complication avoidance. Patients with metastatic spine tumors are often debilitated and at a high risk for surgical morbidity [14]. For patients with limited life expectancies from their underlying disease, the impact of surgical complications with subsequent decrease in quality of life is not ideal. A randomized trial has found a significant benefit with the use of surgical decompression with radiation therapy over radiation alone for high-grade epidural spinal cord compression [15]. However, for the vast majority of patients who do not have highgrade epidural spinal cord compression, or who do not have other indications for surgery, management remains controversial [2]. The primary factor that limits radiation dose for local vertebral body and parspinal tumor control with conventional radiotherapy is the relatively low tolerance of the spinal cord to radiation. Conventional external beam radiotherapy lacks the precision to deliver large singlefraction doses of radiation to vertebral tumors near radiosensitive structures such as the spinal cord. It is the low tolerance of the spinal cord to radiation that often limits the treatment dose to a level that is far below the optimal therapeutic
1204
74
Whole body and spinal radiosurgery
dose [7,16,17]. Therefore, spinal tumors often progress or recur after radiation therapy because of insufficient doses. Precise confinement of the radiation dose to the treatment area, as is the case for intracranial radiosurgery, should increase the likelihood of successful tumor control and clinical response at the same time that the risk of adjacent neurological injury is minimized [17–27]. Spinal radiosurgery offers highly conformal radiation therapy that provides tumorcidal doses of radiation while reducing toxicity to the spinal cord in selected patients. Recent technological innovations in radiotherapy delivery allow for extracranial radiosurgery to be both feasible as well as practical.
Background of Spine Radiosurgery Radiosurgery is defined as the delivery of a highly precise, large radiation dose to a localized area by a stereotactic approach [28]. Radiosurgery is currently widely employed as an effective means for controlling benign and malignant intracranial tumors [29–35]. Conventional frame-based devices used for stereotactic radiosurgery for intracranial lesions use a rigid frame to immobilize the lesion at a known location in space. Stereotactic radiosurgery has been demonstrated to be an effective treatment for brain metastases, either with or without whole-brain radiation therapy, with an 85–95% control rate [36,37]. In the past, stereotactic radiosurgery was limited to intracranial disease because precise localization could be achieved only by surgical frames fixed to the patient’s skull. The frame acts as a fiducial reference system to provide accurate targeting and delivery of the radiation dose using multiple beams of radiation. Intracranial radiosurgery is practical because the lesions are fixed with respect to the cranium, which can be easily immobilized rigidly in a stereotactic frame. As a corollary, treatment is typically limited to single fraction treatments. For spinal applications, if
the radiation dose could be confined more precisely to the treatment volume, as is the case for intracranial radiosurgery, one would be able to deliver high doses of radiation, potentially increasing the likelihood of successful tumor control. Spinal lesions have a fixed relationship to one or more vertebral segments. However, stereotactic radiosurgery techniques developed for spinal lesions over a decade ago using standard linear accelerators required the placement of an invasive rigid external frame system directly to the spine and were therefore not adopted for general use [38]. The Hamilton-Lulu extracranial stereotactic frame for extracranial stereotactic localization employed skeletal fixation via the spinous processes to immobilize the target area and employed noncoplanar radiation beams such as those routinely employed in LINAC-based cranial radiosurgery [39]. The Lax extracranial stereotactic frame developed at Karolinska, Sweden covered both the head and the body down to the level of the midthigh [40,41]. A series of indicators mounted on the frame were visualized during CT imaging and thus defined the stereotactic space enclosing the patient and the target lesion. The patient was maintained within this stereotactic space by the use of a vacuum pillow or foam pad. To deliver the stereotactic radiation, a series of scales were mounted on the frame in the same position that the stereotactic fiducial markers were located during imaging data acquisition. The target would be treated using four to eight non-coplanar beam that were further shaped by employing secondary blocks in the line of the beams. Since Hamilton et al. [42] first described the possibility of linear-accelerator based spinal stereotactic radiosurgery in 1995, multiple centers have attempted to pursue large fraction conformal radiation delivery to spinal lesions using a variety of technologies [2,17,19–22,25,26,28, 42–51]. The emerging technique of spine radiosurgery represents a logical extension of the current
Whole body and spinal radiosurgery
state-of-the-art radiation therapy. Stereotactic radiosurgery for tumors of the spine has more recently been demonstrated to be accurate, safe and efficacious [17,19,20,22,25,26,28,42,44,46,47, 49,51]. Recent technological developments, including imaging technology for three-dimension (3-D) localization and treatment planning, the advent of intensity-modulated radiotherapy (IMRT), and a higher degree of accuracy in achieving target dose conformation while sparing normal surrounding tissue, have allowed clinicians to expand radiosurgery applications to treat malignant and benign lesions of the paraspinal region, vertebral bodies, as well as intradural and even intramedullary locations.
Spinal Cord Tolerance to Radiation and Radiosurgery Radiation-induced spinal cord injury, or myelitis, is one of the most dreaded complications related to spinal radiation. There is little clinical experience regarding the tolerance of the human spinal cord to large single fraction doses, and the tolerance of the spinal cord to a single dose of radiation has not been well defined [26,52,53]. Animal studies support the notion of a dosevolume, i.e., dose-length, radiation effect in spinal cord [54]. However, available clinical data after radiation therapy fail to support such a phenomenon. This fact may be due to the relatively large volume (or length) of spinal cord irradiated in conventional radiotherapy. Therefore, one must still rely upon clinical data derived from external beam irradiation series in which the entire thickness of the spinal cord was irradiated. The total dose (TD) 5/5 (the total dose at which there is a 5% probability of myelitis necrosis at 5 years from treatment) for 5-cm, 10-cm and 20-cm lengths of the spinal cord treated with standard fractionation has been estimated by Emami et al. [55] as 5 Gy, 5 Gy and 4.7 Gy, respectively. These dose levels are estimates
74
based upon extrapolations of data sets that date back to 1948. Emami determined that the spinal cord radiation tolerance varied narrowly between 50 Gy and 47 Gy when one-third versus the entire spinal cord was irradiated, respectively. These estimations have been widely adopted in clinical practice. To minimize the risk of spinal cord necrosis, the radiation tolerance with standard fractionation has been traditionally stated to be 45–50 Gy. A dose of 45–50 Gy in standard fractionation (1.8–2 Gy per fraction) is within the radiation tolerance of the spinal cord (TD 5/5). A single treatment of 8 Gy delivered to a long segment of the spinal cord has been given without reported myelopathy [56,57]. In a review of 172 patients treated with fractionated radiotherapy to the cervical and thoracic spine at the University of California, San Francisco (total dose of 40–70 Gy fractionated over a 2–3 week period), Wara et al. [58] reported nine cases of radiation-induced myelopathy. Three out of nine patients had mild cervical cord neurological deficits without any significant long-term symptoms. The length of the spinal cord that was exposed to radiation averaged from 4 to 22 cm. Hatlevoll et al. [59] reported a series of 387 patients with bronchial carcinoma treated with a split-course regimen using large single fractions. Seventeen patients developed radiation myelitis with average total dose of 38 Gy. Kim et al. [8] reported seven patients with transverse myelopathy from a group of 109 patients treated with definitive radiotherapy (standard fractionation) for head and neck cancers to a total dose of 57–62 Gy with an average field size of 10 by 10 cm. Abbatucci et al. [60] reported 8/203 cases of radiation-induced myelopathy with a total radiation dose of 54–60 Gy to the cervical and thoracic spine. McCunniff et al. [61] reported only one case of radiation myelopathy out of 652 patients who had received greater than 60 Gy using standard fractionation. Phillips et al. [62] reported three cases of transverse myelitis in
1205
1206
74
Whole body and spinal radiosurgery
350 patients treated with tumors to the chest to a total radiation dose of 33–43.5 Gy.
Treatment Planning and Dose Prescriptions The spinal radiosurgery procedure can be divided into four distinct components: (1) immobilization and/or fiducial implantation for image guidance, (2) computed tomography (CT) imaging for treatment planning and generation of digitally reconstructed radiographs (DRR), (3) radiation dose planning and (4) the actual dose delivery. Spinal radiosurgery may be performed entirely in an outpatient setting. The treatment prescription includes quantifications of both volume and dose. The clinical target volume (CTV) includes the gross tumor volume (GTV) and the adjacent anatomical areas with a high likelihood of tumor involvement. Both the CTV and the critical structures for which radiation is to be avoided are identified for planning. The lesion (GTV) is outlined based on CT imaging or from a magnetic resonance (MR) fusion capability. In other words, the lesion may be contoured directly on an MR image that may provide better anatomical resolution. Each spine radiosurgical treatment plan is devised jointly by a team comprised of a surgeon, a radiation oncologist and a radiation physicist. In each case, the radiosurgical treatment plan is designed based upon tumor geometry, proximity to spinal cord and location of the tumor. The prescribed tumor dose is determined based upon the histology of the tumor, spinal cord volume and previous radiation exposure to normal tissue, especially the spinal cord. No large experience with spine radiosurgery has yet resulted in guidelines for the optimal doses and dose constraints. Some centers have administered doses of 6–30 Gy in one to five fractions [20,21,26,63–65]. At the University of Pittsburgh, we chose to use single-fraction radiosurgery
technique as opposed to fractionated therapy because of our large experience with intracranial radiosurgery using the Leksell Gamma Knife. Given the good clinical response as well as the lack of adverse consequences to normal tissue, including the spinal cord, we have continued to employ a single-fraction treatment paradigm for our spinal radiosurgery program. For single-fraction therapy, the tumor dose is maintained at 12–20 Gy to the 80% isodose line contoured at the edge of the target volume. The maximum intratumoral dose ranges from 15 to 30 Gy. A maximum tumor dose of 20 Gy or 16 Gy to the tumor margin appears to provide good tumor control with little risk of radiation induced spinal cord or cauda equina injury [46]. Dose and fractionation schedules vary widely. The Memorial Sloan Kettering group used a maximum does of 20 Gy delivered in 5 fractions [20,65]. The Georgetown group used a mean dose of 21 Gy in 3 fractions [37]. The MD Anderson group used 30 Gy in five fractions [21,27]. The Henry Ford Hospital group used 6–8 Gy boost after conventional irradiation (25 Gy in 10 fractions) [26]. Finally, DeSalles reported a mean dose of 12 Gy [22]. For each spinal radiosurgery case, the spinal cord and/or cauda equina is outlined as a critical structure. At the level of the cauda equina, the spinal canal is outlined. Therefore, at the level of the cauda equina, the critical volume is the entire spinal canal and not actual neural tissue. In terms of single-fraction radiosurgery, it is recommended that the maximum spinal cord dose be kept below 10 Gy. For the cauda equina, a safe maximum dose is probably somewhat higher, perhaps up to 14 Gy [53]. Although these dose limits are likely conservative estimates of true spinal cord and cauda equina tolerance, given the catastrophic nature of radiation induced myelitis, one should err on the side of caution. A limit of 2 Gy is set as the maximum dose to each of the kidneys. This especially becomes important in the treatment of lower thoracic
Whole body and spinal radiosurgery
and lumbar vertebrae, even more so if the patient has undergone a nephrectomy or received nephrotoxic chemotherapy. A limit of 8 Gy is set as the maximum dose to the bowel. The above normal tissue constraints are used in conjunction with the desired prescription dose to perform treatment planning. The physicist used this desired dose distribution to work backwards (inverse treatment planning) to design a field setup (beam angles) and beam intensities (intensity modulated radiotherapy) to accomplish these goals.
Image Guidance Techniques The two major difficulties that must be overcome with spine radiosurgery are issues of target localization and target immobilization. Reliable and accurate patient setup is essential, and various approaches have been used to position and immobilize patients as well as deliver radiation to the spine in a radiosurgical manner [56]. Hamilton et al. [42] adopted an approach similar to that used for intracranial radiosurgery in which an external reference system is rigidly attached to the patient’s bony anatomy. The patient, lying in a prone position, underwent CT imaging with the frame attached, during which the target was delineated and the isocenter located within the coordinate system of the frame. The frame then was transferred to the linear accelerator couch and the immobilized patient underwent treatment. Since the initial experience with frame-based spine radiosurgery, a variety of successful frameless techniques have been developed. It is essential that the patient remain stationary in the correct position throughout dose delivery [2]. The techniques used can be grouped into two approaches. In the first, the patient is immobilized in a stereotactic body frame or immobilization cradle. Such devices are noninvasive and do not guarantee that the patient will remain
74
perfectly positioned. Nonetheless, good immobilization results, obtained by taking pre- and posttreatment imaging, have been reported for the devices currently in use. A second approach is used when localization images can be taken frequently during the actual treatment without greatly lengthening the procedure. It is especially suited to procedures in which dual in-room kilovolt units are available, such is with the CyberKnife. At MD Anderson Cancer Center, a near-simultaneous CT image-guided stereotactic radiotherapy system was developed that integrates a CT-on-rails scanner with a linear accelerator. Patients are immobilized in a supine position by a moldable body cushion vacuum wrapped with a plastic fixation sheet. Patients are transferred directly following CT planning imaging to the linear accelerator couch for the treatment using a rail system [27]. The Memorial Stereotactic Body Frame (MSBF) was developed at Memorial SloanKettering Cancer Center and also utilizes a form of external immobilization using a series of pressure plates. Patients are scanned before each treatment on a CT unit that has been installed in the same room as the linear accelerator. When used in a purely stereotactic fashion, the use of a noninvasive body frame cannot guarantee that the target is positioned within the tight tolerances required to avoid spinal cord irradiation. For this reason, some form of image guidance is required at the time of treatment. When using the MSBF, the pretreatment CT scan taken at 2 mm slice thickness is automatically registered with the planning CT scan using bony landmarks. A second registration of the fiducial system in the treatment and planning scans allows the setup error of the patient in the body frame to be determined. A final verification of the treatment position is done from the comparison of cone beam CT and orthogonal portal images with digitally reconstructed radiographs from the initial planning CT study [65].
1207
1208
74
Whole body and spinal radiosurgery
Innovations in image guidance greatly improved the ability with which bone or some radiographic landmark could be visualized in the patient in the treatment position. Most of these techniques involve the use of kilovoltage imaging. One such approach, developed by John Adler and colleagues at Stanford University, uses two kilovolt sources mounted in the room. The CyberKnife Image-Guided Radiosurgery System1 (Accuray, Inc., Sunnyvale, CA) consists of a 6 mV compact linear accelerator that is smaller and lighter in weight than linear accelerators used in conventional radiotherapy (> Figure 74-1) [66,67]. The smaller size allows it to be mounted on a computer-controlled, sixaxis robotic manipulator that permits a much wider range of beam orientations than can be achieved with conventional radiotherapy devices. . Figure 74-1 The CyberKnife Radiosurgical System (Accuray, Inc., Sunnyvale, CA). Note the two amorphous silicon X-ray screens positioned orthogonally to the treatment couch. The couch can move to position the fiducials in front of the cameras. A radiographic landmark is tracked using near real-time image guidance. The measured position as seen by both cameras is communicated through a real-time control loop to a robotic manipulator that directs the beam to the precise intended target
Two diagnostic X-ray cameras are positioned orthogonally (90 offset) to acquire real time images of the patient’s internal anatomy during treatment. The images are processed to identify radiographic features (skull bony landmarks or implanted fiducials) and then automatically compared to the patient’s CT treatment planning study. The precise tumor position is communicated through a real-time control loop to a robotic manipulator that aligns the radiation beam with the intended target [47,68–70]. The patient can be set up conventionally on the treatment couch without additional immobilization. The positions of the bony structures are checked many times during treatment; thus, any shifts in position can be detected and quickly corrected during the delivery of treatment [2]. Extracranial radiosurgery requires the delivery of precisely shaped radiation beams. A complete description of the radiation physics involved in treatment planning and delivery is beyond the scope of this chapter. In a linear accelerator, electrons are accelerated to high energies and exit as either electrons or are aimed at a target designed to produce high energy X-rays or photons. Radiosurgery treatments utilize the later. A rotatable gantry allows 360 rotation of the source to allow for multiple beam directions. Beam angles or directions are selected to provide the best coverage of the target volume while sparing dose limiting structures. For instance, a lower thoracic tumor will generally be treated with beams angled off the kidneys to avoid delivering a significant dose to this normal tissue structure. The beam is further modified by using collimators within the machine treatment head. These collimators attenuate the primary beam and therefore precisely define the treatment field size. The field size can be continuously adjusted by using multiple collimator leaves which are continuously moving to shape the beam (multileaf collimator or MLC). In radiosurgery, the size of these leaves is very small, allowing for the accurate delivery of radiation
Whole body and spinal radiosurgery
to extremely small field sizes (micro MLC). One can envision each beam of radiation as being divided into many small beamlets, each of which can be modulated into varying intensities using MLC. This is referred to as Intensity Modulated Radiotherapy (IMRT). The Novalis Shaped-Beam Surgery unit (BrainLAB, Westchester, IL) is a specialized treatment machine that utilizes a 6 megavoltage (mV) linear accelerator equipped with a micromultileaf collimator that uses dual in-room X-ray units and amorphous silicon flat-panel digital detectors. The system generates two digitally reconstructed radiographs from the simulation CT scan at the same orientation as the two kilovoltage (keV) X-ray images. The system then automatically compares internal structures noted in the keV images with those in the digitally reconstructed images. The system then adjusts the patient position based upon any isocenter deviations [64]. Newer technologies incorporate a somewhat different approach to the problems of motion and deformation of internal organs and uncertainties in patient setup. In this approach, the kilovolt source is mounted to the gantry of the treatment machine. Orthogonal localization x-rays of the patient are acquired by rotating the gantry to the appropriate angles. An important development is the ability to acquire volumetric pre-treatment images. A volumetric or threedimensional image contains much more information than simply orthogonal two-dimensional projection images [2]. This allows any rotational errors of the patient setup to be readily detected, and it also makes robust automatic registration procedures possible. The TomoTherapy Hi-Art system (TomoTherapy, Inc., Madison, WI) uses this approach by incorporating a miniature linear accelerator that has been integrated directly inside a CT scanner (> Figure 74-2). The system is an integrated treatment planning and CT-based image-guided helical IMRT delivery system
74
. Figure 74-2 The TomoTherapy Hi-Art system (TomoTherapy, Inc, Madison, WI) delivers radiation by using a 6-MV linear accelerator housed on a CT gantry. The beam is modulated by a 64-multileaf collimator that has paired, pneumatically driven, 6.25-mm-wide leaves calculated to open or close at approximately every 7 on linear accelerator rotation, resulting in highly conformal and homogeneous dose distributions
that is capable of highly conformal dose delivery to multiple targets simultaneously [43,71]. The system delivers radiation by using a 6-MV linear accelerator housed on a CT gantry. The beam is modulated by a 64-multileaf collimator that has paired, pneumatically driven, 6.25-mm-wide leaves calculated to open or close at approximately every 7 on linear accelerator rotation, resulting in highly conformal and homogeneous dose distributions. The device is capable of performing image-guided radiotherapy by co-registering pretreatment daily megavoltage CT scans to the initial planning CT scan to ensure accurate patient setup. Cone beam imaging is a new volumetric imaging technique that uses a gantry-mounted
1209
1210
74
Whole body and spinal radiosurgery
kilovolt source and detector. By making a full gantry rotation, the kilovolt system acquires several hundred projection images. These are converted into CT-like axial slices in a process similar to the reconstruction used by conventional CT scanners. The cone beam scans provide extremely high spatial resolution of both bony and soft tissue structures, making possible the setup of sites with submillimeter targeting errors [2]. The Elekta Synergy S (Elekta, Inc., Atlanta, GA) was the first digitally controlled linear accelerator optimized for image-guided radiation therapy enabling the acquisition of a high definition three dimensional volume image at the time of treatment with the patient in the treatment position (> Figure 74-3). The machine combines a linear accelerator with a fully integrated onboard 3-D volume imaging system that allows for the target to be visualized at the precise time of treatment while the patient is on the treatment couch. The robotic couch automatically makes any necessary adjustments in the
patient’s position to ensure that the radiation is directed precisely to the target. A combination of the above image tracking techniques is utilized by the Novalis TX (Varian Medical Systems, Palo Alto, CA and BrainLAB, Westchester, IL) (> Figure 74-4). This system combines cone beam imaging technology for patient positioning prior to treatment with dual in-room X-ray units and amorphous silicon flat-panel digital detectors. The X-ray images are acquired continuously during treatment to compensate for patient movement. A robotic couch can then make the necessary adjustments in the patient’s position. The linear accelerator is equipped with a multi-leaf collimator that allows for precisely shaped radiation beams.
Extracranial Radiosurgery With the ability to perform extracranial radiosurgery, the technique is increasingly being
. Figure 74-3 The Synergy S (Elekta Inc., Atlanta, GA) combines a linear accelerator with an onboard integrated high resolution 3D volume imaging system that allows for the target to be visualized at the precise time of treatment while the patient is in the treatment position on the treatment couch. The couch makes any necessary adjustments in the patient’s position to ensure that the radiation is directed precisely to the target
Whole body and spinal radiosurgery
. Figure 74-4 The Novalis TX (Varian Medical Systems, Palo Alto, CA and BrainLAB, Westchester, IL) combines cone beam imaging technology for patient positioning prior to treatment with dual in-room X-ray units and amorphous silicon flat-panel digital detectors. The X-ray images are acquired continuously during treatment to compensate for patient movement. A robotic couch can then make the necessary adjustments in the patient’s position. The linear accelerator is equipped with a multi-leaf collimator that allows for precisely shaped radiation beams
utilized throughout the body, both as primary definitive therapy, as well as providing cytoreductive therapy for metastatic lesions [72,73]. Both books are on my little round table in my office [72,73]. Stereotactic body radiotherapy (SBRT) has been most widely studied in early stage lung cancer where local control rates with standard fractionation radiotherapy have been extremely poor. In early stage non-small cell lung cancer, surgical therapy remains the standard of care. However, radiosurgical tumor ablation is emerging as an effective, less invasive option for those patients unable to tolerate surgical intervention. The Radiation Therapy Oncology Group (RTOG) has recently completed accrual to a phase II multi-institutional study of radiosurgery (60 Gy in 3 fractions) for peripheral lung tumors less than 5 cm in patients who are not candidates for surgery. This study was based
74
on the results from early work done at Indiana University. The initial Indiana University phase I trial of radiosurgery for medically inoperable non-small cell lung cancer achieved control rates of 90% using 18–24 Gy in 3 fractions [74]. The same group published the results of a phase II trial using 60–66 Gy in 3 fractions for the same patient population. The 2 year control rate was 95% with minimal toxicity for peripheral lesions [75]. Similar promising results have been published from Japan [76–78]. There is also renewed interest in hypofractionated definitive radiation therapy for prostate cancer. There is mounting radiobiological evidence that the therapeutic ratio may be improved by using a hypofractionated schedule of treatment compared to the standard 8 week course of radiation treatment. SBRT is a logical approach to providing accurate high dose per fraction treatment to the prostate. This is under active investigation at various sites but there is currently no long term data using SBRT for prostate cancer [79]. There are also ongoing studies of SBRT in other abdomino-pelvic sites including the liver, kidney and pancreas [80]. Radiosurgical treatment of oligometastases in sites such as liver, spine, and lung may provide a radiotherapeutic cytoreductive therapy. This has been most studied in the liver. Preliminary studies have shown various hypofractionation schedules can provide reasonable local control rates for 1–3 liver metastases with acceptable toxicity [81].
Radiosurgery for Malignant Spine Disease The clinical indications for radiosurgery for spine lesions are currently evolving and will continue to evolve as clinical experience increases (> Figures 74-5 and > 74-6). This is similar to the evolution of indications for radiosurgery for intracranial lesions that occurred during the last
1211
1212
74
Whole body and spinal radiosurgery
. Figure 74-5 Case example of a 42-year-old man with a painful colon metastasis of the L2 vertebral body. He had received prior conventional fractionated irradiation to the lesion with temporary improvement of symptoms. The treatment plan was designed to treat the tumor with a prescribed dose of 18 Gy in a single fraction that was calculated to the 80% isodose line; the maximum tumor dose was 22.5 Gy (CyberKnife Radiosurgical System, Accuray, Inc., Sunnyvale, CA). Axial and sagittal images of the treatment plan (a and b). The tumor volume was 58.8 cm3 and the cauda equina received a maximum dose of 12 Gy (bold red line = contoured tumor; light orange line = 80% isodose line; yellow line = 60% isodose line; dark purple line = 40% isodose line). The dose-volume histogram (DVH) is shown (c). Notice the conformality of the isodose line around the cauda equina
Whole body and spinal radiosurgery
74
. Figure 74-6 Case example of a 57-year-old woman with progression on MR imaging of a T4 breast metastasis after prior conventional fractionated irradiation treatment. The prescribed dose to the planned tumor volume (green line) was 13.5 Gy using 12 co-planar beams (Synergy S, Elekta Inc., Atlanta, GA). Axial and sagittal images of the treatment plan (a and b). The gross tumor volume was 19.3 cm3 and the spinal cord received a maximum dose of 9 Gy (red line). Digitally reconstructed radiograph of one coplanar beam demonstrating the position of the leaves of the multileaf collimator. Cone beam images obtained during patient setup and positioning for treatment provide extremely high spatial resolution of both bony structures as well as soft tissue, making possible the setup of sites with submillimeter targeting errors (d and e)
1213
1214
74
Whole body and spinal radiosurgery
decade. > Table 74-1 summarizes the candidate lesions for spine radiosurgery. > Table 74-2 summarizes some of the primary indications for spine radiosurgery. Spine radiosurgery can deliver radiation to anywhere along the spine, both extradural as well as intradural. Candidate lesions may be those that would require difficult surgical approaches for adequate resection. Candidate patients may have significant medical comorbidities precluding open surgical intervention or a relatively short life expectancy that would deem them inappropriate for open surgical intervention. Another indication would be to halt tumor progression that might lead to spinal instability or neural compromise. Radiosurgery as initial therapy may decrease the need for reirradiation by improving tumor control compared to conventional techniques.
. Table 74-1 Candidate lesions for spine radiourgery Minimal spinal cord compromise Previously irradiated lesions Radioresistant lesions that would benefit for a radiosurgical boost Residual tumor after surgery Recurrent tumor after prior surgical resection Lesions requiring difficult or morbid surgical approaches Relatively short life-expectancy as an exclusion criteria for open surgical intervention Significant medical co-morbidities precluding open surgical intervention No overt spinal instability
. Table 74-2 Indications for spine radiosurgery Pain Primary treatment modality Prevention of tumor progression Radiation boost for radioresistant tumors Progressive neurologic deficit Treatment of residual tumor after surgery Postsurgical tumor progression
Pain The most frequent indication for the treatment of spinal tumors is pain, and pain was the primary indication for spinal radiosurgery in 70% of our cases. Radiation is well known to be effective as a treatment for pain associated with spinal malignancies. Conventional external beam irradiation may provide less than optimal pain relief since the total dose is limited by the tolerance of adjacent tissues (e.g., spinal cord). Spinal radiosurgery was found to be highly effective at decreasing pain in this difficult patient population, with an overall long-term improvement of pain in 374 of the 435 cases (86%), depending upon primary histopathology. Durable pain improvement was demonstrated in 96% of women with breast cancer, 96% of cases with melanoma, 94% of cases with renal cell carcinoma, and 93% of lung cancer cases [82–85]. Pain usually decreases within weeks after treatment, and occasionally within days. Spinal radiosurgery is also effective at alleviating radicular pain caused by tumor compression of adjacent nerve roots. In some cases, post-treatment imaging revealed pathologic fractures, likely the cause of pain and the reason for radiosurgical failure. Such fractures require either open or closed internal fixation to alleviate the pain due to spinal instability. The most extensive published work to date on pain control and quality of life improvement after spinal radiosurgery has been from Georgetown University Hospital [23,37]. Using visual analog scales (VAS) for pain assessment and the 12-item Short Form Health Survey (SF-12), radiosurgery was demonstrated to statistically improve pain control and maintain quality of life with follow-up to 24 months. Early adverse events in their experience were infrequent and minor. The Memorial Sloan Kettering group demonstrated a 90% excellent palliation of symptoms with a median follow-up of 12 months [65]. Other series report similar pain improvement results [17,19,21,22,25,26,86].
Whole body and spinal radiosurgery
Radiographic Tumor Progression Spine radiosurgery is frequently used to treat radiographic tumor progression after conventional irradiation treatment or after prior surgery. The majority of these lesions have received irradiation with significant spinal cord doses, precluding further conventional irradiation delivery. Currently, spine radiosurgery is often being used as a ‘‘salvage’’ technique for those cases in which further conventional irradiation or open surgery are not appropriate. The ideal lesion should be well circumscribed such that the lesion can be easily outlined (contoured) for treatment planning. Overall long-term radiographic tumor control for progressive spinal disease in a series of 500 cases was 88% [46]. Radiographic tumor control differed based upon primary pathology: breast (100%), lung (100%), renal cell (87%) and melanoma (75%). Yamada et al. [65] reported a 90% long term radiographic control rate. Similar radiographic control rates have been reported by others [17,20–22,37,64]. As greater experience is gained, the technique will likely evolve into an initial upfront treatment for spinal metastases in certain cases (e.g., oligometastases). This is similar to the evolution that occurred for the treatment of intracranial metastases using radiosurgery that occurred over the past decade.
Primary Treatment Modality or Boost Radioresistant lesions (e.g., renal cell, sarcoma) that have completed external beam irradiation may undergo radiosurgery as a boost treatment. Another option may be to use radiosurgery as the sole radiation treatment. This clinical scenario is often encountered in a patient undergoing treatment to a symptomatic spine lesion with other significant but asymptomatic spine metastases. The additional asymptomatic lesions may be
74
treated with radiosurgery to avoid further irradiation to the neural elements as well as to avoid further bone marrow suppression and permit subsequent systemic therapy. The benefits for this approach include a single treatment which is radiobiologically larger than can be delivered with standard radiotherapy, with minimal radiation dose to adjacent normal tissue. When used as a primary treatment modality, long-term radiographic tumor control was demonstrated in 90% of cases (in all breast, lung and renal cell carcinoma metastases, and 75% of melanoma metastases) [46,82,83]. Degen et al. [37] reported a 100% tumor control rate in lesions that had not previously undergone irradiation. Radioresistant tumors (e.g., renal cell carcinoma, melanoma, sarcoma) may be treated with spinal radiosurgery after conventional irradiation as a boost treatment with excellent long-term radiographic control. Ryu et al. [26,65] found this to be a highly effective treatment paradigm.
Progressive Neurological Deficit Spine radiosurgery may be used to treat patients with progressive neurological deficits when open surgical intervention is felt to be contraindicated. In most of these particular cases, conventional irradiation has already been delivered to the symptomatic spinal lesion. In our experience, 36 of 42 patients (86%) with a progressive neurological deficit prior to treatment experienced at least some clinical improvement [46]. In most of these cases, open surgical decompression was precluded because of medical co-morbidities. Yamada reported 90% and 92% palliation of symptoms in patients treated for weakness and paresthesias, respectively [65]. Degen reported neurological deficits improved in 16 patients, were unchanged in 24 patients, and worsened in 11 patients in their series [37]. In a series from the Henry Ford Hospital, 12 of 16 cases (75%) with neurologic deficit from spinal
1215
1216
74
Whole body and spinal radiosurgery
cord compression were clinically improved or stable improved after spine radiosurgery (unpublished data).
After Open Surgery If a tumor is only partially resected during an open surgery, radiosurgery may be used to treat the residual tumor at a later date. The spinal tumors can be removed away from neural structures allowing for immediate decompression, the spine can be instrumented if necessary, and the residual tumor can be safely treated at a later date with radiosurgery, thus further decreasing surgical morbidity. Anterior corpectomy procedures in certain cases can be successfully avoided by posterior decompression and instrumentation alone followed by radiosurgery to the remaining anterior lesion. With the ability to effectively perform spinal radiosurgery, the current surgical approach to these lesions might change. Given the steep falloff gradient of the target dose with negligible skin dose, such treatments can be given early in the postoperative period as opposed to the usual significant delay before standard external beam irradiation is permitted. Open surgery for spinal metastases will likely evolve in a similar manner in which malignant intracranial lesions are debulked in such a way as to avoid neurological deficits and minimize surgical morbidity. Rock et al. [87] specifically evaluated the combination of open surgical procedure followed with adjuvant radiosurgery. They found this to be a successful treatment paradigm that was associated with a significant chance of stabilizing or improving neurological function. The technique was well tolerated and associated with little to no morbidity. Our institution has also found radiosurgery combined with open surgery to be a safe and highly successful treatment of the residual tumor bed [46].
Intradural Benign Tumors There is less clinical experience with radiosurgery for benign tumors of the spine than for metastatic lesions. Benign spinal tumors represent a group of intradural extramedullary neoplasms that include meningiomas, schwannomas, and neurofibromas. Microsurgical resection is the primary established treatment for these benign spinal tumors and most spinal meningiomas, schwannomas, and neurofibromas are non-infiltrative. Recurrence is unlikely with complete extirpation [36,88]. However, spinal radiosurgery may be a reasonable alternative treatment in certain clinical scenarios for patients who are less than ideal candidates for open resection because of age, medical comorbitidies, and recurrent tumor [89]. Multiple benign spinal tumors, as are common in familial neurocutaneous disorders, may be another pattern of spinal pathology better suited for the less invasive radiosurgical option. Finally, tumors that have recurred after open surgical resection may make safe surgical resection challenging or not possible. Radiosurgery has been used to successfully treat benign intradural extramedullary lesions with excellent long-term radiographic response, similar to the experience for intracranial radiosurgery [89]. In our institution’s experience with great than one hundred benign intradural extramedullary spinal tumors, long-term pain improvement was demonstrated in 73% of cases and long-term radiographic tumor control was demonstrated in all cases. While surgical extirpation remains the primary treatment option for most benign spinal tumors, radiosurgery has been demonstrated to have both short-term as well as long-term clinical benefits for the treatment of such lesions. Its role in patients with neurofibromatosis will also be further defined with greater clinical experience. Most importantly, spine radiosurgery has been found to be extremely safe at the doses currently used, even for intramedullary lesions. Although a risk for
Whole body and spinal radiosurgery
secondary malignant transformation with radiosurgery of benign tumors is theoretically possible, such a case has never been reported.
Meningiomas Spinal meningiomas are arachnoid cap cellderived tumors of the 5th–7th decade which have a female predominance and occur mainly in the thoracic region. Gross total surgical resection optimizes outcome [90]. Using the CyberKnife, Dodd et al. treated 16 spinal meningiomas (mean dose 20 Gy, mean tumor volume 2.4 cm3, mean follow-up 27 months) and demonstrated radiologic stabilization in 67% and radiologic tumor decrease in 33% of the 15 who had radiographic follow-up. Most patients experienced an improvement in pain and strength with radiosurgery [89]. From our institution’s published series, 13 spinal meningiomas were treated using a singlefraction technique (mean dose 21 Gy, mean tumor volume 4.9 cm3). Eleven of 13 patients had radiosurgery as an adjunctive treatment for residual or recurrent tumor following open surgical resection. Radiographic tumor control was demonstrated in all cases with a median follow-up of 17 months [91].
Schwannomas Schwannoma is the most common spinal tumor and has no proclivity for spinal region or gender [92]. The Stanford series comprised 30 tumors (mean dose 19 Gy, mean tumor volume 5.7 cm3, mean follow-up 26 months) and all but one patient had radiographic tumor control after radiosurgery; one-third of patients reported improvement in pain, weakness, or sensation, but 18% had a clinical decline after treatment [89]. Forty percent of patients in this series had spinal schwannomas in the setting of NF2. From our institution, 35 schwannomas have been treated
74
with spine radiosurgery (mean dose 22 Gy, mean tumor volume 11.0 cm3). Fourteen of seventeen patients (82%) described significant pain improvement for whom pain was the primary indication. Radiographic tumor control was demonstrated in six of seven patients (86%) for which radiosurgery was utilized as the primary treatment. A total of 3 patients went on to undergo upon surgical resection for new or persistent neurological deficits [91].
Neurofibromas Neurofibromas of the spine are often multiple, predominate in the cervical region, and are commonly associated with NF1. At Stanford, 9 neurofibromas in 7 patients with NF1 (mean dose 11 Gy, mean tumor volume 4.31 cm3, mean follow-up 19.9 months) were treated with radiosurgery and tumor stabilization on imaging was documented in six of seven (86%) patients; however, no patients described an improvement in symptoms after radiosurgery and half of the patients complained of a worsening in pain, weakness, or numbness at their last follow-up [89]. Our institution reported 25 neurofibroma cases (mean dose 21 Gy, mean tumor volume 12.6 cm3) [91]. No patient had evidence of radiographic tumor progression on follow-up. Twenty-one of these patients had NF1 and 9 had NF2. Radiosurgery ameliorated discomfort in 8 of 13 (62%) patients treated for pain; however all patients who saw no improvement in pain had NF1. These findings echo outcomes from the Stanford report that found pain control in NF1-associated spinal neurofibromas to be recalcitrant to radiosurgery [89]. Poorer microsurgical results for neurofibromas have also been observed in patients with NF1 [93]. The multiplicity of neurofibromas in NF1 may be partially to blame as this factor makes identifying the symptomatic neurofibroma in need of treatment more difficult. Moreover, the infiltrating of
1217
1218
74
Whole body and spinal radiosurgery
neurofibromas, in contrast to the other benign extramedullary intradural spinal tumors, may engender more irreversible neural damage and increase the susceptibility of the native nerve root to injury from both microsurgical and radiosurgical treatments. Finally, future genomic investigations may reveal that intrinsic genetic differences in NF1-associated neurofibromas predispose to a weaker radiobiologic response.
Arteriovenous Malformations Stereotactic radiosurgery has emerged over the last three decades as a successful alternative to microsurgical resection and embolization for cerebral arteriovenous malformations. Since this technique was first described by Steiner in 1972 [94], over 5,000 patients have successfully been treated with this modality. Radiosurgery causes gradual hyperplasia of the endothelial tissue within the arteries of the nidus of the AVM which leads to vessel occlusion [95]. Numerous reports show that cerebral AVMs have an 80–85% obliteration rate for those vascular malformations smaller than 2.5 cm [45,96–101]. With the success with radiosurgery in the management of brain AVMs, the treatment of spinal cord AVMs with radiosurgery would be a logical next step. Spinal arteriovenous malformations have been successfully treated with radiosurgery [102]. Given the various subtypes of spinal cord AVMs, those with a relatively compact nidus would represent the optimal targets [103]. Type I and IV spinal cord AVMs are dural and perimedullary arteriovenous fistulas, and are often optimally treated with endovascular embolization and or microsurgical resection. Type III spinal cord AVMs, also called juvenile type spinal AVMs, are characterized by large and diffuse intramedullary nidus which can also extend into the extramedullary space. They are less well defined spinal cord AVMs, and also do not represent optimal
radiosurgical targets. Type II spinal cord AVMs, also called glomus AVMs, represent a compact vascular nidus and are often suitable radiosurgery targets. These type II spinal cord AVMs are also difficult if not impossible to treat with embolization and microsurgery alone and often were not treated prior to the development of spinal radiosurgery. The largest experience with spinal AVM radiosurgery comes from the Stanford series [103]. Twenty-three patients with spinal cord AVMs have been treated at Stanford using the Cyberknife technology. Twenty-two of these patients had type II or glomus AVMs while one patient had a type III juvenile spinal cord AVM. Spinal AVM locations included the cervical spine (12 patients), thoracic spine (eight patients), and conus region (three patients). Thirteen of these twenty-three patients (57%) presented with hemorrhage, which was multiple in six of these patients. In the remaining 10 patients, progressive neurologic dysfunction was the presenting symptom. Ten of the 23 patients (43%) underwent prior embolization to their AVM. A median of three treatment sessions was performed. Mean target volume was 2.8 cm3 with a range of 0.26–15 cm3. The mean marginal dose used was 20.3 Gy with a range of 16–21 Gy. A median of two sessions was utilized in this cohort with a range of one to four treatments. The mean maximal internidal dose was 25.8 Gy and ranged from 22.5 to 30 Gy. Clinical follow-up ranged from four to 83 months with a mean of 35 months. Radiographic follow-up averaged 25 months. Although postoperative MRI imaging demonstrated noticeable reduction in the volume of all AVMs over the course of follow-up, only eight of the 23 patients underwent formal spine angiography. Of these eight patients, three patients had complete angiographic obliteration. To date, no patients suffered a rebleed after spine radiosurgery and the clinical outcome was improved or unchanged in 96% of patients. A single patient
Whole body and spinal radiosurgery
who was neurologically severely impaired prior to radiosurgery deteriorated further at 9 months following radiosurgery. MR imaging demonstrated a significant decrease in flow void as well as high signal in the adjacent spinal cord on T2 imaging consistent with radiation induced edema. One other patient experienced significant onset of radiation edema (for a conus AVM) but this AVM obliterated rapidly over 9 months and the patient was one of the three patients who was documented to have complete obliteration of his AVM. The future of radiosurgery for the treatment of spinal AVM has still not been determined.
Potential Advantage of Spine Radiosurgery Spine radiosurgery has several advantages over other treatment alternatives. For patients with intradural tumors, there are obvious advantages to this treatment over the potential complications associated with open surgical techniques. Spine radiosurgery avoids the need to irradiate large segments of the spinal cord as well as delivery of a minimal radiation dose to adjacent normal tissue. Early stereotactic radiosurgery treatment of spinal lesions may obviate the need for extensive spinal surgeries for decompression and fixation in these often debilitated patients. A much larger radiobiologic dose can be delivered compared to external beam irradiation. It may also avoid the need to irradiate large segments of the vertebral column, known to have an important deleterious effect on bone marrow reserve in these cancer patients who frequently require bone marrow suppressive systemic therapy. Avoiding open surgery as well as preserving bone marrow function facilitates continuous chemotherapy in patients with cancer. Furthermore, improved local control, such as has been the case with intracranial radiosurgery. Could translate into more effective palliation and potentially longer survival. With greater
74
clinical experience, upfront radiosurgery perhaps will become more commonly used in certain cases such as patients with single symptomatic spine metastases of a radioresistant histology. Radiosurgery allows for the treatment of lesions previously irradiated using conventional external beam techniques. An advantage to the patient of using single-fraction radiosurgery is that the treatment can be completed in a single day rather than over a course of several weeks, which may be beneficial for patients with a limited life expectancy from cancer. In addition, cancer patients may have difficulty being transported to a radiation treatment facility for prolonged, daily fractionated therapy. A large single fraction of irradiation may be more radiobiologically advantageous to certain tumors such as renal cell carcinoma compared to prolonged fractionated radiotherapy. The radiobiology of large fraction radiotherapy to certain tumor types is currently under investigation. Clinical response such as pain or improvement of a neurological deficit might also be more rapid with a radiosurgery technique. This rapid clinical response is becoming welldocumented in the peer-reviewed literature [19,37]. Finally, the procedure is minimally invasive compared to open surgical techniques and can be performed in an outpatient setting.
Summary Spine radiosurgery is a feasible, safe and clinically effective for the treatment of a variety of spinal tumors using a wide variety of different technologies. Spine radiosurgery represents a logical extension of the current state-of-the-art radiation therapy. It has the potential to substantially (would only use the word significantly if we are speaking of statistics) improve local control of cancer of the spine, which could translate into better palliation. The major potential benefits of radiosurgical ablation of spinal lesions are relatively short treatment time in an outpatient
1219
1220
74
Whole body and spinal radiosurgery
setting combined with potentially better local control of the tumor with minimal risk of side effects, however, further studies confirming these presmed benefits are needed. Spine radiosurgery offers a new and important alternative therapeutic modality for the treatment of spinal tumors in patients who are medically inoperable or who are poor surgical candidates, have had prior irradiation to their tumor, have lesions not amenable to open surgical techniques or as an adjunct to surgery.
References 1. Horn E, Henn J, Lemole J, GM. Thoracoscopic placement of dural-rod instrumentation in thoracic spinal trauma. Neurosurgery 2004;54:1150-4. 2. Yamada Y, Lovelock D, Bilsky M. A review of imageguided intensity-modulated radiotherapy for spinal tumors. Neurosurgery 2007;61:226-35. 3. Clain A. Secondary malignant disease of bone. Br J Cancer 1965;19:15-29. 4. Black P. Spinal metastasis: current status and recommended guidelines for management. Neurosurgery 1979;5:726-46. 5. Gokaslan Z, York J, Walsh G. Transthoracic vertebrectomy for metastatic spinal tumors. J Neurosurgery 1998;89:599-609. 6. Gerszten PC, Welch WC. Current surgical management of metastatic spinal disease. Oncology 2000;14:1013-36. 7. Faul CM, Flickinger JC. The use of radiation in the management of spinal metastases. J Neurooncol 1995;23:149-61. 8. Kim YH, Fayos JV. Radiation tolerance of the cervical spinal cord. Radiology 1981;139:473-8. 9. Markoe AM, Schwade JG. The role of radiation therapy in the management of spine and spinal cord tumors. In: Rea GL, editor. Spine tumors. American Association of Neurological Surgeons; 1994. p. 23-35. 10. Shapiro W, Posner JB. Medical vs surgical treatment of metastatic spinal cord tumors. In: Thompson R, Green J, editors. Controversies in Neurology. New York: Raven Press; 1983. p. 57-65. 11. Sundaresan N, Digiacinto GV, Hughes JEO, et al. Treatment of neoplastic spinal cord compression: results of a prospective study. Neurosurgery 1991;29:645-50. 12. Sundaresan N, Krol G, Digiacinto CV, et al. Metastatic tumors of the spine. In: Sundaresan B, Schmidek H, Schiller A, et al., editors. Tumors of the spine. Philadelphia, PA: W.B. Saunders; 1990:279-304.
13. Lu C, Stomper PC, Drislane FW, et al. Suspected spinal cord compression in breast cancer patients: a multidisciplinary risk assessment. Breast Cancer Res Treat 1998;51:121-31. 14. Vitaz T, Oishi M, Welch W, et al. Rotational and transpositional flaps for the treatment of spinal wound dehiscence and infections in patient populations with degenerative and oncological disease. J Neurosurg Spine 2004;100:46-51. 15. Patchell RA, Tibbs PA, Regine WF, et al. Direct decompressive surgical resection in the treatment of spinal cord compression caused by metastatic cancer: a randomised trial. The Lancet 2005;21:1-6. 16. Loblaw DA, Laperriere NJ. Emergency treatment of malignant extradural spinal cord compression: an evidence-based guideline. J Clin Oncol 1998;16: 1613-24. 17. Ryu S, Chang S, Kim D, et al. Image-guided hypofractionated stereotactic radiosurgery to spinal lesions. Neurosurgery 2001;49:838-46. 18. Amendola B, Wolf A, Coy S, et al. Gamma knife radiosurgery in the treatment of patients with single and multiple brain metastases from carcinoma of the breast. Cancer J 2000;6:88-92. 19. Benzil DL, Saboori M, Mogilner AY, et al. Safety and efficacy of stereotactic radiosurgery for tumors of the spine. J Neurosurgery 2004;101:413-8. 20. Bilsky MH, Yamada Y, Yenice KM, et al. Intensitymodulated stereotactic radiotherapy of paraspinal tumors: a preliminary report. Neurosurgery 2004;54:823-30. 21. Chang EL, Shiu AS, Lii M-F, et al. Phase I clinical evaluation of near-simultaneous computed tomographic image-guided stereotactic body radiotherapy for spinal metastases. Int J Rad Onc Biol Phys 2004;59:1288-94. 22. Desalles AA, Pedroso A, Medin P, et al. Spinal lesions treated with Novalis shaped beam intensity modulated radiosurgery and stereotactic radiotherapy. J Neurosurgery 2004;101:435-40. 23. Gagnon GJ, Henderson FC, Gehan EA, et al. Cyberknife radiosurgery for breast cancer spine metastases: a matched-pair analysis. Cancer 2007;110:1796-802. 24. Jin J-Y, Chen Q, Jin R, et al. Technical and clinical experience with spine radiosurgery: a new technology for management of localized spine metastases. Technol Cancer Res Treat 2007;6:127-33. 25. Milker-Zabel S, Zabel A, Thilmann C, et al. Clinical results of retreatment of vertebral bone metastases by stereotactic conformal radiotherapy and intensitymodulated radiotherapy. Int J Rad Onc Biol Phys 2003;55:162-7. 26. Ryu S, Yin FF, Rock J, et al. Image-guided and intensitymodulated radiosurgery for patients with spinal metastasis. Cancer 2003;97:2013-8. 27. Shiu AS, Chang EL, Ye J-S. Near simultaneous computed tomography image-guided stereotactic spinal
Whole body and spinal radiosurgery
28.
29.
30. 31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
radiotherapy: an emerging paradigm for achieving true stereotaxy. Int J Rad Onc Biol Phys 2003;57:605-13. Yin FF, Ryu S, Ajlouni M, et al. Image-guided procedures for intensity-modulated spinal radiosurgery. J Neurosurgery 2004;101:419-24. Auchter R, Lamond J, Alexander E. A multinstitutional outcome and prognostic factor analysis of radiosurgery for resectable single brain metastasis. Int J Rad Onc Biol Phys 1996;35:27-35. Chang S, Adler J, Hancock S. The clinical use of radiosurgery. Oncology 1998;12:1181-91. Flickinger J, Kondziolka D, Lunsford L. A multiinstitutional experience with stereotactic radiosurgery for solitary brain metastasis. Int J Rad Onc Biol Phys 1994;28:797-802. Kondziolka D, Patel A, Lunsford L. Stereotactic radiosurgery plus whole brain radiotherapy vs radiotherapy alone for patients with multiple brain metastases. Int J Rad Onc Biol Phys 1999;45:427-34. Loeffler J, Alexander EI. Radiosurgery for the treatment of intracranial metastasis. New York: McGraw-Hill; 1993. Loeffler JS, Kooy HM, Wen PY, et al. The treatment of recurrent brain metastases with stereotactic radiosurgery. J Clin Oncol 1990;8:576-82. Sperduto P, Scott C, Andrews D. Stereotactic radiosurgery with whole brain radiation therapy improves survival in patients with brain metastases: report of radiation therapy oncology group phase III study 95–08. Int J Rad Onc Biol Phys 2002;54:3a. Conti P, Pansini G, Mouchaty H, et al. Spinal neurinomas: retrospective analysis and long-term outcome of 179 consecutively operated cases and review of the literature. Surg Neurol 2004;61:34-44. Degen JW, Gagnon GJ, Voyadzis J-M, et al. CyberKnife stereotactic radiosurgical treatment of spinal tumors for pain control and quality of life. J Neurosurg Spine 2005;2:540-9. Hamilton AJ. Linear accelerator (LINAC) – based stereotactic spinal radiosurgery. In: Gildenberg PL, Tasker RR, editors. Textbook of stereotactic and functional neurosurgery. New York: McGraw-Hill; 1998. p. 857-69. Hamilton A, Lulu B. A prototype design for linear accelerator based radiosurgery. Acta Neurochir (Wien) 1995;63:40-3. Blomgren H, Lax I, Naslund I, et al. Stereotactic high dose fraction radiation therapy of extracranial tumors using an accelerator: clinical experience of the first thirty-one patients. Acta Oncol 1995;34:861-70. Lax I, Blomgren H, Naslund I, et al. Stereotactic radiotherapy of malignancies in the abdomen: methodological aspects. Acta Oncol 1994;33:677-83. Hamilton A, Lulu B, Fosmire H, et al. Preliminary clinical experience with linear accelerator-based spinal stereotactic radiosurgery. Neurosurgery 1995;36:311-9.
74
43. Baisden JM, Benedict SH, Sheng K, et al. Helical tomotherapy in the treatment of central nervous system metastasis. Neurosurg Focus 2007;22:1-6. 44. Chang S, Adler J. Current status and optimal use of radiosurgery. Oncology 2001;15:209-21. 45. Colombo F, Pozza F, Chierego G. Linear accelerator radiosurgery of cerebral arteriovenous malformations: an update. Neurosurgery 1994;34:14-21. 46. Gerszten PC, Burton SA, Ozhasoglu C, et al. Single fraction radiosurgery for spinal metastases: clinical experience in 500 cases from a single institution. Spine 2007;32:193-9. 47. Gerszten PC, Welch WC. CyberKnife radiosurgery for metastatic spine tumors. Neurosurg Clin N Am 2004;15:491-501. 48. Hitchcock E, Kitchen G, Dalton E, et al. Stereotactic linac radiosurgery. Br J Neurosurg 1989;3:305-12. 49. Medin P, Solberg T, DeSalles A. Investigations of a minimally invasive method for treatment of spinal malignancies with LINAC stereotactic radiation therapy: accuracy and animal studies. Int J Rad Onc Biol Phys 2002;52:1111-22. 50. Pirzkall A, Lohr F, Rhein B, et al. Conformal radiotherapy of challenging paraspinal tumors using a multiple arc segment technique. Int J Rad Onc Biol Phys 2000;48:1197-204. 51. Ryu S, Rock J, Rosenblum M, et al. Patterns of failure after single-dose radiosurgery for spinal metastasis. J Neurosurgery 2004;101:402-5. 52. Pieters RS, Niemierko A, Fullerton BC, et al. Cauda equina tolerance to high dose fractionated irradiation. Int J Rad Onc Biol Phys 2006;64:251-7. 53. Ryu S, Jin J-Y, Jin R, et al. Partial volume tolerance of the spinal cord and complications of single dose radiosurgery. Cancer 2007;109:628-36. 54. Hopewell J, Morris A, Dixon-Brown A. The influence of field size on the late tolerance of the rat spinal cord to single doses of X rays. Br J Radiol 1987;60:1099-108. 55. Emami B, Lyman A, Brown JT, et al. Tolerance of normal tissue to therapeutic irradiation. J Radiat Oncol Biol Phys 1991;21:109-22. 56. Gerszten PC, Bilsky MH. Spine radiosurgery. Contemp Neurosurg 2006;28:1-8. 57. Tong D, Hendrickson F. The palliation of symptomatic osseous metastases; final results of the study by the radiation therapy oncology group. Cancer 1982;50:893-9. 58. Wara WM, Phillips TL, Sheline GE, et al. Radiation tolerance of the spinal cord. Cancer 1975;35:1558-62. 59. Hatlevoll R, Host H, Kaalhus O. Myelopathy following radiotherapy of bronchial carcinoma with large single fractions: a retrospective study. Int J Radiat Oncol Biol Phys 1983;9:41-4. 60. Abbatucci JS, Delozier T, Quint R, et al. Radiation myelopathy of the cervical spinal cord: time, dose and volume factors. Int J Radiat Oncol Biol Phys 1978;4:239-48.
1221
1222
74
Whole body and spinal radiosurgery
61. McCuniff AJ, Liang MJ. Radiation tolerance of the cervical spinal cord. Int J Radiat Oncol Biol Phys 1989;16:675-8. 62. Phillips TL, Buschke F. Radiation tolerance of the thoracic spinal cord. AJR 1969;105:659-64. 63. Klish MD, Watson GA, Shrieve DC. Radiation and intensity-modulated radiotherapy for metastatic spine tumors. Neurosurg Clin N Am 2004;15:481-90. 64. Rock JP, Ryu S, Yin FF. Novalis radiosurgery for metastatic spine tumors. Neurosurgery Clinics of North America 2004;15:503-9. 65. Yamada Y, Lovelock M, Yenice KM, et al. Multifractionated image-guided and stereotactic intensity modulated radiotherapy of paraspinal tumors: a preliminary report. Int J Rad Onc Biol Phys 2005;62:53-61. 66. Ho AK, Fu D, Cotrutz C, et al. A study of the accuracy of cyberknife spinal radiosurgery using skeletal structure tracking. Neurosurgery 2007;60:147-56. 67. Muacevic A, Staehler M, Drexler C, et al. Technical description, phantom accuracy and clinical feasibility for fiducial-free frameless real-time image-guided spinal radiosurgery. J Neurosurg Spine 2006;5:303-12. 68. Adler J, Murphy M, Chang S, et al. Image-guided robotic radiosurgery. Neurosurgery 1999;44:1-8. 69. Adler JR, Chang SD, Murphy MJ, et al. The CyberKnife: a frameless robotic system for radiosurgery. Stereotactic and Functional Neurosurgery 1997;69:124-8. 70. Murphy MJ, Cox RS. Frameless radiosurgery using realtime image correlation for beam targeting. Med Phys 1996;23:1052-3. 71. Welsh J, Mehta M, Mackie T, et al. Helical tomotherapy as a means of delivering scalpsparing whole brain radiation therapy. Technol Cancer Res Treat 2005;4:661-2. 72. Extracranial Stereotactic Radiotherapy and Radiosurgery. Slotman BJ, Solberg TD, Verellen D, editors. New York: Taylor and Francis Group; 2006. 73. Stereotactic Body Radiation Therapy. Kavanagh BD, Timmerman RD, editors. Philadelphia, PA: Lippincott Williams and Wilkins; 2005. 74. Timmerman RD, Papiez L, McGarry R. Extracranial stereotactic radioablation: results of a phase I study in medically inoperable stage I non-small cell lung cancer. Chest 2003;124:1946-55. 75. Timmerman RD, McGarry R, Yiannoutsos C. Excessive toxicity when treating central tumors in a phase II study of stereotactic body radiation therapy for medically inoperable early-stage lung cancer. J Clin Oncol 2006; 24:4833-9. 76. Nagata Y, Takayama K, Matsuo Y. Clinical outcomes of a phase I/II study of 48Gy stereotactic body radiotherapy in 4 fractions for primary lung cancer using a stereotactic body frame. Int J Radiat Oncol Biol Phys 2005;63:1427-31. 77. Onishi H, Araki T, Shirato H. Stereotactic hypofractionated high-dose irradiation for stage I nonsmall cell lung carcinoma: clinical outcomes in 245 subjects in a Japanes multiinstitutional study. Cancer 2004;101:1623-31.
78. Uematsu M, Shioda A, Tahara K. Computed tomographyguided frameless stereotactic radiotherapy for stage I nonsmall cell lung cancer: 5-year experience. Int J Radiat Oncol Biol Phys 2001;51:666-70. 79. Timmerman RD. Stereotactic body radiation therapy. Curr Probl Cancer 2005;29:120-57. 80. Timmerman R, Kavanagh B, Chos CL. Stereotactic body radiation therapy in multiple organ sites. J Clin Oncol 2007;25:947-52. 81. Shefter TE, Cardenes HR, Kavanagh BD. Stereotactic body radiotherapy for liver tumors. In: Kavanagh BD, Timmerman RD, editors. Stereotactic Body Radiation Therapy. Philadelphia, PA: Lippincott Williams and Wilkins; 2005. 82. Gerszten PC, Burton S, Ozhasoglu C, et al. Stereotactic radiosurgery for spine metastases from renal cell carcinoma. J Neurosurg Spine 2005;3:288-95. 83. Gerszten PC, Burton S, Welch WC, et al. Single fraction radiosurgery for the treatment of breast metastases. Cancer 2005;14:2244-54. 84. Gerszten PC, Burton SA, Belani C, et al. Radiosurgery for the treatment of spinal lung metastases. Cancer 2006;107 (11):2653-61. 85. Gerszten PC, Burton SA, Quinn AE, et al. Radiosurgery for the treatment of spinal melanoma metastases. Stereotact Funct Neurosurg 2006;83:213-21. 86. Ryken TC, Meeks SL, Pennington EC, et al. Initial clinical experience with frameless stereotactic radiosurgery: analysis of accuracy and feasibility. Int J Rad Onc Biol Phys 2001;51:1152-8. 87. Rock J, Ryu S, Shukairy MS, et al. Postoperative radiosurgery for malignant spinal tumors. Neurosurgery 2006;58:891-8. 88. Cohen-Gadol A, Zikel O, Koch C, et al. Spinal meningiomas in patients younger than 50 years of age: a 21-year experience. J Neurosurg Spine 2003;98:258-63. 89. Dodd RL, Ryu MR, Kammerdsupaphon P, et al. CyberKnife radiosurgery for benign intradural extramedullary spinal tumors. Neurosurgery 2006;58:674-85. 90. Peker S, Cerci A, Ozgen S, et al. Spinal meningiomas: evaluation of 41 patients. J Neurosurg Sci 2005;49:7-11. 91. Gerszten PC, Burton SA, Ozhasoglu C, et al. Radiosurgery for benign intradural spinal tumors. J Neurosurgery 2007;106:A742. 92. Seppala M, Haltia M, Sankila R, et al. Long term outcome after removal of spinal schwannoma: a clinicalopathological study of 187 cases. J Neurosurgery 1995;83:621-6. 93. Rampling R, Symonds S. Radiation myelopathy. Curr Opin Neurol 1998;11:627-32. 94. Steiner L, Lesksell L, Greitz T, et al. Stereotaxic radiosurgery for cerebral arteriovenous malformations. Report of a case. Acta Chir Scand 1972;138:459-64. 95. Chang S, Shuster D, Steinberg G, et al. Stereotactic radiosurgery of arteriovenous malformations: pathologic changes in resected tissue. Clin Neuropathol 1997;16:111-6.
Whole body and spinal radiosurgery
96. Betti O, Munari C, Rosler R. Stereotactic radiosurgery with the linear accelerator: treatment of arteriovenous malformations. Neurosurgery 1989;24:311-21. 97. Coffey R, Lunsford L, Bissonette D, et al. Stereotactic gamma radiosurgery for intracranial vascular malformations and tumors: report of the initial North American experience in 331 patients. Stereotact Funct Neurosurg 1990;54:535-40. 98. Colombo F, Benedetti A, Pozza F, et al. Linear accelerator radiosurgery of cerebral arteriovenous malformations. Neurosurgery 1989;24:833-40. 99. Friedman W, Bova F. Linear accelerator radiosurgery for arteriovenous malformations. J Neurosurgery 1992;77: 832-41.
74
100. Steinberg G, Fabrikant J, Marks M, et al. Stereotactic heavy-charged-particle Bragg-peak radiation for intracranial arteriovenous malformations (see comments). N Engl J Med 1990;323:96-101. 101. Steiner L. Radiosurgery in cerebral arteriovenous. In: Flamm ES, Fein J, editors. Cerebrovascular surgery. New York: Springer; 1985. p. 1161-215. 102. Sinclair J, Chang SD, Gibbs IC, et al. Multisession CyberKnife radiosurgery for intramedullary spinal cord arteriovenous malformations. Neurosurgery 2006;58: 6 1081-9. 103. Chang S, Hancock S, Gibbs I, et al. Spinal cord arteriovenous malformation radiosurgery. In: Gerszten PC, Ryu SI, editors. Spine radiosurgery. New York: Thieme; (in press).
1223
80 Anesthesia for Functional Neurosurgery P. H. Manninen . N. Apichatibutra
Stereotactic functional neurosurgery is used for the treatment of many movement and functional disorders. Historically, the awake craniotomy was used for epilepsy surgery, but now it is used increasingly for patients with brain tumors. The role of the anesthesiologist is critical in assuring success for these patients. This chapter will cover the anesthetic considerations and management of patients undergoing surgery for functional disorders, stereotactic biopsy, epilepsy, and brain tumors. The anesthesia management includes the awake patient with monitored anesthesia care or conscious sedation and the use of general anesthesia.
Stereotactic Functional Neurosurgery Functional neurosurgery with the use of deep brain stimulation (DBS) is used in the treatment of patients with movement disorders and other chronic illnesses [1,2]. The initial success was with patients with Parkinson’s disease [3,4]. Now the applications and indications have expanded to many other disorders such as dystonia, tremors, movement disorders, depression, obsessive-compulsive disorder, epilepsy, chronic pain, and other newer areas that are under investigation [1,5,6]. Deep brain stimulation is a minimally invasive procedure that enables structures in the brain such as the subthalamic nucleus to be stimulated electrically by an implanted pacemaker. There are many steps to this procedure including the placement of a rigid frame to the #
Springer-Verlag Berlin/Heidelberg 2009
patient’s head for stereotaxis and radiological imaging to localize brain structures with references to external coordinates. The intraoperative course includes making a burr hole for insertion of the DBS for microelectrode recordings (MER) and clinical testing of an awake patient. The DBS is then secured to the skull and at the same setting or at a future date it is connected to a lead that is tunneled to the chest or abdomen where a pulse generator pacemaker is implanted. In most centers the anesthesiologist plays an important role in the care and monitoring of the patients and in providing sedation or anesthesia during the intraoperative course [4,7–10]. However, the use of sedation and/or general anesthesia remains controversial. The preferred technique of anesthetic management of the patient in many centers is local anesthesia with monitored anesthesia care, that is, no sedation [4,10]. The reasons for having a completely awake patient are to preserve the MER of a single unit, to use stimulation testing for localization, and to observe for improvement in symptoms, adverse effects, and neurological examination. There are concerns that even mild sedative drugs may interfere with neurophysiology and MER. However, performing these procedures without sedation in certain patients may be problematic. Some patients are reluctant to be awake for these long procedures. The patient may have such severe continuous movements, kyphosis, pain, depression, psychotic symptoms, or be uncooperative that they are unable to lie still and/or to cooperate for neurophysiologic testing. The patient’s symptoms may also be worse due to the ‘‘weaning off ’’
1332
80
Anesthesia for functional neurosurgery
from their usual medications. Some of these patients will require sedation or even general anesthesia. There are many challenges and demands for the anesthesiologist whether the procedure is performed with or without sedation. These include keeping the patient comfortable, responsive, and cooperative for a long period of time and if providing sedation, ensuring that it does not interfere with electrophysiological brain mapping and clinical testing. Many of these patients also present with complex medical problems, may be elderly, and on multiple medications. The role of the anesthesiologist is to also monitor the patient’s vital signs, especially cardiovascular and respiratory, ensuring that they remain stable and within normal range. Continuous vigilance is required to rapidly diagnose and treat all complications some of which may be life-threatening.
Preoperative Assessment Prior to the surgical procedure, the anesthesiology team should review all patients scheduled for functional neurosurgery. This can be done in a preoperative anesthesia consult clinic if patients are admitted to the hospital on the day of surgery. All routine assessments including laboratory studies and considerations for anesthesia are reviewed. The patient should be in stable condition with respect to all other ongoing medical conditions. Optimal preparation includes the continuation of routine medications. The anesthesiologist should emphasize to the patient to follow instructions given by the neurologist/neurosurgeon regarding discontinuation of medications for their functional disorder. All patients must follow standard ‘‘nothing by mouth’’ orders. Psychological preparation of the patient is important for the long and potentially difficult awake procedures. Patient should be advised as to what will happen during the course of the day
and the role of the anesthesiologist in providing sedation, if possible, and maximum comfort and support. Specific anesthetic considerations will apply to patients with different disorders. Patients with neurodegenerative disorders such as Parkinson’s disease are often elderly and may have respiratory, cardiovascular, and autonomic system compromise [11–13]. Potential drug interactions and adverse effects from anti-Parkinson’s medications may also occur during anesthesia. Ergotderived dopamine agonists are associated with a higher incidence of cardiac valvular disorders [14]. Drugs that are selective monoamine oxidase inhibitors may have serious interactions with meperidine and sympathomimetic agents [15]. The effect of sedative and anesthetic agents must also be considered in patients with Parkinson’s disease such as propofol induced dyskinesia and the suppression of tremors with remifentanil [16–18]. Patients with dystonia or torticollis may present difficulties in airway management. Pyschiatric patients will also have their own set of challenges and patients with chronic pain will need special consideration in management of their pain medications perioperatively.
Procedure In the operating room or radiology suite the surgeons will apply a rigid head frame to the patient’s skull. There may or may not be an anesthesiologist present and sedation may or may not be used. Local anesthesia is used as a subcutaneous infiltration at the pin sites. Supraorbital and greater occipital nerve blocks are an alternative as they have been shown to be less painful than subcutaneous infiltration but did not result in any difference at the time of pin placement or during surgery [19]. In the radiology suite during imaging no sedative medication is usually required except in cases of extreme anxiety or claustrophobia or if general anesthesia is required for the entire
Anesthesia for functional neurosurgery
procedure. The patient is then transferred to the operating room fully awake or if under sedation or general anesthesia with monitoring and continuation of anesthesia. Proper positioning of patients on the operating table is an important step to ensure maximum comfort and cooperativeness. After the head frame is fixed to the operating room table one needs to ensure that the patient’s neck is comfortable, they are able to open their mouth so that breathing, speech, and swallowing are not restricted. The legs should be flexed and supported under the knees to maintain stability when the head and back are elevated to a sitting position. Slipping downward on the table while head is fixed has resulted in complete upper airway obstruction even in nonsedated patients [10]. All pressure points should be padded. Specific treatment modalities, such as physiotherapy, may be used in specific patients [20]. If a patient with Parkinson’s disease has excessive tremors and can not lie still, a small dose of levodopa may help to decrease the intensity of the tremors. Local anesthesia is infiltrated at the site of the burr hole for insertion of electrodes for the MER.
Intraoperative Anesthetic Considerations Standard monitors used for all procedures include an electrocardiogram, noninvasive blood pressure, oxygen saturation, end-tidal CO2, and respiratory rate. The need for additional monitors such an intra-arterial catheter for blood pressure, temperature, and the insertion of a urinary catheter are dependent on the practice of each institution, condition of the patient, and depth of sedation or anesthesia. Administration of supplemental oxygen is mandatory even for awake patients. This may be delivered via nasal prongs or a mask with an outlet for end-tidal CO2 and respiratory rate monitoring. A secure intravenous catheter is used for medications and fluid administration. Omission of the urinary
80
catheter will be more comfortable for the patient, but fluid administration needs to be restricted. The patient may initially feel very cold in the operating room and may even require a warming blanket. However, with prolonged surgery and with increased levels of anxiety some patients become too hot. Patients who are not sedated will require frequent reassurance, additional comforts such as moving arms and legs, and even ice chips to wet their mouth. Awareness of an awake patient by all operating room personnel and the administration of psychological support by all members of the team will help the patient in this stressful environment. Communication is essential between the neurosurgeon and the anesthesiologist throughout the procedure. The anesthesiologist needs to know when it is appropriate to give the patient sedation or analgesia. The timing of the induction of general anesthesia, if it is to be used, must be clarified. Intubation of the patient’s trachea will be more difficult after the stereotactic frame is pinned to the patient’s skull. If general anesthesia is to be started in the radiology suite, the anesthesiologist must have adequate equipment and support to care for the patient in this potentially ‘‘remote’’ site.
Local Anesthesia The local anesthetic agents usually used for infiltration of the pin sites for the frame and the incision sites are bupivacaine, ropivacaine, and lidocaine with and without epinephrine [21,22]. Compared with lidocaine, bupivacaine has slower onset of action but longer duration. Ropivacaine has less cardiac and central nervous system toxicity compared with bupivacaine. Complications of local anesthesia include toxic blood levels resulting in seizures, respiratory, and cardiac arrest. The maximum doses are based on the patient’s weight and for lidocaine this is 5 mg/kg and up to 7 mg/kg with epinephrine, for bupivacaine 2 mg/kg and
1333
1334
80
Anesthesia for functional neurosurgery
3mg/kg with epinephrine, and for ropivacaine 3.6 mg/kg with epinephrine. If the procedure has been long, additional infiltration of local anesthesia may be required for closure.
other agents. However, they may cause profound respiratory depression. The short acting potent opioid, remifentanil has been reported to decrease tremors in patients with Parkinson’s disease [18].
Conscious Sedation General Anesthesia Sedation during functional neurosurgery is controversial in terms of whether to use it at all, when to use it, how much, and which agents. Sedation may be given throughout the whole procedure or just at the beginning and/or at the end. Most frequently all sedation is stopped during MER and stimulation testing. Commonly agents used are midazolam and propofol with or without opioids such as fentanyl or remifentanil [7,10,23–25]. Benzodiazepines are good anxiolytic agents with amnesic properties but they may alter the threshold for stimulation or prevent precise clinical assessment of the patient. Midazolam has a short duration of action and may be used in low doses prior to insertion of head pins for the frame. Propofol also has a short duration of action and can be given as a bolus or infusion. Propofol has been reported to reduce tremors, as well as to produce myoclonus movements [16,17,26]. Some patients may respond to sedative agents with paradoxical agitation or disinhibition. The management of these patients is to wake up the patient or change the sedative agent. A newer agent, dexmedetomidine, an alpha-2 agonist, has a more favorable pharmacological profile for sedation [27]. Dexmedetomidine provides adequate sedation while maintaining good ventilation and airway patency due to minimal respiratory depression and stable blood pressure [28]. Dexmedetomidine was shown not to impair the intensity of movement disorder in patients with Parkinson’s disease, not to interfere with MER, to provide hemodynamic stability, and decreased use of anti-hypertensive medications. Opioids are often added for analgesia and to improve sedation in conjunction with
The ability to use general anesthesia in specific patients may broaden the scope of patients that are suitable for functional neurosurgery. It would be difficult to treat some patients with severe continuous movements, severe pain and anxiety, or those who are uncooperative and children without the use of general anesthesia [29–31]. There are a few reports of the successful use of general anesthesia. Malteˇte et al. retrospectively reviewed the effects of general anesthesia on the postoperative outcome of patients with Parkinson’s disease following DBS insertion [29]. Fifteen patients with severe anxiety, respiratory problems and ‘‘off medication’’ dystonia requiring general anesthesia were matched with 15 patients treated under local anesthesia. General anesthesia included the administration of propofol sedation with no airway manipulation. The improvement in motor disability was greater in patients who received local anesthesia compared with those who received general anesthesia. Yamada et al. compared 15 patients who required general anesthesia because of severe psychosis and anxiety to 10 patients with local anesthesia [30]. General anesthesia was started after placement of the frame and imaging. Patients were intubated with fiberoptic guidance and had a balanced anesthetic with propofol, fentanyl, nitrous oxide, and sevoflurane. General anesthesia did not adversely effect postoperative improvements in motor and daily activity scores, except for ‘‘off medication’’ bradykinesia. Hertel et al. also looked at the feasibility of a DBS surgery with general anesthesia [31]. General anesthesia was used in nine patients for reasons of fear,
Anesthesia for functional neurosurgery
scoliosis, vertebral pain and heavy coughing. The entire procedure was with total intravenous anesthesia with remifentanil and propofol and endotracheal intubation was with rocuronium. Five minutes before the start of MER and test stimulation, remifentanil was stopped and propofol was lowered as far as possible. The authors concluded that MER were possible with their technique and clinical improvement was achieved in all patients. The adverse effects of general anesthesia in patients with functional disorders must be recognized. With Parkinson’s disease these are well documented and include respiratory and cardiovascular complications and neurological exacerbations including rigidity, confusion, and prolonged recovery time [11–13]. The ability to test the patient during the procedure is lost as well as the ability to identify potential complications and collateral effects of stimulation such as dysarthria and motor responses.
Complications The role of the anesthesiologist also includes the rapid recognition and treatment of complications during and immediately after the procedure. Respiratory complications, though not common, are of great concern [10]. Oversedation may result in decreased respiratory rate, desaturation, and loss of airway patency. There are difficulties in airway management when the patient is in a fixed stereotactic frame, which is rigidly attached to the operating room table. Also the frame may cover some or all of the patient’s mouth and nose. During the procedure if the patient becomes restless or attempts to move, acute airway obstruction may occur as the body shifts but the head remains fixed to the bed. Should this happen one needs to immediately release the frame from the operating room table to release the patient’s airway. This will usually resolve the problem. Supplemental oxygen
80
during the entire procedure will help decrease the incidences of desaturation, especially in sedated patients. Securing the airway is not often required but the anesthesiologist should always be ready. Patients may develop a sudden loss of consciousness from an intracranial bleed or from repetitive seizures. Appropriate airway equipment such as oral and nasal airways, laryngeal mask airways, endotracheal tubes, laryngoscope, and fiberoptic bronchoscopy should be readily available. Ideally, if possible, one should attempt to secure the airway without the removal of the patient’s head frame, so the surgery can be continued if needed. Other respiratory complications relate to the diseases of the patients. Patients with Parkinson’s disease may have restrictive pulmonary dysfunction from poor respiratory muscles function. This may lead to reduced forced vital capacity, reduced baseline arterial oxygen saturation, upper airway obstruction, and obstructive sleep apnea [11,32,33]. Respiratory insufficiency due to the absence of anti-Parkinson’s medications in the postoperative period may also occur. Patients with functional disorders are also at risk as they may have increased sensitivity to sedatives. Cardiovascular parameters are carefully monitored especially blood pressure ensuring that it remains within normal limits. This is frequently difficult in patients who are not sedated and may become anxious, agitated, and are uncomfortable. Hypertension is a common complication [8]. Vasoactive agents may be required to control the blood pressure during the insertion of the electrodes, as higher incidences of intracranial hemorrhage have been associated with higher arterial pressures [34,35]. The optimal level of blood pressure is controversial; one may use a systolic pressure below 160 mmHg systolic or 20% of patient’s preoperative blood pressure. Sedation will help to control blood pressure increases but, if this is not an option, then other agents need to be used. Commonly used agents include labetalol, hydrazaline, nitroglycerine,
1335
1336
80
Anesthesia for functional neurosurgery
sodium nitroprusside, or esmolol. Orthostatic hypotension may occur whenever a patient is placed in a sitting position and this may be aggravated by anti-Parkinson’s drugs [36]. Hypotension is more common after induction of general anesthesia as most anesthetic agents also have peripheral vasodilatation effects, adding to the possible effects of orthostatic hypotension, preoperative hypovolemia, and autonomic dysfunction. Other less common complications include venous air embolism [37–39]. During creation of the burr hole in awake patients, sudden vigorous coughing may be a sign of venous air embolism. Other signs are unexplained hypoxia and hypotension. Earlier detection may be possible with precordial Doppler monitoring. Tension pneumocephalus has also been reported during DBS insertion [40]. Neurological complications may occur during or after the procedure [34,35,41]. Focal deficits such as extremity weakness or confusion may not require any acute treatment by the anesthesiologist. Seizures may occur especially during stimulation testing. Most seizures are focal and do not require treatment. In cases of tonic clonic seizures small doses of midazolam and/or propofol can be given and the procedures can be resumed after control of the seizure. A sudden loss of consciousness due to an intracranial bleed or from a major neurological injury will require rapid treatment. Securing of the airway if needed can be done by any technique the anesthesiologist is comfortable with and may require releasing the head and/or removing the head from the frame. It may be necessary to take the patient for a computerized tomography scan to rule out a hematoma and even to plan for a craniotomy. All this may require the transfer of an unstable or anesthetized patient to other areas of the hospital. The internalization of the wires and pacemaker may be done on the same day as the DBS procedure. Otherwise the patient will return to the operating state at a later date during the same hospitalization. The internalization procedure is
performed under general anesthesia. At this time the influence of anesthetic agents on the DBS is not of concern. The anesthesiologist may use any technique of general anesthesia that is appropriate for each individual patient.
Vagal Nerve Stimulation Vagal nerve stimulation is used for the treatment of medically refractory epilepsy and some neuropsychiatric disorders. The procedure involves the surgical placement of an electrode wrapped around the left vagus nerve and then tunneled and connected to a generator pacemaker inserted into the chest wall. The whole procedure is performed under general anesthesia with endotracheal intubation [42]. The concerns of the anesthetic management relate to the medical condition of the patient. Possible complications during the procedure include cardiovascular events such as bradycardia or atrio-ventricular block, which will respond to appropriate treatment such as atropine, but may result in termination of the procedure. Postoperative complications include lower facial paralysis, laryngeal dysfunction including vocal cord paralysis, and other possible respiratory problems such as decreased respiratory effort, risk of aspiration, and worsening of obstructive sleep apnea.
Patients with DBS for Other Surgery With increased use of DBS for a variety of disorders, the anesthesiologist will encounter patients with an implanted DBS coming for other surgeries [42–45]. This also includes patients with vagus nerve stimulators. The electrical pulse generator pacemaker is usually located in the anterior chest wall of the patient. There is confusion and controversy on the safety of electrocautery use and magnetic resonance imaging (MRI).
Anesthesia for functional neurosurgery
Electrocautery may damage the leads and can temporarily suppress the neurostimulator output and/or reprogram the neurostimulator. During surgery a bipolar mode of electrocautery should be used if possible. When unipolar cautery is needed the ground plate should be as far away as possible from the neurostimulator and leads. External defibrillation may also damage the stimulator. The lowest possible energy for the current should be used and the placement of the defibrillator pads should be as far away as possible from the pacemaker and leads. The effects of the MRI include heating and torque of the DBS and leads, interference with the pacemaker program and image distortion. The stimulator should be turned off for the MRI and all safety recommendations for placement of the radiofrequency coils and parameter selection adhered to. Whenever the DBS is turned off, it important to resume its activity as soon as possible to prevent complications from lack of stimulation, especially if preoperative medications have been withheld. The correct use of a magnet to activate or deactivate the stimulators is complex and differs from cardiac pacemakers. One needs to consult the manufactures information for each generator.
Sterotactic Biopsy Surgery Stereotactic biopsy surgery uses computerized tomography or MRI for obtaining coordinated references to an extracranial system to guide the biopsy needle for accurate localization and sampling of intracranial lesions such as tumor or abscess. Traditionally, the procedure involved the placement of a head frame to the cranium, but fewer frame-based procedures are now performed due to the development of frameless neurological navigation imaging systems. These frameless stereotactic systems use scalp markers as fiducial points to relate the surgical instruments to a computer generated image. Most of these procedures are performed with the patients
80
awake or with minimal sedation to allow for neurological examination of the patient [46,47]. The role of the anesthesiologist is to provide sedation, if needed, and to monitor the patient’s vital signs and neurological system. In some circumstances the anesthesiologist may also need to travel with the patient to or from the radiology suite to the operating room. Acute intracranial complications can lead to a sudden change in level of consciousness. Respiratory complications may be potentially difficult to manage as the patient’s head frame is fixed to the operating room table and the frame may cover the nose and mouth of the patient, limiting access to the airway.
Epilepsy Surgery Epilepsy, a common chronic disorder, is usually treated with a variety of antiepileptic drugs. Medical treatment is deemed refractory if unacceptable side effects associated with the medications preclude adequate seizure control. Some of these patients are candidates for surgical resection of the epileptogenic focus. Surgery for partial seizure disorders involves the resection of a specific epileptogenic focus or a form of temporal lobectomy. Generalized seizures are treated by interruption of the seizure circuits by a corpus callosotomy or a hemispherectomy. A multidisciplinary evaluation including invasive and noninvasive investigations is performed to identify the origin of seizure activity and to evaluate the feasibility of performing surgery safely with minimal risk of neurological or cognitive injury [48,49].
Preoperative Localization of Epileptogenic Focus Many advances in neurological imaging techniques have reduced the need for invasive evaluation. Intracarotid sodium amytal injection (Wada test) is used to test for the lateralization of
1337
1338
80
Anesthesia for functional neurosurgery
language and memory. The drug is injected into the carotid artery via a femoral artery puncture. These tests are often done without the presence of an anesthesiologist. However, more recently the drug of choice has become etomidate, an intravenous anesthetic agent and this has necessitated the presence of an anesthesiologist to monitor and care for the patient. Invasive preoperative investigations for the localization of the epileptic focus in some patients may require surgery for the placement of epidural electrodes, or subdural grids and strip electrodes through burr holes or a craniotomy. These procedures are usually performed under general anesthesia. The anesthetic plan should take into consideration the concerns of a patient with epilepsy and the precautions that apply to any craniotomy. All anesthetic agents may be used as there are no electroencephalogram recordings at this time. Electrode plates or large grids are quite bulky and might require brain shrinkage with the use of mannitol and hyperventilation. These patients might develop postoperative problems with brain edema and require urgent removal of the grid due to the development of intracranial hypertension. After the insertion of intracranial electrodes a thiopental test may be conducted to evaluate some patients before surgery. An anesthesiologist is needed to be present for these tests as potential adverse effects, especially airway compromise, may occur [50].
Preoperative Assessment Epilepsy surgery may be with general anesthesia or with conscious sedation for an awake craniotomy [51,52]. The decision is usually made by the surgeon and is dependent on the location of the seizure focus, the need for intraoperative testing or localization of the seizure focus and eloquent brain function, and the ability of the patient to withstand an awake procedure. The preoperative anesthetic preparation of the patient is performed in the preoperative anesthesia consult clinic and
includes the routine assessment of the patient’s medical conditions, appropriate laboratory tests and investigations. Medications that the patient is on prior to surgery, such as anti-hypertensive agents, should be continued. The administration of anticonvulsant agents prior to surgery is done in consultation with the neurologist and surgeon. Premedication for the purpose of sedation is rarely required as these patients are usually well informed and agents that may influence the electroencephalogram, such as benzodiazepines, should be avoided. Appropriate preparation of the patient for the chosen technique of anesthesia is carried out. The patient should be informed of what to expect during the procedure including a rehearsal of stimulation testing. Specific considerations in patients with epilepsy include the associated medical problems such as psychiatric disorders, rare syndromes such as neurofibromatosis, multiple endocrine adenomatosis, and history of trauma. There are many adverse effects of antiepileptic drugs, which are dose-dependent and usually associated with long-term therapy [48]. Drugs may have neurological side effects such as sedation, confusion, learning impairment, ataxia, and gastrointestinal problems such as nausea and vomiting. Most anticonvulsants are metabolized by the liver. Thus, long-term usage will cause liver enzyme induction, which increases the rate of metabolism of other drugs, particularly anesthetic agents. Long-term therapy with phenytoin causes gingival hyperplasia with poor dentition and, potentially, difficult airway management. Carbamazepine can depress the hemopoietic system and, in rare cases, causes cardiac toxicity. Valproic acid might occasionally result in thrombocytopenia and platelet dysfunction.
Awake Craniotomy The reasons for having an awake patient for all or part of the procedure are for better
Anesthesia for functional neurosurgery
electrocorticographic localization of the seizure focus without the influence of general anesthetic agents, direct electrical stimulation of the cerebral cortex to delineate eloquent areas of brain function in order to preserve them during surgical resection, and for continuous clinical neurological monitoring of the patient [53–57]. The success of an awake craniotomy depends on the proper selection and preparation of the patient. Psychological preparation of the patient by the neurologist and surgeon and should be continued by the anesthesiologist. The challenge is to have the patient comfortable enough to remain immobile through a long procedure, but sufficiently alert and cooperative to comply with testing. The analgesic and sedative drugs employed must have minimal interference with electroencephalographic and stimulation testing.
Intraoperative Anesthetic Considerations The operating room is an unfamiliar and frightening place for most patients. The environment should be made quiet and comfortable with an appropriate room temperature. The operating room table should be as soft as possible with padding for all pressure points and adequate supports. The position of the patient for surgery will be dependent on the procedure. Ideally for an awake patient, the lateral position is more comfortable, allows for better visualization of the patient’s face and treatment of complications such as airway obstruction, nausea and vomiting or seizures. However, the supine position is possible as long as one maintains a good view of the patient’s face and has the ability to manipulate the airway. This can be accomplished by appropriate draping. Careful positioning of the head and neck is required when the headframe is secured to the operating room table to ensure patient comfort.
80
Standard monitors include an electrocardiogram, blood pressure, pulse oximeter, and end tidal capnography. Other monitors such as invasive arterial monitoring are added depending on the practice within each institution and patient needs. To decrease patient discomfort, urinary catheters can be avoided for shorter procedures and if intravenous fluid administration is kept to a minimum. Supplemental oxygen is given to all patients by mask or nasal prongs with an outlet for the monitoring of end tidal CO2 and respiratory rate. Nasal prongs are often tolerated better by the patient and also allow the patient to speak freely for speech testing.
Anesthetic Agents and Techniques Local anesthesia is used for insertion of head pins and for the surgical incision. Scalp nerve blocks are performed using long-acting local anesthetic agents, such as bupivacaine or ropivacaine with the addition of epinephrine [21,22]. Lidocaine, which has a faster onset, may be added and also used to infiltrate areas that are still painful during the procedure, such as dura. There are two commonly used anesthetic techniques for sedation and analgesia; ‘‘conscious sedation’’ and ‘‘asleep awake asleep.’’ The conscious sedation technique implies a level of sedation where the patient has a minimally depressed level of consciousness but is able to maintain their own airway and respond to verbal stimulation [55]. Generally, there is no manipulation of the airway other than the administration of supplemental oxygen via nasal prongs or cannula, or a facemask. Traditionally ‘‘neurolept anesthesia’’ was employed using fentanyl and droperidol [53,54]. Opioids used for analgesia have included fentanyl, sufentanil, and alfentanil [58,59]. Now more commonly an infusion of the short acting potent opioid, remifentanil, is used as this drug is easy to titrate and does not accumulate [60–63]. Though opioids have been shown to produce
1339
1340
80
Anesthesia for functional neurosurgery
epileptiform activity in patients with epilepsy, and may also produce myoclonic movements that may clinically resemble seizure-like activity, this does not appear to be a problem during awake craniotomies as the doses used are generally small [48,63]. Propofol is the usual drug of choice for sedation [60–62,64–66]. It has a rapid onset of action and fast offset so that the level of sedation can be quickly changed and can be given without airway protection. However, propofol may suppress epileptogenic activity and has to be discontinued at least 15–30 min prior to electrocorticographic recordings [65,66]. Propofol may also cause tonic–clonic movements mimicking seizures, but will also effectively stop seizures [26,65]. All these agents can be administered as continuous infusions, intermittent boluses or in any combination. Target controlled sedation and the use of patient controlled anesthesia have also been described [60,66]. Dexmedetomidine has also been used as an adjunct for sedation and analgesia. Its benefits include minimal risk of respiratory depression, hemodynamic stability, and anesthetic sparing effects [27,67–71]. The addition of dexmedetomidine may improve the safety of patients during awake craniotomy and still allow for electrocorticography, cortical mapping, and neurological testing though it is unclear whether it has any anticonvulsant effect [70]. Anti-emetic agents (dimenhydrinate, prochlorperazine, metoclopramide, odansteron, granisetron) may also be needed and do not affect electrocorticography. Sedation is stopped just prior to stimulation testing and may be resumed after all testing is complete. During resection of the lesion heavy sedation should be avoided if ongoing neurological testing is required. In addition to the use of sedation and analgesic medications, continuous communication and reassurance with the patient is critical. Many nonpharmacological measures can help the patient tolerate these procedures. This includes warning the patient in advance of any painful or disturbing stimuli, especially the loud noise
caused by drilling of the bone, which can be frightening though not painful. Often small comforts such as allowing the patient to move intermittently, wetting their lips, ensuring that they are not too hot or cold and just holding the patient’s hand will go a long way to help the patient through the procedure. Another approach to the awake craniotomy is the use of the ‘‘asleep awake asleep’’ technique [67–69,72]. After arrival in the operating room, the patient is administered general anesthesia and the airway is secured with a laryngeal mask or endotracheal tube. The advantage of the laryngeal mask airway is its easier placement, less coughing and possibly less laryngospasm with removal. Both inhalation and intravenous anesthetic agents may be used, with or without controlled ventilation. When the craniotomy is complete, the patient is awakened and the airway removed to allow for the electrocorticographic recordings and stimulation testing. After the testing is completed, general anesthesia is resumed and the airway may or may not be reinserted. The reinsertion of the airway is the main disadvantage of the asleep awake asleep technique. Advantages of this technique include increased patient comfort and tolerance during surgery especially for longer procedures and a secured airway with the ability to use hyperventilation.
Complications Intraoperative complications include respiratory and cardiovascular changes, seizures, the restless patient, nausea and vomiting and the requirement to induce general anesthesia during conscious sedation [73,74]. Other less frequent complications include air embolism [75]. Oxygen desaturation and/or airway obstruction may result from oversedation, seizures, mechanical obstruction, or loss of consciousness from an intracranial event. The anesthesiologist needs to have a preplanned approach to deal with these
Anesthesia for functional neurosurgery
problems. If the problem is just simply oversedation then the treatment is to stop the administration of anesthesia agents and if needed, support the airway with a chin-lift or with a mask and assisted ventilation. However, if the loss of the airway is due to continuous seizures and/or their treatment, or loss of consciousness due to an intracranial event, one may need to secure a definite airway for the remainder of the procedure. The choice of the airway and the technique of insertion will depend on the skill of the anesthesiologist, the anatomy of the patient, and the requirements for the remainder of the procedure. Insertion of a laryngeal mask or endotracheal tube may require the induction of anesthesia, and the use of a direct laryngoscopy or fiberoptic bronchoscopy. During intubation the surgeon can assist by protecting the sterile areas and changing the position of the patient’s head if needed. During the procedure seizures can occur at any time, especially if the patient has been off their usual anticonvulsant medications. Short seizures may not require any treatment. Seizures that are convulsive or generalized need to be treated by protecting the patient from injury, ensuring patent airway, adequate oxygenation, and circulatory stability. Prior to electrocorticographic recordings, seizures can be treated with a small dose of thiopental or propofol, afterward benzodiazepines or longer acting anti-convulsants may be used. Other intraoperative problems include excessive pain and discomfort some of which can be predicted and the patient warned about such as the scalp block and the noise of the drill. Many factors can influence the incidence of nausea and vomiting, such as anxiety, medications, and surgical stimulation, especially the stripping of dura and manipulation of the temporal lobe or meningeal vessels. Intraoperative nausea and vomiting can be treated with any of the common anti-emetic agents. Some patients may become extremely restless, agitated, or uncooperative during the procedure. If the patient
80
becomes disinhibited from the sedative agents, especially propofol, the treatment is to ‘‘lighten’’ the patient’ s level of sedation so one can communicate with them, change the anesthetic agents being used, deepen the level of sedation, or in some situations one may have to convert to general anesthesia. The incidence of conversion to general anesthesia is low [54,73,74]. Other less common complications include local anesthetic toxicity, a tight brain, and cardiovascular changes.
General Anesthesia The reason for choosing general anesthesia is the preference of the surgeon and/or the inability of the patient to tolerate an awake craniotomy. However, with the advances in preoperative neurological imaging, functional testing, and the use of frameless stereotactic surgery for localization of the epileptic focus, the need for an awake patient has greatly decreased. The challenge for general anesthesia if intraoperative localization of the epileptic focus is needed is to provide good conditions for electrocorticography and for motor testing, ensuring that the influence of the anesthetic agents be kept at a minimum, but also avoiding long periods of potential awareness on the part of the patient. Specific preoperative preparation is to inform patients of the possibility that awareness and recall may occur at the time of electrocorticographic recording, but reassuring them that this will be brief and painless. All anesthetics will affect electrocorticography but the use of the shorter acting anesthetic agents, either inhalation agents and/or intravenous agents will allow for faster changes in the depth of anesthesia. During the time of recording all or most of the anesthetic agents are stopped, as much as possible. Another approach would be to do the initial craniotomy and electrocorticography and cortical mapping with the patient awake and then induce general anesthesia.
1341
1342
80
Anesthesia for functional neurosurgery
The difficulty of this technique is the induction of anesthesia and securing of the airway with an exposed brain and the patient’s head in a frame fixed to the operating room table. If no intraoperative testing or recording is to be performed, the anesthetic techniques and agents are at the discretion of the anesthesiologist and the management of the patient is as for any patient undergoing a craniotomy. Some specific concerns with general anesthesia in patients with epilepsy include the effects of long-term anticonvulsant therapy, which may lead to increased dosage requirements of opioids and neuromuscular blocking agents [48]. If the patient has had a recent craniotomy or burr holes for electrode placement, intracranial air might still be present and nitrous oxide should be avoided to prevent complications from an expanding pneumocephalus. Complications that may occur during epilepsy surgery are similar to those for any craniotomy, however, severe bradycardia is a common occurrence with resection of the amygdalo-hippocampus [76].
Intraoperative Recordings and Activation Electrocorticography is performed during surgery after opening of the dura by the placement of electrodes directly on the cortex over the area predetermined to be epileptogenic as well as on adjacent cortex. Additional recordings can be performed with microelectrodes placed into the cortex or depth electrodes into the amygdala and hippocampal gyrus. Stimulation of epileptogenic focus is possible pharmacologically if insufficient information to define the seizure focus adequately is obtained during routine electrocorticography. Traditionally methohexital or thiopental were administrated to gradually increase beta activity in normal functioning neural tissue, but not in the seizure focus. Other agents used include propofol, etomidate, or small dose of opioids in patients who are awake [48,49,63,77,78]. If the patient is
under general anesthesia, other agents can be used such as alfentanil, remifentanil or inhalation agents such as enflurane, sevoflurane with or without hypocarbia [79–82].
Awake Craniotomy for Tumor Surgery Awake craniotomy for tumor surgery is an accepted procedure that allows for mapping of brain function by cortical stimulation to delineate areas of eloquent function, such as speech, sensory, and motor in order to preserve them during resection of the tumor [83–86]. Another possible advantage is the avoidance of general anesthesia and earlier discharge from hospital [87]. The anesthetic management of patients for an awake craniotomy for tumor surgery is in most respects similar to that for epilepsy surgery except the concerns of the anesthetic effects on electroencephalography are not present. The goals of the anesthetic management are to have a comfortable patient who is able to stay immobile on an operating room table for the duration of the procedure and yet be alert and cooperative to comply with cortical mapping. These goals can be accomplished by adequate preparation of the patient, a comfortable environment, appropriate administration of analgesic, and sedative medications, ongoing communication and support of the patient, and rapid treatment of complications.
Preoperative Assessment The success of an awake craniotomy for tumor surgery depends on the proper selection and preparation of the patient. Not all patients are suitable for awake craniotomy such as the extremely anxious patient. Language barriers may make the procedure more difficult, but should not rule out an awake craniotomy if a suitable interpreter is available. The patient should be
Anesthesia for functional neurosurgery
informed of what to expect during the procedure including a rehearsal of stimulation testing. Otherwise the preoperative assessment is similar to any patient with a brain tumor and the concerns of intracranial pathology with respect to intracranial pressure and neurological deficits. Medications that the patient is on prior to surgery, especially steroids (dexamethasone) and anti-convulsants, should be continued. Premedication is not usually necessary. Frequently patients with brain tumors may have only had a short period of time to accept their diagnosis in contrast to patients with epilepsy who have a chronic disorder. The concept of having brain surgery and being awake in the operating room can be terrifying. Prior to bringing the patient in, the operating room should be completely ready so that all members of the team, anesthesiologist, nursing staff and surgeon, can devote their attention to the patient. For the anesthesiologist, preparation of the operating room also includes the preparation of all anesthetic drugs, monitors, and equipment that are required for an awake procedure as well as for general anesthesia should this be required and for the treatment of any complication. The position of the patient for surgery will be dependent on where the lesion is. Supine, lateral, and semi-prone have all been used successfully. It is important to maintain a good view of the patient’s face and have the ability to manipulate the airway. Similar to epilepsy surgery only standard monitors, electrocardiogram, noninvasive blood pressure, pulse oximeter, and end tidal capnography are essential. Other monitors such as invasive blood pressure, urinary catheter are added as indicated by the procedure, the patient and routine practice in each institution.
Anesthetic Agents and Techniques Overall the anesthetic management of the patient is similar to that described above for epilepsy
80
surgery. Many different anesthetic agents and techniques have been successfully used for tumor surgery, as the influence of drugs on the electroencephalogram is not of concern. For a conscious sedation technique the most common combinations used include propofol and/or midazolam and fentanyl or infusion of remifentanil [88,89]. Dexmedetomidine has also been used for tumor surgery [67–69]. Generally there is no manipulation of the airway other than the administration of supplemental oxygen via nasal prongs or cannula, or a facemask. Once the patient is positioned on the operating table, sedation and analgesia can be started. If the surgical approach includes the use of navigational equipment the head will be placed in a head frame with pins using infiltration with local anesthetic. The incision area is infiltrated with local anesthetic agents or scalp nerve blocks may be used. When the dura is approached the amount of sedation should be decreased so that the patient will be alert during testing. In addition to the use of sedation and analgesic medications, continuous communication, reassurance, and the use of nonpharmacological measures is important. In some patients where the resection margins of the tumor are very close to critical eloquent areas, it is useful to continue testing, for speech or motor function, during resection and thus heavy sedation should not be used. The asleep awake asleep technique is a commonly used for tumor surgery [90–92]. The patient is administered general anesthesia with the use of a laryngeal mask or endotracheal tube for airway management during the craniotomy, then awakened and the airway is removed for cortical mapping and testing. After the testing is completed, general anesthesia is resumed with or without reinsertion of the airway for the resection of the tumor and closure. Advantages of this technique for tumor surgery are the increased patient comfort and tolerance for long tumor resections and the ability to use hyperventilation to treat increased brain mass.
1343
1344
80
Anesthesia for functional neurosurgery
Complications Most intraoperative complications are similar to those occurring during epilepsy surgery. Respiratory changes such as desaturation or decrease in respiratory rate may occur from oversedation, an intracranial event or seizures. Rapid treatment is required to prevent airway obstruction. The techniques (describe in section above) used will be dependent on the condition of the patient and the expertise of the anesthesiologist. Some patients with brain tumors initially present with a seizure and thus it is possible for the patient to have an intraoperative seizure. However, most seizures, including patients with no prior history, will occur when electrical stimulation is performed during cortical mapping. Treatment includes protection of the patient, and if needed, a small dose of propofol or midazolam. Repeated doses may be required if seizures continue and longer acting anticonvulsants may also be required such as phenytoin. The incidence of nausea and vomiting during awake craniotomy for tumor surgery is very low compared with epilepsy surgery [73,74]. Whether this is a result of the anesthetic drugs used or the type of surgery is unclear. Most tumor surgery does not involve resection or pulling of the deep brain structures, which are known to cause vomiting. Venous air embolism has also been reported [93]. Other common complications include excessive pain, restlessness, agitation, or uncooperativeness. Rarely one may have to convert to general anesthesia.
possible in the older mature child. In children coexisting conditions with multiple organ system involvement or significant psychological, and behavior problems needs to be reviewed preoperatively. As well the parents may be very closely involved in the child’s management and will require consideration as well. Cerebral hemispherectomy and corpus callosotomy are used in children for the treatment of generalized seizures. These procedures are usually done under general anesthesia as they involve a large craniotomy. The major concerns of these lengthy procedures are the possibility of extensive blood loss and air emboli as the surgical site is close to major vessels and sinuses. The process of Gamma Knife radiosurgery for children with a brain tumor or AVM is a long procedure, and most children are under general anesthesia for most of the procedures. The whole process usually involves the application of the head frame, angiography, scanning in the MRI and CT prior to the radiosurgery and traveling between these sites. The anesthetic agents used may be inhalation or intravenous depending on the supply of gases and scavenging within each of the locations. Standard monitoring is required at all sites and during transfers. The usual anesthetic practice differs in the setting of Gamma Knife room as the anesthesiologist must remain outside of the room and can only observe the patient from monitors displaying real-time pictures of the patient and vital signs.
Postoperative Care Pediatric Patients The management of pediatric patients differs in that very few would be able to tolerate an awake procedure. Most functional neurosurgical procedures are performed under general anesthesia including the use of Gamma Knife. Awake craniotomy, especially using the asleep awake asleep technique of anesthesia, and DBS procedures are
In the initial postoperative period all patients should be closely observed in the postanesthetic care unit and then transferred to an observational unit or ward accustomed to caring for neurosurgical patients with functional disorders and epilepsy. Hemodynamic and respiratory parameters and neurological signs should be repeatedly checked and recorded according to the
Anesthesia for functional neurosurgery
routine practice of the institution. The patient should be nursed in 30 degree head up position to improve ventilation, reduce face, neck, and airway edema and to facilitate cerebral venous and cerebrospinal fluid drainage. Discharge from the postanesthetic care unit to a suitable location should be only when the patient is stable and discharge criteria have been met. Potential for rapid neurological deterioration requires regular and frequent monitoring of the neurological status. Any change in the neurological status of the patient should be immediately assessed. A CT Scan may be needed to rule out an intracranial hemorrhage. Aggressive rapid management of oxygenation, airway, and any ventilation problem is important considering the detrimental effects of hypoxia and hypercarbia on the brain. Cardiovascular monitoring will help to avoid prolonged episodes of hypotension and hypertension both of which may lead to adverse neurological outcomes. Other routine physiological parameters should be considered and monitored for such as temperature, blood glucose, intravascular volume, osmolality, cerebral perfusion pressure and intracranial pressure. Complications may also occur such as seizures, which require immediate attention to the airway, protection of the patient, and pharmacological treatment with benzodiazepines or propofol initially, and then barbiturates and/or phenytoin. Appropriate postoperative analgesia suitable for each patient is important. Some patients may not require any analgesia; others will have a headache or incision pain with varying intensity. A scalp block that is still present may reduce postoperative pain. Initially intravenous fentanyl is a good analgesic. Codeine still remains the drug of choice in many neurosurgical centers [94,95]. Traditionally, potent opioids such as morphine are avoided due to concerns with hypercarbia, miosis, sedation, and nausea. However, modest intravenous doses are generally believed to be safe. Postoperative nausea and vomiting are common complications in neurosurgical patients but
80
the incidence is markedly reduced in patients undergoing awake procedures. Any standard anti-emetic agents are suitable.
Summary The role of the anesthesiologist will continue to be important in the treatment of patients with functional disorders, epilepsy, and brain tumors as more patients will be treated by these different neurosurgical techniques in stereotactic surgery, and awake craniotomy. With the development of new surgical, imaging and monitoring techniques, the anesthetic management of these patients will also continue to evolve and will require continuous change and development. The appropriate anesthetic management of these patients is critical to the success of the operations and will remain a challenge for the anesthesiologist.
References 1. Pereira EAC, Green AL, Nandi D, et al. Deep brain stimulation: indications and evidence. Expert Rev Devices 2007;4:591-603. 2. Halpern C, Hurtig H, Jaggi J, et al. Deep brain stimulation in neurologic disorders. Parkinsonism Relat Disord 2007;13(1):1-16. 3. Lozano AM, Mahant N. Deep brain stimulation surgery for Parkinson’s disease: mechanisms and consequences. Parkinsonism Relat Disord 2004;10:S49-57. 4. Machado A, Rezai AR, Kopell BH, et al. Deep brain stimulation for Parkinson’s disease: surgical technique and perioperative management. Mov Disord 2006;21 Suppl 14:S247-58. 5. Abosch A, Lozano A. Stereotactic neurosurgery for movement disorders. Can J Neurol Sci 2003;30:S72-82. 6. Marangell LB, Martinez M, Jurdi RA, et al. Neurostimulation therapies in depression: a review of new modalities. Acta Psychiatr Scand 2007;116:174-81. 7. Ollinet C, Bedague D, Carcey J, et al. Functional surgery for movement disorders: implications for anaesthesia. Ann Fr Anesth Re´anim 2004;23:428-32. 8. Santos P, Valero R, Arguis MJ, et al. Preoperative adverse events during stereotactic microelectrode-guided deep brain surgery in Parkinson’s disease. Rev Esp Anestesiol Reanim 2004;51:523-30.
1345
1346
80
Anesthesia for functional neurosurgery
9. Fa`bregas N, Craen RA. Endscopic and stereotactic neurosurgery. Curr Opin Anesthesiol 2004;17:377-82. 10. Venkatraghavan L, Manninen P, Mak P, et al. Anesthesia for functional neurosurgery. Review of complications. J Neurosurg Anesthesiol 2006;18:64-7. 11. Mason LJ, Cojocaru TT, Cole DJ. Surgical intervention and anesthetic management of the patient with Parkinson’s disease. Int Anesthesiol Clin 1996;34:133-50. 12. Nicholson G, Pereira AC, Hall GM. Parkinson’s disease and anaesthesia. Br J Anaesth 2002;89(6): 904-16. 13. Burton DA, Nicholson G, Hall GM. Anaesthesia in elderly patients with neurodegenerative disorders: special considerations. Drugs Aging 2004;21:229-42. 14. Zanettini R, Antonini A, Gatto G, et al. Valvular heart disease and the use of dopamine agonist for Parkinson’s disease. N Engl J Med 2007;356(1): 39-46. 15. Gillman PK. Monoamine oxidase inhibitors, opioid analgesics and serotonin toxicity. Br J Anaesth 2005;95(4):434-41. 16. Anderson BJ, Marks PV, Futter ME. Propofol: contrasting effects in movement disorders. Br J Neurosurg 1994;8:387-8. 17. Krauss KJ, Akeyson EW, Giam P, et al. Propofol-induced dyskinesias in Parkinson’s disease. Anesth Analg 1996;83:420-2. 18. Bo¨hmdorfer W, Schwarzinger P, Binder, S, et al. Temporary suppression of tremor by remifentanil in a patient with Parkinson’s disease during cataract extraction under local anesthesia. Anaesthesist 2003;52(9):795-7. 19. Watson R, Leslie K. Nerve blocks versus subcutaneous infiltration for stereotactic frame placement. Anesth Analg 2001;92:424-7. 20. Chevrier E, Fraix V, Krack P, et al. Is there a role for physiotherapy during deep brain stimulation surgery in patients with Parkinson’s disease? Eur J Neurol 2006;13:496-8. 21. Heavner JE. Local Anesthetics. Curr Opin Anaesthesiol 2007;20:336-42. 22. Costello TG, Cormack JR, Hoy C, et al. Plasma ropivacaine levels following scalp block for awake craniotomy. J Neurosurg Anesthesiol 2004;16:147-50. 23. Fukuda M, Kameyama S, Noguchi R, et al. Intraoperative monitoring for functional neurosurgery during intravenous anesthesia with propofol. No Shinkei Geka 1997:25:231-7. 24. Murata J, Sawamura Y, Kitagawa M, et al. Minimally invasive stereotactic functional surgery using an intravenous anesthetic propofol and applying image Fusion and AtlasPlan. No To Skinkei 2001;53:457-62. 25. Gray H, Wilson S, Sidebottom P. Parkinson’s disease and anaesthesia. Br J Anaesth 2003;90:524-5. 26. Reynolds LM, Koh JL. Prolonged spontaneous movement following emergence from propofol/nitrous oxide anesthesia. Anesth Analg 1993;76:192-3. 27. Hsu Y-W, Cortinez LI, Robertson KM, et al. Dexmedetomidine pharmacodynamics: I. Crossover comparison
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
of the respiratory effects of dexmedetomidine and remifentanil in healthy volunteers. Anesthesiology 2004;101:1066-76. Rozet I, Muangman S, Vavilala SM, et al. Clinical experience with dexmedetomidine for implantation of deep brain stimulators in Parkinson’s disease. Anesth Analg 2006;103:1224-8. Malteˆte D, Navarro S, Welter ML. Subthalamic stimulation in Parkinson disease: with or without anesthesia? Arch Neurol 2004;61:390-2. Yamada K, Goto S, Kuratsu J, et al. Stereotactic surgery for subthalamic nucleus stimulation under general anesthesia: A retrospective evaluation of Japanese patients with Parkinson’s disease. Parkinsonism Relat Disord 2007;13:101-7. Hertel F, Zu¨chner M, Weimar I, et al. Implantation of electrodes for deep brain stimulation of the subthalamic nucleus in advanced Parkinson’s disease with the aid of intraoperative microrecording under general anesthesia. Neurosurgery 2006;59:1138-45. Easdown LJ, Tessler MJ, Minuk J, Upper airway involvement in Parkinson’s disease resulting in postoperative respiratory failure. Can J Anaesth 1995;42(4): 344-7. Fa`bregas N, Rapado J, Gambu´s PL, et al. Modeling of the sedative and airway obstruction effects of propofol in patients with Parkinson’s disease undergoing stereotactic surgery. Anesthesiology 2002;97:1378-86. Binder DK, Rau GF, Starr PA. Risk factors for hemorrhage during microelectrode-guided deep brain stimulator implantation for movement disorders. Neurosurgery 2005;56:722-32. Gorgulho A, De Salles AA, Frighetto L, et al. Incidence of hemorrhage associated with electrophysiological studies performed using macroelectrodes and microelectrodes in functional neurosurgery. J Neurosurg 2005;102:888-96. Bouhaddi M, Vuillier F, Fortrat JO, et al. Impaired cardiovascular autonomic control in newly and long-term-treated patients with Parkinson’s disease: involvement of L-dopa therapy. Auton Neurosci 2004;116:30- 8. Suarez S, Ornaque I, Fabregas N, et al. Venous air embolism during Parkinson surgery in patients with spontaneous ventilation. Anesth Analg 1999;88:793-4. Moitra V, Permut TA, Penn RM, et al. Venous air embolism in an awake patient undergoing placement of deep brain stimulators. J Neurosurg Anesthesiol 2004;16:321-2. Deogaonkar A, Avitsian R, Henderson MJ. Venous air embolism during deep brain stimulation surgery in an awake supine patient. Stereotact Funct Neurosurg 2005;83(1): 32-5. Jain V, Prabhakar H, Rath GP, et al. Tension pneumocephalus following deep brain stimulation surgery with bispectral index monitoring. Eur J Anaesthesiol 2007;24(2):203-4.
Anesthesia for functional neurosurgery
41. Higuchi Y, Iacono RP. Surgical complications in patients with Parkinson’s disease after posteroventral pallidotomy. Neurosurgery 2003;52:558-71. 42. Hatton KW, McLarney JT, Pittman T, et al. Vagal nerve stimulation: overview and implications for anesthesiologists. Anesth Analg 2006;103:1241-9. 43. Davies R.G. Deep brain stimulators and anaesthesia. Br J Anaesth 2005;95(3):424-7. 44. Nutt JG, Anderson VC, Peacock JH, et al. DBS and diathermy interaction induces severe CNS damage. Neurology 2001;56:1384-6. 45. Dagtekin O, Berlet T, Gerbershagen JH. Anesthesia and deep brain stimulation: postoperative akinetic state after replacement of impulse generators. Anesth Analg 2006;103:784. 46. Tan TK, Manninen PH. Anesthesia for stereotactic surgery. Semin Anesth 2000;19:292-9. 47. Bilgin H, Mogol EB, Bekar A, et al. A comparison of alfentanil, fentanyl and remifentanil on hemodynamic and respiratory parameters during stereotactic brain biopsy. J Neurosurg Anesthesiol 2006;18:179-84. 48. Kofke WA, Tempelhoff R, Dasheiff RM. Anesthetic implications of epilepsy, status epilepticus, and epilepsy surgery. J Neurosurg Anesthesiol 1997;9:349-72. 49. Sahjpaul RL. Awake craniotomy: controversies, indications and techniques in surgical treatment of temporal lobe epilepsy. Can J Neurol Sci 2000;27:S55-63. 50. Kofke WA, Dasheiff RM, Dong ML, et al. Anesthetic care during thiopental tests to evaluate epileptic patients for surgical therapy. J Neurosurg Anesthesiol 1993;5:164-70. 51. Herrick IA, Gelb AW. Anesthesia for temporal lobe epilepsy surgery. Can J Neurol Sci 2000;27:S64-7. 52. Arango MF, Steven DA, Herrick IA. Neurosurgery for the treatment of epilepsy. Clin Opin Anaesthesiol 2004;17:383-7. 53. Manninen PH, Contreras J. Anesthetic considerations for craniotomy in awake patients. Int Anesthesiol Clin 1986;24:157-74. 54. Archer DP, McKenna JMA, Morin L, et al. Conscioussedation analgesia during craniotomy for intractable epilepsy: a review of 354 consecutive cases. Can J Anaesth 1988;35:338-44. 55. Aglio LS, Gugino LD. Conscious sedation for intraoperative neurosurgical procedures. Tech Neurosurg 2001;7:52-60. 56. Frost EAM, Booj LHDJ. Anesthesia in the patient for awake craniotomy. Curr Opin Anaesthesiol 2007;20:331-5. 57. Hans P, Bonhomme V. Anesthetic management for neurosurgery in awake patients. Minerva Anestesiol 2007;73:507-12. 58. Welling EC, Donegan J. Neuroleptanalgesia using alfentanil for awake craniotomy. Anesth Analg 1989;68:57-60. 59. Gignac E, Manninen PH, Gelb AW. Comparison of fentanyl, sufentanil and alfentanil during awake craniotomy for epilepsy. Can J Anaesth 1993;40:421-4.
80
60. Hans P, Bonhomme V, Born JD, et al. Target-controlled infusion of propofol and remifentanil combined with bispectral index monitoring for awake craniotomy. Anaesthesia 2000;55:255-9. 61. Berkenstadt H, Perel A, Hadani M, et al. Monitored anesthesia care using remifentanil and propofol for awake craniotomy. J Neurosurg Anesthesiol 2001;13:246-9. 62. Keifer JC, Dentchev D, Little K, et al. A retrospective analysis of a remifentanil/propofol general anesthetic for craniotomy before awake functional brain mapping. Anesth Analg 2005;101:502-8. 63. Herrick IA, Craen RA, Blume WT, et al. Sedative doses of remifentanil have minimal effect on ECoG spike activity during awake epilepsy surgery. J Neurosurg Anesthesiol. 2002;14(1):55-8. 64. Silbergeld DL, Mueller WM, Colley PS, et al. Use of propofol (Diprivan) for awake craniotomies: Technical note. Surg Neurol. 1992;38:271-2. 65. Samra SK, Sneyd JR, Ross DA, et al. Effects of proprofol sedation on seizures and intracranially recorded epileptiform activity in patients with partial epilepsy. Anesthesiology. 1995;82:843-51. 66. Herrick IA, Craen RA, Gelb AW, et al. Propofol sedation during awake craniotomy for seizures: patient-controlled administration versus neurolept analgesia. Anesth Analg. 1997;84:1285-91. 67. Mack PF, Perrine K, Kobylarz E, et al. Dexmedetomidine and neurocognitive testing in awake craniotomy. J Neurosurg Anesthesiol. 2004;16:20-5. 68. Ard JL, Bekker AY, Doyle WK. Dexmedetomidine in awake craniotomy: a technical note. Surg Neurol. 2005;63:114-7. 69. Bekker, AY, Kaufman B, Samir H, et al. The use of dexmedetomidine infusion for awake craniotomy. Anesth Analg. 2001;92:1251-3. 70. Souter MJ, Rozet I, Ojemann JG, et al. Dexmedetomidine sedation during awake craniotomy for seizure resection: effects on electrocorticography. J Neurosurg Anesthesiol. 2007;19:38-44. 71. Oda Y, Toriyama S, Tanaka K, et al. The effect of dexmedetomidine on electrocorticography in patients with temporal lobe epilepsy under sevoflurane anesthesia. Anesth Analg. 2007;105:1272-7. 72. Huncke K, Van de Wiele B, Fried I, et al. The asleepawake–asleep anesthetic technique for intraoperative language mapping. Neurosurgery 1998;42;1312-7. 73. Nikas DC, Danks RA, Black PM. Tumor surgery under local anesthesia. Tech Neurosurg. 2001;7:70-84. 74. Skucas AP, Artru AA. Anesthetic complications of awake craniotomies for epilepsy surgery. Anesth Analg. 2006:102:882-7. 75. Scuplak SM, Smith M, Harkness WF. Air embolism during awake craniotomy. Anaesthesia 1995;50:338-40. 76. Sato K, Shamoto H, Yoshimoto T. Severe bradycardia during epilepsy surgery. J Neurosurg Anesthesiol 2001:13:329-32.
1347
1348
80
Anesthesia for functional neurosurgery
77. Smith M, Smith SJ, Scott CA, et al. Activation of the electrocorticogram by propofol during surgery for epilepsy. Br J Anaesth 1996;76:499-502. 78. Herrick IA, Craen RA, Gelb AW, et al. Propofol sedation during awake craniotomy for seizures: electrocorticographic and epileptogenic effects. Anesth Analg 1997;84:1280-4. 79. Manninen PH, Burke SJ, Wennberg R, et al. Intraoperative localization of an epileptogenic focus with alfentanil and fentanyl. Anesth Analg 1999;88:1101-6. 80. McGuire G, El-Beheiry H, Manninen PH, et al. Activation of electrocorticographic activity with remifentanil and alfentanil during neurosurgical excision of epileptogenic focus. Br J Anaesth 2003;91:651-5. 81. Watts AD, Herrick IA, McLachlan RS, et al. The effect of sevoflurane and isoflurane anesthesia on interictal spike activity among patients with refractory epilepsy. Anesth Analg 1999;89:1275-81. 82. Kurita N, Kawaguchi M, Hoshida T, et al. The effects of sevoflurane and hyperventilation on the electrocorticogram spike activity in patients with refractory epilepsy. Anesth Analg 2005;101:517-23. 83. Danks RA, Rogers M, Aglio LS, et al. Patient tolerance of craniotomy performed with the patient under local anesthesia and monitored conscious sedation. Neurosurgery 1998;42:28-36. 84. Taylor MD, Bernstein M. Awake craniotomy with brain mapping as the routine surgical approach to treating patients with supratentorial intraaxial tumors: a prospective trial of 200 cases. J Neurosurg 1999;90:35-41. 85. Danks RA, Aglio LS, Gugino LD, et al. Craniotomy under local anesthesia and monitored conscious sedation for the resection of tumors involving eloquent cortex. J Neurooncol 2001;49:131-9.
86. Berkenstadt H, Ram Z. Monitored anesthesia care in awake craniotomy for brain tumor surgery. IMAJ 2001;3:297-300. 87. Blanshard HJ, Chung F, Manninen PH, et al. Awake craniotomy for removal of intracranial tumor: considerations for early discharge. Anesth Analg 2001;92:89-94. 88. Johnson KB, Egan TD. Remifentanil and propofol combination for awake craniotomy: case report with pharmacokinetic simulations. J Neurosurg Anesthesiol 1998;10:25-9. 89. Manninen PH, Balki M, Lukitto K, et al. Patient satisfaction with awake craniotomy for tumor surgery: a comparison of remifentanil and fentanyl in conjunction with propofol. Anesth Analg 2006:102:237-42. 90. Sarang A, Dinsmore J. Anaesthesia for awake craniotomy evolution of a technique that facilitates neurological testing. Br. J Anaesth 2003;90:161-5. 91. Tongier WK, Joshi GP, Landers DF, et al. Use of laryngeal mask airway during awake craniotomy for tumor resection. J Clin Anesth 2000;12:592-4. 92. Fukaya C, Katayama Y, Yoshino A, et al. Intraoperative wake-up procedure with propofol and laryngeal mask for optimal excision of brain tumour in eloquent areas. J Clin Neurosci 2001;8:253-5. 93. Balki M, Manninen PH, McGuire, et al. Venous air embolism during awake craniotomy in a supine patient. Can J Anaesth 2003;50:835-8. 94. Leslie K, Williams DL. Postoperative pain, nausea and vomiting in neurosurgical patients. Curr Opin Anaesthesiol 2005;18:461-5. 95. Roberts GC. Post-craniotomy analgesia: current practices in British neurosurgical centres-a survey of postcraniotomy analgesic practices. Eur J Anaesthsiol 2005;22:328-32.
77 Evoked Potentials in Functional Neurosurgery J. L. Shils . J. E. Arle
Introduction Intraoperative evoked potentials have a long history in functional neurosurgery and have become a critical component in the success of many of these procedures. Functional neurosurgery seeks to modify or augment neural function in such a way as to bring about a clinical benefit for the patient. Prior to the availability of highquality microelectrode recording (MER) amplifiers, evoked potentials and their associated evoked responses were the sine qua non for reliably localizing relevant functional deep brain targets. Yet, even as MER and imaging capabilities advance, evoked potentials and response characterization remain an important aspect of localization and evaluation during stereotactic and functional procedures. An evoked response is defined as a measured response of the nervous system that is directly produced by either an external or internal stimulus. The stimulus may be mechanical, auditory, visual, thermal, or electrical and may either directly activate neural elements, as during electrical stimulation, or may indirectly activate neural elements as during mechanical movement of a joint or limb. Recording an evoked response is performed by placing a transducer at a point ‘‘downstream’’ from the stimulus that is either: (1) directly effected by the stimulus, or (1) located where the transducer can record far-field effects of the stimulation. The recording can be obtained either in a ‘‘free-running’’ mode or in a ‘‘triggered’’ mode. During functional neurosurgical procedures, both free-running and triggered responses are measured. The electrical recording #
Springer-Verlag Berlin/Heidelberg 2009
of the evoked response is defined as the ‘‘evoked potential.’’ Evoked responses are used for localization during all modern surgical procedures for treating movement disorders and pain. Although the thalamus and its associated nuclei have been the primary targets for evoked potential and evoked response testing [1–28], all targets utilize evoked responses in one form or another. For example, Klostermann et al. [29] first used visual evoked potentials (VEPs) during pallidotomies to determine the location of the optic tract in relation to the base of the internal globus pallidus (GPi). Others have since also used VEPs to localize the optic tract during pallidal surgeries [30–34]. Median nerve somatosensory evoked potentials (SSEPs) have been studied for surgeries targeting the subthalamic nucleus (STN) [28,29,35–37]. And, additionally, free-running evoked responses are currently used for all surgical targets during both MER and non-MER methodologies in stereotactic targeting. Intraoperative language mapping uses a type of inverse evoked response whereby the stimulation is used directly on neural elements to disrupt briefly and help localize functionally relevant language regions in cortex during tumor removal or epilepsy surgery. Finally, motor cortex stimulation therapies use both peripherally activated evoked potentials and cortically activated evoked potentials, allowing the patient to be operated on under general anesthesia. We begin here with a succinct history of evoked response testing during stereotactic and functional procedures, followed by detailed methodologies used during evoked response and evoked potential testing in STN, GPi, thalamic, and
1256
77
Evoked potentials in functional neurosurgery
cortical procedures. Emphasis will be on technical aspects of equipment used for evoked potentials, and specific aspects of intraoperative technique.
History Since the late 1930s, when Meyers [38,39] was performing campotomies, physiology played a role in localizing critical structures1. Initially, intra-operative physiology consisted of direct electrical brain tissue stimulation evoked responses that were visually interpreted at the periphery. However, as technology (both in electronics and surgical methodology) improved, the quality and utility of intra-operative physiology increased. Dawson, in 1951, was the first to describe a reliable technique for detecting small evoked potentials in noise [42,43]. The first digital evoked potential device was based on the Computer of Average Transients in the 1960s which was specifically designed to compute recurring transient waveforms, improving the signal-to-noise ratio [44]. This device was primarily used to analyze evoked potential data and study weak NMR profiles in solid state physics. Another important early use of evoked responses includes work by Bucy [45] who in 1941 operated on two patients, one for severe abnormal movements in the left arm and leg and the other for ‘‘tremor’’-like movements of the right extremities following a traumatic fall and 14 week coma. The surgical procedure removed parts of the primary motor cortex to alleviate the abnormal movements. Bucy hoped to minimize trauma to the face and speech. Intraoperative stimulation to evoke responses in the limb muscles was performed, using 60 Hz, 2 ms, monophasic stimuli up to 20 V in the anesthetized patient. After mapping the cortex
1
It should be noted that many of Hans Berger’s early EEG studies were done on the exposed cortex from trauma [40] and pioneering work by Walker [41] in the 1940’s utilized electrocortocography for the treatment of epileptics.
with this technique, the cortical areas representing the upper limb where the abnormal movements were seen, were removed. The patient woke from surgery without tremor, but with right upper extremity paralysis. Over the next few weeks the tremor returned to the right leg and hand area while voluntary movements of the hand never returned. Sugita and Doi [46] investigated tremor augmentation and synchronization using cortical evoked responses (EEG evoked potential maximum) to localize the areas most important in tremor driving. By stimulating at the level of the anterior commissure-posterior commissure (AC-PC) line, in the ventral lateral nucleus of the thalamus, they could demonstrate a low amplitude evoked response in the sensorimotor cortex (contralateral frontoparietal region) with minimal tremor driving in the contra-lateral hand. At more dorsal locations, within this nucleus, they were able to demonstrate higher amplitude evoked responses with a higher degree of tremor driving. Of note [46] is the fact that when the stimulation was turned off, this stimulation frequency synchronized tremor activity continued for a few seconds, slowly returning to the patient’s resonant tremor frequency. Ablation at this area demonstrated the best control of tremor in patients with extrapyramidal disorders and continues to the present as one of the primary targets for DBS relief of tremor. In the 1960s Hassler et al. [47] discussed the absolute necessity of intraoperative evoked electrical stimulation testing prior to the making of any lesions in the brain. Also, of note is his discussion of the effect on the motor state (situation – e.g., resting, active, raised. . .) and consciousness of the patient with stimulation. Again, they noted that low frequency (4–8 Hz) stimulation will drive the tremor even during pallidal stimulation while high frequency stimulation (25–100 Hz) will reduce tremor [47]. More recent analysis of DBS treatment of tremor in the VIM region by Benabid and colleagues demonstrated a similar
Evoked potentials in functional neurosurgery
high frequency efficiency in tremor suppression, specifically at frequencies greater than 100 Hz [48]. In 1963 Spiegel and Wycis presented their investigations of stimulation evoked responses as a result of stimulation in the thalamic part of Forel’s H field (prerubral area (H3)) [49]. Increases in tremor amplitude were noted with frequencies of 20 and 30 Hz, while inhibition of tremor was noted at stimulation frequencies greater than 100 Hz. They describe conjugate eye movements and various vegetative effects such as changes in heart rate and pupil dilation, but do not describe the stimulation parameters or location of these responses in any detail. There have also been interesting studies utilizing the peripheral H-reflex and the effect of thalamic stimulation on this measure [50,51]. The H-reflex consists of electrical stimuli delivered to afferent Ia sensory fibers which returns through the alpha motor fibers after traversing the monosynaptic connection in the ventral gray of the spinal cord. Laitinen and Ohno [51] found that with lesioning or continuous stimulation to the VL thalamus there was neither consistent facilitation nor inhibition of the H-reflex, but with single pulse thalamic stimulation, using a 2.5 ms delay relative to the H-reflex stimulation pulse, there was a distinct facilitation of the H-reflex. However, Stern et al. [50] found a suppression with continuous thalamic stimulation. It is interesting to note that neither group investigated this as a potential tool for functional localization during the procedure. Much early work on evoked potentials and neurophysiologic studies during functional surgery can be attributed to Albe-Fessard and colleagues [2–9] who developed many techniques for the practical application of evoked responses in the operating room and localization of evoked potential generators. These techniques include detailed microelectrode recordings and evaluation of kinesthetic and voluntary joint movements,
77
sensory responses, thalamic evoked potential generators and their use in localization in the thalamus. In one study, by Yamshiro et al. [2] median nerve SSEPs were recorded in the ventral caudal (VC) and ventral intermediate (VIM) nuclei of the thalamus. The most interesting finding was that they were able to demonstrate a difference in the P15 and N20 latencies with VC recordings of about 0.5 ms later than the VIM latencies. In another study, Albe-Fessard [3] found no latency difference between the thalamus and cortex when recording from an electrode in the VP area. There was an amplitude gradient in the P15 as the electrode moved through the base of the VC nucleus (increase) below the AC-PC line, out of the nucleus. There was no direct correlation to the P15 peak and the primary tremor area, unfortunately, as this would have been very helpful as a localization tool. Fukushima et al. [52] performed several SSEP studies in multiple thalamic nuclei utilizing not only wrist median nerve stimulation, but stimulation at other body locations including the shoulder, torso, ulnar notch, and the posterior tibial nerve at the ankle. They found the largest and most peaked response to peripheral stimulation in the VC nucleus, but could not determine a somatotopic arrangement due to the large overlap between SSEP responses from the different body locations across multiple thalamic nuclei and including the zona incerta and the subthalamic nucleus [52]. Fager [53] used 2.0 V and 25 Hz stimulation to determine a safe distance from the internal capsule when placing lesions in the thalamic or subthalamic areas by noting either delayed responses to stimulation which are considered safe, or non-delayed direct responses to stimulation which would indicate stimulation of the internal capsule. For these procedures Fager was not able to demonstrate benefit with the stimulation, but was able to use the direct motor evoked responses elicited from the internal capsule as an indicator of safe lesioning distance from the capsule.
1257
1258
77
Evoked potentials in functional neurosurgery
Important Aspects of Equipment Used to Stimulate and Generate Evoked Responses Electrode Design Evoked potentials during functional neurosurgery are recorded using three primary types of electrodes: (1) microelectrodes, (2) permanent DBS electrodes, and (3) strip electrodes (> Figure 77-1). Each type of electrode has a distinct geometry and size, and so each recording property will differ. More importantly, the electrical properties and frequency response characteristics of each electrode type will require different amplifiers and conditioning electronics to (1) minimize artifact, (2) improve signal to noise ratios, and (3) optimize energy transfer. Clinical microelectrodes used today are manufactured either from tungsten, or platinum-iridium (Pt-Ir). The tips are etched to micron dimensions and then soldered to a semi-rigid stainless steel insulated wire. Though Pt-Ir electrodes have better signal transfer properties and lower impedances, in the frequency range of interest for evoked response and single unit recordings [54], tungsten is a mechanically stronger material and thus more suited to the harsh environment provided by ancillary operating room services. An evoked potential recording from the center of the ventral intermediate thalamic nuclei (trajectory shown in lower right figure inset) with stimulation of the median nerve is shown in > Figure 77-2. The recording electrode used was a Pt-Ir microelectrode (> Figure 77-1b). The reference electrode in the third trace is located at the tip of the guide canula (> Figure 77-1e) while the reference electrode in the fourth trace is at the 10–20 location Fpz. The top trace is the classical cortical SSEP recording (C3’-Fpz) and the second trace is the sub-cortical montage channel (CV-Fpz). For the recordings shown
in > Figure 77-2, as the electrode moves ventrally, the distance between the active and reference electrode increases and thus the active tip continuously moves away from the stationary reference electrode. This type of recording is called a referential recording. Differences between active electrode positions with referential recording will display as amplitude changes in each of the recordings. One potential drawback with referential recordings, however, is that large artifacts located at the referential electrode can both obscure biologic signals of interest at the active electrode and affect the ‘‘true’’ signal generator response at all positions of the active electrode recording. Utilizing a bipolar recording method keeps both the active and referential electrodes at the same distance by simultaneously moving both along the tract. The bipolar electrode recording technique allows the testing of responses by utilizing a common inter-electrode distance. Evoked potential generators can be localized more easily with this method because phase changes (the primary morphology of such a generator location) are more readily detected. Also, far-field potential effects are minimized with the bipolar montage. The above descriptions demonstrate the difference between referential (monopolar) and bipolar recording montages. In the referential case, changes in neural generator effects will manifest as changes in either amplitude or phase of responses, (> Figure 77-3a) while changes in generator location in the bipolar case will be seen as a phase change in (> Figure 77-3b). One advantage of referential recording is the ability to discriminate large spatial generators that may range over a distance within the same general range as the distance between the two recording surfaces. When changing from microelectrodes to the DBS or paddle type electrodes, (> Figure. 77‐1c, d), for example, one can reliably record both a referential and bipolar montage and thus benefit from the advantages of both methods.
Evoked potentials in functional neurosurgery
77
. Figure 77-1 Examples of different electrodes used for evoked potential recordings during functional neurosurgical procedures. (a) A glass coated Pt-Ir electrode used for micro-electrode recording (Universal electrode, Dothan, AL). (b) A glass coated Pt-Ir electrode used for micro-electrode recordings (FHC, Bowdoinham, ME). (c) Example of a tungsten, nonglass coated, semi-microelectrode (Integra, Burlington, MA). (d) The deep brain stimulating electrode with 1.5 mm long electrodes separated by 1.5 mm (Medtronic model 3387, Minneapolis, MN). (e) The micro-electrode/guide tube system. (f) Paddle electrodes used for cortical or extra-dural surface recordings (Medtronic Resume electrodes, Minneapolis, MN)
Amplifier Design Once the electrode transduces the neural signal into an electrical signal it is passed through an amplifier/conditioning system. Most importantly, the input stage of the amplifier used for
microelectrodes needs to be configured differently than the input stage of an amplifier used to record from larger electrodes. This is related to the fact that the equivalent circuit of all electrodes have both resistive and capacitive elements and, therefore, electrodes themselves will
1259
1260
77
Evoked potentials in functional neurosurgery
. Figure 77-2 Evoked potentials recorded from the scalp and a microelectrode placed in the VIM nucleus of the thalamus. The top trace is the standard cortical montage C3’-Fpz and the second trace is the sub-cortical montage channel CV-Fpz. The third trace shows the response in the thalamus at a distance of 5.0 mm from the base of the VIM nucleus. The distance from the tip of the recording electrode to the canula is 11.0 mm. The bottom trace shows the same active micro-recording electrode yet referenced to the scalp Fpz electrode. This figure is the combination of two recording windows from the Cadwell Cascade intra-operative evoked potential system
color the recorded bio-potential, primarily as a function of frequency (> Figure 77-4 [54,55] and > equation 77-1). > Figure 77-4 shows the equivalent electrical circuit of a metal microelectrode and the frequency response of this electrode type. The primary frequency, and thus impedance, effects are due to the electrode-electrolyte interface (Cma,, Rma, and Ema) and its capacitive nature, with minor contributions from the reference electrode. It should be noted that the distributed capacitance can be an issue with high frequency
noise, and in areas with a lot of wireless traffic this should be considered. Xc ¼
1 2pfC
ð1Þ
The size, shape and material of the electrode recording surface are all important factors that effect resistive and capacitive parameters and are the basis of these electrical differences. Therefore, it is critical that the design of the amplifier’s internal electronics do not adversely modify the signal characteristics; yet if the amplifier’s input stage could ‘‘correct’’ some of the distortion that would also be desirable. For example, microelectrodes act as low frequency filters [56], but as most noise is in the low frequency range, the amplifier electronics can potentially compensate for this if designed properly. Another important design feature of the amplifier relates the impedance of the electrode and its resistive/capacitive properties to the energy transfer from the electrode to the amplifier. The impedance of microelectrodes is typically between 500 kO and 1.5 MO. To maximize energy transfer, the front-end impedance of the amplifier should be at least one and preferably two orders of magnitude greater than the impedance of the electrode. To accomplish this, the amplifier front-end is designed with different configurations of filters and other components to match the characteristics of the electrode. Once the signal is conditioned by the amplifier front-end, the rest of the electronics can be of a more common design. This is particularly important now since most signals today are converted into the digital domain for analysis. Thus, current amplifier design has two primary goals [57]: (1)
(2)
Obtaining very high input impedance to minimize input loading and thus reduce loss of signal transfer from the electrode to the amplifier Assuring that the electrode/amplifier circuit is as highly resistive as possible to minimize frequency related distortions to the original bioelectrical signal
Evoked potentials in functional neurosurgery
77
. Figure 77-3 Referential (a) and bipolar (b) evoked potential recordings in the VIM nucleus of the thalamus using a Medtronic model 3387 electrode. The top trace represents the most ventral electrodes. Note the large amplitude change and relative phase shift with electrode 0 just at the base of the thalamus in the referential montage. The bipolar montage shows an amplitude change, but no phase shift since no contact is fully out of the nucleus
Common Functional Neurosurgical Procedures Using Evoked Potentials Subthalamic Nucleus Procedures Functional neurosurgical procedures that involve targeting the subthalamic region and the subthalamic nucleus (STN) utilize evoked response testing in two ways: (1) recording in the STN region while performing kinesthetic and voluntary manipulation of the patient’s limbs, and
(2) stimulating in the STN region and looking for responses. > Figure 77-5 shows an example of single unit activity changes (firing rate increase) during flexion testing of the elbow joint in a patient with idiopathic Parkinson’s disease. The neural evoked response consists of increased firing rate each time the elbow is flexed which typically can be heard as well as seen. Location of this activity is often of significant value for locating the proper placement of the implanted stimulating electrode in the sensorimotor area of this nucleus.
1261
1262
77
Evoked potentials in functional neurosurgery
. Figure 77-4 Equivalent circuit (modified from [54]) of the extracellular microelectrode recording circuit. Rma, Cma, and Ema represent the metal microelectrode-electrolyte interface parameters. Cd is the distributed capacitance that exists between the shaft of the metal microelectrode and the extracellular fluid resistance Rexc. Rmb, Emb and Cmb are the reference electrodes resistance, potential and capacitance respectively. Finally Cw represents the capacitance between the leads of the electrode. The Figure also shows the experimentally recorded frequency response of the metal micro-electrode [55]
. Figure 77-5 Evoked responses recorded with a Pt-Ir microelectrode showing the change in firing rate each time the patient’s wrist is flexed. The top trace is a recording of the accelerometer that is placed on the patient’s wrist. The shaded areas highlight the responses
Evoked potentials in functional neurosurgery
To be considered appropriate for DBS electrode placement, our surgical protocol seeks a minimum of three distinct evoked responses in at least 4.5 mm of recorded STN under ideal circumstances. In addition to the neural recorded evoked responses to peripheral manipulation, electrical stimulation of the STN region also may be tried. Electrical stimulation in the STN and nearby regions is performed to evoke responses in peripheral muscles. One method uses stimulation threshold testing to help delineate the proximity of the electrode to the corticospinal tract by observing and recording time-locked triggered motor activity in the periphery. Another method looks for changes (both improvement and decrement) in the Parkinsonian symptoms via therapeutic stimulation. Symptomatic observations include changes in tremor, rigidity, speech, eye movements, and bradykineisia. Initial patient preparation and surgical set-up have been previously described [58]. Once MER or Semi(S)-MER has started, the STN is localized using both single/multi-unit physiology and recorded evoked responses as follows:
77
When a single unit is located, evoked response testing of the contralateral joints is begun. For the upper limb the wrist, elbow and shoulder are tested. For the lower limb, the ankle and knee are tested, independently, while the hip is usually tested with the knee. Similar to the wrist example of movement causing a firing rate increase in the related single unit, > Figure 77-6 shows a similar example related to ankle movement. The exact mechanism and pathway for joint activation of the STN is not known, though there are various theories about this action [59]. The most important aspect of these responses, for intra-operative localization purposes, is the fact that these evoked responses are both repeatable and indicative of the electrode being in the sensorimotor region of the STN. Our median number of cells per STN encountered tract where STN cells are recorded is five and the average number of trajectories per patient side is 1.7. In the STN, kinesthetic cells are mostly found in the dorsal and middle regions of the nucleus with few found in the ventral area, similar to findings of Rodriguez-Oroz et al. and Starr et al. [60,61].
. Figure 77-6 Evoked responses recorded with a Pt-Ir microelectrode showing the change in firing rate each time the patient’s ankle is flexed. The top trace is a recording of the accelerometer that is placed on the patient’s ankle. The shaded areas highlight the responses
1263
1264
77
Evoked potentials in functional neurosurgery
As well, we have noted evoked responses mainly located between 9.0 and 14 mm lateral to the midline, corresponding to the sensorimotor area of the STN [60]. Consistent with data from Starr et al. [61], we have also found that the sensorimotor leg region is located more medial than the upper limb region and can be very useful in determining a subsequent movement of the electrode if necessary. During MER in the STN we have yet to find stimulation useful in evoking either beneficial or detrimental evoked responses, due primarily, we believe, to the very focal nature of micro stimulation. The exception to this observation has been testing to determine whether the electrode tip is located in the corticospinal tract. To test for corticospinal tract location, we use very low frequency stimulation (1–5 Hz) at 50–100 mA. Due to the small volume affected by micro-stimulation and the high current density at the tip, a microelectrode can elicit MEP responses in limb or face muscles. EMG electrodes placed on these regions will pick up these responses. We have not performed detailed current versus distance relationships since we commonly do not target the corticospinal tract, yet we find support for these findings from data mapping corticobulbar and corticospinal tracts in monkeys with micro stimulation [62,63]. In contrast, others [64] have found microstimulation useful in the determination of optimal DBS electrode placement. Even though we do not use micro stimulation in the STN, we find macro-stimulation evoked response testing to be critical for proper DBS placement. Stimulation parameters used for this testing are 60 ms and 185 Hz and voltages up to 4.0 V. Due to the design of the Screener Stimulation Device (Model 3682, Medtronic Inc., Minneapolis, MN) we can only test in bipolar mode. Newer devices, such as the ANS MTS trial system (Model 6510, Advanced Neuromodulation Systems, Plano, TX) allow for monopolar testing and constant current
testing. Our standard stimulation sequence includes testing each sequential electrode pair up to 4.0 V. We choose 4.0 V since this is greater than any value we have used in therapeutic stimulation with patients. Also, due to the voltage doubling circuit in the IPG we would rather avoid having the stimulator set above 3.7 V. In cases where we may obtain questionable evoked responses, specifically increased rigidity, or the potential need for tremor reduction at higher voltages, we will test a larger stimulation field using contact patterns 0,1,2,+3 and +0,1,2,3 and up to 4.0 V. Several have used median nerve SSEPs to localize the STN [28,29,35–37]. Pinter et al. [35] discussed the usefulness of SSEPs for localizing STN while Klostermann et al. concluded that SSEPs could not be helpful in localizing STN [29]. More recently Kitagawa et al. [37] did detailed analysis of SSEPs in the subthalamic region finding that electrodes placed posterior to the STN, in the ZI and medial lemniscus (ML), demonstrated a phase reversal at the ZI/ML boarder, while median nerve SSEPs, with the electrode in the STN proper, demonstrated no phase reversal at different locations in the STN [37]. Kitagawa et al. also discussed the fact that these are far-field potentials in this region but also there may be a potential contribution to the P16/N18 from the ZI in the STN region due to excitatory post-synaptic potentials in the ZI region. Dinner et al. [65] found peaks in the STN mirroring the far-field cortical P14(P16)/N18 in all referential DBS electrode recordings lacking specificity. A larger response is noted in the DBS contact 1, but no specific location of this electrode is given which in turn makes using this data for localization purposes undesirable. However, appreciating that the electrode is in the STN as opposed to another region lack of a phase reversal, or definitive amplitude peak localization would still be difficult. Since the primary goal of functional surgery in the STN is to modify extrapyramidal components of the system, it is critical to assure that the stimulation is focused
Evoked potentials in functional neurosurgery
on the sensorimotor segment of the STN and associated sub-thalamic fiber tracts. We have found that SSEPs in this region are very ambiguous and offer no real benefit in localizing the proper functional region of the STN for final DBS localization. > Figure 77-7 shows an example of both bipolar and monopolar recordings from the DBS lead (ANS Libra Deep Brain Stimulation System model, Plano, TX). There are no specific defining features of the evoked potentials to help differentiate the STN from
77
other structures nearby, yet a potential N18 phase shift is noted. Note that the monopolar recordings show no amplitude gradient, as referenced to Fz, with contacts in the STN (0 and 1) and those above the STN (2 and 3). The bipolar recording shows almost a flat line near the 20–25 ms latency demonstrating the similar amplitudes in all channels. There is some phase reversal at 18 ms. in the electrodes around the base of the thalamus, similar to what Kitigawa had discussed [37].
. Figure 77-7 Referential (a) and bipolar (b) evoked potential recordings during placement of a DBS lead in the STN. The recordings are from the DBS lead (Medtronic model 3387, Minneapolis, MN). The top traces represent the most ventral electrodes by looking at the referential recording. It is difficult to see amplitude changes indicating there is no N20 generator nearby, as is known in the bipolar trace there is a small phase shift at a latency of 16 ms. in the most dorsal electrode contacts. This could be related to the electrodes at the dorsal edge of the ZI and the ventral edge of the thalamus
1265
1266
77
Evoked potentials in functional neurosurgery
Internal Globus Pallidal Procedures Surgeries to modify the firing properties of the GPi include DBS and lesioning procedures for the treatment of Parkinson’s disease, Dystonia, Tourette’s, and multiple other movement related disorders [66–71]. With the internal capsule located medial and posterior to the GPi and the optic tract ventral to the GPi, recording of evoked responses is critical to assuring correct lesion or DBS placement. As with the STN procedures, joint movement evoked responses are helpful in localizing the functionally relevant area of the nucleus. As such, all cells recorded in the GPi are evaluated by testing the wrist, elbow, shoulder, ankle, knee, and hip motions of patients. There is lingering debate about the somatotopic nature of the GPi [72–74]. Guridi et al. observed a physiologically defined somatotopy of kinesthetic cells with the face and arm region located ventrolaterally and the leg dorsomedially [72]. Taha et al. found the leg in-between the arm in a rostral caudal axis (the arm representation is concave) [73]. Vitek et al. have found the leg to be medial and dorsal with respect to the arm while the face is more ventral [74]. For PD procedures, the evoked responses to joint manipulation are usually linked to only one joint while for dystonia procedures we have found multiple joints will at times generate an evoked response in a single cell. This data is consistent with some published findings [75], although others have found little multi-joint activity in a single-unit’s receptive field [76]. It should be noted that when testing for kinesthetic evoked responses one needs to be careful not to manipulate the patient in such a way as to shake the recording system generating spurious signals that can be incorrectly identified as inhibitory responses. Also, manipulation with vascular artifact needs to be carefully appreciated. The most common complication in early experience with posterior ventral pallidotomy
procedures was a visual field cut [77]. Mindful of this, locating the optic tract became paramount in reducing the morbidity of pallidotomy. Yokoyama et al. [30] and Tobimatsu et al. [31] used visual evoked potentials to localize the optic tract. Yokoyama et al. used LED goggles to present alternating flash stimuli [30] at various depths until the optic tract was located. The optic tract was localized when the evoked response amplitude increased in a manner similar to what was found in cat studies this group had previously performed [32]. We do use visual evoked responses, recorded at the end of the micro-recording tract. At the point the microelectrode passes the base of the GPi we turn the lights out in the operating room and in 0.5 mm increments stop the electrode and shine a flashlight on and off in a patients eyes. We then listen predominantly to the background noise of the electrode. When we are within 0.5 mm of the optic tract we hear high frequency changes in the neural noise. The closer the electrode tip gets to the optic tract the louder the change. For pallidotomies we would ideally like to be at least 1 mm from the optic tract when placing a lesion. For DBS procedures we try to keep the electrode 2–3 mm from the optic tract. Prior to removing the micro-electrode, stimulation is performed to activate the optic tract. The stimulation parameters are 300 Hz and 100 ms. If the electrode is in the optic tract, optic responses can be noted down to 10 mA. With the micro-electrode we find that we can be only 0.5 mm away from the optic tract even at a stimulation amplitude of 100 mA. Thus, using micro-electrode stimulation is not an optimal technique from a safety standpoint. During the recent pallidotomy era (roughly 1990–2000), after MER and prior to placing the lesion, we would stimulate with the lesioning probe using a 300Hz, 100 ms monophasic and/ or biphasic pulse as the stimulation current is slowly raised. If a visual response is evoked at a stimulation under 1 mA or 2.0 V, the probe is moved dorsally to avoid damage to the optic
Evoked potentials in functional neurosurgery
tract. During DBS procedures, we stimulate through the lowest two contacts of the DBS lead using the Medtronic screener box. Stimulation is raised to 4 V at 210 ms and 130 Hz. If there are no visual evoked responses, defined as phosphenes or visual hallucinations, then the electrode is considered safe in position relative to the optic tract. It is important to note the difference in stimulation voltages between the two electrode types and stimulation devices. A second evoked response is used during pallidal surgeries to test for proximity to the internal capsule. Anatomically, the internal capsule is located posterior and medial to the GPi (> Figure 77-8). Capsular responses are more critical during DBS surgery due to the larger spread of electrical effects than with lesioning [58]. During pallidotomies it was important to localize the distance from the tip of the lesioning probe to that of the internal capsule for two reasons. First, to assure that the lesion would not impinge on the internal capsule and cause a
. Figure 77-8 Axial MRI image demonstrating the relationship of the Internal Capsule to the globus pallidus. The internal capsule is shaded in
77
permanent deficit. Second, contractions in the nasal labial fold are important to demonstrate the appropriate location for the lesioning tip [78]. Utilizing the CF-3 Lesion Generator (Integra, Inc., Burlington, MA) with a 3 mm long by 1.5 mm lesioning probe, the voltage is quickly raised to a maximum stimulation of 6 V while observing the nasal labial fold and contralateral thumb. If stimulation does not induce any changes up to 2.5 V, it is considered safe for the placement of a lesion. Moreover, if stimulation at 5–6 V induces a contraction of the contralateral nasal-labial fold and/or contraction of the contralateral thumb, then its position is considered optimal for lesioning. Such testing is performed prior to all lesions. Note that many patients will have a mild weakness in their nasal-labial fold for approximately 2–4 weeks following surgery. This would be an indicator of a properly placed lesion. This is due to a transient edematous zone around the lesion that will resolve over time. With DBS procedures, stimulation takes place once the DBS electrode is inserted, after MER is completed. At this point it is critical to make sure that therapeutic stimulation will not cause adverse events that will inhibit use of the therapy. Similar to the responses during the recording phase of the surgery, the primary evoked responses are motor and visual in nature. In dystonia patients there may also be some sensory related phenomena that are not as readily elicited in PD patients. Evoked response testing is done in similar fashion to STN testing utilizing a sequential bipolar pattern starting with contacts set at 0, 1+. The pulse width for PD patients is 60 ms, while for dystonia patients it is set at 330 ms. This difference is related to settings used for therapy. It is highly unlikely that PD therapy will require a 330 ms pulse, while it is possible that this may be needed for dystonia therapy. The testing frequency for both PD and dystonia is 130 Hz. First, optic tract evoked responses are tested by turning the OR lights off and asking
1267
1268
77
Evoked potentials in functional neurosurgery
the patient to describe any visual phenomena they may have while raising the voltage. These responses may include ‘‘flashing lights,’’ ‘‘hazy lights,’’ or even visual hallucinations. Stimulation amplitudes are slowly raised to 5.0 V for dystonia patients and 4.0 V for PD patients. If optic responses are noted during cathodal stimulation of contact 0, but not for contact 1, then the electrode will be raised 1.0 mm before securing in place. Following these details in our procedures we have never had optic tract responses when stimulating above contact 0 in the operating room, nor any optic tract adverse events during therapeutic stimulation.
Thalamic Procedures Of all the deep brain structures involved in functional localization, the thalamus is most readily designed for both evoked response testing and evoked potential monitoring. This is because the thalamus contains multiple evoked potential generators and because microelectrode, semimicroelectrode and macro-electrodes can all record evoked responses (event related responses). Additionally, evoked responses are particularly helpful because the single-unit signatures within various thalamic nuclei are difficult to differentiate purely on firing pattern alone. The primary thalamic targets for both movement disorders and pain are in the ventral thalamus. Each of these areas has been neurophysiologically studied in detail [79]. Because the ventral oralis anterior (VOA) and the ventral oralis posterior (VOP) have similar microelectrode recording characteristics [79] it is critical to use evoked potential and evoked response testing when targeting this area. The ventral caudal nucleus (VC), which contains sensory receiving cells and lies just posterior to the ventral intermediate (VIM) nucleus, also has similar recording characteristics to the other three regions. Evoked response testing in the thalamus is separated into both recorded responses and stimulated responses. At our center we do all microelectrode recordings
moving dorsal to ventral while all stimulation testing is done in reverse during the removal of the electrode. There are historically four major naming conventions for the motor thalamic nuclei [80]: Jones [81], Hassler [82,83], Ilinsky and Kultal-Ilinsky [84], and Olszewski [85]. In this discussion we choose to use the nomenclature of Hassler [82,83]. For a helpful cross-reference between each of the nomenclatures see table 1 in reference [80] of this chapter. Beginning with the most posterior of the ventral nuclei, the VC nucleus is the primary receiving area for the spinothalamic and medial lemniscus tracts, both sensory. Microelectrode recorded evoked responses occur in response to light touch. > Figure 77-9 shows two examples of evoked responses in the VC thalamus to light brushing of the thumb. Receptive fields of the cells in this region vary in size with the region of the body. Areas of large receptive fields (4–5 cm) that will evoke a response in a single cell include the shoulder, torso and abdomen whereas areas of very focal (1 cm or less) receptive fields include the thumb, fingertips, mouth and jaw. Cellular responses to evoked skin stimuli in this region are arranged somatotopicaly. Responses follow a medial to lateral alignment (> Figure 77-10 [86]), with mouth responses most medial and leg responses most lateral. > Figure 77-11 shows the map of the responses from two patients who had very focal temporal head pain in which a DBS lead was being placed in VC for pain relief. The medial lateral angles for these two cases were 83 and 85 degrees. One can see the sagittal overlap of some of the more diffuse areas such as the forehead and temporal head area with those of the focal areas of the thumb and lip when stimulating with the DBS lead (1.1 mm diameter) as compared to the micro-electrode (10 mm). Despite identifying the precise area, it was impossible to alleviate any focal temporal head pain without also causing parasthesias in the hand and mouth areas of these patients. One patient was able to tolerate the extra sensations in his
Evoked potentials in functional neurosurgery
77
. Figure 77-9 Two examples of microelectrode recordings in the VC nucleus of the thalamus showing evoked responses to light brushing of the thumb. In the bottom trace expanded time tracings are shown to demonstrate the increased firing rate
. Figure 77-10 Lateral representation of the ventral posterior nuclei of the thalamus. Taken from [86]
hand and mouth with a minor reduction in head pain, while the other patient could not tolerate the extra sensations and the therapy had to be halted. Work by Tasker and Kiss have found that intra-oral representation is between 10–11 mm from midline, face tactile cells are located between 11–14 mm lateral from midline, manual digits 13–16 mm from midline, and foot representation 15–17 mm
lateral to midline [79]. It is important to note that this overall representation is only found in the ventral one half to two-third of the nucleus [79]. Giorgi et al. found a similar arrangement with the addition of fingers being somewhat more caudal and posterior to the face and hand and the hip being just lateral to the shoulder with the thigh; foot and leg having minimal representation [87]. Recording of evoked responses is important during surgeries for the treatment of pain and for the treatment of tremor. Even though the final electrode for tremor is placed in the VIM nucleus, body representations in both VC and VIM are similar. Therefore, knowing the location in the VC can be an important guide for a subsequent electrode trajectory. Moreover, finding the VIM/ VC boarder is critical for tremor therapy since the best location for tremor treatment has been found just inside VIM at this border [88,89]. The VIM nucleus, along with the VOP nucleus, is considered the primary receiving nucleus from the dentate and anterior interposed nuclei of the cerebellum [90]. Similar to organization within the VC nucleus, recorded activity is primarily found in the ventral two-third of the
1269
1270
77
Evoked potentials in functional neurosurgery
. Figure 77-11 Stimulation response locations from two patients who were being operated on for the treatment of focal temporal pain. Patient 1 (black tracts) was injured in a thirty foot fall while at work. Patient 2 (grey tracts) started to have pain after a previous surgery. In both cases post-operative stimulation caused sensations in the hand region at therapeutic levels for alleviation of pain in the temporal region
nucleus. A similar medial-to-lateral somatotopic arrangement is found in this nucleus with face being medial and leg lateral. Kinesthetic responses are the main evoked responses recorded in this region. > Figure 77-12 is an example of such an evoked response. Mechanical stimuli such as the squeezing of the muscle bellies, or the passive movement of joints will evoke response during MER and S-MER in this region. Observations by the authors find the receptive fields in this region to be larger than those of the VC, which is consistent with findings of others [79]. This observation is likely due to the fact that this area receives information from joints rather than from skin surface. In contrast to the purely excitatory nature of the VC evoked responses, VIM evoked responses can be either excitatory (demonstrating an increase in amplitude or firing frequency of a unit) or inhibitory. Inhibitory responses (> Figure 77-13) are found less frequently than excitatory ones, occurring in about 10–15% of the recordings. Lenz et al. has described both fast and slow adapting cells in the VIM to passive joint movements [23]. Thus,
when testing joints, one should test for a minimum of five manipulations. Table 4 in Lenz et al. [23] describes the distribution of evoked responses within different thalamic nuclei. VC stimulation evokes responses at low amplitude levels as compared to evoked responses from stimulation in VIM. Microelectrode stimulation of 5–10 mA can evoke sensations in the face and hand areas. Slightly higher stimulation (20–40 mA) is typically required to evoke responses in the proximal limbs or torso. AlbeFessard et al. [2,3] originally found responses in the VC during micro-stimulation studies from as low as 4 mA whereas at a distance of 2 mm, 100 mA would not generate an evoked response. Ohara et al. [14] performed detailed micro stimulation studies in the VC nucleus and found vibratory sensations ventral to the movement receptive fields in the upper extremity and face. Contrary to this study, however, we have found sensory evoked responses, defined as a tingling, numbness, vibration, or electrical shock, in all areas of the VC thalamus that are adjacent to the VIM, and have used this critical information
Evoked potentials in functional neurosurgery
77
. Figure 77-12 A wrist kinesthetic excitatory evoked response. The top trace is the raster plot for the full test, the second trace is the raw data from the microelectrode recording while the third trace is the accelerometer on the hand. The middle raster plots are peri-event histograms representing the firing rates from 0.5 s before the motion of the wrist to 0.5 s after the wrist motion. At time 0 one can see the increase in firing frequency each time the wrist is flexed. The spike rate histogram at the bottom also shows this effect
in planning a next trajectory of the microelectrode or final electrode. If we do not find tremor cells in a MER trajectory, then VC responses are the next most important response. Using the fact that the somatotopy in the VC and VIM region is the same in a sagittal plane allows this mapping to be done and is a technique that has been used by others [91]. Macro electrode stimulations of 0.1–0.5 mV will, many times, evoke stimulation in the VC when using the DBS lead. When this occurs in an area where there are sensory related single units and that is also the area with the distal limb tremor, then the limb tremor will almost always disappear for a short period of time but return. The parasthesias in that area will most likely stay.
On the other hand if the macro electrode is placed in the center of a tremorgenic zone [89] (in the VIM area) a transient parasthesia may occur at low stimulation levels, but will disappear within a short time, while the distal limb tremor will continue to be reduced or eliminated. VIM stimulation evoked responses are more complex, however, than in the VC. During surgeries for tremor, the primary goal of surgery is to stop tremor, but other motor effects are not the primary intention. In the authors’ experience, tremor reduction or cessation is seen at levels that are much less than those seen for motor responses. Micro stimulation in the VIM at levels reaching 40–100 mA will cause parasthesias in the contralateral body and the majority of these response
1271
1272
77
Evoked potentials in functional neurosurgery
. Figure 77-13 A wrist kinesthetic inhibitory evoked response. The top trace is the raster plot for the single unit recordings, the second trace is the raw data for the microelectrode recording while the third trace is the accelerometer on the hand. The middle raster plots are peri-event histograms representing the firing rates from 1 s before the motion of the wrist to 1 s after the wrist motion. At time 0 one can see the decrease in firing frequency each time the wrist is flexed. The spike rate histogram at the bottom also shows this effect
fields are larger than for VC responses, consistent with past studies[79]. These responses are most easily seen in the face and distal upper extremity. Investigators have also noted stimulation evoked movement sensations, even though no actual movement has occurred [92]. Macro-stimulation, however, evokes transient sensory responses and can also evoke sensations of dizziness or nausea [79]. In our experience these responses are more likely to occur at or close to the time of surgery. After surgery, dizziness and nausea tend to diminish over time and finally disappear. Most peripheral nerve evoked potential studies, and clinical testing related to functional neurosurgery, has occurred during surgeries in the thalamus. > Figure 77-14 shows the normal
path of the median nerve evoked potential with the presumed evoked response generators and their wave pattern. An important question remains although the VC thalamus is thought to be one of the SSEP generator sites, as to whether this knowledge can be used to localize a target during functional surgery. A basis for utilizing the SSEP as a localization tool is the ability to record the near-field components of the resulting waves. The VC nucleus would be the only thalamic nucleus where this could be performed because it is a primary component of the SSEP [93]. To use far field recordings as a consistent localizing tool, either multiple recordings would be needed to triangulate the source, or multiple studies differentiating location would need to be done.
Evoked potentials in functional neurosurgery
. Figure 77-14 Shows the median nerve SSEP pathway from the spinal cord to the cortex and the representative median nerve SSEP traces. Note the N20 peak comes from the cortex, while the N18/P18 peak comes from the thalamus (Drawing used with permission from MJ Shils)
Albe-Fessard and colleagues performed multiple studies of median nerve SSEPs in the VC and VIM nuclei of the thalamus [2] and could demonstrate a difference in the P15 and N20 latencies with the VC recordings set about 0.5 ms later than the VIM latencies. > Figure 77-15 shows the results of microelectrode recorded evoked potentials in a trajectory that passes through the VIM and VC of an essential tremor patient. The microelectrode used was a Pt-Ir glass coated tip (impedance of 800 kO) with a diameter of 10 mm (FHC model MTBPBN(EN2), Bowdoin, ME). Recordings
77
were started at 7.4 mm above the base of the thalamus in the VIM. Stimulation was at the median nerve of the contralateral wrist utilizing 4.7 Hz stimulation, 300 s pulse width, 10 mA and 50 averages. The bandwidth of the signal was from 30 to 750 Hz. The signal was referenced to the canula that is 16.1 mm from the base of the thalamus (15 mm from the original stereotactic target). The amplitude of the N20 starts off low in the VIM and reaches a peak just at the entry to VC (2.2 mm). At the base of the thalamus the amplitude once again decreases. One primary difference in this recording is the large N20 response and small P18 response, possibly due to the reference including the whole length of the canula and not just its tip. Also, this recording comes primarily from VIM and the most anterior portion of the VC nucleus, not the more posterior part. > Figure 77-16 shows the same patient with the evoked potential being recorded from the Medtronic model 3387 DBS lead in a slightly different trajectory than the one used in > Figure 77-15. Stimulation parameters were the same as above. Note the large P18 in the contact 0 derivation, receiving more signal from the posterior part of the VC nucleus while in the other leads the P18 is indistinguishable from background though a lower amplitude N20 is seen. Post-operative testing on this patient demonstrated low intensity parasthesias when using contact 0 in either the monopolar (0.3 V) or bipolar (0.7 V) configuration with a 185 Hz/60 s pulse width signal. Tremor suppression was noted at contact 2 cathode and case positive with a stimulation voltage of 2.0 V. Stimulation in contact 1 was able to reduce the tremor, but caused non-transient parasthesias. This example portrays the utility of evoked potentials in thalamic functional surgery. The primary problem with using SSEPs in the awake patient is both speed and patient comfort. It can take 30–60 s to produce a reliable SSEP and many patients find the stimulation sensations to be disconcerting, whereas simple brush stroke
1273
1274
77
Evoked potentials in functional neurosurgery
. Figure 77-15 Median nerve evoked potentials recorded in the VIM nucleus of the thalamus at different depths during microelectrode recordings. The active microelectrode tip is referenced to Fpz. No phase shift is noted in this recording due to the Fpz reference
and limb movement evoked responses can be done in about 5 s.
Cortical Localization and Stimulation Procedures Motor cortical surgeries for the treatment of pain and movement disorders likely began in the 1940s with Bucy’s direct extirpation of the primary motor cortex [94,95]. Bucy stimulated the cortex to identify appropriate regions for removal. Currently, we use two evoked responses: (1) somatosensory evoked phase-reversal potential mapping (> Figure 77-17a), and (2) direct epidural motor stimulation evoked response mapping (> Figure 77-17b). Intra-operative
phase reversal mapping makes use of the orientation of the pyramidal cells of the cortex and the location of the sensory cortex in relation to that of the motor cortex. Knowing that the cortical generator (N20/P25) of the median nerve somatosensory evoked potential is in the somatosensory cortex [93] allows us then to locate the motor strip just anterior across the central sulcus (> Figure 77-17a). One of the N20 dipoles is located on the posterior wall of the central sulcus [96]. Thus, recording on one side of the sulcus gives one polarity while recording from the other side shows the opposite polarity. The technique using both SSEP and motor mapping has been described in the literature [97]. Briefly, the electrode is placed parallel with and essentially overlying the M1 strip of
Evoked potentials in functional neurosurgery
77
. Figure 77-16 Referential (a) and bipolar (b) median nerve evoked potentials recorded from the DBs electrode (Medtronic model 3387, Minneapolis, MN). The top trace in both montages are recorded from the most distal electrodes. There is a large amplitude change from contact 0 to contact 1, in the referential recording, indicating that contact 0 may be either out of the thalamus or just on the border of the thalamus. For the bipolar recording case there is a minor phase shift moving from contacts 0–1 to contacts 1–2. This indicates that contact 0 is probably just on the board of the thalamic nucleus
cortex and ultimately sutured directly to the dura to avoid post-operative electrode movement (> Figure 77-18). Somatosensory testing consists of placing the 4-contact Resume lead (Medtronic Inc., Minneapolis, MN) on the dura in several directions, mostly perpendicular to the suspected pre-central gyrus. Median and Ulnar nerve somatosensory evoked potentials (SSEPs) are then observed (Cascade EP machine, Cadwell Laboratories, Inc., Kennewick, WA) using a 20 mA, 100 ms monopolar square pulse at a rate of 4.32 Hz. SSEPs were recorded from the Resume lead in both a bipolar (contact 0–1, 1–2, 2–3) and monopolar (all referenced to the 10–20 location of Fz) recording montage. The central sulcus is the point where the N20 response phase reverses (> Figure 77-17a). This is performed across multiple locations to follow the course of the central sulcus. Motor mapping consists of placing an anodal 5 mm stimulation ball probe (Model E1564, Valleylab, Gosport, UK) over the M1 area referenced to a cathode placed at Fz. Stimulation
consists of trains of 5 stimuli each, at an interstimulus duration of 4 ms, a 500 ms pulse width. Stimulation amplitudes were slowly increased at each location starting at 5 mA and increasing to a maximum of 25 mA. Stimulation then stops when the first EMG responses are noted. EMG needles are placed in bipolar fashion (separated by 2 cm in all muscles except the orbicularis oris and orbicularis oculi) in the orbicularis oculi, orbicularis oris, trapezius, deltoid, biceps, triceps, flexor carpi ulnaris, abductor policis brevis, first dorsal interossious, quadraceps, anterior tibialis, and abductor hallicus muscles. Stimulation is performed with a Grass S-88 and two SIU-7 constant current stimulus isolators (Astromed-Grass, West Warwick, RI) and responses are recorded on a Cadwell Cascade (Cadwell Laboratories, Inc., Kwennewick, WA). > Figure 77-18 shows an example of the findings in the extensor and APB muscles with this technique. In this way two types of intraoperative physiology corroborate to locate the precentral gyrus.
1275
1276
77
Evoked potentials in functional neurosurgery
. Figure 77-17 Example of phase reversal mapping during surgery for motor cortex stimulation. The top left traces show both the referential and bipolar phase reversals. The bottom left trace shows the electrode on the dural surface. The right image shows a representation of the location of each electrode over the cortical surface. Looking at the electrode position it can be seen that the P20 responses come from electrodes over the motor cortex, while the N20 responses come from the electrodes over the sensory cortex
The normal homunculus of the motor strip has face and upper limb on the lateral surface with leg on the superior edge and medial surface of the cortex. During our standard mapping cases with focal stimulation we do not generally activate the lower limb muscles. In one patient, a car accident had caused a brachial plexus avulsion and left her with no motor or sensory function below the elbow. The rationale for the surgery was to treat intense pain in the deltoid region. The accident occurred five years prior to the surgery and since then the motor cortex representation had shifted in relation to the new peripheral inputs. In this case, thigh and torso were adjacent to the shoulder and then the face following. The thigh and upper leg was now more lateral than in the normal motor cortex.
In pain patients we have found it beneficial to place the stimulating lead over the cortical motor region that is involved in the pain area. In our experience, we do not derive much benefit for lower limb pain and these findings are similar to some prior studies [98,99], but different from others findings [100,101]. In a study by Tsubokawa et al. [98] they noticed that to benefit patients with lower extremity pain they needed to use stimulation intensities that actually generated motor responses. In the patient described above, although we sought out the deltoid region specifically, we found face, then deltoid, abdominal muscles, and then leg, moving superiorly from a starting position lateral on the cortex. Tumor and epilepsy surgeries utilize both sensory and motor evoked potentials to map
Evoked potentials in functional neurosurgery
77
. Figure 77-18 Example of responses obtained during motor mapping performed during surgery for the placement of a motor cortex stimulator. Each area is labeled as the location where a response is obtained
out eloquent areas of cortex prior to and during resections. In the case of tumor and epilepsy resections, the dura is open and the cortex is visible, which may make one think that locating the central sulcus should be an easy task visually, yet due to the structural abnormality normal cortical arrangement is often distorted. In order to determine true functional cortex, mapping remains a necessary tool, even with new complex imaging algorithms. Similar to motor cortical stimulation cases, median and ulnar nerve SSEPs are used to localize the central sulcus. Multiple positions are tested with either 4 or 6 contact strips of 64 channel grids. In a study with 230 patients, Romsto¨ck et al. [102] found that not only is the N20-P20 phase reversal important, but in large tumors located centrally and postcentrally, later waves (at 35 and 30 ms) are important as well. They found a reduction in amplitude in 18% of patients or complete loss of these waves, yet in about 55% of this group evaluation of potentials between 25 and 30 ms continued to make SSEP mapping a viable tool
with only 8% of patients not being able to be mapped intra-operatively. As with SSEP localization techniques, motor evoked potential mapping is needed to localize functional tissue in anatomy that may not conform to the textbooks. There are two primary methods to stimulate the cortex for mapping. The first is the one described above in the cortical stimulation section utilizes a very short train (5–9 pulses or 20–40 ms) of high frequency (250 Hz) energy [103]. The second method utilizes a long train (2–5 s) of low frequency (50–60 Hz) energy which was originally described by Penfield and Boldrey [104]. Reports of seizures, however, with the Penfield technique approximately 24% [104,105] while with the short train technique vary from 1.2 [106] to 1.6% [107] in patients with a history of epilepsy. As Szele´nyi et al. point out in their review of the literature there, is no greater incidence in patients with a history of seizures as compared to those without a history of seizures in either technique [107]. In one paper by Roux et al. [108] they found good central correlation between fMRI and
1277
1278
77
Evoked potentials in functional neurosurgery
intraoperative motor mapping. One interesting conclusion from the paper was that fMRI localized a larger area of cortex than direct cortical stimulation mapping, thus potentially reducing the amount of tumor the surgeon could remove near the margins.
Conclusion Despite advances in imaging, functional imaging, and multi-channel micro-electrode recording, evoked response testing is still critically important in localizing functional neurosurgical targets. With the advent of neuromodulation therapies and the fact that all of these targets currently are for functional clinical treatment, evoked response testing not only helps in localization, but also in understanding how the therapy may work in a particular patient. Though pure evoked potential studies may have now taken a back-seat to micro-electrode recording, the use of evoked response testing during these procedures is still a vital component of their success. Evoked response testing consists of both peripheral manipulation of the sensory and motor systems to generate a response in the brain and the electrical, or even mechanical manipulation of the neural tissue to evoke responses in the periphery. Even as functional neurosurgery breaks new ground in neuromodulation therapies for treatment of depression, addiction, anorexia, obesity, obsessive compulsive disorder, etc, it will be the functional responses (i.e., evoked responses), be it beneficial or adverse, that may still remain an important measure of how well these therapy works [109,110].
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
References 14. 1. Lin YC. Lenz FA. Distribution and response evoked by micro stimulation of thalamus nuclei in patients with dystonia and tremor. Chin Med J 1994;107(4):265-70. 2. Yamashiro K, Tasker RR, Iwayama K, Mori K, AlbeFessard D, Dostrovsky JO, Chodakiewitz JW. Evoked
15.
potentials from the human thalamus: correlation with micro stimulation and single unit recording. Stereotact Funct Neurosurg 1989;52(2–4):127-35. Albe-Fessard D, Tasker R, Yamashiro K, Chodakiewitz J, Dostrovsky J. Comparison in man of short latency averaged evoked potentials recorded in thalamic and scalp hand zones of representation. Electroencephalogr Clin Neurophysiol 1986;65(6):405-15. Albe-Fessard D. Electrophysiological methods for the identification of thalamic nuclei. Zeitschrift fur Neurologie 1973;205(1):15-28. Albe-Fessard D, Arfel G, Guiot G, Derome P, Guilbaud G. Thalamic unit activity in man. Electroencephalogr Clin Neurophysiol 1967;Suppl 25:132. Albe-Fessard D, Arfel G, Guiot G, Derome P, Hertzog E, Vourc’h G, Brown H, Aleonard P, De la Herran J, Trigo JC. Electrophysiological studies of some deep cerebral structures in man. J Neurol Sci 1966;3(1):37-51. Albe-Fessard D, Arfel G, Guiot G, Derome P, Dela H, Korn H, Hertozog E, Vourch G, Aleonard P. Characteristic electric activities of some cerebral structures in man. Annales de Chirurgie 1963;17:1185-214. Santibanez G, Trouche E, Albe-Fessard D. Study of the evolution, in the course of time, of the amplitude of cortical and thalamic activities evoked by repetitive somatic stimuli. Journal de Physiologie 1963;55:335-6. Krauthamer G, Albe-Fessard D. Inhibition of induced cortical and sub cortical activities by stimulation of the capsule. Comptes Rendus des Seances de la Societe de Biologie et de Ses Filiales 1961;155:1443-50. Lenz FA, Gracely RH, Baker FH, Richardson RT, Dougherty PM. Reorganization of sensory modalities evoked by micro stimulation in region of the thalamic principal sensory nucleus in patients with pain due to nervous system injury. J Comp Neurol 1998;399 (1):125-38. Lin YC, Lenz FA. Effective response evoked by micro stimulation of thalamus nuclei in patients with tremor. Chin Med J 1993;106(5):372-4. Lenz FA, Dostrovsky JO, Kwan HC, Tasker RR, Yamashiro K, Murphy JT. Methods for micro stimulation and recording of single neurons and evoked potentials in the human central nervous system. J Neurosurg 1988;68 (4):630-4. Lee JI, Ohara S, Dougherty PM, Lenz FA. Pain and temperature encoding in the human thalamic somatic sensory nucleus (Ventral caudal): inhibition-related bursting evoked by somatic stimuli. J Neurophysiol 2005;94(3):1676-87. Ohara S, Weiss N, Lenz FA. Micro stimulation in the region of the human thalamic principal somatic sensory nucleus evokes sensations like those of mechanical stimulation and movement. J Neurophysiol 2004;91(2):736-45. Ohara S, Lenz FA. Medial lateral extent of thermal and pain sensations evoked by micro stimulation in somatic
Evoked potentials in functional neurosurgery
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26. 27. 28.
29.
sensory nuclei of human thalamus. J Neurophysiol 2003;90(4):2367-77. Garonzik IM, Hua SE, Ohara S, Lenz FA. Intraoperative microelectrode and semi-microelectrode recording during the physiological localization of the thalamic nucleus ventral intermediate. Mov Disord 2002;17 Suppl 3: S135-44. Hua SE, Garonzik IM, Lee JI, Lenz FA. Microelectrode studies of normal organization and plasticity of human somatosensory thalamus. J Clin Neurophysiol 2000;17 (6):559-74. Lee J, Dougherty PM, Antezana D, Lenz FA. Responses of neurons in the region of human thalamic principal somatic sensory nucleus to mechanical and thermal stimuli graded into the painful range. J Comp Neurol 1999;410 (4):541-55. Zirh TA, Reich SG, Perry V, Lenz FA. Thalamic single neuron and electromyographic activities in patients with dystonia. Adv Neurol 1998;78:27-32. Hua SE, Lenz FA, Zirh TA, Reich SG, Dougherty PM. Thalamic neuronal activity correlated with essential tremor. J Neurol Neurosurg Psychiatry 1998;64(2):273-6. Mandir AS, Rowland LH, Dougherty PM, Lenz FA. Microelectrode recording and stimulation techniques during stereotactic procedures in the thalamus and pallidum. Adv Neurol 1997;74:159-65. Lenz FA, Gracely RH, Romanoski AJ, Hope EJ, Rowland LH, Dougherty PM. Stimulation in the human somatosensory thalamus can reproduce both the affective and sensory dimensions of previously experienced pain. Nat Med 1995;1(9):910-3. Lenz FA, Kwan HC, Martin RL, Tasker RR, Dostrovsky JO, Lenz YE. Single unit analysis of the human ventral thalamic nuclear group. Tremor-related activity in functionally identified cells. Brain 1994;117 (Pt 3):531-43. Lenz FA, Seike M, Richardson RT, Lin YC, Baker FH, Khoja I, Jaeger CJ, Gracely RH. Thermal and pain sensations evoked by micro stimulation in the area of human ventrocaudal nucleus. J Neurophysiol 1993;70 (1):200-12. Tasker RR, Gorecki J, Lenz FA, Hirayama T, Dostrovsky JO. Thalamic microelectrode recording and micro stimulation in central and deafferentation pain. Appl Neurophysiol 1987;50(1–6):414-7. Bakay RAE, Vitek JL, DeLong MR. Thalamotomy for tremor. Neurosurg Atlas 1992;2(4):290-312. Lozano A. Thalamic deep brain stimulation for the control of tremor. Neurosurg Atlas 1998;7:125-135. Hanajima R, Dostrovsky JO, Lozano AM, Hutchison WD, Davis KD, Chen R, Ashby P. Somatosensory evoked potentials (SEPs) recorded from deep brain stimulation (DBS) electrodes in the thalamus and subthalamic nucleus (STN). Clin Neurophysiol 2004;115(2):424-34. Klostermann F, Vesper J, Curio G. Identification of target areas for deep brain stimulation in human basal ganglia
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40. 41.
42. 43.
44.
45.
77
substructures based on median nerve sensory evoked potential criteria. J Neurol Neurosurg Psychiatry 2003;74 (8):1031-5. Yokoyama T, Sugiyama K, Nishizawa S, Yokota N, Ohta S, Uemura K. Visual evoked potentials during posteroventral pallidotomy for Parkinson’s disease. Neurosurgery 1999;44(4):815-24. Tobimatsu S, Shima F, Ishido K, Kato M. Visual evoked potentials in the vicinity of the optic tract during stereotactic pallidotomy. Electroencephalogr Clin Neurophysiol 1997;104(3):274-9. Sugiyama K, Yokoyama T, Yamamoto T, Uemura K. Localization of the optic tract by using sub cortical visual evoked potentials (VEPs) in cats. Clin Neurophysiol 1999;110(1):92-6. Lozano A, Hutchison W, Kiss Z, Tasker R, Davis K, Dostrovsky J. Methods for microelectrode-guided posteroventral pallidotomy. J Neurosurg 1996;84(2):194-202. Yokoyama T, Sugiyama K, Nishizawa S, Yokota N, Ohta S, Yamamoto S, Uemura K. Visual evoked oscillatory responses of the human optic tract. J Clin Neurophysiol 1999;16(4):391-6. Pinter MM. Characteristics of SEP from subthalamic nucleus in human. Electroencephalogr Clin Neurophysiol Suppl 1999;50:364-72. Pesenti A, Priori A, Locatelli M, Egidi M, Rampini P, Tamma F, Caputo E, Chiesa V, Barbieri S. Subthalamic somatosensory evoked potentials in Parkinson’s disease. Mov Disord 2003;18(11):1341-5. Kitagawa M, Murata J-I, Uesugi H, Hanajima R, Ugawa Y, Saito H. Characteristics and distribution of somatosensory evoked potentials in the subthalamic region. J Neurosurg 2007;107:548-54. Meyers R. Surgical interruption of the pallidofugal fibers: its effect on the syndrome of paralysis agitans and technical considerations in its application. NY State J Med 1942;42:317-25. Hayne RA, Meyers R, Knott J. Characteristics of electrical activity of human corpus striatum and neighboring structures. J Neurophysiol 1949;12:185-95. Berger H. Uber das Elektrenzephalogramm des Menschen. Arch, Psychiat. Nevenkrankh 1929;87:527-70. Walker AE, Marshall C, Beresford EM. Electrocortocographic characteristics of the human cerebrum in posttraumatic epilepsy. Assoc Res Nerv Ment Dis 1947;26:502-15. Dawson GD. A summation technique for detecting small signals in large irregular background. J Physiol 1951:115:2-3 Dawson GD. A summation technique for the detection of small evoked potentials. Electroencephalogr Clin Neurophysiol 1954;6(1):65-84. Delorme A, Makeig S. EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. J Neurosci Meth 2004:9–21. Bucy PC. The surgical treatment of extrapyramidal diseases. J Neurol Neurosurg Psychiatry 1951;14:108-17.
1279
1280
77
Evoked potentials in functional neurosurgery
46. Sugita K, Doi T. The effects of electrical stimulation on the motor and sensori systems during stereotaxix operations. Confin Neurol 1967;29:224-9. 47. Hassler R, Riechert T, Mundinger F, Umbach W, Ganglberger JA. Physiological observations in stereotaxic operations in extrapyramidal motor disturbances. Brain 1960;83:337-54. 48. Benabid AL, Pollak P, Gervason C, Hoffmann D, Gao DM, Hommel M, Perret JE, de Rougemont J. Long-term suppression of tremor by chronic stimulation of the ventral intermediate thalamic nucleus. Lancet 1991; 337:403-6. 49. Spiegel EA, Wycis HT, Szekely EG, Soloff L, Adams J, Gildenberg P, Zanes C. Stimulation of Forel’s field during stereotaxic operations in the human brain. Electroencephalogr Clin Neurophysiol 1963;16:537-48. 50. Stern J, Mendell J, Clark K. H reflex suppression by thalamic stimulation and drug administration. J Neurosurg 1968;29:393-6. 51. Laitinen LV, Ohno Y. Effects of thalamic stimulation and thalamotomy on the H reflex. Electroencephalogr Clin Neurophysiol 1970;28:586-91. 52. Fukusima T, Mayanagi Y, Bouchard G. Thalamic evoked potentials to somatosensory stimulation in man. Electroencephalogr Clin Neurophysiol 1976;40:481-90. 53. Fager CA. Evaluation of Thalamic and Subthalamic surgical lesions in the alleviation of Parkinson’s disease. J Neurosurg 1968;28:145-9. 54. Geddes LA. Electrodes and the measurement of bioelectric events. New York: Wiley; 1972. 55. Shils JL, Patterson T, Stecker MM. Electrical properties of metal microelectrodes. Am J Electrodiag Technol 2000;40:71-82. 56. Shils JL, Patterson T, Stecker MM. Electrical properties of metal microelectrodes. Am J Diagn Technol 2000;40:71-82. 57. Gedded LA and Baker LE, Principles of applied biomedical instrumentation. New York: Wiley; 1968. 58. Shils JL, Tagliati M, Alterman RL. Neurophysiological monitoring during surgery for movement disorders. In: Deletis V, Shils JL, editors. Neurophysiology in Neurosurgery: a modern intraoperative approach. New York: Academic Press; 2002. 59. Mink JW. The basal ganglia: focused selection and inhibition of competing motor programs. Prog Neurobiol 1996;50(4):381-425. 60. Rodriguez-Oroz MC, Rodriguez M, Guridi J, Mewes K, Chockkman V, Vitek J, DeLong MR, Obeso JA. The subthalamic nucleus in Parkinson’s disease: somatotopic organization and physiological characteristics. Brain 2001;124(9):1777-90. 61. Starr PA, Theodosopoulos PV, Turner R. Surgery of the subthalamic nucleus: use of movement-related neuronal activity for surgical navigation. Neurosurgery 2003;53 (5):1146-9; discussion 1149.
62. Follett KA, Mann MD. Effective stimulation distance for current from macro electrodes. Exp Neurol 1986;92 (1):75-91. 63. Alterman RL, Sterio D, Beric A, Kelly PJ. Microelectrode recording during posteroventral pallidotomy: impact on target selection and complications. Neurosurgery 1999;44 (2):315-21; discussion 321–3. 64. Pollak P, Krack P, Fraix V, Mendes A, Moro E, Chabardes S, Benabid AL. Intraoperative micro- and macro-stimulation of the subthalamic nucleus in Parkinson’s disease. Mov Disord 2002;17 Suppl 3:S155-61. 65. Dinner DS, Neme S, Nair D, Montgomery EB, Baker KB, Rezai A, Lu¨ders HO. EEG and evoked potential recordings from the subthalamic nucleus for deep brain stimulation of intractable epilepsy. Clin Neurophysiol 2002;113:1391-402. 66. Temel Y, Visser-Vandewalle V. Surgery in Tourette syndrome. Mov Disord 2004;19(1):3-14. 67. Visser-Vandewalle V, Ackermans L, van der Linden C, Temel Y, Tijssen MA, Schruers KR, Nederveen P, Kleijer M, Boon P, Weber W, Cath D. Deep brain stimulation in Gilles de la Tourette’s syndrome. Neurosurgery 2006;58(3):E590. 68. Alterman RL, Snyder BJ. Deep brain stimulation for torsion dystonia. Acta Neurochir Suppl 2007;97 (Pt 2):191-9. 69. Vidailhet M, Vercueil L, Houeto JL, Krystkowiak P, Benabid AL, Cornu P, Lagrange C, Tezenas du Montcel S, Dormont D, Grand S, Blond S, Detante O, Pillon B, Ardouin C, Agid Y, Destee A, Pollak P. French Stimulation du Pallidum Interne dans la Dystonie (SPIDY) study group. Bilateral deep-brain stimulation of the globus pallidus in primary generalized dystonia. New Engl J Med 2005;352(5):459-67. 70. Zhuang P, Li Y, Hallett M. Neuronal activity in the basal ganglia and thalamus in patients with dystonia. Clin Neurophysiol 2004;115(11):2542-57. 71. Coubes P, Cif L, El Fertit H, Hemm S, Vayssiere N, Serrat S, Picot MC, Tuffery S, Claustres M, Echenne B, Frerebeau P. Electrical stimulation of the globus pallidus internus in patients with primary generalized dystonia: long-term results. J Neurosurg 2004;101 (2):189-94. 72. Guridi J, Gorospe A, Ramos E, Linazasoro E, Rodriguez MC, Obeso JA. Stereotactic targeting of the globus pallidus internus in Parkinson’s disease: imagining versus electrophysiological mapping. Neurosurgery 1999;45:278-89. 73. Taha JM, Favre J, Baumann TK, Burchile KJ. Characteristics and somatotopic organization of kinesthetic cells in the globus pallidus of patients with Parkinson’s disease. J Neurosurg 1996;85:1005-12. 74. Vitek JL, Bakay RAE, Hashimoto T, Kaneoke Y, Mewes K, Yu Zhang J, Rye D, Starr P, Baron M, Turner R, Delong MR. Microelectrode guided pallidotomy:
Evoked potentials in functional neurosurgery
75.
76.
77.
78.
79. 80.
81. 82.
83.
84.
85. 86.
87.
88.
Technical approach and application in medically intractable Parkinson’s disease. J Neurosurg 1998;88:1027-43. Vitek JL, Chockkan V, Zhang JY, Kaneoke Y, Evatt M, DeLong MR, Triche S, Mewes K, Hashimoto T, Bakay RA. Neuronal activity in the basal ganglia in patients with generalized dystonia and hemiballismus. Ann Neurol 1999;46(1):22-35. Hutchison WD, Lang AE, Dostrovsky JO, Lozano AM. Pallidal neuronal activity: implications for models of dystonia. Ann Neurol 2003;53(4):480-8. Laitinen LV, Bergenheim AT, Hariz MI. Leksell’s posteroventral pallidotomy in the treatment of Parkinson’s disease. J Neurosurg 1992;76(1):53-61. Alterman RL, Kelly PJ. Pallidotomy technique and results: the New York University experience. Neurosurg Clin N Am 1998;9(2):337-43. Tasker RR, Kiss ZHT. The role of the thalamus in functional neurosurgery. Neurosurg Clin N Am 1995;6(1):73-194. Krack P, Dostrovsky J, Ilinsky I, Kultas-Ilinsky K, Lenz F, Lozano A, Vitek J. Surgery of the motor thalamus: Problems with the present nomenclatures. Mov Disord 2002;17(S3):S2-8. Jones EG. The thalamus. New York: Plenum; 1985. Hassler R. Anatomy of the Thalamus. In: Schaltenbrand G, Bailey P, editors. Introduction to stereotaxis with an atlas of the human brain. Theime; Stuttgart: p. 230-90. Hassler R. Architectonic organization of the thalamic nuclei. In: Schaltenbrand G, Walker EA, editors. Stereotaxy of the human brain: anatomical, physiological and clinical applications. New York: Theime; 1982. p. 140-80. Ilinsky IA, Kultas-Ilinsky K. Neuroanatomy organization and connections of the motor thalamus in primates. In: Kultas-Ilinsky K, Ilinsky IA, editors. Basal ganglia and thalamus in health and movement disorders. New York: Plenum; 2001. p. 77-91. Olszewski J. The thalamus of Macacca mulata. An atlas for use with stereotactic instruments. Basel: Karger; 1952. Albe-Fessard D. Evoked potential recording in functional neurosurgery: part I: use of thalamic evoked potentials to improve the stereotactic localization of electrodes in the human brain. In: Gildenberg PL, Tasker RR, editors. Textbook of stereotactic and functional neurosurgery. New York: McGraw-Hill; 1998. Giorgi C, Kelly PJ, Eaton DC, Guiot G, Dermoe G. A study of the tridimentional distribution of somatosensory evoked responses in human thalamus to aid the placement of stimulating electrodes for treatment of pain. Acta Neurochir Suppl 1980;30:279-87. Atkinson JD, Collins DL, Bertrand G, Peters TM, Pike GB, Sadikot AF. Optimal location of thalamotomy lesions for tremor associated with Parkinson’s disease: a probabilistic analysis based on postoperative magnetic resonance imaging and integrated digital atlas. J Neurosurg 2002;96 (5):854-66.
77
89. Lenz FA, Normand SL, Kwan HC, Andrews D, Rowland LH, Jones MW Seike M Lin YC, Tasker RR, Dostrovskey JO, Lenz YE. Statistical prediction of the optimal site for thalamotomy in Parkinsonian tremor. Mov Disord 1995;10(3):318-28. 90. Stranton GB. Topographical organization of ascending cerebellar projections from the dentate and interposed nuclei in Macaca mulatta: an anterograde degeneration study. J Comp Neurol 1980;190(4):699-731. 91. Kall BA, Goerss SJ, Kelly PJ. A new multimodality correlative imaging technique for VOP/VIM (VL) thalamotomy procedures. Stereotact Funct Neurosurg 1992;58:45-51. 92. Tasker RR, Organ LW, Hawrylyshyn PA. The thalamus and midbrain in man: a physiological atlas using electrical stimulation. Springfield: Charles C. Thomas; 1982. 93. Chiappa K. Evoked potentials in clinical medicine. 3rd ed. Lippincott Williams and Wilkins; 1997. 94. Gabriel EM, Nashold BS. Evolution of neuroablative surgery for involuntary movement disorders: an historical review. Neurosurgery 1998;42(3):575-91. 95. Vilesky JA, Gilman S. Using extirpations to understand the human motor cortex Horsley, Forester, Bucy. Arch Neurol 2003;60(3):446-51. 96. Cakmur R, Towle VL, Mullan JF, Suarez D, Spire JP. Intra-operative localization of sensorimotor cortex by cortical somatosensory evoked potentials: from analysis of waveforms to dipole source modeling. Acta Neurochir 1997;139:1117-25. 97. Arle JE, Apetauerova D, Zani J, Deletis V, Penny D, Hoyt D, Gould G, Shils JL. Motor Cortex Stimulation for Parkinson’s Disease: 12 month follow-up in 4 patients. J Neurosurg 2008;109(1):133-139. 98. Tsubokawa T, Katyama Y, Yamamoto T, Hirayama T, Koyama S. Chronic motor cortex stimulation in patients with thalamic pain. J Neurosurg 1993;78:393-401. 99. Saitoh Y, Kato A, Ninomiya H, Baba T, Shibata M, Mashimo T, Yoshimine T. Primary motor cortex stimulation within the central sulcus for treating deafferentation pain. Acta Neurochir Suppl 2003;87:149-52. 100. Nguyen J-P, Lefaucher J-P, Decq P, Uchiyama T, Carpentier A, Fontaine D, Brugieres P, Pollin B, Feve A, Rostaing S, Cesaro P, Kervel Y. Chronic motor cortex stimulation in the treatment of central and neuropathic pain. Correlations between clinical, electrophysiological and anatomical data. Pain 1999;82:245-51. 101. Smith H, Joint C, Schlugman D, Nandi D, Stein JF, Aziz TZ. Motor cortex stimulation for neuropathic pain. Neurosurg Focus 2001;11(3):1-9. 102. Romsto¨ck J, Fahlbusch R, Ganslandt O, Nimsky C, Strauss C. Localization of the sensorimotor cortex during surgery for brain tumors: feasibility and waveform patterns of somatosensory evoked potentials. J Neurol Neurosurg Psychiatry 2002;72:221-9.
1281
1282
77
Evoked potentials in functional neurosurgery
103. Taniguchi M, Cedzich C, Schramm J. Modification of cortical stimulation for motor evoked potentials under general anesthesia: technical description. Neurosurgery 1993;32:219-26. 104. Penfield W, Boldrey E. Somatic motor and sensory representation in the cerebral cortex of man as studied by electrical stimulation. Brain 1969;60: 339-443. 105. Sartorius CJ, Wright G. Intraoperative brain mapping in a community setting: technical considerations. Surg Neurol 1997;47:380-8. 106. Sala F, Lanteri P. Brain surgery in motor areas: the invaluable assistance of intraoperative neurophysiological monitoring. J Neurosurg Sci 2003;47:79-88. 107. Szele´nyi A, Joksimovicˇ B, Seifert V. Intraoperative risk of seizures associated with transient direct cortical
stimulation in patients with symptomatic epilepsy. J Clin Neurophysiol 2007;24(1):39-43. 108. Roux FE, Boulanouar K, Ranjeva JP, Manelfe C, Tremoulet M, Sabatier J, Berry I. Cortical Intraoperative Stimulation in Brain Tumors as a Tool to Evaluate Spatial Data from Motor Functional MRI. Invest Radiol 1999;34(3):225-9. 109. Arle JE, Apetauerova D, Brophy S, Shils JL. Dyskinesia in Parkinson’s Disease treated by deep brain stimulation once electrode position was revised: case report. Neuromodulation 2007;10(3);238-43. 110. Yingling CD, Ojermann S, Dodson B, et al. Identification of motor pathways during tumor surgery facilitated by multichannel electromyographic recording. J Neurosurg 1999;91:922-7.
Functional Neurosurgery – Technical Aspects
76 Image Guided Functional Neurosurgery S. Khan . N. K. Patel . E. White . P. Plaha . S. Ashton . S. S. Gill
Introduction Since the early descriptions of human stereotactic surgery, image guidance has played an essential role in these procedures [1]. There has been a transition from skull radiographs, ventriculography through to CT scanning and MRI, providing assistance in the planning of surgery. Initially these imaging techniques gave an indirect location for the subcortical structure’s being targeted. The indirect methods are based on brain atlases and typically use the anterior commissure (AC) and posterior commissure (PC) as internal landmarks to co-register the atlas with the patient [2–4]. However there is substantial individual variation in AC-PC based coordinates of subcortical nuclei. Direct visualization of the target morphology using high resolution MRI can demonstrate significant variations in anatomy between individuals and also between different hemispheres in the same individual [5–7]. Normalization of atlas-based coordinates to the dimensions of the patient’s brain is based on the assumption that a linear transformation between two brains is possible, a concept that has never been proven. To compensate for the individual variations when using the indirect method, many centers have developed intraoperative clinical and electrophysiological monitoring procedures. The use of microelectrode recording and macroelectrode stimulation/recording of the target sites is regarded by many to be a pre-requisite for accurate delivery of therapy. Intraoperative stimulation is performed with the patient awake, in the off medication state, so that functional #
Springer-Verlag Berlin/Heidelberg 2009
change can be observed. This can prolong operative time and requires additional staff to be present in theatre, i.e., neurologists. The passing of multiple microelectrodes is associated with greater expense, higher complication rates and potentially prolonged procedures, again exposing patients to many hours of awake surgery, which can be accompanied by significant stress for both the patient and operative staff [8–11]. These constraints highlight the need for more accurate, simpler and quicker targeting methods.
General Overview In this chapter, the authors discuss the issues surrounding the use of MRI for direct target visualization, determining the target position in stereotactic space, accurate targeting and confirmation of targeting. We aim to briefly cover the approaches used by our unit for direct targeting using high resolution MRI, employing an implantable guide tube (> Figure 76-1), with intraoperative MRI for confirmation of accurate placement.
Clinical Methods and Rationale Defining Functional Targets on MR-Images The direct visualization of small subcortical nuclei has in the past been limited by constraints in MR technology, the prolonged scan acquisition times required for high resolution MR
1240
76
Image guided functional neurosurgery
. Figure 76-1 Guide tube with a threaded cylindrical hub and a dome shaped proximal end; and adjacent stylette whose T-shaped proximal end fits within the guide tube’s hub and whose distal end projects beyond the guide tube
appropriate scan parameters in a standard 1.5 T MRI scanner.
MRI Scan Parameters In addition to the hardware related parameters MRI image quality is also dependent on scanning parameters that dictate image resolution (matrix size, slice thickness), SNR and lack of artifact.
images and motion artifact, especially in patients with movement disorders. However advances in MR technology in conjunction with manipulation of scan parameters has enabled the direct visualization of thalamic and subthalamic nuclei.
MRI Scanners MRI Image quality is affected by limitations in the field strength, field homogeneity and acquisition parameters. Advances in major aspects of MR imaging have improved these limitations enabling the direct visualization of brain targets for functional Neurosurgery. The first of these has been the increase in Magnet field strength from 1 to 1.5 and in limited circumstances 3 and 7 T platforms. Most modern scanners also automatically adjust the uniformity of the magnetic field, a process known as shimming, producing a homogonous field. In conjunction with these improvements, increases in magnetic field gradients and improved coil design have resulted in improved signal to noise ratio (SNR), spatial resolution and reduced acquisition times [12]. This has allowed for the direct visualization of functional targets such as the Globus pallidus interna and the subthalamic nucleus using
The SNR can be improved by increasing field of view and slice thickness, however this results in a loss of spatial resolution, which is counterproductive when attempting to visualize some of the small subcortical nuclei targeted in diseases such as PD. The commonest method of increasing SNR is by increasing the number of signal averages (NSA/NEX) performed during a scan. This results in the reduction of noise that is randomly generated and therefore cancelled out over successive averages. This strategy is limited by time constraints. Improvements in SNR are related to the square root of the increase in NSA, whereas increase in scan time is directly related to increase in NSA. Hence a fourfold increase in NSA, will result in a fourfold increase in scan time but only a doubling of SNR [13]. This can be particularly problematic in patients with movement disorders, where significant movement artifact prevents the use of long scan times. In our unit, we rely solely on high-resolution MRI images for planning of functional Neurosurgery and therefore we apply a modified Leksell frame and acquire pre-operative MRI images with the patients anesthetized and paralyzed, under strict stereotactic conditions. This prevents movement artifact, enables the use of long acquisition times with a high NSA and therefore SNR. Image quality is also degraded by ‘‘cross talk’’ from adjacent slices when acquiring data in a
Image guided functional neurosurgery
continuous acquisition. This arises because the profile of a slice is never ideal. This results in signal from tissue adjacent to the selected slice affecting the image contrast/quality. Depending on the exact nature of the sequence, sequence timings (e.g., TE times) and relaxation constants of the tissue being imaged the extent of this cross-talk will vary. Placing a gap between slices reduces the ‘‘cross talk’’ from adjacent tissue [13]. In addition to optimizing SNR and therefore contrast between different tissues, MR sequences used for planning of functional Neurosurgery require a high spatial resolution in order to optimize the visualization of the small subcortical nuclei targeted in diseases such as PD. Image resolution is determined by voxol size, which is related to matrix size, field of view and slice thickness. MRI brain sequences are able to use a small field of view, however reductions in matrix size and slice thickness in order to optimize spatial resolution must be offset against loss of SNR due to these changes [12]. Acquiring images in the AC-PC plane enables correlation of the position of targets such as the STN with surrounding structures like the Red Nucleus and Substantia nigra. This also allows for the use of a brain atlas (e.g., Schaltenbrand and Warren, Stuttgart: Thieme), sliced in the AC-PC plane, as a reference tool during the planning of surgery.
Applying these principles has enabled the visualization of thalamic nuclei [14], the STN [6,15] and GPi [16] in a standard 1 and 1.5 T MR scanner. In our experience the subthalamic nucleus (STN) is best seen on high-resolution T2-weighted images (1.5 T TR 4000, TE 120, TSE 11, NSA 12, 2 mm slice, 0.4 mm gap, voxol size 0.45 0.45 mm) [17–19]. This sequence also provides good visualization of the Red nucleus and the mammillothalamic tract (> Figure 76-2).
76
. Figure 76-2 High-resolution axial T2-weighted MR images showing bilateral STN
We base target planning for the putamen and globus pallidus on a combination of high-resolution T2-weighted and proton density sequences (TR 4000, TE 15, TSE 7, NSA 8) [16].
Clinical Method 1.
2.
Our modified Leksell stereotactic frame (Elekta Instrument AB, Stockholm, Sweden), with non-conducting plastic posts, is positioned low on the head parallel to the orbito-meatal plane, which is used as an approximation of the AC-PC plane, and fixed to the skull using insulated pins. The frame is applied under general anesthesia, which is maintained during the imaging. A mid-sagittal plan scan is acquired and the AC and PC are visualized. Images are acquire parallel (axial) and perpendicular (coronal) to the AC-PC plane to allow direct comparison with an atlas. (> Figure 76-3a and b) If the frame is not parallel then the orientation can be readily adjusted by loosening the fixation of the anterior posts
1241
1242
76
Image guided functional neurosurgery
. Figure 76-3 Midline Saggital MRI scans demonstrating the orientation of axial and coronal images in relation to the anterior and posterior commisure
3.
4.
to the frame and sliding them up or down as appropriate before re-tightening them. Deep brain targets are defined on long acquisition, high-resolution images acquired under strict stereotactic conditions, in both the axial and coronal planes. Different sequences are used to optimize visualization of the different targets. This use of appropriate sequences provides good visualization of the STN, Red nucleus and the mammillothalamic tract (> Figure 76-2) [17–19]. The location of other structures in the subthalamic region such as the zona incerta and prelemniscal radiation can be determined indirectly from these structures and with reference to a brain atlas used as a visual guide. Target visibility may be increased by adjusting the window setting of the T2-weighted images maximizing the gray/white matter contrast and by using magnified hard copy images. The boundaries of visible targets in the subthalamic region, such as the STN, may be enhanced by overlaying inverted images on
to standard T2 images, a method, which neutralizes the gray areas and allows for easier identification of the bright STN edges on a dark background. Often boundaries poorly seen in one imaging plane are better seen on another. With crosscorrelation of STN in axial, coronal and sagittal planes, the boundaries can be further identified, confirmed and a 3-D map of the target constructed.
Transposing the Target Position into Stereotactic Space The accuracy of MRI stereotaxis was problematic early in it development due to intrinsic distortion in the MR field, patient related artifact and image distortion from stereotactic frame systems. As detailed above newer MRI scanners have minimal intrinsic image distortion and can correct for global non-patient related distortion, resulting
Image guided functional neurosurgery
in a homogonous field. Patient related distortion occurs due to the differences in magnetization of varying tissue (magnetic susceptibility artifact) and the local chemical environment of the protons in the structure being scanned (chemical shift). Both can generate artifact as well as image distortion. Magnetic susceptibility artifact can be minimized with the use of spin echo (SE) sequences, high receiver bandwidth (strong gradients) or short time to echo (TE). As the fat and fluid composition of the brain is largely homogenous there is negligible chemical shift, again this can be reduced with the use of strong gradients [20–25]. Distortion of fiducial systems can significantly affect accuracy of anatomical targeting. Fiducials are usually placed at the periphery of the image where there is maximal magnetic susceptibility artifact and fiducials using a fatty medium would also be subject to chemical shift artifact [26–28]. MRI fiducial systems should ideally be placed as close to the patients head as possible and consist of a non-fat containing medium. The Leksell frame has been demonstrated to generate minimal distortion [26]. Any residual distortion can be mapped by the use of Phantom studies. Our own unit currently uses two Phantoms, the first produced by Elekta Intstruments and the second has been designed by Renishaw plc (Wotton-under-edge, Gloucestershire, UK) to fit into the Leksell frame. These are self-contained cylinders of parallel rods, where the cylinders are a solid material and the rods are filled with copper sulfate solution. The first Phantom considers the Axial plane and the second can be rotated to consider both the Coronal and Sagittal planes. Other forms of Phantoms created for distortion correction have been investigated [21,29,30]. A comparison of the rod positions from each MR image and the actual geometric positions of the rods provide a spatial transformation. The transformation obtained is used to correct distorted images in
76
the same plane and location. Specialist software has been created by Renishaw plc to obtain the transformation and apply the distortion correction to the images. An alternate and commonly used method of stereotactic localization is to fuse MRI with CT. Stereotactic space is defined by the CT images and MR is used for target visualization. CTstereotactic accuracy is dependent on mechanical factors and image acquisition parameters. Mechanical movement of the CT Gantry and table must be checked as part of the CT imaging protocol [31]. Modern CT scanners allow the rapid acquisition of thin slices enabling a high degree of stereotactic accuracy [32]. However fusion of MR to CT can introduce an error of up to 1.3 mm [33]. Morphing of MR with CT is performed on a best match or least error basis and true fusion has not yet been demonstrated [34].
Clinical Methods 1.
2.
The 3-D coordinates of a selected target are determined by overlaying a transparency with a 1mm grid scaled to match the magnification of the magnified hard copy images (e.g., 1.6). The center of the grid is positioned in the center of the stereotactic space by aligning four reference points on the transparency with the fiducials visible on the image. The positions of the reference points on the transparency can be adjusted to accommodate for the geometric distortion in the particular MRI scanner after carrying out phantom studies. Alternatively, the specialist software can provide images where the distortions have been corrected, and stereotactic co-ordinates are produced taking this into account. The trajectory is planned. For DBS cases the electrode contacts are outlined on the images and typically the second contact of a quadripolar lead (contacts 1 or 5 of lead
1243
1244
76
Image guided functional neurosurgery
3389 or 3387, Medtronic Inc.) is placed at the target site.
Surgical Procedure and Peri-Operative Confirmation of Accurate Targeting Having visualized and obtained accurate stereotactic co-ordinates for the surgical target a DBS electrode, or other therapeutic device, is delivered to the target. Minor displacement of DBS electrodes from the optimal target position can result in significant sensory, motor and emotional side effects and so confirmation of correct placement is essential. However errors in targeting can occur due to a number of reasons: 1. 2. 3.
Miscalculation of co-ordinates Human error in setting the stereoguide Intraoperative brain shift due to (a) Change in the patients head position relative to the frame (b) Loss of CSF and pneumocephaly [35] (c) Movement of the brain whilst advancing the therapeutic device/probe through the parenchyma (d) Deflection of the therapeutic device/ probe (e.g., dural edge)
It is therefore essential to minimize these factors. The stereotactic frame should be placed under sufficient tension or can be embedded into the outer table of the skull to prevent displacement. A second individual should independently check the target co-ordinates and stereoguide settings. Intraoperatively placing the patient at 45 head up places the burr holes uppermost and minimizes CSF loss. Whilst the dura is open the
burr hole can be constantly irrigated with saline in order to prevent air entry. Conventional methods of refining target position, accounting for intraoperative brain shift and confirming accuracy of placement include MER, intraoperative stimulation with assessment of clinical response and MRI/CT confirmation of electrode position. The shortcoming of these methods, increased hemorrhage risk with MER, prolonged operative time and therefore patient discomfort with both awake MER and intraoperative stimulation have been well described [8,9,11,36]. The use of postoperative MRI/CT for conformation of electrode placement relies on the assumption that the electrode is in the center of the signal void/artifact it produces on these images and image distortion produced by the electrode is minimal. In addition MRI scanning with the presence of an implanted electrode carries some risk of contact heating and tissue damage [37]. The technique described below avoids brainshift, minimize hemorrhage risk and enables intraoperative conformation of electrode placement without causing image distortion [38].
Clinical Methods 1.
2.
3.
Surgery is performed under general anesthesia in a semi-sitting position, such that the frontal burrholes are uppermost. The burrhole is made with a one-quarter inch drill, guided in a pre-planned trajectory by the stereoguide, and is of sufficient size to visualize and coagulate cortical vessels under continuous saline irrigation, minimizing any CSF loss (> Figure 76-4). A elongated stop (Renishaw PLC, UK) is fixed into the upper carriage and with the stereoguide set to the desired coordinates and trajectory, and rechecked, a probe is inserted through the stop to the level of the skull surface (> Figure 76-5).
Image guided functional neurosurgery
. Figure 76-4 Burrhole is made with a one-quarter inch drill, guided in a pre-planned trajectory by the stereoguide
4.
5.
6.
The distance of the probe above the stop, which equates with the distance from the skull to the target, is measured and enables a carbothane guide tube (Renishaw PLC, UK) to be cut to an appropriate length. For the insertion of DBS electrodes the guide tube is generally shortened so that when inserted, its distal end will be 12 mm short of the target thus ensuring that when the DBS lead is implanted all the four contacts will be exposed. The guide tube is inserted into the split guide (Renishaw PLC, UK) that is fixed into the lower carriage of the stereoguide. The probe is then advanced to the target through the stop, and the appropriately sized guide tube, that is held in
76
. Figure 76-5 A probe is inserted through a elongated stop (Renishaw PLC, UK) fixed into the upper carriage to the level of the skull surface and enables the guide tube to be cut to an appropriate length
7.
8.
9.
the instrument carriers split guide block (> Figure 76-6). The split guide is then unclamped from the lower carriage and its halves removed allowing the guide tube to be advanced over the probe to the vicinity of the brain target. Cellulose gauze is laid over the dura around the guide tube and acrylic cement placed into the burrhole. The hub of the guide tube is seated in the acrylic cement, which once set secures the guide tube in place. The probe is now removed. The length from the top of the guide tube dome to the stereoguide datum is measured. The probe is replaced with a radio-opaque stylette, whose T shaped proximal end fits
1245
1246
76
Image guided functional neurosurgery
. Figure 76-6 Probe inserted through stop and guide tube to the brain target
within the hub, cut to length such that its distal end projects beyond the distal end of the guide tube into the target (> Figure 76-7). 10. Our technique of guide tube and stylette implantation performed under GA typically takes 20–30 min per side. 11. The scalp wound is now closed and the patient is transferred to a MRI or CT scanner where the position of the stylette is defined in relationship to the desired target (> Figure 76-8) with images acquired under stereotactic conditions. The peri-operative images are laid onto the plan scans, formatted as inverted images, and any displacement of the stylette from the planned target is measured. 12. Prior to insertion of a DBS lead (e.g., DBS 3389 or 3387 lead Medtronic Inc., Minnea-
. Figure 76-7 Hub of the guide tube is fixed within the burrhole with acrylic cement. Probe is replaced with a radio-opaque stylette cut to the appropriate length such that its distal end projects beyond the guide tube into the target
polis), the length to be inserted is marked off by a sutured stop around the lead, defined by the length of the stylette that has been withdrawn from the guide tube. Once inserted the leads tungsten guide wire is removed and the lead is bent through a 90 arc conforming with that within the slotted hub of the guide tube. The lead is then secured to the skull with a miniplate and screws. The electrode connectors are counter sunk into channels made in the skull. This improves cosmesis by eliminating the prominent profile of the connectors but also reduces complications such as skin erosion and lead displacement
Image guided functional neurosurgery
. Figure 76-8 Peri-operative inverted coronal T2 weighted image (inferior) verifying the position of the radio-opaque stylettes within the planned STN target (pre-operative high-resolution T2-weighted image (superior) from which the STN and surrounding structures can be visualized, and are outlined on the peri-operative inverted image, inclusive of the visible stylette). Perioperative images are obtained in the same slice configuration as the pre-operative planning images
76
. Figure 76-9 Radio-opaque stylette is replaced with a DBS lead with depth of insertion marked off by tying a suture around the lead, as defined by the length of stylette withdrawn. DBS connectors are placed into troughs drilled into the skull
errors that have arisen during target localization, coordinate calculation or during the operative procedure. In these circumstances if the error is large, for example 2 mm or more, then it may be corrected by the implantation of another guide tube and stylette, through another burrhole and trajectory, whilst the sub optimal guide tube and stylette remain in-situ and act as an internal reference and a brain anchor limiting brain shift. On repeat confirmation of target localization, the sub-optimal guide tube and stylette are then removed.
13.
that are associated with the connectors lying in the posterior-auricular position [37] (> Figure 76-9). Implantation of bilateral DBS leads and connection to an implanted generator adds approximately 45–60 min; with total operative time, inclusive of peri-operative MR-imaging and transfers being about 3–3½ h. The peri-operative image may demonstrate a displacement of the radio-opaque stylette from the chosen target. This may result from
If there is a small error in electrode position, for example below 2 mm, in future this will be compensated by the use of DBS electrodes with stearable electric fields (in development) enabling a shift in the center of field by 1.5 mm.
Clinical Indications The guide tube may act as a port for the implantation of electrodes not only for deep brain stimulation (DBS) but also for radiofrequency lesioning; catheters for drug or trophic substance
1247
1248
76
Image guided functional neurosurgery
delivery; neural stem cell or encapsulated cell transplantation; and viral-vector delivery.
Deep Brain Stimulation This method is useful for localizing functional targets without the use of microelectrode recording and or macroelectrode stimulation/recording. Performing our pre-operative plan MRI scans under general anesthesia has enabled us to use long acquisition images, eliminate movement artifact and optimizing image sensitivity whilst maintaining high spatial resolution. This technique has enabled us to directly visualize our target structures, and by cross correlating axial and coronal images, produce a volumetric representation of the intended target. In conjunction with intraoperative MRI in order to confirm guide tube and indwelling stylette placement this technique has resulted in consistent clinical outcomes that have enabled us to discontinue awake surgery, a practice that can be stressful for both patients and surgeon. More recently, the authors have utilized this method for the implantation of deep brain stimulating electrodes into a novel target site, the pedunculopontine nucleus, that had no defined and validated microelectrode signature [39]. The implantable guide tube is also useful for gaining repeat access to functional targets without the need to repeat the whole stereotactic procedure. The device facilitates non-stereotactic replacement of hardware, for example DBS electrodes or catheters, following complications of migration, fracture, malfunction or infection of these therapeutic devices. The device facilitates repeated or staged radiofrequency lesioning at the same target. The radio-opaque stylette may be left in situ within the indwelling guide tube as a permanent marker of the lesion site. Should there be a decline in the functional benefit following a lesion it is simple for the surgeon to repeat the lesion without the extensive work-up and imaging that is usually required. For small
functional targets in eloquent areas, such as the subthalamic nucleus, it may be safer to make an undersized lesion in the first instance and repeat the lesion at a later date to optimize its size. Finally, the guide tube would facilitate replacement of a DBS electrode with a permanent radiofrequency lesion.
Clinical Outcomes The clinical outcome of our patients undergoing subthalamic nucleus stimulation is comparable to outcomes reported in literature. This method has also allowed us to retrospectively correlate the degree of motor improvement with the anatomical location of the optimal stimulation contact in patients with PD. We have subsequently identified the caudal zona incerta as a better target than the STN for deep brain stimulation in the treatment of Parkinson’s disease, into which we now implant deep brain stimulation electrodes [19]. The therapeutic effect of stimulating the pedunculopontine nucleus for the previously treatment resistant axial symptoms of PD has been recently reported [39]. Our results on subthalamic region stimulation for the treatment of essential tremor, and more recently the stimulation of the Zona incerta specifically for multiple forms of tremor have been published [40].
Accuracy of Surgical Technique and Post-Operative Complications Despite concerns regarding the use of the Leksell G frame in MRI stereotaxy, the accuracy of this technique has been repeatedly demonstrated. We have previously reported a mean error of 0.3 0.4 mm in the medial to lateral plane and 0.4 0.4 mm in the AP plane, prior to adopting the use of an indwelling guide tube [17]. Similar findings have been reported by Simon et al,
Image guided functional neurosurgery
and more recently by our group whilst using implanted guide tubes [38,41]. This technique, in contrast to methods employing intraoperative clinical and electrophysiological monitoring procedures using microelectrodes and/or macroelectrode stimulation of the target sites, reduces the number of brain trajectories and the potential associated brain trauma and hemorrhage risk. Following implantation of 506 guide tubes in 250 patients, there were 10 procedure-related complications. One patient developed a non-hemorrhagic paresis with expressive dysphasia following which he continues to make a gradual recovery of function and at the 12-month follow-up he exhibited a milder deficit. One patient developed a small cortical hematoma with a postoperative paresis that has completely resolved after 6 months. Another patient developed dysphagia for 3-months as a consequence of mistargeting secondary to an error in frame relocation, with both initial guide tubes and stylettes implanted into the thalami bilaterally. There was one post-operative self-limiting grand mal seizure and two pulmonary emboli, one of which was fatal. There have been two infections requiring removal of implanted hardware.
MRI Guided Direct Intracranial Drug Delivery In order to circumvent the blood-brain barrier, a number of techniques have been developed to infuse therapeutic agents, including viral vectors, neurotrophic proteins, chemotherapeutics and stem cells, directly into the brain. With the exception of biodegradable carmustine wafers, these techniques have yet to reach routine clinical use, but have been employed in a number of clinical trials.
Injection of drug-producing cells: e.g., Fibroblasts, genetically-modified to produce nerve growth factor (NGF), have been injec-
76
ted into the cholinergic basal forebrain of eight patients with mild Alzheimer’s disease as part of a phase 1 clinical trial [42]. Limited clinical and PETevidence of therapeutic efficacy was demonstrated. Encapsulated cells: e.g., Encapsulated cells, genetically-modified to produce human ciliary neurotrophic factor (CNTF), have previously been implanted into the right lateral ventricle of patients with Huntington’s disease, as part of a phase 1 clinical trial [43]. No significant clinical benefit was observed. Biodegradable polymers: e.g., Carmustine (BCNU) wafers in patients with recurrent high-grade gliomas. Injection: Intraparenchymal: Numerous clinical trials have utilized intraparenchymal injection to administer therapeutic agents into the brain, most commonly to treat high-grade gliomas. This technique has largely been superseded by convection-enhanced delivery. Intraventricular: e.g., Glial cell linederived neurotrophic factor (GDNF) and nerve growth factor (NGF) have been administered intraventricularly in phase 1 clinical trials to patients with Parkinson’s disease [44] and Alzheimer’s disease [45], respectively. Neither trial demonstrated significant clinical benefit. Convection-enhanced delivery (CED): This represents an extremely promising approach to administering therapeutic agents directly into the brain, as in contrast to all other techniques of direct intracranial drug delivery, drug distribution is achieved by bulk flow down a pressure gradient, rather than by diffusion, down a concentration gradient. Consequently it is possible to deliver therapeutic agents, regardless of their molecular size or infused concentration, homogeneously over large, controlled volumes of brain. Unfortunately despite the potential of this approach
1249
1250
76
Image guided functional neurosurgery
and its use in a number of large phase III clinical trials, including the recently completed PRECISE trial (Phase 3 Randomized Evaluation of Convection-Enhanced Delivery of IL-13PE38QQR Compared to Gliadel Wafer with Survival Endpoint in Glioblastoma Multiforme at First Recurrence) CED has yet to be successfully translated into routine clinical practice.
The Role of Imaging in Direct Intracranial Drug Delivery Imaging has a central role in the development of neurosurgical techniques of direct intracranial delivery for two fundamental reasons: 1.
2.
Drug distribution within disease-specific target structures is inextricably linked to drug efficacy. Drug penetration into surrounding structures may be linked to the occurrence of side-effects.
Imaging facilitates the accurate and controlled delivery of therapeutic agents into the brain in a number of important ways:
Conventional stereotactic techniques enable the accurate placement of catheters or needles into target structures prior to infusing therapeutic agents. Molecular imaging to visualize drug efficacy e.g., PET. Visualizing the distribution of therapeutic agents within the brain. This is a rapidly expanding area of research that offers enormous potential in enhancing the clinical utility of cell-based therapies and convectionenhanced delivery.
Imaging Cell-based Therapies Whereas the short half-life of isotopic tracers renders PET and single-photon emission computed tomography (SPECT) ineffectual for prolonged in vivo cellular imaging, it is possible to label cells including embryonic, mesenchymal and neural stem cells with MR contrast agents. To date, both gadolinium chelates and iron oxide particles have been used to label cells in preclinical studies, although due to a paucity of toxicity data, neither has been used in clinical trials. However the potential of this approach has been demonstrated in a number of animal models. For example, ultra-small particle iron oxide (USPIO) labeled embryonic stem cells have been visualized migrating, across the corpus callosum, in a rat stroke model, from their implantation site, into the region of the infarct in the contralateral hemisphere [46].
The Role of Imaging in Convection-Enhanced Delivery A number of imaging strategies are in development to facilitate the administration of therapeutic agents into the brain by CED.
MRI-based Predictive Modeling of Drug Distribution Using patient-specific data, derived from a range of MR-based imaging techniques, including diffusion tensor imaging (DTI), and computational fluid dynamics, software algorithms for predicting drug distribution, by CED, have been developed. These mathematical models facilitate surgical planning of CED, by allowing pre-operative visualization of the predicted drug distribution and extent of infusate reflux along the catheter-brain interface. FDA approved software (iPlan Flow – BrainLAB), incorporating these features and integrated with surgical
Image guided functional neurosurgery
navigation systems, has been developed to facilitate the administration of therapies to patients with brain tumors, in clinical trials. The predictive value of these algorithms in clinical practice is at present relatively limited. For example, when the predicted volume of distribution of cintredekin besudotox in patients with high-grade glioma, was compared to the volume of distribution of 123I-human serum albumin (HSA), observed by SPECT imaging, there was a concordance of only 65.75% (95% CI: 52.0–79.5%) [47]. In addition to the potential limitations of the predictive model used, this disparity may have resulted from the different distribution properties of cintredekin besudotox and 123I-HAS, tissue heterogeneity in the vicinity of high-grade gliomas, and the low spatial resolution of SPECT imaging, compared to MRI.
Surrogate Markers of Drug Distribution An alternative to predicting drug distribution from pre-operative imaging is to coinfuse, with the drug, a contrast agent, designed to act as a surrogate marker of drug distribution. A number of strategies have been developed including the administration of gadolinium-DTPA with glucocerebrosidase to a patient with Gaucher’s disease [48], as well as liposomal gadolinium [49] and iron oxide nanoparticles coated with dextran (ferumoxtran-10) [50], to visualize the distribution of liposomally-encapsulated drugs and adeno-associated virus (AAV), respectively. The obvious limitation of this approach is that the distribution properties of the surrogate marker and therapeutic agent are likely to be significantly different. However this is likely to represent a useful and complementary approach to predictive models of drug distribution.
76
Imaging Drug Distribution There is limited evidence that certain MRI acquisitions, in isolation, may provide limited visualization of drug distribution by CED. For example, diffusion-weighted MRI has been shown to be predictive of the response to paclitaxel, administered by CED, to patients with high-grade gliomas.
Conclusions The use of an implantable guide tube and stylette in stereotactic functional neurosurgery allows the surgeon to verify target localization visually on perioperative high-resolution MR-images. For DBS implantation this method enables direct correlation of the anatomical location of the active contacts with patient outcome and with continuous audit leads to improved practice. The method avoids exposing the patients to long hours of awake surgery and limits the number of brain probings in comparison to techniques employing MER. The device acts as a port for the delivery of DBS and lesioning electrodes, catheters for drug delivery, viral-vector delivery and cell implantation; and for repeated non-stereotactic access to the functional brain targets.
Acknowledgments We wish to thank our nurse specialists, Mrs Lucy Mooney and Mrs Karen O’ Sullivan, for carrying out the programming and assessments of patients whose data has contributed to this book chapter. We also wish to thank Ms Becky Durham and Ms Ruth Blachford for the illustrations. Steven Gill, Sadaquate Khan and Edward White are consultants to Renishaw PLC. None of the other Authors have any conflict of interest to declare. The intended use of the Renishaw ‘neuro| guide” electrode introducer Kit as defined by the manufacturer is “to provides a conduit through
1251
1252
76
Image guided functional neurosurgery
which Deep Brain Stimulation (DBS) electrodes can be delivered to allow stimulation of sub cortical targets within the brain.’ This device is CEmarked for this purpose only in Europe.
References 1. Spiegel EA, et al. Stereotaxic Apparatus for Operations on the Human Brain. Science 1947;106(2754):349-50. 2. Hariz MI, Bergenheim AT. A comparative study on ventriculographic and computerized tomography-guided determinations of brain targets in functional stereotaxis. J Neurosurg 1990;73(4):565-71. 3. Hariz MI. Correlation between clinical outcome and size and site of the lesion in computed tomography guided thalamotomy and pallidotomy. Stereotact Funct Neurosurg 1990;54–55:172-85. 4. Holtzheimer PE III, Roberts DW, Darcey TM. Magnetic resonance imaging versus computed tomography for target localization in functional stereotactic neurosurgery. Neurosurgery 1999;45(2):290-7; discussion 297–8. 5. Brierley JB, Beck E. The significance in human stereotactic brain surgery of individual variation in the diencephalon and globus pallidus. J Neurol Neurosurg Psychiatry 1959;22:287-98. 6. Ashkan K, et al. Variability of the subthalamic nucleus: the case for direct MRI guided targeting. Br J Neurosurg 2007;21(2):197-200. 7. Patel NK, Khan S, Gill S. Comparison of atlas- and magnetic-resonance-imaging-based stereotactic targeting of the subthalamic nucleus in the surgical treatment of Parkinson’s Disease. Stereotact Funct Neurosurg 2008;86:153-61. 8. Chevrier E, et al. Is there a role for physiotherapy during deep brain stimulation surgery in patients with Parkinson’s disease? Eur J Neurol 2006;13(5):496-8. 9. Gorgulho A, et al. Incidence of hemorrhage associated with electrophysiological studies performed using macroelectrodes and microelectrodes in functional neurosurgery. J Neurosurg 2005;102(5):888-96. 10. Hariz M, Blomstedt P, Limousin P. The myth of microelectrode recording in ensuring a precise location of the DBS electrode within the sensorimotor part of the subthalamic nucleus. Mov Disord 2004;19(7):863-4. 11. Hariz MI, Fodstad H. Do microelectrode techniques increase accuracy or decrease risks in pallidotomy and deep brain stimulation? A critical review of the literature. Stereotact Funct Neurosurg 1999;72(2–4):157-69. 12. McRobbie DW, editor. MRI from picture to proton. Cambridge: Cambridge University Press; 2003. p. 372. 13. Westbrook C, editor. MRI in practice. Oxford: Blackwell Publishing; 2005. p. 422.
14. Deoni SC, et al. Visualization of thalamic nuclei on high resolution, multi-averaged T1 and T2 maps acquired at 1.5 T. Hum Brain Mapp 2005;25(3):353-9. 15. Rampini PM, et al. Multiple sequential imagefusion and direct MRI localisation of the subthalamic nucleus for deep brain stimulation. J Neurosurg Sci 2003;47(1):33-9. 16. Hirabayashi H, Tengvar M, Hariz MI. Stereotactic imaging of the pallidal target. Mov Disord 2002; 17 Suppl 3:S130-S134. 17. Patel NK, et al. MRI-directed subthalamic nucleus surgery for Parkinson’s disease. Stereotact Funct Neurosurg 2002;78(3–4):132-45. 18. Patel NK, et al. MRI directed bilateral stimulation of the subthalamic nucleus in patients with Parkinson’s disease. J Neurol Neurosurg Psychiatry 2003;74(12): 1631-7. 19. Plaha P, et al. Stimulation of the caudal zona incerta is superior to stimulation of the subthalamic nucleus in improving contralateral parkinsonism. Brain 2006;129 (Pt 7):1732-47. 20. Reinsberg SA, et al. A complete distortion correction for MR images. II. Rectification of static-field inhomogeneities by similarity-based profile mapping. Phys Med Biol 2005;50(11):2651-61. 21. Doran SJ, et al. A complete distortion correction for MR images. I. Gradient warp correction. Phys Med Biol 2005;50(7):1343-61. 22. Tanner SF, et al. Radiotherapy planning of the pelvis using distortion corrected MR images: the removal of system distortions. Phys Med Biol 2000;45(8):2117-32. 23. Woo JH, Kim YS, Kim SI. The correction of MR images distortion with phantom studies. Stud Health Technol Inform 1999;62:388-9. 24. Maurer CR Jr, et al. Effect of geometrical distortion correction in MR on image registration accuracy. J Comput Assist Tomogr 1996;20(4):666-79. 25. Young IR, et al. The benefits of increasing spatial resolution as a means of reducing artifacts due to field inhomogeneities. Magn Reson Imaging 1988;6(5):585-90. 26. Walton L, et al. A phantom study to assess the accuracy of stereotactic localization, using T1-weighted magnetic resonance imaging with the Leksell stereotactic system. Neurosurgery 1996;38(1):170-6; discussion 176–8. 27. Yu C, et al. A phantom study of the geometric accuracy of computed tomographic and magnetic resonance imaging stereotactic localization with the Leksell stereotactic system. Neurosurgery 2001;48(5):1092-8; discussion 1098–9. 28. Michiels J, et al. On the problem of geometric distortion in magnetic resonance images for stereotactic neurosurgery. Magn Reson Imaging 1994;12(5):749-65. 29. Menuel C, et al. Characterization and correction of distortions in stereotactic magnetic resonance imaging for
Image guided functional neurosurgery
30.
31. 32.
33.
34.
35.
36.
bilateral subthalamic stimulation in Parkinson disease. J Neurosurg 2005;103(2):256-66. Wang D, Doddrell DM, Cowin G. A novel phantom and method for comprehensive 3-dimensional measurement and correction of geometric distortion in magnetic resonance imaging. Magn Reson Imaging 2004;22 (4):529-42. Kondziolka D, editor. Radiosurgery, vol 6, Pittsburgh: Karger; 2006. Landi A, et al. Accuracy of stereotactic localisation with magnetic resonance compared to CT scan: experimental findings. Acta Neurochir (Wien) 2001;143(6): 593-601. Duffner F, et al. Relevance of image fusion for target point determination in functional neurosurgery. Acta Neurochir (Wien) 2002;144(5):445-51. Rezai AR, et al. Deep brain stimulation for Parkinson’s disease: surgical issues. Mov Disord 2006;21 Suppl 14: S197-S218. Miyagi Y, Shima F, Sasaki T. Brain shift: an error factor during implantation of deep brain stimulation electrodes. J Neurosurg 2007;107(5):989-97. Hariz MI. Safety and risk of microelectrode recording in surgery for movement disorders. Stereotact Funct Neurosurg 2002;78(3–4):146-57.
76
37. Oh MY, et al. Long-term hardware-related complications of deep brain stimulation. Neurosurgery 2002;50 (6):1268-74; discussion 1274–6. 38. Patel NK, Plaha P, Gill SSMagnetic resonance imagingdirected method for functional neurosurgery using implantable guide tubes. Neurosurgery 2007;61 5 Suppl 2:358-65; discussion 365–6. 39. Plaha P, Gill SS. Bilateral deep brain stimulation of the pedunculopontine nucleus for Parkinson’s disease. Neuroreport 2005;16(17):1883-7. 40. Plaha P, Khan S, Gill SS. Bilateral stimulation of the caudal zona incerta nucleus for tremor control. J Neurol Neurosurg Psychiatry 2008;79(5):504-13. 41. Simon SL, et al. Error analysis of MRI and leksell stereotactic frame target localization in deep brain stimulation surgery. Stereotact Funct Neurosurg 2005;83(1):1-5. 42. Tuszynski MH, et al. A phase 1 clinical trial of nerve growth factor gene therapy for Alzheimer disease. Nat Med 2005;11(5):551-5. 43. Bloch J, et al. Neuroprotective gene therapy for Huntington’s disease, using polymer-encapsulated cells engineered to secrete human ciliary neurotrophic factor: results of a phase I study. Hum Gene Ther 2004;15(10):968-75. 44. Nutt JG, et al. Randomized, double-blind trial of glial cell line-derived neurotrophic factor (GDNF) in PD. Neurology 2003;60(1):69-73.
1253
45. Eriksdotter Jonhagen M, et al. Intracerebroventricular infusion of nerve growth factor in three patients with Alzheimer’s disease. Dement Geriatr Cogn Disord 1998;9(5):246-57. 46. Hoehn M, et al. Monitoring of implanted stem cell migration in vivo: a highly resolved in vivo magnetic resonance imaging investigation of experimental stroke in rat. Proc Natl Acad Sci USA 2002;99(25): 16267-72. 47. Sampson JH, et al. Clinical utility of a patient-specific algorithm for simulating intracerebral drug infusions. Neuro Oncol 2007;9(3):343-53. 48. Lonser RR, et al. Image-guided, direct convective delivery of glucocerebrosidase for neuronopathic Gaucher disease. Neurology 2007;68(4):254-61. 49. Krauze MT, et al. Real-time visualization and characterization of liposomal delivery into the monkey brain by magnetic resonance imaging. Brain Res Brain Res Protoc 2005;16(1–3):20-6. 50. Szerlip NJ, et al. Real-time imaging of convectionenhanced delivery of viruses and virus-sized particles. J Neurosurg 2007;107(3):560-7.
79 Impedance Recording in Functional Neurosurgery L. Zrinzo . M. I. Hariz
Introduction
General Principles
The stereotactic technique, described by Horsley and Clarke in 1908 and introduced to clinical practice by Spiegel and Wycis in 1947, is essential to anatomical localization in neurosurgery. The introduction of CT and especially MRI guided stereotactic surgery during the last decades of the twentieth century greatly facilitated the use and accuracy of anatomical targeting. Nevertheless, correlation of supposed anatomical location with expected clinical, physiological or electrical properties remains an important corroborative tool. Meyer first measured the electrical properties of brain tissue to localize pathology during surgery in 1921 [1]. Two years later, Grant reviewed tissue impedance patterns in 12 neurosurgical procedures and supplemented these observations with data from fresh and formalin fixed brains as well as animal trials. He concluded that impedance measurements were useful in locating deep seated tumors [2]. These techniques were particularly useful in tumor localization in the pre-CT era but persisted after the introduction of cross sectional imaging [3–7]. Continuous measurement of electrical impedance of brain tissue is readily measured en route to the target during stereotactic procedures and provides safe and reliable information during functional neurosurgical procedures. This has resulted in the widespread availability of commercial lesion generators to measure tissue impedance.
Electrical impedance is a measure of opposition to the flow of an alternating current and is measured in Ohms. Impedance includes resistance, but also takes into account the effects of capacitance and inductance (reactance). Impedance is a more complex measurement than resistance because the effects of capacitance and inductance vary with the frequency of the current passing through the circuit; impedance therefore varies with frequency. Brain tissue impedance is dependent upon numerous anatomical, chemical and physiological factors with the amount of myelin, direction of fibers, number of glial cells, density of neurons, composition of the CSF and blood flow all affecting the absolute values [8]. It is for this reason that the variation or ‘‘pattern’’ of electrical impedance as the probe travels through its trajectory within the brain is more useful than absolute static values [2,4,9,10]. Impedance is highest in white matter, lowest in CSF and intermediate in gray matter [4,9,11,12]. The direction of impedance change in passing between tissues is reproducible and provides supplementary information about the properties of the trajectory and target [4,11,13,14]. Monopolar impedance between a small probe tip and a large indifferent electrode is an old and excellent method to locate different structures inside the brain [15,16]. The indifferent electrode is usually a plate electrode applied to the thigh. The smaller the tip of the active
#
Springer-Verlag Berlin/Heidelberg 2009
1326
79
Impedance recording in functional neurosurgery
electrode in relation to the area of the indifferent electrode, the more accurately impedance changes reflect those in the vicinity of the probe tip [17]. However, safety factors place practical constraints on the size of the active electrode since the risk of hemorrhage increases as probe diameter falls below 0.5 mm [18]. Therefore, non-insulated tips of the order of 1–2 mm in length and diameter are recommended. The standard Elekta electrode (Elekta, Stockholm, Sweden) previously used by the authors measures 2.1 mm in diameter and the non-insulated tip measures 4 mm. Impedance values with that electrode are therefore lower than with the recommended 1.5 2 mm Elekta electrode, (custom made) all other parameters being equal. Nevertheless, this 2.1 mm thick electrode can still disclose differences between gray matter, white matter and CSF spaces, as shown in > Figure 79-1.
Laitinen initially described how a feeding current with a carrier frequency between 1 and 10 kHz resulted in the sharpest difference between gray and white matter impedance values [16]. Impedance measurements in animal and cadavers published shortly afterwards provided such detail that images could be constructed from the acquired data resulting in the term ‘‘impedography’’ [19–21]. A more recent study suggests that carrier frequencies of 8–10 kHz may even offer enough resolution to differentiate between different gray matter structures, a claim that deserves further study in clinical practice [11]. Despite the well documented importance of carrier frequency in the utility of impedance measurements, some commonly available radiofrequency lesion generators (Radionics Inc., Burlington, MA) use suboptimal feeding currents, compromising the utility of this technique in functional neurosurgical practice [22].
. Figure 79-1 Illustration of lesion generator (Elekta, Stockholm, Sweden) showing three different impedance values recorded with a 2.1 mm radiofrequency electrode: (a) gray matter impedance of 455 Ohms. (b) white matter impedance of 616 Ohm. (c) CSF impedance of 166 Ohms
Impedance recording in functional neurosurgery
It has been suggested that coaxial bipolar electrodes may provide a more accurate measurement of local tissue impedance [12,23]. However, others have noted that tissue or electrolyte bridges between the two poles may degrade the quality of the recordings [15]. Monopolar impedance recording is more popular for a number of reasons. Monopolar radiofrequency electrodes serve a dual purpose of impedance measurement and lesion production thus necessitating the passage of only one probe through the brain when performing lesional surgery. In addition, monopolar impedance recording provides information about structures that lie immediately ahead or in close proximity to the probe trajectory. For example, a reduction in monopolar impedance may indicate proximity of the trajectory to a ventricle that would not otherwise be acknowledged by a bipolar recording device [9].
Continual Impedance Monitoring in Functional Procedures Continual impedance monitoring allows the surgeon to assess the impedance pattern in ‘‘real time’’ on a panel meter and as an audio signal, as the stereotactic probe is introduced towards the target area. The acoustic signal allows the surgeon to judge when the probe is passing through gray matter, white matter or CSF boundaries. During functional procedures, the authors pass a custom designed electrode (Elekta, Stockholm, Sweden) with a non-insulated tip of 1.5 mm in diameter and 2 mm in length through a stereotactically defined trajectory and target under continual monopolar impedance monitoring using the Leksell1 Neuro Generator (Elekta, Stockholm, Sweden). With this arrangement, white matter impedance is typically in the range of 800–1,000 Ohms falling to 500–700 Ohms on crossing the border into gray matter with a value of around 300–400 Ohms in CSF. Due to the smaller size of
79
the non-insulated tip of the customized electrode, compared with the standard 2.1 4 mm Elekta electrode, the impedance values are higher and also more local-specific than with the larger electrode.
Pallidal Target When targeting the posteroventral pallidum from a frontal coronal burrhole, 2–2.5 cm from the midline, the probe travels roughly parallel to the sagittal plane traversing the white matter of the corona radiata/internal capsule into the gray matter of the putamen or pallidum depending on the angulation of the electrode trajectory. The fall in impedance is clearly noticed with an abrupt change over a distance of one mm between white matter tracts and the gray matter of the anterodorsal (rostral) putamen or pallidum. One can sometimes notice a slight rise in impedance as the electrode traverses the various medullary laminae or approaches the ansa lenticularis at the base of the pallidum before the probe reaches the cistern between the pallidum and the supra-amygdala where the impedance rapidly falls to 400 Ohms. Continual impedance monitoring would immediately alert the surgeon of a medial deviation in the planned trajectory as the pitch of the auditory signal rises to that typical of white matter with transgression into the internal capsule [22,24–27].
Thalamic Target During thalamic procedures, with a frontal approach, the probe penetrates the lateral part of the head of the caudate nucleus to give impedance values suggestive of gray matter only to rise again as the white matter of the internal capsule is traversed. Impedance then slowly falls as the thalamus is acquired. Once the probe approaches the area below the thalamus, impedance starts to
1327
1328
79
Impedance recording in functional neurosurgery
rise once more. The impedance profile is less sharp during ventrolateral thalamic surgery than during pallidal surgery due to the fact that the probe advances along the thalamocapsular border and reticular thalamic nucleus, during part of its trajectory before reaching the gray matter of the ventrolateral-ventral intermediate thalamus proper [16].
areas link the ventral striatum to the caudate, rendering a mixed impedance value that lies between that of gray and white matter.
Advantages of Impedance Monitoring Figure 79-1a shows the impedance values in a case of pallidal surgery where an electrode with a bare tip 2.1 mm thick and 4 mm long, ended up in capsular white matter. It was subsequently repositioned to end up in pallidal gray matter (> Figure 79-1b). The patient was awake during surgery, and intraoperative stimulation at first location (> Figure 79-1a) with higher impedance confirmed a capsular location: upon stimulating with 120 Hz the patient exhibited tonic cramp in the hand. Should the stereotactic probe traverse or come close to the ventricle, cistern or cystic cavity, a sharp decrease of impedance will be noted. > Figure 79-1c shows a CSF impedance value, with same electrode as in > Figure 79-1a,b, in a case where the electrode traverses a ventricular space. The expected transition from gray to white matter, as determined from preoperative trajectory images, can be corroborated with continual impedance measurements as the probe passes along the trajectory to the desired target. The surgeon is alerted to unexpected deviations or transgressions into CSF spaces without time being spent analyzing or interpreting the acquired data [25]. Reduced operative time minimizes the amount of CSF loss and resulting brain shift; this allows patients to better tolerate surgery under local anesthetic. Finally, the 1–2 mm diameter of the recording probes used tend to displace fibers and vessels away from the trajectory rather than transect them and this may contribute to lower rates of hemorrhage than in cases where microelectrode recording is used [28]. >
Subthalamic Nucleus as Target When targeting the subthalamic nucleus (STN) the impedance profile may vary en route to the target depending on the chosen trajectory. Usually, the trajectory incorporates variable amounts of internal capsule, and thalamus, before entering the dorsolateral zona incerta immediately before entering the STN. This gives a rise of impedance on exiting the thalamus into the immediate subthalamic area, followed by a decrease of impedance upon entering STN. Here, the impedance values are less distinctive than for pallidal surgery due to the fact that the trajectory immediately above the STN, in the dorsal zona incerta, consists of mixed white and gray matter.
Anterior Internal Capsule as Target During anterior capsulotomy for obsessive compulsive disorder, it is important that the electrode, and the subsequent radiofrequency lesions, remain within the confines of the anterior internal capsule between the head of the caudate nucleus and the putamen. Here, continual impedance monitoring should remain white matter impedance until the last 2 mm of the trajectory. At this most ventral point, it may fall slightly, since the most ventral target point is some 2 mm below the extended intercommissural point. At this level, scattered gray matter
Impedance recording in functional neurosurgery
Limitations of Impedance Monitoring As with most physiological and clinical methods, impedance can only provide corroboration of presumed anatomical localization. The existence in some patients of large perivascular (VirchowRobin) spaces in the putamen and posteroventral pallidum may confuse the interpretation of impedance monitoring. Also, a suboptimal current frequency of the radiofrequency machine being used will reduce the discriminative value of impedance in distinguishing between gray and white matter. In all cases, postoperative stereotactic imaging remains the gold standard for accurately documenting anatomical localization after a stereotactic intervention.
79
Conclusions Impedance measurements were first used in the 1920s as a navigational tool in the localization of deep-seated tumors. Functional neurosurgeons can employ continual impedance measurement during stereotactic targeting to reliably document transition from white to gray matter along the trajectory as well as transgression of CSF spaces. This technique provides a rapid, safe and useful corroboration of real-time electrode localization in the operating theatre. Impedance measurements of chronically implanted DBS hardware can assist in the identification of device-related problems as a cause of therapeutic failure.
References Measuring Impedance in Chronically Implanted Systems Currently available deep brain stimulation (DBS) pulse generators (from Medtronic, Inc.) provide impedance measurement as one of a number of tools that can help in identifying device-related problems as a cause of therapeutic failure of DBS. The physician programmer is used to measure impedance for each of the contacts of the quadripolar DBS electrode separately in monopolar mode. Typically, when postoperative edema has subsided, the measured impedance lies between 500 and 1,500 Ohms. For mono-channel devices (Itrel II, Soletra) standard settings of 1 V, 210 ms, and 30 Hz should be used for maximum accuracy in impedance measurement. The Itrel II neurostimulator cannot read impedance values above 2,000 Ohms whereas the dual-channel Kinetra neurostimulator reads values >4,000 Ohms. Values of less than 50 Ohms suggest a short circuit [29,30]. An impedance >2,000 Ohms suggests a broken cable, lead fracture or other connection problem although this has to be confirmed by radiological and other means [31].
1. Meyer AW. Methode zum Auffinden von Hirntumoren bei der Trepanation durch electrische Widerstandmessung. Zbl Chir 1921;18:1824-26. 2. Grant FC. Localization of brain tumors by determination of electrical resistance of the growth. JAMA 1923;81: 2169-71. 3. Bullard DE, Makachinas TT. Measurement of tissue impedence in conjunction with computed tomographyguided stereotaxic biopsies. J Neurol Neurosurg Psychiatry 1987;50:43-51. 4. Gorecki J, Dolan EJ, Tasker RR, Kucharczyk W. Correlation of CT and MR with impedance monitoring and histopathology in stereotactic biopsies. Can J Neurol Sci 1990;17:184-9. 5. Organ L, Tasker RR, Moody NF. Brain tumor localization using an electrical impedance technique. J Neurosurg 1968;28:35-44. 6. Bullard DE. Intraoperative impedance monitoring during CT-guided stereotactic biopsies. Stereotact Funct Neurosurg 1989;52:1-17. 7. Rajshekhar V. Continuous impedance monitoring during CT-guided stereotactic surgery: relative value in cystic and solid lesions. Br J Neurosurg 1992;6:439-44. 8. Birzis L, Aguilar JA, Tachibana S. Cerebral hemodynamics revealed by electrical impedance changes. Confin Neurol 1968;30:1-16. 9. Limonadi FM, Roberts DW, Darcey TM, Holtzheimer PE, III, Ip JT. Utilization of impedance measurements in pallidotomy using a monopolar electrode. Stereotact Funct Neurosurg 1999;72:3-21. 10. Spiegel EA. Methodological problems in stereoencephalotomy. Confin Neurol 1965;26:125-32.
1329
1330
79
Impedance recording in functional neurosurgery
11. Axer H, Stegelmeyer J, Graf von Keyserlingk D. Comparison of tissue impedance measurements with nerve fiber architecture in human telencephalon: value in identification of intact subcortical structures. J Neurosurg 1999;90:902-9. 12. Oran LW, Tasker RR, Moody NF. The impedance profile of the human brain as a localization technique in stereoencephalotomy. Confin Neurol 1967;29:192-6. 13. Tachibana S, Aguilar JA, Birzis L. Scanning the interior of living brain by impedography. J Appl Physiol 1970;28: 534-9. 14. Dierssen G, Marg E. The value of impedance measurements to aid in the localisation in stereotactic surgery. Confin Neurol 1965;26:407-10. 15. Laitinen LV, Johansson GG. Locating human cerebral structures by the impedance method. Confin Neurol 1967;29:197-201. 16. Laitinen L, Johansson GG, Sipponen P. Impedance and phase angle as a locating method in human stereotaxic surgery. J Neurosurg 1966;25:628-33. 17. Ragheb T, Riegle S, Geddes LA, Amin V. The impedance of a spherical monopolar electrode. Ann Biomed Eng 1992;20:617-27. 18. Robinson BW, Bryan JS, Rosvold HE. Locating brain structures. Extensions to the impedance method. Arch Neurol 1965;13:477-86. 19. Tachibana S. Cerebral impedography in cat, rabbit and cadaver. Electroencephalogr Clin Neurophysiol 1969;27:670. 20. Tachibana S. Impedography in tumor localization. Trans Am Neurol Assoc 1970;95:317-9. 21. Tachibana S. Impedance study of brain tissue changes after penetrating injury. Exp Neurol 1971;32:206-17. 22. Laitinen LV. Personal memories of the history of stereotactic neurosurgery. Neurosurgery 2004;55:1420-28; discussion 8-9.
23. Hobza V, Jakubec J, Nemeckova J, Nemecek S, Sercl M. Impedance monitoring in the stereotactic localization of intracranial structures. Sb Ved Pr Lek Fak Karlovy Univerzity Hradci Kralove 1995;38:33-46. 24. Laitinen LV. Pallidotomy for Parkinson’s disease. Neurosurg Clin N Am 1995;6:105-12. 25. Heilbrun MP, Koehler S, McDonald P, Faour F. Optimal target localization for ventroposterolateral pallidotomy: the role of imaging, impedance measurement, macrostimulation and microelectrode recording. Stereotact Funct Neurosurg 1997;69(1-4)(Pt 2):19-27. 26. Iacono RP, Carlson JD, Kuniyoshi SM, Li YJ, Mohamed AS, Maeda G. Electrophysiologic target localization in posteroventral pallidotomy. Acta Neurochir (Wien) 1997;139:433-41. 27. Siemionow V, Yue GH, Barnett GH, Sahgal V, Heilbrun MP. Measurement of tissue electrical impedance confirms stereotactically localized internal segment of the globus pallidus during surgery. J Neurosci Methods 2000;96:113-17. 28. Hariz MI. Safety and risk of microelectrode recording in surgery for movement disorders. Stereotact Funct Neurosurg 2002;78:146-57. 29. Volkmann J, Herzog J, Kopper F, Deuschl G. Introduction to the programming of deep brain stimulators. Mov Disord 2002;17 Suppl 3:S181-7. 30. Butson CR, Maks CB, McIntyre CC. Sources and effects of electrode impedance during deep brain stimulation. Clin Neurophysiol 2006;117(2):447-54. 31. Joint C, Nandi D, Parkin S, Gregory R, Aziz T. Hardware-related problems of deep brain stimulation. Mov Disord 2002;17 Suppl 3:S175-80.
81
Lesions Versus Implanted Stimulators in Functional Neurosurgery W. S. Anderson . R. E. Clatterbuck . K. Kobayashi . J.-H. Kim . F. A. Lenz
Introduction Surgery for movement disorders began with cortical resections by Victor Horsely for chorea in 1906 [1]. In the late 1930s, Meyers pioneered a series of transventricular procedures in the basal ganglia for Parkinson’s disease [2]. The next advance in surgery for movement disorders came with the application of stereotactic techniques to lesioning of structures in the basal ganglia [3]. Lesion targets were refined in the 1950s and 1960s leading to the practice of pallidotomy for the treatment of rigidity and akinesia [4] and thalamotomy for tremor [5]. However, surgery for the treatment of movement disorders was dramatically curtailed with the development of levodopa as pharmacologic therapy for Parkinson’s disease. With the long-term treatment of patients with levodopa, the complications of levodopa became evident leading to renewed interest in the surgical options for treatment of movement disorders [6]. Leksell’s posteroventral pallidotomy was re-established as an effective treatment option for rigidity and bradykinesia in patients with Parkinson’s disease [7]. Stimulation of lesioning targets had long been known to be capable of alleviating symptoms in the operating room. In the early part of this decade, the development of improved technologies for implantable stimulating electrodes made deep brain stimulation (DBS) a viable option to surgical lesioning. We review here several of the important studies demonstrating efficacy of lesioning in the #
Springer-Verlag Berlin/Heidelberg 2009
treatment of movement disorders from the last century. The discussion will primarily be limited to posteroventral pallidotomy and ventral intermediate (Vim) thalamotomy as the most common lesioning operations with a brief discussion of radiosurgical treatments, and lesioning for psychiatric diseases. These procedures will be compared with the three most common DBS procedures: high frequency stimulation (hfs) of the globus pallidus internus (Gpi), Vim, and subthalamic nucleus (STN). The potential complications and benefits of lesioning and stimulation will be discussed and compared.
Posteroventral Pallidotomy and GPi Stimulation Posteroventral pallidotomy for Parkinson’s disease was re-introduced by Laitinen [7]. In his study, 38 patients with Parkinson’s disease and a primary complaint of akinesia underwent stereotactic posteroventral pallidotomy and mean follow up of 28 months. Motor function was assessed using writing, drawing, and gait tests. Near complete relief of rigidity and akinesia were reported in 92% of patients. Eighty-one percent of patients with tremor experienced dramatic improvement. Levodopa induced dyskinesias were also improved. Six of these patients, however, suffered a permanent central homonymous visual field defect. A series of well-designed studies of posteroventral pallidotomy [8–12] have followed the
1350
81
Lesions versus implanted stimulators in functional neurosurgery
lead of Laitinen et al. Several incorporated evaluation protocols in which observers reviewed patients’ exams by videotape and were blinded to the treatment patients had received [8,9,11]. These studies all employed Hoehn and Yahr patient staging [13] and assessment according to the Core Assessment Program for Intracerebral Transplantation (CAPIT) [14] which incorporates the Unified Parkinson’s Disease Rating Scale (UPDRS) [15]. No patient in these studies was better than a Hoehn and Yahr stage III in the on-state. Follow-up intervals from 3 months to 1 year demonstrated a range of improvement in the off-state in the UPDRS motor subscale score from 14 to 70%. These studies demonstrated marked contralateral improvements in rigidity, akinesia, tremor, gait, balance, levodopa related dyskinesias, and on-off fluctuations. These studies documented some ipsilateral improvements as well. The first of these studies reported an off-state improvement in the UPDRS motor score of 71% at 1 year post-operatively [8]. In the study of Lozano et al. [9] UPDRS total motor score in the off-state improved by 30% at 6 months, while the total akinesia score improved by 33%. This study also noted a 15% improvement in the gait score in the off-state and a 92% reduction in contralateral dyskinesias. In a study by Baron et al. [10], a 25% improvement in the UPDRS motor score in the off-state at 3 months was seen. Shannon et al. [12] noted a 15% improvement in the off-state mean motor UPDRS score at 6 months; and a contralateral off-state combined tremor, rigidity, and bradykinesia score improved by 26%. In this study the dyskinesia severity score was improved by 73% at 6 months. In the study of Ondo et al. [11], total off-state UPDRS motor scores improved 14%. Improvement in UPDRS total tremor subscore improved by 59%, gait scores by 22%, and body bradykinesia by 17%. Lang et al. have published a 2-year follow-up study on 11 patients after posteroventral pallidotomy [16]. The initial results in their group of
40 patients were similar to the above studies at 6 months with overall improvement in motor function of 28%. The effect of pallidotomy on contralateral bradykinesia, dyskinesia, and rigidity were maintained at 2 years, while ipsilateral effects were generally lost. Siegfried and others began work with chronic implantable central nervous system stimulation in the early 1980s. Much of the early clinical work with stimulation was done for intractable pain and for tremor. Siegfried and Lippitz, inspired by Laitinen’s work in the posteroventral pallidum, first reported bilateral GPi-hfs in three patients in 1994 [17]. All three patients with advanced Parkinson’s disease (Hoehn and Yahr stage IV or worse) experienced dramatic decreases in on-off phenomena and levodopa induced dyskinesias. Effects were reversed, though not immediately, when stimulation was discontinued. Several other groups have demonstrated marked improvement in patients treated with unilateral and bilateral GPi-hfs. Pahwa et al. reported the treatment of five patients with Parkinson’s disease with GPi-hfs, three with bilateral implants [18]. All patients had disabling symptoms with Hoehn and Yahr stage III disease or worse. At 3-month follow-up the amount of time in the on-state increased from 21 to 65%. UPDRS motor scores in the off-state without stimulation improved 24% at 3 months post-operatively compared to the preoperative off-state. UPDRS motor scores at 3 months post-operatively in the onstate with stimulation were improved 60% over the preoperative on-state. In the off-state at 3 months follow-up, turning on stimulation improved UPDRS scores 21%. In a similar series, Gross et al. reported seven patients with Hoehn and Yahr stage III-IV disease who underwent placement of unilateral GPi-hfs electrodes [19]. Mean improvement over off-state UPDRS motor scores (post-procedure) were 33% with levodopa, 35% with stimulation, and 63% with both. Off-state scores pre- and post-procedure were similar. In addition, four of five patients with
Lesions versus implanted stimulators in functional neurosurgery
tremor experienced considerable improvement. No patients experienced ipsilateral effects. These therapeutic benefits were purchased at a cost. Among the fifteen patients undergoing pallidotomy reported by Baron et al. [10], two (13%) suffered subclinical frontal hemorrhages, one (7%) suffered transient dysarthria, one (7%) suffered persistent worsening of a baseline dysarthria, and one (7%) had a persistent superior quadrantanopsia. Several patients in this study also experienced transient confusion and several experienced transient facial weakness. Ondo et al. [11] reported that among 34 patients undergoing pallidotomy, five (15%) experienced transient side effects which included aphasia (3%) and altered mental status (12%). Shannon et al. [12] reported 26 patients who underwent pallidotomy. One patient (4%) had a fatal hemorrhage and three (12%) had nonfatal hemorrhages. In addition two patients (8%) had cognitive changes, one patient (4%) developed aphasia, three patients (12%) experienced persistent frontal lobe dysfunction, one patient (4%) developed a mild but persistent hemiparesis, and one patient (4%) experienced persistent increase in dysarthria. This study also reported some transient side effects including altered mental status, facial weakness, and dysarthria. Dogali et al. reported no significant complications related to pallidotomy in eighteen patients [8]. Kondziolka et al. reported a 5% rate of transient dysarthria lasting 1–3 weeks in a series of 120 pallidotomies, with no hemorrhages encountered [20]. Additionally, Iacono et al. reported in a series of 126 patients undergoing either unilateral (58) or bilateral (68) pallidotomies, a hemorrhage rate of 3.2% per lesion [21]. De Bie et al. provided a very thorough summary of pallidotomy complications recorded from 1992 to 2000 [22]. These authors found in 334 cases of published unilateral pallidotomies in prospective studies, that 13.8% suffered permanent adverse effects (which were most commonly changes in personality or behavior, dysarthria,
81
dysphagia, and visual field defects). A symptomatic infarction or hemorrhage occurred in 3.9%, and the mortality rate was 1.2%. There was also a group of recorded patients who had undergone bilateral pallidotomy (five cohort studies with 20 total patients). Fourteen suffered adverse effects which included effects on speech and cognition. Interestingly, the patients undergoing microelectrode recording were separately analyzed, and demonstrated an increased rate of adverse effects (14.4% higher) with a frequency of infarct that was also 4.9% higher [20]. In the five patients included in the GPi-hfs studies of Pahwa et al. [18], the only complications reported included a single asymptomatic intracranial hemorrhage, a transient speech difficulty and hemiparesis related to stimulation which resolved in the operating room, and a facial dystonia and paresthesia which required electrode repositioning. Adverse side effects of stimulation included one patient with a visual disturbance (transient) and one patient with stimulation related chorea of a foot. Gross et al. [19] reported seven patients undergoing pallidal stimulation procedures without any complications.
Ventral Thalamotomy and Vim Stimulation One of the first large post-levodopa series of stereotactic thalamotomies for Parkinson’s disease with medically intractable tremor was completed at the Mayo clinic [23]. In this study, 36 patients (mean Hoehn and Yahr stage 2.4) were treated with 37 thalamotomies and 31 (86%) experienced complete relief of their tremor. Another three (5%) were significantly improved. During the follow-up period, which ranged from 14 to 68 months, only two patients suffered from recurrent tremor (both within 3 months). Diederich et al. [24] blindly compared tremor on the operated and unoperated side in 17 patients with Parkinson’s disease at a mean of 10.9 years
1351
1352
81
Lesions versus implanted stimulators in functional neurosurgery
following stereotactic thalamotomy/subthalamotomy using videotaped examinations. At follow-up the mean Hoehn and Yahr stage was 1.8 and UPDRS motor subscore was 17.8. Severity scores for upper extremity tremor were significantly better on the side contralateral to surgery than on the ipsilateral side. In all these patients, the surgical side was initially chosen to treat the side with the more severe tremor. The largest recent series [25] retrospectively reviewed the outcomes in 60 patients with parkinsonian tremor (42 patients), essential tremor (6), cerebellar tremor (6), and post-traumatic tremor (6). These patients all underwent unilateral stereotactic Vim thalamotomy with the exception of two Parkinson’s patients who underwent bilateral procedures, and one patient who underwent lesioning more anteriorly. With a mean follow-up of 53.4 months, Parkinson’s patients experienced moderate to marked improvement in 86% of the cases. Patients with essential tremor showed similar improvement in 83% of cases. Results were not as dramatic for those patients with cerebellar tremor (67%) or post-traumatic tremor (50%). Our own experience with stereotactic Vim thalamotomy for essential tremor provides confirmation of these reports with a blinded measure of pre- and postoperative status [26]. Patients were evaluated preoperatively and at 3 and 12 months postoperatively with a functional disability score and a blinded handwriting/drawing score. Significant improvements in both scores were found postoperatively. Within patients analysis demonstrated statistically significant improvement in 72% of patients. With the advent of chronic DBS in the central nervous system for movement disorders, stimulation of the Vim nucleus has been examined as a treatment for tremor. The largest study [27] included 117 patients (177 operated sides) with movement disorders, including 80 with Parkinson’s disease and 20 with essential tremor. Bilateral implantation was undertaken in 59
patients and 14 patients underwent implantation contralateral to a previous thalamotomy. The follow-up period was as long as 7 years for the earliest procedures. At last follow-up 88% of Parkinson’s patients had complete or near complete relief of tremor, and another 10% had slight to moderate improvement. Global scores including all four limbs were slightly lower. Rigidity and akinesia were not significantly affected. In this series the effect on essential tremor was less dramatic with only 61% of patients experiencing complete or near complete relief of tremor at last follow-up. The effect of Vim DBS was inconsistent for patients with other dyskinesias or tremors. Other studies have reported similar results. Koller et al. [28] studied unilateral Vim DBS in 24 patients with Parkinson’s disease and 29 patients with essential tremor in a multicenter trial using a blinded 3 month postoperative evaluation and an open label 1-year follow-up. Complete resolution of tremor was seen in 31% of essential tremor patients and 58.3% of Parkinson’s patients. Only 3.4% of essential tremor patients and only 4.2% of Parkinson’s disease patients had no change in their tremor. No ipsilateral effects were detected. One direct comparison of thalamotomy and thalamic DBS for tremor found essentially equal efficacy in abolishing tremor (63.7% vs. 62.5%), however, 27% of thalamotomies had to be repeated for tremor recurrence while none of the DBS procedures needed revision [29]. Transient complications of thalamotomy were seen in 58–70% of patients and included contralateral weakness, dysarthria, dysphasia, confusion, dystonia, and sensory disturbances [23,25]. One patient in these two series died at 7 days postoperatively from a pulmonary embolism. The number of patients with long term or permanent complications was much smaller, in the range of 14–23%. Among these complications weakness/dyspraxia and dysarthria figured most prominently. Only one patient in these series experienced permanent cognitive difficulties
Lesions versus implanted stimulators in functional neurosurgery
after a unilateral procedure. Favre et al. reported one hematoma (out of 137 lesioning procedures) in the context of a thalamotomy in a case where the systolic blood pressure had remained elevated at 160 mm Hg [30]. In one series of Vim DBS [28], six of 53 procedures were aborted secondary to intraoperative complications including failure of stimulation to suppress tremor (2 patients), intracranial hemorrhage (1), and subdural hemorrhage (1). One patient in this series experienced a postoperative seizure. Transient side effects of stimulation included paresthesias, experienced by most patients, and gait disorders, that were much less common. The long-term complications seen in this series in the first year included two superficial wound infections, one extension wire erosion, and one failure of the implantable pulse generator. In the large series of Benabid et al. [27], 5.1% of patients experienced small hematomas as a result of electrode passage (half of these were transiently symptomatic), and 31.6% of patients experienced minor side effects that were reversible with discontinuation of stimulation. These included paresthesias (9%), foot dystonia (9%), dysequilibrium (9%), and contralateral dystonia (5%). Dysarthria was seen in 23 patients (19.6%). Interestingly, 14 of these patients were receiving bilateral stimulation and four had a contralateral thalamotomy. Three secondary scalp infections (3%) leading to hardware removal were observed.
Bilateral Subthalamic Nucleus Stimulation Two prospective studies from the 1990s have been published discussing subthalamic nucleus (STN) stimulation for the treatment of Parkinson’s disease symptoms [31]. The evolution of DBS has made STN procedures feasible. Kumar et al. carried out a double-blind evaluation of seven patients with end-stage Parkinson’s disease that underwent bilateral STN-hfs. Patients
81
experienced an improvement in mean UPDRS motor scores post-operatively of 58% in the off-state with stimulation. In the on-state with stimulation, patients’ mean UPDRS motor score improved by 41% compared to the preoperative on-state. Time in the off-state and dyskinesias were both decreased postoperatively. Two patients experienced transient hemichorea as a result of the procedure, and four patients experienced some degree of postoperative cognitive difficulty (one experienced a venous infarction and another a thalamic lesion). Limousin et al. [32] reported 24 patients with Parkinson’s disease (Hoehn and Yahr stage IV-V in the off-state) that underwent bilateral STN-hfs. At 1-year follow-up, mean UPDRS motor scores improved 60% with stimulation in the off-state and 10% in the on-state. As in the Kumar series, patients were significantly improved in the off-state with the stimulator off. Rigidity, akinesia, tremor, gait, and dyskinesias were all improved. Complications included a large intracerebral hematoma (4%) leading to paralysis and aphasia and a subcutaneous infection (4%) requiring hardware removal. Eight patients experienced transient cognitive difficulties, and one patient had permanent worsening of preoperative cognitive deficits. Five patients experienced an eyelid apraxia requiring treatment. In two larger series of DBS for chronic pain [33,34], among a total of 263 patients, ten intracerebral hemorrhages (4%) (including three deaths), 23 infections (9%), ten hardware erosions (4%), and seven foreign body reactions (3%) were seen.
Other Lesioning Procedures Stereotactic radiosurgery for functional procedures is not as widely used as stimulation or traditional thermocoagulation, but is practiced as a form of lesioning by some groups [35–37]. It is particularly useful for patients not tolerating
1353
1354
81
Lesions versus implanted stimulators in functional neurosurgery
implanted hardware, and demonstrates efficacy rates similar to the previous thermocoagulation studies. For instance, Kondziolka et al. found that 69% of 31 patients treated with Gamma Knife thalamotomy for essential tremor demonstrated improvements in action tremor and writing [35]. Complications of radiosurgery for functional procedures have also been reported, and include dysphagia, hemiplegia, visual field deficits, dysarthria, and even pseudobulbar laughter in a set of patients in which poor targeting was to blame [38]. In their series of 31 patients undergoing Gamma Knife thalamotomy for essential tremor, Kondziolka et al. reported one patient who developed a permanent mild hemiparesis and dysarthria 7 months after treatment [35]. Additionally, radiosurgical treatment does not allow for the use of microelectrode mapping to further refine the target location, which might make for higher morbidity rates [39]. Lesioning procedures performed for the treatment of psychiatric disease or pain will also have similar morbidities found in lesioning for movement disorders, including hemorrhage and infection. Seizures as a complication are probably represented more in series of cingulotomies. Wilkinson et al. describe a series of 23 patients undergoing bilateral cingulotomy for chronic pain [40]. Two of these patients suffered intraoperative seizures, and five had late seizures. In the large series of Ballantine et al. involving 714 cingulotomies in 414 patients, there were no deaths reported and no infections [41]. Two of the patients suffered acute subdural hematomas resulting in hemiplegia, and one patient developed a chronic subdural collection. Five of the patients suffered chronic postoperative seizures controlled with phenytoin. Bilateral lesioning of the anterior cingulate can produce cognitive deficits as well. Ochsner et al. reported on deficits picked up in visual cognitive inventories after cingulotomy [38]. These patients had difficulties with sequencing cognitive operations needed to generate complex
or moving images or in rotating images. These deficits are most likely related to the executive system function which the cingulate gyrus participates in [42]. Similar findings were present in a group of chronic pain patients having undergone bilateral anterior cingulotomy [42]. This group of patients, when compared to untreated chronic pain patients, showed deficits in attention and executive function. Self-initiated behaviors as well as behavioral spontaneity were the most affected [42]. This same group of authors also described changes in emotional experience after bilateral anterior cingulotomy [43]. The greatest effects were seen on inventories that correlate most with levels of emotional tension and agitation. This agrees with the clinical effects seen in patients with anxiety disorders. In the recent series of cingulate stimulation for OCD, no conclusive cognitive disturbances were noted [44,45], although this will be an interesting comparison as these cohorts grow larger.
Conclusions In the last decade many new options have become available for the surgical treatment of movement disorders. The principal procedures currently in use include posteroventral pallidotomy, ventrolateral thalamotomy, and DBS in Vim, GPi and STN. With the development of technology allowing routine use of indwelling stimulating devices in the central nervous system, DBS has become a more viable option. Although the collective experience with DBS is considerably smaller (but ever growing) than that with lesioning procedures, present data suggests that stimulation is as effective as lesioning procedures. Posteroventral pallidotomy series demonstrate 14–70% improvement in off-state UPDRS motor scores following surgery. In patients treated with pallidal stimulation, turning on the stimulator in the off-state improved UPDRS motor scores by 21–35%. Similarly, most
Lesions versus implanted stimulators in functional neurosurgery
series place significant improvement in tremor following ventrolateral thalamotomy and Vim stimulation at over 80%. One series that directly compared the two modalities found both effective at around 63% [29]. These studies all suggest that lesioning and stimulation are equally efficacious in the treatment of movement disorders. DBS carries the obvious advantage of being reversible and adjustable. These advantages make bilateral procedures more practical and safe. It remains to be seen if these theoretical advantages will be born out in larger series or randomized trials. Given these advantages however, there are situations in which lesioning is warranted. A lesioning procedure would be more appropriate for people who do not want an implantable device or did not tolerate it for a variety of wound healing or infection issues. Also, lesioning might be more appropriate in patients with no possibility of follow-up for generator programming or replacement, for instance patients living in a developing nation. In general though, stimulation should be used in all patients whenever possible. It would appear from the data presented here that clinically significant hemorrhage risks are similar for permanent lesioning paradigms and stimulator lead placements, about 1.7–4% [30]. The obvious disadvantages of permanent indwelling hardware, increased infection risk (3–9%) and mechanical failure, are born out to some degree with the present series. Other neurologic complications seen with stimulation such as paresthesias and dystonias generally resolved when stimulation was decreased. In comparison, complications such as aphasia, cognitive worsening, mild hemiparesis, and visual field deficits seen with lesioning were in some cases permanent. Such complications were seen in as high as 14–23% of patients. Another aspect to be considered is the increased costs associated with implanting and maintaining DBS. It will take a multicenter, prospective, randomized trial with outcome measures including a cost-benefit analysis to address all these issues.
81
References 1. Horsley V. The function of the so-called motor cortex. Br Med J 1909;2:125-32. 2. Meyers R. Surgical procedure for postencephalitic tremor, with notes on the physiology of the premotor fibers. Arch Neurol Psych 1940;44:455-9. 3. Spiegel EA, Wycis HT, Marks M, Lee AJ. Stereotaxic apparatus for operations on the human brain. Science 1947;106:349-50. 4. Svennilson E, Torvik A, Lowe R, Leksell L. Treatment of parkinsonism by stereotactic thermolesions in the pallidal region. A clinical evaluation of 81 cases. Acta Psych Scand 1960;35:358-77. 5. Hassler R, Riechert T. Indikationen und Lokalisationsmethode der gezielten Hirnoperationen. Nervenarzt 1954;25:441-7. 6. Marsden CD, Parkes JD. Success and problems of longterm levodopa therapy in Parkinson’s disease. Lancet 1977;1:345-9. 7. Laitinen LV, Bergenheim AT, Hariz MI. Leksell’s posteroventral pallidotomy in the treatment of Parkinson’s disease. J Neurosurg 1992;76:53-61. 8. Dogali M, Fazzini E, Kolodny E, Eidelbert D, Sterio D, Devinsky O, Beric A. Stereotactic ventral pallidotomy for Parkinson’s disease. Neurol 1995;45:753-61. 9. Lozano AM, Lang AE, Galvez-Jimenez N, Miyasaki J, Duff J, Hutchison WD, Dostrovsky JO. GPi pallidotomy improves motor function in patients with Parkinson’s disease. Lancet 1995;346:1383-6. 10. Baron MS, Vitek JL, Bakay RAE, Green J, Kaneoke Y, Hashimoto R, Turner RS, Woodward JL, Cole SA, McDonald WM, DeLong MR. Treatment of advanced Parkinson’s disease with microelectrode-guided pallidotomy: 1 year pilot study results. Ann Neurol 1996;40:355-66. 11. Ondo WG, Jankovic J, Lai EC, Sankhla C, Khan M, Ben-Arie L, Schwartz K, Grossman RG, Krauss JK. Assessment of motor function after stereotactic pallidotomy. Neurology 1998;50:266-70. 12. Shannon KM, Penn MD, Kroin JS, Janko KA, York M, Cox SJ. Stereotactic pallidotomy for the treatment of Parkinson’s disease. Neurol 1998;50:434-8. 13. Hoehn MM, Yahr MD. Parkinsonism: onset, progression, and mortality. Neurol 1967;17:427-42. 14. Langston JW, Winder H, Goetz CG, Brooks D, Fahn S, Freeman T, Watts R. Core Assessment Program for Intracerebral Transplantations (CAPIT). Mov Disord 1992;7:2-13. 15. Fahn S, Elton RL, members of the UPDRS Development Committee. Unified Parkinson’s disease rating scale. In: Fahn S, Marsden CD, Calne DB, Goldstein M, editors. Recent developments in Parkinson’s disease, vol. 2. Florham Park: MacMillan Health Care Information; 1987. p. 153-64.
1355
1356
81
Lesions versus implanted stimulators in functional neurosurgery
16. Lang AE, Lozano AM, Montgomery E, Duff J, Tasker R, Hutchinson W. Posteroventral medial pallidotomy in advanced Parkinson’s disease. N Eng J Med 1998;337:262-3. 17. Siegfried J, Lippitz B. Bilateral chronic electrostimulation of the ventroposterolateral pallidum: a new therapeutic approach for alleviating all parkinsonian symptoms. Neurosurg 1994;35:1126-30. 18. Pahwa R, Wilkinson S, Smith D, Lyons K, Miyawaki E, Koller WC. High-frequency stimulation of the globus pallidus for the treatment of Parkinson’s disease. Neurol 1997;49:249-53. 19. Gross C, Rougier A, Guehl D, Boraud T, Julien J, Bioulac B. High-frequency stimulation of the globus pallidus internalis in Parkinson’s disease: a study of seven cases. J Neurosurg 1997;87:491-8. 20. Kondziolka D, Firlik AD, Lunsford LD. Complications of stereotactic brain surgery. Neurol Clin North Am 1998;16(1):35-54. 21. Iacono RP, Shima F, Lonser RR, Kuniyoshi S, Maeda G, Yamada S. The results, indications, and physiology of posteroventral pallidotomy for patients with Parkinson’s disease. Neurosurg 1995;36:1118-25. 22. de Bie RMA, de Haan RJ, Schuurman PR, Esselink RAJ, Bosch DA, Speelman JD. Morbidity and mortality following pallidotomy in Parkinson’s disease. Neurol 2002;58:1008-12. 23. Fox MW, Ahlskog EJ, Kelly PJ. Stereotactic ventrolateralis thalamotomy for medically refactory tremor in postlevodopa era Parkinson’s disease patients. J Neurosurg 1991;75:723-30. 24. Diederich N, Goetz CG, Stebbins GT, Klawans HL, Nitter K, Koulosakis A, Sanker P, Sturm V. Blinded evaluation confirms long-term assymmetric effect of unilateral thalamotomy or subthalamotomy on tremor in Parkinson’s disease. Neurol 1992;42:1311-14. 25. Jankovic J, Cardoso F, Grossman RG, Hamilton WJ. Outcome after stereotactic thalamotomy for parkinsonian, essential and other types of tremor. Neurosurg 1995;37:680-7. 26. Zirh AT, Reich SG, Dougherty PM, Lenz FA. Stereotactic thalamotomy in the treatment of essential tremor of the upper extremity: re-assessment including a blinded measure of outcome. J Neurol Neurosurg Psych 1999;66(6):772-5. 27. Benabid AL, Pollak P, Gao D, Hoffman D, Limousin P, Gay E, Payen I, Benazzouz A. Chronic electrical stimulation of the ventralis intermedius nucleus of the thalamus as treatment of movement disorders. J Neurosurg 1996;84:203-14. 28. Koller W, Pahwa R, Busenbark K, Hubble J, Wilkinson S, Lang AE, Tuite P, Sime E, Lozano A, Hauser R, Malapira T, Smith D, Tarsy D, Myawaki E, Norregaard T, Kormos T, Olanow CW. High frequency unilateral thalamic stimulation in the treatment of essential and parkinsonian tremor. Ann Neurol 1997;42:292-9.
29. Tasker RR, Munz M, Junn FSCK, Kiss ZHT, Davis K, Dostovsky JO, Lozano AM. Deep brain stimulation and thalamotomy for tremor compared. Acta Neurochir Suppl 1997;68:49-53. 30. Favre J, Taha JM, Burchiel KJ. An analysis of the respective risks of hematoma formation in 361 consecutive morphological and functional stereotactic procedures. Neurosurg 2002;50:48-56; discussion 56-7. 31. Kumar R, Lozano AM, Kim YJ, Hutchinson WD, Sime E, Halket E, Lang AE. Double blind evaluation of subthalamic nucleus deep brain stimulation in advanced Parkinson’s disease. Neurol 1998;51:850-5. 32. Limousin P, Krack P, Pollak P, Benazzouz A, Ardouin C, Hoffmann D, Benabid AL. Electrical stimulation of the subthalamic nucleus in advanced Parkinson’s disease. N Engl J Med 1998;339:1105-11. 33. Hosobuchi Y. Subcortical electrical stimulation for control of intractable pain in humans. J Neurosurg 1986;64:543-53. 34. Levy RM, Lamb S, Adams JE. Treatment of chronic pain by deep brain stimulation: Long term follow-up and review of the literature. Neurosurg 1987;21:885-93. 35. Kondziolka D, Ong JG, Lee JY, Moore RY, Flickinger JC, Lunsford LD. Gamma knife thalamotomy for essential tremor. J Neurosurg 2008;108:111-17. 36. Okun MS, Stover NP, Subramanian T, Gearing M, Wainer BH, Holder CA, Watts RL, Juncos JL, Freeman A, Evatt ML, Schuele SU, Vitek JL, DeLong MR. Complications of gamma knife surgery for Parkinson disease. Arch Neurol 2001;58:1995-2002. 37. Friedman DP, Goldman HW, Flanders AE, Gollomp SM, Curran WJ. Stereotactic radiosurgical pallidotomy and thalamotomy with the gamma knife: MR imaging findings with clinical correlation – Preliminary experience. Radiology 1999;212:143-50. 38. Ochsner KN, Kosslyn SM, Cosgrove CR, Cassem EH, Price BH, Nierenberg AA, Rauch SL. Deficits in visual cognition and attention following bilateral anterior cingulotomy. Neuropsych 2001;39:219-30. 39. Jankovic J. Editorial: surgery for Parkinson disease and other movement disorders. Arch Neurol 2001;58:1970-2. 40. Wilkinson HA, Davidson KM, Davidson RI. Bilateral anterior cingulotomy for chronic noncancer pain. Neurosurg 1999;45:1129-34. 41. Ballantine HT, Giriunas IE. Treatment of intractable psychiatric illness and chronic pain by stereotactic cingulotomy. In: Schmidek HH, Sweet WH, editors. Operative neurosurgical techniques. Indications, methods, and results. Philadelphia, PA: W.B. Saunders; 1988. p. 1069-75. 42. Cohen RA, Kaplan RF, Zuffante P, Moser DJ, Jenkins MA, Salloway S, Wilkinson H. Alteration of intention and self-initiated action associated with bilateral anterior cingulotomy. J Neuropsych Clin Neurosci 1999; 11:444-53.
Lesions versus implanted stimulators in functional neurosurgery
43. Cohen RA, Paul R, Zawacki TM, Moser DJ, Sweet L, Wilkinson H. Emotional and personality changes following cingulotomy. Emotion 2001;1:38-50. 44. Greenberg BD, Malone DA, Friehs GM, Rezai AR, Kubu CS, Malloy PF, Salloway SP, Okun MS, Goodman WK, Rasmussen SA. Three-year outcomes in deep brain stimulation for highly resistant obsessive-compulsive disorder. Neuropsychopharm 2006;31:2384-93.
81
45. Nuttin BJ, Gabrie¨ls LA, Cosyns PR, Meyerson BA, Andre´ewitch S, Sunaert SG, Maes AF, Dupont PJ, Gybels JM, Gielen F, Demeulemeester HG. Long-term electrical capsular stimulation in patients with obsessivecompulsive disorder. Neurosurg 2003;52:1263-72.
1357
78 Microelectrode Recording in Functional Neurosurgery W. D. Hutchison . J. O. Dostrovsky . M. Hodaie . K. D. Davis . A. M. Lozano . R. R. Tasker
Introduction Whenever a neurosurgical procedure involves a target or endangers an important neighboring structure that cannot be seen or distinctly imaged intraoperatively, some form of invasive physiological localization is required to assure accuracy and safety. Sometimes a whole structure cannot be visualized clearly (such as STN); on other occasions, the gross structure can be imaged but its important functional subdivisions cannot (such as the case with motor thalamus), even with the highest-quality magnetic resonance imaging (MRI). In both circumstances, the structure must be penetrated and its identity must be established by functional means. The first part of this chapter is concerned with invasive physiological localization of deep brain structures with microor macro-electrodes prior to surgery. The second part of this chapter is devoted to imaging techniques and their use in cortical and subcortical localization; localization relevant to the superficial cerebral cortex regions and to surgery for epilepsy is discussed in the section dealing with epilepsy (see chapters 153 and 157). There is still a need for invasive physiological localization despite any shortcomings since functional stereotactic imaging still does not allow accurate visualization of all stereotactic targets that the surgeon can manipulate. The current consensus is that imaging techniques are still not sufficiently accurate to achieve the best results in most functional stereotactic procedures. If every brain were identical, it would be expected that any target structure would bear a fixed three dimensional relationship in space to brain landmark structures #
Springer-Verlag Berlin/Heidelberg 2009
such as the anterior and posterior commissures (AC and PC). The fact that this is not so is well known from cortical mapping. For example, a given site on the postcentral gyrus a fixed distance from the midsagittal line may in one patient represent the face, in another the arm, and in yet another the leg [1]. However, it is clear to those who routinely perform subcortical mapping, that there is ongoing variation in initial or image-based targeting due to errors from various sources as revealed by the selection of the final target taking into account the mapping results.
Part 1 – Invasive Physiological Localization Methods of Subcortical Physiological Localization Table 78-1 lists some of the available techniques for invasive physiological localization. The most important are microelectrode recording, and electrical stimulation using micro- and macroelectrodes, which will be covered in some detail first. Electroencephalography (EEG), electrocorticography (ECoG), are used mainly during epilepsy surgery and will not be further described here (see chapters 153 and 157) but results from ‘‘deep EEG’’ studies were at one time used for functional localization of deep brain structures [2–6]. The deep EEG technique can be considered to be reinvented, since more and more centers are also recording LFP activity from micro- and macroelectrodes intraoperatively as well as in the immediate post-operative period from DBS electrode >
1284
78
Microelectrode recording in functional neurosurgery
. Table 78-1 Techniques of invasive physiological localization Microelectrode recordings Electrical stimulation, including microstimulation and macrostimulation Local field potential recordings (‘‘deep EEG’’ electroencephalography and electrocorticography) Noninvasive functional imaging Evoked potential recording Impedance monitoring Semimicroelectrode recordings Neural noise recording Microinjection of test substances (local anesthetic or muscimol) Optical imaging of cerebral cortex
contacts. The characteristic oscillation frequencies such as the STN beta band activity in PD patients off medication indeed helps to confirm STN localization. In a similar way, evoked potential recording has been used as a localization method in the past, and recently our group has used focal microelectrode evoked potentials with some further insight into physiological localization. Finally impedance monitoring is rarely used for localization and the techniques of microinjection of test substances are only occasionally used for specific indications or investigations.
Electrical Stimulation and Microelectrode Recording Currently, stimulation and recording are the most widely used techniques for the physiological localization of subcortical structures. Recording can be done with fine tipped microelectrodes capable of discriminating single cells [7–16] or with semimicroelectrodes that cannot [17–22]. Stimulation can be done with a large tipped electrode (macrostimulation) [3–6,17,23–29] or a microelectrode (microstimulation) [7,8,12]. Each technique has its advantages and disadvantages (> Table 78-2). In its simplest form, physiological localization can be reduced to observing the effects of macrostimulation on motor and sensory function.
In peripheral nerves, roots, and certain long tracts, low-frequency (often 2 Hz) stimulation is used to search for motor twitches and higher-frequency (30–300 Hz) stimulation is used for both sensory and motor effects, the latter in the form of tetanization. A correlation between the effect observed and the parameters used allows, with experience, a reasonable estimation of the distance of the stimulation probe from the target structure. Somatotopographic features also can be obtained, as in exploration of the trigeminal nerve, in which selective manipulation is desired. Similar macrostimulation techniques can be used for selective lesioning of the lateral spinothalamic tract. However, in most subcortical explorations, matters may be a little more complicated. Fritsch and Hitzig are said to have been the first to elicit motor responses by stimulating the motor cortex in experimental animals; Bartholomew was first to do so in humans, while Cushing was the first to elicit sensory effects in humans [19,30]. Macrostimulation was employed for confirmation of probe position in early functional stereotactic procedures [6] only after a probe was thought to have reached the appropriate target site. However, macrostimulation also can be done systematically at fixed intervals as the probe is passed into the brain, allowing the results to be plotted in ‘‘figurine charts’’ of the type used by Woolsey in the laboratory [31]. Such mapping provides more comprehensive localization data that are more easily assessed visually; in addition, it provides information about normal and abnormal brain organization, especially if each electrode trajectory is contained in the same sagittal plane. Wetzel and Snider [32] are said to have been the first to use microelectrode recording in humans in 1958 during a pallidotomy, but the technique did not come into common use until after the introduction of thalamotomy for Parkinson’s disease [2,9–11,18,19,21,22,30,33–51]. The earlier work in the thalamus was done with semimicroelectrodes, while later on true microelectrodes were employed. Semimicroelectrodes can more readily monitor
78
Microelectrode recording in functional neurosurgery
. Table 78-2 Comparison of microelectrode recording with stimulation for physiological localization in stereotactic surgery
Safety Speed of exploration Radius of current spread* Dependence on patient cooperation Ability to record and stimulate with same tip Ruggedness for repeated use Sophistication of backup equipment and activities Ability to detect somatotopy Variety of structures identified
Ability to functionally compartmentalize identified structures such as ventrobasal complex Ability to identify nearby structures Risk of technical problems, cost
Electrical Stimulation
Microelectrode Recording
Macrostimulation
Microstimulation
Low Impedance
High Impedance
High Rapid To 3 mm High No
High Medium 0.02–1 mm High Yes
High Medium <1 mm Low Yes
High Slow 20–200 mm Low Yes
High Low
High-medium Medium
High-medium Medium
Medium High
Good High (includes fibers)
Good High
Exquisite Low (recording from fibers limited)
No
Poor
Excellent Low (recording from fibers limited) Medium
High
Good Low
Medium Medium
Poor Medium
Poor Medium-high
*Depends on current used; microstimulation usually limited to 100 mA
neural ‘‘noise’’ as the electrode progresses along its trajectory, with the ‘‘noise’’ being sufficiently characteristic in some regions of each structure traversed to allow its recognition [52–54] (see chapter 93).
The Authors, Stereotactic Technique Stereotactic techniques have been described elsewhere and will be reviewed briefly here [7,8,12– 15,17]. The Leksell G frame is applied to the patient’s head under local anesthesia in the ward or radiology department. A stereotactic MRI is carried out, and the coordinates of AC and PC are determined. There are two general approaches, the more established procedure is to obtain a tentative target using a customized atlas reference to AC-PC and the second is to target
directly the structure of interest, given that MRI sequences are performed that enable visualization of the target. For the first procedure [17,55,56], a series of sagittal brain diagrams are generated using a customized program containing digitized plates from the atlases of Schaltenbrand and Bailey [27,57] and Schaltenbrand and Wahren [27] that are reformatted to conform to the AC-PC distance in each patient. These diagrams are also ruled in a millimeter grid in stereotactic coordinates, reflecting the position of the frame on the patient’s head. The program is also capable of overlaying the position of electrode trajectories and mapping the physiological data. A burr hole or twist-drill hole is now made in the same sagittal plane as the expected target so that all electrode trajectories will lie in the same sagittal plane if possible, facilitating the evaluation of physiological data. With regards the second method of direct targeting, this is obviously best for targets
1285
1286
78
Microelectrode recording in functional neurosurgery
easily recognized on MRI, such as the cingulum or subgenual cingulate cortex, allowing its coordinates can be read directly from the scan. In our opinion, targets such as the globus pallidum internum and the subdivisions of lateral thalamus (i.e., Vim, Vop. Voa, Vc) can not be recognized in this manner; their expected coordinates must be extrapolated from the computergenerated map and confirmed physiologically with an electrode to identify structures with motor and sensory properties respectively (see next section for details).
Macrostimulation Seven studies [17,23,58] have measured the variations in the locations of deep brain structures identified by macrostimulation from their location’s predicted from ventriculography. In each of those studies, discrepancies of 1 mm or more were found in a significant percentage of patients (15, 43, 42, 56, 45, 35, and 5%, respectively; mean, 34%) and discrepancies greater than 2 mm in 50, 32, 33, 23, 16, 13, and 22%, respectively (mean, 27%). A 2-mm discrepancy can lead to surgical failure or complications. We compared the actual locations of tactile neurons in the thalamus with those predicted from computed tomography (CT) imaging, finding no discrepancy in 62.7% of patients in the mediolateral, 63.4% in the dorsoventral, and 44.6% in the anteroposterior dimension; over a 2-mm discrepancy was seen in 10.8, 12.0, and 19.2% of these dimensions, respectively [59]. The largest discrepancies occurred in patients who had previously undergone craniotomies or had suffered massive damage to the brain from multiple sclerosis or stroke. Macrostimulation can be done with a monoor bipolar electrode, with the latter resulting in a little less current spread than the former [17]. Even so, with a 1.1-mm-diameter concentric bipolar macroelectrode employing a 0.5-mm tip
insulated from a surrounding ring by a 0.5-mm insulating band (available from Diros Technology, Toronto, Ontario, Canada), current may spread to involve up to a 3- to 4-mm sphere. We employ trajectories 2–3 mm apart and stimulate at threshold from about 10 mm above to 10 mm below the target in 2-mm steps, using manually controlled trains of 60-Hz monophasic dampened sine wave pulses of 3-ms duration.
Impedance Monitoring Impedance monitoring [60] (see chapter 79) can readily distinguish normal from pathological brain tissue. Collections of cerebrospinal fluid (CSF), abscesses, hematomas, and malignant and soft benign neoplasms have an impedance value of about 400 ohms, while pathological capsules and hard lesions have a much higher impedance. However, the technique is not sufficiently sensitive to reliably distinguish the differences between normal brain tissues. Though dense fiber tracts traversed perpendicularly by a probe may record an impedance as high as 1,600 ohms, the values in different fiber tracts or nuclear masses are not usually sufficiently distinct to provide the sharpness of localization required in stereotactic surgery.
Microinjections of Test Substances Test microinjections of substances that block, enhance, or otherwise alter neurological function can potentially help confirm an electrode’s position at a target site and may constitute a promising technique for the future for some sites, when a more diverse ‘‘library’’ of injectable substances will become available. Meanwhile microinjections of lidocaine [61] have proved useful in the selection of the thalamotomy target site for the control of tremor but are not routinely used in our group due to time constraints.
Microelectrode recording in functional neurosurgery
Evoked Potentials Recording of evoked potentials was used from the earliest days of stereotactic surgery but in its conventional use is now rarely used. Hassler and colleagues [23] recorded evoked cortical potentials, particularly in the precentral cortex, by stimulating the anterior ventral oral nucleus of the thalamus (Voa) and/or the posterior ventral oral nucleus (Vop). Stimulation of Voa at a frequency of 4–8 Hz induced a recruiting response that allowed French and coworkers [62] to identify the passage of a probe into the ventrolateral nucleus. Narabayashi and Ohye [63] evoked an augmenting response in motor and premotor cortex when they stimulated Vo at 6 Hz, but not from the ventral intermediate nucleus (Vim) of the thalamus. The latency of the first negative wave was 20–30 ms. Fukamachi and associates [18] correlated different patterns of cortical evoked potentials, depending on the site in thalamus stimulated. Sano and coauthors [64] were able to determine which part of the internal medullary lamina of the thalamus they were stimulating by virtue of the pattern of ipsilateral scalp evoked potentials produced. Albe-Fessard and coworkers [19] and Yamashiro and associates [65] demonstrated the varying patterns of evoked potentials recorded with microelectrodes from different thalamic nuclei in response to peripheral stimulation (see chapter 77 and 93). More recently our group has used dual microelectrode recordings that has allowed a new technique to be used, that of focal microelectrode evoked potentials within a nucleus (add ref in press). In this technique open filter recordings are made from one electrode while stimulating with single pulses from the other electrode 1 mm away at a similar vertical depth. With single pulses, evoked potentials are routinely found in SNr but not in STN, so this method confirms the location of SNr which can be sparsely populated with spontaneously firing neurons with characteristic properties. The positive going fields are likely of
78
GABAergic origin and can be used as a tool to examine activity- and dopamine - dependent synaptic plasticity at this basal ganglia output station. Within STN a field can also be evoked but with burst stimulation, and it remains to be determined if this is a focal STN response due to GABAergic afferents from GPe or a far-field contamination from SNr.
Microelectrode Construction Methods and Setup The methods for construction of electrodes are briefly covered here for the sake of completeness, but many companies now prepare long electrodes suitable for use with stereotactic guide tubes. Commercially available parylene-C-insulated tungsten or platinum-iridium microelectrodes are used with tip sizes ranging from 15 to 40 mm (height of cone) with initial impedances before plating of 1–2 MOhm. The lengths of these electrodes are in the range of 50–70 mm so they need to be extended for use with deep brain structures. If present, the connector pin is cut off the electrode and the insulation is stripped from the cut end, so that the shaft makes good electrical contact with a stainless steel extender tube. Kapton tubing insulates the extender tube and is glued to the insulation of the electrode shank to make a contiguous seal [12]. Electrodes are plated first with gold and then with platinum to reduce their impedance to about 0.5 MOhm. Electrodes are tested for adequacy of the insulation by using an electrolytic bubble test and/or observing constancy of the tip impedance measurement as it is immersed deeper in the conducting saline. The electrode apparatus is assembled on a table under sterile conditions by using a protective carrier tube that fits inside and is the same length as the outer reinforced guide tube. The outer guide tube is passed through the electrode holders mounted on the stereotactic frame and into the brain a sufficient distance to direct the carrier tube accurately to the target but not as far
1287
1288
78
Microelectrode recording in functional neurosurgery
as the area of the brain that is to be studied. The zero reading is obtained by fixing the top of the microelectrode shaft to the hydraulic microdrive at a point where the tip is visually observed to be flush with the tip of the carrier tube. The inner guide tube containing the microelectrode is then inserted into and fixed with a set screw to the outer guide tube already in place in the brain.
Subcortical Mapping with Microelectrode Recording The microelectrode is then extruded with a manually driven hydraulic microdrive. Continuous recording begins as soon as the microelectrode extrudes into the brain, usually 10–15 mm above the target site, and is continued along the trajectory through the target and a variable distance (5–10 mm) beyond it (> Figure 78-1). Microstimulation is done every 1.0 mm along the way. Currents of up to 100 mA do not damage the electrodes,. Physiological data and a voice channel are recorded for off-line analysis using a commercially available digital interface (CED 1401 mk II) that saves two channels of electrode data, 4 EMGs, 2 accelerometer traces and EOG or any other stimulus triggers as required. These are collected continuously into digital files. Similarly a digital video camera saving direct to DVD is used to monitor events and patient’s movements during the entire recording session. Responses elicited by micro- and macrostimulation are similar in quality, but the threshold and current spread differ. The size of the projected fields at threshold current (the minimal current needed to evoke a sensation) usually is smaller with microelectrodes, but in some cases macrostimulation also results in small projected fields (> Figure 78-2). Frequency of stimulation has a variable effect. Sensory effects may not be perceived with very low (1–5 Hz) or high (over 1,000 Hz) rates of repetition, and for the patient to perceive a sensory effect, trains of
several pulses are required rather than single shocks (> Figure 78-3). Frequency of stimulation in the motor system determines whether individual muscle jerks or tetanization will be produced, and in the extrapyramidal system it determines whether involuntary movements will be driven or inhibited. The main part of this chapter reviews physiological mapping with recording and stimulation in structures in which functional stereotactic procedures are regularly done, describing first the responses considered to be normal and then certain pathological ones.
The Thalamus Tactile Area The main sensory nucleus of the thalamus is the ventrocaudal nucleus (Vc), receiving input from the medial lemniscus and sending relay output to sensory cortex. In this region a microelectrode will record very unmistakable high background noise and the presence of large-amplitude units with so-called tactile’ neurons that respond to superficial cutaneous stimuli such as air puffs, (hair bending), and light brushing with gauze [14,47,66–70]. These neurons frequently respond faithfully to stimulation repetitions of up to at least 50 Hz, and those which adapt slowly and rapidly appear to be intermingled in a medially to laterally oriented homunculus that represents the contralateral half of the body; there may be a few neurons with ipsilateral representation on the lips, but otherwise the homunculus is contralateral. Intraoral responses are located in medial Vc at about 11–12 mm from the midline adjacent to the medial thalamic nuclei, and foot responses are located in lateral Vc at 18–20 mm from the midline next to the internal capsule. At least the labial and manual neurons are organized into parasagittal laminae often slightly concave medially so that a parasagittal electrode trajectory
Microelectrode recording in functional neurosurgery
78
. Figure 78-1 Reconstruction of an electrode trajectory through the sensory thalamus (Vc) 15 mm lateral to the midline in a patient with essential tremor. The receptive fields (RFs) of the cells encountered are shown to the left of the vertical line. The stimulation currents (Int.) and the location and quality of each sensation evoked by stimulation at sites throughout the trajectory are shown to the right of the vertical line. The shaded bar indicates the presumed tactile region of Vc, based on the presence of cells responding to light tactile stimuli. Note also the close correspondence of the projected fields and the RFs. Of particular note is the site marked with an asterix, at which stimulation at 3 mA evoked pain similar in quality and location to appendicitis pain previously experienced by the patient. Note also that 1 mm farther inferior stimulation evoked burning pain on the leg, and 0.5 mm below that it evoked a sensation of warmth. PF = projected field; P = paresthesia; O = other sensation; B = burning; W = warm; N = pain; Au = auditory; Ki = kinesthetic. (Reprinted from Davis et al, [72] with permission.)
1289
1290
78
Microelectrode recording in functional neurosurgery
. Figure 78-2 Projected fields and effect of stimulus intensity on evoked sensation. Examples taken from mapping in two patients with deafferentation pain. Figures show perceived location and spread of paresthesia evoked during microstimulation (1-s train, 300 Hz through a tungsten microelectrode) (a) or macrostimulation (with an electrode with a 0.5-mm exposed tip: 1-s train, 300 Hz) (b) in the ventrocaudal nucleus at various intensities. The patient’s ratings of the intensity of tingling (on a scale from 0 to 10) at various stimulation intensities is shown in each graph. Note the increased size and intensity of paresthesias evoked by increasing stimulus intensity. In A, the figure on the left shows the location of the receptive fields of the neurons recorded at the site of the stimulation. (Reprinted from Dostrovsky et al, [66] with permission.)
Microelectrode recording in functional neurosurgery
78
. Figure 78-3 Dependence of sensory threshold on thalamic Vc microstimulation parameters. Each stimulus-response curve shows the stimulus current required (threshold) to evoke a sensation when the train frequencies (left panel), pulse width (middle panel), or train length (right panel) is varied. (Reprinted from Dostrovsky et al, [66] with permission.)
will record neurons with similar receptive fields (RFs) throughout, except for the dorsal and ventral extremes of the trajectory. This medial concavity sometimes appears to be replaced by a laterally concave arrangement. RFs may be as small as 1–2 mm in diameter in the lips and manual digits but are much larger in the trunk and proximal limbs. Thebulkoftherepresentationisgivenovertothelips and manual digits.RFs arefound only in theinferior half to two-thirds of the basal part of Vc; we have not recorded identifiable cells more dorsally or in the central and dorsal subdivisions of Vc. The described somatotopy may be seriously deranged in patients who have suffered strokes, spinal cord injury, or other major deafferenting illnesses [71]. When stimulation is carried out in the thalamus in the region where tactile cells reside, paresthesias are produced in parts of the body similar in location to those from which RFs were recorded from neurons in the stimulation area. These are termed projected fields (PFs). With microstimulation, the match between the locations of RFs and PFs may be virtually perfect and the PFs may be induced with currents as low as 1 mA (> Figures 78-1 and > 78-2a). As the tip size of the stimulating electrode increases, the PF usually grows larger because of current spread [8,17,71–73]. What may confuse the stereotactic
surgeon is the occurrence of RF/PF mismatch even in an apparently healthy brain, possibly caused by stimulation of fibers of passage that activate a set of sensory cells distant from the ones recorded in the vicinity of the electrode tip. Mismatch is of course much more striking after strokes and other major deafferenting illnesses [71–73]. Suprathreshold stimulation in the region of tactile cells produces increasingly strong paresthesias but usually not pain (> Figure 78-2a).
The Kinesthetic (‘‘MovementSensing’’) Area Immediately rostral, adjacent to the tactile area, are neurons with high-amplitude spontaneous activity whose discharges change phasically or tonically in response to deep pressure (but not light stimulation) of the contralateral skin, passive joint bending, and deep tissue squeezing. This region doubtlesscontainsthespindleafferentrelay,though that relay cannot be specifically identified [74]. Such responses are probably located in Vim, though this has never been proved in humans. Again, there is a medial-to-lateral somatotopographic arrangement but one that appears to be less exquisite than that in the tactile area; among these
1291
1292
78
Microelectrode recording in functional neurosurgery
various neurons, the deep skin receptors appear to lie most caudally, immediately adjacent to the tactile neurons of Vc [40,42,59,63,68,69,75–84]. Electrical stimulation among these cells usually induces contralateral paresthesias indistinguishable from those produced in the tactile area in more or less the same part of the body as that of the receptors that activate the neurons. However, thresholds are higher (10–20 mA) and RFs are larger than in Vc. Thus, with macrostimulation alone, it is very difficult to define the presumptive border between Vim and Vc [17], that is, the junction between deep and kinesthetic neurons and tactile neurons. Occasionally, sensory motor effects are elicited rather than paresthesias as shown in > Table 78-3 [17] which represents all such responses found in about 10,000 brain sites
. Table 78-3 Macrostimulation-induced contralateral sensorimotor effects other than paresthesia in the human thalamus Effect Wants to, has to move Pulling Drawing Pressure Tightening Grabbing Pinching Squeezing Pushing Choking Teeth loose Jaw pushes in and out Pulsation Jumping Flicking Eyelid twitching Throbbing Jerking Cheek sliding to and fro Tremorous feeling Part continuously moving Weak Tired Heavy Loss of control
No. sites 10 10 5 5 9 3 3 1 1 1 2 1 6 3 2 2 1 1 1 1 5 3 3 3 1
that were macrostimulated. Another type of response occasionally induced with macrostimulation in Vim appears to be vestibular in origin. True vertigo [17], nonspecific dizziness, and faintness have all been elicited, presumably arising from stimulation of the rostral vestibulothalamic path. Microelectrode recording of vestibular cells has apparently not been reported in humans, probably because the head is immobilized in a stereotactic frame.
Voluntary Cells Moving more rostral again, probably into Vop and Voa, one encounters neurons generally considered to lie in the terminal fields of pallidal efferents, respectively. Spontaneous activity is less here, and the amplitude of the recorded units lower than in Vim and Vc. Particularly in Vop, cells are found that alter their firing pattern in response to a specific contralateral voluntary movement [43,68,75,84–91], and some of these so-called voluntary cells have RFs from passive movements that oppose the voluntary action to which they are related [86]. Lenz and associates [86] identified neurons here with increased or decreased firing rates occurring 200 ms before the onset of their related particular contralateral voluntary movements. The somatotopographic arrangement of these cells is again a loosely medial-to-lateral one, as is found in Vim. Raeva and colleagues [87–90] studied voluntary cells extensively not only in Voa and Vop but also in the basal ganglia and the reticular nucleus of the thalamus. They found voluntary cells that respond to one or more contralateral (and sometimes ipsilateral) movements with various patterns of change (decreased firing, short inhibition and then increased or decreased firing, changing in and out of synchronous firing patterns, and more complex patterns). Some cells, especially in the reticular nucleus, respond at the command to prepare to make a movement; others, at various stages throughout the actual movement. The cells were
Microelectrode recording in functional neurosurgery
found to be of two types: type A cells, firing spontaneously but with an irregular pattern at 1–20 Hz and tending to increase their firing during voluntary acts (71%), and type B cells, firing with short 3–5 Hz rhythmic discharges that tend to be suppressed during voluntary acts. Some cells become rhythmic at the onset or end of a voluntary movement, sometimes only when the same movement is repeated several times; such rhythmicity was not, however, de pendent on the presence of tremor. Other cells fired at tremor rates but not in phase with tremor; their activity was suppressed during voluntary movements. Stimulation among voluntary cells may sometimes induce a contralateral muscle contraction about a joint related to the voluntary movement that causes the neurons in the vicinity of stimulation to alter their firing [17]. These contractions are very focal at threshold stimulation, tend to fatigue after the repetition of stimulation, and involve more and more muscle groups as the strength of stimulation is increased; this effect needs to be distinguished from tetanization [92].
Tremor Cells In patients with tremor, kinesthetic cells are entrained to fire faithfully in time with the peripheral tremor, with this rhythmic firing ceasing when the tremor stops [20,21,23,34,37,59,63,75, 81,82,93–103]. These are known as tremor cells, and they constitute a very prominent feature of recording in Vim in patients with parkinsonian rest tremor; for obvious reasons, they are less conspicuous in patients with action tremor. Tremor cells also are found in the globus pallidus internus [104] and units have also been identified in the subthalamic area [39], which may correspond to fiber input of the prelemniscal radiation that inputs into the motor thalamus. Additionally some of these tremor synchronous units may belong to the zona incerta which has similar connectivity to STN. Tremor cells can
78
also be found in thalamus and pallidum of dystonia patients with tremor, and if sensory driven, these cells may fire in time with dystonic movements [105]. ‘‘Voluntary’’ cells also may fire synchronously with peripheral tremor, just as kinesthetic cells do, and may fire in this manner even when tremor is absent. Since some of these cells fire 200 ms in advance of a related contralateral voluntary movement, they have been nominated for the role of tremor pacemakers. Lenz and associates [86] found that the firing patterns of voluntary tremor cells with and without kinesthetic input was particularly tightly linked to the pattern of the peripheral tremor electromyogram (EMG) [98,99]. Attempts to prove whether tremor in, say, Parkinson’s disease is a result of pacing by voluntary cells or of deranged feedback in kinesthetic cells have tended to suggest that both processes may be at work [100,106–108]. Obviously, then, tremor cells have something to do with the tremor seen in Parkinson’s disease and other conditions, yet despite the fact that this has been known for decades, it remains unclear what that relationship is or what the mechanism of thalamotomy or chronic electrical stimulation of the thalamus is in the alleviation of tremor. > Figure 78-4 shows the reduction of tremor produced by electrical simulation and by lidocaine microinjection in the anterior Vim. Most surgeons have claimed that the lesion site for thalamotomy or the site for implanting a chronic stimulating electrode in the control of tremor is located in Vim or caudal Vop among kinesthetic or voluntary tremor cells at sites where acute stimulation most effectively arrests the tremor [20,40,42,81,109]. However, plots of different surgeons’ target sites vary considerably [110]. Moreover, we and others have experienced tremor recurrence and the complication of ataxia when making lesions presumably in Vim guided only by the location of kinesthetic tremor cells just rostral to manual digital tactile cells, where effective stimulation-induced
1293
1294
78
Microelectrode recording in functional neurosurgery
. Figure 78-4 Effect of thalamic microstimulation and microinjection of lidocaine on parkinsonian tremor. a. This figure shows a reconstruction (based on the AC-PC coordinates) of a microelectrode trajectory through the thalamus in the 15-mm lateral plane in a patient with Parkinson’s disease. The location of neurons responding to movement of joints and/ or deep tissues or to touch are shown on the pattern of stipple, and the locations where stimulation reduced tremor or evoked a sensation are shown by the solid/open bars. The site of microstimulation and microinjection is indicated by the filled circle in Vim. b. The effect of microstimulation (2-s train, 0. 1 -ms pulses, 300 Hz, 100 mA) on tremor is apparent from EMG recordings from the biceps (B), triceps (T), and wrist extensor (WE) and flexor (WF) muscles. The solid bar indicates the time of the stimulation. c. Effect of microinjection of 0.5 ml of 2% lidocaine on EMG activity. (Reprinted from Dostrovsky et al, [61] with permission.)
tremor arrest occurs [78,111]. It is our opinion that the target size not only must meet these criteria but also must be located about 2–3 mm above the intercommissural line and a similar distance rostral
to the rostral margin of the tactile digital cells, similar in location to the site proposed by AlbeFessard [112]. Such a site coincides well with that proposed by Lenz and associates [86] on the basis
Microelectrode recording in functional neurosurgery
of the fact that tremor cells here are most tightly linked to the peripheral tremor EMG [113] and fits with current functional imaging studies implicating the cerebellum in tremorogenesis [114,115]. Such a site encompasses both kinesthetic and voluntary cells as well as unidentified tremor cells that do not alter their firing in response to known sensory input or movement and encompasses such a small volume of thalamus that it would not seem possible to differentially manipulate only one type of tremor cell. In Parkinson’s disease, any postoperative tremor recurrence tends to take place in the first 3 months; tremor recurrence is rare after that time. Whereas equally effective tremor suppression is also seen in essential tremor, there is a tendency for recurrence with time; in the case of cerebellar tremor, suppression is never complete and recurrence with time is the rule [116] (see chapters 104 and 105).
Pain Pathways The main pain and temperature pathway for sensations below the head ascends in the lateral spinothalamic tract (STT). The analogous pathway for oral and facial pain and temperature is the trigeminothalamic tract (TrT) that originates in the caudal part of the trigeminal spinal tract nucleus (subnucleus caudalis) [117]. Recent anatomic techniques have indicated that the ascending fibers in these tracts terminate in a number of distinct thalamic nuclei [118–121]. In the lateral thalamus, these include primarily the Vc nucleus and a region ventroposterior to it. The latter region includes a number of relatively ill-defined nuclei, including the ventrocaudal parvocellular nucleus (Vcpc) [122], posterior ventromedial nucleus (VMpo) [120] ventroposterior inferior nucleus (VPI), and suprageniculate/limitans. It is unclear at the present time whether pain is processed in each, some, or only one of these regions. The STT also projects to the medial thalamus, in particular to the centralis lateralis and ventrocaudal (MD)
78
[118–120]. The medial thalamus also receives ascending projections from the reticular formation and other brain stem structures. It has been assumed for many years that at least some of these reticulothalamic neurons relay nociceptive information from the spinal cord and are thus part of a spinoreticulothalamic pathway (also termed in older literature the paleospinothalamic tract), although there is not much direct evidence for the existence of such connections from modern tract tracing techniques [123]. It has proved difficult to trace the pain and temperature pathways physiologically in human CNS. Only a few studies have searched for the presence of nociceptive neurons, and reports of responses to noxious stimuli have been primarily anecdotal and have not been replicated. Hitchcock and Lewin [124] recorded increased cellular activity in cuneate and gracile fasciculus as well as caudal trigeminal nucleus and STT in three patients when they stimulated the ipsilateral face and limbs and applied contralateral noxious stimulation during percutaneous cordotomy. Amano and coauthors [125] recorded neurons with widespread bilateral RFs responding to nociceptive stimuli in the mesencephalon medial to the STT during mesencephalic tractotomy. The responses were of low voltage, and the latencies in response to pinprick fell into three groups: less than 250 ms, 400–800 ms, and more than 1,000 ms. It is not clear exactly from which region these responses may have arisen. This is a region of brain stem from which, in our experience, no sensory responses are elicited in most patients by electrical stimulation (see the discussion below) [17]; stimulating a little more laterally in the STT elicits a contralateral feeling of warmth or coldness similar to the responses elicited by stimulation of the STT in the upper cervical cord and elicited during percutaneous cordotomy [17]. Within the medial thalamic nuclei where reticulothalamic and STT neurons project, Sano and colleagues [64,126,127] identified two types of neurons responding to nociceptive stimuli, one with a 30–90 ms latency and the other with a 100–500 ms latency; RFs were
1295
1296
78
Microelectrode recording in functional neurosurgery
extensive and bilateral. Although there have been many reports of the existence of nociceptive neurons in the medial thalamus in monkeys, cats, and rats, our group has been unsuccessful in finding such neurons in these regions in humans. Electrical stimulation of the medial thalamic structures, which are thought to represent the nonspecific pain relay, in our experience seldom elicits any conscious response except the responses sometimes produced at high intensities and attributed to current spread to ascending STT axons. However, Sano and colleagues [64,126,127] reported eliciting pain by stimulating the internal thalamic lamina. In the lateral thalamus, where there have been many reports of nociceptive neurons in animal studies, there are currently reports only from one group of the existence of some thalamic neurons activated by noxious stimuli. These neurons were recorded within and ventroposterior to Vc [128,129]. Our group was recently able to record a few neurons activated by cooling in the region ventroposterior to VC [130] Also in the lateral thalamus, there have been several reports of stimulation-evoked pain and temperature sensations [17,72,122,130–136], and lesions there have been reported to relieve, chronic pain [17,131,137]. These sensations of contralateral pain usually are evoked when the electrode is ventroposterior to Vc, in the Vcpc region [72,122,132–135]. These sensations are in contrast to the paresthetic responses seen throughout the rest of Vc. They are often quite striking; as an electrode passes ventroposteriorly through Vc, a succession of paresthetic responses are invariably evoked, each of which is referred to a small region of the contralateral body. Near the base of Vc, the threshold response can change from paresthetic to warmth or pain and the somatotopy can change abruptly (> Figure 78-1). For example, after a succession of PFs in the hand, PFs can shift to the contralateral leg. The evoked sensations frequently include a burning component and occasionally are referred to internal sites [17,72,133,135,136]
(> Figure 78-1). Interestingly, these sensations can frequently evoke distressful and aversive sensations, qualities usually attributed to medial thalamic processing of pain. At some sites, warmth rather than pain is evoked. Sensations of cold can be elicited at some sites in this region but are much rarer, possibly, as suggested by our preliminary findings, because they are evoked from a more medial region that usually is not explored [130]. In a small number of cases, in particular in central pain patients, stimulation within the tactile relay nucleus Vc itself can elicit unpleasant and painful or sometimes warm sensations [72]. However, in the vast majority of cases, stimulation within Vc, even at intensities more than four times threshold to evoke paresthesia, is not reported as painful. In summary, in most patients, stimulation of Vc elicits paresthesia at threshold and suprathreshold levels. In contrast, stimulation ventroposterior to Vc, presumably in Vcpc/VMpo, and also in the mesencephalic STT frequently elicits feelings of pain, warmth, or cold. Stimulation medial to the STT in the midbrain and the medial thalamic nuclei normally produces no perceptual responses. In certain patients with chronic pain, however, remarkable aberrant responses are seen. We have found that threshold macrostimulation of mesencephalon medial to the lateral STT and of the above-mentioned medial thalamic nuclei medial to the intraoral portion of Vc may elicit contralateral sensations of burning in certain patients with neuropathic pain but usually not in those with cancer pain, in whom no effect is elicited [14,17,138]. In some patients with neuropathic pain, especially that caused by stroke, such macrostimulation may cause frank pain, often similar in quality to that from which the patient suffers [139,140]. We have had only limited opportunities to repeat these studies with microstimulation and have found that microstimulation in medial thalamus is not effective in inducing pain, possibly because of the limited (approximately 100 mA)
Microelectrode recording in functional neurosurgery
amount of current that can be passed through them. In the lateral thalamus, we recently showed that there is an increased incidence of stimulationevoked pain responses to microstimulation within Vc in poststroke central pain patients [72].
Responses in Vc in Patients with Deafferenting Lesions Recording of neuronal activity in Vc tests the integrity of the pathway from the receptor to the thalamus; microstimulation here examines the pathway from thalamus to, presumably, the cortex [14,139–142]. If the patient’s disease interrupts the sensory afferent pathway, the related RFs will be absent but may be replaced by new RFs coming from undamaged input that can sometimes be unusual. If the deafferentation is selective (as in lateral medullary syndrome), the RFs dependent on the medial lemniscus may persist while those dependent on STT will be absent. If the afferent pathway from the thalamus to the cortex has been destroyed, there may be no PFs though RFs persist, or PFs may be elicited in a portion of the body different from that expected for stimulation at that site. It is possible that in brains damaged by, say, stroke, the effects of stimulation depend on its retrograde rather than anterograde propagation, but this has not been proved. It is a feature of brain central pain that even in patients in whom neither PFs nor RFs can be identified throughout the thalamus, ongoing steady pain on the side of the body contralateral to the stroke (often associated with allodynia and hyperpathia) may still be present and sensory perception, though diminished, is not absent. Such observations suggest ipsilateral transmission of sensation as well as ipsilateral generation of central pain, including allodynia and hyperpathia [143], a situation similar to that seen after hemispherectomy [144]. We saw one patient with a thalamus badly damaged from stroke in whom sensory motor cortical stimulation on the side
78
ipsilateral to the stroke produced both ipsilateral motor and sensory effects as well as partial control of the patient’s central pain. We explored another patient with a neuropathic pain syndrome involving the right upper extremity caused by an intramedullary tumor in whom no RFs related to that limb could be identified in contralateral Vc but in whom the PFs related to it appeared normal; thus, loss of input to the thalamus does not necessarily result in transsynaptic degeneration of the thalamocortical circuits. In patients with small infarcts affecting a portion of Vc, microelectrode exploration frequently reveals a ‘‘hole’’ devoid of RFs and PFs corresponding to that lesion [139], with adjacent surviving Vc neurons assuming the input of some of the neurons lost in the hole with a resulting disruption of the usual somatotopographic organization. Ohye and colleagues made similar observations [73]. These changes obviously result in varying degrees of mismatch between RFs and PFs and complicate the conduct of the functional stereotactic procedure. A second abnormality seen in patients with destructive lesions of the nervous system is the presence of bursting cells. Thalamic cells typically fire in a bursting pattern during sleep but not during wakefulness [145–147]. Bursting cells have long been known in animal models and in humans at various sites in the nervous system upstream from deafferenting lesions and frequently have been linked to the pathophysiology of neuropathic pain [148,149]. Our recent observations, however, suggest that bursting cells observed in the lateral thalamus in pain states are markers of deafferentation and do not necessarily imply a role in mediating neuropathic pain, since we have observed their occurrence after a previous thalamotomy for Parkinson’s disease, in patients with stroke-induced dystonia who do not suffer from pain, and in patients with multiple sclerosis being operated on for tremor who do not have pain [139,150,151]. They are common in the thalamus of patients with spinal cord lesions whether or not those patients have pain [139]. Many of these
1297
1298
78
Microelectrode recording in functional neurosurgery
bursting cells fire in a characteristic pattern showing interspike intervals that lengthen as the burst proceeds with the first interspike interval and varying inversely with the number of spikes in the burst [71,151,152]. This pattern strongly suggests that the underlying mechanism for the bursting is the activation of a low-threshold calcium current that results in a calcium spike [147]. The final pathophysiological feature to be considered is related to the effects elicited by stimulation. As has been mentioned previously, in certain patients with chronic pain caused by stroke, macro- or microstimulation of Vc may elicit contralateral somatotopically organized painful responses rather than paresthesia. Such responses seem to occur in stroke patients in whom allodynia and/or hyperpathia are prominent [14,138,139, 153,154]. Thus, thalamic stimulation in these patients seems to induce a central allodynia, just as stimulation of non-nociceptors in the periphery produces pain through disorganization of signal processing in the dorsal hom. It is intriguing to speculate about the pathophysiology of pain production by stimulating Vc. We studied four stroke patients in whom stimulation of the periventricular gray (PVG) suppressed allodynia and hyperpathia [155,156]. The practical implication of these observations is that the usual type of paresthesiaproducing deep brain stimulation (DBS) usually administered in Vc is impractical for pain control in these patients, because it actually produces pain, and may be dependent on the activation of pain pathways capable of suppression by PVG stimulation.
The Auditory and Vestibular Pathways Auditory and vestibular responses have been noted rarely in the course of stereotactic procedures, the former with both micro- and macrostimulation and the latter only with macrostimulation [6,26,31]. In the lateral lemniscus, the medial
geniculate, and their proximal projections, neurons can be recorded that respond to auditory stimuli addressed to the contralateral ear [17,154,157]. However, micro- or macrostimulation at such recording sites usually induces a beelike buzzing heard chiefly in the contralateral ear, with the pitch of the sound induced apparently unrelated to the frequency of the stimulation. Vestibular effects consisting of true vertigo, a feeling of side-to-side translation, nonspecific faintness, or dizziness also may be elicited by macrostimulation at the same sites where auditory findings occur as well as in Vim, as has already been mentioned [17,24,26,158].
The Visual Pathways The optic tract is most often encountered during pallidotomy, since it lies ventral to the internal globus pallidus target used to treat Parkinson’s disease [12]. Axonal responses may be recorded showing action potentials in response to light flashes delivered to the contralateral visual field, and macro- or microstimulationinornearthetractproduceswhite or colored phosphenes, usually in the form of spots of light or stars in the contralateral visual field [12,17]. The lateral geniculate lies lateral to most stereotactic target sites, so that it is seldom penetrated; scanty evidence suggests that stimulation here produces colored phosphenes in the contralateral visual field [17,21,159,160]. Occasionally in the course of performing mesencephalic tractotomies, visual responses have been elicited with macrostimulation in the tectum and more deeply in the midbrain 14–17 mm from the midline (deep to sites of oculomotor activation), presumably arising from activation of the tectopulvinar pathway, which projects to the second visual cortex. This structure does not support vision in humans, as it does in some other species of mammals [17]. The responses recorded here are of three types: white phosphenes, a blanking out of vision, and the sense
Microelectrode recording in functional neurosurgery
of movement of the visual fields in the absence of any obvious movement of the eye itself, all referred to the contralateral fields.
The Anterior and Dorsomedian Nuclei of the Thalamus The anterior and dorsomedian thalamic nuclei are of historical interest in the field of psychosurgery [6]; the usual bilateral dorsomedian nucleus lesions may induce recent memory loss. There is little information concerning microelectrode recording in these structures, and in our experience their stimulation does not elicit detectable responses. We have noted that as a microelectrode is passed parasagittally from rostrodorsally toward the medial parafascicular nucleus, bursting activity is encountered as the electrode traverses the dorsomedian nucleus, which ceases at the presumed dorsal border of the parafascicular nucleus in a plane 2 mm lateral to the wall of the third ventricle. This bursting activity is similar to that observed in the lateral thalamus in that the firing pattern suggests an underlying calcium spike [151,161]. Similar activity has been reported by Rinaldi and associates [162] and Jeanmonod and coworkers [161] who suggested that the bursting activity was associated with the neuropathic pain syndromes from which their patients suffered. However, since similar recordings have not been obtained from this region in patients who do not have pain, this interpretation must be viewed with caution until it is confirmed in animals or nonpain patients.
The Pulvinar Pulvinarotomy was advocated in the past for the relief of spasticity and intractable pain, but it is uncertain whether the procedure is still used [17]. Martin-Rodriguez and coauthors [163] identified neurons in the pulvinar that fired rhythmically but not in time with any tremor the patient might
78
exhibit, while other neurons responded to visual and auditory stimuli or altered their firing in relationship to contralateral voluntary acts.
Periventricular and Periaqueductal Gray Matter Reynolds [164] demonstrated relief of nociceptive pain in rats when he stimulated neurons in the wall of the ventricular system between the rostral third ventricle and the aqueduct, apparently through the effects of a descending pathway relaying in the nucleus raphe magnus that inhibited nociceptive spinothalamic tract neurons. This led Richardson and Akil [165] to introduce chronic stimulation of PVG, a comparable structure in humans, for the relief of chronic pain. Controversy has arisen about whether such stimulation relieves both nociceptive and neuropathic pain; whether PVG stimulation is to be preferred over PAG stimulation, since the latter is unpleasant according to some researchers [162,165–168]; and the precise site is not known for stimulation. Gybels and Kupers [169] concluded that medial PF was the preferred target, 5 mm rostral to PC on the AC-PC line and 2 mm lateral to the wall of the third ventricle. Levy and coworkers [170] provided an important review of the clinical outcome of such stimulation, while Young and associates [168] correlated pain relief from PVG stimulation with release of the opioid peptide met-enkephalin into the ventricular CSF. We have found that stimulation of the periaqueductal gray matter (PAG) is usually unpleasant, and so we prefer target sites in PVG [171]. In our hands, PVG stimulation is ineffective for the control of the steady element of neuropathic pain [171], though it relieved the allodynia and hyperpathia seen in several patients with stroke-induced central pain [155] (see chapters 121 and 132). The physiological localization of PVG and PAG stimulation sites is, however, not straightforward; they frequently can be located only on anatomical basis. The cessation
1299
1300
78
Microelectrode recording in functional neurosurgery
of bursting at the dorsomedian-parafascicular nucleus border has been mentioned [162] and is of some help in target localization. In certain patients, however, macro- or microstimulation of PVG induces acute pain relief and/or a sense of satiety, warmth, inebriation, or well-being, while that of PAG may induce a feeling of horror, dizziness, or diplopia, helping the surgeon select the stimulation site [165,172]. > Table 78-4 summarizes the clinical neurophysiology of the thalamus.
The Subthalamic Area This area (not the subthalamic nucleus of Luys), lying beneath the thalamus, which has clinical physiological characteristics similar to those of Vim and Vop (including the presence of tremor cells), is an alternative target to Vim-Vop for the relief of involuntary movements [39]. Perhaps . Table 78-4 Effects on mood, consciousness, and autonomic function elicited in human midbrain and thalamus Response Effects on mood Laughing Crying, whimpering, grimacing Effects on consciousness Dreaminess Loss of memory, ‘‘foggy head’’ Sleepiness Autonomic Comfortable abdominal feeling, warm or cool Unpleasant abdominal feeling, faint, cold Something going to stop heart Something in chest speeding up, anxiety Gas coming out of stomach As if to take a deep breath Total
Number of Observations 9 1 8 3 1 1 1 23 11 4 3 1 1 1 35
the fact that it contains densely packed fiber tracts of the prelemniscal radiations projecting into Vim and Vop explains its efficacy, but it also explains the enhanced risk that a lesion here will produce the ‘‘cerebellar’’ complications of dysarthria, ataxia, and gait disturbance [111].
Motor Pathways Among the most striking effects of subcortical stimulation are the motor ones, which consist of isolated motor twitches in time with the stimulation pulses at lower frequencies that fuse into tetanization when higher rates are used. These twitches may arise from (1) lower motor neurons, (2) the corticospinal tract, (3) other upper motor neuron pathways, and (4) autonomic pathways [17]. In our experience, lower motor neuron effects have been confined to the oculomotor nerve and have been elicited during midbrain exploration. Motor effects are produced by stimulation of the corticospinal tract in the internal capsule during pallidotomy [12,21,25,50,173–177] and thalamotomy. During thalamic procedures, vocalization also may be elicited by stimulation in the capsule or cerebral peduncle [17,68]. Motor effects are elicited by stimulation of the corticospinal tract of the lower brain stem and spinal cord. Stereotactic physiological explorations have revealed a functional organization of the motor fibers of the internal capsule that is different from that taught in classic texts. Rather than a discrete somatotopy extending caudally from the representation of the face at the genu, Guiot and colleagues [173] found a less discrete separation of the representation of various body parts all located far posteriorly in the posterior limb. Other observers reported similar findings [178,179]. Hawrylyshyn and associates [180] found that the distance from the midline of the motor fibers in the internal capsule was heavily influenced by the width of the third ventricle, so that in a patient with an anatomically normal brain, say, a young
Microelectrode recording in functional neurosurgery
individual suffering from primary dystonia, a thalamotomy lesion 14 mm from the midline, typical for the relief of tremor in a parkinsonian patient, may damage the internal capsule. Stimulation in the motor thalamus usually does not evoke movements, although stimulation in parkinsonian patients with tremor is frequently very effective in blocking their tremor (> Figure 78-4), especially at high frequencies (> Figure 78-5). Macrostimulation and occasionally microstimulation can produce asterixis-like inhibition [181] (> Figure 78-6). More complex responses are elicited in other upper motor neuron pathways, including the oculomotor tracts [17] and are found in the hypothalamus and apparently in the medial forebrain bundle, tectospinal tract, and medial longitudinal fasciculus.
78
Stimulation-induced head tilting, always to the ipsilateral side, arises from sites 2 to 7 mm from the midline in the midbrain posterior to the red nucleus, possibly arising from activation of the medial longitudinal fasciculus or central tegmental tract dentatorubrothalmic tract or interstitial nucleus of Cajal [17,23,182,183]. These are sometimes associated with ipsilateral or bilateral facial tetanization from stimuli delivered 9 to 11 mm from the midline. Oculomotor effects have been elicited frequently in humans at subcortical sites such as the hypothalamus [126,184], subthalamus [29, 58,185–187], thalamus [165,188] and midbrain [176,189,190]. In our experience, they occurred widely from the level of the hypothalamus through the midbrain between 0 and 9 mm from the midline, grouped in four different
. Figure 78-5 Effect of varying thalamic stimulus parameters on parkinsonian tremor. Surface EMG recordings obtained from the wrist extensors (WE) and flexor (WF) muscles during thalamic stimulation are shown for 0.5-s stimulus trains of varying pulse frequency (top panels) and 300-Hz trains stimuli varying in duration (bottom panels). (Reprinted from Dostrovsky et al, [66] with permission.)
1301
1302
78
Microelectrode recording in functional neurosurgery
. Figure 78-6 Effect of thalamic stimulation on voluntary contraction of various muscles. Each trace is the average rectified surface EMG obtained from 100 sweeps. Peaks and troughs greater or less than ±2 SD (short horizontal lines on left) are indicated by black shading. The stimuli, 1-ms pulses of 1 mA, were delivered at time zero from a chronic stimulating electrode implanted in the motor thalamus. Note the pronounced inhibition produced in the contralateral muscles. (Reprinted from Ashby et al, [181] with permission.)
areas. One group lay between 2.5 and 6.5 mm from the midline in the superior colliculus, consisting chiefly of conjugate eye movement without pupillary change. A second group was located in the posterior hypothalamus 4–7 mm from the midline, which coalesced with a third group in the subthalamus. Among hypothalamic responses, ipsilateral eye movement with uni- or bilateral mydriasis was the most common (33%); ipsilateral eye movement without pupillary change occurred at 14% of sites, and bilateral eye movement with bilateral mydriasis occurred at 23%. The most common eye movement in the hypothalamic area was adduction of both eyes, seen at 81% of sites, whereas isolated ipsilateral eye movement most often consisted of depression (68% of sites); adduction occurred at 27% of sites. The fourth group consisted of various oculomotor effects in the subthalamus and midbrain tegmentum. Autonomic and psychic [23,26,186,188,191– 196] responses are associated with functions such as regulation of pulse rate, blood pressure,
and respiration and have been identified chiefly during stereotactic procedures on the hypothalamus for the control of chronic pain and behavioral disorders [17]. Since hypothalamic stimulation in an awake patient is said to be unpleasant, such operations usually have been performed under general anesthesia, and so only objective observations could be made [126,184,197] (> Table 78-4). The most commonly observed effect in our experience, occurring at 81% of sites, was respiratory suppression and apnea elicited 2–7 mm from the midline in the posterior hypothalamus and the immediately adjacent rubral area. Tachypnea occurred rarely (6% of sites) in the rostral hypothalamus, hypertension at 20%, hypotension at 14%, tachycardia at 13%, and bradycardia at 9%. Hypotension was elicited by stimulating throughout the hypothalamus 2–4 mm from the midline, and hypertension by stimulating the inferior part and adjacent rubral area in the same sagittal planes. Tachycardia was produced in the anterior rubral area, and bradycardia throughout the hypothalamicrubral area.
Microelectrode recording in functional neurosurgery
Speech Although a number of authors have reported effects on speech during subcortical macrostimulation [17,26,198], it is difficult to synthesize those observations. In our experience in stimulating nonbehaving patients, the only effect seen was speech arrest seemingly from tetanization originating in the internal capsule or motor cortex. Ojemann and Van Buren [199,200] by contrast, demonstrated that when they stimulated a patient carrying out a speaking paradigm, they elicited interference with object naming in the anterolateral dominant Vo causing repetition of the same wrong name. Stimulating more medially produced perseveration and anomia. Stimulating Vo during tasks of input to memory improved recall, but during the stage of retrieval from memory it impaired it. The nondominant thalamus appeared to be involved in processing letters and numerals [199–207].
Globus Pallidus Umbach and Ehrhardt [43,91] recorded neurons in the globus pallidus that altered their pattern of firing during a voluntary movement. Their spontaneously rhythmic synchronous activity was suppressed at the command to move as well as at the initiation of an actual movement. Raeva and associates [88–90] recorded dense low-voltage, high-frequency (150 Hz) activity in the globus pallidus, sometimes with silent intervals, from cells that did not change their pattern of firing in response to a voluntary movement. Thirty-five percent of globus pallidus (GP) cells fired rhythmically near but not precisely at the frequency of any tremor present, while 50% lacked high frequency discharges. The latter cells changed their firing pattern over a 300–400 ms period at the onset and end of a voluntary act. Some cells increased and others decreased their firing rates, while others responded in a complex manner.
78
In GP, neurons might alter their firing in response to both contralateral and ipsilateral movements (see chapter 95 and 107). In the anteromedial putamen, spontaneously active low voltage units firing at or above 120 Hz were found that displayed silent periods as well as episodes of group discharges. Other cells fired infrequently. Thirty percent fired rhythmically near tremor frequency. Discharges might be inhibited or excited when a contra- or ipsilateral voluntary movement began or ended, or the responses might be complex. The findings in the caudate nucleus were similar; any rhythmic activity present there was not synchronous with tremor.
Neurophysiological Findings of our Group The types of neurons encountered in anterodorsal trajectories penetrating the external globus pallidus (GPe) and internal globus pallidus (GPi) in six patients with Parkinson’s disease in the ‘‘off’’ state have been described [208,209], and the findings from our group will be summarized here. In general, these neuronal firing patterns are similar to those described in monkeys that have been rendered parkinsonian with MPTP (1-methyl-4phenyl- 1,2,3,6-tetrahydropyridine) [210,211]. Although a quantitative analysis of the various types has not been carried out in the human GP, our observations are as follows. The GP consists of an external (GPe) and an internal segment which is further subdivided anatomically into an external (GPi,e) and an internal division (GPi,i) as depicted in sagittal sections (18.5 or 20 mm lateral to the midline) of the stereotactic atlas (see > Figure 78-7, top). Most of the neuronal activity encoded in spike trains recorded in GP of these patients is grouped in irregularly occurring bursts, as demonstrated by firing rates calculated from the reciprocal of the most common interspike interval. In GPe,
1303
1304
78
Microelectrode recording in functional neurosurgery
. Figure 78-7 Cellular firing rates in the globus pallidus (GP). Data obtained from the neuronal recordings encountered throughout a typical microelectrode trajectory through the human GP. The top panel shows the location of the trajectory based on a sagittal reconstruction from the Schaltenbrand and Wahren atlas in the 20-mm lateral plane. Tick marks on the trajectory (S2) indicate 1-mm spacing. The bottom panel shows the mean firing rates (±SD) of neurons at sites throughout the trajectory as a function of the distance to the optic tract (based on physiological findings). (Reprinted from Hutchison et al, [208] with permission.)
we found that most units fired in one of three characteristic patterns and were termed (1) slowfrequency discharge (SFD) units, whose activity may or may not be punctuated by pauses in firing; (2) low-frequency bursting (LFB) units that fire overall at about 10 Hz but whose activity is punctuated with rapid bursts; (3) occasionally, highfrequency discharge (HFD) units encountered in GPe that fire in an irregular pattern at high frequency, sometimes with pauses (> Figure 78-8). The mean neuronal activity in GPe was 60–86 Hz
(SD, n = 40). In GPi,i, the neuronal activity is significantly higher than in GPe. Many of the units encountered fire at a frequency above 100 Hz. As observed in GPe, most of the neuronal activity occurs in irregular bursts in the spike trains. The mean firing rate for GPi,i was calculated to be 82–92 Hz (SD, n = 89). The elevated mean firing rate of neurons in GPi has been observed in MPTP-treated monkeys [210,211] and has been confirmed in humans with Parkinson’s disease [209].
Microelectrode recording in functional neurosurgery
. Figure 78-8 Firing patterns of cells encountered in the globus pallidus. Each example shows the spontaneous firing pattern of typical cell types. In the external segment of the globus pallidus, cells included the low-frequency discharge burst (LFB) and slow-frequency discharge firing with pauses (SFD-P) types typical of GPe. Regular firing-type cells (border cell – Bor) were found in the laminae and pallidal borders, and irregular, high frequency cells were found in the internal segment of the globus pallidus (GPi). (Reprinted from Hutchison et al, [208] with permission.)
Border neurons have been described in nonhuman primates and human GP [208,210] and are readily distinguished from other neurons in human GP by a regular pattern of firing and a lower overall rate of 44 17 Hz (SD, n = 17). They occur at or near the borders of GP and also in the internal laminae. Some of these neurons have been observed to switch into a repetitive bursting activity for several tens of seconds and then switch back to the regular firing pattern mode, which we have tentatively termed ‘‘border-burst’’ cells.
78
With the passage of the electrode tip out of the lower border of GP, the background noise in recordings diminishes and few units are encountered. Occasionally, axons can be recorded. A constant-current stimulator is used to microstimulate through the tip of the electrode. At a distance 1–2 mm from the lower border of GPi, patients report visual sensations from electrical excitation of fibers of the optic tract (‘‘phosphenes’’). Usually the patient is requested to close the eyes so that phosphenes can be more readily perceived. Frequently, white or yellow starbursts or flashes are reported with currents as low as 2 mA. Microstimulation parameters do not exceed 100 mA, 0.2 ms, 1 s, 300 Hz. When the patient does not report phosphenes from microstimulation at sites that correspond anatomically to the optic tract, recordings of flash-evoked axonal potentials from the optic tract are made. In isolated instances, recordings of the flash-evoked potentials have enabled confirmation of the location of the optic tract when microstimulation has failed. Lesions are made by radiofrequency electrocoagulation at cell-dense sites in posteroventral GPi that are at least 3 mm from the optic tract and internal capsule. These are the main guidelines that determine final lesion placement. The presence of units with high-frequency firing, units with movement related activity, and units with oscillations in the firing rate tentatively identified to be synchronous with tremor (termed ‘‘tremor cells’’ here) provides additional confirmation of the optimal target.
Pallidal Tremor Cells Although neurons with tremor-related activity have been well studied in the human motor thalamus, there have been only isolated reports in the literature of pallidal tremor cells. Umbach and Ehrhard [43,91], recorded pallidal units that fired 20 ms after the tremor burst in the contralateral limb. Jasper and Bertrand [68] reported tremor cells at the end of long trajectories curving
1305
1306
78
Microelectrode recording in functional neurosurgery
. Figure 78-9 Example showing tremor cell in GPi. The top trace shows the rectified and filtered EMG from contralateral wrist extensors, and the bottom trace shows the firing of a neuron in GPi. Note the close correspondence between firing pattern and tremor (EMG).
anteriorly from the thalamus that were presumed to be in the medial segment of GP. Raeva [88] also reported pallidal tremor cells but did not find significant correlation with limb tremor. We recently identified and characterized 28 tremor cells in human GP in three patients with tremor [104] II8 (> Figure 78-9). Some pallidal tremor cells have been evaluated with crossspectral analysis and have been shown to have rhythmic neuronal activity that is highly coherent with limb tremor [104]. Interestingly, in one patient in whom limb tremor was different for the upper limb and the lower limb, the tremor cell was coherent with the forearm EMG (4.6 Hz) while the foot EMG was at a higher frequency (5.6 Hz) (> Figure 78-10). Other tremor cells could be found that had rhythmic neuronal activity coherent with the foot tremor at 5.6 Hz. Discordant tremor frequencies in the upper and lower limbs have been known since the early work of Jung [212]. Although further study of these cells is required, there are at least three significant implications of these observations. First, they tend to support the existence of multiple parallel loops at the output level of the human basal ganglia, as demonstrated by primate neuroanatomical tracing techniques [213] and
. Figure 78-10 Example showing the results of spectral analysis (Fourier transform) of the firing of a neuron in GPi and of the EMG of wrist extensors and flexors and of the foot dorsiflexor (tibialis anterior) recorded during the same time interval. Note that the firing of this GPi tremor cell was at the same rate (4.6 Hz) as that of the wrist tremor but not that of the foot tremor.
Microelectrode recording in functional neurosurgery
neurophysiological studies [214,215]. This is in contrast to a pacemaker hypothesis that conceptualizes the pallidum as a functional syncytium where all units have the same frequency and serve as a central generator of an efferent oscillatory drive. Instead, individual tremor cells in the human pallidum appear to show discrete, limbspecific frequencies. Second, they do not support the hypothesis of a peripheral origin of parkinsonian rest tremor via ‘‘long-latency’’ spinal reflex arcs [108], since this hypothesis would predict that tremor in the lower extremity should be at a lower frequency than that in the upper extremity (longer loop time). Third, the frequency of tremor cells in the pallidum is in the same range as the limb tremor: 4–6 Hz, not 12–15 Hz, as predicted by the Llinas and Pare hypothesis [216,217]. This information, along with observations of 4–6 Hz neuronal activity recorded in the sensorimotor cortex of monkeys with tremor [218], tends to support a corticopallidothalamic loop mechanism for parkinsonian tremor generation. Pallidal tremor cells are not uniformly distributed throughout GP but appear to be located mainly or exclusively in GPi. In three patients with idiopathic Parkinson’s disease, 28 tremor cells were localized to the ventral portion of GPi [219]. This suggests that the distribution of tremor cells may be coextensive with the sensorimotor portions of GPi and that the so-called direct pathway from striatum to GPi that is part of a loop including the motor thalamus and motor cortex is probably one neural substrate of parkinsonian tremor. The paucity of tremor cells in GPe suggests that the indirect pathway is not primarily involved in the propagation of neural oscillations of tremor but may play a role in the generation of tremor by means of disinhibition of the subthalamic nucleus and excitation of GPi, as hypothesized in models of basal ganglia dysfunction in Parkinson’s disease [220,221], it is also possible that the elevated firing rate observed in GPi discussed above is related to the tremorogenic mechanism, since we have
78
observed elevated firing rates (95 29 Hz; SD, n = 22) in tremor cells compared to the average firing rate of GPi neurons (67 30 Hz, n = 162) in six patients with Parkinson’s disease [104]. In conclusion, the studies discussed here provide further evidence to support the corticopallidothalamic pathway as a substrate for the central generation of parkinsonian rest tremor.
Cerebral Cortex Observations concerning speech have been mentioned above. Various authors have reported microelectrode recordings of cortical neurons with reference to epileptic activity [38,222–224] and other brain functions [225] but these subjects will not be pursued here.
Cingulum Our group [226] recorded neurons in cingulate cortex that displayed widespread bilateral RFs and responded to noxious stimuli and cooling, complementing a number of published observations implicating this structure in nociception [227]. It was noted that 3 months after cingulotomy, the patient’s assessment of nonnoxious warm stimuli was lessened while that of noxious heat and cold was accentuated [228].
Meningeal Stimulation Referred pain can be induced easily by manipulation of the dura during a wake craniotomies but also during stereotactic subcortical explorations. During mesencephalic tractotomy, when an electrode located near the dorsal surface of the midbrain is stimulated electrically or advanced to distort the meninges, pain is commonly elicited, virtually always referred to the ipsilateral face (40% of sites) or forehead (22%)
1307
1308
78
Microelectrode recording in functional neurosurgery
[17]. Such responses are useful in that they help localize the electrode with respect to the surface of the brain stem, and awareness of such responses avoids confusion.
Miscellaneous Responses Alterations of consciousness and gustatory and olfactory effects are rarely elicited in subcortical exploration [17,23,26,188,191] in contrast to the plethora of experiential effects seen in association cortex. > Table 78-5 lists the rare responses seen by us with macrostimulation at 10,000 sites.
Other Structures Schaltenbrand and Wahren [27] have extensive experience with macrostimulation in the human brain stem that will not be reviewed here.
Applications of Physiological Localization Deep Brain Stimulation This section briefly reviews the use of physiological localization in commonly performed . Table 78-5 Autonomic effects observed under general anesthesia in the hypothalamic region Effect Apnea or depressed rate or amplitude of respiration Increased respiratory rate Elevation of blood pressure Depression of blood pressure Elevation of pulse rate Depression of pulse rate No autonomic effects
functional stereotactic procedures, many directed toward the thalamus. > Table 78-6 summarizes the clinicopathological relationships. The simplest procedure is the implantation of a chronic stimulating electrode to produce paresthesias in the distribution of a patient’s chronic pain syndrome, a procedure that appears to us to be most useful in cases of neuropathic pain [229,230]. DBS has a long history [191], with a rekindling of interest after the proposition of the Melzack-Wall gate theory of pain [231] as a cephalad extension of dorsal column stimulation [232,233]. The usual stereotactic approach is used. A micro- or macroelectrode is directed toward the sites in tactile Vc that are related to the painful part of the body. The use of microelectrodes allows accurate confirmation of location by recording responses to touch. The use of macroelectrodes to identify low-threshold sites for evoking paresthetic responses is less accurate because of the large size of the electrode tip and the resulting greater spread of stimulus current away from the tip. Once the appropriate tactile cells have been suitably mapped, a chronic electrode is introduced into their midst whose stimulation produces paresthesias in the patient’s area of pain. If large portions of the contralateral body are painful, the implant may have to be made in the medial lemniscus or internal capsule, which can be located by extrapolation from the position of tactile cells in Vc. Placement of DBS electrodes to chronically stimulate PVG was mentioned above.
No. Patients 131* 9 33 23 21 14 86
*More than one type of response occurred at some sites
Thalamotomy for Pain Destructive lesions for the relief of pain are no longer made in the lateral thalamus [28,234] except possibly in Vcpc, the identification of which has been reviewed sufficiently [17,64, 126,131,137,235].
Microelectrode recording in functional neurosurgery
78
. Table 78-6 The clinical neurophysiology of the thalamus Method of study Microelectrode Nucleus
Stimulation
Recording
Application
Ventrocaudal (Vc)
Paresthesia
Tactile neurons
Parvocellular ventrocaudal (Vcpcfvmpo) Anterior ventral oral (Voa) Posterior ventral oral (vop) Ventral intermediate (Vim) Anterior (A)
Warm, cool, painful effects and paresthesia
Nociceptors Thermoreceptors
Lesions (historical) and chronic stimulation for pain relief Lesions for pain relief
None recognized
Voluntary cells
Lesions for relief of dyskinesia
Motor ‘‘on’’ effects (see text) Paresthesia, vestibular, sensorimotor effects None recognized
Voluntary cells
Parafascicular (Pf) Periventricular gray (PVG), medial Pf Internal lamina (IL) Pulvinar (Pu)
None Various, acute pain relief, feeling of satiety None or pain or burning None recognized
Nociceptors None recognized
Lesions and chronic stimulation for relief of dyskinesia Lesions and chronic stimulation for control of tremor and dyskinesia Lesions for relief of psychiatric disease (historical) Lesions for pain relief Chronic stimulation for pain relief
Dorsomedian (DM)
None recognized
Bursting cells
Centrum medianum (CM)
None, pain
None recognized
Kinesthetic, and deep sensory neurons None recognized
Nociceptors Various
Medial thalamotomy probably is still widely employed. The plethora of potential medial thalamic targets has been mentioned, along with the difficulty of obtaining useful localizing responses in them [3,28,137,154,156,234,236]. Because of the difficulty in physiologically recognizing medial thalamic structures, we extrapolate the location of our target, parafascicular nucleus from the location of Vc. Frank and coauthors [237] pointed out that though the risks of medial thalamotomy are lower than those of mesencephalic tractotomy, so is the success rate in relieving nociceptive pain usually caused by cancer. Medial thalamotomy has not proved very useful in the treatment of steady neuropathic pain [138,154,156] and has been abandoned for the treatment of movement disorders [238].
Lesions for pain relief Lesions for relief of pain and spasticity (historical) Lesions for relief of psychiatric disorders (historical) Lesions for relief of pain and movement disorders (historical)
Mesencephalic Tractotomy Mesencephalic tractotomy, first by open and then by stereotactic means [31], has long been employed to treat chronic pain, functioning as a rostral extension of spinothalamic tractotomy in the spinal cord [6,17,31,125,165,176,190,228]. We have used macro- or microstimulation to localize lesions for mesencephalic tractotomy. First, the expected position of the medial lemniscus is determined from a stereotactic CT or MRI with the usual stereotactic technique. Stimulation is then carried out, searching for the low-threshold contralateral paresthetic effects typical of medial lemniscus in a series of parasagittal trajectories 2–3 mm apart and approximately 12 mm from the midline. Once the
1309
1310
78
Microelectrode recording in functional neurosurgery
medial lemniscus has been mapped out, further trajectories are made 2–3 mm medial and a few millimeters dorsal to it until a series of points are found where contralateral warm or cold effects, like those seen during cordotomy, are elicited, usually 8–10 mm from the midline. The lesion is planned to destroy the loci of these responses, sparing the medial lemniscus but extending medially almost to the aqueduct to destroy the normally physiologically ‘‘silent’’ region of the spinoreticular pathway as well.
Thalamotomy and Thalamic DBS for Tremor It would appear that exactly the same target is used for the implantation of a chronic stimulating electrode for the relief of tremor as was used in the past for thalamotomy [13,16,40,42, 109,113,239]. Our technique in either procedure is to first locate the tactile representation of the manual digits (about 15 mm from the midline), defining their anterior and inferior boundaries. We then search 2 mm more anterior for tremor cells and for neurons that alter their firing in response to passive or voluntary movement of the tremorous part of the body. Among these tremor cells we locate the sites where stimulation most completely abolishes tremor at the lowest threshold without inducing disturbing paresthesias. The final target must meet these criteria but also must lie 2–3 mm above the AC-PC line (see chapter 92).
the posteroventral portion of the GPi. Leksell considered pallidotomy more beneficial than thalamotomy for bradykinesia and rigidity whereas thalamotomy proved superior for the treatment of tremor. Recently, the effectiveness of posteroventral pallidotomy has been confirmed by clinical assessment of movement disorders carried out by our group [241]. We have summarized (see above) our results with microelectrode recording of single units in GP and microstimulation of adjacent neural structures to be avoided during pallidotomy (see chapter 92). The coordinates of GPi are determined from the stereotactic MRI [12] with reference to an atlas in the 20-mm sagittal plane. The microelectrode is lowered through GPe and GPi and on into the optic tract, recording continuously and stimulating every millimeter near the ventral portion of the pallidum. Cells typical of GPe and GPi are sought as well as any tremor cells or cells that alter their firing in response to contralateral and/or ipsilateral movement (> Figures 78-7 and > 78-8) (see above). The anterior, posterior, and inferior margins of GPi are readily defined by the cessation of background activity, the optic tract by stimulation-induced phosphenes and recording of responses to light and the internal capsule by the induction of tetanizing contralateral facial and upper limb responses. A lesion is then placed to destroy the maximum volume of GPi without encroaching on the motor capsule or optic tract.
Part 2 – Imaging in Structural Imaging Pallidotomy The current interest in stereotactic pallidotomy for the treatment of Parkinson’s disease (PD) arose out of the report of Laitinen and associates [240] of the beneficial effect on bradykinesia and rigidity obtained with lesions directed to
The way of the future may revolve around imaging. Just as it is now possible with imaging to recognize a pathological structure for stereotactic biopsy, the hope is that functional targets eventually will be visualized similarly. Thus, intraoperative functional imaging may play a
Microelectrode recording in functional neurosurgery
major role in functional neurosurgery in the future. Indeed, there is currently controversy about whether functional stereotactic surgery can already be done accurately and safely enough using existing MRI techniques with noninvasive radiosurgery [242,243] (see chapter 93). However, current opinion is that the existing imaging techniques are not sufficiently accurate and require corroboration with invasive physiological studies. In fact, there is still controversy about the relative accuracy of CT and MRI compared with ventriculography for functional stereotactic surgery [244], with some surgeons claiming that ventriculography is currently the most accurate imaging technique despite its disadvantages (> Table 78-2). We and others, however, have found CT and MRI to be at least as accurate as ventriculography for functional stereotactic surgery [245,246], and our unpublished experience clearly shows that imaging is more accurate using the Leksell G frame and a 1.5T General Electric Signa scanner than it is with CT [246]. The superiority of MRI probably is at least partly due to the fact that both AC and PC can be readily visualized with MRI, while AC usually cannot be seen well on CT. However, accurate MRI guidance depends on attention to a number of factors [33,247,248] (see chapter 18 and 39). The target structure should be near the center of the field, the chemical in the fiducial tubes should preferably not be petroleum jelly [249], and the frame used must be truly MRI compatible. To address some of these problems when we use MRI to calculate the coordinates of AC and PC (the usual landmarks on which most functional and stereotactic procedures are based), we first do a midsagittal scan to visualize them and then take 1-mm nonoverlapping axial images using the ‘‘spoiled-grass’’ sequence (excitation time, 13 ms; relaxation time, 43 ms). The single axial image best displaying AC and PC is chosen, and their three-dimensional coordinates are determined by using the MRI console’s
78
computer software. These coordinates are selected by bracketing the locations of AC and PC shown in the midsagittal scan. With the proper precautions outlined above, it seems to us that MRI is the structural imaging technique of choice for functional stereotactic surgery. A new type of structural imaging that may help with stereotactic localization is known as diffusion tensor imaging (DTI). This type of imaging can visualize white matter projections associated with (arising from or to) a specific gray matter structure.
Functional Imaging The last decade has seen the emergence of exciting new imaging techniques. Currently, these technologies are being used primarily for basic research. However, some of the limitations to clinical application will be overcome in the near future. Therefore, functional imaging has great potential as a new tool for neurosurgical procedures that rely on the localization of particular brain sites. Modern imaging technologies include functional magnetic resonance imaging (fMRI), positron emission tomography (PET), magnetoencephalography (MEG), and the combination of MEG and MRI techniques known as magnetic source imaging (MSI). A brief overview of the limitations and applicability of each technique for neurosurgical candidates follows.
Positron Emission Tomography PET is an imaging technique that relies on the ability to detect the emission of positively charged electrons. The source of this emission is typically an injectable molecule containing a positron-emitting isotope such as O15 incorporated into water molecules. The emissions (gamma radiation) from the injected material are captured by a ring of radiation detectors contained
1311
1312
78
Microelectrode recording in functional neurosurgery
in a doughnut-shaped PET device. Simultaneous acquisition of many parallel brain slices can be obtained using these emissions. The involvement of cortical and subcortical areas in somatosensory [33,35–38,250,251], pain [227,250,252,253], motor [254], and complex cognitive functions such as memory [255,256] has been investigated with PET technology. PET is particularly useful in patients who cannot be placed in an MR scanner (e.g., patients who are claustrophobic, have difficulty holding their head very still or who have metal in their body). PET can also be used to identify locations of receptor (dys)function [257,258]. However, PETspatial resolution (several millimeters) limits its use in identifying changes that affect small regions of the brain. Also, to achieve a high level of statistical significance, there is a need to average several images. This usually is achieved by pooling data from several subjects, since a limited amount of radioactivity can safely be injected in a single subject. These technical limitations currently limit the use of PET as a presurgical tool, although newer generation PET scanners with increased sensitivity allow for many more scans to be acquired in individual subjects.
Functional Magnetic Resonance Imaging FMRI is an imaging technique that is sensitive to the oxygen saturation of blood and blood flow. The ability of the microvasculature to supply oxygen by means of increased blood flow exceeds the ability of active neuronal tissue to extract oxygen. As a result, oxygen saturation of postcapillary blood from active neuronal areas is higher than that of inactive areas because of a normal flow ‘‘oversupply’’ to those areas. FMRI is capable of measuring and localizing differences in oxy- and deoxyhemoglobin. Since these two forms of oxygen act differently when exposed to a magnetic field (it is actually the deoxyhemoglobin that is
being detected), brain maps can be generated to show areas of active tissue [259]. Since fMRI is a noninvasive technique that can acquire images quickly (seconds per scan) with fine spatial resolution (1–5 mm of tissue) in a single subject [260]. it can be used with some precision to visualize sites activated by brief stimuli in the active human brain. These features are well suited for the examination of somatosensory and other cognitive processes. The ability to develop protocols that can test brain function has clinical applications such as the localization of functional tissue in tumor patients and presurgical mapping of active zones. The standard method of acquiring fMRI data today is with echo planar imaging (EPI), which can simultaneously image multiple brain regions very rapidly (tens of milliseconds per image). Since EPI has nearly real-time functional imaging capabilities, it is especially well suited for studies of functional systems that change with prolonged or repetitive activation and/or involve multiple spatially segregated brain sites. Thus, FMRI has been used successfully to image motor-related basal ganglia activation [261] and cortical activation associated with visual [262,263], motor [263–266], and somatosensory [267–269] stimuli in normal subjects. Other, more complex functions, such as language [270,271] and pain [269] also have been investigated. More recently, the technique of real-time fMRI has been introduced and made available on most clinical scanners. These ‘‘in-house’’ packages can be quite useful to track brain activations during simple tasks (e.g., motor), but should be used with caution for more complex situations due to the simplicity of the statistical analysis being used ‘‘on the fly’’ to generate activation maps in real time [272]. The ability to map functional cortical tissue physiologically has clinical relevance to presurgical localization. Functional imaging has been used to investigate motor, somatosensory, visual, and language function in patients before neurosurgery. Mapping with FMRI has been shown to corroborate the results of intraoperative mapping with
Microelectrode recording in functional neurosurgery
macrostimulation [268] [273] and thus may become quite useful for procedures that do not have to resolve fine subnuclear borders. One study [271] demonstrated activation of Broca’s area during internal speech generation. Hemispheric laterality was also examined in a study of motor cortex activation in left- versus right-handed subjects [266]. These examples illustrate the usefulness of presurgical mapping with fMRI, although undoubtedly there would still be the need for verification of these data during surgery. Hemispheric dominance as revealed by FMRI could replace the Wada test used before certain neurosurgical procedures to determine the laterality of functions. In the future, intraoperative MRI may provide a means to perform noninvasive functional imaging at the time of surgery. Despite great advances in MRI, there are still some limitations to its application with the current technology. First, the spatial resolution of fMRI is not yet fine enough to distinguish very small regions of the brain such as subcortical nuclei. Therefore, FMRI cannot replace intraoperative electrophysiological mapping as is done before procedures such as thalamotomy and pallidotomy. Second, movement artifact resulting from head movements and motion artifact caused by natural internal oscillations (e.g., spinal cord movement) preclude imaging of some central nervous system (CNS) regions. Third, it may not be possible for some patients to be placed in an MR scanner for the 20–40 min necessary for a FMRI series and MRIs cannot be done in patients with implanted metal devices.
Magnetic Source Imaging MEG is a technique similar to EEG that measures the electric currents in the brain, but with greater spatial accuracy. MSI combines MEG with MRI for anatomic localization of the magnetic signals in a functional setting. The exciting features of MSI are that it is easy to use, is noninvasive, and has
78
good spatial resolution (millimeter range) and very fast acquisition time, allowing for real-time imaging [274,275]. MSI has been used to study many cortical regions involved in audition [276] and vibratory/touch sensation [276–279]. The high spatial resolution of MSI has demonstrated distinct representations for different body regions (e.g., digits, hand, arm, face) in the somatosensory cortex [276–279]. Cortical plasticity resulting from injury has also been demonstrated [280]. MSI can also be combined with magnetic resonance angiography (MRA) to provide visual demarcation of important functional regions before neurosurgery. Some studies have demonstrated very good agreement of MSI data with operative results [275]. Therefore, MRA and MSI are two techniques that may provide useful information before neurosurgical procedures involving the cortex. At this time, these techniques are not applicable to subcortical studies. Also, biomagnetometers are not as widely available as are MRI and PET facilities.
Prediction for Future Protocols These new imaging technologies will no doubt affect neurosurgery in the future. As these techniques are refined and become more widely available, they may provide safe alternative methods for the physiological mapping of viable tissue. This could reduce the need for timeconsuming and invasive modes of mapping. Although the current tools for mapping, such as electrophysiological recording and stimulation, probably will still be required in certain instances, the advent of intraoperative imaging married to functional imaging techniques may provide alternatives for certain patients.
Conclusion Although the hope is that in the future imaging, particularly functional imaging, will allow
1313
1314
78
Microelectrode recording in functional neurosurgery
a stereotactic surgeon to recognize whatever target he or she wishes to manipulate and allow a direct approach similar to that currently used in stereotactic biopsy, for the present, invasive techniques of physiological monitoring are necessary for optimal results with minimal complications. If carried out methodically in the manner of brain mapping as originally advocated by Woolsey, the physiological monitoring not only will satisfy the surgeon’s needs but also will provide further knowledge about the brain’s normal and pathological states.
References 1. Penfield W, Boldrey W. Somatic motor and sensory representation in the cerebral cortex of man as studied by electrical stimulation. Brain 1937;60:389. 2. Albe-Fessard D. Electrophysiological methods for the identification of thalamic nuclei. Z Neurol 1973;205:15-28. 3. Ervin F, Mark V. Stereotactic thalamotomy in the human. II. Physiological observations in the human thalamus. Arch Neurol 1960;3:368. 4. Fairman D, Perlmutter I. Physiological observations during stereotactic surgery of the basal ganglia. Confin Neurol 1965;26:298-305. 5. Hughes J. The electroencephalogram in Parkinsonism. J Neurosurg 1966;24:369. 6. Spiegel E, Wycis H. Clinical and physiological applications. I Stereoencephalotomy. New York: Grune & Stratton; 1962. 7. Lenz FA, Dostrovsky JO, Kwan HC, Tasker RR, Yamashiro K, Murphy JT. Methods for microstimulation and recording of single neurons and evoked potentials in the human central nervous system. J Neurosurg 1988;68:630-4. 8. Dostrovsky JO, Davis KD, Lee L, Sher GD, Tasker RR. Electrical stimulation-induced effects in the human thalamus. In: Devinsky O, Beric A, Dogali M, editors. Electrical and magnetic stimulation of the brain and spinal cord. New York: Raven Press; 1993. p. 219-29. 9. Bertrand G, Jasper H, Wong A, Mathews G. Microelectrode recording during stereotactic surgery. Clin Neurosurg 1969;16:328-55. 10. Jasper H. Recording from microelectrodes in stereotactic surgery for Parkinson’s disease. J Neurosurg 1966;2 suppl: 219. 11. Jasper H, Bertrand G. Exploration of the human thalamus with microelectrodes. Physiologist 1963;7:167.
12. Lozano AM, Hutchison WD, Kiss ZHT, Davis KD, Dostrovsky JO. Methods for microelectrode-guided posteroventral pallidotomy. J Neurosurg 1996;84:194-202. 13. Tasker RR, Lenz FA, Yamashiro K, Gorecki J, Hirayama T, Dostrovsky JO. Microelectrode techniques in localization of stereotactic targets. Neurol Res 1987;9:105-12. 14. Tasker RR. The use of microelectrodes in the human brain. In: Bromm B, Desmedt EF, editors. Pain and the brain: from nociception to cognition. New York: Raven Press; 1995. p. 143-74. 15. Tasker R, Dostrovsky JO, Yamashiro K. Effets sensitifs et moteurs de la stimulation thalamique chez l’homme: applications clinique. Rev Neurol (Paris) 1986;142:316. 16. Tasker RR, Yamashiro K, Lenz FA. Thalamotomy for Parkinson’s disease: microelectrode techniques. In: Lunsford LD, editor. Modern stereotactic surgery. Boston: Martinus Nijhoff; 1988. p. 297-314. 17. Tasker R, Organ L, Hawrylyshyn P. The thalamus and midbrain of man. A physiological atlas using electrical stimulation. Springfield, IL: Charles C Thomas; 1982. 18. Fukamachi A, Oye C, Narabayashi H. Delineation of the thalamic nuclei with a microelectrode in stereotaxic surgery for parkinsonism and cerebral palsy. J Neurosurg 1973;39:214-25. 19. Albe-Fessard D, Guiot G, Hardy J. Electrophysiological localization and identification of subcortical structures in man by recording spontaneous and evoked activities. Electroencephalogr Clin Neurophysiol. 1963;15:1052. 20. Narabayashi H. Tremor mechanisms. In: Schaltenbrand G, Walker AE, editors. Stereotaxy of the human brain. Stuttgart: George Thieme; 1982. p. 510-14. 21. Bertrand C, Martinez SN, Hardy J. Stereotactic surgery for parkinsonism: microelectrode recording, stimulation, and oriented sections with a leucotome. In: Krayenbuhl H, Maspes PE, Sweet WH, editors. Progress in neurological surgery. Basel: Karger; 1973. p. 79-112. 22. Guiot G, Derome P, Arfel G. Electrophysiological recordings in stereotaxic thalamotomy for parkinsonism. In: Krayenbuhl H, Maspes PE, Sweet WH, editors. Progress in neurological surgery. Basel: Karger; 1962. p. 189-221. 23. Hassler R, Riechert T, Mundinger F, Umbach W, Ganglberger JA. Physiological observations in stereotaxic operations in extrapyramidal motor disturbances. Brain 1960;83:337-50. 24. Andy OJ. Sensory-motor responses from the diencephalon. Electrical stimulation in man. J Neurosurg 1966;24:612-20. 25. Bertrand C. Functional localization with monopolar stimulation. J Neurosurg 1966;24:403. 26. Sem-Jacobsen CW, Styri OB. Electrical stimulation of the human brain: pitfalls and results. In: Somjen GG, editor. Neurophysiology studied in man. Amsterdam: Excerpta Medica; 1972. p. 27-36.
Microelectrode recording in functional neurosurgery
27. Schaltenbrand G, Wahren W. Atlas for stereotaxy of the human brain. Georg Thieme; Stuttgart: 28. Monnier M, Fischer R. [Localization, stimulation and coagulation of thalamus in man]. J Physiol (Paris) 1951;43:818. 29. Ito Z. Stimulation and destruction of the prelemniscal radiation or its adjacent area in various extrapyramidal disorders. Confin Neurol 1975;37:41-8. 30. Albe-Fessard D, Arfel G, Guiot G, Derome P, Guilbaud G. Thalamic unit activity in man. Electroencephalogr Clin Neurophysiol 1967;25:132. 31. Woolsey CN. Cortical localization as defined by evoked potential and electrical stimulation studies. In: Schaltenbrand G, Woolsey CN, editors. Cerebral localization and organization. Madison, WI: University of Wisconsin Press; 1964. p. 17-26. 32. Wetzel N, Snider RS. Neurophysiological correlates in human stereotaxis. Q Bull Northwest Univ Med Sch 1958;32:386-92. 33. Carreras M, Mancia D, Pagni CA. Unit discharges recorded from the human thalamus with microelectrodes. Confin Neurol 1967;29:87-92. 34. Albe-Fessard D, Arfel G, Guiot G. Derivations d’activites spontanees et evoquees dans les structures cerebrales profondes de l’homme. Rev Neurol (Paris) 1962;106:89. 35. Albe-Fessard D, Arfel G, Guiot G. Activites elecctriques caracteristiques de quelques structures cerebrales chez l’homme. Ann Chir 1963;17:1185. 36. Albe-Fessard D, Arfel G, Guiot G, Derome P, Hertzog E, Vourc’h, G, Brown H, Aleonard P, De la HJ, Trigo JC. Electrophysiological studies of some deep cerebral structures in man. J Neurol Sci 1966;3:37-51. 37. Albe-Fessard D. Thalamic activity and Parkinson’s disease. In: Umbach W, Koepchen HP, editors. Central rhythm and regulation. Stuttgart: Hyppocrates Verlag; 1974. p. 353-62. 38. LI CL, Van Buren JM. Micro-electrode recordings in the brain of man with particular reference to epilepsy and dyskinesia. In: Somjen GG, editor. Neurophysiology studied in man. Amsterdam: Excerpta Medica; 1972. p. 49-63. 39. Lucking CH, Struppler A, Erbel F, et al. Stereotactic recording from human subthalamic structures. In: Somjen GG, editor. Neurophysiology studied in man. Excerpta Medica; Amsterdam: p. 95-9. 40. Ohye C, Fukamachi A, Miyazaki M, Isobe I, Nakajima H, Shibazaki T. Physiologically controlled selective thalamotomy for the treatment of abnormal movement by Leksell’s open system. Acta Neurochir (Wien.) 1977;37:93-104. 41. Ohye C. Depth microelectrode studies. In: Schaltenbrand G, Walker AE, editors. Stereotaxy of the human brain. Stuttgart: Thieme; 1982. p. 510-14. 42. Ohye C. Selective thalamotomy for movement disorders: micro-recording stimulation techniques and results. In:
43.
44.
45.
46.
47.
48.
49.
50. 51.
52.
53.
54.
55.
56.
57. 58.
78
Lunsford LD, editor. Modern stereotactic neurosurgery. Boston: Martinus-Nijhoff; 1988. p. 315-31. Umbach W, Ehrhardt KJ. [Microelectrode leads in the basal ganglia in man]. Arch Psychiatr Nervenkr 1965;207:106-13. Arutjunov A, Kadin A, Konovalov A. Experience of applying the microelectrode method in the neurosurgical clinic practice during single-stage stereotaxic operations for parkinsonism. Vopr Neirokhir 1970;6:12. Bates JAV. Electrical recording from the thalamus in human subjects. In: Iggo A, editor. Handbook of sensory physiology: somatosensory system. Berlin: Springer; 1972. p. 561-78. Bertrand G, Jasper H. Microelectrode recording of unit activity in the human thalamus. Confin Neurol 1965;26:205-8. Bertrand G, Jasper H, Wong A. Microelectrode study of the human thalamus: functional organization in the ventro-basal complex. Confin Neurol 1967;29:81-6. Donaldson IM. The properties of some human thalamic units. Some new observations and a critical review of the localization of thalamic nuclei. Brain 1973;96:419-40. Gaze RM, Gillingham FJ, Kalyanaraman S, Porter RW, Donaldson AA, Donaldson IM. Microelectrode recordings from the human thalamus. Brain 1964;87:691-706. Gillingham F. Depth recording and stimulation. J Neurosurg 1966;2 382. Guiot G, Hardy J, Be-Fessard D. [Precise delimitation of the subcortical structures and identification of thalamic nuclei in man by stereotactic electrophysiology]. Neurochirurgia (Stuttg) 1962;5:1-18. Fukamachi A, Ohye C, Saito Y, Narabayashi H. Estimation of the neural noise within the human thalamus. Acta Neurochir (Wien) 1977; 121–36. Nakajima H, Fukamachi A, Isobe I, Miyazaki M, Shibazaki T, Ohye C. Estimation of neural noise. Functional anatomy of the human thalamus. Appl Neurophysiol 1978;41:193-201. Yoshida M. Electrophysiological characterization of human subcortical structures by frequency spectrum analysis of neural noise (field potential) obtained during stereotactic surgery. Preliminary presentation of frequency power spectrum of various subcortical structures. Stereotact Funct Neurosurg 1989;52:157-63. Tasker RR, Rowe IH, Hawrylyshyn P, Organ L. Computer mapping of brain-stem sensory centers in man. J Neurosurg 1976;44:458-64. Tasker RR, Dostrovsky JO. Computers in functional stereotactic surgery. In: Kelly PJ, Kall B, editors. Computers in stereotactic neurosurgery. Boston: Blackwell; 1992. p. 154-64. Schaltenbrand G, Baily P. Introduction to stereotaxis with an atlas of the human brain. Stuttgart: Thieme; 1959. Sugita K, Takaoka Y, Mutsuga N, Takeda A, Hirota T. Correlation between anatomically calculated target point
1315
1316
78 59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
Microelectrode recording in functional neurosurgery
and physiologically determined point in stereotaxic surgery. Confin Neurol 1972;34:84-93. Ohye C, Nakamura R, Fukamachi A. Recording and stimulation of the ventralis nuclei of the human thalamus. Confin Neurol 1975;37:258. Organ LW, Tasker RR, Moody NF. The impedance profile of the human brain as a localization technique in stereoencephalotomy. Confin Neurol 1967;29:192-6. Dostrovsky JO, Sher GD, Davis KD, Parrent AG, Hutchison WD, Tasker RR. Microinjection of lidocaine into human thalamus: a useful tool in stereotactic surgery. Stereotact Funct Neurosurg 1993;60:168-74. French LA, Story JL, Galicich JH, Schultz EA. Some aspects of stimulation and recording from the basal ganglia in patients with abnormal movements. Confin Neurol 1962;22:265-74. Narabayashi H, Ohye C. Nucleus ventralis intermedius of human thalamus. Trans Am Neurol Assoc 1974;99:232-3. Sano K, Yoshioka M, Sekino H, Mayanagi Y, Yoshimasu N. Functional organization of the internal medullary lamina in man. Confin Neurol 1970;32: 374-80. Yamashiro K, Tasker RR, Iwayama K, Mori K, be-Fessard D, Dostrovsky JO, Chodakiewitz JW. Evoked potentials from the human thalamus: correlation with microstimulation and single unit recording. Stereotact Funct Neurosurg 1989;52:127-35. Hardy TL, Bertrand G, Thompson CJ. Thalamic recordings during stereotactic surgery. II. Location of quickadapting touch-evoked (novelty) cellular responses. Appl Neurophysiol 1979;42:198-202. Hardy TL, Bertrand G, Thompson CJ. Touch-evoked thalamic cellular activity. The variable position of the anterior border of somesthetic SI thalamus and somatotopography. Appl Neurophysiol 1981;44: 302-13. Jasper HH, Bertrand G. Thalamic units involved in somatic sensation and voluntary and involuntary movements in man. In: Purpura DP, YahrMD, editors. The thalamus. New York: Columbia University Press; 1966. p. 365-90. Lenz FA, Dostrovsky JO, Tasker RR, Yamashiro K, Kwan HC, Murphy JT. Single-unit analysis of the human ventral thalamic nuclear group: somatosensory responses. J Neurophysiol 1988;59:299-316. Ohye C, Fukamachi A, Narabayashi H. Spontaneous and evoked activity of sensory neurons and their organization in the human thalamus. Z Neurol 1972;203:219. Lenz FA, Kwan HC, Martin R, Tasker R, Richardson RT, Dostrovsky JO. Characteristics of somatotopic organization and spontaneous neuronal activity in the region of the thalamic principal sensory nucleus in patients with spinal cord transection. J Neurophysiol 1994;72:1570-87. Davis KD, Kiss ZHT, Tasker RR, DostrovskyJO. Thalamic stimulation-evoked sensations in chronic pain
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
patients and in nonpain (movement disorder) patients. J Neurophysiol 1996;75:1026-37. Ohye C, Shibazaki T, Hirai T, Kawashima Y, Hirato M, Matsumura M. Plastic change of thalamic organization in patients with tremor after stroke. Appl Neurophysiol 1985;48:288-92. Goto A, Kosaka K, Kubota K, Nakamura R, Narabayashi H. Thalamic potentials from muscle afferents in the human. Arch Neurol 1968;19:302-9. Crowell RM, Perret E, Siegfried J, Villoz JP. ‘Movement units’ and ‘tremor phasic units’ in the human thalamus. Brain Res 1968;11:481-8. Hardy TL, Bertrand G, Thompson CJ. Position and organization of thalamic cellular activity during diencephalic recording. II. Joint- and muscle-evoked activity. Appl Neurophysiol 1980;43:28-36. Hardy TL, Bertrand G, Thompson CJ. Position and organization of thalamic cellular activity during diencephalic recording. I. Pressure-evoked activity. Appl Neurophysiol 1980;43:18-27. Jones MW, Tasker RR. The relationship of documented destruction of specific cell types to complications and effectiveness in thalamotomy for tremor in Parkinson’s disease. Stereotact Funct Neurosurg 1990; 54/55:207-11. Maeda T. Lateral coordinates of nucleus ventralis intermedius target for tremor alleviation. Stereotact Funct Neurosurg 1989;52:191-9. McClean MD, Dostrovsky JO, Lee L, Tasker RR. Somatosensory neurons in human thalamus respond to speech-induced orofacial movements. Brain Res 1990;513:343-7. Ohye C. Neurons of the thalamic ventralis intermedius nucleus: their special reference to tremor. Adv Neurol Sci 1985;29:224. Ohye C, Shibazaki T, Hirai T, Wada H, Hirato M, Kawashima Y. Further physiological observations on the ventralis intermedius neurons in the human thalamus. J Neurophysiol 1989;61:488-500. Ohye C, Narabayashi H. Physiological study of presumed ventralis intermedius neurons in the human thalamus. J Neurosurg 1979;50:290-7. Hardy TL, Bertrand G, Thompson CJ. Topography of ‘bilateral-movement-evoked’ thalamic cellular activity found during diencephalic recording. Appl Neurophysiol 1980;43:67-74. Hongell A, Wallin G, Hagbarth KE. Unit activity connected with movement initiation and arousal situations recorded from the ventrolateral nucleus of the human thalamus. Acta Neurol Scand 1973;49:681-98. Lenz FA, Kwan HC, Dostrovsky JO, Tasker RR, Murphy JT, Lenz YE. Single unit analysis of the human ventral thalamic nuclear group. Activity correlated with movement. Brain 1990;113(Pt 6):1795-821.
Microelectrode recording in functional neurosurgery
87. Raeva S. Localization in human thalamus of units triggered during ‘verbal commands,’ voluntary movements and tremor. Electroencephalogr Clin Neurophysiol 1986;63:160-73. 88. Raeva S. Unit activity of some deep nuclear structures of the human brain during voluntary movement. In: Somjen G, editor. Neurophysiology studied in man. Amsterdam: Excerpta Medica; 1972. p. 64-78. 89. Raeva SN. Role of neurons of the neostriatum and nonspecific thalamus in the organization of goal-directed activity. Hum Physiol 1977;3:972. 90. Raeva SN, Livanov MN. Microelectrode studies of neuronal mechanisms of human voluntary amnesic activity. Hum Physiol 1975;1:27. 91. Umbach W, Ehrhardt KJ. Micro-electrode recording in the basal ganglia during stereotaxic operations. Confin Neurol 1965;26:315-17. 92. Strick PL. Activity of ventrolateral thalamic neurons during arm movement. J Neurophysiol 1976;39: 1032-44. 93. Kelly PJ, Ahlskog JE, Goerss SJ, Daube JR, Duffy JR, Kall BA. Computer-assisted stereotactic ventralis lateralis thalamotomy with microelectrode recording control in patients with Parkinson’s disease. Mayo Clin Proc 1987;62:655-64. 94. Albe-Fessard D, Guiot G, Lamarre J, et al. Activation of thalamocortical projections related to tremorogenic processes. In: Purpura DP, Yahr MD, editors. The thalamus. New York: Columbia University Press; 1966. p. 237-53. 95. Bates JAV. The significance of tremor phasic units in the human thalamus. In: Gillingham FJ, Donaldson IML, editors. Third symposium on Parkinson’s disease. Edinburgh: Livingstone; 1996. p. 112-18. 96. Hardy J, Bertrand C, Martinez N. Activites cellulaire thalamiques liees au tremblement parkinsonien. Neurochirurgie 1964;10:449. 97. Hardy TL, Bertrand G, Thompson CJ. Thalamic recordings during stereotactic surgery. I. Surgery topography of evoked and nonevoked rhythmic cellular activity. Appl Neurophysiol 1979;42:185-97. 98. Lenz FA, Tasker RR, Kwan HC, Schnider S, Kwong R, Murphy JT. Cross-correlation analysis of thalamic neurons and EMG activity in parkinsonian tremor. Appl Neurophysiol 1985;48:305-8. 99. Lenz FA, Kwan HC, Martin RL, Tasker RR, Dostrovsky JO, Lenz YE. Single unit analysis of the human ventral thalamic nuclear group. Tremor-related activity in functionally identified cells. Brain 1994;117 (Pt 3):531-43. 100. Lenz FA, Schnider S, Tasker RR, Kwong R, Kwan H, Dostrovsky JO, Murphy JT. The role of feedback in the tremor frequency activity of tremor cells in the ventral nuclear group of human thalamus. Acta Neurochir Suppl (Wien) 1987;39:54-6.
78
101. Lenz FA, Tasker RR, Kwan HC, Schnider S, Kwong R, Murayama Y, Dostrovsky JO, Murphy JT. Single unit analysis of the human ventral thalamic nuclear group: correlation of thalamic ‘‘tremor cells’’ with the 3–6 Hz component of parkinsonian tremor. J Neurosci 1988;8:754-64. 102. Ohye C, Albe-Fessard D. Rhytmic discharges related to tremor in humans and monkeys. In: Chalazonitis N, Boisson M, editors. Abnormal neuronal discharges. New York: Raven Press; 1978. p. 37-48. 103. Ohye C, Saito U, Fukamachi A, Narabayashi H. An analysis of the spontaneous rhythmic and non-rhythmic burst discharges in the human thalamus. J Neurol Sci 1974;22:245-59. 104. Hutchison WD, Lozano AM, Kiss ZHT. Tremor-related activity (TRA) in globus pallidus of Parkinson’s disease (PD). Soc Neurosci 1994;20:783. 105. Lenz FA, Martin R, Kwan HC, Tasker RR, Dostrovsky JO. Thalamic single-unit activity occurring in patients with hemidystonia. Stereotact Funct Neurosurg 1990;54/55:159-62. 106. Schnider SM, Kwong RH, Kwan HC, et al. Detection of feedback in the central nervous system of Parkinson’s patients. IEEE Trans Decis Control 1989; 60:203. 107. Schnider SM, Kwong RH, Lenz FA, Kwan HC. Detection of feedback in the central nervous system using system identification techniques. Biol Cybern 1989;60:203-12. 108. Tatton WG, Lee RG. Evidence for abnormal long-loop reflexes in rigid Parkinsonian patients. Brain Res 1975;100:671-6. 109. Ohye C, Hirai T, Miyazaki M, Shibazaki T, Nakajima H. Vim thalamotomy for the treatment of various kinds of tremor. Appl Neurophysiol 1982;45:275-80. 110. Laitinen LV. Brain targets in surgery for Parkinson’s disease. Results of a survey of neurosurgeons. J Neurosurg 1985;62:349-51. 111. Yasui N, Narabayashi H, Kondo T, Ohye C. Slight cerebellar signs in stereotactic thalamotomy and subthalamotomy for parkinsonism. Appl Neurophysiol 1976;39:315-20. 112. Albe-Fessard D. Physio-pathologie de Parkinson. Presented at Le point sur la maladie de Parkinson symposium, Brussels, Belgium; 1973. 113. Lenz FA, Normand SL, Kwan HC, Andrews D, Rowland LH, Jones MW, Seike M, Lin YC, Tasker RR, Dostrovsky JO. Statistical prediction of the optimal site for thalamotomy in parkinsonian tremor. Mov Disord 1995;10:318-28. 114. Deiber MP, Pollak P, Passingham R, Landais P, Gervason C, Cinotti L, Friston K, Frackowiak R, Mauguie`re F, Benabid AL. Thalamic stimulation and suppression of parkinsonian tremor: evidence of
1317
1318
78 115.
116.
117.
118.
119.
120.
121.
122.
123.
124. 125.
126.
127.
128.
Microelectrode recording in functional neurosurgery
a cerebellar deactivation using positron emission tomography. Brain 1993;116:267-79. Jenkins IH, Bain PG, Colebatch JG, Thompson PD, Findley LJ, Frackowiak RS, Marsden CD, Brooks DJ. A positron emission tomography study of essential tremor: evidence for overactivity of cerebellar connections. Ann Neurol 1993;34:82-90. Shahzadi S, Tasker RR, Lozano A. Thalamotomy for essential and cerebellar tremor. Stereotact Funct Neurosurg 1995;65:11-17. Willis WD Jr. The pain system: the neural basis of nociceptive transmission in the mammalian nervous system. In: Gildenberg PL, editor. Pain and headache. Basel: Karger; 1995. p. 1-346. Apkarian AV, Hodge CJ. Primate spinothalamic pathways. III. Thalamic terminations of the dorsolateral and ventral spinothalamic pathways. J Comp Neurol 1989;288:493-511. Boivie J. An anatomical reinvestigation of the termination of the spinothalamic tract in the monkey. J Comp Neurol 1979;186:343-70. Craig AD, Bushnell MC, Zhang E.-T., Blomqvist A. A thalamic nucleus specific for pain and temperature sensation. Nature 1994;372:770-3. Mehler WR. The anatomy of the so-called ‘‘pain tract’’ in man. An analysis of the course and distribution of the ascending fibers of the fasciculus anterolateralis. In: French JD, Porter RW, editors. Basic research in paraplegia. Springfield, IL: Charles C. Thomas; 1962. p. 26-55. Hassler R. Dichotomy of facial pain conduction in the diencephalon. In: Hassler R, Walker AE, editors. Trigeminal neuralgia. Philadelphia: Sanders; 1970. p. 123-138. Blomqvist A, Berkley KJ. A re-examination of the spinoreticulo-diencephalic pathway in the cat. Brain Res 1992;579:17-31. Hitchcock E, Lewin M. Stereotactic recording from the spinal cord of man. Br Med J 1969;4:44-5. Amano K, Tanikawa T, Iseki H, Kawabatake H, Notani M, Kawamura H, Kitamura K. Single neuron analysis of the human midbrain tegmentum. Rostral mecencephalic reticulotomy for pain relief. Appl Neurophysiol 1978;41:66-78. Sano K. Intralaminar thalamotomy (thalamolaminotomy) and posteromedial hypothalamotomy in the treatment of intractable pain. In: Krayenbuhl H, Maspes PE, Sweet WH, editors. Progress in neurological surgery. Basel: Karger; 1977. p. 50-103. Ishijima B, Yoshimasu N, Fukushima T, Hori T, Sekino H, Sano K. Nociceptive neurons in the human thalamus. Confin Neurol 1975;37:99-106. Lenz FA, Seike M, Lin YC, Baker FH, Rowland LH, Gracely RH, Richardson RT. Neurons in the area of human thalamic nucleus ventralis caudalis respond to painful heat stimuli. Brain Res 1993;623:235-40.
129. Lenz FA, Gracely RH, Rowland LH, Dougherty PM. A population of cells in the human thalamic principal sensory nucleus respond to painful mechanical stimuli. Neurosci Lett 1994;180:46-50. 130. Dostrovsky JO, Davis KD, Kiss ZHT, et al. Evidence for a specific temperature relay site in human thalamus. In IASP World Congress on Pain 1996. 131. Halliday AM, Logue V. Painful sensations evoked by electrical stimulation in the thalamus. In: Somjen GG, editor. Neurophysiology studied in man. Amsterdam: ExcerptsMedica; 1972. p. 221-30. 132. Hassler R. The division of pain conduction into systems of pain sensation and pain awareness. In: Janzen R, Keidl WD, Herz A, Steichele C, editors. Pain; Basic Principles-pharmacology-therapy. London: Churchill Livingstone; 1972. p. 98-112. 133. Davis KD, Tasker RR, Kiss ZH, Hutchison WD, Dostrovsky JO. Visceral pain evoked by thalamic microstimulation in humans. Neuroreport 1995; 6:369-74. 134. Dostrovsky JO, Wells FEB, Tasker RR. Pain sensations evoked by stimulation in human thalamus. In: Inoka R, Shigenaga Y, Tohyama M, editors. Processing and inhibition of nociceptive information, international congress series 989. Amsterdam: Excerpta Medica; 1992. p. 115-20. 135. Lenz FA, Seike M, Richardson RT, Lin YC, Baker FH, Khoja I, Jaeger CJ, Gracely RH. Thermal and pain sensations evoked by microstimulation in the area of human ventrocaudal nucleus. J Neurophysiol 1993;70:200-12. 136. Lenz FA, Gracely RH, Hope EJ, Baker FH, Rowland LH, Dougherty PM, Richardson RT. The sensation of angina can be evoked by stimulation of the human thalamus. Pain 1994;59:119-25. 137. Hitchcock ER, Teixeira MJ. A comparison of results from center-median and basal thalamotomies for pain. Surg Neurol 1981;15:341-51. 138. Tasker RR. Deafferentation pain syndromes. Introduction. In: Nashold BS Jr, Ovelmen-Levitt J, editors. Deafferentation pain syndromes: pathophysiology and treatment. Advances in pain research and therapy. New York: Raven Press; 1991. p. 241-57. 139. Gorecki J, Hirayama T, Dostrovsky JO, et al. Thalamic stimulation and recording in patients with deafferentation and central pain. Stereotact Funct Neurosurg 1989;52:220. 140. Lenz FA, Tasker RR, Dostrovsky JO. Abnormal single unit activity and responses to stimulation in the presumed ventrocaudal nucleus of patients with central pain. In: Dubner R, Gebhart GF, Bond MR, editors. Proceedings of the 5th world congress on pain, vol 18. Amsterdam: Elsevier; 1988. p. 157-64. 141. Lenz FA, Tasker RR, Dostrovsky JO, Kwan HC, Gorecki J, Hirayama T, Murphy JT. Abnormal single-unit activity recorded in the somatosensory thalamus of
Microelectrode recording in functional neurosurgery
142.
143.
144. 145.
146.
147. 148.
149.
150.
151.
152.
153.
154.
155.
156.
a quadriplegic patient with central pain. Pain 1987;31:225-36. Luo L, Davis KD, Kiss ZHT, et al. Phantom sensations elicited by stimulation in thalamus of amputees. In Proceedings of the 8th world congress on pain, Vancouver, BC, Canada,1996. p. 306. Parent AG, Lozano AM, Dostrovsky JO, et al. Central pain in the absence of functional sensory thalamus. Stereotact Funct Neurosurg 1992;59:9. Ralston BL. Hemispherectomy and hemithalamectomy in man. J Neurosurg 1962;19:909-12. McCarley RW, Benoit O, Barrionuevo G., Lateral geniculate nucleus unitary discharge in sleep and waking: state and rate specific aspects. J Neurophysiol 1983;50:798-818. Dostrovsky JO, Tsoukatos J, Kiss ZHT, et al. Neuronal firing patterns in human thalamus during sleep and wakefulness. Physiol Can 1995;26:83. Steriade M, Jones EG, Llinas RR. Thalamic oscillations and signalling. New York: Wiley; 1990. Loeser JD, Ward AAJ, White LE Jr.Chronic deafferentation of human spinal cord neurons. J Neurosurg 1968;29:48-50. Yamashiro K, Iwayama K, Kurihara M, Mori K, Niwa M, Tasker RR, be-Fessard D. Neurones with epileptiform discharge in the central nervous system and chronic pain. Experimental and clinical investigations. Acta Neurochir Suppl (Wien) 1991;52:130-2. Tsoukatos J, Davis KD, Loomis S, et al. Is neuronal bursting activity in lateral thalamus of chronic pain patients related to their pain? In Proceedings of the 8th world congress on pain, Vancouver, BC; 1996. p. 305. Tsoukatos J, Kiss ZHT, Tasker RR, et al. Distinct patterns of bursting activity in the human medial and lateral thalamus. Soc Neurosci Abstr 1995;21:109. Lenz FA, Kwan HC, Dostrovsky JO, Tasker RR. Characteristics of the bursting pattern of action potentials that occurs in the thalamus of patients with central pain. Brain Res 1989;496:357-60. Davis KD, Kiss ZH, Tasker RR, Dostrovsky JO. Thalamic stimulation-evoked sensations in chronic pain patients and in nonpain (movement disorder) patients. J Neurophysiol 1996;75:1026-37. Tasker RR, Dostrovsky JO. Deafferentation and central pain. In: Melzack R, Wall PD, editors. Textbook of pain. Edinburgh: Churchill Livingstone; 1989. p. 154-80. Parent AG, Lozano AM, Tasker RR, et al. Periventricular gray stimulation suppresses allodynia and hyperpathia in man. Stereotact Funct Neurosurg 1992;59:82. Tasker RR. Pain resulting from central nervous system pathology (central pain). In: Bonica J, editor. Management of pain. Philadelphia: Lea & Febiger; 1989. p. 264-83.
78
157. Tasker RR, Organ LW. Stimulation-mapping of the upper human auditory pathway. J Neurosurg 1973;38:320-25. 158. Hawrylyshyn PA, Rubin AM, Tasker RR, Organ LW, Fredrickson JM. Vestibulothalamic projections in man– a sixth primary sensory pathway. J Neurophysiol 1978;41:394-401. 159. Marg E, Dierssen G. Reported visual percepts from stimulation of the human brain with microelectrodes during therapeutic surgery. Confin Neurol 1965;26:57-75. 160. Boethius J, Lindblom U, Meyerson BA, et al. Effects of multifocal brain stimulation on pain and somatosensory functions. In: Zotterman Y, editor. Sensory function of the skin in primates. Oxford/New York: Pergamon; 1976. p. 531-46. 161. Jeanmonod D, Magnin M, Morel A. Thalamus and neurogenic pain: physiological, anatomical and clinical data. Neuroreport 1993;4:475-8. 162. Rinaldi PC, Young RF, Albe-Fessard D, Chodakiewitz J. Spontaneous neuronal hyperactivity in the medial and intralaminar thalamic nuclei of patients with deafferentation pain. J Neurosurg 1991;74:415-21. 163. Martin-Rodriguez JG, Buno W Jr, Garcia-Austt E. Human pulvinar units, spontaneous activity and sensory-motor influences. Electroencephalogr Clin Neurophysiol 1982;54:388-98. 164. Reynolds DV. Surgery in the rat during electrical analgesia induced by focal brain stimulation. Science 1969;164:444-5. 165. Richardson DE, Akil H. Pain reduction by electrical brain stimulation in man. Part 1: Acute administration in periaqueductal and periventricular sites. J Neurosurg 1977;47:178-83. 166. Barbaro NM. Studies of PAG/PVG stimulation for pain relief in humans. In: Fields HL, Besson J-M, editors. Progress in brain research. Pain modulation, vol 77. Amsterdam: Elsevier; 1988. p. 165-73. 167. Young RF, Kroening R, Fulton W, Feldman RA, Chambi I. Electrical stimulation of the brain in treatment of chronic pain. Experience over 5 years. J Neurosurg 1985;62:389-96. 168. Young RF, Bach FW, Van Norman AS, Yaksh TL. Release of beta-endorphin and 0methionine-enkephalin into cerebrospinal fluid during deep brain stimulation for chronic pain. Effects of stimulation locus and site of sampling. J Neurosurg 1993;79:816-25. 169. Gybels J, Kupers R. Deep brain stimulation in the treatment of chronic pain in man: where and why? Neurophysiol Clin 1990;20:389-98. 170. Levy RM, Lamb S, Adams JE. Treatment of chronic pain by deep brain stimulation: long term followup and review of the literature. Neurosurgery 1987;21:885-93. 171. Tasker RR, Vilela O. Deep brain stimulation for the control of intractable pain. In: Youmans JR, editor.
1319
1320
78 172.
173.
174.
175.
176. 177.
178.
179.
180.
181.
182.
183.
184.
185.
186.
Microelectrode recording in functional neurosurgery
Neurological surgery. Philadelphia: Saunders; 1995. p. 3512-27. Tasker RR, Yoshida M, Sima A, et al. Stimulation mapping of the periventricular-periaqueteductal gray matter. In: Samii M, editor. Surgery in and around the brain stem and third ventricle. Heidelberg: Springer; 1986. p. 161-7. Guiot G, Sachs M, Hertzog E, Brion S, Rougerie J, Dalloz JC, Napoleone F. [Electrostimulation & surgical lesions of the internal capsule; anatomical & physiology data.]. Neurochirurgie 1959;5:17-42. Guiot G, Hertzog E, Rondot P, Molina P. Arrest or acceleration of speech evoked by thalamic stimulation in the course of stereotaxic procedures for Parkinsonism. Brain 1961;84:363-79. Martinez SN, Bertrand C, Botana-Lopez C. Motor fiber distribution within the cerebral peduncle. Results of unipolar stimulation. Confin Neurol 1967;29:117-22. Nashold BS. Ocular reactions from brain stimulation in conscious man. Neuroophtalmology 1970;5:92. Velasco FC, Molina-Negro P, Bertrand C, Hardy J. Further definition of the subthalamic target for arrest of tremor. J Neurosurg 1972;36:184-91. Hirayama K, Tsubaki T, Toyokura Y, Okinaka S. The representation of the pyramidal tract in the internal capsule and basis pedunculi. A study based on three cases of amyotrophic lateral sclerosis. Neurology 1962;12:337-42. Bertrand G, Blundell J, Musella R. Electrical exploration of the internal capsule and neighbouring structures during stereotaxic procedures. J Neurosurg 1965;22:333-43. Hawrylyshyn PA, Tasker RR, Organ LW. Third ventricular width and the thalamocapsular border. Appl Neurophysiol 1976;39:34-42. Ashby P, Lang AE, Lozano AM, Dostrovsky JO. Motor effects of stimulating the human cerebellar thalamus. J Physiol 1995;489(Pt 1):287-98. Sano K, Sekino H, Tsukamoto Y, Yoshimasu N, Ishijima B. Stimulation and destruction of the region of the interstitial nucleus in cases of torticollis and see-saw nystagmus. Confin. Neurol 1972;34:331-8. Sano K, Yoshioka M, Ogashiwa M, Ishijima B, Ohye C, Sekino H, Mayanagi Y. Central mechanisms of neck movements in the human brain stem. Confin Neurol 1967;29:107-11. Sano K, Mayanagi Y, Sekino H, Ogashiwa M, Ishijima B. Results of stimulation and destruction of the posterior hypothalamus in man. J Neurosurg 1970; 33:689-707. Kalyanaraman S. Some observations during stimulation of the human hypothalamus. Confin Neurol 1975; 37:189-92. Spiegel EA, Wycis HT, Szekely EG, Dams DJ, Flanagan M, Baird HW, III. Campotomy in
187.
188.
189.
190.
191.
192.
193.
194.
195.
196.
197.
198.
199.
200.
various extrapyramidal disorders. J Neurosurg 1963; 20:871-84. Spiegel EA, Wycis HT, Szekely EG, Soloff L, Dams J, Gildenberg P, Zanes C. Stimulation of forel’s field during stereotaxic operations in the human brain. Electroencephalogr Clin Neurophysiol 1964;16:537-48. Schaltenbrand G, Spuler H, Wahren W, Wilhelmi A. Vegetative and emotional reactions during electrical stimulation of deep structures of the brain during stereotactic procedures. Z Neurol 1973;205:91-113. Wilson WP, Nashold BS. Evoked photic responses from the human thalamus and midbrain. Confin Neurol 1973;35:338-45. Nashold BS Jr, Gills JP Jr. Ocular signs from brain stimulation and lesions. Arch Ophthalmol 1967;77: 609-18. Heath RG, Mickle WA. Evaluation of seven years’s experience with depth electrode studies in human patients. In: Ramey A, O’Doherty C, editors. Electrical studies on unanaesthetized brain. New York: Hoeber; 1961. p. 214-47. Bondartchuk AN, Nikitina LI, Shmatkov YV. The problem of dominant central mechanisms controlling vegetative functions in man. In IV International Congress of Neurosurgery and IX International Congress Neurology, New York, Sept. 20–27,1969. Obrador S, Dierssen G. Mental changes induced by subcortical stimulation and therapeutic lesions. Confin Neurol 1967;29:168. Siegfried J, Wiesendanger M. Respiratory alterations produced by thalamic stimulation during stereotaxic operations. Confin Neurol 1967;29:220-3. Blum B, Israeli J, Askenasy HM. Changes in coronary vessels induced by repeated hypothalamic stimulation. Microcirculation 1976;1:295. Blum B, Ben-Ari W, Yashin T, et al. Regional changes in pial blood supply induced by hypothalamic stimulation. Microcirculation 1976;1:395. Fairman D. Hypothalamotomy as a new perspective for alleviation of intractable pain and regression of metastatic malignant tumors. In: Fusek K, editor. Present limits of neurosurgery. Prague: Avicenum; 1972. p. 525-8. Schaltenbrand G. The effects on speech and language of stereotactical stimulation in thalamus and corpus callosum. Brain Lang 1975;2:70-7. Ojemann G, Fedio P, Van Buren JM. Evidence for dominance within the human lateral thalamus. In IV Interational Congress of Neurosurgery and IX Interational Congress of Neurology, New York, Sept. 20–27, 1969. Ojemann GA. Language and the thalamus: object naming and recall during and after thalamic stimulation. Brain Lang 1975;2:101-20.
Microelectrode recording in functional neurosurgery
201. Creutzfeldt O, Ojemann G, Lettich E. Neuronal activity in the human lateral temporal lobe. I. Responses to speech. Exp Brain Res 1989;77:451-75. 202. Creutzfeldt O, Ojemann G, Lettich E. Neuronal activity in the human lateral temporal lobe. II. Responses to the subjects own voice. Exp Brain Res 1989;77:476-89. 203. Creutzfeldt O, Ojemann G. Neuronal activity in the human lateral temporal lobe. III. Activity changes during music. Exp Brain Res 1989;77:490-98. 204. Ojemann JG, Ojemann GA, Lettich E. Neuronal activity related to faces and matching in human right nondominant temporal cortex. Brain 1992;115 (Pt 1): 1-13. 205. Ojemann GA. Organization of language cortex derived from investigation during neurosurgery. Semin Neurosci 2:297. 206. Fedio P, Van Buren JM. Memory and perceptual deficits during electrical stimulation in the left and right thalamus and parietal subcortex. Brain Lang 1975;2:78-100. 207. Haglund MM, Ojemann GA, Lettich E, Bellugi U, Corina D. Dissociation of cortical and single unit activity in spoken and signed languages. Brain Lang 1993;44:19-27. 208. Hutchison WD, Lozano AM, Davis KD, Saint-Cyr JA, Lang AE, Dostrovsky JO. Differential neuronal activity in segments of globus pallidus in Parkinson’s disease patients. Neuroreport 1994;5:1533-7. 209. Sterio D, Beric A, Dogali M, Fazzini E, Alfaro G, Devinsky O. Neurophysiological properties of pallidal neurons in Parkinson’s disease. Ann Neurol 1994;35:586-91. 210. Filion M, Tremblay L, Bedard PJ. Effects of dopamine agonists on the spontaneous activity of globus pallidus neurons in monkeys with MPTP-induced parkinsonism. Brain Res 1991;547:152-61. 211. Filion M, Tremblay L. Abnormal spontaneous activity of globus pallidus neurons in monkeys with MPTP-induced parkinsonism. Brain Res 1991;547:142-51. 212. Jung R. Untersuchungen uber den Parkinson tremor und andere Zitterformen beim Menschen. Z Ges Neurol Psychiatr 1941;173:263. 213. Hoover JE, Strick PL. Multiple output channels in the basal ganglia. Science 1993;259:819-21. 214. Alexander GE, Crutcher MD, DeLong MR. Basal ganglia-thalamocortical circuits: parallel substrates for motor, oculomotor, ‘‘prefrontal’’ and ‘‘limbic’’ functions. Prog Brain Res 1990;85:119-46. 215. Alexander GE, Crutcher MD. Functional architecture of basal ganglia circuits: neural substrates of parallel processing. Trends Neurosci 1990;13:266-71. 216. Pare D, Curro’Dossi R, Steriade M. Neuronal basis of the parkinsonian resting tremor: a hypothesis and its implications for treatment. Neuroscience 1990;35: 217-26.
78
217. Llinas R, Pare D. Role of intrinsic neuronal oscillations and network ensembles in the genesis of normal and pathological tremors. In Findley LJ, Koller WC, editors. Handbook of tremor disorders. New York: Marcel Dekker; 1995. p. 7-36. 218. Microelectrode studies of unit discharges in the sensory motor cortex of monkeys. In: Gybels JM, editor. The neural mechanism of Parkinson tremor. Brussels: Editions Arsia; 1963. p. 83–122. 219. Hutchison WD, Lozano AM, Tasker RR, Lang AE, Dostrovsky JO. Identification and characterization of neurons with tremor-frequency activity in human globus pallidus. Exp Brain Res 1997;113:557-63. 220. Albin RL, Young AB, Penney JB. The functional anatomy of basal ganglia disorders. Trends Neurosci 1989;12:366-75. 221. DeLong MR. Primate models of movement disorders of basal ganglia origin. Trends Neurosci 1990;13:281-5. 222. Calvin WH. Comments on human epileptic neurons. In: Somjen GG, editor. Neurophysiology studied in man. Excerpta Medica; Amsterdam: p. 110-11. 223. Rayport M. Single neuron studies in human epilepsy. In: Somjen GG, editor. Neurophysiology studied in man. Amsterdam: Excerpta Medica; 1972. p. 100-09. 224. Li CL, Friauf W, Cohen G, Tew JM Jr. A method of recording single cell discharges in the cerebral cortex of man. Electroencephalogr Clin Neurophysiol 1965;18:187-90. 225. Li CL, Tew JM Jr. Reciprocal activation and inhibition of cortical neurones and voluntary movements in man: cortical cell activity and muscle movement. Nature 1964;203:264-5. 226. Hutchison WD, Dostrovsky JO, Davis KD, et al. Single unit responses and microstimulation effects in cingulate cortex of an awake patient. IASP world congress on pain, 1993. 227. Talbot JD, Marrett S, Evans AC, Meyer E, Bushnell MC, Duncan GH. Multiple representations of pain in human cerebral cortex. Science 1991;251:1355-8. 228. Davis KD, Hutchison WD, Lozano AM, Dostrovsky JO. Altered pain and temperature perception following cingulotomy and capsulotomy in a patient with schizoaffective disorder. Pain 1994;59:189-99. 229. Hosobuchi Y, Adams JE, Fields HL. Chronic thalamic and internal capsule stimulation for the control of facial anesthesia dolorosa and the dysesthesia of thalamic syndrome. Adv Neurol 1974;4:783. 230. Mazars G, Merienne L, Cioloca C. [Treatment of certain types of pain with implantable thalamic stimulators]. Neurochirurgie 1974;20:117-24. 231. Melzack R, Wall PD. Pain mechanisms: a new theory. Science 1965;150:971-9. 232. Shealy CN, Mortimer JT, Hagfors NR. Dorsal column electroanalgesia. J Neurosurg 1970;32;560-4.
1321
1322
78
Microelectrode recording in functional neurosurgery
233. Wall PD, Sweet WH. Temporary abolition of pain in man. Science 1967;155:108-9. 234. Mark VH, Ervin FR. Role of thalamotomy in treatment of chronic severe pain. Postgrad Med 1965; 37:563-71. 235. Sano K, Yoshioka M, Ogashiwa M, Ishijima B, Ohye C. Thalamolaminotomy. A new operation for relief of intractable pain. Confin Neurol 1966;27:63-6. 236. Hecaen H, Talairach J, David M, et al. Coagulation limitees du thalamus dans les algies du syndrome thalamique. Rev Neurol (Paris) 1949;81:917. 237. Frank F, Fabrizi AP, Gaist G, Weigel K, mundinger F. Stereotactic mesencephalotomy versus multiple thalamotomies in the treatment of chronic cancer pain syndromes. Appl Neurophysiol 50:314-8. 238. Adams JE, Rutkin BB. Lesions of the centrum medianum in the treatment of movement disorders. Confin Neurol 1965;26:231-45. 239. Benabid AL, Pollak P, Gao DM, Hoffmann D, Limousin P, Gay E, Payen I, Benazzouz A. Chronic electrical stimulation of the ventralis intermedius nucleus of the thalamus as a treatment of movement disorders. J Neurosurg 1996;84:203-14. 240. Laitinen LV, Bergenheim AT, Hariz MI. Ventroposterolateral pallidotomy can abolish all parkinsonian symptoms. Stereotact Funct Neurosurg 1992;58: 14-21. 241. Lozano AM, Lang AE, Galvez-Jimenez N, Miyasaki J, Duff J, Hutchinson WD, Dostrovsky JO. Effect of GPi pallidotomy on motor function in Parkinson’s disease. Lancet 1995;346:1383-7. 242. Desalles A, Hariz M, Functional Radiosurgery. In: Desalles A, Goelsch S, editors. Stereotactic surgery and radiosurgery. Madison, WI: Medical Physics Publishing; 1993. p. 389-406. 243. Rand RW, Jacques DB, Melbye RW, Copcutt BG, Fisher MR, Levenick MN. Gamma Knife thalamotomy and pallidotomy in patients with movement disorders: preliminary results. Stereotact Funct Neurosurg 1993;61 Suppl 1:65-92. 244. Alterman RL, Kall BA, Cohen H, Kelly PJ. Stereotactic ventrolateral thalamotomy: is ventriculography necessary? Neurosurgery 1995;37:717-21. 245. Kondziolka D, Dempsey PK, Lunsford LD, Kestle JR, Dolan EJ, Kanal E, Tasker RR. A comparison between magnetic resonance imaging and computed tomography for stereotactic coordinate determination. Neurosurgery 1992;30:402-6. 246. Tasker RR, Dostrovsky JO, Dolan EJ. Computerized tomography (CT) is just as accurate as ventriculography for functional stereotactic thalamotomy. Stereotact Funct Neurosurg 1991;57:157-66. 247. Taha JM, Lamba MA, Samaratunga C, et al. A method to reduce systemic spatial shift associated with magnetic resonance imaging for stereotactic surgery. American Society for Stereotactic and Functional Neurosurgery, March 8-11 1995. 1995.
248. Burchiel KJ, Coombs BD, Nquyen TT, et al. Stereotactic neurosurgical magnetic resonance imaging and geometric distortion. [American Society for Stereotactic and Functional Neurosurgery, March 8-11, 1995]. 1995. 249. Chen DY, Bradley WG, Ali-Ali F, et al. Stereotactic MRI localization in radiosurgery. [American Society for Stereotactic and Functional Neurosurgery, March 8-11,1995]. 1995. 250. Coghill RC, Talbot JD, Evans AC, Meyer E, Gjedde A, Bushnell MC, Duncan GH. Distributed processing of pain and vibration by the human brain. J Neurosci 1994;14:4095-108. 251. Burton H, Videen TO, Raichle ME. Tactile-vibrationactivated foci in insular and parietal-opercular cortex studied with positron emission tomography: mapping the second somatosensory area in humans. Somatosens Mot Res 1993;10:297-308. 252. Jones AK, Brown WD, Friston KJ, Qi LY, Frackowiak RS. Cortical and subcortical localization of response to pain in man using positron emission tomography. Proc Biol Sci 1991;244:39-44. 253. Carlson M, Nystrom P. Tactile discrimination capacity in relation to size and organization of somatic sensory cortex in primates: I. Old-World prosimian, Galago; II. New-World anthropoids, Saimiri and Cebus. J Neurosci 1994;14:1516-41. 254. Shibasaki H, Sadato N, Lyshkow H, Yonekura Y, Honda M, Nagamine T, Suwazono S, Magata Y, Ikeda A, Miyazaki M. Both primary motor cortex and supplementary motor area play an important role in complex finger movement. Brain 1993;116 (Pt 6):1387-98. 255. Fletcher PC, Frith CD, Grasby PM, Shallice T, Frackowiak RS, Dolan RJ. Brain systems for encoding and retrieval of auditory-verbal memory. An in vivo study in humans. Brain 1995;118 (Pt 2):401-16. 256. Swartz BE, Halgren E, Fuster J, Mandelkern M. An 18FDG-PET study of cortical activation during a short-term visual memory task in humans. Neuroreport 1994;5:925-8. 257. Kapur S, Meyer J, Wilson AA, Houle S, Brown GM. Activation of specific cortical regions by apomorphine: an [15O]H2O PET study in humans. Neurosci Lett 1994;176:21-4. 258. Jones AK, Qi LY, Fujirawa T, Luthra SK, Ashburner J, Bloomfield P, Cunningham VJ, Itoh M, Fukuda H, Jones T. In vivo distribution of opioid receptors in man in relation to the cortical projections of the medial and lateral pain systems measured with positron emission tomography. Neurosci Lett, 1991;126:25-8. 259. Ogawa S, Lee TM, Kay AR, Tank DW. Brain magnetic resonance imaging with contrast dependent on blood oxygenation. Proc Natl Acad Sci USA 1990;87:9868-72. 260. Cohen MS, Bookheimer SY. Localization of brain function using magnetic resonance imaging. Trends Neurosci 1994;17:268-77.
Microelectrode recording in functional neurosurgery
261. Bucher SF, Seelos KC, Stehling M, Oertel WH, Paulus W, Reiser M. High-resolution activation mapping of basal ganglia with functional magnetic resonance imaging. Neurology 1995;45:180-2. 262. Ogawa S, Tank DW, Menon R, Ellermann JM, Kim SG, Merkle H, Ugurbil K. Intrinsic signal changes accompanying sensory stimulation: functional brain mapping with magnetic resonance imaging. Proc Natl Acad Sci USA 1992;89:5951-55. 263. Kwong KK, Belliveau JW, Chesler DA, Goldberg IE, Weisskoff RM, Poncelet BP, Kennedy DN, Hoppel BE, Cohen MS, Turner R. Dynamic magnetic resonance imaging of human brain activity during primary sensory stimulation. Proc Natl Acad Sci USA 1992;89:5675-79. 264. Rao SM, Binder JR, Hammeke TA, Bandettini PA, Bobholz JA, Frost JA, Myklebust BM, Jacobson RD, Hyde JS. Somatotopic mapping of the human primary motor cortex with functional magnetic resonance imaging. Neurology 1995;45:919-24. 265. Kim SG, Ashe J, Georgopoulos AP, Merkle H, Ellermann JM, Menon RS, Ogawa S, Ugurbil K. Functional imaging of human motor cortex at high magnetic field. J Neurophysiol 1993;69:297-302. 266. Kim SG, Ashe J, Hendrich K, Ellermann JM, Merkle H, Ugurbil K, Georgopoulos AP. Functional magnetic resonance imaging of motor cortex: hemispheric asymmetry and handedness. Science 1993;261:615-17. 267. Hammeke TA, Yetkin FZ, Mueller WM, Morris GL, Haughton VM, Rao SM, Binder JR. Functional magnetic resonance imaging of somatosensory stimulation. Neurosurgery 1994;35:677-81. 268. Puce A, Constable RT, Luby ML, McCarthy G, Nobre AC, Spencer DD, Gore JC, Allison T. Functional magnetic resonance imaging of sensory and motor cortex: comparison with electrophysiological localization. J Neurosurg 1995;83:262-70. 269. Davis KD, Wood ML, Crawley AP, Mikulis DJ. fMRI of human somatosensory and cingulate cortex during painful electrical nerve stimulation. Neuroreport 1995;7:321-5. 270. Shaywitz BA, Shaywitz SE, Pugh KR, Constable RT, Skudlarski P, Fulbright RK, Bronen RA, Fletcher JM,
271.
272.
273.
274.
275. 276.
277.
278.
279.
280.
78
Shankweiler DP, Katz L. Sex differences in the functional organization of the brain for language. Nature 1995;373:607-9. Hinke RM, Hu X, Stillman AE, Kim SG, Merkle H, Salmi R, Ugurbil K. Functional magnetic resonance imaging of Broca’s area during internal speech. Neuroreport 1993;4:675-8. Davis KD. Recent advances and future prospects in neuroimaging of acute and chronic pain. Future Neurol 2006;1:203-13. Jack CR Jr, Thompson RM, Butts RK, Sharbrough FW, Kelly PJ, Hanson DP, Riederer SJ, Ehman RL, Hangiandreou NJ, Cascino GD. Sensory motor cortex: correlation of presurgical mapping with functional MR imaging and invasive cortical mapping. Radiology 1994;190:85-92. Naatanen R, Ilmoniemi RJ, Alho K. Magnetoencephalography in studies of human cognitive brain function. Trends Neurosci 1994;17:389-95. Gallen CC, Bloom FE. Mapping the brain with MSI. Curr Biol 1993;3:522-4. Gallen CC, Sobel DF, Lewine JD, Sanders JA, Hart BL, Davis LE, Orrison WW Jr. Neuromagnetic mapping of brain function. Radiology 1993;187: 863-7. Forss N, Hari R, Salmelin R, Ahonen A, Hamalainen M, Kajola M, Knuutila J, Simola J. Activation of the human posterior parietal cortex by median nerve stimulation. Exp Brain Res 1994;99:309-15. Mogilner A, Nomura M, Ribary U, Jagow R, Lado F, Rusinek H, Llinas R. Neuromagnetic studies of the lip area of primary somatosensory cortex in humans: evidence for an oscillotopic organization. Exp Brain Res 1994;99:137-47. Yang TT, Gallen CC, Schwartz BJ, Bloom FE. Noninvasive somatosensory homunculus mapping in humans by using a large-array biomagnetometer. Proc Natl Acad Sci USA 1993;90:3098-102. Yang TT, Gallen CC, Ramachandran VS, Cobb S, Schwartz BJ, Bloom FE. Noninvasive detection of cerebral plasticity in adult human somatosensory cortex. Neuroreport 1994;5:701-4.
1323
82 Radiofrequency Lesions E. R. Cosman Sr. . E. R. Cosman Jr.
The radiofrequency (RF) technique was first put into clinical practice for making controlled therapeutic lesions in the nervous system in the early 1950s, and it have remained an important methodology to this day. This chapter summarizes the physical principles of the RF lesion method and describes the accepted rules for the RF lesion generation that have been verified over the years by clinical experience. Guidelines for making RF lesions of a specific size are stated, and a variety of specific electrode configurations used for various procedures are illustrated. Practical aspects of monitoring the progress of a safe and effective RF lesion are described.
A Brief History of Radiofrequency – Lesion Making Over the years, many physical principles have been utilized to make therapeutic lesions, targeted neurolysis, in the central and the peripheral nervous system. To name a few, these physical principles include: RF heating, direct current coagulation, cryogenics, focused ultrasound, induction heating, chemical destruction, ionizing radiation, stereotactic radiosurgery, mechanical methods such as the leucotomy, focused electromagnetic waves, and lasers. The RF lesion method has remained popular and successful among these techniques because of certain technical advantages that are intrinsic to its application, and these are summarized in the > Table 82-1. Although there have been many contributors to the progress made in the use of the RF technique, there are two names that stand out as particularly significant for their early work in the #
Springer-Verlag Berlin/Heidelberg 2009
field and for the historic magnitude of their contributions: William H. Sweet and Bernard J. Cosman (> Figure 82-1). Sweet wrote a landmark paper [1] in 1953, coauthored by Vernon Mark, which showed that the use of very high frequency current (in the radiofrequency range) for lesion production has decisive advantages over the then established direct current lesion methods that had been used for decades. Sweet and Mark demonstrated that the lesions made by RF are characterized by a well-circumscribed lesion borders, and that there is better control of size and shape of the RF lesion than there is with the direct current method. Sweet made a second major advance to the field a few years later when he performed the first temperature-controlled RF lesions in the trigeminal ganglion for the treatment of trigeminal neuralgia [2–4]. His technique was rapidly adopted by numerous pain clinicians around the world and helped to underscore the power of the RF method. Bernard J. Cosman (> Figure 82-1) made parallel pioneering contributions to the design and engineering of RF lesion generators and electrodes. Prototype RF generators were made by Cosman and Aranow in collaboration with the neurosurgical group at the Massachusetts General Hospital in Boston in the early 1950s. Shortly thereafter, Cosman introduced the first commercial RF lesion generator, the Radionics Model RFG-2, for sale on the United States. For the next 40 years Bernard Cosman, with one of the present authors, his son, Eric Cosman, Sr., continued to make major advances in RF generator and electrode technology, and their designs dominated the field for decades, being offered by their company Radionics, Inc.
1360
82
Radiofrequency lesions
The modern RF lesion generators and electrode designs of today are elaborate and sophisticated. > Figure 82-2 shows two of the latest RF generators, the Model RFG-1A and the Model G4 . Table 82-1 Advantageous features of the radiofrequency lesion method 1. Well-circumscribed lesions 2. Temperature control allows: TQuantifiable lesions Non-charring, sticking, or boiling Differential selection 3. Excellent target control with: Stimulation Impedance monitoring Recording 4. Easily adapted to stereotactic or fixation devices 5. Versatile, robust electrode configurations for numerous clinical indications and target sites 6. Continuous and pulsed RF provides thermal and electric field modification of neural structures that suit specific indications
Graphics-Four (Cosman Medical, Inc., Burlington, Massachusetts). They both incorporate multiple functions and circuitry including an impedance monitor, wide-range stimulator, recording outputs and input connections, temperature monitors, both continuous and pulsed RF generator circuits, automatic temperature control, and lesion timers. They have built-in microprocessors and computers that can provides automatic function of the lesion process as well as safely checks to guard against untoward effects during the procedure. The Model G4 ‘‘GraphicsFour’’ unit enables use of up to four RF electrodes being applied to the patient at the same time, which has utility, for example, in treating multiple level medial branches for spinal pain. It has full color graphic display of the electrode functions with touch-screen features such as built-in standard procedure menus, recording memory, and printout records of procedure parameters, which are helpful, for example, in a busy pain
. Figure 82-l Two pioneers of radiofrequency surgery. William H. Sweet (left) of the Massachusetts General Hospital wrote the landmark paper in 1953, describing the decisive advantage of the radiofrequency lesion. Bernard J. Cosman (right) was the electrical engineer who built and perfected the earliest commercial radiofrequency lesion generators in the early 1950s, working with Dr. Sweet and Dr. Thomas Ballantine
Radiofrequency lesions
82
. Figure 82-2 Two modern RF lesion generators. the Cosman Model RFG-1A (left) incorporates multiple functions and controls, digital readouts, and computer controls for effective and safe lesion making. The Cosman Model G4 ‘‘Graphics – Four’’ (right) has full computer graphic screen display, touch screen controls, stored menus and procedure records, and both digital and graphic screen representations for the function processes. It also has four electrode output jacks so that as many as four electrodes can be used on the patient at during the same procedure. Multiple electrode usage provides time-efficiency in some pain treatments, for example, performing multiple segmental medial branch RF heat lesions during the same patient treatment session
clinic where the throughput of patients being treated for spinal pain is high. The modern RF electrodes are also far more elaborate and sophisticate than in the early days of RF technology. Because the RF electrode can be made in a wide variety of shapes and configurations, a large range of styles have been developed for use at specific target sites. Many examples with be shown in this chapter. Today the use of the RF method, which might be referred to as ‘‘radiofrequency surgery,’’ spans the treatment of numerous disease states and spans a wide range of target sites from the brain to nearly every extremity of the body. In this chapter, we will focus on the RF applications in neurosurgery. However, the RF technique, though having its beginnings in neurosurgery with the work of Sweet and Cosman, has today found far wider application in fields primarily outside of neurosurgery, including, to name a few examples: anesthesiology and the peripheral pain therapy, interventive radiology and cancerous tumor ablation, urology and the treatment of BPH, cardiology and the treatment of WPW syndrome, podiatry, and the treatment of vascular disease.
Physical Principles of Radiofrequency Heat Lesion Generation The proper use of the RF lesion method requires a basic understanding of the physical processes involved [5–10]. > Figure 82-3 shows the basic RF lesioning circuit. The RF generator is the source of an RF voltage that is impressed across its output terminals. The terminals are in turn connected by cables to the so-called active electrode and the dispersive electrode (often called a reference electrode). The active electrode produces the heat lesion, and the dispersive electrode is typically a large-area electrode that is not intended to produce heating, but serves as a return pathway for the RF current. Electrical metering can monitor the RF current and voltage. Appropriate metering can also monitor the impedance and temperature at the active electrode tip. > Figure 82-4 illustrates how the RF current flows through the body between the active and dispersive electrodes. The total RF current IRF in the cables reaches the electrodes and spreads out
1361
1362
82
Radiofrequency lesions
. Figure 82-3 The basic RF circuit
. Figure 82-4 The RF current patterns in the tissue between the active lesion electrode and the dispersive electrodes
in space in the electrolytic medium of the body. This is illustrated by lines of RF current density j, whose patterns are governed by the laws of electricity. The greatest heating takes place in the region of highest current density, which is near the tip of the active electrode. If the dispersive electrode has a significantly large area, the current density near it is low and heating nearby is minimal. For this reason, large-area dispersive electrodes with an area of at least 150 mm2 and an ample amount of conductive gel
between the electrode and the patient’s skin are recommended to avoid any chance of skin burns. The dispersive, or reference, electrode is typically placed on the skin surface over an area of musculature, not far from the site of the active electrode. For example, an active electrode used in the brain may be accompanied by a reference electrode attached and taped to the patient’s shoulder. Galvanic potentials can arise if active and reference electrodes are made from different metals. Thus, when the active lesion tip is made of stainless steel, which is a common type of electrode material, the base material of the reference electrode should also be made of stainless steel to prevent any transient potentials arising between the active and reference electrodes when the electrodes are first connected to the patient and the RF generator. > Figure 82-5 shows the pattern of the electric field and current density near the uninsulated active electrode tip. The mechanism for heating near the tip arises from the RF voltage that is created on the active electrode by the RF generator output. This voltage creates a distribution of the electric field E in the space around the electrode, shown by the E-field lines in > Figure 82-5. The electric field at any given point in space oscillates with the RF
Radiofrequency lesions
82
. Figure 82-5 The electric field line patterns in tissue around the RF electrode, and the associated isotherm surfaces (dashed lines). The RF current lines follow the same pattern and the electric field lines
frequency and causes the nearby charged ions in the electrolyte to move back and forth in space at the same high frequency, which is typically about 500,000 cycles per second (Hertz) for most modern generators. This ionic oscillation is called the ionic current density j, and the pattern the j-field lines in space around the electrode tip follows the same pattern as the E-field lines shown in > Figure 82-5. It is the frictional heating within the tissue resulting from the RF ionic oscillation, i.e., the current density j, which is the basic mechanism by which the tissue heats up, and accordingly, by which the RF heat lesion is made. The ionic oscillations and current density are strongest near the electrode tip because the electric field is strongest there. Thus, the power deposition and tissue temperatures are highest adjacent to the electrode tip. The tissue that is heated by this mechanism in turn heats up the RF electrode tip, and so, by monitoring the tip temperature, an accurate measure of the nearby tissue temperature, and thus the progress of the RF lesioning, is obtained. Temperature monitoring during the
RF lesion making is an essential for control, consistency, and safety of the procedure. The dashed lines in > Figure 82-5 illustrate the surfaces of constant temperature, which are referred to as isotherms. The 45–50 C isothermal surface is critical, since within that surface the tissue will he hotter than 45–50 C and will be permanently killed. That then defines the heat lesion volume. It can be shown that at equilibrium, in a homogeneous medium, and for a specific tip size and tip temperature, the isotherm dimensions are roughly independent of the electrical and thermal conductivities. For a fixed tip size, a higher tip temperature produces a larger 45 C isotherm surface; that is, the lesion volume is larger. However, for a given tip temperature, the lesion size is also dependent on the size of the electrode tip, with larger tips producing, larger lesions. > Figure 82-6 is a schematic diagram of the temperature falloff in the tissue as a function of nominal distance from the electrode tip. This curve can now be calculated by modern
1363
1364
82
Radiofrequency lesions
. Figure 82-6 A schematic diagram of temperature falloff as a function of nominal distance from the RF electrode tip
computer methods. Predicted lesion sizes agree with experimentally and clinically observed data. Certain qualitative features are notable. The highest temperatures occur in the tissue near the tip. The specific shape of the curve, and thus the lesion volume, depend on specific parameters such as: the electrode tip dimensions; the temperature that is maintained in the tissue near the electrode tip (by control of the RF voltage on the electrode); and, the electrical conductivity, thermal conductivity, and convection of the surrounding tissue. As shown in the > Figure 82-6, the higher the measured tip temperature, the larger the distance associated with the 45 C isotherm. It is believed that in the range between 42 and 44 C, neural tissue can sustain “reversible” damage by heating [11]. Thus, there may be a zone of reversibility for a given tip temperature corresponding to the shell of tissue near the 45 C isotherm that can be stunned, but not killed, by thermal lesioning. Measurement of the electrode tip temperature not only quantifies the lesion size but also avoids the tissue near the electrode tip reaching 100 C point, and thus avoids the catastrophic effects of
boiling, explosive gas formation, and searing and charring of the tissue. This dangerous condition can be avoided through careful elevation and monitoring of the tissue temperature to prevent temperature runaway to boiling. > Figure 82-7 illustrates another important aspect the lesion making. That is the time progression toward an equilibrium lesion size. This figure shows the increase of the transverse dimension of the lesion as a function of the time during which the electrode tip temperature is held at a fixed value. The size of the lesion increases until it reaches its asymptotic value. Once this value has been approached, no substantial increase of the lesion size occurs. For consistent lesion making, it is desirable to achieve this equilibrium situation, since it avoids some of the uncertainties associated with variability in tissue impedance, vascularity, and proximity to cerebrospinal fluid (CSF), bone, and other heat sinks. The typical time to reach the asymptotic temperature is 30–60 s for electrode tip sizes that are typically used in neurosurgery and pain therapy and for relatively uniform soft tissue environments [7].
Radiofrequency lesions
. Figure 82-7 DREZ lesion width versus time. The equilibrium lesion size is achieved after 35–45 s for an approximately constant temperature at the electrode tip
82
. Figure 82-8 Irregularities of RF heating can be caused by inhomogenieties in nearby tissue structures and fluid bodies
Figure 82-8 schematically illustrates examples. The proximity of the electrode to CSF bodies such as the ventricles can provide a low-impedance shunt pathway for the RF current density j, and thus sink the heat away and cause irregular lesion shapes and sizes. An example of this situation is RF lesioning in the trigeminal ganglion. The proximity of large blood vessels also has an inhomogeneous cooling effect on the tissue near them. The placement of the RF electrode near a bony structure may have the opposite effect, since a bony mass is an insulator with lower blood circulation. Examples of this can be the RF heating of an intervertebral disk or of the space between other joints in the body. Despite the uncertainties caused by tissue in homogeneities however, the above rules for homogeneous tissue and for nominal conditions have worked well over the past decades, as attested by a mass of successful clinical data using the RF heating method. >
The above discussion can be summarized by some simple rules associated with RF lesion making: Rule 1: The RF current heats the tissue, and the tissue in turn heats the RF electrode tip. Rule 2: Temperature is the basic lesioning parameter, and it should be measured and controlled for consistent and safe RF lesioning. The measurement of electrode tip temperature is directly related to tissue temperature and lesion size. Rule 3: Achievement an equilibrium lesion as a function of time for a fixed tip temperature produces more consistent lesion size (than, for example, the time-buildup lesions). It is desirable to hold the proper tip temperature for 30–60 s to achieve the equilibrium lesion size. Rule 4: The proper electrode size and tip temperature should be selected for a given target site to achieve a consistent and desired lesion size. Specific factors that are difficult to predict or quantify can give rise to variations in the simplified picture of the RF lesion process from one patient to the next. Among these factors are the inhomogenieties in the tissue medium itself.
Typical Lesion Sizes as a Function of Electrode Geometries and Tip Temperatures It is important to have a sense of the size of the lesion that is being made for a given electrode lip geometry and tip temperature. An extensive
1365
1366
82
Radiofrequency lesions
. Table 82-2 Postmortem lesion sizes versus heating parameters Electrode tipa
Lesion size Thalamus Thalamus Cingulum Cingulum Cingulum
Lesion size, mm
Temperature and times
Diameter, mm
Length, mm
Temperature, C
Time, seconds
Time, Postmortem
A
B
1.1 1.2 1.6 1.6 1.6
5 3 5 10 10
72 65 70 80–90 80
360 120 60 60 60
2 years ? 5 months 6 months 6 months
3 2 8 10 10
7 4 8 10–12 12
a
Straight stereotactic electrodes as shown in > Figure 82-17
discussion of this issue is given in papers by Cosman, et al [5–9], and only summary information will be given here. > Table 82-2 shows lesion sizes in the human brain reported by several stereotactic neurosurgeons using conventional electrode of straight, cylindrical tip geometry. These data for the most part were accumulated in postmortem studies at varying times after the lesion was made, and some variation in the lesion size might be expected as a function of time. As an example from > Table 82-2, it is seen that for thalamotomies performed using electrode tips with a diameter of 1.1 mm and a tip length of 3–5 mm and tip temperatures of 65–75 C, lesions sizes of approximately 3 mm minor in diameter and 4–7 mm in length can be achieved. Lesions for cylindrical electrode tip shapes are typically prolate ellipsoids of revolution. It is also seen that larger lesions can be made in the brain with larger electrode tips and higher temperatures. For example, lesions made in the cingulum with electrodes having 1.6 mm diameter and 10 mm tip length at 80–90 C have dimensions of typically 10 mm in minor diameter and 12 mm in length [12]. Electrodes of greater diameter and length for equivalently high temperatures are accordingly larger. > Table 82-2 shows the range of parameters used by several stereotactic neurosurgeons to produce thalamic lesions. This suggests that some variation in lesion size has been used successfully
in RF thalamotomies, but also that there is a reasonable norm for acceptable lesion parameters. As is described further below, electrodes with more complex geometries, such as with having side-outlet electrode tip extensions, have also been used for more precise or more asymmetric targeting in the region of the thalamus, in cases where physiological testing indicates that the initial electrode placement is not ideal. Information on lesion size for very small RF electrodes, such as those used in the spinal cord, is rare but does exist. Cosman and coworkers [7] have reported that electrodes with tip diameters of 0.25 mm and tip lengths of 2 mm, raised to a temperature of about 80 C and for a lesion time of approximately 15 s, will produce lesion sizes of about 2 mm in width and 2 mm in length. A summary of these data is shown in > Figure 82-9. Transverse lesion diameter, which is equivalent to the width of the prolate ellipsoidal lesion, is plotted as a function of the tip temperature for various electrode diameters. The length of the lesion can be assumed to be approximately 1–3 mm longer than the exposed length of the RF electrode, assuming substantially cylindrical electrode tip geometry with either a hemispherical or a sharpened tip end shape. Although this graph may supply canonical curves for lesion size, it must be recognized that the unpredictable factors of tissue inhomogeneity,
Radiofrequency lesions
. Figure 82-9 The canonical curves of lesion diameter (A) versus tip temperature as a function of the electrode tip diameter (top). The dimensions of the heat lesion and the electrode tip (bottom)
3.
82
will not occur from encroachment in nontarget volumes. Proper configuration of the electrode as adapted for a particular target region.
The reader is referred to many articles and procedure technique monographs for specific recommendations on these points for the various RF lesion sites. A brief description is given below of successful RF electrode geometries for specific procedures to illustrate the great range of such designs.
Spinal Cord Electrodes
inhomogeneous conductivities and convectivities, and proximity to bony structures all can produce significant variations in the shape and size of lesions in a given situation.
Radiofrequency Electrode Configurations for Specific Clinical Procedures In addition to the choice of the proper RF electrode size and tip temperature, safe and effective lesion making is dependent on the following conditions: 1. 2.
The proper placement of the electrode in the target region. Adequate physiological testing using stimulation and recording to assess that the target is correct and assure that untoward effects
Targets in the spinal cord are discrete and critical. Associated RF electrodes must be appropriately small, with temperature monitoring, despite the fact that their historical forerunners, such as the cordotomy electrodes of Rosomoff [13], Lin and associates [14], Mullan [15,16], and others, were of relatively larger size and of a nontemperature-measuring variety. > Figure 82-10 shows the pointed tip configuration for the LCE Levin-Cosman Cordotomy Electrode Kit (Cosman Medical), which has an approximately 0.25 mm tip diameter and 2.0 mm tip exposure with an indwelling surfacemounted thermocouple [17]. The electrode can penetrate the pia and make very precise and discrete lesions in the spinothalamic tract for the treatment of intractable pain. The unique design of a surface thermocouple sensor within the sharpened tip allows rapid and faithful lesion temperature readings, which are essential in such tight geometries. For such tiny tips, any small thermal perturbation toward boiling during the temperature raising phase of the procedure, would cause explosive gas and steam formation. Thus, superfast temperature monitoring and fine RF power verniation are essential. This electrode is used through a 2l-gauge special spinal needle that is inserted percutaneously.
1367
1368
82
Radiofrequency lesions
. Figure 82-10 The LCE Levin Cordotomy Electrode (top) and its tip geometry (bottom)
shouldered insulation, staggered tip insulation, and angled-tip configuration are designed for proper electrode placement and target localization in the very tight geometry of the spinal cord that is exposed during laminectomy. In these critical procedures, safe and reliable lesion placement and control require such special tip geometries and rapid surface-mounted thermocouple sensing in the electrode tip.
RF Electrodes for Lesioning in the Trigeminal Ganglion
Figure 82-11 illustrates the tip geometries for the KCTE Kanpolat CT Radionics electrodes, which have both straight and bent-tip geometry [18]. These electrodes are used for percutaneous cordotomy and spinal tractotomies and embody the important innovation of being CT-compatible. Kanpolat and associates [18–22] demonstrated that, with the use of proper materials, a computed tomography (CT) image of the electrode in place in the spinal cord can be made for direct visualization of the positioning of the electrode tip in the lateral spino-thalamic tract. Thus, in addition to the essential physiological testing, which is done through the Levin and Kanpolat electrodes before the lesion is made in percutaneous cordotomies, a new dimension of target position confirmation is achieved by doing the procedure in the CT scanner. > Figure 82-11 shows a confirmational CT slice made with the Kanpolat electrode. > Figure 82-12 shows the Nashold DREZ Electrode tip geometry and the El-Naggar-Nashold Nuclear Caudalis Electrode tip geometry, both from Cosman Medical, Inc. These electrodes have been used by Nashold and coworkers [23–25] to make lesions in the dorsal root entry zone (DREZ) lesion and the nucleus caudalis, respectively. The >
W. H. Sweet and coworkers at the Massachusetts General Hospital in Boston revolutionized the treatment of facial pain from tic douloureux through the use of temperature-monitoring RF electrodes inserted percutaneously into the trigeminal ganglion and posterior rootlets through the foramen ovale [1–4,26,27]. > Figure 82-13 shows the earliest electrode system for this procedure; the Cosman TIC Kit. It evolved from Sweet’s original concepts and was designed by B. J. Cosman, and is still offered by Cosman Medical and used by many neurosurgeons. The set has four insulated cannulas with a removable obdurating stylet. The cannulas having exposed tip lengths 4, 2, 5, 7, and 10 mm. In practice, a cannula, with a stylet in place, is inserted into the trigeminal ganglion, the stylet can be removed, and a thermocouple (TC) temperature electrode can be inserted. The TC electrode is connected to the RF generator, and physiologic testing is done using the stimulation signal output from the generator’s stimulator circuitry. Once the proper target position is thereby confirmed, the RF signal output from the RF generator is applied to the electrode, and the heat lesion is made using while monitoring electrode tip temperature on the generator’s readout display. In this way, Sweet and other workers [28–43] have reported excellent results for the relief of trigeminal pain since the early 1970s.
Radiofrequency lesions
82
. Figure 82-11 The KCTE Kanpolat Cordotomy Electrode (top) The KCTE tip variations (middle figure) include a straight tip and an angled tip to enable adjustment of the tip position in the spinal cord. The diagrams (bottom) illustrate the approach to the lateral spinothalamic tract under CT control. A CT scan shows the KCTE Electrode in the spinal cord
The straight electrodes of the TIC Kit, which have been adequate for most trigeminal neuralgia procedures, are sometimes inadequate to reach selectively the first division of the trigeminal nerve. Therefore, a modification of the TIC Kit was built to overcome this limitation. > Figure 82-14 shows
the Tew Kit that was developed by John M. Tew, MD and Eric Cosman Sr. [33] The Tew Kit includes an insulated cannula through which either an obdurating stylet, a straight emergent RF electrode, or a flexible side-outlet emergent RF electrode can be passed. The tip configurations
1369
1370
82
Radiofrequency lesions
. Figure 82-12 Nashold Thermocouple DREZ Electrode (top) is shown with its holding cannula and sizing clamp. The El-Naggar/ Nashold Angled ENA Electrode (second from the top) with its gripping handle and connection leader wire. The tip detail of the DREZ electrode (lower left, top) shows its shouldered insulation for limited penetration of the dorsal root entry zone region of the spinal cord. The ENA electrode tip (lower left, bottom) that has an angled distal shaft end to facilitate penetration of the nucleus caudalis. The surgical approach for the ENA Electrode shown in the lower right drawing
for the straight and curved Tew electrodes are also shown in > Figure 82-14. Both electrodes have TC thermocouple temperature sensors built into their tip ends. The cannula, with the obdurating stylet in it, is percutaneously inserted into the trigeminal ganglion through the foramen ovale. The stylet is removed, and either the
straight TC electrode is inserted, for making axial tip extensions beyond the end of the cannula, or the curved spring electrode is inserted, for making off-axis tip extensions. This gives the clinician the capability to search in the space of the trigeminal rootlets with the electrode to find the desired trigeminal division corresponding to
Radiofrequency lesions
82
. Figure 82-13 This shows the classical TIC Kit electrodes for trigeminal nerve RF lesioning. The upper figure shows the temperature sensor probe inserted into a cannula. The lower left figure shows the different available cannula tip lengths. The lower right figure shows the electrode cannula’s tip within the trigeminal rootlets
. Figure 82-14 The TEW Electrode accommodates straight and curved RF lesioning tips for use in the trigeminal ganglion. The upper picture shows the TEW cannula with a curved temperature sensor inserted within it. The lower left figure shows the curved spring electrode tip emerging from the cannula’s distal end. The lower right figure shows the curved electrode tip within the posterior rootlets of the trigeminal nerve. The curved tip can be angle upwards or downwards to better reach the painful trigeminal division
the trigeminal pain as determined by sensory stimulation. Each electrode can be connected to the impedance, stimulation, and RF heating functions of the lesion generator with full temperature control. The off-axis electrode illustrates the flexibility that is possible in an RF electrode design to accommodate a specific anatomical target.
RF Lesion Electrodes for Spine to Relieve Neck and Back Pain In the early 1970s, Dr. C. N. Shealy proposed a technique of percutaneous RF Rhizotomy of the medial branch to relieve mechanical lower back pain associated with the lumbar facet joints
1371
1372
82
Radiofrequency lesions
[44–47]. The original SRK Shealy Rhizotomy Kit was developed by E. R. Cosman, Sr. and B. J. Cosman. It had a set of 14-gauge spinal needles that could be inserted percutaneously to approach the area proximate the suspected pain facets. A 16-gauge RF electrode, insulated except for an exposed 7 mm distal tip, could then be passed through the 14 gauge spinal needle, and could be advanced so that the electrode tip contacts the medial branch innervating the selected facet joint. This electrode was designed to produce a heat lesion with temperature control near the medial branch. There was encouraging success in the early days of the technique, but the results were not always consistent. A variant of the SRK electrode referred to as the RRE Ray Rhizotomy Electrode Kit was designed by E. R. Cosman and produced by Radionics Inc. in association with Dr. C. Ray [48,49]. This electrode has a self-penetrating cannula with an insulated tip and temperature monitoring and eliminated the need for the spinal needles. The RRE electrode is
still offered by Cosman Medical, Inc, and it is still preferred by some pain clinicians because of the large lesions it produces and because of its especially rigid shaft. In the late 1970s, Dr. M. Sluijter, and E. R. Cosman developed a more discrete set of electrodes, referred to as the SMK Kit, having finer gauge cannulas for the purpose of treating spinal facet pain in a similar manner to that proposed by Shealey. Sluijter’s approach utilized somewhat different target positions and electrode trajectories [50–55]. Because of its finer gauge electrodes, Sluijter used the system to treat pain in the lumbar as well and the cervical areas. A modern-day version of this electrode system is the CSK Kit produced (Cosman, Medical, Inc), and is shown in > Figure 82-15. The CSK Kit (Cosman Spinal Kit) contains, in one of its versions, 22-gauge (approximately 0.7 mm diameter) disposable cannulas which are offered with uninsulated tip lengths of 2, 5, and 10 mm tip exposures. Other versions of the CSK
. Figure 82-15 The CSK Cosman Spinal Kit for procedures of the spine such as RF medial branch or DRG procedures. Top figure shows the CSK cannula with the temperature sensing TC probe inserted in it. The middle figure shows the range of shaft lengths to accommodate cervical to sacral approaches in the spine for all size patients. The lower figure shows the straight and curved tips and the sharp and blunt points available to suit clinicians’ approaches, insertion techniques, and injection criteria
Radiofrequency lesions
Kit are offered with 20 gauge or with 18 gauge cannulas, shaft lengths of 15 and 20 cm, and a variety of tip lengths to accommodate different target sites and patient sizes. The finer cannulas have the great advantage, over the Shealy and Ray designs, of minimizing the discomfort and problems of percutaneous insertion. The larger gauge CSK cannulas (20 and 18 Gauge) are preferred by some clinicians over the 22 gauge types because they make larger heat lesions in the lumbar and the sacral regions or because they can be used to produce multiple heat lesions for sacroiliac (SI) joint denervations. The cannulas have plastic hubs that have the further advantage of enabling the clinician to visualize the target site under C-arm X-rays a view along the direction of the cannula insertion. That is referred to as the “needle view” approach, and it greatly simplifies the speed and accuracy of placing the electrode tip on the medial branch nerve. The hubs have luer tapers that allow injection of anesthetic and contrast media before the RF lesion is made. The obdurating stylet can be removed, and the CSK-TC thermocouple-sensing electrode can be inserted. This thermocouple electrode has a surface-mounted thermocouple that makes its temperature response very rapid and its temperature-reading accuracy extremely precise. Such fine-gauge cannulas can be inserted without a local anesthetic, and the patient remains fully awake and alert. Once the cannula has been percutaneously inserted, it can be manipulated under simple C-arm fluoroscopy, which is essentially a freehand stereotactic technique. The C-arm is first aligned to the anatomy in the direction desired for the needle insertion; i.e., the needle view direction. The needle is then inserted and manipulated until its radio-opaque metal shaft appears essentially as a dot on the C-arm fluoroscopic image. The cannula is then parallel to the direction of the C-arm view, and the tip is positioned at the proper target location relative to the bony structure, as described by Sluijter [50–55]. > Figure 82-16 shows fluoroscopic views using the needle-view method and the CSK cannulas [36].
82
. Figure 82-16 The RRE Ray Rhizotomy Electrode for RF lesioning of medial branch to treat mechanical back pain. The lower figure shows the tip geometry
Stereotactic Radiofrequency Electrodes The earliest RF electrodes produced by B.J. Cosman in the 1950s were for stereotactic intracranial procedures to treat mood disorders. They comprised straight tubular stainless steel shafts that were insulated over their entire length except for an exposed conductive tip. In the earliest intracranial procedures, the stereotactic electrodes were placed either freehand or with crude guidance and stabilization means. However, as sophisticated stereotactic guidance systems became available by the mid-to-late 1950s, the diameters and lengths of the stereotactic electrodes were precisely tailored to fit the apparatus and their associated microdrive accessories. Straight stereotactic RF electrodes are still available (Cosman Medical, Inc.), and used in intracranial stereotaxy, primarily to treat Parkinson’s disease, pain, and mood disorders. A range of diameters and tip lengths, and shaft lengths are available to suit different clinical needs and modern stereotactic systems. Examples are shown in > Figure 82-17. The smooth hemispherical tips and insulation minimize disruption to brain
1373
1374
82
Radiofrequency lesions
. Figure 82-17 Straight TC Stereotactic electrodes for intracranial RF lesioning, designed for use with a stereotactic guidance system. Tip geometry shown in the lower figure
tissue, and built in thermocouple temperature sensors in the tip enable accurate lesion control. More elaborate stereotactic electrodes have been produced for special stereotactic procedures, and they demonstrate the versatility that is possible in RF electrode design For example, off-axis lesion making to enlarge or to asymmetrically shape the lesion volume lead to electrodes with side extension tips. One of the earliest side-outlet electrodes was designed by N. T. Zervas and built by E. R. Cosman for RF ablation of the pituitary gland [56]. Its tip configuration is shown in the lower right portion of > Figure 82-18. It enabled asymmetrical lesions to be made to ablate all or part of the pituitary gland. An interesting historical note is that the Zervas hypophysectomy was, to the authors’ knowledge, the first example of an RF ablation of a cancerous tumor in humans. Today, RF ablation of tumors is commonly performed on a wide range of target sites throughout the body. > Figure 82-18 also shows a universal, frontfacing, open-lumen cannula that accommodates different inner elements to be passed through, including micro stimulation and/or recording
. Figure 82-18 A range of electrode configurations for straight and off-axis RF lesioning as well as stimulation and recording, after the designs of Zervas, Guildenberg, and Cosman
electrodes and both straight and side projecting RF electrodes. In the upper right is the Gildenberg mini-electrode system, which has a combined recording tip and lesion electrode for more specific targetry during thalamotomy and pallidotomy. J. Siegfried designed an elegant side-issue stereotactic electrode. The SSE Siegfried Stereotactic Electrode is shown in > Figure 82-19. The electrode has a 1.6-mm central shaft with an exposed RF electrode tip of 4 mm for making axial lesions. For stereotactic thalamotomies, it is often necessary to extend the lesion laterally or explore the space around the central electrode tip to determine the optimal target. This is accomplished by a side-emerging electrode with an exposed tip 0.5 mm in diameter and 2.0 mm in length. The tip can be extended a variable amount from the central cannula in any direction and can be used for stimulation, recording, or lesion making. The tiny side-issue tip has an indwelling, surface-mounted thermocouple electrode for very rapid, accurate temperature monitoring if it is desirable to make a lesion from that off-axis tip.
Pulsed Radiofrequency Technique In the modern RF generators, such as shown in > Figure 82-2, there are two modes of RF output that are commonly used for pain therapy. The
Radiofrequency lesions
output waveforms for these two modes are shown schematically in > Figure 82-20. The first mode is the conventional, thermal (or heat) RF mode which has been described so far in this chapter. It uses a continuous sinusoidal RF waveform output, commonly referred to as continuous RF, or CRF. The second mode, which was first reported on by M. Sluijter, E. R. Cosman, and coworkers in 1998 [57], uses a series of pulsed bursts of RF signal, referred to as pulsed RF, or PRF. To date, PRF has been used primarily . Figure 82-19 The SSE Siegfried Side-Outlet Stereotactic Electrode tip geometry for RF lesioning in the basal ganglia, viz. thalamotomies
82
to treat peripheral nerves and the DRG’s, and most commonly it has been applied to treat back and neck pain and neuropathies. The results have been very good and have now been the subject of a large published literature and several clinical trials. One poplar advantage of PRF over CRF is that it can be done with little or no pain to the patient as the PRF output is being delivered. This is in contrast to some CRF applications in which there is considerable pain and discomfort to the patient during the RF heating of the neural tissue. To date there have been limited attempts to use PRF in the central nervous system, although if it were to have success there, it could have significantly expended potential. In both of the continuous and the pulsed RF modes, the amplitude of the RF voltage, V(RF), of the RF waveforms as shown in > Figure 82-20, is measured in units of Volts (V). As describes above, for the continuous RF waveform, a heat lesion is produced by the action of ionic friction of the RF currents in the tissue caused by the voltage V(RF) on the electrode. This means that
. Figure 82-20 A schematic diagram of the RF waveforms for continuous RF mode (CRF) (top) and for pulsed RF mode (bottom). Axes are not to scale
1375
1376
82
Radiofrequency lesions
the neural tissue near the uninsulated, metal electrode tip is heated continuously to destructive temperatures (greater than 45–50 C). Thus, the CRF lesion volume includes all tissue within the 45–50 C isotherm boundary, which tends to have an ellipsoidal shape that encompasses the electrode tip. Within this lesion volume, all cell structures are macroscopically destroyed by heat. The action of pulsed RF on neural tissue is different from continuous RF [9]. Because the RF output is delivered in bursts of short duration relative to the intervening quiescent periods, the average temperature of the tissue near the electrode is not raised continuously or as high as for continuous RF for the same RF voltage V(RF). Since the PRF voltage is typically regulated to keep the average tip temperature in a nondestructive range, other mechanisms must produce the clinically-observed pain relieving effects. The electric field, E, is the fundamental physical quantity that governs all the actions of RF output on neural tissue, both for pulsed RF and for continuous RF modes. The electric field is created in space around an RF electrode that is connected to the output voltage V(RF) from an RF generator. His was illustrated in > Figure 82-5, and also is shown in > Figure 82-21 for a pointed electrode that is commonly used for percutaneous pain procedures. E is represented by an arrow (vector) at every point in space around the electrode tip, indicative of the magnitude and the direction the force it will produce on charged structures and ions in the tissue. The E-field produces various effects on tissue including: oscillations of charges, ionic currents, charge polarizations, membrane voltages, and structuremodifying forces. For continuous RF mode, the dominant consequence of these effects is the production of heat in the tissue caused by frictional energy loss due to the ionic currents that are driven by the E-field. However, for pulsed RF, the effects of E-field are more complex and varied, and range from heat flashes, to modification of neuron ultrastructure, to neural excitation phenomena.
. Figure 82-21 Schematic E-field patterns around a pointed RF electrode (top); and the calculated E-field strength distribution (bottom) in tissue for a 22 gauge electrode at V(RF) = 45 V
All of these effects can play a role in neuronal modification, though exactly how they produce antinociception in PRF treatments is a subject of active scientific investigation. Two consequences of theoretical predictions of the electric field in tissue during PRF are supported by experimental and clinical observations [9]. The first is that, as a consequence of the very high E-fields at the electrode tip, there are hot flashes at the electrode tip that can be thermally destructive to neurons. The second is that there are significant non-thermal effects of the E-field on neurons at positions away from the point of the tip that are certainly related to the pain-relieving effects of PRF. During the brief RF pulse, a hot spot occurs at the tip which can be 15–20 C above the average tissue temperature of the tissue [9] that remains
Radiofrequency lesions
near body temperature of 37–42 C as shown in > Figure 82-22, top. This has been confirmed by ex vivo measurements, > Figure 82-22, bottom, and by finite-element calculations. The intense E-field and hot flashes could be expected to have destructive effects on neural tissue very near the tip point. Evidence for such destruction has been observed in vitro (Cahana et al. [58]). This may play a role in PRF’s clinical effect when electrode point is in the nerve or pressing against the nerve. However, it is unlikely that such focal effects can account for all of PRF pain relief, since the region of extremely high E-fields and T hot flashes are likely confined to less than about 0.2 mm radius from the electrode point (> Figure 82-23). There is evidence that direct, non-thermal effects are important in PRF. It is known that pain relief can be achieved when the side of the electrode tip, not the tip point, is next to an axon or dorsal root ganglion (DRG). While the hot flash fluctuations are less than 1 C at 0.5 mm
82
from the tip in any direction for typical PRF voltages, at lateral distances of greater than 1 mm, the magnitude of the electric field is still large in biological terms. For example, finite-element computation of the E-field for V (RF) = 45 V predict [9] that the E is 20,000 V/m at 0.5 mm, and 12,000 V/m at 1.0 mm laterally (> Figure 82-23). Thus, neuronal modifications in this E-field range should be significant. Comparison of E and T strengths between typical CRF and PRF waveforms show striking differences between these RF modes (> Figure 82-24). Calculations predict that after 60 s of CRF at V(RF) = 20 V, E = 21,000 V/m and T = 60–65 C at the lateral tip surface, and E = 2,750 V/m and T = 50 C at 1.8 mm away. In contrast, after 60 s of PRF with V(RF) = 45 V, E = 46,740 V/m and T = 42 C at the lateral tip surface, and E = 6,100 V/m and T = 38 C at 1.8 mm away. In other words, in PRF, the direct electric field effects are more prominent, whereas in CRF, the
. Figure 82-22 Electric field strength and temperature field strength distributions for the first PRF pulse around a 22 gauge pointed electrode at V(RF) = 45 V and a pulse width of 20 ms
1377
1378
82
Radiofrequency lesions
. Figure 82-23 Hot flashes during a PRF pulse
. Figure 82-24 E-fields dominate over T-fields in PRF. The opposite is true for CRF
Radiofrequency lesions
thermal fields are more prominent and largely mask the E-field effects. Combined with the understanding that PRF has a clinical effect even when the electrode is not placed on the nerve directly, these physical observations suggest that the E-field is directly involved in the analgesic effect of PRF. It is known that PRF Efields produce significant trans-membrane potentials on the neuron membrane and organelles. The E-field can also penetrate the membranes of axon and the DRG soma to disrupt essential cellular substructures and functions. For example, PRF done on the DRG of rabbits causes pronounced neuron ultrastructural modifications that are seen only under electron microscopy [59] and that are likely to modify or disable the cell’s function. This would suggest that PRF can produce sub-cellular, microscopic lesions on neurons in a volume around the electrode, possibly resulting in reduction of afferent pain signals. PRF membrane potentials are also capable of neural excitations (action potentials) by a process called membrane rectification. Because the PRF pulse rate is similar to that of classical conditioning stimulation (1–2 Hz), it has been proposed that PRF may have a similar action [9]. Conditioning stimulation is capable of suppressing synaptic efficiency of A-delta and C-fiber afferent nociception signals [60], a phenomenon know as Long Term Depression (LTD). Therefore, the PRF might be reducing transmission of pain information by LTD of synaptic connections in the dorsal horn. The appropriate exposure of PRF for a given pain syndrome and anatomical target, either for microscopic or LTD mechanisms, should be governed by the PRF “E-dose.” E-dose provides a parametric measure of E-field strength and integral pulse/time exposure [9].
Practical Tips on RF Lesion Making Proper lesion parameters, target control, and clinical judgment are paramount to successful RF lesion making. There are also practical
82
considerations in performing the lesion that can make a decisive difference in regard to a successful result. Although the latest RF generator systems have many automatic features, such as automatic temperature control and automatic ramp up of output levels, it should be kept in mind that, although automatic features may be acceptable for uses at lower risk target sites such as the medial branch nerves in the lower back, their use may not be indicated for very critical target site, such as in the central nervous system. It is for the critical targets that the control of the RF generator and of the progress of the RF lesioning itself should be managed with utmost attention by the hands and eyes of the clinician. One of the most common difficulties that can interrupt an RF procedure is failure of the electrode cables. These cables are the most manipulated, cleaned, sterilized, and mishandled objects in the process and thus are the most vulnerable. Spare cables should be on hand at all times. The impedance monitor on the RF generator should be watched for intermittent open or short circuits. The continuous impedance monitor of the RFG1A and G4 models allows monitoring of circuit integrity before, during, and after the lesion, which is a help to detect any fluctuations or abnormal readings that would indicate faulty connections, untoward shunting of current such as at insulation breaks, or incipient temperature instabilities that could, for example, signal focal boiling. A check of the electrode insulation before and after each case should be carefully done, since breaks in the insulation can cause RF current leakage in tissue regions where lesions are not desired and thus can lead to potentially harmful effects. The use of large-area dispersive electrodes of at least 150-cm2 area is essential to avoid skin burns at the reference electrode contact. The use of needles as a reference electrode should be avoided. Before beginning the lesioning making phase of the procedure and when the electrode is in the patient’s body, a tip temperature of approximately body temperature, i.e., 37 2 C, should always be
1379
1380
82
Radiofrequency lesions
observed on the generator readout. If not, then most likely either the temperature sensing electrode or the connecting cable is faulty, and they should be checked or replaced with spare ones as required. Raising the RF power smoothly and by hand is traditionally good practice, especially when small lesion electrodes that are used in critical anatomical areas such as the spinal cord, brain, and trigeminal ganglion. Runaway temperature to the boiling point can lead to severe and uncontrolled effects, and thus a watchful eye on the temperature meter is required. For smaller electrode tips, the heating process is more sensitive and finicky, because small changes in RF output voltage levels can lead to quick temperature rises and higher risk of a runaway to instability and boiling at the tip. Unusually sluggish or jumpy temperature rise as you raise the RF output control level can be a sign of trouble, and should prompt extra vigilance and even a check of the electrode and cables. While the temperature is the fundamental lesioning parameter and should be measured and carefully watched throughout the procedure, it is good practice to be observant of the voltage, current, and impedance readings. Fluctuations of the voltage and current often signal an erratic effect such as cable interrupt or focal boiling, and under such conditions, the lesion process should be terminated and the system should be checked. It is good practice for a technician to record the lesion temperature, the impedance, and the RF voltage and current during the procedure so that if any question arises, those parameters can be reviewed later.
References 1. Sweet WH, Mark VH. Unipolar anodal electrolyte lesions in the brain of man and cat: report of five human cases with electrically produced bulbar or mesencephalic tractotomies. Arch Neural Psychiatry 1953;70:224-34. 2. Sweet WH, Poletti CE, Roberts JT. Dangerous rises in blood pressure upon heating of trigeminal rootlets; Increased bleeding time in patients with trigeminal neuralgia. Neurosurgery 1985;17:843.
3. Sweet WH, Wepsic JG. Controlled thermocoagulation of trigeminal ganglion and rootlets for differential destruction of pain fibers: I. Trigeminal neuralgia. J Neurosurg 1974;39:143-56. 4. Sweet WH, Wepsic JG. Controlled thermocoagulation of trigeminal ganglion and rootlets for differential destruction of pain fibers. J Neurosurg 1974;40:143-56. 5. Cosman BJ, Cosman EG. Guide to radiofrequency lesion generation in neurosurgery: radionics procedure technique series monographs. Burlington, MA: Radionics; 1974. 6. Cosman ER, Cosman BJ. Methods of making nervous system lesions. In: Wilkins RH, Rengachary SS, editors. Neurosurgery, vol. 3. New York: McGraw-Hill; 1984. p. 2490-8. 7. Cosman ER, Nashold BS, Ovelman-Levitt J. Theoretical aspects of radiofrequency lesions in the dorsal root entry zone. Neurosurgery 1984;15:945-50. 8. Cosman ER, Rittman WJ, Nashold BS, Makachinas TT. Radiofrequency lesion generation and its effect on tissue impedance. Appl Neurophysio1 1988;51:230-42. 9. Cosman ER Jr, Cosman ER, Sr. Electric and thermal field effects in tissue around radiofrequency electrodes. Pain Med 2005;6:(6)405-24. 10. Dieckmann G, Gabriel E, Hassler R. Size, form, and structural peculiarities of experimental brain lesions obtained by thermo controlled radiofrequency. Confin Neurol 1965;26:134-42. 11. Brodkey J, Miyazaki Y, Ervin FR, Mark VH. Reversible heat lesions, a method of stereotactic localization. J Neurosurg 1964;21:49. 12. Hurt RW, Ballantine HT Jr. Stereotactic anterior cingulate lesions for persistent pain: a report on 68 cases. Clin Neurosurg 1974;21:334-51. 13. Rosomoff HL, Carroll F, Brown J, Sheptak T. Percutaneous radiofrequency cervical cordotomy: technique. J Neurosurg 1965;23:639-44. 14. Lin PM, Gildenberg PL, Polakoff PP. An anterior approach to percutaneous lower cervical cordotomy. J Neurosurg 1966;25:553-60. 15. Mullan S. Percutaneous cordotomy. J Neurosurg 1971;35:360-6. 16. Mullan SF, Lichtor T. Percutaneous microcompression of the trigeminal ganglion for trigeminal neuralgia. J Neurosurg 1983;59:1007-12. 17. Levin AB, Cosman ER. Thermocouple-monitored cordotomy electrode. J Neurosurg 1980;53:266-8. 18. Kanpolat Y, Cosman ER. Special radio frequency electrode system for computed tomography-guided painrelieving procedures. Neurosurgery 1996;38:600-3. 19. Kanpolat Y, Savas A, Akyar S, Cosman E. Percutaneous computed tomography-guided spinal destructive procedures for pain control. Neurosurg Q 2004;14(4):229-38. 20. Kanpolat Y, Deda H, Akyar S, Bilgic S. CT guided percutaneous cordotomy. Acta Neurochir Suppl (Wien) 1989;46:67-8.
Radiofrequency lesions
21. Kanpolat Y, Deda H, Akyar S, Bilgic S. CT guided trigeminal tractotomy. Acta Neurochir (Wien) 1989;100: 112-4. 22. Kanpolat Y. The surgical treatment of chronic pain: destructive therapies in the spinal cord. Neurosurg Clin N Am 2004;15:307-17. 23. Friedman AH, Nashold B, Ovelmen-Levitt J. Dorsal root entry zone lesions for the treatment of post-herpetic neuralgia. J Neurosur 1984;60:1258-62. 24. Nashold BS, El-Naggar A, Abdulhak M. New RF lesion DREZ electrodes for relief of facial pin based on a neuroanatomical study in man of the trigeminal nucleus caudalis at the cervicomedullary junction. American Society for Stereotactic and Functional Neurosurgery June 1991. 25. Sampson JH, Nashold BS. Facial pain due to vascular lesions of the brain stem relieved by dorsal root entry zone lesions in the nucleus caudalis. J Neurosurg 1992;77: 473-5. 26. Sweet WH. The treatment of trigeminal neuralgia (tic douloureux). N Engl J Med 1986;315:174-7. 27. White JC, Sweet WH. Pain and the neurosurgeon: a fortyyear experience. Springfield, IL: Charles C Thomas; 1969. pp 169-97, 607–609. 28. Tew JM. Percutaneous electrocoagulation of the trigeminal nerve in the treatment of trigeminal neuralgia. Radionics procedure technique series. Burlington, MA: Radionics; 1974. 29. Tew JM, Keller JT. The treatment of trigeminal neuralgia by percutaneous radiofrequency technique. Clin Neurosurg 1977;24:557-78. 30. Tew JM, Mayfield FH. Trigeminal neuralgia: a new surgical approach (percutaneous electrocoagulation of the trigeminal nerve). Laryngoscope 1973;83:1096. 31. Tew JM, Jr, van Loveren H, Percutaneous rhizotomy in the treatment of intractable facial pain (trigeminal, glossopharyngeal and facial nerves. In: Schmidek HH, Sweet WH, editors. Operative neurosurgica! technique. Orlando, FL: Grune & Stratton; 1988. p. 1111-23. 32. Tew JM, van Loveren HR, Caputi F. Percutaneous stereotactic radiofrequency rhizoromv for trigeminal neuralgia. Radionics procedure technique series. Burlington, MA: Radionics; 1990. 33. Tobler WD, Tew JM, Cosman ER, et al. Improved outcome in the treatment of trigeminal neuralgia by percutaneous stereotactic rhizotomy with a new, curved tip electrode. Neurosurgery 1983;12(3):313-7. 34. Van Loveren H, Tew JM, Keller JT, et al. A ten year experience in the treatment of trigeminal neuralgia: a comparison of percutaneous stereotaxic rhizotomy and posterior fossa exploration. J Neurosurg 1982;57:757. 35. Siegfried J. 500 percutaneous thermocoagulations of the gasserian ganglion for trigeminal pain. Surg Neurol 1977;3:126-31. 36. Siegfried J. Percutaneous controlled thermocoagulation of gasserian ganglion in trigeminal neuralgia: experiences
37.
38.
39.
40.
41.
42.
43.
44. 45.
46. 47.
48.
49.
50.
51.
52.
82
with 1,000 cases. In: Samii M, Jannetta PJ, editors. The cranial nerves. Berlin: Springer; 1981. p. 322-30. Siegfried J, Broggi G. Percutaneous thermocoagulation of the gasserian ganglion in the treatment of pain advanced cancer. In: Bonica JJ, Ventafridda V, editors. Advances in pain research and therapy. New York: Raven Press; 1979. p. 463-8. Siegfried J, Vosmansky M. Technique of the controlled thermocoagulation of trigeminal ganglion and spinal roots. In: Krayenbuhl H, editor. Advanced and technical stan-dards in neurosurgery, vol. 2. Berlin: Springer. p. 199-209. Broggi G, Franzini A. Radiofrequency trigeminal rhizotomy in treatment of symptomatic, non neoplastic facial pain. J Neurosurg 1982;57:483-6. Broggi G, Franzini A, Lasio G, et al. Long-term results of percutaneous retrogasserian thermorhizotomy for “essential” trigeminal neuralgia: considerations in 1000 consecutive patients. Neurosurgery 1990;26:783-7. Broggi G, Siegfried J. The effect of graded thermocoagulation on trigeminal evoked potentials in the cat. Acta Neurochir (Wien) 1977;24:175-8. Lazorthes Y, Verdie JC, Lagarrigue J. Thermocoagulation percutanee des nerfs rashidens a vise analgesique. Neurochirurgie 1976;22:445-53. Lazorthes Y, Verdie JC, Bouyssen M. Interet de 10 υtilisation d’un cadre stereotaxique dans la thermocoagulation selective du ganglion de Gasser. Neurochirugie 1976;22:77-83. Shealy CN. The role of the spinal facets in back and sciatic pain. Headache 1974;14:101-4. Shealy CN. Percutaneous radiofrequency denervation of spinal facets and treatment for chronic back pain and sciatica. J Neurosurg 1975;43:448-51. Shealy CN. Facet denervation in the management of back and sciatic pain. Clin Orthop 1976;115:157-64. Shealy CN. Technique for percutaneous spinal facet rhizotomy. Radionics procedure technique series. Burlington, MA: Radionics; 1973. Ray CD. Percutaneous radio-frequency facet nerve blocks: treatment of the mechanical low-back syndrome. Radionics, procedure technique series. Burlington, MA: Radionics; 1982. Ray CD. Your low-back pain and facet nerve blocks (audiovisual presentation for patient education). Minneapolis, MN: Institute for Low Back Care, Sister Kenny Institute; 1982. Sluijter ME. Percutaneous thermal lesions in the treatment of back and neck pain. Radionics procedure technique series. Burlington, MA: Radionics; 1981. Sluijter ME. Radiofrequency lesions in the treatment of cervical pain syndromes. Radionics procedure technique series. Burlington, MA: Radionics; 1990. Sluijter ME. The use of radiofrequency lesions of the communicating ramus in the treatment of low back pain. Techniques of neurolysis. Boston: Kluwer; 1989.
1381
1382
82
Radiofrequency lesions
53. Sluijter ME. The use of radiofrequency lesions for pain relief in failed back patients. International disability studies. Basel: Eular; 1989. 54. Sluijter ME, Koetsveld-Baart CC. Interruption of pain pathways in the treatment of the cervical syndrome. Anaesthesia 1980;35:302-7. 55. Sluijter ME, Mehta M. Treatment of chronic back and neck pain by percutaneous thermal lesions. In: Lipton S, Miles J, editors. Persistent pain, modern methods of treatment. London: Academic Press; 1981. p. 141-79. 56. Zervas NT. Stereotaxic thermal hypophysectomy. Current techniques in operative neurosurgery. Grune & Stratton; New York: 57. Sluijter ME, Cosman ER, Rittman WJ, Kleef M. The effects of pulsed radiofrequency fields applied to the dorsal root ganglion- a preliminary report. Pain Clin 1998;11: (2)109-18.
58. Cahana A, Vutskits L, Muller D. Acute differential modification of synaptic transmission and cell survival during exposure target position pulsed and continuous radiofrequency energy. J Pain 2003;4:(4)197-202. 59. Erdine S, Yucel A, Cunan A, et al. Effects of pulsed versus conventional radiofrequency current in rabbit dorsal root ganglion morphology. Eur J Pain 2005;9:(3)251-6. 60. Sandkuhler J, Chen JG, Cheng G, Randic M. Low frequency stimulation of the afferent A-delta fibers induces long-term depression at the primary afferent synapses with substantia gelatinosa neurons in the rat. J Neurosci 1997;17:6483-91. 61. Fox JL. Experimental relationship of radio-frequency electrical current and lesion size for application to percutaneous cordotomy. J Neurosurg 1970;33:415-21. 62. Kline MT. Stereotactic radiofrequency lesions as part of the management of pain. Orlando, FL: Paul M. Deutsch Press; 1992.
83 Stimulation Physiology in Functional Neurosurgery A. W. Laxton . J. O. Dostrovsky . A. M. Lozano
Despite substantial progress in the clinical application of deep brain stimulation (DBS) [1–5], the specific mechanisms underlying its effects have yet to be fully determined. In this chapter, following a brief description of relevant anatomy, we discuss some of the physiological principles of electrical neurostimulation, and review research investigating the mechanisms of DBS.
DBS anatomy Because DBS is most commonly used for the treatment of movement disorders, the majority of studies that have explored the mechanisms of DBS have investigated stimulation of the principal movement disorder surgery targets: the subthalamic nucleus (STN), the internal segment of the globus pallidus (GPi), and the ventral intermediate nucleus of the thalamus (Vim). It is therefore worthwhile to briefly outline the relevant neuroanatomy of these targets. More detailed descriptions of this anatomy are to be found in other chapters of this textbook.
output to the GPi and the substantia nigra pars reticulata (SNr) as well as the GPe [6,7].
GPi The GPi receives excitatory glutamatergic input from the STN, inhibitory GABA-ergic input from the striatum and GPe, and dopaminergic input from SNc [8]. The GPi sends inhibitory GABAergic projections to the ventral and intralaminar thalamus, and the pedunculopontine region [6,7]. The rodent homologue of the GPi is the entopeduncular nucleus, embedded within the internal capsule [7].
Vim The Vim receives excitatory glutamatergic input from the cerebral cortex and deep cerebellar nuclei, and inhibitory GABA-ergic input from the reticular nucleus of the thalamus [6,8,9]. The Vim sends glutamatergic efferents to the cortical motor regions and the striatum [6,8,9].
STN The STN receives excitatory glutamatergic input from the frontal cerebral cortex [6,7]. It also receives inhibitory GABA-ergic input from the external segment of the globus pallidus (GPe) [6,7]. The parafascicular nucleus of the thalamus, pedunculopontine nucleus (PPN), and substantia nigra pars compacta (SNc) also project onto the STN [7]. The STN sends excitatory glutamatergic #
Springer-Verlag Berlin/Heidelberg 2009
Basic Physiology of Electrical Neurostimulation DBS is typically delivered in biphasic square wave pulses via cathodal monopolar or bipolar electrodes. The physiological properties of monopolar stimulation are better characterized than they are for bipolar stimulation (see > Table 83-1) [10]. Monopolar stimulation spreads more diffusely
1384
83
Stimulation physiology in functional neurosurgery
. Table 83-1 Basic physiology of electrical neurostimulation* 1. 2. 3. 4. 5. 6.
7.
Axons are more responsive than cell bodies Large axons are more responsive than small axons Heavily myelinated axons are more responsive than less myelinated axons Neural elements are more responsive to cathodal stimulation than anodal High currents (8 times threshold) block action potentials The effect of stimulation on a neural element depends on the electrode’s distance from that neural element Current flow that is parallel to an axon is more likely to produce excitation than current flow that is transverse to an axon
*This table is adapted from Ranck [10]
and can therefore influence neural elements over a greater distance than bipolar stimulation [11]. Compared to monopolar stimulation, the focused current of bipolar stimulation may increase the risk of damage to stimulated regions [12]. Electrical stimulation affects not only the neurons near the electrode but can also excite the axons of neurons at a distance from the stimulation electrode that are passing near by and can also spread to affect neural elements outside of the immediate brain region stimulated. Therefore, although it is common to label stimulation as STN or GPi stimulation, that is really an abbreviated way of saying that electrical impulses are delivered by an electrode placed in the STN or GPi which affect an array of neural elements in the surrounding area [10]. For example, the STN and GPi are surrounded by other important structures, such as the zona incerta and pallidothalamic fiber bundles including the lenticular fasciculus and ansa lenticularis. The effects of DBS may actually result from stimulation of these structures [13]. Thus, the particular brain region, the proximity to surrounding pathways, and even the location within a nucleus will all influence what neural elements are stimulated and therefore what the effect of stimulation will be [14].
The various neural elements differ in their responsiveness to electrical stimulation. Less current is required to excite axons, in particular large myelinated axons, than neuronal cell bodies [10,13]. Furthermore, as the distance of a neural element increases from the stimulating electrode, less current reaches it [3]. The region within which stimulation may influence neural elements increases with stimulation amplitude [1]. This is not a straightforward effect, however, as afferent, efferent, and interneuronal axons may be activated, and this activation may be antidromic or orthodromic [14]. The lowest current (of a theoretically infinite duration) which will generate an action potential in a stimulated neuron is called the rheobase. The time it takes from the onset of electrical stimulation (at twice the rheobase) to the onset of an action potential in the stimulated neuron is termed the chronaxie. A fundamental relationship exists between the amplitude and duration or pulsewidth of stimulation. To maintain a constant effect, the stimulation amplitude must be increased as the duration is decreased, and conversely the stimulation duration must be increased as the amplitude is reduced. This relationship is represented in the following equation: Ith ¼ Irh ð1 þ tad =PWÞ where Ith is the threshold current, Irh is the rheobase, tad is the chronaxie, and PW is the pulse width or duration [8,15]. Different neural elements have different chronaxies. Large myelinated axons have the shortest chronaxies (30–200 ms); smaller axons have longer chronaxies (200–700 ms), and unmyelinated axons, dendrites and neuronal cell bodies have much larger chronaxies (1–10 ms)[8,10].Thismeansthatlargemyelinated axons are more easily activated by electrical stimulationthanotherneuralelements[10,13].Neural tissues also differ in their relative resistivity to electrical impulses. This too affects how stimulation current propagates. White matter generally has 2–3 times greater resistivity than gray matter
Stimulation physiology in functional neurosurgery
which has 4–6 times greater resistivity than cerebrospinal fluid [10]. White matter is also anisotropic which means that the conduction of electrical impulses is directed by the orientation of axons [10]. By comparing the chronaxies for a given DBS effect to known chronaxies, it is possible to infer which neural elements are being stimulated with DBS. Following this reasoning Nowak and Bullier [16] demonstrated that electrical stimulation activated axons and not cell bodies in the cortical gray matter of rodents [8]. This finding was corroborated by a follow-up experiment by Nowak and Bullier [17] in which the orthodromic responses to extracellular electrical stimulation were only slightly lower (15–20%) following N-methyl-D-aspartate (NMDA) block of neuronal cell bodies. Because 80–85% of the normal stimulation response was maintained, the results provide confirmatory evidence that in most cases the neural element most responsible for the effects of electrical brain stimulation is the myelinated axon [1,8]. Holsheimer et al. [18] have extended this work into the clinical realm. They measured the latency of DBS-induced tremor suppression among patients with Parkinson’s disease and found that chronaxies for thalamic and internal pallidal stimulation (129–151 ms) fell within the range of chronaxies of large myelinated axons (30–200 ms). As this is much shorter than the chronaxie of neuronal cell bodies (1–10 ms), these results corroborate that the primary target of DBS mediating the therapeutic benefits of stimulation in those two regions is the myelinated axon [1,8,10]. The distance of the stimulating electrode to the neural elements influences its effect. The rheobase and chronaxie increase as the distance from the stimulating electrode to the neural element increases [8,10]. When the applied current exceeds the threshold by a factor of 8 or more, it can paradoxically block excitation (by inducing a depolarization block) [8]. Therefore, neural elements nearest an electrode may be blocked and distant elements may not receive sufficient
83
stimulation to respond, whereas those elements within an appropriate intermediate perimeter will be excited [8,10]. This range of influence also depends on the properties of each neural element. For example, small axons may be activated close to the stimulating electrode, whereas larger axons may be blocked [10]. The orientation of the stimulating electrode to the axon is also known to influence the effect of stimulation. Current flow that is parallel to an axon is more likely to produce excitation than current flow that is transverse to an axon [10,19].
DBS Mechanisms of Action DBS, surgical lesions, and inhibitory drugs such as the sodium channel blocker lidocaine and the GABAA agonist muscimol, generally produce similar clinical effects when applied to the same neuroanatomical targets [2,20–22]. It is, therefore, reasonable to suppose that DBS and lesions work through a similar mechanism: the inhibition of neuronal activity [3]. While parsimonious and intuitively appealing, this conclusion is difficult to reconcile with the basic physiology of electrical neurostimulation as just described. To better account for the mechanisms of DBS, several explanations have been proposed. It has been suggested that DBS (i) has direct effects onneuronalmembrane properties and ionconductances,(ii)stimulatessynapticactivityleadingtothe release or even depletion of neurotransmitters and postsynaptic desensitization of neurotransmitter receptors, and (iii) alters the frequency or pattern of pathological neuronal activity [3–5,14,22–24].
Direct Cellular and Membrane Effects The ability of high frequency stimulation (HFS) to alter intrinsic membrane properties has been posited as a potential mechanism of DBS [25]. There is evidence that STN HFS can act directly
1385
1386
83
Stimulation physiology in functional neurosurgery
on neuronal membrane sodium and calcium channels [26]. Some in vitro studies suggest that HFS can decrease the excitability of neurons through the inactivation of voltage-gated sodium and calcium currents [27]. By transiently depressing voltage-gated Ca2+ channels, HFS has been shown to block depolarization [27]. HFS may also directly affect Na + channels [28,29]. Using patch-clamp techniques in rat STN slices, Beurrier et al. [27] found that 1 min of high frequency (100–250 Hz) bipolar STN stimulation blocked ongoing STN neuronal firing, and that this blockade lasted for around 6 min after the end of stimulation. Because this neuronal silencing occurred even in the presence of glutamate and GABA antagonists, or cobalt which blocks voltage-gated Ca2+ channels and neurotransmitter release, they concluded that this stimulation effect was not synaptically mediated. Instead, stimulation directly altered neuronal voltage-gated currents leading to a depression of spontaneous neuronal activity. On the other hand, STN HFS (>100 Hz stimulation) in rat brain in vitro slice preparations has been shown to induce depolarization and early rapid firing, then prolonged inhibition [28]. This silencing effect has been attributed to the inactivation of Na+-mediated action potentials. Furthermore, Do & Bean [29] found that the inherent pacemaking rhythmicity of STN neurons is due to persistent Na+ currents flowing at subthreshold voltages. HFS (70 Hz) altered the cells’ inherent rhythmicity through a slow inactivation of Na+ currents. The authors suggest that this stimulation- induced inactivation of sodium currents likely contributes to the clinical effects of DBS at a cellular level. There is also evidence that HFS can cause increased extracellular K+ levels which hyperpolarizes the neuron rendering it less excitable, an effect which may involve the stimulation of glial cells [30]. In rat brain slice preparations, STN HFS (100 Hz) has also been shown to produce long-term alterations in synaptic plasticity [31].
Anderson et al. [32] have identified two types of membrane responses to high frequency extracellular microelectrode stimulation of neurons in rodent thalamic slice preparations using intracellular recording techniques. Although all recorded neurons initially exhibited depolarization and rapid spike activity, some neurons quickly repolarized and ceased to fire (type 1), while others maintained their post-stimulation level of membrane depolarization with or without ongoing spiking (type 2). The authors were able to block the initial depolarization with the sodium channel blocker tetrodotoxin (TTX), the Ca2+ channel antagonist Cd2+, and various glutamate antagonists, including kynurenate. In contrast, the GABAA antagonist picrotoxin did not affect the neuronal depolarization or response types. Overall, their results suggest that the stimulation-induced depolarization is due to the presynaptic release of glutamate, and subsequent activation of postsynaptic glutamate receptors. Furthermore, they found that increasing the applied stimulation current increased the frequency and probability of neuronal firing. Using intraoperative microelectrodes in essential tremor patients undergoing thalamic DBS, Anderson et al. 2006 [33] found that suppression of tremor cell activity occurred in regions beyond the areas of direct current spread from the electrodes. Their results suggest that stimulation produces a functional deafferentation of afferent axons leading to a reversible synaptic depression which thus prevents tremor cell firing in the thalamus. Does DBS inhibit or excite neurons? Studies on the direct cellular effects of HFS suggest that stimulation inhibits neuronal activity [8]. Extracellular bipolar STN stimulation of in vitro rat brain slices can cause an initial increase in STN action potential firing, followed by a longer period of inhibition. In anesthetized rats administered STN stimulation (50 Hz; 300 mA), Lee et al. [34] found an initial increase in STN neuronal firing followed by a longer period of neuronal quiescence.
Stimulation physiology in functional neurosurgery
Boraud et al. [35] have reported that GPi neuronal activity is significantly increased in rhesus monkeys following treatment with MPTP, and that this increased activity is reduced to baseline firing rates with GPi HFS. Bennazouz et al. [36] have also found decreased neuronal activity in the STN and SNr of rats following STN HFS. Anderson et al. [37] recorded ventral thalamic neurons following short trains of GPi HFS stimulation in awake non-Parkinsonian monkeys. Thirty-three of 73 thalamic neurons were inhibited and seven were excited following stimulation. STN HFS (130 Hz; 500 mA; 90 ms) in rats induces c-fos expression but decreases cytochrome oxidase subunit I mRNA levels in STN neurons [38]. This overall reduction in metabolic activity could be compatible with a stimulationinduced inhibition of STN neuronal activity. It should be noted, however, that the intensity of the applied current was quite high (500 mA). In patients with Parkinson’s disease (PD), high frequency microstimulation in STN can decrease firing rates in STN neurons 600 microns from the stimulation site [39,40]. Inhibition of GPi neurons has also been found after GPi microstimulation in PD patients [41]. Pralong et al. [42] showed decreased neuronal firing of tonically active neurons in the pallidal-receiving thalamus of a patient with dystonia following GPi DBS. Other studies show that the effect of stimulation on neuronal firing is variable and can be excitatory. Benazzouz et al. [36] showed that in anesthetized rats, STN HFS led to decreased firing of STN and SNr neurons and increased firing in ventrolateral thalamic neurons in the poststimulation period. The activity of dopaminergic neurons in the SNc of anesthetized rats with and without globus pallidus lesions increased following STN HFS (130 Hz, PW 60 ms, 300 mA) [43]. Based on their previous work showing SNr inhibition with STN HFS [23,36], the authors conclude that STN HFS likely affects SNc neuronal activity by blocking the tonic inhibition of the SNr on the SNc.
83
In human patients the effect of HFS has been more variable. Brief macroelectrode STN HFS (140 Hz; 2 mA; 60 ms) in awake PD patients undergoing DBS surgery caused a decrease in the firing rate and duration of bursting activity of SNr neurons [44]. Low frequency stimulation (LFS; 14 Hz) did not affect SNr neuronal activity. Garcia et al. [45] argue that when stimulation parameters more closely mimic clinical DBS, stimulation excites STN neurons. When electromyographic responses to high frequency (100 Hz) STN stimulation were analyzed in 14 patients with parkinsonism, there was no evidence that stimulation blocked neuronal activity [46]. Instead, the ability of high frequency STN stimulation to attenuate contralateral tremor resulted from the activation of largediameter axonal fibers. Similarly, in 6 PD patients, evoked scalp potentials were recorded following high frequency STN stimulation [47]. The short chronaxie (50 ms) of the stimulated elements suggested that these elements were myelinated axons. The applicability of in vitro findings toward understanding the mechanisms of clinical DBS is unclear [25,27]. The current densities produced in animal studies are often much higher than those used in human DBS, and the slice preparations lack many of the connections, spontaneous activity, and pathological patterns of activity present in human patients, so the results may not be directly applicable to clinical DBS [8,14]. Caution must also be applied when interpreting studies involving anesthetized animals because it is unknown whether the stimulation parameters used produce the same beneficial effects seen with clinical DBS [25]. For example, when stimulation parameters shown to reduce dopamine antagonist induced catalepsy in rats are used, STN HFS increases SNr neuronal firing [48,49]. Anesthetic agents may also alter neuronal responses to stimulation. Furthermore, local inhibitory neuronal effects could still be coupled with the excitation of efferent axons passing, for example, through the STN region from the GPe to the SNr/GPi [50]. Computer modeling of HFS
1387
1388
83
Stimulation physiology in functional neurosurgery
suggests that subthreshold stimulation (relative to a single neuron at a given distance from the stimulating electrode) suppresses intrinsic firing by activating inhibitory presynaptic terminals (depending on the nucleus stimulated), whereas suprathreshold stimulation inhibits the intrinisic firing of the cell body, but generates efferent output at the stimulation frequency via direct activation of the axon arising from the neuron being modeled [24,51]. Finally, even if HFS is excitatory, it may predominantly excite GABAergic axons (e.g., from the thalamic reticular nucleus, putamen, GPe, GPi) and thus reduce neuronal firing rates in the STN, SNr, GPi or Vim [3,35,41,52]. The influence of DBS on neurotransmission will be considered below.
Effects on Neurotransmission STN region stimulation influences glutamatergic and GABA-ergic neurotransmission (see > Figure 83-1) [25,48,53]. In anesthetized rats, Maurice et al. [49] showed that STN HFS decreased firing in the majority of recorded SNr neurons. This inhibition was blocked by the GABA antagonist bicuculline, suggesting that HFS activated inhibitory GABA-ergic projections from the striatum or GPe to the SNr. In freely moving rats, HFS (124 Hz) of the caudate nucleus was associated with increased levels of extracellular GABA in adjacent areas of the caudate [54]. By lesioning the globus pallidus in rats, it is possible to abolish the STN HFS-induced rise in extracellular GABA in the SNr, suggesting that STN-HFS causes GABA-release in the SNr through the activation of pallidonigral axons and/or GPe neurons [55]. When STN HFS was administered to intact and hemi-Parkinsonian rats at an amplitude that produced contralateral limb dyskinesia, increased extracellular glutamate levels were obtained in the ipsilateral SNr [56]. When stimulation was administered below the dyskinesia-inducing
threshold, SNr glutamate levels were unaffected, but ipsilateral extracellular SNr GABA levels increased. Windels [55,57] found that STN HFS in rats leads to increased extracellular glutamate and GABA in the SNr and GPi/entopeduncular nucleus; these results suggest that STN HFS excites STN glutamatergic efferent projections to SNr and GPi as well as GABAergic projections (probably indirectly via GPe). Studies by MacKinnon et al. also suggest that STN stimulation in humans activates pallido-thalamic axons near the dorsal STN [58]. The effects of HFS were also examined in ferret brain slice preparations, in which thalamocortical relay neurons exhibited spontaneous spindle oscillations, and picrotoxin-induced 3–4 Hz absence seizure-like activity [59]. HFS (100 Hz; 10–1,000 mA; 100 ms) generated inhibitory and excitatory postsynaptic potentials, membrane depolarization, and eliminated the spontaneous spindle oscillations and 3–4 Hz absence seizure-like activity. These results suggest that HFS can disrupt abnormal neuronal activity through synaptic neurotransmitter release. Further supporting the synaptic transmission proposal, studies have demonstrated increased firing rates in the GPi following STN HFS [60], and alterations in firing rates in the thalamus following GPi HFS [37,61]. Similarly, glutamate release in the STN [62] and in the downstream GPi and SNr [56,57], has been found to be increased following STN HFS. Windels et al. [55] found that a small proportion of SNr neurons increased firing with STN HFS as stimulation frequency increased from 50 to 130 Hz. This excitation likely represents the activation of glutamatergic projections from the STN to the SNr. In murine thalamic slice preparations, HFS (125–200 Hz; 50 mA) increased the release of adenosine triphosphate (ATP), leading to increased concentrations of the ATP metabolite adenosine [63]. Adenosine A1 receptor activation and adenosine agonists in the thalamus depress excitatory neurotransmission and reduce tremor in mice. Adenosine A1 receptor-null mice experience
Stimulation physiology in functional neurosurgery
83
. Figure 83-1 Summary of potential effects of STN region DBS on neurotransmitter release in GPi/SNr. (a) STN stimulation excites GABA-ergic axons from GPe or striatum; (b) STN stimulation blocks STN neurons and axons; (c) STN stimulation excites STN axons leading to glutamate release
seizures when exposed to even low, subtherapeutic stimulation. These adenosine deficient mice also exhibit involuntary movements without stimulation. These findings provide another plausible mechanism by which thalamic HFS reduces
tremor, and suggest that the effect of HFS in other brain regions may also be due, at least in part, to the stimulation-induced accumulation of adenosine. It is important to note that caffeine, well-known to exacerbate tremor, is an adenosine antagonist.
1389
1390
83
Stimulation physiology in functional neurosurgery
In PD patients, single electrical pulses in the vicinity of GPi neurons produce a cessation of spontaneous activity for 15–25 ms [41]. Because of the latency and duration of this effect, it is likely that stimulation near the GPi causes GABA release from either GPe or striatal axons projecting onto GPi neurons, or local dendritic release of GABA resulting in GPi neuronal inhibition. Although the GPi also receives glutamatergic afferents from the STN, the more numerous GABAergic inputs are believed to overcome these excitatory signals. In patients with dystonia, neuronal firing rates in adjacent regions of GPi and in the downstream Vop decrease during GPi DBS (120 Hz; 3–4 V; 90 ms) suggesting stimulationinduced activation of pre-synaptic GABA-ergic inputs as well as GABA-ergic pallido-thalamic projections [52,61]. Computer modeling suggests that DBS can lead to the propagation of orthodromic and antidromic action potentials [64]. Experimental work in humans is compatible with this conclusion. Following single pulse stimulation of the STN in PD patients, the latency of TMS-induced MEP decreased. This facilitation could be due to orthodromic activation via the subthalamonigropallidal- thalamo-cortical circuit or antidromically via cortico-subthalamic projections [65]. STN HFS in anesthetized rats has also been found to antidromically influence corticosubthalamic projections [49]. Does DBS affect dopaminergic activity? Nigrostriatal axons abut the dorsal STN. Stimulation of the STN may excite these dopaminergic projections, and lead to increased striatal dopamine release. Microdialysis studies in rats have shown increased release of extracellular dopamine and its metabolites following STN HFS [34,66–68]. Using STN HFS in anesthetized rats, Lee et al. [34] found an increased release of striatal dopamine. The authors conclude that STN stimulation excites dopaminergic nigrostriatal projections and the striatal release of dopamine may be responsible for the beneficial clinical effects of STN DBS in patients with movement disorders.
The effects of STN HFS (130 Hz; 80 mA; 80 ms) were studied in a 6-hydroxydopamine-lesioned rat model [69]. Following chronic L-dopa treatment, the rats displayed L-dopa induced dyskinesias (LID). When STN HFS was applied to rats on L-dopa, their LIDs were exacerbated. Off L-dopa, STN HFS did not induce dyskinesias. These results contrast those described above, and suggest that STN HFS may influence L-dopa-modulated transduction pathways, but does not directly increase extracellular dopamine release. Electrophysiological studies in PD patients suggest that STN DBS may normalize pathways adversely affected by L-dopa [70]. It is unlikely, however, that the effect of clinical DBS is mediated through alterations in dopamine neurotransmission [25]. Abosh et al. investigated whether STN DBS increases the release of striatal dopamine in five patients with Parkinson’s disease [71]. Twelve hours after their last dose of L-dopa and 9 h after their DBS systems had been turned off, each patient underwent [11C]raclopride positron emission tomography (PET). Raclopride is a low affinity D2 receptor antagonist. As extracellular dopamine levels increase, raclopride is displaced from the D2 receptor which is reflected in decreased [11C]raclopride binding. Halfway through a 90 min raclopride infusion, the patients’ right STN stimulators were turned on, while their left-sided stimulators were left off. No difference in striatal [11C]raclopride binding was found between the right and left sides. Because unilateral stimulation could cause bilateral dopamine release, bilateral striatal [11C]raclopride binding prior to stimulation was compared to post-stimulation binding, but again no changes were found. With stimulation, the patients’ Unified Parkinson’s Disease Rating Scale (UPDRS) motor scores improved. Hilker et al. [72] obtained similar findings using the same paradigm. These results suggest that the primary mechanism of action of STN DBS is not striatal dopamine release. Does DBS lead to neurotransmitter depletion? Another possibility is that because DBS stimulates
Stimulation physiology in functional neurosurgery
axons, it could produce ongoing neurotransmitter release which could quickly lead to neurotransmitter depletion [13]. Subcortical HFS (125 Hz) in rat brain slices can produce initial depolarization then prolonged depression of excitation in primary motor cortex [73]. This depression was unrelated to GABA-ergic transmission and did not result from a complete block of action potentials. Instead, prolonged HFS reduced excitatory synaptic currents produced by the stimulated pathways, in keeping with a stimulation-induced depletion of excitatory neurotransmitter [50]. Conversely, HFS has been shown to replenish vesicle stores via calcium dependent mechanisms [74]. In a murine brainstem slice preparation, HFS (300 Hz) of presynaptic terminals has been shown to replenish vesicle pools through its influence on voltage-gated Ca2 + channels [74]. It is unlikely that neurotransmitter depletion is a primary mechanism of DBS. Does the effect of DBS vary with stimulation parameters? The effect of DBS depends on stimulation amplitude, pulse width, and frequency as described above. Amplitude. Increasing the current intensity results in an increase in the number of axons/ neurons activated since the effective current spread increases, but can also lead to depolarization block of neurons close to the electrode. The downstream effects can be complex since increased output should increase the magnitude of the stimulation effects on the target nucleus but if the stimulated structure is not homogeneous and/or if the stimulus spreads to affect a different nucleus or axons of passage then additional and/or opposite effects can be induced. In anesthetized rats, STN HFS (50–200 Hz; 20–300 mA; 60 ms) decreased SNr activity at low intensities, but increased it when delivered at high intensity [49]. Because the low current inhibition was blocked by the GABA antagonist bicuculline, it likely resulted from the stimulation-induced release of GABA. In thalamic slices, higher (50 mA) but not lower (10–25 mA) amplitude
83
stimulation led to the increased release of ATP, and only when delivered at frequencies in the 125–200 Hz range [63]. Pulse Width. The effect of variations in stimulation parameters to reduce contralateral wrist rigidity were tested in 10 PD patients with STN DBS [75]. When frequency was held constant, stimulation intensity could be reduced while still achieving clinical effectiveness if the pulse width was increased from 60 to 210 or 450 ms. Similarly, for any particular pulse width, increasing the frequency from 90 to 130 or 170 Hz allowed for a reduction of stimulation intensity while maintaining the desired therapeutic effect. Not surprisingly, the current intensity necessary to induce contralateral limb dyskinesia is higher when shorter pulse widths are used [56]. Frequency. When amplitude and pulse width are held constant, the magnitude of the downstream stimulation effect generally increases with increased stimulation frequency [3]. For example, researchers have found that in some situations the relation between frequency and the clinical or neuronal effects of stimulation are linear [76]. However, at higher frequencies, generally over 100 Hz, the effect of stimulation can frequently change dramatically. For example, abrupt threshold effects can be seen whereby frequency has no effect until a specific threshold of about 80–100 Hz is reached [52]. Similarly, the effect of stimulation may be abruptly blocked once an upper threshold, such as 200 Hz, is reached [30,50]. Moreover, in some situations, low frequency stimulation can have an opposite effect to high frequency stimulation [3]. For clinical applications, the beneficial effects of DBS have generally been achieved with stimulation frequencies within the range of 100–200 Hz [3,77–85]. In rats, STN stimulation induced release of glutamate in GPi and SNr is maximal at frequencies above 130 Hz, and GABA release is increased above frequencies of 60 Hz and only in the SNr [86]. Low frequency stimulation (10 Hz) of STN neurons in naive and dopamine-depleted rat
1391
1392
83
Stimulation physiology in functional neurosurgery
brain slices evoked single 10 Hz spikes, but did not significantly alter the overall ongoing neuronal activity [26]. In contrast, HFS (80–185 Hz) produced a dual effect, completely suppressing previous STN activity and evoking a recurring pattern of spike bursts that were time-locked to stimulation. Because these effects were altered by Na+ (TTX) and Ca2+ (nifedipine) channel blockers, but unaffected by glutamate or GABA antagonists, the authors conclude that STN HFS acts directly on the neuronal membrane, rather than through neurotransmission. They suggest that stimulation directly activates target structures with the effect dependent on the particular characteristics of the stimulated membranes and synapses. Low frequency (5–50 Hz) single pulse microstimulation in the GPi and SNr produces local inhibition [3,41]. This inhibition is not usually seen in the STN, and in the thalamus a brief excitatory response can occur [3]. In the thalamus, activation of local glutamatergic afferents may account for the stimulation-induced excitation. It has been suggested that the balance of glutamatergic and GABA-ergic axons in the STN cancels out any overall inhibitory or excitatory response to stimulation. High frequency (200 Hz) single pulse microstimulation in the GPi and SNr also produces local inhibition [3,41]. When the duration of HFS is increased, the post-stimulation inhibition is reduced, possibly due to the desensitization of GABA receptors [3]. STN microelectrode HFS (100–300 Hz) trains in patients with PD produces, following termination of the train, early inhibition followed by rebound excitation, and then further inhibition of some neurons in STN [39]. This prolonged inhibitory effect is thought to be due to hyperpolarization, possibly due to GABA release from GPe terminals [3,39]. In the thalamus, HFS (100–333 Hz) can also inhibit neuronal firing, particularly in neurons that exhibit spontaneous low threshold spike (LTS) bursting activity, and is likely the result of hyperpolarization.
Highlighting the relevance of frequency for the effects of DBS, in patients with ET, thalamic stimulation above 90 Hz reduces tremor, whereas stimulation below 60 Hz can exacerbate it [87]. Higher frequency stimulation produces a more regular firing pattern which correlates with tremor suppression [87]. Using stimulation parameters that mimic those used in human DBS, Kiss et al. [88] found that the response to stimulation in rodent in vitro thalamic slice preparations began when applied at a frequency of 20 Hz and then increased to a maximum responsiveness at 200 Hz, similar to the response characteristics of human Vim DBS [76,89]. In patients with essential tremor, reductions in tremor have been shown to be maximal at 100 Hz, with increasing tremor reduction between 45 and 100 Hz, and no additional reduction at frequencies above 100 Hz [76,89]. Lee et al. [50] found that the frequency of stimulation significantly altered its effect. Initial excitation of local STN neurons was maximal at 100–140 Hz. At 200 Hz, this activity was entirely blocked, producing poststimulation inhibition alone. Furthermore, the longer the stimulation was applied the longer the poststimulation inhibition persisted. Low frequency stimulation of the STN has been shown to exacerbate b frequency oscillatory activity in the GPi [90]. Using computer models, Grill et al. [91] determined that DBS below 100 Hz is unable to block intrinsic oscillatory neuronal firing patterns, whereas stimulation above 100 Hz completely eliminates it, resulting in a new regular firing pattern.
Effects on Patterns of Neuronal Activity The rate model of basal ganglia function posits that parkinsonism results from reduced rates of neuronal firing in thalamic, cortical, and brainstem components of the motor circuit [6]. The shortcomings of this model are apparent when one considers the clinical features of PD.
Stimulation physiology in functional neurosurgery
The inhibition of motor output seems to fit bradykinesia and rigidity, but not tremor. Moreover, lesions of the thalamus do not result in bradykinesia, and lesions of the GPi do not cause dyskinesias [6]. Rather than rate alone, the underlying pathophysiology of movement disorders likely involves dysfunctional patterns of activity. One notable alteration in the pattern of neuronal activity seen in movement disorders is the presence of oscillatory activity in the STN, SNr, and GPi, particularly in the b frequency range (15–30 Hz) [6,28,92–94]. Additionally, basal ganglia and cortical neurons in patients with parkinsonism exhibit excessive synchronization [6,48]. Successful pharmacological and surgical PD treatments have been associated with a reduction of these anomalous neuronal firing patterns [6,92,95]. It may be the ability of stimulation to reduce burst-firing patterns and de-synchronize pathologic oscillatory activity that is most relevant in explaining its effect [25,96–98]. By interfering with this dysfunctional activity, DBS may allow downstream areas of the motor circuit to perform more normally [6,13,14]. Dopamine alters the oscillatory activity of the STN in PD patients, causing a reduction in b frequency oscillations and an increase in gamma range oscillations (75 Hz) [13,98]. Voluntary movement causes desynchronization of oscillatory activity at 20 Hz and synchronization at 75 Hz [98]. It has been proposed that HFS (STN DBS) may promote higher frequency oscillations similar to dopamine [13]. STN DBS and L-dopa decrease the latency to desynchronization of neuronal activity in the primary sensorimotor and premotor cortices [95]. In a recent rat study of STN DBS, the animals were injected with D1 (SCH-23,390) and D2 (raclopride) receptor blockers to mimic the dopamine-depleted state of PD, and this produced catalepsy [48]. Extracellular microelectrode recordings of SNr neurons were conducted before and after
83
the neuroleptic injections. The activity of SNr neurons changed from regular tonic firing to irregular bursts of firing. STN HFS (130 Hz, 40–100 ms, 2–5 V) was able to reliably abolish the rats’ catalepsy and the abnormal bursting activity in the SNr. Based on these results, the authors conclude that the beneficial effects of STN DBS in PD patients are due to the stimulation’s ability to regulate the pathologic bursting activity in basal ganglia output structures and to restore the balance between the trans-striatal and trans-subthalamic circuits. In an MPTP-treated primate model of parkinsonism, Meissner et al. [99] found that STN HFS (130 Hz; 100 mA; 60 ms) reduced oscillatory activity in the STN. In a different study also in MPTP-treated monkeys, STN HFS (136–185 Hz; 3.5 V) produced regular short latency excitatory responses in GPe and GPi neurons following each stimulation pulse, suggesting the activation of glutamatergic STN projections [60]. This stimulation-locked pattern of increased activity persisted even when stimulation was maintained for more than 5 min. It was also associated with decreased rigidity and increased spontaneous movement in the monkeys. These results further support the theory that DBS effects depend on a disruption or alteration of pre-existing pathological patterns of neuronal activity. DBS produces widespread alterations in cerebral blood flow and metabolism [13]. Computer modeling has demonstrated that STN HFS regularizes GPi firing and restores thalamocortical responsiveness [100]. The effects of STN HFS may be polysynaptic, and effect a variety of downstream targets [48,101]. GPi DBS in patients with dystonia has been associated with alterations in cerebral blood flow in the cerebellum, anterior cingulate cortex, lentiform nucleus, thalamus, pons, and midbrain [102]. The effect of GPi DBS on the neuronal activity in the ventral oralis anterior nucleus of the thalamus (Voa) was investigated in a patient with dystonia [42]. Two types of neuronal firing patterns were
1393
1394
83
Stimulation physiology in functional neurosurgery
observed in the Voa: a low firing rate/high bursting activity type, and a high firing rate/low bursting activity type. GPi DBS reduced the firing rate and increased the bursting activity in the second type of Voa neuron. Because stimulation only affected one type of Voa neuron, it is hard to attribute the effect to the general activation of GABA-ergic pallidal efferents. Thus, while the authors do not propose a specific mechanism for the effect, they do conclude that it reflects the ability of GPi DBS to alter the pattern of pathological neuronal activity. In PD patients, STN DBS is associated with increased cerebral blood flow (CBF; measured with H215O PET) in midbrain, globus pallidus, and thalamus, and decreased CBF bilaterally in frontal, parietal, and temporal cortex [103]. These findings suggest that DBS drives STN activity which increases nigro-pallidal inhibition of thalamocortical projections. Parkinsonian patients with either STN or GPi DBS underwent H215O PET imaging while completing a motor task [104]. With therapeutically effective STN stimulation, movement-related cerebral blood flow changes were seen in the supplementary motor area, cingulate cortex, and dorsolateral prefrontal cortex (DLPFC). During effective GPi stimulation, no significant cerebral blood flow changes occurred. GPi DBS in PD patients increased regional cerebral blood flow, as assessed by H215O PET, in ipsilateral premotor cortical areas [105]. These stimulation-induced changes coincided with improvements in the patients’ rigidity and bradykinesia. In another study examining the H215O PET changes associated with GPi DBS in PD patients, stimulation-induced increases in cerebral blood flow occurred in the left sensorimotor cortex, ventrolateral thalamus, and contralateral cerebellum [106]. In addition to stimulation, participants also performed a motor task during scanning. This may explain why the pattern of activation with stimulation in this study is
somewhat different than that found by Davis et al. [105]. The mechanisms underlying these cortical blood flow changes were not investigated. Nevertheless, these results support the view that DBS influences networks of neuronal activity extending well beyond the site of stimulation, and that these alterations correlate with clinical improvement in humans. Asanuma et al. [107] examined the changes in cerebral glucose metabolism associated with STN DBS and L-dopa therapy. Using FDG-PET, nine PD patients were scanned on and off STN stimulation, and nine others were scanned before and after an intravenous infusion of L-dopa. Following the administration of STN DBS and L-dopa, metabolic reductions were found in the putamen, globus pallidus, sensorimotor cortex, and cerebellar vermis, whereas increases were seen in the precuneus. Relative to L-dopa therapy, STN stimulation was associated with metabolic increases in the STN and decreases in the medial prefrontal cortex. The metabolic alterations associated with STN DBS and L-dopa infusion correlated with clinical improvement. Garcia et al. [26,45] propose that STN HFS has both an activating and inactivating effect. While it silences previous (pathological) neuronal activity, STN HFS also establishes a new beneficial discharge pattern in the gamma frequency (60–80 Hz) range. Such an effect provides a common link between HFS and L-dopa treatment, which is also known to obliterate b frequency (20 Hz) oscillatory activity and replace it with spontaneous synchronization above 70 Hz [45,90]. These results again suggest that, regardless of the specific mechanisms, the success of STN DBS and L-dopa therapy depends on similar alterations to the pathological cerebral activity underlying PD. Vim DBS facilitates TMS-induced motor evoked potentials by activating the primary motor cortex via thalamocortical projections [108]. Vim DBS also facilitates the TMS-induced activation of inhibitory cerebellothalamocortical
Stimulation physiology in functional neurosurgery
projections. Similarly, anterior nucleus of thalamus DBS has been found to drive inhibitory thalamocortical circuits [109]. The influence of HFS on neuronal circuits depends on the specific stimulation site. For example, in a PD patient who had undergone bilateral STN DBS, motor symptoms were improved with left DBS, but episodes of dysphoria were elicited with right DBS alone [110]. Structural imaging revealed that the left electrode was in the inferior STN, whereas the right electrode was marginally superolateral to the STN. Functional magnetic resonance imaging showed that left DBS produced blood oxygen level-dependent (BOLD) increased signal in the premotor and motor cortices, ventrolateral thalamus, putamen, and cerebellum, and decreased signal in the supplementary motor cortex. With right DBS, similar but less pronounced signal changes were seen in these motor areas. However, unique signal increases were seen in the superior prefrontal cortex, Brodmann area (BA) 24, anterior thalamus, caudate, and brainstem, and decreases in the medial prefrontal cortex. These results demonstrate the effect HFS can have on widespread neuronal circuits, and how that effect relates to the neuroanatomy of the stimulation target. They also emphasize the potential range of clinical effects that HFS can influence. Do DBS and surgical lesions produce the same pattern of neuronal activity? Fukuda et al. [111] have identified a Parkinson’s disease related pattern (PDRP) of brain activity using [18F] fluorodeoxyglucose (FDG) PET, characterized by pallidothalamic and pontine hypermetabolism as well as cortical motor hypometabolism. Following unilateral subthalamotomy in PD patients, the PDRP is altered relative to the unlesioned hemisphere. Reduced glucose metabolism is also seen in the ipsilateral midbrain, GPi, ventral thalamus, and pons [112]. These results suggest that STN lesions have widespread effects on motor circuitry. The PDRP is also altered
83
following pallidotomy and GPI DBS, and this alteration is correlated with clinical improvement as measured by UPDRS motor scores [113]. GPi DBS and pallidotomy also increase glucose metabolism in the ipsilateral premotor cortex and cerebellum bilaterally. If GPi DBS inactivates GABA-ergic pallidothalamic axons (i.e., mimics a lesion), one would expect a disinhibition of thalamic neurons, and increase in thalamic spike activity. Anderson et al. [37] examined this hypothesis in two Macaca fascicularis monkeys, and found that the activity of many thalamic neurons decreased during high frequency (120 Hz) bipolar stimulation of the GPi, although some showed increased activity. The authors suggest that the therapeutic effect of GPi DBS is by the activation of stimulated neuronal elements, which may interrupt the pathophysiological corticothalamic circuits producing movement disorders such as PD. Thus, while there may be some similarities in the pattern of neuronal activity associated with stimulation and lesions, the specific mechanisms underlying these effects are not the same. Does DBS promote neurogenesis or neuroprotection? Some researchers have begun to explore the potential neurogenesis-inducing and neuroprotective effects of HFS. For example, increased hippocampal neurogenesis, as measured by the presence of 5’-bromo-2’-deoxyuridine (BrdU) neurons (i.e., new cells) has been shown in rats who had anterior nucleus of thalamus (AN) HFS (130 Hz; 2.5 V; 90 ms) compared with rats who had undergone sham surgery [114]. Furthermore, the reduction of BrdU-positive cells that followed the systemic administration of the neurogenesis suppressor corticosterone was reversed with AN HFS. These results suggest that HFS may promote neurogenesis. Although similar experiments have not been performed with basal ganglia HFS, there is some evidence that STN DBS may promote the survival of midbrain dopamine cells in MPTPtreated monkeys [115]. The mechanism underlying
1395
1396
83
Stimulation physiology in functional neurosurgery
this potential neuroprotective effect may involve stimulation-induced reductions in glutamatemediated excitotoxicty.
Summary and Conclusions DBS affects neuronal function in many ways. The effects of DBS depend on the specific composition of neural elements (axons, neuronal cell bodies, and glia) in the stimulation target, and what set of afferent and efferent pathways are associated with that target. The ways in which DBS influences its target also depend on the amplitude, pulse width, and frequency of the applied stimulation. DBS affects membrane potentials and ion conductances, the synaptic release of neurotransmitters, and general patterns of neuronal activity over widespread networks. Although DBS may have direct inhibitory effects on neuronal cell bodies, it can also excite axons [8,116]. Imaging and electrotrophysiological studies provide evidence that DBS can at least in some situations excite the output pathways from the region stimulated. The ultimate effect of DBS may not be to simply excite or inhibit a specific nucleus, however. Rather, through its combination of excitatory and inhibitory cellular, monosynaptic, and polysynaptic effects, DBS alters the pathological neuronal activity that underlies neurological conditions, such as the synchronized oscillatory patterns in movement disorders, and thereby permits the system to function more normally [3,4,8,14,90,94,116–118].
References 1. Kringelbach ML, et al. Translational principles of deep brain stimulation. Nat Rev Neurosci 2007;8(8):623-35. 2. Benabid AL, Benazzouz A, Pollak P. Mechanisms of deep brain stimulation. Mov Disord 2002;17 Suppl 3:S73-4. 3. Dostrovsky JO, Lozano AM. Mechanisms of deep brain stimulation. Mov Disord 2002;17 Suppl 3:S63-8. 4. Vitek JL. Mechanisms of deep brain stimulation: excitation or inhibition. Mov Disord 2002;17 Suppl 3:S69-72.
5. McIntyre CC, et al. How does deep brain stimulation work? Present understanding and future questions. J Clin Neurophysiol 2004;21(1):40-50. 6. DeLong MR, Wichmann T. Circuits and circuit disorders of the basal ganglia. Arch Neurol 2007;64(1):20-4. 7. Parent A, Hazrati L-N. Functional anatomy of the basal ganglia. II. The place of subthalamic nucleus and external pallidium in basal ganglia circuitry. Brain Res Rev 1995;20(1):128-54. 8. Perlmutter JS, Mink JW. Deep brain stimulation. Annu Rev Neurosci 2006;29:229-57. 9. Parent A, Hazrati L-N. Functional anatomy of the basal ganglia. I. The cortico-basal ganglia-thalamo-cortical loop. Brain Res Rev 1995;20(1):91-127. 10. Ranck JB. Jr., Which elements are excited in electrical stimulation of mammalian central nervous system: A review. Brain Res 1975;98(3):417-40. 11. Follett KA, Mann MD. Effective stimulation distance for current from macroelectrodes. Exp Neurol 1986;92(1):75-91. 12. Temel Y, et al. Monopolar versus bipolar high frequency stimulation in the rat subthalamic nucleus: differences in histological damage. Neurosci Lett 2004;367(1):92-6. 13. Lozano AM, Mahant N. Deep brain stimulation surgery for Parkinson’s disease: mechanisms and consequences. Parkinsonism Relat Disord 2004;10 Suppl 1:S49-S57. 14. Lozano AM, et al. Deep brain stimulation for Parkinson’s disease: disrupting the disruption. Lancet Neurol 2002;1(4):225-31. 15. Weiss G. The law of electrical stimulation in the nerves. Comptes Rendus Des Seances De La Societe De Biologie Et De Ses Filiales 1901;53:466-8. 16. Nowak LG, Bullier J. Axons, but not cell bodies, are activated by electrical stimulation in cortical gray matter I. Evidence from chronaxie measurements. Exp Brain Res 1998;118(4):477-88. 17. Nowak LG, Bullier J. Axons, but not cell bodies, are activated by electrical stimulation in cortical gray matter II. Evidence from selective inactivation of cell bodies and axon initial segments. Exp Brain Res 1998;118(4):489-500. 18. Holsheimer J, et al. Identification of the target neuronal elements in electrical deep brain stimulation. Eur J Neurosci 2000;12(12):4573-7. 19. Rushton WAH. The effect upon the threshold for nervous excitation of the length of nerve exposed, and the angle between current and nerve. J Physiol (Lond) 1927;63(4):357-77. 20. Dostrovsky JO, et al. Microinjection of lidocaine into human thalamus: a useful tool in stereotactic surgery. Stereotact Funct Neurosurg 1993;60(4):168-74. 21. Levy R, et al. Lidocaine and muscimol microinjections in subthalamic nucleus reverse Parkinsonian symptoms. Brain 2001;124(Pt 10):2105-18. 22. Breit S, Schulz JB. Benabid AL. Deep brain stimulation. Cell Tissue Res 2004;318(1):275-88.
Stimulation physiology in functional neurosurgery
23. Benazzouz A, et al. Responses of substantia nigra pars reticulata and globus pallidus complex to high frequency stimulation of the subthalamic nucleus in rats: electrophysiological data. Neurosci Lett 1995;189(2): 77-80. 24. McIntyre CC, et al. Cellular effects of deep brain stimulation: model-based analysis of activation and inhibition. J Neurophysiol 2004;91(4):1457-69. 25. Chang J-Y, et al. Studies of the neural mechanisms of deep brain stimulation in rodent models of Parkinson’s disease. Neurosci Biobehav Rev 2007;31(5):643-57. 26. Garcia L, et al. Dual effect of high-frequency stimulation on subthalamic neuron activity. J. Neurosci 2003;23(25):8743-51. 27. Beurrier C, et al. High-frequency stimulation produces a transient blockade of voltage-gated currents in subthalamic neurons. J Neurophysiol 2001;85(4): 1351-6. 28. Magarinos-Ascone C, et al. High-frequency stimulation of the subthalamic nucleus silences subthalamic neurons: a possible cellular mechanism in Parkinson’s disease. Neuroscience 2002;115(4):1109-17. 29. Do MTH, Bean BP. Subthreshold sodium currents and pacemaking of subthalamic neurons: modulation by slow inactivation. Neuron 2003;39(1):109-20. 30. Bikson M, et al. Suppression of epileptiform activity by high frequency sinusoidal fields in rat hippocampal slices. J Physiol 2001;531(Pt 1):181-91. 31. Shen K-Z, et al. Synaptic plasticity in rat subthalamic nucleus induced by high-frequency stimulation. Synapse 2003;50(4):314-19. 32. Anderson T, et al. Mechanisms of deep brain stimulation: an intracellular study in rat thalamus. J Physiol 2004;559(1):301-13. 33. Anderson TR, et al. Selective attenuation of afferent synaptic transmission as a mechanism of thalamic deep brain stimulation-induced tremor arrest. J Neurosci 2006;26(3):841-50. 34. Lee KH, et al. Dopamine efflux in the rat striatum evoked by electrical stimulation of the subthalamic nucleus: potential mechanism of action in Parkinson’s disease. Eur J Neurosci 2006;23(4):1005-14. 35. Boraud T, et al. High frequency stimulation of the internal Globus Pallidus (GPi) simultaneously improves parkinsonian symptoms and reduces the firing frequency of GPi neurons in the MPTP-treated monkey. Neurosci Lett 1996;215(1):17-20. 36. Benazzouz A, et al. Effect of high-frequency stimulation of the subthalamic nucleus on the neuronal activities of the substantia nigra pars reticulata and ventrolateral nucleus of the thalamus in the rat. Neuroscience 2000;99(2):289-95. 37. Anderson ME, Postupna N, Ruffo M. Effects of highfrequency stimulation in the internal globus pallidus on the activity of thalamic neurons in the awake monkey. J Neurophysiol 2003;89(2):1150-60.
83
38. Salin P, et al. High-frequency stimulation of the subthalamic nucleus selectively reverses dopamine denervationinduced cellular defects in the output structures of the basal ganglia in the rat. J Neurosci 2002;22(12):5137-48. 39. Filali M, et al. Stimulation-induced inhibition of neuronal firing in human subthalamic nucleus. Exp Brain Res 2004;156(3):274-81. 40. Welter M-L, et al. Effects of high-frequency stimulation on subthalamic neuronal activity in Parkinsonian patients. Arch Neurol 2004;61(1):89-96. 41. Dostrovsky JO, et al. Microstimulation-induced inhibition of neuronal firing in human globus pallidus. J Neurophysiol 2000;84(1):570-4. 42. Pralong E, et al. Effect of deep brain stimulation of GPI on neuronal activity of the thalamic nucleus ventralis oralis in a dystonic patient. Neurophysiol Clin 2003;33(4):169-73. 43. Benazzouz A, et al. High frequency stimulation of the STN influences the activity of dopamine neurons in the rat. Neuroreport 2000;11(7):1593-6. 44. Maltete D, et al. Subthalamic stimulation and neuronal activity in the substantia nigra in Parkinson’s disease. J Neurophysiol 2007;97(6):4017-22. 45. Garcia L, et al. High-frequency stimulation in Parkinson’s disease: more or less? Trends Neurosci 2005;28(4):209-16. 46. Ashby P, et al. Neurophysiological effects of stimulation through electrodes in the human subthalamic nucleus. Brain 1999;122(10):1919-31. 47. Ashby P, et al. Potentials recorded at the scalp by stimulation near the human subthalamic nucleus. Clin Neurophysiol 2001;112(3):431-7. 48. Degos B, et al. Neuroleptic-induced catalepsy: electrophysiological mechanisms of functional recovery induced by high-frequency stimulation of the subthalamic nucleus J Neurosci 2005;25(33):7687-96. 49. Maurice N, et al. Spontaneous and evoked activity of Substantia Nigra Pars Reticulata Neurons during high-frequency stimulation of the subthalamic nucleus. J. Neurosci 2003;23(30):9929-36. 50. Lee KH, Roberts DW, Kim U. Effect of high-frequency stimulation of the subthalamic nucleus on subthalamic neurons: an intracellular study. Stereotact Funct Neurosurg 2003;80(1–4):32-6. 51. Miocinovic S, et al. Computational analysis of subthalamic nucleus and lenticular fasciculus activation during therapeutic deep brain stimulation. J Neurophysiol 2006;96(3):1569-80. 52. Wu YR, et al. Does stimulation of the GPi control dyskinesia by activating inhibitory axons? Mov Disord 2001;16(2):208-16. 53. Li T, Qadri F, Moser A. Neuronal electrical high frequency stimulation modulates presynaptic GABAergic physiology. Neurosci Lett 2004;371(2–3):117-21. 54. Hiller A, et al. Electrical high frequency stimulation of the caudate nucleus induces local GABA outflow in freely moving rats. J Neurosci Meth 2007;159(2):286-90.
1397
1398
83
Stimulation physiology in functional neurosurgery
55. Windels F, et al. Pallidal Origin of GABA release within the substantia nigra pars reticulata during high-frequency stimulation of the subthalamic nucleus. J Neurosci 2005;25(20):5079-86. 56. Boulet S, et al. Subthalamic stimulation-induced forelimb dyskinesias are linked to an increase in glutamate levels in the substantia nigra pars Reticulata. J Neurosci 2006;26(42):10768-76. 57. Windels F, et al. Effects of high frequency stimulation of subthalamic nucleus on extracellular glutamate and GABA in substantia nigra and globus pallidus in the normal rat. Eur J Neurosci 2000;12(11):4141-6. 58. MacKinnon CD, et al. Stimulation through electrodes implanted near the subthalamic nucleus activates projections to motor areas of cerebral cortex in patients with Parkinson’s disease. Eur J Neurosci 2005;21(5):1394-402. 59. Lee KH, et al. Abolition of spindle oscillations and 3-Hz absence seizurelike activity in the thalamus by using highfrequency stimulation: potential mechanism of action. J Neurosurg 2005;103(3):538-45. 60. Hashimoto T, et al. Stimulation of the subthalamic nucleus changes the firing pattern of pallidal neurons J Neurosci 2003;23(5):1916-23. 61. Montgomery J, Erwin B. Effects of GPi stimulation on human thalamic neuronal activity. Clin Neurophysiol 2006;117(12):2691-702. 62. Lee KH, et al. High-frequency stimulation of the subthalamic nucleus increases glutamate in the subthalamic nucleus of rats as demonstrated by in vivo enzyme-linked glutamate sensor. Brain Res 2007;1162:121-9. 63. Bekar L, et al. Adenosine is crucial for deep brain stimulation-mediated attenuation of tremor. Nat Med 2008;14(1):75-80. 64. Grill WM, Cantrell MB, Robertson MS. Antidromic propagation of action potentials in branched axons: implications for the mechanisms of action of deep brain stimulation. J Comput Neurosci 2008;24:81-93. 65. Hanajima R, et al. Single pulse stimulation of the human subthalamic nucleus facilitates the motor cortex at short intervals. J Neurophysiol 2004;92(3):1937-43. 66. Paul G, et al. High frequency stimulation of the subthalamic nucleus influences striatal dopaminergic metabolism in the naive rat. Neuroreport 2000;11(3):441-4. 67. Meissner W, et al. High-frequency stimulation of the subthalamic nucleus enhances striatal dopamine release and metabolism in rats. J Neurochem 2003;85 (3):601-9. 68. Bruet N, et al. Neurochemical mechanisms induced by high frequency stimulation of the subthalamic nucleus: increase of extracellular striatal glutamate and GABA in normal and hemiparkinsonian rats. J Neuropathol Exp Neurol 2003;62(12):1228-40. 69. Oueslati A, et al. High-frequency stimulation of the subthalamic nucleus potentiates L-DOPA-induced neurochemical changes in thestriatum in a rat model of Parkinson’s disease J Neurosci 2007;27(9):2377-86.
70. Sailer A, et al. Subthalamic nucleus stimulation modulates afferent inhibition in Parkinson’s disease. Neurology 2007;68(5):356-63. 71. Abosch A, et al. Stimulation of the subthalamic nucleus in Parkinson’s disease does not produce striatal dopamine release. Neurosurgery 2003;53(5):1095-102; discussion 1102-5. 72. Hilker R, et al. Deep brain stimulation of the subthalamic nucleus does not increase the striatal dopamine concentration in parkinsonian humans. Mov Disord 2003;18(1):41-8. 73. Iremonger KJ, et al. Cellular mechanisms preventing sustained activation of cortex during subcortical highfrequency stimulation. J Neurophysiol 2006;96(2): 613-21. 74. Wang L-Y, Kaczmarek LK. High-frequency firing helps replenish the readily releasable pool of synaptic vesicles. Nature 1998;394(6691):384-8. 75. Rizzone M, et al. Deep brain stimulation of the subthalamic nucleus in Parkinson’s disease: effects of variation in stimulation parameters. J Neurol Neurosurg Psychiatry 2001;71(2):215-19. 76. Ushe M, et al. Effect of stimulation frequency on tremor suppression in essential tremor. Mov Disord 2004;19(10):1163-8. 77. Vidailhet M, et al. Bilateral deep-brain stimulation of the globus pallidus in primary generalized dystonia. N Engl J Med 2005;352(5):459-67. 78. Rodriguez-Oroz MC, et al. Bilateral deep brain stimulation in Parkinson’s disease: a multicentre study with 4 years follow-up. Brain 2005;128(Pt 10):2240-9. 79. Deuschl G, et al. A randomized trial of deep-brain stimulation for Parkinson’s disease. N Engl J Med 2006;355(9):896-908. 80. Plaha P, et al. Stimulation of the caudal zona incerta is superior to stimulation of the subthalamic nucleus in improving contralateral parkinsonism. Brain 2006; 129(Pt 7):1732-47. 81. Plaha P, Gill SS. Bilateral deep brain stimulation of the pedunculopontine nucleus for Parkinson’s disease. Neuroreport 2005;16(17):1883-7. 82. Stefani A, et al. Bilateral deep brain stimulation of the pedunculopontine and subthalamic nuclei in severe Parkinson’s disease. Brain 2007;130(Pt 6):1596-607. 83. Mayberg HS, et al. Deep brain stimulation for treatmentresistant depression. Neuron 2005;45(5):651-60. 84. Putzke JD, et al. Thalamic deep brain stimulation for tremor-predominant Parkinson’s disease. Parkinsonism Relat Disord 2003;10(2):81-8. 85. Hariz MI, et al. Multicentre European study of thalamic stimulation for parkinsonian tremor: a 6 year follow-up. J Neurol Neurosurg Psychiatry 2008;79(6): 694-9. 86. Windels F, et al. Influence of the frequency parameter on extracellular glutamate and gamma-aminobutyric acid in substantia nigra and globus pallidus during electrical
Stimulation physiology in functional neurosurgery
stimulation of subthalamic nucleus in rats. J Neurosci Res 2003;72(2):259-67. 87. Kuncel AM, et al. Amplitude- and frequency-dependent changes in neuronal regularity parallel changes in tremor with thalamic deep brain stimulation. IEEE Trans Neural Syst Rehabil Eng 2007;15(2):190-7. 88. Kiss ZHT, et al. Neuronal response to local electrical stimulation in rat thalamus: physiological implications for mechanisms of deep brain stimulation. Neuroscience 2002;113(1):137-43. 89. Ushe M, et al. Postural tremor suppression is dependent on thalamic stimulation frequency. Mov Disord 2006;21(8):1290-2. 90. Brown P, et al. Effects of stimulation of the subthalamic area on oscillatory pallidal activity in Parkinson’s disease. Exp Neurol 2004;188(2):480-90. 91. Grill WM, Snyder AN, Miocinovic S. Deep brain stimulation creates an informational lesion of the stimulated nucleus. Neuroreport 2004;15(7):1137-40. 92. Mallet N, et al. Disrupted dopamine transmission and the emergence of exaggerated beta oscillations in subthalamic nucleus and cerebral cortex. J Neurosci 2008;28(18):4795-806. 93. Hutchison WD, et al. Neuronal oscillations in the basal ganglia and movement disorders: Evidence from whole animal and human recordings. J Neurosci 2004;24 (42):9240-3. 94. Brown P. Oscillatory nature of human basal ganglia activity: relationship to the pathophysiology of Parkinson’s disease. Mov Disord 2003;18(4):357-63. 95. Devos D, et al. Subthalamic nucleus stimulation modulates motor cortex oscillatory activity in Parkinson’s disease. Brain 2004;127(2):408-19. 96. Plenz D, Kital ST. A basal ganglia pacemaker formed by the subthalamic nucleus and external globus pallidus. Nature 1999;400(6745):677-82. 97. Levy R, et al. Synchronized neuronal discharge in the basal ganglia of Parkinsonian patients is limited to oscillatory activity J Neurosci 2002;22(7):2855-61. 98. Levy R, et al. Dependence of subthalamic nucleus oscillations on movement and dopamine in Parkinson’s disease. Brain 2002;125(6):1196-209. 99. Meissner W, et al. Subthalamic high frequency stimulation resets subthalamic firing and reduces abnormal oscillations. Brain 2005;128(10):2372-82. 100. Rubin JE, Terman D. High frequency stimulation of the subthalamic nucleus eliminates pathological thalamic rhythmicity in a computational model. J Comput Neurosci 2004;16(3):211-35. 101. Kita H, et al. Balance of monosynaptic excitatory and disynaptic inhibitory responses of the globus pallidus induced after stimulation of the subthalamic nucleus in the monkey. J Neurosci 2005;25(38):8611-19.
83
102. Yianni J, et al. Effect of GPi DBS on functional imaging of the brain in dystonia. J Clin Neurosci 2005;12(2):137-41. 103. Hershey T, et al. Cortical and subcortical blood flow effects of subthalamic nucleus stimulation in PD. Neurology 2003;61(6):816-21. 104. Limousin P, et al. Changes in cerebral activity pattern due to subthalamic nucleus or internal pallidum stimulation in Parkinson’s disease. Ann Neurol 1997;42(3):283-91. 105. Davis KD, et al. Globus pallidus stimulation activates the cortical motor system during alleviation of parkinsonian symptoms. 1997;3(6):671-4. 106. Fukuda M, et al. Functional correlates of pallidal stimulation for Parkinson’s disease. Ann Neurol 2001;49(2):155-64. 107. Asanuma K, et al. Network modulation in the treatment of Parkinson’s disease. Brain 2006;129(10):2667-78. 108. Molnar GF, et al. Changes in cortical excitability with thalamic deep brain stimulation. Neurology 2005;64(11):1913-9. 109. Molnar GF, et al. Changes in motor cortex excitability with stimulation of anterior thalamus in epilepsy. Neurology 2006;66(4):566-71. 110. Stefurak T, et al. Deep brain stimulation for Parkinson’s disease dissociates mood and motor circuits: a functional MRI case study. Mov Disord 2003;18(12): 1508-16. 111. Fukuda M, et al. Networks mediating the clinical effects of pallidal brain stimulation for Parkinson’s disease: a PET study of resting-state glucose metabolism. Brain 2001;124(8):1601-9. 112. Su PC, et al. Metabolic changes following subthalamotomy for advanced Parkinson’s disease. Ann Neurol 2001;50(4):514-20. 113. Eidelberg D, et al. Regional metabolic correlates of surgical outcome following unilateral pallidotomy for Parkinson’s disease. Ann Neurol 1996;39(4):450-9. 114. Toda H, et al. The regulation of adult rodent hippocampal neurogenesis by deep brain stimulation. J Neurosurg 2008;108(1):132-8. 115. Wallace BA, et al. Survival of midbrain dopaminergic cells after lesion or deep brain stimulation of the subthalamic nucleus in MPTP-treated monkeys. Brain 2007;130(8):2129-45. 116. McIntyre CC, et al. Uncovering the mechanism(s) of action of deep brain stimulation: activation, inhibition, or both. Clin Neurophysiol 2004;115(6):1239-48. 117. Lang AE, Lozano AM. Parkinson’s disease. Second of two parts. N Engl J Med 1998;339(16):1130-43. 118. Liu Y, et al. High frequency deep brain stimulation: what are the therapeutic mechanisms? Neurosci Biobehav Rev 2008;32(3):343-51.
1399
84 Stimulation Technology in Functional Neurosurgery B. H. Kopell . A. Machado . C. Butson
Most of the physician quotes regarding neurostimulation leads of the era [1970’s and 1980’s] are unprintable. (Medtronic Archival Document 1988) The evolution of technology in functional neurosurgery is an interesting mix of rapid development and frustrating stagnation, a constant stream of ideas inspired by serendipity and knowledge from eras long past. Functional neurosurgery has always been intimately associated with cutting-edge technology. For example, the vital technique of stereotactic surgery whose principles today underlie the imagingguided surgical advances of general cranial and spinal surgery was itself first applied as a functional procedure in 1947 [1]. Perhaps the most important arrow in today’s functional neurosurgical quiver is the neurostimulation implant designed to manipulate the electrical signals in a targeted neuronal network in order to alleviate neurological symptoms. Electricity was known to have a potentially therapeutic effect as far back as the Roman Empire. Early medical documents record the experience of a physician by the name of Scribonius Largus who noticed the reduction of pain in his patients suffering from gout after they accidently stepped on an electrical-field generating torpedo fish (> Figure 84‐1) [2]. Despite this initial insight, electricity would not be used therapeutically in the nervous system again for many centuries. Instead, efforts focused on making controlled lesions in the nervous system in order to bring about permanent neuromodulatory effects. Initially, such lesions relied on physical force as Russ Meyers pioneering open-craniotomy approach to disrupting neural #
Springer-Verlag Berlin/Heidelberg 2009
pathways in the 1930s ultimately led to the development of the leukotome in the 1950s [3]. Other means of lesion-generation evolved over the next three decades and included injections of alcohol/wax/oil [4,5], thermal and cryo-energy [6,7], and ultimately radiofrequency energy [8]. The overarching advancement that each technique had was a more precise and better controlled lesion that resulted in decreased patient morbidity. Electricity, however, remained a vital component of these lesional procedures. In the very first human stereotactic procedure, Siegel and Wycis describe the method of stimulating through the lesion-generating needle to confirm that they were not within the internal capsule [1,2]. While most use of electricity during these procedures focused mainly on its usefulness to decrease patient morbidity, eventually clinical effects of electricity became more and more evident. In 1960 Hassler first reported the clinical effects of electrical current on tremor during Vim lesion procedures [2,9]. Soon after, Ron Tasker and others were the first to document carefully the parameters used and the dichotomy between ‘‘low’’ and ‘‘high’’ frequency stimulation [10]. There was a general feeling that low frequency stimulation might drive or increase involuntary movements, especially tremor, and high frequency stimulation might mimic the therapeutic lesional effect, but such observations were inconsistent [2]. Based on these observations of the ability of electrical stimulation to mimic the effects of a lesion, investigators began the era of the chronically implanted electrode. Because there was no
1402
84
Stimulation technology in functional neurosurgery
. Figure 84‐1 An electrical torpedo fish or electric ray. The name comes from the Latin ‘‘torpere,’’ to be stiffened or paralyzed, referring to the effect on someone who handles or steps on a living electric ray
way to provide power to these systems chronically, these electrodes were intermittently powered by direct percutaneous connections to an external power source and stimulus generator. J. L. Pool was the first to employ this technique by implanting silver electrodes in the head of the caudate of a patient suffering from chronic depression [11]. Others soon followed including septal stimulation for chronic pain by Heath [12] and thalamic/pallidal stimulation for movement disorders by Bechtereva [13]. The origin of modern day neurostimulation is generally traced to the publication by Ron Melzack, Ph.D. and Pat Wall, Ph.D., in Science magazine in 1965 (> Figure 84‐2). Their publication ‘‘Pain mechanisms: a new theory,’’ is more commonly known as the Gate Control Theory of Pain [14]. This theory led C. Norman Shealy, M.D., working with his colleagues at Case Institute of Technology, to implant the first Dorsal Column Stimulator (DCS) in 1967. Medtronic became involved at that time in working with Dr. Shealy to develop the first commercially
available unit for clinical investigation. The Medtronic unit was known as the Myelostat Dorsal Column Stimulator (DCS) (> Figure 84‐3) [15]. Sweet and Wespic at the same year developed an implantable system for the treatment of peripheral neuropathic pain that was ultimately commercialized by Roger Avery [16]. Avery Laboratories entered the neurostimulation area several years later, with their design of a DCS system [15]. Spinal cord stimulation (SCS) eventually was utilized beyond neuropathic extremity pain for peripheral vascular disease, angina, spasticity, tremor, and dystonia [17–20]. Shealy, in attempting to better evaluate his patients, began applying external surface stimulation by using the ElecTreat stimulator. He began using this device in the early 1970s as a screening tool for selecting patients for implant. The device had been used extensively in pre-FDA regulation days and was sold with claims that were far in excess of what the device itself could do. Medtronic designed the first external stimulator for use as a screening tool in about 1971.
Stimulation technology in functional neurosurgery
. Figure 84‐2 Ron Melzack, Ph.D. and Pat Wall, Ph.D., in Science magazine in 1965. (Picture courtesy of Medtronic and used with permission)
. Figure 84‐3 The Myelostat DCS. The first commercial spinal cord stimulation (SCS) system manufactured by Medtronic (Picture courtesy of Medtronic and used with permission)
84
. Figure 84‐4 The first commercially available deep brain stimulation (DBS) electrode manufactured by Medtronic (Picture courtesy of Medtronic and used with permission)
This device was known as a Cutaneous Stimulation Device (CSD). It was designed strictly for use by the physician or the patient in a hospital setting to screen for a later implant. Shortly thereafter, it was recognized that such devices had a therapeutic value in their own right, and could be used by the patient in a home setting. Several companies began producing such patient devices in the early 1970s. Medtronic introduced its first TENS device in 1973. This was known as the Neuromod Transcutaneous Nerve Stimulator (TNS). The terminology TNS was soon modified to Transcutaneous Electrical Nerve Stimulator (TENS) [15]. Paralleling the development of spinal cord and peripheral nerve stimulation was the development of brain stimulation. Brain stimulation began in 1969 at the University of California, San Francisco Medical Center. In that year, Dr. Yoshio Hosobuchi implanted a depth electrode in the sensory thalamus for a pain patient (> Figure 84‐4) [21]. During this time, the discovery of the endogenous opiate, endorphin, led to investigation of brain stimulation in the periventricular gray substance. Richardson and Akil reported the first use of this technique in humans [22].
1403
1404
84
Stimulation technology in functional neurosurgery
In 1976, the Medtronic Deep Brain Stimulation (DBS) system was commercially released (> Figure 84‐5) [15]. Soon DBS-type systems were being used in clinical situations for epilepsy, spasticity, and psychiatric disease [23–25]. By the late 1970s there were three companies in the US that manufactured brain stimulation technologies for chronic pain: Avery, Neuromed, and Medtronic. At this time the FDA held a joint commission including the AANS and CNS to examine the technologies [26]. Based on this meeting, it was determined that DBS systems for chronic pain (and other conditions) needed to be proven with regard to safety and efficacy and that studies to date fell short of this bar. Each of the manufacturers were given time to perform studies and provide the documentation for safety and efficacy, but two of the companies felt that such a complex study would not be cost-effective when compared to potential sales, so did not submit such a report. Only Avery documented that DBS met the requirements to be approved for pain management, but it was very shortly after, in 1983, that that Roger Avery retired
and sold the company to Bill Dobelle, who concentrated on visual cortex stimulation for blindness and phrenic nerve stimulation for use in paralyzed patients, and no longer supported DBS. Consequently, DBS for pain management was deapproved and has not yet recovered [2]. This seeming death-knell for brain stimulation implants, ironically, heralded the rebirth of DBS technology, this time in the form of brain stimulation for movement disorders. In 1980 Brice and McLellan reported the first use of DBS for movement disorders with their exploration of thalamic stimulation for MS-type tremor [27]. Soon after this, Alim Benabid’s pioneering work in exploring Vim DBS for tremor and STN DBS for PD led ultimately to an FDA approval of this technology and sparked a renaissance in the neuromodulation field and market [28–30]. There have been several seminal advances in neurostimulation technology in the past three decades. The early stimulators came in two parts. These systems utilize an external transmitter which contains the pulse generating circuitry and power source (battery). This energy is
. Figure 84‐5 The first commercially available deep brain stimulation (DBS) system manufactured by Medtronic (Picture courtesy of Medtronic and used with permission)
Stimulation technology in functional neurosurgery
transmitted through the skin using a radio frequency coupled link. Inside the body is a passive (no battery) receiver which decodes the radio frequency signal and delivers it to the electrode and, therefore, the nervous tissue. Cordis market released the first totally implantable neurostimulator (IPG, or implantable pulse generator) in 1982. Medtronic introduced a fully programmable neurological pulse generator later that year [15]. Other landmark improvements have included increased number of electrode contacts for flexibility in targeting and programming these systems, IPGs capable of controlling more than one electrode array, and transcutaneously rechargeable power sources, extending the time between IPG replacement surgeries.
The Neurostimulation System Today FDA approved neurostimulation systems are the follows: Spinal Cord Stimulation for neuropathic extremity pain and failed back syndrome (1984 510(k)); Peripheral Nerve Stimulation for neuropathic extremity pain (1984 510(k)); Vagus Nerve Stimulation for refractory epilepsy (1997); Vim thalamic DBS for Parkinsonian and Essential tremor (1997); STN/ GPi DBS for Parkinson’s disease (2002); GPi DBS for primary generalized dystonia (limited FDA approval under a Humanitarian Device Exemption); Vagus Nerve Stimulation for treatment resistant Major Depression (2005). In this section, we discuss the currently available types of electrodes and implantable pulse generators. This section reflects the available technology at the first quarter of 2008 and will unquestionably become outdated soon. Nevertheless, some principles are expected to influence the future generations of devices and the reader will be able to apply the information found in this chapter to new devices that will become available in the near future.
84
Implantable Pulse Generators (IPG) Implantable pulse generator output can be voltage-controlled or current-controlled. Currentcontrolled devices adjust the voltage for a given system impedance in order to maintain the desired current output. In this sense, the clinical results obtained with constant current devices are less subject to changes in impedances over time because the voltage will be automatically adjusted to compensate for any such changes. In voltage-controlled systems, the device will automatically adjust the current to compensate for impedances. However, this approach could be problematic if the impedance changes over time: impedance values that fall too low could cause high current densities at the electrode, possibly resulting in tissue damage; impedance values that are too high will result in a large voltage drop at the electrode-tissue interface and decreased effectiveness of stimulation. The implications of safety and the stimulation method will become clearer in the next few sections. Implantable pulse generators can be adjusted by telemetry for activation of individual contacts (as cathodes or anodes) and adjustments in output amplitude, pulse width and frequency. Implantable pulse generators were initially limited in frequency output to a maximum of 185 pps. More recently, rechargeable spinal cord stimulation devices provide greater flexibility, allowing for a maximum of 1200 pps. Pulse widths longer than 500 s are rarely used and so far only one SCS allows for 1,000 s. The amplitude ranges for stimulation range from 0 to 25 mA in different devices. Rechargeable systems have lithium ion batteries that must be periodically recharged with an external radiofrequency antenna. The advantage of the system is longer duration of the batteries (up to 9 years) without need for replacement. A disadvantage is the greater patient compliance
1405
1406
84
Stimulation technology in functional neurosurgery
requirement in keeping with a relatively time consuming recharging schedule. New rechargeable systems also allow for multiple programs displayed in several menu screens in the patient programmer device. Although this technology increases the flexibility for post-operative programming, these menus can be counterintuitive to individuals unfamiliar with current technologies. In addition to multiple programs, it is possible to fractionalize the output of the pulse generators between different stimulation programs or even individual contact fractionation of amplitude. In the latter, each contact has an independent output from the generator and its amplitude can be modulated regardless of the amplitude output to the other active contacts. There is a fast trend for miniaturization of the implantable pulse generators among all competing manufactures. While volumes of IPGs range from 22 to 51 cc, by the time of printing of this text a smaller IPG may already be available.
. Figure 84‐6 The current generation DBS electrodes. The Medtronic 3387 model has 1.5 mm contact spacing and the 3389 model has 0.5 mm spacing. Each individual contact is 1.5 mm high
Leads Brain (subcortical): There are currently only two approved deep brain stimulation leads, both quadripolar. Both have the 1.27 mm diameter and either a 28 or 40 cm length, of which only a segment of 7–12 cm is typically implanted intracranially. Each individual contact is 1.5 mm high. The only specification difference between the two models is the spacing between the contacts. The Medtronic 3387 model has 1.5 mm contact edge-to-edge spacing and the 3389 model has a 0.5 mm edge-to-edge spacing. Hence, the total quadripolar array length (distance from the tip of the most distal contact to the top of the most proximal contact) for the two models is 10.5 and 7.5 mm respectively (> Figure 84‐6). Spinal cord: A large range of lead specifications exist for both percutaneous and paddle (surgical) electrodes.
The initial SCS leads were inserted via a laminotomy approach. Due to the inherent invasiveness of open laminotomy, soon after these systems became commercially available, screening methods were requested. As a result, Medtronic designed the Model 3600 Acute Percutaneous Screening System. This system was the forerunner of present day percutaneous spinal cord stimulators. It used a single, monopolar electrode which was introduced into the epidural space through a needle. The lead was brought percutaneously through the skin and an external TENS electrode was used as the indifferent. This system was used percutaneously by the patient in the hospital for a week’s screening before the decision to implant was made. This system was introduced in about 1973 and was used until 1975 [15].
Stimulation technology in functional neurosurgery
The concept of a percutaneously inserted epidural electrode, which could be converted for chronic stimulation, was developed in 1975. Clinical trials began that year with the Medtronic PISCES Spinal Cord Stimulator. The system was fully market released in the spring of 1976. Avery Laboratories, meanwhile, introduced a similar system which they called their PENS Spinal Cord Stimulator [15]. Cylindrical (percutaneous) electrodes are typically used for minimally invasive procedures aimed at deploying the leads to the epidural space through large caliber needles (15G and larger). The greatest advantage of this methodology is that it virtually does not disrupt the structure of the spinal elements for implantation, except for the needle penetration through the ligamentum flavum. It is well tolerated by most patients under local anesthesia with or without sedation. Furthermore, the ability to adjust the electrodes to the proper point on the spinal cord under local anesthesia using the patient’s verbal feedback to localize electrode location improves clinical effect (> Figure 84‐7). Cylindrical electrodes can be used as trial electrodes, to be removed after a period of a few days once the efficacy of spinal cord stimulation for the patient’s pain syndrome is determined. Alternatively, cylindrical electrodes are also adequate for long term implantation, connected to implantable pulse generators. Cylindrical electrodes vary in diameter from 0.73 to 1.35 mm. Although the total length of a lead influences the implantation strategy, implanters are more commonly concerned about the contact array length, which is the distance from the tip of the most distal contact to the top of the most proximal contact. The contact array length varies
84
depending on the number of contacts per lead (four or eight), height of each contact and contact interspacing. For quadripolar electrodes, the most compact interspacing is 4 mm for a total array span of 24 mm (4 mm contact height). The longest array length, with 6 mm contact height and 12 mm interspacing, is 60 mm. In octopolar electrodes, the most compact interspacing is 1.1 mm, resulting in a total array length of 32 mm. The longest array length is 66 mm, resulting from a combination of 3 mm contact height and 6 mm interspacing. Total length of spinal cord stimulation leads vary dramatically, from 22 to 75 cm. Shorter leads may be easier to steer in the epidural space but most likely will require an extension wire to reach the site of implantation of the IPG. Longer leads may be directly connected to the IPG, assuming a typical scenario of low thoracic SCS lead implantation and abdominal or posterior hip IPG implantation. Cervical lead implantation will frequently require an extension to connect to the IPG without tension in the cables. Paddle-type SCS leads initially were not developed for the treatment of pain. In 1973, Cook and Weinstein reported improvements in the spasticity of their multiple sclerosis patients who had been implanted for pain control. As a result of this observed improvement, several investigators, most notably Don Dooley, M.D., from Miami, reported on a variety of patients with movement disorders. Joe Waltz, M.D. from St. Barnabas Hospital in New York, extended the work of Cook and Dooley to patients with movement disorders resulting from cerebral palsy and trauma. Medtronic worked with Waltz to design a four-plate electrode to be implanted by laminotomy into the epidural space. This electrode was introduced into clinical studies for
. Figure 84‐7 An example of a cylindrical (percutaneous) SCS electrode. This example of SCS electrode is manufactured by Boston Scientific
1407
1408
84
Stimulation technology in functional neurosurgery
movement disorders (spasticity/tremor) in 1978 and eventually became approved by the FDA for the indication of pain in 1982 [15]. Paddle electrodes are preferred over percutaneous electrodes by some clinicians, attempting to achieve more stable and consistent stimulation [31–33]. These electrodes are also often used to replace percutaneous electrodes that have broken or migrated from its original position. Paddle electrodes models vary most significantly in the number of contacts (4–16), number of contact columns (1–3) and length and width of the paddle. The
. Figure 84‐8 Multiple examples of paddle-type SCS leads. These examples are manufactured by Advanced Neuromodulation Systems/St. Jude Medical (Picture courtesy of ANS and used with permission)
smallest paddle electrode is 39 mm in length and 3.8 mm in width, with a single quadripolar column (1 4). The largest has 74 mm length and 13 mm width, with two octopolar columns (2 8). In this lead, each contact is 4.1 mm long and 2.6 mm wide for a total 56 mm array length and 7.2 array width. The most compact 16 contact electrode lead has two octopolar columns distributed over an array length of 34.7 mm and width of 4.7 mm. In addition to single or double columns of four or eight electrodes, there are also three column, 16 contact leads arranged in one central six contact column flanked by five contact columns on either side. In theory, the purpose of these electrodes is to allow for more efficient midline stimulation but, in practice, the tripole configuration has not yet been proven to enhance outcomes (> Figure 84‐8). Peripheral/Cranial nerve: The currently approved peripheral nerve stimulation leads are of the paddle-type and very similar in design to the ones used for spinal cord stimulation. They may also have additional material along the periphery of the electrode array to facilitate anchoringof the lead to the tissue surrounding a targeted peripheral nerve (> Figure 84‐9). The only FDA-approved cranial nerve stimulating electrode is for the vagus nerve. It is a two contact helical electrode designed to wrap around the vagus nerve. It comes in two sizes
. Figure 84‐9 An example of a peripheral nerve stimulating (PNS) lead. The mesh on the periphery is used to anchor the lead to the surrounding tissue (Picture courtesy of Medtronic and used with permission)
Stimulation technology in functional neurosurgery
one for the adult and one for the pediatric populations (> Figure 84‐10). Brain (Surface): There are no FDA-approved electrodes designed specifically to stimulate the surface of the cerebral or cerebellar cortex. Furthermore, there are no FDA approved indications for cortical stimulation. However, there are a number of ‘‘off-label’’ investigational uses for cortical stimulation starting with the pioneering work of Irving Cooper with stimulation of the anterior cerebellar cortex [34]. Chronic stimulation of the cerebral cortex was published by Tsubokawa in 1991 for central neuropathic pain [35]. Other indications that have been published include movement disorders, epilepsy, motor recovery after stroke and tinnitus [36–39]. In general, investigators use paddle-type SCS electrodes to stimulate their targets. . Figure 84‐10 A vagus nerve stimulating lead
84
Tissue Safety: ElectroBiocompatibility of Neurostimulation Implants (CB) Biocompatibility of Materials Used in Neurostimulation Implants Implantable neuromodulation electrical stimulation systems are designed to be safe and effective. Hardware effectiveness can be defined as capacity to generate action potentials in the targeted group of neurons acutely and chronically. It should not be confused with therapeutic effectiveness, which is related to the desired clinical effects with electrical stimulation. Safety issues are not only limited to biocompatibility of the hardware materials in contact to the body but also include a multitude of concerns ranging from damage of neural or surrounding tissues to preventing leakage of harmful substances such as those that compose the battery. Injury to the nervous tissue can result from direct mechanical damage of the chronically implanted electrode upon the surrounding tissue or from electrical stimulation. Although these safety concerns take into assumption that the implantation procedure was uneventful, longevity and durability of the hardware influences the overall surgical risk if repeat surgical interventions are necessary to troubleshoot hardware failure or replace parts. In this sense, even battery life influences the long-term risk of an implantable neurostimulation system as frequent re-interventions necessary to replace depleted implantable pulse generators are more prone to result in complications such as infections or perioperative medical complications. The materials selected for assembling a neuromodulation device can significantly influence safety and effectiveness. In a systematic literature review, Merrill et al. [40] indicate that stimulating electrodes must meet biocompatibility, mechanical adequacy, corrosion, stimulation toxicity and durability requirements. All these
1409
1410
84
Stimulation technology in functional neurosurgery
concerns are influenced by the expected duration of the implantation. Acute implantation followed by removal of the stimulating electrode is less likely to result in tissue damage, electrode corrosion, mechanical failure or stimulation related toxicity than chronic stimulation. Nonetheless, most current standard of care neuromodulation therapies and emerging clinical applicationsfor electrical stimulation with implantable hardware are aimed at alleviation of chronic disorders in, often, otherwise young and healthy individuals. Hardware systems such as deep brain stimulators and spinal cord stimulators should, in principle, outlive the patient and hence need to last several decades. Device parts must be resistant enough not break with repetitive motion at the implanted site (i.e., cervical spine) nor suffer corrosion during these extended implantation periods. Biocompatible materials should not cause necrosis or severe foreign body responses in the surrounding neural tissue. Copper and silver electrodes, although reported in animal research models [41–44], are likely to result in tissue damage [45,46] and should be avoided in clinical devices. Commercially available, FDA approved deep brain and spinal cord stimulation leads have metal electrodes (contacts) composed of platinum iridium alloy. Platinum has been demonstrated to be biocompatible in [47,48]. Iridium is less malleable than platinum, reducing electrode flexibility. Both metals have a large reversible charge storage capacity allowing a larger amount of current to be passed in the electrode before irreversible Faradaic reactions occur at the electrodetissue interface. Irreversible reactions can lead to tissue damage if toxic elements are accumulated or can result in corrosion of the electrode itself, both of which are undesirable. Accumulation of irreversible reactions is also influenced by other factors, including waveform design. Although biphasic square charge balanced waveforms are standard in neuromodulation devices, balancing the cathodic and anodic phases does
not guarantee a perfect electrochemical balance at the electrode-tissue interface. If the anodic phase is longer than necessary for completely reversing the electrochemical process generated in the first phase, the potential at its end will be positive in relation to the pre-stimulation potential. This may lead to accumulation of irreversible reactions that may result in electrode corrosion. Although insulating materials are not subject to the same electrochemical charges as the exposed electrodes, they must also be biocompatible and result in minimal tissue reaction. Current deep brain stimulation and spinal cord stimulation leads use polyurethane for insulation, which has been demonstrated to be minimally reactive [49,50]. Although encapsulation response is expected after implantation of any electrode (or catheter) in the nervous tissue or epimeningeal spaces, it is desirable that chronic inflammation and microglial response should be minimal and that the capsule around the probe should be thin [40,46]. Extension wires and implantable pulse generator cases may not be in direct contact to neural tissue but still need to be inert enough not to cause symptomatic local reactions to the skin or subcutaneous tissue. In addition to platinum-iridium, extension cables may have nickel alloy MP35N and insulation with polyurethane, nylon or silicone, all of which have been demonstrated to be biocompatible or minimally reactive [51–53]. Implantable pulse generator cases are composed of titanium. Allergic and foreign body reactions have been reported to implantable hardware [54,55] and skin allergy kits are available to test a patient’s reactivity to the individual components of the implantable hardware. Note that although non-biocompatible materials may be encased by biocompatible materials that are exposed to human tissue, there is always the risk of breakage or leakage of toxic materials. In this sense, it is advantageous if even internal parts are made of materials that are not toxic to nervous or non-nervous human tissue.
Stimulation technology in functional neurosurgery
Electrical Safety Issues The safety of a neurostimulation system with regard to electrical issues depends on many factors. First, the metal in contact with the tissue must be biocompatible and able to transfer quantities of charge sufficient for therapeutic benefit. Second, the combination of frequency, amplitude and pulse width of the stimulation waveform must be below the damage threshold. Third, if it is to be chronically implanted it must be able to operate safely regardless of the stimulus duration. Damage from stimulation is generally caused by irreversible reactions near the electrode-tissue interface that are a direct result of the different charge transfer mechanisms in the electrode and the tissue: charge is carried by electrons in the stimulation system, but by ions in biological tissue. Despite this multitude of factors, safety considerations in neurostimulation systems can be distilled to two basic principles: 1. 2.
Avoid tissue damage Avoid electrode damage
In many cases, both of these objectives are met by limiting the charge density at the electrodetissue interface. Therefore, one basic component of safe stimulation limits is the relationship between charge density and the possibility of damage. While much is known about the reversible and irreversible reactions that take place at the electrode-tissue interface, the precise boundary between safe and unsafe stimulation is not well understood and has been characterized from a very limited set of empirical data. Additionally, these calculations use several simplifications. First, they assume uniform current density across the electrode contact; however, it is well known that current density is non-uniform with higher density at the edges. Second, they assume that the electrode-tissue interface behaves as a linear device, while routine stimulation requires
84
electrodes to be operated in the nonlinear region where many different electrochemical events are occurring in the metal surface [56]. Nonetheless, a useful system has emerged from this data for predicting safe stimulation limits [57,58]. The basis of this system is to use charge density and charge per phase as cofactors to determine the threshold for tissue damage. This method assumes symmetric waveforms composed of charge-balanced cathodic and anodic pulses [59]. With this approach the electrode is left in the same chemical condition prior to the waveform and as a result there is a much lower probability of tissue damage compared to monophasic pulses. The requisite parameters for this method are the current amplitude, pulse duration and electrode surface area. Once these are know, the charge density in mC/cm2/phase is calculated from D ¼ I PW =A where I is current, PW is the pulse width and A is the surface area. Charge per phase in mC/phase is calculated from Q ¼ I PW Lastly, these two quantities are interrelated with logðDÞ ¼ k logðQÞ which provides a family of curves for different values of k (> Figure 84‐11). Experimental observations have shown damage for values of k = 2 or higher, while no damage was observed for k = 1.5 [58]. To put this in context for neurostimulation, a variety of example data points are included in > Figure 84‐11 for DBS electrodes (1.5 mm height, 1.27 mm diameter, 0.06 cm2 area) and cortical disc electrodes (4 mm diameter, 0.13 cm2 area) at 5 mA stimulus amplitude with a range of pulse widths. It is noteworthy that this approach does not directly take into account stimulation frequency or duration. Rather, the data used to create this system was collected from a variety of experiments of varying frequencies and durations.
1411
1412
84
Stimulation technology in functional neurosurgery
. Figure 84-11 Charge density and charge per phase are cofactors in determining tissue damage. Area where damage has been observed is indicated by gray shading which is above k = 1.9 (ref to Shannon). Also shown are data points for DBS and cortical disc electrodes at pulse widths from 200 to 500 s
Most neurostimulation systems use polarizable electrodes such as Pt/Ir, and therefore charge-transfer takes place via a capacitive, displacement current rather than a direct, faradaic current. It is known that increasing the capacitive properties of the total current flow minimize electrode damage [56], which suggests that damage is causes by faradaic rather than capacitive currents. Each metal or alloy has specific properties, one of which is the charge-carrying capacity of the metal. For Pt/Ir this value is approximately 200 mC/cm2, which is far greater than the threshold for tissue damage. However, the impedance increases as the electrode size decreases, and larger voltages are necessary to overcome the energy barrier at the electrodetissue interface. Larger voltages bring the electrode closer to the water window, wherein anodic polarization causes oxygen gas evolution, while cathodic polarization can result in hydrogen gas evolution. These reactions cause both tissue damage and corrosion of the electrode. Hence, the threshold for neuronal or electrode damage can be reached by increasing the stimulus amplitude,
increasing the stimulus duration or decreasing the size of the electrode contact. Researchers have also examined the effects of stimulation in vivo. One set of studies found that if electrodes that were used to create a stimulusinduced lesion in one part of the brain were implanted elsewhere, they continued to cause lesions without passing any additional current [60]. This suggests that one component of tissue damage results from toxic reactions at the metal surface, which may be acting in additional to thermal or electric-field induced damage. More recently a study was conducted involving postmortem examinations of the brains of eight PD patients who had DBS of the VIM or STN for up to 70 months, and who were programmed with maximum stimulation settings of 4.4 V, 185 Hz, and 120 ms [61]. They used a variety of stains and antibodies to examine myelin sheaths, connective tissue, axons, fibrillary astrogliosis, glial fibrillary acidic protein, neurofilament protein, synaptophysin, and microglia. They found similar results in all patients who underwent longterm stimulation. Tissue changes around the active contact and nonstimulated areas adjacent to the insulated parts of the lead did not differ. Around the lead track, a thin inner capsule of connective tissue was noted. The thickness of this fibrous sheath ranged from 5 to 25 mm, with no correlation to duration of stimulation. A narrow rim of fibrillary gliosis of less than 500 mm abutted the fibrous capsule. In the adjacent brain tissue, a zone of less than 1 mm showed loosely scattered glial fibrillary acidic protein–positive protein astrocytes. They concluded that clinical long-lasting benefit of DBS correlates with the absence of progressive gliotic scar formation, and does not cause damage to the adjacent brain regions. A second group examined the effects of different charge densities and durations on STN DBS in rats [62]. Tissue damage was quantified from morphometric assessment of brain sections stained with Cresyl Violet, which was used to visualize the general tissue reaction, and
Stimulation technology in functional neurosurgery
Fluoro-Jade, which was used to evaluate cell viability and neuronal degeneration. Animals were stimulated at 130 Hz, 60 ms pulse width with charge densities from 0 to 26mC/cm2/phase. They found that while stainless steel electrodes caused significant neuronal damage, Pt/Ir electrodes did not cause damage regardless of the charge density during short term (4 h) stimulation. Nor did they observe damage due to PtIr electrodes during 72 h of stimulation at a modest stimulation intensity of 3mC/cm2/phase. Our understanding of the safety issues related to neurostimulation will continue to evolve as more experimental data becomes available. The method explained above for determining stimulation thresholds for tissue damage is a simplification of a complex, nonlinear system that depends on many factors. However, this approach has proven useful for a variety of neurostimulation devices. Current generation neurostimulation systems pose significant challenges with regards to magnetic resonance imaging (MRI) after implantation. DBS is a partial contraindication to MRI. Scanning a DBS patient without following the manufacturer’s recommendations, especially with a body coil or without a transmit/receive headcoil can result in heat-induced lesions of brain tissue and has been associated with serious permanent neurological deficits [63]. Head coils have been found to generate less heat than body coils [64]. Phantom studies have addressed the risk of heating in DBS implant simulations. Coiling of the excess DBS lead in small concentric loops around the burr hole have been shown to reduce the risk of heating [65]. Similar experiments performed by the same group also demonstrated that the SAR reported by the MRI console is not completely reliable in predicting the risk of heating in DBS implants [66], indicating that safety testing should be performed individually for each MRI system to be used for imaging patients with DBS implants. In 2006, the DBS system manufacturer released a series of recommendations for
84
MR imaging of patients with implanted DBS systems. These included turning the pulse generator output to OFF and setting the stimulation mode to bipolar with the amplitude at 0 V. As for the MRI examination, the recommendation is to use parameters that limit the displayed average head SAR to 1/10 (0.1) W/kg [67]. A recent publication has suggested that this maximum SAR recommendation is excessively low, based on a single institution series of 1071 MRI exams performed in 405 patients with 746 implanted systems. In this series, 1.5-Tesla MRI with an SAR of up to 3 W/kg was used and no adverse events were reported. This series does indicate that higher than recommended SAR may be safe but cannot be completely extrapolated to other MRI systems in the same or other institutions [68]. Publications regarding MRI scanning of patients with SCS or PNS implants are even more scarce and limited only to brain scans in which the stimulation lead and IPG are completely excluded from the transmit/receive head-coil.
Stimulation Protocols The fundamental goal when prescribing neurostimulation therapy is to maximize patient benefit while minimizing side effects. This goal is achieved through judicious selection of the stimulation protocol, which consists of the location of the electrodes, the configuration of cathodes and anodes, and the stimulation waveform. Electrode locations are first chosen during preoperative planning and are sometimes refined during intraoperative recording and stimulation. Once the electrodes are implanted, each of them is assigned one of three states when the IPG is programmed: cathode, anode or inactive. Finally, the stimulation waveform is chosen to control the amplitude of current or voltage through each of the active electrode contacts. The complex interplay of these three components determines the extent of neural activation relative to the
1413
1414
84
Stimulation technology in functional neurosurgery
surrounding anatomical targets. Fortunately, this complexity can be distilled to a few basic principles which can be used to understand the electric field resulting from the stimulation protocol and the neural response to the applied electric field. Over the past few decades several authors have published concise descriptions of the effects of extracellular stimulation on neural elements [69–71]. These descriptions are rooted in physiological observations and careful characterization of the behavior of excitable tissue. More recently, techniques have been developed to make detailed quantitative predictions of the electric field and neural response, and these methods have been applied on an individual patient basis to correlate the effects of stimulation with clinically measured outcomes [72]. These techniques take into account each component of the stimulation protocol, beginning with the electrode location. The current strategy in surgical planning is to implant the stimulating electrodes in close proximity to an anatomical target which is chosen based on the patient disease state. Since most neurostimulation systems have multiple electrodes, this strategy provides maximum flexibility when post-operatively programming the IPG to achieve the therapeutic objectives. Ideally, the size and shape of the electrodes are well matched to the anatomical target: strips or grids are used to cover areas of cortex; multiple cylindrical contacts are used for depth electrodes. In some cases the type and orientation of the electrode can be matched to the shape of the target nucleus. For example, the subthalamic nucleus, which is a common target for DBS, might be best matched to cylindrical electrode leads with small spacing between contacts. In contrast, the external segment of the globus pallidus would be better matched with larger spacing between contacts, and by orienting the electrode along the major axis of the nucleus (roughly in the dorsal-ventral direction). In each of these examples, activation of the anatomical target can then be optimized
by selecting the active electrodes and the stimulation waveform, and to understand this process it is useful to first describe the electric field induced in the tissue. A basic conceptual model of the implanted neurostimulation system includes a voltage or current source, the resistive tissue medium and a reference or return electrode (> Figure 84‐12a). In most systems the stimulating electrode is cathodic (negative) and the reference electrode is anodic (positive), due to the fact that cathodic stimulation is generally much more effective than anodic stimulation (i.e., results in lower neural activation thresholds). The behavior of this system is governed by Ohm’s Law, V ¼ IR where V is voltage, I is current and R is resistance. In 3-dimensional space this relationship is expressed by the Poisson equation, which takes into account the complex anisotropic and inhomogeneous properties of biological tissue. This equation is used to determine voltage as a function of distance from the source electrode, and predicts that voltage falls with a 1/r distribution where r is the distance from the source. While this may be a useful conceptual model for understanding the neurostimulation system, it fails to take into account a critical feature: the electrodetissue interface. The effects of this feature can be demonstrated with two examples. First, during acute cortical stimulation, such as with electrode grids implanted for electrocorticography, there is a layer of highly conductive cerebrospinal fluid (CSF) between the electrode and the tissue which can result in shunt currents that flow around the brain rather than through it. In a second example, chronically implanted depth electrodes such as those used in DBS become surrounded by a layer of high-resistance encapsulation tissue resulting from the foreign body reaction [73]. While the thickness of this layer in the CNS is generally less than that found in the PNS, it is still large enough to drastically increase the
Stimulation technology in functional neurosurgery
84
. Figure 84‐12 Conceptual model of neurostimulation. (a) Basic model includes a voltage or current source, the resistive tissue medium and a distant return (for monopolar stimulation). (b) First principles predict that voltage will fall off with 1/distance (white line), but the voltage drop at the electrode-tissue interface causes voltage to fall off much faster (red line). (c) The fall off is used to scale the time-dependent stimulation waveform according to distance from source, but this fails to take into account the effects of electrode capacitance during voltage-controlled stimulation. (d) The effects of tissue capacitance during current-controlled stimulation and (e) Each of these effects will reduce the voltage observed by the tissue, and subsequently the volume of tissue activated during stimulation
impedance at the electrode-tissue interface. In both of these examples there is a substantial voltage drop across electrode-tissue interface, which in turn causes a reduction in the voltage amplitude observed by the tissue. Hence, the first principle of neurostimulation is that electrodetissue interface causes voltage to fall off faster than we would expect from the tissue properties alone (> Figure 84‐12b). A second component to add to the conceptual model is the biophysical tissue properties.
The simplest representation of neural tissue is that of an isotropic, homogeneous medium which would result in a roughly spherical spread of current away from the stimulation contact. However, there is substantial variability in the properties of brain tissue. White matter is highly conductive along the primary axis of the myelinated fibers but less conductive perpendicular to this axis. Grey matter tends to be roughly isotropic, and CSF is completely isotropic. Theoretical and clinical studies have demonstrated that current spreads
1415
1416
84
Stimulation technology in functional neurosurgery
preferentially along regions of high conductivity, resulting in an asymmetric flow of current away from the source [74], which is the second principle second principle of neural activation. Let us return to our conceptual model with the added encapsulation layer at the electrodetissue interface and biological tissue properties. While our model now more accurately represents the spatial spread of current, we now are faced with a different limitation. Clinical stimulation waveforms are time-dependent, and are almost invariably charge-balanced and biphasic for safety reasons that will be explained shortly. Hence, they consist of an initial cathodic pulse followed by a brief interpulse interval and an anodic pulse (> Figure 84‐12c). However, when we examine our model we note that it predicts voltage as a function of distance but does not take time into account. We could rectify this by scaling the stimulation amplitude up or down according to the voltage distribution (> Figure 84‐12c), but in doing so we would miss another important component of the system, specifically the capacitance of the electrode-tissue interface and the tissue itself. During voltage-controlled stimulation the electrode capacitance causes a decay in the plateau region of the stimulus pulse (> Figure 84‐12d) [75]. For electrodes with large surface area such as DBS (about 6 mm2) and electrocorticography (ECOG) (about 12 mm2) this effect will only be noticeable at longer pulse widths, but will become more pronounced at progressively shorter pulse widths as the electrode size decreases. Further, brain tissue is not only resistive, it also has capacitive properties and these will affect the time course of the stimulation waveform primarily during current controlled stimulation (> Figure 84‐12e). The reasons why electrode capacitance is the primary effect during voltage-controlled stimulation and tissue capacitance during current-controlled stimulation are beyond the scope of this text, but for further details the reader is referred to [75]. However, these effects can be succinctly
summarized in the third principle of neural activation: the amount of charge injected into the tissue during the cathodic phase is less than would be predicted from the stimulation waveform alone; the differences between the two can be calculated from basic properties of the tissue and stimulation system. With the help of this conceptual model we are now better able to understand how the physiological response to stimulation is governed by the interaction between the electric field and surrounding neural elements. Subsequent to implanting the electrodes, the clinical team begins an iterative process of configuring the active electrodes and titrating the stimulation waveform. The philosophy behind this method is that the anatomical target chosen for surgery coincides with the neural stimulation target. Therefore, optimally activating this target with minimal spillover into adjacent areas could provide good symptomatic relief with minimal side effects. This process usually begins with monopolar or bipolar stimulation in the target region. Monopolar stimulation refers to instances where the stimulating electrode is located near the anatomical target with a distant, anodic return electrode that is often the IPG case; bipolar stimulation refers to active electrode contacts in close proximity. The effects of bipolar stimulation are generally more focal than monopolar stimulation, and are preferentially oriented between the cathode and anode. In contrast, monopolar stimulation effects spread radially out from the electrode, resulting in larger volumes of activation with little directional specificity. In both cases, current from the electrodes will travel preferentially along paths of high conductivity such as CSF and the long axis of fiber tracts. The goal during the programming process is to guide the electric field such that the highest current density is in the target region but falls off quickly elsewhere. The final piece of the stimulation protocol is the waveform, the basic parameters of which are
Stimulation technology in functional neurosurgery
the cathodic amplitude, cathodic pulse width and stimulation frequency. At this point it is useful to take into account the different time scales involved in the physiological response. The most excitable neural elements are pyramidal cells and large myelinated axons, and these are also the first to respond to extracellular stimulation. For a given neural target such as a myelinated axon, the amplitude and pulse width interact to determine the pulse width and amplitude necessary to reach threshold as shown earlier. As the stimulus amplitude or pulse width increase a larger volume of tissue is activated. Subsequently, the activity of these elements changes the ongoing activity in the surrounding network. Of the available parameters for neurostimulation, frequency is the least well understood and the most disease dependent. There has been considerable debate in the neuromodulation community over the primary effects of stimulation, specifically whether they are excitatory or inhibitory. Much of the evidence for this debate stems from empirical observations. For example, the early work of Galvani on ‘‘animal electricity,’’ the success of cardiac pacemakers, and ECOGinduced muscle twitches provide good evidence that stimulation has an excitatory effect. On the other hand, DBS often has effects that are similar to surgical ablation, and cortical stimulation is sometimes used to prevent seizures, both of which imply an inhibitory effect. The answer to this debate is most likely captured in the name of this subfield of research. Neuromodulation is the therapeutic alteration of activity in the nervous system by means of implanted devices. Some research suggests that the immediate response to neurostimulation is excitatory [76–78], and that the subsequent effects on the surrounding network are modulatory. In cases such as traumatic brain injury this could be an excitatory effect [79]. However, in diseases of artificial synchronization such as Parkinson’s disease or epilepsy, while the symptomatic response appears to be inhibitory (e.g., tremor suppression or seizure
84
arrest), there is evidence to suggest that the local response is mediated by disrupting pathological activity [80,81]. Consequently the clinician is faced with a wide range of options when choosing frequency. Long-term stimulation at frequencies over 100 Hz has been shown to be effective for movement disorders [82]. In contrast, low frequency activity has been evaluated for seizure arrest [83]. The effects of frequency will likely be an area of rich research for years to come. The current approach to selecting the stimulation protocol is a serial process: first the electrode location is selected and the lead implanted, then the anodes and cathodes are configured, and the stimulation waveform is titrated. However, in the future this is likely to change in two ways. First, the targets of stimulation will become more specific in response to their physiology and primary symptoms. Second, new electrode designs will be implemented to meet the neurostimulation objectives in different anatomical targets [84]. Lastly, neurostimulation systems have historically been open-loop – they provide a fixed stimulation protocol that is independent the physiologic state of the patient. For example, DBS systems provide the same stimulation protocol continuously to the patient until changed by the clinician; some newer systems allow the patient to turn the IPG on or off, or make controlled changes to the stimulation amplitude using a small handheld device. In contrast, closed-loop systems use measurements from the body to modulate the stimulation protocol. To date there have been few neurostimulation systems that incorporate feedback, which is partly due to the increased complexity that is required. The use of feedback requires implantation of additional electrodes that provide stable, chronic recordings. Additionally, the IPG must be designed to interpret those signals and making real-time adjustments to the stimulation protocol. For example, a neurostimulation system for seizure arrest must have recording electrodes to capture ongoing neural activity, a detection
1417
1418
84
Stimulation technology in functional neurosurgery
algorithm to interpret those signals, stimulating electrodes implanted near the seizure focus, and a stimulation waveform that enables seizure arrest. This closed-loop system must operate in real-time without human intervention in order to stop seizures before they spread.
Effects of Extracellular Electricity on Neural Elements The exact mechanisms that underlie the clinical effects of the various forms of neurostimulation remain a matter of debate and investigation. The breadth of this subject is vast and beyond the scope of this chapter although a brief overview will help illustrate the capabilities of the current generation of technology. Much of the theories of how these devices work to bring about their ameliorative effects on neurological maladies have been the result of empirical observation supplemented by neurophysiologic investigations and functional imaging insights. It is undeniable that the clinical effects of neurostimulation modalities resemble their lesional counterparts. It is also self-evident that any lesion will decrease local neural activity by virtue of the fact that there are less neural elements at the site of a lesion once it is made. While this is not the case with neurostimulation systems, some evidence exists why neurostimulation modalities may resemble their lesional counterparts. With regards to the DBS modality, inhibition has been demonstrated to play a role at the level of the soma in neurons exposed to extracellular HFS. Several studies have shown the suppression of STN neuronal activity during STN HFS [85,86]. Based on this observation the effect of this net inhibition would be to remove the effect of increased and abnormal patterns of neuronal activity emanating from the nuclear area being stimulated. Also in support of inhibition both historical and recent evidence have implicated
synaptic failure as a consequence of electrical stimulation associated with DBS. In 2002, Urbano et al. performed a series of experiments looking at the effects of HFS in the thalamus, similar to clinical DBS parameters, on cortical projections [87]. The axons, themselves, were demonstrated to be able to follow high frequency stimulation trains as high as 120 Hz. In contrast, the cortical responses began to decrement starting at frequencies of 60 Hz and higher. SCS AND PNS have also been demonstrated to have mechanisms consistent with a lesion-like effect. Microdialysis studies in the dorsal horns of rats undergoing SCS have demonstrated that SCS decreased local release of glutamate and increased release of GABA [88]. Furthermore WDR neurons in DH of rodents undergoing SCS also show decreased firing rates compared to the off-stimulation state [89]. Similar studies with PNS have demonstrated suppression of A-d fiber nociceptive processing with clinically effective stimulation [90]. However, despite the clinical similarities of stereotactic and functional lesions (e.g., thalamotomy, cordotomy, rhizotomy, neurectomy) and neurostimulation modalities, their mechanisms cannot simply be reduced to a ‘‘lesional effect.’’ One argument against the lesional effect is illustrated by the Marsden Paradox: stereotactic lesions and focal neurostimulation in a supposedly already ‘‘hyperinhibited’’ motor thalamus do not further impair voluntary movement [91]. Studies have underscored this paradox. Electrical stimulation can affect multiple regions of the neuron: dendrite, soma, axon hillock, and axon. In 2003, Shen et al. found multiple effects of high frequency stimulation (HFS) on synaptic function, inhibitory and facilitatory, further adding to the complexity of the effects of electrical stimulation in a neural network [92]. Depending on the extent of what portions of the neuron are being affected by electrical stimulation, experimental data exists to support excitation, inhibition, and changes in network synchrony as
Stimulation technology in functional neurosurgery
mechanisms underlying neurostimulator function. Clearly, more complex effects are critical to the observed clinical results. Chronaxie experiments dating back to the 1970s have clearly demonstrated the effects that extracellular current has on neural tissue. Historical data based on chronaxie experiments have shown that the most sensitive element in gray and white matter undergoing extracellular HFS equivalent to parameters used in clinical neurostimulation is the axon and that stimulation drives rather than inhibits the activity of this structure [71,93]. Insights from these studies have shown that, in general, fibers are more sensitive to stimulation than soma and that the larger diameter fibers are more sensitive than smaller diameter ones. Other studies have demonstrated that the polarity of the contact used also effects the population of elements activated. In general, neural elements perpendicular to an anodal contact or parallel to a cathodal contact tend to be most sensitive to extracellular current [94]. Experimental and clinical evidence from neurostimulation studies corroborates these laboratory findings. Hashimoto et al. have shown that DBS in the STN had the net effect of increasing the mean firing rate of GPi neurons, implying the activation of glutamatergic subthalamopallidal projections [95]. A similar effect was seen with GPi stimulation with reduction in firing rates in the VL thalamus secondary to excitation of inhibitory pallidothalamic GABAergic projections [96]. Looking at concentrations of neurotransmitters ‘‘downstream’’ from implanted STN electrodes, Windels et al. found a significant increase in glutamate in GPi and SNr and a significant increase in GABA in SNr following trains of high-frequency stimulation [97]. In 2000, Baker and Montgomery expanded on this excitatory effect of DBS in their model of DBS mechanisms. This excitation effect may also be fundamental to facilitating a proposed model of DBS mechanism utilizing the concept of
84
stochastic resonance in which a subthreshold normal signal, lost in the noise of a deranged neural network, is amplified by the addition of a regular noise (in this case HFS) by an constructive interference paradigm [98]. Besides simple excitation and inhibition effects of HFS there is a growing body of evidence that DBS produces alterations in oscillatory behavior in the networks undergoing HFS. Hashimoto et al. showed a time-locked alteration of firing patterns in GPi neurons receiving projections of STN neurons influenced by DBS [95]. Devos et al. have shown that this effect on oscillatory activity is reflected at the cortical level in patients undergoing STN DBS for PD [99]. This effect can be explained by taking into account the data above demonstrating the excitatory nature of HFS on axonal projections combined with the orthodromic and antidromic anatomical connections of STN to the cortex. The growing body of functional imaging data appears to corroborate the excitatory influence of DBS on neural networks. PET and fMRI studies have consistently demonstrated increased metabolism/BOLD signal changes in various structures along the subcortical network described above with STN and GPi DBS such as the putamen, pallidum, subthalamic nucleus, and thalamus [66,100,101]. As these increases reflect local changes in synaptic activity, this corroborates the presumed driving effect of DBS on axonal elements [102]. Furthermore this increase in local synaptic activity has also been demonstrated at cortical areas directly connected to this subcortical network, especially SMA in the case of STN DBS and primary motor cortex in the case of Vim DBS [103]. Inhibition of local metabolic and by implication synaptic activity has also been demonstrated in the context of Vim DBS for tremor [104]. In the SCS/PNS experience, one obvious clinical situation that corroborates the axonal stimulation mechanism is that the cathode is universally considered the clinically ‘‘active contact.’’
1419
1420
84
Stimulation technology in functional neurosurgery
Furthermore, the seminal Gate theory of Melzack and Wall is based on the notion that axonal stimulation of DC fibers is the sine qua non of SCS mechanisms. Computer models of SCS also implicate that activation of DC fibers are essential to the mechanisms underlying SCS and that anodal blocking or inhibition of fiber firing is unlikely to be occurring [105]. Finally, rTMS data have demonstrated cortical plasticity changes with clinical successful SCS further implicating the role of neural element activation as an important mechanism [106]. Thus, there are apparent conflicting data in the literature regarding the inhibitory or excitatory effects of neurostimulation devices. Certainly inherent differences in experimental paradigms may explain some contradictions. However, the beneficial effects of neurostimulation may involve all of these apparently contradictory mechanisms. Grill and McIntyre have demonstrated the variable effect on extracellular HFS on neural elements by using DBS as an example [107]. The effective current density decreases along the radial distance from a DBS electrode. The gradient of current density could explain the apparent contradictions presented above. Very close to the epicenter of the area of effect of a DBS electrode would be an area of higher current density. In this region, stimulation may indeed be above the level of somatic activation and could lead to a depolarization block of somatic elements. In addition the axons themselves could fire at a 1:1 ratio between stimulus and axonal spike and result in the ultimate synaptic failure of synapses downstream at frequencies greater than 100 Hz [87]. Further away from this epicenter the current density would decrease. The simulation current could be below the chronaxie of somatic effect while still activating axonal elements, both of passage and emerging from the target nucleus. These axonal elements, due to the lower current density, may fire at ratios less than 1:1 and be below the frequency that would lead to synaptic failure.
The net result could indeed be the overall alteration of patterns of activity to a more regular pattern that is better tolerated by the system and manifested by improved motor behavior. Further refinement of the waveforms of the electrical stimulus, beyond the rectified square wave currently used in DBS, may allow a highly selective activation of different components of the neuron [108]. While the mechanisms of neurostimulation systems remains to be fully elucidated, what is becoming clearer is that these devices can exert both inhibitory and excitatory influences on the targets in which they are implanted. These targets are themselves a part of a larger distributed network so that the net influence of the neuromodulation system goes well beyond local effects.
Stimulation Technology: Future Directions The ‘‘currency’’ of communication between individual elements of the nervous system is electricity. While processes related to cellular and molecular biology do underlie the genesis of neural disease, many neurological disorders can be experimentally and clinically characterized by disordered electrical patterns of activity. As such, if these disordered patterns are returned to a more nominal pattern then the neurological behavior associated with the disease will also be ameliorated. Stem cell and gene therapies are revolutionary in their concept and, no doubt, hold the ultimate keys to many of these illnesses. The dichotomy of utility of these two approaches to neurological disease can be analogized to the situation with computer programming. Stem-cell and gene therapies can be likened to low-level programming languages like assembly code that directly address the ‘‘hardware.’’ However, just like the situation with assembly code, the complex nature of the ‘‘hardware’’ makes the process incredibly complex and challenging.
Stimulation technology in functional neurosurgery
Neurostimulation to follow this analogy is like a high-level computer language, easier to manipulate and achieve end results. If we are able to manipulate the electrical signals in the nervous system to an exquisite degree, in many functional neurological disorders where degeneration is not an issue, it very well may not matter we are not directly addressing the neural cellular machinery. Of course there need not be a firm divide between these two neuromodulatory approaches. In 2007, Aravanis et al. described a potential system utilizing a device-based and gene-therapy approach. By transfecting channelrhodopsin-2 (ChR2), an algal light-activated ion channel into CNS neurons, a solid-state laser delivery system could result in neurostimulation systems with a degree of control that is well-beyond the types of systems we have today [109]. We are, however, quite a ways off from being able to manipulate the electrical signals in a very specific fashion. Indeed, in many ways we are no more adept at neurostimulation today then we were a decade or even two decades ago. A quick glance at some electrodes from the 1970s and 1980s is eerily reminiscent of today’s electrodes (> Figure 84‐13). With the exception of rechargeability, IPGs today are basically similar to ones used in the early 1980s. The material science behind these devices is virtually identical and fraught with issues with regards to MR safety. Instead, much of the progress in neurostimulation has been in the realm of expanded indications rather than in the technology itself. Even despite this increased activity in expanding the use of neurostimulation, the list of approved indications had not radically changed in the last decade. While regulatory issues are certainly a factor, is the lack of approved indications a limitation of neurophysiological knowledge or a limitation of the current technology? In 1977, Phil Gildenberg at the WSSFN meeting in Sao Paolo made a remark that addressed this question: ‘‘The engineers can give us any [device] we need. We just have to know what to ask for. . .’’ [2].
84
. Figure 84‐13 A collection of neurostimulation leads from the 1970s and 1980s
One of the elements lacking in today’s neurostimulation systems is the lack of sensing capability. In the cardiac stimulation arena, approximately a decade passed between the introduction of the first open-loop implantable pacemaker and a closed-loop system. In the same manner that a pacing pulse during certain times of the cardiac cycle may be irrelevant or even counterproductive, so too might be the situation with brain pacing. The brain’s activity is organized on both a spatial and temporal basis. Devices that take advantage of this temporal organization may very well be more efficient and effective than the open-loop systems of today. Despite these limitations and challenges, neurostimulation devices have allowed functional neurosurgeons an ability to modulate nervous activity with arguably more efficacy and certainly more safety than ever before. This technology will continue to evolve to the benefit of millions of patients.
1421
1422
84
Stimulation technology in functional neurosurgery
References 1. Spiegel EA, Wycis HT, Marks M, Lee AJ. Stereotaxic apparatus for operations on the human brain. Science 1947;106:349-50. 2. Gildenberg PL. Evolution of neuromodulation. Stereotact Funct Neurosurg 2005;83:71-9. 3. Obrador S, Dierssen G. [Surgery of the pallidal region in Parkinson’s syndrome; personal technic and early results in the first six cases of surgery]. Rev Clin Esp 1956;61:229-37. 4. Narabayashi H, Shimazu H, Fujita Y, Shikiba S, Nagao T, Nagahata M. Procaine-oil-wax pallidotomy for double athetosis and spastic states in infantile cerebral palsy: report of 80 cases. Neurology 1960;10:61-9. 5. Cooper IS. Chemopallidectomy and chemothalamectomy for parkinsonism and dystonia. Proc R Soc Med 1959;52:47-60. 6. Gildenberg PL. Variability of subcortical lesions produced by a heating electrode and cooper’s balloon cannula. Confin Neurol 1960;20:53-65. 7. Cooper IS, Lee AS. Cryostatic congelation: a system for producing a limited, controlled region of cooling or freezing of biologic tissues. J Nerv Ment Dis 1961;133:259-63. 8. Cosman ER, Nashold BS, Bedenbaugh P. Stereotactic radiofrequency lesion making. Appl Neurophysiol 1983;46:160-6. 9. Hassler R, Riechert T, Mundinger F, Umbach W, Ganglberger JA. Physiological observations in stereotaxic operations in extrapyramidal motor disturbances. Brain 1960;83:337-49. 10. Tasker R, Organ L, Hawrylshyn P. The thalamus and midbrain of man. A physiological atlas using electrical stimulation. Springfield, IL: Charles C Thomas; 1982. 11. Pool J. Psychosurgery in older people. J Am Geriatr Soc 1954;2:456-65. 12. Heath R. Exploring the mind-brain relationship. Baton Rouge, LA: Moran Printing; 1996. 13. Bechtereva N, Bondarchuk A, Smirnov V. Therapeutic electrostimulations of the deep brain structures. Vopr Neirokhir 1972;1:7-12. 14. Melzack R, Wall P. Pain mechanisms: a new theory. Science 1965;150:971-9. 15. Mullett K. Medtronic archive document 1988. 16. Sweet WH, Wepsic JG. Treatment of chronic pain by stimulation of fibers of primary afferent neuron. Trans Am Neurol Assoc 1968;93:103-7. 17. Augustinsson LE, Carlsson CA, Holm J, Jivegard L. Epidural electrical stimulation in severe limb ischemia. Pain relief, increased blood flow, and a possible limbsaving effect. Ann Surg 1985;202:104-10. 18. Mannheimer C, Augustinsson LE, Carlsson CA, Manhem K, Wilhelmsson C. Epidural spinal electrical stimulation in severe angina pectoris. Br Heart J 1988;59:56-61.
19. Siegfried J. [2 different aspects of neurosurgical treatment of spasticity. Cerebral intervention and stimulation of the posterior spinal cord]. Neurochirurgie 1977;23:344-7. 20. Davis R, Gray E. Technical factors important to dorsal column stimulation. Appl Neurophysiol 1981;44:160-70. 21. Hosobuchi Y, Adams JE, Rutkin B. Chronic thalamic stimulation for the control of facial anesthesia dolorosa. Arch Neurol 1973;29:158-61. 22. Richardson DE, Akil H. Pain reduction by electrical brain stimulation in man. Part 1: acute administration in periaqueductal and periventricular sites. J Neurosurg 1977;47:178-83. 23. Cooper IS. The effect of chronic stimulation of cerebellar cortex on epilepsy in man. In: Cooper IS, Riklan M, Snider RS, The cerebellum, epilepsy and behavior. New York: Plenum; 1974. 24. Davis R, Gray E, Ryan T, Schulman J. Bioengineering changes in spastic cerebral palsy groups following cerebellar stimulation. Appl Neurophysiol 1985;48:111-16. 25. Dieckmann G. Chronic mediothalamic stimulation for control of phobias. In: Hitchcock E, Ballantine H, Meyerson B. Modern concepts in psychiatric surgery. New York: Elsevier/North-Holland Biomedical Press; 1979. pp 85-93. 26. Gildenberg P. Symposium on the safety and clinical efficacy of implanted neuroaugmentive devices. Appl Neurophysiol 1977;40:69-240. 27. Brice J, McLellan L. Suppression of intention tremor by contingent deep-brain stimulation. Lancet 1980:1221–2. 28. Benabid AL, Pollak P, Louveau A, Henry S, de Rougemont J. Combined (thalamotomy and stimulation) stereotactic surgery of the vim thalamic nucleus for bilateral Parkinson disease. Appl Neurophysiol 1987;50:344-6. 29. Benabid A, Pollak P, Gross C, Hoffman D, Benazzouz A, Gao DM, Laurnet A, Gentil M, Perret J, Feuerstein C. Stimulation of subthalamic nucleus acutely changes clinical status in Parkinson’s disease. Soc Neurosci Abstr 1993;19:1052. 30. Limousin P, Pollack P, Benazzous A, Hoffman D, LeBas J-F, Broussolle E, Perret JE, Benabid A-L. Effect on parkinsonian signs and symptoms of bilateral subthalamic nucleus stimulation. Lancet 1995;345:91-5. 31. North RB, Kidd DH, Petrucci L, Dorsi MJ. Spinal cord stimulation electrode design: a prospective, randomized, controlled trial comparing percutaneous with laminectomy electrodes: part ii – clinical outcomes. Neurosurgery 2005;57:990-6; discussion 990–6. 32. North RB, Kidd DH, Olin JC, Sieracki JM. Spinal cord stimulation electrode design: prospective, randomized, controlled trial comparing percutaneous and laminectomy electrodes – part i: technical outcomes. Neurosurgery 2002;51:381-9; discussion 389–90. 33. Villavicencio AT, Leveque JC, Rubin L, Bulsara K, Gorecki JP. Laminectomy versus percutaneous electrode
Stimulation technology in functional neurosurgery
34. 35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
placement for spinal cord stimulation. Neurosurgery 2000;46:399-405; discussion 405–6. Cooper IS. Effect of chronic stimulation of anterior cerebellum on neurological disease. Lancet 1973;1:206. Tsubokawa T, Katayama Y, Yamamoto T, Hirayama T, Koyama S. Chronic motor cortex stimulation for the treatment of central pain. Acta Neurochir Suppl (Wien) 1991;52:137-9. Priori A, Lefaucheur JP. Chronic epidural motor cortical stimulation for movement disorders. Lancet Neurol 2007;6:279-86. Fountas KN, Smith JR, Murro AM, Politsky J, Park YD, Jenkins PD. Implantation of a closed-loop stimulation in the management of medically refractory focal epilepsy: a technical note. Stereotact Funct Neurosurg 2005;83:153-8. Brown JA, Lutsep HL, Weinand M, Cramer SC. Motor cortex stimulation for the enhancement of recovery from stroke: a prospective, multicenter safety study. Neurosurgery 2006;58:464-73. Friedland DR, Gaggl W, Runge-Samuelson C, Ulmer JL, Kopell BH. Feasibility of auditory cortical stimulation for the treatment of tinnitus. Otol Neurotol 2007;28:1005-12. Merrill DR, Bikson M, Jefferys JG. Electrical stimulation of excitable tissue: design of efficacious and safe protocols. J Neurosci Meth 2005;141:171-98. Hirakawa K, Hunter-Duvar IM. Chronic electrical stimulation of the chinchilla cochlea. Scanning Electron Microsc 1984:1413–17. Sabato S, Agresta CA, Freeman GM, Salzman SK. Safety versus efficacy of spinal cord stimulation for the generation of motor-evoked potentials in the rat. J Neurotrauma 1991;8:27-44. Stone RG, Wharton RB. Simultaneous multiple-modality therapy for tension headaches and neck pain. Biomed Instrum Technol/Assoc Advancement Med Instrum 1997;31:259-62. Syka J, Popelar J. Inferior colliculus in the rat: neuronal responses to stimulation of the auditory cortex. Neurosci Lett 1984;51:235-40. Dymond AM, Kaechele LE, Jurist JM, Crandall PH. Brain tissue reaction to some chronically implanted metals. J Neurosurg 1970;33:574-80. Stensaas SS, Stensaas LJ. Histopathological evaluation of materials implanted in the cerebral cortex. Acta Neuropathol 1978;41:145-55. Motta PS, Judy JW. Multielectrode microprobes for deep-brain stimulation fabricated with a customizable 3-d electroplating process. IEEE Trans Biomed Eng 2005;52:923-33. Niparko JK, Altschuler RA, Xue XL, Wiler JA, Anderson DJ. Surgical implantation and biocompatibility of central nervous system auditory prostheses. Ann Otol Rhinol Laryngol 1989;98:965-70.
84
49. Oloffs A, Grosse-Siestrup C, Bisson S, Rinck M, Rudolph R, Gross U. Biocompatibility of silver-coated polyurethane catheters and silver-coated dacron material. Biomaterials 1994;15:753-8. 50. Sakas DE, Charnvises K, Borges LF, Zervas NT. Biologically inert synthetic dural substitutes. Appraisal of a medical-grade aliphatic polyurethane and a polysiloxane-carbonate block copolymer. J Neurosurg 1990;73:936-41. 51. Heggers JP, Kossovsky N, Parsons RW, Robson MC, Pelley RP, Raine TJ. Biocompatibility of silicone implants. Ann Plast Surg 1983;11:38-45. 52. Vince V, Thil MA, Veraart C, Colin IM, Delbeke J. Biocompatibility of platinum-metallized silicone rubber: in vivo and in vitro evaluation. J Biomater Sci 2004;15:173-88. 53. Wolfaardt JF, Cleaton-Jones P, Lownie J, Ackermann G. Biocompatibility testing of a silicone maxillofacial prosthetic elastomer: soft tissue study in primates. J Prosthet Dent 1992;68:331-8. 54. Ochani TD, Almirante J, Siddiqui A, Kaplan R. Allergic reaction to spinal cord stimulator. Clin J Pain 2000;16:178-80. 55. Oh MY, Abosch A, Kim SH, Lang AE, Lozano AM. Long-term hardware-related complications of deep brain stimulation. Neurosurgery 2002;50:1268-74; discussion 1274–6. 56. Dymond AM. Characteristics of the metal-tissue interface of stimulation electrodes. IEEE Trans Biomed Eng 1976;23:274-80. 57. McCreery DB, Agnew WF, Yuen TG, Bullara L. Charge density and charge per phase as cofactors in neural injury induced by electrical stimulation. IEEE Trans Biomed Eng 1990;37:996-1001. 58. Shannon RV. A model of safe levels for electrical stimulation. IEEE Trans Biomed Eng 1992;39:424-6. 59. Lilly JC, Hughes JR, Alvord EC, Jr, Galkin TW. Brief, noninjurious electric waveform for stimulation of the brain. Science 1955;121:468-9. 60. Rowland V, Macintyre WJ, Bidder TG. The production of brain lesions with electric currents. II. Bidirectional currents. J Neurosurg 1960;17:55-69. 61. Haberler C, Alesch F, Mazal PR, Pilz P, Jellinger K, Pinter MM, Hainfellner JA, Budka H. No tissue damage by chronic deep brain stimulation in Parkinson’s disease. Ann Neurol 2000;48:372-6. 62. Harnack D, Winter C, Meissner W, Reum T, Kupsch A, Morgenstern R. The effects of electrode material, charge density and stimulation duration on the safety of highfrequency stimulation of the subthalamic nucleus in rats. J Neurosci Meth 2004;138:207-16. 63. Henderson JM, Tkach J, Phillips M, Baker K, Shellock FG, Rezai AR. Permanent neurological deficit related to magnetic resonance imaging in a patient with implanted deep brain stimulation electrodes for Parkinson’s disease:
1423
1424
84 64.
65.
66.
67. 68.
69.
70. 71.
72.
73.
74.
75.
76.
77.
Stimulation technology in functional neurosurgery
case report. Neurosurgery 2005;57:E1063; discussion E1063. Sharan A, Rezai AR, Nyenhuis JA, Hrdlicka G, Tkach J, Baker K, Turbay M, Rugieri P, Phillips M, Shellock FG. MR safety in patients with implanted deep brain stimulation systems (dbs). Acta Neurochir Suppl 2003;87:141-5. Baker KB, Tkach J, Hall JD, Nyenhuis JA, Shellock FG, Rezai AR. Reduction of magnetic resonance imagingrelated heating in deep brain stimulation leads using a lead management device. Neurosurgery 2005;57:392-7; discussion 392–7. Phillips MD, Baker KB, Lowe MJ, Tkach JA, Cooper SE, Kopell BH, Rezai AR. Parkinson disease: pattern of functional MR imaging activation during deep brain stimulation of subthalamic nucleus – initial experience. Radiology 2006;239:209-16. Medtronic. MRI guidelines for medtronic deep brain stimulation systems, 2006, Larson PS, Richardson RM, Starr PA, Martin AJ. Magnetic resonance imaging of implanted deep brain stimulators: experience in a large series. Stereotact Funct Neurosurg 2008;86:92-100. Jayakar P. Physiological principles of electrical stimulation. In: Devinsky O, Berix A, Dogali M, editors. Electrical and magnetic stimulation of the brian and spinal cord. New York: Raven Press; 1993. Lorente de No´ R. A study of nerve physiology. Rockefeller Institute for Medical Research, New York, 1947. Ranck JB, Jr. Which elements are excited in electrical stimulation of mammalian central nervous system: a review. Brain Res 1975;98:417-40. Butson CR, Cooper SE, Henderson JM, McIntyre CC. Patient-specific analysis of the volume of tissue activated during deep brain stimulation. Neuroimage 2007;34:661-70. Butson CR, Maks CB, McIntyre CC. Sources and effects of electrode impedance during deep brain stimulation. Clin Neurophysiol 2006;117:447-54. Butson CR, Cooper SE, Henderson JM, McIntyre CC. Predicting the effects of deep brain stimulation with diffusion tensor based electric field models. Med Image Comput Comput Assist Interv 2006;9(Pt 2):429-37. Butson CR, Cooper SE, Maks CB, Wolgamuth BR, McIntyre CC. Model-based analysis of the effects of electrode impedance, electrode capacitance and 3-D tissue electrical properties on the volume of tissue activated by deep brain stimulation. In: Workshop on NINDS Neural Interfaces, Bethesda, MD, 2005, Holsheimer J, Demeulemeester H, Nuttin B, de Sutter P. Identification of the target neuronal elements in electrical deep brain stimulation. Eur J Neurosci 2000;12:4573-7. Miocinovic S, Parent M, Butson CR, Hahn PJ, Russo GS, Vitek JL, McIntyre CC. Computational analysis of subthalamic nucleus and lenticular fasciculus activation
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
during therapeutic deep brain stimulation. J Neurophysiol 2006;96:1569-80. Hashimoto T, De Elder CM, Long MR, Vitek JL. Responses of pallidal neurons to electrical stimulation of the subthalamic nucleus in experimental parkinsonism. Mov Disord 2000;15 Suppl 3:31. Schiff ND, Giacino JT, Kalmar K, Victor JD, Baker K, Gerber M, Fritz B, Eisenberg B, O’Connor J, Kobylarz EJ, Farris S, Machado A, McCagg C, Plum F, Fins JJ, Rezai AR. Behavioural improvements with thalamic stimulation after severe traumatic brain injury. Nature 2007;448:600-3. Grill WM, Snyder AN, Miocinovic S. Deep brain stimulation creates an informational lesion of the stimulated nucleus. Neuroreport 2004;15:1137-40. Morrell M. Brain stimulation for epilepsy: can scheduled or responsive neurostimulation stop seizures? Curr Opin Neurol 2006;19:164-8. Volkmann J, Moro E, Pahwa R. Basic algorithms for the programming of deep brain stimulation in Parkinson’s disease. Mov Disord 2006;21:S284-9. Halpern CH, Samadani U, Litt B, Jaggi JL, Baltuch GH. Deep brain stimulation for epilepsy. Neurotherapeutics 2008;5:59-67. Butson CR, McIntyre CC. Role of electrode design on the volume of tissue activated during deep brain stimulation. J Neural Eng 2006;3:1-8. Filali M, Hutchison WD, Palter VN, Lozano AM, Dostrovsky JO. Stimulation-induced inhibition of neuronal firing in human subthalamic nucleus. Exp Brain Res 2004;156:274-81. Tai CH, Boraud T, Bezard E, Bioulac B, Gross C, Benazzouz A. Electrophysiological and metabolic evidence that high-frequency stimulation of the subthalamic nucleus bridles neuronal activity in the subthalamic nucleus and the substantia nigra reticulata. Faseb J 2003;17:1820-30. Urbano F, Leznik E, Llinas R. Cortical activation patterns evoked by afferent axons stimuli at different frequencies: an in vitro voltage-sensitive dye imaging study. Thalamus Rel Syst 2002;1:371-8. Cui JG, O’Connor WT, Ungerstedt U, Linderoth B, Meyerson BA. Spinal cord stimulation attenuates augmented dorsal horn release of excitatory amino acids in mononeuropathy via a gabaergic mechanism. Pain 1997;73:87-95. Wallin J, Fiska A, Tjolsen A, Linderoth B, Hole K. Spinal cord stimulation inhibits long-term potentiation of spinal wide dynamic range neurons. Brain Res 2003;973:39-43. Ellrich J, Lamp S. Peripheral nerve stimulation inhibits nociceptive processing: an electrophysiological study in healthy volunteers. Neuromodulation 2005; 8:225-32.
Stimulation technology in functional neurosurgery
91. Marsden CD, Obeso JA. The functions of the basal ganglia and the paradox of stereotaxic surgery in Parkinson’s disease. Brain 1994;117(Pt 4):877-97. 92. Shen KZ, Zhu ZT, Munhall A, Johnson SW. Synaptic plasticity in rat subthalamic nucleus induced by highfrequency stimulation. Synapse 2003;50:314-19. 93. Nowak LG, Bullier J. Axons, but not cell bodies, are activated by electrical stimulation in cortical gray matter. I. Evidence from chronaxie measurements. Exp Brain Res 1998;118:477-88. 94. Gorman AL. Differential patterns of activation of the pyramidal system elicited by surface anodal and cathodal cortical stimulation. J Neurophysiol 1966;29:547-64. 95. Hashimoto T, Elder CM, Okun MS, Patrick SK, Vitek JL. Stimulation of the subthalamic nucleus changes the firing pattern of pallidal neurons. J Neurosci 2003; 23:1916-23. 96. Anderson TR, Hu B, Iremonger K, Kiss ZH. Selective attenuation of afferent synaptic transmission as a mechanism of thalamic deep brain stimulation-induced tremor arrest. J Neurosci 2006;26:841-50. 97. Windels F, Bruet N, Poupard A, Feuerstein C, Bertrand A, Savasta M. Influence of the frequency parameter on extracellular glutamate and gamma-aminobutyric acid in substantia nigra and globus pallidus during electrical stimulation of subthalamic nucleus in rats. J Neurosci Res 2003;72:259-67. 98. Montgomery EB Jr, Baker KB. Mechanisms of deep brain stimulation and future technical developments. Neurol Res 2000;22:259-66. 99. Devos D, Labyt E, Derambure P, Bourriez JL, Cassim F, Reyns N, Blond S, Guieu JD, Destee A, Defebvre L. Subthalamic nucleus stimulation modulates motor cortex oscillatory activity in Parkinson’s disease. Brain 2004;127:408-19. 100. Ceballos-Baumann AO. Functional imaging in Parkinson’s disease: activation studies with PET, FMRI and SPECT. J Neurol 2003;250 Suppl 1:I15-23.
84
101. Jech R, Urgosik D, Tintera J, Nebuzelsky A, Krasensky J, Liscak R, Roth J, Ruzicka E. Functional magnetic resonance imaging during deep brain stimulation: a pilot study in four patients with Parkinson’s disease. Mov Disord 2001;16:1126-32. 102. Logothetis NK. The neural basis of the blood-oxygenlevel-dependent functional magnetic resonance imaging signal. Philos Trans R Soc Lond B Biol Sci 2002;357:1003-37. 103. Eidelberg D, Moeller JR, Kazumata K, Antonini A, Sterio D, Dhawan V, Spetsieris P, Alterman R, Kelly PJ, Dogali M, Fazzini E, Beric A. Metabolic correlates of pallidal neuronal activity in Parkinson’s disease. Brain 1997;120(Pt 8):1315-24. 104. Ceballos-Baumann AO, Boecker H, Fogel W, Alesch F, Bartenstein P, Conrad B, Diederich N, von Falkenhayn I, Moringlane JR, Schwaiger M, Tronnier VM. Thalamic stimulation for essential tremor activates motor and deactivates vestibular cortex. Neurology 2001;56:1347-54. 105. Holsheimer J. Which neuronal elements are activated directly by spinal cord stimulation. Neuromodulation 2002;5:25-31. 106. Schlaier JR, Eichhammer P, Langguth B, Doenitz C, Binder H, Hajak G, Brawanski A. Effects of spinal cord stimulation on cortical excitability in patients with chronic neuropathic pain: a pilot study. Eur J Pain 2007;11:863-8. 107. McIntyre CC, Grill WM. Excitation of central nervous system neurons by nonuniform electric fields. Biophys J 1999;76:878-88. 108. Grill WM, McIntyre CC. Extracellular excitation of central neurons:Implications for the mechanisms of deep brain stimulation.. Thalamus Rel Syst 2001; 1:269-77. 109. Aravanis AM, Wang LP, Zhang F, Meltzer LA, Mogri MZ, Schneider MB, Deisseroth K. An optical neural interface: in vivo control of rodent motor cortex with integrated fiberoptic and optogenetic technology. J Neural Eng 2007;4:S143-56.
1425
85 Therapeutic Lesions Through Chronically Implanted Deep Brain Stimulation Electrodes S. Raoul . D. Leduc . C. Deligny . Y. Lajat
Introduction Surgery for movement disorders began in the late 1930s with Meyers [1]: who perform transventricular procedures in the basal ganglia. Surgery improved with the introduction of stereotactic techniques [2]. Lesion targets were refined in the 1950s and 1960s with pallidotomy for treatment of rigidity and akinesia and thalamotomy for tremor [3]. Surgical treatment of Parkinson’s disease (PD) was less important in the 1970s because of the development of levodopa therapy With the long-term treatment complications of levodopa became evident leading to renewed interest in the surgical options for treatment of movement disorders. Leksell’s posteroventral pallidotomy was an effective treatment for rigidity and akinesia in PD [4]. From the end of 1980s, DBS has been proposed as an alternative treatment to lesion for tremor and Parkinson’s disease (PD) [5–7]. An electrode is stereotacticaly implanted in the targeted nucleus and connected through an internalized cable to a pulse generator. The procedure has the advantage of the reversibility of the effects and the adaptability of the electrical parameters to the symptomatology of the patients. It is however an expensive procedure due to the cost of the
Electronic supplementary material Supplementary material is available in the online version of this chapter at http:// dx.doi.org/10.1007/978-3-540-69960-6_85 and is accessible for authorized users. #
Springer-Verlag Berlin/Heidelberg 2009
pulse generator and requires the dedication of specialized personnel. In the other hand, monopolar radiofrequency (RF) electrodes have been used for many years to make lesions in specific nuclei in the brain to treat essential tremor, Parkinson disease, chronic pain and psychiatric disorders [8–12]. The technic consists to briefly apply a localized elevation of temperature through an implanted electrode. The size of the lesions depends on the electrical parameters (voltage, frequency, current) applied through the electrode [12,13]. Thalamotomies are generally efficient but can provided side-effects such as speech trouble or cognitive dysfunction [14]. Pallidotomies reduces dyskinesias, tremor and rigidity but bilateral pallidotomy was associated with high risks of complications such as cognitive impairment, dysarthrias [15,16]. Subthalamotomies have been performed uni and bilaterally with a relative safety [9,10,17–19] but hemiballism was not exceptional [20,21]. These sideeffects are related to the volume of the lesion [13]. It was then proposed to use DBS lead, as bipolar electrode, to perform progressive and small lesions. We indeed showed that electrodes used for DBS could be implanted, left in place and used to make staged lesion adapted to the symptom progression [22]. The aim of this chapter is to describe the technique, its indications and the precautions it requires. The first part is devoted to an experimental study where the relation between the radiofrequency (RF) condition and the size of
1428
85
Therapeutic lesions through chronically implanted deep brain stimulation electrodes
coagulum is analyzed in order to optimize the condition to make localized lesions in human brain through DBS electrode. The safety of the method is also assessed by evaluating diffusing current to the electrode adjacent to those used for applying RF and by analyzing the ultrastructure of the electrode after RF lesioning. We present in a second part a clinical study and the results obtained in 24 patients. Finally, we discuss the indications of the technique and give a practical guideline.
Experimental Studies in Egg White and Human Cadavers Material and Method Forty milliliters of fresh egg whites was prepared in a square glass container. The quadripolar Medtronic 3387 or 3389 lead was inserted into the egg whites in such a way that all four electrodes were completely immersed. The RF generator (Radionics) was connected to the lead with alligator clips to apply bipolar current between the two distal electrodes. A 250 kHz current was applied with various intensity with steps of 10 mA from 10 to 230 mA (corresponding to voltages with steps of 4V from 2 to 50 V). For each step, 10 lesions (i.e., a total of 250 lesions) were made. This first part of the study was made at constant time (60 s). The second part of the work studied various time duration (30 s, 60 s, 90 s, 120 s, 180 s, 5 min and 10 min) with constant voltage (40 V). For each time 10 lesions were performed. The same experimental procedure was used in six human cadavers. Cadavers were maintained at 37 C in a water bath. Electrodes were implanted in the thalamus and in the pallidum. Part of the skull was removed and transversal sections were made to obtain the pallidum, the caudate nucleus and the thalamus. Sixty (10 in pallidum and 50 in thalamus) lesions were
measured using currents of 10–170 mA with steps of 10 mA (corresponding to 10–80 V with steps of 4 V). The impedance, electrode combination, localization, size of lesions, current, voltage and time were recorded for each bipolar lesion. New DBS Medtronic (one 3389 and three 3387) were used for the experimental studies. Various time (10 s, 20 s, 30 s and 60 s) were applied in human brain tissue. The coagulum size (length and width) was measured using the scale on the miscroscope stage. When a lesion was obtained in the basal ganglia, the color of coagulum is quite different (yellow/white) from the normal color of thalamus or pallidum in cadavers (grey/purple). For some lesions histology was made to confirm more precisely size of lesions Ultrastructure of the DBS electrode after RF lesioning was analyzed with a scan electron microscopy at the end of experimental study in fresh egg whites
Computer Modeling of the Lead and the Coagulation Several parameters such as current density or temperature can not be measured in the neighborhood of the lead during the coagulation. The only way to get informations on these parameters is to simulate the process. This was done using the finite elements method. From experimental observations, we inferred the following mechanism of lesion: the potential difference between active contacts induces a current and the electrical power produced is dissipated in the brain which temperature increases until the coagulation. The first step is to calculate the electrical potential distribution V around the electrode. This is done by solving the Laplace’s equation [23,24]: ~ rV ~ ¼0 rs
ð1Þ
where s is the electrical conductivity of the ~ ¼ rV ~ and the brain. The electrical field E
Therapeutic lesions through chronically implanted deep brain stimulation electrodes
~ are deduced from current density ~ j ¼ sE the potential distribution. At any given point of the brain, a certain amount of electrical . is power 2 ~ dissipated, giving rise to a heat Q ¼ j s. In order to determine the corresponding temperature elevation, we have to solve the bioheat transfer equation [25]: r Cp
@T ~ ~ r k rT þ Ob ðT Ta Þ ¼ Q þ Qm @t ð2Þ
where t is the time, r the brain density, Cp its thermal capacity and k its thermal conductivity. The term Ob ðT Ta Þ corresponds the heat transfer due to blood-flow, it depends on the blood perfusion coefficient Ob and the arterial temperature Ta . In the case of the study with cadavers, this term is null. Qm is the metabolic heat source term. It is very negligible compared to Q. To find solutions of equations (1) and (2) in every point, we used the software Femlab based on finite elements method. Due to the axial symmetry of the problem, the workspace is reduced to two dimensions. The gray matter is assumed to be uniform. Its physical characteristics were found in the literature [26,27]: k ¼ 0:504 W :m1 :K 1 Cp ¼ 3700 J:K 1 :kg1 , and s ¼ 0:18 ½1 þ 0:02ðT 310Þ S=m. The lead is represented by its fingerprints. This surface is not involved in the calculation but rather the areas corresponding to titanium electrodes, since the lead is coated with insulator material. The boundary conditions are contact 0: potential 0 V, contact 1: potential Veff . The contacts 2 and 3 are isolated which means that the current can flow inside these contacts but it cannot pass through them, as in real experiments. The last step is to simulate the coagulation. Experimental studies show that when the brain coagulates, the current decreases abruptly. It means that the grey matter conductivity falls to zero after the coagulation. In order to reproduce this behavior, we assumed a coagulation temperature threshold Tc and introduced the
85
Rt phenomenological variable uðx; y; t; TÞ ¼ 0 jT Tc j þ ðT Tc Þdt which is null while T < Tc and increases monotically with time as soon as T > Tc . Then, we simulated the fall of the conductivity as:
sðx; y; t; TÞ ¼ 0:18 ½1 þ 0:02ðT 310Þ " #1 ð3Þ uðx; y; t; TÞ2 1þ w2 Tc and w are free parameters of the simulation. We choose [23,24] Tc ¼ 60 C and best results were obtained with w ¼ 10 K:s
Results in Egg White and Human Cadavers Effects of Radiofrequency Stimulation Applied to Fresh Egg Whites There was a clear threshold (15 V or 90 mA) to obtain coagulation in egg white. Above this threshold, the size of the lesion in creased with increasing voltage or current (with homogenous and constant impedance) until a plateau was reached (> Figure 85-1). At 40 V, 200 mA and applied during 60 s, the size of lesions ranged from length 5 0.6 mm and width 3 0.4 mm for 25 lesions . Figure 85-1 Diagram showing size of coagulum versus voltage in egg whites
1429
1430
85
Therapeutic lesions through chronically implanted deep brain stimulation electrodes
The coagulation process occurred rapidly (in a few seconds). The current increased during 1 s with a constant voltage and then fell abruptly (see Video).With a stimulation of 40 V, 200 mA, mean lesion size was 4.5 0.4 mm long and 3.0 0.2 mm wide after a stimulation duration of 30 s, 5.0 0.2 mm long and 3.0 0.1 mm wide after a duration of 40 s and 5.1 0.6 mm long and 3.3 0.4 mm wide after a duration of 60 s. When the duration of the stimulation is multiplied by two (from 30 to 60 s) the size of the lesion is only increased by about 10%. This means that the duration of the stimulation is not a critical parameter, the major part of the lesion is made during the first moments (> Figure 85-2). We observed that the coagulum size can not increase more than two adjacent electrodes. In this experimental set-up, we can observe that, after passing the threshold, the coagulation process was a quick phenomenon (less than 20 s). The current increased during 1 s with constant voltage and failed down abruptly.
1,000 O. The threshold to obtain a coagulum was in this case 22 V (about 20 mA) (> Figure 85-3). After that, sizes of lesions increase and reach a plateau level: sizes of lesions in thalamus were represented on > Table 85-1 for 60 s. There were no difference between thalamus and pallidum. It takes about 15 s to create a coagulum in the human brain and then the size of lesions does not increase significantly with increasing stimulation duration. Record sequences during coagulation in eggs and in cadavers showed that the coagulation process took no longer than 20 s. When coagulation was made, the current decreased abruptly from 40 mA to 0 mA in then 10 s as it can be seen on the Radionics Generator. It is the proof that a stable lesion was made. As in egg, size of the lesions can not increase more than two adjacent contacts in cadavers. We have performed 10 lesions in thalamus at 80 V, 80 mA during 10 min and length was 5.8 0.9 mm and width was 3.5 0.7 mm.
Effects of Radiofrequency Stimulation Applied to Human Cadavers
. Figure 85-3 Diagram showing size of coagulum versus voltage in human cadavers
The impedance of human brain tissue measured by the Radionics Generator was usually close to
. Figure 85-2 Size of coagulum versus increasing time at constant voltage (40 V) and constant impedance (400 V) in egg white
. Table 85-1 Sizes of lesions produced by 60 s stimulations Voltage
Length (mm)
Width (mm)
30 volts 40 volts 50 volts
3.3 0.8 4.0 0.8 5.4 0.3
2.3 0.5 3.0 0.6 3.3 0.4
Therapeutic lesions through chronically implanted deep brain stimulation electrodes
Results of Scan Electron Microscopy Study The DBS electrode was composed with titanate and fluor (Teflon isolated component) in spectroscopy microscope. After 200 lesions the electrode showed no clear alterations in its ultrastructure (> Figure 85-4).
Results of the Computer Modeling The temporal evolutions of current density, temperature and conductivity obtained with the numerical simulation of a lesion with 40 V amplitude are showed on > Figures 85-5 – > 85-7. At the beginning, the current is concentrated along the electrode, between the two active contacts (> Figure 85-5a). The temperature strongly increases in this area (> Figure 85-6a) and exceeds the coagulation threshold. A coagulum appears in this area which becomes isolated (> Figure 85-7a). The current must circumvent this new obstacle, its path increases so its amplitude decreases (> Figure 85-5b). But it remains high enough so that the temperature is above the threshold (> Figure 85-6b) and the coagulum
85
grows (> Figure 85-7b). This mechanism repeats, until the two contacts are embedded by the coagulum (> Figure 85-7c) and become isolated. Then, there is no way for the current to join the two active contacts. The current vanishes. The temperature at this moment remains high but the heat is dissipated and the brain recovers its normal temperature after few seconds. The final length and width of coagulum are respectively of the order of 4 mm and 3 mm (> Figure 85-7d). The key feature of this is that the process is self-controlled. As soon as the coagulum becomes wide enough, the coagulation stops. It implies firstly that there is no current flowing to the unused contacts. Secondly, the size of the lesions is limited to the area between the two active contacts.
Clinical Study Material and Method Patient Selection Between 2000 and 2007, 24 patients (5 females and 19 males) were referred to our center for neurosurgical treatment of tremor, dyskinesias
. Figure 85-4 Spectroscopy electronic microscopy photography. (a) contact used (b) contact not used
1431
1432
85
Therapeutic lesions through chronically implanted deep brain stimulation electrodes
. Figure 85-5 Evolution of the current density (A/m2). All the sizes are expressed in mm. (a) t = 0.5 s (b) t = 1.5 s
. Figure 85-6 Evolution of the temperature (K). All the sizes are expressed in mm. (a) t = 0.5 s (b) t = 1.5 s
and parkinsonism. After approvement of the local ethic committee, we performed lesions through DBS electrodes. Fifteen thalamotomies (10 left VIM and 5 right VIM) and three pallidotomies
were performed. Seven subthalamotomies were made in 6 patients (one patient had a bilateral subthalamotomy). Thalamotomies were performed in patients who were not good candidates
Therapeutic lesions through chronically implanted deep brain stimulation electrodes
85
. Figure 85-7 Evolution of the electrical conductivity (S/m). All the sizes are expressed in mm. (a) t = 0.5 s (b) t = 1.5 s (c) t = 3 s (d) t = 60 s
for DBS (age, cardiac pace-maker, general disease). Subthalamotomies were performed in second intention after infection of the hardware. The mean follow-up of this study is 38 months (range between 7 years to 6 months). The mean age of the population was 69 (1 years). The patients who had subthalamotomies were PD patients with medications fluctuations, good response to
Levodopa and no cognitive dysfunction. In the population of thalamotomies, 6 patients were PD patients with unilateral predominant tremor PD, 6 had ET and 3 others tremor (one with a posttraumatic tremor, the other with postsurgical tremor after meningioma and the last one with symptomatic tremor in relation to demyelinisation).
1433
1434
85
Therapeutic lesions through chronically implanted deep brain stimulation electrodes
Clinical Evaluation
. Figure 85-8 Theorical coordinates of the VIM calculated on X-ray
Evaluations were performed before surgery and 3 months, 6 months, 12 months and every year after surgery. Clinical evaluation was based on Unified Parkinson’s Disease Rating Scale (UPDRS) for PD patients and on Tremor Rating Scale for ET patients. Patients were assessed preoperatively and postoperatively off and on medication. Cognitive function was assessed by Mini Mental status and Mattis Dementia Rating scale pre and postoperatively. Preoperative MRI was all normal in the targeting areas. Post-op MRI was performed in 8 patients.
Surgical Procedure After fixing a Talairach frame to the head of the patient, a ventriculography was performed to delineate the midline of the third ventricle, and the anterior (AC) and posterior (PC) commisures. The theorical coordinates of the VIM, GPi and dorsolateral part of the STN were calculated (> Figure 85-8). The surgery was performed under local anesthesia. A 6 macroelectrode lead (PK08, Dixi Medical, Besanc¸on, France) was inserted with the lowest electrode below the calculated target along a double (anteriorly and laterally) oblique trajectory using a robot (ISS medical instrument, Lyon, France). Frontal and sagittal X-rays were used to insure the right position of the lead. The lead was then connected to an external stimulator (Screener 3625, Medtronic, Minneapolis, USA). The clinical effect of a monopolar stimulation was tested through each electrode with 0.5 V incremental voltage from 0 to 10 V (pulse width = 80 ms, frequency = 150 Hz) (Cf. > Figure 85-9). This allowed to identify the therapeutic site, i.e., sites located one or two adjacent electrodes
through which less than 3 V stimulation leads to a clear improvement of symptoms (suppression of tremor for VIM target, suppression of rigidity for GPi or STN targets) without producing significant side-effects. Then, a four-electrode lead (3387 or 3389, Medtronic, Minneapolis, USA) was implanted. The efficiency of stimulation was assessed again in bipolar stimulation configuration (Cf. > Figure 85-10). Then the Medtronics lead was connected with alligator clips to a radiofrequency generator. While the patient was asked to maintain his two arms stretched and to count in a loud voice, a heat coagulation was applied at 40 V for 60 s through the most effective electrode (Cf. > Figure 85-11a,b,c). The coagulation was prematurely stopped if weakness of contralateral stretched arm appeared or if voice deadens. If symptoms are still present after the first coagulation, complementary lesions were performed through the same electrode or to the adjacent electrode. The lead was leaving in place for thalamotomies, pallidotomies and removed for 6 subthalamotomies. The extra-cranial part of the lead was buried under the scalp.
Therapeutic lesions through chronically implanted deep brain stimulation electrodes
85
. Figure 85-9 Stereotactic implantation of the DBS electrode
. Figure 85-10 Clinical testing of the patient
Results Only bipolar lesions were performed with the 250 KHz radiofrequency generator. Electrical parameters settings were as follow: mean voltage = 40 V,
mean current = 38 mA with an impedance of 1,000 O during 60 s. Parameters were adjusted according to the clinical effects and to a previous experimental study. All parameters were summarized in > Table 85-2.
1435
1436
85
Therapeutic lesions through chronically implanted deep brain stimulation electrodes
. Figure 85-11 Practical lesioning. (a) Alligator clips to be connected to the DBS lead and the radiofrequency generator (b) t = DBS lead and alligator clips (c) radiofrequency generator
All patients were improved immediately after the surgical procedure. The Tremor Rating Scale was at 4 preoperatively and at 0 postoperatively. The benefit was stable with time with a follow-up of 3 years. Two of our patients need to make another lesion 3 years and 4 years after the first surgery. The surgical procedure was very simple: under local anesthesia the connector of the lead was reconnected to the external generator and each electrode was tested to assess the most effective electrode. A new lesion was made. For one pallidotomy three staged lesions were performed at 3 months and 6 months
after the initial surgery. After that, the patient had a total control of tremor and dyskinesias. The rigidity was less controlled. For the 2 other patients only one lesion was performed. Seven subthalamotomies were performed in 6 patients, one had bilateral subthalamotoy. The electrode chose for subthalamotomies was the most effective electrode chronically with no-side effects. Before lesioning voltage was increased to 10 V to verify that no capsular effects occur. The bilateral subthalamotomy was performed with a delay of 1 month in between the 2 lesions. Subthalamotomies produced a significant (p < 0.002) reduction (80%) in the off UPDRS part III score
Therapeutic lesions through chronically implanted deep brain stimulation electrodes
85
. Table 85-2 Parameter settings in patients
Patient 1 Patient 2 Patient 3 Patient 4 Patient 5 Patient 6 Patient 7 Patient 8 Patient 9 Patient 10 Patient 11 Patient 12 Patient 13 Patient 14 Patient 15 Patient 16 Patient 17 Patient 18 Patient 19 Patient 20 Patient 21 Patient 22 Patient 23 Patient 24
Voltage (V)
Current (mA)
Time (s)
50 30 40 40 38 40 38 38 30 36 40 40 40 42 50 40 40 30 40 40 40 45 40 40
42 38 25 38 40 32 50 50 50 52 40 52 40 25 30 38 38 60 38 35 42 40 40 38
60 60 30 60 60 60 60 60 60 60 60 60 60 60 60 60 60 30 60 60 60 60 60 60
at the last assessment compared with the preoperative baseline score (52 preop vs. 10 postop). The motor score in the on state was also reduced (79%). The mean daily dose of Levodopa decreased by 41% compared with baseline. One patient was free of Levodopa during 1 year. There was no general complication and no patient died. No postoperatively dyskinesias or hemiballism was observed after subthalamotomies. Two transient confusion were observed after thalamotomies in elder patients (82 years old and 79 years old). One ataxia with tone muscle reduction and difficulties to walk was observed in one patient but this patient had preoperatively a peripherical neuropathy. MRI performed 2 days after surgery show a small and staged lesion in the VIM. The lesion
Electrode 1–2 0–1 0–1 0–1 1–2 0–1 0–1/1–2 0–1 0–1 0–1/1–2 0–1 0–1 0–1 0–1 1–2 0–1 0–1 0–1 0–1 1–2/2–3 0–1 2–3 1–2 2–3
disappears at 3 months on the MRI but the clinical effect was persistant with total abolition of tremor (Cf. > Figure 85-12a,b,c). The other MRI shows a unilateral subthalamotomy and a DBS on the other side. The lesion was surrounding by an edema which disappear between 2 and 5 days (Cf. > Figure 85-13).
Discussion/Conclusion Summary of Results The study was performed to assess the range of parameters that could be used in a surgical procedure to create RF lesions in the brain through DBS leads. The most important result of the
1437
1438
85
Therapeutic lesions through chronically implanted deep brain stimulation electrodes
. Figure 85-12 Coronal T1 MRI. (a) after implantation of the lead (b) 48 h after lesioning (c) 3 months after lesioning
. Figure 85-13 Coronal T2 MRI showing a subthalamotomy
Experiments and computer modeling demonstrate that using a DBS lead to create RF lesions is safe. Clinical studies have shown that it is possible to use this procedure in patients. [22,28]
Discussion of Experimental Results
experimental study is that the process of coagulation was very simple with a threshold and a plateau level, both in egg whites and in human cadaver brains. Computer modeling was performed to assess various parameters than we were unable to measure experimentally, such as current density and temperature. The results of modeling are in agreement with the experimental procedure.
Two models (egg whites and cadavers) were used to estimate lesion size and calculate parameter settings. Egg white or albumin solutions are classical models to study RF lesions [12,29–31]. We found some differences between the two models. The impedance in fresh egg white was lower (300–400 O) than in human cadaver basal ganglia (1,000–1,200 O). This means that, to produce a given lesion size, less current was needed in human brain than in egg white. When using a DBS electrode to create lesions, the impedance of the brain must be checked before and after lesioning. Several important parameters, such as current density or temperature, cannot be directly measured when the lead is embedded in the patient’s brain. To overcome this limitation, we performed numerical simulations because with DBS electrode we can not have directly the temperature applied to the lead. In patients, lesions
Therapeutic lesions through chronically implanted deep brain stimulation electrodes
should be smaller because of blood perfusion. [23,24]The simulations establish the upper limit of the lesion size. Experimentally, the current increases at the beginning of the coagulation and then falls abruptly. Conductivity behaves in a similar manner. It is known [32] that the collagen becomes glucose at around 60 C, the liquids become vapor at around 100 C and tissue charring is initiated at around 200 C, yet the microscopic mechanisms underlying the conductivity transformations are not well established. That is why we used the phenomenological description in which the conductivity decreases as soon as the temperature has exceeded a threshold value. This behavior has not been considered before because previous studies [23,24,32] focused on electrodes specifically designed for RF ablation which are regulated to prevent the temperature exceeding a given threshold. The qualitative agreement between experimental results and simulations confirms the validity of our assumptions and insures that the coagulation is a self-limiting mechanism. Human cadavers serve as a good model for RF lesioning procedures. Nevertheless, there are likely to be modifications in the lesion size in patients because of blood and fluid circulation. Several authors [13,30] sought to correlate the size of lesion in egg and on MRI images after lesioning the brain of pigs or the human brain. Ericksson found a close correlation between the coagulation size and the inner zone seen on MRI [30]. Similar results could be expected in humans. Indeed, the size of coagulum (3–5 mm) did not differ greatly from that of the lesions produced in patients [33] using our new surgical procedure. We have proved that lesions cannot extend beyond two electrodes and that we can create staged, small and controlled lesions in human brains. Another experimental finding was that the DBS electrode was not modified after lesioning. This is an argument in favor of the safety of this surgical procedure in patients. If the electrode is used to produce a lesion and then left in place, it
85
could subsequently be used for DBS. Indeed, this procedure offers various therapeutic alternatives. For example, if tolerance to DBS was observed, lesioning could then be performed under local anesthesia.
Discussion of Clinical Results This study showed that pallidotomies, thalamotomies and subthalamotomies could be performed safely through DBS electrodes. The ration of complications occurs in our series is less than in literature (5% for thalamotomies and 0% for subthalamotomies). Alvarez’s study shows 7/18 patients who had speech deterioration after the surgery and 3/18 who had hemiballism [9]. The present result and data from literature [3,4,18,19, 34–36] indicate that ablative surgery for PD and ET patients may be a useful technique. Efficacy and risk of permanent side-effect appeared to be similar to what is observed with DBS with our procedure. The cost of such technique is far from the cost of DBS since no stimulators nor medical time for electrical parameters fine-tuning are required. However, DBS of the STN is most widely surgical approach for PD [5,6,33,37–41]. Subthalamotomies will be considered in a second hand procedure when infection occurs to the hardware. It should not be considered as the first choice for treating PD. But it is possible to proposed one DBS on one side and lesion on the other side. Subthalamotomies will be restricted to patients treated with cardiac pace-maker, who are not willing to accept the limitations associated with the chronic use of an implanted device according to Alvarez [8,9]. For thalamotomies the discussion is open for patients who need an unilateral lesion. Recent studies have showed that risks are equivalent between the two procedures. Benefit could be better with thalamotomy [42] because same patients develop a tolerance to the stimulation.
1439
1440
85
Therapeutic lesions through chronically implanted deep brain stimulation electrodes
Our surgical procedure could be a good comprise between the both, because if the lead was leaving in place, a new lesion could be simply performed. And the lesions made through DBS electrodes are smaller than classical lesion. It seems not to destroy the brain because in one patient after a subthalamotomy a stimulator was placed with good control of PD 3 years after the lesion (Limousin and Benabib, unpublished result, personal communication).
Some precautions must be respected when using this procedure on patients: check the impedance before lesioning to adjust setting parameters of coagulation, use an RF generator that gives 250 KHz or more (never less, as the electrode could be damaged and disabling pain could occur), use bipolar rather than monopolar lesions, use voltage or current but always check that the relation between these two parameters is consistent with the impedance measured before lesioning. [43–45]
Practical Guidelines to Create Lesions in Human Brain References Lesioning the brain could be done with DBS electrodes. This new surgical procedure was used in patients who could not have deep brain stimulation because of age or clinical presentation and was used if an infection on the hardware occurs. It could an alternative treatment in patients but must be discussed according to experience of the team, clinical presentation of the patients and cost of the hardware. To make staged lesions through DBS electrode, the same surgical procedure may be applied with more cautions about side-effects because of the no reversibility of the procedure. In patients, we recommend that bipolar lesions are performed with a 250-kHz radiofrequency generator at the steady state of the plateau (40 V for 1,000 O impedance in the brain tissue), since experiments in fresh egg whites have shown that, around the threshold, the coagulum could be unstable. This could be the same in human brain if the heat is dissipated by the vessels. Time was not a good parameter in this experiment to increase size of lesions. We recommend using a duration of 30 or 60 s in humans. In our experience in patients (Raoul et al.) we preferred to use one short lesions rather than one long lesion. This was based on the fact that in experimental studies lesions were produced in 10 or 20 s. After this time, no current was observed through the electrodes.
1. Meyers R. Surgical procedure for postencephalitis tremor, with notes on the physiology of premotor fibers. Arch Neurol Psychiatry 1932;4:33–38. 2. Spiegel EA, Wycis HT, Marks M, Lee AS. Stereotactic apparatus for operations on human brain. Science 1947;106:349-50. 3. Svennilson E, Torvik A, Lowe R, Leksell L. Treatment of parkinsonism by stereotactic thermolesions in the pallidal region. A clinical evaluation of 81 cases. Acta Psychiatr Scand 1960;35:358-77. 4. Laitinen LV, Bergenheim AT, Hariz M. Leskell’s posteroventral pallidotomy in the treatment of Parkinson’s disease. J Neurosurg 1992;76:53-61. 5. Benabid AL, Pollak P, Gross C, Hoffmann D, Benazzouz A, Gao DM, et al. Acute and long-term effects of subthalamic nucleus stimulation in Parkinson’s disease. Stereotact Funct Neurosurg 1994;62(1–4):76-84. 6. Fraix V, Pollak P, Chabardes S, Ardouin C, Koudsie A, Benazzouz A, et al. Deep brain stimulation. Rev Neurol (Paris) 2004;160(5 Pt 1):511-21 (Review French). 7. Okun MS, Vitek JL. Lesion therapy for Parkinson’s disease and other movement disorders: update and controversies. Mov Disord 2004;19:375-89. 8. Alvarez L, Macias R, Guridi J, et al. Dorsal subthalamotomy for Parkinson’s disease. Mov Disord 2001;16:72-8. 9. Alvarez L, Macias R, Lopez G, Alvarez E, Pavon N, Rodriguez-Oroz MC, et al. Bilateral subthalamotomy in Parkinson’s disease: initial and long-term response. Brain 2005;128(Pt 3):570-83. 10. Gill SS, Heywood P. Bilateral dorsolateral subthalamotomy for advanced Parkinson’s disease. Lancet 1997;350 (9086):1224. 11. Laitinen LV. Psychosurgery. Stereotact Funct Neurosurg 2001;76(3–4):239-42. 12. Van den Berg J, Van Manen J. Graded coagulation of brain tissue. Acta Physiol Pharmacol Neerlandica 1962;10:353-77.
Therapeutic lesions through chronically implanted deep brain stimulation electrodes
13. Hariz MI, Hirabayashi H. Is there a relationship between size and site of the stereotactic lesion and symptomatic results of pallidotomy and thalamotomy? Stereotact Funct Neurosurgery 1997;69:28-45. 14. Speelman JD, Schuurman R, de Bie RM, Esselink RA, Bosch DA. Stereotactic neurosurgery for tremor. Mov Disord 2002;17 Suppl 3:S84-8. 15. Favre J, Burchiel KJ, Taha JM, Hammerstad J. Outcome of unilateral and bilateral pallidotomy for Parkinson’s disease: patient assessment. Neurosurgery 2000;46 (2):344-53; discussion 353–5. 16. Scott RB, Harrison J, Boulton C, Wilson J, Gregory R, Parkin S, et al. Global attentional-executive sequelae following surgical lesions to globus pallidus interna. Brain 2002;125(Pt 3):562-74. 17. Patel NK, Heywood P, O’Sullivan K, McCarter R, Love S, Gill SS. Unilateral subthalamotomy in the treatment of Parkinson’s disease. Brain 2003;126:1136-45. 18. Su PC, et al. Postural asymmetries following unilateral subthalamotomy for advanced Parkinson’s disease. Mov Disord 2002;17(1):191-4. 19. Su PC, Tseng HM, Liu HM, Yen RF, Liou HH. Treatment of advances Parkinson’s disease by subthalamotomy: one year results. Mov Disord 2003;18:531-8. 20. Chen CC, et al. Hemiballism after subthalamotomy in patients with Parkinson disease: report of 2 cases. Mov Disord 2002;17(6):1367-71. 21. Tseng HM, Su PC, Liu HM. Persistent hemiballism after subthalamotomy: the size of the lesion matters more than the location. Mov Disord. 2003;18(10):1209-11. 22. Raoul S, Faighel M, Rivier I, Verin M, Lajat Y, Damier P. Staged lesions through implanted deep brain stimulating electrodes: a new surgical procedure for treating tremor or dyskinesias. Mov Disord 2003;18:933-8. 23. Chang I. Finite element analysis of hepatic radiofrequency ablation probes using temperature-dependent electrical conductivity. BioMed Eng OnLine 2003;2:12. 24. Chang IA, Nguyen UD. Thermal modeling of lesion growth with radiofrequency ablation devices. BioMed Eng OnLine 2004;3:27. 25. Pennes HH. Analysis of tissue and arterial blood temperatures in the resting human forearm. J Applied Physiol 1948;1(2):93-122. 26. Foster KR, Schwan HP. Dielectric properties of tissues and biological materials: a critical review. Crit Rev Biomed Eng 1989;17(1):25-104. 27. Zhu L, Diao C. Theoretical simulation of temperature distribution in the brain during mild hypothermia treatment for brain injury. Med Biol Eng Comput 2001;39:681-7. 28. Oh MY, Hodaie M, Kim SH, Alkhani A, Lang AE, Lozano AM. Deep brain stimulator electrodes used for lesioning: proof of principle. Neurosurgery 2001;49:363-9. 29. Ericksson O, Wardell K, Bylund NE, Kullberg G, Rehncrona S. In vitro evaluation of brain lesioning
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
85
electrodes (Leskell) using a computer-assisted video system. Neurol Res 1999;21:89-95. Ericksson O, Backlund EO, Lundberg P, Lindstam H, Lindstrom S, Wardell K. Experimental radiofrequency brain lesions: a volumetric study. Neurosurgery 2002;51:781-8. Moriglane JR, Koch R, Schafer H, Ostertag CB. Experimental radiofrequency (RF) coagulation with computer-based on line monitoring of temperature and power. Acta Neurochir (Wien) 1989;96:26-131. Ekstrand V, Wiksell H, Schultz I, Sandstedt B, Rotstein S, Eriksson A. Influence of electrical and thermal properties on RF ablation of breast cancer: is the tumour preferentially heated? Biomed Eng Online 2005;4:41. Rodriguez-Oroz MC, Obeso JA, Lang AE, Houeto JL, Pollak P, Rehncrona S, et al. Bilateral deep brain stimulation in Parkinson’s disease: a multicentre study with 4 years follow-up. Brain 2005;128(Pt 10):2240-9. Barlas O, Hanagasi HA, Imer M, Sahin HA, Sencer S, Emre M. Do unilateral ablative lesions of the sub thalamic nucleus in Parkinson patients lead to hemiballism? Mov Disord 2001;16:306-10. Filho OV. Unilateral subthalamic nucleus lesioning: a safe and effective treatment for Parkinson’s disease. Arq Neuropsiquiatr 2002;60(4):935–48 Patel NK, Plaha P, O’Sullivan K, McCarter R, Heywood P, Gill SS. MRI directed bilateral stimulation of the subthalamic nucleus in patients with Parkinson’s disease. J Neurol Neurosurg Psychiatry 2003;74 (12):1631-7. Jaggi JL, Umemura A, Hurtig HI, Siderowf AD, Colcher A, Stern MB, et al. Bilateral stimulation of the subthalamic nucleus in Parkinson’s disease: surgical efficacy and prediction of outcome. Stereotact Funct Neurosurg 2004;82(2–3):104-14. Kleiner-Fisman G, Fisman DN, Sime E, Saint-Cyr JA, Lozano AM, Lang AE. Long-term follow up of bilateral deep brain stimulation of the subthalamic nucleus in patients with advanced Parkinson disease. J Neurosurg 2003;99(3):489-95. Krack P, Batir A, Van Blercom N, Chabardes S, Fraix V, Ardouin C, et al. Five-year follow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinson’s disease. N Engl J Med 2003;349(20):1925-34. Limousin P, Pollak P, Benazzouz A, Hoffmann D, Broussolle E, Perret JE, et al. Bilateral subthalamic nucleus stimulation for severe Parkinson’s disease. Mov Disord 1995;10(5):672-4. Moro E, Scerrati M, Romito LM, Roselli R, Tonali P, Albanese A. Chronic subthalamic nucleus stimulation reduces medication requirements in Parkinson’s disease. Neurology 1999;53(1):85-90. Pahwa R, Lyons KE, Wilkinson SB, Troster AI, Overman J, Kieltyka J, et al. Comparison of thalamotomy to deep brain stimulation of the thalamus in essential tremor. Mov Disord 2001;16(1):140-3.
1441
1442
85
Therapeutic lesions through chronically implanted deep brain stimulation electrodes
43. Bittar RG, Hyam J, Nandi D, Wang S, Liu X, Joint C, et al. Thalamotomy versus thalamic stimulation for multiple sclerosis tremor. J Clin Neurosci 2005;12(6):638-42. 44. Fox JL. Experimental relationship of radiofrequency electrical current and lesion size for application to percutaneous cordotomy. J Neurosurg 1970;33:415-21.
45. Yen CP, Kung SS, Su YF, Lin WC, Howng SL, Kwan AL. Stereotactic bilateral anterior cingulotomy for intractable pain. J Clin Neurosci 2005;12(8):886-90.
111 Central Procedures for Cervical Dystonia J. Q. Oropilla . Z. H. T. Kiss
Brain surgery has become an excellent option for the long-term control of cervical dystonia. This chapter will briefly review the clinical features important to the surgical management of cervical dystonia, the indications for surgery, the types of central procedures that have been utilized historically and at present, as well as the rationale for each. Finally we will address the complications and long-term outcomes of each procedure.
sleep, relaxation, and various sensory maneuvers. Similar to other forms of dystonia, the abnormal muscle contractions that produce head deviation can be temporarily controlled by a variety of sensory tricks, such as touching the chin, face, or back of the head [2]. Neck pain is a common presenting feature, occurring in more than 70% of patients [3,4]. In addition, associated postural hand tremor is common, present in about 30%. Swallowing functions may be abnormal, especially in patients with extreme retrocollis.
Clinical Features of Cervical Dystonia Epidemiology Focal dystonia affects a single body part, such as the neck. Cervical dystonia is the most common of the focal, idiopathic, adult-onset dystonias encountered in a movement disorder clinic [1,2]. The older term used for this condition is spasmodic torticollis. In cervical dystonia, painful sustained or intermittent tonic contractions of the sternocleidomastoid, trapezius, and deeper neck muscles occur, often unilaterally, and cause abnormal head positions. Sternocleidomastoid muscle contraction causes rotation of the head and lifting of the chin to the contralateral side and lateral bending (tilt) to the ipsilateral side. Rotation may involve any plane but almost always has a horizontal component. Aside from rotational tilting (torticollis), the head can tilt laterally (laterocollis), forward (anterocollis), or backward (retrocollis). The involved neck muscles are often hypertrophied. Cervical dystonia is exacerbated during periods of stress or fatigue and is usually relieved by #
Springer-Verlag Berlin/Heidelberg 2009
Several studies have reported the prevalence of cervical dystonia as ranging from 5.9 to 13 per 100,000 and the incidence from 0.8 to 1.2 per 100,000 person-years, albeit in different population groups [5–9]. It appears to be more common in Caucasians [7,9] and prevalence increases with age. In the study by Le et al. [7], the prevalence for age 0–29 years was 3.1 per 100,000, for age 30–49 was 12.8 per 100,000, and for those over age 50 years, 32.1 per 100,000. Cervical dystonia is 1.9–3.6 times more common in females [5,7,9,10].
Natural History Cervical dystonia tends to worsen for an average of 3–5 years before stabilizing, however progression has occurred as rapidly as within 1 month and over as long as 18 years [11–13]. In about 20–30% of cases, cervical dystonia may progress
1872
111
Central procedures for cervical dystonia
to, or be the initial manifestation of, a more generalized dystonia [2,13]. If becoming generalized, the dystonia usually spreads to the face, jaw, arms or trunk. In the majority of patients, it is a lifelong disorder, but spontaneous remission may occur in 10–20% [13,14]. If this occurs, it typically does so within the first 5 years, but remission may be incomplete or not sustained [12–15]. Some degree of disability is present in the majority of patients, ranging from mild (‘‘subjective feeling of discomfort in social conditions without objective consequences on social life’’) to severe (‘‘qualitative and quantitative modification of the occupational level with resulting impairment of social life’’) [16,17]. Unlike other types of focal dystonia, there is a high incidence of pain in cervical dystonia. It is present in 60–75% of patients at some point and is a major source of disability. The pain is associated with constant head-turning, greater severity of head-turning and the presence of spasms [3,4]. Many patients will have associated depression [18]. Disability is also caused by task-specific limitations (e.g., inability to drive) and avoidance of social interaction resulting from abnormal posture. Patients with cervical dystonia have an increased risk of developing premature degenerative changes of the upper cervical spine. This occurs more commonly on the side towards which the head is turned or tilted [19]. This can also contribute to pain, limitation of head movement, poor response to medical management, and even myelopathy in severe cases [20].
Other Cervical Dystonias Secondary Dystonia A history of preceding head or neck trauma is common in 9–16% of patients with cervical dystonia, reporting this as occurring weeks to months
prior to the onset of symptoms. Acute onset posttraumatic cervical dystonia is a distinct syndrome following head, neck, and shoulder trauma. Its features include a limitation in cervical range of motion, fixed posture, lack of effective sensory tricks, persistence during and lack of improvement after sleep. The response to botulinum toxin injections is poor [21]. Symptoms emerge within days to months after trauma. Diagnostic criteria for posttraumatic dystonia have been proposed [22]: (1) trauma severe enough to cause local symptoms for at least 2 weeks or requires medical evaluation within 2 weeks of the trauma, (2) the initial manifestation of the movement disorder is anatomically related to the site of injury, and (3) the onset of the movement disorder is within days or months (up to 1 year) after the injury. Another subgroup of post-traumatic patients have onset and maximum disability very quickly after injury, with severe pain, a fixed abnormal posture, and resistance to treatment. The term ‘‘posttraumatic painful torticollis’’ was proposed to differentiate this group from classical cervical dystonia patients [23].
Cranial Cervical Segmental Dystonia Blepharospasm is a syndrome of involuntary eye closure produced by spasmodic contractions of the facial and oromandibular muscles. It is a lifelong disorder with a remission rate as high as 11.3% within the first 5 years in one study [24], but this may not be sustained [25]. Generally, patients have progressive worsening of their symptoms during the first 5 years after onset, following which the symptoms stabilize. Up to 15% become functionally blind. In a large proportion of patients with blepharospasm, other facial oromandibular, pharyngeal, laryngeal, and cervical muscles become involved [26,27], and the focal dystonia gradually evolves into a segmental (cranial-cervical) dystonia. While Meige syndrome is an isolated oral facial dystonia, it has been reported to respond to
Central procedures for cervical dystonia
surgery, and will only be discussed in the context of its association with cervical dystonia. The clinical features of cervical dystonia are very relevant to surgeons operating on these patients. It is critical to allow adequate time for medical management because of the spontaneous remissions that can occur. Unusual signs or complaints, failure to respond to adequate doses of botulinum toxin and post-traumatic forms [28] require careful assessment by an experienced movement disorder neurologist on the team, prior to consideration of surgery.
Surgical Indications and Pre-Operative Screening The indications for cervical dystonia are the same as that for any movement disorder surgery and include (1) failure of maximal medical management, (2) no medical contraindications to surgery, such as bleeding diathesis or uncontrolled hypertension, (3) the disability should significantly impact quality of life, (4) the patient should have no cognitive or psychiatric impairment, and (5) should be able to fully cooperate with the procedure and long term follow-up. There are several specific points important for treating cervical dystonia in particular. First, a minimum time period of neurologic observation is required to confirm that remission will not occur spontaneously. Second, botulinum toxin is generally very effective early in the course of the condition; however, only 63% of patients experience long term sustained benefit [29]. Third, severe cervical dystonia can result in cervical spondylosis and myelopathy if left untreated [30– 32].Therefore earlier surgery can be contemplated before severe permanent complications arise. While the other selection criteria for surgery may seem obvious, cooperation with surgery and follow-up are sometimes hard to gauge. Patients selected should be those who will obtain the greatest benefit from treatment, and who will
111
maintain this benefit sufficiently long to justify the time and expense invested, and are physically, cognitively, and emotionally able to tolerate all aspects of surgery and post-operative care [33]. Screening of patients for cervical dystonia surgery should be performed in a multidisciplinary setting using one of the standard motor impairment scales, such as the Tsui Rating Scale [34], the Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS) [17,35], and/or the Burke-Fahn-Marsden dystonia rating scale [36], when a component of a more generalized dystonia is present. Sessions should be videotaped and stored as part of the medical record. This provides not only for quality assurance for the surgeon and the program, but also a good reminder for the patient about their pre-op status years after surgery. Neuropsychological screening prior to ablative or deep brain stimulation surgery has been advocated by several groups and is routinely performed in our center. As discussed in previous chapters, cognitive impairment, defects in emotional processing, and behavioral changes have all been reported after movement disorder surgery for Parkinson’s disease (PD). While neuropsychological evaluation is critical for the PD patient group, it may not be necessary for the younger dystonia populations. There have been no reports of significant cognitive impairments after dystonia surgery; only minor changes have been observed [37]. In fact, improvements have even been noted [38]. Thus, there is no consensus on the requirement for neuropsychological assessment; however, the opportunity to observe a patient by multiple health care personnel may allow other issues to surface. A depression questionnaire performed by the neuropsychometrist allows patients who are suffering from significant depression to be referred for appropriate psychiatric management. Depression has occurred after basal ganglia surgery [39–42] and has been reported in dystonia patients as well [43]. Because depression may mask the
1873
1874
111
Central procedures for cervical dystonia
benefits of the procedure, it must be adequately treated. Other screening procedures include speech pathology and swallowing assessments. While we have not seen significant changes in swallowing function after DBS surgery, it is especially relevant for the cervical dystonia group. These patients may have head positions and muscle weakness related to botulinum toxin injections that make swallowing difficult. Those patients who complain of symptoms undergo formal preoperative swallowing assessments. After screening, the next issue to address is which operation and which target is best for each patient. The following sections will discuss the two generally utilized targets, the thalamus and globus pallidus, the rationale for each and the outcomes of both lesioning and DBS surgery.
Surgical Procedures
Vop, and Forel’s fields [46], and the (mesencephalic) nucleus of Cajal [47]. In 1971, Meares [11] reported a 50% ‘‘moderate’’ improvement in four of eight patients who underwent thalamotomy (target unspecified). Cooper’s results for 160 patients with spasmodic torticollis operated on between 1957 and 1977 were 60% ‘‘satisfactory alleviation’’ of symptoms. However 20% developed dysphonia [48]. In 1980, von Essen reported on 17 cervical dystonia patients with unilateral Voi thalamotomy followed for up to 5 years: 65% had ‘‘good’’ outcomes [49]. Of 27 cases with segmental or focal dystonia seen by Andrew et al. [50], 22 had spasmodic torticollis; of the 16 who underwent bilateral thalamotomies (Vim, Vce/Vci, Cm) 62% were much improved using the author’s rating scale, albeit with a high incidence of residual hemiparesis and/or dysarthria. Most of these thalamic targets have been abandoned. Only the cerebellar and pallidal receiving nuclei, the Voa [51], Vop and Vim [52] remain in use in recent series.
Thalamotomy Functional neurosurgery for dystonia was first used primarily for generalized dystonia, and to a lesser extent for focal or segmental dystonias. These early studies were performed before the modern classification of dystonia was established, and prior to any rating scales; therefore outcomes are difficult to interpret and may include cases with mixed classification, and include secondary and ‘‘psychogenic’’ forms. In addition, there was no ‘‘standard’’ thalamotomy: a specific target within the thalamus was not described and the procedure performed merely named a ‘‘thalamotomy.’’ We will use Hassler’s nomenclature for the thalamic nuclei in the following discussion [44].
Thalamic Targets and Outcomes Reports date back to the 1950s and 1960s when the targets for torticollis included the equivalents to the Vo, Vc, and Ce thalamic nuclei [45], the Voi,
Unilateral versus Bilateral and Relevant Complications In general, unilateral thalamotomy produces only minor improvements in axial and cervical dystonia. In the study of Andrew et al. [50], only two of six patients with unilateral thalamotomy had ‘‘good’’ or ‘‘very good’’ benefit beyond 1 year, compared with 81% (13 of 16) of patients with bilateral thalamotomy. Of these, 63% (10 of 16) with bilateral thalamotomy had ‘‘very good’’ or ‘‘excellent’’ results. While bilateral thalamotomy is more effective, it is associated with a 10–40% risk of serious complications, especially bulbar weakness, resulting in hypophonia or dysphagia, dysarthria, cognitive impairment, and ataxia [53]. In fact, Andrew had a 56% complication rate for bilateral thalamotomy compared to 11% for unilateral thalamotomy. Tasker obtained similar complication rates when performing thalamotomies for generalized dystonia [52].
Central procedures for cervical dystonia
Time Course and Duration of Effect There is little information reported about the time course and duration of effect. Generally, immediate results of thalamotomy in cervical dystonia are unimpressive, but progressive benefit occurs over weeks to months [49,53]. Andrew et al. [50] reported stable results 1 year or more after surgery.
Predictors of Outcome Results of thalamotomy for cervical dystonia vary depending on the presentation. Von Essen et al. [49] noted that those with hypertrophy of the sternocleidomastoid muscle or cervical scoliosis showed less improvement.
Outcomes of Secondary Cervical Dystonia There are no reports of surgical outcomes specific to secondary cervical dystonia. Most studies were performed on patients with secondary generalized dystonia or mixed groups of secondary dystonias. Cooper [48] felt that patients with primary dystonia were more likely to improve markedly after thalamotomy than those with secondary forms. Other studies found the opposite; that patients with secondary dystonia improved more than those with primary. Tasker et al. [52], for example, reported that 68% of patients with secondary dystonia showed a more than 25% improvement, whereas only 46% of patients with primary dystonia improved similarly. In the series by Cardoso et al. [51], 50% of the patients with secondary dystonia were moderately or markedly improved at their most recent follow-up visits, whereas only 43% of patients with primary dystonia obtained similar improvement. Despite further experience now with thalamic and pallidal targets, there still is no consensus on this issue [54].
111
Outcomes for Cranial-Cervical Dystonia There is only one case report on thalamotomy for cranial cervical dystonia. Muta et al. [55] described a case of a woman with Meige syndrome who underwent right Vo complex (Voa + Vop) thalamotomy, followed 1 year later by a left Vo complex thalamotomy. The staged bilateral thalamotomy was not beneficial and she later underwent bilateral pallidal stimulation with good effect.
Thalamic Stimulation Mundinger was the first to implant electrodes into thalamus for cervical dystonia. He targeted unilaterally both the Voa motor thalamus and subthalamic region (zona incerta and the Forel H1 and H2 fields) in seven patients [56]. Only shortterm outcome was reported (up to 8 months), but it seemed favorable. Stimulation parameters applied were intermittent (30 min per day) and low-frequency (2–12 Hz). For an unknown reason, the method was abandoned and further followup was never published. Andy [57] reported using the same type of stimulation (unilateral, short periods of stimulation 3–4 times per day, 50 Hz frequency, 2–5 V) in two patients with torticollis. Although one patient was proclaimed to have a ‘‘good functional result’’ until a parietal scalp infection led to removal of the electrode 15 months after the procedure, the second patient had a good to excellent outcome. Cooper and colleagues [58] reported a series of 21 patients with movement disorders of various types treated with DBS. This series included six patients with dystonia and two with torticollis. The VL thalamic target was utilized, often bilaterally. Only one of the six patients with dystonia benefited from the procedure, whereas the two patients with torticollis showed fair improvement.
1875
1876
111
Central procedures for cervical dystonia
While thalamic DBS may provide some advantages for secondary generalized dystonias [54], it is generally not advocated for cervical dystonia. Instead, the globus pallidus pars interna (GPi) has become the preferred target for these patients.
Pallidal Stimulation Based on the reported efficacy of pallidotomy for the dystonia associated with PD [59], as well as for generalized dystonia [60], posteroventral GPi DBS is the most widely utilized stereotactic target for all types of dystonia. While the observation that pallidal surgery could reduce dystonia in PD was serendipitous, it is not without scientific rationale.
Rationale for Bilateral Surgery Several lines of evidence suggest that cervical dystonia patients have bilateral basal ganglia dysfunction regardless of the clinical manifestations in individual patients. Using positron emission tomography (PET), Magyar-Lehmann et al. [61] demonstrated higher glucose metabolism in the lentiform nucleus bilaterally in cervical dystonia patients than normal controls. There was no correlation with the laterality, specific pattern, or severity of cervical dystonia. In a single photon emission computed tomography study, Naumann et al. [62] also reported bilateral basal ganglia involvement in this patient population. Such bilateral involvement can be at least partially explained by the considerable bi-hemispheric representation of neck muscles in humans [63]. Further evidence exists from the lesion experience in which patients who had bilateral surgery improved more than those who underwent unilateral procedures [53]. Unilateral pallidal stimulation is mainly used when there is a pallidotomy on the contralateral side. However, there are rare case reports of good results with solely unilateral
procedures [64], although which side to stimulate remains contentious [65–67].
Outcome of (Bilateral) Pallidal DBS Bilateral pallidal DBS was first reported for cervical dystonia in 1999 [64,68]. Following this, several case reports and small series further outlined its efficacy [20,69–72]. Other publications had larger series, but combined multiple dystonia types [20,73–77]. However, these publications had significant limitations, including mixing of multiple dystonia subtypes [78], relatively small sample sizes [72,76], inconsistent selection criteria and use of validated clinical scoring methods [74], unblinded outcome measures [79], and short duration of follow-up [78]. > Table 111-1 summarizes the outcomes for all the publications. The largest retrospective series was published by Hung et al. [79] and included ten patients with cervical dystonia and bilateral pallidal DBS followed for a mean of 32 months. At last follow-up, the patients total TWSTRS score improved by 54%, with the severity subscore improving by 55%, disability subscore improving by 59%, and the pain subscore improving by 50%. The only prospective multicenter trial of pallidal DBS for cervical dystonia was performed in Canada [37]. While the study was planned as a pilot, it produced statistically and clinically significant results. Patients were recruited at five centers with experience in DBS surgery and movement disorders. Standard inclusion criteria were required and all patients underwent pre- and postoperative MR imaging, neuropsychologic assessment and barium swallow. The primary outcome was change in TWSTRS severity score 1 year after bilateral GPi-DBS as assessed by two blinded movement disorder neurologists. Secondary outcomes included TWSTRS disability, pain, and total scores, Short Form-36 (SF-36) and the Beck depression inventory. Statistically significant changes were observed in all outcomes. > Figure 111-1 shows
Central procedures for cervical dystonia
111
. Table 111-1 Outcomes of GPi-DBS reported up to March 2008 Reference
N
Type of dystonia
Pallidal DBS
Follow-up (months)
[68]
3
Cervical
Bilateral
6–15
[64]
1
Cervical
Unilateral
[69]
2
Cervical
Bilateral
Not reported 17–24
[71] [70] [73] [20]
1 3 2 5
Cervical + truncal Cervical 2 cervical (of 6) Cervical
Unilateral Bilateral Bilateral Bilaterala
[66] [65]
1 1
Unilateral Unilateral
[74,80]
7
Cervical + truncal Secondary cervical 7 cervical (of 25)
8 2–6 3–12 12–30 (mean 20) 18 mo 12 mo
Bilateral
4–24 mo
[76] [72] [77] [79]
3 4 6 10
3 cervical (of 15) Cervical 6 cervical (of 12) Cervical
Bilateral Bilateral Bilateral Bilateral
6 months 15 months 2 years 32 mo
[37]
10
Cervical
Bilateral
1 year
Outcomes and scales utilized Total TWSTRS improved by 47% (severity by 39%; pain by 43%; disability by 52%) ‘‘Effectively’’ reduced dystonia Small improvement in Tsui scale, marked improvement in pain on visual analogue scale Total TWSTRS improved by 50% Obtained ‘‘substantial benefit’’ Tsui scale improved by 63% TWSTRS severity improved by 63% (pain by 50%; disability by 69%) Tsui score improved by 78% ‘‘marked’’ improvement of dystonia Total TWSTRS improved by 60% (severity by 63%; pain by 59%; disability by 59%) Total TWSTRS improved by 58% Total TWSTRS improved by 73% Total TWSTRS improved by 58% Total TWSTRS improved by 54% (severity by 55%; pain by 50%; disability by 59%) Total TWSTRS improved by 59% (severity by 43%; pain by 64%; disability by 65%)
a
Except one patient had unilateral DBS because of previous thalamotomy
the TWSTRS results in individual patients and the group as a whole. The mean improvement for TWSTRS total, severity, disability, and pain scores was 59, 43, 64, and 65%, respectively. Quality of life scores improved and depression scores dropped significantly as well (> Figure 111-2). All but one patient improved in all outcomes measured.
Stimulation Parameters There is currently no consensus regarding the optimal stimulation parameters required for pallidal DBS in patients with dystonia. The parameters that can be adjusted include pulse duration (in ms), frequency (in Hz), polarity, and amplitude (in V).
Pulse widths for dystonia therapy tend to be less uniform, although generally longer than that required for PD or tremor [54,81,82]. A large pulse-width more rapidly reduces battery life-expectancy, necessitating more frequent surgical replacements and corresponding costs and risk of complications. The only study that specifically addressed pulse duration was reported by Vercueil et al. [83] and focused on 20 patients with primary generalized dystonia treated by GPi DBS. Each patient was examined 10 h after the change in stimulation settings using the Burke-Fahn-Marsden rating scale. Three different pulse widths were applied at similar charge densities (meaning that voltage applied was reduced at higher pulse durations). An independent blinded neurologist failed to find a difference between short (0.06 ms), medium (0.12 ms) and long (0.45 ms) duration
1877
1878
111
Central procedures for cervical dystonia
. Figure 111-1 Results of bilateral pallidal stimulation for cervical dystonia. Each patient is illustrated as an individual symbol and line indicating their outcome. The mean (SD) scores for the whole group are shown as bar graphs. Total TWSTRS scores (A, P < 0.001), severity (B, P = 0.003), disability (C, P < 0.001) and pain subscores (D, P < 0.001, RM-ANOVA) are shown at baseline, 6 and 12 months post-operatively. Adapted from [37] with permission
pulsesin these short term adjustments. There are no similar studies performed under chronic therapeutic conditions or in the setting of cervical dystonia. In DBS terminology, high frequency stimulation is that above 100 Hz. Early reports suggested that GPi DBS for PD had the most beneficial effects at frequencies as high as 185 Hz [84]. Most centers use frequencies ranging from 130 to 185 Hz [80,82,85–87]. In a study by Kupsch et al. [88] also in generalized dystonia patients with GPi-DBS in which patients were examined
4 h after a change in stimulation parameters (noadjustmentsmadeforchargedensity),improvements were noted at 130 Hz with further improvements at 180 and 250 Hz. While 5 and 50 Hz were better than no stimulation at all, they were worse that stimulation applied at 130 Hz. The lack of consensus is further demonstrated by reports of low frequency stimulation improving dystonic symptoms. Kumar et al. [89] reported a frequency-dependent response in one patient with generalized dystonia with the best results at 5 Hz
Central procedures for cervical dystonia
111
. Figure 111-2 Depression (A) and quality of life (SF-36, B) scores pre-operatively and at 6 and 12 months post-operatively, showing significant improvements over time (A: P < 0.001, B: P = 0.003, RM-ANOVA). Each patient score is illustrated as a distinct symbol and line, and the group mean (SD) as the bar graphs. Adapted from [37] with permission
stimulation.Similarly,Altermanetal.[90,91]found benefits at 60 Hz in 15 consecutive patients with generalized dystonia.
Time Course and Duration of Effect Generally the effects of GPi DBS on dystonia are not immediate and the improvement in symptoms follows a sequence of improvement with pain first, then motor disability and finally severity. Dystonic movements (including phasic, myoclonic and tremulous features) may improve immediately or within hours or days after surgery [75,80,82,86]. Dystonic postures (i.e., tonic features) generally have a delayed improvement over weeks to months [79]. The pattern of improvement was found to be linear by Yianni et al. [74] but exponential by Bittar et al. [77] reaching a plateau midway through the first year. When one considers programming of the DBS systems for cervical dystonia patients, we usually apply at least 24 h of continuous stimulation on any one setting before re-programming. Because we found rapid improvements in some
of our patients [92], we do not recommend waiting longer than 1 week between programming sessions, if no improvements are obtained. With respect to long term stimulation it appears that benefits were sustained for 6–9 years in four cervical dystonia patients with pallidal stimulation [93]. However, pulse generators had to be replaced within 1.5–2 years. One case report described a patient in whom ongoing stimulation was not required to maintain good clinical benefit [94]. It is difficult to interpret too much from this case. The patient was implanted with DBS electrodes yet had only 2 years of cervical dystonia documented, therefore the ongoing improvement without stimulation may be attributed to spontaneous remission. Another potential explanation, which requires further investigation, is the ability of stimulation to induce long-term plastic changes.
Predictors of Success Genetically determined primary generalized dystonia (DYT1-positive) predicts improvement following GPi DBS (in generalized dystonia)
1879
1880
111
Central procedures for cervical dystonia
and 90% improvements are seen on the BurkeFahn-Marsden Dystonia Rating Scale [85,87,88, 95,96]. However, most patients with isolated cervical dystonia do not have a known genetic cause or even a family history of similar disorders. Another predictor of success with pallidal stimulation for generalized dystonia is a normal MRI. Patients with abnormal imaging, have by definition secondary dystonia, which responds less well to all types of surgery [76,97].
Neuropsychological risks exist with bilateral surgery and have been particularly well explored in PD patients [104] however, there have been no reports of significant cognitive decline in the dystonia population. In the Canadian multicenter study of GPi DBS for cervical dystonia, we reported details of neuropsychological outcomes [37]. While some declines in phonemic fluency and verbal memory were discovered, these were not significant enough to impact daily life or working ability.
Complications While DBS surgery is generally safe there are potential risks, most of which have been outlined in previous chapters. The specific risks related to cervical dystonia relate to swallowing dysfunction and use of pre-operative botulinum toxin injections. Botulinum toxin injected into cervical muscles may spread to affect pharyngeal muscles and may cause problems with swallowing [98]. Hardware-related complications can occur with any implantable system. However risks for these may be higher in cervical dystonia patients and relate to the abnormal movements of the neck where the extension from DBS lead to pulse generator travels. The general rate of hardware related problems in all DBS surgeries varies from 4.3 to 13% per electrode-year [99–101]. In a review of ten studies investigating adverse effects in 922 patients with different movement disorders, the most commonly reported hardware-related complication was infection (in 6.1% of patients), migration or misplacement of the leads (5.1% of patients), lead fractures (5.0% of patients), and skin erosion (1.3% of patients) [102]. While Yianni et al. [103] found an overall rate of 5.3% hardware malfunction in their series of 133 patients, all failures were in dystonia patients: 18.4% of all dystonia patients and 9.2% of all electrodes implanted in dystonia patient. These occurred more commonly as late complications 6 months post-operatively [99].
Outcomes in Cranial Cervical Dystonia There are only a few case reports in the literature regarding pallidal stimulation for cranial cervical dystonia [55,105–107]. Single cases are reported as part of larger series with mixed types of dystonia, but outcome has not been separately examined. In the largest series by Ostrem et al. [106] 5 of 6 patients had cervical dystonia with blepharospasm. Six months after implantation of bilateral pallidal stimulators, TWSTRS subscores for severity, disability, pain, and total scores had improved by 56, 67, 38, and 56%, respectively. For these five patients, the Burke-Fahn-Marsden Dystonia Rating Scale total movement and disability scores improved by 70 and 42% respectively. Several of the patients experienced a dramatic clinical improvement as early as 1 week after surgery, although it was not stated if the improvements were in the blepharospastic or cervical dystonia component. However, most of the patients reported clear worsening of some motor functions in previously normal body regions after GPi DBS. These included slowness in walking, heaviness in their legs, and subtle difficulty making turns. Others reported new difficulties with writing for long periods of time, typing, or trouble getting out of a bathtub. Similar adverse motor effects have not been reported in previous studies of GPi DBS in (generalized) dystonia.
Central procedures for cervical dystonia
Blomstedt et al. [107] reported the case of a patient with Meige syndrome with ‘‘neck tension’’ in whom a unilateral left pallidal DBS relieved right arm tremor, but provided no benefit for the blepharospasm, oromandibular dystonia, speech or swallowing disturbances. Subsequent implantation of a right pallidal DBS and bilateral stimulation at high amplitudes achieved an effect on the neck tension, blepharospasm, and oromandibular dystonia. The BurkeFahn-Marsden dystonia score improved by 72%.
Summary Overall, bilateral GPi-DBS seems to provide the best outcomes for cervical dystonia, with minimal side effects. However there are many questions that require further investigation, such as the following. Is the GPi the best target? Is bilateral stimulation required or could a unilateral pallidotomy and contralateral GPi-DBS provide as good benefit? Is continuous long-term stimulation required, or can stimulation produce enough plastic change that the devices could be turned off? What are the best electrical parameters to improve cervical dystonia? These questions and many more will keep the stereotactic and functional neurosurgery community active for many years to come.
References 1. Jankovic J, Leder S, Warner D, Schwartz K. Cervical dystonia: clinical findings and associated movement disorders. Neurology 1991;41(7):1088-91. 2. Jankovic J, Tolosa E. Dystonic disorders. In: Jankovic J, Tolosa E, editors. Parkinson’s disease and movement disorders. Philadelphia, PA: Lippincott Williams & Wilkins; 2007. p. 321-47. 3. Chan J, Brin MF, Fahn S. Idiopathic cervical dystonia: clinical characteristics. Mov Disord 1991;6(2):119-26. 4. Kutvonen O, Dastidar P, Nurmikko T. Pain in spasmodic torticollis. Pain 1997;69(3):279-86. 5. Claypool DW, Duane DD, Ilstrup DM, Melton LJ, III. Epidemiology and outcome of cervical dystonia (spasmodic torticollis) in Rochester, Minnesota. Mov Disord 1995;10 (5):608-14.
111
6. Pekmezovic T, Ivanovic N, Svetel M, Nalic D, Smiljkovic T, Raicevic R, Kostic VS. Prevalence of primary late-onset focal dystonia in the Belgrade population. Mov Disord 2003;18(11):1389-92. 7. Le KD, Nilsen B, Dietrichs E. Prevalence of primary focal and segmental dystonia in Oslo. Neurology 2003;61(9):1294-6. 8. Fukuda H, Kusumi M, Nakashima K. Epidemiology of primary focal dystonias in the western area of Tottori prefecture in Japan: comparison with prevalence evaluated in 1993. Mov Disord 2006;21(9):1503-6. 9. Marras C, et al. Minimum incidence of primary cervical dystonia in a multiethnic health care population. Neurology 2007;69(7):676-80. 10. Soland VL, Bhatia KP, Marsden CD. Sex prevalence of focal dystonias. J Neurol Neurosurg Psychiatry 1996;60(2):204-5. 11. Meares R. Natural history of spasmodic torticollis, and effect of surgery. Lancet 1971;2(7716):149-50. 12. Lowenstein DH, Aminoff MJ. The clinical course of spasmodic torticollis. Neurology 1988;38(4):530-2. 13. Jahanshahi M, Marion MH, Marsden CD. Natural history of adult-onset idiopathic torticollis. Arch Neurol 1990;47(5):548-52. 14. Friedman A, Fahn S. Spontaneous remissions in spasmodic torticollis. Neurology 1986;36(3):398-400. 15. Jayne D, Lees AJ, Stern GM. Remission in spasmodic torticollis. J Neurol Neurosurg Psychiatry 1984;47 (11):1236-7. 16. Rondot P, Marchand MP, Dellatolas G. Spasmodic torticollis – review of 220 patients. Can J Neurol Sci 1991;18(2):143-51. 17. Consky E, Lang AE. Clinical assessments of patients with cervical dystonia. In: Jankovic J, editor. Therapy with botulinum toxin. New York: Informa Healthcare; 1994. p. 211-238. 18. Jahanshahi M. Psychosocial factors and depression in torticollis. J Psychosom Res 1991;35(4–5):493-507. 19. Chawda SJ, Munchau A, Johnson D, Bhatia K, Quinn NP, Stevens J, Lees AJ, Palmer JD. Pattern of premature degenerative changes of the cervical spine in patients with spasmodic torticollis and the impact on the outcome of selective peripheral denervation. J Neurol Neurosurg Psychiatry 2000;68(4):465-71. 20. Krauss JK, Loher TJ, Pohle T, Weber S, Taub E, Barlocher CB, Burgunder JM. Pallidal deep brain stimulation in patients with cervical dystonia and severe cervical dyskinesias with cervical myelopathy. J Neurol Neurosurg Psychiatry 2002;72(2):249-56. 21. Frei KP, Pathak M, Jenkins S, Truong DD. Natural history of posttraumatic cervical dystonia. Mov Disord 2004;19(12):1492-8. 22. Jankovic J. Can peripheral trauma induce dystonia and other movement disorders? Yes! Mov Disord 2001;16(1):7-12.
1881
1882
111
Central procedures for cervical dystonia
23. Sa DS, Mailis-Gagnon A, Nicholson K, Lang AE. Posttraumatic painful torticollis. Mov Disord 2003;18(12):1482-91. 24. Castelbuono A, Miller NR. Spontaneous remission in patients with essential blepharospasm and Meige syndrome. Am J Ophthalmol 1998;126(3):432-5. 25. Mauriello JA, Jr, Dhillon S, Leone T, Pakeman B, Mostafavi R, Yepez MC. Treatment selections of 239 patients with blepharospasm and Meige syndrome over 11 years. Br J Ophthalmol 1996;80(12):1073-6. 26. Grandas F, Elston J, Quinn N, Marsden CD. Blepharospasm: a review of 264 patients. J Neurol Neurosurg Psychiatry 1988;51(6):767-72. 27. Svetel M, Pekmezovic T, Jovic J, Ivanovic N, Dragasevic N, Maric J, Kostic VS. Spread of primary dystonia in relation to initially affected region. J Neurol 2007;254(7):879-83. 28. Bramstedt KA, Ford PJ. Protecting human subjects in neurosurgical trials: the challenge of psychogenic dystonia. Contemp Clin Trials 2006;27(2):161-4. 29. Hsiung GY, Das SK, Ranawaya R, Lafontaine AL, Suchowersky O. Long-term efficacy of botulinum toxin A in treatment of various movement disorders over a 10-year period. Mov Disord 2002;17(6):1288-93. 30. Waterston JA, Swash M, Watkins ES. Idiopathic dystonia and cervical spondylotic myelopathy. J Neurol Neurosurg Psychiatry 1989;52(12):1424-6. 31. el-Mallakh RS, Rao K, Barwick M. Cervical myelopathy secondary to movement disorders: case report. Neurosurgery 1989;24(6):902-5. 32. Spitz M, Goncalves L, Silveira L, Barbosa E. Myelopathy as a complication of cervical dystonia. Mov Disord 2006;21(5):726-7. 33. Lang AE, Widner H. Deep brain stimulation for Parkinson’s disease: patient selection and evaluation. Mov Disord 2002;17 Suppl 3:S94-101. 34. Tsui JK, Eisen A, Stoessl AJ, Calne S, Calne DB. Doubleblind study of botulinum toxin in spasmodic torticollis. Lancet 1986;2(8501):245-7. 35. Consky E, Basinki DJ, Ranawaya R, Lang AE. The Toronto Western Spasmodic Torticollis Ratin Scale (TWSTRS): assessment of validity and inter-rater reliability. Neurology 1990;40(4 Suppl 1):445. 36. Burke RE, Fahn S, Marsden CD, Bressman SB, Moskowitz C, Friedman J. Validity and reliability of a rating scale for the primary torsion dystonias. Neurology 1985;35(1):73-7. 37. Kiss ZH, Doig-Beyaert K, Eliasziw M, Tsui J, Haffenden A, Suchowersky O. The Canadian multicentre study of deep brain stimulation for cervical dystonia. Brain 2007;130(Pt 11):2879-86. 38. Pillon B, et al. Preservation of cognitive function in dystonia treated by pallidal stimulation. Neurology 2006;66(10):1556-8.
39. Bejjani BP, et al. Transient acute depression induced by high-frequency deep-brain stimulation. N Engl J Med 1999;340(19):1476-80. 40. Bejjani BP, et al. Bilateral subthalamic stimulation for Parkinson’s disease by using three-dimensional stereotactic magnetic resonance imaging and electrophysiological guidance. J Neurosurg 2000;92(4):615-25. 41. Houeto JL, et al. Behaviouraldisorders,Parkinson’sdisease and subthalamic stimulation. J Neurol Neurosurg Psychiatry 2002;72(6):701-7. 42. Doshi PK, Chhaya N, Bhatt MH. Depression leading to attempted suicide after bilateral subthalamic nucleus stimulation for Parkinson’s disease. Mov Disord 2002;17(5):1084-5. 43. Burkhard PR, Vingerhoets FJ, Berney A, BogousslavskyJ, Villemure JG, Ghika J. Suicide after successful deep brain stimulation for movement disorders. Neurology 2004;63(11):2170-2. 44. Hassler R. Anatomy of the thalamus. In: Schaltenbrand G, Bailey P, editors. Introduction to stereotaxis with an atlas of the human brain. Stuttgart: G. Thieme; 1959. p. 230-90. 45. Cooper IS. Effect of thalamic lesions upon torticollis. N Engl J Med 1964;270:567-72. 46. Hassler R, Hess WR. [Experimental and anatomical findings in rotatory movements and their nervous apparatus.]. Arch Psychiatr Nervenkr Z Gesamte Neurol Psychiatr 1954;192(5):488-526. 47. Sano K, Yoshioka M, Mayanagi Y, Sekino H, Yoshimasu N. Stimulation and destruction of and around the interstitial nucleus of Cajal in man. Confin Neurol 1970;32(2):118-25. 48. Cooper IS. 20-year followup study of the neurosurgical treatment of dystonia musculorum deformans. Adv Neurol 1976;14:423-52. 49. von Essen C, Augustinsson LE, Lindqvist G. VOI thalamotomy in spasmodic torticollis. Appl Neurophysiol 1980;43(3–5):159-63. 50. Andrew J, Fowler CJ, Harrison MJ. Stereotaxic thalamotomy in 55 cases of dystonia. Brain 1983;106 (Pt 4):981-1000. 51. Cardoso F, Jankovic J, Grossman RG, Hamilton WJ. Outcome after stereotactic thalamotomy for dystonia and hemiballismus. Neurosurgery 1995;36(3):501-7. 52. Tasker RR, Doorly T, Yamashiro K. Thalamotomy in generalized dystonia. Adv Neurol 1988;50:615-31. 53. Loher TJ, Pohle T, Krauss JK. Functional stereotactic surgery for treatment of cervical dystonia: review of the experience from the lesional era. Stereotact Funct Neurosurg 2004;82(1):1-13. 54. Krauss JK, Yianni J, Loher TJ, Aziz TZ. Deepbrainstimulationfordystonia.J Clin Neurophysiol 2004;21(1):18-30. 55. Muta D, Goto S, Nishikawa S, Hamasaki T, Ushio Y, Inoue N, Mita S. Bilateral pallidal stimulation for
Central procedures for cervical dystonia
56.
57. 58.
59.
60.
61.
62.
63.
64.
65.
66.
67. 68.
69.
idiopathic segmental axial dystonia advanced from Meige syndrome refractory to bilateral thalamotomy. Mov Disord 2001;16(4):774-7. Mundinger F. [New stereotactic treatment of spasmodic torticollis with a brain stimulation system (author’s transl)]. Med Klin 1977;72(46):1982-6. Andy OJ. Thalamic stimulation for control of movement disorders. Appl Neurophysiol 1983 ;46(1–4):107-11. Cooper IS, Upton AR, Amin I. Chronic cerebellar stimulation (CCS) and deep brain stimulation (DBS) in involuntary movement disorders. Appl Neurophysiol 1982;45 (3):209-17. Baron MS, Vitek JL, Bakay RA, Green J, McDonald WM, Cole SA, DeLong MR. Treatment of advanced Parkinson’s disease by unilateral posterior GPi pallidotomy. 4-year results of a pilot study. Mov Disord 2000;15(2):230-7. Ondo WG, Desaloms JM, Jankovic J, Grossman RG. Pallidotomy for generalized dystonia. Mov Disord 1998;13(4):693-8. Magyar-Lehmann S, Antonini A, Roelcke U, Maguire RP, Missimer J, Meyer M, Leenders KL. Cerebral glucose metabolism in patients with spasmodic torticollis. Mov Disord 1997;12(5):704-8. Naumann M, Pirker W, Reiners K, Lange KW, Becker G, Brucke T. Imaging the pre- and postsynaptic side of striatal dopaminergic synapses in idiopathic cervical dystonia: a SPECT study using [123I] epidepride and [123I] beta-CIT. Mov Disord 1998;13(2):319-23. Thompson ML, Thickbroom GW, Mastaglia FL. Corticomotor representation of the sternocleidomastoid muscle. Brain 1997;120(Pt 2):245-55. Islekel S, Zileli M, Zileli B. Unilateral pallidal stimulation in cervical dystonia. Stereotact Funct Neurosurg 1999;72 (2–4):248-52. Chang JW, Choi JY, Lee BW, Kang UJ, Chung SS. Unilateral globus pallidus internus stimulation improves delayed onset post-traumatic cervical dystonia with an ipsilateral focal basal ganglia lesion. J Neurol Neurosurg Psychiatry 2002;73(5):588-90. Escamilla-Sevilla F, Minguez-Castellanos A, rjonaMoron V, Martin-Linares JM, Sanchez-Alvarez JC, Ortega-Morenoa A, Garcia-Gomez T. Unilateral pallidal stimulation for segmental cervical and truncal dystonia: which side? Mov Disord 2002;17(6):1383-5. Krauss JK. Deep brain stimulation for cervical dystonia. J Neurol Neurosurg Psychiatry 2003;74(11):1598. Krauss JK, Pohle T, Weber S, Ozdoba C, Burgunder JM. Bilateral stimulation of globus pallidus internus for treatment of cervical dystonia. Lancet 1999;354(9181):837-8. KulisevskyJ, LleoA, GironellA, MoletJ, Pascual-Sedano B, Pares P. Bilateral pallidal stimulation for cervical dystonia: dissociated pain and motor improvement. Neurology 2000;55(11):1754-5.
111
70. Parkin S, Aziz T, Gregory R, Bain P. Bilateral internal globus pallidus stimulation for the treatment of spasmodic torticollis. Mov Disord 2001;16(3):489-93. 71. Andaluz N, Taha JM, Dalvi A. Bilateral pallidal deep brain stimulation for cervical and truncal dystonia. Neurology 2001;57(3):557-8. 72. Eltahawy HA, Saint-Cyr J, Poon YY, Moro E, Lang AE, Lozano AM. Pallidal deep brain stimulation in cervical dystonia: clinical outcome in four cases. Can J Neurol Sci 2004b;31(3):328-32. 73. Bereznai B, Steude U, Seelos K, Botzel K. Chronic high-frequency globus pallidus internus stimulation in different types of dystonia: a clinical, video, and MRI report of six patients presenting with segmental, cervical, and generalized dystonia. Mov Disord 2002;17(1):138-44. 74. Yianni J, Bain PG, Gregory RP, Nandi D, Joint C, Scott RB, Stein JF, Aziz TZ. Post-operative progress of dystonia patients following globus pallidus internus deep brain stimulation. Eur J Neurol 2003b;10 (3):239-47. 75. Krause M, Fogel W, Kloss M, Rasche D, Volkmann J, Tronnier V. Pallidal stimulation for dystonia. Neurosurgery 2004;55(6):1361-8. 76. Eltahawy HA, Saint-Cyr J, Giladi N, Lang AE, Lozano AM. Primary dystonia is more responsive than secondary dystonia to pallidal interventions: outcome after pallidotomy or pallidal deep brain stimulation. Neurosurgery 2004a;54(3):613-9. 77. Bittar RG, et al. Deep brain stimulation for generalised dystonia and spasmodic torticollis. J Clin Neurosci 2005;12(1):12-16. 78. Kupsch A, et al. Pallidal deep-brain stimulation in primary generalized or segmental dystonia. N Engl J Med 2006;355(19):1978-90. 79. Hung SW, et al. Long-term outcome of bilateral pallidal deep brain stimulation for primary cervical dystonia. Neurology 2007;68(6):457-9. 80. Yianni J, et al. Globus pallidus internus deep brain stimulation for dystonic conditions: a prospective audit. Mov Disord 2003a;18(4):436-42. 81. Deep-Brain Stimulation for Parkinson’s Disease Study Group. Deep-brain stimulation of the subthalamic nucleus or the pars interna of the globus pallidus in Parkinson’s disease. N Engl J Med 2001;345(13):956-63. 82. Vidailhet M, et al. Bilateral deep-brain stimulation of the globus pallidus in primary generalized dystonia. N Engl J Med 2005;352(5):459-67. 83. Vercueil L, et al. Effects of pulse width variations in pallidal stimulation for primary generalized dystonia. J Neurol 2007;254(11):1533-7. 84. Kumar R. Methods for programming and patient management with deep brain stimulation of the globus pallidus for the treatment of advanced Parkinson’s disease and dystonia. Mov Disord 2002;17 Suppl 3:S198-207.
1883
1884
111
Central procedures for cervical dystonia
85. Cif L, El FH, Vayssiere N, Hemm S, Hardouin E, Gannau A, Tuffery S, Coubes P. Treatment of dystonic syndromes by chronic electrical stimulation of the internal globus pallidus. J Neurosurg Sci 2003;47(1):52-5. 86. Coubes P, et al. Electrical stimulation of the globus pallidus internus in patients with primary generalized dystonia: long-term results. J Neurosurg 2004;101(2):189-94. 87. Starr PA, Turner RS, Rau G, Lindsey N, Heath S, Volz M, Ostrem JL, Marks WJ, Jr. Microelectrode-guided implantation of deep brain stimulators into the globus pallidus internus for dystonia: techniques, electrode locations, and outcomes. J Neurosurg 2006;104(4):488-501. 88. Kupsch A, Klaffke S, Kuhn AA, Meissner W, Arnold G, Schneider GH, Maier-Hauff K, Trottenberg T. The effects of frequency in pallidal deep brain stimulation for primary dystonia. J Neurol 2003;250(10):1201-5. 89. Kumar R, Dagher A, Hutchison WD, Lang AE, Lozano AM. Globus pallidus deep brain stimulation for generalized dystonia: clinical and PET investigation. Neurology 1999;53(4):871-4. 90. Alterman RL, Miravite J, Weisz D, Shils JL, Bressman SB, Tagliati M. Sixty hertz pallidal deep brain stimulation for primary torsion dystonia. Neurology 2007a;69(7):681-8. 91. Alterman RL, Shils JL, Miravite J, Tagliati M. Lower stimulation frequency can enhance tolerability and efficacy of pallidal deep brain stimulation for dystonia. Mov Disord 2007b;22(3):366-8. 92. Kiss ZH, Doig K, Eliasziw M, Ranawaya R, Suchowersky O. The Canadian multicenter trial of pallidal deep brain stimulation for cervical dystonia: preliminary results in three patients. Neurosurg Focus 2004;17(1):E5. 93. Loher TJ, Capelle HH, Kaelin-Lang A, Weber S, Weigel R, Burgunder JM, Krauss JK. Deep brain stimulation for dystonia: outcome at long-term follow-up. J Neurol 2008;255(6):881-4. 94. Goto S, Yamada K. Long term continuous bilateral pallidal stimulation produces stimulation independent relief of cervical dystonia. J Neurol Neurosurg Psychiatry 2004;75(10):1506-7. 95. Vercueil L, Pollak P, Fraix V, Caputo E, Moro E, Benazzouz A, Xie J, Koudsie A, Benabid AL. Deep brain stimulation in the treatment of severe dystonia. J Neurol 2001;248(8):695-700.
96. Tagliati M, Shils J, Sun C, Alterman R. Deep brain stimulation for dystonia. Expert Rev Med Devices 2004;1(1):33-41. 97. Alterman RL, Snyder BJ. Deep brain stimulation for torsion dystonia. Acta Neurochir Suppl 2007;97 (Pt 2):191-9. 98. Borodic GE, Joseph M, Fay L, Cozzolino D, Ferrante RJ. Botulinum A toxin for the treatment of spasmodic torticollis: dysphagia and regional toxin spread. Head Neck 1990;12(5):392-9. 99. Paluzzi A, Belli A, Bain P, Liu X, Aziz TM. Operative and hardware complications of deep brain stimulation for movement disorders. Br J Neurosurg 2006;20(5):290-5. 100. Voges J, Waerzeggers Y, Maarouf M, Lehrke R, Koulousakis A, Lenartz D, Sturm V. Deep-brain stimulation: long-term analysis of complications caused by hardware and surgery – experiences from a single centre. J Neurol Neurosurg Psychiatry 2006;77(7):868-72. 101. Kenney C, Simpson R, Hunter C, Ondo W, Almaguer M, Davidson A, Jankovic J. Short-term and long-term safety of deep brain stimulation in the treatment of movement disorders. J Neurosurg 2007;106 (4):621-5. 102. Hamani C, Lozano AM. Hardware-related complications of deep brain stimulation: a review of the published literature. Stereotact Funct Neurosurg 2006;84 (5–6):248-51. 103. Yianni J, Nandi D, Shad A, Bain P, Gregory R, Aziz T. Increased risk of lead fracture and migration in dystonia compared with other movement disorders following deep brain stimulation. J Clin Neurosci 2004;11(3):243-5. 104. Saint-Cyr JA, Trepanier LL. Neuropsychologic assessment of patients for movement disorder surgery. Mov Disord 2000;15(5):771-83. 105. Opherk C, Gruber C, Steude U, Dichgans M, Botzel K. Successful bilateral pallidal stimulation for Meige syndrome and spasmodic torticollis. Neurology 2006;66(4): E14. 106. Ostrem JL, Marks WJ, Jr, Volz MM, Heath SL, Starr PA. Pallidal deep brain stimulation in patients with cranialcervical dystonia (Meige syndrome). Mov Disord 2007;22(13):1885-91. 107. Blomstedt P, Tisch S, Hariz MI. Pallidal deep brain stimulation in the treatment of Meige syndrome. Acta Neurol Scand 2008;118(3):198-202.
108 Central Procedures for Primary Dystonia X. A. Vasques . L. Cif . B. Biolsi . P. Coubes
From the wide range of neurological disorders, very few are accessible to surgical procedures. In 1911, Oppenheim reported on ‘‘dystonia musculorum deformans’’ which had previously been considered by Gowers (1888) and Schwalbe (1908) as a psychiatric disorder. Dystono-dyskinetic syndromes (DDS) are conditions in which a pathological process occurring in the basal ganglia induces a decrease in the inhibition of the brain cortex [1,2]. The syndrome is characterized by sustained muscle contraction leading to repetitive twisting movements and abnormal postures (> Figure 108-1) [3–7] caused by concurrent contractions of agonist and antagonist muscles. On the basis of results obtained with bilateral pallidotomy for treating the most severe forms of primary generalized DDS and the validated efficiency of deep brain stimulation (DBS) in Parkinson’s disease [8–14], the bilateral chronic electrical stimulation of the Globus Pallidus internus (GPi) was introduced to treat the DDS [15–19]. A review of the clinical features, of the most frequent etiologies, of the surgical procedures proposed over time, of the criteria for DBS patient selection and the results obtained by DBS in DDS in our center will be consecutively discussed.
Clinical Features Dystono-dyskinetic syndromes (DDS) include very different disorders classified according to the age of onset, body distribution and the cause. When symptoms begin before the age of 26 years, DDS is termed ‘‘early-onset’’ and ‘‘late-onset’’ #
Springer-Verlag Berlin/Heidelberg 2009
otherwise [20,21]. In our experience it seems more convenient to speak about early onset before the age of ten. According to distribution, DDS are classified into one of the following categories: focal, segmental, multifocal, hemi-DDS and generalized DDS. When a single body region is affected the term of ‘‘focal’’ is used. The focal DDS includes Meige syndrome, writer’s cramp, blepharospasm, cervical DDS and spasmodic dysphonia. The focal DDS are more common in adults. In segmental DDS, two or more contiguous regions are affected as cranial-cervical DDS (e.g., tardive DDS), crural DDS or brachial-cervical DDS (e.g., DYT11 Myoclonus DDS). When two or more non-contiguous regions are affected, DDS is termed ‘‘multifocal DDS.’’ Generalized DDS starts most often in a limb and tends to spread to other body parts. Hemi-DDS is called when the DDS are confined to one side of the body. Several chromosomal loci and genes have been identified for primary disorders as well as in degenerative diseases including in their phenotype a dystono-dyskinetic syndrome. Primary (DYT1, DYT2, DYT4, DYT6, DYT7, DYT13), secondary (vascular, infectious, tumoral, drug-induced, perinatal cerebral injury, kernicterus, trauma, toxins, metabolic, paraneoplactic syndromes, central pontine myelinolysis), dystonia-plus (Dopa-responsive DDS, DYT11 myoclonus-dystonia, DYT12 rapid-onset dystoniaparkinsonism) and degenerative (mitochondrial diseases, glutaric aciduria, pantothenate kinase associated neurodegeneration, Lesch-Nyhan disease, etc.) groups represent the international etiological classification [7].
1802
108
Central procedures for primary dystonia
. Figure 108-1 Dystonia musculorum deformans
base pair deletion in the gene encoding torsinA, termed the DYT1 gene, is responsible for the disease [22], though only 30–40% of mutation carriers exhibit dystono-dyskinetic symptoms [23]. The DYT1 mutation results in the loss of one glutamic acid residue near the C-terminus of torsinA [22,24]. Several lines of evidence suggest that torsinA is a molecular chaperone that mediates neuroprotection against a variety of cellular insults, and that the disease-associated mutant has a loss-of-function phenotype [25].
Non-DYT1
Primary DDS Primary DDS refer to those syndromes in which DDS is the only clinical sign without other neurological feature (no history of a brain injury, no consistent associated brain pathology, no laboratory findings to suggest a cause for DDS and no clinical improvement with a trial of low-dose levedopa) [7].
DYT1 (Oppenheim’s Dystonia) The Oppenheim’s dystonia usually begins in an arm or leg, rarely in cranial or cervical muscles, and spreads to other areas of the body. The typical phenotype is of early onset starting in one leg and rapidly progressing towards generalization. The DYT1 DDS is the most frequent etiology for childhood onset primary DDS. In adults, there are various phenotypes and often less severe than in children onset. The mutation called DYT1 has been localized to the 9q32–34 region with an autosomal transmission. Its prevalence is ten times higher in the Ashkenase Jewish population (22 per million). A three
Many types of non-DYT1 DDS have been described in individual families but are considered to be rare [26–36]. For example, DYT4 has been used to describe an Australian family with whispering dysphonia and focal to generalized dystono-dyskinetic symptoms [26,37]. DYT6 has been described in German-Mennonite families [28,32] with onset of DDS in the adolescence. The inheritance is autosomal dominant with about 30% of penetrance. The prominent cranial involvement and impaired speech distinguish this ‘‘nonDYT1’’ family from the typical DYT1 phenotype.
Dystonia Plus In dystonia-plus, dystonia is a prominent sign but is usually associated with another movement disorder as parkinsonism (DYT5 and DYT12) [38–43] or myoclonus DDS (DYT11) [5] in which clinical and laboratory findings suggest neurochemical disorders, with no evidence of neurodegeneration.
Dopa-Responsive DDS The Segawa’s disease [44], caused by a gene mapped to chromosome 14, is usually of autosomal dominant inheritance due to GTP-cyclohydrolase
Central procedures for primary dystonia
1 deficiency (DYT5). The initial symptom is a gait difficulty. This syndrome is typically characterized by childhood onset and diurnal fluctuation DDS. The patients present parkinsonism, spasticity, and can suffer from sleep disorders. The main characteristic of the dopa-responsive dystonia is the complete and sustained response to low-dose of levodopa (50–300 mg/day) with a low incidence of complications such as motor fluctuations and dyskinesias. Recessive forms can also occur (tyrosine hydroxylase deficiency, sepiapterin reductase deficiency).
108
Rapid Onset Dystonia-Parkinsonism (DYT12) The rapid onset dystonia-parkinsonism (DYT12) is an autosomal dominant disorder characterized by an abrupt onset with a DDS predominantly affecting bulbar musculature (dysphagia, dysarthia, and mutism) in the second decade of life. Additional parkinsonism is associated (bradykinesia, postural instability). The symptoms evolve over hours or days and generally stabilize within weeks with no or slow further progression. The disease is due to a mutation in the Na/K-ATPase alpha3 subunit (chromosome 19q13).
Myoclonus Dystonia Syndrome Myoclonus dystonia syndrome (MDS) (DYT11) is an autosomal dominant disorder characterized by bilateral, alcohol-responsive myoclonic jerks with onset in childhood or early adolescence [45,46]. Although MDS is genetically heterogeneous [47], heterozygous mutations in the e-sarcoglycan gene on chromosome 7q21 (SGCE) have been identified in several MDS families [48–53]. Myoclonus is the most prominent clinical feature of MDS, with jerks usually involving neck, head, and upper limbs, but occasionally extending to lower limbs. DDS (mainly torticollis and writer’s cramp) often coexists with myoclonus and may occasionally be the only manifestation of the disease [45,50]. Associated psychiatric symptoms (panic attacks, obsessive–compulsive behavior, and anxiety) have been reported [54]. Although the course of the disease is benign in most cases, the abnormal movements can impair daily life activities in a subset of patients. In this population, the chronic alcohol abuse, common when patients discover the beneficial effect of alcohol on their movement, is a frequent and disruptive consequence of this disease. Response to different pharmacological treatments including clonazepam, anticholinergic drugs, levodopa, sodium valproate, piracetam, levetiracetam, 5-HTP is poor [55].
Secondary DDS Secondary DDS represent a large group of conditions leading to DDS often associated to other motor or extra-motor symptoms and result primarily from environmental causes. They are consecutive to drugs, stroke, tumors, infections, and also include DDS secondary to cerebral palsy. The anoxic perinatal injury is a very frequent cause and phenotypes are heterogeneous from mild deficit to severe tetraparetic forms associated to DDS. Delayed onset DDS in cerebral palsy can also occur. Drug induced tardive dystonia and/or dyskinesia generated by exposure to dopamine receptor blocking agents like neuroleptics involve most often orobuccolingual and trunk muscles.
Heredodegenerative DDS This category includes diseases in which DDS can be the prominent feature but most often associated with other neurological symptoms. Many of these diseases are due to genetic abnormalities [7]: 1.
Autosomal dominant: Huntington’s disease, spinocerebellar degenerations, dentaterubral-pallido-luysian atrophy, hereditary spastic paraplegia with DDS,
1803
1804
108 2.
3.
Central procedures for primary dystonia
Autosomal recessive: Wilson’s disease, Pantothenate kinase associated neurodegeneration (PKAN), Metabolic disorders with onset usually in neonatal period or infancy, amino acid disorders, lipid disorders, neuroacanthocytosis, mitochondrial diseases, Lesch-Nyhan syndrome.
Pantothenate Kinase-Associated Neurodegeneration Pantothenate kinase-associated neurodegeneration (PKAN) is a rare autosomal recessive disorder characterized by neurodegeneration and iron accumulation in the brain, due to pantothenate kinase 2 (PANK2) gene mutations [56]. Its frequency is estimated at one case per million. Progressive generalized DDS represent a major clinical feature of PKAN and is responsible for aberrant postures and motor incapacity, which usually worsen in childhood resulting in lifethreatening complications. The radiological hallmark of PKAN is the ‘‘eye of the tiger’’ sign, characterized by bilateral areas of hyperintensity within a region of hypointensity in the medial globus pallidus on T2-weighted magnetic resonance images (MRIs) [57–59]. Pharmacological therapies using dopamine, clonazepam, trihexyphenidyl, tetrabenazine, or baclofen usually carry a poor efficacy in PKAN [57–59].
uric acid, nephrolithiasis, and rarely renal failure. In classic LND, because of complete HPRT deficiency [62], the characteristic neurobehavioral disorder includes delayed acquisition of motor skills, hypotonia evolving towards generalized DDS, weakness, and spasticity mainly in lower limbs. Self-mutilating behavior is widely present, associated with aggressiveness and cognitive dysfunction. Pharmacological treatment have poor efficacy in controlling movement disorders and treatment for behavioral features remains elusive. The most dramatic feature of LND is the self injury. To prevent biting, patients must frequently be submitted to complete teeth extraction, limb attachment for controlling aggressive movements and eye pocking as well as maskwearing against spitting. The patients generally survive with a very poor quality of life.
Mitochondrial Diseases Mitochondrial diseases (MERRF, MELAS, Leber’s disease) are clinically and genetically heterogeneous disorders associating seizures, spasticity, weakness, dystonia, dyskinesia, visual impairment and delay in psychomotor milestones. Mitochondrial encephalopathies are associated with mutations in both, nuclear nDNA and mtDNA [63].
Huntington’s Disease Lesh Nyhan Lesch–Nyhan disease (LND) [60] is an X-linked recessive disorder caused by deficiency in hypoxanthine–guanine phosphoribosyl transferase (HPRT), a purine salvage enzyme. Missense, nonsense, frameshift, and small deletion/insertion mutations have been described in the HPRT1 gene (Xq26-q27.2), the only gene known to be associated with the syndrome [61]. Affected individuals exhibit overproduction of
Huntington’s disease (HD) is a fatal autosomal dominant neurodegenerative disorder caused by the expansion of repeat sequences of DNA triplet cytosine-adenine-guanine (CAG) in exon 1 of the IT15 gene on chromosome 4 encoding a ubiquitous cytosolic protein, huntingtin. The disease is characterized by progressive cognitive impairment, movement disorders (chorea) and psychiatric symptoms. Treatment remains symptomatic and comprises neuroleptic drugs. The clinical efficacy of pharmacological treatment for choreic
Central procedures for primary dystonia
movements remains poor. Furthermore adverse effects are commonly dose limiting.
Wilson’s Disease Wilson’s disease (WD) is a treatable autosomal recessive metabolic disorder characterized by hepatic copper overload and caused by mutations in the gene encoding the copper-transporting P-type adenosine triphosphatase (ATPase) ATP7B. Neurological presentations are variable in respect to both pattern and age of onset. Commonly a movement disorders occurs in the second or third decade of life. WD can mimic several neurological disorders as dystonic, ataxic or parkinsonian syndromes and diagnosis is often delayed [64,65].
Recommendations for Surgical Treatment Several drugs are available to treat dystonodyskinetic symptoms but their efficacy is often limited, transient and difficult to assess. The most important drugs to be used are Levodopa (also as a diagnosis trial in Dopa-responsive DDS), anticholinergics, benzodiazepines, baclofen [66], Botulinum toxin injection especially for the treatment of focal DDS [67]. Surgical treatments for DDS are discussed in patients for whom pharmacotherapy failed to improved symptoms and in patients significantly disabled by the disease. Several surgical treatments have been proposed over time for treating DDS. Ablative surgeries (thalamotomy, pallidotomy) have been performed, previous to the advent of DBS substituting them more and more. Pallidotomy and thalamotomy are rarely maintained, especially in secondary DDS (e.g., hemi-DDS [68]). As ablative surgeries are now rare in the treatment of DDS, we will discuss bellow only the indications for DBS. Very little was published about guidelines for DBS in DDS. Albanese
108
et al. [69] recommended as good practice points to consider pallidal DBS particularly for generalized or cervical dystonia, after medication or when BoNT have failed to provide satisfactory improvement. While in patients with primary DDS refractory to medical treatment DBS of the internal globus pallidus (GPi) can be easily proposed, the selection of the patients with secondary and heredodegenerative DDS is a real challenge. Several variables must be considered to select appropriate patients for surgery including motor, cognitive, overall medical, and psychological status [70].
Primary DDS Independently of the extent of the symptoms, as soon as a primary DDS is refractory to pharmacological therapies and is severe enough to produce disability in daily living, DBS especially of the GPi can be discussed. The onset and the progression of the disease, molecular analysis results (DYT1 mutation), response to pharmacological treatment and MRI are required. Rutledge et al. [71] were considering that structural brain MRI is not routinely required when there is a confident diagnosis of primary dystonia in adult patients because a normal study is expected. Nevertheless, structural brain MRI under general anesthesia is suitable to be performed during the preoperative assessment for at least two reasons: to study the possible signal abnormalities within the future targets of the DBS (GPi) and/or within the other basal ganglia and the GPi volume (probably linked to the therapeutic effect as indicated bellow).
Dystonia-Plus, Secondary DDS and Heredodegenerative DDS From the group of Dystonia-plus syndromes, the myoclonus-dystonia, especially when linked to
1805
1806
108
Central procedures for primary dystonia
mutations in the e-sarcoglycane gene (DYT11), receive the same recommendations that primary DDS. The course of the disease, clinical examination and brain MRI can suggest a heredodegenerative DDS. Further investigations must be performed such as blood work-up, urine sample analysis, CSF testing, electrophysiologic studies, muscle, skin, liver biopsies, PET, eye slit-lamp examination in order to establish etiological diagnosis. The identification of the etiology is important for establishing the prognosis of the DDS and because of the existence of very few diseases having a specific treatment such as Doparesponsive DDS, creatine deficiency or Wilson’s disease. In secondary and heredodegenerative DDS, tone (dystonia) and movement disorders (dyskinesia) are often associated with pyramidal tract signs (spasticity, paresis, and increased tendon reflexes), ataxia and other neurological signs rendering the assessment of movement disorders difficult. In these patients, it is of great interest to perform motor evoked potentials (MEPs) to confirm the involvement of the pyramidal tract by the pathological process. Together with MRI, (18)F-Fluorodeoxyglucose positron emission tomography (18)F-FDG PET, TechnetiumTc-99m ethyl cysteinate dimmer brain single-photon emission CT (Tc99m-ECD SPECT) and 123Iioflupane (DaTSCAN) SPECT complete the morphological informations provided by the MRI data regarding glucose metabolism, perfusion and dopamine transporters in the brain. On MRI, several items are checked: 1. 2.
Cortical atrophy (especially of the motor cortex) and ventricular system size, Basal ganglia atrophy (including caudate, putamen, internal and external pallidus, thalamus); GPi volume is of major importance for the surgical procedure (positioning of the leads) and it seems to be linked to the DBS efficacy,
3. 4.
Basal ganglia lesions or/and necrosis, White matter changes.
(18)F-FDG PET study focus especially in metabolic disorders on the cortical, basal ganglia and thalamic metabolism. Tc99m-ECD SPECT data look after the perfusion of the same areas and dopamine transporter (DAT) SPECT data confirm central dopaminergic dysfunction. From the clinical, elecrophysiological and imaging data, several phenotypes are defined which allow or contraindicate DBS. For the clinical criteria, patients with phasic hyperkinetic movements instead of tonic postures without or with mild pyramidal tract signs are suitable for DBS. Normality of the motor evoked potentials is a complementary argument in favor of the indication for DBS. Severe cortical atrophy, white matter changes, thalamic atrophy or lesions due to the pathological process, GPi atrophy, signal abnormalities or necrosis will contraindicate DBS. On the contrary, isolated putaminal and/or external globus pallidus signal abnormalities not associated to the previously mentioned findings will not influence the GPi efficacy. In secondary and heredodegenerative DDS, the best phenotype profile to receiving DBS is as follows: focal, segmental and or generalized DDS with prominent hyperkinetic component for the movement disorder, no or mild pyramidal tract involvement (no or mild paresis, no spasticity), normal motor evoked potentials, mild sequellae on brain MRI (no or discrete cortical atrophy, no white matter changes, mild striatal signal abnormalities and/or atrophy, no or mild external pallidal abnormalities, and no or GPi and thalamic abnormalities). For the (18)F-FDG PET, normal or mild glucose hypo-metabolism are accepted and no or discrete cortical hypo-perfusion Tc99m-ECD SPECT are required to indicate DBS. It is also suitable to have a normal dopamine transporter SPECT. This type of phenotype can
Central procedures for primary dystonia
be seen in several secondary or heredodegenerative diseases: late onset pure DDS secondary to cerebral palsy, DDS associated with PKAN. In this second example, there are patients with an exclusive DDS phenotype. In this disease, GPi abnormalities located in its anterior part (eye of the tiger sign) will not contraindicate DBS surgery. The sensori-motor part of the GPi in PKAN is preserved and there is room for the DBS lead implantation. At the opposite of patients with the best phenotype for DBS described previously, we meet patients with mild DDS, prominence of dystonic postures, severe pyramidal tract signs, impaired motor evoked potentials, brain hypo-metabolism and hypo-perfusion, altered dopamine transporter SPECT study. In these patients, DBS is contraindicated. These observations highlight the need of integrity of the target and of the motor networks behind the pallidal target for obtaining the therapeutical effect. Numerous combinations between clinical, morphological, metabolic and electrophysiological patterns are seen in patients with secondary and heredodegenerative DDS. If DBS of the GPi represents an effective symptomatic treatment in primary DDS, it could also be proposed in selected cases of secondary and heredodegenerative DDS. In our experience, several patients from these categories could have a real benefit with DBS. The age is generally not a contraindication for DBS. Children [15,16] and adults up to 73 years of age have been successfully treated by DBS under general anesthesia. In our experience, in pharmacologically refractory DDS, in patients with criteria for DBS, there is no any medical reason to delay surgery. Skeleton deformities, fixed postures will impair the benefit of DBS and will limit disability reduction. Associated psychiatric illness (major depression, pathological personality, etc.) can contraindicate DBS and have been considered as criteria for surgery [72,73].
108
Despite the great variability of symptoms in severity and extent between DDS patients, it is not presently possible to statistically validate the clinical pattern carrying always a good prognosis.
Surgical Options As the pharmacological therapies are very limited, many efforts have been made to provide surgical alternatives such as peripheral denervation surgery, pump implantation for intrathecal baclofen delivery and stereotactic functional surgery. Stereotactic neurosurgery was introduced by Spiegel and Wycis in 1950 [74]. Concerning DDS, several targets have been explored over time (various locations within the thalamus, dentate nucleus, internal globus pallidus, subthalamic nucleus), and the results reported in the literature are miscellaneous. Lesioning was first performed, in the fifties and more recently DBS has been implemented. Originally, thalamotomy was the preferred surgical treatment for DDS [75–84]. However, several authors outlined the great variability in clinical outcomes and the high incidence of operative side effects. Case reports and several studies have shown an improvement in DDS after pallidotomy [81,85–93]. Nevertheless, bilateral pallidotomy may be associated with uncontrollable side effects, and the initial improvement in DDS may decrease over time. DBS provides a valuable alternative in the treatment of movement disorders. This procedure was first developed in the treatment of Parkinson’s disease, and applied to several targets (thalamus, internal pallidum, subthalamic nucleus) [8,94,95]. Based on the positive results in Parkinson’s disease, high frequency stimulation has been proposed to treat DDS [15,16,19,96– 99]. With the development of high-frequency stimulation instead of inducing lesions, surgical
1807
1808
108
Central procedures for primary dystonia
procedures have become safer, more reversible [100] and side effects easier to control.
Target Localization Methods While some centers perform simultaneous bilateral surgery, often under general anesthesia, others prefer to surgical procedures under local anesthesia, especially in adults. Target localization methods are also source of debate. With the progresses in image processing, stereotactic methodologies have improved considerably, especially in the treatment of movement disorders. A critical step in all stereotactic procedures remains the accurate target localization [101]. The publication of stereotactic atlases of the human brain allowed the development of various approaches for localizing invisible targets. Most of these protocols, like the Talairach system, used ventriculography, arteriography, and a stereotactic atlas. In the 1980s, the introduction of x-ray CTscanning to guide stereotactic procedures permitted direct visualization of various subcortical structures and lesions. Recently, MR imaging has been used increasingly for target localization in stereotactic surgery, providing improved contrast and exquisite neuroanatomical detail [102,103]. Theoretically, these features, together with the ability to acquire images in any plane (3D imaging), make MR imaging ideal for accurate stereotactic localization of targets, especially deeply located brain structures such as grey nuclei [45,95,103–106]. Using this approach, it is now possible to visualize the borders of the GPi directly [101,107,108]. This raises the possibility of direct targeting of these nuclei by considering individual structural spatial variations [109]. However, the risk of image distortion has often been discussed as a potential source of error that could result in mistargeting [104,110,111]. Given the differences between the protocols used in the MRI series of different centers, distortions are difficult to predict and to compare between centers. Consequently,
the use of MR imaging alone for the determination of stereotactic targets has been called into question and is not widely accepted. Many groups still validate the target coordinates by performing multiple microelectrode penetrations and recordings in the region of the stereotactic target. Technological and surgical advances have increased the accuracy and reliability of MR imaging–based stereotactic procedures when compared to CT scanning and ventriculography based techniques [112]. Many groups continue to use indirect targeting primarily, that is, coordinates calculated by measuring the distance between the target and an internal reference, for example the anteriorposterior commissures (AC–PC line). This line can be located with the aid of ventriculography, computerized tomography, or MR imaging. Distances are then deduced using a human brain atlas, hypothesizing that the proportions given in the atlas are applicable to any individual. It has long been known, however, that there are substantial individual variations in coordinates of subcortical nuclei that are based on the AC–PC line. Consequently, coordinates are usually reported as a range rather than as exact numbers. To compensate for this limited accuracy, several centers have developed intraoperative clinical and electrophysiological monitoring procedures that can be used with local anesthesia. That approach is uncomfortable for the patient, who is subjected to a long procedure and additional risks related to the multiple cerebral tracts necessary to obtain recordings and ventriculographic images. For the GPi there is no consensus regarding the best targeting method and very little published data documenting the accuracy of any method of localization [113,114].
Target Determination Using Atlases Targeting Based on the Schaltenbrand Atlas
The Schaltenbrand and Wahren atlas is composed of consecutive brain slices (plates) obtained from a limited number of brains. Coordinates calculat-
Central procedures for primary dystonia
ed on the basis of structures found in this atlas are not obtained from an average brain but actually from a plate (photograph) corresponding to one slice with a given thickness. The number of the brain from which the slice was extracted to be photographed, appears on each plate. Only three brains were used to compile the plates illustrating basal ganglia anatomy. These three brains were obtained from three male cadavers; in two cases the patient was 40 years old at death and in the other the patient was 51 years old. The target was first selected based on the atlas of Schaltenbrand and Wahren. Among the coronal views (called horizontal views) in this atlas, the neurosurgeon chose the plate on which he wanted to position the target for example DBS of the GPi in patients with DDS. Its projection was then verified on other lateral and frontal views. Once the target had been identified, the distances between the midpoints of the AC and PC, and this target were measured directly on the Schaltenbrand atlas (lateral distance, xSch = 20 mm; anteroposterior distance, ySch = 2 mm; and vertical distance, zSch = 4 mm). Targeting Based on the Talairach Atlas
The Talairach atlas is a proportional atlas, but dimensions expressed in millimeters are valid with precision only for a specific brain corresponding to the dimension of the displayed brain (a 60-year-old right-handed European woman).The concept of proportionality is based on an orthogonal grid system. This system is built according to the maximum dimensions of the brain in the three planes of space. This system is supposed to adapt to the dimensions of individual brains. Any measurement calculated on the basis of this proportional atlas must be normalized according to the brain dimensions of the patient to be studied. Furthermore, the whole methodology is based on the assumption that a homothetic anamorphosis between two brains is possible, a concept that has never been proved.
108
Using the Talairach and Tournoux atlas, the neurosurgeon selected an axial view in which he selected the same target already used in the Schaltenbrand atlas. The coordinates and the distances between the target and the midpoint measured on the Talairach atlas were the following: in the lateral direction, xtal = 18 mm; in the anteroposterior direction, ytal = 4 mm; and in the vertical direction, ztal = 4 mm.
Ventriculography Although in most centers 3D MR imaging is now used in stereotactic functional neurosurgery for movement disorders, targeting is yet achieved in few centers mainly by using ventriculography or CT scanning. These techniques were widely accepted and supported by a number of studies [10,109,115–118]. At the time when digital imaging was not available, ventriculography was considered to be the gold standard for invisible target localization [117] deduced from the visualization of the commissures. However, it is now possible to emphasize some potential difficulties stemming from the systematic use of ventriculography, especially in children. First, it appears that the impact of cerebrospinal fluid leakage on the position of brain structures (AC–PC or grey nuclei) must be precisely evaluated. This has not been specifically studied, but the ability to achieve submillimetric precision during stereotactic procedures will require such a validation. Second, the relevance of using an adult atlas [119] in the pediatric population must be evaluated; error could be introduced in young children because of the small skull perimeter. Third, during ventriculography an additional track through the brain is made; this, together with the use of a chemical agent, adds an additional risk of morbidity.
1809
1810
108
Central procedures for primary dystonia
Intraoperative Clinical Observation and Electrophysiology Many groups validate the target coordinates by performing multiple microelectrode penetrations and recordings in the region of the stereotactic target after administration of local anesthetic agents [18,72,120,121]. The spatial morphology of the target and the trajectory to reach it are assessed via microelectrode recordings with highimpedance microelectrodes. Direct macrostimulation via the DBS electrode is also used to check for threshold of intrinsic and extrinsic response. Intraoperative observation of the immediate clinical effect of DBS (for example, tremor suppression) is probably adequate for adult patients whose symptoms respond immediately to stimulation. However, in patients with DDS there is no on–off effect, and intraoperative electrode testing is not contributive because the effect of the procedure is delayed and can only be assessed after days to several weeks interval. Furthermore, the severity of dystono-dyskinetic movements is often incompatible with a procedure of long duration conducted after administration of local anesthetics, especially in children. Electrophysiological recording seems to be useful to compensate for individual variations. Starr, et al. [101] have calculated GPi coordinates by direct visualization of the GPi boundaries on MR images. Using electrophysiological recordings, they reported significant modifications of the initially determined coordinates in 44% of their adult patients. As the GPi nucleus volume is about 500 mm3 several possible sites for electrode positioning can be chosen within the structure boundaries. Thus, Starr and colleagues in their study chose the optimal position on the basis of subsequent electrophysiological recordings. A study published by our group [112] demonstrated that target localization based solely on anatomical analysis of the GPi is accurate and can be used for functional neurosurgery, especially when general anesthesia is required.
The properties of the pallidal cells during intraoperative recordings in DDS are still controversial [122–127] due to the variability in the type and severity of the disease and the use of anesthesia, notably the propofol. Under general anesthesia, a lower firing rate in the GPi is recorded.
Stereotactic MRI-Based Target Localization Our group has developed a stereotactic procedure based only on stereotactic MRI and described in this chapter [17]. The two electrodes are implanted under general anesthesia without intraoperative microelectrode recordings or clinical control. This technique shortens the procedure to 1 h per electrode. The internal pulse generators are implanted 5 days later also under general anesthesia. This procedure proved to be at low risk of complications (0% bleeding) and high precision [112,128].
Results: Ablative Techniques Ablative stereotactic surgery to treat DDS has been conducted in several targets as the thalamus and the GPi. However, several issues rendered them difficult to assess: the variability in inclusion criteria, the lack of standardized scales to measure outcome, the variability in surgical target, and the fact that imaging studies to assess location and size of the lesions were often not available [129].
Thalamotomy The largest experience with thalamotomy was reached by Cooper [78]. Between 1955 and 1974, he performed thalamotomies in 226 patients with generalized DDS. Cooper reported moderate to marked improvement in 69.7% of 208 patients with idiopathic DDS with a
Central procedures for primary dystonia
mean follow-up of 7.9 years after thalamotomy. The overall best results were obtained in Jewish patients with a family history of DDS (possible DYT1 DDS), and in the patients with childhoodonset of the disease. Gros [130] reported the results of thalamosubthalamotomy lesioning in 25 young patients with severe abnormal movements, including 6 patients with DDS. Concerning these 6 dystonodyskinetic patients, immediate moderate to good results were observed in 5 patients, but this immediate improvement vanished to such a point that after 1 year not a single good result remained. Andrew [76] reported marked improvement in 4 out of 16 patients with generalized DDS, and good results in 62% of the patients with focal or segmental DDS; 59% of the patients needed bilateral surgery. Detailed results of thalamotomy for primary and secondary DDS were provided by Tasker [84]. Tasker reported 25–100% improvement in 68% of 29 patients with secondary DDS (bilateral procedure in 5 patients), but the dystono-dyskinetic symptoms relapsed in one-third of these patients within 1–6 years after surgery. Twenty-five to 100% improvement was observed in 10 out of 20 patients with idiopathic DDS (bilateral procedure in ten patients), but half of the patients who benefited from the surgery regressed within months to 2 years after the surgery. Better results were obtained on appendicular versus axial DDS, and overall immediate and long-term improvement was achieved when bilateral procedure was performed. The relative improvement of DDS after thalamotomy was tarnished by significant complications, and the limiting factor of bilateral procedures is related to these disabling side effects. Dysarthria and dysphonia are observed in 11% of the patients who underwent unilateral thalamotomy, but in 56% of the patients who underwent bilateral surgery [76]. Postoperative hemiparesis is reported in 15% of the patients; postoperative limb ataxia and epilepsy have also
108
been described [76,84]. Cooper reported 0.7% mortality rate after unilateral procedure, but 2% mortality rate after bilateral thalamotomy [78]. Recently, the thalamic lesions have also been applied to treat writer’s cramp and musician’s cramp with a significant improvement in the immediate postoperative period [131–134].
Pallidotomy: Results Pallidotomy has also been performed in patients with severe DDS. Cooper observed no further benefit of pallidotomy after thalamotomy [78]. Buzaco [135] reported almost the same efficacy of pallidotomy compared to thalamotomy but less frequent complications. In 1996, Iacono described the dramatic improvement of a 17-year-old patient with idiopathic generalized DDS after bilateral pallido-ansotomy [136]. Subsequent reports confirmed the efficacy of pallidotomy in alleviating the dystono-dyskinetic symptoms in patients with primary or secondary DDS [93,106,126,137–140]. No long-lasting complication is reported in any of these reports. Overall, pallidotomy was performed mostly in adult patients, follow-up was up to 12 months, and best results were obtained in patients with primary DDS [106,139].
Neuromodulation: Results The GPi is currently the target of choice for DBS surgery in DDS [15–19,72,96–99,112,114,120,141– 146]. Thalamic stimulation is also applied (ventralis oralis anterior, ventralis oralis posterior) but reports are rare. Most recently the stimulation of the subthalamic nucleus has also been proposed [147,148].
Thalamic DBS Mundinger [149] reported satisfactory short-term results of unilateral intermittent low frequency
1811
1812
108
Central procedures for primary dystonia
thalamic DBS in seven patients with cervical DDS. The stimulation parameters included intermittent (30 min/day), low-frequency (2–12 Hz) stimulation. However, no long term results were reported. Andy [150] also found good results with intermittent low frequency thalamic stimulation in two patients with torticollis. Sellal reported in 1993, improvement of an adult patient with hemi DDS after unilateral thalamic stimulation [151]. Two other patients with familial DDS were mildly improved [9]. Vercueil et al. [152] reported on the results of high frequency ventrolateral thalamic stimulation in 12 patients with medically intractable DDS. Although the 12 patients (four with primary and eight with secondary DDS) had improved their global functional outcome, dystono-dyskinetic movement and disability scale scores did not show significant improvement. In four out of five patients treated with thalamic (Vim) DBS, poor results were obtained for secondary DDS. Furthermore, positive results were reported by the bilateral stimulation of the ventralis oralis anterior nucleus [153].
GPI DBS Pallidal stimulation is nowadays the mainstay surgical treatment for patients with DDS, particularlygeneralized DDS. In primary generalized DDS, a significant improvement of the motor and disability scores of the Burke-Fahn-Marsden Dystonia Rating Scale (BFMRS) has been reported (range, 40–80%) [15,16,72,73,142,143,152,154–156]. Similar results have been described in patients with segmental DDS [72,155,157–159]. In 1999, our group reported on the striking improvement of an 8-year-old girl with primary non-DYT1 generalized DDS after bilateral stimulation of the sensorimotor part of the GPi [15]. Furthermore, 36 months follow-up confirmed the long-term efficacy of this procedure
[142]. Kumar and Krauss [19,97] also reported the efficacy of pallidal stimulation in adult patients with generalized or severe focal DDS. One trial of 31 patients with primary generalized DDS reported by our group showed a greater clinical improvement in children compared with adults (12 adults and 19 children) at 2 years follow-up. Efficacy was evaluated by comparing scores on the clinical and functional BFMDRS before and after implantation. The efficacy of stimulation improved with time. After 2 years, compared with preoperative values, the mean (standard deviation) clinical and functional BFMDRS scores had improved by 79 19% and 65 33%, respectively. At the 2-year follow-up examination the improvement was comparable in patients with and without the DYT1 mutation in both the functional (p = 0.12) and clinical (p = 0.33) scores. Vidailhet et al. [154] also reported the clinical outcome in 22 patients with primary generalized DDS with 1 year and 3 years follow-up.The dystono-dyskinetic movement score improved from a mean of 46.3 21.3 before surgery to 21.0 14.1 at 12 months (p < 0.001). The disability score improved from 11.6 5.5 before surgery to 6.5 4.9 at 12 months (p < 0.001). Results were comparable at 3 years and no loss of efficacy has been recorded with this follow-up. In our experience, no significant difference has been recorded between the DYT1 positive and negative primary DDS. Face involvement by dystonia was more frequent in DYT1 negative patients as well as dysarthria. The impact of pallidal stimulation on dysarthria was poor. With a follow-up as long as 10 years, we could not see vanishing of the pallidal DBS therapeutical effect. Nevertheless, in several patients with DYT1 positive and negative patients, a progression of the disease under DBS has been recorded. No significant difference was recorded between the regions involved by the disease, trunk, neck, lower and upper limbs as well as
Central procedures for primary dystonia
facial muscles being all concerned by the new symptoms. As reported in the literature facial involvement is the less frequent in primary DDS. Reoccurrence of signs previously controlled by DBS can also occur. DBS can only partially be adapted, electrical settings change is not always enough to control the reoccurrence of the symptoms or the occurrence of new signs. In several patients we were led to implant a second pair of electrodes within the GPi to complete the effect of DBS. When the stimulation was discontinued, symptoms recurred within minutes to months. This could be seen during off tests for assessment of the progression of the disease, unexpected IPG arrests and at the end of life of the IPG or IPG removal because of infection. No real rebound effect has been seen at the time of the switch of the IPG. No loss of efficacy of DBS has been recorded after off tests. Pallidal stimulation has also been used in patients with dystonia-plus syndromes. In myoclonic DDS, there are reports showing that stimulation improves not only DDS, but also the myoclonic features of the disease [160,161]. Our group attempted bilateral electrical stimulation of the GPi in 4 patients with genetically proven myoclonus dystonia (DYT11) refractory to all pharmacological treatments and reported the outcome in the youngest patient who had the longest follow-up [161]. Patients with secondary DDS represent a heterogeneous population in terms of etiology, phenotype, evolution, and long term prognosis. It has been reported that secondary DDS does not respond to DBS as well as primary DDS. Nevertheless, several reports showed satisfactory clinical outcome in many secondary DDS. In the group of the secondary and heredodegenerative disorders, there is a lack of adapted tools to assess the whole phenotype and its changes under DBS. There is a discrepancy between the mild improvement expressed by the validated movement disorder
108
scales in this population and the consistent improvement assessed by quality of life scales. One posttraumatic and two hemidystonic patients were improved after unilateral pallidal DBS [98,152]. Tardive DDS have been reported to improve by 50–70% [162–165]. For postanoxic, postencephalitic, perinatal, or poststroke DDS, poor results have been published in the literature [122,166,167]. Nevertheless, this is not our experience with postanoxic cerebral palsy patients. In this group, the clinical outcome with 3 years follow-up was of 47 21.8% and of 28.1 24.6% for the disability improvement. Good initial responses to GPi stimulation have also been reported in patients with PKAN [168,169]. With longer follow-up (5 years and more), in our experience maintain of the improvement is heterogeneous among patients. In several patients no vanishing of the benefit was observed meanwhile in others, disease progression impaired the final outcome. In one patient with Lesch-Nyhan disease (LND), Taira et al. reported an improvement on the BFMDRS of 33% and 50% after 2 years and an alleviation of the self-mutilations using a single internal globus pallidus lead implant. Our group developed a surgical protocol with double bilateral simultaneous stimulation of the limbic (anteroventral) and motor (posterior) internal pallidum for controlling both behavioral and movement disorders, respectively. Several patients with LND have been treated in our center following this protocol and the results obtained in the first patient have already been reported (injurious compulsions disappeared; dystonia and dyskinesia were decreased at 28 months follow-up) [170]. In Huntington disease, bilateral implantation of human fetal striatum has given variable clinical results and is of limited access. Recently, bilateral globus pallidus stimulation has been reported to be efficient in HD with short follow-up (inferior to 1 year) [171,172]. Our group performed a bilateral chronic stimulation
1813
1814
108
Central procedures for primary dystonia
of the GPi in one patient with severe choreic movements without marked cognitive impairment. With long-term follow-up (4 years) the benefit of DBS on chorea in HD has been confirmed [173].
Gpi Stimulation for DDS The Surgical Procedure In our center, considering the poor general condition and the permanent restless situation of most patients presenting primary DDS, especially children, we developed a stereotactic procedure under general anesthesia based solely on 3D-MR imaging for target localization without intra-operative microelectrode recordings or clinical control allowing the procedure to be shortened to 1 h per electrode. Under general anesthesia, the hair is shaved transversally in a 1-cm-wide, 16 cm-long line, just in front of the coronal suture. The MR compatible stereotactic frame (Leksell G frame; Elekta Instruments, Stockholm, Sweden) is first stabilized using ear plugs (second hole) and then fixed to the patient’s head (> Figure 108-2), using constant settings of the fixation posts. This procedure aims at a reproducible, symmetrical and optimal positioning of the frame on the patient’s head to align the inferior side of the localizer as parallel as possible to the AC-PC plane in order to localize the axial-transverse plane in which the image of the target was selected. The patient is then immediately transported to the MR imaging unit. Stereotactic MR imaging was performed on a 1.5 T MR unit. One 3D-SPGR sequence is implanted without injection of gadolinium. The images are transmitted to a separated computer through an Ethernet network and the surgical planning (> Figure 108-3) is completed while the patient was prepared and draped for the operation. In the operating room, the stereotactic electrode-guiding device was installed and a
transverse curved incision is made exposing the coronal suture (> Figure 108-4). Two 14-mm burr holes are made at the level of the predetermined trajectories (> Figure 108-4). A small incision of the drug allows sufficient visual control of the cortex and a small corticotomy is performed using the bipolar coagulator. No microelectrode recording was used in these patients. The implanted electrode, with a radius of 0.635 mm, is an MR compatible quadripolar device on which the contacts are numbered from 0 to 3 (0 is the lower contact and 3 the upper contact (model: 3389, Medtronic, ReuilMalmaison, France). The upper border of contact 1 was strictly aligned with the target position and secured under continuous radioscopic control (> Figure 108-5). The temporary extension is cut and left under the galea, close to the lateral extremity of the incision, so it must easily be found during the second operation (connection to the Internal Pulse Generator). Immediately after surgery, the patient was again transported to the MR unit for a postoperative stereotactic MRI control (> Figure 108-6). The post-operative control with the stereotactic frame in place allows to check the electrode position and to measure the possible error, using the same imaging protocol as that used for the surgical planning. The second operation is performed separately on the fifth day, also under general anesthesia. The connector is found through the first incision. The 95-cm extension (Medtronic) connects the electrode with the IPG. The residual length is enrolled behind the IPG to compensate for growth in children and to provide some flexibility with movement in the system (> Figure 108-7).
Magnetic Resonance Imaging Magnetic resonance imaging acquisition is performed with a 1.5-T magnet. Control studies used to assess the homogeneity of the main magnetic
Central procedures for primary dystonia
108
. Figure 108-2 A magnetic compatible Leksell frame is placed under general anesthesia. It is first stabilized using earplugs and then fixed to the patient’s head. The procedure aims to obtain an optimal and symmetric frame placement in order to align it as parallel as possible to the AC-PC plane
field and the calibration of the gradients are obtained the day before the stereotactic operation. All these control studies are performed according to European standards. The procedure included a control with a quality assessment phantom and Quick Shim software under the responsibility of the maintenance department. The MR images are acquired with the head frame in place.
Because the reconstruction program processed images acquired on a 256 256 square matrix with a 260-mm-wide FOV, the pixel size was 1.015 mm. The slice thickness is 1.5 mm. The performed sequence consisted of a 3D fast transformed volume of contiguous transverse axial sections. The parameters used for T1-weighted images are: TE 6 ms, TR 15 ms, tilt angle 25 ,
1815
1816
108
Central procedures for primary dystonia
and two excitations. Because the 256 256 matrix produces 124 images, 1.5-mm-thick sections are chosen to cover the entire cerebral volume.
. Figure 108-3 Preoperative stereotactic MRI from a DYT1 patient
MRI Feasibility The specific absorption rate (SAR) is a measure of the rate at which radio frequency energy is absorbed by the body when exposed to radio-frequency electromagnetic field. Because of the tissue heterogeneity, direct SAR measurement is not possible. However, SAR measurement can be performed in homogeneous liquid and numerically evaluated in biological tissues. A wide range of factors can influence the SAR during MRI including RF, type of coil, concerned tissue volume, and body orientation in respect to vectors field. SAR calculation varies between different models of MR systems, algorithms of calculation, coil design and manufacturers [174]. Furthermore, the mathematical models and algorithms of calculation are not available or published because of industry property. The potential effects of MRI on DBS system are multiple. The RF field, the time-varying magnetic gradient field and the static magnetic field may induce a heating, an alteration in rate or a neurostimulator reprogramming or reset. It has not been possible, to date, to perform an accurate calculation of the heating during MRI.
. Figure 108-4 Installation of the stereotactic electrode-guiding device. Two 14-mm burr holes are made at the level of the predetermined trajectories
Central procedures for primary dystonia
108
. Figure 108-5 Intra-operative strict profile radioscopic monitoring of the electrode position (y, z). The canula does not enter the volume to be stimulated
. Figure 108-6 Postoperative stereotactic MRI from a DYT1 patient
Electrode Heating During MRI The most important safety issue is the electrode heating during MRI. At the frequencies of interest in MRI, some of the RF energy is absorbed and converted to heat. During a MRI,
the power deposited by RF pulses depends on a wide range of factors including the power and the duration of the RF, the type of the RF transmitter coil used, the volume of tissue imaged, and the electrical resistivity of the tissues. During MRI on patients with a DBS system implanted, the magnetic field gradients and the radiofrequency field could be the origin of an induced voltage. This induced voltage produces a current which flows in a conductor involving a local heating. Several factors may influence the degree of heating such as the geometrical structure of the electrodes, the placement in the brain or the insulation of the electrodes. A temperature elevation of more than 5 C causes reversible lesions of the neurons [175]. A temperature of more than 50 C causes irreversible lesions around the electrodes similar to a pallidotomy (80 during 1 min). Several evaluations of MRI-related heating for neurostimulation systems used for DBS have been published [174,176–181]. Significant temperature elevations were found in furthest conditions in contradiction with manufacturer’s advice which cannot be applied in clinical practice.
1817
1818
108
Central procedures for primary dystonia
. Figure 108-7 The excess length is wrapped around the skull burr hole with a mean value of 2 loops per electrode. The excess length is wrapped around IPG with a mean value of 2 loops
Concentric loops placed around the burr hole seem to reduce MRI-related heating of electrodes [178]. Rezai et al. [180] showed that the temperature elevation was dependent on the type of radio frequency coil, level of SAR used and how the electrodes were positioned. The most important temperature elevation found in this study was of 25.3 C with the following configuration: (1) 4 loops around the neurostimulators; (2) 0 loops around the burr hole; (3) a SAR value of 3.9 W/Kg with a transmit/receive body coil. Finelli et al. [179] reported that MRI-related heating for clinical pulse sequences (including fast spin-echo, gradient-echo and echo-planar imaging) using a transmit/receive RF head coil was correlated linearly with local SAR values for single and multislice fast spin-echo images. The authors indicate that the sequences performed at local SARs below 2.4 W/kg (whole-body averaged SAR of 0.09 W/kg) should be safe from a MRI-related heating point of view. The temperature elevation is of approximately 0.9 times the local SAR value. The neurostimulation systems were programmed to the boff Q mode (i.e., no stimulation was delivered) and set to 0 V according to
the manufacturer’s recommendation for patients with this device undergoing an MRI procedure (Medtronic). Baker et al. [174] studied the temperature change at the electrode contacts using two different transmit/receive body coils on two different generation 1.5-T MR systems from the same manufacturer. The result of the study revealed marked differences across two MR systems at the level of radiofrequency-induced temperature changes per unit of whole body SAR for a conductive implant. These data suggest that the temperature elevations were clinically insignificant using clinical sequences for brain imaging and that using SAR to guide MR safety recommendations for neurostimulation systems across different MR systems is unreliable. The calculation or estimation of the SAR values can vary on the basis of the model type and software of the MRI system. This is an issue of major interest because safety information identified to prevent heating for an implant, determined for a given MRI, may not be applied to another MRI, even if it is from the same manufacturer [174].
Central procedures for primary dystonia
Clinical Data Confirm MRI Feasibility Spiegel et al. [182] reported that a 73-year-old patient with bilateral implanted DBS electrodes for Parkinson’s disease showed dystonic and partially ballistic movements of the left leg immediately after undergoing a MRI of the head. This scan was performed with a transmit/receive head coil on a 1.0-T MRI system with the leads externalized and not connected to pulse generators. An important electrode heating and brain damage induced by MRI was published by Henderson et al. [183]. Henderson et al. [183] reported a severe neurological deficit (hemiplegia and aphasia) following electrodes heating during spinal MRI in a patient with a (DBS) system implanted for Parkinson’s disease. The operation mode of the neurostimulation system during the MRI examination was unknown. Multiple scan sequences were performed with a 1.0-T MRI system and a transmit/receive body radiofrequency coil. In comparison to our protocol, we can note the use of a 1.0-T MRI instead of a 1.5-T MRI. The neurostimulators must be switch-off with 0 V during MRI which is not specified in this publication. Several studies [178,180] demonstrated that temperature elevations were inversely proportional to the number of loops wrapped around the skull burr hole. Therefore, the implantation of neurostimulators in the abdominal area with an extension of 66 cm for adult patient (as it is the case in this article) doesn’t allow these loops. Between November 1996 and November 2005, 226 MRI were performed on 161 patients implanted with a deep brain stimulation system. The clinical results obtained during the last ten years confirm the innocuousness of the method used. After the analysis of all the postoperative MRI, no image compatible with a pallidotomy was found. No hyper intense T2 weighted images and no hypo intense T1 weighted images around the artifact of the electrode were found. No case of hemorrhages in our patients was found. No electrode displacements or fractures due to MRI
108
were found. 63 MRI were performed with all the DBS system implanted and no images comparable to a thermal lesion were found. We didn’t record any supplementary neurological deficit (transitory or permanent) in our series of patients.
Target and Trajectory Determination for 3D-MR Imaging The neurosurgeon selected the target by using a 3D cursor following visual recognition of the GPi boundaries on transverse axial sections. Coordinates calculation are performed without reference to an atlas or to the AC-PC line. The final target is chosen on an axial slice located 3 mm above the optical tract in the middle of the GPi. The coordinates are then automatically calculated using dedicated software. The software allowed simultaneous visualization of the three orthogonal sections and the position of the cursor in all directions. The software also added the trace of a vector linked to that point defined the trajectory. The best electrode trajectory was selected upon varying planes and orientations. The corresponding line was then displayed point by point in the cerebral volume. Importantly, it appeared crucial to display slices orthogonally to the electrode trajectory in order to precisely anticipate the relative position of each contact with the surrounding GPi boundaries.
Distortion Control at the Periphery of the Field of View (FOV) The frame and localizer were designed to closely fit the head coil. In this way the fiducial markers were visible at the periphery of the FOV. Using the G-frame localizer as a phantom, we checked that 3D MR imaging of the localizer was consistent with the actual dimensions (120—190 mm) and shape (square). With the stereotactic software ‘‘Measurement Distance’’ function, we calculated the mean distance between fiducial
1819
1820
108
Central procedures for primary dystonia
markers, before and after surgery. These measurements were performed on a routine basis on the axial transverse slice on which the target was to be chosen. The difference between measured and theoretical distance (dt–dm) was designated as error. No significant error due to image distortion at the periphery of the field was observed when this difference was equal to 0.
Distortion Control at the Center of the FOV Anterior (AC) and posterior (PC) white commissures are commonly used as essential landmarks in stereotactic surgery because of their mesial, deep location as well as their fixed position in the brain. In order to calculate the error due to distortion in the center of the field of view, the coordinates of AC and PC, as well as the AC-PC distance (distance between the corresponding midpoints) were calculated, pre- and post-operatively, for each patient. Both structures could be clearly identified on 3D MR images.
Control of the Electrode Position Immediately after surgery the patient was transported to the MR imaging unit, where a control MR image was obtained with the stereotactic frame in place (> Figure 108-6). The coordinates of the final position of the electrode (P2[x2, y2, z2]) upper border of contact 1 were calculated and compared with those of the preoperative selected target (P1[x1, y1, z1]). There was no statistical difference between pre- and postoperative coordinates when considered as distinct sets, in both the x and y directions, and on both the left and right sides (p < 0.05). This indicated that the electrode was positioned in the selected target (> Figure 108-3) with a 1 pixel accuracy.
Electrical Settings The pulse generator can be programmed by telemetry for electrical stimulation settings, contact (cathode or anode), voltage (0–10.5 V), rate (2–185 Hz), pulse width (60–450 ms), and timing (cyclic or continuous stimulation). Two kinds of current flow can be used in the system: unipolar or bipolar. In a unipolar mode, one or more contacts on the electrode are activated to serve as the negative pole. The IPG serves as the positive pole. Therefore, current flows between the electrode (negative pole) and the IPG (positive pole) through the body tissue. The bipolar mode is a system in which the current flows between two or more contacts of the electrode, where the electrode has both positive and negative poles. Stimulation effectiveness is greater near the negative pole than the positive one. The pulse width is a measure, in microseconds, of the duration of each stimulating pulse. The amplitude is a measure of the electrical intensity delivered in a stimulating pulse, measured in volts. And finally, the rate is a measure, in pulse per second, which provides the number of stimulating pulses delivered each second. The higher is the frequency, closer are two stimulation pulses. In our center we use 130 Hz high frequency stimulation continuous stimulation. The frequency is a relevant parameter for the clinical improvement. It is known that high frequencies are required to obtain beneficial effects on motor function [9]. Several articles suggest [87] that low frequency (5 Hz) electrical stimulation of thalamus can exacerbate tremor and more recently applied in the subthalamic nucleus can worsen akinesia [184]. The neural mechanism of this frequencydependent phenomenon is still on debate. The physician will select the electrical settings after the implantation of the whole DBS system and will change them according to the clinical evolution and expected outcome. In our center, for the initial protocol, the following settings were implemented: contact 1 as cathode,
108
Central procedures for primary dystonia
stimulator as anode, rate: 130 Hz, pulse width: 450 ms, voltage: 0.8 V continuous bilateral stimulation. The settings were then adapted by increasing progressively the voltage according to the clinical evolution of each patient to reach a mean value of 1.6 V. After a follow-up of at least six weeks, the volume to be stimulated can be increased by the activation of one additional contact (usually contact 2). Rarely a third contact can be activated if the final outcome with at least 1 year follow-up is not satisfactory. Pharmacological treatment should be maintained unchanged at the initial phase of stimulation in order to not interfere with the assessment of the DBS efficacy on DDS. Several pulse width values have been shown clinical effect but the best value is still mater of debate. Recently, Vercueil et al. performed a study comparing the efficacy of three ranges of pulse width in the treatment of primary DDS [185]. In our center 450 ms pulse width is commonly used. Shorter pulse widths could reduce charge injection and increase the therapeutic window between therapeutic effects and side effects [186]. A double blind study in five primary DDS patients improved at more than 80% was performed in our center in order to compare two different pulse width values. We switched the pulse width from 450 to 210 ms in one side and increased the voltage in order to obtain comparable current drains. A blinded was carried out between 210 and 450 ms: 1. 2.
Each day during five days At least every 2 months after change of the parameter settings during 6 months.
Impedance and the current drain values have been monitored. The results are shown in > Table 108-1. There was no significant change in clinical outcome with chronic assessment confirming the results reported by Vercueil et al. when changing pulse widths during 10 h. The pulse width decrease has been compensated by the voltage increase. The impedance is the resistance measurement of the circuit in Ohm. Different impedances, which are theoretically part of an implanted stimulation system, are involved. However, some of these impedances are much larger than others. In general one may assume that the resistance of the extension cable plus the resistance of the tissues can be neglected compared to all resistances or impedances involved. However, the passage of the current between the electrode boundary and the tissues cells is more difficult. The electron must cross an interface between two different materials and this process is called polarization phenomenon. The positive particles present in the physiological liquids will build-up around the electrode attracted by the negative electron and slow down the passage of the electron to the tissues.
Clinical Examination Dyskinetic movements and abnormal postures were assessed using the two sections of the BFMDRS [187]. The motor part is used to quantify the DDS by reference to nine different body areas by rating both the triggering factor of the
. Table 108-1 Pulse width 450 ms versus 210 ms: results BFMDRS scores evaluation Patient
E3 E2 E1 E0
U [V] 450/210
I [mA] 450/210
BMFDRS/120 450/210
BMFDRS/30 450/210
1 2 3 4 5
0--0 0--0 00-0 --00 00-0
1.0/2.0 1.5/2.5 1.6/2.2 1.7/3.2 1.3/2.9
118/134 93/96 54/53 100/100 47/73
5/5 2/2.5 3/2 0/0 1.5/0
2/2 2/2 1/1 0/0 1/0
1821
1822
108
Central procedures for primary dystonia
dystonic movement and its extent and severity on a scale of 0–120. The disability part is used to quantify the patient’s abilities with regard to activities of daily living and reflects the quality of life on a scale of 0–30. This scale is not so time consuming and is adapted for the assessment and quantification of the hyperkinetic, mobile component of the primary syndromes and dystonic postures. However, the scale is poorly adapted to evaluate secondary DDS and degenerative diseases. The disability part is not appropriate to child assessment before age seven. Another scale used in the assessment of movement disorders is the Unified Myoclonus Rating Scale which permits a precise quantification of the myoclonus but is time consuming. Others scales exist (e.g., for chorea: Unified Huntington’s Disease Rating Scale, for torticollis: Toronto Western Spasmodic Torticollis Rating Scale, for Essential Tremor, Asworth Scale for spasticity, etc.) but there is not one single ideal tool for the global assessment of dystonic and dyskinetic syndromes with associated neurological impairments.
Patients are examined by three physicians (L.C., B.B. and P.C.) who arrive to a consensus opinion. Clinical assessments with video recordings following a standardized video protocol (> Table 108-2) were performed just before implantation of the electrodes, daily during the postoperative hospital stay, monthly during the first year, and at intervals of 3 months thereafter. The pre- and postoperative absolute scores of the BFMDRS were assigned. Benefits of the procedure were calculated as follows: ((preoperative score postoperative score)/preoperative score 100) and are reported in the > Table 108-3 and > 108-4.
Surgical Experience Between November 1996 and October 2006, 118 patients (59 male, 59 female) underwent bilateral electrode implantation for continuous stimulation of the GPi. Mean age at surgery was of 23.7 15.8 years (range, 5–73 years). For
. Table 108-2 Clinical assessments with video recordings following a standardized video protocol Face Open/close the eyes several times Open the mouth Stick out the tongue Read twenty words loudly Lying on the back with the arms beside the trunk At rest With flexed knees alternatively With both knees flexed Passing from lying on the back position to sitting Sitting Finger to nose movement with each arm Finger to nose movement with the two arms alternatively Writing or at least holding the pen Patient standing-up or standing Hold out the arms and spread the fingers With flexed elbows Perform alternative movements of the hands With the arms beside the trunk Patient walking Patient running
Central procedures for primary dystonia
108
. Table 108-3 Clinical scores of the BFMDRS <1
1
2
>3
73.81 39.1% 70.96 25.44% 41.7 26.7% 37.2 41.7% 95.1 7% 55.5 16.1% 58.7 41.8% 27.2 25.4% 36.5 6% 92.2 11.1% 62.6 36.7%
74.3 37.1% 72.3 24.6% 41 16.6% 292 30.4% 84.3 22.5% 11.4 13.9% 63.2 18.8% 20.8 15.5% 19 18.8% 75% (n = 1) 84.1 12%
72.1 31% 76.9 24.5% 46.7 20.1% 52.2% 86.3 8.9% 9.8 22.9% 59.8 24.5% 25.6 26.1% 23.8 12% 100% (n = 1) 87.1 5.3%
73.4 28.8% 75.97 26.7% 47 21.8% 52.2% (n = 1) – 39.5% 48.8 28.3% 39.6% (n = 1) 15.3% (n = 1) 100% (n = 1) 91.4 8.3%
<1
1
2
>3
64.3 41.9% 51.3 32.3% 15.4 24.7% 43.4 17.2% 69.4 52.9% 16.2 0.6% 42.7 30.8% 14.7 6.2% 1.9 2.7% 68.4 33.8% 48.2 26.4%
65.9 40.4% 55 25.3% 24.4 22% 40.3 21.6% 63.9 62.5% 18.4 4.9% 31.5 22.8% 10.5 11.5% 3.8 5.4% 100% (n = 1) 51.4 31.6%
76 23.1% 61.3 33.5% 25.9 18.9% 25% (n = 1) 100% (n = 1) 13.8 16.2% 32 29.2% 11.9 10.6% 7.9 5.2% 100% (n = 1) 71.8 7.3%
72.2 35% 56.2 40.4% 28.1 24.6% 25% (n = 1) – 47.3% (n = 1) 45.2 21.1% 12.5% (n = 1) 11.5% (n = 1) – 81.2 17.1%
Follow-up (years) Group 1A Group 1B Group 2A Group 2B Group 2C Group 2D Group 3A Group 3B Group 3C Group 4A Group 4B
. Table 108-4 Disability scores of the BFMDRS Follow-up (years) Group 1A Group 1B Group 2A Group 2B Group 2C Group 2D Group 3A Group 3B Group 3C Group 4A Group 4B
precise DBS efficacy was assessed by the motor and disability sections of the BMFDRS [187].
Classification The current classification for etiology divides the DDS in four major categories: primary DDS, dystonia plus, secondary DDS and heredegenerative DDS. According to the etiological classification, we divided our DDS population who underwent DBS into several groups and subgroups. The first group, Group 1 is divided into two subgroups, group 1A of idiopathic DDS with the DYT1
mutation present and group 1B of idiopathic DDS without DYT1 mutation. Group 1A (patients with the DYT1 mutation), comprised 17 children and 10 adults, the mean age at surgery was of 21.1 15.2 years (range, 8–66 years) and the mean follow-up was of 50 25.5 months (range, 6–120 months). Group 1B (patients without identified mutation) included 7 children and 20 adults; mean age at surgery was of 29.5 18.5 years (range, 5–68 years) and mean follow-up of 47.6 30.8 months (range, 12–96 months). The second group, Group 2, of secondary DDS included 31 patients with DDS secondary to anoxic (> Figure 108-8) brain injury (Group 2A), 2 strokes (Group 2B), 3 patients with tardive
1823
1824
108
Central procedures for primary dystonia
. Figure 108-8 Postoperative MRI of a patient with DDS secondary to anoxic brain injury
DDS (Group 2C), and 8 other DDS (Group 2D). In this group 18 children and 26 adults have been included, mean age at surgery was of 25.7 16.7 years (range, 7–73 years) and the mean follow-up was of 23.9 19.5 months (range, 6–72 months). The third group, Group 3, of heredodegenerative DDS included six PKAN patients (Group 3A), four mitochondrial diseases (Group 3B), and two Lesch-Nyhan (> Figure 108-9) patients (Group 3C). In this group, nine children and three adults have been operated, mean age at surgery was of 15.6 9.5 years (range, 8–39 years) and the mean follow-up was of 35.5 23 months (range, 6–84 months). The last group, Group 4, of dystonia plus includes three DYT11 myoclonus dystonia patients with mutation present in the e-sarcoglycane gene (Group 4A) and five patients without identified mutation (Group 4B). In this group, four children
. Figure 108-9 Postoperative MRI of a Lesch-Nyhan patient with double bilateral simultaneous stimulation of the limbic (antero-ventral) and motor (posterior) internal pallidum for controlling both behavioral and movement disorders, respectively
and four adults have been operated, mean age at surgery was of 23.5 15.2 years (range, 6–52 years) and the mean follow-up was of 31.1 23.4 months (range, 3–60 months). At the time of surgery, all patients were severely disabled in performing daily activities. All patients were under pharmaceutical treatment with various medications (benzodiazepine, anticholinergic drugs, L-dopa). All patients or their guardians gave written informed consent. The protocol was approved by the French National Ethical Committee (reference number 98.07.02).
Clinical Results The evolution of the clinical score of the BurkeMarsden-Fahn Dystonia Rating Scale (BMFDRS) is reported for the 109 patients separated in previously described groups. In each group,
Central procedures for primary dystonia
the improvement is significant (p < 0.05) (> Tables 108-3 and > 108-4). Despite the high cost and the time consuming clinical management, bilateral chronic electrical stimulation can be proposed as first line treatment for early onset primary generalized DDS when pharmacologically intractable. It is conservative, partially adaptable and well tolerated by the whole population. It must be applied soon, especially in primary DDS before skeleton deformities and life threatening complications occur. The complication rate remains acceptable. For secondary DDS, pallidal stimulation can partially improve dystonic and dyskinetic syndrome with positive impact on painful spasms and swallowing difficulties.
Evolution of Brain Impedance in Dystono-Dyskinetic Patients Treated by GPI Electrical Stimulation Optimal stimulation parameters vary between patients. Our group published a study in order to investigate the influence of electrical brain impedance and delivered current on the brain response to stimulation [188]. Twenty-four patients were included in this study. The patients were bilaterally implanted in the globus pallidus internus through two implanted four contact electrodes. The variation of brain impedance and current measurements was correlated with stimulation parameters, time course, and clinical outcome. When a contact was activated, a statistically significant and reversible decrease of brain impedance was found. Impedance and current values and their variations with time significantly differed between patients. The absolute impedance did not significantly correlate with the final outcome. We conclude that the reversible decrease of impedance reflects an adaptive long-term mechanism, which could be due to a plasticity phenomenon, but has no prognostic value. Impedance and current measurements give new complementary
108
information for parameter adjustment and trouble shooting and should therefore be included in all patients’ follow-up.
Prognostic Value of GPI Volume in Primary DDS Treated by DBS As the improvement is variable from one patient to another, we analyzed the impact of GPi volume on the result of DBS by comparing highly and less improved patients with primary DDS. A three-dimensional model was developed in order to visualize and quantify the relationship between the isofield lines (ISF) generated by the lead used in DBS and the targeted GPi in the stereotactic space (> Figure 108-10). The model was applied to 30 right-handed selected patients with primary DDS who had been treated by bilateral stimulation of the sensory-motor GPi. Ten healthy controls were also included in the study. In twenty patients, improvement was rated above 90% (97 4.6%), and 10 under 60% (56.9 6%) on the BFMDRS motor score. Firstly, we compared the GPi volumes between patients and healthy controls. Secondly, the stimulated GPi volumes, i.e., the intersection between the volume of each ISF value and the GPi volumes, were compared between less and highly improved patients. The results of the study showed that the mean volume of the right (461.8 81.8 mm3) and left (406.6 113.2 mm3) GPi of patients less improved by DBS was significantly smaller than the GPi of patients highly improved (right: 539.9 86.6 mm3, left: 510.6 88.7 mm3) and healthy controls (right: 557.8 109.1 mm3, left: 525.1 40.8 mm3). For the groups who underwent DBS, on the left side, the mean stimulated volumes, from the ISF value 0.2–1 V/mm, were significantly larger in patients highly improved than patients less improved i.e., we calculated a threshold value of the electric field necessary for getting the functional effect at 0.2 V/mm (> Figure 108-8).
1825
1826
108
Central procedures for primary dystonia
. Figure 108-10 Correlation of the anatomical and electrical data in an illustrative case highly improved by DBS. The isofield lines (from 0.025 to 1 V/mm) generated by the electrode during high frequency stimulation of the GPi were represented at the top. The ISF 0.2 V/mm within the left GPi of the patient was represented at the bottom. For a given position and for the electrode we use (3389, Medtronic, France), the increase of the voltage does not compensate for the position. The mean stimulated volume by the ISF line 0.2 V/mm is 37.4 14.8 mm3 for Group I on the left side. On the left side, the volume stimulated by the ISF value from 0.2 to 1 V/mm was significantly (p < 0.03) larger in highly than in less improved patients
a given patients. We previously reported on the GPi functional structure organization and found a more significant somatotopic partition of the motor part on the right side whilst the left (dominant) side did not show a similar organization [189]. The results reported here add supplementary arguments in favor of an asymmetry in the functional anatomy of the GPi. The volume of the left GPi could influence the prognosis after stimulation. In patients with maximum benefits, we found that the optimal volume necessary for obtaining the effect did not cover the whole motor GPi. Furthermore it seems that for a given position and for the electrode we use (3389, Medtronic, France), the increase of the voltage does not compensate for the position. This means that the therapy is only partially adaptable and that the surgical accuracy for electrode positioning is crucial for obtaining the maximum efficacy of the treatment. A suboptimal or wrong targeting and positioning will not be constantly compensated for adapting the electrical settings.
References
The GPi volume, shape, extension and perhaps neuron distribution and orientation can explain the varying benefits of DBS. As the limbic-striatal-pallidal-thalamic in psychiatric disorder, the motor-striatal-pallidal-thalamic circuit appears to play an important role in DDS. Increasing evidence suggests dysfunction in various elements of these circuits, even though the precise mechanism remains unclear. Therefore, the stimulated volume is an individual factor i.e., the volume of tissue that needs to be stimulated depends on the conditions for
1. Marsden CD, Harrison MJ, Bundey S. Natural history of idiopathic torsion dystonia. Adv Neurol 1976;14:177-87. 2. Parent A, Cicchetti F. The current model of basal ganglia organization under scrutiny. Mov Disord 1998;13(2): 199-202. 3. Fahn S. Concept and classification of dystonia. Adv Neurol 1988;50:1-8. 4. Fahn S. Generalized dystonia: concept and treatment. Clin Neuropharmacol 1986;9 Suppl 2:S37-S48. 5. Fahn S, Bressman SB, Marsden CD. Classification of dystonia. Adv Neurol 1998;78:1-10. 6. Fahn S, Eldridge R. Definition of dystonia and classification of the dystonic states. Adv Neurol 1976;14:1-5. 7. de Carvalho Aguiar PM, Ozelius LJ. Classification and genetics of dystonia. Lancet Neurol 2002 Sep;1(5): 316-25. 8. Benabid AL, Pollak P, Gervason C, Hoffmann D, Gao DM, Hommel M, et al. Long-term suppression of tremor by chronic stimulation of the ventral intermediate thalamic nucleus. Lancet 1991 16;337(8738):403-6.
Central procedures for primary dystonia
9. Benabid AL, Pollak P, Gao D, Hoffmann D, Limousin P, Gay E, et al. Chronic electrical stimulation of the ventralis intermedius nucleus of the thalamus as a treatment of movement disorders. J Neurosurg 1996;84(2):203-14. 10. Laitinen LV. CT-guided ablative stereotaxis without ventriculography. Appl Neurophysiol 1985;48(1–6):18-21. 11. Laitinen LV, Bergenheim AT, Hariz MI. Leksell’s posteroventral pallidotomy in the treatment of Parkinson’s disease. J Neurosurg 1992;76(1):53-61. 12. Benabid AL, Pollak P, Louveau A, Henry S, de Rougemont J. Combined (thalamotomy and stimulation) stereotactic surgery of the VIM thalamic nucleus for bilateral Parkinson disease. Appl Neurophysiol 1987;50(1–6):344-6. 13. Benabid AL, Benazzouz A, Hoffmann D, Limousin P, Krack P, Pollak P. Long-term electrical inhibition of deep brain targets in movement disorders. Mov Disord 1998;13 Suppl 3:119-25. 14. Limousin P, Pollak P, Benazzouz A, Hoffmann D, Broussolle E, Perret JE, et al. Bilateral subthalamic nucleus stimulation for severe Parkinson’s disease. Mov Disord 1995;10(5):672-4. 15. Coubes P, Echenne B, Roubertie A, Vayssiere N, Tuffery S, Humbertclaude V, et al. Treatment of early-onset generalized dystonia by chronic bilateral stimulation of the internal globus pallidus. Apropos of a case. Neurochirurgie 1999;45(2):139-44. 16. Coubes P, Roubertie A, Vayssiere N, Hemm S, Echenne B. Treatment of DYT1-generalised dystonia by stimulation of the internal globus pallidus. Lancet 2000 24;355(9222):2220-1. 17. Coubes P, Vayssiere N, El Fertit H, Hemm S, Cif L, Kienlen J, et al. Deep brain stimulation for dystonia. Surgical technique. Stereotact Funct Neurosurg. 2002;78(3–4):183-91. 18. Krauss JK, Loher TJ, Pohle T, Weber S, Taub E, Barlocher CB, et al. Pallidal deep brain stimulation in patients with cervical dystonia and severe cervical dyskinesias with cervical myelopathy. J Neurol Neurosurg Psychiatry 2002;72(2):249-56. 19. Krauss JK, Pohle T, Weber S, Ozdoba C, Burgunder JM. Bilateral stimulation of globus pallidus internus for treatment of cervical dystonia. Lancet 1999 4;354 (9181):837-8. 20. Bressman SB, de Leon D, Brin MF, Risch N, Burke RE, Greene PE, et al. Idiopathic dystonia among Ashkenazi Jews: evidence for autosomal dominant inheritance. Ann Neurol 1989;26(5):612-20. 21. Bressman SB, Sabatti C, Raymond D, de Leon D, Klein C, Kramer PL, et al. The DYT1 phenotype and guidelines for diagnostic testing. Neurology 2000;54(9):1746-52. 22. Ozelius LJ, Hewett JW, Page CE, Bressman SB, Kramer PL, Shalish C, et al. The early-onset torsion dystonia gene (DYT1) encodes an ATP-binding protein. Nat Genet 1997;17(1):40-8.
108
23. Bressman SB. Dystonia genotypes, phenotypes, and classification. Adv Neurol 2004;94:101-7. 24. Ozelius L, Kramer PL, Moskowitz CB, Kwiatkowski DJ, Brin MF, Bressman SB, et al. Human gene for torsion dystonia located on chromosome 9q32-q34. Neuron 1989;2(5):1427-34. 25. Shashidharan P, Paris N, Sandu D, Karthikeyan L, McNaught KS, Walker RH, et al. Overexpression of torsinA in PC12 cells protects against toxicity. J Neurochem 2004;88(4):1019-25. 26. Ahmad F, Davis MB, Waddy HM, Oley CA, Marsden CD, Harding AE. Evidence for locus heterogeneity in autosomal dominant torsion dystonia. Genomics 1993;15(1):9-12. 27. Lossos A, Cohen O, Meiner V, Blumenfeld A, Reches A. Intrafamilial heterogeneity of movement disorders: report of three cases in one family. J Neurol 1997;244 (7):426-30. 28. Bressman SB, Hunt AL, Heiman GA, Brin MF, Burke RE, Fahn S, et al. Exclusion of the DYT1 locus in a non-Jewish family with early-onset dystonia. Mov Disord 1994;9(6):626-32. 29. Holmgren G, Ozelius L, Forsgren L, Almay BG, Holmberg M, Kramer P, et al. Adult onset idiopathic torsion dystonia is excluded from the DYT 1 region (9q34) in a Swedish family. J Neurol Neurosurg Psychiatry 1995;59(2):178-81. 30. Bentivoglio AR, Del Grosso N, Albanese A, Cassetta E, Tonali P, Frontali M. Non-DYT1 dystonia in a large Italian family. J Neurol Neurosurg Psychiatry 1997;62 (4):357-60. 31. Valente EM, Bentivoglio AR, Cassetta E, Dixon PH, Davis MB, Ferraris A, et al. DYT13, a novel primary torsion dystonia locus, maps to chromosome 1p36.13– 36.32 in an Italian family with cranial-cervical or upper limb onset. Ann Neurol 2001;49(3):362-6. 32. Almasy L, Bressman SB, Raymond D, Kramer PL, Greene PE, Heiman GA, et al. Idiopathic torsion dystonia linked to chromosome 8 in two Mennonite families. Ann Neurol 1997;42(4):670-3. 33. Leube B, Rudnicki D, Ratzlaff T, Kessler KR, Benecke R, Auburger G. Idiopathic torsion dystonia: assignment of a gene to chromosome 18p in a German family with adult onset, autosomal dominant inheritance and purely focal distribution. Hum Mol Genet 1996;5(10): 1673-7. 34. Klein C, Ozelius LJ, Hagenah J, Breakefield XO, Risch NJ, Vieregge P. Search for a founder mutation in idiopathic focal dystonia from Northern Germany. Am J Hum Genet 1998;63(6):1777-82. 35. Tezzon F, Zanoni T, Passarin MG, Ferrari G. Dystonia in a patient with deletion of 18p. Ital J Neurol Sci 1998;19(2):90-3. 36. Misbahuddin A, Placzek MR, Chaudhuri KR, Wood NW, Bhatia KP, Warner TT. A polymorphism in the
1827
1828
108 37. 38.
39.
40.
41.
42.
43.
44.
45. 46.
47.
48.
49.
50.
51.
52.
Central procedures for primary dystonia
dopamine receptor DRD5 is associated with blepharospasm. Neurology 2002;58(1):124-6. Parker N. Hereditary whispering dysphonia. J Neurol Neurosurg Psychiatry 1985;48(3):218-24. Nygaard TG, Waran SP, Levine RA, Naini AB, Chutorian AM. Dopa-responsive dystonia simulating cerebral palsy. Pediatr Neurol 1994;11(3):236-40. Steinberger D, Topka H, Fischer D, Muller U. GCH1 mutation in a patient with adult-onset oromandibular dystonia. Neurology 1999;52(4):877-9. Nygaard TG, Marsden CD, Fahn S. Dopa-responsive dystonia: long-term treatment response and prognosis. Neurology 1991;41(2 (Pt 1)):174-81. Dobyns WB, Ozelius LJ, Kramer PL, Brashear A, Farlow MR, Perry TR, et al. Rapid-onset dystoniaparkinsonism. Neurology 1993;43(12):2596-602. Brashear A, DeLeon D, Bressman SB, Thyagarajan D, Farlow MR, Dobyns WB. Rapid-onset dystoniaparkinsonism in a second family. Neurology 1997;48(4): 1066-9. Webb DW, Broderick A, Brashear A, Dobyns WB. Rapid onset dystonia-parkinsonism in a 14-year-old girl. Eur J Paediatr Neurol 1999;3(4):171-3. Segawa M, Hosaka A, Miyagawa F, Nomura Y, Imai H. Hereditary progressive dystonia with marked diurnal fluctuation. Adv Neurol 1976;14:215-33. Gasser T. Inherited myoclonus-dystonia syndrome. Adv Neurol 1998;78:325-34. Quinn NP, Rothwell JC, Thompson PD, Marsden CD. Hereditary myoclonic dystonia, hereditary torsion dystonia and hereditary essential myoclonus: an area of confusion. Adv Neurol 1988;50:391-401. Furukawa Y, Rajput AH. Inherited myoclonus-dystonia: how many causative genes and clinical phenotypes? Neurology 2002;59(8):1130-1. Asmus F, Zimprich A, Naumann M, Berg D, Bertram M, Ceballos-Baumann A, et al. Inherited Myoclonusdystonia syndrome: narrowing the 7q21-q31 locus in German families. Ann Neurol 2001;49(1):121-4. Asmus F, Zimprich A, Tezenas Du Montcel S, Kabus C, Deuschl G, Kupsch A, et al. Myoclonus-dystonia syndrome: epsilon-sarcoglycan mutations and phenotype. Ann Neurol 2002;52(4):489-92. Doheny DO, Brin MF, Morrison CE, Smith CJ, Walker RH, Abbasi S, et al. Phenotypic features of myoclonus-dystonia in three kindreds. Neurology 2002 22;59(8):1187-96. Klein C, Liu L, Doheny D, Kock N, Muller B, de Carvalho Aguiar P, et al. Epsilon-sarcoglycan mutations found in combination with other dystonia gene mutations. Ann Neurol 2002;52(5):675-9. Muller B, Hedrich K, Kock N, Dragasevic N, Svetel M, Garrels J, et al. Evidence that paternal expression of the epsilon-sarcoglycan gene accounts for reduced penetrance
53.
54.
55. 56.
57.
58.
59. 60.
61.
62.
63.
64.
65.
66.
67.
in myoclonus-dystonia. Am J Hum Genet 2002;71(6): 1303-11. Zimprich A, Grabowski M, Asmus F, Naumann M, Berg D, Bertram M, et al. Mutations in the gene encoding epsilon-sarcoglycan cause myoclonus-dystonia syndrome. Nat Genet 2001;29(1):66-9. Saunders-Pullman R, Shriberg J, Heiman G, Raymond D, Wendt K, Kramer P, et al. Myoclonus dystonia: possible association with obsessive-compulsive disorder and alcohol dependence. Neurology 2002;58(2): 242-5. Pueschel SM, Friedman JH, Shetty T. Myoclonic dystonia. Childs Nerv Syst 1992;8(2):61-6. Zhou B, Westaway SK, Levinson B, Johnson MA, Gitschier J, Hayflick SJ. A novel pantothenate kinase gene (PANK2) is defective in Hallervorden-Spatz syndrome. Nat Genet 2001;28(4):345-9. Hayflick SJ. Pantothenate kinase-associated neurodegeneration (formerly Hallervorden-Spatz syndrome). J Neurol Sci 2003;207(1–2):106-7. Hayflick SJ, Westaway SK, Levinson B, Zhou B, Johnson MA, Ching KH, et al. Genetic, clinical, and radiographic delineation of Hallervorden-Spatz syndrome. N Engl J Med 2003;348(1):33-40. Swaiman KF. Hallervorden-Spatz syndrome. Pediatr Neurol 2001;25(2):102-8. Glick N. Dramatic reduction in self-injury in LeschNyhan disease following S-adenosylmethionine administration. J Inherit Metab Dis 2006;29(5):687. Jinnah HA, De Gregorio L, Harris JC, Nyhan WL, O’Neill JP. The spectrum of inherited mutations causing HPRT deficiency: 75 new cases and a review of 196 previously reported cases. Mutat Res 2000;463(3): 309-26. Jinnah HA, Visser JE, Harris JC, Verdu A, Larovere L, Ceballos-Picot I, et al. Delineation of the motor disorder of Lesch-Nyhan disease. Brain 2006;129(Pt 5):1201-17. McFarland R, Chinnery PF, Blakely EL, Schaefer AM, Morris AA, Foster SM, et al. Homoplasmy, heteroplasmy, and mitochondrial dystonia. Neurology 2007;69(9): 911-16. Soltanzadeh A, Soltanzadeh P, Nafissi S, Ghorbani A, Sikaroodi H, Lotfi J. Wilson’s disease: a great masquerader. Eur Neurol 2007;57(2):80-5. Machado A, Chien HF, Deguti MM, Cancado E, Azevedo RS, Scaff M, et al. Neurological manifestations in Wilson’s disease: report of 119 cases. Mov Disord 2006;21(12):2192-6. Ford B, Greene PE, LouisED, Bressman SB, Goodman RR, Brin MF, et al. Intrathecal baclofen in the treatment of dystonia. Adv Neurol 1998;78:199-210. Tsui JK, Fross RD, Calne S, Calne DB. Local treatment of spasmodic torticollis with botulinum toxin. Can J Neurol Sci 1987;14(3 Suppl):533-5.
Central procedures for primary dystonia
68. Kahilogullari G, Ugur HC, Savas A, Dirik EB, Akbostanci MC, Elibol B, et al. Management of a hemidystonic patient with thalamotomy, campotomy and cervical dorsal root entry zone operation. Stereotact Funct Neurosurg 2005;83(4):180-3. 69. Albanese A, Barnes MP, Bhatia KP, Fernandez-Alvarez E, Filippini G, Gasser T, et al. A systematic review on the diagnosis and treatment of primary (idiopathic) dystonia and dystonia plus syndromes: report of an EFNS/MDS-ES Task Force. Eur J Neurol 2006;13(5):433-44. 70. Hamani C, Moro E. Surgery for other movement disorders: dystonia, tics. Curr Opin Neurol 2007;20(4): 470-6. 71. Rutledge JN, Hilal SK, Silver AJ, Defendini R, Fahn S. Magnetic resonance imaging of dystonic states. Adv Neurol 1988;50:265-75. 72. Kupsch A, Benecke R, Muller J, Trottenberg T, Schneider GH, Poewe W, et al. Pallidal deep-brain stimulation in primary generalized or segmental dystonia. N Engl J Med 2006;355(19):1978-90. 73. Hung SW, Hamani C, Lozano AM, Poon YY, Piboolnurak P, Miyasaki JM, et al. Long-term outcome of bilateral pallidal deep brain stimulation for primary cervical dystonia. Neurology 2007;68(6):457-9. 74. Spiegel EA, Wycis HT. Pallidothalamotomy in chorea. Arch Neurol Psychiatry 1950;64(2):295-6. 75. Andrew J, Edwards JM, Rudolf Nde M. The placement of stereotaxic lesions for involuntary movements other than in Parkinson’s disease. Acta Neurochir (Wien) 1974;Suppl 21:39–47. 76. Andrew J, Fowler CJ, Harrison MJ. Stereotaxic thalamotomy in 55 cases of dystonia. Brain 1983;106(Pt 4): 981-1000. 77. Cardoso F, Jankovic J, Grossman RG, Hamilton WJ. Outcome after stereotactic thalamotomy for dystonia and hemiballismus. Neurosurgery. 1995;36(3):501-7; discussion 7–8. 78. Cooper IS. 20-year followup study of the neurosurgical treatment of dystonia musculorum deformans. Adv Neurol 1976;14:423-52. 79. Gros C, Frerebeau P, Privat JM, Perez-Dominguez E, Bazin M. Stereotaxic surgery of non-Parkinsonian dystonias and dyskinesias. Neurochirurgie 1976;22(6):539-52. 80. Hassler R, Dieckmann G. Stereotaxic treatment of torticollis according to animal experiment experiences about direction determined movement. Nervenarzt 1970;41(10): 437-87. 81. Hassler R, Riechert T. Indications and localization of stereotactic brain operations. Nervenarzt 1954;25(11): 441-7. 82. Krayenbuhl H, Siegfried J. Dentatotomies or thalamotomies in the treatment of hyperkinesia. Confin Neurol 1972;34(2):29-33. 83. Mundinger F, Riechert T, Disselhoff J. Long term results of stereotaxic operations on extrapyramidal hyperkinesia
84. 85.
86. 87.
88.
89.
90.
91.
92. 93.
94.
95.
96.
97.
98.
99.
108
(excluding parkinsonism). Confin Neurol 1970;32(2): 71-8. Tasker RR, Doorly T, Yamashiro K. Thalamotomy in generalized dystonia. Adv Neurol 1988;50:615-31. Cooper IS. Dystonia musculorum deformans: natural history and neurosurgical alleviation. J Pediatr 1969;74 (4):585-92. Guiot G, Brion S. Neurosurgery of choreoathetosic and Parkinsonian syndromes. Sem Hop 1952;28(49):2095-9. Hassler R, Riechert T, Mundinger F, Umbach W, Ganglberger JA. Physiological observations in stereotaxic operations in extrapyramidal motor disturbances. Brain 1960;83:337-50. Lozano AM, Kumar R, Gross RE, Giladi N, Hutchison WD, Dostrovsky JO, et al. Globus pallidus internus pallidotomy for generalized dystonia. Mov Disord 1997;12(6):865-70. Mundinger F, Riechert T, Disselhoff J. Long-term results of stereotactic treatment of spasmodic torticollis. Confin Neurol 1972;34(2):41-50. Shima F, Ishido K, Kato M. Roles of the basal ganglia outputs in movement disorders: a viewpoint based on experiences of stereotactic surgery for idiopathic dystonia. No To Hattatsu 1997;29(3):206-12. Speelman D, van Manen J. Cerebral palsy and stereotactic neurosurgery: long term results. J Neurol Neurosurg Psychiatry 1989;52(1):23-30. Vitek JL. Surgery for dystonia. Neurosurg Clin N Am 1998;9(2):345-66. Vitek JL, Zhang J, Evatt M, Mewes K, DeLong MR, Hashimoto T, et al. GPi pallidotomy for dystonia: clinical outcome and neuronal activity. Adv Neurol 1998;78:211-19. Krack P, Pollak P, Limousin P, Hoffmann D, Benazzouz A, Benabid AL. Inhibition of levodopa effects by internal pallidal stimulation. Mov Disord 1998;13(4):648-52. Krack P, Pollak P, Limousin P, Hoffmann D, Xie J, Benazzouz A, et al. Subthalamic nucleus or internal pallidal stimulation in young onset Parkinson’s disease. Brain 1998;121(Pt 3):451-7. Kulisevsky J, Lleo A, Gironell A, Molet J, Pascual-Sedano B, Pares P. Bilateral pallidal stimulation for cervical dystonia: dissociated pain and motor improvement. Neurology 2000;55(11):1754-5. Kumar R, Dagher A, Hutchison WD, Lang AE, Lozano AM. Globus pallidus deep brain stimulation for generalized dystonia: clinical and PET investigation. Neurology 1999;53(4):871-4. Loher TJ, Hasdemir MG, Burgunder JM, Krauss JK. Long-term follow-up study of chronic globus pallidus internus stimulation for posttraumatic hemidystonia. J Neurosurg 2000;92(3):457-60. Tronnier VM, Fogel W. Pallidal stimulation for generalized dystonia. Report of three cases. J Neurosurg 2000;92(3):453-6.
1829
1830
108
Central procedures for primary dystonia
100. Pollak P. Neurosurgical treatment of dyskinesias: pathophysiological consideration. Mov Disord 1999;14 Suppl 1:33-9. 101. Starr PA, Vitek JL, DeLong M, Bakay RA. Magnetic resonance imaging-based stereotactic localization of the globus pallidus and subthalamic nucleus. Neurosurgery 1999;44(2):303-13; discussion 13-4. 102. diPierro CG, Francel PC, Jackson TR, Kamiryo T, Laws ER Jr. Optimizing accuracy in magnetic resonance imaging-guided stereotaxis: a technique with validation based on the anterior commissure-posterior commissure line. J Neurosurg 1999;90(1):94-100. 103. Dormont D, Cornu P, Pidoux B, Bonnet AM, Biondi A, Oppenheim C, et al. Chronic thalamic stimulation with three-dimensional MR stereotactic guidance. AJNR Am J Neuroradiol 1997;18(6):1093-107. 104. Derosier C, Delegue G, Munier T, Pharaboz C, Cosnard G. MRI, geometric distortion of the image and stereotaxy. J Radiol 1991;72(6–7):349-53. 105. Burns JM, Wilkinson S, Kieltyka J, Overman J, Lundsgaarde T, Tollefson T, et al. Analysis of pallidotomy lesion positions using three-dimensional reconstruction of pallidal lesions, the basal ganglia, and the optic tract. Neurosurgery 1997;41(6):1303-16; discussion 16-8. 106. Lin JJ, Lin GY, Shih C, Lin SZ, Chang DC, Lee CC. Benefit of bilateral pallidotomy in the treatment of generalized dystonia. Case report. J Neurosurg 1999; 90(5):974-6. 107. Schneider S, Feifel E, Ott D, Schumacher M, Lucking CH, Deuschl G. Prolonged MRI T2 times of the lentiform nucleus in idiopathic spasmodic torticollis. Neurology 1994;44(5):846-50. 108. Schulz JB, Skalej M, Wedekind D, Luft AR, Abele M, Voigt K, et al. Magnetic resonance imaging-based volumetry differentiates idiopathic Parkinson’s syndrome from multiple system atrophy and progressive supranuclear palsy. Ann Neurol 1999;45(1):65-74. 109. Rosenfeld JV, Barnett GH, Palmer J. Computed tomography guided stereotactic thalamotomy using the Brown-Roberts-Wells system for nonparkinsonian movement disorders. Technical note. Stereotact Funct Neurosurg 1991;56(3):184-92. 110. Sumanaweera TS, Adler JR Jr, Napel S, Glover GH. Characterization of spatial distortion in magnetic resonance imaging and its implications for stereotactic surgery. Neurosurgery 1994;35(4):696-703; discussion 703-4. 111. Kamiryo T, Laws ER Jr. Stereotactic frame-based error in magnetic-resonance-guided stereotactic procedures: a method for measurement of error and standardization of technique. Stereotact Funct Neurosurg 1996;67 (3–4):198-209. 112. Vayssiere N, Hemm S, Zanca M, Picot MC, Bonafe A, Cif L, et al. Magnetic resonance imaging stereotactic
113.
114.
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
target localization for deep brain stimulation in dystonic children. J Neurosurg 2000;93(5):784-90. Ashkan K, Blomstedt P, Zrinzo L, Tisch S, Yousry T, Limousin-Dowsey P, et al. Variability of the subthalamic nucleus: the case for direct MRI guided targeting. Br J Neurosurg 2007;21(2):197-200. Vayssiere N, Hemm S, Cif L, Picot MC, Diakonova N, El Fertit H, et al. Comparison of atlas- and magnetic resonance imaging-based stereotactic targeting of the globus pallidus internus in the performance of deep brain stimulation for treatment of dystonia. J Neurosurg 2002;96(4):673-9. Martinez R, Vaquero J. Image-directed functional neurosurgery with the Cosman-Roberts-Wells stereotactic instrument. Acta Neurochir (Wien) 1991;113(3–4): 176-9. McKean JD, Allen PB, Filipow LJ, Miller JD. CT guided functional stereotaxic surgery. Acta Neurochir (Wien) 1987;87(1–2):8-13. Schuurman PR, de Bie RM, Majoie CB, Speelman JD, Bosch DA. A prospective comparison between threedimensional magnetic resonance imaging and ventriculography for target-coordinate determination in frame-based functional stereotactic neurosurgery. J Neurosurg 1999; 91(6):911-14. Tasker RR, Dostrovsky JO, Dolan EJ. Computerized tomography (CT) is just as accurate as ventriculography for functional stereotactic thalamotomy. Stereotact Funct Neurosurg 1991;57(4):157-66. Niemann K, van den Boom R, Haeselbarth K, Afshar F. A brainstem stereotactic atlas in a three-dimensional magnetic resonance imaging navigation system: first experiences with atlas-to-patient registration. J Neurosurg 1999;90(5):891-901. Krauss JK, Yianni J, Loher TJ, Aziz TZ. Deep brain stimulation for dystonia. J Clin Neurophysiol 2004; 21(1):18-30. Starr PA, Turner RS, Rau G, Lindsey N, Heath S, Volz M, et al. Microelectrode-guided implantation of deep brain stimulators into the globus pallidus internus for dystonia: techniques, electrode locations, and outcomes. J Neurosurg 2006;104(4):488-501. Merello M, Cerquetti D, Cammarota A, Tenca E, Artes C, Antico J, et al. Neuronal globus pallidus activity in patients with generalised dystonia. Mov Disord 2004; 19(5):548-54. Starr PA, Rau GM, Davis V, Marks WJ Jr, Ostrem JL, Simmons D, et al. Spontaneous pallidal neuronal activity in human dystonia: comparison with Parkinson’s disease and normal macaque. J Neurophysiol 2005; 93(6):3165-76. Hutchison WD, Lang AE, Dostrovsky JO, Lozano AM. Pallidal neuronal activity: implications for models of dystonia. Ann Neurol 2003;53(4):480-8.
Central procedures for primary dystonia
125. Vitek JL, Chockkan V, Zhang JY, Kaneoke Y, Evatt M, DeLong MR, et al. Neuronal activity in the basal ganglia in patients with generalized dystonia and hemiballismus. Ann Neurol 1999;46(1):22-35. 126. Lenz FA, Suarez JI, Metman LV, Reich SG, Karp BI, Hallett M, et al. Pallidal activity during dystonia: somatosensory reorganisation and changes with severity. J Neurol Neurosurg Psychiatry 1998;65(5):767-70. 127. Sanghera MK, Grossman RG, Kalhorn CG, Hamilton WJ, Ondo WG, Jankovic J. Basal ganglia neuronal discharge in primary and secondary dystonia in patients undergoing pallidotomy. Neurosurgery 2003;52(6): 1358-70; discussion 70-3. 128. Hariz MI. Complications of deep brain stimulation surgery. Mov Disord 2002;17 Suppl 3:S162-S166. 129. Lang AE. Surgical treatment of dystonia. Adv Neurol 1998;78:185-98. 130. Gros C, Frerebeau P, Perez-Dominguez E, Bazin M, Privat JM. Long term results of stereotaxic surgery for infantile dystonia and dyskinesia. Neurochirurgia (Stuttg) 1976;19(4):171-8. 131. Taira T, Ochiai T, Goto S, Hori T. Multimodal neurosurgical strategies for the management of dystonias. Acta Neurochir Suppl 2006;99:29-31. 132. Shibata T, Hirashima Y, Ikeda H, Asahi T, Hayashi N, Endo S. Stereotactic Voa-Vop complex thalamotomy for writer’s cramp. Eur Neurol 2005;53(1):38-9. 133. Taira T, Hori T. Stereotactic ventrooralis thalamotomy for task-specific focal hand dystonia (writer’s cramp). Stereotact Funct Neurosurg 2003;80(1–4):88-91. 134. Taira T, Harashima S, Hori T. Neurosurgical treatment for writer’s cramp. Acta Neurochir Suppl 2003;87: 129-31. 135. Burzaco J. Stereotactic pallidotomy in extrapyramidal disorders. Appl Neurophysiol 1985;48(1–6):283-7. 136. Iacono RP, Kuniyoshi SM, Lonser RR, Maeda G, Inae AM, Ashwal S. Simultaneous bilateral pallidoansotomy for idiopathic dystonia musculorum deformans. Pediatr Neurol 1996;14(2):145-8. 137. Justesen CR, Penn RD, Kroin JS, Egel RT. Stereotactic pallidotomy in a child with Hallervorden-Spatz disease. Case report. J Neurosurg 1999;90(3):551-4. 138. Lai T, Lai JM, Grossman RG. Functional recovery after bilateral pallidotomy for the treatment of early-onset primary generalized dystonia. Arch Phys Med Rehabil 1999;80(10):1340-2. 139. Lin JJ, Lin SZ, Chang DC. Pallidotomy and generalized dystonia. Mov Disord 1999;14(6):1057-9. 140. Ondo WG, Desaloms JM, Jankovic J, Grossman RG. Pallidotomy for generalized dystonia. Mov Disord 1998;13(4):693-8. 141. Cif L, El Fertit H, Vayssiere N, Hemm S, Hardouin E, Gannau A, et al. Treatment of dystonic syndromes by chronic electrical stimulation of the internal globus pallidus. J Neurosurg Sci 2003;47(1):52-5.
108
142. Coubes P, Cif L, El Fertit H, Hemm S, Vayssiere N, Serrat S, et al. Electrical stimulation of the globus pallidus internus in patients with primary generalized dystonia: long-term results. J Neurosurg 2004;101(2):189-94. 143. Krauss JK, Loher TJ, Weigel R, Capelle HH, Weber S, Burgunder JM. Chronic stimulation of the globus pallidus internus for treatment of non-dYT1 generalized dystonia and choreoathetosis: 2-year follow up. J Neurosurg 2003;98(4):785-92. 144. Kupsch A, Klaffke S, Kuhn AA, Meissner W, Arnold G, Schneider GH, et al. The effects of frequency in pallidal deep brain stimulation for primary dystonia. J Neurol 2003;250(10):1201-5. 145. Lozano AM, Abosch A. Pallidal stimulation for dystonia. Adv Neurol 2004;94:301-8. 146. Tisch S, Zrinzo L, Limousin P, Bhatia KP, Quinn N, Ashkan K, et al. The effect of electrode contact location on clinical efficacy of pallidal deep brain stimulation in primary generalised dystonia. J Neurol Neurosurg Psychiatry 2007;78(12):1314-19. 147. Pastor-Gomez J, Hernando-Requejo V, Luengo-Dos Santos A, Pedrosa-Sanchez M, Sola RG. Treatment of a case of generalised dystonia using subthalamic stimulation. Rev Neurol 2003;37(6):529-31. 148. Chou KL, Hurtig HI, Jaggi JL, Baltuch GH. Bilateral subthalamic nucleus deep brain stimulation in a patient with cervical dystonia and essential tremor. Mov Disord 2005;20(3):377-80. 149. Mundinger F. New stereotactic treatment of spasmodic torticollis with a brain stimulation system (author’s transl). Med Klin 1977;72(46):1982-6. 150. Andy OJ. Thalamic stimulation for control of movement disorders. Appl Neurophysiol 1983;46(1–4):107-11. 151. Sellal F, Hirsch E, Barth P, Blond S, Marescaux C. A case of symptomatic hemidystonia improved by ventroposterolateral thalamic electrostimulation. Mov Disord 1993;8(4):515-8. 152. Vercueil L, Pollak P, Fraix V, Caputo E, Moro E, Benazzouz A, et al. Deep brain stimulation in the treatment of severe dystonia. J Neurol 2001;248(8): 695-700. 153. Ghika J, Villemure JG, Miklossy J, Temperli P, Pralong E, Christen-Zaech S, et al. Postanoxic generalized dystonia improved by bilateral Voa thalamic deep brain stimulation. Neurology 2002;58(2):311-13. 154. Vidailhet M, Vercueil L, Houeto JL, Krystkowiak P, Benabid AL, Cornu P, et al. Bilateral deep-brain stimulation of the globus pallidus in primary generalized dystonia. N Engl J Med 2005;352(5):459-67. 155. Bereznai B, Steude U, Seelos K, Botzel K. Chronic high-frequency globus pallidus internus stimulation in different types of dystonia: a clinical, video, and MRI report of six patients presenting with segmental, cervical, and generalized dystonia. Mov Disord 2002;17(1):138-44.
1831
1832
108
Central procedures for primary dystonia
156. Diamond A, Shahed J, Azher S, Dat-Vuong K, Jankovic J. Globus pallidus deep brain stimulation in dystonia. Mov Disord 2006;21(5):692-5. 157. Andaluz N, Taha JM, Dalvi A. Bilateral pallidal deep brain stimulation for cervical and truncal dystonia. Neurology 2001;57(3):557-8. 158. Foote KD, Sanchez JC, Okun MS. Staged deep brain stimulation for refractory craniofacial dystonia with blepharospasm: case report and physiology. Neurosurgery 2005;56(2):E415; discussion E415. 159. Houser M, Waltz T. Meige syndrome and pallidal deep brain stimulation. Mov Disord 2005;20(9):1203-5. 160. Magarinos-Ascone CM, Regidor I, Martinez-Castrillo JC, Gomez-Galan M, Figueiras-Mendez R. Pallidal stimulation relieves myoclonus-dystonia syndrome. J Neurol Neurosurg Psychiatry 2005;76(7):989-91. 161. Cif L, Valente EM, Hemm S, Coubes C, Vayssiere N, Serrat S, et al. Deep brain stimulation in myoclonusdystonia syndrome. Mov Disord 2004;19(6):724-7. 162. Trottenberg T, Volkmann J, Deuschl G, Kuhn AA, Schneider GH, Muller J, et al. Treatment of severe tardive dystonia with pallidal deep brain stimulation. Neurology 2005;64(2):344-6. 163. Cohen OS, Hassin-Baer S, Spiegelmann R. Deep brain stimulation of the internal globus pallidus for refractory tardive dystonia. Parkinsonism Relat Disord 2007; 13(8):541-4. 164. Franzini A, Marras C, Ferroli P, Zorzi G, Bugiani O, Romito L, et al. Long-term high-frequency bilateral pallidal stimulation for neuroleptic-induced tardive dystonia. Report of two cases. J Neurosurg 2005; 102(4):721-5. 165. Eltahawy HA, Feinstein A, Khan F, Saint-Cyr J, Lang AE, Lozano AM. Bilateral globus pallidus internus deep brain stimulation in tardive dyskinesia: a case report. Mov Disord 2004;19(8):969-72. 166. Krause M, Fogel W, Kloss M, Rasche D, Volkmann J, Tronnier V. Pallidal stimulation for dystonia. Neurosurgery 2004;55(6):1361-8; discussion 8-70. 167. Eltahawy HA, Saint-Cyr J, Giladi N, Lang AE, Lozano AM. Primary dystonia is more responsive than secondary dystonia to pallidal interventions: outcome after pallidotomy or pallidal deep brain stimulation. Neurosurgery 2004;54(3):613-19; discussion 9-21. 168. Castelnau P, Cif L, Valente EM, Vayssiere N, Hemm S, Gannau A, et al. Pallidal stimulation improves pantothenate kinase-associated neurodegeneration. Ann Neurol 2005;57(5):738-41. 169. Umemura A, Jaggi JL, Dolinskas CA, Stern MB, Baltuch GH. Pallidal deep brain stimulation for longstanding severe generalized dystonia in HallervordenSpatz syndrome. Case report. J Neurosurg 2004; 100(4):706-9. 170. Cif L, Biolsi B, Gavarini S, Saux A, Robles SG, Tancu C, et al. Antero-ventral internal pallidum stimulation
171.
172.
173.
174.
175.
176.
177.
178.
179.
180.
181.
182.
183.
improves behavioral disorders in Lesch-Nyhan disease. Mov Disord 2007;22(14):2126-9. Hebb MO, Garcia R, Gaudet P, Mendez IM. Bilateral stimulation of the globus pallidus internus to treat choreathetosis in Huntington’s disease: technical case report. Neurosurgery 2006;58(2):E383; discussion E383. Moro E, Lang AE, Strafella AP, Poon YY, Arango PM, Dagher A, et al. Bilateral globus pallidus stimulation for Huntington’s disease. Ann Neurol 2004;56(2):290-4. Biolsi B, Cif L, El Fertit H, Gil Robles S, Coubes C. Longterm follow-up of Huntington’s disease treated by bilateral deep brain stimulation. J Neurosurg 2008 (in press). Baker KB, Tkach JA, Nyenhuis JA, Phillips M, Shellock FG, Gonzalez-Martinez J, et al. Evaluation of specific absorption rate as a dosimeter of MRI-related implant heating. J Magn Reson Imaging 2004; 20(2):315-20. Brodkey JS, Miyazaki Y, Ervin FR, Mark VH. Reversible Heat Lesions with Radiofrequency Current. a Method of Stereotactic Localization. J Neurosurg 1964;21:49-53. Tronnier VM, Staubert A, Hahnel S, Sarem-Aslani A. Magnetic resonance imaging with implanted neurostimulators: an in vitro and in vivo study. Neurosurgery 1999;44(1):118-25; discussion 25-6. Georgi JC, Stippich C, Tronnier VM, Heiland S. Active deep brain stimulation during MRI: a feasibility study. Magn Reson Med 2004;51(2):380-8. Baker KB, Tkach J, Hall JD, Nyenhuis JA, Shellock FG, Rezai AR. Reduction of magnetic resonance imagingrelated heating in deep brain stimulation leads using a lead management device. Neurosurgery. 2005 Oct; 57(4 Suppl):392-7; discussion 392-7. Finelli DA, Rezai AR, Ruggieri PM, Tkach JA, Nyenhuis JA, Hrdlicka G, et al. MR imaging-related heating of deep brain stimulation electrodes: in vitro study. AJNR Am J Neuroradiol 2002;23(10):1795-802. Rezai AR, Finelli D, Nyenhuis JA, Hrdlicka G, Tkach J, Sharan A, et al. Neurostimulation systems for deep brain stimulation: in vitro evaluation of magnetic resonance imaging-related heating at 1.5 tesla. J Magn Reson Imaging 2002;15(3):241-50. Rezai AR, Baker KB, Tkach JA, Phillips M, Hrdlicka G, Sharan AD, et al. Is magnetic resonance imaging safe for patients with neurostimulation systems used for deep brain stimulation? Neurosurgery 2005;57(5):1056-62; discussion 1056-62. Spiegel J, Fuss G, Backens M, Reith W, Magnus T, Becker G, et al. Transient dystonia following magnetic resonance imaging in a patient with deep brain stimulation electrodes for the treatment of Parkinson disease. Case report. J Neurosurg 2003;99(4):772-4. Henderson JM, Tkach J, Phillips M, Baker K, Shellock FG, Rezai AR. Permanent neurological deficit related to magnetic resonance imaging in a patient with implanted deep brain stimulation electrodes for Parkinson’s disease:
Central procedures for primary dystonia
case report. Neurosurgery 2005;57(5):E1063; discussion E1063. 184. Moro E, Esselink RJ, Xie J, Hommel M, Benabid AL, Pollak P. The impact on Parkinson’s disease of electrical parameter settings in STN stimulation. Neurology 2002;59(5):706-13. 185. Vercueil L, Houeto JL, Krystkowiak P, Lagrange C, Cassim F, Benazzouz A, et al. Effects of pulse width variations in pallidal stimulation for primary generalized dystonia. J Neurol 2007;254(11):1533-7. 186. Kuncel AM, Grill WM. Selection of stimulus parameters for deep brain stimulation. Clin Neurophysiol 2004; 115(11):2431-41.
108
187. Burke RE, Fahn S, Marsden CD, Bressman SB, Moskowitz C, Friedman J. Validity and reliability of a rating scale for the primary torsion dystonias. Neurology 1985;35(1):73-7. 188. Hemm S, Vayssiere N, Mennessier G, Cif L, Zanca M, Ravel P, et al. Evolution of brain impedance in dystonic patients treated by GPi electrical stimulation. Neuromodulation 2004;7(2):67-75. 189. Vayssiere N, van der Gaag N, Cif L, Hemm S, Verdier R, Frerebeau P, et al. Deep brain stimulation for dystonia confirming a somatotopic organization in the globus pallidus internus. J Neurosurg 2004;101 (2):181-8.
1833
115 Destructive Neurosurgical Procedures for Spasticity M. Sindou . P. Mertens
Introduction Spasticity can be defined as a velocity-dependent resistance to passive movement of a joint and its associated musculature. Spasticity is characterized by hyperexcitability of the stretch-reflex related to the loss of inhibitory influences from descending supraspinal structures. Spasticity should not be treated just because it is present as it may be useful for compensating for loss of motor power. Spasticity should be treated only when excess tone leads to further functional losses, impairs locomotion for induces deformities. Functional neurosurgery should be considered when hyperspasticity cannot be controlled by physical therapy and medications. The methods of surgical management are classified according to whether their impact is general or focal and whether the effects are temporary or permanent (> Figure 115-1). They include not only Intrathecal Baclofen Therapy (ITB) and Botulinum toxin injections, but also destructure surgery directed to the peripheral nerves, dorsal roots, dorsal root entry zone or spinal cord. Neurodestructive procedures must be performed so that excess of tone is reduced without suppressing useful muscular tone or impairing any residual motor/sensory functions. In patients who retain some masked voluntary motility, the goal is to re-equilibrate the balance between paretic agonist muscles and spastic antagonist muscles, resulting in improvement in (or reappearance of) voluntary motility. In patients with poor residual or no motor function preoperatively, the aim is to stop the progressive orthopedic deformities and improve comfort. #
Springer-Verlag Berlin/Heidelberg 2009
Because features of disabling spasticity and its consequences differ from one patient to another, the general rule is to tailor the appropriate neurosurgical program to the patient. The first step is to define the objective(s) of the treatment: improvement in function, prevention of deformities, or alleviation of discomfort and pain in very disabled patients; in other words, what can be gained and what will not be obtained by surgery. These issues must be explained carefully to the patient, relatives and care-givers. This step, as well as all the other phases of the neurosurgical program, must be conceived, discussed and applied within the frame of a multidisciplinary team. Our experience with the neurosurgical treatment of spasticity, in more than 1,000 patients over the last 20 years, has led us to believe that the teams treating spasticity should have all requisite technical modalities in their armamentarium [1]. ITB is indicated for adult patients, paraplegic or tetraplegic, with diffuse spasticity, especially from spinal origin. ITB can also be used to treat spasticity related to cerebral palsy but in elder children. Neuro-destructive operations are indicated for focalized spasticity in the limbs – especially after treatment with Botulinum toxin injections has been tried and revealed insufficient. Peripheral neurotomies are most justified when harmful spasticity affects one (or a few) muscular group(s). A preliminary test with an anesthetic block may help to predict the outcome by mimicking the effect of the neurotomy. When harmful spasticity affects the whole limb(s) in hemiplegic or paraplegic patients, surgery involving the dorsal roots (dorsal rhizotomies) or Dorsal Root Entry Zone (microsurgical DREZotomy)
1936
115
Destructive neurosurgical procedures for spasticity
. Figure 115-1 Methods for managing spasticity based on whether it is focal or general and on whether it is permanent or temporary
may be the solution. Complementary orthopedic operations are frequently needed in patients with associated irreducible contractures, tendon retractions, and/or joint deformities.
Selective Peripheral Neurotomies (SPN) Peripheral neurotomies (PN) were introduced for the treatment of localized spastic deformities in the foot by Stoffel [2]. More recently, Gros and associates [3,4] and the present authors [5,6] have made PN more selective by using microsurgery for fine dissection of fascicles and mapping with intraoperative electrical stimulation to better identify the function of the individual nerve fascicles (> Figure 115-2). Neurotomy consists in partial sectioning of one or several motor branches of the nerve(s) innervating targeted muscle(s) in which spasticity is considered to be excessive. It interrupts the segmental reflex arc by acting on both afferent and efferent pathways. Neurotomy eliminates the afferent pathway corresponding to proprioception of the concerned muscle and induces paralysis by section of the efferent pathway. Neurotomies must not involve sensory nerve fibers as even partial section of the extra could be responsible
for deafferentation pain. The motor branches must be either clearly isolated from the nerve trunk, or dissected and identified within the fascicles of the nerve trunk several centimeters proximal to the formation of an identifiable branch. There is no scientific basis for defining the extent of the section. However, all surgeons agree that a partial neurotomy must include the sectioning of 50–80% (usually 75%) of all branches to a targeted muscle in order to be effective.
Technical Principles [5–10] Pre-operative Motor Blocks Before recommending SPN, a test using motor blocks innervating the targeted muscle(s) is of prime importance [5–7]. These blocks, using local anesthetics such as long-lasting bupivacaine, enable the surgeon to evaluate the motor strength of antagonist muscles and determine if articular motion limitations result from spasticity or musculotendinous contractures/articular ankyloses. Botulinum toxin injections can be used as a ‘‘prolonged’’ test for several weeks or months before pursuing neurosurgical treatment, as their effects will mimic the outcome of selective neurotomies on the injected nerves.
Destructive neurosurgical procedures for spasticity
115
. Figure 115-2 Microsurgical procedure of selective peripheral neurotomy in the median nerve: Operative microsurgical views showing the steps of neurotomy for the pronator teres muscular branch. (a) Muscular branch of pronator teres is identified with motor electrical stimulation. (b) According to the preoperative evaluation and subsequent program for this particular patient, 50% of the isolated motor fascicles are resected, 5-mm long, near the muscle to be sure that only its muscular branches are cut. (c) Proximal stump is coagulated with bipolar forceps to prevent regrowth of fibers
This strategy of using pre-operative injection tests allows the patient to appreciate the benefit that can follow a selective neurotomy. Objectives of surgery may be cosmetic (for instance to allow the patient to put his or her hand in a pocket), related to nursing (for instance to allow a caregiver to wash the palm of the patient’s hand), or for functional improvement.
Anesthesia Neurotomy is performed under general anesthesia but without long-lasting curarization. Patient positioning is not always easy because passive mobility is often limited. Most importantly, it
may be useful to test the efficacy of the neurotomy procedure, intraoperatively, by evaluating the stretch reflex implies, which that reflexes be not depressed by the anesthetic drugs. Muscle relaxants must be avoided; nitrogen monoxide and propofol are contraindicated because they modify reflex excitability. The general anesthesia has to be administered without long-lasting curarization so that the motor responses elicited by bipolar electrical stimulation of motor branches can be detected to identify the nerve fascicles.
Mapping Mapping which is the anatomical identification of motor branches fascicles is necessary for an
1937
1938
115
Destructive neurosurgical procedures for spasticity
accurate functional surgery. It is also an essential step to avoid sensory impairment. It usually requires the use of the operating microscope. Frequent variation in the emergence of nerve branches and limited surgical access can make this a difficult step. Identification of fascicles is of course based on descriptive anatomy but needs to be accurate and rigorous to be checked by the study of the muscular responses to bipolar electrical stimulation (NIMBUS: Multifunctional Neuro-Stimulator, Hemodia, Toulouse France). Stimulation is performed with a 2 Hz frequency, at low intensity – ordinarily 1 mA – to avoid electrical diffusion and an incorrect interpretation. Usually, a triple stimulation hook, composed of an anode between two cathodes, is used to grasp the nerve branch to be stimulated. The response to stimulation must be visualized in the form of clinically observable movements of the limb muscles and/or EMG recordings.
Sectioning Once all of motor branches (or fascicles) have been identified by electrical stimulation, those considered responsible for harmful spasticity are marked separately with colored tapes. According to the preoperative or thresds program, variable proportions (50–80% depending on the degree of preoperative spasticity) of the isolated motor branches (or fascicles) are resected under the operating microscope near the muscle to ensure that only the muscular branches are cut. The resection is 5-mm long from the proximal stump, which is coagulated with a fine bipolar forceps to prevent regrowth of fibers. When there are several nerve branches (or fascicles) for one muscle, one or more branches (or fascicles) see divided until the global amount determined to be cut for the considered muscle is attained. The effect of each nerve resection on spasticity is then evaluated by comparing muscle
responses to electrical stimulation, proximally and then distally to the resected portion. If the response after proximal stimulation is still intense, further resection can be performed. The aim is to decrease motor innervation enough to avoid further recurrence of spasticity by ‘‘take-over’’ (that is reinnervation or ‘‘adoption’’ of muscular fibers denervated as a result of the neurotomy, by the surrounding motor fibers).
Operative Techniques Surgery for the Lower Limb [5,11,12] Obturator Neurotomy for the Hip
Obturator neurotomy eliminates spasticity of adductor muscles. It is often proposed to diplegic children with cerebral palsy when their walking is hampered by crossing of the lower limbs. It can also be proposed for paraplegic children to facilitate perineal toilet and self-catheterization. The incision can be performed along the body of the adductor longus in the proximal part of the thigh. Over recent years, transverse incision in the hip flexion fold centered on the prominence of the adductor longus tendon has been preferred. In addition to its more aesthetic appearance, this incision facilitates adductor longus tenotomy when necessary (> Figure 115-3a). To rapidly locate the anterior branch of the obturator nerve, the dissection is conducted laterally to the adductor longus muscle body. The posterior branch is situated more deeply and should be spared to preserve the hip stabilizing muscles (> Figure 115-3b). Hamstring Neurotomy for the Knee
Hamstring neurotomy is indicated in children with spastic diplegia to counter the accentuation of the flexion deformity of the knees observed with growth. The transverse incision is performed in the gluteal fold, centered on the groove between the ischium and the greater trochanter (> Figure 115-4a). After crossing the fibers of the gluteus maximums, the sciatic nerve is iden-
Destructive neurosurgical procedures for spasticity
. Figure 115-3 (a) Skin incision for (right) obturator neurotomy, on the relief of adductor longus muscle (1) or in the hip flexion fold centered on the prominence of the adductor longus tendon (2) which gives a cosmetic advantage. (b) Dissection of the anterior branch (AB) of the right obturator nerve (ON). The adductor longus (AL) is retracted laterally and the gracilis (G) medially. The nerve is anterior to the adductor brevis (AB). Adductor brevis nerve (1), (2), adductor longus nerve (3), gracilis nerve (4), (5). The posterior branch (PB) of the obturator nerve lies under the adductor brevis (AB) and should be spared
tified in the depth of the incision. The branches to the hamstring muscles are isolated at the border of the nerve, primarily based on responses of the semitendinosus muscle, which is often the major muscle responsible for spasticity (> Figure 115-4b). Tibial Neurotomy for the Foot
Tibial neurotomy is indicated for the treatment of varus spastic foot drop with or without claw toes. It consists of exposing all motor branches of the tibial nerve at the popliteal fossa (i.e., the nerves to gastrocnemius and soleus, tibialis posterioris, popliteus, flexor hallucis longus, and flexor digitorum longus). Soleus has been demonstrated to be usually almost fully responsible for the pathogenesis of spastic foot drop, allowing sparing of gastrocnemius [13]. The incision can be vertical, on either side of the popliteal fossa, and extending inferiorly. Over recent years, a transverse incision in
115
. Figure 115-4 (a) The skin incision for hamstring neurotomies (on the right side) is located on the midline between the ischial tuberosity (IT) and the greater trochanter (GT) (1). A transverse incision can also be performed in the gluteal fold (2) centered on the groove between the ischial tuberosity and the greater trochanter, for better aesthetic results. (b) Dissection of the right sciatic nerve (SN), under the piriformis muscle (P), after passing through the fibers of the gluteus maximus muscle (GM). The epineurium of the nerve is opened and fascicles for hamstring muscles (HF) are localized in the medial part of the nerve trunk. Inferior gluteal nerve (IGN), inferior gluteal artery (IGA)
the popliteal fossa, which gives a much better long-term aesthetic result, has been adopted. In addition, this transverse incision allows a high tenotomy of the gastrocnemius insertion fascia at the end of the operation, if necessary (> Figure 115-5a). The first nerve encountered is the sensory medial cutaneous nerve of the leg. Situated immediately anterior to the saphenous vein, it must be spared. More deeply, the tibial nerve trunk, from which the nerves to the gastrocnemius emerge, is easily identified. The superior soleus nerve is situated in the midline, just posterior to the tibial nerve. The effect of a soleus neurotomy is assessed by the immediate intraoperative disappearance of ankle clonus. By retracting the tibial nerve trunk medially by using a traction suture, the other branches can be identified by electrical stimulation as they emerge from the lateral edge of the
1939
1940
115
Destructive neurosurgical procedures for spasticity
. Figure 115-5 (a) Skin incision of the tibial nerve, vertical incision either side of the right popliteal fossa, extending inferiorly (1), or transverse incision in the popliteal fossa (2) for a better long-term aesthetic result. (b) Dissection of the tibial nerve, dorsal view of the right popliteal region. Tibial nerve (1), Peroneal nerve (2). The sensory sural nerve (3) lies superficially just beneath the subcutaneous aponeurosis between the two gastrocnemius muscles. The medial and lateral gastrocnemius nerves (4) may arise either separately from the both sides of the tibial trunk or posteriorly from a common origin, sometimes including the sensory sural nerve. Each gastrocnemius nerve usually divides into two distal branches when approaching the muscle. The one or two soleus nerves (5) may arise from a common origin or quite separately from the tibial nerve. The posterior tibialis nerve (6), like the soleus nerve, originates from the ventro-lateral aspect of the tibial nerve, but more distally at the level of the soleus arch. Sometimes it may originate from a common trunk with the inferior branch of the soleus nerve. The distal trunk of the tibial nerve (7) contains 5–8 fascicles averaging 1 mm in diameter each; two thirds of them are motor fascicles, one third are sensory ones. Muscles: lateral and medial gastrocnemius (LG-MG), Soleus (S)
tibial nerve trunk. The most lateral branch is often the popliteal nerve, followed by the tibial posterior nerve and finally by the inferior soleus nerve and flexor digitorum longus nerves. Some fascicles, often larger, can give a toe flexion response via intrinsic toe flexors (> Figure 115-5b). However, neurotomy of these branches is not recommended, if they cannot be clearly individualized at this level, the more so as they may be mixed with sensory fascicles. Anterior Tibial Neurotomy for Extensor Hallucis
After botulinum toxin injections fail, this neurotomy (rarely performed) is indicated to treat permanent extension of the hallux making it difficult to wear shoes. In practice, this neurotomy may be indicated after unjustified section of the flexor hallucis tendon, responsible for a disequilibrium that favors the extensor. A vertical incision is centered on the junction between the tibialis anterior and the extensor hallucis, at the middle third of the leg. The tibial nerve is situated deeply between these two muscle heads and the neurotomy is performed on the motor branch to the extensor hallucis. Femoral Neurotomy for the Quadriceps
Femoral neurotomy is indicated to treat excessive spasticity of the quadriceps muscle. This muscle is very often spastic and can interfere with gait by limiting knee flexion during the swing phase. Given its ‘‘strategic’’ importance in maintaining upright posture, a motor block is an essential part of the preoperative evaluation. The neurotomy mainly concerns the motor branch to the rectus femoris and vastus intermedius muscles. The incision is horizontal in the hip flexion fold (> Figure 115-6a). The dissection passes medial to the sartorius muscle body and exposes the motor branches of the femoral nerve, first the nerve to the rectus femoris and then, more deeply, the nerve to the vastus intermedius (> Figure 115-6b). Electrical stimulation is essential given the large number of sensory fascicles of this nerve that must be spared.
Destructive neurosurgical procedures for spasticity
. Figure 115-6 (a) Skin incision for right femoral neurotomy, below the inguinal ligament, laterally to the femoral artery (1) or horizontal in the hip flexion fold (2) for better aesthetic results. (b) Dissection of the right femoral nerve (FN) and its branches, after opening the anterior fascia of the psoas muscle (P). The bipolar stimulation allows identification of the two or three branches to sartorius muscle (S) and three or four to rectus femoris muscle, which produces flexion of the hip. The nerve to the vastus intermedius can be found more deeply. Femoral artery (FA), femoral vein (FV)
115
ternal rotation and adduction. The skin incision follows the inner border of teres major, from the lower border of the deltoid muscle’s posterior head to the lower extremity of the scapula. The lower border of the long portion of brachii triceps constitutes the upper limit of the approach. The dissection continues deeply between teres minor and major muscles. In the vicinity of the subscapularis artery, the nerve ending on teres major is identified. The nerve is surrounded by thick fat when approaching the anterior facet of the muscle body [17]. Musculo-cutaneous Neurotomy for the Elbow
Surgery for the Upper Limb [14–16] Pectoralis Major Neurotomy for the Shoulder
Neurotomy of collateral branches of the brachial plexus innervating the pectoralis major are indicated for spasticity of the shoulder with internal rotation and adduction. The skin incision is made at the innermost part of the deltopectoral sulcus and curves along the clavicular axis. First, the clavipectoralis fascia is opened. Then, the upper border of the pectoralis major muscle is reflected downward. Close to the thoracoacromialis artery, the ansa of the pectoralis muscle is identified with the aid of a nerve stimulator [17]. Teres Major Neurotomy for the Shoulder
Neurotomy of collateral branches of the brachial plexus innervating the teres major are also indicated for spasticity of the shoulder with in-
Neurotomy of the musculo-cutaneous nerve is indicated for spasticity of the elbow with flexion, depending on the biceps brachii and the brachialis muscles. The skin incision is performed longitudinally. It extends from the inferior edge of pectoralis major, medial to the biceps brachii, down to 5 cm (> Figure 115-7a). The superficial fascia is opened between the biceps laterally and the brachialis medially. The brachial artery and median nerve exit medially. The dissection proceeds in this space, where the musculocutaneous nerve lies anterior to the brachialis muscle (> Figure 115-7b). Opening the epinerium, allows the fascicles of the nerve to be dissected under high magnification of the operating microscope. The motor fascicles are distinguished from the sensitive ones by using the nerve stimulator. Median Neurotomy for Wrist and Fingers
Neurotomy of the median nerve is indicated for spasticity of the forearm with pronation depending on the pronator teres and quadratus muscle and spasticity of the wrist with flexion depending on the flexor carpi radialis and palmaris longus muscles. In the hand, median neurotomy is indicated for spasticity of the fingers with flexion depending on the flexor digitorum superficialis (flexion of proximal interphalangeal joint and metacarpophalangeal joint) and on the flexor digitorum profondus muscle (flexion of distal interphalangeal, proximal interphalangeal, and
1941
1942
115
Destructive neurosurgical procedures for spasticity
. Figure 115-7 (a) Skin incision for right musculocutaneous neurotomy, along the medial aspect of the biceps brachii, under the inferior edge of the pectoralis major muscle. (b) Dissection of the right musculocutaneous nerve (MC) in the space between the biceps brachii (BB) laterally, the coraco brachialis (CB) medially and the brachialis (B) posteriorly. Branches to brachialis (1), (2) and to biceps brachii (3), (4) are recognized by stimulation giving elbow flexion. Humeral artery (H) with median nerve are situated medially and are not dissected
metacarpophalangeal joint), partly innervated by the median nerve. Swan neck deformation of the fingers depending on the lumbrical and interosseous muscles can be limited by neurotomy, these muscles being innervated by the median and ulnar nerves. As far as the thumb is concerned, neurotomy of the median nerve is indicated for spasticity with flexion and adduction/flexion (thumb-inpalm deformity) depending on the flexor pollicis longus. The skin incision begins 2–3 cm above the flexion line of the elbow, medial to the biceps brachii tendon, passes through the elbow, and curves toward the junction of the upper and middle thirds of the anterior forearm (the convexity of the curve turns laterally) (> Figure 115-8a). Thereafter, the median nerve is searched medially to the brachial artery and recognized at the elbow, deep under the lacertus fibrosus, which is cut. Sharp dissection is used to separate the branches of the median nerve. The pronators teres belly with its two heads is retracted medially and distally so that its muscular branches can be inspected. This muscle is retracted up and laterally while the
flexor carpi radialis is pulled down and medially. The muscular branches to the flexor carpi radialis and to the flexor digitorum superficialis can then be seen. Finally, the latter is retracted medially uncovering the branches to the flexor digitorum profondus, the flexor pollicis longus, and the pronator quadratus. These latter muscular branches may be individualized as separate branches or remain together in the distal trunk of the anterior interosseous nerve. Sometimes, it may be useful to divide the fibrous arch of the flexor digitorum superficialis muscle to make the dissection easier (> Figure 115-8b). Besides this approach, which has a ‘‘wide’’ configuration, a ‘‘minimal’’ approach can be performed. The different fascicles in the trunk of the median nerve, just medial to the brachial artery, are dissected. This latter approach provides a better cosmetic outcome (shorter incision). However, it has the inconvenience of offering a nerve exposure less propitious to identify the various motor branches in the form of fascicles enclosed in the nerve sheath and mixed with the
Destructive neurosurgical procedures for spasticity
sensory ones. This entails the risk of sensory complications, especially of developing a complex regional pain syndrome. Ulnar Neurotomy for Wrist and Fingers
. Figure 115-8 (a) Skin incision on the right forearm for median neurotomy from the medial aspect of the biceps brachii at the level of the elbow longitudinally along the bicipital crest (1). The incision can eventually be continued distally towards the midline above the wrist (2). (b) Dissection of the median nerve in two stages. First stage of the dissection (upper figure), the pronators teres (PT) is retracted upward and laterally, the flexor carpi radialis (FCR) medially. Branches from the median nerve (MN), before it passes under the fibrous arch of the flexor digitorum superficialis (FDS), are dissected: to the pronators teres (1), and two nerve trunks to the flexor carpi radialis, palmaris longus and flexor digitorum superficialis (2), (3). Second stage of the dissection (lower figure), the fibrous arch of the FDS is sectioned to allow a more distal dissection of the median nerve. The FDS is retracted medially and branches from the median nerve are identified: (1) to the flexor pollicis longus (FPL); (2) to the flexor digitorum profondus (FDP); (3) the interosseous nerve and its proper branches to these muscles
115
Neurotomy of the ulnar nerve is also indicated for spasticity of the wrist with flexion and for spasticity with ulnar deviation, both depending on the flexor carpi ulnaris. In the hand, ulnar neurotomy is indicated for spasticity of the fingers with flexion depending on the flexor digitorum profondus muscle (flexion of distal interphalangeal, proximal interphalangeal, and metacarpophalangeal joint), partly innervated by the ulnar nerve. Ulnar neurotomy is also indicated for spasticity with adduction/flexion depending on the adductor pollicis muscle, and in combination with median neurotomy, for swan-neck deformation of the fingers. A separate arched skin incision is performed to expose the ulnar nerve at the medial part of the elbow (> Figure 115-9). After subcutaneous dissection, the ulnar nerve is identified medially to the medial epicondyle, where it enters between the two heads of the flexor carpi ulnaris. There, the motor branches to this latter muscle are identified. More distally, the branches to the medial half of the flexor digitorum profondus are found.
Complications, Side-effects and Causes of Recurrence Local complications include postoperative hematoma and infection. They are rather rare if preventive measures are respected.
. Figure 115-9 Skin incision on the right forearm for ulnar neurotomy, either a longitudinal incision posteriorly to the medial epicondyle and medially to the olecranon at the elbow (1), or a transverse medial incision in the wrist fold (2), depending on the location of the spastic muscle
1943
1944
115
Destructive neurosurgical procedures for spasticity
Sensory disturbances, such as paresthesias and dysesthesias, followed by transient de-afferentation pain (2 months), can be observed if the section accidentally includes sensory fascicles. This complication underscores the importance of precise stimulation. Patients rarely complain of decreased muscle strength after neurotomy. Muscle function is redundant, and no single muscle ensures the movement of a body segment, without the possibility of substitution by another muscle. Specifically in surgery of the upper limb, complications include (rare) transitory hypesthesia of the anterior part of the forearm because of the surgical approach with lesion of subcutaneous sensitive branches rather than of neurotomies themselves (concerning only muscular nerves). Paresis of flexors of the elbow, wrist, fingers, or both (because of excessive nerves sectioning) is rare, generally transient, and responds to physical therapy. Recurrence of spasticity can be observed when the mean amount of sectioning is insufficient, in which case reoperation can be performed.
Functional Improvement The mandate is that neuroablative procedures must be performed so that excessive hypertonia is reduced without suppressing useful muscular tone or impairing any residual motor and/or sensory functions. In patients who retain some masked voluntary motility, the goal is to reequilibrate the balance between paretic agonist muscles and spastic antagonist muscles, resulting in improvement in or reappearance of voluntary motility (> Figure 115-10).
Surgery in Spinal Roots, Dorsal Root Entry Zone, and Spinal Cord Rationale and Overview In 1898, using the animal model of mesencephalic trans-sectioning, Sherrington showed that decerebration rigidity could be abolished by sectioning dorsal roots [18]. From these experimental data, Foerster in 1908 performed the first
. Figure 115-10 Gait analysis with three dimensional joint analysis and poly-EMG. Preoperative Polyelectromyography shows desynchronized activities in the triceps surae, with abnormal cocontractions of the tibialis anterioris and triceps surae (left). After tibial neurotomy (right), there is a reappearance of muscular activity in the tibialis anterior and normal alternation between the contraction of the triceps surae at the end of the stance phase and that of the tibialis anterioris during the swing phase
Destructive neurosurgical procedures for spasticity
dorsal rhizotomies from L1 to S2 (not L4, root of quadriceps) in four patients for the treatment of lower limb spasticity [19]. The results of the first series of 159 patients with cerebral palsy were published in 1913. In his article ‘‘On the indications and results of the excision of posterior spinal nerves in men,’’ Foerster [20] suggested the following recommendations: ‘‘For severe spastic paraplegia, I recommend resecting at least five roots. It is necessary to leave the fourth lumbar root, since this root generally guarantees the extensor reflex of the knee so very necessary for standing and walking. Thus the general rule is resection of the second, third and fifth lumbar, and first and second sacral roots. Unfortunately, there exist individual differences. In some cases, the fourth lumbar does not affect knee extension but knee flexion, as the fifth lumbar and first sacral do; the knee extension is affected only by the second and third lumbar roots. In order to know by which lumbar roots the extension reflex of the knee is effected, we must have recourse to the electrical current during the operation.’’ Considering these results, Foerster concluded that: ‘‘we must bear in mind that the resection of the posterior roots relieves only the spastic symptoms, but not the paralysis, if such exists besides the spastic state. The disappearance of the spasticity after the root resection is the best proof of the sensory origin of the spastic contracture. But a certain degree of spasm sometimes returns, owing to the fact that the spinal grey matter is gradually recharged by the remaining posterior roots’’ [21]. To reduce the sensory secondary effects of Foerster’s original technique, Gros and collaborators [21–23] introduced a slight modification which consisted in preserving one rootlet out of five (on average) for each root, from L1 to S1. This method induced a significant decrease in spasticity in 75% of the 25 patients that were followed, on an average, for a period of 3 years and 8 months. The number of preserved rootlets was sufficient to maintain normal sensation in
115
the patients’ lower limbs in 70% of cases. Apart from the effects on the lower limbs, the authors also observed a decrease in the spasticity of the upper limbs and an improvement in speech and swallowing in 18 out of 25 cases. This effect is termed an indirect effect. To further reduce the incidence of secondary effects on the postural tonus of patients who were capable of walking, Gros and pupils [22,24] introduced the topographical selection of rootlets to preserve the innervation of muscles responsible for the useful tonus (quadriceps, abdominal and gluteal muscles in particular). This technique was termed ‘‘sectorial posterior rhizotomy.’’ The mapping method involves the anatomical identification and stimulation of rootlets, which is essential to determine their different functions. In 1977, Fraioli and Guidetti developed a slight variation of the dorsal rhizotomy: the ‘‘partial dorsal rhizotomy’’ which consisted of sectioning the dorsal part of each rootlet a few millimeters before their entry into the dorsolateral sulcus [25]. In 1976, Fasano and collaborators [26] introduced ‘‘functional dorsal rhizotomy.’’ This technique is based on the intra-operative bipolar stimulation of posterior rootlets in association with observations of clinical responses and electromyography of the lower limbs. The authors consider that these responses, characterized by an exaggeration in the duration or extent of the motor responses, are dependent on the roots that are involved in the abnormal circuits that cause spasticity. These roots must be surgically sectioned. In 1972, Sindou observed that the technique of selective microsurgical destruction of the dorsal root entry zone (for somatogenic or neuropathic deafferentation pain) led to severe hypotonia in the muscles corresponding to the operated medullary segments. Therefore he suggested that this technique should be applied to very severe cases of upper limb spasticity in hemiplegic patients and lower limb spasticity in paraplegic patients [27–29].
1945
1946
115
Destructive neurosurgical procedures for spasticity
In contrast with the lower limb, very few dorsal rhizotomies have been attempted at the cervical level for the upper limb. In his article published in 1913, Foerster [20] also reported on 23 cases of spastic paralysis of the upper limb treated by resection of the dorsal roots from C4 to T2, with the exception of C6. He concluded that for spastic hemiplegias of the upper limb: ‘‘in the majority, the result was not good, satisfactory improvement being obtained in only a few cases; therefore we do not recommend dorsal rhizotomy as a valuable procedure for spasticity of the upper limb.’’ Many years later, Kottke [30] and Heimburger et al. [31] presented the possibility of reducing the spasticity of upper limbs by dorsal rhizotomy at the cervical level: from C1 to C3. The dorsal roots of C4 were not sectioned to avoid affecting the diaphragm, as well as the roots of C5 to T1 to not alter the sensory function of the upper limb. These superior cervical rhizotomies could lead to a slight reduction in the upper limb spasticity, potentially due to an indirect effect through the loss of tonic neck reflexes. Only with the development of selective lesions in the dorsal root entry zone could the surgical procedures on spinal cord afferents produce significant hypotonia of the upper limb [29,32]. These techniques induce severe hypotonia and lead to significant hypoesthesia of operated territories, which must be considered before surgery. In 1945, Munro [33] suggested sectioning of ventral roots from the last thoracic roots to the first sacral root, to treat irreducible spasticities with severe spasms. This type of procedure was recommended in cases of spasticity associated with spontaneous hyperactivity of motor neurones, as observed, following anoxia. In such cases, clinical and experimental data show that the attempts to section dorsal roots are ineffective whereas ventral root sectioning abolishes spasms. In parallel with open surgical techniques, intrathecal chemical rhizotomies, originally
introduced for the treatment of pain associated with cancerous lesions, were also used for the treatment of severe spasticity. Alcoholic solutions were used in 1931 by Doglioti [34] to treat painful cancerous lesions and later in 1953 by Guttman for the treatment of disabling spastic paraplegia [35]. Alcohol was then replaced by phenol (hyperbaric solution) as a neurolytic agent for pain sedation. In 1959, Nathan [36] used this chemical agent for the treatment of spasticity. These techniques are not frequently used as it is very difficult to direct the product onto the roots that are involved in the spasticity, and does not allow a good selectivity of sensitive fibers responsible for the tonus. This explains the frequent appearance of undesirable effects on motor and sphincter functions, and the fact that the results do not always persist in the long term. Percutaneous radiofrequency rhizotomy with thermocoagulation was also introduced to treat chronic pain [37,38]. This technique was then applied to certain spasticities, namely in cases of neurogenic detrusor hyperreflexia of the bladder that are treated by sacral rhizotomy through the sacral foramens [39]. Percutaneous thermocoagulation was then performed on lumbar roots, in particular L2-L3 for the treatment of flexion-adduction of the spastic hip [40]. In 1951, Bischof first described the longitudinal myelotomy [41]. His aim was to interrupt the spinal reflex arc between the ventral and dorsal horns by a vertical coronal incision performed laterally from one side of the spinal cord to the other, from L1 to S1 in cases of total paraplegias. Pourpre [42] modified Bischof ’s myelotomy technique to avoid complete interruption of cortico-spinal fibers. Through a T9 to L1 laminectomy, the procedure consisted of making a posterior longitudinal sagittal incision prior to performing a cruciform myelotomy by transversal incision on either side using a stylet with a right angled extremity. The purpose of this surgically performed lesion was to interrupt the
Destructive neurosurgical procedures for spasticity
spinal reflex arc between the ventral and dorsal horns without sectioning the distribution fibers connecting the pyramidal tract to the motorneurons of the ventral horn. Later on, this technique was popularized by Laitinien and Singounas [43] who created a special surgical knife to perform less damaging myelotomies. This method was then widely used in the treatment of total paraplegia, especially in cases associated with triple flexion and severe sphincter disorders.
The Techniques of Dorsal Rhizotomies Dorsal rhizotomies are the most frequently used methods for children with cerebral palsy. Their techniques will therefore be developed hereafter. Surgical approaches for dorsal rhizotomies may be significantly different from one team to another. The most classical technique – described by Peacock and Arens [47] and Abbott et al. [48] – is as follows. The L1 through S1 laminae are removed using a power saw, which allows replacement of the laminae at the end of the procedure. Bipolar stimulation of the sensory roots (or rootlets), usually of L2 through S1 bilaterally, is carried out using a multichannel EMG recorder to allow electrical monitoring outside the myotome of the root being stimulated. In addition, it is important to palpate the leg muscles for evidence of contraction. Roots which, when stimulated, cause either muscle activity outside of their myotome or activity lasting after cessation of the stimulus current are deemed abnormal and they are separated into their rootlets. The rootlets are in turn stimulated and the same criteria are used to judge their normality. Abnormally responsive rootlets are the candidates to be cut. To limit the extent of the approach [49], (> Figure 115-11), we preferred to perform a limited osteoplastic laminotomy, using a powersaw, in one single piece, from T11 to L1, that will
115
be replaced at the end of procedure and fixed with wires. By doing so, the dorsal (and ventral) L1, L2 and L3 roots can be identified from their muscular responses, evoked by electrical stimulation performed intradurally just before entry into their dural sheaths. The dorsal sacral rootlets are recognized at their entrance into the dorso-lateral sulcus of the conus medullaris. The landmark between S1 and S2 medullary segments is located at 30 mm approximately from the exit of the tiny coccygeal root from the conus. The dorsal rootlets of S1, L5 and L4 can be identified from the motor responses evoked under stimulation; the sensory roots for the bladder (S2-S3) by monitoring vesical pressure, and those for the anal sphincter (S3-S4) by rectomanometry (or simply using the finger introduced into the patient’s rectum) or EMG recordings. Surface spinal cord Somatosensory Evoked Potentials (SEP) recordings from tibial nerve (L5-S1) and pudendal nerve (S1-S3) stimulation may also be helpful. For surgery to be effective, a total amount of 60% of dorsal rootlets must be cut, (with a different quantity cut according to the level and function of the roots involved). In addition, the correspondence of the roots with the muscles having harmful spasticity or useful postural tone must be considered in determining the amount of rootlets to be cut; in most cases L4 (which predominantly gives innervation to the quadriceps femoris) has to be preserved. To reduce the invasiveness of the approach, in 2001 we designed the ‘‘staged interlaminar (IL) approach’’ [49], (> Figure 115-12), the level(s) depending on the roots to be targeted according to the preoperative chart (i.e., the program elaborated with the rehabilitation team). The lumbosacral spine is approached posteriorly on the midline, so that the preselected interlamina spaces can be reached. For the L2 dorsal root, interlaminar approach should be L1-L2, for L3:L2-L3, for L4: L3-L4 for L5: L4-L5, for S1 and/or S2: L5-S1, as shown in figure. After resecting the flavum ligament, the interlaminar space has to be en-
1947
1948
115
Destructive neurosurgical procedures for spasticity
. Figure 115-11 Lumbo-sacral dorsal rhizotomy for children with cerebral palsy (personal technique with osteoplatic laminotomy). The technique consists of performing a limited osteoplastic laminotomy using a power saw, in one single piece, from T11 to L1 (left). The laminae will be replaced at the end of the procedure and fixed with wires (right). The dorsal (and ventral) L1, L2 and L3 roots can be identified by means of the muscular responses evoked by electrical stimulation performed intraduraly just before entry into their dural sheaths. The dorsal sacral rootlets are recognized at their entrance into the dorso-lateral sulcus of the conus medullaris. The landmark between S1 and S2 medullary segments is located at 30 mm approximately from the exit of the tiny coccygeal root from the conus. The dorsal rootlets of S1, L5 and L4 can be identified by their evoked motor responses; the sensory roots for bladder (S2-S3) by monitoring vesical pressure, and those for the canal sphincter (S3-S4) by rectomanometry (or simply using finger introduced into the patient‘s rectum) or EMG recordings. Surface spinal cord SEP recordings from tibial nerve (L5-S1) and pudendal nerve (S1-S3) stimulation may also be helpful. For the surgery to be effective a total amount of 60% of dorsal rootlets must be cut, (with a different quantity cut according to the level and function of the roots involved). Also the correspondence of the roots with the muscles having harmful spasticity or useful postural tone must be considered in determining the amount of rootlets to be cut; in most cases L4 (which predominantly gives innervation to the quadriceps femoris) has to be preserved
larged by resecting (in the order of) the lower half of the sus-jacent and the upper half of the subjacent laminae. Then the dural sheath is opened on the midline over two centimeters. L2 and L3 roots can be reached through the L1-L2 opening, L4 and L5 roots through the L3-L4 IL opening, S1-S2 roots through the L5-S1 IL opening. Generally, the dorsal rootlets (five on average) are easily identified, as they are grouped posteriorly to the ventral root, separated from the latter by an arachnoid fold. Motor responses evoked by stimulation (with a bipolar electrical stimulator) are tested for ventral and then dorsal root (rootlets). The
threshold for obtaining motor responses by stimulating the dorsal root (rootlets) is at least three times the one for the corresponding ventral root (rootlets). After identification of the dorsal rootlets, the appropriate number of them is divided, between one third and two third of the overall dorsal rootlets, according to the preoperative chart. Then, the dural incision is sutured in a tight fashion and the dural suture line is covered with fat tissue harvested from the subcutaneous layer.
Destructive neurosurgical procedures for spasticity
. Figure 115-12 ‘‘Staged interlaminar approach’’
115
level and at a 45 angle at the lumbosacral level. Bipolar coagulation is performed ventrolaterally at the entrance of the rootlets into the dorsolateral sulcus, along all the spinal cord segments selected for operation (> Figure 115-13).
Orthopedic Surgery
The Techniques of Microsurgical DREZ otmy Surgery in the Dorsal Root Entry Zone (DREZ) of the spinal cord was introduced in 1972 for the treatment of persistent pain [27,28,44]. This technique also induced important hypotonia and was therefore used by the authors in cases of severe focalized spasticity [29,45,46], not only in the lower limbs [45,46] but also in the upper limbs [32]. The purpose of the microsurgical DREZotomy technique is to interrupt preferentially both small caliber (nociceptive) and large caliber (myotatic) tonigenic fibers of the dorsal roots situated laterally and in the middle of the entry zone, respectively. The surgical lesion must partially, if not totally, preserve the medial large caliber fibers that project to the dorsal horn of the cord and are located medially in the entry zone. The surgical target includes most of the dorsal horn where the circuits and neurons that activate the segmental circuitry of the spinal cord are situated. This technique is discussed in more detail in a separate chapter in this book. Briefly, the procedure consists of 2–3-mm deep microsurgical incisions at a 35 angle at the cervical
Orthopedic surgical procedures may reduce spasticity by means of muscle relaxation resulting from tendon lengthening. The current techniques available for correcting excessive shortness of the muscle tendon assembly are muscular desinsertion, myotomy, tenotomy, and lengthening tenotomy. Such techniques aim at obtaining a more functional position, avoiding excessive lengthening, which could lead to a decrease in muscular strength. Tendon transfer has a different goal: It normalizes articular orientation disturbed by muscular imbalance. Transfer of spastic muscles must be avoided; if necessary, suppression of spasticity must first be achieved by a neurosurgical procedure osteotomies aim at correcting bone deformation caused by pathology of growth or at treating stiffened joints. Articular surgery (arthrodesis) is indicated only when an osteoarticular deformity cannot be corrected by osterotomy or tendon surgery alone. Arthrodesis must not be performed in children until growth is complete. In children with cerebral palsy, orthopedic procedures are appropriate when tonic imbalance becomes permanent, provoking osteoarticular deformities. In adults, orthopedic procedures can be undertaken to increase comfort in the more severely affected or to improve function in those who have recovered a sufficient level of voluntary control. Whatever the situation, orthopedic surgery must be considered only after spasticity has been reduced.
1949
1950
115
Destructive neurosurgical procedures for spasticity
. Figure 115-13 Technical principles of micro-DREZotomy (MDT) and necessary instruments. Exposure of the dorsolateral aspect of the conus medullaris on the left side. The rootlets of the selected dorsal roots are retracted dorsomedially and held with a (specially designed) ball-tip microsucker (B), which is used as a small hook, to gain access to the ventrolateral part of the DREZ. After division with curved sharp microscissors (S) of the fine arachnoidal filaments that stick the rootlets together with the pia mater, the main arteries running along the dorsolateral sulcus are dissected and preserved, while the smaller ones are coagulated with a pair of sharp bipolar microforceps (F). Then, the incision is performed using a microknife (K), made with a small piece of razorblade inserted within the striated jaws of a curved razorblade holder (K). On average, the cut is a 35 angle descending to a depth of 2–3 mm. The surgical lesion is completed by performing microcoagulations under direct magnified vision (at a low intensity) inside the incision, down to the apex of the dorsal horn. These microcoagulations are made by means of the special sharp bipolar forceps (F), which is insulated except at the tip over 5 mm and graduated every millimeter
Decision-Making in Adults Generally patients do not complain about spasticity; they are more likely to be aware of stiffness, deformity and limitations in functional abilities. ‘‘Stiffness’’ is a useful term because it is widely understood by both clinicians and patients and it does not imply a specific cause. After a period of time, the patients will have a mixture of spasticity and muscle shortening or contracture. In any discussion of spasticity management, an agreed terminology is important in recognizing two principal components of muscle stiffness: (1) ‘‘Dynamic’’ shortening of muscles caused by spasticity. Such patients exhibit hyperreflexia, clonus, and a velocity-dependent resistance to passive joint motion. (2) ‘‘Fixed’’ shortening of muscles described as contracture. They are much less velocity-dependant and remain present under local blocks or anesthesia.
Spasticity should not be treated just because it is present; it should be treated because it is harmful. Patients may use spasticity in functional activities. An extensor pattern in lower limb(s) may aid standing transfers. In this scenario, ‘‘successful’’ spasticity management, if measured by reduction in tone and improved range of motion, might reduce rather than enhance function. Hence the prime goal of spasticity management must be improved function, if possible, and stop or prevent deformities. How we differentiate dynamic from fixed deformities is of prime importance before deciding any surgical treatment, neurosurgical or orthopedic or both. The dynamic range of motion measures are useful starting points, supplemented with instrumental measures of spasticity and its effects on function, such as motion analysis. Methods of spasticity management can be classified according to whether they are focal or general in effect and as to whether the effects are
Destructive neurosurgical procedures for spasticity
permanent or temporary. Within this four-way matrix, practical clinical guidelines may be derived as illustrated in the figure (> Figure 115-14) [50].
115
. Figure 115-14 Algorithms for treating disabling hyperspasticity, paraplegic (top), and hemiplagic (middle and bottom) patients
Intra-Thecal Baclofen Therapy (ITB) ITB therapy can be preceded by a test to screen for adequate response to the medication. The common standard procedure is as follows: the patient receives a test bolus of 25–50 mg intrathecal baclofen via lumbar puncture (or via a temporary intrathecal lumbar catheter connected to a subcutaneous access reservoir to be punctured). In the absence of a positive response, indicated by a two-point reduction in Ashworth score 4–8 h following drug administration, the bolus dose is increased in 25 mg increments up to a maximum bolus of 100 mg. Once a positive response is observed without accompanying unacceptable loss of function, the patient is considered to be a candidate for pump implantation. However, the bolus dose response is a poor guide to the likely daily infusion rate which will be subsequently needed. In addition, the ‘‘bolusmethod’’ is entailed by ‘‘false-negative responses’’ in the sense that the bolus can produce a brutal or exaggerated loss of motor power and muscle tone, which might be interpretated by the patient as a decrease in functional status. Because this could lead to an absence of the indication for pump implantation, especially in patients with the ability to walk, the bolus-test should be replaced by a continuous infusion test, using an external automatic injection pump connected to a line punctured into a subcutaneous reservoir. The test should last from a few days to an entire week so that functional capabilities can be reliably evaluated. The initial post-implantation infusion dose depends, in part, on the effective screening dose. Typically, the initial starting dose is double the effective screening dose. The dose is then increased dailyby 10–30% until the desired effect
is achieved. The most useful criteria for dose adjustment is the effective suppression of the reflexes, i.e., tendon jerk, clonus, spasms, cramps, and the decrease of muscle tone. Once the effective dose has been ascertained and stabilized, the administration of the drug can be finetuned. A programmable pump allowing cyclical
1951
1952
115
Destructive neurosurgical procedures for spasticity
dose adjustments makes it possible to provide levels that correlate with the daily variability of spastic symptoms. A serious risk of intrathecal baclofen administration is overdose, which could be irreversible because of the lack of true baclofen antagonists; therefore this technique requires great care. Other complications include mechanical catheter migration or occlusion and infection, which require revision or removal of the system, respectively. The advantage of the technique is the reversibility of its effects, but high cost, necessity of periodic refilling and reprogramming, and also geographic dependence are serious limitations to this conservative method. ITB is particularly indicated for patients with severe spasticity of spinal cord origin, especially if painful spasms are present, as in advanced multiple sclerosis or after spinal cord injury when physical therapy and rehabilitation did not succeed in preventing the appearance of harmful spasticity. A multicenter study has been carried out to determine doses at 12 months, safety and efficacy of ITB in spasticity of spinal origin (205 patients studied). Doses were between 167 and 462 mg/day (average: 298 mg). Ashworth score decreased from 3–4 to 0.5–1.8 according to the series. ITB can also be indicated for hyperspasticity due to brain stem lesions ITB has also entered over the recent past years the neurosurgical armamentarium to treat cerebral palsy patients. Due to the big size of the available pump, ITB cannot be performed in children under 6 year age. Children with associated choreo-athetosis, hypotonia of neck and trunk, obesity, poor motivation and/or severe multiple deformities, are poor candidates for ITB. For cerebral palsy patients, it must be emphasized that the adequate doses, i.e., the ones effective on the excess of tone without producing motor weakness, are often difficult to establish.
Neuro-destructive Procedures When spasticity cannot be controlled by conservative methods or by botulinum toxin injections, ablative procedures must be considered. The surgery should be performed so that excessive hypertonia is reduced without suppression of useful muscular tone or impairment of the residual motor and sensory functions. Therefore, neuroablative techniques must be as selective as possible. Such therapeutic lesions can be performed at the level of peripheral nerves, spinal roots, spinal cord or the dorsal root entry zone.
Peripheral neurotomies Neurotomies are indicated when spasticity is localized to muscles or muscular groups supplied by a single or a few peripheral nerves that are easily accessible. To help the surgeon decide if neurotomy is appropriate, temporary local anesthetic block of the nerve (with long-lasting bupivacaı¨ne) can be useful. Such a test can determine if articular limitations result from spasticity or musculo-tendinous contractures and/or articular ankyloses (only spasticity is decreased by the test). In addition, these tests give the patient an idea of what to expect from the operation. Botulinum toxin injections may also act as a ‘‘prolonged’’ test for several weeks or months. 1.
For spasticity in the lower limbs, neurotomies of the tibial nerve at the popliteal region and of the obturator nerve just below the subpubic canal, are the most common for the so-called spastic foot and for spastic flexionadduction deformity of the hip, respectively. Selective neurotomy of the branches to the knee flexors (hamstrings) can also be performed at the level of the sciatic trunk through a short skin incision in the buttock. For spastic hyperextension of the first toe (so-called permanent Babinski sign), a selective neurotomy of the branch(es) of the
Destructive neurosurgical procedures for spasticity
deep fibular nerve to the hallux extensor can be useful. 2. Neurotomies can also be indicated for spasticity in the upper limb. Selective fascicular neurotomies can be performed in the musculocutaneous nerve for spastic elbow flexion, and in the median (and ulnar) nerve for spastic hyperflexion of the wrist and fingers. The last procedure – which consists of sectioning the branches to the forearm pronators, wrist flexors and extrinsic finger flexors – is indicated for spasticity in the wrist and the hand, the aim being to open the hand and improve prehension. When the fascicular organization of the median and ulnar nerves does not allow one to accurately differentiate motor from sensory fascicles at the level of their trunks, it is necessary to dissect the motor branches after they have left the nerve trunk in the forearm. Special care must be taken with the sensory fascicles, to avoid painful manifestations. Neurotomies of brachial plexus branches have also been developed for treating the spastic shoulder. The Pectoralis major muscle and Teres major muscle are the main muscles implicated in this condition. This excess spasticity restrains the active (and passive) abduction and external rotation of the shoulder. 3. Basically, selective neurotomies are not only able to reduce excess spasticity and prevent form deformity, but also to improve motor function by re-equilibrating the tonic balance between agonist and antagonist muscles. This is true especially for the spastic foot with equino-varus (> Figure 115-10). With regard to the spastic hand which is a very difficult problem to deal with, a functional benefit in prehension can be achieved only if patients retain a residual motor function in the extensor and supinator muscles, together with a sufficient residual sensory function. If these conditions are not present, only better comfort and better cosmetic aspect can be achieved.
115
Surgery in the dorsal root entry zone When spasticity affects – severely – an entire limb, microsurgical DREZotomy is preferred. For patients with paraplegia, the L2-S5 segments are approached through a T11-L2 laminectomy, whereas for the hemiplegic upper limb, a C4-C7 hemilaminectomy with conservation of the spinous processes is sufficient to reach the C5-T1 segments. MDT is indicated in paraplegic patients, especially when they are bedridden as a result of disabling flexion spasms, and in hemiplegic patients with irreducible and/or painful hyperspasticity in the upper limb. MDT can also be applied to treat neurogenic bladder with uninhibited detrusor contractions resulting in voiding around a catheter. Orthopedic Surgery Whatever the situation and the etiology may be, orthopedic surgery must be considered only after spasticity has been reduced by physical and pharmacological treatments and, when necessary, by neurosurgical procedures. Guidelines for surgical indications have been detailed elsewhere [50] and are summarized in > Figure 115-14. The general rule is to tailor individual treatments as much as possible.
Decision-Making in Children Cerebral palsy is a group of disorders. It is permanent but not unchanging; it involves disorders of movement and/or posture and of motor function; it is due to a non-progressive interference/lesion/ abnormality; this interference/lesion/abnormality is in the developing/immature brain [51]. Spasticity in children as in adults, can be useful to the function or on the contrary can increase disability. Nowadays we use efficient treatment against spasticity in children with cerebral palsy: Botulinum toxin, dorsal rhizotomy, intrathecal Baclofene, selective neurotomy. These treatments are often associated with or-
1953
1954
115
Destructive neurosurgical procedures for spasticity
thopedic surgery at a second stage. The choice between these different treatments is more difficult in children than in adults because a child is in constant development, its orthopedic status changes with increase of growth, in particular, during puberty. To take a decision about which treatment against spasticity, when to propose it, with which orthopedic project after treatment, we need to project into the future, to extrapolate musculoskeletal contractures and their bad consequences but we also need to extrapolate the positive functional evolution with spontaneous psychomotor development. So a global assessment of the psychomotor development of the child is needed associated with an analytic assessment of spasticity, an orthopedic status and a functional level.
The first stage is clinical observation to obtain a global idea of the child’s function. The second stage is measurement of the range of motion to detect contractures that would be/is accessible to neurosurgical treatment. The third stage is the assessment of spasticity (with the Tardieu scale or the Ashworth scale). The final stage is grading the child on the Gross Motor Function Measure and observing the evolution of gross motor function with time (> Figure 115-15). There are several effective neurosurgical treatments for spasticity in children with cerebral palsy [52] (> Figure 115-16). Pursuing neurosurgical treatment requires a consensus about the surgical goal among the surgeon, child, family, and re-education team. The treatment of spasticity must be considered before muscular contractures, and is part of a therapeutic program that
. Figure 115-15 Time-course (in years = anne´es) of five dimensions on the Gross Motor Function Measure (EMFG) in a child with cerebral palsy. Lying (=couche´); Sitting (=Assis); Quadraped crawling (4 pattes); Standing (=debout); Walking/ Running/Jumping (=marche, course et saut); Global score (=score global). Note that Dorsal Rhizotomy (DR) was indicated after cast and Botulinum toxin injections + cast failed. Also note that score improved significantly after DR was performed
Destructive neurosurgical procedures for spasticity
. Figure 115‐16 Algorithms for treatment of disabling spasticity in cerebral palsy
extends over several years. For general spasticity of lower limbs, dorsal rhizotomy or intrathecal baclofen administration are proposed. Dorsal rhizotomy is proposed when definitive action targeted on certain muscular groups is preferred. For focal spasticity, botulinum toxin injection permits delaying surgery until the child is old enough to undergo a selective neurotomy. For upper limb spasticity, botulinum toxin is first proposed. We consider the use of botulinum toxin as a test before neurosurgery. The muscles are injected to simulate the nerve that would be affected by a selective neurotomy. This test allows the child to appreciate the benefit that he or she would enjoy from a selective neurotomy. Lower Limb Spasticity
When a child with global spasticity is assessed every 6 or 12 months with the Gross Motor Function Measure, his future motor function can be predicted and his progress or deterioration observed. The projected treatment is based on a realistic perspective, and we can appreciate if treatment can be delayed (if the curve shows increased function) or if urgent surgical treatment is needed (if the curve plateaus or decreases) (> Figure 115-15).
115
If spasticity of the two lower limbs is global, techniques with global action are discussed: dorsal rhizotomy or intrathecal baclofen. The latter imposes a medical visit every 6 months to fill the pump. Thus, the child and family must be motivated to comply, and probably need to live relatively close to the neurosurgical center. Most families consider the preliminary tests to be an advantage. However, the volume of the pump is a great obstacle in young children. Dorsal rhizotomy can be proposed before 6 years. This procedure permits a targeted action in accordance with the importance of spasticity and the involved muscular groups. When focal spasticity involves the gastrocnemius and soleus muscles, botulinum toxin can be proposed as a good complement to physiotherapy and a plaster cast. This approach enables neurosurgical treatment to be delayed until the child reaches the age for selective neurotomy. After neurotomy, spasticity tends to recur, most frequently in younger children. When spasticity is focal on the adductors, botulinum toxin can again be a good complement to physiotherapy and posture. Sometimes this therapy is sufficient to avoid an obsturator neurotomy, regardless of whether associated with a tenotomy of the adductors, to prevent dislocation of the hip. In a child who can walk, an obturator neurotomy must be pursued cautiously because it can damage the child’s strength and decrease function. Botulinum toxin can be tried on several muscles at one time if spasticity is not very focal but still not global. If many muscles are to be injected, general anesthesia is necessary, particularly for the iliopsoas. The dose cannot exceed a prescribed level, which limits the effectiveness of treatment. Moreover, its action is rarely definitive; it must be repeated every 0–12 months. Upper Limb Spasticity
For the first stage of treatment of the upper limb, botulinum toxin is proposed. The muscles of the upper limbs are small. Even if quite a few must be injected, the maximum allowable dose is rarely
1955
1956
115
Destructive neurosurgical procedures for spasticity
too small to be insufficient. These injections can be considered as a treatment and can be repeated every 6 or 12 months. However, the multiple visits required can be a constraint. Furthermore, patients can develop an immuno-resistance that decreases the effectiveness of treatment with time. It seems more realistic to use the Botulinum toxin as a test before pursuing neurosurgical treatment. Consequently, we prefer to inject the muscles to simulate the outcome of a selective neurotomy. This strategy allows the child to appreciate the benefit that can follow a selective neurotomy. If spasticity involves the shoulder, elbow, wrist and fingers, a posterior rhizotomy of the cervical roots is possible. Dystonia is rarely directly improved by these treatments. However, the decreased range of motion and strength of the dystonic movement can improve the global cosmetic and functional aspects of the upper limb. Surgery in the child requires a multidisciplinary team and the full participation of the child and parents.
Conclusion Neurosurgery for spasticity needs – in essence – a multidisciplinary approach [53,54].
References 1. Sindou M. Neurosurgical management of disabling spasticity. In: Spetzler RF, editor.‘‘Operative techniques in neurosurgery, vol. 7’’. Philadelphia, PA: Elsevier; 2004. p. 95-174. 2. Stoffel A. The treatment of spastic contractures. Am J Orthop Surg 1912;10:611-44. 3. Gros C. La chirurgie de la spasticite´. Neurochirurgie 1972;23:316-88. 4. Gros C, Frerebeau P, Benezech L, et al. Selective radicular neurotomy. In: Simon L, editor. ‘‘Actualities in functional physical therapy’’. Paris: Masson; 1977. p. 230-5.
5. Sindou M, Mertens P. Selective neurotomy of the tibial nerve for the treatment of the spastic foot. Neurosurgery 1988;23:738-44. 6. Mertens P, Sindou M. Selective peripheral neurotomies for the treatment of spasticity. In: Sindou M, Abbott R, Keravel Y, editors. ‘‘Neurosurgery for spasticity. A multidisciplinary approach’’. New York: Springer; 1991. p. 119-32. 7. Decq P. Peripheral neurotomies for the treatment of focal spasticity of the limbs. Neurochirurgie 2003;49:293-305. 8. Sindou M, Simon F, Mertens P, Decq P. Selective peripheral neurotomy (SPN) for spasticity in childhood. Childs Nerv Syst 2007;23:957-70. 9. Mertens P, Sindou M. Surgical management of spasticity. In: Barnes, MP, Johnson GR, editors. ‘‘Clinical management of spasticity’’. Cambridge: Cambridge University Press; 2001. p. 239-65. 10. Decq P, Filipetti P, Lefaucheur JP. Evaluation of spasticity in adults. Oper Tech Neurosurg 2004;7:100-7. 11. Decq P, Cuny E, Filipetti P, Feve A, Keravel Y. Peripheral neurotomy in the treatment of spasticity. Indications, techniques and results in the lower limbs. Neurochirurgie 1998;44:175-82. 12. Decq P, Shin M, Carrillo-Ruiz J. Surgery in the peripheral nerves for lower limb spasticity. Oper Tech Neurosurg 2004;7:136-46. 13. Decq P, Cuny E, Filipetti P, Keravel Y. Role of soleus muscle in spastic equinus foot. Lancet 1998;11(352):118. 14. Brunelli G, Brunelli F. Selective microsurgical denervation in spastic paralysis. Ann Chir Main 1983;2:277-80. 15. Maarrawi J, Mertens P, Sindou M. Surgery in the peripheral nerves for upper limb spasticity. Oper Tech Neurosurg 2004;7:147-52. 16. Maarrawi J, Mertens P, Luaute J, Vial C, Chardonnet N, Cosson M, Sindou M. Long-term functional results of selective peripheral neurotomy for the treatment of spastic upper limb: prospective study in 31 patients. J. Neurosurg 2006;104:215-25. 17. Decq P, Filipetti P, Feve A, Djindjian M, Saraoui A, Keravel Y. Peripheral selective neurotomies of the brachial plexus branches for the spastic shoulder. Anatomical study and clinical results in 5 patients. J Neurosurg 1997;86:648-53. 18. Sherrington CS. Decerebrate rigidity and reflex coordination of movements. J Physiol 1898;22:319-32. 19. Foerster O. Uber eine neue operative methode der behandlung spasticher La¨hmungen mittels resektion hinter Rickenmark swurzeln. Z Orthop Chir 1908;22:203-23. 20. Foerster O. On the indications and results of the excision of posterior spinal nerve roots in men. Surg Gynecol Obstet 1913;16:463-74. 21. Gros C, Frerebeau P, Benezech J, Privat JM. Neurotomie radiculaire se´lective. In: Simon L, editor. ‘‘Actualite´s en re´e´ducation fonctionnelle’’. Paris: Masson; 1977. p. 230-5.
Destructive neurosurgical procedures for spasticity
22. Gros C. Spasticity – clinical classification and surgical treatment. In: Krayenbu¨hl, editor. ‘‘Advances and technical standards in neurosurgery’’, vol. 6. New York: Springer; 1979. p. 55-97. 23. Gros C, Ouaknine G, Vlahovitch B, Frerebeau PH. La radicotomie se´lective poste´rieure dans le traitement neurochirurgical de l’hypertonie pyramidale. Neurochirurgie 1967;13:505-18. 24. Privat JM, Benezech J, Frerebeau P, Gros C. Sectorial posterior rhizotomy, a new technique of surgical treatment for spasticity. Acta Neurochir (Wien) 1976;35:181-95. 25. Fraioli B, Guidetti B. Posterior partial rootlet section in the treatment of spasticity. J Neurosurg 1977;46: 618-26. 26. Fasano VA, Barolat-Romana G, Ivaldi A, Squazzi A. La radicotomie poste´rieure fonctionnelle dans le traitement de la spasticite´ ce´re´brale. Neurochirurgie 1976;22:23-34. 27. Sindou M. Etude de la jonction radiculo-me´dullaire poste´rieure. La radicellotomie se´lective poste´rieure dans la chirurgie de la douleur. Lyon: The`se Me´decine; 1972. p. 182. 28. Sindou M, Quoex C, Baleydier C. Fiber organization at the posterior spinal cord-rootlet junction in man. J Comp Neurol 1974;153:14-26. 29. Sindou M, Fischer G, Goutelle R, Schott B, Mansuy L. La radicellotomie poste´rieure se´lective dans le traitement des spasticite´s. Rev Neurol 1974;30:201-15. 30. Kottke J. Modification of athetosis by denervation of the tonic neck reflexes. Dev Med Child Neurol 1970;12:236-7. 31. Heimburger RF, Slominski A, Griswold P. Cervical posterior rhizotomy for reducing spasticity in cerebral palsy. J Neurosurg 1973;39:30-4. 32. Sindou M, Mifsud JJ, Boisson D, Goutelle A. Selective posterior rhizotomy in the dorsal root entry zone for treatment of hyperspasticity and pain in the hemiplegic upper limb. Neurosurgery 1986;18:587-95. 33. Munro D. The rehabilitation of patients totally paralysed below waist: anterior rhizotomy for spastic paraplegia. Engl J Med 1945;233:456-61. 34. Dogliotti A. Traitement des syndromes douloureux de la pe´riphe´rie par l’alcoolisation sous-arachnoı¨dienne des racines poste´rieures a` leur e´mergence de la moelle ´epinie`re. Presse Me´d 1931;39:1249-52. 35. Guttman L. The treatment and rehabilitation of patients with injuries of the spinal cord. In: Cope 2, editor. ‘‘History of the second world war,’’ vol surgery. England: Her Majesty‘s Stationnery Office; 1953. p. 422-516. 36. Nathan PW. Intrathecal phenol to relieve spasticity in paraplegia. Lancet 1959;2:1099-102. 37. Lazorthes Y, Verdie JC, Lagarrigue J. Thermocoagulation percutane´e des nerfs rachidiens a` vise´e antalgique. Neurochirurgie 1976;22:445-53.
115
38. Uematsu S, Udvarhelyi JB, Benson D, Siebens A. Percutaneous radiofrequency rhizotomy. Surg Neurol 1974;2: 319-25. 39. Young B, Mulchi JJ. Percutaneous sacral rhizotomy for neurogenic detrusor hypereflexia. J Neurosurg 1980;53: 85-7. 40. Segnarbieux F, Frerebeau Ph. The different (open surgical, percutaneous thermal, and intrathecal chemical) rhizotomies for the treatment of spasticity. In: Sindou M, Abbott R, Keravel Y, editors. ‘‘Neurosurgery for spasticity: a multidisciplinary approach’’. New York: Springer; 1991. p. 133-9. 41. Bischof W. Die longitudinal myelotomie. Zbl Neurochir 1951;11:79-88. 42. Pourpre MH. Traitement neuro-chirurgical des contractures chez les paraple´giques post-traumatiques. Neurochirurgie 1960;6:229-36. 43. Laitinen L, Singounas E. Longitudinal myelotomy in the treatment of spasticity of the legs. J Neurosurg 1971;35:536-40. 44. Sindou M, Fischer G, Goutelle A, Mansuy L. La radicellotomie poste´rieure se´lective. Premiers re´sultats dans la chirurgie de la douleur. Neurochirurgie 1974;20:391-408. 45. Sindou M, Millet MF, Mortamais J, Eyssette M. Results of selective posterior rhizotomy in the treatment of painful and spastic paraplegia secondary to multiple sclerosis. Appl Neurophysiol 1982;45:335-40. 46. Sindou M, Jeanmonod D. Microsurgical DREZotomy for the treatment of spasticity on pain in the lower limbs. Neurosurgery 1989;24:655-70. 47. Peacock WJ, Arens LJ. Selective posterior rhizotomy for the relief of spasticity in cerebral palsy. S Afr Med J 1982; 62:119-24. 48. Abbott A, Forem SL, Johann M. Selective posterior rhizotomy for the treatment of spasticity. Childs Nerv Syst 1989;5:337-46. 49. Sindou M. Radicotomies dorsales chez l’enfant. Neurochirurgie 2003;49:312-23. 50. Sindou M, Mertens P. Decision-making for neurosurgical treatment of disabling spasticity in adults. Oper Tech Neurosurg 2004;7:113-18. 51. Hodgkinson I, Berard C. Assessment of spasticity in pediatric patients. Oper Tech Neurosurg 2004;7:109-11. 52. Hodgkinson I, Sindou M. Decision-making for treatment of disabling spasticity in children. Oper Tech Neurosurg 2004;7:120-3. 53. Sindou M, Abbott A, Keravel Y, editors. ‘‘Neurosurgery for spasticity: a multidisciplinary approach’’. New York: Springer; 1991. 54. Decq P, Mertens P, et al. La Neurochirurgie de la Spasticite´. Neurochirurgie 2003;49:135-416.
1957
110 Diagnosis and Medical Management of Cervical Dystonia R. Bhidayasiri . D. Tarsy
Torticollis is not a diagnosis but a physical sign which is characterized by twisting contractions of the neck muscles resulting in forced turning of the head. Therefore, there are many causes of torticollis, both dystonic and non-dystonic in origin. Although primary cervical dystonia (CD) represents the most common cause of torticollis, other non-dystonic (pseudodystonia) causes of torticollis, such as congenital disorders or local musculoskeletal changes in the cervical region, may resemble CD. This chapter will discuss the differential diagnosis of torticollis, particularly how to make a diagnosis of CD and how CD can be distinguished from non-dystonic causes of torticollis and other abnormal head postures. In addition, this chapter will cover a wide variety of disorders affecting the central and peripheral nervous systems, which are associated with secondary forms of CD. The last section of the chapter will focus on the medical management of CD. Although unfortunately relatively limited in efficacy, a variety of pharmacologic agents have been tried, including drugs affecting cholinergic, dopaminergic, serotonergic, and g-aminobutyric acid (GABA) systems.
Many Causes of Torticollis: When to Diagnose Cervical Dystonia The diagnostic evaluation of patients with torticollis requires a determination of the following: Electronic supplementary material Supplementary material is available in the online version of this chapter at http:// dx.doi.org/10.1007/978-3-540-69960-6_110 and is accessible for authorized users. #
Springer-Verlag Berlin/Heidelberg 2009
First, physicians have to observe whether the head deviation represents dystonia, non-dystonic torticollis, or some other type of hyperkinetic movement disorder. Second, it must be determined whether the dystonia is localized to the neck region or is a combination of a segmental or generalized pattern of distribution. Third, it must be determined whether the CD is idiopathic or symptomatic (secondary) to an underlying, identifiable cause. Lastly, other associated movement disorders (e.g., tremor) or secondary complications (e.g., cervical spondylosis) should be identified. By definition, torticollis refers to abnormal cervical postures, which are characterized by tonic or intermittent spasms of neck muscles that cause involuntary deviation of the head from the normal position [1,2]. Therefore, there are many causes of torticollis which are either dystonic or non-dystonic in nature [3–5]. The terms ‘‘nondystonic disorders’’ or ‘‘pseudodystonia’’ refer to disorders that are associated with sustained muscle contractions possibly occurring as a reflex mechanism or reaction to some other disturbance such as, for example, a trochlear nerve palsy or hemianopia which result in cocking of the head to improve vision, or Sandifer syndrome in which gastric reflux causes spasmodic retrocollis [6]. On the other hand, CD is a primary focal dystonia which manifests as sustained involuntary contractions of the neck muscles that result in abnormal movements and postures of the head [7]. With these two operational definitions, several non-dystonic causes of torticollis can be broadly
1858
110
Diagnosis and medical management of cervical dystonia
. Table 110‐1 Non-dystonic causes of torticollis (Modified from [5]) 1. Congenital causes 1.1 Malformation of cervical spine – C1–2 articular anomalies – Klippel–Feil syndrome – Spina bifida 1.2. Hypertrophy of cervical muscles 1.3. Arnold–Chiari malformation 2. Acquired causes 2.1. Cervical musculoskeletal abnormalities – Atlanto-axial dislocation – Atlanto-axial subluxation – C2–3 dislocation – Fracture of the clavicle or scapula 2.2. Infections – Nasopharyngeal infections – Osteomyelitis of the cervical spine 2.3. Neurological causes (excluding dystonia) – Syringomyelia – Posterior fossa tumors – Colloid cyst of the third ventricle – Trochlear nerve palsies 2.4. Miscellaneous causes – Sandifer syndrome – Benign paroxysmal torticollis – Spasmus nutans – Psychogenic
divided into congenital or acquired forms (> Table 110-1) [5]. In infancy, congenital causes such as congenital muscular torticollis and fetal deformations predominate. In children, acquired non-dystonic torticollis is more prevalent than dystonia, with the exception of acute dystonia occurring as an adverse reaction to phenothiazines or metoclopramide [8]. In contrast with children, non-dystonic causes of torticollis in adults are relatively uncommon, but include cervical bony dislocations [9], cervical osteomyelitis [10], fractures of the scapula or clavicle, acquired trochlear nerve palsy [11], and posterior fossa tumors, to name just a few [12,13]. Since non-dystonic causes of torticollis may resemble CD, a treating physician should consider the following factors when evaluating a patient with torticollis before concluding that he/she suffers from primary CD. First, the mean age at onset of the various causes of torticollis differs
significantly. While the peak age at onset of primary CD in adulthood is in the fourth or fifth decade, it is very rare for CD to develop in children [1,5,7]. Therefore, the presence of torticollis in children warrants additional investigations, including cervical radiographs, computed tomography, or magnetic resonance imaging [14]. In children, cervical musculoskeletal abnormalities represent the most common cause of painful non-dystonic torticollis [5]. Among these abnormalities, C1–2 rotatory dislocation is the most common spinal abnormality to present with torticollis. In a retrospective study of 288 pediatric patients with torticollis, 53 children (18% of the study population) had a nonmuscular etiology for their torticollis, in which Klippel–Feil anomalies and associated neurologic disorders were the most common [15]. In children without evidence of dystonia elsewhere, a congenital muscular abnormality or infectious cause of acute ‘‘wry neck’’ should also be sought. In infancy, congenital muscular torticollis and fetal deformation are muscular causes which account for most cases of torticollis. Spasmus nutans is a triad of nystagmus, head nodding, and torticollis, with an onset between 4 and 12 months and which usually has a benign course resolving within 2–3 years [16]. Benign paroxysmal torticollis of infancy is another self-limited disorder of unknown etiology, characterized by recurrent episodes of head tilt, vomiting, pallor, and agitation [17]. Second, the presence of additional neurologic signs, such as ocular nerve palsies, musculoskeletal abnormalities, regional abnormalities in the pharyngeal spaces or soft tissues of the neck as well as a temporal relationship of the onset of torticollis and the occurrence of such a sign, may be supportive of non-dystonic torticollis. For example, patients with congenital fourth nerve palsies who experience large vertical fusional amplitudes may develop contralateral head tilt, resembling CD [18,19]. In such cases, the presence of ocular motility imbalance
Diagnosis and medical management of cervical dystonia
with a new onset of torticollis should prompt consideration of prismatic correction to alleviate double vision before embarking on evaluation and treatment of CD [11]. Importantly, abnormal head posture due to mechanical causes may be distinguished from CD by their fixed persistence while awake and during sleep.
Clinical Features and Diagnosis of Idiopathic Cervical Dystonia Middle-aged patients who develop gradual onset of abnormal head posture associated with involuntary twisting or jerking of the head unaccompanied by other neurological findings (apart from head or hand tremor) or marked cervical spinal abnormalities probably suffer from primary idoiopathic CD. The descriptions of the clinical features of CD in this section are of the idiopathic etiology, which is the most frequent cause of CD in adult patients. By contrast, secondary or symptomatic CD occurs in a minority of adult patients, the details of which are discussed in the next section. The onset of idiopathic CD is usually insidious, with patients complaining of a ‘‘pulling’’ or ‘‘drawing’’ sensation in the neck [7,20]. Women are affected 1.5–1.9 times more often than men, with peak onset in the fifth decade [7]. Symptoms usually begin with mild neck stiffness and subtle postural deviations of the head. When the symptoms are carefully analyzed, many patients complain of vague sensory symptoms before the onset of typical involuntary movements. In typical CD patients, pain and involuntary movements are the two most common reported disabling symptoms [21]. Pain initially occurs in 21% of patients with CD [1]. However, as the disease progresses, neck and shoulder pain becomes prominent in 75% of patients [1,22,23]. Pain is much more common in CD than in other adult-onset focal dystonias and is typically located in the posterior cervical region ipsilateral to the direction of head rotation or head tilt, and
110
almost never occurs in the affected sternocleidomastoid muscle [24]. Most patients with CD consider neck pain the major source of disability even when associated with constant headturning [7,20,25]. Although the pain is more likely to be musculoskeletal, painful cervical spondylosis and cervical radiculopathy can complicate long-standing CD [26]. Furthermore, various types of headache have been reported to have a frequency of approximately 50% in patients with CD [27–29]. Apart from chronic tension-type headache and migraine, headache attributed to CD is currently listed in the new classification of the International Headache Society among headache secondary to disorders of the neck [30]. Several abnormal head and neck postures may occur in CD. Deviation may occur in any single plane or combinations of directions in which the head voluntarily moves. Rotational torticollis is a rotation of the chin around the longitudinal axis towards the shoulder. Laterocollis is a lateral tilt of the head in the coronal plane, moving the ear toward the shoulder. Anterocollis and retrocollis are forward or backward deviations of the head in the sagittal plane; anterocollis brings the chin towards the chest and retrocollis elevates the chin and brings the occiput towards the upper back. Sagittal or lateral shift of the base of the neck from midline may also occur. Most surveys have reported a combination of these deviations to be the most common situation, accounting for 66–80% of cases (> Figure 110-1) [7,20]. Among these deviations, rotational torticollis is the most common, with some degree of rotation noted in 82–97% of patients [7,20]. Laterocollis is the second most common deviation observed and is the predominant deviation in 18% of CD patients [31]. A very characteristic position of the head in CD is rotation to one side, upward deviation of the chin, lateral tilt of the head to the opposite side, and ipsilateral shoulder elevation. Isolated ‘‘pure’’ rotational torticollis is present in only 19–37% of CD patients [20,32]. Pure anterocollis
1859
1860
110
Diagnosis and medical management of cervical dystonia
. Figure 110‐1 A patient with idiopathic cervical dystonia showing predominant left laterocollis with mild anterocollis and left shoulder elevation
. Table 110‐2 Major muscles that are involved in various types of cervical dystonia Patterns of deviation Rotational torticollis
Laterocollis
Anterocollis
and retrocollis are less common in idiopathic CD. Notably, pure retrocollis is often present in tardive dystonia caused by neuroleptic drugs [33], whereas isolated anterocollis can be a hallmark of multiple system atrophy [34]. There is no statistically significant preponderance difference in the incidence of right and left deviation [7,20,31,32]. The active dystonic muscles in CD can usually be identified by visual inspection and palpation, which reveal not only the active muscle contractions but also hypertrophy [31]. It is important to understand the anatomy of neck muscles and the way in which abnormal postures relate to contractions of specific cervical muscles (> Table 110‐2) [35–37]. In order to identify the most involved muscles, it is helpful to instruct the patient to close his or her eyes and to allow the head to draw or deviate into the most comfortable position without any active volitional resistance. Observed abnormal head postures can usually be correlated with a predictable pattern of muscle involvement. In rotational torticollis, as in other dystonias, both agonist and antagonist muscles, including contralateral sternocleidomastoid muscle (SCM) and ipsilateral posterior cervical muscles (splenius capitis, longissimus capitis, and oblique capitis inferior), cocontract. Laterocollis is often produced by
Retrocollis
Involved muscles Contralateral sternocleidomastoid Ipsilateral splenius capitis Ipsilateral levator scapulae Ipsilateral trapezius (cervical part) Ipsilateral semispinalis Ipsilateral sternocleidomastoid Ipsilateral splenius capitis Ipsilateral scalenus medius and posterior Ipsilateral trapezius (cervical part) Longus colli Bilateral sternocleidomastoid Bilateral scalenus medius Bilateral infrahyoid Platysma Bilateral splenius capitis Bilateral trapezius Bilateral semispinalis capitis
contraction of ipsilateral SCM (clavicular more than sternal head), levator scapulae, scalenus, splenius, and trapezius muscles. Retrocollis is usually due to involvement of splenius capitis, trapezius, and deeper posterior cervical muscles such as semispinalis. Muscles involved in anterocollis primarily include deep muscle groups such as longus colli and longus capitus which are prevertebral head flexors, together with a smaller contribution form SCM, scalene, and platysma. The abnormal postures in CD can be so dynamic that they sometimes fluctuate during the course of the illness and on rare occasions have even reversed direction [38]. Symptoms are typically absent first thing in the morning (‘‘honeymoon’’ period), but increase with fatigue and get worse later in the day, sometimes further aggravated by fatigue, stress, and anxiety. Watching television, reading, or writing are specific tasks reported to worsen torticollis, and in some patients standing up and walking. Similar to other primary dystonias, symptoms of CD disappear during sleep.
Diagnosis and medical management of cervical dystonia
Sensory Tricks in Cervical Dystonia Once dystonic postures develop, most patients are able to identify the provocative and palliative factors of their dystonia. A characteristic and unique feature as well as a diagnostic clue in this condition is the presence of a ‘‘sensory trick,’’ ‘‘geste antagoniste,’’ or ‘‘gegendruckphenomen,’’ which is well known to temporarily reduce or even abolish dystonic posturing in CD [39,40]. Touching the face and holding the back of the head are the most common sensory tricks and are present in 85% of patients. Other reported sensory tricks include touching or holding the chin, leaning the head against a high-back chair, holding or pulling the hair, and placing something inside the mouth. The tricks may be tactile or proprioceptive. Even elevating the arm without touching the target area with the finger can impressively diminish involuntary muscle spasms [41]. Most patients have more than one effective trick [40,42]. Patients often camouflage their trick by assuming another posture just before applying the sensory trick and often automatically select the trick when dystonic movements are at their peak [43]. Although the presence of sensory tricks in CD is present in up to 70% of patients and may be clinically variable, the mechanism of these tricks is still unknown [44]. Muller et al. reported that 54% of CD patients presented with more than one effective sensory trick, 82% had a reduction of head deviation of at least 30% during the trick, and 14% had equally effective tricks on either side [40]. Improvement in dystonic movements is typically maintained while the trick is performed, but disappears when the geste ends. During the application of a sensory trick, complete or partial suppression of neck muscle electromyographic (EMG) activity, including both phasic and tonic patterns and tremor bursts, has been demonstrated [31]. Suppression is noted more commonly in rotational torticollis
110
and less frequently in retrocollis, laterocollis, or complex patterns of head deviation. Although the cheek is the most effective area on the ipsilateral side, effects of side and location of trick application appear to play a minor role [44]. While these tricks are generally helpful early in the illness in majority of the patients, they tend to lose effectiveness as the disease progresses [40,42]. Careful observation suggests that the sensory tricks are not simply a counterpressure phenomenon. Wissel et al. [45] demonstrated that 13 out of 25 (52%) patients with CD showed marked reduction of EMG activity (50% in at least one muscle) during arm movement, definitely prior to contact between fingers and the facial target area. The most common EMG findings induced by a sensory trick were decrease or extinction of activity in the muscles primarily responsible for the abnormal head posture and in any synergistic muscles in the neck, together with activation of the contralateral antagonistic muscles [46]. The diversity of effective maneuvers suggests that higher sensorimotor integration processes are involved [44,47–49]. Whereas classic sensory trick maneuvers might influence abnormal sensory input or sensorimotor integration in a direct way, a nonsensory trick such as imagination could only interface at a higher level of complex dynamic mechanism and thereby modify movement: for example, movement preparation. This hypothesis was supported by a recent positron emission tomography (PET) study, showing that the application of a trick resulting in a nearneutral head position led to an increased activation mainly of the inferior and superior parietal cortex, confirming the involvement of higher centers of sensorimotor integration where different contralateral sensory modalities are combined to form a cognitive representation of space [50]. Trick application, therefore, encompasses a variety of mechanisms to restore information about the correct head position that is capable of temporarily switching off dystonic activity (see video).
1861
1862
110
Diagnosis and medical management of cervical dystonia
Other Clinical Features in Cervical Dystonia At least one-third of patients with CD have dystonia in other body parts and many exhibit postural upper limb and/or head tremor [7,20,51,52]. The jaw (oromandibular), eyelids (blepharospasm), arm/hand (writer’s cramp), and trunk (axial) are the most frequently affected parts [7]. Dystonic tremor, now considered as a distinct entity, refers to a postural or kinetic tremor in an extremity or body part affected by dystonia and is usually absent during complete rest [53,54]. It is irregular, has a broad range of frequencies (mainly less than 7 Hz), and may produce a horizontal ‘‘no-no’’ tremor, a vertical ‘‘yes-yes’’ tremor, or a mixed direction tremor. It is often but not always most obvious when the patient voluntarily attempts to rotate the head in the direction opposite to the force of dystonia. Apart from dystonic head tremor, essential-like tremor (tremor associated with dystonia) as well as a combination of dystonic and essential tremor (ET) of the head has been reported in 30 and 8% of CD patients, respectively [7,31,55]. Patients with coexistent essential-type tremor tend to continue to exhibit head oscillation, regardless of the direction of the force of the dystonia. Postural hand tremor is observed in at least 20% of CD patients [20,32,55]. Phenomenologically similar to typical ET, arm tremor in CD may have a mechanism different from that seen in patients with ET [55,56]. However, reports of frequent occurrence of ET in family members of CD patients (one-third of patients) suggest a possible link between the two disorders [57–59]. Abnormality in swallowing is also a frequent complaint among CD patients, particularly in patients with severe retrocollis. Approximately 36–50% of CD patients were found to have abnormal swallowing on the basis of clinical assessment, and the number increased to 72% on electrophysiologic evaluation of oropharyngeal swallowing [60,61]. This incidence is even
greater after botulinum toxin injection or after selective rhizotomy [62,63]. Orthopedic and neurologic complications may arise from CD. Although the exact prevalence of orthopedic complications is unknown, most studies have estimated it to be in the range of 18–41.2%. These include cervical spondylosis (> Figure 110-2), disc herniation, vertebral subluxations and fractures, cervical radiculopathy, and cervical myelopathy [26]. Most changes have occurred at the C2/C3 and C3/C4 levels, with greater severity observed on the side to which the head was turned [64]. If not recognized and treated, CD may produce muscle contractures which result in irreversible cervical deformities. Prior orthopedic conditions such as scoliosis may increase the risk of developing CD [65].
. Figure 110-2 Moderate cervical spondylosis in a patient with idiopathic cervical dystonia
Diagnosis and medical management of cervical dystonia
Not limited to sensorimotor symptoms, the impact of CD on psychosocial functions has been observed to include moderate to severe depression, social phobia, and anxiety [25,66,67]. More recently, ‘‘some’’ or ‘‘severe’’ stigma was reported in a majority of surveyed CD patients encompassing avoidance of others, self-consciousness, feelings of unattractiveness, feeling apologetic, and feeling different from others [68].
Natural History of Idiopathic Cervical Dystonia Idiopathic CD has an insidious onset and typically worsens during the first 5 years after onset, with a range of 1 month to 18 years [1]. The condition eventually stabilizes, sometimes associated with a mild phase of improvement that precedes the stabilization. There is often a day-to-day as well as year-to-year fluctuation in the severity of symptoms, which are invariably worse during periods of physical or emotional stress. One of the major concerns among patients with CD is the possibility of spread of dystonia to other body parts. When primary dystonia begins in childhood or adolescence, it often starts in a leg or arm, before progressing and generalizing to involve multiple body regions. However, generalized dystonia rarely starts in the neck or predominately affects the craniocervical region [69,70]. When it begins in adults, symptoms usually first involve the neck, cranial, or arm muscles, and dystonia tends to remain localized [69,70]. In CD, the dystonia may spread beyond the neck, but rarely becomes generalized. In a study involving 72 CD patients for a mean period of 7.7 years, one-third of patients manifested segmental spread to the arms (16.7%), jaw (11.1%), or trunk (6.9%) [51,71]. While CD can progress, spontaneous remission has been reported in approximately 20% of patients although usually not complete or prolonged [51]. Patients with spontaneous remission
110
tend to have an earlier age at onset, compared to those without remission [1,51,72]. Remissions are most likely to occur during the first year of symptoms and in patients with spasmodic or ‘‘jerky’’ dystonia than in patients with constant neck deviation [72]. In addition, persistent spontaneous remission appears to be influenced by whether it occurs before (nonsustained) or after (sustained) 2 years of illness. The duration of torticollis before remission is significantly longer in those with sustained spontaneous remission [51,72]. Unfortunately, nearly all patients relapse within 5 years, and a cycle of remission and relapse occasionally occurs though rarely more than three times [73]. Interestingly, one clinical study reported long-term remission of up to 4 years in six CD patients following botulinum toxin therapy [74]. Despite these reports, it is difficult to interpret which clinical characteristics predict remission and what percentage of patients will experience dystonic progression because of the absence of clear definitions of remission, the heterogeneous populations that have been studied, and differing durations of follow-up.
Secondary Cervical Dystonia Secondary (symptomatic) forms of CD consist of a number of well-defined diseases, many of which express themselves with torticollis, often accompanied by other neurological deficits [75]. The torticollis may be a major or minor component of the clinical picture. In general clinical practice, secondary CD is uncommon, even in movement disorder clinics [76]. In one study of more than 1000 CD patients, not a single case of Wilson’s disease was found, and central nervous system tumors were uncovered in only two patients [73]. However, it is important for physicians to recognize this group of patients in order to identify the underlying pathologic process. From an etiological point of view, secondary
1863
1864
110
Diagnosis and medical management of cervical dystonia
CD can be caused by focal brain lesions of various origins, neurodegenerative disorders, metabolic disorders, and drugs and chemicals that affect the basal ganglia, thalamus, and brainstem [75–78]. Furthermore, secondary CD can develop following peripheral injury [79–81]. Certain clinical features help to distinguish secondary torticollis from idiopathic CD (> Table 110-3). In general, secondary CD should be suspected when patients have additional neurologic, orthopedic, or medical disorders or have a history of drug exposure or trauma prior to onset of CD. In addition, the appearances of fixed immobile dystonia, which are not typical of primary dystonia unless occasionally in advanced stages, absence of sensory tricks, abrupt onset, and rapid progression should raise the possibility of secondary causes [3,82]. Interestingly, structural lesions associated with torticollis were most commonly localized in the brainstem, cerebellum, and cervical spine, with only very few cases due to basal ganglia lesions [76]. Psychogenic CD is rare, but may be suspected when there are inconsistent directions of torticollis, rapid onset of severe dystonia with fixed posture at rest, active resistance to passive manipulation, absence of sensory tricks, spontaneous correction when the patient is distracted, and a background of somatization disorders or secondary gain [79,83–85]. In contrast with idiopathic CD, patients with psychogenic torticolis often have pronounced laterocollis, less often a rotatory torticollis with
. Table 110‐3 Red flags suggestive of secondary cervical dystonia Sudden onset of dystonia Onset in an infant or a child Severe pain Absence of sensory tricks or geste antagoniste Fixed posture Presence of dystonia during sleep Rapid progression of dystonia Additional focal neurological signs
ipsilateral shoulder elevation, and occasionally contralateral shoulder depression [79]. Pain, which is typically associated with psychogenic dystonia, is also a common feature in CD, and therefore of little help in distinguishing psychogenic from idiopathic CD [83]. The craniocervical region is a common anatomical site to be affected by tardive dystonia. Therefore, a prior history of exposure to neuroleptics should be sought, especially when patients present with predominant retrocollis, spasmodic head movements, and extracervical involvement [33,86]. Despite these clinical supportive clues, differentiation between tardive and idiopathic CD remains difficult in individual cases, and ultimately the definitive diagnosis of tardive CD will still depend on documentation of chronic neuroleptic exposure [33]. Trauma as a causative factor of torticollis has been well documented in the medical literature [2,79–81,87–90]. In most reported cases of traumatic dystonia, the neck and limbs are the most frequently affected sites, and torticollis may develop immediately or after a delay of a month to 1 year after trauma to the head, neck, or shoulder. When the onset is acute (within 4 weeks), the dystonia is characterized by markedly reduced cervical mobility, prominent shoulder elevation with trapezius hypertrophy, absence of involuntary movements, sensory tricks, or active maneuvers, and poor response to botulinum toxin injections [80]. In addition, post-traumatic torticollis usually persists during sleep and does not improve during rest or support. This is in contrast to delayed-onset CD (between 3 months and 1 year), which appears to be clinically indistinguishable from idiopathic CD [80].
Medical Management of Cervical Dystonia Despite the advances that are being made in the treatment of CD, including the use of chemical
Diagnosis and medical management of cervical dystonia
denervation as well as globus pallidum deep brain stimulation, oral pharmacologic agents still remain an alternative in the management of some patients [91–96]. However, most of these agents have not been adequately assessed in controlled clinical trials and in practice are of limited use for most patients with CD. Therefore, medical strategies of CD are usually based on anecdotal and personal experience together with empirical use over many years, rather than evidence-based scientific data [97–99]. A variety of agents have been used, including drugs affecting cholinergic, dopaminergic, serotonergic, and GABA systems. These agents may be used alone, or combined with botulinum toxin chemical denervation therapy or even with deep brain stimulation. In general, when initiating an oral medication for CD, it is important to begin with a low dose, slowly titrate to minimize side effects, and to settle on the lowest effective dose. All drugs should be given in divided doses throughout the day. It is also important to reassure patients of a possible delayed response and that they need to be patient while awaiting this. Among oral pharmacologic agents, anticholinergic drugs are considered to be the most effective for treatment of CD. Trihexyphenidyl and benztropine are the two most commonly used anticholinergic agents. The mechanism of action of triphexyphenidyl is most likely a central antimuscarinic effect. Benztropine has anticholinergic and antihistaminic effects, and also blocks presynaptic dopamine uptake. Trihexyphenidyl showed benefit during a follow-up period of 2.4 years in a double-blinded, randomized, placebo-controlled trial for the symptomatic treatment of segmental and generalized dystonia (mean dose of 30 mg/day) in young patients (mean age 18.9 years; age range 9–32 years) [100]. While useful for treatment of generalized dystonia (71% efficacy in generalized dystonia) [101], anticholinergic drugs provide symptomatic relief for only a limited number of patients with CD [102,103]. Unfortunately, there are no
110
placebo-controlled studies that have demonstrated the efficacy of anticholinergic drugs in CD. The result of one randomized, controlled trial comparing the efficacy of botulinum toxin and trihexyphenidyl in 66 CD patients favored treatment with botulinum toxin [104]. In a retrospective analysis of 71 CD patients, 39% experienced a good response to anticholinergic drugs, with women improving more than men [103]. Greater efficacy (50%) was observed in CD patients with less than 5 years duration, in contrast to 25% of CD patients with disease duration of more than 5 years [103]. Tonic CD may be more responsive to anticholinergics than spasmodic CD [105]. Anticholinergic drugs may also be effective in patients with tardive dystonia who experience predominant retrocollis. In general, the use of anticholinergic agents is limited by their side effects. Dose-limiting side effects of anticholinergic agents are peripheral and central in origin. Peripheral side effects such as dry mouth and blurred vision are common. These symptoms can be ameliorated by coadministration of a peripherally acting anticholinesterase, such as pyridostigmine, synthetic saliva, and eye drops of pilocarpine, a muscarinic agonist. In contrast, central side effects such as forgetfulness, visual hallucinations, confusion, and behavioral changes usually require reducing or discontinuing the anticholinergic drug, which lessens the usefulness of these agents [106]. Anticholinergic therapy is better tolerated if the dose is increased very slowly. Rapid dose escalation can lead to mental confusion. The improvement is sometimes delayed for several weeks. For trihexyphenidyl, treatment should be started at 1 mg/day at bedtime and increased by 2 mg/week up to the maximum tolerated dose. Other slow titration regimens can be used to avoid side effects. Although most patients require relatively high doses of anticholinergics before improvement occurs, the dose should be kept as low as possible so as to avoid unpleasant side effects. Paradoxical worsening of symptoms with anticholinergics has
1865
1866
110
Diagnosis and medical management of cervical dystonia
rarely been observed, although some patients worsen at a low dose followed by improvement at higher doses [107]. In addition to anticholinergic drugs, several GABAergic agents have been used in the treatment of CD. Among these, clonazepam is the most frequently employed agent, with approximately 21% of CD sufferers responding positively in uncontrolled studies [103]. In addition to alleviating dystonic spasms, clonazepam and diazepam are probably effective in reducing the pain associated with dystonia as well as anxiety-exacerbated dystonic postures. Other benzodiazepines may also be effective, but no study has been undertaken to compare the relative effectiveness of these agents. Another GABAergic drug, baclofen, has been shown to be effective in 11% of CD patients, less than those reported efficacy with clonazepam [103]. There are no controlled trials of baclofen for the treatment of CD. Recently, significant improvement of CD was reported in two patients following the use of intrathecal baclofen [108]. Tizanidine has not been demonstrated to be effective in the treatment of CD. Previously employed in the treatment of CD, neuroleptics have been found to have varied efficacies between 9 and 46% [109,110]. Because of long-term side effects, such as tardive dyskinesia [110] and possibly adverse metabolic effects, as well as the availability of chemical denervation therapies, their use is no longer recommended for the treatment of CD. In addition, two uncontrolled trials with clozapine (12.5–300.0 mg/day) failed to establish any improvement in CD [111,112]. Different from neuroleptics, tetrabenazine does not cause tardive syndromes, but evidence for its therapeutic effect in CD is limited [113,114]. Side effects of tetrabenazine include parkinsonism, hypotension, depression, drowsiness, and fatigue. In addition to the above medications, there are isolated reports describing the benefit of lithium (1200–1500 mg/day) for the treatment
of CD and generalized dystonia [115,116]. However, the results were not replicated in a small, controlled study [117]. Similarly, the efficacy of oral mexiletine in a small open-label study of nine CD patients was not confirmed in a subsequent controlled study [118]. Potential benefits of riluzole were reported in a small open-label study involving six CD patients who were refractory to botulinum toxin and other oral pharmacologic agents [119]. There is no evidence to suggest the efficacy of any oral agents as an adjuvant to botulinum toxin for the treatment of CD [120].
Conclusion Patients with various disorders can present with abnormal postures of the head, neck, and shoulder. Among those, CD is the most common cause of adult-onset focal dystonia. Non-dystonic causes of torticollis may resemble CD, but certain clinical features are helpful in distinguishing non-dystonic disorders from the clinical spectrum of dystonic disorders. Non-dystonic causes of torticollis are prevalent in children, but rare in adults. While most cases of CD are idiopathic, secondary causes may be identified in a small number of patients, but should be identified since the management of various secondary CD depends on the underlying disorder. In some instances, the etiology of CD may be psychogenic. It is therefore important to understand the diversity of clinical features associated with CD as well as to accurately identify possible secondary or non-dystonic causes if atypical features are present. Although botulinum toxin is currently considered the treatment of choice in CD, oral pharmacologic agents, which may be prescribed alone or as an adjunct to chemical denervation therapy or even with surgery, still have a significant role to play in some CD patients.
Diagnosis and medical management of cervical dystonia
Acknowledgments Roongroj Bhidayasiri is supported by the Rajchadapiseksompoj faculty grant of Chulalongkorn University and Parkinson disease center development grant of the Thai Red Cross Society.
References 1. Lowenstein DH, Aminoff MJ. The clinical course of spasmodic torticollis. Neurology 1988;38:530-532. 2. Truong DD, Dubinsky R, Hermanowicz N, Olson WL, Silverman B, Koller WC. Posttraumatic torticollis. Arch Neurol. 1991;48:221-223. 3. Bhidayasiri R, Tarsy D. Clinical aspects of spasmodic torticollis and their evolution. In: Bouvier G, de Soultrait F, Molina-Negro P, editors. Spasmodic torticollis. Paris: Expression Sante´; 2006. p. 65-80. 4. Kiwak KJ. Establishing an etiology for torticollis. Postgrad Med. 1984;75(7):126-34. 5. Suchowersky O, Calne DB. Non-dystonic causes of torticollis. Adv Neurol. 1988;50:501-8. 6. Fahn S, Bressman SB, Marsden CD. Classification of dystonia. Adv Neurol. 1998;78:1-10. 7. Jankovic J, Leder S, Warner D, Schwartz K. Cervical dystonia: clinical findings and associated movement disorders. Neurology 1991;41:1088-91. 8. Rodnitzky RL. Drug-induced movement disorders in children. Semin Pediatr Neurol. 2003;10(1):80-7. 9. Tonomura Y, Kataoka H, Sugie K, Hirabayashi H, Nakase H, Ueno S. Atlantoaxialrotatory subluxation associated with cervical dystonia. Spine 2007;32(19):E561-4. 10. McKnight P, Friedman J. Torticollis due to cervical epidural abscess and osteomyelitis. Neurology 1992; 42(3 Pt 1):696-7. 11. Varrato J, Galetta S. Fourth nerve palsy unmasked by botulinum toxin therapy for cervical torticollis. Neurology 2000;55(6):896. 12. Gupta AK, Roy DR, Conlan ES, Crawford AH. Torticollis secondary to posterior fossa tumors. J Pediatr Orthop. 1996;16(4):505-7. 13. Kumandas S, Per H, Gumus H, et al. Torticollis secondary to posterior fossa and cervical spinal cord tumors: report of five cases and literature review. Neurosurg Rev. 2006;29(4):333-8; discussion 338. 14. Maheshwaran S, Sgouros S, Jeyapalan K, Chapman S, Chandy J, Flint G. Imaging of childhood torticollis due to atlanto-axial rotatory fixation. Childs Nerv Syst. 1995; 11(12):667-71. 15. Ballock RT, Song KM. The prevalence of nonmuscular causes of torticollis in children. J Pediatr Orthop. 1996; 16(4):500-4.
110
16. Fernandez-Alvarez E. Transient movement disorders in children. J Neurol. 1998;245(1):1-5. 17. Cohen HA, Nussinovitch M, Ashkenasi A, Straussberg R, Kauschanksy A, Frydman M. Benign paroxysmal torticollis in infancy. Pediatr Neurol. 1993;9(6):488-90. 18. Khan AO. Paradoxical head tilt during fixation with the affected eye in unilateral congenital fourth nerve palsy. J AAPOS. 2005;9(2):200-1. 19. Williams CR, O’Flynn E, Clarke NM, Morris RJ. Torticollis secondary to ocular pathology. J Bone Joint Surg Br. 1996;78(4):620-4. 20. Chan J, Brin MF, Fahn S. Idiopathic cervical dystonia: clinical characteristics. Mov Disord. 1991;6:119-26. 21. Cano SJ, Warner TT, Linacre JM, et al. Capturing the true burden of dystonia on patients: The cervical dystonia impact profile (CDIP-58). Neurology 2004;63: 1629-1633. 22. Kutvonen O, Dastidar P, Nurmikko T. Pain in spasmodic torticollis. Pain 1997;69:279-86. 23. Lobbezoo F, Tanguay R, Thon MT, Lavigne GJ. Pain perception in idiopathic cervical dystonia (spasmodic torticollis). Pain 1996;67(2–3):483-91. 24. Tarsy D, First ER. Painful cervical dystonia: clinical features and response to treatment with botulinum toxin. Mov Disord. 1999;14(6):1043-5. 25. Jahanshahi M, Marsden CD. A longitudinal follow up study of depression, disability, and body concept in torticollis. Behav Neurol. 1990;3:233-46. 26. Konrad C, Vollmer-Haase J, Anneken K, Knecht S. Orthopedic and neurological complications of cervical dystonia-review of the literature. Acta Neurol Scand. 2004;109:369-73. 27. Ondo WG, Gollomp S, Galvez-Jimenez N. A pilot study of botulinum toxin A for headache in cervical dystonia. Headache 2005;45(8):1073-7. 28. Galvez-Jimenez N, Lampuri C, Patino-Picirrillo R, Hargreave MJ, Hanson MR. Dystonia and headaches: clinical features and response to botulinum toxin therapy. Adv Neurol. 2004;94:321-8. 29. Barbanti P, Fabbrini G, Pauletti C, Defazio G, Cruccu G, Berardelli A. Headache in cranial and cervical dystonia. Neurology 2005;64:1308-9. 30. Society HCCotIH. The international classification of headache disorders. Cephalalgia 2004;24:1-160. 31. Deuschl G, Heinen F, Kleedorfer B, et al. Clinical and polymyographic investigation of spasmodic torticollis. J Neurol. 1992;239:9-15. 32. Rondot P, Marchand MP, Dellatolas G. Spasmodic torticollis-review of 220 patients. Can J Neurol Sci. 1991;18:143-51. 33. Molho ES, Feustel PJ, Factor SA. Clinical comparison of tardive and idiopathic cervical dystonia. Mov Disord. 1998;13:486-9. 34. Prakash KM, Lang AE. Reversible dopamine agonist induced anterocollis in a multiple system atrophy patient. Mov Disord. 2007;22(15):2292-3.
1867
1868
110
Diagnosis and medical management of cervical dystonia
35. Van Gerpen JA, Matsumoto JY, Ahlskog JE, Maraganore DM, McManis PG. Utility of an EMG mapping study in treating cervical dystonia. Muscle Nerve. 2000;23(11): 1752-6. 36. Dressler D. Electromyographic evaluation of cervical dystonia for planning of botulinum toxin therapy. Eur J Neurol. 2000;7:713-8. 37. Tijssen MA, Munchau A, Marsden JF, Lees A, Bhatia KP, Brown P. Descending control of muscles in patients with cervical dystonia. Mov Disord. 2002;17(3):493-500. 38. Munchau A, Filipovic SR, Oester-Barkey A, Quinn NP, Rothwell JC, Bhatia KP. Spontaneously changing muscular activation pattern in patients with cervical dystonia. Mov Disord. 2001;16(6):1091-7. 39. Filipovic SR, Jahanshahi M, Viswanathan R, Heywood P, Rogers D, Bhatia KP. Clinical features of the geste antagoniste in cervical dystonia. Adv Neurol. 2004;94:191-201. 40. Muller J, Wissel J, Masuhr F, Ebersbach G, Wenning GK, Poewe WH. Clinical characteristics of the geste antagoniste in cervical dystonia. J Neurol. 2001;248:478-482. 41. Greene PE, Bressman S. Exteroceptive and interoceptive stimuli in dystonia. Mov Disord. 1998;13(3):549-51. 42. Ochudlo S, Drzyzga K, Drzyzga LR, Opala G. Various patterns of gestes antagonistes in cervical dystonia. Parkinsonism Relat Disord. 2007;13(7):417-20. 43. Albanese A. The clinical expression of primary dystonia. J Neurol. 2003;250(10):1145-51. 44. Schramm A, Reiners K, Naumann M. Complex mechanisms of sensory tricks in cervical dystonia. Mov Disord. 2004;19:452-8. 45. Wissel J, Muller J, Ebersbach G, Poewe W. Trick maneuvers in cervical dystonia: Investigation of movementand touch-related changes in polymyographic activity. Mov Disord. 1999;14:994-9. 46. Buchman AS, Comella CL, Leurgans S, Stebbins GT, Goetz CG. The effect of changes in head posture on the patterns of muscle activity in cervical dystonia (CD). Mov Disord. 1998;13(3):490-6. 47. Bhidayasiri R, Bronstein JM. Improvement of cervical dystonia: possible role of transcranial magnetic stimulation simulating sensory tricks effect. Med Hypotheses 2005;64(5):941-5. 48. Tang JK, Mahant N, Cunic D, et al. Changes in cortical and pallidal oscillatory activity during the execution of a sensory trick in patients with cervical dystonia. Exp Neurol. 2007;204(2):845-8. 49. Siggelkow S, Kossev A, Moll C, Dauper J, Dengler R, Rollnik JD. Impaired sensorimotor integration in cervical dystonia. J Clin Neurophysiol. 2002;19:232-9. 50. Naumann M, Magyar-Lehmann S, Reiners K, Erbguth F, Leenders KL. Sensory tricks in cervical dystonia: perceptual dysbalance of parietal cortex modulates frontal motor programming. Ann Neurol. 2000;47(3):322-8.
51. Jahanshahi M, Marion MH, Marsden CD. Natural history of adult-onset idiopathic torticollis. Arch Neurol. 1990;47:548-52. 52. Dauer WT, Burke RE, Greene P, Fahn S. Current concepts on the clinical features, aetiology and management of idiopathic cervical dystonia. Brain 1998;121:547-60. 53. Deuschl G, Bain P, Brin M. Consensus statement of the movement disorder society on tremor. Ad Hoc Scientific Committee. Mov Disord. 1998;13 Suppl 3:2-23. 54. Deuschl G. Dystonic tremor. Rev Neurol (Paris) 2003; 159(10 Pt 1):900-5. 55. Munchau A, Schrag A, Chuang C, et al. Arm tremor in cervical dystonia differs from essential tremor and can be classified by onset age and spread of symptoms. Brain 2001;124:1765-76. 56. Deuschl G, Heinen F, Guschlbauer B, Schneider S, Glocker FX, Lucking CH. Hand tremor in patients with spasmodic torticollis. Mov Disord. 1997;12(4):547-52. 57. Lou JS, Jankovic J. Essential tremor: clinical correlates in 350 patients. Neurology 1991;41(2 (Pt 1)):234-8. 58. Dubinsky RM, Gray CS, Koller WC. Essential tremor and dystonia. Neurology 1993;43(11):2382-4. 59. Pal PK, Samii A, Schulzer M, Mak E, Tsui JK. Head tremor in cervical dystonia. Can J Neurol Sci. 2000; 27(2):137-42. 60. Riski JE, Horner J, Nashold BS. Swallowing function in patients with spasmodic torticollis. Neurology 1990; 40:1443-45. 61. Ertekin C, Aydogdu I, Secil Y, Kiylioglu N, Tarlaci S, Ozdemirkiran T. Oropharyngeal swallowing in craniocervical dystonia. J Neurol Neurosurg Psychiatry 2002; 73:406-11. 62. Munchau A, Good CD, McGowan S, Quinn NP, Palmer JD, Bhatia KP. Prospective study of swallowing function in patients with cervical dystonia undergoing selective peripheral denervation. J Neurol Neurosurg Psychiatry 2001;71(1):67-72. 63. Ranoux D, Gury C, Fondarai J, Mas JL, Zuber M. Respective potencies of Botox and Dysport: a double blind, randomised, crossover study in cervical dystonia. J Neurol Neurosurg Psychiatry 2002;72(4):459-62. 64. Chawda SJ, Munchau A, Johnson D, et al. Pattern of premature degenerative changes of the cervical spine in patients with spasmodic torticollis and the impact on the outcome of selective peripheral denervation. J Neurol Neurosurg Psychiatry 2000;68(4):465-71. 65. Defazio G, Abbruzzese G, Girlanda P, et al. Primary cervical dystonia and scoliosis. A multicenter case-control study. Neurology 2003;60:1012-5. 66. Jahanshahi M, Marsden CD. Depression in torticollis: a controlled study. Psych Med. 1988;18:925-33. 67. Gundel H, Wolf A, Xidara V, Busch R, CeballosBaumann AO. Social phobia in spasmodic torticollis. J Neurol Neurosurg Psychiatry 2001;71(4):499-504.
Diagnosis and medical management of cervical dystonia
68. Papathanasiou I, MacDonald L, Whurr R, Jahanshahi M. Perceived stigma in spasmodic torticollis. Mov Disord. 2001;16:280-5. 69. O’Riordan S, Raymond D, Lynch T, et al. Age at onset as a factor in determining the phenotype of primary torsion dystonia. Neurology 2004;63(8):1423-6. 70. Greene P, Kang UJ, Fahn S. Spread of symptoms in idiopathic torsion dystonia. Mov Disord. 1995;10 (2):143-52. 71. Van Zandijcke M. Cervical dystonia (spasmodic torticollis). Some aspects of the natural history. Acta Neurol Belg. 1995;95(4):210-5. 72. Friedman A, Fahn S. Spontaneous remission in spasmodic torticollis. Neurology 1986;36:398-400. 73. Duane DD. Spasmodic torticollis. Adv Neurol. 1988;49: 135-50. 74. Giladi N, Meer J, Kidan H, Honigman S. Longterm remission of idiopathic cervical dystonia after treatment with botulinum toxin. Eur Neurol. 2000;44(3): 144-6. 75. Calne DB, Lang AE. Secondary dystonia. Adv Neurol. 1988;50:9-33. 76. LeDoux MS, Brady KA. Secondary cervical dystonia associated with structural lesions of the central nervous system. Mov Disord. 2003;18:60-9. 77. Hartmann A, Pogarell O, Oertel WH. Secondary dystonias. J Neurol. 1998;245(8):511-8. 78. Ruegg SJ, Buhlmann M, Renaud S, Steck AJ, Kappos L, Fuhr P. Cervical dystonia as first manifestation of multiple sclerosis. J Neurol. 2004;251(11):1408-10. 79. Sa DS, Mailis-Gagnon A, Nicholson K, Lang AE. Posttraumatic painful torticollis. Mov Disord. 2003;18: 1482-91. 80. Tarsy D. Comparison of acute- and delayed-onset posttraumatic cervical dystonia. Mov Disord. 1998;13:481-5. 81. O’Riordan S, Hutchinson M. Cervical dystonia following peripheral trauma: a case-controlled study. J Neurol. 2004;251:150-5. 82. Schrag A, Trimble M, Quinn N, Bhatia K. The syndrome of fixed dystonia: an evaluation of 103 patients. Brain 2004;127(Pt 10):2360-72. 83. Lang AE. Psychogenic dystonia: a review of 18 cases. Can J Neurol Sci. 1995;22(2):136-43. 84. Fahn S, Williams DT. Psychogenic dystonia. Adv Neurol. 1988;50:431-55. 85. Shill H, Gerber P. Evaluation of clinical diagnostic criteria for psychogenic movement disorders. Mov Disord. 2006;21(8):1163-8. 86. Krack P, Schneider S, Deuschl G. Geste device in tardive dystonia with retrocollis and opisthotonic posturing. Mov Disord. 1998;13(1):155-7. 87. Goldman S, Ahlskog JE. Posttraumatic cervical dystonia. Mayo Clin Proc. 1993;68:443-8. 88. Frei KP, Pathak M, Jenkins S, Truong DD. Natural history of posttraumatic cervical dystonia. Mov Disord. 2004;19:1492-98.
110
89. Krauss JK, Mohadjer M, Braus DF, Wakhloo AK, Nobbe F, Mundinger F. Dystonia following head trauma: a report of nine patients and review of the literature. Mov Disord. 1992;7(3):263-72. 90. Dewey RB, Jr., Maraganore DM, Matsumoto JY. Posttraumatic cervical dystonia manifesting as isolated spasm of the middle scalene muscle. Mayo Clin Proc. 1994;69(2):187-8. 91. Krauss JK, Loher TJ, Pohle T, et al. Pallidal deep brain stimulation in patients with cervical dystonia and severe cervical dyskinesias with cervical myelopathy. J Neurol Neurosurg Psychiatry 2002;72(2):249-56. 92. Bereznai B, Steude U, Seelos K, Botzel K. Chronic highfrequency globus pallidus internus stimulation in different types of dystonia: a clinical, video, and MRI report of six patients presenting with segmental, cervical, and generalized dystonia. Mov Disord. 2002;17(1):138-44. 93. Jankovic J, Schwartz K. Botulinum toxin injections for cervical dystonia. Neurology 1990;40(2):277-80. 94. Jankovic J, Orman J. Botulinum A toxin for cranialcervical dystonia: a double-blind, placebo-controlled study. Neurology 1987;37(4):616-23. 95. Albanese A, Barnes MP, Bhatia KP, et al. A systematic review on the diagnosis and treatment of primary (idiopathic) dystonia and dystonia plus syndromes: report of an EFNS/MDS-ES Task Force. Eur J Neurol. 2006;13(5): 433-44. 96. Bhidayasiri R, Tarsy D. Treatment of dystonia. Expert Rev Neurother. 2006;6(6):863-86. 97. Goldman JG, Comella CL. Treatment of dystonia. Clin Neuropharmacol. 2003;26(2):102-8. 98. Jankovic J. Dystonia: medical therapy and botulinum toxin. Adv Neurol. 2004;94:275-86. 99. Bhidayasiri R, Tarsy D. Medical therapy for dystonia. In: Stacy M, editors. Handbook of dystonia. New York: Informa Healthcare; 2007. p. 301-16. 100. Burke RE, Fahn S. Double-blind evaluation of trihexyphenidyl in dystonia. Adv Neurol. 1983;37:189-92. 101. Burke RE, Fahn S, Marsden CD. Torsion dystonia: a double-blind, prospective trial of high-dosage trihexyphenidyl. Neurology 1986;36(2):160-4. 102. Lang AE, Sheehy MP, Marsden CD. Anticholinergics in adult-onset focal dystonia. Can J Neurol Sci. 1982; 9:313-9. 103. Greene P, Shale H, Fahn S. Analysis of open-label trials in torsion dystonia using high dosages of anticholinergics and other drugs. Mov Disord. 1988;3(1):46-60. 104. Brans JW, Lindeboom R, Snoek JW, et al. Botulinum toxin versus trihexyphenidyl in cervical dystonia: a prospective, randomized, double-blind controlled trial. Neurology 1996;46:1066-72. 105. Jabbari B, Scherokman B, Gunderson CH, Rosenberg ML, Miller J. Treatment of movement disorders with trihexyphenidyl. Mov Disord. 1989;4(3):202-12. 106. Taylor AE, Lang AE, Saint-Cyr JA, Riley DE, Ranawaya R. Cognitive processes in idiopathic dystonia treated
1869
1870
110 107. 108.
109.
110. 111.
112.
Diagnosis and medical management of cervical dystonia
with high-dose anticholinergic therapy: implications for treatment strategies. Clin Neuropharmacol. 1991; 14(1):62-77. Fahn S. High dosage anticholinergic therapy in dystonia. Neurology 1983;33(10):1255-61. Dykstra DD, Mendez A, Chappuis D, Baxter T, DesLauriers L, Stuckey M. Treatment of cervical dystonia and focal hand dystonia by high cervical continuously infused intrathecal baclofen: a report of 2 cases. Arch Phys Med Rehabil. 2005;86(4):830-3. Zuddas A, Cianchetti C. Efficacy of risperidone in idiopathic segmental dystonia. Lancet 1996;347(8994): 127-8. Lang AE. Dopamine agonists and antagonists in the treatment ofidiopathicdystonia.Adv Neurol. 1988;50:561-70. Burbaud P, Guehl D, Lagueny A, et al. A pilot trial of clozapine in the treatment of cervical dystonia. J Neurol. 1998;245:329-31. Thiel A, Dressler D, Kistel C, Ruther E. Clozapinetreatment of spasmodic torticollis. Neurology 1994;44:957-8.
113. Jankovic J, Orman J. Tetrabenazine therapy of dystonia, chorea, tics, and other dyskinesias. Neurology 1988; 38(3):391-4. 114. Paleacu D, Giladi N, Moore O, Stern A, Honigman S, Badarny S. Tetrabenazine treatment in movement disorders. Clin Neuropharmacol. 2004;27(5):230-3. 115. Couper-Smartt J. Lithium in spasmodic torticollis. Lancet 1973;2(7831):741-2. 116. Marti-Masso JF, Obeso JA, Carrera N, Astudillo W, Martinez Lage JM. Lithium therapy in torsion dystonia. Ann Neurol. 1982;11(1):106-7. 117. Koller WC, Biary N. Lithium ineffective in dystonia. Ann Neurol. 1983;13(5):579-80. 118. Ohara S, Hayashi R. Mexiletine in the treatment of spasmodic torticollis. Mov Disord. 1998;13:934-40. 119. Muller J, Wenning GK, Wissel J, et al. Riluzole therapy in cervical dystonia. Mov Disord. 2002;17(1): 198-200. 120. Jankovic J. Treatment of dystonia. Lancet Neurol. 2006;5(10):864-72.
106 Diagnosis and Medical Management of Dystonia S. Fahn
Introduction Dystonia, as the name is used today, goes back to the coinage of the term by Oppenheim in 1911 [1] in his seminal paper describing four patients, with what today we would classify as primary generalized dystonia. Oppenheim was uncertain how to classify or call this disorder, and offered two different names; the one that stuck was dystonia musculorum deformans. He coined ‘‘dystonia’’ to indicate that there could be hypotonia on one occasion and tonic muscle spasms on another. Actually, the disorder was described in the literature prior to Oppenheim’s paper, but under a variety of names (for historical details, see [2]). In the subsequent decades after Oppenheim, the clinical features of dystonia were recognized in Wilson’s disease, cerebral palsy and on recovering from encephalitis, so that dystonia as a primary disorder was more or less lost. It was not until 1944, that dystonia was resurrected as a specific entity as well as a feature of other neurologic disorders [3–5]. The term dystonia had been originally applied to those with generalized involvement, largely children, and it took Marsden [6] to associate a variety of adult-onset focal movement disorders (such as blepharospasm, oromandibular dystonia, torticollis, writer’s cramp), often thought to be psychogenic, to be included under the rubric dystonia. Prior to that recognition, the term formes fruste of generalized dystonia had been applied to some of these disorders, such as torticollis. Today, it is recognized that the vast majority of cases of primary dystonia are the #
Springer-Verlag Berlin/Heidelberg 2009
adult-onset focal dystonias, and only about one-sixth are generalized. Over the years, dystonia had been defined differently by several clinicians. A committee to come up with a unified definition was organized by the Dystonia Medical Research Foundation and offered the following: Dystonia is a syndrome of sustained muscle contractions, frequently causing twisting and repetitive movements, or abnormal postures [2,7].
Clinical Features In contrast to most other types of abnormal movements, agonist and antagonist muscles contract simultaneously (co-contraction) to produce the sustained quality of dystonic movements. Moving a body part in one direction, though, requires a disproportional effort between these opposing muscles. It is common for patients to subconsciously try to straighten out the twisting movements. This produces jerky movements as the dystonia is pulling the affected body part in one direction while the patient attempts to oppose it. It is not uncommon for physicians to mistake such jerky movements as chorea or tremor. By asking the patient not to fight the pulling action of the dystonia, ‘‘just let the muscles go where they want to go,’’ the true sustained twisting properties of the dystonia becomes apparent. Another distinctive feature is that the jerky dystonic movements are patterned, i.e., these contractions continual involve the same muscle
1768
106
Diagnosis and medical management of dystonia
groups repetitively and do not flow to other body parts in a flowing manner, which would be a feature of chorea. The patterned dystonic movements can be rapid, which can also mistake them for choreic movements. The abnormal twisting movements and postures of dystonia are often exacerbated by active voluntary movement of the affected body part and also by normal body parts (so-called ‘‘overflow.’’) The speed of the movement varies widely from slow (athetotic dystonia) to shock-like (myoclonic dystonia). Pain is uncommon in dystonia except when neck muscles are involved (cervical dystonia, torticollis). Rhythmical movements (dystonic tremor) can be present in a dystonic arm or neck. One feature that may help differentiate dystonic tremor from other types of tremor is that there is usually a null point in positioning the affected dystonic body part in which the tremor disappears. When dystonia first appears, the movements typically occur when the affected body part is carrying out a voluntary action (action dystonia) and are not present when that body part is at rest. With progression, dystonic movements can appear at the involved site when other parts of the body are voluntarily moving (i.e., overflow). With further progression dystonic movements become present when the body is ‘‘at rest.’’ Even at this stage, dystonic movements are usually made more severe with voluntary activity, even with the act of speaking. Whereas primary dystonia often begins as action dystonia and may persist as the kinetic (clonic) form, symptomatic dystonia often begins as fixed postures (tonic form). An uncommon form is when the dystonia is present at rest and disappears with action, so-called paradoxical dystonia. Blepharospasm is an exception, because it is just as common to have this paradoxical nature (worse at rest, disappearing with action, such as the act of talking or walking), as being present only with action. Another characteristic and almost unique feature of dystonic movements is that they can
often be diminished by tactile or proprioceptive ‘‘sensory tricks’’ (geste antagoniste). Thus, touching the involved body part or an adjacent body part can often reduce the muscle contractions. For example, patients with torticollis will often place a hand on the chin or side of the face to reduce nuchal contractions, and oromandibularlingual dystonia is often helped by touching the lips or placing an object in the mouth.
Classification Dystonia can be classified in three ways: age at onset, affected body distribution, and etiology (> Table 106-1). Classification by age at onset is useful because this is the most important single factor related to prognosis of primary dystonia, with earlier onset usually progressing to involve several body parts, and late (adult) onset remaining more static and focal. If onset is in the legs, this also indicates a more progressive course leading to generalized dystonia. When a single body part is affected, the condition is referred to as a focal dystonia. Common forms of focal dystonia are spasmodic blepharospasm (upper facial dystonia), oromandibular . Table 106-1 Classification of dystonia By age at onset Early-onset: 26 years Late-onset: >26 years By distribution Focal Segmental Multifocal Generalized Hemidystonia By etiology Primary (also known as idiopathic) dystonia Dystonia-plus Secondary dystonia Heredodegenerative dystonia (usually presents as dystonia-plus) A feature of another neurologic disease (e.g., dystonic tics, paroxysmal dyskinesias, PD, PSP)
Diagnosis and medical management of dystonia
dystonia, torticollis (cervical dystonia), and writer’s cramp (hand and arm dystonia). Involvement of two or more contiguous regions of the body is referred to as segmental dystonia. Generalized dystonia indicates involvement of one or both legs, the trunk, and some other part of the body. Multifocal dystonia involves two or more regions, not conforming to segmental or generalized dystonia. Hemidystonia refers to involvement of the arm and leg on the same side, and usually indicates that the dystonia is secondary to a focal brain injury. The etiologic classification divides the causes of dystonia into four major categories: primary (or idiopathic), secondary (or symptomatic) (environmental causes), dystonia-plus syndromes, and heredodegenerative diseases in which dystonia is a prominent feature. Primary dystonia is characterized as a pure dystonia (with the exception that tremor can be present) and excludes a symptomatic cause; primary dystonia can be familial or sporadic. Dystonia-plus syndromes are non-degenerative (i.e., biochemical) diseases that consist of dystonia and another feature, such as parkinsonism or myoclonus. Dopa-responsive dystonia (incorporates features of parkinsonism) and myoclonus-dystonia (myoclonus can even predominate) are the most common dystoniaplus syndromes. Secondary dystonias are those due to an environmental insult, and heredodegenerative dystonias are due to neurodegenerative diseases that are usually inherited. Other neurologic diseases in which dystonia is often present are dystonic tics, paroxysmal dyskinesias, Parkinson disease, and progressive supranuclear palsy. > Table 106-2 is a listing of the etiologic classification of dystonia.
Adult-onset Focal Dystonias The focal dystonias are the most common forms of dystonias, and these are usually the end result when dystonia first appears in adults, although
106
some may spread to adjacent body parts (segmental dystonia). Focal dystonia is also the first appearance of childhood-onset dystonia, but it soon spreads to become segmental and often generalized. The clinical feature of focal dystonia depends on which body part is affected. The most common focal dystonia involves the neck musculature, known as spasmodic torticollis or cervical dystonia. The head can turn (rotational torticollis), tilt towards an ear, or shift to one side, or bend forward (antecollis) or backwards (retrocollis). Any combination of head positions can be found in individual patients. About 10% have a remission within a year, but relapses usually occur, even many years later. The average age at onset is between 20 and 50 years. The muscles involved are innervated by Cranial Nerve (CN) XI and the upper cervical nerve roots. Some cervical dystonias are manifested as a static pulling of the head into one direction (tonic), but most have a jerky, irregular rhythmic feature (clonic). Some patients try to fight the pulling of the neck by contracting the antagonist muscles, and the physician can be misled by seeing or feeling these contracted muscles, thinking that these are the dystonic muscles, whereas, in fact, they would be the compensatory muscles contracting. To distinguish between involuntary and compensatory/voluntary contractions, the patient should be told to let the movements occur without trying to overcome them. Having them close their eyes and ‘‘let the head go where it wants to go’’ is the technique. Then the true direction of which muscles are involved by the dystonia is revealed. This is especially important when deciding which muscles to inject with botulinum toxin. Common sensory tricks to reduce dystonia are touching the face or the back of the head. When the patient sits with the back of the head touching a wall or the head rest in a car can offer relief. Mechanical devices to place constant cutaneous pressure on the occiput can sometimes be used as a treatment.
1769
1770
106
Diagnosis and medical management of dystonia
. Table 106-2 Principal dystonic disorders by etiology Primary (also known as idiopathic) dystonia Oppenheim dystonia (also called DYT1) Autosomal dominant inheritance of mutated TOR1A gene Early onset (
Diagnosis and medical management of dystonia
106
. Table 106-2 (Continued) Autosomal recessive Wilson disease Gene: Cu-ATPase located at 13q14.3 Niemann-Pick type C (dystonic lipidosis) (sea-blue histiocytosis) defect in cholesterol esterification; gene mapped to chromosome 18 Juvenile neuronal ceroid-lipofuscinosis (Batten disease) GM1 gangliosidosis GM2 gangliosidosis Metachromatic leukodystrophy Lesch-Nyhan syndrome Homocystinuria Glutaric acidemia Neurodegeneration with brain iron accumulation Type 1 PKAN Neuroacanthocytosis Probable autosomal recessive Familial basal ganglia calcifications Progressive pallidal degeneration Mitochondrial Leigh disease Leber disease Dystonia present in other neurologic syndromes Parkinsonism – Parkinson disease, progressive supranuclear palsy, multiple system atrophy, cortical-basal ganglionic degeneration Tourette syndrome (dystonic tics) Paroxysmal dyskinesias Source: Adapted from [8].
Bleparospasm usually occurs in older individuals, women greater than men. It typically begins as excessive blinking, but sometimes just as a lowering of the upper eyelids, commonly mistaking this as ptosis. Many patients complain of eye irritation or dryness, and dry eyes from Sjo¨gren disease need to be ruled out. The early blinking phase often leads to longer stretches of eyelid closing, even to very long durations. Usually there is a combination of eyelid closing and blinking. The most severe form is forceful closing of the eyelids. The muscles involved are the orbicularis oculi innervated by CN VII, and the movements are symmetrical in the two eyes, quite distinct from hemifacial spasm, which is unilateral. One interesting difference is that usually in blepharospasm the eyebrows may elevate due to simultaneous contractions of the frontalis muscles, while in hemifacial spasm the eyebrows come down. Dystonia can spread from the upper face (causing
blepharospasm) to the lower face with movements around the mouth. Common sensory tricks to reduce blepharospasm are touching the corner of the eye, coughing and talking. Bright light notoriously aggravates blepharospasm, and patients have difficulty being in bright light, especially sunlight, and they often wear sunglasses most of the time, even indoors. Driving at night is very difficult because of oncoming headlights. The brightness of a movie screen in a darkened cinema also aggravates blepharospasm. Oromandibular dystonia (OMD) (jaw muscles innervated by CN V) is often associated with lingual dystonia (tongue muscles by CN XII). OMD is less common than blepharospasm. Jawopening dystonia is where the jaw is pulled down by the pterygoids. In jaw-closing dystonia the masseters and temporalis muscles are the prime movers. The jaw can also be moved laterally by the pterygoids. It is important to distinguish the
1771
1772
106
Diagnosis and medical management of dystonia
latter from facial muscle pulling the mouth to one side, which is often a psychogenic movement disorder. OMD can markedly affect chewing and swallowing. Some OMDs appear only with action and are not present at rest. Such actions can involve talking or chewing. With ‘‘at rest’’ OMD, often a patient attempts to overcome jaw-opening and jaw-closing dystonias by purposefully moving the jaw in the opposite direction. This maneuver has often led to an error in misdiagnosis of tardive dyskinesia because the movements superficially appear rhythmic. To distinguish between OMD and tardive dyskinesia, the patient should be told to let the movements occur without trying to overcome them. In this manner, the true direction of where the OMD wants to take the jaw is revealed, and the rhythmic movements stop. Sensory tricks that have been useful in OMD are the placing of an object in the mouth or biting down on an object, such as a tongue blade or pencil. Dental implants have sometimes helped by the physical application of a continual sensory trick. Embouchure dystonia is an action dystonia involving the muscles around the mouth (embouchure) that may develop in professional musicians who play woodwinds and horn instruments. More commonly are musician’s cramps involving the fingers in instrumentalists such as pianists, guitarists, violinists and other string instruments. These are forms of occupational cramps, the most common being writer’s cramp. All are task-specific dystonias. In writer’s cramp, the action of writing brings out the dystonic tightening of the finger, hand, forearm and arm muscles, such as the triceps. If the dystonia progresses, other actions of the arm, like fingerto-nose maneuver, buttoning, sewing, bring out the dystonia. Ultimately, dystonia at rest can develop, but usually does not. About 15% of patients with writer’s cramp have spread to the other arm. Otherwise, the patient can learn to write with the uninvolved arm. Sensory tricks that have been useful are the placing of the pen/ pencil in between other fingers, using large
writing implements, and placing the non-writing hand on top of the writing hand. Dystonia of the vocal cords comes in two varieties: adduction or abduction of the cords with speaking. The former is much more common and produces a tight, constricted, strangulated type of voice with frequent pauses breaking up the voice, and it takes longer to complete what the patient is trying to say. The latter produces a whispering voice. With dystonic adductor dysphonia, the patient is still able to whisper normally, and may present this way to the physician who needs to be aware that this is a compensating mechanism and not primary abductor dysphonia. A major differential diagnosis is vocal tremor, seen fairly commonly in patients with essential tremor; in this condition, the vocal chords are tremulous, like the hands. Focal truncal dystonia can present in adults, both as primary dystonia and as tardive dystonia. The dystonia is usually absent when the patient is lying or sitting, and appears on standing and walking. This is an uncommon form of adultonset focal dystonia. Even less uncommon is focal dystonia of a leg in an adult; but it does happen as a primary dystonia.
Genetic Forms of Dystonia The etiologic classification is arguably the most important type of classification scheme, for knowing the etiology (and pathogenesis) may lead to more specific treatments. A much better understanding of the etiology of dystonia has evolved in the past decade by advances in discovering gene defects that cause dystonia – primary, dystoniaplus and heredodegenerative forms. Many have been assigned the DYT label (> Table 106-3). This DYT label has been applied in chronologic order of their discoveries. Of the 15 with specific gene identifications or chromosomal mapping, most are primary dystonias (DYT1, DYT6, DYT7, DYT13) or Dystonia-plus syndromes
Diagnosis and medical management of dystonia
106
. Table 106-3 Monogenic forms of dystonia Designation
Dystonia type
Inheritance
Gene name and locus
Gene product
DYT1
Early onset Limb onset Progresses Oppenheim dystonia Early onset Generalized Filipino males
Autosomal dominant
TOR1A, 9q34
TorsinA
Autosomal recessive
Unknown
Unknown
X-linked recessive
TAF1, Xq13.1
Multiple transcript system
Unknown
Unknown
GCH1, 14q22.1
GTP cyclo-hydrolase 1
Autosomal recessive Autosomal dominant
11p11.5 8p21-q22
Tyrosine hydroxylase Unknown
Autosomal dominant Autosomal dominant
18p MR-1, 2q34
Unknown Myofibrillogenesis regulator 1
Autosomal dominant
1p21
Unknown
Autosomal dominant
16p11.2, -q12.1 SGCE, 7q21 ATP1A3, 19q12-q13.2 1p36
Unknown
18p11 PANK2, 20p12.3-p13 FTLI, 19q13.3
Unknown Pantothenate kinase deficiency (PKAN) Ferritin light polypeptide
DYT2 DYT3
DYT4 DYT5
DYT6 DYT7 DYT8
DYT9 DYT10 DYT11 DYT12 DYT13 DYT14
Dystonia evolves into parkinsonism Whispering dysphonia Autosomal dominant Dopa-responsive dystonia Segawa syndrome Autosomal dominant Childhood onset Infantile onset Adolescent-onset dystonia (Mennonite/Amish) Familial torticollis Paroxysmal nonkinesigenic dyskinesia Mount-Rebak syndrome Paroxysmal dyskinesia with spasticity Paroxysmal kinesigenic dyskinesia Myoclonus dystonia Rapid-onset dystonia – parkinsonism Cranial-cervical – brachial dystonia Dopa-responsive dystonia
DYT15 Myoclonus dystonia Neurodegeneration with brain iron accumulation Type 1 Neurodegeneration with brain iron accumulation Type 2 (Neuroferritinopathy)
Autosomal dominant Autosomal dominant Autosomal dominant Same as DYT5 (originally mistaken as a new variety) Autosomal dominant Autosomal recessive Autosomal dominant
(DYT5, DYT11, DYT12, DYT15). One is a degenerative disorder (DYT3), and three are clinically part of the paroxysmal dyskinesia disorders (DYT8, DYT9, DYT10). Two are incompletely evaluated families that were designated as DYT2 and DYT4 prematurely. DYT 14 was improperly designated, for the involved family proved to be one of DYT5. DYT1 or Oppenheim dystonia has been the one most well studied, and represents
Epsilon-sarcoglycan Na-K-ATPase Unknown
the prototypic primary dystonia. DYT5 or doparesponsive dystonia (DRD) is most easily treated. Both will be discussed below.
Oppenheim Dystonia Named after Hermann Oppenheim [1] who coined the term dystonia in 1911 to describe
1773
1774
106
Diagnosis and medical management of dystonia
Jewish children with what he called dystonia musculorum deformans, it is now commonly known as DYT1, following genetic studies. The gene (TOR1A) codes for the protein torsinA, found in neuronal endoplasmic reticulum and the nuclear envelope. The abnormal torsinA becomes more prominently located in the nuclear envelope, which is continuous with the endoplasmic reticulum [9]. The universal mutation in this disorder, common in the Ashkenazi Jewish population due to a founder effect introduced about 360 years ago [10], is a deletion of one of an adjacent pair of GAG triplets (codes for glutamate) near the functional domain. The disorder usually begins in childhood, but sometimes in adults, and it almost always starts in either an arm or leg. It tends to spread contiguously and in many (especially if it begins in a leg) becomes generalized and disabling. TorsinA is an ATPase of the heat-shock type that functions in restoring damaged proteins particularly in membranes. The abnormal protein loses ATPase activity, with a resultant decrease as a chaperone protein. The penetrance rate is only 30%. Variations in the TOR1A gene are at least one factor associated with the susceptibility in expressing the disease [11]. The TOR1A gene can be tested commercially for the mutation, which is recommended for all patients with onset of primary dystonia below the age of 26 [12]. The gross and microscopic pathologic evaluations have failed to reveal any consistent pathology, but with special staining techniques, inclusions in the nuclear envelope region have been detected [13].
Dopa-responsive Dystonia (DRD) DRD usually begins in childhood with a peculiar gait (walking on toes). Adult onset cases tend to resemble Parkinson disease. Even in childhood, one can detect parkinsonian features of bradykinesia and loss of postural reflexes, a feature that distinguishes DRD from Oppenheim dystonia. It can resemble, therefore, childhood onset
Parkinson disease with dopamine nigrostriatal neuronal degeneration. The latter condition shows a reduction of ligand binding in FDOPA or dopamine transporter PET and SPECT scanning, whereas these studies are normal in DRD, which does not show neuronal degeneration. DYT5 is due to one of the many discovered mutations in the gene for GTP cyclohydrolase 1 which is the rate-limiting step for synthesis of tetrahydrobiopterin, the cofactor for tyrosine hydroxylase and other hydroxylases, required for the synthesis of biogenic amines, such as dopamine. The dopamine deficiency in the striatum accounts for the symptoms, and the patients respond extremely well to low doses of levodopa. The patients can respond even after years of no treatment, and they do not develop the fluctuations or dyskinesias so commonly seen in patients with Parkinson disease. Interestingly, these patients also respond to low dosages of anticholinergic medications, such as trihexyphenidyl. Some patients show a diurnal pattern of symptoms, being almost normal in the morning, and markedly dystonic at the end of the day. These patients obtain benefit from sleep.
Secondary and Heredodegenerative Dystonias In evaluating patients with dystonia, a history of a neurologic insult, such as exposure to drugs, toxins, trauma, or infection, or the finding of an abnormality on neurological examination would suggest that the patient’s dystonia is secondary or heredodegenerative rather than being a primary dystonia. > Table 106-4 lists the major causes of secondary dystonia, and > Table 106-5, the heredodegenerative dystonias. The most common causes of secondary dystonias seen in a busy movement disorder center are the drug-induced dystonias, especially tardive dystonia, induced by dopamine receptor blocking agents. Tardive dystonia can affect all ages, but as in classic tardive dyskinesia, older people are most susceptible.
Diagnosis and medical management of dystonia
. Table 106-4 Major causes of secondary dystonia Tardive dystonia Birth injury Psychogenic Peripheral trauma Head injury Stroke Encephalitis Note: These etiologies are listed in order of their prevalence as encountered by the Dystonia Clinical Research Center at Columbia University Medical Center
In children and adolescents, it can manifest as generalized dystonia, but it is usually reversible after a long period of being without the offending drugs. Older people are more likely to have a focal form of tardive dystonia, usually OMD, cervical dystonia or blepharospasm. Tardive dystonia resembles primary dystonias unless there is an accompanying tardive akathisia or classic tardive dyskinesia, which allows the diagnosis to be made quite readily. When these other forms of the tardive syndromes are not present, one clinical feature of tardive dystonia that is fairly common to allow this diagnosis is the posture of hyperextension of the neck and trunk with pronated arms, extended elbows and flexed wrists. Also the diagnosis of tardive dystonia may be suspected if the patient has had a recent exposure to dopamine receptor blocking agents.
Prevalence, Neuroimaging, Physiology and Pathology Many studies on the prevalence of dystonia have been conducted, with varying results. Focal dystonias are about tenfold more prevalent than generalized dystonia. In Olmstead County, Minnesota, generalized dystonia occurred in about 1/30,000 individuals, while focal dystonias were found in about 1/3,000 [14]. Although routine MRI fails to show any abnormalities in primary dystonia, voxel based
106
. Table 106-5 Heredodegenerative diseases with prominent dystonia (typically not pure dystonia) X-linked recessive a. Lubag (X-linked dystonia-parkinsonism) (DYT3) b. Deafness-dystonia syndrome (Mohr-Tranebjaerg syndrome) Mutations in DDP1 (deafness-dystonia peptide 1) X-linked dominant Rett syndrome Autosomal dominant Juvenile Huntington disease Neuroferritinopathy Machado-Joseph disease (SCA3) Autosomal recessive Juvenile parkinsonism (parkin gene mutation, presenting with dystonia) Wilson disease Niemann-Pick type C (dystonic lipidosis) (sea-blue histiocytosis) Juvenile neuronal ceroid-lipofuscinosis (Batten disease) GM1 gangliosidosis GM2 gangliosidosis Metachromatic leukodystrophy Lesch-Nyhan syndrome Homocystinuria Glutaric acidemia Triosephosphate isomerase deficiency Methylmalonic aciduria Hartnup disease Ataxia telangiectasia Friedreich ataxia Neurodegeneration with brain iron accumulation Type 1 Neuroacanthocytosis Neuronal intranuclear hyaline inclusion disease Hereditary spastic paraplegia with dystonia Sjo¨gren-Larsson syndrome (ichthyosis, spasticity, mental retardation) Ataxia-amyotrophy-mental retardation-dystonia syndrome Probable autosomal recessive Familial basal ganglia calcifications Progressive pallidal degeneration Mitochondrial Leigh disease Leber disease Other mitochondrial encephalopathies Associated with parkinsonian syndromes Parkinson disease Progressive supranuclear palsy Multiple system atrophy Cortical-basal ganglionic degeneration Source: Adapted from [8], page 317.
1775
1776
106
Diagnosis and medical management of dystonia
morphometry has revealed an increase in the size of sensorimotor structures. Diffusion tensor magnetic resonance imaging indicates a reduced axonal integrity in the subgyral white matter of the sensorimotor cortex in those carrying the DYT1 mutation [15,16]. FDG PET techniques show lenticular-thalamic dissociation and dystonia network patterns [17,18]. These suggest that in this disorder hyperkinetic movements may arise through excessive activity of the direct putaminopallidal inhibitory pathway, resulting in inhibition of the globus pallidus interna. This, in turn, indicates reduced pallidal inhibition of the thalamus with consequent overactivity of medial and prefrontal cortical areas and underactivity of the primary motor cortex during movements. PET scans assessing D2 receptor binding showed a reduction in the striatum in primary dystonia, including nonmanifesting DYT1 gene carriers. There is reduced spinal cord and brainstem inhibition in many reflex studies (long-latency reflexes, cranial reflexes and reciprocal inhibition) in primary dystonia. Physiologic studies of the cerebral cortex show educed preparatory activity in the EEG before the onset of voluntary movements, enhanced premotor and supplementary motor cortical excitability and reduced primary motor cortex activity. These studies suggest that there is defective ‘‘surround inhibition’’ in the basal ganglia and cerebral cortex in primary dystonia [19]. Postmortem examination in the primary dystonias and dystonia-plus syndromes reveals no consistent gross or microscopic pathological changes, except one paper on Oppenheim dystonia that showed perinuclear inclusions using a special immunostaining technique [13]. On the other hand, secondary and heredodegenerative dystonias typically have pathology in the basal ganglia or its connections (thalamus and cortex). Pallidal and thalamic stereotaxic surgical targets that can ameliorate dystonia also support a basal ganglionic pathophysiologic mechanism. Pharmacologic observations (acute and tardive
dystonia; levodopa-induced dystonia; DRD; treatment of dystonia with anticholinergics) also support involvement of the basal ganglia. The pathology of lubag suggests that dystonia may result from an imbalance in the activity between the striosomal and matrix-based pathways (impaired striosomes with intact matrix) [20].
Treatment Treatment of the dystonias continues to evolve. Certain principles remain fundamental. Although there are agents that can often reduce the severity of dystonia, it is most important to identify specific disorders that are treatable, e.g., Wilson’s disease, drug-induced, and infectious, and treat them. Of course, as in all diseases, one needs to educate the patient and family, and provide genetic counseling. The primary focal dystonias of eyelids, vocal cords, jaw, neck and limbs are most easily treated with botulinum toxin (BTX) injections, and this is the preferred approach. Segmental and generalized dystonias require systemic pharmacologic therapy, but residual focal involvement can be treated with BTX. Surgical therapy is reserved for disabling dystonia resistant to medications and BTX injections. But with successful outcomes, surgery is started sooner now, before the patient develops fixed dystonic postures. Surgical therapy is covered in a separate chapter. Pharmacologic therapy, as applied in generalized dystonia, should start with a trial of levodopa to make sure DRD is not overlooked. If that fails, the drugs with the most success are high dosages of anticholinergics (e.g., trihexyphenidyl and benztropine), baclofen and benzodiazepines. High dosage anticholinergics have shown benefit in a controlled clinical trial [21]. The sooner treatment is started, the better the chance for avoiding progressive disabling dystonia [22]. Other medications with some reported benefit are carbamazepine, a dopamine depletor (tetrabenazine or reserpine), dopamine receptor
Diagnosis and medical management of dystonia
. Table 106-6 Order for pharmacologic treatment for segmental and generalized dystonias Levodopa trial to rule out DRD Anticholinergics Baclofen Benzodiazepine Carbamazepine Tetrabenazine or reserpine Dopamine receptor blocker Triple therapy: dopamine depletor, dopamine receptor blocker and anticholinergic Surgery
3.
4. 5. 6.
7. 8.
blockers, and a combination of dopamine depletor, dopamine receptor blocker and anticholinergic medication [22] (> Table 106-6). Because these drugs can produce intolerable side effects, they should be started at low dose and built up gradually until either therapeutic benefit is seen or intolerable side effects are seen. Work with one drug first to determine what it can accomplish. If there is some benefit, a second drug could be added when the final dosage of the first drug is reached. Adding multiple drugs this way may provide benefit not seen with just a single drug. Peripheral side effects of anticholinergics can usually be controlled by employing pilocarpine eye drops (for blurred vision due to dilated pupils), pyridostigmine for urinary hesitation and constipation, and artificial saliva for dry mouth. Central side effects, like impaired memory, are the major cause of being unable to reach the high therapeutic doses needed for benefit, particularly in adults, who cannot tolerate these drugs as well as children can.
References ˝ ber eine eigenartige Krampfkrankheit 1. Oppenheim H. U des kindlichen und jugendlichen Alters (Dysbasia lordotica progressiva, Dystonia musculorum deformans). Neurol Centrabl 1911;30:1090-107. 2. Fahn S, Marsden CD, Calne DB. Classification and investigation of dystonia. In: Marsden CD, Fahn S, editors.
9.
10.
11.
12.
13.
14.
15.
16.
17.
106
Movement disorders 2. London: Butterworths; 1987. p. 332-58. Herz E. Dystonia I. Historical review: analysis of dystonic symptoms and physiologic mechanisms involved. Arch Neurol Psychiatry 1944;51:305-18. Herz E. Dystonia II. Clinical classification. Arch Neurol Psychiatry 1944;51:319-55. Herz E. Dystonia. III. Pathology and conclusions. Arch Neurol Psychiatry 1944;52:20-6. Marsden CD. The problem of adult-onset idiopathic torsion dystonia and other isolated dyskinesias in adult life (including blepharospasm, oromandibular dystonia, dystonic writer’s cramp, and torticollis, or axial dystonia). Adv Neurol 1976;14:259-76. Fahn S. Concept and classification of dystonia. Adv Neurol 1988;50:1-8. Fahn S, Jankovic J. Principles and practice of movement disorders. Philadelphia: Churchill Livingstone Elsevier; 2007. Goodchild RE, Dauer WT. Mislocalization to the nuclear envelope: an effect of the dystonia-causing torsinA mutation. Proc Nat Acad Sci USA 2004; 101(3):847-52. Risch N, De Leon D, Ozelius L, Kramer P, Almasy L, Singer B, Fahn S, Breakefield X, Bressman S. Genetic analysis of idiopathic torsion dystonia in Ashkenazi Jews and their recent descent from a small founder population. Nat Genet 1995;9:152-9. Risch NJ, Bressman SB, Senthil G, Ozelius LJ. Intragenic Cis and Trans modification of genetic susceptibility in DYT1 torsion dystonia. Am J Hum Genet 2007; 80(6):1188-93. Bressman SB, Sabatti C, Raymond D, de Leon D, Klein C, Kramer PL, Brin MF, Fahn S, Breakefield X, Ozelius LJ, Risch NJ. The DYT1 phenotype and guidelines for diagnostic testing. Neurology 2000;54:1746-52. McNaught KS, Kapustin A, Jackson T, Jengelley TA, JnoBaptiste R, Shashidharan P, Perl DP, Pasik P, Olanow CW. Brainstem pathology in DYT1 primary torsion dystonia. Ann Neurol 2004;56(4):540-7. Nutt JG, Muenter MD, Aronson A, Kurland LT, Melton LJ. Epidemiology of focal and generalized dystonia in Rochester, Minnesota. Mov Disord 1988; 3:188-94. Carbon M, Kingsley PB, Su S, Smith GS, Spetsieris P, Bressman S, Eidelberg D. Microstructural white matter changes in carriers of the DYT1 gene mutation. Ann Neurol 2004;56(2):283-6. Carbon M, Kingsley PB, Tang C, Bressman S, Eidelberg D. Microstructural white matter changes in primary torsion dystonia. Mov Disord 2008;23(2): 234-9. Eidelberg D, Moeller JR, Ishikawa T, Dhawan V, Spetsieris P, Przedborski S, Fahn S. The metabolic topography of idiopathic torsion dystonia. Brain 1995;118:1473-84.
1777
1778
106
Diagnosis and medical management of dystonia
18. Eidelberg D, Moeller JR, Antonini A, Kazumata K, Nakamura T, Dhawan V, Spetsieris P, de Leon D, Bressman SB, Fahn S. Functional brain networks in DYT1 dystonia. Ann Neurol 1998;44:303-12. 19. Hallett M. Pathophysiology of dystonia. J Neural Transm Suppl 2006;(70):485-8. 20. Goto S, Lee LV, Munoz EL, Tooyama I, Tamiya G, Makino S, Ando S, Dantes MB, Yamada K, Matsumoto S, Shimazu H, Kuratsu J, Hirano A,
Kaji R. Functional anatomy of the basal ganglia in Xlinked recessive dystonia-parkinsonism. Ann Neurol 2005; 58(1):7-17. 21. Burke RE, Fahn S, Marsden CD. Torsion dystonia: a double-blind, prospective trial of high-dosage trihexyphenidyl. Neurology 1986;36:160-4. 22. Greene P, Shale H, Fahn S. Analysis of open-label trials in torsion dystonia using high dosages of anticholinergics and other drugs. Mov Disord 1988;3:46-60.
109 Functional Stereotactic Procedures for Treatment of Secondary Dystonia H-H. Capelle . J. K. Krauss
Classification and Differential Diagnosis of Secondary Dystonia Dystonia is a movement disorder characterized by patterned, repetitive, phasic, or tonic sustained muscle contractions that produce abnormal, often twisting, postures or repetetive movements [1,2]. There is a wide variety of causes which may result in developing dystonia. It is important to identify those causes in order to optimize treatment for each individual patient. The widely accepted classification by Fahn et al. allows to categorize dystonia along three dimensions [3]. Generally, dystonia can be divided by age of onset, distribution of dystonic symptoms, and by etiology. Primary (idiopathic) dystonia is distinguished from secondary (symptomatic) dystonia by otherwise normal findings in clinical examinations, structural neuroimaging, examination of cerebrospinal fluid, routine electrophysiology, and standard laboratory testing [2,3]. Whereas an increasing number of primary dystonic syndromes have been attributed to a genetic predisposition in recent years [4], secondary dystonia is attributed to a hereditary neurologic disorder or develops as a result of an exogenous insult to the brain [3]. If dystonia is termed secondary, other neurologic abnormalities may be present in addition to dystonic symptoms including other movement disorders. These may include ataxia, spasticity, dementia, paresis, chorea, or parkinsonism which do not occur in primary dystonia [3,5]. Epidemiological data are not available for most forms of secondary dystonia. The prevalence #
Springer-Verlag Berlin/Heidelberg 2009
for secondary dystonia is hard to estimate with regard to the various causes and the frequent misdiagnoses, especially, if dystonia is only part of a more extensive syndrome. Data from crosssectional studies indicate that the prevalence of movement disorders after treatment with dopamine receptor blockers may range between 0.5 and 21% [6–8]. In a study of Miller and Jankovic on 125 patients with drug-induced movement disorders, 24% developed tardive dystonia [9]. The overall incidence rate for tardive movement disorders after neuroleptic medication has been estimated to be 4–12%, hence, tardive dystonia may be present in a small fraction of these patients [10,11]. According to the anatomical distribution of dystonia, focal dystonia is restricted to a single, specific part of the body, involving mostly cranial and cervical muscles. Typical examples are cervical dystonia, blepharospasm, or spasmodic dysphonia [3,5,12]. Segmental dystonia involves two or more adjacent body parts. Multifocal dystonia involves two or more noncontiguous body parts. Hemidystonia involves the ipsilateral arm and leg, and often results from a lesion involving the contralateral basal ganglia. Generalized dystonia involves the body axis, and the lower and upper extremities. The age of onset of dystonia shows two distinct peaks, at 9 years and at 45 years. Accordingly, dystonia can be classified as early-onset or lateonset dystonia [5,12]. The association between age at symptom onset and disease course is somewhat different for patients with secondary dystonia than
1836
109
Functional stereotactic procedures for treatment of secondary dystonia
for patients with primary dystonia [3]. Generally, for those with primary dystonia, the younger the patient at symptom onset, the greater the likelihood that the dystonia will become generalized. However, secondary dystonia beginning during adulthood is more likely to generalize than adultonset primary dystonia [3,5]. In the original etiological classification by Fahn, there are four subdivisions of dystonia including primary dystonia, dystonia-plus syndromes, secondary dystonia, and heredodegenerative disorders [3]. The term ‘‘secondary dystonia’’ was used for a dystonic disorder which develops due to acquired and exogenous causes. In more recent classifications the term ‘‘secondary dystonia’’ or ‘‘symptomatic dystonia’’ encompasses dystonia associated with (1) hereditary neurologic syndromes, including the dystonia-plus syndromes and the heredodegenerative disorders, (2) dystonia due to environmental insults to the brain, (3) dystonia occurring with Parkinson’s disease and other parkinsonian syndromes, and 4) other movement disorders with dystonic phenomenology like tics or paroxysmal dyskinesias [5] (> Table 109-1). The dystonia-plus syndromes which are defined as ‘‘neurochemical disorders’’ are clearly distinguishable from the heredodegenerative disorders by the absence of neuronal degeneration [3,5]. Dystonia, as well as additional features, such as myoclonus characterized by sudden, involuntary, ‘‘shock-like’’ muscle contractions, and parkinsonism are typical features of the dystoniaplus syndromes, which include the following entities: (1) Rapid-onset dystonia-parkinsonism, (2) myclonus-dystonia syndrome, and (3) doparesponsive dystonia [3,13]. Heredodegenerative disorders with dystonia include various conditions which are mostly autosomal recessive syndromes with metabolic abnormalities such as pantothenate kinase deficiency (PKAN) or Wilson’s disease. Other conditions are autosomal dominant, X-linked recessive, or mitochondrially inherited diseases
such as Huntington’s disease, rare disorders like Lubag’s disease and Leigh syndrome. All these conditions include dystonia as a part of their phenotype [3,5]. Many exogenous factors can cause secondary dystonia. These causes include specific drugs, toxins like cyanide or manganese, cerebral palsy, kernikterus, cerebrovascular disease, cerebral infections and postinfectious states, stroke, encephalitis, brain tumors, demyelination, and structural abnormalities [3]. Various forms of dystonia with different anatomic distribution may be caused by manganese intoxication including a dystonicparkinsonian syndrome [14], generalized dystonic movements of the limbs and trunk, or focal dystonias such as blepharospasm and cervical dystonia [15]. Other toxins which were described to cause dystonia are carbon monoxide, carbon disulfide, methanol, cyanide, or disulfiram [16–18]. Structural lesions in the basal ganglia and the thalamus, but also in the brainstem and the parietal lobe resulting from tumors, infections, cerebrovascular disease and demyelination may induce dystonia [19–23]. Severe dystonia may develop with a delay of sometimes several years after perinatal asphyxia or kernikterus. The concept that dystonia can also develop secondary to a traumatic insult to the central nervous system is widely accepted [20,24]. Dystonia may develop after severe head trauma and the most frequent manifestation is hemidystonia [25]. The onset of dystonic symptoms varies from months to years. Injury to the central nervous system may play an important role in ‘‘triggering’’ dystonia onset in patients who have a genetic predisposition to develop dystonia [26]. The concept that peripheral lesions to neural structures may be involved in the pathogenesis of dystonia has been discussed controversially, but it has gained more widespread acceptance recently [27–30]. For example, blepharospasm has been reported following localized eye disease and cervical dystonia subsequent to whiplash injury. Up to 20% of patients with cervical
Functional stereotactic procedures for treatment of secondary dystonia
109
. Table 109-1 Causes for the development of secondary dystonia (Adopted from [5]) ACQUIRED/EXOGENOUS CAUSES Vascular Cerebrovascular or ischemic injury Arteriovenous malformation Perinatal cerebral injuries Perinatal cerebral injury Kernikterus Infectious Viral encephalitis Subacute sclerosing panencephalitis AIDS Creutzfeldt-Jakob disease Trauma Head trauma Peripheral trauma (associated with complex regional pain syndrome) Cervical cord injury Brain tumor Toxins - Manganese, carbon monoxide, carbon disulfide, methanol, disulfiram, wasp sting Drugs - Levodopa, dopamine agonists, antipsychotics, metoclopramide, fenfluramine, flecainide, ergot agents, anticonvulsant agents, certain calcium channel blockers Demyelination - Multiple sclerosis DYSTONIA-PLUS SYNDROMES Myoclonus dystonia Dopa-responsive dystonia (DRD) Rapid-onset dystonia parkinsonism HEREDITARY NEUROLOGIC SYNDROMES ASSOCIATED WITH NEURODEGENERATION (autosomal recessive, autosomal dominant, X-linked recessive, mitochondrial) Wilson disease Amino acid disorders Glutaric academia Methylmalonic academia Homocystinuria Hartnup disease Tyrosinosis Lipid disorders Metachromatic leukodystrophy Neuronal ceroid lipofuscinosis Dystonic lipidoses - Niemann-Pick disease, type C (i.e., sea blue histiocytosis) Primary antiphospholipid antibody syndrome Gangliosidoses (i.e., GM1, GM2) Mitochondrial encephalopathies (e.g., Leigh disease, Leber disease) Lesch-Nyhan syndrome Triosephosphate isomerase deficiency Vitamin E deficiency Biopterin deficiency Lubag or X-linked dystonia parkinsonism Pantothenate kinase-associated neurodegeneration (former known as: Hallervorden-Spatz disease) Hypobetalipoproteinemia, acanthocytosis, retinitis pigmentosa, pallidal degeneration (HARP) syndrome Neuroacanthocytosis Spinocerebellar ataxia (SCA), types 1, 2, or 3 Ataxia telangiectasia Huntington’s disease Dentatorubropallidoluysian atrophy
1837
1838
109
Functional stereotactic procedures for treatment of secondary dystonia
. Table 109-1 (Continued) DYSTONIA DUE TO PARKINSON’S DISEASE AND DEGENERATIVE PARKINSONIAN DISORDERS OF UNKNOWN ETIOLOGY Parkinson’s disease (PD) Progressive supranuclear palsy (PNP) Multiple system atrophy Corticobasal ganglionic degeneration (CBGD) OTHER MOVEMENT DISORDERS WITH DYSTONIC PHENOMENOLOGY Tics Paroxysmal kinesigenic dyskinesias Paroxysmal non-kinesigenic dyskinesias episodic ataxia syndromes PSYCHOGENIC
dystonia reported a previous head or neck trauma. Spasmodic dysphonia may be triggered by laryngeal viral infection [31]. Sensory abnormalities like pain or discomfort have been described as early manifestations of dystonia after peripheral trauma [30,32]. Patients may also develop movement disorders after spinal surgery [33]. In a series of six patients, a movement disorder became manifest with a delay of 1 day to 12 months after surgery. In contrast to secondary dystonia after cerebral injury, peripherallyinduced dystonias usually show a relatively short period between the event and symptom onset and they are generally associated with pain syndromes (e.g., reflex sympathetic dystrophy) [5]. In general, there is no temporary improvement with ‘‘sensory tricks’’ in peripherally induced dystonias (i.e., no gestes antagonistes). Tardive dystonia is a dystonic syndrome resulting from chronic neuroleptic medication whereas acute dystonic disorders occur only for hours after neuroleptic exposure [34–37]. Most neuroleptic agents such as dopamine receptor blocking agents, but also medication for gastrointestinal discomfort may induce tardive dystonia [38]. Patients with Parkinson’s disease (PD) frequently develop painful foot dystonia while reducing levodopa medication. Camptocormia, a focal dystonia with twisting or tonic activation of the abdominal wall muscles has also been reported in association with Parkinson’s disease [39,40]. Paroxysmal dyskinesias and other movement
disorders, for example tics, may produce postures resembling dystonia [3]. As shown in some series Parkinson-plus syndromes like multiple system atrophy and progressive supranuclear palsy are associated with dystonia in up to 46% of patients [41,42]. It is important to delineate secondary dystonia from primary dystonia by detailed clinical evaluation in order to plan treatment. Before surgical treatment is considered for secondary dystonia the underlying disease should be diagnosed. Mostly, in secondary dystonia the symptoms manifest rather in an atypical site of the body for the age of onset [5]. A history of an exogenous insult to the brain may suggest that dystonia is secondary. Early speech abnormalities are another indicator for a symptomatic etiology of the dystonia. Dystonia that occurs during periods of rest, rather than during voluntary action is more likely to be secondary dystonia. The clinical appearance of dystonia might also give a clue about the etiology. The manifestation of hemidystonia is almost exclusively associated with a traumatic injury or other exogenous insult to the brain. The structures involved are the putamen or the caudate nucleus. Another frequent manifestation of secondary dystonia is choreoathetosis which is defined as the occurrence of involuntary movements in a combination of chorea and athetosis. There are multiple causes for choreoathetosis: cerebral palsy (the most common cause) which often occurs with
Functional stereotactic procedures for treatment of secondary dystonia
delayed onset, encephalitis, trauma, tumors, but also several drugs, Lesch-Nyhan syndrome, and neurodegenerative disorders like Huntington’s disease or PKAN. Treatment for secondary dystonias due to a neurological or metabolic disease is usually directed by the specific requirements of that disorder. Patients, especially children, with early-onset dystonia should be evaluated with specific tests to help exclude Wilson’s disease (e.g., slit lamp examination, serum ceruloplasmin). If patient and family history and physical examination reveal certain symptoms, signs, and physical findings suggestive of secondary or heredodegenerative dystonias, more extensive diagnostic testing is recommended, such as enzymatic studies and MRI imaging studies. First, a structural lesion in the basal ganglia associated with dystonia may be found. In Huntington’s disease and neuroacanthocytosis atrophy of the caudate nucleus is a typical finding. Several abnormalities in the MRI scans which were described as the ‘‘face-of-the-giant-panda’’ are found in Wilson’s disease [43]. The ‘‘eye-of-thetiger’’ sign has been associated with PKAN [44]. The evaluation of patients with secondary dystonia may include also examination of the cerebrospinal fluid and blood for the specific metabolites of neurodegenerative and neurochemical disorders listed in > Table 109-1. Factors which should raise the suggestion for a secondary etiology of
109
dystonia are summarized in > Table 109-2 and they should be kept in mind when patients with dystonia are examined.
Surgical Procedures for Secondary Dystonia Stereotactic functional procedures should be considered in individuals who are disabled and do not respond to conservative treatment options. Current functional stereotactic surgical options include: (1) radiofrequency lesioning of the globus pallidus internus (GPi; pallidotomy) and the thalamus (thalamotomy) and (2) chronic deep brain stimulation (DBS) of the GPi and the thalamus. Depending on the distribution of dystonia, the severity, the etiology, the presence of other neurological symptoms, the patients’ age, and the goals to be achieved in each patient, the indications for the available surgical options have to be well considered. Some dystonias, for example hemidystonia after exogenous insult to the caudate nucleus or putamen, may be stable after delayed onset and progression over years. In contrast, other secondary dystonias like tardive dystonia which manifest mostly with a focal or segmental distribution of symptoms may progress over years to other areas of the body and then limit the possible benefit of functional surgery.
. Table 109-2 Summary of clinical signs and hints which may raise the consideration that dystonia is secondary (Adopted from [5]) Clinical signs and diagnostic features suggesting secondary dystonia Clinical history for insult to the brain (head trauma, encephalitis, perinatal hypoxia, drug exposure) Dystonia at rest Atypical site for development of dystonia concerning the age of onset Early development of speech abnormalities Hemidystonia Neurological abnormalities not associated with primary dystonia (parkinsonism, ataxia, dementia, seizures, myoclonus, visual loss, ocular motor deficits, deafness, dysphagia etc.) Other abnormalities in medical examination (hepatomegaly, splenomegaly, Kayser-Fleischer ring, malabsorption etc.) Hints for psychogenic origin (non-physiologic findings like false weakness, false sensory loss, inconsistent movements) Abnormal findings in MR imaging Abnormal findings in laboratory diagnostic
1839
1840
109
Functional stereotactic procedures for treatment of secondary dystonia
The advantage of radiofrequency lesioning procedures over DBS are the lower costs, no risk of equipment failure, the lower risk for infection and the absence of the need of periodic battery replacements. In most centers worldwide, however, DBS of the GPi and thalamus has more and more replaced radiofrequency lesioning because of the lower risk of performing bilateral surgery in one session. Moreover, it avoids concern about the impact of lesioning on the developing brain in children. Compared to ablative surgery, DBS offers a non-lesional modulation of basal ganglia output, its effects are principally reversible, and it is possible to adapt the therapy to the course of the disease and the individual needs of the patient [45–47]. General anesthesia is required only in patients with severe dystonia who may not tolerate stereotactic procedures under local anesthesia. Especially, in children general anesthesia should be chosen. In awake patients often the surgical procedure provides some degree of immediate improvement of the dystonic symptoms. The full effect of surgery has to be expected usually not until weeks or months postoperatively [48].
Vim are usually 13–15 mm lateral and 4 mm posterior to, and at the level of the midcommisural point. If an additional lesion is required, it is placed more anteriorly and medially to the former target.
Thalamotomy
Deep Brain Stimulation
The thalamus was used as a target for decades before the posteroventral lateral GPi has been established as the first line target for treatment of dystonia [47–50]. Over decades it was considered to be advantageous in secondary dystonia. The thalamic target of choice for treatment of dystonia was much more variable among different surgeons than the thalamic target for treatment of tremor. The targets included the nucleus ventralis oralis anterior and posterior (Voa/Vop), the ventralis oralis internus, the ventralis intermedius (Vim), the subthalamic region, the centrum medianum/nucleus parafascicularis complex (CM-Pf), and the pulvinar thalami [49]. Nowadays, the thalamic Vim is used almost exclusively. The coordinates for the thalamic
In the last years, the effect of both pallidal and thalamic DBS has been evaluated in the treatment of various types of secondary dystonia. The target is defined by standard stereotactic imaging or image fusion of CT and MR scans. The standard coordinates of the GPi are: 20–22 mm lateral to and 4 mm below the intercommissural line, and 2–3 mm anterior to the intercommissural midpoint [45]. In most centers bilateral GPi DBS is performed with the quadripolar 3387 DBS electrode (Medtronic, Minneapolis, MN, USA) with 1.5 mm spaces between the single contacts. In a study of the team from Queen’s square the optimal target was found to be in the ventral portion of the GPi or in the medullary lamina or both [51]. The target for
Pallidotomy The target for treatment of dystonia is located in the posteroventral lateral portion of the GPi. Microelectrode recordings are helpful to identify the external and internal borders of the GPi and the globus pallidus externus (GPe). Usually, two lesions are placed with a distance of 2 mm along the trajectory of the radiofrequency probe [45,48]. The effect of surgery on phasic movements may be apparent early postoperatively whereas the tonic dystonic movements may be improved only after a longer period of weeks to months. Pallidotomy for secondary dystonia yields the risk of contralateral weakness or hemiparesis if the lesion is too close to the internal capsule in unilateral surgery, and for speech disturbance in bilateral surgery.
Functional stereotactic procedures for treatment of secondary dystonia
thalamic DBS is the same target which is used in thalamotomy with the following standard coordinates: 12–14 mm lateral to and at the level of the intercommissural line, and 4 mm posterior to the intercommissural midpoint. The implantable pulse generator (IPG) is implanted directly after positioning of the electrodes or in a second session a few days later. To avoid sudden recurrence of dystonia if battery depletion occurs one strategy is to use two IPGs (Soletra, Medtronic) instead of one dual channel IPG (Kinetra, Medtronic).
Clinical Results Variable improvement may be achieved from basal ganglia manipulation in patients with refractory secondary dystonia, and the benefit may be maintained also in the long term. However, whereas the improvement after pallidal DBS in patients with primary dystonia ranges from 40–90% as shown by recent randomized controlled studies which have established GPi DBS as one of the mainstays of therapy in dystonia [52,53], the outcome after functional procedures (radiofrequency lesioning or DBS) in different types of secondary dystonia is much harder to appreciate. Overall, the evaluation of functional stereotactic procedures in the treatment of secondary dystonia is quite complex [54]. The main problems which do not allow to draw definitive conclusions from studies on secondary dystonia are the differences in the used rating scales, the targeted nuclei, the length of follow-up and the underlying various causes for secondary dystonia [49]. Furthermore, dystonia often is only one aspect of the disorder. In general, patients with secondary dystonia respond less well to functional procedures compared to the improvement in patients with primary dystonia, and more disappointing results have to be expected in patients with secondary dystonia due to structural lesions [55]. Radiofrequency lesioning of the thalamus and thalamic DBS has
109
been shown to be more effective in patients with secondary dystonia in some studies [56–58].
Multipatient Series with Mixed Etiology Thalamotomy One major issue of early series is the complete lack of evaluation by standard rating scales. A summary of the major thalamotomy series is given in > Table 109-3 [25,56,57,59–62]. Large historical series including also patients with secondary dystonia were published by Cooper and others to evaluate the effect of thalamotomy [61,63]. If the procedures were initially not effective a second procedure was added. The target of Cooper was usually the Vim, the ventral caudal thalamus and the medial thalamus. The mortality rate was 2%. Overall improvement was seen in 70% of patients with mild to moderate benefit in short term follow-up. Another large series on thalamotomy was published by Andrew et al. in 1983 [62]. The authors evaluated the effect of thalamotomy (uni- or bilateral) in 55 patients with various types of dystonia. Overall, symptoms were described to be improved by up to 50% in secondary dystonia. The patients with hemidystonia experienced improvement by 50–100% [64]. Follow-up was available up to 3 years in some cases. Tasker et al. [56] reported on the outcome after Vim thalamotomy in 49 patients with dystonia, of whom 29 suffered from secondary dystonia. Sixty eight percent of the patients showed a greater improvement than 25%, whereas only 50% of patients with primary dystonia showed similar improvement. In these patients the location of the target was refined by microelectrode recordings. Cardoso et al. described their experience with 17 patients with dystonia and hemiballism who underwent thalamotomy of more ventral Voa/Vop complex [57]. In 50%
1841
1842
109
Functional stereotactic procedures for treatment of secondary dystonia
. Table 109-3 Published series with thalamotomy in patients with various types of dystonia including secondary dystonia (Adopted from Ondo et al., 2001,[49])
Author, year Laitinen, 1965, [59] Cooper, 1969, [60] and 1976, [61] Tasker et al., 1988, [56] Andrew et al., 1983,[62] Krauss et al., 1992, [25] Cardoso et al., 1995,[57]
Type of secondary dystonia included in the series
Overall number of patients
Patients improved (%)
Complications rate (%)
Cerebral palsy, choreoathetosis
10
50
n.a
Cerebral palsy, choreoathetosis, various other types of secondary dystonia Various types of secondary dystonia
208
70
n.a
49
34
21
55
47
n.a
9
50
0
17
47
35
Hemidystonia and other types of secondary dystonia Posttraumatic dystonia, hemidystonia, torticollis Various types of secondary dystonia
of the patients with secondary dystonia there was a moderate to significant improvement at a mean-follow-up of 41 months. Overall 12% of the patients showed further improvement over time. More information on long-term follow-up was reported by Gros et al. [65]. The authors found a postoperative improvement of 51% of their patients, however, after 6 years there was a benefit in 33% from surgery [65]. In the series of Tasker and Cardoso, the patients with secondary dystonia appeared to have a more sustained benefit from thalamotomy than patients with primary dystonia. In the series of Tasker et al. only 31% patients with secondary dystonia versus 65% of patients having primary dystonia showed loss of initial improvement [56,57]. Complications of thalamotomy were found in up to 35% of patients in the series by Cardoso et al. which included confusion and contralateral hemiparesis [57]. One patient showed signs of persistent neurological deficits after a second procedure to enlarge the prior lesion. Overall, complication rates were reported in a wide range of 16–47% by other authors [48]. Bilateral procedures carry the additional risk of dysarthria and pseudobulbar palsy and they are clearly associated
with a higher rate of persistent side effects. Other possible complications include dysphagia, gait disturbance, transient numbness, and seizures.
Pallidotomy While there is only very limited data on the efficacy of pallidotomy on secondary dystonia from early studies, more reliable data became available after the renaissance of pallidotomy in the mid 1990s. When Iacono et al. reported on a pallidotomy series of four patients with dystonia, they noted that a 24-year-old patient with secondary generalized dystonia did not improve [66]. In a larger series Lin et al. reported on 18 patients with secondary dystonia due to cerebral palsy (8), hypoxic encephalopathy (6), carbon monoxide poisoning (2), and encephalitis (2). All patients underwent bilateral pallidotomy but obtained only minor benefit [67,68]. Postoperatively, there was only slow improvement of the dystonic symptoms. At one year follow-up there was an improvement by 13% in the BFM motor score and by 9% in the BFM disability score.
Functional stereotactic procedures for treatment of secondary dystonia
The most pronounced improvement of dystonia was found in the craniocervical region [68]. In a series from Baylor College of Medicine, 16 patients underwent pallidotomy for generalized or hemidystonia [48]. Among these patients there were five patients with secondary dystonia induced by trauma (2), intracerebral bleeding (1), and hypoxia (2). In four of the patients with secondary dystonia, dystonic symptoms started with a focal distribution and then developed to hemidystonia or generalized dystonia. In three patients unilateral surgery was performed, and two patients had bilateral pallidotomy. For evaluation the Burke-Fahn-Marsden (BFM) scale was used. After 3 months of follow-up, four of the patients with secondary dystonia showed a slight to moderate improvement in the BFM scores. However, in the patients with hemidystonia the improvement was much more pronounced which was also reflected by the improvement in the postoperative scores for the activities of daily living (ADL). The patient with global hypoxia experienced no beneficial improvement after a bilateral procedure. In contrast to the results from thalamotomies, the patients with primary dystonia presented with a more marked benefit and consistent improvement than the patients with secondary dystonia. Complications related to surgery included transient postoperative lethargy and mild hemiparesis. When Yoshor et al. compared the effectiveness of thalamotomy versus pallidotomy in patients with secondary dystonia, they found no significant difference in outcome after thalamotomy or pallidotomy [69]. Despite the heterogeneous characteristics of both patient cohorts, patients presenting with secondary dystonia achieved only modest improvement after both procedures. In carefully selected patients concurrent use of thalamotomy and same-side pallidotomy may be useful. There are several reports on patients who underwent thalamotomy with an unsatisfactory response but who benefited from an additional
109
pallidotomy. Krauss and Jankovic, for example, reported on a patient with hemidystonia secondary to gunshot injury of the carotid artery who underwent thalamotomy and pallidotomy later on with satisfactory outcome [24].
Thalamic DBS Thalamic DBS has been used for secondary dystonia before pallidal DBS became more popular. A retrospective study by Vercueil reported on 12 patients (4 with primary and 8 with secondary dystonia) who underwent thalamic DBS [58]. Follow-up was available from 6 months up to 11 years (in one patient). Thalamic stimulation did not result in improvement of the mean BFM scores, but 6 of the 12 patients experienced functional benefit. Two patients with secondary dystonia (generalized and hemidystonia) underwent pallidal DBS later which then resulted in a significant improvement in the dystonia movement scale and disability scores. On the other hand, Villemure et al. described successful thalamic stimulation of the Voa in two patients with secondary dystonia who had undergone pallidal stimulation earlier without benefit [70]. Another patient with postanoxic dystonia due to bilateral necrosis of the basal ganglia was also treated with bilateral thalamic DBS with major improvement after a delay of 4 months. Previous pallidal stimulation had not been effective, however, the stimulation period of only 6 weeks may have been too short to evaluate the true effect of GPi DBS in this case [71].
Pallidal DBS When Eltahawy et al. evaluated their results after pallidotomy and pallidal DBS in dystonia, they found a striking difference in outcome between primary and secondary dystonia. The mean percentage of improvement in postoperative BFM
1843
1844
109
Functional stereotactic procedures for treatment of secondary dystonia
. Figure 109-1 Axial (T2-weighted) and coronal (T1-weighted) MR scans show bilateral implantated quadripolar electrodes (model 3387, Medtronic Inc., Minneapolis) in the posteroventral lateral GPi
scores was only 10.6% for secondary dystonia. Among the six secondary dystonia patients, the only one who had consistent benefit and a normal brain MRI scan was diagnosed as having tardive dystonia [72]. In the already mentioned study by Vercueil et al. four patients with secondary dystonia achieved a mean improvement by more than 50% in the BFM score at a follow-up of 6–24 months [58]. A recent series including video assessments of nine patients with secondary dystonia with different etiology showed moderate to marked improvement in the BFM scores [73].
Multifocal DBS Based on the varying results in some patients as described above, multifocal DBS has been introduced to select the best target [74,75] (> Figure 109-2). The optimal target can be chosen either during test stimulation with the electrodes externalized for a few days or upon chronic stimulation. In the later case, two electrodes will be connected to the IPG, while the two others will be placed under the skin with an
isolated plug. If chronic stimulation does not show satisfactory improvement of symptoms the other electrodes can be connected after weeks or months. Trottenberg et al. reported on a patient who underwent multifocal DBS for treatment of tardive dystonia in whom thalamic DBS was shown to be ineffective, but who benefited from pallidal DBS [75]. In patients with severe choreoathetosis due to cerebral palsy in our center multifocal DBS of the GPi and Vim now is the preferred treatment strategy based on the earlier experience in these patients. Preliminary results of multifocal DBS in seven patients with medically refractory dystonia were reported earlier [74]. Three patients had secondary dystonia including peripherally-induced posttraumatic hemidystonia, head tremor secondary to head injury, and upper limb tremor associated with choreoathetosis due to cerebral palsy. In four patients, it was possible to determine the more effective stimulation site during test stimulation. Rating scores showed that posttraumatic dystonic head tremor was improved more markedly with thalamic stimulation. No improvement of dystonia was observed in the two other patients
Functional stereotactic procedures for treatment of secondary dystonia
109
. Figure 109-2 Axial MR scans of a 30-year-old patient with choreoathetosis who underwent bilateral bifocal implantation (GPi and thalamus) of quadripolar DBS electrodes to select the best site for chronic stimulation [74]
with secondary dystonia during test stimulation, and these two patients underwent chronic alternating stimulation of the GPi and the thalamus after implantation of pacemakers. A patient with dystonia-parkinsonism underwent chronic DBS both of the GPi and STN.
generalized and tardive dystonia [77]. The STN, to our opinion, cannot be recommended as a target of choice for DBS of dystonia, until more data become available. Interestingly, much earlier, lesioning of the subthalamic area also was proposed for treatment of hyperkinesias in choreoathetosis by other groups [65,78].
Subthalamic DBS The subthalamic nucleus (STN) has been introduced as a target in secondary dystonia only recently. In a Chinese study, nine patients with secondary dystonia underwent STN DBS. Among them, two were diagnosed as having tardive dystonia, one had posttraumatic dystonia, three had a history of perinatal anoxia, one had kernikterus, and two had no exact contributory history. Six patients underwent bilateral STN DBS, two had unilateral STN DBS, and one had left STN and right GPi DBS. Follow-up was available up to 3 years. The patients with tardive dystonia were those who responded the most from DBS [76]. Sun et al. observed similar results in patients with
Secondary Dystonia: Specific Entities Tardive Dystonia Tardive dystonia is a late complication of chronic neuroleptic medication which most frequently manifests as focal or segmental dystonia [36,37]. Most often tardive dystonia occurs in the frame of the more common picture of tardive dyskinesia. Both atypical and typical neuroleptics as well as antiemetics with an effect on the central dopamine receptors may cause tardive dystonia. Some antidepressants such as paroxetine and triptans such as sumatriptane are also suspect to induce
1845
1846
109
Functional stereotactic procedures for treatment of secondary dystonia
tardive dystonia. The severity of dystonia does not correlate with the time of exposure to the medication, and also short periods of medication intake may cause tardive dystonia with a certain delay [38,79–82]. In adults tardive dystonia sometimes may be difficult to distinguish from primary adult-onset dystonia [79]. Tardive dystonia may progress over time and than manifest as generalized dystonia [79]. There are only few reports on radiofrequency lesioning in tardive dystonia. One patient with severe antipsychotic drug-induced tardive dyskinesias had modest improvement after bilateral pallidotomy. Also, unilateral thalamotomy was reported to be effective in tardive dystonia [83]. A 66-year-old man with perioral dyskinesias and torticollis benefited markedly at 1 year postoperatively [84]. Tardive dystonia and tardive dyskinesias have been considered good indications for palldial DBS. In a series of five patients with tardive dystonia who underwent bilateral GPi DBS a mean improvement of 87% was reported in the BFM motor subscore [85]. The beneficial effect appeared early after surgery within days and it was sustained after 6 months in all patients and in two patients up to 39 months postoperatively. As indicated above, a patient reported earlier by the same group underwent bifocal DBS targeting the Vim and GPi. After bilateral stimulation of the GPi, the patient showed sustained improvement of the dystonic symptoms within hours. Stimulation of the Vim, however, did not decrease the movements and bifocal stimulation of both the GPi and the Vim did not show any additional benefit [75]. Long-term follow-up is available from a series from Japan including six patients who underwent bilateral pallidal surgery and were followed up for up to 21 months. At the last follow-up, the BFM motor subscore improved by an average of 86% and the BFM disability subscore by 80% [86]. A recent case report is interesting in the light of the effect of pallidal DBS on mood. A 62-year-old woman with major
depression underwent bilateral GPi DBS for neuroleptic-induced tardive dystonia. Postoperatively, the BFM score improved by 36% after 18 months and the Hamilton Depression Rating scale by 50%. This effect on the patient’s mood may reflect the close relationship between basal ganglia circuits sustaining motor as well as limbic functions which may be both modulated by DBS [87]. A major step forward was the French study on tardive dystonia published recently by Damier which included blinded video assessments [88]. In a prospective multicenter setting symptoms improved by more than 40% (mean improvement, 61%; range, 44–75%) in the first enrolled 10 patients, 6 months after bilateral pallidal surgery. The efficacy of the treatment was confirmed by a double-blind videotape based evaluation. The psychiatric status of all patients remained stable. Thobois et al. evaluated the changes of regional cerebral blood flow (rCBF) in five patients of the former study [89]. GPi stimulation resulted in a reduction of rCBF in the primary motor, and the prefrontal cortex. It was concluded that GPi stimulation in tardive dystonia primarily acts via reducing the excess of frontal cortical activation which is consistent with other forms of dystonia. In our experience, patients with tardive dystonia due to neuroleptic medication but without a history for psychiatric disorders respond also well to GPi DBS [90]. Four patients who received dopamine-antagonistic medication to treat ‘‘neurasthenia’’ showed sustained improvement after follow-up ranging from 12 months to 24 months. The STN has been used as a target for treatment of tardive dystonia. In the above mentioned study by Zhang et al. two patients with tardive dystonia benefited from a BFM decrease by more than 90% [76]. In order to have more specific guidelines on pallidal DBS in tardive dystonia a multicenter randomized sham-controlled study has been initiated in Germany.
Functional stereotactic procedures for treatment of secondary dystonia
Choreoathetosis and Dystonia due to Cerebral Palsy Dystonia with choreoathetosis due to infantile cerebral palsy poses a special problem since it might be the most common secondary dystonic movement disorder. Patients often are severely disabled. Overall, there is little contemporary experience with functional stereotactic neurosurgery for treatment of choreoathetosis. Thalamotomy has been used for decades to treat the dystonic symptoms in cerebral palsy. A review of the major reports shows that surgery may have a remarkable impact on patients’ quality of life when motor dysfunction is improved [49]. Usually, the Vim, or more anteriorly, the Voa/Vop complex was targeted. Studies published in the 1960s, in general, reported about 50% improvement in patients with dystonia due to perinatal injury and cerebral palsy [91–93]. Siegfried and colleagues recommended that also subthalamotomy has a beneficial effect on dyskinesias in cerebral palsy [78]. As with other early studies, however, comparison of the results from different series is difficult because of different rating scales used, the length of follow-up, and the heterogeneous groups of patients studied. Broggi et al. showed that the clinical results were stable over 1–4 years in a group of 33 patients after thalamotomy [94]. The best results were achieved for tremor and hyperkinesias whereas dystonia was improved to a lesser extent. Moreover, it was stated that unilateral symptoms were improved more by the operation than axial symptoms. The cerebellar dentate nucleus and other cerebellar structures have also been used as targets for the treatment for dystonia and hyperkinetic movements in cerebral palsy [95,96]. Galanda et al. reported on their experience with lesioning of the cerebellar dentate nucleus which was combined in some patients with radiofrequency lesioning of the pulvinar and other thalamic targets. They found favorable results for dystonia and gradual improvement of choreoathetoid movements [97].
109
More recently chronic stimulation of the anterior cerebellum has been reported to be useful [98]. Pallidotomy has regained popularity for treatment of choreoathetosis in the 1990s. In a series of 24 patients of whom most underwent bilateral pallidotomy, Teo et al. reported that 67% of patients achieved subjective improvement, but objective improvement was seen in only 42% [99]. The rate of transient complications was 50%, including swallowing and speech difficulties, somnolence, eyelid apraxia, mild hemiparesis, and confusion. The permanent complication rate was 17.5%. In a series of 18 patients of whom 8 were described to have cerebral palsy Lin et al. achieved only limited benefit with bilateral pallidotomies [68]. In a large series from Turkey, Imer et al. evaluated the effect of thalamotomy, pallidotomy, subthalamotomy, and lesions in the field of Forel in 85 patients suffering from various dystonias. Thirty four patients among this group had dystonia and choreoathetosis due to cerebral palsy. Persistent improvement was found in 55% of patients after a mean follow-up of 102 months. Combined lesioning in various targets was recommended to achieve optimal benefit [100]. There is only little experience with DBS in choreoathetosis. Excellent improvement of generalized choreoathetosis secondary to cerebral palsy was reported in a 13-year-old boy with a ‘‘dystonic storm’’ [101]. In another report moderate improvement of dystonia with pallidal DBS was noted in a 11-year old boy with generalized choreoathetosis secondary to cerebral palsy [102]. The Montpellier group reported about 30% improvement after 1 year and 40% after 3 years in a series of 21 patients with choreoathetosis [103]. Pallidal DBS in adult patients with choreoathetosis has been the subject of two studies. The mean improvement in the Burke-Fahn-Marsden dystonia rating (BFM) scale in four patients was 12% at 3 months postoperatively, 29% at 1 year, and 23% at 2 years which was not significant as compared to preoperatively [104]. Interestingly,
1847
1848
109
Functional stereotactic procedures for treatment of secondary dystonia
the subjective ratings of outcome in two patients who benefited moderately from pallidal DBS according to the ratings exceeded by far the objective evaluation. The difference between the subjective and objective improvement may reflect that also little improvement means a lot to these severely handicapped patients allowing some ease in daily living. The results of a larger recent French multicenter study were quite similar with an average improvement of 25% after 12 months followup [105]. Based on the current data it is difficult to give definitive recommendations for DBS in patients with choreoathetosis. The primary outcome measures in this group of patients possibly should be re-considered, that is the BFM motor score may not reflect fully the gain these patients achieve.
Dystonic Storm Secondary dystonia patients may rarely present with episodes of generalized, intense and potentially fatal exacerbation of muscle contractures, usually refractory to conventional pharmacological therapy. This clinical situation is referred to as status dystonicus (SD) or dystonic storm [106]. The data on surgical management of SD is rather heterogeneous [101,107]. Salvatory procedures performed include thalamotomy, pallidotomy or pallidal DBS. As indicated above, the Milano group achieved beneficial results in a child with refractory SD [101]. At the time of admission to the hospital, the patient presented with a lifethreatening ‘‘dystonic storm.’’ Preoperatively, continuous sedation and artificial respiration were required. On chronic GPi stimulation, the child’s neurological condition continuously improved. Seven months postoperatively, there was only residual dystonia. More recently, Schrock et al. also reported complete resolution of status dystonicus in two patients and improvement of dystonia by 94% after 1 week in the BFM scores in another patient [108].
Although rare, SD requires prompt diagnosis and therapeutic intervention in the ICU if needed, avoiding metabolic, renal, and ventilatory complications. Conventional clinical interventions may be ineffective and stereotactic neurosurgical procedures need to be explored in more detail [107].
Dystonia due to Acquired Craniocerebral Lesions Movement disorders after severe head injury have been reported to occur in 13–66% of patients [109]. The neuroradiological and pathological findings usually show lesions in the caudate, the putamen or the thalamus. In up to 5% of these patients kinetic tremor and dystonia may be a source of marked disability. The most frequent manifestation of posttraumatic dystonia is hemidystonia. This type of secondary dystonia is usually refractory to medical treatment. Hemidystonia has been shown to respond favorably to thalamotomy in the past with longlasting improvement in individual patients [62,64]. Long-term results after thalamotomy for posttraumatic dystonia were reported by Krauss et al. in a series of nine patients who developed dystonia secondary to head trauma [25]. Seven patients sustained a severe head trauma and two patients had mild head trauma. The hemidystonia occurred on the side of the posttraumatic hemiparesis with a delayed onset from 6 months to 4 years. Seven patients showed lesions of the contralateral putamen or caudate nucleus. During the early postoperative period all patients experienced beneficial improvement of their dystonia. Five patients showed transient side effects with increase of their hemiparesis in four and transient depression in one. In four patients there was still improvement of their hemidystonia after a mean follow-up of 18 years. There is less experience with pallidotomy in hemidystonia. A 15-year-old boy with hemidystonia
Functional stereotactic procedures for treatment of secondary dystonia
was reported to have 84% improvement at 2 years postoperatively [110]. Other groups reported also some beneficial clinical improvement after pallidotomy in hemidystonia [66,77]. In a series on 11 patients with hemidystonia who underwent lesioning of different targets (thalamotomy, pallidotomy, subthalamotomy) separately or in combination, beneficial improvement was obtained in 18.6% of the patients [100]. Thalamic DBS for hemidystonia may also yield some beneficial effect on hemidystonia. The first use of unilateral thalamic DBS in a 16year-old boy for hemidystonia after head injury resulted in excellent outcome with a follow-up of 8 months [111]. Ondo and Krauss reported on a unilateral thalamic stimulation in a 30-year-old woman who developed hemidystonia due to an abscess in her left putamen [48]. Thalamic DBS resulted in marked improvement with sustained benefit for 6 months. On the other hand, the Toronto group found only a minimal effect from thalamic DBS in two patients with hemidystonia. There is limited experience with pallidal DBS in hemidystonia. The group from Grenoble achieved beneficial improvement by pallidal DBS in one patient with hemidystonia who underwent prior thalamic DBS but without response [58].We reported beneficial long-term follow-up on a patient who underwent pallidal stimulation for treatment of posttraumatic hemidystonia [112]. The patient had developed left-sided low-frequency tremor and hemidystonia after severe head injury. After a thalamotomy, tremor was abolished but hemidystonia persisted. DBS of the right posteroventral GPi resulted in remarkable improvement of dystonia-associated pain, phasic dystonic movements and dystonic posture for follow-up of more than 6 years. Overall, the number of patients is too small and the results are too heterogeneous to give definite conclusions which procedure and target is most effective in treatment of disabling hemidystonia.
109
Dystonia due to Peripheral Injury Stereotactic surgery of peripherally-induced dystonia after trauma has been subject of only few reports. In one patient chronic thalamic DBS was effective for peripherally-induced dystonic paroxysmal nonkinesigenic dyskinesia [113]. Unilateral thalamic DBS resulted in a marked improvement and decrease of frequency, duration, and intensity of the dystonic paroxysmal movement disorder. Sustained improvement was achieved over 8 years of follow-up. Then, there was a relative loss of efficacy which was regained, however, when pallidal DBS was instituted [114]. We reported on ineffective DBS in a young woman who developed peripherally-induced dystonia with a tonic posture of her foot after fracture of her metatarsal bone. The patient had been referred with a misplaced electrode in the GPe from another hospital. A multifocal approach was performed targeting both the GPi and Vim, but there was no improvement of her dystonia with chronic stimulation (> Figure 109-3) [115].
Dystonia-plus Syndromes and Neurodegenerative Disorders Thus far, there is only little data available in this group of patients. Nevertheless, the observations and experiences which made may open new perspectives to treat also other neurodegenerative and metabolic disorders associated with dystonia.
Rapid-onset Dystonia Parkinsonism (RDP) There is little experience with functional procedures in RDP. Pittock and colleagues reported on a patient with inherited RDP, who underwent unilateral pallidotomy for torsion dystonia of the left arm which occurred at age 21, but with
1849
1850
109
Functional stereotactic procedures for treatment of secondary dystonia
. Figure 109-3 Peripherally-induced dystonia of the left lower extremity in a 37-year-old patient (a). Axial MR scans show electrodes placed in the right GPi and thalamus. The patient had been referred with a misplaced electrode in the GPe from another hospital (from Capelle et al., 2006; [115], with permission)
no obvious benefit [116]. Deutschlander et al. reported a young patient who underwent bilateral DBS of the GPi which was not successful to improve the dystonic or parkinsonian symptoms [117]. In another patient bilateral GPi DBS showed modest clinical benefit with 30% improvement in the BFM scores 3 months postoperatively [118]. We performed a multifocal approach in a 39-year-old patient with sporadic dystonia-parkinsonism stimulating the STN and GPi, bilaterally [119]. The patient first underwent STN DBS for treatment of his parkinsonian symptoms. Since the dystonic symptoms were not satisfactorily controlled, bilateral GPi DBS was added. With this bifocal approach using chronic bilateral STN and GPi DBS, tremor, rigidity, bradykinesia, and dystonia are well at a follow-up of 4 months. Bilateral GPi and STN DBS should be considered as an alternative treatment option in severe forms of dystoniaparkinsonism.
Myoclonus-Dystonia (MDS) More uniform improvement has been reported in those few cases who underwent chronic DBS for MDS. Trottenberg et al. found that thalamic DBS did not improve dystonia but myoclonus by 80%, while pallidal DBS also ameliorated dystonia [120]. Bilateral pallidal DBS resulted in marked improvement of both dystonia and myoclonus in two other patients [121,122]. Interestingly, in one study single unit firing patterns were modified by the myoclonic jerks [122] and in another study synchronization of pallidal local field potentials was found in the 3–15 Hz band [123].
Pantothenate Kinase-Associated Neurodegeneration (PKAN) Bilateral pallidotomy and bilateral pallidal DBS have been shown to be effective in patients with
Functional stereotactic procedures for treatment of secondary dystonia
PKAN [124–127]. The largest series on pallidal DBS has been reported by Castelnau et al. [128]. Six patients obtained major improvement of their painful spasms, dystonia, and functional autonomy after 6 months. The change in the BFM disability subscore was less distinctive, however, which was assumed to be due to the global neurological impairment. Despite the associated iron accumulation in the basal ganglia found in PKAN, the GPi seems not to be fully damaged still allowing effective stimulation [129]. Longterm follow-up is available in only one patient with sustained benefit 5 years after GPi DBS [130]. Since PKAN is a progressive disorder the benefit of pallidal DBS on dystonia might not outweigh the overall disability in this disorder on long-term. Nevertheless, we think that pallidal DBS should be offered to these patients to achieve benefit for at least a couple of years.
Lubag’s Disease X-linked dystonia-parkinsonism has primarily been reported in young men from the island of Panay in the Philippines. With disease progression, dystonia usually becomes generalized. In some patients, signs of parkinsonism may accompany or precede dystonia. In a patient with Lubag’s disease who underwent bilateral pallidal DBS, the BFM score was improved by 71% at 1-year follow-up [131]. Pallidal DBS may be an option for this syndrome which needs further exploration.
Lesch-Nyhan Syndrome Lesch-Nyhan syndrome is characterized by selfmutilating behavior and dystonia. Based on the proposed organization of the GPi in a limbic (anterior) and motor (posterior) part Cif et al. performed a multifocal approach in a 16-yearold boy with severe generalized dystonia and self-mutilating behavior [132]. Postoperatively,
109
anterior stimulation was found to improve the self-injurious behavior within days, whereas there was no effect on the movement disorder. Posterior GPi stimulation, however, improved only the dystonia. In another patient with bilateral GPi DBS only the self-mutilating behavior disappeared after chronic stimulation [133].
Huntington’s Disease Functional procedures in Huntington’s disease (HD) have been evaluated primarily for treatment of the severe choreatic movements. Pallidotomy and bilateral GPi DBS were reported to yield some clinical benefit [134]. The choreatic movements were reduced, and body weight, mood, and activities of daily living were broadly improved whereas the associated dystonic components were less improved [135,136]. Enforced by the improvement in the animal model of HD some groups evaluated neural transplantation in HD patients with modest improvement of the chorea [137,138]. Further trials are needed to evaluate the effectiveness of the different treatment options.
Dystonia Associated with Parkinson’s Disease Dystonia in PD most often is a consequence of pharmacological treatment and it manifests usually as off-dystonia. If dystonia is unrelated to PD treatment it may occur as blepharospasm or torticollis, more atypically as parkinsonian writer’s cramp or camptocormia. STN DBS in a PD patient with accompanying camptocormia was reported to be efficient for PD symptoms but not for camptocormia [40]. GPi DBS has been evaluated in treating off period dystonia in advanced PD [139]. In 10 patients with bilateral surgery, dystonia improved by 86%. Similar effects have been reported with STN DBS [140].
1851
1852
109
Functional stereotactic procedures for treatment of secondary dystonia
Conclusions The treatment of secondary dystonia is complex and it should be based primarily on the underlying cause. Functional stereotactic procedures are useful in patients with medically refractory dystonia. Since most disorders with secondary dystonia are rare randomized controlled studies with large numbers of patients are not expected to be available in the future. The question whether pallidal or thalamic surgery is more beneficial in some manifestations of secondary dystonia yet has to be resolved. Intraindividual comparison with alternating multifocal stimulation might be an option to approach this goal. It might be worthwhile also to explore new targets such as the CM-Pf for chronic DBS in secondary dystonia [141].
References 1. Fahn S. Concept and classification of dystonia. Adv Neurol 1988;50:1-8. 2. Fahn S. Dystonia: phenomenology, classification, etiology, genetics and pathophysiology. In: Fahn S, Jankovic J, Hallett M, Jenner P, editors. 16th annual course: a comprehensive review of movement disorders for clinical practitioner, Vol. 3. New York: Columbia University; 2006. p. 1-85. 3. Fahn S, Bressman S, Marsden CD. Classification of dystonia. Adv Neurol 1998;78:1-10. 4. Bhidayasiri R. Dystonia: genetics and treatment update. Neurologist 2006;12:74-85. 5. Geyer HL, Bressman SB. Diagnosis of dystonia. In: Warner TT, Bressman SB, editors. Clinical diagnosis and management of dystonia. London: Informa healthcare; 2007. p. 1-14. 6. Friedman JH, Kucharski LT, Wagner RL. Tardive dystonia in a psychiatric hospital. J Neurol Neurosurg Psychiatry 1987;50:801-3. 7. Yassa R, Nair V, Dimitry R. Prevalence of tardive dystonia. Acta Psychiatr Scand 1986;73:629-33. 8. Sethi KD, Hess DC, Harp RJ. Prevalence of dystonia in veterans on chronic antipsychotic therapy. Mov Disord 1990;5:319-21. 9. Miller LG, Jankovic J. Neurologic approach to druginduced movement disorders: a study of 125 patients. South Med J 1990;83:525-32. 10. Burke RE, Fahn S, Jankovic J, Marsden CD, Lang AE, Gollomp S, et al. Tardive dystonia: late-onset and
11.
12. 13.
14.
15.
16. 17.
18.
19.
20.
21.
22.
23.
24. 25.
26.
27.
28.
persistent dystonia caused by antipsychotic drugs. Neurology 1982;32:1335-46. Inada T, Yagi G, Kaijima K, Ohnishi K, Kamisada M, Rockhold RW. Clinical variants of tardive dyskinesia in Japan. Jpn J Psychiatry Neurol 1991;45:67-71. Bressman SB. Dystonia genotypes, phenotypes, and classification. Adv Neurol 2004;94:101-7. Bressman SB, Tagliati M, Klein C. Genetics of dystonia. In: Brin MF, Comella C, Jankovic J, editors. Dystonia monograph. Philadelphia: Lippincott, Williams & Wilkins: 2003. Calne DB, Chu NS, Huang CC, Lu CS, Olanow W. Manganism and idiopathic parkinsonism: similarities and differences. Neurology 1994;44:1583-6. Pal PK, Samii A, Calne DB. Manganese neurotoxicity: a review of clinical features, imaging and pathology. Neurotoxicology 1999;20:227-38. Choi IS, Cheon HY. Delayed movement disorders after carbon monoxide poisoning. Eur Neurol 1999;42:141-4. Grandas F, Artieda J, Obeso JA. Clinical and CT scan findings in a case of cyanide intoxication. Mov Disord 1989;4:188-93. Krauss JK, Mohadjer M, Wakhloo AK, Mundinger F. Dystonia and akinesia due to pallidoputaminal lesions after disulfiram intoxication. Mov Disord 1991;6:166-70. Kostic´ VS, Stojanovic´-Svetel M, Kacar A. Symptomatic dystonias associated with structural brain lesions: report of 16 cases. Can J Neurol Sci 1996;23:53-6. Lee MS, Rinne JO, Ceballos-Baumann A, Thompson PD, Marsden CD. Dystonia after head trauma. Neurology 1994;44:1374-8. Burguera JA, Bataller L, Valero C. Action hand dystonia after cortical parietal infarction. Mov Disord 2001;16:1183-5. Krauss JK, Mohadjer M, Nobbe F, Scheremet R. Hemidystonia due to a contralateral parieto-occipital metastasis: disappearance after removal of the mass lesion. Neurology 1991;41:1519-20. LeDoux MS, Brady KA. Secondary cervical dystonia associated with structural lesions of the central nervous system. Mov Disord 2003;18:60-9. Krauss JK, Jankovic J. Hemidystonia secondary to carotid artery gunshot injury. Childs Nerv Syst 1997;13:285-8. Krauss JK, Mohadjer M, Braus DF, Wakhloo AK, Nobbe F, Mundinger F. Dystonia following head trauma: a report of nine patients and review of the literature. Mov Disord 1992;7:263-72. Jankovic J, Van der Linden C. Dystonia and tremor induced by peripheral trauma: predisposing factors. J Neurol Neurosurg Psychiatry 1988;51:1512-9. Frucht S, Fahn S, Ford B. Focal task-specific dystonia induced by peripheral trauma. Mov Disord 2000;15:348-50. Sankhla C, Lai EC, Jankovic J. Peripherally induced oromandibular dystonia. J Neurol Neurosurg Psychiatry 1998;65:722-8.
Functional stereotactic procedures for treatment of secondary dystonia
29. Weiner WJ. Can peripheral trauma induce dystonia? No! Mov Disord 2001;16:13-22. 30. Jankovic J. Can peripheral trauma induce dystonia and other movement disorders? Yes! Mov Disord 2001;16:7-12. 31. Schweinfurth JM, Billante M, Courey MS. Risk factors and demographics in patients with spasmodic dysphonia. Laryngoscope 2002;112:220-3. 32. Ghika J, Regli F, Growdon JH. Sensory symptoms in cranial dystonia: a potential role in the etiology? J Neurol Sci 1993;116:142-7. 33. Capelle HH, Wo¨hrle JC, Weigel R, Ba¨zner H, Grips E, Krauss JK. Movement disorders after intervertebral disc surgery: coincidence or causal relationship? Mov Disord 2004;19:1202-8. 34. Keegan DL, Rajput AH. Drug induced dystonia tarda: treatment with L-dopa. Dis Nerv Syst 1973;34:167-9. 35. Keepers GA, Casey DE. Prediction of neuroleptic-induced dystonia. J Clin Psychopharmacol 1987;7:342-5. 36. Faurbye A, Rasch Pj, Petersen Pb, Brandborg G, Pakkenberg H. Neurological symptoms in pharmacotherapy of psychoses. Acta Psychiatr Scand 1964;40:10-27. 37. Crane GE. Persistent dyskinesia. Br J Psychiatry 1973;122:395-405. 38. Edwards MJ, Bhatia KP. Drug-induced and tardive dystonia. In: Warner TT, Bressman SB, editors. Clinical diagnosis and management of dystonia. London: Informa healthcare; 2007. p.189-207. 39. Sławek J, Derejko M, Lass P. Camptocormia as a form of dystonia in Parkinson’s disease. Eur J Neurol 2003;10:107-8. 40. Azher SN, Jankovic J. Camptocormia: pathogenesis, classification, and response to therapy. Neurology 2005;65:355-9. 41. Barclay CL, Lang AE. Dystonia in progressive supranuclear palsy. J Neurol Neurosurg Psychiatry 1997;62 (4):352-6. 42. Boesch SM, Wenning GK, Ransmayr G, Poewe W. Dystonia in multiple system atrophy. J Neurol Neurosurg Psychiatry 2002;72:300-3. 43. Jacobs DA, Markowitz CE, Liebeskind DS, Galetta SL. The ‘‘double panda sign’’ in Wilson’s disease. Neurology 2003;61:969. 44. Sethi KD, Adams RJ, Loring DW, el Gammal T. Hallervorden-Spatz syndrome: clinical and magnetic resonance imaging correlations. Ann Neurol 1988;24:692-4. 45. Krauss JK, Grossman RG. Principles and techniques of movement disorders surgery. In: Krauss JK, Jankovic J, Grossman RG, editors. Surgery for parkinson’s disease and movement disorders. Philadelphia: Lippincott, Williams & Wilkins; 2001. p. 74-109. 46. Krauss JK. Deep brain stimulation for dystonia in adults: overview and developments. Stereotact Funct Neurosurg 2002;78:168-82. 47. Krauss JK, Yianni J, Loher TJ, Aziz TZ. Deep brain stimulation for dystonia. J Clin Neurophysiol 2004;21:18-30.
109
48. Ondo WG, Krauss JK. Surgical therapies for dystonia. In: Brin MF, Comella C, Jankovic J, editors. Dystonia monograph. Philadelphia: Lippincott, Williams & Wilkins; 2003. p. 125-147. 49. Ondo WG, Desaloms M, Krauss JK, Jankovic J, Grossman RG. Pallidotomy and thalamotomy for dystonia. In: Krauss JK, Jankovic J, Grossman RG, editors. Surgery for arkinson’s disease and movement disorders. Philadelphia: Lippincott, Williams & Wilkins; 2001. p. 299-306. 50. Krauss JK, Pohle T, Weber S, Ozdoba C, Burgunder JM. Bilateral stimulation of globus pallidus internus for treatment of cervical dystonia. Lancet 1999;354:837-8. 51. Tisch S, Zrinzo L, Limousin P, Bhatia KP, Quinn N, Ashkan K, et al. The effect of electrode contact location on clinical efficacy of pallidal deep brain stimulation in primary generalized dystonia. J Neurol Neurosurg Psychiatry 2007;78:1314-9. 52. Kupsch A, Benecke R, Muller J, Trottenberg T, Schneider GH, Poewe W, et al. Deep-brain stimulation for dystonia study Group: pallidal deep-brain stimulation in primary generalized or segmental dystonia. N Engl J Med 2006;355:1978-1990. 53. Vidailhet M, Vercueil L, Houeto JL, Krystkowiak P, Benabid AL, Cornu P, et al. French Stimulation du Pallidum Interne dans la Dystonie (SPIDY) Study Group: bilateral deep-brain stimulation of the globus pallidus in primary generalized dystonia. N Engl J Med 2005;352:459-67. 54. Albanese A, Barnes MP, Bhatia KP, FernandezAlvarez E, Filippini G, Gasser T, et al. A systematic review on the diagnosis and treatment of primary (idiopathic) dystonia and dystonia plus syndromes: report of an EFNS/ MDS-ES Task Force. Eur J Neurol 2006;13:433-44. 55. Alkhani A, Khan F, Lang AE, Hutchison WD, Dostrovsky J. The response to pallidal surgery for dystonia is dependent on the etiology. Neurosurgery 2000;47:504. 56. Tasker RR, Doorly T, Yamashiro K. Thalamotomy in generalized dystonia. Adv Neurol 1988;50:615-31. 57. Cardoso F, Jankovic J, Grossman RG, Hamilton WJ. Outcome after stereotactic thalamotomy for dystonia and hemiballismus. Neurosurgery 1995;36:501-507. 58. Vercueil L, Pollak P, Fraix V, Caputo E, Moro E, Benazzouz A, et al. Deep brain stimulation in the treatment of severe dystonia. J Neurol 2001;248:695-700. 59. Laitinen LV. Short-term results of stereotaxic treatment for infantile cerebral palsy. Confin Neurol 1965;26:258-63. 60. Cooper IS. Dystonia musculorum deformans: natural history and neurosurgical alleviation. J Pediatr 1969;74:585-92. 61. Cooper IS. 20-year follow-up study of the neurosurgical treatment of dystonia musculorum deformans. Adv Neurol 1976;14:423-52. 62. Andrew J, Fowler CJ, Harrison MJ. Stereotaxic thalamotomy in 55 cases of dystonia. Brain 1983;106:981-1000. 63. Cooper IS. Symposium on diseases of the basal ganglia: physiological and clinical implications of thalamic
1853
1854
109 64.
65.
66.
67. 68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
Functional stereotactic procedures for treatment of secondary dystonia
surgery for dystonia and torticollis. Rev Bras Med 1968;25:774-80. Andrew J, Fowler C, Harrison MJ. Hemi-dystonia due to focal basal ganglia lesion after head injury and improved by stereotaxic thalamotomy. J Neurol Neurosurg Psychiatry 1982;45:276. Gros C, Frerebeau P, Perez-Dominguez E, Bazin M, Privat JM. Long term results of stereotaxic surgery for infantile dystonia and dyskinesia. Neurochirurgia 1976;19:171-8. Iacono RP, Kuniyoshi SM, Schoonenberg T. Experience with stereotactics for dystonia: case examples. Adv Neurol 1998;78:221-6. Lin JJ, Lin SZ, Chang DC. Pallidotomy and generalized dystonia. Mov Disord 1999;14:1057-9. Lin JJ, Lin SZ, Lin GY, Chang DC, Lee CC. Treatment of intractable generalized dystonia by bilateral posteroventral pallidotomy–one-year results. Zhonghua Yi Xue Za Zhi (Taipei) 2001;64:231-8. Yoshor D, Hamilton WJ, Ondo W, Jankovic J, Grossman RG. Comparison of thalamotomy and pallidotomy for the treatment of dystonia. Neurosurgery 2001;48:818-24. Villemure JG, Vingerhoets F, Temperli P. Pallidal or thalamic target. Acta Neurochir 2000;142:1194 (Abstract). Ghika J, Villemure JG, Miklossy J, Temperli P, Pralong E, Christen-Zaech S, et al. Postanoxic generalized dystonia improved by bilateral Voa thalamic deep brain stimulation. Neurology 2002;58:311-3. Eltahawy HA, Saint-Cyr J, Giladi N, Lang AE, Lozano AM. Primary dystonia is more responsive than secondary dystonia to pallidal interventions: outcome after pallidotomy or pallidal deep brain stimulation. Neurosurgery 2004;54:613-19. Pretto TE, Dalvi A, Kang UJ, Penn RD. A prospective, blinded evaluation of deep brain stimulation for the treatment of secondary dystonia and primary torticollis syndromes. Mov Disord 2008;Suppl 1:166 (Abstract). Krauss JK, Capelle HH, Blahak C, Ba¨zner H, Weigel R, Wo¨hrle JC. New treatment paradigm in dystonic movement disorders: multifocal deep brain stimulation. Mov Disord 2004; Suppl 9:285 (Abstract). Trottenberg T, Paul G, Meissner W, Maier-Hauff K, Taschner C, Kupsch A. Pallidal and thalamic neurostimulation in severe tardive dystonia. J Neurol Neurosurg Psychiatry 2001;70:557-9. Zhang JG, Zhang K, Wang ZC, Ge M, Ma Y. Deep brain stimulation in the treatment of secondary dystonia. Chin Med J (Engl) 2006;119:2069-74. Sun B, Chen S, Zhan S, Le W, Krahl SE. Subthalamic nucleus stimulation for primary dystonia and tardive dystonia. Acta Neurochir Suppl 2007;97:207-14. Siegfried J. Surgical therapy of extrapyramidal syndromes. Bibl Psychiatr Neurol 1969;139:693-6.
79. Kiriakakis V, Bhatia KP, Quinn NP, Marsden CD. The natural history of tardive dystonia. A long-term follow-up study of 107 cases. Brain 1998;121:2053-66. 80. Arnone D, Hansen L, Kerr JS. Acute dystonic reaction in an elderly patient with mood disorder after titration of paroxetine: possible mechanisms and implications for clinical care. J Psychopharmacol 2002;16:395-7. 81. Jarecke CR, Reid PJ. Acute dystonic reaction induced by a monoamine oxidase inhibitor. J Clin Psychopharmacol 1990;10:144-5. 82. Garcia G, Kaufman MB, Colucci RD. Dystonic reaction associated with sumatriptan. Ann Pharmacother 1994;28:1199. 83. Weetman J, Anderson IM, Gregory RP, Gill SS. Bilateral posteroventral pallidotomy for severe antipsychotic induced tardive dyskinesia and dystonia. J Neurol Neurosurg Psychiatry 1997;63:554-6. 84. Hillier CE, Wiles CM, Simpson BA.Thalamotomy for severe antipsychotic induced tardive dyskinesia and dystonia. J Neurol Neurosurg Psychiatry 1999;66:250-1. 85. Trottenberg T, Volkmann J, Deuschl G, Kuhn AA, Schneider GH, Muller J, et al. Treatment of severe tardive dystonia with pallidal deep brain stimulation. Neurology 2005;64:344-6. 86. Sako W, Goto S, Shimazu H, Murase N, Matsuzaki K, Tamura T, et al. Bilateral deep brain stimulation of the globus pallidus internus in tardive dystonia. Mov Disord 2008;23:1929-31. 87. Kosel M, Sturm V, Frick C, Lenartz D, Zeidler G, Brodesser D, et al. Mood improvement after deep brain stimulation of the internal globus pallidus for tardive dyskinesia in a patient suffering from major depression. J Psychiatr Res 2007;41:801-3. 88. Damier P, Thobois S, Witjas T, Cuny E, Derost P, Raoul S, et al. French Stimulation for Tardive Dyskinesia (STARDYS) Study Group. Bilateral deep brain stimulation of the globus pallidus to treat tardive dyskinesia. Arch Gen Psychiatry 2007;64:170-6. 89. Thobois S, Ballanger B, Xie-Brustolin J, Damier P, Durif F, Azulay JP, et al. Globus pallidus stimulation reduces frontal hyperactivity in tardive dystonia. J Cereb Blood Flow Metab 2008;28:1127-38. 90. Capelle HH, Blahak C, Schrader C, Kinfe T, Baezner H, Dengler R, et al. Deep brain stimulation for tardive dystonia in patients without a history of psychosis. Mov Disord 2008; Suppl 1:238 (Abstract). 91. Laitinen L, Johansson GG. Stereotaxic treatment of hyperkinesia. Nord Med 1966;75:676-9. 92. Mark Vh, Miyazaki Y, Siegfried J, Ervin Jr, Chato Jc, Eastman Fg. Production of Reversible And Permanent Intracerebral Lesions By Means Of Stereotaxically Localized Cooling. Application to Surgery Of Choreo-Athetosis. Confin Neurol 1963;23:357-8. 93. Cooper IS. An investigation of neurosurgical alleviation of parkinsonism, chorea, athetosis and dystonia. Ann Intern Med 1956;45:381-92.
Functional stereotactic procedures for treatment of secondary dystonia
94. Broggi G, Angelini L, Bono R, Giorgi C, Nardocci N, Franzini A. Long term results of stereotactic thalamotomy for cerebral palsy. Neurosurgery 1983;12:195-202. 95. Krayenbu¨hl H, Siegfried J. Stereotaxic surgery of the dentate nucleus in treatment of hyperkinesia and spastic conditions. Neurochirurgie 1969;15:51-8. 96. Galanda M, Na´dvornı´k P, Sramka M. Combined transtentorial dentatotomy with pulvinarotomy in cerebral palsy. Acta Neurochir 1977; Suppl 24:21-6. 97. Galanda M, Hovath S. Different effect of chronic electrical stimulation of the region of the superior cerebellar peduncle and the nucleus ventralis intermedius of the thalamus in the treatment of movement disorders. Stereotact Funct Neurosurg 1997;69:116-20. 98. Galanda M, Horvath S. Stereotactic stimulation of the anterior lobe of the cerebellum in cerebral palsy from a suboccipital approach. Acta Neurochir Suppl 2007;97:239-43. 99. Teo C. Functional stereotactic surgery of movement disorders in cerebral palsy. In: Krauss JK, Jankovic J, Grossman RG, editors. Surgery for parkinson’s disease and movement disorders. Philadelphia: Lippincott, Williams & Wilkins; 2001. p. 410-415. 100. Imer M, Ozeren B, Karadereler S, Yapici Z, Omay B, Hanag˘asi H, et al. Destructive stereotactic surgery for treatment of dystonia. Surg Neurol 2005;64 Suppl 2: S89-94. 101. Angelini L, Nardocci N, Estienne M, Conti C, Dones I, Broggi G. Life-threatening dystonia-dyskinesias in a child: successful treatment with bilateral pallidal stimulation. Mov Disord 2000;15:1010-2. 102. Gill S, Curran A, Tripp J, Melarickas L, Hurran C, Stanley O. Hyperkinetic movement disorder in an 11year-old child treated with bilateral stimulators. Develop Med Child Neurol 2001;43:350-3. 103. Cif L, El Fertit H, Vayssiere N, Hemm S, Hardouin E, Gannau A, et al. Treatment of dystonic syndromes by chronic electrical stimulation of the internal globus pallidus. J Neurosurg Sci 2003;47:52-5. 104. Krauss JK, Loher TJ, Weigel R, Capelle HH, Weber S, Burgunder JM. Chronic stimulation of the globus pallidus internus for treatment of non-DYT1 generalized dystonia and choreoathetosis: 2-year follow up. J Neurosurg 2003;98:785-92. 105. Vidailhet M, Lagrannge C, Welter ML, Krack P, Krystowiak P, Burbaud P, et al. Bilateral pallidal stimualion in generalized dystonia related to post-anoxic birth injury: multicentre study. Mov Disord. 2008; Suppl 1:162. 106. Manji H, Howard RS, Miller DH, Hirsch NP, Carr L, Bhatia K, et al. Status dystonicus: the syndrome and its management. Brain 1998;121:243-52. 107. Mariotti P, Fasano A, Contarino MF, Della Marca G, Piastra M, Genovese O, et al. Management of status dystonicus: our experience and review of the literature. Mov Disord 2007;22:963-8.
109
108. Schrock LE, Starr PA, Ostrem JL. Bilateral pallidal deep brain stimulation surgery for status dystonicus: a report of 2 cases. Mov Disord 2008; Suppl 1:170 (Abstract). 109. Krauss JK, Jankovic. Head injury and posttraumatic movement disorders. Neurosurgery 2002;50:927-40. 110. Alkhani A, Bohlega S. Unilateral pallidotomy for hemidystonia. Mov Disord 2006;21:852-5. 111. Sellal F, Hirsch E, Barth P, Blond S, Marescaux C. A case of symptomatic hemidystonia improved by ventroposterolateral thalamic electrostimulation. Mov Disord 1993;8:515-8. 112. Loher TJ, Hasdemir MG, Burgunder JM, Krauss JK. Long-term follow-up study of chronic globus pallidus internus stimulation for posttraumatic hemidystonia. J Neurosurg 2000;92:457-60. 113. Loher TJ, Krauss JK, Burgunder JM, Taub E, Siegfried J. Chronic stimulation of the ventrointermediate thalamus is effective for treatment of peripherally-induced dystonic paroxysmal nonkinesigenic dyskinesia. Neurology 2001;56:268-70. 114. Loher TJ, Capelle HH, Kaelin-Lang A, Weber S, Weigel R, Burgunder JM, et al. Deep brain stimulation for dystonia: outcome at long-term follow-up (3 years or longer). J Neurol 2008;255:881-4. 115. Capelle HH, Grips E, Weigel R, Blahak C, Hansjorg B, Wohrle JC, et al. Posttraumatic peripherally-induced dystonia and multifocal deep brain stimulation: case report. Neurosurgery 2006;59:E702. 116. Pittock SJ, Joyce C, O’Keane V, Hugle B, Hardiman MO, Brett F, et al. Rapid-onset dystonia-parkinsonism: a clinical and genetic analysis of a new kindred. Neurology 2000;55:991-5. 117. Deutschla¨nder A, Asmus F, Gasser T, Steude U, Bo¨tzel K. Sporadic rapid-onset dystonia-parkinsonism syndrome: failure of bilateral pallidal stimulation. Mov Disord 2005;20:254-7. 118. Kamm C, Fogel W, Wa¨chter T, Schweitzer K, Berg D, Kruger R, et al. Novel ATP1A3 mutation in a sporadic RDP patient with minimal benefit from deep brain stimulation. Neurology 2008;70:1501-3. 119. Capelle HH, Blahak C, Baezner H, Fogel W, Woehrle JC, Krauss JK. Bifocal pallidal and subthalamic deep brain stimulation in sporadic dystonia-parkinsonism. Acta Neurochir 2006;10:148 (Abstract). 120. Trottenberg T, Meissner W, Kabus C, Arnold G, Funk T, Einhaupl KM, et al. Neurostimulation of the ventral intermediate thalamic nucleus in inherited myoclonusdystonia syndrome. Mov Disord 2001;16:769-71. 121. Cif L, Valente EM, Hemm S, Coubes C, Vayssiere N, Serrat S, et al. Deep brain stimulation in myoclonusdystonia syndrome. Mov Disord 2004;19:724-7. 122. Magarin˜os-Ascone CM, Regidor I, Martı´nezCastrillo JC, Go´mez-Gala´n M, Figueiras-Me´ndez R. Pallidal stimulation relieves myoclonus-dystonia syndrome. J Neurol Neurosurg Psychiatry 2005;76:989-91.
1855
1856
109
Functional stereotactic procedures for treatment of secondary dystonia
123. Foncke EM, Bour LJ, Speelman JD, Koelman JH, Tijssen MA. Local field potentials and oscillatory activity of the internal globus pallidus in myoclonus-dystonia. Mov Disord 2007;22:369-76. 124. Justesen CR, Penn RD, Kroin JS, Egel RT. Stereotactic pallidotomy in a child with Hallervorden-Spatz disease: case report. J Neurosurg 1999;90:551-4. 125. Umemura A, Jaggi JL, Dolinskas CA, Stern MB, Baltuch GH. Pallidal deep brain stimulation for longstanding severe generalized dystonia in’HallervordenSpatz syndrome: case report. J Neurosurg 2004;100: 706-9. 126. Mikati MA, Yehyia A, Darwish H, Karam P, Comair Y. Deep brain stimulation as a mode of treatment of early onset pantothenate kinase-associated neurodegeneration. Eur J Paediatr Neurol 2009;13:61–4. 127. Duker AP, Gilbert DL, Mandybur GT, Espay AJ, Gartner M, Revilla FJ. Transient anti-dystonic benefit after pallidal deep brain stimualtion in a six year old girl with pantothenate kinase-associated neurodegeneration. Mov Disord 2008; Suppl 1:385 (Abstract). 128. Castelnau P, Cif L, Valente EM, Vayssiere N, Hemm S, Gannau A, et al. Pallidal stimulation improves pantothenate kinase-associated neurodegeneration. Ann Neurol 2005;57:738-41. 129. Vidailhet M, Pollak P. Deep brain stimulation for dystonia: make the lame walk. Ann Neurol 2005;57:613-4. 130. Krause M, Fogel W, Tronnier V, Pohle S, Ho¨rtnagel K, Thyen U, et al. Long-term benefit to pallidal deep brain stimulation in a case of dystonia secondary to pantothenate kinase-associated neurodegeneration. Mov Disord 2006;21:2255-7. 131. Evidente VG, Lyons MK, Wheeler M, Hillman R, Helepolelei L, Beynen F, et al. First case of X-linked dystoniaparkinsonism (‘‘Lubag’’) to demonstrate a response to bilateral pallidal stimulation. Mov Disord 2007;22:1790-3.
132. Cif L, Biolsi B, Gavarini S, Saux A, Robles SG, Tancu C, et al. Antero-ventral internal pallidum stimulation improves behavioral disorders in Lesch-Nyhan disease. Mov Disord 2007;22:2126-9. 133. Taira T, Kobayashi T, Hori T. Disappearance of selfmutilating behavior in a patient with lesch-nyhan syndrome after bilateral chronic stimulation of the globus pallidus internus: case report. J Neurosurg 2003;98:414-6. 134. Cubo E, Shannon KM, Penn RD, Kroin JS. Internal globus pallidotomy in dystonia secondary to Huntington’s disease. Mov Disord 2000;15:1248-51. 135. Moro E, Lang AE, Strafella AP, Poon YY, Arango PM, Dagher A, et al. Bilateral globus pallidus stimulation for Huntington’s disease. Ann Neurol 2004;56:290-4. 136. Hebb MO, Garcia R, Gaudet P, Mendez IM. Bilateral stimulation of the globus pallidus internus to treat choreathetosis in Huntington’s disease: technical case report. Neurosurgery 2006;58:E383. 137. Sramka M, Rattaj M, Molina H, Vojtassa´k J, Belan V, Ruzicky´ E. Stereotactic technique and pathophysiological mechanisms of neurotransplantation in Huntington’s chorea. Stereotact Funct Neurosurg 1992;58:79-83. 138. Hauser RA, Sandberg PR, Freeman TB, Stoessl AJ. Bilateral human fetal striatal transplantation in Huntington’s disease. Neurology 2002;58:1704. 139. Loher TJ, Burgunder JM, Weber S, Sommerhalder R, Krauss JK. Effect of chronic pallidal deep brain stimulation on off period dystonia and sensory symptoms in advanced parkinson’s disease. J Neurol Neurosurg Psychiatry 2002;73:395-9. 140. Tolosa E, Compta Y. Dystonia in parkinson’s disease. J Neurol 2006;253 Suppl 7:VII7-13. 141. Krauss JK, Pohle T, Weigel R, Burgunder JM. Deep brain stimulation of the centre median-parafascicular complex in patients with movement disorders. J Neurol Neurosurg Psychiatry 2002;72:546-8.
102 Gene Transfer for Parkinson’s Disease P. A. Starr . K. S. Bankiewicz
Introduction In the therapy of Parkinson’s disease (PD), gene transfer is an investigational technique whose goal is to increase production of a protein or proteins in brain circuits affected by the disease. Three broad strategies are under exploration: (1) Transfer of genes encoding growth factors for dopaminergic cells into the striatum. (2) Transfer of genes encoding enzymes in the dopamine synthetic pathway into the striatum. (3) Transfer of genes that increase GABAergic transmission in the subthalamic nucleus. Current protocols focus on the use of genetically modified neurotropic viruses as the most efficient vectors for gene transfer. In devising a gene transfer approach for PD, the major considerations are: choice of vector, choice of transgene, brain region targeted, means of delivery to the brain, and choice of a noninvasive method to assess gene expression. These considerations, and review of existing clinical trials, are described in this chapter.
Viral Vectors In nature, viruses are extremely efficient at taking over the machinery of host cells to express foreign genes. Cell infection is often receptor mediated and selective for cell type [1]. Viral vectors developed for gene transfer consist of viral genomes from which genes necessary for replication have been removed, and genes expressing a potentially therapeutic protein (the transgene) have been inserted. The vector genome is packaged into a viral ‘‘capsid,’’ or coat, that is similar to the capsid #
Springer-Verlag Berlin/Heidelberg 2009
of the wild type virus. In the laboratory or pharmaceutical manufacturing setting, packaging of the vector is performed in cell culture, using cells genetically modified to express the critical genes for replication and capsid assembly in ‘‘trans’’ to the vector genome. ‘‘In vivo’’ gene transfer refers to delivery of viral vectors directly into the target tissue. ‘‘Ex vivo’’ gene transfer refers to transfection of cells in culture that are then transplanted into the CNS. Both may be accomplished with viral vectors for gene transfer, or other methods such as liposome mediated transfer [2]. The current clinical trials in gene transfer for PD utilize in vivo gene transfer mediated by viral vectors. > Table 102-1 lists the properties of the major viral vectors that have been investigated in PD models, or used in human trials. In all cases, transgene expression is regulated by a promoter, which does not have to be a promotor derived from the original viral genome. The promoters for inserted genes, in those vectors currently in clinical trials, are constitutively active. Therefore, they cannot be deliberately inactivated. The current generation of promoters are also not tissue-specific, thus any tissue specificity of transgene expression is conferred by the cell specificity of the initial receptor mediated cell infection event. The vectors listed in > Table 102-1 differ in several important respects: whether or not they integrate into the host genome, how large a transgene (or genes) that may be accommodated, cell specificity of the vector infection, and durability of transgene expression. Vectors that integrate into the genome, such as those derived from retroviruses, carry a risk of insertional mutagenesis
1720
102
Gene transfer for parkinson’s disease
. Table 102-1 1 Overview of viral vectors in clinical or preclinical studies of gene transfer for Parkinson’s disease. Only AAV2 vectors have been used in humans, but all listed vectors have been studied in animal models of PD Vector type
Maximum insert size
Cell specificity
Chromosomal integration?
AAV-2
4.7 KB
Neurons
Rarely
Adenovirus
37 KB
No
Herpes virus (amplicon) Lentivirus
20 KB
Neurons, glia Neurons, glia Neurons
9 KB
No Yes
. Figure 102-1 Schematic diagram of an AAV2 vector genome. Lltr, left long terminal repeat; Rltr, right long terminal repeat; CAG, CMV enhancer/chicken b-actin (CAG) promoter element; PolyA, noncoding region
which may lead to malignancy. This has been reported in investigational protocols of ex-vivo gene transfer in some non-neurological conditions, including severe combined immunodeficiency (adenosine deaminase deficiency) [12]. A diagram of the vector genome currently used for human trials in PD, based on AAV-2, is shown in > Figure 102-1. expression from AAV2 vectors peaks one month following transfection, but in animal studies appears to be durable up to the longest time points studied, 6 years [3].
Development of Inducible Promoters Inducible promoters for transgenes in viral vectors are under laboratory study [13]. Perhaps the most investigated system has been the tetracycline transactivator (tet) system originally pioneered by Gossen and Bujard [14]. The tet system has been incorporated into AAV vectors
Gene expression
Reference
High expression, long duration High expression, short duration Low expression, short duration High expression, long duration
[3–7] [8] [9] [10,11]
and drives very high level transgene expression in a tetracycline-dependent manner, and this has been demonstrated in a variety of tissues including brain [15–18]. The slope of the doseresponse curve using tetracycline, doxycyline, or their analogs is very steep, thus far precluding the possibility of obtaining intermediate levels of expression by sub-maximal concentrations of agonist. Despite recent improvements that reduced the immunogenic potential of the transactivator protein [19], concerns still linger about constitutive expression of a foreign protein that contains bacterial sequences [20]. An attractive alternative approach to inducible gene expression is the rapamycin-dependent expression system. In this strategy, the active transcription factor can only be formed by dimerization of two inactive protein subunits, FKBP (immunophilin) and FRAP (lipid kinase homolog, mTOR). Dimerization is dependent on the presence of the clinically used immunosuppressant, rapamycin. The components of the expression system are human proteins. The system has been tested in a dual-component AAV2 vector system in rat brain [21], showing induction of expression of hAADC in the striatum of unilaterally 6-OHDA-lesioned rats by injection of rapamycin intraperitoneally. Induction of AADC in the lesioned striatum was associated with the development of L-dopa-induced turning
Gene transfer for parkinson’s disease
behavior. This effect was completely reversible through several on-off cycles. A disadvantage of this system is the need for two types of genomes to be packaged and for each transduced cell to be co-infected with both vector genomes, since the protein transcription factors cannot all be encoded on one vector genome. The need for chronic use of an immunosuppressant drug is another disadvantage. Safety concerns with constitutively active promoters, however, are sufficiently serious that future clinical trials of gene transfer into the nervous system will likely begin to utilize inducible promoters.
the method of choice for human trials in the future, at this time clinical trials of gene transfer for PD have utilized traditional frame-based stereotaxy without real time MRI monitoring. Surgical planning software is an important adjunct to stereotactic gene transfer protocols. Often, multiple brain penetrations must be planned through a limited cranial entry zone. Transgression of the ventricles or deep sulci is to be avoided to minimize the risk of reflux of the injected vector into CSF spaces with potential spread to untargeted regions. Commercially available planning software typically provides the following functions:
Stereotactic Surgical Approaches Viral vectors do not cross the blood-brain barrier. Thus, systemic delivery, even if brain specific expression could be achieved, is not straightforward. Strategies that involve intravascular delivery in combination with temporary disruption of the blood-brain barrier have been studied in the laboratory but not in clinical protocols [22]. Vector delivery in clinical trials has been achieved by stereotactic injection into the brain. Three major stereotactic approaches are currently in use to delivery devices, drugs or biologics to deep brain targets: Frame based stereotaxy using brain images acquired prior to surgery [23], ‘‘frameless’’ neuronavigation-guided stereotaxy using images acquired prior to surgery [24], or ‘‘frameless’’ stereotaxy guided by real time or near real time MR images (interventional MRI or iMRI). The latter strategy has several theoretical advantages: confirmation of correct cannula position prior to infusion, and visualization of the spread of the injected agent in real time. In humans, methodology for interventional MRI guided access to deep subcortical targets has been developed for brain biopsy [25] and for implantation of deep brain stimulator devices [26]. Work on deep brain drug delivery using iMRI is now emerging [27]. Although this seems likely to be
102
Computational fusion of multiple imaging modalities (MR, CT, and PET) Semi-automated registration of headframe fiducial markers Re-formatting of images into standard anatomical planes (which facilitates consistent target selection across subjects) Superposition of a linearly deformable brain atlas onto the patient MRI Independent selection of target and brain entry point Visualization of probe trajectory from entry to target, in oblique or ‘‘trajectory’’ views that are in-plane with the actual trajectory
> Figure 102-2 illustrates a typical target plan for intrastriatal delivery of a vector construct at eight sites, along four stereotactic trajectories, per striatum. > Figure 102-3 illustrates a trajectory plan. The images are taken from two subjects in a Phase I study of intraputaminal injection of AAV2-Neurturin (Cere-120) for Parkinson’s disease, described in greater detail below. The targets are planned on an MRI sequence optimized for visualization of the lentiform nuclei, an inversion recovery-fast spin echo sequence [28]. The trajectories (> Figure 102-3) have been planned so as to avoid crossing ependymal surfaces, sulci, or venous structures as visualized on gadolinium-enhanced T1 images.
1721
1722
102
Gene transfer for parkinson’s disease
. Figure 102-2 Surgical planning for intraputaminal gene transfer: anatomic views. Four putaminal injection tracks and targets are represented in (a) parasagital (b) axial anatomic planes. Software illustrated is Framelink 4.1, Medtronic SofamorDanek. (a) shows the computational fusion of two image sets, a volumetric gradient echo sequence covering the whole brain, with a 2D fast spin echo-inversion recovery sequence covering only the basal ganglia. The images are strictly anatomically accurate only for the third target (purple line in (a), red square in (b)), whereas the other three targets (which are not coplanar) are projected onto the anatomic planes defined by the third target point
Infusion Methods: Diffusion Versus Convection-enhanced Delivery A major challenge for stereotactic delivery of vectors to the brain relates to the size of the potential targets. For putaminal delivery, for example, the volume of the human putamen in PD patients ranges from 3–5 cm3 (depending on age and sex) compared to 0.7 cm3 for the adult rhesus macaque. In rats, the caudate-putamen complex has an approximate volume of 0.05 cm3. When scaling up from successful experiments in rodent models to nonhuman primate models and ultimately to the human, there is an enormous increase in volume of tissue that must be affected. During and following injection of a vector solution through a needle, spread of the agent through the brain tissue may be mediated by diffusion, or by convection (bulk flow). Flow rate, total delivery volume, and geometry of the cannula tip determine whether diffusion or convection predominates [29]. Convective delivery
results in a larger volume of distribution (total volume of brain tissue exposed to vector) for a given injection volume, and a more homogeneous distribution, which may be highly advantageous for delivery to large brain structures such as the striatum. Convection-enhanced delivery also overcomes a major limitation with diffusion-based delivery related to the low diffusivity of vector capsids. When the vector infusate is doped with an MR-visible contrast agent of similar size as the vector particles, such as ferumoxtran-10 [30] or gadolinium-doped liposomes [31], convective delivery may be visualized using interventional MRI. Laboratory animal investigation of this technique is illustrated in > Figure 102-4. Realtime imaging of convection-enhanced delivery of proteins to the brainstem has recently been reported in humans [27]. In grey matter, the ratio of the volume of distribution of the virus particles to the volume of infusate, using convective delivery, was found to be 4.1. This is a promising technique for future clinical trials but has not yet been used for gene transfer in PD.
Gene transfer for parkinson’s disease
102
. Figure 102-3 Surgical planning for intraputaminal gene transfer: trajectory views. Four putaminal injection tracks are shown (different brain from that shown in Figure 2). (a) and (b), ‘‘Trajectory’’ views showing oblique images that are inplane for the most anterior of the four trajectories. (c) ‘‘Probe’s eye view’’ which is perpendicular to the most anterior trajectory, at the brain surface. The trajectories are planned to avoid the cortical veins on the gadoliniumenhanced image. The large ventricles of the brain shown in this case allowed only a narrow corridor between the lateral ventricle and the insula, underscoring the importance of software-based trajectory planning. (d) shows a 3D reconstruction of the head with all trajectories passing though a single point in the burr hole, at different angles. Software illustrated is Framelink 4.1, Medtronic Sofamor-Danek
Design of the Injection Cannula Design of the injection cannula is extremely important to ensure accurate delivery of the infusate to the regions surrounding the cannula outlet. The disappointing failure of a major phase
II trial of intraputaminal GDNF protein infusion for PD may have been related to cannula design [32]. For convection-enhanced delivery from a cannula tip, a cannula whose external surface is smooth along its intraparenchymal course will allow reflux of the vector up the outside of
1723
1724
102
Gene transfer for parkinson’s disease
. Figure 102-4 Laboratory study of convection enhanced delivery of ferumtroxan-10, a high molecular weight MR contrast agent, in the nonhuman primate striatum as visualized by real-time MR imaging. The injection on the left was performed at 1.5 mL/min, that on the right, at 1.75 mL/ min, resulting in some reflux up the injection canula (arrows) (reprinted with permission from [30])
therapy (motor fluctuations and on-period dyskinesias) despite optimal medical management. Patients with dementia, poor motor improvement with levodopa, or major medical comorbidity were excluded. This is the same population of patients who are candidates for subthalamic or globus pallidus deep brain stimulation.
Growth Factor Delivery
the cannula from the injection site. A step-off in the cannula diameter proximal to the injection site is needed to prevent reflux of the solution up the needle track. Multiple step-offs are not necessary and may increase tissue trauma [33]. After injection, a waiting period of several minutes is required prior to withdrawal of the cannula. Flow and reflux characteristics associated with specific cannula designs and infusion protocols have been characterized experimentally [33,34].
Clinical Protocols and Their Experimental basis The major investigational trials of gene transfer in PD (as of late 2007) are summarized in > Table 102-2. Inclusion criteria for all three trials were similar: Patients with moderately advanced idiopathic Parkinson’s disease (Hoehn and Yahr stage 3 or worse when off medication) who have developed complications of medical
Since the discovery of glial cell derived neurotrophic factor (GDNF) in the early 1990s [35], there has been enormous interest in use of dopaminergic cell growth factors as both symptomatic and potentially neuroprotective agents [36]. A number of clinical trials have used direct protein infusion into the striatum, but the only placebocontrolled trial of instrastrial GDNF protein infusion did now show a significant difference between the placebo and experimental groups [32]. In concurrent studies in nonhuman primates, the study sponsor also raised concern over toxicity (cerebellar degeneration) in some nonhuman primates treated with very high doses [32]. Subsequently, gene transfer has attracted interest as a strategy for growth factor delivery that could be more focal (avoiding delivery to unwanted brain areas), and less cumbersome than direct protein infusions that require permanent indwelling pumps. Neurturin is a naturally occurring structural and functional analog of GDNF with similar neurotrophic properties. In preclinical study in parkinsonian nonhuman primates, Kordower et al. [4] studied intrastriatal delivery of AAV-neurturin in the MPTP-treated nonhuman primate model of PD. AAV-Neurturin was delivered to the striatum and substantia nigra, pars compacta four days following MPTP treatment. In comparison with a control group that did not receive AAVNeurturin, the experimental group showed increased survival of nigral dopaminergic neurons, and improvements in motor function that
Gene transfer for parkinson’s disease
102
. Table 102-2 Clinical trials of gene transfer for PD Functional imaging technique
Vector
No. of patients
Brain target
Delivery method
AAV-Neurturin
12 (phase I) 58 (phase II)
Eight stereotactic injections along four needle tracks per side, bilateral
18-F-fluorodopa-PET
AAV-AADC
8 (phase I)
Putamen, anterior and posterior, bilateral Putamen, posterior
18-F-fluoromethyltyrosine PET
AAV-GAD65
12 (phase I)
Two stereotactic injections along two needle tracks per side, convection enhanced, bilateral One stereotactic injection, unilateral
Subthalamic nucleus
18-F-fluorodeoxyglucosePET
Note: PET, positron emission tomography
reached their plateau approximately four months following vector delivery. A Phase I trial of bilateral intrastriatal AAVneurturin has been completed [37]. Twelve subjects received bilateral stereotactic intrastriatal injection of AAV-neurturin at two dose cohorts (low dose and high dose, fourfold difference). The vector was injected at eight sites along four injection tracks spaced equidistantly throughout the anterior and posterior putamen (illustrated in > Figure 102-2). Injection parameters were such that the infusate spread by diffusion, not convection. Unblinded evaluations by movement disorders neurologists were performed at baseline, and at 1, 3, 6, 9 and 12 months posttreatment, on and off medication. The primary endpoint was the change in UPDRS-III off medication at 12 months, compared to baseline. 18-Ffluorodopa was performed at baseline, and at 6 and 12 months post treatment. Since uptake of the ligand 18-F-fluorodopa is mediated by the same receptor system used for dopamine reuptake in the synaptic cleft, the degree of ligand uptake should reflect dopaminergic neurotransmission in nigrostriatal terminals. The mean improvement in the UPDRS-III off was 36% in both low-dose and high does groups (p < 0.001 at 12 months, compared to baseline). The PET studies, however, did not show a significant change in
18-fluorodopa uptake in the striatum. There were no major adverse events. A phase II (placebo controlled) trial is underway. Given its relatively short (1 year) duration of follow-up, the AAV-neurturin phase I trial was designed primarily to detect a symptomatic effect, rather than a neuroprotective one. The symptomatic effect, if demonstrated in larger trials, would presumable be due to the enhancement of dopamine release from existing nigrostriatal neurons, rather than generation of new neurons. Future growth factor trials will need to assess neuroprotection, that is, evidence of a change in the expected natural history of symptom progression over time. The presumed mechanism for this would be prevention of ongoing loss of existing dopaminergic neurons. Even if a neuroprotective benefit for dopaminergic cells is eventually demonstrated in humans, it is unclear if protection would be conferred to nondopaminergic cell types, the loss of which mediates an increasing proportion of patient disability as the disease progresses.
Enzyme Replacement Development of motor fluctuations and onperiod dyskinesias in response to oral levodopa is one of the major shortcomings of medical
1725
1726
102
Gene transfer for parkinson’s disease
therapy of PD. Although the pathophysiology of these phenomena is incompletely understood, they may in part be related to deficiency of the enzyme aromatic amino acid decarboxylase (AADC). Normally, this enzyme is active in striatal dopaminergic nerve terminals, converting endogenous levodopa to dopamine for release at synapses with striatal medium spiny (projection) neurons. In the parkinsonian state, lack of dopaminergic nerve terminals, and resulting lack of AADC, precludes efficient conversion of orally administered levodopa in the striatum. In the MPTP-treated nonhuman primate, intrastriatal injection of AAV-AADC reduced motor complications of orally administered levodopa [3,5]. 18-F-fluoromethyltyrosine PET showed an increase in activity of AADC in the striatal region targeted. With this PET technique, the signal detected in the striatum should reflect local activity of AADC, since any 18-F-fluoromethyltyrosine that is not converted to 18-F-fluoromethyldopa by AADC would be rapidly cleared and thus not detected. 18-F-fluoromethyldopa, conversely, will be taken up by dopaminergic nerve terminals and its signal detected. A phase 1 trial of intrastriatal injection of AAV-AADC in humans with PD is underway [38]. There are two dose cohorts. Like the AAVNeurturin approach, the target is the striatum; but the delivery methods in these two phase I trials differ in important respects. In the AAVAADC trial, delivery is restricted to the postcommissural putamen which, in classical ‘‘segregated circuit’’ models of the basal ganglia, is the somatic motor controlling territory of the striatum. Delivery is accomplished by convection, unlike the AAV-neurturin trial, allowing a larger volume of distribution of the vector with only two stereotactic injections per hemisphere. Early results show clear changes in AADC activity, assessed by 18-F-fluoromethyltyrosine PET, restricted to the regions where postoperative MRI showed T2 changes associated with spread of the vector infusate [38].
Unlike the other two gene therapy approaches discussed in this chapter, there is a safety mechanism built into the AAV2-AADC approach. Since production of the transgene product (dopamine) is dependent on the administration of the precursor drug (L-dopa), excessive AADC gene expression can be modified by limiting administration of L-dopa.
Enhancement of Inhibitory Neurotransmission In Parkinson’s disease, the subthalamic nucleus has excessive and abnormally patterned activity, the interruption of which improves parkinsonian motor signs [39]. This is the theoretical basis for both STN ablation and STN deep brain stimulation in the therapy of PD. Alternatively, pharmacologic or genetic strategies could be used to alter abnormal STN activity. Acutely, STN injection of muscimol, an agonist of the inhibitory neurotransmitter GABA, in humans with PD improves motor deficits [40]. Glutamic acid decarboxylase (GAD) is the rate limiting enzyme for GABA synthesis and occurs in two isoforms, GAD65 and GAD67. In a rat model of PD, injection of AAV-GAD into the STN resulted in increased GABA synthesis and improvement in motor signs [6]. Behavioral improvement was also seen in the MPTP-treated nonhuman primate model of PD [41]. Based on this work, a phase I trial of unilateral STN injection of AAV-GAD for PD was recently completed [7]. There were two dose cohorts with six patients in each. At 12 months, there was an improvement in the UPDRS-III off-medication scores for the contralateral limbs of 24 and 27% in the low and high dose cohorts, respectively. Improvements were first noted at 3 months, and did not decline over the first year, suggesting that a tissue trauma or ‘‘microlesioning’’ effect on the STN injection was not responsible for the improvements seen. PET imaging
Gene transfer for parkinson’s disease
with 18-F-fluorodeoxyglucose showed a reduction in thalamic metabolism restricted to the treated hemisphere, and correlation between clinical motor scores and brain metabolism in the supplementary motor area. This imaging technique does not directly detect activity of GAD, but detects global changes in brain metabolic activity that may have been induced by an alteration in STN physiology.
Risks Associated with Gene Transfer for PD Risks of gene transfer may be divided into two categories: those associated with the surgical procedure and those associated with the vector agent and expression of the transgene. The major risk of stereotactic injection is hemorrhagic stroke. As with any stereotactic procedure, this risk can be reduced but not eliminated by careful trajectory planning and attention to the patient’s blood pressure and coagulation status. For microelectrode guided deep brain stimulation in Parkinson’s disease, the risk of symptomatic hemorrhage in our group’s series is 2% on a per patient basis, with bilateral surgery [42]. Increasing the number of brain penetration is likely to increase the risk of stroke [42,43]. Thus, gene transfer strategies should emphasize delivery methods that allow maximum spread of the infusate with the fewest number of brain penetrations. Risks associated with the vector itself include inflammatory reactions to the viral capsid or transgene, and potential effects associated with unregulated expression of the transgene. In a protocol for hemophilia involving gene transfer to the liver of AAV2 based vectors, immune response to AAV may have decreased transgene expression [44]. Generation of new serum antibodies to AAV2 has been seen in the current clinical trials of gene transfer for PD, but only in a minority of cases. In no case has a significant brain inflammatory response as assessed by serial MRI, been
102
reported. Unregulated expression of the transgene may be associated with unknown risks. In the case of delivery of AAV-GAD to the STN, ablation of the STN could be performed if an unwanted effect occurred from gene expression, as STN ablation is a therapeutic procedure for PD. In the case of delivery of AAV-AADC to the striatum, the transgene is presumably inactive without delivery of the ‘‘prodrug’’ levodopa. For growth factor delivery to the striatum, no simple ‘‘rescue strategy’’ is available if unwanted effects of transgene expression occur, since AAV-2 mediated gene expression appears to be permanent. No undesirable effects attributable to the growth factor were detected in the phase I trial, but the safety profile would be improved by the development of vectors with regulatable gene expression.
Summary Gene transfer is a novel pharmacologic strategy for protein delivery to the central nervous system, with promise for the treatment of Parkinson’s disease. Recently, three different strategies for gene transfer have been employed in clinical trials: Introduction of a dopaminergic growth factor in the striatum, increased expression of the levodopa converting enzyme aromatic amino acid decarboxylase in the striatum, and alteration of GABAergic transmission to reduce basal ganglia output. All current trials use adeno-associated virus serotype 2 as the host vector, and MRI-based stereotactic techniques. Methods for reliable, widespread, homogeneous delivery of viral vectors to deep brain targets are under development.
References 1. Mandel RJ, Burger C, Snyder RO. Viral vectors for in vivo gene transfer in Parkinson’s disease: properties and clinical grade production. Exp Neurol 2008;209(1):58-71. 2. Huwyler J, Wu D, Pardridge WM. Brain drug delivery of small molecules using immunoliposomes. Proc Natl Acad Sci USA 1996;93(24):14164-9.
1727
1728
102
Gene transfer for parkinson’s disease
3. Bankiewicz KS, Forsayeth J, Eberling JL, SanchezPernaute R, Pivirotto P, Bringas J, Herscovitch P, Carson RE, Eckelman W, Reutter B, Cunningham J. Long-term clinical improvement in MPTP-lesioned primates after gene therapy with AAV-hAADC. Mol Ther 2006; 14(4):564-70. 4. Kordower JH, Herzog CD, Dass B, Bakay RA, Stansell J III, Gasmi M, Bartus RT. Delivery of neurturin by AAV2 (CERE-120)-mediated gene transfer provides structural and functional neuroprotection and neurorestoration in MPTP-treated monkeys. Ann Neurol 2006;60 (6):706-15. 5. Bankiewicz KS, Eberling JL, Kohutnicka M, Jagust W, Pivirotto P, Bringas J, Cunningham J, Budinger TF, Harvey-White J. Convection-enhanced delivery of AAV vector in parkinsonian monkeys; in vivo detection of gene expression and restoration of dopaminergic function using pro-drug approach. Exp Neurol 2000; 164(1):2-14. 6. Luo J, Kaplitt MG, Fitzsimons HL, Zuzga DS, Liu Y, Oshinsky ML, During MJ. Subthalamic GAD gene therapy in a Parkinson’s disease rat model. Science 2002; 298(5592):425-9. 7. Kaplitt MG, Feigin A, Tang C, Fitzsimons HL, Mattis P, Lawlor PA, Bland RJ, Young D, Strybing K, Eidelberg D, During MJ. Safety and tolerability of gene therapy with an adeno-associated virus (AAV) borne GAD gene for Parkinson’s disease: an open label, phase I trial. Lancet 2007;369(9579):2097-105. 8. Choi-Lundberg DL, Lin Q, Chang Y, Chiang YL, Hay CM, Mohajeri H, Davidson BL, Bohn MC. Dopaminergic neurons protected from degeneration by GDNF gene therapy. Science 1997;275:838-41. 9. Sun M, Kong L, Wang X, Holmes C, Gao Q, Zhang GR, Pfeilschifter J, Goldstein DS, Geller AI. Coexpression of tyrosine hydroxylase, GTP cyclohydrolase I, aromatic amino acid decarboxylase, and vesicular monoamine transporter 2 from a helper virus-free herpes simplex virus type 1 vector supports high-level, long-term biochemical and behavioral correction of a rat model of Parkinson’s disease. Hum Gene Ther 2004;15(12): 1177-96. 10. Kordower JH, Emborg ME, Bloch J, Ma SY, Chu Y, Leventhal L, McBride J, Chen EY, Palfi S, Roitberg BZ, Brown WD, Holden JE, Pyzalski R, Taylor MD, Carvey P, Ling Z, Trono D, Hantraye P, Deglon N, Aebischer P. Neurodegeneration prevented by lentiviral vector delivery of GDNF in primate models of Parkinson’s disease. Science 2000;290(5492):767-73. 11. Azzouz M, Martin-Rendon E, Barber RD, Mitrophanous KA, Carter EE, Rohll JB, Kingsman SM, Kingsman AJ, Mazarakis ND. Multicistronic lentiviral vector-mediated striatal gene transfer of aromatic L-amino acid decarboxylase, tyrosine hydroxylase, and GTP cyclohydrolase I induces sustained transgene expression, dopamine production, and functional improvement
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
in a rat model of Parkinson’s disease. J Neurosci 2002; 22(23):10302-12. Nienhuis AW, Dunbar CE, Sorrentino BP. Genotoxicity of retroviral integration in hematopoietic cells. Mol Ther 2006;13(6):1031-49. Haberman RP, McCown TJ. Regulation of gene expression in adeno-associated virus vectors in the brain. Methods 2002;28(2):219-26. Gossen M, Bujard H. Tight control of gene expression in mammalian cells by tetracycline-responsive promoters. Proc Natl Acad Sci U S A 1992;89(12):5547-51. Bohl D, Salvetti A, Moullier P, Heard JM. Control of erythropoietin delivery by doxycycline in mice after intramuscular injection of adeno-associated vector. Blood 1998;92(5):1512-17. Chtarto A, Bender HU, Hanemann CO, Kemp T, Lehtonen E, Levivier M, Brotchi J, Velu T, Tenenbaum L. Tetracycline-inducible transgene expression mediated by a single AAV vector. Gene Ther 2003;10(1):84-94. Fitzsimons HL, McKenzie JM, During MJ. Insulators coupled to a minimal bidirectional tet cassette for tight regulation of rAAV-mediated gene transfer in the mammalian brain. Gene Ther 2001;8(22):1675-81. Folliot S, Briot D, Conrath H, Provost N, Cherel Y, Moullier P, Rolling F. Sustained tetracycline-regulated transgene expression in vivo in rat retinal ganglion cells using a single type 2 adeno-associated viral vector. J Gene Med 2003;5(6):493-501. Baron U, Gossen M, Bujard H. Tetracycline-controlled transcription in eukaryotes: novel transactivators with graded transactivation potential. Necleic Acids Res 1997;25(14):2723-9. Favre D, Blouin V, Provost N, Spisek R, Porrot F, Bohl D, Marme F, Cherel Y, Salvetti A, Hurtrel B, Heard JM, Riviere Y, Moullier P. Lack of an immune response against the tetracycline-dependent transactivator correlates with long-term doxycycline-regulated transgene expression in nonhuman primates after intramuscular injection of recombinant adeno-associated virus. J Virol 2002;76(22):11605-11. Sanftner LM, Rivera VM, Suzuki BM, Feng L, Berk L, Zhou S, Forsayeth JR, Clackson T, Cunningham J. Dimerizer regulation of AADC expression and behavioral response in AAV-transduced 6-OHDA lesioned rats. Mol Ther 2006;13(1):167-74. Doran SE, Ren XD, Betz AL, Pagel MA, Neuwelt EA, Roessler BJ, Davidson BL. Gene expression from recombinant viral vectors in the central nervous system after blood-brain barrier disruption. Neurosurgery 1995; 36(5):965-70. Starr PA. Placement of deep brain stimulators into the subthalamic nucleus or globus pallidus internus: technical approach. Stereotact Funct Neurosurg 2003; 79:118-45. Holloway K, Gaede S, Starr PA, Rosenow J, Ramakrishnan V, Henderson J. Frameless stereotaxy
Gene transfer for parkinson’s disease
25.
26.
27.
28.
29.
30.
31.
32.
33.
using bone fiducials for deep brain stimulation. J Neurosurg 2005;103:404-13. Hall WA, Liu H, Martin AJ, Maxwell RE, Truwit CL. Brain biopsy sampling by using prospective stereotaxis and a trajectory guide. J Neurosurg 2001;94:67-71. Martin A, Larson P, Ostrem J, Sootsman K, Weber O, Lindsey N, Myers J, Starr PA. Placement of deep brain stimulator electrodes using real-time high field interventional MRI. Magn Reson Med 2005;54:1107-14. Lonser RR, Warren KE, Butman JA, Quezado Z, Robison RA, Walbridge S, Schiffman R, Merrill M, Walker ML, Park DM, Croteau D, Brady RO, Oldfield EH. Real-time image-guided direct convective perfusion of intrinsic brainstem lesions. Technical note. J Neurosurg 2007;107(1):190-7. Reich CA, Hudgins PA, Sheppard S, Starr PA, Bakay RAE. A high resolution fast spin echo-inversion recovery sequence for preoperative localization of the internal globus pallidus. AJNR Am J Neuroradiol 2000;21:928-31. Chen MY, Lonser RR, Morrison PF, Governale LS, Oldfield EH. Variables affecting convection-enhanced delivery to the striatum: a systematic examination of rate of infusion, cannula size, infusate concentration, and tissue-cannula sealing time. J Neurosurg 1999;90 (2):315-20. Szerlip NJ, Walbridge S, Yang L, Morrison PF, Degen JW, Jarrell ST, Kouri J, Kerr PB, Kotin R, Oldfield EH, Lonser RR. Real-time imaging of convection-enhanced delivery of viruses and virussized particles. J Neurosurg 2007;107(3):560-7. Saito R, Krauze MT, Bringas JR, Noble C, McKnight TR, Jackson P, Wendland MF, Mamot C, Drummond DC, Kirpotin DB, Hong K, Berger MS, Park JW, Bankiewicz KS. Gadolinium-loaded liposomes allow for real-time magnetic resonance imaging of convectionenhanced delivery in the primate brain. Exp Neurol 2005;196(2):381-9. Lang AE, Gill S, Patel NK, Lozano A, Nutt JG, Penn R, Brooks DJ, Hotton G, Moro E, Heywood P, Brodsky MA, Burchiel K, Kelly P, Dalvi A, Scott B, Stacy M, Turner D, Wooten VG, Elias WJ, Laws ER, Dhawan V, Stoessl AJ, Matcham J, Coffey RJ, Traub M. Randomized controlled trial of intraputamenal glial cell line-derived neurotrophic factor infusion in Parkinson disease. Ann Neurol 2006;59(3):459-66. Sanftner LM, Sommer JM, Suzuki BM, Smith PH, Vijay S, Vargas JA, Forsayeth JR, Cunningham J, Bankiewicz KS, Kao H, Bernal J, Pierce GF, Johnson KW. AAV2-mediated gene delivery to monkey putamen: evaluation of an infusion device and delivery parameters. Exp Neurol 2005;194(2):476-83.
102
34. Krauze MT, Saito R, Noble C, Tamas M, Bringas J, Park JW, Berger MS, Bankiewicz K. Reflux-free cannula for convection-enhanced high-speed delivery of therapeutic agents. J Neurosurg 2005;103(5):923-9. 35. Lin LF, Doherty DH, Lile JD, Bextesh S, Collins F. GDNF: a glial cell line-derived neurotrophic factor for midbrain dopaminergic neurons. Science 1993;260: 1130-2. 36. Gash DM, Zhang Z, Ovadia A, Cass WA, Yi A, Simmerman L, Russell D, Martin D, Lapchak P, Collins F, Hoffer BJ, Gerhardt GA. Functional recovery in parkinsonian monkeys treated with GDNF. Nature 1996;380:252-5. 37. Marks WJ, Ostrem JL,Verhagen L, Starr PA, Larson PS, Bakay RA, Taylor R, Cahn-Weiner DA, Stoessl AJ, Olanow CW, Bartus RT. Safety and tolerability of intraputaminal delivery of CERE-120 (adeno-associated virus serotype 2-neurturin) to patients with idiopathic Parkinson’s disease: an open-label, phase I trial. Lancet Neurol 2008;7:400-8. 38. Eberling JL, Jagust WJ, Christine CW, Starr P, Larson P, Bankiewicz KS, Aminoff MJ. Results from a phase I safety trial of hAADC gene therapy for Parkinson disease. Neurology 2008;70(21):1980-3. 39. Bergman H, Wichmann T, DeLong MR. Reversal of experimental parkinsonism by lesions of the subthalamic nucleus. Science 1990;249:1436-8. 40. Levy R, Lang AE, Dostrovsky JO, Pahapill P, Romas J, Saint-Cyr J, Hutchison WD, Lozano AM. Lidocaine and muscimol microinjections in subthalamic nucleus reverse Parkinsonian symptoms. Brain 2001;124(Pt 10):2105-18. 41. Emborg ME, Carbon M, Holden JE, During MJ, Ma Y, Tang C, Moirano J, Fitzsimons H, Roitberg BZ, Tuccar E, Roberts A, Kaplitt MG, Eidelberg D. Subthalamic glutamic acid decarboxylase gene therapy: changes in motor function and cortical metabolism. J Cereb Blood Flow Metab 2007;27(3):501-9. 42. Binder D, Rau G, Starr PA. Risk factors for hemorrhage during microelectrode-guided deep brain stimulator implantation for movement disorders. Neurosurgery 2005;56:722-32. 43. Deep-brain stimulation of the subthalamic nucleus or the pars interna of the globus pallidus in Parkinson’s disease. N Engl J Med 2001;345:956–63. 44. Manno CS, Pierce GF, Arruda VR, Glader B, Ragni M, Rasko JJ, Ozelo MC, Hoots K, Blatt P, Konkle B, Dake M, Kaye R, Razavi M, Zajko A, Zehnder J, Rustagi PK, Nakai H, Chew A, Leonard D, Wright JF, Lessard RR, Sommer JM, Tigges M, Sabatino D, Luk A, Jiang H, Mingozzi F, Couto L, Ertl HC, High KA, Kay MA. Successful transduction of liver in hemophilia by AAV-Factor IX and limitations imposed by the host immune response. Nat Med 2006;12(3):342-7.
1729
95 Globus Pallidus Stimulation for Parkinson’s Disease M. Deogaonkar . J. L. Vitek
Introduction Deep brain stimulation (DBS) for Parkinson’s disease (PD) is routinely performed on patients with medically intractable PD. The targets for DBS in PD have included a number of nodal points in the basal ganglia thalamocortical circuit. These include the ventral intermediate nucleus (VIM) of the thalamus, the internal segment of the globus pallidus (GPi) and the subthalamic nucleus (STN) [1–42]. Given Vim DBS predominately improves tremor; GPi and STN have been the primary targets for the treatment of the motor symptoms associated with PD. Though GPi and STN DBS both improve PD symptoms (e.g., tremor, rigidity and bradykinesia), there is a continued debate over which site is more effective in improving motor symptoms, reducing PD medications and controlling medication associated side effects such as druginduced dyskinesia and motor fluctuations [11,17,20,27,31,35,43–51]. In this chapter, we discuss the historical evolution of GPi as a target, the role of GPi in basal ganglia thalamo-cortical circuitry and the pathophysiology of PD, indications and patient selection for GPi DBS, technical aspects of implanting and programming GPi patients including surgical technique and micro-electrode recordings (MER). We will also review the reported outcomes and complications following GPi DBS and the debate comparing GPi and STN as surgical targets for DBS in PD.
#
Springer-Verlag Berlin/Heidelberg 2009
Historical Evolution of GPi as a Target for Neuromodulation in PD The history of surgical therapy for PD has evolved based on necessity, available technology, our understanding of basal ganglia anatomy and physiology, and the physiological changes in basal ganglia circuitry that occur in the parkinsonian state. Initial approaches targeted a wide variety of structures within and outside the basal ganglia including such sites as the motor cortex [52], cerebral peduncles [53,54] and thalamic and pallidal areas [55]. Earliest ablative procedures by Meyers for the treatment of PD were targeted at the caudate nucleus [56–59]. Guiot then used pallidal coagulation to treat symptoms of PD and dyskinesia [60–63], making his lesions progressively larger and more posterior if initial ablations were not effective. Cooper, led by surreptitious discovery of alleviation of parkinsonian symptoms after ligation of the anterior choroidal artery, went on to perform ligations in a large series of PD patients [64–67]. His experience from ligation and chemo-pallidotomy procedures prompted him to propose the globus pallidus as a possible target for ablation for the treatment of PD [68,69]. This coupled with more accurate targeting using stereotactic technique developed by Spiegel and Wycis [70] and a later report of the effects of pallidotomy on the motor symptoms of PD by Leksell [71] resulted in establishing the globus pallidus as a safe and effective target for the treatment of PD and
1578
95
Globus pallidus stimulation for parkinson’s disease
ushered in an era of stereotactic pallidotomy for this disorder. The first stereotactic pallidotomy was performed by Speigel in 1951 [59] and was further refined in the following four decades [72–81]. Svennilson and Leksell [82] reported 81 patients operated between 1953–1957 and took a systematic approach in defining the optimal lesion site within the pallidum by systematically moving their lesion in different patients from anterior to posterior. They found that lesions posterior and medial in the pallidum, in this case the pallidum refers to GPe and GPi, which would have placed the lesion site in the posterior lateral ‘‘sensorimotor’’ portion of GPi, the vast majority of patients (19/20) had consistently good benefit with ‘‘95% complete relief of tremor and rigidity.’’ Interestingly, the pallidum as a target fell out of favor in the 1960–1980s due to the dramatic benefit of thalamotomy for parkinsonian tremor demonstrated by Hassler [83] together with the discovery of levodopa [59,84,85]. Levodpa however was not the panacea for the treatment of PD it was originally thought to be. It was soon discovered that over time patients on medical therapy developed an increasing incidence of drug-induced dyskinesia and motor fluctuations, symptoms that could be as compromising to the patient as the motor signs associated with PD. Patients with advanced PD were left with few options until the early 1990s when Lauri Laitinen [73,74,86] published a report on the effect of pallidotomy for PD. Laitinen et al. had re-explored Leksells posterior pallidotomy procedure and reported marked benefit in all the cardinal motor signs of PD following this procedure. At the same time advances in our understanding of the functional organization of the basal ganglia and pallido-thalamic circuitry resulted in a refinement of the target region within the pallidum to the postero-lateral ‘‘sensorimotor’’ territory of GPi, an area intimately associated with the development of parkinsonian motor signs [87–89]. Laitinens’ report together
with this rationale based approach for target selection and the development of micro-electrode mapping procedures led to a more accurate method for identifying the target and more consistent placement of lesions in the target area [81,90–92]. This brought about the resurgence of pallidotomy for PD in the early 1990s and led to hundreds of pallidotomy procedures being performed for patients suffering from advanced PD [93–95,75,81]. These procedures provided approximately 30% benefit in the UPDRS III off motor subscore and were very effective in alleviating contralateral tremor, dyskinesia, rigidity and bradykinesia. Benefit for gait and balance however was less consistent. Most approaches at the time were to initially place a unilateral lesion, and if necessary due to refractory gait and balance problems or persistent ipsilateral motor symptoms, to follow with a second lesion on the intact side. Bilateral pallidotomy however was increasingly associated with cognitive and speech disturbances and only a few centers routinely performed such procedures. Eventually bilateral pallidotomy fell out of favor at majority of sites performing these procedures. This led to the search for a therapy to treat the remaining axial and ipsilateral symptoms and the development of DBS [96–100]. The use of electrical stimulation in the basal ganglia for stereotactic surgical procedures was first carried out by Spiegel [101] and Hassler [83]. Though the first clinical use of chronic stimulation of basal ganglia nuclei was reported by Bechtereva [102] it was Benabid [99,103] and his colleagues who refined and popularized DBS for the treatment of movement disorders. GPi as a target for DBS has been commonly used for dystonia, and early trials reported improvement of all the cardinal motor signs of PD with GPi DBS. Multiple non-randomized open label trials however, reported greater improvement in parkinsonian motor signs and the ability to reduce parkinsonian medications with STN DBS that was not possible with GPi DBS, which made STN the target of choice for most major medical
Globus pallidus stimulation for parkinson’s disease
centers performing this procedure. This point however is frequently debated and two double blinded randomized clinical trials are now nearing completion that will address the critical question of which, if either, are better for the treatment of PD [3,4,9,17,24,36,43,46,104–106].
Role of GPi in Parkinson’s Disease Surgical targets for neuromodulation are based largely on their role in mediating the symptoms that are to be treated. The vast majority if not all subcortical sites to date that have been targeted are intimately a part of the basal ganglia thalamocortical circuitry. The functional organization of the basal ganglia and the role of GPi in the development of motor symptoms associated with PD have been refined over years by studies in animal models [88,107–115], imaging studies [116–123] and lesioning procedures [75,90,93,100,124–127]. The initial model of basal ganglia circuitry was based on the concept that [111,128,129] nuclei of the basal ganglia function as components of several segregated parallel circuits that also involve specific portions of the thalamus and cerebral cortex. These circuits take origin from different cortical areas, project to separate portions of the basal ganglia, which in turn project to separate portions of the thalamus, returning to the same areas of the frontal cortex from which they took origin. These parallel circuits are believed to include at least four circuits that are functionally distinct and are related to (1) skeletomotor, (2) oculomotor, (3) associative, and (4) limbic modalities. Of these the skeletomotor or ‘‘motor’’ circuit, has been considered most important in the pathogenesis of hypokinetic movement disorders such as Parkinson’s disease and hyperkinetic movement disorders including dystonia, hemiballismus, and dyskinesia [88]. In this model, the striatum is the principal input structure of the basal ganglia, while the GPi and the substantia nigra pars reticulata (SNr) are
95
the main output structures. Input and output structures are connected by a ‘‘direct’’ pathway that consists of a monosynaptic striato-pallidal (GPi) projection and an ‘‘indirect’’ pathway that is polysynaptic and passes from the striatum to the globus pallidus externus (GPe) and the subthalamic nucleus (STN) before terminating in GPi and SNr. The output nuclei (GPi and SNr) then project to ventralis anterior (VA) and ventralis lateralis pars oralis (VLo), the pallidal receiving area of the motor thalamus, which projects in turn, to motor cortical areas. The human analogue to VA is ventralis anterior and to VLo, ventralis oralis anterior and ventralis oralis posterior, Voa and Vop, respectively. The motor thalamus projects to multiple cortical areas in the frontal cortex including the supplementary motor area (SMA), the motor cortex (MC) and pre-motor cortical areas (PMC). In addition to the thalamic projection, GPi also projects to the brainstem pedunculopontine nucleus (PPN) and midbrain extrapyramidal area (MEA) (> Figure 95‐1). These regions in turn send projections to the thalamus and spinal cord and also play a significant role in mediating the development of parkinsonian motor signs, in particular the postural and gait disorders associated with PD [78,79]. Though inhibitory GABAergic projections and excitatory glutamatergic projections to and from the cortex are present within the intrinsic circuitry of the basal ganglia, dopaminergic projections to the striatum from the substantia nigra, pars compacta (SNc) play a major role in modulating striatal activity. Striatal DA enhances transmission along the direct pathway (via D1 receptors), and reduces transmission over the indirect pathway (via D2 receptors). Extra-striatal DA also acts on receptors in the globus pallidus, SNr, STN and the thalamus [130–135]. In PD the loss of dopaminergic neurons in SNc results in a reduction in striatal and extra-striatal DA. The reduced DAergic input in striatum leads to reduced activity in the direct pathway and
1579
1580
95
Globus pallidus stimulation for parkinson’s disease
. Figure 95‐1 Circuit diagrams of basal ganglia–thalamo-cortical circuitry and direct and indirect pathways under normal conditions (a) and in Parkinson’s disease (b) The black lines represent inhibitory connections, while the grey lines indicate excitatory connections. The thickness of the lines indicate the rate of discharge, thicker is increased, while thinner is decreased. Interrupted lines are used to represent irregular patterns of neuronal activity while the smaller grouped lines perpendicular to the interrupted lines are used to represent bursting activity. SNr, substantia nigra pars reticulata; SNc, substantia nigra pars compacta; STN, subthalamic nucleus; MEA, midbrain extrapyramidal area; PPN, pedunculopontine nucleus; GABA, gamma aminobutyric acid (Figure modified and reproduced with permission from [164].)
increased activity over the indirect pathway resulting in reduction of neuronal activity of GPe and an increase of neuronal discharge rates in the STN. The combination of reduced activity in the direct pathway and increased activity in the indirect pathway lead to increased mean discharge rates of neurons in GPi, and SNr [89,136,137]. These observations led to the hypothesis that hyperactivity of the STN was a major factor leading to the excessive increase in mean discharge rates in GPi and SNr, which led in turn to increased inhibition of thalamocortical pathways, a reduction in cortical activity and the classic motor symptoms associated with PD [138,139]. The ‘‘rate’’ model of basal ganglia function, while elegant in its simplicity has been questioned for its strict separation of direct and indirect
pathways, an inability to explain the improvement in parkinsonian motor symptoms after thalamotomy and reduction in dyskinesia following pallidotomy and has been supplemented by newer theories that place a larger role on altered patterns of activity, synchronized oscillatory activity in particular frequency ranges and disordered sensory processing in the basal ganglia thalamo-cortical circuit [140–143]. A common view held by both models, however, is that GPi is an important nodal point in mediating basal ganglia output and in mediating the development of parkinsonian motor signs. This physiological justification coupled with the ability to obtain high quality images and safely access GPi, made it a prime target for lesioning and one of the prime targets for DBS.
Globus pallidus stimulation for parkinson’s disease
Although the mechanisms of action of DBS remain under debate, most reports now support the hypothesis that DBS in the motor region of GPi (or STN) reduces the altered patterns of activity from GPi to the motor thalamus and brainstem structures leading to improvement in information processing in thalamo-cortical projections and less disruption in cortical-cortical signal processing resulting in a reduction in PD motor signs [144–146]. Recent studies also suggest that GPi in addition to the role played in motor circuits, also plays a major role in information processing in nonmotor cognitive circuits [147–150]. Whether these functions operate in parallel via functionally segregated pathways or through integrated circuits has been a point of contention given the apparent functional segregation demonstrated by lesioning and other studies versus anatomical studies that suggest significant interplay between motor and nonmotor circuits [129,151–155]. Whether operating in parallel or utilizing integration of motor and non-motor pathways, it is clear that the information carried along the cortico-striato-pallidal pathway is critical for proper motor execution and that it also plays an integrative role in cognitive information processing [121,147–150,156,157]. Alterations in neuronal activity in the final output pathway of this circuit, the GPi, have been associated with both hypokinetic and hyperkinetic movement disorders. Lesioning this region to reduce, or stimulation to modify, this output has been associated with improvement in motor function, making GPi a prime target for the treatment of these disorders [24,77,78,80,81,158,159].
Indications and Patient Selection for GPi DBS for PD Similar to the application of DBS for any neurological disorder, identification and selection of patients requires a multidisciplinary team that includes a
95
neurosurgeon trained in functional neurosurgery, a movement disorders neurologist, a neuropsychologist, psychiatrist and neurophysiologist. General selection criteria for patients undergoing GPi DBS include: 1. 2.
3. 4. 5.
6.
7.
Proper diagnosis of idiopathic PD using the British Brain Bank Criterion. Optimum medical management with either loss of efficacy or development of significant motor complications, i.e., drug-induced dyskinesias and/or motor fluctuations. Levodopa responsive disease. Absence of significant neuro-psychological co-morbidity. No major structural abnormality on neuroimaging studies that would increase surgical risk or minimize the benefit from DBS. Stable general medical condition improving the ability of the patients to tolerate all the components of surgery. Adequate social support.
The success of GPi DBS depends on proper selection of the patient, accurate placement of the DBS lead and a programmer who is familiar with programming GPi patients. In patient selection, age of the patient is not a major criterion as most of the published literature fails to show that there is an age benefits from surgery. While it has been demonstrated for pallidotomy that younger patients may receive more improvement following surgery, even patients in the oldest age group in this study received significant benefit from surgery [80,160]. As such one must consider the biological, not the chronological, age of the patient. Proper diagnosis of idiopathic PD is important as atypical parkinsonism, multisystem atrophy (MSA) and progressive supranuclear palsy (PSP) may resemble PD and may demonstrate a temporary response to antiparkinsonian medications. These patients do not respond to GPi DBS [9,160]. Similar to STN, all patients undergoing GPi DBS should be medically optimized prior to surgery
1581
1582
95
Globus pallidus stimulation for parkinson’s disease
and demonstrate a long term response to levodopa [3,4,9,17,24,36,43, 46,104–106]. The degree of response to levodopa pre-operatively as demonstrated from ‘‘on-off’’ testing is a strong indicator for the degree of benefit one can expect from GPi DBS and is an important predictor of outcome. [3,4,9,17,24,36,43,46,104–106] A positive correlation between the amount of benefit as determined from the UPDRS III motor subscore when the patient is off versus when they are on medication, i.e., on-off testing, has been demonstrated in several studies of DBS for PD [27,51]. In assessing levodopa responsiveness as an indication of DBS most studies have used a 25% or more reduction in their Unified Parkinson’s Disease rating Scale (UPDRS) motor score as a cutoff. The on score is determined after administration of therapeutic or supra-therapeutic (1.5 times) dose of levodopa and compared to the off state achieved after withholding all antiparkinsonian medication overnight. Use of the 25% on-off response as a cutoff for surgical candidacy should be used with caution, particularly in patients that have significant functional disability predominally from tremor, dystonia, a narrow window of response to medication or unpredictable on-off episodes. In such cases one has to evaluate the specifics of each patient’s condition and determine whether surgery will provide adequate functional benefits. Patients with pre-existing psychiatric or neuropsychological conditions should have these disorders treated and be in a stable state prior to undergoing DBS surgery. Many of these patients will not be candidates for DBS, particularly those with moderate dementia or an unstable psychiatric state. For patients with depression, therapy should always be initiated and optimized prior to surgery [160,161]. Preliminary imaging using magnetic resonance imaging (MRI) is essential before selecting a surgical candidate to rule out co-existing structural lesions that may affect the outcome of DBS [24,160]. DBS surgery is most commonly performed in awake patients and it is important that patients are in a state of general health that
will allow them to undergo the rigors of surgery and have the social support structure to comply with the demands of post-operative care and subsequent programming. The patient and family members have to understand the potential complications associated with the procedure and have realistic expectations about surgical outcome. In discussing the target site specific indications of GPi DBS over STN DBS in PD patients are still not clearly defined. Though traditionally it is argued that GPi DBS is more effective in controlling levodopa induced dyskinesia [9,24,36, 162] and less effective in controlling tremors and reducing the dosage of medication, the double blind trials performed thus far fail to support these claims [17,31,43,106]. Based on non-randomized studies, patients with significant levodopa induced dyskinesia, ‘‘off’’ or ‘‘on’’ dystonia, and cognitive abnormalities would be favored to undergo GPi DBS though there is as yet no class I evidence to support this. The comparison between GPi and STN as target sites for PD will be discussed later in the chapter.
The Globus Pallidus Target The globus pallidus is divided into two anatomical segments: internal (GPi) and external (GPe). These two segments are separated by the medial medullary lamina (> Figure 95‐2). The pallidal neurons from each segment are morphologically similar but functionally distinct. The GPi is bound laterally by GPe, medially by the internal capsule, ventrally by the optic tract and dorsally by the GPe and striatum. The motor territory of the GPi is ventral and posterior. A somatotopic organization has been described within the sensorimotor territory of GPi with neurons responsive to leg movement found dorsal and medial to arm neurons with the face found ventrally [81,150]. In a subsequent study with a larger sample size 3,869 neurons located within the GPi were examined for their receptive
Globus pallidus stimulation for parkinson’s disease
95
. Figure 95‐2 Location of the globus pallidus internus (GPi) and its relation to the surrounding structures as shown on the axial section of Schaltenbrand and Wahren atlas. The dark boundary between the GPi and GPe is the external lamina (Reproduced with permission from [227].)
field characteristics. This study showed that from dorsal to ventral one encountered a predominance of leg-related neurons more dorsally followed by arm-related and then head related neurons. Furthermore, arm- and head-related neurons were generally more lateral than legrelated neurons, while in the anterior to posterior direction there was a predominance of leg-related neurons found most anteriorly followed by arm and head-related neurons as one moved posteriorly in GPi [163] (> Figure 95‐3). In prior studies of pallidotomy improvement could be observed to occur within separate body parts as the lesion was progressively increased to involve specific portions of the sensorimotor pallidum (Vitek personal observations). While DBS works by activating output from the stimulated structure and may not have the same discrete effect on different body parts it is likely
that placement of a DBS lead either outside the sensorimotor territory or such that it affects only a portion of this region may lead to partial benefit. Similarly, involvement of adjacent fiber pathways or structures may also lead to a differential improvement in symptoms. An example of this has been reported for stimulation in different regions of the pallidum. Bejjani et al. [4,7] have demonstrated different clinical effects after stimulation of dorsal compared to more ventral regions of GP. With stimulation in more dorsal portions of the pallidum they reported improvement in akinesia and rigidity but an exacerbation of dyskinesias, whereas stimulation in the posteroventral portion of GP had a pronounced anti-dyskinetic effect, but worsened bradykinesia. Although the authors reported this as due to a differential effect of stimulation in different regions of the pallidum and
1583
1584
95
Globus pallidus stimulation for parkinson’s disease
. Figure 95‐3 Figure depicting the somatotopic organization within the sensorimotor territory of GPi with neurons responsive to leg movement found dorsal and medial to arm neurons with the face found ventrally. Drawings in this figure represent a composite of 10 patients with different body regions differently shaded. The numbers represent the distance in millimeters from the midline of each parasagittal plane (Figure modified and reproduced with permission from [81].)
activation of adjacent fiber pathways or structures that can either help or hinder ones efforts to ameliorate parkinsonian motor signs using GPi DBS.
Surgical Technique of GPi DBS The surgical technique primarily relies on stereotactic hardware (frames, frameless stereotaxy), image acquisition, navigation systems, stereotactic atlases and neurophysiological mapping techniques. The basic components of DBS implantation surgery involve frame placement, anatomical targeting, physiological mapping, evaluation of macrostimulation thresholds for improvement in motor symptoms or induction of side effects, implantation of the DBS electrode and implantable pulse generator (IPG).
Head-frame Placement essentially ‘‘two opposing targets,’’ it is likely given the 10.5 mm length of the lead (3,387, Medtronic, Minneapolis, MN) used in these patients that dorsal contacts were in GPe, resulting in activation of GPe output and suppression of STN activity leading to the development of dyskinesia [164]. Stimulation through more ventral contacts likely activated corticospinal tract fibers that were responsible for the worsening bradykinesia observed during stimulation. Evidence in support of this is derived from animal studies in which a DBS lead placed in the STN in a nonhuman primate model of PD that activated corticospinal tract fibers worsened bradykinesia while improving rigidity [165]. Stimulation at thresholds that did not activate the corticospinal tract improved both bradykinesia and rigidity in this study. Thus lead placement and choice of stimulation parameters, while intuitively obvious as critically important factors in determining outcome when performing GPi DBS, can lead to
GPi DBS implantation can be performed using a conventional frame based system or a newer frameless system. The frame-based approaches have been in place for decades and have proven accuracy and reliability. The most commonly used frames are the Leksell G frame (Elekta Instruments AB, Stockholm, Sweden) (> Figure 95‐4) and Cosman–Roberts–Wells (CRW) frame (Radionics Inc, Burlington, MA). In a survey of North American centers performing DBS surgeries, the CRW frame is the most commonly used followed by the Leksell frame [38]. The stereotactic accuracy of each frame has been well established [166]. We use the Leksell frame at our center for all the framebased procedures. Placement of the frame is done under local anesthesia unless anxiety or uncontrollable movements necessitate the use of sedation or general anesthesia. The frame is placed so that it is centered in the midline and is parallel to orbito-meatal line, in order to be
Globus pallidus stimulation for parkinson’s disease
. Figure 95‐4 Leksell stereotactic frame (Elekta, Stockholm, Sweden) placed over the head of a patient showing the correct method for placement of the Leksell head-frame. The frame should be placed parallel to orbito-meatal line in order to approximate the AC-PC plane. It is attached to the patient’s head using four pins under local anesthesia
parallel the anterior commissure (AC) – posterior commissure (PC) line. The recent introduction of the frameless systems for DBS lead placement [167,168] has provided an alternative approach for stereotactic targeting. NexFrame (Medtronic, Minneapolis, MN) is a commonly used frameless system (> Figure 95‐5). In this system 5–6 bone fiducial markers are placed over the skull in the outpatient clinic under local anesthesia 1–4 days before surgery. The preoperatively-obtained images are then loaded into a surgical navigation computer to plan a target and trajectory. We have compared the accuracy of Leksell frame based targeting to NexFrame frameless system and have found that though there was more difference between the intended and actual lead location in the frameless group when compared to the frame-based group it was not statistically
95
. Figure 95‐5 Intra-operative picture a frameless stereotactic placement of a DBS lead using Nexframe (Medtronic, MN) and Nexdrive (Medtronic, MN). Nexframe tower fits over the StimLoc base (Medtronic, MN) and needs to be aligned with the intended trajectory using image guided software. Nexframe then functions as a stable skull-mounted guide for the introduction of the DBS lead. It incorporates a microdrive that is used for microelectrode recording and final implant of DBS electrodes
significant (p ¼ 0.07) except in the x-plane (p ¼ 0.03) and the error was not specific to any target including GPi (p ¼ 0.44) [169]. Based on absolute numbers frame-based targeting would appear to be more accurate than frameless targeting, however, the difference was not statistically significant [169,170]. With increasing experience in the use of frameless systems and as newer systems are developed there will likely be little difference in accuracy and decision to use frame versus frameless systems may depend more on familiarity than on any clear differences in the accuracy of the technique.
1585
1586
95
Globus pallidus stimulation for parkinson’s disease
Imaging and Anatomic Targeting Computerized Tomography (CT) scans and MRI are the two main imaging modalities used for targeting when performing DBS implantations. A thin cut stereotactic CT (2 mm slices with no gap and no gantry tilt) is obtained after frame placement and is then fused with the stereotactic MRI on a planning station. The advantage of fusing the CT with MRI is the ability to avoid
image-distortions inherent to MR imaging [171] adding to the stereotactic accuracy [172–175]. The most common MRI sequences used for GPi targeting are a T1-weighted volumetric acquisition of the whole brain with gadolinium enhancement, a T2-weighted axial and coronal acquisition, and inversion recovery (IR) sequences. The T2 and IR sequences are good for delineating the GPi and optic tract (OT) (> Figure 95‐6a). Targeting the GPi is initially determined using
. Figure 95‐6 High resolution axial (a), coronal (c and d) and navigation view of post-contrast T1, T2 weighted and inversion recovery (IR) MR images showing the GPI, optic tract (OT) and structures in close proximity to it. Image A is an IR image showing the GPI (red dot), GPe and putamen. Figures (c) and (d) show the trajectory targeting the OT (image (d)). Image (b) shows the navigation view with vasculature in the sulci and peri-ventricular area
Globus pallidus stimulation for parkinson’s disease
indirect targeting techniques based on the standardized average distances of the site with respect to the AC, PC and the Mid-commissural point (MCP). Typical anatomical coordinates for the postero-ventral motor component of GPi are (19–21 mm lateral to the midline, 2–3 mm anterior to the MCP and 4–5 mm ventral to AC-PC plane). The anatomic targeting is further confirmed using a morphed stereotactic atlas included in the navigation. The common atlases used are the Schaltenbrand and Wahren [176] atlas, Schaltenbrand and Bailey [177] and Talairach and Tournoux [178] atlas. T2 and IR images are then used for ‘‘direct’’ anatomic targeting. Localization of the AC and the PC is done on T1-weighted images. Direct anatomical targeting of GPi is done by visualization of the pallidum on IR and T2 sequences with the ventral most part of the target directed at the OT (> Figures 95‐6c and 95‐6d). The next step is planning the entry point and trajectory. The strategy here is to avoid surface and sub-cortical vessels. In GPi DBS the angle of approach is preferably in a single sagittal plane parallel to the midline medio-lateraly and anterior to the coronal suture anteroposteriorly. The reason for an anterior entry point is to allow entry thru the frontal cortex, which helps position the higher contacts farther away from the internal capsule, which is posterior and medial to GPi. This approach reduces the risk of corticospinal tract activation during stimulation. An alternative approach would be to take a more lateral to medial angle, however, superimposing mapping tracks that are off the parasagital plane can be more difficult than when they are in the same angle as cuts in the atlas upon which the track are superimposed. The trajectory should also pass through the middle of a gyrus rather than into a sulcus and should be away from visualized basal ganglia vasculature and ventricular ependyma so as to minimize the chances of hemorrhagic complications (> Figure 95‐6b).
95
After trajectory planning, the patient is placed supine on the operating table and the frame attached to the table using an adaptor. Prophylactic antibiotics are given at least 30 min prior to incision. The head is prepped and draped in a sterile fashion. Under local anesthesia an approximately 6 cm parasagittal linear incision is made to encompass the entry point determined by the calculated arc and ring angles. Raney clips are then applied to the scalp edges to achieve hemostasis and a selfretaining retractor is placed. A burr-hole is placed with a 14 mm perforating drill bit centered on the calculated entry point marked on the skull. Hemostasis is achieved with bone wax and bipolar cautery. A Medronic Stim-Loc anchoring device (Medtronic, Minneapolis, MN) burr-hole base ring is then placed on the burr-hole and secured with two screws which is used at the end of the procedure to anchor the DBS electrode (> Figures 95‐7a and 95‐7b). The dura is then cauterized and opened exposing the underlying surface of the brain. The microdrive is then assembled and cannulae inserted 20–40 mm above the initial anatomical target. We generally insert the cannulae 25 mm above the target to avoid lenticulostriate vessels found deeper. Gel-form and fibrin glue is applied to minimize cerebrospinal fluid (CSF) loss and air entry into the skull. Subsequently, microelectrode recording and stimulation is undertaken.
Microelectrode Recording/ Mapping To accurately define the target structure and allow for proper placement of the DBS lead, electrophysiological mapping is performed to delineate the posterior and lateral borders of the GPi and identify nearby critical structures such as the optic tract and internal capsule. We believe microelectrode mapping is crucial in order to give one the best chance for optimal
1587
1588
95
Globus pallidus stimulation for parkinson’s disease
. Figure 95‐7 Intra-operative picture of the StimLoc (Medtronic, MN) lead anchoring burr hole system. (a) Shows the StimLoc base fixed over the burr hole before beginning of MER (b) Shows the StimLoc clip and cap in place after placing the electrode with the electrode coiled around the burrhole
placement of the DBS lead given anatomical inaccuracies due to image distortion and intraoperative brain shifts secondary to CSF loss, and pneumocephalus that can lead to inaccuracies in defining the initial target coordinates and shifts in the target itself once the skull is opened [179]. Microelectrode mapping is used to precisely define the target and its boundaries as well as nearby critical structures. This technique is similar whether one is placing a lesion or lead in the GPi. The mapping strategy has been previously described in detail see Vitek et al. [81], a brief summary is provided here. Microelectrode mapping is performed using platinum-iridium glass coated microelectrodes dipped in platinum black with an impedance of around 0.3–0.5 mO [81,179– 181]. These platinum-iridium microelectrodes are capable of recording single unit activity and can also be used for micro-stimulation up to 100 mA without significant breakdown in their recording qualities. A motorized microdrive is used to advance the microelectrode. The objective of the mapping strategy is to identify the sensorimotor territory of GPi, define its posterior and
lateral border, and identify the optic tract ventral to GPi, the internal capsule posterior and medial to GPi and the transition point from GPe to GPi. These objectives are achieved with a minimum of three microelectrode penetrations [4,51,81, 182–184] (> Figures 95‐8a and 95‐8b). Although some have been performed using only one or two penetrations [43,185], this approach does not consistently allow for an accurate map of the borders of GPi and surrounding structures and requires one to translate information derived from one or two tracts that give you a 2-dimensional picture into a 3-dimensional picture of your location within GPi, rather than obtaining that 3-dimensional map directly (> Figures 95‐8a and 95‐8b). When mapping GPi a typical first trajectory passes through striatum, GPe, GPi and ends at the OT (> Figure 95‐9). The striatal cells are relatively silent and show occasional, transient, low frequency discharges usually in response to injury as one advances the microelectrode or stimulates through the microelectrode. The next structure that is encountered is GPe. GPe in PD shows spontaneous, low
Globus pallidus stimulation for parkinson’s disease
95
. Figure 95‐8 (a) Three dimensional pallidal electrophysiological map for DBS placement. Three dimensional (3D) electrophysiological map used for placement of the DBS electrode with electrophysiological guidance. Sagittal (top left), coronal (top right), axial (bottom left), and 3D oblique (bottom right) views are shown. Structures are color labeled with striatum represented in blue, external globus pallidus (GPe) in green, and internal globus pallidus (GPi) in red. The number to the lower right of each figure corresponds to the distance in mm from the anterior-commissure– posterior-commissure line (sagittal and axonal views) and midcommissural point. The 3D map is generated live during the surgical procedure using the OneTrack software system. 32 P = posterior; A = anterior; S = superior; I = inferior; L = left; R = right (Figure modified and reproduced with permission from [37].). (b) Line drawings depicting the strategy of MER in GPi. The diagrams are the physiological maps for a patient that ‘‘form fit’’ on parasagittal planes of the basal ganglia taken from the Schaltenbrand and Bailey atlas. The different shades within each line depict a different pattern of neuronal activity, whereas the x at the end of track 3 depicts the location of the optic tract (OT). The number at the end of track 1 indicates capsular response threshold to microstimulation. The long continuous black line of track 1 in the lower panel represents a track that penetrated through a portion of the internal capsule (Figure modified and reproduced with permission from [81].)
1589
1590
95
Globus pallidus stimulation for parkinson’s disease
. Figure 95‐9 Line diagram showing the different patterns of neural activity encountered in a MER track of the striatum, GPe, GPi and surrounding structures. The line drawing is superimposed over the lateral 21.5 parasagittal plane of the Schaltenbrand and Bailey atlas. It shows the activity in putamen, border cells, the high frequency pausers (HFD-P) and low frequency bursters (LFD-B) in GPe and activity in GPi and nucleus basalis (NB) (Figure reproduced with permission from [81].)
frequency activity generally with two distinct patterns: 1.
2.
Burster cells: these are neurons that discharge as intermittent high frequency bursts (seen in 10–20% of cells) [81]. Pauser cells: these are low frequency tonically discharging cells with pauses and form the majority (80–90%) of cells in GPe [81].
These two types of cells are the hallmark of GPe neurons. Once the electrode leaves GPe it passes through a lamina characterized by the absence of neuronal discharge except for occasional border cells. Border cells are characterized by regular, low frequency, tonic firing. After passing through the lamina one enters GPi. In the GPi, the majority of neurons exhibit high-frequency tonic discharge (HFD cells) [81]. In patients with PD, other patterns that may be present are highfrequency bursts or activity with little pauses between the bursts, creating a ‘‘chugging’’ sound, or lower frequency bursts in the range of 4–6 Hz [81]. In summary characteristics of neurons in the sensorimotor GPi are: 1.
Response to passive and active movement, with approximately 46% of cells responding
2. 3. 4.
to active or passive movement. Sensorimotor responses are found predominately in the postero-lateral portions of the GPi [81]. Increased background activity. High frequency tonic activity. Presence of tremor-related cells in patients with tremor.
Internal accessory laminae are sometimes encountered in GPi in which case one passes from an active region of GPi into a quiet region and subsequently back into an active region again. Once one passes through GPi two regions may be identified based on the location of the first track. These are the internal capsule and optic tract. If one is too lateral neither the optic tract or internal capsule may be identified. Laterally placed trajectories will encounter a large segment of GPe and a small segment of GPi usually with a large lamina between them. Using the Schaltenbrand and Bailey Atlas the optic tract is defined just below the ventral border of GPi in the lateral 18.5–21.5 planes. The distance between the ventral border of GPi and the optic tract will vary based on the relative anterior to posterior location of the track. In more posterior tracks one will pass through the internal capsule before encountering the OT. In such cases microstimulation
Globus pallidus stimulation for parkinson’s disease
just above the OT will elicit evoked movements on the contralateral side of the body, i.e., in the lips, tongue, upper or lower limb. The OT can be identified by audio monitors that translate the recording of the voltage change induced by activation of fibers in the optic tract generated by flashing a light in front of the eyes. Microstimulation in the vicinity of the OT also leads . Figure 95‐10 Intra-operative cross-table lateral fluoroscopic view of the DBS lead in place with routinely used to monitor the DBS lead location and ascertain that it has not been displaced while anchoring and fixing the lead
95
to phosphenes in the visual fields, which may be described by the patient. If the trajectory is placed too lateral or too medial to the motor part of GPi, the OT will not be encountered in the most ventral part of the tract. Similar to posterior trajectories, medial trajectories will elicit capsular stimulation before one encounters the OT. The final location of the GPi electrode in PD is around 5–6 mm anterior to the posterior border and 2–3 mm medial to the lateral border of GPi with the ventral tip sitting just above the OT. Macrostimulation using the DBS electrode itself is then used to determine benefits and side effects. The two commercially available electrodes have four contacts of 1.5 mm height and 1.27 mm in diameter and differ only in the spacing between contacts: 1.5 mm in the 3387 model and 0.5 mm in the 3389 model (Medtronic, Minneapolis, MN). Most often model the 3387 has been used for implantation in GPi; however consideration should be given to using the 3389 based on the length of the trajectory through GPi to allow for the placement of more contacts within the GPi and the option of using double monopolar contacts for stimulation to affect a larger area of the sensorimotor territory. Fluoroscopy is routinely used by many centers to monitor the DBS lead location
. Figure 95‐11 Intra-operative picture of the stage II of the surgery showing tunneling of the electrode in the neck and placement of IPG. The cranial end of the figure shows the parieto-occipital incision to expose the distal lead of the DBS lead. The caudal end of the picture shows the infra-clavicular incision for placement of the IPG. The tunneler used for subcutaneous tunneling is shown between the two incisions to depict the direction of the tunneling
1591
1592
95
Globus pallidus stimulation for parkinson’s disease
and ascertain that it has not been displaced while anchoring and fixing the lead (> Figure 95‐10). Medronic Stim-Loc anchoring device (Medtronic, Minneapolis, MN) (> Figure 95‐7b) is then used to secure the DBS lead. Once secured, the distal end of the DBS lead is tunneled to the parietal/occipital region and placed in the subgaleal space. The second stage of the DBS procedure is implantable pulse generator (IPG) implantation. This is performed under general anesthesia either on the same day or in a staged fashion. A sub-clavicular, subfacial pocket is used for the placement of the IPG which is then connected to the DBS electrode using an extension wire which is tunneled from the parietal incision to the pocket (> Figure 95‐11).
2.
3.
4.
low voltages, while using a contact that is too deep and or posterior may result in worsening bradykinesia due to activation of the corticospinal tract. [164,165,186]. Rate: Most studies suggest that GPi DBS for PD is effective at rates similar to those used for STN although higher or lower rates may be used. Generally the range of effective reported rates range from 130–185 Hz [186,187]. Pulse-width: While higher pulse widths have been used for the treatment of dystonia, pulse widths for PD are most commonly 60–90 ms with some as high as 120 ms or more [186,187]. Amplitude: Amplitude can be optimized once the most effective contact(s) is(are) identified [186,187].
Programming Parameters for GPi DBS for PD There are multiple variables that may be adjusted to maximize clinical efficacy and minimize stimulation induced side effects. Initial programming is always refined by using intra-operative macrostimulation data and a mono-polar review to identify the thresholds of stimulation for improvement in parkinsonian motor signs as well as the thresholds for inducing side effects at the level of each contact. The four variables that are used in programming are choice of contacts (0, 1, 2 or 3 used either as the cathode or anode), frequency of stimulation (hertz), pulse-width (ms) and amplitude (voltage). 1.
Active contact (the cathode): Location of the active contact will depend on the relative depth of lead implantation. We generally implant the lead such that one contact, generally contact 1 or 2 is located near the dorsal border of GPi or just below this site. Stimulation with contacts dorsal to this point lie in the GPe and can be effective but may induce dyskinesia at
Outcomes After GPi DBS in PD Outcomes after GPi DBS for PD have been assessed using the Unified Parkinson’s Disease Rating Scale (UPDRS). Most studies have used the change in the motor subscore (UPDRS III) from the medication OFF and stimulation OFF condition (OFF/OFF) to the medication OFF, stimulation ON (OFF/ON) state to objectively assess the effect of stimulation on parkinsonian motor signs. In addition there are other markers of efficacy that include improvement in Activities of Daily Living scores (ADLs) (UPDRS II), reduction in dyskinesias and reduction in off-time or increase in on-time without dyskinesias usually through patient diaries. Though a variety of outcomes have been reported across studies, the vast majority of studies have reported that GPi DBS is effective in reducing all the cardinal motor signs of PD as well as improving motor fluctuations, reducing dyskinesia and increasing on time [188] Ghika et al. [9] reported
Globus pallidus stimulation for parkinson’s disease
a mean improvement of more than 50% in the Off UPDRS motor score and the activities of daily living (ADL) score. They also reported that the mean off time decreased from 40!10%, and the mean dyskinesia scores were reduced to one-third. Burchiel et al. reported a 40% improvement in the Off UPDRS motor scores [17], while Kumar et al. reported Off UPDRS motor score improvement by 31% and ADL scores by 39%. During the ‘‘on’’ medication period, the reduction in the total ‘‘on’’ dyskinesias score was 66% and improvement in the ADL score was 32% [24]. Loher et al. reported 41% improvement in Off UPDRS motor scores at 12 months and 34% improvement in ADL scores [105]. Volkman et al. reported a 56% improvement in the OFF UPDRS motor score in the 1 year after bilateral GPi DBS, which diminished to 24% five years post-surgery. The study also reported a sustained reduction in dyskinesias at 5 years follow up of 64% [36]. Rodriguez et al. reported a similar reduction in Off UPDRS by
95
46% and reduction in dyskinesia scores by 76% [39]. However, unlike the loss of benefit in the Off UPDRS score reported by Volkmann et al. [36], the benefit reported by Rodriguez et al. was maintained out 4 years [39,40]. A table summarizing studies of GPi DBS and change in motor scores is included in > Table 95‐1. As witnessed from the table results of improvement for GPi DBS vary considerably across studies with some reporting little or no benefit, while others reported 46% improvement sustained out 4 years. Much of this variability is likely due to improper lead location, poor patient selection and/or choice of programming parameters. Given the size of GPi, lead placements that are outside of the sensorimotor territory may not be associated with side effects that would be observed during macrostimulation intraoperatively; information commonly used by the operating team to indicate whether or not the lead has been placed in a reasonable location within the sensorimotor GPi. Because macrostimulation in sensorimotor
. Table 95‐1 Percentage improvement in motor score in studies of bilateral DBS of the internal globus pallidus Percentage improvement during follow-up Study DBS Study Group Burchiel Krack Scotto di Luzio Volkmann Ghika Volkmann Kumar Durif Loher Rodrigues Mean
3 months (%)
6 months (%)
9 months (%)
12 months (%)
24 months (%)
3 years (%)
4 years (%)
5 years (%)
37 39 49
39 44 56 53
42
52
51 55
43
24
50
44 31 36
26 41
36
38 46
39
42
52
43
44
43
46
24
A table summarizing various studies of GPi DBS for PD and change in motor scores (Table modified and reproduced with permission from [37].)
1593
1594
95
Globus pallidus stimulation for parkinson’s disease
GPi may not give immediate benefit in motor signs the lack of a therapeutic effect during macrostimulation is not necessarily an indicator of poor lead placement. However, the lack of capsular effects with high frequency and high voltages are suggestive that one may be too anterior, too lateral or both. If left in this location leads placed outside the sensorimotor territory will not provide optimal motor improvement or patients may lose their benefit over time. Similarly leads placed too close to the internal capsule, should be noted during macrostimulation and moved away from the capsule. If they are left in place it is likely that one will not be able to reach a voltage when programming postoperatively that will affect enough of the sensorimotor GPi to optimally improve motor signs before activating the corticospinal tract.
Comparison of GPi and STN DBS in PD Multiple retrospective and prospective studies have compared the outcomes of GPi and STN DBS in PD [27,31,189]. Both targets have been shown to be beneficial [27,190]. It is often argued based on these studies that STN DBS is more effective and allows for a greater reduction in post-operative medication [27,51,188,191–193]. The only three prospective randomized double blind studies comparing these two targets, however, fail to authenticate any of these arguments. Burchiel et al. [17] randomized ten patients with idiopathic PD, dyskinesia, and motor fluctuations to implantation of bilateral GPi or STN stimulators. Patients and evaluating clinicians were blinded to the stimulation site throughout the study period. After analysis of follow-up data they reported 39 and 44% improvement in Off UPDRS scores in GPi and STN groups, respectively, after 12 months. Rigidity, tremor, and bradykinesia improved equally in both groups. ‘‘On’’ UPDRS scores were improved more by GPi than by STN stimulation. Reduction in dyskinesia was
equally seen in DBS at either site, although the medication requirement was reduced only in the STN group. An extension of this randomized, blinded pilot study was carried out and extended enrollment included 23 patients with idiopathic PD, dyskinesia and motor fluctuations [43]. They were randomized to implantation of bilateral GPi or STN stimulators. Patients and evaluating clinicians were blinded to stimulation site. At 1 year follow up, Off UPDRS motor subscores were improved in both GPi and STN groups (39 vs. 48%). Medication reduction was more marked in the STN group but was not statistically significant from the GPi group (P = 0.08). Dyskinesias were reduced in both GPi and STN groups (89% vs. 62%). Cognitive and behavioral complications were observed only in combination with STN stimulation. Anderson et al. also found increased frequency of postoperative delirium and confusion in STN stimulation patients [43]. A third prospective, randomized, blinded study assessed the effects of unilateral DBS in the STN versus GPi on fine motor skills in 33 patients with advanced PD [194]. Stimulation of either the STN (18 subjects) or GPi (15 subjects) in the off medication state significantly improved movement time and dexterity with no significant difference between the two targets. A recent study that systematically examined the long-term effects of unilateral GPi or STN DBS on the force-producing capabilities of both limbs under unimanual and bimanual motor conditions has found persistent improvements in the control and coordination of grasping forces during maximal efforts and functional dexterous actions [106]. Overall, the results indicate that unilateral GPi or STN DBS is effective in improving overall motor function, maximum force production and the control and specification of grasping forces in both limbs and these effects persist at least 1 year after DBS surgery [106]. Stimulation produced greater clinical gains in motor function and strength for the contralateral limb; however significant improvements were also observed in the ipsilateral limb [106]. In general
Globus pallidus stimulation for parkinson’s disease
there appears to be little difference in the motor scores of patients receiving either STN or GPi DBS in trials where patients were randomized and assessments were blinded. On the other hand these studies suggest that DBS in the STN may result in greater cognitive and neurobehavioral changes than GPi DBS in patients with PD [36,43,193]. In summary, there are multiple open label trials that suggest STN is superior to GPi in terms of motor improvement and reduction in antiparkinsonian medication. Whether this is true or not remains to be determined as the small numbers of randomized blinded studies performed to date do not support this conclusion and a valuable target for the treatment of PD patients that may have less cognitive risk is being ignored by the vast majority of centers performing this procedure. The size of the GPi may make it a more formidable target for DBS in terms of affecting a large enough region of the GPi to optimize benefit with stimulation in this region, however it also allows one to place a lead in the sensorimotor region of GPi where there may be limited encroachment of the DBS current field on non-motor regions. While the small size of the STN obviates the problem of affecting a large enough region of the sensorimotor territory during DBS, it may induce another by the fact that associative and limbic regions of the STN are likely affected by stimulation in this small region where limbic, associative and motor territories all lie within a few millimeters of one another and may account for the increased incidence of cognitive side effects reported during stimulation in this target. Concerning the difference in medication reduction reported by some centers, stimulation in GPi directly suppresses dyskinesia and in some studies investigators may not have reduced medications since they were not forced to consider this option as with STN DBS where stimulation may worsen dyskinesia if antiparkinsonian medication is not reduced. In comparing these two targets one must be careful to take all the nuances into consideration less we fail to take advantage of all the
95
targets available to us and reap the potential advantages that each may offer.
Complications of GPi DBS The complications of DBS surgery can generally be classified under four headings: 1. 2. 3. 4.
Intracranial hemorrhages Infections Hardware related complications Stimulation related complications
The incidence of reported complications varies among centers. Intra-operative hemorrhages during GPi DBS vary from center to center with the range from 0.6 to 10% [188,195–197]. Lenticulostriate arteries that arise from the anterior circulation, are developmentally different and more prone to the effects of hypertension could be responsible for the hemorrhagic complications during GPi DBS [196] (> Figure 95‐12). Additional . Figure 95‐12 Axial non-contrast CT scan showing intra-parenchymal hemorrhage during the placement of a DBS electrode. The hemorrhage is located in the region of GPe
1595
1596
95
Globus pallidus stimulation for parkinson’s disease
factors responsible are coagulopathies, intraoperative hypertension, rapid advancement of the cannula, and the use of anti-platelet medications [195–197]. Reported infection rates in DBS surgery vary widely, from less than 1 to as high as 15% [179,188,195,196,198–220]. Hardware related complications are the most common with a varying incidence from 2.7 to 50% [204,216,221–226]. Stimulation related complications are associated with programming of the DBS lead and are for the most part, reversible and can sometimes be obviated by various programming strategies. If a DBS lead is placed sub-optimally, however, even the best programmer cannot help, and the patient may require revision surgery. In summary, GP DBS has been demonstrated to be a safe and effective target for the treatment of advanced PD. Whether it is equivalent to, better or less effective than STN DBS remains to be determined and will require evaluation with large randomized double blinded controlled studies. Two studies now underway will hopefully answer this question.
References 1. Benabid AL, et al. Chronic VIM thalamic stimulation in Parkinson’s disease, essential tremor and extrapyramidal dyskinesias. Acta Neurochir Suppl (Wien) 1993;58:39-44. 2. Caparros-Lefebvre D, et al. Chronic thalamic stimulation improves tremor and levodopa induced dyskinesias in Parkinson’s disease. J Neurol Neurosurg Psychiatry 1993;56(3):268-73. 3. Iacono RP, et al. Chronic anterior pallidal stimulation for Parkinson’s disease. Acta Neurochir (Wien) 1995;137 (1–2):106-12. 4. Bejjani B, et al. Pallidal stimulation for Parkinson’s disease. Two targets? Neurology 1997;49(6):1564-9. 5. Duff J, Sime E. Surgical interventions in the treatment of Parkinson’s disease (PD) and essential tremor (ET):medial pallidotomy in PD and chronic deep brain stimulation (DBS) in PD and ET. Axone 1997;18(4):85-9. 6. Obeso JA, et al. Surgical treatment of Parkinson’s disease. Baillieres Clin Neurol 1997;6(1):125-45. 7. Bejjani BP, et al. Deep brain stimulation in Parkinson’s disease: opposite effects of stimulation in the pallidum. Mov Disord 1998;13(6):969-70.
8. Galvez-Jimenez N, et al. Pallidal stimulation in Parkinson’s disease patients with a prior unilateral pallidotomy. Can J Neurol Sci 1998;25(4):300-5. 9. Ghika J, et al. Efficiency and safety of bilateral contemporaneous pallidal stimulation (deep brain stimulation) in levodopa-responsive patients with Parkinson’s disease with severe motor fluctuations: a 2-year follow-up review. J Neurosurg 1998;89(5):713-8. 10. Kumar R, et al. Double-blind evaluation of subthalamic nucleus deep brain stimulation in advanced Parkinson’s disease. Neurology 1998;51(3):850-5. 11. Kumar R, et al. Pallidotomy and deep brain stimulation of the pallidum and subthalamic nucleus in advanced Parkinson’s disease. Mov Disord 1998;13 Suppl 1:73-82. 12. Starr PA, Vitek JL. Bakay RA. Deep brain stimulation for movement disorders. Neurosurg Clin N Am 1998;9(2):381-402. 13. Tan AK, et al. Stereotactic microelectrode-guided posteroventral pallidotomy and pallidal deep brain stimulation for Parkinson’s disease. Ann Acad Med Singapore 1998;27(6):767-71. 14. Tasker RR. Deep brain stimulation is preferable to thalamotomy for tremor suppression. Surg Neurol 1998;49(2):145-53; discussion 153-4. 15. Ardouin C, et al. Bilateral subthalamic or pallidal stimulation for Parkinson’s disease affects neither memory nor executive functions: a consecutive series of 62 patients. Ann Neurol 1999;46(2):217-23. 16. Brown RG, et al. Impact of deep brain stimulation on upper limb akinesia in Parkinson’s disease. Ann Neurol 1999;45(4):473-88. 17. Burchiel, KJ, et al. Comparison of pallidal and subthalamic nucleus deep brain stimulation for advanced Parkinson’s disease: results of a randomized, blinded pilot study. Neurosurgery 1999;45(6):1375-82; discussion 1382-4. 18. Caparros-Lefebvre D, et al. Improvement of levodopa induced dyskinesias by thalamic deep brain stimulation is related to slight variation in electrode placement: possible involvement of the centre median and parafascicularis complex. J Neurol Neurosurg Psychiatry 1999;67 (3):308-14. 19. Kumar K, et al. Deep brain stimulation of the ventral intermediate nucleus of the thalamus for control of tremors in Parkinson’s disease and essential tremor. Stereotact Funct Neurosurg 1999;72(1):47-61. 20. Limousin-Dowsey P, et al. Thalamic, subthalamic nucleus and internal pallidum stimulation in Parkinson’s disease. J Neurol 1999;246 Suppl 2:II42-5. 21. Benabid AL, et al. Dyskinesias and the subthalamic nucleus. Ann Neurol 2000;47(4 Suppl 1):S189-92. 22. Benabid AL, et al. Subthalamic stimulation for Parkinson’s disease. Arch Med Res 2000;31(3):282-9. 23. Krack P, et al. Thalamic, pallidal, or subthalamic surgery for Parkinson’s disease? J Neurol 2000;247 Suppl 2: III22-34.
Globus pallidus stimulation for parkinson’s disease
24. Kumar R, et al. Deep brain stimulation of the globus pallidus pars interna in advanced Parkinson’s disease. Neurology 2000;55(12 Suppl 6):S34-9. 25. Lang, AE, Surgery for levodopa-induced dyskinesias. Ann Neurol 2000;47(4 Suppl 1):S193-9; discussion S199-202. 26. Benabid AL, et al. Deep brain stimulation for Parkinson’s disease. Adv Neurol 2001;86:405-12. 27. Krause M, et al. Deep brain stimulation for the treatment of Parkinson’s disease: subthalamic nucleus versus globus pallidus internus. J Neurol Neurosurg Psychiatry 2001; 70(4):464-70. 28. Lozano AM. Deep brain stimulation for Parkinson’s disease. Parkinsonism Relat Disord 2001;7(3):199-203. 29. Luquin MR. Implants of carotid body cells as a treatment alternative for Parkinson disease. Rev Med Univ Navarra 2001;45(2):49-54. 30. Starr PA. Placement of deep brain stimulators into the subthalamic nucleus or Globus pallidus internus: technical approach. Stereotact Funct Neurosurg 2002;79 (3‐4):118-45. 31. Vitek JL. Deep brain stimulation for Parkinson’s disease. A critical re-evaluation of STN versus gpi DBS. Stereotact Funct Neurosurg 2002;78(3‐4):119-31. 32. Benabid AL. Deep brain stimulation for Parkinson’s disease. Curr Opin Neurobiol 2003;13(6):696-706. 33. Lozano AM, Pahwa R. Deep brain stimulation surgery for Parkinson’s disease: mechanisms and consequences. Parkinsonism Relat Disord 2004;10 Suppl 1:49-57. 34. Lyons KE, Mahant N. Deep brain stimulation in Parkinson’s disease. Curr Neurol Neurosci Rep 2004;4 (4):290-5. 35. Volkmann J. Deep brain stimulation for the treatment of Parkinson’s disease. J Clin Neurophysiol 2004;21(1): 6-17. 36. Volkmann J, et al. Long-term results of bilateral pallidal stimulation in Parkinson’s disease. Ann Neurol 2004;55 (6):871-5. 37. Walter BL, Vitek JL. Surgical treatment for Parkinson’s disease. Lancet Neurol 2004;3(12):719-28. 38. Ondo WG, Bronte-Stewart H. The North American survey of placement and adjustment strategies for deep brain stimulation. Stereotact Funct Neurosurg 2005;83 (4):142-7. 39. Rodrigues JP, et al. Globus pallidus stimulation in advanced Parkinson’s disease. J Clin Neurosci 2007;14 (3):208-15. 40. Rodrigues JP, et al. Globus pallidus stimulation improves both motor and nonmotor aspects of quality of life in advanced Parkinson’s disease. Mov Disord 2007;22 (13):1866-70. 41. Hariz MI, et al. Multicenter study on deep brain stimulation in Parkinson’s disease: an independent assessment of reported adverse events at 4 years. Mov Disord 2008;23(3):416-21.
95
42. Limousin P, Martinez-Torres I. Deep brain stimulation for Parkinson’s disease. Neurotherapeutics 2008; 5(2):309-19. 43. Anderson VC, et al. Pallidal vs subthalamic nucleus deep brain stimulation in Parkinson disease. Arch Neurol 2005;62(4):554-60. 44. Rocchi L, et al. Comparison between subthalamic nucleus and globus pallidus internus stimulation for postural performance in Parkinson’s disease. Gait Posture 2004;19(2):172-83. 45. Boucai L, Merello M. Functional surgery for Parkinson’s disease treatment: a structured analysis of a decade of published literature. Br J Neurosurg 2004;18(3):213-22. 46. Alberts JL, et al. Comparison of pallidal and subthalamic stimulation on force control in patient’s with Parkinson’s disease. Motor Control 2004;8(4):484-99. 47. Scotto di Luzio AE, et al. Which target for DBS in Parkinson’s disease? Subthalamic nucleus versus globus pallidus internus. Neurol Sci 2001;22(1):87-8. 48. Benabid AL, et al. Deep brain stimulation of the corpus luysi (subthalamic nucleus) and other targets in Parkinson’s disease. Extension to new indications such as dystonia and epilepsy. J Neurol 2001;248 Suppl 3:III37-47. 49. Allert N, et al. Effects of bilateral pallidal or subthalamic stimulation on gait in advanced Parkinson’s disease. Mov Disord 2001;16(6):1076-85. 50. Katayama Y, et al. Double blinded evaluation of the effects of pallidal and subthalamic nucleus stimulation on daytime activity in advanced Parkinson’s disease. Parkinsonism Relat Disord 2000;7(1):35-40. 51. Krack P, et al. Subthalamic nucleus or internal pallidal stimulation in young onset Parkinson’s disease. Brain 1998;121(Pt 3):451-7. 52. Cobb S, et al. Section of U fibers of motor cortex in cases of paralysis agitans (Parkinson’s disease); report of 9 cases. Arch Neurol Psychiatry 1950;64(1):57-9. 53. Walker AE. Cerebral pedunculotomy for the relief of involuntary movements; hemiballismus. Acta Psychiatr Neurol 1949;24(3‐4):723-9. 54. Walker AE. Cerebral pedunculotomy for the relief of involuntary movements. II. Parkinsonian tremor. J Nerv Ment Dis 1952;116(6):766-75. 55. Spiegel EA, Wycis HT. Effect of thalamic and pallidal lesions upon involuntary movements in choreoathetosis. Trans Am Neurol Assoc 1950;51:234-7. 56. Meyers R. Dandy’s striatal theory of the center of consciousness; surgical evidence and logical analysis indicating its improbability. Trans Am Neurol Assoc 1950; 51:44-9. 57. Meyers R, Hayne R. Electrical potentials of the corpus striatum and cortex in parkinsonism and hemiballismus. Trans Am Neurol Assoc 1948;73(73 Annual Meeting):10-4. 58. Meyers R, Hayne R, Knott J. Electrical activity of the neonstriatum, paleostriatum and neighbouring structures
1597
1598
95 59. 60.
61.
62.
63. 64. 65.
66.
67.
68.
69. 70. 71. 72.
73. 74.
75. 76.
77.
78.
79.
Globus pallidus stimulation for parkinson’s disease
in parkinsonism and hemiballismus. J Neurol Neurosurg Psychiatry 1949;12(2):111-23. Redfern RM. History of stereotactic surgery for Parkinson’s disease. Br J Neurosurg 1989;3(3):271-304. Brion S, Guiot G. [Treatment of dyskinesia by pallidal surgery]. Rev Esp Otoneurooftalmol Neurocir 1956;15 (86):400-4. Guiot G, Brion S. [Neurosurgery of choreoathetosic and Parkinsonian syndromes]. Sem Hop 1952;28(49): 2095-9. Guiot G, Brion S. [Treatment of abnormal movement by pallidal coagulation]. Rev Neurol (Paris) 1953; 89(6):578-80. Guiot G, Brion S. [Surgery of globus pallidus in dyskinesia]. Sem Hop 1955;31(32):1838-45. Cooper IS. Anterior chorodial artery ligation for involuntary movements. Science 1953;118(3059):193. Cooper IS. Ligation of the anterior choroidal artery for involuntary movements; parkinsonism. Psychiatr Q 1953;27(2):317-9. Cooper, IS, Effect of anterior choroidal artery ligation on involuntary movements and rigidity. Trans Am Neurol Assoc 1953;3(78th Meeting):6-7; discussion 8-9. Cooper IS. Surgical occlusion of the anterior choroidal artery in parkinsonism. Surg Gynecol Obstet 1954; 92(2):207-19. Cooper IS. Poloukhine N. Chemopallidectomy: a neurosurgical technique useful in geriatric parkinsonians. J Am Geriatr Soc 1955;3(11):839-59. Cooper IS, Poloukhine N. The globus pallidus as a surgical target. J Am Geriatr Soc 1956;4(12):1182-207. Spiegel E, et al. Stereotactic apparatus for operations on the human brain. Science 1947;106:349-50. Leksell L. Stereotactic apparatus for intracerebral surgery. Acta Chir Scand 1949;99:229-33. Bakay RA, DeLong MR, Vitek JL. Posteroventral pallidotomy for Parkinson’s disease. J Neurosurg 1992;77 (3):487-8. Laitinen LV. Pallidotomy for Parkinson’s disease. Neurosurg Clin N Am 1995;6(1):105-12. Laitinen LV, et al. Leksell’s posteroventral pallidotomy in the treatment of Parkinson’s disease. J Neurosurg 1992;76(1):53-61. Lozano AM, Lang AE. Pallidotomy for Parkinson’s disease. Neurosurg Clin N Am 1998;9(2):325-36. Lozano AM, et al. Microelectrode recording-guided posteroventral pallidotomy in patients with Parkinson’s disease. Adv Neurol 1997;74:167-74. Okun MS, Vitek JL. Lesion therapy for Parkinson’s disease and other movement disorders: update and controversies. Mov Disord 2004;19(4):375-89. Vitek JL, Bakay RA. The role of pallidotomy in Parkinson’s disease and dystonia. Curr Opin Neurol 1997;10 (4):332-9. Vitek JL, Bakay RA, DeLong MR. Microelectrode-guided pallidotomy for medically intractable Parkinson’s disease. Adv Neurol 1997;74:183-98.
80. Vitek JL, et al. Randomized trial of pallidotomy versus medical therapy for Parkinson’s disease. Ann Neurol 2003;53(5):558-69. 81. Vitek JL, et al. Microelectrode-guided pallidotomy: technical approach and its application in medically intractable Parkinson’s disease. J Neurosurg 1998;88(6): 1027-43. 82. Svennilson E, et al. Treatment of parkinsonism by stereotatic thermolesions in the pallidal region. A clinical evaluation of 81 cases. Acta Psychiatr Scand 1960; 35:358-77. 83. Hassler R, et al. Physiological observations in stereotaxic operations in extrapyramidal motor disturbances. Brain 1960;83:337-50. 84. Duvoisin RC. Problems in the treatment of Parkinsonism. Adv Exp Med Biol 1977;90:131-55. 85. Yahr MD, et al. Treatment of parkinsonism with levodopa. Arch Neurol 1969;21(4):343-54. 86. Laitinen LV. Leksell’s unpublished pallidotomies of 1958– 1962. Stereotact Funct Neurosurg 2000;74(1):1-10. 87. Albin RL, Young AB, Penney JB. The functional anatomy of basal ganglia disorders. Trends Neurosci 1989;12 (10):366-75. 88. DeLong MR. Primate models of movement disorders of basal ganglia origin. Trends Neurosci 1990;13(7):281-5. 89. Filion M, Tremblay L. Abnormal spontaneous activity of globus pallidus neurons in monkeys with MPTP-induced parkinsonism. Brain Res 1991;547(1):142-51. 90. Lozano AM, et al. Microelectrode recordings define the ventral posteromedial pallidotomy target. Stereotact Funct Neurosurg 1998;71(4):153-63. 91. Krauss JK, et al. Microelectrode-guided posteroventral pallidotomy for treatment of Parkinson’s disease: postoperative magnetic resonance imaging analysis. J Neurosurg 1997;87(3):358-67. 92. Guridi J, et al. Stereotactic targeting of the globus pallidus internus in Parkinson’s disease: imaging versus electrophysiological mapping. Neurosurgery 1999;45 (2):278-87; discussion 287-9. 93. Baron MS, et al. Treatment of advanced Parkinson’s disease by posterior gpi pallidotomy: 1-year results of a pilot study. Ann Neurol 1996;40(3):355-66. 94. Baron MS, et al. Treatment of advanced Parkinson’s disease by unilateral posterior GPi pallidotomy: 4-year results of a pilot study. Mov Disord 2000;15(2):230-7. 95. Lozano AM, Lang AE. Pallidotomy for Parkinson’s disease. Adv Neurol 2001;86:413-20. 96. Gross RE, et al. Relationship of lesion location to clinical outcome following microelectrode-guided pallidotomy for Parkinson’s disease. Brain 1999;122 (Pt 3):405-16. 97. Benabid AL, et al. Long-term suppression of tremor by chronic stimulation of the ventral intermediate thalamic nucleus. Lancet 1991;337(8738):403-6. 98. Benabid AL, et al. [Treatment of Parkinson tremor by chronic stimulation of the ventral intermediate nucleus of the thalamus]. Rev Neurol (Paris) 1989; 145(4):320-3.
Globus pallidus stimulation for parkinson’s disease
99. Benabid AL, et al. Combined (thalamotomy and stimulation) stereotactic surgery of the VIM thalamic nucleus for bilateral Parkinson disease. Appl Neurophysiol 1987;50(1‐6):344-6. 100. Lombardi WJ, et al. Relationship of lesion location to cognitive outcome following microelectrode-guided pallidotomy for Parkinson’s disease: support for the existence of cognitive circuits in the human pallidum. Brain 2000;123 (Pt 4):746-58. 101. Mogilner AY, Rezai AR. Brain stimulation: history, current clinical application, and future prospects. Acta Neurochir Suppl 2003;87:115-20. 102. Bechtereva NP, et al. Method of electrostimulation of the deep brain structures in treatment of some chronic diseases. Confin Neurol 1975;37(1‐3):136-40. 103. Benabid AL, et al. Acute and long-term effects of subthalamic nucleus stimulation in Parkinson’s disease. Stereotact Funct Neurosurg 1994;62(1‐4): 76-84. 104. Dostrovsky JO, Hutchison WD, Lozano AM. The globus pallidus, deep brain stimulation, and Parkinson’s disease. Neuroscientist 2002;8(3):284-90. 105. Loher TJ, et al. Long-term pallidal deep brain stimulation in patients with advanced Parkinson disease: 1-year follow-up study. J Neurosurg 2002;96(5):844-53. 106. Alberts JL, Okun MS, Vitek JL. The persistent effects of unilateral pallidal and subthalamic deep brain stimulation on force control in advanced Parkinson’s patients. Parkinsonism Relat Disord 2008;14(6):481-8. 107. Breit S, et al. Effects of 6-hydroxydopamine-induced severe or partial lesion of the nigrostriatal pathway on the neuronal activity of pallido-subthalamic network in the rat. Exp Neurol 2007;205(1):36-47. 108. Lenz FA, Vitek JL, DeLong MR. Role of the thalamus in parkinsonian tremor: evidence from studies in patients and primate models. Stereotact Funct Neurosurg 1993;60(1‐3):94-103. 109. Benazzouz A, et al. Implication of the subthalamic nucleus in the pathophysiology and pathogenesis of Parkinson’s disease. Cell Transplant 2000;9(2):215-21. 110. Hamani C, Lozano AM. Physiology and pathophysiology of Parkinson’s disease. Ann NY Acad Sci 2003; 991:15-21. 111. Wichmann T, DeLong MR. Pathophysiology of Parkinson’s disease: the MPTP primate model of the human disorder. Ann NY Acad Sci 2003;991:199-213. 112. DeLong MR. Functional and pathophysiological models of the basal ganglia: therapeutic implications. Rinsho Shinkeigaku 2000;40(12):1184. 113. DeLong MR, Crutcher MD, Georgopoulos AP. Primate globus pallidus and subthalamic nucleus: functional organization. J Neurophysiol 1985;53(2): 530-43. 114. Deogaonkar A, Piallat B, Subramanian T. Effect of L-dopa on neuronal activity of subthalamic nucleus (STN) and substantia nigra (SN) in hemiparkinsonian monkeys. 2005; S75-6.
95
115. Bergman H, et al. The primate subthalamic nucleus. II. Neuronal activity in the MPTP model of parkinsonism. J Neurophysiol 1994;72(2):507-20. 116. Samuel M, et al. Pallidotomy in Parkinson’s disease increases supplementary motor area and prefrontal activation during performance of volitional movements an H2(15)O PET study. Brain 1997;120 (Pt 8):1301-13. 117. Arahata Y, Kato T, Ito K. [Molecular imaging in Parkinson’s disease]. Nippon Rinsho 2007;65(2):327-31. 118. Antonini A, et al. Functional neuroimaging (PET and SPECT) in the selection and assessment of patients with Parkinson’s disease undergoing deep brain stimulation. J Neurosurg Sci 2003;47(1):40-6. 119. Ceballos-Baumann AO. Functional imaging in Parkinson’s disease: activation studies with PET, fmri and SPECT. J Neurol 2003;250 Suppl 1:I15-23. 120. Thobois S, Guillouet S, Broussolle E. Contributions of PET and SPECT to the understanding of the pathophysiology of Parkinson’s disease. Neurophysiol Clin 2001;31 (5):321-40. 121. Fukuda M, et al. Functional correlates of pallidal stimulation for Parkinson’s disease. Ann Neurol 2001;49 (2):155-64. 122. Hirato M, et al. Study on the function of the basal ganglia and frontal cortex using depth microrecording and PET scan in relation to the outcome of pallidotomy for the treatment of rigid-akinesia-type Parkinson’s disease. Stereotact Funct Neurosurg 1997;69(1‐4 Pt 2):86-92. 123. Eidelberg D, et al. Metabolic correlates of pallidal neuronal activity in Parkinson’s disease. Brain 1997;120 (Pt 8):1315-24. 124. Green J, et al. Neuropsychological and psychiatric sequelae of pallidotomy for PD: clinical trial findings. Neurology 2002;58(6):858-65. 125. Baron MS, Noonan JB, Mewes K. Restricted ablative lesions in motor portions of GPi in primates produce extensive loss of motor-related neurons and degeneration of the lenticular fasciculus. Exp Neurol 2006;202 (1):67-75. 126. Filion M. Effects of interruption of the nigrostriatal pathway and of dopaminergic agents on the spontaneous activity of globus pallidus neurons in the awake monkey. Brain Res 1979;178(2‐3):425-41. 127. Poirier LJ, et al. Physiopathology of experimental Parkinsonism in the monkey. Can J Neurol Sci 1975; 2(3):255-63. 128. Delong MR, et al. Functional organization of the basal ganglia: contributions of single-cell recording studies. Ciba Found Symp 1984;107:64-82. 129. Alexander GE, DeLong MR, Strick PL. Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annu Rev Neurosci 1986; 9:357-81. 130. Johnson AE, et al. Characterization of dopamine receptor binding sites in the subthalamic nucleus. Neuroreport 1994;5(14):1836-8.
1599
1600
95
Globus pallidus stimulation for parkinson’s disease
131. Dracheva S, Haroutunian V. Ocomotor behavior of dopamine D1 receptor transgenic/D2 receptor deficient hybrid mice. Brain Res 2001;905(1‐2):142-51. 132. Ito H, et al. Normal database of dopaminergic neurotransmission system in human brain measured by positron emission tomography. Neuroimage 2008; 39(2):555-65. 133. Kaasinen V, et al. Extrastriatal dopamine D2 and D3 receptors in early and advanced Parkinson’s disease. Neurology 2000;54(7):1482-7. 134. Kaasinen V, et al. Upregulation of putaminal dopamine D2 receptors in early Parkinson’s disease: a comparative PET study with [11C] raclopride and [11C]Nmethylspiperone. J Nucl Med 2000;41(1):65-70. 135. Rye DB. Parkinson’s disease and RLS: the dopaminergic bridge. Sleep Med 2004;5(3):317-28. 136. Miller WC, DeLong MR. Parkinsonian symptomatology. An anatomical and physiological analysis. Ann NY Acad Sci 1988;515:287-302. 137. Filion M, Tremblay L, Bedard PJ. Effects of dopamine agonists on the spontaneous activity of globus pallidus neurons in monkeys with MPTP-induced parkinsonism. Brain Res 1991;547(1):52-61. 138. Obeso JA, et al. Pathophysiology of levodopa-induced dyskinesias in Parkinson’s disease: problems with the current model. Ann Neurol 2000;47 4 Suppl 1:S22-32; discussion S32-4. 139. Obeso JA, et al. Pathophysiologic basis of surgery for Parkinson’s disease. Neurology 2000;55 (12 Suppl 6): S7-12. 140. Gatev P, Darbin O, Wichmann T. Oscillations in the basal ganglia under normal conditions and in movement disorders. Mov Disord 2006;21(10):1566-77. 141. Brown P, et al. Effects of stimulation of the subthalamic area on oscillatory pallidal activity in Parkinson’s disease. Exp Neurol 2004;188(2):480-90. 142. Maurer C, Mergner T, Peterka RJ. Abnormal resonance behavior of the postural control loop in Parkinson’s disease. Exp Brain Res 2004;157(3):369-76. 143. Brown, Williams D. Basal ganglia local field potential activity: character and functional significance in the human. Clin Neurophysiol 2005;116(11): 2510-9. 144. Hashimoto T, et al. Stimulation of the subthalamic nucleus changes the firing pattern of pallidal neurons. J Neurosci 2003;23(5):1916-23. 145. Vitek JL, et al. Physiology of hypokinetic and hyperkinetic movement disorders: model for dyskinesia. Ann Neurol 2000;47(4 Suppl 1):S131-40. 146. Guo Y, et al. Thalamocortical relay fidelity varies across subthalamic nucleus deep brain stimulation protocols in a data-driven computational model. J Neurophysiol 2008;99(3):1477-92. 147. Chan CS, Surmeier DJ, Yung WH. Striatal information signaling and integration in globus pallidus: timing matters. Neurosignals 2005;14(6):281-9.
148. Morris G, Crutcher MD. Physiological studies of information processing in the normal and Parkinsonian basal ganglia: pallidal activity in Go/No-Go task and following MPTP treatment. Prog Brain Res 2005; 147:285-93. 149. Rektor I, et al. Cognitive- and movement-related potentials recorded in the human basal ganglia. Mov Disord 2005;20(5):562-8. 150. Romanelli P, et al. Somatotopy in the basal ganglia: experimental and clinical evidence for segregated sensorimotor channels. Brain Res Brain Res Rev 2005; 48(1):112-28. 151. Alexander GE, Crutcher MD. Functional architecture of basal ganglia circuits: neural substrates of parallel processing. Trends Neurosci 1990;13(7):266-71. 152. Alexander GE, Crutcher MD, DeLong MR. Basal ganglia-thalamocortical circuits: parallel substrates for motor, oculomotor, ‘‘prefrontal’’ and ‘‘limbic’’ functions. Prog Brain Res 1990;85:119-46. 153. Parent A, Hazrati LN. Anatomical aspects of information processing in primate basal ganglia. Trends Neurosci 1993;16(3):111-6. 154. Parent A, Hazrati LN. Functional anatomy of the basal ganglia. I. The cortico-basal ganglia-thalamo-cortical loop. Brain Res Brain Res Rev 1995;20(1):91-127. 155. Parent A, et al. Organization of the basal ganglia: the importance of axonal collateralization. Trends Neurosci 2000;23(10 Suppl):S20-7. 156. Fukuda M, et al. Networks mediating the clinical effects of pallidal brain stimulation for Parkinson’s disease: a PET study of resting-state glucose metabolism. Brain 2001;124 (Pt 8):1601-9. 157. Davis KD, et al. Globus pallidus stimulation activates the cortical motor system during alleviation of parkinsonian symptoms. Nat Med 1997;3(6):71-4. 158. Ceballos-Baumann AO, et al. Restoration of thalamocortical activity after posteroventral pallidotomy in Parkinson’s disease. Lancet 1994;344(8925):814. 159. Bakay RA, et al. Posterior ventral pallidotomy: techniques and theoretical considerations. Clin Neurosurg 1997;44:197-20. 160. Lang AE, et al. Deep brain stimulation: preoperative issues. Mov Disord 2006;2(1 Suppl 14):S171-96. 161. Voon V, et al. Deep brain stimulation: neuropsychological and neuropsychiatric issues. Mov Disord 2006;21 Suppl 14:S305-27. 162. Follett KA. Comparison of pallidal and subthalamic deep brain stimulation for the treatment of levodopa-induced dyskinesias. Neurosurg Focus 2004; 17(1):E3. 163. Mavinkurve G, et al. Somatotopic organization of the globus pallidus internus of humans. Stereotact Funct Neurosurg 2007;85:32-33. 164. Vitek JL, et al. Acute stimulation in the external segment of the globus pallidus improves parkinsonian motor signs. Mov Disord 2004;19(8):907-15.
Globus pallidus stimulation for parkinson’s disease
165. Xu W, et al. STN DBS differentially modulates neuronal activity in the pallidal and cerebellar receiving areas of the motor thalamus. In: Movement Disorders Society Meeting, Chicago, IL, 2008. 166. Maciunas RJ, Galloway RL Jr, Latimer JW. The application accuracy of stereotactic frames. Neurosurgery 1994;35(4):682-94; discussion 694-5. 167. Holloway KL, et al. Frameless stereotaxy using bone fiducial markers for deep brain stimulation. J Neurosurg 2005;103(3):404-13. 168. Eljamel MS,Tulley M, Spillane K. A simple stereotactic method for frameless deep brain stimulation. Stereotact Funct Neurosurg 2007;85(1):6-10. 169. Deogaonkar MS. Accuracy of targeting via a frameless stereotactic targeting system for deep brain stimulation surgery. Neurosurgery 2007;61(1):214. 170. Henderson JM. Frameless localization for functional neurosurgical procedures: a preliminary accuracy study. Stereotact Funct Neurosurg 2004;82(4):135-41. 171. Schad LR, et al. Three dimensional image correlation of CT, MR, and PET studies in radiotherapy treatment planning of brain tumors. J Comput Assist Tomogr 1987;11(6):948-54. 172. Voges J, et al. Bilateral high-frequency stimulation in the subthalamic nucleus for the treatment of Parkinson disease: correlation of therapeutic effect with anatomical electrode position. J Neurosurg 2002;96(2):269-79. 173. Slavin KV, et al. Direct visualization of the human subthalamic nucleus with 3T MR imaging. AJNR Am J Neuroradiol 2006;27(1):80-4. 174. Liu X, et al. Localisation of the subthalamic nucleus using Radionics Image Fusion and Stereoplan combined with field potential recording. A technical note. Stereotact Funct Neurosurg 2001;76(2):63-73. 175. Aziz TZ, et al. Targeting the subthalamic nucleus. Stereotact Funct Neurosurg 2001;77(1‐4):87-90. 176. Schaltenbrand G, Wahren W. Atlas for stereotaxy of the human brain. Stuttgart: Thieme; 1977. 177. Schaltenbrand G, Bailey P. Introduction to stereotaxis with an atlas of the human brain. Stuttgart: Thieme; 1959. 178. Talaraich J, Tournoux P. Co-planar stereotactic atlas of the human brain. Stuttgart: Thieme; 1988. 179. Gross RE, et al. Electrophysiological mapping for the implantation of deep brain stimulators for Parkinson’s disease and tremor. Mov Disord 2006;21 Suppl 14: S259-83. 180. Hubel D. Tungsten microelectrode from recording from single units. Science 1957;125:549-50. 181. Lenz FA, et al. Methods for microstimulation and recording of single neurons and evoked potentials in the human central nervous system. J Neurosurg 1988;68(4): 630-4. 182. Guridi J, et al. Targeting the basal ganglia for deep brain stimulation in Parkinson’s disease. Neurology 2000;55 (12 Suppl 6):S21-8.
95
183. Alterman RL, et al. Microelectrode recording during posteroventral pallidotomy: impact on target selection and complications. Neurosurgery 1999;44(2):315-21; discussion 321-3. 184. Gross RE, et al. Variability in lesion location after microelectrode-guided pallidotomy for Parkinson’s disease: anatomical, physiological, and technical factors that determine lesion distribution. J Neurosurg 1999;90(3):468-77. 185. Pahwa R, et al. High-frequency stimulation of the globus pallidus for the treatment of Parkinson’s disease. Neurology 1997;49(1):249-53. 186. Kumar R. Methods for programming and patient management with deep brain stimulation of the globus pallidus for the treatment of advanced Parkinson’s disease and dystonia. Mov Disord 2002;17 Suppl 3: S198-207. 187. Volkmann J, Moro E, Pahwa R. Basic algorithms for the programming of deep brain stimulation in Parkinson’s disease. Mov Disord 2006;21 Suppl 14:S284-9. 188. Deep-brain stimulation of the subthalamic nucleus or the pars interna of the globus pallidus in Parkinson’s disease. N Engl J Med 2001;345(13):956-63. 189. Mazzone P. Deep brain stimulation in Parkinson’s disease: bilateral implantation of globus pallidus and subthalamic nucleus. J Neurosurg Sci 2003;47(1):47-51. 190. Krack P, et al. Five-year follow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinson’s disease. N Engl J Med 2003;349(20):1925-34. 191. Deuschl G, et al. Deep-brain stimulation for Parkinson’s disease. J Neurol 2002;249 Suppl 3:III/36-9. 192. Peppe A, et al. Stimulation of the subthalamic nucleus compared with the globus pallidus internus in patients with Parkinson disease. J Neurosurg 2004; 101(2):195-200. 193. Rodriguez-Oroz MC, et al. Bilateral deep brain stimulation in Parkinson’s disease: a multicentre study with 4 years follow-up. Brain 2005;128 (Pt 10):2240-9. 194. Nakamura K, et al. Effects of unilateral subthalamic and pallidal deep brain stimulation on fine motor functions in Parkinson’s disease. Mov Disord 2007;22(5): 619-26. 195. Kenney C, et al. Short-term and long-term safety of deep brain stimulation in the treatment of movement disorders. J Neurosurg 2007;106(4):621-5. 196. Binder DK, Rau GM, Starr PA. Risk factors for hemorrhage during microelectrode-guided deep brain stimulator implantation for movement disorders. Neurosurgery 2005;56(4):722-32; discussion 722-32. 197. Deogaonkar M, et al. Surgical complications in 800 consecutive DBS implants. J Neurosurg 2007;A774. 198. Voges J, et al. Thirty days complication rate following surgery performed for deep-brain-stimulation. Mov Disord 2007;22(10):1486-9. 199. Tir M, et al. Exhaustive, one-year follow-up of subthalamic nucleus deep brain stimulation in a large,
1601
1602
95 200.
201.
202.
203.
204.
205.
206.
207.
208.
209. 210.
211.
212.
Globus pallidus stimulation for parkinson’s disease
single-center cohort of parkinsonian patients. Neurosurgery 2007;61(2):297-304. Kupsch A, et al. Pallidal deep-brain stimulation in primary generalized or segmental dystonia. N Engl J Med 2006;355(19):1978-90. Novak KE, et al. Two cases of ischemia associated with subthalamic nucleus stimulator implantation for advanced Parkinson’s disease. Mov Disord 2006; 21(9):1477-83. Voges J, et al. Deep-brain stimulation: long-term analysis of complications caused by hardware and surgery– experiences from a single centre. J Neurol Neurosurg Psychiatry 2006;77(7):868-72. Blomstedt P. Hariz MI, Hardware-related complications of deep brain stimulation: a ten year experience. Acta Neurochir (Wien) 2005;147(10):1061-4; discussion 1064. Lyons KE, Hariz MI. Surgical and hardware complications of subthalamic stimulation: a series of 160 procedures. Neurology 2004;63(4):612-6. Umemura A, et al. Deep brain stimulation for movement disorders: morbidity and mortality in 109 patients. J Neurosurg 2003;98(4):779-84. Gill CE, et al. Deep brain stimulation for Parkinson’s disease: the Vanderbilt University Medical Center experience, 1998–2004. Tenn Med 2007;100(4):45-7. Vesper J, et al. Subthalamic nucleus deep brain stimulation in elderly patients – analysis of outcome and complications. BMC Neurol 2007;7:7. Liang GS, et al. Long-term outcomes of bilateral subthalamic nucleus stimulation in patients with advanced Parkinson’s disease. Stereotact Funct Neurosurg 2006; 84(5‐6):221-7. Deuschl G, et al. Deep brain stimulation: postoperative issues. Mov Disord 2006;21 Suppl 14:S219-37. Kleiner-Fisman G, et al. Subthalamic nucleus deep brain stimulation: summary and meta-analysis of outcomes. Mov Disord 2006;21 Suppl 14:S290-304. Blomstedt P, Hariz MI. Are complications less common in deep brain stimulation than in ablative procedures for movement disorders? Stereotact Funct Neurosurg 2006;84(2‐3):72-81. Amirnovin R, et al. Experience with microelectrode guided subthalamic nucleus deep brain stimulation. Neurosurgery 2006;58 Suppl 1:ONS96-102; discussion ONS96-102.
213. Gorgulho A, et al. Incidence of hemorrhage associated with electrophysiological studies performed using macroelectrodes and microelectrodes in functional neurosurgery. J Neurosurg 2005;102(5):888-96. 214. De Salles AA, et al. Functional neurosurgery in the MRI environment. Minim Invasive Neurosurg 2004; 47(5):284-9. 215. Binder DK, Rau G, Starr PA. Hemorrhagic complications of microelectrode-guided deep brain stimulation. Stereotact Funct Neurosurg 2003;80(1‐4):28-31. 216. Kondziolka D, et al. Hardware-related complications after placement of thalamic deep brain stimulator systems. Stereotact Funct Neurosurg 2002;79(3‐4):228-33. 217. Terao T, et al. [Comparison and examination of stereotactic surgical complications in movement disorders]. No Shinkei Geka 2003;31(6):629-36. 218. Coubes P, et al. Deep brain stimulation for dystonia. Surgical technique. Stereotact Funct Neurosurg 2002;78(3‐4):183-91. 219. Beric A, et al. Complications of deep brain stimulation surgery. Stereotact Funct Neurosurg 2001;77(1‐4):73-8. 220. Oh, MY, et al. Long-term hardware-related complications of deep brain stimulation. Neurosurgery 2002;50 (6):268-74; discussion 1274-6. 221. Constantoyannis C, et al. Reducing hardware-related complications of deep brain stimulation. Can J Neurol Sci 2005;32(2):94-200. 222. Hariz MI, Johansson F. Hardware failure in parkinsonian patients with chronic subthalamic nucleus stimulation is a medical emergency. Mov Disord 2001; 16(1):66-8. 223. Hariz MI. Complications of deep brain stimulation surgery. Mov Disord 2002;17 Suppl 3:S162-6. 224. Machado AG, et al. Fracture of subthalamic nucleus deep brain stimulation hardware as a result of compulsive manipulation: case report. Neurosurgery 2005;57 (6):E1318; discussion E1318. 225. Paluzzi A, et al. Operative and hardware complications of deep brain stimulation for movement disorders. Br J Neurosurg 2006;20(5):290-5. 226. Benabid AL, Chabardes S, Seigneuret E. Deep-brain stimulation in Parkinson’s disease: long-term efficacy and safety – What happened this year? Curr Opin Neurol 2005;18(6):23-30. 227. Schaltenbrand G, Warren W. Atlas for Stereotaxy of the Human Brain. New York: Thieme; 1977.
114 History and Current Neurosurgical Management of Spasticity R. D. Penn
The neurosurgical treatment of spasticity has a long history. The entire neuro-axis has been subject to lesions, sections, stimulation, and drug infusions in attempts to control spastic symptoms. The first operations were on peripheral nerves such as the section of obturator nerve by Lorenz in 1897. These types of peripheral nerve operations continue to be useful to this day, and have been made more effective by employing moderate EMG technologies (see Chap. G-31). In essence, botulinum toxic injections, the mainstay of the treatment of focal spasticity, is an extension of selective nerve injuries. Another early approach was Horsley’s use of cortical topectomy in 1895 to control hyperkinetic movements. He noted that spasticity also decreases. This was, however, a short lived effect and was accompanied by motor weakness. Likewise, other brain lesions such as ventral lateral thalamotomy, pulvinarotomy, and dentatotomy have been tried and abandoned because of poor or inconsistent results and side effects. Spinal interventions have proved much more successful. At the beginning of the twentieth century, Foerster introduced dorsal rhizotomies to reduce spasticity. His work was based on Sherrington’s newly discovered spinal reflex arc. The results he published on patients who underwent the procedure were impressive [1] (see > Figure 114-1). Use of this destructive approach waned over the years because of the magnitude of the operation and its complications which can include sensory loss and weakness. Currently, a modified version of the Foerster approach has resulted in a resurgence of interest in the treatment of spasticity of cerebral palsy [2,3]. In these #
Springer-Verlag Berlin/Heidelberg 2009
operations, electrical stimulation of the dorsal roots has been used not only to identify root levels but also to determine which roots should be sectioned because they are responsible for the most spasticity. Lesions in the dorsal root entry zone are also effective and they are reviewed in Chap. G-32. Other spinal lesions such as Bischof ’s lateral longitudinal myelotomy have been tried but are no longer in use because of side effects, in particular bladder dysfunction. Nondestructive neuro-modulation of spasticity by electrical stimulation has been advocated. In particular stimulation of the dorsal spinal cord or cerebellar surface has been performed using standard implanted spinal cord stimulating systems. In spite of initial optimistic reports, these operations have been discontinued because later controlled studies have not supported the original claims. The newest approach has been the delivery of medications to the spinal canal by an implanted drug pump. The first experiments used epidural morphine (Struppler et al. 1983). Morphine is known to decrease polysynaptic nociceptive reflexes in the spinal cord and this clearly reduces spasticity. In 1984, intrathecal baclofen was demonstrated to specifically reduce spasticity with little effect on pain [4]. If one uses an implanted drug pump and catheter to deliver baclofen continuously, the therapeutic effect can be maintained [5]. Chronic spinal intrathecal baclofen has been widely adopted. It is used to treat many types of spasticity of both spinal and cerebral origin (see Chap. G-33).
1926
114
History and current neurosurgical management of spasticity
. Figure 114-1 Photographs from [1] showing a cerebral palsy patient before a dorsal rhizotomy (left) and after (right)
Undoubtedly, new surgical approaches will be developed, and the current successful ones will be perfected. However, how much can we expect from these techniques? Will reducing spasticity improve motor control and improve the activities of daily living? To answer these important clinical questions, the nature of spasticity must be considered in detail. The remainder of this chapter reviews the physiological mechanisms that underlie spasticity and emphasizes a broader definition of the phenomenon then is usually proposed. The argument is made that if spasticity is too narrowly defined, it loses its relevance to movement disorders and that the improvement in motor control and activities of daily living that results from surgical procedures can be understood only in a broader context. Most discussions of spasticity begin with a narrow definition that emphasizes abnormalities in the stretch reflex. Lance is widely quoted and
his view of spasticity has been very influential. Lance defines spasticity as “a motor disorder characterized by a velocity dependent increase in tonic stretch reflexes (muscle tone) with exaggerated tendon jerks resulting from hyperexcitability of the stretch reflex, as one component of the upper motor neuron syndrome” [6]. This concise statement is useful in a number of ways. It centers attention on the apparent pathophysiological mechanism, it provides a simple test to identify the condition, and it makes it part of a larger problem in motor function, namely the upper motor neuron syndrome. The problem with this limited definition is it does not correspond to the general sense in which the word spasticity is understood as denoting the spasms themselves or the abnormal movements in which spasmodic disruption occurs. The dictionary equates spasticity with spasmodic movements and this is the way in which the word is employed
History and current neurosurgical management of spasticity
in ordinary conversation. When we say a person is “spastic” we mean the person is “clumsy” and “uncoordinated” and makes irregular unsmooth movements. The gulf between Lance’s physiological definition and the common notion of spasticity is not just a case of an expert explaining the underlying basis of the general phenomenon. The problem is that Lance’s definition does not in any straight forward way relate to the disorder of movement what we want to understand. What does the hyperactive stretch reflex have to do with not being able to make proper movement? In their excellent review of spinal mechanisms underlying spasticity, Pierrot-Deseilligny and Nazieres recognize this problem [7]. They stated that “contributions of spasticity (as defined by Lance) to motor disabilities of patients with upper motor neuron lesions is probably smaller than previously thought.” The link between improper execution of movements and the abnormal stretch reflexes is weak. This narrow definition of spasticity led Landau to argue that treating spasticity was unlikely to improve a patient’s motor function [8]. Landau drew attention to the positive signs of spasticity noted by the neurologist (hyperactive tendon reflexes and increased muscle tone) and the negative symptoms of weakness, slowness and incoordination which are the patient’s main complaint. The positive signs are viewed as a release phenomenon caused by injury to the motor system. Reducing them obviously cannot repair the damaged motor system and movement production will still be poor. Landau considered new treatments for spasticity to be useless and stated that only a “method of provoking CNS neural regeneration of or improving the potency and control of surviving upper level neurons could provide a direct remedy” [8]. Fortunately, such a nihilistic outlook is not justified. Clinical observations from Foerster to the present have demonstrated that reducing the signs of spasticity often correlates with better movement. For example, Foerster reported that
114
a nonambulatory spastic patient with cerebral palsy was so improved by surgery that he was able to take a train by himself to go to a medical conference for a demonstration. He also had pictures in his original article showing another patient who was able to walk after surgery [1] see > Figure 114-1). Such gains are obviously not possible for patients with profound motor loss resulting from injury. However, even with complete cord transsection, reduction of spasticity may allow a patient to improve overall motor function by facilitating transfers, eliminating spasms, and using a motorized chair. The problem with Lance’s definition and Landau’s arguments is that both ignore the associated physiological abnormalities that occur with the stretch reflex hyperactivity. These associated changes which primarily involve interneuron pathways in the spinal cord make movements difficult. When movement is seen as resulting from the coordinated activation of many muscles, disruptions of these synergies by spasticity can be better understood. Wiesendanger has provided a much more comprehensive definition of spasticity that has clear relevance to voluntary movements: “Spasticity is a movement disorder that develops gradually in response to a partial or complete loss of supraspinal control of spinal cord function. It is characterized by altered activity patterns of motor units occurring in response to sensory and central command signals which lead to co-contraction, mass movements and abnormal postural control” [9]. This definition emphasizes a number of important global aspects of the disordered movement. Central to this concept is the slow development of spasticity after an injury. This distinguishes spasticity from decorticate or decerebrate rigidity which occurs abruptly after an injury to the brain or brainstem and is best interpreted as a release phenomenon. The increased muscle tone and stretch reflexes and dystonic posturing of the limbs in these conditions might be confused with true spasticity, but these conditions have different physiological mechanisms.
1927
1928
114
History and current neurosurgical management of spasticity
Why place so much emphasis on the slow evolution of symptoms? The reason for this is that spasticity is due to a plastic change in the spinal cord circuitry. After a severe spinal cord injury, the patient’s limbs are flaccid for days to weeks. Gradually, muscle tone increases and stretch reflexes develop, and only after months do hypertonia and hyperreflexia emerge and clinically significant spasms appear. Since spasticity is generated at the spinal level, a slow change in the cord produces this progression. What these changes are has not been demonstrated in humans. The most likely explanation is that deafferentation of spinal neurons from their well-formed adult supraspinal input produces collateral sprouting from neighboring axons or denervation hypersensitivity develops (> Figure 114-2). In either case, the spasticity is not due to the simple release of excitatory influence, but instead to morphological or pharmacological events in the cord. The pathophysiological importance of the clinical observation that spasticity develops slowly or that plastic changes occur in the spinal cord is certainly not new.
In 1958, McCouch et al. [10] suggested sprouting as a cause of spasticity and Polistina et al. [11] provided experimental anatomic data for plastic changes in the spinal cord synapses after deafferentation. However, the evidence that this happens in patients is lacking and would be difficult to obtain. This means that animal models are needed, but unfortunately, good chronic models of spasticity have not been developed. Decerebrate rigidity, which superficially resembles spasticity, has been often taken as a model. The fact that tone can be reduced with dorsal rhizotomies in both the decerebrate rigidity animals and spasticity in humans makes the resemblance appear close. However, the time course is wrong and the reflex pathways are different. Some progress in creating more realistic models is being made but the physiological changes have not been well characterized [12]. In addition to drawing attention to the slow development of spasticity, Wiesendanger’s definition emphasizes the prominent clinical features which are likely to impair motor performance; co-contraction, mass movements and abnormal
. Figure 114-2 The phenomenon of collateral sprouting. Left. Presynaptic fiber terminals degenerate as a consequence of lesion B, and synaptic sites are vacated. Middle. The intact fiber terminals from source A form new preterminal collaterals that occupy the vacated synaptic sites. Right. The phenomenon of denervation supersensitivity: Partial degeneration of presynaptic terminals leads to an increased postsynaptic receptor sensitivity of intact synapses to the released transmitter (reprinted with permission of the publisher from [9])
History and current neurosurgical management of spasticity
truncal control. Electromyographic (EMG) recordings made in patients with spasticity illustrate these phenomena. Consider a typical spastic patient trying to dorsiflex the ankle (> Figure 114-3). The result of this effort is inappropriate cocontraction of the soleus muscle as well as spread to distant muscles, the hamstring and quadriceps. Furthermore, this attempted movement results in extensor or flexor spasms in one or both legs interrupting any movement being made at the time. In addition to these problems, the patient cannot adjust muscle tone for proper postural control of the limb or trunk position, thus, maintaining a stable sitting position is not possible.
114
The dyssynergic activation of muscle groups is at the core of the movement disorder caused by spasticity. Smooth muscle performance becomes spastic in the common sense of the word because spinal mechanisms and integrated control have been disrupted and new, inappropriate reflex patterns have emerged. Some of these are primitive reflex patterns that were present during development. Others may be newly formed. In any case, activation of these abnormal circuits by supraspinal input or sensory input at the spinal cord level produces poorly coordinated jerking movements and frequently extensor or flexor spasms.
. Figure 114-3 EMG recordings of a spastic patient trying to dorsiflex his ankle. Note in (a) the cocontraction of the soleus muscle and the spread of responses to the hamstring and quadriceps. (b) After intrathecal baclofen. Note that the response in the soleus muscle is now absent and that there is much less spread of reflexes (reprinted with permission of the publisher from Latash ML, Penn RD, Corcos DM, Gottlieb GL. Effects of intrathecal baclofen on voluntary motor control in spastic paresis. J Neurosurg 1990;72:388-92)
1929
1930
114
History and current neurosurgical management of spasticity
It should be pointed out that hyperactive reflexes alone can sometimes interfere with smooth voluntary movement [13]. The induction of clonus is an obvious example. In > Figure 114-4, a tracing of the EMGs and kinetic records of ankle position show how a hyperactive stretch reflex disrupts ankle dorsiflexion. On the normal side, the movement elicits only a small response in the soleus and the ankle goes to the target without difficulty. On the spastic side, a large reflex is seen in the soleus and movement stops short of the target. However, most of the time the degree of reflex hyperactivity does not correlate with the degree of motor impairment [14]. While the mechanisms underlying the plastic changes in the spinal cord are unknown, the specific abnormal circuitry that results has been studied intensively. For details, a number of reviews can be consulted [7,14–18]. The exaggerated tendon jerk and the velocity dependent increase in tone and stretch reflexes have been the starting place for most neurophysiological studies of spasticity. Why do they occur? The final motor pathway is the alpha motor neuron, and so any abnormal reflex is reflected in the firing abnormalities of these neurons. One hypothesis that could explain the exaggerated output to sensory stimulation seen in spasticity is hyperexcitability of the alpha motor neuron. This could be due to intrinsic changes in the membrane characteristics or supraspinal excitatory input. The latter occurs in cats made decerebrate by the anemic method. In humans, no independent method has been devised to measure the supraspinal or membrane properties of the motor neurons and so this hypothesis is unproven. A more testable hypothesis has been that fusimotor hyperactivity leads to an increased sensitivity of the muscle spindles’ primary endings to stretch so that an exaggerated reflex is produced. In the decerebrate cat, this mechanism is present and helps explain rigidity in the limbs. The most conclusive test of this hypothesis in spastic humans
. Figure 114-4 Recordings of agonist (TA EMG) and antagonist (SOL EMG) EMG and angle and velocity data for an 18degree ankle dorsiflexion of the spastic limb to an 8-degree target (solid lines) and a 24-degree dorsiflexion of the healthy limb to a 4-degree target (dotted lines). The marking on the right side of the angle trace shows the distance moved and the site of the target (i.e., plus or minus half the target size). The important point here is that the spastic limb reverses direction before reaching even the near boundary of the target, whereas the healthy limb moves past the target before reversing direction. The reversal in direction that occurs in the healthy limb is a natural oscillation that often is associated with rapid movements (reprinted with permission of the publisher from [13])
History and current neurosurgical management of spasticity
has been done by Hagbarth [19] who recorded directly from nerves using microelectrodes to demonstrate the firing of small afferents. Compared with normals, no increase in gamma motor neuron output to the muscle spindle was seen in spastic patients. Several other less direct tests of muscle spindle excitability have also demonstrated normal levels of sensitivity [7]. Accumulated evidence against fusion motor hyperactivity is strong enough to eliminate this hypothesis from serious consideration. The lack of gamma hyperactivity definitely differentiates human spasticity from the animal model of decerebrate rigidity. Another potential way in which the output of the alpha motor neurons could be exaggerated is through reduction of recurrent inhibition via Renshaw cells. These inhibitory interneurons are strategically placed to provide negative feedback on motor neuron output (see > Figure 114-5). The level of activity, if decreased, will increase motor neuron firing rates by means of disinhibition. Electrophysiological evidence in spastic patients making voluntary movements demonstrates this type of disinhibition and may account for clonus. In contrast, at rest, with purely passive provoked response reflexes Renshaw activity is not inhibited. Another location where disinhibition might occur is at the presynaptic terminals of the 1A monosynaptic fibers going to motor neurons. Controversy exists about whether such disinhibition is found in patients, but experiments by Mailis and Ashby suggest that it does not [20]. What other pathways could be important? 1A fibers not only make direct synapses on alpha motor neurons but also have polysynaptic connections (> Figure 114-6). These interneurons are in turn modulated by descending input. Facilitation of this polysynaptic pathway could increase motor neuron excitation. Experiments using vibration induced stimulation of these pathways suggests in some spastic patients such hyperexcitability exists [7]. Another important abnormality is loss of reciprocal inhibition
114
. Figure 114-5 Diagram of the reciprocal spinal circuits that control reflex activity. Note how Ia fibers excite anterior tibial motoneurons and, via Ia interneurons, inhibit soleus motoneurons. These pathways are affected by descending input (reprinted with permission of the publisher from [7])
(> Figure 114-5). Normally, excitation of antagonistic motor neuron pool is linked by inhibitory interneurons to the antagonistic muscles’ motor neurons. As was previously illustrated, one of the characteristics of poorly made movement in a spastic patient is co-contraction of antagonistic muscle groups. Thus, a loss of facilitation of this interneuron pathway is likely to be present in spasticity. This brief summary of spinal cord circuitry in spastic patients is a simplification of a large group of experiments made by various investigations on spastic patients produced by spinal as well as cerebral lesions. Many controversies exist and disparate views continuing to be expressed. However, there is a consensus on a number of points. First, several spinal cord circuitry
1931
1932
114
History and current neurosurgical management of spasticity
. Figure 114-6 Diagram showing the descending fibers that control polysynaptic pathways in the spinal cord going to the alpha motor neurons. Their inhibitory effects may be reduced in spastic patients (reprinted with permission of the publisher from Pierrot-Deseilligny E. Pathophysiology of spasticity. Triangle 1983;22:165-74)
cannot be considered in isolation and therapy for spasticity must deal with abnormalities in the spinal cord circuitry and the abnormal compliance of the limbs [14].
References
problems are likely to be important and the degree to which they are involved may vary from patient to patient. Second, decerebrate rigidity is no longer considered a good model of spasticity. Third, gamma loop hyperexcitability does not occur in spastic patients. Fourth, emphasis needs to be placed on the slow emergence of spasticity and the way in which resulting abnormal spinal circuits interfere with voluntary movement. A few points should be added about changes in muscles and joints that may complicate the assessment of spasticity. As clinicians who work with spastic patients know, contractures caused by ligament shortening or joint ankylosis often occur [21]. Furthermore, as muscle function changes, fiber types change and result in different stiffness characteristics [22]. This means a patient with chronic spasticity develops a variety of mechanical abnormalities that can interfere with movement. The central nervous system
1. Foerster O. On the indications and results of the excision of posterior spinal nerve roots in men. Surg Gynecol Obster 1913;16:463-74. 2. Fasano VA, Barolat-Romana G, et al. Electrophysiological assessment of spinal circuits in spasticity by direct dorsal root stimulation. Neurosurgery 1979;4(2):146-51. 3. Engsberg JR, Ross SA, et al. Effect of selective dorsal rhizotomy in the treatment of children with cerebral palsy. J Neurosurg 2006;105 Suppl 1:8-15. 4. Penn RD, Kroin JS. Intrathecal baclofen alleviates spinal cord spasticity. Lancet 1984;1(8385):1078. 5. Penn RD, and Kroin JS. Continuous intrathecal baclofen for severe spasticity. Lancet 1985;2(8447):125-7. 6. Lance J. Symposium synopsis. In: Young R, Feldman RG, Kaella WP, editors. Spasticity: disordered motor control. Chicago: Yearbook, 1980. p. 485-94. 7. Pierrot-Deseilligny. Spinal mechanisms underlying spasticity. In: Young R, Delwside PJ, editors. Restorative neurology. Amsterdam: Elsevier; 1985. p. 63–76. 8. Landau W. The fable of the neurological demon and the emperor’s new therapy. Arch Neurol 1974;31:217-19. 9. Wiesendager M. Neurophysiological bases of spasticity. In: Abbott R, Sindou M, Keravel Y, editors. Neurosurgery for spasticity: a multidisciplinary approach. New York: Springer-Verlag; 1991. p. 15-19. 10. McCouch G, Austin GM, et al. Sprouting as a cause of spasticity. J Neurophysiol 1958;21(3):205-16. 11. Polistina DC, Murray M, et al. Plasticity of dorsal root and descending serotoninergic projections after partial deafferentation of the adult rat spinal cord. J Comp Neurol 1990;299(3):349-63. 12. Carter RL, Ritz LA, et al. Correlative electrophysiological and behavioral evaluation following L5 lesions in the cat: a model of spasticity. Exp Neurol 1991; 114(2):206-15. 13. Corcos DM, Gottlieb GL, et al. Movement deficits caused by hyperexcitable stretch reflexes in spastic humans. Brain 1986;109 (Pt 5):1043-58. 14. Dietz V, Sinkjaer T. Spastic movement disorder: impaired reflex function and altered muscle mechanics. Lancet Neurol 2007;6(8):725-33. 15. Penn RD, Corcos DM. Spasticity and its management. In: Youmans J, editor. Neurological surgery. Philadelphia: Saunders; 1990. p. 4371-85. 16. Ashby PMD. Neurophysiology of spinal spasticity. In: Davidoff RA, editor. Handbook of the spinal cord. New York: Marcel Decker; 1987.
History and current neurosurgical management of spasticity
17. Young R, Delwaide PJ. Drug therapy: spasticity. N Engl J Med 1981;304:28-33. 18. Young RR, Delwaide PJ. Drug therapy: spasticity (second of two parts). N Engl J Med 1981;304(2):96-9. 19. Hagbarth KE, Wallin G, et al. Muscle spindle responses to stretch in normal and spastic subjects. Scand J Rehabil Med 1973;5(4):156-9. 20. Mailis A, Ashby P. Alterations in group Ia projections to motoneurons following spinal lesions in humans. J Neurophysiol 1990;64(2):637-47. 21. Lowenthal M, Tobis JS. Contractures in chronic neurologic disease. Arch Phys Med Rehabil 1957;38(10):640-5.
114
22. Young JL, and Mayer RF. Physiological alterations of motor units in hemiplegia. J Neurol Sci 1982;54 (3):401-12. 23. Latash ML, Penn RD, Corcos DM, Gottlieb GL. Effects of intrathecal baclofen on voluntary motor control in spastic paresis. J Neurosurg 1990;72:388-392. 24. Struppler A, Burggmayer B, OCHS GB, Pleiffer HG. The effect of epidural application of opioids on spasticity of spinal origin. Life Sci. 1983;33(suppl.): 607-610. 25. Pierrot-Deseilligny E, Pathiophysiology of spasticity Tringle. 1983;22:165-74.
1933
87 History of Surgery for Movement Disorders A. G. Parrent
Cortical Ablation Since the beginnings of neurosurgery, operations have been done throughout the central nervous system to treat movement disorders. Horsley felt that athetosis resulted from abnormal cortical discharge and proposed excision of the area of motor cortex somatotopically related to the affected limb [1]. In 1909, he completely relieved hemi-athetosis in a 15-year-old boy by removing the precentral gyrus for the upper extremity at the expense of enduring dyspraxia and hemiparesis [2]. In 1931, Bucy removed the right precentral arm area in a patient to alleviate seizures, but postoperatively and in long-term follow-up, the patient’s accompanying left choreoathetosis was completely abolished [3]. Bucy believed that athetoid movements were mediated by descending extrapyramidal fibers arising largely from area 6 of the frontal lobe cortex. In his earlier cortical removals, area 4 was spared along the anterior aspect of the precentral sulcus, since this area gave rise to a large portion of the pyramidal tract. However, when he observed partial return of involuntary movements in some of his patients, Bucy decided to remove area 4 as well as area 6 to include the portion of the extrapyramidal system arising from area 4. This involved removing the entire precentral gyrus down to the bottom of the central sulcus plus the posterior portion of the third frontal convolution. Subsequent reports suggested that this resulted in more complete abolition of the athetoid movements than that achieved with lesser ablations [4–6]. Similar results for cortical ablation for athetosis were reported later [7–10]. #
Springer-Verlag Berlin/Heidelberg 2009
It was well known that the resting tremor of Parkinson’s disease would disappear after a cerebral vascular accident that rendered the person hemiplegic [11,12]. In such patients, the tremor would frequently reappear as recovery ensued. It had also been shown that removal of the precentral area could abolish experimental cerebellar tremor in the monkey [13]. Based on this information implicating the corticospinal system in the generation or at least the mediation of tremor, in 1937 Bucy and Case removed the precentral and adjacent premotor area in a patient with incapacitating tremor after a head injury [6]. This produced complete tremor abolition and, initially, contralateral hemiplegia, but motor function improved substantially with time on the affected side, leaving enduring impairment mainly of dexterous hand and finger movements. After seeing this patient through a 2-year follow-up without tremor recurrence, Bucy used the same procedure in a patient with parkinsonian tremor, producing permanent alleviation of the tremor [4]. Klemme [10] carried out premotor corticectomies in patients with parkinsonian tremor and reported his results together with those for athetosis; there was a 77% incidence of significant tremor reduction or elimination without an additional neurological deficit and a mortality rate of 17%. Unfortunately, the report did not include details of the surgical technique and follow-up. Putnam [14] reported a patient with parkinsonian tremor and one with posttraumatic tremor in whom he removed an area of the precentral gyrus. Inonepatienttremorwasabolished,andintheother it was significantly reduced, in both cases at the expense of fine motor control on the affected side.
1468
87
History of surgery for movement disorders
In a subsequent report, Bucy [17] noted that cortical removal limited to the precentral gyrus (area 4) was sufficient to abolish tremor but that the removal needed to encompass the posterior portion of the adjacent frontal gyri (areas 4 and 6) to abolish athetosis. Immediately after resection of the motor cortex, patients demonstrated marked contralateral weakness, but there was recovery of a ‘‘remarkable degree of voluntary control and power in the involved extremity if it had not been previously paralyzed’’ [4]. Such ablations induced hyperreflexia, clonus, dyspraxia, and impairment of distal fine limb movements, replacing the patient’s preexisting disorder with a more acceptable deficit. Bucy’s indications for the procedure were patients with the affected limb incapacitated and relatively useless who were willing to accept the operative 10% mortality risk [15,16]. Given the inevitable development of some degree of permanentparesisaftertheoperation,hefeltthatthe procedure should be limited to patients with unilateral or markedly asymmetrical choreoathetosis and patients with tremor in whom the resulting hemiparesis would result in a lesser disability than the tremor. He stated: ‘‘Only occasionally does one see a unilateral, nonprogressive parkinsonism in a relatively young patient in whom the tremor is the greatest disability and for whom the tremor is so severe that both he and his physician consider it wise for him to exchange it for a partial paralysis of the affected extremities’’ [17].
Cordotomy Experience with cortical ablation led to an approach to the pyramidal tract at other levels. In 1931, Putnam [18,19] started cutting the descendingextrapyramidal(vestibulospinal,reticulospinal, tectospinal) tracts in the high cervical anterior quadrant of the cord (anterolateral cordotomy (> Figure 87‐1a)). There was substantial relief in patients with choreoathetosis but not in five
with parkinsonian tremor. Oldberg’s experience was similar [19]. Putnam reasoned that since cortical ablation of pyramidal and extrapyramidal areas could alleviate tremor while sectioning of the spinal extrapyramidal tract alone had no effect on tremor, impulses descending in the spinal pyramidal tract might be responsible for the production of tremor [20]. Rothmann [21] had previously demonstrated that sectioning the lateral pyramidal tract in the monkey produced a minimal deficit, and so it seemed reasonable that the same procedure might be tolerated in humans. In 1940, Putnam reported the results of lateral corticospinal tract interruption (lateral pyramidotomy (> Figure 87‐1b)) in the high cervical region in seven patients [14] In early follow-up, tremor and rigidity were reduced but not abolished, and less impairment of motor function was seen than after extirpation of the motor cortex [8]; there was about a 30–80% reduction in power. . Figure 87‐1 Diagrams of the various high cervical cordotomy procedures used to treat movement disorders: (a) Putnam’s anterolateral cordotomy for choreoathetosis; (b) Putnam’s lateral pyramidotomy for parkinsonian rigidity and tremor; (c) Ebin’s combined lateral and ventral pyramidotomy; (d) Oliver’s modification of Putnam’s pyramidotomy, the equivalent of a lateral funiculotomy
History of surgery for movement disorders
By 1950, Putnam had operated on 22 patients, 15 followed for more than 1 year, more than half with satisfactory reduction of tremor, the rest unimproved [22]; none of these patients died. Ebin [23] thought that the residual tremor after Putnam’s lateral pyramidotomy might be eliminated if pyramidal fibers traveling with the ventral corticospinal tract were sectioned along with the lateral corticospinal tract (> Figure 87‐1c). He reported nine unilateral procedures for parkinsonian tremor with complete elimination of resting tremor in four and virtual elimination in five. In two additional patients in whom rigidity was the major problem, there was substantial improvement. In most cases, power returned to 35–40% of the preoperative baseline. Ebin also showed, in a single case, that the procedure could be carried out bilaterally and still allow substantial return of voluntary motor control. Oliver [24,25] reported his experience with Putnam’s procedure in 26 parkinsonian patients, of whom 16 enjoyed significant reduction of tremor. Since tremor was almost never completely eliminated by this procedure, he modified Putnam’s lateral pyramidotomy by increasing the depth and extent of the section in subsequent patients, many of whom underwent staged bilateral procedures. In half, tremor was reduced and motor performance was improved though the benefits faded in many patients over subsequent weeks to months. Six patients were worse, and three died. Oliver then cut the entire lateral funiculus in 14 patients (> Figure 87‐1d). Hemiplegia was complete immediately after the procedure but tended to improve over a few months, with the patients regaining useful levels of function, especially in the lower limbs, but with persisting contralateral analgesia and reduction of sexual function.
Pedunculotomy In seeking a site at which corticospinal tract fibers might be interrupted with minimal danger
87
of injuring other structures, Walker [26] sectioned the lateral two-thirds to three-fifths of the cerebral peduncle in a patient with hemiballismus (> Figure 87‐2a). He achieved excellent control of the movement disorder, and although there was reduced strength in the affected arm and hand, fine finger movements could be dexterously performed 1 year after the procedure. Walker later used the same procedure to treat parkinsonian tremor, reporting the results in four patients followed for almost 3 years [27]. Two patients with complete relief of tremor exhibited a greater degree of hemiparesis than did the other two patients, who showed slight residual tremor but mild weakness. Thus, the best result represented a compromise. Guiot and Pecker [28] performed a shallower incision (3 mm) in approximately the same area of the cerebral peduncle as Walker for parkinsonian tremor (> Figure 87‐2b). Postoperative motor function was excellent, but tremor tended to recur. Division of the lateral [29], or medial segment [30] of the cerebral peduncle was shown to be ineffective in treating movement disorders of the ballistic, athetotic, or parkinsonian type.
. Figure 87‐2 Diagrammatic cross section of the midbrain just below the emergence of the oculomotor nerves. (a) Walker’s lateral pedunculotomy; (b) Guiot and Pecker’s section for hemiparkinsonian tremor; (c) Meyers’ intermediate midbrain crusotomy for athetosis
1469
1470
87
History of surgery for movement disorders
Meyers, who had been attempting to find better surgical treatment for patients with severe athetotic and dystonic movement disorders, sequentially sectioned the medial and then the lateral segments of the cerebral peduncle in one of his patients [31]. In spite of the division of two-thirds to three-fifths of that structure, there was no improvement in the movement disorder, but his patient did not lose any of her preoperative motor function. Meyers surmised that it might be possible to divide the intermediate portion of the cerebral peduncle bilaterally (intermediate midbrain crusotomy (> Figure 87‐2c)) without severe impairment of motor function [31,32]. In his series of 48 patients [33], there was marked improvement of athetotic movements in 28, moderate improvement in 12, and slight improvement in 4. Athetotic movements were never completely relieved, and four patients died. Bucy [34] showed that the same procedure was effective in treating hemiballism. In an autopsy study performed 2 years after the procedure, he demonstrated the loss of 83% of the axons in the ipsilateral medullary pyramid and degeneration of the uncrossed ventral corticospinal tract and the crossed lateral corticospinal tract, along with loss of 90% of Betz’s cells in the precentral cortex. Thus, there was pathological evidence of virtually complete unilateral destruction of the pyramidal tract in an individual who exhibited only a very mild contralateral hemiparesis with only slight impairment of fine finger and fine toe movements. Bucy’s findings called into question the role of the ‘‘pyramidal’’ tract in the voluntary control of movement.
Open Subcortical Procedures Meyers carried out the first direct operation on the basal ganglia in an attempt to treat parkinsonian tremor [35]. Such surgery had up to that time been considered contraindicated because Dandy in 1930 [36] asserted that the ‘‘center of
consciousness’’ lay within the part of the brain supplied by the anterior cerebral artery. Dandy [37] later assigned that role to the anterior part of the striatum. Discussing Meyers’s first report [35] of excising caudate nucleus in three patients, Joseph King noted that Jefferson Browder had performed a frontal lobectomy down to the caudate nucleus in a patient with Parkinson’s disease, noting afterward that tremor had ceased and the parkinsonian facies had receded. Meyers performed his operation under local anesthetic for clinical monitoring, using a transcortical, transventricular approach [35,38,39], with progressive removal of portions of the basal ganglia and sequential examination. Excisions included (1) the caudate head, (2) the caudate head with section of the anterior limb of the internal capsule, (3) the caudate head and the oral half of the putamen, with section of the anterior limb of the internal capsule, (4) the caudate head, oral half of the putamen, and oral pole of globus pallidus, with section of the anterior limb of internal capsule, and (5) section of the ansa lenticularis (ansotomy) either alone or in combination with the above procedures. His first 58 operations, carried out between 1939 and 1949 [22,33] are shown in > Table 87‐1, along with the results. Of these procedures, Meyers found that sectioning of the pallidofugal efferent pathways (ansotomy) produced the best results for parkinsonian tremor, with removal of the caudate head and section of the anterior limb of the internal capsule a close second. Isolated removal of the caudate head and body was insufficient to alleviate tremor. Between 1949 and 1954, Meyers performed 55 additional procedures, most including section of the ansa lenticularis. Two-fifths of the patients were ‘‘much improved,’’ one-fifth were ‘‘improved,’’ and operative mortality dropped to 10%. Browder [40,41] carried out removal of the caudate head and section of the anterior limb of the internal capsule in 15 parkinsonian patients. Tremor was abolished or significantly reduced in
87
History of surgery for movement disorders
. Table 87‐1 Description and Results of Meyers’ Procedures (see text) Operation
No. patients
Extirpation of caudate head Extirpation of caudate head, section of anterior limb internal capsule Extirpation of caudate head, section of anterior limb internal capsule, removal of oral third of putamen Extirpation of caudate head, section of anterior limb internal capsule, removal of oral third of putamen and oral pole of globus pallidus Ansotomy (section of pallidofugal fibers) Ansotomy, extirpation of caudate head, section of anterior limb internal capsule Ansotomy, extirpation of caudate head, section of anterior limb internal capsule, removal of oral third of putamen Extirpation of caudate head, linear separation of motor and premotor cortex, undercutting of premotor cortex Total
1 11 6
58
Results
No. patients
Much Improved (tremor-free, some rigidity, no clinical neurological deficits) Improved (tremor and rigidity reduced, not abolished; if abolished, attended by some degree of postoperative dyspraxia, paresis, or spasticity) Unimproved Mortality Total
11 (19%) 25 (43%)
10 patients, but there were three deaths. Hamby [42] reported a single patient with dystonia musculorum deformans (idiopathic torsion dystonia) who underwent this procedure without benefit. Meyers felt that the operative mortality for these procedures was excessive for elective surgery and that open surgery at the level of the basal ganglia had a very limited role to play in the treatment of Parkinson’s disease. However, he had shown that surgery on the basal ganglia did not cause coma and could eliminate tremor and significantly reduce rigidity without the neurological consequences of corticectomy, pedunculotomy, or cordotomy.
Anterior Choroidal Artery Ligation By accident, Cooper [43,44] discovered another way to perform Meyers’ operation when in 1952 he inadvertently tore the anterior choroidal artery and abandoned a cerebral pedunculotomy
4 22 10 3 1
15 (26%) 7 (12%) 58
on a 39-year-old man with postencephalitic Parkinson’s disease. Postoperatively, tremor and rigidity were reduced without hemiparesis. Cooper concluded that he had infarcted a structure involved in tremor and proceeded to do anterior choroidal artery occlusion in 55 patients with parkinsonism [45]; contralateral tremor in 65% was nearly eliminated, rigidity decreased in 75%, and 11% suffered hemiplegia at the expense of 10% operative mortality. Anatomic studies showed that the anterior choroidal artery irrigates most of the globus pallidus, subthalamic nucleus, ventrolateral thalamus, red nucleus, and ansa lenticularis and contributes blood to the hippocampus, optic tract, lateral geniculate body, and posterior limb of the internal capsule. Abbie [46] stated that occlusion of the anterior choroidal artery produced consistent necrosis in most of the globus pallidus, accompanied by partial destruction of the retrolenticular portion of the internal capsule and the optic radiations. However, human studies showed that
1471
1472
87
History of surgery for movement disorders
destruction could be minimal or could involve the medial segment of the globus pallidus, the optic tract, and the posterior limb of the internal capsule [47]. Variability made the procedure rather unpredictable [48] and limited its popularity.
improvement in athetosis [57], choreoathetosis [60], and hemiballismus [60]. Gioino and associates [61] and Tsubokawa and Mariyasu [62] also reported excellent relief of hemiballismus after pallidal lesions.
Human Stereotactic Surgery
Subthalamic Region
Pallidotomy and Ansotomy
In performing pallidotomy, Spiegel and associates worried about concomitant damage to the internal capsule or the anterior choroidal artery that lay near the pallidum. To avoid these risks, Spiegel and Wycis made lesions more distally at the junction of the ansa lenticularis and lenticular fasciculus in Forel’s field H [63,64] (campotomy) in 25 parkinsonian patients, all of whom had tremor and 20 of whom had significant rigidity. Tremor was significantly reduced in 22 patients, and rigidity decreased in 17 without operative mortality and with only two cases of transient hemiparesis. A variety of targets in the subthalamic area have been used to treat parkinsonian tremor. Given the inherent variability in the diencephalic area from patient to patient as well as the potential inaccuracies in target localization, it is difficult to know exactly which structures were affected in the reported studies. Those implicated include Forel’s fields H and H2, the zona incerta, and the prelemniscal radiations [63,65–70]. Lesions in this area produce results comparable to those of ventrolateral thalamotomy with similar complications, but effective lesions can be smaller than those required for effective thalamotomy.
In 1947, Spiegel and Wycis [49] introduced the first effective human stereotactic device with which to approach deep structures by using intracranial radiological guidance. This was a revolutionizing form of surgery for movement disorders. Their first lesions were made in the dorsomedial thalamic nucleus for psychiatric illness, the mesencephalic pain pathways for intractable pain, and the medial thalamus for epilepsy [50]. They were afraid to lesion the globus pallidus or its outflow for fear of enhancing rigidity, as this procedure did in experimental animals. Thus, their first cautious pallidal lesion was made in a patient with Huntington’s chorea [51], followed by patients with choreoathetosis [52]. When no rigidity ensued, they lesioned the pallidum and ansa lenticularis (pallidoansotomy) in parkinsonian patients [53], improving tremor in 78% of 50 patients, with an operative mortality of 2.8%, permanent hemiplegia in 4% of patients, and transient hemiparesis in 12% of patients [54]. Pallidotomy and pallidoansotomy were then widely used in Parkinson’s disease, with tremor reduction or abolition being claimed in 60–80% of patients and rigidity suppression in 75–90% [55–58]. The procedures seemed more successful for rigidity than for tremor, with the latter often recurring if the operation first abolished it. Cooper [59] found pallidotomy useful in patients with nonparkinsonian movement disorders. Among 30 patients, 19 with choreoathetosis and 8 with dystonia, 20 were significantly improved. Narabayashi and colleagues reported
Thalamotomy Hassler concluded that the ventrolateral nucleus of the thalamus should be an effective target for the relief of parkinsonian symptoms and, with Riechert, carried out thalamotomy in
History of surgery for movement disorders
1952 [71,72]. At autopsy, it was concluded that rigidity was relieved by lesions in nucleus ventralis oralis anterior (Voa) and the field of Forel and tremor was relieved in the nucleus ventralis oralis posterior (Vop) [73,74]. Cooper was in the habit of performing chemopallidectomy freehand under X-ray control with ventriculography and skull landmarks. If his first injection was unsuccessful, he made a second one more posteriorly. Subsequent autopsy studies showed that this second lesion lay in the ventrolateral thalamus [55,75]. He soon concluded that ‘‘a lesion in the ventrolateral thalamus has a more complete and lasting effect on contralateral tremor than does the pallidal lesion’’ [76]. Thalamotomy became his treatment of choice for most parkinsonian patients, with long-term tremor relief in 89% and improved rigidity in approximately 70% of 203 parkinsonian patients reported [77]. With the support of both Spiegel and Wycis and Cooper, thalamotomy then took over as the main surgical treatment for Parkinson’s disease. Thalamotomy proved more effective than pallidotomy for parkinsonian tremor and at least as effective for rigidity, with fewer complications [78–80]. Mundinger’s multicenter review of 2,033 patients [81] revealed that rigidity improved in 76 11% of patients after pallidotomy and 75.2 0.5% after thalamotomy; tremor responded to pallidotomy in 55 22% after pallidotomy and 65 15% after thalamotomy. In his own 945 patients, these data were 90, 85, 59, and 86%, respectively. Within the thalamus, lesions in the posterior part of the ventrolateral nucleus (Vop, Vim) had a greater effect on tremor than did those in the anterior VL (Voa); the latter had a greater effect on rigidity and dopa dyskinesias [82–84]. Ventrolateral thalamotomy was also used successfully in the treatment of essential tremor [85–89], post-stroke or post-traumatic intention tremor [89–91], dystonia [92–94,116], hemiballismus [91,95], writer’s cramp [96], writing tremor [89], and intention tremor associated with
87
multiple sclerosis [97,98], though for many of these conditions the reported numbers are low. The results in essential tremor are similar to those for parkinsonian tremor, while in the other conditions, significant improvement is seen in onethird to one-half of the patients and moderate improvement is seen in an additional one-third.
Target Localization Successful human stereotactic surgery depends on accurate localization of subcortical structures. Spiegel and Wycis initially used the pineal, then the posterior commissure (PC) for procedures in the posterior thalamus and mesencephalon, and the anterior commissure (AC) for procedures in the region of the globus pallidus as outlined by ventriculography [99]. Their stereotactic atlas [100] related subcortical structures to these intracranial landmarks. Talairach constructed an atlas based on the anterior-posterior (AC-PC) commissure line [101], while others utilized a line from the PC to the foramen of Monro [72]. Various stereotactic atlases were produced, some of which included variability tables allowing for variations in the length of the AC-PC line [102–104]. Variability of the position of the posterior commissure with respect to thalamic structures was stressed [105]. Despite the approximate locations of targets that could be achieved with an atlas, some means of physiological confirmation of the location of a target was necessary.
Physiological Localization One technique of physiological localization was that of first making a temporary lesion before proceeding to a permanent one, as Cooper [106] and Narabayashi [57] did with a local anesthetic. Then electrical stimulation was used to distinguish motor and sensory structures and to avoid the optic tract during pallidal surgery and the internal
1473
1474
87
History of surgery for movement disorders
capsule during pallidal and thalamic surgery [107,108]. In the thalamus, contralateral somatotopographically arranged sensory phenomena could be elicited in the sensory nucleus [108,109]. In tremor patients, thalamic target sites could be identified by stimulation-induced tremor reduction or arrest at high frequencies and tremor drive at low frequencies. In the field of Forel [63] and globus pallidus, similar effects were achieved, less consistently in the latter [110]. Albe-Fessard and colleagues [111,112] pioneered the techniques of experimental neurophysiology in the operating room with the use of microelectrodes. Microelectrode recording distinguishes white from gray matter, thereby establishing nuclear boundaries. It allows the physiologist to ‘‘interrogate’’ the neurons with differing sensory stimuli and can help determine functional nuclear boundaries and intranuclear somatotopy. This technique has subsequently been widely adopted [113–117] and has contributed to an understanding of normal and abnormal physiology. While modern magnetic resonance imaging (MRI) and computed tomography (CT) technology is steadily enhancing the stereotactic technique with an accuracy comparable to that of ventriculography, it still cannot satisfy the need for physiological corroboration of the target [94]. Physiological studies are still striving to identify the ‘‘ideal’’ lesion site in thalamus [47,118].
fibers in the subthalamic region were the main procedures used [65,70,87,97,119–121] in preference to pallidotomy. A number of factors were responsible for the rekindling of interest in pallidotomy in the late 1980s: 1.
2.
3.
The Modern Era – The Last 20 Years Posteroventral Pallidotomy The late 1960s and early 1970s witnessed a substantial decline in the number of procedures performed for the treatment of Parkinson’s disease and for movement disorders in general, largely a result of the introduction of l-dopa. The surgical treatment of Parkinson’s disease was limited to the treatment of tremor refractory to l-dopa and thalamotomy and/or interruption of pallidofugal
Many of the limitations and side effects of chronic l-dopa therapy were becoming evident: loss of efficacy, disabling motor fluctuations and dopa dyskinesias. In the late 1950s, Lars Leksell modified his pallidotomy target from the classical anterodorsal portion of the medial pallidum to the posteroventral pallidum. He reported improvement of tremor and rigidity, but also saw ‘‘improved mobility in terms of strength, range, speed and precision’’ [58]. This statement was taken to indicate an improvement in bradykinesia. At the time, the larger community of functional neurosurgeons overlooked the significance of these results. Using the same procedure in 38 patients, Laitinen [122] reported improvement of tremor in 81% of patients, rigidity in 92%, and hypokinesia in 92%. He also noted a beneficial effect on dopa dyskinesias. Laitinen presented this material at number of international meetings and generated interested in revisiting the pallidotomy target. Better pathophysiological models of basal ganglia function were being developed [123,171], and it was becoming understood that many of the ‘‘negative’’ symptoms of Parkinson’s disease were due to neuronal overactivity in the medial pallidum and the subthalamic nucleus (> Figure 87‐3). Models of basal ganglia circuitry proposed that the loss of striatal dopamine in Parkinson’s disease caused overactivity of the striatal projection to the lateral segment of the globus pallidus. The resulting decrease in lateral pallidal activity results in disinhibition of the subthalamic nucleus, its main projection site.
History of surgery for movement disorders
87
. Figure 87‐3 Simplified model of basal ganglia function. (a) Normal Basal Ganglia: Outflow from the striatum is directed along two pathways. The ‘‘direct’’ pathway is an inhibitory pathway from the striatum to the internal segment of the globus pallidum. In the ‘‘indirect’’ pathway the striatum inhibits the external segment of the globus pallidus which, in turn inhibits the subthalamic nucleus. The subthalamic nucleus is excitatory on the internal segment of the pallidum. The net outflow of the basal ganglia is inhibitory on the thalamus and the pedunculopontine nucleus. (b) Basal Ganglia in Parkinson’s Disease: In Parkinson’s disease, the dopaminergic connections from the substantia nigra pars compacta are impaired, and striatal output is reduced. In the ‘‘indirect’’ pathway, the net effect is overactivity of the subthalamic nucleus. The net effect on the internal segment of the globus pallidus is increased excitation from the STN outflow and decreased inhibition from the striatum resulting in overactivity. Green arrows indicate excitation, red arrows indicate inhibition: GPe – globus pallidus external segment; GPi – globus pallidus internal segment; STN – subthalamic nucleus; PPN – pedunculopontine nucleus; Thal – pallidal receiving area in the thalamus; SNc – substantia nigra pars compacta
Increased subthalamic activity in turn causes overactivity of the internal segment of the globus pallidus, which projects to the pedunculopontine nucleus (PPN) and the ventrolateral (VL) thalamus [123]. Presumably, overactivity in the subthalamic nucleus and internal pallidum produces the parkinsonian symptoms of tremor, bradykinesia, and hypokinesia through projections to the PPN and VL thalamus. In 1991 the first North American ‘‘modernera’’ pallidotomies were carried out [127], and interest rapidly grew in the functional neurosurgical community. A survey of neurosurgeons invited to the ‘‘Pallidotomy. . .where are we’’ meeting in Irvine California 1995 sampled 28 centers in North America and showed how rapidly pallidotomy had been integrated into the treatment of Parkinson’s Disease [125]. The
survey reported that 1,015 patients had undergone 1,219 pallidotomies in these centers and almost certainly underestimated the total number of such procedures that had been performed in total. Posteroventral pallidotomy produced improvement in contralateral dopa-dyskinesias, tremor, on/off motor fluctuations, akinesia and rigidity [125,126,130,131,133,136,158,207,208]. The most dramatic effects appeared to be on dyskinesias and off-period dystonia. The benefits were sustained in long-term (4–5 year) follow-up [133,158] although with a tendency for motor scores to slowly worsen over time, either due to disease progression or to loss of surgical benefit. It became evident that parkinsoniansymptoms thatwere dopa-responsive were responsive to pallidotomy. Autonomic, psychological, cognitive, speech and swallowing problems were not helped by pallidotomy. On-period postural instability and gait problems were also resistant to pallidotomy. The ideal patient was one
1475
1476
87
History of surgery for movement disorders
with predominant bradykinesia, severe ‘‘off’’states, dyskinesias during the ‘‘on’’ state, and excellent response to l-dopa. Unilateral pallidotomy was considered a safe and effective procedure. But bilateral pallidotomies were associated with a higher incidence (up to 60%) of speech problems (hypophonia, dysarthria), dysphagia, drooling [124,134,135,137,172,207,210,211,212], as well as an increased incidence of cognitive problems [137,207,210,212]. Some proposed making smaller contralateral lesions to avoid these problems. With the development of chronic implantable stimulation technology, implantation of a chronic pallidal stimulator contralateral to a pre-existing pallidotomy became another option [126].
Subcortical Stimulation Acute stimulation in the thalamus during stereotactic surgery has long been used to identify the appropriate target area for tremor alleviation. Stimulation at high frequencies (>100 Hz) suppresses tremor at such sites. Chronic stimulation of the thalamic VPL nucleus has been used to treat patients with pain and sensory deafferentation [144]. Andy chronically stimulated the parafascicular and center median nuclei in patients with pain and dyskinesias and stimulated various sites in the thalamus and subthalamic area in patients with a heterogeneous group of movement disorders [145,146]. Siegfried et al. [147] noted that thalamic stimulation in patients with deafferentation pain and dyskinesias helped not only the pain but also the movement disorder. Benabid and his group [148], carried out chronic stimulation of the thalamic ventral intermediate nucleus (Vim) in 26 patients with Parkinson’s disease and 6 patients with essential tremor, reporting 92% relief or major improvement in parkinsonian tremor and 67% improvement in essential tremor. Similar results were reported by Blond [149,150] who also noted a
beneficial effect on dopa dyskinesias [151]. Thalamic deep brain stimulation (dbs) has also proved effective for severe cerebellar intention tremor [203] and hemiballismus [204] in small numbers of patients. When assessed in large multicenter studies thalamic stimulation improved parkinsonian and essential tremor with results comparable to thalamotomy [143,154]. Adverse effects occurring with thalamotomy are also seen with chronic Vim stimulation (dysarthria, ataxia, paresthesias), but reducing stimulation amplitude and accepting mild residual tremor can eliminate them. A rebound effect may be seen in which tremor of increased amplitude occurs for a short period after the discontinuation of stimulation. The potential superiority of thalamic stimulation over lesioning lay in the ability to adjust the stimulation parameters to regain tremor control when tremor recurs and to reduce or eliminate permanent adverse effects [139,142]. With the efficacy and safety of dbs in the thalamus having been established it was reasonable to extend its application to other brain targets. Stimulation of the pallidum for Parkinson’s disease was found to produce different effects depending on location, with dorsal contacts improving akinesia and inducing dyskinesias, and ventral contacts reducing akinesia and dyskinesias [152,155]. Published series showed that unilateral pallidal dbs for Parkinson’s disease was as effective as unilateral pallidotomy and bilateral pallidal dbs probably safer [140,141,159,160,161,165,213]. But there was also added morbidity related to stimulation including: lead fracture, infection, skin erosion and premature battery failure. These procedures were more expensive than lesioning and necessitated continuing follow-up for stimulation adjustment. DBS applied to the standard dystonia targets (pallidum, Vim and Vop thalamus) produces benefits that mirror ablative surgery [129,196,197,198,199,200]. The delayed improvement following pallidotomy is also seen in pallidal dbs. There is still uncertainty whether bilateral
History of surgery for movement disorders
pallidotomy is more effective that a combination of lesioning and stimulation [201].
Subthalamic Nucleus Experimental animal models of Parkinson’s disease demonstrated an increase in neuronal activity in the subthalamic nucleus (STN) and the globus pallidus corresponding to the development of parkinsonian signs (> Figure 87‐3) [123,168,174]. In MPTP monkeys excitotoxic lesions, radiofrequency lesions and high frequency stimulation all alleviate these signs, often with the simultaneous production of dyskinesias and hemiballism (transient or permanent) [123,153,163,168,169,170,171,172,173,206]. In the early 1990s the first reports of bilateral STN-dbs in humans started to emerge, demonstrating improvement in all of the cardinal signs and symptoms of Parkinson’s [164,167,183]. Tremor, off-period akinesia, rigidity and dystonia are improved. The initiation of STN stimulation induces dyskinesias that can be controlled by reducing the dopaminergic medications, resulting in an overall improvement in on-period dyskinesias. In several studies improvements in off-period motor function were maintained over long-term follow-up [164,175,176], with gradual worsening of akinesia, speech, postural instability and gait freezing consistent with the slow natural progression of Parkinson’s disease. Comparisons between pallidal surgery and subthalamic dbs suggested similar overall benefit [141,156,176,177,178]. Akinesia was improved more with STN-dbs and dyskinesias more with pallidotomy or pallidal-dbs. STN-stimulation allowed greater medication reduction than pallidotomy or pallidal-dbs. Stimulators in the pallidum generally required higher voltages leading to earlier battery replacement and higher cost of treatment. There was evidence that cognitivebehavioral morbidity was somewhat higher with STN-dbs [156,157,176].
87
STN lesioning in humans was introduced with the hope that it would provide a cost-effective method of achieving many of the benefits seen with STN-dbs. STN lesioning improves rigidity, bradykinesia, tremor and reduces l-dopa requirements [180,181,182,185]. The durability of the results varies between the reporting centers and may relate to lesion location. Some suggest that the best lesions extend from the dorsal STN into the pallidofugal fibers [182,185]. Dyskinesias were common and often transient, but additional surgery was required (thalamotomy [185], subthalamic area dbs [182]) when these persisted. Additional morbidity included ataxia and dysarthria.
Transplantation Attempts to physiologically ‘‘reconstruct’’ the basal ganglia by implanting dopamine-producing tissue started with Madrazo’s transplantation of adrenal medullary tissue into the caudate nucleus of parkinsonian patients [205] and have extended to the use of fetal nigral tissue [190]. The results of adrenal transplantation have been collected in two central registries. The American Association of Neurological Surgeons Registry documented significant clinical improvement in 20% of 140 patients [191]. The United Parkinson Foundation Registry reported modest improvement in the first year after transplantation, but this waned during the second year. At 2 years, 31% of patients demonstrated increased daily ‘‘on’’ time and improved quality of function during ‘‘off’’ periods. Nine percent of these patients died of complications that might have been related to surgery, and there were persistent psychiatric problems in 22%. Autopsy after adrenal transplantation demonstrated few if any surviving adrenal cells [192]. Fetal nigral grafts in patients with Parkinson’s disease and MPTP-induced parkinsonism produced some degree of clinical improvement along with evidence of graft survival on positron emission tomography (PET) studies [193–195].
1477
1478
87
History of surgery for movement disorders
Subsequent clinical trials have demonstrated that grafted neurons survive, reinnervate the striatum and produce dopamine in the brains of parkinsonian patients. However, the clinical benefit is limited and not always correlated with changes in 18 F-dopa uptake as demonstrated on PET. There is some suggestion that patients under 60 years old do better. Troubling off-period dyskinesias develop in 15–50% of graft recipients [179,184,186,187,188,189]. Research is currently investigating the way that graft tissue interacts with the host brain, and the possibility of genetically engineered cells that can serve as graft tissue. The role that transplantation will ultimately play in the management of patients with Parkinson’s disease is yet to be determined.
Summary The history of the surgical treatment of movement disorders has been a story of experimental neurophysiology applied to the human brain. Current and future trends will be guided by a growing understanding of the chemical and electrical circuitry of the basal ganglia and will be aided by technical advances that allow more precise and safer surgical procedures.
References 1. Horsley V. Surgery of the central nervous system. Br Med J 1890;2:1286-92. 2. Horsley V. The function of the so-called motor area of the brain. Br Med J 1909;2:125-32. 3. Bucy PC, Buchanan DN. Athetosis. Brain 1932; 55:479-92. 4. Bucy PC. Cortical extirpation in the treatment of involuntary movements. Res Publ Assoc Res Nerv Ment Dis 1942;21:551-95. 5. Bucy PC, Case TJ. Athetosis: II. Surgical treatment of athetosis. Arch Neurol Psychiatry 1939;41:721-46. 6. Bucy PC, Case TJ. Tremor, physiologic mechanism and abolition by surgical means. Arch Neurol Psychiatry 1939;41:721-46.
7. Sachs E. The subpial resection of the cortex in the treatment of Jacksonian epilepsy (Horsley operation) with observations on areas 4 and 6. Brain 1935;58:492-503. 8. Putnam TJ. The operative treatment of diseases characterized by involuntary movements (tremor, athetosis). Res Publ Assoc Res Nerv Ment Dis 1942;21:666-96. 9. Klemme RM. Surgical treatment for dystonia, paralysis agitans and athetosis. Arch Neurol Psychiatry 1940;44:926. 10. Klemme RM. Surgical treatment of dystonia, with report of one hundred cases. Res Publ Assoc Res Nerv Ment Dis 1942;21:596-601. 11. Parkinson J. An essay on the shaking palsy. London: Sherwood, Neely & Jones, 1817. Cited in Ostheimer AJ: A bibliographic note on ‘‘Essay on Shaking Palsy,’’ by James Parkinson, MD, Member of the Royal College of Surgeons. Arch Neurol Psychiatry 1922;7: 681–710. 12. Patrick HT, Levy DM. Parkinson’s disease. Arch Neurol Psychiatry 1922;7:711-20. 13. Aring CD, Fulton JF. Relation of the cerebrum to the cerebellum: cerebellar tremor in the monkey and its absence after removal of the principal excitable areas of the cerebral cortex (areas 4 and 6a, upper part): accentuation of cerebellar tremor following lesions of the premotor area (area 6a, upper part). Arch Neurol Psychiatry 1936;35:439-66. 14. Putnam TJ. Treatment of unilateral paralysis agitans by section of the lateral pyramidal tract. Arch Neurol Psychiatry 1940;44:950-76. 15. Bucy PC. The surgical treatment of abnormal involuntary movements. J Surg Gynecol Obstet 1950;90:376-9. 16. Bucy PC. Surgical treatment of extrapyramidal diseases. J Neurol Neurosurg Psychiatry 1951;14:108-17. 17. Bucy PC. Cortical extirpation in the treatment of involuntary movements. Am J Surg 1948;75:257-63. 18. Putnam TJ. Treatment of athetosis and dystonia by section of exrapyramidal motor tracts. Arch Neurol Psychiatry 1933;29:504-21. 19. Putnam TJ. Results of treatment of athetosis by section of extrapyramidal tracts in the spinal cord. Arch Neurol Psychiatry 1938;39:258-75. 20. Putnam TJ. Relief from unilateral paralysis agitans by section of the pyramidal tract. Arch Neurol Psychiatry 1938;40:1049-50. 21. Rothmann M. Ueber die Ergebnisse der experimentellen Ausschaltungen der motorischen Funktion und ihrer Bedeutung fur die Pathologie. Z Klin Med 1903;48:10-29. 22. Meyers R. Historical background and personal experiences in the surgical relief of hyperkinesia and hypertonus. In: Fields WS, editor. Pathogenesis and treatment of parkinsonism. Springfield, IL: Charles C Thomas; 1958. p. 229-70. 23. Ebin J. Combined lateral and ventral pyramidotomy in treatment of paralysis agitans. Arch Neurol Psychiatry 1949;62:27-47.
History of surgery for movement disorders
24. Oliver LC. Surgery in Parkinson’s disease: complete section of the lateral column of the spinal cord for tremor. Lancet 1950;1:847-8. 25. Oliver LC. Parkinson’s disease and its surgical treatment. London: HK Lewis; 1953. 26. Walker AE. Cerebral pedunculotomy for the relief of involuntary movements: hemiballismus. Acta Psychiatr Scand 1949;24:723-9. 27. Walker AE. Cerebral pedunculotomy for the relief of involuntary movements: parkinsonian tremor. J Nerv Ment Dis 1952;116:766-75. 28. Guiot G, Pecker J. Tractotomie mesencephalique anterieure pour tremblement parkinsonien. Rev Neurol (Paris) 1949;81:387-8. 29. Bucy PC. Is there a pyramidal tract? Brain 1957;80:376-92. 30. Meyers R. The human frontocorticopontine tract: functional inconsequence of its surgical interruption. Neurology 1951;1:341-56. 31. Meyers R. Results of bilateral intermediate midbrain crusotomy in seven cases of severe athetotic and dystonic quadriparesis. Am J Phys Med Rehabil 1956;35: 84-105. 32. Meyers R. Physiological and therapeutic effects of bilateral intermediate midbrain crusotomy for atheto-dystonia. Surg Forum 1956;6:486-8. 33. Meyers R. The surgery of the hyperkinetic disorders. In: Vinken PJ, Bruyn GW, editors. Handbook of clinical neurology, vol. 65. Amsterdam: North-Holland; 1969. p. 844-78. 34. Bucy PC, Keplinger JE, Sequeira EB. Destruction of the ‘‘pyramidal tract’’ in man. J Neurosurg 1964;21:385-98. 35. Meyers R. A surgical procedure for the alleviation of postencephalitic tremor, with notes on the physiology of the premotor fibres. Arch Neurol Psychiatry 1940;44:455-9. 36. Dandy WE. Changes in our conceptions of localization of certain functions in the brain. Am J Physiol 1930;93:643-7. 37. Dandy WE. The location of the conscious center in the brain: the corpus striatum. Bull Johns Hopkins Hosp 1946;79:34-46. 38. Meyers R. The modification of alternating tremors, rigidity and festination by surgery of the basal ganglia. Res Publ Assoc Res Nerv Ment Dis 1942;21:602-65. 39. Meyers R. Surgical experiments in the therapy of certain ‘‘extrapyramidal diseases’’: a current evaluation. Acta Psychiatr Neurol 1951;67:1-42. 40. Browder J. End results following the capsular operation for parkinsonism. Surg Clin North Am 1948;28:390-5. 41. Browder J. Section of the fibers of the anterior limb of the internal capsule in parkinsonism. Am J Surg 1948;75:264-8. 42. Hamby WB. The surgical treatment of dystonia musculorum deformans. J Neurosurg 1953;10:490-5.
87
43. Cooper IS. Ligation of the anterior choroidal artery for involuntary movements of parkinsonism. Psychiatr Q 1953;27:317-9. 44. Cooper IS. Surgical occlusion of the anterior choroidal artery in parkinsonism. Surg Gynecol Obstet 1954;99:207-19. 45. Cooper IS. The neurosurgical alleviation of parkinsonism. Springfield, IL: Charles C Thomas; 1956. 46. Abbie AA. The morphology of the fore-brain arteries with special reference to the evaluation of the basal ganglia. J Anat 1934;68:433-70. 47. Mettler FA, Cooper I, Liss H, et al. Patterns of vascular failure in the central nervous system. J Neurol Exp Neurapathol 1954;13:528-39. 48. Rand RW, Brown WJ, Stern WE. Surgical occlusion of anterior choroidal arteries in parkinsonism: clinical and neuropathologic findings. Neurology 1956;6:390-401. 49. Spiegel EA, Wycis HT, Marks M, Lee AJ. Stereotactic apparatus for operations in the human brain. Science 1947;106:349-50. 50. Spiegel EA, Wycis HT, Freed H, Lee AJ. Stereoencephalotomy. Proc R Soc Exp Biol Med 1948;69:175-7. 51. Spiegel EA, Wycis HT. Pallidothalamotomy in chorea. Arch Neurol Psychiatry 1950;64:295-6. 52. Spiegel EA, Wycis HT, Baird HW. Effect of thalamic and pallidal lesions upon involuntary movements in choreoathetosis. Trans Am Neurol Assoc 1950;75:234. 53. Spiegel EA, Wycis HT. Ansotomy in paralysis agitans. Arch Neurol Psychiatry 1953;69:652-3. 54. Wycis HT, Spiegel EA. Long range results of pallidoansotomy in paralysis agitans and parkinsonism. In: Fields WS, editor. Pathogenesis and treatment of parkinsonism. Springfield, IL: Charles C Thomas; 1958. p. 294-8. 55. Cooper IS, Bravo G. Chemopallidectomy and chemothalamectomy. J Neurosurg 1958;15:244-50. 56. Housepian EA, Pool JL. An evaluation of pallido-ansal surgery. In: Fields WS, editor. Pathogenesis and treatment of parkinsonism. Springfield, IL: Charles C. Thomas; 1958. p. 317-24. 57. Narabayashi H, Okuma T, Shikiba S. Procaine oil blocking of the globus pallidus. Arch Neurol Psychiatry 1956;75:36-48. 58. Svennilson E, Torvik A, Lowe R, Leksell L. Treatment of parkinsonism by stereotactic thermal lesions in the pallidal region. Acta Psychiatr Neurol Scand 1960;35:358-77. 59. Cooper IS. Relief of juvenile involuntary movement disorders by chemopallidectomy. JAMA 1957;164:1297-331. 60. Narabayashi H, Okuma R, Shikiba S. Discussion on the topic ‘‘Stereotactic Method.’’ Trans 1st Internat Cong Neurol Surg, Excerpta Medica, 34–35 (1957), cited in Speigel EA (ed): Progress in Neurology and Psychiatry. New York: Grune & Stratton, 1959, vol 14. 61. Gioino GG, Dierssen G, Cooper IS. The effect of subcortical lesions on production and alleviation of hemiballic or hemichoreic movements. J Neurol Sci 1966;3:10-36.
1479
1480
87
History of surgery for movement disorders
62. Tsubokawa T, Moriyasu N. Lateral pallidotomy for relief of ballistic movement – its basic evidences and clinical application. Confin Neurol 1975;37:10-15. 63. Spiegel EA, Wycis HT, Szekely EG, et al. Campotomy in various extrapyramidal disorders. J Neurosurg 1963:20;871-84. 64. Spiegel EA, Wycis HT, Szekely EG, et al. Campotomy. Trans Am Neurol Assoc 1962;87:240-2. 65. Andy OJ, Jurko MF, Sias FR, Jr. Subthalamotomy in treatment of parkinsonian tremor. J Neurosurg 1963;22:860-70. 66. Hullay J, Velok J, Gombi R, Boczan G. Subthalamotomy in Parkinson’s disease. Confin Neurol 1970;32:345-8. 67. Kjellberg RN. Pallido-thalamic mechanisms in movement disorders. Confin Neurol 1965;26:328-35. 68. Mundinger F. Results of 500 subthalamotomies in the region of the zona incerta. In: Third symposium on Parkinson’s disease. Edinburgh and London: E & S Livingstone; 1969. p. 261-5. 69. Story JL, French LA, Chou SN, Meier MJ. Experience with subthalamic lesions in patients with movement disorders. Confin Neurol 1965;26:218-21. 70. Velasco F, Molina-Negro P, Bertrand C, Hardy J. Further definition of the subthalamic target for arrest of tremor. J Neurosurg 1972;36:184-91. 71. Hassler R. Die extrapyramidalen rindensysteme und die zentrale regelung der motorik. Dtsch Z Nervenheilk 1956;175:233-58. 72. Hassler R, Riechert T. Indikationen und lokalisations methode der gerzielten himoperationen. Nervenarzt 1954;25:441-7. 73. Hassler R, Mundinger F, Riechert T. Correlations between clinical and autoptic findings in stereotaxic operations in parkinsonism. Confin Neurol 1965;26:282-90. 74. Hassler R, Riechert T, Mundinger F, et al. Physiological observations in stereotaxic operations in extrapyramidal motor disturbances. Brain 1960;83:337-51. 75. Cooper IS, Bravo GJ, Riklan M, et al. Chemopallidectomy and chemothalamectomy for parkinsonism. Geriatrics 1958;13:127-47. 76. Cooper IS, Bravo GJ. Implications of a five-year study of 700 basal ganglia operations. Neurology 1958;8:701-7. 77. Cooper IS. Cryogenic technique of thalamic surgery for parkinsonism and other involuntary movement disorders. Prog Neurol Surg 1973;5:159-88. 78. Broager B. The surgical treatment of parkinsonism. Acta Neurol Scand 1963; Suppl 4:181-7. 79. Krayenbuhl H, Wyss OAM, Yasargil MG. Bilateral thalamotomy and pallidotomy as treatment for bilateral parkinsonism. J Neurosurg 1961;18:429-44. 80. Selby G. Stereotactic surgery for the relief of parkinsons disease: 1. A critical review. J Neurol Sci 1967;5:315-42. 81. Mundinger F. Die stereotaksische operative behandlung des parkinsyndroms: Pathophysiologie, ergebnisse und indikationsstellung. Med Klin 1963;58:1181-6.
82. Kawashima Y, Yakahashi A, Hirato M, Ohye C. Stereotactic Vim Vo thalamotomy for choreatic movement disorders. Acta Neurochir Suppl (Wien) 1991;52:103-6. 83. Narabayashi H, Maeda T, Yokochi F. Long-term follow-up study of nucleus ventralis intermedius and ventrolateral is thalamotomy using a microelectrode technique in parkinsonism. Appl Neurophysiol 1987;50:330-7. 84. Narabayashi H, Yokochi T, Nakajima Y. Levodopainduced dyskinesia and thalamotomy. J Neurol Neurosurg Psychiatry 1984;47:831-9. 85. Cooper IS. Heredofamilial tremor abolition by chemothalamectomy. Arch Neurol 1962;7:129-31. 86. Goldman MS, Kelly PJ. Stereotactic thalamotomy for medically intractable essential tremor. Stereotact Funct Neurosurg 1992;58:22-5. 87. Goldman MS, Ahlskog JE, Kelly PJ. The symptomatic and functional outcome of stereotactic thalamotomy for medically intractable essential tremor. J Neurosurg 1992;76:924-8. 88. Mohadjer M, Goerke H, Milios E, et al. Long-term results of stereotaxy in the treatment of essential tremor. Stereotact Funct Neurosurg 1990;54/55:125-9. 89. Ohye C, Hirai T, Miyazaki M, et al. Vim thalamotomy for the treatment of various kinds of tremor. Appl Neurophysiol 1982;45:275-80. 90. Marks PV. Stereotactic surgery for post-traumatic cerebellar syndrome: an analysis of seven cases. Stereotact Funct Neurosurg 1993;60:157-67. 91. Sarosta-Rubinstein S, Bjork RJ, Snyder BD. Post traumatic intention myoclonus. Surg Neurol 1983; 20:131-2. 92. Caracalos A. Results of 103 cryosurgical procedures in involuntary movement disorders. Confin Neurol 1972;34:74-81. 93. Cooper IS. Dystonia reversal by operation on basal ganglia. Arch Neurol 1962;7:132-45. 94. Tasker RR, Yamashiro K, Lenz F, Dostrovsky JO. Thalamotomy for Parkinson’s disease. In: Lunsford LD, editor. Modern stereotactic neurosurgery. Boston: Nijhoff; 1988. p. 297-314. 95. Narabayashi H. Neurophysiological ideas on pallidotomy and ventrolateral thalamotomy for hyperkinesias. Confin Neurol 1962;22:291-303. 96. Hirai T, Miyazaki M, Nakajima H, et al. The correlation between tremor characteristics and the predicted volume of effective lesions in stereotaxic nucleus ventralis intermedius thalamotomy. Brain 1983;106:1001-18. 97. Goldman MS, Kelly PJ. Symptomatic and functional outcome of stereotactic ventralis lateralis thalamotomy for intention tremor. J Neurosurg 1992;77:223-9. 98. Mundinger F, Kuhn I. Postoperative and long-term results after stereotactic operations for action myoclonus in cases of encephalomyelitis disseminata. Appl Neurophysiol 1982;45:299-305. 99. Spiegel EA, Wycis HT. Mesencephalotomy in treatment of intractable facial pain. Arch Neurol Psychiatry 1953;69:1-13.
History of surgery for movement disorders
100. Spiegel EA, Wycis HT. Stereoencephalotomy: part 1. New York: Grune & Stratton; 1952. 101. Talairach J, David M, Tournoux P, et al. Atlas d’Anatomie Stereotaxique: Reperage Radiologique Indirect des Noyaux Gris Centraux, des Region Mesencephalo-sous-optique et Hypothalamique de l’Homme. Paris: Masson; 1957. 102. Andrew J, Watkins ES. A stereotaxic atlas of the human thalamus and related structures: a variability study. Baltimore: Williams & Wilkins; 1969. 103. Schaltenbrand G, Bailey P. Introduction to stereotaxis with an atlas of the human brain. Stuttgart: Thieme; 1959. 104. Van Buren MJ, MacCubbin DA. An outline atlas of the human basal ganglia with estimation of anatomical variation. J Neurosurg 1962;19:811-39. 105. Brierley JB, Beck E. The significance in human stereotactic brain surgery of individual variation in the diencephalon and globus pallidus. J Neurol Neurosurg Psychiatry 1959;22:287-98. 106. Cooper IS. Intracerebral injection of procaine into the globus pallidus in hyperkinetic disorders. Science 1954;119:417-8. 107. Bertrand C, Martinez SN, Poirier L, Gauthier C. Experimental studies and surgical treatment of extrapyramidal diseases. In: Fields WS, editor. Pathogenesis and treatment of parkinsonism. Charles C Thomas; Springfield, IL: p. 299-316. 108. Gillingham FJ, Watson WS, Donaldson AA, Naughton JAL. The surgical treatment of parkinsonism. Br Med J 1960;2:1395-402. 109. Bertrand C, Martinez SN. Functional localization with monopolar stimulation. J Neurosurg 1960;24:440-2. 110. Alberts WW, Feinstein B, Levin G, et al. Stereotaxic surgery for parkinsonism-clinical results and stimulation thresholds. J Neurosurg 1963;23:174-83. 111. Albe-Fessard D, Arfel G, Guiot G, et al. Identification et delimitation precise de certaines structures souscorticales de l’homme par l’electrophysiologie: Son interet dans la chirurgie stereotaxique de dyskinesies. C R Acad Sci 1961;253:2412-4. 112. Guiot G, Hardy J, Albe-Fessard D. Delimitation precise des structures sous-corticales et identification de noyaux thalamiques chez l’homme par l’electrophysiologie stereotaxique. Neurochirurgie 1962;5:1-18. 113. Bertrand G, Jasper H, Wong A. Microelectrode study of the human thalamus: functional organization in the ventro-basal complex. Confin Neurol 1967;29: 81-6. 114. Fukamachi A, Ohye C, Narabayashi H. Delineation of the thalamic nuclei with a microelectrode in stereotaxic surgery for parkinsonism and cerebral palsy. J Neurosurg 1973;39:214-25. 115. Lenz FA, Tasker RR, K wan HC, et al. Selection of the optimal lesion site for the relief of parkinsonian tremor on the basis of spectral analysis of neuronal firing patterns. Appl Neurophysiol 1987;50:338-43.
87
116. Ohye C, Shibazaki T, Hirato M, et al. Strategy of selective Vim thalamotomy guided by microrecording. Stereotact Funct Neurosurg 1990;54/55:186-91. 117. Velasco F, Molina-Negro P. Electrophysiological topography of the human diencephalon. J Neurosurg 1973;38:204-14. 118. Jones MW, Tasker RR. The relationship of documented destruction of specific cell types to complications and effectiveness in thalamotomy for tremor in Parkinson’s disease. Stereotact Funct Neurosurg 1990;54/ 55:207-11. 119. Fager CA. Evaluation of thalamic and subthalamic surgical lesions in the alleviation of Parkinson’s disease. J Neurosurg 1968;28:145-9. 120. Laitinen LV. Brain targets in surgery for Parkinson’s disease. J Neurosurg 1985;62:349-51. 121. Riechert T. Stereotaxic surgery for the treannent of Parkinson’s syndrome. Prog Neurol Surg 1973;5:1-78. 122. Laitinen LV, Bergenheim AT, Hariz MI. Leksell’s posteroventral pallidotomy for the treatment of Parkinson’s disease. J Neurosurg 1992;76:53-61. 123. DeLong MR. Primate models of movement disorders of basal ganglia origin. Trends Neurosci 1990;13:281-5. 124. Burchiel KJ, Favre J, Taha J, et al. Risk of speech deterioration after pallidotomy [abstract]. In: Proceedings of the congress of neurological surgeons 46th annual meeting, Montreal, Quebec, Canada, 1996. p. 259-60. 125. Favre J, Taha JM, Nguyen TT, Gildenberg PL, Burchiel KJ. Pallidotomy: a survey of current practice in North America. Neurosurgery 1996;39:883-90. 126. Lozano AM, Lang AE. Pallidotomy for Parkinson’s disease. Neurosurg Clin N Am 1998;9:325-36. 127. Dogali M, Fazzini E, Kolodny E, Eidelberg D, Sterio D, Devinsky O, Beric´ A. Stereotactic ventral pallidotomy for Parkinson’s disease. Neurology 1995;45:753-61. 128. Dostrovsky JO, Hutchison WD, Lozano AM. The globus pallidus, deep brain stimulation, and Parkinson’s disease. Neuroscientist 2002;8:284-90. 129. Eltahawy HA, Saint-Cyr J, Giladi N, Lang AE, Lozano AM. Primary dystonia is more responsive than secondary dystonia to pallidal interventions: outcome after pallidotomy or pallidal deep brain stimulation. Neurosurgery 2004;54:613-9. 130. Alkhani A, Lozano AM. Pallidotomy for Parkinson disease: a review of contemporary literature. J Neurosurg 2001;94:43-9. 131. Lang AE, Lozano A, Montgomery EB, Tasker RR, Hutchison WD. Posteroventral medial pallidotomy in advanced Parkinson’s disease. Adv Neurol 1999;80: 575-83. 132. Lozano AM, Lang AE, Galvez-Jimenez N, Miyasaki J, Duff J, Hutchinson WD, Dostrovsky JO. Effect of GPi pallidotomy on motor function in Parkinson’s disease. Lancet 1995;346(8987):1383-7.
1481
1482
87
History of surgery for movement disorders
133. Fazzini E, Dogali M, Sterio D, Eidelberg D, Beric A. Stereotactic pallidotomy for Parkinson’s disease: a longterm follow-up of unilateral pallidotomy. Neurology 1997;48:1273-7. 134. Counihan TJ, Shinobu LA, Eskandar EN, Cosgrove GR, Penney JB, Jr. Outcomes following staged bilateral pallidotomy in advanced Parkinson’s disease. Neurology 2001;56:799-802. 135. Parkin SG, Gregory RP, Scott R, Bain P, Silburn P, Hall B, Boyle R, Joint C, Aziz TZ. Unilateral and bilateral pallidotomy for idiopathic Parkinson’s disease: a case series of 115 patients. Mov Disord 2002; 17:682-92. 136. Baron MS, Vitek JL, Bakay RA, Green J, Kaneoke Y, Hashimoto T, Turner RS, Woodard JL, Cole SA, McDonald WM, DeLong MR. Treatment of advanced Parkinson’s disease by posterior GPi pallidotomy: 1-year results of a pilot study. Ann Neurol 1996;40:355-66. 137. Ghika J, Ghika-Schmid F, Fankhauser H, Assal G, Vingerhoets F, Albanese A, Bogousslavsky J, Favre J. Bilateral contemporaneous posteroventral pallidotomy for the treatment of Parkinson’s disease: neuropsychological and neurological side effects. Report of four cases and review of the literature. J Neurosurg 1999;91:313-21. 138. Alterman RL, Kelly PJ. Pallidotomy technique and results: the New York experience. Neurosurg Clin NA 1998;9:337-43. 139. Tasker RR. Deep brain stimulation is preferable to thalamotomy for tremor suppression. Surg Neurol 1998;49:145-54. 140. Siegfried J, Lippitz B. Bilateral chronic electrostimulation of ventroposterolateral pallidum: a new therapeutic approach alleviating all parkinsonian symptoms. Neurosurgery 1994;35:1126-9. 141. Kumar R, Lozano AM, Montgomery E, Lang AE. Pallidotomy and deep brain stimulation of the pallidum and subthalamic nucleus in advanced Parkinson’s disease. Mov Disord 1998;Suppl 1:73-82. 142. Schuurman PR, Bosch DA, Bossuyt PM, Bonsel GJ, van Someren EJ, de Bie RM, Merkus MP, Speelman JD. A comparison of continuous thalamic stimulation and thalamotomy for suppression of severe tremor. N Engl J Med 2000;342:461-8. 143. Koller W, Pahwa R, Busenbark K, Hubble J, Wilkinson S, Lang A, Tuite P, Sime E, Lozano A, Hauser R, Malapira T, Smith D, Tarsy D, Miyawaki E, Norregaard T, Kormos T, Olanow CW. High-frequency unilateral thalamic stimulation in the treatment of essential and parkinsonian tremor. Ann Neurol 1997;42:292-9. 144. Mazars G, Merienne L, Cioloca C. Control of dyskinesias due to sensory deafferentation by means of thalamic stimulation. Acta Neurochir Suppl (Wien) 1980;30:239-43. 145. Andy OJ. Parafascicular-center median nuclei stimulation for intractable pain and dyskinesia (painfuldyskinesia). Appl Neurophysiol 1980;43:133-44.
146. Andy OJ. Thalamic stimulation for control of movement disorders. Appl Neurophysiol 1983;46:107-11. 147. Siegfried J, Lippitz B. Chronic electrical stimulation of the VL-VPL complex and of the pallidum in the treatment of movement disorders: personal experience since 1982. Stereotact Funct Neurosurg 1994;62:71-5. 148. Benabid AL, Pollak P, Gervason C, et al. Longterm suppression of tremor by chronic stimulation of the ventral intermediate thalamic nucleus. Lancet 1991;337:403-6. 149. Blond S, Siegfried J. Thalamic stimulation for the treatment of tremor and other movement disorders. Acta Neurochir Suppl (Wien) 1991;52:109-11. 150. Blond S, Caparros-Lefebvre D, Parker F, et al. Control of tremor and involuntary movement disorders by chronic stereotactic stimulation of the ventral intermediate nucleus. J Neurosurg 1992;77:62-8. 151. Caparros-Lefebvre D, Blond S, Vermersch P, et al. Chronic thalamic stimulation improves tremor and levodopa induced dyskinesias in Parkinson’s disease. J Neurol Neurosurg Psychiatry 1993;56:268-73. 152. Krack P, Pollak P, Limousin P, Hoffmann D, Benazzouz A, Le Bas JF, Koudsie A, Benabid AL. Opposite motor effects of pallidal stimulation in Parkinson’s disease. Ann Neurol 1998;43:180-92. 153. Wichmann T, DeLong MR. Functional and pathophysiological models of the basal ganglia. Curr Opin Neurobiol 1996;6:751-8. 154. Limousin P, Speelman JD, Gielen F, Janssens M, Multicentre European study of thalamic stimulation in parkinsonian and essential tremor. J Neurol Neurosurg Psychiatry 1999;66:289-96. 155. Bejjani B, Damier P, Arnulf I, Bonnet AM, Vidailhet M, Dormont D, Pidoux B, Cornu P, Marsault C, Agid Y. Pallidal stimulation for Parkinson’s disease: two targets? Neurology 1997;49:1564-9. 156. Anderson VC, Burchiel KJ, Hogarth P, Favre J, Hammerstad JP. Pallidal vs. subthalamic nucleus deep brain stimulation in Parkinson disease. Arch Neurol 2005;62:554-60. 157. Smeding HMM, Esselink RAJ, Schmand B, KoningHaanstra M, Nijhuis I, Wijnalda EM, Speelman JD. Unilateral pallidotomy versus bilateral subthalamic nucleus stimulation in PD: a comparison of neuropsychological effects. J Neurol 2005;252:176-82. 158. Fine J, Duff J, Chen R, Chir B, Hutchison W, Lozano AM, Lang AE. Long-term follow-up of unilateral pallidotomy in advanced Parkinson’s disease. N Engl J Med 2000;342:1708-14. 159. Blomstedt P, Hariz GM, Hariz MI. Pallidotomy versus pallidal stimulation. Parkinsonism Rel Disord 2006;12:296-301. 160. Loher TJ, Burgunder JM, Pohle T, Weber S, Sommerhalder R, Krauss JK. Long-term pallidal deep brain stimulation in patients with advanced Parkinson disease: 1-year follow-up study. J Neurosurg 2002;96:844-53.
History of surgery for movement disorders
161. Troster AI, Fields JA, Wilkinson SB, Pahwa R, Miyawaki E, Lyons KE, Koller WC. Unilateral pallidal stimulation for Parkinson’s disease: neurobehavioral functioning before and 3 months after electrode implantation. Neurology 1997;49:1078-83. 162. Benazzouz A, Gross C, Feger J, Boraud T, Bioulac B. Reversal of rigidity and improvement in motor performance by subthalamic high-frequency stimulation in MPTP-treated monkeys. Eur J Neurosci 1993;5: 382-9. 163. Guridi J, Herrero MT, Luquin MR, Guillen J, Ruberg M, Laguna J, Vila M, Javoy-Agid F, Agid Y, Hirsch E, Obeso JA. Subthalamotomy in parkinsonian monkeys: behavioural and biochemical analysis. Brain 1996;119:1717-27. 164. Krack P, Batir A, Van Blercom N, Chabardes S, Fraix V, Ardouin C, Koudsie A, Limousin PD, Benazzouz A, LeBas JF, Benabid AL, Pollak, P. Five-year follow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinson’s disease. N Engl J Med 2003;349:1925-34. 165. Kumar R, Lang AE, Rodriguez-Oroz MC, Lozano AM, Limousin P, Pollak P, Benabid AL, Guridi J, Ramos E, van der Linden C, Vandewalle A, Caemaert J, Lannoo E, van den Abbeele D, Vingerhoets G, Wolters M, Obeso JA. Deep brain stimulation of the globus pallidus pars interna in advanced Parkinson’s disease. Neurology 2000;55(12 Suppl 6):S34-39. 166. Tasker RR, Doorly T, Yamashiro K. Thalamotomy in generalized dystonia. In: Fahn S, Marsden C, Caine DB, editors. Advances in neurology, Dystonia 2. vol. 5. New York: Raven Press; 1988. p. 615-31. 167. Benabid AL, Pollak P, Gross C, Hoffmann D, Benazzouz A, Gao DM, Laurent A, Gentil A, Perret J. Acute and long-term effects of subthalamic nucleus stimulation in Parkinson’s disease. Stereotact Funct Neurosurg 1994;62:76-84. 168. Bergman H, Wichmann T, DeLong MR. Reversal of experimental parkinsonism by lesions of the subthalamic nucleus. Science 1990;249:1436-8. 169. Aziz TZ, Peggs D, Sambrook MA, Crossman AR. Lesion of the subthalamic nucleus for the alleviation of 1-methyl-4-phenyl-1,2,3,6 tetrahydropyridine (MPTP)induced parkinsonism in the primate. Mov Disord 1992;6:288-92. 170. Bergman H, Wichmann T, DeLong MR. Amelioration of parkinsonian symptoms by inactivation of the subthalamic nucleus in MPTP treated monkeys. Mov Disord 1990;5:284. 171. Brotchie JM, Mitchel IJ, Sambrook MA, Crossman AR. Alleviation of parkinsonism by antagonism of excitatory amino acid transmission in the medial segment of the globus pallidus in rat and primate. Mov Disord 1991;6:133-8. 172. Benazzouz A, Gross C, Feger J, et al. Reversal of rigidity and improvement in motor performance by subthalamic
173.
174.
175.
176.
177.
178.
179.
180.
181.
182.
183.
87
high frequency stimulation in MPTP treated monkeys. Eur J Neurosci 1993;5:382-9. Brotchie JM, Mitchell IJ, Sambrook MA, Crossman AR. Alleviation of parkinsonism by antagonism of excitatory amino acid transmission in the medial segment of the globus pallidus in rat and primate. Mov Disord 1991;6:133-8. DeLong MR, Crutcher MD, Georgopoulos AP. Primate globus pallidus and subthalamic nucleus: functional organization. J Neurophysiol 1985;53:530-43. Goodman RR, Kim B, McClelland SIII, Senatus PB, Winfield LM, Pullman SL, Yu Q, Ford B, McKhann GMII. Operative techniques and morbidity with subthalamic nucleus deep brain stimulation in 100 consecutive patients with advanced Parkinson’s disease. J Neurol Neurosurg Psychiatry 2006;77:12-17. Rodriguez-Oroz MC, Obeso JA, Lang AE, Houeto JL, Pollak P, Rehncrona S, Kulisevsky J, Albanese A, Volkmann J, Hariz MI, Quinn NP, Speelman JD, Guridi J, Zamarbide I, Gironell A, Molet J, PascualSedano B, Pidoux B, Bonnet AM, Agid Y, Xie J, Benabid AL, Lozano AM, Saint-Cyr J, Romito L, Contarino MF, Scerrati M, Fraix V, Van Blercom N. Bilateral deep brain stimulation in Parkinson’s disease: a multicentre study with 4 years follow-up. Brain 2005;128:2240-9. The Deep-Brain Stimulation for Parkinson’s Disease Study Group. Deep-brain stimulation of the subthalamic nucleus or the pars interna of the globus pallidus in Parkinson’s disease. N Engl J Med 2001;345:956–63. Esselink RAJ, de Bie RMA, de Haan RJ, Steur ENHJ, Beute GN, Portman AT, Schuurman PR, Bosch DA, Speelman JD. Unilateral pallidotomy versus bilateral subthalamic nucleus stimulation in Parkinson’s disease: one year follow-up of a randomised observer-blind multi centre trial. Acta Neurochir (Wien) 2006;148:1247-55. Graff-Radford J, Foote KD, Rodriguez RL, Fernandez HH, Hauser RA, Sudhyadhom A, Rosado CA, Sanchez JC, Okun MS. Deep brain stimulation of the internal segment of the globus pallidus in delayed runaway dyskinesia. Arch Neurol 2006;63:1181-4. Alvarez L, Macias R, Lopez G, Alvarez E, Pavon N, Rodriguez-Oroz MC, Juncos JL, Maragoto C, Guridi J, Litvan I, Tolosa ES, Koller W, Vitek J, DeLong MR, Obeso JA. Bilateral subthalamotomy in Parkinson’s disease: initial and long-term response. Brain 2005;128:570-83. Su PC, Tseng HM, Liu HM, Yen RF, Liou HH. Subthalamotomy for advanced Parkinson disease. J Neurosurg 2002;97:598-606. Patel NK, Heywood P, O’Sullivan K, McCarter R, Love S, Gill SS. Unilateral subthalamotomy in the treatment of Parkinson’s disease. Brain 2003;126:1136-45. Limousin P, Pollak P, Benazzouz A, et al. Bilateral subthalamic nucleus stimulation for severe Parkinson’s disease. Mov Disord 1995;10:672-4.
1483
1484
87
History of surgery for movement disorders
184. Olanow CW, Goetz CG, Kordower JH, Stoessl AJ, Sossi V, Brin MF, Shannon KM, Nauert GM, Perl DP, Godbold J, Freeman TB. A Double-blind controlled trial of bilateral fetal nigral transplantation in Parkinson’s disease. Ann Neurol 2003;54:403-14. 185. Vilela Filho O, da Silva DJ. Unilateral subthalamic nucleus lesioning. A safe and effective treatment for Parkinson‘s disease. Arq Neuropsiquiatr 2002;60(4): 935-48. 186. Mendez I, Sanchez-Pernaute R, Cooper O, Vinuela A, Ferrari D, Bjorklund L, Dagher A, Isacson O. Cell type analysis of functional fetal dopamine cell suspension transplants in the striatum and substantia nigra of patients with Parkinson’s disease. Brain 2005; 128:1498-510. 187. Piccini P, Pavese N, Hagell P, Reimer J, Bjorklund A, Oertel WH, Quinn NP, Brooks DJ, Lindvall O. Factors affecting the clinical outcome after neural transplantation in Parkinson’s disease. Brain 2005;128:2977-86. 188. Freed CR, Greene PE, Breeze RE, Tsai WY, DuMouchel W, Kao R, Dillon S, Winfield H, Culver S, Trojanowski JQ, David D, Eidelberg D, Fahn S. Transplantation of embryonic dopamine neurons for severe Parkinson’s disease. N Engl J Med 2001;344:710-9. 189. Greene PE, Fahn S. Status of fetal tissue transplantation for the treatment of advanced Parkinson disease. Neurosurg Focus 2002;13(5):e3. 190. Lindvall O, Brundin P, Widner H, et al. Grafts of fetal dopamine neurons survive and improve motor function in Parkinson’s disease. Science 1990;247:574-7. 191. Bakay RAE. Selection criteria for CNS grafting into Parkinson’s disease patients. In: Lindvall O, Bjorkland A, Widner H, editors. Intracerebral transplantation in movement disorders. Amsterdam: Elsevier; 1991. p. 137-8. 192. Goetz CG, Stebbing GT, Klawans HL, et al. United Parkinson foundation neurotransplantation registry on adrenal medullary transplant: presurgical, and I-year and 2-year follow up. Neurology 1991;41:1719-22. 193. Lindvall O, Wichner H, Rehncrona S, et al. Transplantation of fetal dopamine neurons in Parkinson’s disease: one-year clinical and neurophysiological observation in two patients with putaminal implants. Ann Neurol 1992;31:155-65. 194. Sawle G, Bloomfield PM, Bjorklund A, et al. Transplantation of fetal dopamine neurons in Parkinson’s disease: PET [18F]6-L-fluorodopa studies in two patients with putaminal implants. Ann Neurol 1992;31:166-73. 195. Wichner H, Tetrud J, Rehncrona S, et al. Bilateral fetal mesencephalic grafting in two patients with parkinsonism induced by 1-methyl-4-phenyl-l,2,3,6tetrahydropyridine (MPTP). N Engl J Med 1992; 327:1556-63. 196. Alterman RL, Miravite J, Weisz D, Shils JL, Bressman SB, Tagliati M. Sixty hertz pallidal stimulation deep brain
197.
198.
199.
200.
201.
202.
203.
204.
205.
206.
207.
stimulation for primary torsion dystonia. Neurology 2007;69:681-8. Vidailhet M, Vercueil L, Houeto JL, Krystkowiak P, Lagrange C, Yelnik J, Bardinet E, Benabid AL, Navarro S, Dormont D, Grand S, Blond S, Ardouin C, Pillon B, Dujardin K, Hahn-Barma V, Agid Y, Deste´e A, Pollak P, and The French SPIDY Study Group. Bilateral, pallidal, deep-brain stimulation in primary generalised dystonia: a prospective 3 year follow-up study. Lancet Neurol 2007;6:223-9. Kupsch A, Benecke R, Mu¨ller J, Trottenberg T, Schneider GH, Poewe W, Eisner W, Wolters A, Mu¨ller JU, Deuschl G, Pinsker MO, Skogseid IM, Roeste GK, Vollmer-Haase J, Brentrup A, Krause M, Tronnier V, Schnitzler A, Voges J, Nikkhah G, Vesper J, Naumann M, Volkmann J, Deep-Brain Stimulation for Dystonia Study Group. Pallidal deep-brain stimulation in primary generalized or segmental dystonia. N Engl J Med 2006;355:1978-90. Vidailhet M, Vercueil L, Houeto JL, Krystkowiak P, Benabid AL, Cornu P, Lagrange C, Te´zenas du Montcel S, Dormont D, Grand S, Blond S, Detante O, Pillon B, Ardouin C, Agid Y, Deste´e A, Pollak P, for the French Stimulation du Pallidum Interne dans la Dystonie (SPIDY) Study Group. Bilateral deep-brain stimulation of the globus pallidus in primary generalized dystonia. N Engl J Med 2005;352:459-67. Vercueil L, Krack P, Pollak P. Results of deep brain stimulation for dystonia: a critical reappraisal. Mov Disord 2002;17 Suppl 3:S89-93. Cersosimo MG, Raina GB, Piedimonte F, Antico J, Graff P, Micheli FE. Pallidal surgery for the treatment of primary generalized dystonia: long-term follow-up. Clin Neurol Neurosurg 2008;110:145-50. Shima F, Iacono R, Sakata S, et al. Mechanisms of posteroventral pallidotomy in Parkinson’s disease. XIth Meeting of WSSFN, Ixtapa, Abstract 123, 1993. Nguyen JP, Geny C, Yepes C, et al. Chronic thalamic stimulation for the relief of severe postural cerebellar tremor: a series of 11 cases. Mov Disord 1992; Suppl 1:159. Tsubokawa T, Yamamoto T, Katayama Y. Chronic thalamic stimulation for relief of ballistic involuntary movement. XIth Meeting of WSSFN, Ixtapa, Abstract 144, 1993. Madrazo I, Drucker-Colin R, Diaz V, Martinez-Mata J. Open microsurgical autograft of adrenal medulla to the right caudate nucleus in two patients with intractable Parkinson’s disease. N Engl J Med 1987; 316:831-4. Bergman H, Wichmann T, DeLong MR. Reversal of experimental parkinsonism by lesions of the subthalamic nucleus. Science 1990;249(4975):1436-8. Lang AE, Duff J, Saint-Cyr JA, Trepanier L, Gross RE, Lombardi W, Montgomery E, Hutchinson W,
History of surgery for movement disorders
208. 209.
210.
211.
Lozano AM. Posteroventral medial pallidotomy in Parkinson’s disease. J Neurol 1999;246 Suppl 2:II28-41. Lozano AM, Lang AE. Pallidotomy for Parkinson’s disease. Adv Neurol 2001;86:413-20. Lang AE, Lozano AM, Montgomery E, Duff J, Tasker R, Hutchinson W. Posteroventral medial pallidotomy in advanced Parkinson’s disease. N Engl J Med 1997;337:1036-42. Scott R, Gregory R, Hines N, Carroll C, Hyman N, Papanasstasiou V, Leather C, Rowe J, Silburn P, Aziz T. Neuropsychological, neurological and functional outcome following pallidotomy for Parkinson’s disease. A consecutive series of eight simultaneous bilateral and twelve unilateral procedures. Brain 1998;121(Pt 4):659-75. Intemann PM, Masterman D, Subramanian I, De Salles A, Behnke E, Frysinger R, Bronstein JM.
87
Staged bilateral pallidotomy for treatment of Parkinson disease. J Neurosurg 2001;94:437-44. 212. York MK, Lai EC, Jankovic J, Macias JA, Atassi F, Levin HS, Grossman RG. Short and long-term motor and cognitive outcome of staged bilateral pallidotomy: a retrospective analysis. Acta Neurochir (Wien) 2007; 149:857-66. 213. Kumar R, Lang,AE, Rodriguez-Oroz MC, Lozano AM, Limousin P, Pollak P, Benabid AL, Guridi J, Ramos E, van der Linden C, Vandewalle A, Caemaert J, Lannoo E, van den Abbeele D, Vingerhoets G, Wolters M, Obeso JA. Deep brain stimulation of the globus pallidus pars interna in advanced Parkinson’s disease. Neurology 2000;55 Suppl 6:S34-9.
1485
103 Intraparenchymal Drug Delivery for Parkinson’s Disease R. D. Penn . A. A. Linninger
Just as focal stimulation of the basal ganglion structures can improve symptoms of Parkinson’s disease, so can focal infusion of medications. This has been demonstrated in a series of tests conducted in patients who were undergoing stereotactic creation of a lesion or placement of an electrode for Parkinson’s disease. In the first experiment performed, drug was infused into the globus pallidus just prior to pallidotomy [1]. Muscimol, a GABA-A agonist was injected over 5 min while the patient performed a simple motor movement of opening and closing his thumb and second finger. As > Figure 103-1 shows, prior to infusion the motion was slow, irregular and fatigued rapidly. Ten minutes after the infusion, the patient’s movement speed increased and he showed less fatigue. Examination of the patient’s arm and wrist demonstrated reduced tone. Of note was the small volume of infusion (2.5 mL) and low dose (2.5 mg) Also of importance was the time delay between infusion and the onset of improvements and then the slow return to baseline over the next 30 min. Once the pallidotomy was performed, the same improvements in the movement tasks were seen. Later experiments by a second group of investigators extended these observations to other sites [2]. Muscimol slowly infused into the thalamus (VIM) dramatically stopped tremor with the same time course as seen in the pallidum. Likewise, muscimol injected into STN improved tremor. Electrophysiological recordings (see > Figure 103-2), confirmed that the drug was acting as expected by inhibiting electrical activity in the nucleus as it spread from the injection site to the more distant tissue. #
Springer-Verlag Berlin/Heidelberg 2009
Although muscimol infusion and electrical stimulation appeared in these experiments to have the same motoric effects, they operate by very different mechanisms. Electrical stimulation has many diverse effects on neural tissue. It causes excitation of nearby axons and activates recurrent inhibitory circuits as well as directly inhibiting neuronal firing [3]. High frequency stimulation also entrains neuronal firing going to distant deep nuclei and cortical regions. Muscimol in contrast works on the GABA-A receptors and presynaptically inhibits neuronal firing. Whether continuously infused muscimol would be better than stimulation has not yet been tested, but the two modalities work by very different physiological mechanisms and are likely to have different positive and negative clinical affects. A key question yet to be answered is if continuous infusion of muscimol will lead to tolerance to the drug. Human experiments have not been done but relevant rodent tests suggest muscimol will continue to provide inhibition. Rats with unilateral 5-hydroxy dopamine lesions to mimic Parkinson’s disease showed sustained effects on turning when muscimol was given via an implanted osmotic pump for 2 weeks (unpublished results Kang and Penn) (> Figure 103-3). These proof of principle experiments show that a local infusion of muscimol can be used to inhibit specific groups of neurons and might improve motor performance in movement disorders. However, this treatment has not been pursued clinically. No pharmaceutical company has been willing to produce a medical grade formulation and test it in humans. Muscimol is an old medication which cannot be patented
1732
103
Intraparenchymal drug delivery for parkinson’s disease
. Figure 103-1 Injection of muscimol into the GPi of a patient with Parkinson’s disease. (a) and (b) show the opening and closing of the patients first and second fingers. The movements after muscimol injection are more rapid and uniform, and do not fatigue [1]
and it might not be better than the already well established standard DBS treatment. The most likely pathway for its development would be for other purposes such as the treatment of epilepsy or behavioral disorders. If muscimol were found to be effective for those disorders, it would then become available for trials in motor disorders. Could dopaminergic medications be given selectively to the basal ganglion areas, thus avoiding side effects due to systemic administration? This application is much more difficult than providing a drug which selectively inactivates a small well localized malfunctioning basal ganglion nucleus. Dopamine or dopamine agonists would have to reach wide areas of the striatum to effectively mimic the distribution of natural
dopamine release. As will be discussed, diffusion in the brain can only be expected to distribute a medication a millimeter or two away from the point that it is introduced by catheter [4]. Also, dopamine is rapidly metabolized and this further limits distribution. Similarly, infusion of dopamine into the CSF would be ineffective because the distribution would be to the superficial periventricular regions rather than the deep basal ganglion structures. Another approach to Parkinson’s disease that requires parenchymal delivery is the use of neurotrophins. The goal in neurotrophin delivery is twofold: to rescue dying dopaminergic neurons and to increase dopaminergic activity. The promise, as yet unproven, is that neurotrophins will not
Intraparenchymal drug delivery for parkinson’s disease
. Figure 103-2 The slow diffusion of drug into the STN. The infusion of lidocaine decreases neural firing in the cells near the injection site but not 1.2 mm away. Note that the cell 0.6 mm away does not stop firing until 3.5 mL have been injected and 5 min have past. The cell 1.20 mm away does not stop firing even after 11 min [2]
only provide symptomatic relief by increasing available dopamine, but also decrease the rate of loss of neurons. This is different from all other treatments for Parkinson’s disease which are symptomatic only and require continuous dosing (some investigators have suggested that DBS slows neural degeneration but this has not been proven in clinical trials. See Chapter X).
103
. Figure 103-3 The continuing effects of chronic muscimol infusion. Rats with 5-hydroxy dopamine lesions are infused with muscimol into the STN for 4 weeks. Their stepping response is then tested after a bolus IV dose of LDopa. Note that the muscimol group improves more than the control group (Ding, Y, Kang U, Penn R, unpublished results)
The experimental background for using neurotrophins in Parkinson patients has been thoroughly reviewed in the literature [5]. Neurotrophins are large proteins which do not cross the blood brain barrier. Various ways to get them to their site of action have been utilized which include gene therapy with in vivo viral vectors or implantation of genetically modified cells, intranasal administration, implantation of columnar release systems, or direct infusion into the CSF or brain tissue [6]. The most extensively studied neurotrophin in man is glial derived neurotrophic factor (GDNF). The experiments with the infusion of this molecule provide a good example of the potential for this type of treatment and for its inherent problems. GDNF is a member of the larger transforming growth factor/beta family and it is closely related to neurturin and persephin. It has a strong trophic influence on dopaminergic cells [7] and is important in the development and maintenance of their function. Both rodent and primate models of Parkinson’s disease respond to infusion of GDNF [8,9]. Histological studies
1733
1734
103
Intraparenchymal drug delivery for parkinson’s disease
show neuroprotective effects as well as fiber outgrowth when GDNF is infused into the striatum or the substantia nigra. Most importantly, the motor symptoms in these Parkinson-like animals are markedly decreased. In the primate studies, both intraparenchymal and intraventricular infusions have been performed and appeared equally effective. This observation led to the first GDNF study in Parkinson patients in which intermittent ventricular injections were given. The results were discouraging. Patients had nausea and vomiting and sensory dysesthesias associated with the injections and no improvement in their Parkinsonian
symptoms [10]. In retrospect, it is clear that GDNF was unable to reach the striatum or nigral cells from the ventricular surface and the flooding of the CSF spaces with high doses of GDNF resulted in side effects. In fact, an autopsy on one of the treated patients who died of unrelated causes showed none of the changes seen in the animal studies (see > Figure 103-4a) [11]. To test whether more direct delivery of GDNF to the striatum would work, Gill and coworkers used implanted drug pumps connected to catheters stereotactically placed into the putamen bilaterally [12]. GDNF was initially infused continuously at 14 mcg a day and then raised to
. Figure 103-4 Lack of effect of intraventricular GDNF on dopamine cells (TH-staining) compared to a marked increase with parenchymal infusion. (a) shows the putamen in a patient with Parkinson’s disease who was given multiple intraventricular injections of GDNF and no significant TH staining is seen [11]. (b) Arrow shows site of the catheter injection of GDNF and the dark TH staining around it [13]
Intraparenchymal drug delivery for parkinson’s disease
higher levels over the following months. Major improvements in the UPDRS (United Parkinson Disease Rating Scale) were seen as shown in > Figure 103-5 for the first year of treatment and then maintained for another year. PET scans to measure dopamine uptake demonstrated a marked increase in the region of the infusion catheter tip (> Figure 103-6). One patient died later of a heart attack and his autopsy clearly showed increase in dopaminergic cells and streaming of neurons towards the site of infusion [13], as was seen in the animal histology, but not seen when GDNF was given into the CSF (see > Figure 103-4 compare a and b). A confirmatory open label phase II study was done by Gash’s group in ten patients using convection enhanced delivery to only one putamen. The improvements over the year study period were similar to those of Gill [14]. Interestingly in both studies the clinical improvement took several months to occur suggesting a slow biological process. When GDNF was withdrawn several years later, the patients remained well for many months before slowly deteriorating. The next step in the investigation of GDNF was a phase two double blind study of 34 patients sponsored by Amgen. The results of the six month constant infusion were negative [15]. The primary outcome measure of 25% or better improvement in ‘‘off ’’ motor scores was not met. Interestingly, the secondary measure of increase dopamine uptake near the catheter as measured by fluorodopa PET scanning was strongly positive. Thus, the study seemed to indicate that a local biological effect occurred but no corresponding clinical improvement. Why the controlled study was unable to duplicate the results of the first two trials is unclear and various explanations have been proposed. The obvious one is that the blinded study was more objective and GDNF simply does not work. However, analysis of the double blind study shows it to be underpowered due unanticipated variability of the baseline scores. Over 300
103
patients would have to have been tested to show a significance change [16]. Most importantly, there were major differences in infusion rates, drug doses, catheter designs, and targeting between the three studies [17]. Thus too much variability in GDNF delivery to the putamen could be the cause of the discordant results. Unfortunately, the explanation is likely to remain a mystery because GDNF has been withdrawn from further human study by the drug company which controls the patent (for a discussion of the scientific and public reactions to GDNF withdrawal, see Hsieh, Penn 2008). The experience with GDNF underlines the importance of understanding distribution of the drug when it is directly infused into the brain tissue. Stereotaxic techniques are now available to place catheters or needles for acute or chronic infusion of medications into specific targets practically anywhere in the brain. Getting a drug to cover the whole targeted region is much more difficult. Tools to predict in a given patient the likely distribution for an infused medication are currently being developed and are still in the experimental stage [18]. The basic principles are clear. If a molecule is introduced into the extracellular space of the gray matter its movement will be driven by the physics of diffusion and size and configuration of the tortuous spaces between cells [6]. For a molecule such as inulin that remains within the extracellular space and is not eliminated by metabolism or uptake into cells or capillaries, the rate of diffusion depends on its concentration and the tissue porosity and tortuosity. The higher the concentration difference the greater the driving force and the more porous and less tortuosity the faster the diffusion. In practice, diffusion is very slow in the gray matter. It takes a water soluble molecule over 12 times as long to move a given distance in the brain as it does in agar solution [19]. Calculations and in vivo experiments show that for molecules infused at rates which do not cause bulk flow, the concentration will be reduced
1735
1736
103
Intraparenchymal drug delivery for parkinson’s disease
. Figure 103-5 Improvement in five Parkinson patients infused with intraparenchymal GDNF. The total UPDRS, the motor scores and the activities of daily living when ‘‘on’’ or ‘‘off’’ L-dopa therapy. The effect of treatment was seen by 3 months but continued improvement is seen to 6–12 months and then appears to stabilize [12]
Intraparenchymal drug delivery for parkinson’s disease
103
. Figure 103-6 GDNF infusion increases F-dopa turnover. (a) PET image of a patient with unilateral Parkinson’s disease with decreased uptake on left. (b) Same patient after 12 months of GDNF infusion into the putamen. The circle shows the region of interest. Coronal scans showing increased uptake after treatment in the substantia nigra (c) and putamen (d) [12]
100-fold by 1–2 mm from the injection site. Thus, the volume effectively perfused is less than 10 mm cubed. The size of human STN is approximately 3 3 6 mm or 54 cu mm. The putamen is many times larger. These considerations alone indicate how diffusive flow is unlikely to be very effective when used for large targets but is best applied in applications where a small nucleus can be targeted. That is why dopamine infusion to wide regions of the basal ganglion will not succeed but using muscimol to reduce neural activity of the ventral third of the STN might be useful. It should be pointed out that diffusion from the CSF can be useful if one waits long enough
for a steady state to be reached and if the needed concentration is very low. Octreotide, a somatostatin analog, was screened as a potential treatment for Alzheimer’s disease and slow IVC infusion created a level as high as naturally occurring somatostatin [20]. In an effort to increase the size of region that is perfused, convection techniques have been introduced [21–24]. The basic concept is to move fluid and the drug in it through the extracellular space by bulk flow. The primary driving force is a pressure gradient induced by the flow of fluid pumped out of a catheter. The molecule to be distributed is carried within the fluid mass.
1737
1738
103
Intraparenchymal drug delivery for parkinson’s disease
Unlike diffusion in which the size of the molecule is a key parameter determining its rate of movement, large and small sized molecules are convected in similar fashion. Thus, convective flow can increase the volume of distribution of large as well as small molecules. The key to using convective flow is determining the flow rates that are high enough to induce convective transport but not so large as to damage areas around the catheter or distant brain tissue. Gash found that short bursts of rapid of flow can produce convection [25]. Using his technique the distribution of GDNF in monkeys was substantially increased compared to diffusive flow. Gash also found that the larger the area of distribution of GDNF in the striatum, the larger the histological and clinical effect. The differences in distribution and response in the animal experiments could explain the difference between the double blind trial which did not use convection flow and that of the open unilateral trial that did. Another factor that must be considered in drug distribution is catheter design. Large diameter catheters allow fluid to leak back along the insertion tract to other regions of the brain or to the brain surface [17,26]. The physics of fluid flow out of the catheter is also important. A single end hole catheter creates a jet stream
downward through the tissue which may change the shape of the drug distribution and potentially injury tissue. Experiments have shown a larger and less variable flow when multi-hole catheters have been used [27]. As the distribution volume becomes greater with convective flow, the non-homogeneous structure of the brain and white matter tracts influence drug movement. White matter has a much lower resistance to flow of water along the axons than across them [28]. This means that flow which reaches a white matter tract will be carried preferentially along the tract. > Figure 103-7 shows a point infusion going from the deep gray structures to the white matter above and then along it [29]. To predict the distribution accurately, the anisotropic and heterogeneous properties of the brain must be considered. Fortunately, these can be measured using MRI techniques [18]. > Figure 103-8 shows an example of a DTI image of the human brain and the anisotropy in various regions. Finally, the binding properties, charge, size and metabolic pathways of the molecule must be known. Two examples illustrate the problems. Liposomes with a diameters greater than 100 mm. will be blocked by the extracellular space and so will not move beyond the region
. Figure 103-7 Infusion of GDNF from a catheter in the striatum of a rat, with and without co-infusion with heparin. A shows CED which goes upward along the catheter and then along the white matter tracts. When heparin (B) is added the CED spreads the GDNF much further [29]
Intraparenchymal drug delivery for parkinson’s disease
103
. Figure 103-8 Axial and coronal DTI measurements a normal subject showing anisotropy of brain tissue. The local diffusion tensor in each voxel of axial and coronal brain sections is represented by an ellipsoid. Four sample locations – (a) corpus callosum, (b) corona radiata, (c) internal capsule, and (d) gray matter regions show the orientation of apparent water diffusion tensor ellipsoids in more detail [18]
of infusion, liposomes less than 50 mcg will. Molecules that are bound by extracellular matrix material may also be limited in distribution. GDNF distribution is much larger if it is coinfused with heparin which blocks the binding of the neurotrophin in the extracellular matrix (see > Figure 103-7b) [29]. All these factors have to be considered when contemplating the use of intraparenchymal delivery. Otherwise, the medication may not reach the intended targets. One approach
to determining drug distribution is to measure distribution of the medication in an appropriate primate model. An educated guess at the type of catheter and its placement must be made, as well as, the infusion protocol. Then a series of empirical tests need to be run. If the distribution is not the desired one, then adjustments to the various parameters have to be made and the experiments re-run. Finally, the results have to be scaled to human dimensions. Clearly this testing can be a slow and expensive process.
1739
103
. Figure 103-9 Overview of computer-assisted brain analysis for the design of drug delivery therapies. The first step entails the collection of medical images from magnetic resonance imaging (MRI). Geometry reconstruction detects sharp boundaries and functional regions inside the brain. Grid generation partitions surfaces and volumes into small tetrahedrons for finite volume discretization of the transport equations. Physical properties such as drug diffusion and hydraulic conductivity tensors are estimated from apparent water diffusion tensor measurement obtained by diffusion tensor imaging (DTI). Computational analysis solves the discretized transport equations to predict the drug distribution [18]
1740 Intraparenchymal drug delivery for parkinson’s disease
Intraparenchymal drug delivery for parkinson’s disease
103
. Figure 103-10 Prediction of a neurotrophic factor distribution in the basal ganglia with single hole catheter in anisotropic heterogeneous tissue model over 3 weeks [18]
Imaging and computational methods can streamline these steps. Such a process is outlined in > Figure 103-9 [18]. First, anatomical MRI and infusion tensor imaging is performed on the patient who is to be treated. Then the brain geometry is placed into a three-dimensional computational grid. Data from experiments in animals is used to provide transport and kinetic characteristics of the molecule to be infused. The DTI data from the patient is used to measure the apparent water diffusion and this information is incorporated into the grid. Finally, the laws of fluid mechanics are used in the grid based simulations which predict drug distribution from a catheter placed in a selected location of the brain. This modeling can also be used to compute where multiple catheters should be placed and the rates of flow to produce an optimum distribution of medication in the targeted area. > Figure 103-10 shows the results of the simulation for a neurotrophin with 4 weeks of conductive flow from a point source. In spite of the significant advantage of using advanced computational methods to accurately predict drug distribution, much work needs to be done to fully implement and make practical a program for general use. Such programs will
have to be test by comparing their predictions to actual drug distribution in animal and human infusions. This is the same type of development used to perfect radiation therapy for tumors with complex configurations and tissue characteristics. The fact that such techniques for radiation treatment have been developed and are widely used suggests that drug delivery could undergo the same evolution. Hopefully, such tools will provide the rational basis for designing drug infusion protocols. As the GDNF trials illustrate until we can predict the distribution of a drug accurately, we will not be able to adequately test it in clinical studies. Considerable work needs to be done both in the laboratory and in patients to determine whether drug delivery directly into the brain will help Parkinson patients and change the natural history of disease.
References 1. Penn RD, Kroin JS, et al. Injection of GABA-agonist into globus pallidus in patient with Parkinson’s disease. Lancet 1998;351(9099):340-1. 2. Levy R, Lang AE, et al. Lidocaine and muscimol microinjections in subthalamic nucleus reverse Parkinsonian symptoms. Brain 2001;124(Pt 10):2105-18.
1741
1742
103
Intraparenchymal drug delivery for parkinson’s disease
3. Dostrovsky JO, Levy R, et al. Microstimulation-induced inhibition of neuronal firing in human globus pallidus. J Neurophysiol 2000;84(1):570-4. 4. Kroin JS, Kao LC, et al. Dopamine distribution and behavioral alterations resulting from dopamine infusion into the brain of the lesioned rat. J Neurosurg 1991;74(1):105-11. 5. Hurelbrink CB, Barker RA. The potential of GDNF as a treatment for Parkinson’s disease. Exp Neurol 2004;185(1):1-6. 6. Thorne RG, Frey WH II. Delivery of neurotrophic factors to the central nervous system: pharmacokinetic considerations. Clin Pharmacokinet 2001;40(12): 907-46. 7. Lin LF, Doherty DH, et al. GDNF: a glial cell line-derived neurotrophic factor for midbrain dopaminergic neurons. Science 1993;260(5111):1130-2. 8. Gash DM, Zhang Z, et al. Functional recovery in parkinsonian monkeys treated with GDNF. Nature 1996;380(6571):252-5. 9. Hoffer BJ, Hoffman A, et al. Glial cell line-derived neurotrophic factor reverses toxin-induced injury to midbrain dopaminergic neurons in vivo. Neurosci Lett 1994;182 (1):107-11. 10. Nutt JG, Burchiel KJ, et al. Randomized, double-blind trial of glial cell line-derived neurotrophic factor (GDNF) in PD. Neurology 2003;60(1):69-73. 11. Kordower JH, Palfi S, et al. Clinicopathological findings following intraventricular glial-derived neurotrophic factor treatment in a patient with Parkinson’s disease. Ann Neurol 1999;46(3):419-24. 12. Gill SS, Patel NK, et al. Direct brain infusion of glial cell line-derived neurotrophic factor in Parkinson disease. Nat Med 2003;9(5):589-95. 13. Love S, Plaha, P, et al. Glial cell line-derived neurotrophic factor induces neuronal sprouting in human brain. Nat Med 2005;11(7):703-4. 14. Slevin JT, Gerhardt GA, et al. Improvement of bilateral motor functions in patients with Parkinson disease through the unilateral intraputaminal infusion of glial cell line-derived neurotrophic factor. J Neurosurg 2005; 102(2):216-22. 15. Lang AE, Gill S, et al. Randomized controlled trial of intraputamenal glial cell line-derived neurotrophic factor infusion in Parkinson disease. Ann Neurol 2006; 59(3):459-66.
16. Hutchinson M, Gurney S, et al. GDNF in Parkinson disease: an object lesson in the tyranny of type II. J Neurosci Methods 2007;163(2):190-2. 17. Morrison PF, Lonser RR, et al. Convective delivery of glial cell line-derived neurotrophic factor in the human putamen. J Neurosurg 2007;107(1):74-83. 18. Linninger AA, Somayaji MR, et al. Prediction of convection-enhanced drug delivery to the human brain. J Theor Biol 2008;250(1):125-38. 19. Nicholson C, Sykova E. Extracellular space structure revealed by diffusion analysis. Trends Neurosci 1998;21 (5):207-15. 20. Kroin JS, O’Dorisio, TM, et al. Distribution of a somatostatin analog after continuous intraventricular administration. Neurosurgery 1989;24(5):744-8. 21. Lieberman DM, Laske DW, et al. Convection-enhanced distribution of large molecules in gray matter during interstitial drug infusion. J Neurosurg 1995;82(6):1021-9. 22. Nguyen TT, Pannu YS, et al. Convective distribution of macromolecules in the primate brain demonstrated using computerized tomography and magnetic resonance imaging. J Neurosurg 2003;98(3):584-90. 23. Raghavan R, Brady ML, et al. Convection-enhanced delivery of therapeutics for brain disease, and its optimization. Neurosurg Focus 2006;20(4):E12. 24. Bobo RH, Laske DW, et al. Convection-enhanced delivery of macromolecules in the brain. Proc Natl Acad Sci USA 1994;91(6):2076-80. 25. Gash DM, Zhang Z, et al. Trophic factor distribution predicts functional recovery in parkinsonian monkeys. Ann Neurol 2005;58(2):224-33. 26. Krauze MT, Saito R, et al. Reflux-free cannula for convection-enhanced high-speed delivery of therapeutic agents. J Neurosurg 2005;103(5):923-9. 27. Oh S, Odland R, et al. Improved distribution of small molecules and viral vectors in the murine brain using a hollow fiber catheter. J Neurosurg 2007;107(3):568-77. 28. Sarntinoranont M, Chen X, et al. Computational model of interstitial transport in the spinal cord using diffusion tensor imaging. Ann Biomed Eng 2006;34(8):1304-21. 29. Hamilton JF, Morrison PF, et al. Heparin coinfusion during convection-enhanced delivery (CED) increases the distribution of the glial-derived neurotrophic factor (GDNF) ligand family in rat striatum and enhances the pharmacological activity of neurturin. Exp Neurol 2001;168(1):155-61.
117 Intrathecal Drugs for Spasticity R. D. Penn
Intrathecal Baclofen has proved to be extremely effective in the treatment of severe spasticity. The idea of using medication intrathecally came from the observation that patients who had significant side effects from oral morphine could be treated effectively via the spinal route. The high cerebral spinal fluid (CSF) concentration at the cord level produces excellent analgesia and avoids drowsiness and nausea because little morphine gets to the supraspinal levels. In an analogous way, oral baclofen therapy often is limited by central side effects before a significant reduction in spasticity occurs. However, if the drug is introduced into the lumbar space, a complete reversal of spasticity is achieved without central side effects [1]. If one uses an implantable drug pump and catheter system to deliver baclofen continuously, the therapeutic effect can then be maintained [2]. This technique has been used for over 25 years and has proved to be successful in treating many type of spasticity. Baclofen is the most effective oral antispastic agent available. It was designed to mimic the actions of gamma amino butyric acid (GABA), one of the major inhibitor neurotransmitters in the nervous system. It binds to GABA-B receptors that are found throughout the neuro-axis but are particularly abundant in Rexed lamina II and III in the spinal cord [3]. Neurophysiological studies have shown that when an isolated spinal cord is perfused with baclofen, both monosynaptic and polysynaptic reflexes are decreased [4]. Most of its action occurs at the presynaptic terminals, decreasing calcium influx and consequently reducing neurotransmitter release, although it also increases postsynaptic potassium conductance [5]. These perfusion #
Springer-Verlag Berlin/Heidelberg 2009
studies on the normal spinal cord have been extended to rats with chronic spinal cord injuries that result in hyperactive mono and polysynaptic motor reflexes. When low doses of baclofen are applied to the spinal cord, these reflex pathways are suppressed without significantly changing the inward currents of the motor neurons. This means that baclofen works primarily presynaptically in the injured spinal cord Li et al. [6]. When baclofen is taken orally, the slightly lipid soluble molecule can penetrate the blood brain barrier. Unfortunately, oral baclofen has two drawbacks. First the penetration of the blood brain barrier is poor; the ratio of drug in the plasma to the CSF is variable but is often less than 1:10 [7]. Second, as the oral doses increase, central side effects of drowsiness and confusion occur and limit the total dose that can be given. For most severely spastic patients, central toxicity is reached well before significant relief of spasticity is achieved. For example, Ashworth measurements of spasticity (see > Table 117-1) rarely change by more than a single point on the scale of 5 with oral baclofen. The problem of adequate delivery to the nervous system without causing side effects is solved by the use of direct lumbar injection. Baclofen given in the lumbar subarachnoid space produces a locally high concentration at the spinal level and low concentration supraspinally [8]. The first tests in three patients demonstrated that a 50 mcg bolus of intrathecal baclofen was enough to reduce Ashworth scores from a range of 4–5 to 1 (normal tone) for 6–12 h (see > Figure 117-1). After this initial success with a bolus dose these patients had programmable infusion pumps implanted with a catheter going into the subarachnoid
1974
117
Intrathecal drugs for spasticity
. Table 117-1 Spasticity and spasm rating scales Rigidity: Ashworth scale 1 No increase in tone 2 Slight increase in tone, giving a ‘‘catch’’ when affected parts are moved in flexion or extension 3 More marked increase in tone but affected parts are easily flexed 4 Considerable increase in tone; passive movement difficult 5 Affected parts rigid in flexion or extension Spasm Spasms are measured by the number of sustained flexor and extensor muscle spams over a 1-h period 0 None 1 No spontaneous spasms; vigorous sensory and motor stimulation results in spasms 2 Occasional spontaneous spasms and easily induced spasms 3 More than one but fewer than ten spontaneous spasms per hour 4 More than ten spontaneous spasms per hour
. Figure 117-1 The effect of 50 mcg of intrathecal baclofen on the first patient tested. The patient’s severe rigidity was completely gone by 1 h and did not begin to return for 6 h
space. When baclofen was given continuously, the patients’ hyperactive stretch reflexes, increased muscle tone and spasms were all eliminated. Several of the initial patients are still being successfully treated 25 years later. Many single and multicenter studies have been published that confirmed these original observations. The first trials were on spinal cord
injury and multiple sclerosis patients. Then trials were extended to patients with cerebral palsy and cerebral lesions causing spasticity. > Table 117-2 lists representative publications. A number of reviews have recently appeared on adult and childhood spasticity in [9] and a large section of Operative neuromodulation, Acta Neurochirugica, Supplement 97,1 (1997) covers spasticity. Intrathecal baclofen has been tried for other nonspastic movement disorders such as primary and secondary dystonia, focal leg dystonias [10] and Stiff-man syndrome [11] and often has been helpful. It also has been suggested for autonomic dysreflexia after severe brain injury [12]. The two clinical measures of outcome used in most of the studies are the Ashworth scale and the spasm score (see > Table 117-1). These are easy to apply and provide a simple way of measuring spastic changes over time. The results of a representative multicenter study on 74 patients with spasticity due to multiple sclerosis or spinal cord injury are shown at the > Figure 117-2 [13]. At baseline, the average Ashworth score was 4 and this was reduced to 1.5 with a bolus intrathecal injection. Continuous intrathecal baclofen via drug pump low Ashworth scores were maintained for the 42 months of the study. The results on spasm scores were similar.
Intrathecal drugs for spasticity
117
. Table 117-2 Representative intrathecal baclofen trials Author
Year
Country
Penn Ochs
1989 1989
Coffey Lazorthes Muller Albright Van Schaeybroeck
1993 1990 1991 2003 2003
USA Germany, Sweden, Belgium, Holland, UK USA France Germany USA Belgium
No. patients 20 28 74 56 211 68 11
The average dose of intrathecal baclofen started at 200 mcg a day and increased over time for the first year or two and then stabilized below 400 mcg a day. These typical results illustrate how effective baclofen is when given intrathecally. No oral medication has ever produced such a large decrease in spasticity. While spasm and Ashworth scores are useful objective ways of following patients receiving intrathecal baclofen, they do not reflect the important changes that occur in activities of daily living. Studies by Loubser and by Parke have found that a wide variety of activities of daily living are improved with treatment [14,15]. This includes better self care, dressing, transfers, less dependence on caregivers and faster movement times. For patients who have spasms at night, being able to sleep uninterrupted was a major benefit. A few patients have been able to go from a wheelchair to independent walking, although most do not have the residual motor control needed to achieve major gains in ambulation. These conclusions have been confirmed by more recent quality of life survey of 49 patients treated [16]. In that study, 88% of the patients treated stated that their quality of life had improved, 8% were unsure and 4% saw no change, but no one was worse. The patients noted control of spasticity without sedation, ease of care for caregivers, easier positioning, less pain and improved transfers as reasons they felt improved. Although the implantation of the drug pump
Type of spasticity
Type
Spinal/MS Mixed
Double blind Prospective multicenter
Spinal/MS Mixed Mixed Cerebral (CP) Cerebral
Prospective multicenter Prospective Prospective multicenter Prospective multicenter Double blind
system and maintenance of therapy is expensive, cost effectiveness studies have suggested it is ‘‘cost effective’’ [17]. There are several reasons for these gains. > Figure 117-3 in chapter G-30 on spasticity demonstrates a patient’s attempt at dorsiflexion of the ankle before and after intrathecal baclofen. Co-contraction is reduced as well as the abnormal spread of activation to the distant muscle groups when baclofen is given. This effect underlies some of the improvement seen in some patients. Another factor is that reduced hypertonia allows for passive movement of legs for dressing and self catheterization even though voluntary movements may not be possible. Also, control of leg spasms means that the upper extremity movements are not disrupted. This is particularly important for work and school, in which stable sitting is critical. A number of investigators have studied changes in bladder function after intrathecal baclofen. Increased bladder capacity and reduced bladder spasms and dyssynergy have been seen in some patients [18,19]. The variable response to baclofen undoubtedly is due to variations in bladder pathology. A few patients had worse bladder control from intrathecal baclofen due to relaxation of the external sphincter producing incontinence. Bowel habits may change because of baclofen, but management has not been a problem and fecal incontinence has not been a complaint.
1975
1976
117
Intrathecal drugs for spasticity
. Figure 117-2 A typical prospective study (A and B) of intrathecal baclofen for spasticity of spinal origin. Note the marked decrease in Ashworth and spasm scores after a bolus trial and the with constant infusion
To achieve good clinical effects, the amount of baclofen infused usually has to be increased in the first 6–12 months. In most patients, it stabilizes by a year to 2 years and further increases in dosage are not necessary. The range of effective dosing is quite large. Some patients are well controlled with 25 mcg per day and other patients may need over 1000 mcg a day. Tolerance to intrathecal baclofen is rare. If it develops patients have to be taken off medication for
several weeks to reduce the tolerance and then it can be restarted again at lower levels. During the withdrawal period, morphine can be used as a substitute [20]. The use of intrathecal baclofen is not without problems. > Table 117-3 lists the complications found in a survey of 40 centers with almost 1,000 patients. The rate was low for both the initial hospitalization and afterwards with chronic infusion [21]. The most frequent
Intrathecal drugs for spasticity
117
. Figure 117-3 111 In flow study of drug pump and catheter going into the lumbar subarachnoid space. Left: 111In does not exit the pump; the pump was malfunctioning. Middle: The 111In goes to the top of the catheter and stops; the end of the catheter was closed by a fibrosis scar. Right: Normal flow pattern with distribution of 111In along the spinal canal
catheter problems are kinks and migrations. Pump failure and pump stalls are rare but some have occurred causing sudden withdrawal. Other centers have reported higher complication rates, especially in children [22]. Most of the drug related side effects due to too high a dose of baclofen have proved to be reversible by simply reducing the dose. The most frequent central nervous system symptoms are drowsiness, dizziness, blurred vision and slurred speech. Nausea and vomiting can sometimes occur. These problems resolve within 24 h after the dose is decreased or a bolus dose is dissipated. Hypotension is potentially a problem. If test boluses are used for a trial, blood pressure should be monitored. Large bolus doses, 500 mcg and up, can cause coma and respiratory depression. Bolus doses in the range of 100–400 mcg may cause somnolence. For such moderate overdoses, physostigmine given intravenously may reverse the drowsiness without affecting the reduction in spasticity [23]. Unfortunately, no specific antagonist is clinically available for large overdoses of baclofen. Physostigmine given in multiple doses can have severe side effects by raising blood pressure and causing cardiac symptoms. Several patients in an early German multicenter studies developed pulmonary emboli [24]. For that reason, patients should not be given so much intrathecal baclofen that they become hypotonic and have no spasms. Some muscle
activity in the legs is necessary to avoid venous stasis. Proper muscle tone can be achieved by means of careful dose adjustment. The Medronic Inc. programmable pump has been used by most physicians to delivery intrathecal baclofen. The advantages of the programmable unit are that the dose can be adjusted easily and great precision is possible and variations of dose during the day can be programmed. Frequently patients need higher doses at night to control spasms and less during the waking hours to allow increased tone in the legs for transfers. The disadvantages of this pump are that the battery life is limited to 4–5 years and it has a complex mechanical and electronic design which is more prone to failure. In practice, Medronic pumps have functioned well with a yearly failure rate of 1–2%. A different approach involves using a constant infusion pump, originally available from Infusaid Inc. This has been employed successfully at a number of centers [24]. Adjustments of the dose require removing and replacing all the drugs in the pump reservoir and then waiting for the new drug concentration to go through the coils and the pump and catheter, which can take 12–24 h. For patients who do not need frequent or precise adjustments, the simpler pump is adequate. The most frequent problem with the current drug delivery systems are with their catheters. Stronger catheter material and better placement
1977
1978
117
Intrathecal drugs for spasticity
. Table 117-3 Complications of intrathecal baclofen multicenter study Survey of complications At 115 centers, 936 patients [21] Initial hospitalization CSF collection Constipation Headache CSF leaks Infection Hematoma Hydrocephalus Worsened gait Urinary retention Flipped pump Catheter revision Post discharge Infection Seroma Hydrocephalus Seizures System complications Catheter Kinks/migration Infection Occlusion Arachnoiditis Cause not stated Total Pumps Infection Patient request Hypermobility w/effusion Pump failure CSF Leaks Dehiscence Cause not stated Total
3.3% 2.9% 2.4% 2.2% 0.9% 0.8% 1 case 1 case 1 case 1 case 2 revisions 1.7% 0.8% 0.2% 0.1% N
%
40 3 1 1 19 64
4 <1 <1 <1 2% 7%
16 11 8
17 1 4
4 2 1 1 43
4 4 4 1 4%
techniques have cut down on the number of disruptions in flow but the catheters are far from perfect. For proper operating techniques to place the catheter see [25] for adults and [26] for children. The key points are that the catheter should be inserted under fluoroscopy at oblique angle. This is to avoid the intraspinous ligaments and to make cephalad advancement easier.
The catheter has to be secured at the fascial level with a purse-string absorbable suture and close to that a silastic holder is used to reduce catheter movement. A two-piece catheter is preferable because of the heavier proximal part of the catheter is strong and does not kink. Until better catheters are available, neurosurgeons will have to contend with diagnosing and correcting drug delivery problems. When a previously well controlled patient has an abrupt onset of spasticity, the pump and catheter systems can be checked by a series of steps. First, the pump is checked electronically using a programmer to make sure that the system has the correct prescription and that the battery is functioning. Then the fluid in the reservoir is aspirated to see if the reservoir was filled correctly and if the amount corresponds to what was to have been delivered. Since the reservoir volume cannot be measured better than plus or minus 1 ml, this is only a gross test of infusion. The typical patient receives 0.33 ml a day, so it takes several days after the flow stops to note a discrepancy. X-rays of the catheter system may demonstrate dislodgments or disconnections and sometimes kinks are obvious. A bolus of 50–100 mcg of baclofen can be programmed to check the patient’s response. The total lack of response usually indicates a lack of drug delivery. A small but definite response may mean that tolerance has developed or that there is a small leak from the catheter. If the problem still cannot be found, then Indium-111 flow study through the pump can be done [27]. > Figure 117-3 shows a normal Indium study, one in which the isotope does not get out of the pump and one in which the flow is stopped because of fibrosis around the tip of the catheter within the lumbar subarachnoid space. The alternative to an Indium-111 study is the use of the side port on the drug pump to inject a radio-opaque dye under fluoroscopy; It must be remembered that the catheter may have a high concentration of baclofen (at 2,000 mcg/cc and 0.2 cc in the catheter and amount that would be
Intrathecal drugs for spasticity
injected is 400 mcg) which could potentially cause an overdose. Furthermore, catheter injections sometimes miss small leaks in the catheter system because they are hard to see. When all the tests fail to determine the cause of the problem, some clinicians simply change the catheter and for reasons that are unclear, that may resolve the issue. Sudden withdrawal from intrathecal baclofen can cause a rare, but potentially lethal side effect.
117
The syndrome consists of high fever, seizures, altered mental status and profound muscle spasm and/or rigidity that can lead to rhabdomyolysis. These signs are similar to autonomic dysreflexia and indeed many patients who have experienced severe withdrawal syndrome have mid thoracic or higher spinal cord injuries making then susceptible to autonomic instability. The baclofen withdrawal syndrome can be
. Table 117-4 Differential Diagnosis of ITB Withdrawal ITB Withdrawal
Autonomic Dysreftexia
MH
NMS
Ryanodine receptor mutation causing leakage of calcium from sarcoplasmic reticulum Acute onset during or after anesthesia, especially halothane; patient may have prior or family history; muscle biopsy establishes diagnosis Tachycardia
Acute loss of hypothalamic dopaminergic transmission
Tachycardia
Hypertension
Hypertension
Normal, but skin temperature may be elevated Decreased level of consciousness
Elevated, sometimes followed by hypothermia Decreased level of consciousness
Spasticity and rigidity below the level of the spinal cord lesion
Generalized, sustained, rigorous (tetanic) muscle contractions
Tremor, worsening to profound, generalized rigidity
Piloerection and skin pallor below level of injury; flushing, vasodilation, and profuse sweating above level of injury
CK elevation at onset; centrally acting muscle relaxants are ineffective
Labile or hyperactive autonomic function, leukocytosis at onset
Mechanism
Abrupt decrease in CNS GABAB transmission
Disconnection of major splanchnic sympathetic outflow from supraspinal control
Clinical setting and timing
One to 3 day evolution of the life-threatening syndrome after abrupt cessation of ITB
Heart rate
Tachycardia
Blood pressure
Hypotension or labile blood pressure Elevated
Ongoing nociceptive stimulus below the level of the spinal cord lesion (ie, urinary obstruction) in a T6 level or higher SCI patient Bradycardia, sometimes followed by tachycardia Hypertension
Body temperature Mental status
Muscle activity
Other clinical features
Dysphoria or malaise that evolves to decreased level of consciousness Rebound spasticity and rigidity greater than patient’s baseline Prodromal itching or paresthesias, priapism in men; seizures may occur during advanced syndrome
Elevated
Normal level of consciousness with dysphoria or apprehension
Hours after initiation of dopamine-blocking neuroleptic drugs, or abrupt cessation of dopamine agonist administration
1979
1980
117
Intrathecal drugs for spasticity
confused with malignant hyperthermia (MT) or neuroleptic/malignant syndrome (NMS). A full review of this problem is provided by Coffey [28]. The differentiating features are listed in a table from that paper (> Table 117-4). The only effective therapy is to rapidly restore intrathecal baclofen while providing supportive care in the ICU. This means giving intrathecal baclofen by lumbar puncture or preferentially by an external pump and intrathecal catheter if the pump system is not working. Since many withdrawal events are due to failure to refill patients’ pumps in a timely manner, refilling the pump and giving a large bolus followed by a high concentration infusion will be adequate if done soon enough. The key to recognizing baclofen withdrawal is early itching which almost always occurs [29]. For supportive care, benzodiazepines are often used. Cyproheptadine to reverse presumed serotonergic syndrome may also be helpful [30]. The recognition of this syndrome in emergency rooms is important so that it is correctly diagnosed and treated as early as possible. Patients and caregivers should also be warned of the rare, but serious threat that sudden baclofen withdrawal represents. The question of who can be treated with intrathecal baclofen is easy to answer. Any patient with spasticity that is not adequately controlled by oral medications and is causing significant pain, spasms or interfering with activities of daily living should be considered. If the spasticity is due to multiple sclerosis or spinal cord injury, a test bolus might not even be necessary. If there is the possibility that the patient will not respond, an intrathecal dose can be carried out. Bolus dosing will often have side effects or make the patient too weak. The patient must be warned that this could happen. The trial is only to see if intrathecal baclofen works for that particular patient’s type of spasticity. Continuous infusion with a pump is, in almost all cases, able to control drug side effects. Whether baclofen is better than alternative ablative
surgery or rhizotomies depends on many factors including the operative experience of the team offering surgery, the age of the patient and the preferences of the caregiver. Baclofen is an effective treatment, but has real complications and costs. It does not cure the cause of spasticity, so it commits physicians and patients to a lifetime of treatment. Until we have ways to repair CNS injuries directly, intrathecal baclofen will continue to have an important role in reducing the suffering of spastic patients.
References 1. Penn RD, Kroin JS. Intrathecal baclofen alleviates spinal cord spasticity. Lancet 1984;1(8385):1078. 2. Penn RD, Kroin JS. Continuous intrathecal baclofen for severe spasticity. Lancet 1985;2(8447):125-7. 3. Price G, Wilkin GP, Turnbull MJ, Bowery NG. Are baclofensensitive GABA receptors present on primary afferent terminals of the spinal core? Nature 1984;307:71-4. 4. Davidoff RA, Sears ES. The effects of Lioresal on synaptic activity in the isolated spinal cord. Neurology 1974;24(10):957-63. 5. Zieglglansberger W, Howe JR, Sutor B. The neuropharmacology of baclofen. In: Z. J. Muller H, Penn RD, editors. Local spinal therapy of spasticity. Berlin: Springer-Verlag; p. 37-49. 6. Li Y, Li X, et al. Effects of baclofen on spinal reflexes and persistent inward currents in motoneurons of chronic spinal rats with spasticity. J Neurophysiol 2004;92(5):2694-703. 7. Knutsson E, Lindblom U, et al. Plasma and cerebrospinal fluid levels of baclofen (Lioresal) at optimal therapeutic responses in spastic paresis. J Neurol Sci 1974;23(3):473-84. 8. Kroin JS, Penn R. Cerebrospinal fluid pharmacokinetics of lumbar intrathecal baclofen. In: Lakke DE, JPWF, Rutgers AWF, editors. Parenteral drug therapy in spasticity and Parkinson’s disease. Carnforth, UK: Parthenon; p. 67-77. 9. Park TS, A., L. Treatment of spasticity. Neurosurgical Focus 2006;21(2). 10. Penn RD, Gianino JM, et al. Intrathecal baclofen for motor disorders. Mov Disord 1995;10(5):675-7. 11. Penn RD, Mangieri EA. Stiff-man syndrome treated with intrathecal baclofen. Neurology 1993;43(11):2412. 12. Becker R, Benes L, et al. Intrathecal baclofen alleviates autonomic dysfunction in severe brain injury. J Clin Neurosci 2000;7(4):316-19.
Intrathecal drugs for spasticity
13. Coffey JR, Cahill D, et al. Intrathecal baclofen for intractable spasticity of spinal origin: results of a long-term multicenter study. J Neurosurg 1993;78(2):226-32. 14. Loubser PG, Narayan RK, et al. Continuous infusion of intrathecal baclofen: long-term effects on spasticity in spinal cord injury. Paraplegia 1991;29(1):48-64. 15. Parke B, Penn RD, et al. Functional outcome after delivery of intrathecal baclofen. Arch Phys Med Rehabil 1989;70(1):30-2. 16. Staal C, Arends A, et al. A self-report of quality of life of patients receiving intrathecal baclofen therapy. Rehabil Nurs 2003;28(5):159-63. 17. de Lissovoy G, Matza LS, et al. Cost-effectiveness of intrathecal baclofen therapy for the treatment of severe spasticity associated with cerebral palsy. J Child Neurol 2007;22(1):49-59. 18. Nanninga JB, Frost F, et al. Effect of intrathecal baclofen on bladder and sphincter function J Urol 1989;142(1):101-5. 19. Talalla A, Grundy D, et al. The effect of intrathecal baclofen on the lower urinary tract in paraplegia. Paraplegia 1990;28(7):420-7. 20. Penn RD, Kroin JS. Long-term intrathecal baclofen infusion for treatment of spasticity. J Neurosurg 1987;66(2):181-5. 21. Stempien L, Tsai T. Intrathecal baclofen pump use for spasticity: a clinical survey. Am J Phys Med Rehabil 2000;79(6):536-41. 22. Motta F, Buonaguro V, et al. The use of intrathecal baclofen pump implants in children and adolescents:
23.
24.
25.
26. 27.
28.
29.
30.
117
safety and complications in 200 consecutive cases. J Neurosurg 2007;107 Suppl 1:32-5. Muller-Schwefe G, Penn RD. Physostigmine in the treatment of intrathecal baclofen overdose. Report of three cases. J Neurosurg 1989;71(2):273-5. Muller H. Treatment of severe spasticity: results of a multicenter trial conducted in Germany involving the intrathecal infusion of baclofen by an implantable drug delivery system. Rev Eur Technom Biomed 1991;13:184-6. Penn R. Placement of morphine or baclofen pumps. In: Fessler R, Sekhar L, editors. Atlas of neurosurgical techniques spine and peripheral Nerves. New York: Thieme; 2006. Albright AL, Ferson SS. Intrathecal baclofen therapy in children. Neurosurg Focus 2006;21(2):e3. Rosenson AS, Ali A, Fordham EW, Penn RD. Indium-111 DTPA flow study to evaluate surgically implanted drug pump delivery system. Clin Nuclear Med 1990;15:154-6. Coffey RJ, Edgar TS, et al. Abrupt withdrawal from intrathecal baclofen: recognition and management of a potentially life-threatening syndrome. Arch Phys Med Rehabil 2002;83(6):735-41. Ben Smail D, Hugeron C, et al. Pruritus after intrathecal baclofen withdrawal: a retrospective study. Arch Phys Med Rehabil 2005;86(3):494-7. Meythaler JM, Roper JF, et al. Cyproheptadine for intrathecal baclofen withdrawal. Arch Phys Med Rehabil 2003;84(5):638-42.
1981
104 Management of Essential Tremor J. M. Nazzaro . K. E. Lyons . R. Pahwa
Essential tremor (ET) is one of the most common movement disorders with prevalence estimates ranging from 0.4 to 5% [1]. It is common to see several members of the same family affected and it is inherited in an autosomal dominant fashion. ET may present as early as childhood, but most commonly occurs in older adults and males and females are equally affected. Tremor frequency is generally 4–12 Hz with variable amplitude. The tremor is most commonly postural and/or kinetic and is generally not present in the rest position. It is not uncommon for the tremor to be increased with anxiety or stress [2]. It has been estimated that upper extremity tremor is present in 95% of ET patients, head tremor is present in 34%, lower extremity tremor in 20%, voice tremor in 12% and involvement of the tongue, face or trunk in 5% [3]. The main disability is related to activities requiring use of the dominant hand such as writing, eating, drinking, and other activities requiring fine movements.
Pharmacologic Treatment of Essential Tremor Pharmacologic treatment is generally initiated when the tremor causes functional disability and consequent impairment in activities of daily living. Initially, treatment may only be required on an as needed basis such as during stressful or anxiety provoking situations. In these cases, propranolol, a beta-adrenergic blocker, or a benzodiazepine such as clonazepam or alprazolam can be taken. In addition, the majority of ET patients have a reduction in tremor with alcohol; therefore, when used #
Springer-Verlag Berlin/Heidelberg 2009
judiciously, alcohol may be beneficial during social engagements or other stressful situations [3]. If persistent disability is present, a regular medication regimen should be initiated. The most effective and most commonly used medications in the treatment of ETare the beta-adrenergic blocker propranolol and the anticonvulsant primidone [4]. Propranolol is the only medication approved by the United States Food and Drug Administration (FDA) for ET. When given in doses of 60–320 mg/day, it has been shown to reduce tremor by approximately 50–60% [5–8]. The majority of studies have reported a reduction of upper extremity tremor with inconsistent benefit in head tremor [9,10]. A long-acting formulation of propranolol is available and has been found to reduce tremor comparably to the standard formulation [11,12]. The most common side effects with propranolol are nausea, vomiting, bradycardia, diarrhea, hypotension, drowsiness, fatigue, lightheadedness, weakness, and paresthesia. In patients that do not respond to or cannot tolerate propranolol, small studies have suggested that other beta-adrenergic blockers such as metoprolol, atenolol, sotalol, timolol, nadolol, and arotinolol may effectively reduce tremor in some patients [4]. Primidone, at doses of 50–1000 mg/day, has also been shown in multiple studies to reduce upper extremity tremor by approximately 50–70% with less consistent effects on head tremor [7,13–15]. Side effects generally resolve after initial use and most commonly include ataxia, vertigo, nausea, vomiting, fatigue, impotence and rash. Primidone and propranolol have been compared in several studies and have been
1744
104
Management of essential tremor
shown to have comparable control of tremor up to 1 year after initiation [16–18]. In patients that do not have satisfactory tremor control with either propranolol or primidone alone, a combination of the two medications can provide additional tremor control [3]. If satisfactory tremor control is not achieved with a beta-adrenergic blocker and/or primidone, there are several other treatment options including topiramate, gabapentin, zonisamide, pregabalin, benzodiazepines, and a host of other medications with undetermined efficacy that can be tried [4]. The anticonvulsant, topiramate, has been shown to provide significant tremor reduction in some ET patients. In two double-blind, placebocontrolled studies [19,20], topiramate, up to 400 mg/day, provided a significant reduction in tremor compared to placebo. However, in both studies, over 30% of the subjects withdrew prior to study completion, the majority due to adverse events. The most common adverse events were loss of appetite, taste perversion, weight loss, nausea, fatigue, somnolence, headache, paresthesia, dizziness, and disorientation. Gabapentin, another anticonvulsant, has been shown to have mixed results as a treatment for ET. In one study examining gabapentin, propranolol and placebo as monotherapy [21], tremor reduction with gabapentin (1,200 mg/day) was comparable to that of propranolol (120 mg/day) and both drugs were superior to placebo. Another study compared gabapentin (1,800 or 3,600 mg/day) and placebo as adjunct therapy [22]. Significant tremor reduction was observed with gabapentin compared to placebo on clinical ratings; however there were no differences between the two groups according to accelerometry. A third study found no significant differences between gabapentin (1,800 mg/day) and placebo [23]. The most common side effects were drowsiness, fatigue, dizziness, nervousness, shortness of breath, and nausea. Small double-blind, placebo-controlled studies of the anticonvulsants, zonisamide [24] and
pregabalin [25] have been reported. Zonisamide (up to 200 mg/day) demonstrated no tremor reduction compared to placebo as measured by clinical rating scales; however, 40% of the patients reported a slight improvement in tremor with zonisamide and it resulted in a significant reduction in tremor amplitude according to accelerometry when compared to placebo. Approximately 30% of the zonisamide subjects did not complete the study due to side effects which included diarrhea, fatigue, headache, nausea, and paresthesia. Pregabalin (up to 600 mg/day) resulted in a significant reduction in upper extremity tremor compared to placebo according to both clinical ratings of tremor and accelerometry. Although 67% of the pregabalin subjects reported some improvement in tremor, according to total tremor rating scale scores, activities of daily living, drawing and pouring there were no differences between pregabalin and placebo. Side effects related to pregabalin, including dizziness, flu symptoms and malaise, led to early study withdrawal in approximately 30% of subjects. Benzodiazepines such as alprazolam or clonazepam may be beneficial for ET patients, particularly those with associated anxiety. In a small double-blind study [26], alprazolam provided tremor relief comparable to primidone at a daily dose of 0.75mg. In a second study [27], alprazolam significantly reduced tremor compared to placebo. In both studies, the primary side effect was sedation. Conflicting results have been reported for clonazepam. In one study, clonazepam (up to 4 mg/day) was no different from placebo and a large proportion of the subjects withdrew from the study due to sedation. However, in another study [28], clonazepam (up to 6 mg/day) significantly reduced kinetic tremor in 100% of subjects. Although there are several potential treatment options for ET, the currently available pharmacologic treatments provide satisfactory tremor reduction in only about 50% of ET patients [4].
Management of essential tremor
When functional ability and quality of life are significantly affected by tremor refractory to medical measures, neurosurgical interventions should be considered. The surgeries currently available for ET are thalamotomy and deep brain stimulation (DBS) directed to the ventralis intermedius (Vim) nucleus of the thalamus [4]. Presently the neurosurgical procedure of choice for ET is Vim DBS which delivers high frequency electrical stimulation via implanted hardware.
Surgical Treatment of Essential Tremor History Prior to Vim DBS, lesioning of the thalamus, thalamotomy, directed to the Vim was the operation of choice for ETand other tremor disorders. There is an extensive neurosurgical history dating to the 1950s regarding stereotactic lesioning techniques [29,30]. Lesions within the thalamus for ET were increasingly performed in the setting of technologic advances together with a more precise understanding of the thalamic anatomy and its relationship to clinical outcome. Significant and lasting benefit was reported in the majority of patients with Vim thalamotomy. A primary limitation of ablative surgery however, is that the surgical lesion is irreversible and Vim thalamotomy came to be associated with permanent neurologic deficits in many patients [31]. During thalamotomy, it was noted that stimulation above 100 Hz within the Vim resulted in tremor relief [32,33]. These observations led investigators in the 1980s to implant stimulating electrodes directed to the Vim for tremor control [34,35]. This led to approval in Europe, Canada and other countries outside the United States in 1993 and approval by the FDA in 1997 for unilateral Vim DBS for upper extremity tremor control in patients with disabling ET and Parkinsonian tremor.
104
Deep Brain Stimulation In DBS, a stimulating lead (Medtronic, Minneapolis, MN) comprised of four in-line electrode platinum-iridium contacts each measuring 1.5 mm in length and 1.27 mm in diameter and separated from the next contact by 0.5 mm (model 3389) or 1.5 mm (model 3387), is implanted. The lead is connected to an implantable pulse generator via a tunneled extension wire. The pulse generator is most commonly implanted within a pocket created in the outer soft tissues of the upper anterior chest though in some cases it may be implanted in the anterior abdominal area with use of a longer DBS extension wire. Stimulation may be delivered via single or multiple electrode contacts of the DBS lead. Parameters including stimulation voltage, pulse width, and frequency may be adjusted with telemetry. Monopolar or bipolar stimulation may be delivered. In ET patients, typical parameters in a well positioned lead are stimulation at or below 3.0 Volts (V), with pulse widths between 60–90 microseconds (ms), and frequencies of 135–160 Hertz (Hz), though higher parameters may be required and depend upon the location of the lead contacts [36,37]. The pulse generator requires replacement approximately every 2–5 years due to voltage depletion secondary to normal usage [38]. Many ET patients turn the DBS system off at night to conserve battery life. The mechanism of action of DBS and the underlying physiologic substrate accounting for the beneficial effects of DBS in the ET patient remain unknown [39–43]. The ultimate effects of DBS are similar to that of lesioning. Investigators have postulated activation as well as inactivating and/or resetting of neural pathways and networks by the stimulation. Neuronal and axon characteristics in regards to morphology, firing rate, and distance from the stimulation may be contributing factors to the underlying mechanisms of the ultimate clinical expression. Several have postulated
1745
1746
104
Management of essential tremor
different effects of stimulation on cell bodies in comparison to fiber tracts. In regards to long term brain morphologic effects of DBS, reports have indicated no significant detrimental long term effects upon brain tissue study at time of autopsy with current DBS hardware used at recommended stimulation settings. The most common finding is mild gliosis associated with the DBS lead [44].
Thalamic Anatomy and Physiology The thalamus is involved in central processing of sensory, motor, cognitive, and emotional functions. The thalamus is oval shaped, located within the diencephalon. The mesencephalon of the brain stem is situated ventrally (below) and the telencephalon is located dorsally (above). Medially is the third ventricle while the internal capsule comprises the lateral thalamic border. The foramen of Monroe comprises the rostral (anterior) border. The thalamus is oriented approximately 20–30 in reference to the mid-sagitial plane, with the caudal (posterior) aspect of the thalamus projecting further lateral than the anterior thalamus. The human thalamus is comprised of approximately 120 nuclei. The specific functional significance of several thalamic areas remains to be defined [45,46]. In regards to stereotactic surgery for ET, the area of interest is the ventral lateral tier of thalamic nuclei. The most frequently referred to human stereotactic atlas is the Schaltenbrand and Wahren atlas which employs the terminology of Hassler [47]. The Vim nucleus measures approximately 2–4 mm anterior-posterior, 8–10 mm medial to lateral, and 8–10 mm in height. Anteriorly the Vim is bordered by the ventralis oralis posterior (Vop) nucleus and posteriorly by the ventralis caudalis (Vc) nucleus. Medially the Vim is bordered by the central thalamic nucleus and laterally lies the internal capsule. Dorsal to the Vim lies the zentrolateralis
nucleus and the dorsal group of thalamic nuclei, such as the dorsalis intermedius nucleus and ventral lies the radiations of the medial lemniscus and the zona incerta. The Vim is viewed according to the method of Hassler as receiving cerebellar afferents, the Vop as the pallidal receiving area and the Vc as receiving sensory afferents from the medial lemniscus [45]. Neurons within the Vim respond with increased activity to contralateral passive joint movement and these have been referred to as kinesthetic cells. Tremor cells are also localized within the Vim. Tremor cells characteristically fire rhythmically and in a burst-like fashion reflecting the postural and kinetic tremor in ET patients. Also, voluntary or combined cells that increase their activity just before or during active movements may be encountered in the Vim [48]. Importantly, there is a homunculus representation within the Vim with face medial, leg lateral, and arm between leg and face. Vop generally demonstrates a notably decreased background and the units encountered are of a general lower frequency and amplitude than that most commonly encountered in the Vim. The predominant cells encountered within the Vop are voluntary cells. Kinesthetic cells may also be encountered though much less frequent than that encountered in the Vim. Presence of kinesthetic cells increases near the Vim. Tremor cells may also be encountered in the Vop, primarily the posterior aspect, adjacent to the Vim [48]. Characteristically, cells within Vc demonstrate responsiveness to light touch most often from a well defined and limited body area. These units have been labeled tactile cells or cutaneous sensory cells. Cells responsive to deep pressure may be encountered, particularly in the ventralanterior aspect of the Vc. Tremor cells may be encountered in the anterior Vc, near the Vim border. Intraoperatively, background signal intensity may be very similar to or generally further increased in comparison to the Vim. Importantly, the homunculus representation in the Vc reflects
Management of essential tremor
that found in the Vim. It is important to note that the Vc extends further lateral than the Vim which in turn extends further lateral than the Vop. These differences in laterality as it pertains to the homunculus within the Vc and Vim need to be taken into account during surgery [48].
Surgical Technique Operative stages include stereotactic frame placement or placement of feducials for frameless methods, brain imaging, target planning, patient positioning during operation, method of exposing the cortex, guide tube insertion, electrophysiology, DBS lead insertion, test stimulation, and securing the lead to the skull. An unrecognized error, for example of 1 mm, during any one of these steps confounds an unrecognized error at another step, leading to not only potentially unsatisfactory surgery but also limitations in reproducing the surgery in a consistent manner and in the setting of individual variations in thalamic anatomy. There are several aspects of the surgery which are based largely on the surgeon’s preference and experience. Stereotactic implantation of a DBS lead within the thalamus may be performed with the use of a stereotactic frame or may be accomplished via frameless methods. The surgery may incorporate microelectrode recordings with or without microstimulation, recordings and/or stimulation with semi-microelectrodes, or no electrophysiologic recordings and only test stimulation with the DBS lead. Targeting may be based solely on MRI or may incorporate MRI-CT fusion and various computer packages have been developed which incorporate stereotactic algorithms for multiple brain imaging data sets. Ventriculography or surgery based solely on CT or X-ray guidance is now rarely performed. Targeting methods also vary but these are overall minor variations based on standard landmarks with
104
some adjustments made for individual variation. If microelectrodes are used, these may be passed one at a time or more than one at a time via various arrays and microlectrode recording criteria before implantation of the DBS lead varies. Other variables include patient positioning for the surgery (head up or supine), guide tube length, specific intraoperative test stimulation parameters, positioning of excess lead wire, and timing of pulse generator and extension wire implantation. There are little data addressing any of these variables in relation to clinical outcomes [49,50]. A more detailed description of our approach to DBS surgery in the thalamus for ET is outlined below. Initially, the stereotactic frame is placed such that the base of the frame is roughly parallel to the orbital-meatal line which approximates the intercommissural plane. It is held in place via four skull fixation pins which are placed after local anesthesia. A high definition MRI for stereotactic localization is obtained. The MRI is obtained within the stereotactic frame in order to reduce patient movement. Some prefer to obtain the brain MRI before placing the frame. Another method would be to incorporate CTMRI fusion. The imaging data are then transferred to a computer system for stereotactic localization and target planning. With the use of the stereotactic computer system, the anterior commissure (AC) and the posterior commissure (PC) are localized and the AC/PC line is determined. Stereotactic coordinates for the Vim target localization are initially placed 11.5 mm lateral to the lateral wall of the third ventricle and are anteriorly/posteriorly located at the midpoint between the mid commissural point and the PC at the level of the AC/PC line. The target point is further adjusted according to the location of the internal capsule. In regards to laterality, the DBS lead should be situated at least 3 mm from the thalamic-internal capsule border. Unfortunately, this border may not always be well visualized by MRI. The trajectory
1747
1748
104
Management of essential tremor
of approach is selected to avoid cortical vessels, sulci, the ventricular system, and paraventricular vessels. We have evolved to prefer a more posterior frontal approach in our Vim DBS surgeries such that the intent is to come along the long axis of the Vim, and ideally situate the DBS lead within Vim, approximately 2–3 mm anterior to the Vc, and thus very near the Vim/ Vop border. The patient is positioned supine with the head not elevated to maintain the same position used during imaging and therefore reducing error associated with brain shifts. This also reduces the chance of air entry into the subdural space. A linear incision is made in the frontal area following local anesthesia. The precise location of the incision and entry point is determined by the trajectory of the planned stereotactic approach. A small and deliberate dural opening will help avoid loss of Cerebrospinal fluid. Upon entry of the guide tube within brain, gelfoam is placed extradural around the guide cannula. It is then slowly advanced within the brain to 20 mm above the planned target. The stylet within the guide cannula is then removed following which the microelectrode system is placed. At the beginning of the microelectrode recording, the microelectrode is slowly advanced such that it is situated 15 mm above the planned target. Microelectrode impedance is again checked to confirm preoperative values. In ET patients, microelectrodes are used for electrophysiologic localization. We do not routinely stimulate with the microelectrode for further localization although some surgeons may choose to do this. There are no clear data addressing patient outcome with recording versus recording and stimulation with microelectrodes. Further, there are little data available comparing thalamic surgeries in ET patients performed with or without microelectrode guidance. However, anatomic variability between individuals is well documented. To assume that the DBS lead will be optimally placed solely via atlas coordinates
based primarily on averaged distances from the AC and PC landmarks and in relation to the internal capsule may be suboptimal. Following satisfactory target localization, the microelectrode is withdrawn and the DBS lead is placed with a microdrive. Test stimulation is conducted to document both efficacy and unwanted side effects. If side effects are obtained, contact combinations are changed to help define the area producing the side effect. Unwanted motor responses such as twitching or tonic movements, generally indicates that the DBS lead is situated too far lateral. Paresthesias indicate Vc involvement and that the DBS lead is situated too posterior or that contact 0 is situated within the anterior caudal Vc and a higher contact combination should be tested. Sensory involvement of the leg suggests lateral Vc involvement whereas facial sensations suggest medial Vc involvement. Dizziness and related constitutional symptoms may occur if the lead is too deep. Sensory phenomena may also be elicited upon stimulation of the underlying (deep) lemniscal pathways. If test stimulation is clearly unsatisfactory, a decision must then be made whether to conduct additional microlectrode runs, reinsert the DBS lead at new coordinates, or stop the procedure. Upon satisfactory test stimulation, the external generator is disconnected from the brain electrode and the DBS lead is secured. A postoperative CT is obtained to ensure no complications such as intracranial hemorrhage and is fused with the preoperative MRI to ensure that the lead is in the intended location. If there are no complications, patients are discharged the morning following surgery. We generally implant the DBS pulse generator and extension wire 1–2 weeks after brain lead implantation. This is performed as a day surgery procedure, under general anesthesia. Some may perform implantation of all of the DBS hardware in one sitting. The DBS system is turned on approximately four weeks following surgery, to allow for healing and resolution of any changes (i.e., edema) at the stimulation site.
Management of essential tremor
Thalamotomy The surgery for lesioning the Vim is similar to that followed for Vim DBS. In thalamotomy for ET, the Vim is localized using electrophysiological recording and/or stimulation methods. Given the complexity of the anatomy involved, together with the irreversible nature of ablative surgery, accurate localization is paramount. The area to be ablated corresponds to the area in which the DBS lead would be implanted [51–53] Thermal techniques using radiofrequency evolved to be the preferred lesioning method. The area to be lesioned is within the Vim, at least 2–3 mm anterior to Vc, 2–3 mm dorsal to the ventral Vim border, and at least 3–4 mm medial to the thalamus-internal capsule border. Lesions may be made with an insulated electrode with an exposed tip. Other methods utilize two probes, between which current is passed for lesioning purposes [54]. Following test stimulation to ensure no unwanted side effects, a graduated lesion or lesions are made in a progressive fashion. Methods vary in regards to temperature and duration of heating [51–53]. In one method, a test lesion is made at 42 C for 60 s using a lesioning electrode, neurological symptoms are monitored following which the temperature is increased to 75 C for 60 s. If the tremor is not completely ablated following observation for approximately 15 min, the lesion may be enlarged by heating to 80 C for an additional 60 s or the probe may be withdrawn 3–5 mm and an additional lesion made at 80 C for 30 s. A lesion measuring 30–60 mm3 has been estimated to be adequate for ET [51].
Clinical Outcomes The short and long term benefits of unilateral Vim DBS on upper extremity tremor have been repeatedly demonstrated [35–37,55–58]. Significant resolution of upper extremity tremor has been
104
reported in up to 90% of patients. Details of the larger, long-term studies are discussed below. Pahwa et al. [36] reported outcomes of thalamic DBS after five years for 22 ET patients (15 unilateral; 7 bilateral). Targeted hand tremor was improved by 75% with unilateral implants and up to 86% with bilateral implants. Total body tremor was improved by 46% in those with unilateral implants and 78% in those with bilateral implants. After both unilateral and bilateral procedures, functional ability such as writing, drawing and pouring, was improved by up to 57%. There were no significant changes in stimulation parameter settings at one versus five years post-surgery. At 5 years, the average amplitude ranged from 3.2 to 3.6 V, average frequency ranged from 153 to 158 Hz and average pulse width ranged from 111 to 129 ms. Surgical and device related adverse events were calculated for 45 patients which included those with both ET and parkinsonian tremor. In this study, hardware complications, excluding pulse generator replacements occurred in approximately 25% of patients and over 10% reported stimulation related adverse events. The most common adverse events were paresthesia, pain, asthenia, dysarthria, hypophonia, incoordination, gait difficulties, abnormal thinking, headache, depression, dysphagia, and hallucinations. Several adverse events were more common with bilateral procedures, particularly, speech disorders, abnormal gait and incoordination. Sydow et al. [37] reported outcomes after an average of 6.5 years after Vim DBS for 19 ET patients (12 unilateral, 7 bilateral). There was an overall improvement in tremor of 41% at the 6-year follow-up which was reduced from a 53% improvement reported after 1 year. Similarly, activities of daily living were improved by 39% after 6 years which was reduced from 82% after one year. It should be noted that activities of daily living scores without stimulation at 6 years were significantly worse than baseline scores, suggesting that disease progression may be responsible for the decline in benefit seen
1749
1750
104
Management of essential tremor
between years one and six. Stimulation parameter settings were changed slightly between initial programming and the 6-year follow-up. Average amplitude was initially 2.0 V compared to 2.6 V at 6 years, average frequency was increased from 156 to 173 Hz and average pulse width was decreased from 103 to 89 ms. Surgical adverse events included infection, erosion and skin irritation each in two patients and subcutaneous hematoma, paresis and lead repositioning due to poor placement each in one patient. Hardware complications included loss of effect in two patients and intermittent stimulation in one. The most common stimulation related adverse events were paresthesia, gait disorders, dystonia and dysarthria which was more common in those with bilateral implants. Koller et al. [59] reported long-term outcomes up to 69 months (average 40 months) in 25 ET patients after unilateral Vim DBS. At the longest follow-up, total body tremor was improved by 50% and targeted hand tremor was improved by 78%. A decline in benefit with time was not reported. The average stimulation parameter settings at the longest follow-up were an amplitude of 3.6 V, frequency of 161 Hz and pulse width of 100 ms. The only surgical adverse event reported was seizure in one patient. Hardware complications related to the lead included replacement in six patients due to loss of effect, migration in one patient and fracture in one patient. Other hardware complications included erosion from the extension wire in three patients, pulse generator malfunction in one patient and complete removal of the system in one patient due to loss of effect. The most common stimulation related adverse events included paresthesia, headache, paresis, dysarthria, and disequilibrium which were mild and resolved with parameter adjustments. Putzke et al. [58] reported outcomes 3–36 months after Vim DBS in 52 ET patients (29 unilateral, 23 bilateral). There was a minimum improvement of 83% in upper extremity tremor and 63% in activities of daily living at all follow-up
visits and there was no reduction in benefit over time. Stimulation parameters were relatively constant over time with average amplitude at 1 month being 2.9 V compared to 3.0 V at 3 years, an increase in average frequency from 154 to 171 Hz and a slight increase from 86 to 88 ms in pulse width. Loss of effect requiring lead repositioning occurred in eight patients, infection occurred in one patient and lead breakage occurred in two patients. The most common stimulation related adverse events included dysarthria, disequilibrium, paresthesia, and motor disturbances. In those with bilateral implants, staged procedures were performed and adverse events were examined after both the first and second implants. There was a significant increase in dysarthria after the second procedure and disequilibrium was also more common after the second implant. Benefit may also be gained in head and voice tremor after Vim DBS. There is only one case report of two ET patients who received bilateral Vim DBS specifically for head tremor. Total resolution of head tremor was reported 9 months after implantation [60]. Effects on head and voice tremor have been examined in several cohorts for which Vim DBS was performed primarily for upper extremity tremor [37,61–63]. In these studies, head and voice tremor were improved from 15 to 51% with unilateral procedures and from 39 to 100% with bilateral procedures. Several studies have compared the clinical outcomes and complications related to Vim DBS and Vim thalamotomy [64–67]. The efficacy of Vim DBS and Vim thalamotomy in ET are comparable. Approximately 70–80% of patients have marked reduction in contralateral upper extremity tremor with both procedures. Both methods produce complete or nearly complete resolution of contralateral upper extremity tremor in approximately 60–90% of patients. Long term benefits are similar between both methods, with significant lasting improvement seen in the majority of patients although with both therapies, over time there may be a reduction in
Management of essential tremor
benefit in some patients. The primary difference between the two therapies is an increase in serious neurological complications associated with thalamotomy compared to DBS. An advantage of Vim DBS is the reversibility and adjustability of the system to improve efficacy and reduce adverse events. The primary drawbacks of DBS are the costs associated with the hardware, system adjustments and maintenance and complications related to the hardware (infection, breakage, migration, and wound breakdown). Presently, Vim thalamotomy is only considered in select patients with medication resistant disabling tremor who would not be good candidates for Vim DBS. Considerations include medical issues precluding tolerance of the DBS hardware, known need for repeated MRI scanning, inability to comply with follow up appointments for DBS programming and cost. In rare cases, an ET patient may not be able to tolerate DBS or thalamotomy due to increased surgical risk secondary to other medical conditions. In these cases, radiosurgical thalamotomy may be a possible treatment option [4].
Radiosurgical Thalamotomy Stereotactic radiosurgical thalamotomy has been used as a treatment for ET, though the experience is limited and only the gamma knife method has been reported. Targeting is based primarily on atlas coordinates, with adjustments made for associated anatomy that may be visualized on planning brain MRI such as the location of the thalamic-internal capsule border or proportional in relation to total thalamus length. Niranjan et al [68] reported eight ET patients treated with gamma knife thalamotomy. Radiosurgery dosages range from 130 to 150 Gy and were delivered using a 4 mm collimator. Patients were followed a median of 6 months. All patients had improvement on objective measures and 75% reported complete tremor resolution while 25%
104
reported greater than 50% improvement. Tremor improvement began an average of 6 weeks following radiosurgery and continued over the next 6 months. One patient developed hemiparesis and dysarthria 8 months following radiosurgery. Kondzioka et al [69] reported 26 ET patients 4–96 months after gamma knife thalamotomy, eight of which were previously reported [68]. Radiosurgery was delivered to a single isocenter using a 4 mm collimator at 130–140 Gy. Patients returned for clinical and imaging evaluations 4 months after radiosurgery, and yearly thereafter. Improvement in both action tremor and handwriting was present in 69% of patients, 23% showed improvement only in action tremor, and 12% had no improvement in tremor or handwriting. Complications included transient mild right hemiparesis in one patient and mild right hemiparesis and speech impairment in another. Brain MRI 6 months following radiosurgery in one of these patients demonstrated a lesion much larger than anticipated; however, 4 years later there was no evidence of a thalamotomy lesion. Young et al [70] reported 51 ET patients treated with gamma knife thalamotomy. Radiation was delivered via a 4 mm collimator to a dose of 120–160 Gy; 92% of patients had near-complete or complete tremor relief on formal blinded assessments 6 months following treatment. There were four (8%) treatment failures. Two ET patients lost benefit by 24 months. Seventeen ET patients were followed for 48 months; 71% were tremor free, 18% were near tremor free, 6% had no benefit and the long term outcome of the remaining 6% was not clearly specified. Complications specific to the ET group were not reported though of the total study group of 158 gamma knife thalamotomy patients (102 PD, 52 ET, 4 other), adverse events secondary to gamma knife thalamotomy were rare. There were two patients who experienced minor neurologic adverse events; one patient had mild persistent hemiparesis and speech difficulty while another had mild paresthesia involving the face and upper extremity. Six months following
1751
1752
104
Management of essential tremor
treatment, lesion sizes for the entire group varied from 60 to 523 mm3 while at 36 months the range of lesion sizes were 10 to 2030 mm3. Ohye et al. [71] reported a gamma knife thalamotomy series that included 11 ET patients. Outcomes specific to the ET group are difficult to discern from the data provided. For the larger group that included PD, ET, and tremor secondary to other etiologies, benefit was not seen until one year following radiation, though follow-up was overall random and several patients did not return. Two of the eleven ET patients had no improvement and were again treated with gamma knife 2 and 4 years later with no benefit. The authors also reported marked variability in lesion sizes and areas of involvement on follow up MRIs despite standardization of collimator size (4 mm) and dose (130 Gy). Siderowf et al [72] reported serious neurologic sequelae in an ET patient treated with gamma knife radiation, though details regarding the radiation delivered were not provided. Factors that have tempered enthusiasm for gamma knife thalamotomy include lesioning based on indirect targeting without physiologic guidance, delay of benefit for several weeks or months, documented variability in lesion size and area of brain involvement, and delayed onset of complications. At this time, there are insufficient data regarding gamma knife thalamotomy in ET patients [7]. Future advances in brain imaging and radiosurgery equipment and techniques may extend the application of radiosurgical thalamotomy for ET patients; however this awaits further examination in carefully controlled long-term studies.
Summary ET is a common movement disorder that can cause significant disability. If functional disability is present, the most commonly used medications are propranolol and primidone either alone or
in combination. Other pharmacologic therapies commonly prescribed include other beta-adrenergic blockers such as atenolol or metoprolol, other anticonvulsants such as topiramate and gabapentin and benzodiazepines such as clonazepam and alprazolam. Although there are several treatment options available, they provide satisfactory tremor control in only about 50% of patients. For patients with medication resistant ET, Vim DBS is the preferred surgical therapy and provides satisfactory tremor control in up to 90% of patients. Vim DBS and thalamotomy have been shown to provide comparable efficacy but increased complication rates have been reported with thalamotomy. For patients who are not candidates for DBS due to other medical conditions, cost of the procedure, or an inability to return to the surgical center for follow-up adjustments and system maintenance, Vim thalamotomy may be an option. In the rare case that medical conditions prohibit both DBS and thalamotomy, gamma knife thalamotomy may be considered.
References 1. Louis ED, Ottman R, Hauser WA. How common is the most common adult movement disorder? Estimates of the prevalence of essential tremor throughout the world. Mov Disord 1998;13(1):5-10. 2. Lou JS, Jankovic J. Essential tremor: clinical correlates in 350 patients. Neurology 1991;41(2 Pt 1):234-8. 3. Pahwa R, Lyons KE. Essential tremor: differential diagnosis and current therapy. Am J Med 2003;115 (2):134-42. 4. Lyons KE, Pahwa R, Comella CL, Eisa MS, Elble RJ, Fahn S, et al. Benefits and risks of pharmacological treatments for essential tremor. Drug Saf 2003;26 (7):461-81. 5. Tolosa ES, Loewenson RB. Essential tremor: treatment with propranolol. Neurology 1975;25(11):1041-4. 6. Calzetti S, Findley LJ, Gresty MA, Perucca E, Richens A. Effect of a single oral dose of propranolol on essential tremor: a double-blind controlled study. Ann Neurol 1983;13(2):165-71. 7. Zesiewicz TA, Elble R, Louis ED, Hauser RA, Sullivan KL, Dewey RB, Jr., et al. Practice parameter: therapies for essential tremor: report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 2005;64(12):2008-20.
Management of essential tremor
8. Calzetti S, Sasso E, Baratti M, Fava R. Clinical and computer-based assessment of long-term therapeutic efficacy of propranolol in essential tremor. Acta Neurol Scand 1990;81(5):392-6. 9. Calzetti S, Sasso E, Negrotti A, Baratti M, Fava R. Effect of propranolol in head tremor: quantitative study following single-dose and sustained drug administration. Clin Neuropharmacol 1992;15(6):470-6. 10. Koller WC. Propranolol therapy for essential tremor of the head. Neurology 1984;34(8):1077-9. 11. Cleeves L, Findley LJ. Propranolol and propranolol-LA in essential tremor: a double blind comparative study. J Neurol Neurosurg Psychiatry 1988;51(3):379-84. 12. Koller WC. Long-acting propranolol in essential tremor. Neurology 1985;35(1):108-10. 13. Findley LJ, Cleeves L, Calzetti S. Primidone in essential tremor of the hands and head: a double blind controlled clinical study. J Neurol Neurosurg Psychiatry 1985;48 (9):911-5. 14. Koller WC, Royse VL. Efficacy of primidone in essential tremor. Neurology 1986;36(1):121-4. 15. Sasso E, Perucca E, Fava R, Calzetti S. Primidone in the long-term treatment of essential tremor: a prospective study with computerized quantitative analysis. Clin Neuropharmacol 1990;13(1):67-76. 16. Gorman WP, Cooper R, Pocock P, Campbell MJ. A comparison of primidone, propranolol, and placebo in essential tremor, using quantitative analysis. J Neurol Neurosurg Psychiatry 1986;49(1):64-8. 17. Dietrichson P, Espen E. Primidone and propranolol in essential tremor: a study based on quantitative tremor recording and plasma anticonvulsant levels. Acta Neurol Scand 1987;75(5):332-40. 18. Koller WC, Vetere-Overfield B. Acute and chronic effects of propranolol and primidone in essential tremor. Neurology 1989;39(12):1587-8. 19. Connor GS. A double-blind placebo-controlled trial of topiramate treatment for essential tremor. Neurology 2002;59(1):132-4. 20. Ondo WG, Jankovic J, Connor GS, Pahwa R, Elble R, Stacy MA, et al. Topiramate in essential tremor: a doubleblind, placebo-controlled trial. Neurology 2006;66 (5):672-7. 21. Gironell A, Kulisevsky J, Barbanoj M, Lopez-Villegas D, Hernandez G, Pascual-Sedano B. A randomized placebocontrolled comparative trial of gabapentin and propranolol in essential tremor. Arch Neurol 1999;56(4):475-80. 22. Ondo W, Hunter C, Vuong KD, Schwartz K, Jankovic J. Gabapentin for essential tremor: a multiple-dose, doubleblind, placebo-controlled trial. Mov Disord 2000;15 (4):678-82. 23. Pahwa R, Lyons K, Hubble JP, Busenbark K, Rienerth JD, Pahwa A,et al. Double-blind controlled trial of gabapentin in essential tremor. Mov Disord 1998;13(3):465-7. 24. Zesiewicz TA, Ward CL, Hauser RA, Sanchez-Ramos J, Staffetti JF, Sullivan KL. A double-blind placebo-controlled
25.
26.
27. 28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39. 40.
104
trial of zonisamide (zonegran) in the treatment of essential tremor. Mov Disord 2007;22(2):279-82. Zesiewicz TA, Ward CL, Hauser RA, Salemi JL, Siraj S, Wilson MC, et al. A pilot, double-blind, placebocontrolled trial of pregabalin (Lyrica) in the treatment of essential tremor. Mov Disord 2007;22(11):1660-3. Gunal DI, Afsar N, Bekiroglu N, Aktan S. New alternative agents in essential tremor therapy: double-blind placebo-controlled study of alprazolam and acetazolamide. Neurol Sci 2000;21(5):315-7. Huber SJ, Paulson GW. Efficacy of alprazolam for essential tremor. Neurology 1988;38(2):241-3. Biary N, Koller W. Kinetic predominant essential tremor: successful treatment with clonazepam. Neurology 1987;37(3):471-4. Speelman JD, Bosch DA. Resurgence of functional neurosurgery for Parkinson’s disease: a historical perspective. Mov Disord 1998;13(3):582-8. Ohye C, Maeda T, Narabayashi H. Physiologically defined VIM nucleus: its special reference to control of tremor. Appl Neurophysiol 1976;39(3–4):285-95. Jankovic J, Cardoso F, Grossman RG, Hamilton WJ. Outcome after stereotactic thalamotomy for parkinsonian, essential, and other types of tremor. Neurosurgery 1995;37(4):680-6; discussion 686–7. Hassler R, Riechert T, Mundinger F, Umbach W, Ganglberger JA. Physiological observations in stereotaxic operations in extrapyramidal motor disturbances. Brain 1960;83:337-50. Ohye C, Kubota K, Hongo T, Nagao T, Narabayashi H. Ventrolateral and subventrolateral thalamic stimulation: motor effects. Arch Neurol 1964;11:427-34. Benabid AL, Pollak P, Seigneuret E, Hoffmann D, Gay E, Perret J. Chronic VIM thalamic stimulation in Parkinson’s disease, essential tremor and extra-pyramidal dyskinesias. Acta Neurochir Suppl 1993;58:39-44. Benabid AL, Pollak P, Gao D, Hoffmann D, Limousin P, Gay E, et al. Chronic electrical stimulation of the ventralis intermedius nucleus of the thalamus as a treatment of movement disorders. J Neurosurg 1996;84(2):203-14. Pahwa R, Lyons KE, Wilkinson SB, Simpson RK, Jr., Ondo WG, Tarsy D, et al. Long-term evaluation of deep brain stimulation of the thalamus. J Neurosurg 2006;104 (4):506-12. Sydow O, Thobois S, Alesch F, Speelman JD. Multicentre European study of thalamic stimulation in essential tremor: a six year follow up. J Neurol Neurosurg Psychiatry 2003;74(10):1387-91. Ondo WG, Meilak C, Vuong KD. Predictors of battery life for the Activa Soletra 7426 Neurostimulator. Parkinsonism Relat Disord 2007;13(4):240-2. Dostrovsky JO, Lozano AM. Mechanisms of deep brain stimulation. Mov Disord 2002;17 Suppl 3:S63-8. Vitek JL. Mechanisms of deep brain stimulation: excitation or inhibition. Mov Disord 2002;17 Suppl 3: S69-72.
1753
1754
104
Management of essential tremor
41. Anderson T, Hu B, Pittman Q, Kiss ZH. Mechanisms of deep brain stimulation: an intracellular study in rat thalamus. J Physiol 2004;559(Pt 1):301-13. 42. Anderson TR, Hu B, Iremonger K, Kiss ZH. Selective attenuation of afferent synaptic transmission as a mechanism of thalamic deep brain stimulation-induced tremor arrest. J Neurosci 2006;26(3):841-50. 43. McIntyre CC, Savasta M, Walter BL, Vitek JL. How does deep brain stimulation work? Present understanding and future questions. J Clin Neurophysiol 2004;21(1):40-50. 44. Haberler C, Alesch F, Mazal PR, Pilz P, Jellinger K, Pinter MM, et al. No tissue damage by chronic deep brain stimulation in Parkinson’s disease. Ann Neurol 2000;48(3):372-6. 45. Hassler R. Architectonic organization of the thalamic nuclei. In: Schaltenbrand G, Walker A, editors. Stereotaxy of human brain. Anatomical, Thieme; physiological and clinical applications. New York: Thieme; 1982. p.140-80. 46. Percheron G. Thalamus. In: Paxinos G, Mai JK, editors. The human nervous system. New York: Elsevier; 2004. p. 592-675. 47. Schaltenbrand G, Wahren W. Atlas for stereotaxy of the human brain. New York: Thieme; 1977. 48. Hamani C, Dostrovsky JO, Lozano AM. The motor thalamus in neurosurgery. Neurosurgery 2006;58(1):146-58; discussion 146–58. 49. Rezai AR, Kopell BH, Gross RE, Vitek JL, Sharan AD, Limousin P, et al. Deep brain stimulation for Parkinson’s disease: surgical issues. Mov Disord 2006;21 Suppl 14: S197-218. 50. Machado A, Rezai AR, Kopell BH, Gross RE, Sharan AD, Benabid AL. Deep brain stimulation for Parkinson’s disease: surgical technique and perioperative management. Mov Disord 2006;21 Suppl 14:S247-58. 51. Bakay R, Vitek J, Delong M. Thalamotomy for tremor. In: Rengachary S, Wilkins R, editors. Neurosurgical operative atlas. New York: Thieme; 1992. p. 299-312. 52. Kelly P. Contemporary stereotactic ventralis lateral thalamotomy in the treatment of parkinsonian tremor and other movement disorders. In: Heilbrun M, editor. Concepts in neurosurgery. Stereotactic neurosurgery. Baltimore: Williams and Wilkins; 1988. p. 133-48. 53. Slavin K, Burchiel K. Thalamotomy without microelectrode recording. In: Lozano A, editor. Movement disorder surgery. Basel: Karger; 2000. p. 172-80. 54. Ohye C. Thalamotomy for Parkinson’s disease. Part 1. Historical background and technique. In: Gildenberg P, Tasker R, editors. Textbook of stereotactic and functional neurosurgery. New York: McGraw-Hill; 1998. p. 1167-78. 55. Koller W, Pahwa R, Busenbark K, Hubble J, Wilkinson S, Lang A, et al. High-frequency unilateral thalamic stimulation in the treatment of essential and parkinsonian tremor. Ann Neurol 1997;42(3):292-9.
56. Limousin P, Speelman JD, Gielen F, Janssens M. Multicentre European study of thalamic stimulation in parkinsonian and essential tremor. J Neurol Neurosurg Psychiatry 1999;66(3):289-96. 57. Rehncrona S, Johnels B, Widner H, Tornqvist AL, Hariz M, Sydow O. Long-term efficacy of thalamic deep brain stimulation for tremor: double-blind assessments. Mov Disord 2003;18(2):163-70. 58. Putzke JD, Wharen RE, Jr., Obwegeser AA, Wszolek ZK, Lucas JA, Turk MF, et al. Thalamic deep brain stimulation for essential tremor: recommendations for long-term outcome analysis. Can J Neurol Sci 2004;31(3):333-42. 59. Koller WC, Lyons KE, Wilkinson SB, Troster AI, Pahwa R. Long-term safety and efficacy of unilateral deep brain stimulation of the thalamus in essential tremor. Mov Disord 2001;16(3):464-8. 60. Berk C, Honey CR. Bilateral thalamic deep brain stimulation for the treatment of head tremor: report of two cases. J Neurosurg 2002;96(3):615-8. 61. Koller WC, Lyons KE, Wilkinson SB, Pahwa R. Efficacy of unilateral deep brain stimulation of the VIM nucleus of the thalamus for essential head tremor. Mov Disord 1999;14(5):847-50. 62. Ondo W, Almaguer M, Jankovic J, Simpson RK. Thalamic deep brain stimulation: comparison between unilateral and bilateral placement. Arch Neurol 2001;58 (2):218-22. 63. Putzke JD, Uitti RJ, Obwegeser AA, Wszolek ZK, Wharen RE. Bilateral thalamic deep brain stimulation: midline tremor control. J Neurol Neurosurg Psychiatry 2005;76(5):684-90. 64. Pahwa R, Lyons KE, Wilkinson SB, Troster AI, Overman J, Kieltyka J, et al. Comparison of thalamotomy to deep brain stimulation of the thalamus in essential tremor. Mov Disord 2001;16(1):140-3. 65. Schuurman PR, Bosch DA, Bossuyt PM, Bonsel GJ, van Someren EJ, de Bie RM, et al. A comparison of continuous thalamic stimulation and thalamotomy for suppression of severe tremor. N Engl J Med 2000;342 (7):461-8. 66. Tasker RR, Munz M, Junn FS, Kiss ZH, Davis K, Dostrovsky JO, et al. Deep brain stimulation and thalamotomy for tremor compared. Acta Neurochir Suppl 1997;68:49-53. 67. Tasker RR. Deep brain stimulation is preferable to thalamotomy for tremor suppression. Surg Neurol 1998;49(2):145-53; discussion 153-4. 68. Niranjan A, Kondziolka D, Baser S, Heyman R, Lunsford LD. Functional outcomes after gamma knife thalamotomy for essential tremor and MS-related tremor. Neurology 2000;55(3):443-6. 69. Kondziolka D, Ong JG, Lee JY, Moore RY, Flickinger JC, Lunsford LD. Gamma Knife thalamotomy for essential tremor. J Neurosurg 2008;108(1):111-7.
Management of essential tremor
70. Young RF, Jacques S, Mark R, Kopyov O, Copcutt B, Posewitz A, et al. Gamma knife thalamotomy for treatment of tremor: long-term results. J Neurosurg 2000;93 Suppl 3:128-35. 71. Ohye C, Shibazaki T, Zhang J, Andou Y. Thalamic lesions produced by gamma thalamotomy for movement disorders. J Neurosurg 2002;97(5 Suppl):600-6.
104
72. Siderowf A, Gollump SM, Stern MB, Baltuch GH, Riina HA. Emergence of complex, involuntary movements after gamma knife radiosurgery for essential tremor. Mov Disord 2001;16(5):965-7.
1755
105 Management of Tremors other than Essential Tremor and Parkinson’s Disease J. P. Nguyen . S. Raoul . C. Deligny . V. Roualdes . Y. Keravel
Resting tremor is mainly observed in Parkinson’s disease, and postural tremor mainly corresponds to essential tremor. These two types of tremor are well known and generally respond very well to stereotactic techniques targeted on the ventral intermediate nucleus (Vim) of the thalamus: thalamotomy or Vim stimulation. As indicated in other chapters, the currently preferred stimulation techniques are associated with a much lower risk of neurological sequelae. The third main group of tremor consists of so-called action tremor. This type of tremor is more complex, comprising a postural tremor component and an intention tremor component. Action tremor generally has a larger amplitude than the other two forms and predominantly involves proximal muscles. Action tremor very rapidly becomes disabling and surgery may be considered soon after its onset. There are many causes of action tremor, but the leading causes are multiple sclerosis and head injury. Stereotactic techniques are the same as for Parkinsonian and essential tremor, but the results are generally less favorable, as the functional result largely depends on the neurological signs associated with the tremor: motor deficit, cognitive deficit, cerebellar syndrome, and especially dystonia, which is often present in these patients. This chapter will describe the clinical classification, etiology, assessment and surgical management of action tremor.
#
Springer-Verlag Berlin/Heidelberg 2009
Clinical Classification Action tremor is a complex tremor, associating a postural component and an intention component [1,2]. Action tremor often has a large amplitude with a relatively low frequency (2.5–4 Hz). Severe action tremor can involve the distal and especially proximal segments of the upper limbs. Action tremor affects rotation, abduction-adduction and flexion-extension of the shoulder. Proximal tremor is easily demonstrated with the patient’s forearms in flexion and arms in abduction. Several terms have been used to describe these severe forms of action tremor, which can be a source of confusion: severe postural cerebellar tremor, rubral tremor, superior cerebellar peduncle tremor, peduncular tremor, cerebellar outflow tremor. When action tremor is less severe, it mainly affects flexion-extension movements of the wrist and fingers. Distal action tremor is more rapid than the proximal form (2.5–10 Hz). It is demonstrated by asking the patient to keep the upper limb extended. Several terms have also been used to describe these milder forms of action tremor: mild postural cerebellar tremor, postural tremor. The presence of an intention component is one of the characteristics of action tremor: the tremor becomes more severe as the finger approaches the target. This component is clearly
1758
105
Management of tremors other than essential tremor and parkinson’s disease
demonstrated during the finger-nose test and explains why patients have trouble grasping and using an object. Once again, several terms have been used to describe this intention component: cerebellar intention tremor, kinetic cerebellar tremor, ataxic tremor, hyperkinetic tremor. Action tremor may be isolated, but it is very often associated with several other symptoms depending on the cause of the tremor. It can be associated with resting tremor and then constitutes Holmes’ tremor, classically described as a low frequency tremor present at rest and during voluntary movements and frequently comprising a postural component. This type of tremor is generally observed in a context of sequelae of head injury, but remains relatively rare, especially in multiple sclerosis [3]. Action tremor is more frequently associated with cerebellar syndrome, especially in multiple sclerosis. Dysmetria, often incorrectly called cerebellar tremor is theoretically not a true tremor [4]. The finger does not reach the target and oscillates around the target, which may resemble a tremor. It can be difficult to distinguish dysmetria from action tremor, especially when both are present. The presence of other signs of cerebellar syndrome, such as hypotonia and asynergia (Stewart-Holmes sign [5]), can guide the diagnosis. Action tremor can be associated with a motor deficit that minimizes the tremor, which can disappear in the case of complete motor deficit. A partial motor deficit can be responsible for a poor functional result after surgery, even when the tremor has been completely eliminated and must therefore be taken into account in the preoperative assessment. Various forms of tremor, other than action tremor, can also justify surgical treatment due to the disability that they induce: dystonic, myoclonic and orthostatic tremor, but these relatively rare conditions will not be discussed in detail in this chapter.
Etiologies Action tremor is classically due to lesions involving the superior cerebellar peduncles, but it can also be related to lesions involving the lateral cerebellar nuclei or cerebellar vermis. Theoretically, any lesion situated in the efferent pathways of the cerebellum, especially involving the cerebellothalamic pathways, can be responsible for action tremor [6–8]. Recent studies suggest that Holmes’ tremor may be related to lesions involving the ventromedial tegmentum, including the red nucleus, subthalamic nucleus and cerebellothalamic pathways [9]. However, the role of the red nucleus in the pathogenesis of action tremor remains controversial [2,8]. Lesions of the cerebellar vermis can be responsible for bilateral action tremor and axial tremor involving the head, neck and trunk. The role of the cerebellar cortex was suggested by the work by Axelrad et al. based on post-mortem studies of patients with essential tremor [10]. The role of dysfunction of the GABAergic system was also suggested by Kralic et al. [11] and Louis ED et al. [12]. In summary, clinical observations, animal studies and certain experimental studies in man suggest that the cerebellum, thalamic nuclei and their reciprocal connections are involved in the pathogenesis of intention tremor, but further studies are necessary to evaluate the role of the basal ganglia and other structures [3]. Multiple sclerosis is the leading cause of action tremor. Cerebellar or brain stem lesions are observed in more than 80% of cases during the course of multiple sclerosis [13]. It is therefore not surprising that a large number of patients develop action tremor. However, the figures vary considerably, as Kelly et al. reported a frequency of only 3% [14], while Alusi et al., in 1999, estimated that 75% of patients with multiple sclerosis presented varying degrees of tremor [15]. A recent study estimated that 25% of
Management of tremors other than essential tremor and parkinson’s disease
patients with MS experience tremor, which is functionally disabling in 6% of cases [16]. In another recent study, Alusi SH [17] reported tremor in 58% of patients. Tremor was mild in 27% of cases, moderate in 16% of cases and severe in 15% of cases. Most patients present isolated intention tremor or severe postural tremor associated with intention tremor. Isoniazid can be effective on this mixed form, but has no effect on isolated intention tremor [18]. Resting tremor is extremely rare, observed in only 1% of cases in the series published by Pittock et al. [16]. The site of tremor was precisely described in the most recent publication by Alusi et al. [17]. It affects the upper limbs in 56% of cases, bilaterally in 36% of cases, and involves the lower limbs in 10% of cases. The head is affected in 9% of cases and the trunk is affected in 7% of cases. Other abnormal movements may be observed in the course of multiple sclerosis. Tranchant et al. [19] reported a series of 14 cases of MS, with dystonia in nine cases, Parkinsonian syndrome in three cases and myoclonus in two cases. The authors also reviewed 135 cases published in the literature. Paroxysmal dystonia is the most frequent form of abnormal movements and can be related to lesions situated in the spinal cord, internal capsule, thalamus or mesencephalon. Several cases of paroxysmal tremor, comprising postural tremor associated with dystonic features and considered to be a form of paroxysmal dystonia, have also been described [20]. Sequelae of head injury represent the second leading cause of action tremor. Tremor is usually observed after serious head injury responsible for prolonged coma. Most patients present signs indicating a brain stem lesion. The frequency of action tremor after head injury varies according to the authors from 13 to 66% of cases [21]. Tremor appears fairly rapidly after the trauma, but can resolve spontaneously after several months. The tremor is usually an action tremor
105
but some patients may also present resting tremor. In the review of the literature by Krauss et al. [22] based on 221 patients, persistent action tremor was reported in 7.7% of cases and persistent dystonic features were reported in 4.4% of cases. Myoclonus and Parkinsonian syndromes were observed in 0.5 and 0.9% of cases, respectively. Tremor is generally associated with other neurological signs: hemiparesis was observed in 40% of cases and signs of brain stem lesions were observed in 20% of cases. Lesions considered to be responsible for post-traumatic tremor are situated in the cerebellum and brain stem. Diffuse axonal lesions are observed in more than 90% of cases [22] and lesions of the dentatothalamic tract are revealed in a similar proportion of patents. Haggard et al. [23] reported the MRI findings in patients with post-traumatic action tremor present for several years. These patients essentially presented lesions of the superior cerebellar peduncles and also presented oculomotor disorders suggesting the existence of more extensive brain stem lesions. Focal brain stem lesions represent another relatively frequent cause of action tremor. Geny et al. [7] reported a case of isolated cavernoma of the brain stem responsible for severe action tremor, in which MRI demonstrated a lesion situated in the superior cerebellar peduncle homolateral to the tremor. Action tremor is rarely due to supratentorial lesions. However, thalamic lesions have been described as a possible cause of action tremor. Lee and Marsden [24] studied 62 cases of thalamic and subthalamic lesions and identified 10 cases presenting this type of tremor. In these cases, tremor was not isolated but associated with other symptoms, mainly dystonia. Luijckx et al. [25] reported three patients with infarction confined to the posterior part of the internal capsule, who presented isolated action tremor. These authors suggested that the tremor could be related to interruption of both ascending (dentatorubrothalamocortical) and descending (corticocerebellopontine) cerebellar pathways.
1759
1760
105
Management of tremors other than essential tremor and parkinson’s disease
More recently, cases of action tremor related to phenylalanine hydroxylase deficiency [26], neuropathy associated with monoclonal gammopathy [27], Wilson’s disease [28] or spinocerebellar ataxia type 2 [29] have been described.
tremor by analysis of spiral drawing [34,35]. The largest amplitude action tremors are usually observed in the context of multiple sclerosis, generating marked fatigue with the slightest attempt to move the upper limbs. Reduction of this fatigue, which is difficult to quantify, is one of the objectives of surgery in these patients [36].
Assessment Clinical assessment is an important step, as many of these patients with action tremor are candidates for stereotactic surgery, as action tremor generally induces severe disability and medical treatment tends to be relatively ineffective. However, this type of surgery is not devoid of risks and, as in all forms of functional surgery, the benefit-risk balance must be carefully evaluated. The functional benefit is not always easy to evaluate, as these patients do not always present isolated action tremor. Associated disorders, such as motor deficits or dystonia, are generally not improved by surgery and may participate in persistence of functional disability and persistence of impaired activities of daily living. Surgery is essentially indicated in patients with severe isolated action tremor, responsible for major functional disability. The severity of tremor can be assessed by clinical scales such as Webster’s score [30]. Due to the complexity of action tremor, clinical assessment must be performed during various maneuvers: arms extended, elbows flexed and finger-nose test. The intensity of tremor can be measured by accelerometry, electromyography [31] or analysis of video recordings. Bain et al. [32] proposed assessment of the severity of action tremor on a 10-point scale, which is closely correlated with impaired activities of daily living. Berk et al. [33] evaluated activities of daily living by means of the SF-36 questionnaire and a scale from 0 to 4 concerning six activities (personal hygiene, social activities, writing, dressing, eating and work). Several authors have emphasized the importance of assessing action
Surgical Treatment Medical treatment is often very disappointing. However, several treatments can be tried: clonazepam, hyoscine, isoniazid, glutethimide, carbamazepine, primidone and ondansetron. Propranolol (160–320 mg per day) is classically effective on postural tremor, but has little effect on the tremors encountered in multiple sclerosis [15]. Surgical treatment is therefore often considered due to the very disabling nature of action tremor. For a long time, thalamotomy with lesion of the ventral intermediate nucleus (Vim) was the only technique proposed. This technique was often effective, but the Vim lesion generally had to be larger than that used to treat Parkinsonian tremor. There was therefore a slightly greater risk of the lesion extending onto the internal capsule and causing a motor deficit. This technique has been gradually replaced by chronic high frequency stimulation of Vim, which is better supported especially in the case of bilateral treatment. More recently, radiosurgical (gamma knife) thalamotomy has been proposed [37] and chronic motor cortex stimulation [38] has been shown to be effective to treat this type of tremor. The results of these various treatment modalities will be analyzed below.
Thalamotomy Jankovic et al. [39] reported the long-term results of a study of 60 patients with tremor treated by thalamotomy. Twelve of these patients
Management of tremors other than essential tremor and parkinson’s disease
presented action tremor secondary to ischemic stroke in four cases (thalamic in three cases), olivopontocerebellar atrophy in two cases and severe head injury in six cases. As for all patients in the series, the stereotactic target was the Vim. Functional improvement and reduction of tremor amplitude were obtained in seven patients (58.3% of cases). Patients with post-traumatic tremor were the least improved and the best results were obtained in patients with thalamic infarction. In this group of 12 patients, 33% developed a lasting complication (ataxia, dysarthria, motor deficit). These results are in agreement with those published in the literature and confirm that the results of thalamotomy in patients with action tremor are relatively modest at the price of a high complication risk. This negative impression was counterbalanced by the results published by Krauss et al. [21], who reported the results of a series of 35 patients with post-traumatic tremor treated by thalamotomy. The stereotactic target was the Zona Incerta in 12 patients and the Zona Incerta and the inferior part of the ventrolateral nucleus of the thalamus in 23 patients. Tremor was abolished or markedly reduced in 65% of patients. Long-term functional improvement was obtained in only 37% of patients and lasting complications were observed in 38% of cases. In another publication, Krauss et al. [22] analyzed the results of 22 published series of patients with post-traumatic tremor, comprising a total of 128 cases. Tremor was improved in 88% of cases with a mean followup of 2.9 years. Functional improvement was obtained in 86% of cases, but the lasting complication rate remained high (37% of cases), mostly consisting of deterioration of preoperative disorders such as dysarthria and gait disorders. Koch et al. [3] recently reviewed the results of thalamotomy for the treatment of action tremor in patients with multiple sclerosis. He reviewed 26 series of patients published between 1960 and 2005. The most relevant results were derived from nine series (303 patients) each
105
comprising at least 20 patients. The mean minimum follow-up was 13.2 months and the mean longest follow-up was 110.6 months. Tremor has markedly improved in 77.5% of cases, but a good functional result was obtained in only 39.7% of cases. Several authors have proposed Gamma Knife Vim thalamotomy. The first series included patients with Parkinsonian tremor. In view of the encouraging results, Mathieu et al. [37] proposed extending the indications to action tremor in patients with multiple sclerosis. He treated six patients with multiple sclerosis presenting disabling action tremor with a mean follow-up of 27.5 months. All patients were improved an average of 2.5 months after treatment. Four patients obtained a marked functional improvement with, in particular, improvement of writing and drawing. One patient developed postoperative hemiparesis, which regressed after administration of corticosteroids. The authors suggested that Gamma Knife thalamotomy could be proposed when thalamic stimulation is considered to be too dangerous in these patients who often have a very frail general and neurological status. However, it should be stressed that the target in the context of treatment of action tremor has a more variable site and is more extensive than that used for the treatment of Parkinsonian tremor. Larger series would be expected to reveal a higher failure rate and a greater number of complications in this indication.
Thalamic Stimulation In 1996, Benabid et al. [40] reported the results of a series of 117 patients with tremor treated by chronic high frequency stimulation of Vim. Eleven patients presented action tremor, due to multiple sclerosis in four cases and head injury or mesencephalic hemorrhage in seven cases. These patients were part of a group of 17 patients considered to suffer from dyskinesia. The results
1761
1762
105
Management of tremors other than essential tremor and parkinson’s disease
on action tremor are difficult to evaluate, as the results were presented globally for the overall patient population. Assessment of the results was mainly based on analysis of the Fahn score [41], a 5-point clinical scale, with no functional assessment. The authors considered that 57% of dyskinetic patients were improved. Action tremor was improved in two of the four patients with multiple sclerosis. The only complication observed in this group was delayed onset (after 8 days) of motor neglect, which resolved spontaneously over 3 months. In the personal experience reported by Siegfried and Lippitz [42], based on 60 patients with disabling tremor treated by thalamic stimulation, 10 presented action tremor related to multiple sclerosis and tremor was perfectly controlled in seven cases. In 1996, we reported our experience of 13 cases of action tremor related to multiple sclerosis treated by unilateral Vim stimulation [36]. All patients were severely disabled (mean EDSS score [43]: 6.8) and presented severe action tremor interfering with use of the upper limb. The absence of an acute attack during the previous 6 months was one of the main selection criteria. Tremor, particularly its proximal component, was significantly improved in 69.2% of cases. Functional improvement was mainly obtained in patients whose disability was almost exclusively related to tremor (functional improvement in 80% of cases). Six patients were able to eat unassisted after the procedure. One patient developed a transient postoperative motor deficit. No deterioration of the EDSS was observed at 3 months. In 2001, we analyzed a larger series comprising 37 patients presenting similar clinical features [44]. In addition to the EDSS, clinical assessment comprised Webster’s score [30], Brown’s functional score [45], Bain’s visual analogue scale [32] and a self-assessment scale. With a mean follow-up of 21 months, 78.3% of patients were considered to be improved with a statistically significant improvement for three of the four scales. Brown’s functional score was not significantly
improved, probably because this score comprises several bimanual tasks that remain difficult to perform despite a marked unilateral improvement (32 procedures were exclusively unilateral). Tremor gradually recurred in three patients (8.1% recurrence rate). An incomplete but permanent deficit of the lower limbs, contralateral to the procedure, was observed postoperatively. Gait was transiently worsened in two patients and one case of infection was observed. This study confirms that thalamic stimulation is less invasive than thalamotomy in this indication and is effective in at least 70% of patients, as reported in the series by Siegfried and Lippitz [42]. Koch et al. [3] recently reviewed the main publications concerning this indication. He reported the results of 20 publications, but only eight publications concerned real patient series (between 9 and 14 consecutive patients) with a mean follow-up of 12 months (range: 3–48 months). Quantitative assessment of the improvement of tremor was reported in five publications. Improvement ranged from 69 to 100% (mean: 91.4%). Complete absence of functional improvement was reported in two publications. Three publications reported functional improvement ranging from 64 to 92% (mean: 81.3%). This meta-analysis demonstrated a certain discordance between almost constant improvement of action tremor and the functional result which was variable or assessed in various ways, raising the problem of functional assessment scales. Berck’s study [33] is particularly interesting, as this author assessed the results of thalamic stimulation in terms of quality of life, in a prospective study including 12 patients with a minimum follow-up of 1 year. Quality of life was assessed in terms of six activities of daily living and by the SF-36 questionnaire. Assessment at 1 year revealed no significant improvement of SF-36, but a significant improvement of the ability to eat and dress unassisted, maintain social relationships and ensure personal hygiene.
Management of tremors other than essential tremor and parkinson’s disease
105
The patients most markedly improved were younger patients with a good intellectual level and recent-onset disease. Experience of thalamic stimulation on posttraumatic action tremor is relatively limited, as, in a review of the literature in 2002, Krauss et al. [22] identified only 17 cases reported in eight publications. The mean follow-up was 10 months. Tremor was markedly improved in all but two publications and a marked functional improvement was reported in all but one publication, in which it was not assessed. No adverse effect was reported.
was combined with more anterior stimulation involving Voa and Vop. These findings were not confirmed by Lim et al. [54], who found that simultaneous stimulation of these two targets was not more effective than stimulation of a single target, either Vim or Voa. All these results indicate that Vim remains the major target, but, when intraoperative clinical testing shows that Vim stimulation is not effective on the tremor, the target should include Vop or Voa and Zona Incerta.
Stereotactic Target
Thalamotomy has been demonstrated to be very effective in both multiple sclerosis and posttraumatic tremor, but is associated with a high complication rate. Thalamic stimulation may be less effective, but appears to induce much fewer complications. This impression was confirmed in a study by Bittar et al. [55] comparing 10 cases of thalamotomy to 10 cases of thalamic stimulation in the treatment of action tremor related to multiple sclerosis. These authors concluded that thalamic stimulation should be the first-line treatment option proposed to these patients, as it is associated with a much lower risk than thalamotomy. In a review of the literature published by Yap et al. [56], thalamotomy and thalamic stimulation induced an equivalent degree of improvement, but also similar complication rates. Functional improvement was observed more frequently with stimulation. Four deaths following thalamotomy have been reported in the literature, but three of the four cases of postoperative hemorrhage occurred after insertion of a stimulation electrode.
Most authors consider the Vim to be the most effective target to treat tremor, either resting tremor, postural tremor or action tremor, whether by inducing a lesion or by chronic high frequency stimulation [21,39,46,47]. However, Ohye et al. [48] already reported that treatment of action tremor required a larger thalamotomy, especially in terms of the height of the lesion, i.e., involving the inferior part of Vim but also its middle or even superior parts. We reported similar findings based on our experience of chronic stimulation [44]. The size of the target varies from patient to patient, emphasizing the need to adapt the target to each case, hence the importance of intraoperative clinical testing, which cannot be performed with Gamma Knife thalamotomy. Whittle et al. [49] as other authors [15,50,51] observed that the results of deep brain stimulation on action tremor were better when the target was situated more anteriorly and more inferiorly than the Vim, which results in stimulation of the Zona Incerta. Hamel et al. [52] also suggested that a subthalamic target (including Zona Incerta, prelemniscal projections and cerebellothalamic projections), would be more effective on action tremor than stimulation of Vim. Foote et al. [53] reported better results when stimulation of Vim
Thalamotomy or Thalamic Stimulation
Other Treatment Options Chronic Vim stimulation is the first treatment option to be proposed to patients, but what other
1763
1764
105
Management of tremors other than essential tremor and parkinson’s disease
options can be proposed in the case of failure? We have reported a case of marked long-term improvement of action tremor after chronic motor cortex stimulation [38], but this result was not confirmed in other patients (unpublished results). However, Sandyk et al. [57] reported good results in three patients treated by external cortical stimulation. Recent studies [58] suggest that chronic vagus nerve stimulation could also be useful in the treatment of action tremor in multiple sclerosis.
Conclusion Action tremor is a complex tremor that is difficult to treat medically. Chronic stimulation or a lesion of the Vim can markedly improve tremor and provide functional improvement. This improvement is much better in young subjects with a short history of disease, presenting few neurological deficit associated with tremor. A consensus has almost been reached to propose deep brain stimulation directed onto the Vim as first-line treatment. When Vim stimulation is ineffective on intraoperative clinical testing, the Zona Incerta or Voa-Vop complex should be tested. In the presence of an obvious thalamic lesion, motor cortex stimulation can be considered when a trial of external cortical stimulation is positive.
References
6. 7.
8.
9.
10.
11.
12.
13. 14.
15. 16.
17. 18.
1. Elbe RJ, Koller WC. Cerebellar tremor. In: Elbe RJ, Koller WC, editor. Tremor. Baltimore/London: The Johns Hopkins University Press; 1990. p. 106-17. 2. Sabra AF, Hallet M. Action tremor with alternating activity in antagonist muscles. Neurology 1984;34:151-6. 3. Koch M, Mostert J, Heersema D, De Keyser J. Tremor in multiple sclerosis. J Neurol 2007;254:133-45. 4. Rondot P, Jedynak CP, Ferrey G. Pathological tremors: nosological correlates. Prog Clin Neurophysiol 1978; 5:95-113. 5. Waubant E, Tezenas du Montcel S, Jedynak C, Obadia M, Hosseini H, Damier P, Lubetzki C, Agid Y, Degos JD.
19. 20.
21.
Multiple sclerosis tremor and the Stewart-Holmes manoeuvre. Mov Disord 2003;18:948-52. Anouti A, Koller WC. Tremor disorders. Diagnosis and management. West J Med 1995;162:510-13. Geny C, Nguyen JP, Cesaro P, Goujon C, Brugie`res P, Degos JD. Thalamic stimulation for severe action tremor after lesion of the superior cerebellar peduncle. J Neurol Neurosurg Psychiatry 1995;59:641-2. Remy P, de Recondo A, Defer G, Loc’h C, Amarenco P, Plante-Bordeneuve V, Dao-Castellana MH, Bendriem B, Crouzel C, Clanet M. Peduncular ‘‘rubral’’ tremor and dopaminergic denervation: a PET study. Neurology 1995;45:472-7. Ohye C, Shibazaki T, Hirai T, Kawashima Y, Hirato M, Matsumura M. Tremor mediating thalamic zone studied in humans and in monkeys. Stereotact Func Neurosurg 1993;60:136-45. Axelrad JE, Louis ED, Honig LS, Flores I, Ross GW, Pahwa R, Lyons KE, Faust PL, Vonsattel JP. Reduced Purkinje cell number in essential tremor: a postmortem study. Arch Neurol 2008;65:101-7. Kralic JE, Criswell HE, Osterman JL, O’Buckley TK, Wilkie ME, Matthews DB, Hame K, Breese GR, Homanics GE, Morrow AL. Genetic essential tremor in gamma-aminobutyric acidA receptor alpha1 subunit knockout mice. J Clin Invest 2005;115:584-6. Louis ED. A new twist for stopping the shakes? Revisiting GABAergic therapy for essential tremor. Arch Neurol 1999;56:807-8. Bauer HJ. Problems of symptomatic therapy in multiple sclerosis. Neurology 1978;28:8-20. Kelly R. Clinical aspects of multiple sclerosis. In Handbook of clinical neurology: demyelinating diseases, vol. 3. New York: Elsevier Science Publishers; 1988. p. 49-78. Alusi SH, Glickman S, Aziz TZ. Tremor in multiple sclerosis. J Neurol Neurosurg Psychiatry 1999;66:131-4. Pittock SJ, McClelland RL, Mayr WT, Rodriguez M, Matsumoto JY. Prevalence of tremor in multiple sclerosis and associated disability in the Olmsted County population. Mov Disord 2004;19:1482-5. Alusi SH, Worthington J, Glickman S, Bain PG. A study of tremor in multiple sclerosis. Brain 2001;124:720-30. Bozec CB, Kastrukoff LF, Wright JM, Perry TL, Larsen TA. A controlled trial of iszoniazid therapy for action tremor in multiple sclerosis. J Neurol 1987;234:36-9. Tranchant C, Bhatia P, Marsden CD. Movement disorders in multiple sclerosis. Mov Disord 1995;4:418-23. Nardocci N, Zorzi G, Savoldelli M, Rumi V, Angelini L. Paroxysmal dystonia and proximal tremor in a young patient with multiple sclerosis. Ital J Neurol Sci 1995;16:315-9. Krauss JK, Mohadjer M, Nobbe F, Mundinger F. The treatment of post-traumatic tremor by stereotactic surgery. Symptomatic and functional outcome in a series of 35 patients. J Neurosurg 1994;80:810-19.
Management of tremors other than essential tremor and parkinson’s disease
22. Krauss JK, Jankovic J. Head injury and posttraumatic movement disorders. Neurosurgery 2002;50:927-40. 23. Haggard P, Miall C, Wade D, Fowler S, Richardson A, Anslow P, Stein J. Damage to cerebellocortical pathways after closed head injury: a behavioural and magnetic resonance imaging study. J Neurol Neurosurg Psychiatry 1995;58:439-43. 24. Lee MS, Marsden CD. Movement disorders following lesions of the thalamus or subthalamic region. Mov Disord 1994;5:493-507. 25. Luijckx GJ, Boiten J, Lodder J, Heuts-van Raak L, Wilmink J. Isolated hemiataxia after supratentorial brain infarction. J Neurol Neurosurg Psychiatry 1994;57:742-4. 26. Payne MS, Brown BL, Rao J, Payne BR. Treatment of phenylketonuria-associated tremor with deep brain stimulation: case report. Neurosurgery 2005;56:E868. 27. Ruzicka E, Jech R, Zarubova K, Roth J, Urgosik D. VIM thalamic stimulation for tremor in a patient with IgM paraproteinaemic demyelinating neuropathy. Mov Disord 2003;18:1192-5. 28. Pal PK, Sinha S, Pillai S, Taly AB, Abraham RG. Successful treatment of tremor in Wilson’s disease by thalamotomy: a case report. Mov Disord 2007;22:2287-90. 29. Freund HJ, Barnikol UB, Nolte D, Treuer H, Auburger G, Tass PA, Samii M, Sturm V. Subthalamic-thalamic DBS in a case with spinocerebellar ataxia type 2 and severe tremor-a unusual clinical benefit. Mov Disord 2007;22:732-5. 30. Webster DD. Critical analysis of the disability in Parkinson’s disease. Mod Treat 1968;5:257-82. 31. Spieker S, Boose A, Jentgens C, Dichgans J. Long-term recordings in parkinsonian and essential tremor. J Neural Transm 1995;46:339-49. 32. Bain PG, Findley LJ, Atchison P, Behari M, Vidailhet M, Gresty M, Rothwell JC, Thompson PD, Marsden CD. Assessing tremor severity. J Neurol Neurosurg Psychiatry 1993;56:868-73. 33. Berck C, Carr J, Sinden M, Martzke J, Honey CR. Thalamic deep brain stimulation for the treatment of tremor due to multiple sclerosis: a prospective study of tremor and quality of life. J Neurosurg 2002;97:815-20. 34. Bain PG, Findley LJ. Spirography. In: Bain PJ, Findley LJ, editors. Standards in neurology. Book 1: Assessing tremor severity. London: Smith-Gordon; 1993. p. 9-19. 35. Elbe RJ, Brilliant M, Leffler K, Higgins C. Quantification of essential tremor in writing and drawing. Mov Disord 1996;11:70-8. 36. Geny C, Nguyen JP, Pollin B, Fe`ve A, Ricolfi F, Cesaro P, Degos JD. Improvement of severe postural cerebellar tremor in multiple sclerosis by chronic thalamic stimulation. Mov Disord 1996;11:489-94. 37. Mathieu D, Kondziolka D, Nirajan A, Flickinger J, Lunsford D. Gamma knife thalamotomy for multiple sclerosis tremor. Surgical neurol 2007;68:394-9.
105
38. Nguyen JP, Pollin B, Fe`ve A, Geny C, Cesaro P. Improvement of action tremor by chronic cortical stimulation. Mov Disord 1998;13:84-8. 39. Jankovic C, Cardoso F, Grosmann RG, Hamilton WJ. Outcome after stereotactic thalamotomy for parkinsonian, essential, and other types of tremor. Neurosurgery 1995;37:680-7. 40. Benabid AL, Pollak P, Gao D, Hoffman D, Limousin P, Gay E, Payan I, Benazzouz A. Chronic electrical stimulation of the ventralis intermedius nucleus of the thalamus as a treatment for movement disorders. J Neurosurg 1996;84:203-14. 41. Fahn S, Elton RL, UPDRS development committee. Unified Parkinson’s disease rating scale. In: Fahn S, Marsden CD, Goldstein M, Calne DM, editors. Recent developments in Parkinson’s disease. Florham Park, NJ: Macmillan Health Care Information; 1987. p. 153-64. 42. Siegfried J, Lippitz B. Chronic electrical stimulation of the VL-VPL complex and of the pallidum in the treatment of movement disorders: personal experience since 1982. Stereotact Funct Neurosurg 1994;62:71-5. 43. Kurtzke JF. Rating neurological impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology 1983;33:1444-52. 44. Capparos-Lefebvre D, Blond S, Nguyen JP, Pollak P, Benabid AL. Chronic deep brain stimulation for movement disorders. In: Cohadon F, editor. Advances and technical standards in neurosurgery, vol. 25. Vienna: Springer; 1999. p. 61-138. 45. Brown RJ, MacCarthy B, Jahanshahi M. Accuracy of self reported disability in patients with parkinsonism. Arch neurol 1989;46:955-9. 46. Derome PJ, Jedynak CP, Visot A, Delalande O. Treatment of involuntary movements by thalamic lesions. Rev Neurol 1986;142:391-7. 47. Nguyen JP, Degos JD. Thalamic stimulation and proximal tremor. A specific target in the ventrointermedius thalami. Arch Neurol 1993;50:498-500. 48. Ohye C, Shibazaki T, Hirai T, Wada H, Hirato M, Kawashima Y. Further physiological observations on the ventralis intermedius neurons in the human thalamus. J Neurophysiol 1989;61:488-500. 49. Whittle IR, Yau YH, Hooper J. Mesodiencephalic targeting of stimulating electrodes in patients with tremor caused by multiple sclerosis. J Neurol Neurosurg Psychiatry 2004;75:1210. 50. Nandi D, Liu X, Bain P, Parkin S, Joint C, Winter J, Stein J, Scott R, Gregory R, Aziz T. Electrophysiological confirmation of the zona incerta as a target for surgical treatment oa`f disabling involuntary arm movements in multiple sclerosis: use of local field potentials. J Clin Neurosci 2002;9:64-8. 51. Plaha P, Khan S, Gill SS. Bilateral stimulation of the caudal zona incerta nucleus for tremor control. J Neurol Neurosurg Psychiatry 2008;79:504-13.
1765
1766
105
Management of tremors other than essential tremor and parkinson’s disease
52. Hamel W, Herzog J, Kopper F, Pinsker M, Weinert D, Mu¨ller D, Krack P, Deuschl G, Mehdorn HM. Deep brain stimulation in the subthalamic area is more effective than nucleus ventralis intermedius stimulation for bilateral intention tremor. Acta Neurochir 2007;149:749-58. 53. Foote KD, Okun MS. Ventralis intermedius plus ventralis oralis anterior and posterior deep brain stimulation for posttraumatic Holmes tremor: two leads may be better than one: technical note. Neurosurgery 2005;56:E445. 54. Lim DA, Khandhar SM, Heath S, Ostrem JL, Ringel N, Starr P. Multiple target deep brain stimulation for multiple sclerosis related and poststroke Holmes’ tremor. Stereotact Funct Neurosurg 2007;85:144-9. 55. Bittar RG, Hyam J, Nandi D, Wang SY, Liu X, Joint C, Bain PG, Gregory R, Stein J, Aziz TZ. Thalamotomy
56.
57.
58.
59.
versus thalamic stimulation for multiple sclerosis tremor. J Clin Neurosci 2005;12:638-42. Yap L, Kouyialis A, Varma TR. Stereotactic neurosurgery for disabling tremor in multiple sclerosis: thalamotomy or deep brain stimulation ? Br J Neurosurg 2007;21:349-54. Sandyk R, Dann LC. Weak electromagnetic fields attenuate tremor in multiple sclerosis. Int J Neurosci 1994;79:199-212. Marrosu F, Maleci A, Cocco E, Puligheddu M, Barberini L, Marrosu MG. Vagal nerve stimulation improves cerebellar tremor and dysphagia in multiple sclerosis. Mult Scler 2007;13:1200-2. Alusi SH, Aziz TZ, Glickman S, Jahanshahi M, Stein JF, Bain PG. Stereotactic lesional surgery for the treatment of tremor in multiple sclerosis: a prospective case-controlled study. Brain 2001;124:1576-89.
90 Medical Management of Parkinson’s Disease E. V. Encarnacion . R. A. Hauser
Introduction Parkinson’s disease is a progressive neurodegenerative disorder characterized by loss of dopaminergic neurons of the substantia nigra pars compacta (SNc). This occurs predominantly in the ventrolatral tier and is accompanied by the presence of proteinacious intracytoplasmic Lewy bodies that contain aggregated a-synuclein. This neuronal loss results in decreased dopaminergic innervation of the striatum, most prominent in the dorsal putamen. Motor dysfunction occurs when approximately 50% of nigral dopaminergic cells are lost, and striatal dopamine is reduced by approximately 80% [1]. It is now recognized that Lewy body pathology in Parkinson’s disease begins in the brain in the olfactory bulbs and lower brainstem, and ascends over time to affect the substantia nigra and later the cerebral cortex. In addition, other areas of neuronal loss include noradrenergic neurons of the locus coeruleus (LC), serotonergic neurons of the dorsal raphe (DR), cholinergic neurons of the nucleus basalis of Meynert (NBM), neurons of the dorsal motor nucleus of the vagus nerve, and peripheral autonomic neurons. Clinically, PD is characterized by the presence of bradykinesia, rigidity, and tremor. The tremor is typically a 3–4 Hz resting tremor, but it may also include lower amplitude postural and kinetic components. About 30% of patients with PD do not have tremor, but one needs to be more cautious about the diagnosis in these individuals. PD is usually asymmetric with one side being
#
Springer-Verlag Berlin/Heidelberg 2009
affected first and remaining worse throughout the course of the disease. Another important feature is a robust and sustained response to dopamine replacement therapy. Improvement with dopaminergic medications is usually most obvious for bradykinesia and rigidity as the response of tremor is variable. Nonmotor symptoms are increasingly recognized as a source of disability in PD, and some of these can occur at any time during the course of the disease, such as depression and anxiety, fatigue, sleep disturbances, and autonomic dysfunction. Although medications provide good benefit for motor features of PD for several years, most patients’ response is complicated by the development of motor fluctuations and dyskinesias. In the modern era, the incidence of motor fluctuations after 4–6 years of levodopa therapy has been reported as 12–60% and for dyskinesias 8–64% [2]. These problems tend to worsen over time and 59–100% of patients experience dyskinesias by 10 years [3–5]. It has been estimated that in approximately 43% of cases, fluctuations and dyskinesias can be adequately controlled with medication adjustment [5]. Fluctuations and dyskinesias can have a large negative impact on quality of life if not adequately controlled, and these patients are commonly considered for surgical intervention. However, it is important to note that by 15 years, most PD patients have substantial disability, usually due to non-dopaminergic features such as cognitive impairment and balance difficulty, features that are not improved with surgical intervention.
1508
90
Medical management of parkinson’s disease
The differential diagnosis of PD consists chiefly of essential tremor (ET) and the Atypical Parkinsonisms (AP). In contrast to PD, ET is characterized by a relatively symmetric postural or kinetic tremor without rigidity and little or no bradykinesia. There may be a family history, and a transient response to alcohol is common. The APs include Multiple System Atrophy (MSA) and Progressive Supranuclear Palsy (PSP) in which there is more widespread neuronal loss than that seen in PD, including degeneration in the striatum. These conditions are characterized by relatively symmetric parkinsonism (bradykinesia, rigidity) with little or no tremor, early speech and balance difficulty, axial rigidity, and little or no response to dopaminergic medical treatment. Conceptually, the ideal pharmacological approach is to identify individuals with PD as early as possible and to administer a neuroprotective agent to slow or stop disease progression. It may be possible to identify individuals with PD prior to their developing classic motor features (bradykinesia, rigidity, tremor). Most patients have olfactory dysfunction prior to onset of motor features, and REM behavior disorder (RBD, loss of normal atonia during REM sleep) occurs prior to motor features in some individuals. Mid-life risk factors for PD include constipation, sleepiness, and adiposity. At risk individuals might be further evaluated using PET or SPECT imaging to identify loss of dopaminergic neurons prior to development of motor features. At the current time, there is no medication proven to slow progression of disease, but multiple candidate agents are being evaluated. For now, treatment is symptomatic with the goal being to provide the best quality of life over time as possible. There is a growing awareness that both motor and non-motor symptoms must be addressed, and this has reshaped modern medical management of PD.
Motor Symptoms Medications Carbidopa/Levodopa (Sinemet) Levodopa remains the most efficacious treatment for the motor symptoms of PD but is associated with the development of motor complications, including motor fluctuations (wearing off) and dyskinesia. Because dopamine itself does not cross the blood-brain barrier (BBB), attention turned to its metabolic precursor, levodopa (l-dihydroxyphenylalanine). Levodopa is an aromatic amino acid that readily crosses the BBB. After oral ingestion, levodopa is absorbed in the proximal small intestine by way of a carriermediated large neutral amino acid transport system and crosses the BBB through a similar transport system. Large neutral amino acids in meals can reduce levodopa absorption and brain transport, but this is usually only noticeable clinically in patients who have developed motor fluctuations and are sensitive to small changes in levodopa delivery to the brain. Once absorbed, levodopa can be decarboxylated peripherally to form dopamine which causes nausea and vomiting. Therefore, levodopa is combined with a peripheral dopa decarboxylase inhibitor (DDCI), such as carbidopa or benzaride, to reduce nausea and vomiting. The addition of a DDCI increases levodopa’s central bioavailability and increases its plasma half-life from approximately 60 to 90 min. Despite this short plasma half-life, levodopa can initially be administered on a TID or QID schedule and provide benefit that lasts from dose to dose. This presumably reflects levodopa uptake into remaining dopaminergic neurons, conversion to dopamine, intraneuronal storage, and slow release of dopamine into the synapse over time. As more and more dopaminergic neurons are lost, this storage and release mechanism may be lost, and
Medical management of parkinson’s disease
levodopa’s clinical effect becomes shorter and shorter in duration until clinical benefit closely mirrors its serum pharmacokinetics. Carbidopa/levodopa is available in immediate release (IR) 10/100, 25/100, and 25/250 mg tablets and controlled release (CR) 25/100 and 50/200 mg tablets. It has been estimated that it takes approximately 80 mg of carbidopa per day to saturate peripheral decarboxylase, although some patients may need as much as 200 mg per day, to avoid nausea. The initial levodopa target dose is typically carbidopa/levodopa IR 25/100 TID or QID or CR 50/200 BID. A slow titration is usually employed in an effort to minimize side effects. Carbidopa/levodopa IR can be started at 25/100 ½ tablet per day, increasing by ½ tablet every week until the target dose is reached. Potential side effects include nausea (secondary to dopamine stimulation of the area postrema in the medulla), orthostatic hypotension, confusion, hallucinations, and somnolence. Patients are monitored clinically and the daily levodopa dose is usually slowly increased over time in an effort to maintain control of motor symptoms as the disease progresses. Long term administration of levodopa is associated with the development of motor fluctuations and dyskinesias. The risk of developing fluctuations and dyskinesias is correlated to the total daily levodopa dose, so unnecessarily high dosages should be avoided. Although the exact mechanism responsible for the development of motor complications is unknown, it appears that pulsatile stimulation of dopamine receptors leads to downstream changes that are expressed clinically as fluctuations and dyskinesia. Patients who are only experiencing wearing off motor fluctuations are usually easily treated by one of several strategies including moving levodopa doses closer together, switching from IR to CR formulations, or adding adjuncts such as MAO-B inhibitors, COMT inhibitors, or dopamine agonists. The situation is much more difficult in patients
90
who are experiencing both fluctuations and dyskinesias. In these individuals, decreasing dopaminergic medication is likely to worsen OFF time (time when medication has worn off) and increasing dopaminergic medication is likely to worsen dyskinesia. Amantadine is often helpful to reduce dyskinesia but must be used with caution in patients with underlying cognitive dysfunction. Typically, smaller levodopa doses are administered more frequently. Adjunct medications are then used to ‘‘fine tune’’ the response. Nonetheless, a balance between OFF time and dyskinesia may be the best possible outcome. Because the development of motor complications appears to be related in large measure to pulsatile stimulation caused by levodopa, there are efforts underway to develop very long acting levodopa preparations (8–24 h) that might capture the marked efficacy and good side effect profile of levodopa and avoid the development of motor complications. Carbidopa/Levodopa Oral Disintegrating Tablet – ODT (Parcopa)
Parcopa is an orally disintegrating immediate release form of carbidopa/levodopa that was approved based on demonstration of bioequivalence with carbidopa/levodopa IR tablets. Its clinical efficacy and side effect profile are therefore similar to carbidopa/levodopa IR. Parcopa rapidly dissolves on the tongue, travels down the gut in saliva, and is absorbed in the small intestine. It does not require water intake, thereby providing increased convenience for patients who at times may not have easy access to water. It is also helpful for patients with dysphagia as it does not need to be swallowed. Parcopa comes in 10/100, 25/100 and 25/250 mg tablets and its dosing schedule is similar to carbidopa/ levodopa IR, initially given 3–4 times daily. It contains variable amounts of phenylalanine and should be used with caution in individuals with phenylketonuria.
1509
1510
90
Medical management of parkinson’s disease
Carbidopa/levodopa/entacapone (Stalevo)
Stalevo is a combination of levodopa, carbidopa, and entacapone and is discussed below.
Dopamine Agonists Dopamine agonists directly stimulate post-synaptic striatal dopamine receptors. In early PD, oral dopamine agonists provide antiparkinsonian efficacy that although not as great as levodopa, is usually sufficient to control motor symptoms for several years. Available oral dopamine agonists are associated with a relatively low risk of development of fluctuations and dyskinesias compared to levodopa. In contrast to levodopa, which has a very short halflife, oral dopamine agonists have relatively long half-lives (>6 h) and may therefore avoid pulsatile dopaminergic stimulation. Multicenter trials noted lower incidences of dyskinesias [6–9] and wearing off [7,9] with dopamine agonists compared to levodopa as initial treatment. Dopamine agonists are therefore now widely accepted as an option to treat early PD to forestall the occurrence of motor complications. Most patients will eventually need levodopa after 1–3 years of agonist monotherapy. Another potential rationale for the use of dopamine agonists in early PD is the possibility of neuroprotection. In the laboratory, dopamine agonists have been shown to exert direct antioxidant and receptor-mediated anti-apoptotic effects [10]. A putative neuroprotective effect was supported by PET and SPECT studies that showed slower rates of progression of imaging markers in patients receiving early agonist monotherapy compared to levodopa monotherapy [11,12]. However, possible pharmacokinetic and pharmacodynamic effects confound interpretation of these findings and a neuroprotective effect of dopamine agonists has not been proven in PD patients. Oral dopamine agonists include the newer non-ergots, pramipexole and ropinorole, and the older ergots, bromocriptine, pergolide and
cabergoline. Recently, a rotigotine (non-ergot) transdermal patch was approved for use in early PD but was voluntarily withdrawn from the U.S. market due to crystallization within the patch. The ergot-derived compounds have been associated with pleural, pericardial, and retroperitoneal fibrosis [13–15] and cardiac valvulopathy [16–19]. Pergolide has been voluntarily withdrawn from the market because of valvulopathy, while bromocriptine is now rarely used in the U.S. These two medications will not be discussed further. Common side effects of dopamine agonists include somnolence, nausea, edema, orthostatic hypotension, constipation, and hallucinations. Comorbid illnesses are important risk factors for the development of somnolence, edema, and hallucinations [20] while decline in cognitive function [20,21], older age [20,21], duration of disease [21], history of depression [21], and history of sleep disorder [21] increase the risk of hallucinations. Sudden sleep attacks can occur [22–24] but are relatively infrequent. Nonetheless, patients should be warned about the potential for sleepiness and should not drive if they are falling asleep at inappropriate times. Impulse control disorders have recently been described as a potential side effect of dopamine agonists and can include pathologic gambling, shopping, buying, internet use, or sexual activity. Pramipexole (Mirapex)
Pramipexole is a non-ergot synthetic aminobenzathiazol derivative that is active at D2, D3, and D4 receptors. It has lower affinity for betaadrenergic receptors, muscarinic acetylcholine receptors and 5-HT receptors. After oral ingestion, pramipexole reaches peak plasma levels in 1–3 h and has an elimination half-life of 8–12 h. It is less than 20% protein bound and is excreted mostly unchanged in the urine. It is available in 0.125, 0.25, 0.5, 1.0 and 1.5 mg tablets. Pramipexole is commonly introduced at 0.125 mg TID and slowly escalated to 0.5 mg TID within 4–5 weeks. The usual maximum dosage is 4.5 mg per day.
Medical management of parkinson’s disease
The CALM-PD study (Comparison of the Agonist Pramipexole With Levodopa on Motor Complications of Parkinson’s Disease) showed that pramipexole monotherapy in early PD delayed the occurrence of motor complications compared to initial treatment with carbidopa/ levodopa IR [7,25]. As an adjunct to levodopa in patients with motor fluctuations, pramipexole has been shown to decrease ‘‘off ’’ time, improve activities of daily living and motor scores, and allow a reduction in levodopa dosage [26,27]. It was also demonstrated to be superior to placebo as an adjunct to levodopa in reducing drug resistant tremor [28]. Ropinorole (Requip)
Ropinorole is a nonergot dopamine agonist with high affinity for D2 receptors, and no significant D1 or D5 activity. Unlike pramipexole, it lacks affinity for adrenergic, cholinergic and serotonergic receptors. It reaches peak plasma levels in 1–2 h in fasted patients and approximately 4 h when taken with meals. Ropinirole is 40% protein bound. It is metabolized by the P450 CYP1A2 hepatic enzyme system with an elimination half-life of 6 h. Patients on ciprofloxacin may experience increased plasma levels of ropinorole. Ropinirole is available in 0.25, 0.5, 1.0, 2.0, 3.0, 4.0 and 5.0 mg tablets. It is commonly started at 0.25 mg TID and slowly titrated upward. Clinical efficacy is usually seen at 9 mg a day in early disease, but higher doses are often necessary as the disease advances. The usual maximum dose is 24 mg per day. Initial treatment with ropinorole has been shown to delay the onset of motor complications compared to initial treatment with levodopa [6]. As an adjunct to levodopa in patients with motor fluctuations, ropinorole has been shown to reduce ‘‘off ’’ time, and allow a reduction in levodopa dose [29]. Ropinirole 24-h prolonged release (ropinirole 24-h) is currently being investigated and has been shown to be safe and well tolerated as an adjunct to levodopa [30]. The prolonged release
90
formulation was developed as a once-a-day formulation allowing a simple and faster dosetitration schedule. Ropinirole 24-h allows a steady rate of absorption and reduces plasma concentration fluctuations. Cabergoline
Cabergoline is an ergot-derived agonist with specificity for D2 receptor, available outside the U.S. It is much longer-acting than other dopamine agonists with a half-life of greater than 65 h and is given once daily. Cabergoline has demonstrated efficacy as monotherapy in early PD [31] and as an adjunct in patients with motor fluctuations it reduces off time [32], increases on time [33], and allows a reduction of levodopa dose [33]. It has been associated with valvulopathy, similar to other ergot derived agonists [34]. Rotigotine Transdermal System (Neupro)
Rotigotine [()-5, 6, 7, 8-tetrahydro-6-[propyl[2-(2-thienyl-ethyl) amino-1-naphthalenol HCL] was the most recent dopamine agonist to be approved. It was recently withdrawn from the U.S. market due to crystallization in some batches. It is a lipid soluble non-ergot full D3/D2/D1 and partial D4 dopamine receptor agonist that has demonstrated efficacy in both early and advanced Parkinson’s disease and was approved as a treatment for early PD. It is formulated in a siliconebased transdermal patch and delivered over 24h with once daily application, achieving relatively constant plasma concentrations. It avoids variation in gastrointestinal absorption due to delayed gastric emptying and also avoids hepatic firstpass effects. The simple dosing regimen may improve compliance and it is useful in patients with dysphagia. Rotigotine patch is generally well tolerated and has a side effect profile similar to oral dopamine agonists. It provided dose-related antiparkinsonian effects as monotherapy in early disease [35,36], and significantly improved ‘‘off ’’ time in those not optimally controlled with levodopa in advanced disease [37].
1511
1512
90
Medical management of parkinson’s disease
Rotigotine is initiated at 2 mg/24 h and titrated in 2 mg/24 h increments each week until optimal effect is observed, up to 6 mg/24 h. The most common side effects are nausea, application site reactions, somnolence, dizziness, and headache. The patch can easily be removed if adverse effects occur. Apomorphine (Apokyn)
Apomorphine is a non ergot-derived, directacting dopamine agonist with strong D1 and D2 dopamine receptor-stimulating properties. When administered orally, apomorphine has poor bioavailability because of extensive first pass hepatic metabolism. It is currently approved in the US as a subcutaneous injection. When administered in this way, maximal plasma concentration is reached in 8 min, and its half-life life is about 30 min. For patients with motor fluctuations, it can reverse ‘‘off ’’ periods within 10–15 min and usually lasts about 90 min. It is therefore useful for patients with motor fluctuations, particularly those who wish to reverse ‘‘off ’’ periods rapidly. This can include refractory or unexpected ‘‘off ’’ periods, and nighttime or early morning ‘‘off ’’ periods. Initiation of subcutaneous apomorphine is usually undertaken in the physician’s office while orthostatic blood pressures are monitored, and the effective dose delineated. An initial test dose of 0.2 mL (2 mg) is administered and if there are no meaningful side effects and no apparent efficacy, additional doses are administered up to a maximum recommended dose of 0.6 mL (6 mg). It is currently recommended that patients receive the antiemetic trimethobenzamide (Tigan) 300 mg TID for at least 3 days prior to initiation of medication. It is continued during the first two months of therapy in an effort to avoid nausea. Common adverse events include yawning, dizziness, nausea, somnolence and dyskinesias [38]. One study reported that there were no significant differences between apomorphine and placebo in the incidence of hypotension [38].
Outside the US, apomorphine is available as a subcutaneous infusion. When administered in this way, it reduces ‘‘off ’’ time, allows a reduction in levodopa dose, and reduces dyskinesias [39].
COMT Inhibitors COMT (catechol-O-methyltransferase) is an enzyme that metabolizes dopamine and other catecholamines including levodopa. It is found throughout the body, in the liver, kidneys, gastrointestinal tract, spleen, and brain. Centrally, COMT is primarily localized in nonneuronal cells, such as glia, and has not been detected in nigrostriatal dopaminergic neurons. When levodopa is administered with a peripheral DDCI, its primary peripheral route of metabolism is via COMT to form 3-O-methyldopa (3-OMD). Adding a peripheral COMT inhibitor to levodopa/DDCI reduces peripheral levodopa metabolism, thereby extending the levodopa half-life and making more levodopa available for transport across the BBB over a longer time. There are two available COMT inhibitors, tolcapone and entacapone. Tolcapone has greater efficacy but is associated with rare fatal hepatotoxicity and therefore requires liver function test (LFT) monitoring. Entacapone has not been associated with hepatotoxicity and LFT monitoring is not required. Entacapone (COMTAN)
Entacopone is a selective and reversible peripheral inhibitor of COMT. It is readily absorbed across the intestinal mucosa and is not significantly affected by first-pass metabolism in the liver. Its bioavailability ranges from 30–40%, and it is highly protein bound, with an elimination half-life is 0.4–0.7 h [40]. Entacopone comes in 200 mg tablets and is routinely administered with each dose of levodopa/DDCI to a usual maximum of 8 tablets (1,600 mg) per day in the US and 10 tablets (2,000 mg) per day elsewhere. Adding entacapone 200 mg to carbidopa/levodopa IR
Medical management of parkinson’s disease
increases levodopa’s elimination half-life from 1.3–2.4 h. In single dose studies it does not increase peak levodopa plasma concentration or time to maximal plasma concentration [41,42], but in multiple dose studies it can increase peak levodopa concentrations through the day [41]. Entacapone increase the levodopa half-life and area under the curve, thereby extending levodopa’s duration of action (42). Entacapone also extends the levodopa area under the curve plasma concentration of carbidopa/levodopa CR. Entacapone is approved as an adjunct to levodopa/DDCI IR for patients experiencing wearing off fluctuations. Clinical trials demonstrate that adding entacapone to levodopa/DDCI increases ‘‘on’’ time, decreases ‘‘off’’ time, and enhances motor function [43,44]. The efficacy of entacapone and cabergoline were found to be comparable, but the clinical benefit was more quickly apparent with entacapone, presumably because cabergoline must be slowly escalated whereas entacapone can be initiated at the full maintenance dose [45]. Entacapone also showed a more favorable adverse event profile than cabergoline [45]. Adverse events are mostly those due to enhanced dopaminergic activity, such as increased dyskinesia, nausea, and drowsiness. Orange, yellow or brown discoloration of urine, saliva or sweat can also be seen. Diarrhea can occur, and typically appears within 4–12 weeks after entacapone is started, but may appear as early as the first week and as late as many months after initiation of treatment. In the MPTP marmoset model of PD, the addition of entacapone to each dose of levodopa/DDCI administered QID increased ‘‘on’’ time and notably, also reduced the development of dyskinesia, presumably by providing more continuous dopaminergic stimulation [46]. Carbidopa/Levodopa/Entacapone (Stalevo)
Stalevo is a combination of levodopa, carbidopa, and entacapone. It was approved based on demonstration of bioequivalence with carbidopa/levodopa plus entacapone administered as separate tablets. It is indicated for PD patients
90
receiving carbidopa/levodopa IR who are experiencing wearing off. Carbidopa/levodopa/entacapone provides increased convenience as patients can take fewer pills [47]. Each Stalevo tablet contains 200 mg of entacapone, and carbidopa/levodopa in a 1:4 ratio. The tablet is named for the mg amount of levodopa it contains. Stalevo is available in formulations of 50 (12.5 mg carbidopa/50 mg levodopa/200 mg entacapone), 100 (25 mg carbidopa/100 mg levodopa/200 mg entacapone), 150 (37.5 mg carbidopa/150 mg levodopa/200 mg entacapone), and 200 mg (50 mg carbidopa, 200 mg levodopa, 200 mg entacapone). Switching from carbidopa/levodopa IR to the corresponding carbidopa/levodopa/entacapone dose is analogous to adding entacapone to carbidopa/ levodopa IR. In switching patients from carbidopa/levodopa CR to carbidopa/levodopa/entacapone, it should be noted that the bioavailability of levodopa from carbidopa/levodopa CR is approximately 70–75% that of carbidopa/levodopa IR. A majority of patients suboptimally controlled on carbidopa/levodopa CR can be switched to carbidopa/levodopa/entacapone with improvements in motor function, quality of life, and sleepiness [48]. Investigations are currently underway to determine if carbidopa/levodopa/ entacapone causes fewer motor complications, especially dyskinesia, compared to carbidopa/ levodopa IR when administered as the first levodopa formulation in early PD. Tolcapone (Tasmar)
Like entacapone, tolcapone (3, 4-dihydrox-40 methyl-5-nitrobenzo-phenone) is rapidly absorbed after oral administration. It reaches Tmax in 1.5–2 h. It is highly protein bound with a bioavailability of 60% and an elimination half-life of 2–3 h. It is usually administered on a TID schedule. Tolcapone is somewhat lipophilic and may cross the BBB, but evidence of possible central COMT inhibition in humans is less convincing than that seen in animal models [49]. Tolcapone is efficacious as an adjunct to levodopa/DDCI to
1513
1514
90
Medical management of parkinson’s disease
increase ‘‘on’’ time, reduce ‘‘off ’’ time and allow a reduction of the daily levodopa dose [50–52]. The initial dose is 100 mg TID and it can be increased to 200 mg TID. Side effects are mostly due to enhanced dopaminergic activity, such as dyskinesia, which can be alleviated by decreasing levodopa dose. For patients who are experiencing dyskinesia, a pre-emptive reduction of the levodopa dose of 25–50% prior to the addition of tolcapone may be prudent to avoid severe dyskinesia. Further levodopa titration is undertaken as clinically necessary. Diarrhea can occur, can be severe, and is dose-related. Like entacapone, yellowish discoloration of urine can occur. Rare cases of fatal hepatotoxicity have been reported and strict LFT monitoring guidelines have been recommended by the FDA. Levels of serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) should be monitored every 2–4 weeks during the first six months and then as clinically indicated. If the transaminase levels exceed twice the normal limit, tolcapone should be discontinued. Tolcapone is a very efficacious adjunct to levodopa but because of the potential for hepatotoxicity it is usually reserved until other medical options have been exhausted. Nonetheless, it may be quite appropriate for patients who are being considered for PD surgery.
Monoamine Oxidase (MAO)-B Inhibitors MAO is an enzyme that is involved in the breakdown of catecholamines, including dopamine, norepinephrine and serotonin. There are two types of MAO, A and B. MAO-B is predominantly found in the human brain and platelets and has an affinity for dopamine. MAO-A is found predominantly in the intestinal tract and has an affinity for serotonin and norepinephrine. MAO inhibitors were originally used for depression, but were later shown to have an antiparkinsonian
effect. By blocking MAO-B, MAO-B inhibitors reduce dopamine clearance in the brain and increase dopamine availability at post-synaptic dopamine receptors. MAO-A normally metabolizes ingested tyramine and other catecholamines. MAO-A inhibition allows absorption of tyramine which can lead to hypertensive crisis (‘‘cheese effect’’). MAO-B inhibitors are relatively selective for MAO-B, but at high doses begin to lose their specificity and can also inhibit MAO-A. Selegiline and rasagiline are both irreversible MAO-B inhibitors as they covalently bind to MAO-B. Both are propargylamine molecules and have been demonstrated to block apoptosis (programmed cell death) in vitro, independent of MAO-B inhibition [53–55]. They are potent anti-apoptotic agents that act by maintaining glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as a dimer, thereby preventing its nuclear translocation where it blocks upregulation of anti-apoptotic proteins, and by holding the mitochondrial transition pore closed to prevent activation of the apoptosis pathway [56]. Selegiline (Eldepryl)
Selegiline (N-propynyl-methamphetamine) is a relatively selective irreversible MAO-B inhibitor that is readily absorbed from the intestine. Oral selegiline reaches peak plasma concentration in 30–120 min and has a half-life of 2 h. Because it binds irreversibly to MAO-B in the brain, its clinical effect lasts several months. Selegiline metabolites include l-methamphetamine, l-amphetamine and desmethylselegiline. Oral selegiline is administered at a dose of 5 mg BID, at breakfast and at lunch. Because of its amphetamine metabolites, it is usually not given later in the day to avoid insomnia. At the recommended dosage, selegiline provides mild antiparkinsonian effect without risk of tyramine interactions. There has been longstanding interest in possible neuroprotective effects from selegiline but such an effect has not been proven. Selegiline
Medical management of parkinson’s disease
derives its protective benefit from its metabolite desmethylselegiline which provides greater protective effects in cell culture and animal models than selegiline [54]. In the DATATOP (deprenyl and tocopherol anti-oxidative therapy of parkinsonism) study [57], selegiline monotherapy delayed the need for levodopa compared to placebo. However, a small symptomatic effect was identified, thereby confounding the study’s ability to clearly detect neuroprotection. Oral selegiline is currently approved as an adjunct to levodopa. The most common side effects of selegiline are nausea, constipation, diaphoresis, hallucinations and dyskinesias. The serotonin syndrome, characterized by diaphoresis, hypertension and confusion, can rarely occur when selegiline is combined with selective serotonin reuptake inhibitors (SSRI’s). Oral Disintegrating Selegiline – Zydis Selegiline (Zelapar)
Zydis selegiline dissolves on contact with saliva and undergoes pregastric absorption. First-pass metabolism is minimized and in theory it might be possible to get greater brain MAO inhibition before MAO-A in the gut is inhibited. Treatment is initiated at a dose of 1.25 mg once a day for at least 6 weeks then escalated to 2.5 mg once a day if the desired benefit has not been attained. It is placed on top of the tongue where it disintegrates in several seconds. It is taken in the morning before breakfast and without liquid. Orally disintegrating selegiline is approved as an adjunct to levodopa in patients experiencing wearing off motor fluctuations and has been shown to significantly reduce daily off time and increase the average number of dyskinesia-free on hours when added to levodopa [58]. Adverse events are similar to oral selegiline. Rasagiline (Azilect)
Rasagiline (N-propargyl-(1-R)-aminoindan) is also an irreversible MAO-B inhibitor. Like other
90
propargylamines, rasagiline has been demonstrated to provide anti-apoptotic effects independent of MAO-B inhibition. In vitro, it protects against a variety of toxins, increases survival of cultured fetal mesencephalic dopaminergic neurons and protects dopamine cells against apoptotic cell death [59,60]. It has also been shown to provide neuroprotective effects in a variety of in vivo models [61]. Rasagiline provides antiparkinsonian benefit and is approved as monotherapy in early disease and as an adjunct to levodopa for wearing off fluctuations in later disease. The TEMPO (rasagiline mesylate (TVP1012) as early monotherapy in PD outpatients) study was performed in early untreated PD patients using a delayed start design [62]. Untreated PD patients were randomized to receive rasagiline or placebo during the first 6 months. At the end of this phase, the rasagiline-treated patients demonstrated significant improvement in UPDRS motor scores in comparison with the placebo treated group. All patients subsequently received rasagiline for another 6 months. At the end of 12 months, patients who received rasagiline for 12 months experienced significantly greater benefit than patients who received placebo for 6 months followed by rasagiline for 6 months. This result is potentially consistent with a neuroprotective effect and can not be explained by a simple symptomatic effect. A larger study to evaluate initial treatment with rasagiline versus placebo using a delayed-start design is now under way. The PRESTO study (Parkinson’s Rasagiline: Efficacy and Safety in the Treatment of Off) [63] and the LARGO study (Lasting effect in Adjunct therapy with Rasagiline) [64] evaluated the efficacy of rasagline as an adjunct to levodopa in patients with motor fluctuations. Rasagiline administered once daily significantly reduced off time and the efficacy was comparable to that of entacapone 200 mg administered with each levodopa dose. An ancillary analysis of data
1515
1516
90
Medical management of parkinson’s disease
from the LARGO trial found that rasagiline also improved freezing compared to placebo. Rasagiline can be administered with or without food. The typical dosage is 1 mg per day for monotherapy, and 0.5 mg per day if used as an adjunct. It can be increased to 1 mg once daily if the response is not adequate. Absorption is rapid, and peak plasma concentration is reached after 30 min. As with selegiline, clinical effect lasts several months and is dependent on MAO-B turnover in the brain. Rasagiline does not have amphetamine metabolites. Its major metabolite is 1-R-aminoidan which also has antiparkinsonian efficacy. Rasagiline is metabolized via CYP1A2 and serum concentrations can be increased when administered concomitantly with medications that inhibit CYP1A2 such as cimetidine, ciprofloxacin and omeprazole. Rasagiline is generally well tolerated, particularly as monotherapy. Potential side effects include postural hypotension, anorexia, and increased dyskinesia.
Anticholinergics Anticholinergic medications are currently used to reduce tremor, particularly in patients who experience inadequate tremor control with dopaminergic medications. The two most commonly prescribed anticholinergics are trihexyphenidyl (Artane) and benztropine (Cogentin). Trihexyphenidyl is initiated at a dose of 1 mg per day and slowly escalated to a maintenance dose of 4–6 mg per day as tolerated. Benztropine is prescribed at doses of 0.5–2 mg TID. Common central side effects include memory impairment, confusion, and hallucinations. These side effects are more common in patients with pre-existing cognitive impairment and the elderly. Peripheral side effects include dry mouth, blurred vision, constipation, nausea, urinary retention, impaired sweating and tachycardia. Most side effects are dose dependent and can be minimized by reducing the dose.
Amantadine (Symmetrel) Amantadine is an anti-viral agent that was accidentally discovered to have antiparkinsonian effect. It provides mild benefit in alleviating rigidity and bradykinesia in early disease, and delays the need for levodopa. In advanced disease, it is useful as an adjunct to levodopa to reduce ‘‘off’’ time and to reduce dyskinesia. It is a glutamate antagonist and may also promote endogenous dopamine release, directly stimulate dopamine receptors, and inhibit dopamine reuptake. It also has anticholinergic properties. The bioavailability of amantadine is 100% in oral form. It is excreted virtually unmetabolized via the kidneys. Peak plasma concentration occurs in 1–4 h and it has a long plasma half-life of 10–28.5 h. Amantadine can be initiated at a starting dose of 100 mg QD, increasing by 1 tab every week to a usual target dose of 100 mg TID or QID. Amantadine has been shown to provide an antidyskinetic effect without worsening parkinsonism [65,66]. There is a 60% reduction in peak dose ‘‘on’’ dyskinesias [65]. The antidyskinetic effect can be sustained for a year and has been attributed to antagonism of central glutamatergic N-methyl-D-aspartate (NMDA) receptors [67]. The most common peripheral side effects are livedo reticularis, a mottled bluish-red reticular discoloration which blanches to pressure, and pedal edema. Livedo reticularis typically occurs after weeks of treatment and may take weeks to resolve. Other notable side effects include psychiatric and cognitive disturbances, such as confusion and hallucinations, although these are more common in patients who already have underlying cognitive dysfunction. Amantadine’s anticholinergic activity can cause orthostatic hypotension, dry mouth, urinary retention and blurry vision. It should be used cautiously in those with poor renal clearance. Abrupt withdrawal should be avoided as this can cause neuroleptic malignant syndrome and dramatic worsening of parkinsonism in more advanced cases. Amantadine is an important medication
Medical management of parkinson’s disease
for individuals who may be considered for PD surgery, as it is currently the only available medication that can reduce dyskinesia without worsening PD.
Treatment Strategies
90
are less likely to develop motor complications and the elderly and cognitively impaired are more likely to experience hallucinations and confusion. Therefore, in younger patients, dopamine agonists are often initiated and when no longer sufficient to control symptoms, levodopa is added. For older patients, levodopa may be the prudent choice.
Early Disease In general, symptomatic medications for PD are introduced when the patient has some functional disability or finds symptoms annoying. Nonetheless, the optimal time to introduce medications to achieve the best overall outcome is not clear, and there is interest in the notion that dopaminergic agents other than levodopa (MAO-B inhibitors and dopamine agonists) might provide the greatest benefit if started as early as possible, possibly to maintain compensatory mechanisms [68]. For patients with early disease who only require mild benefit, MAO-B inhibitors can be considered. They provide mild benefit and are usually well tolerated. Amantadine can also be considered but is generally associated with more adverse events including edema and hallucinations. Once greater symptomatic benefit is required, the choice comes down to starting a dopamine agonist and adding levodopa later on, or going directly to levodopa. Levodopa is the most efficacious medication but is associated with the development of motor fluctuations and dyskinesias (motor complications). Dopamine agonists are not associated with the development of fluctuations or dyskinesias, but are moderately efficacious and have more short term side effects including somnolence, edema, hallucinations, and impulse control disorders. The choice between dopamine agonists and levodopa as early treatment is usually made on the basis of age. Younger individuals (<65–70) are more prone to develop motor complications and have a longer treatment time horizon. Older individuals (>70)
Advanced Disease with Motor Complications Patients with motor fluctuations and without dyskinesia should undergo medication adjustments to minimize or resolve off time. This can be accomplished by moving levodopa doses closer together, adding an MAO-B inhibitor, COMT inhibitor, or dopamine agonist. If necessary, the levodopa dose for each administration can be increased. The situation is more complicated in patients with both fluctuations and dyskinesias as increases in dopamine medication are likely to lessen off time but increase dyskinesia, and decreases in medication are likely to decrease dyskinesia but worsen off time. A core strategy is to administer smaller levodopa doses more frequently and to add adjunctive medications including MAO-B inhibitors, COMT inhibitors and dopamine agonists, in an effort to smooth dopamine stimulation in the brain. For patients who are relatively intact cognitively, amantadine is an important medication because it can reduce both dyskinesia and off time. Tolcapone may also be helpful, although the levodopa dose must commonly be reduced. An additional option for patients who have fluctuations that cannot be controlled with oral medication manipulations is apomorphine subcutaneous injections on an as needed basis.
Non-motor Symptoms The occurrence of non-motor symptoms (NMS) is now recognized as an integral part of PD.
1517
1518
90
Medical management of parkinson’s disease
A significant number of patients experience autonomic dysfunction, sleep disturbances, mood disorders, and dementia. There is also increasing evidence that certain non-motor features can precede the onset of motor symptoms, such as constipation, hyposmia, and REM behavior disorder (RBD). Non-motor features in PD can have a devastating effect on quality of life and therefore proper identification and treatment are necessary.
Autonomic Dysfunction Autonomic symptoms are important features of PD and the incidence increases with age, disease severity, and dopaminergic medication usage [69]. The most significant differences in autonomic function between PD patients and controls are gastrointestinal and urinary disturbances [69]. Constipation
Infrequent bowel movements are associated with an elevated risk of future PD [70,71], and constipation may be one of the earliest signs of the disease [71,72]. Once motor signs appear, constipation is very common and is likely due to the disease itself, PD medications, and possibly reduced water intake secondary to diminished thirst sensation [71]. Slow colonic transit time, decreased phasic rectal contraction, weak abdominal strain, and paradoxical sphincter contraction on defecation were found in patients with frequent constipation [73]. Treatment includes increasing fluids, adding more fiber to the diet, stool softeners, polyethylene glycol (Miralax), and if necessary, lactulose. Anticholinergic medications should be discontinued if possible. Urinary Disturbances
There is an increased incidence of urinary disturbances in the more advanced stages of Parkinson disease [74–76]. One report found substantially more urinary problems in PD patients than controls, with urgency and incontinence
being particularly more common [69]. Dopamine dysfunction may play a role [77] and urinary function improves with dopaminergic therapy in some cases. When a patient experiences a change in voiding pattern, urinary tract infection must be considered and treated if present. A urologic evaluation is warranted for persistent urinary dysfunction in order to determine whether detrusor hyper or hypoactivity is occurring. Nocturnal frequency can be minimized by reducing fluid intake after the evening meal. If this measure is not effective, peripheral anticholinergics can be used. These include oxybutinin chloride (Ditropan, Ditropan XL), porpantheline bromide, and tolterodine tartarate (Detrol, Detrol LA). Anticholinergics should be used with caution as they can lead to urinary retention in those with detrusor hypoactivity or outlet obstruction and may increase the risk for gastric obstruction. Orthostatic Hypotension (OH)
Cardiac sympathetic denervation [78–80] and neurogenic orthostatic hypotension [80,81] are very common in patients with PD and can occur early in the disease [82]. Those with OH can suffer from debilitating dizziness and life threatening falls. OH can occur independent of levodopa treatment [79] but can also be aggravated by levodopa [83] or other dopaminergic agents [84–87]. OH is defined as a drop of 20 mmHg in systolic blood pressure or 10 mmHg in diastolic blood pressure when a patient rises from a supine to a standing position. Once a patient becomes symptomatic, treatment is generally considered necessary. Treatment includes reducing antihypertensives, hydration, liberalization of dietary salt, and use of compression stockings. Patients should also be instructed to arise slowly. Reducing dopaminergic medications may be considered when feasible. When conservative measures fail, the mineralcorticoid fludrocortisone (0.1–0.4 mg a day) or the a-1-agonist midodrine
Medical management of parkinson’s disease
(2.5–20 mg/day) can be added, but must be used with caution in patients with cardiovascular disease. Sexual Dysfunction
Sexual dysfunction is common and may be the initial autonomic disturbance. Its severity may not correlate with overall disease severity, unlike other autonomic problems [88]. Medical screening for impotence should be performed. Depression and anxiety should be addressed, and medications such as propranolol and beta adrenergic blockers should be discontinued if possible. More extensive work-up includes appropriate urological consultation and endocrinological evaluation may be warranted. Small studies have shown that Sildenafil (Viagra), a potent inhibitor of phosphodiesterase type 5, may be useful in treating erectile dysfunction [89,90]. Hypersexuality can be caused or aggravated by dopaminergic agents. In this case, reduction of the offending agent may be beneficial. Hypersexuality is an impulse control disorder and will be discussed further under this topic.
90
if feasible. Management also includes proper sleep hygiene by asking the patient to minimize daytime naps in order to sleep better at night. If medication adjustment does not resolve the problem, a polysomnographic study should be performed to exclude sleep disorders. Disrupted sleep is common in PD patients and can predate future PD [96]. Polysomnography may also identify sleep apnea, periodic limb movements of sleep, or REM behavior disorder. If the sleep study is negative for a sleep disorder, then alerting agents, such as modafanil (Provigil) can be tried although the efficacy is variable [97–99]. It is well tolerated and does not appear to worsen parkinsonian symptoms [97–99]. The typical dosage is 200–400 mg/day. Sudden sleep attacks have been reported [22], particularly in association with dopamine agonist administration [100]. Patients should be educated about this potential side effect and advised to pull over and stop driving if they feel sleepy. Patients who fall asleep at inappropriate times should be advised not to drive. Insomnia
Sleep Disturbances Excessive Daytime Sleepiness (EDS)
Excessive daytime sleepiness (EDS) is a common problem in PD. It is associated with the disease itself, PD medications, and sleep disorders. Increased daytime sleepiness is more frequent in patients with PD than in elderly controls [91]. Duration of disease [92] and advanced disease [92,93] correlate with the degree of daytime sleepiness. Dopaminergic agents have been implicated, particularly dopamine agonists [92,94,95]. For patients experiencing EDS, consideration should first be given to discontinuing non-dopaminergic sedating medications. If a dopamine agonist was recently introduced or if there is no other apparent cause, consideration should be given to reducing or discontinuing the dopamine agonist
Difficulty initiating or maintaining sleep has been reported in PD patients [101] and may be a component of a primary sleep disorder or secondary to the disease [101], dementia or depression. A careful medication history is needed and alerting agents should be eliminated. Proper sleep hygiene is recommended. Alcohol, caffeine and tobacco should be avoided during the latter part of the day and evening. Nocturia can be avoided by reducing fluids at night. Trouble sleeping can also be caused by difficulty turning or getting comfortable [102] and may be due to inadequate control of PD symptoms. A bedtime dose of a longer duration medication, such as carbidopa/ levodopa/entacapone, carbidopa/levodopa CR, or a dopamine agonist, may be useful in providing a longer dopaminergic effect to alleviate rigidity and bradykinesia later into the night. When necessary, sleep medications can be
1519
1520
90
Medical management of parkinson’s disease
considered. Shorter acting sedative hypnotics are preferred, but physical dependence may occur. An open label trial showed that quetiapine, an atypical antipsychotic which is frequently used to treat hallucinations and psychosis in PD, may be a safe and effective alternative treatment of insomnia [103]. Depression should be treated if present. Rapid Eye Movement (REM) Behavior Disorder (RBD)
REM behavior disorder is characterized by the loss of normal muscular atonia during REM sleep, accompanied by complex motor behaviors [104]. RBD should be suspected if aggressive or unusual behavior during sleep is reported. The bed partner characteristically complains that the patient is hitting, kicking, or talking in his sleep. RBD is common in the ‘‘synucleinopathies,’’ including PD [105], dementia with Lewy bodies (DLB) and MSA. RBD can precede the onset of motor symptoms [106] and has been detected in more than one-third of patients with PD without dementia [107]. The presence of RBD may be a risk factor for the development of cognitive impairment and dementia [108] and has been associated with EEG slowing [109]. Clonazepam (Klonopin) is usually helpful to reduce symptoms of RBD and can be used in dosages of 0.5 mg to 2 mg at bedtime. Sleep Apnea
Obstructive sleep apnea (OSA) is defined by intermittent absent or reduced airflow during sleep despite respiratory effort. It is unclear if OSA is more common in PD than the general population but this problem needs to be considered in individuals with EDS. Polysomnography is required for diagnosis. Those with OSA are usually treated with continuous positive airway pressure (CPAP) mask although it is often not well tolerated in older individuals.
legs usually associated with unpleasant sensations, and is typically worse at night and when resting quietly. Sensory disturbances can include aches, pain, crawling sensations, burning, and tingling, and voluntary movement of the legs typically improve the symptoms. The classic clinical picture is the patient who tries to lay down at night to go to sleep but feels uncomfortable sensations in the legs and must get out of bed and walk around to get relief. Although dopamine deficiency is implicated in RLS, it is a different disease from PD. Dopamine agonists, such as ropinorole and pramipexole, are the medications of choice. Lower doses are used compared to those used to control motor symptoms of PD. Levodopa may also be efficacious but is often associated with rebound or augmentation. Secondary causes of RLS should be sought, such as iron deficiency anemia and treated if present.
Neuropsychiatric/Cognitive Depression
Depression affects up to 50% of patients with PD and has a strong negative effect on quality of life [110]. It is common and oftentimes undertreated early in the disease [111], and is associated with increased disability [112,111]. Clinically significant depression in PD is underdiagnosed [113,114]. Antidepressants that are effective in the general population appear to be effective in PD as well. Serotonin reuptake inhibitors are most commonly administered since they do not possess the anticholinergic side effects of the tricyclic antidepressants. Caution is advised when SSRI’s are administered concomitantly with MAO-B inhibitors, but at recommended dosages interactions appear to be rare. Anxiety
Restless Legs Syndrome (RLS)
RLS can contribute to nighttime sleep difficulties. It is characterized by an urge to move the
A majority of PD patients experience anxiety and panic attacks can occur. Symptoms of anxiety sometimes occur specifically during ‘‘off ’’ time
Medical management of parkinson’s disease
and these can be controlled by adjusting PD medications to reduce ‘‘off ’’ time. Anxiolytic drugs, particularly the short acting benzodiazepines, can be used, although they should be used with caution in the cognitively impaired population. There is the potential for addiction and spiraling upward doses should be avoided. SSRIs are commonly prescribed and may be helpful for both anxiety and depression. Impulse Control Disorders (ICD)
Impulse control disorders, particularly pathological gambling, may have increased frequency in Parkinson’s disease [115–118]. Levodopa treatment has been implicated [115], but more recent studies have associated dopamine agonists in the occurrence of these disorders [116–120]. Other ICDs include compulsive buying and hypersexuality. Lifetime prevalence of these behaviors is 6.1% and this increases to 13.7% in patients on dopamine agonists [120]. Some patients develop a pattern of compulsive dopaminergic drug use, referred to as the dopamine dysregulation syndrome (DDS). It has been speculated that there is sensitization of brain dopamine systems mediating reward by dopaminergic drugs. Those experiencing DDS demand increases in medications early in the course of treatment. Compulsive gambling, hypersexuality, and mood changes also frequently accompany DDS. A history of ICD symptoms prior to PD onset [118], younger age at motor symptom onset, male gender and longer duration of treatment with dopamine agonists [119] contribute to the risk of developing one or more of these behavioral disorders. Further, higher novelty seeking traits and a personal or family history of alcohol use may present a greater risk for pathological gambling with dopamine agonists [121]. Hypersexuality has also been observed after pallidotomy [122] and deep brain stimulation (DBS) of the pallidum [123]. Conversely, pathological gambling has improved after chronic subthalamic nucleus stimulation [124– 126]. An ICD screening prior to dopaminergic
90
drug use may be helpful to identify susceptible individuals. For individuals who experience an ICD, consideration should be given to lowering the dosage of dopamine agonist, and possibly levodopa, when feasible. Dementia
Cognitive dysfunction and dementia may be the biggest problem that PD patients face over the long term. Prevalence rates of mild cognitive impairment without dementia and dementia after 15 years are 84 and 48%, respectively [127]. Increasing age, later age of disease onset, longer duration of PD symptoms, presence of hallucinations, and impairment of memory and language function are all predictive factors for the development of dementia [128]. Dementia appears to be due to loss of cholinergic neurons in the nucleus basalis of Meynert and the presence of cortical Lewy bodies [129]. Acetycholinesterase inhibitors are used to treat PDD. Rivastigmine (Exelon) is a carbamate-type dual inhibitor of brain acetyl- and butyrylcholinesterases and is approved for treatment of mild to moderate PDD. In large placebocontrolled studies, rivastigmine was associated with moderate improvements in dementia for 24 weeks [130] that was sustained up to 48 weeks [131]. Rivastigmine treated groups had higher rates of nausea, vomiting, and tremor than placebo [130]. It is titrated slowly from 3 mg a day to 12 mg a day. Visual hallucinations appear to predict more rapid decline and possibly greater therapeutic benefit from rivastigmine treatment in PD [132]. Recently, a rivastigmine transdermal patch has been approved for treatment of PDD. It offers the important advantage of less nausea and vomiting than the oral formulation [133]. In a small double blind study, donepezil at dosages of 5–10 mg per day provided modest benefit for PDD [134]. Memantine (Namenda), an NMDA antagonist, has been approved for the treatment of Alzheimer’s dementia but has not been evaluated for PDD.
1521
1522
90
Medical management of parkinson’s disease
Psychosis
Hallucinations are relatively common in PD and the incidence increases with age. Most patients who experience hallucinations have underlying cognitive dysfunction and may be prone to developing frank dementia. Hallucinations generally occur in more advanced disease and are associated with medication use, especially amantadine, dopaminergic agents, anticholinergics, sedatives and antidepressants. Patients are not usually threatened, but frightening hallucinations occasionally occur. Visual hallucinations are more common than auditory ones. Treatment includes possible evaluation and treatment of acute medical problems and adjustment or reduction of medications that might be causing them. Amantadine and anticholinergic drugs, such as trihexyphenydyl, should be reduced or stopped first. Next would be dopamine agonists. Patients may do best on levodopa alone. If decreasing antiparkinsonian medications compromises motor function, atypical antipsychotics such as quietiapine (Seroquel) or clozapine (Clozaril) can be administered. Both have less parkinsonian side effects than other atypical antipsychotics. Although there are conflicting reports on the benefits of quietiapine on psychosis [135–137], it is widely used, since hematological monitoring for possible agranulocytosis is required while using clozapine.
Conclusion Optimal medical management of PD requires recognition and treatment of both motor and nonmotor symptoms. Levodopa is the most effective agent for motor symptoms but is associated with the development of motor fluctuations and dyskinesias. Because of this, a levodopa sparing strategy is often utilized in younger patients by administering MAO-B inhibitors and dopamine agonists before levodopa is added. Investigations are underway to determine if smoothing levodopa concentrations by using Stalevo rather than
carbidopa/levodopa IR will lead to a lower incidence of motor complications. For patients with motor fluctuations and dyskinesias, smaller doses of levodopa are given more frequently and adjunctive medications are added. For patients with more difficult motor complications, amantadine can reduce both dyskinesias and off time. Additional considerations include tolcapone with adjustment of levodopa dosing and subcutaneous apomorphine for refractory off periods. Nonmotor features are treated symptomatically but non-dopaminergic features, especially dementia, still dominate long term disability. Ultimately medications that can slow disease progression are needed.
References 1. Reiderer P, Wuketich S. Time course of nigrostriatal degeneration in Parkinson’s disease. J Neural Transm Park Dis Dement Sect 1976;38:277-301. 2. Ahlskog JE, Muenter MD. Frequency of levodoparelated dyskinesias and motor fluctuations as estimated from the cumulative literature. Mov Disord 2001;16:448-58. 3. Colosimo C, De Michele M. Motor fluctuations in Parkinson’s disease: pathophysiology and treatment. Eur J Neurol 1999;6:1-21. 4. Grandas F, Galiano ML, Tabernero C. Risk factors for levodopa-induced dyskinesias in Parkinson’s disease. J Neurol 1999;246:1127-33. 5. Van Gerpen JA, Kumar N, Bower JH, Weigand S, Ahlskog JE. Levodopa-associated dyskinesia risk among Parkinson disease patients in Olmsted County, Minnesota, 1976–1990. Arch Neurol 2006;63:205-9. 6. Rascol O, Brooks DJ, Korczyn AD, De Deyn PP, Clarke CE, Lang AE. A five-year study of the incidence of dyskinesia in patients with early Parkinson’s disease who were treated with ropinirole or levodopa: 056 Study Group. N Engl J Med 2000;342:1484-91. 7. Parkinson Study Group. A randomized controlled trial comparing pramipexole with levodopa in early Parkinson’s disease: design and methods of the CALM-PD Study. Clin Neuropharmacol 2000;23:34-44. 8. Hauser RA, Rascol O, Korczyn AD, Jon Stoessl, Watts RL, Poewe W, DE Deyn PP, Lang AE. Follow-up of Parkinson’s disease patients randomized to initial therapy with ropinirole or levodopa. Mov. Disord 2007;22 (16):2409-17.
Medical management of parkinson’s disease
9. The Parkinson Study Group. Pramipexole vs. levodopa as initial treatment for Parkinson disease. Arch Neurol 2004;61:1044-53. 10. Olanow C, Jenner P, Brooks D. Dopamine agonists and neuroprotection in Parkinson’s disease. Ann Neurol 1998;44 Suppl 1:167-74. 11. Parkinson Study Group. Dopamine transporter brain imaging to assess the effects of pramipexole versus levodopa on Parkinson disease progression. JAMA 2002;287:1653-61. 12. Whone AL, Watts RL, Stoessl AJ, Davis M, Reske S, Nahmias C, Lang AE, Rascol O, Ribiero MJ, Remy MJ, Remy P, Poewe WH, Hauser RA, Brooks DJ; REAL-PET Study Group. Slower progression of Parkinson’s disease with ropinirole versus levodopa: The REAL-PET study. Ann Neurol. 2003;54(1):93-101. 13. Todman DH, Oliver WA, Edwards RL. Pleuropulmonary fibrosis due to bromocriptine treatment for Parkinson’s disease. Clin Exp Neurol 1990;27:79-82. 14. Ling LH, Ahlskog JE, Munger TM, et al. Constrictive pericarditis and pleuropulmonary disease linked to ergot dopamine agonist (cabergoline) for Parkinson’s disease. Mayo Clin Proc 1999;74:371-5. 15. Danoff SK, Grasso ME, Terry PB, Flynn JM. Pleuropulmonary disease due to pergolide use for restless legs syndrome. Chest 2001;120:313-6. 16. Serratrice J, Disdier P, Habib G, Viallet F, Weiller PJ. Fibrotic valvular heart disease subsequent to bromocriptine treatment. Cardiol Rev 2002;10:334-6. 17. Baseman DG, O’Suilleabhain PE, Reimold SC, et al. Pergolide use in Parkinson’s disease is associated with cardiac valve regurgitation. Neurology 2004;63:301-4. 18. Horvath J, Fross RD, Kleiner-Fisman G, et al. Severe multivalvular heart disease: a new complication of the ergot derivative dopamine agonists. Mov Disord 2004;19:656-62. 19. Pinero A, Marcos-Albertca P, Fortes J. Cabergolinerelated severe restrictive mitral regurgitation. N Engl J Med 2005;353:18-19. 20. Biglan KM, Holloway RG Jr, McDermott MP, Richard IH; Parkinson Study Group CALM-PD Investigators. Risk factors for somnolence, edema, and hallucinations in early Parkinson disease. Neurology 2007;69:187-95. 21. Sanchez-Ramos JR, Ortoll R, Paulson GW. Visual hallucinations associated with Parkinson disease. Arch Neurol 1996;53:1265-68. 22. Frucht S, Rogers JD, Greene PE, Gordon MF, Fahn S. Falling asleep at the wheel: motor vehicle mishaps in persons taking pramipexole and ropinirole. Neurology 1999;52:1908-10. 23. Paus S, Brecht HM, Koster J, Seeger G, Klockgether T, Wullnder U. Sleep attacks, daytime sleepiness, and dopamine agonists in Parkinson’s disease. Mov Disord 2003;18:659-67.
90
24. Avorn J, Schneeweiss S, Sudarsky LR, Benner J, Kiyota Y, Levin R, Glynn RJ. Sudden uncontrollable somnolence and medication use in Parkinson disease. Arch Neurol 2005;62:1242-8. 25. Parkinson Study Group. Pramipexole versus levodopa as initial treatment for Parkinson disease: a randomized controlled trial. JAMA 2000;284:1931-8. 26. Lieberman A, Ranhosky A, Korts D. Clinical evaluation of pramipexole in advanced Parkinson’s disease: results of a double-blind, placebo-controlled, parallel-group study. Neurology 1997;49:162-8. 27. Pinter MM, Pogarell O, Oertel WH. Efficacy, safety, and tolerance of the non-ergoline dopamine agonist pramipexole in the treatment of advanced Parkinson’s disease: a double blind, placebo controlled, randomised, and multicentre study. J Neurol Neurosurg Psychiatry 1999;66:436-41. 28. Pogarell O, Gasser T, van Hilten JJ, Spieker S, Pollentier S, Meier D, Oertel WH. Pramipexole in patients with Parkinson’s disease and marked drug resistant tremor: a randomised, double blind, placebo controlled multicentre study. J Neurol Neurosurg Psychiatry 2002;72(6):713-20. 29. Lieberman A, Olanow CW, Sethi K, et al. A multicenter trial of ropinirole as adjunct treatment for Parkinson’s disease. Ropinirole Study Group. Neurology 1998;51:1057-62. 30. Pahwa R, Stacy MA, Factor SA, Lyons KE, Stocchi F, Hersh BP, Elmer LW, Truong DD, Earl NL; EASE-PD Adjunct Study Investigators. Ropinirole 24-hour prolonged release: randomized, controlled study in advanced Parkinson disease. Neurology 2007;68 (14):1108-15. 31. Rinne UK, Bracco F, Chouza C, Dupont E, Gershanik O, Marti Masso JF, Montastruc JL, Marsden CD, Dubini A, Orlando N, Grimaldi R. Cabergoline in the treatment of early Parkinson’s disease: results of the first year of treatment in a double-blind comparison of cabergoline and levodopa. The PKDS009 Collaborative Study Group. Neurology 1997;48(2):363-8. 32. Lieberman A, Imke S, Muenter M, Wheeler K, Ahlskog JE, Matsumoto JY, Maraganore DM, Wright KF, Schoenfelder J. Multicenter study of cabergoline, a long-acting dopamine receptor agonist, in Parkinson’s disease patients with fluctuating responses to levodopa/carbidopa. Neurology 1993;43(10):1981-4. 33. Hutton JT, Koller WC, Ahlskog JE, Pahwa R, Hurtig HI, Stern MB, Hiner BC, Lieberman A, Pfeiffer RF, Rodnitzky RL, Waters CH, Muenter MD, Adler CH, Morris JL. Multicenter, placebo-controlled trial of cabergoline taken once daily in the treatment of Parkinson’s disease. Neurology 1996;46(4):1062-5. 34. Yamamoto M, Uesugi T, Nakayama T. Dopamine agonists and cardiac valvulopathy in Parkinson disease. Neurology 2006;67:1225-9.
1523
1524
90
Medical management of parkinson’s disease
35. Jankovic J, Watts RL, Martin W, Boroojerdi B. Transdermal rotigotine: double-blind, placebo-controlled trial in Parkinson disease. Arch Neurol 2007;64(5):676-82. 36. Parkinson Study Group. A controlled trial of rotigotine monotherapy in early Parkinson’s disease. Arch Neurol 2003;60:1721-8. 37. LeWitt PA, Lyons KE, Pahwa R. SP 650 Study Group. Advanced Parkinson disease treated with rotigotine transdermal system: PREFER study. Neurology 2007;68(16):1262-7. 38. Pahwa R, Koller WC, Trosch RM, Sherry JH. APO303 Study Investigators. Subcutaneous apomorphine in patients with advanced Parkinson’s disease: a doseescalation study with randomized, double-blind, placebo-controlled crossover evaluation of a single dose. J Neurol Sci 2007;258(1–2):137-43. 39. Katzenschlager R, Hughes A, Evans A, Manson AJ, Hoffman M, Swinn L, Watt H, Bhatia K, Quinn N, Lees AJ. Continuous subcutaneous apomorphine therapy improves dyskinesias in Parkinson’s disease: a prospective study using single-dose challenges. Mov Disord 2005;20(2):151-7. 40. Keranen T, Gordin A, Karlsson M, et al. Inhibition of soluble catechol-O-methyltransferase and single-dose pharmacokinetics after oral and intravenous administration of entacapone. Eur J Clin Pharmacol 1994;46:151-7. 41. Nutt JG, Woodward WR, Beckner RM, et al. Effect of peripheral catechol-O-methyltransferase inhibition on the pharmacokinetics and pharmacodynamics of levodopa in parkinsonian patients. Neurology 1994;44: 913-9. 42. Ruottinen HM, Rinne UK. A double-blind pharmacokinetic and clinical dose-response study of entacapone as an adjuvant to levodopa therapy in advanced Parkinson’s disease. Clin Neuropharm 1996;19:283-96. 43. Parkinson’s Study Group. Entacapone improves motor fluctuations in levodopa-treated Parkinson’s disease patients. Ann Neurol 1997;42:747-55. 44. Poewe WH, Deuschl G, Gordin A, Kultalahti ER, Leinonen M; Celomen Study Group. Efficacy and safety of entacapone in Parkinson’s disease patients with suboptimal levodopa response: a 6-month randomized placebo-controlled double-blind study in Germany and Austria (Celomen study). Acta Neurol Scand 2002;105(4):245-55. 45. Deuschl G, Vaitkus A, Fox GC, Roscher T, Schremmer D, Gordin A; CAMP Study Group. Efficacy and tolerability of Entacapone versus Cabergoline in parkinsonian patients suffering from wearing-off. Mov Disord 2007;22 (11):1550-5. 46. Smith LA, Jackson MJ, Al-Barghouthy G, Rose S, Kuoppamaki M, Olanow W, Jenner P. Multiple small doses of levodopa plus entacapone produce continuous dopaminergic stimulation and reduce dyskinesia induction in MPTP-treated drug-naive primates. Mov Disord 2005;20(3):306-14.
47. Brooks DJ, Agid Y, Eggert K, Widner H, Ostergaard K, Holopainen A; TC-INIT Study Group. Treatment of end-of-dose wearing-off in parkinson’s disease: stalevo (levodopa/carbidopa/entacapone) and levodopa/DDCI given in combination with Comtess/Comtan (entacapone) provide equivalent improvements in symptom control superior to that of traditional levodopa/DDCI treatment. Eur Neurol 2005;53(4):197-202. 48. Lyons KE, Pahwa R. Conversion from sustained release carbidopa/levodopa to carbidopa/levodopa/entacapone (stalevo) in Parkinson disease patients. Clin Neuropharmacol 2006;29(2):73-6. 49. Napolitano A, Zu¨rcher G, Da Prada M. Effects of tolcapone, a novel catechol-O-methyltransferase inhibitor, on striatal metabolism of L-dopa and dopamine in rats. Eur J Pharmacol 1995;273:215-21. 50. Adler CH, Singer C, O’Brien C, Hauser RA, Lew MF, Marek KL, Dorflinger E, Pedder S, Deptula D, Yoo K. Randomized, placebo-controlled study of tolcapone in patients with fluctuating Parkinson disease treated with levodopa-carbidopa. Tolcapone Fluctuator Study Group III. Arch Neurol 1998;55(8):1089-95. 51. Baas H, Beiske AG, Ghika J, Jackson M, Oertel WH, Poewe W, Ransmayr G. Catechol-O-methyltransferase inhibition with tolcapone reduces the ‘‘wearing off ’’ phenomenon and levodopa requirements in fluctuating parkinsonian patients. Neurology 1998;50(5 Suppl 5): S46-S53. 52. Rajput A, Martin W, Saint-Hilaire M-H, Dorflinger E, Pedder S. Tolcapone improves motor function in parkinsonian patients with ‘‘wearing-off ’’ phenomenon: a double-blind, placebo-controlled, multicenter trial. Neurology 1997;49:1066-71. 53. Tatton WG, Chalmers-Redman RME. Modulation of gene expression rather than monoamine oxidase inhibition: ()deprenyl-related compounds in controlling neurodegeneration. Neurology 1996;47:171-83. 54. Mytilineou C, Radcliffe PM, Olanow CW. L-()Desmethylselegiline, a metabolite of L-()-selegiline, protects mesencephalic dopamine neurons from excitotoxicity in vitro. J Neurochem 1997;68:434-6. 55. Weinreb O, Amit T, Bar-Am O, Chillag-Talmor O, Youdim MB. Novel neuroprotective mechanism of action of rasagiline is associated with its propargyl moiety: interaction of Bcl-2 family members with PKC pathway. Ann NY Acad Sci 2005;1053:348-55. 56. Olanow CW. Rationale for considering that propargylamines might be neuroprotective in Parkinson’s disease. Neurology 2006;66(10 Suppl 4):S69-79. 57. LeWitt PA. Deprenyl’s effect at slowing progression of parkinsonian disability: the DATATOP study. The Parkinson Study Group. Acta Neurol Scan suppl 1991;136:79-86. 58. Waters CH, Sethi KD, Hauser RA, Molho E, Bertoni JM; Zydis Selegiline Study Group. Zydis selegiline reduces off time in Parkinson’s disease patients with motor
Medical management of parkinson’s disease
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
fluctuations: a 3-month, randomized, placebo-controlled study. Mov Disord 2004;19(4):426-32. Finberg JP, Takeshima T, Johnston JM, Commissiong JW. Increased survival of dopaminergic neurons by rasagiline, a monoamine oxidase-B inhibitor. Neuroreport 1998;9:703-7. Goggi J, Theofilipoulos S, Riaz SS, Jauniaux E, Stern GM, Bradford HF. The neuronal survival effects of rasagiline and deprenyl on fetal human and rat ventral mesencephalic neurons in culture. Neuroreport 2000;11:3937-41. Blandini F, Armentero MT, Fancellu R, Blaugrund E, Nappi G. Neuroprotective effects of rasagiline in a rodent model of Parkinson’s disease. Exp Neurol 2004;187:455-9. Parkinson Study Group. A controlled, randomized, delayed-start study of rasagiline in early Parkinson disease. Arch Neurol 2004;61:561-6. Parkinson Study Group. A randomized placebocontrolled trial of rasagiline in levodopa-treated patients with Parkinson disease and motor fluctuations: the PRESTO study. Arch Neurol 2005;62(2):241-8. Rascol O, Brooks DJ, Melamed E, Oertel W, Poewe W, Stocchi F, Tolosa E; LARGO study group. Rasagiline as an adjunct to levodopa in patients with Parkinson’s disease and motor fluctuations (LARGO, Lasting effect in Adjunct therapy with Rasagiline Given Once daily, study): a randomised, double-blind, parallel-group trial. Lancet 2005;365(9463):947-54. Metman LV, Del Dotto P, Munckhof van den P, Fang J, Mouradian MM, Chase TN. Amantadine as treatment for dyskinesias and motor fluctuations in Parkinson’s disease. Neurology 1998;50:1323-6. Del Dotto P, Pavese N, Gambaccini G, Bernardini S, Metman LV, Chase TN, Bonucceli U. Intravenous amantadine improves levodopa-induced dyskinesias: an acute double-bling placebo-controlled study. Mov Disord 2001;16(3):515-20. Metman LV, Del Dotto P, LePoole K, Konitsiotis S, Fang J, Chase TN. Amantadine for levodopa-induced dyskinesias: a 1-year follow-up study. Arch Neurol 1999;56(11):1383-6. Schapira AH, Obeso J. Timing of treatment initiation in Parkinson’s disease: a need for reappraisal? Ann Neurol 2006;59(3):559-62. Verbaan D, Marinus J, Visser M, van Rooden SM, Stiggelbout AM, van Hilten JJ. Patient-reported autonomic symptoms in Parkinson disease. Neurology 2007;69:333-41. Abbott RD, Petrovitch H, White LR, et al. Frequency of bowel movements and the future risk of Parkinson’s disease. Neurology 2001;57:456-62. Ueki A, Otsuka M. Life style risks of Parkinson’s disease: association between decreased water intake and constipation. J Neurol 2004;251 Suppl 7:vII18-vII23. Abbott RD, Ross GW, Petrovitch H, Tanner CM, Davis DG, Masaki KH, Launer LJ, Curb JD, White LR. Bowel
73.
74.
75.
76.
77.
78.
79.
80.
81.
82. 83. 84. 85.
86.
87.
88.
90
movement frequency in late-life and incidental Lewy bodies. Mov Disord 2007;22(11):1581-6. Sakakibara R, Odaka T, Uchiyama T, et al. Colonic transit time and rectoanal videomanometry in Parkinson’s disease. J Neurol Neurosurg Psychiatry 2003;74 (2):268-72. Hattori T, Yasuda K, Kita K, Hirayama K. Voiding dysfunction in Parkinson’s disease. J Psychiatry Neurol 1992;46:181-6. Stocchi F, Carbone A, Inghilleri M, et al. Urodynamic and neurophysiological evaluation in Parkinson’s disease and multiple system atrophy. J Neurol Neurosurg Psychiatry 1997;62:507-11. Hobson P, Islam W, Roberts S, Adhiyman V, Meara J. The risk of bladder and autonomic dysfunction in a community cohort of Parkinson’s disease patients and normal controls. Parkinsonism Relat Disord 2003;10:67-71. Brusa L, Petta F, Pisani A, Miano R, Stanzione P, Moschella V, Galati S, Finazzi Agro` E. Central acute D2 stimulation worsens bladder function in patients with mild Parkinson’s disease. J Urol 2006;175(1):202-6; discussion 206-7. Goldstein D, Holmes C, Li S, Bruce S, Metman L, Cannon R. Cardiac sympathetic denervation in Parkinson disease Ann Intern Med 2000;133:338-47 Goldstein DS, Eldadah BA, Holmes C, et al. Neurocirculatory abnormalities in Parkinson disease with orthostatic hypotension. Independence from levodopa treatment. Hypertension 2005;46:1-7. Golstein DS. Cardiac denervation in patients with Parkinson disease. Cleve Clin J Med 2007;74 Suppl 1: S91-S94. Senard JM, Raı¨ S, Lapeyre-Mestre M, Brefel C, Rascol O, Rascol A, Montastruc JL. Prevalence of orthostatic hypotension in Parkinson’s disease. J Neurol Neurosurg Psychiatry 1997;63:584-89. Golstein DS. Orthostatic hypotension as an early finding in Parkinson’s disease. Clin Auton Res 2006;16(1):46-54. Calne DB, Brennan J, Spiers ASD, Stern GM. Hypotension caused by L-dopa. BMJ 1960;1:474-5. Schoenberger JA. Drug-induced orthostatic hypotension. Drug Saf 1991;6:402-7. Kujawa K, Leurgans S, Raman R, Blasucci L, Goetz C. Acure Hypotension when starting dopamine agonists in Parkinson’s disease. Arch Neurol 2000;57:1461-3. Etminan M, Gill S, Samii A. Comparison of the risk of adverse events with pramipexole and ropinirole in patients with Parkinson’s disease: a meta-analysis. Drug Saf 2003;26(6):439-44. Obering CD, Chen JJ, Swope DM. Update on apomorphine for the rapid treatment of hypomobility (‘‘off ’’) episodes in Parkinson’s disease. Pharmacotherapy 2006;26(6):840-52. Visser M, Marinus J, Stiggelbout AM, Van Hilten JJ. Assessment of autonomic dysfunction in Parkinson’s disease: the SCOPA-AUT. Mov Disord 2004;19(11):1306-12.
1525
1526
90
Medical management of parkinson’s disease
89. Zesiewicz TA, Helal M, Hauser RA. Sildenafil citrate (Viagra) for the treatment of erectile dysfunction in men with Parkinson’s disease. Mov Disord 2000;15(2):305-8. 90. Raffaele R, Vecchio I, Giammusso B, Morgia G, Brunetto MB, Rampello L. Efficacy and safety of fixeddose oral sildenafil in the treatment of sexual dysfunction in depressed patients with idiopathic Parkinson’s disease. Eur Urol 2002;41(4):382-6. 91. Hogl B, Seppi K, Brandauer E, Glatzl S, Frauscher B, Niedermuller U, Wenning G, Poewe W. Increased daytime sleepiness in Parkinson’s disease: a questionnaire survey. Mov Disord 2003;18(3):319-23. 92. Ondo WG, Dat VK, Khan H, Atassi F, Kwak C, Jankovic J. Daytime sleepiness and other sleep disorders in Parkinson’s disease. Neurology 2001;57:1392-6. 93. Kumar S, Bhatia M, Behari M. Sleep disorders in Parkinson’s disease. Mov Disord 2002l;17(4):775-81. 94. Hauser RA, Gauger L, Anderson WM, Zesiewicz TA. Pramipexole-induced somnolence and episodes of daytime sleep. Mov Disord 2000;15(4):658-63. 95. Pal S, Bhattacharya KF, Agapito C, Chaudhuri KR. A study of excessive daytime sleepiness and its clinical significance in three groups of Parkinson’s disease patients taking pramipexole, cabergoline and levodopa mono and combination therapy. J Neural Transm 2001;108:71-7. 96. Abbott RD, Ross GW, White LR, Tanner CM, Masaki KH, Nelson JS, Curb JD, Petrovitch H. Excessive daytime sleepiness and subsequent development of Parkinson disease. Neurology 2005;65(9):1442-6. 97. Nieves AV, Lang AE. Treatment of excessive daytime sleepiness in patients with Parkinson’s disease with modafinil. Clin Neuropharmacol 2002;25(2):111-4. 98. Adler CH, Caviness JN, Hentz JG, Lind M, Tiede J. Randomized trial of modafinil for treating subjective daytime sleepiness in patients with Parkinson’s disease. Mov Disord 2003;18(3):287-93. 99. Ondo WG, Fayle R, Atassi F, Jankovic J. Modafinil for daytime somnolence in Parkinson’s disease: double blind, placebo controlled parallel trial. J Neurol Neurosurg Psychiatry 2005;76(12):1636-9. 100. Hobson DE, Lang AE, Martin WRW, Razmy A, Rivest J, Fleming J. Excessive daytime sleepiness and sudden sleep onset in Parkinson’s disease: a survey by the Canadian Movement Disorders Group. JAMA 2002;287:455-63. 101. Gjerstad MD, Wentzel-Larsen T, Aarsland D, Larsen JP. Insomnia in Parkinson’s disease: frequency and progression over time. J Neurol Neurosurg Psychiatry 2007;78(5):476-9. 102. Stack EL, Ashburn AM. Impaired bed mobility and disordered sleep in Parkinson’s disease. Mov Disord 2006;21(9):1340-2. 103. Juri C, Chana´ P, Tapia J, Kunstmann C, Parrao T. Quetiapine for insomnia in Parkinson disease: results from an open-label trial. Clin Neuropharmacol 2005;28(4):185-7.
104. Mahowald MW, Schenck CH. Insights from studying human sleep disorders. Nature 2005;437:1279-85. 105. Boeve BF, Silber MH, Parisi JE, et al. Synucleinopathy pathology and REM sleep behavior disorder plus dementia or parkinsonism. Neurology 2003;61:40-5. 106. Schenck CH, Bundlie SR, Mahowald MW. Delayed emergence of a parkinsonian disorder in 38% of 29 older men initially diagnosed with idiopathic rapid eye movement sleep behaviour disorder. Neurology 1996;46 (2):388-93. 107. Gagnon JF, Bedard MA, Fantini ML, et al. REM sleep behavior disorder and REM sleep without atonia in Parkinson’s disease. Neurology 2002;59:585-9. 108. Vendette M, Gagnon JF, De´cary A, MassicotteMarquez J, Postuma RB, Doyon J, Panisset M, Montplaisir J. REM sleep behavior disorder predicts cognitive impairment in Parkinson disease without dementia. Neurology 2007;69:1843-9. 109. Gagnon JF, Fantini ML, Bedard MA, et al. Association between waking EEG slowing and REM sleep behavior disorder in PD without dementia. Neurology 2004;62:401-6. 110. Dooneief G, Mirabello E, Bell K, Marder K, Stern Y, Mayeux R. An estimate of the incidence of depression in idiopathic Parkinson’s disease. Arch Neurol 1992;49:305-7. 111. Ravina B, Camicioli R, Como PG, Marsh L, Jankovic J, Weintraub D, Elm J. The impact of depressive symptoms in early Parkinson disease. Neurology 2007;69:342-7. 112. Weintraub D, Moberg PJ, Duda JE, Katz IR, Stern MB. Effect of psychiatric and other nonmotor symptoms on disability in Parkinson’s disease. J Am Geriatr Soc 2004;52:784-8. 113. Shulman LM, Taback RL, Rabinstein AA, Weiner WJ. Non-recognition of depression and other non-motor symptoms in Parkinson’s disease. Parkinsonism Relat Disord 2002;8:193-7. 114. Weintraub D, Moberg PJ, Duda JE, Katz IR, Stern MB. Recognition and treatment of depression in Parkinson’s disease. J Geriatr Psychiatry Neurol 2003;16:178-83. 115. Molina JA, Sa´inz-Artiga MJ, Fraile A, et al. Pathologic gambling in Parkinson’s disease: a behavioral manifestation of pharmacologic treatment? Mov Disord 2000;15:869-72. 116. Driver-Dunckley E, Samanta J, Stacy M. Pathological gambling associated with dopamine agonist therapy in Parkinson’s disease. Neurology 2003;61:422-3. 117. Dodd ML, Klos KJ, Bower JH, et al. Pathological gambling caused by drugs used to treat Parkinson disease. Arch Neurol 2005;62:1377-81. 118. Weintraub D, Siderowf AD, Potenza MN, Goveas J, Morales KH, Duda JE, Moberg PJ, Stern MB. Association of dopamine agonist use with impulse control disorders in Parkinson disease. Arch Neurol 2006;63 (7):969-73.
Medical management of parkinson’s disease
119. Giladi N, Weitzman N, Schreiber S, Shabtai H, Peretz C. New onset heightened interest or drive for gambling, shopping, eating or sexual activity in patients with Parkinson’s disease: the role of dopamine agonist treatment and age at motor symptoms onset. J Psychopharmacol 2007;21(5):501-6. 120. Voon V, Hassan K, Zurowski M, de Souza M, Thomsen T, Fox S, Lang AE, Miyasaki J. Prevalence of repetitive and reward-seeking behaviors in Parkinson disease. Neurology 2006;67(7):1254-7. 121. Voon V, Thomsen T, Miyasaki JM, de Souza M, Shafro A, Fox SH, Duff-Canning S, Lang AE, Zurowski M. Factors associated with dopaminergic drug-related pathological gambling in Parkinson disease. Arch Neurol 2007;64(2):212-6. 122. Mendez MF, O’Connor SM, Lim GT. Hypersexuality after right pallidotomy for Parkinson’s disease. J Neuropsychiatry Clin Neurosci 2004;16(1):37-40. 123. Roane DM, Yu M, Feinberg TE, Rogers JD. Hypersexuality after pallidal surgery in Parkinson disease. Neuropsychiatry Neuropsychol Behav Neurol 2002;15(4):247-51. 124. Ardouin C, Voon V, Worbe Y, Abouazar N, Czernecki V, Hosseini H, Pelissolo A, Moro E, Lhomme´e E, Lang AE, Agid Y, Benabid AL, Pollak P, Mallet L, Krack P. Pathological gambling in Parkinson’s disease improves on chronic subthalamic nucleus stimulation. Mov Disord 2006;21(11):1941-6. 125. Witjas T, Baunez C, Henry JM, Delfini M, Regis J, Cherif AA, Peragut JC, Azulay JP. Addiction in Parkinson’s disease: impact of subthalamic nucleus deep brain stimulation. Mov Disord 2005;20(8):1052-5. 126. Bandini F, Primavera A, Pizzorno M, Cocito L. Using STN DBS and medication reduction as a strategy to treat pathological gambling in Parkinson’s disease. Parkinsonism Relat Disord 2007;13(6):369-71. 127. Hely MA, Morris JG, Reid WG, Trafficante R. Sydney Multicenter Study of Parkinson’s disease: non-L-dopa-responsive problems dominate at 15 years. Mov Disord 2005;20(2):190. 128. Hobson R, Meara J. Risk and incidence of dementia in a cohort of older subjects with Parkinson’s disease in the United Kingdom. Mov Disord 2004;19(9):1043-9.
90
129. Hurtig HI, Trojanowski JQ, Galvin J, Ewbank D, Schmidt ML, Lee VM, Clark CM, Glosser G, Stern MB, Gollomp SM, Arnold SE. Alpha-synuclein cortical Lewy bodies correlate with dementia in Parkinson’s disease. Neurology 2000;54(10):1916-21. 130. Emre M, Aarsland D, Albanese A, Byrne EJ, Deuschl G, De Deyn PP, Durif F, Kulisevsky J, van Laar T, Lees A, Poewe W, Robillard A, Rosa MM, Wolters E, Quarg P, Tekin S, Lane R. Rivastigmine for dementia associated with Parkinson’s disease. N Engl J Med 2004;351 (24):2509-18. 131. Poewe W, Wolters E, Emre M, Onofrj M, Hsu C, Tekin S, Lane R; EXPRESS Investigators. Long-term benefits of rivastigmine in dementia associated with Parkinson’s disease: an active treatment extension study. Mov Disord 2006;21(4):456-61. 132. Burn D, Emre M, McKeith I, De Deyn PP, Aarsland D, Hsu C, Lane R. Effects of rivastigmine in patients with and without visual hallucinations in dementia associated with Parkinson’s disease. Mov Disord 2006;21 (11):1899-907. 133. Cummings J, Lefe`vre G, Small G, Appel-Dingemanse S. Pharmacokinetic rationale for the rivastigmine patch. Neurology 2007;69(4 Suppl 1):S10-S13. 134. Ravina B, Putt M, Siderowf A, Farrar JT, Gillespie M, Crawley A, Fernandez HH, Trieschmann MM, Reichwein S, Simuni T. Donepezil for dementia in Parkinson’s disease: a randomised, double blind, placebo controlled, crossover study. J Neurol Neurosurg Psychiatry 2005;76(7):934-9. 135. Klein C, Prokhorov T, Miniovich A, Dobronevsky E, Rabey JM. Long-term follow-up (24 months) of quetiapine treatment in drug-induced Parkinson disease psychosis. Clin Neuropharmacol 2006;29 (4):215-19. 136. Kurlan R, Cummings J, Raman R, Thal L; Alzheimer’s Disease Cooperative Study Group. Quetiapine for agitation or psychosis in patients with dementia and parkinsonism. Neurology 2007;68(17):1356-63. 137. Rabey JM, Prokhorov T, Miniovitz A, Dobronevsky E, Klein C. Effect of quetiapine in psychotic Parkinson’s disease patients: a double-blind labeled study of 3 months’ duration. Mov Disord 2007;22(3):313-8.
1527
113 Microvascular Decompression for Trigeminal Neuralgia J. R. Pagura
Although it is a well-known clinical entity, trigeminal neuralgia (TN) has, over the course of years, been the object of controversial discussion, not only as regards etiology, but mainly as to treatment. Even after Nicholas Andre´ [1] published a description of TN in 1756, it was most probably mistaken for other nosological entities, with diverse modes of treatment attempted [2]. After several centuries, despite technological developments affording more precise diagnoses and improved conditions for surgical treatment, this pathology is fat from being completely understood, and the multiplicity of treatments advocated over this period corroborates this. A lack of better anatomic and physiological knowledge and uncertain anesthesia techniques were limiting factors for the surgical treatment of TN in early times. For this reason, the simplest form of treatment invariable proved the best choice, even though it may not always have been the one based on the best pathophysiological concept. In writing on microvascular decompression (MVD), we should like to focus on the work of some authors. Dandy [3], in 1932, laid the foundations for MVD when he stood for approaching TN by a cerebellar route via the posterior fossa, emphasizing that, in a number of instances, he had found a vascular loop in contact with this nerve. It is of interest to note that 27 years elapsed before Gardner and Miklos [4] confirmed Dandy’s assumption that the vascular loop might well produce alterations in the myelin at the level of the root entry zone. There were considerable advances in MVD in 1966 with the work of Jannetta and Rand [5]. #
Springer-Verlag Berlin/Heidelberg 2009
These authors performed what was probably the first MVD with the use of the microscope, using the transtentorial subtemporal approach that, in our view, was more complicated than that advocated by Dandy and Gardner. These same authors were later to make use of the cerebellar route through the posterior fossa. There is no doubt that improved anesthesia techniques with routine use of a microscope were factors that were fundamental in popularizing MVD. However, the possibility to treat the trigeminal neuralgia maintaining the facial sensibility is, in my opinion, the most important thing. The good results and the lower complication rates shown in many author‘s series encourage surgeons to choose the microvascular decompression as their first option in the treatment of trigeminal neuralgia.
Pathophysiology Vascular compression (VC) and MVD surgery, even if widely accepted, are still a source of controversy in the literature. Cranial nerves are usually positioned in close relation to arteries and veins; their proximity probably fluctuates with the degree of pulsation of the vessels and of the cerebrospinal fluid and the position of the head [6]. For reasons as yet unknown, the trigeminal, facial, and glossopharyngeal nerves seem more susceptible to VC than the other nerves [6]. Those who do not accept VC as a casual factor for TN hold that the surgeon is simply uncovering a normal nerve-vessel relationship that would also be found in persons free of pain.
1912
113
Microvascular decompression for trigeminal neuralgia
Some authors have tried to elucidate this controversy. Hamlyn and Kind [7] developed clinical and experimental work concomitantly on vascular compression in patients with TN. From the clinical standpoint, they encountered 90% of VC in patients submitting to MVD. An experimental study in cadavers of patients who died with no previous background of TN did not show evidence of vascular compression of depression of the nerve, although perfusion of the vessels augmented contact of the vascular structures with the nerve. Hardy and Rhoton [8] in 1978 referred to finding VC in cadavers with no history of TN, so that they concluded that there must be asymptomatic compression of the trigeminal nerve. At what moment would compression become symptomatic? Would there be a trigger factor? Jannetta [9] holds that TN is typically manifested at an advanced age because cerebral atrophy and the process of arteriosclerosis – with resulting hardened, twisted arteries – would be conditions fundamental for the onset of TN. How can the presence of young patients with TN in almost all of the series published be explained? How can venous compression be explained? According to the same author, if arterial compression were to occur up to approximately 2.2 mm from the apparent visible emergence of the nerve from the brain stem, it would be taking place within an area of transition from central myelin to peripheral myelin. This area of transition, known as root entry zone (REZ), when subjected to contact with a vascular structure, would suffer demyelination with resulting dysfunction of the nerve, manifest clinically as a syndrome of hyperactive dysfunction. Short-circuiting between fibers of different diameters secondary to demyelination at the REZ may provide the mechanism for trigger points [10]. According to Moller [11], demyelination would lead to alterations at the level of the nucleus of the spinal tract of the trigeminal
nerve. This may be one explanation for the significant percentage of relapses after any method of surgical treatment for TN. Although a pathophysiological explanation may perhaps encompass several theories, what seems clear to us is that removal of contact of the vascular loop with the nerve in any position in which it may be found results, in the great majority of cases, in relief from the symptomatology with no alteration in sensitivity, signifying decompression without apparent lesion of the nerve. Cases described without confirmation of VC may occur under some circumstances. The position of the head could alter the vessel-nerve relationship, explaining a lesser occurrence of nocturnal TN crisis in the majority of patients. In clinical observations, Jannetta [9] confirmed that the majority of patients lie in lateral decubitus contralateral to the pain, a posture that would tend to remove the impact of the VC. Besides the positioning of the head and opening of the dura mater, retraction of the cerebellum and opening of the cistern in the region of the cerebellopontine angle should significantly alter the relationship between nerves and vessels in the posterior fossa, so that the operative field may not always reveal the true relationship between them. Another limiting factor in identifying VC would be a lack of experience on the part of the neurosurgeon in identifying it; several experienced surgeons report a similar fact in literature [12–15]. This controversy may perhaps be summarized in one point: attempts have been made for many years to treat the symptom of TN; MVD in an endeavor to treat a possible cause.
Clinical Manifestations Paroxysmal stabbing pain of short duration that can be triggered by a discrete tactile stimulus (trigger point) generally affecting one or more divisions of the trigeminal nerve are characteristic
Microvascular decompression for trigeminal neuralgia
of TN. For many years, the vast majority of citations regarded the absence of alteration in sensitivity of the face to be characteristic of TN. In our series we found patients with typical Trigeminal Neuralgia with mild alterations of the facial sensibility [16]. Pain was generally unilateral and could, in 5–10% of patients, be bilateral at some stage of the disease. In our experience, we observed the onset of bilateral TN after surgical treatment on the side on which the pain began. Trigeminal neuralgia most often affects women close to age 50 [10].
Investigation Classically, investigation of patients with TN was carried out with either a computed tomography (CT) or a Magnetic Resonance of the brain. I believe that the MRI should be the first choice when we talk about complementary exams. The diagnosis of trigeminal neuralgia is done based on the history of the disease and the neurological examination. The image exams are done just to avoid some pathology which could cause symptoms of trigeminal neuralgia like tumors or AVMs. Despite having developed the image like magnetic resonance in which the neurovascular conflict could be seen the choice of the surgery is independent from these findings. The cerebral angiography by MRI which can show us a vascular loop in the region of the cerebellopontine angle is important when we suspect of the presence of dolicobasilar artery.
Selection of Patients for Surgery All of the patients with TN are initially submitted to clinical treatment with carbamazepine, phenytoin, clonazepan, or baclofen. In cases that did not respond to the medication or
113
where the collateral effects were very intense, we then went on to consider surgical treatment. On this point, we share Apfelbaum1s [12] concern that the procedure chosen for each case represents the best effort on the part of the doctor, weighing pros and cons of each possible method of treatment. The doctor is responsible for suggesting the appropriate procedure, which must then be discussed extensively with the patient, mainly with regard to the possibility of recurrence or sequelae. We believe, however, that once our basic criteria for suggesting a particular procedure have been met, the patient may opt for whatever type of surgical treatment he or she chooses. We feel comfortable enough with this because we perform all the main types of surgery for the treatment of TN. Age over 65 years was formerly a contraindication to an approach via the posterior fossa. However, after we acquired more experience with the procedure, we did away with this limitation. Today, if the patient is in good clinical condition, age is not a contraindication to MVD, our principal strategy for treating TN. We always recommend MVD as a procedure of choice for young patients, the same being the case for those patients with compromise of the first division of the trigeminal nerve, owing to the risks of anesthesia of the cornea and a greater incidence of dysesthetic phenomena.
Material Between January 1981 and July 2007, 324 patients with TN submitted to MVD. Not included in these statistics were patients with a diagnosis of aneurysm of the basilar artery, arteriovenous malformation, meningioma, neurinoma, and epidermoid tumor. Some of the patients in the series have been submitted to MVD in other centers and,
1913
1914
113
Microvascular decompression for trigeminal neuralgia
owing to the persistence of pain, submitted to a second MVD on our service. For purposes of analysis of the material, we considered those patients as having submitted to a first surgery. Second operations in our series are discussed separately. There was predominance of the female sex: there were 187 woman and 137 man. The right side was affected 181 times and the left side 143; symptomatology was bilateral, always with unilateral onset, in 8 patients (3.9%) The age of the patients at surgical treatment in our service varied from 19 to 78 years (> Table 113-1). In the majority of cases more than one division of TN was affected. The combination of TN in V2 and V3 division was present in 117 patients and was predominant (> Table 113-2). Of the patients operated upon, 52 (had previously submitted to some type of ablation procedure.
. Table 113-1 Distribution of patients by age Age 19–30 31–40 41–50 51–60 61–70 71–80
Number of patients 6 16 39 82 96 85
. Table 113-2 Anatomic distribution of pain Trigeminal nerve V1 V2 V3 V1–V2 V2–V3 V1–V2–V3
Number of patients 13 39 54 78 117 23
Surgical Technique Each Patient submitted to MVD underwent a detailed preoperative assessment. General anesthesia with orotracheal intubation is used with central venous and arterial access and a Foley catheter placed in the bladder. Trichotomy of the occipital region is restricted to the area of the incision. Earphones are adapted when we use auditory evoked potentials. Prophylactic antibiotic therapy are utilized at the onset of surgery. The patient is placed in the lateral decubitus position and securely fixed with the careful use of straps so that there is pressure of the neurovascular bundles of the upper limb close to the table. Caudal traction is exerted on the shoulder, which is turned upward in order to augment the space in which the surgeon will work. The head is fixed on the Mayfield support and must be parallel to the floor of the room. Following antisepsis and asepsis of the nuchal and occipital region, we perform a vertical and paramedian incision approximately 8 cm long about 3 cm from the mastoid apophysis. Two-thirds of this incision must be below the nuchal line and on-third above. With the aid of the electrical scalpel and an orthostatic retractor, we divide and release the nuchal muscles until we have exposed the occipital bone. The emissary veins are coagulated and the bone orifices sealed with bone wax. We effect a burr hole 1 cm below the asterion, which is enlarged with the assistance of gouges varying from 2.5 to 3.5 cm. The upper limit of the craniectomy must be the transverse sinus and the lateral edges of the sigmoid sinus. The aerated cavity of the opened mastoid must be sealed with was bone powder. The dura mater is opened in a curvilinear fashion, beginning medially and extending superolaterally to the junction of the transverse and sigmoid sinus and inferolaterally to the edge of the sigmoid sinus (> Figure 113-1). The free edge of the dura mater is sutures to the
Microvascular decompression for trigeminal neuralgia
. Figure 113-1 Diagram showing opening of the dura mater
muscular layer over the mastoid, augmenting the exposure. At this stage of the surgery, we introduce the microscope. With the aid of a 10-mm spatula, we perform a mild retraction of the upper part of the cerebellum, below the superior petrous vein, opening the arachnoid in the region of the cerebello-pontine angle. We place a small sponge and aspirate the fluid over this until the cerebellum is well below the dura mater. If the cerebellum is tense, it is possible to try to drain the cerebellomedullaris cistern located inferomedially. Only after the cerebellum is found to be in an ideally relaxed state, do we commend the gently retraction of the same. The cerebellar surface must be protected with Gelfoam and a sponge to prevent lesions to the parenchyma. The spatula is then advanced progressively until the superior petrosal vein is seen. When we reach the superior petrous complex, we shall be able to see the trigeminal nerve in an antero inferior position. In the great majority of cases, we coagulate the superior petrous vein, which permits a good view of the nerve. We recommend caution at the moment of coagulation and resection. Routinely, at this stage of the proceeding, we place a sponge between the vein and the region of trigeminal nerve, for in the event of bleeding, we may safely aspirate, preventing lesions of the nerves in the region. In increasing traction of the cerebellum, we must
113
keep the arachnoid in the region of the cerebellopontine angle close to the facial and vestibulocochlear nerves totally open, thus avoiding traction from being transmitted to those nerves. Removal of the arachnoid around the trigeminal nerve is preceded by careful inspection in the search for the VC. At this phase of the surgery, the lateral inclination of the table may be useful to better expose the nerve. Once the VC has been identified the trigeminal nerve must be completely freed from its arachnoid and the vessel must be separated from contact with the nerve. Then we must position Teflon, totally isolating the nerve. It is important to observe the two free edges of the Teflon with the vessel lying on the middle of its surface. More than one prosthesis was utilized on various occasions. When the compression is venous, the vein must be coagulated. If there is any doubt as to the effectiveness of the venous drainage, which may possibly be prejudiced when veins of la large caliber exert pressure on the nerve, dissection may be attempted and isolation with Teflon utilized. In closure, the dura mater must be carefully sutured: biological glue and fragments of muscle must be used to close possible orifices. A cranioplasty using titanium plate and bone source is done. Nuchal muscles are sutured in two layers with non absorbable sutures. The subcutaneous layer is closed with absorbable suture, with staples on the skin.
Surgical Findings Surgical findings are listed in (> Table 113-3). The artery most often found as the cause of compression of the trigeminal nerve was the superior cerebellar artery, alone (> Figure 113-2) or in combination with another artery or vein. In eight cases we could not find any evidence of compression of the trigeminal nerve.
1915
1916
113
Microvascular decompression for trigeminal neuralgia
Results This surgery was effective in alleviating the pain from TN in approximately 96% of the patients. The majority awoke from anesthesia without pain; however, a certain number continued to experience pain of considerable less intensity that persisted from a few days to weeks. As a routine, we generally continue the specific medication used by the patient in the preoperative . Table 113-3 Intraoperative findings Number of patients Arterial compression Superior cerebellar artery Anterior inferior cerebellar artery (AICA) Basilar artery Posterior inferior cerebellar artery (PICA) Unidentified artery Mixed compression Unidentified artery/vein Venous compression Vein Negative operation
197 48 2 1 29 14 23 10
. Figure 113-2 Proximal anterior and lateral compression by SCA
stage in much smaller doses for some days and then gradually taper down the dosage depending on how the patient’s symptoms evolve. Our results are listed in (> Table 113-4). Results were apparently not influenced by sex, age, duration of symptoms, or type of vascular compression. Patients treated with ablative procedures could presents of residual pain or paresthesia immediately postoperatively. The group with partial relief consisted of patients who improved considerably but who, nevertheless, still required small doses of carbamazepine or clonazepam. Of the 12 patients from the group without immediate response, 6 were reoperated upon; in 2 of these there was a technical fault in the endeavor to isolate the artery. In 4 cases, we did not observe anything to account for the continued presence of pain. Both cases with technical failure improved on the second attempt and required medication only in small doses. The other six cases continued with pain and another type of procedure and medication proved necessary. Four patients were reoperated upon owing to late recurrence of pain (over 6 months). The reoperated procedure was technically more laborious owing to adhesions from the prior intervention, and the surgical findings were not convincing of VC. All of these were submitted to other procedures and required medication. Since then, we have not subjected patients with recurrence of pain to a further MVD. The medium follow up period was 5.8 years and the recurrence of the pain occurred in 9, 3% of our patients. The great majority of cases . Table 113-4 Results of microvascular decompression Groups
Immediate
Asymptomatic Partial relief Without response Total
267 (82.4%) 45 (13.9%) 12 (3.7%) 324
a
9.3% recurrence
Microvascular decompression for trigeminal neuralgia
occurred in the first 2 years, similar to the findings of Barker [17]. Writing about the results, I believe we have to discuss changing the criteria of classification from good, poor and bad to good and bad only. In my opinion, good results must be considered when the patient is completely pain free after the surgery, without using any medicine and without having any complications. In all the other cases, even when the patient has improved, but still needs to use medicine, we need to consider it a bad result. Usually the authors consider favorable results patients totally pain free or better of their symptomatology even still using small amount of medicine. In all the cases when we have definite complications, even if the patient is completely pain free, we need to consider the results as bad. Having definite complications in the treatment of the trigeminal neuralgia should not be accepted. When we consider doing a surgery in a patient with pain like a trigeminal neuralgia, the indication is either if the patient still has pain despite the medication, or the patient presents important side effects due to the use of medicine. For this reason, the only good result we could consider is to abolish the pain completely without using any medication. In all other situations the results must be considered bad. I believe that with this simplified classification we can be sure about the effectiveness of the different surgical methods and its results.
Complications The majority of complications in MVD in our experience was transient and occurred in the beginning of our series. A great number of patients complain of headache, nausea, and dizziness, generally in the immediate postoperative period, improving over the subsequent hours. Aseptic meningitis, referred to by Jannetta [13], was not frequent in our material, even not
113
used steroids routinely as I used to use in the beginning. > Table 113-5 lists our main complications. Cerebrospinal fluid fistulas were treated with rest and external lumbar drainage; 6 cases necessitated reintervention. We later began to make use of biological glue, which practically did away with this complication. Bacterial meningitis was satisfactorily treated with systemic antibiotic therapy. In the cases where there was a cranial nerve deficit, this was usually transient. One patient who awakened after surgery without pain, developed drowsiness resulting from cerebellar edema, and died on the sixth postoperative day from pulmonary embolism. This case occurred in 1983, with a patient who had been submitted to a two radiofrequency percutaneous rhizotomy. Since 1995 I have not had any complications, but very few cases of transitory and mild dizziness.
Discussion Pain is a symptom and not a disease [18]. The many types of surgery proposed for the treatment of TN, all with reported good results and similar recurrence rates, leads us to question what the several surgical procedures used in the treatment of TN have in common that would
. Table 113-5 Postoperative complications Cerebrospinal fluid fistula Meningitis Labial herpes Death Nerves affected: IV V motor V sensitive VII VIII
16 6 18 1 Transient 2 3 – 2 11
Persistent – – 1 1 1
1917
1918
113
Microvascular decompression for trigeminal neuralgia
account for the disappearance of painful symptomatology. Sweet [19] believes that all of the procedures including MVD inflict a lesion on the nerve. Perhaps, because of this type of explanation, ablative procedures, which are generally simpler to do, have for very long taken place of more extensive surgery. Could it be that the simplicity of some of these ablative techniques would justify the loss of facial sensation? It was this question and the findings of Dandy [19], Gardner [3], and Janetta [5] that led to the recommendation that MVD be carried out in the posterior fossa to minimize the disorders of sensation occurring in other types of procedure. With the increased experience of several neurosurgeons elaborating its methodology [6,12–15], there has been every endeavor to consider treatment of TN by MVD, which eliminates the cause of its pathology rather than the other procedures that treat only its symptoms. However, although there may be a cause-andeffect relationship between the finding of a vascular loop compressing the trigeminal nerve and the abolition of pain, because of the controversy not only about the pathophysiology, but also as to the type of surgical treatment, it has not as yet been possible to change the name idiopathic trigeminal neuralgia to compressive trigeminal neuralgia. However elegant and probable the VC theory may be in explaining TN, a series of questions has as yet remained unanswered. 1. 2. 3. 4. 5.
How does one explain the incidence in young people, although rare? Why is it unilateral? Why are there frequent periods of spontaneous relief from pain? Why are there negative surgical explorations in symptomatic patients? What accounts for recurrence or persistence of pain in patients after MVD, even in the series of experienced neurosurgeons whose procedure was technically satisfactory?
6.
7.
How can asymptomatic vascular compression found in anatomic studies be explained? If the compression occurred only in the posterior fossa, why does TN improve with manipulation of this nerve in any segment, from its peripheral portion to the region of the nucleus of the spinal tract of the same?
These unanswered questions still lead to statements such as the following: ‘‘I do, although I am not sure what it is that I do’’ [20]. About the negative exploration during the surgery, I have cited for many years that in cases in which the neurovascular compression in a typical trigeminal neuralgia is not found, it is probably caused by the lack of experience of the surgeon by failing to recognize the vascular conflict in the cerebellopontine angle. In the cases that I have published, all of them occurred in the early years of my experience with the microvascular decompression. However, in one of my last cases in a patient with the ‘‘probable’’ typical trigeminal neuralgia which was submitted, for two times before to a microcompression using a ballon I could‘t find any compression at all, even after exhaustingly looking for one. I left the surgical Center without doubt that there was no compression, despite having noticed a slight depression in the distal part of the nerve. Of course even having done a ‘‘compression’’ using a micro fo´rceps in the nerve for 50 s, the result was poor. Because of cases like this we could not change the name of trigeminal neuralgia for ‘‘compression neuralgia’’ yet. Looking at the results in the literature, there was no big difference between the early results when we compare the microvascular decompression with other methods like Radiofrequency Trigeminal Rhizotomy and Micro Compression of the ganglion. In all these cases, the good results are above 90%. When a patient, after being submitted to one of these three methods, by a highly regarded team, still remains with
Microvascular decompression for trigeminal neuralgia
pain, even if the pain is milder, we must be careful to indicate another kind of surgery. We are probably before an atypical neuralgia or we even though the symptoms could seem to be a trigeminal neuralgia In cases like this, when the patient has been submitted to another type of surgery and either remains with pain, mainly the sharp pain, immediately after the surgery or has a very early recurrence of the pain, we could be before a pain caused by anatomical transformation in the nerve pathways, mainly in its central part. The absence of simultaneous compromise of divisions I and III of the trigeminal nerve in our material and generally in the literature, reinforces the VC theory, for it would be difficult, in view of their somatotopic relationship, for compression by a single vessel to compromise those two divisions. Two different arteries would be necessary, compressing separately roots VI and V3, a fact never observed in our material. Any of the other combinations of compromise of the divisions of the trigeminal nerve are, however, easily explained and are found clinically. Depending on the nerve territory affected, we may anticipate the position of the VC. According to the Jannetta’s [21] observations, a rostral compression of the trigeminal nerve cause pain in the third division, the medial and more distant compression is responsible for pain in the second and the caudal compression for pain in the first division (> Figure 113-3). Although incomplete understanding of VC pathophysiology in undeniable, the great benefit that MVD has brought to patients with TN is the possibility of being freed of pain through a safe method, without prior commitment to lesion a nerve. Like Dahle and colleagues [22], we do not consider MVD a cure for TN, but I should dare to say that MVD is the mode of treatment that possible is closer to the cause. Every time we perform MVD, we neither partially section the pars major nor massage the nerve but one time. Any neuropraxis is involuntary and only that needed during dissection. Our
113
. Figure 113-3 Big compression in the anterior and proximal portion of the nerve causing pain in the territory of V1, V2 and V3
patients normally enjoy total pain relief immediately postoperatively, the majority without any clear-cut sensory disorder. Jannetta [9,13] believes that VC would cause TN only if it occurred up to the REZ area, and that if VC were to occur beyond, the patient would present atypical facial pain. Tashiro and co-workers [23] consider VC is symptomatic only if it occurs up to the REZ area and proposes straightforward transposition just beyond the REZ in cases such as compression by an artery transfixing the nerve. We do not believe that a neurosurgeon can identify the referred area of transition of the myelin during surgery. In our material, various patients with typical TN presented VC situated beyond the REZ area. In these cases, VC was tread in the usual manner, and patients went on to relief of pain. I don’t think that the compression only in the entry zone could be the cause of the trigeminal neuralgia We have observed in all these years patients with typical trigeminal neuralgia in which the nerve vascular conflict was done either in the posterior part of the nerve or in its distal portion, mainly when it was venous compression.
1919
1920
113
Microvascular decompression for trigeminal neuralgia
Removal of these compressions led to the total recovery of the symptoms Venous compression is generally more difficult to identify, and even in cases of compression through tumors, the possibility of VC of arterial or venous origin must not be discarded. One instructive case was that of a girl of 9 years not included in this series. The patient had a facial pain secondary to an astrocytoma of the brain stem with growth exophytic to the cerebellopontine angle on the same side as the pain. A percutaneous radiofrequency rhizotomy relieved her to pain for some months. Because her pain returned, we chose to approach the posterior fossa and, having removed the expohytici portion of the tumor, we observed the presence of venous compression of the trigeminal nerve that was, in fact, grooved buy the vein. This patient has been asymptomatic for the last 15 years. Although routine coagulation of the superior group of veins did not present any clinical repercussion in our cases, we recommend that, before coagulation of same, every attempt be made to identify the vascular compression. If the compression on the nerve is of venous origin, it will probably be produced by an inferior venous group. I these cases, we try to avoid coagulation of the superior venous group and will do so only if there is no other alternative. In some cases instead of coagulate the vein responsible for the compression, we place a Teflon to separate the vessel from the nerve (> Figures 113-4 and > 113-5). Sometimes, according to preoperative images we can think about giving up the microvascular decompression. One of the cases is when you find a big ecstasy or a dollycomega basilar artery. Normally, because of its anatomy position, it is not responsible for the compression although we have two cases in our series. To be sure about what artery is compressing the nerve, the basilar artery must be displaced (> Figures 113-6 and > 113-7), and this situation could harm the final results of the surgery,
. Figure 113-4 Venous compression in distal position. Endoscopy view
. Figure 113-5 Decompression using teflon. Note the coagulation of the superior petrosal complex (arrow)
mainly in elderly people in which the artery is hardened and calcified. Concerning the false diagnostic impressions arising from preoperative investigation, we cite two cases of dolichomegabasilar artery implicated by preoperative MRI as the cause of VC with subsequent TN. However, upon surgical exploration after the dolichomegabasilar artery had been moved aside, we observed that that
Microvascular decompression for trigeminal neuralgia
113
. Figure 113-6 Basilar ecstasy (arrow) pulled and compression by AICA
. Figure 113-7 Removal of neurovascular compression
structure did not participate in the VC, which was produced instead in one case by AICA and in the other by a nonidentified artery, both cases being treated in the usual fashion. It‘s not so difficult to recognize the vascular compression in the cerebellopontine angle. When you don‘t find a typical compression in the proximal and anterior part of the trigeminal nerve we can use the endoscope, looking around the corner to be completely sure about the absence of the
anterior compression. With the technical development of the microscope I don‘t think it is necessary. Otherwise Chen [24], report an additional diagnose in 14.76% using endoscopies during the MVD and Kabil [25] an improved result using the endoscopies when compared to MVD. ‘‘Compression is compression’’ The relationship between nerve and artery anatomically happen. It is completely different from the compression. Either you have a convincing compression or
1921
1922
113
Microvascular decompression for trigeminal neuralgia
you are before a case like the one I described above – a patient with a typical trigeminal neuralgia without a clear compression. Retraction of the cerebellum is fundamental for good surgical exploration. Lumbar puncture as suggested by Wilkins [6] is not necessary. Use of mannitol and furosemide, as proposed by Apfelbaum [12], may also prove useful, but does not substitute for cisternal drainage of cerebrospinal fluid at the start of the procedure. The cerebellum must not be subjected to excessive retraction because we have observed severe alterations in the auditory evoked potential during the initial phase of cerebellar retraction. The same thing was observed by Sindou and colleagues [26]. According to Brock [27] the perioperative brainstean auditory evocated potencials was useful in identify changes of cochlear nerve function during the surgery. We are reluctant to reoperated on a patient subjected to MVD, appoint of view supported by our experience, and that of the authors [12,28]. In cases of early or tardy recurrence of pain, we used to reoperate only upon those cases that were forwarded to us from other services. Nowadays either in the patients treated by us or operated by experienced surgeons we suggested another type of procedure. The death in our series occurred in a patient of 78 years old who had been previously subjected to other surgical methods of treatment and who presented with pain that was extremely resistant to medication. Surgery was completed uneventfully, one artery having been isolated from TN. In the postoperative period, the patient presented cerebellar edema, remained incubated for a prolonged period of time, and finally died of pulmonary embolism. In contrast, another patient of 73 years old, also previously operated upon over three times by other methods submitted to MVD following implantation of an external cardiac pacemaker and ended up well and free of pain. Abolishing pain is the final objective of the procedures for treatment of TN. We know
greatest complaint of patients in the postoperative phase when they are free of pain revolves around the sensory disorders caused by ablative surgery that are at times more undesirable than the very tic for which the operation was done. For this reason, we advocate MVD as a first option for the treatment for TN, a procedure that abolishes pain while preserving facial sensation. The last procedure to be considered is lesioning of the nerve.
References 1. Andre´ NA. Observations pratiques sur les maladies de l’uretre. Chez Delaguette; Paris: p. 318-82. 2. Wilkins RH. Historical perspectives. In: Rovit RL, Murali R, Jannetta PJ, editors. Trigeminal neuralgia. Baltimore, MD: Williams & Wilkins; 1990. p. 1-25. 3. Dandy WE. The treatment of trigeminal neuralgia by the cerebellar route. Ann Surg 1932;96:787. 4. Gardner WJ, Miklos MV. Response of trigeminal neuralgia to ‘‘decompression’’ of sensory root: discussion of the cause of trigeminal neuralgia. JAMA 1959;170: 1773. 5. Jannetla PJ, Rand RW. Transtentorial retrogasserian rhizotomy in trigeminal neuralgia by microneurosurgical technique. Bull LA Neurol Soc 1966;31:93. 6. Wilkins RH. Neurovascular decompression procedures in the surgical management of disorders of cranial nerves V, VII, IX and X lo treat pain. In: Schmidek HH, Sweet WH, editors. Operative neurosurgical techniques. 3rd ed. Philadelphia, PA: Saunders; 1995. p. 1457-67. 7. Hamlyn PJ, King TT. Neurovascular compression in trigeminal neuralgia: a clinical and anatomical study. J Neurosurg 1992;76:948. 8. Hardy DG, Rhoton AL, Jr. Microsurgical relationship of the superior cerebellar artery and the trigeminal nerve. J Neurosurg 1978;49:669. 9. Jannetta PJ. Treatment of trigeminal neuralgia by microoperative decompression. In: Youmans JR editor. Neurological surgery, 3rd ed. Philadelphia, PA: Saunders; 1980. p. 3928-42. 10. Dott NM. Facial pain. Proc R Soc Med 1951;44:1034. 11. Moller AR. The cranial nerve vascular compression syndrome: II. A review of pathophysiology. Acta Neurochir 1991;113:24. 12. Apfelbaum RI. Surgical management of disorders of the lower cranial nerves. In: Schmidek, HHSweet WR, editors. Operative neurosurgical techniques: indications, methods and results, 2nd ed. Orlando, FL: Grune & Stratton; 1989. p. 1097-109.
Microvascular decompression for trigeminal neuralgia
13. Jannetta PJ. Microvascular decompression of the trigeminal nerve root entry zone. In: Rovit RL, Murali R, Jannetta PJ editors. Trigeminal neuralgia. Baltimore, MD: Williams & Wilkins; 1980. p. 201-22. 14. KIan B. Microvascular decompression and partial sensory rhizotomy in the treatment of trigeminal neuralgia: personal experience with 220 patients. Neurosurgery 1992;30:49. 15. Rand RW, Hunstock AT. Trigeminal neuralgia: Gardner neurovascular decompression operation: glossopharyngeal neuralgia. In: Rand RW editor. Microneurosurgery, 3rd ed. Mosby; ST Louis, MO: p. 666-82. 16. Pagura JR, Lima WC, Rabello JP. Microvascular decompression trigeminal neuralgia. Textbook of stereotactic and functional neurosurgery, chapter 175. 1996. p. 1715–21. 17. Fred G, Barker MD, Jannetta PJ, Bissonette DJ, PA-C, Mark VL, Hae Dong J. The long-term outcome of microvascular decompression for trigeminal neuralgia. N Engl J Med 1996;334(17):1077-1084. 18. Gardner WJ. Trigeminal neuralgia. Clin Neurosurg 1967;15:1. 19. Sweet WH. Complications of treating trigeminal neuralgia: an analysis of the literature and response lo questionnaire. In: Rovit RL, Murali R, Jannetta PJ editors. Trigeminal neuralgia. Baltimore, MD: Williams & Wilkins; 1990. p. 251-79. 20. Adams CBT. Microvascular compression: an alternative view and hypothesis. J Neurosurg 1989;57:1. 21. Jannetta PJ, McLaughlin MR, Casey KF. Technique of microvascular decompression. Technical note. Neurosurg Focus 2005;18(5):E5.
113
22. Dahle L, Essen CV, Kourtpoulos H, et al. Microvascular decompression for trigeminal neuralgia. Acta Neurochir 1989;99:109. 23. Tashiro H, Kondo A, Aoyama I, et al. Trigeminal neuralgia caused by compression from arteries transfixing the nerve. J Neurosurg 1991;75:783. 24. Chen MJ, Zhang WJ, Yang C, Wu YQ, Zhang ZY, Wang Y. Endoscopic neurovascular perspective in microvascular decompression of trigeminal neuralgia. J Craniomaxillofac Surg 2008. 25. Kabil MS, Eby JB, Shahinian HK. Endoscopic vascular decompression versus microvascular decompression of the trigeminal nerve. Minim Invasive Neurosurg 2005; 48(4):207-12. 26. Sindou M, Fobe´ JL, Ciriano D, Fischer C. Hearing prognosis and intraoperative guidance of brainstem auditory evoked potential in microvascular decompression. Laryngoscope 1992;102:678. 27. Brock S, Scaioli V, Ferroli P, Broggi G. Neurovascular decompression in trigeminal neuralgia: role of intraoperative neurophysiological monitoring in the learning period. Stereotact Funct Neurosurg 2004;82(5–6):199-206. 28. Yamaki T, Hashi K, Niwa J, et al: Results of reoperation for failed microvascular decompression. Acta Neurochir 1992;115:1. 29. Maxwell RE. Clinical diagnosis of trigeminal neuralgia and differential diagnosis of facial pain. In: Rovit RL, Murali R, Jannetta PJ editors. Trigeminal neuralgia. Baltimore, MD: Williams & Wilkins; 1990. p. 53-77. 30. Parkinson D. Microvascular compression-decompression: a recollection. Neurosurgical forum. J Neurosurg 1989; 70:819.
1923
100 Motor Cortex Stimulation for Parkinson’s Disease M. Meglio . B. Cioni
In recent years there has been increasing interest in stimulation of the motor cortex for therapeutic purposes. Tsubokawa and his group [1] pioneered the application of motor cortex stimulation (MCS) for chronic pain. In the near future data from different groups will be available on the effects of MCS for dystonia. In this chapter we will summarize what is actually known on the results achieved with MCS in treating symptoms of advanced Parkinson’s disease (PD). The possible mechanisms underlying such effects as well as an algorithm for surgical treatment of PD will be discussed. Neurologists and clinical neurophysiologists with their active support are essential not only in patient selection and evaluation but in collecting and interpreting data useful to increase our knowledge of the physiology of cerebral cortex and ultimately to improve surgical therapy of such distressing and difficult diseases.
Literature An analysis of the Literature shows that a total of 79 patients with Parkinson’s disease have been treated by MCS [2–13], but some of the studies may be counting previously reported cases. In 2000, Canavero and Paolotti [5] reported the case of a 72-year-old woman with advanced Parkinson’s disease who showed an improvement of symptoms following unilateral extradural MCS. The quadripolar electrode was positioned in the extradural space overlying M1, controlateral to her worst clinical side. Chronic stimulation was delivered subthreshold for any #
Springer-Verlag Berlin/Heidelberg 2009
movement or sensation, 3 V, 180 ms, 25 Hz, 3+/0 setting, off during sleep. The clinical improvement was bilateral, she could stand without assistance, climb the stairs, walk for a short distance – The UPDRS III in On-Med condition decreased from 44 to 23 after 3 months. She had a moderate to severe PD-associated dementia and this aspect also improved. L-dopa was reduced by 80%. These results were confirmed in a new patient in 2002, by the same group [6]. In 2003, Pagni et al. [13] described three new cases; and in 2005 they summed up the results obtained in 6 patients. All of them were submitted to unilateral MCS, opposite to the worst clinical side; chronic stimulation was delivered bipolarly at 2.5–6 V, 150–180 ms, 25–40 Hz, continuously; the follow ups ranged from 4 months to 2.5 years. The patients were evaluated only in the On-Med state: global UPDRS decreased by 42–62%, UPDRS III (evaluating motor performances) decreased by 32–83%. It was possible to decrease L-dopa by 11–33% in three cases and by 70–73% in 2 patients. The improvement was bilateral. They reported a case of postoperative misplacement of the electrode and no negative side effects of stimulation. In 2003, a multicenter open label study was prompted by Pagni on behalf of the Italian Society of Neurosurgery. The preliminary results have been published [3,11]. Twenty-nine patients with advanced PD were treated by MCS, confirming that any symptom of PD (tremor, rigor, akinesia, motor dexterity, posture and gait, instability, freezing) may improve. L-dopa could be reduced. In an ‘‘intention to treat analysis’’, after
1680
100
Motor cortex stimulation for parkinson’s disease
6 months of MCS the mean UPDRS III decreased by 21% in off-med condition and by 34% in onmed condition. The effect of MCS, was less evident at 12 months of follow up: decrease of UPDRS III by 13 and 21%, respectively, in off and on-med. Notably, the stimulation parameters differed in the various centers: monopolar or bipolar stimulation, 2–8 V, 60–400 ms, 20–120 Hz, continuously or only during day time; some patients were unresponsive to MCS; many of these patients had MRI findings of leucoencephalopathy, white matted ischemic foci, cerebral atrophy suggesting a diagnosis of Parkinsonism rather than PD. In 2003, a paper was published [10] on the efficacy of subdural MCS in 5 patients with Parkinsonism due to multiple systemic atrophy (MSA). The Authors concluded that MCS failed to improve motor disability in these MSA patients at 3 and 6 months and that this was probably likely a function of disease progression. In 2006, Cilia et al. [7] reported the results obtained in 5 patients with PD who fulfilled CAPSIT criteria for DBS, with the exception of age >70 years. Patients were assessed preoperatively and after 6 months of MCS of the left hemisphere, on and off medication, with stimulator on and 2 weeks later with stimulator off. Stimulation was monopolar, 3–4 V, 40–60 Hz, 180–210 ms. Outcome measures included UPDRS and dyskinesia, and changes in medication level. They found no significant mean change following MCS, but there was a trend for reduction of medications and improvement in dyskinesia; 3 patients reported a reduced off time and 4 out of 5 patients had a subjective amelioration of stability, posture and gait. The study was conducted within a selected population, but the number of patients is very small to allow any clear-cut conclusion. One more case was described by Benvenuti et al. [4]. Finally, Arle et al. [2] reported the results of bilateral MCS in 4 patients with PD
followed for 12 months. They found benefits within the first 6 months in UPDRS III (decreased by 60%) but most benefits seen initially were lost by the end of 12 months. In the Literature we found only open-label studies, no control, no randomization, no double blind. Patients population was not homogeneous, as well as parameters of stimulation. MCS appears to induce some improvement but no clear evidence is present in the Literature.
Surgery A multicontact electrode (Resume lead, model 3587A, Medtronic, Minneapolis, MN, USA) is usually placed through one or two burr holes or through a small craniotomy in the epidural space over the motor cortex unilaterally or bilaterally, parallel to the motor knob for the hand [2,14,8]. In Parkinson’s disease, the best electrode orientation and position has not been established, but it is important to know the precise electrode position for future correlation between the clinical effect of MCS and the electrode site. The key point of surgery is the accurate placement of the electrode over the motor cortex. The identification of the motor cortex is the result of the integration of anatomical, neuroradiological, functional, and neurophysiological data, taking into account the huge population variability. Anatomically there is great variability in the superficial gyri of the hemispheres; the so called ‘‘hand knob’’ is a knee of the precentral gyrus toward the postcentral gyrus. It looks like an inverted omega or epsilon in axial slices and like a posterior directed hook in sagittal slices. It represents the motor hand area and is the less variable part of the motor cortex [15]. The motor cortex may be localized morphologically by craniometers landmarks, such as the Taylor– Haughton lines or as the points of the 10–20 International EEG system. It is possible to identify the typical inverted omega shape of the hand
Motor cortex stimulation for parkinson’s disease
knob on MRI performed with fiducial markers and integrated in a neuronavigation workstation [16]. Mogilner AY et al. [17] suggested the use of magnetoencephalography (MEG) to identify the dipole over the sensorymotor cortex; the data from MEG were integrated in a frameless stereotactic database by using a three-dimensional coregistration algorithm. Pirotte et al. [18] utilized fMRI to identify the hand and tongue motor area. fMRI data were coregistered on 3D T1 weighted MRI anatomical scans and matched with data from intraoperative neurophysiology. Recently, it has been suggested DTI (Diffusion Tensor Imaging) tractography to identify the motor cortex [19], but it should be taken in mind that TDI is a mathematical probability function, not an anatomical image. All these sophisticated tools may be of help for the accurate planning of the burr hole or of the craniotomy, but a precise neurophysiological location is mandatory. We use craniometer landmarks (10–20 EEG System:CZ, C3 and C4 points) to draw the central sulcus over the scalp; the anatomical location of the motor strip is confirmed by MRI and neuronavigation. A burr hole is drilled in front of the central sulcus, medially to the presumed hand motor area, then the electrode paddle is slipped epidurally, parallel to the motor strip at the hand knob (> Figure 100-1). Then the position of the electrode is verified neurophysiologically. We use the phase reversal technique to identify the central sulcus. We stimulate the controlateral median nerve at the wrist (0.5 ms, 4.7 Hz, 20 mA) and record from each contact of the strip electrode; a cortical N20 potential is recorded over the sensory cortex, a cortical P20 potential is recorded over the motor cortex; the central sulcus is between the two contacts showing the phase reversal. In our experience it was possible to record a clear phase reversal in all the cases; Nguyen et al. also found this technique very reliable: all the P20 potentials were on the anterior bank of the central fissure while all the N20
100
potentials were on the posterior bank [20,21]. However this is true only for median nerve SEPs. Polarity inversion of potentials across the sulcus is a less reliable criterion for trigeminal SEPs than for median nerve SEPs [22]; and only occasionally a phase reversal has been described for tibial nerve SEPs [23]. It is not possible to interfere the position of the motor hand area from the position of the maximum amplitude median nerve SEP: Woolsey et al. demonstrated that the face–arm boundary is situated more laterally on the postcentral gyrus than on the pre central gyrus, by 1–2 cm [24]. It is therefore necessary to map the motor cortex. Penfield first systematically stimulated the sensory–motor cortex and described the sensory and motor ‘‘homunculus’’ [25]. He utilized a bipolar direct stimulation of the cortex, applying 50–60 Hz stimuli up to 20 mA for 1–4 s, in the awake patient and looked for movements or sensations. This technique required awake surgery, often induced complex movements involving more than one muscle and provoked epileptic seizures in an high percentage of cases (20–25%). The classical homunculus was not always reproducible: as regards to the leg representation Woolsey found that in only 1/3 of the cases the lower extremity is on the medial surface of the hemisphere, in 2/3 of the cases it extends on the lateral surface and in 27% of the cases the whole lower extremity is on the lateral surface [24]. Furthermore fMRI studies on the somatotopic representation of the hand in the primary motor cortex showed that neuronal populations involved in movements of different fingers overlap extensively and that the control of each finger movement utilizes a neural network distributed throughout the hand area rather than a spatially segregated population [26]. Recently an enlarged and displaced motor map for the hand area has been described in Parkinson’s patients. Map shifts were found in the majority of the patients (12/15), in untreated early cases as well as treated cases of long duration, and there was a correlation between inter-side difference in the
1681
1682
100
Motor cortex stimulation for parkinson’s disease
. Figure 100-1 Surgical planning for implantation of epidural quadripolar electrode paddle
severity of Parkinson’s disease symptoms (UPDRS) and interhemispheric map displacement [27]. Consequently, we suggest the motor mapping in every case of motor cortex electrode placement. Motor mapping is obtained by motor cortex focal anodal stimulation through each contact of the same strip electrode (reference Fz) with a short train of stimuli (5 stimuli, 0.5 ms, ISI 4 ms, 10–30 mA). Muscle responses are recorded from muscle bellies of the controlateral hemibody, with needle electrodes. The morphological and neurophysiological positions are then integrated in an anatomo-functional
location. This mapping technique allows the use of general anesthesia (a total intravenous anesthesia using Propofol and Remifentanyl, avoiding muscle relaxants after intubation) and has a very low rate of induced epileptic seizures (less than 4%) compared with the classical so called ‘‘Penfield’s technique’’ for motor cortex mapping. After the precise cortical location, the electrode lead(s) is (are) tunneled to a subclavicular site and connected to a neurostimulator placed in a subcutaneous pocket (Kinetra, model 7428, Medtronic).
Motor cortex stimulation for parkinson’s disease
Personal Experience In 2003, we started a prospective study to evaluate the efficacy of MCS in patients with advanced Parkinson’s disease. The inclusion criteria were: - idiopathic Parkinson disease (PDSBB criteria); - at least 5 years disease’s length; - disease in the advanced state (UPDRS in off >/=40/180; Hoehn and Yahrs >/=3; motor complications: fluctuations and disabling dyskinesias); - positive response to L-dopa; - DBS not accepted by the patient or contraindicated; - patient ability to give informed consent to the study. The exclusion criteria were: - history of epilepsy or EEG epileptic
100
activity; - alcohol or drug abuse; - mental deterioration; - psychiatric symptoms; - previous basal ganglia surgery; - other major illness. Eleven patients met the above-mentioned criteria and were submitted to the implant of an epidural plate electrode over the motor cortex controlateral to the worst clinical side in three cases, and to a bilateral implant in the remaining eight cases (> Figure 100-2), according to the technique described above. Therapeutic stimulation during the first year was through the electrode controlateral to the worst clinical side; parameters were: 120 ms, 80 Hz, 3–6V (subthreshold for movements, and motor or sensory feelings), delivered continuously through contacts 0 (anode) and 3 (cathode)
. Figure 100-2 X-rays and CT-scan post op. controls showing electrodes position
1683
1684
100
Motor cortex stimulation for parkinson’s disease
After 12 months, in cases of bilateral implant, stimulation became bilateral (same parameters for the side omolateral to the worst clinical side). The clinical assessment before implant and at 1, 3, 6, 12, 18, and 24 months included: - UPDRS (Unified Parkinson Disease Rating Scale); - finger tapping; - walking time; - PDQL (Parkinson Disease Quality of Life Scale); - neuropsychological evaluation including MMSE (Mini Mental State Evaluation); - EEG: - oral medications and adverse events. The clinical motor evaluation was performed both in the off and in the on medication state and the motor assessment was videotaped. The ‘‘OFF’’ condition was achieved by withdrawing antiparkinsonian medications as follows: Levodopa for at least 12 h, pergolide, pramipexole, ropinirole for at least 48 h, cabergoline for at least 168 h, apomorphine for at least 3 h. The ‘‘ON’’ condition was achieved 60 min after administering a suprathreshold dose of standard L-dopa, according to daily schedule. Furthermore our patients were submitted to an extensive neuropsychological test battery including a mini mental state examination (MMSE), behavioural assessment of mood and anxiety, tests for verbal short-term memory, spatial short-term memory, episodic verbal memory, nonverbal abstract reasoning, frontal executive functions and verbal fluency. Cognitive and behavioral assessment were performed preoperatively and at 6, 12, and 18 months, in the on med status. Seven patients reached the 24 months follow up evaluation (3 unilateral and 4 bilateral MCS) A statistical (Wilcoxon’s test) significant improvement was present after 12 months of unilateral MCS, regarding total UPDRS offmed, UPDRS II, UPDRS III off-med, subscore for axial symptoms (UPDRS III: items 27–31), UPDRS IV, PDQL (> Table 100-1). The effect of unilateral MCS was bilateral, with no significant difference between the two sides. It was evident after 1–2 weeks of stimulation, and in a case of
. Table 100-1 Mean clinical improvement following MCS at 12 and 24 months Mean improvement compared to preoperative scores Global UPDRS Off Med UPDRS III Off MED Axial symptoms subscore Off Med UPDRS IV PDQL
At 12 months (%)
At 24 months (%)
20.34 13.44 22.20
15.31 17.40 23.94
32.94 15.54
22.40 23.94
accidental switching off of the stimulator, the patient became aware of something going wrong after 2–3 weeks. After 1 year of unilateral stimulation, 4 patients underwent bilateral MCS, the remaining 3 patients continued with unilateral stimulation. The statistical analysis at 24 months demonstrated a significant (Wilcoxon’s test) improvement in total UPDRS off-med (by 15.3% compared to the preoperative score), in UPDRS III off-med (by 17.4%), in the subscore for axial symptoms (by 23.9%), in UPDRS IV (by 22.4%) and in PDQL (by 13%), (> Table 100-1). A statistical analysis comparing the results of unilateral versus bilateral stimulation at 24 months was impossible because of the small samples, but in the group submitted to bilateral MCS, there was a trend toward an enhancement of the clinical effect. Notably, in all the 7 patients, the UPDRS III off med at 24 months was lower than UPDRS III off med at preoperative evaluation, as well as PDQL-39; UPDRS IV was lower in 6 cases and unchanged in 1 patient. Drug treatment could be decreased by 21%, at 24 months. Cognitive assessment in the overall group of patients showed a significant postoperative improvement on the MMSE and on tasks of episodic verbal memory; but this most likely reflects a practice effect [28]. No significant postoperative decline was observed on any cognitive task,
Motor cortex stimulation for parkinson’s disease
including those of phonological and semantic verbal fluency; on the contrary in DBS patients a significant decline on verbal fluency has been consistently reported [29]. Unilateral stimulation of the left hemisphere shows a statistical trend toward a postoperative improvement of phonological verbal fluency, along with an increase of depressive symptoms. On the contrary, stimulation of the right hemisphere shows a trend toward a decrease of depression, and no effect on verbal fluency. This reflects the different role of the two hemispheres in mood regulation [30]. No complication occurred; no adverse events, particularly no epileptic seizures nor EEG epileptic activity, were encountered. In an effort to find a metabolic indicator confirming the effect of MCS and helping in hypothesizing the mechanisms of action, we studied the functionality of the presynaptic and postsynaptic dopaminergic system. We utilized single-photon emission computed tomography (SPECT) with 1231Ioflupane (DAT-scan) to evaluate binding to the dopamine transporters before and after 6, 12, and 28 months of MCS, and IBZM-SPECT to evaluate activity of striatal postsynaptic D2 receptors, before and after 6 and 12 months of MCS. At DAT-scan at 6 and 12 months there was an increase in the activity in both putamens, particularly evident on the side omolateral to the stimulation (> Figure 100-3). Postsynaptic receptor activity was unregulated at 6 and 12 months.
Discussion Our personal data and the reports present in Literature so far suggest that MCS can modulate some symptoms of Parkinson disease, as well as other motor symptoms in movement disorders. Axial symptoms, gait, akinesia, and freezing are improved.
100
. Figure 100-3 DAT-scan: percentage changes in uptake after 6, 12 and 28 months of motor cortex stimulation (bilateral stimulation). Como: omolateral caudato; Ccontro: controlateral caudato; Pomo: omolateral putamen; Pcontro: controlateral putamen
Many old and new data suggest a strong involvement of the motor cortex in PD. The extirpation of motor cortex abolishes parkinsonian tremor [31]. Intraoperative acute stimulation of the motor cortex subthreshold for movements, relieved tremor and rigidity in a Parkinson patient [24]. rTMS (repetitive Transcranial Magnetic Stimulation) of the motor cortex improved motor performances in Parkinson disease [32–38]. Many of these studies assessed the effects of rTMS by UPDRS (Unified Parkinson Disease Rating Scale). Lefaucheur [34] used real and sham stimulation in 12 ‘‘off-med’’ patients with PD and compared the results with those obtained by a single dose of L-dopa. Real rTMS but not sham stimulation, improved motor performances. They concluded that these results support the perspective of the primary motor cortex as a possible target for neuromodulation in PD.
1685
1686
100
Motor cortex stimulation for parkinson’s disease
rTMS of the motor cortex increased dopamine release in ipsilateral putamen in healthy volunteers, as demonstrated by 11Craclopride positron emission tomography (PET) [39], while rTMS of the prefrontal cortex induced dopamine release in the ipsilateral caudate nucleus [40]; furthermore in Parkinson’s disease patients, rTMS of the motor cortex induced an increase of dopamine release in ipsilateral putamen along with a spatially enlarged area of dopamine release [41]. A bilateral overactivity of the motor cortex is present in PD [42] and it was reduced by dopaminergic treatment and by STN stimulation. Left STN stimulation reduced the abnormal overactivity in the ipsilateral primary sensorymotor cortex and in premotor cortical areas [43]. An abnormal synchronization between cortical and basal ganglia has been demonstrated in human PD using coherence analysis and movement-related frequency-specific changes in synchronization [44]. This activity was modulated by STN stimulation, which decreased the abnormal spreading of desynchronization and increased primary motor cortex activity during movement preparation and execution, with a correlated improvement in bradykinesia [45]. Animal studies suggest that the cerebral cortex plays an important role in regulating the activity of STN. Indeed, the STN represents the second entry point of cortical information to the basal ganglia [46,47]. The most extensive cortical innervation of STN originates from motor areas following a somatotopic distribution: fibers from the motor cortex hand area being located mainly in the most lateral and dorsal part of the STN [46–48]. The putamen is the principal input nucleus for somatic motor control in the basal ganglia and receives somatotopically organized cortico-striatal projections from the frontal motor areas [47,49–51]. Anterograde tracing studies in monkeys have shown that cortico-striatal fibres originating in the motor cortex project to the lateral part of
the putamen with a dorsoventral arrangement, the leg represented in the dorsal putamen, the face more ventrally,and the arm lying in between these two areas [49–51]. Strafella AP et al. in 2004 [52], demonstrated that 74.9% of neurons in dorsolateral STN respond to TMS of the ipsilateral motor cortex in humans. This response is characterized by a short-latency short-duration excitation followed by a long-lasting inhibition (more than 100 ms). They conclude that this findings ‘‘clearly indicate that the human motor cortex exerts a powerful modulatory influence over the STN’’. Drouot X et al. [53] described a functional recovery of Parkinson’s motor signs in MPTP baboons following motor cortex stimulation. High-frequency motor cortex stimulation significantly reduced akinesia and bradykinesia. The effect was present only in animals displaying moderate and severe 18F-DOPA striatal uptake depletion. The behavioral benefit was associated with an increased metabolic activity in the supplementary motor area as assessed with 18-F-deoxiglucose PET, a normalization of mean firing rate in the internal globus pallidus and the subthalamic nucleus, and a reduction of synchronized oscillatory neural activities in these two structures. In MPTP monkeys, GPi neurons showed an increased firing rate, 1 min of MCS restored normal firing rate in GPi neurons. This effect was long-lasting following the stimulation of the ipsilateral motor cortex, while the effect immediately disappeared after switching off the controlateral motor cortex stimulation. On the contrary, Wu et al. [54] were unable to demonstrate a reliable reduction in parkinsonian symptoms when MCS was delivered daily for 1 h in MPTP macaques, despite they explored a broad range of stimulation parameters. These authors observed increased activity and improved Kluver performance during the first 24 h of continuous biweekly stimulation. The effects faded after 24 h and were context specific. Finally, motor cortex stimulation has been reported to be effective not only to relieve
Motor cortex stimulation for parkinson’s disease
pain but also to improve associated movement disorders in patients with thalamic hand [55] and in patients with post-stroke movement disorders [56]. Brown JA et al. [57] reported a case of enhanced motor recovery after stroke following motor cortex stimulation. As regard to the possible mechanisms of action, the motor cortex is part of the corticobasal ganglia loop and modulation at one of the stations, will interfere with all the others. DBS influences the motor cortex activity as well as the motor cortex influences the STN, the dopamine release in the putamen, and other cortical areas involved in movement [58]. MCS is subthreshold for any movement, so we can rule out the pyramidal cells and axons as point of action. MCS is probably interfering with small inhibitory axons in the cortex itself and/or with the axons of afferents and efferents running parallel to the stimulating electrodes as postulated by a computer modeling study [59,60]. So MCS may orthodromically and/or antidromically activate fibers connecting the motor cortex to the basal ganglia or it may act at local level decreasing cortical excitability or disrupting oscillatory rhythms and abnormal patterns of activity. The clinical course of some MCS effects suggests the possibility of time consuming processes such as synaptic plasticity, long term potentiation or depression, expression of secondary messengers, polarization of brain tissue. Fibers recruitment is more likely to occur than cells activation, and it will depend on the modality of stimulation. The distance between the electrodes and the neural elements is important; a modelling study stresses the importance of CSF thickness: every 1 mm of CSF we need 6.6 V to obtain the same effect on the neural tissue [59]. Probably this problem was overestimated; in the clinical practice the straight paddle implanted over the convex surface of the dura at the hand knob, will squeeze the CSF underlying the paddle. The bipolar stimulation between two contacts of the electrode paddle corresponds to a
100
bifocal monopolar stimulation, due to the wide distance between contacts. The anode excites the fibres that run perpendicular to the electrode surface, while the cathode excites the fibres running horizontally under the paddle [59–61]. The intensity of stimulation is intuitively very important for the fibres recruitment as well as the pulse width. Frequency of stimulation is another crucial variable [9]: with frequency up to 130 Hz, the fibres are more likely to be depolarized and excited. However if the stimulation is maintained for long periods of time, some synapses may not follow the stimulus train and be blocked, with consequent inhibition. Specific frequencies may be necessary to impose specific patterns of activity, or to suppress abnormal rhythms, or in time consuming processes. The clinical effect of MCS cannot be compared with that of STNDBS due to the different inclusion criteria. DBS is usually contraindicated in the patients submitted to MCS, because of age or because of the presence of MRI anatomical abnormalities (cerebral atrophy, vascular lesions). However, STNDBS appears to be more effective on motor symptoms, but MCS seems to be more effective on axial symptoms subscore; complication rate and adverse events rate is lower for MCS; particularly, verbal fluency is not impaired by MCS; finally, our data from DAT-Scan suggest that MCS may have a protective effect on putaminal degeneration at least at short follow up. The clinical effect of MCS seems to decrease with time. This may be due to a placebo effect, and/or to the progressive nature of the disease; or it may reflect a true loss of effectiveness of the stimulation; changing the parameters of MCS may be of help in resuming the clinical effect. Changing the frequency or the intensity of stimulation according to the impedances (some fibrosis may develop between electrode surface and dura layer) may be useful. Neurostimulators delivering impulses in current may overcome this
1687
1688
100
Motor cortex stimulation for parkinson’s disease
problem. Usually the stimulation is delivered continuously. The slight decline in the clinical benefit may be due to a sort of habituation of the cortex. To change type of stimulation may restore the effect. If this is true, alternate stimulation (right side or left side) may be a solution. The clinical effect is long-lasting, therefore a cyclic stimulation (only during daytime or 30 min on and 2–3 h off) may be proposed. Not all the patients respond to MCS: this may be due to the rather large inclusion criteria, or to the different electrode position and different stimulation parameters. The effect of rTMS may be predictive of the clinical outcome following MCS. However, the frequency and the duration of rTMS certainly differs from that used for MCS. As regards the electrode position and stimulation parameters the number of patients treated with MCS is still too low to allow statistical correlation analysis to identify prognostic factors. Different surgical techniques may be used to place the epidural cortical electrode: general versus local anaesthesia, burr hole versus craniotomy, craniometer landmarks versus neuronavigation with MRI or with fMRI. No matter the technique used, we believe that a neurophysiological precise location is mandatory if we want to exactly know where our electrode is. Our methodology allows motor mapping under general anaesthesia with a very low incidence of epileptic seizures (4–5% compared to 20–25% of the so called Penfield’s technique). The need for a bilateral implant has still to be demonstrated. In our experience unilateral MCS improves motor performances bilaterally, but bilateral stimulation seems to increase such an improvement. According to the data reported so far, we suggest the use of MCS in PD patients with prominent axial symptoms, gait disturbances, and therapy complications. It has to be stressed that this neuromodulation procedure should be taken into consideration in other movement disorders such as pure akinesia. Only a controlled randomized
study will bring stronger evidence on the clinical usefulness of MCS in Parkinson’s disease and in other movement disorders [62].
References 1. Tsubokawa T, Katayama Y, Yamamoto T, Hirayama T, Koyama S. Chronic motor cortex stimulation for the treatment of central pain. Acta Neurochir. 1991;52:137-9. 2. Arle JE, Apetauerova D, Zani J, Deletis V, Penny D, Hoit D, Gould C, Shils JL. Motor cortex stimulation for Parkinson disease: 12 months follow up in 4 patients. J Neurosurg. 2008;109(1):133-9. 3. Bentivoglio AR, Cavallo MA, Cioni B, Contarino F, Eleopra R, Lavano A, et al. Motor cortex stimulation for movement disorders In: Meglio, editors. Proceedings of the 14th Meeting of the Worls Society for Stereotactic and Functional Neurosurgery. Bologna; Medimond, 2005. p. 89-97. 4. Benvenuti E, Cecchi F, Colombini A, Gori G. Extradural motor cortex stimulation as a method to treat advanced Parkinson’s disease: new perspectives in geriatric medicine. Aging Clin Exp Res. 2006;18:347-8. 5. Canavero S, Paolotti R. Extradural motor cortex stimulation for advanced Parkinson’s disease. Mov Disord. 2000;15:169-71. 6. Canavero S, Paolotti R, Bonincalzi V, Castellano G, Greco-Crasto S, Rizzo L, et al. Extradural motor cortex stimulation for advanced Parkinson’s disease: report of two cases. J Neurosurg. 2002;97:1208-11. 7. Cilia R, Landi A, Vergari F, Sganzerla E, Pezzotti G, Antonini A. Extradural motor cortex stimulation in Parkinson’s disease. Mov Disord. 2007;22:111-4. 8. Cioni B. Motor cortex stimulation for Parkinson’s disease. In: Sakas D, Simpson B, Krames E, editors. Operative neuromodulation. Springer; Acta Neurochir. 2007;Suppl 97/2:233-8. 9. Fasano A, Piano C, De Simone C, Cioni B, Meglio M, Bentivoglio AR. High frequency extradural motor cortex stimulation dramatically improves axial symptoms in a patient with Parkinson’s disease. Mov Disord, 2008; 23(13):1916-19. 10. Kleiner-Fisman G, Fisman DN, Kahn FI, Sime E, Lozano A, Lang AE. Motor cortical stimulation for Parkinsonism in Multiple Systemic Atrophy. Arch Neurol 2003;60:1554-1558. 11. Pagni CA, Altibrandi MG, Bentivoglio AR, Caruso G, Cioni B, Contarino F, et al. Extradural motor cortex stimulation for Parkinsin’s disease: History and first results by the study group of the Italian Neurosurgical Society. Acta Neurochir. 2005;93 Suppl: 113-9.
Motor cortex stimulation for parkinson’s disease
12. Pagni CA, Zeme S, Zenga F, Maina R. Extradural motor cortex stimulation in advanced Parkinson’s disease: the Turin experience. Neurosurg 2005;57 (ONS Suppl 3): ONS-402. 13. Pagni CA, Zeme S, Zenga F, Maina R, Mastropietro A, Papurello D. Further experience with extradural motor cortex stimulation for treatment of advanced Parkinson’s diseases. Report of 3 new cases J Neurosurg Sci. 2003;47:189-93. 14. Canavero S, Bonincalzi V. Extradural cortical stimulation for movement disorders. In: Sakas D , Simpson B, Krames E, editors. Operative Neuromodulation. Springer: Acta Neurochir 2007;Suppl 97/2:223–32. 15. Yousry TA, Schmid UD, Alkadhi H, Schmidt D, Peraud A, Buettner A, et al. Localization of the motor hand area to a knob on the precentral gyrus. A new landmark. Brain. 1997;120:141-57. 16. Gharabaghi A, Hellwig D, Rosahl SK, Shahidi R, Schrader C, Freund HJ, et al. Volumetric image guidance for motor cortex stimulation: integration of threedimensional cortical anatomy and functional imaging. Neurosurgery 2005;57 Suppl 1:114-20. 17. Mogilner AY, Rezai AR. Epidural motor cortex stimulation with functional imaging guidance Neurosurg Focus 2001;11:article 4. 18. Pirotte B, Voordecker P, Neugroschl C, Baleriaux D, Wikler D, Metens T, et al. Combination of functional magnetic resonance imaging-guided neuronavigation and intraoperative cortical brain mapping improves targeting of motor cortex stimulation in neuropathic pain. Neurosurgery 2005;56 Suppl 2:344-59. 19. Kamada K, Sawamura Y, Takeuchi F, Kawaguchi H, Kuriki S, Todo T, et al. Functional identification of the primary motor area by corticospinal tractography. Neurosurgery 2005;56 Suppl 1:98-109. 20. Nguyen JP, Lefaucheur JP, Decq P, Uchiyama T, Carpentier A, Fontaine D, et al. Chronic motor cortex stimulation in the treatment of central and neuropathic pain. Correlation between clinical, electrophysiological and anatomical data. Pain. 1999;82:245-51. 21. Nguyen JP, Lefaucheur JP, Le Guerinel C, Eizenbaum JF, Nakano N, Carpentier A, et al. Motor cortex stimulation in the treatment of central and neuropathic pain. Arch Med Res. 2000;31:263-265. 22. McCarthy G, Allison T, Spencer DD. Localization of the face area of human sensorimotor cortex by intracranial recording of somatosensory evoked potentials. J Neurosurg 1993;79(6):874-84. 23. Maegaki Y, Najm I, Terada K, Morris HH, Bingaman WE, Kohaya N, et al. Somatosensory evoked highfrequency oscillations recorded directly from the human cerebral cortex. Clin Neurophysiol 2000;111:1916-26. 24. Woolsey CN, Erickson T, Gilson WE. Localization in somatic sensory and motor areas of human cerebral cortex as determined by direct recording of evoked potentials and electrical stimulation. J Neurosurg. 1979;51:476-506.
100
25. Penfield W, Boldrey E. Somatic motor and sensory representation in the cerebral cortex of man as studied by electrical stimulation. Brain. 1937;60:389-443. 26. Beinsteiner R, Windischberger C, Lanzenberger R, Edward W, Cunnington R, Erdler M. Finger somatotopy in human motor cortex. NeuroImage. 2001;13:1016-26. 27. Thickbroom GW, Byrnes ML, Walters S, Stell R, Mastaglia FL. Motor cortex reorganisation in Parkinson’s disease. J Clin Neurosci. 2006;13(6):639-42. 28. Daniele A, Albanese A, Contarino MF, Zinzi P, Barbier A, Gasparini F, Romito A, et al. Cognitive and behavioural effects of chronic stimulation of the subthalamic nucleus in patients with Parkinson’s disease. J Neurol Neurosurg Psychiatry. 2003;74:175-82. 29. Contarino MF, Daniele A, Sibilia AH, Romito L, Bentivoglio AR, Gainotti G, et al. Cognitive outcome 5 years after bilateral chronic stimulation of the subthalamic nucleus in patients with Parkinson’s disease. J Neurol Neurosurg Psychiatry. 2007;78:248-52. 30. Rotenberg VS. The peculiarity of the right hemisphere function in depression: solving the paradoxes. Prog Neuropsychopharmacol Biol Psychiatry. 2004;28:1-13. 31. Bucy PC. Special article: The neural mechanisms of athetosis and tremor. Ann Surg. 1945;122:943-54. 32. Ikeguci M, Touge T, Nashiyama Y, Takeuchi H, Kurijama S, Ohkawa M. Effects of successive repetitive transcranial magnetic stimulation on motor performances and brain perfusion in idiopathic Parkinson’s disease J Neurol Sci 2003;209:41-6. 33. Khedr EM, Farweez HM, Islam H. Therapeutic effect of repetitive transcranial magnetic stimulation on motor function in Parkinson’s disease patients. Eur J Neurol 2003;10:567-572. 34. Lefaucheur JP, Drouot X, Von Raison F, MenardLefaucheur I, Cesaro P, Nguyen JP. Improvement of motor performances and modulation of cortical excitability by repetitive transcranial magnetic stimulation of the motor cortex in Parkinson’s disease. Clin Neurophysiol 2004;115:2530-2541. 35. Mally J, Stone TW. Improvement in Parkinsonian symptoms after repetitive transcranial magnetic stimulation. J Neurol Sci. 1999;162:179-84. 36. Pascual-Leon A, Valls-Sole J, Brasil-Neto JP, Cammarota A, Grafman J, Hallett M. Akinesia in Parkinson’s disease. Effects of subthreshold ripetitive transcranial megnetic stimulation of the motor cortex. Neurology 1994;44:892-8. 37. Shimamoto H, Takasaki K, Shigemori M, Imaizumi T, Ayabe M, Shoji H. Therapeutic effect and mechanism of repetitive transcranial magnetic stimulation in Parkinson’s disease J Neurol 2001;248 Suppl 3:48-52. 38. Siebner HR, Mentschel C, Auer C, Conrad B. Repetitive transcranial stimulation has a beneficial effect on bradykinesia in Parkinson’s disease. Neuroreport 1999;10:589-94.
1689
1690
100
Motor cortex stimulation for parkinson’s disease
40. Strafella AP, Paus T, Barrett J, Dagher A. Repetitive transcranial magnetic stimulation of the human prefrontal cortex induces dopamine release in the caudate nucleus. J Neurosci. 2001;21:RC157(1–4). 41. Strafella AP, Ko JH, Grant J, Fracaccio M, Monchi O. Corticostriatal functional interactions in Parkinson’s disease: a rTMS/11Craclopride PET study. Eur J Neurosci. 2005;22:2946-52. 42. Ridding MC, Inzelberg R, Rothwell JC. Changes in excitability of motor cortical circuitry in patients with Parkinson’s disease. Ann Neurol. 1995;37:181-8. 43. Payoux F, Remy P, Damier P, Miloudi M, Loubinoux J, Pidoux B, et al. Subthalamic nucleus stimulation reduces abnormal motor cortical overactivity in Parkinson’s disease. Arch Neurol. 2004;61:1307-13. 44. Brown P. Oscillatory nature of human basal ganglia activity. Relationship to the pathophysiology of Parkinson’s disease. Mov Disord. 2003;18:357-63. 45. Devos D, Labyt E, Derambure P, Bourriez JL, Cassim F, Reyns N, et al. Subthalamic nucleus stimulation modulates motor cortex oscillatory activity in Parkinson’s disease. Brain 2004;127:408-19. 46. Nambu A, Tokuno H, Hamada I, Kita H, Imanishi M, Akazawa T, et al. Excitatory cortical inputs to pallidal neurons via the subthalamic nucleus in the monkey. J Neurophysiol 2000;84:289-300. 47. Parent A, Hazrati LN. Functional anatomy of the basal ganglia. The place of subthalamic nucleus and external pallidum in basal ganglia circuitry. Brain Res Rev. 1995;20:128-54. 48. Afsharpour S. Topographical projection of the cerebral cortex to the subthalamic nucleus. J Comp Neurol. 1985;236:14-28. 49. Jones EG, Coulter JD, Burton H, Porter R. Cells of origin and terminal distribution of corticostriatal fibers arising in the sensory-motor cortex of monkeys J Comp Neurol 1977;173:53-80. 50. Takada M, Tokuno H, Nambu A, Irase M. Corticostriatal projections from the somatic motor areas of the frontal cortex in the macaque monkey: segregation versus overlap of input zones from the primary motor cortex, the supplementary motor area, and the premotor cortex. Exp Brain Res. 1998;120:114-28.
51. Tokuno H, Irase M, Nambu A, Akazawa T, Miyachi T, Takada M. Corticostriatal projections from distal and proximal forelimb representations of the monkey primary motor cortex Neurosci Lett. 1999;269:33-6. 52. Strafella AP, Vanderwerf Y, Sadikot AF. Transcranial magnetic stimulation of the human motor cortex influences the neural activity of subthalamic nucleus. Eur J Neurosci 2004;20:2245-9. 53. Drouot X, Oshino S, Jarraya B, Besret L, Kishima H, Dauquet J, et al. Functional recovery in a primate model of Parkinson’s disease following motor cortex stimulation. Neuron 2004;44:769-78. 54. Wu AK, McCairn KW, Zada G, Wu T, Turner RS. Motor cortex stimulation: mild transient benefit in a primate model of Parkinson’s disease. J Neurosurg. 2007;106:695-700. 55. Franzini A, Ferroli P, servello D, Broggi G. Reversal of thalamic hand syndrome by long term motor cortex stimulation. J Neurosurg 2000;94:873-5. 56. Katayama Y, Oshima H, Fukaya C, Kawamata T, Yamamoto T. Control of post-stroke movement disorders using chronic motor cortex stimulation. Acta Neurochir 2002;79:Suppl 89-92. 57. Brown JA, Lutsep H, Cramer SC, Weinand M. Motor cortex stimulation for enhancement of recovery after stroke: case report. Neurol Res. 2003;25:815-18. 58. Priori A, Lefaucheur JP. Chronic epidural motor cortical stimulation for movement disorders Lancet Neurol 2007;6:279-86. 59. Manola L, Roelofsen BH, Holsheimer J, Marani E, Geelen J. Med Biol Eng Comput. 2005;43:335-43. 60. Manola L. Holsheimer J, Veltink P, Buitenweg JR. Anodal vs cathodal stimulation of motor cortex: a modeling study. Clin Neurophysiol. 2007;118:464-74. 61. Patton HD, Amassian VE. Single and multiple unit analysis of cortical stage of pyramidal tract activation. J Neurophysiol. 1954;17:345-63. 62. Ohnishi T, Hayashi T, Okabe S, Nonaka K, Matsuda H, Imabayashi E, et al. Endogenous dopamine release induced by repetitive transcranial magnetic stimulation over the primary motor cortex: an (11C)raclopride positron emission tomography study in anesthetized macaque monkeys Biol Psychiatry. 2004;55:484-9.
99 Other Targets to Treat Parkinson’s Disease (Posterior Subthalamic Targets and Motor Cortex) F. Velasco . S. Palfi . F. Jime´nez . J. D. Carrillo-Ruiz . G. Castro . Y. Keravel
Introduction In 1817, James Parkinson made noted in cases of shaking palsy, “the abolition of tremor with the onset on hemiplegia secondary to capsular hemorrhage”. He also quoted that tremor and muscular tension decreased or disappeared during sleep [1]. This lead was not followed by surgeons for many years. In regard to participation of cortical motor areas in the physiopathology of tremor, Aring and Fulton reported in 1936 that the experimental tremor, induced by cerebellar lesions in monkeys, could be stopped by ablations of cortical areas 4 and 6 [2]. In 1945, Bucy relieved tremor with “little associated motor impairement” by extirpation of area 4 alone [3]. Few years later, Walker proposed to interrupt pyramidal fibers in the lateral part of the cerebral peduncle to avoid speech impairement and decrease the likehood of epilepsy that may complicate cortical excisions. The “cerebral pedunculotomy” resulted in the improvement of tremor, rigidity and hemibalism [4]. These experiences, together with the failed attempts to control involuntary movements by sectioning the dorsal spinal roots [5], or interrupting ascending spinal cord tracts [6,7], gave support to the idea of a “central pacemaker” in the physiopathology of tremor and other motor disturbances. The clinical expression of involuntary movements would be mediated through pyramidal and extrapyramidal motor cortices. #
Springer-Verlag Berlin/Heidelberg 2009
The report, made by Meyers that tremor, rigidity and festination could be improved by lesioning the caudate nucleus and the afferent and efferent fibers to the basal ganglia (“pallidofugal fibers”) without causing motor deficit [8] attracted the surgical efforts towards the basal ganglia and away from the pyramidal and extrapyramidal tracts. On the other hand, the participation of mesencephalic structures in the genesis of involuntary movements had been suspected for long time on the basis of pathological findings in autopsy cases of idiopathic and postencephalitic Parkinson?s disease (PD) [9]. In 1948, Ward et al. [10] reported that experimental lesions in the ventromedial mesencephalon of rhesus monkeys resulted in “alternating tremor at rest” contralateral to lesions. Monkeys also presented hypertonia and fixed posture in flexion of the trembling extremities. Ventromedial mesencephalic lesions in monkeys were extensively used as an experimental model of PD. However, during a debate on this modelof tremor, Mettler [11] broughtthe attention to the fact that only one-third of the lesioned monkeys developed tremor. The remaining twothirds not only failed in doing so, but the systemic administration of oxotremorine, used to induced tremor in intact animals, was ineffective. Those “nontrembling” monkeys had lesions placed but few millimeters dorsal and lateral to the intended target. The interpretation provided by Dr. Jasper
1666
99
Other targets to treat parkinson’s disease (posterior subthalamic targets and motor cortex)
in this debate was that, within the mesencephalic tegmentun, they were two contiguous antagonic systems. One prevented the development of tremor and rigidity placed in the pars compacta of substantia nigra (SNc) and the other participated in tremorogenesis placed in the dorsal tegmentun. To test that hypothesis, and taking into account the observation of James Parkinson in that tremor disappears during sleep, we performed two consecutive radiofrequency lesions in the mesencephalic tegmentum in a group of rhesus monkeys. A first lesion aiming SNc (A 7.0 to 9.0; L 2.5; H –2.0) to induce contralateral tremor and rigidity and a second one to suppress those symptoms aiming the mesencephalic reticular formation (MRF) (A 5.0 to 7.0; L 3.0; H + 2,0
to + 3,0). Guided by multi-unit activity recordings and microstimulation, lesions could be precisely placed as intended (> Figure 99-1). All monkeys developed tremor at rest and rigidity in flexion of contralateral extremities with lesions restricted to SNc and fibers immediatly dorsal. The second lesion in the dorsal tegmentum abolished tremor and induced hypotonia. Even with small lesions (average 3.2 mm), they involved MRF, zona incerta (Zi), central tegmental tract (Ttc), middle longitudinal fasciculus (MLF) and part of medial lemniscus (Lm) and brachium conjunctivum (Bcj) [12]. Although these experiments confirmed the hypothesis of mesencephalic structures involved in tremor production and suppression, they failed to elucidate which specific lesioned structure was responsible for tremor arrest.
. Figure 99-1 Electrophysiological exploration of mesencephalic region of rhesus monkey, performed with multiunitary recording (right) and microstimulation through the electrode cannulae (left). Lesions to induce tremor were centered in the area with large units on a densely packed background, corresponding to substantia nigra pars compacta (SNpc), immediately above the fibers of the cerebral peduncle (Cped). Microstimulation in SNpc and Cped induced eyelid retraction and internal rotation of ipsilateral eye. Recording at the nucleus reticularis mesencephalic (NRM or MRF) in the animals under light barbiturate, anesthesia had only background activity; however, awaking the animal with a noxious stimulus induced firing of middle size spikes. High-frequency stimulation of NRM awakened the animal. Lesions in this point suppressed contralateral tremor. Other abbreviations: IN, substantia inominata between SNpc and red nucleus (RN); ML, midline thalamic nuclei; PAG, periaqueductal gray; SNpd, substantia nigra pars reticulata. (From Velasco et al. Exp Neurol 64:516–527, 1979) [12]
Other targets to treat parkinson’s disease (posterior subthalamic targets and motor cortex)
The Posterior Subthalamic Target During the 50s, stereotactic lesions to treat movement disorders and PD in particular were performed on the globus pallidum (Gp) [13] and ventrolateral thalamus [14]. It became evident that in order to obtain complete and permanent effect on symptoms, lesions should include the whole nucleus, which often derived in “excessively large lesions that increased the risk of complications” [15]. Therefore, lesions became oriented to destroy fiber connections of basal ganglia in areas where they become compacted. Spiegel and Wycis, in 1962, reported their experience with anterior subthalamic lesions directed to Forel’s field H2; to interrupt pallidal connections with VL and red nucleus (Ru), as well as the thalamic and cortical fibers projecting to Ru. Although these lesions, called “campotomy”, were intended to preserve Zi, they invariably included Zi medial to the subthalamic nucleus of Luys (STN) [15,16,17]. Later on, Story et al, moved the subthalamic target 7 mm behind mid-commissural point (MCP) with improvement of tremor and rigidity in 62% of their cases [18]. Andy et al. in 1963, reported that lesions immediately lateral to Ru-controlled rest and intention tremor [19]. Although the authors speculated that lesions were successful because they interrupted output in the Bcj, this seems unlikely since midbrain tegmental stimulation elicits tremor in the presence of Bcj degeneration [20]. Besides, lesions or high-frequency electrical stimulation of Ru and Bcj induce dizziness, instability and dysmetria [21] that were not reported as complications in Andy’s work. In 1965, Mundinger made note that small lesions in the basal part of VL (Voa-Vop subnucleus), which most likely interrupted its afferents, were sufficient to produce the best improvement in PD symptoms [15]. He used an electrode for coagulation, which protruded perpendicular to the carring cannula for a distance of 7.0 mm,
99
and induced lesions 1.5–2.5 mm around the electrode. The lesion intended to destroy Zi dorsal and posterior to STN as well as prelemniscal radiations (Raprl) and the rostro-latero-dorsal Ru. This author also related improvement to the destruction of dentate thalamic connections. It is important to remark that all these subthalamic targets were designed to preserve STN, in the belief, that lesions of this nucleus were responsible for hemibalismus seen in vascular lesions of the subthalamic region. Recently, STN lesions have been used to treat PD without inducing hemibalismus [22]. In 1969, Bertrand et al. quoted that the simple introduction of a 1.5-mm diameter electrode in the posterior subthalamic region arrested contralateral tremor [23]. The tip of electrodes, at the moment this effect occurred, was invariably below the AC–PC line as evidenced in the ventriculograms. Neither the arrest of tremor was accompanied by any motor or sensory deficits nor low amplitude 60 Hz electrical stimulation through the electrode induced them. Subsequent studies defined the stereotactic location and extension of this exquisite subthalamic area to arrest tremor. In order to standardize the stereotactic coordinates, the length of the anterior–posterior commissure (AC–PC) line was divided in tenths, using the resulting units to measure the distances below the AC–PC line and at the side of the midsagittal plane in the ventriculograms. In the initial analysis, we noticed that the target volume did not exceed 1/10 in “x”, 2/10 in “y” and 3/10 in “z” coordinates (about 2.5 5.0 7.5 mm3) and was located 5/10 lateral to midsagittal plane, 8/10 behind AC and 2/10 below AC–PC line [24]. The target area was definitely located in the posterior third of the subthalamus. Thereafter, these stereotactic coordinates were used to target subsequent cases. A larger number of cases were used to perform a statistical outline of the target volume through confidence intervals. The resultant volume allows to predict with 98% probability the
1667
1668
99
Other targets to treat parkinson’s disease (posterior subthalamic targets and motor cortex)
place where insertion of electrodes will arrest tremor in futures cases (> Figure 99-2) [25]. Lesions in this target have been used in over 700 cases for the past 37 years and confirmed the reliability of the statistical outline to determine its location. It is important to remark that the effect occurred in some cases when electrodes were as low as 12 mm below the AC–PC line and therefore already in the mesencephalic tegmentum. In our experience, limitation of lesioning the posterior subthalamic target were that it might induce a transient or even permanent neglect of contralateral extremities in cases with advanced PD. Besides, bilateral lesions also increased the state of bradykinesia [26]. Therefore, for many years indications were restricted to cases with mainly unilateral tremor. Rigidity was treated
with additional lesioning in VL and bradykinesia was considered a contraindication. These limitations do not apply in cases treated by electrical stimulation in the same target. Other authors had validated the results of lesioning this optimum area to treat mainly tremor [27,28].
Electrical Stimulation of the Posterior Subthalamic Target When deep brain stimulation (DBS) was accepted as a minimal invasive technique, which has low morbidity and mortality, and undesirable effects may be reversed by adjusting stimulation parameters, we decided to apply the technique to the posterior subthalamic target. Initial indications
. Figure 99-2 Standardization of thalamic and subthalamic areas made from AC–PC distance. Drawings represent lateral (left) and AP (right) views of ventriculogram. AC–PC distance has been divided in 1/10 and resultant units used to divide the areas above and below AC–PC level and at the side of the midline. In this frame, the place, where the tip of the electrodes introduced through a frontal parasagittal burr hole suppressed or decreased tremor, has been plotted. Electrodes tip position was seen in lateral and AP radiograms, and black dots indicate those electrodes that suppressed contralateral tremor temporarly and clear dots indicate those electrodes that significantly decreased tremor. Outside this area, none of the additional 31 trajectories induced significant changes in tremor amplitude. Notice that the critical area is slender in the lateral direction (“x” coordinate), somehow larger in the AP direction, particularly immediately below AC–PC line, (“y” coordinate) and definitely larger in ventrodorsal direction (“z” coordinate), (From Velasco et al. Appl Neurophysiol. 38:38–46, 1975) [25]
Other targets to treat parkinson’s disease (posterior subthalamic targets and motor cortex)
were the same as for lesioning, i.e. mainly unilateral disease, with predominant tremor and rigidity and minimal degree of bradykinesia (Hoeh-Yahr II-III). Electrical stimulation of the fibers, surrounding Ru and posterior to STN as seen in the MRI, resulted in significant decrease of contralateral tremor and rigidity (p < 0.01) without increase in bradykinesia [29]. Since bradykinesia was not increased, we decided to explore the effect of the posterior subthalamic target in advanced PD cases (Hoeh-Yahr V) presenting severe tremor, rigidity and bradykinesia, as well as imbalance and gait problems. Those patients were evaluated through UPDRS, and had an off-medication average score in UPDRS part III of 63. Bilateral stimulation induced significant decrease in tremor and rigidity (p < 0.001) as
99
seen in unilateral cases, and to a lesser extend in bradykinesia (p < 0.01) imbalance and festination. Stimulation was always bipolar between adjacent contacts with a center-to-center distance of 3 mm. Again, MRI studies showed the cathode located between Ru and STN, medial to Zi, in the area labeled in the stereotaxic atlas as Raprl [29,30,31] (> Figure 99-3). Other authors have reported similar experiences stimulating the “posterior subthalamic white matter” in PD cases with dominant tremor [32,33], as well as tremor of other ethiologies [34]. However, there is little information in regard to the effects on other PD symptoms. The stereotactic coordinates used in those reports were similar to the ones we have used. On the other hand, some reports on electrical stimulation of STN have found the most
. Figure 99-3 MRI axial sections in cases with unilateral (right) and bilateral (left) implantation of DBS-Raprl; MRI sections were taken at the level of the contact acting as cathode. Notice that the electrodes are placed immediately lateral to red nucleus, posterior to subthalamic nucleus and most likely medial to caudal zona incerta
1669
1670
99
Other targets to treat parkinson’s disease (posterior subthalamic targets and motor cortex)
effective contact to relief PD symptoms immediately dorsal and medial to STN in the area of rostral Zi (rZi) [35,36,37]. These observations called the attention to Zi as a possible target to treat PD. Plaha et al. in 2006, published a wellelaborated report that compares the efficacy of electrical stimulation in STN with stimulation of rZi and caudal Zi (cZi). In their experience, stimulation of cZi improves PD symptoms, in particular, tremor and rigidity to a higher degree than STN; cZi target would be 2–3 lateral and 1–2 mm above Raprl [38]. > Figure 99-4 compares the location of electrodes position in cZi reported by Plaha with 36 electrodes positioned in Raprl in our cases. To plot the position of electrodes in Raprl, we used a fusion-imaging technique between axial CT-MRI and horizontal –3.5 section of Schaltenbrand and Wahren atlas [39]. For the moment, it is not possible to make a statement on the possible advantages or disadvantages of the posterior subthalamic target over STN in the treatment of PD, mainly because experience with the former is still restricted to a small number of cases as compared to STN. However, it may be that posterior subthalamic target is the treatment of choice for those cases with prominent tremor. This brings forth the question “is there a universal target to treat all PD cases, or if treatment should be tailored with the choice of more than one target according to patient’s symptom severity”. > Table 99-1 has been elaborated with the percentages of improvement of different PD symptoms, induced by DBS in different targets. Data was collected from reports that evaluated patients within the same follow-up period and using the same international scale (UPDRS part III).
. Figure 99-4 Plotting of electrodes position on the horizontal section 3.5 mm below AC–PC plane (Hv-3.5) of the Schaltenbrand and Wahren stereotactic atlas. Electrodes in Raprl (yellow circles) were plotted using a CT-MRI-atlas imaging fusion technique (Praezis Plus, Tamed Germany). Electrodes in caudal zona incerta (cZi) (red crosses) were copied from a diagram published by Plaha et al (Figure 4, Brain, 129:1732–1747, 2006) [38]. The two groups of electrodes are in different structures. Abbreviations: Cos, Superior colliculus; Bcoi, Brachium conjuntivum; Aaq, Periaqueductal gray; Apbga, Area parabigeminalis anterior; Lm, Lemniscus medius; Ll, Lemniscus lateralis; Puig, Pulvinar; Gl, Lateral genuculate; Ppd, Peripeduncular nucleus; Zi, Zona incerta; Ru, Red Nucleus; FM, Meynert Fasciculus; Sth, Subthalamic Nucleus; Cpip, posterior branch internal capsule; H2, Forel’s field H2; Tmth, Mamilothalamic tract; Hpth, Hypothalamus; Vet, Third ventricle; Fx: Fornix; pdthif, Pedunculus thalami inferior; Pmi, Pallidus medial internus; Lapi, Lamina pallidal interna; Pme, Pallidus medial externus; Lapm, Medial pallidal lamina
Surgical Clues
X-Ray films. Medication was discontinued the night before surgery and under local anesthesia, the head was positional in the stereotactic frame. A burr hole centered at 15 mm from the midline
For many years, the “optimum” target was approached using ventriculography and biplanar
Other targets to treat parkinson’s disease (posterior subthalamic targets and motor cortex)
99
. Table 99-1 Percent improvement of different PD symptoms treated by DBS of different targets. Data was obtained from reports that evaluated improvement through UPDRS part III scores from 6 to 12 months follow-up [31,38,42,43,44]. Abbreviations: STN, subthalamic nucleus or corpus Luysii; rZi, rostral zona incerta; cZi, caudal zona incerta; Raprl, prelemniscal radiations; Gpi, globus pallidus internus; Vim, ventralis intermidius thalami; NA, not available. Notice that contiguous targets such as STN and rZi or Raprl and cZi seem to have similar effects on tremor, rigidity and bradykinesia. Posterior subthalamic target is more efficient to treat tremor and rigidity Target STN rZi cZi Raprl Vim Gpi
Tremor (%) 80 86 93 92 80 80
Rigidity (%)
Bradykinesia (%)
Gait (%)
Posture (%)
Author
65 52 76 94 None 41
51 56 65 65 None 41
55 NA NA 50 NA 32
53 NA NA 35 NA 26
Benabid et al Plaha et al Plaha et al CarriIIoRuiz et al Benabid et al Loher et al
over the coronal suture was performed. The ventricular needle was directed posteriorly taking as reference the external auditory meatus. This very important maneuver allowed bringing the needle tip behind the foramen of Monro and therefore, the injected air invariably filled the 3rd ventricule. Twenty to 30 cc of air were gently exchanged by 15 to 20 cc of CSF, and the lateral X-Ray film was taken as the surgeon injected the final 2 cc of air. The AC and PC were clearly demonstrated and in almost all cases injected air delineated also Sylvian aqueduct and the 4th ventricule. AC–PC distance is measured and the target calculated through a simple procedure of standardization: AC PC length 10 8 x 8 ¼}y} in mm ðbehind AC in > > > > > lateral X-Ray filmÞ > > > > < x 1 2 ¼ }z} in mm ðbelow AC-PC > line in lateral filmÞ > > > > > x 5 ¼ }x} in mm ðfrom the middle of > > > : 3rd ventricule in AP filmÞ For example, in a case with an AC–PC distance of 25 mm the coordinates would be estimated as follows.
25 ¼ 2:5 8 ¼ 20 mm behind AC 10 25 ¼ 2:5 1 or 2 ¼ 2:5 to 5:0 mm z¼ 10 below AC-PC level 25 ¼ 2:5 5 ¼ 12:5 lateral to x¼ 10 mid 3rd verticule plane
y¼
The insertion of a leukotome or radiofrequency electrode and even a DBS will decrease or arrest contralateral tremor temporarily [26,29]. More recently, we have positioned DBS tetrapolar electrodes or radiofrequency lesioning electrodes using CT-MRI imaging fusion corroborated with the fusion of anatomical horizontal sections of stereotactic atlas. We added microelectrode unitary recording, either with a single electrode or in an array of 5 electrodes (BenGun); microstimulation through the cannula of microelectrodes and bipolar macrostimulation through DBS contacts or the tip of radiofrequency electrode. We have also used somatosensory evoked potentials (SEP) stimulating contralateral median nerve and recording through each DBS contact. These additional tests have been performed to define the anatomical boundaries of the target and their physiological enviroment (see below).
1671
1672
99
Other targets to treat parkinson’s disease (posterior subthalamic targets and motor cortex)
Indirect CT-MRI targeting would be similar to that described for ventriculography. Direct targeting would be the white matter lateral to the anterior half of Ru, posterior to STN and medial to Zi. For lesioning, the target may be approached in an oblique trajectory in the frontal plane to avoid the lateral ventricule wall. However, for DBS electrode placement it is better to stay parallel to the sagittal plane to have two contacts in Raprl. Inclination in the frontal plane would bring one of the contacts too medial and close to III cranial nerve or too lateral and close to pyramidal fibers in the internal capsule. It is also important to place the electrode with an angle of 45 to 60 in regard to AC–PC line, which is the direction of Raprl. Angle less than 45 will bring the lowest contact close or within the medial lemniscus, which will restrict the range of stimulation. Lesioning is performed by radiofrequency at 80 C during 1 min. Additional 1 min heating could be necessary. DBS typically requires pulse amplitude of 2.5 to 3.5 V, 130 Hz and 180–210 ms.
Anatomic-physiologic Correlations of Posterior Subthalamic Target Subthalamus is densely packed with fibers and nuclei within a few millimeter area. This triangular-shaped region with an anterior vertex lays lateral to the hypothalamus and periaqueductal gray, posteromedial to the internal capsule and anterior to the tegmental area (parabigemina anterior in the Schaltenbrand and Wahren atlas) and the medial geniculate body. While its upper limit is clearly defined by the thalamus, it extends without clear-cut boundaries towards the mesencephalon. Subthalamic nuclei include the STN or corpus Luysi, Zi and upper part of Ru. A nucleus, named “substantia Q of Sano”, extends between Zi, SNr and peripeduncular nucleus in the upper mesencephalon. Zi extends rostrally (rZi) above and
medial to STN and caudally (cZi) behind STN and around Raprl and medial lemniscus (Lm). Subthalamic fibers may be classified as follows: 1. Fibers interconnecting basal ganglia such as ansa lenticularis and prerubral area, thalamic fasciculus, lenticular fasciculus, (Forel’s fields H, H1, H2) and subthalamic fasciculus. They traverse the anterior part of the subthalamus, medial to STN. 2. Ascending fibers from the cerebellum (Bcj) and Ru that end in the thalamus; most of them traversing Ru and some placed immediately lateral to this nucleus within Raprl (peri-rubral fibers). 3. Ascending tracts of specific sensory systems such as Lm, lemniscus lateralis (LL) and spino-thalamic tract placed in the posterior part of the Subthalamus. 4. Fibers connecting nonspecific structures that include Tractus tegmentales centralis (Ttc) and Raprl and project to intralaminar and ventrolateral thalamic nuclei. 5. Descending cortical fibers originated mainly from motor, prefrontal and cingulated cortices, and end in all subthalamic nuclei. 6. Multiple fibers originated in the subthalamic nuclei and hypothalamus (Meynert’s Retroreflex Bundle) that interconnect those nuclei among them, with other basal ganglia, brain stem and spinal nuclei [15,39,45,46,47,48]. Therefore, even small lesions in this area most likely interfere with more than one of these nuclei and fiber bundles. Moreover, recent reports that study the volume of tissue that DBS in STN involves, have demonstrated that current densities within therapeutic ranges extend to contiguous fibers and nucleus, and induce complex patterns of activation [49]. So, claiming that therapeutic effects derive from activation or deactivation of a single structure and even more of a specific segment of these structures seems for the moment inappropriate. Still, spreading of the current to contiguous structures may not necessarily induce efficient excitation or inhibition of all structures involved in the total volume of tissue [50]. On the other hand, experience has proven that precise localization of discrete lesions
Other targets to treat parkinson’s disease (posterior subthalamic targets and motor cortex)
and DBS electrodes is essential to obtain the optimum control of symptoms and avoid undesirable collateral effects. Therefore, lesions or electrical stimulation separated by only 2–3 mm distance
. Figure 99-5 The position of 28 electrodes in Raprl (red trajectories) is compared with 100 electrodes positioned in STN (Black trajectories) of Benabid series [40], in a sagittal plane of the standardization technique used by Talairach et al [41]. Plotting was performed through a computerized program that considers the position of the contacts used for stimulation in “y” and “z” stereotactic coordinates. Notice that trajectories in Raprl are longer as the two contacts used for bipolar stimulation have been plotted, while STN includes only the contact used as cathode. Zero line corresponds to the AC–PC level, the length of AC–PC line is divided in 1/12 and STN is below from 4/12 to 8/12. None of the Raprl trajectories traverses STN area (Our data and STN data were elaborated by courtesy of Prof. A. Benabid)
99
induce different therapeutic effects (see > Table 99-1). In > Figure 99-5, we present the analysis of the trajectories of DBS in Raprl and STN, made by a computerized program, that clearly indicates that DBS aimed to those targets are contiguous but not superimposed. For that reason, it seems important to define subthalamic targets stereotactically (anatomically) and electrophysiologically. > Table 99-2 contains information published on the stereotactic coordinates of 3 subthalamic targets: STN, Raprl and Zi, as well as the direct target aimed in MRI [30,35,38]. The microelectrode recording pattern using similar microelectrode design, tip size, impedance and filtering has been published for STN [35,36] and Zi [51]. We have recently recorded Raprl with similar techniques and found that there is no neuronal firing in the target (> Figure 99-6), which confirms our report made with multiunitary recordings long ago [52,53]. Electrical stimulation over therapeutic charge density levels more often induced dyskinesia and dysarthria in STN [54]. This is not the case in Raprl, where paresthesias and diplopia are more often induced, indicating the proximity of Lm and 3rd cranial nerve fibers [29,30]. Finally, while decrease in tremor amplitude is
. Table 99-2 Intraoperative observations described for the most effective subthalamic targets in the treatment of Parkinsons disease: indirect target coordinates, direct target position, microelectrode recording, sensory-motor responses to electrical stimulation at intensities above therapeutic level and the effect of electrodes insertion on contralateral tremor (for explanation, see text). Notice that coordinates for the targets are contiguous, but do not superimpose. Abbreviations as in > Table 99-1 Indirect Target
STN
Raprl
cZi
X (mm) Y (mm) Z (mm) Direct MRI Target Microelectrode Recording
x ¼ 11.23 0.9 y ¼ 1.7 0.9 z ¼ 1.7 1.5 Dorsal Medial STN Dense High-Amplitude Unitary Recording Dyskinesias Dysarthria
x ¼ 11.69 0.66 y ¼ 6.73 1.62 z ¼ 4.38 1.02 Perirubral White Matter Few, if any, Low-Amplitude units Paresthesias Diplopia
x ¼ 14.0 1.56 y ¼ 5.8 1.46 x ¼ 2.1 1.05 Behind Posterior end of STh Middle-Amplitude Sparced Units (4–20 Hz bursts) NA
Ocassionally decreases contralateral tremor
Decreases or supresses contralateral tremor
NA
Electrical Stimulation over therapeutic charge density Insertion of DBS
1673
1674
99
Other targets to treat parkinson’s disease (posterior subthalamic targets and motor cortex)
. Figure 99-6 Microelectrode recordings made in an array of 5 electrodes, separated by a distance of 2 mm. On the right, a diagram Hv-1,5 of the atlas shows the anatomic position of each electrode. On the left, the recordings of electrode placed in Raprl (green), Ru (light blue), nucleus ventralis caudalis (yellow), Zi (dark blue) and H1 Forel’s field (red) are shown. Recordings at Raprl and H1 do not show neuronal firing
occasionally seen when inserting electrodes in STN, it is a rule for inserting electrodes in Raprl. This information is not available for cZi. Raprl share some clinical and electrophysiological similarities with mesencephalic reticular formation (MRF): Bilateral lesions in Raprl induce a state of bradykinesia that mimics what is seen in diseases that induce neural loss of mesencephalic tegmentum, such as progressive supranuclear palsy [28]. Bilateral implantation of DBS induce a transient state of somnolence accompanied by 3-Hz delta waves in the frontopolar EEG leads, as described in the discrete experimental lesions or mesencephalic compression in humans [12]. In Raprl and MRF, only late components of SEP (P200 and P300) are recorded, with maximal amplitude during a paradigm of selective attention [55]. Therefore, we have related Raprl to MRF in regard to tremor production, without ignoring that alleveation of rigidity by DBS in Raprl may result from deactivation of cerebellar fibers around red nucleus (perirubral fibers). Zi has been related to various functions in experimental animals: visceral control, arousal, locomotion and posture, organized from rostral to caudal, most likely expressed through thalamic intralaminar (Ce-pf ) and ventrolateral
(Vim, Vop) nuclei, as well as brainstem centers (for review see references [38] and [56]). In summary, clinical experience has confirmed the hypothesis advanced by Mundinger in that targeting compacted fibers or nuclei in the subthalamus is more efficient to control PD symptoms than targets in the thalamus or Gpi. On the other hand, electrophysiological and imaging studies of the posterior subthalamic target suggest that it is related to a motor system different from that suggested by Alexander et al. [57]. Posterior subthalamic target seems to be linked with motor centers in the brain stem and cerebellum.
Motor Cortex Stimulation for Movement Disorders Stimulation of the motor cortex for movement disorders is a less invasive, emerging, therapeutic approach. Functional benefit following motor cortex stimulation in movement disorders is supported by both preclinical studies and clinical observations. One of the first clinical observation was reported by Woolsey et al., as early as 1979 [58]. In this study, stimulation of the sensorymotor cortex, at current values subthreshold for
Other targets to treat parkinson’s disease (posterior subthalamic targets and motor cortex)
movement, induced a reduction of tremor and rigidity in two Parkinson disease (PD) patients. Woolsey observed that during stimulation voluntary movements were unaffected and strength of movement was improved. Using a different stimulation approach, Pascual-Leone [59] observed also that subthreshold transcranial motor cortex stimulation is capable of improving motor performances in akinetic parkinsonian patients. The clinical results were supported by other groups on Parkinson disease using transcranial motor cortex stimulation. Chronic electrical cortical stimulation consists in placing one or more epidural or subdural electrodes, unilaterally or bilaterally over the primary motor cortex by using MRI-guided neuronavigation system and electrophysiological recordings such as sensitive and motor evoked potentials. However, the surgical technique remains at present hetereogeneous between different surgical teams. Theexactplacementoftheelectrodesalongoracross the primary motor cortex, the number of contacts and distance between contacts greatly varied between groups, and need more clinical studies and devices development to establish an optimal surgical technique to treat specific motor symptoms. Most of the clinical observations of motor control following cortical stimulation were made in patients suffering from central neuropathic pain (due to thalamic haemorrhage or infarct, midbrain haemorrhage, multiple lacunar striatal or thalamic infarct and cerebellar abnormalities) associated with distal and proximal tremor, hemichoreoathetosis or dystonia. In these case reports, MCS could reduce pain and the severity of abnormal movements such as dystonic movements or tremor at a frequency between 40 and 125 Hz. [60,61]. Concerning Parkinson disease (PD) and related symptoms such as akinesia and locomotor activity, it has been demonstrated that unilateral high-frequency stimulation of motor cortex could induce a significant behavioral recovery in MPTP nonhuman primates model of advanced
99
PD. [62]. These behavioral observations were associated with an increased metabolic activity of the ipsilateral supplementary motor areas (SMA) using positron emission tomography (PET) imaging study. [63]. In PD patients and Parkinson disease syndromes, the beneficial clinical effect of cortical stimulation appears to be more heterogeneous as some pioneer open label studies reported a clinical efficacy in akinesia, rigidity and tremor in advanced PD patients [64,65,66,67,68] whereas other phase I/II clinical trials did not demonstrate any significant clinical efficacy [63,69]. This can be explained by the heterogeneous selection of patients (age, severity of the disease), the heterogeneous surgical technique and stimulation parameters used among different studies. It should be noted that a robust improvement of locomotor activity using bilateral high-frequency motor cortex stimulation has been demonstrated in a patient suffering from severe progressive primary freezing syndrome [70]. The behavioral improvement following motor cortex stimulation observed in this patient was associated with an increased cerebral blood flow over the SMA as observed in nonhuman primate study with the same stimulating parameters [62]. Indeed, functional neuroimaging has characterized various changes in cortical activity in patients with movement disorders. PET or functional MRI showed that the SMA is hypoactive in PD. Because the SMA is involved in automated or complex movements, hypoactivation of this area could play a critical role in akinesia symptom. Electrophysiological studies are consistent with imaging studies showing also an hypoactivity of the SMA in PD patients and increased cortical excitability of the corticospinal projections at rest, concomitant to, or resulting from a reduced intracortical inhibition. In contrast, dystonic movements may induce an increased metabolic activity in the dorsolateral prefrontal cortex, the pre-SMA and also in the lateral premotor cortex. Thus, movement disorders have been characterized by various functional cortical
1675
1676
99
Other targets to treat parkinson’s disease (posterior subthalamic targets and motor cortex)
changes, which could represent either primary or compensatory mechanisms of motor symptoms. Motor cortical stimulation could modulate these cortical electrophysiological abnormalities. It may produce extensive functional effects in subcortical structures that are directly involved in the pathological abnormality restoring a pattern of desynchronisation within specific neural pathways or loops.
References 1. Parkinson J. An essay on the shaking palsy. London: Sherwood Neely and Jones. 1817, pp. 66. 2. Aring CD, Fulton JF. Relation of the cerebrum to the cerebellum. Arch. Neurol. and Psychiat 1936;35:439-466. 3. Bucy PC. Surgical relief of tremor at rest. Ann Surg 1945;122:933-941. 4. Walker AE. Cerebral pedunculotomy for the relief of involuntary movements I: Hemibalismus. Acta Psychiat Neurol 1949;24:713-729. 5. Puusepp L. Chirurgishe Neuropahtologie. Dorpat Estonia Kruger 1931;1:416-417. 6. Puusepp L. Cordotomia posterior lateralis (fasc Burdachi) on account of trembling and hypertonia of the muscles in the hand. Folia Neuropath Estonia 1930;10:62-66. 7. Putman TJ. Treatment of athetosis and dystonia by section of the extrapyramidal motor tracts. Arch Neurol and Psychiat 1933;29:504-521. 8. Meyers RA. Surgical procedure for post-encephalitic tremors, rigidity and festination by surgery of the basal ganglia. A Research Nerv And Ment Dis Proc 1942;21:602-665. 9. Greenfied JC, Bosanket FD. The brain stem lesions in parkinsonism. J Neurol Neurosurg Psychiat 1953;16:213-226. 10. Ward AA, McCulloch WS, Magoun HW. Production of an alternating tremor at rest in monkeys. J Neurophysiol 1948;11:317-330. 11. Mettler FA, Experimental Tremor. Discussion. In Purpura DP and Yahr MD (eds). The Thalamus. New York, Columbia University Press. 1966, pp. 249. 12. Velasco F, Velasco M, Romo R, Maldonado H. Production and suppression of tremor by mesencephalic tegmental lesions in monkeys. Exp Neurol 1979;64:516-527. 13. Laitinen LV. Leksell’s unpublished pallidotomies of 1958– 1962. Stereotact Funct Neurosurg 2000;74(1):1-10. 14. Hassler R, Riechter T, Mundinger F, Umbach N, Gangelbreger JA. Physiological observations in stereotactic operations in extrapyramidal motor disturbances. Brain 1960;83:337-350.
15. Mundinger F. Stereotaxic interventions on the zona incerta area for treatment of extrapyramidal motor disturbances and their results. Confin Neurol 1965;26:222-230. 16. Spiegel EA, Wycis HT, Szekely EG, Baird HW, Adams J, Flanagan M. Campotomy. Trans Am Neurol Assoc 1962;87:240-242. 17. Spiegel EA, Wycis HT, Szekely EG, Adams J, Flanagan M and Baird HW. Campotomy in various extrapyramidal disorders. J Neurosurg 1963;20:871-881. 18. Story JL, French LA, Chou SN, Meier MJ. Experiences with subthalamic lesions in patients with movement disorders. Confin Neurol 1965;26:218-221. 19. Andy OJ, Jurko MF, Sias FR. Subthalamotomy in treatment of Parkinsonian tremor. J Neurosurg 1963;20:860-870. 20. Wycis HT, Szekely EG, Spiegel EA. Tremor on stimulation of the midbrain tegmentum after degeneration of the brachium conjunctivum. J Neuropath Exp Neurol 1957;16:79-84. 21. Velasco F, Velasco M, Ogarrio C, Fanghanel G. Electrical stimulation of centromedian thalamic nucleus in the treatment of convulsive seizures. A preliminary report. Epilepsia 1987;28:421-230. 22. Patel NK, Heywood P, O’Sullivan K, McCarter R, Love S, Gill SS. Unilateral subthalamotomy in the treatment of Parkinson’s disease. Brain 2003;126:1136-1145. 23. Bertrand C, Hardy J, Molina Negro P, Martinez SN. Optimum physiological target for the arrest of tremor. In: Gillingham JF and Donaldson IML (Eds). 3rd Symposium on Parkinson’s Disease. Edinburgh, Livingstone. 1969, pp. 251-254. 24. Velasco F, Molina-Negro P, Bertrand C, Hardy J. Further definition of the subthalamic target for tremor arrest. J Neurosurg 1972;36:184-191. 25. Velasco F, Velasco M, Machado JP. A statistical outline of the subthalamic target for the arrest of tremor. Appl Neurophysiol 1975;35:28-46. 26. Velasco F, Velasco M, Ogarrio C, Olvera A. Neglect induced by thalamotomy in humans: A quantitative appraisal of the sensory and motor deficits. Neurosurgery 1986;19:744-751. 27. Driollet R, Schvarcz JR, Orlando J. Optimum target for arrest tremor. Confin Neurol 1974;36:355. 28. Ito Z. Stimulation and Destruction of the Prelemniscal Radiation or its adjacent area in various Extrapyramidal Disorders. Confin Neurol 1975;37:41-48. 29. Velasco F, Jime´nez F, Perez ML, Carrillo-Ruiz JD, Velasco AL, Ceballos J, Velasco M. Electrical stimulation of the prelemniscal radiations in the treatment of Parkinson’s Disease: An Old Target Revised with New Techniques. Neurosurgery 2001;44:293-306. 30. Carrillo-Ruiz JD, Velasco F, Jime´nez F, Castro G, Velasco AL, Herna´ndez JA, Velasco M. Bilateral electrical stimulation of the prelemniscal radiations in the treatment of advanced Parkinson’s disease. Neurosurgery 2008;62:1-15.
Other targets to treat parkinson’s disease (posterior subthalamic targets and motor cortex)
31. Carrillo-Ruiz JD, Velasco F, Jime´nez F, Velasco AL, Velasco M, Castro G. Neuromodulation of prelemniscal radiations in the treatment of Parkinson’s disease. Acta Neurochir 2007;(Suppl)97(2):185-190. 32. Kitagawa M, Murata J, Uesugi H, Kikuchi S, Saito H, Tashiro K, Sawamura Y. Two year follow-up of chronic stimulation of the posterior subthalamic white matter for tremor-dominant Parkinson’s disease. Neurosurgery 2005;56:281-289. 33. Espinoza J, Arango JG. Surgical management of PD. Deep brain stimulation of prelemniscal radiations. Mov Dis 2005;20(suppl 10)5:159-160. 34. Murata JI, Kitagawa M, Uesegi H, Hisatoshi S, Iwasaki Y, Kikuchi S, Tashiro K, Sawamura Y. Electrical stimulation of the posterior subthalamic area for the treatment of intractable proximal tremor. J. Neurosurg 2003;99:708-715. 35. Lanotte MM, Rizzone M, Mergamasco B, Faccani G, Melcane A, Lopiano L. Deep brain stimulation of the subthalamic nucleus: anatomical, neurophysiological and outcome correlations with the effects of stimulation. J Neurol Neurosurg Psychiatry 2002;72:53-58. 36. Littlechild P, Vasma TR, Eldrige PR, Fox S, Foister A, Fletcher N, Steiger M, Byine P, Tyler K, Flintham S. Variability in position of the subthalamic nucleus targeted by magnetic resonance imaging and microelectrode recordings as compared to atlas coordinates. Stereotact Funct Neurosurg 2003;80:82-87. 37. Vogues J, Volkmann J, Allert N, Lehrke R, KoulousakisA, Freund HJ, Sturm V. Bilateral high frequency stimulation in the subthalamic nucleus for the treatment of Parkinson’s disease: correlation of therapeutic effect with anatomical electrode position. J Neruosurg 2002;96:269-279. 38. Plaha P, Ben-Shlomo Y, Patel NK, Gill SS. Stimulation of the caudal zona incerta is superior to stimulation of the subthalamic nucleus in improving contralateral parkinsonism. Brain 2006;129:1732-1747. 39. Schaltenbrand G, Wahren W. Atlas of Stereotaxy of the human brain. Stuttgart Thieme Pub. 1977. 40. Benabid A, Koudsie´ A, Benazzouz A, Fraix V, Ashraf A, Le Bas JF, Chabarde S, Pollak P. Subthalamic stimulation for Parkinson’s disease. Arch Med Res 2000;31:282-289. 41. Talairach J, David M, Tournoux P, Corredor H, Krasina T. Atlas d’anatomie ste´re´otaxique. Re´pe´rage radiologique des noyaux gris centraux, des regions mesencephalo-sous optique et hypothalamique de l’homme. Paris: Masson et Cie. 1957, pp. 293. 42. Benabid AL, Pollack P, Gervason C, Hoffmann D, Gao DM, Hommer M, Perret JE, de Rougemont J. Long term suppression of tremor by chronic stimulation of the ventral intermediate thalamic nucleus. Lancet 1991;337(8738):403-406. 43. Limousin P, Pollack P, Benazzouz A, Hoffmann D, Lebas JF, Broussolle E, Perret JE, Benabid AL. Effect on
44.
45. 46.
47. 48. 49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
99
Parkinsonian signs and syptoms of bilateral subthalamic stimulation. NEJM 1998;339(16):1105-1111. Loher TJ, Burgunder JM, Pohle T, Weber S, Sommorhalder R, Krauss JK. Long-term pallidal deep brain stimulation in patients with advanced Parkinson disease: 1 year follow up study. J Neurosurg 2002;96 (5):844-853. Talairach J. Atlas stereotactique de cerveau. Paris Masson, 1950, pp 73-225. Mitrofanis J, Askan K, Wallace BA, Benabid AL. Chemoarchitectonic heterogeneities in the primate zona incerta: clinical and functional implications. J Neurocytol 2004;33:429-440. Tepper JM, Abercrombie ED, Bolam JP. Basal ganglia macrocircuits. Prog Brain Res 2007;160:3-7. Garcia-Rill E. The basal ganglia and the locomotor regions. Brain Res 1986;396(1):47-63. Butson RC, Cooper SE, Henderson MJ, McIntyre CC. Patient specific analysis of the volume of tissue activated during deep brain stimulation. Neuroimage 2007;34:661-670. Miocinovic S, Parent M, Butson C, Gahn PJ, Russo GS, Vitek JL, McIntyre CC. Computational analysis of subthalamic nucleus and lenticular fascicularis activation during therapeutic deep brain stimulation. J Neurophysiol 2006;96:1569-1580. Merello M, Tenea E, Cerquetti D. Neuronal activity of the zona incerta in Parkinson’s disease patients. Movement Disorders 2006;21:937-943. Velasco F, Velasco M. A reticulothalamic system engaged in selective attention and tremor in man. Neurosurgery 1979;4:30-36. Jime´nez F, Velasco F, Velasco M, Brito F, Morel C, Marquez I, Pe´rez ML. Subthalamic prelemniscal radiations stimulation for the treatment of Parkinson’s disease. I.-Electrophysiological characterization of the area. Arch Med Res 2000;31:62-74. Starr PA. Placement of Deep Brain stimulators into the subthalamic nucleus or globus pallidus internus: technical approach. Stereotact Funct Neurosurg 2002;79:118-145. Velasco M, Velasco F, Maldonado H, Machado J. Differential effect of thalamic and subthalamic lesions on early and late components of somatic evoked potentials in man. Electroenceph Clin Neurophysiol 1975;39:163-171. Mitrofanis J. Some certainly for the “zone of uncertainly”? Exploring the function of the zona incerta. Neuroscience 2005;130:1-15. Alexander GE, DeLong MR, Strick PL. Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annu Rev Neurosci 1986;9:357-378. Woolsey CN, Erickson TC, Gilson WE. Localization in somatic sensory and motor areas of human cerebral cortex as determined by direct recording of evoked potentials and electrical stimulation. J Neurosurg 1979;51: 476-506.
1677
1678
99
Other targets to treat parkinson’s disease (posterior subthalamic targets and motor cortex)
59. Pascual-Leone A, Valls-Sole` J, Brasil-Neto JP, Cammarota A, Grafman J, Hallett M: Akinesia in Parkinson’s disease. II: effects of subthreshold repetitive transcranial motor cortex stimulation. Neurology 1994;44:892‐898. 60. Katayama Y, Oshima H, Fukaya C, Kawamata T, Yamamoto T. Control of post-stroke movement disorders using chronic motor cortex stimulation. Acta Neurochir 2002;79(suppl):89-92. 61. Nguyen JP, Pollin B, Feve A, Geny C, Cesaro P. Improvement of action tremor by chronic cortical stimulation. Mov Disord 1998;13:84-88. 62. Drouot X, Oshino S, Jarraya B, . et al. Functional recovery in a primate model of Parkinson’s disease following motor cortex stimulation. Neuron 2004;44:769-778. 63. Strafella AP, Lozano AM, Lang AE, Ko JH, Poon YY, Moro E. Subdural motor cortex stimulation in Parkinson’s disease does not modify movement-related rCBF pattern. 64. Mov Disord. 2007 Oct 31;22(14):2113‐6. 64. Canavero S, Paolotti R. Extradural motor cortex stimulation for advanced Parkinson’s disease: case report. Mov Disord 2000;15:169-171.
65. Canavero S, Paolotti R, Bonicalzi V, et al. Extradural motor cortex stimulation for advanced Parkinson disease. Report of two cases. J Neurosurg 2002;97:1208-1211. 66. Canavero S, Bonicalzi V, Paolotti R, et al. Therapeutic extradural cortical stimulation for movement disorders: a review. Neurol Res 2003;25:118-122. 67. Franzini A, Ferroli P, Dones I, Marras C, Broggi G. Chronic motor cortex stimulation for movement disorders: a promising perspective. Neurol Res 2003;25:123-126. 68. Pagni CA, Zeme S, Zenga F, Maina R. Extradural motor cortex stimulation in advanced Parkinson’s disease. Neurosurgery 2005;57:E402. 69. Kleiner-Fisman G, Fisman DN, Kahn FI, Sime E, Lozano AM, Lang AE. Motor cortical stimulation for parkinsonism in multiple system atrophy. Arch Neurol 2003;60:1554-1558. 70. Tani N, Saitoh Y, Kishima H, Oshino S, Hatazawa J, Hashikawa K, Yoshimine T. Motor cortex stimulation for levodopa-resistant akinesia: case report. Mov Disord. 2007 Aug 15;22(11):1645-9.
92 Pallidotomy for Parkinson’s Disease M. I. Hariz
Introduction After the introduction of L-dopa in the treatment of Parkinson’s disease (PD) in the late 1960ies, stereotactic surgery for PD, which had been a common procedure until then, went into a long period of hibernation. The renaissance of surgery for PD started in the mid-eighties when Lauri Laitinen in Umea˚, Sweden, resurrected Leksell’s posteroventral pallidotomy published in 1960 [1], for treatment of post-L-dopa PD [2,3]. PostL-dopa PD was a ‘‘different’’ disease compared to pre-L-dopa PD, in as much as the patients who had been on L-dopa medication for several years presented not only with the four cardinal symptoms of PD (tremor, bradykinesia, rigidity, and postural instability) but also with so-called motor fluctuations and dopa-induced dyskinesias (DID). Following the first reports of Laitinen about posteroventral pallidotomy (PVP), this method experienced a worldwide spread in clinical and academic practice [4], and the original publication of Laitinen from 1992 [2] remains the most quoted paper in the literature on surgery for PD. Even though today, deep brain stimulation (DBS), especially subthalamic nucleus (STN) DBS has become the dominant surgical procedure for PD, PVP and other lesional surgeries for PD are still considered, performed and evaluated [5,6], albeit very rarely. According to the American Academy of Neurology, level of evidence A rating of a given treatment requires at least two consistent Class I studies; a class I study consists of a prospective, randomized, controlled clinical trial with masked #
Springer-Verlag Berlin/Heidelberg 2009
outcome assessment in a representative population. So far, the only surgical procedure for PD that qualifies for level of evidence A is in fact PVP, thanks to at least two published studies that fulfill the above mentioned criteria [7,8]. Furthermore, PVP is the only surgical procedure for PD that has published outcome studies beyond 5 years [9], in fact up to 13 years after surgery in individual patients [10]. Finally, PVP has been shown in several and recent reviews of controlled studies to improve quality of life [11–13]. This Chapter will review the rational and indications for PVP. The surgical technique in the hands of the author, and the effects and side effects of PVP will be detailed including a mention of some controversies that have surrounded these issues in the literature. Finally, some reasons for the decline of this procedure will be provided, along with an opinion on the place that PVP ought to have today within the armamentarium of surgical therapies for advanced PD.
Rationale for Pallidotomy There is a widely spread claim related to PVP: it is the claim about the role of the animal model [14,15] in the renaissance of pallidal surgery for PD: There is no need to re-write history: The pioneer publication that contributed to the resurrection of surgery for Parkinson’s disease (PD) is that of Laitinen et al. on Leksell’s posteroventral pallidotomy published in January 1992 [2]. This paper described the results of a series of patients operated on between 1985 and 1990, that is, patients operated on before even the
1540
92
Pallidotomy for parkinson’s disease
existence of the 1-methyl-4-phenyl-1-2-3-6 tetrahydropyridine (MPTP) animal model of basal ganglia circuitry and function in PD. It is following discussions with Leksell that Laitinen tried pallidotomy on his patients, as he clearly stated in the acknowledgements part of the pioneering paper of 1992. ‘‘Our idea to revisit the pallidum was born during and after many fruitful discussions with Professor Lars Leksell, 1907–1986. We are deeply grateful to this great man and friend’’ [2]. Needless to mention that when Leksell performed his pallidotomies in the fifties, and moved progressively his target further posterior and ventral in the pallidum [1,2], there was neither L-dopa nor animal model around. Hence the information obtained later on through meticulous basic science experimental studies using non-human primate models of PD confirmed the clinical observations and experience of Leksell and Laitinen, i.e., that a stereotactic lesion in the (sensorimotor) posteroventral pallidum alleviated the cardinal symptoms of PD. Furthermore, the most robust and long-lasting effect of PVP, has been on the dopa induced dyskinesias (DID) of PD patients [9,10], a phenomenon that did not exist in the patients of Svennilson and Leksell in the 1950s. The animal model provided thus an a posteriori confirmation of the clinical observation in humans, and provided a scientific pathophysiological rationale for the effect of pallidotomy on parkinsonian symptoms [16,17]. In summary, L-dopa deficiency in the putamen provokes through a GABAergic direct pathway to the globus pallidus internus (GPi) and an indirect GABAergic and glutaminergic pathway through external pallidum and subthalamic nucleus (STN), a pathological pattern of activity (some qualify it as a pattern of overactivity) of the GPi, which provokes an inhibition of movement time and inhibition of initiation of movements by inhibiting the thalamo-cortical circuitry. This inhibition may account for the
development of akinesia, rigidity and tremor in PD. The dyskinesia seen in PD patients after long duration L-dopa treatment may also be mediated by a dysinhibition of the inhibitory STN which provokes a pathologically decreased pattern of activity of the inhibitory globus pallidus and hence, the initiation of choreoathetotic and dystonic movements, i.e., dyskinesias. Therefore, a lesion in the posteroventral sensorimotor part of the globus palldus internus GPi will contribute to elimination of the pathological pattern of neuronal activity of the GPi and a release or ‘‘normalization’’ of the thalamo-cortical activity, accounting for symptom improvement.
Indications Cognitively preserved patients, without major brain atrophy, and suffering from L-dopa sensitive PD with motor fluctuations and on-off phenomena, on- or off- dystonias, dyskinesias (hyperkinesias), and muscular cramps may benefit most from PVP. Even tremor may benefit from PVP although perhaps to a lesser extend than from thalamotomy. There is an impact profile of PVP: Dyskinesias respond best, while the effect on akinesia, and gait freezing are clearly less robust. Ideally, the patient should present with asymmetric symptoms. PVP should not be performed bilaterally in one session. A second contralateral pallidotomy may be performed at least 6 months after the first one and only if the first pallidotomy has been without side effects, and has been verified to lie entirely within the posteroventral pallidum without encroachment on the internal capsule. One should keep in mind that bilateral pallidotomies, even staged, may increase the risk for side effects, especially the risk for dysarthria, dysphonia, and cognitive decline [18–21]. Besides, it has been observed that the second contralateral pallidotomy usually is not as efficacious on
Pallidotomy for parkinson’s disease
contralateral symptoms as the first one [20]. Unilateral pallidotomy is better tolerated in elderly patients than bilateral STN DBS, most probably due to the unilaterality of the surgery, but also due to the fact that in the STN, associative, limbic and motor circuitries are less segregated than in the larger GPi.
Preoperative Imaging The pallidal target can be visualized with a good quality MRI using proton density or inversion recovery sequences [22,23]. Furthermore, the optic tract, which is very close to the ventral border of the internal globus pallidus (GPi), can be visualized. Even the lamina medullare between the internal and external globus pallidus (GPi and GPe) and between the GPe and putamen can be visualized. Provided these visualizations, there is no need to follow blindly anatomical coordinates based on the anterior commissure (AC) and posterior commissure (PC) of the third ventricle and the Atlas. The initial Leksell-Laitinen coordinates for the pallidal target had been defined as follows: 2–3 mm in front of the midcommissural point (MCP), 20– 22 mm lateral and 4–6 mm below the AC-PC line. In many patients this is indeed the case, but sometimes, depending on the shape of the pallidum, these coordinates can be fine-tuned if the target structures and their surroundings are visualized on thin-slice axial and coronal MRI. The MRI study is typically done on a patient off-medication and thus, without dyskinesias, during the imaging. Two mm-thick axial slices are scanned through the target area from above the foramen of Monro until the upper brain stem. Then 3–5 coronal 2 mm-thick slices are scanned perpendicular to the AC-PC line and at the level of the mamillary bodies. On these slices, the depth of the individual target in relation to the base of pallidum, supra-amygdala, and optic tract is assessed. The laterality of the target point
92
can be defined both on axial slices and on coronal slices. It should be in the posterior pallidum, 2 mm lateral to the border between medial pallidum and internal capsule.
Surgery Many authors who have published on pallidotomy have used microelectrode recording (MER) techniques for physiological confirmation of the target, prior to performing the lesion. Much has been written about the advantages of this technique to ensure a proper placement of lesions in the sensorimotor GPi and to provide the most optimal results of the surgery on PD symptoms. It is beyond the scope of this chapter to discuss the issue of MER in pallidotomy and it is referred instead to papers analyzing some of the MER literature [24–27]. Undoubtedly, MER is an exquisite research tool and has provided valuable insight and knowledge about the behavior of the neurons of the GPi in PD patients [28]. But MER is not a uniform procedure since its use varies considerably between surgeons, with some using up to 17 MER tracks to ‘‘map’’ the pallidal area prior to lesioning [29,30], while others use fewer tracks [30,31]. The present author does not use microelectrode recording. At surgery, impedance recording is used to differentiate between grey matter, white matter and CSF space. Impedance recording, macro-stimulation, and lesioning are carried out with a non-insulated electrode tip of 2 mm in length and 1.5 mm in diameter. The main aim of stimulation is not to provoke a block of the symptoms (like tremor arrest during thalamic stimulation) but to avoid internal capsule and optic nerve. The effects of acute intraoperative pallidal stimulation on the parkinsonian symptoms per se are not consistent. Sometimes there is no response, sometimes intraoperative stimulation increases the speed of movement of the leg or arm, or even provoke slight dyskinesias.
1541
1542
92
Pallidotomy for parkinson’s disease
In order to maximize the symptoms for the time of surgery, all parkinsonian medication should be stopped at least 6 h before the operation. The burr hole should be 2–2.5 cm from the midline at the level of the coronal suture. The probe is stopped at a level 6 mm above the zero target (the ‘‘zero target’’ being a point in the ventralmost GPi) for an observation period, and reading of impedance which should be grey matter impedance. Electrical stimulation is carried on with 6 Hz up to 10 mA and 120 Hz up to 5 mA. During stimulation, the patient should be carefully observed and the following should be checked: alertness, orientation, memory, speech articulation and voice, facial expression, limb strength, limb movements, limb dexterity, limb coordination, sensation at fingertips, cheek, tongue and lips, vision, and of course eventual effect on tremor, rigidity, limb akinesia etc., and a conversation should be kept with the patient. If stimulation does not give rise to any undesirable reactions (capsular, optic, mental), a radiofrequency lesion is produced with 80 C during 60 s. This coagulation may result in improvement of bradykinesia in the leg. Dyskinesias in the foot may appear simultaneously, which is considered to be a proof that the probe is in a good target. Then the probe is moved in 2 mm steps further down to the zero target level. Impedance recordings, stimulations and coagulations are performed as above. The tremor -if present- and bradykinesia of the arm and hand may be reduced at the levels 2 mm above the zero level and below. During stimulation at these lowest levels, extreme care should be taken to assess any visual response to stimulation, before coagulation of the zero level is performed. If the impedance suddenly drops below 350 Ohms or stimulation gives rise to a visual response, surgery is interrupted. With this technique the previously published side effect of 14% scotoma [2] has been reduced to less than 1%. The effect of a successful pallidotomy is often visible on the operating table. The patient usually
reports that the contralateral leg weighs less than the other one upon raising the legs alternatively. Movements such as cycling with the leg and finger tap are much faster than on the non-operated side. Even the muscular cramp, when present, disappears during the surgery. Eventual tremor bursts may still occur. Muscular hypotonia is rare after pallidotomy.
Postoperative Imaging To document lesion’s location and size and its relation to effects and side effects of surgery is but evident. For this, it is mandatory that the postoperative MRI is performed some time after the operation when the edema surrounding the lesion has subsided. Also the MRI study should be done with thin axial slices parallel to AC-PC and thin coronal slices perpendicular to AC-PC. The scanning should use the same parameters as for the preoperative one. Only then can one properly evaluate the extent of the lesion in relation to the targeted area and in relation to the subdivisions of the pallidum (> Figure 92-1). Several publications have shown postoperative MRI scans on which one could see that the lesions were not where they were intended to be notwithstanding the authors’ claims about the accuracy of the methods used for targeting [29,32]. Other authors documented in detail the anatomical disparity of their pallidotomy lesions and tried to correlate the locations of lesions to the differential results of surgery on the various symptoms of the patient, and the impact of lesion location on side effects [33–36]. Asides from obvious capsular symptoms in case the lesion encroaches on the internal capsule (paresis, dysarthria, dysphagia in bilateral capsular encroachment), it has been shown that cognitive and behavioral side effects were more common in patients in whom the lesions were placed into the anteromedial, limbic-associative pallidum [36].
Pallidotomy for parkinson’s disease
92
. Figure 92-1 Stereotactic MRI, performed one year after surgery, with four axial contiguous 2 mm-thick scans showing a rightsided pallidotomy located in the posteroventral GPi
Side Effects of Pallidotomy Generally speaking, the risks of pallidotomy are less than those of thalamotomy. PVP is better tolerated by elderly patients, despite their having the disease for a longer time than the typical ‘‘thalamotomy patient’’ [37]. PVP in either hemisphere may provoke worsening of memory, injury to optic tract, paresis, depression, stroke, drooling. PVP in dominant hemisphere may provoke confusion, dysarthria, dysphonia. Some side
effects of pallidotomy may become evident only days or weeks after surgery. This concerns mainly side effects such as brain infarction or stroke [19], dysarthria, and possible memory deterioration. Especially after the second contralateral pallidotomy, there are increased risks of negative affection of speech, swallowing and cognition [18–20,38]. As for the severe complications of the surgery, such as brain hemorrhage, paralysis, and the like, several studies have shown that
1543
1544
92
Pallidotomy for parkinson’s disease
pallidotomy is safe in experienced hands but that upon review of all available literature it was shown that in patients operated on using MER there were more bleeding complications than in those who had a macrostimulation-guided pallidotomy [24,26,39]. The Toronto group, in their paper published in 2001 [39], provided the following details: In 1959 patients from 85 papers from 40 centers in 12 countries, MER was used in 46.2% and macrostimulation in 53.8% of the patients. Cerebral hemorrhage occurred in 2.7% of MER patients and in 0.5% of macroelectrode patients. Overall complications occurred in 26% of MER patients and in 19.2% of macroelectrode patients [39].
Results The results of pallidal surgery depend on what symptoms are assessed. The degree of improvement of the symptoms, dyskinesias excepted, may be variable. Following pallidotomy, ‘‘on-off ’’ fluctuations may still occur; however, the ‘‘on’’ periods generally last longer and provide better mobility without dyskinesias, and the ‘‘off ’’ periods usually are not as profound and as longlasting as before surgery.
Regardless of what surgical technique had been used, MER-guided or macrostimulationguided, the most consistent finding in the literature is that pallidotomy exerts its main effect on limb dyskinesia/dystonia, rigidity, and tremor, in that order, and least on gait freezing and other axial symptoms. Although the percentages of improvement in various aspects of the Unified Parkinson’s Disease Rating Scale (UPDRS) reported in the literature were rather disparate, this disparity, is not between reports from MER groups versus non-MER groups, but within either group, as has been shown by Starr et al. in their comprehensive survey on effect of unilateral pallidotomy, published in 1998 [40], and as is shown here in selected well performed studies listed in > Table 92-1. Hence, unilateral posteroventral pallidotomy results in general in a 20–30% improvement of motor symptoms when evaluated with the UPDRS in patients who are off-medication state at the time of the scoring. This is certainly a modest improvement, although it has been stated by prominent movement disorder neurologists, who assessed pallidotomy effects, that ‘‘large changes in motor function in specific tasks or specific parts of the body can be concealed if the usual Parkinson’s disease rating scales are employed’’ [51].
. Table 92-1 Percentage reduction of UPDRS Part III (motor scores, off-medications) in various pallidotomy studies Follow-up postop
Author (year)
MER
No. of patients
% Reduction
24 months 52 months 6 months 12 months 12 months 24 months 12 months 6 months 12 months 6 months 6 months 12 months
Vitek (2003) Fine (2000) Giller (1998) Kondziolka (1999) De Bie (2001) Eskandar (2000) Masterman (1998) Kopyov (1997) Uitti (1998) Kumar (1998) Melnick (1999) Lai (2000)
Yes Yes No No No No No Yes Yes Yes Yes Yes
20 20 47 58 32 68 32 29 41 39 29 89
25% [7] 19% [9] 31% [41] 23% [42] 26% [43] 20% [44] 24% [45] 25% [46] 21% [47] 31% [48] 16% [49] 35% [50]
Pallidotomy for parkinson’s disease
Is There Still a Place for Pallidotomy? In 2001, ‘‘Moving Along’’ the newsletter of the Movement Disorders Society published controversy articles about pallidotomy, with neurologist Mahlon DeLong defending the use of pallidotomy [52] and neurosurgeon Alim-Louis Benabid stating that pallidotomy should be abandoned [53]. DeLong argued that pallidotomy is less expensive, without need for time consuming adjustment of stimulation parameters, does not require battery replacement, and is without risk of infection and hardware problems. Benabid was of the argument that DBS can be safely performed bilaterally, is adjustable to needs of patients, is reversible, and does not preclude later surgical therapy trials. It is peculiar for the ‘‘modern’’ posteroventral pallidotomy that it has had such a great appeal on movement disorders neurologists who were extremely prolific in studying and publishing its effects, even more than neurosurgeons [4]. Pallidotomy has been extensively documented in long-term follow up studies. Besides it is the only surgical procedure for PD so far that fulfils criteria for level A, class I, evidence. It provides an improvement of PD motor symptoms of the magnitude of 20–30%, at the same time allowing for unchanged or even increased doses of dopaminergic medications without fear for severe dyskinesias. It is safe and well tolerated even by elderly patients, with few of the non-motor side effects inherent to other allegedly safer surgical therapies such as STN DBS [5,6,54]. It is cheap, readily accessible for patients living far away from movement disorders centers or in areas and countries lacking means for sophisticated expensive healthcare. It is certainly not as safe as DBS if performed bilaterally but, even when unilateral, evidence shows that it is still by far better for patients with advanced PD than medications alone [7,8].
92
So why has pallidotomy in effect been abandoned after less than a decade of renewed interest? In this author’s opinion, there are several reasons for that: The first reason is the advent of DBS, especially DBS in STN. This is a procedure than can be done simultaneously bilaterally, the side effects of which may be in part reversible, at least in the short term, and it is so efficient on motor symptoms that it allows decrease of medications, and to some extent it is adaptable to the needs of the patient over time. But above all, DBS provides for the first time a convenient living human model for safe and ethical clinical studies and research, since stimulation can, at any time point after surgery, be shut off, manipulated, or altered at will. Notwithstanding the numerous documented side effects of STN DBS, especially the non-motor side effects [55], that have put very strict inclusion criteria on patients eligible for this therapy – in fact more severe inclusions criteria than for pallidotomy – this procedure has completely overshadowed the more lenient pallidotomy. Other important reasons for the marked decline of use of pallidotomy is that there is no money involved in this procedure since no company is interested in sponsoring meetings, conferences, education or research on pallidotomy. Additionally, there is no sensational marketing, or impressive audiovisual promotion for pallidotomy as is the case for DBS, on the internet and elsewhere. Also there is a ‘‘denigration’’ of pallidotomy (and other lesional surgical procedures for PD) based on the notion that ‘‘pallidotomy creates an additional lesion in an already sick brain’’ [53]. In this author’s opinion, this may be a non-argument, and the history of functional neurosurgery is full of examples where ablation of pathological brain areas in PD, in psychiatric illness, and in epilepsy have indeed improved the patients: It is well known for example that no epileptic temporal lobe is better for the patient than an epileptic one, and, as Lozano put it once
1545
1546
92
Pallidotomy for parkinson’s disease
in a meeting, ‘‘no pallidum is better than a dysfunctional pallidum’’ for patients with advanced PD. This apparent ‘‘paradox’’ by which destroying a specific and limited dysfunctional area of the brain in PD in fact improves the patient because it removes a disturbing signal, is indeed an issue that eminent movement disorder neurologist have tried to understand and explain [56,57] Another important issue related to the decline of pallidotomy is the current lack of skills to do proper lesions among the new generation of young functional neurosurgeons who started their functional career with DBS and who do not have a prior experience or training in lesional stereotactic surgery. It is very easy to implant electrodes in basal ganglia targets for somebody who has been trained to perform lesional surgery, but it is not at all evident that a surgeon who is skilful at DBS surgery can automatically perform safe lesional procedures; hence, in many centers, pallidotomy is not even offered to PD patients who for any reason are not eligible for DBS, but may be in fact readily eligible for unilateral pallidotomy, because of their motor fluctuations, dyskinesias or parkinsonian dystonia. For the sake of these, far from rare, patients, it falls on experienced neurosurgeons to make sure that the skills of performing proper and safe stereotactic lesional surgery do not get lost, lest neurosurgeons who know how to do proper functional lesions, including posteroventral pallidotomy, become an extinguished species.
Dedication This Chapter is dedicated to the memory of Lauri Laitinen (1928–2005), who resurrected functional neurosurgery for Parkinson’s disease by pioneering the revival of leksell’s posteroventral pallidotomy.
References 1. Svennilson E, Torvik A, Lowe R, Leksell L. Treatment of parkinsonism by stereotactic thermolesions in the pallidal region. Acta Psychiatr Neurol Scand 1960;35: 358-77. 2. Laitinen LV, Bergenhein AT, Hariz MI. Leksell’s posteroventral pallidotomy in the treatment of Parkinson’s disease. J Neurosurg 1992;76:53-61. 3. Laitinen LV, Bergenheim AT, Hariz MI. Ventroposterolateral pallidotomy can abolish all parkinsonian symptoms. Stereotact Funct Neurosurg 1992;58:14-21. 4. Hariz MI. From functional neurosurgery to ‘‘interventional’’ neurology: a review of publications on thalamotomy, pallidotomy, and deep brain stimulation for Parkinson’s disease from 1966 to 2001. Mov Disord 2003;18:845-53. 5. Hooper AK, Okun MS, Foote KD, Fernandez HH, Jacobson C, Zeilman P, et al. Clinical cases where lesion therapy was chosen over deep brain stimulation. Stereotact Funct Neurosurg 2008;86:147-52. 6. Okun MS, Vitek JL. Lesion therapy for Parkinson’s disease and other movement disorders: update and controversies. Mov Disord 2004;19:375-89. 7. Vitek JL, Bakay RA, Freeman A, Evatt M, Green J, McDonald W, et al. Randomized trial of pallidotomy versus medical therapy for Parkinson’s disease. Ann Neurol 2003;53:558-69. 8. de Bie RM, de Haan RJ, Nijssen PC, Rutgers AW, Beute GN, Bosch DA, et al. Unilateral pallidotomy in Parkinson’s disease: a randomised, single blind, multicentre trial. Lancet 1999;354:1665-9. 9. Fine J, Duff J, Chen R, Chir B, Hutchison W, Lozano AM, et al. Long-term follow-up of unilateral pallidotomy in advanced Parkinson’s disease. N Engl J Med 2000;342:1708-14. 10. Hariz MI, Bergenheim AT. A 10-year follow up review of patients who underwent Leksell’s posteroventral pallidotomy for Parkinson’s disease. J Neurosurg 2001;94: 552-8. 11. Krack P, Hamel W, Mehdorn HM, Deuschl G. Surgical treatment of Parkinson’s disease. Curr Opin Neurol 1999;12:417-25. 12. Goetz CG, Poewe W, Rascol O, Sampaio C. Evidencebased medical review update: pharmacological and surgical treatments of Parkinson’s disease: 2001 to 2004. Mov Disord 2005;20:523-39. 13. Martinez-Martin P, Deuschl G. Effect of medical and surgical interventions on health-related quality of life in Parkinson’s disease. Mov Disord 2007;22:757-65. 14. Albin RL, Young AB, Penney JB. The functional anatomy of basal ganglia disorders. Trends Neurosci 1989;12: 366-75.
Pallidotomy for parkinson’s disease
15. Alexander GE, Crutcher MD, DeLong MR. Basal ganglia-thalamocortical circuits: parallel substrates for motor, oculomotor, ‘‘prefrontal’’ and ‘‘limbic’’ functions. Prog Brain Res 1990;85:119-46. 16. Lang AE, Lozano AM. Parkinson’s disease. First of two parts. N Engl J Med 1998;339:1044-53. 17. Lang AE, Lozano AM. Parkinson’s disease. Second of two parts. N Engl J Med 1998;339:1130-43. 18. York MK, Lai EC, Jankovic J, Macias A, Atassi1 F, Levin HS, et al. Short and long-term motor and cognitive outcome of staged bilateral pallidotomy: a retrospective analysis. Acta Neurochir (Wien) 2007;149:857-66. 19. Intemann PM, Masterman D, Subramanian I, DeSalles A, Behnke E, Frysinger R, et al. Staged bilateral pallidotomy for treatment of Parkinson disease. J Neurosurg 2001;94:437-44. 20. De Bie RMA, Schuurman PR, Esselink RAJ, Bosch DA, Speelman JD. Bilateral pallidotomy in Parkinson’s disease: a retrospective study. Mov Disord 2002;17:533-8. 21. Counihan TJ, Shinobu LA, Eskandar EN, Cosgrove GR, Penney JB Jr. Outcomes following staged bilateral pallidotomy in advanced Parkinson’s disease. Neurology 2001;56:799-802. 22. Hirabayashi H, Tengvar M, Hariz MI. Stereotactic imaging of the pallidal target. Mov Disord 2002;17 Suppl 3: S130-S134. 23. Starr PA, Vitek JL, DeLong M, Bakay RA. Magnetic resonance imaging-based stereotactic localization of the globus pallidus and subthalamic nucleus. Neurosurgery 1999;44:303-13. 24. Hariz MI, Fodstad H. Do microelectrode techniques increase accuracy or decrease risks in pallidotomy and deep brain stimulation? A critical review of the literature. Stereotact Funct Neurosurg 1999;72:157-69. 25. Hariz MI. Safety and risk of microelectrode recording in surgery for movement disorders. Stereotact Funct Neurosurg 2002;78:146-57. 26. Palur RS, Berk C, Schulzer M, Honey CR. A metaanalysis comparing the results of pallidotomy performed with microelectrode recording or macroelectrode stimulation. J Neurosurg 2002;96:1058-62. 27. Honey CR, Berk C, Palur RS, Schulzer M. Microelectrode recording for pallidotomy: mandatory, beneficial or dangerous? Stereotact Funct Neurosurg 2001;77:98-100. 28. Levy R, Dostrovsky JO, Lang AE, Sime E, Hutchison WD, Lozano AM. Effects of apomorphine on subthalamic nucleus and globus pallidus internus neurons in patients with Parkinson’s disease. J Neurophysiol 2001;86:249-60. 29. Dogali M, Fazzini E, Kolodny E, Eidelberg D, Sterio D, Devinsky O, et al. Stereotactic ventral pallidotomy for Parkinson’s disease. Neurology 1995;45:753-61. 30. Alterman RL, Sterio D, Beric A, Kelly PJ. Microelectrode recording during posteroventral pallidotomy: impact on target selection and complications. Neurosurgery 1999;44:315-23.
92
31. Vitek JL, Bakay RA, Hashimoto T, Kaneoke Y, Mewes K, Yu Zhang J, et al. Microelectrode-guided pallidotomy: technical approach and its application in medically intractable Parkinson’s disease. J Neurosurg 1998;88: 1027-43. 32. Kirschman DL, Milligan B, Wilkinson S, Overman J, Wetzel L, Batnitzky S, et al. Pallidotomy microelectrode targeting: neurophysiology-based target refinement. Neurosurgery 2000;46:613-24. 33. Hariz MI, Hirabayashi H. Is there a relationship between size and site of the stereotactic lesion and symptomatic results of pallidotomy and thalamotomy? Stereotact Funct Neurosurg 1997;69:28-45. 34. Gross RE, Lombardi WJ, Lang AE, Duff J, Hutchison WD, Saint-Cyr JA, et al. Relationship of lesion location to clinical outcome following microelectrode-guided pallidotomy for Parkinson’s disease. Brain 1999;122:405-16. 35. Trepanier LL, Kumar R, Lozano AM, Lang AE, SaintCyr JA. Neuropsychological outcome of GPi pallidotomy and GPi or STN deep brain stimulation in Parkinson’s disease. Brain Cogn 2000;42:324-47. 36. Lombardi WJ, Gross RE, Trepanier LL, Lang AE, Lozano AM, Saint-Cyr JA. Relationship of lesion location to cognitive outcome following microelectrode-guided pallidotomy for Parkinson’s disease: support for the existence of cognitive circuits in the human pallidum. Brain 2000;123:746-58. 37. Hariz MI, DeSalles AAF. The side-effects and complications of posteroventral pallidotomy. Acta Neurochir Suppl 1997;68:42-8. 38. de Bie RMA, de Haan RJ, Schuurman PR, Esselink RAJ, Bosch DA, Speelman JD. Morbidity and mortality following pallidotomy in Parkinson’s disease. A systematic review. Neurology 2002;58:1008-12. 39. Alkhani A, Lozano AM. Pallidotomy for Parkinson disease: a review of contemporary literature. J Neurosurg 2001;94:43-9. 40. Starr PA, Vitek JL, Bakay RA. Ablative surgery and deep brain stimulation for Parkinson’s disease. Neurosurgery 1998;43:989-1013. 41. Giller CA, Dewey RB, Ginsburg MI, Mendelsohn DB, Berk AM. Stereotactic pallidotomy and thalamotomy using individual variations of anatomic landmarks for localisation. Neurosurgery 1998;42:56-65. 42. Kondziolka D, Bonaroti E, Baser S, Brandt F, Kim YS, Lunsford LD. Outcomes after stereotactically guided pallidotomy for advanced Parkinson’s disease. J Neurosurg 1999;90:197-202. 43. de Bie RM, Schuurman PR, Bosch DA, de Haan RJ, Schmand B, Speelman JD. Outcome of unilateral pallidotomy in advanced Parkinson’s disease: cohort study of 32 patients. J Neurol Neurosurg Psychiatry 2001;71: 375-82. 44. Eskandar EN, Shinobu LA, Penney JB Jr, Cosgrove GR, Counihan TJ. Stereotactic pallidotomy performed without
1547
1548
92 45.
46.
47.
48.
49.
50.
Pallidotomy for parkinson’s disease
using microelectrode guidance in patients with Parkinson’s disease: surgical technique and 2-year results. J Neurosurg 2000;92:375-83. Masterman D, DeSalles A, Baloh RW, Frysinger R, Foti D, Behnke E, et al. Motor, cognitive, and behavioral performance following unilateral ventroposterior pallidotomy for Parkinson disease. Arch Neurol 1998;55:1201-8. Kopyov O, Jacques D, Duma C, Buckwalter G, Kopyov A, Lieberman A, et al. Microelectrode-guided posteroventral medial radiofrequency pallidotomy for Parkinson’s disease. J Neurosurg 1997;87:52-9. Uitti RJ, Wharen RE Jr, Turk MF. Efficacy of levodopa therapy on motor function after posteroventral pallidotomy for Parkinson’s disease. Neurology 1998;51:1755-7. Kumar R, Lozano AM, Montgomery E, Lang AE. Pallidotomy and deep brain stimulation of the pallidum and subthalamic nucleus in advanced Parkinson’s disease. Mov Disord 1998;13 Suppl 1:73-82. Melnick ME, Dowling GA, Aminoff MJ, Barbaro NM. Effect of pallidotomy on postural control and motor function in Parkinson disease. Arch Neurol 1999;56:1361-5. Lai EC, Jankovic J, Krauss JK, Ondo WG, Grossman RG. Long-term efficacy of posteroventral pallidotomy in the
51.
52. 53. 54.
55.
56.
57.
treatment of Parkinson’s disease. Neurology 2000;55: 1218-22. Obeso JA, Linazasoro G, Rothwell JC, Jahanshahi M, Brown R. Assessing the effects of pallidotomy in Parkinson’s disease. Lancet 1996;347:1490. Delong MR. Why pallidotomy should not be abandoned. Moving Along 2001;3:3. Benabid AL. Why should we abandon pallidotomy? Moving Along 2001;3:3-5. Blomstedt P, Hariz MI. Are complications less common in deep brain stimulation than in ablative procedures for movement disorders. Stereotact Funct Neurosurg 2006;84:72-81. Voon V, Kubu C, Krack P, Houeto JL, Tro¨ster AI. Deep brain stimulation: neuropsychological and neuropsychiatric issues. Mov Disord 2006;21 Suppl 14: S305-S327. Marsden CD, Obeso JA. The functions of the basal ganglia and the paradox of stereotaxic surgery in Parkinson’s disease. Brain 1994;117:877-97. Brown P, Eusebio A. Paradoxes of functional neurosurgery: clues from basal ganglia recordings. Mov Disord 2008;23:12-20.
107 Pathophysiology of Dystonia J. A. Bajwa . M. D. Johnson . J. L. Vitek
Introduction First coined by Hermann Oppenheim in 1911, dystonia is a neurological disorder which manifests as sustained muscle contractions that frequently cause twisting and repetitive movements or abnormal postures. Dystonia can be classified by etiology (primary and secondary), age (early and late onset), and affected body region (focal, multi-focal, segmental, hemidystonia and generalized). When the cause is genetic or not recognized, the dystonia is considered to be primary but when the cause is identifiable (as in the case of stroke, brain trauma, or metabolic disease), it is referred to as secondary [1,2]. The majority of people with generalized dystonia first exhibit symptoms in childhood and adolescence that progressively worsen within the first 5 years after onset, and then remain affected throughout their lifetime [3]. Patients with cervical dystonia or spasmodic torticollis tend to develop the disorder later in life and while it may spread to adjacent body parts in some patients it is more likely not to generalize. The genetic etiology for primary dystonia has been linked to DYT gene mutations. A three pair gene deletion (CAG) has been identified for TOR1A (DYT1) gene at the 9q34 locus [4]. To date, there are fourteen other dystonic disorders classified under DYT gene mutations. All of these disorders are autosomal dominant except for DYT2 which is autosomal recessive and DYT3 (Lubag disease) which is X linked [5,6]. Secondary causes of dystonia include malfunction of specific anatomical pathways [7], drugs, especially dopamine receptor blockers and neuroleptics [8], and brain lesions such as those produced by #
Springer-Verlag Berlin/Heidelberg 2009
stroke [9], anoxia [10], trauma [11,12], or those associated with neurodegeneration [13]. There is also increasing recognition that repetitive movements of certain body parts can be a risk factor for focal dystonias such as writer’s cramp and musician’s dystonia including embouchure dystonia [14–16]. Dystonia has traditionally been thought of as a hyperkinetic movement disorder originating from abnormal sensorimotor integration within the basal ganglia thalamo-cortical circuit [17,18]. Lesions within this pathway (especially in the putamen) can lead to the appearance of dystonic movements, but symptoms do not typically emerge until weeks or months after the initial injury. This suggests that dystonia may, in some cases, be triggered by a pathological reorganization of specific brain regions and not necessarily by the direct effects of a degenerated pathway in the brain [19,20]. Several other physiological hallmarks of dystonia have been proposed. Hallett and colleagues have argued that dystonia derives from impaired inhibition at multiple levels of the central nervous system [20]. Mink and others have suggested that the co-contraction of agonist and antagonist muscles that occurs in dystonia derives from disturbed ‘‘surround’’ inhibition of competing motor programs during voluntary movements [21]. Abnormal sensory plasticity within brain regions encoding the functional representations of affected body parts may also contribute to the development of dystonia [22,23]. The goal of this chapter is to review the studies investigating the neurophysiological basis for dystonia with particular emphasis on the brain structures that exhibit abnormal activity
1780
107
Pathophysiology of dystonia
and how these disturbances translate within the sensorimotor network to the development of dystonia. Future studies will need to address whether dystonias that affect different parts of the body are caused by different pathophysiological mechanisms or the same pathophysiological mechanism in somatotopic portions of different regions of the brain. Determining how changes in neuronal activity in these areas lead to dystonia will help us to refine current and provide new therapeutic approaches for the treatment of dystonia that are consistently effective for all forms of this disorder.
Functional Organization of the Sensorimotor Circuit The basal ganglia are organized into a family of functionally segregated circuits, including motor, oculomotor, associative, and limbic [24]. These circuits take origin from different cortical areas, project to separate portions of the basal ganglia and thalamus, and return to areas of the frontal cortex from which they took origin. Of these circuits, the ‘‘motor’’ circuit is most closely associated with the development of dystonia [25].
Basal Ganglia – Thalamocortical ‘‘Motor’’ Circuit As shown in > Figure 107-1, the motor circuit includes excitatory projections from precentral motor and postcentral somatosensory cortex that target the sensorimotor portions of striatum (primarily the putamen) [26–28]. Striatal projection neurons in turn send inhibitory efferents that follow one of two routes. The ‘‘direct’’ pathway consists of striatal GABAergic neurons that primarily target the globus pallidus pars interna (GPi) and substantia nigra pars reticulata (SNr),
and that exhibit excitatory responses to dopamine by means of a dopamine D1 receptor binding mechanism. In contrast, dopamine produces an inhibitory effect on neurons along the ‘‘indirect’’ pathway through D2-like receptor activation. The indirect pathway consists of striatal GABAergic neurons projecting predominately to neurons in the globus pallidus pars externa (GPe). GPe neurons in turn send inhibitory projections to the subthalamic nucleus (STN) and GPi. Projections to GPi from GPe largely arise from axon collaterals of projections from GPe to STN. The GPi also receives excitatory input from the STN, a nucleus modulated by the cortex, GPe, pedunculopontine nucleus (PPN) and thalamic subnuclei including the centromedian and parafascicular nucleus (CM/Pf, respectively). Most of the outputs of the basal ganglia arise from GPi which sends inhibitory projections to the PPN in the brain stem [29,30] and subnuclei of the thalamus including ventralis lateralis pars oralis (VLo) and ventralis anterior (VA) [31,32]. These areas of thalamus in turn send excitatory projections to the supplementary motor area (SMA) and premotor areas (PM), respectively but also have smaller projections to the primary motor and arcuate premotor cortices [33–36]. The motor region of the GPi also sends inhibitory projections to the CM/Pf of thalamus that project back to the striatum [37] such that CM projections primarily target putamenal neurons involved in the direct pathway [38]. Motor regions of SNr also sends inhibitory projections to the thalamus projecting largely to ventralis anterior magnocellularis (VAmc), which sends excitatory projections to the prefrontal cortex [39]. GPi and SNr also have smaller inhibitory projections to the midbrain tegmentum and the superior colliculus [40,41]. Within the motor circuit of the basal ganglia, somatotopic organization is preserved at each nodal point in the motor circuit. In addition anatomically segregated motor subcircuits have
Pathophysiology of dystonia
107
. Figure 107-1 Sensorimotor networks, including the predominant basal ganglia thalamo-cortical and cerebellar thalamo-cortical pathways, have been implicated in the pathophysiology of dystonia. Synaptic terminal shape (square, diamond, circle) signifies the type of neurotransmitter involved, whereas the size of the shape reflects the degree of axonal collateralization in the target nucleus. Within the basal ganglia, line thicknesses represent proportions of each type of projection neuron. Abbreviations are as follows for the cortex (M1: primary motor cortex, PM: premotor cortex, S1: primary somatosensory cortex, SMA: supplementary motor area); the basal ganglia (GPe: globus pallidus pars externa, GPi: globus pallidus pars interna, SNc: substantia nigra pars compacta, SNr: substantia nigra pars reticulata, STN: subthalamic nucleus); the thalamus (CM: centromedian nucleus, Pf: parafascicular nucleus, R: reticular formation of thalamus, VA: ventralis anterior, VLc: ventralis lateralis pars caudalis, VLo: ventralis lateralis pars oralis, VPLo: ventralis posterolateralis pars oralis); the cerebellum (DN: dentate nucleus, FN: fastigial nuclei, IH: intermediate hemisphere of cerebellum, IN: interposed nuclei, LH: lateral hemisphere of cerebellum, V: vermis); and the brain stem (PN: pontine nucleus, PPNc: caudal pedunculpontine nucleus, PPNd: dorsal pedunculopontine nucleus)
been identified [35]. Each motor subcircuit takes origin in different motor and premotor cortical areas (motor cortex, supplementary and arcuate premotor cortex) and involves different portions of the basal ganglia and thalamus [35,42]. Although their exact function remains to be defined, these motor subcircuits have been proposed to play a differential role in normal motor control and in the development of the abnormal movements present in hyperkinetic and hypokinetic movement disorders [43].
Scaling Versus Focusing That the basal ganglia mediate primary dystonia and many forms of secondary dystonia has not been questioned, how it does so however, remains unclear. To understand the potential role of the basal ganglia in dystonia, it is important to briefly review current hypotheses concerning the role of these structures in planning, executing, and regulating movements. Based on the above anatomical model, several theories
1781
1782
107
Pathophysiology of dystonia
concerning the role of the basal ganglia in motor control have been proposed. One theory proposes that neural activity within the basal ganglia–thalamocortical motor circuit acts to scale movement. Scaling of movement is thought to occur by a combination of inhibition of GPi/ SNr neurons via the direct pathway and excitation of these same neurons by means of the indirect pathway. Inhibition of these output neurons would facilitate movement by disinhibition of thalamocortical projections excitatory to the cortex. Activation of the same GPi/SNr neurons would lead to inhibition of movement by inhibiting these same thalamocortical projections [44,45]. The balance between excitatory and inhibitory inputs to GPi/SNr output neurons would then modulate the amount of disinhibition of thalamocortical neurons, providing a mechanism by which movement could be scaled. A second hypothesis proposes that this circuit focuses motor activity. Focusing of motor activity is proposed to occur by means of activation of the direct pathway, reducing GPi/SNr output, thereby disinhibiting movement-facilitating thalamic neurons, ultimately resulting in activation of prime mover muscles and execution of the desired movement. At the same time, activation of different GPi/SNr neurons by the indirect pathway would increase GPi/SNr output, thus increasing inhibition on a separate population of thalamocortical neurons that suppress antagonistic movements [21,46]. Based on this hypothesis of basal ganglia function, phasic reductions of GPi/ SNr output by means of the direct pathway facilitate cortically initiated movements by disinhibition of thalamus, whereas phasic increases in GPi/ SNr output by means of the indirect pathway act to inhibit antagonistic or unwanted movements. A major function of the basal ganglia according to this hypothesis is to inhibit competing motor commands and the inability of the basal ganglia to do that results in hyperkinetic movements and/or the co-contraction of agonist and antagonist muscle activity [21].
Neurophysiology of Dystonia Network Dysfunction Imaging studies, while at times contradictory, have suggested that multiple regions of the brain exhibit abnormal activity in both patients with genetic predispositions for dystonia and patients expressing dystonic motor signs. Useful imaging modalities in the study of dystonia have included magnetic resonance imaging (MRI), functional magnetic resonance imaging (fMRI), diffusion tensor imaging (DTI), positron-emission tomography (PET) and single-photon emission computed tomography (SPECT). We will briefly discuss the role of imaging in two important etiological forms of dystonia: DYT1 dystonia and focal dystonia.
DYT1 Dystonia There are limited reports of neuropathological examination of patients with DYT1 dystonia. Most have reported no pathological changes in the brains of DYT1 patients [47]. However, a recent study of four patients has revealed perinuclear inclusion bodies in the PPN, cuneiform nucleus, and griseum centrale mesencephali, which stain positive for TorsinA, ubiquitin and nuclear envelope protein laminin A/C [48]. In the absence of any significant neuropathological abnormality in DYT1 dystonia, functional neuroimaging is valuable in defining disease pathophysiology. Diffusion tensor imaging, which is used to analyze fiber tracts in the brain by measuring water diffusivity, has shown white matter changes within sensorimotor cortex in asymptomatic DYT1 carriers [49]. FDG PET studies in patients at rest, during wakefulness and in sleep have shown that DYT1 dystonia is associated with hypermetabolism in a network including SMA, the lateral cerebellum
Pathophysiology of dystonia
and posterior putamen extending into the pallidum [50,51]. The same finding has been discovered in non-manifesting carriers of the DYT1 gene, in DYT6 patients, in patients with essential blepharospasm, and in ungenotyped primary generalized dystonia (PGD) patients [50–53]. In DYT1 dystonia, hypermetabolism in a network including the midbrain, cerebellum and thalamus is observed only during dystonic movements [50]. PET studies during motor tasks in patients with DYT1 dystonia have consistently found hypermetabolism in the SMA, cerebellum and the basal ganglia and decreased metabolic activity in the contralateral sensorimotor cortex [54,55].
Focal Dystonia Certain forms of focal dystonia are thought to derive from impaired sensory processing, as evident by the clinical observations of geste antagoniste (‘‘sensory trick’’) [56] and vibrationinduced posturing [57]. Results from cortical PET and fMRI studies during dystonia-inducing motor tasks in patients with focal dystonia have been less consistent, with some studies reporting decreased activation of premotor areas and the SMA [58–60], while others showed enhanced activation in these regions [60]. However, there is evidence to suggest disrupted signaling in the basal ganglia – thalamocorticol circuit in patients with cervical dystonia and writer’s cramp [51, 58, 62]. As we shall see in the following sections, imaging studies that show altered metabolic activity often represent multiple disease processes that occur simultaneously in a particular nucleus. Some of these processes have been revealed to us using electrophysiological and neurochemical techniques. We will discuss these findings for each nodal point within the subcortical-cortical circuit, beginning first with the primary input structure to the basal ganglia – the striatum.
107
Striatum Structural Lesions in the Striatum Associated with Dystonia Striatal lesions, especially those extending into the posterior parts of putamen, often lead to the delayed emergence of dystonia [63–65]. A comprehensive study of 240 patients with caudate nucleus, putamen, and/or globus pallidus lesions reported that 64% of the patients with lesions affecting primarily the putamen exhibited dystonia, whereas only 9% of patients with lesions selective to the caudate showed signs of dystonia [66]. In non-human primates, selective unilateral lesions of the posterior (sensorimotor) putamen were reported to trigger dystonic posturing [67]. Yet, when those lesions were accompanied by ablation of the anterior (associative) putamen, dystonic movements were no longer present, suggesting that some dystonias may depend on the interplay between motor and associative pathways through the basal ganglia thalamo-cortical network. Decreased putamenal volume, however, does not appear to be a generalizable feature for all forms of dystonia. In fact, the volume of putamen in patients with idiopathic focal dystonia was actually shown to be larger (by 10%) than in control patients [68].
Abnormal Striatal Activity in Dystonia Studies have reported increased glucose metabolism in the putamen of patients with spasmodic torticollis, [69] and decreased GABA in the putamen of patients with focal dystonia [70], which suggest a deficiency in intrastriatal inhibition. An fMRI study reported that both putamen and caudate activity remained abnormally high in focal hand dystonia patients even after the cessation of a finger-tapping task and after the return of sensorimotor cortical activity
1783
1784
107
Pathophysiology of dystonia
to a baseline level [71]. However, increased striatal metabolism may not be indicative of all types of dystonia. Striatal metabolism levels were higher in DYT1 patients than in DYT6 patients [72], and striatal hypermetabolism was observed in DYT1 patients who did and did not express sustained muscle contractions [50], leaving one to question whether or not there may be a causal relationship between the presence of increased metabolic activity in the lentiform nucleus and the development of dystonia. Recent studies have instead focused on the implications of impaired striatal interneuron activity, which has been hypothesized to adversely affect the integration of sensorimotor input.
Disinhibition of Striatal Output Through Decreased Interneuron Activity Cholinergic interneuron signaling in the striatum appears to be dysfunctional in some forms of dystonia [73]. Though these interneurons represent only 1–2% of the total neuronal population in the striatum [74], cholinergic interneurons have widely distributed processes targeting spatially diverse networks of medium spiny neurons projecting to the pallidum. Cholinergic interneurons are specifically modulated by sensory stimuli that bring about a particular behavior and are involved in long-term plastic changes within the striatal network [75–76]. Consequently, abnormal cholinergic signaling would be expected to have broad network effects in the striatum [77]. Experimental studies in mouse models of earlyonset torsion dystonia (DYT1) have found that striatal cholinergic interneurons are more sensitive than in control mice to D2-like agonists, which have an excitatory effect on these cells [78]. Patients with spasmodic torticollis exhibit decreased vesicular acetylcholine transporter binding [73]. Vesicular acetylcholine transporter binding is indicative of cholinergic signaling
within the striatum and decreased levels of this transporter may suggest a decrease in the density of striatal cholinergic interneurons. Impaired intrastriatal inhibition may also stem from the malfunction of other types of interneurons. In dtsz hamsters expressing agerelated paroxysmal dystonia, no difference in the density of striatal cholinergic interneurons was evident when compared with control animals [79]. Instead, in this model of dystonia, a significant decrease in the number of striatal parvalbumin (PVþ) GABAergic interneurons was observed, which correlated with the severity of paroxysmal symptoms [80]. Unique to this model is the eventual remission of symptoms as the hamster ages, which was shown to parallel the maturation of the PV þ interneurons [81]. The lack of striatal inhibition was also shown to lead to a decrease in firing rate and abnormal patterns of activity in the entopeduncular nucleus (rodent homologue of GPi) [82]. Consistent with these studies, Yamada et al. showed that injection of the GABAA antagonist, bicuculline, in the putamen of cats could elicit dystonic head movements, which paralleled a strong inhibition of neuronal spike activity in the SNr [83].
Altered Striatal D2 Receptor Expression Dysfunction of the nigrostriatal dopaminergic pathway has been implicated in dystonia [84]. Pharmacological agents that block dopamine receptors are known to induce acute dystonia [85,86], and transient hemidystonia has been observed following unilateral injections of the dopaminergic cell neurotoxin MPTP (1-methyl 4-phenyl 1,2,3,6-tetrahydropyridine) in nonhuman primates [87]. In terms of the latter, it was hypothesized that the combination of low striatal dopamine levels immediately following MPTP-intoxication (66% reduction) and a transient down-regulation of D2-like receptors
Pathophysiology of dystonia
selectively disinhibited the indirect pathway [88]. According to the focusing hypothesis, abnormal firing patterns along the indirect pathway would encumber the suppression of unwanted involuntary movements during the execution of a motor program [87,89]. This may explain why clinical manifestations of dystonia are more commonly associated with the initiation of movements and less commonly occur at rest. Imaging techniques have been useful to investigate whether disturbances in dopaminergic signaling trigger changes in levels of expression of D1 and D2 receptors. PET measurements showed lower putamenal binding of [18F]spiperone, a dopaminergic radioligand relatively selective for D2-like receptors, in patients with idiopathic focal dystonia than in normal controls [90]. Similar decreases in D2-like binding in the putamen have been reported in patients with DYT1 dystonia [91] and cervical dystonia [92]. However, opposite effects on D2-like receptors have been observed in patients with doparesponsive dystonias (DRD). DRDs are characterized by mutations in the genes encoding GTP cyclohydrolase 1 or tyrosine hydroxylase, which in turn reduce the synthesis of catecholamines, including dopamine and serotonin [93]. In studies with DRD patients, D2 receptor binding to the radioligand [11C]raclopride in the putamen was shown to be elevated [94,95], whereas D1 and DAT binding was not significantly different from controls [96]. While the changes in D2 receptor expression levels have been assumed to affect the principal medium spiny neurons, certain interneurons within the striatum, including cholinergic interneurons that primarily express D2-like receptors, may also contribute to the loss of inhibition and subsequent lack of focusing of striatal output.
Summary Dysfunction in the striatum has been linked to both primary and secondary dystonia [17,50,84].
107
Recent experimental evidence in patients with dystonia as well in animal models of dystonia suggests that disinhibition of striatal neurons projecting to the globus pallidus may enable the co-activation and sustained contractions of agonist and antagonist muscles. Multiple mechanisms appear to generate abnormal striatal disinhibition, which may explain how different etiologies of dystonia result in the manifestation of similar motor signs.
Globus Pallidus Pallidal Lesions can Generate Hyperkinetic and Hypokinetic Conditions In some cases, dystonia has been linked to selective unilateral [10] or bilateral [97] lesions of the globus pallidus. According to the scaling theory of the basal ganglia, lesions in the GPe would be expected to further disinhibit and alter the discharge patterns of STN and GPi, leading to decreased thalamocortical and PPN activity and the emergence of hypokinetic movements. Indeed, patients with selective GPe lesions show akinetic-rigid abnormalities [98,99], which in the parkinsonian state can exacerbate motor symptoms. Lesions of the GPi, on the other hand, would be expected to disinhibit thalamocortical and PPN activity, thereby facilitating the development of hyperkinetic movements including dystonia. Two studies reported that dystonia was present when lesions affected the GPi in particular [10,97], but these cases are difficult to reconcile with the fact that GPi lesions are also effective at alleviating dystonia in patients with primary dystonia [100,101]. As such, it would appear that pallidal lesions can have opposite effects depending on which segment of the pallidum is lesioned, what neural elements are damaged, and the pathological state of the sensorimotor network.
1785
1786
107
Pathophysiology of dystonia
Decreased Pallidal Spike Rates One of the first models proposed to explain the development of movement disorders was based on changes in mean discharge rate occurring at key nodal points in the basal ganglia – thalamocortical circuit [44,102]. The ‘‘rate’’ model of Parkinson’s disease, for example, postulated that depletion of dopaminergic signaling from the SNc would produce an increase in GPi activity, excessive inhibition of the pallidal receiving areas in thalamus, and a corresponding reduction in thalamocortical activity resulting in slower and/ or fewer movements [44,102]. According to the rate model, the development of hyperkinetic disorders would occur from opposite changes at these nodal points including reduced mean discharge rates in STN and GPi and increased mean discharge rates in GPe and VLo. Although proposed originally to explain the development of chorea [44], the hypothesis soon became adapted to explain the development of dystonia [102]. In support of the rate model, we and others have reported a reduction in the mean discharge rate of neurons in the GPi of patients with dystonia and hemiballismus [103–106]. Similarly, studies with humans and animal models of Parkinson’s disease have shown that the mean discharge rate in GPi is reduced and in GPe is increased during drug-induced dyskinesias [107,108]. Although some have argued that the reduction in mean discharge of GPi neurons derives from the use of anesthetic agents during recording [109], a number of studies have now demonstrated that such reductions are also present in the absence of anesthetic agents [110,111]. GPi firing rates in dystonia patients undergoing microelectrode mapping for deep brain stimulation or for placement of a lesion (i.e., pallidotomy) have also been reported to exhibit an inverse relationship to the severity of dystonia (BFMDRS score) [18,103,105]. Lenz et al. also observed that the mean discharge rates recorded from GPi neurons later in the surgical mapping
procedure as the patient’s dystonic symptoms worsened were lower than those collected in the earlier stages of the procedure when the patient was less dystonic [112]. Unlike recordings in GPi, pathological changes in GPe neuron firing rates are less certain with dystonia. While most studies show parity in firing rate between pallidal segments in patients with dystonia, some studies have documented significant reductions in mean discharge rates of GPe neurons [106,113], whereas others have found little [103] to no change [114,115]. One hypothesis to account for the variability among these studies is that only the affected region of the pallidal complex expresses pathological firing rates. When grouping recordings from patients with different types of dystonia (e.g., generalized, cervical, focal, etc.), abnormal firing rates could get washed out by relatively normal rates in other regions that are not directly involved in the pathology, i.e., regions of the pallidum whose somatosensory responses are related to unaffected body part(s). It is also possible that different types of dystonia have different underlying pathophysiologies or that the recordings in these studies reflect different degrees of severity in dystonia. Although compelling in its simplicity the rate model failed to explain the improvement in dystonic symptoms, hemiballismus, and druginduced dyskinesia following pallidal lesions, a procedure that would further reduce the output from this region. More emphasis has now been placed on abnormal patterns of activity rather than alterations in mean discharge rates to explain the pathophysiological mechanisms of dystonia [18,103,106,112,116,117].
Increased Pallidal Spike Bursts and Oscillations In the pallidum of dystonic patients, most neurons do not discharge tonically as reported in normal monkeys but fire in irregularly
Pathophysiology of dystonia
grouped discharges with intermittent bursts [103,106,109,111,113,115–118]. Similar bursting activity has been observed in the entopenducular nucleus of mutant dtsz hamsters, but only when paroxysmal dystonia was present, disappearing when dystonic symptoms resolved [119]. These pallidal burst patterns in dystonia are similar to those reported in patients with Parkinson’s disease [103,111]. Starr and colleagues have hypothesized that the differences between dystonia and PD are due to the same changes in bursting activity superimposed on a lower mean discharge rate in the case of dystonia and higher mean discharge rate in the case of PD [103] (> Figure 107-2a). Another group recording neuronal activity from dystonic and PD patients, however, reported that the
107
GPi of patients with cervical dystonia had more irregular discharge patterns, more frequent and longer pauses, and more bursting activity than in the GPi of patients with PD [111]. The long pauses between bursts were also observed within the GPi of a patient with hemiballismus [106]. In this study, the duration of pauses between bursts were strongly correlated with EMG activity in the biceps muscle, whereas the duration of GPi bursts were more strongly associated with EMG activity in the triceps muscle (> Figure 107-2b). Given the strong correlation between GPi and EMG neuronal activity in this patient it was suggested that there was likely an increase in synchronization of GPi neurons in this patient. Taken together, these data suggest that increased synchronization and the temporal code of GPi
. Figure 107-2 (a) GPi activity in patients with dystonia consisted of increased burstiness superimposed on an overall reduced discharge rate. Spike recordings in the GPi of PD patients exhibited similar burstiness except with higher overall discharge rates as compared with recordings in normal monkeys. (Modified with permission from The American Physiological Society [103]). (b) GPi firing patterns were shown to correlate with EMG activity in a hemiballistic patient. Pauses in GPi neuron discharge were associated with biceps muscle activity, whereas GPi spike bursts were related to triceps muscle activity. (Modified with permission from John Wiley & Sons, Inc. [106])
1787
1788
107
Pathophysiology of dystonia
neurons have a robust relationship to the expression of disordered movements and muscle activity in dystonia and hemiballismus. Oscillatory field potentials within the pallidum, time-locked with neuronal spike activity [120], have also been observed in dystonia patients [121–123]. In comparison to PD patients, dystonia patients were reported to have pallidal oscillations with less power between 11 and 30 Hz and greater power between 4 and 10 Hz [116,121]. These oscillations have also been reported to be minimal when dystonia patients are at rest, but significantly increased during dystonic movements, suggesting that enhanced low frequency oscillations within the globus pallidus were associated with the development of phasic but not necessarily tonic dystonia. Indeed, multiple studies have described a strong correlation of low frequency oscillatory activity in the GPi to EMG activity in the dystonic limb [116,122], whereas patients performing a sensory trick to relieve cervical dystonia have been found to express desynchronized GPi activity in the 6–8 Hz frequency band [123]. While it is difficult to demonstrate whether the globus pallidus drives EMG activity or the involuntary movements drive GPi activity, Sharott et al. used a statistical tool known as a directed transfer function to argue that there is a higher likelihood oscillations within the globus pallidus drive muscle activity [122].
Reorganization of Functional Receptive Fields in the Pallidum The distribution of abnormal neuronal activity within the pallidum was reported to be different for focal versus generalized dystonia in that those patients with generalized dystonia had a broad distribution of changes throughout the pallidum, while those patients with focal dystonia including patients with cervical dystonia and patients with hand dystonia had changes restricted to a smaller portion of the pallidum [117]. Another
study also noted that the observed neuronal changes in patients with dystonia were restricted to the more ventral portions of GPi, a region previously found to contain a higher proportion of cells related to neck movement [124]. These findings would be consistent with the hypothesis that different phenotypes of dystonia are the result of changes in not only neuronal firing pattern, but also altered receptive fields of neurons found within certain somatotopic regions of the GPi. In a single case report, Lenz and colleagues reported significantly more GPi cells that were responsive to sensory stimulation in dystonia (53%) compared with patients with hemiballismus (13%), both of which are hyperkinetic disorders [112]. We also have observed widened receptive fields and cells responding to movement of multiple joints in patients with dystonia as compared with those recorded in normal monkeys [106]. In the dystonia patients, 43% of GPi neurons were modulated by passive manipulation or active movement and of these, 68% also responded to multiple joints in one or more limbs. On the other hand, only 1/34 GPi cells examined in a hemiballistic patient responded to passive limb manipulation. A recent study by Chang et al, however, has challenged the view that widened receptive fields of GPi neurons contribute to the development of dystonia. In their study they reported that somatotopic representations were not distorted in the basal ganglia of dystonia patients and therefore argued that the basal ganglia did not contribute to alterations in sensory processing observed at the level of the cortex in patients with dystonia [118]. In this study a comparison of somatosensory responses in the limb of patients with generalized dystonia was compared to those in craniocervical dystonia where the limb was defined as clinically normal. There were no significant differences found between these two patient groups in terms of the proportion of multi-joint responses or the volume of the pallidum represented by each body region. As a result they
Pathophysiology of dystonia
concluded that abnormal body map representations reported previously for the sensorimotor cortex and putamen were not present at the level of the internal pallidum. These findings are supported by a study by Hutchison and colleagues who also reported a low incidence of multi-joint responses (only 2 cells out of >200 cells examined) in dystonia patients and only 12% of cells with a somatosensory response [109]. However, they did report an increase in the proportion of cells that decreased activity in response to movement and suggested that this was consistent with ‘‘movement-related overactivity in thalamocortical pathways.’’ Although these studies question the role of widened receptive fields and distorted body maps at the level of the pallidum in the development of dystonia, these discrepancies can be explained by differences in the design of the study. First is the comparison to the ‘‘normal’’ limbs of patients with craniocervical dystonia. It is rare to find multi-joint movements in normal primates, yet multi-joint movements were found as frequently in the patients with cranciocervical dystonia as they were in patients with generalized dystonia suggesting the ‘‘clinically normal’’ limb may have some abnormalities not detected clinically. Eidelberg demonstrated that asymptomatic carriers of the DYT1 gene exhibit abnormalities in cortical activity, suggesting that even in the ‘‘normal’’ state, basal ganglia dysfunction is present and that these changes propagate through the neural network [125]. We would further suggest that such abnormalities may express themselves phenotypically only when a sufficiently large population of cells are affected. Thus, one may find changes in neuronal activity that may not be associated with the expression of dystonia until a sufficiently large number of neurons are affected, i.e., the manifestation of dystonia may be dependent on reaching a critical number of affected cells in order to generate a large enough change in downstream structures to manifest the movement disorder.
107
Summary In primary dystonia, and perhaps in secondary dystonia, the pallidum expresses disordered activity characterized by changes in rate, bursting, lowfrequency oscillations (especially in the 4–10 Hz frequency range), and distortions in the functional somatotopy of GPi neurons. These features have been proposed as the predominant physiological factors that contribute to the disruption of function in the thalamocortical pathway and the development of dystonic motor signs.
Thalamus Thalamic Lesions and Dystonia Thalamic infarcts have also been linked to the manifestation of dystonia, though lesions in the lentiform nuclei appear to have a much stronger correlation [17]. In a series of 22 patients with selective lesions of the thalamus, 13 exhibited abnormal involuntary movements, which were contralateral to the lesion and characteristic of dystonia, myoclonus, and/or tremor [126]. Most commonly, lesions associated with dystonia were centered in the cerebellar-receiving area of thalamus (ventralis intermedius, [Vim]) and the somatosensory nucleus of thalamus (ventralis caudalis, [Vc]). Interestingly, similar to the GPi, lesions within the Vim and Vop can also produce an immediate decrease in phasic dystonia [127].
Abnormal Thalamic Activity in Dystonia Imaging studies indicate that the motor thalamus is hyperactive in patients with primary cervical dystonia [128], consistent with the rate model hypothesis that there is decreased inhibitory output from the basal ganglia to the thalamus in patients with dystonia. Specific manifestations
1789
1790
107
Pathophysiology of dystonia
of dystonia, however, may depend on the particular subnuclei of the motor thalamus that are impaired. A non-human primate study by Guehl et al. [129] found that injections of the GABAA antagonist bicuculline into the pallidal-receiving area of thalamus (VLo) resulted in severe dystonic postures, whereas GABAA antagonists infused within the cerebellar receiving area (VPLo) evoked myoclonic jerks. In a follow-up study [130], this group reported that discharge frequencies of neurons in the motor thalamus increased significantly and discharge patterns became burstier following the bicuculline injection in either VLo or VPLo. Burst patterns were found to correlate with myoclonic jerks. In human patients with primary and secondary dystonias, Vop and Vim neuronal discharge rates were reported to be highly correlated with the frequency of EMG activity [117]. In another study, Vim spike bursts in dystonia patients that correlated with EMG were reported to precede the onset of EMG activity [131]. Similar correlations between firing patterns of neurons in the somatosensory thalamus Vc and EMG activity were found to be less significant; however the activity in many sensory cells in Vc nevertheless preceded EMG activity, which was unexpected since Vc is a relay nucleus for somatosensory input from the periphery [127]. These findings emphasize that the pathophysiology of dystonia likely involves both motor and sensory thalamocortical pathways and that abnormalities in sensory processing may drive the motor dysfunction in at least some dystonias.
Overlapping Functional Representations in Thalamus
in the Vc as compared with similar experiments in essential tremor patients. Cerebellar relay nucleus neurons have also been shown to exhibit widened receptive fields with increased representational area of dystonic parts as compared with patients with essential tremor [127,131]. Microstimulation in the Vim was found to evoke contractions in multiple muscles simultaneously, which is similar to the co-contractions of agonist and antagonist muscle activity in dystonia [127]. Neurons in the pallidal receiving area, VLo, were also reported to be more likely to respond to passive limb movements about multiple joints, suggesting that the distorted somatotopic maps present in the GPi are also present within the thalamus in dystonia [130]. Together, this disruption of sensorimotor processing propagating through the thalamus has been hypothesized to contribute to the lack of focusing and overflow of EMG activity to nonprime mover muscles leading to the erratic movements and loss of motor control associated with dystonia.
Summary Similar to pallidum, neuronal recordings within the sensorimotor thalamus have indicated that dystonia is associated with increased bursting activity and a somatotopic reorganization of receptive fields in thalamic subnuclei. Both pallidal and cerebellar receiving areas of motor thalamus as well as somatosensory regions appear to exhibit these changes suggesting that different dystonic movements may be generated depending on the relative changes within each region of the thalamus.
Cerebellum Neuronal recordings in the cutaneous core of Vc have been reported to have increased responsiveness to multiple parts of the body in dystonia patients [23]. In this study, dystonia patients reported a greater degree of sensation in multiple parts of the body during electrical stimulation
Structural Abnormalities Associated with Dystonia The cerebellum is not conventionally considered to be a pathogenic source for dystonia, but
Pathophysiology of dystonia
lesions affecting the cerebellum and olivocerebellar pathway have been associated with dystonia, particularly cervical dystonia [132]. Cervical dystonia has also been linked to the presence of cerebellar tumors through autopsy studies, and removal of cerebellar tumors resulted in improvement in motor symptoms [133,134]. Similarly, cerebellectomy in the dystonic rat model eliminated dystonic movements [135]. Implication of a role for cerebello-thalamic pathways in the development of dystonia is also derived from the observation that thalamic lesions associated with the development of dystonia tend to occur in the cerebellar-receiving areas of thalamus and not those directly targeted by basal ganglia projections [136].
Imaging Abnormalities Associated with Dystonia Eidelberg and colleagues have shown that DYT1 carriers have increased metabolic activity in the cerebellum when at rest irrespective of whether they exhibit the motor signs of dystonia [50]. However, during movement, affected gene carriers expressing dystonia showed higher metabolic activity in the cerebellum (as well as the lentiform nuclei and SMA) than non-affected DYT1 carriers. Hypermetabolism using [18F] FDG and PET has also been demonstrated in the cerebellum of patients with cervical dystonia when compared with normal controls [62]. Hypermetabolism in the cerebellum of patients with essential blepharospasm [52] has also been described using FDG-PET.
Altered Cerebellar Activity in Dystonia Abnormal cerebellar output has also been implicated in the emergence of dystonia in animal
107
models. In a genetic rat model of dystonia, deep cerebellar nuclei exhibited increased burstiness [137] and decreased GABA receptor density [138], presumably from increased GABAergic signaling at Purkinje cell synapses. Pizoli and colleagues also observed abnormal cerebellar signal processing in a kainic acid induced dystonic rat model in which neuronal activation was assessed by in situ hybridization for c-fos. C-fos mRNA expression was increased in the cerebellum, locus ceruleus, and red nucleus [139].
Integration of Cerebellar and Pallidal Circuits While the basal ganglia thalamo-cortical and cerebellar thalamo-cortical circuits have been considered anatomically segregated, Strick and colleagues have recently identified a disynaptic connection from the deep cerebellar nuclei (especially the dentate nucleus) to the striatum and GPe by way of several thalamic subnuclei, including the intralaminar nuclei and VA/VL [140]. These data support the argument that pathological activity in the cerebellum may directly affect basal ganglia activity and basal ganglia activity may affect the cerebellum. This finding could explain the observations that have implicated both the both basal ganglia and cerebellum in the pathogenesis of dystonia.
Summary While the cerebellum is likely to be involved in the production of dystonia as suggested by imaging and electrophysiological studies, it has been difficult to conclude whether changes in cerebellar activity reflect compensatory or causative processes given the lack of detailed electrophysiological studies investigating cerebellar activity in patients with dystonia [141].
1791
1792
107
Pathophysiology of dystonia
Cortex Hypermetabolism in the Cortex Several studies have used imaging techniques to investigate whether differences in cortical activity occur between patients with and without dystonia. It is important to consider that metabolic activity in the cortex of dystonia patients may depend on the type of dystonia and the setting in which the imaging was performed (i.e., whether a task was being performed, the type of task, the severity of dystonia), which may explain why many of the imaging studies in dystonia have seemed at first glance contradictory. In both manifesting and nonmanifesting DYT1 dystonia patients at rest, supplementary motor association areas exhibit abnormal activation [50]. Increased activity in SMA is consistent with higher firing rates in VLo, which sends its excitatory projections primarily to SMA. During dystonic movements, two other studies found hypermetabolism in the pre-supplementary motor cortex as well as in the contralateral premotor cortex, anterior cingulate, and sensorimotor cortex [72,142]. A study that took a series of 12 sequential PET scans in patients with writer’s cramp performing four different motor tasks (three scans per task) showed an increase in sensorimotor and premotor cortical activity as the dystonia in these patients progressively worsened during tasks [143]. Together, these studies suggest that metabolic activity in areas of cortex involved in movement execution and planning are different from that seen in the normal state in patients with dystonia and are consistent with dysfunction of intracortical inhibitory mechanisms [144].
Impaired Intracortical Inhibition Several studies have indicated that patients with focal dystonias have impaired intracortical
inhibition [145–148]. Paired-pulse transcranial magnetic stimulation (TMS) has provided a useful technique to examine intracortical inhibition in dystonia [149]. With this method, a subthreshold conditioning TMS pulse applied over motor cortex and paired with a second suprathreshold TMS pulse with a latency of less than 6 ms will normally suppress an EMG response to the second pulse. In dystonia, however, the EMG response to paired pulse stimulation is not inhibited. These findings are consistent with reduced cortico-cortical inhibition via local interneurons [145,148,150]. In support of this hypothesis, Levy et al. reported a decrease in GABA signaling in sensorimotor cortex contralateral (but not ipsilateral) to the dystonic hand [70]. Another study found that the cortical silent period, which is produced by a single TMS pulse and thought to be mediated by GABAB receptors [151], was reduced [150]. While it is possible the impairment to intracortical interneuron signaling derives from physiological alterations in interneurons themselves, another explanation is that there are changes in afferent activity to these interneurons that result in reduced intracortical inhibition. Voluntary movement, for instance, appears to reduce the excitability of inhibitory cortical neurons targeting the pyramidal projection neurons [152].
Abnormal Sensory Processing Previous studies have implicated the sensory system in the pathophysiology of focal dystonia [153–155]. Byl and colleagues reported that monkeys performing a repetitive movement task developed a focal dystonia coincident with an alteration of cortical receptive field representations in the somatosensory cortex of the working hand [153,156]. Broadening of sensory receptive fields [157] and abnormal sensory discrimination [158,159] have also been reported in humans with focal dystonia. Sanger and coworkers, in
Pathophysiology of dystonia
agreement with previous reports by Tinazzi and colleagues [160], observed changes not only in spatial discrimination, but also in temporal discrimination to sensory stimuli in patients with writer’s cramp [160,161]. Both primary and secondary sensory cortices in patients with focal dystonia appear to exhibit spatially overlapped representations of multiple body parts [162]. As shown in > Figure 107-3, fMRI representations of individual finger digits were distinct in normal controls, but overlapped in patients with focal hand dystonia. Hallett has argued that the sensorimotor nervous system is a predictive entity and abnormal processing of sensory information that occurs on multiple levels in focal dystonia might in turn impair the development of movement preparation paradigms [154]. Murase [163] and colleagues further suggest that the added sensory input of a ‘‘sensory trick,’’ which is a patient-specific technique to decrease motor signs of dystonia, might act to ‘‘rebalance the motor system’’ [163]. It is unclear, however, whether the rebalancing approach derives from sensory input directly into cortex, from the effect of this sensory information on basal ganglia and thalamic activity or whether it results from the movement itself, in effect as a result of corollary discharge.
Summary The pathophysiology of dystonia appears to involve loss of intracortical inhibition, abnormalities in sensory function, and changes in plasticity within the sensorimotor cortex. Whether one or more of these features are true for all types of dystonia remains to be determined.
A Neuron Model for Dystonia Consistent with previous hypotheses that striatal dysfunction underlies the development of primary
107
dystonia is the study by Albin et al. [73] who reported widespread reduction in [123I]Iodobenzovesamicol (IBVM) binding in the putamen and total striatum [73]. These data suggested that in primary dystonia there is a decrease in the density of striatal cholinergic interneurons. Such a loss could lead to increased responsiveness of striatal neurons to afferent input and account for the dynamic changes in neuronal activity observed through the basal ganglia – thalamocortical network (see > Figure 107-4). Differences in the rate and pattern of activity reported in downstream nuclei including the GPe and GPi could occur as the results of this altered processing of sensorimotor input to the striatum. While mean discharge rates in GPi are reduced, there also appears to be a reduction in mean discharge rates of neurons in GPe as well. Based upon the presumptive roles of the direct and indirect pathways in mediating various aspects of motor control, together with previous imaging studies and the observation of altered somatosensory responses of pallidal neurons, we contend that in primary dystonia, activity in both the direct and indirect pathways is increased. One also needs to account for the preponderance of bursting and low frequency oscillatory activity in patients with dystonia together with the fact that lesions in GPi that further reduce output from GPi still improve dystonic symptoms [100,101]. In the neuron model we described in 2002 [18], we hypothesized that the increasing severity and overflow of dystonia to antagonist muscle groups that occur during movement would be associated with a further reduction in mean discharge rates and increase in low frequency synchronized oscillations in neuronal activity in GPi. With the prevalence of low frequency oscillatory activity reported to occur in patients with dystonia, we would also propose that the development of low frequency uncontrolled synchronization of neurons in the pallido-thalamic circuit underlies the development of dystonic
1793
1794
107
Pathophysiology of dystonia
. Figure 107-3 Cortical representations generated from finger tip stimulation to digit 2 and digit 5 were found to be distinct in area 1 of sensory cortex for (a) control patients, but not for (b) patients with focal hand dystonia. (Reproduced with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc. [162])
movement. Furthermore, given the reported similarity of changes in neuronal activity in patients with PGD and CD [116,117], we would hypothesize that phenotypic differences between primary generalized dystonia and cervical dystonia arise as a result of differences in the distribution of these changes in the pallidum affecting portions of the somatotopically organized sensorimotor
pallidum with similar effects cascading throughout the network. Yet, multiple sensorimotor networks appear to be dysfunctional in dystonia. In addition to the changes in neuronal activity in the basal ganglia and thalamus [106,131], metabolic changes in the cerebellum [50], loss of inhibition of brainstem and spinal reflexes [20], and
Pathophysiology of dystonia
107
. Figure 107-4 Basal ganglia – thalamocortical network model for primary dystonia showing changes in firing rates and patterns between the (a) normal and (b,c) dystonic conditions. Abbreviations are the same as in > Figure 1 with the following addition: MEA, midbrain extrapyramidal area. Multiple lines of different lengths exiting a nucleus represent asynchronous neuronal activity; multiple broken lines of different lengths illustrate altered patterns of asynchronous neuronal activity; whereas multiple broken lines of the same length illustrate altered patterns of synchronous activity. The width of the lines depicts the amount of neuronal activity. Consistent with previous reports, striatal interneuron activity as well as thalamic and pallidal activity is reduced at rest. During movement, pallidal activity is further reduced, leading to an increase in thalamic activity and the development of uncontrolled synchrony throughout the subcortical-cortical network. This reduction leads to a disruption in cortical and brainstem output and the disordered movement that occurs in dystonia. This model is not encompassing of the changes in TMS and spinal reflexes, nor does it completely attempt to fully depict the changes in intrathalamic circuitry that contribute to these changes. This is left for further speculation. Modification of the present model will occur as new data concerning the neuronal activity changes in the pallido-thalamic and cerebello-thalamic circuits under the above conditions becomes available. Reproduced and modified with permission of John Wiley & Sons, Inc. [18]
excitability and metabolic changes in the cortex [150] have also been reported. Future neuron models of dystonia must account for each of these physiological changes whether one believes they are causal or epiphenomenal if the model is to accurately reflect the physiological state. And in so doing, increasing our understanding of the pathophysiological basis for the loss of focusing and scaling of muscle activity in dystonia will no doubt lead to improvements in
current therapies and to the development of new therapeutic approaches for this disabling movement disorder.
References 1. Fahn S, Bressman SB, Marsden CD. Classification of dystonia. Adv Neurol 1998;78:1-10. 2. Tarsy D, Simon DK. Dystonia. N Engl J Med 2006;355 (8):818-29.
1795
1796
107
Pathophysiology of dystonia
3. Greene PE, Kang UJ, Fahn S. Spread of symptoms in idiopathic torsion dystonia. Mov Disord 1995;10:143-52. 4. Ozelius LJ, et al. The early-onset torsion dystonia gene (DYT1) encodes an ATP-binding protein. Nat Genet 1997;17:40-48. 5. Khan NL, Wood NW, Bhatia KP. Autosomal recessive, DYT2-like primary torsion dystonia: A new family. Neurology 2003;61(12):1801-3. 6. Nolte D, Niemann S, Muller U. Specific sequence changes in multiple transcript system DYT3 are associated with X-linked dystonia parkinsonism. Proc Natl Acad Sci USA 2003;100(18):10347-52. 7. Gibb WR, Kilford L, Marsden CD. Severe generalized dystonia associated with a mosaic pattern of striatal gliosis. Mov Disord 1992;7(3):217-23. 8. Miller LG, Jankovic J. Sulpiride-induced tardive dystonia. Mov Disord 1990;5:83-84. 9. Alarcon F, et al. Post-stroke movement disorders: report of 56 patients. J Neurol Neurosurg Psychiatry 2004;75 (11):1568-74. 10. Munchau A, et al. Unilateral lesions of the globus pallidus: report of four patients presenting with focal or segmental dystonia. J Neurol Neurosurg Psychiatry 2000;69(4):494-8. 11. Frucht S, Fahn S, Ford B. Focal task-specific dystonia induced by peripheral trauma. Mov Disord 2000;15 (2):348-50. 12. Lee MS, et al. Dystonia after head trauma. Neurology 1994;44:1374-78. 13. Jankovic J, Tintner R. Dystonia and parkinsonism. Parkinsonism Relat Disord 2001;8:109-121. 14. Frucht S, Fahn S, Ford B. French horn embouchure dystonia. Mov Disord 1999;14(1):171-3. 15. Frucht SJ. Focal task-specific dystonia in musicians. Adv Neurol 2004;94:225-30. 16. Hallett M. Pathophysiology of writer’s cramp. Hum Mov Sci 2006;25(4–5):454-63. 17. Marsden CD, et al. The anatomical basis of symptomatic hemidystonia. Brain 1985;108(Pt 2):463-483. 18. Vitek JL. Pathophysiology of dystonia: a neuronal model. Mov Disord 2002;17 Suppl 3:S49-62. 19. Wichmann T, DeLong MR. Physiology of the basal ganglia and pathophysiology of movement disorders of basal ganglia origin. In: Watts RL, Koller WC, eds. Movement Disorders: Neurologic principles and practice. New York: McGraw-Hill, 1997; 87-97. 20. Hallett M. The neurophysiology of dystonia. Arch Neurol 1998;55:601-603. 21. Mink JW. The basal ganglia: focused selection and inhibition of competing motor programs. Prog Neurobiol 1996;50(4):381-425. 22. Garraux G, et al. Changes in brain anatomy in focal hand dystonia. Ann Neurol 2004;55(5):736-9. 23. Lenz FA, Byl NN. Reorganization in the cutaneous core of the human thalamic principal somatic sensory nucleus
24.
25.
26. 27.
28.
29. 30.
31.
32.
33.
34.
35. 36.
37.
38.
39.
40.
(Ventral caudal) in patients with dystonia. J Neurophysiol 1999;82:3204-3212. Alexander G, Delong M, Strick P. Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Ann Rev Neurosci 1986;9:357-81. Wichmann T, DeLong MR. Functional and pathophysiological models of the basal ganglia. Curr Opin Neurobiol 1996;6:751-58. Kemp JM, Powell TP. The cortico-striate projection in the monkey. Brain 1970;93(3):525-46. Kunzle H. Bilateral projections from precentral motor cortex to the putamen and other parts of the basal ganglia. An autoradiographic study in macaca fascicularis. Brain Res 1975;88:195-209. Kunzle H. Projections from the primary somatosensory cortex to basal ganglia and thalmus in the monkey. Exp Brain Res 1977;30:481-92. Pahapill PA, Lozano AM. The pedunculopontine nucleus and Parkinson’s disease. Brain 2000;123(Pt 9):1767-83. Parent M, Parent A. The pallidofugal motor fiber system in primates. Parkinsonism Relat Disord 2004;10 (4):203-11. Uno M, Yoshida M. Monosynaptic inhibition of thalamic neurons produced by stimulation of the pallidal nucleus in cats. Brain Res 1975;99:377-80. Ilinsky I, Kultas-Ilinsky K. Sagittal cytoarchitectonic maps of the macaca mulatta thalamus with a revised nomenclature of the motor-related nuclei validated by observations on their connectivity. J Comp Neurol 1987;262:331-64. Schell GR, Strick PL. The origin of thalamic inputs to the arcuate premotor and supplementary motor areas. J Neurosci 1984;4:539-560. Jones EG, et al. Cells of origin and terminal distribution of corticostriatal fibers arising in the sensory-motor cortex of monkeys. J Comp Neurol 1977;173:53-80. Hoover JE, Strick PL. Multiple output channels in the basal ganglia. Science 1993;259:819-21. Kievit J, Kuypers HG. Subcortical afferents to the frontal lobe in the rhesus monkey studied by means of retrograde horseradish peroxidase transport. Brain Res 1975;85 (2):261-6. Sadikot AF, Parent A, Francois C. Efferent connections of the centromedian and parafascicular thalamic nuclei in the squirrel monkey: a PHA-L study of subcortical projections. J Comp Neurol 1992;315:137-59. Sidibe M, Smith Y. Differential synaptic innervation of striatofugal neurones projecting to the internal or external segments of the globus pallidus by thalamic afferents in the squirrel monkey. J Comp Neurol 1996;365:445-65. Ilinsky I, Jouandet ML, Goldman-Rakic PS. Organization of the nigrothalamocortical system of the rhesus monkey. J Comp Neurol 1985;236:315-30. Harnois C, Filion M. Pallidofugal projections to the thalalmus and midbrain: A quantitative antidromic
Pathophysiology of dystonia
41.
42.
43.
44.
45.
46.
47. 48. 49.
50. 51. 52.
53.
54.
55.
56. 57.
58.
59.
activation study in monkeys and cats. Exp Brain Res 1982;47:277-85. Parent A, De Bellefeuille L. Organization of efferent projections from the internal segment of globus pallidus in primate as revealed by fluorescence retrograde labeling method. Brain Res 1982;245:201-213. Oka H, Jinnai K. Common projection of the motor cortex to the caudate nucleus and the cerebellum. Exp Brain Res 1978;31:31-42. Vitek JL, Giroux M. Physiology of hypokinetic and hyperkinetic movement disorders: model for dyskinesia. Ann Neurol 2000;47:S131-140. Albin RL, Young AB, Penney JB. The functional anatomy of basal ganglia disorders. Trends Neurosci 1989;12:366-75. Alexander G, Crutcher M. Functional architecture of basal ganglia circuits: Neural substrates of parallel processing. Trends Neurosci 1990;13:266-71. Mink JW. The basal ganglia and involuntary movements: impaired inhibition of competing motor patterns. Arch Neurol 2003;60:1365-68. de Carvalho Aguiar PM, Ozelius LJ. Classification and genetics of dystonia. Lancet Neurol 2002;1(5):316-25. McNaught KS, et al. Brainstem pathology in DYT1 primary torsion dystonia. Ann Neurol 2004;56(4):540-7. Carbon M, et al. Microstructural white matter changes in carriers of the DYT1 gene mutation. Ann Neurol 2004;56 (2):283-6. Eidelberg D, et al. Functional brain networks in DYT1 dystonia. Ann Neurol 1998;44:303-312. Eidelberg D, et al. The metabolic topography of idiopathic torsion dystonia. Brain 1995;118(Pt 6):1473-1484. Hutchinson M, et al. The metabolic topography of essential blepharospasm: a focal dystonia with general implications. Neurology 2000;55(5):673-7. Trost M, et al. Primary dystonia: is abnormal functional brain architecture linked to genotype? Ann Neurol 2002;52(6):853-6. Ceballos-Baumann AO, et al. Overactive prefrontal and underactive motor cortical areas in idiopathic dystonia. Ann Neurol 1995;37:363-372. Playford ED, et al. Increased activation of frontal areas during arm movement in idiopathic torsion dystonia. Mov Disord 1998;13(2):309-18. Greene PE, Bressman S. Exteroceptive and interoceptive stimuli in dystonia. Mov Disord 1998;13:549-551. Kaji R, et al. Tonic vibration reflex and muscle afferent block in writer’s cramp. Ann Neurol 1995;38 (2):155-62. Ibanez V, et al. Deficient activation of the motor cortical network in patients with writer’s cramp. Neurology 1999;53(1):96-105. Oga T, et al. Abnormal cortical mechanisms of voluntary muscle relaxation in patients with writer’s cramp: an fMRI study. Brain 2002;125:895-903.
107
60. Pujol J, et al. Brain cortical activation during guitarinduced hand dystonia studied by functional MRI. Neuroimage 2000;12(3):257-67. 61. Ceballos-Baumann AO, et al. Botulinum toxin does not reverse the cortical dysfunction associated with writer’s cramp. A PET study. Brain 1997;120(Pt 4):571-82. 62. Stoessl AJ, et al. PET studies of cerebral glucose metabolism in idiopathic torticollis. Neurology 1986;36:653-657. 63. Burton K, et al. Lesions of the putamen and dystonia: CT and magnetic resonance imaging. Neurology 1984;34:962-965. 64. Fross RD, et al. Lesions of the putamen: their relevance to dystonia. Neurology 1987;37(7):1125-9. 65. Marsden CD, et al. The anatomical basis of symptomatic hemidystonia. Brain 1985;108:463-83. 66. Bhatia KP, Marsden CD. The behavioural and motor consequences of focal lesions of the basal ganglia in man. Brain 1994;117(Pt 4):859-76. 67. Burns LH, et al. Selective putaminal excitotoxic lesions in non-human primates model the movement disorder of Huntington disease. Neuroscience 1995;64(4):1007-17. 68. Black KJ, Ongur D, Perlmutter JS. Putamen volume in idiopathic focal dystonia. Neurology 1998;51:819-824. 69. Magyar-Lehmann S, et al. Cerebral glucose metabolism in patients with spasmodic torticollis. Mov Disord 1997;12:704-8. 70. Levy LM, Hallett M. Impaired brain GABA in focal dystonia. Ann Neurol 2002;51(1):93-101. 71. Blood AJ, et al. Basal ganglia activity remains elevated after movement in focal hand dystonia. Ann Neurol 2004;55(5):744-8. 72. Carbon M, et al. Regional metabolism in primary torsion dystonia: effects of penetrance and genotype. Neurology 2004;62:1384-1390. 73. Albin RL, et al. Diminished striatal [123I]iodobenzovesamicol binding in idiopathic cervical dystonia. Ann Neurol 2003;53(4):528-32. 74. Graveland GA, Williams RS, DiFiglia M. A Golgi study of the human neostriatum: neurons and afferent fibers. J Comp Neurol 1985;234:317-333. 75. Aosaki T, Kimura M, Graybiel AM. Temporal and spatial characteristics of tonically active neurons of the primate’s striatum. J Neurophysiol 1995;73:1234-52. 76. Raz A, et al. Neuronal synchronization of tonically active neurons in the striatum of normal and parkinsonian primates. J Neurophysiol 1996;76(3):2083-8. 77. Pisani A, et al. Re-emergence of striatal cholinergic interneurons in movement disorders. Trends Neurosci 2007;30 (10):545-53. 78. Pisani A, et al. Altered responses to dopaminergic D2 receptor activation and N-type calcium currents in striatal cholinergic interneurons in a mouse model of DYT1 dystonia. Neurobiol Dis 2006;24(2):318-25. 79. Hamann M, et al. Acetylcholine receptor binding and cholinergic interneuron density are unaltered in a genetic
1797
1798
107 80.
81.
82.
83.
84. 85.
86.
87.
88.
89.
90.
91.
92.
93. 94.
95. 96.
Pathophysiology of dystonia
animal model of primary paroxysmal dystonia. Brain Res 2006;1099(1):176-82. Gernert M, et al. Deficit of striatal parvalbumin-reactive GABAergic interneurons and decreased basal ganglia output in a genetic rodent model of idiopathic paroxysmal dystonia. J Neurosci 2000;20(18):7052-8. Hamann M, et al. Age-related changes in parvalbuminpositive interneurons in the striatum, but not in the sensorimotor cortex in dystonic brains of the dt(sz) mutant hamster. Brain Res 2007;1150:190-9. Bennay M, Gernert M, Richter A. Spontaneous remission of paroxysmal dystonia coincides with normalization of entopeduncular activity in dt(SZ) mutants. J Neurosci 2001;21(13):RC153. Yamada H, Fujimoto K, Yoshida M. Neuronal mechanism underlying dystonia induced by bicuculline injection into the putamen of the cat. Brain Res 1995;677:333-36. Perlmutter JS, Mink JW. Dysfunction of dopaminergic pathways in dystonia. Adv Neurol 2004;94:163-70. Garver DL, et al. Dystonic reactions following neuroleptics: time course and proposed mechanisms. Psychopharmacologia 1976;47(2):199-201. Kolbe H, et al. Neuroleptic-induced acute dystonic reactions may be due to enhanced dopamine release on to supersensitive postsynaptic receptors. Neurology 1981;31 (4):434-9. Perlmutter JS, et al. MPTP induces dystonia and parkinsonism. Clues to the pathophysiology of dystonia. Neurology 1997;49:1432-38. Tabbal SD, et al. 1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine-induced acute transient dystonia in monkeys associated with low striatal dopamine. Neuroscience 2006;141(3):1281-7. Mink JW, Thach WT. Basal ganglia intrinsic circuits and their role in behavior. Curr Opin Neurobiol 1993;3 (6):950-7. Perlmutter JS, et al. Decreased [18F]spiperone binding in putamen in idiopathic focal dystonia. J Neurosci 1997;17 (2):843-50. Asanuma K, et al. Decreased striatal D2 receptor binding in non-manifesting carriers of the DYT1 dystonia mutation. Neurology 2005;64(2):347-9. Naumann M, et al. Imaging the pre- and postsynaptic side of striatal dopaminergic synapses in idiopathic cervical dystonia: a SPECT study using [123I] epidepride and [123I] beta-CIT. Mov Disord 1998;13(2):319-23. Breakefield XO, et al. The pathophysiological basis of dystonias. Nat Rev Neurosci 2008;9(3):222-34. Kishore A, et al. Striatal D2 receptors in symptomatic and asymptomatic carriers of doparesponsive dystonia measured with [11C]-raclopride and positron-emission tomography. Neurology 1998;50(4):1028-32. Kunig G, et al. D2 receptor binding in dopa-responsive dystonia. Ann Neurol 1998;44(5):758-62. Rinne JO, et al. Striatal dopaminergic system in doparesponsive dystonia: a multi-tracer PET study shows
increased D2 receptors. J Neural Transm 2004;111 (1):59-67. 97. Bucher SF, et al. Pallidal lesions. Structural and functional magnetic resonance imaging. Arch Neurol 1996;53(7):682-6. 98. Munro-Davies LE, et al. Lateral pallidotomy exacerbates akinesia in the Parkinsonian patient. J Clin Neurosci 1999;6(6):474-6. 99. Zhang J, et al. Lesions in monkey globus pallidus externus exacerbate parkinsonian symptoms. Exp Neurol 2006;199:446-453. 100. Vitek JL, et al. GPi pallidotomy for dystonia: clinical outcome and neuronal activity. Adv Neurol 1998;78:211-219. 101. Lozano AM, et al. Globus pallidus internus pallidotomy for generalized dystonia. Mov Disord 1997;12:865-870. 102. DeLong MR. Primate models of movement disorders of basal ganglia origin. Trends Neurosci 1990;13:281-5. 103. Starr PA, et al. Spontaneous pallidal neuronal activity in human dystonia: comparison with Parkinson’s disease and normal macaque. J Neurophysiol 2005;93 (6):3165-76. 104. Starr PA, et al. Microelectrode-guided implantation of deep brain stimulators into the globus pallidus internus for dystonia: techniques, electrode locations, and outcomes. Neurosurg Focus 2004;17(1):E4. 105. Suarez JI, et al. Pallidotomy for hemiballismus: efficacy and characteristics of neuronal activity. Ann Neurol 1997;42(5):807-11. 106. Vitek J, et al. Neuronal activity in the basal ganglia in patients with generalized dystonia and hemiballismus. Ann Neurol 1999;46(1):22-35. 107. Levy R, et al. Effects of apomorphine on subthalamic nucleus and globus pallidus internus neurons in patients with Parkinson’s disease. J Neurophysiol 2001;86 (1):249-60. 108. Merello M, et al. Apomorphine induces changes in GPi spontaneous outflow in patients with Parkinson’s disease. Mov Disord 1999;14(1):45-9. 109. Hutchison WD, et al. Pallidal neuronal activity: implications for models of dystonia. Ann Neurol 2003;53 (4):480-8. 110. Steigerwald F, et al. Effect of propofol anesthesia on pallidal neuronal discharges in generalized dystonia. Neurosci Lett 2005;386(3):156-9. 111. Tang JK, et al. Neuronal firing rates and patterns in the globus pallidus internus of patients with cervical dystonia differ from those with Parkinson’s disease. J Neurophysiol 2007;98(2):720-9. 112. Lenz FA, et al. Pallidal activity during dystonia: somatosensory reorganisation and changes with severity. J Neurol Neurosurg Psychiatry 1998;65(5):767-70. 113. Sanghera MK, et al. Basal ganglia neuronal discharge in primary and secondary dystonia in patients undergoing pallidotomy. Neurosurgery 2003;52(6):1358-70; discussion 1370-3.
Pathophysiology of dystonia
114. Gernert M, Richter A, Loscher W. In vivo extracellular electrophysiology of pallidal neurons in dystonic and nondystonic hamsters. J Neurosci Res 1999;57(6):894-905. 115. Merello M, et al. Neuronal globus pallidus activity in patients with generalised dystonia. Mov Disord 2004;19 (5):548-54. 116. Liu X, et al. Different mechanisms may generate sustained hypertonic and rhythmic bursting muscle activity in idiopathic dystonia. Exp Neurol 2006;198(1):204-13. 117. Zhuang P, Li Y, Hallett M. Neuronal activity in the basal ganglia and thalamus in patients with dystonia. Clin Neurophysiol 2004;115(11):2542-57. 118. Chang EF, et al. Neuronal responses to passive movement in the globus pallidus internus in primary dystonia. J Neurophysiol 2007;98(6):3696-707. 119. Gernert M, et al. Altered discharge pattern of basal ganglia output neurons in an animal model of idiopathic dystonia. J Neurosci 2002;22(16):7244-53. 120. Chen CC, et al. Neuronal activity in globus pallidus interna can be synchronized to local field potential activity over 3–12 Hz in patients with dystonia. Exp Neurol 2006;202(2):480-6. 121. Silberstein P, et al. Patterning of globus pallidus local field potentials differs between Parkinson’s disease and dystonia. Brain 2003;126(Pt 12):2597-608. 122. Sharott A, et al. Is the synchronization between pallidal and muscle activity in primary dystonia due to peripheral afferance or a motor drive? Brain 2008;131 (Pt 2):473-84. 123. Tang JK, et al. Changes in cortical and pallidal oscillatory activity during the execution of a sensory trick in patients with cervical dystonia. Exp Neurol 2007;204 (2):845-8. 124. Vitek JL. Surgery for dystonia. Neurosurg Clin N Am 1998;9:345-366. 125. Eidelberg D. Abnormal brain networks in DYT1 dystonia. Adv Neurol 1998;78:127-33. 126. Lehericy S, et al. Clinical characteristics and topography of lesions in movement disorders due to thalamic lesions. Neurology 2001;57(6):1055-66. 127. Zirh TA, et al. Thalamic single neuron and electromyographic activities in patients with dystonia. Adv Neurol 1998;78:27-32. 128. Galardi G, et al. Basal ganglia and thalamo-cortical hypermetabolism in patients with spasmodic torticollis. Acta Neurologica Scandia 1996;96:167-72. 129. Guehl D, et al. Bicuculline injections into the rostral and caudal motor thalamus of the monkey induce different types of dystonia. Eur J Neurosci 2000;12(3):1033-7. 130. Macia F, et al. Neuronal activity in the monkey motor thalamus during bicuculline-induced dystonia. Eur J Neurosci 2002;15(8):1353-62. 131. Lenz FA, et al. Thalamic single neuron activity in patients with dystonia: dystonia-related activity and somatic sensory reorganization. J Neurophysiol 1999;82:2372-2392.
107
132. LeDoux MS, Brady KA. Secondary cervical dystonia associated with structural lesions of the central nervous system. Mov Disord 2003;18(1):60-9. 133. Grey EG. Studies on the localization of cerebellar tumors: the position of the head and suboccipital discomforts. Ann Surg 1916;63:129-139. 134. Krauss JK, Seeger W, Jankovic J. Cervical dystonia associated with tumors of the posterior fossa. Mov Disord 1997;12:443-447. 135. LeDoux MS, Lorden JF, Ervin JM. Cerebellectomy eliminates the motor syndrome of the genetically dystonic rat. Exp Neurol 1993;120(2):302-10. 136. Lee MS, Marsden CD. Movement disorders following lesions of the thalamus or subthalamic region. Mov Disord 1994;9:493-507. 137. LeDoux MS, Lorden JF. Abnormal spontaneous and harmaline-stimulated Purkinje cell activity in the awake genetically dystonic rat. Exp Brain Res 2002;145 (4):457-67. 138. Beales M, et al. Quantitative autoradiography reveals selective changes in cerebellar GABA receptors of the rat mutant dystonic. J Neurosci 1990;10(6):1874-85. 139. Pizoli CE, et al. Abnormal cerebellar signaling induces dystonia in mice. J Neurosci 2002;22(17):7825-33. 140. Hoshi E, et al. The cerebellum communicates with the basal ganglia. Nat Neurosci 2005;8(11):1491-3. 141. Jinnah HA, Hess EJ. A new twist on the anatomy of dystonia: the basal ganglia and the cerebellum? Neurology 2006;67(10):1740-1. 142. Ceballos-Baumann AO, et al. Motor reorganization in acquired hemidystonia. Ann Neurol 1995;37(6):746-57. 143. Odergren T, Stone-Elander S, Ingvar M. Cerebral and cerebellar activation in correlation to the action-induced dystonia in writer’s cramp. Mov Disord 1998;13 (3):497-508. 144. Hallett M. Disorder of movement preparation in dystonia. Brain 2000;123(Pt 9):1765-1766. 145. Chen R, et al. Impaired inhibition in writer’s cramp during voluntary muscle activation. Neurology 1997;49(4):1054-9. 146. Gilio F, et al. Effects of botulinum toxin type A on intracortical inhibition in patients with dystonia. Ann Neurol 2000;48(1):20-6. 147. Ikoma K, et al. Abnormal cortical motor excitability in dystonia. Neurology 1996;46:1371-6. 148. Ridding MC, Taylor JL, Rothwell JC. The effect of voluntary contraction on cortico-cortical inhibition in human motor cortex. J Physiol (London) 1995;487 (2):541-8. 149. Curra A, et al. Transcranial magnetic stimulation techniques in clinical investigation. Neurology 2002;59 (12):1851-9. 150. Defazio G, Berardelli A, Hallett M. Do primary adultonset focal dystonias share aetiological factors? Brain 2007;130(Pt 5):1183-93.
1799
1800
107
Pathophysiology of dystonia
151. Werhahn KJ, et al. Differential effects on motorcortical inhibition induced by blockade of GABA uptake in humans. J Physiol 1999;517(Pt 2):591-7. 152. Ridding MC, et al. Changes in the balance between motor cortical excitation and inhibition in focal, task specific dystonia. J Neurol Neurosurg Psychiatry 1995;59(5):493-8. 153. Byl NN, et al. A primate model for studying focal dystonia and repetitive strain injury: effects on the primary somatosensory cortex. Phys Ther 1997;77(3):269-84. 154. Hallett M. Is dystonia a sensory disorder? Ann Neurol 1995;38:139-40. 155. Sanger TD, Merzenich MM. Computational model of the role of sensory disorganization in focal task-specific dystonia. J Neurophysiol 2000;84(5):2458-64. 156. Byl NN, Merzenich MM, Jenkins WM. A primate genesis model of focal dystonia and repetitive strain injury: I. Learning-induced dedifferentiation of the representation of the hand in the primary somatosensory cortex in adult monkeys. Neurology 1996;47:508-20.
157. Bara-Jimenez W, et al. Abnormal somatosensory homunculus in dystonia of the hand. Ann Neurol 1998;44:828-31. 158. Molloy FM, et al. Abnormalities of spatial discrimination in focal and generalized dystonia. Brain 2003;126 (Pt 10):2175-82. 159. Bara-Jimenez W, et al. Sensory discrimination capabilities in patients with focal hand dystonia. Ann Neurol 2000;47(3):377-80. 160. Tinazzi M, et al. Temporal discrimination of somesthetic stimuli is impaired in dystonic patients. Neuroreport 1999;10:1547-50. 161. Sanger T, Tarsy D, Pascual-Leone A. Abnormalities of spatial and temporal sensory discrimination in writer’s cramp. Mov Disord 2001;16:94-9. 162. Butterworth S, et al. Abnormal cortical sensory activation in dystonia: an fMRI study. Mov Disord 2003;18 (6):673-82. 163. Murase N, et al. Abnormal premovement gating of somatosensory input in writer’s cramp. Brain 2000;123 (Pt 9):1813-29.
89 Pathophysiology of Parkinson’s Disease M. R. DeLong . T. Wichmann
Introduction The characteristic motor impairments that accompany Parkinson’s disease, including paucity and slowness of movement (akinesia, bradykinesia), muscle stiffness (rigidity), and tremor at rest, are summarily called ‘‘Parkinsonism.’’ Parkinsonism is considered to result from the degeneration of dopaminergic neurons in the substantia nigra pars compacta (SNc), and dopamine loss in brain areas that receive inputs from SNc neurons. Although neurons in non-dopaminergic brain regions also degenerate in Parkinson’s disease, producing symptoms such as gait instability, cognitive decline, or autonomic dysfunction, the pathophysiology of these non-dopaminergic signs and symptoms is essentially unknown, because of the lack of valid animal models for these signs. The discussion of the pathophysiology of Parkinson’s disease in this chapter will therefore focus largely on the pathophysiology of (dopaminedependent) Parkinsonism which can be modeled in animals through the use of selective dopaminergic toxins, such as 1-methyl-4-phenyl-1, 2, 3, 6-tetrahydropyridine (MPTP) [1–3], or 6-hydroxydopamine (6-OHDA). In the following, we will first outline the circuit anatomy of the basal ganglia, followed by a review of the activity changes known to result from the loss of dopaminergic cells in the midbrain. A detailed discussion of the secondary morphological and biochemical changes in the basal ganglia that are seen in the parkinsonian state is beyond the scope of this chapter.
#
Springer-Verlag Berlin/Heidelberg 2009
Circuit Anatomy of the Basal Ganglia The basal ganglia include the striatum (caudate nucleus and putamen), the external and internal pallidal segments (GPe, GPi), the subthalamic nucleus (STN), and the substantia nigra, pars reticulata and pars compacta (SNr, SNc). Together with specific areas in thalamus and cortex, these structures form a family of segregated circuits that subserve ‘‘motor,’’ ‘‘associative’’ and ‘‘limbic’’ functions. Within each circuit, the striatum and the STN receive input from cortex (and other areas), which is then transferred to the basal ganglia output nuclei, GPi and SNr. GPi and SNr, in turn, project to the ventral anterior and ventrolateral nuclei of the thalamus (VA/VL), which then send efferents back to the cerebral cortex, largely targeting frontal cortical areas. The ‘‘motor’’ loop is particularly early affected by the degeneration of SNc neurons in Parkinson’s disease. This circuit originates in precentral motor areas, and involves the post-commissural putamen and motor regions of GPe, GPi, SNr, STN, and VA/VL. Basal ganglia projections also reach the intralaminar centromedian and parafascicular thalamic nuclei (CM/Pf), as well as brainstem structures, including the superior colliculus, pedunculopontine nucleus (PPN), and the reticular formation. CM/Pf and PPN project back to the basal ganglia, forming part of feedback loops that may help to regulate basal ganglia activity. Striatal output reaches GPi/SNr by one of two pathways, either the monosynaptic ‘‘direct’’
1498
89
Pathophysiology of parkinson’s disease
connection, or the polysynaptic ‘‘indirect’’ projection, which passes through GPe and STN. Direct and indirect pathways have opposing actions on GPi/SNr activity: Direct pathway activation reduces the inhibitory output from GPi/ SNr, thus disinhibiting thalamocortical neurons which may help to facilitate movements. Activation of the indirect pathway, in contrast, would increase GPi/SNr activities, resulting in greater inhibition of the thalamocortical system, and, with regard to motor functions, fewer movements. Dopaminergic inputs to the striatum are thought to regulate corticostriatal transmission. Striatal neurons that give rise to the direct pathway express facilitatory dopamine D1-receptors. In contrast, striatal neurons in the indirect pathway express inhibitory D2-receptors. Because of the opposite actions of the direct and indirect pathways, activation of D1- and D2-receptors by endogenous dopamine in the striatum may reduce GPi/SNr activity and facilitate the activity in associated thalamocortical projection neurons.
Neuronal Activity Changes in Basal Ganglia and Thalamus Throughout the 1980s and early 1990s, authors emphasized that striatal dopamine loss is associated with changes in firing rates of neurons in GPe, STN and GPi. Electrophysiologic studies in dopamine-depleted animals and parkinsonian patients showed an increase of activity in STN and GPi, and a decrease in GPe [4–12]. Together with measurements of brain metabolism [e.g., 13,14], these findings led to the development of the ‘‘rate model’’ of the pathophysiology of Parkinsonism [15,16] in which the activity changes in the basal ganglia are explained as resulting from a shift of the balance of activity in the direct and indirect pathways towards the movement-inhibiting actions of the indirect pathway (> Figure 89-1, right panel). Dopamine loss in the striatum would result in excessive activation of the indirect pathway via
reduction of D2-receptor mediated inhibition, and reduced DI-receptor mediated activation of the direct pathway. The end result of these changes would be inhibition of GPe activity, disinhibition of STN neurons, excessive activity in GPi and SNr, and greater inhibition of thalamocortical connections. The ‘‘rate’’ model of Parkinsonism is supported by studies of the effects of inactivation of STN or GPi in parkinsonian animals and patients. These interventions quickly reverse the motor signs of Parkinsonism [17–26]. As a reflection of increased GPi output to the thalamus, metabolic studies in MPTP-treated monkeys have suggested that the metabolism in the ‘‘basal ganglia-receiving’’ thalamic VA and VL nuclei is increased in Parkinsonism [13,27]. In addition, there is biochemical and metabolic evidence that GABAergic basal ganglia output directed at these nuclei is increased in Parkinsonism [28,29], suggesting changes in the overall synaptic or neuronal activity. However, the few existing studies of thalamic firing rate changes in Parkinsonism have not conclusively shown rate changes in the thalamus, suggesting that changes in basal ganglia output patterns may more strongly modulated metabolic changes than changes in firing rates alone [30–32]. Although the rate model has great appeal because of its simplicity, it has been criticized because predictions made based on this model were not met. For instance, biochemical studies examining activity-dependent changes in the GABA-synthesizing enzyme glutamate decarboxylase (GAD), did not confirm the expected changes in GPe and STN [7,33,34]. It has also been found that global increases or decreases in GPi activity, produced by local drug injections, do not necessarily produce Parkinsonism or involuntary movements, respectively, as would be predicted by the rate model. Also, against the predictions of the rate model, neither GPe nor VA/VL lesions reliably induce Parkinsonism. The antiparkinsonian effects of focal electrical ‘‘deep brain’’ stimulation (DBS) of the STN, a procedure
Pathophysiology of parkinson’s disease
89
. Figure 89-1 Parkinsonism-related changes in overall activity (‘‘rate model’’) in the basal ganglia-thalamocortical motor circuit. Black arrows indicate inhibitory connections; gray arrows indicate excitatory connections. The thickness of the arrows corresponds to their presumed activity. CM, centromedian nucleus of thalamus; CMA, cingulate motor area; D1, D2, dopamine receptor subtypes; GPe, external segment of the globus pallidus; GPi, internal segment of the globus pallidus; M1, primary motor cortex; Pf, parafascicular nucleus of the thalamus; PMC, premotor cortex; PPN, pedunculopontine nucleus; SMA, supplementary motor area; SNc, substantia nigra pars compacta; SNr, substantia nigra pars reticulata; STN, subthalamic nucleus; VA, ventral anterior nucleus of thalamus; VL, ventrolateral nucleus of thalamus
that may increase GPi output to the thalamus [35,36], are also contrary to the prediction of the rate model. These findings suggest that changes in the firing patterns of individual neurons or neuronal ensembles, and changes in the processing of sensory and other types of information may contribute significantly to Parkinsonism. Some of the alterations in firing patterns are readily apparent in the typical electrophysiological recordings of the spiking activity of single GPi neurons that are presented in > Figure 89-2. In the following paragraphs we will review the literature on these abnormalities. One of the most prominent pattern abnormalities in the extrastriatal basal ganglia and thalamus is an increase in the incidence of burstdischarges [see > Figure 89-2 and 5,7,8,37,38–43]. The generation of these bursts has not been clarified. Striatal dopamine loss may play a role in the production of bursts, as may loss of dopamine
outside of the striatum. As an example in favor of the latter possibility, absence of dopamine in the STN may enhance the effects of GABAergic inhibition of the STN, increasing the likelihood of post-inhibition rebound bursting [44–46]. The inhibitory GPe inputs may contribute strongly to the development of burst discharges in the STN [45,47–49]. Bursts are not only found in the basal ganglia, but also in the ventral thalamus [30–32,40,50–52]. Burst firing in the thalamus may be the result of hyperpolarization of thalamic neurons through increased GABAergic basal ganglia input [40,50], although this is disputed by some authors [31,52]. A second, related, characteristic of basal ganglia activity in parkinsonian animals and patients is the emergence of oscillatory activity (see > Figure 89-2). Abnormal oscillations in the alpha- and beta- frequency ranges have been documented in single-cell recordings in GPe,
1499
1500
89
Pathophysiology of parkinson’s disease
. Figure 89-2 Activity of single cells in GPi, recorded in a Rhesus monkey before and after treatment with MPTP. Shown are examples of separate neurons, recorded with standard extracellular recording methods in the same animal before and after MPTP. Each data segment is 5 s in duration
GPi, and STN in parkinsonian animals and patients with Parkinson’s disease whose basal ganglia were recorded during functional neurosurgical procedures [see, e.g., 53,54,55]. Field potential recordings made with DBS electrodes in GPi and STN of parkinsonian patients have also demonstrated an increased amount of beta activity, and a reduction of gamma-band oscillations in STN and GPi. These abnormalities disappeared when the patients were treated with dopaminergic agents [56,57]. It is likely that these local field potentials reflect joint synaptic or spiking activities generated by large groups of neurons. Pathological oscillations in single cell firing, and in field potential recordings have also been described in the thalamus [e.g., 40,51,52,58,59]. Closely related to bursts and oscillatory firing behavior is the development of abnormal synchrony between basal ganglia neurons and
between thalamic neurons in the dopaminedepleted state [reviewed by 32,53,57]. The pathological synchrony in the basal ganglia is likely to be a direct result of the loss of dopamine in the parkinsonian brain, as systemically administered dopaminergic agents rapidly reduce the synchrony [60,61]. How the loss of dopamine leads to increased synchrony remains unclear. Conceivably, synchronous activities could be produced in the striatum, for instance, through enhancement of electrotonic coupling between striatal cells [62–65], or through altered axon collateral activity [66]. Changes in the striatopallidal inhibition may also induce synchrony in the pallidum and other basal ganglia structures [67]. The interplay between GPe and STN is a possible source of synchronous oscillatory activity that is then transmitted to the basal ganglia output structures and thalamus. STN cells generate rebound bursts in response to transient
Pathophysiology of parkinson’s disease
volleys of inhibitory inputs from GPe [48], which may then trigger additional bursting activity in GPe. Inputs from the striatum, cortex or thalamus may modulate the function of this STN-GPe ‘‘pacemaker’’ [see, e.g., 67,68,69–73]. Another possible source of synchronous oscillatory activity in the basal ganglia is the cerebral cortex. Cortical inputs are known to strongly modulate basal ganglia activity, and overly synchronized cortical inputs may lead to an abnormal degree of synchronous oscillations in the basal ganglia [reviewed in ref. 53]. Besides the changes in spontaneous activity of the basal ganglia, the processing of sensory or movement-related information in the basal ganglia appears also to be altered in Parkinsonism. In the striatum, sensory responses can be assumed to be the result of cortical inputs to this structure. In the extrastriatal basal ganglia, other inputs, such as those reaching the STN via the corticosubthalamic projection [see > Figure 89-1, left panel and, e.g., 71,74,75] may also play a role. In VA/VL, such responses may be either transmitted through basal ganglia inputs, or (more likely) result from cortico-thalamic projections. In pallidal recordings in MPTP-treated animals, researchers found a reduction in the specificity of sensory responses [76,77], and an increase in the proportion of excitatory responses [78,79]. MPTP-induced Parkinsonism in monkeys was associated with reduced specificity of proprioceptive responses in the thalamus as well [32]. Such sensory changes may contribute to abnormal scaling of movements and bradykinesia in Parkinson’s disease by disrupting cortico-subcortical feedback mechanisms that control the extent and speed of movement [discussed in greater detail by 80].
Changes in Cortical Activity The fact that the basal ganglia are closely linked to areas in the cerebral cortex makes it likely that Parkinsonism is associated with substantial changes in cortical activity. To date, such changes
89
have been primarily explored with imaging methods. Early 2-deoxyglucose studies suggested that resting cortical activation is globally reduced in MPTP-treated monkeys [81]. Later functional imaging work in parkinsonian patients showed that cortical activation is specifically reduced in the supplementary motor area (SMA) and in the anterior cingulate cortex [82–89], and that such changes are linked to the changes in cortical activity and the previously described subcortical changes [90,91]. Recent electrophysiologic studies have found reduced activation and increased interneuronal synchrony in motor cortex or the SMA in MPTP-treated monkeys [92,93]. In parkinsonian patients, there is also evidence for abnormal EEG synchrony with inefficient movement-related modulation in the beta-band of frequencies [94–96], very likely related to the oscillatory activity in basal ganglia and thalamus (see above). The excessively synchronized oscillatory activity may be specifically relevant for development of akinesia [e.g., 97]. Interestingly, dopamine depletion not only disturbs the function of frontal cortical areas that receive input from the basal ganglia via the thalamus, but also that of other areas of cortex. Thus, in parkinsonian patients, motor and other tasks often recruit brain areas that are not activated in non-parkinsonian individuals, such as areas in the lateral premotor cortex, cerebellum and posterior parietal and occipital lobes [e.g., 82,87,88,89,98–101].
Changes in Brainstem Activity There is strong evidence that abnormalities in brainstem regions, such as the PPN, may also be involved in some aspects of Parkinsonism, specifically in akinesia. Thus, PPN lesions in normal monkeys produce hypokinesia [see, e.g., 102,103,104], and recent studies indicate that PPN-DBS may be effective for gait impairment difficulties that do not respond satisfactorily to anti-parkinson medications or STN DBS [105].
1501
1502
89
Pathophysiology of parkinson’s disease
Conclusion
References
Abnormalities throughout the basal gangliathalamocortical network contribute to Parkinsonism. Changes in neuronal activity, produced by dopamine loss in the putamen and other basal ganglia nuclei appear to severely disrupt the activity of neurons throughout the basal ganglia, thalamus and cortex and lead to the aberrant activation of brain areas that are not part of the immediate basal ganglia-thalamocortical circuitry. Interestingly, many of the pattern changes in the basal ganglia are not necessarily generated at the site of the greatest dopamine loss, the striatum, but in other areas of the circuitry, particularly in STN and GPe. At this time, there is no clear hypothesis unifying the various findings mentioned above into a coherent model that would explain how activity changes in basal ganglia, thalamus, cortex and brainstem produce Parkinsonism. However, even without such a model, continued efforts at characterizing in detail the abnormalities in brain activity in Parkinsonism will help us to develop better and more specific antiparkinsonian treatments. Specifically the insight that Parkinsonism is a network disorder in which abnormalities in small brain areas have far-reaching consequences on the entire motor portion of the brain has encouraged the development of focal neurosurgical interventions such as lesion and DBS treatments. These treatments have revolutionized the treatment of advanced Parkinsonism.
1. Burns RS, Chiueh CC, Markey SP, Ebert MH, Jacobowitz DM, Kopin IJ. A primate model of parkinsonism: selective destruction of dopaminergic neurons in the pars compacta of the substantia nigra by N-methyl-4phenyl-1,2,3,6-tetrahydropyridine. Proc Natl Acad Sci USA 1983;80:4546-50. 2. Langston JW, Ballard P, Tetrud JW, Irwin I. Chronic Parkinsonism in humans due to a product of meperidine-analog synthesis. Science 1983;219:979-80. 3. Forno LS, DeLanney LE, Irwin I, Langston JW. Similarities and differences between MPTP-induced parkinsonsim and Parkinson’s disease. Neuropathologic considerations. Adv Neurol 1993;60:600-8. 4. Miller WC, DeLong MR. Altered tonic activity of neurons in the globus pallidus and subthalamic nucleus in the primate MPTP model of parkinsonism. In: Carpenter MB, Jayaraman A, editors. The basal ganglia II. New York: Plenum Press, 1987. p. 415-27. 5. Bergman H, Wichmann T, Karmon B, DeLong MR. The primate subthalamic nucleus. II. Neuronal activity in the MPTP model of parkinsonism. J Neurophysiol 1994;72:507-20. 6. Filion M, Kemblay L, Bedard PJ. Effects of dopamine agonists on the spontaneous activity of globus pallidus neurons in monkeys with MPTP-induced parkinsonism. Brain Res 1991;547:152-161. 7. Soares J, Kliem MA, Betarbet R, Greenamyre JT, Yamamoto B, Wichmann T. Role of external pallidal segment in primate parkinsonism: comparison of the effects of MPTP-induced parkinsonism and lesions of the external pallidal segment. J Neurosci 2004;24:6417-26. 8. Wichmann T, Bergman H, Starr PA, Subramanian T, Watts RL, DeLong MR. Comparison of MPTP-induced changes in spontaneous neuronal discharge in the internal pallidal segment and in the substantia nigra pars reticulata in primates. Exp Brain Res 1999;125:397-409. 9. Hassani OK, Mouroux M, Feger J. Increased subthalamic neuronal activity after nigral dopaminergic lesion independent of disinhibition via the globus pallidus. Neuroscience 1996;72:105-15. 10. Dogali M, Beric A, Sterio D, Eidelberg D, Fazzini E, Takikawa S, Samelson DR, Devinsky O, Kolodny EH. Anatomic and physiological considerations in pallidotomy for Parkinson’s disease. Stereotact Funct Neurosurg 1994;62:53-60. 11. Lozano A, Hutchison W, Kiss Z, Tasker R, Davis K, Dostrovsky J. Methods for microelectrode-guided posteroventral pallidotomy. J Neurosurg 1996;84:194-202. 12. Taha JM, Favre J, Baumann TK, Burchiel KJ. Tremor control after pallidotomy in patients with Parkinson’s disease: correlation with microrecording findings. J Neurosurg 1997;86:642-7.
Acknowledgements This work was supported by grants from the National Institute of Neurological Disorders and Stroke (NS042250, NS049474 and NS054976), and an institutional NIH grant (RR-000165) to the Yerkes National Primate Research Center.
Pathophysiology of parkinson’s disease
13. Mitchell IJ, Clarke CE, Boyce S, Robertson RG, Peggs D, Sambrook MA, Crossman AR. Neural mechanisms 0underlying parkinsonian symptoms based upon regional uptake of 2-deoxyglucose in monkeys exposed to 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine. Neuroscience 1989;32:213-26. 14. Schwartzman RJ, Alexander GM. Spinal cord metabolism of the 1-methyl-4-phenyl-1,2,3, 6-tetrahydropyridine-treated monkey. Brain Res 1985;337:263-8. 15. Albin RL, Young AB, Penney JB. The functional anatomy of basal ganglia disorders. Trends Neurosci 1989;12:366-75. 16. DeLong MR. Primate models of movement disorders of basal ganglia origin. Trends Neurosci 1990;13:281-5. 17. Bergman H, Wichmann T, DeLong MR. Reversal of experimental parkinsonism by lesions of the subthalamic nucleus. Science 1990;249:1436-8. 18. Aziz TZ, Peggs D, Sambrook MA, Crossman AR. Lesion of the subthalamic nucleus for the alleviation of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)induced parkinsonism in the primate. Mov Disord 1991;6:288-92. 19. Guridi J, Herrero MT, Luquin R, Guillen J, Obeso JA. Subthalamotomy improves MPTP-induced parkinsonism in monkeys. Stereotact Funct Neurosurg 1994;62:98-102. 20. Gill SS, Heywood P. Bilateral subthalamic nucleotomy can be accomplished safely. Mov Disord 1998;13:201. 21. Alvarez L, Macias R, Lopez G, Alvarez E, Pavon N, Rodriguez-Oroz MC, Juncos JL, Maragoto C, Guridi J, Litvan I, Tolosa ES, Koller W, Vitek J, DeLong MR, Obeso JA. Bilateral subthalamotomy in Parkinson’s disease: initial and long-term response. Brain 2005;128:570-83. 22. Laitinen LV. Pallidotomy for Parkinson’s diesease. Neurosurg Clin N Am 1995;6:105-12. 23. Baron MS, Vitek JL, Bakay RA, Green J, Kaneoke Y, Hashimoto T, Turner RS, Woodard JL, Cole SA, McDonald WM, DeLong MR. Treatment of advanced Parkinson’s disease by posterior GPi pallidotomy: 1-year results of a pilot study. Ann Neurol 1996;40:355-66. 24. Lozano AM, Lang AE, Galvez-Jimenez N, Miyasaki J, Duff J, Hutchinson WD, Dostrovsky JO. Effect of GPi pallidotomy on motor function in Parkinson’s disease. Lancet 1995;346:1383-7. 25. Dogali M, Fazzini E, Kolodny E, Eidelberg D, Sterio D, Devinsky O, Beric A. Stereotactic ventral pallidotomy for Parkinson’s disease. Neurology 1995;45:753-61. 26. Vitek JL, Bakay RA, Freeman A, Evatt M, Green J, McDonald W, Haber M, Barnhart H, Wahlay N, Triche S, Mewes K, Chockkan V, Zhang JY, DeLong MR. Randomized trial of pallidotomy versus medical therapy for Parkinson’s disease. Ann Neurol 2003;53:558-69. 27. Rolland AS, Herrero MT, Garcia-Martinez V, Ruberg M, Hirsch EC, Francois C. Metabolic activity of cerebellar and basal ganglia-thalamic neurons is reduced in parkinsonism. Brain 2007;130:265-75.
89
28. Chadha A, Dawson LG, Jenner PG, Duty S. Effect of unilateral 6-hydroxydopamine lesions of the nigrostriatal pathway on GABA(A) receptor subunit gene expression in the rodent basal ganglia and thalamus. Neuroscience 2000;95:119-26. 29. Palombo E, Porrino LJ, Bankiewicz KS, Crane AM, Sokoloff L, Kopin IJ. Local cerebral glucose utilization in monkeys with hemiparkinsonism induced by intracarotid infusion of the neurotoxin MPTP. J Neurosci 1990; 10:860-9. 30. Vitek JL, Ashe J, Kaneoke Y. Spontaneous neuronal activity in the motor thalamus: alteration in pattern and rate in parkinsonism. Soc Neurosci Abstr 1994; 20:561. 31. Molnar GF, Pilliar A, Lozano AM, Dostrovsky JO. Differences in neuronal firing rates in pallidal and cerebellar receiving areas of thalamus in patients with Parkinson’s disease, essential tremor, and pain. J Neurophysiol 2005;93:3094-101. 32. Pessiglione M, Guehl D, Rolland AS, Francois C, Hirsch EC, Feger J, Tremblay L. Thalamic neuronal activity in dopamine-depleted primates: evidence for a loss of functional segregation within basal ganglia circuits. J Neurosci 2005;25:1523-31. 33. Pedneault S, Soghomonian JJ. Glutamate decarboxylase (GAD65) mRNA levels in the striatum and pallidum of MPTP-treated monkeys. Brain Res Mol Brain Res 1994;25:351-4. 34. Schneider JS, Wade TV. Experimental parkinsonism is associated with increased pallidal GAD gene expression and is reversed by site-directed antisense gene therapy. Mov Disord 2003;18:32-40. 35. Hashimoto T, Elder CM, Okun MS, Patrick SK, Vitek JL. Stimulation of the subthalamic nucleus changes the firing pattern of pallidal neurons. J Neurosci 2003;23:1916-23. 36. McIntyre CC, Grill WM, Sherman DL, Thakor NV. Cellular effects of deep brain stimulation: model-based analysis of activation and inhibition. J Neurophysiol 2004;91:1457-69. 37. Filion M. Effects of interruption of the nigrostriatal pathway and of dopaminergic agents on the spontaneous activity of globus pallidus neurons in the awake monkey. Brain Res 1979;178:425-41. 38. Wichmann T, Soares J. Neuronal firing before and after burst discharges in the monkey Basal Ganglia is predictably patterned in the normal state and altered in parkinsonism. J Neurophysiol 2006;95:2120-33. 39. Hutchison WD, Lozano AM, Davis K, Saint-Cyr JA, Lang AE, Dostrovsky JO. Differential neuronal activity in segments of globus pallidus in Parkinson’s disease patients. Neuroreport 1994;5:1533-7. 40. Magnin M, Morel A, Jeanmonod D. Single-unit analysis of the pallidum, thalamus and subthalamic nucleus in parkinsonian patients. Neuroscience 2000;96:549-64.
1503
1504
89
Pathophysiology of parkinson’s disease
41. Vila M, Perier C, Feger J, Yelnik J, Faucheux B, Ruberg M, Raisman-Vozari R, Agid Y, Hirsch EC. Evolution of changes in neuronal activity in the subthalamic nucleus of rats with unilateral lesion of the substantia nigra assessed by metabolic and electrophysiological measurements. Eur J Neurosci 2000;12:337-44. 42. Ni ZG, Bouali-Benazzouz R, Gao DM, Benabid AL, Benazzouz A. Time-course of changes in firing rates and firing patterns of subthalamic nucleus neuronal activity after 6-OHDA-induced dopamine depletion in rats. Brain Res 2001;899:142-7. 43. Breit S, Bouali-Benazzouz R, Popa RC, Gasser T, Benabid AL, Benazzouz A. Effects of 6-hydroxydopamineinduced severe or partial lesion of the nigrostriatal pathway on the neuronal activity of pallido-subthalamic network in the rat. Exp Neurol 2007;205:36-47. 44. Shen KZ, Johnson SW. Dopamine depletion alters responses to glutamate and GABA in the rat subthalamic nucleus. Neuroreport 2005;16:171-4. 45. Bevan MD, Hallworth NE, Baufreton J. GABAergic control of the subthalamic nucleus. Prog Brain Res 2007;160:173-88. 46. Bevan MD, Atherton JF, Baufreton J. Cellular priniciples underlying normal and pathological activity in the subthalamic nucleus. Curr Opin Neurobiol 2006;16:621-628. 47. Ni Z, Bouali-Benazzouz R, Gao D, Benabid AL, Benazzouz A. Changes in the firing pattern of globus pallidus neurons after the degeneration of nigrostriatal pathway are mediated by the subthalamic nucleus in the rat. Eur J Neurosci 2000;12:4338-44. 48. Plenz D, Kitai S. A basal ganglia pacemaker formed by the subthalamic nucleus and external globus pallidus. Nature 1999;400:677-82. 49. Beurrier C, Congar P, Bioulac B, Hammond C. Subthalamic nucleus neurons switch from single-spike activity to burst-firing mode. J Neurosci 1999;19:599-609. 50. Kaneoke Y, Vitek JL. The motor thalamus in the parkinsonian primate: enhanced burst and oscillatory activities. Soc Neurosci Abstr 1995;21:1428. 51. Guehl D, Pessiglione M, Francois C, Yelnik J, Hirsch EC, Feger J, Tremblay L. Tremor-related activity of neurons in the ‘motor’ thalamus: changes in firing rate and pattern in the MPTP vervet model of parkinsonism. European Journal of Neuroscience 2003;17:2388-400. 52. Zirh TA, Lenz FA, Reich SG, Dougherty PM. Patterns of bursting occurring in thalamic cells during parkinsonian tremor. Neuroscience 1998;83:107-21. 53. Gatev P, Darbin O, Wichmann T. Oscillations in the basal ganglia under normal conditions and in movement disorders. Mov Disord 2006;21:1566-77. 54. Rivlin-Etzion M, Marmor O, Heimer G, Raz A, Nini A, Bergman H. Basal ganglia oscillations and pathophysiology of movement disorders. Curr Opin Neurobiol 2006;16:629-37. 55. Levy R, Hutchison WD, Lozano AM, Dostrovsky JO. Synchronized neuronal discharge in the basal ganglia
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
of parkinsonian patients is limited to oscillatory activity. J Neurosci 2002;22:2855-61. Brown P, Oliviero A, Mazzone P, Insola A, Tonali P, Di Lazzaro V. Dopamine dependency of oscillations between subthalamic nucleus and pallidum in Parkinson’s disease. J Neurosci 2001;21:1033-8. Hammond C, Bergman H, Brown P. Pathological synchronization in Parkinson’s disease: networks, models and treatments. Trends Neurosci 2007;30:357-64. Raeva S, Vainberg N, Dubinin V. Analysis of spontaneous activity patterns of human thalamic ventrolateral neurons and their modifications due to functional brain changes. Neuroscience 1999;88:365-76. Sarnthein J, Jeanmonod D. High thalamocortical theta coherence in patients with Parkinson’s disease. J Neurosci 2007;27:124-31. Heimer G, Bar-Gad I, Goldberg JA, Bergman H. Dopamine replacement therapy reverses abnormal synchronization of pallidal neurons in the 1-methyl-4phenyl-1,2,3,6-tetrahydropyridine primate model of parkinsonism. J Neurosci 2002;22:7850-5. Levy R, Ashby P, Hutchison WD, Lang AE, Lozano AM, Dostrovsky JO. Dependence of subthalamic nucleus oscillations on movement and dopamine in Parkinson’s disease. Brain 2002;125:1196-209. O’Donnell P, Grace AA. Dopaminergic modulation of dye coupling between neurons in the core and shell regions of the nucleus accumbens. J Neurosci 1993;13:3456-71. Cepeda C, Walsh JP, Hull CD, Howard SG, Buchwald NA, Levine MS. Dye-coupling in the neostriatum of the rat: I. Modulation by dopamine-depleting lesions. Synapse 1989;4:229-37. Onn SP, Grace AA. Amphetamine withdrawal alters bistable states and cellular coupling in rat prefrontal cortex and nucleus accumbens neurons recorded in vivo. J Neurosci 2000;20:2332-45. Berretta N, Paolucci E, Bernardi G, Mercuri NB. Glutamate receptor stimulation induces a persistent rhythmicity of the GABAergic inputs to rat midbrain dopaminergic neurons. Eur J Neurosci 2001;14:777-84. Guzman JN, Hernandez A, Galarraga E, Tapia D, Laville A, Vergara R, Aceves J, Bargas J. Dopaminergic modulation of axon collaterals interconnecting spiny neurons of the rat striatum. J Neurosci 2003;23:8931-40. Terman D, Rubin JE, Yew AC, Wilson CJ. Activity patterns in a model for the subthalamopallidal network of the basal ganglia. J Neurosci 2002;22:2963-76. Loucif KC, Wilson CL, Baig R, Lacey MG, Stanford IM. Functional interconnectivity between the globus pallidus and the subthalamic nucleus in the mouse brain slice. J Physiol 2005;567:977-87. Bevan MD, Magill PJ, Terman D, Bolam JP, Wilson CJ. Move to the rhythm: oscillations in the subthalamic nucleus-external globus pallidus network. Trends in Neurosciences 2002;25:525-31.
Pathophysiology of parkinson’s disease
70. Stanford IM. Independent neuronal oscillators of the rat globus pallidus. J Neurophysiol 2003;89:1713-17. 71. Hartmann-von Monakow K, Akert K, Kunzle H. Projections of the precentral motor cortex and other cortical areas of the frontal lobe to the subthalamic nucleus in the monkey. Exp Brain Res 1978;33:395-403. 72. Nambu A, Tokuno H, Takada M. Functional significance of the cortico-subthalamo-pallidal ‘hyperdirect’ pathway. Neurosci Res 2002;43:111-17. 73. Castle M, Aymerich MS, Sanchez-Escobar C, Gonzalo N, Obeso JA, Lanciego JL. Thalamic innervation of the direct and indirect basal ganglia pathways in the rat: ipsi- and contralateral projections. J Comp Neurol 2005;483:143-53. 74. Hamada I, DeLong MR. Excitotoxic acid lesions of the primate subthalamic nucleus result in reduced pallidal neuronal activity during active holding. J Neurophysiol 1992;68:1859-66. 75. Nambu A, Kaneda K, Tokuno H, Takada M. Organization of corticostriatal motor inputs in monkey putamen. J Neurophysiol 2002;88:1830-42. 76. Rothblat DS, Schneider JS. Alterations in pallidal neuronal responses to peripheral sensory and striatal stimulation in symptomatic and recovered parkinsonian cats. Brain Res 1995;705:1-14. 77. Schneider JS, Rothblat DS. Alterations in intralaminar and motor thalamic physiology following nigrostriatal dopamine depletion. Brain Res 1996;742:25-33. 78. Boraud T, Bezard E, Bioulac B, Gross CE. Ratio of inhibited-to-activated pallidal neurons decreases dramatically during passive limb movement in the MPTP-treated monkey. J Neurophysiol 2000;83:1760-3. 79. Leblois A, Meissner W, Bioulac B, Gross CE, Hansel D, Boraud T. Late emergence of synchronized oscillatory activity in the pallidum during progressive Parkinsonism. Eur J Neurosci 2007;26:1701-13. 80. Wichmann T, DeLong MR. Pathophysiology of parkinsonian motor abnormalities. In: Narabayashi H, Nagatsu T, Yanagisawa N, Mizuno Y , editors. Advances in neurology, vol 60. New York: Raven Press; 1993. p. 53-61. 81. Schwartzman RJ, Alexander GM. Changes in the local cerebral metabolic rate for glucose in the 1-methyl4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) primate model of Parkinson’s disease. Brain Res 1985;358:137-43. 82. Turner RS, Grafton ST, McIntosh AR, DeLong MR, Hoffman JM. The functional anatomy of parkinsonian bradykinesia. Neuroimage 2003;19:163-79. 83. Brooks DJ. PET and SPECT studies in Parkinson’s disease. Baillieres Clin Neurol 1997;6:69-87. 84. Playford ED, Jenkins IH, Passingham RE, Nutt J, Frackowiak RS, Brooks DJ. Impaired mesial frontal and putamen activation in Parkinson’s disease: a positron emission tomography study. Ann Neurol 1992;32:151-61. 85. Jenkins IH, Fernandez W, Playford ED, Lees AJ, Frackowiak RS, Passingham RE, Brooks DJ. Impaired activation of the supplementary motor area in Parkinson’s
89
disease is reversed when akinesia is treated with apomorphine. Ann Neurol 1992;32:749-57. 86. Jahanshahi M, Jenkins IH, Brown RG, Marsden CD, Passingham RE, Brooks DJ. Self-initiated versus externally triggered movements. I. An investigation using measurement of regional cerebral blood flow with PET and movement-related potentials in normal and Parkinson’s disease subjects. Brain 1995;118:913-33. 87. Thobois S, Dominey P, Decety J, Pollak P, Gregoire MC, Broussolle E. Overactivation of primary motor cortex is asymmetrical in hemiparkinsonian patients. Neuroreport 2000;11:785-9. 88. Samuel M, Ceballos-Baumann AO, Blin J, Uema T, Boecker H, Passingham RE, Brooks DJ. Evidence for lateral premotor and parietal overactivity in Parkinson’s disease during sequential and bimanual movements. A PET study. Brain 1997;120:963-76. 89. Haslinger B, Erhard P, Kampfe N, Boecker H, Rummeny E, Schwaiger M, Conrad B, Ceballos-Baumann AO. Event-related functional magnetic resonance imaging in Parkinson’s disease before and after levodopa. Brain 2001;124:558-70. 90. Antonini A, Moeller JR, Nakamura T, Spetsieris P, Dhawan V, Eidelberg D. The metabolic anatomy of tremor in Parkinson’s disease. Neurology 1998;51:803-10. 91. Eidelberg D. Functional brain networks in movement disorders. Current Opinion in Neurology 1998;11: 319-26. 92. Watts RL, Mandir AS. The role of motor cortex in the pathophysiology of voluntary movement deficits associated with parkinsonism. Neurologic Clinics 1992;10: 451-69. 93. Goldberg JA, Boraud T, Maraton S, Haber SN, Vaadia E, Bergman H. Enhanced synchrony among primary motor cortex neurons in the 1-methyl-4-phenyl-1,2,3,6tetrahydropyridine primate model of Parkinson’s disease. J Neurosci 2002;22:4639-53. 94. Silberstein P, Pogosyan A, Kuhn AA, Hotton G, Tisch S, Kupsch A, Dowsey-Limousin P, Hariz MI, Brown P. Cortico-cortical coupling in Parkinson’s disease and its modulation by therapy. Brain 2005;128:1277-91. 95. Brown P, Marsden CD. What do the basal ganglia do? Lancet 1998;351:1801-4. 96. Brown P. Oscillatory nature of human basal ganglia activity: relationship to the pathophysiology of Parkinson’s disease. Mov Disord 2003;18:357-63. 97. Timmermann L, Wojtecki L, Gross J, Lehrke R, Voges J, Maarouf M, Treuer H, Sturm V, Schnitzler A. Ten-Hertz stimulation of subthalamic nucleus deteriorates motor symptoms in Parkinson’s disease. Mov Disord 2004;19:1328-33. 98. Dagher A, Owen AM, Boecker H, Brooks DJ. The role of the striatum and hippocampus in planning: a PET activation study in Parkinson’s disease. Brain 2001;124: 1020-32. 99. Moody TD, Bookheimer SY, Vanek Z, Knowlton BJ. An implicit learning task activates medial temporal lobe
1505
1506
89
Pathophysiology of parkinson’s disease
in patients with Parkinson’s disease. Behav Neurosci 2004;118:438-42. 100. Grossman M, Cooke A, DeVita C, Lee C, Alsop D, Detre J, Gee J, Chen W, Stern MB, Hurtig HI. Grammatical and resource components of sentence processing in Parkinson’s disease: an fMRI study. Neurology 2003;60:775-81. 101. Peters S, Suchan B, Rusin J, Daum I, Koster O, Przuntek H, Muller T, Schmid G. Apomorphine reduces BOLD signal in fMRI during voluntary movement in Parkinsonian patients. Neuroreport 2003;14:809-12. 102. Kojima J, Yamaji Y, Matsumura M, Nambu A, Inase M, Tokuno H, Takada M, Imai H. Excitotoxic lesions of the pedunculopontine tegmental nucleus produce
contralateral hemiparkinsonism in the monkey. Neurosci Lett 1997;226:111-14. 103. Nandi D, Aziz TZ, Giladi N, Winter J, Stein JF. Reversal of akinesia in experimental parkinsonism by GABA antagonist microinjections in the pedunculopontine nucleus. Brain 2002;125:2418-30. 104. Mena-Segovia J, Bolam JP, Magill PJ, Pedunculopontine nucleus and basal ganglia: distant relatives or part of the same family? Trends Neurosci 2004;27:585-8. 105. Stefani A, Lozano AM, Peppe A, Stanzione P, Galati S, Tropepi D, Pierantozzi M, Brusa L, Scarnati E, Mazzone P. Bilateral deep brain stimulation of the pedunculopontine and subthalamic nuclei in severe Parkinson’s disease. Brain 2007;130:1596-607.
91 Patient Selection for Surgery for Parkinson’s Disease E. K. Tan . J. Jankovic
The cornerstone of therapy for Parkinson’s disease (PD) is dopamine replacement with levodopa. The early years of treatment with levodopa are most predictable in obtaining a satisfactory therapeutic response. However, chronic administration can lead to problematic side effects, chiefly motor fluctuations and dyskinesias, in the majority of patients [1,2]. There has been an enormous resurgence of interest in functional surgery in PD in recent years as evidenced by the growing number of reported clinical trials in this area. Better understanding of the pathophysiology of basal ganglia dysfunction underlying PD and advancements in neuroradiological, neurosurgical and neurophysiological techniques have contributed to this trend. PD surgery involves lesioning or deep brain stimulation (DBS) of specific nuclei of the basal ganglia and cell replacement, details of which will be covered in other chapters of this book. It is worth emphasizing that surgical therapy can complement medical treatment in the more advanced cases of PD and improve quality of life [3]. A randomized study showed that DBS of the subthalamic nucleus (STN) was more effective than medical management alone in patients under 75 years of age with severe motor complications [4]. Selecting the most appropriate patient is important in enhancing the success of surgery. While there are no ‘‘gold standard’’ guidelines, enormous experience has been accumulated over the years, particularly with DBS surgery, that has helped in developing useful criteria for selection of the best surgical candidates [5,6]. While most surgical procedures for PD are still performed in #
Springer-Verlag Berlin/Heidelberg 2009
academic centers, more and more communitybased neurologists and even primary care physicians encounter patients who are considering surgery or have undergone surgery for PD and, therefore, dissemination of knowledge about selection of patients, expectation and postoperative management is increasingly important not only for specialists but also for other health care providers. Here we provide a concise review of the general and specific indications in selecting patients for PD surgery and share our experiences in this area. It is beyond the scope of this review to discuss surgical patient selection for other indications, such as tremor, dystonia, Tourette syndrome, and other movement disorders [7].
Selection of Patients As a general rule, functional stereotactic surgery should be performed in specialized centers where clinical expertise in movement disorders, technology and interpretation of imaging, surgical equipment, and skilled personnel are available to provide a coordinated and fully integrated multidisciplinary approach in order to maximize the benefits and minimize the risks of surgery. The decision as to which specific type of surgery should be recommended must be based on a balanced review of current knowledge of the indications and surgical outcomes, personal experience by the surgeon, and a consensus opinion by both the neurologist and neurosurgeon, along with active participation of the patient and the family. There are no rigid rules to follow in
1530
91
Patient selection for surgery for parkinson’s disease
. Table 91-1 Selection of patients for deep brain stimulation surgery Medical team requirements Multidisciplinary team approach involving neurologists, neurosurgeons, and neuropsychologists Clinical expertise in recognition and treatment of movement disorders Skills and experience in stereotactic surgery Adequate imaging Patient’s requirements Fully understands the risk/benefits/limitations of surgery Willingness to commit time and personal resources for postoperative medication and programming adjustments Realistic expectations Medical criteria Duration of PD symptoms more than 5 years (to allow for atypical features to emerge and assess response to dopaminergic therapy) Dopaminergic responsiveness (>30% reduction in motor UPDRS) Troublesome dyskinesias, motor fluctuations despite optimal medical therapy Disabling medication resistant tremor Normal MRI No atypical and secondary parkinsonism No dementia and depression or other psychiatric problems Good medical health Preferably not >80 years of age
the selection process, as every case has to be individualized according to symptomatology and specific needs of the patient. A careful selection process, however, is critical to beneficial outcome (> Table 91-1).
Inclusion Criteria Diagnosis of Parkinson’s Disease Many studies have demonstrated that patients with atypical parkinsonism, such as progressive supranuclear palsy, multiple system atrophy, dementia with Lewy bodies, or vascular parkinsonism do not respond well to surgery and their symptoms may substantially worsen after surgery
[8]. Therefore, the correct diagnosis of PD is one of the most important criteria for selection of patients for surgery. In the absence of a diagnostic test, several clinical criteria have been proposed for the diagnosis of PD [9,10]. Despite refinements in these clinical criteria, there may be up to 8.1–25% inaccuracy in the initial diagnosis even in the hands of PD experts [11]. Since the diagnostic inaccuracy improves with followup, it is advisable to consider patients for surgery only when there is a considerable duration of symptoms (at least 5 years). This is reasonable considering the fact that most patients develop complications of fluctuations and dyskinesias, the chief indications for surgery, only after 5–6 years of levodopa therapy [1].
Response to Levodopa Kazumata et al. showed that response to levodopa was a preoperative indicator of good clinical outcome for stereotaxic pallidotomy [12]. They suggested that assessment of clinical response to levodopa could provide a measure of the patient’s capacity to reduce baseline pallidal hyperactivity and to modulate the complex cortico-striatothalamic motor network. In a recent study on the predictive value of preoperative levodopa response to long-term DBS benefit in 33 PD patients who underwent bilateral STN-DBS, Piboolnurak and colleagues [13] demonstrated that levodopa response significantly decreased postoperatively by 31.1% at 3 years and 32.3% at 5 years. These findings could be explained by a reduction in medication requirement, direct STN stimulation effect or PD progression. The magnitude of the preoperative response to levodopa did not predict DBS benefit at 3 and 5 years. Most authors, however, would agree that levodopa responsiveness remains one of the best predictors for outcome after STN DBS [6]. Some have indicated that carefully selected PD patients intolerant to levodopa may also have significant
Patient selection for surgery for parkinson’s disease
improvement [14]. Wider et al. suggested that STN DBS can compensate for the short-duration response and long-duration response to levodopa [15]. Based on the CAPIT and other assessments, minimum 30% reduction in the true off state (at least 12 h after the last dose of levodopa), as assessed by the motor part (part III) of the Unified Parkinson’s Disease Rating Scale (UPDRS) after a morning levodopa is recommended as evidence of levodopa responsiveness. Using a suprathreshold (1.5–2 times the usual) dose of levodopa tends to assure an optimal response and also may bring out dyskinesias which are helpful to observe as this is usually one of the symptoms that is targeted for improvement with surgery. Patients should understand that it is highly unlikely that their condition after surgery will be better than their best ‘‘on’’ status after optimal dose of levodopa. In this regard we find it very helpful to video patient before surgery during their off state and on state using a standardized video protocol. This is useful not only in documenting their condition and their ‘‘best on time,’’ including dyskinesias, but also to review retrospectively if in the future there are any concerns whether atypical features were present.
Dyskinesias and Motor Fluctuations Levodopa-related complications, such as wearing off and dyskinesias, are among the most important reasons for referring PD patients for surgery. Most pallidotomy series reported robust effect on levodopa-induced dyskinesias, with improvement in the contralateral dyskinesia ranging from 65 to 100% [16]. Similar effects have been reported with DBS [16–18]. Krack et al. [17] showed that bilateral STN DBS can lead to marked improvements over 5 years in motor function while ‘‘off ’’ medication and in dyskinesia while ‘‘on’’ medication. Marked improvement in dyskinesias could also be achieved during the
91
‘‘on’’ state following chronic bilateral DBS of globus pallidus interna (GPi) [18]. Rodrigues et al. noted that reduction in dyskinesia can play a major role in the improvement of quality of life in GPi-DBS treated patients [19]. More comprehensive review of the effect of DBS on dyskinesias and motor fluctuations will be discussed in the other chapters of this book.
Tremor Thalamotomy [20] and high frequency stimulation of ventral intermediate nucleus of thalamus (VIM) [21] are very effective for the treatment of PD tremor. In a multi-center study of thalamic DBS revealed that a robust symptomatic control of PD tremor in selected patients can be achieved and maintained during up to 6 years of follow up [22]. Excellent results in PD tremor have been also reported with STN stimulation [4,16–18,23]. There is some controversy as to which target should be used in patients with PD in whom large amplitude tremor is the main or only disabling feature. Although VIM has been traditional considered the optimal target for the treatment of essential tremor and other tremors, STN DBS has been demonstrated to be perhaps as effective as VIM DBS and may offer additional advantage of also improving other features of PD [23].
Bradykinesia Of all the motor symptoms of PD, bradykinesia appears to correlate best with nigrostriatal dopaminergic deficiency [24]. Bilateral STN and GPi stimulation have also been reported to improve bradykinesia and ‘‘off ’’ motor scores [4,16–18]. In a prospective study to measure upper extremity bradykinesia using a quantitative measure of angular velocity, Koop et al. reported a significant improvement in upper limb bradykinesia
1531
1532
91
Patient selection for surgery for parkinson’s disease
just after microelectrode recording before inserting or activating the DBS electrode in PD patients undergoing STN DBS [25]. Other authors provided electrophysiologic evidence that STN DBS increases strength and movement velocity at the ankle joint [26]. Although bradykinesia is not necessarily the main reason for selecting patients for surgery, it should not be considered a contraindication and, as noted, this cardinal PD symptom may also improve with DBS. Thus, in summary, the ideal surgical candidate for surgery, particularly DBS, would be a patient who, despite optimal levodopa therapy, continues to be troubled with levodopa-induced dyskinesias, motor fluctuations and disabling tremor. Surgery is also a consideration in those who could not tolerate the adverse effects of the various drugs.
Exclusion Criteria Atypical Parkinsonism Atypical clinical features indicative of multiple system involvement such as absence of tremor, the presence of supranuclear gaze palsy, dysautonomia, dementia, pyramidal signs, and poor response to levodopa suggest the presence of a neurodegenerative disease other than PD, respond poorly to surgical interventions [8].
Severe Brain Atrophy or Ischemic Changes on Limaging Studies Although mild cerebral atrophy and ischemic lesions are relatively common in the elderly, severe cerebral atrophy may pose technical difficulties during surgery and the presence of severe ischemic changes greatly increases operative risk. However, mild vascular changes on preoperative magnetic resonance imaging (MRI), have little or no effect on benefit from surgery [27].
Depression, Psychiatric Illness, Dementia, and Alcohol Abuse Besides the lack of insight and poor judgment that could impair their full understanding of the relative risks and benefits of surgery, patients with depression, dementia, or other neurobehavioral disorders are poor candidates for surgical intervention because their outcome may be difficult to assess. A meta-analysis of 10-year experience with DBS has concluded that the prevalence of depression was 2–4%, suicidal ideation/suicide attempt was 0.3–0.7%, and completed suicide rate was 0.16–0.32% over 2.4 years (compared to the overall 0.02% annual suicide rate in the US) [28]. Therefore, patients with history of significant psychiatric problems should be first referred for psychiatric assessment. Some centers actively screen for such problems using the structured clinical interview and detailed history taking. Depression is a common co-morbid condition in patients with PD, but when it markedly impacts on the overall functioning, despite optimal antidepressant therapy, it should be considered a reason to exclude such patients from consideration for surgery. There are several instruments used to assess depression, but the Hamilton Depression Scale (HAM-D), Beck Depression Inventory (BDI), Hospital Anxiety and Depression Scale (HADS), Montgomery-Asperg Depression Rating Scale (MADRS), and Geriatric Depression Scale (GDS) have been found particularly useful for screening purposes and HAM-D, MADRS, BDI and the Zung Self-Rating Depression Scale (SDS) have been recommended for assessment of severity [29]. Patients with dementia have been shown to have less overall benefit from DBS surgery. Mild cognitive deficits, frequently associated with advanced PD, may not, however, be an absolute contraindication to surgery. Studies have shown that there may be an initial worsening in short and long delay memory, executive functions and verbal fluency during the first 3 months
Patient selection for surgery for parkinson’s disease
postoperatively, but 12 months after surgery most of these functions return to baseline with the exception of speed of mental processing [30]. More recently, others have highlighted that bilateral STN DBS can worsen various aspects of frontal executive functioning, especially in the elderly patients [31,32]. However, long term follow-up studies have suggested that DBS STN does not predispose to dementia [33] and is associated with only minimal cognitive decline over 5 years [34]. A metaanalysis involving 28 cohort studies revealed small but significant declines in executive functions and verbal learning and memory. Moderate declines were only reported in semantic and phonemic verbal fluency [35]. Careful assessment of cognitive function must be part of the pre-surgical work-up and various factors (such as medications, ‘‘on’’ or ‘‘off’’ states etc) should be taken into consideration in the evaluation. The issue on the most appropriate neuro-cognitive screening tools pre-operatively has been debated. Some have suggested Mini Mental State Exam (MMSE) cutoff scores of 23 or 24 or scores on frontal/executive function screening batteries such as the Frontal Assessment Battery, while others have suggested more extensive battery of tests [36]. While MMSE has been used traditional to screen for cognitive deficits, it often fails to detect early cognitive decline because of its ceiling effect, and, therefore, the Montreal cognitive assessment (MoCA) has been developed to detect mild cognitive impairment in PD. In a study designed to compare the two scales in 88 patients with PD, the percentage of subjects scoring below a cutoff of 26/30 (used by others to detect mild cognitive impairment) was significantly higher on the MoCA (32%) than on the MMSE, suggesting that the MoCA may be a more sensitive tool to identify early cognitive impairment in PD [37]. Before a final decision is made whether a patient is a surgical candidate, a full neuropsychological assessment preoperatively for those who pass the MMSE or MoCA screening is recommended.
91
Advanced Age PD patients and caregivers frequency inquire about age as a potential contraindication for surgery. Uitti et al. [38] in an early study, found that the safety and benefits of pallidotomy were similar in the young (<65 years) and the older (<65 years, mean age 71.4 years) group of patients. Other authors have revealed greater improvement of the UPDRS motor scores in the ‘‘off ’’ state with age [39]. A recent study of DBS STN in 45 PD patients revealed a significant negative correlation between age and the improvement of three dimensions of PDQ 39 (mobility, activities of daily life, and cognition) at 24 months post surgery. Apathy and depression, but not cognitive impairment was positively correlated with age [40]. Derost et al. in a 2-year follow-up study of DBS STN found younger patients (<65 years) do better than older ones (>65 years) in terms of their motor function as assessed by various quality of life measures [3,41,42]. Another frequent concern is the risk of hemorrhage in elderly patients with hypertension. Sansur et al. [43] studied incidence of symptomatic hemorrhage after stereotactic electrode placement over a 15-year period. They found that history of hypertension was the most significant factor associated with hemorrhage, followed by older age, male sex, and a diagnosis of PD. Thus caution needs to be exercised in selecting elderly patients with a history of poorly controlled hypertension or other vascular risk factors. However, we feel that the state of patient’s physical and mental health, rather than age, should be the primary consideration in selecting suitable surgical candidates. Thus, there is no ‘‘cut-off ’’ age, and each patient’s risk/benefit ratio for surgery should be assessed individually. Because of high operative risk and low life expectancy, most surgeons do not operate on patients above 80 years of age. Patients with terminal illness or systemic disease
1533
1534
91
Patient selection for surgery for parkinson’s disease
associated with organ dysfunction are usually excluded because of the limited life span and higher surgical risk of complications.
available, and thus some patients would choose lesion over DBS surgery after properly informed discussions by the medical team.
Practical Considerations
Specific Considerations
While standard medical criteria are frequently emphasized in the pre-operative selection of patients, numerous individual ‘‘patient’’ factors need to be seriously considered in the evaluation process. Of these, patients’ or caregivers’ unrealistic expectations is one of the most frequent reasons for unhappiness with surgical outcome and this can complicate post-operative management. It is important for the medical team to properly educate the patients and their caregivers and help them understand what their realistic expectations from the surgery should be. Any false expectations of an immediate fix for their symptoms may result in disappointment. Furthermore, the patients should understand that the levodopa-resistant symptoms such as ‘‘on’’ freezing, postural instability, and dysarthria usually do not significantly improve with surgery, and these observations have to be clearly spelled out preoperatively. Some authors have developed a mnemonic device which may be useful in patient education [44]. Post-operative management for DBS surgery also requires considerable time and commitment by both patients and their caregivers. These involve frequent travels to the medical center for time consuming adjustments of DBS parameters, especially in the initial 3–6 months period. Other issues such as social and financial problems, cultural beliefs and practices and availability for long term follow-up can also be confounding factors for post-operative management. This is especially so for international centers which manage patients referred to them from different parts of the world. In many third world countries, post-operative DBS adjustment expertise is not
Despite the general guidelines and the ideal patient profile for PD surgery, dealing with potential PD surgery candidates in clinical practice is not always so straight forward. In some instances, the medical team can also consider a patient for surgery even if he or she falls outside the common inclusion and exclusion criteria. For instance, there may be strong humanitarian or medical reasons to carry out a lesion or DBS surgery even if the patient has certain relative contraindications (such as cognitive dysfunction, advanced age etc). Patients with medically refractory disabling tremor, those who cannot tolerate the side effects of dopaminergic medications, and those with severe disabling dyskinesias could be considered even if they happen to have other contraindications. The final decision is based on the understanding between the medical team and the patients and their caregivers. Importantly, details of the collective recommendation of the multidisciplinary team, the risk/ benefits and the primary objective for the surgery should be fully documented. There may also be instances where patients have had previous functional surgery but additional DBS procedures need not be a contraindication for them. These patients include those with previous unsuccessful stimulation due to misplaced DBS electrodes or unilateral pallidotomy. Those with runaway dyskinesias after neurotransplantation, GPi DBS contralateral to the most disabling dyskinesias may be suitable. Those with clearly defined genetic forms of PD can also be considered as long as they satisfy the inclusion and exclusion criteria as applied to the sporadic PD patients.
Patient selection for surgery for parkinson’s disease
Future Considerations It is evident that other than medical indications, many factors such as general medical conditions, culture and social-related issues, development of dementia and psychiatric problems, and lifestyle changes need to be considered in selecting the most appropriate candidate for PD surgery. Studies are needed to examine how and to what extent these factors influence the final outcome of surgery. It would also be useful to examine the correlation of imaging abnormalities (such as severity of small vessel disease and atrophy) with postoperative adverse events and outcomes. These investigations should also be extended using advanced functional imaging techniques. Prospective studies are needed to better define the group of ‘‘mild cognitive impairment’’ patients who might benefit from surgery and the most ideal neuropsychological assessment tools that should be applied. Although DBS surgery is currently preferred over lesion surgery, there is still a role for the latter in some instances. Even though most centers seem to prefer STN as the target for DBS in patients with PD, both STN and DBS are generally regarded as excellent targets. In fact, patients who are mostly troubled by dyskinesias or whose cognitive functioning is borderline, might be better candidates for GPi DBS as the latter target seems to be particularly effective for controlling dyskinesias and may be less frequently associated with psychiatric complications. If drug reduction is the primary goal, however, then STN may be the preferred target. Surgery for PD is continuously evolving. Further studies to evaluate additional surgical targets (or a combination of known targets) or improving current surgical techniques aimed at relieving specific symptoms or at a subset of PD patients may improve the pre-operative selection process. In addition to improvements in techniques, novel targets, such as pedunculopontine
91
nucleus, and caudal part of the zona incerta, are being explored for the treatment of specific symptoms [45,46]. A standardized evaluation pre- and post-surgical protocol would be most useful (on and off assessments of the UPDRS ADL and motor sections, dyskinesia severity and duration, on and off diary cards differentiating on function with and without disabling dyskinesias). Quality of life and other specific scales aimed at selected symptoms would also be useful additions. Reports, such as the one from ‘‘Consensus on DBS for PD’’ conference, commissioned by the Congress of Neurological Surgeons and the Movement Disorder Society (MDS) and endorsed by the Scientific Issues Committee of the MDS and the American Society of Stereotactic and Functional Neurosurgery, create better awareness among the scientific community with regards to the various aspects of DBS preoperative decisionmaking process [36]. Other individual groups have also developed and validated screening tools, such as the Florida Surgical Questionnaire for Parkinson Disease (FLASQ-PD) for potential surgical candidates [47]. This brief questionnaire is designed to screen for ‘‘red flags’’ that may indicate potential unfavorable outcome. The highest FLASQ-PD score is 34 with zero red flags, and the lowest/best possible score is zero with 8 red flags. A score of about 25 with no red flags indicates a potential good surgical subject. The FLASQ-PD can be administered by a general or nurse practitioner or physician assistant, and potential surgical candidates can then be referred to the medical team for further evaluation. These efforts to improve the selection of candidates are likely to be expanded and improved upon in the near future. Prospective cost effect analysis comparing DBS and medical therapy will be most invaluable for the patient and the healthcare sponsors and providers [48]. The impact is likely to be greater among the younger patients who are still economically active. A recent study suggests
1535
1536
91
Patient selection for surgery for parkinson’s disease
that DBS may be cost effective in treating PD if quality of life improves 18% or more compared with those receiving best medical management [49]. While the beneficial effects of STN DBS have been clearly demonstrated in advanced PD and the procedure is currently performed after a mean symptom duration of 14 years, when severe motor complications have resulted in marked loss of quality of life, increasing number of studies are attempting to address the question whether patients in earlier stages of the disease should be considered for this surgical treatment [50,51]. In this light, future studies should attempt to address the risk/benefit ratio (quality of life, psychological problems, surgical complications etc) of operating on patients prior to the development of troublesome or disabling motor complications. It would also be interesting to examine whether early surgery modulates the risk of developing non motor complications. In conclusion, functional stereotactic PD surgery should be performed in specialized centers where expertise, technology, equipment, and skilled personnel are available to ensure the best outcome possible. This will ensure not only the most optimal benefits but also reduced the risk of immediate and long-term postoperative complications [52]. While there are general indications for this type of surgery, the actual selection process is highly individualized, catering to each patient’s specific needs and expectations. With greater experience and new developments in this field, PD sufferers can look forward to a better quality of life in the horizon.
Acknowledgments The National Neuroscience Institute, Singapore and Baylor College of Medicine PD and MD Clinic are National Parkinson Foundation (NPF) International Centers of Excellence, and their support is acknowledged.
References 1. Jankovic J. Motor fluctuations and dyskinesias in Parkinson’s disease: clinical manifestations. Mov Disord 2005;20 Suppl 11:S11-S16. 2. Jankovic J, Stacy M. Medical management of levodopaassociated motor complications in patients with Parkinson’s disease. CNS Drugs 2007;21(8):677-92. 3. Diamond A, Jankovic J. The effect of deep brain stimulation on quality of life in movement disorders. J Neurol Neurosurg Psychiatry 2005;76(9):1188-93. 4. Deuschl G, Schade-Brittinger C, Krack P, Volkmann J, Scha¨fer H, et al. German Parkinson Study Group, Neurostimulation Section. A randomized trial of deep-brain stimulation for Parkinson’s disease. N Engl J Med 2006;355(9):896-908. 5. Tan EK, Jankovic J. Movement disorder surgery: patient selection and evaluation of surgical results. In: Lozano AM, editor. Movement disorder surgery: progress and challenges, progress in neurological surgery. Basel, Switzerland: Karger; 2000. p. 78-90. 6. Rodriguez RL, Fernandez HH, Haq I, Okun MS. Pearls in patient selection for deep brain stimulation. Neurologist 2007;13:253-60. 7. Halpern C, Hurtig H, Jaggi J, Grossman M, Won M, Baltuch G. Deep brain stimulation in neurologic disorders. Parkinsonism Relat Disord 2007;13:1-16. 8. Shih LC, Tarsy D. Deep brain stimulation for the treatment of atypical parkinsonism. Mov Disord 2007;22: 2149-55. 9. Gelb DJ, Oliver E, Gilman S. Diagnostic criteria for Parkinson’s disease. Arch Neurol 1999;56:33-9. 10. Langston JW, Widner H, Goetz CG, Brooks D, Fahn S, Freeman T, Watts R. Core assessment program for intracerebral transplantations (CAPIT). Mov Disord 1992;7:2-13. 11. Jankovic J, Rajput AH, McDermontt MP, Perl DP. Evolution of diagnosis in early Parkinson’s disease. Arch Neurol 2000;57(3):369-72. 12. Kazumata K, Antonini A, Dhawan V, Moeller JR, Alterman RL, Kelly P, Sterio D, Fazzini E, Beric A, Eidelberg D. Preoperative indicators of clinical outcome following stereotaxic pallidotomy. Neurology 1997;49: 1083-90. 13. Piboolnurak P, Lang AE, Lozano AM, Miyasaki JM, Saint-Cyr JA, Poon YY, Hutchison WD, Dostrovsky JO, Moro E. Levodopa response in long-term bilateral subthalamic stimulation for Parkinson’s disease. Mov Disord 2007;22(7):990-7. 14. Lyons KE, Davis JT, Pahwa R. Subthalamic nucleus stimulation in Parkinson’s disease patients intolerant to levodopa. Stereotact Funct Neurosurg 2007;85(4): 169-74. 15. Wider C, Russmann H, Villemure JG, Robert B, Bogousslavsky J, Burkhard PR, Vingerhoets FJ.
Patient selection for surgery for parkinson’s disease
16. 17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
Long-duration response to levodopa in patients with advanced Parkinson disease treated with subthalamic deep brain stimulation. Arch Neurol 2006;63(7):951-5. Walter BL, Vitek JL. Surgical treatment for Parkinson’s disease. Lancet Neurol 2004;3(12):719-28. Krack P, Batir A, Van Blercom N, Chabardes S, Fraix V, Ardouin C, Koudsie A, Limousin PD, Benazzouz A, LeBas JF, Benabid AL, Pollak P. Five-year follow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinson’s disease. N Engl J Med 2003;349 (20):1925-34. Weaver F, Follett K, Hur K, Ippolito D, Stern M. Deep brain stimulation in Parkinson disease: a metaanalysis of patient outcomes. J Neurosurg 2005; 103(6):956-67. Rodrigues JP, Walters SE, Watson P, Stell R, Mastaglia FL. Globus pallidus stimulation improves both motor and nonmotor aspects of quality of life in advanced Parkinson’s disease. Mov Disord 2007;22(13):1866-70. Jankovic J, Cardoso F, Grossman RG, Hamilton WJ. Outcome after stereotactic thalamotomy for parkinsonian, essential and other types of tremor. Neurosurgery 1995;45:1743-6. Ondo WG, Jankovic J, Schwartz K, Almaguer M, Simpson RK. Unilateral thalamic deep brain stimulation for refractory essential tremor and Parkinson’s disease tremor. Neurology 1998;51:1063-9. Hariz MI, Krack P, Alesch F, Augustinsson LE, Bosch A, Ekberg R, Johansson F, Johnels B, Meyerson BA, Nguyen JP, Pinter M, Pollak P, von Raison F, Rehncrona S, Speelman JD, Sydow O, Benabid AL. Multicentre European study of thalamic stimulation for Parkinsonian tremor: a 6-year follow-up. J Neurol Neurosurg Psychiatry 2008;79:694-9. Diamond A, Shahed J, Jankovic J. The effects of subthalamic nucleus deep brain stimulation on parkinsonian tremor. J Neurol Sci 2007;260:199-203. Vingerhoets FJ, Schulzer M, Calne DB, Snow BJ. What clinical signs of Parkinson’s disease best reflects the nigrotriatal lesion? Ann Neurol 1997;41:58-64. Koop MM, Andrzejewski A, Hill BC, Heit G, BronteStewart HM. Improvement in a quantitative measure of bradykinesia after microelectrode recording in patients with Parkinson’s disease during deep brain stimulation surgery. Mov Disord 2006;21(5):673-8. Vaillancourt DE, Prodoehl J, Sturman MM, Bakay RA, Metman LV, Corcos DM. Effects of deep brain stimulation and medication on strength, bradykinesia, and electromyographic patterns of the ankle joint in Parkinson’s disease. Mov Disord 2006;21(1):50-8. Desaloms JM, Krauss JK, Lai EC, Jankovic J, Grossman RG. Posteroventral medial pallidotomy for treatment of Parkinson’s disease: preoperative magnetic resonance imaging features and clinical outcome. J. Neurosurg 1998;89:194-9.
91
28. Appleby BS, Duggan PS, Regenberg A, Rabins PV. Psychiatric and neuropsychiatric adverse events associated with deep brain stimulation: a meta-analysis of ten years’ experience. Mov Disord 2007;22:1722-8. 29. Schrag A, Barone P, Brown RG, et al. Depression rating scales in Parkinson’s disease: critique and recommendations. Mov Disord 2007;22:1077-92. 30. Rettig GM, Lai EC, Krauss JK, Grossman RG, Jankovic J. Neuropsychological evaluation of patients with Parkinson’s disease before and after pallidal surgery. In: Krauss JK, Grossman RG, Jankovic J, editors. Pallidal surgery for the treatment of Parkinson’s disease and movement disorders. Philadelphia: Lipincott-Raven; 1998. p. 211-31. 31. Saint-Cyr JA, Tre´panier LL, Kumar R, Lozano AM, Lang AE. Neuropsychological consequences of chronic bilateral stimulation of the subthalamic nucleus in Parkinson’s disease. Brain 2000;123 (Pt 10):2091-108. 32. Tre´panier LL, Kumar R, Lozano AM, Lang AE, SaintCyr JA. Neuropsychological outcome of GPi pallidotomy and GPi or STN deep brain stimulation in Parkinson’s disease. Brain Cogn 2000;42(3):324-47. 33. Ory-Magne F, Brefel-Courbon C, Simonetta-Moreau M, Fabre N, Lotterie JA, Chaynes P, Berry I, Lazorthes Y, Rascol O. Does ageing influence deep brain stimulation outcomes in Parkinson’s disease? Mov Disord 2007; 22(10):1457-63. 34. Contarino MF, Daniele A, Sibilia AH, Romito LM, Bentivoglio AR, Gainotti G, Albanese A. Cognitive outcome 5 years after bilateral chronic stimulation of subthalamic nucleus in patients with Parkinson’s disease. J Neurol Neurosurg Psychiatry 2007;78:248-52. 35. Parsons TD, Rogers SA, Braaten AJ, Woods SP, Tro¨ster AI. Cognitive sequelae of subthalamic nucleus deep brain stimulation in Parkinson’s disease: a meta-analysis. Lancet Neurol 2006;5(7):578-88. 36. Lang AE, Houeto JL, Krack P, Kubu C, Lyons KE, Moro E, Ondo W, Pahwa R, Poewe W, Tro¨ster AI, Uitti R, Voon V. Deep brain stimulation: preoperative issues. Mov Disord 2006;21 Suppl 14:S171-S196. 37. Zadikoff C, Fox SH, Tang-Wai DF, Thomsen T, de Bie RM, Wadia P, Miyasaki J, Duff-Canning S, Lang AE, Marras C. A comparison of the mini mental state exam to the montreal cognitive assessment in identifying cognitive deficits in Parkinson’s disease. Mov Disord 2008;23 (2):297-9. 38. Uitti RJ, Wharen RE Jr, Turk MF, Lucas JA, Finton MJ, Graff-Radford NR, Boylan KB, Goerss SJ, Kall BA, Adler CH, Caviness JN, Atkinson EJ. Unilateral pallidotomy for Parkinson’s disease: comparison of outcome in younger versus elderly patients. Neurology 1997;49:1072-7. 39. Kishore A, Turnbull IM, Snow BJ, de la FuenteFernandez R, Schullzer M, Mark E, Yardley S, Calne DB. Efficacy, stability and predictors of outcome of pallidotomy for Parkinson’s disease: six-month follow-up
1537
1538
91 40.
41.
42.
43.
44.
45.
46.
Patient selection for surgery for parkinson’s disease
with additional 1-year observations. Brain 1997;120: 729-37. Aybek S, Gronchi-Perrin A, Berney A, Chiuve´ SC, Villemure JG, Burkhard PR, Vingerhoets FJ. Long-term cognitive profile and incidence of dementia after STN-DBS in Parkinson’s disease. Mov Disord 2007; 22(7):974-81. Derost PP, Ouchchane L, Morand D, Ulla M, Llorca PM, Barget M, Debilly B, Lemaire JJ, Durif F. Is DBS-STN appropriate to treat severe Parkinson disease in an elderly population? Neurology 2007;68(17):1345-55. Martinez-Martin P, Deuschl G. Effect of medical and surgical interventions on health-related quality of life in Parkinson’s disease. Mov Disord 2007;22:757-65. Sansur CA, Frysinger RC, Pouratian N, Fu KM, Bittl M, Oskouian RJ, Laws ER, Elias WJ. Incidence of symptomatic hemorrhage after stereotactic electrode placement. J Neurosurg 2007;107(5):998-1003. Okun MS, Foote KD. A mnemonic for Parkinson disease patients considering DBS: a tool to improve perceived outcome of surgery. Neurologist 2004;10(5):290. Hamani C, Stone S, Laxton A, Lozano AM. The pedunculopontine nucleus and movement disorders: anatomy and the role for deep brain stimulation. Parkinsonism Relat Disord 2007;13:S276-S280. Stefani A, Lozano AM, Peppe A, et al. Bilateral deep brain stimulation of the pedunculopontine and
47.
48.
49.
50.
51.
52.
subthalamic nuclei in severe Parkinson’s disease. Brain 2007;130:1596-607. Okun MS, Fernandez HH, Pedraza O, Misra M, Lyons KE, Pahwa R, Tarsy D, Scollins L, Corapi K, Friehs GM, Grace J, Romrell J, Foote KD. Development and initial validation of a screening tool for Parkinson disease surgical candidates. Neurology 2004;63(1):161-3. Martinez-Martin P, Deuschl G. Effect of medical and surgical interventions on health-related quality of life in Parkinson’s disease. Mov Disord 2007;22:757-65. Tomaszewski KJ, Holloway RG. Deep brain stimulation in the treatment of Parkinson’s disease: a cost-effectiveness analysis. Neurology 2001;57(4):663-71. Diamond A, Jankovic J. Quality of life and cost effectiveness of deep brain stimulation in movement disorders. In: Vitek J, Starr P, Okun M, Tarsy D, editors. Deep brain stimulation for neurological and psychiatric disorders. Current clinical neurology series. Totowa, NJ: Human Press; 2008 (in press). Schupbach WM, Maltete D, Houeto JL, et al. Neurosurgery at an earlier stage of Parkinson disease: a randomized, controlled trial. Neurology 2007;68:267-71. Kenney C, Simpson R, Hunter C, Ondo W, Almaguer M, Davidson A, Jankovic J. Short-term and long-term safety of deep brain stimulation in the treatment of movement disorders. J Neurosurg 2007;106: 621-5.
112 Peripheral Procedures for Cervical Dystonia T. Taira
Introduction Cervical dystonia is a type of focal dystonia in which simultaneous and sustained contraction of both agonist and antagonist muscles are confined only to the neck [1]. The term ‘‘spasmodic torticollis’’ has been used historically for many years. However, because the symptom is not always ‘‘spasmodic’’ and the etiology of this condition is proven to be a type of dystonia due to dysfunction of central motor control, the term ‘‘cervical dystonia (CD) is more appropriate. Spasmodic torticollis is sometimes mistaken with other non-dystonic abnormal neck posture like congenital muscular contracture. In 1930, Dandy [2] scorned that medical students and even doctors were often misled into the belief that a psychogenic factor is an important, even the sole, cause of spasmodic torticollis. Cooper [3] wrote an autobiographical article ‘‘Victim is Always the Same’’ [4] indicating that many patients with dystonia were misdiagnosed as psychogenic origin despite the fact that surgical interventions to the motor part of the pallidum or thalamus alleviate the symptoms. The prevalence of cervical dystonia ranges between 9 and 30 per 100,000 [5,6] and this is the most frequent among focal dystonias. Women are affected 1.3–2.0-fold more than men. According to a series of 266 CD patients of Chan et al. [7], family history was positive Electronic supplementary material Supplementary material is available in the online version of this chapter at http:// dx.doi.org/10.1007/978-3-540-69960-6_112 and is accessible for authorized users. #
Springer-Verlag Berlin/Heidelberg 2009
in 12%, and remission was seen in 9.8% of the patients.
History of Surgical Treatment Peripheral procedure including intradural rhizotomy for CD has a long history. The ancient Greeks are reported to have sectioned the SCM [8]. Academic record indicates that German surgeon, Issac Minnius, performed sectioning of SCM in 1641. In 1867 de Morgan first reported surgery to the accessory nerve, while Bujalski had reported ligation of the XIth nerve in 1834. Collier invented silver wire constriction of the XIth nerve in 1890. Thus, until about 1890, CD or torticollis had been considered as a disorder of XIth nerve or the SCM muscle. In 1891, Keen [9] noticed importance of posterior neck muscles in CD and performed sectioning of the posterior neck muscles with denervation of the upper three cervical nerves. Several years later, de Quervain reported resection of both SCM and muscles in the posterior suboccipital triangle. Based on the Foerster’s experience of posterior rhizotomy for spasticity, Taylor reported posterior rhizotomy of the upper four cervical roots in 1915, but his attempts proved unsuccessful. McKenzie, working with Cushing in 1923, sought a simpler method of denervation of neck muscles by cutting the upper cervical anterior roots intradurally [10]. He performed this procedure unilaterally with partial sectioning of the accessory nerve. In his operation, because of Taylor’s influence, posterior roots were also sacrificed. Dandy, in 1930, reported successful
1886
112
Peripheral procedures for cervical dystonia
relief of torticollis in seven out of eight patients with bilateral upper three anterior rhizotomy [2]. Later he added peripheral denervation of the XIth nerve instead of intradural section. In 1949 Putnam et al. reported successful treatment of 17 out of 18 patients with McKenzie-Dnady technique, and in 1951 Poppen reported more encouraging results in 36 cases [11]. As the cases of intradural rhizotomy accumulated, it was proven safe to cut the upper four cervical anterior roots on the more involved side and the upper three on the other, without affecting the respiratory function of the diaphragm. Initially McKenzie had been sectioning the XIth nerve intradurally, but he, like Dandy, soon came to prefer peripheral selective denervation of SCM to preserve trapezius branch and to avoid postoperative dysfunction of the shoulder. Sorensen and Hamby [12,13] then reported their experience in 71 cases, and Hamby and Schiffer [14,15] one of 50 more. There were three death, and one-third of the patients suffered dysphasia, about 60% complained of continuing pain, and two-thirds developed paralysis of the trapezius muscle even after selective peripheral denervation of SCM. Taker [8] reported his 73 consecutive cases in 1976, showing initial good results in 29 among 47 patients who underwent both peripheral denervation of SCM and bilateral upper cervical anterior rhizotomy. Two patients developed temporally hemiparesis and hypesthesia, probably due to cord ischemia. Intradural high cervical anterior rhizotomy ‘‘McKenzie-Dandy operation’’ became a standard surgical treatment for many for CD. This procedure was widely performed all over the world to over several hundred patients in 1960s and 1970s [16]. However, this surgical procedure was almost abandoned after introduction of extradural denervation procedure in early 1980s. In the intradural procedure, because the rootlets from the spinal cord contain fibers innervating both anterior and posterior branches of the spinal nerve and because, during surgery, we can
not effectively distinguish posterior components from anterior, rhizotomy should be confined to C1–C3 or C4 to avoid phrenic nerve (C3, C4). Complications associated with anterior rhizotomy were not negligible; neck instability, dysphasia, trapezius paralysis, spinal cord infarction, and so on are documented [17]. Hernesniemi and Keranen [18] studied long-term outcome of myotomy and intradural rhizotomy in 1990. The mean follow-up time was 4 years, and only two out of 23 patients had a good overall outcome in contrast to the poor or very poor results in 13 of the patients. They conclude that both myotomy and high cervical anterior rhizotomy seemed to be only rarely indicated for surgical treatment of spasmodic torticollis. Friedman et al. [16], on the contrary, reported more favorable effects of ventral cervical and selective spinal accessory nerve rhizotomy in 58 patients. Forty-nine patients (85%) had a marked improvement in their condition, with 33 (57%) attaining an excellent result and 16 (28%) noting significant improvement. Muscle spasms were completely relieved in 42 patients (72%) and markedly reduced in 10 (17%). Thirty-four patients (59%) reported that their resting head posture was restored to a neutral position. The likelihood that a patient’s head posture returned to normal was inversely proportional to the preoperative duration of the spasmodic torticollis. Twenty-six patients (45%) suffered mild transient difficulty with swallowing solid foods in the immediate postoperative period. In most cases these minor difficulties abated in the months following surgery. The problem is that the effect of intradural rhizotomy was not evaluated in any report by using a validated standard rating scale, and assessment tended to be subjective. Another surgical approach to CD was stereotactic operations targeting the pallidum or the thalamus. Numerous reports appeared in 1960s. Cooper [19] reported in 1964 that he obtained poor results with unilateral or bilateral lesions in the globus pallidus or ventrolateral nucleus of
Peripheral procedures for cervical dystonia
the thalamus but good results when bilateral lesions were made in the ventrolateral and ventroposterior nuclei and centrum medianum. Hassler and Dieckmann [20,21] found that their results were unsatisfactory when the lesions were made in the nucleus ventralis oralis internus alone but that, in 16 patients, successful results were obtained when the lesion extended inferiorly into Forel H1 area especially in whom head turning was horizontal. However, still it was difficult to approach the stereotactic treatment of spasmodic torticollis with conviction. Because of the recent success of deep brain stimulation for the treatment of generalized dystonia, some clinical trials to study the effect of deep brain stimulation for CD have been reported with promising results [22–28]. In 1970s, Claude Bertrand in Montreal had been involved extensively in stereotactic operations for various types of movement disorders including CD [29]. However, he was not satisfied
112
with the results of thalamotomy for CD and he started more extensive denervation of posterior neck muscles by approaching to the extradural peripheral nerves [30–32] (> Figure 112-1). Bertrand wrote as follows [29]. ‘‘Undoubtedly, if thalamotomy had produced more consistent results, we would not have been tempted to replace it with peripheral denervation. When performing the first few procedures, we were afraid that denervation would suppress not only the abnormal movements but also normal movements of the neck, but we were soon reassured that there are enough proximal collateral branches to maintain turning of the head to the side operated on.’’ In the previous edition Textbook of Stereotactic and Functional Neurosurgery published in 1998, very little was mentioned about Bertrand procedure. Even in 1998, Tasker [33] noted ‘‘Though not as widely utilized, Bertrand’s technique of open posterior ramus section appears to offer certain advantages, and
. Figure 112-1 Operation field of traditional Bertrand procedure for posterior neck muscle denervation (left) and sternocleidomastoid mescle denervation (right)
1887
1888
112
Peripheral procedures for cervical dystonia
. Figure 112-2 Anatomical comparison of denervation between intradural and extradural procedures
it is hoped that more widespread experience with this technique will become available.’’ Since then, there appeared many and worldwide reports of favorable results of Bertrand’s selective peripheral denervation for CD [34–43], and it is now a standard and established procedure for medically refractory CD. > Figure 112-2 shows the locations of denervation in intradural and extradural procedures. > Figure 112-3 shows cross section of the neck indicating the innervation pattern of the muscles (left). More selective complete denervation is accomplished in the selective peripheral denervation (right, upper), while incomplete and unnecessary denervation occurs in intradural rhizotomy (right, lower).
Types and Features of Cervical Dystonia CD is characterized by involuntary sustained abnormal contractions of cervical muscles that produce abnormal head movements or postures.
There are 27 pairs of muscles in the neck, and abnormal contractions of muscles occur with various combinations of these muscles, resulting in different abnormal head and neck postures. There are four major typical postures as shown in > Figure 112-4. The contemporary nomenclature defines rotation as movement of the chin to the opposite shoulder in the horizontal plane, tilt (laterocollis) as moving the ear to the ipsilateral shoulder, and flexion/extension in the sagittal plane as anterocollis/retrocollis. Transition of the axis of the head from the axis of the body may be called lateral or sagittal shift. The muscles responsible for these symptoms are summarized in > Table 112-1. The most common type of CD is horizontal rotation type in which the sternocleidomastoid muscle and the contralateral posterior neck muscles mainly the splenius capitus muscle show abnormal involuntary contraction. Often there are also tremors or jerky movements of the head. About half of the patients have essential tremor like tremor in the upper extremities. The frequency of each of CD symptoms has been studied, indicating
Peripheral procedures for cervical dystonia
112
. Figure 112-3 Innervation of cervical muscles (left) and comparison of muscles denervated in intradural and extradural procedures (right)
. Figure 112-4 Classification of symptoms of cervical dystonia
that rotation component is the commonest (82–97%), though pure rotation is less common (19–37%). Laterocollis is the second most common type, occurring in 42–74%, and predominant lateroccolis was seen in 18%. Pure laterocollis is rare, noted in 2–3.8%. Retrocollis occurs in 24–38%. ‘‘Sensory trick’’ (gestes antagonistiques) is a pathognomonic feature of CD. This is transient correction of head position by
maneuvers such as touching various locations on the face, neck, or head with the hand or fingers. Almost all patients with CD acquire such kinds of sensory trick and use to relieve the symptoms. Electromyography (EMG) may be useful to identify the abnormally contracting muscles. Podivinsky [44] classified spasmodic torticollis in five types based on EMG patterns (> Figure 112-5). The problem of EMG recording in CD patients
1889
1890
112
Peripheral procedures for cervical dystonia
. Table 112-1 Muscles involved in CD Symptom
Responsible muscles
Horizontal rotation
Unilateral sternocleidomastoid Contralateral splenius Inferior oblique Semispinalis cervicalis Bilateral posterior cervical muscles Semispinalis capitus Semispinalis cervicalis Splenius Suboccipital rectus (major and minor) Bilateral trapezius Sternocleidomastoid Levator scapulae Scalene Semispinalis capitus Semispinalis cervicalis Superior oblique Trapezius Bilateral sternocleidomastoid Longus colli Submental
Retrocollis
Laterocollis
Antecollis
. Figure 112-5 Classification of cervical dystonia based on electromyography pattern. A: sternocleidomastoid type, B:splenius type, C: trapezius type, D: unilateral type, E: bilateral diffuse type
is that the symptom is not often marked in supine or reclining sitting position generally used in electrophysiology suite. Therefore, EMG should be recorded in various postures that make
the symptom more marked. Another problem is that we can not clear distinction of abnormal contractions of the trapezius and splenius muscles with surface electrodes, because these
Peripheral procedures for cervical dystonia
muscles are too closely located. I stopped using EMG as general preoperative evaluation for CD. I believe that the most reliable method to identify the target muscles to be denervated is palpation of the neck muscles. If the symptom is more marked while walking, I walk along with the patient with my fingers palpating the neck muscles. There are 27 muscles in the neck, and among them, only six pairs (SCM, TRP, SPL,
112
LS, SScap, scalene) are easily palpable on clinical examination. This is, however, enough to determine which muscles should be denervated. Nowadays, some kind of validated rating scale is essential to evaluate the CD symptoms and effect of any treatment modalities. For this purpose, Toronto Western Spasmodic rating Scale (TWSTRS, > Table 112-2) [45] and Tsui score (> Table 112-3) [46] are commonly used. If a
. Table 112-2 Toronto Western Spasmodic rating Scale (TWSTRS) [45] Severity items Rotation
Tilt
Anterocollis or retrocollis
Range of motion and effort (rate for plane of motion that appears most limited)
Tremor and jerking
Global assessment severity
0 None 1 Mild (1/3 range) 2 Moderate (1/3 to 2/3 range) 3 Severe (>2/3 range) 0 None 1 Mild (<15 degrees) 2 Moderate (16–35 degrees) 3 Severe (>35 degrees) 0 None 1 Mild downward deviation 2 Moderate (1/2 range) 3 Severe 0 Full range of motion, without obvious effort 1 Can move head well past midline, or can achieve full range of motion only be exerting obvious effort 2 Not able to move head past midline, or past midline only with great effort 3 Barely able to move head past the abnormal posture, or moves past abnormal posture briefly only with great effort 0 Absent 1 Mild: occasional 2 Moderate: often present 3 Severe: almost always 0 None 1 Mild 2 Moderate 3 Severe Duration factor 0 Dystonia not present 1 Occasional dystonia (<25% of the time), usually submaximal in severity 2 Intermittent (<50% of time), usually submaximal or occasional (<25%) maximal dystonia 3 Frequent (>50%), usually submaximal or intermittent maximal 4 Constant dystonia or frequent maximal 5 Constant maximal dystonia
1891
1892
112
Peripheral procedures for cervical dystonia
. Table 112-3 Tsui score A. Amplitude of sustained movements A1: Rotation A2: Lateral head tilt A3: Antero/retrocollis B. Duration of sustained movements Intermittent: 1, Constant: 2 C. Axial distortion C1: Scoliosis C2: Shoulder elevation/depression D. Unsustained head movements (head tremor/jerk) D1: Severity none:0 moderate:1 severe:2 D2: Duration none:0 occasional:1 continuous:2 Total score = (A1 + A2 + A3) X B + C1 + C2 + D1 + D2
Absent 0 0 0
<15 1 1 1
<30 2 2 2
<45 3 3 3
Absent 0 Absent 0
<15 1 <7 1
<30 2 <15 2
>30 3 >15 3
>45 4 4 4
Definition of therapeutic effect with modified Tsui ScoreExcellent: score decrease >10 or final score = 0Good: 5 = < score decrease < = 9Fair: 3 = < score decrease < = 4No change: 2 = < score change < = 2Worse: score increase > = 3
patient has extra-cervical symptoms, which may be seen occasionally in segmental or generalized dystonia, Burke-Fahn-Marsden scale [47] is more appropriate. There is no consensus on which evaluation method is the best in case of selective peripheral denervation. Odergren et al. [48] compared various types of evaluation methods in cases of CD treated with botulinum injections. They suggested that the efficacy of treatment in CD is best evaluated by using a combination of the visual analogue scale for pain and the Tsui score for dystonic posture and movement ability.
Innervation of the Cervical Muscles Posterior neck muscles are most often involved in CD, and their innervation comes mainly from the posterior branches of C1–C8 cervical nerves. > Table 112-4 summarizes the innervation pattern
of the major cervical muscles, and > Figure 112-3 (left) shows the cross section of the neck indicating important muscles in selective peripheral denervation for CD.
Indication of Surgical Treatment In the management of CD, botulinum toxin injection to the abnormally contracting muscles is generally accepted as the first treatment of choice. Some medications such as benzodiazepines, anticholinergics, antidepressants and mexilletine may be used, but the effect is usually mild and modest. As in other surgical treatment of movement disorders, conservative treatment including botulinum injections should be tried first. Neurologists who have enough experience and are specialized in this field should do botulinum treatment. The injections should be done at least 4–5 times and this may take more than 12 months.
Peripheral procedures for cervical dystonia
112
. Table 112-4 Innervation of major cervical muscles Platysma
Facial nerve
Sternocleidomastoid Trapezius Semispinalis capitus Semispinalis cervicalis Splenius Suboccipital rectus (major and minor) Superior oblique Inferior oblique Levator scapulae Scalene muscles Longus colli Longus capitus Rectus capitus anterior Rectus capiyus lateralis
Accessory nerve Accessory nerve Posterior branches of C1–C8 Posterior branches of C1–C8 Posterior branches of C3–C6 Posterior branches of C1–C2 Posterior branches of C1–C2 Posterior branches of C1–C3 Anterior branches of C3, C4 Anterior branches of C5–C8 Anterior branches of C1–C3 Anterior branches of C1–C3 Anterior branches of C1 Anterior branches of C1
If botulinum toxin fails in the treatment of cervical dystonia, selective peripheral denervation is now accepted as the best surgical option. When the patient is unwilling to have botulinum injections because of financial problem or social circumstances, we can offer surgical treatment as an option. The National Institute for Clinical Excellence of the UK has produced a guideline for selective peripheral denervation in cervical dystonia which was issued in August 2004 [49]. Selective peripheral denervation should not be confused with intradural rhizotomy, which has a higher incidence of complications; it is indicated in patients with cervical dystonia who do not achieve adequate response with medical treatment or repeated botulinum injections. It is indicated in nonresponders to botulinum injections. Additional myectomy may be carried out if necessary. Patients with prominent (phasic or myoclonic) dystonic movements or with dystonic head tremor are not good candidates for this procedure, and such patents should be treated with pallidal deep brain stimulation [22–28,50,51]. In some patients selective peripheral denervation can also be an alternative to botulinum injections. Overall, about one to two-thirds of patients achieve useful long-term improvement. This proportion has been higher, up to 90%, in some studies [32];
however, it is unclear how follow-up was performed in these studies. In traditional Bertrand procedure, Denervation of C2 invariably involves numbness in the territory of the greater occipital nerve in the early post-operative period. Patients should be informed about the invariable procedure related numbness; neuropathic pain can develop. Swallowing difficulties have been noted in some studies. In about 1–2% of patients the procedure causes weakness in non-dystonic muscles, in particular in the trapezius. Re-innervation can occur and may require further surgery. The European Federation of Neurological Societies proposed a guideline on the diagnosis and treatment of primary dystonia in 2006 [52], indicating recommendations and good practice points on selective peripheral denervation for CD as follows; ‘‘Selective peripheral denervation is a safe procedure with infrequent and minimal side effects that is indicated exclusively in cervical dystonia. This procedure requires a specialized expertise.’’ If a patient has CD symptoms with extracervical symptoms of dystonia, such as blephalospasm, oro-laryngeal dystonia, or limb dystonia, I do not consider peripheral denervation as the first choice treatment, because suppression of cervical
1893
1894
112
Peripheral procedures for cervical dystonia
symptoms may make the extracervical symptoms severer. For such segmental or generalized dystonia, pallidal deep brain stimulation should be considered first. Peripheral denervation of cervical muscles may be indicated for some patients with segmental/generalized dystonia with residual CD symptoms even after pallidal stimulation. Despite the very promising results of selective peripheral denervation, however, there is still a substantial group of patients who do not benefit from this procedure. According to Braun et al., positive response to prior botulinum toxin therapy seems to be a very good predictor of outcome after selective peripheral denervation [53]. Ford et al. [34] reported outcome of selective ramisectomy for botulinum toxin ‘‘resistant’’ torticollis and concluded that about one third of patients with torticollis resistant to injections of botulinum toxin may derive modest long term functional improvement from selective denervation, with a reduction in dystonia by about 30%, but remain unable to work. Fixed posture secondary to soft tissue fibrosis or skeletal changes may develop in CD. Arce [54] examined the range of movement of the neck under general anesthesia with maximum muscle relaxant. Only 18% of 45 patients had a normal range of movement and position of the head. When such fixed organic changes are severe, the effect of denervation surgery becomes modest.
Anesthesia Bertrand’s original procedure is generally performed in sitting position, which makes less bleeding from the cervical venous plexus. Denervation of SCM is easily performed with this position. The only drawback of this position is a risk of air embolism, and in many centers, anesthetists are not accustomed to the intraoperative management of problems related to sitting position, and they tend not to allow this position. When using sitting position, monitoring with precordial
Doppler and a multiorifice right atrial catheter has been recommended. Lobato et al. studied the incidence of air embolism during this operation in sitting position, and reported that Doppler detected one episode of air embolism that lasted <20 s and had no clinical sequelae in the prospective group of 69 patients. The incidence of complications associated with right atrial catheter insertion was 8% (carotid puncture, hematoma, inability to cannulate) but with no permanent sequelae. They concluded that air embolism is infrequent and self-limited in association with selective denervation for torticollis. While monitoring with precordial Doppler for patients undergoing denervation for torticollis is indicated, the use of a right atrial catheter is of limited value because of associated complications and increased operating room time and cost. Girard et al. [55] focused on air embolism retrospectively and patent foramen ovale in patients undergoing selective peripheral denervation in the sitting position. Among 342 patients, seven patients (2%) exhibited venous air embolism. The severity of venous air embolism was 2/5 for three patients, 3/5 for three patients, and 4/5 for one patient. Air could be aspirated from the central venous catheter for three patients. No deaths occurred. Among the 96 transesophageal echocardiographic examinations performed, five cases (5.2%) of patent foramen ovale were detected. For those patients, they performed surgery in the prone or park-bench position. No paradoxical air embolism was detected. Based on their study, they recommend that the detection of a patent foramen ovale prompt a change in position for this surgical procedure. Usually, endotracheal intubation is not a problem in patients with CD. Mac et al. [56] studied frequency of difficult airway and intubation in CD patients and concluded that patients with CD do not appear to have a higher frequency of difficult airway or difficult intubation.
Peripheral procedures for cervical dystonia
In my practice, I prefer prone position because OR staffs are much more accustomed to this setting. Before induction of anesthesia, I mark the incision for XIth nerve denervation. Central venous line is not necessary, and I have not encountered situation necessary for blood transfusion. Total intravenous anesthesia with propofol and fentanyl is generally used with short acting muscle relaxant only at induction. We fix the patient head with a Mayfield cramp. The upper body is raised about 30 degrees and the neck is slightly flexed. We do not shave the head but we clip the hair along the nuchal skin incision.
Surgical Procedures Denervation of the Posterior Cervical Muscles Skin Incision In the original technique developed by Bertrand, he stressed importance of hockey stick incision for denervation of the posterior rami of the cervical nerves, because a straight skin incision necessitates excessive lateral retraction, which
112
may cause neuroapraxia and make identification of the nerves difficult with electrical stimulation. Cohen-Gadol and coworkers also describe the use of a Bertrand’s hockey stick incision to avoid neuroapraxia. However, in my opinion, there is no scientific evidence that neurapraxia is more readily induced by a smaller straight incision. In my experience and in others [35,57], if we use an operating microscope with fine microsurgical technique, a straight suboccipital midline incision is sufficient to expose the posterior rami of the cervical spinal nerves with excellent surgical results [58]. Even with a short straight incision (> Figure 112-6), by tilting the microscope as in transsphenoidal pituitary surgery, the posterior branches of C1–C6 are easily visualized. Another drawback of hockey stick incision is that the peripheral C2 nerve (great occipital nerve) is always sacrificed. As I discuss subsequently, all the patients who underwent Bertrand’s original technique suffer C2 numbness and may experience denervation neuropathic pain. If we try to preserve C2 sensation, the skin incision should be straight. Based on my experience in almost 200 procedures, I believe a small straight incision is enough for this surgery in the era of microneurosurgery.
. Figure 112-6 Skin incision and direction of the microscope. Skin incision can be short, and by tilting the microscope, we can access to the C1–C6 nerves
1895
1896
112
Peripheral procedures for cervical dystonia
Anatomy According to the embryological and comparative anatomy study by Kato and Sato [59], the semispinalis capitus originates from two parts. Because of this, SSCap receives innervation form both the medial and lateral surfaces. SSCap separates the posterior branch into medial and lateral rami. SPL is innervated from the lateral branch of the lateral rami, while SSCer receives nerve supply from the medial branch of the medial rami (> Figure 112-7). The semispinalis capitus has double innervation from both the medial branch of the lateral rami and the medial branch of the lateral rami. This anatomical relation is complex and difficult, but I realized that this is very important when performing the Bertrand operation as shown in > Figure 112-1. In most recurrence cases, I had only dissected the plane between the semispinalis capitus and cervicalis muscles where only the medial rami of the posterior branch were exposed. For complete denervation of SPL, the lateral rami of the posterior branch should be cut. The lateral rami of the posterior branch run in the anterolateral aspects of SSCap,
and we have to cut the insertions of SSCap to the vertebral body. After cutting the insertions of SSCap, we can observe another muscle of which fibers run parallel to the rostro-caudal direction. This muscle is distinct form SSCap of which fibers are oblique to the vertebral column, and it is the longissimus muscle. The lateral rami are found on the dorsal aspect of the longissimus muscle, and they run perpendicular to the vertebral column. I believe understanding such anatomy is the key for successful operation. After realizing the detailed anatomy of the spinal posterior nerves, incidence of reoperation dramatically decreased. Bogduk and Anat reported excellent review on the posterior branches of the cervical nerves [60]. Layer by layer anatomy of the cervical muscles and posterior branches (> Figure 112-8) helps our understanding of the 3-D anatomy.
Bertrand’s Original Technique and Taira’s Modified Method One of the disadvantages of Bertrand’s original technique is aesthesia or dysesthesia in the C2
. Figure 112-7 Cross section of surgical route and posterior neck muscles and nerves. AB, anterior branch of the spinal nerve; PB, posterior branch of the spinal nerve; LR-PB, lateral rami of the posterior branch; MR-PB, medical rami of the posteriro branch; SScap, semispinalis capitus muscle; SScer, semispinalis cervicalis muscle
Peripheral procedures for cervical dystonia
region. This is because both the sensory and motor fibers run in a common nerve trunk at the proximal extradural part of the C2 nerve. Bleeding from the paravertebral venous plexuses around the C1 and C2 nerves is often troublesome especially in case of prone position. Care should also be taken not to injure the vertebral
112
artery adjacent to the C1 nerve (> Figure 112-9). To overcome such disadvantages, I modified of Bertrand’s operation in that the intradural ventral roots of C1 and C2 are sectioned through a limited C1 hemilaminectomy [42,43,58]. The denervation of posterior rami from the C3–C6 nerves and the accessory nerve is same as in the
. Figure 112-8 Layers of posterior neck muscles and relation with the posterio branches of the spinal nerves. SPL, splenius muscle; LR, lateral rami; MR, medial rami (modified from [60])
. Figure 112-9 Microsurgical anatomy of extradural C1 and C2 nerves. post. men. a, posterior meningeal artery; post. sp.a, posterior spinal artery; br, branch; SOM, superior oblique muscle; IOM, inferior oblique muscle; Rec, suboccipital rectus muscle
1897
1898
112
Peripheral procedures for cervical dystonia
original Bertrand’s procedure. > Figure 112-10 shows the comparison between Bertrand’s procedure and Taira’s modification. In Taira’s modification, after making a straight skin incision from the inion to the C4 level (> Figure 112-6), the midline nuchal ligament is cut and the inferior oblique muscle and the posterior arch of C1 are exposed first. The suboccipital rectus muscles are cut unilaterally at their insertions, but the inferior oblique muscle should be always preserved. C1 hemilaminectomy (> Figure 112-11) is extended as far lateral as to expose the lateral aspect of the dural sac. The dura is opened for 1.5 cm length. Drilling the occipital bone is not necessary. After dissection of the arachnoid membrane and the dentate ligament, C1 and C2 ventral rootlets are exposed and sectioned (> Figure 112-12). There are usually 4–6 anterior rootlets in C2, and 3–4 in C1. C1 posterior roots are often absent. For the intradural procedure from a small dural opening, we need maximum magnification of a well-balanced high quality operation microscope and very fine microsurgical instruments. The small vessels running with the rootlets should be preserved as much as possible. After dural closure, the space caudal to the inferior oblique muscle and between the
. Figure 112-10 Comparison between traditional Bertrand procedure and Taira’s modification
semispinalis capitus and the semispinalis cervicalis muscles is dissected to expose the posterior branches of the C3–C6 spinal nerves. The nerves we see in this plane are the medial rami of the posterior branches of the spinal nerves. They innervate to SScap and SScer muscles, and not to SPL. For denervation of SPL, which is very important, we have to reach to the lateral rami of the posterior branches. To expose these nerves to SPL, we have to cut the insertions of SScap muscle. After cutting the insertions, there appears a muscle of which fibers run parallel to the body axis. Because the muscle fibers of SScap run obliquely to the body axis, the difference is apparent. This is the longissimus colli muscle, and on the surface of this muscle, we can identify the nerve braches going to SPL. These nerves are confirmed with electrical stimulation, and coagulated and cut. Peripheral nerves usually run . Figure 112-11 C1 hemilaminectomy for C1,C2 anterior rhizotomy
Peripheral procedures for cervical dystonia
112
. Figure 112-12 Intradural anterior C1,C2 rhizotomy
with vessels. Large venous plexuses may have to be also coagulated. We confirmed complete denervation with strong monopolar electrical stimulation (5 Hz, 1 ms pulse width, 5–10 V). By tilting the operation microscope as in transsphenoidal pituitary surgery (> Figure 112-5), the procedure is performed through much smaller opening than in traditional Bertrand’s method. At closure of the wound, the space surrounded by the bilateral semispinalis capitus muscles and the inferior oblique muscle was closed in watertight fashion to prevent cerebrospinal fluid (CSF) leakage. We usually do not place a drainage tube. In my practice, Bertrand’s traditional procedure is indicated for bilateral denervation of the posterior cervical muscles in patients with retrocollis. This is because we can not make distinction between intradural anterior rootles serving
anterior and posterior branches of the spinal nerves. Sacrificing bilateral rootlet components of anterior branches result in dysphagia and dysarthria as was experienced in McKenzie-Dandy operation. In bilateral procedure, the most common complication is difficult of extending the neck which results in difficulty of swallowing. This is especially true when the patient is aged woman with thin neck muscles. In such cases, I preserve bilateral C6 nerves. Bilateral C6 denervation should be avoided as Bertrand reported [61]. In extradural denervation of C1 and C2, we usually cut the inferior oblique muscle. C2 peripheral nerve runs under this muscle, and once the peripheral part is identified, this is followed proximally to the point where the anterior and posterior branches are divided. The anterior branch should be preserved, while the posterior branch is coagulated and cut. For C1
1899
1900
112
Peripheral procedures for cervical dystonia
denervation, the rostral edge of the C1 posterior arch is followed laterally and, with electrical stimulation, the posterior branches embedded in the fatty tissue are identified. It is not wise to secure the proximal part of C1 nerve, because it is surrounded by venous plexuses. Recently, Arce [54] reported lateral musclesplitting approach. He stresses that the main advantage of this approach is easier access to the superior posterior cervical rami (C1–C5) and preservation of midline attachment of trapezius and cervical muscles. He mentions that this technique appears to be associated with significant decrease of postoperative pain and a faster recovery.
Denervation of the Sternocleidomastoid Muscle The sternocleidomastoid muscle is generally considered to have nerve supply from the accessory
nerve. However, the innervation pattern to SCM is not so simple and it also receives fibers from the C2 and C3 anterior branches as studied by Caliot et al. [62,63]. According to their study, there are three main types of variations of innervation pattern to SCM (> Figure 112-13). The classical ‘‘anastomotic’’ type of innervation (also referred to as Maubrac’s type of innervation) was the most frequent, although present in only half of all cases. Because there are fibers arising from the high cervical (C2, C3) nerves to SCM, abnormal contractions may still remain even after intradural section of the XIth nerve trunk. Therefore denervation of SCM should be carried out at the peripheral area where fibers of both XIth nerve and the cervical plexus enter the SCM. In the original Bertrand’s technique, the peripheral main trunk of XIth nerve is completely exposed by sectioning the SCM with a large skin incision along the direction of XIth nerve. This makes easier to identify all the motor nerves going
. Figure 112-13 Variation of innervation pattern to the sternocleidomastoid muscle
Peripheral procedures for cervical dystonia
to SCM. Because of cosmetic reason, we use a small (2–3 cm) skin incision along the posterior border of SCM. The center of the incision comes to a point where the posterior border of SCM crosses a line drawn between the lower border of the ear lobe and neck-shoulder angle. This point should be marked before introduction of anesthesia when the patient is still in supine position. After skin incision, the great auricular nerve coming up from the posterior border of SCM is identified. This point is called ‘‘auricular point’’ and is a nice landmark, because the main trunk of XIth nerve runs under this point. With a fine monopolar stimulation (1–5 V, 5 Hz, 0.2–0.5ms), the location of the XIth nerve trunk is roughly confirmed by strong contractions of the trapezius muscle. Dissection should be carried out until the XIth nerve trunk is identified. Once the nerve is found, we dissect the surrounding tissue distally and proximally to expose the nerve trunk completely. We may partially cut the SCM itself. An operative microscope is essential to find fine branches to SCM and to preserve the XIth main trunk. The great auricular nerve should be also preserved, otherwise the patient may suffer dysesthetic sensation in the earlobe. In most textbooks of anatomy, the XIth nerve trunk is described that it runs behind SCM, but in my experience, the nerve pierces a part of SCM in most cases. Dissection of the XIth nerve should be carried out proximally where the fat tissue behind SCM appears. There are usually three to four branches from the main trunk to SCM. These branches should be checked with electrical stimulation. When cutting the branches to SCM, care should be taken not to damage the main trunk with current spread of bipolar coagulation. We stimulate the tissue along the posterior border of SCM to confirm that there are no more nerve supplies to SCM. We usually do not denervate the trapezius muscle, because weakness of this muscle results in marked disabilities such difficulty of raising the arm and shoulder joint problems. However,
112
there are some patients in whom abnormal contraction of the trapezius muscle is a major symptom of CD. Krauss et al. reported a technique for performing partial sectioning and myectomy of the trapezius muscle. They used asleep-awakeasleep anesthesia that allowed intraoperative control of the sectioning procedure to avoid causing postoperative weakness of arm elevation above the horizontal plane. This technique can be used as an adjunct to other peripheral surgical procedures in patients with marked laterocollis and dystonic elevation and ante-version of the shoulder.
Denervation of the Levator Scapulae Muscle Understanding the surgical anatomic relationships of the motor nerves to the levator scapulae (LS) muscle (> Figure 112-14) is imperative for the surgical denervation of this muscle in patients with laterocollis [64]. LS muscle originates on the anterior aspect of the scapula and inserts into the upper cervical vertebrae and the mastoid process. It travels beneath the upper fibers of the trapezius. This muscle acts to elevate the scapula and tilt the head towards the same side. Typical clinical feature of this relatively rare CD symptom is lateral tilting of the neck and elevation of the shoulder (> Figure 112-15). Rotation of the head is minimal, and so the head tilts laterally only in the coronal plane. Sometimes the ear touches to the shoulder, and because of tonic nature of the symptom, there is no space between the neck and shoulder. Palpation of a firm and tight LS muscle bundle in the posterior cervical triangle is the most reliable diagnostic physical examination. According to the anatomical study by Frank et al. [65], an average of approximately two nerves from the cervical plexus (range 1–4 nerves) emerge from beneath the posterior border of the sternocleidomastoid muscle in a cephalad to caudad progression to enter the posterior triangle of the neck on their way to innervating the levator
1901
1902
112
Peripheral procedures for cervical dystonia
scapulae. These cervical plexus contributions exhibited a fairly regular relationship to the emergence of XIth nerve and the auricular point along the posterior border of the sternocleidomastoid muscle. After emerging from the posterior border of the sternocleidomastoid to enter the posterior triangle of the neck, cervical plexus contributions to the levator scapulae travel for a variable distance posteriorly and inferiorly, sometimes branching or coming together. Ultimately these nerves cross the anterior border of the levator . Figure 112-14 Gross anatomy of the levator scapulae muscle
scapulae as 1–3 nerves in a regular superior to inferior progression. The dorsal scapular nerve from the brachial plexus exhibited highly variable anatomic relations in the inferior aspect of the posterior triangle, and penetrates or gives branches to the levator scapulae in about 30% of the examined cases. The levator scapulae muscle receives predictable motor supply from the cervical plexus. Contribution of nerve supply is reported to be 5.5% from C2, 100% from C3 and C4, and 19% from C5 anterior branches [59]. The nerve supply to LSM is not from the dorsal part of the muscle, but from the ventrolateral surface (> Figure 112-16), and because of this anatomy, the denervation procedure is not difficult in the anterior approach. Patients are operated on under endotracheal anesthesia and in the supine position with the head rotated to the opposite side of the surgical field. Muscle relaxant is not used except at the induction of anesthesia to monitor intraoperative muscle contraction. I use an operative microscope from the skin incision. The skin incision is about 8 cm long along the posterior border of SCM (> Figure 112-17). The great auricular nerve, the branches of the external jugular vein, and the accessory nerve are carefully identified and preserved. The great auricular nerve is followed along the posterior surface of SCM to the surface layer to LSM. At the anterior
. Figure 112-15 Typical example of laterocollis due to levator scapulae muscle contraction. Before (left) and after (right) surgical treatment
Peripheral procedures for cervical dystonia
112
. Figure 112-16 Microsurgical anatomy of innervation to the levator scapulae muscle
. Figure 112-17 Skin incision for levator scapulae denervation
border of LSM, high cervical nerve plexuses were identified. With electrical stimulation (5 Hz or 50 Hz, 0.2–1 ms pulse width, 0.2–1 V), the nerve branches to LSM are identified and cut (> Figure 112-18). The distal part of the nerve was pulled out from the muscle to prevent reinnervation. The C5 nerve is identified with electrically induced contraction of the deltoid muscle. If there are branches from the C5 nerve to LSM, these are also denervated. The phrenic nerve is identified with electrically induced diaphragm contraction and carefully preserved.
This nerve runs on the anterior surface of the anterior scalene muscle. In some patients with laterocollis, scalene muscles are also involved. Because of the innervation pattern of scalene muscles, complete denervation may be difficult, and we may add myotomy of the scalene muscles. Arce [54] described briefly on posterolateral approach to the levator scapulae, but he noted that complete denervation of the levator scapulae is difficult to achieve with this approach and additional myotomy is recommended.
1903
1904
112
Peripheral procedures for cervical dystonia
. Figure 112-18 Operative view of levator scapulae denervation in the posterior cervical triangle
Postoperative Rehabilitation and other Related Problems After selective peripheral denervation for CD, patients should be well-informed of the importance of physiotherapy [32,54]. It is very important that patients do posture exercises to regain a sense of midline, and to improve the range of movement. However, there is no standard of physiotherapy or rehabilitation programs after peripheral denervation of CD [66]. Patients with CD have higher incidence of symptomatic degenerative cervical spondylosis that may require surgical treatment [67–70]. Loher et al. [71] studied characteristics of features of degenerative spinal disease in patients with dystonia and analyzed operative strategies. Degenerative spinal disorders in patients with dystonia and choreoathetosis occur much earlier than in the physiological aging process. Dystonic movement disorders more often affect the higher cervical levels (C2–C5). Degenerative changes of the cervical spine are more likely to occur on the side where the chin is rotated or tilted. Posterior approaches are often used for decompression, but additional anterior fusion also becomes necessary. Anterior approaches with or without instrumented fusion yielded more favorable
results, but drawbacks are pseudarthrosis and adjacent-level disease. Although psychogenic origin of CD is denied, mental problems seem to be common in CD patients in daily clinical practice [72]. A recent study showed that depression and anxiety disorder occur in patients with CD with a 3.7-fold increase compared to a matched control group, which suggests that psychiatric comorbidity is not just secondary to chronic disease and disfigurement and may have its own pathogenesis related to dystonia [73]. A study comparing 201 CD patients with 135 control subjects with spine pain showed elevated rates of depression among the CD patients [74]. Jahanshahi and Marsden [75,76] reported controlled studies on depression and personality in CD patients, and concluded that the prevalence of psychiatric disorder, any of the personality dimensions evaluated, and their self-reports of events prior to onset of the illness did not differ in CD and control groups. However, they found that self-referent negative cognitions such as self-blame, self-accusation, self-punitive thoughts, and negative body-image emerged as the prominent component of depression in torticollis. Kashiwase and Kato [77] reported that personality of CD patients has tree types; type I: overadaptive type, type II: maladaptive type:,
Peripheral procedures for cervical dystonia
and type III: compatible type. Type I is a typical psychosomatic with high frustration tolerance. Type II is personality disorder with low frustration tolerance. In type III, frustration tolerance varies depending on social circumstances. In type I, the prognosis of ST is generally unfavorable, since it is associated with recurrence and prolongation of the symptoms. In type II, the prognosis of ST is generally favorable. However, type II patients experience social difficulties. One characteristic of type III is that the onset of symptoms is usually at older age. We have to consider such mental aspects of CD patients and appropriate mental support is often necessary.
Clinical Results Although a denervation procedure for torticollis is not a curative treatment but merely a symptomatic suppression of abnormal contractions of the neck muscles, we often experience excellent postoperative ‘‘cure’’ of the condition as described by Cohen-Gadol, et al. [38]. The overall improvement of the symptoms is generally 80–90% in many reports. Although good surgical outcome with low morbidity has been reported, little information is available on factors determining the outcome. Ford et al. [34] studied the effects of peripheral denervation in patients with torticollis resistant to injections of botulinum toxin. They found that one third of the patients may derive modest long term functional improvement from selective denervation, with a reduction in dystonia by about 30%, but remain unable to work. Munchau et al. [41] performed prospective study of selective peripheral denervation for botulinum-toxin resistant patients, and concluded that that selective peripheral denervation is an effective treatment for patients with secondary, but probably not for those with primary, botulinum toxin treatment failure. They also noted that reinnervation is not infrequent and can compromise outcome.
112
Postoperative morbidity is low, but there is a risk of dysphagia. Examples of clinical effects of my own series are shown in the attached DVD (See Video).
Other Peripheral Surgical Procedures Microvascular Decompression Because of clinical similarities between CD and hemifacial spasm, some neurosurgeons have attempted microvascular decompression of the spinal XIth nerve. Shima et al. [78,79] reported seven cases of spasmodic torticollis treated by microsurgical decompression of the XIth nerve. In these patients, there was an intermittent horizontal torticollis characterized by aggravation of the symptoms when in a resting posture, presenting with a striking contrast to the torticollis of extrapyramidal origin that was alleviated while in the resting posture and aggravated by postural stress. They observed a tight neurovascular contact at the C1 level, occurring between the principal XIth nerve and the vertebral or posterior inferior cerebellar artery. They decompressed the nerve in two by transposing the compressing artery and in five by sectioning at C1 or C2 the branching root of the XIth nerve that had caused the tight cross contact by locking the nerve trunk to the dura mater. The symptoms improved after an interval of 1–4 weeks. After an average follow-up of 3 years, full or satisfactory relief was obtained in five and some improvement in two patients. Jho and Jannetta [80] treated 20 patients with spasmodic torticollis by microvascular decompression of the spinal accessory nerves, the upper cervical nerve roots and the brainstem. Ten had right horizontal; nine, left horizontal; and one, retrocollis. Twenty-two operations were performed, suboccipital craniectomy and C1 laminectomy in 18 and retromastoid craniectomy in four
1905
1906
112
Peripheral procedures for cervical dystonia
operations. The most common compressing blood vessels were the vertebral artery and/or the posterior inferior cerebellar artery. No nerve section was performed. During the 5–10 year follow-up, 13 (65%) were cured, four (20%) improved with minimal spasm, one (5%) improved with moderate spasm, and two (10%) improved minimally or unchanged. In most cases the cure or improvement was noticed gradually over 6 months to 2 years following the operation. MVD for ST is a nondestructive benign procedure with high probability of cure or significant improvement. Despite such excellent results of vascular decompression surgery on spasmodic torticollis, there have been no further reports on this surgery since then. Vascular compression theory does not match clinical findings that muscles innervated from other than the accessory nerve are hyperactive. It is hard to consider that unilateral decompression affects both SCM and contralateral posterior neck muscles. In my experience of clinical examination in over 200 patients, I have never seen patients in whom only SCM played a role in clinical presentation of CD. It is still mysterious whether torticollis treated by Shima and Jho is a different clinical entity from ordinary cervical dystonia.
Spinal Cord Stimulation Many patients with CD know that mild finger touch to a certain part of the face or the neck by themselves relieves the symptoms. The symptoms become less marked by leaning against a wall or lying down. Such a phenomenon is called ‘‘sensory trick,’’ and this indicates that sensory inputs play an importance role in pathophysiology of CD. Gildenberg [81] considered that adding artificial sensory input with electrical stimulation to the cervical cord might alleviate the symptom of CD. After transcutaneous trial stimulation, he implanted a dorsal
column stimulator at the C1–C2 level. Among 22 patients, three had sufficient relief with transcutaneous stimulation only. An additional 6 patients underwent surgical implantation of dorsal column stimulators. It was empirically determined that a frequency of 800–1,100 Hz gave the best relief from torticollis. One patient had an excellent result; three have had good results; one had a fair result, and one had a poor result [82]. Currently available implantable stimulators have no ability of outputting such high frequency pulses. Dieckmann and Veras [83] reported the effect of cervical cord stimulation for 18 patients suffering from spasmodic torticollis. Permanent 1,100-Hz stimulation of the C2–C4 level resulted in a marked improvement in 50% of the patients, in a satisfactory result in 27.8% and in an unsatisfactory result in 22.2%. Muscular tension and related pain were reduced within 8 days to 4 months postoperatively. A measurable improvement of head posture rendered objective by a mechanoelectrical measuring device could only be observed after a continuous stimulation over 8–12 months. Although these reports indicated that spinal cord stimulation might be a good treatment of choice for CD, Fahn [84] warned the use of spinal cord stimulation for dystonias and there have been no further detailed studies since 1990s.
Conclusion Selective peripheral denervation is generally indicated in patients with CD who do not achieve adequate response with medical treatment or repeated injections of botulinum toxin. This is a safe procedure with infrequent and minimal side effects. Because this procedure requires a specialized expertise with understanding of detailed anatomy of cervical peripheral nerves, it is still underutilized all over the world.
Peripheral procedures for cervical dystonia
References 1. Duane DD. Spasmodic torticollis: clinical and biologic features and their implications for focal dystonia. Adv Neurol 1988;50:473-92. 2. Dandy W. An operation for the treatment of spasmodic torticollis. Arch Surg 1930;20:1021-32. 3. Das K BD, Rovit RL, Murali RAJ, Couldwell WT. Irving S. Cooper (1922–1985): a pioneer in functional neurosurgery. J Neurosurg 1998;89:865-73. 4. Cooper IS. Victim is always the same. New York: Harper & Row; 1973. 5. Nutt JG MM, Aronson A, Kurland LT, Melton LJ III. Epidemiology of focal and generalized dystonia in Rochester, Minnesota. Mov Disord 1988;3:188-94. 6. Claypool DW, Duane DD, Ilstrup DM, Melton LJ III. Epidemiology and outcome of cervical dystonia (spasmodic torticollis) in Rochester, Minnesota. Mov Disord 1995;10(5):608-14. 7. Chan J, Brin MF, Fahn S. Idiopathic cervical dystonia: clinical characteristics. Mov Disord 1991;6(2):119-26. 8. Tasker RR. The treatment of spasmodic torticollis by peripheral denervation: the McKenzie operation. In: Morley TP, editors. Current controversies in neurosurgery. Philadelphia: WB Saunders; 1976. p. 448-54. 9. Keen WW. New operation for spasmodic wry neck: namely, division or exsection of the nerves supplying the posterior rotator muscles of the head. Ann Surg 1891;13:44. 10. McKenzie KG. Intrameningeal division of the spinal accessory and roots of the upper cervical nerves for the treatment of spasmodic torticollis. Surg Gynecol Obstet 1924;39:5-10. 11. Poppen JL, Martinez-Niochet A. Spasmodic torticollis. Surg Clin North 1951;31:883-90. 12. Sorenson BF, Hamby WB. Spasmodic torticollis: results in 71 treated patients. Neurology 1966;16:867-78. 13. Sorensen BF, Hamby WB. Spasmodic torticollis. Results in 71 surgically treated patients. JAMA 1965;194:706-8. 14. Hamby WB, Schiffer S. Spasmodic torticollis: results after cervical rhizotomy in 80 cases. Clin Neurosurg 1970;17:29-37. 15. Hamby WB, Schiffer S. Spasmodic torticollis: results after rhizotomy in 50 cases. J Neurosurg 1969;31:323-6. 16. Friedman AH, Nashold BS Jr, Sharp R, Caputi F, Arruda J. Treatment of spasmodic torticollis with intradural selective rhizotomies. J Neurosurg 1993;78:46-53. 17. Speelman JD, van Manen J, Jacz K, van Beusekom GT. The Foerster-Dandy operation for the treatment of spasmodic for the treatment of spasmodic torticollis. Acta Neurochir Suppl (Wien) 1987;39:85-7. 18. Hernesniemi J, Keranen T. Long-term outcome after surgery for spasmodic torticollis. Acta Neurochir 1990;103 (3–4):128-30.
112
19. Cooper IS. Effect of thalamic lesions upon torticollis. N Engl J Med 1964;7;270:567-72. 20. Hassler R, Dieckmann G. Stereotaxic treatment of torticollis according to animal experiment experiences about direction determined movement. Nervenarzt 1970;41(10):437-87. 21. Hassler R, Dieckmann G. Stereotactic treatment of different kinds of spasmodic torticollis. Confin Neurol 1970;32(2):135-43. 22. Eltahawy HA, Saint-Cyr J, Poon YY, Moro E, Lang AE, Lozano AM. Pallidal deep brain stimulation in cervical dystonia: clinical outcome in four cases. Can J Neurol Sci 2004;31(3):328-32. 23. Botzel K, Steude U. First experiences in deep brain stimulation for cervical dystonia. Nervenarzt 2006; 77(8):940-5. 24. Kiss ZH, Doig-Beyaert K, Eliasziw M, Tsui J, Haffenden A, Suchowersky O. The Canadian multicentre study of deep brain stimulation for cervical dystonia. Brain 2007;130(Pt 11):2879-86. 25. Kiss ZH DK, Eliasziw M, Ranawaya R, Suchowersky O. The Canadian multicenter trial of pallidal deep brain stimulation for cervical dystonia: preliminary results in three patients. Neurosurg Focus 2004;15;17(1):E5. 26. Goto S, Mita S, Ushio Y. Bilateral pallidal stimulation for cervical dystonia. An optimal paradigm from our experiences. Stereotact Funct Neurosurg 2002;79(3–4):221-7. 27. Hung SW, Hamani C, Lozano AM, Poon YY, Piboolnurak P, Miyasaki JM, Lang AE, Dostrovsky JO, Hutchison WD, Moro E. Long-term outcome of bilateral pallidal deep brain stimulation for primary cervical dystonia. Neurology 2007;68(6):457-9. 28. Islekel S, Zileli M, Zileli B. Unilateral pallidal stimulation in cervical dystonia. Stereotact Funct Neurosurg 1999;72 (2–4):248-52. 29. Bertrand CM. Surgery of involuntary movements, particularly stereotactic surgery: reminiscences. Neurosurgery 2004;55(3):698-703. 30. Bertrand C, Molina-Negro P, Martinez SN. Combined stereotactic and peripheral surgical approach for spasmodic torticollis. Appl Neurophysiol 1978;41(1–4):122-33. 31. Bertrand CM M-NP. Selective peripheral denervation in 111 cases of spasmodic torticollis: rationale and results. Adv Neurol 1988;50:637-43. 32. Bertrand CM. Selective peripheral denervation for spasmodic torticollis: surgical technique, results, and observations in 260 cases. Surg Neurol 1993;40(2):96-103. 33. Tasker RR. Overview of the surgical treatment of spasmodic torticollis. In: Tasker RR, Gildenberg PL, editors. Textbook of stereotactic and functional neurosurgery. New York: McGraw-Hill; 1998. p. 1053-8. 34. Ford B, Louis E, Greene P, Fahn S. Outcome of selective ramisectomy for botulinum toxin resistant torticollis. J Neurol Neurosurg Psychiatry 1998;65:472-8.
1907
1908
112
Peripheral procedures for cervical dystonia
35. Braun V RH. Selective peripheral denervation for spasmodic torticollis: 13-year experience with 155 patients. J Neurosurg 2002;97 Suppl 2:207-12. 36. Jang KS, HKPH, Joo W II, Ji C, Kyung Jin Lee KJ, Choi CR. Selective peripheral denervation for the treatment of spasmodic torticollis. J Korean Neurosurg Soc 2005;37:250-3. 37. Chen X, Ma A, Liang J, Ji S, Pei S. Selective denervation and resection of cervical muscles in the treatment of spasmodic torticollis: long-term follow-up results in 207 cases. Stereotact Funct Neurosurg 2000;75(2–3):96-102. 38. Cohen-Gadol AA, Ahlskog JE, Matsumoto JY, Swenson MA, McClelland RL, Davis DH. Selective peripheral denervation for the treatment of intractable spasmodic torticollis: experience with 168 patients at the Mayo Clinic. J Neurosurg 2003;98(6):1247-54. 39. Davis HD AJ, Litchy WJ, Root LM. Selective peripheral denervation for torticollis: preliminary results. Mayo Clin Proc 1991;66:365-71. 40. Meyer CH. Outcome of selective peripheral denervation for cervical dystonia. Stereotact Funct Neurosurg 2001;77(1–4):44-7. 41. Munchau A, Palmer JD, Dressler D, O’Sullivan JD, Tsang KL, Jahanshahi M, Quinn NP, Lees AJ, Bhatia KP. Prospective study of selective peripheral denervation for botulinum-toxin resistant patients with cervical dystonia. Brain 2001;124(Pt 4):769-83. 42. Taira T, Hori T. Peripheral neurotomy for torticollis: a new approach. Stereotact Funct Neurosurg 2001; 77(1–4):40-3. 43. Taira T, Hori T. A novel denervation procedure for idiopathic cervical dystonia. Stereotact Funct Neurosurg 2003;80(1–4):92-5. 44. Podivinsky F. Torticollis. In: Vinken PJ, Bruyn GW, editors. Handbook of clinical neurology. Amsterdam: North-Holland Publishing; 1968. p. 567-603. 45. Consky ES, Lange AE. Clinical assessment of patients with cervical dystonia. In: Jankovic J, Hallett M, editors. Therapy with botulinum toxin. New York: Marcel Dekker; 1994. p. 211-37. 46. Tsui JKC, Eisen A, Mak E, Carruthers J, Scott AB, Calne DB. A pilot study on the use of botulinum toxin in spasmodic torticollis. Can J Neurol Sci 1985;12:314-16. 47. Burke RE, Fahn S, Marsden D, Bressman SB, Moskowitz C, Friedman J. Validity and reliability of a rating scale for the primary torsion dystonias. Neurology 1985;35:73-7. 48. Odergren T, Tollback A, Borg J. Efficacy of botulinum toxin for cervical dystonia. A comparison of methods for evaluation. Scand J Rehabil Med 1994;26(4):191-5. 49. NIICE. The National Institute for Clinical Excellence. Selective peripheral denervation of cervical dystonia. In: August www.niceorguk; 2004. 50. Krauss JK, Loher TJ, Pohle T, Weber S, Taub E, Barlocher CB, Burgunder JM. Pallidal deep brain
51. 52.
53.
54.
55.
56.
57.
58.
59.
60. 61.
62.
63.
64.
65.
stimulation in patients with cervical dystonia and severe cervical dyskinesias with cervical myelopathy. J Neurol Neurosurg Psychiatry 2002;72(2):249-56. Krauss JK. Deep brain stimulation for treatment of cervical dystonia. Acta Neurochir Suppl 2007;97(Pt 2):201-5. Albanese A, Barnes MP, Bhatia KP, Fernandez-Alvarez E, Filippini G, Gasser T, Krauss JK, Newton A, Rektor I, Savoiardo M, et al. A systematic review on the diagnosis and treatment of primary (idiopathic) dystonia and dystonia plus syndromes: report of an EFNS/MDS-ES Task Force. Eur J Neurol 2006;13(5):433-44. Braun V, Richter HP, Schroder JM. Selective peripheral denervation for spasmodic torticollis: is the outcome predictable? J Neurol 1995;242(8):504-7. Arce CA. Selective peripheral denervation in cervical dystonia. In: Stacy MA, editor. Handbook of dystonia. New York: Informa healthcare; 2007. p. 381–92. Girard F, Ruel M, McKenty S, Boudreault D, Chouinard P, Todorov A Molina-Negro P, Bouvier G. Incidences of venous air embolism and patent foramen ovale among patients undergoing selective peripheral denervation in the sitting position. Neurosurgery 2003;53(2):316-9. Mac TB, Girard F, McKenty S, Chouinard P, Boudreault D, Ruel M, Bouvier G. A difficult airway is not more prevalent in patients suffering from spasmodic torticollis: a case series. Canad J Anaesth 2004;51(3):250-3. Braun V, Richter H. Selective peripheral denervation for the treatment of spasmodic torticollis. Neurosurgery 1994;35:58-63. Taira T, Kobayashi T, Takahashi K, Hori T. A new denervation procedure for idiopathic cervical dystonia. J Neurosurg 2002;97 Suppl 2:201-6. Kato K, Sato T. Morphological analysis of the levator scapulae, rhomboideus and serratus anterior. J Anat 1978;53:339-56. Bogduk N, Anat D. The clinical anatomy of the cervical dorsal rami. Spine 1982;7:319-30. Bertrand CM. Operative management of spasmodic torticollis and adult-onset dystonia with emphasis on selective denervation. In: Schmidek HH, Sweet WH, editors. Operative neurosurgical techniques. 2nd ed. Orlando: Grune & Stratton; 1988. p. 1261-9. Caliot P, Cabanic P, Bousquet V, Midy D. A contribution to the study of the innervation of the sternocleidomastoid muscle. Anat Clin 1984;16(1):21-8. Caliot P, Bosquet V, Midy D, Cabanie P. A contribution to the study of the accessory nerve: surgical implications. Surg Radiol Anat 1989;11(1):11-15. Taira T, Kobayashi T, Hori T. Selective peripheral denervation of the levator scapulae muscle for laterocollic cervical dystonia. J Clin Neurosci 2003;10(4):449-52. Frank DK, Wenk E, Stern JC, Gottlieb RD, Moscatello AL. A cadaveric study of the motor nerves to the levator scapulae muscle. Otolaryngol Head Neck Surg 1997;117:671-80.
Peripheral procedures for cervical dystonia
66. Smania N, Corato E, Tinazzi M, Montagnana B, Fiaschi A, Aglioti SM. The effect of two different rehabilitation treatments in cervical dystonia: preliminary results in four patients. Funct Neurol 2003;18(4):219-25. 67. Waterston JA, Swash M, Watkins ES. Idiopathic dystonia and cervical spondylotic myelopathy. J Neurol Neurosurg Psychiatry 1989;52(12):1424-6. 68. Hanakita J, Suwa H, Nagayasu S, Nishi S, Ohta F, Sakaida H. Surgical treatment of cervical spondylotic radiculomyelopathy with abnormal involuntary neck movements. Report of three cases. Neurol Med Chir 1989;29(12):1132-6. 69. Polk JL, Maragos VA, Nicholas JJ. Cervical spondylotic myeloradiculopathy in dystonia. Arch Phys Med Rehab 1992;73(4):389-92. 70. Wong AS, Massicotte EM, Fehlings MG. Surgical treatment of cervical myeloradiculopathy associated with movement disorders: indications, technique, and clinical outcome. J Spinal Disord Tech 2005;18 S107-S114. 71. Loher TJ, Barlocher CB, Krauss JK. Dystonic movement disorders and spinal degenerative disease. Stereotact Funct Neurosurg 2006;84(1):1-11. 72. Rondot P, Marchand MP, Dellatolas G. Spasmodic torticollis – review of 220 patients. Can J Neurol Sci 1991;18(2):143-51. 73. Gundel H, Wolf A, Xidara V. High psychiatric comorbidity in spasmodic torticollis: a controlled study. J Nerv Ment Dis 2003;191:465-73. 74. Duane DD, Vermilion KJ. Cognition and affect in patients with cervical dystonia with and without tremor.
75. 76. 77.
78.
79.
80.
81.
82. 83.
84.
112
In: Fahn S, Hallet M, Delong M, editors. Dystonia 4: advances of neurology. Philadelphia: Lippincott Williams & Wilkins; 2004. p. 179-89. Jahanshahi M, Marsden CD. Depression in torticollis: a controlled study. Psychol Med 1988;18(4):925-33. Jahanshahi M, Marsden CD. Personality in torticollis: a controlled study. Psychol Med 1988;18(2):375-87. Kashiwase H, Kato M. The classification of idiopathic spasmodic torticollis: three types based on social adaptation and frustration tolerance. Psychiatry Clin Neurosci 1997;51(6):363-8. Shima F, Fukni M, Matsubara T, Kitamura K. Spasmodic torticollis caused by vascular compression of the spinal accessory nerve. Surg Neurol 1986;26(5):431-4. Shima F, Fukni M, Kitamura K, Kuromatsu C, Okamura T. Diagnosis and surgical treatment of spasmodic torticollis of 11th nerve origin. Neurosurgery 1988;22:358-63. Jho HD, Jannetta P. Microvascular decompression for spasmodic torticollis. Acta Neurochir (Wien) 1995;134 (1–2):21-6. Gildenberg PL. Treatment of spasmodic torticollis by dorsal column stimulation. Appl Neurophysiol 1978;41:113-21. Gildenberg PL. Comprehensive treatment of spasmodic torticollis. Appl Neurophysiol 1981;44:233-43. Dieckmann G, Veras G. Bipolar spinal cord stimulation for spasmodic torticollis. Appl Neurophysiol 1985;48 (1–6):339-46. Fahn S. Lack of benefit from cervical cord stimulation for dystonia. N Engl J Med 1985;313(19):1229.
1909
98 PPN Stimulation for Parkinson’s Disease S. Stone . C. Hamani . A. M. Lozano
Introduction
PPN Anatomy
Parkinson’s disease (PD) is the second most common neurodegenerative disorder next to Alzheimer’s disease, affecting up to 1% of individuals aged 65–69 years and 3% of those over 80 years of age [1]. Among the cardinal features of parkinsonism (resting tremor, bradykinesia, rigidity, and postural instability), postural and gait disfunction leading to falls represents the largest single contributor to the number of emergency room visits and overall cost to the healthcare system relating to PD [2–4]. In addition, the fear of falling is associated with its recurrence, and frequently leads to a loss of independence and depression [5]. Postural and gait disfunction have proven particularly resistant to current dopamine and surgical therapies, which suggests a greater involvement of non-dopaminergic pathways and other brain loci distinct from the pallidal and subthalamic nuclei in their pathophysiology [6–14]. The pedunculopontine nucleus (PPN) is a brainstem locomotive center that also processes sensory and behavioral information. Its connections with basal ganglia structures and the spinal cord suggest that the PPN could play a role in the mechanisms that underlie axial motor symptoms in PD. First, anatomical, pharmacological, and physiological aspects of the PPN in regard to locomotion will be reviewed. Second, its potential pathophysiological role in PD, and prospects as a target for neuromodulatory therapy in PD using deep brain stimulation (DBS) will be discussed.
The rostral extent of the PPN begins immediately posterior and caudal to the substantia nigra pars compacta (SNc). It then travels caudally between the fibers of the superior cerebellar peduncle and medial lemniscus, inferiorly ending adjacent to the locus coeruleus > Figure 98‐1. The PPN can be regionally separated into two subdivisions based on cell density, and these in turn harbor different neurotransmitter profiles [15–19]. The pars compacta of the PPN (PPNc) is situated caudally and dorsolaterally, while neurons of the pars dissipata of the PPN (PPNd) are sparsely distributed with a more rostral tendency [20]. The PPNc predominantly consists of cholinergic neuron clusters, with a lesser proportion of non-cholinergic neurons. The PPNd also includes cholinergic and noncholinergic neurons, but exhibits considerably greater regional variation and a more, even overall representation of these two cell groups than the PPNc [21,22]. Non-cholinergic PPNc and PPNd neurons are primarily glutamatergic, but lesser numbers are noradrenergic, dopaminergic, and GABAergic [17]. Neither the average number of total neurons nor the exact breakdown of non-cholinergic neuron types in the human PPN are well established [17]. However, the approximate number of cholinergic neurons in the adult human PPN is 17,000 per side, with 5% variation roughly [21]. One third of these are located in the PPNc, and two thirds in the PPNd [21]. In humans, the
#
Springer-Verlag Berlin/Heidelberg 2009
1650
98
PPN stimulation for parkinson’s disease
. Figure 98‐1 Axial sections through the brainstem showing cellular architecture (left) and tracts (right) at (a) intercollicular level (rostral extent of PPN) and (b) inferior collicular level (caudal extent of PPN). The PPNd is labelled 8 in A and the PPNc is labelled 9 in B. (a) Intercollicular level (rostral extent of PPN). 1 = nucleus intercollicularis; 2 = griseum central mensencephali; 3 = nucleus paralemniscalis; 4 = nucleus centralis colliculi inferioris; 5 = nucleus mesencephalicus nervi trigemini; 6 = nucleus nervi trochlearis; 7 = nucleus cuneiformis; 8 = nucleus tegmentalis pedunculopontinus, pars dissipata (PPNd); 9 = substantia nigra, pars compacta; 10 = nucleus interpeduncularis; 11 = nucleus pontis; 12 = commissural colliculi inferioris; 13 = brachium colliculi inferioris; 14 = fasciculus longitudinalis dorsalis; 15 = tractus mesencephaliicus nervi trigemini; 16 = fasciculus anterolateralis; 17 = tractus tectospinalis; 18 = tractus trigeminothalamicus dorsalis; 19 = nervus trochlearis; 20 = fasciculus longitudinalis medialis; 21= tractus tegmentalis centalis; 22 = lemniscus medialis; 23 = pedunculus cerebellaris superior; 24 = decussatio pedunculorum cerebellarium superiorum; 25 = pedunculus mamillaris; 26 = tractus parietotemporopontinus; 27 = tractus pyramidalis; 28 = tractus frontopontinus; 29 = fibrae pontocerebellares. (b) Inferior collicular level (caudal extent of PPN). 1 = nucleus intercollicularis; 2 = colliculus inferior, nucleus centralis; 3 = colliculus inferior, zona lateralis; 4 = griseum centrale mesencephali; 5 = locus coeruleus; 6 = nucleus mesencephalicus nervi trigemini; 7 = nucleus cuneiformis; 8 = corpus parabigeminum; 9 = nucleus tegmentalis peduculopontinus, pars compacta (PPNc); 10 = nucleus centralis superior; 11 = substantia nigra, pars compacta; 12 = nucleus interpeduncularis; 13 = nuclei pontis; 14 = commissural colliculi inferioris; 15 = fasciculus longitudinalis dorsalis; 16 = nervus trochlearis; 17 = tractus mesencephalicus nervi trigemini; 18 = lemiscus lateralis; 19 = tractus tectopontinus; 20 = fasciculus anterolateralis; 21 = fasciculus longitudinalis medialis; 22 = tractus tedmentalis centralis; 23 = lemniscus medialis; 24 = pedunculus cerebellaris superior; 25 = decussatio pedunculorum cerebellarium superiorum; 26 = fibrae corticotegmentales; 27 = pedunculus mamillaris; 28 = fibrae pontocerebellares; 29 = tractus parietotemporopontinus; 30 = tractus pyramidalis; 31 = tractus frontopontinus. From Nieuwenhuhs et al. (1998), with permission
PPN stimulation for parkinson’s disease
number of cholinergic neurons in the PPN region does not consistently decrease with age [21,23].
PPN Connections Human studies examining the PPN’s connectivity throughout the brain are relatively scarce. Muthusamy et al. recently performed diffusion tensor magnetic resonance imaging (MRI) on eight human volunteers, and demonstrated prominent tracts between the PPN and cerebellum via the superior cerebellar peduncle and between the PPN and spinal cord [24]. Rostral subcortical regions with PPN connections included the thalamus, pallidum, and STN. Cortical regions connecting to the PPN, via the internal capsule, included the primary motor cortex, premotor areas, and frontal lobe. It should be noted that connections were not always observed in all subjects, which likely reflect a combination of some degree of inter-individual variability in these pathways and the current limits of sensitivity, resolution, and target specificity inherent to diffusion tensor imaging. In order to describe PPN connectivity in more detail and with directional specificity, the following discussion will include data derived from non-human primates, and other mammals. > Figure 98‐2 provides a summary model for our current understanding of interconnections between the PPN, basal ganglia, and other relevant structures.
PPN Afferents The best characterized PPN afferents stem from the globus pallidus internus (GPi) and the substantia nigra reticulata (SNr) (> Figure 98‐2) [25,26]. The majority of GPi neurons send axon collaterals to both the ventrolateral thalamus and the PPN, with the latter possibly being the
98
more dominant target given the relatively larger diameter of the PPN axonal branches [27]. Pallidal projections to the PPN, along with the midbrain tegmentum, descend dorsomedially along the pallidotegmental tract without appreciable branching. Once at the PPN, these pallidotegmental projections break into collaterals that innervate both the PPNc and PPNd [26,27]. Human and non-human primate studies indicate that the GPi projections are GABAergic and preferentially terminate on the noncholinergic cells of the PPNd [28,29]. Conversely, SNr inputs are not known to collateralize and have shown similar preferential termination only on non-cholinergic PPNd neurons in rodents. SNr projections are also GABAergic [28,29]. Whether or not GPi and SNr inputs segregate to some extent into separate regions within the PPN, or converge onto common neurons, remains to be determined. Several cortical areas send afferents to the PPN, all of which are likely glutamatergic (> Figure 98‐2). The dorsal PPN receives afferents from the primary motor cortex in non-human primates. These corticotegmental projections provide somatotopically arranged orofacial, forelimb, and hindlimb representations in a medial to lateral orientation [30]. Similarly mediolaterally-arranged inputs to the middle portion of the PPN stem from the supplementary and pre-supplementary motor areas, as well as the dorsal and ventral premotor cortices. Cortical afferents also arise from the frontal eye fields, terminating throughout the PPN [30]. Other PPN afferents include inputs from the subthalamic nucleus (STN), brainstem, and cervical and lumbar regions of the spinal cord (> Figure 98‐2) [15,19,31,32]. However, these connections have not been histologically confirmed in primates or humans, and, with the exception of glutamatergic STN terminals, none of their neurotransmitter systems have been identified [17].
1651
1652
98
PPN stimulation for parkinson’s disease
. Figure 98‐2 Pedunculopontine nucleus (PPN, indicated in yellow) connectivity within the basal ganglia circuitry (see text for details). Structures sharing connections with the PPN are colored blue, and those relevant to basal ganglia circuits but lacking direct PPN connections are indicated in faded blue. Predominantly stimulatory connections are depicted by solid green lines with arrow heads, and inhibitory ones with dashed red lines with round heads. Solid black PPN connections are presumed stimulatory but are relatively poorly characterized, possibly exerting more heterogeneous effects. BS brain stem; CBC cerebellar cortex; D1 and D2 dopaminergic receptor types 1 and 2; DCN deep cerebellar nuclei; GPe external globus pallidus; GPi/SNr internal globus pallidus/substantia nigra pars reticulata; ML/IL midline and intralaminar nuclei of the thalamus; PN pontine nuclei; PPNd/PPNc pedunculopontine nucleus pars dissipata/pars compacta; SC spinal cord; STN subthalamic nucleus; SNc substantia nigra pars compacta; VA/VL ventral anterior/ventrolateral nuclei of the thalamus; Vim ventral intermediate nucleus of the thalamus
PPN Efferents PPN efferents can be grouped into ascending and descending components, both of which consist of cholinergic and non-cholinergic neurons [16,17]. Although this grouping is useful for circuitry modeling, it should be noted that some PPN neurons collateralize extensively and project in both directions and/or contralaterally. Ascending cholinergic PPN neurons travel via the dorsal and ventral tegmental bundles, projecting mainly, though not exclusively, to thalamic associative non-specific midline, and intralaminar nuclei (> Figure 98‐2) [17,25,33]. Other ascending ventral tegmental projections densely innervate basal ganglia structures (> Figure 98‐2). These basal ganglia afferents ultimately ascend along the lenticular fasciculus and ansa lenticularis,
major output projection pathways of the GPi [25]. The majority of fibers terminating in the basal ganglia target the STN, which receives bilateral PPN projections in non-human primates, and the substantia nigra pars compacta (SNc) [16,25). Lesser numbers also terminate in the SNr, globus pallidus, and striatum [15,16,19,25]. The neurotransmitters involved in PPN-STN and PPN-pallidal synapses are unknown; however, both cholinergic and glutamatergic PPN neurons are known to project to dopaminergic SNc, and to a lesser extent to SNr, neurons [17,34]. Descending PPN projections are thought to collateralize extensively and innervate deep cerebellar, midbrain, pontine, medullary, and uniand bilateral spinal cord nuclei (> Figure 98‐2) [15,18,19,35]. Specific brainstem targets include the medullary reticular formation, locus
PPN stimulation for parkinson’s disease
coeruleus, and raphe nuclei. Targets that are relatively close to the PPN, such as the medullary reticular formation, mainly receive cholinergic terminals. Conversely, spinal cord projections are primarily glutamatergic [17].
PPN Neuronal Pharmacology In vitro electrophysiology and in vivo behavioral studies in animals suggest numerous neurotransmitters directly influence PPN neurons [17]. GABA, acetylcholine (ACh), serotonin, noradrenaline, and opiates decrease PPN activity, whereas glutamate and histamine increase activity (reviewed in depth by [17]). > Table 98‐1 summarizes findings for the three major neurotransmitter systems in the PPN, namely glutamate, acetylcholine, and GABA. It should be noted that no studies have shown simultaneous neuronal and behavioral effects of these agents on PPN function, and electrophysiological data beyond cholinergic PPN neurons is lacking. In addition, only M2, 3, and 4 muscarinic cholinergic receptors have been localized on PPN neurons, while NMDA and glycine receptors have been found in the general area of the PPN [36–38]. Accordingly, our current understanding of PPN neuronal pharmacology remains a proposed model.
PPN Electrophysiology Three types of neurons have been identified in the PPN using intra- and extracellular recordings . Table 98‐1 Summary of the major PPN neurotransmitter effects on PPN neuronal activity and behaviour in animal studies (adapted from [17]) Neurotransmitter system
Cholinergic PPN neuronal activity
Locomotion
Glutamate Acetylcholine GABA
↑ ↓ ↓
↑ ↓ ↓
98
of the rat, cat, and non-human primate [17,28,31,39–41]. The first is a bursting type, with narrow extracellular spikes, that tends to fire at around 20 Hz. Depolarization or hyperpolarization of these neurons can elicit this phasic pattern [39,41]. These bursting neurons are most likely glutamatergic and are mainly located in the PPNd [41]. They are felt to represent the target neurons for GABAergic GPi and SNr fibers both anatomically and functionally, with the latter evident from experiments showing PPN inhibition in response to electrical stimulation of the SNr [28,40]. They also provide the main PPN output to the spinal cord, of importance to the initiation of programmed movements [17]. The second type of PPN neuron responds to current injection with single action potentials, large afterhyperpolarizations, no bursts, and produces relatively broad spikes at less than 12 Hz. These properties are ideal for producing slow tonic repetitive firing patterns [39,41]. The nonbursting neurons include a large cholinergic population and likely correspond to PPNc neurons, which relay feedback sensory information from the spinal cord and provide significant inputs to the thalamus and SNc. These units appear important for the maintenance of gait [17]. A third less-characterized type of PPN neuron combines the characteristics of both bursting and nonbursting types [17]. In addition to the aforementioned intrinsic electrophysiological features observed in the PPN, recent studies indicate that the PPN can respond to changes in the activity of input nuclei (> Figure 98‐2). In particular, the response of PPN neurons to STN stimulation has been investigated in anesthetized rodents. STN stimulation with single pulses induces excitatory responses in 20% of recorded PPN neurons [42]. In contrast, 1–5 second trains of stimulation at 130 Hz lead to a response in approximately 40% of recorded PPN neurons [42]. Of responding neurons, 85% are inhibited and 15% are excited by STN stimulation [42]. This stimulation parameterdependent pattern of responsiveness is maintained
1653
1654
98
PPN stimulation for parkinson’s disease
after nigral 6-OHDA injections, commonly used to generate rodent models of PD [42]. In animals with entopeduncular (EP, the rodent homologue of GPi) lesions, 75% of PPN neurons become responsive to STN stimulation. In this scenario, the proportions of neurons responding with excitation (85%) or inhibition (15%) are reversed, and those distributions remain constant in animals additionally treated with nigral saline or 6-OHDA injections. These results suggest that the inhibition of PPN neurons after trains of STN stimulation is likely mediated by EP neurons [42]. This agrees with our current understanding of PPN connectivity which predicts that STN-driven GABAergicoutput from GPi and SNr inhibits the PPN (> Figure 98‐2). Recent studies have also demonstrated the PPN’s ability to influence activity in the STN, SNr, and SNc (> Figure 98‐2) [43,44]. Following PPN lesions in anesthetized rodents, STN firing rates increase from 10 to 15 Hz [43]. In addition, the proportion of irregularly firing and bursting STN units increases from 16 to 31% and 4 to 18% respectively [43]. Similar changes occur in the SNr following PPN lesions, namely an increase in the firing rate (from 20 to 29 Hz) and the percentage of units firing irregularly (from 18 to 32%) [43]. A decrease in the number of spontaneously firing dopaminergic SNc cells were also seen, which may represent the key pathway whereby PPN neurons modulate STN and SNr activity (> Figure 98‐2) [43]. Similar experiments performed in SNc-lesioned animals find that PPN lesioning reduces STN and SNr activity back to baseline, suggesting that stimulatory PPN inputs to the STN and SNr may become dominant in the absence of the SNc-mediated PPN output pathway [44]. Although these preliminary findings require replication in nonhuman primates and further support from human data, they underscore the potentially significant influence of the PPN on basal ganglia structures and propose a mechanism through
which the modulation of PPN activity might exert clinical effects on motor function.
Role of PPN in Locomotion and Tone In order to consider a role for the PPN in locomotion and tone, it is necessary to appreciate a theoretical framework underlying neurological aspects of posture and gait. Normal gait requires the proper functioning and interaction of three primary neurological processes: locomotion (including its initiation and maintenance), balance, and adaptation to the environment [45]. These processes are felt to rely upon the corresponding neuronal systems that include spinal cord central pattern generators (CPGs, capable of independently producing rhythmic locomotion), peripheral proprioceptive afferents (providing adaptive feedback), brainstem locomotion control centers (providing top-down control of both CPGs and reflexes that mediate peripheral inputs to the spinal cord), and corticospinal and corticobulbar voluntary control pathways [46]. Pathology in one or more of these neurological components of gait and posture control can result in a clinical gait disorder [45]. > Table 98‐2 summarizes some key features of different neurological gait syndromes, the differentiation of which is vital to ensure appropriate diagnostic and therapeutic actions are taken. As summarized by Snijders et al. (45), it should be noted that a number of medications can have side-effects that manifest as neurological gait, posture, and balance impairments, and thus a thorough global assessment is always required for these patients [45]. > Table 98‐3 expands upon hypokinetic-rigid, or ‘‘Parkinsonian’’ gait, balance, and postural disorders in order to emphasize that several underlying etiologies can produce similarly appearing clinical manifestations. Using this framework, the following
PPN stimulation for parkinson’s disease
98
. Table 98‐2 Major Neurologic Gait Syndromes (adapted from [45]) Gait Syndrome Antalgic gait Paretic/ hypotonic gait Spastic gait
Vestibular gait
Main features of gait Short stance phase on affected limb, limping High steppage, dropping foot, waddling
Circumduction, intermittent abduction of ipsilateral arm with each step, foot dragging: ‘‘scuffing toe’’, Scissoring Deviation to one side
Useful clinical tests
Trendelenburg’s sign
Aggravated by eye closure, Unterberger positive
Sensory ataxic gait Cerebellar ataxic gait
Staggering, wide based
Aggravated by eye closure
Staggering, wide based
Not aggravated by eye closure
Dyskinetic gait Hypokineticrigid gait
Extra movements that affect gait
Can be task-specific (e.g., dystonic gait) Improves with external cues, aggravation by secondary task
Cautious gait
‘‘Walking on ice’’ (slow, wide base, short steps), striking improvement with external support Severe balance impairment (no rescue reactions with pull test; ‘‘falls like a log’’), inadequate synergies, inappropriate foot placement, crossing of legs, leaning wrong direction when turning or standing, variable performance (influenced by environment and emotion), hesitation and freezing (ignition failure) Incongruous with known gait disorders, bizarre/non-physiologic, variable/inconsistent, abrupt onset, extreme slowness, exaggerated effort, sudden buckling, unusual/uneconomic posture
Higher level gait disorder
Psychogenic gait disturbance
Shuffling (slow speed, short stride, rigidity, reduced step height), Hesitation and freezing
Abnormal interaction with environment (e.g., trouble adapting with walking aids, no benefit from cues), sometimes better able to perform cycling leg movements while recumbent (gait apraxia)
Exaggerated effort,
Associated signs and symptoms Pain, limited range of movements Lower motor neuron features (e.g., weakness, atrophy, low to absent tendon reflexes) Pyramidal syndrome, anterior-medial side of the shoe sole worn out Vestibular features (e.g., nystagmus, abnormal tilting test) Disturbed proprioception Cerebellar ataxia (e.g., dysarthria, hypermetria, nystagmus) Features of dystonia, chorea, myoclonus or tics Hypokinetic-rigid features (e.g., bradykinesia, resting tremor) Postural instability (mild to moderate), excessive fear of falling Frontal release signs, executive dysfunction, depression, frequent falls
Incongruous affect, secondary gain, rare falls or injuries, history of psychiatric disease
1655
1656
98
PPN stimulation for parkinson’s disease
. Table 98‐3 Hypokinetic-rigid or ‘‘Parkinsonian-like’’ gait disorders (adapted from [45]) Condition
Main anatomical substrate(s) (Disease Process)
Parkinson’s disease
Substantia nigra (neurodegenerative)
Multiple system atrophy, parkinsonian type Progressive supranuclear palsy
Basal ganglia, cerebellum, pyramidal tracts, autonomic nervous system (neurodegenerative) Diffuse brainstem pathology (neurodegenerative)
Corticobasal degeneration
Basal ganglia, cortex (neurodegenerative)
Dementia with Lewy bodies Subcortical arteriosclerotic encephalopathy Vascular parkinsonism
Basal ganglia, cortex (neurodegenerative) Subcortical white matter (vascular)
Strategic vascular lesion
Putamen, globus pallidus, thalamus, dorsal mesencephalon (vascular) Frontostriatal/periventricular (ventricular widening)
Normal pressure hydrocephalus Drug-induced parkinsonism
Diffuse white matter, basal ganglia (vascular)
Basal ganglia
Characteristic features
Associated features
Narrow-based gait, asymmetric, stooped posture, early freezing and falls rare Early phase like PD gait, later phase more wide-based, Pisa syndrome, antecollis, falls due to syncope
Good response to levodopa, resting tremor Cerebellar ataxia, autonomic features, pyramidal signs
Wide-based gait, freezing common, erect posture with retrocollis, early spontaneous backward falls, motor recklessness, frequent and severe injuries Asymmetrical (e.g., unilateral leg apraxia, dystonia, or myoclonus), later onset of wide-based gait with freezing and shuffling Like PD gait but more symmetric
Vertical gaze palsy, pseudobulbar palsy, frontal dementia, applause sign
Magnetic gait (small steps, widebased, start hesitation), variable step timing and amplitude Lower body parkinsonism, wide based > stooped, relatively preserved arm swing Lower body parkinsonism, freezing/ severe gait akinesia, severe disequilibrium, drifting to one side Wide-based gait, freezing, gait apraxia, truncal imbalance, preserved arm swing Mild gait impairment, rarely freezing, preserved postural reflexes, Pisa syndrome
evidence will build the case for the PPN’s role as an important brainstem control center for the initiation and continuance of gait as well as the maintenance of postural stability.
Evidence from PPN Stimulation The PPN, along with the cuneiform nucleus, forms part of the functionally defined mesencephalic locomotor region (MLR). This area of the brainstem can be chemically excited, using
Apraxia, alien limb, cortical sensory loss
Dementia, fluctuations, visual hallucinations Urinary incontinence, cognitive decline, stepwise progression Urinary incontinence, cognitive decline, stepwise progression
Urinary incontinence, cognitive decline Upper limb tremor, symmetrical presentation
microinjections of glutamate agonists, or electrically stimulated to produce controlled locomotion on a treadmill in decerebrate animals [47,48]. Certain MLR stimulation parameters are necessary to induce specific effects on locomotion and tone. For example, several seconds of continuous midfrequency (20–60 Hz) stimulation induces locomotor activity, whereas high-frequency (>100 Hz) stimulation suppresses muscle tone. Although the PPNc has been considered the optimal specific site for inducing locomotion [47], some evidence suggests that this phenomenon may not
PPN stimulation for parkinson’s disease
be consistently found in, and unique to, the PPN [49]. Because of the PPN’s complexity and many adjacent nuclei and fibers, different experimental protocols using electrical stimulation or drug infusion could produce differing effects through a variety of mechanisms. Indeed, the exact role and interplay of MLR nuclei in the control of postural muscle tone and locomotion remains controversial and further studies are necessary to clarify these issues. Mesopontine centers, including the PPN, are thought to regulate tone and locomotion through the modulation of lower brainstem reticulospinal neurons [50]. These neurons, in turn, activate neuronal networks in the spinal cord that include CPGs and thus ultimately exert control over the use of motor patterns governing actions such as locomotion. Certain regions of the reticular formation contain populations of reticulospinal neurons which appear to exert the downstream effects of, and respond to, PPN activity. For instance, stimulation of the pontine inhibitory area (PIA) of decerebrate animals, induces muscle hypotonia by inhibiting centers that facilitate muscle tone and locomotion, such as the locus coeruleus and the MLR [51]. In rodents, the PIA includes the middle portions of the caudal and oral pontine reticular nuclei, and parts of the gigantocellular and dorsal gigantocellular reticular nuclei [52]. The PPN’s ability to influence the PIA is suggested by long lasting responses in the caudal pontine reticular nucleus in response to mid-frequency PPN stimulation in rat brain slices and decerebrated cats [53,54]. It has been further suggested that PPN connections to the pontine reticular formation may play an important role in the hypotonia that occurs during REM sleep [15,17]. Other mechanisms through which the PPN might influence tone and gait may involve its projections to the medioventral medulla and spinal cord [15]. Approximately 50% of medioventral medullary neurons receive short-latency inputs following MLR stimulation. Although the exact role of this communication is unknown, initial suggestions
98
point to its participation in ‘‘triggering’’ locomotive patterns [17]. Despite logically representing important routes for PPN control over spinal CPGs, the role of direct PPN projections to the spinal cord remains to be elucidated.
Evidence from PPN Inhibition PPN function has also been inferred from lesion, high-frequency stimulation, and chemical inhibition studies in non-human primates. Radiofrequency and excitotoxic lesions (with kainic acid injections) induce a contralateral flexed posture and hypokinesia [55,56]. While animals with unilateral lesions tend to recover near normal mobility after approximately 1 week, those with bilateral lesions suffer a significant long-lasting decrease of motion followed by partial recovery after approximately 1 week [57]. Unilateral low frequency stimulation (below 30 Hz) induces a 5 Hz tremor and tendency to turn towards the ipsilateral side, whereas high frequency stimulation induces akinesia [58]. GABA and Ach agonist delivery to the PPN generally has a negative effect on locomotion [59–62]; however, a recent study using non-human primates found that topical muscimol (an Ach agonist) applied to the PPN improves scores on the primate parkinsonism motor rating scale [62]. As is the case for PPN stimulation, it is not always clear which parts of the PPN or adjacent structures are affected in these drug delivery studies and thus discrepant results may arise from different patterns of local network disruption.
Theorized Sensory Role In addition to its role in descending locomotive circuits, the PPN responds to somatosensory stimuli and possibly receives spinal cord input from lamina I (posteromarginal nucleus in the spinal cord, predominantly consisting of second order spinothalamic neurons) [18,63,64]. Given its
1657
1658
98
PPN stimulation for parkinson’s disease
cholinergic projections to the thalamus, as well as efferent connections with deep cerebellar nuclei, the PPN may relay sensory information to thalamic and cerebellar areas that may then adapt motor patterns to external events [17]. This suggests a mechanism whereby the PPN could take part in the sensory modulation of posture and gait.
Role of PPN in Gait Disfunction Related to PD Gait and postural disfunction in PD is characterized by bradykinesia, rigidity, freezing or difficulty with gait initiation, and a loss of balance with ensuing falls [11,65]. Despite the former two aspects proving relatively responsive to levodopa and STN/Gpi-based medical and surgical therapies, the latter two remain particularly treatmentresistant and incapacitating manifestations of the disease. Given that the PPN appears to play a role in the initiation and maintenance of gait and postural tone, and its pharmacological and electrophysiological properties rely heavily upon cholinergic and other non-dopaminergic neurotransmitters, pathology of this structure could theoretically play a significant role in the refractory gait and postural manifestations of PD [17]. Direct evidence linking PPN disfunction to PD posture and gait disturbances is lacking; however, several lines of rodent, non-human primate, and human evidence suggest an involvement of the PPN in the pathology of PD. The PPN in anesthetized 6-OHDA nigral-lesioned rats is characterized by a high percentage of cells firing irregularly and in bursts, as well as an increased firing rate, as compared to controls (18–20 Hz vs. 10–11 Hz) [66]. Breit and colleagues recently demonstrated that some of the electrophysiological changes in the basal ganglia of these PD-model animals could be reversed by PPN lesions. These included settling down the firing rate of the STN (from 17 to 11 Hz) and SNr (from 26 to 17 Hz), without a remarkable
change in the firing pattern within these structures [43,44]. Matsumura et al. (2001) observed a smaller degree of nigral degeneration and less parkinsonian symptoms in an MPTP monkey model of PD when animals were previously treated with PPN lesioning [67]. This may be related to a reduction in excitatory outflow from the PPN to the SNc (> Figure 98‐2). Furthermore, Nandi et al. have shown that PPN microinjections of the GABAergic antagonist bicuculline, but not saline or muscimol, improve parkinsonian symptoms in MPTP primates [62]. Concomitant changes in PPN metabolic and electrophysiological activity have been documented in parkinsonian rodents and non-human primates [68,69]. An initial theory is that this reflects a compensatory mechanism attempting to increase dopamine release via PPN-SNc projections [70,71]. Alternatively, it has also been suggested that nuclei sending afferents to the PPN, particularly the STN, GPi, and SNr, could be driving or inhibiting its activity. Pathological studies in humans have reported that approximately 50% of the large cholinergic neurons of the lateral part of the PPNc degenerate in PD [72–74]. While direct connections between all of these individual lines of evidence are absent, they collectively suggest that PD-related degeneration of non-dopaminergic PPN neurons could lead to electrophysiological and metabolic disturbances that culminate in disregulation of PPN-mediated gait and postural functions. This has led to the hypothesis that medical or surgical interventions aimed at modulating PPN activity in PD patients might facilitate the normalization of gait initiation, maintenance, and postural stability.
PPN as a Surgical Target in PD DBS represents a relatively safe and wellestablished means of focally, titratably, and
PPN stimulation for parkinson’s disease
reversibly modulating brain targets to return pathologically functioning circuits to a more normal state, and this technique is being employed as the initial strategy to influence PPN activity in PD patients (> Figure 98‐3).
Patient Selection A particular challenge in applying this experimental therapy concerns the appropriate selection of patients. Although definitive criteria must await further study, certain predictions can be made based on our background understanding of both PPN function and the neurological framework for gait and postural control. Current thinking suggests that PD patients with refractory and functionally significant impairment in gait initiation (manifested as debilitating freezing) and postural stability (associated with an unacceptable fall risk) warrant consideration on an investigational basis. Those with significant impairments in other aspects of gait and postural
. Figure 98‐3 Post-operative axial T1-weighted MRI of a PD patient demonstrating a DBS electrode in the region of the left PPN (arrow)
98
control, such as those associated with the nonPD conditions listed in > Tables 98‐2–98-3, would theoretically have a significantly lower chance for benefit since a rationale for PPN disfunction underlying their gait disturbance is lacking. For instance, impairments in cognition (including dementia), and primary affective disorders, are associated with walking instability and falls that appear to relate to a disfunction in cortical control of gait and posture [75–78]. Similarly, patients with comorbid conditions affecting gait, such as proprioceptive deficits, are theoretically less ideal candidates for PPN DBS. In general, PPN modulation would not be expected to address gait impairments stemming from pathologies other than basal ganglia disorders.
Outcomes of PPN DBS in PD Patients Initial reports of PPN DBS in human PD patients documented technical aspects and electrophysiological findings of the procedure, as well as short-term indicators of clinical outcome [79,80]. A more recent study provided greater insight into the clinical outcome of six patients who underwent the simultaneous insertion of bilateral PPN and STN DBS [81] systems. Bipolar stimulation was delivered using the following parameters: 25 Hz frequency, 60 s pulse width, and 1.5–2 V. At 6 months follow-up, Unified Parkinson’s Disease Rating Scale (UPDRS) motor scores in the ‘‘off ’’ medication/’’on’’ stimulation condition were improved by 33% with PPN DBS. As expected, this did not reach the benefits obtained with STN-only DBS (54%), or simultaneous STN and PPN DBS (56%). Of greater interest, improvements in axial symptoms, such as rising from a chair, posture, gait, and postural stability, were similar with PPNonly, STN-only, and simultaneous PPN and STN DBS (60–70%) [81].
1659
1660
98
PPN stimulation for parkinson’s disease
In the ‘‘on’’ medication condition, Stefani and colleagues found a greater improvement in UPDRS motor scores when STN and PPN stimulation were combined compared to either target stimulated alone [81]. More importantly, STN stimulation alone did not improve axial symptoms when the patients were taking their medication. If only the axial symptoms were taken into account in the ‘‘on’’ medication condition, both PPN-only DBS and simultaneous PPN, and STN DBS led to a 50–60% improvement versus the ‘‘off ’’ stimulation condition, suggesting PPN stimulation provided a unique benefit. Importantly, no major complications were described in either series. The single minor side effect reported with high-frequency PPN stimulation was paresthesias, likely resulting from the target’s proximity to the medial lemniscus [81]. The exciting potential for PPN DBS can be readily visualized in this sample movie that depicts a PD patient with significant gait and postural symptoms who underwent bilateral implantation of PPN DBS electrodes. While examples such as this and the small amount of available human data is encouraging, PPN DBS for the treatment of PD-related gait and postural dysfunction remains investigational and requires further evaluation before considering it a mainstream treatment option.
Conclusions The PPN has extensive anatomical, electrophysiological, and neurochemical connections with numerous brain regions involved in motor control. Studies involving the modulation of PPN neuronal activity (either directly, or through manipulations of upstream or downstream targets) can produce electrophysiological changes in motor nuclei and affect motor function. Recent pilot studies using PPN DBS to treat PD patients suggest that the procedure is safe and may be capable of improving axial symptoms. Future efficacy studies with larger
series of patients, longer follow-up, and ultimately double-blinding are now necessary in order to further characterize the safety and benefits of PPN DBS in the treatment of levodopa and surgically-resistant postural instability and gait disfunction in PD.
References 1. Tanner CM, Goldman SM. Epidemiology of Parkinson’s disease. Neurol Clin. 1996;14:317-335. 2. Bloem BR, Boers I, Cramer M, Westendorp RG, Gerschlager W. Falls in the elderly. I. Identification of risk factors. Wien Klin Wochenschr. 2001a;113:352-362. 3. Bloem BR, Grimbergen YA, Cramer M, Willemsen M, Zwinderman AH. Prospective assessment of falls in Parkinson’s disease. J Neurol. 2001b;248:950-958. 4. Schrag A, Ben-Shlomo Y, Quinn N. How common are complications of Parkinson’s disease? J Neurol. 2002; 249:419-423. 5. Adkin AL, Frank JS, Jog MS. Fear of falling and postural control in Parkinson’s disease. Mov Disord. 2003; 18:496-502. 6. Bonnet AM, Pichon J, Vidailhet M, Gouider-Khouja N, Robain G, Perrigot M, et al. Urinary disturbances in striatonigral degeneration and Parkinson’s disease: clinical and urodynamic aspects. Mov Disord. 1997;12:509-513. 7. Horak FB, Carlson-Kihtz P, Stephens M, Gross A, Hogarth P, Burchiel K. Effects of deep brain stimulation and levodopa on a variety of postural tasks in Parkinson’s patients. Mov Disord. 2002;17:s194. 8. Kemoun G, Defebvre L. Gait disorders in Parkinson disease. Clinical description, analysis of posture, initiation of stabilized gait. Presse Med. 2001a;30:452-459. 9. Kemoun G, Defebvre L. Gait disorders in Parkinson disease. Gait freezing and falls: therapeutic management. Presse Med. 2001b;30:460-468. 10. Klawans HL. Individual manifestations of Parkinson’s disease after ten or more years of levodopa. Mov Disord. 1986;1:187-192. 11. Morris ME, Iansek R, Matyas TA, Summers JJ. Stride length regulation in Parkinson’s disease. Normalization strategies and underlying mechanisms. Brain. 1996;119 (Pt 2)551-568. 12. Rascol O, Payoux P, Ory F, Ferreira JJ, Brefel-Courbon C, Montastruc JL. Limitations of current Parkinson’s disease therapy. Ann Neurol. 2003;53 Suppl 3:S3-S12; discussion S12-S5. 13. Robertson LT, Horak FB, Anderson VC, Burchiel KJ, Hammerstad JP. Assessments of axial motor control during deep brain stimulation in parkinsonian patients. Neurosurgery. 2001;48:544-551; discussion 551-552.
PPN stimulation for parkinson’s disease
14. Rocchi L, Chiari L, Horak FB. Effects of deep brain stimulation and levodopa on postural sway in Parkinson’s disease. J Neurol Neurosurg Psychiatr. 2002;73:267-274. 15. Inglis WL, Winn P. The pedunculopontine tegmental nucleus:where the striatum meets the reticular formation. Prog Neurobiol. 1995;47:1-29. 16. Lavoie B, Parent A. Pedunculopontine nucleus in the squirrel monkey: distribution of cholinergic and monoaminergic neurons in the mesopontine tegmentum with evidence for the presence of glutamate in cholinergic neurons. J Comp Neurol. 1994a;344:190-209. 17. Pahapill PA, Lozano AM. The pedunculopontine nucleus and Parkinson’s disease. Brain. 2000;123 (Pt 9):17671783. 18. Reese NB, Garcia-Rill E, Skinner RD. The pedunculopontine nucleus–auditory input, arousal and pathophysiology. Prog Neurobiol. 1995;47:105-133. 19. Winn P, Brown VJ, Inglis WL. On the relationships between the striatum and the pedunculopontine tegmental nucleus. Crit Rev Neurobiol. 1997;11:241-261. 20. Kus L, Borys E, Ping Chu Y, Ferguson SM, Blakely RD, Emborg ME, et al. Distribution of high affinity choline transporter immunoreactivity in the primate central nervous system. J Comp Neurol. 2003;463:341-357. 21. Manaye KF, Zweig R, Wu D, Hersh LB, De Lacalle S, Saper CB, et al. Quantification of cholinergic and select non-cholinergic mesopontine neuronal populations in the human brain. Neuroscience. 1999;89:759-770. 22. Mesulam MM, Geula C, Bothwell MA, Hersh LB. Human reticular formation: cholinergic neurons of the pedunculopontine and laterodorsal tegmental nuclei and some cytochemical comparisons to forebrain cholinergic neurons. J Comp Neurol. 1989;22;283(4):611-633. 23. Ransmayr G, Faucheux B, Nowakowski C, Kubis N, Federspiel S, Kaufmann W, et al. Age-related changes of neuronal counts in the human pedunculopontine nucleus. Neurosci Lett. 2000;288:195-198. 24. Muthusamy KA, Aravamuthan BR, Kringelbach ML, Jenkinson N, Voets NL, Johansen-Berg H, Stein JF, Aziz TZ. Connectivity of the human pedunculopontine nucleus region and diffusion tensor imaging in surgical targeting. J Neurosurg. 2007;107(4):814-820. 25. Lavoie B, Parent A. Pedunculopontine nucleus in the squirrel monkey: projections to the basal ganglia as revealed by anterograde tract-tracing methods. J Comp Neurol. 1994b;344:210-231. 26. Shink E, Sidibe M, Smith Y. Efferent connections of the internal globus pallidus in the squirrel monkey: II. Topography and synaptic organization of pallidal efferents to the pedunculopontine nucleus. J Comp Neurol. 1997; 382:348-363. 27. Parent M, Levesque M, Parent A. Two types of projection neurons in the internal pallidum of primates: single-axon tracing and three-dimensional reconstruction. J Comp Neurol. 2001;439:162-175.
98
28. Granata AR, Kitai ST. Inhibitory substantia nigra inputs to the pedunculopontine neurons. Exp Brain Res. 1991;86:459-466. 29. Noda T, Oka H. Distribution and morphology of tegmental neurons receiving nigral inhibitory inputs in the cat: an intracellular HRP study. J Comp Neurol. 1986; 244:254-266. 30. Matsumura M, Nambu A, Yamaji Y, Watanabe K, Imai H, Inase M, et al. Organization of somatic motor inputs from the frontal lobe to the pedunculopontine tegmental nucleus in the macaque monkey. Neuroscience. 2000;98:97-110. 31. Hammond C, Rouzaire-Dubois B, Feger J, Jackson A, Crossman AR. Anatomical and electrophysiological studies on the reciprocal projections between the subthalamic nucleus and nucleus tegmenti pedunculopontinus in the rat. Neuroscience. 1983;9:41-52. 32. Jackson A, Crossman AR. Nucleus tegmenti pedunculopontinus: Efferent connections with special reference to the basal ganglia, studied in the rat by anterograde and retrograde transport of horseradish peroxidase. Neuroscience. 1983;10:725-765. 33. Krout KE, Belzer RE, Loewy AD. Brainstem projections to midline and intralaminar thalamic nuclei of the rat. J Comp Neurol. 2002;448:53-101. 34. Charara A, Smith Y, Parent A. Glutamatergic inputs from the pedunculopontine nucleus to midbrain dopaminergic neurons in primates: Phaseolus vulgaris-leucoagglutinin anterograde labeling combined with postembedding glutamate and GABA immunohistochemistry. J Comp Neurol. 1996;364:254-266. 35. Rye DB, Saper CB, Lee HJ, Wainer BH. Medullary and spinal cord efferents of the pedunculopontine tegmental nucleus and adjacent mesopontine tegmentum in the rat. J Comp Neurol. 1988;269:315-341. 36. Probst A, Cortes R, Palacios JM. The distribution of glycine receptors in the human brain. A light microscopic autoradiographic study using [3H]strychnine Neuroscience. 1986;17:11-35. 37. Rye DB, Thomas J, Levey A. Distribution of molecular muscarinic (m1-m4) receptor subtypes and choline acetyltransferase in the pontine reticular formation of man and non-human primates. J Sleep Res. 1995;24:59. 38. Stone TW, Burton NR. NMDA receptors and ligands in the vertebrate CNS. [Review]. Prog Neurobiol. 1988;30:333-368. 39. Kang Y, Kitai ST. Electrophysiological properties of pedunculopontine neurons and their postsynaptic responses following stimulation of substantia nigra reticulata. Brain Res. 1990;535:79-95. 40. Noda T, Oka H. Nigral inputs to the pedunculopontine region: intracellular analysis. Brain Res. 1984; 322: 332-336. 41. Takakusaki K, Shiroyama T, Yamamoto T, Kitai ST. Cholinergic and noncholinergic tegmental pedunculopontine
1661
1662
98 42.
43.
44.
45.
46. 47. 48.
49.
50.
51.
52.
53.
54.
PPN stimulation for parkinson’s disease
projection neurons in rats revealed by intracellular labeling. J Comp Neurol. 1996;371:345-361. Florio T, Scarnati E, Confalone G, Minchella D, Galati S, Stanzione P, et al. High-frequency stimulation of the subthalamic nucleus modulates the activity of pedunculopontine neurons through direct activation of excitatory fibres as well as through indirect activation of inhibitory pallidal fibres in the rat. Eur J Neurosci. 2007;25:1174-1186. Breit S, Lessmann L, Benazzouz A, Schulz JB. Unilateral lesion of the pedunculopontine nucleus induces hyperactivity in the subthalamic nucleus and substantia nigra in the rat. Eur J Neurosci. 2005;22:2283-2294. Breit S, Lessmann L, Unterbrink D, Popa RC, Gasser T, Schulz JB. Lesion of the pedunculopontine nucleus reverses hyperactivity of the subthalamic nucleus and substantia nigra pars reticulata in a 6-hydroxydopamine rat model. Eur J Neurosci. 2006;24:2275-2282. Snijders AH, van de Warrenburg BP, Giladi N, Bloem BR. Neurological gait disorders in elderly people: clinical approach and classification. Lancet Neurol. 2007; 6(1):63-74. Review. Dietz V. Proprioception and locomotor disorders. Nat Rev Neurosci. 2002;3(10):781-790. Review. Garcia-Rill E. The pedunculopontine nucleus. Prog Neurobiol. 1991;36:363-389. Garcia-Rill E, Skinner RD. The mesencephalic locomotor region. I. Activation of a medullary projection site. Brain Res. 1987a;411:1-12. Takakusaki K, Habaguchi T, Ohtinata-Sugimoto J, Saitoh K, Sakamoto T. Basal ganglia efferents to the brainstem centers controlling postural muscle tone and locomotion: a new concept for understanding motor disorders in basal ganglia dysfunction. Neuroscience. 2003;119:293-308. Garcia-Rill E, Skinner RD. The mesencephalic locomotor region. II. Projections to reticulospinal neurons. Brain Res. 1987b;411:13-20. Mileykovskiy BY, Kiyashchenko LI, Kodama T, Lai YY, Siegel JM. Activation of pontine and medullary motor inhibitory regions reduces discharge in neurons located in the locus coeruleus and the anatomical equivalent of the midbrain locomotor region. J Neurosci. 2000;20:8551-8558. Hajnik T, Lai YY, Siegel JM. Atonia-related regions in the rodent pons and medulla. J Neurophysiol. 2000;84:1942-1948. Garcia-Rill E, Skinner RD, Miyazato H, Homma Y. Pedunculopontine stimulation induces prolonged activation of pontine reticular neurons. Neuroscience. 2001;104:455-465. Homma Y, Skinner RD, Garcia-Rill E. Effects of pedunculopontine nucleus (PPN) stimulation on caudal pontine reticular formation (PnC) neurons in vitro. J Neurophysiol. 2002;87:3033-3047.
55. Aziz TZ, Davies L, Stein J, France S. The role of descending basal ganglia connections to the brain stem in parkinsonian akinesia. Br J Neurosurg. 1998;12:245-249. 56. Kojima J, Yamaji Y, Matsumura M, Nambu A, Inase M, Tokuno H, et al. Excitotoxic lesions of the pedunculopontine tegmental nucleus produce contralateral hemiparkinsonism in the monkey. Neurosci Lett. 1997;226:111-114. 57. Munro-Davies L, Winter J, Aziz TZ, Stein J. Kainate acid lesions of the pedunculopontine region in the normal behaving primate. Mov Disord. 2001;16:150-151. 58. Nandi D, Liu X, Winter JL, Aziz TZ, Stein JF. Deep brain stimulation of the pedunculopontine region in the normal non-human primate. J Clin Neurosci. 2002b;9:170-174. 59. Brudzynski SM, Wu M, Mogenson GJ. Modulation of locomotor activity induced by injections of carbachol into the tegmental pedunculopontine nucleus and adjacent areas in the rat. Brain Res. 1988;451:119-125. 60. Childs JA, Gale K. Circling behavior elicited from the pedunculopontine nucleus: evidence for the involvement of hindbrain GABAergic projections. Brain Res. 1984;304:387-391. 61. Milner KL, Mogenson GJ. Electrical and chemical activation of the mesencephalic and subthalamic locomotor regions in freely moving rats. Brain Res. 1988;452:273-285. 62. Nandi D, Aziz TZ, Giladi N, Winter J, Stein JF. Reversal of akinesia in experimental parkinsonism by GABA antagonist microinjections in the pedunculopontine nucleus. Brain. 2002a;125:2418-2430. 63. Grunwerg BS, Krein H, Krauthamer GM. Somatosensory input and thalamic projection of pedunculopontine tegmental neurons. Neuroreport. 1992;3:673-675. 64. Hylden JL, Hayashi H, Bennett GJ, Dubner R. Spinal lamina I neurons projecting to the parabrachial area of the cat midbrain. Brain Res. 1985;336(1):195-198. 65. Murray MP, Sepic SB, Gardner GM, Downs WJ. Walking patterns of men with parkinsonism. Am J Phys Med. 1978;57:278-294. 66. Breit S, Bouali-Benazzouz R, Benabid AL, Benazzouz A. Unilateral lesion of the nigrostriatal pathway induces an increase of neuronal activity of the pedunculopontine nucleus, which is reversed by the lesion of the subthalamic nucleus in the rat. Eur J Neurosci. 2001;14:1833-1842. 67. Matsumura M, Kojima J. The role of the pedunculopontine tegmental nucleus in experimental parkinsonism in primates. Stereotact Funct Neurosurg. 2001;77:108-115. 68. Chang JW, Yang JS, Jeon MF, Lee BH, Chung SS. Effect of subthalamic lesion with kainic acid on the neuronal activities of the basal ganglia of rat parkinsonian models with 6-hydroxydopamine. Acta Neurochir Suppl. 2003;87:163-168. 69. Mitchell IJ, Clarke CE, Boyce S, Robertson RG, Peggs D, Sambrook MA, et al. Neural mechanisms underlying parkinsonian symptoms based upon regional uptake of 2-deoxyglucose in monkeys exposed to 1-methyl-4-phenyl-1,2,3, 6-tetrahydropyridine. Neuroscience. 1989;32:213-226.
PPN stimulation for parkinson’s disease
70. Futami T, Takakusaki K, Kitai ST. Glutamatergic and cholinergic inputs from the pedunculopontine tegmental nucleus to dopamine neurons in the substantia nigra pars compacta. Neurosci Res. 1995;21:331-342. 71. Kitai ST, Shepard PD, Callaway JC, Scroggs R. Afferent modulation of dopamine neuron firing patterns. Curr Opin Neurobiol. 1999;9:690-697. 72. Hirsch EC, Graybiel AM, Duyckaerts C, Javoy-Agid F. Neuronal loss in the pedunculopontine tegmental nucleus in Parkinson disease and in progressive supranuclear palsy. Proc Natl Acad Sci USA. 1987;84:5976-5980. 73. Jellinger K. The pedunculopontine nucleus in Parkinson’s disease, progressive supranuclear palsy and Alzheimer’s disease. J Neurol Neurosurg Psychiatr. 1988;51: 540-543. 74. Zweig RM, Jankel WR, Hedreen JC, Mayeux R, Price DL. The pedunculopontine nucleus in Parkinson’s disease. Ann Neurol. 1989;26:41-46. 75. Camicioli RM, Howieson DB, Lehman S, Kaye JA. Talking while walking: The e. ect of a dual task in aging and Alzheimer’s disease. Neurology. 1997;48:955-958.
98
76. Hausdor JM, Peng CK, Goldberger AL, Stoll AL. Gait unsteadiness and fall risk in two affective disorders: A preliminary study. BMC Psychiatry. 2004;4:39. 77. Springer S, Giladi N, Peretz C, Yogev G, Simon ES, Hausdor JM. Dual-tasking effects on gait variability: The role of aging, falls, and executive function. Mov Disord. 2006;21:950-957. 78. Woollacott M, Shumway-Cook A. Attention and the control of posture and gait: A review of an emerging area of research. Gait Posture. 2002;16:1-14. 79. Mazzone P, Lozano A, Stanzione P, Galati S, Scarnati E, Peppe A, et al. Implantation of human pedunculopontine nucleus: A safe and clinically relevant target in Parkinson’s disease. Neuroreport. 2005;16:1877-1881. 80. Plaha P, Gill SS. Bilateral deep brain stimulation of the pedunculopontine nucleus for Parkinson’s disease. Neuroreport. 2005;16:1883-1887. 81. Stefani A, Lozano AM, Peppe A, Stanzione P, Galati S, Tropepi D, et al. Bilateral deep brain stimulation of the pedunculopontine and subthalamic nuclei in severe Parkinson’s disease. Brain. 2007;130:1596-1607.
1663
88 Psychiatric Considerations in Management of Movement Disorders M. Zurowski . V. Voon . V. Valerie
Deep Brain Stimulation (DBS) is performed for a variety of movement disorder indications chief among which are Parkinson’s Disease (PD) and dystonia. Other indications include essential tremor, tremor of Multiple Sclerosis and tardive dyskinesia (TD). While the motor outcomes of DBS in these disorders are reviewed in other chapters of this text, this chapter will focus on the psychiatric sequelae of DBS. Areas reviewed will be subthalamic nucleus (STN) and globus pallidus interna (GPi) targeting for PD and GPi stimulation for dystonia as these indications have been investigated most comprehensively. Treatment of the neuropsychiatric symptoms post DBS will not be covered in this chapter. Some management strategies are offered in Voon et al. [1]
Parkinson Disease Neuropsychiatric complications of Parkinson’s Disease are common. More than 60% of patients report one or more psychiatric symptoms at some point in the course of their illness [2]. Common psychiatric symptoms include changes in cognition, mood (emotional processing, depression, mania), as well as prominent anxiety, and also psychosis, apathy, and sleep disorders [2,3]. More recently there has also been described a group of behavioral disturbances called Impulse Control Disorders (ICDs). These behaviors include ‘‘hedonistic homeostatic dysregulation’’ or a form of excessive and pathological
#
Springer-Verlag Berlin/Heidelberg 2009
use of dopaminergic medications for nonmotor purposes [4], hypersexuality, pathological gambling, hyperphagia, excessive shopping and punding [5]. These symptoms are likely related to dopaminergic medications used for the treatment of PD [6]. The effects of DBS on ICD symptoms are not clear but are slowly being elucidated and will be reviewed below. The relationship of neuropsychiatric symptoms to PD is unclear. Depressive symptoms often begin before the motor onset of PD [7] and may in fact involve systems independent of the motor pathways. For instance, dopamine has been implicated in mood, hedonistic drive, and reward in the non-PD populations [8]. Therefore when dopaminergic medications are withdrawn following successful DBS, other dopaminergic pathways may be undertreated resulting in apathy (anhedonia, aboulia) and even depression [9]. This is more of an issue post STN DBS than GPi DBS seeing that the former allows for greater dopaminergic medication reduction. Similarly, it is also unclear how stimulation parameters are related to neuropsychiatric effects of DBS since these parameters are set to optimize motor results of the surgery and not psychiatric ones.
Electrode Placement Electrode placement has a profound effect on the success of surgery in terms of movement
1488
88
Psychiatric considerations in management of movement disorders
and psychiatric outcomes. Most psychiatric outcome literature is based on the effect of STN stimulation. Targeting the STN is difficult. It is small, measuring only 0.18 cm3 [10]. It is not homogenous as it is comprised of limbic and motor components. It is also surrounded by a rich assortment of white matter tracts serving a variety of functions [11]. Because of this complexity some of the results presented below may be contaminated by subtle variations in electrode placement within the nucleus or possible current spread to adjoining structures.
Methodological Overview of Outcome Studies Neuropsychiatric outcomes, other than cognition, have not been the focus of DBS research until relatively recently. Consequently literature pertaining to psychiatric effect of globus pallidus internus (GPi) and ventrointermedial nucleus (Vim) DBS is limited as compared to subthalamic nucleus (STN) stimulation given that the latter is now the favored intervention. Therefore the apparently greater rate of cognitive and psychiatric adverse events reported after STN compared to GPi DBS may be a reflection of this report bias. Possible other confounders are the reduction of dopaminergic medication following STN DBS that may in itself account for post-operative depression and apathy [12]. Anatomic differences between the two nuclei may also predispose STN DBS to a greater neuropsychiatric side-effect profile. For instance, the STN is anatomically smaller with a more variable orientation than the GPi, this makes electrode misplacement more likely and makes it theoretically more susceptible to possible current spread to adjoining nonmotor circuits. Studies are further limited by relatively small numbers of subjects that may not be sufficient to detect even large effect size. Furthermore, most studies do not contain control groups making the interpretation of results and drawing
of conclusions with regard to neuropsychiatric effects of DBS problematic.
Cognitive Outcomes The vast majority of literature following DBS for PD has been published following bilateral STN electrode placement. Most studies suggest that this procedure is benign from a cognitive perspective in well selected patients, but conclusions have been hampered by relatively small sample sizes. The most robust finding appears to be a decline in verbal fluency [13–17], although other studies have also reported declines in verbal memory [14,17–19], conditional associative learning [20], visuospatial memory [17–19], processing speed [17], response inhibition [21] and some measures of executive function [19]. Some of the same studies also document mild improvements in mental flexibility [18,20,22], working memory [20], visuomotor sequencing [13,14,16,20], conceptual reasoning [14,20], and overall cognitive function [14]. It has been postulated that older patients, and those with moderate cognitive impairment prior to surgery are at greater risk [17–19,23,24], but this has not been clearly demonstrated. Dementia has been an exclusion criterion of most studies. In a 5-year follow up study Krack et al. have reported the gradual development of dementia in 3 of 49 patients [25]. These findings seem to reflect a gradual cognitive decline that may be commensurate with illness progression although this is uncertain as the study did not include a control group. However, there have also been isolated reports of acute cognitive changes following electrode implantation in the STN [26,27]. It is unclear how medication changes contribute to these cognitive findings. Most studies have not documented any significant cognitive decline following GPi DBS [19,28,29]. Only mild declines in semantic word fluency [30,31] and visuoconstruction [30] scores have been reported.
Psychiatric considerations in management of movement disorders
Mood Emotional Processing Several studies have addressed the issue of the effects of STN stimulation on emotional processing in PD. A mood induction study demonstrated greater emotional experience and recall of emotionally valenced immediate memory [32]. Behavioral studies have suggested that PD patients with on-STN stimulation have impaired selective recognition of negative emotional facial stimuli (anger, sadness, disgust and fear) [33–35]. In contrast, using a verbal affective priming task, Castner et al. have shown that PD patients offstimulation had impaired processing of negative stimuli (i.e. lack of the expected delay in reaction time to negative stimuli) which normalized in on-stimulation [36]. Finally, local field potential recordings during STN DBS surgery demonstrate greater event related desynchronization during exposure to both positive and negative emotional stimuli than to neutral stimuli [37]. The proposed pathophysiology of these emotional processing changes may be increased anterior cingulate and decreased putaminal activity in response to emotional stimuli secondary to STN stimulation [38]. However, conclusions drawn from these studies must still be tentative as studies differ with respect to stimuli type and task and comparison groups which are either pre-surgery or off-stimulation. Nevertheless, it does appear that processing of emotional material by PD patients is affected by STN DBS and may result in clinical sequelae.
Depression The prevalence of baseline depression in PD is between 40 and 50% [39], with the incidence rate reported to be 1.86% per year [40]. A preoperative survey documented that 60% of patients presenting for STN surgery had a past history of depression requiring treatment [41].
88
The immediate post-DBS period is associated with increased emotional reactivity, or excessive mood-congruent emotional responses to minor triggers that have been identified in 75% of STN DBS patients [24]. However, episodes of postoperative depression, or sustained low mood or anhedonia with concomitant neurovegetative changes, have been reported in only 1.5–25% of patients [24,25,42–45]. Immediate post-op depression following STN may be associated with acute dopaminergic medication withdrawal [12]. In the long term depression scores tend to remain constant [25] or improve [15] following surgery. One study documented moderate to severe depression in 33% of patients before surgery, followed by 20% of patients in the first post-operative year, and 15% in the third post-operative year [15]. Approximately half of those with severe depression preoperatively continued to be severely depressed at 3 years postoperatively. However, in another retrospective review only 33% of patients with preoperative history of depression developed postoperative depression [24], suggesting that a history of preoperative depression may be a risk factor but is not necessarily predictive. In contrast, a prospective study found that a personal history of depression, family psychiatric history, and preoperative depression scores did not differentiate between the postoperatively depressed and nondepressed groups [46]. This study found that depression was better predicted by postoperative confusion and female gender. It is interesting to note that there have been individual cases reporting acute mood effects when STN stimulators were turned on, or when voltage was increased. Depending on the positioning of the electrodes, some patients report the quality of their mood change to be different than their previous experience of depression. For instance, two patients with targets located in the STN have been described with stimulation linked tearfulness that was qualitatively different from their premorbid experience of major depression [47], In contrast, acute stimulation of the left
1489
1490
88
Psychiatric considerations in management of movement disorders
substantia nigra pars reticulata [48] led to a depressive state resembling major depression that may have been related to stimulation of nigral or nigrothalamic fibers. A similar state was reported with stimulation of the right zona incerta/fields of Forel [49]. These reports suggest that while dysphoric states may occur with stimulation of electrodes located within the STN, the full spectrum of Major Depression associated with the range of vegetative, cognitive, and motivational symptoms may be more likely related to electrodes misplaced outside the STN. It has been hypothesized that stimulation dorsal or ventral to optimal contact, misplacement of the electrode, or current leakage to adjacent tracts maybe responsible for these acute changes [1]. Stimulation of the GPi as compared to the STN is reported to have much lower mood effects [45,50,51] although studies are very limited in terms of specificity of diagnosis and patient numbers. Therefore it is premature to speculate if this is a more appropriate target for those with more severe mood disorders.
suicides and 9 completed suicides were compared to 70 controls to determine factors associated with attempted suicides it was found that postoperative depression (P < 0.0001), being single (P = 0.007), previous history of impulse control disorders or compulsive medication use (P = 0.005), and postoperative apathy (P = 0.004) were associated with 52% of the variance for attempted suicide risk. Completed suicides were associated only with postoperative depression (P < 0.001). The authors conclude that suicide is thus one of the most important potentially preventable risks for mortality following STN DBS for PD (Voon et al., unpublished data). Given these findings, PD patients should be screened for depression and suicidal ideation preoperatively. Patients with severe treatment refractory depression should be contraindicated from surgery given the potential of postoperative decompensation. Preoperative suicidal ideation resistant to treatment can also be considered a contraindication.
Mania Suicide Whereas patients with PD are seen to have 10% of the risk of completed suicide as compared to the general population, STN DBS may increase this risk. The presence of postoperative depression may relate to postoperative suicide. In a recent international multicenter retrospective survey of 55 movement disorder and surgical centers, the completed suicide percentage was 0.45% (24/5,311) and attempted suicide percentage was 0.90% (48/5,311) (Voon et al., unpublished data) in those who have undergone STN DBS. Suicide rates in the first postoperative year (261/100,000/year) (0.26%/year) were higher than the age-matched country-specific rates (Standardized Mortality Ratio for suicide: SMR 12.51; P < 0.0001) and near baseline by the third postoperative year (37/100,000/year) (0.037%/ year) (SMR 1.78; P = 0.43). When 27 attempted
Euphoria is frequently seen but rarely complained about in the post-operative period, as patients experience a dramatic improvement in their activities of daily living and a relief of painful off-period dystonia [12]. Transient mania has also been reported in 4–15% of patients usually in the first 3 postoperative months [25,52]). Mania has mood (euphoria, irritability), motivational (increased goal directed activity), cognitive (increased rate of thought processes, rapid speech, poor judgment), and motor (restlessness, agitation) symptoms that are likely related to stimulation affecting the limbic STN regions, although an influence on neighboring structures such as the medial forebrain bundle or the lateral hypothalamus due to current diffusion cannot be excluded [12]. Symptoms may be related to ongoing use of relatively high doses of dopaminergic medication with the addition of STN
Psychiatric considerations in management of movement disorders
stimulation. Clinically, as the medications are reduced symptoms of mania abate. In the rare event of psychotic mania clozapine or quetiapine and decreased stimulation parameters have been recommended [1].
Sleep Patients with PD experience a number of sleep disorders including insomnia, daytime somnolence, and parasomnias including Rapid Eye Movement Sleep Behavior Disorder (RBD) [53]. The pathophysiology of these problems is uncertain and likely including cerebral pathology of PD, dopaminergic medication and co-existing neuropsychiatric illness such as Major Depression among common causes. In a series of small studies assessing sleep post STN DBS it has been found that subjective and objective sleep qualities were improved with this intervention. The duration of the longest period of uninterrupted sleep [54], total sleep time [55], slow wave sleep and REM sleep all increased and sleep efficiency improved [56]. Stimulation of the STN reduced nighttime akinesia by 60% and completely suppressed axial and early morning dystonia [55] all of which likely contribute to improved sleep. Interestingly, none of the studies showed an improvement in RBD or periodic leg movement often seen in PD.
Impulse Control Disorders Impulse Control Disorders (ICDs) are increasingly being recognized in PD patients. Risk factors appear to be male gender, younger age, family history of substance abuse, and use of dopamine agonist medications [6]. It is yet unclear how STN stimulation affects these symptoms as small observational studies have documented both, the resolution of ICDs [57–59], as well as their triggering [60,61]. For instance, in an
88
observational study of seven PD patients with active preoperative dopamine agonist-induced pathological gambling, all patients had a long term improvement of gambling symptoms following STN DBS surgery that paralleled the decrease in dopaminergic medication dose [57]. However, two of the patients went on to develop persistent apathy following surgery. Compulsive dopaminergic medication use has also been demonstrated to improve following STN DBS in two young highly motivated PD patients with surgery early in the course of their illness [59]. Conversely, there is a report of pathological gambling symptoms worsening in two patients in the early postoperative period suggesting that like in postoperative mania, there exists an interaction between dopaminergic medications and STN stimulation. New onset pathological gambling following STN DBS surgery [61] has also been reported as has the worsening of compulsive medication use. Given the role of a potential interaction between stimulation and dopaminergic tone, careful preoperative assessment and preoperative reduction of dopamine agonist and dopaminergic dose is suggested in this patient population. Close postoperative monitoring, medication management and access to a multidisciplinary team are also recommended. Studies seeking to understand impulsivity in PD and DBS have examined impulsivity subdivided into motoric impulsivity (or motor response inhibition), the tendency to rapid decision making without adequate evaluation of choices, and impulsive choice. The effects of STN DBS on the first two factors have been assessed. With respect to cognitive studies on impulse control, van den Wildenberg et al. demonstrated that STN stimulation improves both choice reaction time and response inhibition (stop-signal task) [62]. The authors suggest that this result is due to an overall improvement of parkinsonian motor symptoms rather than a specific effect of STN stimulation on the decision making process. However, some [14] but not all studies [16] have
1491
1492
88
Psychiatric considerations in management of movement disorders
demonstrated STN-stimulation induced impairments in a response conflict task (Stroop wordcolor interference task). These findings have been parsed out further by Frank et al. who demonstrated that PD patients with on-STN stimulation actually have faster than expected reaction times to high conflict rewarding choices, suggesting that these patients do not spend an adequate amount of time deliberating over difficult decisions. This effect was not seen with high conflict punishment choices [63]. Therefore, the impulsivity seen with STN stimulation may be due to impaired decision making under conditions of high conflict reward situations, and not because of motor response disinhibition. This is in contrast to medications that seem to impair patients’ ability to learn from negative decision outcomes [63].
Apathy Apathy in nondepressed patients with PD is characterized by an isolated lack of motivation and initiative. In the general PD population the rate of apathy is between 16.5 and 42% [2,64] and appears to be associated more with cognitive dysfunction, particularly executive dysfunction, rather than depression [64]. In this population it appears to be somewhat responsive to dopaminergic medications implicating an underlying dopaminergic etiology [65,66]. In keeping with this hypothesis it has been reported that acute STN stimulation improves self-reported apathy [67]. Transient apathy seen in the immediate postoperative period also responds to increasing levodopa or acute STN stimulation [68] both of which induce similar subjective improvement of apathy that may be related to dopaminergic medication withdrawal [69]. However, worsening of apathy scores have been reported in the first six months after surgery [70] in 12 and 25% of patients 3–5 years after STN DBS surgery
[15,25]. In the long-term follow up of patients with PD after STN DBS apathy is the most frequently reported psychiatric symptom [71]. Because of the lack of control groups it is uncertain whether this is secondary to DBS, or if it is simply a representation of underlying PD progression, especially as apathy is more related to age [72] and progressive frontal executive dysfunction that is part of PD.
Quality of Life and Social Functioning Quality of Life measures post STN and GPi DBS show an improvement ranging from 14 to 62% [73] depending on methodology used. Improvement is usually seen in all dimensions measured with mobility, ADL, stigma, emotional wellbeing, and bodily discomfort showing greater improvement than social support, cognition, and communication subscales. However, using qualitative data from unstructured interviews two groups have found that some patients experience a deterioration of their social and occupational functioning following successful motor STN [74,75]. These findings do not appear to be captured by Quality of Life questionnaires such as the PDQ-39. In a group followed by Schupbach et al. it was found that marital conflict occurred in 17 of 24 couples post STN DBS, in some of whom this represented a continuation of presurgical marital difficulties [75]. There was also a 33% incidence of depression in spouses of DBS patients studied. Similar difficulties were found in patients’ professional roles as only 9 of 16 patients who had professional activity before DBS surgery went back to work after surgery [75]. Perozzo et al. have speculated that these difficulties may represent a loss of control or a sense of impotence and passivity [74] over PD, as stimulator parameters can only be adjusted by physicians. Caregivers may be worried of losing
Psychiatric considerations in management of movement disorders
their newly acquired autonomy and feel disinclined to sustain a caregiver role again. Patients may also have difficulty giving up the benefits derived from the illness condition such as more attention from the relative or irresponsibility toward everyday matters [74]. These findings suggest that social and psychological accommodation to even successful DBS may be difficult for patients most of whom have a long history of disability and dependence on others. Problems in postoperative social adjustment may be related to a variety of factors already mentioned in this chapter, including apathy, mild cognitive changes, aberrant emotional processing, and euphoria. These factors may be quite subtle in their effect. For instance, difficulties in nonverbal information processing of facial emotional expression such as anger and sadness to which patients have no insight [34] can have profound consequences in terms of interpersonal relationships. As can impulsivity in high conflict reward choice situations [63]. Therefore despite significant motor improvement and greater independence patients may still fail to make the proper social and vocational adjustments to take full advantage of the benefits of DBS.
Dystonia There is less known about psychiatric aspects of DBS for dystonia than for PD. The DBS target that has been most investigated is the GPi. Published studies suggest that GPi DBS does not cause cognitive decline [76,77]. There is a suggestion that depression improves or at least does not worsen following the procedure [76,78]. However there is one report [79] of 2 out of 16 patients committing suicide following GPi DBS. Both patients had a pre-operative history of depression. One of the patients was 53-years old with a 12 year history of cervical dystonia, he killed himself by overdose of medications and
88
alcohol within a month of surgery, before any benefit of the procedure was noticed. The second patient was 44-years old with a 35 year history of primary generalized dystonia. He had an excellent response to GPi DBS. Six months postoperatively he regained social life, was free of medication; at fourteen months he was able to resume his work at a children’s farm. Two weeks before suicide by auto-strangulation he appeared to be in good spirits when assessed in clinic. There is one other report of suicide post GPi DBS for secondary dystonia after an anoxic brain injury [80]. It is hard to draw any firm conclusions from these isolated events other than to note that there may be psychiatric risks to undergoing DBS for dystonia. These risks may have to do with electrode placement in the GPi, stimulation parameters, adaptations that patients have to make to a new life less encumbered by movement disability, or failure of the DBS as a final solution for the dystonia. As yet there are no guidelines for DBS for dystonia with regards to psychiatric illness. However it appears that patients suffering from severe depression should be followed closely by a psychiatrist familiar with movement disorders, and depression should be treated before surgery is attempted. Severe depression or suicidal ideation unresponsive to treatment may be contraindications to patients undergoing this procedure.
Conclusions Deep brain stimulation is a neuropsychiatrically safe procedure for the vast majority of patients with PD and dystonia. However there is a subset of patients that develop psychiatric problems postoperatively that may be related to electrode stimulation factors, medication changes, illness progression, or the effect of a major life altering event. Some guidelines have been developed for preoperative screening [1,81,82] of patients and their postoperative management [1,12,83]. These
1493
1494
88
Psychiatric considerations in management of movement disorders
guidelines are currently based largely on expert opinion and will continue to change as we gain more experience with DBS that is quickly becoming the mainstay of treatment for severe movement disorders.
References 1. Voon V, et al. Deep brain stimulation: neuropsychological and neuropsychiatric issues. Mov Disord 2006;21 Suppl 14:S305-S327. 2. Aarsland D, et al. Range of neuropsychiatric disturbances in patients with Parkinson’s disease. J Neurol Neurosurg Psychiatry 1999;67(4):492-6. 3. Ferreri F, Agbokou C, Gauthier S. Recognition and management of neuropsychiatric complications in Parkinson’s disease. CMAJ 2006;175(12):1545-52. 4. Giovannoni G, et al. Hedonistic homeostatic dysregulation in patients with Parkinson’s disease on dopamine replacement therapies. J Neurol Neurosurg Psychiatry 2000;68(4):423-8. 5. Voon V, et al. Prevalence of repetitive and reward-seeking behaviors in Parkinson disease. Neurology 2006;67(7): 1254-7. 6. Voon V, et al. Prospective prevalence of pathologic gambling and medication association in Parkinson disease. Neurology 2006;66(11):1750-2. 7. Mayeux R, et al. Altered serotonin metabolism in depressed patients with parkinson’s disease. Neurology 1984;34(5):642-6. 8. Bressan RA, Crippa JA. The role of dopamine in reward and pleasure behaviour – review of data from preclinical research. Acta Psychiatr Scand Suppl 2005;427:14-21. 9. Krack P, et al. Mirthful laughter induced by subthalamic nucleus stimulation. Mov Disord 2001;16(5):867-75. 10. Yelnik J, Functional anatomy of the basal ganglia. Mov Disord 2002;17 Suppl 3:S15-S21. 11. Parent A, Hazrati LN. Functional anatomy of the basal ganglia. II. The place of subthalamic nucleus and external pallidum in basal ganglia circuitry. Brain Res Brain Res Rev 1995;20(1):128-54. 12. Krack P, et al. Postoperative management of subthalamic nucleus stimulation for Parkinson’s disease. Mov Disord 2002;17 Suppl 3:S188-S197. 13. Ardouin C, et al. Bilateral subthalamic or pallidal stimulation for Parkinson’s disease affects neither memory nor executive functions: a consecutive series of 62 patients. Ann Neurol 1999;46(2):217-23. 14. Daniele A, et al. Cognitive and behavioural effects of chronic stimulation of the subthalamic nucleus in patients with Parkinson’s disease. J Neurol Neurosurg Psychiatry 2003;74(2):175-82.
15. Funkiewiez A, et al. Long term effects of bilateral subthalamic nucleus stimulation on cognitive function, mood, and behaviour in Parkinson’s disease. J Neurol Neurosurg Psychiatry 2004;75(6):834-9. 16. Pillon B, et al. Neuropsychological changes between ‘‘off ’’ and ‘‘on’’ STN or GPi stimulation in Parkinson’s disease. Neurology 2000;55(3):411-18. 17. Saint-Cyr JA, et al. Neuropsychological consequences of chronic bilateral stimulation of the subthalamic nucleus in Parkinson’s disease. Brain 2000;123(Pt 10): 2091-108. 18. Alegret M, et al. Effects of bilateral subthalamic stimulation on cognitive function in Parkinson disease.[see comment]. Arch Neurol 2001;58(8):1223-7. 19. Trepanier LL, et al. Neuropsychological outcome of GPi pallidotomy and GPi or STN deep brain stimulation in Parkinson’s disease. Brain Cogn 2000;42(3):324-47. 20. Jahanshahi M, et al. The impact of deep brain stimulation on executive function in Parkinson’s disease. Brain 2000;123(Pt 6):1142-54. 21. Hershey T, et al. Stimulation of STN impairs aspects of cognitive control in PD. Neurology 2004;62(7):1110-14. 22. Witt K, et al. Deep brain stimulation of the subthalamic nucleus improves cognitive flexibility but impairs response inhibition in Parkinson disease. Arch Neurol 2004;61(5):697-700. 23. Dujardin K, et al. Influence of chronic bilateral stimulation of the subthalamic nucleus on cognitive function in Parkinson’s disease. J Neurol 2001;248(7):603-11. 24. Houeto JL, et al. Behavioural disorders, Parkinson’s disease and subthalamic stimulation. J Neurol Neurosurg Psychiatry 2002;72(6):701-7. 25. Krack P, et al. Five-year follow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinson’s disease. N Engl J Med 2003;349(20):1925-34. 26. Hariz MI, et al. Bilateral subthalamic nucleus stimulation in a parkinsonian patient with preoperative deficits in speech and cognition: persistent improvement in mobility but increased dependency: a case study. Mov Disord 2000;15(1):136-9. 27. Morrison CE, et al. Neuropsychological functioning following bilateral subthalamic nucleus stimulation in Parkinson’s disease. Arch Clin Neuropsychol 2004; 19(2):165-81. 28. Fields JA, et al. Cognitive outcome following staged bilateral pallidal stimulation for the treatment of Parkinson’s disease. Clin Neurol Neurosurg 1999; 101(3):182-8. 29. Vingerhoets G, et al. Cognitive outcome after unilateral pallidal stimulation in Parkinson’s disease. J Neurol Neurosurg Psychiatry 1999;66(3):297-304. 30. Troster AI, et al. Declines in switching underlie verbal fluency changes after unilateral pallidal surgery in Parkinson’s disease. Brain Cogn 2002;50(2):207-17. 31. Volkmann J, et al. Long-term results of bilateral pallidal stimulation in Parkinson’s disease. Ann Neurol 2004; 55(6):871-5.
Psychiatric considerations in management of movement disorders
32. Schneider F, et al. Deep brain stimulation of the subthalamic nucleus enhances emotional processing in Parkinson disease. Arch Gen Psychiatry 2003;60(3):296-302. 33. Biseul I, et al. Fear recognition is impaired by subthalamic nucleus stimulation in Parkinson’s disease. Neuropsychologia 2005;43(7):1054-9. 34. Dujardin K, et al. Subthalamic nucleus stimulation induces deficits in decoding emotional facial expressions in Parkinson’s disease. J Neurol Neurosurg Psychiatry 2004;75(2):202-8. 35. Schroeder U, et al. Facial expression recognition and subthalamic nucleus stimulation. J Neurol Neurosurg Psychiatry 2004;75(4):648-50. 36. Castner JE, et al. Semantic and affective priming as a function of stimulation of the subthalamic nucleus in Parkinson’s disease. Brain 2007;130(Pt 5):1395-407. 37. Kuhn AA, et al. Activation of the subthalamic region during emotional processing in Parkinson disease. Neurology 2005;65(5):707-13. 38. Geday J, Ostergaard K, Gjedde A. Stimulation of subthalamic nucleus inhibits emotional activation of fusiform gyrus. Neuroimage 2006;33(2):706-14. 39. Cummings JL. Depression and Parkinson’s disease: a review. Am J Psychiatry 1992;149(4):443-54. 40. Dooneief G, et al. An estimate of the incidence of depression in idiopathic Parkinson’s disease. Arch Neurol 1992;49(3):305-7. 41. Voon V, et al. Psychiatric symptoms in patients with Parkinson disease presenting for deep brain stimulation surgery. J Neurosurg 2005;103(2):246-51. 42. Herzog J, et al. Two-year follow-up of subthalamic deep brain stimulation in Parkinson’s disease. Mov Disord 2003;18(11):1332-7. 43. Martinez-Martin P, et al. Bilateral subthalamic nucleus stimulation and quality of life in advanced Parkinson’s disease. Mov Disord 2002;17(2):372-7. 44. Ostergaard K, Sunde N, Dupont E. Effects of bilateral stimulation of the subthalamic nucleus in patients with severe Parkinson’s disease and motor fluctuations. Mov Disord 2002;17(4):693-700. 45. Volkmann J, et al. Safety and efficacy of pallidal or subthalamic nucleus stimulation in advanced PD. Neurology 2001;56(4):548-51. 46. Berney A, et al. Effect on mood of subthalamic DBS for Parkinson’s disease: a consecutive series of 24 patients. Neurology 2002;59(9):1427-9. 47. Doshi PK, et al. Depression leading to attempted suicide after bilateral subthalamic nucleus stimulation for Parkinson’s disease. Mov Disord 2002;17(5):1084-5. 48. Bejjani BP, et al. Transient acute depression induced by high-frequency deep-brain stimulation. N Engl J Med 1999;340(19):1476-80. 49. Stefurak T, et al. Deep brain stimulation for Parkinson’s disease dissociates mood and motor circuits: a functional MRI case study. Mov Disord 2003;18(12):1508-16.
88
50. Anderson VC, et al. Pallidal vs subthalamic nucleus deep brain stimulation in Parkinson disease. Arch Neurol 2005;62(4):554-60. 51. Rodriguez-Oroz MC, et al. Bilateral deep brain stimulation in Parkinson’s disease: a multicentre study with 4 years follow-up. Brain 2005;128(Pt 10):2240-9. 52. Romito LM, et al. Transient mania with hypersexuality after surgery for high frequency stimulation of the subthalamic nucleus in Parkinson’s disease. Mov Disord 2002;17(6):1371-4. 53. Larsen JP, Sleep disorders in Parkinson’s disease. Adv Neurol 2003;91:329-34. 54. Iranzo A, et al. Sleep symptoms and polysomnographic architecture in advanced Parkinson’s disease after chronic bilateral subthalamic stimulation. J Neurol Neurosurg Psychiatry 2002;72(5):661-4. 55. Arnulf I, et al. Improvement of sleep architecture in PD with subthalamic nucleus stimulation. Neurology 2000;55(11):1732-4. 56. Monaca C, et al. Effects of bilateral subthalamic stimulation on sleep in Parkinson’s disease. J Neurol 2004; 251(2):214-18. 57. Ardouin C, et al. Pathological gambling in Parkinson’s disease improves on chronic subthalamic nucleus stimulation. Mov Disord 2006;21(11):1941-6. 58. Bandini F, et al. Using STN DBS and medication reduction as a strategy to treat pathological gambling in Parkinson’s disease. Parkinsonism Relat Disord 2007; 13(6):369-71. 59. Witjas T, et al. Addiction in Parkinson’s disease: impact of subthalamic nucleus deep brain stimulation. Mov Disord 2005;20(8):1052-5. 60. Lu C, Bharmal A, Suchowersky O. Gambling and Parkinson disease. Arch Neurol 2006;63(2):298. 61. Smeding HM, et al. Pathological gambling after bilateral subthalamic nucleus stimulation in Parkinson disease. J Neurol Neurosurg Psychiatry 2007;78(5):517-19. 62. van den Wildenberg WP, et al. Stimulation of the subthalamic region facilitates the selection and inhibition of motor responses in Parkinson’s disease. J Cogn Neurosci 2006;18(4):626-36. 63. Frank MJ, et al. Hold your horses: impulsivity, deep brain stimulation, and medication in parkinsonism. Science 2007;318(5854):1309-12. 64. Pluck GC, Brown RG. Apathy in Parkinson’s disease. J Neurol Neurosurg Psychiatry 2002;73(6):636-42. 65. Chatterjee A, Fahn S. Methylphenidate treats apathy in Parkinson’s disease. J Neuropsychiatry Clin Neurosci 2002;14(4):461-2. 66. Marin RS, et al. Apathy: a treatable syndrome. J Neuropsychiatry Clin Neurosci 1995;7(1):23-30. 67. Czernecki V, et al. Does bilateral stimulation of the subthalamic nucleus aggravate apathy in Parkinson’s disease? J Neurol Neurosurg Psychiatry 2005;76 (6):775-9.
1495
1496
88
Psychiatric considerations in management of movement disorders
68. Funkiewiez A, et al. Acute psychotropic effects of bilateral subthalamic nucleus stimulation and levodopa in Parkinson’s disease. Mov Disord 2003;18(5):524-30. 69. Krack P, et al. Subthalamic nucleus or internal pallidal stimulation in young onset Parkinson’s disease. Brain 1998;121(Pt 3):451-7. 70. Drapier D, et al. Does subthalamic nucleus stimulation induce apathy in Parkinson’s disease? J Neurol 2006; 253(8):1083-91. 71. Funkiewiez A, et al. Effects of levodopa and subthalamic nucleus stimulation on cognitive and affective functioning in Parkinson’s disease. Mov Disord 2006;21(10):1656-62. 72. Ory-Magne F, et al. Does ageing influence deep brain stimulation outcomes in Parkinson’s disease? Mov Disord 2007;22(10):1457-63. 73. Diamond A, Jankovic J. The effect of deep brain stimulation on quality of life in movement disorders. J Neurol Neurosurg Psychiatry 2005;76(9):1188-93. 74. Perozzo P, et al. Deep brain stimulation of subthalamic nucleus: behavioural modifications and familiar relations. Neurol Sci 2001;22(1):81-2. 75. Schupbach M, et al. Neurosurgery in Parkinson disease: a distressed mind in a repaired body? Neurology 2006; 66(12):1811-16. 76. Halbig TD, et al. Pallidal stimulation in dystonia: effects on cognition, mood, and quality of life. J Neurol Neurosurg Psychiatry 2005;76(12):1713-16.
77. Pillon B, et al. Preservation of cognitive function in dystonia treated by pallidal stimulation. Neurology 2006; 66(10):1556-8. 78. Kupsch A, et al. Pallidal deep-brain stimulation in primary generalized or segmental dystonia. N Engl J Med 2006;355(19):1978-90. 79. Foncke EM, Schuurman PR, Speelman JD. Suicide after deep brain stimulation of the internal globus pallidus for dystonia. Neurology 2006;66(1):142-3. 80. Burkhard PR, et al. Suicide after successful deep brain stimulation for movement disorders. Neurology 2004;63(11):2170-2. 81. Lang AE, et al. Deep brain stimulation: preoperative issues. Mov Disord 2006;21 Suppl 14:S171-S196. 82. Saint-Cyr JA, Albanese A. STN DBS in PD: selection criteria for surgery should include cognitive and psychiatric factors. Neurology 2006;66(12):1799-800. 83. Deuschl G, et al. Deep brain stimulation: postoperative issues. Mov Disord 2006;21 Suppl 14:S219-S237.
93 Selective Thalamotomy and Gamma Thalamotomy for Parkinson Disease C. Ohye
Selective Thalamotomy Introduction Since the first edition of this book, stereotactic surgery has changed considerably, because of the rapid progress in computerized imaging systems and neuroscience. Technically, one of the most remarkable changes is the development of deep brain stimulation (DBS) initiated in 1996 [1] and used all over the world now. In fact, this method is nowadays used as “a la mode”. However, DBS has an intrinsic weak point as will be described in this chapter. As a consequence of this weak point the number of thalamotomies performed have reduced but the technical and scientific progress holds promise of a more precise operation to get reliable, safe results. Therefore, in this chapter, a comparatively new aspect of modern thalamotomy will be described.
Understanding the Thalamus During the past decade, several important new concepts of the thalamus, beginning with the most well-known Hassler’s Atlas [2] in higher primates including humans, have emerged (Percheron [3], Hirai and Jones [4], Parent [5], Ohye [6], Morel [7]). Among them, in relation to stereotactic thalamotomy, new percellation of the thalamic nuclei and hence the evolution #
Springer-Verlag Berlin/Heidelberg 2009
of nomenclature are remarkable. Although there is still some confusion, we can distinguish rather clearly (whatever name is used) [8] VA, VO, Vim, VC in this order [9] from rostral to caudal, in the lateral part of the human thalamus where stereotactic intervention is most frequently undertaken. These nuclei, (part of the so called ventral tier), receive nigral, striatal, cerebellar spinal and lemniscal input and send fibers to the related cerebral cortex [10,11]. The border of each thalamic nucleus is not always very clearly defined; the difference has to be kept in mind before invasion into these areas. In fact, as shown already and as will be shown again later Vim, for example, is specifically related to the tremor because a selective lesion in a part of Vim after physiological identification results in immediate, complete arrest of the tremor [12]. Further, there is also the topographic representation of the contra-lateral body part within the Vim nucleus itself [13], roughly, lower limb area in the dorsolateral part, upperlimb area in the middle and face area in the most ventromedial part. The thalamic representation area and the clinically affected patients’ body part should be matched when final coagulation or stimulation is made. In order to perform such a precise operation, microrecording to get physiological information first, prior to coagulation, and a small coagulation of less than 100 mm3 has been shown to be enough to treat Parkinsonian tremor [14]. Once the tremor disappears on the operating table, the effect continues for a long period of 5–10 years in excellent cases [15].
1550
93
Selective thalamotomy and gamma thalamotomy for parkinson disease
Chemical substances present in the human thalamus are not yet well understood, although the presence or distribution of several transmitters and related substances have been described [16–18]. For example, the presence of aminoacids (glutamate,GABA), acetylcholine, monoamines (dopamine, norepinephrine, serotonin), calcium binding proteins (calbidin, calretinin, palvalbumis), gut brain peptides (cholecystokinin, neurotensin, tachykinin, neurotensin) have been known for some time. However, in Vim, which is our area of interest there is no definite substance identified. The Toronto group examined the activity of the Vim neuron by microinjection of muscimol (GABA agonist) and found that the tremor was suppressed [19]. Therefore it may be possible that GABA is related to synaptic transmission of Vim neuron. The study of the human thalamus is not easy; however we need to obtain more information. The advances in basic research have changed our ideas of thalamic function as a whole. We now know that there are many neuronal pathways that go through the thalamus; the cerebello thalamo cortico-cerebellar loop is well-known and there are others like the cortico-striatopallido-thalamo-cortical loop, cortico-thalamo (N, reticularis)-cortical loop [20,21] and corticothalamo-(VA-VL and MD) – cortical loop. Moreover, the thalamus receives afferent fibers from the spinal cord, (to Vim, VC), the brain stem nucleus, the pedunculo pontine nucleus, (to CM–Pf), substantia nigra (to VA), etc. and also sends fibers to the respective connecting areas. Thus, the interrelations of the thalamus with its surroundings are more complex than was once thought. The thalamus is now known to be more involved with higher neuronal functions rather than serving as a simple relay station. It plays a role not only in sensory motor function but also in the cognitive, associative, (corollary), and perception memory attention consciousness [22]. In view of these developments, the target within the thalamus for stereotactic surgery is now
increasing, for example, the CM–Pf complex is considered for treating pain and epilepsy. Precise evaluation is yet to be done.
Technical Improvement One of the big differences in modern stereotactic surgery is the use of the computer; the Leksell surgiplan [23,24], for example. This device is used along with the computerized imaging system (MRI or CT image). Three dimensional brain images are shown synchronized together in the same 3D coordinates. This means that we can always see 3D images with the same reference points and 3D values. It provides us clear visual information of the morphological features of the brain. Use of Surgiplan (sps). Sps is a very useful computer assisted support system for identifying the target point. This system provides 3D brain images (MRI, CT) with the same reference points and the same coordinate system. The system goes through the following steps: 1. 2. 3.
4.
Register patient file. Take MR, CT images and send them to the main computer. AC–PC decision. Using Fase image (heavy T2), anterior and posterior commissure positions are determined by pointing on the image. The respective 3D coordinates are registered automatically. Functional target. The tentative target point is settled on the images after registration of the intended coordinates. This point is not the coagulation center but may pass through it toward the center. We choose, for example, 5 mm anterior to PC(Y), 16–17 mm from the mid line (X), and on the level of IC line (Z), the point being shown by a cross on each 3D image. Then the point is modified on the 3D images, taking into account 45% of the thalamic length and 2–3 mm medial from the thalamo cortical border.
Selective thalamotomy and gamma thalamotomy for parkinson disease
5.
Paths definition. Determine the safety path from the cortical surface (entrance) to the tentative target point (zero). Use 45 to IC line on the lateral view, and 10 on the A-P view as the standard. Care is taken not to damage the cortical vein around the entrance point, and not to invade the edge of the lateral ventricle. A line connecting the target and cortical entrance point is drawn, simulating the electrode trajectory and along this line each image is successively shown at the vertical plane. This ensures that we can see (through the surgeon’s eye), the cerebral structure of each point and hence to avoid undue damage by the electrode or coagulation needle. The distance from entry point to target is also shown. If there are no objects in the path, the 3D coordinates of the tentative target and the trajectory pathway to the target is determined. Thus after setting the 3D coordinates on the frame, the electrode or coagulation needle is oriented exactly to the reference point choosing any angle of the arch.
Surgiplan is of great use in target planning. Even the advanced computer imaging system cannot visualize the thalamic target nucleus clearly but only give a global idea. As already illustrated, our standard target for tremor surgery is 7 mm anterior to CP, 4 mm above the level of IC (inter commissural line)-line and 2 mm inside the thalamo-capsular border. This point is easily shown on MR (or CT) images but we should be careful about individual differences of the thalamus. In this regard, a classic work by Brierly and Beck [25] leads us to possible individual differences. They claim that each thalamic nucleus in humans is more precisely represented by the proportion of the thalamic length and not by its distance from the posterior commissure. In general most stereotactic neurosurgeons use Hassler’s atlas as the reliable standard atlas to orient the recording or
93
stimulation electrode. However, there are some pitfalls in using this atlas. Therefore, to modify the target point according to the size and shape of the individual patient, the method mentioned earlier serves as a useful guide. In fact there are considerable differences in the thalamic length and shape that cannot be ignored. For example, the variation of the thalamic axis at the depth to the mid line is 23–50 parallel to the width of the third ventricle. And the mid point of the Vim nucleus in the axial section, from the posterior commissure which we often refer to, is located at 53–66% of the thalamic length, though the center of the Vim nucleus is almost constantly found at around 45% of the thalamic length [26,27]. Since we began applying these new methods to our stereotactic thalamotomy in almost every case we have been able to get enough neurophysiological information as described later here, and consequently the operations have resulted in almost 100% success even on the operating table (immediate effect). In some exceptional cases with thalamic deformity due to a previous hemorrhage, tumor, or severe head injury, we had difficulty finding the exact target point, even with the help of microrecording. In such cases, reoperation becomes necessary.
Microrecording The recording electrode is a steel-steel concentric needle type originally made in a French laboratory. Its details have already been presented [12,28]. The patient is always in an awake state, so that he can communicate any change in his status or condition that he senses. The microrecording records general background activity; and it also records the spike discharge with the help of a micromanipulator, that enables us to advance the electrode slowly or quickly, continuously or step by step, as required.
1551
1552
93
Selective thalamotomy and gamma thalamotomy for parkinson disease
The recording process starts after making a small burr hole over the pre frontal area. The Leksell stereotactic system uses a guide hole attached to the electroholder with three small holes in a line at 3 mm intervals (It is also available at 4 mm intervals). We use the center hole and one posterior hole. Usually, the center hole is used for setting the zero point and for coagulation while the posterior position is for depth recording. This standard setting is based on the fact that statistically 80% of the important physiological information is obtained from the posterior electrode, 3 mm away from the center and parallel to the parasagittal midline. In fact, in our earlier operations we used two electrodes at the same time, but after several such operations it was clear that the posterior electrode more frequently recorded useful neuronal activity (kinesthetic and /or tremor response: see later section); therefore in recent times depth recording is done only through the posterior electrode to avoid undue damage of the brain. Today we recognize that the posterior electrode and hence the posterior coagulation needle, is a good indicator of the posterior wall (limit) where the final coagulation occurs. The final coagulation occurs between the center and posterior needles. Data recorded by the microelectrode are fed into the oscilloscope for on line monitoring, together with surface EMG from the contra lateral extremities (only directly related EMG activity is selected and shown on the oscilloscope). At the same time these data are recorded on paper, through the EEG machine. Using the nine-channel EEG machine, thalamic activities, its integrated background activity, and thalamic spike (changed to pulse pattern) are recorded. The thalamic activity and the EMGs of the four channels are registered in Data analyzer (PowerLab, A-D Instrument). The latter two data are used later for analysis using PowerLab, which has several programs to do precise spike analysis (spike histograms, correlation, and configuration of the spike itself etc.). PowerLab facilitates data analysis.
Results of Microrecording, Data Acquisition With this type of electrode, we can record the neural activity in the depth of the brain, the global background electrical activity, and spike discharge (multiple or single neuronal activity) depending on the position of the electrode tip. As we approach the thalamic nucleus (mostly toward the lateral part of the Vim nucleus in the case of patients with tremor) the recording electrode passes through the cerebral cortex, white matter, caudate nucleus, white matter again and descends to the thalamus. As already described, these structures exhibit characteristic electrical activity. Briefly the spikes are seen superposed on the slow oscillation at the cerebral cortex, small positive spikes are seen at the white matter, 15–20 Hz slow waves and spikes of wide duration at the caudate nucleus, and small positive numerous spikes in the thalamus. Therefore we can determine the approximate position of the electrode by the characteristic electrical activity. In this regard, the audio monitor is very useful. Within the thalamus, the neuronal activity changes depending on the different nucleus. In general, at first, in the dorsal thalamus, small or medium size spikes, sometimes with slow waves are seen. In the dorsal thalamic area, however, no particular response is seen in the VO nucleus. This is because our electrode is relatively large, and not enough fine to record small neurons in the VO area. If a finer electrode is used, there will be voluntary movement-related activity as shown by Jasper, Raeva etc. Between the dorsal thalamus and VC area, large spikes indicate that we are in the Vim nucleus where electrical activity exhibits marked spontaneity. In the Vim, especially in its lateral part, in the typical case, rhythmic neuronal activity, time-locked with the contra lateral tremor movement, exists (> Figure 93-1). Also, the same spike discharge responds to passive movement of the muscle (flexion or
Selective thalamotomy and gamma thalamotomy for parkinson disease
93
. Figure 93-1 Examples of thalamic neuronal activity. (a) Upper three traces show tremor rhythmic activity of a Vim neuron (the first trace) with EMG of the upper arm (second and third traces). It also responds to passive movement of contralateral elbow (next two traces); an example of kinesthetic response. (b) Tactile response in the Vc nucleus. Light touch on the left mental-chin area (shaded area) provoked sharp response
extension). After several analyses, these responses are assumed to originate from the stretching muscle spindle. These are the most important landmarks to decide on the final coagulation point in our microrecording method. Other physiological findings about the Vim-Kinesthetic neuron are presented elsewhere [27,29–31].
Distribution of Kinesthetic and Tremor Neurons in Vim In order to accomplish selective thalamotomy, depth recording to get exact information of
neuronal activity in each individual case is indispensable, because Vim neurons receive signals from trembling muscle and transmit the same to the cortical 3a area. [29,32–34] In theory, by stereotactic surgery, we destroy most of the tremor mediating neurons. In this regard, we looked for the optimal area of the Vim nucleus but, in each individual case, the effective area of coagulation is variable in mms. So we studied the precise distribution of the kinesthetic and/or rhythmic neurons in the Vim nucleus. We found evidence in the literature, from studies by our group and by other centers, that rhythmic neurons are distributed in a wide area in and
1553
1554
93
Selective thalamotomy and gamma thalamotomy for parkinson disease
around the standard Vim nucleus. But these studies showed the distribution of tremor-related cells by plotting the recorded points and showing them all together in one image so that the individual differences were not considered [35–37]. Therefore in the recent 50 Vim thalamotomy cases, we studied thoroughly the distribution in each patient separately (> Figure 93-2 and > Figure 93-3). It was revealed that tremor neurons had really diffused, occupying the field in and around the Vim nucleus. The dorsal most point was 12 mm above the level of the IC line and the lowest point was on the same level as the IC line. It was interesting to note that the population of kinesthetic and tremor rhythmic neurons are most densely packed between 6 and 1–2 mm dorsal to the level of the IC line. Therefore the optimal center for tremor surgery is 3 mm above the level of the IC line. It is noteworthy that this value is exactly the same as the one we had already deduced from the results of selective thalamotomy. Another important fact is that if we draw a circle of 3 mm radius, which mimics a lesion of coagulation or Gamma irradiation, 80% of the points were involved and destroyed, but the rest 20% escaped from the therapeutic lesion, . Figure 93-2 Two hundred thirty two Vim neurons recorded from recent 50 operations. All recorded points were demonstrated by unfolding the electrode penetrating area of cone shape zone, and arranged according to the order of the highest point of recording track in each case. Abscissa is the number of 50 patients and the ordinate is mm from the level of the intercommissural line
revealing theoretical (mathematical) evaluation of the ablative surgery. As our Vim thalamotomy is always assisted by microrecording the volume of the coagulation area is decided by physiological finding, and the center of coagulation can be adjusted in each case. On the contrary, in case of Gamma thalamotomy in which no adjuvant method is available, we take only the average value as mentioned above, and then inevitably, about 20% of the area is not covered. In the case of such a blind operation, the deficit is only theoretically estimated. In view of this, the importance of microrecording cannot be emphasized enough.
Lamellar Organization Within the Vim nucleus, microrecording revealed another important neuronal organization – rostrocaudal lamellar arrangement as reported from animal experiments. In thalamic surgery, when two recording electrodes are introduced in parallel to the sagittal plane, in our case 3 mm apart, natural stimulation of a certain contralateral area induced reaction in two electrodes at the same time [30] in some cases. It is interesting to note that in the central nervous system, (in
. Figure 93-3 Bar graph made from the previous image ( > Figure 93-3). Each recorded points were summed up to make side bars. Note that most of the responding Vim neurons were distributed from zero to 500–600 mm zone dorsoventrally, suggesting the optimal target range
Selective thalamotomy and gamma thalamotomy for parkinson disease
humans also), neurons are frequently arranged in the rostrocaudal sense. Consequently, when we make final coagulation by Leksell’s pair coagulation needle system, the lesion is made along the rostrocaudal direction rather than in the mediolateral sense. We have reason to believe that the result is more effective in this case.
Tap Response In recent years, we have been interested in the response to tapping on the hand. When the electrode is introduced into the Vim nucleus, not infrequently a large positive slow wave
93
superimposed by a couple of spike discharges is recorded [38] (> Figure 93-4). This happens soon after passive pronation supination movement of the forearm and slightly rostral to the cutaneous response to a light touch on the hand. Usually the positive deflexion is very large and similar to the EEG evoked responses to electrical stimulation of the peripheral nerve. But tap response is easily obtained by natural stimulation by skin tap and not by a light touch on the skin surface. The rough estimated latency of the tap response was about 15 ms. The response was always phasic, it was assumed to be related to the Paccini corposcule of subcutaneous origin. As this particular response always occurs just
. Figure 93-4 Example of a tap response, by tapping on the forearm flexor muscle. One of the large evoke response recorded by slow sweep is extended by fast sweep in the upper right corner
1555
1556
93
Selective thalamotomy and gamma thalamotomy for parkinson disease
before the electrode enters the so-called main sensory nucleus of VC (ventralis caudalis), it is recognized as a warning sign that the tap of the recording electrode is close to the VC nucleus, lesion of which results in paresthsia. In practice, we stop at this point and the coagulation is made including this point but not invading into deeper into the area. Until the tap response area is reached, the coagulation can be made without noticeable complication.
. Figure 93-5 A part of the sagittal view of the thalamus (drawn on the right upper corner) illustrating the proper target zone for tremor surgery (a red circle). Three parallel lines denote direction of recording or coagulating needle of our standard Vim thalamotomy. A horizontal line is the intercommissural line. Corresponding Vim nucleus is shown by zigzag zone. Shaded area is a model of our standard coagulation area by Leksell’s dual needle system
Selective Lesion The clinical importance of the particular neuron group in the lateral part of Vim is that a small selective lesion within this area is enough to arrest various kinds of tremors. When we make a therapeutic lesion, we always depend on the physiological findings, estimating the extent of the high amplitude Vim zone and recording points of kinesthtic and tremor neuron. The first coagulation is made between the center and posterior coagulation needle after adjusting the direction of the posterior needle and the height, in the dorso-ventral sense. Even earlier in the 1970s before the time of MRIs and CTs, we estimated the volume of the minimum effective lesion for tremor as about 60 mm3. [14] Since then we have continued to make the most effective lesions within the lateral part of the Vim nucleus and now it is widely recognized and accepted so. This type of minimally effective lesion is realized by partly overlapping coagulations 3–4 times around the center needle (> Figure 93-5). Using a pair of 4 mm coagulation needles with intervals of 3 mm, the direction of the posterior needle is adapted to match the clinical symptom (hand or leg etc.) and physiological finding. If the upper limb neuron is recorded and the patient has upper limb tremor, the first coagulation is made between the center needle and posterior needle (3 mm apart), adjusting the dorso ventral height so as to coagulate including the active Vim area. Then
we add a second and third lesion by adjusting the posterior needle to cover the necessary area of Vim. Thus, in Parkinsonian and essential tremor, a small coagulation of about 40–60 mm3 is sufficient to arrest the tremor [7]. In contrast, in patients with coarse tremor after stroke or severe head injury and in whom kinesthetic neurons are often difficult to find within the standard Vim areas as anatomically defined, the effective coagulation volume tends to be larger – 100–200 mm3 or more. In these cases, the recording process may be repeated and be successful even in a deteriorated Vim area. In practice, in our recent series with microrecording, the immediate effect of our lesions has been almost 100% for tremor patients. The long-term effect, as long as 10 years after the procedure, of selective Vim thalamotomy is also satisfactory, being comparable or superior to that of other reported cases. Therefore, in this group, tremor can be regarded as cured. In one out of 10–15 cases, Parkinsonian tremor may recur 2–3 weeks after the operation, and in such a case, reoperation with recording
Selective thalamotomy and gamma thalamotomy for parkinson disease
reveals rhythmic discharge near the coagulated area, indicating that the former lesion was not large enough. In this case an additional small lesion gives a good result. For coarse tremors after stroke and head injury, the risk of residual or recurrent tremor is higher, probably because the morphological and functional changes in and around the thalamus make the intraoperative physiological study difficult. Meticulous exploration is required in such cases. For the treatment of Parkinsonian rigidity and other abnormal hypertonic states, the coagulative lesion is placed mainly in Vo. For this purpose, our strategy is to identify Vim first and then make a lesion anterior to Vim, because Vo neurons exhibit no specific (or at least definitive) physiological activity.
Complications Untoward side effects resulting from a physiologically guided selective Vim thalamotomy are minimal. In almost every case, however, the patient complains of a “light” feeling in the contra lateral limb just after the coagulation. This feeling usually continues for several weeks and then gradually subsides. It may continue in some cases for several months before the patient gets used to it. This may be inevitable because we more or less destroy the projection area of deep sensation, and the feeling of lightness is assumed to reflect the loss of deep sensation. Certainly, this is better than weakness of capsular origin; global muscle power is maintained, and deep tendon reflexes are normal. In one patient among 20, there will be temporary numbness or paresthesia, often around the lips or fingertips, as a result of secondary or temporary damage to Vc. Generally, this subsides within a month. Persistent paresthesias are rare, occurring once in approximately 100 cases. Dysarthria is another annoying side effect resulting from destruction of the deep pharyngeal representative area either on the left or right side of the thalamus. When kinesthetic responses
93
from lip, tongue, or pharyngeal compression are encountered, it is wise not to make lesions including these points. Aphasia or dysphasia is apt to occur after bilateral thalamic lesions. However, restricted selective Vim thalamotomy usually doses not cause such language dysfunction, even after a bilateral invasion. Severe, unexpected accidents such as intracerebral hemorrhage (direct or indirect) and central nervous system infection may occur once in 200 cases. Because we introduce a small electrode or needle into the brain, such accidents are unavoidable, even with a highly sophisticated technique.
Summary Although the modern tendency in thalamotomy is deep brain stimulation, exactly defined selective thalamotomy aided by microrecording is a valuable procedure to treat many cases of involuntary movement, without noticeable complications. In this section, the basic idea and practical procedures of thalamotomy were described. Recent basic research on the human thalamus relevant to our thalamotomy for Parkinson’s disease and related movement disorders were also mentioned.
Gamma Thalamotomy Introduction Application of the gamma knife to functional disorders takes a back seat among the fields of radiosurgery, despite several pioneering works [39–44] mainly because the target of functional disorders is not visualized even with modern advanced neuroimaging techniques. In 2006, functional disorders accounted for only 7.6% of gamma knife treatments worldwide. In Japan, about 4% of gamma knife treatments were for functional disorders. We were interested in treating cases with movement disorders and central pain by gamma
1557
1558
93
Selective thalamotomy and gamma thalamotomy for parkinson disease
knife, as introduced by Leksell [45] in the early era of gamma knife treatment. For us, this is an extension of what we have done as selective thalamotomy with the aid of microrecording [12,46]. Here, in relation to the previous section, we describe briefly the principal technique used and the results on movement disorders.
Indication In using the gamma knife for the treatment of movement disorders, we considered applying the idea of stereotactic surgery that we have practiced for more than 40 years; we started on cases with tremor type Parkinson disease. Since 1992, we have operated on 150 cases of involuntary movement (mainly tremor type Parkinson’s disease and essential tremor) and on 15 cases with central pain. In our experience, Gamma thalamotomy for cases with tremor, whatever the cause, has been successful in 80% of the case. For cases of central pain, the results will be presented elsewhere. The criteria for the selection of patients are almost the same as for usual stereotactic thalamotomy [12]. The patients eligible for gamma thalatomy are: 1. 2. 3. 4.
5. 6. 7.
Those with tremor and/or rigid type PD, who cannot have drug therapy Elderly patients Those with no psychiatric problems Those with no other severe complications such as hypertension, diabetes mellitus or cerebral infraction Those with one side dominance Those who refuse to have open surgery Those who accept a delay of about 6 months for effective results
thalamotomy. We explain to the patient and his or her family the advantages and disadvantages in each operation, and the final decision is often with the patient. Full consensus is obtained in every case.
Hospitalization Three days hospitalization is the rule in our clinic (cf. 2 weeks for open thalamotomy). On the first day general physical and laboratory examinations (blood analysis, urinalysis), X-ray for chest and cranium, EKG, EEG and EMG are performed. CT examinations are useful to get approximate estimation of the morphological characteristics in each individual case. For example, if any cortical atrophy or deformity of the thalamus exists, the CT would reveal it. In the evening of the first day of hospitalization, all the above data obtained are shown to the patients and the details of operative procedures explained. On the second day, Gamma thalamotomy will be done under local anesthesia. Details of the operative procedures are described in the next part. In brief, the frame fixation starts at 9:00 followed by MRI and CT. Target planning and dosemetry follow. After obtaining the target 3D coordinates, the patient is led to the Gamma knife room and the irradiation treatment starts. This usually takes about an hour or an hour and a half to finish, and the whole procedure ends around 13:00. The patient is requested to rest on that day. Mild headache may occur just after releasing the stereotactic frame or the pain at the pin screw point may remain for some time. On the third day, the patient can be discharged.
Operation Among these conditions, 1, 3, and 5 are the same as for usual stereotactic operation. After almost 10 years of experience, we are now convinced of safe and reliable results from gamma knife
The operation starts with fixation of the stereotactic frame (Leksell’s G-frame) on the skull of the patient. At this moment, care is taken to fix
Selective thalamotomy and gamma thalamotomy for parkinson disease
the angle of the frame relative to the position of the head, not compressing respiration when the head is fixed to the headrest during MRI examination and Gamma irradiation. After fixation, MRI and CT are taken. For MRI, 2 mm slices of heavy T2 weighted image (FASE) are used to get the axial image from the level of the lateral ventricle to the cerebral aqueduct, 2 mm slices of proton density weighted axial and coronal images for the level of the thalamus. All images including CTare transferred to Leksell’s Surgiplan and Gamma plan [24]. Target planning, the most important process of gamma thalamotomy with the aid of Surgiplan, comes next. It is better to check possible displacement in MRI, by plotting the ventricular system on FASE images and superimposing these images on the CT images (> Figure 93-6). A difference within 1 mm is acceptable. As described in the previous section [12,30], the optimal target for the treatment of tremor (the lateral part of the thalamic ventralis intermedius nucleus) is not visualized clearly and we depend at first only on the posterior
93
commissure defined in the FASE image. The standard coordinates deduced from our experiences of microrecording guided stereotactic thalamotomy are: 7 mm anterior to the posterior commissure, 4 mm above the level of the intercommissural (IC) line, and 2 mm medial to the thalamo-capsular border as shown in > Figure 93-6. These coordinates are variable in each case depending on the shape of the ventricular system, and shape of the thalamus itself, notably the distance of the IC line. At this moment, it is necessary to check the tentative focus by anatomical fact, namely the position of the Vim nucleus usually found at about 45% from the anterior tip of the thalamus in the horizontal plane [25,27,31] (> Figure 93-6). If the patient has already had stereotactic thalamotomy on the other side, the target planning on the contralateral side is more easily obtained because we can see the thalamic lesion on MRI and then the symmetric point will be a suitable target. We have had about 40 such cases. After setting the 3D coordinates of the tentative target point, these values are transferred
. Figure 93-6 Horizontal sections of the thalamus illustrating our standard optimal target point in a MR image (left) and the corresponding atlas (right) to show ventral tire nuclei with their respective input–output relations
1559
1560
93
Selective thalamotomy and gamma thalamotomy for parkinson disease
to the Gamma plane to plan for the dose. Examining the dose distribution curve projected on the axial and coronal planes, we verify the safety planning for our standard maximum dose of 130 Gy (> Figure 93-7). If the initial planning is not suitable, it is modified to a safer one not invading too much (10–15% isodose curve) into the internal capsule laterally with the main sensory nucleus of VC located posteriorly. Then, some plugging pattern is applied to mainly avoid the internal capsule and eye lens. Usually about 20~30 plugs are used.
Results The clinical course of Gamma thalamotomy for tremor reduction is characterized by slow changes
over several months, variable in each case, in contrast to the immediate effect of radiofrequency coagulation. The first case in our series, an old lady with Parkinson’s disease took one of the typical slow recovery courses (150 Gy was used on our first six cases) – the tremor did not change for several months. But it began to reduce around 1 year after the treatment and almost subsided [43]in time. This time course was considered as our standard one until renewal of the cobalt source, which accelerated tremor reduction by almost 3–6 months as will be shown below. Thalamic reaction. Thalamic reaction after irradiation revealed by MRI follow up was variable in each case. But roughly, two kinds of reactions were noticed. One was a round, restricted low signal zone surrounded by a ring-like high
. Figure 93-7 An example of the images by Leksell’s gamma plan for target in 3D fashion. In this patient with Parkinson disease, target planning was done by surgiplan and transferred to gamma plan for dosemetry (shown by isodose rings)
Selective thalamotomy and gamma thalamotomy for parkinson disease
signal area (> Figure 93-8) like the radiofrequency lesion in usual stereotactic thalamotomy. The other was a round zone with streaking along the thalamocapsular border (> Figure 93-9). In the other reactions, sometimes an irregular shaped high signal area extending to the internal capsule or medial thalamic area was noticed. Often it was also accompanied by a high signal streaking along the thalamocapsular border or rail like streaking along the thalamocapsular and pallidocapsular borders [47,48]. We could not predict which type of reaction might occur before or at the moment of irradiation. We do not know yet the ongoing pathological sequence to explain these changes, except the central low signal zone that may correspond to the real necrotic area. Time sequential, systematic examination of the thalamic lesion revealed somewhat more complicated features [48]. However, in general the thalamic lesion became visible after 3 months. Usually it was seen as a round high signal zone with ambiguous borders. For the most part, it
93
remained so for the next 3 months, but some of them increased in volume up to several hundred mm. After 6 months, in most of the cases, the lesion volume decreased or remained at the same volume, and only in exceptional cases, did the volume increase further. But in any case, no remarkable clinical complications were noticed. At present, we cannot predict the result of such thalamic reactions. There is a tendency to believe that the younger the patient, the smaller the lesion (including quick wound healing), but this is not the general rule. Therefore, regular follow up study in each individual case is very important. The time course for clinical improvement is parallel to that of UPDRS as shown in > Figure 93-10.
Reoperation by Gamma Knife If the clinical effect is not manifested within 1 year, or the reduction of the tremor is not adequate, a second Gamma thalamotomy is attempted on
. Figure 93-8 Two MR images of horizontal section of the thalamus. Similar thalamic reaction was seen (white, high signal round point), made by different lesioning; radiofrequency coagulation (left) and by gamma knife irradiation (right). In both cases, the clinical result was excellent, almost no tremor
1561
1562
93
Selective thalamotomy and gamma thalamotomy for parkinson disease
. Figure 93-9 Typical two different thalamic reactions after the same gamma irradiation (130 Gy, 4 mm collimator, one shot) in our recent series. Simple round or oval reaction (left) was more often seen than that with streaking along the thalamocapsular border (right). In some cases, the shape of the thalamic reaction changes with time course, but it did not affect the clinical course
the same side. A thorough examination of the irradiated lesion on MRI, often shows it is too small or has deviated slightly from the proper target zone. Then, the new target is decided on considering the remaining tremor site and matching it with the thalamic representation of the body. We did ten such cases, and all of them resulted in good improvement of the tremor.
Reoperation with Microrecordings In four cases in our early series, in which the clinical effect was not sufficient even after observation for more than 1 year, usual stereotactic thalamotomy with microrecording was performed afterward [49]. In these cases of insufficient results, the irradiated lesion was too
medial or anterior, the real crucial point for tremor arrest having escaped direct irradiation. So, after discussion with the patient, we attempted a reoperation. The new target was settled at 1–2 mm more posterior-lateral point, within the area of the previous center of irradiation and within the surrounding high signal zone as shown in > Figure 93-11. When a microrecording electrode was introduced into the point of the supposed damaged Vim nucleus, to our surprise, almost normal activity was recorded, including kinesthetic response, and rhythmic grouped discharge time locked to the corresponding peripheral tremor. After coagulation of this point, the tremor stopped immediately without any noticeable complications. We concluded that the high signal area surrounding the center of irradiation had not been severely
Selective thalamotomy and gamma thalamotomy for parkinson disease
93
. Figure 93-10 An example of the parallel development of clinical effect on tremor measured by UPDRS together with thalamic reaction after usual gamma knife irradiation
damaged by the gamma beam and had contained functionally normal neurons that survived as well.
Gamma Thalamotomy After Stereotactic Thalamotomy As a reverse order strategy, gamma thalamotomy in a case with early recurrence of tremor after insufficient stereotactic thalamotomy (radiofrequency lesion) also seems to be a useful procedure. Relatively early recurrence of tremor may occur due to insufficient coagulation even with microrecording control. In our opinion, it may be encountered in younger patients in whom the coagulative lesion recovers within a couple of months after radiofrequency thalamic
coagulation. In fact, in our recent series, we had 11 such cases. An example is shown in > Figure 93-8a. This was the case of a young lady with essential tremor, who had been suffering for long from spontaneous tremor in her bilateral hands without notable improvement after medication. She was operated in our clinic in 2000 and underwent left sided thalamotomy aided by microrecording. Its immediate effect was excellent, the tremor stopped completely. But about 2 months later the tremor recurred as before. MRI examination revealed that the coagulation lesion shrank to become a small lesion located too posteromedially. Therefore, after discussion with the patient and her family, we tried to extend the lesion in a more anterolateral direction, covering more of the Vim nucleus, by gamma knife.
1563
1564
93
Selective thalamotomy and gamma thalamotomy for parkinson disease
. Figure 93-11 The thalamic electrical activity (upper most trace) adjacent to the irradiated high signal zone (lower MRI) with the contralateral EMG (second and third traces) showing almost normal neuronal activity and rhythmic burst activity of tremor 1 year 7 months after gamma thalamotomy. In this case, right sided thalamic coagulation was successfully performed 6 years ago
The target planning was done as discussed above and 130 Gy was given in one shot, with 4 mm collimator as the usual gamma irradiation. The residual tremor decreased gradually about a couple of months after irradiation, and almost completely disappeared in 1 year. She can now use her right hand almost normally for writing, eating, cooking etc. In other cases with similar conditions of recurrent tremor, an additional lesion was made, considering the somewhat deviated radiofrequency lesion, to cover the escaped effective area. The results were good as far as the tremor and rigidity were concerned. In this sense, gamma thalamotomy could be a useful optional method to help correct insufficient results after radiofrequency coagulation.
Summary and Conclusion The outline of Gamma thalamotomy for Parkinson’s disease was presented. It should be emphasized that the basic idea was deduced from our own experiences of selective thalamotomy with microrecording. The 3D coordinates of the tentative target point were taken from the most effective and proper coagulation lesions obtained in the previous selective thalamotomy with microrecording (7 mm from the posterior commissure, 2 mm from the thalamo capsular border and 4 mm above the level of the intercommissural line) and these values were further corrected by a particular anatomical study by Brierley and Beck [25]. By this method, although there remain a few ambiguous points, the overall results were
Selective thalamotomy and gamma thalamotomy for parkinson disease
satisfactory, being at least about 80% successful for tremor and rigidity in Parkiinson’s disease and related disorders. For akinesia and pure gait disturbance as freezing, it is not so effective and the real target is still an open question. Because no information is available on thalamic neuronal reactions to the high dose single fraction GK irradiation that we use [50–56], we must be adequately cautious in follow-up of patients on a long term basis to assess ongoing clinical changes and thalamic reactions [57,58]. We have, in fact, recognized several different features of thalamic reactions with approximately 10 years of follow-up. In this sense, we emphasize that, for the treatment of PD by GK thalamotomy, the maximum dose is an essential factor in avoiding severe untoward complications as described in a critical report [59]. The standard dose of 130 Gy in our series appears to be both safe and effective, producing no major complications. Target planning is another essential factor which needs to be standardized. As described herein and earlier, the configuration of the thalamius, i.e., the length and the relative position of the posterior commissure, which we use very often, are quite variable [48]. In performing GK thalamotomy without a direct visible target on computerized images and no adjuvant method of overcoming individual variations, knowledge of routine stereotactic thalamic surgery is highly recommended. We are now conducting a multicenter trial to determine the safest and most reliable GK treatment for PD.
References 1. Benabid A, Pollak P, Gao G, et al. Chronic electrical stimulation of the ventralis intermedius nucleus of the thalamus as a treatment of movement disorders. J Neurosurg 1996;84:203-14. 2. Hassler R. Architectonic organization of the thalamic nuclei. In: Schaltenbrand G, Wahren W (Ed). Stereotaxy of the Human Brain, 2nd ed. Stuttgart, G Thieme; 1977. 3. Percheron G. Thalamus. In: Paxinos G, Mai JK(eds). Human Nervous System. 2nd ed. Amsterdam: Elsevier; 2004. p. 592-673.
93
4. Hirai T, Jones EG. A new parcellation of the human thalamus on the basis of histochemical staining. Brain Res Rev 1989;141-34. 5. Parent A, Hazrati NL. Functional anatomy of the basal ganglia. 1. The cortico-basal ganglia-thalamo-cerebral loop. Brain Res Rev 1995;20:91-127. 6. Ohye C. Thalamus and Thalamic Damage. In: Ramachandran VS (Chief Ed) Encyclopedia of Human Brain. Vol 4. Amsterdam et. Academic Press; 2002. p. 575-87. 7. Morel A, Margnin M, Jeanmonod D. Multiarchitectonic and stereotactic atlas of the human thalamus. J comp Neurol 1997;387:588-630. 8. Macchi G, Jones EG. Toward agreement on terminology of nuclear and subnuclear divisions of the motor thalamus. J Neurosurg 1997;86:670-85. 9. Fang PC, Stepniewska I, Kaas JH. The thalamic connections of motor, premotor, and prefrontal areas of cortex in a prosimian primate (Otolemur garnetti). Neuroscience 2006;143:987-1020. 10. McFarland NR, Haber SN. Thalamic relay nuclei of the basal ganglia from both reciprocal and nonreciprocal cortical connections, linking multiple frontal cortical areas. J Neurosci 2002;22(18):8117-32. 11. Ohye C. Thalamotomy for Parkinson’s disease and Other Types of Tremor. Par1. Historical Background and Technique. In: Gildenberg PL, Tasker RR (Eds). Textbook of Stereotactic and Functional Neurosurgery. New York: McGraw Hill; 1998. p. 1167-78. 12. Hassler R, Mundinger, F Riechert T. Stereotaxis in Parkinson symdrome. Berlin Heiderberg, NY: SpringerVerlag; 1977. 13. Hirai T, Miyazaki M, Nakajima H, Shibazaki T, Ohye C. The correlation between tremor characteristic and the predicted volume of effective lesions in stereotaxic. Nucleus ventralis intermedius thalamotomy. Brain 1983;106:1001-18. 14. Nagaseki Y, Shabazaki T, Hirai T, Kawashima Y, Hirato M, Wada H, Miyazaki M, Ohye C. Long term follow-up after selective thalamotomy. J Neurosurg 1986;65:296-302. 15. Blomqvist A, Ericson AC, Craig AD, Broman J. Evidence for glutamate as a neurotransmitter in spinothalamic tract terminals in the posterior region of owl monkeys. Exp Brain Res 1996;(1):33-44. 16. Salt TE. Glutamate receptor functions in sensory relay in the thalamus. Phil Trans R Soc London B 2002;357:1759-66. 17. Steriade M, Jones EG, McCormick DA. Thalamic organization and chemical neuroanatomy. In: Thalamus. Vol 1, Amsterdam: Elsevier; 1997. p. 31-174 18. Pahapill PA, Levy R, Dostr ovsky JO, Davis KD, Rezai AR, Tasker RR, Lozano AM. Tremor arrest with thalamic microinjections of muscimol in patient with essential tremor. Ann Neurol 1999;46:249-52. 19. Jones EG. Thalamic organization and function after Cajal. Prog Brain Res 2002;136:333-57.
1565
1566
93
Selective thalamotomy and gamma thalamotomy for parkinson disease
20. Nakano K. Neural circuits and topographic organization of the basal ganglia and related regions. Brain & Dev 2000;22:S5-S16. 21. Sidibe M, Pare JF, Smith Y. Nigral and pallidal inputs to functionally segregated thalamostriatal neurons in the centromedian/parafascicular complex in monkey. J Comp Neurol 2002;447(3):286-99. 22. Parent M, Parent A. Single-axon tracing and threedimensional reconstruction of centre medien-parafascicular thalamic neurons in primates. J Comp Neurol 2005; 481(1):127-44. 23. McFarland NR, Haber SN. Convergent inputs from thalamic motor nuclei and frontal cortical areas to the dorsal striatum in the primate. J Neurosci 2000;20(10): 3798-813. 24. Hikosaka O. GABAergic output of the basal ganglia. Prog Brain Res 2007;160:209-26. 25. Steriade M, Deschenes M. The thalamus as an oscillator. Brain Res Rev 1984;8:1-63. 26. Van der Werf YD, Witter MP, Groenewegen HJ. Anatomical and functional evidence for participation in processes of arousal. The intralaminar and midline nuclei of the thalamus. Anatomical and functional evidence for participation in processes of arousal and awareness. Brain Res Rev 2002;39(2-3):107-40. 27. Weigel R, Krauss JK. Center median-parafascicular complex and pain control. Review from neurosurgical persprctive. Stereotac Func Neurosurg 2004;82(2-3):115-26. 28. Ohye C, Shibazaki T. [Leksell Surgeplan(sps)] (Japanese) In Frontiers of Neurosurgery No.2, New Technology Toward 21st Century 2000. p. 375-38. 29. Shibazaki T, Ohye C. [Leksell Gamma Plan.] (Japanese) In Frontiers of Neurosurgery No.2, New Technology Toward 21st Century 2000. p. 369-74. 30. Brierley JB, Beck E. The significance in human stereotactic brain surgery of individual variations in the diencephalons and globus pallidus. J Neurol Neurosurg Psychiat 1959;22:287-98. 31. Ohye C, Shibazaki T, Sato S. [Location of the thalamic Vim nucleus. Its relation to the whole thalamic length.] Funct Neurosurgery 2002;41(2):52-3 (Jpn). 32. Ohye C. Neural Noise Recording in Functional Neurosurgery. In: Gildenberg PL, Tasker RR (Eds). Textbook of Stereotactic and Functional Neurosurgery. New York, McGraw Hill; 1998. p. 941-7. 33. Ohye C, Shibazaki T, Hirai T, Wada H, Hirato M, Kawashima Y. Further physiological observations on the ventralis intermedius neurons in the human thalamus. J Neurophysiol 1989;61:488-500. 34. Ohye C. Functional Organization of the Human Thalamus. In: Steriade M, Jones EG, McCormick DA (eds). Thalamus, Amsterdam: Elsevier; 1997. p. 517-42 35. Ohye C, Shibazaki T, Sato S, Cai XT. Distribution of kinesthetic neurons in the thalamic ventralis intermedius nucleus. Its relation to gamma thalamotomy. Funct Neurosurg 2003;42(2):108-9.
36. Miyagishima T, Takahashi A, Kikuchi S, Watanabe K, Hirato M, Saito N, Yoshimoto Y. Effect of ventralis intermedius thalamotomy on the area in sensorimotor cortex activated by passive hand movement: MRI image study. Stereotact. Funct Neurosurg 2007;85:225-34. 37. Ohye C. Neuronal Mechanism of Tremor. Progr Neurol 1981;25:106-17. 38. Ohye C. Production and physiological study of Holmes tremor in monkeys. In: LeDoux M (ed). Animal model of movement disorders. Birlington MA: Elsevier Academic Press; 2005. p. 377-91. 39. Albe-Fessard D, Arfel G, Guiot G et al. Acitvites electriques caracteristiques de quelques structures cerebrales chez l’homme. Ann Chir 1963;17:1185-214. 40. Hamani, C, Dostrovsky JO, Lozano, AM. The motor thalamus in Neurosurgery. Neurosurgery 2006;58(1): 146-58. 41. Magnin M, Morel A, Jeanmonod D. Single unit analysis of the pallidum, thalamus, and subthalamic nucleus in parkinsonian patients. Neuroscience 2000;96:549-64. 42. Ohye C, Shibazaki T, Zama A, Sato S. Thalamic neuron responding to tapping on the contralateral extremity. (Japanese) Func Neurosurg 2005;44:24-5. 43. Duma CM, Jacques DB, Kopyov OV, Mark RJ, Copcutt B, Farokhi HK. Gamma knife radiosurgery for thalamotomy in parkinsonian tremor; a five- year experience. J Neurosurg 1998;88:1014-49. 44. Larsson B, Leksell L, Rexed R, Sourander P, Mair W, Andersson B. The high energy proton as a neurosurgical tool. Nature 1958;182:1222-3. 45. Leksell LA. Stereotaxis and radiosurgery. An operative system. Springfield II: CC Thomas; 1971. 46. Lindquist C, Steiner L, Hindmarsh T. Gamma knife thalamotomy for tremor. Report of two cases. In: Steiner L (ed). Radiosurgery, New York: Raven Press; 1992. p. 237-43. 47. Rand RW, Jaques DB, Melbye RW, Copcutt BG, Fisher MR, Levenick MN. Gamma knife thalamotomy and pallidotomy in patients with movement disorders: Preliminary results. Stereotact Funct Neurosurgery 1993;61:65-92. 48. Young RF, Jacques S, Mark R, et al. Gamma knife thalamotomy for treatment of tremor: long term results. J Neurosurg 2000;93 Suppl 3:128-35. 49. Leksell L. Cerebral radiosurgery. I. Gamma thalamotomy in two cases of intractable pain. Acta Chir Scand 1968;134:585-95. 50. Ohye C. From selective thalamotomy with microrecording to Gamma Thalamotomy for movement disorders. Stereotac Func Neurosurgery 2006;84:155-61. 51. Ohye C, Shibazaki T, Hirato H, Inoue H, Andou Y. Gamma thalamotomy for Parkinsonian and other kinds of tremor. Stereotac Funct Neurosurg 1995;66 Suppl 1 33-42. 52. Sato S, Ohye C, Shibazaki T, Zama A. Neurophysiological evaluation of the optimum target in Gamma
Selective thalamotomy and gamma thalamotomy for parkinson disease
53.
54.
55.
56.
57.
58.
thalamotomy: Indirect evidence. Stereotact Funct Neurosurg 2005;83:108-14. Ohye C, Shibazaki T, Zhang J, Andou Y. Thalamic lesion produced by gamma thalamotomy for movement disorders. J Neurosurg 2002;97 Suppl 5:600-6. Ohye C, Shibazaki T, Sato S. Gamm knife thalamotomy for movement disorders. Evaluation of the thalamic lesion and clinical results. J Neurosurg 2005;102 Suppl:234-40. Ohye C, Shibazaki T, Ishihara J, Zhang J. Evaluation of gamma thalamotomy for parkinsonian and other tremors: survival of neurons adjacent to the thalamic lesion after gamma thalamotomy. J Neurosurg 2000;93 Suppl 3 120-7. Berry RJ, Hall EJ, Forster DW, et al. Survival of mammalian cells exposed to X-rays at ultra-high dose rate. Br J Radiol 1969;42:102-7. Friehs GM, Ojakangas CL, Pachatz P, Schrottner O, Otto E, Pencil G. Thalamotomy and caudatotomy with the gamma knife as a treatment for parkinsonism with a comment on lesion sizes. Stereotac Func Neurosurg 1995;64 Suppl 1 209-221. Friehs GM, Noren G, Ohye C, DUma CM, Marke R, Plombon J, Young RF. Lesion size following gamma
59.
60.
61.
62.
63.
64.
93
knife treatment for functional disorders. Stereotact Funct Neurosurg 1996;66 Suppl 1:320-8. Friedman DP, Goldman W, Flanders AE, et al. Stereotactic radiosurgical pallidotomy and thalamotomy with the gamma knife: MR imaging findings with clinical correlation- preliminary experience. Radiology 1999;212:143-50. Hall EJ. Radiobiology for the Radiologist Fourth Ed, (M Urano, translated in Japanese) Tokyo: Shinohara Pub. 1995. Kondziolka D, Lunsford LD, Flickinger JC. The radiobiology of radiosurgery. Neurosurg Clin N Am 1999;10(2): 157-66. Flickinger JC, Lunsford LD, Wu A, et al. Predicted dose-volume isoeffect curves for stereotactic radiosurgery with the 60Co gamma unit. Acta Oncol 1991;30:363-7. Flickinger JC, Kondziolka D, Lunsford LD. Radiobiological analysis of tissue responses following radiosurgery. Technol Cancer Res Treat 2003;2(2): 87-92. Okun MS, Stover NP, Subramanian T, Gearing M, Wainer BH, Holder CA, Watts RL, Juncos JL, Freeman, A Evatt ML, Schuele SU, Vitek JL, DeLong MR. Complications of gamma knife surgery for Parkinson disease. Arch Neurol 2001;58:1995-2002.
1567
96 Subthalamic Nucleus Stimulation for Parkinson’s Disease A. L. Benabid . J. Mitrofanis . S. Chabardes . E. Seigneuret . N. Torres . B. Piallat . A. Benazzouz . V. Fraix . P. Krack . P. Pollak . S. Grand . J. F. LeBas
Introduction The discovery in 1987 of the effects of stimulation at high-frequency able to mimic in a reversible and adjustable manner the effects of local destruction of functional targets has revived the functional neurosurgery of movement disorders, and currently other fields of functional neurosurgery such as psychosurgery. VIM thalamotomy allowed demonstrating the main properties of high frequency stimulation on the most spectacular symptom of Parkinson disease, which is tremor. However, the therapeutic effect was quickly clearly recognized as almost strictly limited to the alleviation of tremor and had no effect on bradykinesia and rigidity. The fact is that, although it is the most visible of these symptoms, tremor is not the most disabling: the difficulties of advanced parkinsonian patients are essentially related to akinesia and rigidity. The publication in 1990 by the groups of Mahlon DeLong [1] and of Crossman [2] of the prominent role of the subthalamic nucleus (STN) in the control of motor function and of its importance when destroyed to improve akinetic and rigidity symptoms in MPTP monkeys has opened new horizons for deep brain stimulation (DBS) in movement disorders. However, due to its reputation as the source of hemiballism when destroyed by hemorrhages, the STN did not appear to be a very attractive surgical target [3]. For this reason, subthalamotomy was not a good procedure to exploit this basic science discovery, #
Springer-Verlag Berlin/Heidelberg 2009
the experience acquired during high-frequency stimulation (HFS) of the thalamus suggested that STN could be a target for neuroinhibition methods as provided by DBS. This assumption was supported by the results of experiments in MPTP monkeys replicating the conclusions of Bergmann’s and Aziz’s reports using HFS instead of lesioning [4]. The ultimate confirmation of the interests of this target was given when the first patients with advanced Parkinson’s disease were implanted, showing that tremor, rigidity, and bradykinesia were very significantly improved by this method [5,6], allowing to decrease the drug dosage by 60% in average [7], and therefore alleviating the levodopa induced motor fluctuations and dyskinesias [8]. Since that time, this has been used all over the world and several thousands of patients have been operated and improved, making this method the reference surgical procedure for advanced Parkinson’s disease.
Functional Anatomy of the Subthalamic Nucleus Preamble In recent times, rather than just forming a minor link in a greater basal ganglia circuitry loop, most authorities now consider the subthalamic nucleus (STN) as a major driving force of basal ganglia activity and hence a key influence on movement
1604
96
Subthalamic nucleus stimulation for parkinson’s disease
control, together with cognitive and limbic-type functions [9–16]. In particular, the cortex is thought to drive basal ganglia circuitry, not only through its classical input to the striatum, but also through the STN. With this new view in mind, it is perhaps not surprising that dysfunction of this nucleus generates quite striking clinical afflictions and that it has formed a focus for therapeutic intervention. This new found and pivotal significance attributed to the STN, quite surprisingly, much belie its small size and apparent simplicity of cellular make up and connectivity. In the section that follows, the anatomy of the STN will be explored, from its cytoarchitecture to its connections with other neural centers. From this, an insight into the functional significance of this rather remarkable nucleus may hopefully be gained.
Topography, Cytoarchitecture and Chemoarchitecture The STN is generally oval-shaped and lies on the inner surface of the peduncular portion of the white matter of the internal capsule (> Figure 96-1a). More caudally, the medial part of the nucleus overlies the rostral portions of the substantia nigra complex. It is particularly welldeveloped in primates and, perhaps as a consequence, lesions to the STN in these species, as against those in non-primates, generate very distinct clinical symptoms [10,12,16]. The STN has a high density of darkly stained cells; in Nissl-stained sections, its boundaries can often be made out clearly from surrounding diencephalic structures (> Figure 96-1a). The cells of the STN, although thought to be made up of a homogenous functional group of cells, have been described to have either spindle-shaped, pyramidal, or rounded somata (> Figure 96-1b). Each gives rise to several spiny dendritic processes that often align parallel with the rostrocaudal axis of the nucleus. These processes may be very long,
some up to nearly 1 mm. In primates, these processes are thought to remain within the boundaries of the STN, while in non-primates, STN dendrites tend to ‘‘trespass’’ onto the territories of adjacent structures, such as the zona incerta. The greater bulk of the STN cells project to other neural centers, but a small population of interneurones is suspected. There is no doubt that the dominant neurotransmitter associated with the STN cells is glutamate (> Figure 96-1b), although other neurochemicals have been reported, for example parvalbumin (calcium buffering protein; > Figure 96-1d) and nitric oxide synthase (> Figure 96-1e). It should be noted that classically, the STN was thought not to be glutamatergic and excitatory, but on the contrary, GABAergic and inhibitory! Indeed, most, if not all STN projections have now been revealed as glutamatergic and they exert extremely powerful excitatory effects on their target structures. For this reason, among others, the STN has been suggested to be a major driving force – and central feature – of the basal ganglia circuitry [10–12,16]. The STN cells have high spontaneous tonic activity with irregular bursts [17]. During a typical electrode track through the brain, the STN is characterized by a sudden increase in the background noise (from the relatively quiet of the zona incerta) reflecting the high cell density and cells with bursting activity (> Figure 96-1f). The electrode location within the sensorimotor sector (see below) of the STN is often verified by the response of cells (increased firing audible through signal transduction and amplification) to passive movement of the contralateral limbs. As with other major centers of the basal ganglia, the STN can be divided into distinct functional sectors. First, there is a large sensorimotor sector that occupies the dorsolateral regions of the nucleus. STN cells in this sector respond to somatosensory stimulation, changing discharge rates during movements of the contralateral limbs (in fact, there is a segregated map of
Subthalamic nucleus stimulation for parkinson’s disease
96
. Figure 96-1 Photomicrographs of the anatomy (a–e) and electrophysiology (f) of the subthalamic nucleus (STN) in the Rhesus monkey. (a) Nissl-stained section of the location of the STN in the diencephalon. The small box indicates the approximate region where photomicrograph in b was taken from. (b) Nissl-stained STN cells. (c) Glutamate immunoreactive STN cells. (d) parvalbumin (calcium binding protein) immunoreactive STN cells. (e) nitric oxide synthase immunoreactive cells. (f) indicates the characteristic firing pattern seen in the STN by microelectrodes; high background noise and cells that show bursting activity. All anatomical images are of coronal sections, with dorsal to top and lateral to right. Scale bar = 100 mm. ST: subthalamic nucleus. ZI: zona incerta. ic: internal capsule. FF: fields of Forel. III: third ventricle
1605
1606
96
Subthalamic nucleus stimulation for parkinson’s disease
the body found within the STN). Second, there is a small associative territory that occupies the ventromedial region of the nucleus. Cells here are activated during visual oculomotor tasks. Finally, there is a limbic sector, occupying the medial tip of the nucleus. This sector receives inputs from limbic cortex and ventral regions of the globus pallidus. In primates, these sectors involve distinct cell groups that respond to the different types of stimuli; in non-primates, however, the sectors are less clear, perhaps because individual STN cells tend to have multiple projection sites and inputs [10–12,16].
Connections The STN connections with other parts of the brain have been well described in non-primates, as well as primates. As with most neural centers, the STN has favored and major connections with some regions, as well as some less favored and minor connections with others. In general, the major connections are with primary motor cortex, the external segment of the globus pallidus (GPe), substantia nigra pars reticulata (SNr) and the pedunculopontine tegmental nucleus; the minor connections are with the striatum, intralaminar nuclei of the thalamus and various brainstem nuclei [10–12,16].
the STN cells. Through these inputs, the STN is thought to be the pivotal nucleus of the basal ganglia through which the cortex influences the outputs of the basal ganglia [10–12,16].
Globus Pallidus The STN projects to all parts of greater pallidal complex. In primates, the bulk of STN cells have been shown to project to either internal segment of the globus pallidus (GPi) or the GPe, not to both. In non-primates, however, nearly all STN cells project to both pallidal segments, as well as the SNr. There is some topography evident in the STN-pallidal projection. STN cells lying caudal and medial in the nucleus project to the GPi, while cells lying centrally in nucleus project to the GPe. In each case, the STN terminations in the globus pallidus are arranged in quite striking arrays parallel to medullary laminae. These arrays or ‘‘bands’’ are organised similar to and in register with those from striatum. Indeed, they have been shown to converge upon the same pallidal cells. Unlike striatal inputs, however, the STN inputs are excitatory and form asymmetric synapses. Although the STN projections projects to both GPi and GPe, the STN receives inputs mainly from the GPe only. Indeed, the GPe has been described as providing one of the heaviest inputs to the STN [10–12,15].
Cortex Substantia Nigra Complex Although the STN has no projections to the cortex, the cortex provides a major input to the STN. This cortical projection is rather selective, arising mainly from primary motor cortex, and to a lesser extent, from prefrontal cortex. Most of the cortical cells projecting to the STN in these areas are from layer V, and the inputs are mainly via collaterals of axons terminating elsewhere [12]. These cortical axons terminate in small boutons on spines and fine dendrites of
The STN also projects to the other basal ganglia output nucleus, the SNr. The STN axonal terminals in the SNr have again been shown to be glutamatergic and form asymmetric synapses with dendritic shifts and to a lesser extent somata. There have also been reports of a direct STN projection to the SNc, but these are sparse [12,18]. Nevertheless, these could still generate glutamate excitotoxicity in the parkinsonian
Subthalamic nucleus stimulation for parkinson’s disease
condition. In this context, it should be noted that the STN projections to the SNr are thought to be related strongly with the SNr projection to the SNc [12]. Since the SNr sends projections to the SNc, the STN may ‘‘switch off’’ the dopaminergic SNc cells via its heavy projection to the SNr. Hence, during deep brain stimulation of the STN – of which inhibits the abnormally overactive STN cells in Parkinson disease – there is an increase in the activity of SNc dopaminergic neurons [19] and dopaminergic levels in striatum [20]. This may counteract the direct STN projection to the SNc, which is weaker than the SNr projection [18], but may not necessarily prevent the release of glutamate onto SNc cells by the overactive STN in parkinsonian cases (and hence glutamate excitotoxicity). Since the STN projects to both the globus pallidus and SNr, it is thought to influence activities of cells that constitute the output systems of basal ganglia. These STN efferents may control the initiation of movement, before setting the physiologic conditions (eg, membrane potential) of pallidal and SNr cells to appropriate levels prior to arrival of striatal signals [12]. Thus, the STN is in a position to influence the whole net output of the basal ganglia.
Pedunculopontine Tegmental Nucleus In view of recent experimental evidence, this brainstem nucleus has recently become included by most authors in the basal ganglia scheme. In particular, and with reference to the STN, it has been shown recently that the pedunculopontine nucleus may in fact be a contributing factor in generating the characteristic overactivity of the STN in parkinsonian cases [21]. In terms of connections, reciprocal ones between the pedunculopontine nucleus and STN have been documented well in primates and non-primates. It has been shown that individual STN cells project directly to
96
the pedunculopontine nucleus (pars compacta portion) and not elsewhere; interestingly, a distinct population of STN cells project to the basal ganglia output nuclei, the globus pallidus and SNr. In turn, the pedunculopontine nucleus sends an excitatory cholinergic (+glutamatergic) projection back to the STN; some recent reports suggest that some pedunculopontine pathways may even be GABAergic inhibitory [22].
Other The STN has been described to have other, generally more sparse connections with various neural centers, including the intralaminar nuclei of the thalamus, the dorsal raphe, as well as a direct projection with the striatum [9,11–16].
Function of the STN in Experimental and Clinical Settings Dysfunction of the STN leads to some striking motor afflictions. For instance, if the STN becomes abnormally underactive or lesioned in humans, a distinct set of clinical symptoms are revealed – namely a series of violent chorea-like movements, referred to as hemiballismus. Conversely, if the STN becomes abnormally overactive, usually after a loss of midbrain dopaminergic cells, the symptoms associated with Parkinson disease manifest – namely akinesia, rigidity and tremor [12]. For example, in Parkinson disease patients, 6-OHDA (6-Hydroxydopamine)-lesioned rats and MPTP (methyl-4-phenyl-1,2,3,6-tetrahydropyridine)-treated monkeys (latter two are wellknown animal models of Parkinson disease), there is an increase in the firing rate of cells in the STN. Accompanying this increase in mean firing rate, there is a change in the pattern of firing. STN cells show an increase in oscillatory bursts and synchronicity [9,12,13,16,23]. In addition, in MPTP-treated monkeys, there is an
1607
1608
96
Subthalamic nucleus stimulation for parkinson’s disease
increase in glucose metabolism and mitochondrial enzyme activity in the STN. In fact, there are reports that STN cells enhance their activity during the course of the MPTP treatment, even before the first appearance of clinical signs [17]. Further, in Parkinson disease patients and MPTPtreated monkeys, both lesion and deep brain stimulation (at high frequency) of the STN alleviate the parkinsonian symptoms, presumably by reducing the activity of the basal ganglia output streams (see above). For these reasons, the STN is viewed as a centerpiece of pathophysiology in Parkinson disease and has become a popular surgical target for relief of motor symptoms [24,25].
Summary For such a small collection of cells, with such apparent simplicity of neurochemical make-up and connectivity, it is truly extraordinary that its dysfunction can cause such havoc in normal neurological function. One feels that the story is not complete, however, and that there are still many secrets of functional anatomy that the STN holds. One hopes that these will be revealed soon, in order for a better appreciation of STN function in normal and abnormal cases.
Material and Methods Indications and Contraindications Clinical Indications, and Good Prognosis Factors Idiopathic Parkinson’s Disease
The best responsive patients present with the clinically diagnosed Parkinson’s disease in whom it can be predicted that the main symptoms of the triad (bradykinesia, rigidity, and tremor), will be significantly improved [6,7,26–28].
Levodopa Responsiveness
It has been shown [29] that the percentage of improvement with the best optimal adjustment of antiparkinsonian medication or suprathreshold dose of levo-dopa (for instance 300 mg in a single dose) is highly predictive of a similar improvement after optimal placement of the electrodes into the subthalamic target. Moreover, the levodopa response is one of the mandatory criteria of diagnosis of idiopathic Parkinson’s disease. Motor Fluctuations
Severe levodopa related motor complications despite optimal adjustment of antiparkinsonian medication are usually significantly improved after surgery, which plays a major role in the improvement of the quality of life [27,30]. This is explained by the mechanism of dyskinesia induction, related to the pulsatile administration of levodopa [31]. The decrease or arrest of this pharmacogical aggression allowed by the beneficial effects of STN stimulation restores a more normal pharmacokinetic regimen of the striatal dopaminergic receptors.
Contraindications Dementia
Dementia, as well as cognitive deficits, are definitely not improved by STN DBS and might even be increased by the multifactorial traumatism constituted by the different steps of the surgical management. More over, they participate to the prognostic evaluation of the evolution of the Parkinson’s disease, outside of the range of the early evaluation of the disease and may constitute elements to suspect an atypical parkinsonian syndrome, or the entrance of the patient in a stage of additional system degeneration, in particular, cholinergic. Moreover, this means also that the patients are entering a stage of the disease where they might not take the full benefit of the motor improvement, their quality of life being mostly impaired by the cognitive disorder.
Subthalamic nucleus stimulation for parkinson’s disease
Surgical Contraindications
All surgical contraindications (anticoagulants, terminal cancer, infectious disease and immunodeficiency, etc.) apply to the selection of the patient for DBS in general, particularly when the risks related to the penetration of the brain by a probe are concerned, such as bleeding. Additional contraindications are related to the generation of high frequency electrical side effects which may interfere with the correct functioning of sensing devices such as cardiac pacemakers and defibrillators, which need to have specific features preventing the artifacts to be considered as cardiac signals.
Bad Prognosis Factors Age
Age, as in all surgical therapies is more or less negatively related to the general outcome. However this must be carefully evaluated for each patient in whom the effects of age vary considerably. Gait Disturbance
This is a topic which has to be carefully assessed by the neurological team before surgery. Freezing is part of the patterns of akinesia, which usually responds to the levodopa treatment. When the freezing of gait happens, persists and is not improved in the on-medication period, this is usually not improved by STN stimulation (and might be improved by the low-frequency stimulation of a new target, which is the pedunculopontine nucleus [32–34]) and therefore should be taken into account into the decision-making process or at least the patient, the family, and the caregivers should be clearly informed [35]. Speech Problems
Speech is usually improved, but much less than the other motor symptoms [36,37]. When patients have hypophonia before surgery, this might be impaired or worsened after, particularly when the
96
medication doses are significantly decreased. The improvement due to DBS might not replace the improvement due to medication: this leads to severe hypophonia from which the patient and the family usually complain. This also should be taken into account in the decision-making process, particularly when there is a pre-existing hypophonia, and should be part of the information delivered to the patient, family, and caregivers. Atypical Parkinsonism (Multiple Systemic Atrophy, Progressive Supranuclear Palsy)
Although there are no specifically oriented studies to address those questions, the common experience is globally negative; the symptoms outside of the idiopathic Parkinson’s disease triad (dementia, autonomic disorders, cognitive decline and dementia, oculomotor disturbance, etc.) are usually not improved, at least significantly. However, particularly during the initial phases, the improvement of the motor symptoms might help significantly the patients, although this is for a limited period of time, the benefits being secondarily obliterated by the appearance of the other symptoms, particularly of the cognitive decline. This should be carefully balanced, the improvement of the motor symptoms being sometimes of a real value for the patients even for a relatively short period of time.
Frequently Asked Questions Previous Ablative Surgery (Thalamotomy, Pallidotomy, Sub thalamotomy, etc.)
There are several reports showing that previous ablative surgery is not a contraindication for DBS in general, and of STN in particular, provided that the ablative procedure has not destroyed the target itself [38,39]. Previous DBS (STN, Other Targets)
The implantation of an electrode into the target does not destroy it; therefore it is always possible
1609
1610
96
Subthalamic nucleus stimulation for parkinson’s disease
to reimplant targets when DBS was not successful, while during the selection of the patient all criteria were fulfilled to predict a beneficial effect of DBS. Moreover, it is not even necessary most of the time to remove the inefficient electrode as, by definition, this inefficient electrode is situated elsewhere than in the right target, which is where the new implantation is aimed at. We have reoperated several patients in whom the electrode was not satisfactory or leading to artifacts, and the new electrode was in all cases more efficient and beneficial [40].
Surgical Procedure for implantation of Electrodes into the Subthalamic Nucleus The procedure described in this chapter is the one followed in Grenoble, and is applied to all targets (STN, internal pallidum GPI, thalamic VIM, Pedunculopontine Nucleus PPN, subgenual cortex CG25). This does not pretend to be the ultimate procedure to be followed, but just describes the state of the art of the method in our team. There are obviously other ways to perform it, as reported by other teams, as well as, depending on the future technical improvements, the method might evolve in our own team.
Pre-operative Imaging and Planning [41] Stereotactic Ventriculography Under General Anesthesia
The patient is anesthetized, the head is shaved, and placed on a Talairach modified frame, using four transcranial hollow screws (Sofamor, France), in which four pins held by verniers of the four posts of the frame are inserted. The millimetric values recorded on the verniers will be reused for subsequent repositioning of the patient in an exactly similar position.
The frame is installed within a biorthogonal X-ray setup (distance from the tube to the X-ray film: 3.5 m) and is equipped with an angiography localizer made of, in each X-ray direction, two plastic plates bearing four metallic beads, which will serve as fiducials on the X-ray images. The X-ray images are obtained using flat digital detectors. (for full description, see the chapter on the Talairach system, in this handbook). Twist drill is performed after a small skin incision through the skull, usually on the right side, at 9 cm from the nasion and 2.5 cm from the midline. The right frontal horn of the ventricle is hand free tapped, using a Cushing cannula at a depth of 6.5 cm from the skin surface. A 2-ml air bubble is injected to check on X-rays that the tip of the cannula is correctly placed in the frontal horn. During the injection of 6.5 ml of contrast medium (Iopamiron, Schering), a 12 s sequence of X-ray images is recorded, in the lateral direction, immediately followed by X-rays taken in the anterior posterior direction, usually without any new injection of contrast medium (> Figure 96-2). Stereotactic MRI is performed in the afternoon, when the patient is awake, using a stereotactic head holder MRI compatible with an MRI localizer made of plates containing an N figure tube filled with copper sulfate, the dimensions of which are coherent with the stereotactic frame used for ventriculography and with the angiolocalizer features. The planning is made by merging the images from the ventriculography and the MRI, and constructing the stereotactic target using graphic tools and software subroutines which have been included into the navigation software (> Figure 96-3a, b). On the digital X-ray films of the ventriculography, the selection of two images, clearly showing the ventricular features on which the targeting is based, the AC (anterior commissure) and PC (posterior commissure), and the height of the thalamus, which corresponds to the floor of
Subthalamic nucleus stimulation for parkinson’s disease
96
. Figure 96-2 Pretargeting of STN: Planning on the Ventriculogram (lateral and frontal views) based on AC-PC line, height of the thalamus (floor of the lateral ventricle) and midplane of the third ventricle: an oblique line, passing through the midAC-PC point, and the 10/12 of the tangent to the top of the thalamus, parallel to AC-PC, crosses the floor of the third ventricle (showing the best estimate of the STN target) and the inner table of the skull just ahead of the coronal suture (showing a good estimate of the entry point). On the frontal view, the laterality of the entry point and of the STN target is set at 35 and 12 mm respectively
the lateral ventricle, on the lateral view, and the midline of the third ventricle, as seen on an antero-posterior stereo X-ray image, allows the automatic drawing of the intended target (in the present situation STN, but the coordinates of any other stereotactic target can be introduced in the software) (> Table 96-1). This X-ray target is then imported into the Voxim neuronavigation software of the robotized arm of the Neuromate, along with the MRI images, through the hospital data exchange network. The importation of both X-ray and MRI images obtained in stereotactic condition in the same patients, the same day allows checking the coherence of the data by matching two
anatomical structures, AC and PC, which are clearly identified and visible on the X-ray ventriculographic image and on the T1 and T2 MRI images (> Figure 96-3c). In a large number of cases, these two sets of data correctly match, but in a significant number of cases, there is a discrepancy, which might be non-negligible with up to 2–3 mm in distance between the AC images for instance (> Figure 96-3d). This stresses the non-perfect reliability of the MRI as a localization method, which is a problem which is not yet fully solved. When the matching between the images of the commissures is satisfactory, one checks the matching between the theoretical targeting,
1611
1612
96
Subthalamic nucleus stimulation for parkinson’s disease
. Figure 96-3 Neuronavigation on the Voxim1 software: pretargeting of STN on the lateral (a) and frontal (b) ventriculograms. Projection of the AC and PC ventriculographic points on the lateral T1-weighted MRI midplane images demonstrate the correct match (c) or the mismatch (d) of these anatomically defined features. The pretarget is projected on the T2-weighted axial MRI image where STN is seen as a hypointense signal (e). This comforts the value of the targeting when the AC and PC ventriculographic and MRI images match correctly as in c. The projection of the trajectory on the cortical surfaces allows checking that large vessels (veins and arteries) as well as the sulci (where small size vessels are packed) are not hit by the penetrating tools
which is provided by STN target position obtained with the X-rays software on the ventriculography, and the MRI image of the STN, which is mostly visible on T2 weighted sequences
(> Figure 96-3e). If there is a significant mismatch, particularly in the lateral direction, the laterality is corrected. According to the MRI informations, if the MRI ventriculoraphy coherence as previously
Subthalamic nucleus stimulation for parkinson’s disease
96
. Table 96-1 Coordinates of the STN target STN Target Coordinates Antero-Post 1/12 AC-PC Mean SD
Verticality 1/8 HT
5.19 0.70
1.25 0.70
Antero-Post (mm) 10.91 1.47
Verticality (mm) 2.70 1.51
Laterality (mm)
AC-PC (mm)
HT (mm)
11.57 1.76
24.04 2.68
10.49 1.55
The coordinates are expressed in the bicommissural reference frame (Positive values are ahead of PC, above the level of the AC-PC line, and towards the left side) They are given as proportional relative values in 1/12 of the AC-PC distance for the antero-posterior coordinate, and in 1/8 of the height of the thalamus (HT) for the vertical coordinates. They are also expressed in mm using the average values of AC-PC distance and HT. In all cases the laterality is given in mm. A 1.05 magnification coefficient has been applied to the X-ray measures
described is not satisfactory, then, one keeps the theoretical target as the practical one. The planning also allows choosing the entry point, both avoiding hitting the big vessels (arteries and veins) of the cortical surface but also the sulci where are most of the deep vessels (> Figure 96-3f). The entry point should be chosen also in order to avoid as much as possible the ventricle, but even more the caudate nucleus, as it seems that bilateral injury of the caudate nucleus with the microelectrode cannulas might be related to the postoperative confusion (> Figure 96-4). The planning data are imported into the controller of the robotized arm through the neuronavigation software (> Figure 96-5).
Electrode Implantation Three days after the planning session, the implantation session is performed under local anesthesia. Using the readings of the four verniers, the patient is reinstalled on the stereotactic frame, using the pins, which are reinserted into the hollow screws. X-ray images are taken to ensure that replacement is correct. Using the preplanning data, which have been stored into the neuronavigation software, the robotized arm is launched to reach the position corresponding to the first side to be operated
(in general, the site contralateral to the worst clinical side of the patient). The tool holder of the robotized arm allows to mark the central penetrating trajectory on the skin, which serves to cut an arciform incision, the concavity of which is oriented posterior and medial. The skin, subcutaneous tissue, and periosteum are reclined in block from the skull by a rugination and a 6 or 9 mm in diameter trephination is performed, using a drilling tool held by the tool holder of the robot. The dura is not opened and the electrode guide tool (Ben Gun with five parallel channels, distant by 2 mm axis to axis from each other, accommodating guidetubes as well as the chronic DBS electrodes 1.27 mm in diameter) is introduced down to the dura. The guidetubes are introduced by perforation of the dura matter using sharp stylettes and then lowered into the brain using blunt stylettes. When the stylettes are removed, microelectrodes protected by their own guidetubes (FHC microelectrode) are inserted at the depth corresponding to the target point ( 15 mm), which is controlled by X-ray images. Electrophysiological Exploration
The electrodes can be moved towards the target using the micromanipulator driven by the electrophysiological system (AlphaOmega), the inner tip of the microelectrode (1 m tip diameter, impedance 1–10 MOhms) is connected to the pre-amplifier and then to the electronic stages
1613
1614
96
Subthalamic nucleus stimulation for parkinson’s disease
. Figure 96-4 The ‘‘caudate confusion hypothesis’’: the body of the caudate nucleus may be hit bilaterally by the two electrodes and particularly during the Microrecording phase by the five guide tubes on each side. this might be one cause of the temporary postoperative confusion seen in about 20% of the patients
of the data acquisition and processing system. Two types of electrophysiological investigation are performed: Microrecording of the neuronal activity: The pattern of the electrical activity into the STN nucleus has been reported in several previous papers, by our team as well as by others [42,43]. This pattern is the signature of the target (> Figure 96-6). It is essentially made of asymmetrical spikes at rather high frequency, and typically exhibiting bursting patterns in Parkinson’s disease. These units respond to passive movements of the limb joints and proprioceptive inputs such as muscle pressure, as well as they exhibit synchronous activity when the
patient has tremor on the controlateral side of the microrecording. The length of the recording of such patterns varies between 5 and 6 mm, inserted between two silent zones corresponding to white matter, the first one between 0 and 2–3 mm below the AC PC plane (subthalamic area and anterior zona incerta), the other one between 9 and 10 or 11 mm corresponding to the white matter just above the substantia nigra reticulata (SNR). When the electrode enters the SNR, another type of neuronal activity is typically recorded, made of symmetrical spikes of large amplitude and regular activity and unresponsive to external stimuli in general (> Figure 96-7).
Subthalamic nucleus stimulation for parkinson’s disease
96
. Figure 96-5 Integration of the Xray and MRI modalities into the neuronavigation software of the robotized arm neuromate1
. Figure 96-6 Intraoperative microrecording: Upper line: intraoperative set-up of the patient on the frame, under local anesthesia and of the five channel microdrive (AlphaOmega system) Lower line: Microrecording discharge patterns in the ST and in the SNr (SN)
1615
1616
96
Subthalamic nucleus stimulation for parkinson’s disease
. Figure 96-7 Postop imaging of two permanent STN electrodes (DBS model 3389, Medtronic) and the pulse generator (IPG Kinetra, Medtronic)
Micro stimulation can be performed even using the microelectrode which has been used to record neuronal activity, with current intensities up to 10 mA for short periods, about 10–30 s. Microstimulation is essential as it allows the observation of the beneficial effects inside the target and of the side effects, outside of it [44]. The beneficial effects consist in the improvement of the clinical symptoms of PD. All symptoms, except gait, could be tested, but in practice, the rigidity of the wrist appeared to be the most convenient as it does not require an active participation of the patient, can be scored by experienced neurologists in the OR using a semi-quantitative scale, validated inside the team. Speech and akinesia are difficult to test consistently over the whole duration of the surgery. Tremor is obviously an excellent symptom to use for testing but is often absent in the advanced akine´to-rigid stages of the patients selected for surgery. The side effects depend on the structure surrounding STN which is reached by the spread of current when the electrode is either outside STN or close to its boundaries. Regardless of the frequency, the fibers are excited and then symptoms generated by the activation of these fibers can be observed. Laterally to STN, muscular contraction are induced by the excitation of the cortico-spinal fibers of the internal capsule, most often in the face and the
upper limb, and of the cortico-nuclear fibers induced conjugated binocular deviation towards the side contralateral to stimulation. More posterior, paresthesias are induced in the fibers of the lemniscus medialis. Medially to STN, and deeper, fibers of the oculomotor nerve induce monocular deviation of the eye, towards the midline, or either upward or downward. These side effects must be taken into account to avoid placing the chronic electrode in these places where postoperative side effects would limit the possibilities of efficient stimulation? Post operative imaging and IPG implantation: Three days after electrode implantation, postoperative MRI is made in a very similar manner, than in the preoperative stage: under local anesthesia or even without it, the patient is replaced into the MRI localizer by resetting the readings of the four pins at the previously recorded data. Axial T1 and coronal T2 weighted sequences are performed, without gadolinium injection, and are imported into the neuronavigation software for further control and comparison of the electrodes position as seen on the last X-ray pictures and on the post operative MRI. In addition to the control of the localization, one may again check the accuracy of the MRI and the magnitude of the mismatch by projecting the extremities of the four contact sets of the DBS electrodes onto the
Subthalamic nucleus stimulation for parkinson’s disease
T1 and T2 MRI images, and to observe whether or not they do match correctly. This postoperative control is systematically performed for the STN cases (as well as for the other targets such as VIM, GPI, and more recently, the PPN nucleus), and there was no clinically or radiologically observable side effect of complications during these examinations. This observation is important, as currently for safety reasons, the recommended threshold values of the SAR (surface absorption rate) are lowered, making it more difficult to perform and jeopardizing the quality control of the surgical procedures. At the end of the MRI examination, the four transcranial screws are removed and the skin incisions sutured with non-resorbent material. Five days after electrode implantation, under general anesthesia, the IPG (Kinetra Medtronic double chamber or two single chamber Soletras) are inserted in the subclavicular area, in a subcutaneous pouch. Careful hemostasis must be performed and the skins, closed in two layers after local Rifampycine irrigation, as all skin incisions in the previous procedures have been treated. Care must be taken that no incision crosses or overlays over implanted material. The extension leads from the head to the subclavicular area must be placed sufficiently deep under the skin to prevent adhesion to this subdermal area of the skin. In several instances, the IPG has been implanted under the breast, using a subaxillar skin incision on the lateral border of the mammal gland.
Programming Programming is started by the neurologists during the week following IPG implantation. There are not 64 combinations, but only four to start with. The frequency must be set at 130 Hz, the pulse width at 16 ms, the polarity is positive for the case and is negative for the DBS contact. The four contacts (zero for the deepest, distal contact
96
to three for the more proximal) are subsequently investigated. The voltage is set at 0 V and progressively increased while the neurologist checks the efficiency of stimulation, initially on only one particularly sensitive sign, easy to perform, not implicating patient’s goodwill, which is the rigidity of the wrist as explored by passive manipulation by the neurologists. The degree of improvement compared to the off stimulation period is noted. At the same time, and when the maximal improvement has been reached, one looks at the possible appearance of side effects such as paresthesias, dyskinesias, eye deviation, usually monocular and ipsilateral to stimulation, witnessing the involvement of the third nerve fibers, oculomotor nerve, paresthesias due to diffusion to the lemniscus medialis, or muscular contraction (most often at the level of the face or of the arm, sometimes at the level of the lower limb). This strategy is repeated for the four contacts 0, 1, 2 and 3. At the end of this systematic exploration, one or sometimes two, contacts appear to be the best (the highest threshold for side effects and the lowest threshold for beneficial improvement of symptoms). The typical setting is voltage 2–3.5 V, 130 Hz, 60 ms, case positive. Using this strategy, one cannot miss a good setting. Conversely, by playing with all possible combinations of contacts and parameters, one cannot transform a bad response into success as well as one cannot transform a success into failure. Quite often, trying to reach quickly the best setting might result in the induction of paresthesias of dyskinesias, very similar to the levodopa induced dyskinesias. They tend to amend in the evolution and particularly their induction threshold increases along time. When the best contact has been selected, it is convenient to increase the voltage progressively over several days to avoid the dyskinesias. At the same time, the medication, which has been already significantly reduced before the operation, is further decreased in accordance with the clinical
1617
1618
96
Subthalamic nucleus stimulation for parkinson’s disease
improvement, and set as a compromise value, low enough to prevent the appearance of dyskinesias, and high enough to prevent apathy and hypophonia.
Results Since the first application in 1993, more than 1,000 patients have been implanted around the world and several papers have reported clinical results. The following data will synthesize in a summarized manner the results coming from three main publications, reporting short term results at 3 and 6 months of a multicenter double-blind study [27], long term results at 1, 3, and 5 years of one group’s follow up [26], and a consensus meta- analysis of the literature [45]. In addition complications have been also evaluated on the complete series of 325 STN patients in our institution, in a retrospective, unblinded study (Unpublished data).
Complications and Side Effects Among 512 cases of bilateral implantations of various targets operated in our institution since 1987 (thalamus VIM, pallidal GPi, subthalamic nucleus STN, etc), 325 consecutive bilateral STN cases operated since 1993, (97.5% bilaterally operated patients = 641 implanted sides) have been reviewed. All complications and adverse effects (AE) have been considered, regardless of their severity, in an attempt to be exhaustive and in order to avoid the distortion of the evaluation of the complications, which could result from forgetting or neglecting some complications, due to the lack of a precise definition. Therefore, complications and AE have been classified in 3 quantitative categories: Benign: not symptomatic (e.g., only MRI visible), no need for a prolonged stay, no need for reoperation of any kind.
Significant: symptomatic, need for prolonged stay or reoperation, but no permanent deficit or sequelae. Severe: symptomatic, leading to reoperation, hardware revision or ablation, or to a specific treatment, and responsible for permanent deficit or sequelae. Each category has been quantitatively evaluated. An important parameter, related to the overall safety of the method, is the number of patients without any complication, as previously defined. Complications are also considered from the qualitative point of view, distributed among hemorrhages, infections, neurocognition. Specific categories are also related to targeting (ventriculography, frame setting), electrode implantation, and hardware implantation. 57.5% of the STN patients had no adverse effect of any kind, 42.5% had (at least one) adverse effects: 19.8% were severe, 32.5% were significant, and 47.7% were benign. Comparing the occurrence of the severe and significant adverse effects between the main targets, they are significantly more frequent in STN (22.2%/325 patients) than in VIM (8.6%/138 patients) and in GPi (7.9%/63 patients). Considering the surgical phases, 7.4% are related to ventriculography and frame setting, 20.9% to electrode implantation, 13.5% to IPG and hardware, and 31.1% to stimulation. From this point of view, this is again mostly in STN: the reason might be that most STN patients have a long history of Parkinson’s disease; they are older, in a severe condition. VIM stimulation was performed in younger patients, with essentially tremor, and therefore in a better condition than the patients who have akinesia and rigidity, at a more advanced stage of the disease. GPi stimulation involves also young patients with dystonia, but also the situation of the internal pallidum is less critical, surrounded by structures less sensitive than STN is. Infections are mostly superficial, hardware related and happen in 4.4% of the cases, 1.1%
Subthalamic nucleus stimulation for parkinson’s disease
are severe, 1.3% are significant, and 1.9% are mild or benign. Hemorrhages occur in 8.4% of the cases, mostly at the entry point or subcortical, rarely in the target. Of these hemorrhages 95.2% were either asymptomatic or transiently symptomatic, only 6.8% of them were responsible for permanent symptoms. The asymptomatic hemorrhages (42.1%) were seen only on the MRI, which enhances the importance of preoperative MRI. This stresses the importance of MRI, in the preoperative stage to avoid superficial vessels and penetration in a sulcus, as well as the ventricle and the caudate nucleus (which bilateral traumatism seems to be related to the postoperative confusion), as well as in the postoperative stage, where asymptomatic bleedings (42.9%) are only seen on MRI. They are asymptomatic in 3.4%, symptoms are transient in 4.4%, and permanent in only 0.6%. The depression and suicidality were multifactorial, and one must keep in mind that the treatment which is evaluated is purely medical before surgery and is a combination of highfrequency stimulation and drugs after surgery. There was 10.3% of transient post operative confusion, possibly related to the bilateral traumatism of the caudate nucleus by the guidetubes during the electrophysiological exploration, 12.4% of the patient exhibited postoperative depression, lasting for several weeks to several months,always recessive either spontaneously or using antidepressant drugs, there were 1.3% of suicide attempts, and only one patient (0.2%) committed suicide. These neuropsychological and behavioral complications occurred in total in 24.3% of the cases and depression is related to societal issues. There is clearly an evolution along time of the complications: over 300 bilateral STN cases, the number of complications free patients were the 27.3% in the first 150 cases and 72.7% in the last 150 cases. The reason for this improvement in the complication rate is unclear: is that
96
an improvement of technique? The fact was that the screws used for the replacement of the patient were redesigned; the MRI was more systematically used to determine the entry point in the second half of those patients. Could it be also that there was a less careful record of those effects, although this seems improbable? One has also to take into account that in the postoperative period, the treatment is a combination of high-frequency stimulation and drugs, and the respective parts of those two components to the side effects has to be further evaluated. One must also consider the complications from the point of view of the intention to treat, which means that complication happened during the selection or pre-implantation phases must be taken into account when the patient were finally not operated. This accounted for 2.6% of complication in 540 cases including all indications and all targets. They were mainly hemorrhages, coagulation problems, cognitive breakdown, during the pre-targeting phase. In a five year follow-up [26] three patients died (one, 3 years after a post operative intracerebral hemorrhage, which was evacuated, but left the patient bedridden, one myocardial infarction 11 months after surgery, and one suicide 6 months after surgery). Transient, post operative confusion occurred in 24% of the patients during the first few days postoperative (from temporospatial disorientation to psychosis). Device related complications were rare, and one patient required temporary removal of the extension lead and pulse generator, due to infection. Thirty-one percent of patients had eye lid opening apraxia in the first 3 months, which remained a problem in 19% at 5 years. Stimulation induced dyskinesias could be observed and was reversible by decreasing the voltage. 41 of 42 patients gained weight (mean 3 kg, maximum 5 kg). Transient hypomania developed in 8%, and transient depressive episodes with longer follow-up were observed in 17%, and transient apathy in 5%, responding to medication. Apathy did not respond
1619
1620
96
Subthalamic nucleus stimulation for parkinson’s disease
to dopaminergic treatment in 12% and dementia appeared in three patients (7%) between the third and the fifth year. In a 6 month multi-cellular study [27] serious adverse events were more common with neurostimulation than with medication alone (12.8% vs. 3.8%, P < 0.04) and included a fatal intracerebral hemorrhage. The overall frequency of adverse events was higher in the medication group (64% vs. 50%, P = 0.08). A total of 173 adverse events were reported in 89 patients: 39 (50.0%) in the neurostimulation group and 50 (64.1%) in the medication group (P = 0.08). Most adverse events were well-known medical problems associated with advanced Parkinson’s disease.
physical summary score of the SF-36, a generic quality-of-life scale. The mean UPDRS-III score improved by 41% in the off medication state (from 48.0 12.3 at baseline to 28.3 14.7 at 6 months) and by 23% in the on medication state, (from 18.9 9.3 at baseline to 14.6 8.5 at 6 months). UPDRS-II, markedly improved (by 39%). The mean PDQ-39 summary index score improvement was 23.9% (41.8 13.9 at baseline and 31.8 16.3 at 6 months), the dyskinesia scale obtained while the patient was not taking medication improved from 6.7 5.3 to 3.1 3.5 (54%). The dopaminergic equivalents were reduced by 50%. The amplitude of stimulation was 2.9 0.6 V, the frequency, 139 18 Hz; and the pulse duration, 63 7.7 ms.
Clinical Results Clinical Improvement at Six Months [27]
Clinical Improvements at Five Years [26]
In an unblinded randomized-pairs trial on 156 patients under 75 years of age with advanced Parkinson’s disease and severe motor symptoms, neurostimulation of the subthalamic nucleus plus medication was compared to medical management [27]. The primary end points were the changes from baseline to 6 months in the quality of life, (Parkinson’s Disease Questionnaire PDQ39), and the severity of symptoms without medication, (Unified Parkinson’s Disease Rating Scale, part III UPDRS-III). Pairwise comparisons from baseline to six months showed that neurostimulation, as compared with medication alone, caused greater improvements in the PDQ-39 (50 of 78 pairs, P = 0.02) and the UPDRS-III (55 of 78, P < 0.001), with mean improvements of 9.5 and 19.6 points, respectively. Neurostimulation resulted in improvements of 24–38% in the PDQ-39 subscales for mobility, activities of daily living, emotional well-being, stigma, and bodily discomfort and 22% improvement in the
Off Medication Evaluation in the Off Medication State
STN stimulation improved the total score of UPDRS part III, as compared to the baseline value (55.7 11.9), by 66% at 1 year, 59% at 3 years and 54% at 5 years. At five years, the improvement compared to baseline was 75% for tremor, 71% for rigidity, and 49% for akinesia. Postural stability and gait also improved, but speech improved only during the first year and then progressively returned to the baseline at five years. The total score of UPDRS part II, as compared to baseline value (30.4 6.6%), improved by 66% at 1 year, 51% at 3 years, and 49% at 5 years, the worsening between 1 and 5 years was significant (p < 0.001). The improvement was dramatic, postoperatively in the off medication condition, at five years. Most patients were independent in the activities of daily living in the off medication condition (mean score on Schwab and England’s scale 73%), whereas before surgery most had
Subthalamic nucleus stimulation for parkinson’s disease
been fully dependent on a caregiver (Schwab and England’s scale 33%). Before surgery, off medication painful dystonia was observed in 71% of the patients while it was present only in 19%, at 1 year and 33% at 5 years. On Medication Evaluation
Motor function and activities of daily living in the on medication state did not improve after the stimulation of the STN, there was no significant changes for tremor and rigidity but scores for akinesia, speech, postural stability, and freezing of gait worsened (P < 0.001), which resulted in the worsening of the total score of UPDRS part III and part II. Schwab and England score was unchanged, and compared to the baseline, the disability related to dyskinesia decreased by 58% and their duration by 71%. 29 of 42 patients (69%) had levodopa induced dyskinesias before surgery, and only four at three months and two at five years. Neuropsychological evaluation: there were no significant changes on the Beck Depression inventory, the Mattis dementia rating scale was worse at five years reflecting progressive dementia in three patients, but the changes were not significant (131 18 vs. 136 10 at baseline, p = 0.07). The average score for frontal lobe function was slightly worse at five years. (37.3 11.2 vs. 40.4 9.2, p = 0.03). Medications and Stimulation Settings
The levodopa equivalent daily dose decreased from 1409 605 mg at baseline to 584 366 mg at 1 year, 526 328 mg at 3 years, and 518 333 mg at 5 years (p < 0.001). At 5 years, 11 or 42 patients (26%) were no longer taking levodopa and three were not taking any dopaminergic drugs. The stimulation settings after 1 year (2.8 1.6 V, 143 19 Hz, 61 6 ms) were not significantly different at five years (3.1 0.4 V, 145 19 Hz, 64 12 ms). Monopolar stimulation was used in 90% of patients at one and five years, only one patient required replacement of the stimulators within the first five years.
96
Discussion Costs They are related to hardware at the time of implantation and replacement [46]: In 49 pts bilaterally implanted with Itrel II, continuous stimulation with usual parameters (amplitude: 3.2 0.3 V, pulse width: 65 10 ms, frequency:145 16 Hz) the replacement had to be anticipated in 25% due to unilateral depletion, and the average duration of life of the IPGs was 83 14(40–113) months. The electrodes, the extensions, as well as the IPG (implantable programmable generator), must be purchased altogether at the time of implantation. This actually corresponds to an amount of several thousand dollars (which for the moment does not change even if new companies are challenging the previous monopole), which must be paid at the time of implantation. However, if one compares between implanted patients and patients treated with the best medical treatment, [47], it is clearly shown that the cost of the hardware and all of the expenses related to the surgery, including hospitalization, is lower than the cost of medication, caregivers, accessories such as wheelchair, etc, during a period equivalent to the duration of life of the IPG. The problem however remains as the medical costs are paid progressively along time when drugs are needed. A different regimen of payment, such as leasing, could make the burden to the paying organizations (Social Security, insurances, etc.) similar for the two forms of treatment.
Surgical Complications, Side Effects and Non-surgical Complications Why is the morbidity of STN surgery different from GPI or Vim, for instance? Is it the duration of surgery? But STN and GPI procedures are
1621
1622
96
Subthalamic nucleus stimulation for parkinson’s disease
equivalently long. Is it the number of passes? But in STN as well as in GPI, both targets are targeted using the Ben Gun system with five parallel microelectrodes. The difference might be due more in terms of entry point and target itself, and also of age and disease. There are biases which are the evolution of the methods, the careful record of adverse effect along time, and the multiplicity of operators. Implantation related complications could be avoided using a very careful planning on MRI (provided that there is no distortion and mismatch between MRI informations and reality), to avoid the vessels, the sulci (in which a high density of vessels are contained), the caudate nucleus (where preliminary data suggest that bilateral traumatism of these nuclei by the exploring tracks might be related to the post operative confusion), and finally the ventricle (where some vessels, such as the thalamo-caudate vein, maybe hit when the guidetubes penetrate the ventricle or ependyma). Also, the reentry of the cannula and electrodes into the parenchyma at the level of ependyma might be a cause of deviation of electrodes. Targeting on MRI, to be reliable, assumes that the brain is in the same situation than it was at the time of the acquisition of the images. This is definitely not the case when the dura is opened, which might cause a loss of CSF, which is sometimes significant. This is the reason why we do not open the dura, even using five tracks for each side, the guidetubes puncturing the dura and therefore preventing the CSF leakage, Opening the dura is performed most of the time to observe cortical vessels, and spare them or allows coagulation. This is actually a false security, as the vessels situated in deeper layers, and particularly those contained in the sulci, which are almost parallel to the electrode track, are not visible, even when the dura is opened. A permanent improvement of the design of the different parts of the hardware is necessary and must be pursued by the companies under the advices of the surgeons: for instance, the
shape and size of the transcranial screws used in our procedure have been modified, and are better long than short in order to avoid that the scalp covers them and creates inflammation and an opportunity for external to internal infection. Low profile extensions are currently provided, and we may expect new perspectives in nanotechnologies, leading to miniaturized systems. Duration of surgery, microelectrode recording, the number of passes, have been poorly related to the clinical outcome and to the complication rate, although several publications have addressed this issue [48,49]. The neurocognitive complications are multifactorial. A careful inventory of the pre-operative neuropsychiatric episodes (depression, delusion, suicide attempts or ideas, etc.), must be done and may help eliminating patients who could be predicted to significantly impair due to the various events, which are classically associated to the term of operative shock. It has become clear during the recent years, that societal changes occurring after the surgical period may be responsible for depression and even suicide, because the patient cannot cope with the important life, social, familial, professional, changes to which he has not been prepared. A more careful and serious preoperative psychological preparation should probably help those patients to readapt more easily in their new life. These changes are not specifically related to the target [50] but are mainly related to the magnitude of the improvement, which is why they are not so clearly reported after ablative surgery, where, in particular due to more often unilateral procedures, the results are not so visible. Depression and suicide [51] have been also observed in all the major surgeries, providing the patient with nearly significant improvements, such as breast plastic surgery, as well as the recovery of freedom for prisoners who spent a rather long time in jail [52–55]. The complication related to stimulation can be improved by a precise positioning of the electrode avoiding diffusion to the neighboring
Subthalamic nucleus stimulation for parkinson’s disease
structures which are responsible for adverse effects,such as muscular contractions due to spreading of current to the pyramidal tract, [56] which may be limiting factors, and therefore preventing the optimal use of the stimulation benefits. Clear recognition of the specific role of stimulation, combined with medication, can explain some neurocognitive outcomes: for instance, speech might be impaired when dysarthria is induced by stimulation of the cortico-nuclear fibers in the internal capsule leading to the undesired contraction of the laryngeal muscles, but also hypophonia might be related to the strong diminution of medication, as it is clear that even if STN stimulation improves speech [36], it cannot compensate the effect of medication, which must be therefore re-increased. Similarly dyskinesias induced by stimulation are mostly due to an overdose of treatment and, therefore can be alleviated by decreasing either the drug dosage or the stimulation parameters, particularly the voltage.
Mechanism of Action of DBS at High Frequency High frequency stimulation (HFS) mimics the effects of ablative lesions in all the targets which have been used so far (thalamus, pallidum, subthalamic nucleus, accumbens nucleus, hypothalamus, etc.). However, HFS does not create a lesion, as all symptoms created by stimulation or all improvements are only temporarily contemporary to the duration of the stimulation. As another difference to lesioning, HFS probably induces specific effects by excitation or inhibition of structures at distance, such as the motor or sensory side effects obtained by diffusion to the motor pyramidal track or to the lemniscus medialis fibers. A large series of papers have been published, dealing with the mechanism of action of HFS, and there is an important movement of basic research aimed at this goal. The complete mode of action is still
96
currently globally unknown, but one can consider that it is the combination of a rather large number of sub mechanisms [57]. A jamming of the neuronal message transiting through the stimulated structure is highly probable as we have suggested that it might happen in the very early stages of the procedure [57,58]. This could be coherent within the observation that pathological states such as Parkinson’s disease create synchronization of neuronal activities, which are expressed by bursting patterns visible during the microrecording, which was shown in the pallidum in monkeys [9]. This is supported by several experimental data obtained in monkeys showing that STN HFS resets subthalamic firing and reduces abnormal oscillations of STN neurons [59]. The hypothesis of extinction or strong inhibition of neuronal firing is also supported by direct observation of the decrease in discharge rate during stimulation [60–66] as well as dual effects, associating excitation and induction of high-frequency bursts. Finally, it has been shown in our laboratory that HFS inhibits the production and/or the release of certain neurotransmitters and hormones in cultures, suggesting that this phenomenon could partcipate to the functional inhibition induced by HFS [62].
Neuroprotection It has been demonstrated that in parkinsonian patients as well as in animal models, the neuronal activity of the STN nucleus is profoundly altered, associating the appearance of a rhythmic pattern made of burst, in addition to a general increase in firing rate. STN neurons are glutamatergic and glutamate is an excitatory amino acid which has excitotoxic effects on the dopaminergic neurons of the projection area of the STN. The increased glutamate output of STN on the dopaminergic neurons of the Substantia Nigra pars compacta participate to their degeneration, which lead to the idea that decreasing output by antagonists such as MK 801 would slow down the
1623
1624
96
Subthalamic nucleus stimulation for parkinson’s disease
degeneration process involving the dopaminergic neurons [63]. As one of the mechanisms of HFS might be a decrease in the firing rate of neurons submitted to this type of stimulation, one can suppose that this, (as well as the ablation of this structure) would play a similar role and would be slowing down the neurodegenerative process involving the nigral dopaminergic system. In order to evaluate this hypothesis, we have studied the effects of STN lesions in rats receiving unilateral injection of 6 hydroxydopamine (6-OHDA, a specific neurotoxin of dopaminergic neurons) into the striatum ipsilateral to the STN lesion [64,65]. The cell loss of dopaminergic neurons was significantly reduced as compared to rats in which the STN nucleus had not been destroyed. These data were confirmed by other studies [59,61,66] using the same 6-OHDA toxin or another one such as 3NP [67]. Experiments in rats using HFS of STN instead of ablation [68,69] led to similar results, confirming at the same time that lesion and HFS were equivalent, as we had assumed it in the chapter of the mechanisms. A large study in MPTP monkeys [70] rendered parkinsonian by systemic injection of the neurotoxin MPTP and receiving unilaterally, either a kainic acid lesion of the STN, or chronic, high-frequency stimulation, confirmed also that in both paradigms, there was a significantly increased survival of these neurons, in the substantia nigra compacta ipsilateral to the STN lesioned side. There was no difference in survival or loss of dopaminergic neurons between the two Substantia Nigra compacta when the STN nucleus was left intact or sham operated. In human patients, the only study using PET scan did not confirm these experimental data, but it was performed in patients at a too advanced stage of the disease [71]. In one unpublished series of our patients, we evaluated the evolution of the off-off UPDRS score at various times as compared to the preoperative stage.
Three groups of patients could be isolated, the first one (25%) was continuing to impair along time, as it is classical in idiopathic Parkinson’s disease. More surprisingly, 36% of them had a UPDRS score which was maintained within 15% of variation, which is considered as the range of non-significant variation of this scale. A third group of 38% of the patients did significantly improve their UPDRS at one year as compared to the preoperative stage, and half of them kept improving over five years. These data, from animal experiments altogether with these clinical data, would tend to suggest that HFS could be able to alter the natural history of the neurodegeneration of the dopaminergic system, but so far, there was no relationship which could be established between those three groups, and causes or factors, which would characterize them. One may however consider that those responders were having a different susceptibility or a slower rate of degeneration, allowing HFS of STN, not only slowing down the neurodegenerescence, but also allowing the dopaminergic neurons, which had lost their dopaminergic production, but were still alive [72] could recover sufficiently to produce again dopamine, which would explain why unexpectedly and apparently paradoxically these patients would not only stabilize but even improve. Before this could be considered as demonstrated, there is an urgent need for controlled randomized clinical trials of patients, investigated using both clinical and non clinical (such as PETscan) tests and operated early enough. This raises several problems, as at the beginning of the disease, the risk of implanting patients who would not have idiopathic PD is increased. In addition, differently from the animal experiments, the patients, already at the time of diagnosis, would have a loss of dopaminergic cells, which is around 60–70%. This would also raise problems about the cost of the management and about the management itself of Parkinson’s disease, which would be profoundly changed.
Subthalamic nucleus stimulation for parkinson’s disease
Perspectives
96
being performed, the results of which are not reported at the present time.
Alternatives to STN HFS Duodopa Gene Therapy
Based on the assumption that the hyperactive STN would deliver an increased amount of the glutamate excitotoxin to the nigral dopaminergic cells, a experiment using AAV GAD gene therapy has been performed in rats [73], demonstrating that the glutamatergic STN nucleus, could be transformed into a GABAergic structure. This has led to a similar study targeting the STN of parkinsonian patients. Preliminary data [74] have reported a clinical improvement as well as PET scan based evidence of metabolic improvement. Recent data have been reported concerning the transcription of Nurturin into the human striatum of PD patients with also preliminary encouraging data [75]. Growth Factors Infusion
Based on experimental data in rats and monkeys showing the therapeutic effects of the glial derived nerve factor (GDNF) on animal models of parkinsonism, chronic infusion of GDNF using an implanted pump through catheters inserted into the striatum has been performed in several patients and highly significant improvement was reported [76–78], which was not confirmed by an international multicenter double-blind controlled study [79]. Criticisms have been made about the methodology used in the multicenter study, which might not have completely replicated the design of the princeps study. Cortical Stimulation
Experiments in monkeys of chronic cortical stimulation of the motor area [80] have provided an improvement of the experimental parkinsonian symptoms in MPTP monkeys with a concomitant reduction in the firing rate of STN neurons. Human clinical trials are
Levodopa induced dyskinesias are the principal drawback of medical therapy, and several attempts are being made to suppress this complication. Dyskinesias are considered to be the result of a loss of an optimal response by the striatal dopaminergic receptors, induced by pulsatile administration of levodopa. Attempts have been made to produce a more stable and regular dopamine concentration in the brain by continuous infusion of dopamine agonists such as Apomorphine [81] or Lisuride [82]. Although these methods clearly improved the treatment by decreasing the dyskinesias, local complications appeared such as cutaneous nodules at the level of the needle insertion on the abdominal wall and restrict the use of this method. Currently, another route of continuous administration is being tested, which uses a duodeno-gastrostomy with an intra-gastric catheter. Despite the invasiveness of this method, and the discomfort which is imposed upon the patients, satisfactory results are being reported [83]. Grafting Methods (Mesencephalic Fetal Cells, Stem Cells, Retinal Epithelial Pigmented Cells Encapsulated Cells, etc.)
Since several decades, an impressive amount of work in basic science and in animal models have been performed in highly expert labs around the world as well as several transfers of the method to parkinsonian patients in controlled trials. Encouraging, but always partial results have been reported in several occasions, showing clinical improvement, evidence of dopaminergic reinnervation of the striatum, good survival of the grafted neurons, efficient production of dopamine and increased tyrosine hydroxylase immunoreactivity. Various types of cells have been used (adrenal gland, mesencephalic fetal grafts,
1625
1626
96
Subthalamic nucleus stimulation for parkinson’s disease
and more recently, epithelial retinal cells). Stem cells are also being investigated as a potential material, which could be much more immunologically tolerated, but raising their own problems. Despite that this approach is ultimately the most elegant, it is still experimental and cannot be proposed to patients as a part of the therapeutic panel currently available [84].
tested. Moreover, the paradigm of stimulation at high frequency might be improved and there are reports [85] of smart combinations of pulses, such as phase resetting pulses, which should be more efficient, and also needing less energy, therefore, saving the batteries and decreasing the costs.
Physiology Improvements New Medications: Levodopa Without Dyskinesias
The pharmaceutical companies are working hard on designing dopaminergic agonists, which would have the beneficial effects of levodopa, which remains a miracle drug during the honeymoon of its utilization, without its major complication, which are the dyskinesias. It is clear that if such a molecule could be designed, this would make most of the surgical approaches, including HFS on various targets, useless.
Hardware Improvements Hardware is currently at the first generation, and has achieved proof of concept that is far from being optimal. There is an urgent need for miniaturized stimulators, allowing implantation at the level of the head, with a very short extension and miniature connectors to prevent the skin complications. Rechargeable batteries should be, and actually are being, developed, which would suppress the need for replacement and would decrease the costs. New electrode designs are being expected from the development of nanotechnologies, allowing a more precise targeting of the stereotactic structures.
The lower morbidity of high-frequency stimulation of the brain structures has allowed to investigate the effects of its application to targets suggested by the results of basic research. This was the case of the subthalamic nucleus for Parkinson’s disease, for the accumbens nucleus for psychosurgery, of the posterior hypothalamus for cluster headaches. The subthalamic nucleus, although it represents the current ideal therapeutic target for Parkinson’s disease at advanced stages, is however not efficient on the whole spectrum of symptoms, particularly the gait problems unresponsive to levodopa, such as the freezing of gait [35]. From the basic studies of the group of Oxford [86,87] showing that the pedunculopontine nucleus, which is degenerating in Parkinson’s disease could improve the gait and the motricity of the lower limbs in monkeys when stimulated at low, and not at high, frequency. This has led already to several clinical trials, the preliminary results of which are confirming the basic science assumptions and providing preliminary data very encouraging. Additional electrodes connected to an IPG at low-frequency can be added into the PPN of parkinsonian patient already successfully stimulated with bilateral STN electrodes, improving the triad of symptoms, but not controlling their freezing of gait.
Software Improvements There is no reason that the current waveform, which is basically a square pulse, biphasic and asymmetrical to equilibrate the electrical charges, might be the best one. New waveforms must be
Conclusion The subthalamic nucleus is currently the first choice target for treatment of advanced
Subthalamic nucleus stimulation for parkinson’s disease
Parkinson’s disease cases, although a good double-blind controlled study to precisely establish the respective values of the internal pallidum and of STN is still lacking. This prominent position of STN as the most favored target is due to: Its strategic anatomo functional situation, which makes it a central station, where several circuits and networks intercross and interact. Its MRI visibility. Its typical microelectrode recording pattern. Its relatively small size. These advantages are also counteracted by drawbacks for the same reasons. Small size: the beneficial motor effects are associated to side effects due to the simultaneous involvement of other functions, such as mood in particular. Easy spreading of the current to internal capsule (motor side effect), lemniscus medialis (sensitive side effects), oculomotor nerve fibers (ocular deviation), hypothalamus (behavioral changes). The clinical effects are depending upon careful patient selection, accuracy of placement of the electrodes, delicate balance between medical treatment and stimulation. Complications seem more frequent than in the thalamus or in the pallidum, because again of its small size and situation, which would call for a better electrode design. Despite the general agreement that STN is the best target and that STN stimulation is the most efficient surgical treatment for advanced Parkinson’s disease cases, the best reported improvement ratios (about 70%), could be increased, the coverage of symptoms could be optimized and the side effects could be reduced. This calls for improvements at several technological levels, including progresses in MRI visualization, correction of the MRI distortion, improved electrophysiological methods to delineate the functional target (such as online more sophisticated data
96
processing), new design of electrodes aiming at the three-dimensional steerable stimulated volume, new paradigms and waveforms pulses, multiple targets simultaneously implanted, optimized hardware design to decrease the extracerebral complications. Keeping in mind that deep brain stimulation techniques are not aimed at becoming the unique treatment of neurodegenerative diseases, but are indicated in advanced cases, unless their neuroprotective effect would be demonstrated, which is not currently the case, one should resist the temptation of making these techniques too fast and expedite at the expense of the patients’ benefit, which remains the only goal of these approaches.
References 1. Bergman H, Wichmann T, DeLong MR. Reversal of experimental parkinsonism by lesions of the subthalamic nucleus. Science 1990;249:1436-8. 2. Aziz TZ, Peggs D, Sambrook MA, Crossman AR. Lesion of the subthalamic nucleus for the alleviation of 1-methyl4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-induced parkinsonism in the primate. Mov Dis 1991;6:288-92. 3. Aebischer P, Goddard M. Treating Parkinson’s disease with lesions of the subthalamic nucleus. Science 1991;252:133-4. 4. Benazzouz A, Gross C, Feger J, Boraud T, Bioulac B. Reversal of rigidity and improvement in motor performance by subthalamic high-frequency stimulation in MPTP-treated monkeys. Eur J Neurosci 1993;5:382-9. 5. Benabid AL, Pollak P, Gross C, Hoffmann D, Benazzouz A, Gao DM, Laurent A, Gentil M, Perret J. Acute and long-term effects of subthalamic nucleus stimulation in Parkinson’s disease. Stereotact Funct Neurosurg 1994;62:76-84. 6. Limousin P, Pollak P, Benazzouz A, Hoffmann D, Le Bas JF, Broussole E, Perret JE, Benabid AL. Effect of parkinsonian signs and symptoms of bilateral subthalamic nucleus stimulation. Lancet 1995;345:91-5. 7. Limousin P, Krack P, Pollak P, Benazzouz A, Ardouin C, Hoffmann D, Benabid AL. Electrical stimulation of the subthalamic nucleus in advanced Parkinson’s disease. N Engl J Med 1998;339:1105-11. 8. Krack P, Limousin-Dowsey P, Benabid AL, Pollak P. Chronic stimulation of subthalamic nucleus improves levodopa-induced dyskinesias in Parkinson’s disease. Lancet 1997;350:1676.
1627
1628
96
Subthalamic nucleus stimulation for parkinson’s disease
9. Bergman H, Wichmann T, Karmon B, DeLong MR. The primate subthalamic nucleus. II. Neuronal activity in the MPTP model of parkinsonism. J Neurophysiol 1994;72:507-20. 10. Parent A, Smith Y. Organization of efferent projections of the subthalamic nucleus in the squirrel monkey as revealed by retrograde labeling methods. Brain Res 1987;436:296-310. 11. Parent A. Extrinsic connections of the basal ganglia. Trends Neurosci 1990;13:254-8. 12. Parent A. Carpenters human neuroanatomy. 9th ed. Philadelphia, PA: Williams and Wilkins; 1996. 13. Blandini F. The role of the subthalamic nucleus in the pathophysiology of Parkinson’s disease. Funct Neurol 2001;16:99-106. 14. Nambu A, Tokuno H, Takada M. Functional significance of the cortico-subthalamo-pallidal ‘hyperdirect’ pathway. Neurosci Res 2002;43:111-7. 15. Groenewegen HJ. The basal ganglia and motor control. Neural Plast 2003;10:107-20. 16. Temel Y, Blokland A, Steinbusch HWM, Visser-Vandewalle V. The functional role of the subthalamic nucleus in cognitive and limbic circuits. Prog Neurobiol 2005;76:393-413. 17. Bezard E, Boraud T, Bioulac B, Gross CE. Involvement of the subthalamic nucleus in glutamatergic compensatory mechanisms. Eur J Neurosci 1999;11:2167-70. 18. Hammond C, Shibazaki T, Rouzaire-Dubois B. Branched output neurons of the rat subthalamic nucleus: electrophysiological study of the synaptic effects on identified cells in the two main target nuclei, the entopeduncular nucleus and the substantia nigra. Neurosci 1983;9:511-20. 19. Benazzouz A, Gao D, Ni Z, Benabid AL. High frequency stimulation of the STN influences the activity of dopamine neurons in the rat. Neuroreport 2000;11:1593-6. 20. Meissner W, Harnack D, Paul G, Reum T, Sohr R, Morgenstern R, Kupsch A. Deep brain stimulation of subthalamic neurons increases striatal dopamine metabolism and induces contralateral circling in freely moving 6-hydroxydopamine-lesioned rats. Neurosci Lett 2002;328:105-8. 21. Breit S, Lessmann L, Unterbrink D, Popa RC, Gasser T, Schulz JB. Lesion of the pedunculopontine nucleus reverses hyperactivity of the subthalamic nucleus and substantia nigra pars reticulata in a 6-hydroxydopamine rat model. Eur J Neurosci 2006;24:2275-82. 22. Pahapill PA, Lozano AM. The pedunculopontine nucleus and Parkinson disease. Brain 2000;123:1767-83. 23. Ni ZG, Bouali-Benazzouz R, Gao DM, Benabid AL, Benazzouz A. Time-course of changes in firing rates and firing patterns of subthalamic nucleus neuronal activity after 6-OHDA-induced dopamine depletion in rats. Brain Res 2001;899:142-7. 24. Ashkan K, Wallace B, Bell BA, Benabid AL. Deep brain stimulation of the subthalamic nucleus in Parkinson’s
25. 26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
disease 1993–2003: where are we 10 years on? Br J Neurosurg 2004;18:19-34. Benabid AL, Benazzouz A, Pollak P. Mechanisms of deep brain stimulation. Mov Disord 2002;17 Suppl 3:S73-4. Krack P, Batir A, Van Blercom N, Chabardes S, Fraix V, Ardouin C, Koudsie A, Dowsey-Limousin P, Benazzouz A, Le Bas JF, Benabid AL, Pollak P. Five-year follow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinson’s disease. N Engl J Med 2003;349:1925-34. Deuschl G, Schade-Brittinger C, Krack P, Volkmann J, Scha¨fer H, Bo¨tzel K, Daniels C, Deutschla¨nder A, Dillmann U, Eisner W, Gruber D, Hamel W, Herzog J, Hilker R, Klebe S, Kloss M, Koy J, Krause M, Kupsch A, Lorenz D, Lorenzl S, Mehdorn HM, Moringlane JR, Oertel W, Pinsker MO, Reichmann H, Reuss A, Schneider GH, Schnitzler A, Steude U, Sturm V, Timmermann L, Tronnier V, Trottenberg T, Wojtecki L, Wolf E, Poewe W, Voges J. A randomized trial of deep-brain stimulation for Parkinson’s disease. N Engl J Med 2006;355:896-908 (Erratum in: N Engl J Med. 2006;355:1289). Lang AE., Houeto JL, Krack P, Kubu C, Lyons KE, Moro E, Ondo W, Pahwa R, Poewe W, Tro¨ster AI, Uitti R, Voon V. Deep brain stimulation: preoperative issues. Mov Dis 2006;21:S171-96. Charles PD, Van Blercom N, Krack P, Lee SL, Xie J, Besson G, Benabid AL, Pollak P. Predictors of effective bilateral subthalamic nucleus stimulation for PD. Neurology 2002;59:932-4. Fraix V, Pollak P, Van Blercom N, Xie J, Krack P, Koudsie A, Benabid AL. Effect of subthalamic nucleus stimulation on levodopa-induced dyskinesia in Parkinson’s disease. Neurology 2000;55:1921-3. Obeso JA, Grandas F, Herrero MT, Horowsdi R. The role of pulsatile versus continuous dopamine receptor stimulation for functional recovery in Parkinsons disease. Eur J Neurosci 1994;6:889-97. Mazzone P, Lozano A, Stanzione P, Galati S, Scarnati E, Peppe A, Stefani A. Implantation of human pedunculopontine nucleus: a safe and clinically relevant target in Parkinson’s disease. Neuroreport 2005;16:1877-81. Stefani A, Lozano AM, Peppe A, Stanzione P, Galati S, Tropepi D, Pierantozzi M, Brusa L, Scarnati E, Mazzone P. Bilateral deep brain stimulation of the pedunculopontine and subthalamic nuclei in severe Parkinson’s disease. Brain 2007;130:1596-607. Plaha P, Gill SS. Bilateral deep brain stimulation of the pedunculopontine nucleus for Parkinson’s disease. Neuroreport 2005;16:1883-7. Xie J, Krack P, Benabid AL, Pollak P. Effect of bilateral subthalamic nucleus stimulation on parkinsonian gait. J Neurol 2001;248:1068-72. Gentil M, Chauvin P, Pinto S, Pollak P, Benabid AL. Effect of bilateral stimulation of the subthalamic nucleus on parkinsonian voice. Brain Lang 2001;78:233-40.
Subthalamic nucleus stimulation for parkinson’s disease
37. Gentil M, Pinto S, Pollak P, Benabid AL. Effect of bilateral stimulation of the subthalamic nucleus on parkinsonian dysarthria. Brain Lang 2003;85:190-6. 38. Fraix V, Pollak P, Moro E, Chabardes S, Xie J, Ardouin C, Benabid AL. Subthalamic nucleus stimulation in tremor dominant parkinsonian patients with previous thalamic surgery. J Neurol Neurosurg Psychiatry 2005;76:246-8. 39. Goto S, Yamada K, Ushio Y. Subthalamic nucleus stimulation in a parkinsonian patient with previous bilateral thalamotomy. J Neurol Neurosurg Psychiatry 2004; 75:164-165. 40. Anheim M, Batir A, Fraix, V, Silem M, Chabardes S, Seigneuret E, Krack P, Benabid, AL, Pollak, P. Improvement in parkinsonism by STN stimulation depends on electrode placement: effects of reimplantation. Arch Neurol 2008;65:612-616. 41. Breit S, LeBas JF, Koudsie A, Schulz J, Benazzouz A, Pollak P, Benabid AL. Targeting for the implantation of stimulation electrodes into the subthalamic nucleus: a comparative study of magnetic resonance imaging and ventriculography. Neurosurgery 2006;58 1 Suppl: ONS83-95. 42. Hutchison WD, Allan RJ, Opitz H, Levy R, Dostrovsky JO, Lang AE, Lozano AM. Neurophysiological identification of the subthalamic nucleus in surgery for Parkinson’s disease. Ann Neurol 1998;44:622-8. 43. Benazzouz A, Breit S, Koudsie A, Pollak P, Krack P, Benabid AL. Intraoperative microrecordings of the subthalamic nucleus in Parkinson’s disease. Mov Disord 2002;17 Suppl 3:S145-9. 44. Pollak P, Krack P, Fraix V, Mendes A, Moro E, Chabardes S, Benabid AL. Intraoperative micro- and macrostimulation of the subthalamic nucleus in Parkinson’s disease. Mov Disord 2002;17 Suppl 3: S155-61. 45. Deuschl G, Herzog J, Kleiner-Fisman G, Kubu C, Lozano AM, Lyons KE, Rodriguez-Oroz MC, Tamma F, Tro¨ster AI, Vitek JL, Volkmann J, Voon V. Deep brain stimulation: postoperative issues. Mov Dis 2006;21: S219-37. 46. Anheim M, Fraix V, Chabarde`s S, Krack P, Benabid AL, Pollak P. Lifetime of Itrel II pulse generators for subthalamic nucleus stimulation in Parkinson’s disease. Mov Disord 2007;22:2436-9. 47. LePen C, Wait S, Moutard-Martin F, Dujardin M, Zie´gler M. Cost of illness and disease severity in a cohort of French patients with Parkinson’s disease. Pharmacoeconomics 1999;16:59-69. 48. Hamel W, Schrader B, Weinert D, Herzog J, Mu¨ller D, Deuschl G, Volkmann J, Mehdorn HM. Technical complication in deep brain stimulation. Zentralbl Neurochir 2002;63:124-7. 49. Okun MS, Tagliati M, Pourfar M, Fernandez HH, Rodriguez RL, Alterman RL, Foote KD. Management of referred deep brain stimulation failures: a retrospective
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
96
analysis from 2 movement disorders centers. Arch Neurol 2005;62:1250-5. Foncke EM, Schuurman PR, Speelman JD. Suicide after deep brain stimulation of the internal globus pallidus for dystonia. Neurology 2006;66:142-3. Lieb K, Schlaepfer TE. Deep-brain stimulation for Parkinson’s disease. N Engl J Med 2006;355:2256. author reply 2256. Bostwick JM, Rackley SJ. Completed suicide in medical/ surgical patients: who is at risk? Curr Psychiatry Rep 2007;9:242-6 (Review). Kariminia A, Law MG, Butler TG, Levy MH, Corben SP, Kaldor JM, Grant L. Suicide risk among recently released prisoners in New South Wales, Australia. Med J Aust 2007;187:387-90. Pratt D, Piper M, Appleby L, Webb R, Shaw J. Suicide in recently released prisoners: a population-based cohort study. Lancet 2006;368:119-23. Sarwer DB. The psychological aspects of cosmetic breast augmentation. Plast Reconstr Surg 2007;120 7 Suppl 1:110S-117S (Review). Tommasi G, Krack P, Fraix V, LeBas J-F, Chabardes S, Benabid A-L, Pollak P. Pyramidal tract side effects induced by deep brain stimulation of the subthalamic nucleus. J Neurol Neurosurg Psychiatry 2008;79:813-819. Benabid AL, Wallace B, Mitrofanis J, Xia R, Piallat B, Chabardes S, Berger F. A putative generalized model of the effects and mechanism of action of high frequency electrical stimulation of the central nervous system. Acta Neurol Belg 2005;105:149-57. Benabid AL, Pollak P, Gao DM, Hoffmann D, Limousin P, Gay E, Payen I, Benazzouz A. Chronic electrical stimulation of the ventralis intermedius nucleus of the thalamus as a treatment of movement disorders. J Neurosurg 1996;84:203-14. Meissner W, Leblois A, Hansel D, Bioulac B, Gross CE, Benazzouz A, Boraud T. Subthalamic high frequency stimulation resets subthalamic firing and reduces abnormal oscillations. Brain 2005;128:2372-82. Welter ML, Houeto JL, Bonnet AM, Bejjani PB, Mesnage V, Dormont D, Navarro S, Cornu P, Agid Y, Pidoux B. Effects of high-frequency stimulation on subthalamic neuronal activity in parkinsonian patients. Arch Neurol 2004;61:89-96. Chen L, Liu Z, Tian Z, Wang Y, Li S. Prevention of neurotoxin damage of 6-OHDA to dopaminergic nigral neuron by subthalamic nucleus lesions. Stereotact Funct Neurosurg 2000;75:66-75. Xia R, Berger F, Piallat B, Benabid AL. Alteration of hormone and neurotransmitter production in cultured cells by high and low frequency electrical stimulation. Acta Neurochir (Wien) 2007;149:67-73; discussion 73. Turski L, Bressler K, Rettig KJ, Lo¨schmann PA, Watchel H. Protection of substantia nigra from MPP+ neurotoxicity by NMDA antagonists. Nature 1991;349:414-8.
1629
1630
96
Subthalamic nucleus stimulation for parkinson’s disease
64. Piallat B, Benazzouz A, Benabid AL. Subthalamic nucleus lesion in rats prevents dopaminergic nigral neuron degeneration after striatal 6-OHDA injection: behavioural and immunohistochemical studies. Eur J Neurosci 1996;8:1408-14. 65. Piallat B, Benazzouz A, Benabid AL. Neuroprotective effect of chronic inactivation of the subthalamic nucleus in a rat model of Parkinson’s disease. J Neural Transm Suppl 1999;55:71-7. 66. Paul G, Meissner W, Rein S, Harnack D, Winter C, Hosmann K, Morgenstern R, Kupsch A. Ablation of the subthalamic nucleus protects dopaminergic phenotype but not cell survival in a rat model of Parkinson’s disease. Exp Neurol 2004;185:272-80. 67. Nakao N, Nakai E, Nakai K, Itakura T. Ablation of the subthalamic nucleus supports the survival of nigral dopaminergic neurons after nigrostriatal lesions induced by mitochondrial toxin 3-nitropropionic acid. Ann Neurol 1999;45:640-51. 68. Maesawa S, Kaneoke Y, Kajita Y, Usui N, Misawa N, Nakayama A, Yoshida J. Long-term stimulation of the subthalamic nucleus in hemiparkinsonian rats: neuroprotection of dopaminergic neurons. J Neurosurg 2004;100:679-87. 69. Temel Y, Visser-Vandewalle V, Kaplan S, Kozan R, Daemen MA, Blokland A, Schmitz C, Steinbusch HW. Protection of nigral cell death by bilateral subthalamic nucleus stimulation. Brain Res 2006;1120:100-5. 70. Wallace BA, Ashkan K, Heise CE, Foote KD, Torres N, Mitrofanis J, Benabid AL. Survival of midbrain dopaminergic cells after lesion or deep brain stimulation of the subthalamic nucleus in MPTP-treated monkeys. Brain 2007;130:2129-45. 71. Hilker R, Portman AT, Voges J, Staal MJ, Burghaus L, van Laar T, Koulousakis A, Maguire RP, Pruim J, de Jong BM, Herholz K, Sturm V, Heiss WD, Leenders KL. Disease progression continues in patients with advanced Parkinson’s disease and effective subthalamic nucleus stimulation. J Neurol Neurosurg Psychiatry 2005;76:1217-21. 72. Javoy-Agid F, Hirsch EC, Dumas S, Duyckaerts C, Mallet J, Agid Y. Decreased tyrosine hydroxylase messenger RNA in the surviving dopamine neurons of the substantia nigra in Parkinson’s disease: an in situ hybridization study. Neurosci 1990;38:245-53. 73. Luo J, Kaplitt MG, Fitzsimons HL, Zuzga DS, Liu Y, Oshinsky ML, During MJ. Subthalamic GAD gene therapy in a Parkinson’s disease rat model. Science 2002; 298:425-9. 74. Kaplitt MG, Feigin A, Tang C, Fitzsimons HL, Mattis P, Lawlor PA, Bland RJ, Young D, Strybing K, Eidelberg D, During MJ. Safety and tolerability of gene therapy with an adeno-associated virus (AAV) borne GAD gene for Parkinson’s disease: an open label, phase I trial. Lancet 2007;369:2097-105. 75. Marks WJ, Jr, Ostrem JL, Verhagen L, Starr PA, Larson PS, Bakay RAE, Taylor R, Cahn-Weiner DA, Stoessl AJ,
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
Olanow CW, Bartus RT. A phase I, open-label study of CERE-120 (Adeno-Associated Virus Serotype 2 [AAV2]Neurturin [NTN]) to assess the safety and tolerability of intraputaminal delivery to subjects with idiopathic Parkinson’s disease. Lancet Neurol 2008, submitted. Gill SS, Patel NK, Hotton GR, O’Sullivan K, McCarter R, Bunnage M, Brooks DJ, Svendsen CN, Heywood P. Direct brain infusion of glial cell line-derived neurotrophic factor in Parkinson disease. Nat Med 2003;9:589-95. Love S, Plaha P, Patel NK, Hotton GR, Brooks DJ, Gill SS. Glial cell line-derived neurotrophic factor induces neuronal sprouting in human brain. Nat Med 2005; 11:703-4. Patel NK, Bunnage M, Plaha P, Svendsen CN, Heywood P, Gill SS. Intraputamenal infusion of glial cell linederived neurotrophic factor in PD: a two-year outcome study. Ann Neurol 2005;57:298-302. Lang AE, Gill S, Patel NK, Lozano A, Nutt JG, Penn R, Brooks DJ, Hotton G, Moro E, Heywood P, Brodsky MA, Burchiel K, Kelly P, Dalvi A, Scott B, Stacy M, Turner D, Wooten VG, Elias WJ, Laws ER, Dhawan V, Stoessl AJ, Matcham J, Coffey RJ, Traub M. Randomized controlled trial of intraputamenal glial cell line-derived neurotrophic factor infusion in Parkinson disease. Ann Neurol 2006;59:459-66 (Erratum in: Ann Neurol. 2006;60:747). Drouot X, Oshino S, Jarraya B, Besret L, Kishima H, Remy P, Dauguet J, Lefaucheur JP, Dolle´ F, Conde´ F, Bottlaender M, Peschanski M, Ke´ravel Y, Hantraye P, Palfi S. Functional recovery in a primate model of Parkinson’s disease following motor cortex stimulation. Neuron 2004;44:769-78. Pollak P, Champay AS, Hommel M, Perret J, Benabid AL. Subcutaneous apomorphine in Parkinson’s disease. J Neurol Neurosurg Psychiatry 1989;52:544. Stibe C, Lees A, Stern G. Subcutaneous infusion of apomorphine and lisuride in the treatment of parkinsonian on-off fluctuations. Lancet 1987;8537:871. Johnston TH, Fox SH, Brotchie JM. Advances in the delivery of treatments for Parkinson’s disease. Expert Opin Drug Deliv 2005;2:1059-73 (Review). Morizane A, Li JY, Brundin P. From bench to bed: the potential of stem cells for the treatment of Parkinson’s disease. Cell Tissue Res 2008;331:323-36. Tass PA, Hauptmann C. Therapeutic modulation of synaptic connectivity with desynchronizing brain stimulation. Int J Psychophysiol 2007;64:53-61. Munro-Davies LE, Winter J, Aziz TZ, Stein JF. The role of the pedunculopontine region in basalganglia mechanisms of akinesia. Exp Brain Res 1999;129:511-7. Nandi D, Aziz TZ, Giladi N, Winter J, Stein JF. Reversal of akinesia in experimental parkinsonism by GABA antagonist microinjections in the pedunculopontine nucleus. Brain. 2002;125:2418-30.
94 Subthalamotomy for Parkinson’s Disease J. A. Obeso . L. Alvarez . R. Macias . N. Pavon . G. Lopez . R. Rodriguez-Rojas . M. C. Rodriguez-Oroz . J. Guridi
Introduction Deep brain stimulation (DBS) of the subthalamic nucleus (STN) or the globus pallidus pars interna (GPi) has become a standard treatment of motor complications in Parkinson’s disease (PD). Compared with classic ablative surgery, DBS offers the advantage of some degree of reversibility and adaptability, avoids causing new lesions in a brain undergoing a progressive neurodegenerative process and may be less hazardous to undertake in surgical practice. Altogether, the general medical and public opinion has evolved towards favoring DBS over potentially damaging procedures. On the other hand, ablative surgery may be the only or best option on some occasions. This may occur when, for instance, a previously implanted device for DBS has to be removed due to persistent infection or some other technical problem. In addition, DBS may not be easily applied to patients living in remote regions or may be contraindicated for patients suffering from immune deficiency or personality disorders. It should be noted that the cost of DBS and its associated consumables is not affordable for many patients in countries where medical care is not guaranteed by public institutions. Finally, there is a small proportion of PD patients with severe parkinsonian features confined to one hemisphere in whom unilateral surgery may be a reasonable therapeutic option. In such cases, a unilateral lesion may have many practical advantages over DBS.
#
Springer-Verlag Berlin/Heidelberg 2009
Scientific Background and Recent Historical Perspective The STN in the Parkinsonian State Studies conducted in the MPTP (1-methyl, 4-phenyl, 1,2,3,6-tetrahydropyridine) monkey model in the 1990s led to the recognition that hyperactivity of the STN is one of the most fundamental pathophysiological changes in the parkinsonian state. This was based on several experimental approaches. Studies of 2-deoxyglucose (2-DG) regional uptake were performed by Crossman’s group in Manchester [1], extracellular recording of neuronal activity by DeLong’s group at Emory University in Atlanta [2] and assessment of metabolic markers of cellular activity (i.e., glutamic acid decarboxylase = GAD and cytochrome oxidase = CO) by in situ hybridization, which was undertaken by our group in collaboration with Agid et al. in Paris [3]. All of these indicated that dopaminergic depletion led to overactivity of the STN, which by virtue of its then recently recognized glutamatergic excitatory activity [4] overdrives GPi and substantia nigra reticulata (SNr) leading to excessive basal ganglia inhibitory output to the thalamus and brainstem. In keeping with this, Crossman et al. [5] showed that local administration in the GPi of the glutamate antagonist kynurenic acid induced motor improvement in MPTP monkeys. Final proof of the role of the STN in the parkinsonian condition arose with the demonstration that
1570
94
Subthalamotomy for parkinson’s disease
lesion of the STN induces marked permanent motor improvement in MPTP monkeys. This was associated with reduced neuronal activity in GPi/SNr neurons as assessed electrophysiologically and by in situ hybridization studies [6–8]. These findings set the scenario for a revitalization of surgery for PD, but the GPi (i.e., pallidotomy) was chosen as the preferred approach instead of subthalamotomy. This occurred despite the fact that at the time there were no experimental data indicating the anti-parkinsonian effect of pallidotomy. However, the coincidental publication of Laitinen et al. [9] experience with pallidotomy following Leksell’s targeting approach and, particularly, the fear that lesion of the STN would be associated with hemiballism led subthalamic lesion to be discarded as a surgical option in the early 1990s.
Hemichorea-ballism (HCB) and Lesion of the STN In 1949, Whittier and Mettler [10] in a classic paper described for the first time how an electrolytic lesion of the STN in normal monkeys induced contralateral hemiballism (known as choreoid hyperkinesia). They also stated that a lesion with a minimum volume of 20% of the STN was needed and integrity of the pallidum and pallidothalamic pathway was required in order to obtain hemi-dyskinesia. One year later, Carpenter et al. [11] showed that lesion of the GPi or the pallidal efferent projection (ansa lenticularis) abolished the hemiballism induced by lesion of the STN. A few years earlier, Purdon-Martin (1938) had described a patient with a hemorrhage in the STN who showed hemichorea-ballism (HCB) contralateral to the lesion side [12]. On the basis of all these findings, the STN was conceived at the time as exerting inhibition of the basal ganglia output. It is noteworthy that while these studies in monkeys established the involvement of the pallidum and its efferent projections in the pathophysiology of dyskinesias, they also were at
odds with the model of the basal ganglia described in the late 1980s. Indeed, the data indicating a paradoxical anti-dyskinetic effect of pallidotomy, so amply described in the last decade [13], had already been available since the early 1950s. The experience with lesion of the STN in MPTP monkeys indicated that HCB could indeed be induced, but was less severe than in normal animals [8,14]. This led to the concept that in the parkinsonian state the threshold for inducing HCB by lesion of the STN was higher than in the normal state, thus allowing subthalamomy to be considered as a surgical option [15,16]. Indeed, a review of the early literature with stereotactic lesions of the thalamus that were equivocally placed below the inter-commissural line indicated that cases of severe HCB were rare [16] and not even necessarily associated with lesion of the STN [17]. We therefore decided to explore the possibility of using subthalamotomy as another surgical option for the treatment of PD. This differs from previous experiences with lesions in the subthalamic region performed during the 1960s. At that time, so-called campotomies or subthalamotomies were used as an alternative to thalamotomy for alleviation of tremor and rigidity [18,19]. The target for these lesions was Forel’s field, the zona incerta and the prerubral field but not the Corpus Luisii proper.
Technical Aspects Surgery is performed under local anesthesia and stereotactic frame placement and the surgical procedure is similar that that currently used for DBS until the very end of the surgery. The anterior and posterior commissures are identified and the intercommissural line (ICL) measured by using CT and MRI fusion. The image target is placed 12 mm lateral to the midline (X coordinate), 2–3 mm posterior to mid intercommissural point of ICP (Y coordinate) and 3–4 mm below the intercommissural plane (Z coordinate). In the CIREN, where the majority
Subthalamotomy for parkinson’s disease
of our experience was obtained, the STN is defined by semi-microrecording of multiunit action potential and on-line integration of the amplitude and number of spikes at the recording sites [20]. The software program for surgical planning system (STASSIS) is routinely used to localize the different structures along the recording tracks. Typically, the beginning of the STN is defined by an abrupt large increment in the integrated electrical activity. The modification of neuronal activity in response to passive limb displacement and movements of the limbs, neck, and trunk and recording ‘‘tremor cells’’ served to localize the sensorimotor region of the STN in its dorsolateral portion. The number of recording tracks required to define the region to lesion has decreased with experience and it is now about 4–5 per nucleus. A thermolytic lesion is applied with a radiofrequency lesioning probe of 1.1 mm diameter and a 2 or 4 mm exposed tip. The region to be injured is first warmed to 42, 50, and 60 C during periods of 10 s separated by 1–2 min intervals. Subsequently, a lesion is made with a power of 8 W, at 70 C for 60 s. Lesions made
94
with this strategy had an approximate volume of 50–70 mm3 (> Figure 94-1).
Clinical Effects Our first three patients were operated on (unilateral subthalamotomy) at the CIREN in 1995 and the first seven patients reported in 1997 [21]. Another independent pilot study [22], also with a limited number of patients, indicated that unilateral subthalamotomy could be performed safely in most patients. Our experience has increased considerably and the technique refined over the years. In what follows, we summarize the main clinical outcomes as observed in our patients and also reported by other groups.
Parkinsonian Signs In general unilateral subthalamotomy is associated with a reduction of about 50% in the
. Figure 94-1 Magnetic resonance image (T-1 weighted image) in the sagittal (top left), axial (top right) and coronal (bottom) planes with a typical unilateral subthalamic lesion in size and location on the right hemisphere (left side of picture)
1571
1572
94
Subthalamotomy for parkinson’s disease
‘‘off’’ UPDRS (Unified Parkinson’s Disease Rating Scale)-III (motor) and UPDRS-II (Activities of Daily Living) [23–26]. The antiparkinsonian benefit is mainly in the limbs contralateral to the lesion but also in axial features. These effects have been reported to persist for 12–24 months post-operatively in previous publications. The ‘‘on’’ state UPDRS-III also improved significantly, an effect mainly related to the potent antitremor effect of STN surgery [23,27]. The global anti-parkinsonian effect of subthalamotomy is achieved with a concomitant 40–50% reduction in daily dose requirements of levodopa. We have recently assessed the evolution of 25 patients evaluated periodically up to 36 months after surgery. The significant improvement persisted throughout follow-up, except for the ‘‘on’’ state at the last evaluation period (3 years). However, there was a marked increase in the ‘‘off ’’ UPDRS-III by the third year that was secondary to worsening of parkinsonian signs axially and ipsilaterally to the lesion side. In other words, the beneficial effects on the limbs contralateral to the lesion are maintained 3 years after the lesion, but disease progression is noticeable on the ipsilateral side and axial signs. This is a very similar evolution to that observed with pallidotomy [28,29] and nowadays with the occasional parkinsonian patient with unilateral DBS. The percentage of improvement after STN lesion in our group is similar to the benefit reported by other surgical groups [24–26]. Bilateral subthalamotomy has also been performed in PD but the experience is more limited. We reported 17 patients operated with bilateral STN lesion. Seven patients were given staged operations with several months between the procedures and ten patients were operated on the same surgical day (simultaneous surgery). The UPDRS motor part III in the ‘‘off ’’ and ‘‘on’’ state was reduced by 47.5 and 32% respectively 2 years after surgery and the daily dose of levodopa was reduced by 72%.
The scores in the dyskinesia scale decreased significantly throughout the study period [27].
Levodopa-Induced Dyskinesias L-dopa dyskinesias were significantly reduced at 24 months (p < 0.01) and motor fluctuations were also alleviated (p < 0.01). The effect on dyskinesias is predominantly on the contralateral side to the lesion. In our largest series of 89 patients, subthalamotomy increased mean LID score in the side contralateral to the lesion at the first year after surgery but this effect was modest and disappeared upon longer followup. Interestingly, LIDs continue to increase in the non-operated side. Reduction of LIDs in PD patients treated with subthalamotomy[24] or STN-DBS is often explained as a result of decreased levodopa intake but in our patients changes in levodopa daily dose after surgery did not avoid a progressive increment of LIDs on the non-operated side. Thus, we favor the idea that the basal ganglia of the operated hemisphere changes functionally to be less sensitivity to produce LIDs.
Lesion-Induced Hemichorea-ballism The risk of inducing HCB is still the major hurdle to use of subthalamic lesion in patients. We recently assessed the incidence of HCB in 89 consecutive patients evaluated for a minimum of 12 months post-surgery. Dyskinesias appeared in a majority of patients (58.8%) but generally the evolution was benign, tending towards self-resolution within the first few days post-operatively (Alvarez et al., submitted for publication). In 14 patients the dyskinesias persisted and failed to abate over the next few months. In eight of those fourteen patients (9% of the total series), the dyskinesias were severe,
Subthalamotomy for parkinson’s disease
considered as HCB, and required a second surgery (pallidotomy) to resolve the movement disorder (> Figure 94-2). This second procedure was welltolerated in keeping with previous reports [30,31]. What factors determine the occurrence of HCB after subthalamotomy? We still have no good answer to this fundamental practical question. We noticed that patients who developed HCB had a significantly higher levodopa-induced dyskinesia score pre-operatively. Interestingly, a similar observation has been made in PD patients who developed ‘‘off’’ dyskinesias after striatal fetal nigral transplant (Olanow et al., submitted for publication). Together, this perhaps suggests a pre-disposition or increased sensitivity to developing dyskinesias after modifying basal ganglia activity. Intuitively, the location and volume of the lesion should be ‘‘a priori’’ determining factors. However, our experience indicates that the size of the lesion is not a critical factor, since both large and small lesions have been associated with severe dyskinesias. The lesion site may be more relevant. Lozano [32] suggested that subthalamotomy does not generally provoke dyskinesias because the lesion reached the thalamic fasciculus running dorsal to the STN, thus producing a
94
pallidotomy-like effect. This view coincides by and large with the experience described in the monkey by Carpenter et al. [11]. Accordingly, larger lesions which usually extend dorsally, reaching the zona incerta and Forel’s field should be associated with fewer dyskinesias. Our experience with unilateral and bilateral lesions does not support this explanation. We and others have encountered severe and persistent HCB in patients with lesions clearly affecting Forel’s field [25,33] and even reaching the ventral thalamus [17]. A different explanation may be that some regions of the STN may be critically important in the control of complex movement patterns and involved in the induction of HCB. This awaits future results from research currently under way in our study group.
Cognitive Functions McCarter et al. [34] described no major cognitive defect in 12 PD patients treated with unilateral subthalamotomy except for reduced verbal fluency after a follow up averaging six months. We used an ample battery of cognitive and neuropsychiatric tests to evaluate ten patients
. Figure 94-2 Magnetic resonance image (T-1 weighted image) in the axial (left) and coronal (right) planes showing a left subthalamotomy and pallidotomy (right side of picture)
1573
1574
94
Subthalamotomy for parkinson’s disease
subjected to bilateral simultaneous subthalamotomy. Surgery induced no deterioration on cognitive assessments and, on the contrary, a significant improvement on the Hamilton Depression Rating Scale scores, as well as in apathy and depression test scores were encountered. Hyperactive behaviors significantly increased (NPI -neuropsychiatric Inventory-disinhibition and euphoria) immediately following surgery. Such hyperactivity peaked one month after surgery but improved in the following months until becoming normal by the 12-month evaluation. Hyperactive behaviors were most notable in five of the ten patients, three of whom suffered from severe dyskinesias after surgery. This may therefore be related to some factors regarding the volume or location of the lesion, as discussed in the previous section [27]. More recently, we used a similar neuropsychological battery to evaluate 33 PD patients treated with unilateral subthalamic lesion (Bickel et al., submitted for publication). No difference was found in general cognition in the MMSE (mean score 28 1.5 vs. 29.3 2.3), visuo-constructive functions (mean score 20 vs. 22) and FAB (mean score 15.6 2.6 vs. 14.1 1.4). The verbal fluency was reduced pre-operatively (with respect to control subjects) and improved, but remained below the normal range (mean score in control subjects 11 3.1 vs. 16.2 4.3 in patients pre-operatively and 14.0 4.1 post-operatively). Depression improved moderately as shown by a change in the Hamilton scale (mean score 22 5.9 vs. 15 7.5). It thus appears that lesion of the STN in PD is not associated with any major aggravation of cognition. A negative effect on speech, particularly with bilateral lesion, is possible and should be carefully considered in patients who may exhibit speech problems pre-operatively. We have not observed abnormalities related to defect in movement inhibition or behavioral disinhibition in patients with unilateral subthalamotomy, nor has there been any indication of wrong judgments in daily situations which require
taking a decision or making a choice. This contrasts with theoretical and experimental findings in the rat [35] as well as a recent report concerning PD patients treated with bilateral STN-DBS [36]. According to such studies, the STN is crucially involved in the inhibition of movements once they are initiated and in taking the time to make the right choice in situations of decision conflict. It could be that unilateral surgery of the STN is less capable of inducing behavioral deficits than bilateral DBS or subthalamotomy.
Complications STN lesion is generally not associated with a high incidence of adverse events other than lesioninduced dyskinesias. In 89 consecutive patients we found a transient or mild dysarthria in four (4.4%), infection of the scalp incision in three patients (3.3%), epileptic seizure in two patients (2.2%) and asymptomatic bleeding along the recording track trajectory in three patients (3.3%). None of these problems were associated with disability or permanent neurological deficit (Alvarez et al., submitted for publication).
Conclusions Subthalamotomy has a great impact on the basal ganglia (BG) circuit in parkinsonian condition. Experimental studies in the 6-OHDA (hydroxydopamine) in rats and MPTP treated monkeys show that the lesion restores functionally the BG, reducing neuronal hyperactivity and abnormal firing patterns [37]. Indeed, after unilateral STN lesion the BG output nuclei evolve towards a state of functional normalization as indicated by metabolic markers (i.e., mRNA expression of cytochrome oxidase and GAD) [8,38]. Furthermore, in patients with FDG-PET (Fluorodeoxyglucose Positron Emission Tomography) in PD shows a characteristic pattern of increased activity in the globus pallidus, thalamus and
Subthalamotomy for parkinson’s disease
brainstem, and relative reduction in the pre-motor and posterior parietal cortical areas [39]. Infusion of levodopa, pallidotomy, DBS of the STN or GPi are all capable of reducing such abnormal network activity [40]. However, the largest normalizing impact is achieved by STN lesion [39]. Subthalamic lesion induces a marked and significant reduction in the ‘‘off ’’ state severity leading to a reduction in motor fluctuations. It may also be possible that the abnormal signals generated by the BG in response to levodopa in the parkinsonian state are dampened by the lesion. In other words, the lesion may convert the BG into a more stable and less fluctuating system compared with the highly unstable and oscillating state that characterizes the levodopa-treated situation [41]. This may be also a relevant mechanism by which subthalamotomy (as well DBS) reduces motor complications in PD. In practice, we acknowledge that STN lesion is not an approach favored by the groups currently involved in the surgical treatment of PD, while DBS is seen as a less aggressive and reversible technique. Moreover, both the neuroscientific community and patients and society in general are expecting to ‘‘resolve’’ PD by means of more sophisticated and modern techniques such as stem-cell-derived grafts, gene therapy or the application of neuroprotective drugs. None of such potential developments has a high possibility of becoming a therapeutic option in the next 5–10 years. On the other hand, a wealth of experimental evidence indicates that lesion of the STN reduces neuronal death induced by toxins as MPTP or 6-OHDA [42]. Perhaps, we are all missing the chance of modifying PD progression by not applying a well-known, classic, surgical principle, i.e., eliminating what is not working well.
References 1. Mitchell IJ, Clarke CE, Boyce S, Robertson RG, Peggs D, Sdambrook MA, Crossman AR. Neural mechanism underlying parkinsonian symptoms based upon regional uptake of 2-deoxiglucose in monkeys exposed to
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12. 13.
14.
15.
16.
94
1-methyl-4phenyl-1,2,3,6-tretrahydropyridine. Neuroscience 1989;32:213-26. Miller WC, DeLong MR. Altered tonic activity of neurons in the globus pallidus and subthalamic nucleus in the primate model of parkinsonism. In: Carpenter MB, Jayaraman A, editors. The basal ganglia II. Structure and function. New York: Plenum; p. 415-27; 1987. Levy R, Herrero MT, Ruberg M, Villares J, Faucheux B, Guridi J, Guillen J, Luquin MR, Javoy-Agid F, Obeso JA, et al. Effects of nigrostriatal denervation and L-dopa therapy on the GABAergic neurons in the striatum, in MPTP-treated monkeys and Parkinson’s disease: an in situ hybridization study of GAD67 mRNA. Eur J Neurosci 1995;7:1199-209. Smith Y, Parent A. Neurons of the subthalamic nucleus in primates diplay glutamate but not GABA immunoreactivity. Brain Res 1988;453:353-6. Crossman AR, Peggs D, Boyce S, Luquin MR, Sambrook MA. Effecto of NMDA antagonist MK-801 on MPTPinduced parkinsonism in the monkey. Neuropharmacology 1989;28:1271-3. Bergman H, Wichmann T, DeLong MR. Reversal of experimental parkinsonism by lesions of the subthalamic nucleus. Science 1990;249:1436-8. Aziz TZ, Peggs D, Sambrook MA, Crossman AR. Lesion of the subthalamic nucleus for the alleviation of 1-methyl, 4 phenyl, 1,2,3,6 tetrahydropiridine (MPTP)-induced parkinsonism in the primate. Mov Disord 1991;6:288-92. Guridi J, Herrero MT, Luquin MR, Guille´n J, Ruberg M, Laguna J et al. Subthalamotomy in parkinsonian monkeys. Behavioural and biochemical analysis. Brain 1996; 119:1717-27. Laitinen LV, Bergenheim AT, Hariz MI. Leksell’s posteroventral pallidotomy in the treatment of Parkinson’s disease. J Neurosurg 1992;76:53-61. Whittier JR, Mettler FA. Studies on subthalamus of rhesus monkey: hyperkinesia and other physiologic effects of subthalamic lesions, with special reference to the subthalamic nucleus of Luys. J Com Neurol 1949;90:319-72. Carpenter MB, Whittier JR, Mettler FA. Analysis of choreoid hyperkinesia in the rhesus monkey. J Comp Neurol 1950;92:293-322. Purdon Martin J, Alcock NS. Hemichorea associated with a lesion of the corpus Luysii. Brain 1934;57:504-16. Marsden CD and Obeso JA. The functions of the basal ganglia and the paradox of stereotaxic surgery in Parkinson’s disease. Brain 1994;117:877-97. Obeso JA, Rodriguez MC, DeLong MR. Basal ganglia pathophysiology: a critical review. Adv Neurol 1997; 74:3-18. Guridi J, Herrero MT, Luquin MR, Obeso JA. The subthalamic nucleus: a new possible stereotaxic target for Parkinson’s disease. Mov Disord 1993;8:421-9. Guridi J, Obeso JA. The subthalamic nucleus, hemiballismus and Parkinson’s disease: reappraisal of a neurosurgical dogma. Brain 2001;124:5-19.
1575
1576
94
Subthalamotomy for parkinson’s disease
17. Dierssen G, Bergmann L, Gioino L, Cooper IS. Hemiballism following surgery for Parkinson’s disease. Arch Neurol 1961;5:627-37. 18. Andy OJ, Jurko MF, Sias FR. Subthalamotomy in treatment of parkinsonian tremor. J Neurosurg 1963; 20:860-70. 19. Fager CA. Evaluation of thalamic and subthalamic surgical lesions in the alleviation of Parkinson’s disease. J Neurosurg 1963;28:145-9. 20. Lopez G, Morales J, Tejeiro J, Vitek JL, Perez-Paara S, Fernandez R, Maragoto C, et al. Anatomic and neurophysiological methods for the targeting and lesioning of the subthalamic nucleus. Cuban experience and review. Neurosurgery 2003;52:817-31. 21. Obeso JA, Alvarez L, Macias R, Guridi J, Juncos JL, Tejeiro J, et al. Neurology 1997;48 Suppl 3:A138. 22. Gill SS, Heywood P. Bilateral dorsolateral subthalamotomy for advanced Parkinson’s disease. Lancet 1997; 350:1224. 23. Alvarez L, Macias R, Guridi J, Lopez G, Alvarez E, Maragoto CT, et al. Dorsal subthalamotomy for Parkinson’s disease. Mov Disord 2001;16:72-8. 24. Patel NK, Heywood P, O’Sullivan K, McCarter R, Love S, Gill SS. Unilateral subthalamotomy in the treatment of Parkinson’s disease. Brain 2003;126:1136-45. 25. Su PC, Tseng HM, Liu HM, Yen RF, Liou HH. Subthalamotomy for advanced Parkinson disease. J. Neurosurg 2002;97:598-606. 26. Vilela F, da Silva DJ. Unilateral subthalamic nucleus lesioning: a safe and effective treatment for Parkinson’s disease. Arq. Neuropsiquiatr 2002;60:935-48. 27. Alvarez L, Macias R, Lopez G, Alvarez E, Pavon N, Rodriguez-Oroz MC, Juncos JL, Maragoto C, Guridi J, Litvan I, Tolosa ES, Koller W, Vitek J, DeLong MR, Obeso JA. Bilateral subthalamotomy in Parkinson’s disease: initial and long-term response. Brain 2005; 128:570-83. 28. Lang AE, Lozano AM, Montgomery E, Duff J, Tasker R, Hutchison W. Posteroventral pallidotomy in advanced Parkinson’s disease. N Engl J Med 1997;337:1036-42. 29. Baron MS, Vitek JL, Bakay RAE, Kaneoke Y, Hashimoto T, Turner RS, et al. Treatment of advanced Parkinson’s disease by posterior GPi pallidotomy: 1-year results of a pilot study. Ann Neurol 1996;40:355-66. 30. Suarez JI, Metman LV, Reich SG, Dougherty PM, Hallet M, Lenz FA. Pallidotomy for hemiballismus efficacy and characteristics of neuronal activity. Ann Neurol 1997;42:807-11. 31. Vitek JL, Chockkan V, Zhang JY, Kaneoke Y, Evatt M, DeLOng MR, et al. Neuronal activity in the basal ganglia in patients with generalized dystonia and hemiballismus. Ann Neurol 1999;46:22-35. 32. Lozano AM. The subthalamic nucleus: myth and opportunities. Mov Disord 2001;16:183-4. 33. Merello M, Perez-Lloret S, Antico J, Obeso JA. Dyskinesias induced by subthalamotomy in Parkinson’s
34.
35.
36.
37.
38.
39.
40.
41.
42.
disease are unresponsive to amantadine. J Neurol Neurosurg Psychiatry 2006;77:172-4. McCarter RJ, Walton NH, Rowan AF, Gill SS, Palomo M. Cognitive functioning after subthalamic nucleotomy for refractory Parkinson’s disease. J Neurol Neurosurg Psychiatry 2000;69:60-6. Baunez C, Nieoullon A, Amalric M. In a rat model of parkinsonism, lesions of the subthalamic nucleus reverse increases of reaction time but induce a dramatic premature responding deficit. J Neurosci 1995; 15:6531-41. Frank MJ, Samanta J, Moustafa AA, Sherman SJ. Hold your horses: impulsivity, deep brain stimulation, and medication in parkinsonism. Science 2007;23318 (5854):1309-12. Wichmann T, Bergman H, DeLong MR. The primate subthalamic nucleus. III. Changes in motor behavior and neuronal activity in the internal pallidum induced by subthalamic inactivation in the MPTP model of parkinsonism. J Neurophysiol 1994;72:521-30. Blandini F, Garcia-Osuna M, Greenamyre JT. Subthalamic ablation reverses changes in the basal ganglia oxidative metabolism and motor response to apomorphine induced by nigrostriatal lesion in rats. Eur J Neurosci 1997;9:1407-13. Trost M, Su PC, Barnes A, Su SL, Yen RF, Tseng HM, Ma Y, Eidelberg D. Evolving metabolic changes during the first postoperative year after subthalamotomy. J Neurosurg 2003;99:872-8. Asanuma K, Tang C, Ma Y, Dhawan V, Mattis P, Edwards C, Kaplitt MG, et al. Network modulation in the treatment of Parkinson’s disease. Brain 2006;129:2667-78. Obeso JA, Rodriguez-Oroz MC, Marin C, Alonso F, Zamarbide I, Lanciego JL, Rodriguez-Diaz M. The origin of motor fluctuations in Parkinson’s disease: importance of dopaminergic innervation and basal ganglia circuits. Neurology 2004;62 Suppl 1:17-30. Piallat B, Benazzouz A, Benabid AL. Subthalamic nucleus lesion in rats prevents dopaminergic nigral neuron degeneration after striatal 6-OHDA injection behavioural and immunohistochemical studies. Eur J Neurosc 1996;8:1408-14.
Functional Neurosurgery for Movement and Motor Disorders – Clinical Aspects
86 Surgery for Movement Disorders: An Overview K. M. Prakash . A. E. Lang
Introduction Surgery has become a well-established form of therapy in movement disorders such as in Parkinson’s disease (PD) and essential tremor (ET), especially for patients with symptoms that are refractory to medications or who have intolerable side effects related to medical therapy. Other movement disorders also appear to benefit from stereotactic surgery including dystonia, tremors associated with multiple sclerosis, and tics seen in Tourette’s syndrome (TS). In this chapter we will provide an overview of various movement disorders with an emphasis on those that may benefit from functional neurosurgical interventions. We will also briefly outline those treatments that have been effective; the reader is directed to many other chapters in this book which cover these issues in greater detail.
Definition of Movement Disorders Movement disorders are clinically, pathologically and genetically heterogeneous and are characterized by impairment of the planning, control or execution of movement. They are generally divided into hypokinetic versus hyperkinetic disorders. The hypokinetic disorders are characterized by akinesia, bradykinesia, and rigidity. Akinesia is defined as a paucity of movement while bradykinesia refers to slowness of movement. Rigidity is an involuntary increase in muscle tone, appreciated equally in flexors and extensors that may have cog-wheeling when tremor is superimposed. Hypokinetic disorders, also referred to #
Springer-Verlag Berlin/Heidelberg 2009
as akinetic-rigid syndromes, include PD and various forms of parkinsonism. Hyperkinetic disorders are disorders in which there is an excess amount of movement, either spontaneous or in response to a volitional movement or another stimulus. They are often, involuntary though some, most notably tics, have a voluntary component as well. The most common types of hyperkinetic disorders include tremor, dystonia, tics, chorea, ballismus and myoclonus. It is not uncommon for patients to present with a combination of movement disorders, such as the resting tremor and akinetic-rigid features of Parkinson’s disease or dystonia and tremor.
Parkinson’s Disease and Atypical Parkinsonism The classical features of rest tremor, rigidity, bradykinesia and postural instability characterize PD [1]. The onset of PD is slow and the deficits are usually asymmetric. It has a gradual progression and a sustained response to dopaminergic medications. Pathologically, neuronal loss predominates in the substantia nigra zona compacta, particularly the ventrolateral tier; this leads to decreased dopamine in the striatum with a rostrocaudal gradient (maximum loss caudally). In addition to motor dysfunction, patients with PD also have a wide spectrum of nonmotor manifestations due to varied patterns of degeneration in dopaminergic (mesolimbic, mesocortical, retinal) and nondopaminergic (cholinergic, noradrenergic, serotoninergic) neuronal systems [1]. These nonmotor manifestations may occur early
1446
86
Surgery for movement disorders: an overview
in the disease but are generally more disabling later on with disease progression and include fatigue, cognitive decline, depression, anxiety, behavioral disturbances, visual dysfunction, dysautonomia, weight loss, sleep abnormalities, abnormal sensations, and pain [2,3]. Atypical parkinsonism or parkinson-plus disorders may mimic PD in the early stages, although often ‘‘red flags’’ are present to indicate an alternative diagnosis. In addition to a variety of historical features and clues on physical examination, one important aspect of these disorders is the response to levodopa which is either poor from the outset, short-lived or unsustained. Atypical parkinsonisms include other neurodegenerations such as multiple system atrophy (MSA), progressive supranuclear palsy (PSP), and corticobasal degeneration (CBD). ‘‘Secondary parkinsonism’’ often refers to parkinsonism due to definable nondegenerative causes such as neuroleptic drugs, toxins, hydrocephalus, brain tumor etc. (see > Table 86-1 for differential diagnosis of parkinsonism). The most commonly utilized research scale to evaluate disease severity and progression in PD is the ‘‘Unified Parkinson’s Disease Rating Scale’’ (UPDRS) which includes mentation, behavior, and mood; activities of daily living and motor sections and a drug-related motor complications section. The first 2 and fourth sections are evaluated by interview and the motor section is rated by examination. The severity of PD is often described by Hoehn and Yahr scale, which defines five ‘‘stages’’ (I–V) of the disease. Stage I includes unilateral motor features (which include tremor, rigidity, and bradykinesia), stage II involves bilateral motor features, stage III worsening bilateral features with balance difficulties however still independent, stage IV includes patients who are dependant but able to ambulate with assistance, and stage V includes patients who are wheelchair-bound or bed ridden. The Schwab and England Activities of Daily Living Scale is commonly used to assess disability in PD. Scoring is rated by a rater or the patient and scored
0–100% (where 0% signifies bedridden patients with non functioning vegetative abilities like swallowing, bladder and bowel function; 100% signifies complete independence and essentially normal function). In the early stages, PD is responsive to medical therapies, which include anticholinergic drugs for tremor, dopamine agonists, levodopa, monoamine oxidase-B and catechol-O-methyl transferase inhibitors and others. With the advance of the underlying degeneration, PD patients develop motor and nonmotor features, which are less responsive to medications, and more complications related to the long-term medications that include motor fluctuations and dyskinesias. Motor fluctuations consist of variation in clinical status that occurs hour to hour over the course of a day and dyskinesia refers to abnormal involuntary movements (typically mixtures of chorea, athetosis, and dystonia) occurring in association with medication therapy. Such patients may rapidly alternate between ‘‘off-periods’’ and ‘‘onperiods’’ with dyskinesias. The periods of optimal balance become shorter or absent as the disease progresses. It is these patients that have been helped significantly by surgical procedures. Surgery for PD has been attempted since the start of the twentieth century [5,6]. Stereotactic procedures were introduced in the 1940s [7,8]. Different targets in the brain (especially the thalamus and palladium) were explored to control PD symptoms. However with the introduction of levodopa in 1967, there was a rapid decline in these surgical procedures. Levodopa caused a marked improvement of motor symptoms and reduction of morbidity and mortality. However, with the long-term use of levodopa patients developed motor fluctuations and dyskinesia, which often resulted in marked disability. With the better understanding of basal ganglia circuitry and possible surgical targets for PD, improved neuroimaging and electrophysiological recording techniques there has been a resurgence of surgery in PD.
Surgery for movement disorders: an overview
86
. Table 86-1 Differential Diagnoses of Parkinsonism (Adapted from [4]) 1. Parkinson’s disease Sporadic Genetic Autosomal dominant (e.g.,a synuclein gene mutations, duplications, triplications; LRRK2 mutations) Autosomal recessive (e.g., parkin, DJ-1, PINK-1) 2. Secondary parkinsonism Neurodegenerative diseases (sporadic or genetic) Multiple system atrophy (MSA) Progressive supranuclear palsy (PSP) Corticobasal degeneration (CBD) Dementia with Lewy bodies (DLB) Alzheimer’s disease ALS-parkinsonism-dementia complex of Guam Spinocerebellar ataxias (e.g., SCA 2, SCA 3) Huntington’s disease Neuroacanthocytosis Wilson’s disease Pantothenate kinase-associated neurodegeneration (PKAN) due to PANK2 mutations Calcification of the basal ganglia (Fahr’s syndrome) Neuroferritinopathy Others Dopa-responsive dystonia Drugs Neuroleptics, prochlorperazine, metoclopramide, tetrabenazine, reserpine, cinnarizine, flunarizine, alpha-methyldopa, lithium Toxic MPTP, manganese, carbon monoxide, mercury Infectious/inflammatory Encephalitis lethargica Other encephalitis including HIV Subacute sclerosing panencephalitis (SSPE) Creutzfeldt-Jakop disease Multiple sclerosis Vascular Atherosclerosis Amyloid angiopathy Neoplastic Primary brain tumor or brain metastasis Normal pressure hydrocephalus Head trauma Psychogenic parkinsonism
In general, PD patients who respond to levodopa and have motor fluctuations and/or dyskinesia that cannot be controlled with medications and patients with disabling and medically resistant tremor are candidates for surgery [5]. However, symptoms unresponsive to levodopa (e.g., postural reflex impairment, hypophonia/ dysarthria, swallowing, and mentation) will most
likely not improve with surgery. Significant cognitive, behavioral or psychiatric problems such as active severe depression and important concomitant medical illnesses represent major exclusions for surgery. Two types of surgeries, ablative surgery and deep brain stimulation (DBS), are available for PD patients [6,7]. The ablative surgeries include
1447
1448
86
Surgery for movement disorders: an overview
thalamotomy, pallidotomy and subthalamotomy. Ablative surgeries are relatively cheaper compared to DBS and they are favored in less developed countries. However, nowadays DBS is favored more than ablative surgeries due to its advantages, which include adjustability and reversibility. The targets for DBS include the thalamus (ventral intermediate (Vim) nucleus), globus pallidus interna (GPi) and subthalamic nucleus (STN). Thalamic stimulation may be considered for patients with disabling tremor-predominant form who are medication resistant and have minimal signs of bradykinesia and rigidity, as these latter symptoms are not improved with thalamic stimulation. However most tremor-predominant patients are eventually impaired by these other symptoms and so Vim surgery has been largely abandoned in PD except for elderly patients with long standing disabling tremor. Patients with a combination of PD tremor and essential tremor are also candidates for this surgery, although again other targets such as the STN are generally preferred. GPi stimulation has been shown to improve tremor, bradykinesia and rigidity and markedly reduce levodopa-induced dyskinesia. Benefit may not be sustained and is overall somewhat less than with STN DBS, therefore, the most common surgical target for PD is now the STN. DBS of the STN improves all motor symptoms, which include tremor, bradykinesia, rigidity, posture and gait in the medication off state [9]. There is also marked improvement of dyskinesia largely due to reduction in antiparkinsonian medications after surgery [10] and improvement in the duration of ‘‘on’’ time. Activities of daily living show a significant improvement too. The best predictor of this response is the preoperative response to levodopa and generally any symptoms that remain resistant to levodopa do not respond to STN DBS. More recently, reports have suggested an important role for the upper brainstem, and in particular, the pedunculopontine nucleus (PPN),
in the development of motor symptoms, such as akinesia, gait dysfunction and postural instability [11,12]. The PPN is part of the mesencephalic locomotor reticular region [13] and maintains dense interconnections with the basal ganglia and several other pontine and medullary areas [14,15]. Preliminary open-label evaluations of PPN DBS in a very small number of patients have provided promising results [16,17]. The hope is that this treatment may improve ambulatory features that are resistant to current medical and surgical therapies. Other surgical therapies in PD are either in the experimental or early stages. These include motor cortex stimulation, intraparenchymal drug delivery, gene therapy directed at altering basal ganglia network function (STN GAD), dopamine metabolism (AADC) or providing trophic support (Neurturin), and cell based therapies for the purposes of providing dopamine replacement (Spheramine) or striatal reinnervation directly or indirectly through trophic stimulation. These will be further elaborated in subsequent chapters. The management of atypical parkinsonism currently includes symptomatic and palliative strategies, as well as family education and support; the ultimate goal is to improve the quality of life for patients and their caregivers. DBS and other surgical therapies, while promising for idiopathic PD, have been largely ineffective for atypical parkinsonism [18,19]. Multiple system atrophy (MSA) represents the largest cohort of patients with atypical parkinsonism who have undergone functional surgery including pallidotomy, transplantation procedures and DBS and nearly all authors have reported poor outcomes with surgery [20–23]. Deep brain stimulation of the PPN may be useful in treating drug-resistant balance and gait disorders in PD and so there is interest in evaluating this technique in the atypical parkinsonian disorders, particularly in patients with PSP and MSA.
Surgery for movement disorders: an overview
Tremor Tremor is one of the most common involuntary movement disorders seen in clinical practice. It is defined as an involuntary, rhythmic, and sinusoidal oscillation of one or more body parts produced by alternating or synchronous contractions of agonist and antagonist muscles. Tremor is classified as resting or action tremor. A resting tremor is a tremor in a limb that is in a resting position, with its weight fully supported against gravity. It is typically seen in parkinsonism, more commonly in Parkinson’s disease than other parkinsonian syndromes. Action tremor is further divided into postural, kinetic or intention tremor. They all imply an active contraction of the muscles involved. A postural tremor is seen with the maintenance of a posture, against gravity, such as when the arms are outstretched in front of the body. A kinetic tremor is seen with a voluntary movement of the limb, such as a tremor in an upper limb during finger-to-nose maneuver. Intention tremor increases in amplitude when approaching a target. > Table 86-2 provides a differential diagnosis of tremor as well as listing a variety of other types of rhythmical movement disorders. Essential tremor (ET) is the commonest pathological form of action tremor in movement disorders that is either postural or kinetic in character and mainly affecting the hands. It is usually bilateral with a frequency of 4–12 Hz and largely symmetrical [24]. The upper limbs are affected in about 95% of patients, followed by head (34%), lower limbs (20%), voice (12%), face and trunk (5%) [25]. A diagnosis of dystonic tremor should be strongly suspected when tremor of the head occurs in isolation or precedes the development of hand tremor. With the passage of time, the frequency of the tremor in ET decreases and the amplitude may increase [26]. The prevalence ranges from 0.4 to 6.7% in persons over 40 years old so it is the most common type of tremor apart from the
86
. Table 86-2 Differential diagnosis of tremor and rhythmical movement disorders (Adapted from [4]) 1. Physiologic tremor Enhanced physiologic tremor Metabolic Hypoglycemia Hyperthyroidism Hyperparathyroidism Pheochromocytoma Drugs Caffeine Beta-agonists Theophylline Amiodarone Valproic acid Antidepressants Amphetamines Lithium Others Withdrawal of drugs Alcohol Benzodiazepines Others Anxiety, stress, fatigue Fever, sepsis 2. Primary or idiopathic Essential tremor Task-specific tremor Orthostatic tremor 3. Tremor associated with CNS diseases Tremor with parkinsonian syndromes Idiopathic Parkinson’s disease Other ‘‘atypical’’ parkinsonisms (PSP, MSA, CBD) Neuroleptic-induced parkinsonism Wilson’s disease Multiple sclerosis Fragile X premutation – tremor/ataxia syndrome (FXTAS) Stroke Brain tumor Head trauma Midbrain tremor (Holme’s tremor) 4. Ttremor associated with peripheral neuropathies 5. Psychogenic tremor 6. Other rhythmical movement disorders Palatal tremor (‘‘essential’’ or ‘‘isolated’’ and symptomatic) Rhythmical myoclonus (including ‘‘myoclonic tremor’’) Rhythmical movements in dystonia (‘‘dystonic tremor’’) Clonus Asterixis Epilepsia partialis continua
1449
1450
86
Surgery for movement disorders: an overview
. Table 86-2 (Continued) Hereditary chin quivering Spasmus nutans Nystagmus Head bobbing with hydrocephalus
normal physiological tremor found in everyone which may be accentuated by a variety of factors including many medications. The age of onset of ET is typically 60–70 years, but not uncommonly well before 60 years (even in childhood), and both sexes are equally affected. A majority of ET patients note a profound reduction of tremor in response to alcohol but this is temporary and may be followed by rebound worsening. In multiple sclerosis (MS), tremor can be seen in the upper extremities (55%), lower extremities (8%), head (7%) and trunk (5%) [27]. The pathophysiology of tremor in MS is not fully understood, mainly because of the presence of multiple central nervous system (CNS) lesions, which prevent precise neuroanatomical correlation of structure to function [28]. It is often one component in a complex movement disorder that includes dysmetria and other ataxia features [28– 32]. The tremor is generally believed to be cerebellar in nature and accompanied by other cerebellar, sensory and corticospinal impairment [33]. Holmes’ tremor, previously known as midbrain tremor, rubral tremor, thalamic tremor, myorhythmia, and Benedikt’s syndrome [34], predominantly occurs in proximal limbs. It is of low frequency (<4.5 Hz), commonly with a rest component that clearly worsens with postural maintenance, worsening still during movement and goal directed tasks. Holmes’ tremor is almost always attributable to lesions in upper brain stem, thalamus, or cerebellum, interrupting pathways in the midbrain tegmentum (rubro-olivocerebellorubral loop, rubrospinal fibers, dopaminergic nigrostriatal fibers and the serotonergic brain stem telencephalic fibers). The disability of these patients is extreme, because they are neither able to
keep their hands still nor can they reach and grasp, when the syndrome is fully expressed. The Fahn-Tolosa-Marin Tremor Rating Scale (TRS) is a widely used clinical rating scale for tremor [35], rating severity of tremor by body part from 0 (none) to 4 (severe). The scale is divided into three parts. Part A assesses examiner-reported tremor location/severity (amplitude), Part B assesses examiner-reported ability to perform specific motor tasks/functions (writing, drawing, and pouring with dominant and nondominant hands), and Part C assesses patient-reported functional disability resulting from the tremor (speaking, eating, drinking, hygiene, dressing, writing, working, and social activities). Finally, the TRS includes one separate item dealing with global assessment of tremorrelated disability, rated by both patient and examiner on a 5-point scale. The two most often used drugs in ETare nonselective b-blockers (for example, propranolol) and primidone. Although the mechanism action of these drugs in ET is not exactly known, they are significantly effective and generally well tolerated [36–42]. Apart from these first line drugs, topiramate, gabapentin and alprazolam have also been used as monotherapy or adjunctive treatment [43,44]. Intramuscular injections of botulinum toxin have been considered in medically resistant cases especially for head and voice tremor [45]. In general, the less pronounced the symptoms, the better the chances of a favorable outcome with medical treatment for patients with ET. However in a minority of patients the disability becomes unacceptable despite medical therapy when the tremor significantly interferes with feeding, drinking, and writing, or, in the case of vocal tremor, disrupts communication. The results of medical treatment for MS tremor are often less than satisfactory. Several drugs have been tried, but with variable success, including drugs used for ET as well as odansetron (5-HT3 antagonist), isoniazid, physostigmine, carbamazepine, and clonazepam [46,47].
Surgery for movement disorders: an overview
The progressive nature of the disease makes medical treatments empirical and generally unrewarding. Similarly, it is difficult to treat a Holmes’ tremor. Although levodopa and clonazepam have been reported to be effective [48–50], these are single case reports and generally pharmacotherapy is not effective [51,52]. In addition, no controlled studies have been performed because of the rarity of this disorder. In medically refractory tremulous patients, an alternative to pharmacotherapy is stereotactic neurosurgery typically involving the thalamus. An alternative to ‘‘open’’ surgery is gamma knife thalamotomy which may be useful in elderly patients with medical contraindications to more conventional thalamotomy or DBS. ET patients with pure postural tremor of the upper extremities are highly likely to improve after either lesioning [53–59] or DBS [60–67] of the VIM or the subthalamic area [68]. If intention tremor or a more proximal tremor predominates, the success rate of surgery is slightly decreased. If head, voice, or trunk tremors are the main indications for surgery, a bilateral procedure is usually necessary [65–67] and bilateral DBS is favored over thalamotomy as the latter carries higher risk of complications although bilateral thalamic DBS does have a moderate risk of dysarthria [69,70]. In refractory cases of tremor in MS patients, thalamotomy and chronic DBS of the VIM (or less commonly nucleus ventralis oralis posterior and zona incerta or subthalamic area) may provide benefit [71–74]. Although few studies used highly standardized quantitative outcome measures, and follow up periods were generally one year or less, the data has suggested that chronic DBS of the VIM produces improved tremor control in multiple sclerosis [70]. However, complete cessation of tremor is generally not achieved. There have been some reported cases in which tremor control decreased over time, and frequent reprogramming became necessary. One potential method of predicting response of the disability in MS is the use of electrophysiological
86
studies. Patients with a very narrow tremor frequency band may respond better than those with a broad band that suggest considerable superimposed ataxia. The surgical treatment of Holmes’ tremors with stereotactic radiofrequency lesioning (posteroventral pallidotomy) and high frequency thalamic stimulation has been reported [38,52,74–76]. The VIM nucleus has been reported to be the effective target of stereotactic surgery in Holmes’ tremors [52,76,77]. Nevertheless, there is some uncertainty about whether these patients do respond favorably to stereotactic intervention and much depends on the extent of the causative lesion and the specific symptom pattern of the patient.
Dystonia Dystonia is defined as a neurological syndrome characterized by involuntary, patterned, sustained, or repetitive muscle contractions of opposing muscles, causing twisting movements and abnormal postures. This is one of the most disabling movement disorders [78]. It may affect the trunk, neck, face, arms or legs. Action dystonia refers to abnormal postures that occur during voluntary activity, which are sometimes taskspecific, for example writer’s cramp. Dystonias are classified based on anatomical distribution (focal, segmental, or generalized), age at onset, etiology, and genetics [79,80]. Childhoodonset dystonia usually progresses from focal limb dystonia to a severe generalized form, whereas dystonia that begins after the age of 25 years usually involves craniocervical muscles, remains localized or segmental, and is usually non-progressive [79]. The classification based on etiology (see > Table 86-3) includes primary dystonia, secondary dystonia, dystonia-plus syndromes, and paroxysmal dystonia [80]. The genetic classification is based on the loci of genes involved (loci DYT1 through DYT15) that include autosomal
1451
1452
86
Surgery for movement disorders: an overview
. Table 86-3 Classification and causes of dystonia (Adapted from [4]) 1. Primary dystonias (primary torsion dystonia; previously known as dystonia musculorum deformans) Familial (DYT1, 2, 4, 6, 7, and 13) Sporadic (usually adult-onset, focal or segmental) 2. Dystonia-plus Dystonia with parkinsonism Dopa-responsive dystonia Dopamine-agonist responsive dystonia (e.g., aromatic acid decarboxylase deficiency) Myoclonus-dystonia 3. Heredodegenerative dystonias X-linked Lubag Deafness-dystonia-optic atrophy syndrome (Mohr-Tranebjaerg) Pelizaeus-Merzbacher disease Autosomal-dominant Rapid-onset dystonia-parkinsonism Juvenile parkinsonism (e.g., due to mutations in the parkin gene) Huntington’s disease SCAs and Dentato-rubro-pallido-luysian atrophy (DRPLA) Autosomal-recessive Wilson’s disease Niemann-Pick type C GM1 gangliosidosis GM2 gangliosidosis Metachromatic leukodystrophy Lesch-Nyhan syndrome Homocystinuria Glutaric academia Triose-phosphate isomerase deficiency Hartnup’s disease Ataxia-telangiectasia Ataxia with oculomotor apraxia I & II Pantothenate kinase-associated neurodegeneration (PANK2 mutations) Juvenile neuronal ceroid lipofuscinosis Neuroacanthocytosis Intranuclear hyaline inclusion disease Hereditary spastic paraplegia with dystonia Probable autosomal recessive Familial basal ganglia calcifications Progressive pallidal degeneration Rett’s syndrome Mitochondrial Leigh’s disease Leber’s disease Other mitochondrial cytopathies
. Table 86-3 (Continued) Sporadic, with parkinsonism Parkinson’s disease Progressive supranuclear palsy Multiple system atrophy Corticobasal degeneration 4. Secondary dystonias Drug-induced (acute and tardive dystonia) Perinatal cerebral injury Delayed-onset dystonia Athetoid cerebral palsy Pachygyria Kernicterus Encephalitis Subacute sclerosing leukoencephalopathy Wasp sting Creutzfeldt-Jakob disease Human immunodeficiency syndrome Vascular Stroke Arteriovenous malformation Primary antiphospholipid syndrome Hypoxia Head trauma Thalamotomy Brainstem lesion Brain tumor Multiple sclerosis Central pontine myelinolysis (extrapontine myelinolysis) Peripheral injury Toxins Hypoparathyroidism Psychogenic
dominant, autosomal recessive and X-linked causes of primary dystonia, dystonia-plus syndromes and paroxysmal dystonias. Primary (generalized, focal or segmental) dystonias are unaccompanied by other neurologic abnormalities, except tremor and occasionally myoclonus, and have no known cause except for the genetic mutations in some cases. About 90% of primary generalized dystonia in Ashkenazi Jews and 50% of primary generalized dystonia in other populations is due to deletion of a GAG triplet from the DYT1 gene located at chromosome 9q34 [81]. DYT1 dystonia is inherited in an autosomal dominant fashion with penetrance of only 30% and in 94% of cases symptoms begin
Surgery for movement disorders: an overview
in a limb [82]. In about two-thirds of cases there is progression to generalized or multifocal dystonia [81], which usually occurs within 5 years of onset, but can occur later [83]. Primary focal dystonias are more common than primary generalized dystonia [84] and nearly always occur in adults typically beginning in midlife or later. These may involve the neck, face, or arm, whereas the leg is rarely involved. Cervical dystonia, the most common focal dystonia, usually begins between the ages of 30 and 50 years often with neck stiffness, restricted head mobility and abnormal head postures which are sometimes associated with irregular head tremor. Sensory tricks, for example lightly touching the face or chin reduce the severity of symptoms, are common and particularly effective early in the course; this phenomenon is also known as the geste antagoniste. Cranial dystonia (one form is sometimes referred to as Meige’s syndrome) comprises involvement of a variety of facial, jaw and oropharyngeal muscles. It is often associated with cervical involvement (craniocervical dystonia) and sometimes with involvement of the larynx (spasmodic dysphonia). Blepharospasm causes abnormal contraction of the orbicularis oculi resulting in excessive blinking and forced eyelid closure, sometimes causing functional blindness. Oromandibular dystonia causes abnormal activity in lower facial, tongue, jaw, and pharyngeal muscles that can interfere with speaking or swallowing. Spasmodic dysphonia is dystonia of the vocal cords; abnormal adduction, which causes a strained, strangled voice, is more common than abduction, in which the voice sounds whispering and breathy. Brachial dystonia is another form of focal dystonia that can be primarily present with writing (writer’s cramp). Writer’s cramp may spread from the dominant to the contralateral arm in 15% of patients. Similar problems can also occur in pianists and string musicians. Such ‘‘task specific focal dystonia’’
86
can affect a variety of complex highly learned skills. The dystonia-plus syndromes are disorders associated with other neurologic signs such as parkinsonism in dopa-responsive dystonia and myoclonus in myoclonus–dystonia [85,86]. Dopa-responsive dystonia (DRD) is an autosomal dominant disorder, which usually presents in early childhood with foot dystonia, gait abnormality and hyperreflexia most often due to mutations in the gene that codes for GTP cyclohydrodase I. Initial symptoms may be followed by progressive generalized dystonia [87]. Diurnal fluctuation with worsening of symptoms later in the day is a unique feature. A small proportion of patients present with adult-onset parkinsonism. The hallmark of this disorder is a dramatic, sustained response to low to moderate doses of levodopa and generally without the development of motor complications seen in PD. Myoclonus–dystonia (DYT11) is a rare autosomal dominant disorder often caused by a mutation in the gene encoding e-sarcoglycan [88]. Myoclonus is usually the more predominant component and dystonia is usually mild and sometimes absent. When present, dystonia begins in childhood or adolescence and affects the arms, trunk, and bulbar muscles. Both myoclonus and dystonia improve with the ingestion of alcohol [89] (see myoclonus section below for more discussion of this disorder). Rapid-onsetdystonia–parkinsonism(DYT12) is a rare autosomal dominant disorder that begins in adolescence or young adulthood with the rapid appearance of dystonia and parkinsonism, usually followed by a plateau in symptoms and has no clinical response to levodopa [90]. Recently this has been found to be due to mutations in the Na+/K + ATPase a3 gene ATP1A3 [86]. Secondary dystonia includes heredodegenerative diseases with known neuropathological features, drug-induced dystonia, and dystonia caused by acquired structural abnormalities. Heredodegenerative disorders are heterogeneous
1453
1454
86
Surgery for movement disorders: an overview
group of degenerative, metabolic and genetic disorders with other neurological abnormalities [91]. One very important disorder that must be excluded when dealing with dystonia in children and young adults is Wilson’s disease. Acute drug-induced dystonia is caused by levodopa, dopamine agonists, antipsychotic drugs, anticonvulsant agents, serotonin-reuptake inhibitors, and rarely by other miscellaneous drugs. On the other hand, persistent tardive dystonia may occur after prolonged use of dopamine-receptor-blocking antipsychotic drugs and metoclopramide. Acquired brain lesions may produce either hemidystonia or focal limb dystonia. A hemidystonia pattern should always suggest this possibility and brain imaging is frequently abnormal. Lesions of the basal ganglia involving the putamen and thalamus are particularly common [92] and occur after perinatal injury, kernicterus, infarcts, hemorrhage, infection, trauma, anoxia, multiple sclerosis, and brain tumors [93]. Peripheral trauma is sometimes followed by dystonic postures in the injured body part. Most often this is a tonic dystonia present at rest typically accompanied by considerable pain and, in some, additional features of ‘‘reflex sympathetic dystrophy’’ are present. This is a controversial disorder; some groups feel that this is a well defined ‘‘organic’’ condition while others feel that psychological factors play a major role in the genesis of the symptoms. Paroxysmal dystonias (loci DYT8 through DYT10) are characterized by episodic dystonia and other involuntary movements (generally choreoathetosis) typically without symptoms or neurologic findings between episodes. Their relation to other dystonias is uncertain since they overlap clinically with other episodic disorders such as epilepsy, migraine and episodic ataxia and may be ion-channel disorders. They are broadly divided into brief kinesigenic dystonia precipitated by sudden movements that usually respond to anticonvulsant agents, more prolonged spontaneous (non-kinesigenic) dystonia that are more
resistant to treatment, exercise-induced dystonia, and mixed forms [94]. A variety of evaluation tools have been used to assess various types of dystonia. The most commonly used scale in assessment of cervical dystonia is the Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS) [95]. Another more recently reported scale designed to capture the burden of dystonia on patients is the Cervical Dystonia Impact Profile (CDIP-58) scale [96]. Other scales, such as the Burke-Fahn-Marsden scale (BFMS) and the Unified Dystonia Rating Scale (UDRS), have been used to assess patients with generalized dystonia [97–99]. Treatment options for dystonia include medical and surgical therapies. Medical treatment can be subdivided into oral drugs and local botulinum toxin injections, while surgical treatment may involve selective peripheral denervation or functional brain surgery. Oral medications are the first line therapy especially in patients with childhood and adolescent-onset dystonia. Patients with DRD dramatically respond to levodopa, and a levodopa trial should be performed in all early-onset dystonia patients. However, levodopa usually does not improve or may even worsen other forms of dystonia. Drugs frequently used in non-DRD dystonia include anticholinergics (trihexphenidyl), benzodiazepines (diazepam, clonazepam), dopamine depletors (tetrabenazine), gamma aminobutyric acidergics (baclofen), atypical neuroleptics (clozapine, olanzapine) and even typical neuroleptics (e.g., haloperidol) in very disabled patients. The first treatment of choice is usually anticholinergics with about 40 50% of dystonia patients showing a moderate response (high doses are usually required) [97,100], followed by baclofen [100,101]. Dopamine depletors such as tetrabenazine may be of particular benefit in tardive dystonia [102]. Most focal dystonias are relatively unresponsive to systemic drug treatment but are frequently well controlled by periodic botulinum toxin (BTX) injections into the muscles directly involved in the dystonia [103]. High doses and
Surgery for movement disorders: an overview
short intervals between injections may accelerate or increase the rate of antibody formation with the development of secondary non-responsiveness [104]. Patients refractory to BTX type A due to antibodies can now be treated with BTX type B [105]. If oral medications and botulinum toxin fail, in appropriately chosen, disabled patients intrathecal baclofen and surgery could be considered. Intrathecal baclofen may be especially useful for patients with secondary forms of dystonia who have additional spasticity. Until recently, unilateral or bilateral stereotactic radiofrequency ablations of the thalamus or the pallidum were the preferred surgical methods to treat patients with severe and otherwise medically refractory dystonia. Deep-brain stimulation of the GPi is now the preferred surgical treatment for dystonia because it has a lower risk of complications than does lesion surgery, and the stimulation parameters can be customized for each individual patient. It has been observed that the improvement of dystonia following DBS implants follows a specific sequence. Whilst dystonic movements (including phasic, myoclonic and tremulous features) may improve immediately or within hours or days after surgery, dystonic postures (i.e., tonic features) generally have a delayed improvement over weeks to months [106–108]. When pain is present (most often with cervical dystonia) it may also respond very early. Primary generalized dystonia seems to respond better than secondary dystonia to GPi-DBS, although uncontrolled evidence also suggests that patients with cervical dystonia [109], pantothenate kinase associated neurodegeneration [106], and tardive dystonia [110] also substantially improve. Apart from the GPi, the ventrolateral thalamus for secondary dystonia and subthalamic nucleus for primary dystonia have also been considered for surgical targets [111,112]. In addition to stereotactic procedures targeting thalamus and globus pallidus, peripheral denervation procedures have been used extensively
86
before the advent of botulinum toxin, especially for focal and segmental dystonias. Three procedures have been used previously in the treatment of cervical dystonia: extradural selective sectioning of posterior (dorsal) rami (posterior ramisectomy) combined with sectioning the spinal accessory nerve with or without myotomy; intradural sectioning of anterior cervical roots (anterior cervical rhizotomy) usually with spinal accessory nerve section; and microvascular decompression of the spinal accessory nerve. Surgical treatments, such as facial nerve lysis and orbicularis oculi myectomy, once used extensively in the treatment of blepharospasm, have been almost abandoned because botulinum-toxin treatment is very effective in most cases and without frequent postoperative complications such as ectropion, exposure keratitis, facial droop, and postoperative swelling and scarring [113]. Rarely, an extensive orbicularis oculi myectomy procedure is still used for patients with functional blindness due to botulinum toxin resistant blepharospasm. When this is combined with lower cranial and cervical involvement GPi DBS would be the preferred surgical treatment. Likewise, recurrent laryngeal nerve section, once used in the treatment of spasmodic dysphonia [114], is rarely used currently and only when botulinum toxin fails to provide a satisfactory relief. Another once popular procedure, namely spinal-cord stimulation for cervical dystonia, was ineffective in a controlled trial.
Chorea and Ballism Chorea refers to irregular, flowing, non-stereotyped, random, involuntary movements that often possess a writhing quality referred to as choreoathetosis. When chorea is proximal and of large amplitude, it is called ballism. Chorea is usually worsened by anxiety and stress and subsides during sleep. Most patients attempt
1455
1456
86
Surgery for movement disorders: an overview
to disguise chorea by incorporating it into a purposeful activity. Whereas ballism is most often encountered as hemiballism due to contralateral structural lesions of the subthalamic nucleus, the striatum or their connections, chorea may be the expression of a wide range of disorders, including drugs, metabolic, infectious, inflammatory, vascular, and neurodegenerative diseases. The causes of chorea and ballism are listed in > Tables 86-4 and > 86-5, respectively. In clinical practice, Huntington’s disease is an important cause of adult-onset chorea. Huntington’s disease is an autosomal dominantly inherited progressive neurodegenerative disorder. The mutant gene has been localized to chromosome 4p16.3 [115,116]. The gene product huntingtin is widely distributed in both neurones and extra neuronal tissues. The mutation in Huntington’s disease involves the expansion of a trinucleotide cytosine-adenosine-guanidine (CAG) repeat encoding glutamine. Expansion of the CAG repeat beyond the critical threshold of 36 repeats results in disease, and forms the basis of the polymerase chain reaction based genetic test. The pathogenesis of Huntington’s disease is yet unknown but increasing evidence suggests important role of altered gene transcription, mitochondrial dysfunction and excitotoxicity. The expanded polyglutamine stretch leads to a conformational change and abnormal proteinprotein interactions. Huntington’s disease is a progressive disabling neurodegenerative disorder characterized by the triad of movement disorders, dementia, and behavioral disturbances. Illness may emerge at any time of life, with the highest occurrence between 35 and 40 years of age. The involuntary choreiform movements are the hallmark of Huntington’s disease. However it is the mental alterations that often represent the most debilitating aspect of the disease and place the greatest burden on families of Huntington’s disease patients.
Chorea, in patients with Huntington’s disease, usually starts with slight movements of the fingers and toes and progresses to involve facial grimacing, eyelid elevations, and writhing limb movements. Another important associated feature is motor impersistence, whereby individuals are unable to maintain tongue protrusion, eyelid closure or a firm grip. Other common features include eye movement abnormalities (slowing of saccades and increased latency of response), rigidity, myoclonus, dysarthria and ataxia [117]. Dystonia tends to occur as the disease progresses or is associated with the use of dopamine antagonist medications. Parkinsonism is also common especially later in the course; some patients especially those with juvenile onset, have a parkinson-predominant phenotype (akineticrigid or Westphal variant). Dysphagia, which occurs prominently in the terminal stage, can result in fatal aspiration. Cognitive impairment is inevitable in all patients with Huntington’s disease [118,119] and typically begins as selective deficits involving psychomotor, executive, and visuospatial abilities and progresses to more global impairment, although higher cortical language tends to be spared. A wide range of psychiatric and behavioral disturbances are recognized in Huntington’s disease, with affective disorders among the most common [119]. Depression occurs up to 50% of patients. The suicide rate in Huntington’s disease is fivefold that of the general population [120]. Psychosis is also common, usually with paranoid delusions. Caregivers not uncommonly also report apathy and aggressive behavior. Current treatments in Huntington’s disease are largely symptomatic, aimed at reducing the motor and psychological dysfunction of the individual patient. The sections below could also pertain to the management of other neurodegenerative diseases manifesting chorea and other neuropsychiatric features (e.g., neuroacanthocytosis). In general, treatment of chorea is not recommended unless it is causing disabling functional
Surgery for movement disorders: an overview
. Table 86-4 Differential diagnosis of chorea (Adapted from [4]) Genetic Benign hereditary chorea Huntington’s disease Huntington-like conditions Neuroacanthocytosis DRPLA Wilson’s disease Pantothenate kinase-associated neurodegeneration Spinocerebellar ataxias Ataxia-telangiectasia Ataxia oculomotor apraxia I Infections/Parainfectious causes Sydenham’s chorea Acquired immunodeficiency syndrome Encephalitis and post-encephalitic Creutzfeldt-Jakob disease Drugs Levodopa, dopaminergic agonists, anticonvulsants, neuroleptics, antidepressants, amphetamines, anticholinergics, antihistamines,oral contraceptives Endocrinologic/Metabolic Hyperthyroidism Hypoparathyroidism Acquired hepatolenticular degeneration Chorea gravidarum Immunologic Systemic lupus erythematosus Henoch-Schonlein purpura Antiphospholipid syndrome Vascular Stroke Arterio-venous malformation Moya moya syndrome Polycythemia rubra vera Other Cardiopulmonary bypass with hypothermia Cerebral palsy Kernicterus Head trauma Neoplastic and paraneoplastic
or social impairment. Olanzapine or risperidone, atypical antipsychotics, have been found to reduce chorea with less risk of the extrapyramidal side effects, compared to the typical dopamine receptor blocking agents. Traditional neuroleptics such as haloperidol can improve chorea but are associated with increased risk of tardive dyskinesia, dystonia, difficulty swallowing, and gait
86
. Table 86-5 Differential diagnosis of ballism (Adapted from [4]) Focal lesions in basal ganglia Vascular Stroke (including infarction and hemorrhage) Cavernous angioma Post-surgical complications Neoplastic Metastases Primary CNS tumors Infections Toxoplasmosis Tuberculoma Cryptococcosis Inflammatory Multiple sclerosis Immunologic Systemic lupus erythematosus Scleroderma Behcet’s disease Iatrogenic Subthalamotomy Thalamotomy Non-ketotic hyperglycemia Hypoglycemia Syndenham’s chorea Head injury Drugs Anticonvulsants Levodopa Oral contraceptives
disturbances, and should not be considered first line agents. Tetrabenazine, a dopamine depletor may be very effective in suppressing chorea without the risk of tardive dyskinesia [121]. However drug-induced parkinsonism, akathisia, increased dysphagia and gait dysfunction as well as increased or de novo depression may occur with this drug and so patients should be followed closely for these features. Other agents including amantadine and possibly riluzole may also improve chorea [122]. The selective serotonin reuptake inhibitors (SSRIs) have become the first line agents in the treatment of depression in Huntington’s disease. In addition, SSRIs may suppress chorea and reduce aggression in Huntington’s disease [123]. The dose should be started low and could be doubled every two weeks
1457
1458
86
Surgery for movement disorders: an overview
if necessary. The atypical antipsychotic agents, such as clozapine, quetiapine, and olanzapine may be required to treat psychosis in Huntington’s disease [124]. Valproic acid may be useful in the long-term management of aggression and irritability [125]. In severely affected HD patients who have failed various medical treatments, surgery may be a possible option. There have been recent reports of bilateral pallidal stimulation in very carefully selected severely affected and medically intractable HD patients that produced a significant reduction in choreoathetoid movements, dystonia and improvement in overall motor functioning [126,127]. A number of groups have been investigating alternative approaches to the treatment of HD, including cell and tissue transplantation [128,129]. The aim of transplantation is to restore the neuronal circuitry and provide a substrate for functional restoration. To date, cells from the developing central nervous system (fetal brain, brainstem and spinal cord) are the only appropriate sources for transplantation. The available evidence suggests that human fetal striatal grafts can survive transplantation and may induce clinical benefits in patients with Huntington’s disease [128]. Functional neuroimaging studies have shown increased metabolic activity and small improvements in motor, cognitive, and behavioral measures in some patients [129]. This treatment approach is still experimental and information about long-term outcome is not yet available. A recent study of 2 patients demonstrated little long-term clinical benefit despite survival of transplanted fetal neurons from the ganglionic eminence, possibly due to a failure of integration into the host striatum [130]. Pallidal lesioning or DBS may also be considered for other disabling drug resistant choreic disorders especially when caused by a static or non-progressive cause. For example, rare patients with persistent hemiballism may be candidates for unilateral GPi surgery (DBS or lesions) while
patients with disabling tardive dyskinesia (usually with more prominent dystonic components) may improve considerably with bilateral GPi DBS.
Tics and Tourette’s Syndrome Tourette’s syndrome (TS) is characterized by multiple motor and vocal tics that wax and wane over time. Tics are sudden, brief, intermittent, involuntary or semivoluntary movements (motor tics) or sounds (phonic or vocal tics). Motor tics are stereotyped repetitive involuntary or semivoluntary movements that typically involve the face, head, and upper body. Phonic (vocal) tics are sounds such as sniffing, grunting, or barking that are associated with muscle contractions of the oropharynx and diaphragm. Tics may be simple (contraction of single or multiple muscles or simple unformed sounds) or complex (coordinated patterned movements or partial or full words). Coprolalia, the utterance of profanities is a widely known feature of TS however only a small proportion of patients ever experience this symptom. The etiological classification of tic disorders is presented in > Table 86-6. Most tics are primary or idiopathic meaning that they have no identifiable cause. Secondary tics are caused by a defined underlying brain disease or environmental factor. Tic severity is based on the frequency, intensity, and complexity of movements and sounds and can range from barely perceptible and infrequent to intense and nearly continuous. Tics may also change their anatomical location, pattern, severity, and complexity over time. Indeed waxing and waning in their nature and severity are characteristic features of tics in TS. Tics are influenced by environmental factors: stressful and exciting activities are associated with transient increases in tic severity, and relaxation calm, or focused activities are associated with transient reduction of tic severity. Simple tics of childhood are extremely common. The relationship
Surgery for movement disorders: an overview
. Table 86-6 Etiological classification of tics (Adapted from [4]) 1. Primary or idiopathic Transient motor or phonic tics Chronic motor or phonic tics Adult-onset tics Tourette Syndrome 2. Secondary Tics Genetic Neuroacanthocytosis Huntington’s disease Pantothenate kinase-associated neurodegeneration (PANK2 mutations) Tuberous sclerosis Chromosomal disorders Infections Sydenham’s chorea PANDASa Encephalitis and post-encephalitic Creutzfeldt-Jakob disease Neurosyphilis Drugs Methylphenidate, cocaine, amphetamines, levodopa, phenobarbital, carbamazepine, phenytoin, lamotrigine, neuroleptics Developmental Mental retardation Pervasive developmental disorders/Autism Other causes Head trauma Stroke Cardiopulmonary bypass with hypothermia Carbon monoxide poisoning a Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal infections – the existence of this disorder remains somewhat controversial
between these and TS, which comprises multiple motor and at least one phonic tic, is unclear. TS has important but poorly understood genetic determinants and some obligate carriers may manifest no symptoms, only behavioral features such as obsessive compulsive behavior (see below) or only simple transient tics of childhood. Tics in TS most often begin in the first decade of life, usually between 5 and 7 years of age. They wax and wane, with peak severity in early adolescence with a gradual decrease in severity into adulthood [131,132]. Remission of tics may occur in the third decade of life in up
86
to 50% of patients, but to date, there are no prognostic features that predict which patients will have a remission in their symptoms. A small percentage of patients have severe, disabling tics refractory to standard medical treatment that continue or even increase in adulthood. The Yale Global Tic Severity Scale (YGTSS) [133] is a clinician-completed rating scale used to rate tic severity along several dimensions based on parent and child reports and clinician observations during the interview. Each dimension is represented by a subscale designed to quantify the number, frequency, duration, intensity, and complexity of both motor and vocal tics. Each subscale includes several descriptions to help the clinician make his or her ratings. Guided by these descriptions, each subscale is issued a rating between 0 and 5, with higher scores indicating greater severity. The five subscales are rated separately for motor and vocal tics. The motor subscales are then summed to produce an overall motor tic severity rating, and the vocal tic subscales are summed to provide an overall vocal tic severity rating; each ranges from 0 to 25. The motor and vocal tic severity ratings are then summed to produce an overall tic severity score that ranges from 0 to 50. A recent report of the response of a patient with TS to functional surgery raised a concern that the YGTSS may be insufficiently responsive to important changes in severely affected patients [134]. A video assessment and rating of tic severity is also frequently used in the evaluation of experimental therapies [135]. Although tics are the defining symptoms in TS, many individuals with TS have other symptoms, including obsessive-compulsive behavior (OCB), attention deficit hyperactivity disorder (ADHD), anxiety, and mood disorders [136]. Up to 50% of patients with TS have OCB. Similarly, up to 50% of people with TS have ADHD symptoms. Smaller, but significant, percentages have anxiety, depression, or other affective symptoms and learning disorders are particularly common. A small proportion of patients may
1459
1460
86
Surgery for movement disorders: an overview
have self-injurious behavior either as a part of or separate from their tics. Identification of these comorbid symptoms is critical as they usually contribute more to the patient’s impairment and disability than the tics themselves. Most tics do not require drug treatment. Education including the family members, teachers and friends is a critical component of optimal patient care. Comorbid symptoms such as ADHD and OCB may require treatment in their own right. Behavioral modification, special education and other approaches are often required for patients with more severe problems. When the tics per se are a source of physical or emotional disability they require treatment. The standard treatment is pharmacologic involving mainly neuroleptics, a2adrenergic agonists and sometimes benzodiazepines. In some cases, behavioral treatment may provide temporary control of symptoms. A small proportion of patients have disabling, medically refractory tics or experience unbearable side effects from the medication. It is these patients who have been potential candidates for neurosurgical interventions. Many different ablative surgeries have been carried out in TS throughout the history of surgical treatment of this disorder. Frontal lobe operations have included prefrontal lobotomies and bimedial frontal leucotomies. The limbic system has been targeted using limbic leucotomy and anterior cingulotomy. Thalamic operations have included lesioning of the medial, intralaminar, and ventrolateral thalamic nuclei. Infrathalamic lesions have been carried out at the level of Forel’s fields (campotomies) and the zona incerta, and cerebellar surgery included dentatotomies. In an attempt to achieve total control of symptoms, more complex operations have been carried out, such as combined anterior cingulotomies and infrathalamic lesions. DBS for TS is still in its early stages but results from preliminary trials appear very promising. Since the first DBS procedures were carried
out in 1999 for intractable TS, there have been several reports of dramatic clinical improvement in a small number of TS patients treated with thalamic and pallidal stimulation. This has brought marked enthusiasm and interest concerning the application of DBS in TS [137–139]. Given the complexity of TS and concerns about the inappropriate selection of patients a working group has published recommendations for the selection and evaluation of patients [140] which should be carefully considered if patients are being proposed for this still experimental procedure.
Myoclonus Myoclonus comprises sudden, brief, shock-like, involuntary movements resulting from both active muscle contraction and inhibition of ongoing muscle activity, positive or negative myoclonus, respectively. > Table 86-7 shows the etiological classification of myoclonus. Myoclonus can also be classified electrophysiologically, based on the origin of the jerks in the cerebral cortex (cortical), brainstem (reticular), spinal cord (spinal; segmental or propriospinal) and peripheral nerves (rare). For the purpose of this chapter only essential myoclonus will be discussed. In essential myoclonus, the myoclonus is the most prominent or the only clinical finding. Sporadic and hereditary forms of this type of myoclonus exist. Sporadic essential myoclonus probably has various heterogeneous, yet undiscovered, causes, and many patients may have false negative family histories. Hereditary essential myoclonus has been clinically characterized by the following: onset before age 20 years; dominant inheritance with variable severity; a benign course compatible with an active life and normal longevity; and absence of cerebellar ataxia, spasticity, dementia, and seizures [141]. The myoclonus is usually distributed throughout
Surgery for movement disorders: an overview
. Table 86-7 Classification and causes of myoclonus (Adapted from [4]) 1. Physiologic myoclonus Sleep myoclonus Anxiety-induced myoclonus Exercise-induced myoclonus Hiccups Benign infantile myoclonus during feeding 2. Essential myoclonus Essential myoclonusa Hereditary Sporadic Myoclonus-dystoniaa 3. Epileptic myoclonus Fragments of epilepsy Isolated epileptic myoclonic jerks Photosensitive myoclonus Myoclonic absences Epilepsia partialis continua Idiopathic stimulus-sensitive myoclonus Childhood myoclonic epilepsies Infantile spasms Lennox-Gastaut syndrome Cryptogenic myoclonus epilepsy Juvenile myoclonic epilepsy of Janz Benign familial myoclonic epilepsy Baltic Myoclonus (Unverricht-Lundborg) 4. Symptomatic myoclonus Toxic Bismuth Heavy-metal poisoning Methyl bromide, DDT Drugs (multiple) Metabolic Hyponatremia Hypoglycemia Nonketotic hyperglycemia Hepatic failure Renal failure Dialysis dysequilibrium syndrome Infantile myoclonic encephalopathy Multiple carboxylase deficiency Biotin deficiency Storage disease Lafora body disease Lipidoses Neuronal ceroid lipufuscinosis Sialidosis Mitochondrial cytopathies Spinocerebellar degeneration Friedreich’s ataxia Ataxia-telangiectasia
86
. Table 86-7 (Continued) Other spinocerebellar degenerations Basal ganglia degenerations Parkinson’s disease Progressive supranuclear palsy Corticobasal degeneration Multiple system atrophy Huntington’s disease Wilson’s disease DRPLA Dementias Alzheimer’s disease Creutzfeldt-Jakob disease Dementia with Lewy bodies Viral encephalopathies HIV Subacute sclerosing panencephalitis Encephalitis lethargica Herpes Simplex encephalitis Arbovirus encephalitis Postinfectious encephalitis Physical encephalopathies Post-hypoxic myoclonus (Lance-Adams) Post-traumatic Heat stroke Electric shock Decompression injury Focal CNS damage Post-thalamotomy Stroke Spinal cord lesions Tumor Trauma Paraneoplastic syndromes Psychogenic myoclonus a
Probably represents the same entity
the upper body, exacerbated by muscle activation, and substantially decreased with alcohol ingestion. The term myoclonus-dystonia syndrome has been introduced because of the common occurrence of dystonia in these cases and this is now more commonly classified as a form of ‘‘dystonia-plus’’ (see above), although myoclonus may be the only clinical feature. Recent progress in the genetics of the myoclonus-dystonia syndrome has provided insights into pathogenesis and added information to the clinical picture. Mutations in the
1461
1462
86
Surgery for movement disorders: an overview
e-sarcoglycan gene have had the strongest association with the myoclonus-dystonia syndrome [142]. By contrast to the other types of sarcoglycan, e-sarcoglycan is expressed in the brain and other non-muscle tissues, but its function is unknown. Most mutations are loss-of-function mutations and maternal imprinting has been associated with reduced penetrance [143]. Exactly how the loss of function of e-sarcoglycan results in myoclonus or dystonia is not known, but e-sarcoglycan dysfunction interferes with normal GABA inhibitory receptors. This may explain the striking benefit obtained with ethanol mentioned earlier. In essential myoclonus (including myoclonus-dystonia), pharmacological treatment generally does not match the amelioration seen with alcohol, and as a result there is a real danger of alcoholism in this disorder. Separate from alcoholism (and possibly contributing to it) myoclonus-dystonia may also be associated with other behavioral/psychiatric disturbances such as depression, anxiety and OCB. Treatment with anticholinergic drugs may help both myoclonus and dystonia in about a third of the cases, and diazepam or clonazepam may relieve the jerks to some extent, with little effect on dystonia. There have been only few reports on the success of surgery in essential myoclonus or myoclonus-dystonia patients. The myoclonic symptoms may be improved with VIM lesioning or DBS [144]. On the other hand, GPi stimulation may have beneficial effects on dystonia as well as myoclonus [145,146].
References 1. Lang AE, Lozano AM. Parkinson’s disease: first of two parts. N Engl J Med 1998;339:1044-53. 2. Uc EY, Struck LK, Rodnitzky RL, Zimmerman B, Dobson J, Evans WJ. Predictors of weight loss in Parkinson’s disease. Mov Disord 2006;21:930-6. 3. Uc EY, Rizzo M, Anderson SW, Qian S, Rodnitzky RL, Dawson JD. Visual dysfunction in Parkinson disease without dementia. Neurology 2005;65:1907-13.
4. Cloutier M, Lang AE. Movement disorders: an overview. In: Factor SA, Lang AE, Weiner WJ, editors. Drug induced movement disorders. Malden, MA: Blackwell; 2005. p. 3-19. 5. Volkmann J. Deep brain stimulation for the treatment of Parkinson’s disease. J Clin Neurophysiol 2004; 21:6-17. 6. Koller WC, Pahwa R, Lyons KE, Albanese A. Surgical treatment of Parkinson’s disease. J Neurol Sci 1999; 167:1-10. 7. Lyons KE, Pahwa R. Deep brain stimulation in Parkinson’s disease. Curr Neurol Neurosci Rep 2004;4: 290-5. 8. Spiegel EA, Wycis HT. Thalamotomy and pallidotomy for treatment of choreic movements. Acta Neurochir (Wein) 1952;2:417-22. 9. Deuschl G, Schade-Brittinger C, Krack P, Volkmann J, Schafer H, Botzel K, et al. A randomized trial of deepbrain stimulation for Parkinson’s disease. N Engl J Med 2006;355:896-908. 10. Russmann H, Ghika J, Combrement P, Villemure JG, Bogousslavsky J, Burkhard PR, et al. L-Dopa-induced dyskinesia improvement after STN-DBS depends upon medication reduction. Neurology 2004;63:153-5. 11. Nandi D, Aziz TZ, Giladi N, Winter J, Stein JF. Reversal of akinesia in experimental parkinsonism by GABA antagonist microinjections in the pedunculopontine nucleus. Brain 2002;125:2418-30. 12. Munro-Davies LE, Winter J, Aziz TZ, Stein JF. The role of the pedunculopontine region in basalganglia mechanisms of akinesia. Exp Brain Res 1999; 129:511-7. 13. Jordan LM. Initiation of locomotion in mammals. Ann NY Acad Sci 1998;860:83-93. 14. Delwaide PJ. Parkinsonian rigidity. Funct Neurol 2001;16:147-56. 15. Pahapill PA, Lozano AM. The pedunculopontine nucleus and Parkinson’s disease. Brain 2000;123:1767-83. 16. Mazzone P, Lozano A, Stanzione P, Galati S, Scarnati E, Peppe A, et al. Implantation of human pedunculopontine nucleus: a safe and clinically relevant target in Parkinson’s disease. Neuroreport 2005;16:1877-81. 17. Plaha P, Ben-Shlomo Y, Patel NK, Gill SS. Stimulation of the caudal zona incerta is superior to stimulation of the subthalamic nucleus in improving contralateral parkinsonism. Brain 2006;129:1732-47. 18. Goto S, Kunitoku N, Soyama N, Yamada K, Okamura A, Yoshikawa M, et al. Posteroventral pallidotomy in a patient caused by hypoxic encephalopathy. Neurology 1997;49:707-10. 19. Krauss JK, Jankovic J, Lai EC, Rettig GM, Grossman RG. Posteroventral medial pallidotomy in levodopa-unresponsive parkinsonism. Arch Neurol 1997;54:1026-9. 20. Tarsy D, Apetauerova D, Ryan P, Norregaard T. Adverse effects of subthalamic nucleus DBS in a patient with multiple system atrophy. Neurology 2003;61:247-9.
Surgery for movement disorders: an overview
21. Lezcano E, Gomez-Esteban JC, Zarranz JJ, Alcaraz R, Atares B, Bilbao G, et al. Parkinson’s disease-like presentation of multiple system atrophy with poor response to STN stimulation: a clinicopathological case report. Mov Disord 2004;19:973-7. 22. Visser-Vandewalle V, Temel Y, Colle H, van der Linden C. Bilateral high-frequency stimulation of the subthalamic nucleus in patients with multiple system atrophyparkinsonism. Report of four cases. J Neurosurg 2003; 98:882-7. 23. Lang AE, Lozano A, Duff J, Tasker R, Miyasaki J, Jimenez NG, et al. Medial pallidotomy in late-stage parkinson’s disease and striatonigral degeneration. In: Obeso JA, DeLong MR, Ohye C, Marsden CD, editors. Advances in Neurology, vol. 74. Philadelphia: Lippincort-Raven; 1997. p. 199-211. 24. Louis ED. Essential tremor. N Engl J Med 2001;345: 887-91. 25. Elble RJ. Diagnostic criteria for essential tremor and differential diagnosis. Neurology 2000;54:S2-6. 26. Elble RJ. Essential tremor frequency decreases with time. Neurology 2000;55:1547-51. 27. Bain PG. The management of tremor. J Neurol Neurosurg Psychiatry 2002;72 Suppl 1:I3-I9. 28. Alusi SH, Glickman S, Aziz TZ, Bain PG. Tremor in multiple sclerosis. J Neurol Neurosurg Psychiatry 1999; 66:131-4. 29. Geny C, Nguyen JP, Pollin B, Feve A, Ricolfi F, Cesaro P, et al. Improvement of severe postural cerebellar tremor in multiple sclerosis by chronic thalamic stimulation. Mov Disord 1996;11:489-94. 30. Hooper J, Taylor R, Pentland B, Whittle IR. A prospective study of thalamic deep brain stimulation for the treatment of movement disorders in multiple sclerosis. Br J Neurosurg 2002;16:102-9. 31. Montgomery EB, Jr, Baker KB, Kinkel RP, Barnett G. Chronic thalamic stimulation for the tremor of multiple sclerosis. Neurology 1999;53:625-8. 32. Schulder M, Sernas T, Mahalick D, Adler R, Cook S. Thalamic stimulation in patients with multiple sclerosis. Stereotact Funct Neurosurg 1999;72:196-201. 33. Deuschl G. New treatment options for tremors. N Engl J Med 2000;342:505-7. 34. Deuschl G, Bain P, Brin M. Consensus statement of the Movement Disorder Society on Tremor. Ad Hoc Scientific Committee. Mov Disord 1998;13 Suppl 3:2-23. 35. Fahn S, Tolosa E, Concepcion M. Clinical rating scale for tremor. In: Jankovic J, Tolosa E, editors. Parkinson’s disease and movement disorders, 2nd ed. Baltimore: Williams and Wilkins; 1993. p. 271-80. 36. LorenzD, Deuschl G. Update on pathogenesis and treatment of essential tremor. Curr Opin Neurol 2007;20:447-52. 37. Larsen TA, Calne DB. Essential tremor. Clin Neuropharmacol 1983;6:185-206. 38. Dietrichson P, Espen E. Effects of timolol and atenolol on benign essential tremor: placebo-controlled studies based
39.
40.
41. 42.
43.
44.
45.
46. 47.
48.
49. 50.
51.
52.
53.
54.
55.
86
on quantitative tremor recording. J Neurol Neurosurg Psychiatry 1981;44:677-83. Koller WC. Dose-response relationship of propranolol in the treatment of essential tremor. Arch Neurol 1986;43:42-3. Cleeves L, Findley LJ. Propranolol and propranolol-LA in essential tremor: a double blind comparative study. J Neurol Neurosurg Psychiatry 1988;51:379-84. Young RR. Essential-familial tremor and other action tremors. Semin Neurol 1982;2:386-91. Jefferson D, Jenner P, Marsden CD. Beta-adrenoreceptor antagonists in essential tremor. J Neurol Neurosurg Psychiatry 1979;42:904-9. Connor GS. A double-blind placebo-controlled trial of topiramate treatment for essential tremor. Neurology 2002;59:132-4. Ondo W, Hunter C, Vuong KD, Schwartz K, Jankovic J. Gabapentin for essential tremor: a multiple-dose, doubleblind, placebo-controlled trial. Mov Disord 2000;15: 678-82. Brin MF, Lyons KE, Doucette J, Adler CH, Caviness JN, Comella CL, et al. A randomized, double masked, controlled trial of botulinum toxin type A in essential hand tremor. Neurology 2001;56:1523-8. Sandyk R. Successful treatment of cerebellar tremor with clonazepam. Clin Pharm 1985;4:615-8. Trelles L, Trelles JO, Castro C, Altamirano J, Benzaquen M. Successful treatment of two cases of intention tremor with clonazepam. Ann Neurol 1984;16:621. Shepard GMG, Tauboll E, Bakke SJ, Nyberg-Hansen R. Midbrain tremor and hypertrophic olivary degeneration after pontine hemorrhage. Mov Disord 1997;12:432-7. Biary N, Cleeves L, Findley L, Koller W. Post-traumatic tremor. Neurology 1989;39:103-6. Remy P, de Rocondo A, Defer G, Loc’h C, Amarenco P, Plante-Bordeneuve V, et al. Peduncular ‘‘rubral’’ tremor and dopaminergic denervation: a PET study. Neurology 1995;45:472-7. Bala VM. Uncommon forms of tremor. In: Watts RL, Koller WC, editors. Movement disorders. Neurologic principles and practice. New York: McGraw-Hill; 1997. p. 391-2. Riechert T, Richter D. Surgical treatment of multiple sclerosis tremor and essential tremor. Munch Med Wochenschr 1972;114:2025-8. Hirai T, Miyazaki M, Nakajima H, Shibazaki T, Ohye C. The correlation between tremor characteristics and the predicted volume of effective lesions in stereotaxic nucleus ventralis intermedius thalamotomy. Brain 1983;106:1001-18. Nagaseki Y, Shibazaki T, Hirai T, Kawashima Y, Hirato M, Wada H, et al. Long-term follow-up results of selective VIM-thalamotomy. J Neurosurg 1986;65: 296-302. Mohadjer M, Goerke H, Milios E, Etou A, Mundinger F. Longterm results of stereotaxy in the treatment of essential tremor. Stereotact Funct Neurosurg 1990;55:125-9.
1463
1464
86
Surgery for movement disorders: an overview
56. Lakie M, Arblaster LA, Roberts RC, Varma TR. Effect of stereotactic thalamic lesion on essential tremor. Lancet 1992;340:206-7. 57. Goldman MS, Ahlskog JE, Kelly PJ. The symptomatic and functional outcome of stereotactic thalamotomy for medically intractable essential tremor. J Neurosurg 1992;76:924-8. 58. Jankovic J, Cardoso F, Grossman RG, Hamilton WJ. Outcome after stereotactic thalamotomy for parkinsonian, essential, and other types of tremor. Neurosurgery 1995;37:680-6. 59. Benabid AL, Pollak P, Seigneuret E, Hoffmann D, Gay E, Perret J. Chronic VIM thalamic stimulation in Parkinson’s disease, essential tremor and extrapyramidal dyskinesias. Acta Neurochir Suppl (Wien) 1993;58:39-44. 60. Benabid AL, Pollak P, Gao D, Hoffmann D, Limousin P, Gay E, et al. Chronic electrical stimulation of the ventralis intermedius nucleus of the thalamus as a treatment of movement disorders. J Neurosurg 1996;84:203-14. 61. Koller W, Pahwa R, Busenbark K, Hubble J, Wilkinson S, Lang A, et al. High-frequency unilateral thalamic stimulation in the treatment of essential and parkinsonian tremor. Ann Neurol 1997;42:292-9. 62. Benabid AL, Benazzouz A, Hoffmann D, Limousin P, Krack P, Pollak P. Long-term electrical inhibition of deep brain targets in movement disorders. Mov Disord 1998;13:119-25. 63. Limousin P, Speelman JD, Gielen F, Janssens M. Multicentre European study of thalamic stimulation in parkinsonian and essential tremor. J Neurol Neurosurg Psychiatry 1999;66:289-96. 64. Taha JM, Janszen MA, Favre J. Thalamic deep brain stimulation for the treatment of head, voice, and bilateral limb tremor. J Neurosurg 1999;91:68-72. 65. Koller WC, Lyons KE, Wilkinson SB, Pahwa R. Efficacy of unilateral deep brain stimulation of the VIM nucleus of the thalamus for essential head tremor. Mov Disord 1999;14:847-50. 66. Troster AI, Fields JA, Pahwa R, Wilkinson SB, StraitTroster KA, Lyons K, et al. Neuropsychological and quality of life outcome after thalamic stimulation for essential tremor. Neurology 1999;53:1774-80. 67. Pahwa R, Lyons KL, Wilkinson SB, Carpenter MA, Troster AI, Searl JP, et al. Bilateral thalamic stimulation for the treatment of essential tremor. Neurology 1999; 53:1447-50. 68. Herzog J, Hamel W, Wenzelburger R, Po¨tter M, Pinsker MO, Bartussek J, et al. Kinematic analysis of thalamic versus subthalamic neurostimulation in postural and intention tremor. Brain 2007;130:1608-25. 69. van Manen J. Stereotaxic operations in cases of hereditary and intention tremor. Acta Neurochir 1974;21:49-55. 70. Goldman MS, Kelly PJ. Symptomatic and functional outcome of stereotactic ventralis lateralis thalamotomy for intention tremor. J Neurosurg 1992;77:223-9.
71. Speelman JD, Schuurman PR, de Bie RM, Bosch DA. Thalamic surgery and tremor. Mov Disord 1998;13:103-6. 72. Wishart HA, Roberts DW, Roth RM, McDonald BC, Coffey DJ, Mamourian AC, et al. Chronic deep brain stimulation for the treatment of tremor in multiple sclerosis: review and case reports. J Neurol Neurosurg Psychiatry 2003;74:1392-7. 73. Nandi D, Aziz TZ. Deep brain stimulation in the management of neuropathic pain and multiple sclerosis tremor. J Clin Neurophysiol 2004;21:31-9 74. Miyagi Y, Shima F, Ishido K, Moriguchi M, Kamikaseda K. Postereoventral pallidotomy for midbrain tremor after a pontine hemorrhage. J Neurosurg 1999;91:885-8. 75. Samie MR, Selhorst JB, Koller WC. Post-traumatic midbrain tremors. Neurology 1990;40:62-6. 76. Krauss JK, Mohadjer M, Nobbe F, Mundinger F. The treatment of posttraumatic tremor by stereotactic surgery: symptomatic and functional outcome in a series of 35 patients. J Neurosurg 1994;80:810-9. 77. Andrew J, Fowler CJ, Harrison MJG. Tremor after head injury and its treatment by stereotaxic surgery. J Neurol Neurosurg Psychiatry 1982;45:815-9. 78. Fahn S, Marsden CD, Calne DB. Classification and investigation of dystonia. In: Marsden CD, Fahn S, editors. Movement disorders 2. London: Butterworths; 1987. p. 332-58. 79. Bressman SB. Dystonia genotypes, phenotypes, and classification. Adv Neurol 2004;94:101-7. 80. Fahn S, Bressman SB, Marsden CD. Classification of dystonia. Adv Neurol 1998;78:1-10 81. Bressman SB, Sabatti C, Raymond D, de Leon D, Klein C, Kramer PL, et al. The DYT1 phenotype and guidelines for diagnostic testing. Neurology 2000;54: 1746-52. 82. Bressman SB, de Leon D, Kramer PL, Ozelius LJ, Brin MF, Greene PE, et al. Dystonia in Ashkenazi Jews: clinical characterization of a founder mutation. Ann Neurol 1994;36:771-7. 83. Edwards M, Wood N, Bhatia K. Unusual phenotypes in DYT1 dystonia: a report of five cases and a review of the literature. Mov Disord 2003;18:706-11. 84. Greene P, Kang UJ, Fahn S. Spread of symptoms in idiopathic torsion dystonia. Mov Disord 1995;10:143-52. 85. Bressman SB. Dystonia: phenotypes and genotypes. Rev Neurol (Paris) 2003;159:849-56. 86. de Carvalho Aguiar P, Sweadner KJ, Penniston JT, Zaremba J, Liu L, Caton M, et al. Mutations in the Na+/K+-ATPase alpha3 gene ATP1A3 are associated with rapid-onset dystonia parkinsonism. Neuron 2004; 43:169-75. 87. Nygaard TG, Marsden CD, Fahn S. Dopa-responsive dystonia: long-term treatment response and prognosis. Neurology 1991;41:174-81. 88. Zimprich A, Grabowski M, Asmus F, Naumann M, Berg D, Bertram M, et al. Mutations in the gene encoding
Surgery for movement disorders: an overview
epsilon-sarcoglycan cause myoclonus-dystonia syndrome. Nat Genet 2001;29:66-9. 89. Quinn NP. Essential myoclonus and myoclonic dystonia. Mov Disord 1996;11:119-24. 90. Dobyns WB, Ozelius LJ, Kramer PL, Brashear A, Farlow MR, Perry TR, et al. Rapid-onset dystoniaparkinsonism. Neurology 1993;43:2596-602. 91. Friedman J, Standaert DG. Dystonia and its disorders. Neurol Clin 2001;19:681-705. 92. Marsden CD, Obeso JA, Zarranz JJ, Lang AE. The anatomical basis of symptomatic hemidystonia. Brain 1985;108:463-83. 93. Calne DB, Lang AE. Secondary dystonia. Adv Neurol 1988;50:9-33. 94. Demirkiran M, Jankovic J. Paroxysmal dyskinesias: clinical features and classification. Ann Neurol 1995; 38:571-9. 95. Consky ES, Lang AE. Clinical assessments of patients with cervical dystonia. In: Jankovic J, Hallett M, editors. Therapy with botulinum toxin. New York: Marcel Dekker; 1994. p. 211-37. 96. Cano SJ, Warner TT, Linacre JM, Bhatia KP, Thompson AJ, Fitzpatrick R, et al. Capturing the true burden of dystonia on patients: the Cervical Dystonia Impact Profile (CDIP-58). Neurology 2004;63: 1629-33. 97. Burke RE, Fahn S, Marsden CD. Torsion dystonia: a double-blind, prospective trial of high-dosage trihexyphenidyl. Neurology 1986;36:160-4. 98. Ondo WG, Desaloms M, Jankovic J, Grossman R. Surgical pallidotomy for the treatment of generalized dystonia. Mov Disord 1998;13:693-8. 99. Comella CL, Leurgans S, Wuu J, Stebbins GT, Chmura T. Dystonia Study Group, Rating scales for dystonia: a multicenter assessment. Mov Disord 2003;18: 303-12. 100. Greene PE, Shale H, Fahn S. Analysis of open-label trials in torsion dystonia using high dosages of anticholinergics and other drugs. Mov Disord 1988;3:46-60. 101. Greene PE, Fahn S. Baclofen in the treatment of idiopathic dystonia in children. Mov Disord 1992;7: 48-52. 102. Jankovic J, Orman J. Tetrabenazine treatment in dystonia, chorea, tics, and other dyskinesias. Neurology 1988;38:391-4. 103. Beradelli A, Curra A. Pathophysiology and treatment of cranial dystonia. Mov Disord 2002;17 Suppl 2:S70-S74. 104. Dauer WT, Burke RE, Greene P, Fahn S. Current concepts on the clinical features, aetiology and management of idiopathic cervical dystonia. Brain 1998;121:547-60. 105. Brashear A, Lew MF, Dykstra DD, Comella CL, Factor SA, Rodnitzky RL, et al. Safety and efficacy of NeuroBloc (botulinum toxin type B) in type A-responsive cervical dystonia. Neurology 1999;53:1439-46. 106. Vidailhet M, Vercueil L, Houeto JL, Krystkowiak P, Benabid AL, Cornu P, et al. Bilateral deep-brain
107.
108.
109.
110.
111.
112.
113.
114.
115.
116. 117. 118.
119.
120.
121.
86
stimulation of the globus pallidus in primary generalized dystonia. N Engl J Med 2005;352:459-67. Coubes P, Cif L, El Fertit H, Hemm S, Vayssiere N, Serrat S, et al. Electrical stimulation of the globus pallidus internus in patients with primary generalized dystonia: long-term results. J Neurosurg 2004;101:189-94. Yianni J, Bain P, Giladi N, Auca M, Gregory R, Joint C, et al. Globus pallidus internus deep brain stimulation for dystonic conditions: a prospective audit. Mov Disord 2003;18:436-42. Krauss JK, Loher TJ, Pohle T, Weber S, Taub E, Barlocher CB, et al. Pallidal deep brain stimulation in patients with cervical dystonia and severe cervical dyskinesias with cervical myelopathy. J Neurol Neurosurg Psychiatry 2002;72:249-56. Trottenberg T, Volkmann J, Deuschl G, Kuhn AA, Schneider GH, Muller J, et al. Treatment of severe tardive dystonia with pallidal deep brain stimulation. Neurology 2005;64:344-6. Vercueil L, Pollak P, Fraix V, Caputo E, Moro E, Benazzouz A, et al. Deep brain stimulation in the treatment of severe dystonia. J Neurol. 2001;248:695-700. Sun B, Chen S, Zhan S, Le W, Krahl SE. Subthalamic nucleus stimulation for primary dystonia and tardive dystonia. Acta Neurochir Suppl. 2007;97:207-14. Chapman KL, Bartley GB, Waller RR, Hodge DO. Follow-up of patients with essential blepharospasm who underwent eyelid protractor myectomy at the Mayo Clinic from 1980 through 1995. Ophthal Plast Reconstr Surg 1999;15:106-10. Dedo HH, Izdebski K. Intermediate results of 306 recurrent laryngeal nerve sections for spastic dysphonia. Laryngoscope 1983;93:9-15. Gusella JF, Wexler NS, Conneally PM, Naylor SL, Anderson MA, Tanzi RE, et al. A polymorphic DNA marker genetically linked to Huntington’s disease. Nature 1983;306:234-8. Conneally PM. Huntington’s disease: genetics and epidemiology. Am J Hum Genet 1984;36:506-26. Lasker AG, Zee DS. Ocular motor abnormalities in Huntington’s disease. Vis Res 1997;37:3639-45. Hahn-Barma V, Deweer B, Durr A, Dode C, Feingold J, Pillon B, et al. Are cognitive changes the first symptoms of Huntington’s disease? A study of gene carriers. J Neurol Neurosurg Psychiatry 1998;64:172-7. Naarding P, Kremer HP, Zitman FG. Huntington’s disease: a review of the literature on prevalence and treatment of neuropsychiatric phenomena. Eur Psychiatry 2001;16:439-45. Schoenfeld M, Myers RH, Cupples LA, Berkman B, Sax DS, Clark E. Increased rate of suicide among patients with Huntington’s disease. J Neurol Neurosurg Psychiatry 1984;47:1283-7. Ondo WG, Tintner R, Thomas M, Jankovic J. Tetrabenazine treatment for Huntington’s disease-associated chorea. Clin Neuropharmacol 2002;25:300-2.
1465
1466
86
Surgery for movement disorders: an overview
122. Rosas HD, Koroshetz WJ, Jenkins BG, Chen YI, Hayden DL, Beal MF, et al. Riluzole therapy in Huntington’s disease (HD). Mov Disord 1999;14: 326-30. 123. Fava M. Psychopharmacologic treatment of pathologic aggression. Psychiatr Clin North Am 1997;20:427-51. 124. Bonuccelli U, Ceravolo R, Maremmani C, Nuti A, Rossi G, Muratorio A. Clozapine in Huntington’s chorea. Neurology 1994;44:821-3. 125. Grove VE, Quintanilla J, De Vaney GT. Improvement of Huntington’s disease with olanzapine and valproate. N Engl J Med 2000;343:973-4. 126. Moro E, Lang AE, Strafella AP, Poon YY, Arango PM, Dagher A, et al. Bilateral globus pallidus stimulation for Huntington’s disease. Ann Neurol. 2004;56:290-4. 127. Hebb MO, Garcia R, Gaudet P, Mendez IM. Bilateral stimulation of the globus pallidus internus to treat choreathetosis in Huntington’s disease: technical case report. Neurosurgery 2006;58:E383. 128. Hauser RA, Furtado S, Cimino CR, Delgado H, Eichler S, Schwartz S, et al. Bilateral human fetal striatal transplantation in Huntington’s disease. Neurology 2002;58:687-95. 129. Bachoud-Levi AC, Gaura V, Brugieres P, Lefaucheur JP, Boisse MF, Maison P, et al. Effect of fetal neural transplants in patients with Huntington’s disease 6 years after surgery: a long-term follow-up study. Lancet Neurol 2006;5:303-9. 130. Keene CD, Sonnen JA, Swanson PD, Kopyov O, Leverenz JB, Bird TD, et al. Neural transplantation in Huntington disease: long-term grafts in two patients. Neurology 2007;68:2093-8. 131. Leckman J, Zhang H, Vitale A, Lahnin F, Lynch K, Bondi C, et al. Course of tic severity in Tourette syndrome: the first two decades. Pediatrics 1998;102:14-9. 132. Erenberg G, Cruse R, Rothner A. The natural history of Tourette syndrome: a follow-up study. Ann Neurol 1987;22:383-5. 133. Leckman JF, Riddle MA, Hardin MT, Ort SI, Swartz KL, Stevenson J, et al. The Yale Global Tic Severity Scale: initial testing of a clinician-rated scale of tic severity. J Am Acad Child Adolesc Psychiatry 1989;28:566-73. 134. Bajwa RJ, de Lotbinie`re AJ, King RA, Jabbari B, Quatrano S, Kunze K, et al. Deep brain stimulation in Tourette’s syndrome. Mov Disord 2007;22:1346-50.
135. Goetz CG, Pappert EJ, Louis ED, Raman R, Leurgans S. Advantages of a modified scoring method for the Rush Video-Based Tic Rating Scale. Mov Disord 1999; 14:502-6. 136. Robertson M. Tourette syndrome, associated conditions and the complexities of treatment. Brain 2000;123: 425-62. 137. Vandewalle V, van der Linden C, Groenewegen HJ, Caemaert J. Stereotactic treatment of Gilles de la Tourette syndrome by high frequency stimulation of thalamus. Lancet 1999;353:724. 138. Visser-Vandewalle V, Temel Y, Boon P, Vreeling F, Colle H, Hoogland G, et al. Chronic bilateral thalamic stimulation: a new therapeutic approach in intractable Tourette syndrome: report of three cases. J Neurosurg 2003;99:1094-100. 139. Houeto JL, Karachi C, Mallet L, Pillon B, Yelnik J, Mesnage V, et al. Tourette’s syndrome and deep brain stimulation. J Neurol Neurosurg Psychiatry 2005;76:992-5. 140. Mink JW, Walkup J, Frey KA, Como P, Cath D, Delong MR, et al. Tourette Syndrome Association, Inc. Patient selection and assessment recommendations for deep brain stimulation in Tourette syndrome. Mov Disord 2006;21:1831-8. 141. Mahloudji M, Pikielny RT. Hereditary essential myoclonus. Brain 1967;90:669-74. 142. Asmus F, Gasser T. Inherited myoclonus-dystonia. In: Fahn S, Hallett M, DeLong M, editors. Advances in neurology, vol 94: dystonia 4. Philadelphia: Williams and Wilkins; 2004. p. 113-9. 143. Muller B, Hedrich K, Kock N, Dragasevic N, Svetel M, Garrels J, et al. Evidence that paternal expression of the e-sarcoglycan gene accounts for reduced penetrance in myoclonusdystonia. Am J Hum Genet 2002;71:1303-11. 144. Kupsch A, Kuehn A, Klaffke S, Meissner W, Harnack D, Winter C, et al. Deep brain stimulation in dystonia. J Neurol 2003;250:I47-52. 145. Liu X, Griffin IC, Parkin SG, Miall RC, Rowe JG, Gregory RP, et al. Involvement of the medial pallidum in focal myoclonic dystonia: a clinical and neurophysiological case study. Mov Disord 2002;17:346-53. 146. Magarinos-Ascone CM, Regidor I, Martinez-Castrillo JC, Gomez-Galan M, Figueiras-Mendez R. Pallidal stimulation relieves myoclonus-dystonia syndrome. J Neurol Neurosurg Psychiatry 2005;76:989-91.
116 Surgery in the Dorsal Root Entry Zone for Spasticity M. P Sindou . P. Mertens
Introduction Surgery in the dorsal root entry zone (DREZ) was introduced in 1972 [1,2] to treat intractable pain. Because of its inhibitory effects on muscular tone, it has been applied to patients with focalized hyperspasticity [3–6]. This method – named microsurgical DREZotomy (MDT) – attempts to selectively interrupt the small nociceptive and the large myotatic fibers (situated laterally and centrally, respectively), while sparing the large lemniscal fibers which are regrouped medially. It also enhances the inhibitory mechanisms of Lissauer’s tract and dorsal horn [7] (> Figure 116‐1). MDT [8] consists of microsurgical incisions that are 2–3 mm deep and at 35 angle for cervical level and at 45 angle at the lumbo-sacral level followed by bipolar coagulations performed ventrolaterally at the entrance of the rootlets into the dorso-lateral sulcus, along all the cord segments selected for operation. MDT is indicated in paraplegic patients, especially when they are bedridden as a result of disabling flexion spasms [9], and in hemiplegic patients with irreducible and/or painful hyperspasticity in the upper limb [10]. MDT can also be used to treat neurogenic bladder with uninhibited detrusor contractions resulting in voiding around a catheter [11].
Rationale Anatomical Bases By definition [1], the DREZ is an entity that includes the central portion of the dorsal rootlet, #
Springer-Verlag Berlin/Heidelberg 2009
the tract of Lissauer, and the layers I to V of the dorsal horn of the spinal cord where the afferent fibers synapse with the cells of the sensory pathways, especially the spinoreticulothalamic tract. Dorsal horn is also penetrated by the myotatic fibers that cross through to reach the ventral horn where they terminate onto motoneurons.
Dorsal Rootlets Depending on the level of the spinal cord, each dorsal root divides into 4–10 rootlets of 0.25–1.5 mm in diameter. Each rootlet is a distinct anatomic–functional entity, that is, a root in miniature. Anatomic studies [1,2] revealed a spatial segregation of afferent fibers in the DREZ according to their sizes and destinations (> Figure 116‐2), and because of this, the lateral regrouping of the fine fibers allows them to be preferentially interrupted without destruction of the large fibers. Whether all nociceptive fibers reach the spinal cord through the dorsal roots is unclear. Anatomic and electrophysiologic studies in animals showed that about 30% of the fibers in the ventral roots were afferent C-axons originating from the dorsal root ganglion cells and projecting into the dorsal horn. These findings, which challenge the Bell-Magendie law, have been clarified. Most (but not all) of the central root afferents do not enter the cord through the lamina cribrosa of the ventral root but instead make a U-turn to reach the dorsal horn via the dorsal root [12].
1960
116
Surgery in the dorsal root entry zone for spasticity
. Figure 116‐1 Schematic representation of the DREZ area and the target of micro-DREZotomy (MDT). Upper part: each rootlet can be divided (owing to the transition of its glial support) into a peripheral and a central segment. The transition between the two segments is at the pial ring (PR), which is located approximately 1 mm outside the penetration of the rootlet into the dorsolateral sulcus. Peripherally, the fibers are mixed together. As they approach the PR, the fine fibers (considered nociceptive) move toward the rootlet surfaces. In the central segment, they group in the ventrolateral portion of the DREZ and enter the dorsal horn (DH) through the tract of Lissauer (TL). The large myotatic fibers (myot) are situated in the middle of the DREZ, whereas the large lemniscal fibers are located dorsomedially. Lower part: schematic data on DH circuitry. Note the monosynaptic excitatory arc reflex, the lemniscal influence on a DH cell and an interneuron (IN), the fine fiber excitatory input onto DH cells, and the IN, the origins in layer I and Layers IV to VII of the anterolateral pathways (ALP), and the projection of the IN onto the motor neuron (MN). (DC, dorsal column.) Rexed laminae are marked from I to VI. The MDT (arrowhead) cuts most of the fine and myotatic fibers and enters the medial (excitatory) portion of the LT and the apex of the dorsal horn. It should preserve most lemniscal presynaptic fibers, the lateral (inhibitory) portion of TL, and most of the DH [1]
Lissauer’s Tract The tract of Lissauer is situated dorsolateral to the dorsal horn and comprises (1) a medial
part, through which the small afferents enter and where they trifurcate to reach the dorsal horn, either directly or through a two-metamere ascending or descending pathway, and (2) a
Surgery in the dorsal root entry zone for spasticity
116
. Figure 116‐2 Course of nerve fibers at the DREZ in humans [2,3]. The diameter of the cervical rootlets chosen as examples is 1 mm. Axons are stained by the Bodian method. (a) Longitudinal section before entry into the spinal cord. At the peripheral segment (P), the large and small fibers have no particular organization. Just before the pial ring (PR), the small fibers reach the rootlet surface (arrows), mainly in the lateral region (L). In the central segment (C), the small fibers are arranged in two bundles (asterisk) located on either side of the large fibers. (b) Longitudinal section at the entry into the spinal cord. The large fibers constitute the center of the rootlet and run toward the dorsal column (DC). The small fibers form two bundles. One is lateral (triangle); the other is medial (asterisk). The medial portion runs obliquely across the rootlet (arrows) to reach the tract of Lissauer (TL). Thus, most of the small fibers are regrouped at the lateral region of the DREZ. (c) Longitudinal section of the rootlet with its afferent endings in the spinal cord. The large lemniscal fibers (thick curved arrow) are grouped medially to enter the dorsal column (DC). The large myotatic fibers (straight arrow) penetrate deeper into the posterior horn (PH) to reach the ventral horn. The small nociceptive fibers (thin curved arrow) are regrouped laterally to enter the TL [1]
lateral part, through which a large number of longitudinal endogenous propriospinal fibers interconnect different levels of the substantia gelatinosa. The tract of Lissauer plays an important role in the intersegmental modulation of the nociceptive afferents [13]. Its medial part transmits the excitatory effects of each dorsal root to the adjacent segments, and its lateral part conveys the inhibitory influences of the substantia gelatinosa into the neighboring metameres [14]. Selective destruction of the medial
part of the tract of Lissauer should cause a reduction in the regional excitability of the nociceptive afferents.
Dorsal Horn Most of the fine nociceptive afferents enter the dorsal horn through the medial part of the tract of Lissauer and the dorsal aspect of the substantia gelatinosa. Ramon y Cajal’s [15] recurrent
1961
1962
116
Surgery in the dorsal root entry zone for spasticity
collaterals of the large lemniscal fibers approach the dorsal horn through the ventral aspect of the substantia gelatinosa [16,17]. As the dendrites of some of the spinoreticulothalamic cells make synaptic connections with the primary afferents inside the substantia gelatinosa layers, substantia gelatinosa exerts a strong segmental modulating effect on nociceptive input. When the large lemniscal afferents within peripheral nerves or dorsal roots are altered, a reduction in the inhibitory control of the dorsal horn occurs. This situation presumably results in excessive firing of the dorsal horn neurons. This phenomenon, thought to be at the origin of deafferentation pain, has been identified in patients by electrophysiologic recordings [18– 20] and has been reproduced in animal experiments [21–23]. Destruction of these hyperactive neurons should suppress the nociceptive impulses generated in the spinoreticulothalamic pathways. Pain-generating neurotransmitters should also be favorably modified by destruction of the dorsal horn apex neurons [24].
of the dorsal horn (I to V layers of Rexed’s classification) are also destroyed if microbipolar coagulations are performed inside the dorsal horn. The procedure is presumed to (partially) preserve the inhibitory structures of the DREZ (i.e., the lemniscal fibers reaching the dorsal horn and the substantia gelatinosa propriospinal interconnecting fibers running through the lateral part of the tract of Lissauer. The method, named microsurgical DREZotomy (MDT), was conceived to prevent complete abolition of tactile and proprioceptive sensations and to avoid deafferentation phenomena [1,25]. Depth and extent of the lesion depend on the degree of the desired therapeutic effect and on the preoperative sensory and functional status of the patient.
Principles of DREZotomy
Dorsal Roots
DREZotomy consists of a longitudinal openingincision of the dorsolateral sulcus, performed ventrolaterally at the entrance of the rootlets into the sulcus, and of microbipolar coagulations, performed inside the sulcus, down to the apex of the dorsal horn, continuously along all the spinal cord segments selected for surgery. The lesion, which penetrates the lateral part of the DREZ and the medial part of the tract of Lissauer, extends to the apex of the dorsal horn, which can be recognized by its browngray color. The average lesion is 2–3 mm deep and is made at a 35 angle medially and ventrally and is presumed to destroy preferentially the nociceptive fibers grouped in the lateral bundle of the dorsal rootlets and the excitatory medial part of the tract of Lissauer. The upper layers
Poorly individualized on leaving the ganglion, the rootlets separate approximately 1 cm before they penetrate the dorsolateral sulcus. They remain joined, however, by fine leptomeningeal membranes, which are easily separated with microdissection. The dorsal roots, which have a mostly symmetric distribution, show different types of division and penetration of their rootlets according to their spinal cord level:
Surgical Anatomy Working in the DREZ requires knowledge of the regional anatomy. Details of this anatomy have been given in previous publications [1,26,27].
1. 2.
The posterior element of C1 (1 mm in diameter) exists in 80% of cases. The superior cervical roots (C2–C4) divide into an average of four rootlets, which are approximately 0.75 mm in diameter and are well separated from one another. Each rootlet has a cylindric type of penetration.
Surgery in the dorsal root entry zone for spasticity
3.
4.
5.
6.
7.
The inferior cervical roots (C5–C8) usually divide into six rootlets with a diameter of 1.50 mm. They are juxtaposed against one another. They also have a cylindric type of penetration. The thoracic roots divide into an average of five small-diameter rootlets (approximately 0.25 mm) and are widely spaced. Penetration is filiform or ribbon shaped. The superior lumbar roots (L1–L3) divide into 8–10 well-grouped rootlets. Often, each rootlet subdivides further into several small secondary filiform rootlets with diameters of less than 0.25 mm. They penetrate the sulcus separately. This type of penetration is called filiform penetration. The lumbosacral roots (L4–S3) usually divide into seven rootlets that are approximately 1.50 mm in diameter. At entry into the spinal cord, they are often imbricated. Penetration is usually cylindric. The sacrococcygeal roots (S4–Co), often adherent to the filum terminale, usually divide into three slender rootlets with diameters of less than 0.25 mm. Penetration is filiform. Sometimes, coccygeal nerve bundles are included in the fibrous sheath of the filum.
Before it deeply penetrates into the spinal cord, a rootlet sometimes takes a subpial course that is superficial to but embedded three quarters in the spinal cord tissue. This course can run as long as 1 mm along the dorsolateral sulcus, and such a segment cannot be dissected, as is frequently the case in the lower part of the cord, where the rootlets are obliquely orientated.
Dorsal Horn The angulation of the DREZ lesion is determined by the axis of the dorsal horn in relation to the sagittal plane crossing the dorsolateral sulcus.
116
According to 82 measurements performed by Young (personal communication), the mean DREZ angle is at 30 , at C6, 26 at T4, 37 at T12, and 36 at L3. The site and extent of the DREZ lesion is also determined by the shape, width, and depth of Lissauer’s tract and the dorsal horn [28], as is shown in figure 134‐3 in chapter 134 (this book).
Selection and Assessment of Patients MDT is recommended only for severe disabling hyperspastic states, such as those encountered in upper limb hemiplegia with abnormal postures in flexion, of the elbow, wrist, and fingers (MDT has to be performed from C5 to T1) and paraplegia with spontaneous flexion or irreducible extension, making washing and dressing difficult, seating in a wheelchair uncomfortable, and kinestherapy ineffective (MDT has to be performed on both sides from L2 to S2, or to S5, if there is a spastic bladder). Of course, MDT, like all other ablative methods, is indicated only when spasticity has resisted all forms of physiotherapy and drug treatments, especially with diazepam, baclofen, and dantrolene, and after having considered the option of intrathecal baclofen therapy. Assessment of the patient must be carried out by a multidisciplinary team. Aside from the neurosurgeon, the team should include an anesthesiologist, a neurologist, a psychiatrist, an orthopedic surgeon, a specialist in urodynamics, and a physiotherapist. Once the spastic state has been demonstrated as harmful, one has to determine the respective involvement in the abnormal posture of (1) spasticity, to be treated with MDT, and (2) only, articular, muscular, tendinous, and/or ligamentous limitations, to be relieved by orthopedic procedures. If doubt persists after detailed clinical and radiological examination, a testing
1963
1964
116
Surgery in the dorsal root entry zone for spasticity
of passive articular mobility should be performed under brief general anesthesia with curare derivatives. When spasticity plays the larger role in the articular limitations, abnormal postures are significantly diminished during the test. If this is not the case, orthopedic surgery may be the initial or the only treatment. Nerve blocks with bupivacaine may also be used for testing.
Surgical Technique Surgery in the DREZ is a microsurgical procedure, requiring an appropriate microsurgical training. Surgery is performed in the patient under general anesthesia, with an initial shortlasting curarization to allow intraoperative observation of motor responses to bipolar electrical stimulation of the nerve roots. Stimulated ventral roots have a motor threshold at least three times lower than the dorsal roots. Standard microsurgical techniques are used with the magnification of 10–25 times. Special microinstruments for MDT have been made by Leibinger-Fischer (Freiburg, West Germany) (> Figure 116‐3).
Operative Procedure at the Cervical Level The prone position with the head and neck flexed in the ‘‘Concorde’’ position has the advantage of avoiding brain collapse caused by cerebrospinal fluid (CSF) depletion. The head is fixed with a three-pin head holder. The level of laminectomy is determined after identification of the prominent spinous process of C2 by palpation. For unilateral DREZ-surgery, a hemilaminectomy – generally from C4 to C7 – with preservation of the spinous processes, allows sufficient exposure to the posterolateral aspect of the cervical spinal cord segments that correspond to the upper limb
innervation, that is, the rootlets of C5 to T1 (> Figure 116‐4). Dura and arachnoid are opened longitudinally. Then the exposed roots are dissected free by separating the tiny arachnoid filaments that bind them to each other, to the arachnoid sheath and to the cord pia mater. The radicular vessels are preserved. Each ventral and dorsal root from C4 to T1 is electrically stimulated at the level of its corresponding foramen to identify precisely its muscular innervation and its functional value. Stimulated ventral roots have a motor threshold at least three times lower than the dorsal roots. Responses are in the diaphragm for C4 (the response is palpable below the lower ribs), in the shoulder abductors for C5, in the elbow flexors for C6, in the elbow and wrist extensors for C7, and in the muscles intrinsic of the hand for C8 and T1. Microsurgical lesions are performed at the selected levels, that is, those that correspond to the upper limb myotomes. The technique is summarized and illustrated in > Figure 116‐6. The incision is made with a microknife (razor blade in a blade-holder or ophtalmologic micro-scalpel). Then microcoagulations are made in a ‘‘chain’’ (i.e., dotted) manner. Each microcoagulation is performed – under direct magnified vision – by short-duration (a few seconds: one to three), low intensity, bipolar electrocoagulation, with a specially designed sharp bipolar forceps incremented at every millimeter. The depth and extent of the lesion depend on the desired therapeutic effect and the preoperative status of the limb. If the laxity of the root is sufficient, the incision is performed continuously in the dorsolateral sulcus, ventrolaterally along all of the rootlets of the targeted root, thus accomplishing a sulcomyelotomy. If not, a partial ventrolateral section is made successively on each rootlet of the root, after the surgeon has isolated each one by separating the tiny arachnoid membranes that hold them together.
Surgery in the dorsal root entry zone for spasticity
116
. Figure 116‐3 Instruments for MDT (‘‘Sindou Instrument-Kit for Microsurgical DREZ-tomy,’’ catalogue ref: 12–29000. Howmedica Leibinger, FL Fischer GmbH. Bo¨tzinger Strasse 41, D-7911 Freiburg, Germany and 14540 Beltwood Parkway East, Dallas, Texas 75244). From top to bottom (a) curved and buttoned microhook for manipulating and holding the spinal roots, (b) malleable microprobe (of the Jacobson type) for gentle dissection and sustained retraction of the rootlets, (c) buttoned microsucker (of an original design) that can be used not only as a sucker but also as a probe and/or a retractor (ref: 12–04220 from Microsurgical DREZ-tomy Kit’’), (d) curved sharp microscissors to divide the fine arachnoidal filaments, the pia mater, and the tiny pial vessels (ref: 12–30108 from Microsurgical DREZ-tomy Kit.), (e) curved razor blade holder whose jaws are striated to allow better stability of the piece of razor blade (A ophthalmic microknife can be used instead of razor blade), and (f) bipolar bayonet-shaped forceps insulated except over 5 mm at the tip, sharp and graduated every millimeter as shown in the magnified view (ref: 12–30179 from microsurgical DREZ-tomy Kit)
Operative Procedure at the Lumbo-Sacral Level The patient is positioned prone on thoracic and iliac supports and the head placed 20 cm lower
than the level of the surgical wound to minimize the intraoperative loss of cerebrospinal fluid. The desired vertebral level is identified by palpation of the spinous processes or, if this is difficult, by lateral X-ray study that includes the S1
1965
1966
116
Surgery in the dorsal root entry zone for spasticity
. Figure 116‐4 Micro-DREZotomy (MDT) technique at the cervical level. Exposure of the right dorsolateral aspect of the cervical cord at C6. (a) The rootlets of the selected dorsal root (DR) are displaced dorsally and medially with a hook or a microsucker to obtain access to the ventrolateral aspect of the DREZ in the dorsolateral sulcus. Using microscissors, the arachnoid adhesions are cut between the cord and the dorsal rootlets. (DC, dorsal cord; DLF, dorsolateral funiculus). (b) After having coagulated exclusively the tiny pial vessels, an incision of 2 mm in depth at 35 ventrally and medially is made with a microknife in the lateral border of the dorsolateral sulcus. (c) Then microcoagulations are performed down to the apex of the dorsal horn using a sharp graduated bipolar microforceps
vertebra. Interspinous levels identified by a needle can then be marked with a nontoxic dye (methylene blue). A laminectomy is performed from T11 to L2. The dura and arachnoid are opened longitudinally, and the filum terminale is isolated. Identification of roots is performed by electrical stimulation (> Figure 116‐5). The L1 and L2 roots are easily identified at their penetration into their respective dural sheaths. Stimulation of L2 produces a response of the iliopsoas and adductor muscles. Identification of L3 to L5 is difficult for several reasons, (1) the exit through their respective dural sheaths is caudal to the exposure, (2) the dorsal rootlets enter the sulcus along an uninterrupted line, (3) the ventral roots are hidden in front of the dentate ligament, and [4] the motor responses in the leg to stimulation of the roots are difficult to observe intraoperatively because of the patients’s prone position. Stimulation of L3 produces a preferential response in the adductors and quadriceps, of L4 in quadriceps, and of L5 in the anterior tibialis. Stimulation of the S1 dorsal root produces a motor response of the gastrocnemius-soleus group that can be
confirmed later, by repeatedly checking the Achilles ankle reflex before, during, and after MDT. Stimulation of the S2 to S4 dorsal roots (or better, directly, the corresponding spinal cord segments at the DREZ) can be assessed by recording of the motor vesical or anal response by use of cystomanometry, rectomanometry, or electromyography of the anal sphincter (or simply with a finger into the sphincter). Because neurophysiologic investigations are time-consuming to perform in the operative room, we have found that measurements at the conus medullaris can be sufficient in the patients who already have severe preoperative impairment of their vesicoanal functions. These measurements, based on human postmortem anatomic studies, have shown that the landmark between the S1 and S2 segments is situated around 30 mm above the exit from the conus of the tiny coccygeal root [26,27]. MDT at the lumbosacral levels has the same principles as the ones at the cervical level. The technique is summarized and illustrated in > Figure 116‐7. At the lumbosacral level, MDT is difficult and possibly dangerous because of the rich vasculature of the conus. The dorsolateral
Surgery in the dorsal root entry zone for spasticity
116
. Figure 116‐5 MDT technique at the lumbosacral level. Top (drawings): Upper left: exposure of the conus medullaris through a T11 to L1 laminectomy. Upper right: approach of the dorso-lateral sulcus. For doing so, the dorsal the rootlets of the selected roots are displaced dorsally and medially to obtain proper access to ventrolateral aspect of the DREZ. Bottom left (operative view): the selected dorsal roots are retracted dorso-medially and held with a (specially designed) ball-tip micro-sucker, used as a small hook, to gain access to the ventro-lateral part of the DREZ. After division of the fine arachnoidal filaments sticking the rootlets together with the pia-mater with curved sharp microscissors (not shown), the main arteries running along the dorso-lateral sulcus are dissected and preserved, whilst the smaller ones are coagulated with a sharp bipolar micro-forceps (not shown). Then, a continuous incision is performed using a micro-knife, made with a small piece of razor blade inserted within the striated jaws of a curved razor-blade-holder. The cut is, on average, at a 45 angle and to a depth of 2 mm. Bottom right (operative view): the surgical lesion is completed by doing micro-coagulations under direct magnified vision, at a low intensity, inside the dorso-lateral sulcomyelotomy down to the apex of the dorsal horn. These microcoagulations are made all along the segments of the cord selected to be operated on, by means of the special sharp bipolar forceps, insulated except at the tip over 5 mm and graduated every millimeter
spinal artery courses along the dorsolateral sulcus. Its diameter is 0.1–0.5 mm, and it is fed by the posterior radicular arteries and joins caudally with the descending anterior branch of the Adamkiewicz artery through the conus medullaris anastomotic loop of Lazorthes. This artery has to be preserved by being freed from the sulcus.
Intraoperative Neurophysiologic Monitoring In addition to the study of the muscular responses to (> Figures 116‐6–> 116‐8) bipolar stimulation of the ventral and dorsal roots – as previously
described-, surface somato-sensori evoked potentials (SSEPs) recordings can be useful for identification of the spinal cord segments. SSEPs monitoring includes (1) dorsal root presynaptic potentials, and (2) dorsal horn postsynaptic potentials. Potentials have a maximal intensity at C6–C7 and C8 for stimulation of the median and ulnar nerve, respectively, and at L5 to S2 and S2 to S4 for stimulation of the tibial nerve and the dorsal nerve of the penis (or clitoris), respectively [11–13]. Recordings of surface SSEPs can also be helpful to monitor the surgical lesion itself. (1) The dorsal column potentials can be monitored for checking the integrity of the ascending dorsal
1967
1968
116
Surgery in the dorsal root entry zone for spasticity
. Figure 116‐6 Right C5-T1 MDT for spasticity in the upper limb. The gray matter potential (N13) is recorded at the level of the sixth cervical segment after stimulation of the median nerve; before and during C5,C6, and C7 MDT (left); at the level of the eighth cervical segment after stimulation of the ulnar nerve; and before and during C8 and T1 MDT (right). Note the amplitude reduction of N13 (in the order of two-thirds of the reference value, which is considered close to the optimal result)
column fibers, especially when the dorsolateral sulcus is not clearly marked. (2) The dorsal horn potentials can be monitored to follow the extent and depth of MDT, particularly when good sensory functions are present before surgery.
Postoperative Care and Rehabilitation Program Because of the precarious preoperative state of a large majority of spastic patients and the severity of their basic disease, hospitalization in the intensive care unit for a few days is strongly recommended. Intensive physiotherapeutic measures for respiration and positioning in bed are necessary
to prevent pulmonary complications and pressure sores. Then the patient must be transferred to a rehabilitation center for intensive physiotherapy. After discharge from the center, an ambulatory physiotherapy program is mandatory to allow the patient to live at home. Thereafter, the patient is followed regularly as an outpatient by the multidisciplinary team.
Illustrative Results Our series consists of three groups of patients. (1) 94 patients, mostly hemiplegic, underwent MDT at the cervical level for hyperspasticity in the upper limb; MDT was performed from C5 to T1
Surgery in the dorsal root entry zone for spasticity
116
. Figure 116‐7 Bilateral L4-S2 MDT for spasticity and pain in the lower limbs. The gray matter potential (N22) is recorded on both sides at the L5-S1 cord level after stimulation of the posterior tibial nerve before, during, and at the end of MDT. Note the amplitude reduction of N22 (in the order of two-thirds of the reference value, which is considered close to the optimal result)
. Figure 116‐8 In intraoperative monitoring, the fact that the dorsal horn potential is maximal at the level(s) corresponding to the main level of entry into the cord of the stimulated peripheral nerve can be exploited to identify cord segments that cannot be easily recognized by evaluation of the muscular response to dorsal root stimulation. This is the case for the sacral metameres. The levels at which the N15 dorsal horn potential evoked by stimulation of the dorsal nerve of the clitoris (or penis) show the largest amplitude are considered to correspond to the S2–S4 metameres
1969
1970
116
Surgery in the dorsal root entry zone for spasticity
segments through C3–C7 hamilaminectomy. (2) 175 patients had MDTat the lumbosacral level for excessive spasticity complicating severe paraplegic states such as those observed in multiple sclerosis patients; MDT was performed bilaterally, through a T11 to L2 laminectomy, from L2 down to S2, and additionally down to S5 when there was a hyperactive neurogenic bladder with urine leakage around the catheter. (3) 15 patients underwent MDT at the sacral (S2 to S3 or S4) level for hyperactive neurogenic bladder. Results have been detailed elsewhere [8–10, 29–31]. Only brief summary is given here. Follow-up ranged from 2 to 25 years; average: 9 years. For hemiplegic patients with harmful spasticity in the upper limb, a good effect was obtained in 83%. The effect on the upper limb was significant and lasting, only at the level of the shoulder and elbow, allowing there reappearance of some voluntary movements when hidden behind hypertonia (> Tables 116‐1 and > 116‐2). Effect was much less beneficial at the level of wrist and fingers, so that additional peripheral neurotomies, together with orthopedic surgery, were often required. For paraplegic patients, a useful effect on lower limbs (i.e., a lasting decrease in tone allowing easy passive mobilization) was obtained in 80% of the patients (> Table 116‐3) (> Figure 116‐9). Bladder capacity was significantly improved in 85%; the patients who improved were those in whom detrusor was not irreversibly fibrotic. MDT constantly produced a decrease in sensation in the operated territories: mild in 40%, marked in 40%, and severe in 20%. When present, pain was durably relieved in 88% in both groups.
Indications Based on our 25 year experience, indications for MDT for hyperspastic states can be summarized as follows:
. Table 116‐1 Functional score for hemiplegic patients with spasticity in the upper limb (personal score, see ref. [30]) Grade
Description
I
Absence of useful active motility; uneasy and painful passive mobilization, making it difficult to dress and wash Easy passive mobilization but without any useful voluntary movements Little but useful voluntary motor function, as for instance for blocking objects with hand Good voluntary motor function with ability of prehension with hand and fingers
II III IV
. Table 116‐2 Comparison between pre-operative and post-operative functional status in hemiplegic. Patients with spasticity in the upper limb (see ref. [30]) Postoperative Preoperative IV III II I Total
1.
2.
I
II
III
IV
4
0
1 6 7
5 5
4
total 0 4 1 16 16
The hyperspastic hemiplegic upper limb can benefit from MDTwhen spasticity predominates in shoulder and elbow. Wrist and fingers are less favorably influenced, especially when there are irreducible contractures and deformities in flexion and/or poor motor function in the extensors; in the later eventuality, peripheral neurotomies together with tendon surgery may be preferred. For lower limbs, as MDT generally has a dramatic effect on tone, surgical indications must be restricted to paraplegic patients with severe disability, unable to walk autonomously. MDT is indicated if patients cannot be seated comfortably in wheelchair or are exposed to pressure sores in bed, especially if additional pain – resulting from
Surgery in the dorsal root entry zone for spasticity
. Table 116‐3 Global functional score for paraplegic patients with spasticity in the lower limb(s) [30] Pain 0 absent 1 mild and rare; no influence on daily life 2 mild, but frequent; inducing depression 3 marked and frequent; participating to the disability 4 permanent and severe; inducing suicide desire Spasms 0 absent 1 mild and rare spasms, only during mobilization; no disability 2 mild but frequent spasms during mobilization; moderate disability 3 marked and frequent, spontaneous spasms; making sitting position uncomfortable, 4 almost constant severe spasms; making sitting position impossible Sitting position 0 normal and comfortable 1 normal posture, but slightly uncomfortable 2 marked difficulty, causing reduction of sitting periods 3 severe difficulty; patients to be tied down in position 4 impossible Body transfers 0 normal 1 mild difficulty 2 moderate difficulty; can do but only from bed to chair 3 marked difficulty; need for a person helping 4 severe difficulty; need for two persons helping Washing and dressing 0 normal 1 mild difficulty 2 moderate difficulty; can do but slowly 3 marked difficulty; need for a person helping 4 severe difficulty; totally dependant This score developed by Millet and associate ([1] cited in ref. [30]) quantifies five components that are directly influenced by spasticity, abnormal postures, and articular limitations and are part of the patient’s everyday life. The score goes from 0 to 4 for each component, with a total of 20/20 denoting a bedridden and totally dependent patient. A score of 10/20 was seen to correspond reproducibly to the threshold between a decent and an unacceptable condition and thus as the lowest position at which to consider surgery
3.
spasms, contractures and/or neurotrophic disturbances – is present. Intrathecal baclofen therapy is an alternative to MDT. MDT can be indicated to treat neurogenic bladder when there is no voluntary
116
. Figure 116‐9 Distribution of preoperative (a) and postoperative (b) global functional scores (GFS) in patients with spasticity in the lower limbs. Nb = number of patients [30])
micturition and if there are uninhibited detrusor contractions resulting in voiding around the catheter or in between intermittent self-catheterization. Surgery in the DREZ must be considered within the frame of all the methods belonging to the armamentarium of hyperspastic disabling states [29–31].
References 1. Sindou M. Etude de la jonction radiculo-me´dullaire poste´rieure: la radicellotomie poste´rieure se´lective dans la chirurgie de la douleur. Lyon: These med; 1972. 2. Sindou M, Quoex C, Baleydier C. Fiber organization at the posterior spinal cord-rootlet junction in man. J Comp Neurol 1974;153:15-26. 3. Sindou M, Fisher G, Goutelle A, et al. La radicellotomie poste´rieure se´lective dans le traitement des spasticite´s. Rev Neurol 1974;130:201-15. 4. Sindou M, Millet MF, Mortamais J, et al. Results of selective posterior rhizotomy in the treatment of painful
1971
1972
116 5.
6.
7.
8.
9.
10.
11.
12. 13. 14.
15. 16.
17.
18.
Surgery in the dorsal root entry zone for spasticity
and spastic paraplegia secondary to multiple sclerosis. Appl Neurophysiol 1982;45:335-40. Sindou M, Abdennebi B, Sharkey P. Microsurgical selective procedures in the peripheral nerves and the posterior root-spinal cord junction for spasticity. Appl Neurophysiol 1985;48:97-104. Sindou M, Pregelj R, Boisson D, et al. Surgical selective lesions of nerve fibers and myelotomies for the modifications of muscle hypertonia. In: Sir Eccles J, Dimitrijevic, MR, editors. Recent achievements in restorative neurology: upper motor neuron functions and dysfunctions. Basel: S. Karger; 1985. p. 10-26. Eccles J, Eccles R, Magni F. Central inhibitory action attributable to presynaptic depolorization produced by muscle afferent volleys. J Physiol 1961;159:147-66. Sindou M, Jeanmonod D, Mertens P. Surgery in the DREZ: microsurgical DREZotomy for treatment of spasticity. In: Sindou M, Abbott R, Keravel Y, editors. Neurosurgery for spasticity. Wien: Springer; 1991. p. 165-82. Sindou M, Jeanmonod D. Microsurgical DREZ-tomy for the treatment of spasticity and pain in the lower limbs. Neurosurgery 1989;24:655-70. Sindou M, Mifsud JJ, Boisson D, et al. Selective posterior rhizotomy in the dorsal root entry zone for treatment of hyperspasticity and pain in the hemiplegic upper limb. Neurosurgery 1986;18:587-95. Beneton C, Mertens P, Leriche, et al. The spastic bladder and its treatment. In: Sindou M, Abbott R, Keravel Y. Neurosurgery for spasticity: a multidisciplinary approach. New York: Springer; 1991. p. 193-9. Willis WD. Pain system. Basel: Karger; 1985. Rand R. Further observations on Lissauer’s tractolysis. Neurochirurgica 1960;3:151-68. Denny-Brown D, Kirk EJ, Yanagisawa N. The tract of Lissauer in relation to sensory transmission in the dorsal horn of spinal cord in the macaque monkey. J Comp Neurol 1973;151:175-200. Ramon y Cajal S. Histologie du syste´me nerveux, vol. 1. Paris: Maloine; 1901. Szentagothai J. Neuronal and synaptic arrangement in the substantia gelatinosa. J Comp Neurol 1964; 122:219-39. Wall PD. Presynaptic control of impulses at the first central synapse in the cutaneous pathway. In: Eccles JC, Schade´ JP, editors. Physiology of spinal neurons. Amsterdam: Elsevier; 1964. p. 92-118. Jeanmonod D, Sindou M, Magnin M, et al. Intra-operative unit recordings in the human dorsal horn with a
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30. 31.
simplified floating microelectrode. Electroencephalogr Clin Neurophysiol 1989;72:450-4. Guenot M, Hupe JM, Mertens P, et al. New type of microelectrode for obtaining unitary recordings in the human spinal cord. J Neurosurg (Spine) 1999;91: 25-32. Guenot M, Bullier J, Rospars J, et al. Single-unit analysis of the spinal dorsal horn in patients with neuropathic pain. J Clin Neurophysiol 2003;20:142-50. Loeser JD, Ward AA, Jr, White LE, Jr. Some effects of deafferentation of neurons. J Neurosurg 1967; 17:629-36. Albe-Fessard D, Lombard MC. Use of an animal model to evaluate the origin of and protection against deafferentation pain. In: Bonica JJ, et al. editors. Advances in pain research and therapy, vol. 5. New York: Raven Press; 1983. p. 691-700. Guenot M, Bullier J, Sindou M. Clinical and electrophysiological expression of deafferentation pain alleviated by dorsal root entry zone lesions in rats. J Neurosurg 2002;97:1402-9. Mertens P, Ghaemmaghami C, Bert L, et al. Microdialysis study of amino-acid neurotransmitters in the spinal dorsal horn of patients undergoing microsurgical dorsal root entry zone lesioning. J Neurosurg (Spine 1) 2001;94:165-73. Jeanmonod D, Sindou M. Somatosensory function following dorsal root entry zone lesions in patients with neurogenic pain or spasticity. J Neurosurg 1991;74:916-32. Sindou M, Fischer G, Mansuy L. Posterior spinal rhizotomy and selective posterior rhizidiotomy. In: Krayenbu¨hl H, Maspes PE, Sweet WH, editors. Progress in neurological surgery, vol. 7. Basel: Karger; 1976. p. 201-50. Sindou M, Goutelle A. Surgical posterior rhizotomies for the treatment of pain. In: Krayenbu¨hl H, editor. Advances and technical standards in neurosurgery, vol. 10. Vienna: Springer; 1983. p. 147-85. Mertens P, Gue´not M, Hermier M, et al. Radiologic anatomy of the spinal dorsal horn at the cervical level (anatomy – MRI correlations). Surg Radiol Anat 2000; 22:81-8. Sindou M. Microsurgical DREZotomy (MDT) for pain, spasticity and hyperactive bladder: a 20 year experience. Acta Neurochir 1995;137:1-5. Sindou M, Abbott R, Keravel Y (eds.). ‘‘Neurosurgery for spasticity.’’ Wien: Springer; 1991. Decq P, Mertens P. La Neurochirurgie de la Spasticite´, Neurochirurgie. Masson: Paris; 2003.
97 Thalamic Stimulation for Parkinson’s Disease R. E. Wharen . R. J. Uitti . J. A. Lucas
Introduction The application of deep brain stimulation (DBS) has proven to be a tremendous step forward for treatment of patients with Parkinson’s Disease (PD). Today DBS, along with medications, is a fundamental component in the management of PD patients. Although newer experimental techniques focus on restoring or preventing the progressive neural degeneration that occurs in PD, DBS remains a valuable treatment option to help control the disabling symptoms of patients with PD. Stimulation of the ventrointermedius nucleus of the thalamus (Vim) is a treatment alternative that can be considered for PD patients suffering from tremor in whom other features of PD are not a source of disability. Vim DBS does not substantially improve akinesia, rigidity, or dyskinesia [1–5]. Vim DBS can control extremity tremor both short [6–10] and long term [2,11–18], and can be safely performed unilaterally or bilaterally without postoperative medication changes [1]. Compared to thalamotomy, thalamic DBS is generally a preferred treatment because of the inherent advantages of reversibility and reduction of side effects from large or suboptimal lesions, and adaptation of stimulation over time to the needs of the individual patient [1,5,19,20].
bilateral ablative procedures [21]. The development of thalamic DBS in fact was based upon the observation that stimulation of the thalamic target at high frequencies, a standard technique used to localize a target for thalamotomy, had the same effect as does its destruction [22–25]. The application of thalamic DBS to select patients with PD remains an art. The majority of PD patients suffer from motor (akinetic-rigid) disabilities, and DBS using a subthalamic nucleus (STN) or internal globus pallidus (GPi) target is more appropriate and may also eliminate tremor [26,27]. STN DBS, in fact, has been performed in PD patients because of progressive parkinsonism with disabling motor fluctuations who had previously undergone thalamic DBS for treatment of tremor [28]. Consensus pragmatic recommendations [1] for the use of Vim DBS in PD patients are that it can be considered as an option for the treatment of long-standing tremor-dominant patients in whom other features of PD are not a source of disability [1]. Tremorpredominant PD patients frequently have minimal progression of other parkinsonian features over many years despite severe tremor. Vim DBS can also be an effective option for patients previously treated with STN or GPi DBS who continue to have (or subsequently develop) severe tremor despite good control of other motor symptoms of PD.
Patient Selection Surgical Procedure DBS of the thalamus was first employed in PD patients who had previously undergone a unilateral thalamotomy to avoid the necessity for #
Springer-Verlag Berlin/Heidelberg 2009
The goal of surgery is to accurately place a DBS electrode in a thalamic location such that
1632
97
Thalamic stimulation for parkinson’s disease
tremor will be abolished at as low a voltage or current as possible maximizing battery life, and at a level that avoids side effects from the stimulation. There are numerous techniques to accomplish this goal. Described below is our technique that has evolved from experience gained from more than 200 thalamic DBS cases.
Headframe Application A COMPASS stereotactic headframe is applied the day of surgery. Local anesthesia is used by many, but we prefer to use a brief propofol anesthetic titrated to tremor relief during the headframe application to minimize patient discomfort. Benzodiazepines should be avoided because of possible prolonged tremor reduction that can occur which could interfere with the evaluation of electrode placement at the time of surgery. One advantage of the COMPASS headframe is that the software corrects for any rotation or lateral tilt and thus does not require orientation of the headframe along the orbitomeatal line as may be necessary with other types of stereotactic headframes.
Surgical planning Computer-assisted stereotactic planning software (COMPASS) is used to plan the initial target and trajectory. Since the ventalis intermedius (Vim) nucleus of the thalamus cannot be distinguished on MRI, initial target planning relies on indirect methods. The target is based on the Guiot method for localization of Vim [30,31,38]. The anterior (AC) and posterior (PC) commissures are chosen from MRI, with the initial thalamic target 11.5 mm lateral to the lateral border of the third ventricle on the AC-PC plane. In the sagittal plane the initial target is 1 mm anterior to the posterior border of the Vim nucleus on the AC-PC line (> Figure 97-1). The target can be correlated with a computer overlay of a stereotactic atlas (e.g., Schaltenbrand and Wahren) co-registered and adjusted to the patient’s intercommissural line (> Figure 97-2). A trajectory is chosen to avoid the lateral ventricle (> Figure 97-3). The anterior angle of approach is usually 60–70 from a perpendicular to the AC-PC plane and depends upon the location of the incision and burr hole, which is usually placed behind the hair line when possible. This angle of approach usually traverses the borders between thalamic nuclei and facilitates mapping and physiologic target localization [32].
Imaging Although several groups continue to use ventriculography [19,20], most now use MRI and/or CT imaging [29]. MRI is the modality of choice to visualize the anterior and posterior commissures. Our preferred technique is to obtain a volumetric, spoiled gradient echo sequence using 1-mm contiguous slices in a 256 256 matrix in the axial plane prior to surgery. On the day of surgery a CT scan using 1.0-mm contiguous slices is obtained after application of the headframe, and the MRI is co-registered to the CT using image fusion techniques. If the patient has a pacemaker then only a CT is obtained.
Surgical Technique Initial Placement of the DBS Electrode The patient is positioned is a semi-sitting position (> Figure 97-4) and the sterile preparation and draping performed to permit easy access for patient assessment by the movement disorder neurologist and nurse. Using local anesthesia and a c-shaped incision, a burr hole is made and a NAVIGUS burr hole cap inserted for eventual securing of the electrode. A microdrive
Thalamic stimulation for parkinson’s disease
97
. Figure 97-1 Vim target planning using Guiot’s geometric scheme. The Vim target is 1 mm anterior to the posterior border of the Vim on the AC-PC line
(Atlantic Research Systems) utilizing three guide tubes spaced 3 mm apart and oriented in an anterior to posterior direction is attached to the stereotactic headring. (> Figure 97-5a,b). The Medtronic (Minneapolis, MN) Model 3387 electrode is then advanced through the middle guide tube to the target site. While being advanced, the patient’s contralateral postural tremor is observed to determine any changes related to disruption of tremor activity by virtue of electrode insertion alone.
Intra-operative Confirmation of the Target Some physiologic localization and confirmation of the target site is necessary, but the techniques used to analyze and verify the target site vary greatly across centers, and may even vary within the same group depending upon the target site, as techniques for location of the subthalamic nucleus (STN) may differ from localization of
the Vim. Techniques of physiologic localization include microelectrode recording (MER), semi-microelectrode recording, macroelectrode recording, microelectrode stimulation, and macroelectrode stimulation. A consensus review recently concluded that there are no data to support one strategy over another [29]. For targeting theVim, several groups still utilize microelectrode recording [32,33]. Our group discontinued the use of MER and microstimulation for localization of Vim and currently use only macrostimulation through the DBS electrode for physiologic confirmation [34]. In our experience as well as others [35], MER can sometimes itself produce a microthalamotomy effect that can inhibit the critical step of the clinical evaluation of tremor relief with the DBS electrode. Since the ultimate goal of surgery is to accurately place a DBS electrode in Vim such that tremor will be abolished at as low a voltage or current as possible, and at a level that avoids side effects from the stimulation, the physiologic criteria for an acceptable electrode placement are
1633
. Figure 97-2 Computer overlay of the Vim target on the Schaltenbrand and Wahren atlas
1634
97 Thalamic stimulation for parkinson’s disease
. Figure 97-3 Final target and trajectory planning – the lateral coordinate is 11.5 mm lateral to the lateral border of the third ventricle. The trajectory should avoid the lateral ventricle
Thalamic stimulation for parkinson’s disease
97 1635
1636
97
Thalamic stimulation for parkinson’s disease
. Figure 97-4 Operative setting – the patient is in the semi-sitting position for comfort and to minimize loss of cerebrospinal fluid
based upon such parameters. After the DBS electrode is placed at the target site and radiographic confirmation obtained, macrostimulation is performed. Evaluation of the lateral coordinate is obtained by increasing the voltage until sensory stimulation occurs. The optimal site of sensory stimulation involves the hand (preferably the thumb and index finger) and the corner of the mouth. If stimulation occurs in the leg the electrode is too lateral, and if only in the face it is too medial. In our experience a medial or lateral adjustment is necessary in less than 10% of cases. The empiric criteria used to define an acceptable physiologic target localization include elimination of the tremor at a voltage of less than 3 volts and a therapeutic to side effect ratio of at least 2:1 (> Figure 97-6). It is acceptable for the patient to experience some sensory stimulation initially but this should not persist. Speech also needs to be monitored to avoid intolerable dysarthria, particularly when bilateral DBS is being performed. The stimulation parameters used for intraoperative evaluation include bipolar stimulation with the most distal contact negative and the most proximal contact positive, a pulse width of 90 ms, and a frequency of 165 hertz. An
attempt is made to achieve effective tremor control using these parameters with at least two distal contacts. The electrode can be easily adjusted along the trajectory with the micromanipulator, and is frequently tested several millimeters beyond the AC-PC line. If these parameters are not achieved along the initial trajectory despite a good medial to lateral localization, the electrode can be quickly adjusted along a parallel trajectory either posteriorly or anteriorly by 3 mm using the parallel guide tubes. The most frequent adjustment in 25% of cases was repositioning of the DBS electrode 3 mm anterior to the initial target. In 54% of the cases a single trajectory was sufficient (> Table 97-1).
Completion of DBS Implantation After physiologic localization of the target site is completed, the electrode is secured and the final position confirmed by a lateral skull x-ray. The DBS electrode is then fixed and protected with a cap and tunneled subgaleally in a parietooccipital direction, the wound is closed, and the headframe removed. The extension lead and
Thalamic stimulation for parkinson’s disease
97
. Figure 97-5 (a) Operative setting demonstrating the COMPASS headring and the Atlantic Research System microdrive. (b) Closeup view of the microdrive and the three cannula system with 3-mm spacing between the cannula. The cannula are oriented in the anterior to posterior direction
pulse generator are inserted using general anesthesia as an outpatient procedure usually 3–7 days after electrode implantation. At that time a small incision is made over the lead cap and the initial incision is not reopened. The impulse generator is placed in a subclavicular location.
Programming the Impulse Generator Methods and timing of programming the impulse generator vary greatly and have not been standardized [36]. Our approach has been to attempt to program the stimulator immediately
1637
1638
97
Thalamic stimulation for parkinson’s disease
. Figure 97-6 Intraoperative confirmation and adjustment – bipolar stimulation is performed with the Medtronic trial stimulator. The goal is to abolish tremor at a threshold of <3 volts and achieve a therapeutic to side effect ratio of at least 2 to 1. Stimulation parameters are 165 hertz and the pulse width is 90 ms. A lateral x-ray is obtained to verify the location of the DBS lead to the planned target
following implantation of the impulse generator. Sometimes a microthalamotomy effect is still present which can limit the programming at this time. This generally portends a good long term outcome for tremor relief. Follow-up visits with reprogramming are scheduled at the time of suture removal, and at 1, 3, and 12 months following surgery and yearly thereafter. Although changes in the stimulation parameters are frequent in the first few weeks, after 1 month following surgery the stimulation parameters are generally stable. Stimulation is attempted using first a bipolar configuration. Monopolar stimulation using case positive is used if necessary to improve tremor control and reduce side effects from the stimulation.
Results Patient Characteristics A clinical series of 24 patients with tremorpredominant PD treated with placement of 28 thalamic DBS electrodes for intractable tremor between May 1997 and September 2006 are detailed in > Table 97-1. Diagnosis of PD was made by one movement disorder neurologist (RJU) based on the presence of resting tremor, bradykinesia, and rigidity without evidence of an atypical parkinsonian syndrome. All patients benefited from levodopa therapy. Surgical indications included disabling tremor despite optimal medical therapy, which included trials of
2.9 2.8 2.5 2.6 2.0
1.5 2.0 1.0 2.8 2.8 2.8 3.0 2.9 3.2 1.3 3.5 1.0 2.0 4.5 2.6 2.4 2.4 3.0 3.2 3.0 2.5 1.7 1.8
2.5
6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Average
Volts
1 2 3 4 5
Patient #
77
60 60 60 60 60 60 90 90 60 60 90 60 90 90 60 60 60 90 90 60 150 90 60
90 90 90 90 90
PW
160
130 130 100 160 160 160 185 160 160 130 160 135 170 185 130 185 170 185 160 160 185 160 130
185 160 170 185 185
Rate
0–3+ 0–3+ 0–case + 0–case+ 0–3+ 0–3+ 0–3+ 0–3+ 0–3+ 0–3+ 0–3+ 0–3+ 0–3+ 0–3+ 1–3+ 0–3+ 0–3+ 0–case+ 1–case+ 0–3+ 1–case+ 0–3+ 1,2,3–0+
0–3+ 1–3+ 0–3+ 0–3+ 0,1–case +
Contacts
Parameters at 1 month
. Table 97-1 Stimulation parameters and outcomes
3.1
2.5 2 2.8 3.2 3 2.8 3.5 2 3.6 1.6 4.7 3 2.9 4.5 2.5 3 4 3.7 3.3 3 3 2.5 3
3.9 4.5 2.5 2.6 2
Volts
84
90 60 60 90 60 60 90 90 60 60 90 90 90 90 90 60 90 90 120 60 90 90 60
90 90 90 90 150
PW
167
130 130 100 185 160 130 145 185 160 130 185 160 170 185 170 185 185 185 160 160 185 185 185
185 185 170 185 185
Rate 1–3+ 0–case+ 0–3+ 0–3+ 0,1,2– case + 0–3+ 0–3+ 1–3+ 0–3+ 0–3+ 0–3+ 1,2–3+ 3–case+ 0–3+ 0–3+ 0–2+ 0–3+ 0–3+ 0–3+ 0–2+ 0–3+ 0–3+ 0–case+ 1–case+ 0–3+ 0–case+ 1–3+ 1,2,3–0+
Contacts
Parameters at last followup
4
3 0.1 6.3 7 4.1 1.1 6.3 5.1 5.2 2 7.1 4.9 2.2 2.4 6.6 3.6 2.7 2.1 2.1 0.1 2 7 8.6
1.6 0.6 4.7 4 10.2
Years after surgery
1.4
1.0 1.5 1.5 1.0 2.0 2.0 1.0 1.0 2.0 2.0 1.5 2.0 2.0 2.0 1.0 1.0 1.0 1.5 1.5 2.5 2.0 0.0 0.4
0 1.0 2 2 2
Stimulation at surgery (volts)
no no yes no no no yes no no yes no no no no no no yes no no yes no yes no
yes no no no no
Microthalamotomy effect
ant med no no no med no no lat no lat + post ant no ant ant no no ant + lat lat ant post no no
no ant no no no
Adjustment of target
Thalamic stimulation for parkinson’s disease
97 1639
1640
97
Thalamic stimulation for parkinson’s disease
levodopa and dopaminergic agents. Staged, bilateral surgery (four patients) was performed in cases of severe bilateral tremor. Two patients had previous pallidotomies (patients #7 and #28) and one patient (patient #25) had a previous subthalamic DBS procedure with excellent improvement of all symptoms except tremor. A thalamic DBS procedure was subsequently performed on the same side for tremor control (patient #25). Nineteen of these patients have been previously reported in 2003 [16]. For this subgroup of patients a battery of subjective and objective measures of tremor was completed at planned pre- and post-operative intervals. The Tremor Rating Scale (TRS) [37], a wellestablished patient and clinician-based rating scale of tremor, and activities of daily living (ADL) were used to analyze this subgroup. Objective measures included the Purdue Pegboard Test, accelerometer testing of reaction and movement time, tremor frequency, and tremor power [16]. Outcome measures collected under both the stimulation ‘‘on’’ and ‘‘off ’’ condition at each post-operative interval were compared to each other and to the pre-operative functioning (i.e., baseline). Paired comparisons were made using the Wilcoxon sign rank test. Significance was set at p < 0.05, although three significance levels were reported (p < 0.05, p < 0.01, and p < 0.001). Patients were predominantly male (n = 24, 86%) with a mean age of 68.8 years (standard deviation (SD), 7.8 years) at the time of surgery. A total of 28 Vim DBS electrodes were placed (n = 24 unilateral stimulation; 86%) predominantly on the left side (left n = 21, 75%, right n = 7, 25%). The course of follow-up consisted of a mean of 4.0 years (SD, 2.6 years) ranging from 0.1 to 10.2 years (> Table 97-1).
occurred 24 months after electrode placement. This lead was subsequently replaced 5 months later with scalp revision using a rotational scalp flap. Following this case the scalp incision was changed from a linear incision to a c-shaped incision so that the incision is not directly over the implant. In addition, the use of the Navigus burr hole cap with a much lower profile replaced the original Medtronic burr hole cap. One lead required repositioning 24 months after surgery despite adequate tremor control due to contralateral ‘‘burning and tingling’’ of the face associated with stimulation (patient #28). There were no symptomatic hemorrhages in this group of patients. In our experience of over 300 patients with PD treated with DBS, there has been one symptomatic hemorrhage which occurred in a patient undergoing STN DBS following insertion of a second microelectrode for microelectrode recording (MER). In our opinion, there is unquestionably a small but not insignificant risk associated with placement of microelectrodes for microelectrode recording. A recent analysis [29] of the issue of the risk of hemorrhage with MER concluded that only class IV data exist and the data are inconclusive, but a meta-analysis of reported cases [39] noted that non-MER techniques were at least five times less likely to have hemorrhagic complications. This hemorrhagic risk combined with the questionable efficacy of microelectrode recording in the thalamus, where the clinical evaluation of the elimination of tremor is the gold standard, and the risk of the creation of a microthalamotomy effect from the use of MER, that impedes the clinical evaluation of relief of tremor by intraoperative stimulation, influenced our group against the use of MER for thalamic DBS.
Surgical Complications
Long-term Efficacy
One lead (patient #10) was removed due to erosion of the scalp wound over the burr hole cap (prior to the use of the Navigus cap) which
Three general findings emerge from the summary of mean TRS scores at the various postoperative intervals (> Table 97-2). First, the
Thalamic stimulation for parkinson’s disease
97
. Table 97-2 Summary of clinical tremor rating scale
Off stimulation On stimulation IL upper extremitya Off stimulation On stimulation Off stimulation On stimulation IL lower extremitya Off stimulation On stimulation Midline tremorb Off stimulation On stimulation Midline tremor* Off stimulation On stimulation ADL Off stimulation On stimulation
Presurgical M SD
Month 1 M SD
Month 3 M SD
Year 1 M SD
Year 2 M SD
Year 3 M SD
7.6 2.7
5.9 3.4* 0.5 1.1{, #
6.1 3.2 0.6 1.0{,#
7.5 2.3 0.8 1.0{,#
5.4 3.7 0.9 1.2{,*
5.2 4.5 0.8 1.3*
3.0 3.1
2.7 2.7 2.4 2.0 1.8 1.7 0.6 0.9**#
3.7 3.0 2.1 2.0{ 2.5 2.0 0.3 0.5**,##
3.7 3.2 2.7 3.0{ 2.2 2.1 0.1 0.4*#
2.0 2.4 1.9 2.2 2.0 2.1 0.0 0.0{
1.7 2.1 1.3 1.5 3.6 1.5 0.0 0.0{
0.8 1.5
0.9 1.3 0.4 0.9
1.2 1.2 0.5 0.8
1.6 1.8 0.8 1.4
1.0 1.3 0.2 0.4
0.7 0.6 0.3 0.6
0.5 0.7
1.0 (–) 0.5 0.7
1.0 0.0 1.0 0.0
3.0 (–) 0.0 (–)
3.0 (–) 0.0 (–)
3.0 (–) 0.0 (–)
2.9 3.5
2.2 2.4 1.3 1.0#
2.4 2.3 0.9 0.9*{
3.8 4.3 1.1 1.4*#
3.7 3.0 1.3 1.3#
3.0 1.7 1.7 1.5
11.8 5.3
– 5.5 3.1{
– 5.9 4.7**
– 5.2 3.6**
– 4.0 4.1*
– 6.3 3.7
2.3 2.3
CL contralateral; IL ipsilateral; ADL activities of daily living. Summary of tremor scores represent all forms of tremor (rest, postural, kinetic); * p < 0.05 versus baseline. **p < 0.01 versus off. {p < 0.01 versus baseline. {p < 0.05 versus off. #p < 0.001 versus off. ##p < 0.001 versus baseline (From Putzke JD, et al Parkinsonism and Relat Disord 10: 81–83, 2003, with permission) a Only those with unilateral DBS b Only those with bilateral DBS
stimulation ‘‘on’’ condition produces significant improvement when compared to pre-operative ratings of ADL ability, midline tremor, and contralateral upper and lower extremity tremor. Second, comparisons between the stimulation ‘‘on’’ and ‘‘off’’ conditions showed significant improvement with stimulation at nearly every postoperative interval. Third, ipsilateral upper and lower extremity tremor in the stimulation ‘‘on’’ condition showed some improvement compared to baseline over the first 3 months, but no significant improvement was observed after 12 months. Objective quantitative measures (> Table 97-3) demonstrated significant improvement of contralateral resting and postural tremor with stimulation compared to both baseline functioning and the ‘‘off ’’ condition at nearly every post-operative interval. In contrast,
measurements of reaction time and movement time, finger dexterity, and ipsilateral tremor showed little or no effect with stimulation.
Side-effects, Stimulator Settings, Anti-PD Medication Usage, Neuropsychological Outcomes Stimulation parameters in all 24 patients have remained stable over a mean follow-up of 4 years and extending to more than 10 years in some patients (> Table 97-1). The mean parameters at 1 month after surgery were 2.5 volts, pulse width of 77 ms, and a rate of 160 hertz (Hz) compared to mean parameters of 3.1 volts, pulse with of 84 ms, and a rate of 167 Hz at the time of the last follow-up. Stimulation parameters did
1641
CL reaction time (ms) Off stimulation On stimulation IL reaction time (ms)a Off stimulation On stimulation CL movement time (ms) Off stimulation On stimulation IL movement time (ms)a Off stimulation On stimulation CL rotations Off stimulation On stimulation IL rotationsa Off stimulation On stimulation CL resting tremor frequency (Hz) Off stimulation On stimulation IL resting tremor frequency (Hz)a Off stimulation On stimulation CL resting tremor power (mgrav) Off stimulation On stimulation 486.2 109.9 411.4 134.1{ 507.7 158.9 469.2 104.2 406.8 30.6 388.2 75.0 325.5 11.7 383.6 124.0 1.2 0.7 1.3 0.6 1.6 0.9 1.5 1.0 3.2 2.3 0.0 0.0**,# 2.7 2.4 0.8 1.8 214.6 228.1 1.5 3.52,b
501.5 166.8
387.5 110.2
350.6 131.9
1.0 1.0
1.4 1.0
3.8 1.9
2.4 2.4
154.6 182.2
Month 1 m (SD)
504.9255.5
Baseline m (SD)
148.8 323.2 0.6 2.1**,#
2.8 2.4 1.4 2.2**
2.7 2.1 0.4 1.3**,#
1.2 0.9 1.5 0.9
0.7 0.4 1.2 0.9
349.3 97.0 386.5 12.2{
427.1 124.5 382.1 89.7
502.9 152.6 511.8 192.0
503.9 109.3 467.7 117.0
Month 3 m (SD)
142.8 318.7 3.7 11.72,b
2.5 2.4 1.6 2.6
3.0 2.0 0.4 1.3**,#
1.0 0.9* 1.4 1.1
0.9 0.7 0.7 0.7
497.7 450.0 528.2 417.9
519.1 274.3 448.8 175.3
598.4 240.4 571.1 247.3
532.4 129.5 517.8 162.9
Year 1 m (SD)
58.3 69.5 1.2 2.7
0.8 2.0 0.0 0.0
3.7 1.8 0.8 1.8
1.1 1.1* 1.2 1.3*
0.5 0.3 0.6 0.5
409.1 158.9 397.4 209.7
425.9 130.7 423.4 178.9
482.1 145.9 455.6 65.8
471.4 63.1 480.0 70.0
Year 2 m (SD)
76.0 100.5 0.0 0.0
0.0 0.0 0.0 0.0
1.6 2.8 0.0 0.0
1.1 1.9 1.0 1.7
0.5 0.3 0.3 0.6
370.0 61.8 374.7 26.4
460.3 35.9 475.7 142.6
468.0 143.9 417.0 64.6
467.3 103.8 449.0 143.5
Year 3 m (SD)
97
. Table 97-3 Summary of Objective hand Measurements (From Putzke JD, et al. Parkinsonism and Relat Disord 10: 81-88, 2003, with permission
1642 Thalamic stimulation for parkinson’s disease
8.3 2.3 8.8 3.0 5.6 2.5 8.0 2.4 8.6 2.6 6.1 2.1
3.9 2.6 2.4 2.8 361.8 540.8 5.3 10.32{ 34.4 51.3 5.0 5.2* 8.4 2.4 9.2 2.7 6.4 2.7 8.2 1.7 8.3 2.4{ 6.3 1.5
3.2 2.5
454.1 532.9
58.7 119.5
7.9 3.0 8.8 2.8 6.1 2.7
112.9 206.4 34.7 64.6
338.4 519.8 12.5 26.8**,{
4.8 0.8 2.5 2.8{
4.0 1.8 1.8 2.5*,{
3.6 2.1** 0.6 1.4*,#
4.2 1.70
35.8 63.2 17.1 38.2{
43.3 96.9 37.8 121.5*
48.8 126.3
8.0 2.8 7.9 3.5 6.1 3.1{
6.1 3.4 7.1 3.1 4.4 3.1
138.9 226.2* 129.5 223.2
357.0 532.6 84.7 234.8**,{
4.0 2.1 2.3 2.5
4.5 0.6 3.3 2.5
29.1 44.5 14.6 31.2
Higher purdue pegboard scores reflect better functioning CL contralateral; IL ipsilateral *p < 0.05 versus baseline. **p < 0.01 versus baseline. {p < 0.001 versus baseline. {p < 0.05 versus off. #p < 0.01 versus off. a Only those with unilateral DBS
IL resting tremor power (mgrav)* Off stimulation On stimulation CL postural tremor frequency (Hz) Off stimulation On stimulation IL postural tremor frequency (Hz)a Off stimulation On stimulation CL postural tremor power (mgrav) Off stimulation On stimulation IL postural tremor power (mgrav)* Off stimulation On stimulation Purdue pegboard Off stimulation CL hand IL hand* Both hands On stimulation CL hand IL hand* Both hand 7.0 2.5 9.2 1.1 5.8 2.4
6.6 4.0 9.3 3.6 4.6 3.4
39.7 63.4 0.0 0.0
402.9 577.2 2.4 3.3
2.5 2.4 0.0 0.0
3.9 1.8 1.4 1.9
10.3 21.7 1.4 3.1
7.3 4.6 10.7 1.2 5.7 1.5
6.3 3.2 10.0 2.0 6.3 2.9
10.7 6.4 2.7 4.6
62.3 108.0 0.0 0.0
2.8 2.3 2.7 4.6
1.6 2.8 0.0 0.0
0.0 0.0 2.3 4.0
Thalamic stimulation for parkinson’s disease
97 1643
1644
97
Thalamic stimulation for parkinson’s disease
not change significantly after the third postoperative month [16]. Within the first 3 months after surgery, stimulation parameters were frequently adjusted at clinic visits (69%) to minimize side effects (27%) and to maximize tremor control (81%). Among stimulation changes to minimize side effects, the most common were parasthesias (26%) and dysarthria (21%). Anti-parkinsonian medication use, based on a daily levodopa equivalence dose, revealed a non-significant decrease from the pre-surgical to the longest available follow-up interval for the mean daily levodopa equivalence dose (M = 939 mg vs. 867 mg, SD = 275 and 295 mg) [16]. Over the follow-up interval, the levodopa equivalence dose remained the same (within 100 mg of pre-surgial dose) for 39%, decreased (>100 mg decrease) for 39%, and increased (>100 mg increase) for 22% of patients. These results are consistent with other studies demonstrating little or no change in the long term use of anti-parkinsonian medications [3,4,8]. The lack of any significant increase in pharmacotherapy requirements after prolonged follow-up in these
patients may be explained by the relatively slow progression of non-tremor parkinsonian signs in patients with tremor-predominant PD. A battery of neuropsychological measures (> Table 97-4) were administered to 18 of our 24 patients with medically intractable Parkinson’s disease (15M, 3F) prior to unilateral thalamic DBS surgery (12 L-VIM, 6 R-VIM). The mean age of this sample was 67.2 years (SD = 9.1, Range = 42–81). Participants had a mean education of 12.4 years (SD = 3.8, Range = 4–18) and a mean baseline Dementia Rating Scale score in the low average to mildly impaired range (Mean = 126.9, SD = 10.3, Range = 95–141). The assessment was repeated approximately 3-months post-surgery (Mean test-retest interval = 110 days, SD = 33, Range = 85–186. Given multiple comparisons, a more stringent alpha of p < .01 was applied. Repeated measures ANOVA revealed a significant main effect for time on a measure of semantic verbal fluency (i.e., patients were given three, one-minute trials to generate as many words as possible belonging to the categories Animals, Fruits, and
. Table 97-4 Pre- and post-operative neuropsychological performances Baseline a
Measures Digit span Letter-number sequencing Phonological fluency Category fluency Benton JLO Trails A Trails B HVLT-R Total learning 25-min delay %Retention Recognition POMS anxiety POMS depression
Follow-up
n 17 15 17 17 16 17 17
M 13.7 6.9 27.2 34.8 20.3 62.6 164.7
(SD) (3.2) (2.9) (11.7) (10.6) (6.9) (58.2) (83.3)
M 14.9 7.1 23.6 31.2 19.0 68.6 151.3
(SD) (3.8) (2.8) (11.5) (12.1) (7.5) (57.9) (85.4)
F 2.7 0.1 5.5 8.9* 1.0 1.8 1.1
15 15 15 15 16 16
17.1 5.3 71.4 8.9 52.3 49.4
(6.6) (2.4) (23.5) (2.5) (11.1) (13.3)
17.5 4.5 56.6 8.3 46.3 48.6
(5.8) (2.8) (31.8) (1.9) (10.8) (10.3)
0.1 4.4 4.5 0.7 3.5 0.1
*p < .01 a All data as raw scores with the exception of POMS Anxiety and Depression scores, which are presented as T-scores. JLO Judgment of Line Orientation; HVLT-R Hopkins Verbal Learning Test-Revised; POMS Profile of Mood States
Thalamic stimulation for parkinson’s disease
97
. Figure 97-7 Neuropsychological outcomes for patients with bilateral thalamic stimulation (includes patients with PD and essential tremore) reveals no significant changes in verbal fluency, visualspatial testing, or learning tests at 3 months following surgery
Vegetables). There were no other significant differences; however, there may have been insufficient power in this small sample to detect more subtle effects (> Table 97-4). Neuropsychological evaluation of patients who have received bilateral thalamic DBS implantation for either essential tremor or PD did not show any significant differences in measures of verbal fluency, visual-spatial abilities, or learning (> Figure 97-7).
Summary Stimulation of the Vim nucleus of the thalamus is an effective treatment that can be considered for select PD patients suffering from tremor in whom other features of PD are not a source of disability. Vim DBS can control extremity tremor both short and long term, and can be safely performed unilaterally or staged bilaterally without postoperative medication changes. Stimulation parameters remain stable for as long as at least
10 years following surgery, and stimulation tolerance was not observed. Vim DBS can also be an effective treatment for select patients that have disabling tremor despite control of other symptoms of PD with previous pallidotomy or STN or GPi DBS.
References 1. Lang AE, Houeto JL, Krack P, Kubu C, Lyons KE, Moro E, et al. Deep brain stimulation: preoperative issues. Mov Disord. 2006;21:S171-196. 2. Blond S, Caparros-Lefebvre D, parker F, Assaker R, petit H, Guieu J, et al. Control of tremor and involuntary movement disorders by chronic stereotactic stimulation of the ventral intermedius thalamic nucleus. J Neurosurg. 1992;77:62-68. 3. Benabid AL, Pollak P, Gao DM, Hoffman D, Limousin P, Gay E, et al. Chronic electrical stimulation of the ventralis intermedius nucleus of the thalamus as a treatment of movement disorders. J Neurosurg. 1996;84:203-14. 4. Kumar K, Kelly M, Toth C. Deep brain stimulation of the ventral intermediate nucleus of the thalamus for control of tremors in parkinson’s disease and essential tremor. Stereot Funct Neurosurg. 1999;72:47-62.
1645
1646
97
Thalamic stimulation for parkinson’s disease
5. Tasker RR. Deep brain stimulation is preferable to thalamotomy for tremor suppression. Surgical Neurol. 1998;49:145-53. 6. Alesh F, Pinter MM, Helscher RJ, Fertl L, Benabid AL, Koos WT. Stimulation of the ventral intermediate thalamic nucleus in tremor dominated Parkinson’s disease and essential tremor. Acta Neurochir. 1995;136:75-81. 7. Koller W, Pahwa R, Busenbark K, Hubble J, Wilkinson S, lang A, et al. High-frequency unilateral thalamic stimulation in the treatment of essential and parkinsonian tremor. Ann Neurol. 1997;42:292-99. 8. Limousin P, Speelman JD, Gielen F, Janssens M: Multicentre European study of thalamic stimulation in parkinsonian and essential tremor. J Neurol Neurosurg Psychiatry. 1999;66:289-96. 9. Tarsy D, Norreggard T, Hubble J. Thalamic deep brain stimulation for Parkinson’s disease and essential tremor. In: Tarsy D, Vitek J, Lozano A, editors. Surgical treatment of parkinson’s disease and other movement disorders. Totowa: Humana Press; 2003. p. 152-61. 10. Tarsy D, Norregaard T. Thalmaic deep brain stimulation. I In Pahwa R, Lyons KE, Koller WC, editors. Therapy of parkinson’s disease. New York: Dekker; 2004. p. 293-308. 11. Pollak P, Benabid AL, Limousin P Benazzouz P, A: Chronic intracerebral stimulation in Parkinson’s disease. Adv Neurol. 1997;74:213-20. 12. Pollak P, Fraix V, krack P, Moro E, mendes A, Chabardes S, et al. Treatment results: Parkinson’s disease. Mov Disord. 2002;17:S75-83. 13. Benabid AL, Benazzouz A, Gao D, Hoffman D, Limousin P, Koudsie A, et al. Chronic electrical stimulation of the ventralis intermerdius nucleus of the thalamus and of other nuclei as a treatment for Parkinson’s disease. Tech Neurosurg. 1999;5:5-30. 14. Lyons KE, Koller WC, Wilkonson SB, Pahwa R. Long-term safety and efficacy of unilateral deep brain stimulation of the thalamus for parkinsonian tremor. J Neurol Neurosurg Psychiatry. 2001;71:682-84. 15. Rehncrona S, Johnels B, Widner H, Tornqvist A, hariz M, Sydow O. Long-term efficacy of thalamic deep brain stimulation for tremor: double-blind assessments. Mov Disord. 2002;18:163-70. 16. Putzke JD, Wharen RE Jr, Wszolek ZK, Turk MF, Strongosky AJ, Uitti RJ. Thalamic deep brain stimulation for tremor-predominant Parkinson’s disease. Parkinsonism Relat Disord. 2003;10:81-88. 17. Kumar R, Lozano AM, Sime E, lang AE. Long-term followup of thalamic deep brain stimulation for essential and parkinsonian tremor. Neurology 2003;61:1601-04. 18. Tarsy D, Scollins L, Corapi K, O’Herron SO, Apetauerova D, Norregaard T. Progression of parkinson’s disease following thalamic deep brain stimulation for tremor. Stereotact Funct Neurosurg. 2005;83:222-27. 19. Benabid AL, Pollak P, Gervason C, Hoffman D, Gao DM, Hommel M, et al. Long-term suppression of tremor
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
by chronic stimulation of the ventral intermediate thalamic nucleus. Lancet 1991;337:403-06. Schuurman PR, Bosch DA, Bossuyt PMM, Bonsel GJ, Van Someren EJW, DeBie RMA, et al. A comparison of continuous thalamic stimulation and thalamotomy for suppression of severe tremor. N Engl J Med. 2000;342:461-68. Benabid AL, Pollak P, Louveau A, Henry S, de Rougemont J. Combined (thalamotomy and stimulation) stereotactic surgery of the VIM thalamic nucleus for bilateral Parkinson’s disease. Appl Neurophysiol. 1987;50:344-46. Guiot G, Derome, P, Arfel G, Walter S. Electrophysiological recordings in stereotaxic thalamotomy for parkinsonism. Prog Neurol Surg. 1973;5:189-221. Ohye C, Shibazaki T, Hirai T, wada H, Hirato M, Kawashima Y. Further physiological observations on the ventralis intermedius neurons in the human thalamus. J Neurophysiol. 1989;61:488-500. Ohye C, Maeda T, Harabayashi H. Physiologically defined Vim nucleus: its special reference to control of tremor. Appl Neurophysiol. 1977;39:285-295. Tasker R. Effets sensitifs et moteurs de la stimulation thaalamique chez l’homme: Applications cliniques. Rev Neurol (Paris). 1986;142:316-326. Krack P, Batir A, Van Blercom N, Chabardes S, Fraix V, Ardouin C, et al. Five-year followup of bilateral stimulation of the subthalamic sucleus in advanced Parkinson’s disease. N Engl J Med. 2003;349:1925-34. Volkmann J, Sturm V, Weiss P, Kappler J, Voges J, Koulousakis A, et al. Bilateral high-frequency stimulation of the internal globus pallidus in advanced Parkinson’s disease. Ann Neurol. 1998;44:953-61. Fraix V, Pollak P, Moro E, Chabardes S, Xie J, Ardouin C, et al. Subthalamic nucleus stimulation in tremor dominant parkinsonian patients with previous thalamic surgery. J Neruol Neurosurg Psychiatry. 2005;76:246-48. Rezai AR, Kopell BH, Gross RE, Vitek JL, Sharan AD, Limousin P, Benabid AL. Deep brain stimulation for Parkinson’s Disease: surgical issues. Mov Disord. 2006;21:S197-218. Burchiel KJ. Thalamotomy for movement disorders. In: Gildenberg PL, editor. Neurosurgery clinics of North America. Philadelphia: WB Saunders; 1995. p. 55-71. Taren T, Guiot G, Derome P, Trigo JC. Hazards of stereotaxic thalamotomy. Added safety factor in corroborating x-ray target localization with neurophysiological methods. J Neurosurg. 1968;29:173-82. Gross RE, Krack P, Rodriguez-Oroz, Rezai AR, Benabid AL. Electrophysiological mapping for the implantation of deep brain stimulators for Parkinsons’s disease and tremor. Mov Disord.2006:21;S259-83. Garonzik AM, Hua SE, Ohara S, Lenz FA. Intraoperative microelectrode and semi-microelectrode recording during the physiologic localization of the thalamic nucleus ventral intermediate. Mov Disord.2002;17:S135-44.
Thalamic stimulation for parkinson’s disease
34. Obwegeser AA, Uitti RJ, Witte RJ, Lucas JA, Turk MF, Wharen RE Jr. Quantitative and qualitative outcome measures after thalamic deep brain stimulation to treat disabling tremor. Neurosurgery 2001;48:274-81. 35. Giller CA, Dewey RB Jr. VIM thalamotomy can succeed when VIM stimulation fails: report of 2 cases for tremor. Stereotact Funct Neurosurg. 2002;79:51-56. 36. Deuschl G, Herzog J, Kleiner-Fisman G, Kubu C, Lozano AM, Lyons KE, et al. Deep brain stimulation: postoperative issues. Mov Disord. 2006;21:S219-37. 37. Fahn S, Tolosa E, Marin C. Clinical rating scale for tremor: In: Jankovic J, Tolosa E, editors. Parkinson’s
97
disease and movement dirorders. Baltimor: Urban and Schwarzenberg; 1993. p. 271-80. 38. Guiot G, Derome P. The principle of steroataxic thalamotomy. In: Kahn EA, Crosby EC, Schneider RC, Taren J, editors. Correlative neurosurgery. Springfield: Charles C Thomas; 1969, p. 376-401. 39. Hariz MI, Fodstad H. Do microelectrode techniques increase or decrease the risks in pallidotomy and deep brain stimulation? A critical review of the literature. Stereotac Funct Neurosurg. 1999;72:157-69.
1647
101 Tissue Transplantation for Parkinson’s Disease K. Mukhida . M. Hong . I. Mendez
Introduction ‘‘A gentleman, otherwise well, had the idea that his brain was rotten. He went to the King, begging him to command M le Grand, Physician, M Pigray, King’s Surgeon-in-Ordinary, and myself to open his head, remove his diseased brain and replace it with another. We did many things to him but it was impossible for us to restore his brain.’’1 The first attempts at neural transplantation were not undertaken for over three centuries after this, Ambroise Pare´’s allusion to brain restoration in his 1564 publication, Dix livres de la chirurgie. The history of the scientific investigation of the central nervous system (CNS) transplantation is shorter than that for other organ systems due to the ‘‘complexity of the pathological subsystems of the CNS in disease and the lack of available and appropriate cadaver donor tissue’’ [268, p. 1032]. Initial studies were not concerned with the potential for cell replacement strategies to restore neurological functions. Thompson is considered to have been the first to study neural transplantation when, in 1890, he transplanted cortical tissue from adult dogs and cats into adult dogs [66,264]. Inspired by Charles Darwin’s work on evolution, Thompson’s interest was to use transplantation as a tool to better understand the ‘‘vitality of tissues in living organisms’’ [66,264], p. 389): ‘‘I had no expectation 1 This quotation was cited by Dr. Stephen Dunnett in his lecture entitled ‘‘Functional plasticity and repair of the striatum with striatal transplants’’ delivered on December 3, 2004 at Dalhousie University, Halifax, Nova Scotia, Canada. #
Springer-Verlag Berlin/Heidelberg 2009
of being able to restore abolished function by the operation, but the question of vitality of the brain tissue and the course if its degeneration is a subject which is of very wide interest’’ [264, p. 701]. Although the transplanted tissue survived only 7 weeks at most, these findings enabled Thompson to postulate that the age of the donor tissue influenced its ability to survive: ‘‘the higher the original development of a tissue or cell has been . . . the more profoundly it is affected by alterations in environment or nutrition’’ [264, p. 701]. Building on Thompson’s work, Dunn utilized younger animals (10 days old) in her experiments [66,76]. Like Thompson, her scientific motivation was not for therapeutic application but to determine whether the ‘‘vitality’’ of transplanted cortical cells could be maintained in a foreign cerebral environment [76, p. 571]. The survival of some grafts confirmed the importance of donor age on influencing graft survival, as hypothesized by Thompson [66,76]. Also around the beginning of the twentieth century, other investigators, such as Saltykow [235], Nissl [192], and Altobelli [5], studied the trophic effects of neuronal grafts by replanting cortical tissue into the donor animals and observing their effects on axonal regeneration in areas of the brain surrounding the grafts [66]. Even as late as the 1940s, though, neural transplantation research was still not clinically-oriented and was focused mainly on the ability of grafted tissues to survive in the brain and their cells to differentiate [144]. Although Das concedes that it is not known how the work of scientists engaged in neural transplantation research was received by the larger neuroscience community, he suggests that it was
1692
101
Tissue transplantation for parkinson’s disease
‘‘very likely that they were not taken seriously, either due to skepticism from other investigators or the findings being peripheral to the mainstream interests of neuroscientists’’ [66, p. 397]. The ‘‘contemporary period in neural transplantation,’’ as Das [66] refers to it, began in the 1970s with the recognition of the potential clinical applications of neural transplantation in the treatment of neurodegenerative diseases by replacing neurons lost to disease with grafted cells (p. 398). Parkinson’s disease (PD) is considered to be the most relevant disorder for such strategies (reviewed in [140]). First, it has been argued that the predominant motor symptoms of PD are due to the selective degeneration of the discrete population of dopaminergic neurons in the substantia nigra pars compacta (SNc), implying that there is a clearly-defined group of neurons that could be replaced by transplantation.2 Surgical procedures to replace those cells are technically feasible in parkinsonian patients, either directly in the SNc or in the caudate and putamen, to which the degenerating SNc dopaminergic neurons project [175,177]. Second, medical management that supplies exogenous dopamine, such as the provision of the dopamine precursor L-dihydroxyphenylalanine (levodopa), improves the cardinal motor symptoms of PD, suggesting that transplantation of dopaminergic neurons could also provide similar benefits by acting as dopamine pumps. Third, unlike current therapies for PD that treat its symptoms and are not curative, neural transplantation therapies aim to reconstruct the damaged dopaminergic nigrostriatal circuitry that is affected [20,29]. For example, currently available surgical therapies, such as deep brain stimulation (DBS) of basal ganglia structures, effectively treat parkinsonian motor symptoms (reviewed in [74,83,110,159,198]), but can be associated with mechanical [63,197], psychiatric [271], motor 2
The validity of this assumption will be discussed later (see ‘‘Future directions’’).
[51], and cognitive complications [284]. Furthermore, the durability of benefits cannot be guaranteed, especially in the context of the continuing neurodegeneration that occurs [117,139,228,269]. Although initially effective, levodopa loses its efficacy because of a reduced ability to store levodopa presynaptically and loss of capacity to convert levodopa to dopamine as a number of endogenous dopaminergic neurons continue to degenerate [119,167]. The use of levodopa is also limited by the development of a number of deleterious side effects, such as dykinesias and the ‘‘on-off ’’ phenomenon [166]. Instead of the pulsatile delivery of dopamine associated with pharmacological management, transplantation of dopaminergic cells may enable dopamine provision to occur in a more physiological and continuous manner [291,292], thus preventing motor performance fluctuations [49,195]. Finally, studies using animal models of PD have demonstrated the ‘‘proof-of-principle’’ that transplantation therapies can effectively improve the motor symptoms seen following the loss of the dopaminergic cells of the substhantia nigra pars compacta. Initial studies, performed in the mid-1970s transplanting fetal dopaminergic neurons into the rodent model of dopamine cell loss as seen in PD, focused mainly on the histological aspects of graft survival [31,255]; later, the functional effects of transplanted cells were assessed [32,78,209]. Since these initial studies, transplanted dopaminergic neurons have been shown to survive post-transplantation [191], produce and secrete dopamine [267], form synaptic connections with host neurons [77], and attenuate both simple motor behaviors, such as asymmetrical rotational behaviors induced by the administration of dopamine agonists, as well as more complex sensorimotor deficits, such as akinesia and motor forelimb function, that resemble those experienced by parkinsonian patients [181]. Animal models of PD have also been used to refine the transplantation procedure, such as the incorporation of stereotactic techniques to deliver
Tissue transplantation for parkinson’s disease
cells to basal ganglia targets, microtransplantation techniques that minimize damage to host tissue during the cell implantation procedure, and methods to store cells prior to transplantation to facilitate screening for pathogens (reviewed in [293]).
Clinical Trials with Post-mitotic Tissue Open-label Trials On the basis of the concept that provision of dopamine to the caudate and putamen by transplanted cells would be sufficient to improve parkinsonian symptoms, without the requirement for those cells to integrate into host neural circuitries, non-neural tissues have been used for transplantation [123,124]. The cells of the adrenal medulla were considered attractive candidates for therapeutic use because they secrete catecholamines; there was evidence of their efficacy in animal models of PD and they could be autologously harvested avoiding the ethical issues associated with the use of fetal neural tissues [136]. Thus, the first clinical trials of neural transplantation for PD, performed in Sweden, utilized adrenal medullary tissue [14,150]. The procedure was found to be safe but did not impart any clinical benefits [14,150]. A subsequent report using an open microsurgical technique that involved the placement of autologous adrenal medulla on top of the caudate nucleus through the lateral ventricle was associated with significant improvement of parkinsonian symptoms [162]. This was followed by the transplantation in several hundred other patients around the world (reviewed in [75]) and the development of stereotactic procedures for the implantation of the tissue [12]. Follow-up studies in these transplanted patients demonstrated that any clinical improvements, which did not replicate those observed in the Madrazo and colleagues’ study
101
(1987), lasted only 18 months [102,158,200], however, and the procedure was associated with significant perioperative morbidity and mortality [103]. Post-mortem studies of transplanted patients demonstrated poor survival of dopaminergic cells [120,138,211]. In retrospect, it has been suggested that the preclinical data upon which the open-label trials were based did not convincingly justify the use of adrenal medullary tissue for clinical transplantation strategies [268]; even in parkinsonian rodents and monkeys, behavioral improvements were modest and shortlived, and transplanted cells poorly survived in the striatum [19,116,258]. Open-label trials in which neural tissue has been transplanted have also been conducted, with the typical source of dopaminergic neurons being those derived from the fetal ventral mesencephalon (FVM). Since these cells were first transplanted unilaterally into the caudate and putamen of patients with PD in 1987 [153,163], over 350 patients worldwide have received such grafts and the technical aspects of transplantation have been refined. Initially, solid pieces of FVM tissue were placed on the caudate nucleus via craniotomy the approach being through the lateral ventricle [164,180]. Later, stereotactic methods were devised to deliver FVM either as minced pieces [93,115,137], strands or ‘‘noodles’’ [90], or cell suspensions [175,176]. Cell suspension grafts integrate, form synaptic connections with host tissue, and become vascularized to a greater extent than cells in solid tissue grafts [100,147] and induce less displacement of the host structures into which they are transplanted [177]. As a disadvantage, cell suspension grafts must be derived from fetuses between 6 and 8 weeks gestational age in order to minimize the axotomy-induced cell death that occurs in older tissues during tissue preparation, in contrast to the up to 12-week gestational age tissue that can be used for solid FVM grafts because of the lesser degree of tissue processing that is required prior to their transplantation [94]. Proper dissection of the FVM has also been shown
1693
1694
101
Tissue transplantation for parkinson’s disease
to be an additional technical factor that influences the efficacy of neural transplantation. In two postmortem studies, improper dissection of the FVM was thought to have led to the proliferation of non-neural cells within the grafts [88,165]. While cases such as these highlight the risks associated with neural transplantation [68,170], open label clinical trials have demonstrated that it is a relatively safe neurosurgical procedure3 that can impart beneficial outcomes in at least a subset of patients. Graft-derived effects become apparent within 6–12 months following transplantation and plateau after 3–4 years, suggesting the need for graft maturation [123,122,212]. Some transplanted patients have demonstrated an increase in time spent in the ‘‘on’’ phase and concomitant decrease in time spent in the ‘‘off ’’ phase [109,135,154,155,274], improved motor function as determined by pronation/supination tests [41,175,176], timed motor tasks [71,220], and the motor component of the Unified Parkinson’s Disease Rating Scale (UPDRS) [41,109,175, 176,152,155,274], improved bradykinesia and rigidity [213], and decreased duration and intensity of dyskinesias [175,176]. In some cases, patients have been able to reduce their daily levodopa requirements [109,212,274] and the benefit has been substantial enough for patients’ quality of life to improve to the extent that they are able to return to employment [109,212, 152,176]. Anatomical studies have confirmed that transplanted FVM cells survive and remain functional for as long as a decade after transplantation [212,213]. Positron emission tomography (PET) has confirmed that intrastriatally-transplanted FVM cells can restore [18F]-dopa uptake in the dopamine-denervated striatum [212,213] (> Figure 101-1). Moreover, monitoring of 3
The most commonly reported adverse effects associated with stereotactic implantation of FVM tissue have been asymptomatic hemorrhages, subdural hematoma, transient confusion, and psychiatric symptoms (reviewed in Anonymous [10]).
synaptic dopamine release using [11C]-raclopride demonstrated that such grafts can release dopamine both basally as well in response to the administration of a dopamine agonist, and thus prevent the up-regulation of striatal D2 receptors that occurs as a consequence of dopamine depletion in the striatum [212]. These in vivo findings are complemented by those obtained from postmortem analyses of grafts from transplanted patients. Despite ongoing disease progression, transplanted dopaminergic neurons have been observed to survive, typically appearing along the periphery of the grafts (> Figure 101-2), and reinnervate up to 80% of the host striatum [92,136,137,138,177]. Axo-dendritic and axoaxonic synapses have formed between grafted and host cells, as visualized by electron microscopy [136]. Transplanted fetal cells show an ageappropriate absence of melanin and remain devoid of signs of neurodegeneration, such as the development of Lewy bodies that characterize PD [148,136,178]. Despite the promising results of some openlabel trials, there is insufficient evidence of the clinical efficacy of neural transplantation to recommend its use as a routine therapeutic procedure for PD or as sole therapy for PD symptom management [10]. This is due to the great variability in outcomes reported after transplantation. In the best cases, the benefits of transplantation have been obvious: motor function has improved to the extent that patients have no longer required any levodopa [109,212,274], FD uptake in the grafted striatum has nearly normalized [212,213], and over 100,000 surviving dopaminergic cells have been found in the grafts [137,177]. In other cases, however, transplantation has not yielded any clinically meaningful benefits [254], and in general does not improve postural stability, tremor, gait, or swallowing and speech dysfunction [17,152,239]. Indeed, clinical outcomes have varied not only between but also within centers [28,151,208,279]. These variable outcomes are likely related, at least in part, to the
Tissue transplantation for parkinson’s disease
101
. Figure 101‐1 Magnetic resonance images of a patient who received FVM transplants in both the putamen and substantia nigra bilaterally for the treatment of advanced PD (a,b,c). The tracts of the needle that was used to deliver the cell suspension into the putamen is apparent in the sagittal (a) and axial (c) views. Maps of fluorodopa uptake as visualized by positron emission tomography are shown superimposed on the magnetic resonance images (d–g). An increase in uptake was evident 28 months after transplantation (f,g) compared to pre-operatively (d,e)
. Figure 101‐2 A representative coronal section through the putamen of a patient who was transplanted with a cell suspension of FVM cells demonstrates the graft-host interface. The transplanted cells have been identified immunohistochemically with tyrosine-hydroxylase, the rate-limiting enzyme in the production of dopamine, that is present in dopaminergic neurons. Scale bar = 200 mm
1695
1696
101
Tissue transplantation for parkinson’s disease
variability in the manner in which neural transplantation procedures are performed. For example, there is variation in the age of donor tissues, manner of tissue storage and preparation for transplantation, the number of cells transplanted and the basal ganglia sites of their transplantation, and the immunosuppression regimes utilized [28,151,208,279]. The formulation of the Core Assessment Program of Intracerebral Transplantation (CAPIT) at the Fernstrom Symposium on Intracerebral Transplantation in 1990 attempted to standardize the clinical and imaging aspects of transplantation and has enabled some measure of comparison of results from different clinical trials [72,143].
Double-blind Randomized Controlled Trials To more objectively assess the effects of FVM transplantation for PD, two double-blind, randomized-controlled trials sponsored by the National Institute of Neurological Diseases and Stroke (NINDS) of the National Institutes of Health (NIH) were undertaken [90,199]. The first study by Freed and colleagues would be the first neurosurgical double-blind clinical trial [91], but even at its inception it was marked by controversy [60,277]. It was argued that the study’s base at only a single centre and utilization of only one type of transplantation procedure would be unlikely to provide definitive conclusions regarding the efficacy of cell restoration strategies for PD [60,277], and that there were ethical concerns regarding the use of placebo controls in which patients received a burr hole, with no dural opening or transplantation of cells [60,161,276]. Nevertheless, the need to control for investigator bias and placebo effects [69,95,243] were deemed necessary in order to more rigorously assess the efficacy of neural transplantation than could be done with open label trials. As the principal investigator of the
study explained: ‘‘we decided that valid conclusions could be drawn only from a large group of subjects who received an identical transplant operation while being compared with a control surgical group that underwent a surgical procedure identical to the transplant except for penetration of the brain’’ [91]. In the first clinical trial, 40 patients with advanced PD were randomly assigned to either receive FVM tissue stereotactically delivered as 200 mm diameter strands bilaterally into the putamen or placebo [90]. At 1 year post-transplantation, only transplanted patients younger than 60 years demonstrated significant improvement in subjective global rating scores, total UPDRS scores, which decreased 28% compared to pre-operative values, motor UPDRS scores, which decreased 34% compared to pre-operative values, and Schwab and England scores. These improvements were observed in the first 6 months after transplantation; they then reached a plateau. Additionally, 15% of the transplanted patients developed dystonia and dyskinesias that were difficult to control by reducing or stopping levodopa therapy. These less-than-promising results were criticized as being due to flaws in the study’s design. Only 2 donor FVMs were transplanted per putamen, which is less than what had been transplanted in many open label studies [293], and none of the patients received immunosuppression therapy. Despite PET imaging that showed a significant increase in FD uptake in the grafted putamen in both younger and older patients, in two patients whose grafts were examined at post-mortem after they died of causes unrelated to the transplantation procedures, only between 2,060 and 22,760 transplanted dopaminergic cells were detected, which is much less than what has been observed in openlabel trials [136,137,138,177]. The technique utilized to process the cells, which involved culturing them for up to 4 weeks, was considered detrimental, since it may have selected for the survival of fetal A10 dopaminergic neurons of the ventral
Tissue transplantation for parkinson’s disease
tegmental area, which do not innervate the striatum, instead of the A9 dopaminergic neurons that do [265]. Moreover, the transplantation method, a transfrontal approach delivering the cells as strands, was ‘‘unconventional’’ and contrasted with the standard approach used in most other open-label trials [142,293]. Finally, the methods of patient evaluation were considered unsatisfactory since patients were followed for only 1 year, the primary outcome measure was subjective (a patient-scored rating of clinical improvement or deterioration), and CAPIT guidelines were not followed [293]. In the more recent double-blind clinical trial [199], patients were transplanted in the putamen bilaterally with solid FVM pieces derived from either one or four fetuses per side and stored for no more than 2 days in culture prior to transplantation. Despite PETevidence of graft survival, the study’s primary endpoint was not met since the transplanted patients did not demonstrate any significant improvement in UPDRS motor scores compared to the patients who received sham transplants. Clinical improvements were observed up to 9 months after transplantation in patients who received transplants derived from four fetuses, but these diminished by the end of the study period at 24 months. Post-mortem analyses of the grafts in two patients who received grafts derived from four fetuses and died of causes unrelated to the transplantation procedure demonstrated greater transplanted dopaminergic cell survival (between 70,000 and 120,000 cells per putamen). As in the first double-blind clinical trial, a significant proportion of patients (56%) developed ‘‘off ’’ phase dyskinesias due to the transplants that in some patients required management with DBS. The results of the NIH-sponsored doubleblind clinical trials were disappointing and did not provide evidence to support neural transplantation for routine surgical management of PD. The results highlighted the need for further research to
101
optimize the variables that are thought to influence the efficacy of neural transplantation, including better determination of the patients who would be most likely to benefit from this experimental procedure, the basal ganglia structures that should be targeted to provide enhanced clinical improvements, the mechanisms of graft-induced dyskinesias, the type of immunosuppression that should be administered, and methods to improve transplanted dopaminergic cell survival and host reinnervation.
Unresolved Transplantation Issues Appropriate Patients for Transplantation The double-blind clinical trial performed by Freed and colleagues [90] identified younger patients (60 years) as benefiting the most from FVM putaminal transplants. Studies in rodent models of PD suggest that dopaminergic cell survival, neuritic outgrowth, and function are influenced by the age of the host striatal environment into which they are transplanted. For example, 75% fewer fetal dopaminergic neurons survived after transplantation into the aged compared to younger parkinsonian rats [252]. Furthermore, dopaminergic grafts that were able to ameliorate behavioral deficits in younger PD rats were ineffective in older animals, which was related to decreased transplanted dopaminergic cell survival and neuritic outgrowth in the aged striatum [61]. This age-related graft survival is associated with the decrease in the levels of neurotrophic factors, such as brain-derived neurotrophic factor (BDNF) and glial cell line-derived neurotrophic factor (GDNF), found in the striatum with advancing age [157,286,287,290]. Although the patients involved in neural transplantation clinical trials to date typically have had advanced
1697
1698
101
Tissue transplantation for parkinson’s disease
medically-recalcitrant PD, these clinical and experimental studies suggest that patients may experience greater therapeutic benefits if the transplant is done at an earlier age [91]. Alternatively, clinical outcomes following transplantation may be related to the severity of patients’ disease and not their age, per se. In open-label studies, patients who required lower preoperative doses of levodopa (850 mg/day vs. 1,350 mg/day) demonstrated greater reduction of PD symptoms after transplantation, which may reflect that their stage of PD was comparatively less advanced at the time of transplantation. This parallels the results of Freed and colleagues [90], who found that patients’ preoperative responses to levodopa correlated better with graft-induced benefits than their ages. Patients older than 60 years who responded with a greater than 60% improvement in motor scores after levodopa administration demonstrated clinical benefits similar to those observed for younger patients [90]. In the second doubleblind clinical trial [199], patients with milder PD (UPDRS scores 49 prior to transplantation) exhibited better clinical responses after transplantation compared to those with higher scores and, thus, more advanced disease [151,279]. In a rat model that mimics the ongoing dopaminergic degeneration that occurs with disease progression in PD, graft-induced amelioration of sensorimotor deficits associated with lesions of the nigrostriatal dopaminergic pathway were only attenuated following lesions that caused degeneration of additional dopaminergic systems [38]. Indeed, patients with no significant decreases in FD uptake in non-transplanted areas of the striatum have not demonstrated clinical deterioration [212], whereas patients with disease outside of the nigrostriatal dopaminergic system have experienced few graft-derived benefits [254]. Thus, transplantation therapies may be more effective in patients with neurological deficits predominantly localized to the nigrostriatal dopaminergic system.
Optimal Targets for Transplantation The putamen has been the primary target for cell placement in clinical trials of neural transplantation for PD. The degeneration of A9 substantia nigra cells that denervates the putamen of its dopaminergic input is thought to be primarily responsible for the motor dysfunction observed in PD [91,118,134,194,208]. Correspondingly, functional recovery correlates better with dopaminergic reinnervation of the putamen than the caudate in the non-human primate model of PD [219]. FVM grafts have been placed ectopically in the putamen instead of homotypically in the substantia nigra because of the inability of grafts to extend dopaminergic processes as far rostrally as their striatal targets when placed in the substantia nigra alone [16,62,78,181]. As a consequence of this inability of FVM grafts in the SN to supply dopamine to the striatum, behavioral recovery in animal models of PD remains incomplete [16,62,78,181]. Provision of dopaminergic innervation to the putamen alone may be insufficient to completely restore all PD-related motor deficits, however. In rat models of PD, striatal grafts effectively attenuate dopamine agonist-induced rotational asymmetry behaviors but only partially improve more complex sensorimotor behaviors that more closely mimic the motor deficits exhibited by patients with PD, such as skilled forelimb use [16,181,189,201,278]. Even when greater numbers of fetal dopaminergic cells are transplanted into the striatum [171] or when cells are grafted so as to provide more extensive dopaminergic reinnervation of the striatum [189,201,278], recovery of more complex behaviors remains incomplete. This may be related to the inability for FVM grafts in the striatum to influence activity in other basal ganglia nuclei that are also affected in PD. For example, FVM striatal grafts fail to normalize the metabolic activities of neurons in the subthalamic nucleus, globus
Tissue transplantation for parkinson’s disease
pallidus, and substantia nigra [186], structures implicated in the pathogenesis of parkinsonian symptomatology [73], and only partially normalize down-regulation of substance P and dynorphin expression in striatonigral output pathways that serve to inhibit the substantia nigra pars reticulata and globus pallidus [48,247]. Enhancement of neural transplantation efficacy may therefore require more complete reconstruction of the nigrostriatal dopaminergic pathway that cannot be accomplished with putaminal grafts alone. Simultaneous grafting of dopaminergic cells to the striatum and substantia nigra can partially recapitulate that circuitry, as demonstrated in animal models [173,174] in which injection of the retrograde tracer fluorogold labeled 11.5% of neurons in the nigral graft, suggesting that the nigral graft contributes to striatal reinnervation [173]. Parkinsonian animals with so-called ‘‘double grafts’’ exhibit more complete restoration of both pharmacologicalinduced and spontaneous motor behaviors [16,174,181]. This is not entirely surprising, given the crucial role of dendritically-released dopamine from substantia nigra pars compacta neurons in modulation of the basal ganglia function [52,53,99,227]. A recent phase 1 clinical study has demonstrated that simultaneous and bilateral nigral and striatal dopaminergic transplants are efficacious and can be safely performed [175]. Improvements were observed in UPDRS, Hoehn and Yahr, Schwab and England, and pronation/ supination scores in three patients followed postoperatively for up to 13 months [175]. Further, PET imaging revealed increases in FD uptake in both the putamen and substantia nigra [175]. Animal studies have also identified the subthalamic nucleus as an attractive target for transplantation [8,121,181]. Transplantation paradigms that have simultaneously targeted the striatum, substantia nigra, and subthalamic nucleus with dopaminergic grafts have induced functional recovery of complex sensorimotor behaviors in parkinsonian rats, which was thought to be related in part to
101
the role of dopamine in modulating subthalamic nucleus activity [39,43,87,89,112,179,181,294]. Recognition that the globus pallidus also receives dopaminergic afferents from the substantia nigra par compacta have supported its investigation as another potential therapeutic target [22]. It has been suggested, therefore, that transplantation could be tailored to individual patients by using FD PET to identify specific areas and extents of dopaminergic loss and accordingly varying the doses and locations of dopaminergic grafts [279]. Effective reconstruction of the basal ganglia circuitry that is affected in PD may not only require a multitarget strategy, but also a multiphenotypic approach that addresses the pathological activities of various basal ganglia structures Winkler and colleagues [278] have found, for example, that a combination of intrastriatal dopaminergic grafts, to address dopamine deficiency here, and intranigral GABAergic grafts, to inhibit the pathologically overactive nigral neurons, produced improved restoration of parkinsonian motor behaviors in a rat model of PD. These findings have recently been extended; additional inhibition of the subthalamic nucleus using either fetal- or human neural precursor cell (HNPC)-derived GABAergic grafts significantly improves forelimb motor function and akinesia [182].
Graft-induced Dyskinesias The etiology of the graft-induced dyskinesias observed in the two NIH-sponsored clinical trials is poorly understood [90,199]. Freed and colleagues [90] proposed that dyskinesias resulted from excess release of dopamine from the grafts and that this could be remedied by transplanting less FVM tissue. Although a recent study using parkinsonian rats with levodopa-induced dyskinesias supports the hypothesis that graft volume may affect the development of abnormal involuntary movements [141], others have not shown dyskinesia severity to be related to
1699
1700
101
Tissue transplantation for parkinson’s disease
dopamine production by intrastriatal grafts. Retrospective analysis of patients transplanted in open-label series found no correlation between FD uptake levels on PET and the severity of dyskinesias [108,214]. Also, there was no difference in such levels in patients who did or did not develop graft-induced dyskinesias in the second double-blind clinical trial [199]. This agrees with observations from experimental models of PD in which grafted animals demonstrate decreased propensity to develop levodopa-induced dyskinesias [45,145], which is thought to be related to the ability of transplanted cells to regulate their own dopamine release via autoreceptors [27,212]. An alternative explanation for the cause of graft-induced dyskinesias relates to the degree of striatal reinnervation provided by FVM grafts. Using voxel-based analysis to study FD uptake levels in the striatum, Ma and colleagues [160] found that patients who developed dyskinesias demonstrated small graft-derived areas of high FD uptake surrounded by deficient FD uptake. Such patchy reinnervation was not observed by Olanow and colleagues [199], however, who have suggested instead that graft-derived dyskinesias, which in their study resembled diphasic dyskinesias that occur at the end of levodopa dosing, were due to incomplete reinnervation of the striatum that led to insufficiently low intrastriatal dopamine levels. The influence of the cellular composition of FVM grafts on graft-derived reinnervation suggests another possible etiology for the dyskinesias observed in transplanted patients. The type of dopaminergic reinnervation of the striatum afforded by FVM cell transplants is related to the relative proportion of dopaminergic neuron subtypes contained within the tissue [265]. Inappropriate reinnervation of the striatum may result from grafts comprised predominantly of A10 dopaminergic neurons of the FVM [122,124], which are derived from the ventral tegmental area and normally project to non-striatal forebrain regions, instead of A9 dopaminergic neurons derived
from the substantia nigra pars compacta, whose normal target is the striatum [265]. Moreover, grafts comprised predominantly of A10 dopaminergic neurons may be unable to regulate dopamine release in a normal manner, since A10 neurons contain fewer presynaptic dopamine autoreceptors and dopamine transporters than A9 neurons [34,59,236]. Refinement of FVM tissue dissection to include only dopaminergic neurons from the A9 region may be necessary therefore, for appropriate graft-derived striatal dopaminergic reinnervation and dopamine release and thus, the prevention of graft-induced dyskinesias [123,124,177]. Dissection of FVM tissue to control graft cellular composition may also be necessarily due to the possible contribution of serotonergic cells within the tissue to the development of graft-induced dyskinesias. Serotonin neurons can store dopamine, such as that derived from levodopa administration. However, their inability to autoregulate dopamine release can lead to fluctuations in extracellular dopamine levels that subsequently induce dyskinesias. Accordingly, levodopa-induced dyskinesias could be attenuated in parkinsonian rodents by either selective serotonergic lesions or administration of serotonergic antagonists [46], and parkinsonian rats that received grafts rich in serotonergic neurons, which contained few dopaminergic neurons, exhibited worsening levodopa-induced dyskinesias over time [45]. Inclusion of sufficient quantities of dopaminergic neurons within FVM grafts to modulate the extracellular levels of dopamine released from serotonergic cells within the grafts is likely of critical importance to prevent graft-induced dyskinesias [45].
Improving Transplanted Cell Survival Clinical improvement after transplantation is related, in part, to the number of surviving
Tissue transplantation for parkinson’s disease
transplanted dopaminergic cells. Patients who have received FVM cells derived from at least three donors per hemisphere grafted have demonstrated greater clinical improvements compared to patients who have received less tissue. It has been suggested that there is a threshold of dopaminergic cell survival and reinnervation that is required for grafts to induce clinical benefits [41,90,107,109,113,135,175,176,199]. Recovery of FD uptake in the striatum to 50% of normal levels (reviewed in [107]), reinnervation of at least onethird of the striatum (reviewed in [107]), and survival of at least 100,000 dopaminergic neurons in the putamen [136,137,138] are thought to be required to induce a satisfactory clinical response [at least 30% decrease in UPDRS scores [151]]. In the two NIH-sponsored double-blind clinical trials, decreased graft viability related to lack of immunosuppression may have been partly responsible for the disappointing results [90,199]. In the study by Freed and colleagues [91], none of the patients were immunosuppressed and there was poor transplanted dopaminergic cell survival in the two patients whose grafts were assessed post-mortem. In the trial conducted by Olanow and colleagues [199], clinical benefits observed in transplanted patients began to decline once immunosuppression was stopped 6 months after transplantation. In other clinical trials, in contrast, no functional decline or poor graft survival was reported using a similar immunsuppression paradigm [175,176,177]. This outcome may be related to the use of cell suspension grafts, which are less immunogenic than solid FVM grafts [18,90,100,147,177,199,219,248]. Procurement of multiple FVMs for transplantation per patient is required due to poor transplanted dopaminergic cell survival. It is recognized that between 1 and 20% of transplanted dopaminergic neurons survive post-transplantation and their deaths occur during all stages of the procedure [79,80,81,136,138,253]. Cells undergo apoptotic and necrotic death during FVM dissection due to hypoxia and hypoglycemia, during cell
101
preparation and transplantation due to mechanical trauma, and after transplantation due to deficient neurotrophic support in the host striatal environment [79,81,140,208]. Transplantation of tissue derived from as many as 15 FVMs may therefore be required per patient [124]. A variety of technical and pharmacological methods have been used to enhance transplanted dopaminergic cell survival and, thus, reduce the number of FVMs required for transplantation (reviewed in [40]). A microtransplantation technique that utilizes a glass microcapillary for cell implantation has been shown to decrease the mechanical trauma induced to the host brain during transplantation in animal models [190]. The consequent reduction of host inflammatory response to the procedure enhanced transplanted dopaminergic cell survival [190]. A reduced host inflammatory response was also elicited by transplantation of lower doses of dopaminergic cells, which were as efficacious as higher dose grafts in attenuating dopamine agonist-induced rotational behavior over the long-term [22]. The exposure of FVM tissue during cell preparation and transplantation to neurotrophic factors [11,50,84,114,171,229], anti-apoptotic agents [58,218,240,288], anti-oxidants [1,185,238], antiexcitotoxicity agents [240], and non-ionic surfactants [216] have all been shown to enhance transplanted dopaminergic cell survival. In the clinical setting, glial cell line-derived neurotrophic factor [175,176] and the lipid peroxidation inhibitor tirilazad mesylate [41] have been used adjunctively in neural transplantation. Ultimately, widespread therapeutic application of neural transplantation is limited by logistical and ethical issues related to its reliance on fetal tissue. Logistically, FVM tissue is limited in availability, can be difficult to dissect [188], may transfer infection to the host [230], and cannot be stored for extended periods of time [231]. Ethically, concerns exist regarding the use of aborted fetal material [7,44,105,187,268], which vary depending upon whether the tissue
1701
1702
101
Tissue transplantation for parkinson’s disease
is considered a pathological specimen, research subject, or cadaveric tissue [268]. Dopaminergic neurons need to be produced in unlimited quantities in a standardized fashion for neural transplantation therapies for PD to become both practical and ethically acceptable [231].
nervous system neuraxis in embryonic and adult mammals (reviewed in [6,97,104]), expand them in the presence of mitogens for long periods of time [295] and in potentially unlimited quantities in culture without adversely modifying their karyotypes [128], and to potentially differentiate them into clinically-useful phenotypes, like dopaminergic neurons for PD [20].
Stem Cell Transplantation Strategies Defining Stem Cells As an alternative to post-mitotic dopaminergic cells derived from FVM tissue, stem cells are attractive candidates for cell restoration strategies for PD. Mitotically-active cells were originally identified in the adult mammalian hippocampus and olfactory bulb in the 1960s [3,4]. Since then, it has been shown that cells can be isolated from both the embryonic and adult mammalian nervous system, expanded in culture, and later differentiated into all three classes of cells in the central nervous system [221,222]. Although these have been looselytermed stem cells or neural stem cells, Seaberg and colleagues [242] argue that the terminology used to describe ‘‘stem cells’’ should depend upon their biological behaviors. According to them, inherent stem cell characteristics are indefinite self-renewal and multipotentiality [242]. Embryonic stem cells are isolated from the inner cell mass of the blastocyst and have the potential to differentiate into cells that comprise all of the embryonic germ layers (totipotent) [85,168]. In contrast, neural stem and precursor cells4 display more restricted abilities to differentiate and self-renew over long periods of time (reviewed in [15,242]). The properties that make neural precursor cells potentially applicable to neural transplantation are the ability to isolate them from all levels of the central 4
Henceforward referred to as ‘‘neural precursor cells.’’
Differentiation of Stem Cells Recent studies in animal models of PD have demonstrated the feasibility of using embryonic stem cells and neural precursor cells as an alternative to FVM tissue. The propensity for neural precursor cells to differentiate into neurons decreases with time in culture, with GABA as the default differentiation phenotype [125]. The central nervous system environment into which the cells are transplanted is also thought to influence their phenotypic differentiation. Neural precursor cells demonstrate region-specific differentiation when transplanted into neurogenic regions [96,98,244,260], such as the dentate gyrus of the hippocampus or the subventricular zone of the lateral ventricles, but primarily become astrocytes when transplanted into nonneurogenic regions, such as the striatum and substantia nigra [126,183,261]. Differentiation of neural precursor cells into dopaminergic neurons has been reported when the cells are transplanted into lesioned environments [282], but in general spontaneous dopaminergic differentiation is rare [183,261]. In contrast, embryonic stem cells can more readily differentiate into dopaminergic cells [70], especially when transplanted in low numbers [30]. Methods have been developed to increase the proportion of embryonic stem cells and neural precursor cells that differentiate into a dopaminergic phenotype. Modification of the environment in which the cells are cultured can enhance their dopaminergic differentiation, which has
Tissue transplantation for parkinson’s disease
been accomplished by exposure of cells to low oxygen levels [257], cytokines [156,215], ascorbic acid [280], retinoic acid [205], forskolin [224,225,225,273], and neurotrophic factors [106,205,224,232,225,225]. Recapitulation of the environment in which dopaminergic neurons develop has also been used to guide the differentiation of stem and precursor cells into a dopaminergic phenotype. In culture, this has been done by exogenously exposing the cells to factors involved in midbrain neuron specification, such as sonic hedgehog and fibroblast growth factor 8 [132,149,210,233,270], or by genetically modifying them to overexpress key regulatory genes involved in dopaminergic neuron differentiation, such as Nurr1 [55,131,132,234,237], Pitx3 [169,193], Lmx1a [9], Lmx1b [249], Engrailed 1 and 2 [245,246], or Wnt1 [207]. Embryonic stem cells and neural precursor cells differentiated in these ways have ameliorated both simple [259,296,297] and more complex sensorimotor behaviors [233,298] after transplantation into animal models of PD.
Appropriate Stem Cell Differentiation and Transplantation In order for transplanted stem and precursor cells to elicit functional improvements in PD, it may not be sufficient to simply differentiate them into dopaminergic cells since not all dopaminergic cells are alike [123,133,151]. The A9 dopaminergic neurons of the substantia nigra pars compacta, which express the G-protein-gatedinwardly rectifying K+ channel subunit, preferentially innervate the striatum [65,265,299], whereas the A10 dopaminergic neurons of the ventral tegmental area, which express calbindin and cholecystokinin, preferentially innervate non-striatal forebrain regions [101,265,299]. Both of these dopaminergic neuron subtypes
101
are included in FVM grafts, with the A9 neurons typically observed around the periphery of the graft at the interface with the host striatum and A10 neurons within the centre of the graft [177]. Differentiation of stem and precursor cells into dopaminergic cells with the A9 phenotype may be preferable than their differentiation into nonspecific or the A10 dopaminergic cells in order to improve graft-derived innervation of the transplanted striatum [123]. Although embryonic stem cell-derived grafts may be comprised of large numbers of non-specific dopaminergic cells, some do not produce extensive dopaminergic reinnervation of the striatum [25,210]. Indeed, parkinsonian animals transplanted with cells derived from the medial VM, which contain aldehyde dehydrogenase expressing A9 dopaminergic neurons, demonstrated appropriate reinnervation of striatal regions important for motor control and behavioral recovery of PD motor symptoms [111]. Over-expression of Pitx3 in stem and precursor cells may be a method of enhancing the yield of A9 differentiated dopaminergic cells [56]. Would the best transplantation strategy, then, involve transplantation of a homogenous population of A9 dopaminergic differentiated stem or precursor cells? It has been argued that non-dopaminergic cells contained within transplants could form ‘‘aberrant or unnecessary’’ synapses with host neurons that could dampen or worsen functional improvements mediated by the transplanted dopaminergic cells [13]. Certainly, it is possible to generate pure populations of dopaminergic cells for transplantation using fluorescence activated cell sorting technology [57,285,289] and parkinsonian animals have been transplanted with these cells suspensions [285]. It is clear, however, that nondopaminergic cells contained within either FVM- or stem- or precursor cell-derived grafts may be critical to the differentiation and survival of the dopaminergic cells of the grafts.
1703
1704
101
Tissue transplantation for parkinson’s disease
For example, a variety of cellular feeder layers are employed to facilitate the differentiation of embryonic stem cells into dopaminergic cells in culture [130,132,210,233,262,263], suggesting that cell-cell contact or factors secreted by feeder layer cells are necessary for dopaminergic differentiation [13]. Astrocytes secrete factors that enhance neuronal differentiation in general [241], such as those within the hippocampus [250], and dopaminergic differentiation in particular [272], such as those within the FVM (Parish et al. 2007). Neuronal maturation and synapse formation [251,266] and neuritic outgrowth of neurons is also beneficially influenced by astrocytes [127], suggesting that their inclusion in dopaminergic grafts is crucial.
Integration of Stem Cells with Host Neural Circuits The efficacy of stem and precursor cells in cell restoration strategies for PD will likely depend upon their ability to integrate into host neural circuitries at least as well as transplanted FVM cells [29,231]. In patients, the demonstration that FVM grafts can activate frontal motor cortical areas and that increases in blood oxygenation levels are observed in functional magnetic resonance imaging studies in response to repetitive motor tasks suggests that these grafts become functionally integrated [21,35,212]. In animal models of PD, transplanted FVM cells demonstrate the electrophysiological characteristics of substantia nigra dopaminergic neurons, including the ability to generate action potentials, and receive excitatory and inhibitory synaptic inputs from host neurons [300]. Synaptic integration of transplanted dopaminergic cells is thought to be necessary for dopamine to be released in a regulated way from the graft [123,124]. Transplanted stem and precursor cells can also demonstrate appropriate integration with host neural circuitries. Anatomically, synaptophysin expression on host axon terminals adjacent to
transplanted cell dendrites, and dendritic expression of postsynaptic density 95 on transplanted cells has suggested the presence of synaptic connections between transplanted embryonic stem cells and host neurons [275]. Electrophysiological experiments have confirmed that transplanted stem and precursor cells can fire action potentials [82,132,145,256,298] and receive synaptic inputs from host neurons [26,82,145,301], including inhibitory post-synaptic potentials that are unique to midbrain dopaminergic neurons [132]. Extracellular stimulation of transplanted cells causes excitatory post-synaptic potentials in host neurons, extracellular stimulation of host neurons elicit inhibitory post-synaptic potentials in transplanted cells, further confirming graft-host integration [132].
Therapeutic Applications of Stem Cell Trophic Effects Transplanted stem and precursor cells need not necessarily integrate into host neural circuitries to induce beneficial functional effects in PD. Neural precursor cells can constitutively secrete a variety of trophic factors that enhance the survival of dopaminergic neurons, including glial cell line-derived neurotrophic factor [202], sonic hedgehog [217], and stem cell factor [283]. When transplanted into parkinsonian animals, these cells protect endogenous dopaminergic neurons from neurotoxin-induced cell death [202,283] and prevent the compensatory increase in dopaminergic neurons that is observed in the striatum [33], as well as increase the survival of co-transplanted fetal dopaminergic neurons, thus reducing the number of FVMs required for transplantation [184,217].
Stem Cell Transplantation Caveats As promising as stem- and precursor-derived dopaminergic cells are as an alternative cell
Tissue transplantation for parkinson’s disease
source to FVM tissue for neural transplantation for PD, their clinical use may be hampered by a number of factors. First, the efficiency of dopaminergic differentiation needs to be optimized. Differentiation of human embryonic stem cells is more easily performed than of precursor cells [25,210,233,281], but even so, reports of the derivation of homogenous populations of dopaminergic cells are rare [54]. A recent study found that the proportion of transplanted embryonic stem cells that demonstrated a dopaminergic phenotype decreased with time, suggesting that additional instructive cues may be required in order for the cells to maintain that phenotype [233]. Additionally, incomplete differentiation of cells, such that they do not express all of the transcription factors of subtantia nigra pars compacta dopaminergic neurons, may lead to their inappropriate integration into host neural circuitries, with potentially detrimental consequences [275]. Second, the survival of transplanted stemand precursor-derived dopaminergic cells is low, ranging from as low as 0.13–5% [47,206,259]. Third, grafts derived from these cells can form tumors [30,37,233] due to the presence of undifferentiated cells within the grafts that continue to proliferate. Human embryonic stem cells [302,303] and mouse neural precursor cells [128] have been shown to remain karyotypically normal after extended passage in culture, but long-term studies of the tumorigenic potential of grafts are still lacking [104]. Methods to terminally differentiate the cells may inadvertently make the cells tumorigenic, for example, by insertational mutagenesis during genetic modification [170]. It may be possible to decrease the potential for grafts to form tumors by removing undifferentiated cells from suspensions prior to transplantation (using FACS, for example), removing genes not essential for dopaminergic differentiation that affect proliferation (such as cripto, for example) [204], or inserting genes that render the cells lethally susceptible to specific medications (such as the herpes virus thymidine kinase gene) [24,86]. Fourth, differentiated cells that require
101
co-culture with non-human cell feeder layer systems for differentiation may become infected with animal pathogens [226]. It has been recommended that only cells that have not been in contact with animal products during their derivation, expansion, and differentiation would be potentially clinically applicable [68]. Finally, migration of transplanted stem and precursor cells away from sites of transplantation [42,126,183] may lead to the formation of heterotopias or aberrant synaptic connections with host neurons. As has been recently reviewed, a variety of safety and efficacy issues need to be resolved before stem and precursor cells can be considered suitable for clinical application [170].
Conclusions Almost 450 years after Ambroise Pare´’s allusion to brain restoration, the proof-of-principle that neural transplantation therapies for PD are feasible has been well-established. Current research trends have focused on the identification of the ideal cell source for transplantation (dopaminergic cells that can be generated in unlimited quantities and in a standardized fashion). By itself, however, this approach fails to adequately address the shortcomings that limit the interpretation of the open-label and double-blind clinical trials as less-than-promising. Improvement of the efficacy of neural transplantation will require greater understanding of the pathophysiology of PD, the biology of transplanted cells and the factors that influence graft survival and host reinnervation. The clinical variables that will lead to optimal patient selection also have to be understood. Cell restoration to repair the brain in PD will likely require a multimodality approach that may include the use of neurotrophic factors and targeting of multiple basal ganglia structures since it is clear the symptoms of PD also result from neurodegeneration outside of the nigrostriatal dopaminergic system [2,36,142]. As knowledge of the mechanisms of brain repair improves and progress in stem cell technologies
1705
1706
101
Tissue transplantation for parkinson’s disease
advances, the realization of Pare’s vision of brain restoration may be brought closer to reality.
References 1. Agrawal AK, Chaturvedi RK, Shukla S, Seth K, Chauhan S, Ahmad A, Seth PK. Restorative potential of dopaminergic grafts in presence of antioxidants in rat model of Parkinson’s disease. J Chem Neuroanat 2004; 28:253-64. 2. Ahlskog JE. Beating a dead horse – dopamine and Parkinson disease. Neurology 2007;69:1701-11. 3. Altman J. Autoradiographic and histological studies of postnatal neurogenesis IV. Cell proliferation and migration in the interior forebrain, with special reference to persisting neurogenesis in the olfactory bulb. J Comp Neurol 1969;137:433-57. 4. Altman J, Das GD. Autoradiographic and histological evidence of postnatal hippocampal neurogenesis in rats. J Comp Neurol 1965;124:319-35. 5. Altobelli R. Inesti cerebrali. Gazz Gazzetta Internazionale di Medicina e Chirurgia 1914;17:25-34. 6. Alvarez-Buylla A, Garcia-Verdugo JM. Neurogenesis in adult subventricular zone. J Neurosci 2002;22:629-34. 7. American Medical Association. Medical applications of fetal tissue transplantation: council on scientific affairs, American Medical Association. J Am Med Assoc 1990;263:565-70. 8. Anderson L, Caldwell MA. Human neural progenitor cell transplants into the subthalamic nucleus lead to functional recovery in a rat model of Parkinson’s disease. Neurobiol Dis 2007;27:133-40. 9. Andersson E, Tryggvason U, Deng Q, Friling S, Alekseenko Z, Robert B, Perlmann T, Ericson J. Identification of intrinsic determinants of midbrain dopamine neurons. Cell 2006;124:393-405. 10. Anonymous. Surgical treatment for Parkinson’s disease: neural transplantation. Mov Dis 2002;17 Suppl 4: S148-55. 11. Apostolides C, Sanford E, Hong M, Mendez I. Glial cell line-derived neurotrophic factor improves intrastriatal graft survival of stored dopaminergic cells. Neuroscience 1998;83(2):363-72. 12. Apuzzo ML, Neal JH, Waters CH, Appley AJ, Boyd SD, Couldwell WT, Wheelock VH, Veiner LP. Utilization of unilateral and bilateral stereotactically placed adrenomedullary-striatal autografts in parkinsonian humans: rationale, techniques, and observations. Neurosurgery 1990;26:746-57. 13. Arenas E. Engineering a dopaminergic phenotype in stem/precursor cells: role of Nurr1, glia-derived signals, and Wnts. Ann NY Acad Sci 2005a;1049:51-66.
14. Backlund E-O, Granberg P-O, Hamberger B, Knutson E, Martensson A, Sedvall G, Seiger A, Olson L. Transplantation of adrenal medullary tissue to striatum in parkinsonism. First clinical trials. J Neurosurg 1985;62:169-73. 15. Baizabal J-M, Furlan-Magaril M, Santa-Olalla J, Covarrubias L. Neural stem cells in development and regenerative medicine. Arch Med Res 2003;34:572-88. 16. Baker KA, Sadi D, Hong M, Mendez I. Simultaneous intrastriatal and intranigral dopaminergic grafts in the parkinsonian rat model: role of the intranigral graft. J Comp Neurol 2000;426(1):106-16. 17. Baker KK, Ramig LO, Johnson AB, Freed CR. Preliminary voice and speech analysis following fetal dopamine transplants in 5 individuals with Parkinson disease. J Speech Lang Hear Res 1997;40:615-26. 18. Baker-Cairns BJ, Sloan DJ, Broadwell RD, Puklavec M, Charlton HM. Contributions of donor and host blood vessels in CNS allografts. Exp Neurol 1996;142:36-46. 19. Bankiewicz KS, Plunkett RJ, Kopin IJ, Jacobowitz DM, London WT, Olfield E. Transient behavioral recovery in hemiparkinsonian primates after adrenal medullary allografts. Prog Brain Res 1988;78:543-50. 20. Barker RA. Repairing the brain in Parkinson’s disease: where next? Mov Dis 2002;17(2):233-41. 21. Barker RA, Dunnett SB. Functional integration of neural grafts in Parkinson’s disease. Nat Neurosci 1999;2(12): 1047-8. 22. Bartlett LE, Sadi D, Lewington M, Mendez I. Functional improvement with low-dose dopaminergic grafts in hemiparkinsonian rats. Neurosurgery 2004;55:405-15. 23. Bartlett LE, Mendez I. Dopaminergic reinnervation of the globus pallidus by fetal nigral grafts in the rodent model of Parkinson’s disease. Cell Transplant 2005; 14(2–3):119-27. 24. Barzon L, Bonaguro R, Castagliuolo I, Chilosi M, Gnatta E, Parolin C, Boscaro M, Palu` G. Transcriptionally targeted retroviral vector for combined suicide and immunomodulating gene therapy of thyroid cancer. J Clin Endocrinol Metab 2002;87:5304-5311. 25. Ben-Hur T, Idelson M, Khaner H, Pera M, Reinhartz E, Itzik A, Reubinoff BE. Transplantation of human embryonic stem cell-derived neural progenitors improves behavioral deficit in parkinsonian rats. Stem Cells 2004; 22:1246-55. 26. Benninger F, Beck H, Wernig M, Tucker KL, Bru¨stle O, Scheffler B. Functional integration of embryonic stem cellderived neurons in hippocampal slice cultures. J Neurosci 2003;23(18):7075-83. 27. Bjo¨rklund A. Dopaminergic transplants in experimental parkinsonism: cellular mechanisms of graft-induced functional recovery. Curr Opin Neurobiol 1992;2:683-9. 28. Bjo¨rklund A, Dunnett SB, Brundin P, Stoessl AJ, Freed CR, Breeze RE, Levivier M, Peschanski M, Studer L, Barker R. Neural transplantation for the treatment of Parkinson’s disease. Lancet Neurol 2003;2(7): 437-45.
Tissue transplantation for parkinson’s disease
29. Bjo¨rklund A, Lindvall O. Cell replacement therapies for central nervous system disorders. Nat Neurosci 2000;3:537-44. 30. Bjo¨rklund LM, Sa´nchez-Pernaute R, Chung S, Andersson T, Chen IYC, McNaught KSTP, Brownell A-L, Jenkins BG, Wahlestedt C, Kim K-S, Isacson O. Embryonic stem cells develop into functional dopaminergic neurons after transplantation in a Parkinson rat model. Proc Natl Acad Sci USA 2002;99(4):2344-9. 31. Bjo¨rklund A, Stenevi U, Svendgaard NA. Growth of transplanted monoaminergic neurones into the adult hippocampus along the perforant path. Nature 1976; 262:787-90. 32. Bjo¨rklund A, Stenevi U. Reconstruction of the nigrostriatal dopamine pathway by intracerebral nigral transplants. Brain Res 1979;177:555-60. 33. Bjugstad KB, Redmond DE, Jr, Teng YD, Elsworth JD, Roth RH, Blanchard BC, Snyder EY, Sladek JR, Jr. Neural stem cells implanted into MPTP-treated monkeys increase the size of endogenous tyrosine hydroxylasepositive cells found in the striatum: a return to control measures. Cell Transplant 2005;14(4):183-92. 34. Blanchard V, Raisman-Vozari R, Vyas S, Michel PP, Javoy-Agid F, Uhl G, Agid Y. Differential expression of tyrosine hydroxylase and membrane dopamine transporter genes in subpopulations of dopaminergic neurons of the rat mesencephalon. Brain Res Mol Brain Res 1994;22(1–4):29-38. 35. Blu¨ml S, Kopyov O, Jacques S, Ross BD. Activation of neurotransplants in humans. Exp Neurol 1999;158: 121-5. 36. Braak H, Del Tredici K, Rub U, de Vos RA, Jansen Steur EN, Braak E. Staging of brain pathology related to sporadic Parkinson’s disease. Neurobiol Aging 2003;24: 197-211. 37. Brederlau A, Correia AS, Anisimov SV, Elmi M, Paul G, Roybon L, Morizane A, Berqqist F, Riebe I, Nannmark U, Carta M, Hanse E, Takahashi J, Sasai Y, Funa K, Brundin P, Eriksson PS, Li JY. Transplantation of human embryonic stem cell-derived cells to a rat model of Parkinson’s disease: effect of in vitro differentiation on graft survival and teratoma formation. Stem Cells 2006;24(6):1433-40. 38. Breysse N, Carlsson T, Winkler C, Bjo¨rklund A, Kirik D. The functional impact of the intrastriatal dopamine neuron grafts in parkinsonian rats is reduced with advancing disease. J Neurosci 2007;27(22):5849-56. 39. Brown L, Makman MH, Wolfson LI, Dvorkin B, Warner C, Katzman R. A direct role of dopamine in the rat subthalamic nucleus and adjacent intrapedencular area. Science 1979;206:1416-8. 40. Brundin P, Karlsson J, Emga˚rd M, Kaminski Schierle GS, Hansson O, Peterse´n A˚, Castilho RF. Improving the survival of grafted dopaminergic neurons: a review over current approaches. Cell Transplant 2000b;9:179-95.
101
41. Brundin P, Pogarell O, Hagell P, Piccini P, Widner H, Schrag A, Kupsch A, Crabb L, Odin P, Gustavii B, Bjo¨rklund A, Brooks DJ, Marsden CD, Oertel WH, Quinn NP, Rehncrona S, Lindvall O. Bilateral caudate and putamen grafts of embryonic mesencephalic tissue treated with lazaroids in Parkinson’s disease. Brain 2000;123:1380-90. 42. Burnstein RM, Foltynie T, He X, Menon DK, Svendsen CN, Caldwell MA. Differentiation and migration of long term expanded human neural progenitors in a partial lesion model of Parkinson’s disease. Int J Biochem Cell Biol 2004;36:702-13. 43. Campbell GA, Eckardt MJ, Weight FF. Dopaminergic mechanisms in the subthalamic nucleus of rat: analysis using horseradish peroxidase and microiontophoresis. Brain Res 1985;333:261-70. 44. Caplan AL. Should foetuses or infants be utilized as organ donors? Bioethics 1987;1:119-40. 45. Carlsson T, Carta M, Winkler C, Bjo¨rklund A, Kirik D. Serotonin neuron transplants exacerbate L-DOPAinduced dyskinesias in a rat model of Parkinson’s disease. J Neurosci 2007;27(30):8011-22. 46. Carta M, Carlsson T, Kirik D, Bjo¨rklund A. Dopamine released from 5-HT terminals is the cause of L-dopainduced dyskinesia in parkinsonian rats. Brain 2007; 130:1819-33. 47. Carvey PM, Ling ZD, Sortwell CE, Pitzer MR, McGuire SO, Storch A, Collier TJ. A clonal line of mesencephalic progenitor cells converted to dopamine neurons by hematopoietic cytokines: a source of cells for transplantation in Parkinson’s disease. Exp Neurol 2001;171(1): 98-108. 48. Cenci MA, Campbell K, Bjo¨rklund A. Neuropeptide messenger RNA expression in the 6-hydroxydopaminelesioned rat striatum reinnervated by fetal dopaminergic transplants: differential effects of the grafts on preproenkephalin, preprotachykinin and prodynorphin messenger RNA levels. Neuroscience 1993;57:275-96. 49. Chase TN, Baronti F, Fabbrini G, Heuser IJ, Juncos JL, Mouradian MM. Rationale for continuous dopaminomimetic therapy of Parkinson’s disease. Neurology 1989;39 (11 Suppl 2):7-10. 50. Chaturvedi RK, Shukla S, Seth K, Agrawal AK. Nerve growth factor increases survival of dopaminergic graft, rescue nigral dopaminergic neurons and restores functional deficits in rat model of Parkinson’s disease. Neurosci lett 2006;398:44-9. 51. Chen CC, Bru¨cke C, Kempf F, Kupsch A, Lu CS, Lee ST, Tisch S, Limousin P, Hariz M, Brown P. Deep brain stimulation of the subthalamic nucleus: a two-edged sword. Curr Biol 2006;16(22):R952-3. 52. Cheramy A, Nieoullan A, Glowinski J. In vivo evidence for a dendritic release of dopamine in cat substantia nigra. Appl Neurophysiol 1979;42:57-9. 53. Cheramy A, Leviel V, Glowinski J. Dendritic release of dopamine in the substantia nigra. Nature 1981;289: 537-42.
1707
1708
101
Tissue transplantation for parkinson’s disease
54. Cho MS, Lee YE, Kim JY, Chung S, Cho YH, Kim DS, Kang SM, Lee H, Kim MH, Kim JH, Leem JW, Oh SK, Choi YM, Hwang DY, Chang JW, Kim DW. Highly efficient and large-scale generation of functional dopamine neurons from human embryonic stem cells. Proc Natl Acad Sci USA 2008;105(9):3392-7. 55. Chung S, Sonntag K-C, Andersson T, Bjorklund LM, Park JJ, Kim D-W, Kang UJ, Isacson O, Kim K-S. Genetic engineering of mouse embryonic stem cells by Nurr1 enhances differentiation and maturation into dopaminergic neurons. Eur J Neurosci 2002;16:1829-38. 56. Chung S, Hedlund E, Hwang M, Kim DW, Shin BS, Hwang DY, Jung Kang U, Isacson O, Kim KS. The homeodomain transcription factor Pitx3 facilitates differentiation of mouse embryonic stem cells into AHD2expressing dopaminergic neurons. Mol Cell Neurosci 2005;28:241-52. 57. Chung S, Shin BS, Hedlund E, Pruszak J, Ferree A, Kang UJ, Isacson O, Kim KS. Genetic selection of sox1GFP-expressing neural precursors removes residual tumorigenic pluripotent stem cells and attenuates tumor formation after transplantation. J Neurochem 2006;97: 1467-80. 58. Cicchetti F, Costantini L, Belizaire R, Burton W, Isacson O, Fodor W. Combined inhibition of apoptosis and complement improves neural graft survival of embryonic rat and porcine mesencephalon in the rat brain. Exp Neurol 2002;177:376-84. 59. Ciliax BJ, Drash GW, Staley JK, Haber S, Mobley CJ, Miller GW, Mufson EJ, Mash DC, Levey AI. Immunocytochemical localization of the dopamine transporter in human brain. J Comp Neurol 1999;409(1):38-56. 60. Cohen J. New fight over fetal tissue grafts. Science 1994;263:600-1. 61. Collier TJ, Sortwell CE, Daley BF. Diminished viability, growth, and behavioral efficacy of fetal dopamine neuron grafts in aging rats with long-term dopamine depletion: an argument for neurotrophic supplementation. J Neurosci 1999;19(13):5563-73. 62. Collier TJ, Sortwell CE, Elsworth JD, Taylor JR, Roth RH, Sladek, Jr, JR, Redmond, Jr, DE. Embryonic ventral mesencephalic grafts to the substantia nigra of MPTP-treated monkeys: feasibility relevant to multipletarget grafting as a therapy for Parkinson’s disease. J Comp Neurol 2002;442(4):320-30. 63. Constantoyannis C, Berk C, Honey CR, Mendez I, Brownstone RM. Reducing hardware-related complications of deep-brain stimulation. Can J Neurol Sci 2005;32(2):194-200. 64. Dahlstrom A, Fuxe K. Evidence for the existence of monoamine-containing neurons in the central nervous system. I. Demonstration of monoamines in the cell bodies of brain stem neurons. Acta Phyiological Scandinavica 1964;62 Suppl 232:1-55. 65. Damier P, Hirsch EC, Agid Y, Graybiel AM. The substantia nigra of the human brain. I. Nigrosomes and the
66. 67.
68.
69. 70.
71.
72.
73. 74.
75.
76.
77.
78.
nigral matrix, a compartmental organization based on calbindin D(28K) immunohistochemistry. Brain 1999; 122:1421-36. Das GD. Neural transplantation: an historical perspective. Neurosci Biobehav Rev 1990;14:389-401. Dass B, Olanow CW, Kordower JH. Gene transfer of trophic factors and stem cell grafting as treatments for Parkinson’s disease. Neurology 2006;66 Suppl 4: S89-103. Dawson L, Bateman-House AS, Agnew DM, Bok H, Brock DW, Chakravarti A, Greene M, King PA, O’Brien SJ, Sachs DH, Schill KE, Siegel A, Solter D, Suter SM, Verfaillie CM, Walters LB, Gearhart JD, Faden RR. Safety issues in cell-based intervention trials. Fertil Steril 2003;80(5):1077-85. de la Fuente-Ferna´ndez R, Stoessl AJ. The placebo effect in Parkinson’s disease. Trends Neurosci 2002;25(6):302-6. Deacon T, Dinsmore J, Costantini LC, Ratliff J, Isacson O. Blastula-stage stem cells can differentiate into dopaminergic and serotonergic neurons after transplantation. Exp Neurol 1998;149:28-41. Defer GL, Geny C, Ricolfi F, Fenelon G, Monfort JC, Remy P, Villafane G, Jeny R, Samson Y, Keravel Y, Gaston A, Degos JD, Peschanski M, Cesaro P, Nguyen JP. Long-term outcome of unilaterally transplanted parkinsonian patients. I. Clinical approach. Brain 1996;119:41-50. Defer GL, Widner H, Marie´ RM, Re´my P, Levivier M. Core assessment program for surgical interventional therapies in Parkinson’s disease (CAPSIT-PD). Mov Dis 1999;14:572-84. DeLong MR. Primate models of movement disorders of basal ganglia origin. Trends Neurosci 1990;13:281-5. Deuschl G, Schade-Brittinger C, Krack P, Volkmann J, Scha¨fer H, Bo¨tzel K, Daniels C, Deutschla¨nder A, Dillmann U, Eisner W, Gruber D, Hamel W, Herzog J, Hilker R, Klebe S, Kloss M, Koy J, Krause M, Kupsch A, Lorenz D, Lorenzl S, Mehdorn HM, Moringlane JR, Oertel W, Pinsker MO, Reichmann H, Reuss A, Schneider GH, Schnitzler A, Steude U, Sturm V, Timmermann L, Tronnier V, Trottenberg T, Wojtecki L, Wolf E, Poewe W, Voges J. German Parkinson Study Group, Neurostimulation Section. A randomized trial of deep-brain stimulation for Parkinson’s disease. N Engl J Med 206;355(9): 896-908. Drucker-Colı´n R, Verdugo-Dı´az L. Cell transplantation for Parkinson’s disease: present status. Cell Mol Neurobiol 2004;24(3):301-16. Dunn EH. Primary and secondary findings in a series of attempts to transplant cerebral cortex in the albino rat. J Comp Neurol 1917;27:565-82. Dunnett SB. Functional repair of striatal systems by neural transplants: evidence for circuit reconstruction. Behav Brain Res 1995;66:133-42. Dunnett SB, Bjo¨rklund A, Schmidt RH, Stenevi U, Iversen SD. Intracerebral grafting of neuronal cell
Tissue transplantation for parkinson’s disease
suspensions. IV. Behavioral recovery in rats with unilateral 6-OHDA lesions following implantation of nigral cell suspensions in different forebrain sites. Acta Physiol Scand Suppl 1983;522:29-37. 79. Emga˚rd M, Blomgren K, Brundin P. Characterisation of cell damage and death in embryonic mesencephalic tissue: a study on ultrastructure, vital stains and protease activity. Neuroscience 2002;115(4):1177-87. 80. Emga˚rd M, Karlsson J, Hansson O, Brundin P. Patterns of cell death and dopaminergic neuron survival in intrastriatal nigral grafts. Exp Neurol 1999;160:279-88. 81. Emga˚rd M, Hallin U, Karlsson J, Bahr BA, Brundin P, Blomgren K. Both apoptosis and necrosis occur early after intracerebral grafting of ventral mesencephalic tissue: a role for protease activation. J Neurochem 2003;86:1223-32. 82. Englund U, Bjo¨rklund A, Wictorin K, Lindvall O, Kokaia M. Grafted neural stem cells develop into functional pyramidal neurons and integrate into host cortical circuitry. Proc Natl Acad Sci USA 2002;99(26): 17089-94. 83. Eskander E, Cosgrove GR, Shinobu LA. Surgical treatment of Parkinson’s disease. J Am Med Assoc 2001;286(4):3056-9. 84. Espejo M, Cutillas B, Arenas E, Ambrosio S. Increased survival of dopaminergic neurons in striatal grafts of fetal ventral mesencephalic cells exposed to neurotrophin-3 or glial cell line-derived neurotrophic factor. Cell Transplant 2000;9:45-53. 85. Evans MJ, Kaufman MH. Establishment in culture of pluripotential cells from mouse embryos. Nature 1981; 292:154-6. 86. Fareed MU, Moolten FL. Suicide gene transduction sensitizes murine embryonic and human mesenchymal stem cells to ablation on demand – a fail-safe protection against cellular misbehavior. Gene Ther 2002;9:955-62. 87. Flores G, Liang LL, Sierra A, Martı´nez-Fong D, Quirion R, Aceves J, Srivastava LK. Expression of dopamine receptors in the subthalamic nucleus of the rat: characterization using reverse transcriptase-polymerase chain reaction and autoradiography. Neuroscience 1999;91:549-56. 88. Folkerth RD, Durso R. Survival and proliferation of nonneural tissues, with obstruction of cerebral ventricles, in a parkinsonian patient treated with fetal allografts. Neurology 1996;46:1219-25. 89. Franc¸ois C, Savy C, Jan C, Tande D, Hirsch EC, Yelknik J. Dopaminergic innervation of the subthalamic nucleus in the normal state, in MPTP-treated monkeys, and in Parkinson’s disease patients. J Comp Neurol 2000;425:121-9. 90. Freed CR, Greene PE, Breeze RE, Tsai W-Y, DuMouchel W, Kao R, Dillon S, Winfield H, Culver S, Trojanowski JQ, Eidelberg D, Fahn S. Transplantation of embryonic dopamine neurons for severe Parkinson’s disease. N Engl J Med 2001;344(10):710-9.
101
91. Freed CR, Leehey MA, Zawada M, Bjugstad K, Thompson L, Breeze RE. Do patients with Parkinson’s disease benefit from embryonic dopamine cell transplantation? J Neurol 2003;250 Suppl 3:III44-46. 92. Freed CR, Breeze RE, Rosenberg NL, Schneck SA, Kriek E, Qi J-X, Lone T, Zhang Y-B, Snyder JA, Wells TH, Ramig LO, Thompson L, Mazziotta JC, Huang SC, Grafton ST, Brooks D, Sawle G, Schroter G, Ansari AA. Survival of implanted fetal dopamine cells and neurologic improvement 12 to 46 months after transplantation for Parkinson’s disease. N Eng J Med 1992; 327:1549-55. 93. Freeman TP, Olanow CW, Hauser RA, Nauert GM, Smith DA, Borlongan CV, Sanberg PR, Holt DA, Kordower JH, Vingerhoets FJ, et al. Bilateral fetal nigral transplantation into the postcommissural putamen in Parkinson’s disease. Ann Neurol 1995a; 38(3):379-88. 94. Freeman TP, Sanberg PR, Nauert GM, Boss BD, Spector D, Olanow CW, Kordower JH. The influence of donor age on the survival of solid and suspension intraparenchymal human embryonic nigral grafts. Cell Transplant 1995b; 4(1):141-54. 95. Freeman TB, Vawter DE, Leaverton PE, Godbold JH, Hauser RA, Goetz CG, Olanow CW. Use of placebo surgery in controlled trials of a cellular-based therapy for Parkinson’s disease. N Eng J Med 1999;341 (13):988-92. 96. Fricker RA, Carpenter MK, Winkler C, Greco C, Gates MA, Bjo¨rklund A. Site-specific migration and neuronal differentiation of human neural progenitor cells after transplantation in the adult rat brain. J Neurosci 1999; 19:5990-6005. 97. Gage FH. Mammalian neural stem cells. Science 2000;287:1433-8. 98. Gage FH, Coates PW, Palmer TD, Kuhn HG, Fisher LJ, Suhonen JO, Peterson DA, Suhr ST, Ray J. Survival and differentiation of adult neuronal progenitor cells transplanted to the adult brain. Proc Natl Acad Sci USA 1995;92(25):11879-83. 99. Gauchy C, Kemel ML, Desban M, Romo R, Glowinski J, Besson MJ. The role of dopamine released from distal and proximal dendrites of nigrostriatal dopaminergic neurons in the control of GABA transmission in the thalamic nucleus ventralis medialis in the cat. Neuroscience 1987;22:935-46. 100. Geny C, Naimi-Sadaoui S, Jeny R, Belkadi AM, Juliano SL, Peschanski M. Long-term delayed vascularization of human neural transplants to the rat brain. J Neurosci 1994;14:7553-62. 101. Gerfen CR, Herkenham M, Thibault J. The neostriatal mosaic: II. Patch- and matrix-directed mesostriatal dopaminergic and non-dopaminergic systems. J Neurosci 1987;7(12):3915-34. 102. Goetz CG, Olanow CW, Koller WC, Penn RD, Cahill D, Morantz R, Stebbins G, Tanner CM, Klawans HL, Shannon KM, Comella CL, Witt T, Cox C,
1709
1710
101 103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
Tissue transplantation for parkinson’s disease
Wasman M, Cauger L. Multicenter study of autologous adrenal medullar transplantation of the corpus striatum in patients with advanced Parkinson’s disease. N Eng J Med 1989;320:337-41. Goetz CG, Stebbins, GT, III, Klawans HL, Koller WC, Grossman RG, Bakay RA, Penn RD. United parkinson foundation neurotransplantation registry on adrenal medullary transplants: presurgical, and 1- and 2-year follow-up. Neurology 1991;41:1719-22. Goldman SA. Stem and progenitor cell-based therapy of the human central nervous system. Nat Biotechnol 2005;23(7):862-71. Greeley HT, Hamm T, Johnson R, Price CR, Weingarten R, Raffin T. The ethical use of human fetal tissue in medicine. N Eng J Med 1989;320:1093-6. Grothe C, Timmer M, Scholz T, Winkler C, Nikkhah G, Claus P, Itoh N, Arenas E. Fibroblast growth factor-20 promotes the differentiation of Nurr1-overexpressing neural stem cells into tyrosine hydroxylase-positive neurons. Neurobiology Dis 2004;17:163-70. Hagell P, Brundin P. Cell survival and clinical outcome following intrastriatal transplantation in Parkinson disease. J Neuropathol Exp Neurol 2001; 60(8):741-52. Hagell P, Piccini P, Bjo¨rklund A, Brundin P, Rehncrona S, Widner S, Crabb L, Pavese N, Oertel WH, Quinn N, Brooks DJ, Lindvall O. Dyskinesias following neural transplantation in Parkinson’s disease. Nat Neurosci 2002;5(7):627-8. Hagell P, Schrag A, Piccini P, Jahanshahi M, Brown R, Rehncrona S, Widner H, Brundin P, Rothwell JC, Odin P, Wenning GK, Morrish P, Gustavii B, Bjo¨rklund A, Brooks DJ, Marsden CD, Quinn NP, Lindvall O. Sequential bilateral transplantation in Parkinson’s disease: effects of the second graft. Brain 1999;122 (6):1121-32. Hamani C, Neimat J, Lozano AM. Deep brain stimulation for the treatment of Parkinson’s disease. J Neural Transm Suppl 2006;70:393-9. Haque NS, LeBlanc CJ, Isacson O. Differential dissection of the rat E16 ventral mesencephalon and survival and reinnervation of the 6-OHDA-lesioned striatum by a subset of aldehyde dehydrogenase-positive TH neurons. Cell Transplant 1997;6:239-48. Hassani OK, Franc¸ois C, Yelknik J, Fege´r J. Evidence for a dopaminergic innervations of the subthalamic nucleus in the rat. Brain Res 1997;749:88-94. Hauser RA, Freeman TB, Snow BJ Nauert M, Gauger L, Kordower JH, Olanow CW. Long-term evaluation of bilateral fetal nigral transplantation in Parkinson disease. Arch Neurol 1999;56(2):179-87. Hebb AO, Hebb K, Ramachandran AC, Mendez I. Glial cell line-derived neurotrophic factor-supplemented hibernation of fetal ventral mesencephalic neurons for transplantation in Parkinson’s disease: long-term storage. J Neurosurg 2003;98(5):1078-83.
115. Henderson BT, Clough CG, Hughes RC, Hitchcock ER, Kenny BG. Implantation of human fetal ventral mesencephalon to the right caudate nucleus in advanced Parkinson’s disease. Arch Neurol 1991;48(8):822-7. 116. Herrera-Marchitz M, Stro¨mberg I, Olsson D, Ungerstedt U, Olson L. Adrenal medullary implants into the dopamine denervated rat striatum. II. Acute behavior as a function of graft amount and location and its modulation by neuroleptics. Brain Res 1984; 297:53-61. 117. Hilker R, Portman AT, Voges J, Staal MJ, Burghaus L, van Laar T, Koulousakis A, Maguire RP, Pruim J, de Jong BM, Herholz K, Sturm V, Heiss W-D, Leenders KL. Disease progression continues in patients with advanced Parkinson’s disease and effective subthalamic nucleus stimulation. J Neurol Neurosurg Psychiatr 2005;76:1217-21. 118. Holthoff-Detto VA, Kesller J, Herholz K, Bo¨nner H, Pietrzyk U, Wu¨rker M, Ghaemi M, Wienhard K, Wagner R, Heiss WD. Functional effects of striatal dysfunction in Parkinson disease. Arch Neurol 1997; 54(2):145-50. 119. Hornykiewicz O. Dopamine and brain function. Pharmacol Rev 1966;18:925-64. 120. Hurtig H, Joyce J, Sladek JR, Trojanowski JQ. Postmortem analysis of adrenal-medulla-to-caudate autograft in a patient with Parkinson’s disease. J Neurol 1989; 25:607-14. 121. Inden M, Kim D-H, Qi M, Kitamura Y, Yanagisawa D, Nishimura K, Tsuchiya D, Takata K, Hayashi K, Taniguchi T, Yoshimoto K, Shimohama S, Sumi S, Inoue K. Transplantation of mouse embryonic stem cell-derived neurons into the striatum, subthalamic nucleus and substantia nigra, and behavioral recovery in hemiparkinsonian rats. Neurosci Lett 2005;387:151-6. 122. Isacson O, Deacon T. Neural transplantation studies reveal the brain’s capacity for continuous reconstruction. Trends Neurosci 1997;20:477-82. 123. Isacson O, Bjorklund LM, Schumacher JM. Toward full restoration of synaptic and terminal function of the dopaminergic system in Parkinson’s disease by stem cells. Ann Neurol 2003;53 Suppl 3:S135-48. 124. Isacson O, Costantini L, Schumacher JM, Cicchetti F, Chung S, Kim K-S. Cell implantation therapies for Parkinson’s disease using neural stem, transgenic or xenogeneic donor cells. Parkinsonism Relat Disord 2001;7:205-12. 125. Jain M, Armstrong RJE, Tyers P, Barker RA, Rosser AE. GABAergic immunoreactivity is predominant in neurons derived from expanded human neural precursor cells in vitro. Exp Neurol 2003;182:113-23. 126. Jain M, Armstrong RJ, Elneil S, Rosser AE, Barker RA. Migration and differentiation of transplanted human neural precursor cells. Neuroreport 2003b;14(9): 1257-62. 127. Johansson S, Stro¨mberg I. Guidance of dopaminergic neuritic growth by immature astrocytes in organotypic
Tissue transplantation for parkinson’s disease
128.
129.
130.
131.
132.
133.
134.
135.
136.
137.
138.
139.
cultures of rat fetal ventral mesencephalon. J Comp Neurol 2002;443:237-49. Kallos MS, Behie LA, Vescovi AL. Extended serial passaging of mammalian neural stem cells in suspension bioreactors. Biotechnol Bioeng 1999b;65:589-99. Kaminski-Schierle GS, Hansson O, Brundin P. Flunarizine improves the survival of grafted dopaminergic neurons. Neuroscience 1999;94(1):17-20. Kawasaki H, Suemori H, Mizuseki K, Watanabe K, Urano F, Ichinose H, Haruta M, Takahashi M, Yoshikawa K, Nishikawa S-I, Nakatsuji N, Sasai Y. Generation of dopaminergic neurons and pigmented epithelia from primate ES cells by stromal cell-derived inducing activity. Proc Natl Acad Sci USA 2002; 99(3):1580-5. Kim D-W, Chung S, Hwang M, Ferree A, Tsai H-C, Park J-J, Chung S, Nam TS, Kang UJ, Isacson O, Kim K-S. Stromal cell-derived inducing activity, Nurr1 and signaling molecules synergistically induce dopaminergic neurons from mouse embryonic stem cells. Stem Cells 2006;24(3):557-67. Kim J-H, Auerbach JM, Rodrı´guez-Go´mez JA, Velasco I, Gavin D, Lumelsky N, Lee S-H, Nguyen J, Sa´nchez-Pernaute R, Bankiewicz K, McKay R. Dopamine neurons derived from embryonic stem cells function in an animal model of Parkinson’s disease. Nature 2002;418(6893):50-6. Kirik D, Bjo¨rklund A. Histological analysis of fetal dopamine cell suspension grafts in two patients with Parkinson’s disease gives promising results. Brain 2005;128:1478-9. Kish SJ, Shannak K, Hornykiewicz O. Uneven pattern of dopamine loss in the striatum of patients with idiopathic Parkinson’s disease. Pathophysiologic and clinical implications. N Eng J Med 1988;318:876-80. Kopyov OV, Jacques DS, Lieberman A, Duma CM, Rogers RL. Outcome following intrastriatal fetal mesencephalic grafts for Parkinson’s patients is directly related to the volume of grafted tissue. Exp Neurol 1997;146:536-45. Kordower JH, Freeman TB, Olanow CW. Neuropathology of fetal nigral grafts in patients with Parkinson’s disease. Mov Dis 1998;13 Suppl 1:88-95. Kordower JH, Freeman TP, Snow BJ, Vingerhoets FJG, Mufson EJ, Sanberg PR, Hauser RA, Smith DA, Nauert GM, Perl DP, Olanow CW. Neuropathological evidence of graft survival and striatal reinnervation after the transplantation of fetal mesencephalic tissue in a patient with Parkinson’s disease. N Eng J Med 1995;332:1118-24. Kordower JH, Goetz CG, Freeman TB, Olanow CW. Dopaminergic transplants in patients with Parkinson’s disease: neuroanatomical correlates of clinical recovery. Exp Neurol 1997;144:41-6. Krack P, Batir A, Van Blercom N, Chabardes S, Fraix V, Ardouin C, Koudsie A, Limousin PD, Benazzouz A,
140.
141.
142.
143.
144.
145.
146.
147.
148. 149.
150.
151. 152.
153.
154.
101
LeBas JF, Benabid AL, Pollak P. Five-year follow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinson’s disease. N Eng J Med 2003;349 (20):1925-34. Kuan W-L, Barker RA. New therapeutic approaches to Parkinson’s disease including neural transplants. Neurorehabil Neural Repair 2005;19(3):155-81. Lane EL, Winkler C, Brundin P, Cenci MA. The impact of graft size on the development of dyskinesia following intrastriatal grafting of embryonic dopamine neurons in the rat. Neurobiol Dis 2006;22:334-45. Lang AE, Obeso JA. Time to move beyond nigrostriatal dopamine deficiency in Parkinson’s disease. Ann Neurol 2004;55(6):761-5. Langston JW, Widner H, Goetz CG. Core assessment program for intracerebral transplantation (CAPIT). Mov Dis 1992;7:2-13. Le Gros Clark WE. Neuronal differentiation in implanted foetal cortical tissue. J Neurol Psychiatry (London) 1940;3:263-72. Lee S-H, Lumelsky N, Studer L, Auerbach JM, McKay RD. Efficient generation of midbrain and hindbrain neurons from mouse embryonic stem cells. Nat Biotechnol 2000;18:675-9. Lee CS, Cenci MA, Schulzer M, Bjo¨rklund A. Embryonic ventral mesencephalic grafts improve levodopainduced dyskinesias in a rat model of Parkinson’s disease. Brain 2000;123:1365-79. Leigh K, Elisevich K, Rogers KA. Vascularization and microvascular permeability in solid versus cell-suspension embryonic neural grafts. J Neurosurg 1994;81:272-83. Lewy F. Zur pathologischen Anatomie der Paralysis agitans. Zeitschrift Nervenheilk 1913;50:50-5. Lin JC, Rosenthal A. Molecular mechanisms controlling the development of dopaminergic neurons. Semin Cell Dev Biol 2003;14:175-80. Lindvall O, Backlund EO, Farde L. Transplantation in Parkinson’s disease: two cases of adrenal medullary autografts to the putamen. Ann Neurol 1987;22:457-68. Lindvall O, Bjo¨rklund A. Cell therapy in Parkinson’s disease. NeuroRx 2004;1(4):382-93. Lindvall O, Hagell P. Clinical observations after neural transplantation in Parkinson’s disease. Prog Brain Res 2000;127:299-320. Lindvall O, Rehncrona S, Brundin P, Gustavii B, Astedt B, Widner H, Lindholm T, Bjo¨rklund A, Leenders KL, Rothwell JC, et al. Human fetal dopamine neurons grafted into the striatum in two patients with severe Parkinson’s disease. A detailed account of the methodology and a 6-month follow-up. Arch Neurol 1989;46:615-31. Lindvall O, Brundin P, Widner H, Rehncrona S, Gustavii B, Frackowiak R, Leenders KL, Sawle G, Rothwell JC, Marsden CD, et al. Grafts of fetal dopamine neurons survive and improve motor function in Parkinson’s disease. Science 1990;247(4942):574-7.
1711
1712
101
Tissue transplantation for parkinson’s disease
155. Lindvall O, Sawle G, Widner H, Rothwell JC, Bjo¨rklund A, Brooks D, Brundin P, Frackowiak R, Marsden CD, Odin P, et al. Evidence for long-term survival and function of dopaminergic grafts in progressive Parkinson’s disease. Ann Neurol 1994;35(2):172-80. 156. Ling ZD, Potter ED, Lipton JW, Carvey PM. Differentiation of mesencephalic progenitor cells into dopaminergic neurons by cytokines. Exp Neurol 1998;149: 411-23. 157. Ling ZD, Collier TJ, Sortwell CE, Lipton JW, Vu TQ, Robie HC, Carvey PM. Striatal trophic activity is reduced in the aged rat brain. Brain Res 2000;856:301-9. 158. Lo´pez-Lozano JJ, Bravo G, Brera A, Milla´n I, Dargallo J, Salmea´n J, Urı´a J, Insausti J, and the Clı´nica Puerta de Hierro Neural Transplantation Group. Long-term improvement in patients with severe Parkinson’s disease after implantation of fetal ventral mesencephalic tissue in a cavity of the caudate nucleus: 5-year follow up in 10 patients. J Neurosurg 1997;86:931-42. 159. Lozano AM, Dostrovsky J, Chen R, Ashby P. Deep brain stimulation for Parkinson’s disease: disrupting the disruption. Lancet Neurol 1(4):225–31. 160. Ma Y, Feigin A, Dhawan V, Fukuda M, Shi Q, Greene P, Breeze R, Fahn S, Freed C, Eidelberg D. Dyskinesia after fetal cell transplantation for parkinsonism: a PET study. Ann Neurol 2002;52(5):628-34. 161. Macklin R. The ethical problems with sham surgery in clinical research. N Eng J Med 1999;341:992-6. 162. Madrazo I, Drucker-Colı´n R, Dı´az V, Martı´nez-Mata J, Torres C, Becerril JJ. Open microsurgical autograft of adrenal medulla to the right caudate nucleus in two patients with intractable Parkinson’s disease. N Eng J Med 1987;316:831-4. 163. Madrazo I, leon V, Torres C, Aguilera MC, Varela G, Alvarez F, Fraga A, Drucker-Colı´n R, Ostrosky F, Skurovich M, et al. Transplantation of fetal substantia nigra and adrenal medulla to the caudate nucleus in two patients with Parkinson’s disease. N Eng J Med 1988;318:51. 164. Madrazo I, Franco-Bourland R, Ostrosky-Solis F. Neural transplantation (autoadrenal, fetal nigral and fetal adrenal) in Parkinson’s disease: the Mexican experience. Prog Brain Res 1990;82:593-602. 165. Mamelak AN, Eggerding FA, Oh DS, Wilson E, Davis RL, Spitzer R, Hay JA, Caton WL, III. Fatal cyst formation after fetal mesencephalic allograft transplant for Parkinson’s disease. J Neurosurg 1998;89:592-8. 166. Marsden CD, Parkes JD. ‘‘On-off’’ effects in patients with Parkinson’s disease on chronic levodopa therapy. Lancet 1976;1(7954):292-6. 167. Marsden CD, Parkes JD. Success and problems of longterm levodopa therapy in Parkinson’s disease. Lancet 1977;1(8007):292-6.
168. Martin GR. Isolation of a pluripotent cell line from early mouse embryos cultured in medium conditioned by teratocarcinoma stem cells. Proc Natl Acad Sci USA 1981;78:7634-8. 169. Martinat C, Bacci J-J, Leete T, Kim J, Vanti WB, Newman AH, Cha JH, Gether U, Wang H, Abeliovich A. Cooperative transcription activation by Nurr1 and Pitx3 induces embryonic stem cell maturation to the midbrain dopamine neuron phenotype. Proc Natl Acad Sci USA 2006;103(8):2874-9. 170. Master Z, McLeod M, Mendez I. Benefits, risks and ethical considerations in translation of stem cell research to clinical applications in Parkinson’s disease. J Med Ethics 2007;33:169-73. 171. Mehta V, Hong M, Spears J, Mendez I. Enhancement of graft survival and sensorimotor behavioral recovery in rats undergoing transplantation with dopaminergic cells exposed to glial cell line-derived neurotrophic factor. J Neurosurg 1998;88:1088-95. 172. Mehta V, Spears J, Mendez I. Neural transplantation in Parkinson’s disease. Can J Neurol Sci 1997;24:292-301. 173. Mendez I, Hong M. Reconstruction of the striato-nigrostriatal circuitry by simultaneous double dopaminergic grafts: a tracer study using fluorogold and horseradish peroxidase. Brain Res 1997;778(1):194-205. 174. Mendez I, Sadi D, Hong M. Reconstruction of the nigrostriatal pathway by simultaneous intrastriatal and intranigral dopaminergic transplants. J Neurosci 1996;16(22):7216-27. 175. Mendez I, Dagher A, Hong M, Gaudet P, Weerasinghe S, McAlister V, King D, Desrosiers J, Darvesh S, Acorn T, Robertson H. Simultaneous intrastriatal and intranigral fetal dopaminergic grafts in patients with Parkinson disease: a pilot study. J Neurosurg 2002;96:589-96. 176. Mendez I, Dagher A, Hong M, Hebb A, Gaudet P, Law A, Weerasinghe S, King D, Desrosiers J, Darvesh S, Acorn T, Robertson H. Enhancement of survivel of stored dopaminergic cells and promotion of graft survival by exposure of human fetal nigral tissue to glial cell line-derived neurotrophic factor in patients with Parkinson’s disease. Report of two cases and technical considerations. J Neurosurg 2000;92(5):863-9. 177. Mendez I, Sanchez-Pernaute R, Cooper O, Vin˜uela A, Ferrari D, Bjo¨rklund L, Dagher A, Isacson O. Cell type analysis of functional fetal dopamine cell suspension transplants in the striatum and substantia nigra of patients with Parkinson’s disease. Brain 2005;128: 1498-510. 178. Mendez I, Vin˜uela A, Astradsson A, Mukhida K, Hallett P, Robertson H, Tierney T, Holness R, Dagher A, Trojanowski JQ, Isacson O. Dopamine neurons implanted into Parkinson’s disease patients survive without pathology for 14 years. Nat Med 2008;14:507-9.
Tissue transplantation for parkinson’s disease
179. Mintz I, Hammond C, Fe´ger J. Excitatory effect of iontophoretically applied dopamine on identified neurons of the rat subthalamic nucleus. Brain Res 1986;375:172-5. 180. Molina H, Quinones AL, Ortega I. Stereotactic transplantation of foetal ventral mesencephalic cells: Cuban experiences from five patients with idiopathic Parkinson’s disease. J Neural Transplant Plast 1992;3:338-9. 181. Mukhida K, Baker KA, Sadi D, Mendez I. Enhancement of sensorimotor behavioral recovery in hemiparkinsonian rats with intrastriatal, intranigral, and intrasubthalamic nucleus dopaminergic transplants. J Neurosci 2001; 21(10):3521-30. 182. Mukhida K, Hong M, McLeod M, Miles G, Kobayashi N, Baghbaderani B, Sen A, Behie L, Brownstone R, Mendez I. Enhancement of sensorimotor behavioural recovery in parkinsonian rats with a multitarget basal ganglia dopaminergic and GABAergic transplantation strategy. Can J Neurol Sci 2007;34 Suppl 2:S26. 183. Mukhida K, Baghbaderani BA, Hong M, Lewington M, Phillips T, McLeod M, Sen A, Behie LA, Mendez I. Survival, differentiation, and migration of bioreactorexpanded human neural precursor cells in a model of Parkinson disease. Neurosurg Focus 2008;24:E8. 184. Mukhida K, Hong M, Phillips T, Quinn M, Baghbaderani BA, Massoud S, Sen A, Behie LA, Mendez I. Co-transplantation with human neural precursor cells enhances fetal dopaminergic cell survival and rotational behaviour in a rat model of Parkinson’s disease. Cell Transplanation (submitted). 185. Nakao N, Frodl EM, Duan W, Widner H, Brundin P. Lazaroids inprove the survival of grafted rat embryonic dopamine neurons. Proc Natl Acad Sci USA 1994;91:12408-12. 186. Nakao N, Ogura M, Nakai K, Itakura T. Intrastriatal mesencephalic grafts affect neuronal activity in basal ganglia nuclei and their target structures in a rat model of Parkinson’s disease. J Neurosci 1998;18:1806-17. 187. National Institutes of Health. NIH interim guidelines for the support and conduct of therapeutic human fetal tissue transplantation research. NIH Guide Grants Contracts 1993;22:2-3. 188. Nauert GM, Freeman TP. Low-pressure aspiration abortion for obtaining embryonic and early gestational fetal tissue for research purposes. Cell Transplant 1994;3:147-51. 189. Nikkhah G, Duan WM, Knappe U, Jo¨dicke A, Bjo¨rklund A. Restoration of complex sensorimotor behaviour and skilled forelimb use by a modified nigral cell suspension transplantation approach in the rat Parkinson model. Neuroscience 1993;56:33-43. 190. Nikkhah G, Olsson M, Eberhard J, Bentlage C, Cunningham MG, Bjo¨rklund A. A microtransplantation approach for cell suspension grafting in the rat Parkinson
191.
192.
193.
194.
195.
196.
197.
198. 199.
200.
201.
202.
203.
204.
101
model: a detailed account of the methodology. Neuroscience 1994;63(1):57-72. Nishino H. Intracerebral grafting of catecholamine producing cells and reconstruction of disturbed brain function. Neurosci Res 1993;16:157-72. Nissl F. Experimentellanatomische Untersuchungen u¨ber die Hirnrinde. Verhandlungen Gesellschaft Deutscher ¨ rzte 1911;83:353-5. Naturforscher und A Nunes I, Tovmasian LT, Silva RM, Burke RE, Goff SP. Pitx3 is required for development of substantia nigra dopaminergic neurons. Proc Natl Acad Sci USA 2003;100:4245-50. Nyberg P, Nordberg A, Wester P, Winblad B. Dopaminergic deficiency is more pronounced in putamen than in nucleus caudatus in Parkinson’s disease. Mol Chem Neuropathol 1983;1(3):193-202. Obeso JA, Grandas F, Herrero MT, Horowski R. The role of pulsatile versus continuous dopamine receptor stimulation for functional recovery in Parkinson’s disease. Eur J Neurosci 1994;6:889-97. Odorico JS, Kaufman DS, Thomson JA. Multilineage differentiation from human embryonic stem cell lines. Stem Cells 19:193–204. Oh MY, Abosch A, Kim SH, Lang AE, Lozano AM. Long-term hardware-related complications of deep brain stimulation. Neurosurgery 2002;50(6): 1268-74. Olanow CW. Surgical therapy for Parkinson’s disease. Eur J Neurol 2002;9 Suppl 3:31-9. Olanow CW, Goetz CG, Kordower JH, Stoessl AJ, Sossi V, Brin MF, Shannon KM, Nauert GM, Perl DP, Godbold J, Freeman TB. A double-blind controlled trial of bilateral fetal nigral transplantation in Parkinson’s disease. Ann Neurol 2003;54:403-14. Olanow CW, Koller WC, Goetz CG, Stebbins GT, Cahill DW, Gauger LL, Morantz R, Penn RD, Tanner CM, Klawans HL. Autologous transplantation of adrenal medulla in Parkinson’s disease. 18-month results. Arch Neurol 1990;47:1286-9. Olsson M, Nikkhah G, Bentlage C, Bjo¨rklund A. Forelimb akinesia in the rat Parkinson model: differential effects of dopamine agonists and nigral transplants as assessed by a new stepping test. J Neurosci 1995; 15:3863-75. Ourednik J, Ourednik V, Lynch WP, Schachner M, Snyder EY. Neural stem cells display an inherent mechanism for rescuing dysfunctional neurons. Nat Biotechnol 2002;20(11):1103-10. Parish CL, Castelo-Brano G, Rawal N, Tonnesen J, Sorensen AT, Salto C, Kokaia M, Lindvall O, Arenas E. Wnt5a-treated midbrain neural stem cells improve dopamine cell replacement therapy in parkinsonian mice. J Clin Invest 2008;118:149-60. Parish CL, Parisi S, Persico MG, Arenas E, Minchiotti G. Cripto as a target for improving embryonic
1713
1714
101 205.
206.
207.
208.
209.
210.
211.
212.
213.
214.
215.
216.
Tissue transplantation for parkinson’s disease
stem cell-based therapy in Parkinson’s disease. Stem Cells 2005;23:471-6. Park S, Lee KS, Lee YJ, Shin HA, Cho HY, Wang KC, Kim YS, Lee HT, Chung KS, Kim EY, Lim J. Generation of dopaminergic neurons in vitro from human embryonic stem cells treated with neurotrophic factors. Neurosci Lett 2004;359:99-103. Park CH, Minn YK, Lee JY, Choi DH, Chang MY, Shim JW, Ko JY, Koh HC, Kang MJ, Kang JS, Rhie DJ, Lee YS, Son H, Moon SY, Kim KS, Lee SH. In vitro and in vivo analyses of human embryonic stem cell-derived dopaminergic neurons. J Neurochem 2005; 92:1265-76. Patapoutian A, Reichardt LF. Roles of Wnt proteins in neural development and maintenance. Curr Opin Neurobiol 2000;10(3):392-9. Paul G, Ahn YH, Li J-Y, Brundin P. Transplantation in Parkinson’s disease: the future looks bright. Adv Exp Med Biol 2006;557:221-48. Perlow MJ, Freed WJ, Hoffer BJ, Seiger A, Olson L, Wyatt RJ. Brain grafts reduce motor abnormalities produced by destruction of nigrostriatal dopamine system. Science 1979;204:643-7. Perrier AL, Tabar V, Barberi T, Rubio ME, Bruses J, Topf N, Harrison NL, Studer L. Derivation of midbrain dopamine neurons from human embryonic stem cells. Proc Natl Acad Sci USA 2004;101(34):12543-8. Peterson DI, Price ML, Small CS. Autopsy findings in a patient who had an adrenal-to-brain transplant for Parkinson’s disease. Neurology 1989;39:235-8. Piccini P, Brooks DJ, Bjo¨rklund A, Gunn RN, Grasby PM, Rimoldi O, Brundin P, Hagell P, Rehncrona S, Widner H, Lindvall O. Dopamine release from nigral transplants visualized in vivo in a Parkinson’s patient. Nat Neurosci 1999;2(12):1137-40. Piccini P, Lindvall O, Bjo¨rklund A, Brundin P, Hagell P, Ceravolo R, Oertel W, Quinn N, Samuel M, Rehncrona S, Widner H, Brooks DJ. Delayed recovery of movement-related cortical function in Parkinson’s disease after striatal dopaminergic grafts. Ann Neurol 2000;48:689-95. Piccini P, Pavese N, Hagell P, Reimer J, Bjo¨rklund A, Oertel WH, Quinn NP, Brooks DJ, Lindvall O. Factors affecting the clinical outcome after neural transplantation in Parkinson’s disease. Brain 2005;128(12): 2977-86. Potter ED, Ling ZD, Carvey PM. Cytokine-induced conversion of mesencephalic-derived progenitor cells into dopamine neurons. Cell Tissue Res 1999;296: 235-46. Quinn M, Mukhida K, Sadi D, Hong M, Mendez I. Adjunctive use of the non-ionic surfactant Poloxamer 188 improves fetal dopaminergic cell survival and reinnervation in a neural transplantation strategy for Parkinson’s disease. Eur J Neurosci 2008;27(1): 43-52.
217. Rafuse VF, Soundararajan P, Leopold C, Robertson HA. Neuroprotective properties of cultured neural progenitor cells are associated with the production of sonic hedgehog. Neuroscience 2005;131(4):899-916. 218. Rawal N, Parish C, Castelo-Branco G, Arenas E. Inhibition of JNK increases survival of transplanted dopamine neurons in Parkinsonian rats. Cell Death Differ 2007;14:381-3. 219. Redmond DE, Jr, Vinuela A, Kordower JH, Isacson O. Influence of cell preparation and target location on the behavioral recovery after striatal transplantation of fetal dopaminergic neurons in a primate model of Parkinson’s disease. Neurobiol Dis 2008;29:103-16. 220. Remy P, Samson Y, Hantraye P, Fontaine A, Defer G, Mangin JF, Fe´nelon G, Ge´ny C, Ricolfi F, Frouin V, et al. Clinical correlates of [18F]fluorodopa uptake in five grafted parkinsonian patients. Ann Neurol 1995;38 (4):580-8. 221. Reynolds BA, Weiss S. Generation of neurons and astrocytes from isolated cells of the adult mammalian central nervous system. Science 1992;255:1707-10. 222. Reynolds BA, Weiss S. Clonal and population analysis demonstrate that an EGF-responsive mammalian embryonic CNS precursor is a stem cell. Dev Biol 1996;175:1-13. 223. Riaz SS, Bradford HF. Factors involved in the determination of the neurotransmitter phenotype of developing neurons of the CNS: applications in cell replacement treatment for Parkinson’s disease. Prog Neurobiol 2005;76:257-78. 224. Riaz SS, Jauniaux E, Stern GM, Bradford HF. The controlled conversion of human neural progenitor cells derived from foetal ventral mesencephalon into dopaminergic neurons in vitro. Dev Brain Res 2002;136:27-34. 225. Riaz SS, Theofilopoulos S, Jauniaux E, Stern GM, Bradford HF. The differentiation potential of human foetal neuronal progenitor cells in vitro. Dev Brain Res 2004;153:39-51. 226. Richards M, Fong CY, Chan WK, Wong PC, Bongso A. Human feeders support prolonged undifferentiated growth of human inner cell masses and embryonic stem cells. Nat Biotechnol 2002;20:933-6. 227. Robertson HA. Dopamine receptor interactions: some implications for the treatment of Parkinson’s disease. Trends Neurosci 1992;15:201-6. 228. Rodriguez-Oroz MC, Obeso JA, Lang AE, Houeto J-L, Pollak P, Rehncrona S, Kulisevsky J, Albanese A, Volkmann J, Hariz MI, Quinn NP, Speelman JD, Guridi J, Zamarbide I, Gironell A, Molet J, Pascual-Sedano B, Pidoux B, Bonnet AM, Agid Y, Xie J, Benabid A-L, Lozano AM, Saint-Cyr J, Romito L, Contarino MF, Scerrati M, Fraix V, Van Blercom N. Bilateral deep brain stimulation in Parkinson’s disease: a multicentre study with 4 years follow-up. Brain 2005;128:2240-9. 229. Rosenblad C, Martinez-Serrano A, Bjo¨rklund A. Glial cell line-derived neurotrophic factor increases survival,
Tissue transplantation for parkinson’s disease
230.
231.
232.
233.
234.
235.
236.
237.
238.
239.
240.
241.
growth and function of intrastriatal fetal nigral dopaminergic grafts. Neuroscience 1996;75(4):979-85. Rosser AE, Barker RA, Harrower T, Watts C, Farrington M, Ho AK, Burnstein RM, Menon DK, Gillard JH, Pickard J, Dunnet SB. Unilateral transplantation of human primary fetal tissue in four patients with Huntington’s disease: NEST-UK safety report ISRCTN no 36485475. J Neurol Neurosurg Psychiatr 2002; 73(6):678-85. Rossi F, Cattaneo E. Neural stem cell therapy for neurological diseases: dreams and reality. Nat Rev Neurosci 2002;3:401-9. Roussa E, Krieglstein K. GDNF promotes neuronal differentiation and dopaminergic development of mouse mesencephalic neurospheres. Neurosci Lett 2004;361: 52-5. Roy NS, Cleren C, Singh SK, Yang L, Beal MF, Goldman SA. Functional engraftment of human ES cell-derived dopaminergic neurons enriched by cocultured with telomerase-immortalized midbrain astrocytes. Nat Med 2006;12(11):1259-68. Sakurada K, Ohshima-Sakurada M, Palmer TD, Gage FH. Nurr1, an orphan nuclear receptor, is a transcriptional activator of endogenous tyrosine hydroxylase in neural progenitor cells derived from the adult brain. Development 1999;126:4017-26. Saltykow S. Versuche u¨ber Gehirnreplantation, zugleich ein Beitrag zur Kenntniss der Vorga¨nge an den zelligen Gehirnelementen. Arch Psychiatr (Berlin) 1905;40: 329-90. Sanghera MK, Manaye K, McMahon A, Sonsala PK, German DC. Dopamine transporter mRNA levels are high in midbrain neurons vulnerable to MPTP. Neuroreport 1997;8(15):3327-31. Saucedo-Cardenas O, Quintana-Hau J, Le W-D, Smidt MP, Cox JJ, De Mayo F, Burbach JPH, Conneely OM. Nurr1 is essential for the induction of the dopaminergic phenotype and the survival of ventral mesencephalic late dopaminergic precursor neurons. Proc Natl Acad Sci USA 1998;95:4013-8. Sautter J, Ho¨glinger GU, Oertel WH, Earl CD. Systemic treatment with GM1 ganglioside improves survival and function of cryopreserved embryonic midbrain grafted to the 6-hydroxydopamine-lesioned rat striatum. Exp Neurol 2000;164:121-9. Sayles M, Jain M, Barker RA. The cellular repair of the brain in Parkinson’s disease – past, present and future. Transplant Immunol 2004;12:321-42. Schierle GS, Hansson O, Leist M, Nicotera P, Widner H, Brundin P. Caspase inhibition reduces apoptosis and increases survival of nigral transplants. Nat Med 1999;5:97-100. Schulz TC, Noggle SA, Palmarini GM, Weiler DA, Lyons IG, Pensa KA, Meedeniya AC, Davidson BP, Lambert NA, Condie BG. Differentiation of human embryonic stem cells to dopaminergic neurons in
242.
243.
244.
245.
246.
247.
248.
249.
250.
251.
252.
253.
254.
101
serum-free suspension culture. Stem Cells 2004;22: 1218-38. Seaberg RM, van der Kooy D. Stem and progenitor cells: the premature desertion of rigourous definitions. Trends Neurosci 2003;26(3):125-31. Shetty N, Friedman JH, Kieburtz K, Marshall FJ, Oakes D. The placebo response in Parkinson’s disease. Clin Neuropharmacol 1999;22(4):207-12. Shihabuddin LS, Horner PJ, Ray J, Gage FH. Adult spinal cord stem cells generate neurons after transplantation in the adult dentate gyrus. J Neurosci 2000; 20:8727-35. Simon HH, Saueressig H, Wurst W, Goulding MD, O’Leary DD. Fate of midbrain dopaminergic neurons controlled by the engrailed genes. J Neurosci 2001; 21(9):3126-34. Simon HH, Saueressig H, Wurst W, Goulding MG, O’Leary DD. En-1 and En-2 control the fate of the dopaminergic neurons in the substantia nigra and ventral tegmentum. Eur J Neurosci 1998;10 Suppl 10:389. Sirinathsinghji DJS, Dunnett SB. Increased proenkephalin mRNA levels in the rat neostriatum following lesion of the ipsilateral nigrostriatal dopamine pathway with 1-methyl-4-phenylpryridium ion (MPP(+)): reversal by embryonic nigral dopamine grafts. Mol Brain Res 1991;9:263-9. Sloan DJ, Baker BJ, Puklavec M, Charlton HM. The effect of site of transplantation and histocompatibility differences on the survival of neural tissue transplanted to the CNS of defined inbred rat strains. Prog Brain Res 1990;82:141-52. Smidt MP, Asbreuk CH, Cox JJ, Chen H, Johnson RL, Burbach JP. A second independent pathway for development of mesencephalic dopaminergic neurons requires Lmx1b. Nature Neurosci 2000;3(4):337-41. Song H, Stevens CF, Gage FH. Astroglia induce neurogenesis from adult neural stem cells. Nature 2000;417:39-44. Song HJ, Stevens CF, Gage FH. Neural stem cells from adult hippocampus develop essential properties of functional CNS neurons. Nature Neurosci 2002;5 (5):438-45. Sortwell CE, Camargo MD, Pitzer MR, Gyawali S, Collier TJ. Diminished survival of mesencephalic dopamine neurons grafted into aged hosts occurs during the immediate postgrafting interval. Exp Neurol 2001;169:23-9. Sortwell CE, Pitzer MR, Collier TJ. Time course of apoptotic cell death within mesencephalic cell suspension grafts: implications for improving grafted dopamine neuron survival. Exp Neurol 2000;165(2):268-77. Spencer DD, Robbins RJ, Naftolin F, Marek KL, Vollmer T, Leranth C, Roth RH, Price LH, Gjedde A, Bunney BS, et al. Unilateral transplantation of human fetal mesencephalic tissue into the caudate nucleus of
1715
1716
101 255.
256.
257.
258.
259.
260.
261.
262.
263.
264. 265.
266.
Tissue transplantation for parkinson’s disease
patients with Parkinson’s disease. N Eng J Med 1992;327 (922):1541-8. Stenevi U, Bjo¨rklund A, Svendgaard NA. Transplantation of central and peripheral monoamine neurons to the adult rat brain: techniques and conditions for survival. Brain Res 1976;114:1-20. Storch A, Lester HA, Boehm BO, Schwarz J. Functional characterization of dopaminergic neurons derived from rodent mesencephalic progenitor cells. J Chem Neuroanat 2003;26:133-42. Storch A, Paul G, Csete M, Boehm BO, Carvey PM, Kupsch A, Schwarz J. Long-term proliferation and dopaminergic differentiation of human mesencephalic neural precursor cells. Exp Neurol 2001;170:317-25. Stro¨mberg I, Herrera-Marschitz M, Hultgren L, Ungerstedt U, Olson L. Adrenal medullary implants into the dopamine denervated rat striatum. I. Acute catecholamine levels in grafts and host caudate as determined by HPLC-electrochemistry and fluorescence histochemical image analysis. Brain Res 1984;297: 41-51. Studer L, Tabar V, McKay RDG. Transplantation of expanded mesencephalic precursors leads to recovery in parkinsonian rats. Nat Neurosci 1998;1(4):290-5. Suhonen JO, Peterson DA, Ray J, Gage FH. Differentiation of adult hippocampus-derived progenitors into olfactory neurons in vivo. Nature 1996;383(6601):624-7. Svendsen CN, Caldwell MA, Shen J, ter Borg MG, Rosser AE, Tyers P, Karmiol S, Dunnett SB. Longterm survival of human central nervous system progenitor cells transplanted into a rat model of Parkinson’s disease. Exp Neurol 1997;148:135-46. Tabar V, Panagiotakos G, Greenberg ED, Chan BK, Sadelain M, Gutin PH, Studer L. Migration and differentiation of neural precursors derived from human embryonic stem cells in the rat brain. Nat Biotechnol 2005;23(5):601-6. Takagi Y, Takahashi J, Saiki H, Morizane A, Hayashi T, Kishi Y, Fukuda H, Okamoto Y, Koyanagi M, Ideguchi M, Hayashi H, Imazato T, Kawasaki H, Suemori H, Omachi S, Iida H, Itoh N, Nakatsuji N, Sasai Y, Hashimoto N. Dopaminergic neurons generated from monkey embryonic stem cells function in a Parkinson primate model. J Clin Invest 2005;115(1): 102-109. Thompson WG. Successful brain grafting. NY Med J 1890;51:701-2. Thompson L, Barraud P, Andersson E, Kirik D, Bjo¨rklund A. Identification of dopaminergic neurons of nigral and ventral tegmental area subtypes in grafts of fetal ventral mesencephalon based on cell morphology, protein expression, and efferent projections. J Neurosci 2005;25(27):6467-77. Toda H, Takahashi J, Mizoguchi A, Koyano K, Hashimoto N. Neurons generated from adult rat hippocampal stem cells form functional glutamatergic
267.
268.
269.
270.
271.
272.
273.
274.
275.
276. 277. 278.
279.
and GABAergic synapses in vitro. Exp Neurol 2000;165(1): 66-76. Triarhou LC, Stotz EH, Low WC, Norton J, Ghetti B, Landwehrmeyer B, Palacios JM, Simon JR. Studies on the striatal dopamine uptake system of weaver mutant mice and effects of ventral mesencephalic grafts. Neurochem Res 1994;19:1349-58. Turner DA, Kearney W. Scientific and ethical concerns in neural fetal tissue transplantation. Neurosurgery 1993;33(6):1031-7. Visser-Vandewalle V, Van der Linden C, Temel Y, Celik H, Ackermans L, Spincemaille G, Caemaert J. Long-term effects of bilateral subthalamic nucleus stimulation in advanced Parkinson disease: a four year follow-up study. Parkinsonism Relat Disord 2005;11: 157-65. Volpicelli F, Perrone-Capano C, Da Pozzo P, ColucciD’Amato L, di Porzio U. Modulation of nurr1. gene expression in mesencephalic dopaminergic neuronsJ Neurochem 2004;88:1283-94. Voon V, Moro E, Saint-Cyr JA, Lozano AM, Lang AE. Psychiatric symptoms following surgery for Parkinson’s disease with an emphasis on subthalamic stimulation. Adv Neurol 2005;96:130-47. Wagner J, A˚kerud P, Castro DS, Holm PC, Canals JM, Snyder EY, Perlmann T, Arenas E. Induction of a midbrain dopaminergic phenotype in Nurr. 1-overexpressing neural stem cells by type 1 astrocytes. Nat Biotechnol 1999;17:653-9. Wang X, Li X, Wang K, Zhou H, Xue B, Li L, Wang X. Forskolin cooperating with growth factor on generation of dopaminergic neurons from human fetal mesencephalic neural progenitor cells. Neurosci Lett 2004;362: 117-21. Wenning GK, Odin P, Morrish P, Rehncrona S, Widner H, Brundin P, Rothwell JC, Brown R, Gustavii B, Hagell P, Jahanshahi M, Sawle G, Bjo¨rklund A, Brooks DJ, Marsden CD, Quinn NP, Lindvall O. Short- and long-term survival and function of unilateral intrastriatal dopaminergic grafts in Parkinson’s disease. Ann Neurol 1997;42(1):95-107. Wernig M, Benninger F, Schmandt T, Rade M, Tucker KL, Bu¨ssow H, Beck H, Bru¨stle O. Functional integration of embryonic stem cell-derived neurons in vivo. J Neurosci 2004;24(22):5258-68. Weijer C. I need a placebo like I need a hole in the head. J Law Med Ethics 2002;30(1):69-72. Widner H. NIH neural transplantation funding. Science 1994;263:737. Winkler C, Bentlage C, Nikkhah G, Samii M, Bjo¨rklund A. Intranigral transplants of GABA-rich striatal tissue induce behavioral recovery in the rat Parkinson model and promote the effects obtained by intrastriatal dopaminergic transplants. Exp Neurol 1999;155: 165-86. Winkler C, Kirik D, Bjo¨rklund A. Cell transplantation in Parkinson’s disease: how can we make it work? Trends Neurosci 2005;28(2):86-92.
Tissue transplantation for parkinson’s disease
280. Yan J, Studer L, McKay RD. Ascorbic acid increases the yield of dopaminergic neurons derived from basic fibroblast growth factor-expanded mesencephalic precursors. J Neurochem 2001;76(1):307-11. 281. Yang D, Zhang Z-J, Oldenburg M, Ayala M, Zhang S-C. Human embryonic stem cell-derived dopaminergic neurons reverse functional deficit in parkinsonian rats. Stem Cells 2008;26(1):55-63. 282. Yang M, Donaldson AE, Jiang Y, Iacovitti L. Factors influencing the differentiation of dopaminergic traits in transplanted neural stem cells. Cell Mol Neurobiol 2003;23(4/5):851-64. 283. Yasuhara T, Matsukawa N, Hara K, Yu G, Xu L, Maki M, Kim SU, Borlongan CV. Transplantation of human neural stem cells exerts neuroprotection in a rat model of Parkinson’s disease. J Neurosci 2006;26(48): 12497-511. 284. York MK, Dulay M, Macias A, Levin HS, Grossman R, Simpson R, Jankovic J. Cognitive declines following bilateral subthalamic nucleus deep brain stimulation for the treatment of Parkinson’s disease. J Neurol Neurosurg Psychiatry 2008;79:789-95. 285. Yoshizaki T, Inaji M, Kouike H, Shimazaki T, Sawamoto K, Ando K, Date I, Kobayashi K, Suhara T, Uchiyama Y, Okano H. Isolation and transplantation of dopaminergic neurons generated from mouse embryonic stem cells. Neurosci lett 2004;363:33-7. 286. Yurek DM, Fletcher-Turner A. Lesion-induced increase of BDNF is greater in the striatum of young versus old rat brain. Exp Neurol 2000;161:392-6. 287. Yurek DM, Fletcher-Turner A. Differential expression of GDNF, BDNF, and NT-3 in the aging nigrostriatal system following a neurotoxic lesion. Brain Res 2001;891:228-35. 288. Zawada WM, Meintzer MK, Rao P, Marotti J, Wang X, Esplen JE, Clarkson ED, Freed CR, Heidenreich KA. Inhibitors of p38 MAP kinase increase the survival of transplanted dopamine neurons. Brain Res 2001;891:185-96. 289. Zhao SS, Maxwell S, Jimenez-Beristain A, Vives J, Kuehner E, Zhao J, O’Brien C, de Felipe C, Semina E, Li M. Generation of embryonic stem cells and transgenic mice expressing green fluorescent protein in midbrain dopaminergic neurons. Eur J Neurosci 2004;19(5): 1133-40. 290. Zhou J, Pliego-Rivero B, Bradford HF, Stern GM. The BDNF content of postnatal and adult rat brain: the effects of 6-hydroxydopamine lesion in adult brain. Dev Brain Res 1996;97:297-303. 291. Zetterstrom T, Brundin P, Gage FH, Sharp T, Isacson O, Dunnett SB, Ungerstedt U, Bjorklund A. In vivo
292.
293.
294.
295.
296.
297.
298.
299. 300.
301.
302.
303.
101
measurement of spontaneous release and metabolism of dopamine from intrastriatal grafts using intracerebral dialysis. Brain Res 1986;362:344-49. Strecker RE, Sharp T, Brundin P, Zetterstrom T, Ungerstedt U, Bjorklund A. Autoregulation of dopamine release and metabolism by intrastriatal nigral grafts as revealed by intracerebral dialysis. Neuroscience 1987; 22:169-78. Dunnett SB, Bjorklund A, Lindvall O. Cell therapy in Parkinson?s disease ? stop or go? Nat Rev Neurosci 2001; 2:365-9. Hedreen JC. Tyrosine hydroxylase-immunoreactive elements in the human globus pallidus and subthalamic nucleus. J Comp Neurol 1999;409:400-10. Chang MY, Park CH, Lee SY, Lee SH. Properties of cortical precursor cells cultured long term are similar to those of precursors at later developmental stages. Brain Res Dev Brain Res 2004;153:89-96. Kim JH, Auerbach JM, Rodriguez-Gomez JA, Velasco I, Gavin D, Lumelsky N, Lee SH, Nguyen J, SanchezPernaute R, Bankiewicz K, McKay R. Nature 2002; 418:50-6. Yang D, Zhang ZJ, Oldenburg M, Ayala M, Zhang SC. Human embryonic stem cell-derived dopaminergic neurons reverse functional deficit in parkinsonian rats. Stem Cells 2008;26:55-63. Parish CL, Castelo-Branco G, Rawal N, Tonnesen J, Sorensen AT, Salto C, Kokaia M, Lindvall O, Arenas E. Wnt5a-treated midbrain neural stem cells improve dopamine cell replacement therapy in parkinsonian mice. J Clin Invest 2008;118:149-60. Dahlstrom A, Fuxe K. Localization of monoamines in the lower brain stem. Experientia 1964;20:398-99. Sorensen AT, Thompson L, Kirik D, Bjorklund A, Lindvall O, Kokaia M. Functional properties and synaptic integration of genetically labelled dopaminergic neurons in intrastriatal grafts. Eur J Neurosci 2005;21: 2793-99. Wernig M, Benninger F, Schmandt T, Rade M, Tucker KL, Bussow H, Beck H, Brustle O. Functional integration of embryonic stem cell-derived neurons in vivo. J Neurosci 2004;24:5258-68. Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, Swiergiel JJ, Marshall VS, Jones JM. Embryonic stem cell lines derived from human blastocysts. Science 282: 1145-47. Odorico JS, Kaufman DS, Thomson JA. Multilineage differentiation from human embryonic stem cell lines. Stem Cells 2001;19:193-204.
1717
128 Ablative Spinal Cord Procedures for Cancer Pain P. L. Gildenberg
Spinal cord surgery for the management of pain has, through the years, mainly concerned the lateral spinothalamic tract. The earliest and most popular procedure generally advocated for pain management and used primarily until very recently was surgical anterolateral cordotomy. Although this was satisfactory for the management of cancer pain, it has been far less successful for persistent non-cancer pain, which is discussed in Chap. I-1.
Surgical Anatomy The lateral spinothalamic tract conducts mainly acute pain and temperature sensation. Fibers originate in the dorsal root entry zone. They may ascend one or more segments before decussating in the anterior white commissure to ascend in the anterolateral quadrant, on the side opposite the pain, where it lies just posteromedial to the fibers of the spinocerebellar tract. There is considerable clinical evidence, however, that the decussation may variably occur many segments above the input segment [1], or that some fibers may not cross in the spinal cord, but decussate at brain stem levels [2,3]. Fibers in the lateral spinothalamic tract are somatotopically arranged, with decussating axons lying on the medial edge of the tract at each spinal cord level. Consequently, as the tract ascends, the most lateral and posterior fibers represent the lowest portion of the body, and the more medial and anterior fibers represent
#
Springer-Verlag Berlin/Heidelberg 2009
one or two levels (or more) below the segment under study. In the highest cervical levels, the most anteromedial fibers represent the upper extremity and neck. There is some evidence that fibers representing temperature sensation tend to lie somewhat posterior to those subserving pain sensation [4]. Thalamic projections of the spinothalamic system in man are more complex than the classical description would suggest [5,6]. The ascending fibers tend to contribute collaterals to the multisynaptic grey matter in the midbrain and medial thalamus, those areas projecting to the limbic affective part of the brain (> Figure 128-1). Those few spinothalamic fibers that arrive at the thalamus are distributed primarily in the ipsilateral ventral posterolateral nucleus and bilaterally in intralaminar nuclei [8], The former may represent discriminative acute pain and the latter the ‘‘suffering’’ that is associated with clinical pain [9]. Our knowledge about the anatomical and neurophysiological connections has been recently reviewed by Willis [5]. Surgical section of the lateral spinothalamic pathway, called anterolateral cordotomy, produces a loss of pain sensation on acute noxious stimulation, such as pin prick, usually starting one or several levels below the spinal interruption. In the area of analgesia, even though a single pin prick may not be appreciated as being sharp, sometimes repeated pin stick stimulation can be appreciated [10]. Loss of deep pain sensation is somewhat less consistent [11], and is more nearly correlated with successful pain relief after cordotomy [12].
2160
128
Ablative spinal cord procedures for cancer pain
. Figure 128-1 Pain pathways project to the limbic brain, as well as somatosensory areas. This shows also a multisynaptic central grey pathway (from [7])
Somatic pain of the body wall or extremities may be relieved by such a maneuver, but usually not pain of visceral origin. Because persistent pain may continue or resume after an apparently successful cordotomy, the existence of multisynaptic pain pathways was assumed, in addition to the lateral spinothalamic pathways, even if such polysynaptic tracts can be demonstrated only with difficulty by usual anatomical means. They have been classified as the paleospinothalamic and archispinothalamic systems [7,13]. In addition, a postsynaptic dorsal column (PSDC) pathway carrying visceral pain
sensation is located in the vicinity of the central canal [14]. Because there are multiple pathways carrying pain information to the brain, it is not uncommon for pain and/or suffering to persist or to recur after an initially successful cordotomy. The anatomy of the spinothalamic tract was first described by Edinger [15] in 1889, but its function was not known. Spiller [16] reported in 1905 on a patient in whom pain and temperature sensation in the lower body was lost secondary to bilateral tuberculomas involving the anterolateral quadrants of the spinal cord, an observation that led soon thereafter to the concept of surgical
Ablative spinal cord procedures for cancer pain
anterolateral cordotomy. Schu¨ller [17] demonstrated in monkeys that one could section the anterolateral tracts without causing paralysis, and he coined the term ‘‘chordotomie’’ for that procedure. The original theory that pain was specifically transmitted by the lateral spinothalamic pathways led Spiller [16], early in the twentieth century, to study loss of pain sensation after section of the anterolateral quadrants of the spinal cord in animals. He persuaded Martin [18] to make similar lesions in patients suffering from pain, and they reported the first successful planned anterolateral cordotomy in 1912. The operation was refined further by Foerster [19,20] in Europe and Frazier [21] in America, and eventually became the most commonly performed neurosurgical procedure for the treatment of pain. Surgical cordotomy for pain in the lower body, usually performed at the thoracic area, became a standard neurosurgical procedure [11,22,23]. It required a laminectomy, however, in a group of patients who were often too debilitated to tolerate such a procedure. Usually it was done with the patient under general anesthesia, so it was not possible to monitor the location or extent of the lesion during surgery, except in the rare case done under local or regional anesthesia [11]. This led Mullan [24] to develop a percutaneous technique whereby a radioactive strontium needle was passed into the anterolateral spinal canal at the C2 level, where it was brought into contact with the spinal cord for a measured time to gradually produce a lesion involving the spinothalamic tract. The procedure was simplified and modified by Rosomoff [25], who used a similar approach, but he produced the lesion with a radiofrequency electrode, providing an immediate lesion and better control over the location and extent of the lesion. A danger with interrupting the anterolateral quadrant at the C2 level, however, is that fibers concerned with respiration run adjacent to the lateral spinothalamic tract in the anterolateral spinal cord down to the C3 level [26]. Consequently, when a cordotomy
128
lesion is made above the C4 level, it may include fibers concerned with respiration, which may produce a fatal sleep-induced apnea, Ondine’s curse, if bilateral lesions are made. In addition patients with chest wall cancer pain after a pneumonectomy may succumb from sleep induced apnea after a unilateral lesion [27]. The respiratory impairment appears to be an inability to integrate sensory information into the respiratory drive, rather than paralysis of the diaphragm, so the patient is able to breath when awake (but may express a vague feeling of apprehension), but respiratory drive fails when the patient falls asleep [28,29]. This impairment may be identified by a failure of the awake patient to respond to a hypercapnia challenge on breathing CO2 [27]. Percutaneous cervical cordotomy was modified by Lin and Gildenberg [28] in 1966 to approach the spinal cord with a radiofrequency needle electrode inserted diagonally through an intervertebral disk at lower cervical levels, that is, below C4, which made the procedure more difficult but safer, with complete elimination of the risk of respiratory depression (> Figure 128-2).
. Figure 128-2 Percutaneous lower cervical cordotomy. The needle electrode is inserted diagonally through an intervertebral disk to the lateral spinal cord (from [28])
2161
2162
128
Ablative spinal cord procedures for cancer pain
Until the mid-1970s, cordotomy was done for persistent pain regardless of etiology. By that time, however, it gradually became recognized that cordotomy was much more successful in management of cancer pain than chronic pain of benign origin, so its use in the latter has been discouraged more and more [30–32]. Various physiologic and anatomic observations were made in studies of patients who had a cordotomy. As early as 1926, Greenfield [33] suggested to Armour that the spinothalamic fibers could be interrupted bilaterally in selected segments by cutting the spinal cord through the midline, aiming for the decussation of the spinothalamic axons. Armour [34] successfully relieved the pain of one patient, but the patient died of pneumonia postoperatively and the project was not pursued. However, a decade later, Putnam [35] reintroduced the procedure, named commissural myelotomy, and it became accepted. It was soon recognized, however, that the amount of analgesia was quite small, even with good relief of pain, and the mechanism of that observation has become apparent only in the past few years. Until the mid-1960s, interruption of the spinothalamic tract with either cordotomy or commissural myelotomy became the standard neurosurgical management of pain, regardless of etiology. It gradually became apparent, however, (1) that many patients had no or only temporary pain relief, (2) the different types of pain demanded different strategies, that is, acute pain, cancer pain, and chronic pain of non-cancer origin, (3) that it was necessary to evaluate psychological attributes as well as the patients description of somatic pain, and (4) there must be some anatomical explanation for all the conflicting observations. It was in 1965 that Melzack and Wall [36] presented the gate control theory of pain (> Figure 128-3). This concept reconciled seemingly conflicting results from both experimental studies and clinical observations. They proposed that there was a ‘‘gate’’ at the dorsal root entry zone
. Figure 128-3 The Melzack-Wall gate theory (from [36])
in the substantia gelatinosa that opened or closed to allow or to block transmission of pain sensation, depending on whether noxious or non-noxious stimulation predominated (> Figure 128-1). The concept was elaborated further to include additional concepts of the mechanism of the emotional (limbic) contribution to pain perception [37]. As the understanding of the pain system became more sophisticated, the use of cordotomy for chronic non-cancer pain declined, although it was still used appropriately for cancer pain. The further development of spinal cord [38–40] or deep brain stimulation [41–43] for management of chronic pain intruded even more on the use of cordotomy, and the decline was accelerated further by the introduction of implantable spinal morphine pumps [44–46]. Meanwhile, there was further elaboration of an additional spinal cord pain pathway that had been hinted at with observations after commissural myelotomy. In 1970, Hitchcock [47] presented a procedure he called central myelotomy. His stereotactic apparatus allowed insertion of an electrode through the foramen magnum or into the upper cervical spinal cord. He used his stereotactic apparatus to introduce a needleelectrode from dorsally to perform high cervical cordotomy, in order to produce the same percutaneous cervical cordotomy lesion as produced by Mullan [24] and Rosomoff [25]. In one case,
Ablative spinal cord procedures for cancer pain
the electrode was inadvertently inserted into the center of the spinal cord at the cervicomedullary junction, when the patient suddenly moved as the electrode contacted the pia. The patient had immediate relief of his pain, and the relief lasted when only a small lesion was made at that site. This target at the center of the spinal cord at the cervico-meddulary junction was adopted for central myelotomy, and other patients had similar successful relief of pain by the production of this lesion [48,49]. Although this procedure could not be done readily without a stereotactic apparatus which allowed this unusual approach, other neurosurgeons, including myself, reported similar success by producing a mechanical lesion at the same site after the cervicomeddulary junction had been exposed surgically. One must consider the results of Hitchcock’s stereotactic myelotomy in light of the old procedure of commissural myelotomy, both of which resulted in pain relief without detectable analgesia or reports of pain relief far exceeding the amount of analgesia, especially if visceral pain predominated [22,50,51]. The observation suggested that in some patients the spinothalamic fibers may ascend more than only one or two segments before decussating, an observation consistent with examination of patients after percutaneous cervical cordotomy [3]. It also suggested that there is some tract other than the decussating spinothalamic fibers being interrupted when a midline myelotomy is performed. The sum of these observations led Gildenberg [52] to conclude that there is a pathway, previously undescribed, ascending at the center of the spinal cord carrying predominately visceral pain perception, and that interruption of that pathway could relieve such cancer pain. It was assumed that the pathway was multisynaptic and probably incorporated into the central grey. There seemed to be little advantage in interrupting the pathway at the cervicomedullary area, as Hitchcock had done, for patients who had pelvic or perineal
128
pain, which occurred commonly in patients with cancer of pelvic organs, such as uterus or rectum. Consequently Gildenberg interrupted the central cord at the spinal cord lumbothoracic level with a single level T9 laminectomy, often providing excellent relief of such cancer pain. He recruited Hirshberg to collaborate in accumulating sufficient independent clinical experience, and they published a series of patients thus treated with ‘‘limited myelotomy’’ in 1981 [52]. When a manuscript by Gildenberg and Hirshberg, discussing these observations and speculating that there might be a multisynaptic pathway at the anterior border of the dorsal columns, was submitted to two US neurosurgical journals, it was rejected because ‘‘there is no such pathway,’’ which led to publication in a British journal [52]. Hirshberg provided the spinal cord of one of his patients who had died of his cancer to Willis [53], who identified a new long-tract pathway that was not polysynaptic, but lay at the anterior part of the medial borders of the posterior columns. He further demonstrated in rats that the pathway was associated with visceral pain [53,54]. This led to Willis consulting with his neurosurgical colleagues to perform limited myelotomy, with the target that Willis had identified in the dorsal columns. Nauta [55] interrupted the pathways by making a number of small radiofrequency lesions with multiple small electrode penetrations through the dorsal surface and called the procedure ‘‘punctate myelotomy,’’ although it is the same procedure as limited myelotomy with minor modifications.
Techniques Surgical cordotomy: The technique of surgical cordotomy has been described numerous times with many minor modifications [11,21,23,56], but the basic approach has not changed much since Spiller and Martin’s [18] original report in 1912.
2163
2164
128
Ablative spinal cord procedures for cancer pain
One representative technique is presented, although others are equally as safe and effective. A one or two segment laminectomy is done at a level at least several segments above the pain input, care being taken to recognize spinal cord rather than vertebral levels. Since the T9 spinal cord level has the most vulnerable vascular supply, it is best to avoid that area. The dura is opened and the spinal cord is visualized under magnified vision. The dentate ligament above and below the planned incision are freed. If there is any traction on the dorsal roots when the cord is rotated, or if they prevent the rotation of the spinal cord, the dorsal roots just above and below the incision are also sectioned. The dentate ligament is grasped by a hemostat or other instrument, and the spinal cord is rotated 45 to expose the anterolateral surface, taking care to rotate the entire cord and not to twist it, so the internal structures retain as close to normal orientation as possible. A sharp blade is inserted at the origin of the dentate ligament, and an incision is made to a point just medial to the emergence of the fibers of the anterior root, with care to avoid injury to the anterior spinal artery. The incision should be deep enough into the cord to transect a pie-shaped segment of about 90 , rather than interrupting only the pia and superficial fibers. If the dentate ligament originates far dorsal to the lateral midpoint, the later should be used to identify the posterior margin of the incision to avoid interruption of the corticospinal pathway. It has been reported that approximately half of patients with cancer pain in the appropriate distribution will have total immediate relief and one-fourth will have partial relief, but the remaining one-fourth will have no benefit [26]. By the end of six months, however, pain will return in approximately one-half of the originally successful patients [57]. After unilateral cordotomy for pelvic or abdominal cancer pain, an additional 10% or more may have the rapid increase of pain on the side opposite the analgesic side [3,58,59].
The mortality rates vary greatly from one series to the next, particularly in relationship to whether non-cancer pain is included in the study. White and Sweet [60] report 8% mortality rate in cancer patients after high thoracic cordotomy, but none when the procedure was done for non-cancer pain. They also report a 13% risk of lower extremity weakness after unilateral cordotomy [11]. Commissural myelotomy: The technique of commissural myelotomy begins like that of surgical cordotomy. The usual patients have bilateral pelvic or lower extremity pain, or pelvic visceral pain. The spinal cord is exposed throughout all segments for which pain denervation is intended and perhaps two additional cephalad segments. The spinal level does not correspond to the spinal cord segment, and that must be taken into account when the multiple level laminectomy is planned. Today it is advisable to use an operating microscope to visualize the spinal cord during the midline bisection. The arachnoid must be carefully dissected to expose the cord and identify the dorsal vasculature. The midline can be identified by the vessels diving between the posterior columns in the posterior median sulcus. The pia may be firm and require sharp dissection to begin the dissection. The posterior median septum is a single fibrous layer that lies between the posterior columns, and one can dissect along either side of it to the central canal area. Because the septum is a valuable landmark and because the neural tissue to be separated is friable, I prefer to use a ball dissector. Earlier techniques recommended sectioning only the posterior commissure, but it is doubtful that structure alone could be adequately identified. With good control over the depth of incision, complete myelotomy may provide a more consistent result. The anterior median septum can first be palpated and then visualized as the anterior extent of the dissection, and the surgeon should not continue beyond that for fear of damaging the anterior spinal artery.
Ablative spinal cord procedures for cancer pain
Results of midline commissural myelotomy vary greatly from one series to another [19], possibly because of differences in patient selection or differences in technique, with some surgeons sectioning more cranially than others or others perhaps not sectioning deeply enough. There are, unfortunately many reported undesirable side effects, such as hyperesthesia, diminished proprioception, paresthesia or dysesthesia of the legs, radicular pain, paresis, or incoordination of gait [61], along with a mortality rate of up to 8% [62]. Gybels and Sweet [61] review the literature and note that there may be a variable amount of analgesia, and sometimes good pain relief is obtained with little analgesia [1,50,51,63,64]. Although they declined to explain these results on the basis of understanding of pathways at the time of their writing, our observations with limited myelotomy [52] and the new findings of Willis [65] would account for those discrepancies very well. Limited myelotomy: The theoretical consideration of a pain pathway ascending in the central spinal cord of Hitchcock’s [66] central myelotomy served as the basis for our limited myelotomy [52], where a lesion is made in the center of the spinal cord at only a single segment above the input from the painful areas. For pelvic pain, particularly visceral pain of rectal or uterine cancer, the most common indication for limited myelotomy, the lesion is made at the thoraco-lumbar junction of the spinal cord. An exposure by making a T9 laminectomy exposes the spinal cord at approximately the T12 level, so all the lumbar and sacral dermatomes are included. The procedure is just the same as for commissural myelotomy, but involves only a single level. It has been our custom to move a small ball dissector around the depth of the midline incision to assure damage to the adjacent pathways. The success rate of limited myelotomy is comparable to that of commissural myelotomy, but so far no untoward side effects or mortality
128
have been seen [52]. When there is a component of somatic pain, it may not necessarily be relieved by limited myelotomy, but may require a separate procedure. Lesions in the dorsal root entry zone are also used to treat pain of various types, but that is discussed in Chap. I-5 and I-6. Percutaneous cervical cordotomy, C2: The technique of percutaneous cervical cordotomy depends on a cooperative patient and excellent fluoroscopy, and has not been modified significantly since the first reports of Rosomoff [25,27]. It requires a much different orientation than usual neurosurgery, along with attention to detail and patience in working with awake patients in great pain. The patient lies supine with a C-arm X-ray apparatus adjusted to visualize either an AP or lateral view across the base of the skull. On the lateral view, it is not difficult to identify the crotch between the posterior arch of C1 and the lamina of C2. The needle is introduced horizontally to a point just anterior to the middle of the canal, care being taken to avoid spearing the spinal cord if the needle advances suddenly on penetrating the dura. The use of a head support and needle microdrive, such as that designed by Rosomoff, are very helpful [25]. A drop of hyperbaric or other contrast material allows visualization of the dentate ligament, and a small amount of air may outline the anterior surface of the cord. The obturator is removed from the needle and the electrode introduced. The electrode tip is advanced with care to the pia half way between the anterior cord and the dentate ligament, as seen on the lateral X-ray. The patient may complain of pain as the pia is touched, and it may be necessary to pierce the pia with a sharp thrust. Impedance may be measured to assure penetration into the spinal cord [67]. Stimulation at approximately 50 Hz can be used to verify proper position within the spinothalamic tract, as evidenced by the projection of sensation to the contralateral body, and the tip of the
2165
2166
128
Ablative spinal cord procedures for cancer pain
needle-electrode adjusted accordingly. A very low voltage may be required, and segmental sensation must be differentiated from spinothalamic tract response. The threshold for nerve root sensation is lower and the sensation is much more painful [23,68]. A lesion is made, for instance, at 70 to 90 C for up to 30 seconds in increments of 10 seconds [69]. The lesion may be repeated, if necessary. When adequate analgesia is demonstrated, the electrode is withdrawn. The results after percutaneous cervical cordotomy are similar to those after surgical cordotomy, but the complication rate may be significantly less [60]. However, the complication of sleep induced apnea is significant. Although Kreiger and Rosomoff [70,71] recommended the use of a CO2 challenge prior to production of lesions on the second side, many neurosurgeons fear that complication sufficiently to avoid bilateral C2 cordotomy completely. In our early percutaneous cervical cordotomy experience, a patient who had had a pneumonectomy with recurrent chest wall pain had a C2 lesion on the opposite side with good relief of pain. She died the second night post-operatively; only in retrospect did we realize she had died of sleep induced apnea. Three other patients developed that complication after a lesion was produced on the second size. All had had excellent analgesia on the first side and no respiratory problems for at least several days before the second lesion was made. It was after the reassuring overnight observation of the first patient that two more patients had their procedure, both on the following day. All three patients developed sleep induced apnea, so all three were on ventilators at the same time. It was then that we recognized that the single contralateral lesion in the first patient denervated the respiratory sensory innervation from his only remaining lung, and this patient who was already critically ill was found dead in bed during the night, without our realizing the cause [28]. Within a few weeks, we had moved the lesion down to a lower cervical level.
Lower cervical percutaneous cordotomy: The procedure of lower cervical cordotomy evolved over several years to the following protocol [59,72]. The patient is in the supine position with the neck slightly extended, similar to that for a cervical discogram. The spinal cord ordinarily rests against the posterior canal in this position, and that may be verified by the instillation of contrast material and/or air as for the C2 cordotomy. The target point depends on the segment representing the pain. The cord is 10 20 mm at that level, with the dentate ligament at the midpoint on the lateral projection. For pain involving the sacrum or lower extremity, the target is 8 mm lateral and 5 mm anterior to the posterior wall of the canal. For chest pain, the lesion is 3–4 mm lateral and 8 mm anterior to the posterior wall. AP and lateral X-ray tubes are adjusted to provide identical magnification without moving either the patient or the X-ray apparatus or cassettes. The needle is inserted under local anesthesia between the carotid sheath and the tracheo-esophageal complex. By pressing on the skin at that site, it is possible to bring the subcutaneous tissues almost in contact with the anterior surface of the vertebral body near the midline. The desired interspace is identified by relating the level of the shoulders with the scout lateral X-ray. Just the tip of the needle is inserted into the disk, and AP and lateral films are taken. The trajectory is either calculated mathematically or determined graphically. If one pictures the cross section of the neck with the needle just entering the anterior disk, the line between the position of the needle tip and the target represents the hypotenuse of a right triangle. The relative distance representing the base of the triangle can be measured on the AP film as the distance from the tip of the needle to the target point. The relative distance representing the height of the triangle can be measured on the lateral film as the distance from the tip of the needle to the target point. An equivalent triangle can be drawn
Ablative spinal cord procedures for cancer pain
at the corner of a piece of paper by measuring relative base and height distances along the edge of the paper and connecting those point by drawing the hypotenuse. That paper can then be held by the patient’s chin and the needle aligned by eye with the hypotenuse, which will place it on the proper trajectory to hit the target. The needle is advanced along that trajectory until it is locked in its path by the posterior annulus fibrosis and posterior longitudinal ligament. The X-rays films are repeated using the same technique, the measurements are repeated, an equivalent triangle is drawn (the angles of the triangle should be identical, even if the length of the sides are shorter), and it is verified that the needle lies along the path of the hypotenuse. When it is in satisfactory trajectory, the needle is advanced into the spinal canal, and the obturator is replaced by the electrode. The electrode assembly is advanced until both the patient and surgeon recognize that the pia is being touched, and the pia is punctured carefully with a sharp thrust. Films are again taken to assure that the tip of the electrode is at the proper target point. As with C2 percutaneous cervical cordotomy, penetration of the spinal cord can be verified by impedance measurement. Stimulation, however, is less reliable at lower cervical levels, since root pain at a low threshold may interfere with stimulation of the ascending tracts. Results after lower percutaneous cervical cordotomy are equivalent to those after open surgical cordotomy or C2 percutaneous cervical cordotomy. The complication of weakness, however, is higher but of a different nature. Approximately 15% of patients have weakness of the ipsilateral hand due to trauma to the emerging motor nerve root. Two-thirds of those cases return to normal within three weeks, leaving a 5% risk of permanent hand weakness. Most patients are so gratified about their relief from pain that they accept that problem when it occurs. A modification of percutaneous cervical cordotomy has been developed by Kanpolat [73,74].
128
Ordinarily percutaneous cervical cordotomy depends on calculating or adjusting the target as seen on orthostatic AP and lateral fluoroscopy or X-ray films [25,28,72]. The surgeon does not see the actual trajectory of the needle-electrode on any single view. However, a transverse CT slice can be taken along the path of the needle so that the needle, its trajectory, the spine and the spinal cord can all be seen on one projection slice or film, which facilitates the targeting immensely. The procedure is otherwise the same as a C2 or lower cervical cordotomy.
Conclusions Interruption of pain pathways within the spinal cord has proven to be of great help to a number of patients. However, it is performed less and less, for several reasons. During the last two decades, we have recognized more and more that simple interruption of pain pathways, although of great use in cancer patients, is inappropriate for treatment of most patients with chronic non-cancer pain. Such pain is complicated by many physical and emotional factors [30,31,75], which require a more comprehensive pain management program [76], as discussed in Chap. I-1. Selected patients with cancer pain, however, are still candidates for cordotomy or myelotomy, since their prognosis for survival may be too short for recurrence of pain and since the somatic aspects and localization of their pain are often clearly identifiable [77]. During this same time, alternative methods of treatment have emerged. Intraspinal administration of morphine with either an external (for pre-terminal care) or an implanted pump (for patients with a prognosis of greater than three months) may provide excellent relief of cancer pain with fewer risks. If the cancer progresses to involve the other side or other sites, the pump may be reprogrammed to accommodate the increased pathology. In fact, I have performed
2167
2168
128
Ablative spinal cord procedures for cancer pain
only a few cordotomies (percutaneous or surgical) since I began to implant morphine pumps. I would still advocate limited myelotomy for pelvic visceral pain. It has little enough risk and, if successful, may be more cost effective and requires less follow-up care than an implanted spinal pump.
References 1. Sweet WH, Poletti CE. Operations in the brain stem and spinal canal with an appendix on open cordotomy. In: Wall PD, Melzack R, editors. Textbook of pain. Edinburgh: Churchill-Livingstone; 1984. p. 615-31. 2. Truex RC, Taylor MJ, Smythe MQ, Gildenberg PL. The lateral cervical nucleus of cat, dog and man. J Comp Neurol 1970;139:93-104. 3. Gildenberg PL. Physiologic observations during percutaneous cervical cordotomy. In: Somjen GG, editor. Neurophysiology studied in man. Excerpta Medica; Amsterdam; 1972. p. 231–6. 4. Carpenter MB. Human neuroanatomy. 7th ed. Baltimore: Williams and Wilkins; 1976. 5. Willis WD. The pain system. The neural basis of nociceptive transmission in the mammalian nervous system. Basel: Karger; 1985. 6. Willis WD Jr. Central nervous system mechanisms for pain modulation. Appl Neurophysiol 1985;48:153-65. 7. Struppler A. Verhandlungen der Deutschen Gesellschaft fur Innere Medizin 1980;86:1560-2. 8. Nathan PW, Smith MC. Some tracts of the anterior and lateral columns of the spinal cord. In: Knighton RS, Dumke PR, editors. Pain. Boston: Little, Brown; 1966. p. 44-57. 9. Lahuerta J, Bowsher D, Campbell J, Lipton S. Clinical and instrumental evaluation of sensory function before and after percutaneous anterolateral cordotomy at cervical level in man. Pain 1990;42:23-30. 10. Cook AW, Nathan PW, Smith MC. Sensory consequences of commissural myelotomy. A challenge to traditional anatomical concepts. Brain 1984;107:547-68. 11. White JC, Sweet WH. Pain and the neurosurgeon. A forty year experience. Springfield: Charles C Thomas; 1969. 12. Nathan PW, Smith MC. Clinico-anatomical correlation in anterolateral cordotomy. Adv Pain Res Ther 1979;3: 921-6. 13. Bishop GH. The relation between nerve fiber size and sensory modality: phylogenetic implication of the afferent innervation of cortex. J Nerv Ment Dis 1959; 128:89-114.
14. Palecek J, Paleckova V, Willis WD. The roles of pathways in the spinal cord lateral and dorsal funiculi in signaling nociceptive somatic and visceral stimuli in rats. Pain 2002;96:297-307. 15. Edinger L. Vergleichend-entwicklingsgeschichtliche und anatomische Studien im Bereiche des Central-nervensys¨ ber die Forsetzung der hinteren Ru¨ckenmarkstems. II. U wurzeln zum Gehirn. Anat Anz 1889;4:121-8. 16. Spiller WG. The location within the spinal cord of the fibers for temperature and pain sensations. J Nerv Ment Dis 1905;32:318-20. ¨ ber operative Durchtrennung der Ru¨chen17. Schu¨ller A. U marksstra¨nge (Chordotomie). Wien med Wchenschr 1910;60:2292-5. 18. Spiller WG, Martin E. The treatment of persistent pain of organic origin in the lower part of the body by division of the anterolateral column of the spinal cord. J Am Med Assoc 1912;58:1489-90. 19. Foerster O. Vorderseitenstrangdurchschneidung im Ru¨ckenmark zur Beseitigung von Schmerzen. Berlin klin Wschr 1913;50:1499. 20. Foerster O, Gagel O. Die Vorderseitenstrangdurchschneidung beim Menschen. Eine Klinische, pathophysiologisch, anatomische Studie. Ztschr f d ges Neurol u Psychiat 1932;138:1-92. 21. Frazier CH. Section of the anterolateral columns of the spinal cord for the relief of pain. Arch Neurol Psychiat 1920;4:137-47. 22. Gybels JM, Sweet WH. Neurosurgical treatment of persistent pain. Pain and headache, vol 11. Basel: Karger; 1989. 23. White JC, Sweet WH. Pain, its mechanism and neurosurgical control. Springfield: Charles C Thomas; 1955. 24. Mullan S, Harper PV, Hekmatpanah J, Torres H, Dobben G. Percutaneous interruption of spinal pain tracts by means of a strontium-90 needle. J Neurosurg 1963;20: 931-9. 25. Rosomoff HL, Carrol F, Brown J, Sheptak P. Percutaneous radiofrequency cervical cordotomy: technique. J Neurosurg 1965;23:639-44. 26. Nathan PW. The descending respiratory pathway in man. J Neurol Neurosurg Psychiatry 1963;26:487-99. 27. Rosomoff HL, Krieger AJ, Kuperman AS. Effects of percutaneous cervical cordotomy on pulmonary function. J Neurosurg 1969;31:620-7. 28. Lin PM, Gildenberg PL, Polakoff PP. An anterior approach to percutaneous lower cervical cordotomy. J Neurosurg 1966;25:553-60. 29. Tenicela R, Rosomoff HL, Feist J, Safar P. Pulmonary function following percutaneous cervical cordotomy. Anesthesiology 1968;29:7-16. 30. Gildenberg PL, DeVaul RA. The chronic pain patient. Evaluation and management. 7th ed. Basel: Karger; 1985. 31. Gildenberg PL. General and psychological assessment of the pain patient. In: Tindall GT, Cooper PR, Barrow DL,
Ablative spinal cord procedures for cancer pain
32.
33. 34. 35.
36. 37.
38.
39.
40.
41.
42. 43.
44.
45.
46.
47. 48.
49. 50.
editors. The practice of neurosurgery. Baltimore: Williams and Wilkins; 1996. p. 2987-96. Bonica JJ. Organization and function of a pain clinic. In: Bonica JJ, editor. International symposium on pain. New York: Raven Press; 1974. Walker AE. History of neurological surgery. Baltimore: Williams and Wilkins; 1951. Armour D. Surgery of the spinal cord and its membranes. Lancet 1927;i:691-7. Putnam TJ. Myelotomy of the commissure. A new method of treatment for pain of the upper extremities. Arch Neurol Psychiatry 1934;32:1189-92. Melzack R, Wall PD. Pain mechanisms: a new theory. Science 1965;150:971-9. Melzack R, Casey KL. Sensory, motivational, and central control determinants of pain. In: Kenshalo DR, editor. The skin senses. Springfield: Charles C Thomas; 1968. p. 423-39. Long DM, Erickson DE. Stimulation of the posterior columns of the spinal cord for relief of intractable pain. Surg Neurol 1975;41:134-41. Gildenberg PL. The use of pacemakers (electrical stimulation) in functional neurological disorders. In: Rasmussen T, Marino R, editors. Functional neurosurgery. New York: Raven Press; 1979. p. 59-74. Augustinsson LE, Sullivan L, Sullivan M. Physical, psychologic, and social function in chronic pain patients after epidural spinal electrical stimulation. Spine 1986;11: 111-19. Richardson DE, Akil H. Pain reduction by electrical brain stimulation in man. 1. Acute administration in periaqueductal and periventricular sites. J Neurosurg 1977;47: 178-83. Foley KM. Opioids. Neurologic clinics 1993;11:503-22. Gybels J, Kupers R. Central and peripheral electrical stimulation of the nervous system in the treatment of chronic pain. Acta Neurochir Suppl (Wien) 1987;38:64-75. Brazenor GA. Long term intrathecal administration of morphine: a comparison of bolus injection via reservoir with continuous infusion by implanted pump. Neurosurgery 1987;21:484-91. Payne R. Role of epidural and intrathecal narcotics and peptides in the management of cancer pain. Med Clin N Amer 1987;71:313-27. Onofrio BM, Yaksh TL. Long-term pain relief produced by intrathecal morphine infusion in 53 patients. J Neurosurg 1990;72:200-9. Hitchcock E. Stereotactic cervical myelotomy. J Neurol Neurosurg Psychiatry 1970;33:224-30. Schvarcz JR. Spinal cord stereotactic techniques re trigeminal nucleotomy and extralemniscal myelotomy. Appl Neurophysiol 1978;41:99-112. Hitchcock E. Stereotactic myelotomy. Proc R Soc Med 1974;67:771-2. Cook AW, Kawakami Y. Commissural myelotomy. J Neurosurg 1977;47:1-6.
128
51. King RG. Anterior commissurotomy for intractable pain. J Neurosurg 1977;47:7. 52. Gildenberg PL, Hirshberg RM. Limited myelotomy for the treatment of intractable cancer pain. J Neurol Neurosurg Psychiatry 1984;47:94-6. 53. Al Chaer ED, Lawand NB, Westlund KN, Willis WD. Pelvic visceral input into the nucleus gracilis is largely mediated by the postsynaptic dorsal column pathway. J Neurophysiol 1996;76:2675-90. 54. Al Chaer ED, Feng Y, Willis WD. A role for the dorsal column in nociceptive visceral input into the thalamus of primates. J Neurophysiol 1998;79:3143-50. 55. Nauta HJ, Hewitt E, Westlund KN, Willis WD Jr. Surgical interruption of a midline dorsal column visceral pain pathway. Case report and review of the literature. J Neurosurg 1997;86:538-42. 56. Sweet WH, Poletti CE, Gybels JM. Operations in the brainstem and spinal canal, with an appendix on the relationship of open to percutaneous cordotomy. In: Wall PD, Melzack R, editors. Textbook of pain. Edinburgh: Churchill Livingstone; 1994. p. 1113-35. 57. Cowie RA, Hitchcock ER. The late results of anterolateral cordotomy for pain relief. Acta Neurochir (Wien) 1982;64:39-50. 58. Mansuy L, Sindou M, Fischer G, Brunon J. Spinothalamic cordotomy in cancerous pain. Results of a series of 124 patients operated on by the direct posterior approach. Neurochirurgie 1976;22:437-44. 59. Gildenberg PL. Percutaneous cervical cordotomy. Clin Neurosurg 1974;21:246-56. 60. White JC, Sweet WH. Anterolateral cordotomy: open versus closed comparison of end results. Adv Pain Res 1979;3:911-19. 61. Gybels JM, Sweet WH. Neurosurgical treatment of persistent pain. Physiological and pathological mechanisms of human pain. Pain Headache 1989;11:1-402. 62. Grunert V, Kraus H, Sunder-Plassmann M, Gestring GF. Die kommissurale myelotomie: indikation und ergebnis. Wien klin Wschr 1970;82:865-8. 63. Sourek K. Commissural myelotomy. J Neurosurg 1969;31:524-7. 64. Wertheimer P, Lecuire J. La myelotomie commissurale posteriure. A propos de 107 observations. Acta Chir Belg 1953;52:568. 65. Willis WD Jr, Westlund KN. The role of the dorsal column pathway in visceral nociception. Curr Pain Headache Rep 2001;5:20-6. 66. Hitchcock E. Stereotactic cervical myelotomy. J Neurol Neurosurg Psychiatry 1970;33:224-30. 67. Gildenberg PL, Zanes C, Flitter M, Lin PM, Lautsch EV, Truex RC. Impedance measuring device for detection of penetration of the spinal cord in anterior percutaneous cervical cordotomy. Technical note. J Neurosurg 1969;30:87-92. 68. Tasker RR. Percutaneous cordotomy. Compr Ther 1975;1:51-6.
2169
2170
128
Ablative spinal cord procedures for cancer pain
69. Levin AB, Cosman ER. Thermocouple-monitored cordotomy electrode. Technical note. J Neurosurg 1980;53:266-8. 70. Krieger AJ, Rosomoff HL. Sleep-induced apnea. I. A respiratory and autonomic dysfunction syndrome following bilateral percutaneous cervical cordotomy. J Neurosurg 1974;40:168-80. 71. Krieger AJ, Rosomoff HL. Sleep-induced apnea. II. Respiratory failure after anterior spinal surgery. J Neurosurg 1974;40:181-5. 72. Gildenberg PL, Lin PM, Polakoff PP, Flitter MA. Anterior percutaneous cervical cordotomy: determination of target point and calculation of angle of insertion. Technical note. J Neurosurg 1968;28:173-7.
73. Kanpolat Y, Akyar S, Caglar S, Unlu A, Bilgic S. CT-guided percutaneous selective cordotomy. Acta Neurochir (Wien) 1993;123:92-6. 74. Kanpolat Y, Deda H, Akyar S, Caglar S. C.T.-guided pain procedures. Neurochirurgie 1990;36:394-8. 75. Gildenberg PL, DeVaul RA. Management of chronic pain refractory to specific therapy. In: Youmans JR, editor. Neurological surgery. Philadelphia: W.B.Saunders; 1982. p. 3749-68. 76. Bonica JJ, Loeser J. Management of pain. 2nd ed. Philadelphia: Lea and Febiger; 1990. 77. Kanner R. Diagnosis and management of pain in patients with cancer. Basel: Karger; 1988.
Functional Neurosurgery for Pain
118 Anatomy and Physiology of Cancer Pain W. D. Willis Jr. . K. N. Westlund
The pain system, like the visual system, is a complex neural apparatus that consists of specialized peripheral sensory receptor organs, an initial central neural processing network, several divergent higher-level relay stations that mediate a variety of reflex reactions to painful stimuli, and variously distributed thalamocortical and limbic circuits for sensory perception, learning, memory and other cognitive activity, and emotional responses (> Figure 118-1) [1,2]. In addition to the ascending part of the pain system, there are descending pathways that modulate pain either by inhibiting nociceptive transmission (endogenous analgesia systems) or by enhancing it (amplifier effect). This chapter deals only with the ascending component of the pain system. This was the topic of a previous review [2], and the emphasis here is on findings since that review. There are detailed accounts of earlier work on the pain system, including both the ascending and the descending components [2–11].
Nociceptors and the Encoding of Pain Sensation Certain peripheral sensory receptors are clearly specialized for signaling noxious events. The term nociceptor was coined by Sherrington [12] to describe receptors that respond to damaging or potentially damaging stimuli. However, convincing recordings from nociceptors were not made for more than another half century [2,10,13,14]. Most experimental work on nociceptors has been concerned with cutaneous nociceptors. However, nociceptors have been described in muscle, joint, #
Springer-Verlag Berlin/Heidelberg 2009
and visceral nerves [15–18]. By now, nociceptors have been well characterized not only in animals but in human subjects [19–21]. Cutaneous pain can have a pricking or a burning (or dull) quality, whereas musculoskeletal pain has an aching quality [22]. The quality of cutaneous pain has been shown to be related to the activation of particular classes of nociceptors. For example, stimulation of cutaneous A delta nociceptors in human nerves produces pricking pain and stimulation of C polymodal nociceptors causes burning or dull pain [21,23,24]. However, stimulation of some C polymodal nociceptors causes a sensation of itch rather than pain [21]. Intraneural stimulation of muscle nociceptors causes aching or cramping pain [25,26]. Evidently, there are labeled line coding mechanisms for the sensory qualities of cutaneous pricking and burning pain and for muscular aching pain as well as for itch. There is still controversy about how visceral pain is encoded [27]. In one view, the intensity of visceral pain is a reflection of the total input over visceral mechanoreceptors. In another, there are specific visceral nociceptors with thresholds higher than those of mechanoreceptors. An argument in favor of the proposal that there are specific visceral nociceptors comes from the discovery of ‘‘silent’’ nociceptors in visceral as well as joint and cutaneous nerves.
Silent Nociceptors Schaible and Schmidt [28,29] found that a large proportion of the group IV (or C) fibers that supply the knee joint in cats are unresponsive to
1986
118
Anatomy and physiology of cancer pain
. Figure 118-1 Schematic diagram showing elements of the pain system (Redrawn from Willis [1])
joint movements unless the joint is inflamed. Inflammation sensitizes previously silent joint afferents, causing them to become spontaneously active as well as highly responsive to movements, even in the innocuous range (> Figure 118-2). Since this discovery, ‘‘silent’’ nociceptors have been discovered in visceral nerves [30] and cutaneous nerves [31], although not in muscle nerves [15]. ‘‘Silent’’ nociceptors provide an extreme example of a characteristicofmany nociceptors: the propertyof sensitization after damage [32–36]. Under normal
conditions, these receptors are quiescent and are insensitive to mechanical stimuli. However, they can be sensitized by inflammation, after which they become spontaneously active and show responses even to weak mechanical stimuli. The background activity found in sensitized primary afferent nociceptors presumably underlies the pain felt in the area of damage. In addition, sensitization of primary afferent nociceptors accounts for a reduced threshold for evoking pain and an increase in pain evoked by a suprathreshold noxious stimulus
Anatomy and physiology of cancer pain
118
. Figure 118-2 Activity recorded from an initially silent and unresponsive group IV or C nociceptor supplying the knee joint of a cat. The top traces in (a) and (b) show the responses of the afferent fiber, whereas the lower traces signal passive movements of the knee. Flexion, innocuous outward rotation (OR), and noxious outward rotation (n.OR), of the knee had no effect initially, as shown in the control records in (a) and (b). The unit also could not be activated by local pressure on the joint capsule (c). However, after kaolin and carrageenan injection (times after injection are indicated) and the development of inflammation, the unit displayed spontaneous activity and responses to the stimuli (a and b, middle records). Furthermore, it could now be activated by localized pressure (c, middle and right) [29]
in a damaged area [34–37]. This change in pain sensitivity is called primary hyperalgesia [22,38]. The mechanism of sensitization involves the activation of second messenger systems in the nociceptive fibers by inflammatory mediators [39]. For example, bradykinin enhances the discharges of C fibers by means of the protein kinase C (PKC) signal transduction cascade [40]. Bradykinin, through bradykinin receptors, activates phospholipase C, resulting in an increased production of inositol 1,4,5,-triphosphate (IP3) and diacylglycerol (DAG). IP3 increases the intracellular concentration of Ca2+, and DAG activates PKC. PKC in turn phosphorylates ion channel proteins, resulting in enhanced permeability when the channels are opened.
Receptors on Receptors Another argument in favor of specific nociceptors is the sensitivity of certain sensory receptors to capsaicin, the active ingredient in chili peppers [41–44]. The responses of nociceptors to capsaicin can be attributed to the presence of capsaicin (or ‘‘vanilloid’’) receptors in the membranes of these afferent fibers. These receptors have been mapped by resiniferatoxin binding and are found in dorsal root and cranial nerve ganglia and also in the brain stem in the nucleus of the solitary tract and the spinal nucleus of the trigeminal nerve [45]. Activation of capsaicin receptors opens nonselective cation channels that powerfully activate and in high doses block conduction
1987
1988
118
Anatomy and physiology of cancer pain
in the sensory axons [44]. An excess of capsaicin can destroy nociceptors through an excitotoxic action, especially if it is administered neonatally. Even in adults, capsaicin can cause selective degeneration of unmyelinated sensory fibers containing substance P and calcitonin gene-related peptide in organs such as the ureter [46,47]. This suggests that capsaicin could be used as a therapeutic agent. However, the ability of capsaicin to desensitize capsaicin receptors and block nerve conduction at doses below those needed for the destruction of sensory axons provides an alternative approach. The capsaicin receptor agonist resiniferatoxin has been found to be 1,000 times more potent than capsaicin in desensitizing capsaicin receptors but only about 100 times as potent in provoking inflammation. Resiniferatoxin introduced into the urinary bladder has been found to desensitize bladder afferents for several months, and it may be possible to produce analgesia while minimizing inflammation in the treatment of bladder conditions by employing this agent rather than capsaicin itself [48]. A number of other pharmacological receptors have been discussed in association with the functions of primary afferents, including nociceptors [49]. These receptors include those related to excitatory amino acids, opiate, serotonin, and gamma-aminobutyric acid (GABA) receptors. Transmitter-receptor interactions include rapid, short-lived ligand-gated cation channel events as well as slower, more persistent second messenger events that can result in the sensitization of dorsal horn neurons. Sensitization will be discussed below. The prolonged duration of the effects of receptor activation can be due initially to coactivation by cooperative transmitter substances and then to effects of transmitter substances coupled with second messenger cascades. In the case of the excitatory amino acid receptors, the glutamate N-methyl-D-aspartate (NMDA) receptor has been shown to be greatly potentiated by other transmitter substances,
including glycine acting at the glycine site [50], substance P [51,52], and neurokinin A [53]. In the case of neurokinin A, Duggan and coworkers [54] suggested that neurokinin’s resistance to enzymatic degradation also contributes to the potentiation of NMDA action for up to several hours after an inflammatory injury. Primary afferent calcitonin gene-related peptide has been shown to block the degradation of substance P [55,56], prolonging its effect. Specific discussion exists in the literature suggesting that these receptor events can occur on the terminal membranes of primary afferent fibers as well as on the structures they innervate [54,57–59]. Recently, specific antibody localization methods have demonstrated presynaptic primary afferent sites for NMDA autoreceptors [60] and serotonin 5HT3 receptors [61]. Specific antibodies against synthetic delta [62] and mu [63] opioid receptors have been localized in primary afferent ganglia and terminals. Terminals containing both opioid receptor types were localized adjacent to enkephalin-containing terminals in the dorsal horn, suggesting a role for these receptors in the presynaptic regulation of transmitter release in the dorsal horn. The delta opioid receptor localization was coincident with staining for calcitonin gene-related peptide, and both opioid receptors colocalized in the same primary afferent neurons. Putative substance P-containing dorsal root ganglion cells, which express preprotachykinin A mRNA, have also been shown to express mu (90%) and kappa (30%) opioid receptor subtypes with a double in situ hybridization method [64]. Although substance P receptors have not been found in association with primary afferent neurons, a dense population of fusiform-shaped immunoreactive neurons was identified in lamina I [65]. Similarly, no cells were found in the primary afferent termination zone in lamina II, which contains the greatest number of substance P terminals. Immunoreactive cells in deeper laminae (III through V), however, were labeled along
Anatomy and physiology of cancer pain
the length of large dendrites extending dorsally into lamina II. The mRNA for another peptide, neuropeptide Y, has been colocalized in small dorsal root ganglion neurons with calcitonin gene-related peptide, substance P, and galanin [66]. There have been several reports of specific peripheral receptor action and neuronal localization, including excitatory amino acid and opioid receptors. In pharmacological studies in which the hind-paw was superfused, antidromic stimulation-induced release of substance P in the periphery was inhibited by intraarterial morphine (mu) and [D-Ala2,D-Leu5] enkephalin (delta) [67]. The inhibitory effect of morphine was antagonized by pretreatment with naloxone, suggesting that opioid receptors can regulate the release of substance P from peripheral nerve endings into the extravascular space. Some evidence has been reported for the existence of transmitter receptors on the peripheral endings of primary afferents. The noxious effects of peripheral application of serotonin to painful blister bases in human skin can be blocked by 5HT3 serotonin receptor antagonists [68]. Peripheral administration of glutamate and its agonists cutaneously [69–71] or of combinations of excitatory amino acids directly into the knee joint [72] can produce acute pain-related behaviors in rats. Electron microscopic localization of NMDA, kainate, and a-amino-3-hydroxy-5methylisoxazoleproprionic acid (AMPA) glutamate receptors in glabrous skin has also been reported [69]. Several studies have shown that dorsal horn transmitter content is dependent on primary afferent input to the dorsal horn. This includes, for instance, a significantly reduced GABA content in laminae I through III interneurons after neurectomy [73] and increases after carrageenaninduced inflammation in the hindpaw [74]. Increases in dorsal horn glutamate content after knee joint inflammation may include increases within the interneurons as well as in primary afferent terminals [75].
118
Dorsal Root and Lissauer’s Tract Most of the axons of nociceptors enter the dorsal horn of the spinal cord by way of the dorsal roots [2]. In primates, the fine primary afferent fibers migrate to a lateral position in the dorsal root as it enters the spinal cord, allowing the fine afferent fibers to follow a direct route of entry into Lissauer’s tract [76]. Some nociceptive afferents also enter the spinal cord through ventral roots (> Figure 118-3) [77–81], although how many is controversial [82]. The physiological effects of stimulation of ventral root afferents appear to be mediated by branches of the afferents that enter the spinal cord through dorsal roots [83–86]. . Figure 118-3 Course of a ventral root afferent into the gray matter of the spinal cord. The fiber was labeled with horseradish peroxidase applied to the ventral root. It entered the ventral horn (A), continued dorsally into the dorsal horn through the nucleus proprius (NP), and ended (B) in the substantia gelatinosa (SG) and lamina I [77]
1989
1990
118
Anatomy and physiology of cancer pain
However, after peripheral nerve injury, ventral root afferents sprout [87–90], and the number that penetrate into the spinal cord increases after a delay of about 6 months [91]. Thus, a component of neuropathic pain could involve nociceptive ventral root afferents [92]. Lissauer’s tract contains the branches of fine primary afferent fibers and the axons of dorsal horn interneurons [93,94]. The axons bifurcate as they enter Lissauer’s tract, and their branches extend rostrally and caudally. Individual cutaneous C-fiber nociceptors have been traced for only a short distance rostrocaudally in Lissauer’s tract [95,96]. Interestingly, visceral C fibers have an extensive distribution, traveling for up to six segments in Lissauer’s tract [96].
Dorsal Horn The spinal cord has been described, using Nisslstained sections, as a laminated structure [97]. The dorsal horn of the enlargements consists of
laminae I through VI (> Figure 118-4). Lamina I, or the marginal layer, is a thin layer that caps the dorsal horn; the apical part of lamina I expands beneath Lissauer’s tract. Lamina II is the substantia gelatinosa. Laminae III through IV complete the superficial part of the dorsal horn, lamina V the neck, and lamina VI the base. Laminae I, II, and V are regarded as particularly important for nociceptive processing [2,10]. Another important region for nociception, especially visceral nociception, is the gray matter around the central canal: lamina X [98]. The primary afferent fibers that supply mechanoreceptors of the skin and muscle synapse in characteristic termination zones in the dorsal horn, intermediate region, and ventral horn (> Figure 118-4, left) [11]. Cutaneous A delta nociceptors terminate chiefly in laminae I and V, although some branches reach lamina X (> Figure 118-4, right) [99]. Cutaneous C nociceptors end primarily in laminae I and II (> Figure 118-4, right) [95]. Muscle afferent terminals in the dorsal horn are concentrated in
. Figure 118-4 Termination zones of primary afferent fibers in the spinal cord gray matter. The dotted lines in the drawing of a transverse section of the spinal cord show the boundaries of the gray matter and of laminae I through VI of the dorsal horn. The characteristic zones of termination of large muscle afferents and cutaneous mechanoreceptors and of thermoreceptors are shown on the left, and those of cutaneous nociceptors are shown on the right [11]
Anatomy and physiology of cancer pain
laminae I and V [100], as are joint afferent endings [101]. Visceral C fibers have an extensive distribution, with endings in laminae I, II, V, and X; some terminals also distribute contralaterally [96]. Many of the dorsal horn neurons that receive input from primary afferent nociceptors are interneurons. They may be excitatory or inhibitory interneurons, and their role in nociceptive processing is complex. Afferent input may be influenced by intraspinal and brain stem relays that affect the outcome of the information relayed by the synaptic transmission. A detailed understanding of interneuronal effects on nociceptive projection neurons will depend on careful morphological studies of the neural circuitry of the dorsal horn, combined with electrophysiological and neuropharmacological characterization of the interactions of nociceptive and nonnociceptive afferents, interneurons, and projection neurons. Detailed descriptions of the responses of dorsal horn interneurons to noxious and innocuous stimulation of the skin and descriptions of muscle, joint, and visceral receptors can be found elsewhere [2,6,10,15–18]. Recent work on substance P (SP) receptors in the dorsal horn has provided a number of interesting insights into the cellular mechanisms of nociceptive processing. SP receptors are especially concentrated in lamina II; however, none of the interneurons of the substantia gelatinosa appear to have SP receptors [65]. Instead, the SP receptors are on neurons of lamina I and on the dendrites of neurons in the deeper layers of the dorsal horn that extend dorsally into lamina II. After noxious stimuli, SP receptors undergo endocytosis; they regain their normal distribution within an hour [102].
Windup and Central Sensitization Repeated volleys in C fibers result in progressively enhanced responses of dorsal horn neurons,
118
a process called windup [103]. Windup is attributed to a prolonged depolarization after each C-fiber volley (> Figure 118-5a) [104–106]. The depolarization produced by a C-fiber volley consists of an early component lasting tens of milliseconds that is due to mono- and polysynaptic responses mediated by non-NMDA glutamate receptors [107]. A second slow component lasting hundreds of milliseconds is due to actions of excitatory amino acids on NMDA glutamate receptors and actions of peptides on neurokinin receptors [103,107–109]. A third prolonged component is produced by activation of neurokinin receptors [104,110]. Repetitive stimulation of C fibers results in a buildup of the slow depolarization that triggers progressively more action potentials (> Figure 118-5b). The early component of windup can be blocked by antagonists of the NMDA type of glutamate receptor
. Figure 118-5 (a) Components of the prolonged depolarization evoked by a volley in C fibers. The recording is an extracellular potential electronically propagated into a ventral root in an in vitro preparation of rat spinal cord. The depolarization is thus a compound synaptic potential recorded from motoneurons in the ventral horn. The fast component, early slow component, and prolonged slow component are indicated by the white, black, and shaded areas [104] (b) Windup in action potential generation is shown to be due to summation of the prolonged depolarization in response to repetitive stimulation of C fibers. The intracellular recording was made from a large neuron of the ventral horn in an in vitro preparation of rat spinal cord [105]. □ fast component (0–0.1 s); ▄ slow component (0.1–1 s); ░ prolonged component (1–10 s)
1991
1992
118
Anatomy and physiology of cancer pain
[111–113]. whereas the late component is blocked by antagonists of neurokinin (NK1 and NK2) receptors [108,109]. With strong and repeated C-fiber input to the spinal cord, there is a long-lasting enhancement of the responses of dorsal horn neurons as a result of the cooperative action of excitatory amino acids and peptides [113–115]. (see the section on the sensitization of spinothalamic tract neurons, below). This ‘‘central sensitization’’ can occur in vitro and thus independently of peripheral sensitization of nociceptors and can last for an hour [116]. Central sensitization involves the activation of second messenger systems such as nitric oxide, arachidonic acid, and PKC [117]. It accounts for the development of secondary hyperalgesia and allodynia in areas of skin away from the zone of damage [35,118– 122]. Although central sensitization depends on the activation of C fibers, the pain of allodynia is triggered by innocuous tactile stimuli such as brushing of the skin. Thus, the proximate cause of allodynia is the excitation of A beta mechanoreceptor afferents [123], although the underlying cause is the activation of C fibers. Normally, these tactile afferents never cause pain, but they do after central sensitization [120]. A proposed mechanism is that wide-dynamic-range projection neurons such as spinothalamic tract cells become much more responsive to tactile stimuli [124]. If these neurons signal only pain, hyperresponsiveness to tactile stimuli will result in allodynia. Longer-term changes can be attributed to other factors, such as sprouting and subsequent reorganization of dorsal horn circuits [125,126]. A number of morphological changes, including changes in the expression of peptide neurotransmitters and their receptors and in substances such as enzymes, have been described in the dorsal horn or the dorsal root ganglia after peripheral inflammation or injury [127–130]. These changes are presumably mediated through an action of immediate, early genes such as c-fos and c-jun, which are expressed by nociceptive dorsal horn neurons
after noxious stimulation [131–134]. Immunocytochemical and pharmacological studies find visceral afferent modulation occurs through a variety of neuropeptides (CGRP, CCK, enkephalins, VIP, and somatostatin), while substance p is equally important in somatic and visceral afferent modulation of dorsal horn events [135].
The Efferent Role of Afferents in Amplification of Nociceptive Input Recent studies have described an efferent role for primary afferents that, in conjunction with dorsal horn sensitization, can result in amplification . Figure 118-6 Role of dorsal root reflexes (DRRs) in peripheral inflammation. Left. The hindlimb of a rat with a knee inflamed by injection of kaolin and carrageenan into the joint capsule. There is increased activity in the fine joint afferents, leading to greatly increased input into the central nervous system. This causes not only pain but also central sensitization and hyperalgesia. In addition, there is upregulation of GABAergic inhibitory processes, resulting in the triggering of DRRs by the enhanced primary afferent depolarization. The DRRs in turn invade the peripheral terminals of the joint afferents, presumably releasing peptides and other substances. Calcitonin gene-related peptide (CGRP) would cause vasodilation, and substance P (SP) would cause plasma extravasation. Joint swelling is reduced to half if nonNMDA glutamate receptors are blocked with CNQX or if GABAA receptors are blocked with bicuculline administered in the spinal cord dorsal horn [136]
Anatomy and physiology of cancer pain
of joint inflammation and nociception (> Figure 118-6) [137]. It has been hypothesized that the hyperalgesia accompanying knee joint inflammation by kaolin and carrageenan is initiated when afferent fibers release excitatory amino acids that act on non-NMDA glutamate receptors in the superficial dorsal horn, affecting both excitatory and inhibitory interneuronal circuits. Using in vivo spinal microdialysis, it was determined that after initiation of the inflammatory process, there are increasing concentrations of excitatory amino acids in the dorsal horn [138,139], further increasing the excitability of superficial dorsal horn interneuronal circuits. Sensitized excitatory interneurons activate projection neurons through NMDA receptors that transmit signals that lead to the perception of pain by higher brain centers [140]. Sensitized interneurons can also exert a depolarizing presynaptic effect through an action on GABAA receptors on primary afferent endings, probably through dendroaxonic synapses [141,142]. If the primary afferent depolarization is powerful enough, antidromic activity in primary afferent fibers is initiated and relayed back out to the periphery as a dorsal root reflex [143– 150]. In arthritis, at least some of the dorsal root reflexes are in A delta and C fibers [137]. Peptides and perhaps other inflammatory mediators in the afferents are released into the peripheral tissue, contributing to inflammatory processes and the recruitment of previously ‘‘silent’’ afferents [28,29]. Studies in our laboratory have shown that inhibition of GABAA or non-NMDA glutamate receptors in the spinal cord is sufficient to reduce the edema by one-half and to reduce significantly the behavioral hyperalgesia and thermal increase seen in this acute knee joint model [75,151–157]. Thus, the neurogenic component contributing to peripheral inflammation is in part mediated by spinal and dorsal root reflex mechanisms. It is plausible that spread of the spinal sensitization and dorsal root reflexes contralaterally explains the mirror pain and inflammation seen clinically. All these amplification
118
events contribute not only to the developing neurogenic inflammation but also to the degree of persistence of the pain state.
Nociceptive Dorsal Horn Projection Neurons Most of the evidence from experiments on animals and from clinical studies indicates that the principal nociceptive pathways in primates, including humans, and rats ascend in the anterolateral quadrant of the spinal cord [2,10,158,159]. The nociceptive pathways of the anterolateral quadrant include the spinothalamic, spinoreticular, spinoparabrachial, spinomesencephalic, and spinohypothalamic tracts.
Spinothalamic Tract Spinothalamic tract (STT) neurons in primates are located mostly in the dorsal horn, especially in laminae I and IV through VI [160–162]. Individual STT cells have been injected intracellularly with horseradish peroxidase and examined at the light and ultrastructural levels (> Figure 118-7a) [163–168]. Immunocytochemical staining at the ultrastructural level can reveal the neurotransmitters in synapses made with STT cells. Almost half the synaptic profiles that contact the cell bodies and proximal dendrites of STT cells contain glutamate (> Figure 118-7b) [163]. Another quarter of these profiles contain GABA [167]. A small fraction contain SP [169] or calcitonin gene-related peptide (CGRP) [166]. and some contain catecholamines [168] or serotonin [170]. The axons of STT cells cross in the ventral white commissure at a level close to that of the cell body (> Figure 118-7a). The axons then turn rostrally in the ventral funiculus, eventually migrating into the ventrolateral funiculus as they ascend toward the brain stem. The axons of lamina I STT cells travel rostrally in the middle [171] or even the dorsal part of the lateral funiculus [162].
1993
1994
118
Anatomy and physiology of cancer pain
. Figure 118-7 (a) A primate spinothalamic tract neuron injected intracellularly with horseradish peroxidase (HRP). The STT neuron was of the high-threshold (nociceptive-specific) type. The cell body was located in the lateral part of lamina IV. It had long dendrites that extended dorsolaterally as far as lamina I and ventromedially nearly to lamina X. The axon is shown crossing the midline just ventral to lamina X. Forty-six percent of the terminal profiles that contacted the cell body of this neuron were immunoreactive for glutamate (b) Glutamate-containing synaptic terminal (L2 type, containing a large number of large, dense core vesicles) on the HRP-labeled STT cell shown in (a). Glutamate was stained by the immunogold method (gold particles are indicated by open arrows). Note the active zone (arrowheads), small clear vesicles, and dense core vesicles in the terminal. The dense core vesicles presumably contain a peptide that is colocalized with glutamate in this ending [163]
Primate STT axons terminate in the following thalamic nuclei: ventral posterior lateral (VPL), ventral posterior inferior (VPI), medial part of the posterior complex (POm), and intralaminar complex, including the central lateral (CL) and parafascicular nuclei (Pf) as well as the nucleus submedius (> Figure 118-8) [172–176]. Recently, a projection from lamina I to the posterior part of the ventral medial nucleus (VMpo) was described in monkeys [177]. STT neurons in laminae IV through VI in the lumbosacral enlargement generally respond best to stimulation of the skin [10]. Some are activated best by innocuous mechanical stimuli, but these neurons are in the minority. Most STT cells in the deep dorsal horn are classified as wide-dynamicrange neurons (activated by innocuous mechanical stimulation of the skin but more responsive to noxious intensities) or high-threshold neurons (activated entirely or chiefly by noxious cutaneous stimuli). The receptive fields are discrete and
contralateral to the thalamic target of the axon. The size of the receptive fields varies from very small to large. The same STT cells generally also respond to noxious heating (or cooling) of the skin and to intradermal injection of capsaicin. Some have been found to respond to noxious stimulation of muscle [178] or a joint [179]. Primate STT cells located in segments away from the enlargements are likely to have strong inputs from muscle and viscera [180]. The same neurons typically have a convergent input from the skin at the same dermatomal level as the muscle or visceral input [181,182]. However, visceral inputs entering distant segments of the spinal cord are likely to be inhibitory [183,184]. This form of inhibition appears to involve a supraspinal loop and a relay in the gray matter of the upper cervical spinal cord [184]. Primate STT cells in the deep dorsal horn can readily be sensitized by strong noxious stimulation of the skin. The stimulation can be mechanical
Anatomy and physiology of cancer pain
118
. Figure 118-8 Schematic drawing of the organization of the spinothalamic tract (STT), spinoreticular tract (SRT), and spinomesencephalic tract (SMT). The gray matter of the left side of the spinal cord is shown at the left from a dorsal perspective (the cervical and lumbar enlargements are indicated by C and L). The synapses of primary afferent fibers on second-order projection cells are shown only in the lumbar enlargement. The axons of the projection cells cross the midline and then ascend to the brain. SRT and SMT axons end in the medulla (Med), pons (P), and midbrain (M). The STT is shown to distribute to a number of thalamic nuclei, including the ventral posterior lateral nucleus (both caudal and oral parts, VPLc and VPLo), the medial part of the posterior group (Pom), the central lateral nucleus (CL), and the nucleus submedius (SM) [10]
[185], thermal [121,186]. or chemical (> Figure 118-9) [188,189], or it can involve inflammation, as in experimental arthritis [140]. Increased responses to innocuous and weak noxious mechanical
stimuli are found when parts of the receptive field that have not been damaged are stimulated. Also, the size of the cutaneous receptive field expands, and so stimuli from a wider area can activate a
1995
1996
118
Anatomy and physiology of cancer pain
. Figure 118-9 Sensitization of a primate spinothalamic tract cell after an intradermal injection of capsaicin. The background discharge of the neuron is shown in (a). The capsaicin injection produced a dramatic increase in activity (b). Before the injection, the neuron responded to brushing, pressure, and pinch stimuli at five points in the cutaneous receptive field (c, e, g; the points stimulated are shown in the drawing of the leg at the bottom). The responses were enhanced after the capsaicin injection (d, f, and h). The receptive field also enlarged (compare doubly hatched to hatched areas in drawing) [187]
given neuron. Thus, central sensitization of STT cells (and other projection neurons) or of the neural pathways that excite them can help account for secondary hyperalgesia and allodynia [124].
Sensitization of primate STT cells can be produced by coapplication of an excitatory amino acid such as NMDA or quisqualic acid along with SP onto STT cells [189,190]. Central
Anatomy and physiology of cancer pain
sensitization resulting from activation of C fibers by intradermal injection of capsaicin results from the release in the dorsal horn of excitatory amino acids and peptides such as SP, since central sensitization can be prevented or reduced by spinal cord administration of excitatory amino acid receptor antagonists as well as by antagonists of neurokinin receptors [189,191]. Central sensitization results from the activation of signal transduction pathways, including the PKC pathway [192,193]. A special set of deep STT cells projects to the CL nucleus and other nuclei of the medial thalamus [194]. These STT cells have very large receptive fields, often encompassing the entire body surface. They respond only to intense noxious stimuli and are powerfully excited by stimulation in the pontine reticular formation. These neurons are clearly nociceptive, but their properties do not suit them for a role in sensory discrimination of pain. They often have convergent input from visceral afferent fibers [195]. STT cells in lamina I [196–198] respond best to noxious mechanical stimulation of the skin and can be classified as wide-dynamic-range or high-threshold cells [196,197] or as nociceptive specific, heat-pinch-cold, cold, warm, or widedynamic-range cells [199,200]. Lamina I cells belong to four morphological types (fusiform, pyramidal, multipolar, and flattened neurons); this was based on tissue sections cut in three planes [201]. STT cells in rats are either pyramidal or flattened cells [202]. According to Zhang and Craig [203], lamina I STT cells in monkeys examined in transverse sections are fusiform, pyramidal, or multipolar cells. On the basis of recordings from lamina I STT cells in cats, fusiform cells are predicted to be nociceptive-specific; pyramidal cells, cold-specific; and multipolar cells, multimodal or nociceptive-specific [204]. A number of differences have been observed between STT cells in laminae I and V. Besides cell shape, a striking morphological difference is that there are many fewer synaptic contacts on the cell bodies of lamina I cells compared with lamina V STT cells [168,205]. The receptive fields of
118
nociceptive lamina I STT cells tend to be small [206]. Lamina I STT cells are almost never classified as low-threshold cells but instead are generally considered high-threshold or widedynamic-range cells [198]. Many lamina I STT cells are thermoreceptive rather than nociceptive [199,200], whereas lamina V STT cells do not appear to receive an input from specific thermoreceptors [207,208]. Finally, STT cells of lamina I are harder to sensitize than are STT cells in the deep layers of the dorsal horn [188]. This observation may relate to the weak or absent input to many of these neurons from sensitive mechanoreceptors. McMahon and Wall [209] have suggested that lamina I cells contribute to a positive feedback loop that involves excitatory connections to the brain stem and back to lamina I. This reverberatory circuit has an output from brain stem structures that inhibit cells of the deep dorsal horn, including STT cells. Stimulation in the anterior pretectal nucleus has an antinociceptive effect and excites neurons in lamina I while inhibiting neurons in lamina V, including in both cases STT cells [156,157]. Evidently, nociceptive neurons in laminae I and V play quite different roles in nociception. There is a large group of neurons in the gray matter of the C1 and C2 segments that project to the thalamus [10,161,210,211]. Neurons in the same region also project to the reticular formation and the spinal cord. Neurons in this region have large, complex receptive fields [212,213] and resemble neurons of the reticular formation and STT cells in the lumbosacral enlargement that project to the intralaminar nuclei [194].
Spinoreticular, Spinoparabrachial, Spinomesencephalic, and Spinohypothalamic Tract Neurons Spinal neurons that project chiefly to the medial pontomedullary reticular formation (> Figure 118-8) are located chiefly in the deeper
1997
1998
118
Anatomy and physiology of cancer pain
layers of the dorsal horn [214]. These cells often have large, complex receptive fields, and their response properties are more suited to nociceptive functions other than sensory discrimination, such as arousal, attention, and other functions generally attributed to the caudal reticular formation [10,215,216]. Spinal projections to the parabrachial region, midbrain reticular formation, and periaqueductal gray matter are from neurons in lamina I and in deeper layers of the spinal cord gray matter (> Figure 118-8) [10,177,217–224]. Such neurons have properties similar to those of STT cells in the superficial or deep dorsal horn [10,225,226]. Spinomesencephalic neurons with large, complex receptive fields have been described in the upper cervical spinal cord [220]. The axons of lamina I projection neurons have been followed as they course through the brain stem, using an anterograde tracing technique [177]. These axons pass through the column formed by the catecholaminergic cell groups of the medulla and pons and have collaterals that terminate on and among these cells (> Figure 118-10) [227]. This pathway can provide the catecholamine system with information about the state of the body and thus contribute to appropriate regulatory adjustments through an action on autonomic and other control centers. Subsequent modulation of nociception also may occur through the descending noradrenergic system. A spinohypothalamic tract (SHT) has been described in rats and cats [210,226,228–232]. The cells are in locations similar to those of STT cells [210], and SHT neurons have responses to cutaneous stimuli that are similar to those of STT cells [226,233]. There are also direct projections to other limbic structures, such as the amygdaloid nucleus, preoptic area, and septal nuclei, as well as to the basal ganglia [224,229]. These pathways, as well as relays through the parabrachial region and the periaqueductal gray matter, can provide access for nociceptive information to
the telecephalic limbic system. This may help explain the profound motivational-affective responses commonly seen in individuals suffering from pain. In addition to the nociceptive pathways that ascend in the anterolateral quadrant of the spinal cord, more posteriorly situated pathways have been suggested to play a role in pain. These pathways include the dorsal column–medial lemniscus system (including the postsynaptic dorsal column path) and the spinocervicothalamic path.
Dorsal Column–Medial Lemniscus System In addition to the ascending branches of large mechanoreceptive primary afferent fibers, the dorsal column in animals is known to contain fine primaryafferentCfibers[234–236].Atleastsomeof these fine afferents terminate in the dorsal column nuclei. In addition, the axons of neurons in the dorsal horn project in the dorsal column (and some axons in the adjacent dorsal lateral funiculus) to the dorsal column nuclei. The latter neurons are termed postsynaptic dorsal column neurons [10,237–243]. Many postsynaptic dorsal column neurons respond to noxious stimuli [244–250]. Some neurons in the nucleus gracilis respond to noxious stimulation of the skin [245,251,252]. Such responses recorded in cats in the rostral part of the nucleus gracilis, which is not the main location of thalamic relay neurons, have been attributed to activation of postsynaptic dorsal column neurons [245]. However, nociceptive responses in gracile neurons that project to the VPL nucleus also have been recorded in monkeys and cats [251,252]. In humans, the dorsal column–medial lemniscus system is not regarded as an important pathway for cutaneous pain, since such pain is not affected by lesions in the dorsal part of the spinal cord [253] and since cutaneous pain can be eliminated by anterolateral cordotomy [160].
Anatomy and physiology of cancer pain
118
. Figure 118-10 Drawings of sagittal sections at different planes taken through the side of the brain stem of a monkey contralateral to the injection site in the spinal cord of an anterograde tracer (1–8, lateral to medial). The open circles show the locations of catecholamine-containing neurons as demonstrated by immunocytochemical staining for tyrosine hydroxylase. Several catecholaminergic cell groups are indicated (A1, A2, A5, A6, A7, C1). Fine dots indicate the locations of the terminals of spinal projection neurons anterogradely labeled by an injection of Phaseolus vulgaris leucoagglutinin (PHA-L) into lamina I in the cervical enlargement [227]
1999
2000
118
Anatomy and physiology of cancer pain
Pain can return months to years after an initially successful cordotomy [159]. However, it is unclear if this return of pain should be attributed to alternative pathways, the development of a central pain state, spread of the disease process, or another cause. Bilateral pain, especially when it is of visceral origin, is difficult to treat, since bilateral cordotomies are needed to interrupt the spinothalamic tract and have many side effects; furthermore, pain often recurs [158–160]. An alternative procedure that was tried by a number of neurosurgeons is midline commissural myelotomy [159,254–264]. The idea was to interrupt the decussating axons of spinothalamic neurons of both sides in a single procedure. The length of the myelotomy was tailored to match the involved segments from which the pain originated. However, pain was often relieved not only in the area where this was expected to occur but also in distant parts of the body [264]. A proposed explanation was that there may be a multisynaptic pain pathway that ascends in the central gray matter or nearby. Hitchcock [265,266] introduced a limited midline myelotomy procedure that involved a midline lesion at C1 [267], and pain was relieved in distant parts of the body. A limited midline myelotomy at T10 developed by Gildenberg and Hirshberg [268] was effective for the relief of pelvic visceral cancer pain. The spinal cord lesion relieving cancer pain in one of Hirshberg’s cases extended from the posterior surface of the spinal cord to the posterior commissure, but no damage to the gray matter was apparent [269]. We developed the hypothesis that a major pelvic visceral pain pathway ascends next to the midline in the dorsal column and tested it in animal experiments. Our initial morphological experiments in rats, using retrograde and anterograde tracing techniques, indicate that neurons that are concentrated in the dorsal commissural gray matter of the sacral cord and project to the nucleus gracilis are suitable candidates to mediate visceral nociceptive signals [269]. These
neurons belong to the postsynaptic dorsal column pathway. The nociceptive signals are then relayed to the VPL nucleus and then presumably to the cerebral cortex. It is possible that other spinal cord, brain stem, and thalamic nuclei are also involved. Noxious visceral stimuli such as colorectal distension and injection of mustard oil into the colon activate neurons in the nucleus gracilis and the VPL nucleus of the rat thalamus [269–271]. The visceroceptive responses of the gracile and VPL neurons (as well as the responses of these neurons to innocuous cutaneous stimuli) are dependent on the dorsal column pathway, since a lesion placed in the dorsal column at T10 essentially eliminates these responses (> Figure 118-11a). By contrast, a lesion of the ventrolateral column at T10 eliminates the responses to noxious cutaneous stimuli but has only a small effect on the responses to noxious visceral or innocuous cutaneous stimuli (> Figure 118-11b). Two alternative explanations for the ability of a dorsal column lesion to eliminate the responses of gracile and VPL neurons to noxious visceral stimuli are (1) interruption of the fine primary afferent fibers known to ascend in the dorsal column to the nucleus gracilis and (2) interruption of the axons of postsynaptic dorsal column neurons. To distinguish between activity mediated by this pathway and the directly projecting fine primary afferents of the dorsal column, a microdialysis fiber was placed with its open dialysis membrane in the gray matter of the cord at S2 (> Figure 118-12a) [271]. The idea was to block synaptic activation of postsynaptic dorsal column neurons in L6–S1 by administrating morphine or the non-NMDA glutamate receptor antagonist CNQX into the sacral gray matter. However, since the responses of neurons in the gracile or VPL nuclei to colorectal distension depend on relays not only in the sacral cord through the pelvic nerves but also in the lower thoracic and upper lumbar cord through the hypogastric nerves, the latter nerves were severed before the experiment. The result of
Anatomy and physiology of cancer pain
118
. Figure 118-11 (a) Effects of dorsal column (DC) and ventrolateral column (VLC) lesions of the spinal cord at T10 on the responses of a neuron in the rat ventral posterior lateral (VPL) nucleus to cutaneous and visceral stimuli. The location of the unit is shown in a drawing of a section through the thalamus, and the lesions are indicated on a drawing of a cross section of the spinal cord. The unit responded initially to mechanical stimulation of the skin (BR = brush; PR = press; PI = pinch) and to graded colorectal distension (CRD = innocuous, 20 mmHg; noxious = 40–80 mmHg). The DC lesion eliminated the responses to BR and PR and nearly eliminated the responses to CRD. The VLC lesion eliminated the remaining responses (b) Effects of the reverse sequence of lesions on another VPL unit. A VLC lesion eliminated only the PI response, whereas a later DC lesion eliminated all the other responses [270]
the study was that morphine or CNQX blocked the responses of gracile and VPL neurons to noxious visceral stimuli almost completely (> Figure 118-12b), indicating that the visceral input is relayed through the postsynaptic dorsal column pathway.
The effectiveness of dorsal column lesions in eliminating the responses of gracile and VPL neurons to noxious visceral stimuli underscores the significance of the axons that form the part of the pathway in the dorsal columns for visceral pain. More experimental work is needed to document
2001
2002
118
Anatomy and physiology of cancer pain
. Figure 118-12 (Continued)
Anatomy and physiology of cancer pain
the presence of this pathway in primates and determine in more detail the nature of this pathway. There is growing evidence that visceral pain may be conveyed through the dorsal column–medial lemniscus system. Visceral stimuli activate neurons in the dorsal column–medial lemniscus pathway [272–275]. Recently, responses to noxious stimulation of the female reproductive organs were recorded by Berkley and Hubscher [276] in the nucleus gracilis and by Berkley and associates [277] in the VPL nucleus of rats. The visceral responses in the nucleus gracilis were reduced or eliminated by lesions of the dorsal column and dorsolateral funiculus [276]. Both Foreman’s group [278] and Apkarian’s group [279] were able to record responses to noxious visceral stimuli from neurons of the monkey VPL nucleus. The cells had convergent inputs from the skin as well as from one or more viscera.
Spinocervical Tract The cells of origin of the spinocervical tract are in the dorsal horn, and this tract relays in the lateral cervical nucleus, which is in the C1–C2 segments just ventrolateral to the superficial dorsal horn [10]. A lateral cervical nucleus has been described in humans and monkeys, although it is more prominent in carnivores [10,280]. The lateral cervical nucleus projects to the contralateral VPL thalamic nucleus as well as to several sites in the brain stem [10].
118
Neurons of the spinocervical tract and the lateral cervical nucleus can respond to noxious stimuli, although their predominant function appears to be tactile [281–283].
Thalamus A number of thalamic nuclei have been implicated in nociceptive processing, including the ventral posterior lateral and medial nuclei (VPL and VPM), the VPI nucleus, the posterior group (PO), the nucleus submedius and the VMpo nucleus, and several of the intralaminar nuclei [2].
VPL, VPM, and VPI Nuclei Since the VPL nucleus in primates, including humans, receives substantial input from the STT, it is not surprising that nociceptive responses can be recorded from neurons in this region [279,284– 291]. In addition, visceral nociceptive responses occur in the VPL nucleus [278]. These responses are mediated by the dorsal column–medial lemniscus pathway [270,271,279]. Similarly, nociceptive neurons have been observed in the VPM nucleus [292–294]. Apkarian and Shi [291] also explored the VPI and PO nuclei and demonstrated nociceptive neurons in these areas. Most of the nociceptive neurons in the VPL and VPM nuclei are of the wide-dynamic-range (WDR) type, although highthreshold (HT) (or nociceptive-specific) neurons
. Figure 118-12 (a) An experimental arrangement to test the hypothesis that the postsynaptic dorsal column pathway conveys visceral input to the nucleus gracilis (NG). A recording was made from a unit in the NG. The hypogastric nerves (h.n.) were cut, and so the input from the colon was restricted to the pelvic nerve (p.n.) input to the sacral cord. A microdialysis fiber was inserted into the sacral cord at S1 to allow the infusion of morphine into the dorsal horn (b) Recordings from a neuron of the NG show that the unit responded to cutaneous stimuli (BR = brush; PR = press; PI = pinch) and to graded colorectal distensions. The responses to visceral stimuli (but not those to cutaneous mechanical stimuli) were nearly eliminated when morphine was administered into the dorsal horn. This effect was reversed by systemically administered naloxone. A lesion of the dorsal column (DC) at T10 eliminated all responses [271]
2003
2004
118
Anatomy and physiology of cancer pain
are also found in these nuclei. The nociceptive neurons in the VPI nucleus are evenly divided between WDR and HT cells [291].
Ventral Posterior Nuclei in Humans Recent studies by Lenz and coworkers demonstrated the presence of nociceptive neurons in the VPL, VPM, and probably VPI nuclei in human patients [295–299]. Stimulation through the same microelectrode that was used for recording often evoked a sensation of pain in these subjects. In a patient with angina, stimulation of the VPL nucleus evoked anginal pain [297].
Intralaminar Nuclei Early studies of the nociceptive responses of neurons in the intralaminar nuclei were reviewed previously [2]. More recently, Bullett [300] demonstrated in rats the induction of c-fos immunoreactivity in the central lateral and parafascicular nuclei after noxious stimulation. Reyes-Vazquez and associates [301] in rats and Bushnell and Duncan [302] in monkeys found nociceptive responses by many of the neurons they sampled in the intralaminar nuclei. Bushnell and Duncan [302] suggested that these neurons can help encode the intensity of noxious stimulation, although not the location. They further suggested that these cells may contribute to the motivational-affective component of pain.
Other Medial Thalamic Nuclei In one report, most of the recorded neurons in the nucleus submedius were found to respond to innocuous cooling, although some were highthreshold nociceptive neurons [303]. In another study, a large proportion of the submedius neurons that responded to cutaneous stimulation
were nociceptive [304]. C-fos labeling was induced in the nucleus by noxious stimulation [300]. Recently, Roberts and Dong [305] found that bilateral lesions that included the nucleus submedius caused signs of hyperalgesia to appear, suggesting that a function of the nucleus is antinociception. Zhang and colleagues [306] have provided evidence that stimulation in the nucleus submedius causes antinociception, as shown by prolongation of the tail flick reflex latency, and that this is mediated through the ventrolateral orbital cortex and the periaqueductal gray matter. Craig and associates [307] found a high concentration of neurons in the VMpo nucleus in monkeys that respond either to cold or to noxious stimuli. They observed a comparable area in the human thalamus that has similar staining properties for calcium-binding proteins. This nucleus projects to the insula. Craig and associates [307] proposed that it plays a role in pain and temperature sensations.
Imaging of the Thalamus During Pain Increased metabolic activity in the thalamus has been reported in several studies of experimental pain that used positron emission tomography (PET) [308–310]. However, in several PETstudies of patients with chronic pain, there was a decrease in blood flow in the thalamus [311,312]. Furthermore, anterolateral cordotomy normalized the blood flow [311]. However, a single photon emission computed tomography (SPECT) study showed hyperactivity in the thalamus of patients with central poststroke pain [313].
Cerebral Cortex Although recording studies were conducted in the past decade that demonstrated nociceptive
Anatomy and physiology of cancer pain
responses of cortical neurons [314–317], the greatest progress in determining the role of the cerebral cortex in nociceptive processing has resulted from imaging studies. Noxious stimuli have been shown to result in increases in metabolism in the SI and SII cortex, as well as in the anterior cingulate gyrus, insula, and other cortical areas [308–310,318,319]. Similarly, increases in cortical activity were found in the SI and cingulate cortices in response to painful electrical stimulation, using functional magnetic resonance imaging [320]. However, changes in cerebral blood flow were not seen in SI and SII in patients with chronic neuropathic pain, although there was an increase in the insula, posterior parietal cortex, and anterior cingulate gyrus [321]. A reduction in cortical blood flow has been reported in the SI cortex during persistent pain [322,323].
4.
5.
Conclusions The following conclusions can be drawn. 1.
2.
3.
The pain system mediates pain sensation, reflex reactions to nociceptive stimuli, learning and memory of pain, and emotional responses to pain. It also includes feedback circuits that modulate nociceptive transmission. Nociceptors are peripheral sensory receptors that respond to damaging or potentially damaging stimuli. These receptors have been found in cutaneous, muscle, joint, and visceral nerves. They contribute to labeled line pathways for different qualities of pain sensation, such as pricking and burning cutaneous pain and aching muscular pain. ‘‘Silent’’ nociceptors normally are unresponsive to mechanical stimuli but become sensitized, for example, by chemical or heat stimuli or inflammation. Sensitized nociceptors have spontaneous activity and are
6.
7.
8.
118
very responsive to previously ineffective stimuli. They contribute to spontaneous pain and primary hyperalgesia. Sensitization can be attributed to activation of signal transduction pathways in the afferent neurons. Primary afferent nociceptors are sensitive to a variety of chemical substances, such as capsaicin, excitatory amino acids, opioids, and GABA. The actions of these substances are mediated by receptors coupled to ion channels or to G proteins or other second messenger pathways. These actions can affect pre- or postsynaptic functions. Primary afferent nociceptors generally enter the spinal cord dorsal horn by way of the dorsal roots. However, a few enter through ventral roots. After peripheral nerve injury, the number of these ventral root nociceptors increases, and so ventral root afferents may contribute to the plastic rearrangements that occur in chronic pain. Primary afferent nociceptors are distributed to the dorsal horn through Lissauer’s tract. Cutaneous afferents travel only a short distance in that tract, but visceral afferents may travel as many as six segments. Different types of primary afferent fibers terminate in characteristic ways in the laminae of the spinal cord dorsal horn. Cutaneous A delta nociceptors end preferentially in laminae I, V, and X, whereas cutaneous C fibers end chiefly in laminae I and II. Muscle and joint nociceptors project to laminae I and V, and visceral nociceptors project to laminae I, II, V, and X. Interneurons of lamina II (substantia gelatinosa) do not have substance P receptors despite the rich plexus of substance P–containing terminals in this lamina. The substance P receptors are on neurons of lamina I and on the dendrites of neurons in deeper laminae whose dendrites ascend dorsally into lamina II. Substance P receptors
2005
2006
118 9.
10.
11.
12.
13.
Anatomy and physiology of cancer pain
are internalized after noxious stimulation and then are recycled. The responses of nociceptive dorsal horn neurons become greater with repeated stimulation of C-fiber nociceptors, a process known as windup. This is due to summation of long-lasting excitatory potentials mediated by NMDA glutamate receptors and neurokinin receptors. Strong, repeated C-fiber input causes a long-lasting (hours) central sensitization of dorsal horn nociceptive neurons. This involves activation of NMDA and neurokinin receptors and second messenger pathways. Central sensitization is responsible for secondary hyperalgesia and allodynia elicited by stimulation of undamaged areas. Long-lasting plastic changes in dorsal horn circuitry can be produced by alterations in peptides and enzymes contained in dorsal horn neurons and by sprouting and rewiring of the dorsal horn circuits. These changes may be mediated by an action of immediate, early genes such as c-fos and c-jun. Inflammation such as arthritis results not only in central sensitization of nociceptive pathways of the dorsal horn but also in upregulation of GABAergic inhibitory transmission. Increased presynaptic inhibition elicits dorsal root reflexes, which then increase the peripheral release of neurotransmitters, including excitatory amino acids and peptides, and enhance neurogenic edema and other signs of inflammation. Ascending nociceptive pathways in the anterolateral quadrant include the STT. A variety of studies have been done on the location, ultrastructure, synaptic neurochemistry, thalamic projections, response properties, and central sensitization of STT cells of laminae I and V. Other ascending pathways that accompanytheSTTarethespinoreticular,spinoparabrachial, spinomesencephalic, and
14.
15.
16.
spinohypothalamic tracts. Some spinal neurons make direct connections with the amygdala, septal nuclei, and basal ganglia. Nociceptive post-synaptic dorsal column neurons project their axons through the dorsal column while matter to synapse in the dorsal column nuclei, where spinocervical tract neurons send their axons to the lateral cervical nucleus through the dorsal part of the lateral funiculus. Evidence suggests that visceral pain is mediated chiefly by axons of the post-synaptic dorsal column pathway. This provides an explanation of the success of commissural myelotomies and limited midline myelotomies in relieving visceral pain in cancer patients. Recordings from thalamic neurons in primates and humans have revealed that there are nociceptive neurons in and around the ventrobasal nuclei as well as in the nuclei of the intralaminar complex and medial thalamus. Imaging studies in human subjects are consistent with this finding. Similarly, recordings have been made from nociceptive neurons in the cerebral cortex, both in SI and in SII, as well as in the anterior cingulate gyrus. Imaging studies show increases in functional activity in SI, SII, anterior cingulate, and insular regions of the cortex after noxious stimulation. However, some studies indicate that there can be a reduction in cerebral blood flow in SI during painful stimulation.
References 1. Willis WD. Physiology of pain perception. In: Takagi H, Oomura Y, Ito M, Otsuka M, editors. Biowarning system in the brain. A Naito Foundation Symposium. Tokyo: University of Tokyo Press; 1988. p. 69-85. 2. Willis WD. The pain system: The neural basis of nociceptive transmission in the mammalian nervous system. In: Gildenberg PL, editor. Pain and headache, vol. 8. Basel: Karger; 1985. p. 1-346.
Anatomy and physiology of cancer pain
3. Price DD, Dubner R. Neurons that subserve the sensorydiscriminative aspects of pain. Pain 1977;3:307-38. 4. Basbaum AI, Fields HL. Endogenous pain control mechanisms: review and hypothesis. Ann Neurol 1978;4:451-62. 5. Willis WD. Control of nociceptive transmission in the spinal cord. In: Ottoson D, editor. Progress in sensory physiology, vol. 3. Berlin: Springer; 1982. p. 1-159. 6. Besson JM, Chaouch A. Peripheral and spinal mechanisms of nociception. Physiol Rev 1987;67:67-186. 7. Fields HL, Besson JM. Pain modulation. Progress in brain research. Amsterdam: Elsevier; 1988. p. 1-29. 8. Price DD. Psychological and neural mechanisms of pain. New York: Raven Press; 1988. 9. Bonica JJ. The management of pain. 2nd ed. Philadelphia: Lea & Febiger; 1990. 10. Willis WD, Coggeshall RE. Sensory mechanisms of the spinal cord. New York: Plenum Press; 3rd ed., 2004; 1-962. 11. Light AR. The initial processing of pain and its descending control: spinal and trigeminal systems. In: Gildenberg PL, editor. Pain and headache. Basel: Karger; 1992. p. 1-306. 12. Sherrington CS. The integrative action of the nervous system. New Haven, CT: Yale University Press; 1906 (2nd ed, 1947, Yale, 1981). 13. Iggo A. Cutaneous heat and cold receptors with slowly conducting (C) afferent fibres. Q J Exp Physiol 1959;44:363-70. 14. Iriuchijima J, Zotterman Y. The specificity of afferent cutaneous C fibres in mammals. Acta Physiol Scand 1960;49:267-78. 15. Mense S. Nociception from skeletal muscle in relation to clinical muscle pain. Pain 1993;54:241-89. 16. Ness TJ, Gebhart GF. Visceral pain: a review of experimental studies. Pain 1990;41:167-234. 17. Schaible HG, Grubb BD. Afferent and spinal mechanisms of joint pain. Pain 1993;55:5-54. 18. Cervero F. Sensory innervation of the viscera: peripheral basis of visceral pain. Physiol Rev 1994;74:95-138. 19. Adriaensen H, Gybels J, Handwerker HO, Van Hees J. Response properties of thin myelinated (A-d) fibers in human skin nerves. J Neurophysiol 1983;49:111-22. 20. Adriaensen H, Gybels J, Handwerker HO, Van Hees J. Suppression of C-fibre discharges upon repeated heat stimulation may explain characteristics of concomitant pain sensations. Brain Res 1984;302:203-11. 21. Ochoa J, Torebjo¨rk E. Sensations evoked by intraneural microstimulation of C nociceptor fibres in human skin nerves. J Physiol 1989;415:583-99. 22. Lewis T. Pain. New York: Macmillan; 1942. 23. Konietzny F, Perl ER, Trevino D, et al. Sensory experiences in man evoked by intraneural electrical stimulation of intact cutaneous afferent fibers. Exp Brain Res 1981;42:219-22. 24. Torebjo¨rk HE, Ochoa JL. New method to identify nociceptor units innervating glabrous skin of the human hand. Exp Brain Res 1990;81:509-14.
118
25. Torebjo¨rk HE, Ochoa JL, Schady W. Referred pain from intraneural stimulation of muscle fascicles in the median nerve. Pain 1984;18:145-56. 26. Simone DA, Marchettini P, Caputi G, Ochoa JL. Identification of muscle afferents subserving sensation of deep pain in humans. J Neurophysiol 1994;72:883-9. 27. Cervero F, Ja¨nig W. Visceral nociceptors: a new world order? Trends Neurosci 1992;15:374-78. 28. Schaible HG, Schmidt RF. Effects of an experimental arthritis on the sensory properties of fine articular afferent units. J Neurophysiol 1985;54:1109-22. 29. Schaible HG, Schmidt RF. Time course of mechanosensitivity changes in articular afferents during a developing experimental arthritis. J Neurophysiol 1988;60: 2180-95. 30. Ha¨bler HJ, Ja¨nig W, Koltzenburg M. Activation of unmyelinated afferent fibres by mechanical stimuli and inflammation of the urinary bladder in the cat. J Physiol 1990;425:545-62. 31. Handwerker HO, Kilo S, Reeh PW. Unresponsive afferent nerve fibres in the sural nerve of the rat. J Physiol 1991;435:229-42. 32. Bessou P, Perl ER. Response of cutaneous sensory units with unmyelinated fibers to noxious stimuli. J Neurophysiol 1969;32:1025-43. 33. Fitzgerald M, Lynn B. The sensitization of high threshold mechanoreceptors with myelinated axons by repeated heating. J Physiol 1977;265:549-63. 34. Meyer RA, Campbell JN. Myelinated nociceptive afferents account for the hyperalgesia that follows a burn to the hand. Science 1981;213:1527-29. 35. LaMotte RH, Thalhammer JG, Torebjo¨rk HE, Robinson CJ. Peripheral neural mechanisms of cutaneous hyperalgesia following mild injury by heat. J Neurosci 1982;2:765-81. 36. LaMotte RH, Thalhammer JG, Robinson CJ. Peripheral neural correlates of magnitude of cutaneous pain and hyperalgesia: a comparison of neural events in monkey with sensory judgments in human. J Neurophysiol 1983;50:1-26. 37. Koltzenburg M, Lundberg LER, Torebjo¨rk HE. Dynamic and static components of mechanical hyperalgesia in human hairy skin. Pain 1992;51:207-19. 38. Hardy JD, Wolff HG, Goodell H. Pain sensations and reactions. New York: Williams & Wilkins; 1952 (Re-printed by Hafner, New York, 1967). 39. Dray A. Chemical activation and sensitization of nociceptors. In: Besson JM, Guilbaud G, Ollat H, editors. Peripheral neurons in nociception: physio-pharmacological aspects. Paris: John Libbey Eurotext; 1994. p. 49-70. 40. Dray A, Bettaney J, Forster P, Perkins MN. Bradykinininduced stimulation of afferent fibres is mediated through protein kinase C. Neurosci Lett 1988;91:301-7. 41. Petsche U, Fleischer E, Lembeck F, Handwerker HO. The effect of capsaicin application to a peripheral nerve on
2007
2008
118 42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
Anatomy and physiology of cancer pain
impulse conduction in functionally identified afferent nerve fibres. Brain Res 1983;265:233-40. Baumann TK, Simone DA, Shain CN, LaMotte RH. Neurogenic hyperalgesia: the search for the primary cutaneous afferent fibers that contribute to capsaicin-induced pain and hyperalgesia. J Neurophysiol 1991;66:212-27. LaMotte RH, Lundberg LER, Torebjo¨rk HE. Pain hyperalgesia and activity in nociceptive C units in humans after intradermal injection of capsaicin. J Physiol 1992;448:749-64. Szolcsa´nyi J, Po´rsza´sz R, Petho¨ G. Capsaicin and pharmacology of nociceptors. In: Besson JM, Guilbaud G, Ollat H, editors. Peripheral neurons in nociception: physiopharmacological aspects. Paris: John Libbey Eurotext; 1994. p. 109-24. Szallasi A, Nilsson S, Farkas-Szallasi T, et al. Vanilloid (capsaicin) receptors in the rat: Distribution in the brain, regional differences in the spinal cord, axonal transport to the periphery, and depletion by systemic vanilloid treatment. Brain Res 1995;703:175-83. Chung JM, Lee KH, Kim J, Coggeshall RE. Activation of dorsal horn cells by ventral root stimulation in the cat. J Neurophysiol 1985;54:261-72. Sann H, Jancso´ G, Ambrus A, Pierau FK. Capsaicin treatment induces selective sensory degeneration and increased sympathetic innervation in the rat ureter. Neuroscience 1995;67:953-66. Craft RM, Cohen SM, Porreca F. Long-lasting desensitization of bladder afferents following intravesical resiniferatoxin and capsaicin in the rat. Pain 1995;61: 317-23. Wilcox GL. Excitatory neurotransmitters and pain. In: Bond MR, Charlton JE, Woolf CJ, editors. Proceedings of the sixteenth world congress on pain. Amsterdam: Elsevier; 1991. p. 97-117. Johnson JW, Ascher P. Glycine potentiates the NMDA response in cultured mouse brain neurons. Nature 1987;325:529-31. Murase K, Ryu PD, Randic M. Excitatory and inhibitory amino acids and peptide-induced responses in acutely isolated rat spinal dorsal horn neurons. Neurosci Lett 1989;103:56-63. Dougherty PM, Willis WD. Enhancement of spinothalamic neuron responses to chemical and mechanical stimuli following combined micro-iontophoretic application of N-methyl-D-aspartic acid and substance P. Pain 1991; 47:85-93. Hope PJ, Schaible HG, Jarrott B, Duggan AW. Release and persistence of immunoreactive neurokinin A in the spinal cord is associated with chemical arthritis. Pain 1990;5 Suppl:S230. Duggan AW, Hall JG, Headley PM. Suppression of transmission of nociceptive impulses by morphine: selective effects of morphine administered in the region of the substantia gelatinosa. Br J Pharmacol 1977;61:65-76.
55. Le Greves P, Nyberg F, Ho¨kfelt T, Terenius L. Calcitonin gene-related peptide is metabolized by an endopeptidase hydrolyzing substance P. Regul Pept 1989;25:277-86. 56. Mao J, Coghill RC, Kellstein DE, et al. Calcitonin generelated peptide enhances substance P–induced behaviors via metabolic inhibition: in vivo evidence for a new mechanism of neuromodulation. Brain Res 1992;574: 157-63. 57. Jessell TM, Iversen LL. Opiate analgesics inhibit substance P release from rat trigeminal nucleus. Nature 1977;268:549-51. 58. Mudge AW, Leeman SE, Fischbach GD. Enkephalin inhibits release of substance P from sensory neurons in culture and decreases action potential duration. Proc Natl Acad Sci USA 1979;76:526-30. 59. Fields HL, Emson PC, Leigh BK, et al. Multiple opiate receptor sites on primary afferent fibres. Nature 1980;284:351-53. 60. Liu H, Wang H, Sheng M, et al. Evidence for presynaptic N-methyl-D-aspartate autoreceptors in the spinal cord dorsal horn. Proc Natl Acad Sci USA 1994;91:8383-7. 61. Kia HK, Miguel MC, McKernan RM, et al. Localization of 5-HT3 receptors in the rat spinal cord: Immunohistochemistry and in situ hybridization. Neuroreport 1995;6: 257-61. 62. Dado RJ, Law PY, Loh HH, Elde R. Immunofluorescent identification of a d-opioid receptor on primary afferent nerve terminals. Neuroreport 1993;5:341-4. 63. Arvidsson U, Riedl M, Chakrabarti S, et al. Distribution and targeting of a mu-opioid receptor (MOR1) in brain and spinal cord. J Neurosci 1995;15:3328-41. 64. Minami M, Maekawa K, Yabuuchi K, Satoh M. Double in situ hybridization study on coexistence of mu-, deltaand kappa-opioid receptor mRNAs with preprotachykinin A mRNA in the rat dorsal root ganglia. Brain Res 1995;30:203-10. 65. Brown JL, Liu H, Maggio JE, et al. Morphological characterization of substance P receptor-immunoreactive neurons in the rat spinal cord and trigeminal nucleus caudalis. J Comp Neurol 1995;356:327-44. 66. Zhang X, Wiesenfeld-Hallin Z, Ho¨kfelt T. Effect of peripheral axotomy on expression of neuropeptide y receptor mRNA in rat lumbar dorsal root ganglia. Eur J Neurosci 1994;6:43-57. 67. Yonehara N, Imai Y, Chen JQ, et al. Influence of opioids on substance P release evoked by antidromic stimulation of primary afferent fibers in the hind instep of rats. Regul Pept 1992;38:13-22. 68. Richardson BP, Engel G. The pharmacology and function of 5-HT3 receptors. Trends Pharmacol Sci 1986;7:424-8. 69. Carlton SM, Hargett GL, Coggeshall RE. Localization and activation of glutamate receptors in unmyelinated axons of rat glabrous skin. Neurosci Lett 1995;197:25-28. 70. Jackson DL, Graff CB, Richardson JD, Hargreaves KM. Glutamate participates in the peripheral modulation
Anatomy and physiology of cancer pain
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
of thermal hyperalgesia in rats. Eur J Pharmacol 1995;284:321-5. Zhou S, Bonasera L, Carlton SM. Peripheral administration of NMDA, AMPA or KA results in pain behaviors in rats. Neuroreport 1996;7:895-900. Lawand NB, Willis WD, Westlund KN. Excitatory amino acid receptor involvement in peripharal nociceptive transmission in rats. Eur J Pharm 1997;324:169-177. Castro-Lopes JM, Tavares I, Coimbra A. GABA decreases in the spinal cord dorsal horn after peripheral neurectomy. Brain Res 1993;620:287-91. Castro-Lopes JM, Tavares I, To¨lle TR, et al. Increase in GABAergic cells and GABA levels in the spinal-cord in unilateral inflammation of the hindlimb in the rat. Eur J Neurosci 1992;4:296-301. Sluka KA, Willis WD, Westlund KN. Joint inflammation and hyperalgesia are reduced by spinal bicuculline. Neuroreport 1993;5:109-12. Sindou M, Quoex C, Baleydier C. Fiber organization at the posterior spinal cord-rootlet junction in man. J Comp Neurol 1974;153:15-26. Light AR, Metz CB. The morphology of the spinal cord efferent and afferent neurons contributing to the ventral roots of the cat. J Comp Neurol 1978;179:501-16. Coggeshall RE, Coulter JD, Willis WD. Unmyelinated axons in the ventral roots of the cat lumbosacral enlargement. J Comp Neurol 1974;153:39-58. Maynard CW, Leonard RB, Coulter JD, Coggeshall RE. Central connections of ventral root afferents as demonstrated by the HRP method. J Comp Neurol 1977;172:601-8. Yamamoto T, Takahashi K, Satomi H, Ise H. Origins of primary afferent fibers in the spinal ventral roots in the cat as demonstrated by the horseradish peroxidase method. Brain Res 1977;126:350-4. Mawe GM, Bresnahan JC, Beattie MS. Primary afferent projections from dorsal and ventral roots to autonomic preganglionic neurons in the cat sacral spinal cord: light and electron microscopic observations. Brain Res 1984;290:152-7. Risling M, Hildebrand C. Occurrence of unmyelinated axon profiles at distal, middle and proximal levels in the ventral root L7 of cats and kittens. J Neurol Sci 1982;56:219-31. Chung K, Schwen RJ, Coggeshall RE. Ureteral axon damage following subcutaneous administration of capsaicin in adult rats. Neurosci Lett 1985;53:221-6. Shin HK, Kim J, Chung JM. Flexion reflex elicited by ventral root afferents in the cat. Neurosci Lett 1985;62:333-58. Shin HK, Kim J, Nam SC, et al. Spinal entry route for ventral root afferent fibers in the cat. Exp Neurol 1986;94:714-25. Kim J, Shin HK, Nam SC, Chung JM. Proportion and location of spinal neurons receiving ventral root afferent inputs in the cat. Exp Neurol 1988;99:296-314.
118
87. Oh UT, Kim KJ, Baik-Han EJ, Chung JM. Electrophysiological evidence for an increase in the number of ventral root afferent fibers after neonatal peripheral neurectomy in the rat. Brain Res 1989;501:90-99. 88. Nam SC, Kim KJ, Leem JW, et al. Fiber counts at multiple sites along the rat ventral root after neonatal peripheral neurectomy or dorsal rhizotomy. J Comp Neurol 1989;290:336-42. 89. Nam SC, Kim KJ, Leem JW, et al. Increased number of unmyelinated fibers in the ventral root after neurectomy in adult rat. Neurosci Lett 1990;116:40-4. 90. Park MJ, Chung K, Chung JM. Immunocytochemical evidence for sprouting of ventral root afferents after neonatal sciatic neurectomy in the rat. Neurosci Lett 1994;165:125-8. 91. Chung BS, Sheen K, Chung JM. Evidence for invasion of regenerated ventral root afferents into the spinal cord of the rat subjected to sciatic neurectomy during the neonatal period. Brain Res 1991;552:311-19. 92. Coggeshall RE. Law of separation of function of the spinal roots. Physiol Rev 1980;60:716-55. 93. Chung K, Langford LA, Applebaum AE, Coggeshall RE. Primary afferent fibers in the tract of Lissauer in the rat. J Comp Neurol 1979;184:587-98. 94. Coggeshall RE, Chung K, Chung JM, Langford LA. Primary afferent axons in the tract of Lissauer in the monkey. J Comp Neurol 1981;196:431-42. 95. Sugiura Y, Lee CL, Perl ER. Central projections of identified, unmyelinated (C) afferent fibers innervating mammalian skin. Science 1986;234:358-61. 96. Sugiura Y, Terui N, Hosoya Y. Difference in distribution of central terminals between visceral and somatic unmyelinated (C) primary afferent fibers. J Neurophysiol 1989;62:834-40. 97. Rexed B. The cytoarchitectonic organization of the spinal cord in the cat. J Comp Neurol 1952;96:415-95. 98. Honda CN. Visceral and somatic afferent convergence onto neurons near the central canal in the sacral spinal cord of the cat. J Neurophysiol 1985;53:1059-78. 99. Light AR, Perl ER. Spinal termination of functionally identified primary afferent neurons with slowly conducting myelinated fibers. J Comp Neurol 1979;186:133-50. 100. Craig AD, Mense S. The distribution of afferent fibers from the gastrocnemius-soleus muscle in the dorsal horn of the cat, as revealed by the transport of horseradish peroxidase. Neurosci Lett 1983;41:233-8. 101. Craig AD, Heppelmann B, Schaible HG. Projection of the medial and posterior articular nerves of the cat’s knee to the spinal cord. J Comp Neurol 1988;276:279-88. 102. Mantyh PW, DeMaster E, Malhotra A, et al. Receptor endocytosis and dendrite reshaping in spinal neurons after somatosensory stimulation. Science 1995;268:1629-32. 103. Mendell LM. Physiological properties of unmyelinated fiber projection to the spinal cord. Exp Neurol 1966;16: 316-32.
2009
2010
118
Anatomy and physiology of cancer pain
104. Nagy I, Woolf CJ. Lignocaine selectively reduces C fibre-evoked neuronal activity in rat spinal cord in vitro by decreasing N-methyl-D-aspartate and neurokinin receptor-mediated post-synaptic depolarizations: implications for the development of novel centrally acting analgesics. Pain 1996;64:59-70. 105. Sivilotti LG, Thompson SWN, Woolf CJ. The rate of rise of the cumulative depolarization evoked by repetitive stimulation of small-calibre afferents is a predictor of action potential windup in rat spinal neurones in vitro. J Neurophysiol 1993;69:1621-31. 106. Thompson SWN, King AE, Woolf CJ. Activity-dependent changes in rat ventral horn neurones in vitro: summation of prolonged afferent evoked postsynaptic depolarizations produce a D-APV sensitive windup. Eur J Neurosci 1990;2:638-49. 107. Thompson SWN, Gerber G, Sivilloti LG, Woolf CJ. Long duration ventral root potentials in the neonatal rat spinal cord in vitro: the effects of ionotropic and metabotropic excitatory amino acid receptor antagonists. Brain Res 1992;595:87-97. 108. Nagy I, Maggi CA, Dray A, et al. The role of neurokinin and N-methyl-D-aspartate receptors in synaptic transmission from capsaicin-sensitive primary afferents in the rat spinal cord in vitro. Neuroscience 1993;52: 1029-37. 109. Nagy I, Miller BA, Woolf CJ. NK1. and NK2 receptors contribute to the C-fibre evoked slow potentials in the rat spinal cord. Neuroreport 1994;5:2105-8. 110. Thompson SWN, Urban L, Dray A. Contribution of NK1 and NK2 receptor activation to high threshold afferent fibre evoked ventral root responses in the rat spinal cord in vitro. Brain Res 1993;625:100-8. 111. Davies SN, Lodge D. Evidence for involvement of N-methylaspartate receptors in ‘‘wind-up’’ of class 2 neurones in the dorsal horn of the rat. Brain Res 1987;424: 402-6. 112. Dickenson AH, Sullivan A. Evidence for a role of the NMDA receptor in the frequency dependent potentiation of deep rat dorsal horn nociceptive neurons following C-fibre stimulation. Neuropharmacology 1987;26:1235-8. 113. Woolf CJ, Thompson SWN. The induction and maintenance of central sensitization is dependent upon N-methyl-D-aspartic acid receptor activation: implications for the treatment of post-injury pain hypersensitivity states. Pain 1991;44:293-9. 114. Coderre TJ, Katz J, Vaccarino AL, Melzack R. Contribution of central neuroplasticity to pathological pain: review of clinical and experimental evidence. Pain 1993;52:259-85. 115. Urban L, Thompson SWN, Dray A. Modulation of spinal excitability: cooperation between neurokinin and excitatory amino acid neurotransmitters. Trends Neurosci 1994;17:432-8. 116. Randic M, Hecimoviec H, Ryu PD. Substance P modulates glutamate-induced currents in acutely isolated rat
117.
118. 119.
120.
121.
122. 123.
124.
125.
126.
127.
128.
129.
130.
131.
132.
spinal cord dorsal horn neurones. Neurosci Lett 1990; 117:74-80. Coderre TJ, Yashpal K. Intracellular messengers contributing to persistent nociception and hyperalgesia induced by L-glutamate and substance P in the rat formalin pain model. Eur J Neurosci 1994;6:1328-34. Woolf CJ. Evidence for a central component of postinjury pain hypersensitivity. Nature 1983;306:686-8. LaMotte RH, Shain CN, Simone DA, Tsai EFP. Neurogenic hyperalgesia: psychophysical studies of underlying mechanisms. J Neurophysiol 1991;66:190-211. Torebjo¨rk HE, Lundberg LER, LaMotte RH. Central changes in processing of mechanoreceptive input in capsaicin-induced secondary hyperalgesia in humans. J Physiol 1992;448:765-80. Kenshalo DR, Leonard RB, Chung JM, Willis WD. Facilitation of the responses of primate spinothalamic cells to cold and to tactile stimuli by noxious heating of the skin. Pain 1982;12:141-52. Willis WD. Hyperalgesia and allodynia. New York: Raven Press; 1992. Koltzenburg M, Torebjo¨rk HE, Wahren LK. Nociceptor modulated central sensitization causes mechanical hyperalgesia in acute chemogenic and chronic neuropathic pain. Brain 1994;117:579-91. Willis WD. Mechanical allodynia: a role for sensitized nociceptive tract cells with convergent input from mechanoreceptors and nociceptors? Am Pain Soc J 1993;2:23-33. Woolf CJ, Shortland P, Coggeshall RE. Peripheral nerve injury triggers central sprouting of myelinated afferents. Nature 1993;355:75-78. Woolf CJ, Shortland P, Reynolds M, et al. Reorganization of central terminals of myelinated primary afferents in the rat dorsal horn following peripheral axotomy. J Comp Neurol 1995;360:121-34. Iadarola MJ, Brady LS, Draisci G, Dubner R. Enhancement of dynorphin gene expression in spinal cord following experimental inflammation: stimulus specificity, behavioral parameters and opioid receptor binding. Pain 1988;35:313-26. Noguchi K, Ruda MA. Gene regulation in an ascending nociceptive pathway: inflammation-induced increase in preprotachykinin mRNA in rat lamina I spinal projection neurons. J Neurosci 1992;12:2563-72. Vizzard MA, Erdman SL, De Groat WC. Increased expression of neuronal nitric oxide synthase (NOS) in visceral neurons after nerve injury. J Neurosci 1995;15:4033-45. Abbadie C, Brown JL, Mantyh PM, Basbaum AI. Spinal cord substance P receptor immunoreactivity increases in both inflammatory and nerve injury models of persistent pain. Neuroscience 1996;70:201-9. Hunt SP, Pini A, Evan G. Induction of c-fos-like protein in spinal cord neurons following sensory stimulation. Nature 1987;328:632-4. Mene´trey D, Gannon A, Levine JD, Basbaum AI. Expression of c-fos protein in interneurons and projection
Anatomy and physiology of cancer pain
133.
134.
135. 136.
137.
138.
139.
140.
141.
142.
143.
144. 145. 146.
147.
148.
neurons of the rat spinal cord in response to noxious somatic, articular, and visceral stimulation. J Comp Neurol 1989;285:177-95. Herdegen T, To¨lle TR, Bravo R, et al. Sequential expression of JUN B, JUN D and FOS B proteins in rat spinal neurons: cascade of transcriptional operations during nociception. Neurosci Lett 1991;129:221-4. Abbadie C, Besson JM. c-fos. expression in rat lumbar spinal cord during the development of adjuvant-induced arthritis. Neuroscience 1992;48:985-93. de Groat WC, Neuropepdites in pelvic afferents. Experientia 1987;43:801-813. Willis WD, Sluka KA, Rees H, Westlund KN. A contribution of dorsal root reflexes to peripheral inflammation. In: Mendell L, Rudomin P & Romo R, eds. Presynaptic inhibition and neural control. New York: Oxford University Press; 1996. (in press). Sluka KA, Rees H, Westlund KN, Willis WD. Fiber types contributing to dorsal root reflexes induced by joint inflammation in cats and monkeys. J Neurophysiol 1995;74:981-9. Sluka KA, Westlund KN. An experimental arthritis in rats: dorsal horn aspartate and glutamate increases. Neurosci Lett 1992;145:141-4. Sorkin LS, Westlund KN, Sluka KA, et al. Neural changes in acute arthritis in monkeys: IV. Time-course of amino acid release into the lumbar dorsal horn. Brain Res Rev 1992;17:39-50. Dougherty PM, Palecek J, Paleckova V, et al. The role of NMDA and non-NMDA excitatory amino acid receptors in the excitation of primate spinothalamic tract neurons by mechanical, chemical, thermal and electrical stimuli. J Neurosci 1992;12:3025-41. Gobel S. Neural circuitry in the substantia gelatinosa of Rolando: anatomical insights. Adv Pain Res Ther 1979;3:175-95. Carlton SM, Hayes ES. Light microscopic and ultrastructural analysis of GABA-immunoreactive profiles in the monkey spinal cord. J Comp Neurol 1990;300: 162-82. Barron DH, Matthews BHC. The interpretation of potential changes in the spinal cord. J Physiol 1938;92: 276-321. Toennies JF. Reflex discharge from the spinal cord over the dorsal roots. J Neurophysiol 1938;1:378-90. Eccles JC, Kozak W, Magni F. Dorsal root reflexes of muscle group I afferent fibres. J Physiol 1961;159:128-46. Eccles JC, Kostyuk PG, Schmidt RF. Central pathways responsible for depolarization of primary afferent fibres. J Physiol 1962;161:237-57. Eccles JC, Kostyuk PG, Schmidt RF. Presynaptic inhibition of the central actions of flexor reflex afferents. J Physiol 1962;161:258-81. Eccles JC, Magni F, Willis WD. Depolarization of central terminals of group I afferent fibres from muscle. J Physiol 1962;160:62-93.
118
149. Rees H, Sluka KA, Westlund KN, Willis WD. Do dorsal root reflexes augment peripheral inflammation? Neuroreport 1994;5:821-4. 150. Rees H, Sluka KA, Westlund KN, Willis WD. The role of glutamate and GABA receptors in the generation of dorsal root reflexes by acute arthritis in the anaesthetized rat. J Physiol 1995;484:437-45. 151. Sluka KA, Westlund KN. Centrally administered nonNMDA but not NMDA receptor antagonists block peripheral knee joint inflammation. Pain 1993;55:217-25. 152. Sluka KA, Westlund KN. Behavioral and immunohistochemical changes in an experimental arthritis model in rats. Pain 1993;55:367-77. 153. Sluka KA, Jordan HH, Westlund KN. Reduction in joint swelling and hyperalgesia following post treatment with a non-NMDA glutamate receptor antagonist. Pain 1994;59:95-100. 154. Sluka KA, Willis WD, Westlund KN. Inflammationinduced release of excitatory amino acids is prevented by spinal administration of a GABAA and not by a GABAB receptor antagonist in rats. J Pharmacol Exp Ther 1994;271:76-82. 155. Sluka KA, Willis WD, Westlund KN. The role of dorsal root reflexes in neurogenic inflammation. Pain Forum 1995;4:141-9. 156. Rees H, Tsuruoka M, Al-Chaer ED, Willis WD. Excitation and inhibition of STT cells by stimulation of the pretectal region in the anaesthetized primate. J Physiol 1994;476:47P. 157. Rees H, Terenzi MG, Roberts MHT. Anterior pretectal nucleus facilitation of superficial dorsal horn neurones and modulation of deafferentation pain in the rat. J Physiol 1995;489:159-69. 158. White JC, Sweet WH. Pain and the neurosurgeon: a forty-year experience. Springfield, IL: Thomas; 1969. 159. Gybels JM, Sweet WH. Neurosurgical treatment of persistent pain: physiological and pathological mechanisms of human pain. In: Gildenberg PL, editor. Pain and headache. Basel: Karger; 1989. 160. Willis WD, Kenshalo DR, Leonard RB. The cells of origin of the primate spinothalamic tract. J Comp Neurol 1979;188:543-74. 161. Apkarian AV, Hodge CJ. Primate spinothalamic pathways: I. A quantitative study of the cells of origin of the spinothalamic pathway. J Comp Neurol 1989;288: 447-73. 162. Apkarian AV, Hodge CJ. Primate spinothalamic pathways: II. The cells of origin of the dorsolateral and ventral spinothalamic pathways. J Comp Neurol 1989;288:474-92. 163. Westlund KN, Carlton SM, Zhang D, Willis WD. Glutamate-immunoreactive terminals synapse on primate spinothalamic tract cells. J Comp Neurol 1992;322:519-27. 164. Surmeier DJ, Honda CN, Willis WD. Natural groupings of primate spinothalamic neurons based on cutaneous
2011
2012
118 165.
166.
167.
168.
169.
170.
171.
172.
173.
174.
175.
176.
177.
178.
179.
Anatomy and physiology of cancer pain
stimulation: physiological and anatomical features. J Neurophysiol 1988;59:833-60. Carlton SM, LaMotte CC, Honda CN, et al. Ultrastructural analysis of axosomatic contacts on functionally identified primate spinothalamic tract neurons. J Comp Neurol 1989;281:555-66. Carlton SM, Westlund KN, Zhang D, et al. Calcitonin gene-related peptide containing primary afferent fibers synapse on primate spinothalamic tract cells. Neurosci Lett 1989;190:76-81. Carlton SM, Westlund KN, Zhang D, Willis WD. GABA-immunoreactive terminals synapse on primate spinothalamic tract cells. J Comp Neurol 1992;322: 528-37. Westlund KN, Carlton SM, Zhang D, Willis WD. Direct catecholaminergic innervation of primate spinothalamic tract neurons. J Comp Neurol 1990;299:178-86. Carlton SM, LaMotte CC, Honda CN, et al. Ultrastructural analysis of substance P and other synaptic profiles innervating an identified primate spinothalamic tract neuron. Neurosci Abstr 1985;11:578. LaMotte CC, Carlton SM, Honda CN, et al. Innervation of identified primate spinothalamic tract neurons: ultrastructure of serotonergic and other synaptic profiles. Neurosci Abstr 1988;14:852. Craig AD. Spinal distribution of ascending lamina I axons anterogradely labeled with Phaseolus vulgaris. leucoagglutinin (PHA-L) in the cat. J Comp Neurol 1991;313:377-93. Boivie J. An anatomical reinvestigation of the termination of the spinothalamic tract in the monkey. J Comp Neurol 1979;186:343-70. Berkley KJ. Spatial relationships between the terminations of somatic sensory and motor pathways in the rostral brainstem in cats and monkeys: I. Ascending somatic sensory inputs to lateral diencephalon. J Comp Neurol 1980;193:283-317. Craig AD, Burton H. Spinal and medullary lamina I projection to nucleus submedius in medial thalamus: a possible pain center. J Neurophysiol 1981;45:443-66. Mantyh PW. The spinothalamic tract in the primate: a re-examination using wheatgerm agglutinin conjugated to horseradish peroxidase. Neuroscience 1983;9:847-62. Apkarian AV, Hodge CJ. Primate spinothalamic pathways: III. Thalamic terminations of the dorsolateral and ventral spinothalamic pathways. J Comp Neurol 1989;288:493-511. Craig AD. Distribution of brainstem projections from spinal lamina I neurons in the cat and monkey. J Comp Neurol 1995;361:225-48. Foreman RD, Schmidt RF, Willis WD. Effects of mechanical and chemical stimulation of fine muscle afferents upon primate spinothalamic tract cells. J Physiol 1979;286:215-31. Dougherty PM, Sluka KA, Sorkin LS, et al. Neural changes in acute arthritis in monkeys: I. Parallel
180.
181.
182.
183.
184.
185.
186.
187.
188.
189.
190.
191.
192.
enhancement of responses of spinothalamic tract neurons to mechanical stimulation and excitatory amino acids. Brain Res Rev 1992;17:1-13. Hobbs SF, Chandler MJ, Bolser DC, Foreman RD. Segmental organization of visceral and somatic input onto C3-T6 spinothalamic tract cells of the monkey. J Neurophysiol 1992;68:1575-88. Blair RW, Ammons RN, Foreman RD. Responses of thoracic spinothalamic and spinoreticular cells to coronary artery occlusion. J Neurophysiol 1984;51:636-48. Milne RJ, Foreman RD, Giesler GJ, Willis WD. Convergence of cutaneous and pelvic visceral nociceptive inputs onto primate spinothalamic neurons. Pain 1981;11:163-83. Brennan TJ, Oh UT, Hobbs SF, et al. Urinary bladder and hindlimb afferent input inhibits activity of primate T2-T5 spinothalamic tract neurons. J Neurophysiol 1989;61:573-88. Hobbs SF, Oh UT, Chandler MJ, et al. Evidence that C1 and C2 propriospinal neurons mediate the inhibitory effects of viscerosomatic spinal afferent input on primate spinothalamic tract neurons. J Neurophysiol 1992;67: 852-60. Owens CM, Zhang D, Willis WD. Changes in the response states of primate spinothalamic tract cells caused by mechanical damage of the skin or activation of descending controls. J Neurophysiol 1992;67:1509-27. Kenshalo DR, Leonard RB, Chung JM, Willis WD. Responses of primate spinothalamic neurons to graded and to repeated noxious heat stimuli. J Neurophysiol 1979;42:1370-89. Dougherty PM, Sluka KA, Sorkin LS, et al. Neural changes in a arthritis in monkeys: I. Parallel enhancement of responses of spinothalamic tract neurons to mechanical stimulation and excitatory amino acids. Brain Res Rev 1992;17:1-13. Simone DA, Sorkin LS, Oh U, et al. Neurogenic hyperalgesia: central neural correlates in responses of spinothalamic tract neurons. J Neurophysiol 1991;66:228-46. Dougherty PM, Willis WD. Enhanced responses of spinothalamic tract neurons to excitatory amino acids accompany capsaicin-induced sensitization in the monkey. J Neurosci 1992;12:883-94. Dougherty PM, Palecek J, Zorn S, Willis WD. Combined application of excitatory amino acids and substance P produces long-lasting changes in responses of primate spinothalamic tract neurons. Brain Res Rev 1993;18:227-46. Dougherty PM, Palecek J, Paleckova V, Willis WD. Neurokinin 1 and 2 antagonists attenuate the responses and NK1 antagonists prevent the sensitization of primate spinothalamic tract neurons after intradermal capsaicin. J Neurophysiol 1994;72:1464-75. Palecek J, Paleckova V, Dougherty PM, Willis WD. The effect of phorbol esters on the responses of primate spinothalamic neurons to mechanical and thermal stimuli. J Neurophysiol 1994;71:529-37.
Anatomy and physiology of cancer pain
193. Lin Q, Peng YB, Willis WD. Possible role of protein kinase C in the sensitization of primate spinothalamic tract neurons. J Neurosci 1996;16:3026-34. 194. Giesler GJ, Yezierski RP, Gerhart KD, Willis WD. Spinothalamic tract neurons that project to medial and/or lateral thalamic nuclei: evidence for a physiologically novel population of spinal cord neurons. J Neurophysiol 1981;46:1285-308. 195. Ammons WS, Girardot MN, Foreman RD. T2-T5 spinothalamic neurons projecting to medial thalamus with viscerosomatic input. J Neurophysiol 1985;54:73-89. 196. Willis WD, Trevino DL, Coulter JD, Maunz RA. Responses of primate spinothalamic tract neurons to natural stimulation of hindlimb. J Neurophysiol 1974;37: 358-72. 197. Craig AD, Kniffki KD. Spinothalamic lumbosacral lamina I cells responsive to skin and muscle stimulation in the cat. J Physiol 1985;365:197-221. 198. Ferrington DG, Sorkin LS, Willis WD. Responses of spinothalamic tract cells in the superficial dorsal horn of the primate lumbar spinal cord. J Physiol 1987;388: 681-703. 199. Craig AD, Hunsley SJ. Morphine enhances the activity of thermoreceptive cold-specific lamina I spinothalamic neurons in the cat. Brain Res 1991;558:93-7. 200. Craig AD, Serrano LP. Effects of systemic morphine on lamina I spinothalamic tract neurons in the cat. Brain Res 1994;636:233-44. 201. Lima D, Coimbra A. A Golgi study of the neuronal population of the marginal zone (lamina I) of the rat spinal cord. J Comp Neurol 1986;244:53-71. 202. Lima D, Coimbra A. The spinothalamic system of the rat: structural types of retrogradely labeled neurons in the marginal zone (lamina I). Neuroscience 1988;27: 215-30. 203. Zhang EF, Craig AD. Morphological classes of retrogradely labeled lamina I spinothalamic neurons in the monkey. Neurosci Abstr 1995;21:645. 204. Han ZS, Craig AD. Morphological characteristics of physiologically identified lamina I cells in cats. Neurosci Abstr 1994;20:547. 205. Lekan HA, Carlton SM. Glutamatergic and GABAergic input to rat spinothalamic tract cells in the superficial dorsal horn. J Comp Neurol 1995;361:417-28. 206. Willis WD. Neural mechanisms of pain discrimination. In: Lund JS, editor. Sensory processing in the mammalian brain. New York: Oxford University Press; 1989. p. 130-43. 207. Surmeier DJ, Honda CN, Willis WD. Responses of primate spinothalamic neurons to noxious thermal stimulation of glabrous and hairy skin. J Neurophysiol 1986;56:328-50. 208. Surmeier DJ, Honda CN, Willis WD. Temporal features of the responses of primate spinothalamic neurons to noxious thermal stimulation of hairy and glabrous skin. J Neurophysiol 1986;56:351-68.
118
209. McMahon S, Wall PD. Descending excitation and inhibition of spinal cord lamina I projection neurons. J Neurophysiol 1988;59:1204-19. 210. Burstein R, Cliffer KD, Giesler GJ. Cells of origin of the spinohypothalamic tract in the rat. J Comp Neurol 1990;291:329-44. 211. Burstein R, Dado RJ, Giesler GJ. The cells of origin of the spinothalamic tract of the rat: A quantitative reexamination. Brain Res 1990;511:329-37. 212. Carstens E, Trevino DL. Anatomical and physiological properties of ipsilaterally projecting spinothalamic neurons in the second cervical segment of the cat’s spinal cord. J Comp Neurol 1978;182:167-84. 213. Smith MV, Apkarian AV, Hodge CJ. Somatosensory response properties of contralaterally projecting spinothalamic and nonspinothalamic neurons in the second cervical segment of the cat. J Neurophysiol 1991;66: 83-102. 214. Kevetter GA, Haber LH, Yezierski RP, et al. Cells of origin of the spinoreticular tract in the monkey. J Comp Neurol 1982;207:61-74. 215. Haber LH, Moore BD, Willis WD. Electrophysiological response properties of spinoreticular neurons in the monkey. J Comp Neurol 1982;207:75-84. 216. Kinomura S, Larsson J, Gulya´s B, Roland PE. Activation by attention of the human reticular formation and thalamic intralaminar nuclei. Science 1996;271:512-15. 217. Hylden JLK, Hayashi H, Bennett GJ, Dubner R. Spinal lamina I neurons projecting to the parabrachial area of the cat midbrain. Brain Res 1985;336:195-8. 218. Wiberg M, Westman J, Blomqvist A. Somatosensory projection to the mesencephalon: an anatomical study in the monkey. J Comp Neurol 1987;264:92-117. 219. Yezierski RP. Spinomesencephalic tract: projections from the lumbosacral spinal cord of the rat, cat, and monkey. J Comp Neurol 1988;267:131-46. 220. Yezierski RP, Broton JG. Functional properties of spinomesencephalic tract (SMT) cells in the upper cervical spinal cord of the cat. Pain 1991;45:187-96. 221. Lima D, Coimbra A. Morphological types of spinomesencephalic neurons in the marginal zone (lamina I) of the rat spinal cord, as shown after retrograde labeling with cholera toxin subunit B. J Comp Neurol 1989;279: 327-39. 222. Bernard JF, Besson JM. The spino(trigemino)-pontoamygdaloid pathway: electrophysiological evidence for an involvement in pain processes. J Neurophysiol 1990;63: 473-90. 223. Slugg RM, Light AR. Spinal cord and trigeminal projections to the pontine parabrachial region in the rat as demonstrated with Phaseolus vulgaris. leucoagglutinin. J Comp Neurol 1994;339:49-61. 224. Newman HM, Stevens RT, Apkarian AV. Direct spinal projections to limbic and striatal areas: anterograde transport studies from the upper cervical spinal cord
2013
2014
118 225.
226.
227.
228.
229.
230.
231.
232.
233.
234.
235.
236.
237.
238.
239.
Anatomy and physiology of cancer pain
and the cervical enlargement in squirrel monkey and rat. J Comp Neurol 1996;365:640-58. Yezierski RP, Sorkin LS, Willis WD. Response properties of spinal neurons projecting to midbrain or midbrainthalamus in the monkey. Brain Res 1987;437:165-70. Dado RJ, Katter JT, Giesler GJ. Spinothalamic and spinohypothalamic tract neurons in the cervical enlargement of rats: II. Responses to innocuous and noxious mechanical and thermal stimuli. J Neurophysiol 1994;71:981-1002. Westlund KN, Craig AD. Spinal cord lamina I terminations in brainstem catecholamine cell groups. Exp Brain Res 1996;110:151-62. Burstein R, Cliffer KD, Giesler GJ. Direct somatosensory projections from the spinal cord to the hypothalamus and telencephalon. J Neurosci 1987;7:4159-64. Cliffer KD, Burstein R, Giesler GJ. Distributions of spinothalamic, spinohypothalamic, and spinotelencephalic fibers revealed by anterograde transport of PHA-L in rats. J Neurosci 1991;11:852-68. Katter JT, Burstein R, Giesler GJ. The cells of origin of the spinohypothalamic tract in cats. J Comp Neurol 1991;303:101-12. Dado RJ, Katter JT, Giesler GJ. Spinothalamic and spinohypothalamic tract neurons in the cervical enlargement of rats: III. Locations of antidromically identified axons in the cervical cord white matter. J Neurophysiol 1994;71:1003-21. Zhang X, Kostarczyk E, Giesler GJ. Spinohypothalamic tract neurons in the cervical enlargement of rats: descending axons in the ipsilateral brain. J Neurosci 1995;15:8393-407. Burstein R, Dado RJ, Cliffer KD, Giesler GJ. Physiological characterization of spinohypothalamic tract neurons in the lumbar enlargement of rats. J Neurophysiol 1991;66:261-84. Patterson JT, Coggeshall RE, Lee WT, Chung K. Long ascending unmyelinated primary afferent axons in the rat dorsal column: immunohistochemical localizations. Neurosci Lett 1990;108:6-10. Patterson JT, Head PA, McNeill DL, et al. Ascending unmyelinated primary afferent fibers in the dorsal funiculus. J Comp Neurol 1989;290:384-90. Conti F, De Biasi S, Giuffrida R, Rustioni A. Substance P–containing projections in the dorsal columns of rats and cats. Neuroscience 1990;34:607-21. Rustioni A, Kaufman AB. Identification of cells of origin of non-primary afferents to the dorsal column nuclei of the cat. Exp Brain Res 1977;27:1-14. Rustioni A, Hayes NL, O’Neill S. Dorsal column nuclei and ascending spinal afferents in macaques. Brain 1979;102:95-125. Brown AG, Fyffe REW. Form and function of dorsal horn neurones with axons ascending the dorsal columns in cat. J Physiol 1981;321:31-47.
240. Bennett GJ, Seltzer Z, Lu GW, et al. The cells of origin of the dorsal column postsynaptic projection in the lumbosacral enlargements of cats and monkeys. Somatosens Res 1983;1:131-49. 241. Giesler GJ, Nahin RL, Madsen AM. Postsynaptic dorsal column pathway of the rat: I. Anatomical studies. J Neurophysiol 1984;51:260-75. 242. De Pommery J, Roudier F, Mene´trey D. Postsynaptic fibers reaching the dorsal column nuclei in the rat. Neurosci Lett 1984;50:319-23. 243. Enevoldson TP, Gordon G. Postsynaptic dorsal column neurons in the cat: a study with retrograde transport of horseradish peroxidase. Exp Brain Res 1989;75:611-20. 244. Uddenberg N. Functional organization of long, secondorder afferents in the dorsal funiculus. Exp Brain Res 1968;4:377-82. 245. Angaut-Petit D. The dorsal column system: II. Functional properties and bulbar relay of the postsynaptic fibres of the cat’s fasciculus gracilis. Exp Brain Res 1975;22:471-93. 246. Brown AG, Brown PB, Fyffe REW, Pubols LM. Receptive field organization and response properties of spinal neurones with axons ascending the dorsal columns in the cat. J Physiol 1983;337:575-88. 247. Bennett GJ, Nishikawa N, Lu GW, et al. The morphology of dorsal column postsynaptic spinomedullary neurons in the cat. J Comp Neurol 1984;224:568-78. 248. Kamogawa H, Bennett GJ. Dorsal column postsynaptic neurons in the cat are excited by myelinated nociceptors. Brain Res 1986;364:386-90. 249. Noble R, Riddell JS. Cutaneous excitatory and inhibitory input to neurones of the postsynaptic dorsal column system in the cat. J Physiol 1988;396:497-513. 250. Giesler GJ, Cliffer KD. Postsynaptic dorsal column pathway of the rat: II. Evidence against an important role in nociception. Brain Res 1985;326:347-56. 251. Ferrington DG, Downie JW, Willis WD. Primate nucleus gracilis neurons: responses to innocuous and noxious stimuli. J Neurophysiol 1988;59:886-907. 252. Cliffer KD, Hasegawa T, Willis WD. Responses of neurons in the gracile nucleus of cats to innocuous and noxious stimuli: basic characterization and antidromic activation from the thalamus. J Neurophysiol 1992;68: 818-32. 253. Nathan PW, Smith MC, Cook AW. Sensory effects in man of lesions of the posterior columns and of some other afferent pathways. Brain 1986;109:1003-41. 254. Armour D. On the surgery of the spinal cord and its membranes. Lancet 1927;2:691-7. 255. Putnam TJ. Myelotomy of the commissure: a new method of treatment for pain in the upper extremities. Arch Neurol Psychiatr 1934;32:1189-93. 256. Leriche R. Du traitment de la douleur dans les cancers abdominaux et pelviens inope´rables ou re´cidive´s. Gaz Hoˆpitaux Paris 1936;109:917-22.
Anatomy and physiology of cancer pain
257. Mansuy L, Lecuire J, Acassat L. Technique de la mye´lotomie commissurale poste´rieure. J Chir (Paris) 1944;60: 206-13. 258. Mansuy L, Sindou M, Fischer G, Brunon J. La cortotomie spinothalamique dans les douleurs cance´reuses: Re´sultats d’une se´rie de 124 malades ope´re´s par abord direct poste´rieure. Neurochirurgie 1976;22: 437-44. 259. Wertheimer P, Lecuire J. La mye´lotomie commissurale ´ propos de 107 observations. Acta Chir poste´rieure: A Belg 1953;52:568-74. 260. Dargent M, Mansuy L, Colon J, De Rougement J. Les proble´mes pose´s par la douleur dans l’e´volution des cancer gyne´cologiques. Lyon Chirurg 1963;59:62-83. 261. Sourek K. Commissural myelotomy. J Neurosurg 1969;31:524-7. 262. Cook AW, Kawakami Y. Commissural myelotomy. J Neurosurg 1977;47:1-6. 263. King RB. Anterior commissurotomy for intractable pain. J Neurosurg 1977;47:7-11. 264. Cook AW, Nathan PW, Smith MC. Sensory consequences of commissural myelotomy: a challenge to traditional anatomical concepts. Brain 1984;107:547-68. 265. Hitchcock ER. Stereotactic cervical myelotomy. J Neurol Neurosurg Psychiatry 1970;33:224-30. 266. Hitchcock ER. Stereotactic myelotomy. Proc R Soc Med 1974;67:771-2. 267. Schvarcz JR. Stereotactic extralemniscal myelotomy. J Neurol Neurosurg Psychiatry 1976;39:53-7. 268. Gildenberg PL, Hirshberg RM. Limited myelotomy for the treatment of intractable cancer pain. J Neurol Neurosurg Psychiatry 1984;47:94-6. 269. Hirshberg RM, Al-Chaer ED, Lawand NB, et al. Is there a pathway in the posterior funiculus that signals visceral pain? Pain 1996; 67:291–305. 270. Al-Chaer ED, Lawand NB, Westlund KN, Willis WD. Visceral nociceptive input into the ventral posterolateral nucleus of the thalamus: a new function for the dorsal column pathway. J Neurophysiol 1996;76: 2661-74. 271. Al-Chaer ED, Lawand NB, Westlund KN, Willis WD. Pelvic visceral input into the nucleus gracilis is largely mediated by the postsynaptic dorsal column pathway. J Neurophysiol 1996;76:2675-90. 272. Aidar O, Geohegan WA, Ungewitter LH. Splanchnic afferent pathways in the central nervous system. J Neurophysiol 1952;15:131-8. 273. Amassian VE. Fiber groups and spinal pathways of cortically represented visceral afferents. J Neurophysiol 1951;14:445-60. 274. Rigamonti DD, Hancock MB. Analysis of field potentials elicited in the dorsal column nuclei by splanchnic nerve A-beta afferents. Brain Res 1974;77:326-9. 275. Rigamonti DD, Hancock MB. Viscerosomatic convergence in the dorsal column nuclei. Exp Neurol 1978;61: 337-48.
118
276. Berkley KJ, Hubscher CH. Are there separate central nervous system pathways for touch and pain? Nature Med 1995;1:766-73. 277. Berkley KJ, Guilbaud G, Benoist J, Gautron M. Responses of neurons in and near the thalamic ventrobasal complex of the rat to stimulation of uterus, cervix, vagina, colon, and skin. J Neurophysiol 1993;69:557-68. 278. Chandler MJ, Hobbs SF, Fu CG, et al. Responses of neurons in ventroposterolateral nucleus of primate thalamus to urinary bladder distension. Brain Res 1992;571: 26-34. 279. Bru¨ggemann J, Shi T, Apkarian AV. Squirrel monkey lateral thalamus: II. Viscerosomatic convergent representation of urinary bladder, colon and esophagus. J Neurosci 1994;14:6796-814. 280. Truex RC, Taylor MJ, Smythe MQ, Gildenberg PL. The lateral cervical nucleus of the cat, dog, and man. J Comp Neurol 1965;139:93-104. 281. Brown AG, Hamann WC, Martin HF. Effects of activity in nonmyelinated afferent fibres on the spinocervical tract. Brain Res 1975;98:243-59. 282. Cervero F, Iggo A, Molony V. Responses of spinocervical tract neurones to noxious stimulation of the skin. J Physiol 1977;267:537-58. 283. Downie JW, Ferrington DG, Sorkin LS, Willis WD. The primate spinocervicothalamic pathway: Responses of cells of the lateral cervical nucleus and spinocervical tract to innocuous and noxious stimuli. J Neurophysiol 1988;59: 861-85. 284. Gaze RM, Gordon G. The representation of cutaneous sense in the thalamus of the cat and monkey. Q J Exp Physiol 1954;39:279-304. 285. Perl ER, Whitlock DG. Somatic stimuli exciting spinothalamic projections to thalamic neurons in cat and monkey. Exp Neurol 1961;3:256-96. 286. Pollin B, Albe-Fessard D. Organization of somatic thalamus in monkeys with and without section of dorsal spinal tracts. Brain Res 1979;173:431-49. 287. Kenshalo DR, Giesler GJ, Leonard RB, Willis WD. Responses of neurons in primate ventral posterior lateral nucleus to noxious stimuli. J Neurophysiol 1980;43: 1594-614. 288. Chung JM, Lee KH, Surmeier DJ, et al. Response characteristics of neurons in the ventral posterior lateral nucleus of the monkey thalamus. J Neurophysiol 1986;56:370-90. 289. Chung JM, Surmeier DJ, Lee KH, et al. Classification of primate spinothalamic and somatosensory thalamic neurons based on cluster analysis. J Neurophysiol 1986;56:308-27. 290. Casey KL, Morrow TJ. Ventral posterior thalamic neurons differentially responsive to noxious stimulation of the awake monkey. Science 1983;221:675-7. 291. Apkarian AV, Shi T. Squirrel monkey lateral thalamus: I. Somatic nociresponsive neurons and their relation to spinothalamic terminals. J Neurosci 1994;14:6779-95.
2015
2016
118
Anatomy and physiology of cancer pain
292. Yokota T, Nishikawa Y, Koyama N. Distribution of trigeminal nociceptive neurons in nucleus ventralis posteromedialis of primates. In: Dubner R, Gebhart GF, Bond MD, editors. Pain research and clinical management, vol. 3. Amsterdam: Elsevier; 1988, p. 555-9. 293. Bushnell MC, Duncan GH, Tremblay N. Thalamic VPM nucleus in the behaving monkey: I. Multimodal and discriminative properties of thermosensitive neurons. J Neurophysiol 1993;69:739-52. 294. Duncan GH, Bushnell MC, Oliveras JL, et al. Thalamic VPM nucleus in the behaving monkey: III. Effects of reversible inactivation by lidocaine on thermal and mechanical discrimination. J Neurophysiol 1993;70:2086-96. 295. Lenz FA, Seike M, Lin YC, et al. Neurons in the area of human thalamic nucleus ventralis caudalis respond to painful heat stimuli. Brain Res 1993;623:235-40. 296. Lenz FA, Seike M, Richardson RT, et al. Thermal and pain sensations evoked by microstimulation in the area of human ventrocaudal nucleus. J Neurophysiol 1993; 70:200-12. 297. Lenz FA, Gracely RH, Hope EJ, et al. The sensation of angina can be evoked by stimulation of the human thalamus. Pain 1994;59:119-25. 298. Lenz FA, Gracely RH, Rowland LH, Dougherty PM. A population of cells in the human thalamic principal sensory nucleus respond to painful mechanical stimuli. Neurosci Lett 1994;180:46-50. 299. Lenz FA, Kwan HC, Martin R, et al. Characteristics of somatotopic organization and spontaneous neuronal activity in the region of the thalamic principal sensory nucleus in patients with spinal cord transection. J Neurophysiol 1994;72:1570-87. 300. Bullett E. Induction of c-fos-like protein within the lumbar spinal cord and thalamus of the rat following peripheral stimulation. Brain Res 1989;493:391-7. 301. Reyes-Vazquez C, Prieto-Gomez B, Dafny N. Noxious and non-noxious responses in the medial thalamus of the rat. Neurol Res 1989;11:177-80. 302. Bushnell MC, Duncan GH. Sensory and affective aspects of pain perception: is medial thalamus restricted to emotional issues? Exp Brain Res 1989;78: 415-18. 303. Dostrovsky JO, Broton JG, Warma NK. Functional properties of subnucleus caudalis lamina I neurons projecting to nucleus submedius. In: Schmidt RF, Schaible HG, Vahle-Hinz C, editors. Fine afferent nerve fibers and pain. Weinheim: VCH; 1987. p. 357-66. 304. Miletic V, Coffield JA. Responses of neurons in the rat nucleus submedius to noxious and innocuous mechanical cutaneous stimulation. Somatosens Mot Res 1989;6:567-87. 305. Roberts VJ, Dong WK. The effect of thalamic nucleus submedius lesions on nociceptive responding in rats. Pain 1994;57:341-9.
306. Zhang YQ, Tang JS, Yuan B, Jia H. Inhibitory effects of electrical stimulation of thalamic nucleus submedius area on the rat tail flick reflex. Brain Res 1995;696:205-12. 307. Craig AD, Bushnell MC, Zhang ET, Blomqvist A. A thalamic nucleus specific for pain and temperature sensation. Nature 1994;372:770-3. 308. Jones AKP, Brown WD, Friston KJ, et al. Cortical and subcortical localization of response to pain in man using positron emission tomography. Proc R Soc Lond [Biol] 1991;244:39-44. 309. Casey KL, Minoshima S, Berger KL, et al. Positron emission tomographic analysis of cerebral structures activated specifically by repetitive noxious heat stimuli. J Neurophysiol 1994;71:802-7. 310. Coghill RC, Talbot JD, Evans AC, et al. Distributed processing of pain and vibration by the human brain. J Neurosci 1994;14:4095-108. 311. DiPiero V, Jones AKP, Iannotti F, et al. Chronic pain: a PET study of the central effects of percutaneous high cervical cordotomy. Pain 1991;:46:9-12. 312. Iadarola MJ, Max MB, Berman KF, et al. Unilateral decrease in thalamic activity observed with positron emission tomography in patients with chronic neuropathic pain. Pain 1995;63:55-64. 313. Cesaro P, Mann MW, Moretti JL, et al. Central pain and thalamic hyperactivity: a single photon emission computerized tomographic study. Pain 1991;47:329-36. 314. Kenshalo DR, Chudler EH, Anton F, Dubner R. SI cortical nociceptive neurons participate in the encoding process by which monkeys perceive the intensity of noxious thermal stimulation. Brain 1988;454:378-82. 315. Dong WK, Salonen LD, Kawakami Y, et al. Nociceptive receptor of trigeminal neurons in SII-7b cortex of awake monkeys. Brain 1989;484:314-24. 316. Chudler EH, Anton F, Dubner R, Kenshalo DR. Responses of nociceptive SI neurons in monkeys and pain sensation in humans elicited by noxious thermal stimulation: effect of interstimulus interval. J Neurophysiol 1990;63:559-69. 317. Sikes RW, Vogt BA. Nociceptive neurons in area 24 of rabbit cingulate cortex. J Neurophysiol 1992;68:1720-32. 318. Talbot JD, Marrett S, Evans AC, et al. Multiple representations of pain in human cerebral cortex. Science 1991;251:1355-8. 319. Derbyshire SWG, Jones AKP, Devani P, et al. Cerebral responses to pain in patients with atypical facial pain measured by positron emission tomography. J Neurol Neurosurg Psychiatry 1994;57:1166-72. 320. Davis KD, Wood ML, Crawley AP, Mikulis DJ. fMRI of human somatosensory and cingulate cortex during painful electrical stimulation. Neuroreport 1995;7: 321-5. 321. Hsieh JC, Belfrage M, Stone-EIander S, et al. Central representation of chronic ongoing neuropathic
Anatomy and physiology of cancer pain
pain studied by positron emission tomography. Pain 1995;63:225-36. 322. Apkarian AV, Stea RA, Manglos SH, et al. Persistent pain inhibits contralateral somatosensory cortical activity in humans. Neurosci Lett 1992;140:141-7.
118
323. Canavero S, Pagni CA, Castellano G, et al. The role of cortex in central pain syndromes: preliminary results of a long-term techetium-99 hexamethypropyl-eneamineoxime single photon emission computed tomography study. Neurosurgery 1993;32:185-191.
2017
143 Balloon Compression for Trigeminal Neuralgia J. A. Brown . J. G. Pilitsis
Introduction In the 1950s, neurosurgeons frequently performed intra-operative manipulation of the trigeminal ganglion, maxillary and mandibular divisions to treat trigeminal neuralgia (TN) pain [1]. When neurosurgeons decompressed the mandibular division in the middle fossa, they learned that middle fossa injury to the nerve was more successful in relieving pain than when the division was decompressed [1]. Based on this concept, Sean Mullan, in 1983, pioneered the concept of percutaneous balloon compression for trigeminal neuralgia [2]. Since that first publication, results from more than 800 of these procedures have been reviewed in the literature [3]. In this chapter, we will review the indications for balloon compression, the surgical and radiographic technique used to perform it, postoperative management, results of treatment and potential complications of treatment.
Patient Selection The classical characteristics of trigeminal neuralgia (TN) are intermittent, short, one-sided, sharp, electric shock-like pains in one or more of the divisions mediated by the trigeminal nerve. (Burchiel type 1) [3,4] TN pain is often relieved medically with an anticonvulsant such as carbamazepine, gabapentin or diphenylhydantoin. If pain improves with one of these medications, one should be confident about the diagnosis.
#
Springer-Verlag Berlin/Heidelberg 2009
Once the diagnosis of TN is made, magnetic resonance imaging (MRI) of the trigeminal nerve is done using T1 weighted 1mm axial slices through the pons. This technique identifies any secondary cause of trigeminal neuralgia present, such as a trigeminal schwannoma or an arteriovenous malformation and may show a vascular association with the nerve by a vein or artery. Even if a secondary cause of TN is present, balloon compression is a surgical option. The selection of a surgical approach to TN depends on many factors. These factors include the patient’s general health, personal concerns regarding the morbidity of a posterior fossa operation, and the division in which the pain occurs. Age is not necessarily a factor, though older patients with many medical conditions may be better candidates for balloon compression than microvascular decompression. Contralateral jaw weakness is not a contraindication, despite the expectation of temporary masseter and pterygoid muscle weakness. Balloon compression has the advantage of avoiding injury to the small, unmyelinated fibers that mediate the corneal reflex. This provides relative protection to corneal sensation in patients with first division TN [5]. When intraluminal pressure is measured, balloon compression rarely causes anesthesia dolorosa or corneal keratitis. It is also done using general anesthesia. This limits the pain and anxiety that is often present during other percutaneous procedures done in awake patients [6]. The introducing cannula does not pass through the foramen ovale, which reduces the small, but significant, risks that are a result of passing a needle into the middle fossa.
2458
143
Balloon compression for trigeminal neuralgia
Pre-operative Preparation MRI is useful to identify any vascular associations. Vascular distortion of the trigeminal nerve may be a prognostic sign [7] (> Figure 143-1a and b). Vertebral-basilar ectasia, if present, may make balloon compression more risky because of the danger of vascular injury from rupture of the large vessel. Pre-operative clearance includes an electrocardiogram. The mean age of patients undergoing balloon compression is 65 years and the anesthesiologists require the screening study before inducing general anesthesia. Also, an electrocardiogram identifies patients with cardiac arrhythmias who may be at risk when the bradycardia caused by the trigeminal depressor response occurs during compression. Oral acyclovir can be prescribed several days before surgery to patients who have a history of recurrent cold sores, to try to alleviate a post-operative outbreak. There is, however, no clinical evidence to support its effectiveness,
Operative Procedure Use the angiography suite for the procedure if possible. High intensity, multi-plane imaging
makes placement of the cannula and catheter easier than if a portable imaging device is used in the operating room. Request that the anesthesiologist not use anticholinergic agents or muscle relaxants during induction. A light general anesthesia is induced often with propofol, then, an external pacemaker is positioned on the chest. The pacemaker is set to trigger immediately if the heart rate drops below 45 beats/min. Test the pacemaker to make sure the stimulus is being captured by heart muscle. The pacer blocks the trigeminal depressor response more rapidly than an intravenous atropine injection. If bradycardia persists during compression, give 0.4 mg atropine intravenously. The depressor response consists of both bradycardia and brief hypotension, often with a reflex hypertension after the pacemaker is triggered. Pre-operative atropine will inhibit this depressor response but it also prevents physiologic confirmation of nerve compression. Pre operative muscle relaxants block the facial muscular contraction that confirms placement of the cannula at the foramen ovale. The patient lies supine, a roll located under the shoulders providing 15 degrees of neck extension. Turn the head 15–30 degrees towards the opposite side. Older patients often have degenerative cervical spondylosis and further
. Figure 143-1 (a) Magnetic resonance imaging scan of the brain showing vascular compression of the right trigeminal nerve with secondary distortion at the nerve root entry zone. (b) Magnetic resonance imaging scan in the same patient showing vascular compression of the right trigeminal nerve
Balloon compression for trigeminal neuralgia
extension and rotation is difficult for them. Sterilely prepare the perioral region. Lubricate and tape the eyes shut. This reduces any risk of a corneal abrasion. This would be a significant problem should there be diminished corneal sensation. Mark the entrance site on the cheek 2.5 cm lateral to the angle of the lip. For first division pain treatment, enter slightly more lateral. For third division pain, enter slightly more cephalad. Place small sterile plastic drapes around the puncture site to isolate it, followed by a larger drape to cover the body. These drapes allow one to keep the balloon catheter and insoufflation syringe sterile during the operation. Fix the digital pressure monitor to an adjacent pole at the surgeon’s eye level or to the fluoroscopy table. Drape the fluoroscope imaging unit. Place the balloon catheter and insoufflation syringe on the sterile field. Use the FDA approved kit for this procedure. It includes sharp and blunt trocars, an introducing cannula, curved and straight guiding stylets and a #4 French diameter balloon. (Cook Vascular, Inc., Leechburg, PA). Fill the insoufflation syringe (Merit Medical, Inc., Salt Lake City, UT) with several milliliters of radio opaque dye and eliminate any air bubbles. The syringe measures intraluminal pressures in atmospheres. The target intraluminal pressure is 1.3–1.5 atmospheres. Intraluminal pressure does not need to be measured. Many neurosurgeons control pressure indirectly by varying the volume of dye used to fill the balloon in order to obtain the pear shape. Inflate the #4 French balloon catheter with 0.75–1.0 of contrast until a pear shape occurs. The pear shape may not always appear. Pressure measurement provides a consistent way to determine how much to squeeze the trigeminal nerve. Obtain a pure lateral skull image by aligning the plane of the planum sphenoidale and clivus so that they each form a single line. If the image is not aligned, the imaged position of the catheter tip may be misleading. Insert the sharp obturator
143
in the #14 gauge cannula and place it preliminarily at the planned entrance point on the cheek. Nick the skin with the #15 blade that is provided in the kit. This incision accommodates the cannula, and is closed with a steri-strip. Use a lateral view first. Again, set the cannula’s point of entry so that that it approaches the foramen ovale on a plane parallel to the floor of the middle fossa and the foramen ovale. The cannula is nearly parallel to the slope of the petrous bone for third division pain. For maxillary division pain choose a more oblique angle. For first division pain the oblique angle is about 30 degrees above the plane of the petrous bone. Regardless, do not insert the #14 gauge cannula beyond the foramen ovale. Only insert the guiding stylets and balloon catheter intracranially. For third division pain, direct the cannula towards the radiographic intersection of clivus and petrous bone. Once the cannula passes through the skin of the cheek, remove the sharp obturator and replace it with the blunt one. This will prevent vascular injury in the cheek or skull base. Puncture the deep epidermal layers of the check using the cannula with the sharp obturator. Then remove the sharp obturator and replace it with the blunt one. Direct the cannula to the skull base. Once the cannula reaches the skull base, switch to a modified submental view. Slightly extend the patient’s neck and rotate it laterally about 15 degrees to the opposite side. Aim the imaging unit at an angle of about 30 degrees under the chin. In this view, the foramen ovale is sited medial to the mandible, lateral to the maxilla, and directly above the petrous bone and is directly visualized. Once the foramen ovale is visible, advance the cannula using repeated imaging. Center the imaging unit so that the barrel of the cannula points directly to the foramen ovale. (> Figure 143-2) When the foramen is engaged you will feel resistance from the cannula. The depressor response usually occurs, but only briefly and not strongly. There may be a slight facial muscular twitch from stimulation of the
2459
2460
143
Balloon compression for trigeminal neuralgia
. Figure 143-2 Radiographic modified submental image showing the foramen ovale with a #14 Gauge cannula and straight guiding stylet directed towards the center of the foramen. The foramen ovale is found just above the petrous bone, lateral to the maxillary sinus and medial to the mandible
nerve. The cannula should not penetrate any further. Cerebrospinal fluid does not drip through the cannula, since the cannula has not entered the subarachnoid space surrounding the trigeminal ganglion. This is different from the expectations regarding appropriate positioning that occurs with glycerol and thermal rhizotomy. Once the cannula engages the foramen ovale, remove the blunt obturator and insert a straight guiding stylet. Venous bleeding occurs from the epidural venous complex. This can be stopped by advancing the cannula deeper into the foramen ovale. Insert a straight guiding stylet until it passes beyond the edge the cannula and the inner aspect of the foramen ovale. You will feel a characteristic ‘‘pop,’’ similar to the tactile feeling noted when performing a lumbar puncture. If there is excessive resistance to advancing the stylet, then the cannula is not ideally positioned at the foramen ovale. Either the cannula
slipped out of the foramen, or approaches it at an oblique angle that prevents passage of the stylet through the foramen. Next, obtain an anterior-posterior imaging view. Center the petrous bone radiographically in the center of the ipsilateral orbit on the image intensifier. The medial dip in the petrous bone is the entrance to Meckel’s cave, the porus trigeminus. The dura splits over this erosion in the skull base to allow the trigeminal root to pass into the subarachnoid space. Direct the stylet toward the center of the porus for second division pain or multidivisional pain, the lateral porus for third division pain and the medial porous for first division pain. In patients with predominantly first division pain, aim the cannula from a more lateral to medial position to enable the stylet to pass into the medial segment of the porus. The entrance to the porus is located approximately 17 mm beyond the foramen ovale. The cannula system is designed so that when the cannula is placed at the foramen ovale, it is not possible to pass the guiding stylet beyond the porous trigeminous into the posterior fossa. Use the curved stylet to reach the medial or central aspect of the porous more readily. Its curve is positioned down during passage through the cannula. Then rotate the curve superomedially, after it is at the porous in order to decrease the low risk of dural perforation. If the dura is perforated, the stylet should be redirected to remain intradural so that the balloon adequately squeezes the nerve. The only indication that dural perforation has occurred may be that a pear shape fails to appear during inflation despite the anterior-posterior and lateral images indicating appropriate balloon position. Obtain a lateral view to use during balloon inflation. To position the balloon catheter at the best angle for third, multidivisional or first division pain, position the fluoroscopy imaging beam parallel to the skull base. The planum sphenoidale and posterior clinoids when superimposed in the fluoroscopic images also provide
Balloon compression for trigeminal neuralgia
the best image of the sella turcica. This view will then show the cannula tip at the base of the middle fossa. The lateral view is then used for the balloon compression. For second or third division pain, the stylet should remain parallel and adjacent to the petrous bone. For first division pain, the stylet should be visualized more superiorly above the petrous bone. Once the proper trajectory is established, the stylet is removed. Test the balloon function by removing the stylet and inflating the balloon with air using the tuberculin syringe provided in the kit, then replace the stylet. Place the tip of the balloon just beyond the edge of the petrous bone as seen in the anterior posterior view for third division pain. For first division pain the balloon is positioned slightly more posterior, approximately 3 mm beyond the petrous edge. This is because the catheter is being directed superomedial to reach the first division fibers that are located in the superior portion of the trigeminal root. The third division fibers are located in the inferior portion of the trigeminal root. Again, because the catheter is approaching the nerve root from the foramen ovale, it is traversing the root obliquely. Once properly positioned radiographically, secure the locking device firmly on the balloon catheter at the edge of the cannula. This will prevent the balloon from sliding forward into the posterior fossa when it is inflated. Remove the inner stylet. Connect the catheter to the insoufflation syringe and evacuate all air from the balloon using the tuberculin syringe provided in the kit and a three-way stopcock. Finally, attach the syringe to the digital monitor, which will then stabilize at a zeroed pressure. Slowly inflate the balloon with radioopaque dye while repeatedly monitoring blood pressure and observing intermittent lateral fluoroscopic images. When the balloon inflates within the porus, a pear shape appears on the lateral fluoroscopic image. (> Figure 143-3) If the catheter tip
143
. Figure 143-3 Lateral radiographic image of the skull showing alignment of the planum sphenoidale, clival line and base of middle fossa. The balloon is inflated and has pear shape indicating that its tip is located within the entrance to Meckel’s cave. Note that the balloon extends beyond the clival line when properly compressing the trigeminal nerve at the porus trigeminus
is not within the porous, the pear is not seen and there will be less numbness created. If pain relief does occur, it is likely to be delayed and brief. The reason for this is as follows: In the middle fossa, over the ganglion, the balloon lifts the dura off the ganglion. Within the porus, it compresses the retrogasserian fibers against the firm edge of the dura and the petrous ridge as the dura splits allowing the nerve to pass into Meckel’s cave. It is possible to achieve compression at a higher pressure here. Rupture of the balloon does not cause any injury. If the patient is known to be allergic to the iodine based dye used, then premedicate the patient with intravenous steroids. When properly inflated, using a #4 French balloon catheter, the intraluminal balloon pressure should be raised to 1.3–1.6 atmospheres. Once that pressure is reached and a pear shape is present, the balloon should remain inflated
2461
2462
143
Balloon compression for trigeminal neuralgia
for 1 min-longer if more significant numbness is sought. The depressor response usually occurs, which triggers the pacemaker, but only briefly. There may be a secondary elevation of blood pressure especially after triggering the pacemaker. Control any hypertensive response by using additional anesthetic. After deflation remove the catheter and cannula together. If the catheter is removed separately, cerebrospinal fluid may drip through the cannula. The subarachnoid space surrounding the ganglion in Meckel’s cave is opened when the balloon inflates. Compress the skin incision against the maxilla for 5 min to reduce venous bleeding in the cheek. Close the incision with a steri tape. Begin the extubation sequence after the blood pressure is stable.
Post-operative Management In the recovery room, reduce post operative swelling in the cheek by using an ice pack. The patient may be discharged home a few hours after the procedure or remain hospitalized until the next day. Postoperatively, evaluate for the adequacy of their relief. Typically, the patient awakens without pain, except for local discomfort at the needle entry site. Trigeminal pain may persist for a day or two post-operatively and then subside if the compression has been mild. Assess for the extent of sensory loss. A decrease in touch and pinprick sensation occurs in two-thirds of patients. Patients may find the associated subjective numbness slightly uncomfortable at first, but usually adjust in the first 3–4 weeks. The numbness generally decreases substantially by 3–6 months. Pain relief tends to persist even after the numbness resolves. Decreased sensation is most common in the third division and sometimes occurs in patients being treated with first or second division pain. Despite this absence of sensory change, the patient may have complete pain relief. Rarely, patients report dysesthesias,
i.e. unpleasant hypalgesias (decreased sensitivity to pain) and hypoesthesias (decreased sensitivity to touch). The corneal reflex is usually not decreased, perhaps because compression selectively preserves A-delta and C fibers [8]. This selectivity makes balloon compression especially useful as a percutaneous treatment of patients with first division TN pain. Thermal rhizotomy injures all fiber types nonselectively at temperatures required to create a clinical lesion. Gamma knife radiosurgery has similar global effects on the nerve. Glycerol is a mild demyelinating agent. The operation does not easily cause divisional nerve injury. Patients who have a history of herpes simplex can have a labial eruption within days of surgery. Premedication does not seem to prevent this from happening. Mild ipsilateral temporal and masseter muscle weakness occurs in two-thirds of patients, but usually is of minimal clinical relevance as patients generally have avoided using that side of their mouths for years. If temporormandibular joint pain develops because of the muscular imbalance created by the masseter muscle weakness, treat it with oral anti-inflammatory medication until resolution.
Discussion Numerous series have been published over the last two decades and have recognized other infrequent complications [9–23]. One case report described balloon compression resulting in subarachnoid hemorrhage and fatal complications secondary to the sharp obturator being advanced well beyond the foramen ovale [24]. A second death occurred following rupture of a dural AV fistula during the procedure. A carotid cavernous fistula and an external carotid fistula have also been reported [25]. This experience suggests that a cannula with a blunt obturator should be used to reach the foramen ovale through the cheek.
Balloon compression for trigeminal neuralgia
Temporo-mandibular joint (TMJ) symptoms are present in 44% of patients before balloon compression [26]. One week and one month after balloon compression there were significant increases from baseline in jaw mobility, jaw deviation and limited jaw opening. These normalized at 7 months after surgery. In this study the authors conclude that TN causes TMJ symptoms; balloon compression causes temporary masticatory muscle injury and recovery from TN after surgery leads to improved jaw mobility [26]. In a review of 290 patients followed for a mean of 19 months, 87% had immediate pain relief after compression for 3–10 min at surgery. There was a 5% recurrence rate using this duration of compression. The authors did not control pressure [27]. Hearing, taste and/or smelling loss, and visual difficulties after compression have been reported [28]. Hearing difficulties may occur because of eustacian tube dysfunction from motor sequelae of trigeminal nerve injury. Taste and smell relate to sensory input from the tongue and could exhibit dysfunction associated with severe facial and tongue numbness. In a separate study significant difference in the olfactory threshold at the immediate post-operative period is noted (p = 0.048) [29]. The authors of this study do not correlate the degree of numbness with the likelihood of reporting such symptoms. Compression for longer than 1 min, though reducing the recurrence rate, may alter the incidence and extent of associated morbidity. In our most recent review of results in 56 patients, the recurrence rate in patients who had initial relief was 16%. The mean time until recurrence in patients who experienced pain relief after surgery was 13 months (range 3–23 months). Mild numbness immediately after surgery was observed in 83% of patients. At the most recent evaluation, 17% of patients reported persistent, non troublesome numbness and none had moderate or severe numbness. Minor dysesthesia was present in two patients (4%). Mild masseter/pterygoid muscle weakness occurred in 24% of patients and resolved
143
within a maximum period of 1 year [30]. Dan reviewed his 496 patients with typical symptoms of unilateral trigeminal neuralgia with a mean evaluation time of nearly 11 years. Pain recurrence occurred in 19% of patients within 5 years and in 32% over the 11-year review period. Symptomatic dysesthesias occurred in 4% without any corneal anesthesia or anesthesia dolorosa [21]. This is the largest and longest study of the procedure yet available.
Summary Percutaneous balloon compression is a simple and effective treatment for trigeminal neuralgia successfully employed for more than two decades. It is especially useful in patients with first division pain because it does not injure the myelinated fibers that mediate the blink reflex. It is also most helpful in patients with pain that has spread across multiple division because it does not require multiple lesions; It also is of advantage in older patients with whom it would be difficult to communicate during selective thermal rhizotomy. It is a relatively inexpensive, when compared with gamma knife radiosurgery, for example, and it is a technically and technologically simple operation in a period of medicine where cost considerations have become an important issue.
References 1. Shelden CH, Pudenz RH, Freshwater DB, Crue BL. Compression rather than decompression for trigeminal neuralgia. J Neurosurg 1955;12(2):123-6. 2. Mullan S, Lichtor T. Percutaneous microcompression of the trigeminal ganglion for trigeminal neuralgia. J Neurosurg 1983;59(6):1007-12. 3. Brown JA. Percutaneous trigeminal nerve compression. In: Batjer HH, Loftus CM, editors. Textbook of neurological surgery: principles and practice. Philadelphia: Lippincott Williams and Wilkins; 2002. p. 2980-5. 4. Burchiel KJ. A new classification for facial pain. Neurosurgery 2003;53(5):1164-6. 5. Brown JA, Hoeflinger B, Long PB, et al. Axon and ganglion cell injury in rabbits after percutaneous
2463
2464
143 6. 7.
8.
9.
10.
11.
12. 13.
14.
15.
16.
17.
18.
Balloon compression for trigeminal neuralgia
trigeminal balloon compression. Neurosurgery 1996;38 (5): 993-1003. Mullan S, Brown JA. Trigeminal neuralgia. Neurosurg Quart 1996;6(4):267-88. Sindou M, Leston J, Howeidy T, Decullier E, Chapuis F. Micro-vascular decompression for primary Trigeminal Neuralgia (typical or atypical). Long-term effectiveness on pain; prospective study with survival analysis in a consecutive series of 362 patients. Acta Neurochir (Wien) 2006;148(12):1235-45. Cruccu G, Inghilleri M, Fraioli B, Guidetti B, Manfredi M. Neurophysiologic assessment of trigeminal function after surgery for trigeminal neuralgia. Neurology 1987;37(4):631-8. Abdennebi B, Bouatta F, Chitti M, Bougatene B. Percutaneous balloon compression of the Gasserian ganglion in trigeminal neuralgia. Long-term results in 150 cases. Acta Neurochir (Wien) 1995;136(1–2):72-4. Belber CJ, Rak RA. Balloon compression rhizolysis in the surgical management of trigeminal neuralgia. Neurosurgery 1987;20(6):908-13. Brown JA, Gouda JJ. Percutaneous balloon compression of the trigeminal ner ve. Neurosurg Clin N Am 1997;8(1):53-62. Connelley TJ. Balloon compression and trigeminal neuralgia. Med J Aust 1982;2(3):119. Correa CF, Teixeira MJ. Balloon compression of the Gasserian ganglion for the treatment of trigeminal neuralgia. Stereotact Funct Neurosurg 1998;71(2):83-9. Fiume D, Scarda G, Natali G, Della VG. Percutaneous microcompression of the gasserian ganglion. New treatment for trigeminal neuralgia. Riv Neurol 1985;55 (6):387-91. Lee ST, Chen JF. Percutaneous trigeminal ganglion balloon compression for treatment of trigeminal neuralgia, part II: results related to compression duration. Surg Neurol 2003;60(2):149-53. Lobato RD, Rivas JJ, Sarabia R, Lamas E. Percutaneous microcompression of the gasserian ganglion for trigeminal neuralgia. J Neurosurg 1990;72(4):546-53. Meglio M, Cioni B, d’Annunzio V. Percutaneous microcompression of the gasserian ganglion: personal experience. Acta Neurochir Suppl (Wien) 1987;39:142-3. Meglio M, Cioni B. Percutaneous procedures for trigeminal neuralgia: microcompression versus radiofrequency thermocoagulation. Personal experience. Pain 1989;38(1):9-16.
19. Mizuno M, Saito K, Takayasu M, Yoshida J. Percutaneous microcompression of the trigeminal ganglion for elderly patients with trigeminal neuralgia and patients with atypical trigeminal neuralgia. Neurol Med Chir (Tokyo) 2000;40(7):347-50. 20. Natarajan M. Percutaneous trigeminal ganglion balloon compression: experience in 40 patients. Neurol India 2000;48(4):330-2. 21. Skirving DJ, Dan NG. A 20-year review of percutaneous balloon compression of the trigeminal ganglion. J Neurosurg 2001;94(6):913-17. 22. Urculo E, Arrazola M, Gereka L, . et al. Evaluation of the Mullan’s technique in the treatment of trigeminal neuralgia. Rev Neurol 1998;27(157):477-84. 23. Zanusso M, Curri D, Landi A, Colombo F, Volpin L, Cervellini P. Pressure monitoring inside Meckel’s cave during percutaneous microcompression of gasserian ganglion. Stereotact Funct Neurosurg 1991;56(1):37-43. 24. Spaziante R, Cappabianca P, Peca C, de Divitiis E. Subarachnoid hemorrhage and ‘‘normal pressure hydrocephalus’’: fatal complication of percutaneous microcompression of the gasserian ganglion. Case report. Neurosurgery 1988;22(1 Pt 1):148-51. 25. Langford P, Holt ME, Danks RA. Cavernous sinus fistula following percutaneous balloon compression of the trigeminal ganglion. Case report. J Neurosurg 2005;103 (1):176-8. 26. de S, Sr., da Nobrega JC, Teixeira MJ, de Siqueira JT. Masticatory problems after balloon compression for trigeminal neuralgia: a longitudinal study. J Oral Rehabil 2007;34(2):88-96. 27. Liu HB, Ma Y, Zou JJ, Li XG. Percutaneous microballoon compression for trigeminal neuralgia. Chin Med J (Engl) 2007;120(3):228-30. 28. de S, Sr., da Nobrega JC, de Siqueira JT, Teixeira MJ. Frequency of postoperative complications after balloon compression for idiopathic trigeminal neuralgia: prospective study. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2006;102(5):e39-e45. 29. Siqueira SR, Nobrega JC, Teixeira MJ, Siqueira JT. Olfactory threshold increase in trigeminal neuralgia after balloon compression. Clin Neurol Neurosurg 2006;108(8):721-5. 30. Brown JA, Pilitsis JG. Percutaneous balloon compression for the treatment of trigeminal neuralgia: results in 56 patients based on balloon compression pressure monitoring. Neurosurg Focus 2005;18(5):E10.
125 Bulbar DREZ Procedures for Facial Pain J. P. Gorecki
The goal of ablative surgical procedures at the level of the brainstem and dorsal root entry zone (DREZ) is the permanent destruction of secondorder neurons implicated in nociceptive afferent pathways. This destruction removes cell bodies, axons of the second-order neurons, interneurons, and primary afferent axons before they synapse with second-order neurons within the central nervous system. The axons of the second-order neurons comprise the tracts that terminate in the thalamus and brainstem. This tract is the spinothalamic tract. It is a widely held generalization that ablative procedures are useful only for pain of the cancerous kind. NCD and DREZ operations are unique exceptions to this generalization. NCD is also effective for facial pain associated with anesthesia dolorosa, atypical facial pain, postherpetic pain, trauma induced pain, and some deafferentation pains. In the spinal cord the DREZ operation has been shown to be effective for the pain of brachial plexus avulsion and to some extent for pain associated with spinal cord injury. Midbrain tractotomy, in my opinion, is the surgical procedure of choice for nociceptive pain associated with cancer that is unilateral and extends above the C5 dermatome. The physiological justification for destructive procedures is based on some simple assumptions. Ablation of the normal nociceptive pathway should remove pain that is perceived in response to nociceptive stimuli. Such ablation within the central nervous system at the level of second-order neurons should be both permanent and effective for the induced pain of cancer. Nerve regeneration often leads to loss of #
Springer-Verlag Berlin/Heidelberg 2009
effectiveness following ablation at the level of the peripheral nervous system. The conscious perception of pain is influenced by both nociceptive input and spontaneous neuronal activity within the central nervous system. Abnormally increased electrophysiological activity is identified within the dorsal horn in many central pain syndromes [1,2]. This abnormal activity is theorized to be causative in the generation of pain; therefore, elimination of this activity should be helpful in the management of such central pain syndromes. Destruction of the dorsal root entry zone accomplishes this goal. Somatic sensory input from the face is carried by cranial nerves V, VII, IX, and X. Fibers that carry nociceptive input from all four of these cranial nerves are located centrally in the descending trigeminal tract. The descending trigeminal tract terminates in the nucleus caudalis. The nucleus caudalis is one of the three subdivisions of the nucleus of the spinal tract. The nucleus caudalis contains second-order neurons for nociception from the head and face. Axons from the nucleus caudalis normally cross, travel in the quintothalamic tract, and terminate with synapses on third-order neurons in the thalamus or periaqueductal gray. For this reason a number of surgical ablative procedures specifically target the descending trigeminal tract, nucleus caudalis or both. Ablative procedures of the descending trigeminal tract and nucleus caudalis are performed as either open microsurgical operations or operations using stereotactic methods. The ablative lesions consist of a single transverse incision, a single cylindrical or spherical lesion, or a series
2126
125
Bulbar DREZ procedures for facial pain
of lesions that coalesce over a length of the medulla and upper spinal cord. The following terms are commonly used to refer to these ablative operations: nucleus caudalis DREZ, trigeminal nucleotomy, and descending trigeminal tractotomy. Given the limitations of the current technology it is difficult to distinguish the actual physiological structures that are ultimately ablated by these various procedures. The trigeminal nerve is the main sensory nerve for the face and the head. Sensation from this area is also transmitted via cranial nerves VII, IX, and X as well as the upper cervical branches. The nuclei of the trigeminal nerve include the motor nucleus of V, the chief sensory nucleus, the mesencephalic nucleus, and the nucleus of the spinal tract. The nucleus of the spinal tract, located in the medulla, is divided into three subnuclei based on the work of Olszewki [3]; pars caudalis, pars interpolaris (or oralis), and pars rostralis. The nucleus of the trigeminal tract is histologically indistinguishable from the gray matter of the dorsal horn of the upper cervical cord. The nucleus of the spinal tract receives input from the descending trigeminal tract. The descending trigeminal tract contains fibers transmitting pain and nociceptive signals that originate in cranial nerves V, VII, IX, and X. Some fibers in the spinal tract descend as far as the uppermost three segments of the cervical cord where they mingle within Lissaur’s zone. Nociceptive fibers terminate in the nucleus caudalis in the medulla. The nucleus caudalis is a rostral extension of the dorsal horn of the spinal cord. Cell bodies of the trigeminal primary sensory neurons are located within the gasserian ganglion and the mesencephalic nucleus. The gasserian ganglion corresponds anatomically to the segmental dorsal root ganglia. The mesencephalic nucleus contains unipolar primary sensory neurons implicated in proprioception. The mesencephalic nucleus, located beneath the lateral edge of the fourth ventricle, is made up of primary sensory neuronal cell bodies.
The presence of these cell bodies within the central nervous system rather that in ganglia is anatomically unique. The neurons in the sensory trigeminal nuclei represent second-order neurons. Efferent fibers from these neurons terminate in the nucleus ambiguous, hypoglossal nucleus, reticular formation, cerebellum, and both the ventral and dorsal trigeminothalamic tracts to the thalamus. Somarotopic organization is described within the descending trigeminal tract and the nucleus caudalis. The segmental organization of sensory representation is described in two different patterns. The first is a concentric pattern of rings that has been compared to onion rings. Sensation from the middle portion of the face immediately surrounding the mouth is represented at more rostral levels within the medulla. Clinical observation, that it is more difficult to achieve dense analgesia close to the midline of the face with nucleotomy or tractotomy supports this pattern. Preservation of pain appreciation in the midline of the face is most apparent when nucleotomy is performed at more caudal levels of the medulla. The second pattern is based on the segmental trigeminal divisions. Fibers that originate in the first trigeminal division reach a more caudal level within the nucleus caudalis and the descending tract than do fibers that originate in the third trigeminal division. This is supported by both direct sensory evoked potential recording during surgery and by the observation that following nucleotomy or tractotomy dense analgesia is easier to achieve in the first trigeminal division than in the third division. The third division of the trigeminal nerve is not represented as far as the caudal in the medulla. Segmentation corresponding to trigeminal divisions is also present from the medial to the lateral within the medulla. Fibers from cranial nerves VII, IX, and X are located most medially in the descending tract, immediately adjacent to the dorsal column. Fibers from the third trigeminal division are located immediately lateral to
Bulbar DREZ procedures for facial pain
the fibers from the nerves VII, IX, and X. Fibers from the first trigeminal division are located most laterally within the descending tract, furthest from the dorsal column and closer to the motor roots of the vagus and spinal accessory nerve. For these reasons, in order to achieve dense analgesia that includes the third trigeminal division, the lesion must be made close to the dorsal column and relatively rostral in the medulla. Such a lesion location increases the risk of undesired injury to the dorsal column and a subsequent proprioceptive deficit. Nucleus caudalis DREZ is used to treat medically intractable pain in the face. The pharmacological management of deaffreneation pain in patients who are candidates for NCD consists of analgesic agents of increasing potency combined with antidepressant agents, anticonvulsant agents, gamma aminobutyric acid (GABBA) modifying agents, and topical agents. Topical agents include DMSO, aspirin, anesthetic agents, and capsaicin. NCD is only considered when maximal pharmacological management fails and the severity of the pain negatively impacts the patient’s quality of life. Most patients treated for trigeminal neuralgia have undergone prior surgery. Psychological evaluation is used to verify that the patient does not have an associated psychiatric illness. The most common indication for NCD is for the treatment of pain due to benign pathology. The operation is performed under general anesthesia and involves the exposure of the upper spinal cord and lower brainstem. Since many stereotactic and percutaneous techniques are more applicable to the pain of cancer, offering such an invasive operation to patients debilitated by malignancy seems overly aggressive. NCD can and has been used to treat bilateral pain. NCD is and should be considered for patients with end-stage trigeminal neuralgia. Such patients have failed all medical intervention and have failed at least one of the more common surgical procedures for trigeminal neuralgia such
125
as microvascular decompression, radiofrequency retrogasserian rhizolysis, glycerol, balloon compression, GAMMA Knife, avulsion, alcohol ablation, or open nerve section. NCD is especially effective for first division pain that in many instances is more difficult to treat with the more common operations listed previously. A patient with a preserved corneal reflex prior to NCD is unlikely to develop corneal anesthesia following surgery. This fact drove Sjoqvist to perform the first descending trigeminal tractotomy in 1937. NCD is also employed to treat postherpetic neuralgia, atypical facial pain, anesthesia dolorosa and various deafferentation pain syndromes in the face due to prior surgery or trauma. NCD is more effective for nociceptive type posttraumatic pain. A few cases of pain from tumors compressing the fifth nerve have been treated in this manner. The earliest reports about NCD indicate that there is a high success rate for postherpetic neuralgia [4,5]. This finding has not been supported by reviews with longer followup evaluations (see > Table 125‐1). Several authors describing descending trigeminal tractotomy specifically state there is no benefit for postherpetic neuralgia [6,7]. NCD has also been used in the treatment of cluster and migraine headaches. NCD can be considered as an extrapolation of DREZ coagulation from the spinal cord to involve the equivalent structure for the face. The nucleus caudalis is comparable to the dorsal root entry zone. NCD consistently results in analgesia in the absence of anesthesia. The incidence of anesthesia dolorosa or painful dysesthesia following NCD is low. The corneal reflex is preserved following NCD. For these reasons lesions at the level of the nucleus candalis are preferable to sectioning of the nerve or root. NCD is most effective at producing analgesia in the first trigeminal division and avoids the risk of corneal ulceration. NCD should be considered for the treatment of
2127
2128
125
Bulbar DREZ procedures for facial pain
. Table 125‐1 Character of pain
Etiology
Number
Outcome
Change in pain score
Anesthesia dolorosa
Tic douloureux
18
Decrease 0.3
Anesthesia dolorosa
Stroke
5
Anesthesia dolorosa
Schwannoma
1
Anesthesia dolorosa
Multiple sclerosis
1
Anesthesia dolorosa
Trauma deafferentation
4
Nociceptive
Trauma nociceptive
6
Nociceptive
Cluster headache
1
Nociceptive
Postherpetic neuralgia
12
Unknown
Atypical facial pain
10
Excellent 1 Good 0 Fair 3 Poor 12 Lost 2 Excellent 0 Good 0 Fair 2 Poor 2 Lost 1 Excellent 0 Good 0 Fair 0 Poor 1 Lost 0 Excellent 0 Good 0 Fair 0 Poor 0 Lost 1 Excellent 0 Good 0 Fair 1 Poor 3 Lost 0 Excellent 4 Good 2 Fair 0 Poor 0 Lost 0 Excellent 0 Good 0 Fair 0 Poor 0 Lost 1 Excellent 3 Good 0 Fair 3 Poor 4 Lost 2 Excellent 1 Good 0 Fair 5 Poor 4 Lost 0
Decrease 0.5
Decrease 0
Lost
Decrease 0
Decrease 5.0
Lost
Decrease 3.5
Decrease 2.5
Outcome from NCD performed at Duke University between January 1990 and 1999
first division trigeminal pain. NCD is most effective for pain that is intermittent and induced, or pain that can be classified as nociceptive. For dental pain it may be necessary to include lesions
of the pontine trigeminal nucleus or to use midbrain tractotomy to achieve dense analgesia. The usefulness of NCD for postherpetic neuralgia remains to be proven. Falconer [8] and Sjoqvist
Bulbar DREZ procedures for facial pain
[9] both state that descending tractotomy is not effective for postherpetic neuralgia. In contrast, early articles about NCD suggest that postherpetic neuralgia responds the best to NCD. Data from more cases and longer follow-up may clarify this inconsistency. We should recall that DREZ in the spinal cord lost its appeal for treating postherpetic neuralgia. Class I data evaluating pharmachological management, intraspinal narcotic analgesia, NCD, midbrain tractotomy, and maybe deep brain stimulation would seem to be in order. Facial pains that should be considered for NCD include atypical facial pain, tic that has failed other therapy, postherpetic neuralgia, and posttraumatic pain. A previous review of the results for NCD suggested that the most benefit was obtained for patients with atypical facial pain; however additional review of the data suggests that the best results are in fact obtained by patients with intermittent, induced, or nociceptive type pain. Lesser preoperative denervation correlates with a better outcome. NCD is performed under general anesthesia. Muscle relaxation is discontinued in order to avoid interference with intraoperative EMG recording. The patient is positioned prone and the head is supported in pin fixation. The exposure is midline. Muscle is removed from the occiput to expose the foramen magnum and removed unilaterally from C1 and C2. An eccentric craniectomy or craniotomy is performed on the side ipsilateral to the pain and the dura is exposed overlying the cerebellar tonsil. The arch of C1 is removed on the symptomatic side. The bone of C2 is left intact. The dura is opened eccentric to the midline with a curved or Y shaped incision and retracted to expose the symptomatic side. The landmarks that are exposed include the obex, the rootlets of C1 and C2, the rootlets of cranial nerve XI, the vertebral artery, and the dorsolateral sulcus. A series of radiofrequency lesions is made extending the cephalad from the uppermost sensory rootlet of C2, along the dorsolateral sulcus. Lesion parameters are 80 C for 15–20 s. The electrodes used to produce the
125
lesions have a diameter of 0.25 mm and incorporate insulation on the most proximal segment of the electrode that penetrates the central nervous system. Two different sized electrodes are used. The longer electrode is used to make lesions more cephalad, beyond the rootlets of C1. The nucleus caudalis has a larger cross sectional diameter at more cephalad levels. The longer electrode can be used more safely at these more cephalad levels since the pyramidal tract does not lie in immediate contact with the deep surface of the nucleus caudalis at these more cephalad levels. The shorter electrode is used at more caudal levels to reduce the chance of injury to the pyramidal tract. The electrodes contain a curve close to the active tip to make placement of the lesion more ergonomic for the surgeon. The purpose of the insulation is to prevent coagulation of the central nervous system tissue close to the surface at the site of penetration. The spinocerebellar tract overlies the descending trigeminal tract, especially at the more cephalad locations. More aggressive electrophysiological monitoring is also available. Prior to creating a lesion, the anatomy is mapped by recording trigeminal somatosensory evoked potentials (TSSEPs) directly from the electrode tip that is penetrated into the spinal cord and medulla. TSSEPs are simultaneously recorded from the scalp. The evoked potentials arise from electrical triggers passed through bipolar needle electrodes placed adjacent to the mental nerve, inraorbital nerve, and supraorbital nerve. The amplitude of the TSSEP along the dorsolateral sulcus and the threshold stimulus required to produce a recordable response is documented. It is therefore possible to localize which portion of the nucleus caudalis or descending tract is obtaining maximal input from various divisions of the trigeminal nerve. It is also possible to tailor the lesion based on the anatomical location of the pain. The trigeminal first division is represented at the most caudal locations, while the representation of the third division is limited to more cephalad locations.
2129
2130
125
Bulbar DREZ procedures for facial pain
The lateral margin of the nucleus can be mapped by recording similar evoked potentials induced by a stimulus applied to the ipsilateral median nerve. As the recording electrode is progressively moved to more lateral positions, a point is reached beyond which evoked potentials are no longer recorded following stimulation of the median nerve unless the intensity of the stimulus is increased many fold. Similarly, the amplitude of the recorded SSEP drops sharply at this point. This point corresponds to the location from which low-threshold TSSEPs are recorded. Passing a stimulating current directly through the electrode penetrating the central nervous system identifies proximity to the pyramidal tract. Stimulation results in an EMG response that can be recorded from leg or arm muscles with a threshold of less than 1 volt. Lesions should not be created at these locations. Electrophysiologists are evaluating potential techniques to record from or demonstrate the location of the spinocerebellar tract. With this technique of more aggressive monitoring it is possible to significantly reduce the number of lesions necessary to achieve analgesia and consequently can reduce the incidence of complications. Statistically each time a lesion is created there is a finite risk of a complication. Completing the operation with fewer lesions should reduce the overall risk. No cases of ataxia occurred following the few cases performed with this technique. Very few complications are described for stereotactic nucleotomy. This reflects one clear advantage of producing lesions in an awake, cooperative patient. One must also question the accuracy of the reporting for this procedure with a healthy skepticism. The most common complication following NCD was first described by Sjoqvist. He used the term ataxia and attributed the clinical picture on an injury to the spinocerebellar tract and restiform body. This complication affects the
ipsilateral arm. Occasionally the leg is also affected, resulting in gait impairment. The patient may describe inability to use the affected arm. The complaint is often beyond proportion to the physical examination, especially after the patient has completed a course of physical therapy. Deficits that can be identified on examination include, in order of frequency, ataxia, passpointing, loss of two-point discrimination, and weakness; Often there is a modest weakness present as the patient is emerging from anesthesia that resolves completely. Although the residual symptoms may be subjective, patients can report significant disability. We believe that this complication results from injury to the spinocerebellar tract. Less commonly injured adjacent tracts include the cuneate tract and nucleus and the pyramidal tract. Our patients are placed on high doses of systemic steroids prophylacticly to reduce this complication. NCD has been modified three times to reduce the risk of ataxia. The first modification was the addition of proximal insulation to the electrode. The second modification was the introduction of electrodes of two different lengths for the lesions of different parts of the nucleus. The so-called Nashold/El-Naggar electrode also includes a bend in the electrode to make placement of the lesions physically more accurate. We then added more specific and rigid electrophysiological monitoring. Unfortunately the methodology used to document the electrophysiological data for all of the historical cases at Duke was such that accurate data are not available for retrospective analysis. All of the data are tainted by bias that cannot be corrected in the absence of the original data. The incidence of ataxia in the earliest reports of NCD was 90% [10]. The incidence of ataxia using the Nashold/El-Naggar electrode was reported to be 33% [11]. In the author’s experience, a total of 113 NCD operations were performed at Duke between 1982 and 2001. The first procedure was performed on a patient with a 44 month history
Bulbar DREZ procedures for facial pain
of postherpetic neuralgia. The operation has been performed using a standard technique since January 1990. The one exception is the increased use of electrophysiological monitoring of trigeminal evoked potentials in the last six cases. A single row of 16–20 lesions is made using two Nashold/El-Naggar electrodes. This more standardized technique was used in 58 patients and retrospective data was available for review for 51 of these patients. Follow-up evaluation was obtained by an interview in person or by telephone by a disinterested third party. Surgery was performed on 42 female patients and 16 male patients. The mean age for the group was 55.9 years. The follow-up was for a mean of 23.7 months with a range of 11–84 months. The etiology for the preoperative pain was classified as follows: postherpetic neuralgia in 12, atypical facial pain in 10, stroke in 5, trauma in 10, trigeminal neuralgia in 18, cluster headache in one, multiple sclerosis with tic douloureux in one, and schwanoma in one. The outcome is
125
classified as excellent in nine, good in two, fair in 14, poor in 26, and seven patients were lost to follow-up. The outcome results are shown in > Tables 125‐2 and > 125‐3. When asked to subjectively evaluate their quality of life, 16 patients described it as improved, six as somewhat improved, 16 as unchanged, and 13 as worse. A verbal digital score (VDS) is used to measure the severity of pain. The VDS is such that 0 represents no pain and 10 represents the worst imaginable pain. The preoperative VDS was 9 (range 6–10), suggesting a very high level of pain. The postoperative mean VDS was 5.2. Prior to surgery 100% of the patients were using narcotic medications regularly. Following NCD only 19 patients (37%) were still using narcotics. For the group as a whole there were no episodes of serious wound complications, CSF leak, or death. Twenty three of the 51 patients (45%) reported no untoward effects from the surgery. The reported complications include ataxia in 21 (41%), aseptic meningitis in one,
. Table 125‐2 Character
Etiology
Number
Outcome
Anesthesia Dolorosa
Trigeminal Neuralgia Stroke Trauma Multiple Sclerosis Tumor
29
Noceceptive
Trauma Postherpetic Neuralgia Cluster Headache
19
Neither
Atypical Facial Pain
10
Excellent 1 Good 0 Fair 6 Poor 18 Lost 4 Excellent 7 Good 2 Fair 3 Poor 4 Lost 3 Excellent 1 Good 0 Fair 5 Poor 4 Lost 0 Excellent 9 Good 2 Fair 14 Poor 26 Lost 7
Total
58
Patient would repeat the procedure
No 24 Yes 23 Unsure 4 Lost 7
2131
2132
125
Bulbar DREZ procedures for facial pain
. Table 125‐3 First author and reference
Procedure
Number
Siquera [12] Nashold [13] Nashold [4]
NCD NCD NCD
2
Bernard [14]
NCD
18
Nashold [15]
NCD
Bernard [5]
NCD
27
Ishijima [16]
NCD
4
Rossitch [17] Rawlings [18]
NCD NCD
5
Bernard [19] Young [10] Rossitch [20] Sampson [21]
NCD NCD NCD NCD
Few
Nashold [22] Rawlings [18] Chen [23] Nashold [11]
NCD NCD NCD NCD
2 21
El-Naggar [24] Spiegelmann [25] Grigoryan [26]
NCD
10
NCD
14
11 pain free
Bullard [101]
NCD
27
Excellent 7
Bullard [102]
20
Gorecki [103]
Redo NCD NCD
35
Gorecki [104] Sjoqvist [9] Grant [106] Hamby [107]
NCD IT IT IT
46 9 20 48
13
Outcome
3 months (13 good) 6 months (5 good, 1 fair, 7 poor) Excellent 11 Good 6 Fair 1 Poor 1
Immediate 85% Delayed 52%
Complication
Comment First report Abstract 5/6 postherpetic good
Incoordination 17/21
Dysmetria 20 3
5 immediate
Better outcome for less sensory loss
V1 ventrolateral, V2 dorsomedial 67% relief herpetic Complete relief for herpetic Cancer Dense packing, two rows, electrode insulation
5% Electrode insulation Technique El-Naggar/Nashohld electrode
18 2
100% Excellent 48% Good 5% Fair 5% Poor 43% 5 excellent
NCD
Excellent 12 Good 14 Fair 3 Poor 6 3 pain free 7 success 10/28 pain free
3–5% 3–5% None Ataxia 33%
4 + 5 year fu New electrode
0%
New electrode
Ataxia 33%
Best for postherpetic
Paresis 1 Ataxia 3 Brown-Sequard 1 Hypesthesia 1 Ataxia 50% Meningitis 2 Ataxia 60%
Ultrasound
Redo Good for atypical facial pain
Vocal cord and ataxia No risk to restiform Mortality 5.7%
Moved lesion caudal Corneal reflex preserved
Bulbar DREZ procedures for facial pain
125
. Table 125‐3 (Continued) First author and reference
Procedure
Number
Outcome
Complication
Comment
Mortality 46% in cancer Falconer [8] Guidetti [109] McKenzie [110]
IT IT IT
20 124 42
Pain recurred 46 25% spots of pain
Moffie [111] Hosobuchi [112] Young [113] Hitchcock [114] Hitchcock [115] Crue [116] Fox [117] Schvarcz [118]
IT IT
8 6
4 pain free All
IT TN
7
TN
3
All good for herpes
TN TN TN
104
Herpetic 87.5% Anesthesia dolorosa 57% Dysesthesia 72%
Cerebellar 19% No corneal anesthesia
No lemniscal
Bilateral possible Dysesthesia in 8 Developed medullary spinothalamic tractotomy 13–15 year f/u SSEP Dental pain in pons
retinal artery thrombosis in one, hearing loss in two, neck pain in one, “turtle neck” tightness in the neck in three, vertigo in one, suicide in one, and a suicide attempt in one. The suicide and attempted suicide occurred in patients with continued pain and attests to the severity of the underlying pain problem. The summary of the published results for NCD are presented in > Table 125‐3. In conclusion, Nucleus Caudalis DREZ coagulation is the definitive surgical intervention for intractable pain secondary to brachial plexus avulsion. Similarly, DREZ provides definitive management for the nociceptive pain that is usually located in a dermatomal pattern close to the level of injury in patients with spinal cord injury and intractable pain. Further investigation of the role of DREZ is warranted in the management of phantom pain. DREZ and micro-DREZotomy are largely similar.
STT, dorsal column
Short fu
Free hand Procedure of choice for postherpetic
Stereotactic midbrain tractotomy is the procedure of choice for nociceptive pain due to cancer above the C5 dermatome. Given the effectiveness of midbrain tractotomy in such situations, this surgery should be offered to patients earlier rather than later. NCD, descending trigeminal tractotomy, and trigeminal nucleotomy should appropriately be considered together. These procedures are particularly effective for pain located in a V1 distribution. If corneal anesthesia is not present prior to surgery, these procedures should be accompanied by preservation of the corneal reflex. The risk of denervation dysesthesia is less following these procedures than following other operations that result in deafferentation. NCD is more effective for nociceptive type pain than central pain. The major risk associated with these procedures is the development of ipsilateral ataxia. The risk of ataxia has been substantially
2133
2134
125
Bulbar DREZ procedures for facial pain
reduced with the use of appropriate monitoring of TSSEPs to direct the placement of lesions. Since the introduction of routine SSEP monitoring, ataxia has not been documented. It may be possible to eliminate the risk to adjacent white matter tracts by lesioning the nucleus caudalis with neurotoxins that directly attack the perikaryon and spare passing axons. Given the author’s experience with Nucleus Caudalis Dorsal Root Entry Zone (NCD) lesion creation, we no longer offer this surgical intervention. There are simpler more effective procedures, such as midbrain tractotomy or spinothalamic tractotomy, available for patients with chronic cancer pain. The results for NCD lesioning are not consistently favorable in a sufficient number of patients to justify the relatively high risk. Significant and dramatic improvements have been accomplished with the adoption of direct sensory evoked potential monitoring from the treatment electrode during NCD. Such advances would justify additional prospective, randomized, research about NCD in patients with deafferentation facial pain, post traumatic facial pain, and possibly post herpetic neuralgia. It is our anecdotal impression that the risk of complications during brainstem lesioning is less when fewer more precisely placed lesions are employed. It is quite possible that the risk to benefit the ratio is reasonable for trigeminal nucleotomy, as described by Jusef Kanpolat, in these patients.
References 1. Lombard MC, Larabi Y. Electrophysiological study of cervical dorsal horn cells in partially deafferentiated rats. Adv Pain Res Ther 1983;5:147-154. 2. Ovelmen-Levitt J, Johnson B, Bedenbaugh P. Dorsal root rhizotomy and avulsion in the can a comparison of the long term effects on the dorsal horn neuronal activity. Neurosurgery 1984;15:921-927. 3. Olszewski J. On the anatomical and functional organization of the spinal trigeminal nucleus. J Comp Neurol 1950;92:401-413.
4. Nashold BS Jr, Lopez H, Chodakiewitz JW, et al. Trigeminal DREZ for craniofacial pain. In: Samll M, editor. Surgery in and around the brainstem. Heldelberg: Springer; 1986, p. 54-59. 5. Bernard EJ, Nashold BS Jr, Caputi F. Clinical review of nucleus caudalis dorsal root entry zone lesions for facial pain. Appl Neurophysiol 1988;51:218-224. 6. Dogliotti M. First surgical sections, in man, of the lomniscus lateralis (pain-temperature path) at the brain stem, for the treatment of diffuse rebellious pain. Anesth Analg 1938;17:143-145. 7. Kanpolat Y, Deda H, Akyar S, et al. CT guided trigeminal tractotomy. Acta Neuracbir (Wien) 1989;100:112-114. 8. Falconer MA. Intramedullary trigeminal nactotomy and its place in the treatment of facial pain. J Neurol Neurosurg Psychiatry 1949;12:297-311. 9. Sjo¨qvist O. Studies on pain conduction in the trigeminal nerves: a contribution to the surgical treatment of facial pain. Acta Psychiatr Neurol Suppl 1938;17:1-139. 10. Young JN, Nashold BS Jr, Cosman ER. A new insulated caudalis nucleus DREZ electrode. Technical note. J Neurosurg 1989;70:283-284. 11. Nashold BS Jr, El-Naggar Amr O, Ovelmen-Levitt J, et al. A new design of radiofrequency lesion electrodes for use in the caudalis nucleus DREZ operation. J Neurosurg 1994;80:1116-1120. 12. Sequeira JM. A method for bulbospinal trigeminal nucleotomy in the treatment of facial deafferentation pain. Appl Neurophysiol 1985;48:277-280. 13. Nashold BS Jr, Caputi F, Bernard E. Trigeminal DREZ. Caudalis nuclear lesions for relief of facial pain. Neurosurgery 1986;19:150. 14. Bernard EJ, Nashold BS Jr, Caputi F, et al. Nucleus caudalis DREZ lesions for facial pain. Br J Neurosurg 1987;1:81-92. 15. Nashold BS Jr. Neurosurgical technique of the dorsal root entry zone operation. Appl Neurophysiol 1988;51:136-145. 16. Ishljima B, Shlmoji K, Shimizu H, et al. Lesions of spinal and trigeminal dorsal root entry zone for deafferentation pain. Appl Neurophysiol 1988;51:175-187. 17. Rossitch E Jr, Zeidman SM, Nashold BS Jr. Nucleus caudalis DREZ for facial pain due to cancer. Br J Neurosurg 1989;3:45-49. 18. Rawlings CE III, El-Naggar Amr O, Nashold BS Jr. The DREZ procedure: an update on technique. Br J Neurosurg 1989;3:633-642. 19. Bernard EJ Jr, Nashold BS Jr, Caputi F. Clinical review of nucleus caudalis dorsal root entry zone lesions for facial pain. Appl Neurophysiology 1988;51:218-224. 20. Rossitch E Jr, Young JN, Nashold BS Jr. Nucleus caudalis and dorsal root entry zone lesions for pain relief: an update. In: Wilkins RH, Rengachary SS, editors. Neurosurgery update II. Vascular, spinal, pediatric, and functional neurosurgery. New York: McGraw-Hill; 1990, p. 360-365.
Bulbar DREZ procedures for facial pain
21. Sampson JH, Nashold BS Jr. Facial pain due to vascular lesions of the brain stem relieved by dorsal root entry zone lesions in the nucleus caudalis. J Neurosurg 1992;77:473-475. 22. Nashold BS Jr, El-Naggar Amr O. Dorsal root entry zone (DREZ) lesioning. In: Rengachary SS, editor. Neurosurgical operative atlas. Lebanon, NH: AANS; 1992, p. 9-24. 23. Chen HJ. Facial pain relieved by dorsal root entry zone lesions in the trigeminal nucleus caudalis: report of two cases. J Formosa Med Assoc 1993;92:583-585.
125
24. El-Naggar Amr O, Nashold BS Jr. Nucleus caudalis DREZ lesions for relief of intractable facial pain. In: Wilkins RR, Rengachary SS, editors. Neurosurgery, 2nd ed. New York: McGraw Hill; 1995. 25. Splegelmann R, Friedman WA, Ballinger WE, et al. Anatomic examination of a case of open trigeminal nucleotomy (nucleus caudalis dorsal root entry zone lesions) for facial pain. Stereotactic Funct Neurosurg 1991;56:166-178. 26. Grigoryan YUA, Slavin KV, Ogleznev KYA. Ultrasonic lesion of the trigeminal nucleus caudalis for deafferentation pain. Acta Neurochir 1994;131:229-235.
2135
122 Comprehensive Management of Cancer Pain Including Surgery P. S. Kalanithi . J. M. Henderson
Introduction Cancer pain presents a unique challenge for both clinician and patient. For clinicians, cancer pain consists of numerous complex syndromes that not only challenge clinical decision making but also can strain the doctor-patient relationship. For patients, pain is often their principal complaint, leading to decreased ability to care for themselves, difficulty with ambulation, decreased appetite, weakness and depression. The pathophysiology of cancer pain has received extended study, and in parallel, clinical therapies and guidelines for pain treatment have been developed, principally focusing on pharmacologic therapy. Surgical management of cancer pain has advanced significantly, and many more options are now available to cancer patients suffering from debilitating pain. In the United States in 2006, an estimated 1.4 million cases of invasive cancer were diagnosed. Worldwide, over 7 million patients are diagnosed with cancer every year. 70–95% of cancer patients develop clinically significant pain [1–3]. Patients with gastrointestinal cancers constitute up to one-third of patients referred to pain services, but all cancer types are represented [4]. Over half of all cancer patients, despite therapy, have moderate to severe pain [5]. Consultation with a pain specialist is often necessary to ameliorate the patient’s pain, and often results in improved pain control. However, even with aggressive pain therapy monitored by pain specialists following WHO guidelines, 24%
#
Springer-Verlag Berlin/Heidelberg 2009
will fail to achieve good pain control over the course of their cancer [4]. After 1 month of care by pain specialists, 9% will remain in severe to maximal pain, a percentage that remains stable over 10 years of follow-up [4]. In one study of cancer patients, the 3% who had received surgical intervention were able to discontinue oral opioids, suggesting an important role for surgery in pain control for some patients [3]. Pain is a significant determinant of quality of life, and uncontrolled pain may lead to suicide [6,7]. Given the high prevalence of pain among cancer patients, especially among those with terminal diagnoses, pain control should be paramount. This chapter will provide a review of the major principles of cancer pain management for the neurosurgeon.
Barriers to Treatment of Pain Despite recent movements to increase awareness of pain among clinicians, pain control remains suboptimal [8]. Particularly when disease is incurable, the patient’s quality of life should assume central importance in the treatment plan. Unfortunately, multiple obstacles exist to appropriate treatment of pain, and are particularly acute in the surgical treatment of pain. Patient reluctance. Numerous beliefs of patients hamper their own pain control: the belief that pain is inevitable with cancer, the belief that reporting pain will distract the physician from treating or curing the cancer, the belief that medication tolerance rapidly develops, the
2062
122
Comprehensive management of cancer pain including surgery
belief that if they are bothering their doctor they are not ‘‘good’’ patients, and the belief that people inevitably become addicted to ‘‘strong’’ pain medicines [9]. In addition, patients often fear side effects of pain medication, and particularly, fear addiction to pain medications. It is worth noting that higher levels of education and patient instruction can help remove these barriers [10]. Social barriers. No matter the health care system, obtaining pain medication can be a problem for patients. In systems with socialized medicine, heavy regulation can impede physician prescription. In the United States, a patient’s insurance may limit coverage for medications (or the patient may be uninsured entirely). In developing nations, a general debility of a healthcare infrastructure can make pain medication availability erratic. Physician-related factors. A study of pain medication prescription found that the most powerful predictor of patient analgesic use was not the patient’s pain, but the providers’ pain management practices [11]. In other words, the individual clinician’s habits often dictate how much medication a patient receives, regardless of the patient’s condition. As such, physician behaviors may be the most important to examine. Inadequate assessment. Physicians often underestimate the severity of pain [12,13]. Obviously, underdiagnosis and underestimation will lead to undertreatment of pain, and thus, poor control. Unsurprisingly, the patient’s rating of his pain correlates better with pain-related outcomes than the physician’s ratings [14]. Not only is pain severity underestimated, assessment of pain phenomenology is often imprecise [15,16]. Reluctance to prescribe opioids. Physicians are often overly concerned about addiction [17] and side effects, and fail to prescribe opioids in adequate doses [18]. Ironically, addictive behaviors can develop as a result of undertreatment of pain. In addition, a significant number of physicians believe that pain is principally a marker of disease progression, rather than a condition to be
treated, and may hesitate to provide adequate analgesia to avoid masking progression [19]. Lack of training. Physicians in general may lack adequate training in pain assessment and management. In addition, surgical palliation is a relatively new concept, and may not be adequately represented in residency training. Studies of physician education indicate that caregivers with higher pain management knowledge were significantly more likely to provide appropriate analgesia [19]. Lack of familiarity with options. It is often unclear which, if any, specialty, should be responsible for expertise in pain management, with some aspects handled by internists, oncologists, anesthesiologists, palliative care physicians, general surgeons, and surgical subspecialties, including, of course, neurosurgery [20]. As a result, many primary physicians of cancer patients are not aware of, or are suspicious of, available treatments outside their specialty, denying patients possible relief and improved quality of life. Attesting to this divide, most national pain management guidelines either do not mention surgical options, or mention them only in passing without specific guidance, including the 2005 AHRQ guidelines, the French national guidelines [21] and, most significantly, the WHO cancer pain management guidelines [22]. These unfortunate omissions may be partly due to the difficulty of producing multicenter randomized controlled trials for surgical procedures [23]. However, there are published studies suggesting the efficacy of surgical procedures for treating pain, and the near-complete absence of these procedures from major guidelines clearly does not serve the patients in extreme pain who may benefit from such therapy.
Patient Assessment Not all pain is alike, nor does all of it respond to the same treatments. It is worth remembering
Comprehensive management of cancer pain including surgery
that, while the majority of pain among cancer patients is caused by the cancer or consequences of the disease, a significant percentage (17%) is caused by the antineoplastic therapy itself [24]. In assessing pain syndromes in cancer patients, several aspects of pain should be elicited. 1.
Temporality. Pain duration can be roughly divided into acute and chronic. In cancer patients, acute pain can be related to a large number of insults, including procedures, both diagnostic and therapeutic; chemotherapy or radiotherapy, in the form of mucositis, myelosuppressive pain, myalgias, among others; and damage from the cancer itself, as in pathologic fractures. Chronic pain similarly may be related to both neoplasms and attempts to treat them. Chronic pain can occur from tumor
2.
122
invasion of bone or nerve tissue. Surgical resection can result in damage to nerves or neuroma formation; amputation can produce phantom pains. Radiation can cause plexus fibrosis or myelopathy, while chemotherapy can cause painful polyneuropathies (e.g., with vincristine use). In addition, non-cancer pain syndromes can develop, especially if general debilitation occurs (> Table 122-1). Intensity. Intensity of pain is the most common means of assessing pain control. Rapid increase in pain intensity may signal increased tumor activity, including recurrence or progression. Intensity of pain also may guide pain medication dosage. The most commonly used system is the verbal analog scale, in which pain is rated on scale of 1–10, or the NRS-11, with pain rated on a
. Table 122-1 Chronic pain syndromes associated with cancer therapy Mechanism I. After surgery A. Nerve trauma B. Entrapment of nerves in scar tissue C. Amputation of limb D. After mastectomy or thoracotomy II. After radiotherapy A. Radiation fibrosis of nerves or plexuses B. Myelopathy of spinal cord
C. Peripheral nerve tumors caused by radiation III. After chemotherapy A. Vinca alkaloid–induced peripheral neuropathy B. Steroid pseudorheumatism
C. Aseptic necrosis of bone (chronic steroid treatment) D. Postherpetic neuralgia
Common sites and characteristics Neuralgic pain Peripheral nerve or dermatomal distribution Superficial wound scar hypersensitivity Localized stump pain (neuroma) or phantom pain Neuralgic pain, dysesthesia Diffuse pain 6 months to many years later lymphedema and local skin changes motor loss Brown-Se´quard syndrome (ipsilateral sensory and contralateral motor loss) Pain at level of damage or referred pain Painful enlarging mass in area of radiation along line of peripheral nerve or plexus Burning pain in hands and feet associated with symmetrical polyneuropathy Diffuse joint and muscle pain with associated tenderness to palpation but no inflammatory signs Pain resolves when steroid reinstituted Pain in knee, leg, or shoulder with limitation of movement Bone scan changes delayed after onset of pain Burning pain, dysesthesia
source: Cousins MJ. Introduction to acute and chronic pain: implications for neural blockade. In: Cousins MJ, Bridenbaugh PO, editors. Neural blockade in clinical anesthesia and management of pain. 2nd ed. Philadelphia, PA: Lippincott; 1988. p. 750
2063
2064
122 3.
4.
5.
Comprehensive management of cancer pain including surgery
0–10 scale. A number of other rating scales exist to assess intensity, including formal questionnaires, pain diaries, and other modalities [8]. Location. Regions and systems affected by the main pain syndrome show large variation depending on the site of cancer origin [24]. In one study, pain syndromes were located in the lower back (36%), abdominal region (27%), thoracic region (23%), lower limbs (21%), head (17%) and pelvic region (15%) [24]. Direct involvement of a particular site by primary or metastatic tumor also produces specific syndromes (see > Table 122-2). Quality. Additionally, different pathophysiologies produce different types of pain [24]. Careful assessment of the quality of pain directs both further work-up as well as treatment options. Generally, the distinction can be made between nociceptive pain (arising from intact signaling pathways) and neuropathic pain (arising from damaged or interrupted pathways). Pain can be classified as originating from nociceptors in bone (35%), soft tissue (45%) or visceral structures (33%) or can be of a neuropathic origin (34%) [24]. Social history. In addition to clarifying a patient’s pain complaints, other patient factors should be assessed. While pain quality (neuropathic versus nonneuropathic) and temporal characteristics (chronic versus acute) were found to affect rates of successful pain treatment, so did previous opioid dose, psychological distress, tolerance to medication, and past history of alcohol or drugs [25]. Unsurprisingly, rates of depression are higher among cancer patients than in the general population [26,27]. The patient’s psychological state in cancer pain often includes exhaustion and frustration, as well as fear and hopelessness [28]. These emotional burdens can increase pain, and may complicate treatment [29].
Completely assessing a patient’s pain can be difficult. A detailed phenomenological history is the main tool for diagnosis, and should guide the physical exam and imaging studies. The role of imaging in cancer pain is discussed elsewhere. Assessment is further complicated by the fact that patients often have multiple pain syndromes. Thirty percent of the patients presented with one, thirty-nine percent with two and thirty-one percent with three or more distinct pain syndromes [24]. However, a detailed assessment of the patient’s pain is essential in determining the etiology, ordering appropriate diagnostic studies, and executing successful treatment plans (> Table 122-3).
Pain Mechanisms in Cancer Patients As described above, cancer pain can be generally classified as nociceptive or neuropathic. Other forms of pain, including central pain and psychogenic pain, are briefly discussed. The physiology of pain is briefly discussed below; for a more detailed discussion, please see relevant Chapter (> Table 122-4).
Nociceptive Pain In non-pathological states, nociceptive receptors (nociceptors) in end organs are usually activated by thermal or mechanical stimuli, which activate unmyelineated C and small, myelineated A fibers. These fibers, via the dorsal root ganglia, send signals into spinal lamina I and V, into the spinothalamic tract, and on to the thalamus. Stimulation of nociceptors releases numerous neurotransmitters including glutamate, substance P and calcitonin gene-related peptide. However, inflammatory mediators can also activate the nociceptive system, including bradykinin, serotonin, and prostaglandins. In
Comprehensive management of cancer pain including surgery
122
. Table 122-2 Pain syndromes directly caused by cancer (primary or metastatic) Mechanism
Common sites and characteristics
I. Infiltration of bone by tumor A. Base of skull metastases 1. Jugular foramen
Dull, constant aching muscle spasm
2. Clivus 3. Sphenoid sinus 4. Orbital syndrome 5. Middle fossa B. Cavernous sinus syndrome C. Vertebral body metastases 1. Subluxation of atlas 2. C7–T1 metastases
3. L1 metastases 4. Sacral metastases
D. Metastatic fracture close to nerves II. Infiltration or compression of nerve tissue by tumor A. Cerebral metastases B. Cranial neuralgias C. Peripheral nerve ( peripheral and perivascular lymphangitis) D. Brachial plexus E. Lumbar plexus F. Sacral plexus G. Meningeal carcinomatosis H. Epidural spinal cord compression III. Obstuction of hollow viscus (e.g., gut, genitourinary tract) IV. Occlusion of arteries and veins V. Stretching of periosteum or fascia (tissues with tight investment) VI. Inflammation ulceration (necrosis and infection of tumors) VII. Soft tissue infiltration VIII. Raised intracranial pressure
Occipital pain and tenderness, exacerbated by head movement Hoarseness, dysarthria, dysphagia Vertex headache exacerbated by neck flexion Dysfunction of VII–XII cranial nerves Bifrontal headache Nasal stuffiness, diplopia Supraorbital pain Blurred vision, diplopia, ophthalmoplegia Trigeminal distribution pain, numbness, paraesthesia Headache, diplopia, dysarthria, dysphagia Frontal headache and supraorbital pain Dysfunction of III–VI cranial nerves Neck pain radiating to top of skull Progressive neurological deficit Constant paraspinal pain radiating to both shoulders Local tenderness on percussion Sensory loss and hand and triceps weakness Dull, aching pain in midback referred to sacroiliac joints Radicular pain Dull, aching low back pain exacerbated by lying or sitting Perianal sensory loss Bowel and bladder dysfunction Acute onset pain and muscle spasm
Headache, especially multiple or posterior fossa metastases Glossopharyngeal neuralgia syncope and hypotension Trigeminal neuralgia Burning constant pain in area of sensory loss dysethesia and hyperalgesia signs of sympathetic overactivity Radicular pain in shoulder and arm Horner’s syndrome (>50%) Radicular pain to anterior thigh and groin or to leg and foot Dull aching perianal pain Bowel and bladder dysfunction Constant headache neck stiffness or low back and buttock pain Severe pain locally over involved vertebral body or radicular pain Poorly localized dull or cramping pain Ischemic rest pain (skin) or claudication (muscle) or pain venous engorgement Severe localized pain (e.g., periosteum) Visceral pain (e.g., ovary) Severe localized pain (e.g., perineum) Visceral pain (e.g., cervix) Localized pain Severe headache postural changes, confusion, behavioral changes, etc.
source: Cousins MJ. Introduction to acute and chronic pain: implications for neural blockade. In: Cousins MJ, Bridenbaugh PO, editors. Neural blockade in clinical anesthesia and management of pain, 2nd ed. Philadelphia, PA: Lippincott; 1988. p. 749
2065
2066
122
Comprehensive management of cancer pain including surgery
. Table 122-3 Definition of pain terms Pain
Allodynia Hyperalgesia Hyperesthesia
Hyperpathia
Dysesthesia
An unpleasant sensory and emotional experience associated with actual or potential tissue damage or described in terms of such damage Pain caused by a stimulus that does not normally provoke pain An increased response to a stimulus that is normally painful Increased sensitivity to stimulation, excluding the special senses; may refer to various modes of cutaneous sensibility, including touch and thermal sensation without pain, as well as to pain A painful syndrome characterized by increased reaction to a stimulus, especially a repetitive stimulus, as well as increased threshold An unpleasant abnormal sensation, whether spontaneous or evoked
addition, inflammatory pain states are associated with increases in substance P, calcitonin generelated peptide, protein kinase C gamma, and substance P receptor [30]. Chronic activation of nociceptors can lead to chronic nociceptive pain. Inflammatory reactions to tumors may generate chronic pain states, and some tumor cytokines may increase DRG activity by increasing calcium channels [31], suggesting that chronic nociceptive pain in cancer patients may be both due to the body’s reaction to the tumor, as well as the tumor itself. The general therapy of these pain states may involve control of inflammation, blockade of secondary messengers, or removal of tumor, among other modalities. However, the different locations of nociceptors throughout different tissues produce different qualitative pains: somatic, visceral, and osteogenic. Osteogenic pain, (‘‘bone pain’’), is thought to be more responsive to anti-inflammatory medications, though opioids may also be effective [32–37]. For soft-tissue nociceptive pain, anti-inflammatory medication may be less effective [38]. Visceral pain signals
have been hypothesized to travel in or medial to the dorsal columns [39,40] which may form the basis for the effectiveness of procedures such as punctate midline myelotomy [39,40].
Neuropathic Pain Neuropathic pain results from damaged or dysfunctional nociceptive fibers, creating abnormal impulse generation. Damage to axons, supporting cells, or the surrounding environment can produce neuropathic changes. Unlike in nociceptive pain, neuronal damage is associated with decreases in substance P and calcitonin generelated peptide and increases in galanin and neuropeptide Y in both primary afferent neurons and the spinal cord [30]. These changes may create areas of hyperexcitability or ectopic ‘‘pacemakers’’ of unregulated firing. In addition, membrane properties of neurons may be altered. Because of their differing pathophysiologies, treatments for nociceptive pain might be expected to be poorly effective for neuropathic pain. In fact, neuropathic pain tends to respond poorly to opioids [25], and may require treatments that alter neuron and ion-channel function, including tricyclic antidepressants, anti-epileptic drugs, ketamine [41], or surgery [42,43] for alleviation.
Central Pain Central pain is a severe form of chronic pain resulting from damage to certain portions of the central nervous system, including the spinal cord, thalamus and brain stem. While the mechanisms are unclear, patients typically suffer from burning pain and dysesthesias. Reorganization of thalamic representations in response to injury may play a role in its pathogenesis. Treatment options remain very limited. Fortunately, central pain is not a common feature in cancer patients.
Comprehensive management of cancer pain including surgery
122
. Table 122-4 Pain syndromes in patients with cancer: exacerbating psychological factors Psychological factor
Possible causes
Anxiety
Sleeplessness Fear of death, loss of self-control Fear of surgical mutilation Uncontrolled pain Loss of social position and work Confused understanding of disease because of poor communication Family and financial problems Sleeplessness Loss of physical abilities Sense of helplessness Disfigurement Loss of valued social position Frustration with therapeutic failures Resentment of sickness Irritability caused by pain and general discomfort
Depression
Anger
A vicious cycle usually develops:
source: Cousins MJ. Introduction to acute and chronic pain: implications for neural blockade. In: Cousins MJ, Bridenbaugh PO, editors. Neural blockade in clinical anesthesia and management of pain, 2nd ed. Philadelphia, PA: Lippincott; 1988. p. 750
Cancer Pain Whether cancer pain represents a separate kind of pain from normal nociceptive or neuropathic pain remains controversial. Some animal models suggest some different biological features, including a lack of detectable changes in standard pain markers in either primary afferent neurons or the spinal cord, with the novel presence of massive astrocyte hypertrophy without neuronal loss, increase in the neuronal expression of c-Fos, and increase in the number of dynorphinimmunoreactive neurons [30]. Other models depict cancer pain as a particular syndrome with nociceptive and neuropathic features [44]. However, most authors do not distinguish cancer pain as an entirely distinct pathophysiologic phenomenon. Often ‘‘cancer pain’’ simply means
‘‘any pain accompanying cancer,’’ and may be nociceptive, neuropathic, or both.
Psychogenic Aspects Pain is a complex phenomenon, and while peripheral mechanisms of pain are understood in their basics, higher cortical processing of pain is poorly understood. The impact of any disease state is related to the amount of social support available to a person. This phenomenon is particularly true of pain. Patients with more complex social issues and emotional issues may undergo somatization [29], and conceptualizing pain purely as nociceptive burden may limit effective treatment. Patients suffering from cancer are at particular risk for having increasing
2067
2068
122
Comprehensive management of cancer pain including surgery
emotional, financial and social burdens, and successful pain treatment may require addressing these other aspects of cortical functioning. Pain states and emotional states are intricately linked, and while pain may induce negative affect, negative affect may also increase pain perception. The neural basis for this phenomenon is unclear; however, the prevalence of anxiety among cancer patients may make them more prone to pain. Effective treatment of pain, particularly pain with a psychogenic component, may require treatment of the underlying emotional states. Tricyclic antidepressants may be effective in alleviating psychogenic pain states [45].
carefully weighed prior to initiation. The goal of pain treatment is to improve quality of life. However, pain is not the only determinant of quality of life, and other symptoms are important to control as well, including pain medication side effects [46]. Pain control may also increase patient mobility, improve patient compliance with other therapies, and reduce morbidity. Treatment of pain in cancer patients comprises two basic strategies: tumor control and direct pain control. Each strategy can be divided into three modalities: pharmacology, radiotherapy, and surgery.
Antineoplastic Therapy
Other Causes of Pain Having cancer does not exclude other diagnoses from the differential. While neuropathic pain is a particular challenge for cancer patients, it is worth remembering that neuropathic pain can also be caused by other causes of neuropathy, including diabetes. Similarly, radicular pain may represent disc herniation, rather than vertebral metastasis. Normal causes of pain should be considered in the work-up of pain in the cancer patient.
Other Symptoms While pain is one of the most distressing symptoms affecting cancer patients, they are subject to a large variety of disabling symptoms, including fatigue, weakness, nausea, and appetite loss, all of which can contribute to, and worsen, pain [2]. Addressing these associated symptoms may be helpful in alleviating pain as well.
Treatment Options Formulating the Treatment Plan As in all medical or surgical treatments, the risks and benefits of any intervention must be
While antineoplastic therapy has an obvious role in the treatment of cancer, it may also be the most effective mode of analgesia. Even in patients with terminal diagnoses, therapy directed principally at alleviating tumor burden may provide the best available palliation. Pharmacologic means of treating tumors include a variety of chemotherapeutic regimens. Hormonal treatments may be helpful in select tumors. These will vary by tumor type, and will need to be initiated in conjunction with the patient’s oncologist. Radiotherapy directed against the tumor can play a significant role in palliative care. Numerous modalities are available, including external beam radiotherapy and radioisotope therapy [47,48], among others. Of particular interest to the neurosurgeon are methods of stereotactic radiosurgery. These are increasingly being used to treat tumors in palliative settings, in for example, tumors causing facial pain [49] or for spinal tumors [50], especially in palliative care patients, who may be without the physiologic reserve to tolerate surgery, but who may still benefit from focused antineoplastic therapy. Surgery directed at tumor removal can play an important role in maintaining quality of life. Besides its use in the management of acute emergencies (e.g., hemorrhagic tumors), surgical tumor resection without curative intent may be
Comprehensive management of cancer pain including surgery
appropriate analgesia. Tumors that impinge on crucial neurological or vascular structures may significantly alter a patient’s level of functioning. Decompression of stenosis and stabilization of the vertebral column in the setting of metastases to the spine is a common example of surgical intervention in a palliative setting [51]. Surgical treatment may also be used to treat the complications of tumor invasion, as in kyphoplasty or vertebroplasty for pathologic fracture [52].
Direct Analgesic Therapy Direct analgesic therapy can be divided into the same three modalities: pharmacologic, radiotherapeutic, and surgical. While developments in radiosurgery and neurosurgery are providing more options of which the clinician should be aware, pharmacologic intervention remains the mainstay of analgesic therapy. While a large variety of pain medications exist, opioids are the principal agents employed (> Table 122-5). Several models for pain control algorithms have been proposed, the most predominant of which is the World Health Organization (WHO) Model. The basic principles of the WHO model are (1) oral medications are to be used when possible, (2) background analgesia should be provided with scheduled dosing and additional medication for breakthrough pain (3) treatment . Table 122-5 Equianalgesic narcotic ratios Analgesic Morphine to methadone Morphine to hydromorphone Morphine to oxycodone
Parenteral conversion
Oral conversion
1:1
3:1
6.67:1
4:1
0.7:1
1:1
Source: Adapted from > Table 122-3, Anderson 2001 (Anderson, 2001:222). Initial conversion dose is usually 50–75% of equianalgesic dose, with additional doses provided for breakthrough pain
122
efficacy should be assessed at regular intervals, and (4) pain medication adjustment should be according to the WHO ‘‘ladder.’’ The WHO ladder has three steps (see > Figure 122-1). Step 1 begins with nonopioid analgesia, including non-steroidal anti-inflammatory drugs (NSAIDs) and acetaminophen. If non-opioids fail to control pain, Step 2 includes ‘‘weak’’ opioids, such as codeine and oxycodone, and partial opioid agonists, like buprenorphine. Step 3, for those whose pain remains uncontrolled, involves strong opioids. Morphine is the most common example of a strong opioid. Because pure opioid agonists have no ceiling for their analgesic effect, doses may be increased indefinitely. In the WHO model, no further steps were developed. At each step, adjuvant therapy may be considered for other symptoms, including treatment of opioid side effects, including anti-emetics, laxatives, and anxiolytics. The WHO stepladder has been validated, with strict adherence to its guidelines providing good or satisfactory pain control in 88% of patients [4,53]. However, several modifications and criticisms have been proposed. The data on so-called ‘‘weak’’ and partial opioids does not definitively support their use [54] and, thus, a simple 2-step model of the WHO guidelines has been advocated [55]. If pain cannot be controlled with simple antiinflammatory medication, pain control can be achieved more quickly by directly starting strong opioids. This strategy may decrease days of significant pain by as much as 20%, though it will also increase rates of opioid side effects [55]. Other authors have proposed adding a fourth step to the ladder for interventional pain control procedures [56]. While opioids may have no theoretical dose limit for analgesic effect, the presence of side-effects and adverse reactions may create maximum tolerable doses. In these cases, which may occur as frequently as 25% of the time, surgical or other intervention may be indicated to achieve optimal pain control (see > Figure 122-1).
2069
2070
122
Comprehensive management of cancer pain including surgery
. Figure 122-1 From Miguel, 2000. {Miguel, 2000 #67} fourth step (b).
> Figure
1A-B. Three step WHO Cancer pain stepladder (a), with proposed
Finally, it is worth noting that the WHO guidelines emphasize pain intensity over pathophysiology, as is appropriate given our limited understanding of pain mechanisms. However, as our understanding of the pathophysiology of pain increases, specific therapies for particular kinds of pain may be developed. Nociceptive pain and neuropathic pain appear to respond to different treatments, with neuropathic pain being somewhat more resistant to opioid therapy. Similarly, osteogenic pain may also have unique treatments. In the future, we may anticipate that the management of pain may become more tailored to specific pain phenomena.
Classes of Medications While the primary agents of analgesia are opioids, many other drugs can be used. Some of the major classes of analgesics are described
below. Drugs not part of the WHO ladder are considered ‘‘adjuvant.’’ They may be considered in cases where the primary analgesics have limited efficacy, restrictive side effects, or simply where they may provide more effective analgesia based on the patient’s particular pain syndrome [34]. In general, these drugs can be safe and effective in combination, though combined use does not always show a clear direct benefit [57,58]. It is worth noting that increasing the number of drugs can create drug interactions, complicate dosing, and may harm patients [59]. Non-steroidal anti-inflammatory drugs (NSAIDs) (WHO Step 1). Aspirin and other NSAIDs decrease pain by inhibiting synthesis of the inflammatory mediator, prostaglandin, via interfering with the enzyme cyclooxygenase. As noted above, the inflammatory cascade plays an important role in nociceptive pain, both activating and sensitizing nociceptors. Prostaglandins may be particularly important in neoplastic bone
Comprehensive management of cancer pain including surgery
nociception [37]. NSAIDs, particularly ketorolac and diclofenac, are effective agents in treating nociceptive pain, and, potentially, in bone nociception [58,60–63]. NSAIDs have well known side effects limiting their use, including gastrointestinal ulceration, impaired platelet function, and nephrotoxicity. In addition, supratherapeutic doses of NSAIDs do not increase their analgesic effect. However, they may be given in combination with opioids to improve analgesia. Acetaminophen/paracetomol (WHO Step 1). Acetaminophen has similar antipyretic and analgesic activity to NSAIDs, but has less antiinflammatory activity than NSAIDs. Its exact mechanism of action is unknown. Peripheral prostaglandin inhibition may play a role, though acetaminophen also has central effects [64,65]. Hepatotoxicity limits its dosing to <4 g/day. Acetaminophen may be given per rectum if the oral route is unavailable. Around the clock dosing of acetaminophen may provide effective background analgesia. Acetaminophen may also be given alone or in combination with opioids in the treatment of cancer pain [66]. Corticosteroids (Adjuvant). Corticosteroids do not have a definite role as analgesic agents in the treatment of cancer patient. Addition of oral dexamethasone to opioid regimens has been shown to provide no benefit [67]. However, intrathecal beclamethasone has been noted to reduce pain, particularly in the setting of vertebral metastasis [68,69]. Corticosteroids may also play a role in preventing moderate-to-severe emesis after chemotherapy [70]. Tricyclic antidepressants (Adjuvant). Tricyclic antidepressants (TCAs) have multiple actions at different sites; centrally they block the re-uptake of both serotonin and norepinephrine, while peripherally they block sodium ion channels. In addition, they may also interact with opioid receptors [71]. Amytriptyline has been most commonly used, and is a first line agent for treating neuropathic pain [72]. Their analgesic effect is thought to be separate from their
122
psychotropic effects [73]. They have shown very good efficacy in treating pain syndromes associated with psychiatric conditions, as well as for neuropathic pain [74]. It is likely that TCAs are underprescribed among cancer patients [75]. Further study of these drugs in cancer patients in warranted. Their use in cancer patients should be considered in patients with co-morbid psychiatric conditions, including depression, and in patients with significant neuropathic pain, which may be resistant to opioid therapy. Anti-epileptic drugs (Adjuvant). The most studied anti-epileptic drug for treatment of cancer pain is gabapentin. Gabapentin leads to increased GABA activity, via non-NMDA receptor antagonism and binding to the alpha-2-delta subunit of the voltage dependent calcium channel, which inhibits the release of excitatory neurotransmitters [76]. Both when added to opioids and used alone, it has shown significant efficacy in treating neuropathic pain as well as allodynia [77,78]. It should be used as a first line agent for treating neuropathic pain [72]. Gabapentin likely remains underprescribed among cancer patients [75]. Use of other anti-epileptic drugs in cancer patients has not been adequately studied. Bisphosphonates (Adjuvant). Bisphosphonates inhibit osteoclast activity, and appear to significantly decrease nociceptive bone pain in the setting of bony metastases [33]. However, they have also been associated with osteonecrosis of the jaw [79,80]. Weak opiates (WHO Step 2). While opiates can be divided into ‘‘weak’’ and ‘‘strong,’’ the mechanism of action of these drugs is similar (see below). Strong opioids are pure opioid agonists, while weak opiates have mixed actions. The list of weak opiates includes partial agonists (e.g., buprenorphine), combination agonistanagonists, (e.g., pentazocine and butorphanol), mixed opioid/neurotransmitter modulators (e.g., tramadol) and premetabolites (e.g., codeine) [81]. Hydrocodone is derived from both codeine and thebaine (the precursor to oxycodone). Given
2071
2072
122
Comprehensive management of cancer pain including surgery
the mixed pharmacologic profiles of weak opiates, they are thought to have maximum analgesic doses, i.e., doses above which no further analgesia is achieved. Pure opioid agonists (strong opiates) do not. In addition, weak opiates are often combined with NSAIDs or acetaminophen, which have dose-limiting toxicities. The role of weak opiates in cancer pain management has been poorly studied [54]. Some have argued while the evidence for weak opiates may not support their use, tramadol merits further consideration [82]. In studies comparing weak opiates, pentazocine was found to be less effective, while other weak opiates had relatively good efficacy [81,83,84]. Side effects are similar as for strong opioids, including nausea, vomiting, constipation, confusion, and respiratory depression. However, some weak opiates have novel adverse effects; for example, pentazocine can induce neuropathy [85]. Strong opioids (WHO Step 3). Strong opioids represent the most commonly prescribed analgesic in cancer pain, used on almost 50% of treatment days [4]. Common opioids include morphine, oxycodone, hydromorphone and fentanyl. The three main classes of opioid receptor (m, ∂, K) are G-protein coupled, and when activated, result in neuron hyperpolarization and decreased neurotransmitter release. Opioid receptors are located throughout pain signaling pathways. K and m receptors are located in the dorsal gray of the spinal cord, which are a significant junction in pain processing. The neurons of the dorsal spinal cord both receive input from incoming A and C fibers, as well as neuromodulatory input from supraspinal noradrenergic, serotonergic and cholinergic nuclei. These brainstem nuclei, including the ventral medulla and the nucleus raphe magnus, also contain opioid receptors. Supratentorially, opioid receptors are located in other key areas of pain regulation, including the limbic system (amygdala, hypothalamus, nucleus accumbens), the periaqueductal grey and the thalamus. While low doses of opioids effectively control pain in opioid-naı¨ve patients, tolerance to
their analgesic effects can develop. Four patterns of opioid use commonly occur in cancer patients: (1) discontinuation or reduction of opioids after antineoplastic therapy or analgesic procedure; (2) stable doses over long periods; (3) gradual development of tolerance, with increasing doses providing effective analgesia; or (4) rapid escalation of opioid dose. Rapid escalation of opioid dose in the setting of increasing pain does not correlate with addictive behavior [86]; a work-up for disease progression should be undertaken and an alternative pain regimen should be considered. In the setting of tolerance, opioid doses may be increased indefinitely, limited only by the development of side effects. Because the relationship between dose and analgesic effect is not linear, large increases in doses may be needed to achieve analgesia. Tolerance to one opioid does not imply tolerance to other opioids. Switching to another opioid may provide good analgesia at relatively lower doses. However, conversion rates between different opioids are highly variable. Individual drugs have different potency in different individuals, which preclinical studies suggest may be due to differential expression of the m receptor [87]. Further complicating the transition between drugs, opioid conversion ratios appear to vary with the amount given [88]. Thus, a conversion ratio for low dose opioids may not apply at higher doses of the same opioid. In addition, conversation ratios vary in the acute and chronic setting [89]. Opioid conversion should begin very conservatively and then proceed under close monitoring to avoid either prolonged ineffective analgesia or opioid overdose. If another opioid also fails to provide analgesia, changing the route of administration or a neurosurgical procedure should be considered. Opioid side effects are usually temporary, as tolerance to side effects develops much more rapidly than to analgesia. 90% develop tolerance to sedation within 7 days [1]. Nausea and sedation similarly abate. However, constipation usually
Comprehensive management of cancer pain including surgery
remains a significant problem, and often requires aggressive use of laxatives and stool softeners [90,91]. Peripherally acting opioid antagonists are being studied to counteract opioid-induced bowel dysfunction while preserving analgesia [92]. In addition, chronic opioid use has been associated with significant neurotoxicity. Opioid-induced neurotoxicity is an increasingly recognized phenomenon, whose cardinal features are delirium, hallucinosis, myoclonus, seizures and paradoxical hyperalgesia [1,93–95]. This has been most commonly associated with meperidine, which has toxic metabolites, including MPTP and norpethidine [96,97]. While meperidine’s toxicity is well-known, it is still in use, and continues to demonstrate significant adverse effects [98]. However, opioid induced toxicity is not restricted to meperidine, and may occur with use of other opioids. This phenomenon may be more pronounced in the elderly [99]. Current recommendations for treating opioid induced toxicity are to use an opioid without active metabolites, such as fentanyl or methadone, or to consider surgical treatment of pain. Despite the possibility of these adverse effects, opioids remain a very effective and relatively safe mode of analgesia. One of the distinct advantages of opioids is the great variety of routes for administration. 80% of cancer patients will be unable to tolerate oral medication at some point during their illness [100]. Opioids are variously available in oral, intravenous, sublingual, transdermal, transmucosal, inhaled, intramuscular, subcutaneous, and rectal formulations. In addition, surgical procedures may provide direct opioid access to the central nervous system (see below). As such, patients should always be able to have access to adequate pain medication.
Surgical Procedures for Pain While the indications for the various procedures for pain control have not been fully elaborated,
122
certainly it is reasonable to consider an interventional approach when more conservative methods of analgesia have failed. Interventions for pain control can be divided into three categories: 1. 2. 3.
Neuraxial drug delivery Neurolytic therapy Neural electromodulation
These methods are discussed in detail in their respective chapters; as such, a brief overview is presented here.
Neuraxial Drug Delivery The therapeutic effect of opioids occurs largely within the central nervous system, while the peripheral effects are largely adverse. The cerebral pharmacology of opiates is understudied, but likely is a significant determinant of opiate efficacy [101]. By administering analgesics closer to their target receptors, side effects can be minimized, effectively widening the therapeutic index of a particular drug. The three routes to direct neuraxial delivery are (1) epidural, (2) lumbar intrathecal or subarachnoid, and (3) intraventricular. Epidural administration differs from the other three routes in that the blood brain barrier is not bypassed, and will usually require significantly higher dosing than intradural delivery. Several different catheter types are available, which vary according to implantation technique (percutaneous vs surgical placement) and rate of medication delivery (patient-controlled, fixed, or programmable). Most catheters for long-term use require surgical implantation. The most commonly infused drugs are morphine and hydromorphone. If needed, especially for neuropathic pain, the opioid may be accompanied by bupivicaine or clonidine. Other drugs may be used, including fentanyl and ziconotide [102]. Neuraxial pain relief can provide effective analgesia in patients for whom non-invasive
2073
2074
122
Comprehensive management of cancer pain including surgery
analgesia has failed [103]. In one study, severe pain decreased from 86 to 17% with no difference between the intrathecal and epidural groups. In addition, oral opioid doses were reduced by over 50%, and cognitive side-effects were significantly decreased [104]. Reviews of data from uncontrolled studies report excellent pain relief in the large majority of patients receiving intrathecal and epidural therapy. Epidural and lumbar intrathecal infusion was associated with greater peripheral side effects than ventricular infusion, including nausea, urinary retention, pruritus, and constipation. Central side effects, including respiratory depression, sedation and confusion, were more common with intraventricular delivery [105]. One infrequent but notable complication of high dose intrathecal opiates is the formation of a inflammatory granuloma, which may require surgical resection to avoid cord compression [106]. Cost modeling predicts that total cost of therapy through surgical implant breaks even with oral therapy by 2 years; however, pain control must be the primary indication [107]. Other factors involved in choosing a route for drug delivery include the length of expected therapy, the presence of spinal metastases, and patient preference. Complete analgesia can be obtained without significant neurological changes or side-effects, and tolerance is less marked than with parenteral opiates. Neuraxial delivery of analgesia is appropriate for long term use, either in a health-care facility or at home, and is generally safe [108,109].
Neurolytic Therapy Neurolytic therapy includes interventions in both the peripheral and central nervous system. The major types of neurolytic techniques are nerve blocks, cordotomies, and cerebral ablations. Neurolytic methods in the peripheral nervous system include chemical ablation, radiofrequency thermal ablation, radiosurgery, and
rarely, sharp section. The most common procedure is regional nerve blockade, or nerve block, which involves local injection of an anesthetic substance into or near nerves. Like most procedure, efficacy is operator-dependant, and complete blocks provide more permanent analgesia [110]. Using imaging or anatomical guidance, an alcohol or phenol solution is injected into the nerve or plexus. Nerve blocks are commonly used for visceral pain from invasive abdominal tumors. Sites for nerve blocks include the celiac plexus, inferior mesenteric plexus, superior hypogastric plexus, the stellate ganglion, paravertebral nerves, and, in some cases, the brachial plexus [111–113]. A common and well-studied nerve block is the celiac plexus block, which is used for pancreatic, hepatic, gastric, or colon cancer [114–117]. Some data suggests celiac blocks, compared to oral morphine alone, significantly decrease pain, reduce morphine consumption, improve performance scores, reduce side effects, and prolong survival [118,119]. A related procedure is the thorascopic splanchnicectomy with sympathectomy, which has similar efficacy in pain control for visceral pain [120,121]. Sympathectomy can also be performed with radiofrequency ablation and phenol injection [122]. Complications of these procedures include orthostatic hypotension, diarrhea, and rarely, retroperitoneal hemorrhage and paralysis [123]. Other peripheral sites of neurolysis include nerves in the head and neck (e.g., trigeminal nerve), dorsal spinal roots (e.g., DREZ procedures), and others [124]. Neuroablative procedures are also performed in the central nervous system. Spinothalamic cordotomy can be performed to control pain in when other less invasive therapies fail. The aim of the surgery is to unilaterally ablate the spinothalamic tract, usually in the cervical or upper thoracic region. While ablation of tracts may interrupt pain signaling, thalamic blood flow appears to alter after cordotomy, suggesting more complicated mechanisms of analgesia [125]. Percutaneous cordotomy may be performed, as high as C1-C2,
Comprehensive management of cancer pain including surgery
and may be CT-guided [126]. On rare occasions, bilateral cordotomy may be performed [127]. In one study, good results (pain rating <3) were obtained in 95% of patients immediately following cordotomy [128]. In another study, 83% had a reduction in pain such that their dose of opioid could be at least halved, and 38% were able to stop opioids completely [129]. While new pain may develop in one-third of patients, this pain is typically much more responsive to medication [130]. Midline myelotomy, a similar procedure, has been used to ablate the decussating spinothalamic tracts [131–133]. Intracranial ablative procedures for pain may also be performed. Stereotactic mesencephalic tractotomy provides relief from pain in up to 80% of patients, but carries a mortality rate between 0.5 and 1.8%, as well as significant morbidity (up to 10% rate of anesthesia dolorosa) [134,135]. Hypophysectomy has been used to treat osteogenic pain, via a variety of methods, including stereotactic radiosurgery [136,137]. Thalamotomy has been used in the past to control cancer pain, and occasionally is still carried out with newer methods, including stereotactic LINAC, Gamma Knife, and Cyberknife radiosurgery [138–140].
Electrical Stimulation Manipulation of neural signaling may be accomplished by electrical as well as chemical means. Numerous modes of electrical intervention exist. Transcutaneous electrical nerve stimulation has been used for many years; though it may not be more effective than placebo, it may still reduce opioid use [141]. Spinal cord stimulation, in small studies of cancer patients, improves pain scores, reduces medication use, and improves gait [142]. Randomized trials support the use of spinal cord stimulation in failed back syndrome and complex regional pain syndrome [143]. Its role in cancer pain is still to be defined. Motor
122
cortex stimulation has been used for neuropathic and deafferentation pain, suggesting a potential role in cancer pain [144–146]. Applications of deep brain stimulation are being developed for numerous neuropsychiatric pathologies, and while one of its earliest uses was in pain, it has yet to play a significant role in treatment of cancer pain [147,148]. Electrical stimulation is a promising, but as yet underresearched, area of treatment for cancer pain.
Conclusions Pain in the setting of malignant disease is a multifactorial entity. Undertreatment of cancer pain is, unfortunately, all too common. Comprehensive treatment requires familiarity with numerous syndromes and pathophysiological mechanisms. Despite these potential complexities, there are many effective therapies for cancer pain, both surgical and non-surgical. Familiarity with the basic principles of cancer pain treatment can help neurosurgeons provide optimal care for this challenging patient population.
References 1. Bruera E, Kim HN. Cancer pain. JAMA 2003;290:2476-9. 2. Teunissen SC, Wesker W, Kruitwagen C, de Haes HC, Voest EE, de Graeff A. Symptom prevalence in patients with incurable cancer: a systematic review. J Pain Symptom Manage 2007;34:94-104. 3. Grond S, Zech D, Schug SA, Lynch J, Lehmann KA. Validation of World Health Organization guidelines for cancer pain relief during the last days and hours of life. J Pain Symptom Manage 1991;6:411-22. 4. Zech DF, Grond S, Lynch J, Hertel D, Lehmann KA. Validation of World Health Organization Guidelines for cancer pain relief: a 10-year prospective study. Pain 1995;63:65-76. 5. Dahl JL. Pain: impediments and suggestions for solutions. J Natl Cancer Inst Monogr 2004;124–6. 6. Filiberti A, Ripamonti C, Totis A, . et al. Characteristics of terminal cancer patients who committed suicide during a home palliative care program. J Pain Symptom Manage 2001;22:544-53.
2075
2076
122
Comprehensive management of cancer pain including surgery
7. Ripamonti C, Filiberti A, Totis A, De Conno F, Tamburini M. Suicide among patients with cancer cared for at home by palliative-care teams. Lancet 1999;354:1877-8. 8. Dgvis MP, Walsh D. Cancer pain: how to measure the fifth vital sign. Cleve Clin J Med 2004;71:625-32. 9. Abbas SQ, Abbas Z. Is opiate compliance a problem in cancer pain? A survey of health-care professionals’ views. Int J Palliat Nurs 2003;9:56-63. 10. Gunnarsdottir S, Serlin RC, Ward S. Patient-related barriers to pain management: the Icelandic Barriers Questionnaire II. J Pain Symptom Manage 2005;29:273-85. 11. Dawson R, Sellers DE, Spross JA, Jablonski ES, Hoyer DR, Solomon MZ. Do patients’ beliefs act as barriers to effective pain management behaviors and outcomes in patients with cancer-related or noncancer-related pain? Oncol Nurs Forum 2005;32:363-74. 12. Staton LJ, Panda M, Chen I, et al. When race matters: disagreement in pain perception between patients and their physicians in primary care. J Natl Med Assoc 2007;99:532-8. 13. Cleeland CS. The measurement of pain from metastatic bone disease: capturing the patient’s experience. Clin Cancer Res 2006;12:6236s-6242s. 14. Reis S, Borkan J, Vanraalte R, Tamir A, Dahan R, Hermoni D. The LBP patient perception scale: a new predictor of LBP episode outcomes among primary care patients. Patient Educ Couns 2007;67:191-5. 15. Bogefeldt J, Grunnesjo M, Svardsudd K, Blomberg S. Diagnostic differences between general practitioners and orthopaedic surgeons in low back pain patients. Ups J Med Sci 2007;112:199-212. 16. Horowitz SH. The diagnostic workup of patients with neuropathic pain. Med Clin North Am 2007;91:21-30. 17. Aranda S, Yates P, Edwards H, Nash R, Skerman H, McCarthy A. Barriers to effective cancer pain management: a survey of Australian family caregivers. Eur J Cancer Care (Engl) 2004;13:336-43. 18. Jacobsen R, Sjogren P, Moldrup C, Christrup L. Physician-related barriers to cancer pain management with opioid analgesics: a systematic review. J Opioid Manag 2007;3:207-14. 19. Vallerand AH, Collins-Bohler D, Templin T, Hasenau SM. Knowledge of and barriers to pain management in caregivers of cancer patients receiving homecare. Cancer Nurs 2007;30:31-7. 20. Hargus EP. Aggressive cancer pain management – where does the buck stop? Conn Med 1992;56:603-6. 21. Krakowski I, Theobald S, Balp L, et al. Summary version of the Standards, Options and Recommendations for the use of analgesia for the treatment of nociceptive pain in adults with cancer (update 2002). Br J Cancer 2003;89 Suppl 1:S67-S72. 22. Ventafridda V, Saita L, Ripamonti C, De Conno F. WHO guidelines for the use of analgesics in cancer pain. Int J Tissue React 1985;7:93-6.
23. Cherny NI, Arbit E, Jain S. Invasive techniques in the management of cancer pain. Hematol Oncol Clin North Am 1996;10:121-37. 24. Grond S, Zech D, Diefenbach C, Radbruch L, Lehmann KA. Assessment of cancer pain: a prospective evaluation in 2266 cancer patients referred to a pain service. Pain 1996;64:107-14. 25. Bruera E, Schoeller T, Wenk R, et al. A prospective multicenter assessment of the Edmonton staging system for cancer pain. J Pain Symptom Manage 1995;10:348-55. 26. Gagliese L, Gauthier LR, Rodin G. Cancer pain and depression: a systematic review of age-related patterns. Pain Res Manag 2007;12:205-11. 27. Rodin G, Katz M, Lloyd N, Green E, Mackay JA, Wong RK. Treatment of depression in cancer patients. Curr Oncol 2007;14:180-8. 28. Sela RA, Bruera E, Conner-spady B, Cumming C, Walker C. Sensory and affective dimensions of advanced cancer pain. Psychooncology 2002;11:23-34. 29. Strasser F, Walker P, Bruera E. Palliative pain management: when both pain and suffering hurt. J Palliat Care 2005;21:69-79. 30. Honore P, Rogers SD, Schwei MJ, et al. Murine models of inflammatory, neuropathic and cancer pain each generates a unique set of neurochemical changes in the spinal cord and sensory neurons. Neuroscience 2000;98:585-98. 31. Khasabova IA, Stucky CL, Harding-Rose C, et al. Chemical interactions between fibrosarcoma cancer cells and sensory neurons contribute to cancer pain. J Neurosci 2007;27:10289-98. 32. Davis MP, Walsh D, Lagman R, LeGrand SB. Controversies in pharmacotherapy of pain management. Lancet Oncol 2005;6:696-704. 33. Gralow J, Tripathy D. Managing metastatic bone pain: the role of bisphosphonates. J Pain Symptom Manage 2007;33:462-72. 34. Knotkova H, Pappagallo M. Adjuvant analgesics. Med Clin North Am 2007;91:113-24. 35. Miaskowski C, Mack KA, Dodd M, et al. Oncology outpatients with pain from bone metastasis require more than around-the-clock dosing of analgesics to achieve adequate pain control. J Pain 2002;3:12-20. 36. Mouedden ME, Meert TF. Pharmacological evaluation of opioid and non-opioid analgesics in a murine bone cancer model of pain. Pharmacol Biochem Behav 2007;86:458-67. 37. Payne R. Mechanisms and management of bone pain. Cancer 1997;80:1608-13. 38. Collins D, Penman I, Mishra G, Draganov P. EUSguided celiac block and neurolysis. Endoscopy 2006;38:935-9. 39. Willis WD Jr, Westlund KN. The role of the dorsal column pathway in visceral nociception. Curr Pain Headache Rep 2001;5:20-6.
Comprehensive management of cancer pain including surgery
40. Willis WD, Al-Chaer ED, Quast MJ, Westlund KN. A visceral pain pathway in the dorsal column of the spinal cord. Proc Natl Acad Sci USA 1999;96:7675-9. 41. Tarumi Y, Watanabe S, Bruera E, Ishitani K. High-dose ketamine in the management of cancer-related neuropathic pain. J Pain Symptom Manage 2000;19:405-7. 42. Clubb B. Management of neuropathic pain following treatment for breast cancer in the absence of recurrence: a challenge for the radiation oncologist. Australas Radiol 2004;48:459-65. 43. Grond S, Radbruch L, Meuser T, Sabatowski R, Loick G, Lehmann KA. Assessment and treatment of neuropathic cancer pain following WHO guidelines. Pain 1999;79:15-20. 44. Cain DM, Wacnik PW, Eikmeier L, Beitz A, Wilcox GL, Simone DA. Functional interactions between tumor and peripheral nerve in a model of cancer pain in the mouse. Pain Med 2001;2:15-23. 45. Jann MW, Slade JH. Antidepressant agents for the treatment of chronic pain and depression. Pharmacotherapy 2007;27:1571-87. 46. Grond S, Zech D, Diefenbach C, Bischoff A. Prevalence and pattern of symptoms in patients with cancer pain: a prospective evaluation of 1635 cancer patients referred to a pain clinic. J Pain Symptom Manage 1994;9:372-82. 47. Rockhill JK. Advances in radiation therapy for oncologic pain. Curr Pain Headache Rep 2007;11:270-5. 48. Bauman G, Charette M, Reid R, Sathya J. Radiopharmaceuticals for the palliation of painful bone metastasis-a systemic review. Radiother Oncol 2005;75:258-70. 49. Chang JW, Kim SH, Huh R, Park YG, Chung SS. The effects of stereotactic radiosurgery on secondary facial pain. Stereotact Funct Neurosurg 1999;72 Suppl 1:29-37. 50. Degen JW, Gagnon GJ, Voyadzis JM, et al. CyberKnife stereotactic radiosurgical treatment of spinal tumors for pain control and quality of life. J Neurosurg Spine;2005;2:540-9. 51. Cappelletto B, Del Fabro P, Meo A. Decompression and surgical stabilization in the palliative treatment of vertebral metastases. Chir Organi Mov 1998;83:167-76. 52. Burton AW, Reddy SK, Shah HN, Tremont-Lukats I, Mendel E. Percutaneous vertebroplasty–a technique to treat refractory spinal pain in the setting of advanced metastatic cancer: a case series. J Pain Symptom Manage 2005;30:87-95. 53. Grond S, Zech D, Lynch J, Diefenbach C, Schug SA, Lehmann KA. Validation of World Health Organization guidelines for pain relief in head and neck cancer. A prospective study. Ann Otol Rhinol Laryngol 1993;102:342-8. 54. Weak opiate analgesics: modest practical merits. Prescrire Int 2004; 13:22-5. 55. Maltoni M, Scarpi E, Modonesi C, et al. A validation study of the WHO analgesic ladder: a two-step vs. threestep strategy. Support Care Cancer 2005;13:888-94.
122
56. Miguel R. Interventional treatment of cancer pain: the fourth step in the World Health Organization analgesic ladder? Cancer Control 2000;7:149-56. 57. Grond S, Zech D, Schug SA, Lynch J, Lehmann KA. The importance of non-opioid analgesics for cancer pain relief according to the guidelines of the World Health Organization. Int J Clin Pharmacol Res 1991;11:253-60. 58. Jenkins CA, Bruera E. Nonsteroidal anti-inflammatory drugs as adjuvant analgesics in cancer patients. Palliat Med 1999;13:183-96. 59. Bernard SA, Bruera E. Drug interactions in palliative care. J Clin Oncol 2000;18:1780-99. 60. Pannuti F, Robustelli della Cuna G, Ventaffrida V, Strocchi E, Camaggi CM. A double-blind evaluation of the analgesic efficacy and toxicity of oral ketorolac and diclofenac in cancer pain. The TD/10 recordati Protocol Study Group. Tumori 1999;85:96-100. 61. Toscani F, Piva L, Corli O, et al. Ketorolac versus diclofenac sodium in cancer pain. Arzneimittelforschung 1994;44:550-4. 62. Carlson RW, Borrison RA, Sher HB, Eisenberg PD, Mowry PA, Wolin EM. A multiinstitutional evaluation of the analgesic efficacy and safety of ketorolac tromethamine, acetaminophen plus codeine, and placebo in cancer pain. Pharmacotherapy 1990;10:211-16. 63. Gordon RL. Prolonged central intravenous ketorolac continuous infusion in a cancer patient with intractable bone pain. Ann Pharmacother 1998;32:193-6. 64. Stockler M, Vardy J, Pillai A, Warr D. Acetaminophen (paracetamol) improves pain and well-being in people with advanced cancer already receiving a strong opioid regimen: a randomized, double-blind, placebo-controlled cross-over trial. J Clin Oncol 2004;22:3389-94. 65. Pickering G, Loriot MA, Libert F, Eschalier A, Beaune P, Dubray C. Analgesic effect of acetaminophen in humans: first evidence of a central serotonergic mechanism. Clin Pharmacol Ther 2006;79:371-8. 66. Hardy J, Reymond E, Charles M. Acetaminophen in cancer pain. J Clin Oncol 2005;23:1586; author reply 186–7. 67. Mercadante SL, Berchovich M, Casuccio A, Fulfaro F, Mangione S. A prospective randomized study of corticosteroids as adjuvant drugs to opioids in advanced cancer patients. Am J Hosp Palliat Care 2007;24:13-19. 68. Taguchi H, Oishi K, Sakamoto S, Shingu K. Intrathecal betamethasone for cancer pain in the lower half of the body: a study of its analgesic efficacy and safety. Br J Anaesth 2007;98:385-9. 69. Inada T, Kushida A, Sakamoto S, Taguchi H, Shingu K. Intrathecal betamethasone pain relief in cancer patients with vertebral metastasis: a pilot study. Acta Anaesthesiol Scand 2007;51:490-4. 70. Vardy J, Chiew KS, Galica J, Pond GR, Tannock IF. Side effects associated with the use of dexamethasone for prophylaxis of delayed emesis after moderately emetogenic chemotherapy. Br J Cancer 2006;94:1011-15.
2077
2078
122
Comprehensive management of cancer pain including surgery
71. Benbouzid M, Gaveriaux-Ruff C, Yalcin I, et al. Deltaopioid receptors are critical for tricyclic antidepressant treatment of neuropathic allodynia. Biol Psychiatry 2008;63(6):633-6. 72. Dworkin RH, O’Connor AB, Backonja M, et al. Pharmacologic management of neuropathic pain: evidencebased recommendations. Pain 2007;132:237-51. 73. Panerai AE, Bianchi M, Sacerdote P, Ripamonti C, Ventafridda V, De Conno F. Antidepressants in cancer pain. J Palliat Care 1991;7:42-4. 74. Bryson HM, Wilde MI. Amitriptyline. A review of its pharmacological properties and therapeutic use in chronic pain states. Drugs Aging 1996;8:459-76. 75. Berger A, Dukes E, Mercadante S, Oster G. Use of antiepileptics and tricyclic antidepressants in cancer patients with neuropathic pain. Eur J Cancer Care (Engl) 2006;15:138-45. 76. Bennett MI, Simpson KH. Gabapentin in the treatment of neuropathic pain. Palliat Med 2004;18:5-11. 77. Keskinbora K, Pekel AF, Aydinli I. Gabapentin and an opioid combination versus opioid alone for the management of neuropathic cancer pain: a randomized open trial. J Pain Symptom Manage 2007;34:183-9. 78. Ross JR, Goller K, Hardy J, et al. Gabapentin is effective in the treatment of cancer-related neuropathic pain: a prospective, open-label study. J Palliat Med 2005;8:1118-26. 79. Malden NJ, Pai AY. Oral bisphosphonate associated osteonecrosis of the jaws: three case reports. Br Dent J 2007;203:93-7. 80. Gutta R, Louis PJ. Bisphosphonates and osteonecrosis of the jaws: science and rationale. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2007;104:186-93. 81. De Conno F, Ripamonti C, Sbanotto A, et al. A clinical study on the use of codeine, oxycodone, dextropropoxyphene, buprenorphine, and pentazocine in cancer pain. J Pain Symptom Manage 1991;6:423-7. 82. Leppert W, Luczak J. The role of tramadol in cancer pain treatment – a review. Support Care Cancer 2005;13:5-17. 83. Rodriguez RF, Castillo JM, Del Pilar Castillo M, et al. Codeine/acetaminophen and hydrocodone/acetaminophen combination tablets for the management of chronic cancer pain in adults: a 23-day, prospective, double-blind, randomized, parallel-group study. Clin Ther 2007;29:581-7. 84. Rodriguez RF, Bravo LE, Castro F, et al. Incidence of weak opioids adverse events in the management of cancer pain: a double-blind comparative trial. J Palliat Med 2007;10:56-60. 85. Sinsawaiwong S, Phanthumchinda K. Pentazocineinduced fibrous myopathy and localized neuropathy. J Med Assoc Thai 1998;81:717-21. 86. Lowe SS, Nekolaichuk CL, Fainsinger RL, Lawlor PG. Should the rate of opioid dose escalation be included as a feature in a cancer pain classification system? J Pain Symptom Manage 2008 Jan; 35(1):51–57.
87. Nagashima M, Katoh R, Sato Y, Tagami M, Kasai S, Ikeda K. Is there genetic polymorphism evidence for individual human sensitivity to opiates? Curr Pain Headache Rep 2007;11:115-23. 88. Anderson R, Saiers JH, Abram S, Schlicht C. Accuracy in equianalgesic dosing. conversion dilemmas. J Pain Symptom Manage 2001;21:397-406. 89. Patanwala AE, Duby J, Waters D, Erstad BL. Opioid conversions in acute care. Ann Pharmacother 2007;41: 255-66. 90. Panchal SJ, Muller-Schwefe P, Wurzelmann JI. Opioidinduced bowel dysfunction: prevalence, pathophysiology and burden. Int J Clin Pract 2007;61:1181-7. 91. Max EK, Hernandez JJ, Sturpe DA, Zuckerman IH. Prophylaxis for opioid-induced constipation in elderly long-term care residents: a cross-sectional study of Medicare beneficiaries. Am J Geriatr Pharmacother 2007;5:129-36. 92. Becker G, Galandi D, Blum HE. Peripherally acting opioid antagonists in the treatment of opiate-related constipation: a systematic review. J Pain Symptom Manage 2007;34:547-65. 93. Lawlor PG, Bruera E. Side-effects of opioids in chronic pain treatment. Curr Opin Anaesthesiol 1998;11:539-45. 94. Gallagher R. Opioid-induced neurotoxicity. Can Fam Physician 2007;53:426-7. 95. Daeninck PJ, Bruera E. Opioid use in cancer pain. Is a more liberal approach enhancing toxicity? Acta Anaesthesiol Scand 1999;43:924-38. 96. Quaglia MG, Farina A, Donati E, Cotechini V, Bossu E. Determination of MPTP, a toxic impurity of pethidine. J Pharm Biomed Anal 2003;33:1-6. 97. McDermot P. Recognizing normeperidine toxicity. Nursing 2003;33:24. 98. Seifert CF, Kennedy S. Meperidine is alive and well in the new millennium: evaluation of meperidine usage patterns and frequency of adverse drug reactions. Pharmacotherapy 2004;24:776-83. 99. Vigano A, Bruera E, Suarez-Almazor ME. Age, pain intensity, and opioid dose in patients with advanced cancer. Cancer 1998;83:1244-50. 100. Ripamonti C, Zecca E, De Conno F. Pharmacological treatment of cancer pain: alternative routes of opioid administration. Tumori 1998;84:289-300. 101. Upton RN. Cerebral uptake of drugs in humans. Clin Exp Pharmacol Physiol 2007;34:695-701. 102. Hassenbusch SJ, Portenoy RK, Cousins M, et al. Polyanalgesic Consensus Conference 2003: an update on the management of pain by intraspinal drug delivery – report of an expert panel. J Pain Symptom Manage 2004;27:540-63. 103. Sloan PA. Neuraxial pain relief for intractable cancer pain. Curr Pain Headache Rep 2007;11:283-9. 104. Burton AW, Rajagopal A, Shah HN, et al. Epidural and intrathecal analgesia is effective in treating refractory cancer pain. Pain Med 2004;5:239-47.
Comprehensive management of cancer pain including surgery
105. Ballantyne JC, Carwood CM. Comparative efficacy of epidural, subarachnoid, and intracerebroventricular opioids in patients with pain due to cancer. Cochrane Database Syst Rev 2005;CD005178. 106. Coffey RJ, Burchiel K. Inflammatory mass lesions associated with intrathecal drug infusion catheters: report and observations on 41 patients. Neurosurgery 2002;50:78-86; discussion 86–7. 107. Hassenbusch SJ. Cost modeling for alternate routes of administration of opioids for cancer pain. Oncology (Williston Park) 1999;13:63-7. 108. Gestin Y, Vainio A, Pegurier AM. Long-term intrathecal infusion of morphine in the home care of patients with advanced cancer. Acta Anaesthesiol Scand 1997;41:12-17. 109. Lobato RD, Madrid JL, Fatela LV, Sarabia R, Rivas JJ, Gozalo A. Intraventricular morphine for intractable cancer pain: rationale, methods, clinical results. Acta Anaesthesiol Scand Suppl 1987;85:68-74. 110. De Cicco M, Matovic M, Bortolussi R, et al. Celiac plexus block: injectate spread and pain relief in patients with regional anatomic distortions. Anesthesiology 2001;94:561-5. 111. Kitoh T, Tanaka S, Ono K, Ohfusa Y, Ina H, Otagiri T. Combined neurolytic block of celiac, inferior mesenteric, and superior hypogastric plexuses for incapacitating abdominal and/or pelvic cancer pain. J Anesth 2005;19:328-32. 112. Nadig M, Ekatodramis G, Borgeat A. Continuous brachial plexus block at the cervical level using a posterior approach in the management of neuropathic cancer pain. Reg Anesth Pain Med 2002;27:446; author reply 446–7. 113. Igarashi H, Sato S, Shiraishi Y. Saddle block using 10–20% tetracaine for patients with perineal pain due to recurrent rectal cancer. Anesthesiology 2003;98:781-3. 114. Caratozzolo M, Lirici MM, Consalvo M, Marzano F, Fumarola E, Angelini L. Ultrasound-guided alcoholization of celiac plexus for pain control in oncology. Surg Endosc 1997;11:239-44. 115. Nerve block can decrease pancreatic cancer pain. Mayo Clin Health Lett 2004;22:4. 116. Ischia S, Polati E, Finco G, Gottin L. Celiac block for the treatment of pancreatic pain. Curr Rev Pain 2000;4:127-33. 117. Suleyman Ozyalcin N, Talu GK, Camlica H, Erdine S. Efficacy of coeliac plexus and splanchnic nerve blockades in body and tail located pancreatic cancer pain. Eur J Pain 2004;8:539-45. 118. Carroll I. Celiac plexus block for visceral pain. Curr Pain Headache Rep 2006;10:20-5. 119. Jain PN, Shrikhande SV, Myatra SN, Sareen R. Neurolytic celiac plexus block: a better alternative to opioid treatment in upper abdominal malignancies: an Indian experience. J Pain Palliat Care Pharmacother 2005;19:15-20. 120. Stefaniak T, Basinski A, Vingerhoets A, et al. A comparison of two invasive techniques in the management of
121.
122.
123.
124.
125.
126.
127.
128.
129.
130.
131.
132.
133.
134.
122
intractable pain due to inoperable pancreatic cancer: neurolytic celiac plexus block and videothoracoscopic splanchnicectomy. Eur J Surg Oncol 2005;31:768-73. Kang CM, Lee HY, Yang HJ, et al. Bilateral thoracoscopic splanchnicectomy with sympathectomy for managing abdominal pain in cancer patients. Am J Surg 2007;194:23-9. Gofeld M, Faclier G. Bilateral pain relief after unilateral thoracic percutaneous sympathectomy. Can J Anaesth 2006;53:258-62. de Leon-Casasola OA. Interventional procedures for cancer pain management: when are they indicated? Cancer Invest 2004;22:630-42. Koizuka S, Saito S, Kubo K, et al. Percutaneous radiofrequency mandibular nerve rhizotomy guided by CT fluoroscopy. AJNR Am J Neuroradiol 2006;27: 1647-8. Di Piero V, Jones AK, Iannotti F, et al. Chronic pain: a PET study of the central effects of percutaneous high cervical cordotomy. Pain 1991;46:9-12. Kanpolat Y, Savas A, Ucar T, Torun F. CT-guided percutaneous selective cordotomy for treatment of intractable pain in patients with malignant pleural mesothelioma. Acta Neurochir (Wien) 2002;144:595-9; discussion 599. Bekar A, Kocaeli H, Abas F, Bozkurt M. Bilateral high-level percutaneous cervical cordotomy in cancer pain due to lung cancer: a case report. Surg Neurol 2007;67: 504-7. Crul BJ, Blok LM, van Egmond J, van Dongen RT. The present role of percutaneous cervical cordotomy for the treatment of cancer pain. J Headache Pain 2005;6:24-9. Jackson MB, Pounder D, Price C, Matthews AW, Neville E. Percutaneous cervical cordotomy for the control of pain in patients with pleural mesothelioma. Thorax 1999;54:238-41. Nagaro T, Adachi N, Tabo E, Kimura S, Arai T, Dote K. New pain following cordotomy: clinical features, mechanisms, and clinical importance. J Neurosurg 2001;95:425-31. Hong D, Andren-Sandberg A. Punctate midline myelotomy: a minimally invasive procedure for the treatment of pain in inextirpable abdominal and pelvic cancer. J Pain Symptom Manage 2007;33:99-109. Nauta HJ, Soukup VM, Fabian RH, . et al. Punctate midline myelotomy for the relief of visceral cancer pain. J Neurosurg 2000;92:125-30. Nauta HJ, Hewitt E, Westlund KN, Willis WD Jr. Surgical interruption of a midline dorsal column visceral pain pathway. Case report and review of the literature. J Neurosurg 1997;86:538-42. Frank F, Fabrizi AP, Gaist G, Weigel K, Mundinger F. Stereotactic mesencephalotomy versus multiple thalamotomies in the treatment of chronic cancer pain syndromes. Appl Neurophysiol 1987;50:314-18.
2079
2080
122
Comprehensive management of cancer pain including surgery
135. Frank F, Fabrizi AP, Gaist G. Stereotactic mesencephalic tractotomy in the treatment of chronic cancer pain. Acta Neurochir (Wien) 1989;99:38-40. 136. Hayashi M, Taira T, Chernov M, et al. Gamma knife surgery for cancer pain-pituitary gland-stalk ablation: a multicenter prospective protocol since 2002. J Neurosurg 2002;97:433-7. 137. Sloan PA, Hodes J, John W. Radiosurgical pituitary ablation for cancer pain. J Palliat Care 1996;12:51-3. 138. Frighetto L, De Salles A, Wallace R, et al. Linear accelerator thalamotomy. Surg Neurol 2004;62:106-13; discussion 113–14. 139. Young RF. Functional neurosurgery with the Leksell Gamma knife. Stereotact Funct Neurosurg 1996;66: 19-23. 140. Whittle IR, Jenkinson JL. CT-guided stereotactic antero-medial pulvinotomy and centromedian-parafascicular thalamotomy for intractable malignant pain. Br J Neurosurg 1995;9:195-200. 141. Robb KA, Newham DJ, Williams JE. Transcutaneous electrical nerve stimulation vs. transcutaneous spinal electroanalgesia for chronic pain associated with breast cancer treatments. J Pain Symptom Manage 2007;33:410-19.
142. Cata JP, Cordella JV, Burton AW, Hassenbusch SJ, Weng HR, Dougherty PM. Spinal cord stimulation relieves chemotherapy-induced pain: a clinical case report. J Pain Symptom Manage 2004;27:72-8. 143. Lee AW, Pilitsis JG. Spinal cord stimulation: indications and outcomes. Neurosurg Focus 2006;21:E3. 144. Cioni B, Meglio M. Motor cortex stimulation for chronic non-malignant pain: current state and future prospects. Acta Neurochir Suppl 2007;97:45-9. 145. Saitoh Y, Yoshimine T. Stimulation of primary motor cortex for intractable deafferentation pain. Acta Neurochir Suppl 2007;97:51-6. 146. Maarrawi J, Peyron R, Mertens P, et al. Motor cortex stimulation for pain control induces changes in the endogenous opioid system. Neurology 2007;69: 827-34. 147. Pereira EA, Green AL, Nandi D, Aziz TZ. Deep brain stimulation: indications and evidence. Expert Rev Med Devices 2007;4:591-603. 148. Owen SL, Green AL, Nandi DD, Bittar RG, Wang S, Aziz TZ. Deep brain stimulation for neuropathic pain. Acta Neurochir Suppl 2007;97:111-16.
127 CT-Guided Percutaneous Cervical Cordotomy for Cancer Pain Y. Kanpolat
Sectioning of the lateral spinothalamic tract in the anterolateral spinal cord is known as cordotomy. The name cordotomy was first used by Schu¨ller [1]. The first cordotomy in man was carried out by Martin at Spiller’s instigation [2]. Cordotomy operations were performed with the open technique, and applied through the upper thoracic region by the posterior approach. High cervical cordotomy was performed by Foerster (1927) and anterior open cordotomy was described by Collis (1963) and Cloward (1964) [3–5]. Due to the high mortality and morbidity risk of open cordotomy the authors developed a less invasive procedure like the percutaneous approach. The first application of percutaneous cordotomy with a radioactive strontium needle was described by Mullan in 1963 [6], who caused electrolytic coagulation of the anterolateral spinal cord with a needle electrode system in 1965 [7,8]. In the same year, Rosomoff described the technique of percutaneous cordotomy with a radiofrequency (RF) needle electrode system [9]. Percutaneous cordotomy at the lower cervical region using the anterior approach was described by Lin and Gildenberg in 1966 and 1968 [10,11]. Percutaneous cordotomy is superior to conventional methods in many ways. First of all, it is less invasive, and also, the impedance measurements with the needle electrode system provide valuable information about the structure of the tissue in which the electrode is located [7–9]. In addition, with the cooperation of the patient in response to neurostimulation the functional #
Springer-Verlag Berlin/Heidelberg 2009
features of the spinal cord region where the needle electrode system is placed is easily evaluated for correct positioning of the lesion [12,13]. The most important disadvantage of conventional percutaneous cordotomy was in the imaging technique used. X-ray imaging was routinely performed during the operation despite the use of positive and negative contrast agents (with X-ray imaging.) It is not possible to demonstrate the real shape and diameters of the spinal cord and the target–electrode relations. In the last 30 years, due to extensive usage of morphine pumps and electrode stimulating systems, percutaneous cordotomy lost its popularity in intractable pain surgery. We have performed percutaneous destructive pain procedures with CT-imaging and a special needle electrode system since 1987 [13–17]. CT-guided percutaneous cordotomy can be safely, effectively, and selectively performed with the help of this special needle electrode kit. Percutaneous cordotomy was used with CT and MRI guidance in neurosurgical practice [18–21]. In this chapter, essentials of the technique of CT-guided percutaneous cordotomy are presented.
Pertinent Anatomy and Rationale The spinal cord receives sensory information from the skin, joints, and muscles of the trunk and limbs and from the internal organs, in turn containing the motor neurons responsible for both voluntary and reflex movements, and has
2150
127
CT-guided percutaneous cervical cordotomy for cancer pain
clusters of neurons that control many visceral functions [22]. The spinal cord is composed of gray matter, which contains the cell bodies, glial cells, and white matter, consisting mainly of axons grouped into tracts, in which there is a supporting framework of neuroglia. The axons of the white matter consist of: (1) axons to the ventral and from the dorsal nerve roots, (2) fibers of the anterior white commissure, (3) intersegmental tracts, and (4) fibers interconnecting the spinal cord and the brain [22]. Somatic sensory signals are conveyed along two major ascending systems in the spinal cord: (1) the anterolateral system, and (2) the dorsal column-medial lemniscal system. The dorsal column-medial lemniscal system mediates tactile sensations and extremity proprioception; the anterolateral system carries information chiefly about pain and temperature, and relays some tactile information. In the dorsal column-medial lemniscal system the axons of second-order neurons cross the midline in the medulla. On the other hand, the anterolateral system decussates in the spinal cord. Whereas both systems transmit sensory information predominantly to the contralateral thalamus and cortex, the anterolateral system also projects ipsilaterally. Since most of the fibers in the anterolateral system decussate over 2–5 segments before entering the anterolateral columns, spinothalamic cordotomy aims to interrupt the spinothalamic tract ascending contralaterally to the painful side. Although the majority of pain-transmitting fibers decussate normally, the decussated axon rate and their position change between individuals they have been reported to be highly variable, thus, a unilateral cordotomy can, on rare occasions, produce ipsilateral analgesia due to nondecussated neurons (encountered) in some individuals [22]. The anterolateral sensory system has a somatotropic relation with fibers from higher levels laminating medially and ventrally. The topographic representation within the pain tracts
usually places the sacral segments most posterolaterally with the segments, the segments lying more ventromedially as one ascends the cord, i.e., cervical segments are situated more medially and anteriorly [22,23]. On the other hand, from the distribution of the pain-conducting fibers within the antero-lateral spinal tract, it seems that the small ventrally located fibers mainly conduct pain sensation, i.e., the organization of fibers from outside inward is: superficial pain, temperature, and deep pain (> Figure 127-1). In 1926, Peet stated that the depth of the cordotomy incision, particularly in the anterior portion of the anterolateral tract, caused analgesia in the upper parts of the body [24]. In 1939, Hyndman and Van Epps suggested the possibility of producing high-level analgesia with the preservation of pain and temperature sensibility in the lower half of the body [25]. Brihaye et al. [26] stated that they have tried to confirm the value of such limited transsection on numerous occasions but were able to obtain analgesia of the upper extremity while sparing the leg only once. Brihaye and Sweet concluded that this type of operation carries a high risk of recurrence of pain because of incomplete destruction of the painconducting pathway [26,27]. In 1965, Mullan indicated the possibility of selective lesions with the percutaneous technique ‘‘As experience has been gained we have achieved greater precision in obtaining selective cordotomies of arm, leg and trunk’’ [8]. Our proposal in managing unilateral local intractable pain is selective destruction of the pain pathway, which innervates the related painful area, instead of causing a hemianalgesia. Between the anterior extent of the pyramidal tracts and the posterior extent of the lateral spinothalamic tracts a narrow ‘‘safety zone’’ is the white matter. On the other hand, there is also much variation in the size and importance of the ventral corticospinal tract, e.g., it may sometimes not decussate at all. Since the motor decussation may extend from the obex to the C1 level, contralateral leg weakness can also occur if the lesion
CT-guided percutaneous cervical cordotomy for cancer pain
127
. Figure 127-1 Cross section of human spinal cord at C2 level
is too high [23]. Due to the ventral spinocerebellar tract overlying the lateral spinothalamic tract, a lesion which eliminates the ventral spinocerebellar tract produces ipsilateral ataxia of the arm. Fibers mediating respiration, as well as the descending autonomic pathways for vasomotor and genitourinary control, which lie adjacent to the anterior horn in close proximity to cervical spinothalamic fibers intermingle with the spinothalamic tract and are likely to be sectioned at the same time as the deeper nociceptive fibers. In previous studies [28], at the level of the first cervical vertebra, respiratory center in man was located in the anterior quadrant of the cord, suggesting that it lies 3 mm from the lateral surface of the anterior quadrant, was 2.5-mm in width, and possibly extends to C7.
Indications Cases with unilateral localized pain due to malignancy are ideal candidates for percutaneous
cordotomy. In such cases, CT-guided percutaneous cordotomy should be the first choice before administration of narcotic drugs, especially if the primary malignant disease is under control. Since these patients can probably return to active life, the use of neurostimulation systems and/or morphine pumps should be avoided as initial therapy [29–33]. We advise bilateral cordotomy only for cases with bilateral abdominal, pelvic, or lower extremity pain, and do not recommend it for upper trunk and extremity pain because of the high risk of respiratory complications. In benign diseases, cordotomy must be considered as the final choice but before the morphine pump [29–31]. Before performing the procedure, identification of the pain type is essential. For example, a case suffering from brachialgia due to Pancoast’s tumor can benefit from cordotomy at the beginning. However, it must be explained to the patient that in future neuropathic pain and functional disorders may occur due to lesioning of the plexus.
2151
2152
127
CT-guided percutaneous cervical cordotomy for cancer pain
Patients with severe pulmonary dysfunction and those who are not able to stay in supine position for 30–40 min are not suitable for the procedure. The other important point is that the cooperation and behavior of both the patient and his or her family should be considered carefully before such therapies are attempted and should be attempted with due care in patients treated for a long time with opiate alkaloids who have possibly developed dependency and in cases of psychopathic family-patient relations [29–33].
Target The target in percutaneous cordotomy is the lateral spinothalamic tract located at the anterolateral part of the spinal cord at the C1–C2 level. At the level of C1–C2, anteroposterior diameter of spinal cord has been reported to be 10 mm and transversally 14 mm [6–9]. White and Sweet also reported the diameter of the spinal cord as 12 1 mm at the equator [27]. The most important advantage of CT-guided cordotomy is the measurement of spinal cord diameters for each patient during the procedure. In our studies of 63 patients, the spinal cord diameters at C1–C2 level measured 7.0–11.4 mm (average: 8.66 mm) anteroposteriorly and 9.0–14 mm (average: 10.96 mm) transversely [34]. After the dural puncture ideal localization of the tip of the cannula is 1 mm anterior to the dentate ligament (equator of the spinal cord) for lumbosacral fibers and 2–3 mm anterior to the dentate ligament for thoracic and cervical fibers [33]. With the help of these diametral measurements, the inserted part of the electrode system can be adjusted. The shape of the spinal cord is also very important for the placement of the electrode into the target. It must be kept in mind that the transsection of the spinal cord can be either ovoid or circular. In the lateral spinothalamic
tract, the pain fibers have such a somatotopic arrangement that those related to the cervical region are located in the anteromedial part of the tract and lumbosacral fibers are located in the posterolateral part. The tip of the electrode must be placed according to this somatotopic organization to denervate the part of the body where the pain is dominant.
Electrode System and Its Calibration In practice, there are different types of needle electrode systems available for percutaneous cordotomy, their own standards, and therefore electrical parameters valid for one type cannot be applied effectively to another. Some authors indicate electrical values such as milliamperes and RF power values for the special electrode systems used by them. Therefore, they must not be accepted without testing each electrode and generator system in egg-white lesioning. For testing impedance measurements and evaluation of CT images, we used a simple artificial spinal cord model set up by using a segment of the spinal cord of a sheep placed in a hollowed orange filled with a saline solution. With this model, it is possible to observe impedance changes while the electrode is in the saline solution, touching the cord and within the cord. We use, the KCTE electrode kit (Cosman Medical1 Inc., Burlington, MA, USA). In this system, 20–22 gauge thin wall needles with plastic hubs were designed to avoid imaging artifact problems. There are 2 mm open tip termocouple electrode in the kit. One of them is 0.25 mm in diameter with a straight tip, and the other one is 0.25 mm in diameter with a curved tip. The depth of the inserted part of both the straight and curved active electrodes should also be adjusted according to the measured diameters of the spinal cord before application. The crew
CT-guided percutaneous cervical cordotomy for cancer pain
of the sizing clamps is tightened in the same direction as the tip of the curved electrode to facilitate sensing of the direction.
Technique Preoperative Preparation Patients should have been fasting for 5 h before the operation. Before the procedure, the required dose of analgesics should be given parenterally. At this stage, patients should be re-informed about the details of the procedure. If required, neuroleptic anesthesia should be given in which will a dose, not affect cooperation of the patient during the procedures.
Positioning Since CT-guided percutaneous cordotomy is performed in the CT unit, the patient is placed on the CT-table in the supine position. With the help of the head-support, the head of the patient is kept in slight flexion, and the longitudinal axis of the cervical spine must be kept strictly straight. If the primary disease does not allow the patient to lie in supine position, it is possible to carry out the procedure in the lateral decubitis position. The head is fixed with a fixation band.
in the Trandelenburg position before admission to the CT unit. If the general condition of the patient does not permit lumbar puncture, 5 mL of contrast material is injected during the procedure at the C1–C2 level. Diametral measurements of the spinal cord are taken, and the inserted part of the active electrode is adjusted accordingly. CT images from a 1,200 SX device, and an IQ Premier 512*512 matrix, with 2 mm slice thickness are used and the quality of the image is enhanced by diminishing the diameter of the image formation area.
Insertion of the Needle Electrode System Twenty gauge plastic hub needles specially designed for CT-guided procedures are used. Following the injection of the local anesthetic agent, the cordotomy needle is inserted inferior to the tip of the mastoid process in a vertical plane perpendicular to the axis of the spinal cord. The placement of the needle at C1–C2 level can be seen in the lateral scanogram, and the direction of the needle can be manipulated toward the anterior part of the spinal cord with the help of axial CT sections (> Figures 127-2– > 127-3). Ideal placement is 1 mm anterior to the dentate ligament for lumbosacral fibers (> Figure 127-4), . Figure 127-2 Correct position of the needle
Injection of the Contrast Material In CT-guided percutaneous cordotomy, the patient should fast for 5 h before the procedure. The patient is informed before the procedure by the surgeon. The required dose of analgesics is given parenterally. A contrast material should be administered into the subarachnoidal space of the spinal cord. Iohexol (7–8 mL of 240 mg/mL) is given 20 min before the operation by lumbar puncture. In this case, the patient should be kept
127
2153
2154
127
CT-guided percutaneous cervical cordotomy for cancer pain
. Figure 127-3 Classical Rosomoff cordotomy electrodes placed in the anterolateral spinal cord
. Figure 127-4 Final position of the termocoupled electrode in the posterolateral part of the lateral spinothalamic tract
2–3 mm anterior for thoracic and cervical fibers (> Figure 127-5). The needle is in ideal position if it is perpendicular to the spinal cord. After achieving the ideal position of the needle tip, the straight or curved electrode is inserted. Impedance measurement is very important to identify the position of the active electrode tip in the CSF, in contact with the spinal cord or in the spinal cord. Target–electrode relationships can be easily detected by direct visualization of the needle electrode system by CT-guidance. Due to this direct visualization possibility, impedance measurements are no longer vital for the
. Figure 127-5 Final position of the termocoupled electrode in the anterior part of the lateral spinothalamic electrode
. Figure 127-6 Improperly placed electrode in front of the anterior region of the spinal cord
control of the penetration of the active electrode into the spinal cord. Impedance measurements supply information about the penetration of the active electrode into the spinal cord but not about the depth of penetration. However, it is still an important indication of passage into a new medium along the path of the needle electrode. In some cases, pial punctures cannot be achieved (> Figure 127-6), instead the spinal cord is displaced. This situation can easily be visualized on CT, and therefore proper measures can be taken, and placement of the electrode can
CT-guided percutaneous cervical cordotomy for cancer pain
be repeated to achieve puncturing. Such rare distortions of the spinal cord in contact with the needle can be detected on the CT, therefore misinterpretation is avoided. Finally, with the help of CT guidance, the needle electrode can be directed into the required area within the lateral spinothalamic tract. For arm and chest pain anteromedial and for leg pain posterolateral placement of the active electrode system are chosen in the lateral spinothalamic tract [16,17].
Physiologic Localization With the help of neurophysiological confirmation via impedance measurement, it identifies whether the active electrode tip is in the cerebrospinal fluid (CSF) (100–200 O) in contact with the spinal cord (300–400 O) or inside the spinal cord (more than 700 O). The target–electrode relationships are detected by CT guidance. Another important step is functional confirmation of the electrode obtained by stimulation. An indifferent electrode is routinely placed in the contralateral side. Stimulation with low frequencies (2–6 Hz, 0.4–1.5 V) cause ipsilateral trapezius muscle contractions indicating that the electrodes are within or near the anterior gray matter. Stimulation of 100 Hz with 0.2–1.5 V causes pain, paresthesia, or warmth in the spinothalamic tract, and that is the target [29–32]. In our practice, there is no possibility of mislocalization of the active electrode system with the help of CT-imaging but the presence of some variations of the tracts, especially of the corticospinal tract, ipsilateral motor response can be observed. This is the most important contribution of the stimulation during this procedure. In more than 350 CTguided procedures, we did not observe any anatomical variations in practice.
Lesion With our special needle electrode system, persistent lesions can usually be achieved at
127
a tip temperature of over 60 C within 20 s. During the lesioning, the energy and tip temperature are increased gradually and ipsilateral motor functions must be continuously tested. The final lesion is made at 70–80 C and for 60 s. Level of the heat is gradually increased during the lesioning in 60 s. Although the target is confirmed morphologically by CT and neurophysiological by impedance measurements and stimulation, if the required level of analgesia is not obtained, the lesion is repeated with the same parameters in 60 s. Number of the lesions is not more than three in 70–80 C; in bilateral cordotomy we prefer to minimize the number of lesions but sometimes use two or three lesions for dominant pain side. After the procedure, the patient should be transported on a stretcher because there may be a temporary walking difficulty due to ataxia. The patient should have been previously informed about this, and be hospitalized for one night after the operation especially just after the bilateral cordotomy in intensive care unit. Neurological examination is repeated postoperatively. The patient must also be warned about headaches due to the contrast matter and CSF loss. We usually inform the patient regarding his or her post-cordotomy life [33].
Results Between 1987 and 2008, CT-guided percutaneous cordotomy was applied to 221 cases 248 times. The procedure was applied bilaterally in 12 cases. Most (107 cases) were suffering from intractable unilateral pain due to malignancy. Pulmonary malignancies (60 cases), mesothelioma (26 cases), pancoast (15 cases), and breast carcinoma (6 cases) had the highest rate (52%). Among the others, there were gastrointestinal carcinomas, osteogenic malignancies, metastatic carcinomas, genitourinary malignancies, and patients with several malignancies.
2155
2156
127
CT-guided percutaneous cervical cordotomy for cancer pain
The procedure was applied to 16 cases of benign pain like failed-back syndrome, post-thorocotomy syndrome, electrical burn, perineural cyst, etc. The pain measurement scale was determined as follows: I – no pain, II – partial satisfactory pain relief, III – partial nonsatisfactory pain relief, and IV – no change in pain. In this grading system, we considered grade I and II patients as having pain relief after surgical intervention. The initial success rate of CT-guided percutaneous cordotomy was 94.6%. In cases with malignancies, this rate was 96.9%. In 182 cases (88.78%), only the painful region of the body was relieved from pain thus achieving selective cordotomy. In four cases (three Pancoast’s tumors), pain due to malignancy was relieved but neuropathic pain related to the deafferentation occurred afterwards. In 13 cases, the procedure was applied twice because of failure of the treatment, in 1 case the procedure was applied three times. Pulse RF procedure was applied to the patient with electrical burns, and the pain resolved thereafter. However, 15 days later the pain recurred and a second cordotomy was applied with CTguided percutaneous RF cordotomy. Two years later the patient returned to our department and insisted on a third cordotomy, following which successful pain relief was obtained. In 12 cases the procedure was applied bilaterally [17]. The initial success rate of 3,742 percutaneous cordotomy cases collected by Lorenz was 75–96% [35]. In another study reported by Sindou [36], results taken from an analysis of 171 personal cases and a review of 37 series from the literature, totaling 5,770 cases pain relief was evaluated independently in cancerous and noncancerous patients. In the cancer group (2,022 cases), anterolateral cordotomy was effective in providing satisfactory pain relief during the survival period in a significant number of patients, i.e., in 75% at 6 months and 40% after 1 year.
Complications In percutaneous cordotomy, there are two reasons for complications. First, the needle electrode may be mislocalized therefore may not carry out its function on the required tract but affect other tracts. In CT-guidance, this possibility is almost totally overruled due to direct imaging. The second important reason for complications is involuntary enlargement of the lesioned area causing interruption of unintended functions. This possibility is also minimal due to 2 mm diameter termic lesions with the special electrode system. As a result of these advantages, no mortality or persistent complications were encountered in our series. Only three temporary ataxia cases and two temporary hemiparesis were observed. In cervical cordotomies, it has already been mentioned that respiratory dysfunction is the principal complication and can be fatal [28,37–39]. The risk is higher in patients with preexisting functional respiratory disorders. When a bilateral cordotomy is carried out, respiratory functions must be carefully observed postoperatively. Cardiovascular disturbances may also be expected after a cordotomy. Sleep-induced apnoea has been reported as a serious and fatal complication [37–39]. Ipsilateral hemiparesis may occur, being mostly of transient nature but in other series irreversible hemiparesis was reported in up to 8% of cases. Such complications may vary with the condition of the patient and the level of cordotomy. Walking difficulty due to ataxia was noticed as the most common complication, which tends to disappear in most cases. Some authors indicated that if difficulty with urination is present, it is also usually temporary. As reported previously, importance especially after bilateral cordotomy must be expected in most cases; however, explanation of the mechanism is speculative. Dysesthetic syndromes may occasionally occur but their long-term persistence is not common.
CT-guided percutaneous cervical cordotomy for cancer pain
Conclusion In the last 30 years, destructive pain procedures have lost their popularity in the context of surgical treatment of pain due to malignancy. The reason is not only the high efficiency of new surgical methods like morphine pumps or neurostimulation but also neglect in the field of stereotactic pain procedures and electrode technology. CT-guided pain procedures can be used easily, effectively, and selectively in the treatment of unilateral intractable pain. With the help of CT imaging and new electrode technology, this procedure can be used efficiently by neurosurgeons in the treatment of intractable pain, a practice they seem to have lost to other disciplines during the last 30 years.
Acknowledgments I would like to express our deepest appreciation to Mr. Ender Arkun for editing this article, and to Associate Professor Dr. Toygun Orbay for his creative drawings.
References ¨ ber operative Durchtrennung der Ru¨cken1. Schu¨ller A. U marksstra¨nge (Chordotomie). Wien Med Wschr 1910;60:2291-6. 2. Spiller WG, Martin E. The treatment of persistent pain of organic origin in the lower part of the body by division of the antero-lateral column of the spinal cord. J Am Med Assoc 1912;58:1489-90. 3. Foester O. Die Leitungsbahnen des Schmerzgefu¨hls und die Chirurgische Behandlung des Schmerzzusta¨nde. Berlin: Urban und Schwarzenberg; 1927. p. 36. 4. Collis JS Jr. Anterolateral cordotomy by an anterior approach; report of a case. J Neurosurg 1963;20:445-6. 5. Cloward RB. Cervical cordotomy by an anterior approach; report of a case. J Neurosurg 1964;20:445-6. 6. Mullan S, Harper PV, Hekmatpanach J, Torres H, Dobbin G. Percutaneous interruption of spinal pain tracts by means of a strontium 90 needle. J Neurosurg 1963;20:931-9.
127
7. Mullan S, Hekmatpanach J, Dobbin G, Beckman F. Percutaneous intramedullary cordotomy utilizing the unipolar anodal electrolytic lesion. J Neurosurg 1965;22:548-53. 8. Mullan S. Percutaneous cordotomy. J Neurosurg 1971;35:360-6 9. Rosomoff HL, Carroll F, Brown J, and Sheptak P. Percutaneous radiofrequency cervical cordotomy: technique. J Neurosurg 1965;23:639-44. 10. Lin PM, Gildenberg PL, Polakoff PO. An anterior approach to percutaneous lower cervical cordotomy. J Neurosurg 1966;25:553-60. 11. Gildenberg PL, Lin PM, Polakoff PP, II, Flitter MA. Anterior percutaneous cervical cordotomy. Determination of target point and calculation of angle of insertion. Technical note. J Neurosurg 1968;28:173-7. 12. Gildenberg PL, Zanes C, Flitter MA, Lin PM, Lautsch EV. Impedance measuring device for detection of penetration of the spinal cord in anterior percutaneous cervical cordotomy. Technical note. J Neurosurg 1969;30:87-92. 13. Kanpolat Y, Atalag M, Deda H, Siva A. CT-Guided extralemniscal myelotomy. Acta Neurochir 1988;91:151-2. 14. Kanpolat Y, Deda H, Akyar S, Bilgic S. CT-Guided percutaneous cordotomy. Acta Neurochir 1989;46:67-8. 15. Kanpolat Y, Deda H, Akyar S, Caglar S, Bilgic S. CT-Guided trigeminal tractotomy. Acta Neurochir 1989;100:112-4. ¨ nlu¨ A, Bilgic S. CT16. Kanpolat Y, Akyar S, Caglar S, U Guided percutaneous selective cordotomy. Acta Neurochirur 1993;123:92-7. 17. Kanpolat Y, Savas A, Caglar S, Temiz C, Akyar S. Computerized tomography-guided percutaneous bilateral selective cordotomy. Neurosurg Focus 1997;2:e4. 18. Fenstermaker RA, Sternau LL, Takaoka Y. CT-assisted percutaneous anterior cordotomy: technical note. Surg Neurol 1995;43:147-50. 19. Bekar A, Kocaeli H, Abas¸ F, Bozkurt M. Bilateral highlevel percutaneous cervical cordotomy in cancer pain due to lung cancer: a case report. Surg Neurol 2007;67:504-7. 20. Raslan A. Percutaneous computed tomography-guided transdiscal low cervical cordotomy for cancer pain as a method to avoid sleep apnea. Stereotact Funct Neurosurg 2005;83:159-64. 21. McGirt MJ, Villavicencio AT, Bulsara KR, Gorecki J. MRI-guided frameless stereotactic percutaneous cordotomy. Sterotact Funct Neurosurg 2002;78:53-63. 22. Walker EA. The spinothalamic tract in man. Arch Neurol Psychiatry 1940;43:284-98. 23. Taren JA, Davis R, Crosby EC. Target physiologic corroboration in stereotactic cervical cordotomy. J Neurosurg 1969;30:569-619. 24. Peet MM. The control of intractable pain in lumbar region, pelvis and lower extremities. Arch Surg 1926;13:153-204.
2157
2158
127
CT-guided percutaneous cervical cordotomy for cancer pain
25. Hyndman OR, Van Epps C. Possibility of differential section of the spinothalamic tract. A clinical and histologic study. Arch Surg 38:1036-53. 26. Brihaye J, Thiry S, Le Clerco R, et al. Le traitement chirurgical de la douleur. Acta Chir Belg (Suppl) 1962;2:255-475. 27. White JC, Sweet WH. Spinothalamic tractotomy. Pain and neurosurgeon. Springfield, IL: Charles C. Thomas; 1969. p. 678-772. 28. Belmusto L, Brown E, and Owens G. Clinical observations on respiratory and vasomotor disturbance as related to cervical cordotomies. J Neurosurg 1963;20:225-32. 29. Kanpolat Y. Percutaneous stereotactic pain procedures: percutaneous cordotomy, extralemniscal myelotomy, trigeminal tractotomy-nucleotomy. In: Burchiel K, editor. Surgical management of pain. Stuttgart/New York: Thieme; 2002. p. 745-62. 30. Kanpolat Y. Cordotomy for pain. In: Schulder M, editor. Handbook of stereotactic and functional neurosurgery. New York: Marcel & Dekker; 2003. p. 459-72. 31. Kanpolat Y. The surgical treatment of chronic pain: destructive therapies in the spinal cord. Neurosurg Clin N Am 2004;15:307-17. 32. Kanpolat Y. Percutaneous cordotomy, tractotomy, and midline myelotomy, minimally invasive stereotactic pain procedures. Semin Neurosurg 2004;15:203-20.
33. Kanpolat Y. Percutaneous destructive pain procedures on the upper spinal cord and brain stem in cancer pain: CT-guided techniques, indication and results. In: Pickard JD, editor. Advances and technical standards in neurosurgery. Wien: Springer; 2007;147-173. 34. Kanpolat Y, Akyar S, Caglar S. Diametral measurements of the upper spinal cord for stereotactic pain procedures: experimental and clinical study. Surg Neurol 1995;43:478-83. 35. Lorenz R. Methods of percutaneous spinothalamic tract section. In: Krayenbu¨hl H, Brihaye J, editors. et al. Advances and technical standards in neurosurgery, vol. 3, Wien: Springer; 1976. p.123-54. 36. Sindou M, Jeanmonod D, Mertens P. Ablative neurosurgical procedures for the treatment of chronic pain. Neurophysiol Clin 1990;20:399-423. 37. Onofrio BM. Cervical spinal cord and dentate delineation in percutaneous radiofrequency cordotomy at the level of the first to second cervical vertebrae. Surg Gynecol Obstet 1971;133:30-4. 38. Rosomoff HL. Bilateral percutaneous cervical radiofrequency cordotomy. J Neurosurg 1969;31:41-6. 39. Rosomoff HL, Krieger AJ, Kuperman AS. Effects of percutaneous cervical cordotomy on pulmonary function. J Neurosurg 1969;31:620-7.
131 DBS for Persistent Non-Cancer Pain C. Hamani . D. Fontaine . A. Lozano
Introduction
Rationale/mechanisms of Action
The use of electrical stimulation to treat pain dates back from the 1950s, when Heath [1,2] and Pool [3] reported analgesic effects in patients receiving stimulation in the septal region. Despite of these earlier trials, most of the pioneering studies that led to the development of deep brain stimulation (DBS) as currently employed in the clinical practice took place in the early 1970s. Electrical stimulation of the thalamus and internal capsule was initially developed by Mazars et al. [4–7], White and Sweet [8], Hosobuchi et al. [9,10], Adams et al. [11], and Fields et al. [12]. A few years later, Richardson and Akil [13–15] and Hosobuchi et al. [16] described the first clinical results with periaqueductal (PAG)/periventricular gray matter (PVG) stimulation. Inthe1970sand1980s,thalamicandPAG/PVG stimulation were commonly used for the treatment of chronic refractory pain [7,10,11,15,17–25]. However, within the last decades there has been a progressive declinein boththenumberofpublished studies and the number of chronic pain patients treated with DBS. This has been partially attributed tothedevelopmentandemploymentoflessinvasive alternatives for the management of nociceptive pain, including catheters and pumps for opioid administration, new pharmacological agents, and spinal cord stimulation. Despite of these facts, DBS continues to be routinely offered to patients with chronic refractory neuropathic pain. We will review the rationale and mechanisms of action, selection criteria, technical aspects of the procedure, surgical targets, outcome, and complications of DBS for the treatment of pain.
PAG/PVG
#
Springer-Verlag Berlin/Heidelberg 2009
The first to report the antinociceptive role of PAG stimulation was Reynolds in 1969 [26]. He initially described a series of rodents that underwent laparotomies only under PAG stimulationinduced analgesia [26]. Since then various studies have been conducted in rats, cats and non-human primates, and stimulation-induced analgesia has been reported in several brainstem structures, particularly the PAG, raphe, and noradrenergic nuclei (for a review see [27–29]). Within the PAG, antinociceptive responses seem to be more prominent when stimulation is applied to the ventral portions of the nucleus, in close relationship to the dorsal raphe nucleus [30–33]. Stimulation of this region may also lead to stereotyped behavioral responses, including gnawing, rotation and tremor [28]. When the dorsal or dorsolateral PAG is stimulated, analgesia is often less pronounced and accompanied by aversive responses [31,34,35]. In rodents, these include neurovegetative reactions, avoidance and flight responses, vocalizations, rage, among others [36,37]. These events seem to correlates with the fear/anxiety reported by humans undergoing DBS [38,39]. One of the suggested mechanisms for the antinociceptive effects of PAG/PVG stimulation is the activation of the endogenous opioid system [16,40]. In rodents, microinjections of morphine into the PAG, particularly the ventral portion of the nucleus, induce analgesic effects similar to those described with stimulation [40–43]. Moreover, the analgesic effects of both PAG
2228
131
Dbs for persistent non-cancer pain
microinjections of morphine or electrical stimulation are reversed by naloxone [28,44]. Interestingly, these findings were in line with preliminary human studies showing (1) that patients treated with PAG/PVG DBS had an increased level of endogenous opioids in the cerebrospinal fluid [45,46] and (2) that stimulation-induced analgesia was reversed by naloxone [16,47]. Further studies conducted on a double-blinded fashion however, did not corroborate these findings as the analgesic effects of PAG/PVG stimulation were equally reversed by either naloxone or placebo [48]. It has then been suggested that additional mechanisms independent on the endogenous opioid system could be involved in the analgesia induced by PAG/PVG stimulation. These non-opioid effects could be potentially mediated by the antidromic and/or orthodromic activation of the raphe, adjacent reticular formation, parabrachial nucleus, locus coeruleus, the A5 cell group, and spinal cord [49,50]. In anesthetized non-human primates, PAG stimulation inhibited the activity of spinal cord cells as well as their responses to innocuous and noxious cutaneous stimuli [51]. It has been postulated that the activation of antinociceptive serotoninergic and noradrenergic systems following PAG stimulation might have contributed to these effects. Ascending projections from the PAG/PVG region also innervate cerebral structures involved in the mechanisms of pain and antinociception, including the intralaminar nuclei of the thalamus and the medial hypothalamus [29]. Functional imaging studies in patients undergoing DBS have shown that PAG/PVG stimulation activates not only the medial thalamus but also the anterior cingulate cortex [52,53]. This later structure is involved in the modulation of emotional response to pain and might comprise another mechanism through which PAG/PG stimulation could be exerting its effects.
Sensory Thalamus The first author to employ thalamic stimulation for the treatment of pain was Mazars in 1960 [6]. The rational for choosing the target was based on a theory by Head and Holmes [54] that pain could be caused by two fundamental mechanisms: Either an excess of ‘‘protopathic/nociceptive’’ stimuli, or an insufficient control of these two by ‘‘epicritic/ proprioceptive’’ mechanisms. The idea was to deliver supplementary stimulation to the thalamus of pain patients whenever ‘‘proprioceptive’’ mechanisms were felt to be insufficient to control ‘‘nociceptive’’ stimuli [4–7]. Although this theory is now outdated, almost 50 years after the first use of the technique the mechanisms through which thalamic stimulation induces analgesia are still unknown. There is little doubt that the thalamus is involved in pain signaling pathways. During surgical procedures in patients with chronic pain, microelectrode recording and stimulation techniques have shown that (1) thalamic neurons respond to noxious stimuli [55,56] and (2) stimulation of the thalamus can evoke pain [57,58]. In addition, it has been clearly established that patients with chronic pain have altered somatotopic distributions of receptive fields (RF) and projected fields (PF) as well as abnormal neuronal firing patterns in the thalamus (particularly an increased bursting activity) [20,55,59–61]. Whether these events occur as a result of chronic pain or contribute to the development of pain-related symptoms is still unclear. Bearing in mind the changes described above, one hypothesis would be that DBS could exert part of its effects by modulating the altered firing patterns observed in the thalamus of chronic pain patients [62]. In addition, thalamic stimulation could also influence the activity of other areas of the brain involved in antinociception. Corroborating this later assertion, stimulation
Dbs for persistent non-cancer pain
of the somatosensory thalamus inhibits spontaneous activity and pain-evoked responses of parafascicularis nucleus cells in rodents [63]. In addition, stimulation of this same region in nonhuman primates decreased behavioral responses induced by noxious stimuli and inhibited activity of lamina I spinal cord neurons [51]. Initially, it has been suggested that the influence of thalamic stimulation on spinal cord neurons could be related to the antidromic activation of spinothalamic pathways as well as the modulation of serotoninergic and noradrenergic brainstem nuclei [64,65]. However, the fact that patients with spinal cord lesions benefit from thalamic DBS opposes this hypothesis and urges the need for additional investigation in the field [66]. Imaging studies have also been conducted in patients with chronic pain treated with thalamic stimulation [53,67–69]. Positron emission tomography has shown a significant activation of the amygdala, insula, and anterior cingulate cortex during stimulation [67–69]. In functional MRI studies, Vc stimulation at intensities that were able to induce paresthesias activated the primary and secondary somatosensory cortices, thalamus and insula [53]. These findings were somewhat similar to those reported in rodents receiving 2-deoxyglucose during thalamic stimulation [70].
Selection Criteria/surgical Targets Recognizing the etiology of the pain syndrome for each patient and distinguishing whether there is a predominance of nociceptive or neuropathic components are crucial initial steps for an adequate therapeutic management. Though no proper studies comparing the outcome of DBS in different targets have been conducted, it is a general consensus that neuropathic pain is more likely to respond to stimulation of the sensory thalamus, whereas nociceptive pain responds
131
better to PAG/PVG stimulation [71–75]. Patients with mixed pain may be implanted with electrodes in both structures. In addition to these more commonly used targets, DBS in the internal capsule has routinely been offered to post-stroke patients with significant thalamic atrophy [11,39]. Other targets leading to a good outcome in small clinical series or case reports are the septal region [76,77], medial thalamus (including the centromedianparafasciculaar nuclei) [78–81], and the cingulate gyrus [82]. In addition to the clinical characteristics of pain, some authors also advocate the use of the so-called morphine-naloxone test [16,19,39,83]. During this test, patients are initially given morphine (usually i.v. to a total of 25 mg over 45 min) and asked to rate their pain. Once the response is established, patients receive naloxone to assess whether pain recurs. Patients who respond to the test are considered to have a more prominent nociceptive component and are offered DBS in the PAG/PVG. Despite of the rationale for the morphine-naloxone test, its validity in predicting a good outcome to DBS has been disputed [48]. As a result, most centers do not offer the test at present. Common etiological diagnoses in patients with neuropathic pain undergoing thalamic DBS are post-stroke pain, atypical facial pain, spinal cord injury, multiple sclerosis and phantom limb pain. Patients with nociceptive pain treated with PAG/PVG stimulation most commonly have failed back syndrome (FBS). In the past PAG/PVG DBS was also commonly used for the treatment of cancer pain. Independent of the target or condition to be treated, all patients referred for DBS have to have severe (often indicated by a minimal visual analog scale score (VAS) of 6/10 or greater) and refractory pain. Patients must have tried and failed all reasonable medications and behavioral treatments. Another important aspect is the timeframe
2229
2230
131
Dbs for persistent non-cancer pain
between the onset of pain and surgery. Though six months is often stated as a reasonable interval, most patients are only deemed surgical candidates years after the development of pain. As a final remark, surgery is often declined to patients with significant psychological or psychosocial overlay, secondary gain, and to those with clinical and neurological conditions that may significantly increase the risk of the procedure (e.g., coagulopathies) (> Table 131-1).
Surgical Technique Anatomical Target Localization and Surgical Planning For anatomical targeting, most centers to date use stereotactic magnetic resonance imaging (MRI) or computed tomography (CT) (this later to be merged with MRI images acquired prior to surgery). The use of neuronavigation is not crucial if stereotactic MRI alone is used. Yet, it certainly
facilitates the procedure by (1) decreasing the time spent for calculating the surgical coordinates, (2) allowing the target to be determined based on triplanar images, and (3) reducing MRI related distortions [87,88]. If one chooses to merge a stereotactic CT to a previously acquired MRI, either specific software or neuronavigation systems are required. Though surgery has been previously done with ventriculography and CT alone, we believe that MRI is essential for targeting due to its higher resolution for visualizing cerebral structures. Thalamic electrodes are implanted in somatosensory region of the thalamus (ventralis caudalis nucleus; Vc) contralateral to the side of the worse pain. Patients with severe bilateral pain are often treated with bilateral electrodes. As the internal thalamic anatomy cannot be visualized on MRI, we often rely on coordinates relative to the anterior (AC) and posterior commissures (PC). The coordinates for Vc are 2–3 mm anterior to PC (y coordinate), with the z coordinate (dorsalventral) set at the level of the AC-PC plane. Due
. Table 131-1 Outcome of DBS surgery for the treatment of pain according to etiology
Etiology Post-stroke pain Phantom limb Anesthesia dolorosa Spinal cord injury Causalgia/ CRPS FBSS
Number of patients treated
Percentage of patients who had a successful stimulation trial
Percentage of patients who underwent surgery and were using DBS at long-term
Percentage of patients implanted with an IPG that were using DBS at long-term
79
53%
24%
45%
16
81%
50%
62%
38
68%
32%
46%
45
47%
22%
48%
11
81%
73%
89%
72
90%
76%
85%
Note: Data was acquired from studies that provided a quantitative follow-up and included patients with chronic non-cancer pain of known organic origin, who failed all conventional treatments, and had no psychiatric or psychosocial overlay [13,19,24,25,38,73,83–86]. In some of these studies, it was not possible to clearly discern the surgical target for the treatment of each condition. Yet, patients with failed back syndrome were most commonly treated with PAG/PVG DBS, whereas those with predominant neuropathic pain had electrodes implanted in either Vc or Vc and PAG/PVG. CRPS, complex regional pain syndrome; FBSS, failed back surgery syndrome; IPG, internal pulse generator; PAG/PVG, periaqueductal/periventricular gray matter Vc, ventralis caudalis nucleus of the thalamus
Dbs for persistent non-cancer pain
to the somatotopical representation of the body in Vc, an x (medial-lateral) coordinate of 12–13 mm from the midline is used for facial pain, 14–15 mm for upper extremity pain, and 16–17 mm for lower extremity pain (> Figure 131-1). As previously stated, in patients with thalamic infarcts DBS in the posterior limb of the internal capsule may be an alternative approach. This structure can be targeted through direct visualization or trough coordinates relative to the commissures (25 mm lateral to the midline, 12–14 mm posterior to the midcommissural point with a z coordinate set at the level AC-PC plane) [13,38,39,73,86]. The direct visualization of the boundaries of the III ventricle and cerebral aqueduct on MRI scans can certainly help during the targeting of the PAG/PVG. In earlier trials, PAG electrodes used to be implanted in close relationship to the aqueduct [14]. As stimulation at this more ventral location often led to side effects, more
131
recent series have favored the use of the PVG as a target [25,38,39,85,86]. Standard coordinates for placement of the electrodes in this region are 2–5 mm anterior to PC, 2 mm lateral to the medial wall of the third ventricle, at the level of the AC-PC plane [38,39,85,86]. In the PAG/ PVG, unilateral stimulation often leads to bilateral analgesic effects [13,39].
Physiological Target Localization In the operating room, various centers use microelectrode recordings (MER) and stimulation to physiologically map the targets. Thalamic extracellular recordings are used to assess the activity of individual neurons and their receptive field, defined as the body site(s) whereby tactile stimulation or movement evoked changes in electrical firing patterns occur. Microstimulation is used to define projected fields,
. Figure 131-1 Location of the electrode contacts in a patient with chronic pain. (a) Post-operative MRI showing electrode contact artifacts in the nucleus ventralis caudalis (arrow) and the periventricular region (arrowheads). (b) Schematic representation of the same contacts (black dots) over a Schaltenbrand and Warren atlas section (from Schaltenbrand and Wahren. Atlas of stereotaxy of the human brain. New York: Thieme; 1977; modified and reprinted with permission)
2231
2232
131
Dbs for persistent non-cancer pain
characterized as the body location where the patients perceive an electrical stimulus-induced sensation [59,89]. In most patients, MER and microstimulation in Vc reveal a somatotopic representation of body parts, including tactile RFs and microstimulation induced paresthesias into corresponding PFs [20,59,61,89,90]. The MER characteristics of the PAG/PVG region have not been well established. Once the surgical site is mapped, a DBS electrode is implanted into the target. Stimulation through each of the contacts of the DBS leads is then conducted to ensure that no adverse effects are present. In the case of Vc, stimulation elicits sensory changes (most often paresthesias) in the regions of the body correspondent to the patients’ pain. If paresthesias are not covering the areas of pain (i.e., they are felt in the leg when pain is in the arm), the electrode has to be moved. Patients with PAG/PVG electrodes sometimes report a warmth sensation during stimulation. This feeling might be described as relaxing and pleasurable [38,39,83]. Oscillopsia and eventually loss of upward gaze have also been reported (some authors advocate that the presence of these effects indicate a good placement of the electrodes) [73,85]. When high frequency stimulation is used or when the electrodes are anterior to target, sympathetic responses and sensations of anxiety and fear may also be observed [38,39,83]. Posterior and deeper placement of the electrodes may lead to smothering, vertigo or nausea [83]. Once the electrodes are implanted, they are fixed to the skull and connected to percutaneos extensions. These are externalized through the skin and used to test the patients in the postoperative period.
Test Stimulation Trial Patients with chronic pain treated with DBS can often tell in the postoperative period when there
is a lack of response to the procedure. As a result, various centers conduct the so-called ‘‘test stimulation trial’’ when patients are still in the hospital (before implanting the internal pulse generator; IPG). During these tests, several stimulation settings are explored in each of the electrode contacts. A trial is considered to be successful if a reduction of more than 50% in VAS scores is achieved with stimulation. Under these circumstances, the electrodes are connected to an IPG, implanted under general anesthesia in the infraclavicular chest wall (usually above the pectoralis fascia). When two electrodes are internalized, the patients have either two IPGs or a dual channel generator implanted. Patients who do not have a successful stimulation trial have the DBS electrode(s) and the percutaneous extension(s) removed. In our experience, some patients have a substantial reduction (60–100%) in pain scores immediately after the insertion of the electrodes in the absence of stimulation [38]. In these patients, we often cannot determine whether stimulation is effective right after the procedures. As a result, patients that develop this insertional effect have their external cables cut, the DBS electrodes buried underneath the galea, and are subsequently discharged from the hospital. Upon recurrence of the pain, these patients are re-hospitalized and new percutaneous extensions are connected to the DBS electrodes under local anesthesia. A stimulation trial is then conducted and, whenever stimulation is deemed to be effective, the patients’ leads are connected to an IPG [38]. Overall, approximately 60% of the patients initially implanted with DBS electrodes have a successful trial and subsequently undergo the implantation of an IPG. At long-term, thalamic stimulation is often delivered at parameters that induce pleasant paresthesias in the regions of pain [19,38,39,83]. This is often achieved at higher frequencies (around 100 Hz), with 60–210 ms of pulse width and 2–5 V [19,24,38,39,83,86]. It is worth mentioning that, whenever the intensity of
Dbs for persistent non-cancer pain
stimulation gets too strong, the sensation induced often becomes unpleasant and most patients are not able to tolerate it. The most common settings for PAG/PVG stimulation are 1–5 V, with a pulse width of 60–210 ms at 10–25 Hz [13,48,83,85,86]. Patients that have simulation-induced warmth sensation often ask for the stimulation amplitude to set at that level. Though most authors advocate the use of continuous stimulation, cyclic stimulation with the electrodes turned ‘‘on’’ for 1 min and ‘‘off ’’ for 10 min has also been proposed [83].
Results The long-term outcome of DBS for the treatment of chronic neuropathic pain is quite variable in the clinical literature, with most studies showing a response in 20–70% of the patients treated [7,10,15,17–19,21,23,24,38,39,71–75,83–86,91–94]. The response rate in patients with nociceptive pain is slightly higher, in the order of 40–80% [13,19,48,71–75,83,86,91]. Part of the variability in the literature may be attributed to the different conditions treated, the lack of consistency in the methods used to evaluate a clinical response, length of follow up, the use of analgesic drugs or a combination of the above mentioned factors [95]. Invariably however, the number of patients that have a successful stimulation trial is significantly higher than the number of patients who benefit from the procedure at long-term (particularly when neuropathic pain is considered) [73,75,83,96]. The reason for this is still unclear. It is possible that plasticity of neuronal circuits might play a role, culminating with the adaptation of the cerebral tissue to stimulation with time. It is also possible that some of the patients might overstate the benefits of stimulation to ‘‘pass’’ the postoperative trial and get an IPG implanted. Another factor that needs to be taken into account,
131
is the phenomenon of tolerance, predominantly described after PAG/PVG stimulation [39,97]. This consists on the need for deliverance of higher and higher currents to obtain similar effects over time. It is unclear whether this phenomenon might be related to that reported for opiates. Although animal experiments predicted facilitation or cross-tolerance between DBS and opiates [98–100], such effects were not observed when various drugs or stimulation holidays were used to prevent or treat tolerance in humans [101]. As a final remark, one cannot underestimate the role of a placebo effect in patients with chronic pain undergoing DBS [102]. Independent on the reasons for the decrease in the number of patients benefiting from surgery at long-term, a worst outcome has frequently been reported in more recent trials, as compared to older ones [38,86]. These poorer results are in line with the findings of two open-labeled multicenter studies sponsored by one of the manufacturers of the stimulators (Medtronic) [101]. Because of the uninspiring results of these studies, the company did not apply for marketing approval to treat pain with DBS [101]. Based on the findings described above, one may then question: ‘‘Should we continue to offer DBS for the treatment of chronic pain?’’ Our impression is that we should. Though less invasive effective therapeutic alternatives exist for nociceptive pain, patients with refractory neuropathic pain have basically two surgical options: DBS or motor cortex stimulation (also of unknown benefit across different pain pathologies). Most importantly, even though the proportion of responders to DBS at long-term is not very high, patients that do respond have substantial longstanding improvements, with reductions in VAS scores in the order of 50– 80% [38,71,72,74,75,83,86]. Characterizing prognostic factors for a successful outcome would then be crucial for selecting the best candidates for the procedure.
2233
2234
131
Dbs for persistent non-cancer pain
Potential Prognostic Factors Identified to Date Patients with nociceptive pain seem to respond better to DBS than patients with neuropathic pain [71–75,83,86]. If only the later diagnosis is taken into account, patients with brachial plexus avulsion, complex regional pain syndrome, and peripheral neuropathies have a better response to stimulation than those with postherpetic neuralgia or thalamic pain [72–75,86,103]. As for other surgical treatments, the presence of psychological or litigation problems forecast a poor prognosis [72,73]. Aside from the proper selection of patients, the presence of an insertional effect seems to predict a good response to the postoperative stimulation trials [38]. It is also our impression that long-term responders spontaneously report a clear-cut analgesic effects of stimulation as soon as the test trials are conducted [38]. Patients with Vc electrodes that benefit from the procedure often describe pleasant paresthesias in the regions of pain, feeling really comfortable with the quality of the sensation experienced [10,38,39,72,73,75,83,86]. A good pain relief with PAG/PVG DBS is often described in patients that experience a pleasant warmth feeling during stimulation [13,19,48,72–75,83,86]. In a recent series of reports, field potentials were recorded in the thalamus during PVG stimulation [104,105]. When the electrodes were either ‘‘off’’ or ‘‘on’’ at 50 Hz (a frequency that was not associated with analgesic effects), 0.2–0.4 Hz potentials were systematically recorded in the region of Vc. Interestingly, a significant reduction in the amplitude of these potentials was observed when the PVG electrode was turned ‘‘on’’ at frequencies that induced analgesia (5–35 Hz). Based on these results, the authors suggested that a decrease in low frequency thalamic potentials after PVG stimulation could predict a good therapeutic response [104,105].
Complications Hardware and surgical complications of the DBS surgery for pain are similar to those of DBS in other disorders [72–75,106]. There is a 2–3% risk of intracranial hemorrhages, mostly asymptomatic. Lead problems occur in 4–5% of the patients, including lead migration, breakage of the wires, and leads that need to be repositioned. Infections may occur in 3–5% of the patients, with approximately 50% requiring the removal of parts or even the entire DBS system [96]. Stimulation-induced adverse effects have mainly been reported with PAG/PVG DBS. These are similar to the ones previously described during the surgical procedures, when the electrodes are initially turned ‘‘on’’ in the operating room. In addition, migraine-like headaches have also been reported in patients treated with PAG/PVG DBS, but those are likely unrelated to stimulation [73,83].
Conclusion and Perspectives of DBS for Pain The long-term outcome of DBS for the treatment of chronic pain is quite variable, with most studies showing a response in 20–80% of the patients treated. Part of this variability may be attributed to the different conditions treated, surgical targets used, the lack of consistency in the methods to evaluate a clinical response, different lengths of follow-up, the use of analgesic drugs or a combination of the above mentioned factors. Recently published series as well as two double blinded multicenter trials sponsored by one of the manufacturers of the device have shown a worst outcome with the procedure when compared to older studies [38,86,95]. These poorer results were a consequence of either a smaller percentage of patients considered to be responders, or a significantly higher proportion of
Dbs for persistent non-cancer pain
patients that discontinue stimulation at longterm. In spite of these facts, DBS is still considered part of the armamentarium for the treatment of chronic pain because patients who do respond have significant long-lasting improvements. This urges future studies to more appropriately investigate these procedures. Limitations of the previously published series should be taken into account and features such as the inclusion of appropriate control groups, the concomitant use of analgesics, and the use accepted methods of data collection and analysis should be a systematic part of newly planned trials. In addition, it will be necessary to investigate more homogeneous groups of patients, classified according to the etiological diagnosis of their pain syndromes. This will likely contribute to a better understanding of potential predictors of response and help to establish strict inclusion criteria for DBS.
References 1. Heath RG. Studies in schizophrenia: a multidisciplinary approach to mind-brain relationships. Cambridge, Massachusetts: Harvard University Press; 1954. 2. Heath RG, Mickle WA. Evaluation of seven years’ experience with depth electrode studies in human patients, In: Ramey ER, O’Doherty DS, editors. Electrical studies on the unanesthetized human brain. New York: Paul B. Hoeber; 1960. p. 214-47. 3. Pool JL, Clark WK, Hudson P, Lombardo M. Steroid hormonal response to stimulation of electrodes implanted in the subfrontal parts of the brain. In: Fields In:WS, Guillemin R, Carton CA, editors. Hypothalamic-hypophysial interrelationships, a symposium. Sprigfield, IL: Charles C. Thomas; 1956. p. 114-24. 4. Mazars G, Merienne L, Cioloca C. Treatment of certain types of pain with implantable thalamic stimulators. Neurochirurgie 1974;20:117-24. 5. Mazars G, Merienne L, Ciolocca C. Intermittent analgesic thalamic stimulation. Preliminary note. Rev Neurol (Paris) 1973;128:273-9. 6. Mazars G, Roge´ R, Mazars Y. Re´sultats de la stimulation du faisceau spinothalamique et leur incidence sur la physiopathologie de la douleur. Rev Neurol (Paris) 1960;103:136-8.
131
7. Mazars GJ. Intermittent stimulation of nucleus ventralis posterolateralis for intractable pain. Surg Neurol 1975;4:93-5. 8. White JC, Sweet WH. Pain and the neurosurgeon: a 40yearexperience.Springfield, IL: Charles C. Thomas; 1969. 9. Hosobuchi Y, Adams JE, Rutkin B. Chronic thalamic stimulation for the control of facial anesthesia dolorosa. Arch Neurol 1973;29:158-61. 10. Hosobuchi Y, Adams JE, Rutkin B. Chronic thalamic and internal capsule stimulation for the control of central pain. Surg Neurol 1975;4:91-2. 11. Adams JE, Hosobuchi Y, Fields HL. Stimulation of internal capsule for relief of chronic pain. J Neurosurg 1974;41:740-4. 12. Fields HL, Adams JE. Pain after cortical injury relieved by electrical stimulation of the internal capsule. Brain 1974;97:169-78. 13. Richardson DE, Akil H. Long term results of periventricular gray self-stimulation. Neurosurgery 1977;1:199-202. 14. Richardson DE, Akil H. Pain reduction by electrical brain stimulation in man. I. Acute administration in periaqueductal and periventricular sites. J Neurosurg 1977;47:178-83. 15. Richardson DE, Akil H. Pain reduction by electrical brain stimulation in man. II. Chronic self-administration in the periventricular gray matter. J Neurosurg 1977;47: 184-94. 16. Hosobuchi Y, Adams JE, Linchitz R. Pain relief by electrical stimulation of the central gray matter in humans and its reversal by naloxone. Science 1977;197:183-6. 17. Dieckmann G, Witzmann A. Initial and long-term results of deep brain stimulation for chronic intractable pain. Appl Neurophysiol 1982;45:167-72. 18. Gybels J, Kupers R. Central and peripheral electrical stimulation of the nervous system in the treatment of chronic pain. Acta Neurochir Suppl (Wien) 1987;38:64-75. 19. Hosobuchi Y. Subcortical electrical stimulation for control of intractable pain in humans. Report of 122 cases (1970–1984). J Neurosurg 1986;64:543-53. 20. Lenz FA, Tasker RR, Dostrovsky JO, Kwan HC, Gorecki J, Hirayama T, Murphy JT. Abnormal single-unit activity recorded in the somatosensory thalamus of a quadriplegic patient with central pain. Pain 1987; 31:225-36. 21. Plotkin R. Results in 60 cases of deep brain stimulation for chronic intractable pain. Appl Neurophysiol 1982;45:173-8. 22. Siegfried J, Lazorthes Y, Sedan R. Indications and ethical considerations of deep brain stimulation. Acta Neurochir Suppl (Wien) 1980;30:269-74. 23. Tasker RR, Vilela Filho O. Deep brain stimulation for neuropathic pain. Stereotact Funct Neurosurg 1995;65:122-4.
2235
2236
131
Dbs for persistent non-cancer pain
24. Turnbull IM, Shulman R, Woodhurst WB. Thalamic stimulation for neuropathic pain. J Neurosurg 1980;52:486-93. 25. Young RF, Kroening R, Fulton W, Feldman RA, Chambi I. Electrical stimulation of the brain in treatment of chronic pain. Experience over 5 years. J Neurosurg 1985;62:389-96. 26. Reynolds DV. Surgery in the rat during electrical analgesia induced by focal brain stimulation. Science 1969;164:444-5. 27. Dostrovsky JO. Stimulation-produced antinociception. Prog Brain Res 1988;77:159-64. 28. Oliveras JL, Besson JM. Stimulation-produced analgesia in animals: behavioural investigations. Prog Brain Res 1988;77:141-57. 29. Reichling DB, Kwiat GC, Basbaum AI. Anatomy, physiology and pharmacology of the periaqueductal gray contribution to antinociceptive controls. Prog Brain Res 1988;77:31-46. 30. Gebhart GF, Toleikis JR. An evaluation of stimulationproduced analgesia in the cat. Exp Neurol 1978;62:570-9. 31. Oliveras JL, Besson JM, Guilbaud G, Liebeskind JC. Behavioral and electrophysiological evidence of pain inhibition from midbrain stimulation in the cat. Exp Brain Res 1974;20:32-44. 32. Oliveras JL, Guilbaud G, Besson JM. A map of serotoninergic structures involved in stimulation producing analgesia in unrestrained freely moving cats. Brain Res 1979;164:317-22. 33. Oliveras JL, Woda A, Guilbaud G, Besson JM. Inhibition of the jaw opening reflex by electrical stimulation of the periaqueductal gray matter in the awake, unrestrained cat. Brain Res 1974;72:328-31. 34. Olds ME, Olds J. Approach-escape interactions in rat brain. Am J Physiol 1962;203:803-10. 35. Wolfe TL, Mayer DJ, Carder B, Liebeskind JC. Motivational effects of electrical stimulation in dorsal tegmentum of the rat. Physiol Behav 1971;7:569-74. 36. Kiser RS, Lebovitz RM, German DC. Anatomic and pharmacologic differences between two types of aversive midbrain stimulation. Brain Res 1978;155:331-42. 37. Waldbillig RJ. Attack, eating, drinking, and gnawing elicited by electrical stimulation of rat mesencephalon and pons. J Comp Physiol Psychol 1975;89:200-12. 38. Hamani C, Schwalb JM, Rezai AR, Dostrovsky JO, Davis KD, Lozano AM. Deep brain stimulation for chronic neuropathic pain: long-term outcome and the incidence of insertional effect. Pain 2006;125:188-96. 39. Kumar K, Wyant GM, Nath R. Deep brain stimulation for control of intractable pain in humans, present and future: a ten-year follow-up. Neurosurgery 1990;26:774-81; discussion 772–81. 40. Jacquet YF, Lajtha A. Morphine action at central nervous system sites in rat: analgesia or hyperalgesia depending on site and dose. Science 1973;182:490-2.
41. Sharpe LG, Garnett JE, Cicero TJ. Analgesia and hyperreactivity produced by intracranial microinjections of morphine into the periaqueductal gray matter of the rat. Behav Biol 1974;11:303-13. 42. Yaksh TL, Rudy TA. Narcotic analgestics: CNS sites and mechanisms of action as revealed by intracerebral injection techniques. Pain 1978;4:299-359. 43. Yaksh TL, Yeung JC, Rudy TA. Systematic examination in the rat of brain sites sensitive to the direct application of morphine: observation of differential effects within the periaqueductal gray. Brain Res 1976;114:83-103. 44. Cannon JT, Prieto GJ, Lee A, Liebeskind JC. Evidence for opioid and non-opioid forms of stimulation-produced analgesia in the rat. Brain Res 1982;243:315-21. 45. Akil H, Richardson DE, Hughes J, Barchas JD. Enkephalin-like material elevated in ventricular cerebrospinal fluid of pain patients after analgetic focal stimulation. Science 1978;201:463-5. 46. Hosobuchi Y, Rossier J, Bloom FE, Guillemin R. Stimulation of human periaqueductal gray for pain relief increases immunoreactive beta-endorphin in ventricular fluid. Science 1979;203:279-81. 47. Adams JE. Naloxone reversal of analgesia produced by brain stimulation in the human. Pain 1976;2:161-6. 48. Young RF, Chambi VI. Pain relief by electrical stimulation of the periaqueductal and periventricular gray matter. Evidence for a non-opioid mechanism. J Neurosurg 1987;66:364-71. 49. Millan MJ. Descending control of pain. Prog Neurobiol 2002;66:355-474. 50. Willis WD, Westlund KN. Neuroanatomy of the pain system and of the pathways that modulate pain. J Clin Neurophysiol 1997;14:2-31. 51. Gerhart KD, Yezierski RP, Wilcox TK, Willis WD. Inhibition of primate spinothalamic tract neurons by stimulation in periaqueductal gray or adjacent midbrain reticular formation. J Neurophysiol 1984;51: 450-66. 52. Pereira EA, Green AL, Bradley KM, Soper N, Moir L, Stein JF, Aziz TZ. Regional cerebral perfusion differences between periventricular grey, thalamic and dual target deep brain stimulation for chronic neuropathic pain. Stereotact Funct Neurosurg 2007;85:175-83. 53. Rezai AR, Lozano AM, Crawley AP, Joy ML, Davis KD, Kwan CL, Dostrovsky JO, Tasker RR, Mikulis DJ. Thalamic stimulation and functional magnetic resonance imaging: localization of cortical and subcortical activation with implanted electrodes. Technical note. J Neurosurg 1999;90:583-90. 54. Head H, Holmes G. Sensory disturbances from cerebral lesions. Brain 1911;34:102-254. 55. Hua SE, Garonzik IM, Lee JI, Lenz FA. Microelectrode studies of normal organization and plasticity of human somatosensory thalamus. J Clin Neurophysiol 2000;17:559-74.
Dbs for persistent non-cancer pain
56. Lenz FA, Seike M, Lin YC, Baker FH, Rowland LH, Gracely RH, Richardson RT. Neurons in the area of human thalamic nucleus ventralis caudalis respond to painful heat stimuli. Brain Res 1993;623:235-40. 57. Dostrovsky JO. Role of thalamus in pain. Prog Brain Res 2000;129:245-57. 58. Lenz FA, Seike M, Richardson RT, Lin YC, Baker FH, Khoja I, Jaeger CJ, Gracely RH. Thermal and pain sensations evoked by microstimulation in the area of human ventrocaudal nucleus. J Neurophysiol 1993;70:200-12. 59. Lenz FA, Dostrovsky JO, Tasker RR, Yamashiro K, Kwan HC, Murphy JT. Single-unit analysis of the human ventral thalamic nuclear group: somatosensory responses. J Neurophysiol 1988;59:299-316. 60. Lenz FA, Gracely RH, Rowland LH, Dougherty PM. A population of cells in the human thalamic principal sensory nucleus respond to painful mechanical stimuli. Neurosci Lett 1994;180:46-50. 61. Tasker RR. Microelectrode findings in the thalamus in chronic pain and other conditions. Stereotact Funct Neurosurg 2001;77:166-8. 62. Vilela Filho O. Thalamic ventrobasal stimulation for pain relief. Probable mechanisms, pathways and neurotransmitters. Arq Neuropsiquiatr 1994;52:578-84. 63. Benabid AL, Henriksen SJ, McGinty JF, Bloom FE. Thalamic nucleus ventro-postero-lateralis inhibits nucleus parafascicularis response to noxious stimuli through a non-opioid pathway. Brain Res 280:217–31. 64. Gerhart KD, Yezierski RP, Fang ZR, Willis WD. Inhibition of primate spinothalamic tract neurons by stimulation in ventral posterior lateral (VPLc) thalamic nucleus: possible mechanisms. J Neurophysiol 1983;49:406-23. 65. Tsubokawa T, Yamamoto T, Katayama Y, Moriyasu N. Clinical results and physiological basis of thalamic relay nucleus stimulation for relief of intractable pain with morphine tolerance. Appl Neurophys 1982;45:143-55. 66. Vilela Filho O, Tasker RR. Pathways involved in thalamic ventrobasal stimulation for pain relief: evidence against the hypothesis VB stimulation→rostroventral medulla excitation→dorsal horn inhibition. Arq Neuropsiquiatr 1994;52:386-91. 67. Davis KD, Taub E, Duffner F, Lozano AM, Tasker RR, Houle S, Dostrovsky JO. Activation of the anterior cingulate cortex by thalamic stimulation in patients with chronic pain: a positron emission tomography study. J Neurosurg 2000;92:64-9. 68. Duncan GH, Kupers RC, Marchand S, Villemure JG, Gybels JM, Bushnell MC. Stimulation of human thalamus for pain relief: possible modulatory circuits revealed by positron emission tomography. J Neurophysiol 1998;80:3326-30. 69. Kupers RC, Gybels JM, Gjedde A. Positron emission tomography study of a chronic pain patient successfully treated with somatosensory thalamic stimulation. Pain 2000;87:295-302.
131
70. Aiko Y, Shima F, Hosokawa S, Kato M, Kitamura K. Altered local cerebral glucose utilization induced by electrical stimulations of the thalamic sensory and parafascicular nuclei in rats. Brain Res 1987;408:47-56. 71. Bittar RG, Kar-Purkayastha I, Owen SL, Bear RE, Green A, Wang S, Aziz TZ. Deep brain stimulation for pain relief: a meta-analysis. J Clin Neurosci 2005;12:515-19. 72. Levy RM. Deep brain stimulation for the treatment of intractable pain. Neurosurg Clin N Am 2003;14:389-399, vi. 73. Levy RM, Lamb S, Adams JE. Treatment of chronic pain by deep brain stimulation: long term follow-up and review of the literature. Neurosurgery 1987;21:885-93. 74. Rezai AR, Lozano AM. Deep brain stimulation for chronic pain. In Burchiel KJ, editor. Surgical management of pain. New York: Thieme; 2002. p. 565-74. 75. Wallace BA, Ashkan K, Benabid AL. Deep brain stimulation for the treatment of chronic, intractable pain. Neurosurg Clin N Am 2004;15:343-57, vii. 76. Schvarcz JR. Chronic stimulation of the septal area for the relief of intractable pain. Appl Neurophysiol 1985;48:191-4. 77. Schvarcz JR. Long-term results of stimulation of the septal area for relief of neurogenic pain. Acta Neurochir Suppl (Wien) 1993;58:154-5. 78. Andy OJ. Parafascicular-center median nuclei stimulation for intractable pain and dyskinesia (painful-dyskinesia). Appl Neurophysiol 1980;43:133-44. 79. Andy OJ. Thalamic stimulation for chronic pain. Appl Neurophysiol 1983;46:116-23. 80. Schvarcz JR. Chronic self-stimulation of the medial posterior inferior thalamus for the alleviation of deafferentation pain. Acta Neurochir Suppl (Wien) 1980;30:295-301. 81. Thoden U, Doerr M, Dieckmann G, Krainick JU. Medial thalamic permanent electrodes for pain control in man: an electrophysiological and clinical study. Electroencephalogr Clin Neurophysiol 1979;47:582-91. 82. Spooner J, Yu H, Kao C, Sillay K, Konrad P. Neuromodulation of the cingulum for neuropathic pain after spinal cord injury. Case report. J Neurosurg 2007;107:169-72. 83. Kumar K, Toth C, Nath RK. Deep brain stimulation for intractable pain: a 15-year experience. Neurosurgery 1997;40:736-46; discussion 737–46. 84. Bittar RG, Burn SC, Bain PG, Owen SL, Joint C, Shlugman D, Aziz TZ. Deep brain stimulation for movement disorders and pain. J Clin Neurosci 2005;12:457-63. 85. Owen SL, Green AL, Stein JF, Aziz TZ. Deep brain stimulation for the alleviation of post-stroke neuropathic pain. Pain 2006;120:202-6. 86. Rasche D, Rinaldi PC, Young RF, Tronnier VM. Deep brain stimulation for the treatment of various chronic pain syndromes. Neurosurg Focus 2006;21:E8. 87. Andrade-Souza YM, Schwalb JM, Hamani C, Hoque T, Saint-Cyr J, Lozano AM. Comparison of 2-dimensional magnetic resonance imaging and 3-planar reconstruction
2237
2238
131 88.
89.
90.
91.
92.
93.
94.
95.
96.
Dbs for persistent non-cancer pain
methods for targeting the subthalamic nucleus in Parkinson disease. Surg Neurol 2005;63:357-62; discussion 353–62. Shenai MB, Ross DA, Sagher O. The use of multiplanar trajectory planning in the stereotactic placement of depth electrodes. Neurosurgery 2007;60:272-6; discussion 276. Davis KD, Kiss ZH, Tasker RR, Dostrovsky JO. Thalamic stimulation-evoked sensations in chronic pain patients and in nonpain (movement disorder) patients. J Neurophysiol 1996;75:1026-37. Lenz FA, Kwan HC, Martin R, Tasker R, Richardson RT, Dostrovsky JO. Characteristics of somatotopic organization and spontaneous neuronal activity in the region of the thalamic principal sensory nucleus in patients with spinal cord transection. J Neurophysiol 1994;72:1570-87. Nandi D, Aziz TZ. Deep brain stimulation in the management of neuropathic pain and multiple sclerosis tremor. J Clin Neurophysiol 2004;21:31-9. Nandi D, Smith H, Owen S, Joint C, Stein J, Aziz T. Peri-ventricular grey stimulation versus motor cortex stimulation for post stroke neuropathic pain. J Clin Neurosci 2002;9:557-61. Owen SL, Green AL, Nandi DD, Bittar RG, Wang S, Aziz TZ. Deep brain stimulation for neuropathic pain. Acta Neurochir Suppl 2007;97:111-16. Siegfried J. Monopolar electrical stimulation of nucleus ventroposteromedialis thalami for postherpetic facial pain. Appl Neurophysiol 1982;45:179-84. Coffey RJ, Lozano AM. Neurostimulation for chronic noncancer pain: an evaluation of the clinical evidence and recommendations for future trial designs. J Neurosurg 2006;105:175-89. Hamani C, Lozano AM. Hardware-related complications of deep brain stimulation: a review of the published literature. Stereotact Funct Neurosurg 2006;84:248-51.
97. Tsubokawa T, Yamamoto T, Katayama Y, Hirayama T, Sibuya H. Thalamic relay nucleus stimulation for relief of intractable pain. Clinical results and beta-endorphin immunoreactivity in the cerebrospinal fluid. Pain 1984;18:115-26. 98. Jacquet YF, Lajtha A. Paradoxical effects after microinjection of morphine in the periaqueductal gray matter in the rat. Science 1974;185:1055-7. 99. Millan MJ, Czlonkowski A, Herz A. An analysis of the ‘tolerance’ which develops to analgetic electrical stimulation of the midbrain periaqueductal grey in freely moving rats. Brain Res 1987;435:97-111. 100. Nichols DS, Thorn BE. Stimulation-produced analgesia and its cross-tolerance between dorsal and ventral PAG loci. Pain 41:347-52. 101. Coffey RJ. Deep brain stimulation for chronic pain: results of two multicenter trials and a structured review. Pain Med 2001;2:183-92. 102. Marchand S, Kupers RC, Bushnell MC, Duncan GH. Analgesic and placebo effects of thalamic stimulation. Pain 2003;105:481-8. 103. Whitworth LA, Fernandez J, Feler CA. Deep brain stimulation for chronic pain. Semin Neurosurg 2004;15:183-93. 104. Nandi D, Aziz T, Carter H, Stein J. Thalamic field potentials in chronic central pain treated by periventricular gray stimulation – a series of eight cases. Pain 2003;101:97-107. 105. Nandi D, Liu X, Joint C, Stein J, Aziz T. Thalamic field potentials during deep brain stimulation of periventricular gray in chronic pain. Pain 2002;97:47-51. 106. Hamani C, Ewerton FI, Bonilha SM, Ballester G, Mello LE, Lozano AM. Bilateral anterior thalamic nucleus lesions and high-frequency stimulation are protective against pilocarpine-induced seizures and status epilepticus. Neurosurgery 2004;54:191-5; discussion 195–7.
135 Facet Denervation R. R. Tasker . Wen Ching Tzaan
Introduction Chronic pain in the midline of the low back, posterior neck, or less often, the dorsal spine, is a common disability that is difficult to treat. The belief that facet joint disease might be responsible for such pain goes back many years [1–4]. Currently, pain of facet origin is thought to be steady, felt over a variable number of vertebrae, extending laterally as far as the proximal limb, but not associated with radicular pain or neurological deficit [5–13]. Occasionally it is not midline or bilateral, but rather strictly unilateral. It is often said to be aggravated by spinal extension. Attempts to ameliorate it surgically appear to have begun in the 1970s, first with Rees’ [14] attempts to cut the facet nerves with a hooked knife and then with Shealy’s [15,16] work with a radio frequency (RF) lesioning electrode. The latter procedure attracted much interest [17–37]. Our own experience [38] will be described here amplified by the work of others.
Anatomy Facet joints are synovial and innervated especially by the medial branch of the primary posterior ramus of the segmental spinal nerve at its level, as well as by other contributions including the medial branch one level above, even with contributions from one level above that again, and one level below the joint. The medial branch also innervates the paraspinal muscles at the same level. Though there are varying opinions as to whether the denervation should be restricted to the articular branch, paraspinal muscle pain may #
Springer-Verlag Berlin/Heidelberg 2009
be present as well and could be ameliorated if the muscle branches are also divided [39–47].
Indications Patients suitable for facet denervation (often referred to as facet rhizotomy) should have had pain compatible with facet disease as described above for at least 6 months without relief by more conservative measures. Imaging should have excluded other pathology but confirmed the existence of facet degenerative disease. All patients in our series were first referred for local anesthetic blockade of the appropriate facet nerves [48] by an independent therapist enjoying 50% or more temporary reduction of their pain during the test. It doesn’t seem important whether the test consists of a nerve block (reference [16,17,29,30,33,36,49–54] or whether it is a facet block [51,55]. Both tests seem to be suitable predictors of the surgical result. Kwan and Fiel [56] have not found facet denervation useful in the pain of whiplash though it has been used apparently successfully in the treatment of cervicogenic headache [57,58]. Overall, placebo controlled studies revealed better relief after the real operation than after the administration of a placebo [50].
Technique The surgical procedure was performed under general anesthesia without muscle paralysis so that the effects of motor stimulation could be observed.
2292
135
Facet denervation
We have used local anesthetic but the discomfort of RF lesion making proved unacceptable. With sufficient sedation for the lesioning the purpose of using local anesthetic was defeated since lesions had to be made successively as soon as a lesion site was found interfering with the physiological localization of the next site. The denervation was carried out bilaterally in cases of bilateral or midline pain from one vertebral level above to one below the level of the patient’s pain. The denervation in the low back area would be carried out down to the level of the first sacral vertebra. In strictly unilateral pain, not affecting the midline unilateral denervation was done. Though disruption of only the articular branch is necessary to denervate the joint, the medial branch of the posterior primary ramus is usually interrupted in the lumbosacral spine, the primary posterior ramus itself in the cervical and thoracic region. The procedure is done by carrying out radiological and physiological localization as well as lesion making at each site sequentially. Radiological localization is first performed [39] though CT guidance has been described [60]. With the patient in the prone position, the C-arm of the image intensifier is introduced in the AP plane. In the cervical area, the facet rhizotomy needle is directed to the lateral notch of the lateral mass of the vertebral body at the level where innervation is being performed. In the thoracic area, it is introduced between two adjacent ribs adjacent to the thoracic vertebra. In the lumbosacral spine, the needle is aimed to a point where the superior margin of the transverse process abuts the lateral rim of the pedicle. At L5, the needle is introduced into the lumbosacral notch and at S1 into the first dorsal sacral foramen. The facet denervation electrodes used are 15 cm long, 1.1 mm in diameter, insulated except for their 5 mm terminal tip (See > Figure 135-1). These are available from Diros Technology Ontario. They are introduced through a #14 lumbar puncture (LP) needle, which in turn is
. Figure 135-1 Facet denervation electrode courtesy Diros Technology, ON, Canada
inserted through a stab wound. When the LP needle reaches the desired target as seen in the AP image of the image intensifier, its stylet is replaced with the facet electrode and the LP needle is withdrawn sufficiently so that the bare tip does not ground out against the LP needle. Once the radiological position is satisfactory, physiological localization is done [39]. Using the Owl RF generator from Diros Technology, (See > Figure 135-2) two Hertz stimulation is carried out. Once the electrode is suitably positioned, contraction occurs in the ipsilateral paraspinal muscles at thresholds below two volts but contraction should not occur in the somatotopically related limb muscles until the voltage is raised to at least 7 V indicating a safe separation between the electrode tip and the ventral primary ramus. The electrode position needs to be serially changed until satisfactory responses occur. Sometimes no response or only a response at very high voltage is seen despite all electrode position adjustments. This response occurs most often at the L5-S11evel or in patients with previous spinal surgery. If this happens, the surgeon must depend on the radiological localization alone. Once the surgeon is satisfied with electrode position, the temperature monitor (see > Figure 135-3) is inserted and an RF lesion is made at each site starting at 80 C for 90 s, gradually increasing recorded temperature and current flow at the electrode tip until current ‘‘fall-off ’’ occurs signifying a maximum lesion has been made. Single lesions were made at each site unless premature ‘‘fall-off ’’ occurred when the electrode was repositioned and a new lesion attempted. Tissue impedance reflects the tissues at the lesion site and determines current flow. Laser lesioning has been attempted [61].
Facet denervation
135
. Figure 135-2 OWL Universal RF System UR7-3AP Courtesy DIROS Technology, ON, Canada
. Figure 135-3 OWL facet rhizotomy temperature probe – Courtesy DIROS Technology, Ontario, Canada (Diros Technology is located at 232 Hood Rd. Markham, ON L3R 3K8, Canada, [email protected])
Patients could usually be sent home the same day with suitable oral analgesics for the next 5 days [19,23,30].
Results Our own experience [38] will be described. Facet denervations were done on 95 patients, repeated in 23% after immediate failure or early recurrence, totaling 13 in the cervical, 17 in the thoracic and 88 in the lumbosacral spine. Two to five joints were denervated in each patient, failure being most common at the lumbosacral level. Forty-one percent of operations were successful on the first trial (successful meant more than 50% reduction in the patient’s pain rated on a 1–10 scale), with similar results in cervical, thoracic and lumbosacral spine. Mooney’s results were
similar to ours [48]. There was no significant difference in the success rate whether unilateral or bilateral denervation was done, nor did success depend on the number of joints denervated. We abandoned using local anesthetic because lesion making under local anesthesia is too painful. There was no significant difference between results with the two types of anesthesia despite the loss of guidance by subjective responses in patients operated on under general anesthetic. Despite the often alleged diagnostic feature of facet pain that it is aggravated by extension of the spine, we found no difference in the surgical results in the presence or absence of this feature. Though it is often claimed that previous spinal surgery makes facet denervation less successful [15,17,31], we did not find this to be so though a fusion mass may make it difficult to reach the facet nerve. When we repeated the procedure after failure, the second time results were statistically significantly consistent with the first, suggesting that, if a first facet denervation fails to relieve pain, repetition has a low chance of success. Since facet denervations are neurectomies, nerve regeneration and pain recurrence should be expected requiring repetition of the procedure over varying periods of time up to years as observed by Lord et al. [62] who reported pain relief for 3–6 months in 8.3% of his patients, 6–12 months
2293
2294
135
Facet denervation
in 25%, 12–24 months in 8.3%, and over 24 months in 16.7%. The reported success rates of others ranged from 20 to 90% [17–37,63–65].
Complications Three percent of our patients reported aggravation of their preoperative pain after attempted facet denervation for no apparent reason. Ten percent developed neuropathic pain always in the distribution of the medial branch of the primary posterior ramus located in the paravertebral area and it always disappeared in a few months. This complication was rare in the lumbosacral area (3 out of 86 cases), commoner in the cervical and thoracic area) (9 out of 30 cases) reflecting the fact that in the lumbosacral area there are said to be no cutaneous branches of the posterior primary ramus. One patient (1.8%) complained postoperatively of transient ipsilateral leg pain attributed to inadverant heating of the sensory portion of the ventral ramus and two (1.7%) reported subjective ipsilateral leg weakness, persistent in one, without clinical evidence of muscle weakness, wasting or fibrillation.
References 1. Goldthwait JE. The lumbosacral articulation. An explanation of many cases of ‘‘lumbago, sciatica and paraplegia.’’ Boston Med Surg J 1911;164:356-72. 2. Putti V. New concepts in the pathogenesis of sciatic pain. Lancet 1927;2:53-60. 3. Ghormley RK. Low back pain with special reference to the articular facets, with presentation of an operative procedure. JAMA 1933;101:1773-7. 4. Badgley CEO. The articular facets in relationship to low back pain and sciatic radiation. J Bone Joint Surg 1941;23:481-96. 5. Stacey B, Colantonio A, Vookles JL, Sibell D, Kubawiak L. Management of pain by anesthetic techniques. In: Winn HR, editor. Youmans neurological surgery, Vol 3. 5th ed. Philadelphia, PA: Saunders; 2004. p. 2970-86.
6. Fox JL, Rizzoli HV. Identification of radiological coordinates for the posterior articular nerve of Luschka in the lumbar spine. Surg Neuro1 1973;1:343-6. 7. Bogduk N, Marsland A. The cervical zygapophysial joints as a source of neck pain. Spine 1988;13:610-7. 8. Donovan WH, Dwyer AP, White BW, et al. A multidisciplinary approach to chronic low-back pain in western Australia. Spine 1981;6:591-7. 9. Lilius G, Laasonen EM, Myllynen P, Harilainen A, Groniund G. Lumbar facet joint syndrome: a randomized clinical trial. J Bone Joint Surg Br 1989;71:681-4 10. Mehta M, Parry CB. Mechanical back pain and the facet joint syndrome. Disabil Rehabil l994;16:2-12. 11. Mooney V, Robertson J. The facet syndrome. Clin Orthop 1976;115:149-56. 12. Snewing G. Facet joint syndrome: a review. Physiother Can 1984;36:141-4. 13. Wetzel FT. Chronic benign cervical pain syndrome: surgical considerations. Spine 1992;17:S367-74. 14. Rees WES. Multiple bilateral subcutaneous rhizolysis of segmental nerves in the treatment of the intervertebral disc syndrome. Ann Gen Pract 1971;26:126-7. 15. Shealy CN. Percutaneous radiofrequency denervation of spinal facet: treatment for chronic back pain and sciatica. J Neurosurg 1975;43:448-51. 16. Shealy CN. Facet denervation in the management of back and sciatic pain. Clin Orthop 1976;115:157-64. 17. Babur H. Facet rhizotomy for cervical radiculitis. Mt Sinai J Med 1994;61:265-71 18. Banerjee T, Pittman HH. Facet rhizotomy: another armamentarium for treatment of low backache. NCMH 1976;37:354-60. 19. Bogduk N, Long DM. Percutaneous lumbar medial branch neurotomy: a modification of facet denervation. Spine 1980;5:193-200. 20. Bogduk N, Macintosh J, Marsland A. Technical limitations to the efficacy of radiofrequency neurotomy for spinal pain. Neurosurgery 1987;20:529-35. 21. Burton C. Percutaneous radiofrequency facet denervation. Appl Neurophysiol 1976;39:80-6. 22. Florez G, Eiras J, Dcar S. Percutaneous rhhizotomy of the articular nerve of Luschka for low back and sciatic pain. Acta Neurochir (suppl) 1977;24:67-71. 23. Florez G. Eiras J. Ulcar S. Radiofrequency facet denervation in the treatment of persistent headache associated with chronic neck pain. J Neurol Orthop Surg 1980;1:127-30. 24. King JS. Randomized trial ofthe Rees and Shealy methods for the treatment of low back pain. In: Morley TP, editor. Current controversies in neurosurgery. Philadelphia: WB Saunders; 1976. p. 89-94. 25. Lord SM, Barnsley L, Bogduk N. Percutaneous radiofrequency neurotomy in the treatment of cervical zygapophysial joint pain: a caution. Neurosurgery 1995;36:732-9.
Facet denervation
26. Lord M, Barnsley L, Wallis BJ, McDonald GJ, Bogduc N. Percutaneous radiofrequency neurotomy for chronic cervical zygapophyseal-joint pain. N Engl J Med 1996;335:1721-6. 27. McCulloch JA. Percutaneous radiofrequency lumbar rhizolysis (rhizotomy). Appl Neurophysiol 1976;39:87-96. 28. McCulloch JA, Organ LW. Percutaneous radiofrequency lumbar rhizolysis. CMAJ 1977;116:30-2. 29. Mehta M, Sluijter ME. The treatment of chronic back pain: a preliminary survey of the effect of radiofrequency denervation of the posterior vertebral joints. Anaesthesia 1979;34:768-75. 30. Ogsbury JS, Simon RH, Lehman RW. Facet ‘‘denervation’’ in the treatment of low back syndrome. Pain 1977;3:257-63. 31. Oudenhoven RC. Articular rhizotomy. Surg Neurol 1974;2:275-8. 32. Pawl RP. Results in the treatment of low back syndrome from sensory neurolysis of the lumbar facets (facet rhizotomy) by thennal coagulation. Proc Inst Med Chic 1974;30:151-2. 33. Rashbaum RF. Radiofrequency facet denervation: a treatment alternative in refractory low back pain with or without leg pain. Orthop Clin North Am 1983;14:569-75. 34. Schaerer JP. Radiofrequency facet rhizotomy in the treatment of chronic neck and low back pain. Int Surg 1978;63:53-9. 35. Schaerer JP. Treatment of prolonged neck pain by radiofrequency facet rhizotomy. J. Neural Orthop Med Surg1988;9:74-6. 36. Silvers HR. Lumbar percutaneous facet rhizotomy. Spine 1990;15:36-40. 37. Vervest ACM. Stoler RI. The treatment of cervical pain syndromes with radiofrequency procedures. Pain Clinic: 1991;4:103-12. 38. Tzaan, WC, Tasker RR.Percutaneous radiofrequency facet rhizotomy experience with 118 procedures and reappraisal of its value. Can Neuro Sci 2000;27 (2):125-30. 39. Auteroche P. Innervation of the zygapophysial joints of the lumbar spine. Anat Clin 1983;5:17-28. 40. Bogduk N, Long DM. The anatomy of the so-called ‘‘articular nerves’’ and their relationship to facet denervation in the treatment of low back pain. J Neurosurg 1979;51:172-7. 41. Bogduk N. The clinical anatomy ofthe cervical dorsal rami. Spine 1982;7:319-30. 42. Bogduk N, Wilson AS, Tynan W. The human lumbar dorsal rami. J Anat 1982;134:383-97. 43. Edgar MA, Ghadially LA.Innervation of the lumbar spine. Clin Orthop 1976;115:35-41. 44. Groen GJ, Baljet B, Drukker L. Nerves and nerve plexuses of the human vertebral column. Am J Anat 1990;188:282-96.
135
45. Lewin T, Moffett B, Viidik A. The morphology of the lumbar synovial interveretebral joints. Acta Morphol Neerl Scand 1962;4:299-319. 46. Pedersen HE, Blunck CFJ, Gardner E. The anatomy of lumbar posterior rami and meningeal branches of spinal nerves (sinu-vertebral nerves) with an experimental study of their function. J Bone Joint Surg Am 1956;38:377-91. 47. Bradley KC. The anatomy of backache. Aust NZ J Surg 1974;44:227-32. 48. Mooney V. Facet Syndrome. In: Weinstein JN, Wiesle SW, editors. The lumbar spine: the international society for the study of the lumbar spine. Philadelphia: WB Saunders 1990; 422-41. 49. Lippitt AB. The facet joint and its role in spine pain: management with facet joint injections. Spine 1984;9:746-50. 50. Lord SM, Barnsley L, Bogduk N. The utility of comparative local anesthetic blocks versus placebo-controlled blocks for the diagnosis of cervical zygapophysial joint pain. Clin J Pain 1995;11:208-13. 51. Marks RC, Houston T, Thulboume T. Facet joint injection and facet nerve block: a randomized comparison in 86 patients with chronic low back pain. Pain 1992;49:325-8. 52. Revel ME, Listrat VM, Chevalier XJ, et al. Facet joint block for low back pain: identifying predictors of a good response. Arch Phys Med Rehabil 1992;73:824-8. 53. Stolker RJ, Vervest ACM, Groen GJ. The management of chronic spinal pain by blockades: a review. Pain 1994;58:1-20. 54. Wood L. Acute locked facet syndrome and its treatment by manipulation under local periarticular anaesthesia: Part 1. Clinical perspective and pilot study proposal. J Manipulative Physiol Ther 1984;7:211-7. 55. Barnsley L, Lord S, Wallis B, Bogduk N. Falsepositive rates of cervical zygaphophyseal joint blocks. Clin J Pain 1993;9:124-30. 56. Kwan O, Fiel J. Critical appraisal of facet joints injections for chronic whiplash. Med Sci Monitor 2002;8(8): RA191-5. 57. van Suijlekom AA, Weber WE, et al. Radiofrequency cervical zygapophyseal joint neurotomy for cervicogenic headache: a short term follow-up study. Funct Neuro1 1998;13(1):82-3. 58. Blume HG. Treatment of cervicogenic headaches; radiofrequency neurotomy to the sinuvertebral nerves to the upper cervical disc and to the outer layer of the C3 nerve root or C4 nerve root respectively. Funct Neurol 1998;13 (1):83-4. 59. Lord SM, Barns1ey L, Wallis BJ, McDonald GJ, Bogduk N. Percutaneous radiofrequency neurotomy for chronic cervical zygaphophysea1 joint pain. N Eng1 J Med 1996;335:1721-6. 60. Koizuka S, Saito S, et al. Percutaneous radiofrequency lumbar facet rhizotomy guided by computed tomography fluoroscopy. J Anesthesia 2005;19(2):167-9.
2295
2296
135
Facet denervation
61. Iwatsuki K, Yoshimine T, et al. Alternative denervation using laser irradiation in lumbar facet syndrome. Lasers Surg and in Med 2007;39(3):225-9. 62. Lord SM, Barnsley L, Bogduk N. Percutaneous radiofrequency neurotomy in the treatment of cervical zygapophyseal joint pain: a caution. Neurosurgery 1995;36:732-9. 63. Akatov AV, Dreval ON, et al. Transcutaneous radiofrequency destruction of the articular nerves in treat-
ing low back pains. Zhurnal Voprosy Neirokhirurgii Imeni N- N- Burdenko 1997;2:17-20. 64. Van Kleet M, Barendse GA. et al. Randomized trial of radiofrequency lumbar facet denervation for chronic low back pain. Spine 1999;24(18):1937-42. 65. Seres JL. Long-term follow-up of patients treated with cervical radiofrequency neurotomy for chronic neck pain [comment]. Neurosurgery 1999;45 (6):1499-500.
145 Gamma Knife Surgery for Trigeminal Neuralgia and Facial Pain A. C. J. de Lotbinie`re
Historical Introduction The concept of using ionizing radiation for the treatment of facial pain is not new. Shortly following the discovery of ‘‘X-rays’’ by Wilhelm Konrad Roentgen, Hermann Moritz Gocht introduced in 1897 the use of radiation therapy for the relief of pain in a case of breast carcinoma [1]. Later that same year he successfully treated a patient with trigeminal neuralgia, in whom all previous treatments had been met with failure [2]. Complete remission of the neuralgic syndrome was observed 2 days following the intervention. He subsequently treated 20 patients, pain relief occurring in no less than 17 [3]. Author of a popular textbook of practical radiology that was re-edited a number of times [4], he died in 1938 at the age of 68, victim of the effects of unprotected exposure to ionizing radiation. Although radiation therapy continued to be used in a number of cases of trigeminal neuralgia over the next few decades [5], it gradually fell out of use presumably secondary to a higher recurrence rate and/or lack of efficacy. Over half a century after Gocht’s initial use of therapeutic X-rays for the treatment of pain, Lars Leksell cautiously began to use his pioneering technique of stereotactic radiosurgery [6] for patients suffering from trigeminal neuralgia. However it was not until 1971 that he published his first report on two patients with trigeminal neuralgia, each of whom had been treated in 1953 [7]. Both
#
Springer-Verlag Berlin/Heidelberg 2009
patients became pain-free within months following the procedure and were noted to have normal facial sensation at the time of their last follow-up examination 18 years later. Despite the evident success of these cases, it was not until 1996 that an interest was rekindled among clinicians to use this technique following the publication of a multicenter trial using the Gamma Knife for the treatment of medically refractory trigeminal neuralgia [8]. In this study 50 patients underwent radiosurgery focused on the proximal portion of the trigeminal nerve adjacent to the root entry zone, the majority of whom had undergone prior surgical procedures. The proportion of patients who maintained complete pain relief was noted to be significantly greater in those who received a dose of 70 Gy or more when compared to those who received 60–65 Gy. Three patients developed paresthesias and decreased sensation after the radiosurgical intervention, the symptoms resolving completely in one, whereas in another there was subtotal improvement. It was concluded that as a minimally invasive technique, stereotactic radiosurgery should continue to be evaluated as part of a surgical approach to the management of recurrent trigeminal neuralgia or as a primary procedure for the elderly or medically infirm. Since then it has gained its place as a standard therapeutic intervention for the treatment of typical trigeminal neuralgia in those patients who have failed to respond to medical management [9].
2476
145
Gamma knife surgery for trigeminal neuralgia and facial pain
Trigeminal Neuralgia The syndrome of trigeminal neuralgia (TN) or tic douloureux is well known. According to the Revised International Classification of Headache Disorders (ICHD-II) [10] the diagnostic criteria are as follows: 1.
2.
3. 4. 5.
Paroxysmal attacks of pain lasting from a fraction of a second to 2 min, affecting one or more divisions in the trigeminal nerve and fulfilling criteria B and C Pain has at least one of the following characteristics: a. Intense, sharp, superficial, or stabbing b. Precipitated from trigger areas or by trigger factors Attacks are stereotyped in the individual patient There is no clinically evident neurologic deficit Not attributed to another condition
This needs to distinguished from so-called ‘‘atypical’’ trigeminal neuralgia in which pain is noted to arise in the territory of the trigeminal nerve, but lacks the typical characteristics of trigeminal neuralgia. Frequently the pain is described as being constant, triggers are often lacking and examination of the facial sensation may reveal abnormalities such as allodynia or hyperpathia. A more appropriate term to describe this condition would be painful trigeminal neuropathy. The problem for the clinician arises when there is a mixture of the two, as can be noted occasionally in patients with long-standing TN [11]. Not infrequently patients are labeled as having trigeminal neuralgia by referring physicians, when in fact the characteristics of their pain suggest another diagnosis. Given that surgical interventions (including radiosurgical procedures) targeting the trigeminal nerve or its branches are less effective in patients with atypical features when compared to those presenting
with the classical syndrome of trigeminal neuralgia, an accurate diagnosis is essential in order to give a realistic prognosis with regards to a specific procedure. Following the initial publication of Kondziolka et al. noted above, multiple publications have appeared documenting the effectiveness of Gamma Knife stereotactic radiosurgery for medically refractory TN. These can be roughly sorted into two groups: those in whom the root entry zone of the trigeminal nerve is included in the target [12–15] and those in whom the target is the distal portion of the posterior root of the trigeminal nerve proximal to the pars triangularis [16,17]. As might be expected, higher doses (i.e., 90 Gy) in those patients whose target includes the proximal portion of the nerve and root entry zone produce a higher incidence of facial numbness and dysesthesias [18], as is the case in patients who undergo repeat radiosurgical interventions [19]. Results vary from center to center depending on the dose used and reporting criteria, however it can be estimated that approximately 75% of patients will become pain-free initially, but less than 60% maintain this outcome at 2 years, and just over half of the treated patients remain pain-free on or off medications at 3 years [20]. On the other hand those reporting the results of the more distal target at a higher dose of 90 Gy report a success rate that appears to be somewhat better without producing a significant increase in bothersome dysesthesias or numbness. Patients with multiple sclerosis and those with atypical features have outcomes that are significantly worse, as do patients who have undergone prior surgical ablative procedures. As regards the outcome of patients who undergo repeat radiosurgery for idiopathic TN, studies suggest that a similar rate of pain relief is achieved following a second intervention [21]. Although studies have examined irradiating a longer segment of the trigeminal nerve either by adding another 4 mm collimator [22] or
Gamma knife surgery for trigeminal neuralgia and facial pain
8 mm collimator [23], no convincing published evidence has demonstrated the superiority of these techniques compared to the use of a single isocenter despite the finding that the rate of numbness and dysesthesias was higher in the former. Recently a report of a preliminary multicenter study retrospectively analyzed the pooled data for 95 patients who underwent stereotactic radiosurgery using the CyberKnife, a linear accelerator-based technology, to treat patients with idiopathic TN [24]. A nonisocentric treatment plan was used to treat a longer segment of nerve using either a 5 mm or 7.5 mm collimator. Although the initial pain relief was similar to those undergoing a Gamma Knife radiosurgical intervention, the complication rate was significantly higher, 47% of patients experiencing new post-treatment numbness on follow-up examinations. In addition, anesthesia dolorosa was noted to occur in 2 patients, masticator weakness in 2 patients, trismus in 2 patients, decreased hearing in 2 patients, dry eye syndrome in 2 patients, diplopia in 1 patient, and foot paresthesias in 1 patient, complications that the authors ascribe to ‘‘an excessive dose’’ and ‘‘poor planning.’’ Although this technology does not require the application of a stereotactic frame, the results point to a higher complication rate when compared to patients undergoing a Gamma Knife radiosurgical intervention. However with modification of dose planning it is anticipated that the complication rate for patients undergoing CyberKnife radiosurgery will decrease in the future. As has been pointed out by Lopez et al [25], the results of stereotactic radiosurgery for trigeminal neuralgia are comparable to other ablative techniques. However, the lack of uniformity in reporting techniques using universal outcome measures and standardized questionnaires, together with the absence of long-term outcomes in a sufficient population of patients, makes a meaningful comparison of results between different centers difficult, if not impossible. Suffice it to say at present Gamma Knife radiosurgery appears to be
145
the safest technique for the treatment of trigeminal neuralgia, especially in elderly patients in whom the side effects of medications are often poorly tolerated.
Glossopharyngeal Neuralgia Glossopharyngeal neuralgia or vagoglossopharyngeal neuralgia is a syndrome characterized by severe transient stabbing pain that may occur in the ear, at the base of the tongue, in the tonsillar fossa, or beneath the angle of the jaw. Common provoking factors include: swallowing, talking, and coughing. Like trigeminal neuralgia, it may remit and relapse and is often confused with the latter. According to the Revised International Classification of Headache Disorders (ICHD-II) the diagnostic criteria are as follows: 1.
2.
3. 4. 5.
Paroxysmal attacks of facial pain lasting from fractions of a second to 2 min and fulfilling criteria B & C Pain has all of the following characteristics a. Unilateral location b. Distribution within the posterior part of the tongue, tonsillar fossa, and pharynx or beneath the angle of the lower jaw and/or ear c. Sharp, stabbing, and severe d. Precipitated by swallowing, chewing, talking, coughing, and/or yawning Attacks are stereotyped in the individual patient There is no clinically evident neurologic deficit Not attributed to another disease
On occasion glossopharyngeal neuralgia may be associated with hemodynamic instability resulting from reflexive autonomic outflow that can lead to life-threatening syncopal episodes [26]. Following Dandy’s seminal article on the subject [27] in 1927, the surgical treatment of
2477
2478
145
Gamma knife surgery for trigeminal neuralgia and facial pain
glossopharyngeal neuralgia has consisted of sectioning the glossopharyngeal nerve proximal to its entrance into the jugular foramen. Subsequently the technique was modified to include adjacent selected rootlets of the vagus nerve [28]. Fifty years later Laha and Jannetta [29] advocated microvascular decompression for the treatment of glossopharyngeal neuralgia, since which time it has earned its place in the therapeutic armamentarium for the treatment of this disease, long-term followup studies demonstrating sustained relief of pain in the majority of patients [30–32]. Less invasive, percutaneous thermorhizotomy has been used to treat a more limited number of cases with promising results [33]. Based on the success of treating TN with radiosurgery, Stieber et al [34] proposed a noninvasive treatment of glossopharyngeal neuralgia using the Gamma Knife for patients who were deemed medically inoperable or who refused microsurgery. Their patient was offered a microvascular decompression but refused to undergo surgery. The nerve root complex was targeted at its entry into the osseous canal of the jugular foramen based on stereotactic high-resolution CT and MRI scans, 80 Gy was delivered to the nerve with a 4 mm collimator helmet. Six weeks following the procedure the patient noted significant improvement in her pain and by 3 months the pain had resolved allowing the patient to discontinue her medication. At 6 months she experienced a mild recurrence of her pain that did not require medications or another surgical intervention. Since the presentation of this paper in 2004, the author of this chapter has treated two patients with glossopharyngeal neuralgia with Gamma Knife surgery using a similar described target, but at a higher dose of 90 Gy, each patient remaining pain free with follow-up extending over 2 years in one and 3 years in the other. It remains to be seen if these results can be reproduced in a larger series of patients with a longer duration of follow-up.
Cluster Headache Cluster headache is a syndrome characterized by attacks of strictly unilateral excruciatingly severe pain with orbital, supraorbital, or temporal location. Attacks typically last 15–180 min if untreated. According to the Revised International Classification of Headache Disorders (ICHD-II), the headache is accompanied by at least one of the following: 1. 2. 3. 4. 5. 6.
Ipsilateral conjunctival injection and/or lacrimation Ipsilateral nasal congestion and/or rhinorrhea Ipsilateral eyelid edema Ipsilateral forehead and facial sweating Ipsilateral miosis and/or ptosis or A sense of restlessness or agitation
As the name of the syndrome suggests, the attacks are episodic and occur in clusters in between which the patient is free of pain, but in about 10–15% of patients the attacks become chronic such that remissions occur less than a month per year [35]. In some patients the chronic form of the disease may occur at the onset, whereas in others it may develop from the episodic form. Presently this condition is considered to fall under the group of entities entitled trigeminal autonomic cephalgias that include paroxysmal hemicrania and short-lasting unilateral neuralgiform headache attacks with conjunctival injection and tearing (SUNCT syndrome), as well as cluster headaches [36]. It is now recognized that other terms were previously used to describe similar entities: erythroprosopalgia of Bing, ciliary or migrainous neuralgia of Harris, erythromelalgia of the head, Horton’ headache, histaminic cephalgia, petrosal neuralgia of Gardner, sphenopalatine neuralgia, Vidian or Sluder’s neuralgia, and hemicrania periodica neuralgiformis [37]. Surgical ablative procedures utilized to treat medically-refractory cluster headaches generally
Gamma knife surgery for trigeminal neuralgia and facial pain
fall into two groups: those that are focused on the trigeminal nerve and its branches and those that are aimed at interrupting the efferent parasympathetic pathways. The former include sensory trigeminal nerve root section [38], percutaneous trigeminal radiofrequency thermorhizotomy [39], percutaneous glycerol rhizolysis [40] and trigeminal tractotomy [41]. The latter include nervus intermedius section [42], greater superficial petrosal neurectomy [43], and sphenopalatine ganglionectomy [44]. Recently posterior hypothalamic stimulation has been used successfully to treat intractable chronic cluster headache [45] in a small group of patients, the target having been selected on the basis of PET scan activation of this region in patients experiencing an attack of pain [46]. In 1998 Ford et al. published a paper on the treatment of refractory cluster headache in a small group of patients using the Gamma Knife targeting the trigeminal nerve with a 4 mm collimator [47]. Six patients were reported with follow-up outcomes extending from 8 to 14 months. Good to excellent results were reported for all patients, the majority having the chronic cluster headache variant. However, subsequent long-term follow-up (unpublished data) revealed that the majority of patients went on experience a relapse. In another study [48] ten patients were treated with Gamma Knife surgery using a similar target, albeit with a slightly higher dose (80 vs. 70 Gy). The mean follow-up for these patients (13.2 months) was similar to the previous study. Three patients had excellent results and a further three had good results, the remaining four patients being failures. Three patients developed paresthesias, one hypoesthesia, and one deafferentation pain necessitating motor cortex stimulation. Based on the reports of Ford et al. and that of Pollock and Kondziolka [49] describing the Gamma Knife treatment of a patient with sphenopalatine neuralgia, this author began to treat patients with medically intractable chronic cluster headache in 1999 using a combination of the
145
trigeminal target and the sphenopalatine ganglion target described in the above paper. The working hypothesis was that both targets were involved in the symptomatology of chronic cluster headaches, a syndrome known to depend on the integrity of trigeminal-parasympathetic reflexes activated during attacks of pain [50]. At the 12th International Meeting of the Leksell Gamma Knife Society in Vienna, Austria a series of seven patients were reported with a minimum follow-up of 3 years for each case [51]. A 4 mm collimator was used to target the intracisternal portion of the trigeminal nerve with a maximum dose of 80 Gy in three patients and a maximum dose of 90 Gy in four, whereas for the sphenopalatine target (described in the paper by Pollock and Kondziolka) an 8 mm collimator was used with a maximum dose of 90 Gy in each case being directed at the target. Five patients became pain-free, one patient had greater than 50% pain relief, and one patient with bilateral cluster headaches had excellent relief on one side and poor relief on the other. Two patients experienced a relapse of their pain and were treated 15 months and 26 months again following their initial intervention, both of whom experienced excellent results. Four of seven patients developed numbness in the territory of the maxillary division of the trigeminal nerve, one of whom experienced bothersome paresthesias as well. These preliminary results will need to be reproduced in a larger number of patients before this technique can be claimed to offer the definitive treatment for medically intractable chronic (or episodic) cluster headache, although the data suggest that this may be the intervention of choice with the least amount of risk to the patient.
Conclusions The role of stereotactic radiosurgery for the treatment of trigeminal neuralgia is now accepted as being the least invasive of all the surgical
2479
2480
145
Gamma knife surgery for trigeminal neuralgia and facial pain
interventions, having gained acceptance as the procedure of choice in the elderly and in those who tolerate medications poorly. As a greater understanding of the functional neuroanatomy underlying other conditions manifesting with facial pain grows, targets for the interruption or modulation of pathways will be explored and new opportunities for the relief of pain will present themselves to the neurosurgeon dedicated to the relief of pain in this particularly challenging group of patients. As was so eloquently stated by Lars Leksell, ‘‘radiosurgery has established its efficacy and safety and offers an operative system which, combined with sophisticated modern diagnostic methods, makes the depths of the brain more accessible [52].’’
References 1. Gocht HM. Fortschritte auf dem Gebiete der Ro¨ntgenstrahlen; 1897. 2. Artico M, et al. Celebrating the Centennial – Hermann Moritz Gocht and radiation therapy in the treatment of trigeminal neuralgia. Acta Neurochir (Wien) 1997;139: 761-3. 3. Calchi-Novati G. La roentgenterapia nelle nevralgie del trigemino. Rivista di Clinica Medica (Firenze) 1937;38:121-51. 4. Gocht HM. Handbuch der Ro¨ntgen-lehre zum Gebrauche fu¨r Mediziner. Stuttgart: Verlag von Ferdinand Enke; 1911. 5. Pohle EA. Clinical Roentgen therapy. Philadelphia: Lea & Febiger; 1938. p. 511-12. 6. Leksell L. The stereotaxic method and radiosurgery of the brain. Acta Chir Scand 1951;102:316-19. 7. Leksell L. Stereotaxic radiosurgery in trigeminal neuralgia. Acta Chir Scand 1971;137:311-14. 8. Kondziolka D. Stereotactic radiosurgery for trigeminal neuralgia: a multiinstitutional study using the gamma unit. J Neurosurg 1996;84:940-45. 9. Lundsford LD, et al. Radiosurgery practice guideline report #1–03. Harrisburg, PA: International RadioSurgery Association; 2003. 10. Headache Classification Subcommittee of the International Headache Society. The international classification of headache disorders. Cephalgia 2004;24 Suppl 1:1-150. 11. Burchiel KJ, Slavin KV. On the natural history of trigeminal neuralgia. Neurosurgery 2000;46:152-5. 12. Young RF, et al. Gamma knife radiosurgery for treatment of trigeminal neuralgia: idiopathic and tumor related. Neurology 1997;48:1107-09.
13. Rogers CL, et al. Gamma knife radiosurgery for trigeminal neuralgia: the initial experience of the Barrow Neurological Institute. Int J Radiat Oncol Biol Phys 2000;47:1013-19. 14. Kondziolka D, Lundsford LD, Flickinger JC. Stereotactic radiosurgery for the treatment of trigeminal neuralgia. Clin J Pain 2002;18:42-7. 15. Sheehan J, et al. Gamma knife surgery for trigeminal neuralgia: outcomes and prognostic factors. J Neurosurg 2005;102:434-41. 16. Re´gis J, et al. Prospective controlled trial of gamma knife surgery for essential trigeminal neuralgia. J Neurosurg 2006;104:913-24. 17. Massager N, et al. Gamma knife surgery for idiopathic trigeminal neuralgia performed using a far-anterior cisternal target and a high dose of radiation. J Neurosurg 2004;100:597-605. 18. Pollock XX, et al. High-dose trigeminal neuralgia radiosurgery associated with increased risk of trigeminal nerve dysfunction. Neurosurgery 2001;49:58-64. 19. Hasegawa T, et al. Repeat radiosurgery for refractory trigeminal neuralgia. Neurosurgery 2002;50:494-502. 20. Lopez BC, Hamlyn PJ, Zakrzewska JM. Stereotactic radiosurgery for primary trigeminal neuralgia: state of the evidence and recommendations for future reports. J Neurol Neurosurg Psychiatry 2003;75:1019-24. 21. Hasegawa T, et al. Repeat radiosurgery for refractory trigeminal neuralgia. Neurosurgery 2002;50:494-502. 22. Flickinger JC, et al. Does increased nerve length within the treatment volume improve trigeminal neuralgia radiosurgery? A prospective double-blind, randomized study. Int J Radiat Oncol Biol Phys 2001;51:449-54. 23. Kanner AA, et al. Gamma knife radiosurgery for trigeminal neuralgia: comparing the use of a 4-mm versus concentric 4- and 8-mm collimators. Stereotact Funct Neurosurg 2004;82:49-57. 24. Villavavicencio AT, et al. CyberKnife radiosurgery for trigeminal neuralgia treatment: a preliminary multicenter experience. Neurosurgery 2008;62:647-55. 25. Lopez BC, Hamlyn PJ, Zakrzewska JM. Stereotactic radiosurgery for primary trigeminal neuralgia: state of the evidence and recommendations for future reports. J Neurol Neurosurg Psychiatry 2003;75:1019-24. 26. Reddy K, et al. Painless glossopharyngeal ‘‘neuralgia’’ with syncope: a case report and literature review. Neurosurgery 1987;21:916-19. 27. Dandy W. Glossopharyngeal neuralgia (tic douloureux): its diagnosis and treatment. Arch Surg 1927; 15:198-214. 28. Sampson JH, et al. Microvascular decompression for glossopharyngeal neuralgia: long-term effectiveness and complication avoidance. Neurosurgery 2004;54:884-90. 29. Laha RK, Jannetta PJ. Glossopharyngeal neuralgia. J Neurosurg 1977;47:316-20. 30. Resnick DK, et al. Microvascular decompression for glossopharyngeal neuralgia. Neurosurgery 1995;36: 64-9.
Gamma knife surgery for trigeminal neuralgia and facial pain
31. Patel A, et al. Microvascular decompression in the management of glossopharyngeal neuralgia; analysis of 217 cases. Neurosurgery 2002;50:705-11. 32. Kondo A. Follow-up results of using microvascular decompression for treatment of glossopharyngeal neuralgia. J Neurosurg 1997;88:221-25. 33. Giorgi C, Broggi G. Surgical treatment of glossopharyngeal neuralgia and pain from cancer of the nasopharynx. J Neurosurg 1984;61:952-55. 34. Stieber VW, Bourland JD, Ellis TL. Glossopharyngeal neuralgia treated with gamma knife surgery: treatment outcome and failure analysis. J Neurosurg :2005;102 Suppl:155-57. 35. Ekbom K, De Fine Olivarius B. Chronic migrainous neuralgia: diagnostic and therapeutic aspects. Headache 1971;11:97-101. 36. Goadsby PJ. Pathophysiology of cluster headache: a trigeminal autonomic cephalgia. Lancet Neurol 2002;1: 251-57. 37. Goadsby PJ, Tfelt-Hansen P. Cluster headaches: introduction and epidemiology. In: Olesen J, Goadsby PJ, Ramadan NM, Tfelt-Hansen P, Welch KMA, editors. ‘‘The headaches.’’ 3rd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2006. p. 743-45. 38. Onofrio BM, Campbell JK. Surgical treatment of chronic cluster headache. Mayo Clin Proc 1986;61:537-44. 39. Mathew NT, Hurt W. Percutaneous radiofrequency trigeminal gangliorhizolysis in intractable cluster headache. Headache 1988;28:328-31. 40. Sweet WH, Poletti CE, Macon JB. Treatment of trigeminal neuralgia and other facial pains by retrogasserian injection of glycerol. Neurosurgery 1981;9:647-53.
145
41. Green MW. Long-term follow-up of chronic cluster headache treated surgically with trigeminal tractotomy. Headache 2003;43:479-81. 42. Rowed DW. Chronic cluster headache managed by nervus intermedius section. Headache 1990;30:401-06. 43. Gardner MJ, Stowell A, Dutlinger R. Resection of the greater superficial petrosal nerve in the treatment of unilateral headache. J Neurosurg 1968;28:54-60. 44. Meyer JS, et al. Sphenopalatine ganglionectomy for cluster headache. Arch Otolaryngol 1970;92:475-84. 45. Leone M, et al. Long-term follow-up of bilateral hypothalamic stimulation for intractable cluster headache. Brain 2004;127:2259-64. 46. May A, et al. Hypothalamic activation in cluster headache attacks. Lancet 1998;352:275-78. 47. Ford RG, et al. Gamma knife treatment of refractory cluster headache. Headache 1998;38:3-9. 48. Donnet A, Valade D, Re´gis J. Gamma Knife treatment for refractory cluster headache: prospective open trial. J Neurol Neurosurg Psychiatry 2005;76:218-21. 49. Pollock BE, Kondziolka D. Stereotactic radiosurgical treatment of sphenopalatine neuralgia: case report. J Neurosurg 1997;87:450-53. 50. Hardebo JE. On pain mechanisms in cluster headache. Headache 1991;31:314-20. 51. de Lotbiniere ACJ, Knisely JP, Bond JE. Long-term outcome for chronic cluster headaches treated with Gamma Knife stereotactic radiosurgery. Proceedings of the 12th international meeting of the Leksell Gamma Knife Society. Leksell Gamma Knife Society; 2004. p. 115 (Abstract). 52. Leksell L. Stereotactic radiosurgery. J Neurol Neurosurg Psychiatry 1983;46:797-803.
2481
121 History of DBS for Pain D. Richardson
Introduction My interest in techniques for pain control stem from the early 1960s when the only pain procedures available for the control of chronic pain were destructive including rhizotomy, cordotomy (open and percutaneous), open medullary tractotomy and saline frontal lobotomy. The control of pain in the lower extremities was fairly well developed, with use of anterior-lateral cordotomy at T2-3 to sever the ascending spinothalamic tracts in the spinal cord. It was a more difficult to control pain in the upper extremities requiring the cordotomy to be done in the mid or high cervical area and bilateral high cervical cordotomy was complicated by, sometimes fatal, sleep apnea or ‘‘Ondine’s curse’’ [1]. Face pain, at that time, was treated by rhizototomy including the fifth cranial nerve, nervus intermedius, glossopharyngeal nerve, the upper fibers of the vagus nerve, and upper cervical sensory nerve rootlets. Medullary tractotomy was an alternative requiring less surgical exposure but the open posterior fossa approach still required 3–4 h under general anesthesia. These were major undertakings in patients who were seriously ill from head or neck cancer and the results reflected that in my series of five cases, I had two deaths from aspiration pneumonia. For these reasons, I attempted to develop new techniques in the brain itself above the sensory input from the face ‘‘but below the level of awareness,’’ which elicited great interest. Stereotactic surgery had been developed by Irving Cooper for Parkinson’s Disease [2] and I had been exposed to its use as a research technique for deep brain stimulation and recording carried out by Robert G. Heath at Tulane in New Orleans for the study of #
Springer-Verlag Berlin/Heidelberg 2009
schizophrenia and other mental diseases and had access to stereotactic equipment [3]. Stereotactic procedures could be preformed under local anesthesia and would be much safer for debilitated patients with terminal cancer. Looking at the then known anatomical pathways of pain input, through the dorsal root ganglion into the dorsal horn, synapsing and crossing the midline into the contralateral anterior-lateral spinal cord and ascending through the spinothalamic tract and medial lemniscus to the thalamus, the latter seemed an obvious site for a lesion to control pain using stereotactic techniques (> Figure 121-1). A search of the literature revealed little in the way of attempts to control pain by central lesions. Sensory cortical resection had been carried out with poor results [4] and medullary tractotomy, while successful, had a high potential for complications and late failure. Earl Walker, then at Johns Hopkins Hospital [5], had mapped the thalamus in monkeys showing that the sensory receptive areas in the ventromedial and ventrolateral thalamic nuclei (VPM and VPL) of the thalamus were topographically represented and had mapped the VPM/VPL of the monkey thalamus, which I thought was comparable to the human but on a smaller scale. With this information, and the equipment at hand, a young lady with carcinoma of the breast with brachial-plexus invasion and severe intractable pain in the upper extremities was selected to be the first case of coagulation of the VPM/VPL nucleus of the thalamus for pain control. She was operated upon under local anesthesia and an electrode introduced into the VPM/VPL and stimulation carried out. Paresthesias were obtained in the area of pain and numbness in her
2050
121
History of DBS for pain
. Figure 121-1 The spinothalamic pain pathways as known in 1960
knowledge of the final pathway serving chronic pain above the medulla. They postulated a change in function of pathways leading to pain perception by opening a pain gate. This prompted a great deal of interest and attempts to manipulate the paingate leading to the development of peripheral nerve stimulation initially by Wall and Sweet [7] and spinal cord stimulation later by Norman Shealy [8] to manipulate the gate by activating touch/proprioception pathways which would, according to the Melzack-Wall Gate Theory, close the gate for pain pathways in this system.
Neurophysiological Studies
upper extremity and a 5 6 mm lesion was made using radio-frequency thermo-coagulation in this area. The procedure went well, and post-operative by the patient had further decrease in touch and pin prick sensation in the arm. But, her pain was markedly increased despite the numbness produced by the lesion. Failure and poor results can be powerful motivators to achieve new goals. Shortly after this disaster, Melzack and Wall published their paper describing the mechanism for pain perception, which later became known as the Melzack-Wall Gate Theory [6]. This theoretical interpretation of scientific information was made to explain how pain could be perceived in view of the fact that the pain pathway while well known to extend cephalad in the spinothlalmic tract, once the brain stem was reached, there was no
Obviously the anatomical studies that were published up to that time, did not explain the central projections of the chronic pain pathways to centers above the brain stem that were assumed to be necessary for perception of pain. For that reason, we considered physiological studies would be more rewarding in identifying this pathway. The development of the averaging computer to allow reduction of random neural ‘‘noise’’ and recording of evoked potentials was being developed and image retaining oscilloscopes allowed records to be retained and photographed. We were fortunate enough to obtain a Nuclear of Chicago averaging computer on loan from the department of psychiatry and neurology and using this we develop a cat model for sciatic nerve stimulation as input and recordings in the brain stem and thalamus (> Figure 121-2). The durations of recordings for evoked potentials (EP) in most physiological experiments were up to 100 ms following stimulation, but firing of pain related cells was known to be prolonged after activation so we prolonged the averaging over 250–500 ms following sciatic nerve stimulation and a large late wave was observed that could be obliterated by intravenous morphine (MS) (> Figure 121-3). It was also obliterated by dorsal column stimulation and distant peripheral nerve stimulation.
History of DBS for pain
Mapping studies of the brain stem were carried out and the late MS related component from pain input was observed in the reticular formation. A re-review of the anatomical literature revealed that actually the spinothalamic tract should be more correctly called the spinoreticular tract,
. Figure 121-2 Physiological experimental evoked potential recording set-up for sensory input from sciatic nerve. Shows Faraday cage, temperature monitoring, notch filter, ventilator attachment, recording electrode in brain, in cat
121
since approximately 80% of the fibers in the ascending ventrolateral spinal tract terminate in the reticular formation. The other 20% ascend to the VPM/VPL thalamic nucleae. Obviously there were two pathways combined in the spinal cord but once the brain stem was reached these diverged into what we would later call the Paleospinothalamic system and the Neo-spinothalamic system. ‘‘Paleo’’ meaning (philologically) ‘‘old’’ and ‘‘Neo’’ meaning ‘‘new’’ or chronic and acute systems [9] (> Figure 121-4). This led us to believe, based the on the patients that we had operated upon and our animal studies, that the chronic pain system was not related to the spinothalamic pathway, but to the spinoreticular pathway. Our dilemma was compounded by the fact that this was not a system that could be easily lesioned in the brain stem with safety because of the compactness of the structures of the brain stem and the poorly anatomically isolated reticular formation. For this reason, we carried out mapping EP studies more cephalad into the thalamus and were able
. Figure 121-3 Sensory noxious evoked response recorded over 250 ms showing late pain related response (tracing) and obliteration by intravenous morphine 3 mg per kilogram weight (tracing)
2051
2052
121
History of DBS for pain
. Figure 121-4 Cartoon of pain perception based on evoked responses from our experiments in the cat. Direct spinothalamic tract in clear and spinoreticular tract through centromedianum and pulvinar is shaded (reference [9,30])
to obtain the same type of late pain evoked responses in the centromedian nuclei of the thalamus which had also been described by neuroanatomists as the cephalad extension of the reticular system (> Figure 121-5).
Lesions for Chronic Pain The application of these studies was then carried to patients with severe chronic pain from terminal cancer involving the head, neck, and upper extremities. It was found that lesions of the centromedian nucleus produced complete obliteration of chronic pain and could be carried out bilaterally without damage to the patient. The chronic pain component was completely resolved, but the patient’s clinical examination including appreciation of pin prick, touch, pressure and vibratory sensation continued to be completely normal [10].
The most intriguing insight into the neo- and paleo-spinothalamic systems was the observation in a patient at the New Orleans VA Hospital. Following a unilateral CM lesion for pain arising from cancer of the lung invading the lower brachial plexus, his pain was controlled but, while filling his cigarette lighter, he lit spilled lighter fluid on his (previously painful) left arm and received second and third degree burns. The burns were extremely painful for about 10 to 15 min and then the pain completely resolved, he had burn dressings changes and debridement daily without the need for any medication. We believe this demonstrates the difference between acute and chronic pain. The ‘‘neo-spinothalamic’’ tract for new/fast well localized pain fatigues after a few minutes and is replaced by the ‘‘paleospinothalamic’’ system, the older/slower, chronic, poorly localized, non-fatiguing system through the reticular pathway following the painful event. The results of our initial studies of CM lesions for chronic cancer pain were very successful, but then, as some of the patients began to live longer than expected, we began to notice recurrence of the pain after 9–12 months even in the patients with bilateral lesions. In an attempt to prolong the results from this technique, we extended the lesions posteriorly from the centro-median nucleus on into the pulvinar, where we had also found evoked pain potentials [10]. But while it did not damage the patient neurologically it did not seem to prolong the results. We were using a new angular coordinate skull attached stereotactic device at that time which was much smaller than the original device but more prone to error. An incident occurred in the operating room that gave us further insight into modulation of the pain system. A 45-yearold gentlemen with cancer of the lung with extension into the brachial plexus, was being operated upon under local anesthesia following the electrode insertion, but before the x-rays were obtained for position conformation, stimulation was carried out to obtain paresthesias
History of DBS for pain
121
. Figure 121-5 Evoked potential mapping study in the thalamus of the cat. Multiple passes arranged in atlas order, 23 passes in one A-P plain. One millimeter offset, 200 ms recordings, 50 stimuli averaged
which we used to identify the CM (stimulation at high current produces generalized contralateral paresthesias). When the stimulation was started the patient immediately asked if the female nurse or x-ray technician in the operating room was wearing perfume, and then stated that his pain was completely resolved. No one was wearing perfume and once the stimulation was discontinued the olfactory hallucination disappeared and the pain returned in about 60 s. Repeating the stimulation replicated the patient’s pain relief with stimulation, but it returned as the stimulation effects resolved. X-rays of the electrode placement revealed the electrode to be near
the mid-line, but, because of the need to finish the operative procedure, the exact electrode tip position was not measured However, these observations did demonstrate that stimulation in the human central nervous system could very effectively obliterate chronic pain.
Initial Studies on Deep Brain Stimulation for Pain Modulation In an attempt to proceed with studies based on our belief that central modulation of pain was possible with deep brain stimulation we were
2053
2054
121
History of DBS for pain
preparing to do a group of animal studies when I had the opportunity to collaborate with Huda Akil who had finished her graduate research in John Leibskind’s laboratory at the University of California in Los Angles. Here, along with David Mayor, Wotfle, Carter had already carried out the multiple studies in the rat of stimulation-produced analgesia (SPA) in the ventral peri-aqueductal gray (PAG) or raphae nuclei (serotonin system) [11]. In actual fact, this area had been stimulated by Reynolds as published in 1969 [12], but the importance of this site for stimulation-produced analgesia was not appreciated until it was re-discovered in Liebskind’s laboratory. Actually, the studies were started as a test of learning enhancement because lesions in the raphe system had been shown to inhibit learning in rats. The reverse theory, that
increased activity by stimulation of this area could improve learning was to be tested. An attempt to demonstrate this in rats was a total failure and produced absolutely no learning in the rats. Interestingly, the rats were being taught in a Skinner apparatus where they were trained by shocking the floor of the cage to avoid the shock by pressing a lever. The rats ignored the learning shock and, when the system was tested it was found that the rats had good motivational shocking but ignored it. It was realized to be due to the fact that the rats were analgesic as Reynolds had found in his studies previously. These experiments led us to test PAG SPA in our cat model and demonstrated that stimulation of the raphae nucleus produces the same changes in the late pain related evoked potential changes (> Figure 121-6).
. Figure 121-6 Averaged noxious input from sciatic nerve recorded in the parafascicular/centrum medianum complex. Demonstrates obliteration of pain EP by dorsal collum stimulation, central gray (PAG) stimulation and IV morphine. (500 ms, 50 averaged responses,) Shows obliteration of late response by all three manipulations
History of DBS for pain
Initial Deep Brain Stimulation in the Human Akil and I then proceeded to stimulate five patients who were being operated upon to produce CM lesions for cancer pain. With the patient’s permission, a single electrode tract to the ventral PAG area to allow stimulation of the raphe nuclei was designed. In this blinded study of five patients, stimulation was started well above the PAG as control stimulation (> Figure 121-7). Much to our surprise we found that stimulation of the wall of the third ventricle also produced analgesia without the side effects of stimulation of the raphe system directly [13]. Direct stimulation of the PAG area in these patients produced analgesia along with oscillopsia, nausea, and, in some patients, a complaint of impending doom similar to the epigastric rising syndrome described by Penfield from stimulation of the insular cortex [14]. After
121
reassuring ourselves that good analgesia with minimal side effects could be obtained from stimulating the wall of the third ventricle we implanted our first patient in 1973 using electrodes that had been devised for chronic recordings of deep brain structures in schizophrenic patients [15]. The first patient operated upon was implanted with this improvised system using a radio frequency coupled spinal cord stimulation device made by Medtronic1 for spinal cord stimulation. The patient had had multiple major back procedures for intractable pain and had been unable to work for many years. After the electrode was placed, and test stimulation carried out, he obtained complete pain relief without side effects. In the operating room his straight leg raising test improved from pain at 15 to 85 degrees without any pain (> Figure 121-8). He was implanted for chronic stimulation and used the improvised electrode for 3 years before it was replaced with a
. Figure 121-7 Cartoon summery of 5 passes to raphae nucleus (PAG) through paraventricular gray (PVG). All sights gave analgesia except the most dorsal pass (marked with half dark circles) posterior to aqueduct that produces noxious sensation (reference [13])
2055
2056
121
History of DBS for pain
commercially available Medtronic1 electrode that was then available. He returned to full time work as an electrician for many years and is still using his deep brain stimulator for pain control [16]. . Figure 121-8 Photo of intraoperative examination of the first patient chronically implanted for control of chronic pain secondary to post laminectomy syndrome
We had a great deal of consternation about whether this was a placebo effect, knowing that the placebo effect in pain treatment can be as high as 40% of patients. But in 1974, we implanted a terminally ill patient with carcinoma of the lung and bilateral chest and arm pain, who was bed ridden and taking large amounts of narcotics without relief. Following surgery he had complete pain relief from a unilateral deep brain electrode insertion in the PVG and was able to stop all narcotics and leave the hospital to spend the rest of his life pain free at home with his family. Autopsy studies, done by Bill Mehler at Ames Research Center on patients who died of cancer, confirmed that the electrodes were indeed in the wall of the third ventricle [17] (> Figure 121-9). At that time there were no known nuclei in this area and no known tracts that we could associate with the results to explain the pain relief obtained from stimulating the wall of the third ventricle. Demonstration of the ability of naloxone (a specific opiate antagonist) to obliterate SPA in animals [18] and also in our patients [19] made us aware of a connection between opiates and SPA.
. Figure 121-9 Autopsy specimen of patient with successful pain control. Site of electrode in wall of third ventricle (reference [17])
History of DBS for pain
After discovery of the endogenous opiates we were able to measure the release by SPA of opiates into ventricular fluid [20]. It was found that the opiate level in the ventricular cerebrospinal
. Figure 121-10 Endogenous opiates levels in human ventricular fluid following PVG stimulation, shows rapid elevation following SPA by two methods of measurement [20]
121
fluid specimens obtained before and after stimulation showed a dramatic rise in opiate levels (> Figure 121-10). In the lab, single cell recordings from raphe cells showed extreme activation of firing rates by intravenous morphine. This is unlike the response of cells almost anywhere else in the brain, which are usually depressed. Studies in normal humans [28] and in stimulator patients [29] were found to be consistent with the involvement of the endogenous opiate system. Tracing studies carried out by Watson and Barcus [21] using specific antibodies to beta-lipotropin, in the mouse, showed that this was indeed a tract from the hypothalamus along the walls of the third ventricle into the raphe nuclei and locus caeruleus (> Figure 121-11). We then did a study in patients, inserting electrodes along the tract that Watson had identified in mice, and found that SPA could be obtained along this tract in the wall of the third ventricle and into the hypothalamus [22]. However, in the hypothalamus and the more rostral extent of the tract, marked elevation of systemic blood pressure by 80–100 mm/hg systolic occurred within seconds of stimulating in some of these sites in a few patients making clinical use of stimulation here too dangerous for clinical use. In addition to this, some mood
. Figure 121-11 Drawing of Beta-endorphin/beta-lipotropin/ACTH pathway from hypothalamus to periaqueductal gray and locus ceruleus in mouse by immuno-histochemical technique (Watson and Barcus, reference [21])
2057
2058
121
History of DBS for pain
elevation occurred with anterior ventricular wall stimulation and several patients elected to use that electrode for pain control rather than the more posterior electrode just above the posterior third ventricle (> Figures 121-12 and > 121-13). A re-review of the literature revealed that Heath had produced pain relief as well as euphoria with stimulation in this area in a patient that had carcinoma of the cervix serving as a control . Figure 121-12 Composite drawing of SPA sites along the Betaendorphin tract in the human (reference [22])
. Figure 121-13 Lateral X-ray of electrodes placed in anterior and posterior beta-endorphin tract. The patient chose to keep the anterior electrode following a peri-cutaneous trial of stimulation, even though both gave pain relief
for his studies in schizophrenia [3,23]. In 1967, Gol published a single case study showing stimulation in the septal area did indeed produce pain relief [24].
Stimulation of the VPM/VPL Nuclei and Internal Capsule In 1974, when we presented our clinical studies at the American Association of Neurological Surgery meeting, Hosobushi and Adams presented a paper at the same meeting reporting stimulation of the VPM for anesthesia dolorosa following rhizotomy for trigeminal neuralgia with good pain control [25] which confirmed earlier animal studies in our laboratory demonstrating that cross inhibition between neo – and paleo – spinothalamic systems could be produced by stimulation in the thalamus as well as in the peripheral nerve and spinal cord [31,32]. Human studies had shown that stimulation of the VPM/VPL nuclei acted very much like stimulation of the dorsal columns in that the paresthesias had to be located in the patient’s painful area and pain returned very quickly following cessation of stimulation. They later reported that, over time, patients became tolerant to stimulation. However, one patient found that running his voltage up and down (using the external radio frequency induction pulse generator) reduced tolerance [26,27] (> Figure 121-14). This prompted the new stimulator systems developed by the Medtronic Corporation to include a ramping setting to allow this to be done automatically by the completely internalized pulse generators.
Present Indications and Use of Deep Brain Stimulation for Pain Control The more recent development of the endogenous drug pump for analgesic drugs administration has almost done away with the need for stimulation of
History of DBS for pain
. Figure 121-14 AP X-Ray of electrodes placed in both PVG and VPM/ VPL for postoperative trial stimulation
the PVG (opiate system) and VPM/VPL stimulation is used primarily for deafferentation pain in the head and neck or central pain following injury to the brain, since for pain in the extremities or trunk, spinal cord stimulation is usually available. Thus the indication for deep brain stimulation for pain control at this time is primarily for deafferentation pain located in the head, neck and face and usually electrode placement is in the VPM/ VPL or internal capsule. However, looking at the broader view, deep brain stimulation for pain control was the initial study used for the development of the deep brain stimulation systems and was proof of concept for the use of DBS in the modulation and treatment of nervous system diseases now primarily used for the treatment of movement disorders. We are in the early phases of chronic deep brain stimulation for treatment of psychiatric illnesses such as the obsessive-compulsive disorder and depression. The use of deep brain stimulation for upregulation or down-regulation in modulation systems of the nervous system is still in its infancy and hopefully, in the next 10–20 years, many behavioral psychiatric and neurological diseases will be found to be modified, corrected, regulated or cured using this technique.
121
References 1. Mullan S, Hosobuchi Y. Respiratory hazards of high cervical percutaneous cordotomy. J Neurosurg 1968;28(4):291-7. 2. Cooper IS. The neurosurgical alleviation of Parkinsonism. Springfield, IL: Charles C. Thomas; 1956. p. 104. 3. Heath RG. Electrical self-stimulation of the brain in man. Am J Psychiatry 1963;120:571-7. 4 Echols D. Personal communication; 1959. 5. Walker AE. The Primate thalamus. Chicago, IL: The University of Chicago Press; 1938. p. 321. 6. Melzack R, Wall PD. Pain mechanisms: a new theory. Science 1965;150(699):971-9. 7. Wall PD, Sweet WH. Temporary abolition of pain in man. Science 1967;155(758):108-9. 8. Shealy CN, Mortimer JT, Becker DP. Electrical inhibition of pain: experimental evaluation. Anes Analg Curr res 1967;46:299-305. 9. Zorub DS, Richardson DE. An extralemniscal projection to the centrum medianum and pulvinar of the thalamus. Confinia Neurologica 1973;35:356-67. 10. Richardson DE. Thalamotomy for intractable pain. Confinia Neurologica 1967;29:139-45. 11. Mayer OJ, Wotfle TL, Akil H, et al. Analgesia from electrical stimulation in the brainstem of the rat. Science 1971;174:1351-4. 12. Reynolds DV. Surgery in the rat during electrical analgesia induced by focal brain stimulation. Science 1969;164:444-5. 13. Richardson DE, Akil H. Pain reduction by electrical brain stimulation in man. Part I: Acute administration in periaqueductal and periventricular sites. J Neurosurg 1977;47:178-83. 14. Penfield W, Faulk ME Jr. The Insula; further observations on its function. Brain 1955;78:445-70. 15. Llewellyn RC, Heath RG. A Surgical technique for chronic electrode implantation in humans. Confin Neurolgica 1962;22:223-7. 16. Richardson DE, Akil H. Long-term results of periventricular gray self-stimulation. Neurosurgery 1977;1: 199-202. 17. Baskin DS, Mehler WR, Hosobuchi Y, Richardson DE, Adams JE, Flitter MA. Autopsy analysis of the safety, efficacy and cartography of electrical stimulation of the central gray in humans. Brain Res 1986;371(2):231-6. 18. Akil H, Mayer OJ, Liebeskind JC. Antagonism of stimulation-produced analgesia by Naloxone, a narcotic antagonist. Science 1976;191:961-2. 19. Richardson DE, Akil H. Pain reduction by electrical brain stimulation in man. Part 2: Chronic self‐administration in the periventricular gray matter. J Neurosurg 1977;47:184-94. 20. Akil H, Richardson DE, Barchas JO, Li CH. Appearance of beta-endorphin like Immunoreactivity in human
2059
2060
121 21.
22.
23. 24. 25.
26.
History of DBS for pain
ventricular cerebrospinal fluid upon analgesic electrical stimulation. Proc Natl Acad Sci 1978;75:5170-2. Watson SJ, Barchas JD, Li CH. Beta-lipotropin: localization of cells and axons in rat brain by immuno‐ cytochemistry. Proc Natl Acad Sci 1977;74:5155-8. Richardson DE. Analgesia produced by stimulation of various sites in the human beta-endorphin system. Appl Neurophysiol 1982;45:116-22. Heath RG. Studies in Schizophrenia. Cambridge, MA: Harvard University Press; 1954. p. 1054. Gol A. Relief by electrical stimulation of the septal area. J Neurosci 1967;5:115-20. Hosobuchi Y, Adams JE, Rutkin B. Chronic thalamic stimulation for the control of facial anesthesia dolorosa. Arch Neurol 1973;29:158-61. Hosobuchi Y, Adams JE, Rutkin B. Chronic thalamic and internal capsule stimulation for the control of central pain. Surg Neurol 1975;4:91-2.
27. Adams IE, Hosobuchi Y, Fields HL. Stimulation of internal capsule for relief of chronic pain. J Neurosurg 1974;41:740-4. 28. Akil H, Watson SJ, Sullivan S, Barchas JO. Enkephalin like material in normal human CSF: Measurement and levels. Life Sci 1978;23:121-5. 29. Hosobuchi Y, Adams JE, Linchitz R. Pain relief by electrical stimulation of the central gray matter in humans and its reversal by naloxone. Science 1977;197:183-6. 30. Richardson DE, Zorub DS. Sensory function of the pulvinar. Confinia Neurologica 1970;32:165-73. 31. Richardson DE. Autoinhibition in the sensory system of the cat. Surg Forum 1970;21:447-9. 32. Richardson DE. Single unit responses in the neo‐ spinothalamic system of the cat. J Surg Res. 1973;14:472-7.
148 Hypothalamic Stimulation for Cluster Headache A. Franzini . M. Leone . G. Messina . R. Cordella . C. Marras . G. Bussone . G. Broggi
Introduction
Material and Methods
Cluster headache is the most severe of the primary headaches. Chronic cluster headache is not a life threatening condition but may kill the patients either due to drugs abuse, sleep deprivation, or severity of pain leading to suicide. The rationale for high frequency stimulation (HFS) of the posteromedial hypothalamus (pHyp) has been based on advanced functional studies that identified the hypothalamus as the origin of Cluster Headache attacks [1]. We sought to interact with the origin of cluster headache pain by rebalancing the allegedly hyperfunctioning hypothalamic neuronalpools.Deepbrainstimulation(DBS)isthe available methodology to modulate a discrete brain volumeataspecifiedtarget.Thepreviousexperience ofneurologicaldiseasescontrolledbyfocalelectrical stimulation such as Parkinson disease, dystonia, tremor, convinced us to attempt pHyp modulation in a desperate patient affected by chronic cluster headache (CCH) refractory to any conservative and classical treatment including trigeminal thermorhizotomy [1–8] and sphenopalatine ganglion Lidocaine injection. After the first successfully operated patient [3], 15 more patients affected by CCH and 10 more patients affected by painful syndromes affecting the face underwent DBS. Chronic high frequency stimulation of the pHyp has been the first direct therapeutic application of functional neuroimaging data in a restorative reversible procedure for the treatment of an otherwise refractory neurological condition. The surgical technique and the long term results are reported and discussed.
The Diagnosis of CCH
#
Springer-Verlag Berlin/Heidelberg 2009
Intense pain bouts lasting about 10–45 min, often occurring during sleep Pain localized behind the eye or in the eye region and radiating to the forehead, temple, nose, cheek or upper gum on the affected side The affected eyelid may become swollen or droop and the pupil may constrict The nostril on the affected side of the head is often congested Nasal discharge and tearing of the eye is on the same side as the pain Excessive sweating during attacks Face may become flushed on the affected side Four to twelve pain bouts daily for more than a year with no remission or with pain-free periods lasting less than 2 weeks
Selection Criteria for DBS
CCH diagnosis supported by two independent neurologists dedicated to headaches treatment. All drugs for CH prophylaxis tried in sufficient dosages alone and in combination. These comprise verapamil, lithium carbonate, methysergide, valproate, topiramate, gabapentin, melatonin,pizotifen, indomethacin and steroids.
2518
148
Hypothalamic stimulation for cluster headache
Normal CT scan. Normal cerebral MRI including cranio-cervical junction and MRI arterial and venous angiography. Previous ineffective sphenopalatine ganglion endoscopic block by local anesthetic; this procedure alone in our experience may relieve 10% of CCH patients.
Since 2005 we included also chronic stimulation of the great occipital nerve [2–6, 9–16] to rule out patients suitable for peripheral neuromodulation; this procedure in our experience may relieve 30% of CCH patients. One of the two channels of the implanted IPG (Kinetra Medtronic) is connected to the peripheral nerve electrode while the second channel is let free and may be later utilized for DBS just connecting the deep hypothalamic electrode to the free extension of the IPG (> Figure 148-1).
Patients
Sixteen patients fulfilled the selection criteria and therefore the diagnosis of CCH (14 males, mean age at operation 43 years); two of these patients had bilateral pain bouts. One patient had refractory ‘‘Short-lasting Unilateral Neuralgiform Headache attacks with Conjunctival injection and Tearing’’ (SUNCT) without spontaneous or drug induced remissions [2–4,16]. Five patients had neuropathic pain and atypical facial neuralgia localized within the first trigeminal branch (two patients were misdiagnosed as cluster headache before hospital admission). Five patients had refractory trigeminal neuralgia secondary to multiple sclerosis (MS) with involvement of the first division of the
. Figure 148-1 Stimulation of the great occipital nerve performed before the pHyp electrode implant. Note the free extension (E) of the dual channel Kinetra (Medtronic inc. Minneapolis) IPG ready for the connection to the deep brain electrode
Hypothalamic stimulation for cluster headache
fifth nerve (previous microvascular decompression and repeated thermorizotomies had been only temporarily effective).
Stereotactic Methodology The stereotactic implantation was performed with the Leksell frame (Eleckta, Stockholm, Sweden) under local anesthesia. Preoperative antibiotics were administrated to all patients. A preoperative MRI (brain axial volumetric fast spin echo inversion recovery and T2 images) was used to obtain high definition images for the precise determination of the locations of both anterior and posterior commissures and midbrain structures below the commissural plane such as the mammillary bodies and the red nucleus. MR images were fused with 2-mm thick CT slices that were
148
obtained under sterotactic conditions by using an automated technique that is based on a mutualinformation algorithm (Frame-link 4.0, Sofamor Danek Steathstation, Medtronic, Minneapolis, MN). The work-station also provided stereotactic coordinates of the pHyp ipsilateral to the involved side: 3 mm behind the midcommissural point, 5 mm below this point, and 2 mm lateral from the midline (> Figures 148-2–148-4). This same area had been targeted in the seventies for relief of facial pain by K. Sano [17].
Targeting The target planning based exclusively on the midcommissural point caused electrode misplacement in one patient previously reported [13]. This kind of error is due to the anatomical
. Figure 148-2 Hypothalamic target representation (red dots) on AP and lateral ventriculograms. Other targets widely utilized in stereotactic neurosurgery are represented in colors: yellow (Vim complex), pale blue (Subthalamic complex), green (Globus pallidus internus). Lower right: MRI showing the tips of two pHyp. electrodes
2519
2520
148
Hypothalamic stimulation for cluster headache
. Figure 148-3 The interpeduncolar point as utilized to refine the Y stereotactic coordinate in Hypothalamic implants
. Figure 148-4 MRI showing the tips of two pHyp. electrodes and microrecording pattern
individual variability of the angle between the brainstem and the commissural plane. To correct this possible error, we introduced a third anatomical landmark, which allowed final target registration. We called this landmark ‘‘interpeduncular nucleus’’ or ‘‘interpeduncular point’’ [15] and it is placed in the apex of the interpeduncular cistern 8 mm below the commissural plane at the level of the maximum diameter of the mammillary bodies (> Figure 148-3). The Y value of the definitive target (anteroposterior coordinate to the mid-commissural point in the classical midcommissural reference system) was corrected in our patients and the definitive target coordinate was chosen 1–2 mm posterior to the interpeduncular point instead of 3 mm posterior to themidcommissuralpoint. Adedicatedprogram and atlas has been developed and is freely available on the internet to get the proper coordinates of the target (www.angelofranzini.com/BRAIN.htm). A rigid cannula was inserted through a 3 mm, coronal, paramedian twist-drill hole and placed up to 10 mm from the target. This cannula was used as both a guide for microrecording [11] and for the placement of the definitive electrode (Quad 3389; Medtronic).
Microrecording at the Target Three patients were submitted to intraoperative microrecording (> Figure 148-4). They did not take prophylactic drugs for 1 day before the implantation and remained awake throughout the surgical session.Continuousphysiologicalrecordingsbegan as the microelectrode reached the presumptive coordinates of the target, and were performed by means of a Medtronic Leadpoint system. Post-operative data analysis was performed by the Spike2 analysis package (CED, Cambridge, UK). Single unit events were discriminated, and confirmed to arise from a single neuron, using template-matching spike sorting software. The firing rate was calculated by dividing the total
Hypothalamic stimulation for cluster headache
148
number of the isolated spikes by the length of the recording. Properties of the firing pattern were inspected by plotting inter-spike interval histograms (ISIH; 5 ms bin width and lag up to 100 ms). Autocorrelograms (5 ms bin width and lags up to 1,000 ms) were plotted to evaluate the rhythmicity of the spike trains. The average firing rate was around 24 spikes per second. All neurons generated for most of the recordings isolated action potentials, as shown by the highest concentration of intervals in the 10–15 ms range, with 7.2% of ISI shorter than 5 ms, which reflect very high intraburst frequencies. Autocorrelograms of two cells did not display any regularity in the occurrence of peaks and troughs, which indicates a lack of periodicity of the firing discharge. Only one autocorrelogram displayed some regularity in the occurrence of peaks and troughs, with an oscillatory pattern at around 1 Hz. In one patient, firing rate was reduced by contralateral but not by ipsilateral tactile stimulation of the ophthalmic branch [10].
patients had immediate disappearance of pain after surgery before the IPG was activated. Post-operative stereotactic CT was performed toruleoutmalpositionandwasmergedwiththepreoperative MRI to confirm the correct electrode placement [12]. Bilateral implantable pulse generators (IPG) (Soletra, Medtronic, inc.) were placed in subclavicular pockets and connected to the brain electrode for chronic continuous electrical stimulation. Since 2005 we have connected the deep brain electrode to the subclavicular dual channel IPG (Kinetra, Medtronic) previously implanted for Great Occipital Nerve (GON) stimulation.
Macrostimulation at the Target
Results
Increasing progressively the current amplitude at a given frequency (60 Hz) and pulse duration (60 ms) we observed the appearance of ocular movement (deviation toward the stimulated side (3–4 V) followed by an ipsilateral third nerve motor response (4–5 V)) and then the patient began to experience a sense of fear and panic (5–6 V). Vegetative responses and/or cardiovascular effects were not evoked by intraoperative macrostimulation at the amplitudes tested. When side effects were ruled out at the standard parameters of stimulation (1–3 V), the guiding cannula was removed and the electrode secured to the skull with microplates. We never observed adverse mechanical effects due to electrode insertion at the target during surgery but it has to be remarked that three
CCH Series
Definitive Stimulation Parameters The parameters of chronically delivered electrical stimulation were 185 Hz, 60–90 ms and the amplitude ranged between 1 and 3 V in unipolar configuration with case positive. When the IPG is turned on a few days or weeks after surgery, the current amplitude is progressively increased remaining subthreshold for collateral side effects [14,15].
The mean follow-up was 24 months (range 12–62 months). The detailed results recently reported [5] are summarized in the following remarks:
In the whole series, 71% of postoperative days were pain free (> Figure 148-5) and intensity and duration of pain bouts was significatively reduced. Medications have been reduced to less than 20% of the preoperative level. The mean time until the patient was pain free or enjoyed pain reduction was 42 days (range 1–86 days). The mean stimulation amplitude was 2.4 V (range 0.6–3.3 V).
2521
2522
148
Hypothalamic stimulation for cluster headache
. Figure 148-5 Percentage of preoperative days with pain bouts (A) versus the postoperative condition (B) in ten patients affected from CCH monitored along 1 year. The areas at the top of the cylinders represent the intensity of pain attacks (black areas) which were also reduced after surgery (A)
Twelve of the stimulators (nine patients) have been switched off at least once in single-blind fashion. After switching off, pain recurred after an interval of 2 months that seemed to be unrelated to the duration of previous stimulation; the pain improved or disappeared when the stimulator was turned back on. In patients with bilateral crises, the beneficial effects occurred only on the ipsilateral side.
Sunct Pain and autonomic phenomena disappeared after surgery; the amplitude was gradually increased to 1.8 V and after 8 pain-free months the stimulator was turned off with the patient being unaware of it; she remained pain-free for the next 3 months; in month 11, the attacks gradually reappeared and persisted, hence, the stimulator was turned on again and the attacks disappeared. DBS allowed long lasting periods without pain and a major control of pain during sporadic
attacks. The patient returned to a normal social, family and working life [4].
Neuropathic Pain and Atypical Facial Pain After surgery, the patients reported no reduction in their pain. The stimulation parameters were the same as for CCH and SUNCT patients (180 Hz, 60 ms, mean voltage 1.3). After 4 months of continuous stimulation (6, 8, and 10 months, respectively) the continuous pain was the same as preoperatively. Increasing amplitude did not offer any pain relief. Amplitude higher than 3 V induced dizziness and oculomotor activation in all cases. Bipolar stimulation did not offer any improvement. When the IPG was switched off with the patient being unaware of it, the episodes of paroxysmal pain were described by the patient as being slightly more intense than those that occurred during stimulation.
Multiple Sclerosis Trigeminal Neuralgia At 1–3 years follow-up, two out of five operated patients were pain-free and off medication after chronic stimulation, while the remaining patients improved and felt their pain was controlled by adding medication to pHyp stimulation. The DBS had beneficial effects on pain limited to the first trigeminal branch for an average of 23 months. After the implant (median 20 months), three patients underwent a non-DBS surgical procedure to alleviate the pain in the II and III branch, but not in the first.
Discussion and Conclusion DBS in CCH patients resulted in a significant reduction of pain bouts (> Figure 148-5). The
Hypothalamic stimulation for cluster headache
procedure was well tolerated. Transient, reversible diplopia is the main limitation to increasing amplitude. Before the operation, none of the patients was able to work. As a result of stimulation, most patients’ lives have gradually returned to normal; most have resumed work. Nevertheless some problems were identified in our experience:
The diagnosis of CCH must be precise and meet the reported criteria [16]. Comorbidity with other painful syndromes of the face and sometimes with personality disorders [18] may result in an incorrect diagnosis. To avoid this bias in patient selection we suggest strict cooperation with headache specialists, psychiatrists and headache dedicated Units. PHyp stimulation benefits only CCH patients, not other pain syndromes of the face such as atypical facial pain and neuropathic pain [15]. About 30% of CCH patients may have significant improvement after peripheral neuromodulation procedures (GON) suggesting the existence of different subtypes of patients in the same nosographic class. In other words in
148
certain CCH patients the peripheral pathogenetic mechanisms may be more relevant than the central ones. To address this problem we suggest GON stimulation and sphenopalatine ganglion local anesthetic blocks prior to DBS surgery. In the future, PET studies and RM spectroscopy may provide preoperative imaging of hypothalamic involvement in patients affected by CCH [11] allowing further definitions of the indications for DBS. The targeting problems have been resolved adding a third reference point to the stereotactic bicommissural registration procedure. Nevertheless, the use of our target is limited in that we cannot increase the amplitude of delivered current to over 3 V because of the appearance of induced ocular movements. Newtargetswithintheposteriorhypothalamus may be tested to try to avoid this limitation.
Although the current literature suggests the percentage of patients responding to DBS is 50–60% [19,20] the refinement of targeting and patient selection may improve the success rate of Phyp stimulation in CCH patients. Anyway, the use of DBS has changed the prospects of CCH patients.
. Figure 148-6 Schematic representation of the neural network involved in the origin of cluster headache pain and vegetative phenomena
2523
2524
148
Hypothalamic stimulation for cluster headache
Finally, we have to remember that CCH is a dramatic disabling condition leading to abuse of steroids (two patients of the operated series were unable to walk due to severe lower limb myopathy induced by chronic steroid abuse). Also the abuse of triptans may be life-threatening (one patient died before the implant, due to myocardial infarct). DBS benefits this condition and the cost of the procedure is largely compensated for within the first year of induced remission even if the disease cannot be definitively cured by DBS. As for the other painful syndromes of the face, SUNCT responded to stimulation in a similar way to CCH. Atypical facial pain did not respond at all. Trigeminal paroxysmal pain responded only when limited to the first trigeminal division. These data suggest pathogenetic and anatomic links between the pHyp, the first trigeminal division, the reticular formation and the autonomic system of the face [7] (> Figure 148-6).
8.
9.
10.
11.
12.
13.
14.
References 15. 1. May A, Bahra A, Buchel C, Frackowiak RS, Goadsby PJ. Hypothalamic activation in cluster headache attacks. Lancet 1998;352:275-8. 2. Jarrar RG, Black DF, Dodick DW, Davis DH. Outcome of trigeminal nerve section in the treatment of chronic cluster headache. Neurology 2003;60:1360-62. 3. Leone M, Franzini A, Bussone G. Stereotactic stimulation of posterior hypothalamic gray matter in a patient with intractable cluster headache. N Engl J Med 2001;345:1428-9. 4. Leone M, Franzini A, D’Andrea G, Broggi G, Casucci G, Bussone G. Deep brain stimulation to relieve drugresistant SUNCT. Ann Neurol 2005;57:924-7. 5. Leone M, Franzini A, Broggi G, Bussone G. Hypothalamic stimulation for intractable cluster headache: longterm experience. Neurology 2006;67(1):150-2. 6. Leone M, Franzini A, Cecchini AP, Broggi G, Bussone G. Stimulation of occipital nerve for drug-resistant chronic cluster headache. Lancet Neurol 2007;6(4):289-91. 7. Malick A, Strassman AM, Burstein R. Trigeminohypothalamic and reticulohypothalamic tract neurons in
16.
17.
18.
19.
20.
the upper cervical spinal cord and caudal medulla of the rat. J Neurophysiol 2000;84:2078-112. Matharu MS, Goadsby PJ. Persistence of attacks of cluster headache after trigeminal nerve root section. Brain 2002;125:976-84. Burns B, Watkins L, Goadsby PJ. Treatment of medically intractable cluster headache by occipital nerve stimulation: long-term follow-up of eight patients. Lancet 2007;369(9567):1099-106. Cordella R, Carella F, Leone M, Franzini A, Broggi G, Bussone G, Albanese A. Spontaneous neuronal activity of the posterior hypothalamus in trigeminal autonomic cephalalgias. Neurol Sci 2007;28(2):93-5. Lodi R, Pierangeli G, Tonon C, Cevoli S, Testa C, Bivona G, Magnifico F, Cortelli P, Montagna P, Barbiroli B. Study of hypothalamic metabolism in cluster headache by proton MR spectroscopy. Neurology 2006;66(8): 1264-6. Ferroli P, Franzini A, Marras C, Maccagnano E, D’Incerti L, Broggi G. A simple method to assess accuracy of deep brain stimulation electrode placement: preoperative stereotactic CT-postoperative MR image fusion. Stereotact Funct Neurosurg 2004;82:14-19. Franzini A, Ferroli P, Leone M, Bussone G, Broggi G. Hypothalamic deep brain stimulation for the treatment of chronic cluster headaches: a series report. Neuromodulation 2004;7:1-8. Franzini A, Ferroli P, Leone M, Broggi G. Stimulation of the posterior hypothalamus for treatment of chronic intractable cluster headaches: first reported series. Neurosurgery 2003;52:1095-99. Franzini A, Marras C, Tringali G, Leone M, Ferroli P, Bussone G, Bugiani O, Broggi G. Chronic high frequency stimulation of the posteromedial hypothalamus in facial pain syndromes and behaviour disorders. Acta Neurochir Suppl 2007;97(Pt 2):399-406. Headache classification committee of the international headache society. The international classification of headache disorders (2nd ed.). Cephalalgia 2004;24:1–195. Sano K, Mayanagi Y, Sekino H, Ogashiwa M, Ishijima B. Results of stimulation and destruction of the posterior hypothalamus in man. J Neurosurg 1970;33:689-707. Torelli P, Manzoni GC. Pain and behaviour in cluster headache. A prospective study and review of the literature. Funct Neurol 2003;18:205-10. Schoenen J, Di Clemente L, Vandenheede M, et al. Hypothalamic stimulation in chronic cluster headache: a pilot study of efficacy and mode of action. Brain 2005;128:940-7. Starr PA, Barbaro NM, Raskin NH, Ostrem JL. Chronic stimulation of the posterior hypothalamic region for cluster headache: technique and 1-year results in four patients. J Neurosurg 2007;106(6):999-1005.
129 Intrathecal Opiates for Cancer Pain J. C. Sol . J. C. Verdie . Y. Lazorthes
According to the World Health Organization (WHO), an estimated 6.6 million people die from cancer every year [1]. With cancer progression, 62–86% of patients will experience pain [2,3]. Up to 80% of cancer patients described their pain as being moderate to severe intensity [4]. The WHO has established an algorithm of the three-step analgesic ladder that can be use to treat cancer pain [1,5,6]. However, about 10–20% of cancer patients fail to achieve pain relief from using the three-step analgesic ladder and opioid rotation [7]. This failure arise either from pain that is refractory to opioids or the inability of the patients to tolerate the side effects of the opioids at higher doses. It has been suggested that a fourth, ‘‘interventional’’ step be added onto the three-step WHO analgesic ladder [8,9]. One major class of interventions is neuraxial therapy with the administration of intrathecal (I-Th) analgesics at the level of spinal cord [10]. Administration of I-Th offers many advantages over irreversible neurodestructive procedure to control pain. In 1976, Yaksh and Rudy demonstrated the efficacy of I-Th opioids in abolishing pain in animal models [11]. Spinal administration of morphine for pain control in cancer patients was first reported in 1979 by Wang et al. and it was well documented by ventafridda et al [12,13]. Since the 1980s intraspinal drug delivery therapy has been increasingly utilized, first in cancer patients, later on, in patients with non malignant intractable pain who failed to respond to conventional treatment [14–17]. I-Th drug delivery involves drug administration through a small catheter directly into the cerebrospinal fluid (CSF) where medications are #
Springer-Verlag Berlin/Heidelberg 2009
adsorbed directly onto central nervous system receptor sites. I-Th opioid administration produces a direct inhibitory action on modulatory interneurons to produce a decrease in the central transmission of nociceptive impulses [18]. Compared to the systemic intravenous application, the longer lasting therapeutic effect is achieved with a smaller dose and therefore is associated with a reduce rate of side effects. Techniques include external catheters for short term needs and internal pumps for long term use. Opioids have been successfully infused by the I-Th route in the treatment of intractable cancer pain with 64–95% success [19]. Adjunctive I-Th agents such as local anesthetics and a-2 agonists, which cannot be administered effectively or safely by any other route, add to the armamentarium of the pain specialist treating the patient with refractory cancer pain [20,21]. Although direct I-Th administration of morphine in the treatment of intractable cancer pain remains a well established therapeutic alternative, its therapeutic indications are more and more limited because of the development of galenic forms of oral opioids with slow release, new routes of systemic administration, strategies with opioid rotation and limitations related to the administration technique [22,23].
Patients Selection Inclusion Criteria Choosing optimal candidates is a critical factor that dictates the success or failure of I-Th opioid
2172
129
Intrathecal opiates for cancer pain
administration. Although it is not clear at what point in the course of the treatment spinal opioids should be initiated and which patients are ideal candidates, in some cases, the indiscriminate use of spinal opioids is inappropriate and should be discouraged [24,25]. However, in rare cases, it should be considered earlier in the progressive course of the disease and not reserved as a last resort for patients with a short life expectancy [26]. The true incidence of pain requiring spinal analgesia remains unknown, as the size of the group from which patients are selected for spinal analgesia is rarely reported, and spinal opioids are often started before systemic opioid administration is optimized [24]. Only a small proportion of patients (less than 2%) with cancer pain have been reported to be candidates for spinal treatment [6,27]. According to the prevalent and more accepted opinion in the field of cancer pain, indications for the use of spinal opioids should include patients treated by systemic opioids with effective pain relief but with unacceptable side effects, or unsuccessful treatment with sequential strong opioid drug trials despite escalating doses [10,28–30]. Cancer pain with a strong neuropathic component (i.e., plexopathy) may also respond better to I-Th therapy, particularly when adjunctive agents such as local anesthetics and clonidine are used. In the setting of severe pain and the prospect of a very aggressive chemotherapy regimen, when the oral route will likely be an unreliable means of administering analgesics, strong consideration should be given to adopting an I-Th approach to pain and symptom management [31]. Before proceeding to the implantation of a permanent system of I-Th drug delivery (implantable pump with continuous or programmable flow), a scrupulous patient selection process, which rests on a multidisciplinary pain evaluation is essential. One advantage of intraspinal analgesia is the powerful and selective effect, furthermore, a trial of therapy can be undertaken with minimal risk to the patient. If no benefit is seen with trial intraspinal therapy, a more
permanent catheter or pump system need not to be implanted. The decision to employ this technique is complex, involves multiple factors and must be carefully considered. Consequently, a comprehensive multidisciplinary approach to the treatment of cancer pain is optimal [32]. It includes the appropriate palliative antineoplastic therapy, management of analgesics and adjuvant pain medications, behavioral and psychiatric support, and finally interventional therapies [33]. Before proceeding with any invasive pain procedure, communication between the pain physician and the relevant parties (patient, family, members of the care team) is absolutely essential, a thorough history and physical examination of the patient is also critical [34]. Onofrio and Yaksh have developed the classics following inclusion/exclusion criteria [35]: 1.
2.
3. 4.
The patient suffers from a pain state that is unilateral or bilateral including midline and is limited to spinal level T2 and caudally. If arm, head or neck pain is present, intracerebroventricular (ICV) catheters may be more appropriate. Failure to achieve pain relief with doses of systemic narcotics raised to the point of unacceptable side effects that cannot be controlled by opiate rotation or adjunctive therapy. Life expectancy believed to be in excess of 3 months. Favorable response to a test spinal dose of morphine. Several pain states are notably less sensitive to opiates than others, and this sensitivity can be assessed directly by the acute delivery of the test drug.
Exclusion Criteria Several factors should be considered before proceeding with a pump implantation:
Intrathecal opiates for cancer pain
1. 2. 3. 4.
5.
Inability to tolerate the agent that will be infused. Skin infection over the intended site, current bacteremia or sepsis. Coagulopathy that increases the risk of epidural or subdural hematoma. Non patent CSF drainage or the presence of a spinal mass that precludes free distribution. A short anticipated survival time does not automatically exclude the spinal route as a therapeutic modality. However, a short survival time may prevent the patient from obtaining a benefit that outweighs the physical and psychological pressures of a pump implant. Short survival times, however may be more consistent with the placement of a port that allows bolus treatment. Patients with anticipated survival less than 3 months are not candidate for pump placement.
129
techniques and include an assessment of pain, function, mood, and side effects. There are no studies providing guidelines on the level of improvement that is required in this assessment to predict long-term success. The quality of the analgesic activity obtained rather than the duration of test determines efficacy. A response is generally regarded as positive after a test of I-Th administration of morphine if it produces a reduction of more than 50% of the pain intensity without limiting side effects [41]. As we do not know the true relationship between systemic and I-Th dosages of opiates we recommend a single I-Th test dose of 1–2 mg of morphine [42]. Systemic studies in which pain patients are screened for opiate sensitivity are surprisingly few; however, approximately half of cancer pain patients examined for responsiveness to spinal opiates typically fail to respond to the acute test [43].
Intrathecal Drug Delivery System Screening Procedure Before Pump Placement Once intraspinal analgesia has been decided on, first an I-Th trial must be accomplished. The use of a systematic prepump screening procedure serves two purposes. First, it establishes conclusively that the patient’s pain state is sensitive to the drug demonstrating the efficacy in improving pain control [36]. Second, it facilitates selection of the starting drug concentration after the pump placement. There is no standard for neuraxial trials, although numerous approaches have been advocated [37–40]. In general, intrathecally screening trials can be divided into the following: (1) single injection, (2) multiple injections, and (3) continuous infusion. There are no studies supporting one over the other, and clinicians should use their own judgment to decide which technique best suits their practice [40]. Assessments that should be made to determine the success of the trial are similar for all the three
There are basically three types of I-Th delivery techniques in cancer patients: (1) externalized system, (2) Access-port system, and (3) totally implanted system [44]. Each technique has different risks and costs that are taken into account when deciding which approach to use. Parameters of drug delivery (infusions rates and bolus injection volumes) depend on a variety of factors.
Externalized Systems Simple percutaneous I-Th catheter are placed by a minimally invasive technique that can be done at the bedside. Such systems are designed for short term use (weeks). It may be implemented as a trial method to assess efficacy of I-Th opioids, with the intention of placing an implantable system, or it can be used as a definitive method of delivering I-Th drug when life expectancy
2173
2174
129
Intrathecal opiates for cancer pain
is extremely short. Because it is not tunneled, infection risk and mechanical failure are higher, markedly reducing successful long term use. Tunneled I-Th catheters are placed in the operating room under sterile conditions. A dorsally placed catheter is tunneled for a variable distance and usually exits from the lateral abdominal wall. With appropriate infection control measures, including the use of antimicrobial filters and meticulous exit site care, tunneled I-Th can be used for months or years [45,46]. An advantage of the exteriorized system is that non-pain specialist palliative care clinicians may be become proficient in its management with minimal training. Toward the end of life, when analgesic requirements may be escalating rapidly, hospice registered nurses can easily administer I-Th boluses, increase the infusion rate as indicated, and alter the drug(s) used with relative ease. Disadvantages of a tunneled system are the need for an externalized pump apparatus, infection, and the possibility of inadvertent catheter dislodgement or removal.
Access-Port System Partially externalized systems are those in which the catheter is placed by a needle inserted into the target site through a small incision of the skin. The external end of the catheter is then connected to an access port, which is placed under the skin. Injections are made by placing a needle through the skin and into the access port [47]. The subcutaneous system permits freedom of movement, reduces the risk of catheter removal and eases the impediment to patient movement. Several studies showed that subcutaneous port systems are superior to externalized systems if long term use is anticipated [44]. For patients with life expectancy less than 3 months the access port could be an economical pump alternative but needs a daily injection with ambulatory
follow-up period limitation and the increased risk of local and meningeal infection [19,48].
Totally Implanted System Totally implanted systems are those with the catheter and delivery system completely implanted. They have the advantage of a lower risk of infection and allowing the patient more independence. The totally implanted systems are more expensive and require a more involved surgical intervention than the externalized or partially extrenalized system. In 1981, the first I-Th pump was implanted for the delivery of morphine in the treatment of chronic intractable pain [49]. The implantable drug delivery system (IDDS) are based on two uniquely different technologies related to their pumping mechanisms with either fixed or programmable flow rate. All implantable pumps come with a refillable drug reservoir. In addition, they all have an access port system that allows bypassing of the pumping device to permit sampling of CSF or direct injection of a drug into the implanted catheter system. FDA-approved fixed infusion pumps system include Codman’s model 400 (Raynham, MA), Medtronic’s Isomed (Minneapolis, MN), and Arrow’s M-3,000 (Walpole MA). I-Th infusion is based on a valve-controlled drug delivery, which is driven by a two-phased gas bellow. With this gas driven pump, its simplicity of construction gives an indefinite life expectancy. The rate is set by a fixed rate valve based on delivered volume per day. Changing the dose in a fixed rate pump requires the medication concentration to be changed mandating pump refill each time the dose is adjusted [44]. These implantable pump come in a variety of reservoir volumes (up to 60 ml), allowing selection based on patient’s needs and body habitus. Modern implantable pumps such as the Synchromed pump I and II (Medtronic)
Intrathecal opiates for cancer pain
(> Figure 129-1) are programmable computerized electronic pumps with a power source and drug reservoir (20–40 ml). Pump can be refilled through a port accessed by a needle through the skin and programmed by an external hand-held device that can alter the rate of infusion, deliver boluses, and perform other functions. The battery life typically is 5–7 years, so it is unlikely that a replacement would be necessary during the average lifetime of the cancer patient with advanced disease. Drug refill are office-based and minimally uncomfortable, frequency is typically 1–6 months and depends on the infusion rate, which is influenced by the drug concentration and total daily dose [50]. Advantages of these systems, compared with an exteriorized system, include patient freedom, low maintenance, and low infection risk. Disadvantages are the need for an initially more invasive surgery and the logistical problems concerning pump refills and programming. With an implanted system, drug changes are more involved and rapid dose titration is not as easily accomplished because of the need for pump reprogramming by specifically trained specialists. . Figure 129-1 The synchromed II infusion pump (Medtronic Inc)
129
A variety of catheter materials have been employed in long term human implants including nylon, polyamid, polyurethane and silicone [44].
Surgical Technique Before implantation, it is important to examine the patient’s abdomen and decide on which side to implant the pump. The choice of anesthetic should be explained in detail to the patient. Practitioners may elect to perform intrathecal pump implantation with percutaneous procedure under general anesthesia, spinal anesthesia, or local infiltration [34,35,51]. Catheter introduction under local infiltration allows the patient to communicate to the surgeon any paresthesia reducing the risk of neurologic injury. Prophylactic antibiotics for staphylococcal coverage should be administered 30 min before local anesthesia and surgical incision [30]. Patients are placed in either the right or the left lateral decubitus position on the operating table, the skin at the lower back including the midline and paramedian area are prepared and draped in sterile fashion. The placement of an implantable intrathecal pump consists of the catheter placement followed by implantation of the pump [50,52]. Fluoroscopy is used to identify the L2-L3 interspace (to avoid any damage to the conus medullaris), and the skin is marked at the midline. Approximately 5 cm caudal and 1 cm lateral to this point is the desired entry point of the introducer needle for a shallow angled, and close to the dorsal processes, entering in midline away from the roots. This is important to reduce neurologic injury and kinking of the silastic catheter. The Tuohy needle is inserted until CSF is obtained. The catheter is then inserted through the needle into the intrathecal space and advanced slowly under fluoroscopic guidance so that the catheter tip reaches the desire location. The distal end of the catheter
2175
2176
129
Intrathecal opiates for cancer pain
is advanced progressively, to lie approximately 5 cm cephalad to the conus (between T10 and L1) depending on the level of the most intense pain. The distal end of the catheter should be clamped to prevent further CSF leakage. A small 3–4 cm vertical incision is made alongside the needle (protecting the catheter) through the dermis. Metzenbaum scissors are used to dissect carefully along the needle until the supraspinous ligament or dorsal lumbar fascia is identified. The needle is removed and the catheter is secured to the fascia with an anchor device to minimize the risk of migration or dislodgement. Therefore, before proceeding the end of the catheter should be checked for free flowing CSF and then clamped to prevent excessive leakage. Once the catheter is in place, the pump must be prepared for implantation according to the manufacturer’s recommendations [53]. The incision for the pump pocket is made in the right or left lower quadrant of the abdomen at or about the ombilical level with a plane of dissection between the fascia of the external oblique muscle and subcutaneous fat. The pump should be placed below the belt line, but not too close to the anterior rib or iliac crest as it may lead to prolonged discomfort. The pocket with blunt dissection, should be created to just accommodate insertion of the pump. If the pocket is too large, the resultant dead space will increase the likelihood of postoperative seroma development or allow the pump to flip over. The pump should not be placed deeper than 2.5 cm from the skin surface to allow reliable telemetry, palpation, and refill procedures. After creating the pump pocket, tunneling from the pump pocket to the back wound is be done with the tunneling device. It is always important to be able to palpate the tip of the rod as it is passes, thereby reducing the risk of peritoneal or pleural entry and viscera injury. Once the tunnel has been created, the catheter can now be pulled through the tunnel to the pump site. The catheter should be measured (to determine
dead space volume) and connected to the pump with a silk 2–0 suture. Prior to connecting the catheter, it is important to once again verify the free flow of CSF from the catheter. The pump is then placed into the pocket reservoir side up. Any excess catheter should be placed behind the pump to prevent damage to the catheter during refilling. The pump should be secured to the pocket by suturing to the abdominal fascia under the pump pocket. Before closing the wounds, it is important to establish adequate hemostasis to prevent the risk of postoperative infections. The wounds are closed in a two layer closure according to the preference of the surgeon.
Surgical Complications Implantation of a pump for intrathecal opioid delivery must be done with the highest level of care to minimize surgical complications, which may include bleeding, tissue damage, CSF leak and infection (> Table 129-1) [53,54].
Bleeding Excessive bleeding is not a major problem during implantation because the surgery does not involve heavily vascularized areas. Uncontrolled bleeding may be avoided if the surgeon carefully screens patients for coagulopathies and those who are taking medications that can increase the risk of bleeding. Patients who are anticoagulated are not candidates for this type of procedure until the coagulation returns to normal [38]. Bleeding can occur within the subdural or epidural space during catheter placement and lead to hematoma with spinal cord compression. Post-operatively, it will present first with increasing back pain that rapidly progresses to neurological deficits, including motor weakness, sensory changes, and sphincter dysfunction. Superficial
Intrathecal opiates for cancer pain
. Table 129-1 Side effects and complications of I-Th opioid administration Medication side effects Pruritus Nausea and vomiting Urinary retention Constipation Fluid retention Respiratory depression Sedation Surgical complications Bleeding Neurological deficits related to tissue damage Cerebrospinal fluid leaks Infections Technical (delivery system) complications Catheter Breaking Kinking Disconnections Tip obstruction Dislodgement Pump Overfilling Battery failure Pump torsion Wrong refill Programming errors
post-operative bleeding around the wound is another potential complication. An hematoma can appear at the pump site, with swelling, pressure and pain. Abdominal binders are sometimes helpful by applying direct pressure over the site to reduce the swelling and discomfort.
129
recognized over the past two decades. Since the first case report by North et al. in 1991, on intrathecal catheter tip mass causing spinal cord compression [59] more cases studies and systematic review have been published [60,61]. The incidence of intrathecal catheter tip inflammatory mass has been reported to be 0.4 after 2 years of therapy, increasing to 1.16% after 6 years of therapy [62]. The incidence of asymptomatic lesions may be much higher [63]. Speculative hypotheses regarding the cause of the lesions have included the following: the physical or chemical characteristics or the concentration of the infused drugs or diluents; infection with slow-growing or fastidious bacteria, pyrogen or endotoxin effects; hypersensitivity to silicon catheter material; and/or delayed effects of surgical trauma from the catheter implantation procedure [60,64,65]. The formation of a granulomatous mass may be indicated by complaints of new pain, numbness, weakness, or changes in bladder or bowel habits [66,67]. The diagnosis of a mass around the tip of the catheter and spinal cord compression should be confirmed by a gadolinium-enhanced MRI; failing the diagnose the problem properly could lead to permanent neurologic sequelae [68]. Once a diagnosis of a granulomatous mass around the catheter tip has been confirmed, rapid surgical decompression of the spinal cord, removal of the mass, and removal of the spinal catheter are required [38,60].
Cerebrospinal Fluid Leaks Neurologic Injury Implantation of the catheter into the spinal canal can lead to damage of the nerve roots or the spinal cord [55–57]. Neurologic injury can also develop later. One patient developed progressive necrotic myelopathy leading to paraplegia, a rare form of transverse myelitis [58]. Inflammatory mass lesion at the tip of intrathecal catheters or catheter tip granulomas have been increasingly
A CSF leak into the epidural space is inevitable because of the nature of the catheterization procedure. The catheter that is placed into the intrathecal sac is smaller than the needle, and when the needle is removed, it leaves a hole in the dural sac that is larger than the catheter itself. This leak usually stops within several days but may occasionally persist. CSF leaks also may occur as a result of the following: incomplete tissue sealing
2177
2178
129
Intrathecal opiates for cancer pain
around the catheter at the insertion site, multiple subarachnoid punctures during catheter placement, dislodgement, disconnection, and break or migration of the catheter. Leaks of CSF through the incision always require resuturing of the surgical wound, and it is recommended that a complete wound takedown and repair be performed. Postdural puncture headache can result from CSF leakage around the catheter or loss of CSF during the implant procedure. The overall reported incidence of postdural puncture headache is 1–30%. In the majority of cases the headache is moderate and disappears in a few day without treatment. However, in 15% of the patients the headache is severe and incapacitating, and can last for weeks [69]. In cases of persistent postdural puncture headache, a pharmacologic approach is the first-line treatment (caffeine, NaCl 0.9% IV. . .) [69]. If drug therapy is not effective and the patient’s headache persists longer than 5–7 days, autologous epidural patching is required [29]. CSF leakage can occur anywhere along the catheter or in the pocket site and may result in CSF hygromas. CSF hygromas are subcutaneous collections of CSF that are usually relatively small and self limiting, resolve in 1–2 weeks, and are not considered to be clinically significant. Aspirations should be avoided with small hygromas because of the risk of infection, whereas large hygromas, which rarely occur, may require aspiration.
Seroma, Necrosis, and Skin Perforations Wound healing impairment, skin perforations, and necrosis at the pump pocket may be caused by improper positioning or a reaction to the size of the pump. Formation of a seroma, a serosanguinous fluid collection is also common around the pump pocket, especially if the pocket created
was much larger than the device. Seromas are usually sterile, can last for 1–2 months postimplantation, and are usually not clinically significant. Puncture and drainage should be avoided as the seroma is usually self-limited [29] and will only serve to introduce infection. Abdominal binders are somewhat helpful in reducing the size and discomfort of the seroma and may accelerate the resorption.
Infections Guidelines regarding the prevention and treatment of IT drug delivery system infection have been published [70]. Infections may occur in the pump insertion site, along the catheter, in the epidural or intrathecal space. In general, the reported incidence of infection with I-Th drug delivery is low (0–9%). Preventive measures against infection should always be implemented while implanting foreign bodies. Preventing infections requires the use of strict sterile techniques, perioperative antibiotics, some practitioners advocate the use of intra-operative antibiotic irrigation as well. Since most operative infections are caused by staphylococcus aureus or S. epidermidis, cephalosporin or vancomycin are used by most surgeons. Wound infections need to be identified early and treated aggressively to prevent serious complications. Symptoms of pump insertion site (pump pocket) infection include pain, erythrema, tenderness, swelling, fever and increased temperature over the pump pocket. Superficial infections should be cultured and treated with antibiotics. More serious infections involving the catheter or pocket will require usually the removal of all the implanted hardware, followed by appropriate antibiotic therapy. Once the material has been removed, the wound should be left open and wet to dry, with sterile normal saline dressings done until the wound closes on
Intrathecal opiates for cancer pain
its down. Recently local treatment and leaving the pump in place had been advocated in some cases [71,72]. Infections involving the epidural space require immediate removal of all the implanted devices and intravenous antibiotics. Epidural infections can lead to epidural abscess which can compress the thecal sac and cause neurologic injury. Intrathecal infection (meningitis), while rare, requires immediate explantation of all foreign body materials and the start of intravenous antibiotic therapy. Symptoms of intrathecal infection may include fever, stiff neck, and positive meningeal stretch signs. These signs are also common under normal circumstances after implantation; hence, clinicians should use caution when diagnosing infectious meningitides. If the patient does not appear toxic, has a normal complete blood count, and the CSF drawn from the pump’s side port reveals leukocytosis and elevated protein with gram negative stains, then the patient can be observed until CSF cultures come back. This is usually a limited situation. Late infectious complications can also occur [65], one case of transverse myelitis associated with an acinetobacter baumanii was reported with intrathecal pump catheter related infection [73].
Pharmacology of Intrathecal Opioids Mechanism of Action Spinal delivery of opiates has the advantage of providing local regulation of nociceptive processing with a minimal impact of the agent on supraspinal centers relevant to higher-order function. It is assumed that the analgesic effect of opioids is mediated via conformation changes of selective receptors in the dorsal horn of the
129
spinal cord and in relevant areas of the brain. After direct I-Th application, very high concentration of morphine can reach the receptor sites. Koulousakis et al. showed that morphine concentration in the CSF was about three thousand times higher than the concentration of its metabolites M6G and M3G and conclude that the main analgesic effect of I-Th application can be attributed to morphine ‘‘itself ’’ [74]. Opioid agents have spinal analgesic actions, which are reversed by naloxone and exhibit a structure activity relationship. Opioid receptors are concentrated in the dorsal horn in the lamina I,II, V and X of the spinal cord. At the receptor level, membrane hyperpolarization can be induced by m -, d- and occasionally κ-receptors through activation of an inwardly rectifying potassium channel. A membrane G0-protein mediates the m receptor effect, and Gi/0-proteins mediate the d receptor effects. In addition to hyperpolarization induced by m and d agonist receptor occupancy, a simultaneous inhibition of the opening of voltage-sensitive calcium channels subsequently reduces the terminal release of neurotransmitters from the cell. These receptors are widely distributed on the spinal terminals of primary afferent neurons (presynaptic) and on cell that originate in the dorsal horn (postsynaptic) [75,76]. How spinally administered opioids affect behavior (antinociception) depends on the presynaptic inhibition of neurotransmitter release from small primary afferents (substance P, calcitonin gene-related peptide) and on the hyperpolarization of post-synatic neurons resulting from G-protein-mediated activation of potassium channels [77,78]. Opioids have a differential effect, reducing dorsal horn neuronal activity evoked by C-fiber stimulation more than activity evoked by AD-fiber stimulation [79]. The selectivity of the spinal opioid receptors results in analgesia without the side effects associated with spinal anesthesia (motor, sensory or sympathetic blockade).
2179
2180
129
Intrathecal opiates for cancer pain
Intraspinal Opioids Intraspinal delivery of opioids and other agents can be achieved by a variety of approaches, including epidural bolus, intermittent I-Th injection, and continuous epidural and I-Th infusions. There is a lack of consensus over what constitutes an appropriate method to affect an equianalgesic conversion from systemic to I-Th morphine. Conversion tables have been proposed. If the patient is taking other opioids, a conversion to intramuscular morphine is necessary. Then 1/10 of the intramuscular dose is chosen for epidural route [80]. The intrathecal dose requirement is about 10 times lower than epidural morphine requirement (> Table 129-2) [29]. Tolerance to systemic opioids is an important factor contributing to the wide variation in the starting dose as well as maximum dose. Cross-tolerance between spinal and systemic opiates has been described. However, there is not a clear trend to increasing I-Th dose escalation with time, as dose escalation is only poorly correlated with duration of use [81]. It has been shown that over extended periods a significant number of patients show a plateau in their dose requirements and, after some time, a precipitous increase in drug requirement. As in the case of oral administration, progression of the disease is the major factor influencing spinal opioid escalation [82,83]. The upper limit to which this can go is constricted practically by side effects. An abrupt return of pain in patients who had achieved initially excellent results from I-Th opioids suggest alterations in drug delivery (withdrawal, disconnection, kinking of the catheter. . .) or
change in the medical status (progression of the disease, generation of neuropathic pain. . .). I-Th drug delivery has largely replaced the epidural route for various reasons [31,84,85]. Epidural infusions require an approximately 10-fold greater volume and dose of opioid to traverse the dura and enter the subarachnoid space. This large difference in infusion volumes and doses has a major impact on cost, and the frequency of bag changes, which necessitates breaking the sterile system more frequently. Comparative studies have shown that in long term treatment, I-Th morphine administration may give more satisfactory pain relief with lower doses of morphine and fewer side effects than epidural administration [86]. Furthermore, treatment failure and loss of analgesic efficacy is more likely with the epidural route, probably because of catheter obstruction or epidural fibrosis that impedes diffusion of drug to the subarachnoid space. Crul and Delhaas reported a greater incidence of complications with the I-Th catheter than with the epidural catheter during the first 20 days after implantation (CSF leakage) [84]. Subsequently, technical complications rates have been shown to be higher with long term epidural (55%) compared to I-Th (5%) infusions. No significant differences in pain intensity, pain relief, satisfaction scores and neuropsychological function between bolus and infusion treatment have been evidenced. However, there was a significantly greater degree of dose escalation in patients receiving continuous infusion compared with the patients receiving repeated bolus doses [87]. Quality of analgesia appears
. Table 129-2 Equianalgesic opioid conversion (mg) [44] Opioid
Oral
Parenteral
Epidural
Intrathecal
Hydrophilicity
Morphine Hydromorphone Fentanyl Sufentanyl
300 60 – –
100 20 1 0.1
10 2 0.1 0.01
1 0.2 0.01 0.001
High Intermediate Low Low
Intrathecal opiates for cancer pain
to be better when using continuous infusion compared with intermittent bolus of morphine. Many factors are considered in the decision for an external pump versus an implantable. Burton et al. published their decision making algorithm [37]. Factors which lead them to consider an implantable I-Th pump include: a longer life expectancy (>3 months), access to pump refill/reprogramming capabilities, diffuse pain and favorable response to an I-Th trial. The cost of an implanted I-Th pump plays a role in decision making for implantation in the cancer patients. The economic analysis of an implantable versus externalized pumps revealed the 3 months life expectancy to be the approximate ‘‘break-even’’ point for the implanted pump to become more cost-effective than the externalized pump system [88–92]. Patients who do not have at least 3 months to live, and who do not tolerate systemic delivery of opioids, who have successfully completed a trial of intraspinal delivery of opioids, are candidates for externalized catheter and pump delivery system [34,93]. More than a dozen different agents have been used intrathecally in the management of cancer pain but many of these are used off-label [31,94]. Opioids as a class have been most extensively studied [95–97]. Morphine sulphate is the standard first-line drug used in I-Th therapy and the standard of comparison for other intraspinally active analgesics [22,98]. Typically, additional agents are then added in a stepwise fashion depending on the response to morphine alone, and it is not uncommon that one to three adjunctive agents are added to the infused solution. Clinical guidelines have emerged in 2000 and had been revised in 2003 to form a basis of decision-making for treatment using intraspinal therapy with a specific set of algorithms published for cancer pain [39,99,100]. The exact dosage comparison of different opioid analgesic agents for intraspinal use is difficult. As a general rule, the higher the lipid solubility the lesser the analgesic potency when
129
the drug is administered intraspinally [101]. However, there is wide variation among patients in many aspects including age, weight and the degree of opioid tolerance which affects dose determination. Most importantly, the physical and chemical properties of the drug itself will determine its overall efficacy [96]. Morphine sulphate is the only FDAapproved opioid for I-Th use, and it generally remains the first-line agent for I-Th analgesia. Numerous others I-Th opioids have been used successfully including hydromorphone, fentanyl, sufentanyl, meperidine and methadone [50,100,102]. Morphine with high hydrophilicity, penetrates the spinal cord slowly, allowing a considerable amount of the drug to ascend cephalad in the CSF. The mechanism of cephalad migration of morphine is thought to be due to the bulk flow of CSF which ascend from the lumbar region, reaching the cisterna magna by 1–2 h and the fourth and lateral ventricles by 3–6 h [103]. Intrathecally administered morphine produces slower onset, longer clearance from the spinal, longer duration of anti-nociception, but a higher incidence of certain side effects. Fentanyl and sufentanil are more lipophilic than morphine or hydromorphone, which results in relatively more rapid adsorption into neural structures proximate to drug delivery and receptor saturation and less migration to the brainstem. This has some theoretical advantages when a more dermatomal analgesia is desired, without the adverse effects associated with opioid activity at the brainstem level (respiratory depression).
Others Intraspinal Agents Many other agents have also been used intraspinally either alone or in combination with opioids in the treatment of intractable cancer pain (> Table 129-3) [10,104]. Table lists different non opioids agents which have been used intraspinally in pain management; only a few
2181
2182
129
Intrathecal opiates for cancer pain
. Table 129-3 Opioid and other agents used for intraspinal analgesia Drug category
Agents
Opioids
Morphine Hydromorphone Fentanyl Sufentanyl Methadone Meperidine Bupivacaine
Sodium channel antagonists (local anesthetics) Alpha 2-adrenergic agonists NMDA antagonists Calcium channel antagonists Somatostatins GABA agonists
Adenosine agonists Acetylcholinesterase inhibitors Corticosteroids
Ropivacaine Clonidine Ketamine Ziconotide Somatostatin Baclofen (GABAB) Midazolam (GABAA) Adenosine Neostigmine Prostigmine Betamethasone
of these are currently used in cancer pain management. Clonidine is an a-2 adrenergic agonist that acts at the dorsal horn modulating the transmission of noxious sensory information by mimicking the activation of descending of descending noradrenergic pathways and inhibiting neurotransmitter release [14,105]. Epidural and I-Th clonidine have been studied extensively in the post-operative pain and labor analgesia setting, but few data exists on cancer pain management [106,107]. It is generally considered a very useful adjunct to opioids and local anesthetics when there is a predominant neuropathic pain component. The consensus of best clinic practice recommend using clonidine as a third line-agent when treating refractory pain, but should be used as a second-line agent when there is a neuropathic pain component. Local anesthetics, such as bupivacaine and lidocaine, are sodium-channel blockers that
interrupt potentials along sensory and motor nerves. When administered intrathecally local anesthetics can block incoming sensory nerves and produce complete analgesia [108]. Studies, primarily in cancer patients with chronic infusion of bupivacaine and morphine combination, show no significant neurotoxicity [109,110]. Spinally administered local anesthetics have a synergic effect when combined with opioids. Bupivacaine is commonly used in post-operative epidural infusions, but is also widely used in the intrathecal route in combination with opioids for cancer pain. Like clonidine, bupivacaine is more effective when used in combination with opioids than opioids alone in the treatment of neuropathic pain; however bupivacaine is also used to treat nociceptive pain [15,111,112]. As with other local anesthetics, bupivacaine dosing is limited by the occurrence of motor blockade at higher concentrations. In the terminal patient who is bed-bound and does not object to lower extremity weakness, higher doses of local anesthetics may be used to provide profound analgesia with no cognitive side effects. At lower doses, with an I-Th patient-controlled demand dose, an interesting case report shows promising efficacy in treating patients with refractory pain [113]. Ziconotide is a novel snail peptide that is an N-type voltage-sensitive calcium channel blocker [114–116]. Calcium channel blockade prevents calcium ion influx and neurotransmitter release. This selectively inhibits transmission of primary afferents located in the dorsal horn of the spinal cord. Ziconotide is FDA-approved, and is the first agent developed specifically for I-Th administration in chronic pain state. Multicenter, randomized, double blind, placebocontrolled studies have evaluated the safety and efficacy of I-Th ziconotide in intractable chronic pain associated with cancer or AIDS with improvement in 75% of patients. Clinical experience has shown that most of the adverse effects are dose related and can be avoided by slow and careful titration.
Intrathecal opiates for cancer pain
Neostigmine is another interesting non opioid that may have a role as an intraspinal analgesic [117,118]. Cholinergic mechanisms are involved in the transmission of pain pathways within the spinal cord [119]. Neostigmine, a water soluble anticholinergic, does not cross the barrier; hence, when injected intrathecally has a long acting central effect. The drug has efficacy and minimal toxicity, but initial trials have seen some troublesome systemic side effects [120]. A number of other alpha-2 adrenergic agonists, local anesthetics, calcium-channel blockers, and drug from others classes (Somatostatin, analogues, GABA agonists, NMDA antagonists, betamethasone) have been delivered via the intraspinal route to treat intractable cancer pain, with varying degree of success [121,122].
Efficacy of Intraspinal Opioids in Cancer Pain Assessment of the efficacy of chronic spinal opiate infusion for cancer pain has been of major importance since the earliest description of the approach. Over the 30 last years, many studies have attempted to determine efficacy of long term I-Th morphine therapy [121]. However,
129
only few controlled study evaluating efficacy of long-term I-Th morphine were published and review of these articles shows large discrepancies in the results [24,44]. An important question involves the definition of the efficacy of the therapy [43]. Several methodologies have been employed (pain relief, decreased side effects or adjuvant systemic opioids, improved function. . .) and it remains difficult to establish population outcome statistics. I-Th delivery of opioids has been shown to reduce significantly pain levels in terminal adult or pediatric malignancies in a large percentage of the clinical studies [36,37,84,121,123–126]. A review of literature from 1999 showed found that the use of I-Th delivery morphine by an implanted infusion pump provided ‘‘good to excellent’’ pain relief in patient with intractable pain [51,127]. > Table 129-4 summarizes the principal parameters of various clinical series with significant numbers of patients. Morphine is the drug most commonly used by the I-Th route, although some investigators include other opioids or bupivacaine when there is an associated neuropathic component. Furthermore, the results of five recent studies of I-Th therapy demonstrate the effectiveness of this therapy in the management of severe pain
. Table 129‐4 Selected published reports on the use of intrathecal morphine for the treatment of cancer pain Dose (mg/24h)
Follow-up (mo)
Reference
No. of Patients
Range
Average
Average
Pain relief Excellent + good (%)
Laugner et al. [123] Lazorthes et al. [42] Meynadier et al. [128] Penn and Paice [125] Onofrio and Yaksh [36] Follett et al. [129] Chambers et al. [130] Paice et al. [124] Gestin et al. [131] Sallerin-Caute et al. [83]
41 48 25 35 53 35 12 133 50 159
1.5–7.5 1–10 – – 1–5 0.65–15 4–40 – 0.4–94 1–80
– – 1 – 4 5.4 – 14.2 9.2 7.8
2 4 5 5.4 0.5 7.7 6.2 12 4.7 3
+++ 75% 96% 80% 65% 77% 100% 95% 50% 80%
Bup: Bupivacaine; Cinf: Continuous infusion
Mode, opioid Bolus, morphine Bolus, morphine Bolus, morphine Cinf, morphine Cinf, morphine Cinf, morphine Cinf, morphine Morphine + Bup External pump Bolus, morphine
2183
2184
129
Intrathecal opiates for cancer pain
from cancer [37,132–135]. Four of the studies examined the effectiveness of IDDS on the management of continuous severe cancer pain, the fifth on episodic or breakthrough pain. Burton et al. analyzed retrospectively the effectiveness of intraspinal analgesia in 87 patients with a 8-week follow-up [37]. They showed that after administration of intraspinal analgesia via the epidural or I-Th route, there was a significant reduction in the proportion of patients with severe pain. Oral opioid intake, self-reported drowsiness and mental clouding also significantly decreased. Recently Mercadante et al. evaluated the clinical response to a combination of I-Th morphine and levobupivacaine in advanced cancer patients who were highly opioid tolerant, being previously treated with multiple opioid trials unsuccessfully [132]. The intrathecal treatment provided a long-term improvement of analgesia, with a decrease in adverse effects and opioid consumption until death. A prospective randomized multicenter clinical trial investigated the safety and efficacy of an I-Th drug delivery system (IDDS) plus comprehensive medical management (CMM) versus CMM alone in patients with refractory cancer pain [85,135,136]. The authors conclude that the patients receiving IDDS plus CMM had reduced pain, fewer common drug toxicities, and improved survival, compared with patients receiving CMM alone. Although the IDDS and CMM group achieved better outcomes than the control group, results of the trial also demonstrated that algorithm usage by pain specialists in the control group receiving CMM alone also reduced cancer pain by 39% and pain medication toxicity by 17%. According to previous results, the panel thus recommended the use of algorithm for the management of cancer pain [137]. In 2005, the same group of authors evaluated whether IDDS could help the most refractory patients failed by expert CMM [134]. CMM patients, who crossed over to IDDS for the most refractory pain had significant reductions in pain and drug toxicity. Furthermore, the
survival time of 3 months may be long enough for the IDDS implant to be cost effective. In a fourth study, Rauck et al. evaluated Intrathecal drug delivery system in the management of episodic or breakthrough pain in a prospective, multicenter open-label study [133]. In this study of 119 cancer patients with refractory cancer pain and/or uncontrollable side effects, better analgesia was achieved when the patients managed their pain with an implantable, patientcontrolled I-Th drug delivery system similar to patient-controlled analgesia. Results of the study showed a long-lasting reduction in numerical analog pain, systemic opioid use and side effects. Administration of I-Th opioids and adjuvant medications also allows reductions of up to 200% in the amount of administered oral or parenteral medication [37,39]. In addition to reduces dosages, I-Th opiate and adjuvant medication enhance pain control with minimal side effects.
Side Effects of Intrathecal Morphine Pharmacologic Side Effects Pharmacologic side effects of systemic and I-Th opioids have been well characterized (> Table 129-1). Although several of the side effects associated with opioids are dose-dependant and temporary, some are long-lasting [22,138]. Side effects are less common in patients chronically exposed to either I-Th, epidural or systemic opioids, some are mediated via interaction with specific opioid receptors while others are not [22].
Pruritus Pruritus is one of the most common side effects associated with I-Th morphine administration [138]. It occurs unpredictably and is mostly
Intrathecal opiates for cancer pain
localized in the face, neck or upper thorax although it may also be generalized. The onset of pruritus, usually occurrings within a few hours of injection, is often difficult to control, and may precede the onset of pain relief [139]. Pruritus is generally not severe, and decreases with repeated doses. The mechanism of action is not thought to be due to histamine release and may be due to cephalad migration of opioids in CSF and subsequent interaction with the trigeminal nucleus located superficially in the medulla [140]. Pruritus can be readily treated with the mu antagonist, naloxone [141].
Nausea and Vomiting Nausea usually occurs within 4 h of I-Th opioid delivery and may be immediately followed by vomiting. Chaney et al. noted the incidence of nausea and vomiting following acute I-Th opioid administration to be 30% [22]. Because nausea and vomiting are dose-related, these side effects can be prevented by initiating I-Th morphine delivery with a low dose; the dose is then slowly increased during the first few days of treatment [139,142]. The mechanism of nausea and vomiting is likely due to cephalad migration of the opioid in CSF and subsequent interaction with opioid receptors located in the chemoreceptor trigger zone (area postrema) [143]. Sensitization of the vestibular system to motion and decrease gastric emptying may also contribute to the development of nausea and vomiting [144,145]. Symptoms can be relieved in most patients with antiemetic agents such as metoclopramide or haloperidol [146].
129
prostates [147]. Urinary retention may occur with the test doses and may require catheterization. It is believed to be dose dependant [148]. Compensatory changes appear to occur over time so that the retention does not appear to be a chronic problem in patients with otherwise functioning bladders. This effect reflect a direct suppression of the micturation reflex at the spinal level, leading to an increase in bladder capacity, relaxation of the detrusor muscle, and inhibition of external sphincter relaxation [149,150]. The detrusor relaxation caused by epidural morphine is reversed with naloxone although reversal of analgesia is also likely, adjuvant treatment with colinomimetic drugs are also effective in reducing urinary retention [151].
Constipation Anderson et al. found that 30% of the patients in their study experienced constipation at least once over a 2-year follow-up period [152]. The decreased gastrointestinal motility caused by I-Th morphine is because of the interaction with opioid receptors in the spinal cord, rather than systemic absorption [145,153]. Patient on I-Th morphine therapy may present signs and symptoms of ileus, which in turn, may cause nausea and vomiting [153,154]. Some authors have recommended a prophylactic approach with cathartics [155]. The recommended guidelines include starting with a stool softner and a bowel stimulant, increasing dosages are needed before adding laxatives [155]. Dietary changes and increased fluid intake are also recommended.
Urinary Retention
Fluid Retention
The incidence of urinary retention following I-Th opioids ranges between 42 and 80% and occurs most often in older males with enlarged
Oliguria, water retention and peripheral edema caused by I-Th opioids can be quite problematic [156]. The incidence of long-term I-Th
2185
2186
129
Intrathecal opiates for cancer pain
opioid induced edema ranges from 6.1 to 21% [147,157]. Pharmacokinetic studies suggest the cephalad migration of morphine in the CSF, and subsequent interaction with opioid receptors in the posterior pituitary gland and vasopressin release [22]. Treatment of I-Th opioid-induced edema starts with simple measures such as leg raising, elastic stockings and salt and fluid restriction. Diuretics may also be used with benefit. Aldrete et al. recommended that preexisting leg venous insufficiency and edema be considered relative contraindication for I-Th opioid therapy [157].
Respiratory Depression Respiratory depression is the most feared complication of I-Th morphine therapy. Clinically important respiratory depression has been reported following I-Th morphine infusion [158]. Respiratory depression may occur within minutes of injection or may be delayed for hours. In patients with prior narcotic exposure, the incidence of respiratory depression with gradually increasing opioid dosing is rare. The incidence of clinically relevant early respiratory depression (within 2 h of injection) following spinal opioid delivery is 1%. Delayed respiratory depression often occurs 6–12 h following I-Th delivery of morphine and may last for 24 h. Delayed respiratory depression results from the cephalad migration of the opioid in the CSF and subsequent interactions with the opioid receptors located in the ventral medulla [76,159]. The complication of respiratory depression may be reduced by starting I-Th opioid therapy at a low dose, then slowly titrating to the lowest analgesic dose, and by closely monitoring the patient during the first 24–48 h of I-Th opioid delivery. Reversal of respiratory depression can be readily accomplished with the administration of mu antagonist naloxone or a kappa agonist/mu antagonist nalbuphine [160].
Sedation and Cognitive Symptoms Sedation and mental cloudiness (cognitive impairment) occur frequently with all opioids, although the severity of these symptoms is related to route of administration. The incidence of sedation is lower with I-Th administration compared with systemic (oral or subcutaneous) administration of opioids. Mental status change, presented as sedation and lethargy, occurred in 10–14% of patients receiving long-term I-Th morphine infusion therapy [152]. The degree of sedation appears to be dose-related [139]. Central nervous system depression may be profound and coma has been reported [161]. Any time sedation occurs following I-Th administration of opioids, respiratory depression must be suspected [162]. Additional mental status changes other than sedation may also occur including paranoia, catatonia, euphoria, anxiety, delirium and hallucinations [163]. Cephalad drug migration in the CSF and subsequent interactions with opioid receptors in the brain such as the thalamus, limbic system and cerebral cortex are suggested. When these side effects are evident, opioid dose reduction should be tried first. Delirium and hallucination are treated with neuroleptics or benzodiazepines.
Hyperalgesia Hyperalgesia induced by I-Th morphine administration has been studied in animals [164–168] and humans [169,170]. High dose I-Th morphine administration may produce hyperalgesia and allodynia in rare occasions [171]. Early reports suggested causality by I-Th high-dose morphine to block glycine or gamma-aminobutyric acid (GABA)-mediated inhibition [172]. Conjugated metabolites of morphine, several hundred times more potent at producing behavioral excitation may also be involved [173]. More and more evidence suggests that the development of
Intrathecal opiates for cancer pain
tolerance and hyperalgesia share common mechanisms. Management approaches include opioid dose reduction, opioid rotation or low dose infusion of naloxone [174–177].
Neurotoxicity Delivery of agents into the spinal space often results in local concentrations which far exceed levels noted after systemic delivery. This may be result in changes in myelin sheath integrity, cell viability, and inflammatory reactions (activation of astrocytes and resident microglial macrophages) [178]. Additional proposed toxicity mechanisms include growth factors that induce cellular proliferation, vasoconstriction and apoptosis [179]. It remains unclear however what severity of such changes should be important. In animals epidural or I-Th morphine causes spinal cord damage or inflammatory changes in the meninges [180181182]. In humans, I-Th morphine has been implicated as possible cause of neurological dysfunction [183]. On the other hand, administration of larges doses of I-Th morphine for prolonged periods of time has proved to be safe [81]. Autopsies after I-Th drug delivery for intervals from weeks to several years have revealed no evidence of arachnoidis around the catheter or degenerative disease related to the constant morphine infusion [109,110,184].
Technical (Delivery System) Complications The majority of delivery system complications involve the catheter not the pump. Catheter complications can include dislodgement, kinks, breakage, occlusion or migration [185,186]. Resulting problems from catheter complications may include drug under infusion leading to a loss of analgesia, inconsistent analgesia, or drug
129
withdrawal symptoms. Mechanical catheter failures should be confirmed using imaging techniques. A radiograph of the spine and pump will indicate whether the catheter has dislodged, is broken or disconnected. Injecting non-ionic radiographic dye through the side port of the pump under fluoroscopy will allow verification of minor catheters breaks, kinks, and patency. Complications associated with programmable pump may include overfilling of the pump, battery failure, pump failure and pump torsion or flipping [38,187]. Algorithms for device assessment have been published, and if the practitioner is in doubt, the pump manufacturer can be consulted [188].
Respective Indications of Intrathecal and Intracerebroventricular Morphine Administration Since the initial finding of leavens et al in 1982 that low doses of morphine (1 mg) given by the intracerebroventricular (ICV) route to four patients presenting intractable cancer pain produced profound analgesic activity (from 80 to 100%) without respiratory depression or neurological modification [189], many prospective clinical studies have confirmed this observation [43,190–195]. In all the studies, the indications related to only chronic pain of malignant origin secondary to a physiologic mechanism of nociception, after failure of systemic opioids. Some of these patients had been incompletely controlled by intrathecal spinal administration because of a diffuse or cephalic topography of their pain. Comparative analysis of these clinical series shows significant (80% of good or excellent results) and durable (mean followup, 3 months), analgesic effectiveness for small intracerebroventricular morphine doses (on an average 1–2 mg:24 h). In our experiment, the effective daily dose increased by 0.3–2.4 mg/24 h
2187
2188
129
Intrathecal opiates for cancer pain
[41,193,196]. Central side effects were drowsiness, somnolence, confusion, or hallucination; these effects were transitory and reversed spontaneously or by temporarily reducing the dose. Some cases of respiratory depression were reported; these occurred either during the test period or following an error of dosage. The risk of respiratory depression justifies careful monitoring particularly during the initial period. In practice, the two sites of CSF administration are complementary and based on the topography of pain [190]. I-Th spinal morphine is indicated in pelvic, abdominal, and lower thoracic pain. It also can be effective in higher thoracic pain and when the painful area involves cervical dermatomes. In such cases, the spinal catheter needs to be implanted through the lumbar route to the cervicothoracic subarachnoid space so that morphine diffuses and activates opioid receptors in the cervical spinal cord [197]. ICV administration is indicated for cranial or cervical cancer pain. In the event of failure or partial effectiveness of spinal I-Th administration for pain in the cervicothoracic area, it is a secondary approach because it is more invasive. Implantation of an intraventricular catheter makes this technique invasive and has considerably limited its indication in the last few years. It is, however, a simple, effective method accompanied by a high percentage of success and has limited iatrogenic complications. It is cost-effective as well. Physicians who specialize in pain management and oncologists should be aware of this alternative. Its use should be considered in intractable pain of cancerous origin, after failure of longacting orally administered narcotics and when the topography of the pain is diffuse or cephalic.
Future Directions: I-Th Cell and Gene Therapy Although, in selected patients, direct spinal delivery of opioids increase pain relief and reduce
opioid-related central side effects, this alternative route of morphine administration is limited by the cost, specialized maintenance, and mechanical malfunctions of implantable delivery systems and by the risk of bacterial contamination [41,42,198]. In an answer to these limitations, I-Th cell and gene therapy are promising approaches for providing sustained pain control in chronic cancer pain states. I-Th cell therapy by transplantation of living cells, can act as pharmacological drug pumps, providing a continually renewable local source of pain reducing agents eliminating the need for narcotic administration [199]. In addition, cell transplantation can utilize naturally derived neuroactive substances which have biological halflives that are too short to be delivered by another means. Chromaffin cells localized in the medullary portion of the adrenal glands were selected as they produce and release in culture high levels of opioid peptides, catecholamines as well as other neuropeptides (somatostatin, neurotrophic factors) [200,201]. Over the past several year, numerous studies have demonstrated that transplantation of adrenal medullary allografts into the spinal subarachnoid space of rats results in increased CSF level of opioids peptides and catecholamines [202] and produces analgesic effects in both acute and chronic pain models without the development of significant tolerance [199,203–205]. Based on studies in rodent, clinical trials using intrathecal human adrenal medullary allografts gave promising results in patients with cancer-related pain intractable to drug therapy [206,207]. Lazorthes et al. reported the results of a prospective phase 2 open clinical study in 15 patients with intractable cancer pain, demonstrating the feasibility of this innovative approach and its long term safety [208]. Furthermore our laboratory reported morphological and functional evidence of lumbar chromaffin cell allograft survival in two patients autopsied beyond 1 year of follow-up [209]. The practical application of human medullary tissue allografts
Intrathecal opiates for cancer pain
in clinical settings was, however hampered by the complex one site preparation before implant and, more importantly, by the limited availability of human donor tissue [198]. Xenogeneic donors may provide a cell source alternative permitting wide scale application of this therapeutic approach but infectious risks and particularly the viral risk must be evaluated [210–212]. It is likely that the field will move into fetal or geneticallyengineered cell lines, which can be tailored to produce necessary neuroactive substances and allow for the safe transplantation of a characterized and homogeneous cell population [199,213]. Eaton et al. reported recently the initial immortalization and characterization of different chromaffin cell lines derived from fetal human adrenal glands using tag, v-myc, or hTERT for immortalization [214,215]. The ability to successfully create and initiate stable chromaffin cell lines from human source is an immense first step in the development of a clinical strategy to provide a transplant source for therapeutic chromaffin tissue. The potential analgesic effect of human hNT2 GABA- secreting cell in a pre-clinical model of chronic pain has also recently been discussed [216,217]. Transplants of other cell lines genetically engineered to synthetize and secrete potentially antinociceptive molecules (endorphin, metenkephalin, galanin, neurotrophin BDNF, 5HT. . .) have been examined in preclinical studies in the last 15–20 years with various results [199,218]. A veritable explosion of research in gene therapy for pain has occurred in the last 20 years [219]. Gene therapy with direct somatic gene transfer for pain will eventually overcome the problems associated with transplantation of non autologous cells such as rejection of the cell grafts by the patient. These virus-mediated methods, although at the early stages of evolution and use, offer large-scale production of biologic agents that can be conveniently and
129
confidently used for the long-term relief of chronic pain [219–223].
Conclusion Cancer pain management is a complicated challenge. It requires a thorough understanding of the cancer disease process, the pain diagnosis, and the treatment modalities available to treat the painful condition. Full utilization of all available treatment modalities should help to optimize the patient’s pain control and quality of life. I-Th administration of morphine is an effective and conservative method to control refractory chronic pain of malignant origin after failure of opioids by the systemic route. Before institution of this approach, the physician must have a clear understanding of indications, contraindications, relevant anatomy, and the ability to assess and treat potential complications. Patients must be carefully selected and should undergo a prognostic trial. Nevertheless, the incidence of major complications is very low with intrathecal pumps, and they remain very useful tools in treating patients with refractory cancer pain. However, I-Th therapy for the management of refractory cancer pain is yet to become a standard practice. Many patients are referred too late in the course of their disease. It should be considered earlier in the progressive course of the disease and not reserved as a last resort for patients with a short life expectancy, particularly since such patients are then unable to benefit from the implantation of a drug delivery system. In this area of aggressive cancer management, optimal symptom control is an integral part of the treatment plan, and undoubtedly reduces the likelihood of therapy delays and attrition. Improved awareness of the indications, potential benefits, and limitations of I-Th therapy hopefully will ensure that patients are not denied an opportunity to ease their discomfort at the conclusion of life.
2189
2190
129
Intrathecal opiates for cancer pain
References 18. 1. World Health Organization: Cancer pain relief and palliative care (ed 2). Geneva, World Health Organization 1996;12–15. 2. Silver J, Mayer RS. Barriers to pain management in the rehabilitation of the surgical oncology patient. J Surg Oncol 2007;95:427-35. 3. van den Beuken-van Everdingen MH, de Rijke JM, Kessels AG, et al. Prevalence of pain in patients with cancer: a systematic review of the past 40 years. Ann Oncol 2007;18:1437-49. 4. Lesage P, Portenoy RK. Trends in cancer pain management. Cancer Control 1999;6:136-45. 5. Azevedo Sao Leao Ferreira K, Kimura M, Jacobsen Teixeira M. The WHO analgesic ladder for cancer pain control, twenty years of use. How much pain relief does one get from using it? Support Care Cancer 2006;14:1086-93. 6. Zech DF, Grond S, Lynch J, et al. Validation of World Health Organization guidelines for cancer pain relief: a 10-year prospective study. Pain 1995;63:65-76. 7. Krames E. Interventional pain management: appropriate when less invasive therapies fail to provide adequate analgesia. Med Clin North Am 1999;83:787-808. 8. Miguel R. Interventional treatment of cancer pain: the fourth step in the World Health Organization analgesic ladder? Cancer Control 2000;7:149-56. 9. Waldman S. The role of spinal opioids in management of cancer pain. J Pain Symptom Manage 1990;5:163. 10. Sloan PA. Neuraxial pain relief for intractable cancer pain. Curr Pain Headache Rep 2007;11:283-9. 11. Yaksh TL, Rudy TA. Analgesia mediated by a direct spinal action of narcotics. Science 1976;192:1357-8. 12. Ventafridda GV, Fighuzzi M, Tamburini M, et al. Clinical observation on analgesia elicited by intrathecal morphine in cancer patients. In: Bonica JJ, Ventafridda V, editors. Advances in pain research and therapy, vol. 2. New-York: Raven Press; 1979. p. 559-65. 13. Wang JK, Nauss LA, Thomas JE. Pain relief by intrathecally applied morphine in human. Anesth 1979; 50:149-51. 14. Cohen SP, Dragovich A. Intrathecal analgesia. Anesthesiol Clin 2007;25:863-82. 15. Deer TR, Caraway DL, Kim CK, . et al. Clinical experience with intrathecal bupivacaine in combination with opioid for the treatment of chronic pain related to failed back surgery syndrome and metastatic cancer pain of the spine. Spine J 2002;2:274-8. 16. Hassenbusch SJ. Epidural and subarachnoid administration of opioids for nonmalignant pain: technical issues, current approaches and novel treatments. J Pain Symptom Manage 1996;11:357-62. 17. Turner JA, Sears JM, Loeser JD. Programmable intrathecal opioid delivery systems for chronic noncancer pain:
19.
20.
21.
22. 23. 24.
25.
26.
27.
28.
29.
30.
31. 32. 33. 34. 35.
a systematic review of effectiveness and complications. Clin J Pain 2007;23:180-95. Yaksh TL, al-Rodhan NR, Jenson TS. Sites of action of opiates in production of analgesia. Prog Brain Res 1988;77:371-4. Devulder J, Ghys L, Dhondt W, et al. Spinal analgesia in terminal care: risk versus benefit. J Pain Symptom Manage 1994;9:75-81. Bennett G, Burchiel K, Buchser E, et al. Clinical guidelines for intraspinal infusion: report of an expert panel. Polyanalgesic consensus conference 2000. J Pain Symptom Manage 2000;20:S37-43. van Dongen RT, Crul BJ, van Egmond J. Intrathecal coadministration of bupivacaine diminishes morphine dose progression during long-term intrathecal infusion in cancer patients. Clin J Pain 1999;15:166-72. Chaney MA. Side effects of intrathecal and epidural opioids. Can J Anaesth 1995;42:891-903. Mercadante S. Opioid rotation for cancer pain: rationale and clinical aspects. Cancer 1999;86:1856-66. Mercadante S. Controversies over spinal treatment in advanced cancer patients. Support Care Cancer 1998;6: 495-502. Mercadante S, Agnello A, Armata MG, et al. The inappropriate use of the epidural route in cancer pain. J Pain Symptom Manage 1997;13:233-7. Lazorthes Y, Gouarderes C, Verdie JC, et al. [Analgesia by intrathecally applied morphine. Pharmacokinetics study and application to intractable pain (author’s transl)]. Neurochirurgie 1980;26:159-64. Hogan Q, Haddox JD, Abrams S, et al. Epidural opiates and local anesthetics for the management of cancer pain. Pain 1991;46:271-9. Coyne PJ, Smith T, Laird J, et al. Effectively starting and titrating intrathecal analgesic therapy in patients with refractory cancer pain. Clin J Oncol Nurs 2005;9:581-3. Krames ES. Intrathecal infusional therapies for intractable pain: patient management guidelines. J Pain Symptom Manage 1993;8:36-46. Stearns L, Boortz-Marx R, Du Pen S, et al. Intrathecal drug delivery for the management of cancer pain: a multidisciplinary consensus of best clinical practices. J Support Oncol 2005;3:399-408. Brogan SE. Intrathecal therapy for the management of cancer pain. Curr Pain Headache Rep 2006;10:254-9. Blond S, Meynadier J. [Drug therapy of pain in oncology: keep it simple!]. Bull Cancer 1994;81:653-8. Wallace MS. Treatment options for refractory pain: the role of intrathecal therapy. Neurology 2002;59:S18-24. Phan PC, Are M, Burton AW. Neuraxial infusions. Tech Reg Anesth Pain Manage 2005;9:152-60. Onofrio BM, Yaksh MC. Continuous delivery of opiates for persistent pain states. In: Gildenberg PL, Tasker RR, Franklin P, editors. Textbook of stereotactic and functional neurosurgery. New York: Mc Graw-Hill; 1996. p. 1463-76.
Intrathecal opiates for cancer pain
36. Onofrio BM, Yaksh TL. Long-term pain relief produced by intrathecal morphine infusion in 53 patients. J Neurosurg 1990;72:200-9. 37. Burton AW, Rajagopal A, Shah HN, et al. Epidural and intrathecal analgesia is effective in treating refractory cancer pain. Pain Med 2004;5:239-47. 38. Krames ES. Intraspinal opioid therapy for chronic nonmalignant pain: current practice and clinical guidelines. J Pain Symptom Manage 1996;11:333-52. 39. Smith TJ, Coyne PJ. How to use implantable intrathecal drug delivery systems for refractory cancer pain. J Support Oncol 2003;1:73-6. 40. Sunil J, Panchal MD, Julio A, et al. Intrathecal pumps. Tech Reg Anesth Pain Manage 2000;3:137-42. 41. Lazorthes Y, Sallerin B, Verdie JC, et al. [Management of intractable cancer pain: from intrathecal morphine to cell allograft]. Neurochirurgie 2000;46:454-65. 42. Lazorthes Y, Verdie JC, Bastide R, et al. Spinal versus intraventricular chronic opiate administration with implantable drug delivery devices for cancer pain. Appl Neurophysiol 1985;48:234-41. 43. Lazorthes Y, Sallerin B, Verdie JC. Intracerebroventricular administration of morphine for control of irreducible cancer pain, chapter 150. In: Gildenberg PL, Tasker RR, editors. Textbook of stereotactic and functional neurosurgery. New York: Mc Graw-Hill; 1996. p. 1477-82. 44. Wallace M, Yaksh TL. Long-term spinal analgesic delivery: a review of the preclinical and clinical literature. Reg Anesth and Pain Med 2000;25:117-57. 45. Baker L, Lee M, Regnard C, et al. Evolving spinal analgesia practice in palliative care. Palliat Med 2004;18:507-15. 46. Nitescu P, Hultman E, Appelgren L, et al. Bacteriology, drug stability and exchange of percutaneous delivery systems and antibacterial filters in long-term intrathecal infusion of opioid drugs and bupivacaine in ‘‘refractory’’ pain. Clin J Pain 1992;8:324-37. 47. Cherry DA, Gourlay GK, Cousins MJ, et al. A technique for the insertion of an implantable portal system for the long term epidural administration of opioids in the treatment of cancer pain. Anaesth Intensive Care 1985;13:145. 48. Holmfred A, Vikerfors T, Berggren L, et al. Intrathecal catheters with subcutaneous port systems in patients with severe cancer-related pain managed out of hospital: the risk of infection. J Pain Symptom Manage 2006;31:568-72. 49. Onofrio BM, Yaksh TL, Arnold PG. Continuous low-dose intrathecal morphine administration in the treatment of chronic pain of malignant origin. Mayo Clin Proc 1981;56:516-20. 50. Krames E. Implantable devices for pain control: spinal cord stimulation and intrathecal therapies. Best Pract Res Clin Anaesthesiol 2002;16:619-49. 51. Lazorthes Y, Sallerin B, Verdie JC, et al. Intrathecal and intra-cerebro-ventricular opioids: past uses and current indications, chapter 49. In: Burchiel, editor.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
129
Surgical management of pain. New York: Thieme Medical Publ Inc.; 2001. p. 625-32. Gillespie CS, Sherman DL, Fleetwood-Walker SM, . et al. Peripheral demyelination and neuropathic pain behavior in periaxin-deficient mice. Neuron 2000;26:523-31. Knight KH, Brand FM, McHaourab AS, et al. Implantable intrathecal pumps for chronic pain: highlights and updates. Croat Med J 2007;48:22-34. Follett KA, Naumann CP. A prospective study of catheterrelated complications of intrathecal drug delivery systems. J Pain Symptom Manage 2000;19:209-15. Albrecht E, Durrer A, Chedel D, et al. Intraparenchymal migration of an intrathecal catheter three years after implantation. Pain 2005;118:274-8. Harney D, Victor R. Traumatic syrinx after implantation of an intrathecal catheter. Reg Anesth Pain Med 2004;29:606-9. Huntoon MA, Hurdle MF, Marsh RW, et al. Intrinsic spinal cord catheter placement: implications of new intractable pain in a patient with a spinal cord injury. Anesth Analg 2004;99:1763-5, table of contents. Levin GZ, Tabor DR. Paraplegia secondary to progressive necrotic myelopathy in a patient with an implanted morphine pump. Am J Phys Med Rehabil 2005;84:193-6. North R, Cutchis RN, Epstein JA, et al. Spinal cord compression complicating subarachnoid infusion of morphine: case report and laboratory experience. Neurosurgery 1991;29:778-84. Coffey RJ, Burchiel K. Inflammatory mass lesions associated with intrathecal drug infusion catheters: report and observations on 41 patients. Neurosurgery 2002;50:78-86; discussion 86–7. Deer TR, Raso LJ, Garten TG. Inflammatory mass of an intrathecal catheter in patients receiving baclofen as a sole agent: a report of two cases and a review of the identification and treatment of the complication. Pain Med 2007;8:259-62. Yaksh TL, Hassenbusch S, Burchiel K, et al. Inflammatory masses associated with intrathecal drug infusion: a review of preclinical evidence and human data. Pain Med 2002;3:300-12. Toombs JD, Follett KA, Rosenquist RW, et al. Intrathecal catheter tip inflammatory mass: a failure of clonidine to protect. Anesthesiology 2005;102:687-90. Allen JW, Horais K, Tozier N, et al. Opiate pharmacology of intrathecal granulomas. Anesthesiology 2006;105: 590-8. Lehmberg J, Scheiwe C, Spreer J, et al. Late bacterial granuloma at an intrathecal drug delivery catheter. Acta Neurochir (Wien) 2006;148:899-901; discussion 901. Schuchard M, Krames ES, Lanning RM. Intraspinal analgesia for non malignant pain: a retrospective analysis for efficacy, safety, and feasibilityin 50 patients. Neuromodulation 1998;1:46-56. Schuchard M, Lanning RM, North R, et al. Neurologic sequelae of intraspinal drug delivery systems: results of a
2191
2192
129 68.
69.
70.
71.
72.
73.
74.
75. 76.
77.
78.
79.
80.
81.
82.
83.
Intrathecal opiates for cancer pain
survey of American implanters of implantable drug delivery systems. Neuromodulation 1998;1:137-48. Hassenbusch S, Burchiel K, Coffey RJ, et al. Management of intrathecal catheter-tip inflammatory masses: a consensus statement. Pain Med 2002;3:313-23. Choi A, Laurito CE, Cunningham FE. Pharmacologic management of posdural puncture headache. Ann Pharmacother 1996;30:831-9. Follett KA, Boortz-Marx RL, Drake JM, et al. Prevention and management of intrathecal drug delivery and spinal cord stimulation system infections. Anesthesiology 2004;100:1582-94. Boviatsis EJ, Kouyialis AT, Boutsikakis I, et al. Infected CNS infusion pumps. Is there a chance for treatment without removal? Acta Neurochir (Wien) 2004;146:463-7. Kallweit U, Harzheim M, Marklein G, et al. Successful treatment of methicillin-resistant Staphylococcus aureus meningitis using linezolid without removal of intrathecal infusion pump. Case report. J Neurosurg 2007;107: 651-3. Ubogu EE, Lindenberg JR, Werz MA. Transverse myelitis associated with Acinetobacter baumanii intrathecal pump catheter-related infection. Reg Anesth Pain Med 2003;28: 470-4. Koulousakis A, Kuchta J, Bayarassou A, et al. Intrathecal opioids for intractable pain syndromes. Acta Neurochir Suppl 2007;97:43-8. Dickenson AH. Mechanisms of the analgesic actions of opiates and opioids. Br Med Bull 1991;47:690-702. Holtsman M, Fishman SM. Opioid receptors, chapter 10. Essentials of pain medicine and regional anesthesia. 2nd ed. 2005. p. 87–91. Saeki S, Yaksh TL. Suppression of nociceptive responses by spinal mu opioid agonists: effects of stimulus intensity and agonist efficacy. Anesth Analg 1993;77:265-74. Yaksh TL. Spinal opiates: a review of their effect on spinal function with emphasis on pain processing. Acta Anaesthesiol Scand Suppl 1987;85:25-37. Brennum J, Arendt-Nielsen L, Horn A, et al. Qualitative sensory examination during epidural anaesthesia and analgesia in man: effects of morphine. Pain 1991;52: 75-83. Samuelsson H, Malberg F, Ericksson M, et al. Outcomes of epidural morphine treatment in cancer pain: nine years of clinical experience. J Pain Symptom Manage 1995;10: 105-12. Yaksh TL, Onofrio BM. Retrospective consideration of the doses of morphine given intrathecally by chronic infusion in 163 patients by 19 physicians. Pain 1987;31:211-23. Appelgren L, Nordborg C, Sjoberg M, et al. Spinal epidural metastasis: implications for spinal analgesia to treat ‘‘refractory’’ cancer pain. J Pain Symptom Manage 1997;13:25-42. Sallerin-Caute B, Lazorthes Y, Deguine O, et al. Does intrathecal morphine in the treatment of cancer pain induce the development of tolerance? Neurosurgery 1998;42:44-9; discussion 49–50.
84. Crul BJ, Delhaas EM. Technical complications during long-term subarachnoid or epidural administration of morphine in terminally ill cancer patients: a review of 140 cases. Reg Anesth 1991;16:209-13. 85. Smith TJ, Swainey C, Coyne PJ. Pain management, including intrathecal pumps. Curr Pain Headache Rep 2005;9:243-8. 86. Gesten Y, Vainio A, Pe´gurrier AM. Long-term intrathecal infusion of morphine in the home care of patients with advanced cancer. Acta Anaesthesiol Scand 1997;41: 12-17. 87. Gourlay GK, Plummer JL, Cherry DA, et al. Comparison of intermittent bolus with continuous infusion of epidural morphine in the treatment of severe cancer pain. Pain 1991;47:135-40. 88. Bedder MD, Burchiel K, Larson A. Cost analysis of two implantable narcotic delivery systems. J Pain Symptom Manage 1991;6:368-73. 89. Erdine S, De Andres J. Drug delivery systems. Pain Pract 2006;6:51-7. 90. Erdine S, Talu GK. Cost effectiveness of implantable devices versus tunneled catheters. Anesth Tech Pain Manag 1998;2:157-162. 91. Hassenbusch SJ, Paice JA, Patt RB, et al. Clinical realities and economic considerations: economics of intrathecal therapy. J Pain Symptom Manage 1997;14: S36-48. 92. Mueller-Schwefe G, Hassenbusch S, Reig E. Cost effectiveness of intrathecal therapy for pain. Neuromodulation 1999;2:77-84. 93. Van Dongen RT, Crul BJ, De Bock M. Long-term intrathecal infusion of morphine and morphine/bupivacaine mixtures in the treatment of cancer pain: a retrospective analysis of 51 cases. Pain 1993;55:119-23. 94. Ghafoor VL, Epshteyn M, Carlson GH, et al. Intrathecal drug therapy for long-term pain management. Am J Health Syst Pharm 2007;64:2447-61. 95. Gaudette KE, Weaver SJ. Intraspinal use of morphine. Ann Pharmacother 2003;37:1132-5. 96. Yaksh TL, Nouheihed R. The physiology and pharmacology of spinal opioids. Annu Rev Pharmacol Toxicol 1985;25:433. 97. Zieglgansberger W. Opiate actions on mamalian spinal neurons. Int Rev Neurobiol 1984;25:243. 98. Ventafridda V, Spoldi E, Caraceni A, et al. Intraspinal morphine for cancer pain. Acta Anaesthesiol Scand Suppl 1987;85:47-53. 99. Bennett G, Burchiel K, Buchser E, et al. Clinical guidelines for intraspinal infusion: report of an expert panel. J Pain Symptom Manage 2000;20:537-43. 100. Hassenbusch SJ, Portenoy RK, Cousins M, et al. Polyanalgesic consensus conference 2003: an update on the management of pain by intraspinal drug delivery – report of an expert panel. J Pain Symptom Manage 2004;27:540-63.
Intrathecal opiates for cancer pain
101. McQuay HJ, Sullivan AF, Smallman K, et al. Intrathecal opioids, potency and lipophilicity. Pain 1989;36: 111-15. 102. Waara-Wolleat KL, Hildebrand KR, Stewart GR. A review of intrathecal fentanyl and sufentanil for the treatment of chronic pain. Pain Med 2006;7:251-9. 103. Di Chiro G. Observations on the circulation of CSF. Acta Radiol 1996;5:988-1002. 104. Kedlaya D, Reynolds L, Waldman S. Epidural and intrathecal analgesia for cancer pain. Best Pract Res Clin Anaesthesiol 2002;16:651-65. 105. Eisenach J, Detweiler D, Hood D. Hemodynamic and analgesic actions of epidurally administered clonidine. Anesthesiology 1993;78:277-87. 106. Eisenach SC, DuPen S, Dubois M, et al. Epidural clonidine analgesia for intractable cancer pain. Pain 1995;61:391-9. 107. Hassenbusch SJ, Gunes S, Wachsman S, et al. Intrathecal clonidine in the treatment of intractable pain: a phase I/II study. Pain Med 2002;3:85-91. 108. Mercadante S. Problems of long-term spinal opioid treatment in advanced cancer patients. Pain 1999;79: 1-13. 109. Sjoberg M, Karlsson PA, Nordborg C, et al. Neuropathologic findings after long-term intrathecal infusion of morphine and bupivacaine for pain treatment in cancer patients. Anesthesiology 1992;76: 173-86. 110. Wagemans MF, van der Valk P, Spoelder EM, et al. Neurohistopathological findings after continuous intrathecal administration of morphine or a morphine/ bupivacaine mixture in cancer pain patients. Acta Anaesthesiol Scand 1997;41:1033-8. 111. Krames ES. The chronic intraspinal use of opioid and local anesthetic mixtures for the relief of intractable pain: when all else fails! Pain 1993;55:1-4. 112. Walker SM, Goudas LC, Cousins MJ, et al. Combination spinal analgesic chemotherapy: a systematic review. Anesth Analg 2002;95:674-715. 113. Buchser E, Durrer A, Chedel D, et al. Efficacy of intrathecal bupivacaine: how important is the flow rate? Pain Med 2004;5:248-52. 114. Prommer E. Ziconotide: a new option for refractory pain. Drugs Today (Barc) 2006;42:369-78. 115. Staats PS, Yearwood T, Charapata SG, et al. Intrathecal ziconotide in the treatment of refractory pain in patients with cancer or AIDS: a randomized controlled trial. Jama 2004;291:63-70. 116. Thompson JC, Dunbar E, Laye RR. Treatment challenges and complications with ziconotide monotherapy in established pump patients. Pain Physician 2006;9: 147-52. 117. Almeida RA, Lauretti GR, Mattos AL. Antinociceptive effect of low-dose intrathecal neostigmine combined with intrathecal morphine following gynecologic surgery. Anesthesiology 2003;98:495-8. 118. Grant GJ, Piskoun B, Bansinath M. Intrathecal administration of liposomal neostigmine prolongs analgesia in mice. Acta Anaesthesiol Scand 2002;46:90-4.
129
119. Naguib M, Yaksh TL. Antinociceptive effects of spinal cholinesterase inhibition and isobolographic analysis of the interaction with mu and alpha 2 receptor systems. Anesthesiology 1994;80:1338-48. 120. Hood DD, Eisenach JC, Tuttle R. Phase I safety assessment of intrathecal neostigmine methylsulfate in humans. Anesthesiology 1995;82:331-43. 121. Bennett G, Serafini M, Burchiel K, et al. Evidence-based review of the literature on intrathecal delivery of pain medication. J Pain Symptom Manage 2000;20:S12-36. 122. Inada T, Kushida A, Sakamoto S, et al. Intrathecal betamethasone pain relief in cancer patients with vertebral metastasis: a pilot study. Acta Anaesthesiol Scand 2007;51:490-4. 123. Laugner B, Muller A, Thiebaut JB, et al. [Analgesia with an implanted device for repetitive intrathecal injections of morphine]. Ann Fr Anesth Reanim 1985;4: 511-20. 124. Paice JA, Penn RD, Shott S. Intraspinal morphine for chronic pain: a retrospective, multicenter study. J Pain Symptom Manage 1996;11:71-80. 125. Penn RD, Paice JA. Chronic intrathecal morphine for intractable pain. J Neurosurg 1987;67:182-6. 126. Winkelmuller W, Burchiel K, Van Buyten JP. Intrathecal opioid therapy for pain: efficacy and outcomes. Neuromodulation 1999;2:67-76. 127. Gilmer-Hill HS, Boggan JE, Smith KA, et al. Intrathecal morphine delivered via subcutaneous pump for intractable cancer pain: a review of the literature. Surg Neurol 1999;51:12-15. 128. Meynadier J, Dubar M, Blond S, et al. Intrathecal morphine in treatment of intractable pain in cancer patients. Pain Clin 1985;1:87-91. 129. Follett KA, Hitchon PW, Piper J, et al. Response of intractable pain to continuous intrathecal morphine: a retrospective study. Pain 1992;49:21-5. 130. Chambers FA, MacSullivan R. Intrathecal morphine in the treatment of chronic intractable pain. Ir J Med Sci 1994;163:318-21. 131. Gestin Y, Vainio A, Pegurier AM. Long-term intrathecal infusion of morphine in the home care of patients with advanced cancer. Acta Anaesthesiol Scand 1997;41:12-17. 132. Mercadante S, Intravaia G, Villari P, . et al. Intrathecal treatment in cancer patients unresponsive to multiple trials of systemic opioids. Clin J Pain 2007;23:793-8. 133. Rauck RL, Cherry D, Boyer MF, . et al. Long-term intrathecal opioid therapy with a patient-activated, implanted delivery system for the treatment of refractory cancer pain. J Pain 2003;4:441-7. 134. Smith TJ, Coyne PJ. Implantable drug delivery systems (IDDS) after failure of comprehensive medical management (CMM) can palliate symptoms in the most refractory cancer pain patients. J Palliat Med 2005; 8:736-42. 135. Smith TJ, Staats PS, Deer T, et al. Randomized clinical trial of an implantable drug delivery system compared
2193
2194
129 136.
137.
138. 139.
140. 141.
142.
143.
144.
145.
146. 147.
148. 149.
150.
Intrathecal opiates for cancer pain
with comprehensive medical management for refractory cancer pain: impact on pain, drug-related toxicity, and survival. J Clin Oncol 2002;20:4040-9. Smith TJ, Coyne PJ, Staats PS, et al. An implantable drug delivery system (IDDS) for refractory cancer pain provides sustained pain control, less drug-related toxicity, and possibly better survival compared with comprehensive medical management (CMM). Ann Oncol 2005;16:825-33. Du Pen S, Du Pen AR, Polissar N, . et al. Implementing guidelines for cancer pain management: results of a randomized controlled clinical trial. J Clin Oncol 1999;17:361-70. Ruan X. Drug-related side effects of long-term intrathecal morphine therapy. Pain Physician 2007;10:357-66. Bailey PL, Rhondeau S, Schafer PG, . et al. Doseresponse pharmacology of intrathecal morphine in human volunteers. Anesthesiology 1993;79:49-9; discussion 25A. Ballantyne JC, Loach AB, Carr DB. Itching after epidural and spinal opiates. Pain 1988;33:149-60. Charuluxananan S, Kyokong O, Somboonviboon W, . et al. Nalbuphine versus ondansetron for prevention of intrathecal morphine-induced pruritus after cesarean delivery. Anesth Analg 2003;96:1789-93, table of contents. Raffaeli W, Marconi G, Fanelli G, et al. Opioid-related side-effects after intrathecal morphine: a prospective, randomized, double-blind dose-response study. Eur J Anaesthesiol 2006;23:605-10. Simoneau II, Hamza MS, Mata HP, et al. The cannabinoid agonist WIN55,212–2 suppresses opioid-induced emesis in ferrets. Anesthesiology 2001;94:882-7. Loper KA, Ready LB, Dorman BH. Prophylactic transdermal scopolamine patches reduce nausea in postoperative patients receiving epidural morphine. Anesth Analg 1989;68:144-6. Thoren T, Wattwil M. Effects on gastric emptying of thoracic epidural analgesia with morphine or bupivacaine. Anesth Analg 1988;67:687-94. Frederich ME. Nonpain symptom management. Prim Care 2001;28:299-316. Winkelmuller M, Winkelmuller W. Long-term effects of continuous intrathecal opioid treatment in chronic pain of nonmalignant etiology. J Neurosurg 1996;85:458-67. Morgan M. The rationale use of intrathecal and extradural opioids. Br J Anaesth 1989;63:165-88. Patt RB, Hassenbusch S. Implantable technology for paincontrol: identification and management of problems and complications, chapter 64. In: Waltman S, editor. Interventional pain management, 2nd ed. Orlando, FL: Saunders; 2001. p. 663. Rawal N, Mollefors K, Axelsson K, et al. An experimental study of urodynamic effects of epidural morphine and of naloxone reversal. Anesth Analg 1983;62:641-7.
151. Wang J, Pennefather S, Russel G. Low dose naloxone in the treatment of urinary retention during extradural fentanyl causes excessive reversal of analgesia. Br J Anaesth 1998;80:565-6. 152. Anderson VC, Burchiel KJ. A prospective study of longterm intrathecal morphine in the management of chronic nonmalignant pain. Neurosurgery 1999;44:289-300; discussion 300–1. 153. Porreca F, Filla A, Burks TF. Spinal cord mediated opiate effects on gastrointestinal transit in mice. Eur J Pharmacol 1983;86:135-6. 154. Yaksh TL, Noveihed R. The physiology and pharmacology of spinal opiates. Ann Rev Pharmacol Toxicol 1985;25:433-62. 155. Mahajan G, Fishman SM. Opioid therapy: adverse effects including addiction, chapter 13. Essentials of pain medecine and regional anesthesia. 2nd ed. 2005. p. 113–23. 156. Anderson VC, Cooke B, Burchiel K. Intrathecal hydromorphone for chronic non malignant pain: a retrospective study. Pain Med 2001;2:287-97. 157. Aldrete JA, Couto da Silva JM. Leg edema from intrathecal opiate infusions. Eur J Pain 2000;4:361-5. 158. Scherens A, Kagel T, Zenz M, et al. Long-term respiratory depression induced by intrathecal morphine treatment for chronic neuropathic pain. Anesthesiology 2006;105:431-3. 159. Shook JE, Watkins WD, Camporesi EM. Differential roles of opioid receptors in respiration, respiratory disease, and opioid-induced respiratory depression. Am Rev Respir Dis 1990;142:895-909. 160. Hammond JE. Reversal of opioid associated late onset respiratory depression by nalbuphine hydrochloride. Lancet 1984;2:1208. 161. Sidi A, Davidson JT, Behar M, et al. Spinal narcoticsand central nervous depression. Anaesthesia 1981;36: 1044-7. 162. Paulus DA, Paul W, Munson ES. Neurological depressionafter intrathecal morphine. Anesthesiology 1981;54: 517-8. 163. Christie JM, Meade WR, Markowsky S. Paranoid psychosis after intrathecal morphine. Anesth Analg 1993;77:1298-9. 164. Dunbar SA, Karamian IG. Periodic abstinence enhances nociception without significantly altering the antinociceptive efficacy of spinal morphine in the rat. Neurosci Lett 2003;344:145-8. 165. Johnston IN, Milligan ED, Wieseler-Frank J, . et al. A role for proinflammatory cytokines and fractalkine in analgesia, tolerance, and subsequent pain facilitation induced by chronic intrathecal morphine. J Neurosci 2004;24:7353-65. 166. Mao J, Sung B, Ji RR, et al. Chronic morphine induces downregulation of spinal glutamate transporters: implications in morphine tolerance and abnormal pain sensitivity. J Neurosci 2002;22:8312-23.
Intrathecal opiates for cancer pain
167. Vanderah TW, Gardell LR, Burgess SE, et al. Dynorphin promotes abnormal pain and spinal opioid antinociceptive tolerance. J Neurosci 2000;20: 7074-9. 168. Yaksh TL, Harty GJ, Onofrio BM. High dose of spinal morphine produce a nonopiate receptor-mediated hyperesthesia: clinical and theoretic implications. Anesthesiology 1986;64:590-7. 169. Parisod E, Siddall PJ, Viney M, et al. Allodynia after acute intrathecal morphine administration in a patient with neuropathic pain after spinal cord injury. Anesth Analg 2003;97:183-6, table of contents. 170. Wilson GR, Reisfield GM. Morphine hyperalgesia: a case report. Am J Hosp Palliat Care 2003;20:459-61. 171. Angst MS, Clark JD. Opioid induced hyperalgesia: a qualitative systematic review. Anesthesiology 2006; 104:570-87. 172. Yaksh TL, Harty G. Pharmacology of the allodynia in rats evoked by high dose intrathecal morphine. J Pharmacol Exp Ther 1988;244:501-7. 173. Sakurada T, Komatsu T, Sakurada S. Mechanisms of nociception evoked by intrathecal high-dose morphine. Neurotoxicology 2005;26:801-9. 174. Davis AM, Inturrisi CE. d-Methadone blocks morphine tolerance and N-methyl-D-aspartate-induced hyperalgesia. J Pharmacol Exp Ther 1999;289:1048-53. 175. Davis MP, Lasheen W, Gamier P. Practical guide to opioids and their complications in managing cancer pain. What oncologists need to know. Oncology (Williston Park) 2007;21:1229-38; discussion 1238–46, 1249. 176. Inturrisi CE. Pharmacology of methadone and its isomers. Minerva Anestesiol 2005;71:435-7. 177. Mercadante S, Villari P, Ferrera P. Naloxone in treating central adverse effects during opioid titration for cancer pain. J Pain Symptom Manage 2003;26:691-3. 178. Bennett G, Deer T, Du Pen S, et al. Future directions in the management of pain by intraspinal drug delivery. J Pain Symptom Manage 2000;20:S44-50. 179. Mao J, Sung B, Ji RR, et al. Neuronal apoptosis associated with morphine tolerance: evidence for an opioid-induced neurotoxic mechanism. J Neurosci 2002;22:7650-61. 180. Coombs DW, Colburn RW, DeLeo JA, et al. Comparative spinal neuropathology of hydromorphone and morphine after 9- and 30-day epidural administration in sheep. Anesth Analg 1994;78:674-81. 181. Hodgson PS, Neal JM, Pollock JE, et al. The neurotoxicity of drugs given intrathecally (spinal). Anesth Analg 1999;88:797-809. 182. Rawal N, Nuutinen L, Raj PP, et al. Behavioral and histopathologic effects following intrathecal administration of butorphanol, sufentanil, and nalbuphine in sheep. Anesthesiology 1991;75:1025-34. 183. Krames ES, Gershow J, Glassberg A, et al. Continuous infusion of spinally administered narcotics for the relief
184.
185.
186.
187.
188.
189.
190.
191.
192.
193.
194.
195.
196.
197.
198.
129
of pain due to malignant disorders. Cancer 1985;56: 696-702. Coombs DW, Fratkin J, Meier FA, et al. Neuropathologic lesions and CSF morphine concentrations during chronic continuous intraspinal morphine infusion. A clinical and post-mortem study. Pain 1985; 22:337-51. Dickerman RD, Schneider SJ. Recurrent intrathecal baclofen pump catheter leakage: a surgical observation with recommendations. J Pediatr Surg 2002;37:E17. Ko WM, Ferrante FM. New onset lumbar radicular pain after implantation of an intrathecal drug delivery system: imaging catheter migration. Reg Anesth Pain Med 2006;31:363-7. Dickerman RD, Stevens QE, Schneider SJ. The role of surgical placement and pump orientation in intrathecal pump system failure. A technical report. Pediatr Neurosurg 2003;38:107-9. Burton AW, Conroy B, Garcia E, et al. Illicit substance abuse via an implanted intrathecal pump. Anesthesiology 1998;89:1264-7. Leavens ME, Hill CS, Jr, Cech DA, et al. Intrathecal and intraventricular morphine for pain in cancer patients: initial study. J Neurosurg 1982;56:241-5. Ballantyne JC, Carwood CM. Comparative efficacy of epidural, subarachnoid, and intracerebroventricular opioids in patients with pain due to cancer. Cochrane Database Syst Rev2005;CD005178. Blond S, Meynadier J, Dupard T, et al. [Intracerebroventricular morphine therapy. Apropos of 79 patients]. Neurochirurgie 1989;35:52-7. Houdek M, Opavsky J, Ostrizek J. Intracerebroventricular application of morphine in the treatment of intractable malignant pain. Acta Univ Palacki Olomuc Fac Med 1990;128:101-6. Lazorthes YR, Sallerin BA, Verdie JC. Intracerebroventricular administration of morphine for control of irreducible cancer pain. Neurosurgery 1995;37: 422-8; discussion 428–9. Sandouk P, Serrie A, Urtizberea M, et al. Morphine pharmacokinetics and pain assessment after intracerebroventricular administration in patients with terminal cancer. Clin Pharmacol Ther 1991;49:442-8. Seiwald M, Alesch F, Kofler A. [Intraventricular morphine administration as a treatment possibility for patients with intractable pain]. Wien Klin Wochenschr 1996;108:5-8. Lazorthes Y. Intracerebroventricular administration of morphine for control of irreducible cancer pain. Ann N Y Acad Sci 1988;531:123-32. Baker L, Balls J, Regnard C, et al. Cervical intrathecal analgesia for head and neck/upper limb cancer pain: six case reports. Palliat Med 2007;21:543-5. Lazorthes Y, Bes JC, Sol JC, et al. Intrathecal chromaffin cell allograft for cancer pain. In: Burchiel, editor. Surgical management of pain. New York, 2002.
2195
2196
129
Intrathecal opiates for cancer pain
199. Czech KA, Sagen J. Update on cellular transplantation into the CNS as a novel therapy for chronic pain. Prog Neurobiol 1995;46:507-29. 200. Livett BG, Dean DM, Whelan LG, et al. Co-release of enkephalin and catecholamines from cultured adrenal chromaffin cells. Nature 1981;289:317-9. 201. Unsicker K. The trophic cocktail made by adrenal chromaffin cells. Exp Neurol 1993;123:167-73. 202. Sagen J, Kemmler JE. Increased levels of met-enkephalinlike immunoreactivity in the spinal cord CSF of rats with adrenal medullary transplants. Brain Res 1989; 502:1-10. 203. Ginzburg R, Seltzer Z. Subarachnoid spinal cord transplantation of adrenal medulla suppresses chronic neuropathic pain behavior in rats. Brain Res 1990;523:147-50. 204. Guenot M, Lee JW, Nasirinezhad F, et al. Deafferentation pain resulting from cervical posterior rhizotomy is alleviated by chromaffin cell transplants into the rat spinal subarachnoid space. Neurosurgery 2007; 60:919-25; discussion 919–25. 205. Wang H, Sagen J. Absence of appreciable tolerance and morphine cross-tolerance in rats with adrenal medullary transplants in the spinal cord. Neuropharmacology 1994;33:681-92. 206. Lazorthes Y, Bes JC, Sagen J, et al. Transplantation of human chromaffin cells for control of intractable cancer pain. Acta Neurochir Suppl (Wien) 1995;64:97-100. 207. Pappas GD, Lazorthes Y, Bes JC, et al. Relief of intractable cancer pain by human chromaffin cell transplants: experience at two medical centers. Neurol Res 1997;19:71-7. 208. Lazorthes Y, Sagen J, Sallerin B, et al. Human chromaffin cell graft into the CSF for cancer pain management: a prospective phase II clinical study. Pain 2000; 87:19-32. 209. Bes JC, Tkaczuk J, Czech KA, et al. One-year chromaffin cell allograft survival in cancer patients with chronic pain: morphological and functional evidence. Cell Transplant 1998;7:227-38. 210. Buchser E, Goddard M, Heyd B, et al. Immunoisolated xenogenic chromaffin cell therapy for chronic pain. Initial clinical experience. Anesthesiology 1996;85: 1005-12; discussion 1029A–30A. 211. Sagen J, Pappas GD, Pollard HB. Analgesia induced by isolated bovine chromaffin cells implanted in rat spinal cord. Proc Natl Acad Sci USA 1986;83:7522-6. 212. Sol JC, Sallerin B, Larrue S, et al. Intrathecal xenogeneic chromaffin cell grafts reduce nociceptive behavior in a rodent tonic pain model. Exp Neurol 2004;186: 198-211.
213. Jozan S, Aziza J, Chatelin S, et al. Human fetal chromaffin cells: a potential tool for cell pain therapy. Exp Neurol 2007;205:525-35. 214. Duplan H, Li RY, Vue C, . et al. Grafts of immortalized chromaffin cells bio-engineered to improve met-enkephalin release also reduce formalin-evoked c-fos expression in rat spinal cord. Neurosci Lett 2004;370:1-6. 215. Eaton MJ, Martinez M, Karmally S, et al. Initial characterization of the transplant of immortalized chromaffin cells for the attenuation of chronic neuropathic pain. Cell Transplant 2000;9:637-56. 216. Eaton MJ. Cell and molecular approaches to the attenuation of pain after spinal cord injury. J Neurotrauma 2006;23:549-59. 217. Eaton MJ, Wolfe SQ, Martinez M, et al. Subarachnoid transplant of a human neuronal cell line attenuates chronic allodynia and hyperalgesia after excitotoxic spinal cord injury in the rat. J Pain 2007;8:33-50. 218. Eaton MJ. Emerging cell and molecular strategies for the study and treatment of painful peripheral neuropathies. J Peripher Nerv Syst 2000;5:59-74. 219. Cope DK, Lariviere WR. Gene therapy and chronic pain. ScientificWorldJournal 2006;6:1066-74. 220. Meunier A, Latremoliere A, Dominguez E, et al. Lentiviral-mediated targeted NF-kappaB blockade in dorsal spinal cord glia attenuates sciatic nerve injuryinduced neuropathic pain in the rat. Mol Ther 2007; 15:687-97. 221. Pohl M, Braz J. Gene therapy of pain: emerging strategies and future directions. Eur J Pharmacol 2001;429:39-48. 222. Pohl M, Meunier A, Hamon M, et al. Gene therapy of chronic pain. Curr Gene Ther 2003;3:223-38. 223. Rainov NG, Heidecke V. Experimental therapies for chronic pain. Acta Neurochir Suppl 2007;97:473-7.
130 Management of Pain of Benign Versus Cancer Origin P. L. Gildenberg . R. A. DeVaul
Pain remains a perplexing problem to the clinician, despite numerous advances in the understanding of the physiology of pain perception. Rather than improving pain management, advances in understanding the physiology of pain have perhaps added to the confusion, since laboratory studies generally involve acute pain, whereas the pain generally referred for management is chronic pain. The general trend in management of pain is more often toward procedures, many of which are inappropriate, with less concern about the type of pain, options for treatment, and the many factors that influence pain perception. Patients with acute pain are generally managed by an acute care team, primary physician or surgeon, and are not referred to the neurosurgeon. It is only after the primary care doctor becomes frustrated by pain that does not seem to go away that the patient is referred to a neurosurgeon with even less experience managing chronic pain, expecting that some new miraculous procedure will cure the pain. There aren’t any! The wise neurosurgeon will keep the following in mind:
#
‘‘There ain’t no magic!’’ The key to management of chronic pain is to listen to the patient The keys to managing intractable chronic pain are identifying what factors are causing the pain, if possible identifying what factors make the pain worse, and managing those factors that can be managed Springer-Verlag Berlin/Heidelberg 2009
Both the patient and the doctor must set mutually agreed realistic goals The goal of some patients may be to remain a chronic pain patient ‘‘I’ve got to do something!’’ is NOT an indication for surgery Beware the Pain Clinic where everyone gets the same procedure The treatment the chronic pain patient receives is more closely related to the type of specialist to whom the patient is referred, rather than to the pain itself It is ill advised to do a procedure first, and if the patient is not cured then send him or her to a pain doctor. First evaluate what is best for this individual patient
Questions for management of pain
When is it appropriate To give analgesics To avoid analgesics To perform a procedure – Stimulation procedure – Ablative lesion What factors make pain worse? Which factors can be controlled to lessen pain
Definitions The pain the physician sees in the office differs significantly from the pain the physiologist studies. Although both relate to an intolerably
2198
130
Management of pain of benign versus cancer origin
disagreeable sensation, the physiologist’s definition is narrow, relating to acute sensation that is carried by the primary pain pathways. Clinical pain, however, is a perception that is influenced by many things not directly related to what caused the pain in the first place. The type of pain which most commonly vexes the physician and is most commonly referred in frustration to the neurosurgeon is chronic pain. To make matters worse, the term chronic pain has been used in a number of different ways, so that different authors use different definitions when writing about different clinical circumstances. The first step toward establishing a rational approach to management of pain is to define the types of clinical pain specifically, so the reader can relate information about pain management to patients with the appropriate type of pain. To that end we would recommend the nomenclature in > Table 130-1, which will be used throughout this chapter. Although these terms are not used universally, they can serve to establish specific definitions for each type of clinical pain so the clinician may establish criteria for a rational management program for each type of pain. Indeed, the management strategies of the various types of clinical pain are quite different. As noted in > Table 130-1, most authors refer to pain as either ‘‘acute’’ or ‘‘chronic,’’ if they make any distinction at all. By this usual definition, acute pain is immediate and short lasting. Chronic pain is any pain that lasts for an arbitrarily long time. According to the usual loose nomenclature, a number of types of pain
. Table 130-1 Types of clinical pain Usual nomenclature
Nomenclature we prefer
Acute pain Chronic pain Cancer pain Pain of ‘‘benign’’ origin
Acute pain Persistent pain Cancer pain Chronic pain
are classified together as chronic pain. Thus, there is no differentiation between types of pain that should be treated very differently when they are classified together, which may prompt the treating physician to treat many patients inappropriately. For instance, the doctor may seek a neurosurgical ‘‘cure’’ of a chronic pain for which appropriate therapy may not be surgical. There is a further naive tendency to consider that chronic pain is either ‘‘physiological’’ or ‘‘psychological,’’ that is, the pain is either ‘‘real’’ (physiological) or ‘‘imagined’’ (psychological). Unfortunately, the usual criterion for this distinction is that if the pain is successfully treated it is ‘‘organic’’ but if the physician fails to ‘‘cure’’ the pain it is the patient’s fault and consequently totally psychological. Obviously, this classification of pain is useless. Most of the patients we see with in-tractable chronic pain have an underlying organic etiology for their pain and psychological factors that intensify and perpetuate the pain. We propose the classification that appears in the right column of > Table 130-1. Acute pain is indeed pain that, because of its clinical nature, lasts a brief period. This is ordinarily pain of an acute injury, postsurgical pain, or pain which accompanies an acute clinical condition, such as an inflammation or abscess. Acute pain ordinarily resolves when the underlying etiology begins to ease, or, at any rate, within a few days. The pain that is not acute (that is, pain that persists for more than a few days) is sometimes due to cancer and sometimes due to nonneoplastic conditions. Because the characteristics of long-lasting pain caused by cancer are much different than the characteristics and the management of other types of long-lasting pain, cancer pain is unique, with an approach that is reasonably well defined. We prefer to reserve the term chronic pain for long-lasting non-cancer pain, since there is no other reasonable term for it and since that term is already commonly used (but unfortunately not exclusively used) by default. This
Management of pain of benign versus cancer origin
type of chronic pain is the pain that is most difficult to treat and the pain which often prompts a neurosurgical consultation or a referral to a comprehensive pain management program – note that the goal is management and not necessarily cure. It is also the type of pain with the least likelihood of successful treatment, and consequently is the most frustrating for physician and patient alike, especially when both have unrealistic expectations. Because both cancer pain and chronic pain share a number of characteristics, chief among which is that they last for a long time and demand treatment onto themselves, the encompassing term persistent pain can be used to define long-lasting pain, referring either to long-lasting chronic pain or cancer pain.
General Rules of Pain Management The question of whether pain is ‘‘real’’ is generally moot. Pain is a totally subjective sensation that can be perceived only by the patient. If the patient perceives pain, it is real enough and requires understanding and management, whether or not it is of an organic etiology. The clinician has only the patient’s description of the pain to know what is to be treated. The physician also has only the patient’s report to assess whether the treatment has been successful. Unfortunately, interpretation of the mechanism of chronic pain is often complicated by psychological factors of the physician, who may feel threatened or accused because the patient still has pain after treatment. In our experience, purely psychogenic pain is unusual in a chronic pain clinic practice. However, psychological factors usually influence perception of organic pain and consequently the severity of pain, regardless of the etiology [10]. Unsophisticated physicians commonly assume that any pain for which they cannot find an etiology is psychogenic. For instance, a patient
130
with a badly sprained back with chronic pain may have normal x-ray studies, from which the doctor concludes that ‘‘it is all in your head.’’ This may help the physician believe he or she is absolved of responsibility for the patient, provide an excuse for the insurance company to deny further management, or provide an employer an excuse for terminating benefits, but the pain may be just as disabling as if the x-ray showed an identifiable fracture. However, it is an equally common error to see benign nonpainful degenerative disease on an x-ray of a chronic pain patient and automatically assume that it is related to the pain. The sensation of pain is not like any of the primary senses, such as vision, smell, touch, temperature, or hearing. Those sensations are specific, perceived by everyone identically, and are immediate responses to a particular stimulus. However, appreciation of pain requires considerable processing of the information before it reaches consciousness. Much of that information processing is influenced by factors other than the immediate noxious stimulus. External factors may influence pain perception, such as the circumstances in which the injury is acquired. For example, an injury that is sustained in a hostile environment may cause very little pain until the individual escapes to a safer place. Pain perception is particularly influenced by formidable internal influences which are significant factors in the perception and management of chronic pain. If the psychological need for pain and/or disability is sufficient, perception of pain may be magnified. If there is sufficient secondary gain, pain may be magnified. Stress and depression also magnify pain perception significantly, even if unrelated to the physical problem, and must be dealt with as part of pain management. Differentiating between pain and ‘‘suffering,’’ both of which the patient may express as pain, is especially important. Even with a small physiological component to the pain, there may be a great deal of suffering, and the patient may
2199
2200
130
Management of pain of benign versus cancer origin
be significantly disabled from a minor injury. Although the immediate somatic stimulus for pain may be alleviated, the patient may continue to have distress and disability if the suffering component of the problem is not also dealt with. Each of the three types of pain we described in > Table 130-1 requires different management, which makes it particularly important to categorize patients appropriately right from the start. Management of acute pain is directed toward treating the underlying etiology while allowing the patient to be comfortable, so that the pain is only a secondary consideration. Analgesics and rest are appropriate. Since most acute pain lasts only a few days, a week or two at the most, there is little concern for addiction to pain medications except in those patients with a preexisting addictive personality or history. Analgesics such as narcotics should be given in sufficient amounts, particularly when the pain is severe, as in postsurgical pain. Inappropriate concern about addiction or long-term effects of narcotics may deprive the patient of pain relief. However, even in the acute phase, a plan must be formulated to discontinue narcotics at the appropriate time. As a general rule, the physician should have a general idea of when the narcotics will be discontinued even before the first prescription is written. However, the dose of narcotic must be appropriate to the severity of the pain, even if that requires taking large doses initially, especially postoperatively. It fact, withholding narcotics during that period may provoke other problems that intensify pain perception, make the pain worse, and prolong the need for pain management. Cancer pain has many of the characteristics of both acute pain and chronic pain. As in acute pain, cancer pain may occur from tissue damage secondary to malignancy or sometimes secondary to treatment [3,8]. Treating the underlying etiology may or may not be possible, since it may
be secondary to an infiltrative neoplasm that has already been treated to the maximum. Nevertheless, just as in acute pain, treatment of the underlying etiology and administration of sufficient analgesics constitutes the treatment. If the pain persists or progresses despite treatment of the underlying cancer, the pain becomes a focus of management, and pain treatment becomes proactive. Cancer pain also has many of the characteristics of chronic pain. Many of the same psychological factors that magnify chronic pain also magnify and potentiate cancer pain and require attention. Indeed, depression and regression may be the result of the cancer itself, and they can make the pain worse and consequently must be addressed as part of a pain management program. Doses of analgesics must be escalated as the pain progresses and as the cancer becomes more severe. Ordinarily by that time the prognosis is so short that one need not worry about eventual withdrawal of narcotics. Fully 85–90% of cancer patients can be kept reasonably pain free by oral analgesics and related medications. As the pain becomes severe, however, administration of narcotics by mouth or intramuscularly may become ineffective, especially as side effects limit the dose that can be administered. If the pain is still disabling at that point, narcotics may be administered parenterally or into the cerebrospinal fluid, or ablative surgical procedures might be considered. Chronic pain most often begins with acute injury or disease of tissues, and ordinarily starts as an acute pain. However, the pain does not ease as the acute condition resolves, and subsequently the pain itself becomes the problem. The pain may persist because the tissue does not return to normal functioning, which is common with muscle or myofascial injury, where muscle may remain tense and in spasm for a long time. There may be no specific treatment for chronic peripheral neuropathy, which may result in
Management of pain of benign versus cancer origin
persistent hyperpathia because of malfunction of sensory nerves. The arbitrary clinical definition of chronic pain is pain that lasts for 6 months or more. This pain outlasts its biological function and thus no longer has a physiological usefulness. Although acute pain is protective in nature, chronic pain may be destructive. The pain pathways associated with acute pain may be only minimally active in chronic pain, whereas other factors may magnify and potentiate the chronic pain. The original etiology of the pain may no longer exist—the pain itself becomes the abnormality to be treated. Two patients with virtually identical injuries may have vast differences in the severity of chronic pain. One patient will return to work and to normal functioning rapidly, even before the injured tissues have healed. The other patient will have terrible pain and remain disabled, sometimes permanently. What is the difference in these two patients? Even though the injuries may be identical, external factors such as litigation, narcotics, or significant regression may cause a worsening of the pain over time. Severity of chronic pain may be intensified significantly by a variety of psychological and pharmacological factors in some patients. This complex multifaceted syndrome requires management in a comparably complex and multidisciplinary manner. The identification of those factors is the key to chronic pain management. A management program must first identify those factors which magnify, promote, and potentiate the pain. Usually by the time the patient is referred for a chronic pain program, treatable physical factors have already been dealt with, or, more commonly, have been overtreated. Those factors which prevent rehabilitation must be identified and managed. Narcotics are withdrawn, the patient is urged to become more active, depression and regression are addressed, and the patient is
130
encouraged to return to productive adult behavior. In general, the management program is exactly the opposite of the program for acute pain. Only after factors which intensify pain have been evaluated and managed may a few patients become candidates for surgical intervention. Indeed, surgery was found to be indicated in only 5% of our Chronic Pain Clinic patients. Because the physiological etiology of the pain very often plays only a minor role in perpetuating the pain, and because pain pathways are sufficiently redundant so that interruption of one pathway may only lead to the opening of another, simple ablative procedures are rarely the answer to chronic pain. The use of temporary blocks in and of themselves is rarely successful and is generally indicated only as part of an overall comprehensive pain management program. Consequently, the physician must consider how much of each component of the three p’s may be involved in an individual patient – physiological factors, pathological factors, or psychological factors. Physiological pain is that pain which results from the appropriate response of an intact nervous system to injured tissue. This is usually the major factor in acute pain. It is appropriate to feel pain, which in this scenario is a normal response to injury. Pathological factors, from the standpoint of this discussion, refers to pain that is produced because of a malfunction in the nervous system. This is the pain of causalgia or dysesthesia. Attention is directed toward treating the nervous system, rather than or in addition to somatic injury. Psychological factors are the most difficult to evaluate and to manage. It is only through talking with the patient in an unhurried manner that such factors can be revealed. These factors are the ones most often ignored, although the patient can often derive great benefit of lessened pain from dealing with psychological factors. Indeed, it may not be possible to treat the physical or
2201
2202
130
Management of pain of benign versus cancer origin
pathological factors, but the patient may benefit significantly by addressing those psychological factors that make the pain worse over time. Much confusion and many mistakes in pain management occur because the clinician does not consider the distinctions among the three major categories of clinical pain and may assign the patient to the wrong category or use a management strategy that is appropriate for a different category of pain. Consequently, the first consideration is to decide whether the pain is acute pain, cancer pain, or chronic pain. Patients with acute pain are rarely referred to a neurosurgeon, so little further will be said about the management of acute pain. It is normal for patients who have just suffered an injury or have just gone through surgery to have pain, sometimes very severe. Management begins at the time of the acute episode. Patients should receive a sufficient amount of analgesics, narcotic if necessary, to have relief of most if not all the pain. Patients with a prior history of substance abuse must be titrated closely, since they may have significant tolerance and may be happy to return to a state of addiction. With both cancer pain and chronic pain, however, factors that intensify pain perception must be evaluated and managed, in addition to the somatic component of the pain. Many of those factors are psychological. Consequently, the neurosurgeon concerned with pain management needs a firm appreciation of the psychological factors involved with pain perception and should work closely with an expert in that field to provide the patient with the best evaluation and management. Psychological factors not only magnify pain perception, they also set up vicious cycles (discussed below) that potentiate chronic pain and interfere with management. Unless these factors are tended to, patients will be selected inappropriately for surgery, and even those patients who are good candidates for surgery may end up with results less favorable
than if the patient underwent a comprehensive pain program as well as the surgery. The time to evaluate and address the factors that make pain worse is in the initial interview. Reserving such management for only those patients who have failed a surgical procedure or multiple anesthetic blocks is not good practice.
What Makes Pain Worse? As one reads about selection of patients for pain procedures, there is always an admonition to exclude patients with ‘‘psychological factors’’ or comment that ‘‘psychological factors may have an adverse influence on results.’’ However, the ‘‘psychological factors’’ are rarely defined. Sometimes they are lumped under the diagnosis of ‘‘chronic pain syndrome,’’ but, again, the details of that syndrome are rarely specified. However, these psychological factors form the basis of management in a multidisciplinary chronic pain clinic setting and are important to consider in patient selection for surgery. To help the neurosurgeon (as well as other pain specialists) evaluate patients with persistent pain, whether it be cancer pain or chronic pain, the factors that make pain worse are divided into five categories: [1] inappropriate use of narcotics, [2] depression, [3] physical regression, [4] psychological regression, and [5] intolerance to stress. These five factors must be managed as part of the effort to treat the pain. If these factors are not considered, there is high risk of failure. This constellation of these five factors comprises a condition referred to herein as the chronic pain syndrome. All five of these factors make pain worse, so that the patient gets into an ever-escalating spiral of pain and disability. Successful management requires attention to all five of these factors, which actually constitute the basis for comprehensive multimodality pain clinics. These five factors must be addressed as
Management of pain of benign versus cancer origin
part of the preoperative evaluation and management when pain procedures are contemplated. Treating chronic pain is frustrating to both physician and patient. Although the mechanism of the pain may not be identifiable, those factors that make the pain worse are often not only apparent after initial patient evaluation, but treatable, as well. In this respect, the physician more often manages chronic pain than treats or cures it. The key to managing intractable chronic pain is [1] identifying what factors are causing the pain, if possible, [2] identifying what factors make the pain worse, and then [3] managing those factors that can be managed. The goal for patient and doctor must be clearly redefined as not cure by the doctor but the best social function by the patient – rehabilitation with medical guidance.
Inappropriate Use of Narcotics Narcotics were designed for acute pain and were not intended to be taken over a long period. In acute pain, the dose of narcotic must be appropriate to the severity of the pain; in fact, withholding narcotics during that period may provoke the other problems that intensify pain perception and make the pain worse and prolong the need for pain management. However, the usual patients who present for a chronic pain program use narcotics inappropriately in escalating doses and have other manifestations of the chronic pain syndrome. They frequently have addictive personalities with a history of abuse of other drugs, alcohol, or other medications as well, or they may be taking narcotics inappropriately because these have been prescribed by their physician, who did not know any better. There is no drug designed for chronic pain. All analgesics are designed on the acute pain physiological model. Since the physiology of chronic pain is much different than that of
130
acute pain, narcotics and other analgesics are often ineffective. Successful management does not involve shifting from one narcotic to another, escalating the dose, or adding another analgesic drug. Most chronic pain patients have less pain when they are taking no narcotics at all than when they are taking doses of narcotics that they find frustratingly ineffective. When narcotics are administered over a long period, they suppress the normal production of endorphins. One can see how the loss of this internal resource may increase pain perception. As tolerance to the administered narcotic develops, the patient is left both without the benefit of endorphins and without the benefit of administered analgesics, with a resultant increase in pain perception. Patients who take narcotics over a long period may develop a pattern of recurrent withdrawal. Some of the first manifestations of withdrawal are heightened pain perception and increased muscle tone (which may increase pain of muscle origin) and agitation. If a patient has been taking narcotics regularly four to six times a day, a pattern of tolerance develops within a few weeks. As the time for the next dose of narcotic approaches, the patient begins to go into withdrawal, the first manifestation of which is heightened pain perception. The patient recognizes that the pain is getting worse and takes the next dose of narcotic. The narcotic treats the withdrawal with its heightened pain perception, so that the pain level may decrease to its basal state. However, because there is so much tolerance, the basal state may be the same physiological level of pain as if narcotics had not been administered at all. The decrease in pain perception after a dose of narcotic may not be due to analgesia. The narcotic may treat the withdrawal and not the pain. This can happen many times a day, and each time it happens it reinforces the narcotic seeking and heightens the intensity of the pain during the withdrawal. The patient complains to
2203
2204
130
Management of pain of benign versus cancer origin
the doctor that the narcotic really doesn’t take the pain away, but that he or she needs the narcotic ‘‘because it takes the edge off.’’ This catch phrase usually indicates that the narcotic is being used to treat recurrent withdrawal rather than pain. Thus we have a vicious cycle prompted by the inappropriate chronic administration of narcotics. The pain leads to narcotic use, which leads to recurrent withdrawal, which leads to more pain which leads to more narcotics, more withdrawal, and so forth (> Figure 130-1). Treatment of recurrent withdrawal is necessary to discontinue narcotic use. Before the narcotic can be discontinued, the reasoning must be explained to the patient—why the narcotics are making things worse rather than better and why it is necessary to decrease the dose of narcotic, usually stopping it completely. Once the need to discontinue the narcotics has been explained to the patient, withdrawing the narcotics abruptly is often possible if the patient is well enough motivated. A period of increased pain perception may last for about 3 days, although it occasionally may last for several weeks. If necessary, the dose can be tapered over 10–14 days. In general, if the dose is decreased by 15% of the initial dose, and each day sees a decrease of that same amount, the narcotics can be successfully tapered over 8 days with little additional withdrawal. An overly defensive refusal to discontinue narcotics often indicates a very poor prognosis. Such patients frequently have a history of alcohol or other substance abuse and are substituting the narcotic for the other drug. . Figure 130-1
Note that there has been a recent movement to justify the long-term use of narcotics for chronic pain, as is used in cancer pain. A significant aspect of this movement is the availability and marketing of oxycodone, a slow releasing heroin derivative [13]. While this drug may have some advantages in reducing the unevenness of shorter acting narcotic blood levels and being especially useful in cancer treatment or the rare patient with chronic pain, it carries the same problems as all narcotic treatment for chronic pain syndrome [17]. Marketing for OxyContin for benign pain has led to several epidemics of hillbilly heroin abuse [2,16]. OxyContin has a sustained release coating. If this coating is damaged by cutting or crushing the tablet, the oxycodone is released abruptly, providing a large addicting dose that produces a narcotic high. Oxycodone is not a heroin derivative. It is a derivative of opium, and is similar in chemical structure to codeine. Oxycodone has similar effects to morphine and heroin. It is 50% more powerful than hydrocodone and highly addictive [13]. However, one must recognize that the three groups of physicians behind this movement to treat chronic pain with unlimited narcotics are [1] physicians who are knowledgeable about management of cancer pain, for which longterm narcotic administration is appropriate, who are worried about or inappropriately fail to distinguish between cancer pain and chronic pain, [2] physicians who may present themselves as pain specialists but emphasize a monodisciplinary approach to chronic pain, and [3] physicians who are concerned that narcotic use may be legislated and want to keep patient care in the hands of physicians. Selected patient reports are presented by these groups as examples of successful long-term administration of narcotics being appropriate and benign, but the patients thus presented generally have very similar characteristics. First, they are well-motivated patients and take narcotics to allow them to increase their
Management of pain of benign versus cancer origin
level of activity, frequently to the point of working full-time or tending to a family. Second, they take low and appropriate doses of narcotics and may take the same low doses over a long period. Third, they do not escalate the dose even when they are under stress. Fourth, they are generally not depressed. Such patients do not really suffer from the psychological aspects of the chronic pain syndrome described above, even though they may have chronic pain. Not only do such patients do well on appropriately low closes of analgesics over long periods, but, if their pain were intractable enough to consider it, they would be ideal candidates for stimulation procedures. A common problem confronts the physician when a patient requests ever increasing doses of a narcotic based drug. Which patients should be given the increased doses and which should not? The suspicion of prescription narcotic abuse should be evaluated. Is the patient asking for and/or receiving the drug from more than one physician? If so, a single doctor should be identified as the only one to write prescriptions for pain meds for that patient. Most important, does the increasing use of the drug result in more or less regression? Does the medication make it possible for the patient to return to work, or at least self-sufficiency? If so, the patient is less disabled when taking the analgesics, in which case they are indicated. We have had patients who are active and doing well on longterm use of a narcotic. On the other hand, if the patient becomes increasingly disabled or regressed as the dose is increased, this is counter to the reason for giving the patient pain medication, so narcotics should be avoided.
130
or not it is related to a pain complaint), but the patient has a history of drug abuse or may be an alcoholic. Similarly, a patient with chronic pain who had previously been successfully withdrawn from narcotics may present with a new acute injury or need for surgery. Such patients are at risk of slipping hack into narcotic abuse. The best maneuver is to address the problem directly and in advance. A verbal contract can he made with the patient indicating how much narcotic is generally required for such procedures and for how long. Because there may be considerable residual tolerance, it may be necessary to provide the patient with a larger initial dose than would ordinarily be considered for such surgery. However, the duration the narcotics are administered should be no different than in the usual patients. Negotiate in advance when the high doses of narcotics will he discontinued, when the transition will be made to oral narcotics, and when narcotics will he completely stopped. For instance, we might negotiate that the patient will receive unlimited narcotics for the first 2 or 3 days after a procedure such as a laminectomy, often by a PCA pump to minimize the highs and lows. After the third day, the transition will be made to oral analgesics. No narcotic-based analgesics will he administered after the third week postoperatively, regardless of the circumstance. Because the protocol is negotiated in advance, the patient knows what to expect and realizes that he or she cannot manipulate hack into chronic narcotic administration. Patients generally are reasonable and compliant with such a contract when it is administered with understanding firmness.
Depression How are Narcotics Prescribed in a Patient with a History of Drug Abuse? It is not uncommon for the neurosurgeon to have a patient present for appropriate surgery (whether
Depression is common in patients with chronic pain and, indeed, is one of the hallmarks of the chronic pain syndrome. Recently, depression has been demonstrated to be due to a Mendelian dominant/recessive gene for a neurotransmitter
2205
2206
130
Management of pain of benign versus cancer origin
precursor. The gene is interactive with stress to cause clinical depression, not unlike type two diabetes where a genetic predisposition may make the disease become active after dietary indiscretion and weight gain. If a patient has one or two short alleles, he or she gets depressed under stress [6,14]. Three-fourths of the population have such a predisposition to depression, and its high incidence in any medical population and especially chronic pain is predictable. Well motivated patients especially become very depressed when unable to continue work. The depression is intensified with the chronic stress resulting from disruption of income and disruption of the family. The patient has unlimited time to ‘‘sit and stew’’ about his or her circumstances. Unfortunately, managed care, insurance, and disability hassles contribute significantly to the stress and depression associated with the disability of the chronic pain syndrome, and such frustrations make the pain worse. For the most part, socioeconomic developments in medicine have been designed for quick management of acute conditions but do not recognize chronic problems. The depression that accompanies the chronic pain state is often the direct result of a well-motivated patient being disabled and, in that regard, should ideally be considered as compensable as the physical injury itself and as demanding of management. Once this issue is settled, patients have less stress, depression, and pain, and the door to rehabilitation opens. One peculiar observation in patients with chronic pain is that a surprising number suffer from unresolved grief and its attendant depression. The grief may be preexisting and result from loss of a loved one, sometimes years before. The patient’s symptoms may in fact relate to the symptoms of the grieved for loved one or to the deceased’s final illness. Sometimes the patient may worry that he or she has the same potentially fatal illness and is not being told the truth about it. A common pattern, however, is that the patient is grieving for himself or herself, and
that grief is unresolved. The patient may have been vigorous and active prior to the illness or injury that leads to the chronic pain state. In fact, the patient’s self-perception may revolve around activities that are no longer possible. The patient may be a compulsive individual who has had no sedentary activities to fall back on, whose entire life involved work or vigorous physical activity. The patient grieves for the individual who can no longer participate in such activity, the same as grieving for any other lost loved one. Often a patient will insist on continued narcotic administration because it helps the patient to avoid facing the stresses in his or her life or the grief that needs to be addressed. Depression and unresolved grief often require professional counseling to improve defining the importance of these factors, which is an essential first step in successful management. Ordinarily, the same pattern of grieving occurs as in the loss of a family member. The initial stage of a normal grieving process is denial. The initial period of denial often coincides with the time that both the patient and the physician anticipate an unremarkable recovery and return to vigorous health. The next step of grieving is anger, which may or may not be expressed overtly but is common when the chronic pain is a result of an injury. If the injury occurred at work, the patient may be angry at the employer or the circumstance, and that anger may be magnified if the patient feels that she or he is not getting appropriate compensation and understanding for the work-related injury. If grief continues to be unresolved, the patient may remain in a state of depression. The combination of depression of the chronic pain syndrome along with additional depression from an unresolved grief reaction intensifies both the disability and the chronic pain state. For any given condition, patients who are depressed feel pain more severely than patients who are not depressed. Patients commonly recognize this when it is pointed out. Unfortunately, it is
Management of pain of benign versus cancer origin
often ignored by physicians who may hope for a miraculous cure from a procedure in a patient who is so depressed that any recuperation or rehabilitation takes more energy than the patient can muster. With depression comes a vicious cycle. The pain and disability cause depression, which causes heightened pain perception, which causes more pain and more depression, and so forth (> Figure 130-2). Successful management of depression helps to break up the cycle and lessen the pain. The most commonly used drugs in any chronic pain program are antidepressants [5,7,9]. There is some evidence that anti-depressants lessen perception of chronic pain, although it is not an analgesic as defined by response to a noxious stimulus. Many factors are involved with the perception of chronic pain, and antidepressants may work in a number of ways. Sleep disturbance is a common concomitant of depression associated with chronic pain and is often expressed by waking during the night. Patients may be able to go to sleep, perhaps with the help of analgesics or other medication. Then they wake at 2:00 or 3:00 in the morning and lie there in pain. Patients interpret this as ‘‘the pain woke me up.’’ Actually, the depression woke the patient, who wakes only to find that the pain was still there. With nothing to do but stare at the darkened ceiling and concentrate on the pain, pain perception becomes heightened. The 2:00 A.M. awakening may coincide with the timing of recurrent withdrawal, which also increases the pain during the night. . Figure 130-2
130
Discontinuation of narcotics helps to prevent the nighttime episode of recurrent withdrawal, and that, in turn, helps treat both the sleep disturbance and the depression. Antidepressant medications may help significantly in reestablishing a normal diurnal sleep cycle, which is an important factor in management of chronic pain. When normal sleep is resumed, the patient becomes less fatigued and less depressed. The reestablishment of a normal sleep pattern often coincides with the patient beginning to make rapid recovery and good clinical improvement. Antidepressants, typically tricyclics, are the pharmacological component for most chronic pain programs, and they should be administered as part of the preoperative evaluation when a chronic pain procedure is contemplated. In large studies all antidepressants work. The choice of drug is obvious if the patient has used one that has worked well before. If there is no past history of use, the key is finding one that is well tolerated with few side effects for a first time patient. Compliance is very low because the side effects are immediate and the intended antidepressant and antianxiety effects are often delayed 2–6 weeks. It is critical to always assess suicide risk and inform a depressed patient that this is a part of depression. Most physicians don’t realize the greatest risk for suicide in a patient in treatment is a few weeks after starting the medications. All drugs for depression treat the lethargy and sleep disturbance before the mood, causing some patients to have the energy to act on their hopelessness when they start to respond. This is why electroshock therapy (ECT) may be a lifesaver as it treats mood first. This was also the reason that Prozac was almost taken off the market; uninformed people interpreted the increase in suicide attempts after treatment as caused by the Prozac and not due to the sequence of response to it. Its worth noting that we found Amitriptyline (Elavil) in low, non- antidepressant doses
2207
2208
130
Management of pain of benign versus cancer origin
(50–75mg as opposed to the antidepressant dose of 100–150mg) were effective in several kinds of chronic pain, e.g., cancer, shingles, and trauma induced pain. It is often worth a clinical trial on its own after the chronic pain syndrome issues have been identified and addressed.
. Figure 130-3
Physical Regression Physical regression directly results sometimes from chronic pain and sometimes from inappropriate advice given by the physician. In acute pain, it is appropriate to withdraw from physical activity to rest the injured painful part. However, rest may be continued ad absurdium, so that many months after the acute episode the patient is still being advised to be less and less active, even though it is obvious that inactivity is not helping. Thus, another vicious cycle begins. The inactivity causes deterioration of muscle strength, shortening and stiffness of muscles, and increased pain, especially when the pain is of muscular origin to begin with. Any attempt at activity increases the pain, so the patient’s normal response is to decrease the activity even more. That inappropriate response is reinforced by inappropriate agreement by the physician. The muscles become weaker and more painful still, so that the activity decreases even more. Thus, the vicious cycle is established when pain causes inactivity, which causes muscle weakness, which causes more pain on activity, which causes more inactivity and more weakness and more pain, and so forth (> Figure 130-3). The basis of any rehabilitation program, whether it be pain rehabilitation or physical rehabilitation, is to gradually and progressively increase physical activity to the maximum that the patient can perform within mechanical limitations. Patients who are physically regressed beyond what their injury would dictate may have much less pain when physical remobilization is instituted. That should be considered
prior to surgical intervention. The neurosurgeon must also assess whether the lack of patient motivation to get better has contributed to the physical regression, in which case lack of motivation would also adversely affect the outcome of any surgery, so counseling about motivation and mobilization are required. It is tempting to fantasize that a surgical procedure will alleviate the patient’s pain so that the patient will get up off of the operative table and resume normal activity. However, patients who have lost their physical conditioning will require rehabilitation effort to regain it, even when pain symptoms abate. ‘‘If you don’t use it, you lose it’’ is axiomatic. The physiological exercise curve is the same for all but those who regress. Start very low on the curve. It takes time before improvement is obvious.
Psychological Regression Psychological regression is one of the most important aspects of chronic pain and is one of the most overlooked. Patients who have acute pain are appropriately assigned to a passive role, with the assumption that they will resume their adult responsibilities when their painful condition heals. However, most patients with chronic pain have not resumed an active adult role. As patients become less and less involved with activities and more involved with the pain itself, they become more and more severely regressed
Management of pain of benign versus cancer origin
psychologically. Although some patients do everything they can to become rehabilitated and return to normal life, others become more regressed and savor the role of patient. They adopt patienthood as a permanent lifestyle. These patients obtain considerable psychological reward from being taken care of, from being the center of attention, from becoming increasingly dependent, and from avoiding adult stresses and responsibility. Such psychological secondary gain rein-forces pain behavior even more strongly than financial reward, although many adopt patienthood as a new occupation. Sometimes this is reinforced by the caretaker. It is not uncommon for a caretaker spouse to relish increased importance in the family and increased independence, just as the patient is becoming more and more dependent. The physician must be aware that the spouse may sabotage efforts at pain management or rehabilitation to keep the patient dependent. It is especially important for patients who may relish too much the psychological regression and the escape from adult activity, since such patients already have a poor response to pain management and rehabilitation. Another indication of a poor prognosis is the patient who is unduly regressed because he or she has no option. The patient’s disability (and consequently income) may depend on the patient continuing to have pain. The patient may not be able to return to physically demanding work, even if her or his pain were to lessen or resolve. Consequently, the only practical option is to remain regressed and not rehabilitate, for fear of losing both employment opportunity and disability income. This presents a conflict between the patient’s income goal and the goal of pain management, which may make successful pain alleviation impossible. This situation is unfortunately programmed into many occupations. Some patients may have no skills except for heavy labor. When young and entering the job market, they are proud of their physical abilities and their ability to handle a
130
physically demanding job. However, the job is too demanding for middle-aged workers, who then have no option for promotion or transfer to a less physically demanding job. They are trapped into a job that is too harsh for their age. One day, they become injured on the job and become physically disabled with pain. As long as the pain continues they remain disabled from their occupation, and their income continues. However, if the pain were to go away, they would lose their disability income and have nothing to fall back on except a job they are no longer able to perform – they have no options. Medical disability seems to be the usual road to retirement in many such jobs, almost as if it were intended as a retirement plan. After all, how many middle aged laborers do you see at a work site?
Intolerance to Stress Intolerance to stress may begin as a manifestation of depression, physical regression and/or psychological regression, but soon assumes an identity of its own. Intolerance to stress may begin as a manifestation of depression, but soon assumes an identity of its own. Although intolerance to stress may affect patients with any type of chronic pain, it is a particularly important in patients with pain of myofascial origin, that is, pain originating from musculoskeletal injury, disease, or a malfunction in adjusting muscle tension. Stress causes an unconscious reflex that leads to increasing or tightening muscle tone, particularly neck or low back muscles, and may affect only a limited part of any given muscle. When the muscle tightening exceeds threshold, the muscle may go into spasm [11]. After injury, the threshold for muscle spasm may be much lower. Either physical or emotional stress may cause muscle tightness. The physical stress may occur when patients exceed his or her mechanical
2209
2210
130
Management of pain of benign versus cancer origin
capabilities, which makes muscles tighten and go into spasm. Emotional stress of any kind can increase muscle tone to the point where it increases the pain and, when severe, may actually cause the original muscle spasm. Thus, a vicious cycle is set up where a muscle injury may cause muscle tension, and muscle spasm in turn causes pain. The pain is a stressor that causes a reflex muscle tightening, possibly enough to provoke more muscle tension and possibly spasm. That causes more pain, which causes more stress, and more muscle tension and so forth (> Figure 130-4). Many things about the chronic pain state lead to stress. The pain itself is a stressor. The disability and attendant insurance and financial difficulties are stressful, even under the best of circumstances, and such stress has increased significantly in the age of managed care. The new role of the chronic pain patient as a dependent member of the family may be stressful to all, since the entire family must establish new relationships. One group of patients who are particularly prone to stress and to having myofascial pain, are men who brag that they have ‘‘worked hard all my life.’’ Their entire self-identity may be that of a compulsive, hard-working breadwinner, the respected head of the family. These men have few interests outside the workplace, and even their interest in the family may revolve around their being the provider. Such people find it . Figure 130-4
extremely stressful to become disabled and lose their identity. They have no experience with activities other than work and have nothing to fill their time except to mourn for lost physical capabilities and lost roles. Such individuals are motivated to the point where their motivation itself becomes detrimental, in that they set unrealistic goals for themselves, as they had all their life, so they are repeatedly frustrated. Indeed, the more motivated this patient is, the more stressful his plight may be, and the more resistant to any management program. It is only with considerable counseling and reestablishment of a more realistic revised identity and goals that any progress can be made. Note that successful management of this patient population involves repeated definition of barriers with the patient and realistic goals for improvement. Not only do these patients hold unrealistic goals for themselves, but they hold unrealistic expectations for their physicians as well. They may come into your office extolling your reputation and flattering you by telling you they know that you are the person who will make them better. They freely criticize their prior physician who failed to cure their pain. They will accept no less than total return to their prior physical fitness (which may have been years ago) as a satisfactory result, so they are invariably disappointed, which leads to additional frustration and stress and worsening of their symptoms. It is important to recognize such patients before embarking on a surgical management program, since surgery as a sole modality of treatment is usually doomed to failure, and the blame for such failure will be entirely cast upon you as the physician. Thus, we have identified two patients at the opposite ends of the spectrum who are equally inclined to fail surgical or invasive management programs. At one end of the scale we have the need-to-suffer patient who relishes the role of illness, pain, and disability and resists any attempt at changing that status. On the other
Management of pain of benign versus cancer origin
end of the spectrum, we have a need-to-work patient who is motivated to a fault and whose insistence on total cure precludes the possibility for a reasonable and realistic improvement. Both of these patients fail to set realistic goals for themselves and for the treating physicians. The need-to-suffer patient sets unstated goals that are unrealistically low—basically the goal is to remain a patient. The need-to-work patient sets goals that are unrealistically high and also are doomed to failure. Thus, it is important that both the patient and physician enter any treatment program with mutually accepted realistic goals. (> Table 130-1) Let us digress to define the term ‘‘myofascial’’ as used in this chapter. The word itself refers to muscle and associated ligaments. A muscle injury may cause pain that is particularly susceptible to the vicious cycles described herein. It can occur when the muscle is suddenly stressed (as in a fall) or stressed because of over-stretching. The most common site for myofascial pain is the low back followed closely by the neck. Low back myofascial pain may radiate down the ipsilateral proximal leg, but ordinarily does not take on a radicular distribution. Neck myofascial pain may be referred to the head, particularly the frontal or temporal area, and present as a headache. The diagnosis may be made by deep palpation of the painful area to identify a trigger point. On deep palpation of the back, the patient may suddenly report, ‘‘That’s it’’. Deep palpation of the suboccipital muscles, particularly at their origin at the C2 lateral mass, may cause the headache to be suddenly more severe. Injection of the trigger point with local anesthetic may ease the pain significantly but temporarily. I (plg) have treated patients with severe localized myofascial pain with Botox with good relief that may last for weeks, and believe this should be evaluated further [1,4,15]. My (plg) hypothesis is that the pain generator is an area of injury, perhaps with fibrosis involving the muscle spindle rather than the muscle fibers, so the tension regulating system
130
becomes reset at a point where the local muscle remains in painful contraction.
The Chronic Pain Syndrome Much has been written about the chronic pain syndrome, but no accepted definition exists for it. In general, the chronic pain syndrome is considered to be a constellation of conditions relating to chronic pain, which make a simple treatment program unsuccessful. The key question is, ‘‘When I interview and examine a patient, how do I know whether he or she has a chronic pain syndrome?’’ For the sake of the discussion in this chapter, chronic pain syndrome is defined as a chronic pain condition that is complicated by several of or all the five factors enumerated above—narcotic dependence or inappropriate use of narcotics, depression, physical regression, psychological regression, and intolerance to stress—the same five factors that make pain worse and impede recovery. The majority of patients with chronic pain who are referred to the neurosurgeon for surgery usually have chronic pain syndrome or they would not have gotten that far in the referral pattern. Consequently it behooves the neurosurgeon who agrees to see such patients to have an appreciation for the chronic pain syndrome and the knowledge and skill to define how it affects a given patient as well as the ability to both evaluate and manage the chronic pain syndrome. When any of these above-mentioned five factors is present, it significantly influences perception of pain, that is, it makes pain worse. Any of these five factors significantly affects patient evaluation and decisions concerning patient management, particularly if invasive procedures are contemplated. When any of these five factors is present, successful management of these factors is a significant part of the treatment program, regardless of whatever other modality is employed.
2211
2212
130
Management of pain of benign versus cancer origin
The five components of the chronic pain syndrome should be addressed prior to embarking on a treatment program, or as an initial part of that program. If the patient presents as a candidate for a spinal cord stimulator, for instance, it is difficult to evaluate the patient until chronic pain syndrome factors are under control. Certainly the evaluation of the effectiveness of the stimulation during the trial period is more accurate if the above factors have already been managed. Following implantation of the spinal cord stimulator, one could anticipate a greater likelihood of successful pain management if the patient no longer has uncontrolled chronic pain syndrome. It is particularly advantageous that depression and narcotic dependence he under control prior to any invasive procedure, since both interfere significantly with the likelihood of success. It is important to relate to the patient that consideration of the psychological factors is not a suggestion that ‘‘the pain is all in your head.’’ In fact, most patients have already been told that by frustrated referring physicians and resent it, since they recognize that the pain is in their back and not in their head. Indeed, by far the majority of patients who present for chronic pain management have a physiological basis for the onset of their pain, but pain perception is magnified and prolonged significantly by psychological factors which impact adversely on any evaluation or management program. You should reassure the patient, ‘‘I do not believe the pain is in your head, but what is happening in your head may make the pain worse and interfere with success of any treatment program.’’
Evaluating the Chronic Pain Patient Evaluation of the chronic pain patient begins with the first view of the patient walking down the hall to the examination room, or even earlier
if the physician’s office allows it. The observation of patients when they do not realize they are being observed and the comparison of those observations with observations during direct examinations sometimes reveal great contrast which may imply that patient is magnifying his or her symptoms. Who should do the initial psychological interview? Certainly the initial psychological evaluation must be done by the neurosurgeon who will make the final decision about an invasive procedure. Consequently, any neurosurgeon taking on the responsibility for such procedures must also take on the responsibility for learning how to do a basic psychological interview with chronic pain patients, and obtaining a psychological consultation afterwards or referring the patient to a multimodality pain clinic. Many times a formal psychological evaluation in addition to the neurosurgeon’s interview is helpful, especially a consultation with a psychologist who is knowledgeable about chronic pain management (which excludes the majority of clinical psychologists). However, that is in addition to and not instead of the evaluation by the neurosurgeon. It is a grave mistake and can be extremely misleading to have an independent psychological evaluation by a psychologist who may not be conversant with problems of chronic pain patients. Additional problems arise when the neurosurgeon also has not done a psychological interview. It is important that there be good communication between the psychologist and the neurosurgeon. Any neurosurgeon seeing chronic pain patients should develop a longterm relationship with an individual clinical psychologist knowledgeable about this field. There is no psychometric written test that takes the place of a personal interview. There is no written test that can explore the various aspects of the chronic pain syndrome described above or can evaluate the personal and family psychodynamics that may contribute to chronic pain disability. The Minnesota
Management of pain of benign versus cancer origin
Multiphasic Personality Inventory (MMPI) detects depression and somatization, which are present in most chronically ill patients with pain and disability, but does not give a clue as to appropriate management or psychodynamics. It is usual in chronic pain patients to see elevations in both of those scales. That does not indicate a preexisting personality defect. It may be the result of rather than the cause of the chronic pain syndrome, so is not specifically related to prognosis. However, need-to-suffer patients may have very little overt depression, since they are satisfied with their disabled state. The absence of depression when it is appropriate is a sign of poor prognosis. However, need-to-suffer patients who are particularly difficult to manage are best identified by history. All will have histories of childhood abuse, poor and often abusive marriages, many and unsuccessful medical interventions and an urgent need for you to treat them and make it all better. Redefining their goal from treatment by the doctor to management by the patient to minimize social dysfunction will be unacceptable to them. It is often helpful to engage a physiatrist (rehabilitation specialist) in patient evaluation as well. A knowledgeable physiatrist who has a great deal of experience with chronic pain patients may help with the evaluation of the patient’s motivation and whether the patient might successfully engage in a rehabilitation program, which is, after all, the bottom line. A physiatrist also may be helpful in evaluating the mechanics of the patient’s problem, how those impaired mechanics might contribute to the painful condition, and how they might interfere with the patient’s rehabilitation potential. They also know that the realistic goal for this patient group is REHABILITATION, not cure, and they can reinforce this approach. Approximately 20% of patients who were referred to our Chronic Pain Clinic were those we refer to as the ‘‘need to suffer’’ patients.
130
In reviewing their histories, they have had a lifelong history of impaired relationships, impaired health, major and minor accidents, andmarriages to abusive or addictive spouses. Their pain is far greater than the observable etiology can explain. Estimates are that 20% of chronic pain patients are responsible for 80% of chronic pain visits, and our impression is that many or most of those patients fall into the need to suffer category. Treatment of such patients will invariably fail, since their need is not for relief but for continued pain and disability with the attendant sympathy and attention. Often, the best the physician can do is to keep such patients off narcotics and out of the operating room. This patient group frequently has a past history of significant childhood sexual, emotional, or physical abuse, which leads to translating all stress into physical symptoms with the high risk of excessive medical treatments. Treatment of such patients will invariably fail, as the pain is a stress symptom and will not respond to medical treatment. Management of this patient group is most effective if the patient is told there is no medical cure by procedure or medication and they must carry on for the benefit of their family, children, etc. Often antidepressants will be of some help, but the major goal is to protect the patient from other physicians, who will willingly resort to invasive procedures.
The Interview The key to evaluation of the chronic pain patient is the interview, sometimes even more than the physical examination. Not only should the pain be discussed, but the patient should be asked directly about things related to the chronic pain syndrome. Throughout the interview the patient should be observed for signs of inappropriate or exaggerated pain behavior. The physician should review the medical records and diagnostic studies personally. Usually
2213
2214
130
Management of pain of benign versus cancer origin
by the time the patient gets to the pain specialist the problem is not one of a misdiagnosis but one of overtreatment. Do not accept the label applied by prior physicians. Assigning a label is required in present day practice. Unfortunately, once the wrong label is assigned, the patient carries it forever. Nevertheless, review of the films will occasionally demonstrate something that was previously missed. In our experience this patient group had a high incidence of undiagnosed medical problems unrelated to the pain complaint. Insistence on pain relief had led to an exclusive focus that neglected obvious other medical issues. Review of the medical records does not take the place of obtaining a detailed history directly from the patient. A good opening statement is, ‘‘Tell me about your problem.’’ This leaves the door open for the patient to start the interview without being led. It is not uncommon for the initial comments, or those offered soon after, to concern something other than the pain itself, which may give a great clue into the dynamics of the patient’s problem. The patient might begin with a comment about compensation or disability, rather than the pain, which may indicate what the patient considers most important about the visit. The initial comment may concern one of the aspects of the chronic pain syndrome, such as depression or narcotic-seeking behavior. The initial comment may concern the spouse or family. The patient may begin the interview by criticizing prior doctors. The patient may reveal in the initial comments what she or he thinks is the reason for referral, which may include unrealistic expectations for a surgical cure; the patient may believe that he or she is being referred for yet another laminectomy rather than a chronic pain program or for larger doses of narcotic. How does the neurosurgeon recognize the depressed patient during the interview? Look at the patient. It is amazing how many physicians record the history while looking only at the desk or chart, thereby losing important information
from the evaluation. Is the patient sitting alert and erect or slouching and haggard looking? Does the patient become tearful in discussing the pain, the disabilities, the effect the pain has had on the family, or the dire finances or other stresses? Does the patient speak in helpless and hopeless terms? Is the patient clutching a tissue? (We have found this last observation an excellent indicator of a severe chronic pain state and generally a poor prognosis.) Ask Patients to Describe Their Pain. It is amazing how many patients arrive in the office of the neurosurgical consultant without any of the prior physicians asking them to describe their pain. The patient uses the word pain, and the physician automatically attaches his or her own definition, which may be different than the patient’s definition. In describing the pain, the patient should be specific. Exactly where is the pain? Is the pain constant or intermittent? What makes the pain worse? What makes the pain better? What is the character of the pain? What does the pain feel ‘‘like’’? Is there more than one pain, and, if so, which pain is disabling? Patients and physicians may become overly concerned with putting a label on the diagnosis. When the patients give their history, they may relate what they have been told by prior physicians, rather than describing what they feel. This can lead to erroneous conclusions, especially since prior diagnoses may not have been accurate or patients may be conveying information that they misunderstand. Rather than being trapped into a misperception of what patients have been told previously, the interviewer should insist that patients describe their symptoms rather than diagnostic conclusions. The new examiner may reach quite a different diagnosis than had been reached previously. Ask how long the patient has had pain. When did it begin? How did it begin? Did it begin with an incident or episode or did it begin spontaneously? Did it begin suddenly or did it begin gradually? What is the mechanism of the event
Management of pain of benign versus cancer origin
that started the pain, and was the injury and subsequent pain appropriate to the accident? Ask about what the pain has been ‘‘like’’. Did it come on and stay? Did it get gradually and progressively worse (a significant indication of the chronic pain syndrome)? Did the pain get better and then worse again? (Is there a new factor concerning the pain that may not have been evaluated?) Was the onset of pain delayed? It is typical for pain of a myofascial syndrome to begin not at the time of injury but some time over the next 24 or 48 h. For a classic example, a male patient injures himself at work, recognizes that he has ‘‘pulled his back muscle’’, but is able to (or required to) continue to work for the rest of the day. That evening he may feel somewhat stiff but the next morning may be unable to get out of bed to return to work. This is physiologically consistent with a myofascial or muscle injury. However, that pattern is sometimes used by the employer or insurance company to suggest that the patient does not have ‘‘real’’ pain related to the injury, since it did not occur immediately. If the patient had prior surgery, one must ask what was the pain like before the surgery and what was it like afterward. Is it the same pain? It is not uncommon for a patient who has a laminectomy for a radiculopathy to have back pain afterward, but he or she is only asked if they have pain, not about the pain prior to surgery. The pain may be very different. This presents several important implications. If the radicular pain is considerably less after surgery, one may be dealing with a healing neuropathy which may ease in time. If the pre-operative pain is unchanged or worse, however, one must consider whether the surgery was appropriate or inadequate and whether there still might be primary surgery required rather than a pain program or procedure. If the patient has back pain rather than continued radicular pain, this must be considered as a separate and new diagnosis and require a rehabilitation program.
130
Patients commonly develop myofascial pain as the result of surgical exposure, even though the original radiculopathy pain may have been alleviated. Fortunately, this is less common since limited exposure laminectomies have become widely used. Unfortunately, this distinction is sometimes missed in asking patients about pain postoperatively. Patients may say that they still have pain, without mentioning that it is a different pain. The surgeon may think that the laminectomy was unsuccessful, even to the point of subjecting the patient to yet another laminectomy. However, if the pain is myofascial pain at the site of the incision, further laminectomy is contraindicated, and the program should change from one for a radiculopathy to one for a local back injury. It is amazing how frequently this distinction is missed, even by the experts. I (plg) have seen research protocols about laminectomy results, wherein the distinction between back pain and radicular pain (both pre-operative and post-operative) is never addressed. The ‘‘failed back syndrome’’ is that the patient ‘‘still has pain after back surgery’’. Ask about prior treatments and the patient’s responses to each of those treatments. If a prior treatment showed some encouraging results, it may be premature to consider surgery. It is a bad prognostic sign if the description of prior treatments is too detailed or too dramatic, and especially if the patient has seen a large number of doctors for unsuccessful treatment in the past. It is a dangerous sign for the patient to relate how one treatment after another just made the pain worse and worse. This may reveal that the patient has unrealistic goals or may suggest that the patient’s motivation is to remain in pain and disabled, or it may be manipulation to obtain more narcotics. This is often accompanied by flattery that you have been recommended as the best doctor in the world and the patient knows you will find the cause of the pain and you will save the patient from a lifetime of pain. Beware, because the patient will speak just as unkindly
2215
2216
130
Management of pain of benign versus cancer origin
about you to the doctors he or she sees subsequently since the pain is not curable. If the patient has switched doctors, ask why the patient changed in his or her own words, and not the words of the referring physician. Ask very specific questions about the medications the patient is taking. These questions include not only the names of the medication and how much was prescribed but also the amounts that were actually taken. Have the doses of narcotics gradually increased, have they decreased, remained stable? What is the patient’s attitude toward taking narcotics? Does the patient seek more medication than the prior physician is willing to give, or does the patient recognize that narcotics are not the answer? In asking patients about medications, it is important to ask two separate questions, [1] what has been prescribed and [2] what is the patient actually taking. If one asks the patient how much narcotic has been prescribed, the answer may be ‘‘one every 6 hours.’’ However, if the physician asks how many tablets you actually take per day, the answer may be ‘‘8’’ or ‘‘10.’’ Another way of eliciting this information is asking how many days a refill prescription of 100 tablets (or 10 or 50) may last. Occasionally patients will report that they are not actually taking the medication as prescribed. This is a serious consideration for antidepressants, which will be helpful only if taken regularly, and for narcotics because of the development of tolerance and the risk of addiction. The interview should be combined with the initial efforts at educating the patient about the chronic pain syndrome and why it is important to manage the various aspects of that condition. The initial interview is a good time to discuss that narcotics are designed for acute pain and not chronic pain and that taking narcotics over a long period in escalating doses is inappropriate, especially if they do not improve the patient’s depression and regression. This is also a good time to point out to the patient that if drugs
were the answer, they would have controlled the pain long ago and the patient would not be presenting to the neurosurgeon. It is amazing how many patients are under the impression that they are referred to a chronic pain clinic just to receive more pain medicine. This view may be the result of the frustration that led to the initial referral. It is unfortunately common that the initial treating physician gives the patient more and more narcotic until things have gotten out of hand. At that point the patient is demanding more and more narcotics and the referring physician finally realizes that the demand for medication is excessive, especially when the patient insists on phoning nights and weekends asking for more drugs. In that case, the reason for referral to a pain management program may not he an altruistic and sincere desire to help the patient but an attempt to pass the patient’s drug-seeking behavior to another physician. Ask the patient if there have been any other crises in the family. It is not uncommon for a patient to be referred because the pain has become more severe. In talking to the patient, however, it may turn out that a parent or loved one has just died, and that is the basis for this setback, rather than a change in physical status. Management of grieving may provide considerable improvement in pain. The caretaker may no longer be able to provide the care the patient wants or needs. Ask how the pain interferes with activity. The patient should describe the activities on a typical day. If the patient has considerable physical regression and spends much of the day sitting down or lying down, what occupies the patient’s mind during that time? Does the patient lie staring at the ceiling thinking about the pain or is the patient engaged in sedentary activities or hobbies that would be appropriate for his or her level of physical activity? This is the chance for the neurosurgeon to evaluate how much physical regression the patient has and how much the pain may
Management of pain of benign versus cancer origin
interfere with physical activity. However, it is also an important time for the physician to evaluate how much psychological regression and depression there may be and whether either type of regression is out of proportion to the physical abnormality itself. This is also an opportunity to ask whether the issues of the initial injury have been resolved. If the patient is still in the midst of a disability hearing, especially if the patient has a great deal of hostility toward the employer, it may not be the opportune time for successful pain management. Indeed, establishing the patient’s disability status and treatment of pain may represent conflicting interests. For the patient to get the optimal disability determination there should be maximum pain, maximum regression (especially physically), maximal administration of narcotics, and severe depression. If these factors are successfully managed, the disability will be determined to be less severe. Both the patient and her or his representative recognize that and thus may be motivated not to give up the severe pain prior to the conclusion of the hearings. It may be better to wait until legal issues are settled before starting a pain management program. Ask who is in the household with the patient. Who does the household chores? How much does the patient participate in that? Is the patient encouraged to participate in household activities or encouraged to lie passively and be taken care of ? The family history may be revealing. It is not uncommon for chronic pain patients to have grown up in a household where a parent was also disabled. If so, what was the relationship of the parent to the patient? Another peculiar pattern in chronic pain patients is that many grew up in a household where they were given a great deal of responsibility very early. They may have only gotten attention or been allowed to play only when they were sick, which may have led to a lifelong pattern
130
of being sick or regressed to obtain love and attention. What was the patient’s vocational history prior to the problem that led to the chronic pain? The need-to-suffer patient may have a history of numerous prior injuries at work. There may be a history of frequent job changes or frequently getting fired, since such individuals tend to create problems just as things are going well. Ask about the patient’s marital history. Again, the need-to-suffer patients may have a chaotic family history as well as erratic vocational and medical histories. Were spouses abusive or alcoholic (typical for a need-to-suffer patient)? Did children have chaotic lifestyles? Did siblings also have chronic pain or other chronic problems? Ask if the patient had psychological trauma as a child, especially if there was any history of sexual abuse. There is a significant incidence of early sexual abuse in need-to-suffer chronic pain patients. Ask the patient about short-term and longterm goals. If the patient does not have goals, there can be no means for establishing progress. If the, goals are unrealistic, both the patient and physician become frustrated, and the pain may become worse. It is common for chronic pain patients to be devoid of goals. They have spent so much time thinking of the pain that they may never have considered what they might do if the pain were less. It is important for both the patient and the physician to set realistic goals at the beginning of the treatment program. Finally, the physician should assess unrelated symptoms. It is common for a patient to be referred for chronic pain and turn out to have an unrelated significant physical or medical problem that has been overlooked because all the attention has been directed to the pain itself. This is a particular problem with patients who may have a work-related disability and have lost their general medical insurance, so that the only coverage they have relates to their injury-related disability.
2217
2218
130
Management of pain of benign versus cancer origin
Physical Examination of the Chronic Pain Patient Physical examination should focus on the pain complaint but include a general assessment. Much of the physical examination involves the observation of the patient during the interview and during the time the patient is moving about the office and waiting room. Note whether there is a discrepancy between the description of the patient’s disability and the ability to move around the office. Patients who indicate that they are unable to rise from their beds or who grimace and groan when rising from their chairs during the examination but then bend over freely to tie their shoe laces at the conclusion of their examinations may not be good candidates for surgery or any other program. They should be confronted with this observation in a non-threatening manner. If the patient describes total lack of physical activity and exercise, is there a lack of muscle mass and muscle tone consistent with that? Patients who are totally disabled physically develop characteristic muscle atrophy and tightening. Does the patient report a total lack of physical activity but have grimy fingernails and calluses that belie the accuracy of the description of activity? Physical examination of patients with myofascial pain is particularly important. Are there tight and tender muscles consistent with the patient’s complaint? Is there a specific trigger point of tender muscles that corresponds to the area of initial injury? The most common place for such trigger points is at the iliac crest, and sometimes repeated local anesthetic and steroid injection into those trigger points may provide considerable relief [18]. In general, much of the physical examination involves detective work as well as diagnostic work. It is very difficult to obtain an accurate impression of the chronic pain patient on just a single office interview. The physician is totally
dependent on what the patient says, since pain defies direct measurement.
The Management Plan While the interview and assessment are still fresh in mind, consider what factors cause pain primarily and what factors may make the pain worse. In general, there are three contributions to clinical pain. The physical component is that aspect of the pain that can be directly attributed to body tissue disease or injury. The psychological aspects concern the chronic pain syndrome and what makes pain worse. The pharmacologic aspect may involve inappropriate use of long-term narcotics, with addiction, recurrent withdrawal, and depression. Any plan for management involves consideration of all three components. As a general rule, whatever plan is embarked upon will be more successful if the psychological components are evaluated and managed first. Although many surgeons feel that a patient can deal much better with the psychological issues after the physical factors are taken care of (if ever), the chance of improvement of the pain by treating just the physical factors is much less in the presence of the chronic pain syndrome, so the patient may never get to the point where there is sufficient improvement to embark on the psychological and rehabilitation program secondarily. However, management of the psychological factors first has several benefits. There is often sufficient improvement so that the invasive procedure need no longer be considered. During the management of the psychological factors the patient may set realistic goals which allow better assessment of improvement from the pain procedure. It was not uncommon as we got involved with the patient on a day-to-day or week-to-week basis during management of psychological factors that we realized that the initial evaluation was too optimistic and inappropriate, so that plans for an
Management of pain of benign versus cancer origin
invasive procedure were abandoned. The referral to a neurosurgeon may have been for an intended procedure, but the surgeon must take the responsibility to make the final decision, as for any neurosurgical procedure, not the referring doctor. The same holds true for dealing with pharmacologic aspects of the patient’s problem. It is tempting to think that the patient will awaken from the neurosurgical procedure and have so little pain that narcotics are no longer required so they might be easily discontinued. This is rarely the case. If the patient is taking habituating doses of narcotics prior to the procedure, they will continue to require narcotics to prevent recurrent withdrawal afterwards. Also, the drive to prevent withdrawal symptoms may reinforce the perception of pain after the procedure so the patient is far more likely to continue to have pain. These issues are managed better postoperatively if the groundwork to stop narcotics has been laid prior to surgery. The surgeon must be part of an overall chronic pain management program in order to optimize the results of pain surgery. The emphasis for chronic pain should be on management rather than cure. The goal is to decrease pain to a level where the patient may participate in a rehabilitation program more successfully and more meaningfully. It is unusual for a pain procedure to stop pain completely, but it is not uncommon for the chronic pain to be modified sufficiently that the patient has considerable benefit and less disability. To make a valid decision concerning surgery, nonsurgical options must be available, either through appropriate referral or preferably with the surgeon acting as a member of a multidisciplinary team. The treatment the patient ends up getting is more often determined by the specialty of the physician to whom the patient is referred than whether the patient should have any procedure at all and, if so, which procedure. Pain clinics run by anesthesiologists tend to concentrate on blocks or drugs; pain clinics run by oncologists
130
or internists are more likely to center on pain medications; pain clinics run by neurosurgeons are more likely to advise a procedure, and it is not uncommon for every patient to have the same procedure. When treatment fails, the patient may be referred to a pain clinic run by a different specialty, so the approach will be vastly different – surely there is little consensus among various types of pain doctors about how to evaluate and manage chronic pain.
Evaluation of the Cancer Pain Patient The initial interview of the cancer pain patient is similar to the initial interview of chronic pain patients in many respects, and the following questions should be asked: Does the patient have symptoms of depression, such as a feeling of helplessness and hopelessness and waking in the middle of the night? Does the patient appear to be depressed, or does the family report withdrawn behavior that suggests depression? Depression heightens pain perception and makes pain worse. Patients with cancer often are depressed not only from the contemplation their own mortality but as a side effect to their treatment. They are often fatigued from the disease process itself, which predisposes to depression, and additionally fatigued by the medications or other treatment. How does the patient describe the pain? The same questions should be asked about cancer pain as for chronic pain. The distribution of the pain is particularly important, since patients with cancer pain may have a number of separate pains, only one of which requires intervention. The patient may report generalized pain. If one is considering an ablative procedure, it is important to know what part of the body is affected by the pain that requires treatment, as well as what pains there may be in other parts of the body. A curious observation is that if the patient has pain that is very severe on one side of the body and
2219
2220
130
Management of pain of benign versus cancer origin
mild on the other side, the mild pain may become much more severe if the severe pain is successfully treated, suggesting that the severe pain inhibits perception of the lesser pain. What does the pain feel like? The answer to this question may provide a clue as to whether the pain is due to involvement of somatic structures by neoplasm (which may respond to denervation procedures) or whether nerve tissue is involved (which may be more difficult to treat) or viscera (which may not respond to interruption of somatic pain pathways). What makes the pain worse? This global question can evoke description of many factors of the chronic pain syndrome and may give some insight into the mechanism of the pain. Did the pain become more severe as the cancer spread or did the pain become more severe as the patient became more depressed or developed other psychological factors? If the latter, it is the psychological factor that must be managed initially. What makes the pain better? Again, the surgeon is looking for clues for potential management strategies that might not involve surgery. How has the pain progressed? What were the responses to prior treatments? If the patient reports frustration with failure of prior treatments, the resultant anxiety and depression must be dealt with as a separate issue. Finally, have all other means of pain management been exhausted before ablative procedures are considered? If the only thing that occupies the patient’s mind is the pain itself, the pain perception is intensified. The patient should be encouraged to be as active and involved as the physical condition would permit. It is not uncommon for the family to respond to the diagnosis of cancer by hovering about the patient, promoting the patient’s dependence, and depriving the patient of any opportunity to be active or self-sufficient. Although such an approach is laudably sympathetic, it is counterproductive in that it may
make the patient more regressed and more disabled than is necessary from the physical condition.
Comparison Between Management of Chronic Pain and Cancer Pain The role for surgery in the management of chronic pain is limited. [3,4] Nonsurgical management of chronic pain takes priority, and surgery should be considered as only a concluding part of that comprehensive program in carefully selected patients. As a general rule, stimulation procedures are mostly indicated for chronic pain, whereas ablative procedures may be more appropriate for cancer pain. Procedures are generally indicated only for those problems which have readily identified etiology, with pain corresponding to anatomical involvement of the disease or injured part. Procedures for ‘‘atypical’’ pain or ill-defined pain rarely succeed. It is very tempting to consider pain surgery when the etiology of the pain is obscure, but that is the very time that pain surgery should be avoided. If there is no procedure with a reasonable chance of success, the correct decision is not to do surgery. Management of chronic pain can be frustrating, and chronic pain patients can be very trying. ‘‘I have got to do something’’ is NOT an indication for surgery. The something that is done under those circumstances is almost always the wrong thing. There has been a recent swing toward invasive procedures for management of chronic pain. Very often this takes the form of repeated blocks, sometimes with neurolytic substances. There is little evidence that such an approach is helpful on a long-term basis, even though such a simplistic procedure has great appeal to third-party payers and referring physicians. On the other hand, it may be useful to obtain such temporary
Management of pain of benign versus cancer origin
relief as part of a comprehensive chronic pain management program. The general approach to the patient with chronic pain is conservative with the emphasis being on management of the factors that make the pain worse rather than aggressively trying to influence the etiology of the pain. Indeed, most patients have already had extensive attempts at influencing the underlying cause of pain prior to being referred to the neurosurgeon. Management emphasizes rehabilitation, avoidance of narcotics, and return to activity with minimizing disability. Some of these vicious cycles influence cancer pain, such as depression, but many do not. Things that make chronic pain worse may also make cancer pain worse, so the approach to cancer pain should include the identification and control of those same factors that constitute the chronic pain syndrome. Attempts to treat the underlying etiology in cancer patients may be unsuccessful because of the very nature of the disease. As the cancer progresses it may involve more and more tissues in a pain-producing manner. Consequently, by definition, the management of the underlying etiology is the management of the cancer itself, and the nature of the disease is such that the disease often outruns the oncologist’s ability to keep it under control. Thus, cancer pain patients who are referred to neurosurgeons for pain control may still require management of the underlying etiology, but it is the referring physician rather than the neurosurgeon who more often attempts to deal with that aspect of the program. It is important to recognize that there are a number of different specific mechanisms that may cause cancer pain [5]. As cancer invades tissue it may cause pain. Treatment of the cancer may cause pain, particularly as it affects the nervous system. Thus, local radiation may lead to a painful complication, but that is a late effect that generally occurs 8–12 months afterward. Modern intensity modulated radiation makes such late
130
complications far less likely. Surgery may leave residual pain. Such pain must be dealt with just as if it is caused by the cancer itself. Most cancer pain can be managed with a combination of oral analgesics and attention to chronic pain syndrome factors. In those patients who have pain despite narcotics, a more aggressive approach is taken to management of cancer pain than with chronic pain, first by adjusting the dose of narcotics and sometimes by interrupting primary nervous pathways concerned with pain perception or suffering. Ablative procedures are indicated much more commonly for cancer pain than for chronic pain for several reasons. First, the localization of the pain may lend itself more favorably to interrupting pain pathways to that specific part of the body, such as with cordotomy. Second, the ability of the nervous system to develop alternate pathways for pain perception is less effective, since the prognosis may not allow the patient to live long enough for such reorganization to occur. Third, the goal is for comfort and pain management, usually without regard to resumption of prior physical capabilities, since that may already have been rendered impossible because of the disease process itself. For the same reason, narcotics are used much more freely in cancer pain than they would be for chronic pain. Tolerance is less of an issue, since life expectancy is most often limited. Addiction is not an issue, since usually the patient will never get to the point where withdrawal becomes either desirable or necessary. Interference with normal mentation is less of an issue, since the patient will remain a patient and not be required to function independently. The soporific effects of narcotics may be a benefit rather than a problem, since calming sedation is often desirable even if it would interfere with the normal intellectual or physical capabilities. Because of these considerations, the analgesic dose must be readjusted as tolerance develops,
2221
2222
130
Management of pain of benign versus cancer origin
with only secondary concern that narcotics may interfere with mentation. Eventual withdrawal of narcotics may never be needed. The protocol is to give the lowest does that does the job and to escalate narcotics successively to achieve and maintain adequate pain relief as long as possible. A program of progressive dose escalation is employed. When pain first develops, non narcotic analgesics are indicated (but this frequently occurs before the patient would be referred for pain management). When the pain progresses to the point where stronger analgesics are indicated, narcotics is given in adequate dose. A good balance must he achieved between giving adequate doses for satisfactory pain relief and not giving any more than is necessary, since it is desirable not to sedate the patient more than is necessary (at least not with narcotics), not to induce depression, and not to use up the narcotic tolerance any sooner than required, since the dose of narcotics must be readjusted upward as the disease progresses and tolerance grows. The narcotic dose can be minimized, however, by attending to those factors that may make the pain worse. Providing long-acting agents to prevent highs and lows will often prolong the time that oral narcotics are successful. The physician should be aware of the duration of the action of the narcotic. It is a common mistake to order the dose of narcotic too infrequently, in hopes of minimizing the daily dose. This results in the pattern that the patient alternates between too much and too little narcotic effects. The patient continues to have pain intermittently, and the result is to escalate the total daily dose of narcotic even more rapidly than would be necessary. It is better to give multiple smaller doses than large doses that are given too infrequently, and it is easier for patients to adjust their narcotic medications if they are administered every 3 or 4 h rather than 6 or 8 h. It may be desirable for the patient to take narcotics at different dose
schedules during the day and night, depending on the individual requirements. When the patient no longer responds to normal short-acting narcotics, a change to long-acting doses may be very helpful. This allows smooth escalation of the dose and helps the patient sleep uninterrupted during the night. A combination of short-acting agents during the day and a long-acting agent at bedtime may be most desirable. Eventually, however, the pain may become more severe than can be managed with usual oral narcotics. The general philosophy of administering narcotics in patients with cancer is to continue to give narcotics in whatever dose is necessary to provide pain relief, with other aspects of pain management being attended to. The limiting factor is one of side effects. Some patients do not tolerate the sedative effects of narcotics even at relatively small doses, so higher doses cannot be used. Other patients may have pain eventually become so severe that the dose of narcotic required for pain management is a hypnotic dose, so that the patient is drowsy or sleeping before the dose provides adequate analgesia. As an alternative to oral narcotics, skin patches that administer a slow dose of narcotic (usually fentanyl) can provide high analgesic doses at a constant rate. When the patient no longer responds to oral or transdermal narcotics, parenteral narcotics would be the next line of choice. Such narcotics might be administered in several different ways. If the patient’s prognosis is fairly short and if the patient has sufficient analgesia without undue sedation, intramuscular or intravenous narcotic may provide good pain relief. At one point it becomes more practical to administer the medication intravenously than to give repeated intramuscular doses throughout the day and night, for which an indwelling intravenous catheter can be employed. The patient or family can provide intravenous injection every few hours if that is
Management of pain of benign versus cancer origin
adequate. An infusion pump can be used to provide a continuous dose of narcotic. If the patient is bedridden, a bedside infusion pump can be used. If the patient is ambulatory, small battery-operated pumps are available that the patient can wear to provide a continuous dose of intravenous narcotics throughout the day and night, even when the patient is active. When it is no longer possible to obtain adequate pain relief with acceptable sedation with oral or intravenous narcotics, alternate means of narcotic administration may be sought. If the patient is preterminal and bedridden, the best and most humanitarian approach may be to provide the narcotics intravenously in whatever dose is necessary, even if the patient may become somnolent. However, if the patient still has the opportunity to spend time with the family and would have a reasonable quality of life if the pain were controlled, intraspinal morphine may be considered. Preservative-free morphine may be administered at very small doses through an indwelling spinal catheter. The total daily dose when administered epidurally is approximately one-tenth of the dose that is required parenterally for adequate pain relief. An epidural catheter may not remain patent for too long, so a subarachnoid catheter may remain more effective for a longer period. The subarachnoid dose is approximately one-tenth of the epidural dose, or one-hundredth the daily parenteral dose. The prognosis for life and the available resources may dictate the means of administering the intraspinal narcotic. If the prognosis is greater than 3 months, an implanted morphine pump may be costeffective, particularly if it allows the patient to return to activity or enables the patient to remain out of the hospital or hospice. If the prognosis is less than 3 months or if the socioeconomic opportunities do not allow a totally implanted pump, the spinal catheter may emerge through the skin for direct infusion. The infusion may be
130
made twice a day by nurse or family member, if good safety can be maintained. A micro filter can be used to minimize the risk of contamination. A dose of 1–4 mg morphine administered once or twice a day may provide dramatic pain relief, although tolerance may develop, and this dose may require escalation. A continuous infusion pump can be used to provide approximately the same dose, which must be adjusted to the individual requirement of the patient. If the patient does not respond to narcotics administered orally, parenterally, or intraspinally, or if the type or distribution of pain precludes the use of spinal narcotics, a neurosurgical pain procedure may be considered. Ordinarily, the first procedures considered for cancer pain are ablative, rather than implantation of a stimulator. As a general rule the ablative procedure should be made as low in the nervous system as possible. Consequently, cordotomy is often the procedure of choice. Because of the possibility of performing the cordotomy percutaneously, the preferable level may be at the upper or a lower cervical spinal level, since cordotomy at thoracic levels requires open surgery. Indeed, percutaneous cervical cordotomy has displaced open surgical cordotomy at most centers. Limited myelotomy may be effective in alleviating visceral pain of pelvic structures even without significant interruption of the decussating pain pathways or significant amount of analgesia by pin stick sensation [12]. If the pain involves the head and neck or is too extensive to consider cordotomy, a stereotactic procedure involving the brain may be in order. Such procedures should generally concern the portion of the pain perception pathways that project to the medial or limbic areas of the brain, as well as or instead of the spinothalamic pathways, in order to alleviate the suffering component of the pain as well as the somatic pain component, which is generally required for successful pain management. Thus, a
2223
2224
130
Management of pain of benign versus cancer origin
mesencephalotomy lesion is more successful if it encroaches on the central gray, even if the lateral spinothalamic tract is spared. A thalamotomy is more likely to be effective if it involves the basal or intralaminar area, even if the primary somatosensory pathways are left intact. For example, an intralaminar or basal thalamotomy, which involves the medial limbic area of the thalamus but spares the somatosensory fibers, may provide excellent relief of the pain of a Pancoast syndrome, since much of that pain is neuropathic rather than truly somatic in nature. Even though the approach to management of cancer pain is much more aggressive than chronic pain, such procedures are not indicated until all nonsurgical means of pain management have been exhausted. This means that the same factors that make chronic pain worse may also make cancer pain worse, and these factors should be managed along with the management of the pain per se.
Summary How does the neurosurgeon make decisions about pain management? [1] The type of pain must be established. Is it acute pain, chronic pain, or cancer pain? [2] The proportion of the components of chronic pain should be considered – physiological factors, psychological factors, and pharmacologic factors [3]. Those factors that make pain worse must be identified, and an evaluation and treatment protocol set up to address these factors, which include the five components of the chronic pain syndrome—narcotic dependence, depression, psychological regression, physical regression, and intolerance to stress [4]. The physician should select patients for pain management procedures only after the chronic pain syndrome has been evaluated and brought under optimal control [5]. Physicians should select only those patients who still are in need of a pain procedure after specific management of the
chronic pain syndrome has been completed. Patients who resist or reject management of the chronic pain syndrome or refuse management of psychological factors should be rejected for invasive procedures for pain treatment [6]. The least destructive and most peripheral procedures should be considered first. The emphasis in chronic pain is on nondestructive and stimulation procedure. The emphasis in cancer pain is first to devise the most efficient means for narcotic administration—oral narcotics, then parenteral narcotic administration, then means for intraspinal administration, and only after that considering ablative procedures [7]. Both the patient and physician must recognize that it is not possible to control all types of pain, and both must have a realistic appraisal of the possibility and realistic goals prior to embarking on a management program. The evaluation and management of pain require much time and patience and a natural empathy to evaluate psychological factors, which some neurosurgeons do not possess. Consequently, management of pain should be left to those neurosurgeons who have an interest in the field, who have the innate ability to evaluate all aspects of pain perception, who have appropriate consultation for the management of nonsurgical means of pain control or may be a member of a multidisciplinary team, and who have developed an armamentarium of many different techniques in pain management so the most appropriate one can be selected for the individual patient > Table 130-2.
. Table 130-2 Components of chronic pain syndrome Narcotic dependence or inappropriate use of narcotics Depression Physical regression Psychological regression Intolerance to stress
Management of pain of benign versus cancer origin
References 1. Barwood S, Baillieu C, Boyd R, Brereton K, Low J, Nattrass G, Graham HK. Analgesic effects of botulinum toxin A: a randomized, placebo-controlled clinical trial. Dev Med Child Neurol. 2000;42:116-121. 2. Carise D, Dugosh KL, McLellan AT, Camilleri A, Woody GE, Lynch KG. Prescription OxyContin abuse among patients entering addiction treatment. Am J Psychiatry. 2007;164:1750-1756. 3. Cherny NI, Coyle N, Foley KM. Suffering in the advanced cancer patient: a definition and taxonomy. J Palliat Care. 1994;10:57-70. 4. De Andres J, Cerda-Olmedo G, Valia JC, Monsalve V, Lopez A, Minguez A. Use of botulinum toxin in the treatment of chronic myofascial pain. Clin J Pain. 2003;19:269-275. 5. Doan BD, Wadden NP. Relationships between depressive symptoms and descriptions of chronic pain. Pain. 1989;36:75-84. 6. Eley TC, Sugden K, Corsico A, Gregory AM, Sham P, McGuffin P, Plomin R, Craig IW. Gene-environment interaction analysis of serotonin system markers with adolescent depression. Mol Psychiatry. 2004;9:908-915. 7. Feinmann C. Pain relief by antidepressants: A possible mode of action. Pain. 1985;23:1-8. 8. Foley KM, Payne RM. Current therapy of pain. TorontoPhiladelphia: Decker; 1989.
130
9. France RD, Houpt JL, Ellinwood EH. Therapeutic effects of antidepressants in chronic pain. Gen Hosp Psychiatry. 1984;6:55-63. 10. Gildenberg PL, DeVaul RA. The Chronic Pain Patient. Evaluation and Management. Basle: Karger; 1985. 11. Gildenberg PL, DeVaul RA. Medical management of chronic pain, In: Youmans JR, editor. Neurological Surgery. Philadelphia: Saunders; 1995. p. 5073-5089. 12. Gildenberg PL, Hirshberg RM. Limited myelotomy for the treatment of intractable cancer pain. J Neurol Neurosurg Psychiatry. 1984;47:94-96. 13. Gutstein HB, Akil H. Opioid analgesics, in Hardman JG, In: Limbird IE, editor. The Pharmacological Basis of Therapeutics. New York: McGraw-Hill; 2001;521-556. 14. Kendler KS, Kuhn JW, Vittum J, Prescott CA, Riley B. The interaction of stressful life events and a serotonin transporter polymorphism in the prediction of episodes of major depression: a replication. Arch Gen Psychiatry. 2005;62:529-535. 15. Lang AM. Botulinum toxin therapy for myofascial pain disorders. Curr Pain Headache Rep. 2002;6:355-360. 16. Moulton D. ‘‘Hillbilly heroin’’ arrives in Cape Breton. CMAJ. 2003;168:1172. 17. Savage SR. Opioid use in the management of chronic pain. Med Clin North Am. 1999;89:761-786. 18. Volinn E, Turczyn KM, and Loeser JD. Theories of back pain and health care utilization. Neurosurg Clin N Am. 1991;2:739-748.
2225
138 Mechanisms of Action of Spinal Cord Stimulation B. Linderoth . R. D. Foreman . B. A. Meyerson
Background Spinal cord stimulation (SCS) has since long been an effective mode of therapy for treating neuropathic pain of peripheral origin, peripheral ischemic pain states, vasospastic conditions and therapy resistant angina pectoris. Estimates indicate that presently more than 18,000 new systems for SCS are implanted annually worldwide. However, the mode of action for SCS is still poorly understood although in recent years some data on the underlying physiological mechanisms have emerged [1–4]. For the implementation of evidence-based and mechanismdirected therapies, which is to-date requested by the medical profession and health care providers, a thorough knowledge about the mode of action is required [5]. This is also a necessary condition for the further development of the techniques used in neuromodulation. Depending on the level of the neuroaxis where SCS is applied, the stimulation may also affect viscero-somatic reflexes that can modify a multitude of physiological functions. Possible effects of SCS on different organ systems when applied at various sites are illustrated in > Figure 138‐1 [6]. The mechanisms involved in the SCS effects may be quite different depending on the targeted organ; for example, the mode of action for producing pain relief differs fundamentally when SCS is applied in neuropathic and in ischemic pain conditions [3,4,7,8]. In this chapter, firstly, the physiological bases for the use of SCS in neuropathic pain and in ischemic extremity pain will be elucidated. Secondly, the putative mechanisms behind #
Springer-Verlag Berlin/Heidelberg 2009
the use of SCS in therapy resistant angina pectoris will be discussed; the possible application of SCS in various visceral pain conditions will also be briefly addressed. Numerous possibilities for explaining the modes of action of electric neuromodulation have been generated from clinical studies, but animal experiments are required for exploration of the underlying mechanisms [cf. 9]. Such investigations were extensively conducted in the 1970s and 1980s but the use of normal, intact and anesthetized animals, the application of only noxious and phasic peripheral stimuli and the application of SCS for short periods of time only (<1 min) and using parameters different from those applied clinically limited the clinical relevance of these studies. Even with these limitations studies conducted in normal animals have provided background data that are of value for interpretation of the results gathered from animal models of disease (e.g., [10]). There is an on-going discussion between clinicians and basic scientists about the clinical relevance of animal data (e.g., [11]), particularly when a model is designed to mimic a condition like pain that cannot be evaluated objectively but only indirectly by employing behavioral measures. One way of circumventing this problem is a translational approach to pain research, which necessitates a reciprocal transfer of bench-to-bedside data for promoting the further advancement and refinement of treatments [12,13]. This approach, however, will require the development of better animal models that more adequately mimic specific pain conditions as well as investigations to confirm animal findings in human experimental and clinical studies.
2332
138
Mechanisms of action of spinal cord stimulation
. Figure 138‐1 Schematic picture illustrating how SCS applied to different spinal cord segments, besides the effects on neuropathic pain, may induce changes in different target organs mediated via stimulation-induced changes in local autonomic activity, dorsal root reflexes, or on viscero-somatic reflexes. The numbers next to the red lightning bolts correspond with the numbers listed under the organ response. Some of these changes in target organ function may be beneficial for the individual and SCS at a certain site may thus be utilized therapeutically (redrawn after [3]). ICNS = intrinsic cardiac nervous system
SCS in Neuropathic Pain Several models of nerve injury-induced ‘‘painlike behavior’’ have been described (e.g., [14– 19]). After a nerve lesion (sciatic nerve or its peripheral branches; spinal roots) the animals soon develop a change in the posture of the nerve-injured extremity as well as increased sensitivity to normally innocuous mechanical and thermal stimuli. In fact, such hypersensitivity, similar to the clinically observed ‘‘allodynia,’’ is the most common behavioral sign that serves to monitor ‘‘pain’’ in animal models of neuropathy. However, the
pathophysiological mechanisms behind this phenomenon are still not fully identified [20,21]. The most common method of evaluating the tactile hypersensitivity is to determine the threshold that induces a withdrawal response to innocuous stimuli produced by probing the nerve injured hind paw with von Frey filaments. Normal rats generally tolerate a stiff filament (i.e., 30 g) without withdrawal while most nerve-lesioned animals develop severe hypersensitivity that may lead to a brisk withdrawal in response to the application of soft filaments calibrated to below 2–7 g of bending force. This quantifiable ‘‘symptom’’ thus mimics a ‘‘stimulus-evoked pain-like reaction,’’
Mechanisms of action of spinal cord stimulation
that can be interpreted as being equivalent to the ‘‘allodynia’’ observed in patients with painful neuropathic conditions (e.g., [22]). A major concern in this context is that not more than 20–40% of neuropathic pain patients present with mechanical allodynia (e.g., [11]) whereas tactile hypersensitivity occurs in a much larger proportion of nerve injured rats. An important aspect regarding the clinical relevance of animal models of ‘‘neuropathic pain’’ is that these animals almost never display behavioral signs indicating the presence of ongoing, spontaneous pain. These characteristics of animal models assumed to mimic neuropathic pain should therefore be taken into account when findings in studies using such animals subjected to neuromodulation, e.g., SCS, are interpreted in terms of clinical signs and symptoms – i.e., when data are translated from bench to bedside (> Figure 138‐2).
Dorsal Horn and Spinal Circuitry The presence of paresthesiae, indicating the activation of the dorsal columns (DC), is a
138
prerequisite for pain relief, but it has also been suggested that the tingling and vibratory sensations could be merely epiphenomena. If so, the therapeutic effects could instead be exerted via the activation of other pathways than the DC, notably the dorsolateral funiculus (DLF) containing descending, pain-controlling pathways. However, this tract is located at a distance from an SCS lead overlying the DCs and electric stimuli would therefore most likely first activate the interpositioned dorsal root fibers that enter horizontally and have a low threshold [23]. These roots would then generate segmental paresthesiae at the level of the active electrodes [24]. Thus, it is most likely that activation of the DCs proper evokes mechanisms that provide the pain relief. The pivotal role of the DCs in the SCS effect is further supported by the observation that preservation of somatosensory responses evoked from the painful region is, as a rule, a prerequisite for a positive effect. This is also indicated by the observation that pain associated with extensive deafferentation or direct injury of the DC fibers (where it is not possible to obtain paresthesiae at the painful site) fails to respond to SCS. Most clinicians consider
. Figure 138‐2 Radiographic appearance of a quadripolar SCS lead in a human (a) and a monopolar extradural electrode in a rat (b) (lateral projections). Arrows indicate active cathodes
2333
2334
138
Mechanisms of action of spinal cord stimulation
paresthesia-coverage of the painful area as a requirement for a beneficial effect in neuropathic pain. There is some experimental evidence that SCS may also inhibit nociceptive input at a segmental spinal level [10,25] and this has gained some support by the finding that the stimulation may depress a nociceptive flexor reflex in patients [26] – as well as in animals. Electrical stimuli applied to the sural nerve territory induce a contraction of the biceps femoris when the intensity of the stimulation is perceived as a ‘‘pricking’’ pain sensation. This flexor response conceivably represents the activation of Ad afferent fibers that may be attenuated by SCS. This effect seems to be related to the clinical pain relieving effect and has been proposed as an objective correlate to the pain relieving effect of SCS. This is, however, difficult to explain in view of the fact that SCS does not otherwise influence either novel acute pain or evoked, experimental pain resulting from Ad-fiber activation [67,68]. Studies have been performed at the Karolinska Institutet using ‘‘models of mononeuropathy,’’ i.e., rats with injury of the sciatic nerve or its branches resulting in hind paw hypersensitivity, in order to explore the mechanisms behind the SCS effects in neuropathic pain [4,8,27]. These animals have been implanted with a miniaturized SCS system so that the effect of stimulation on evoked pain can be monitored in the awake, freely moving animal. It has been demonstrated that in some of the rats SCS may effectively suppress the hypersensitivity, comparable to the effect on allodynia observed in patients. Thus, SCS applied for 20–30 min with stimulus parameters similar to those employed clinically may lead to a significant elevation of the abnormally low withdrawal threshold to innocuous mechanical and thermal (von Frey filaments) stimuli and this effect may outlast the SCS for up to 1 h. There is much evidence that the phenomenon of tactile allodynia is mediated mainly via low threshold Ab fibers and that it represents
a central state of hyperexcitability [e.g., 28]. The plasticity changes in the spinal cord following peripheral nerve injury are manifested by persistently augmented responsiveness and a high degree of spontaneous discharge of primarily wide-dynamic range dorsal horn neurons. In acute experiments we have demonstrated that SCS may induce a significant and long-lasting inhibition of both the after-discharges and the exaggerated principal response in such neurons in nerve-lesioned rats [29]. In the clinical setting, this suppression of dorsal horn neuronal activity may correspond to the beneficial effect of SCS not only on the allodynia but also on the spontaneous neuropathic pain. These observations suggest that SCS may preferentially influence Ab-fiber related functions. This notion is further supported by the finding that the threshold of the early component of the flexor reflex, which is Ab-fiber mediated, is elevated whereas the late C-fiber dependent late phase is unaffected ([27]; see also [30]). However, it has also been reported that the C-fiber flexor reflex can be significantly attenuated, but this observation was made in normal, intact animals [31]. It has further been shown that SCS significantly decreased the duration of long-term-potentiation (LTP) response to C-fiber activation from about 6 h to about 30 min [32]. It should be noted that in these experiments only the sensitized C-fiber response was influenced while neither the normal C- nor Ab- functions were affected. The mechanisms involved in the phenomenon of cutaneous hypersensitivity and on-going pain as a result of nerve injury are incompletely understood, and the emphasis on large, lowthreshold fiber related functions as pivotal for explaining the effect of SCS is necessarily an oversimplification. It might well be that the mode of action of SCS instead relates to a generalized state of peripheral and central sensitization (involving sensitized or awakened nociceptors), descending spinal facilitation, etc. The original conceptual basis for SCS presupposes antidromic activation of ascending
Mechanisms of action of spinal cord stimulation
dorsal column fibers and this implies that the region of action is primarily segmental. Our experimental data support this interpretation but a research group in Beirut has provided some evidence that instead the effect is predominantly exerted via a supraspinal loop [33,34].
Possible Transmitter Mechanisms Involved in SCS For obvious reasons, electric current applied to the dorsal aspect of the spinal cord activates multiple transmitter/receptor systems, but little is known about systems that are critically involved in the attenuation of chronic, neuropathic pain by SCS. Human data from analyses of lumbar CSF in conjunction with SCS are sparse and inconclusive. However, it appears that opioid mechanisms conceivably are not involved. There is some evidence that SCS tends to increase substance P (SP) content in human CSF and spinal release of SP and serotonin in cats [35,36]. However, it might well be that the SCSinduced changes of SP are not necessarily related to its pain relieving effect. In a series of acute experiments using microdialysis in the dorsal horn of nerve lesioned rats we have demonstrated that SCS reduces the release of excitatory amino acids (glutamate, aspartate) and at the same time augments the GABA release [37]. It is of special interest that this effect on the GABA system occurred only in rats that in preceding experiments had been found to respond to SCS with significant suppression of hind paw hypersensitivity [38]. These results confirm earlier observations that the state of central hyperexcitability manifested in the development of allodynia after peripheral nerve injury relates to dysfunction of the spinal GABA systems (e.g., [39]), and it appears that SCS may act by restoring normal GABA levels in the dorsal horn. These findings were supplemented by behavioral experiments where we showed that that the allodyniasuppressive effect of SCS could be counteracted by
138
intrathecal injection of a GABAB antagonist whereas the GABAA antagonist bicuculline was less effective. Conversely, intrathecal administration of GABA or a GABAB agonist, baclofen, markedly enhanced the effect of SCS [40]. In subsequent studies it was found that rats that were non-responders to SCS, i.e., their hind paw mechanical hypersensitivity was not attenuated, could be converted to responders with intrathecal administration of low, by themselves ineffective, doses of baclofen. The same potentiating effect was found with adenosine and it can thus be concluded that both the GABA- and adenosine related systems are directly involved in the pain relieving effect of SCS [40,41]. These results initiated a clinical study where it was demonstrated that the SCS effect can be enhanced by simultaneously administering intrathecal baclofen in low doses [42,43]. This appears to be a good example of translational research enabling direct transfer of results ‘‘from the bench to bedside.’’ Later studies also demonstrated that gabapentin, pregabalin and clonidine may have similar potentiating effects in non-responding rats [44,45]. In particular, the results obtained with clonidine are of interest since it is known that the antinociceptive effect of this substance is related to activation of the spinal cholinergic system [46]. If so, the effect of SCS might act also via involvement of these mechanisms, and recent studies in the rat have provided evidence supporting this notion [47]. A spinal microdialysis study showed that the release of acetylcholine (Ach) in the dorsal horn was augmented in nerve-injured rats responding to SCS while the Ach levels in the non-responders did not change during SCS treatment. Further behavioral studies using Ach receptor (muscarinic and nicotinic) antagonists administered intrathecally indicate the pivotal importance of activation of the muscarinic M4 and M2 receptors for the SCS effect. A recent immunohistochemical study appears to confirm the crucial role of the M4 muscarinic receptor in the response to SCS after
2335
2336
138
Mechanisms of action of spinal cord stimulation
peripheral nerve injury [48]. Moreover, intrathecal administration of a muscarinic receptor agonist may potentiate the SCS hypersensitivity suppressive effect (Song et al., Submitted). In conclusion, a cascade of transmitters is probably released by SCS and recent publications point to the complex interactions among the different neuronal circuits that may relate to the SCS effect (e.g., [49–51]). > Figure 138‐3 depicts a tentative scheme of some essential features of the mode of action of SCS when applied for neuropathic pain. This conceptualization of SCS is necessarily incomplete, in particular with regard to
the possible involvement of transmitter/receptor mechanisms. Further, it is primarily based on experiments performed on animal models of mononeuropathy with no definite signs of ongoing, spontaneous pain. Thus, such data should be interpreted with caution.
Clinical Pain States Associated with Dysautonomia Recent evidence strongly supports the notion that SCS may be efficacious in complex regional pain syndromes (CRPS) e.g., [22,52–55].
. Figure 138‐3 Schematic representation of the possible mode of action of SCS in neuropathic pain based on present knowledge derived predominantly from experiments performed on animal (rat) models of mononeuropathy. Both segmental and supraspinal mechanisms are represented. Possible supraspinal relays are not included because of insufficient knowledge about the organization of a proposed supraspinal loop. Broken arrow lines represent antidromic, and full line arrows ortodromic activation in the dorsal columns, their collaterals and in primary A-afferents. The diagram does not depict a possible SCS activation of the dorsolateral funiculus (DLF). It is conceivable that numerous transmitters and modulators are involved in the modulation exerted by interneurons (represented by ‘‘X’’). Descending control of second order neurons is here represented both as inhibitory (black arrows) and facilitatory (white arrows) influences from supraspinal centers. (SP – substance P; EAA – excitatory amino acids (glutatmate, aspartate); Ach- acetylcholine)) (redrawn after [4])
Mechanisms of action of spinal cord stimulation
A sympathicolytic action of SCS may be part of the mode of action behind the pain relieving effect in conditions associated with signs of sympathetic dysfunction (skin discoloration, temperature changes, sweating, change in dermal hairing, atrophy, etc.) that may be present in CRPS of both types (reflex sympathetic dystrophy – RSD, as well as in causalgia) [56–59]. However, these effects are only partially understood and are still a matter of controversy [57,60,61]. SCS may positively influence CRPS type I that has not been responsive to sympathetic blocks (e.g., [55]) although the probability of positive effects seems more likely in patients responding to diagnostic sympathetic blocks [22]. The effects of SCS in ischemic states (to be further discussed below) have in animal studies been found to depend also on antidromic activation of small diameter afferents that may result in a peripheral release of vasoactive substances. This type of mechanism could conceivably also be involved in the effects of SCS in CRPS. However, it has been argued that SCS induced peripheral vasodilatation is not a prerequisite for pain relief in CRPS I [57,60]. In pain syndromes associated with signs of autonomic disturbance SCS may hypothetically act on the symptoms in several ways: (1) by direct inhibitory actions onto central hyperexcitable neurons (as indicated above) (2) by decreasing sympathetic efferent output acting on the activated adrenoreceptors on the damaged sensory neurons, and (3) by reducing peripheral ischemia both by a sympathicolytic action and via e.g., antidromic mechanisms. This proposed third action is related to the ‘‘indirect-coupling hypothesis’’ for dysautonomic pain conditions where the damaged afferent neurons are supposed to develop hypersensitivity to even mild hypoxia [62]. Some animal models of CRPS have been developed ([63,64] type I; [65] type II) but their clinical significance has been questioned [66] and there are as yet no data from such models where SCS has been used.
138
SCS in Ischemic Pain Ischemic pain is generally characterized as essentially nociceptive. Several studies have indicated that SCS does not alleviate acute nociceptive pain (e.g., [2,3,67,68]). The pain relief induced by SCS is presumably secondary to attenuation of tissue ischemia that occurs as a result of either increasing/redistributing blood flow to the ischemic area or decreasing tissue oxygen demand (reviews, [69,70]). Further support for the notion that relief of ischemic pain in vascular disease is the result of activation of other mechanisms than those involved in neuropathic pain is the observation that stimulation may be effective also when applied below the threshold for paresthesiae [70,71]. No established animal models of peripheral arterial occlusive disease (PAOD) that gives rise to ischemic pain have yet been developed. Therefore, anesthetized animal models under normal physiological conditions have been used to investigate mechanisms of SCS-induced changes in peripheral blood flow during SCS [70,72–78]. Cutaneous blood flow and calculated vascular resistance in the glabrous surfaces of the ipsilateral and contralateral hind paws have been recorded using laser Doppler flowmetry. In some of the studies also perfusion in muscle tissue has been investigated [35,75]. Skin temperature was measured with a thermistor probe placed on the plantar aspect of the foot, next to the laser Doppler probe. This technique has made it possible to explore underlying mechanisms of peripheral microcirculation by using various interventions such as hexamethonium, CGRP antagonists (e.g., (CGRP 8–37)), adrenergic agonists and antagonists, nitric oxide synthetase inhibitors, sympathetic denervation, dorsal rhizotomies, and local paw cooling. Experimental studies using these interventions have provided evidence that SCS suppresses efferent sympathetic activity causing attenuation of peripheral vasoconstriction, which secondarily could
2337
2338
138
Mechanisms of action of spinal cord stimulation
lead to relief of pain [74,75]. In addition to vasodilatory effects obtained with suppression of sympathetic nervous activity, more recent data have demonstrated that SCS also depends on antidromic mechanisms involving the sensory afferent fibers of the dorsal roots that release peripheral CGRP with subsequent peripheral vasodilatation [72]. A salient observation is that SCS induced-vasodilatation of a cooled hind paw (<25 C) evoked an early phase of vasodilatation via the sensory afferent fibers and a late phase via suppression of the sympathetic efferent activity [79]. Thus, the control of each of these mechanisms most likely is related to the activity level of the sympathetic system. Later studies have confirmed that sensory afferent fibers are important mediators of SCS-induced vasodilatation, and that at higher, but not painful, SCS intensities C-fibers may also participate in the effect [78–80]. Thus, SCS at the spinal L2–L5 segments activates interneurons, which subsequently stimulate spinal terminals of Transient Receptor Potential V1 (TRPV1) containing sensory fibers, which are mainly of the C-type [81,82]. These fibers transmit action potentials antidromically from the site of stimulation in the spinal segments to the nerve endings in the peripheral tissues. The action potentials cause the production and release of vasodilators, including CGRP that binds to receptors in endothelial cells in vascular smooth muscle. The activation of endothelial cells leads to the production and subsequent release of nitric oxide (NO) that results in relaxation of vascular smooth muscle cells (review, [80]). The overall result is that vascular smooth muscle cell relaxation decreases vascular resistance and increases peripheral blood flow. In rats with experimental diabetes the TRPV1 containing sensory fibers appear to be among the first to degenerate and this may be one reason why in such rats SCS is less likely to produce peripheral vasodilatation at higher stimulation intensities [83]. However, SCS at lower intensities
was still effective in producing this effect in these animals. The skin flap model in rat is another way to demonstrate the effects of SCS on vasospasm and ischemia [84,85]. The important aspect of these studies is that they were designed to explore if pre-emptive SCS could increase the length of survival of a long-term groin skin flap and to identify possible neuromediators. The superficial epigastric artery was exposed and a detachable microvascular clip used to occlude this single feeding branch to the flap. The clip was removed after 12 h. SCS was applied for 30 min prior to the occlusion. In addition, one group received the CGRP-antagonist CGRP 8–37. After 7 days, the flaps of the control group were necrotized, but the majority of flaps in animals receiving pre-emptive SCS survived the 12-h occlusion. In addition, decreased survival was observed in a group of animals receiving CGRP 8–37. These results provide evidence that preemptive SCS may counteract the consequences of tissue ischemia and that CGRP is involved in this effect. The hypothetical mechanisms behind SCSinduced peripheral vasodilatation discussed above are schematically outlined in > Figure 138‐4.
SCS in Angina Pectoris Angina pectoris is often present in ischemic heart disease, and is characterized clinically by intense pain and discomfort in the chest, jaw, shoulder, back, or arm. The development of angina pectoris most commonly occurs when there is an imbalance between the supply and the demand of oxygen in the heart. The common mechanisms that decrease blood supply to the heart are vasospasm and occlusion of the coronary vessels. A large population of patients with chronic angina pectoris is unresponsive to conventional treatments [86]. However, SCS has been used to treat such
Mechanisms of action of spinal cord stimulation
138
. Figure 138‐4 A diagram illustrating effects of spinal cord stimulation (SCS) applied to the low thoracic-lumbar dorsal columns on mechanisms that produce vasodilation of peripheral vasculature. SCS activates interneurons that may (A) reduce the activity of spinothalamic tract (STT) cells (less probable in the clinical setting); (B) decrease the activity of sympathetic preganglionic neurons (2); (C) reduce the release of norepinephrine from sympathetic postganglionic neurons; (D) activate antidromically the dorsal root afferent fibers (1) with (E) release of calcitonin gene related peptide (CGRP) and nitric oxide (NO). In addition (not illustrated) intra- and extracellular changes increasing survival probability tissue in severe ischemia may be induced by the electrical activation
therapy-resistant angina pectoris since the 1980s [87,88] and has proven to very efficacious – but the mechanisms producing pain relief and improved heart function still remain unclear. Although early animal data demonstrated direct inhibitory effects of SCS on cardiac nociception, subsequent clinical studies have clearly proven that SCS does not merely relieve pain but it also improves the function of the heart. Infact, it appears that resolution of cardiac ischemia is the primary factor. Some investigators have proposed a stimulation-induced flow increase or redistribution of blood supply, while others interpret the reduction of coronary ischemia (decreased ST changes; reversal of lactate production) as being mainly due to decreased cardiomyocyte oxygen demand (e.g., [89,90]).
Studies have been performed to determine the role of blood flow changes in relieving angina pectoris with SCS. In a human experimental study, PET was utilized to provide some, though weak evidence for flow redistribution with SCS [91]. The same problem was addressed in an animal study by utilizing the distribution of isotope-labeled micro spheres in the hearts of anesthetized and artificially ventilated adult mongrel dogs [92]. The results of this experimental study failed to confirm the existence of a local flow increase in the myocardium or to show any changes in the pressure-volume relationships during SCS. However, a limitation of this study was that occlusions of the left anterior descending coronary arteries were performed in normal hearts. Considering that patients have
2339
2340
138
Mechanisms of action of spinal cord stimulation
long-term coronary ischemic disease, it would be appropriate to conduct such studies in canine hearts with previous infarctions and long-term ischemic episodes. In patients with compromised coronary arterial blood supply, SCS applied during standardized workloads, comparable to exercise and rapid cardiac pacing, markedly reduced the magnitude of ST segment changes of the electrocardiogram [90,93]. These results support the supposition that SCS may improve the working capacity of the heart. To mimic the development of chronic ischemic heart disease in an animal model of myocardial ischemia, an ameroid constrictor was implanted around the proximal left circumflex coronary artery of a group of canines [94]. The material of the constrictor ring slowly swells and progressively reduces blood flow through the artery and induces development of collaterals [95,96]. This process creates a collateral-dependent myocardial ischemia substrate. In subsequent experiments the chest was opened and the exposed heart was paced at a basal rate of 150 beats/min. An ECG plaque containing unipolar contacts was used to record from 191 sites on the left ventricle distal to the left coronary artery occluded by the ameroid constrictor. In order to stress the heart either angiotensin II was administered via the coronary artery blood supply to the right atrial ganglionated plexus or rapid ventricular pacing was applied via a standard pacemaker. Both stressors produced an elevation of the ST segments that, however, was markedly attenuated during SCS. In a similar way, ST segment responses were largely unchanged when rapid ventricular pacing (240 beats/min during 60 s) was applied during SCS. These data indicate that SCS may attenuate the deleterious effects that stressors associated with chemical activation of the intrinsic cardiac nervous system exert on a myocardium with reduced coronary reserve. It could be concluded that SCS appears to produce anti-ischemic
effects that contribute to improved cardiac function. Further evidence to support the anti-ischemic effects of SCS on the heart is the observation that pre-emptive SCS seems to have a protective effect on the myocardium. This was illustrated by the finding that the infarct size after controlled coronary artery occlusion is reduced. However, the protective effects of SCS therapy are lost if it is initiated after ischemia induction. Recent studies indicate that SCS-induced release of local catecholamines in the myocardium may trigger protective changes related to mechanisms behind ‘‘ischemic pre-conditioning’’ [97,98] – but without any signs of other ischemic changes. There are also other signs indicating that SCS may induce a state similar to that following a short ischemic period, e.g., the activation of protein-kinase C (for discussion, see e.g., [98]). In ischemia, the intrinsic cardiac nervous system is profoundly activated [99,100]. If this activity persists it may result in spreading dysrhythmias that lead to more generalized ischemia. An exciting observation is that SCS appears to stabilize the activity of these intrinsic cardiac neurons especially during the ischemic challenge of coronary artery occlusions. As in patients with angina, SCS can reduce the symptoms and signs of ischemia for long periods after the stimulation is terminated. Modulation of the intrinsic cardiac nervous system may be at least one mechanism that protects the heart from more severe ischemic threats due to generalized arrhythmias (DeJongste et al., unpublished data; [101]. Some of the pathways and putative mechanisms behind effects of SCS on cardiac function discussed above are briefly summarized in > Figure 138‐5. As noted in the introduction a variety of organ functions may be affected by SCS applied at various levels of the spinal cord (> Figure 138‐1). In the concluding paragraphs some further examples of SCS applications are described.
Mechanisms of action of spinal cord stimulation
138
. Figure 138‐5 A diagram illustrating effects of SCS applied to the T1-T2 dorsal columns (DC) on neuronal mechanisms that reduce pain and improve cardiac function resulting from ischemic heart disease. SCS activates interneurons that may (1) reduce the activity short-term of spinothalamic tract (STT) cells; (2) modulate the activity of sympathetic preganglionic neurons and (3) stabilize the intrinsic cardiac nervous system (ICN), reduce ischemia and decrease infarct size. In addition a protective effect on ischemic cardiomyocytes related to local release of catecholamines has recently been demonstrated (see text)
Irritable Bowel Syndrome Functional bowel disorders, including irritable bowel syndrome (IBS), are common abnormalities of the gastrointestinal tract that are associated with crampy abdominal pain, abnormal bowel habits, and somatic hypersensitivity [102– 104]. The mechanisms underlying chronic visceral symptoms of IBS are not well understood, and presently, no effective therapy is available. Since SCS is beneficial in reducing some types of visceral pain and effectively suppresses hyperexcitable somatosensory and viscerosomatic (bladder) reflexes in patients experiencing spasticity (e.g., [105]), our research team decided to study the effects of SCS as a potential therapy
for visceral pain originating from the gastrointestinal tract [106]. We used a rat model developed by Ness & Gebhart [107] to quantify the level of visceral pain. In this model abdominal muscle contractions are recorded during colorectal distension employed to induce a nociceptive reflex. To resemble the clinical condition of IBS, the model was modified to produce visceral hypersensitivity by infusing a low concentration of acetic acid into the colon, which causes hypersensitivity in the absence of mucosal damage [108–110]. In this model, a miniature SCS electrode system was chronically implanted with the technique used in our studies on neuropathic pain. After 1 week, animals were anesthetized briefly with
2341
2342
138
Mechanisms of action of spinal cord stimulation
isoflurane to suture a strain gauge force transducer on the right external oblique abdominal muscle. A colorectal balloon was then used to distend normal colons as well as colons irrigated with acetic acid; the number of abdominal contractions both with and without SCS was recorded from the strain gauge. The results showed that SCS significantly suppressed the visceromotor responses that were produced with colorectal distension in both normal rats and in those with sensitized colons. In a subsequent study, a rat model of post-inflammatory colonic hypersensitivity was used and also in this condition SCS significantly reduced abdominal contractions during innocuous distension [111]. The suppressive effect of SCS on colonic sensitivity provides evidence that SCS may have therapeutic potential for the treatment of visceral pain of gastrointestinal origin associated with abdominal cramping and painful abdominal spasms. In fact, there is a case report of a patient suffering from severe IBS who with SCS experienced reduced hypersensitivity and relief of diarrhea [112]. Furthermore, Khan et al. [113] in a retrospective study have shown that SCS can be used effectively to treat a variety of visceral pain syndromes including generalized abdominal pain, chronic non-alcoholic pancreatitis and pain following post traumatic splenectomy. These clinical observations are in agreement with the animal studies and support the notion that SCS might be used to treat various functional bowel as well as other visceral disorders. As a direct consequence of our experiments a randomized controlled trial of SCS therapy in IBS is presently underway.
Other Organ Dysfunctional Syndromes Bronchial tree. Only one group so far has explored the possible effects of SCS for bronchospasm
[114]. They used a sheep model in which bronchospasm was produced by inhalation of an Ascaris suum extract. High cervical SCS (C1–C2) markedly decreased bronchomotor tone. If such effects can be demonstrated with more common allergens applied in humans, SCS could develop into a therapy to treat conditions where bronchial constriction is involved. Urinary bladder. In the 1980s, SCS was commonly used to treat spasticity in multiple sclerosis but with the introduction of intrathecal baclofen this indication was given up. However, decreasing the urgency of voiding was the most marked effect of SCS on urinary bladder spasticity [115,116]. There are also several case reports that discuss the beneficial effects of using neuromodulation on other syndromes such as interstitial cystitis, and mixed low mid-line pain syndromes using a retrograde approach with low sacral, conus or root stimulation (reviews, see [117,118]). These two examples illustrate that SCS may provide benefits for various autonomic functions and improve organ function as we move the stimulating electrode along the neuro-axis.
Conclusions SCS induces effects in multiple systems and the benefit for a certain condition may depend on (1) the site on the spinal cord activated (2) selection of a certain biological effect that may be selectively relevant and of benefit in a certain pain syndrome. Knowledge about physiological mechanisms behind the beneficial effects provides a corner stone for further development of neurostimulation as well as for strategies to support the technique with receptor-active pharmaceuticals in cases with unsatisfactory response to the stimulation per se [42,43]. In order to further explore the physiological mechanisms of SCS in various painful (and other) conditions, a tight dialogue
Mechanisms of action of spinal cord stimulation
between clinicians and basic researchers is essential. Questions asked by the clinician should initiate applied research projects for the basic scientist who has the possibility of testing the ideas in experimental simplified systems. The clinician and experimentalist should design and evaluate the animal models and their data outputs together in order to ascertain a maximal relevance of each model for the therapeutic problem. SCS is a therapy that may be effective in some pain syndromes otherwise resistant to treatment; it is well tolerated for patients, minimally invasive, reversible and with few side effects as compared to chronic pharmacotherapy. Furthermore, in some syndromes, SCS may have its primary effect by improving an organ function, which secondarily can result in attenuation of the pain generating mechanisms associated with the disease. We firmly believe that SCS at present is an under-used treatment modality. Furthermore, our health care system demands ‘‘evidence based’’ and ‘‘mechanism based’’ therapies, and this amplifies the need to expand our knowledge through research projects aimed at further exploration of physiological mechanisms that are activated by neuromodulation.
Acknowledgments Research data from the laboratories of Karolinska Institute and Oklahoma University reported in this chapter have been obtained with support of The Swedish Medical Research Council, several NIH funds, Karolinska Institutet Funds and from Medtronic Europe SA.
References 1. Foreman RD, DeJongste MJL, Linderoth B. Integrative control of cardiac function by cervical and thoracic spinal neurons. In: Armour JA, Ardell JL, editors. Basic and clinical neurocardiology, Chap. 5. London: Oxford University Press; 2004. p. 153-86.
138
2. Linderoth B, Foreman RD. Physiology of spinal cord stimulation. Review and update. Neuromodulation 1999;2(3):150-64. 3. Linderoth B, Foreman RD. Mechanisms of spinal cord stimulation in painful syndromes. Role of animal models. Pain Med 2006;7 Suppl 1:14-26. 4. Meyerson BA, Linderoth B. Mode of action of spinal cord stimulation in neuropathic pain. J Pain Symptom Manage 2006;31:6-12. 5. Woolf CJ, Bennett GJ, Doherty M, et al. Towards a mechanism-based classification of pain? Pain 1998; 77:227-9. 6. Foreman RD. Integration of viscerosomatic sensory input at the spinal level. Prog Brain Res 2000; 122:209-21. 7. Linderoth B, Meyerson BA. Central nervous system stimulation for neuropathic pain. In: Hansson P, Hill RG, Fields HL, Marchettini P, editors. Neuropathic pain: pathophysiology and treatment. Progress in pain research and management, vol 21, Chap. 13. Seattle: IASP Press; 2001. p. 223-49. 8. Meyerson BA, Linderoth B. Spinal cord Stimulation: mechanisms of action in neuropathic and ischemic pain. In: Simpson BA, editor. Electrical stimulation and the relief of pain, vol 15, Chap. 11. New York: Elsevier; 2003. p. 161-82. 9. Ho¨kfelt T, Zhang X, Wiesenfeld-Hallin Z. Messenger plasticity in primary sensory neurons following axotomy and its functional implications. TINS 1994;17:22-30. 10. Chandler MJ, Brennan TJ, Garrison DW, Kim KS, Schwartz PJ, Foreman RD. A mechanism of cardiac pain suppression by spinal cord stimulation: implications for patients with angina pectoris. Eur Heart J 1993;14:96-105. 11. Hansson P. Difficulties in stratifying neuropathic pain by mechanisms. Eur J Pain 2003;7(4):353-7. 12. Klein T, Magerl W, Rolke R, Treede RD. Human surrogate models of neuropathic pain. Pain 2005;115 (3):227-33. 13. Mao J. Translational pain research: bridging the gap between basic and clinical research. Pain 2002;97:183-7. 14. Bennett GJ, Xie YKA. Peripheral mononeuropathy in rat that produces disorders of pain sensation like those seen in man. Pain 1988;33(1):87-107. 15. Decosterd I, Woolf CJ. Spared nerve injury: an animal model of persistent peripheral neuropathic pain. Pain 2000;87(2):149-58. 16. Gazelius B, Cui JG, Svensson M, Meyerson B, Linderoth B. Photochemically induced ischaemic lesion of the rat sciatic nerve. A novel method providing high incidence of mononeuropathy. Neuroreport 1996;7:2619-23. 17. Kim SH, Chung JM. An experimental model for peripheral neuropathy produced by segmental spinal nerve ligation in the rat. Pain 1992;50(3):355-63.
2343
2344
138
Mechanisms of action of spinal cord stimulation
18. Seltzer Z, Dubner R, Shir Y. A novel behavioral model of neuropathic pain disorders produced in rats by partial sciatic nerve injury. Pain 1990;43(2):205-18. 19. Wall PD, Gutnick M. Ongoing activity in peripheral nerves: the physiology and pharmacology of impulses originating from neuroma. Exp Neurol 1974;43:580-93. 20. Baba H, Ji RR, Kohno T, Moore KA, Ataka T, Wakai A, Okamoto M, Woolf CJ. Removal of GABAergic inhibition facilitates polysynaptic A fiber-mediated excitatory transmission to the superficial spinal dorsal horn. Mol Cell Neurosci 2003;24:818-30. 21. Scholz J, Woolf CJ. Can we conquer pain? Nature Neurosci 2002;5 Suppl:1062-7. 22. Harke H, Gretenkort P, Ladleif HU, Rahman S. Spinal cord stimulation in sympathetically maintained complex regional pain syndrome type I with severe disability. A prospective clinical study. Eur J Pain 2005;9:363-73. 23. Holsheimer J. Computer modelling of spinal cord stimulation and its contribution to therapeutic efficacy. Spinal Cord 1998;36(8):531-40. 24. Feirabend HKP, Choufoer S, Ploeger S, Holsheimer J, van Gool JD. Morphometry of human superficial dorsal and dorsolateral column fibres: significance to cord stimulation. Brain 2002;125:1137-49. 25. Foreman RD, Beall JE, Applebaum AE, Coulter JD, Willis WD. Effects of dorsal column stimulation on primate spinothalamic tract neurons. J Neurophysiol 1976;39:534-46. 26. Garcia-Larrea L, Sindou M, Mauguie`re F. Nociceptive flexion reflexes during analgesic neurostimulation in man. Pain 1989;39:145-56. 27. Meyerson BA, Ren B, Herregodts P, Linderoth B. Spinal cord stimulation in animal models of mononeuropathy: effects on the withdrawal response and the flexor reflex. Pain 1995;61:229-43. 28. Woolf C, Doubell T. The pathophysiology of chronic pain – increased sensitivity to low threshold A-beta fiber inputs. Curr Opin Neurobiol 1994;4:525-34. 29. Yakhnitsa V, Linderoth B, Meyerson BA. Spinal cord stimulation attenuates dorsal horn neuronal hyperexcitability in a rat model of mononeuropathy. Pain 1999;79:223-33. 30. Linderoth B, Meyerson BA. Dorsal column stimulation: modulation of somatosensory and autonomic function. In: McMahon SB, Wall PD, editors. The neurobiology of pain. Seminars in the neurosciences, vol 7. London: Academic Press; 1995. p. 263-77. 31. Saade´ NE, Tabet, Soueidan SA, Bitar M, Atweh SF, Jabbur SJ. Supraspinal modulation of noceception in awake rats by stimulation of the dorsal column nuclei. Brain Res 1986;369:307-10. 32. Wallin J, Fiska A, Tjolsen A, Linderoth B, Hole K. Spinal cord stimulation inhibits long-term potentiation of spinal wide dynamic range neurons. Brain Res 2003;973:39-43.
33. El-Khoury C, Hawwa N, Baliki M, Atweh SF, Jabbur SJ, Saade NE. Attenuation of neuropathic pain by segmental and supraspinal activation of the dorsal column system in awake rats. Neuroscience 2002;112(3):541-53. 34. Saade´ NE, Al Amin H, Chalouhi S, Baki SA, Jabbur SJ, Atweh SF. Spinal pathways involved in supraspinal modulation of neuropathic manifestations in rats. Pain 2006;126:280-93. 35. Linderoth B, Gazelius B, Franck J, Brodin E. Dorsal column stimulation induces release of serotonin and substance P in the cat dorsal horn. Neurosurgery 1992;31 (2):289-97. 36. Meyerson BA, Brodin E, Linderoth B. Possible neurohumoral mechanisms in CNS stimulation for pain suppression. Appl Neurophysiol 1985;48:175-80. 37. Cui JG, O’Connor WT, Ungerstedt U, Meyerson BA, Linderoth B. Spinal Cord Stimulation attenuates augmented dorsal horn release of excitatory amino acids in mononeuropathy via a GABAergic mechanism. Pain 1997;73:87-95. 38. Stiller CO, Cui J-G, O’Connor WT, Brodin E, Meyerson BA, Linderoth B. Release of GABA in the dorsal horn and suppression of tactile allodynia by spinal cord stimulation in mononeuropathic rats. Neurosurgery 1996;39: 367-75. 39. Castro-Lopes JM, Tavares I, Coimbra A. GABA decreases in the spinal cord dorsal horn after peripheral neurectomy. Brain Res 1993;620:287-91. 40. Cui JG, Linderoth B, Meyerson BA. Effects of spinal cord stimulation on touch-evoked allodynia involve GABAergic mechanisms. An experimental study in the mononeuropathic rat. Pain 1996;66:287-95. 41. Cui JG, Sollevi A, Linderoth B, Meyerson BA. Adenosine receptor activation suppresses tactile hypersensitivity and potentiates effect of spinal cord in mononeuropathic rats. Neurosci Lett 1997;223:173-6. 42. Lind G, Meyerson BA, Winter J, Linderoth B. Intrathecal baclofen as adjuvant therapy to enhance the effect of spinal cord stimulation in neuropathic pain: a pilot study. Eur J Pain 2004;8(4):377-83. 43. Lind G, Schechtmann G, Winter J, Meyerson BA, Linderoth B. Baclofen-enhanced spinal cord stimulation and intrathecal baclofen alone for neuropathic pain: longterm outcome of a pilot study. Eur J Pain 2007;12:132-6. 44. Schechtmann G, Wallin J, Meyerson BA, Linderoth B. Intrathecal clonidine potentiates suppression of tactile hypersensitivity by spinal cord stimulation in a model of neuropathy. Anesth Analg 2004;99:135-9. 45. Wallin J, Cui J-G, Yahknitsa V, Schechtmann G, Meyerson BA, Linderoth B. Gabapentin and pregabalin suppress tactile allodynia and potentiate spinal cord stimulation in a model of neuropathy. Eur J Pain 2002;6:261-72. 46. Obata H, Li X, Eisenach JC. a2-Adrenoceptor activation by clonidine enhances stimulation-evoked acetylcholine
Mechanisms of action of spinal cord stimulation
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
release from spinal cord tissue after nerve ligation in rats. Anesthesiology 2005;102:657-62. Schechtmann G, Song Z, Ultenius C, Meyerson BA, Linderoth B. Cholinergic mechanisms in the pain relieving effect of spinal cord stimulation in a model of neuropathy. Pain 2008;139:136-45. Song ZY, Ultenius C, Schechtmann G, Meyerson BA, Linderoth B. Downregulation of M4 but not M2 muscarinic receptors in the dorsal horn after peripheral nerve injury relates to responsiveness to SCS. Abstract of the Second International Congress on Neuropathic Pain, Berlin; June 2007. Eur J Pain 2007;11 Suppl 1:175. Li DP, Chen SR, Pan YZ, Levey AI, Pan HL. Role of presynaptic muscarinic and GABA(B) receptors in spinal glutamate release and cholinergic analgesia in rats. J Physiol 2002;543 (Pt 3):807-18. Obata H, Saito S, Sasaki M, Goto F. Possible involvement of a muscarinic receptor in the anti-allodynic action of a 5-HT2 receptor agonist in rats with nerve ligation injury. Brain Res 2002;932:124-8. Wang XL, Zhang HM, Li DP, Chen SR, Pan HL. Dynamic regulation of glycinergic input to spinal dorsal horn neurones by muscarinic receptor subtypes in rats. J Physiol 2006;571:403-13. Kemler MA, Barendse GA, van Kleef M, de Vet HC, Rijks CP, Furnee CA, van den Wildenberg FA. Spinal cord stimulation in patients with chronic reflex sympathetic dystrophy. N Engl J Med 2000b;343:618-24. Kemler MA, de Vet HC, Barendse GA, van den Wildenberg FA, van Kleef M. Spinal cord stimulation for chronic reflex sympathetic dystrophy – five-year follow-up. N Engl J Med 2006;354:2394-6. Kumar K, Hunter G, Demeria D. Spinal cord stimulation in treatment of chronic benign pain: challenges in treatment planning and present status, a 22-year experience. Neurosurgery 2006;58:481-96. Olsson GL, Meyerson BA, Linderoth B. Spinal cord stimulation in adolescents with Complex Regional Pain Syndrome type I (CARPS-I). Eur J Pain 2008;12:53-9. Baron R, Levine JD, Fields HL. Causalgia and reflex sympathetic dystrophy: does the sympathetic nervous system contribute to the generation of pain? Muscle Nerve 1999;22(6):678-95. Kemler MA, Barendse GA, van Kleef M, Egbrink MG. Pain relief in complex regional pain syndrome due to spinal cord stimulation does not depend on vasodilation. Anesthesiology 2000;92:1653-60. Kumar K, Nath RK, Toth C. Spinal cord stimulation is effective in the management of reflex sympathetic dystrophy. Neurosurgery 1997;40:503-9. Wasner G, Heckmann K, Maier C, Baron R. Vascular abnormalities in acute reflex sympathetic dystrophy (CRPS I): complete inhibition of sympathetic nerve activity with recovery. Arch Neurol 1999;56:613-20.
138
60. Ather M, Di Vadi P, Light D, Wedley JR, Hamann WC. Spinal cord stimulation does not change peripheral skin blood flow in patients with neuropathic pain. Eur J Anaesthesiol 2003;20(9):736-9. 61. Max MB, Gilron I. Sympathetically maintained pain: has the emperor no clothes? Neurology 1999;52(5):905-7. 62. Michaelis M. Coupling of sympathetic and somatosensory neurons following nerve injury: mechanisms and potential significance for the generation of pain. In: Devor M, Rowbotham MC, Wiesenfeld Z, editors. Progress in pain research and management, vol. 16. IASP Press; Seattle: p. 645-56. 63. Coderre TJ, Xanthos DN, Francis L, Bennett GJ. Chronic post-ischemia pain (CPIP): a novel animal model of complex regional pain syndrome-type I (CRPS-I; reflex sympathetic dystrophy) produced by prolonged hindpaw ischemia and reperfusion in the rat. Pain 2004;112:94-105. 64. Guo TZ, Wei T, Kingery WS. Glucocorticoid inhibition of vascular abnormalities in a tibia fracture rat model of complex regional pain syndrome type I. Pain 2006;121:158-67. 65. Kingery WS, Davies MF, Clark JD. A substance P receptor (NK1) antagonist can reverse vascular and nociceptive abnormalities in a rat model of complex regional pain syndrome type II. Pain 2003;104:75-84 66. Baron R. Can we model CRPS type 1? Pain 2004;11:8-9. 67. Lindblom U, Meyerson BA. Influence on touch, vibration and cutaneous pain of dorsal column stimulation in man. Pain 1975;1:257-70. 68. Nashold BS. Overview of neuroaugmentation. Neurosurgery 1977;1:230-2 69. Augustinsson LE, Linderoth B, Mannheimer C. Spinal cord stimulation in different ischemic conditions. In: Illis LS, editor. Spinal cord dysfunction III: functional stimulation, Chap. 12. Oxford: Oxford University Press; 1992. p. 270-93. 70. Linderoth B. Spinal cord stimulation in ischemia and ischemic pain. In: Horsch S, Claeys L editors. Spinal cord stimulation: an innovative method in the treatment of PVD and angina. Darmstadt: Steinkopff Verlag; 1995. p. 19-35. 71. Eddicks S, Maier-Hauff K, Schenk M, Muller A, Baumann G, Theres H. Thoracic spinal cord stimulation improves functional status and relieves symptoms in patients with refractory angina pectoris: the first placebo-controlled randomised study. Heart 2007;93:585-90. 72. Croom JE, Foreman RD, Chandler MJ, Barron KW. Cutaneous vasodilation during dorsal column stimulation is mediated by dorsal roots and CGRP. Am J Physiol Heart Circ Physiol 1997;272:H950-H957. 73. Croom JE, Foreman RD, Chandler MJ, Barron KW. Reevaluation of the role of the sympathetic nervous system in cutaneous vasodilation during dorsal spinal cord stimulation: are multiple mechanisms active? Neuromodulation 1998;1:91-101.
2345
2346
138
Mechanisms of action of spinal cord stimulation
74. Linderoth B, Gunasekera L, Meyerson BA. Effects of sympathectomy on skin and muscle microcirculation during dorsal column stimulation: animal studies. Neurosurgery 1991;29:874-9. 75. Linderoth B, Herregodts P, Meyerson BA. Sympathetic mediation of peripheral vasodilation induced by spinal cord stimulation: animal studies of the role of cholinergic and adrenergic receptor subtypes. Neurosurgery 1994;35:711-19. 76. Tanaka S, Barron KW, Chandler MJ, Linderoth B, Foreman RD. Low intensity spinal cord stimulation may induce cutaneous vasodilation via CGRP release. Brain Res 2001;896:183-7. 77. Tanaka S, Barron KW, Chandler MJ, Linderoth B, Foreman RD. Local cooling alters neural mechanisms producing changes in peripheral blood flow by spinal cord stimulation. Auton Neurosci 2003a;104:117-27. 78. Tanaka S, Komori N, Barron KW, Chandler MJ, Linderoth B, Foreman RD. Mechanisms of sustained cutaneous vasodilation induced by spinal cord stimulation. Auton Neurosci 2004;114(1–2):55-60. 79. Tanaka S, Barron KW, Chandler MJ, Linderoth B, Foreman RD. Role of primary afferents in spinal cord stimulation-induced vasodilatation: characterization of fiber types. Brain Res 2003b;959:191-8. 80. Wu M, Linderoth B, Foreman RD. Putative mechanisms behind effects of spinal cord stimulation on vascular diseases: a review of experimental studies. Auton Neurosci 2008;138:9-23. 81. Wu M, Komori N, Qin C, Farber J, Linderoth B, Foreman RD. Sensory fibers containing vanilloid receptor-1 (VR-1) participate in spinal cord stimulation-induced vasodilation. Brain Res 2006;1107:177-84. 82. Wu M, Qin C, Farber JP, Linderoth B, Foreman RD. Roles of peripheral terminals of transient receptor potential vanilloid-1 containing sensory fibers in spinal cord stimulation-induced peripheral vasodilation. Brain Res 2007b;1156:80-92. 83. Wu M, Thorkilsen M, Qin C, Farber JP, Linderoth B, Foreman RD. Effects of spinal cord stimulation on peripheral circulation in stretozotocin-induced diabetic rats. Neuromodulation 2007;10:216-23. 84. Gherardini G, Lundeberg T, Cui J, Eriksson SV, Trubek S, Linderoth B. Spinal cord stimulation improves survival in ischemic skin flaps: an experimental study of the possible mediation by calcitonin gene-related peptide. Plast Reconstr Surg 1999;103(4):1221-8. 85. Linderoth B, Gheradini G, Ren B, Lundeberg T. Pre-emptive spinal cord stimulation reduces ischemia in an animal model of vasospasm. Neurosurgery 1995;37:266-72. 86. DeJongste MJL, Haaksma J, Hautvast RW, Hillege HL, Meyler PW, Staal MJ, Sanderson JE, Lie KI. Effects of spinal cord stimulation on daily life myocardial
87.
88. 89.
90.
91.
92.
93.
94.
95. 96.
97.
98.
ischemia in patients with severe coronary artery disease. A prospective ambulatory ECG study. Br Heart J 1994;71:413-18. Mannheimer C, Augustinsson LE, Carlsson CA, Manhem K, Wilhelmsson C. Epidural spinal electrical stimulation in severe angina pectoris. Br Heart J 1988;59:56-61. Murphy DF, Giles. Dorsal column stimulation for pain relief from intractable angina. Pain 1987;28:365-8 Eliasson T, Augustinnson LE, Mannheimer C. Spinal cord stimulation in severe angina pectoris – presentation of current studies, indications, and clinical experience. Pain 1996;65:169-79. Mannheimer C, Eliasson T, Andersson B, Bergh CH, Augustinsson LE, Emanuelsson H, Waagstein F. Effects of spinal cord stimulation in angina pectoris induced by pacing and possible mechanism of action. Br Med J 1993;307:477-80. Hautvast RW, Blanksma PK, DeJongste MJ, et al. Effect of spinal cord stimulation o myocardial blood flow assessed by positron emission tomography in patients with refractory angina pectoris. Am J Cardiol 1996;77:462-7. Kingma J, Linderoth B, Ardell JL, et al. Neuromodulation therapy does not influence blood flow distribution or left-ventricular dynamics during acute mycardial ischemia. Auton Neurosci 2001;91(1–2):47-54. Hautvast RW, Brouwer J, DeJongste MJ, et al. Effect of spinal cord stimulation on heart rate variability and myocardial ischemia in patients with chronic intractable angina pectoris – a prospective ambulatory elecrocardiographic study. Clin Cardiol 1998;21:33-8. Cardinal R, Ardell J, Linderoth B, Vermeulen M, Foreman RD, Armour JA. Spinal cord activation differentially modulates ischemic electrical responses to different stressors in canine ventricles. Auton Neurosci 2004;111 (1):34-47 Schaper W. The collateral circulation of the heart. NorthHolland; Amsterdam: p. 183-7. Tomoike H, Inou T, Watanabe K, Mizukami M, Kikuchi Y, Nakamura M. Functional significance of collaterals during ameroidinduced coronary stenosis in conscious dogs. Circulation 1983;67:1001-8. Ardell JL, DellI´talia LJ, CC, Milhorn DM, Linderoth B, DeJongste MJL, Foreman RD, Armour JA. Spinal cord stimulation modulates catecholamine release into interstitial fluid in the canine myocardium. Abstracts of FASEB Meeting, Baltimore, 2002. FASEB J 2002;16(4): A 118, part I. Southerland EM, Milhorn D, Foreman RD, Linderoth B, DeJongste MJL, Armour JA, Subramanian V, Singh M, Singh K, Ardell JL. Pre-emptive, but not reactive, spinal cord stimulation mitigates transient ischemia-induced myocardial infarction via cardiac adrenergic neurons. Am J Physiol Heart Circ Physiol 2007;292(1):H311-H317.
Mechanisms of action of spinal cord stimulation
99. Armour JA, Linderoth B, Arora RC, DeJongste MJ, Ardell JL, Kingma JG Jr, Hill M, Foreman RD. Long-term modulation of the intrinsic cardiac nervous system by spinal cord neurons in normal and ischaemic hearts. Auton Neurosci 2002;95(1–2):71-9. 100. Foreman RD, Linderoth B, Ardell JL, Barron KW, Chandler MJ, Hull SS Jr, TerHorst GJ, DeJongste MJL, Armour JA. Modulation of intrinsic cardiac neurons by spinal cord stimulation: implications for therapeutic use in angina pectoris. Cardiovasc Res 2000;47:367-75. 101. Issa ZF, Zhou X, Ujhelyi MR, Rosenberger J, Bhakta D, Groh WJ, Miller JM, Zipes DP. Thoracic spinal cord stimulation reduces the risk of ischemic ventricular arrhythmias in a postinfarction heart failure canine model. Circulation 2005;111(24):3217-20. 102. Lembo T, Fullerton S, Diehl D, Raeen H, Munakata J, Naliboff B, Mayer EA. Symptom duration in patients with irritable bowel syndrome. Am J Gastroenterol 1996;91:898-905. 103. Mayer EA, Gebhart GF. Basic and clinical aspects of visceral hyperalgesia. Gastroenterology 1994; 107:271-93. 104. Whitehead WE, Holtkotter B, Enck P, Hoelzl R, Holmes KD, Anthony J, Shabsin HS, Schuster MM. Tolerance for rectosigmoid distention in irritable bowel syndrome. Gastroenterology 1990;98:1187-92. 105. Illis LS, editor. Spinal cord dysfunction III: functional stimulation. Oxford: Oxford University Press; 1992. 106. Greenwood-Van Meerveld B, Johnson AC, Foreman RD, Linderoth B. Attenuation by spinal cord stimulation of a nociceptive reflex generated by colorectal distention in a rat model. Auton Neurosci 2003;104:17-24. 107. Ness TJ, Gebhart GF. Colorectal distention as a noxious visceral stimulus: physiologic and pharmacologic characterization of the pseudaffective reflexes in the rat. Brain Res 1988;450:153-69. 108. Gunter WD, Shepard JD, Foreman RD, Myers DA, Greenwood-Van Meerveld B. Evidence for visceral hypersensitivity in high-anxiety rats. Physiol Behav 2000;69:379-82. 109. Langlois A, Pascaud X, Junien JL, Dah SG, Riviere PJ. Response heterogeneity of 5-HT3 receptor antagonists in
110.
111.
112.
113.
114.
115.
116.
117. 118.
119.
120.
138
a rat visceral hypersensitivity model. Eur J Pharmacol 1996;318:141-4. Plourde V, St-Pierre S, Quirion R. Calcitonin generelated peptide in viscerosensitive response to colorectal distension in rats. Am J Physiol 1997;273:G191-G196. Greenwood-Van Meerveld B, Johnson AC, Foreman RD, Linderoth B. Spinal cord stimulation attenuates visceromotor reflexes in a rat model of post-inflammatory colonic hypersensitivity. Auton Neurosci 2005;122 (102):69-76. Krames E, Mousad DG. Spinal cord stimulation reverses pain and diarrheal episodes of irritable bowel syndrome: a case report. Neuromodulation 2004;7(2):82. Khan YN, Raza SS, Khan EA. Application of spinal cord stimulation for the treatment of abdominal visceral pain syndromes: case reports. Neuromodulation 2005;8:14-27. Gersbach P, Gardaz J-P, Soler M, et al. Influence of high-cervical spinal cord stimulation on antigen-induced bronchospasm in an ovine model. Abstract ALA/ATS Internat Congress, San Diego; 1999. Illis LS, Sedgwick EM, Tallis RC. Spinal cord stimulation in multiple sclerosis: clinical results. J Neurol Neurosurg Psychiatry 1980;43(1):1-14. Meglio M, Cioni B, Amico ED, Ronzoni G, Rossi GF. Epidural spinal cord stimulation for the treatment of neurogenic bladder. Acta Neurochir (Wien) 1980;54 (3–4):191-9. Bernstein AJ, Peters KM. Expanding indications for neuromodulation. Urol Clin North Am 2005;32(1):59-63. Van Balken MR, Vergunst H, Bemelmans BL. The use of electrical devices for the treatment of bladder dysfunction: a review of methods. J Urol 2004;172(3):846-51. Song Z, Meyerson BA, Lineroth B. Muscarinic receptor activation potentiates the effect of spinal cord stimulation on pain related behaviour in rats with mononeuropathy. Neurosci Lett 2008 Feb 26; [Epub ahead of print] 436:7-12. Harke H, Gretenkort P, Ladleif HU, Rahman S. Spinal cord stimulation in sympathetically maintained complex regional pain syndrome type I with severe disability. A prospective clinical study. Eur J Pain 2005;9:363-373.
2347
150 Mesencephalotomy for Cancer Pain P. L. Gildenberg
Mesencephalotomy as a treatment for pain has evolved in a manner that reflects the changing approach to pain management through the years. As such, a review of mesencephalotomy provides insight into the evolution of concepts of surgical pain management. More than a century ago, the primary pain pathway was demonstrated, first clinically, then in the laboratory and finally in the operating room. The rationale of the spinothalamic tract was introduced after the post-mortem examination of a patient who was known to have suffered a gun shot wound in 1878. The patient had lost unilateral pain sensation but had retained touch sensation. A second patient with a similar loss of sensation from a spinal tuberculoma led Spiller [1] to recognize in 1905 that the pain pathway within the spinal cord was located in the anterolateral columns. His observations reassured Martin [2] sufficiently for them to report in 1912 their success in treating pain by surgical section of the anterolateral spinal cord. Since that time, anterolateral cordotomy has been used more often than any other procedure for pain management. However, if the pain occurs in the upper extremities, upper body, or even in the face, it becomes necessary to interrupt the primary pain pathway at a level above the spinal cord. This led to an attempt by Dogliotti [3] in 1938 to section the spinothalamic tract in the pons, but the results were inconsistent and neurological complications were common. The level at which the spinothalamic pathway lies closest to the surgically approachable surface is in the mid-brain, and sectioning it at that level is #
Springer-Verlag Berlin/Heidelberg 2009
called mesencephalotomy (> Figure 150-1). In 1942, Walker [4] described a procedure wherein he surgically exposed the mesencephalon and made a pie-shaped incision to interrupt the spinothalamic pathway with good hemianalgesia and usually pain relief. However, it often interrupted the medial lemniscus as well, and produced a feeling of numbness on the entire contralateral side and frequently dysesthesia. The illustration in his article demonstrated that the lesion also involved the brachium of the inferior colliculus, the lateral edge of the periaqueductal grey and the intervening reticular formation (> Figure 150-2). A similar procedure by Drake and McKenzie [5] was marked by similar complications. White and Sweet [6] went so far as to state that the procedure was not acceptable because of the occurrence of severe dysesthesia, which often left the patient worse than before, a high postoperative mortality rate, and the tendency of the analgesia to diminish with time. The search for a better controlled technique to interrupt the primary pain pathways in the mesencephalon came on the scene just a few years after Walker’s report, when in 1947 Spiegel and Wycis [7] introduced stereotactic surgery. Within a year, they reported a stereotactic mesencephalic spinoquintothalamic tractotomy for facial dysesthesia of iatrogenic origin [8] with a resultant pain relief that lasted for at least 18 years, and stereotactic surgery became the modality of choice for attacking the pain pathways within the brain. The medial lemniscus was avoided, and dysesthesia was not produced. The lesion was designed to interrupt both the spinothalamic and quintothalamic tracts, but it probably also encroached on the reticular
2534
150
Mesencephalotomy for cancer pain
. Figure 150-1 Anatomy of the mesencephalon. Tracts involved in mesencephalotomy are labeled on the right
. Figure 150-2 Area sectioned by Walker [4] adapted from his report in 1942, showing the encroachment of his incision in the mesencephalon that includes the medial lemniscus and reticular formation and spinothalamic tract
formation dorsal to the red nucleus as it entered the central grey. In some later procedures, they included a lesion in the dorsomedial nucleus of the thalamus in order to interrupt its connections to the pre-frontal area. Their concept was that the
spinothalamic lesion treated the pain transmission, but the dorsomedial lesion helped the ‘‘suffering’’ component of intractable pain, which is projected via the spinoreticular system to limbic structures [9]. They recognized, in that era of prefrontal lobotomy, that there is frequently an affective component to both pain and movement disorders, so they attempted to blunt that adverse influence as well [10]. Several other authors demonstrated that damage to the medial lemniscus could cause severe dysesthesia. Frank [11] correlated injury to that area with dysesthesia after the type of mesencephalotomy that Walker had described [4]. Colombo [12] recorded sensory evoked potentials before, during, and after stereotactic mesencephalotomy in eight patients to map the extent of the lesion. He concluded that dysesthesia correlated with damage to the medial lemniscus. The search was on to define the ideal mesencephalic lesion. Leksell [13] initially advocated
Mesencephalotomy for cancer pain
including the medial lemniscus along with the spinothalamic tract for cancer pain, but later dropped that advice. Mark [14] advocated combining the mesencephalic lesion with a lesion in the somatosensory thalamus. Roeder and Orthner [15] observed a significant reduction in the affective response to pain when the spinoreticulothalamic area was included in the lesion. In 1954, Spiegel et al. [16,17] demonstrated both in the laboratory and in patients with thalamic syndrome that pain was transmitted via
150
spinoreticular pathways, which had been recognized as being involved with pain transmission (> Figure 150-3). They advocated that the mesencephalotomy lesion extend medially to include spinoreticular areas, but advised caution about the extent of such lesions, especially in patients with a strong emotional component to the chronic pain syndrome [19]. They had incidentally also used dorsomedial nucleus lesions to blunt the emotional effect on symptoms of movement disorders since the first reported sterotactic patients [7],
. Figure 150-3 Illustration from Struppler [18] showing the lateral spinothalamic tract and reticulospinothalamic projection
2535
2536
150
Mesencephalotomy for cancer pain
and advocated its inclusion for treatment of chronic pain. They noted that it was not necessary to include spinothalamic or somatosensory areas to obtain successful pain relief, but advocated lesions in the centrum medianum and intralaminar thalamus [20]. On the other hand, Whisler and Voris [21] freely included the spinoreticular pathway in patients with cancer, and reported good results even with large bilateral lesions. In a move to define the physiology and anatomy of chronic pain, it was necessary to study the mesencephalon in awake human patients. In 1969, Nashold et al. [22] inserted electrodes into putative midbrain targets and stimulated over a several week period prior to making a mesencephalotomy lesion. They verified that the somatotopic organization of the spinothalamic and quintothalamic tracts extended from lateral to medial, with the thorax and abdomen represented medially near the central grey. Although low frequency stimulation might lessen the pain, high frequency stimulation in the central grey produced unpleasant sensations involving midline structures, as well as strong negative or fearful emotional reactions, sometimes hyperventilation, a feeling of panic, involuntary verbalization, and flushing with slowed pulse and respiration, which has since been attributed to projections to the hypothalamus [16,23]. Stimulation at the level of the superior colliculus produced sensation in the face, arm, chest and trunk. Stimulation at the level of the inferior colliculus produced sensation in just the arm and face. Complex ocular movements, sometimes unilateral, defined encroachment on the emerging oculomotor fibers, areas to be avoided. Stimulation of the spinothalamic tract produced contralateral burning pain, numbness, or a cold sensation. They localized their ideal target to be between 5 and 10 mm lateral to the midline [24], medial to their original site, which encroached on the medial reticular area and lateral edge of the central grey (> Figure 150-4). The result was improved relief of intractable pain with less sensory loss and no
. Figure 150-4 The two areas defined by Nashold [22,25]. The medial area is where he recommended lesioning which overlaps the central grey and involves the spinoreticular area. Stimulation produces diffuse pain in the center of the body with a strong fearful emotional reaction, often coinciding with the area of denervation pain for which mesencephalotomy is being done. The lateral area is where stimulation produces pain and numbness of the contralateral face, arm, chest or leg
dysesthesia, documenting that pain relief could occur without significant impingement on the spinothalamic pathway [26]. In addition, the pain relief was often bilateral, confirming the concept that the midbrain reticular system projects bilaterally to the thalamus. Shieff and Nashold [27] concluded that chronic pain is invariably associated with emotional distress, which is not purely a secondary psychological phenomenon but involves intrinsic neural changes. The interruption of the extralemniscal pathway alleviated the emotional aspect of pain and consequently the pain perception itself, even if the spinothalamic pathway was not included in the lesion (> Figure 150-3). Thus, the search for the ideal lesion that began in the spinothalamic pathway led to an area that left that pathway in tact. Tasker [28] performed extensive mapping on stimulation of the mesencephalon in awake patents during stereotactic surgery. He found the medial lemniscus to lie 10–12 mm lateral and the reticulospinothalamic tract 7–9 mm lateral. Stimulation of the reticulothalamic tract 5–7 mm
Mesencephalotomy for cancer pain
lateral produced no sensation unless the current was large enough to stimulate the periaqueductal grey, when bizarre emotional responses were seen. Production of lesions at midbrain levels may encroach on emerging oculomotor fibers to produce disturbance of ocular motility, which can be significantly disabling. Spiegel and Wycis [9] reported an incidence as high as 50% at their original campotomy lesion site at the level of the superior colliculus, although that was in conjunction with lesions intended to be placed in Forel’s field. Nashold [27,29] as well as Amano [30] recommended a target at the level of the inferior colliculus, 5 mm below the intercommissural plane, which significantly alleviated that complication. Thus, mesencephalotomy has come full circle. The first attempt, sectioning the lateral spinothalamic pathway by making an incision in the peduncle was fraught with unacceptable side effects. When stereotactic surgery became available, it became possible to interrupt that initial target with greater precision. When a search was made for the ideal localization, it became evident that the procedure worked better if the lesion intruded on the multisynaptic reticulospinal pathway, which led to the realization that enhanced benefit and fewer side effects could be obtained by limiting the lesion to the multisynaptic area, sparing the spinothalamic tract which had been the original target.
Mesencephalotomy for Cancer or Denervation Pain Although lesions in or just above the mesencephalic area had been made for motor disorders, this discussion will include only management of cancer and central denervation pain. It has become evident that interruption of pain pathways for persistent pain or chronic pain of benign origin might provide some temporary pain relief,
150
the specific indications are few and the pain almost invariably returns, so mesencephalotomy is rarely indicated for pain other than cancer pain and thalamic syndrome. If there is a large component of suffering, either physical or emotional, as there often is in patients with extensive facial or pharyngeal cancer, the mesencephalic lesion may be coupled with an addition lesion in the dorsomedial, intralaminar, basal thalamus [20], or cingulum [31] which may provide enhanced comfort. If there is considerable neuropathic pain, as in Pancoast syndrome, Nashold [32] reported that patients who had been taking large doses of narcotics for their neurogenic pain syndromes, such as Pancoast syndrome, may be withdrawn from narcotics abruptly after obtaining pain relief from mesencephalotomy without precipitating narcotic withdrawal symptoms. I have also found an intralaminar thalamotomy without the mesencephalic lesion may provide exceptional pain relief, and high doses of narcotics can be discontinued without withdrawal symptoms [33–35], which may be related to the observation that centrum medianum lesions in self-administering addicted rats may change withdrawal seizure threshold and drug seeking behavior [36]. Mesencephalotomy has been used for pain of central origin, as in a stroke involving the ascending pain pathways within the medulla, midbrain or thalamus, the so-called lateral medullary syndrome, as well as stroke causing severe pain. Nashold [25] obtained significant relief of thalamic syndrome and related pain with mesencephalotomy. It is interesting to note that it had considerable benefit whether the primary pathology lay above or below the mesencephalon, suggesting that the pain was driven by interference with the spinothalamic of spinoreticulothalamic input to the thalamus. Nashold also obtained relief of severe postherpetic facial or upper extremity pain by mesencephalotomy, but cautioned that it should be used only if the pain were severe enough to
2537
2538
150
Mesencephalotomy for cancer pain
justify the risk. Similarly, facial dysesthesia following denervation after surgery for trigeminal neuralgia may be improved. One must caution, however, that so-called ‘‘atypical facial neuralgia’’ is not likely to benefit from any surgical procedure, since it appears that this ill-defined syndrome may be a distortion of somatosensory sensation rather than a sensation that involves pain perception per se. The primary indication for mesencephalotomy is management of cancer pain involving the head, neck, or upper extremities when all else has failed. If an ablative procedure is required for pain below those levels, anterolateral [37] or limited [38] cordotomy is usually the recommended technique. As in any stereotactic procedure, the most important consideration for mesencephalotomy is patient selection. Its use in cancer pain should be reserved until all non-invasive modalities have failed, and only when there is good correlation between the site of cancer involvement and the generation of pain. Mesencephalotomy may be particularly helpful in patients who have persistent pain along with intense psychological generation of the suffering that often accompanies cancer pain of the head, face, and neck. Surgery alone is often not sufficient to deal with these emotional components of having an intensely painful terminal illness, and postoperative psychotropic medication or counseling may still be indicated. It is helpful to consider that the pain is generated by cancer involving the tissues, whereas the suffering may be a secondary but still significant part of the cancer pain syndrome. Consequently treatment should be directed to both components of the syndrome. From the anatomical standpoint one may consider interruption of the primary pain pathways as pain management, but interruption of the spinoreticulothalamic pathways as management of suffering. It must be recognized, with that consideration, that it is often possible to control the pain without interrupting the primary pain pathway, but with a lesion interrupting the system concerned with
suffering. Consequently, if psychological factors are the overwhelming component, the patient may benefit from either a mesencephalotomy [26] or an intralaminar or basal thalamotomy [33], or even a cingulotomy [39]. The target for mesencephalotomy is not a specific anatomical structure that can be identified on MRI. The classical target, actually an amalgam of the historical target points of several authors, who used the intercommissural line as their reference landmark, is 5 mm behind the posterior commissure, 5–10 mm lateral, and 5 mm below the intercommissural plane [9,15,16,25,29,32]. Intraoperative stimulation is done at several frequencies to identify physiologically the localization of the electrode [32]. The surgeon looks particularly for responses that indicate stimulation of surrounding structures, particularly extraocular movements that indicate that the electrode is too high and may involve the oculomotor fibers, fear or emotional outburst that may indicate that the electrode is too medial, or projection of spinothalamic sensation to the contralateral body that indicates that the electrode is too lateral. With the electrode properly placed, the patient may exhibit a muted emotional response, and sensation generally projected to the central core of the body. Microelectrode recording, as reported by Amano [40], demonstrates neurons n the target area responding only to peripheral pin prick stimulation with a latency of 250–1,000 ms. There is ordinarily no discernible analgesia on pin stick testing, but patients may report instant pain relief. If the pain is bilateral, the lesion is made contralateral to the more severe pain, since many patients experience bilateral relief of pain. If necessary, a contralateral lesion may be made three weeks later, in order to evaluate whether the patient has had delayed relief. The results of mesencephalotomy for cancer pain are included in several literature reviews by Nashold [32], Tasker [41], and Gybels and Sweet [42], that include persistent pain of benign origin, as well. These three authors have reviewed essentially the same series [12,19,21,26,43–49], but
Mesencephalotomy for cancer pain
results are most clearly tabulated in the book by Gybels and Sweet [42]. The authors reported considerable success in a total of 270 patients treated with mesencephalotomy for cancer pain. In all, 86% of patients had significant pain relief, usually for the remaining life of the patient. Mortality rates for the entire series varied from 1.8 to 8%, but it was not always clear which patients may have died from their disease and which from a surgical complication. (In my experience, there is a group of terminal cancer patients who just go to sleep after being successfully relieved of their pain and die within days.) Oculomotor dysfunction was seen in 13–20% of patients, even with intraoperative stimulation control and directing the lesion to the safer more caudal target. Dysesthesia was reported in 15–21%. One additional study by Laitinen [50] reported as complications 15.8% ‘‘lemniscal damage’’ with 42.1% dysesthesia, but it is not clear why this complication rate was greater than other authors, except perhaps for stricter criteria for dysesthesia. There have been relatively few recent reports on series of mesencephalotomy, the most recent by Amano [51] who encourages neurosurgeons to consider mesencephalotomy and reports an optimistic result in both cancer and non-cancer pain, with no mortality and long-lasting relief in some patients. The present status of mesencephalotomy is that it can be a valuable procedure in management of cancer pain or central denervation pain. It is necessary to select patients carefully, since the procedure is not without danger. Only patients with pain too high to be treated with cordotomy or a morphine pump should be considered. Many neurosurgeons, including myself, prefer to use a target in the intralaminar thalamus, which may have equal benefit but less risk, particularly the risk of extraocular palsy. If there is a strong somatic component of ‘‘pain’’ and relatively little ‘‘suffering,’’ mesencephalotomy may be beneficial in cancer pain, as well as central denervation pain.
150
References 1. Spiller WG. The location within the spinal cord of the fibers for temperature and pain sensations. J Nerv Ment Dis 1905;32:318-20. 2. Spiller WG, Martin E. The treatment of persistent pain of organic origin in the lower part of the body by division of the anterolateral column of the spinal cord. J Am Med Assoc 1912;58:1489-90. 3. Dogliotti M. First surgical sections, in man, of the lemniscal lateralis (pain-temperature path) at the brain stem, for the treatment of diffuse rebellious pain. Anesth Analg 1938;17:143-5. 4. Walker AE. Relief of pain by mesencephalic tractotomy. Arch Neurol Psychiatry 1942;48:865-83. 5. Drake CG, McKenzie KG. Mesencephalic tractotomy for pain. Experience with six cases. J Neurosurg 1953;10: 457-62. 6. White JC, Sweet WH. Pain, Its Mechanism and Neurosurgical Control. Springfield: Charles C Thomas; 1955. 7. Spiegel EA, Wycis HT, Marks M, Lee AS. Stereotaxic apparatus for operations on the human brain. Science 1947;106:349-50. 8. Spiegel EA, Wycis HT. Present status of stereoencephalotomies for pain relief. Confin Neurol 1966;27:7-17. 9. Spiegel EA, Wycis HT. Mesencephalotomy in treatment of intractable facial pain. AMA Arch Neurol Psychiatry 1953;69:1-13. 10. Spiegel EA, Wycis HT, Baird HW. Effect of thalamic and pallidal lesions upon involuntary movements in choreoathetosis. Trans Am Neurol Assoc 1950;75:234. 11. Frank F, Tognetti F, Gaist G, Frank G, Galassi E, Sturiale C. Stereotaxic rostral mesencephalotomy in treatment of malignant faciothoracobrachial pain syndromes. A survey of 14 treated patients. J Neurosurg 1982;56:807-11. 12. Colombo F. Somatosensory-evoked potentials after mesencephalic tractotomy for pain syndromes. Neuroradiologic and clinical correlations. Surg Neurol 1984;21: 453-8. 13. Leksell L. Gezielte Hirnoperationen. In: Hassler R, Riechert T, editors. Handbuch der Neurochirurgie. Berlin: Springer; 1957. p. 178-92. 14. Mark VH, Ervin FR, Hackett TP. Clinical aspects of stereotactic thalamotomy in the human. Arch Neurol 1960;3:17-32. 15. Roeder F, Orthner H. Uber zentrale Schmerzoperationen, insbesondere mediale Mesencephalotomie bei thalamischer Hyperpathie u. bei Anaesthesia doloroa. Confin Neurol 1961;21:51-67. 16. Spiegel EA, Kletzkin M, Szekely EG, Wycis HT. Pain reactions upon stimulation of the tectum mesencephali. J Neuropathol Exp Neurol 1954;13:212-20. 17. Spiegel EA, Kletzkin M, Szekely EG, Wycis HT. Role of hypothalamic mechanisms in thalamic pain. Neurology 1954;4:739-45.
2539
2540
150
Mesencephalotomy for cancer pain
18. Struppler A. Clinical pain syndromes [German]. Med Klin (Munich) 1966;61:862-3. 19. Wycis HT, Spiegel EA. Long-range results in the treatment of intractable pain by stereotaxic midbrain surgery. J Neurosurg 1962;19:101-7. 20. Spiegel EA, Wycis HT, Szekely EG, Gildenberg PL. Medial and basal thalamotomy in so-called intractable pain. In: Knighton RS, Dumke PR, editors. Pain. Boston: Little Brown; 1966. p. 503-17. 21. Whisler WW, Voris HC. Mesencephalotomy for intractable pain due to malignant disease. Appl Neurophysiol 1978;41:52-6. 22. Nashold BS Jr, Wilson WP, Slaughter DG. Sensations evoked by stimulation in the midbrain of man. J Neurosurg 1969;30:14-24. 23. Spiegel EA, Wycis HT. The central mechanism of emotions. Am J Psychiatry 1951;109:426-31. 24. Nashold BS Jr, Wilson WP. Central pain. Observations in man with chronic implanted electrodes in the midbrain tegmentum. Confin Neurol 1966;27:30-44. 25. Nashold BSJ, Wilson WP, Slaughter DG. Stereotaxic midbrain lesions for central dysesthesia and phantom pain. Preliminary report. J Neurosurg 1969;30:116-26. 26. Nashold BS Jr. Extensive cephalic and oral pain relieved by midbrain tractotomy. Confin Neurol 1972;34:382-8. 27. Shieff C, Nashold BSJ. Stereotactic mesencephalotomy. Neurosurg Clin N Am 1990;1:825-39. 28. Tasker RR. Identification of pain processing systems by electrical stimulation of the brain. Hum Neurobiol 1982;1:261-72. 29. Shieff C, Nashold BSJ. Stereotactic mesencephalic tractotomy for the relief of thalamic pain. Br J Neurosurg 1987;1:305-10. 30. Amano K. Destructive central lesions for persistent pain. Outcome. In: Gildenberg PL, Tasker RR, editors. Textbook of stereotactic and functional neurosurgery. New York: McGraw-Hill; 1998. p. 1425-9. 31. Turnbull IM. Cingulotomy. In: Voris HC, Whistler WW, editors. Treatment of pain. Springfield: Charles C Thomas; 1975. p. 143-50. 32. Nashold BS Jr. Brainstem stereotaxic procedures. In: Schaltenbrand G, Walker AE, editors. Stereotaxy of the Human Brain. Stuttgart, NY; Georg Thieme Verlag;1982. p. 475-83. 33. Spiegel EA, Wycis HT, Szekely EG, Gildenberg P, Zanes C. Combined dorsomedial, intralaminar and basal thalamotomy for relief of so-called intractable pain. J Int Coll Surg 1964;42:160-8. 34. Gildenberg PL, Frost EAM. The effect of neurological disease on narcotic actions. In: Adler MW, Manara L, Samanin R, editors. Factors affecting the action of narcotics. New York: Raven Press; 1978. p. 703-16.
35. Gildenberg PL. Stereotactic treatment of head and neck pain. Res Clin Stud Headache 1978;5:102-21. 36. Adler MW, Lin C, Smith KP, Tresky R, Gildenberg PL. Lowered seizure threshold as a part of the narcotic abstinence syndrome in rats. Psychopharmacologia 1974;35:243-7. 37. Gildenberg PL. Percutaneous cervical cordotomy. Clin Neurosurg 1974;21:246-56. 38. Gildenberg PL, Hirshberg RM. Limited myelotomy for the treatment of intractable cancer pain. J Neurol Neurosurg Psychiatry 1984;47:94-6. 39. Hurt RW, Ballentine HT Jr. Stereotactic anterior cingulate lesions for persistent pain: a report of 68 cases. Clin Neurosurg 1974;21:334-51. 40. Amano K, Tanikawa T, Iseki H, Kawabatake H, Notani M, Kawamura H, et al. Single neuron analysis of the human midbrain tegmentum. Rostral mecencephalic reticulotomy for pain relief. Appl Neurophysiol 1978;41:66-78. 41. Tasker RR. Stereotactic surgery. In: Wall PD, Melzack R, editors. Textbook of pain. Edinburgh: Churchill Livingstone; 1994. p. 1137-57. 42. Gybels JM, Sweet WH. Neurosurgical treatment of persistent pain. Pain and headache, vol 11. Basel: Karger; 1989. 43. Whisler WW, Voris HC. Mesencephalotomy for intractable pain due to malignant disease. Appl Neurophysiol 1978;41:52-6. 44. Schvarcz JR. Paraqueductal mesencephalotomy for facial central pain. In: Sweet WH, editor. Neurosurgical treatment in psychiatry, pain, and epilepsy. Baltimore: University Park Press; 1977. p. 661-7. 45. Amano K, Kawamura H, Tanikawa T, Kawabatake H, Notani M, Iseki H, et al. Long-term follow-up study of rostral mesencephalic reticulotomy for pain relief – report of 34 cases. Appl Neurophysiol 1986;49:105-11. 46. Shieff C, Nashold BS Jr. Mesencephalotomy for thalamic pain. Neurol Res 1987;9:101-4. 47. Helfand MH, Leksell L, Strang RR. Experiences with intractable pain treated by stereotaxic mesencephalotomy. Acta Chir Scand 1965;129:573-80. 48. Mazars G. Etat actuel de la chirurgie de la douleur. Neurochirurgie 1976;22 Suppl 1:1-158. 49. Frank F, Sturiale C, Gaist G, Fabrizi A, Frank Ricci R. Stereotactic mesencephalic tractotomy in the treatment of Pancoast syndrome. Appl Neurophysiol 1985;48:274-6. 50. Laitinen LV. Mesencephalotomy and thalamotomy for chronic pain. In: Lunsford LD, editor. Modern stereotactic neurosurgery. Boston: Martinus Nijhoff; 1988. p. 269-77. 51. Amano K, Kawamura H, Tanikawa T, Kawabatake H, Iseki H, Taira T. Stereotactic mesencephalotomy for pain relief. A plea for stereotactic surgery. Stereotact Funct Neurosurg 1992;59:25-32.
144 Microvascular Decompression for Trigeminal Neuralgia S. Sup Chung
Although there are various surgical methods for the treatment of trigeminal neuralgia (TN), microvascular decompression (MVD) is a widely accepted method. Dandy [1,2] first noted frequent compressing lesions, such as tumors, arteries or veins crossing the trigeminal nerve root, while he was doing partial rhizotomy in the posterior cranial fossa for the treatment of trigeminal neuralgia. Dandy [3] analyzed 215 cases of trigeminal neuralgia who were treated by the posterior fossa approach and found an artery (66 cases or 30.7%), a petrosal vein (30 cases or 14.0%), and other pathological lesions (32 cases or 14.9%) were in contact with the trigeminal nerve root. About 45% of the cases had contact with vessels, while 87 cases (~40%) had no gross abnormal findings. Based on his findings, Dandy concluded that some extrinsic lesions to the sensory root were usually the cause of trigeminal neuralgia. Gardner and Miklos [4] explored the posterior fossa in patients who had recurrent neuralgic pain after operation on the middle cranial fossa. They found an anomalous arterial loop lying against the nerve. The pain was completely relieved after separating this vessel from the nerve root by the interposition of a piece of gelfoam. Another patient who had a large menigioma compressing the nerve root was relieved of pain after the removal of tumor. Based on these findings, Gardner and Miklos concluded that the cause of TN was related to a sensory root within the middle or posterior fossa. Gardner [5,6] continued with further investigations and explored the cerebellopontine angle #
Springer-Verlag Berlin/Heidelberg 2009
in 18 patients who had recurrent neuralgic pain after middle fossa operation. Arterial loop compressed the trigeminal nerve in six cases, while another six cases had abnormal lesions such as acoustic tumors, cirsoid aneurysms of the basilar artery, etc. No explanation could be ascertained for the remaining six cases, probably indicating no abnormal findings. Gardner concluded that compression to the nerve root causes demyelination and a short-circuiting phenomenon between touch and pain fibers, ultimately resulting in neuralgic pain. Kerr [7] found severe demyelination in the trigeminal nerve root of patients with TN using light and electron microscopes. He suggested that localized demyelinateion can produce neural short-circuitings or ephaptic transmissions, which can produce paroxysmal pain. Hilton et al. [8] histopathologically confirmed the hypothesis of ephaptic transmission in the pathogenesis of trigeminal neuralgia related to vascular compression. Using a surgical microscope, Jannetta [9,10] was able to observe more abnormalities than Dandy and Gardner’s macroscopic analysis. Jannetta ascertained that in most cases the offenders were blood vessels and many of them were aberrant arteries. Jannetta suggested that the offending blood vessels compressed the root entry zone at the pons, and that decompression of the trigeminal root at the root entry zone could produce permanent relief of TN without destruction of the neural tissue. Jannetta standardized the procedure, and MVD for TN became a treatment of choice. MVD is a nondestructive procedure
2466
144
Microvascular decompression for trigeminal neuralgia
which deals with the underlying causes of the condition. The procedure has the least impact on sensory changes and inducement of dysesthesia on the face than other procedures, while also having higher incidence of long-term relief and low recurrence rate of pain.
Patient Selection Patients who are refractory to medical treatment or who have to stop using medications due to side effects can be candidates for surgery. The selection criteria of appropriate candidates for various treatment options are important procedures. A successful outcome depends on a well-considered selection of patients with typical neuralgia. With a careful assessment of their history, combined with physical examination, diagnosis of typical trigeminal neuralgia can be easily established. At first the pain is reported to be unilateral, brief, paroxysmal, and electric shock-like or lancinating pain in one or more distributions of the trigeminal nerve. In many cases, there is a trigger point around the mouth. But in a chronic state, the neuralgic pain usually lasts longer, is non-lancinating, and sometimes reported to be a constant burning pain. Some investigators have referred to this condition as transitional trigeminal neuralgia or atypical trigeminal neuralgia, which can also be treated with MVD [11,12]. Atypical facial pain should be differentiated from the trigeminal neuralgia or atypical trigeminal neuralgia. Atypical facial pain is a poorly localized burning, aching or throbbing pain in the deep portion of face. Atypical facial pain is not a candidate for surgical therapy. One interesting contrasting observation is that patients with atypical facial pain press their face with hands which help their suffering, whereas patient with typical trigeminal neuralgia cannot do that. Current treatment modalities include percutaneous radiofrequency rhizotomies, percutaneous glycerol rhizotomies, trigeminal balloon
microcompressions, stereotactic radiosurgeries, and MVDs. The selection of a particular treatment modality is decided by several factors, including medical condition, the age of the patient, history of prior surgery, severity of facial pain, and patient’s preference. To be a candidate for the decompressive procedure, the patient’s age is preferably <70, have good general health to be able to tolerate a small craniectomy under general anesthesia, and also be willing to accept the side effects of surgery. Percutaneous procedures leave patients with some degree of sensory loss and dysesthesia on the face, while recurrence rate is high. Radiosurgery appears to be a valuable addition to the existing treatment modalities. However, for patients who have extremely severe neuralgic pain radiosurgery is not recommended as sometimes it takes long time to control the pain. Furthermore, we need additional information on long-term results and possible late complications of the radiosurgery. MVD is perhaps the procedure of choice for relatively young and healthy patients. However, elderly patients who have history of failed percutaneous surgery can also be a candidate for MVD. There have been cases where multiple sclerosis, brain-stem infarction, or inflammation at the level of the trigeminal root entry zone has shown typical pain resembling trigeminal neuralgia. In these cases percutaneous radiofrequency trigeminal rhizotomy, percutaneous glycerol rhizotomy, or stereotactic radiosurgery is better indicated rather than MVD [13–18]. The involvement of ophthalmic division varies from 13.5 to 32.8% (> Table 144-1) [19–22]. When patients have neuralgic pain on the ophthalmic division, MVD is recommended. Hypalgesia or analgesia of the cornea after ablative procedures may cause keratitis or blindness of the ipsilateral eye. The incidence of hearing loss after MVD is low, but we should be very careful
Microvascular decompression for trigeminal neuralgia
144
. Table 144-1 Distribution of pain Pain distribution (%) V1 only V2 only V3 only V1, V2 V2, V3 V1, V2, V3
Apfelbaum [19]
Barker [20]
Chang [21]
Sindou [22]
0.7 20.4 26.9 11.8 31.5 8.7
3 18 15 17 36 12
1 21.2 21.2 8.7 44.2 3.8
2.7 21.3 15.2 22.1 30.7 8
when patients exhibit opposite side hearing difficulty. Manipulation of dolichoectatic vessels and traction of small perforators during MVD is dangerous and should be avoided. Therefore, MVD is not recommended in severe cases of vertebrobasilar dolichoectasia [21,23].
images, more vascular contact at the root entry zone can be detected with CISS imaging, including small arteries or veins, helping surgeons to see compressing vessels before surgery [21,24]. If a dolichoectasia of vertebrobasilar system is detected, a percutaneous procedure is recommended in place of MVD in patient with severe degree of dolichoectasia (> Figure 144-2).
Investigation Operative Technique Preoperative audiometry (pure tone and speech) is necessary as part of a baseline study of possible postoperative hearing impairment. Preoperative brain stem auditory evoked potentials (BAEP) is also necessary as baseline information of intraoperative BAEP monitoring, which is helpful in preventing excessive traction of acoustic nerve root, cerebellum, and brain stem, and resulting hearing impairment and injury to the cerebellum and brain stem. Magnetic resonance imaging and angiography (MRI and MRA) can detect tumors, plaques of multiple sclerosis, vascular anomalies, or cerebral aneurysms, while unexpected bleeding of cerebral aneurysm or vascular anomalies can be prevented during or after the operation. Vascular contact at the trigeminal root entry zone can be seen on the three-dimensional timeof-flight MRA (3D-TOF MRA) image and constructive interference in steady-state (CISS) image (> Figure 144-1). Compared with 3D-TOF
After general endotracheal anesthesia, a threepoint head-fixation device is employed, and the patient is placed in the lateral decubitus position on the operating table. Pressure points are secured by axillary roll, pillow, and pads. The head is slightly tilted away from the operating side and flexed to the sternum, while vertex of the head is positioned parallel to the floor. The patient is secured to the bed with straps and the shoulder is pull down caudally with tape to have maximum working room. The electrodes for BAEP are fixed for intraoperative monitoring. A 6–7 cm scalp incision is made just posterior to the hairline, with one-third of the incision above the inion-meatal line and two-thirds below (> Figure 144-3). Soft tissue is dissected with an electrical scalpel and mastoid eminence is exposed. The emissary veins are coagulated and sealed with bone wax. A small burr hole is made just posterior to the mastoid air cell and extended to expose the transverse and
2467
2468
144
Microvascular decompression for trigeminal neuralgia
. Figure 144-1 (a) Axial image of three-dimensional time-of-flight (3D-TOF) image showing arterial contact to trigeminal nerve root (long arrow: artery, short arrow: nerve root). (b) Axial constructive interference in steady-state (CISS) image showing the vascular contact to trigeminal nerve root (long arrow: vessel, short arrow: nerve root)
. Figure 144-2 3D-TOF MRA images in a 63-year-old man with severe dolichoectactic vertebrobasilar artery. Neuralgic pain persisted after microvascular decompression (MVD). Hearing loss and facial palsy developed after MVD. The pain relieved after stereotactic gamma knife radiosurgery. (a) Axial 3D-TOF image shows dolichoectactic vertebrobasilar artery compressing the nerve root and brain stem (long arrow: vertebrobasilar artery, short arrow: nerve root). (b) MRA demonstrates the dolichoectactic vertebrobasilar artery
sigmoid sinuses. The size of the craniectomy is a rectangular shape about 2.5 3.5 cm size, while the longer edge is along the sigmoid sinus. Mastoid air cell is sealed with bone wax. Dura mater is incised in cruciate form, whose two bases are transverse and sigmoid sinuses (> Figure 144-3b). A cottonoid is placed over cerebellum while the CSF is gently sucked out. The surgical microscope is introduced at this point. When the cerebellum becomes slack, the
microsurgery spatula is inserted over the cottonoid. The cerebellum is retracted gently inferomedially, and the tentorium and acoustic nerve root are identified. Direction of retraction is to acoustic nerve side to prevent excessive traction of the nerve. The cerebellum becomes more slack after puncture of the trigeminal cistern. If the superficial branch of the petrosal vein obstructs the vision, it can be coagulated and cut. However, no arterial branch should be cut, regardless of
Microvascular decompression for trigeminal neuralgia
144
. Figure 144-3 (a) The head is slightly tilted away from the operating side and flexed to the sternum, and vertex of the head is positioned parallel to the floor. The shoulder is pulled down caudally with tape to have maximum working room (a: inion-meatal line, b: line of scalp incision). (b) A rectangular shape retromastoid craniectomy (a: transverse sinus, b: sigmoid sinus, c: cruciate form dura incision). (c) Cerebellum is retracted over the cottonoid (a: cerebellum, b: brain stem, c: trigeminal nerve root, d: acoustic nerve root, e: SCA, f: AICA). (d) Arachnoid over the nerve root dissected and the nerve root become free (a: SCA is pulled away from nerve root to tentorium using Teflon felt. b: AICA is pulled away from nerve root to the side of internal acoustic meatus using Teflon felt)
size. The arachnoid around the trigeminal root should be dissected in order to make it free. The compressing artery is pulled away from the trigeminal root using decompressive prosthesis, Teflon felt. In case of the superior cerebellar artery, it is pulled away from nerve root to tentorium using Teflon felt. If the vessel is the anterior inferior cerebellar artery, the artery is pulled away to the acoustic nerve side. The Teflon felt is fixed to the tentorium or dura around internal acoustic meatus using fibrin glue (> Figure 144-3c and > Figure 144-3d). If the compressing vessel is a vein, then coagulate the vein and cut. If there is no compressing vessel, approximately one-third of the
root is cut in the caudolateral part for the lower facial pain. Dura mater is closed watertightly using muscle pieces or fibrin glue after meticulous hemostasis around the root, brain stem, and cerebellum. Muscles, subcutaneous tissues, and scalp are closed as usual procedures.
Surgical Results The reported postoperative results of pain relief after MVD has been summarized from a number of studies in > Table 144-2. > Table 144-2 shows the variable rate of pain relief, follow up period,
2469
2470
144
Microvascular decompression for trigeminal neuralgia
. Table 144-2 Surgical outcomes after microvascular decompression for trigeminal neuralgia Initial results (%) Reference Apfelbaum [19] Bederson [25] Klun [14] Rath [26] Barker [20] Kondo [27] Broggi [28] Chang [21] Tyler-Kabara [12] Pamir [29] Sindou [22]
No. of patients 289 252 178 135 1155 281 141 104 1739 90 362
Long-term results (%)
Excellent
Good
Failure
Excellent
Good
Recurrence
94.1 NS 96.1 80 82 NS 85 85.6 80.3 85.6 85.6
4.0 NS 2.8 10.4 16 94.8a 7.5 11.5 16.5 11.1 3
1.8 NS 1.1 9.6 2 5.2 7.5 2.9 3.2 3.3 11.4
70.4 75 93.8 NS 64 NS 75.2 79.8 73.7 NS 80
18.7 8 NS 73.3a 4 86.6a 9.2 11.5 6.8 85a 4.9
9.1 12 6.2 17.1 30 8.2 7.8 5.8 16.3 11.7 15.1
Mean follow-up (yr) 4.6 5.1 5.2 2.5 6.2 9.2 3.2 5.7 11.3 5 8
Excellent, pain-free; good, infrequent pain controlled with medication; failure, little or no improvement after MVD. NS, not specified a Good to excellent
. Table 144-3 Compressing vessels in operative fields Vessel (%) SCA AICA SCA + AICA PICA VB Unidentified artery Vein and artery Vein only No compression Tumor
Apfelbaum [19]
Barker [20]
Chang [21]
Sindou [22]
64 6.6 8 0 1.4 0 19 11.8 3.5 3.8b
75 10 NS 1 3 15 68 13 NS NS
41.3 10.6 6.7 0 5.8 a NS 17.4 3.8 8.7 5.8
74.3 6.1 16.3 0 5.1a NS 26.5 3.3 4.7 1.4c
SCA, superior cerebellar artery; AICA, anterior inferior cerebellar artery; VB, vertebrobasilar artery; PICA, posterior inferior cerebellar artery; NS, not specified a dolichoectatic VB b AVM; 0.3% + tumor; 3.5% c Tumor, AVM, etc
and recurrence rates [12,14,19–22,25–29]. Immediate postoperative pain relief was excellent (complete relief) in 80–96.1% and good (partial relief) in 2.8–16.5%, while significant pain relief (complete + partial relief) was exhibited in 88.6– 98.9% of patients. Initial failure rate was 1.1– 11.4%. Long-term pain relief was excellent in 64–93.8%, and good in 4–18.7%, while significant pain relief was reported by 68–91.3% of
patients. Average follow-up duration ranged from 2.5 to 11.3 years. As noted in > Table 144-3, in the operating procedure, superior cerebellar artery alone was the most common compressing vessel to the trigeminal nerve root (41.3–75%). Next common compressing vessel is the anterior inferior cerebellar artery. The vertebral or basilar artery compression is 1.4–5.8%, venous compression
Microvascular decompression for trigeminal neuralgia
is 3.3–11.8%, and arterial and venous mixed compression is 17.4–68%, while no vascular compression ranges from 3.5 to 8.7% [19–22]. There have been attempts to establish factors affecting the outcomes of MVD for TN. Investigators analyzed factors such as age, sex, site of pain, symptom duration, compressing vessels, preoperative destructive lesioning, bilaterality of pain, and presence of trigger points [12,14,19– 21,30–34]. Investigators generally agree that several combined factors, such as presence of trigger point, kinds of compressing vessels, and symptom duration affect the prognosis of pain relief in trigeminal neuralgia. Tyler-Kabara et al. [12] reported that patients who have trigger points have excellent immediate postoperative and long-term outcome of pain relief. Many patients with typical neuralgia exhibit trigger points, calculated to be up to 85.9%. Therefore, the presence of trigger points indicates typical trigeminal neuralgia, which subsequently means that their chances of a better outcome are higher than patients with atypical trigeminal neuralgia [12,31]. The type of vascular compression may also play a role in the recurrence of TN following MVD. In the cases of arterial cross-compression, they have more long-term complete pain relief and less recurrence [30,31]. It has also been reported that venous compression is a sign of high recurrence rate [8,12,14,30,33,34]. It is postulated that neighboring small veins have enlarged after interruption of the compressing vein, recollateralization, or that interrupted veins recanalized again after the surgery [34,35]. Jannetta and Bissonette [35] reported that the recanalization of vein occurs commonly at about 4 months after the operation. Although dolichoectactic vertebrobasilar artery is not a common cause of neuralgic pain, a number of cases have been reported to be successfully treated by MVD [36–39]. However, severe dolichoectasia is a serious disorder which can lead
144
to occlusive strokes or cerebral hemorrhages, and manipulation of ecstatic vessel and tense perforators can be hazardous [40,41]. There are reports recommending cautious approach with patients whose neuralgic pain may be caused by the dolichoectatic vertebrobasilar system [21,23]. Trigeminal neuralgia without compressing vessels is well-known and the incidence varies from 3.5 to 8.7% (> Table 144-3) [19–22]. Generally in patients having no visible compressing vessels, partial sensory rhizotomy is performed. However, recurrence rate is high. Klun [14] reported a 49% recurrence rate within 5.2 years after the operation. In contrast, the long-term recurrenceratewas6%inpatientswithcompressing vessels. In patients without vascular compression, MVD without partial sensory rhizotomy can relieve the neuralgic pain; however, long-term recurrence rate is also high [42]. The patients with long duration of neuralgic symptom before MVD may have poor outcome. Barker et al. [20] and Bederson and Wilson [25] reported that the patients who have neuralgic pain >8 years may have higher recurrence rate after MVD. They postulated that the patients with long-standing extrinsic compression might develop intrinsic lesion and resulting poor outcome. This suggests that the success of MVD depends on the reversibility of the dysfunction caused by arterial compression of the nerve root [25]. There are reports on scarred Teflon felts cause recurrent neuralgic pain and the pain relieved by removal of the scarred decompressive material [43–45]. Rate of recurrence varies widely, 5.8–30%, in long-term follow-up (> Table 144-2). A 2–3.5% annual recurrence rate of trigeminal neuralgic pain is reported after MVD, while the majority of recurrent neuralgic pain occurs within 2 years after the operation [25,31,33,46]. Kolluri and Heros [30] reported 75% and Mendoza and Illingworth [47] reported 90% of recurrence occurred within 2 years of MVD. It appears that patients who remain pain-free for 2 years after
2471
2472
144
Microvascular decompression for trigeminal neuralgia
surgery have a relatively high probability of remaining pain-free [30]. Therefore, good candidates for MVD are relatively young and healthy patients exhibiting obvious offending artery on magnetic resonance images.
Mortality in these cases has been attributed to unexpected intracranial bleeding, trauma, or infarction of the cerebellum and brain stem. However, mortality rates have also decreased since the introduction of MRI and MRA and BAEP monitoring, in conjunction with improvement of surgical techniques. In very rare occasions, packed decompressing materials cause kinking of the arteries around the trigeminal root that results in infarction of cerebellum and brain stem, which ultimately leads to mortality. Current mortality rate ranges from 0 to 1.43%. > Table 144-4 [12,14,19–22,25] shows the changes of incidences of morbidity and mortality. Facial sensory changes are mainly related to partial sensory rhizotomies. The incidence of hearing loss after MVD ranges from 0.4 to 4.2%.
Complications After the introduction of MRI and BAEP, complication rates substantially decreased. The MRI and MRA detects unruptured cerebral aneurysms or vascular anomalies, and can prevent the unexpected intracranial hemorrhage during or after the operation. Incidence of hearing impairment decreased remarkably by using BAEP monitoring during surgery. This has been proven by the precipitous decline of hearing loss since the adoption of BAEP monitoring by most institutes in 1990 [48]. Trauma to the cerebellum and brain stem by excessive retraction can also be prevented by using intraoperative BAEP monitoring. Incidences of trauma have also decreased since the introduction of BAEP monitoring. McLaughlin et al. [48] reported that incidence of hearing loss was 1.98% before introduction of BAEP monitoring and has since declined to 0.8% after BAEP monitoring. The incidence of cerebellar injury has also declined from 0.87 to 0.45%.
Management of Recurrent Cases In recurrent patients, treatment needs to be started all over again. If medical treatment fails to control neuralgic pain, then re-evaluation of the patient is required, and additional surgical procedures may be recommended. In patients who have previous MVD, other percutaneous procedures are recommended. Although there are reports on good surgical results after repeated MVD in patients with recurrent pain after initial decompressive surgery
. Table 144-4 Complication after microvascular decompression
Reference
Mortality (%)
Cerebellum/ brainstem infarct (%)
Apfelbaum [19] Bederson [25] Klun [14] Barker [20] Chang [21] Tyler-Kabara [12]
1.04 0 1.3 0.14 0 0.2
1.04 0 0.4 0.07 0 0.1
a
IV + V deficit
Hearing loss (%)
Facial weakness (%)
Facial sensory change (%)
Diplopia (%)
0.7 4.2 0.6 1.2 0.4 1.4
0.35 0.4 0 0.15 0 1.7
0 3.0 0.6 1.65 7.9 4.6
0 0 0 0.15 0a –
Microvascular decompression for trigeminal neuralgia
[35,49], results of re-operation after failed MVD for TN are not as good as initial decompressive procedures. There are low incidence of slipped Teflon felt or compression of nerve root by new offending vessels [19,26,33]. Kureshi and Wilkins [50] reported 70% negative exploration rate in patients with persistent or recurrent trigeminal neuralgia. Furthermore, about 30% of reexplorations have adverse effects. In these patients glycerol rhizotomy is recommended. Glycerol rhizotomy has higher recurrence rate than percutaneous radiofrequency rhizotomy, but has less sensory change. Stereotactic radiosurgery can be applied if the patients do not exhibit severe neuralgic pain. For patients who report recurrent pain after percutaneous procedures, MVD is recommended. If patients have a history of failed percutaneous procedure, repeated open surgery will be suggested. If there is no new compressing vessels, slipped Teflon or scarred Teflon in the operating field partial sensory rhizotomy is recommended.
References 1. Dandy WE. Section of the sensory root of the trigeminal nerve at the pons. Preliminary report of the operative procedure. Bull Johns Hopkins Hosp 1925;36:105-6. 2. Dandy WE. The treatment of trigeminal neuralgia by the cerebellar route. Ann Surg 1932;96:787-93. 3. Dandy WE. Concerning the cause of trigeminal neuralgia. Am J Surg 1934;24:447-55. 4. Gardner WJ, Miklos MV. Response of trigeminal neuralgia to ‘‘decompression’’ of sensory root: discussion of the cause of trigeminal neuralgia. JAMA 1959;170:1773-6. 5. Gardner WJ. Concerning the mechanism of trigeminal neuralgia and hemifacial spasm. J Neurosurg 1962;19:947-58. 6. Gardner WJ. Trigeminal neuralgia. Clin Neurosurg 1967;15:1-56. 7. Kerr FWL. Pathology of trigeminal neuralgia: light and electron microscopic observations. J Neurosurg 1967;26:151-6. 8. Hilton DA, Love S, Gradidqe T, Coakham HB. Pathological findings associated with trigeminal neuralgia caused by vascular compression. Neurosurgery 1994;35:299-303.
144
9. Jannetta PJ, Rand RW. Transtentorial subtemporal retrogasserian neurectomy in trigeminal neuralgia by microsurgical technique. Bull LA Neurol Soc 1966;31:93-9. 10. Jannetta PJ. Arterial compression of the trigeminal nerve at the pons in patients with trigeminal neuralgia. J Neurosurg 1967;26:159-62. 11. Burchiel KJ, Slavin KV. On the natural history of trigeminal neuralgia. Neurosurgery 2000;46:152-5. 12. Tyler-Kabara EC, Kassam AB, Horowitz MH, et al. Predictors of outcome in surgically managed patients with typical and atypical trigemial neuralgia: comparison of results following microvascular decompression. J Neurosurg 2002;96:527-31. 13. Lazar ML, Kirkpatrik JB. Trigeminal neuralgia and multiple sclerosis: demonstration of the plaque in an operative case. Neurosurgery 1979;5:711-7. 14. Klun B. Microvascular decompression and partial sensory rhizotomy in the treatment of trigeminal neuralgia: personal experience with 220 patients. Neurosurgery 1992;30:49-52. 15. Kondziolka D, Lunsford LD, Bissonette DJ. Long-term results after glycerol rhizotomy for multiple sclerosisrelated trigeminal neuralgia. Can J Neurol Sci 1994;21 (2):137-40. 16. Levy RM, Foroohar M, Herman M, et al. Radiofrequency trigeminal rhizolysis for the treatment of trigeminal neuralgia secondary to brainstem infarction. Stereotact Funct Neurosurg 1995;65:120-1. 17. Globy AJ, Norbash A, Silverberg GD. Trigeminal neuralgia resulting from infarction of the root entry zone of the trigeminal nerve: case report. Neurosurgery 1998;43:620-3. 18. Chang JW, Choi JY, Yoon YS, et al. Unusual causes of trigeminal neuralgia treated by gamma knife radiosurgery. J Neurosurg (Suppl 5) 2002;97:533-5. 19. Apfelbaum RI. Surgery for tic doulreaux. Clin Neurosurg 1983;31:351-68. 20. Barker FG, Jannetta PJ, Bissonnette DJ, et al. The longterm outcome of microvascular decompression for trigeminal neuralgia. N Engl J Med 1996;33417:1077-83. 21. Chang JW, Chang JH, Park YG, Chung SS. Microvascualr decompression in trigeminal neuralgia: a correlation of three-dimentional time-of-flight magnetic resonance angiography and surgical findings. Streotact Funct Neurosurg 2000;74(3–4):167-74. 22. Sindou M, Leston J, Howeidy T, et al. Micro-vascular decompression for primary trigeminal neuralgia. Longterm effectiveness on pain; prospective study with survival analysis in a consecutive series of 362 patients. Acta Neurochir (Wien) 2006;148:1235-45. 23. Hanakita J, Kondo A. Serious complications of microvascular decompression operations for trigeminal neuralgia and hemifacial spasm. Neurosurgery 1988;22:348-52. 24. Yoshino N, Akimoto H, Yamada, et al. Trigeminal neuralgia: evaluation of neuralgic manifestation and site
2473
2474
144 25.
26.
27.
28.
29.
30. 31.
32.
33.
34.
35.
36.
Microvascular decompression for trigeminal neuralgia
of neurovascular compression with 3D CISS MR imaging and MR angiography. Radiology 2003;228:539-45. Bederson JB, Wilson CB. Evaluation of microvascular decompression and partial sensory rhizotomy in 252 cases of trigeminal neuralgia. J Neurosurg 1989;71:359-67. Rath SA, Klein HJ, Richter HP. Findings and long-term results of subsequent operations after failed microvascular decompression of trigeminal neuralgia. Neurosurgery 1996;39:933-40. Kondo A. Follow-up results of microvascular decompression in trigeminal neuralgia and hemifacial spasm. Neurosurgery 1997;40:46-52. Broggi G, Ferroli P, Franzini A, et al. Microvascular decompression for trigeminal neuralgia: Comments Micro-vascular decompression for primary Trigeminal Neuralgia 1243 on a series of 250 cases, including 10 patients with multiple sclerosis. J Neurol Neurosurg Psychiatry 2000;68(1):59-64. Pamir MN, Peker S. Microvascular decompression for trigeminal neuralgia: a long-term follow-up study. Minim Invasive Neurosurg 2006;49(6):342-6. Kolluri S, Heros RC. Microvascular decompression for trigeminal neuralgia. Surg Neurol 1984;22:235-40. Burchiel KJ, Clarke H, Hagland M, Loeser JD. Longterm efficacy of microvascular decompression in trigeminal neuralgia. J Neurosurg 1988;69:35-8. Hamlyn PJ, King TT. Neurovascular compression in trigeminal neuralgia: a clinical and anatomical study. J Neurosurg 1992;76:948-54. Sun T, Saito S, Nakai O, Ando T. Long-term results of microvascular decompression for trigeminal neuralgia with reference to probability of recurrence. Acta Neurochir (Wien) 1994;126:144-8. Lee SH, Levy EI, Scarrow AM, et al. Recurrent trigeminal neuralgia attributable to veins after microvascular decompression. Neurosurgery 2000;46:356-61. Jannetta PJ, Bissonette DJ. Management of the failed patients with trigeminal neuralgia. Clin Neurosurg 1985;32:334-47. Richard R, Shawdon H, Hollingworth R. Operative findings on microsurgical exploration of the cerebelopontine angle in trigeminal neuralgia. J Neurol Neurosurg Psychiatry 1983;46:1098-101.
37. Piatt JH, Wilkins RH. Treatment of Tic douloureux and hemifacial spasm by posterior fossa exploration: therapeutic implication of various neurovascular relationships. Neurosurgery 1984;14:462-71. 38. Linskey ME, Jho HD, Jannetta PJ. Microvascular decompression for trigeminal neuralgia caused by vertebrobasilar compression. J Neurosurg 1994;81(1):1-9. 39. Kirsch E, Hausmann O, Kaim A, et al. Magnetic resonance imaging of vertebrobasilar ectasia in trigeminal neuralgia. Acta Neurochir (Wien) 1996;138(11):1295-8. 40. Passero S, Filosomi G. Posterior circulation infarcts in patients with vertebrobasilar dolichoectasia. Stroke 1998;29(3):653-9. 41. Passero SG, Calchetti B, Bartalini S. Intracranial bleeding in patients with vertebrobasilar dolichoectasia. Stroke 2005;36:1421-5. 42. Baechli H, Gratzl O. Microvascular decompression in trigeminal neuralgia with no vascular compression. Eur Surg Res 2007;39:51-7. 43. Fujimaki T, Hoya K, Sasaki T, et al. Recurrent trigeminal neuralgia caused by an inserted prosthesis: report of two cases. Acta Neurochir 1996;138:1307-10. 44. Smucker P, Bonnin JIN, Pritz MB. Teflon granuloma with midbrain cysts after microvascular decompression of the trigeminal nerve. Acta Neurochir 2007;149:537-9. 45. Vitali AM, Sayer FT, Honey CR. Recurrent trigeminal neuralgia secondary to Teflon felt. Acta Neurochir 2007;149:719-22. 46. Olson S, Atkinson L, Weidmann M. Microvascular decompression for trigeminal neuralgia: recurrences and complications. J Clin Neurosci 2005;12(7):787-9. 47. Mendoza N, Illingworth RD. Trigeminal Neuralgia treated by microvascular decompression: a long-term followup study. Br J Neurosurg 1995;9:13-9. 48. McLaughlin MR, Jannetta PJ, Clyde BL, et al. Microvascular decompression of cranial nerves: lessons learned after 4400 operations. J Neurosurg 1999;90(1):1-8. 49. Cho DY, Chang CGS, Wang YC, et al. Repeat operations in failed microvascular decompression for trigeminal neuralgia. Neurosurgery 1994;35:665-70. 50. Kureshi SA, Wilkins RH. Posterior fossa reexploration for persistent or recurrent trigeminal neuralgia or hemifacial spasm: surgical findings and therapeutic implications. Neurosurgery 1998;43:1111-17.
132 Motor Cortex Stimulation for Persistent Non-cancer Pain A. G. Machado . A. Y. Mogilner . A. R. Rezai
Introduction The management of patients with chronic pain of benign origin is one of the most difficult and often frustrating tasks of the stereotactic and functional neurosurgeon. These patients are typically seen by a multitude of health care professionals providing a variety of interventions spanning all treatment modalities. Neurosurgery is considered a treatment of last resort and only patients that are refractory to all other pain modalities are considered surgical candidates. Although this approach respects the stepwise approach to the management of chronic pain, which reserves higher risk interventions for patients that are refractory to less invasive alternatives, it also creates a selection bias. Most patients seen by the neurosurgeon have had several years of pain and have been deemed refractory to any intervention. These very refractory disorders are most likely less amenable to any treatment, reducing the chances that neurosurgical intervention may work. In addition, the patient may no longer remember the description of the original pain disorder, which has been modified by the summation of multiple interventions including intentional peripheral or central deafferentation procedures. Often, the treatment is aimed at treating the complications of previous treatments, as the ‘‘original’’ pain syndrome could even be deemed tolerable by the patient. To add further complexity to the matter, behavioral problems are typically present in the chronic and treatment refractory pain population, ranging from depression and hopelessness to personality disorders. In this context, motor #
Springer-Verlag Berlin/Heidelberg 2009
cortex stimulation (MCS) is considered an alternative to the treatment of patients with otherwise intractable neuropathic pain syndromes. Although the overall experience with this treatment modality is still limited, it has been more extensively evaluated in patients with post-stroke central pain and trigeminal neuropathic pain [1]. In the United States, the use of implantable neuromodulation systems for motor cortex stimulation in patients with chronic pain is considered ‘‘off label,’’ as the equipment used for MCS is not approved by the United States Food and Drug Administration for this indication. Despite the rising interest in this application and the encouraging outcomes reported by some groups, the mechanisms underlying the analgesic effects are unclear [2]. When originally suggested as a possible method for pain modulation, it was aimed at alleviating central deafferentation pain [3]. The animal model used for proof of principle entailed ablation of the anterior spinothalamic tract in the cat [4]. In this model, the sensory nucleus of the thalamus was found to be hyperactive, with increased mean spike density. The finding was consistent with data collected previously from chronic pain patients undergoing microelectrode recording [5]. When motor cortex stimulation was applied, a reduction in density with wider inter-spike intervals was observed, indicating that MCS could reduce deafferentation-induced thalamic hyperactivity. This animal model of spinal cord injury was then translated into a human surgical procedure. In patients, however, MCS has been more frequently reported for the treatment of post-stroke pain than chronic pain related
2240
132
Motor cortex stimulation for persistent non-cancer pain
to myelopathy. Although both can be considered central pain syndromes, patients with post spinal cord injury pain have preservation of the thalamocortical integration. Thalamocortical disruption may be the underlying cause of the inconsistent efficacy of MCS in patients with post-stroke pain [6–8]. This is corroborated by indications that patients with at least partially preserved or recovered motor function are more likely to have good outcomes than those with significant motor deficits [9]. Functional imaging studies have recently contributed to our understanding of the mechanisms involved in MCS induced analgesia. Positron emission tomography (PET) has demonstrated increased blood flow in the anterior cingulate cortex, orbital frontal cortex, basal ganglia and periqueductal gray matter [10]. It is possible that activation of these areas is mediated by corticocortical association fibers.
Historical Background and Outcomes Motor cortex stimulation for the treatment of chronic pain syndromes was first reported in 1991 by Tsubokawa et al. [3,11,12]. The first series was composed mostly of patients with central pain syndrome following ischemic or hemorrhagic strokes. The authors had observed limited benefits from thalamic stimulation and were motivated to translate MCS into clinical use. In the preceding decades, thalamic deep brain stimulation had been reported with optimistic outcomes before more critical analysis of outcome indicated that the results were not as promising [13,14]. Early results of MCS indicated that five of 12 patients had complete resolution of the pain, which was maintained for the first year of follow-up while three additional patients had partial pain alleviation. At this time, MCS seemed to be a promising therapy for post-stroke central pain, with results that were superior to DBS. However, these results were not
reproduced by the second institution reporting on a cohort of patients. In 1993, a group of three patients with post-stroke central pain operated at the Karolinska Institute failed to respond to MCS [6]. However, the seven patients with peripheral neuropathies and trigeminal neuropathic pain presented with more favorable results. Again, central deafferentation pain proved to be a challenging pain condition for which first DBS and then MCS demonstrated results that were initially promising but not consistently reproducible. Nevertheless, additional groups embraced the method as a promising alternative. Larger case series contributed to the current understanding of outcomes and allowed for technical improvements to be proposed. Nguyen, Keravel and collaborators have reported on a series of 32 patients with chronic pain [7,8,15]. The mean duration of pain at the time of surgery was approximately 8 years. Postoperative follow-ups ranged significantly (3–50 months). In this series, patients with both poststroke central pain syndromes and facial pain presented with favorable outcomes. Ten of 13 patients with central pain had >40% pain relief while nine of 12 patients with facial pain had >40% relief, eight of whom with >70% relief. The successful results in central pain patients were attributed to accurate localization of the motor cortex using computerized image guidance and thorough electrophysiological testing. Nuti et al. [16] reported on the long term results of MCS in a series predominantly composed of patients with central pain. Of note, although the average follow-up was 4 years, the range was very wide, varying from 2 to 104 months. Half of the patients implanted presented with >40% pain relief. However, even patients with lesser percentage reductions in pain levels indicated some benefit. Eight of 11 patients with 10–39% pain reduction indicated that they would undergo the procedure again in order to gain the same benefits. This suggests that, in this group of patients, small changes in quantitative pain measuring may be associated with
Motor cortex stimulation for persistent non-cancer pain
disproportionately better quality of life results. Rasche et al. [17] have recently reported on the longest follow-up in a large series. Seventeen patients underwent a 1 week period of test stimulation. Three of seven patients with post-stroke central pain and five of ten patients with facial pain presented with positive results (50% pain reduction or more). One patient with facial pain had sustained the positive result for a period of 10 years while the others averaged 3.5 years of follow-up. Maintenance of positive results without gradual decline in efficacy is not consistently reported. Henderson et al. [18] demonstrated that effectiveness is gradually lost with MCS in average 7 months after implantation. For most patients, the benefits can be recaptured with intensive reprogramming efforts. Motor cortex stimulation has a role in managing patients with intractable facial pain and central pain syndromes. Outcomes from the largest series in the literature indicate that positive results can be accomplished with this method and may be long-lasting. Double-blinded assessment of a heterogeneous group of patients with implanted motor cortex stimulators has recently corroborated the efficacy of this method [19]. In this study, stimulation was randomized to OFF 60 or 90 days post implantation (and active stimulation) for a period of 30 days and pain levels reassessed. The results indicate that pain levels were significantly reduced by stimulation and significantly increased when the stimulation was turned off. Nevertheless, when considering MCS as a treatment modality, it should be expected that a proportion of patients will not respond to MCS and that the extent of pain alleviation is not complete but partial.
Patient Selection and Trial Period Although MCS can potentially benefit patients with a number of diagnoses [20,21] the bulk
132
of the reported outcomes are in patients with chronic facial or central pain syndromes [7,12, 15,17,22,23]. It is common for patients to have been seen and treated by a number of physicians of various subspecialties before a neurosurgeon is consulted. A careful review of the previous interventions may indicate a treatment modality that can be explored before MCS is finally considered. Patients may either be considered individually for candidacy or may be enrolled in a clinical trial. In the latter, recommendations for the study design and selection of a homogeneous group have been previously published [24]. Patients considered as candidates for treatment outside a study should understand that MCS is still a new procedure with limited worldwide experience and that ‘‘off-label’’ use of medical equipment will be necessary. The specific diagnosis should be considered and the extent of disability taken into consideration when deciding whether the risks are justifiable. Individuals suffering from unrelenting pain over several years have at least some extent of emotional and psychological overlap aggravating or influencing pain perception [25–34]. A formal psychological evaluation by a professional experienced in chronic pain may help in identifying those with personality disorders or at risk for secondary gain or poor outcome. As neurosurgical intervention will undoutedly be viewed by the patient and family as a highly invasive, last resort procedure, their expectations may be overly optimistic and perhaps even unrealistic. Preoperatively, a hypothetical reduction in pain by 40–50% should be contemplated by a good surgical candidate as beneficial and with good potential for improving quality of life, enhancing productivity and independence. If an externalized trial (see below) will be performed, it is useful to emphasize that the results of the trial are not completely predictable. Indeed, unlike a percutaneous spinal cord stimulator trial, assessing pain relief in the period immediately following a
2241
2242
132
Motor cortex stimulation for persistent non-cancer pain
craniotomy may be quite difficult, due to multiple confounding factors including the duration of hospitalization and other sedative/analgesic medications provided during that time period. The patient must be ready to accept another failure and undergo removal of the electrodes if the trial fails to significantly alleviate the pain. As with any other neuromodulatory device, hardware maintenance should be discussed prior to implantation with the patient and family. Those with limited understanding of current electronic technology may find it difficult to understand how programming is adjusted. Geography may play a role in patient selection. Families must have the commitment to bring the patient for programming at the specialized center as needed. This may not be feasible to those living in remote locations. After implantation, magnetic resonance imaging (MRI) will be contraindicated. Those who may need MRIs for evaluation of other health problems may have to be reconsidered as non candidates for MCS. History, pain characteristics, physical examination and additional tests may be used as possible outcome predictors. Katayama et al. [9] evaluated a series of 31 patients and noted that spared motor function was correlated to positive outcomes from cortical stimulation. Thirteen of 18 patients with minimal or no motor deficits had satisfactory pain relief but only two of 13 patients with significant motor deficits presented with positive outcomes. Likewise, intraoperative motor evoked responses indicated a greater chance for pain alleviation. Pain characteristics such as allodynia, dysesthesia or hyperpathia were not predictive of outcomes. Transcranial magnetic stimulation (TMS) has been used as noninvasive tool for research and treatment of neuropathic pain syndromes [35–38]. Although feasible, the use of repetitive TMS may not be practical as a long-term treatment modality. Nevertheless, it has been tested as a method to predict outcomes from implanted motor cortex stimulators [39]. In this trial, six of 11 patients benefited from MCS implantation. Five
of these six patients had experienced pain alleviation with rTMS. There were no false positives, corroborating to the value of this modality as a screening tool.
MCS Externalized Trial Implantation of the motor cortex stimulation electrodes and connection to an external pulse generator for a period of approximately 1 week is an option to primary permanent implantation. This approach follows a similar rationale of externalized trials routinely used in other neuromodulation modalities for chronic pain syndromes such as spinal cord stimulation. The purpose of the trial is to determine whether the method will be efficacious or not in alleviating the chronic pain and serves to prevent life long commitments to implantable hardware without benefit. It is also likely to be cost-effective as it reduces the overall number of implanted pulse generators, the most expensive component of the hardware system. The results observed by the larger series with chronic follow-up corroborate the predictive value of externalized trials. Patients who respond to stimulation during the trial period are more likely to benefit from permanent implantation. However, patients who fail subacute stimulation are not likely to become responders with chronic stimulation [16,17]. Double-blinded assessments are feasible with subthreshold MCS as it produces no paresthesias or subjective sensations that can hint the patient to the status of the stimulator. In one study, double blinded stimulation demonstrated lack of efficacy in six of nine patients who failed the trial period. In those cases, stimulation and placebo resulted in reduction of reported pain intensities. Only blinded stimulation allowed the investigators to conclude that the effect was related to placebo effect. Permanent implantation was not offered to these patients [17].
Motor cortex stimulation for persistent non-cancer pain
In our experience, the trial period typically ranges from 5 to 10 days. Although longer externalized trials could be useful in predicting response to permanent implantation, the risk of infection may increase [40]. During the first days of the trial, the patient’s ability to report pain accurately may be confounded by the new incisional pain. This is particularly critical among those with facial pain, due to the proximity between the chronic pain topography and the implantation site. Programming of motor cortex stimulation during the externalization period is a time consuming process due to the large number of possible permutations of cathodes, anodes, pulse widths, amplitudes and stimulation frequencies. We initiate the process by identifying a pair of contacts that generates motor evoked responses at the lowest amplitudes as these are likely to be in the immediate vicinity of the motor cortex. The effects of electrical stimulation can be influenced by the polarity of the activated contacts [41,42], by the distance between cortical neurons and electrode contacts and their relative orientation in the crown of the gyrus or sulcus. The thresholds may vary depending on whether the patient is laying or sitting. Stimulation is set to approximately 70% of the motor threshold and the nursing team is advised to disconnect the stimulator immediately if seizures occur and to have intravenous autiseizure medications ready. Depending on the level of training of the team, it may be safer to admit the patient to a monitored unit or epilepsy unit for the time period. It may take the patients 30 min or more to appreciate the analgesic effect of each new stimulation setting. Trials are considered successful when at least one stimulation setting generates a reproducible reduction in pain by 40–50% or more and placebo stimulation is ineffective. When the response is equivocal or the extent of pain reduction is limited, removal of the electrode is likely to be the best option, even if there is a subjective impression that MCS will be beneficial.
132
Surgical Technique and Complications Planning Preoperative planning is focused on identifying the somatotopy of the motor cortex that corresponds to the painful area. This can be accomplished with currently available frameless neuronavigation systems. In our experience, reformatted volumetric T1 magnetic resonance images provide good resolution for localization (> Figure 132-1). The technique has been described and demonstrated to be effective by others [7,15,43,44]. Other options include the use of stereotactic CT scans or simply skull landmarks. Functional MRI (fMRI) assisted localization of the motor cortex is an alternative method for localization and has been reported to have a high correlation with intraoperative electrophysiologic mapping of the motor cortex [45,46].
Burr Hole Versus Craniotomy Motor cortex stimulation was first attempted with electrode arrays inserted through burr holes [3,11]. This remains a valid method and the choice of very experienced groups, with good results [17]. Alternatively, a craniotomy can be planned around the area identified as the precentral gyrus during the preoperative planning. This option allows for a greater exposure of the corresponding dura mater and accommodates more epidural electrodes for intraoperative physiology. As with other functional neurosurgery procedure, electrophysiology is often used to confirm or refine the anatomical location determined by frame-based or frameless image guidance [47–53]. There is currently a trend for the use of epidural recordings and stimulation through a 5–7 cm craniotomy among most groups. This has been linked to more accurate localization of the motor cortex and better
2243
2244
132
Motor cortex stimulation for persistent non-cancer pain
. Figure 132-1 Reformatted triplanar MRI views are routinely used for image guidance and are reliable for identifying the sensorimotor cortex region. This axial view shows the location of the central sulcus and precentral gyrus (arrow). Intraoperative electrophysiology is then used to confirm or refine the location of the motor cortex
outcomes in patients with post-stroke central pain [7]. Somatosensory evoked potentials (SSEP’s) and electrical cortical mapping (ECM) can be performed using electrode grids similar to those used for corticoelectroencephalography (> Figure 132-2). During SSEP recordings, the N20-P20 phase reversal (> Figure 132-3) is used to identify the central sulcus and its orientation. The grid allows for localization of the central sulcus at more than one point, determining its course within the craniotomy, which is important for planning the orientation of electrode implantation. Electrical cortical mapping can also be performed intraoperatively with the same grid used for SSEPs. Stimulation is attempted with individual contacts of the grid in the nonparalyzed patient. The goal is to locate the contact that produces motor responses at the lowest thresholds in the painful area, which
should be nearest to the motor cortex itself. Motor or generalized seizures can be triggered by stimulation. The position of the identified electrodes can be marked on the corresponding dura mater, indicating the position and orientation for the electrodes to be implanted. The most common electrodes used for chronic cortical stimulation are laminectomy spinal cord stimulation electrodes such as the Medtronic Resume (> Figure 132-4). Although the most proximal segment of the electrode can be anchored in a burr hole, the more ample exposure of the craniotomy allows for secure anchoring of the electrode array(s) with sutures on both ends (> Figure 132-5). Dual electrodes can be implanted through craniotomies, side by side, increasing the options for postoperative programming. This approach has been shown to be useful when trying to recapture benefits from chronic stimulation that were lost over time [18]. The advent of dual channel stimulators has allowed for connection of both electrodes to a single pulse generator. The electrode arrays can be positioned parallel or perpendicular to the central sulcus. Parallel implantation in theory increases the chances that the correct somatotopy of the motor cortex will be stimulated. Alternatively, perpendicular implantation increases the chances that at least one contact will be directly on the motor cortex and its transition to the central sulcus. Internalization of the system is offered to patients who experience significant and nonplacebo analgesic effects during the trial. The skin flap is partially reopened but there is no need to open the craniotomy since the electrode extensions are placed under the galea. A subcutaneous pocket is created in the subclavicular region to fit the chosen IPG model, in a similar fashion to DBS procedures. In very thin patients, it is useful to create the pocket deep to the fascia to add another layer of protection and minimize the risk for skin erosion and hardware exposure.
Motor cortex stimulation for persistent non-cancer pain
132
. Figure 132-2 A craniotomy is shown with an epidural grid of 4 4 electrodes. The grid is centered in the location defined by image guidance as the precentral gyrus. Somatossensory evoked potentials are recorded and the effects of epidural stimulation noted. Each individual contact can be activated as the cathode or anode for stimulation. The contact that generates contralateral motor responses at the lowest thresholds is likely to be overlying the motor cortex
. Figure 132-3 A diagrammatic representation of the 4 4 electrode grid is shown with the labels for each of the 16 contacts (a) and then oriented (b) to the surgical view. Contacts seven and ten (dashed arrow) showed the first negative (N20) and positive (P20) potentials, respectively. The phase reversal indicates the location of the central sulcus (solid arrow) in relation to the grid. Courtesy of Dr. Dileep Nair, Cleveland Clinic Epilepsy Center
2245
2246
132
Motor cortex stimulation for persistent non-cancer pain
. Figure 132-4 One or two spinal cord stimulation paddle leads are placed either in parallel or transverse to the central sulcus, over the area defined anatomically and electrophysiologically as the motor cortex. The use of dual electrodes increases the flexibility for postoperative programming and may be particularly important when reprogramming becomes necessary months after implantation, to recapture gradually decreasing benefits
. Figure 132-5 The electrodes can be directly sutured to the dura for optimal anchoring. Note that, in this case, the final location of implantation was in the edge of the craniotomy and not in the center, indicating that the anatomically defined target was refined significantly by electrophysiological findings
Motor cortex stimulation for persistent non-cancer pain
Complications Complications associated with motor cortex stimulation can be related to the surgical procedure and hardware or to stimulation. In our experience, infections related to neurostimulator implants are more common in the IPG site than in the lead site, with or without associated erosion. However, the infection is likely to track the electrodes and contaminate the entire system. If no injury to the dura occurs during implantation, infections should be epidural and related to the flap, with a reduced risk for meningitis or cerebritis. It is possible that trials with externalized electrodes increase the risk for hardware contamination and postoperative infections [40]. If the infection involves any segment of the hardware, it is usually recommended to explant the entire system to allow for proper antimicrobial therapy. However, successful treatment of infections related to the implantable pulse generators with partial removal of the hardware and preservation of the intracranial lead have been reported [54]. Although the limited experience reported in the literature does not allow for accurate calculation of complication rates, it is expected (based on other functional neurosurgical procedures) that epidural hematomas and electrode migration may arise after implantation. Failure of the hardware is a common issue in deep brain stimulation and spinal cord stimulation. The hardware used for motor cortex stimulation is the same and one can expect a similar rate of failure. Electrodes, anchors, extension wires, implantable pulse generators and connectors are at risk for failure. Pain control is likely to be lost either immediately or gradually as the electrodes fail. Interrogation and electronic analysis of the system may indicate abnormal impedances suggestive of circuit shorting or electrode breakage. In our experience with DBS, connectors implanted below the neck impose a risk for breakage of the more fragile DBS leads. Implantation with the connector in the cranial segment
132
is likely to reduce the rate of lead breaks as the bulkier extensions are more likely to tolerate repetitive neck motion. In either case, it is preferred to replace a subcutaneous extension than the DBS lead. Stimulation related complications tend to be transient and readily resolved by interrupting stimulation. Seizures can occur during programming and resolve with administration of intravenous benzodiazepines and cessation of stimulation. Headaches related to stimulation have been reported by a few patients primarily ipsilateral to the electrode location, and thus controlateral to the site of the chronic pain. It is possible that these arise from stimulation of the dura’s innervation and attempts have been made to denervate the dura with bipolar coagulation prior to implantation. Motor cortex stimulation is an alternative for the management of patients with otherwise intractable benign pain syndromes. To-date, the experience indicates that not all patients will respond to this modality and a trial period may be helpful in identifying those who are likely to have long-term benefits prior to permanent implantation. Although results vary from series to series, MCS tends be more efficacious for patients with chronic facial pain and pain related to peripheral deafferentation than for those with post-stroke central pain. The mechanisms underlying the analgesic effects of motor cortex stimulation are still unclear but preservation of corticofugal pathways seem to be important for successful outcomes.
References 1. Burchiel KJ. A new classification for facial pain. Neurosurgery 2003;53(5):1164-6; discussion 1166–7. 2. Meyerson B. Motor cortex stimulation–effective for neuropathic pain but the mode of action remains illusive. Pain 2005;118(1–2):6-7. 3. Tsubokawa T. et al. Treatment of thalamic pain by chronic motor cortex stimulation. Pacing Clin Electrophysiol 1991;14(1):131-4.
2247
2248
132
Motor cortex stimulation for persistent non-cancer pain
4. Hirayama T, et al. Chronic changes in activity of the thalamic relay neurons following spinothalamic tractotomy in cats. Effects of motor cortex stimulation. Pain 1990;5:273. 5. Hirayama T, et al. Recordings of abnormal activity in patients with deafferentation and central pain. Stereotact Funct Neurosurg 1989;52(2–4):120-6. 6. Meyerson BA, et al. Motor cortex stimulation as treatment of trigeminal neuropathic pain. Acta Neurochir Suppl (Wien) 1993;58:150-3. 7. Nguyen JP, et al. Motor cortex stimulation in the treatment of central and neuropathic pain. Arch Med Res 2000;31(3):263-5. 8. Nguyen JP, et al. Chronic motor cortex stimulation in the treatment of central and neuropathic pain. Correlations between clinical, electrophysiological and anatomical data. Pain 1999;82(3):245-51. 9. Katayama Y, Fukaya C, Yamamoto T. Poststroke pain control by chronic motor cortex stimulation: neurological characteristics predicting a favorable response. J Neurosurg 1998;89(4):585-91. 10. Peyron R, et al. Motor cortex stimulation in neuropathic pain. Correlations between analgesic effect and hemodynamic changes in the brain. A PET study. Neuroimage 2007;34(1):310-21. 11. Tsubokawa T, et al. Chronic motor cortex stimulation for the treatment of central pain. Acta Neurochir Suppl (Wien) 1991;52:137-9. 12. Tsubokawa T, et al. Chronic motor cortex stimulation in patients with thalamic pain. J Neurosurg 1993;78 (3):393-401. 13. Coffey RJ. Deep brain stimulation for chronic pain: results of two multicenter trials and a structured review. Pain Med 2001;2(3):183-92. 14. Levy RM, Lamb S, Adams JE. Treatment of chronic pain by deep brain stimulation: long term follow-up and review of the literature. Neurosurgery 1987;21(6):885-93. 15. Nguyen JP, et al. Treatment of deafferentation pain by chronic stimulation of the motor cortex: report of a series of 20 cases. Acta Neurochir Suppl 1997;68:54-60. 16. Nuti C, et al. Motor cortex stimulation for refractory neuropathic pain: four year outcome and predictors of efficacy. Pain 2005;118(1–2):43-52. 17. Rasche D, et al. Motor cortex stimulation for long-term relief of chronic neuropathic pain: a 10 year experience. Pain 2006;121(1–2):43-52. 18. Henderson JM, et al. Recovery of pain control by intensive reprogramming after loss of benefit from motor cortex stimulation for neuropathic pain. Stereotact Funct Neurosurg 2004;82(5–6):207-13. 19. Velasco F, et al. Efficacy of motor cortex stimulation in the treatment of neuropathic pain: a randomized doubleblind trial. J Neurosurg 2008;108(4):698-706. 20. Son UC, et al. Motor cortex stimulation in a patient with intractable complex regional pain syndrome type II with
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32. 33. 34. 35.
36.
37.
hemibody involvement. Case report. J Neurosurg 2003;98(1):175-9. Son BC, et al. Motor cortex stimulation for central pain following a traumatic brain injury. Pain 2006;123(1–2): 210-6. Rainov NG, et al. Epidural electrical stimulation of the motor cortex in patients with facial neuralgia. Clin Neurol Neurosurg 1997;99(3):205-9. Rainov NG, Heidecke V. Motor cortex stimulation for neuropathic facial pain. Neurol Res 2003;25(2): 157-61. Coffey RJ, Lozano A.M. Neurostimulation for chronic noncancer pain: an evaluation of the clinical evidence and recommendations for future trial designs. J Neurosurg 2006;105(2):175-89. Beltrutti D, et al. The psychological assessment of candidates for spinal cord stimulation for chronic pain management. Pain Pract 2004;4(3):204-21. Carlson CR, Psychological considerations for chronic orofacial pain. Oral Maxillofac Surg Clin North Am 2008;20(2):185-95. Hoffman BM, et al. Meta-analysis of psychological interventions for chronic low back pain. Health Psychol 2007;26(1):1-9. Manchikanti L, et al. Psychological factors as predictors of opioid abuse and illicit drug use in chronic pain patients. J Opioid Manag 2007;3(2):89-100. Manchikanti L, et al. Evaluation of psychological status in chronic low back pain: comparison with general population. Pain Physician 2002;5(2):149-55. Tang NK, Crane C. Suicidality in chronic pain: a review of the prevalence, risk factors and psychological links. Psychol Med 2006;36(5):575-86. van Wijk RM. et al. Psychological predictors of substantial pain reduction after minimally invasive radiofrequency and injection treatments for chronic low back pain. Pain Med 2008;9(2):212-21. Cano A. et al. Marital functioning, chronic pain, and psychological distress. Pain 2004;107(1–2):99-106. Gamsa A. The role of psychological factors in chronic pain. I. A half century of study. Pain 1994;57(1):5-15. Gamsa A. The role of psychological factors in chronic pain. II. A critical appraisal. Pain 1994;57(1):17-29. Saitoh Y, et al. Reduction of intractable deafferentation pain due to spinal cord or peripheral lesion by highfrequency repetitive transcranial magnetic stimulation of the primary motor cortex. J Neurosurg 2007;107(3): 555-9. Lefaucheur JP. The use of repetitive transcranial magnetic stimulation (rTMS) in chronic neuropathic pain. Neurophysiol Clin 2006;36(3):117-24. Lefaucheur JP, et al. Neuropathic pain controlled for more than a year by monthly sessions of repetitive transcranial magnetic stimulation of the motor cortex. Neurophysiol Clin 2004;34(2):91-5.
Motor cortex stimulation for persistent non-cancer pain
38. Kanda M, et al. Transcranial magnetic stimulation (TMS) of the sensorimotor cortex and medial frontal cortex modifies human pain perception. Clin Neurophysiol 2003;114(5):860-6. 39. Andre-Obadia N, et al. Transcranial magnetic stimulation for pain control. Double-blind study of different frequencies against placebo, and correlation with motor cortex stimulation efficacy. Clin Neurophysiol 2006;117 (7):1536-44. 40. Constantoyannis C, et al. Reducing hardware-related complications of deep brain stimulation. Can J Neurol Sci 2005;32(2):194-200. 41. Manola L, et al. Anodal vs cathodal stimulation of motor cortex: a modeling study. Clin Neurophysiol 2007;118 (2):464-74. 42. Holsheimer J, et al. Cathodal, anodal or bifocal stimulation of the motor cortex in the management of chronic pain? Acta Neurochir Suppl 2007;97(Pt 2):57-66. 43. Velasco M, et al. Motor cortex stimulation in the treatment of deafferentation pain. I. Localization of the motor cortex. Stereotact Funct Neurosurg 2002; 79(3–4):146-67. 44. Tirakotai W, et al. Localization of precentral gyrus in image-guided surgery for motor cortex stimulation. Acta Neurochir Suppl 2007;97(Pt 2):75-9. 45. Pirotte B, et al. Comparison of functional MR imaging guidance to electrical cortical mapping for targeting selective motor cortex areas in neuropathic pain: a study based on intraoperative stereotactic navigation. AJNR Am J Neuroradiol 2005;26(9):2256-66. 46. Pirotte B, et al. Combination of functional magnetic resonance imaging-guided neuronavigation and intraoperative
47.
48.
49.
50.
51.
52.
53.
54.
132
cortical brain mapping improves targeting of motor cortex stimulation in neuropathic pain. Neurosurgery 2005; 56(2 Suppl):344-59; discussion 344–59. Vitek JL, et al. Microelectrode-guided pallidotomy: technical approach and its application in medically intractable Parkinson’s disease. J Neurosurg 1998;88 (6):1027-43. Vitek JL, Bakay RA, DeLong MR. Microelectrode-guided pallidotomy for medically intractable Parkinson’s disease. Adv Neurol 1997;74:183-98. Zonenshayn M, et al. Comparison of anatomic and neurophysiological methods for subthalamic nucleus targeting. Neurosurgery 2000;47(2):282-92; discussion 292–4. Narabayashi H, Maeda T, Yokochi F. Long-term followup study of nucleus ventralis intermedius and ventrolateralis thalamotomy using a microelectrode technique in parkinsonism. Appl Neurophysiol 1987;50(1–6):330-7. Machado A, et al. Deep brain stimulation for Parkinson’s disease: Surgical technique and perioperative management. Mov Disord 2006;21(S14):S247-58. Gross RE, et al. Electrophysiological mapping for the implantation of deep brain stimulators for Parkinson’s disease and tremor. Mov Disord 2006;21 Suppl 14:S259-83. Starr PA, et al. Microelectrode-guided implantation of deep brain stimulators into the globus pallidus internus for dystonia: techniques, electrode locations, and outcomes. J Neurosurg 2006;104(4):488-501. Sillay KA, Larson PS, Starr PA. Deep brain stimulator hardware-related infections: incidence and management in a large series. Neurosurgery 2008;62(2):360-6; discussion 366–7.
2249
119 Neuroimaging and Pain R. Peyron
Introduction The use of functional imaging techniques in the assessment of pain processes began in the early 1990s with Positron Emission Tomography (PET). In the first two reports, brain regions that were supposed to mediate pain processes included primary (SI) and secondary (SII) somatosensory areas, and thalamus. In the very first reports, brain areas that were reported to actually differ between a noxious and a nonnoxious stimulus were anterior cingulate cortex [1,2], secondary somatosensory, and primary somatosensory cortices [2], thalamus and lenticular nucleus [1]. Since then, many other groups in the world have reported the brain areas in which noxious stimuli (compared to non-noxious stimuli) were associated with increased activity. Surprisingly, activation in the operculo-insular cortices, bilaterally, were the most frequently and consistently reported brain activations [3]. This is precisely the interest of functional imaging to investigate the whole brain and to point out brain regions (i.e., operculo-insular cortices) that would not have been explored by using a priori hypothesis, and that therefore would not have been known as major sites for pain integration. In the last 15 years, many other human studies, using functional Magnetic Resonance Imaging (fMRI), but also Laser Evoked Potentials (LEPs), Magneto-Encephalography (MEG), clinical case-reports, direct intra-cerebral recordings or direct intra-cerebral stimulation in humans have converged to highlight the operculo-insular cortices as major sites for integration of pain processes. Other brain regions such as anterior cingulate cortex (ACC), dorso-lateral-pre-frontal #
Springer-Verlag Berlin/Heidelberg 2009
(DLPF) and posterior parietal (PP), primary motor (MI) cortices, cerebellum, Supplementary Motor Area (SMA) and peri-aqueductal grey matter (PAG) have also shown that they were involved in pain processing. Thus, from a physiological point of view, it can be considered that the so-called ‘‘pain-matrix’’ is identified and concerns a limited number of brain regions [3,4]. Then, in a second step of physiological investigations, the question of the functional contribution of these areas to the pain sensation, was addressed by using experimental designs that were adapted to dissociate the different components of the brain response to pain. In other words, these studies were dedicated to characterize the functional significance of each activation (i.e., the ‘‘what’’ question), the pain sensation being defined as a multi-dimensional experience made of different components [5]. The second axis in recent research was to dissociate in time scale, these brain responses that are known to be consecutive and non-simultaneous (i.e., the ‘‘when’’ question). This issue could not be easily addressed with PET or standard fMRI that averaged signal over 1 min or 3–5s respectively, while early responses to noxious stimuli are known to reach brain target in 150–170 ms. Data obtained from direct intra-cerebral recordings and from MEG should allow exact specification, with a better temporal resolution than PET and standard fMRI, the time-profile of these brain responses. Finally, in a third axis of research, the underlying chemical mediation of pain processes has been investigated by ligand-PET studies to investigate selectively the endogenous opioid system. Such a knowledge of physiological pain processes is of major interest to assess which brain
2020
119
Neuroimaging and pain
regions may be involved in the abnormal pain processes that may be associated with clinical and/or chronic pain situations. Finally, the ultimate challenge of researchers and physicians in the fields of physiological and clinical pain conditions is to investigate the mechanisms and the brain areas that may be crucial for the strengthening of analgesic processes. This is indeed the more recent development of functional imaging techniques to look at different conditions of analgesia or pain relief and their correlates in terms of brain activities and potential therapies for chronic pain.
Physiological Pain Conditions and Components Brain regions which have been described as increasing their activity in response to painful stimuli are now identified as the so-called ‘‘pain matrix.’’ Thus, ‘‘pain-related’’ processes in the brain are distributed within a limited number of activated brain regions for which an exhaustive inventory can be found in previous reviews or meta-analysis [3,4].
The Discriminative Component What is important within this network is the generally bilateral activations, particularly in SII and insular cortices [6–9]. Over the last 15 years, a second important finding is that various in vivo investigations in humans have pointed out insular and SII cortices as major sites for pain processes, while a priori hypothesis did not present these areas as such. Probably one main reason for such ignorance was the incomplete exploration of this structure, because of the depth of insular cortex that was difficult to reach either for deep recording, or for direct stimulation, obviously in humans but also in animal studies. Accordingly, in their per-operative and almost systematic
functional mapping of the human cortex, Penfield and Boldrey [10] and Penfield and Jasper [11] did only rarely evoke unpleasant sensations during superficial stimulation around the SI cortex. The conclusion that could be drawn from these papers was that stimulating the brain did not evoke much pain. However, such a conclusion would elude the bias that for technical reasons, the depth of SII and insular cortices could not be explored. Using stimulation procedures on electrodes that were implanted before surgery in patients with refractory epilepsy, a recent study collecting 4,160 stimulations within the brain first replicated the finding that stimulating various areas within the brain cortex, including anterior cingulate and SI, did not evoke pain (> Figure 119-1 – Mazzola L, Isnard J, Peyron R, Mauguie`re F. Where is pain in the brain? Results of neurostimulations 2008, in preparation). The worthy exception of that was that only 60 sites of stimulation in the brain acutely evoked pain (1%), and that these sites were systematically located in SII and insula for 10 and 50 stimulations, respectively. Considering the 472 stimulations in this area, the frequency of pain-evoked sensation was as frequent as 13%, suggesting definitely that this structure is the most sensitive to evoked pain by direct stimulation of the cortex. Similar findings were recently reported if, instead of stimulating, the same electrodes were used to record, an electrical activity in response to noxious stimuli delivered through a CO2 laser would be found. To date, only the posterior-midCC [12] and the operculo-insular [13] cortices have shown an electrical response to noxious laser stimuli, and only the operculo-insular cortices have shown the ability to encode for pain intensity suggesting that it is the only area in the brain that could be qualified as a true primary cortex for painful events. What is also known, and is consistent with other data (see above), is that this area encodes also for intensities of stimulation below the pain threshold, at least for heat [7], cold [14] and for laser [15,16] sensations, and
Neuroimaging and pain
119
. Figure 119-1 Direct electrical stimulations in medically refractory patients with epilepsy whose brain was explored with StereoElectro-Encephalo-Graphy (SEEG) in order to guide the extent and the location of the resection. Most of them had a temporal lobe epilepsy, explaining why the larger sample of patients is those with temporal explorations (1,997 stimulations). Black letters indicate brain regions where trains of electrical pulses did not elicit any painful sensation. Conversely, red letters indicated the brain area where stimulations induced a truly painful sensation. This area was restricted to the insular cortex and the adjacent SII operculum. Another interesting finding is that when considering the proportion of electrical shocks that lead to a painful sensation, the frequency was a high as 13%. Based on these results, this cortical area that was often not explored previously because of its anatomical location, seems to be very sensitive in terms of painful evoked responses, and conversely, at the level of the brain, this area seems almost specific for pain perception (from Mazzola L, Isnard J, Peyron R, Mauguie`re F. Where is pain in the brain? Results of neurostimulations 2008, in preparation)
that, when painful, the electrical response is bilateralized in the operculo-insular cortex [17,18]. If we now consider the contribution of functional imaging to the knowledge of physiological brain responses to pain, it is noteworthy that in conditions where the attentional component of pain was controlled, the ‘‘pain-matrix’’ can be restricted to a bilateral activation in SII-insular cortices [19]. All these results emphasized the contribution of these areas in discrimination processes devoted to the encoding of the sensory aspects, including identification of the quality
of the stimulus, the evoked sensation and its intensity. In addition to SII-insular cortices, functional imaging, but not data from intra-cerebral recordings, suggest that SI could also assume this function at least for the aspect of intensity (but not pain) coding [20,21], in agreement with electrophysiological studies showing precise coding of noxious stimulus intensity in SI neurons in monkeys [22]. Taken together, these findings converge to suggest that the discriminative aspects of pain sensation primarily involve the operculo-insular cortices and that other
2021
2022
119
Neuroimaging and pain
brain areas that also respond to noxious stimuli may mediate other aspects or other components of the pain sensation.
The Associated Components Many studies have tried to separate, one after the other, the components that are known to contribute to pain perception, in agreement with the definition of the multi-dimensional aspects of pain proposed by Melzack and Casey [5]. The first component to be described with imaging techniques was the emotional aspect of pain which was investigated by manipulating the emotional state with hypnotic suggestion [23]. Specific manipulation of pain unpleasantness modulates pain-related activity in the ACC, suggesting a specific encoding of pain unpleasantness in the ACC [24]. Attention to pain was the second component to be manipulated and identified with PET [19], fMRI [25–27] and Laser Evoked Potentials (LEPs,[28]). As a stimulus (painful or not) contacts the skin, attention is automatically driven toward this stimulus. Spatial attention is known to involve posterior parietal (BAs 40) and pre-frontal (BAs 44–47) cortices [19,29]. Attention is also known to modulate pain intensity [30] and interactions between the pain matrix and the attentional matrix involves the anterior portion of cingulate gyrus [27]. This area is likely to participate in attentional shifts [19,26] directed towards pain. Sub-cortically, evidence for activation related to attention have also been found in sub-cortical structures such as thalamus [19] and PAG [31]. Anticipation of pain was associated with decreased activity in brain areas which are parts of the pain matrix, namely in SII bilaterally and in SI ipsilaterally and contralateral to stimulation in subdivision which were not congruent with the localization of the actual stimulus [32]. Co-existence of increased and decreased activities have also been repeatedly reported in
anterior cingulate and medial pre-frontal cortices [33–39] during anticipation, possibly in relation with stress and/or anxiety [35]. Interestingly, these imaging findings match data from single units recordings in macaques [40] and humans [41] showing that nociceptive neurons in the anterior cingulate also responded to pain anticipation. The motor aspects of pain perception was less frequently investigated since motor withdrawal is an almost natural and uncontrollable component as a subject is exposed to a painful stimulation. In a preliminary study, we found that painful stimulations, intense enough to induce a flexion reflex in the leg, were associated with recruitment of sensory, motor, pre-motor, anterior cingulate and cerebellar activations, which are likely to support the motor aspect, visible or not, of the pain sensation [42]. Investigation of such a motor component has to consider recent data on direct intra-cerebral recordings (see above, [12]) showing that an electrical response can be elicited in the posterior part of mid-ACC by noxious laser stimuli, with a latency that is unconsistent with late components of the pain response, generally considered to mediate emotional, cognitive or affective informations. Conversely, such an early response whose latency is similar to the SII response would fit with fast attentional orienting or reflex motor planning, the exact location of the response being highly consistent with the motor subdivision of ACC.
Self or Others Pain. The Example of Empathy These trials in the dissection of pain components have led to a recent series of studies investigating whether observing others’ pain or having empathy for them could recruit the pain network. Interestingly, it was shown that empathy for the other’s pain involved a part of the pain matrix including anterior insulae and ACC while experiencing pain concerned SII, posterior insula, SI and caudal
Neuroimaging and pain
ACC [43]. Again, this study whose main findings have been replicated by others [44–47] allows the separation of component which is almost emotional and affective and another component whose function is mainly sensory-discriminative and whose distribution of activation fits with the known function of these structures.
Interactions Between Components Investigations of interactions between different components of the pain are more difficult to interprete. For example, anterior cingulate is known to participate, at least in attentional, emotional, empathizing, discriminative, anticipatory and motor components. Each function involved different sub-divisions in the ACC and each sub-division is known to interact one with the others, so that the net result of the activations includes different proportions of these several aspects. As a limiting factor for ‘‘macroscopic’’ imaging studies, these interactions have been evidenced at the microscopic level, for example, by single neuron recordings in SII cortex of the monkey, in which attentional tasks modulated firing activities of neurons devoted to either visual or tactile functions [48]. These results imply that interpretation of functional imaging data should be driven with humility, at least when every component that underlies a brain response cannot be completely segregated by the experimental design of the study and/or when the spatial resolution did not permit segregation of two distinct functional subdivisions in a given brain area.
Application to Clinical Pain Situations Once advances have been made in the knowledge of physiological pain components, the application of these physiological data to the study of clinical conditions in abnormal pain situations is possible. It may be of interest for the understanding of
119
pathophysiology or, alternatively, for the exploration of mechanisms that underlie pain control. To date, the main application of pain imaging to clinical situations has been to investigate processes that underlie neuropathic pain or conversive disorders (hysteria). Investigations of mechanisms that control pain including opioid-induced analgesia, but also hypnotic suggestion, and placebo effect are also of interest for the understanding and the improvement of therapies that concur to relieve clinical pain. All these investigations lead to challenging investigations such as ligand PET that allows, in vivo, the study of direct changes in neurotransmitter systems (such as the endogenous opioid system) that may underlie clinical pain situations and/or pain relief.
Neuropathic Pain Conditions: Basal State Since patients with neuropathic pain consistently have spontaneous pain, the first issue was whether they have basic abnormalities in their brains. If yes, then, the second issue is whether the basic brain abnormalities actually depend on processes related to chronic pain. Because of the lesion, it is indeed conceivable that the abnormalities could depend either on a post-lesional diaschisis, or on somatosensory reorganization, or on more specific reorganization that would lead to pain. To answer this last issue one needs to investigate not only the basic (rest) state of these patients, but also of patients with similar lesions, without neuropathic pain. To date, this issue has not been addressed in a functional imaging study. However, several studies converge to find abnormally decreased Cerebral Blood Flow in the thalamus contralateral to chronic pain [49–52]. In all cases, this abnormality was found to be reversed by various analgesic procedures (from cordotomy to anesthetic blocks to motor cortex stimulation). These findings suggest that thalamic abnormalities may be related to pain rather than to a
2023
2024
119
Neuroimaging and pain
causative lesion. Accordingly, a recent case-report of neuropathic pain in Wallenberg’s syndrome also argues in favor of this view. This patient had, contralateral to facial pain, a thalamic hypoperfusion that was reversed after successful treatment of his neuropathic pain by motor cortex stimulation. Conversely to pain, his facial deafferentation was unchanged after treatment, arguing in favor of pain-related, rather than deafferentation-related hemodynamic changes. As a second argument, the hypoperfused thalamus had a loss of input from a limited (facial) receptive field, while the contralateral thalamus had a loss of input from a large hemibody receptive field, suggesting that, if thalamic hypoperfusion was related to deafferentation, the thalamus ipsilateral to facial pain should have been hypoperfused. This singular observation showed that thalamic abnormality does not merely reflect deafferentation but rather abnormal pain mechanisms, and that these abnormalities are also reversible with pain relief after therapy [53].
Neuropathic Pain Conditions: Allodynia Since patients with neuropathic pain also frequently have allodynic pain, the issue of the functional reorganizations that leads to allodynia could be assessed with functional imagery, by comparing (abnormal) allodynic responses to those obtained with the same physical stimulus applied in a region without neuropathic pain. Accordingly, several studies have investigated allodynia in normal subjects after exposure to capsaicin [54–57]. Nevertheless, only a few studies investigated this phenomenon in patients with neuropathic pain. In these patients, the discriminative system was damaged and they were often unable to analyze correctly the physical characteristics of the stimulus. For example, they often cannot detect whether the stimulus is cold, warm, or neutral, or whether or not they
have been touched by the examinator; in contrast, they immediately feel pain as an innocuous tactile stimulus grazes their skin. In this particular model of sensory distortion called allodynia, reliable abnormal activities have been reported in insular and SII cortices [58–60], suggesting postlesion reorganization in a system devoted to the discriminative aspect of pain. In a larger population of patients with neuropathic pain, it was shown that an innocuous stimulus which normally (on the non-painful side) activated contralateral SI, SII and insular cortices recruited these regions bilaterally as the same stimulus was applied on the painful side and induced allodynic sensation [61]. In another study on Wallenberg’s syndrome, allodynia was shown also to recruit an abnormal activity in the lateral thalamus contralateral to pain [58]. These findings suggest that abnormal activities may develop (after a lesion inducing neuropathic pain) in brain areas which are physiologically devoted to the integration of pain, particularly, in its discriminative component. Cortical areas which are located ipsilaterally to pain may be involved both in sensory recovery and in the genesis of such allodynic pain (> Figure 119-2).
Neuropathic Pain: The Challenge of Neurotransmission This field of research on neuropathic pain has moved now to the study of the brain distribution of specific receptors both in clinical conditions and during analgesic procedures. In a recent study, Willoch et al [62,63] demonstrated that decreased binding of opioids concerned all the main regions involved in the processing of pain. These findings were replicated in a larger and independent study [64], suggesting that these neurobiological abnormalities are linked to the occurrence of neuropathic pain. These abnormalities in opioid receptor distribution may be explained by increased binding of endogenous opioids in these
Neuroimaging and pain
119
. Figure 119-2 Brain responses to pain in a group of 27 patients (from [61]). Images are displayed in a neurological convention (left side of the brain is on the left side of the image). Control stimulations (on the non-painful side of the body) are delivered on the right side. Brain responses to control stimuli are presented on the left part of the figure. Responses contralateral to stimulations are presented on the top rows while responses ipsilateral to stimulation are on the bottom rows. Note that, for the innocuous stimuli used in this experiment, responses were mainly contralateral. Allodynic stimulations are delivered on the left side of the body. Brain response to allodynia are presented on the right part of the figure. Note that brain responses contralateral to stimulations are lower in terms of both amplitude (size of effects) and extent (volume of clusters). An exactly inverse pattern of response is observed ipsilaterally to the allodynic stimulation
2025
2026
119
Neuroimaging and pain
patients because of a lesion changing the imbalance between excitatory and inhibitory processes. Interestingly, various non-opioid procedures devoted to alleviate pain, as well as opioid therapy [65–68], have been shown to induce changes of activity in the orbito-frontal, rostral cingulate and brainstem areas. These procedures included placebo-induced analgesia [68,69] – and we have to remember that placebo analgesia that may be reversed by nalaxone, may therefore be mediated through endogenous opioids [70,71] – anesthetic blocks [50], thalamic [72] or motor cortex [51,52] stimulations and distraction from pain [27,73]. These results suggest that activations in orbitofrontal and rostral cingulate cortices with various procedures resulting in pain control may correspond to non-specific top-down mechanisms exerted through endogenous opioids towards brainstem relays.
Hysteric Pain and Fibromyalgia Finally, the application of pain imaging has been recently used for the study of another clinical situation in which there is a distortion between a given stimulus and the intensity of evoked pain. Hysterical anesthesia is a typical conversive situation where the distortion between a normally painful stimulus and the absence of perceived pain is maximal. Subjects with hysterical anesthesia were shown to have a dysfunction of pain processing [74]. Painful stimulations were associated, not only with the absence of activation in expected brain regions, but also to de-activation involving mainly the somato-sensory and attentional network (SI, SII, posterior parietal and pre-frontal cortices). On the other hand, they were found to have abnormal activation to pain in the rostral ACC. Their disorder extended to the somato-sensory (non-painful) system since abnormalities to innocuous brushing were also observed. These abnormalities suggest a clinically relevant dysfunction in the somato-sensory and
the pain systems but may also indicate abnormal function in the systems that have been listed above (rostral ACC) as involved in pain controls. In fibromyalgia, another clinical situation that includes enhanced pain perception without an identified lesion in the nervous system, an abnormal recruitment of the pain matrix [75–77] is the first finding that has been reported as a correlate of perceptual abnormalities in these patients. More recent studies investigating structural brain changes have revealed less grey matter density in mid-posterior cingulate, insular and medial frontal cortices [78], but these findings remain to be replicated, another study reporting a very different pattern of structural changes in the brain [79]. In any case, these abnormalities need to be confirmed or replicated in further studies to corroborate either decreased pain perception (hysterical anesthesia) or abnormally enhanced sensitivity to pain (fibromyalgia), or, if present, to individualize brain abnormalities in which the disease could find a neurological explanation in addition to the major psychiatric or psychological problems known to be critical in these patients.
Conclusion Functional imaging has provided many pieces of knowledge on physiological pain mechanisms in humans and in vivo. This information is almost consistent across different techniques that can be used to investigate such processes, and the last 15 years have seen the epicenter of the pain processing moving from SI and anterior cingulate cortices to SII and insulae. The application of functional imaging techniques to clinical pain is still a difficult challenge because of the need to pool a large number of patients, who may have different lesions, and in whom the lesion may lead to reorganizations that may or may not be related to pain. In addition, the prospective aspect of this issue – i.e., to know why some
Neuroimaging and pain
patients with a given lesion have neuropathic pain and others with the same lesion do not – remains to be answered in the future. Finally, the finding that most of the physiological or clinical painful conditions are associated with detectable changes in binding to opioid receptors opens a new perspective for functional imaging in the study of biochemical mechanisms underlying pain processes and also in analgesic mechanisms.
References 1. Jones AK, Brown WD, Friston KJ, Qi LY, Frackowiak RS. Cortical and subcortical localization of response to pain in man using positron emission tomography. Proc Biol Sci 1991;244:39-44. 2. Talbot JD, Marrett S, Evans AC, Meyer E, Bushnell MC, Duncan GH. Multiple representations of pain in human cerebral cortex. Science 1991;251:1355-8. 3. Peyron R, Laurent B, Garcia-Larrea L. Functional imaging of brain responses to pain. A review and metaanalysis (2000). Neurophysiol Clin 2000;30:263-88. 4. Apkarian AV, Bushnell MC, Treede RD, Zubieta JK. Human brain mechanisms of pain perception and regulation in health and disease. Eur J Pain 2005;9:463-84. 5. Melzack R, Casey KL. Sensory, motivational and central control determinants of pain: a new conceptual model. In: Kenshalo DR, editor. The skin senses. Springfield, IL: Thomas; 1968. p. 423-43. 6. Casey KL, Minoshima S, Berger KL, Koeppe RA, Morrow TJ, Frey KA. Positron emission tomographic analysis of cerebral structures activated specifically by repetitive noxious heat stimuli. J Neurophysiol 1994;71:802-7. 7. Coghill RC, Sang CN, Maisog JM, Iadarola MJ. Pain intensity processing within the human brain: a bilateral, distributed mechanism. J Neurophysiol 1999;82: 1934-43. 8. Vogt BA, Derbyshire S, Jones AK. Pain processing in four regions of human cingulate cortex localized with coregistered PET and MR imaging. Eur J Neurosci 1996;8:1461-73. 9. Tolle TR, Kaufmann T, Siessmeier T, et al. Regionspecific encoding of sensory and affective components of pain in the human brain: a positron emission tomography correlation analysis. Ann Neurol 1999;45:40-7. 10. Penfield W, Boldrey E. Somatic motor and sensory representation in the cerebral cortex of man as studied by electrical stimulation. Brain 1937;60:389-443. 11. Penfield W, Jasper H. Epilepsy and the functional anatomy of the human brain. In: Brown L, editor. Little, Brown: Boston, MA; 1954.
119
12. Frot M, Mauguiere F, Magnin M, Garcia-Larrea L. Parallel processing of nociceptive A-delta inputs in SII and midcingulate cortex in humans. J Neurosci 2008;28: 944-52. 13. Frot M, Magnin M, Mauguiere F, Garcia Larrea L. Human SII and posterior insula differently encode thermal stimuli. Cereb Cortex 2007;Mar;17(3):610-20. Epub 2006 April 13. 14. Craig AD, Chen K, Bandy D, Reiman EM. Thermosensory activation of insular cortex. Nat Neurosci 2000;3: 184-90. 15. Derbyshire SW, Jones AK, Gyulai F, Clark S, Townsend D, Firestone LL. Pain processing during three levels of noxious stimulation produces differential patterns of central activity. Pain 1997;73:431-45. 16. Bornhovd K, Quante M, Glauche V, Bromm B, Weiller C, Buchel C. Painful stimuli evoke different stimulus-response functions in the amygdala, prefrontal, insula and somatosensory cortex: a single-trial fMRI study. Brain 2002;125:1326-36. 17. Frot M, Mauguiere F. Timing and spatial distribution of somatosensory responses recorded in the upper bank of the sylvian fissure (SII area) in humans. Cereb Cortex 1999;9:854-63. 18. Frot M, Mauguiere F. Dual representation of pain in the operculo-insular cortex in humans. Brain 2003;126: 438-50. 19. Peyron R, Garcia-Larrea L, Gregoire MC, et al. Haemodynamic brain responses to acute pain in humans: sensory and attentional networks. Brain 1999;122:1765-80. 20. Hofbauer RK, Rainville P, Duncan GH, Bushnell MC. Cortical representation of the sensory dimension of pain. J Neurophysiol 2001;86:402-11. 21. Ringler R, Greiner M, Kohlloeffel L, Handwerker HO, Forster C. BOLD effects in different areas of the cerebral cortex during painful mechanical stimulation. Pain 2003;105:445-53. 22. Kenshalo DR, Jr, Chudler EH, Anton F, Dubner R. SI nociceptive neurons participate in the encoding process by which monkeys perceive the intensity of noxious thermal stimulation. Brain Res 1988;454:378-82. 23. Rainville P, Hofbauer RK, Paus T, Duncan GH, Bushnell MC, Price DD. Cerebral mechanisms of hypnotic induction and suggestion. J Cogn Neurosci 1999;11:110-25. 24. Rainville P, Duncan GH, Price DD, Carrier B, Bushnell MC. Pain affect encoded in human anterior cingulate but not somatosensory cortex. Science 1997;277:968-71. 25. Davis KD, Taylor SJ, Crawley AP, Wood ML, Mikulis DJ. Functional MRI of pain- and attention-related activations in the human cingulate cortex. J Neurophysiol 1997;77: 3370-80. 26. Frankenstein UN, Richter W, McIntyre MC, Remy F. Distraction modulates anterior cingulate gyrus activations during the cold pressor test. Neuroimage 2001;14:827-36.
2027
2028
119
Neuroimaging and pain
27. Bantick SJ, Wise RG, Ploghaus A, Clare S, Smith SM, Tracey I. Imaging how attention modulates pain in humans using functional MRI. Brain 2002;125:310-9. 28. Garcia-Larrea L, Peyron R, Laurent B, Mauguiere F. Association and dissociation between laser-evoked potentials and pain perception. Neuroreport 1997;8:3785-9. 29. Posner MI, Dehaene S. Attentional networks. Trends Neurosci 1994;17:75-9. 30. Miron D, Duncan GH, Bushnell MC. Effects of attention on the intensity and unpleasantness of thermal pain. Pain 1989;39:345-52. 31. Tracey I, Ploghaus A, Gati JS, et al. Imaging attentional modulation of pain in the periaqueductal gray in humans. J Neurosci 2002;22:2748-52. 32. Drevets WC, Burton H, Videen TO, Snyder AZ, Simpson JR, Jr, Raichle ME. Blood flow changes in human somatosensory cortex during anticipated stimulation. Nature 1995;373:249-52. 33. Chua P, Krams M, Toni I, Passingham R, Dolan R. A functional anatomy of anticipatory anxiety. Neuroimage 1999;9:563-71. 34. Hsieh JC, Stone-Elander S, Ingvar M. Anticipatory coping of pain expressed in the human anterior cingulate cortex: a positron emission tomography study. Neurosci Lett 1999;262:61-4. 35. Simpson JR, Jr, Drevets WC, Snyder AZ, Gusnard DA, Raichle ME. Emotion-induced changes in human medial prefrontal cortex: II. During anticipatory anxiety. Proc Natl Acad Sci USA 2001;98:688-93. 36. Porro CA, Cettolo V, Francescato MP, Baraldi P. Temporal and intensity coding of pain in human cortex. J Neurophysiol 1998;80:3312-20. 37. Porro CA, Baraldi P, Pagnoni G, et al. Does anticipation of pain affect cortical nociceptive systems? J Neurosci 2002;22:3206-14. 38. Ploghaus A, Tracey I, Gati JS, et al. Dissociating pain from its anticipation in the human brain. Science 1999;284:1979-81. 39. Ploghaus A, Narain C, Beckmann CF, et al. Exacerbation of pain by anxiety is associated with activity in a hippocampal network. J Neurosci 2001;21:9896-903. 40. Koyama T, Tanaka YZ, Mikami A. Nociceptive neurons in the macaque anterior cingulate activate during anticipation of pain. Neuroreport 1998;9:2663-7. 41. Hutchison WD, Davis KD, Lozano AM, Tasker RR, Dostrovsky JO. Pain-related neurons in the human cingulate cortex. Nat Neurosci 1999;2:403-5. 42. Peyron R, Kupers R, Jehl JL, et al. Central representation of the RIII flexion reflex associated with overt motor reaction: an fMRI study. Neurophysiol Clin 2007;37: 249-59. 43. Singer T, Seymour B, O’Doherty J, Kaube H, Dolan RJ, Frith CD. Empathy for pain involves the affective but not sensory components of pain. Science 2004;303:1157-62. 44. Botvinick M, Jha AP, Bylsma LM, Fabian SA, Solomon PE, Prkachin KM. Viewing facial expressions of pain
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
engages cortical areas involved in the direct experience of pain. Neuroimage 2005;25:312-9. Jackson PL, Brunet E, Meltzoff AN, Decety J. Empathy examined through the neural mechanisms involved in imagining how I feel versus how you feel pain. Neuropsychologia 2006;44:752-61. Ogino Y, Nemoto H, Inui K, Saito S, Kakigi R, Goto F. Inner experience of pain: imagination of pain while viewing images showing painful events forms subjective pain representation in human brain. Cereb Cortex 2007;17:1139-46. Benuzzi F, Lui F, Duzzi D, Nichelli PF, Porro CA. Does it look painful or disgusting? Ask your parietal and cingulate cortex. J Neurosci 2008;28:923-31. Steinmetz PN, Roy A, Fitzgerald PJ, Hsiao SS, Johnson KO, Niebur E. Attention modulates synchronized neuronal firing in primate somatosensory cortex. Nature 2000;404:187-90. Di Piero V, Jones AK, Iannotti F, et al. Chronic pain: a PET study of the central effects of percutaneous high cervical cordotomy. Pain 1991;46:9-12. Hsieh JC, Belfrage M, Stone-Elander S, Hansson P, Ingvar M. Central representation of chronic ongoing neuropathic pain studied by positron emission tomography. Pain 1995;63:225-36. Garcia-Larrea L, Peyron R, Mertens P, et al. Electrical stimulation of motor cortex for pain control: a combined PET- scan and electrophysiological study. Pain 1999;83:259-73. Peyron R, Garcia-Larrea L, Deiber MP, et al. Electrical stimulation of precentral cortical area in the treatment of central pain: electrophysiological and PET study. Pain 1995;62:275-86. Garcia-Larrea L, Maarrawi J, Peyron R, et al. On the relation between sensory deafferentation, pain and thalamic activity in Wallenberg’s syndrome: a PET-scan study before and after motor cortex stimulation. Eur J Pain 2006;10:677-88. Iadarola MJ, Berman KF, Zeffiro TA, et al. Neural activation during acute capsaicin-evoked pain and allodynia assessed with PET. Brain 1998;121:931-47. Baron R, Baron Y, Disbrow E, Roberts TP. Brain processing of capsaicin-induced secondary hyperalgesia: a functional MRI study. Neurology 1999;53:548-57. Witting N, Kupers RC, Svensson P, Arendt-Nielsen L, Gjedde A, Jensen TS. Experimental brush-evoked allodynia activates posterior parietal cortex. Neurology 2001;57: 1817-24. Lorenz J, Cross D, Minoshima S, Morrow T, Paulson P, Casey K. A unique representation of heat allodynia in the human brain. Neuron 2002;35:383-93. Peyron R, Garcia-Larrea L, Gregoire MC, et al. Allodynia after lateral-medullary (Wallenberg) infarct. A PET study. Brain 1998;121(Pt 2):345-56. Petrovic P, Ingvar M, Stone-Elander S, Petersson KM, Hansson P. A PET activation study of dynamic mechanical
Neuroimaging and pain
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
allodynia in patients with mononeuropathy. Pain 1999;83: 459-70. Peyron R, Garcia-Larrea L, Gregoire MC, et al. Parietal and cingulate processes in central pain. A combined positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) study of an unusual case. Pain 2000;84:77-87. Peyron R, Schneider F, Faillenot I, et al. An fMRI study of cortical representation of mechanical allodynia in patients with neuropathic pain. Neurology 2004;63:1838-46. Willoch F, Tolle TR, Wester HJ, et al. Central pain after pontine infarction is associated with changes in opioid receptor binding: a PET study with 11C-diprenorphine. AJNR Am J Neuroradiol 1999;20:686-90. Willoch F, Schindler F, Wester HJ, et al. Central poststroke pain and reduced opioid receptor binding within pain processing circuitries: a [11C]diprenorphine PET study. Pain 2004;108:213-20. Maarrawi J, Peyron R, Mertens P, et al. Differential brain opioid receptor availability in central and peripheral neuropathic pain. Pain 2007;127:183-94. Firestone LL, Gyulai F, Mintun M, Adler LJ, Urso K, Winter PM. Human brain activity response to fentanyl imaged by positron emission tomography. Anesth Analg 1996;82:1247-51. Adler LJ, Gyulai FE, Diehl DJ, Mintun MA, Winter PM, Firestone LL. Regional brain activity changes associated with fentanyl analgesia elucidated by positron emission tomography. Anesth Analg 1997;84:120-6. Casey KL, Svensson P, Morrow TJ, Raz J, Jone C, Minoshima S. Selective opiate modulation of nociceptive processing in the human brain. J Neurophysiol 2000;84: 525-33. Petrovic P, Kalso E, Petersson KM, Ingvar M. Placebo and opioid analgesia – imaging a shared neuronal network. Science 2002;295:1737-40. Wager TD, Rilling JK, Smith EE, et al. Placebo-induced changes in FMRI in the anticipation and experience of pain. Science 2004;303:1162-7.
119
70. Levine JD, Gordon NC, Fields HL. The mechanism of placebo analgesia. Lancet 1978;2:654-7. 71. Benedetti F. The opposite effects of the opiate antagonist naloxone and the cholecystokinin antagonist proglumide on placebo analgesia. Pain 1996;64:535-43. 72. Duncan GH, Kupers RC, Marchand S, Villemure JG, Gybels JM, Bushnell MC. Stimulation of human thalamus for pain relief: possible modulatory circuits revealed by positron emission tomography. J Neurophysiol 1998;80:3326-30. 73. Valet M, Sprenger T, Boecker H, et al. Distraction modulates connectivity of the cingulo-frontal cortex and the midbrain during pain – an fMRI analysis. Pain 2004;109:399-408. 74. Mailis-Gagnon A, Giannoylis I, Downar J, et al. Altered central somatosensory processing in chronic pain patients with ‘‘hysterical’’ anesthesia. Neurology 2003;60:1501-7. 75. Gracely RH, Petzke F, Wolf JM, Clauw DJ. Functional magnetic resonance imaging evidence of augmented pain processing in fibromyalgia. Arthritis Rheum 2002;46: 1333-43. 76. Cook DB, Lange G, Ciccone DS, Liu WC, Steffener J, Natelson BH. Functional imaging of pain in patients with primary fibromyalgia. J Rheumatol 2004;31:364-78. 77. Pukall CF, Strigo IA, Binik YM, Amsel R, Khalife S, Bushnell MC. Neural correlates of painful genital touch in women with vulvar vestibulitis syndrome. Pain 2005;115:118-27. 78. Kuchinad A, Schweinhardt P, Seminowicz DA, Wood PB, Chizh BA, Bushnell MC. Accelerated brain gray matter loss in fibromyalgia patients: premature aging of the brain? J Neurosci 2007;27:4004-7. 79. Schmidt-Wilcke T, Luerding R, Weigand T, et al. Striatal grey matter increase in patients suffering from fibromyalgia – a voxel-based morphometry study. Pain 2007;132 Suppl 1:S109-16.
2029
147 Occipital Neuralgia D. B. Cohen . M. Y. Oh . D. M. Whiting
Occipital neuralgia is a disorder characterized by lancinating, paroxysmal, electric shock-like pain radiating from the occipital region to the vertex of the head, mainly in the distribution of the greater occipital nerve. As there are several different pain syndromes that can involve this region, a strict definition of occipital neuralgia can help distinguish this disorder from other pain syndromes including migraine and tension headaches, and cervical muscular strain [1–4]. More attention has been paid to this disorder in recent years, at least in part because of the emergence of occipital nerve stimulation, a minimally invasive and non-destructive treatment [5,6].
Anatomy The C2 dorsal roots are typically the most rostral dorsal roots in the human spinal cord. Approximately 2–4 mm from their dural exit, they converge to form the C2 dorsal root ganglion, which averages 4.5 mm in width and is located lateral to the atlantoaxial ligament and under the obliquus inferior [1,3,7]. The C2 roots and ganglion are located dorsal to the lateral atlantoaxial joint, opposite to the inferior articular process of the atlas, and are held against the joint capsule by fascia. Much of the ganglion lies caudal to the C1 arch but some portion is often directly underneath it [8]. Distal to the ganglion, the C2 spinal nerve quickly divides into ventral and dorsal rami. The ventral ramus, after merging with the cervical plexus, provides most of the contribution to the lesser occipital nerve, which transmits cutaneous sensation from the mastoid eminence
#
Springer-Verlag Berlin/Heidelberg 2009
and posterior surface of the ear [1]. The dorsal ramus splits into four branches: a lateral branch, superior branch, branch to the obliquus inferior, and sensory medial branch. The sensory medial branch gives rise to the greater occipital nerve (GON), whose anatomic course and innervation are important to the understanding and treatment of occipital neuralgia. While running transversely across the obliquus inferior, the GON lies under the splenius capitis and cervicis, longissimus capitis, and semispinalis capitis muscles. It then passes through the aponeurotic attachment of the trapezius and sternocleidomastoid muscles at the superior nuchal line and divides into terminal branches. These terminal branches innervate the region of the C2 dermatome, extending from the occipital area to the vertex of the head and forward to the coronal suture, and mediolaterally from the midline to the medial margin of the mastoid [1,8–10]. Several points of susceptibility to entrapment of the C2 root/ganglion and GON have been suggested based on anatomical considerations. These include: (1) compression of the root/ ganglion between the bony structures of C1 and C2; (2) compression of the ganglion/dorsal ramus at the lateral edge of the posterior atlantoaxial ligament; (3) GON compression at the point at which it emerges from the semispinalis capitis; and (4) compression of the GON at its site of emergence at the superior nuchal line [1]. It has been postulated, with limited clinical evidence, that compression of the C2 elements or GON at one or more of these sites could lead to the pain of occipital neuralgia. Hunter and Mayfield [11] proposed that traumatic extension of
2508
147
Occipital neuralgia
the cervical spine could cause a crush injury of the C2 roots and ganglion between the posterior elements of C1 and C2, but in a cadaver study by Bogduk [8], there was no compromise of the C2 elements by bony structures during extension of the cervical spine. However, when rotation was combined with extreme extension, the posterior arch of the atlas approximated the dorsal edge of the superior articular process of the axis and could reduce the available space for the ganglion and nerve on the side opposite to the direction of rotation. It is unclear in how many trauma patients this mechanism would actually occur [8]. In another cadaver study, Stechison et al. [7] found that the C2 ganglion could not be compressed during extreme neck movement, but removal of the soft tissues during the dissection could have allowed the ganglion to move away from the atlantoaxial joint.
Clinical Features and Non-surgical Treatment Pain from occipital neuralgia tends to follow the distribution of the GON, and is usually unilateral, stereotypical, and can be triggered by palpation over the nerve or neck movement. Typical terms used to describe the pain include deep, aching, shock-like, electric, shooting, jabbing, stabbing, sharp, or exploding, and the pain is paroxysmal [4]. Adherence to these diagnostic criteria is important for evaluating the results of any proposed treatments [1,2,12,13]. Pain described as aching, throbbing, pounding, dull, or pressure-like, and that is steady and diffuse in nature and bilateral in location, is less likely to represent true occipital neuralgia and has been suggested to respond less well to treatment [1,2]. Along with the pain, common associated findings can include loss of sensory function or hyperpathia in the C2 distribution, and Tinel’s sign over the involved GON [1].
Prior to initiating treatment for presumed occipital neuralgia, it is important to rule out underlying cervical spinal disorders, including instability or neural compression resulting from spondylotic disease [14], through the appropriate imaging modalities (MRI, myelogram/CT, etc.) If such a structural abnormality of the spine is present, treatment should be directed towards it. Additionally, metabolic and inflammatory causes of neuralgic pain should be investigated. Various casereports have identified neurosyphilis [15], temporal arteritis [16], and various other maladies in association with occipital neuralgia [1]. Initial treatment for occipital neuralgia consists of analgesic medications, especially nonsteroidal anti-inflammatory medications. Cervical collars and acupuncture can also be used [1,17]. Local anesthetic blocks, often used as a ‘‘trial’’ prior to surgical intervention, can be used as a treatment in their own right, although the relief obtained is often short-lived. In one double-blind, randomized, controlled trial, patients were given either a GON and lesser occipital nerve block using lidocaine, bupivacaine, fentanyl, and clonidine, or an injection of normal saline. After 2 weeks, the patients receiving the block reported approximately 50%, statistically significant decreases on both the visual analog scale and total pain index, as well as significant decreases in pain medicine consumption and overall symptomatology. Patients receiving a block also had a significantly longer pain-free interval until requiring the resumption of pain medications (3.67 days vs. 1.52 days) [18]. More recently, injections of botulinum toxin have been reported to improve symptoms in various headache syndromes including occipital region pain [19,20]. It is not clear yet whether the reported efficacy of botulinum toxin for this indication lies in its effects on the cervical musculature, and/or if it has a central and peripheral neuromodulatory, antinociceptive effect distinct from its muscular effects [20].
Occipital neuralgia
Ablative and Decompressive Surgical Treatments When non-surgical treatments are unsuccessful, a variety of surgical techniques can be employed. (> Table 147‐1) Most authors describe using an anesthetic block of either the GON or the C2 roots prior to surgery as a trial [10,21–23]. Alternatively, symptom provocation (by sublingual nitroglycerin, oxygen inhalation, or subcutaneous ergotamine) can also serve as a trial [22]. C2 blocks have been advocated as more useful than GON blocks, in that a positive response to a C2 block may implicate pathology proximal to the GON that may otherwise be missed, but definitive evidence is lacking [13]. The rationale of utilizing blocks is that a positive response, in the form of at least a temporary decrease or resolution of the patient’s pain syndrome, signifies that the C2/GON complex is indeed involved in the patient’s pathophysiology and is likely to respond favorably to surgery [1]. This approach, however, has been questioned by some authors, who feel that a block is not strongly predictive of ultimate surgical success [2,24]. In an analysis of factors predicting a positive
147
response to C2 decompression in their own patient population, Pikus et al. did not find response to a C2 root/ganglion block to be a prognostic factor [12]. Perhaps the most basic surgical approach to occipital neuralgia is occipital neurectomy. Section of the GON does tend to result in pain relief, though it is generally short-lived. In an overall series of 500 patients with various headache syndromes, Anthony et al. [25] describe 60 patients who underwent an occipital neurectomy. Although 42/60 (70%) were initially headachefree postoperatively, the mean duration of pain relief was only 8.1 months. Recurrence of pain with this method may be due to regeneration of the nerve, as the ganglion is left intact. Additionally, the proximal nerve stump can be the site of pathology, such as a neuroma [1]. GON neurolysis has also been described, with the rationale being to relieve pain resulting from entrapment of the nerve at some point along its peripheral course. Gille et al. [9] published a retrospective series of ten patients who had occipital pain exacerbated by neck flexion. The patients underwent neurolysis of the GON and detachment of the obliquus inferior from
. Table 147‐1 Summary of surgical treatments of occipital neuralgia Treatment modality
Advantages
Disadvantages
Occipital neurectomy
Technically simple Short-term efficacy
GON neurolysis
Short-term efficacy
C2 ganglionectomy C2 ganglion decompression
No neuroma formation
Very high pain recurrence rate Proximal nerve stump can develop pathology Destructive; sensory loss High rate of pain recurrence Non-destructive Destructive; sensory loss
Non-destructive Rhizotomy
Sensory loss minimized
Occipital nerve Stimulation
Non-destructive Minimally invasive
Questionable contribution of ganglion compression to occipital neuralgia Invasiveness, technical difficulty Does not interrupt all nociceptive fibers Infection risk due to implanted device
2509
2510
147
Occipital neuralgia
the lateral aspect of the spinous process of the axis, to prevent muscular compression of the GON during flexion. After a mean followup period of 37 months, the average visual analog scale (VAS) score had decreased from 80/100 to 20/100, analgesic consumption decreased in all patients, and 7/10 patients were either very satisfied or satisfied. On the other hand, Bovim et al. [21] noted similar good responses initially, but over a shorter (14.5 month) follow-up interval noted pain recurrence in 46/50 (92%) of their patients. Although most of the patients felt that they had benefited from surgery, the authors did not recommend neurolysis due to the high rate of pain recurrence. C2 ganglionectomy may be performed in order to avoid the problem of nerve regeneration and neuroma formation [1,3]. Lozano et al. [2] performed C2 ganglionectomy on 39 patients, of whom 17 had failed a prior occipital neurectomy or decompressive procedure. Overall, excellent results (greater than 90% decrease in pain) were obtained in 19 patients and good results (50–90% decrease in pain) were noted in seven patients over an average follow-up period of 28 months. The authors did note a better response to surgery in patients whose pain was post-traumatic (64% excellent response versus 29% in those patients with non-traumatic, spontaneous onset of pain). Stechison [7] and Jansen [22] also noted good results in smaller patient series. Sensory loss in the distribution of the GON results from any such destructive procedure, and a deafferentation pain syndrome (analogous to anesthesia dolorosa) is also a concern, but Lozano et al. [2] noted this complication in only 1 of their 39 patients as a result of surgery. Other authors have described decompression of the C2 ganglion instead of removal. Compression of the C2 ganglion by the atlanto-epistrophic ligament has been postulated as a cause of occipital pain. Stechison et al. [7] described excellent results in five patients, two of whom underwent decompression of the C2 ganglion by division of
this ligament. Pikus et al. [13] obtained an overall success rate, defined as complete or adequate pain relief, of 90% in their series of 35 patients with an average of 21 months of follow-up. They also performed an analysis of factors predicting success after surgery [13], but the only statistically significant prognostic factors identified were adherence to accepted diagnostic criteria and a successful C2 root/ganglion block. Vascular compression, from either the vertebral venous plexus or arterial loops, was noted by Jansen et al. [22] in 16 patients who were operated on for unilateral hemicranial pain. Six of these patients underwent C2 root decompression, with similarly good results. The authors did note macroscopic signs of vascular irritation or compression of the C2 root in all 16 patients, and in two patients an arterial loop was mobilized off the nerve and held away with a piece of Gelfoam. In a concomitant electron microscopic study involving excised roots, the authors noted morphological changes that were possibly the result of chronic vascular compression. Additionally, in a cadaver study performed by the same authors, examination of 100 C2 nerve roots and ganglions from presumably headache-free individuals did not reveal any large, congested veins. Other authors [13,26,27] have noted vascular compression in their patients as well. Some authors, though, have questioned the contribution of vascular compression to occipital neuralgia [13,28]. Pikus et al. [13] noted both venous compression and the presence of a thickened atlanto-epistrophic ligament in high percentages of their patients (85 and 66%, respectively) but a causative link could not be assumed. Lastly, rhizotomy can be performed in an effort to control pain, although it is more complicated and risky than the aforementioned procedures [1]. An early report described section of the entire dorsal root at C2 and C3, with good results in 16 of 22 patients in whom follow-up was available [29]. More recent efforts have
Occipital neuralgia
involved selective dorsal rhizotomies, sectioning only the ventrolateral aspect of the involved root (s) in order to minimize sensory loss [30,31]. One concern, though, is that the procedure does not interrupt the nociceptive fibers that may be present in the anterior rootlets [1]. Some surgeons [10,23] extend the rhizotomy as far caudally as C4 in order to interrupt all possible contributions to the GON. Kapoor et al. [10] noted complete or partial relief in 76.5% of their patients at an average of 20 months of follow-up. They also observed a trend towards better results in patients who had not had a prior surgical procedure, although this difference was not statistically significant due to the small number of patients involved. Dubuisson [30] noted excellent or good results in 10 of 14 patients, with preserved scalp sensation in all patients, and similar results were also obtained by Horowitz et al. [23].
Occipital Nerve Stimulation The limitations of the previously described procedures include pain recurrence (all procedures), sensory loss in the scalp or deafferentation syndrome (ablative procedures), and invasiveness (rhizotomy). Stimulation of the occipital nerve has emerged in recent years as a non-destructive and minimally invasive procedure that can be considered as a first-line surgical option after nonsurgical treatments have been exhausted [5,32]. Typically, a period of trial stimulation will take place first, in which an electrode (either percutaneous or paddle-type) is inserted via a retromastoid or midline incision (> Figures 147‐1 and > 147-2). With the patient awake, the electrode is advanced subcutaneously in the region of the GON, and test stimulation is used to produce paresthesias in the distribution of the GON to confirm appropriate positioning. Ideally, the middle two electrodes in a four electrode array cover the appropriate area. The lead is then anchored to
147
the fascia to prevent migration [34]. A percutaneous extension is then attached to a controller, and the patient receives stimulation at home for several days to a week. If stimulation appears efficacious, then the system is internalized via implantation of a pulse generator in either the infraclavicular or buttock region [6,33]. Typical stimulation parameters that have been reported for occipital nerve stimulation are: pulse width 90–360 ms, frequency 30–130 Hz, and amplitude 0.5–4 V [6,35]. The most common complication of stimulation appears to be lead migration, which can result in loss of stimulation that may require operative revision of the electrode. The incidence of migration is highly variable among series, but has been reported to be as high as 60% in one study [36]. Although definitive evidence is lacking, it seems likely that paddle-type electrodes may be less likely to migrate, due to their larger, flat shape, than cylindrical percutaneous electrodes [33,36,37]. Additionally, paddle-type electrodes deliver stimulation to the nerve without stimulating as much of the surrounding tissue, since these leads only have electrical contacts on the side facing the nerve. This feature may enable the use of lower amplitude stimulation, prolonging the life of the generator [33,37]. The results of occipital nerve stimulation (ONS) have so far been encouraging. Weiner’s report [6] described excellent results (>75% pain relief) in two-thirds and good results (>50% pain relief) in one-third of 13 patients. Oh et al. [33] placed occipital nerve stimulators in ten patients with occipital neuralgia and obtained similar results after 6 months of follow-up. Kapural et al. [38] noted significant improvement after stimulation using the VAS and pain disability index (PDI) as outcome measures. Several other reports outline similar results [35–37,39,40]. Despite early promising results with ONS, there has so far been a dearth of well-designed studies with long-term follow-up. One comprehensive review called for future studies to more
2511
2512
147
Occipital neuralgia
. Figure 147‐1 Depiction of bilateral occipital nerve stimulator electrode placement (reprinted from [33])
clearly delineate diagnostic criteria, to be randomized, include parallel control groups, utilize sham stimulation, blind the patients, investigators, and programmers, and to follow patients for at least 1 year postoperatively [41]. ONS has been postulated to exert its effects via both peripheral effects and central neuromodulation. Stimulation of large sensory fibers in the occipital nerve is thought to suppress smallfiber nociceptive input at the level of the spinal cord dorsal horn, and may also suppress A-delta fiber activity through a peripheral conduction blockade [24]. Evidence has also been provided
for a central neuromodulatory effect of ONS by a recent study utilizing positron emission tomography (PET). The study examined metabolic markers of brain activation in eight patients who had bilateral occipital nerve stimulators implanted for migraine headaches. Patients underwent a PET scan with the stimulator on and off, and the metabolic activity in the dorsal rostral pons, anterior cingulate cortex, cuneus, and pulvinar were noted to be altered by the peripheral stimulation. The authors suggest, based on this and other evidence, an association between pulvinar activation and pain relief,
Occipital neuralgia
147
. Figure 147‐2 Illustration of bilateral occipital nerve stimulator electrode placement (reprinted from [33])
although it is unclear if this simply represents an association or if the pulvinar is a mediator of pain relief. Also, this and other studies have suggested a role for the dorsal rostral pons in migraine headache. In these patients the dorsal rostral pons appears to be in a state of constant dysfunctional activation, which may help explain the recurrence of patients’ pain when the stimulator is turned off [42].
Future Directions The encouraging results so far obtained with ONS for occipital neuralgia has led to interest in the technique for other indications, most notably migraine headache. It has been known for some time [11] that a connection exists between occipital nerve elements and the trigeminal nerve complex. Specifically, these systems appear to overlap in the upper cervical spinal cord, in the region of the spinal trigeminal nucleus [1]. Evidence for the functional, as well as anatomical, connection between these two systems was provided by a study performed by Goadsby et al. [43]. Cats were subjected to occipital nerve stimulation, and the authors
measured an increase in metabolic activity in the cervical dorsal horn and trigeminal nucleus caudalis, as would occur when trigeminal structures are stimulated. Thus, there appears to be an overlap in the processing of nociceptive information, which could help explain the occurrence of both anterior and posterior head pain in various headache syndromes. This also lends support to the idea that interventions such as anesthetic blocks or surgery on the C2 root, ganglion, or GON may have more of a central neuromodulatory effect than a peripheral effect, and explains how ONS may help alleviate a variety of headache syndromes [43–45]. With this in mind, several researchers have begun to report results of ONS for migraine-type headaches. Migraines are estimated to occur in 17% of adult females, 6% of adult males, and 5–10% of children, with women experiencing an average of 37.4 attacks/year and men 34. The estimated cost of migraines in terms of missed workdays and impaired performance is $18 billion [44]. Oh et al. [33] implanted occipital nerve stimulators in ten patients with transformed migraine, a condition involving nonparoxysmal cervical tension with secondary radiation to the posterior head. The pain occurs daily or almost
2513
2514
147
Occipital neuralgia
daily, and patients have a previous history of episodic migraine with an increased frequency and decreased severity of migrainous features. The results obtained were comparable to the results of ONS for occipital neuralgia. Popeney et al. [45] implanted 25 migraine patients with peripheral nerve stimulators, and found a significant decrease in both headache days/month (75.6 per 3 months preoperatively vs. 37.5 per 3 months postoperatively) and headache severity (9.32 on the VAS preoperatively vs. 5.72 postoperatively). Over an average of 18.3 months of follow-up, 88% of patients experienced at least a 50% improvement in headache frequency and severity, and all patients reported that their headache was wellcontrolled and that they would repeat the procedure. Several prospective, randomized, blinded studies are currently planned, involving larger numbers of patients with longer follow-up [46].
Conclusion The advent and application of peripheral neurostimulation techniques to the treatment of occipital neuralgia has provided an additional treatment option for these often difficult-to-treat patients. Its lack of invasiveness, adjustability, reversibility, and favorable safety profile do give it an advantage over earlier surgical procedures. More recent anatomic and pathophysiologic insight has provided the basis for studying ONS for migraine, with early promising results. Larger and better-designed studies with longer follow-up periods are essential for the future investigation of ONS.
References 1. Lozano AM. Treatment of occipital neuralgia. In: Gildenberg PL, and Tasker RR, editors. Textbook of stereotactic and functional neurosurgery. New York, NY: McGraw-Hill; 1998. p. 1729-33.
2. Lozano AM, Vanderlinden G, Bachoo R, Rothbart P. Microsurgical C-2 ganglionectomy for chronic intractable occipital pain. J Neurosurg 1998;89:359-65. 3. Wang MY, Levi ADO. Ganglionectomy of C-2 for the treatment of medically refractory occipital neuralgia. Neurosurg Focus 2002;12:E14. 4. Merskey H, Bogduk N. Classification of chronic pain: descriptions of chronic pain syndromes and definitions of pain terms. Seattle: IASP Press; 1994. p. 64-5. 5. Alo´ KM, Holsheimer J. New trends in neuromodulation for the management of neuropathic pain. Neurosurgery 2002;50:690-704. 6. Weiner RL, Reed KL. Peripheral neurostimulation for control of intractable occipital neuralgia. Neuromodulation 1999;2:217-21. 7. Stechison MT, Mullin BB. Surgical treatment of greater occipital neuralgia: an appraisal of strategies. Acta Neurochir (Wien) 1994;131:236-40. 8. Bogduk N. The anatomy of occipital neuralgia. Clin Exp Neurol 1980;17:167-84. 9. Gille O, Lavignolle B, Vital JM. Surgical treatment of greater occipital neuralgia by neurolysis of the greater occipital nerve and sectioning of the inferior oblique muscle. Spine 2004;29:828-32. 10. Kapoor V, Rothfus WE, Grahovac SZ, Kassam SZA, Horowitz MB. Refractory occipital neuralgia: preoperative assessment with CT-guided nerve block prior to dorsal cervical rhizotomy. AJNR Am J Neuroradiol 2003;24:2105-10. 11. Hunter CR, Mayfield FH Role of the upper cervical roots in the production of pain in the head. Am J Surg 1949;48:743-52. 12. Pikus HJ, Phillips JM. Characteristics of patients successfully treated for cervicogenic headache by surgical decompression of the second cervical root. Headache 1995;35:621-9. 13. Pikus HJ, Phillips JM. Outcome of surgical decompression of the second cervical root for cervicogenic headache. Neurosurgery 1996;39:63-71. 14. Ehni G, Benner B. Occipital neuralgia and C1-C2 arthrosis. N Engl J Med 1984;310:127. 15. Smith DL, Lucas LM, Kumar KL. Greater occipital neuralgia: an unusual presenting feature of neurosyphilis. Headache 1987;27:552-4. 16. Jundt JW, Mock D. Temporal arteritis with normal erythrocyte sedimentation rates presenting as occipital neuralgia. Arthritis Rheum 1991;34:217-19. 17. Hongbin W, Hong C, Hongrui J. Experience in acupuncture treatment of occipital neuralgia. J Tradit Chin Med 2002;22:183. 18. Naja ZM, El-Rajab M, Al-Tannir MA, Ziade FM, Tawfik OM. Occipital nerve blockade for cervicogenic headache: a double-blind randomized controlled clinical trial. Pain Pract 2006;6:89-95.
Occipital neuralgia
19. Freund BJ, Schwartz M. Treatment of chronic cervicalassociated headache with botulinum toxin A: a pilot study. Headache 2000;40:231-6. 20. Loder E, Biondi D. Use of botulinum toxins for chronic headaches: a focused review. Clin J Pain 2002;18:S169-S176. 21. Bovim G, Fredriksen TA, Stolt-Nielsen A, Sjaastad O. Neurolysis of the greater occipital nerve in cervicogenic headache. Headache 1992;32:175-9. 22. Jansen J, Bardosi A, Hildebrandt J, Lu¨cke A. Cervicogenic, hemicranial attacks associated with vascular irritation or compression of the cervical nerve root C2. Clinical manifestations and morphological findings. Pain 1989;39:203-12. 23. Horowitz MB, Yonas H. Occipital neuralgia treated by intradural dorsal nerve root sectioning. Cephalalgia 1993;13:354-60. 24. Schwedt TJ, Dodick DW, Trentman TL, Zimmerman RS. Response to occipital nerve block is not useful in predicting efficacy of occipital nerve stimulation. Cephalalgia 2007;27:271-4. 25. Anthony M. Headache and the greater occipital nerve. Clin Neurol Neurosurg 1992;94:297-301. 26. Hildebrandt J, Jansen J. Vascular compression of the C2 and C3 roots – yet another cause of chronic intermittent hemicrania? Cephalalgia 1984;4:167-70. 27. Jho HD, Jannetta PJ. Entrapment of C2 ganglion: the socalled occipital neuralgia or migraine. Presented at the 61st Annual Meeting of the American Association of Neurological Surgeons, April 24–29, Boston, MA; 1993. 28. Stechison MT. Outcome of surgical decompression of the second cervical root for cervicogenic headache. Neurosurgery 1997;40:1105-6. 29. Chambers WR. Posterior rhizotomy of the second and third cervical nerves for occipital pain. J Am Med Assoc 1954;155:431-2. 30. Dubuisson D. Treatment of occipital neuralgia by partial posterior rhizotomy at C1-3. J Neurosurg 1995;82:581-6. 31. Rasskazoff S, Kaufmann AM. Ventrolateral partial dorsal root entry zone rhizotomy for occipital neuralgia. Pain Res Manage 2005;10:43-5. 32. Oh M, Whiting D. Minimally invasive peripheral nerve stimulation for the treatment of occipital neuralgia. In: Burchiel KJ, editor. Pain news. Pain; Chicago, IL: AANS/ CNS Section on 1999. p. 3-5. 33. Oh MY, Ortega J, Bellotte JB, Whiting DM, Alo´ K. Peripheral nerve stimulation for the treatment of occipital neuralgia and transformed migraine using a C1-2-3
34. 35.
36.
37. 38.
39.
40.
41.
42.
43.
44.
45.
46.
147
subcutaneous paddle style electrode: a technical report. Neuromodulation 2004;7:103-12. Gofeld M. Anchoring of suboccipital lead: case report and technical note. Pain Pract 2004;4:307-9. Slavin KV, Nersesyan H, Wess C. Peripheral neurostimulation for treatment of intractable occipital neuralgia. Neurosurgery 2006;58:112-19. Schwedt TJ, Dodick DW, Hentz J, Trentman TL, Zimmerman RS. Occipital nerve stimulation for chronic headache – long-term safety and efficacy. Cephalalgia 2007;27:153-7. Jones RL. Occipital nerve stimulation using a Medtronic Resume II1 electrode array. Pain Physician 2003;6:507-8. Kapural L, Mekhail N, Hayek SM, Stanton-Hicks M, Malak O. Occipital nerve electrical stimulation via the midline approach and subcutaneous surgical leads for treatment of severe occipital neuralgia: a pilot study. Anesth Analg 2005;101:171-4. Slavin KV, Colpan ME, Munawar N, Wess C, Nersesyan H. Trigeminal and occipital peripheral nerve stimulation for craniofacial pain: a single-institution experience and review of the literature. Neurosurg Focus 2006;21:E6. Melvin EA, Jordan FR, Weiner RL, Primm D. Using peripheral stimulation to reduce the pain of C2-mediated occipital headaches: a preliminary report. Pain Physician 2007;10:453-60. Coffey RJ, Lozano AM. Neurostimulation for chronic noncancer pain: an evaluation of the clinical evidence and recommendations for future trial designs. J Neurosurg 2006;105:175-89. Matharu MS, Bartsch T, Ward N, Frackowiak RSJ, Weiner R, Goadsby PJ. Central neuromodulation in chronic migraine patients with suboccipital stimulators: a PET study. Brain 2004;127:220-30. Goadsby PJ, Knight YE, Hoskin KL. Stimulation of the greater occipital nerve increases metabolic activity in the trigeminal nucleus caudalis and cervical dorsal horn of the cat. Pain 1997;73:23-8. Rogers LL, Swidan S. Stimulation of the occipital nerve for the treatment of migraine: current state and future prospects. Acta Neurochir Suppl 2007;97:121-8. Popeney CA, Alo´ KM. Peripheral neurostimulation for the treatment of chronic, disabling transformed migraine. Headache 2003;43:369-75. Goadsby PJ, Dodick D, Mitsias P, Khan K, Khan A, Brewer AR, Saper J, Silberstein SD. ONSTIM: occipital nerve stimulation for the treatment of chronic migraine. Eur J Neurol 2005;12 Suppl 2:198.
2515
126 Percutaneous Cordotomy R. R. Tasker
Introduction The more liberal use of narcotics to treat chronic pain appears to have diminished interest in the use of destructive procedures such as percutaneous cordotomy. This was once a very popular and frequently performed operation which a review of recent publications suggests has lost much of its attraction [1]. Originally it was done by open laminectomy after the conceptual advances of Spiller [2] and Spiller and Martin [3]. Mullan et al. [4] introduced a percutaneous technique with a radiostrontiun source. This was soon replaced, however, by lesion making through the use of anodal current [5] until radio frequency lesioning [6,7] became the method of choice. Myelography was added for localization [8] and the operation was assisted with the use of electrical impedance monitoring [9] coupled with electrophysiological corroboration of the target site [10–15]. The most recent refinement is the addition of computerized tomography [16]. The procedure can be done by various approaches: the low cervical, through the disc, method [17] the high dorsal cervical [18,19] and the high lateral cervical with which this author has most experience and which will be discussed here. CT guided cordotomy appears to have gained interest [20–23] and, when used to guide low cervical through the disc cordotomy, may help avoid post operative respiratory problems [21].
Indications The procedure is effective for the control of chronic pain that is dependent on transmission #
Springer-Verlag Berlin/Heidelberg 2009
in pain pathways such as that of cancer as well as shooting and evoked elements of neuropathic pain [24,25]. Simpler treatment attempts should have first failed before moving to percutaneous cordotomy, taking into account the expected success and complication rates; Gybels ([26]; > Table 126‐1) gives useful guidelines. The high lateral approach seems to have been most popular because it can be done without the use of a stereotactic frame as is necessary for the high dorsal cervical approach and it gives a level of analgesia up to C5, unachievable with the low, through the disc; anterior cervical approach which avoids damage to the reticulospinal pathway that governs automatic respiratory function but which makes small incremental movements of the electrode embedded as it is in the disc, difficult. Attacks on the somatosensory system, as distinct from the spinothalamic tract, though simpler to carry out, may damage non-expendable somatosensory and motor function while the use of modulatory surgery, such as drug infusion or chronic stimulation, requires expensive equipment.
Mechanism Even after all these years, there is limited consensus on the mechanism by which cordotomy eases pain [27–31]. Be that as it may, the operation succeeds in reducing pain by severing the lateral spinothalamic tract which includes temperature appreciation as well. Since the temperature and pain fibers travel separately in the tract, the dermatomal extent of the loss of the two types of sensation after cordotomy are similar though not identical. It is interesting that Rosso and
2138
126
Percutaneous cordotomy
. Table 126‐1 Neurosurgical indications in cancer pain Life expectancy, months 1–2
2–5
>5
Procedure CSF opioid infusion and percutaneous neurolytic procedures. CSF opioid infusion and percutaneous neurolytic procedures; if these fail, cordotomy, if feasible, or other destructive surgery as the situation demands. Ablative procedures as above; PVG stimulation and CSF opioid infusion.
Aglioti et al. [32] have made the surprising finding that nociceptiye deafferentation, such as by cordotomy, rapidly produces modulation of cortical neuronal activity along the leminiscal pathway.
Contraindications Percutaneous cordotomy is relatively ineffective in midline truncal pain [33] and it is dangerous if performed contralateral to a non-respiring hemithorax (the result of pneumonectomy, lung damage, phrenic nerve interruption or spinal cord lesions including cordotomy) because the proposed cordotomy lesion can interrupt the only remaining functional ipsilaterally distributed reticulospinal tract [34–41] (see > Figure 126‐1). For the same. reason, bilateral percutaneous cordotomy is dangerous, especially if a high level of analgesia was produced on the first side, since the reticulospinal tract lies close to the spinothalamic fibers in the high cervical area. Price and Pounder et al. [42], have studied the issue of post cordotomy respiratory function finding, surprisingly, that poor respiratory function preoperatively should not, in itself, be a barrier.
Technique Percutaneous cordotomy is usually done under local anesthesia with intravenous sedation allowing communication with the patient for physiological localization. If the patient is unable to cooperate or is very young, general anesthesia without muscle paralysis can be used instead [43–46]. The patient is positioned with the cervical spine as horizontal as possible to best retain contrast medium when the operation is done, using a specially designed head rest (> Figure 126‐2) with a mechanical stage to move the electrode under lateral image intensification. A spinal puncture cordotomy electrode assembly (> Figure 126‐3) is used whose stylet is replaced by the cordotomy electrode once its position is satisfactory. The electrode has a 4 mm projecting tip extending 2 mm beyond its shrink fit Teflon insulation (> Figure 126‐4) and the latter extends 2 mm beyond the open end of the spinal needle. The purpose of this design is to impale the spinal cord to a depth of 2 mm where the cuff of Teflon will arrest its progress at an ideal depth for cordotomy. The spinal needle, controlled by the manipulator with its stylet in place is first advanced perfectly horizontally into the subarachnoid space which is identified by repeated removal of the stylet until spinal fluid flow is identified. The stylet is then withdrawn and a few milliliters of a mixture of spinal fluid, air and oil based positive contrast medium are shaken up in a 10 ml syringe and forcefully injected into the subarachnoid space using a suitable connector. This outlines the dentate ligament within the spinal canal (> Figure 126‐5). Water soluble contrast medium does not persist long enough to give adequate imaging. The image of the dentate ligament outlined by the bubbles of contrast medium lies midway between the anterior and posterior margins of the bony canal and must be distinguished from dye lying on the anterior cord
Percutaneous cordotomy
126
. Figure 126‐1 Percutaneous cordotomy showing impedance changes as the electrode is advanced and the anatomic relations of the spinothalamic tract. This view is oriented with the dorsal surface of the cord at the top, not as the surgeon encounters the structures with the patient in the supine position. S: sacral, L, lumbar, T, thoracic, C, cervical dermatomes in the spinothalamic tract
margin or the contralateral dentate ligament. Sometimes dye is held up on the ventral or dorsal root lines causing some confusion. The progress of the advancing of the spinal needle/electrode complex is marked by a sense of resistance and perhaps a twinge of pain as it traverses first the ligamentum flavum and then the dura. It is important not to penetrate the cord with other than the sharpened tip of the electrode. The progress can be monitored in two ways: with an AP x-ray in which the appropriately positioned electrode tip is superimposed almost on the image of the contralateral edge of the dens (> Figure 126‐6) and by impedance monitoring. With a ground
electrode, such as a hollow intramuscular needle, inserted into the ipsilateral deltoid muscle, the electrical impedance between electrode and ground is about 400 ohms when the tip of the electrode is in spinal fluid and it should rise abruptly to 1,000–2,000 ohms once the cord is penetrated. This is also accompanied by a sense of resistance felt by the surgeon and twinge of pain by the patient. Proper electrode positioning is determined by several factors. Using the OWL electronic backup equipment (> Figure 126‐7), electrical stimulation is used to verify the electrode’s position in the spinothalamic tract. Using two hertz
2139
2140
126
Percutaneous cordotomy
. Figure 126‐2 Head holder and mechanical stage for percutaneous cordotomy (available from Diros Technology, Toronto)
. Figure 126‐5 Lateral x-ray study showing cordotomy electrode tip just anterior to the dentate ligament (outlined with lipiodol). The ventral and dorsal root lines, anterior cord margin, and dorsal subarachnoid space are also outlined
. Figure 126‐3 Percutaneous cordotomy electrode capable of temperature monitoring (available from Diros Technology, Toronto)
. Figure 126‐6 Anteroposterior x-ray study showing cordotomy electrode impaled in cord near midline ofCI-C2 interspace . Figure 126‐4 Percutaneous cordotomy electrode tip projecting 2 mm from introductory LP needle (scale in mm)
stimulation, motor responses are first sought, either under general anesthetic without paralysis or in the wake sedated patient with local anesthetic. The main purpose is to avoid the
corticospinal tract where two hertz stimulation causes contraction in the appropriate ipsilateral musculature while 100 Hz stimulation causes tetanization in the same muscles and sometimes
Percutaneous cordotomy
126
. Figure 126‐7 The OWL Cordotomy System (available from Diros Technology, Inc.) Diros Technology is located at 232 Hood Rd., Markham, ON L3P 3A8, Canada e-mail: [email protected]
ipsilateral paresthesiae as well. With the electrode ideally located in spinothalamic tract, two hertz stimulation at 1–3 Vactivates ipsilateral nuchal or upper limb muscles, sometimes ipsilateral upper limb muscles, and rarely causes a contralateral, sensory affect (see > Table 126‐2). There should be little or no ipsilateral nuchal tetanization and no ipsilateral effects below the upper extremity at threshold. One hundred hertz stimulation in spinothalamic tract should not tetanize anywhere at threshold but should cause contralateral sensory effects. Suprathreshold stimulation that causes tetanizing especially in the ipsilateral leg suggests. a safe proximity to corticospinal tract. It is difficult to explain the ipsilateral motor effects at 1–3 Hz seen when the electrode is stimulated in the spinothalamic tract, unaccompanied by any motor effects at 100 Hz at the same site. > Table 126‐2 lists physiological effects at various cord sites. With ideal localization in the spinothalamic tract, 100 Hz stimulation induces a contralateral usually warm or cool sensory effect. If it is felt in the lower extremity, the electrode is likely in the more caudal aspects of the tract where a resultant lesion will give a low level of analgesia but also spare damage to the reticulospinal tract and hence respiration. Sensory effects referred to the
contralateral hand suggest a central position in the spinothalamic tract where a lesion will yield a higher level of analgesia but could damage the reticulospinal tract (see > Figure 126‐1). Sensory responses sparing the upper or lower extremities suggest an inadequate lesion site. Uncommon ipsilateral paresthesiae are difficult to explain and when they occur, suprathreshold stimulation should be done to rule out tetanization in the same body part where a lesion might damage the corticospinal tract. Radiofrequency lesioning is done once the localizing criteria have been satisfied. This requires an increased level of sedation and analgesia or even a brief general anesthetic. A test lesion is first made at 50 C and untoward effects sought. Such a small lesion may not produce analgesia. If all is well, the lesion is gradually enlarged up to 70 or 80 C for up to 60 s with serial neurological checking. Usually it is not necessary to reposition the electrode though this can be done if the result warrants and to obtain a suitable level of analgesia it may occasionally be necessary to move the electrode to the occipital cervical level if all else fails. Assuming no complications evolve, the maximum lesion is usually made by allowing the radiofrequency current to flow until ‘‘fall off ’’ occurs
2141
2142
126
Percutaneous cordotomy
. Table 126‐2 Stimulation results for a variety of possible electrode positions at CI-C2 Threshold simulation Location
2 Hz
100 Hz
Ipisilateral anterior horn
Contractions in ipsilateral neck, often at very low threshold.
Tetanization of ipsilateral neck
Ipsilateral corticospinal tract Contralateral cord
Contractions in neck, any ipsilateral musculature
Tetanization of any ipsilateral musculature, sometimes ipsilateral paresthesias. Tetanization in contralateral neck
Ipsilateral spinothalamic tract
Contractions in ipsilateral neck or upper limb
No response to stimulation
Stimulator turned off, defects in equipment, patient’s muscles paralyzed if under general anesthetic, gross misplacement of electrode outside cord or even in subarachnoid space
Contractions in neck stronger on contralateral side
as monitored by the OWL equipment. Analgesia to both superficial and deep painful stimuli is sought as the lesion-making proceeds and power in the ipsilateral leg should be repeatedly tested. Sacral sparing should be avoided if the pain is located in lower dermatomes. Following completion of an adequate lesion, there may be temporary reduction of ipsilateral hemithorax excursion but this is usually transient. The ipsilateral lower extremity may demonstrate paresis at this stage but if the patient is still able to lift the leg even weakly off the operating table, the end result is usually satisfactory. Post operatively, the patient can be nursed in the regular ward but must be cautioned about the risk of post operative ipsilateral leg weakness which might cause him to fall should he get out of bed alone. When respiratory difficulties or
No tetanization anywhere, sensory experience of contralateral warmth, sometimes pain, cold, or paresthesias. Suprathreshold stimulation may produce ipsilateral leg and/or neck tetanization
Other observations Electrode appears anterior in x-ray. Electrode appears dorsal in x-ray. Often follows difficulty in electrode positioning. X-ray positioning appears ideal.
other problems occur, he should be nursed in the intensive care area. Urinary retention should be expected but usually passes once the patient is up and about. Preoperative analgesic narcotics can nearly always be reduced to a modest level post operatively; addiction and withdrawal symptoms are rare. Percutaneous cordotomy can be done under general anesthetic without paralysis in the same way, missing only the very important feature of sensory responses to 100 Hz stimulation.
Results In our patients, the spinothalamic tract was lesioned on the first attempt in 95.5% of patients, and in 99% after early repetition. Ninety-eight
Percutaneous cordotomy
. Table 126‐3 Published success after unilateral percutaneous cordotomy – percent of patients with pain relief Complete
Significant
Reference
63 75 77 75 79 – 75 –
– 96 89 83 – 68 – 79.1 50% after 3 months 82.3 75 80 71 87 92.5 89 87.8
54 55 56 57 58 59 53 60
82 59–96 84 81.1 89
67 68 46 68 69
71 – – – 64 76 – 74.5 90 immediately 84 at 3 months 61 at 1 year 43 at 1–5 years 37 at 5–10 years 64 – 72 – –
43 61 62 63 64 65 66 45 47
percent had apparently adequate analgesic levels at this stage. However, only 94.4% had immediate satisfactory pain relief, persisting in 84% to the latest follow-up: Rosomoff et al. [47] reported 90% immediate pain relief and 84% at 3 months. After 1 year, 61% had adequate analgesia, 43% at 1–5 years, 37% after 5–10 years. In a review of 136 of our patients, inadequate pain relief contralateral to the cordotomy lesion affected 7% because the analgesic level was inadequate to begin with or because it fell post operatively. In 2%, disease progress caused new pain especially above the originally adequate level of analgesia. 39.8% suffered persisting or new pain below an originally adequate analgesia level, usually the result of neuropathic pain that fails to respond to cordotomy, brought on by the
126
progress of cancer causing lumbosacral plexus damage. In 5% the cause of the persisting pain was unclear, 1% developed post cordotomy dysaesthesia and 40.8% mirror pain. Mirror pain is usually less severe than the original but may require a second-side cordotomy as may significant bilateral pain prior to cordotomy. It has been suggested [48–52] that this is the result of opening up of previously inactive synapses in the cord allowing stimuli on the post operatively analgesic side to trigger pain on the opposite side of the body. If pain recurs over time, usually in patients who did not suffer from cancer, and analgesia fades, repeated cordotomy in our patients yielded 50% pain relief despite achieving adequate analgesia; observations that are difficult to explain since there is no evidence published of regeneration in the spinothalamic tract of which this author is aware [44–46,53]. > Table 126‐3 lists the results of others.
Complications > Table 126‐4 lists the complications of percutaneous cordotomy. Published mortality ranged up to 6.2% after unilateral cordotomy, severe respiratory complications up to 27%, mild respiratory problems up to 65.7%. Nearly every patient develops Homer’s syndrome post operatively though it is usually transient. The incidence of respiratory complications probably reflects patient selection since a higher level of analgesia must be achieved in patients with neck or upper extremity pain and the risk is higher in those whose respiratory function has already been prejudiced on one side by cordotomy, phrenic nerve, or severe pulmonary problems. Avoiding such patients reduces the mortality. Post operative mortality also reflects life expectancy associated with the primary disease. One-half percent of the author’s cases died of their disease
2143
Homer’s syndrome Neck pain Other significant Other transient, minor Reference
3.2
13
54
56
14.1 (0.7 severe)
2.9
11
18
2
5.5
3.2
2
1.5
69
20
2.5
6.2
58
2.6
5.3
10.5
7.9
9
59
50
7.5
3
8.7
8
4
48.5
6
5.7
4.5
60
1 50
2.35
2–100
1–15
0.5 20
61
100
8
63
94
8.7
7.2
10.1
1.4
62
100
100
8–20
64
26
100
6
6
19
3
69
4
5
Unilateral 6.2
65
Most
6.8
8.7
31
3.9
3.9
4.2
66
58
42
16
10.5
4.7
16
70
Most
6.1
4.3
2.4
0.6
0.6
47
1
10
2
25
6
2
3
1
68
>100
–’>20
0–8.7
–’> 19
100
1–2
2.6 27
0–6
46
23.0 (1/ 2 persist) 2.4 0.3 0.6
8.3
0
1.8
2.1
7.0
3.5
17.9
1.0
0.3
1.2
0.3
70
0.7
1.1
1.5
0.8
3.3
69
4
15
2
39
5
2
2
59
54.5
36.3
9.9
18.1
27.0
9.9
71
59
18.1
67.8
13.6
4.5
29.2
27.2
63
100
36.1
58
36.1
2.8
67
Common
6.7
0
0
0
0
Bilateral
68
40
+100
–’>67
–’>36
–
46
23 (1/4 persist) 1.6 1.6 1.6
0
8.1
9.8
9.8
21.3
29.5
1.6
3.2
1.6
1.6
126
Mild respiratory problems Transient respiratory problems Significant paresis or ataxia Mild or transient paresis or ataxia Significant worsening of micturition Mild, transient worsening of micturition Transient bowel incontinence Transient hypotension Contralateral limb weakness Postcordotomy dysesthesia
Death (mostly respiratory Severe respiratory failure
. Table 126‐4 Complications of Percutaneous Cordotomy
2144 Percutaneous cordotomy
Percutaneous cordotomy
before the 30th postoperative day, but in some of the published series 11% did so. Post cordotomy dysesthesia cannot be avoided and usually affects up to 10% of patients to some degree, 5% significantly. Interference with bladder function is unlikely (up to 12%) after cervical cordotomy because the bladder control path, unlike the reticulospinal respiratory control path, is bilaterally distributed. However, since the bladder control pathways lie close to the spinothalamic tract [72–74], bladder control is at risk after bilateral cordotomy (see > Figure 126‐1).
Bilateral Percutaneous Cordotomy In our experience, the success and complication rates of bilateral percutaneous cordotomy are the square of the rates for the unilateral operation except for respiratory and urinary complications. Patients who are at high risk for respiratory complication can be identified reflecting the risk for unconscious respiratory failure due to lesioning of the reticulospinal tract (Ondine’s curse). However, voluntary respiration mediated by the corticospinal tract is not affected. It is probably wise to monitor patients after bilateral cordotomy for about 3 days in an ICU with blood gas monitoring. Brief periods of ventilatory assistance are usually all that are required if the blood gases deteriorate. Urinary difficulties occur because the tract controlling the bladder is so close to the spinothalamic tract. It is, however, bilaterally distributed, unlike the reticulospinal tract for respiratory function, which is strictly ipsilaterally distributed, so that bladder failure occurs only if both tracts are affected either by bilateral cordotomy or cordotomy plus pre-existing disease [75,76]. The average survival after percutaneous cordotomy was 3 months in our series which consisted mostly of cancer cases. As time goes
126
on, in non-cancer patients who do not die of their disease, the incidence of pain recurrence increases, for virtually no pain operation, even those for tic douloureux, gives permanent pain relief. In tic douloureux, for example, 40–50% show recurrence within 15 years after microvascular decompression and 80% after 10 years following radiofrequency lesions [77].
Neuropathic Pain We have been particularly interested in the history of neuropathic pain and have identified three basic types. The commonest is a steady burning or dysesthetic discomfort which appears to be unrelated to transmission in the pain pathways but rather reflects loss of modulation in the brain itself. Sometimes, however, such patients have painful shocks of pain in the same site or else they show allodynia and/or hyperpathia. These latter two features appear to be relieved by cordotomy, whereas the steady pain is not, presumably because they depend on impulse transmission in pain pathways.
References 1. De Conno F, Panzeri C, et al. Palliative care in a National Cancer Center: Results in 1987 vs. 1993 vs. 2000. J Pain Symptom Manag. 2003;25(6):499-511. 2. Spiller WG. The occasional clinical resemblance between caries of the vertebrae and lumbothoracic syringomyelia and the location within the spinal cord of the fibres for the sensation of pain and temperature. Univ Penn Med Bull. 1905;18:147-54. 3. Spiller WG, Martin E. The treatment of persistent pain of organic origin in the lower part of the body by division of the anterolateral column of the spinal cord. JAMA 1912;58:1489-90. 4. Mullan S, Harper PV, Hekmatpanah J, et al. Percutaneous interruption of spinal-pain tracts by means of a strontium-90 needle. J Neurosurg. 1963;20:931-9. 5. Mullan S, Hekmatpanah J, Dobben G, Beckman F. Percutaneous intramedullary cordotomy utilizing the unipolar anodal electrolytic system. J Neurosurg. 1965;22:548.
2145
2146
126
Percutaneous cordotomy
6. Rosomoff HL, Carrol F, Brown J, Sheptak P. Percutaneous radiofrequency cervical cordotomy: Technique. J Neurosurg. 1965;23:639-44. 7. Sweet WH, Mark VH, Hamlin H. Radiofrequency lesions in the central nervous system of man and cat: including case reports of eight bulbar pain-tract interruptions. J Neurosurg. 1960;17:213-25. 8. Onofrio BM. Cervical spinal cord and dentate delineation in percutaneous radiofrequency cordotomy at the level of the first to second cervical vertebrae. Surg Gynecol Obstet. 1971;133:30-4. 9. Gildenberg PL, Zanes C, Flitter MA, et al. Impedance monitoring device for detection of penetration of the spinal cord in anterior percutaneous cervical cordotomy: Technical note. J Neurosurg. 1969;30:87-92. 10. Hitchcock ER, Tsukamoto Y. Distal and proximal sensory responses during stereotactic spinal tractotomy in man. Ann Clin Res. 1973;5:68-73. 11. Sweet WH, White JC, Selverstone B, Nilges R. Sensory responses from anterior roots and from surface and interior of spinal cord in man. Trans Am Neuro Assoc. 1950;75:165. 12. Taren JA. Physiologic corroboration in sterotaxic high cervical cordotomy. Confin NeuroI. 1971;33:285-90. 13. Taren JA, Davis R, Crosby EC. Target physiologic corroboration in stereotaxic cervical cordotomy. J Neurosurg. 1969;30:569-84. 14. Tasker RR, Organ LW. Percutaneous cordotomy: physiological identification of target site. Confin Neurol. 1973;35:110-7. 15. Tasker RR, Organ LW, Smith KC. Physiological guidelines for the localization of lesions by percutaneous cordotomy. Acta Neurochir Suppl (Wien) 1974;21:111-7. 16. Kanpolat Y, Deda H, Akyar S, Bilgic S. CT-guided percutaneous cordotomy. Acta Neurochir Suppl (Wien) 1989;46:67-8. 17. Lin PM, Gildenberg PL, Polakoff PP. An anterior approach to percutaneous lower cervical cordotomy. J Neurosurg. 1966;25:553-60. 18. Crne BL, Todd EM, Carregal EJA. Posterior approach for high cervical percutaneous radiofrequency cordotomy. Confin Neurol. 1968;30:4l. 19. Hitchcock ER. An apparatus for sterotactic spinal surgery: a preliminary report. J Neurosurg. 1969;31:386-92. 20. Bekar A, Kocaeli H, et al. Bilateral high-level percutaneous cervical cordotomy in cancer pain due to lung cancer: a case report. Surg Neurol. 2007;67(5):504-7. 21. Raslan AM. Percutaneous computed tomography-guided transdiscal low cervical cordotomy for cancer pain as a method to avoid sleep apnea. Stereot Funct Neuros. 2005;83(4):159-64. 22. Yegul I, Erhan E. Bilateral CT -guided percutaneous cordotomy for cancer pain relief. Clin Radiol. 2003;58 (11):886-9. 23. Kanpolat Y, Savas A, et al. CT-guided percutaneous selective cordotomy for treatment of intractable pain in
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
patients with malignant pleural mesothelioma. Acta Neurochirurgica 2002;144(6):595-9; discussion 599. Tasker RR, Dostrovsky JO. Deafferentation and central pain. In: Wall PD, Melzack R, editors. Textbook of pain. 2nd ed. Edinburgh: Churchill Livingstone; 1989. p. 154-80. Tasker RR. The problem of deafferentation pain in the management of the patient with cancer. J Palliat Care 1987;2:812. Gybels JM. Indications for the use of neurosurgical techniques in pain control. In: Bond MR, Charlton IE, Woolf J, editors. Proceedings of the VI World Congress on Pain. Amsterdam: Elsevier; 1991. p. 475-82. Willis WD. The pain system: The neural bases of nociceptive transmission in the mammalian nervous system. In: Gildenberg PL, editor. Pain and headache. vol 8. Basel: Karger; 1985. Noordenbos W, Wall PD. Diverse sensory functions with an almost totally divided spinal cord: a case of spinal cord transection with preservation of one anterolateral quadrant. Pain 1976;2:185-95. Wall PD. The design of experimental studies in the future development of restorative neurology of altered sensation and pain. In: Dimitrijevic MR, Wall PD, Lindblom U, editors. Recent achievements in restorative neurology. vol 3: altered sensation and pain. Karger; Basel: 1990. p. 197-205. Lahuerta J, Bowsher D, Campbell J, Lipton S. Clinical and instrumental evaluation of sensory function before and after percutaneous anterolateral cordotomy at cervical level in man. Pain 1990;42:23-30. DiPiero V, Jones AKP, Ianotti F, et al. Chronic pain: a PET study of the central effects of percutaneous high cervical cordotomy. Pain 1991;46:9-12. Rosso T, Aglioti SM, et al. Functional plasticity in the human primary somatosensory cortex following acute lesion of the anterior lateral spinal cord: neurophysiological evidence of short-term cross-modal plasticity. Pain 2003;101(1–2):117-27. Meyerson BA, Arner S, Linderoth B. Pelvic cancer pain (somatogenic pain): pros and cons of different approaches to the management of pelvic cancer pain. Acta Neurochir Suppl (Wien)1984;33:407-19. Belmusto L, Brown E, Owens G. Clinical observations on respiratory and vasomotor disturbances as related to cervical cordotomies. J Neurosurg. 1963;201:225-32. Belmusto L, Woldring S, Owens G. Localization and patterns of potentials of the respiratory pathways in the cervical spinal cord in the dog. J Neurosurg. 1965; 22:277-83. Fox JL. Localization of the respiratory pathway in the upper cervical spinal cord following percutaneous cordotomy. Neurology 1969;19:1115-8. Hitchcock E, Leece B. Somatotopic representation of the respiratory pathways in the cervical cord of man. J Neurosurg. 1967;27:320-9.
Percutaneous cordotomy
38. Mullan S, Hosobuchi Y. Respiratory hazards of high cervical percutaneous cordotomy. J Neurosurg. 1968;28:291-7. 39. Nathan PW. The descending respiratory pathway in man. J Neurol Neurosurg Psychiatry 1963;26:487-99. 40. Rosomoff HL, Krieger AJ, Kuperman AS. Effects of percutaneous cervical cordotomy on pulmonary function. J Neurosurg. 1969;31:620-7. 41. Tenicela R, Rosomoff HL, Feist J, Safar P. Pulmonary function following percutaneous cervical cordotomy. Anesthesiology 1968;29:7-16. 42. Price C, Pounder D, et al. Respiratory function after unilateral percutaneous cervical cordotomy. Pain Symptom Manag. 2003;25:459-63. 43. Izumi J, Hirose Y, Yazaki T. Percutaneous trigeminal rhizotomy and percutaneous cordotomy under general anesthesia. Stereotact Funct Neurosurg. 1992;59:62-8. 44. Tasker RR. Percutaneous cordotomy-the lateral high cervical technique. In: Schmidek HR, Sweet WH, editors. Operative neurosurgical techniques: indications, methods and results. New York: Grune and Stratton; 1982. p. 1137-53. 45. Tasker RR. Percutaneous cordotomy: the lateral high cervical technique. In: Schmidek HR, Sweet WH, editors. Operative neurosurgical techniques: indications, methods and results. Orlando, FL: Grune and Stratton; 1988. p. 1191-205. 46. Tasker RR. Percutaneous cordotomy. In: Schmidek HR, Sweet WH (editors). Operative neurosurgical techniques. 3rd ed. Philadelphia: Saunders; 1995. p. 1595-611. 47. Rosomoff HL, Pape I, Loeser JD, Bonica JJ. Neurosurgical operations on the spinal cord. In: Bonica JJ editor. The management of pain. 2rd ed. Philadelphia: Lea and Febiger; 1990. p. 2067-81. 48. Bowsher D. Contralateral mirror-image pain following anterolateral cordotomy. Pain 1988;33:63-5. 49. Ischia S, Ischia A. A mechanism of new pain following cordotomy (letter) Pain 1988;32:383-4. 50. Nagaro T, Amakawa K, Kimura S, Arai T. Reference of pain following percutaneous cervical cordotomy. Pain 1993;53:205-11. 51. Nagaro T, Kumura S, Arai T. A mechanism of new pain following cordotomy: reference of sensation. Pain 1987;30:89-91. 52. Nathan PW. Reference of sensation at the spinal level. J Neurol Neurosurg Psychiatry 1956;19:88-100. 53. Lipton S. Percutaneous cervical cordotomy. Acta Anaesthesiol Belg. 1981;32:81-5, with additional personal communication. 54. O’Connell JEA. Anterolateral chordotomy for intractable pain in carcinoma of the rectum. Proc R Soc Med. 1969;62:31-3, 1223–5. 55. Lorenz R. Methods of percutaneous spinothalamic tract section. In: Krayenbiihl H, editor. Advances and technical standards in neurosurgery. vol 3. Vienna: Springer; 1976. p. 123-45. 56. Grote W, Roosen CW. Dieperkutane chordotomie. Langenbecks Arch Chir. 1976;342:101-8.
126
57. Yon Schrottner O. Die perkutane zervikale anterolaterale Chordotomie. Wien Klin Wochenschr. 1978;90:372-4. 58. Meglio M, Cioni B. The role of percutaneous cordotomy in the treatment of chronic cancer pain. Acta Neurochir (Wien) 1981;59:111-21. 59. Kiihner A. La cordotomie percutane´e: Sa place actuelle dans 1a chirurgie de 1a douleur. Anesth Analg 1981; 38:357-9. 60. Ventafridda V, De Conno F, Fochi C. Cervical percutaneous cordotomy. In: Bonica JJ, Ventafridda V, Pagni CA, editors. Advances in pain research and therapy. vol 4. Management of superior sulcus syndrome (Pancoast syndrome). New York: Raven Pres; 1982. p. 185-98. 61. Siegfried J, Ku¨hner A, Sturn V. Neurosurgical treatment of cancer pain: recent results. Cancer Res. 1984; 89:148-52. 62. Lipton S. Percutaneous cordotomy. In: Wall PD, Melzack R (editors). Textbook of pain. Edinburgh: Churchill Livingstone; 1984. p. 632-8. 63. Ischia S, Luzzani A, Ischia A, Pacini L. Role of unilateral percutaneous cervical cordotomy in the treatment of neoplastic vertebral pain. Pain 1984;19:123-31. 64. Lahuerta T, Lipton S, Wells JCD. Percutaneous cervical cordotomy: results and complications in a recent series of 100 patients. Ann R Coll Surg Engl. 1985;67:41-4. 65. Ischia S, Ischia A, Luzzani A, et al. Results up to death in the treatment of persistent cervico-thoracic (Pancoast) and thoracic malignant pain by unilateral percutaneous cervical cordotomy. Pain 1985;21:339-55. 66. Farcot J-M, Mercky F, Tritschler J-L, Schaeffer F. Cordotomies cervicales percutane´es dans les douleurs cancereuses throraciques primitives ou secondaires (a propos de 19 cas). Agressologie 1988;29:87-9. 67. Amano K, Kawamura H, Tanikawa T, et al. Bilateral versus unilateral percutaneous high cervical cordotomy as a surgical method of pain relief. Acta Neurochir Suppl (Wien) 1991;52:143-5. 68. Tasker RR. Ablative central nervous system lesions for control of cancer pain. In: Arbit E (editor): Management of cancer-related pain. Mt, Kisko, NY: 1993 Futura p. 231-55. 69. Ischia S, Luzzani A, Ischia A, et al. Subarachnoid neurolytic block (L5-S1) and unilateral percutaneous cervical cordotomy in the treatment of pain secondary to pelvic malignant disease. Pain 1984;20:139-49. 70. Stuart G, Cramond T. Role of percutaneous cervical cordotomy for pain of malignant origin. Med J. Aust. 1993;158:667-70. 71. Lipton S. Percutaneous cervical cordotomy and the injection of the pituitary with alcohol. Anaesthesia 1978;33:953-7. 72. Palma A, Hozer J, Cuadra O, Palma J. Lateral percutaneous spinothalamic tractotomy. Acta Neurochir (Wien) 1988;93:100-3. 73. Koulousakas A, Nittner K. Bilateral C1–C2 cordotomies: can complications be avoided? Appl Neurophysiol. 1982;45:500-3.
2147
2148
126
Percutaneous cordotomy
74. Nathan PW, Smith MC. The centripetal pathway from the bladder and urethra within the spinal cord. J Neurol Neurosurg Psychiatry 1951;14:262-80. 75. Nathan PW, Smith MC. The centrifugal pathway for micturition within the spinal cord. J Neurol Neurosurg Psychiatry 1958;21:177-89.
76. Nathan PW. Reuslts of anterolateral cordotomy for pain in cancer J Neurol Neurosurg Psychiatry 1963;26:353-62. 77. Loeser JD. Tic douloureux and atypical face pain. In: Wall PD, Melzack R, editors. Textbook of pain. 3rd ed. Edinburgh: Churchill Livingstone; 1994. p. 699-710.
139 Peripheral Nerve Stimulation for Neuropathic Pain A. G. Shetter
The gate control theory of pain [1], which was proposed in 1965, predicted that the activation of large-diameter afferent fibers would have an inhibitory effect on small-diameter afferent fibers at the spinal cord level and would reduce the central transmission of pain messages. Shortly afterward, Wall and Sweet [2] tested this hypothesis clinically by electrically stimulating peripheral nerves and nerve roots with surface or subcutaneous electrodes in a group of eight patients with chronic pain syndromes. There were three patients with posttraumatic neuralgias, and all of these patients had their pain temporarily alleviated by stimulation. Wall and Sweet [2] noted that pain relief could outlast the period of stimulation by many minutes and that paresthesias had to cover the area of pain to be effective. They also cautioned that the ‘‘clinical implications [of this technique] are at present equivocal because two of the first group of patients, who were stimulated many times per day, reported a decreased effect on their pain after several months [2].’’ These initial observations, made over 40 years ago in a small group of patients, have been confirmed by subsequent investigators and have proved to be remarkably accurate. On the basis of this favorable experience, White and Sweet [3] implanted a permanent peripheral nerve-stimulating system in a patient with pain after a prior median nerve injury. This case represents the first use of neuroaugmentative surgery for pain control. Peripheral nerve stimulation (PNS) has since become established as an effective technique for treating carefully selected patients with chronic pain secondary to nerve injury. #
Springer-Verlag Berlin/Heidelberg 2009
Mechanism of Action Although the exact mechanism by which PNS reduces chronic pain is unknown, experimental evidence suggests that it may have both a central effect and a peripheral effect on acute pain perception. Chung and colleagues [4,5] recorded from spinothalamic cells in the lumbosacral spinal cord of anesthetized monkeys. They found that the response of those neurons to both noxious thermal and electrical stimuli could be markedly inhibited by applying a repetitive conditioning stimulus to the common peroneal nerve or tibial nerve. The inhibition often outlasted the period of conditioning stimulation by 20–30 min. This effect was observed in both spinalized and intact animals, indicating that spinal cord neuronal circuitry must be partially responsible. Inhibition was segmentally organized, since stimulation of a nerve innervating the receptive field of the spinothalamic neuron that was tested had the most potent effect. However, stimulation of the nerves innervating the limb other than the one on which the receptive field was found also produced a degree of inhibition. Woolf and associates [6] suggested that stimulation of alpha fibers may activate descending inhibitory pathways from the brain stem in addition to its effect on dorsal horn inhibitory currents. They applied non-noxious electrical stimulation to the base of the tail in rats and measured the flexor withdrawal response after immersion of the tail in hot water. Electrical stimulation in the intact animals produced a marked prolongation in the withdrawal reaction
2350
139
Peripheral nerve stimulation for neuropathic pain
time. An antinociceptive effect also was seen in rats with complete spinal cord transection at T10–T11, but it was less pronounced. Both the spinal and the supraspinal inhibitory mechanisms could be blocked by naloxone. Pretreatment depletion of 5-hydroxytryptamine attenuated the effect of electrical stimulation in the intact animals but not in the spinal preparations. Positron emission tomography (PET) studies in patients undergoing PNS have also shown changes in central neuronal activity. Matharu et al. [7] used PET scans to measure regional cerebral blood flow in eight patients undergoing bilateral occipital nerve stimulation for treatment of migraine headaches. Paresthesias produced by stimulation modulated blood flow in the left pulvinar and anterior cingulate cortex, while pain relief correlated with blood flow alterations in the dorsal rostral pons, anterior cingulate cortex, and cuneus. In contrast to the central mechanisms for PNS analgesia proposed by these authors, other investigators have demonstrated effects that are peripherally mediated. Ignelzi and Nyquist [8] stimulated cat superficial radial nerves for 5- to 10-min intervals using stimulus parameters comparable to those employed for clinical PNS. They then recorded from single nerve fibers proximal to the site where the conditioning stimulus was applied. The majority of the fibers tested demonstrated excitability changes after repetitive stimulation. These changes included transient slowing of conduction velocity, an increase in electrical threshold, and/or a decrease in response probability. Wall and Gutnick [9] studied the physiological properties of sciatic nerve neuromas created in rats. Recordings from dorsal root filaments originating from a neuroma disclosed the presence of axons that were continuously active in the absence of a peripheral stimulus. This degree of spontaneous activity was never seen in normal nerves. Tetanic stimulation of nerve fibers proximal to a neuroma markedly suppressed the rate of spontaneous firing for as long as 1 h. Similar
changes in excitability were not observed in nerve fibers with intact sensory endings. These authors [9] suggested that the clinical pain relief observed after PNS might partially reflect antidromic invasion of damaged nerve fibers, which inhibits ongoing neuronal activity of the type observed in their animal model. Experimental studies such as these indicate that PNS may relieve pain by acting at multiple sites and through various mechanisms. The likelihood that part of its effect may occur peripherally has practical consequences, because it implies that PNS may produce results different from those of spinal cord stimulation in treating certain pain syndromes.
Results After the initial application of PNS in 1965, a number of clinical series were reported during the ensuing 10–15 years [10–15]. The experience was generally favorable, but the use of PNS declined in the 1980s as surgeons active in neuroaugmentative procedures for pain control turned their attention to spinal cord and deep brain stimulation. In recent years, there has been renewed interest in this technique [16–21]. Evaluation of the literature to determine the true efficacy of PNS is difficult for several reasons. There are no uniform criteria for the definition of success, and patient selection criteria vary considerably. Reported follow-up intervals are often less than 1–2 years despite the known tendency for surgical results to decline with time. There is also considerable confusion and lack of standardization regarding the use of diagnostic categories such as reflex sympathetic dystrophy (RSD), causalgia, post–traumatic neuralgia, and complex regional pain syndrome (CRPS). Finally, no randomized series have compared the results of PNS with best medical care or with alternative surgical options such as spinal cord stimulation and peripheral nerve neurolysis. With these reservations in mind, some tentative conclusions can be drawn.
Peripheral nerve stimulation for neuropathic pain
Most patients treated with PNS have had pain after trauma to a single peripheral nerve. Typically, but not always, their symptoms may be accompanied by physical signs of sympathetic nervous system hyperactivity or trophic tissue changes. For pain of this type, 40–60% of patients experience worthwhile relief after the implantation of a peripheral nerve-stimulating system [10–20]. However, variation among series is wide, and meaningful comparisons are impossible for the reasons outlined above. Some surgeons have observed that success is more likely when treating purely sensory nerves (e.g., superficial radial nerve) than mixed nerves [15] and that upper extremity pain is more likely to respond than pain in the lower extremities [10,13,19]. The results with sciatic nerve stimulation have tended to be disappointing in part because of the difficulty of activating sensory fibers without producing excess motor drive in this large, functionally important nerve. Initial experience using PNS to treat cancer pain [10], sciatica, pain after lower back surgery and nerve root injury and idiopathic pain has tended to be unfavorable. Although there have been isolated instances of success in all these categories, success rates seem to be substantially less than 50%. One of the best early reports on PNS was that of Nashold and associates [13]. It is distinguished by its lengthy follow-up (4–9 years) and unusually stringent criteria for success. Patients so designated reported more than 90% subjective pain relief. The patients were off all analgesic medications and continued to use the stimulator regularly. Over an 8-year interval (1970–1977), 35 patients were implanted. The long-term success rate for patients with upper extremity pain secondary to peripheral nerve injury was 53% (9 of 17 patients). Their results with sciatic nerve implantation were less favorable (31% success rate) because of the technical difficulty of stimulating this nerve effectively and the inclusion of some patients whose pain was related to lumbar spine pathology. If the
139
latter group is eliminated, their long-term success rate increases to 38% (5 of 13 patients). The authors believed that their results improved over the course of the study and that future advances in equipment technology and patient selection criteria would lead to better results in the future. By far the largest experience with PNS is that of Racz and colleagues [16,17]. Much of their information is contained in text books rather than peer-reviewed journals. A publication [17] in 1997 described 125 nerve implants performed in 117 patients. Two-thirds of the patients had pain in an upper extremity. Some had poorly localized pain [‘‘whole-body reflex sympathetic dystrophy (RSD)’’] that was thought to have been triggered by an initial injury to a peripheral nerve. Stimulation of the affected nerve was observed to help with pain outside the sensory distribution of the nerve itself. There were 101 patients available for followup at postoperative intervals of 1–53 months. Seventy-eight (77%) patients were described as having ‘‘good to excellent pain relief.’’ In recent years, several groups have reported moderately large series of peripheral nerve implants that seem to equal or possibly exceed the results described by earlier investigators. Hassenbusch et al. [18] prospectively evaluated 34 patients with RSD or CRPS Type II who had 4-contact plate electrodes placed on a single nerve in an upper or lower extremity. During a 2–4 day trial, 32 patients experienced more than 50% pain reduction and had their electrode connected to a batterypowered internal pulse generator for long-term use. At a mean follow-up of 2.2 years, 63% were considered to have a good or fair outcome characterized by pain reduction of at least 25% associated with improvement in vasomotor tone, trophic changes, or motor function. Patients with pain and clinical findings in the distribution of a single major peripheral nerve were more likely to benefit than those with more diffuse pain. Eisenberg et al. [19] performed peripheral nerve implants in 46 patients and followed them for a median of 10.8 years. All patients had
2351
2352
139
Peripheral nerve stimulation for neuropathic pain
medically intractable pain related to injury of a single peripheral nerve, and all experienced complete but temporary pain relief with local anesthetic blockade of the nerve. None of their patients were thought to meet the diagnostic criteria for CRPS. At last follow-up, 78% of the patients had a good result, which was defined as a more than 50% reduction in pain intensity on the Visual Analogue Scale with abstinence from analgesic medications. Both cuff and plate electrodes were used, with no difference in outcomes between the two systems. Given the length of follow-up, the favorable results reported in this series are particularly impressive. Mobbs et al. [20] performed a temporary trial of peripheral nerve stimulation in 42 patients and then implanted permanent stimulating systems in 38 patients. All of their patients had traumatic or iatrogenic injury to a single nerve, but they do not mention whether any of their patients were considered to have CRPS. At a mean follow-up of 31 months, 61% of the patients receiving a permanent implant had more than 50% decrease in pain intensity and 47% reported a significant improvement in activity levels. The results of nerve stimulation in the lower extremities were substantially poorer than those for the upper extremities. Eventually, the authors discontinued lower extremity peripheral nerve stimulation in favor of spinal cord stimulation. Their series also differs from other reports in that they implanted a modified plate electrode with a Silastic mesh wrapped circumferentially around the nerve. In 1999 Weiner and Reed [22] described the successful treatment of occipital neuralgia through stimulating electrodes inserted percutaneously in the vicinity of the greater and lesser occipital nerves. Since then, use of this technique has increased and its indications have expanded to include cervicogenic headache, transformed migraine, cluster headache, and tension headache. Small, retrospective series from a number of centers [23] have indicated a positive response in 50–90% of implanted patients. Weiner [24] has reported 75% good to excellent long-term pain
control in more than 150 patients treated from 1993 through 2005. In this group of patients, pain scores on the Visual Analogue Scale decreased from a preoperative mean of 9 (range 5–10) to a postoperative mean of 3 (range 0–6). The apparent success of occipital nerve stimulation encouraged other investigators to attempt percutaneous electrode placement adjacent to the supraorbital and infraorbital nerves for the treatment of postherpetic and posttraumatic trigeminal neuralgia. Good outcomes have been described in several small series [23,25]. Despite the promising results for occipital nerve stimulation, most reports are based on retrospective analysis of patients with a variety of headache syndromes who have been followed for relatively short intervals. Well-designed prospective studies with rigid entry and outcome criteria and longer follow-up data are needed before this procedure can be endorsed for widespread use.
Complications Although the efficacy of PNS is difficult to state with certainty based on review of the literature, its complication rate is well known. The most frequently encountered problem is the need for surgical revision to repair breakage of the lead wire or to reposition electrodes that are not producing satisfactory paresthesias. Surgical revision was required in as many as 50% of patients in earlier series [11]. In recent years the incidence of technical malfunction probably has declined as a result of improvements in the stimulating equipment. Multicontact-electrode systems that permit the physician to change electrode combinations and polarities externally without the need for surgical exploration have been particularly helpful in this regard. In most series the infection rate has averaged 4–5% [10,11,13–20]. A proportion of these infections has been superficial and did not require removal of the implanted material.
Peripheral nerve stimulation for neuropathic pain
Most of the earlier experience with PNS involved the use of electrodes mounted in a Silastic cuff that was wrapped circumferentially around the implanted nerve. If the cuff was too tight or if excess scarring occurred in response to the foreign body, nerve damage could result. In an attempt to avoid this problem, more recent investigators [16–18,21] have advocated the use of a longitudinally oriented electrode that does not encircle the nerve. However, the likelihood of nerve injury with a circumferential electrode is low. A review of five series [10,11,13–15] involving 228 patients who were implanted with cuff electrodes disclosed five instances of nerve injury, for an incidence rate of 2.2%. Adverse occurrences other than technical malfunction, infection, and nerve injury are infrequent and include those that might be experienced after any peripheral nerve exploration under general anesthesia. It is apparent that the risk of injury related to PNS is low and that complications are rarely serious.
Surgical Techniques The operative technique for implanting a PNS system is straightforward but does require knowledge of peripheral nerve surgical anatomy. There are three components to an implant: a multicontact electrode, a radiofrequency receiver or a lithium battery-powered pulse generator, and a length of lead wire joining the electrode to the power source. At present no implantable hardware has been designed specifically for PNS. The Avery peripheral nerve cuff electrode used by many of the early implanters is no longer manufactured, and the Cyberonics electrode (Houston, TX) for vagus nerve stimulation is too small to encompass a major peripheral nerve. All of the more recent devices in use were originally developed for spinal cord stimulation and have been adapted for peripheral nerve applications. The only device approved by the Food and Drug Administration for PNS is
139
made by Medtronics (Minneapolis, MN). It employs a radiofrequency transmitter-receiver system as the power source rather than a totally implantable lithium battery-powered pulse generator. All other systems developed for spinal cord stimulation are used for PNS on an off-label basis [21]. Over the past 15 years, the most commonly utilized electrode described in the literature has been a rectangular plate or paddle with four oval contact points aligned longitudinally. The plate or paddle is mated to an internal pulse generator powered by a lithium battery, because most patients prefer the convenience of a totally implantable power source compared to a radiofrequency transmitter-receiver system, which requires an antenna taped to the skin. The patient is provided with an external programmer that allows him or her to activate the power source and to alter the intensity, rate, and pulse width of the stimulus. A physician-controlled programmer can be used to change combinations of electrode contact points or polarities. These variables are adjusted as needed to produce sensory paresthesias covering the area of pain without excess motor drive. Several manufacturers of spinal-cord stimulating systems now make a variety of plate electrodes with as many as 16 contact points arrayed in bipolar, tripolar, or asymmetric combinations. They can be paired with implanted pulse generators that allow transcutaneous battery recharging, thereby decreasing the need for frequent battery replacement. These newer devices have theoretical advantages over the simpler systems described in most of the PNS literature, but experience with their use is still limited. Percutaneous electrodes have been used as an alternative to plate electrodes for occipital nerve stimulation [25]. These electrodes are thin, cylindrical leads that can be inserted through needles and have 4–8 contact points spaced at variable intervals along the distal shaft of the electrode. They require less surgical dissection than plate electrodes, but some surgeons believe that the
2353
2354
139
Peripheral nerve stimulation for neuropathic pain
leads are more prone to migrate. Paddle electrodes tend to require lower stimulus intensities for effective stimulation than percutaneous leads, which can prolong pulse generator battery life. This issue is less of a concern now that rechargeable battery systems are available. Both types of electrodes have been used to good effect by different investigators, and there is no clear evidence to favor one over the other for occipital nerve stimulation. Percutaneous implanted electrodes have also been used in a limited number of patients to stimulate branches of the trigeminal nerve for neuropathic facial pain syndromes [23,25]. The specific surgical dissection required for PNS depends on the nerve to be implanted. This information is available in standard textbooks on peripheral nerve anatomy. However, several general principles need to be considered. The stimulating electrode should be placed proximal to the presumed site of injury. In patients who have undergone prior peripheral nerve explorations, a new incision may be needed to visualize a more normal portion of the nerve. The exposure also should be planned so that the lead cable joining the electrode to the pulse generator or the radiofrequency receiver does not cross more than one joint space. Such placement lessens the likelihood of lead breakage resulting from repetitive flexion and extension. The pulse generator is relatively bulky and must be implanted subcutaneously in the infraclavicular region, lower abdomen, lateral thigh, or upper buttock to minimize incisional discomfort. In obese patients, the pulse generator should not be placed more than 1.5 cm below the surface of the skin to allow effective programming or recharging. The most commonly implanted nerves in the upper extremity are the median and ulnar nerves. These nerves usually are exposed at the midhumeral level with the pulse generator or receiver placed in a subcutaneous pocket approximately 1 inch inferior to the clavicle. The most frequently implanted lower extremity nerves are the posterior tibial and common peroneal nerves. In most
instances, these structures are best exposed above the medial malleolus and the tibial head, respectively. The power source is buried subcutaneously in the anterolateral thigh. The nerve need not be dissected over a greater distance than required to position the electrode. Care should be taken to preserve as much adventitial tissue intact around the nerve as possible. If a longitudinal electrode is used, a bed is created adjacent to the nerve and the electrode plate is sutured to soft tissues. The nerve surface must appose the electrode contact points satisfactorily, and pressure on the nerve by the existing lead cable must be avoided. Racz and colleagues [16,17] have advocated suturing a piece of fascia to the face of the plate electrode so that the electrode does contact the nerve directly. Because a fibrous envelope rapidly surrounds Silastic material implanted in the body, it is unclear whether the addition of a fascial barrier provides a further protective effect. An occipital nerve electrode is usually placed through a small midline incision at the level of C1 [26]. Using fluoroscopic guidance, either a percutaneous or a plate electrode is directed laterally toward the tip of the mastoid process. The electrode must be positioned in the subcutaneous tissues deeply enough to avoid skin erosion but superficially enough to avoid painful stimulation of the underlying suboccipital fascia and musculature. The nerve itself is not visualized directly. Depending on the location of the pain, bilateral or unilateral leads may be placed. Some surgeons prefer a lateral electrode insertion site just below and medial to the mastoid tip, with a percutaneous electrode extending up to or across the midline. A pulse generator may be positioned in the upper buttocks or infraclavicular area. The lead cable joining the electrode to the pulse generator or radiofrequency receiver should be tunneled subcutaneously between the two incisions before final electrode positioning so that the electrode is not inadvertently dislodged. Sometimes a third incision is needed midway between the first 2 to assist in subcutaneous passage of the
Peripheral nerve stimulation for neuropathic pain
cable. Some surgeons prefer to bring temporary external lead wires through the skin for a trial period of stimulation before implanting a permanent system. In this case, the sequence described above is performed in two stages. Patients are usually discharged 1–2 days after the final implant. They are instructed in the use of the stimulating equipment and encouraged to try different stimulus parameters empirically. Most patients select stimulus intensity and a pulse width that produce barely perceptible sensory paresthesias without activation of motor fibers. However, some patients report that brief periods of very high-intensity stimulation are more effective. Some individuals prefer a pulse rate of 100 Hz or higher, which gives a ‘‘buzzing’’ or ‘‘tingling’’ sensation, while others elect a lower stimulus rate. The amount of time required for stimulation also varies considerably from person to person, and no set guidelines can be given. Many patients can stimulate for relatively brief intervals and obtain pain relief for hours afterward. Others note that their pain returns to baseline levels within minutes after the cessation of stimulation. Such patients tend to use their units continuously during waking hours.
Patient Selection Criteria Patients with intolerable pain after peripheral nerve trauma are the best candidates for PNS. They may or may not have signs of sympathetic nervous system overactivity or underactivity. Common clinical examples of this type of problem include pain after penetrating injury of the peroneal nerve or median nerves, pain after excision of a Morton’s neuroma, and persistent pain after ulnar nerve transposition. Ideally, the pain is confined to the sensory distribution of a single nerve. Some authors believe that PNS can be used to treat pain that began with a discrete nerve injury but later
139
evolved into CRPS or RSD encompassing an entire limb [17,18]. This indication for PNS should be considered questionable pending further documentation. There also have been reports of implants involving two separate nerves [17] or elements of the brachial plexus [11]. Such implants may be warranted in unusual instances, but most surgeons favor spinal cord stimulation as a better alternative if the pain is diffuse rather than focal. Pain associated with sciatica, prior lower back surgery, cancer, and nerve root injury did not consistently benefit from PNS in earlier trials, and these syndromes should no longer be considered indications for this technique. There is no adequate documentation regarding the use of PNS for painful metabolic neuropathies or postherpetic neuralgia. Several tactics have been suggested to further refine the selection criteria for PNS. Sweet [15], Nashold and colleagues [13], and Eisenberg et al. [19] strongly advocate the routine preoperative use of nerve blocks. If pain is not stopped or markedly reduced on a temporary basis by local anesthetic blockade of the nerve that will be implanted, they believe that the patient should be excluded as a surgical candidate. Unfortunately, a favorable response to a nerve block does not guarantee success with PNS. Picazza and associates [14] analyzed the predictive value of transcutaneous neurostimulation for PNS. Patients who experienced pain reduction with a transcutaneous stimulator were more likely to benefit from the later implantation of a peripheral nerve electrode than were those who failed to respond. However, a substantial number of patients with a negative trial of transcutaneous stimulation were later helped by PNS. In recent years, several investigators have described the use of an externalized lead wire for a temporary trial of PNS before permanent implantation [18,20,21]. The lead extension from the electrode is tunneled away from the incision through a separate stab wound and is connected to an external battery-powered pulse generator.
2355
2356
139
Peripheral nerve stimulation for neuropathic pain
Stimulation continues for 2–7 days, and the patient is asked to make a subjective estimate of the degree of pain relief. Those with little or no improvement have the electrode removed, while the remainders undergo permanent implantation. We believe that all three screening methods have merit. Patients who do not experience marked pain reduction after local anesthetic blockade of a single peripheral nerve are not good candidates for electrode implantation. Individuals whose pain is lessened by transcutaneous neurostimulation tend to have a higher likelihood of success with PNS. However, a negative response should not be used to exclude patients who are otherwise appropriate candidates for this technique. Patients who do not experience significant pain relief after a temporary trial of PNS should not have a permanent system implanted.
Summary PNS has a good long-term success rate in treating carefully selected patients with pain in the distribution of a single traumatized peripheral nerve. More diffuse pain symptoms characteristic of CRPS or RSD may be helped by this technique, but the documentation for this indication is less clear-cut. Preliminary experience with occipital nerve stimulation for occipital neuralgia and various headache syndromes has shown promise and warrants further evaluation. Serious operative complications after PNS are rare. Improved equipment has decreased the incidence of technical malfunction. PNS should be considered the surgical procedure of choice for prosttraumatic neuralgias that do not respond to simpler treatments.
References 1. Melzack R, Wall PD. Pain mechanisms: a new theory. Science 1965;150:971-8. 2. Wall PD, Sweet WH. Temporary abolition of pain in man. Science 1967;155:108-9.
3. White JC, Sweet WH. Pain and the neurosurgeon: a fortyyear experience. Springfield, IL: Charles C Thomas; 1969. p. 895-6. 4. Chung JM, Fang ZR, Hori Y, et al. Prolonged inhibition of primate spinothalamic tract cells by peripheral nerve stimulation. Pain 1984;19:259-75. 5. Chung JM, Lee KH, Hori Y, et al. Factors influencing peripheral nerve stimulation produced inhabitation of primate spinothalamic tract cells. Pain 1984;19:277-93. 6. Woolf CJ, Mitchell D, Barrett GD. Antinociceptive effect of peripheral segmental electrical stimulation in the rat. Pain 1980;8:237-52. 7. Matharu MS, Bartsch T, Ward N, Frackowiak RS, Weiner R, Goadsby PJ. Central neuromodulation in chronic migraine patients with suboccipital stimulators: a PET study. Brain 2004;127 (Pt 1):220-30. 8. Ignelzi RJ, Nyquist JK. Excitability changes in peripheral nerve fibers after repetitive electrical stimulation: implications in pain modulation. J Neurosurg 1979;51:824-33. 9. Wall PD, Gutnick M. Properties of afferent nerve impulses originating from a neuroma. Nature 1974;248:740-3. 10. Campbell JN, Long DM. Peripheral nerve stimulation in the treatment of intractable pain. J Neurosurg 1976;45:692-9. 11. Law JD, Swett J, Kirsch WM. Retrospective analysis of 22 patients with chronic pain treated by peripheral nerve stimulation. J Neurosurg 1980;52:482-5. 12. Long DM, Erickson DL, Campbell JN, North RB. Electrical stimulation of the spinal cord and peripheral nerves of pain control: a 10-year experience. Appl Neurophysiol 1981;44:207-17. 13. Nashold BS, Goldner JL, Mullen JB, Bright DS. Longterm pain control by direct peripheral nerve stimulation. J Bone Joint Surg 1982;64A:1-10. 14. Picaza JA, Hunter SE, Cannon BW. Pain suppression by peripheral nerve stimulation. Appl Neurphysiol 1977/ 78;40:223-34. 15. Sweet WH. Control of pain by direct electrical stimulation of peripheral nerves. Clin Neurosurg 1976;23:103-11. 16. Racz GB, Lewis R Jr, Heavner JE, Scott J. Peripheral nerve stimulation implant for treatment of causalgia. In: Stanton-Hicks M, editor. Pain and sympathetic nervous system. Boston: Kluwer; 1990. p. 225-39. 17. Shetter AG, Racz GB, Lewis R Jr, Heavner JE. Peripheral nerve stimulation. In: North RB, Levi RN, editors. Neurosurgical management of chronic pain. NY: Springer; 1997. p. 261-70. 18. Hassenbusch SJ, Stanton-Hicks M, Schoppa D, Walsch JG, Covington EC. Long-term results of peripheral nerve stimulation for reflex sympathetic dystrophy. J Neurosurg 1996;84:415-23. 19. Eisenberg E, Waisbrod H, Gerbershagen HU. Long-term peripheral nerve stimulation for painful nerve injuries. Clin J Pain 2004;20:143-6. 20. Mobbs NJ, Naire S, Blum P. Peripheral nerve stimulation for the treatment of chronic pain. J Clin Neurosci 2007;14:216-21.
Peripheral nerve stimulation for neuropathic pain
21. Slavin KV. Peripheral nerve stimulation for neuropathic pain. Neurotherapeutics. 2008;5:100-6. 22. Weiner RL, Reed KL. Peripheral neurostimulation for the control of intractable occipital neuralgia. Neuromodulation 1999;2:369-75. 23. Slavin KV, Colpan ME, Munawar N, Weiss C, Nersesyan H. Trigeminal and occipital peripheral nerve stimulation for craniofacial pain: a single-institution experience with review of the literative. Neurosurg Focus 2006;21:1-5.
139
24. Weiner RL. Occipital neurostimulation for treatment of intractable headache syndromes. Acta Neurochir Supp1 2007;97:129-33. 25. Johnson MD, Burchiel KJ. Peripheral stimulation for treatment of trigeminal postherpetic neuralgia and trigeminal posttraumatic neuropathic pain. A pilot study. Neurosurgery 2004;55:135-40. 26. Trentman TL, Zimmerman RS. Occipital nerve stimulation. Technical and surgical aspects of implantation. Headache 2008;48:319-27.
2357
133 Radiofrequency Dorsal Root Entry Zone Lesions for Pain P. Konrad . F. Caputi . A. O. El-Naggar
The goal of this chapter is to summarize the relevant anatomy and physiology of pain transmission into the spinal cord through the dorsal root entry zone (DREZ), and in using this knowledge, to select appropriate patients for precise radiofrequency lesioning of the DREZ in the treatment of a variety of pain syndromes. Furthermore, details regarding operative technique from the authors experience should aid the reader in optimizing the outcomes and minimizing complications in appropriately selected patients for this procedure.
Anatomy and Physiology of the Dorsal Root Entry Zone Details regarding the anatomy and physiology of pain perception and transmission in the nervous system are discussed elsewhere in this text. The following section is a summary of the pertinent concepts when considering whether a lesion of the DREZ would be appropriate in treating a painful condition. In our experience, the treatment of chronic pain is best optimized when the physician first understands the etiology of the pain and then arrives at an anatomical localization of the generator for the pain. Through understanding the exact character of the pain and its location, the surgeon can then make a thoughtful judgment as to whether the pain is Electronic supplementary material Supplementary material is available in the online version of this chapter at http:// dx.doi.org/10.1007/978-3-540-69960-6_133 and is accessible for authorized users. #
Springer-Verlag Berlin/Heidelberg 2009
first of all, accessible, and secondly, where to make the lesion. Furthermore, experienced surgeons then are able to weigh the accessibility of whether DREZ lesions would have a reasonable chance of producing significant pain relief versus the potential complications that might be encountered with this procedure. Anatomy of pain pathways in the spinal cord. Several methods of classification of pain pathways in the spinal cord have been published [1–5]. One classification process is based on the physiological understanding of the modality of pain transmission, namely nociceptive versus non-nociceptive pain fibers [4,5]. This classification distinguishes the pathways mediating pain generated by noxious stimuli (nociception) or non-noxious stimuli. This differentiation allows the surgeon to (1) differentiate whether transmission of painful sensation occurs primarily in the dorsal horn or whether alternate pain-enhancing pathways (for example mechanical or temperature sensitivity) are also involved, (2) whether appropriate medical therapy or other interventional therapy has been tried, and (3) what the potential prognosis for lesioning the DREZ will be for a given patient. > Figure 133-1 illustrates the nociceptive and other pain pathways as they enter the dorsal root entry zone on cross section of the spinal cord. This anatomy holds true for cervical, thoracic and lumbar segments, but with different proportions [3,7–11]. Several anatomical features are worth noting from a surgical standpoint: (1) as fibers enter the dorsal horn, large fibers of proprioception are located medially, large myotactic fibers are located in the middle of the dorsal root, and smaller (C) fibers
2252
133
Radiofrequency dorsal root entry zone lesions for pain
. Figure 133-1 Cross section diagram of the spinal cord, illustrating the relevant histological components related to the DREZ. (a) Nociceptive information is carried in C-type fibers along with other pain-related phenomenon such as thermal and cutaneous mechanoreceptive information, which typically enter the dorsal root entry zone in the lateral aspect of the rootlet. First order neurons in the dorsal ganglion mediating nociceptive information synapse ipsilaterally with second order neurons in Rexed lamina I–V. These second order neurons project rostrally in the contralateral spinothalamic tract (STT) after crossing in the anterior commissure of the spinal cord. Large A-type fibers which carry proprioceptive information from Golgi tendon organs or intrafuscal stretch receptors are located in the medial aspect of the root entry zone and project up the cord ipsilaterally in the dorsal columns (DC) to their respective cervical nuclei (n. cuneatus and n. gracilus) (revised from [6]). (b) Nociceptive information associated with first order neurons located in the dorsal root ganglion project to Rexed lamina I–V (substantia gelatenosa, Sgo) either in the same segment of entry or adjacent spinal cord segments in Lissauers tract, located immediately lateral to lamina I–II (revised from [7])
associated with nociception, autonomic function and light touch are located on the lateral edge of the entry zone, (2) Lissauer’s tract is located immediately lateral to the DREZ and is responsible for longitudinal transmission of nociceptive information at least two segments superior and inferior to the point of entry into the cord, (3) the corticospinal tract, responsible for voluntary control motor function, is located in the white matter immediately lateral to the DREZ, and (4) the majority of nociceptive and other
small diameter fibers synapse within Rexed lamina I–V of the dorsal horn of the spinal cord. In addition to understanding the segmental innervation of the dorsal root, it is also important to frame the distribution of nociceptive information along the longitudinal axis of the spinal cord. Lissauer’s tract is a key pathway that conducts nociceptive information at least two segments above and below the DREZ (> Figure 133-2). Through this tract, the first order neurons synapse with multiple segments of the spinal cord
Radiofrequency dorsal root entry zone lesions for pain
133
. Figure 133-2 Longitudinal diagram of the spinal cord, illustrating the relevant tract anatomy and relationship of the gray matter to tract location and blood supply. Angle of approach to DREZ lesions is slightly greater in the cervical (35 degrees [10]) than thoracic or lumbar-sacral segments. Also, note the proximity and somatotopic arrangement of the corticospinal tract and dorsal columns to the DREZ. Lissauer’s tract is thought to mediate nociception into adjacent spinal segments as much as two segments from the point of entry into the DREZ (> Figure 133-1; Revised from [13])
and distribute nociceptive information to nearby somatic zones which are involved in reflexive behavior [12]. However, it blurs the margin of nociceptive information, leading to a less distinct border of a painful zone described by the patient [14–17]. It is important, thus, for surgeons who contemplate the extent of DREZ lesioning to understand that up to two segments above and below a specific dermatomal segment may be involved in nociceptive transmission. Another important anatomical concept to keep in mind is the relationship of the dorsal horn to adjacent white matter tracts (funiculi), and that these dimensions change throughout the spinal cord [10,18–20]. > Figure 133-2 illustrates this relationship for the cervical, thoracic, and lumbar-sacral regions of the spinal cord. The following are important concepts: (1) the angle of the dorsal horn with respect to the sagittal axis is greater in the cervical enlargement (35 degrees [10]) of the cord than in the lumbarsacral enlargement (approximately 20 degrees), (2) lateral to the dorsal horn is the corticospinal tract organized in a somatotopic fashion, with
cervical, thoracic, and lumbar fibers arranged in a medial to lateral direction, and (3) the dorsal columns that are medial to the dorsal horn are organized with converse somatotopy, namely cervical proprioceptive fibers are lateral to lower extremity fibers. These relationships mean the following to the surgeon contemplating a lesion of the DREZ. If dorsal root fibers are avulsed, as commonly seen in pain associated with brachial or lumbar-sacral plexus trauma, then deviating laterally from the DREZ may result in ipsilateral hemiparesis below the level of the lesion [21–24]. Alternatively, if lesions deviate too medially from the DREZ, loss of proprioception may occur. These risks should be included in the consideration of this procedure [25].
Pathophysiology of Pain Related to the Dorsal Root Entry Zone Although pain is a subjective complaint, there are common symptoms and signs associated with complaints of pain in patients who respond
2253
2254
133
Radiofrequency dorsal root entry zone lesions for pain
favorably to DREZ lesioning. These symptoms and signs are directly related to the pathology of deafferentation of the first order neurons mediating not only nociceptive (tissue-destructive) information but thermal, cutaneous mechanical, vasomotor, chemical, and hair follicle sensory neurons [2,4,5,26,27]. These neurons all share several common histological features that perhaps explains the overlap in symptoms when injury occurs. First, the symptoms of pain associated with light touch (hyperesthesia with allodynia), mild changes in temperature, movement of hair, and the associated flushing that may occur are all mediated by either Ad or C-type fibers [1,16]. Ad fibers are thinly myelinated, and multiple axons share a common Schwann cell for myelination. C fibers are mostly non-myelinated and represent perhaps the most ‘‘fragile’’ of axonal structures. These small fibers conduct at a slow speed (Ad fibers = 12–30 m/s; C fibers = 0.5–2.0 m/s), and as a result, account for the delay in pain perception upon injury. Typical complaints of hypersensitivity in zones adjacent to complete disruption of these fibers is most likely related to the distribution of homologous information through Lissauer’s tract [28–31]. Patients with partial or complete deafferentation of cutaneous sensory information will typically report sensitivity to nonnoxious stimuli (allodynia) in regions adjacent to but clearly within the zone of spinal or peripheral nerve injury. Due to the persistence of multiple sensory modalities surrounding an injury zone, it is thought that this may contribute to persistent pathological activity in nociceptive fibers that the patient reports as ‘‘painful’’ despite its benign nature [29,30]. For example, light touch in the skin surrounding an anesthetic zone associated with avulsed root segments may be perceived as painful (nociceptive) and will be a driving complaint of the patient. Causalgia may be perceived with little or no stimulation and is thought to be pathologically mediated via the
sympathetic chain [15,16]. Hence, successful DREZ lesioning may not address alternate autonomic pathways that mediate pain outside the spinal cord or bypass the region of the DREZ lesion. It is important, then, for the surgeon to recognize the extent to which this pathological circuitry may or may not be addressed by a DREZ lesion. And alternative therapies, such as intrathecal medications or neurostimulation therapy (see other Chapters) may also need to be considered when the patient’s symptoms suggest that the DREZ procedure may not address the complete syndrome of pain experienced by the patient. Physiological observations in deafferented spinal cords. Whether induced experimentally [11,16,32] or incidentally recorded in humans undergoing surgery [33,34], increased and abnormal firing patterns are seen in neurons residing in the dorsal root entry zone when deafferentation occurs. This abnormal firing is seen throughout neurons located in dorsal horn. In animals that underwent deafferentation, increased firing is seen in dorsal horn sensory neurons within hours of the injury and has been documented for months after the injury. This correlates with patients who complain of onset of pain within an anesthetic zone (anesthesia dolorosa) or burning sensation (causalgia) in the anesthetic or adjacent zone of injury. Extracellular recordings acquired from patients just prior to DREZ lesioning reveal similar hyperactivity in the dorsal horn. Similar to denervation of muscle fibers, these data suggest that the deafferented second order neurons of the dorsal horn have increased receptors, and display heightened sensitivity to normal neurotransmission from nearby neurons or a pathological increase in neurotransmitter analogs. These changes occur over days to months and involve complex receptor pharmacology mediated through neuropeptides and polysynaptic changes [35]. Known neurotransmitters involved in normal and pathological function in the dorsal horn are summarized in > Figure 133-3. Response to opioids and opioid antagonists can provide in-
Radiofrequency dorsal root entry zone lesions for pain
133
. Figure 133-3 Localization of opioid receptors in the substantia gelatinosa of the spinal cord (lamina II and III). mu receptors (o), delta receptors (small delta symbal), and k receptors (triangle). Modified from Burchiel (Burchiel K. Surgical Management of Pain. Thieme Medical Publishers, Inc., New York 2002; p. 582)
sight into the pathways mediating complaints of pain by the patient. Most patients receive benefit from opioid agonist therapy, which correlates with reduction in hyperactivity in the substantia gelatinosa of the dorsal horn [36–39]. However, opioid receptors undergo adaptation in the form of down-regulation, which in turn diminishes the long term success of this therapy. Recently, pharmacotherapy oriented towards providing gamma-amino butyric acid (GABA) agonists [40], NMDA antagonists, or diazepam agonists also provide some relief [41–44]. However, despite multiple medication regimens, patients typically consult surgical options because lasting relief from pain associated with peripheral or spinal trauma is not achieved. The neurosurgeon who contemplates DREZ lesioning must understand that these patients have altered receptor function in the associated spinal segments of the injury and may have return of painful symptoms even years after successful surgical DREZ lesioning. Yet, DREZ lesioning can result in successful reduction in medications used for chronic pain management.
What has not been explained is the mechanism of how patients experience return of similar symptoms and signs after years of successful relief after DREZ lesioning. Several possibilities exist. One is the regrowth of neurons into deafferented zones, resulting in expanded regions of hyperesthesia and allodynia [33,45–47]. Another reason might be that reorganization of existing circuits occurs, resulting in a pathological form of neural plasticity [48,49]. A third possibility may be that secondary structural changes occur in the region of the painful segment, such as syrinx formation [50,51], adhesions of the rootlets [52,53] or compressive lesions associated with spinal instability [54,55], that present years after the initial treatment. These possibilities need to be ruled out before contemplating further lesions or other surgical interventions for treatment of the pain.
Selection and Workup of Patients Traumatic injury of the cervical plexus. The classic history given by the patient is usually the result
2255
2256
133
Radiofrequency dorsal root entry zone lesions for pain
of a high speed motorcycle or motor vehicle accident. The patient typically has immediate partial or complete loss of sensation in the arm, as well as significant loss of motor function. Most patients describe a partial return of sensory and motor function within six months of the onset of the injury. However, within this time period, the accompanying neuropathic pain arises in the affected limb. The mechanism of the avulsion is thought to occur from the sudden depression of the shoulder associated with a sudden medial tilting of the head and neck causing the avulsion of the nerve roots. Wynn Parry [56] observed that 90% of these patients will develop deafferentation pain early in the injury, but it may also be delayed for 3–4 months. At least a third of these patients continue to experience pain after 3 years [57]. Complete cervical avulsion patients usually describe anesthesia up to the level of the clavicle. However patients with partial avulsions may have patchy areas of preserved sensations. The patient will typically complain of severe burning pain (causalgia) which is worst in the lateral aspect of the forearm and hand. If the T1 or T2 roots are involved the patient usually presents with a Horner’s syndrome on the side of the avulsion. A cervical MRI [58–60] or a cervical myelogram [61–63] will typically reveal the presence of pseudomeningoceles ipsilateral to the side of the trauma. The sizes and levels of the pseudomeningoceles however, are not necessarily proportional to the extent and levels of avulsions. EMG studies may be helpful in determining the presence of muscle deafferentation, but should not be used to deny the presence of isolated sensory denervation of the limb. Use of somatosensory evoked potentials may show delayed or absent conduction of signal through the cord [64,65]. The pain of brachial plexus avulsion usually does not respond to conventional pain medications including narcotic medications and anticonvulsants. Stellate ganglion blocks, acupuncture, transcutaneous electric stimulation, and physical therapy could be effective in some cases, however the response
to the DREZ operation is much more significant and more complete [21,64,65]. Patients who fail narcotic and other medical management that have clear clinical evidence of brachial plexus avulsion are excellent candidates for the DREZ operation. Traumatic injury of the lumbar plexus and cauda equina. Similar to brachial plexus injuries, lumbar plexus and spine injuries can result in root avulsion from the spinal cord and deafferentation of the spinal cord from first order neuron damage. The history usually involves high speed motor vehicle accidents, in particular motorcycle accidents. The patient usually has severe associated injuries such as pelvic fractures, and possibly traumatic amputation of the lower extremities. Like the cervical injury, the patient loses sensation and motor function in one or both legs at the time of the accident. Furthermore, bowel and bladder function are usually diminished or absent. Burning pain associated with dense sensory loss is common, and depending on the extent of dorsal root avulsion, there may be hyperpathia and allodynia bordering the anesthetic region. Some estimate that 25–50% of patients with this injury have severe pain that is refractory to narcotics and other medications [66–68]. Even though the avulsion occurs in the area of the conus medullaris, a lumbar myelogram or MRI will typically reveal the pseudomeningoceles in the lumbosacral junction area usually at the L5 and S1 levels. Patients who have both mixed conus medullaris lesions and cauda equina syndrome (as evidenced by deafferentation of lower extremity muscles) are usually good candidates to consider for DREZ. Neurosurgeons who are contemplating DREZ on these patients, however, must have a thorough understanding of the altered anatomy typically present with these patients (such as intrathecal adhesions and altered spinal cord anatomy) and contemplate the associated risks of obtaining adequate exposure of the lower spinal cord. Bilateral DREZ procedures for these patients are not uncommon.
Radiofrequency dorsal root entry zone lesions for pain
Post-amputation (phantom limb) pain. Phantom pain may result after traumatic amputations of the upper or lower extremities. It may also result after surgical amputations subsequent to severe peripheral vascular disease. Excellent results are achieved with the DREZ operation in patients with phantom pain due to traumatic amputations, but not in patients with stump pain only due to surgical amputations [69]. Moreover, the best results are seen in patients with phantom pain due to traumatic amputations associated with lumbosacral nerve root avulsions noted during surgery [70]. It is not clear why this difference exists, but the distinction clinically can be discerned by noting the presence of hypalgesia in the affected limb and the description of neuropathic pain in the perceived phantom limb. Saris et al. [69] noted that 67% of patients had good or excellent pain relief following DREZ lesioning of the lower spinal cord. Patients who present with phantom pain, especially following trauma, should first be given a reasonable trial of medication and consideration for less risky surgical options such as a trial of spinal stimulation. However, should these options fail; a DREZ lesion should be considered. Traumatic injury of the spinal cord. Anywhere between 25 and 50% of paraplegic patients due to spinal cord injuries develop intractable pain [14,53,66,71]. Such injuries may result from motor vehicle accidents, gunshot injuries and sport injuries. Traumatic injuries usually create several types of painful phenomenon, and the history is essential in determining the relationship of the complaint to the likely cause of the pain. The most common type of pain associated with spinal cord injuries is a segmental or a girdle- like pain, which develops just above the level of sensory loss [14,72]. This so called ‘‘end-zone’’ pain is a transition zone between the dermatomes with normal sensations and the dermatomes with total analgesia. This zone also harbors trigger spots (hyperpathia) which when touched can reproduce the severe segmental pain as well as pain radiating into the paralyzed extremities. Such pain
133
can also be triggered by a full bladder or bowel. The onset of pain may be delayed for weeks, months or even years after the injury. Quadriplegic patients may present with a radicular kind of pain corresponding to the level of the cervical cord injury, which may be a variant of the ‘‘endzone’’ pain associated with paraplegics. When the patient provides the above history, this ‘‘end-zone pain’’ responds very well to the DREZ operation. Another type of pain associated with spinal cord injury is a rather diffuse, burning kind of pain arising below the level of the injury which may involve the whole body below that level or maybe localized to the sacral area or to certain visceral structures such as the rectum or bowels. Friedman and Nashold reported less than 20% success in treatment of this type of pain with the DREZ operation [14,73]. It is incumbent upon the neurosurgeon contemplating DREZ lesioning in this patient population to determine if there are other extrinsic factors that contribute to the pain. Patients who describe pain that is worse with mechanical movement, slowly progressive sensory or motor deficits, ‘‘electric, shock-like’’ sensation with movement, or Lhermitte’s phenomenon have non-characteristic neuropathic pain complaints. A thorough workup prior to offering a DREZ lesion would include determining the presence of mechanical instability, syringomyelia, compressive lesions (such as a herniated disc, hypertrophied ligamentum flavum, or arachnoid cyst), or arachnoiditis. Even if the history suggests central neuropathic pain (burning, deep, steady pain at or below the level of the injury), an MRI of the affected region of the spine routinely obtained prior to surgery may prepare the surgeon for distorted anatomy that may be encountered with the planned exposure. If the spinal cord injury is associated with a significant syrinx as seen on the preoperative MRI or a post myelogram CT scan, the latter should be dealt with during surgery. Depending on the size of the syrinx, a syringo-peritoneal or syringo-
2257
2258
133
Radiofrequency dorsal root entry zone lesions for pain
subarachnoid shunt placement should also be considered prior to lesioning the spinal cord [74,75]. Simple drainage of the syrinx without performing DREZ lesions, however, may not help the patient’s complaints related to deafferentation pain [76]. Patients with spinal cord trauma due to gunshot injuries usually have the spinal cord injured at several segments. Adequate exploration of all segments involved as well as at least three segments above the level of the injury should be performed [14,20]. It is not unusual in those patients to have several segments of spinal cord damage and nerve root damage as well as nerve root avulsions for several segments above the level of the injury [14,20]. Failure to include such segments with the DREZ lesions will cause incomplete relief of pain. Cancer pain. Neoplasms that may produce deafferentation pain usually result from local compression or infiltration of either the brachial or lumbar-sacral plexus. Unilateral upper extremity pain associated with Pancoast tumors have had good to excellent pain reduction reported in the literature [77,78]. Conceivably, a homologous pain syndrome associated with retroperitoneal or pelvic tumors invading or displacing the lumbar-sacral plexus may also benefit from DREZ lesioning of the lower spinal cord, although no cases have been specifically described in the literature. Patients who have significant sensory loss with hyperpathia and allodynia may be eligible for DREZ lesioning. The risks of DREZ lesioning should be weighed against those for cordotomy and intrathecal pharmacotherapy. A DREZ lesioning may be an appropriate choice in patients with very focal, unilateral upper extremity pain associated with a treated lesion in the thoracic apex that have a relatively long life expectancy and lead a relatively independent life [79,80]. In such patients, the low risk of ipsilateral hemiparesis and minimal sensory loss outside the affected limb may be less risky than a high cervical cordotomy or intrathecal medications delivered to the upper cervical spinal cord region.
Post-herpetic pain. Fortunately, only about 10% of patient affected by the varicella zoster virus present post herpetic neuralgia [28,73]. The incidence of chronic, refractory pain increases with age, and 50% of patients who present for surgical options are typically octogenarians [81]. The complaint of post-herpetic neuralgia from spinal nerve root involvement localizes mostly in a thoracic dermatome (85%). This pain originates from an intense myelo-radiculitis that may cause significant deafferentation of the second order neurons of the spinothalamic tract. The residual pain is characterized by hypalgesia, hyperpathia and allodynia in the affected dermatomes. In this sense it is similar to traumatic deafferentation pain. However, patients with post-herpetic neuralgia have a discrete dermatomal manifestation (usually associated with visible scars resulting from previous vesicular eruption) and no motor or myelopathic changes. Candidates for a DREZ lesion should have failed conservative therapy, such as topical anesthetics (e.g., lidocaine) and oral agents (e.g., GABA agonist, tricyclic antidepressant, and/or antiepileptic drugs). DREZ lesioning has better success for pain that is characterized as ‘‘shooting’’ and paroxysmal with allodynia over the affected dermatome [22,82]. There may also be an associated deep cramping or itching sensation. In counseling patients about DREZ lesioning for this condition, long term remission of pain is not common. Friedman et al. [73] reports that good to excellent pain reduction occurred in 60% of patients in the in the postoperative period, but then slowly declined to 18% over an average of 6 years follow up. Another consideration in counseling patients regarding this procedure is the immediate post-operative risk of ipsilateral leg weakness and hyperesthesia seen with thoracic DREZ lesions in 52% of patients, which persisted in 25% of the patients. This number may be lower today when using localizing intraoperative motor and sensory evoked potentials [83]. Unlike other conditions treated with DREZ lesions noted above, post her-
Radiofrequency dorsal root entry zone lesions for pain
petic neuralgia often recurs and surgery should be considered carefully.
133
. Figure 133-4 Relative location of spinal cord segments versus vertebral segments, useful in planning appropriate exposure for DREZ lesioning
Operative Technique Preoperative measures. Preoperative MRI or myelography and post-myelography CT scan can provide essential anatomical details regarding pseudomeningoceles, adhesions of the spinal cord, distorted spinal cord anatomy, or spinal deformities [3,59,84,85]. The value of such studies for each group of patients is highlighted in the previous sections. High doses of opioid agonists need to be reduced if at all possible in patients several weeks prior to surgery to reduce postoperative medical complications and allow for easier pain management post-operatively. Typically patients are given antibiotics (cefazolin 1 g or vancomycin 1 g) one-half hour prior to surgery. Steroids are also commonly given at the beginning of the procedure to reduce post-operative spinal cord edema (e.g., dexamethasone 6–10 mg IV every 6 h tapered over a few days). Exposure of appropriate segments. Once a DREZ lesion is planned, the first question that should be addressed is the extent to which the lesion must be carried out for success versus the risks associated with overextending the lesion. > Figure 133-4 illustrates the relationship of spinal nerve roots with vertebral segments. Notice that for the typical brachial plexus DREZ lesion, an exposure of the dorsal spinal cord from C4-T1 will allow a view of the related dorsal roots (C5-T1) [3,10,86]. For thoracic DREZ lesions for herpetic neuropathy, exposure of two vertebral segments above and below the dermatome is suggested. For DREZ lesioning of the lower spinal cord related to spinal cord injury or cauda equina syndrome, it is advisable to expose at least two vertebral segments above the superior aspect of the painful zone. Typically, a bilateral laminectomy from T10-L2 is performed for treatment of pain related to conus medullaris and cauda equina lesions.
A hemilaminectomy that is performed for unilateral DREZ lesions may reduce the likelihood of post-laminectomy kyphosis. This approach is appropriate for most cases of non-spinal trauma (such as brachial plexus avulsion or herpetic neuropathy) in which no previous laminectomy has been performed. Yet it is much wiser to perform a standard laminectomy when intradural adhesions, syringomyelia, or distorted spinal cord anatomy are seen on pre-operative imaging. For a unilateral DREZ lesion, one can perform a hemilaminect-
2259
2260
133
Radiofrequency dorsal root entry zone lesions for pain
omy with a high speed drill exposing the dorsolateral spinal canal. For a bilateral DREZ lesion, a more standard laminectomy can be performed. The end result of the exposure needs to ensure that the appropriate spinal cord segments are exposed with an adequate opening to allow the dura to be reflected and the lesioning electrode placed at the appropriate angle (20–30 degrees lateral from the midsagittal plane) for approach into the dorsal horn. Although the spinal cord segments align with vertebral segments in the cervical spine, inferiorly in the spine this relationship becomes discordant, and surgeons must make adjustments as seen in > Figure 133-4. If significant spinal cord injury has occurred, it is advisable to perform a bilateral laminectomy to ensure a view of both normal and abnormal anatomy. Use of the operating microscope with moderate to high powered magnification is essential. Orientation of the midline, dorsal columns, vascular structures, dorsal rootlets, dentate ligament all provide the neurosurgeon with
appropriate entry landmarks and estimates of the location of the dorsal horn and corticospinal tract [10]. If the dura and arachnoid are opened separately, blood from decompression of the dura can be removed and controlled before it enters the subarachnoid space, reducing the potential for post-operative adhesions. Once the arachnoid is opened, an assistant who can continue to aspirate blood and cerebrospinal fluid from the field is very helpful because it allows the neurosurgeon to maintain attention on the DREZ lesioning line (see Video). Once lesioning starts, staying focused on the orientation and direction of the DREZ minimizes deviation from the intended zone for lesioning, and the location of the previous lesion is not lost. Lesion location and technique. > Figure 133-5 and Video shows the typical Nashold DREZ lesioning electrode and dimensions. Piercing the pia usually requires a sharp push, so as to minimize the deformation of the cord and mini-
. Figure 133-5 The Nashold DREZ lesioning electrode for spinal lesioning procedures (Model NTCD, Cosman Medical, Burlington, MA, USA). Note that the exposed tip measures 0.25 2 mm, thereby matching the appropriate depth necessary for most spinal cord applications. The electrode also contains a thermocouple which allows for temperature measurement and feedback control for precise thermal lesioning when used with the Cosman lesion generator (RFG-1A, Cosman Medical, Burlington, MA, USA) (reproduced from Cosman Medical, Inc)
Radiofrequency dorsal root entry zone lesions for pain
mize unwanted injury. The ideal location for the lesion should be at the lateral edge of the spinal rootlet as it enters the cord, where the nociceptive fibers are gathered (see discussion of Anatomy of the Pain Pathways in the Spinal Cord). The electrode should be inserted to the full depth of the exposed tip (2 mm), and in so doing, impedance measurements can be made to identify zones of injury [87]. Nashold and colleagues described low impedance values (around 500– 1,000 ohms) associated with areas of injury, versus 1,200–2,000 ohms for normal grey and white matter of the spinal cord respectively. This may be useful in delineating the DREZ area and avoiding deviating into adjacent spinal tracts. Somatosensory evoked potentials (SSEPs) [83,88,89] and motor evoked potentials (MEPs) [90–92] may also aid in the identification of these adjacent tracts when there is significant anatomical distortion from previous injury. For cervical regions the angle of the electrode is approximately 30 degrees from midline whereas in the lower thoracic region it is approximately 20 degrees. The weight of the lesioning electrode is usually adequate to hold the electrode still without inadvertent dislodgement or movement during the brief lesion.
133
Lesions are made with the radiofrequency generator set to 75 C for 15–20 s [86,93]. This usually results in a 1 2 mm lesion [46,94]. The lesion is repeated down the length of the DREZ spaced about 1–1.5 mm apart, or essentially the width of the insulated end of the electrode. A typical unilateral DREZ procedure may result in a total of 40–60 lesions spanning four or more spinal cord segments [21]. Thus, an efficient DREZ lesioning technique requires a coordinated effort among the neurosurgeon, the surgical assistant, and the individual running the lesion generator. Once a lesion is created, a small tan discolored area is left or a small puncture is seen where the needle penetrated the cord (> Figure 133-6). It is important to continually re-evaluate that DREZ lesions are following the dorsal lateral sulcus if no dorsal rootlets are seen (as illustrated in Video). If avulsed rootlets are seen throughout the intended lesioning zone, the neurosurgeon may find it useful to begin rostral or caudal to the avulsed segments in the region containing spinal rootlets to identify the dorsolateral sulcus and progress into the avulsed region. Finally, it is very unusual to see significant arteries cross the dorsolateral sulcus, since it is a watershed zone
. Figure 133-6 DREZ lesions along the spinal cord in the region of T12-L2 for treatment of end-zone pain related to spinal cord injury. Note the avulsed segment of dorsal roots inferior to the T12 dorsal root. Several DREZ lesions can be seen as small dark spots along the dorsal root line. Note also relationship of the DREZ lesions to the dorsal columns, lateral spinal cord and dentate ligament
2261
2262
133
Radiofrequency dorsal root entry zone lesions for pain
between the dorsal spinal arteries and the anterior spinal artery [7,9]. However, a prudent neurosurgeon should always avoid injuring any significant arterial supply to the spinal cord. The DREZ lesions are completed when either the lesions encompass one or two spinal cord segments above the painful zone or the impedances of the cord have normalized. Closure of the arachnoid and dura can be accomplished in one layer with a fine suture (4-0 or 5-0). The use of dural sealants such as thrombogenic derivatives (e.g., Tisseel1, Baxter, Deerfield, IL, USA) or synthetic derivatives (e.g., DuraSeal1, Confluent Surgical, Waltham, MA, USA) may reduce the incidence of cerebrospinal fluid leak. Fascia and cutaneous closure are performed in routine fashion. Post-operative care. Patients are typically observed closely in the intensive care unit for the first 24 h for any new neurological deficits, such as mild weakness or diminished proprioception ipsilateral and inferior to the lesion. If new postoperative deficits emerge, urgent imaging with MRI or CT myelography is appropriate to rule out hematoma or other anatomical complications. Most patients however, can be mobilized by the first post-operative day. To reduce likelihood for CSF leak, patients who underwent cervical or upper thoracic DREZ lesions are instructed to keep the head of the bed elevated at least 30 degrees, whereas those who underwent thoracolumbar exposures are kept flat for 24 h. Most patients are encouraged to get out of bed by the first post-operative day and encouraged to mobilize. Dexamethasone in tapering amounts is usually administered over the course of several days, but may be prolonged if new deficits are encountered that may result from edema secondary to the lesion. Patients are mobilized typically on the first post-operative day. Pain management is converted from IV to oral medications in anticipation of discharge within 3 days post-operatively. Most patients who benefit from DREZ lesioning will notice significant pain reduction by the
second post-operative day, and may require very little pain medication at the time of discharge. Successful outcome from DREZ lesioning is the best encouragement to consider this therapy for others. The neurosurgeon is encouraged therefore, to follow the outcomes of such patients through intermittent visits over the course of his/her practice to improve on patient selection and serve as a reminder of the benefits and limits of this therapy.
Outcomes Complications. Several reviews have been published regarding complications and outcomes of the DREZ procedure for RFL [25,95], laser [96] and microcoagulation [97]. > Table 133-1 summarizes the potential complications associated with radiofrequency DREZ lesioning of the spinal cord. In general, the most serious complications are usually associated with lesions that inadvertently are placed too far laterally and injure the corticospinal tract resulting in permanent ipsilateral weakness below the lesion. This occurred in 3–14% of patients reported since 1990. It is most frequent with thoracic DREZ lesions where the dorsal horn is the thinnest and the margin for error is the least. Permanent sensory loss, namely ipsilateral loss of proprioception and light touch below the DREZ site, is tolerated better by patients, but is reported at a higher rate (2–70% in reports since 1990). It appears that the smaller electrode [105] and experience with the technique has resulted in a lower complication rate. Other complications such as bowel and bladder dysfunction, CSF leak, infection, and hematoma formation are more rare. These complications are more typical of surgery related to exposure of the spinal cord and closure of the wound. In addition, post-laminectomy kyphosis is more likely to occur in patients with multiplelevel laminectomies that extend laterally into the
Radiofrequency dorsal root entry zone lesions for pain
133
. Table 133-1 Complications that potentially can occur with radiofrequency DREZ lesioning of the spinal cord reported in studies with ten or more patients No. of patients
Permanent sensory or motor loss
Author
Year
Reason for DREZ
Samii [23]
1984
BPI, SCI
35
0
Richter [98] Thomas [99] Garcia-March [100] Friedman [73]
1984 1984 1987 1988
10 34 11 39 56 32 10
30% S, 10% both 12% M/S
Campbell [101]
1988
BPI, SCI BPI BPI BPI SCI PHN BPI
Ishijima [102] Saris [103] Saris [104]
1988 1988 1988
BPI, SCI, PHN Post-amputation Peripheral
30 22 12
Young [105]
1990
21
17 102 29 10 73 47 41 21 60
Kumagai [106] Edgar [107] Sampson [20]
1992 1993 1995
Various – lg electrode Various – sm electrode Various SCI Conus
Rath [22] Samii [108] Falci [109] Tomas [89] Chen [110]
1997 2001 2002 2005 2006
Cauda equina Various BPI SCI BPI BPI
37
60% M/S 5% M, 5% minor 69% M 20% hyperreflexia 62% S 41% mild M/S >50% S, 8% M
Transient motor or sensory loss 3% M, 25% M, 23% both 10% M 50% M/S 9% M
0 20
16 10
17% M, 8% sphincter, 74% mild dysmetria
24% S, 19%M, 5% both 3% S, 5% M 71% S, 41% M 2% S, 3% M 3% S, 14% M, 10% sphincter
Other, %
14 1 7
1
3% M
35 5 7
10% M 4% M 70% S, 14% M 14% S/M
10% M
2 9
25% S
Note: BPI, brachial plexus injury; M, motor; PHN, post-herpetic neuralgia; S, sensory; SCI, spinal cord injury
facet joint or pars interarticularis and in patients with significant pre-existing spondylosis [111]. Prognosis of pain relief. > Table 133-2 summarizes a review of the literature reporting outcomes in ten or more patients undergoing radiofrequency DREZ lesioning for various neuropathic conditions other than post-amputation pain. Very few reports on the outcome for postamputation pain exist and these results from small studies of radiofrequency DREZ lesioning are also included. Overall, the success of radiofrequency DREZ lesioning has improved over the years. It con-
tinues to remain true that the best results are obtained with patients who have brachial plexus avulsion. Patients can expect good to excellent reduction in brachial plexus avulsion pain 54–91% of the time, and it appears to last in at least 50% of patients over 5 years. Patients who had end-zone pain rather than diffuse distal pain related to spinal cord injury (SCI) had better outcomes (78% vs. 20%). Although follow-up studies for SCI pain are not nearly as long, the results also appear to hold for more than 3 years [25]. Both Tomas et al. [89] and Falci et al. [109] have indicated that intraoperative electrophysi-
2263
2264
133
Radiofrequency dorsal root entry zone lesions for pain
. Table 133-2 Outcomes of pain relief in patients who underwent radiofrequency DREZ lesioning for the treatment of various painful conditions. Good results are considered when pain relief is reported as either good or excellent Author
Year
No. of patients
Good results
Follow-up period (months)
34 22 11 10 17 18 56 47 21 40
21 (62%) 20 (91%) 6 (54%) 8 (80%) 14 (82%) 13 (72%) 33 (59%) 30 (64%) 13 (62%) 32 (80%)
4–44
1990 1993 1995 2002
20 31 25 14 46 39 41
10 (50%) 23 (78%) – end-zone pain 5 (20%) – diffuse pain 8 (57%) 42 (92%) 21 (54%) 33 (80%)
1988 1990
32 11
8 (25%) 6 (54%)
6–72
1985 1988
7 9 7
1 (14%) 6 (67%) 2 (29%)
1–28 6–60 6–60
Brachial plexus avulsion pain Thomas [99] 1984 Samii [23] 1984 Garcia-March [100] 1987 Campbell [101] 1988 Ishijima [102] 1988 Young [105] 1990 Freidman [112] 1990 Samii [108] 2001 Tomas [89] 2005 Chen [110] 2006 SCI pain Weigand [113] 1985 Freidman [14] 1986 Young [105] Edgar [107] Sampson [20] Falci [109] Post-herpetic pain Freidman [73] Young [105] Phantom-limb pain Weigand Saris
ology of the dorsal horn during these procedures is likely to enhance the outcomes and reduce complications. Both of these groups suggest that tailoring the lesioning procedure to include areas of hyperactivity in the DREZ region will capture additional levels mediating pain not normally anticipated in the preoperative plan. Patients on the other hand who have a ‘‘dull, aching, burning pain’’ distal to the region of spinal cord injury are similar to those who complain of phantom-limb stump pain, and less optimal results are seen. These data suggest that lesioning of the DREZ will not encompass the pain pathways mediating this type of pain. In fact, autonomic pathways extrinsic to the spinal cord may mediate the refractory portions of the pain not treated by DREZ lesioning [16,25,73].
8–58 (mean, 17) 7–52 6–57 12–156 24–216 (mean 168) 4–96 36–120 1–28 6–72 6–72 2–96 (mean, 44) (mean 36) 12–72
DREZ lesioning for post-herpetic neuralgia pain is associated with poor outcomes and increased morbidity. Although initial pain relief was seen in 29 of 32 patients in the first several months, Friedman et al. [73] found that only 8 of these patients had good pain relief by a year. Considering the increased risk for motor deficits following thoracic DREZ lesioning (> Table 1331), one should be cautious in offering good results in the long run in patients with this type of pain. Laser or microsurgical DREZotomy (MDT) lesioning has been extensively described by Sindou since 1972 [78]. When others have compared the results with radiofrequency lesions, similar results were found [25] in a few reports. The advantage of radiofrequency lesioning is that the lesions
Radiofrequency dorsal root entry zone lesions for pain
are usually around 1 mm round and are highly reproducible [93]. Furthermore, insertion of the Nashold DREZ electrode will ensure a lesion depth of 2.5 mm with equal spacing around the electrode tip (> Figures 133-5 and 133-6). Stimulation can easily be performed just prior to each lesion in areas in which the anatomy is obscure. These are advantageous over an open lesioning technique of the DREZ in which the spinal cord is visually disrupted. A disadvantage of the radiofrequency technique is the lack of visualization of the actual lesion within the spinal cord. There may be skip areas when lesions are not spaced tightly [25,107], which may result in less optimal outcomes and potential for increased morbidity due to wandering from the DREZ line.
Conclusions Radiofrequency DREZ lesioning is an excellent procedure to offer patients with medically refractory pain due to a variety of syndromes. In particular, patients with pain due to brachial plexus avulsion and end-zone pain related to traumatic spinal cord injury are to be considered for this procedure. The neurosurgeon contemplating this procedure should have a solid understanding of the microanatomy in the DREZ region of the spinal cord and be familiar with contemporary intraoperative physiological testing to optimize the outcomes from this surgical procedure.
References 1. Almeida TF, Roizenblatt S, Tufik S. Afferent pain pathways: a neuroanatomical review. Brain Res 2004;1000: 40-56. 2. Coggeshall RE. Fos, nociception and the dorsal horn. Prog Neurobiol 2005;77:299-352. 3. Mertens P, Guenot M, Hermier M, et al. Radiologic anatomy of the spinal dorsal horn at the cervical level (anatomic-MRI correlations). Surg Radiol Anat 2000;22: 81-8.
133
4. Romanelli P, Esposito V. The functional anatomy of neuropathic pain. Neurosurg Clin N Am 2004;15: 257-68. 5. Willis WD Jr. Dorsal horn neurophysiology of pain. Ann N Y Acad Sci 1988;531:76-89. 6. Burchiel K. Surgical management of pain. New York: Thieme; 2002. 7. Light AR. Normal anatomy and physiology of the spinal cord dorsal horn. Appl Neurophysiol 1988;51:78-88. 8. Elliot KA. Cross-sectional diameters and areas of the human spinal cord. Anatomical Record 1945;93: 287-93. 9. McCormick PC, Stein BM. Functional anatomy of the spinal cord and related structures. Neurosurg Clin N Am 1990;1:469-89. 10. Xiang JP, Liu XL, Xu YB, et al. Microsurgical anatomy of dorsal root entry zone of brachial plexus. Microsurgery 2008;28(1):17-20. 11. Young PA. The anatomy of the spinal cord pain paths: a review. J Am Paraplegia Soc 1986;9:28-38. 12. Zhang H, Xie W, Xie Y. Spinal cord injury triggers sensitization of wide dynamic range dorsal horn neurons in segments rostral to the injury. Brain Res 2003;1055: 103-10. 13. Nashold B, Pearlstein R. The DREZ Operation, Park Ridge, IL: The American Association of Neurological Surgeons, 1996. 14. Friedman AH, Nashold BS Jr. DREZ lesions for relief of pain related to spinal cord injury. J Neurosurg 1986; 65:465-9. 15. Furue H, Katafuchi T, Yoshimura M. Sensory processing and functional reorganization of sensory transmission under pathological conditions in the spinal dorsal horn. Neurosci Res 2004;48:361-8. 16. Nashold BS Jr. Deafferentation pain in man and animals as it relates to the DREZ operation. Can J Neurol Sci 1988;15:5-9. 17. Rosenow JM, Henderson JM. Anatomy and physiology of chronic pain. Neurosurg Clin N Am 2003;14: 445-62, vii. 18. Karatas A, Caglar S, Savas A, et al. Microsurgical anatomy of the dorsal cervical rootlets and dorsal root entry zones. Acta Neurochir (Wien) 2005;147:195-9; discussion 199. 19. Romanelli P, Esposito V, Adler J. Ablative procedures for chronic pain. Neurosurg Clin N Am 2004;15:335-42. 20. Sampson JH, Cashman RE, Nashold BS Jr, et al. Dorsal root entry zone lesions for intractable pain after trauma to the conus medullaris and cauda equina. J Neurosurg 1995;82:28-34. 21. Nashold BS, Jr., Friedman A, Bullitt E. The status of dorsal root entry zone lesions in 1987. Clin Neurosurg 1989;35:422-8. 22. Rath SA, Seitz K, Soliman N, et al. DREZ coagulations for deafferentation pain related to spinal and peripheral nerve lesions: indication and results of 79 consecutive procedures. Stereotact Funct Neurosurg 1997;68: 161-7.
2265
2266
133
Radiofrequency dorsal root entry zone lesions for pain
23. Samii M, Moringlane JR. Thermocoagulation of the dorsal root entry zone for the treatment of intractable pain. Neurosurgery 1984;15:953-5. 24. Sindou M, Mertens P, Wael M. Microsurgical DREZotomy for pain due to spinal cord and/or cauda equina injuries: long-term results in a series of 44 patients. Pain 2001;92: 159-71. 25. Denkers MR, Biagi HL, Ann O’Brien M, et al. Dorsal root entry zone lesioning used to treat central neuropathic pain in patients with traumatic spinal cord injury: a systematic review. Spine 2002;27:E177-E184. 26. Blumenkopf B. Neurochemistry of the dorsal horn. Appl Neurophysiol 1988;51:89-103. 27. O’Neill OR, Burchiel KJ. Role of the sympathetic nervous system in painful nerve injury. Neurosurg Clin N Am 1991;2:127-36. 28. Baron R. Peripheral neuropathic pain: from mechanisms to symptoms. Clin J Pain 2000;16:S12-S20. 29. Cervero F. Spinal cord mechanisms of hyperalgesia and allodynia: role of peripheral input from nociceptors. Prog Brain Res 1996;113:413-22. 30. Christensen MD, Hulsebosch CE. Chronic central pain after spinal cord injury. J Neurotrauma 1997;14: 517-37. 31. Roberts WJ. A hypothesis on the physiological basis for causalgia and related pains. Pain 1986;24:297-311. 32. Suzuki R, Dickenson A. Spinal and supraspinal contributions to central sensitization in peripheral neuropathy. Neurosignals 2005;14:175-81. 33. Guenot M, Bullier J, Rospars JP, et al. Single-unit analysis of the spinal dorsal horn in patients with neuropathic pain. J Clin Neurophysiol 2003;20:143-150. 34. Jeanmonod D, Sindou M, Mauguiere F. Intraoperative electrophysiological recordings during microsurgical DREZ-tomies in man. Stereotact Funct Neurosurg 1990;54–55:80-5. 35. Blumenkopf B. Neuropharmacology of the dorsal root entry zone. Neurosurgery 1984;15:900-3. 36. Gardell LR, King T, Ossipov MH, et al. Opioid receptormediated hyperalgesia and antinociceptive tolerance induced by sustained opiate delivery. Neurosci Lett 2006;396:44-9. 37. Inturrisi CE. Clinical pharmacology of opioids for pain. Clin J Pain 2002;18:S3-S13. 38. Przewlocki R, Przewlocka B. Opioids in neuropathic pain. Curr Pharm Des 2005;11:3013-25. 39. Schug SA, Saunders D, Kurowski I, et al. Neuraxial drug administration: a review of treatment options for anaesthesia and analgesia. CNS Drugs 2006;20:917-33. 40. Blumenkopf B. Combination analgesic-antispasmodic therapy in postoperative pain. Spine 1987;12:384-7. 41. Balazy TE. Clinical management of chronic pain in spinal cord injury. Clin J Pain 1992;8:102-10. 42. Erdine S, De Andres J. Drug delivery systems. Pain Pract 2006;6:51-7. 43. Stacey BR. Management of peripheral neuropathic pain. Am J Phys Med Rehabil 2005;84:S4-S16.
44. Zagari MJ, Mazonson PD, Longton WC. Pharmacoeconomics of chronic nonmalignant pain. Pharmacoeconomics 1996;10:356-77. 45. Golding J, Shewan D, Cohen J. Maturation of the mammalian dorsal root entry zone – from entry to no entry. Trends Neurosci 1997;20:303-8. 46. Iacono RP, Aguirre ML, Nashold BS Jr. Anatomic examination of human dorsal root entry zone lesions. Appl Neurophysiol 1988;51:225-9. 47. Tessler A. Neurotrophic effects on dorsal root regeneration into the spinal cord. Prog Brain Res 2004;143:147-54. 48. Masuda T, Shiga T. Chemorepulsion and cell adhesion molecules in patterning initial trajectories of sensory axons. Neurosci Res 2005;51:337-47. 49. Ramer MS, McMahon SB, Priestley JV. Axon regeneration across the dorsal root entry zone. Prog Brain Res 2001;132:621-39. 50. Brammah TB, Jayson MI. Syringomyelia as a complication of spinal arachnoiditis. Spine 1994;19:2603-5. 51. Frisbie JH, Aguilera EJ. Chronic pain after spinal cord injury: an expedient diagnostic approach. Paraplegia 1990;28:460-5. 52. Hoffman GS. Spinal arachnoiditis. What is the clinical spectrum? I. Spine 1983;8:538-40. 53. Koulousakis A, Kuchta J, Bayarassou A, et al. Intrathecal opioids for intractable pain syndromes. Acta Neurochir Suppl 2007;97:43-8. 54. Butler JC, Whitecloud TS III. Postlaminectomy kyphosis. Causes and surgical management. Orthop Clin North Am 1992;23:505-11. 55. Wiggins GC, Shaffrey CI. Dorsal surgery for myelopathy and myeloradiculopathy. Neurosurgery 2007;60: S71-S81. 56. Parry CB. Pain in avulsion lesions of the brachial plexus. Pain 1980;41–53. 57. Parry CB. Pain in avulsion of the brachial plexus. Neurosurgery 1984;15:960-5. 58. Nakamura T, Yabe Y, Horiuchi Y, et al. Magnetic resonance myelography in brachial plexus injury. J Bone Joint Surg Br 1997;79:764-9. 59. Ochi M, Ikuta Y, Watanabe M, et al. The diagnostic value of MRI in traumatic brachial plexus injury. J Hand Surg [Br] 1994;19:55-9. 60. Yoshikawa T, Hayashi N, Yamamoto S, et al. Brachial plexus injury: clinical manifestations, conventional imaging findings, and the latest imaging techniques. Radiographics 2006;26 Suppl 1:S133-S143. 61. Carvalho GA, Nikkhah G, Matthies C, et al. Diagnosis of root avulsions in traumatic brachial plexus injuries: value of computerized tomography myelography and magnetic resonance imaging. J Neurosurg 1997;86:69-76. 62. Murphey F, Kirklin J. Myelographic demonstration of avulsing injuries of the nerve roots of the brachial plexus–a method of determining the point of injury and the possibility of repair. Clin Neurosurg 1973;20:18-28.
Radiofrequency dorsal root entry zone lesions for pain
63. Walker AT, Chaloupka JC, de Lotbiniere AC, et al. Detection of nerve rootlet avulsion on CT myelography in patients with birth palsy and brachial plexus injury after trauma. AJR Am J Roentgenol 1996;167:1283-7. 64. Synek VM. Somatosensory evoked potentials from musculocutaneous nerve in the diagnosis of brachial plexus injuries. J Neurol Sci 1983;61:443-52. 65. Synek VM, Trubuhovich RV. Important abnormalities in recordings of somatosensory evoked potentials in coma. Clin Electroencephalogr 1991;22:118-26. 66. Beric A. Post-spinal cord injury pain states. Pain 1997;72:295-8. 67. Harrop JS, Hunt GE Jr, Vaccaro AR. Conus medullaris and cauda equina syndrome as a result of traumatic injuries: management principles. Neurosurg Focus 2004; 16:e4. 68. Nashold BS Jr, Bullitt E. Dorsal root entry zone lesions to control central pain in paraplegics. J Neurosurg 1981; 55:414-19. 69. Saris SC, Iacono RP, Nashold BS Jr. Dorsal root entry zone lesions for post-amputation pain. J Neurosurg 1985; 62:72-6. 70. Iacono RP, Linford J, Sandyk R. Pain management after lower extremity amputation. Neurosurgery 1987;20: 496-500. 71. Meyerson BA. Neurosurgical approaches to pain treatment. Acta Anaesthesiol Scand 2001;45:1108-13. 72. Rawlings C III, Rossitch E Jr, Nashold BS Jr. The history of neurosurgical procedures for the relief of pain. Surg Neurol 1992;38:454-63. 73. Friedman AH, Bullitt E. Dorsal root entry zone lesions in the treatment of pain following brachial plexus avulsion, spinal cord injury and herpes zoster. Appl Neurophysiol 1988;51:164-9. 74. Lee TT, Alameda GJ, Camilo E, et al. Surgical treatment of post-traumatic myelopathy associated with syringomyelia. Spine 2001;26:S119-S127. 75. Schaller B, Mindermann T, Gratzl O. Treatment of syringomyelia after posttraumatic paraparesis or tetraparesis. J Spinal Disord 1999;12:485-8. 76. Nashold BS Jr, Vieira J, el-Naggar AO. Pain and spinal cysts in paraplegia: treatment by drainage and DREZ operation. Br J Neurosurg 1990;4:327-35. 77. Esposito S, Delitala A, Nardi PV. Microsurgical DREZlesion in the treatment of deafferentation pain. J Neurosurg Sci 1988;32:113-15. 78. Sindou M. Microsurgical DREZotomy (MDT) for pain, spasticity, and hyperactive bladder: a 20-year experience. Acta Neurochir (Wien) 1995;137:1-5. 79. Kori SH. Diagnosis and management of brachial plexus lesions in cancer patients. Oncology (Williston Park) 1995;9:756-60; discussion 765. 80. Zeidman SM, Rossitch EJ, Nashold BS Jr. Dorsal root entry zone lesions in the treatment of pain related to radiation-induced brachial plexopathy. J Spinal Disord 1993;6:44-7.
133
81. Kennedy PG. Varicella-zoster virus latency in human ganglia. Rev Med Virol 2002;12:327-34. 82. Friedman AH, Nashold BS Jr, Ovelmen-Levitt J. Dorsal root entry zone lesions for the treatment of post-herpetic neuralgia. J Neurosurg 1984;60:1258-62. 83. Makachinas T, Ovelmen-Levitt J, Nashold BS Jr. Intraoperative somatosensory evoked potentials. A localizing technique in the DREZ operation. Appl Neurophysiol 1988;51:146-53. 84. Bodley R. Imaging in chronic spinal cord injury– indications and benefits. Eur J Radiol 2002;42:135-53. 85. Potter K, Saifuddin A. Pictorial review: MRI of chronic spinal cord injury. Br J Radiol 2003;76:347-52. 86. Rawlings CE III, el-Naggar AO, Nashold BS Jr. The DREZ procedure: an update on technique. Br J Neurosurg 1989;3:633-42. 87. Vieira JF, Shieff C, Nashold BS Jr, et al. Impedance measurements of the spinal cord of man and animals. Appl Neurophysiol 1988;51:154-63. 88. Nashold BS Jr, Ovelmen-Levitt J, Sharpe R, et al. Intraoperative evoked potentials recorded in man directly from dorsal roots and spinal cord. J Neurosurg 1985;62: 680-93. 89. Tomas R, Haninec P. Dorsal root entry zone (DREZ) localization using direct spinal cord stimulation can improve results of the DREZ thermocoagulation procedure for intractable pain relief. Pain 2005;116:159-63. 90. Husain AM, Elliott SL, Gorecki JP. Neurophysiological monitoring for the nucleus caudalis dorsal root entry zone operation. Neurosurgery 2002;50:822-7; discussion 827–8. 91. Konrad PE, Tacker WA. Pyramidal versus extrapyramidal origins of the motor evoked potential. Neurosurgery 1991;29:795-6. 92. Oberle J, Antoniadis G, Kast E, et al. Evaluation of traumatic cervical nerve root injuries by intraoperative evoked potentials. Neurosurgery 2002;51:1182-8; discussion 1188–90. 93. Nashold BS Jr. Neurosurgical technique of the dorsal root entry zone operation. Appl Neurophysiol 1988;51: 136-45. 94. Yoshida M, Noguchi S, Kuga S, et al. MRI findings of DREZ-otomy lesions. Stereotact Funct Neurosurg 1992; 59:39-44. 95. Raslan AM, McCartney S, Burchiel KJ. Management of chronic severe pain: spinal neuromodulatory and neuroablative approaches. Acta Neurochir Suppl 2007; 97:33-41. 96. Sindou MP, Blondet E, Emery E, et al. Microsurgical lesioning in the dorsal root entry zone for pain due to brachial plexus avulsion: a prospective series of 55 patients. J Neurosurg 2005;102:1018-28. 97. Prestor B. Microcoagulation of junctional dorsal root entry zone is effective treatment of brachial plexus avulsion pain: long-term follow-up study. Croat Med J 2006;47:271-8.
2267
2268
133
Radiofrequency dorsal root entry zone lesions for pain
98. Richter HP, Seitz K. Dorsal root entry zone lesions for the control of deafferentation pain: experiences in ten patients. Neurosurgery 1984;15:956-9. 99. Thomas DG, Jones SJ. Dorsal root entry zone lesions (Nashold’s procedure) in brachial plexus avulsion. Neurosurgery 1984;15:966-8. 100. Garcia-March G, Sanchez-Ledesma MJ, Diaz P, et al. Dorsal root entry zone lesion versus spinal cord stimulation in the management of pain from brachial plexus avulsion. Acta Neurochir Suppl (Wien) 1987; 39:155-8. 101. Campbell JN, Solomon CT, James CS. The Hopkins experience with lesions of the dorsal horn (Nashold’s operation) for pain from avulsion of the brachial plexus. Appl Neurophysiol 1988;51:170-4. 102. Ishijima B, Shimoji K, Shimizu H, et al. Lesions of spinal and trigeminal dorsal root entry zone for deafferentation pain. Experience of 35 cases. Appl Neurophysiol 1988;51:175-87. 103. Saris SC, Iacono RP, Nashold BS Jr. Successful treatment of phantom pain with dorsal root entry zone coagulation. Appl Neurophysiol 1988;51: 188-97. 104. Saris SC, Vieira JF, Nashold BS Jr. Dorsal root entry zone coagulation for intractable sciatica. Appl Neurophysiol 1988;51:206-11. 105. Young RF. Clinical experience with radiofrequency and laser DREZ lesions. J Neurosurg 1990;72:715-20.
106. Kumagai Y, Shimoji K, Honma T, et al. Problems related to dorsal root entry zone lesions. Acta Neurochir (Wien) 1992;115:71-8. 107. Edgar RE, Best LG, Quail PA, et al. Computer-assisted DREZ microcoagulation: posttraumatic spinal deafferentation pain. J Spinal Disord 1993;6:48-56. 108. Samii M, Bear-Henney S, Ludemann W, et al. Treatment of refractory pain after brachial plexus avulsion with dorsal root entry zone lesions. Neurosurgery 2001;48:1269-75; discussion 1267–1275. 109. Falci S, Best L, Bayles R, et al. Dorsal root entry zone microcoagulation for spinal cord injury-related central pain: operative intramedullary electrophysiological guidance and clinical outcome. J Neurosurg 2002;97: 193-200. 110. Chen HJ, Tu YK. Long term follow-up results of dorsal root entry zone lesions for intractable pain after brachial plexus avulsion injuries. Acta Neurochir Suppl 2006;99:73-5. 111. Albert TJ, Vacarro A. Postlaminectomy kyphosis. Spine 1998;23:2738-45. 112. Konrad PE, Tacker WA Jr. Suprathreshold brain stimulation activates non-corticospinal motor evoked potentials in cats. Brain Res 1990;522:14-29. 113. Weigand H, Winkelmuller W. Gehandlung des Deafferentierungsschmerzes durch Hochfrequenzlasion der Hinterwurzeleintrittszone. Dtsch Med Wochenschr 1985;110:216-220.
141 Radiofrequency Rhizotomy for Trigeminal Neuralgia E. Taub
Clinical Features and Differential Diagnosis Idiopathic, or essential, trigeminal neuralgia is characterized by severe, lightning-like, shooting (‘‘lancinating’’) pain in the face in the distribution of one or more of the divisions of the trigeminal nerve. The pain is usually felt in the second and/or third division of the nerve, occasionally in the first. The disorder typically arises in middle age. The pain is always in the same place and can be provoked by lightly touching specific areas on the skin (‘‘trigger points’’) or by simple activities such as speaking, chewing, shaving, or brushing the teeth. A cool wind striking the face can precipitate an attack. These episodic pains may be accompanied by a dull ‘‘background’’ pain that is continually present; background pain thus does not rule out the diagnosis of idiopathic trigeminal neuralgia. The neurological examination generally reveals no abnormalities. There is no sensory deficit in the trigeminal distribution. Important facts for the clinician to know are that idiopathic trigeminal neuralgia is bilateral in a small percentage of cases, and that pain-free periods lasting months or years are common, even without treatment. The latter fact implies that there is no pressing need to treat patients with idiopathic trigeminal neuralgia if they are having a prolonged spontaneous remission. It may, however, still be wise to do so if the attacks have recurred again and again and have been severe and medically intractable with each recurrence. #
Springer-Verlag Berlin/Heidelberg 2009
Trigeminal neuralgia is sometimes caused by another illness. It is then ‘‘symptomatic,’’ rather than ‘‘essential’’ or ‘‘idiopathic,’’ trigeminal neuralgia. The major symptomatic causes are multiple sclerosis and masses in the cerebellopontine angle impinging on the trigeminal nerve. Symptomatic trigeminal neuralgia should be suspected in patients under age 50 and in those with a trigeminal sensory deficit or any other accompanying neurological signs and symptoms. On the other hand, a case of trigeminal neuralgia unaccompanied by any of these ‘‘red flags’’ may still turn out to be due to demyelinating disease or to a posterior fossa lesion. For this reason, any patient in whom the diagnosis of trigeminal neuralgia is made should undergo magnetic resonance imaging of the brain to establish the diagnosis of essential versus symptomatic types. The MRI also sometimes reveals a vascular loop adjacent to the trigeminal root entry zone, in accordance with the hypothesis of neurovascular contact as a cause of essential trigeminal neuralgia. This, too, is useful to know in making therapeutic decisions. Treatment algorithms differ among neurosurgeons treating trigeminal neuralgia, however, and the demonstration of a neurovascular contact by MRI in no way constrains the neurosurgeon to perform an open decompressive microvascular procedure, as opposed to a percutaneous rhizotomy. In summary, the diagnosis of idiopathic trigeminal neuralgia is based on its typical history, normal findings on neurological examination, and normal MRI findings (with or without a vascular loop). Percutaneous radiofrequency
2422
141
Radiofrequency rhizotomy for trigeminal neuralgia
rhizotomy is a good treatment option not only for patients with idiopathic trigeminal neuralgia, but also for those with trigeminal neuralgia due to multiple sclerosis [1,2]. If the MRI should reveal a posterior fossa lesion as the cause of the pain, this must, of course, be treated directly, usually by resection.
Pharmacotherapy Idiopathic trigeminal neuralgia generally responds well to medical treatment, at least at first. The agent of choice is still carbamazepine, usually at a dose of 600–1,000 mg/day, or the related drug oxcarbazepine (beware of hyponatremia when either of these is used). Gabapentin can also be highly effective. Pregabalin is a further drug that was recently introduced for the treatment of neuropathic pain in general. Phenytoin and the orphan drug L-baclofen seem to be used less commonly. The pain may be well controlled with medications for several years, but it typically worsens over time, so that the doses of medication must be raised or multiple drugs must be given in combination. Ultimately, a point is often reached where the pain can only be adequately controlled by medication at the cost of unacceptable side effects (usually sedation), or not at all. Once this happens, surgical treatment is clearly indicated.
The Choice of Surgical Treatment The choice between open microvascular decompression and percutaneous procedures for the treatment of trigeminal neuralgia is a contentious issue among neurosurgeons. Individual specialists naturally tend to advocate the treatments that they themselves perform (or to perform the ones that they advocate). The unique advantage of microvascular decompression is clearly that, when successful, it relieves the pain
of trigeminal neuralgia without producing any sensory deficit. All percutaneous procedures – among which radiofrequency rhizotomy is the most commonly performed type – rely on the generation of some degree of hypesthesia in the painful area. Experience shows that, if sensation in the painful area remains fully normal after the procedure, the pain will not be relieved. On the other hand, microvascular decompression is a highly invasive procedure with a relatively high frequency of major complications. If one adds up all of the serious complications encountered in the Pittsburgh series of 1,336 patients [3], one arrives at a figure of 6.6%: the most common ones were severe facial numbness (1.6%), cerebrospinal fluid leak (1.5%), and ipsilateral deafness (1.1%). ‘‘Chemical meningitis,’’ defined as self-limited symptoms of meningeal irritation, occurred in 16.8% of all patients. The potential complications of radiofrequency rhizotomy are not trivial, either (see below). Thus, the choice between these two types of procedure is by no means clear-cut. In general, it can be said that radiofrequency rhizotomy is clearly preferred over microvascular decompression in patients who are not candidates for open surgery by reason of advanced age or concomitant illness. The criteria for ‘‘advanced age’’ differ from institution to institution: for example, age over 60, age over 70, or advanced ‘‘biological age,’’ loosely defined. Many neurosurgeons consider microvascular decompression to be the better option for patients who are young enough and healthy enough to undergo it, yet others (including the author), in view of the complications, hold radiofrequency rhizotomy to be the treatment of choice for all patients. In any event, the neurosurgeon is obliged to discuss all potentially beneficial treatment options with the patient, including those that are thought to be less good, so that the patient can make his or her own informed decision. While it is certainly proper for the neurosurgeon to recommend
Radiofrequency rhizotomy for trigeminal neuralgia
one form of treatment over another based on his or her own experience and judgment, the patient’s autonomy in this matter must be respected: the patient may legitimately choose to go elsewhere for a different treatment.
Rationale of Percutaneous Radiofrequency Rhizotomy The procedure is commonly called a ‘‘rhizotomy,’’ yet this is actually a misnomer, as it does not involve the transection of a root either in an anatomical or in a functional sense. ‘‘Thermocoagulation’’ is a less familiar, but more accurate designation. The idea of the procedure is to use a radiofrequency electric current to generate heat locally in a portion of the trigeminal root, either in or just behind the Gasserian ganglion, in order to destroy some, but not all, of the nerve fibers within it. When this is done with ‘‘increments of graded heating,’’ as described by Sweet [4,5], the desired result is a near-total or total analgesia to pinprick in the cutaneous area that corresponds to the heated part of the trigeminal root, accompanied by only partial hypesthesia to light touch in the same area. The production of analgesia in the symptomatic area is well correlated with the relief of pain. It seems to be practically impossible to achieve analgesia without some degree of hypesthesia. On the other hand, total anesthesia in the symptomatic area is unnecessary for pain relief, and counts as a complication when it occurs.
141
discomfort it entails. Any anticoagulant medication that the patient may be taking, including aspirin, should be discontinued or reversed before the procedure. The procedure is performed in an operating room, either on an ambulatory basis or during a brief hospitalization (patients whose procedures are performed in the afternoon can go home the next day). An anesthesiologist is present at all times to administer sedation and to monitor the patient’s vital signs and O2 saturation; nasal prongs should be in position from the outset for the delivery of supplemental oxygen, as the patient’s respiratory rate may be temporarily depressed by the intraoperatively administered sedation. The tubes leading to these prongs should not overlie the patient’s cheek where the needle will be inserted; rather, they should be taped back over the forehead. A prophylactic intravenous dose of an antibiotic (e.g., cefuroxime) is given before the procedure is begun. The patient, initially fully awake, lies supine on the operating table with the head in the neutral position or tilted slightly backward. (If the head is tilted forward, the neurosurgeon’s working space will be limited by the patient’s chest). A fluoroscope is brought into position to obtain a precisely lateral image, including a sharp view of the sella turcica and the retrosellar shadow of the clivus (the ‘‘clival line’’). The skin of the patient’s cheek is disinfected with a colorless solution. This is important: if povidoneiodine or other colored solutions are used, it will be impossible to see the unilateral, local skin erythema typically induced by heating of the nerve root, as discussed below.
Technique Discontinuing the patient’s medications against trigeminal neuralgia on the day of the procedure may be a useful way to bring back the pain at full intensity and thus enable a robust check of its relief by the procedure. The author generally does not do this, however, because of the
Needle Placement The guide needle is inserted in essentially the same manner as described by Ha¨rtel in 1914 for injection treatments [6]. The anesthesiologist is asked to give the patient a first dose of short-acting
2423
2424
141
Radiofrequency rhizotomy for trigeminal neuralgia
intravenous sedation (generally methohexital or propofol) for the placement of the guide needle, and a skin bleb is raised with local anesthetic at the cutaneous insertion site, 2–3 cm lateral to the corner of the mouth on the symptomatic side of the face. The guide needle is inserted here and then advanced in a superior, posterior, and medial direction, under the buccal mucosa, toward the foramen ovale. A 10-cm needle is almost always long enough to reach the foramen and the Gasserian ganglion. Longer ones should also be at hand for the rare cases in which they are needed. If the procedure is performed on the right side, the neurosurgeon advances the guide needle with the left hand while palpating its submucosal course inside the mouth with the right index finger (which, of course, now becomes unsterile). With this finger positioned on the lateral pterygoid wing, the neurosurgeon feels the needle passing over this bony structure in a superoposterior direction while always remaining under the mucosa, i.e., one must be able to confirm that the needle never passes into the oral cavity. Looking at the patient’s face in a direct frontal view, the neurosurgeon verifies that the frontal projection of the externally visible course of the needle is in line with the ipsilateral pupil, or just medial to it. Meanwhile, on a lateral fluoroscopic image of the head, the tip of the needle should be pointing roughly in the direction of the midpoint of the clival line. As the tip of the needle is advanced, it will either pass directly through the foramen ovale or else strike the bony skull base next to it, in which case the neurosurgeon should slightly modify its position empirically until it slips into the foramen. At the moment that the needle passes through the foramen ovale, coming into close contact with the trigeminal nerve as it does so, two kinds of response may be observed. The patient, if he or she is not very deeply sedated at this time, may wince; and there may be an immediate vagal response, leading to transient bradycardia or even a few
seconds of asystole. This is a further reason why an anesthesiologist must be present. The foramen ovale cannot be seen directly on a lateral radiographic view of the skull, and the insertion of the needle into the foramen occasionally presents technical difficulties. Depending on the shape of the patient’s head, it may not be possible for the guide needle to traverse the entire bony canal of the foramen ovale if it is inserted precisely lateral to the corner of the mouth; it may need to be reinserted at either a higher or a lower point to obtain the proper angle. Occasionally, it may be useful to reposition the fluoroscope for an anteroposterior view, with the beam passing upward from the patient’s chin toward the back of the head, in order to obtain a head-on view of the foramen ovale (this is only rarely necessary). With the use of lateral fluoroscopy as described, one should be able to avoid passing the guide needle too far forward and into the inferior orbital fissure. In a small number of reported cases reviewed by Gybels and Sweet [5], a needle tip that was placed too medially entered the foramen lacerum and punctured the carotid artery: when the stylet was removed, bright red blood pumped forth from the needle. In most cases, there were no permanent complications after the needle was removed. Once the tip of the needle has passed through the foramen ovale, it is advanced further until it just overlies the clival line (> Figure 141‐1). This is the initial position for test stimulation through the trigeminal electrode. The stylet of the guide needle is now removed. A few drops of cerebrospinal fluid (CSF) generally emerge from the needle at this point; this finding is consistent with, but not conclusive proof of, a proper position of the needle [7]. There may also be no CSF at all, even if the needle is properly positioned, either because the needle tip lies directly within the neural tissue or because there is scarring in the area from a previous procedure.
Radiofrequency rhizotomy for trigeminal neuralgia
. Figure 141‐1 Intraoperative lateral fluoroscopic image of a correctly placed guide needle and trigeminal electrode in a 59-year-old woman with idiopathic trigeminal neuralgia in the left V2 distribution. A single lesion placed at this position (80 C for 60 s) resulted in dense hypalgesia and mild hypesthesia in the V2 distribution, as well as mild hypalgesia and mild hypesthesia in the V3 distribution. The pain of trigeminal neuralgia was relieved. ST, sella turcica; CL, clival line; FO, foramen ovale
Although puncturing the foramen ovale under fluoroscopic guidance in the manner outlined above is a relatively straightforward matter in experienced hands, a number of other techniques have been devised that, while being more cumbersome, may increase the likelihood of hitting the target on the first pass of the needle. Krol and Arbit [8] placed the guide needle in the computerized tomography suite under real-time CT guidance; Laitinen [9] did so stereotactically with a special-purpose ‘‘trigeminus stereoguide’’; and Bale et al. [10] have described a frameless stereotactic method.
Physiological Localization The neurosurgeon notes the impedance that is measured at the electrode tip: typical values are
141
in the range of 250–300 ohms. If the resistance is too low, the electrode tip may be lying in the subarachnoid space at an excessive distance from the neural elements, and the guide needle and electrode should be repositioned. By this time, the patient has generally awakened from the initial sedation, and a nondestructive stimulation test can be performed. In order to elicit paresthesia, a square-wave current is delivered at a frequency of 50 Hz and a pulse duration of 1 ms, at slowly rising intensity. The patient generally reports tingling in the face at a threshold value of 0.20–0.30 volts. The location of tingling depends on the particular population of nerve fibers that is stimulated, which, in turn, depends on the location of the electrode tip: tingling may be felt in the ophthalmic, maxillary, or mandibular distribution. The needle should be withdrawn or advanced, if necessary, so that the site of tingling coincides with the area of the patient’s pain. Taha and Tew [7] state that the electrode tip should be ca. 5 mm beyond the clival line for V1 stimulation, just at the clival line for V2 stimulation, and 5 mm in front of the clival line for V3 stimulation, although, of course, the optimal position will vary from patient to patient. Following Sweet [5], an additional stimulation test should be performed to prevent unintentional lesioning of the motor branch of the mandibular division: single shocks of a squarewave signal are given (e.g., at a frequency of 2 Hz) at gradually rising intensities so that the threshold for jerking of the masseter can be determined. This threshold should be at least three times as high as the sensory threshold at the same site. If not, the electrode should be repositioned laterally and the thresholds should be retested. A final aspect of physiological localization is the generation of a reversible ‘‘test lesion’’ at 60 C for 60 s. The patient is sedated for this and reawakened afterward. There will often, but not always, be cutaneous erythema due to vasodilatation in the area supplied by the fibers that
2425
2426
141
Radiofrequency rhizotomy for trigeminal neuralgia
are heated; if it is seen, this response should be in the same area as the patient’s pain. The erythema may be only slight—if so, it is best seen on the vermilion border of the lips—or it may be quite florid and unmistakable. Above all, a vasomotor response in the V1 area is a sign that the needle should be repositioned (except in the rare cases when the pain to be treated is in this area), as a permanent lesion here might cause corneal anesthesia. Once the patient is awake again, one tests cutaneous sensation to light touch and pinprick in all three trigeminal areas bilaterally, and checks both corneal reflexes. The findings, if positive, are generally a good indication of what a permanent lesion is likely to produce.
Lesioning Permanent thermal lesion-making then follows in increments of 5–10 C for 1 min or less. The patient may not need to be sedated to the same degree, or at all, for the later lesions, as these will be less painful because of the loss of sensation already induced by the earlier ones. After each lesion is made, sensory testing is performed once more in the awake patient, including testing of cutaneous sensation to light touch and pinprick in all three trigeminal areas bilaterally and a check of both corneal reflexes. The endpoint of lesion-making is robust and reproducible hypalgesia or analgesia in the initially painful area, with no more than partial hypesthesia. Of course, if the patient’s typical neuralgic pain was present right at the beginning of the procedure (despite whatever medications the patient had been taking up to that point), then the abolition of this pain is a further important endpoint. Lesioning should be stopped even before these endpoints are reached, however, if cutaneous anesthesia is produced or if the corneal reflex is abolished. Taha and Tew [7] recommend waiting 15 min after an effective, and putatively last,
lesion has been made before withdrawing the needle, as the effect may turn out to be transient and require reinforcement with another lesion. Once the desired endpoint has been reached, the needle is withdrawn and the puncture site is covered with a small dressing. The patient is sent to the recovery room for a brief period of observation, in accordance with the duration of sedation, and then back to the ward.
Postoperative Care The patient is discharged home on the day of the procedure, or on the next day at latest, and may return to work the day after discharge. Artificial tears are prescribed if the corneal reflex has been unintentionally abolished. The patient’s medications for trigeminal neuralgia are tapered to off in linear fashion over a period of 3–4 weeks.
Results and Complications Because of the brevity of the procedure (1–2 h), the ease with which it can be carried out in experienced hands, the relatively small number of neurosurgeons that are well versed in it, and the large number of patients who need it, it has been possible for multiple neurosurgical centers to accumulate very extensive case series with more than 500 patients each. Gybels and Sweet [5] tabulated the results of 11 such series that had been published up to 1988; a very large, more recent series is that of Kanpolat, including 1,600 patients, published in 2001 [11]. The results can be summarized as follows:
Early, complete relief of pain was obtained in over 95% of patients in all series, though early reoperation was sometimes necessary to regain initial relief. The rate of late recurrence (at some time after 1 year) requiring reoperation is generally
Radiofrequency rhizotomy for trigeminal neuralgia
in the range of 20–30%. Relief of pain after the procedure persists much longer if dense hypalgesia is produced than if only mild hypalgesia is produced [12]. The procedure can generally be repeated without difficulty if the pain recurs. The pain relief that results is equally good as after the first procedure, and it will probably last longer. Dysesthesia in the treated area is a common postoperative phenomenon (5–24% in various series) that sometimes requires treatment with medication. ‘‘Anesthesia dolorosa’’ is defined as dysesthesia arising in a totally anesthetic area and is much rarer (1.2%). It is not clear whether the distinction of anesthesia dolorosa from dysesthesia in a still at least partially sensate area is clinically useful. Nor is it clear from the literature how long the reported cases of dysesthesia tended to persist: many cases are presumably self-limited (in analogy to the dysesthesia that occurs after spinal rhizotomy and ganglionectomy procedures, which has been carefully studied) [13]. Nonetheless, persistent and severe dysesthesia or anesthesia dolorosa can be a significant therapeutic problem. In medically intractable cases, a further functional neurosurgical procedure may have to be considered (e.g., motor cortex stimulation). Weakness of the muscles of mastication occurred in 10–15% of patients in typical series, and in as many as 25% in some series. The duration of this side effect is not indicated in all reports. In the author’s experience, it tends to resolve over several weeks or months. Physical therapy may be beneficial. Corneal anesthesia is not synonymous with an absent corneal reflex, and the two are not always carefully distinguished in published reports. The frequency of one or the other, or both, ranges from 6 to 20% or more. Corneal anesthesia cannot always be avoided with careful technique, as it may occur even when
141
the intraoperatively observed vasomotor response spares the ophthalmic distribution [5]. It may often be transient, as the author has observed in two personal cases, and it seems to lead to keratitis in only a minority of affected patients (e.g., 2 of 6 patients in Sweet’s series [5,14]). Neuroparalytic keratitis is a trophic phenomenon that predisposes to infection but is not necessarily accompanied by it. At worst, it can lead to loss of vision or even of an eye, as has rarely been reported. All patients with corneal changes should be given artificial tears to use several times a day and should be referred to an ophthalmologist for close follow-up. Cranial nerve palsies (of nerves other than the intended trigeminal nerve) are rare: they are encountered in at most 1–2% of all patients in published series. These generally affect the oculomotor, trochlear, or abducens nerves, because of their proximity to the trigeminal nerve. The author has had a personal experience of two postoperative trochlear nerve palsies, both of which resolved completely (one within 2 months, the other within 6 months). Such cases presumably represent transient neurapraxia due to a mild thermal injury of the affected nerve. Transient reactivation of herpes simplex lesions (cold sores) on the lips is an occasional complication that is probably well known to all neurosurgeons who perform this procedure, though it is perhaps considered too trivial to be mentioned in most published series. Patients with bothersome cold sores can be referred to a dermatologist for antiviral and/or symptomatic treatment until these resolve. Likewise, buccal hematoma along the path of the guide needle may seem too trivial to be mentioned in published reports but is a source of transient distress to some patients (perhaps 10%). Like any subcutaneous hematoma, this resolves within a few weeks.
2427
2428
141
Radiofrequency rhizotomy for trigeminal neuralgia
Conclusion Percutaneous radiofrequency trigeminal rhizotomy is a remarkably effective treatment for trigeminal neuralgia, after which most treated patients are free of pain and can discard the medications that they took for this condition. Its complications are rare and mostly mild and transient. The pain may recur within a few years, especially if the hypalgesia produced by the procedure is not dense. In such cases, a second procedure can be performed and will generally be at least as effective as the first one.
6. 7.
8.
9.
10.
References 11. 1. Kanpolat Y, Berk C, Savas A, Bekar A. Percutaneous controlled radiofrequency rhizotomy in the management of patients with trigeminal neuralgia due to multiple sclerosis. ActaNeurochir (Wien) 2000;142(6):685-9. 2. Berk C, Constantoyannis C, Honey CR. The treatment of trigeminal neuralgia in patients with multiple sclerosis using percutaneous radiofrequency rhizotomy. Can J Neurol Sci 2003;30(3):220-3. 3. Barker FG 2nd, Jannetta PJ, Bissonette DJ, Larkins MV, Jho HD. The long-term outcome of microvascular decompression for trigeminal neuralgia. N Engl J Med 1996;334 (17):1077-83. 4. Sweet WH. Trigeminal neuralgias. In: Alling CC, editor. Facial pain. Philadelphia: Lea and Febiger; 1968. p. 89-106. 5. Gybels JM, Sweet WH. Thermal rhizotomy, Chap IA. In: Neurosurgical treatment of persistent pain:
12.
13.
14.
physiological and pathological mechanisms of human pain. Basel: Karger; 1989. p. 17-43. ¨ ber die intracranielle Injektionsbehandlung der Ha¨rtel F. U Trigeminusneuralgie. Med Klin 1914;10:582-4. Taha JM, Tew JM Jr. Radiofrequency rhizotomy for trigeminal and other cranial neuralgias, Chap 172. In: Gildenberg PL, Tasker RR, editors. Textbook of functional and stereotactic neurosurgery. NY: McGraw-Hill; 1996. p. 1687-96. Krol G, Arbit E. Percutaneous electrocoagulation of the trigeminal nerve using CT guidance. Technical note. J Neurosurg 1988;68(6):972-3. Bale RJ, Laimer I, Martin A, Schlager A, Mayr C, Rieger M, Czermak BV, Kovacs P, Widmann G. Frameless stereotactic cannulation of the foramen ovale for ablative treatment of trigeminal neuralgia. Neurosurgery 2006; 59(4 Suppl 2):ONS394-401. Laitinen LV. Trigeminus stereoguide: an instrument for stereotactic approach through the foramen ovale and foramen jugulare. Surg Neurol 1984;22(5):519-23. Kanpolat Y, Savas A, Bekar A, Berk C. Percutaneous controlled radiofrequency trigeminal rhizotomy for the treatment of idiopathic trigeminal neuralgia: 25-year experience with 1,600 patients. Neurosurgery 2001;48(3):524-32. Taha JM, Tew JM Jr, Buncher CR. A prospective 15-year follow up of 154 consecutive patients with trigeminal neuralgia treated by percutaneous stereotactic radiofrequency thermal rhizotomy. J Neurosurg 1995;83(6): 989-93. Taub A, Robinson F, Taub E. Dorsal root ganglionectomy for intractable monoradicular sciatica. In: Schmidek HH, Sweet WH, editors. Operative neurosurgical techniques: indications, methods, and results, 3rd ed. Philadelphia: W. B. Saunders; 1995. p. 1585-93. Lewis RA, Keltner JL, Cobb CA. Corneal anesthesia after percutaneous radiofrequency trigeminal rhizotomy. A retrospective study. Arch Ophthalmol 1982;100(2):301-3.
142 Retrogasserian Glycerol Injection for Trigeminal Neuralgia B. Linderoth . G. Lind
Introduction Many patients with trigeminal neuralgia (TN) are elderly, suffer from chronic diseases or are generally weak. Hence therapeutic minimally invasive methods with low surgical risk, little impact on facial sensibility, and the possibility to be performed in local anesthesia with only slight sedation are necessary. Retrogasserian glycerol rhizolysis has these advantages. In such patients even the usual pharmaceutical regimen with carbamazepine, other anticonvulsants, or baclofen is known to carry severe side effects. These problems apply particularly to patients with paroxysmal facial pain associated with multiple sclerosis (MS). The discovery of the beneficial effects of glycerol in patients with TN was purely accidental. In a first trial series in the 1970s, to use the Leksell Gamma Knife in Stockholm to create lesions in the Gasserian ganglion as a therapy for TN, glycerol was used as vehicles for a radiopaque metal dust (tantalum powder) [1,2]. Surprisingly it was noted that merely injecting glycerol and tantalum dust mixture in patients abolished the paroxysmal pain before the Gamma knife procedure was performed. On the basis of these observations, Sten Ha˚kanson, neurosurgeon at the Karolinska Hospital in Stockholm, developed the technique for treating TN by glycerol injection into the trigeminal cistern. The first series of patients was presented in 1981 [3] and the method was then rapidly adopted in several neurosurgical centers. Over the years, many series of patients treated using Ha˚kanson’s procedure, or some #
Springer-Verlag Berlin/Heidelberg 2009
variation of the original method, have been reported. The results from different series have been highly variable. In many centers the outcome has been quite satisfactory [4–8] and glycerol rhizolysis has continued to be the method of choice, in particular for elderly and weak patients. In other services, the results have been so discouraging (Siegfried, 1985, and Rhoton, 1985, unpublished results, both cited in Sweet [9]; Price, 1985, unpublished results, cited in Sweet [10] and Fujimaki and colleagues [11]) that some neurosurgeons rapidly abandoned the procedure. In this chapter, the possible reasons for these discrepancies are examined and a standard procedure to ensure maximum efficacy and safety is described.
Physiological Background The etiology of trigeminal neuralgia is probably multifactorial but degradation of the myelin sheath due to advancing age, neurovascular conflicts with compression of the nerve root by an arterial branch in the posterior fossa or demyelination and formation of plaques in multiple sclerosis (MS) seem to form a common denominator behind the syndrome. In many instances, however, the etiology remains obscure. In the literature we find no satisfactory animal models of trigeminal neuralgia, and it is difficult to obtain relevant histologic data from patients. However, trigeminal neuralgia presents with such idiosyncratic signs and symptoms, and
2430
142
Retrogasserian glycerol injection for trigeminal neuralgia
responds to so distinctive a set of therapeutic modalities, that scientific deduction can be used to generate likely hypotheses. The ‘‘ignition hypothesis’’ of trigeminal neuralgia [12–14] is based on advances in the understanding of abnormal electrical activation of injured sensory neurons [15], supported by histopathological examinations of biopsy specimens from patients with trigeminal neuralgia who are undergoing microvascular decompression (MVD) of the trigeminal root in the posterior fossa [12]. According to this hypothesis, trigeminal neuralgia results from specific abnormal activation of trigeminal afferent neurons in the trigeminal root or ganglion. Injury renders both axons and axotomized somata hyperexcitable. The hyperexcitable afferents, in turn, induce pain paroxysms as a result of synchronized after-discharge activity. The ignition hypothesis accounts for the major positive and negative signs and symptoms of trigeminal neuralgia, for its pathogenesis, and for the efficacy of treatment modalities (for discussion cf [13,14]. The only therapeutic method currently in use for tic doloeureux directed onto one of the above-mentioned etiological factors is microvascular decompression (MVD) where the surgeon resolves a neurovascular conflict if found during posterior fossa exploration. The other surgical approaches all include graded lesioning of the trigeminal nerve (i.e., postganglionic root fibers or proximal root) using minimally invasive approaches chemically, or with heat, physical compression or radiation (Gamma Knife treatment). The attained spectrum of fibers in the nerve seems to differ between the methods but the clinical outcome is somewhat similar, i e relief from the pain paroxysms with no or only slight side-effects. A more detailed discussion of physiological/ morphological and neurochemical effects of glycerol on nervous tissue follows after the description of the methodology, outcomes and side-effects.
Indications The main indication for glycerol rhizolysis remains classic idiopathic TN. Common reasons for progressing to surgical treatment include deficient control of paroxysms in spite of an adequate pharmaceutical regimen, severe medication side effects, development of drug allergy or intolerance, or signs of hepatic or renal malfunction ascribed to medication. Paroxysmal facial pain in MS is another prime indication. The initial outcome in this group of patients is as satisfactory as for idiopathic TN, but the long-term results are, as with other available methods, less encouraging. This will be further discussed below. Patients with signs of deafferentation should, in principle, not be submitted to a procedure causing additional damage to the nerve. However, many patients with TN previously treated by other methods display signs of neural lesion, such as hypesthesia, allodynia, hyperalgesia, and some degree of continuous deafferentation pain. Such patients should be accepted for treatment only if the paroxysmal pain component is dominant and after careful evaluation of sensory deficits. If such patients are accepted for glycerol rhizolysis, the procedure should be carried out, using only a small amount of glycerol. The same considerations also apply to the use of glycerol rhizolysis in atypical facial pain/painful trigeminal neuropathy. The general rule is that glycerol rhizolysis is not indicated in such cases. Only when a dominant paroxysmal component is present and the signs of deafferentation are slight, may glycerol injection be considered. Both neurosurgeon and patient should be aware that the procedure might aggravate the deafferentation and thereby the constant pain component. Preoperative examination. The preoperative evaluation should be focused on the presence of typical signs of TN, previous treatments, the
Retrogasserian glycerol injection for trigeminal neuralgia
pharmaceutical regimen, the presence of sensory deficits, constant pain components, and ipsilateral hearing loss. Because we recommend the use of contrast medium injection in all cases, intolerance to iodine and previous adverse reactions to contrast medium should be investigated. It is recommended that a magnetic resonance imaging (MRI) study preferably with an MRI sequence optimized for detection of vascular structures adjacent to the nerve root, or at least a computed tomography scan with and without contrast injection, should be performed before surgery. The surgeon and the anesthetist must evaluate the patient before the procedure to individualize premedication and describe the details of the procedure to the patient. Most patients tolerate the procedure well in local anesthesia with adequate premedication, and with only slight sedation but very anxious patients may require general intubation anesthesia.
Technique The original technique of Ha˚kanson has been subject to many modifications by various neurosurgeons. These variations encompass the type of anesthesia selected, general or local; patient position and fluoroscopic projection; whether or not cisternography is performed; other modes of localization of the needle tip (electric stimulation; reactions to drop-by-drop injection of glycerol); the dose of glycerol used; instillation of glycerol in one step or as minute volumes in an incremental fashion, with intermittent sensory testing; trials to empty the cistern after attaining a satisfactory effect according to intraoperative testing; and the time period the patient is kept sitting with the head flexed after the procedure. Some of these modifications have resulted in less satisfactory results [9,11,16,17]. We consider retrogasserian glycerol rhizolysis to be an anatomically oriented method aimed at graded
142
lesioning of fibers in a certain locus (postganglionically). Thus, the localization procedure should also be anatomic and the treatment should be meticulously performed, using the smallest possible volume of pure, sterile glycerol considered to be effective in each case. The procedure as it is currently performed at the Department of Neurosurgery, Karolinska University Hospital, Stockholm is described below. Anesthesia. The entire procedure, at the Karolinska University Hospital, was, between the late 1970s up to the late 1990s, carried out with the patient awake and premedicated approximately 45 min before the start of the session with 5–10 mg of morphine hydrochloridescopolamine administered subcutaneously and 2.5 mg of droperidol administered intramuscularly. The doses were adjusted according to the age and condition of the patient. In some cases, it was helpful to give 0.5 mg of atropine intravenously immediately before the procedure to protect against bradycardia during needle insertion. An intravenous line with slow infusion of Ringer’s solution is maintained during and for some hours after the session. This type of approach may still be the only possible in some settings where anesthetists are lacking. During recent years sedation with i v propofol using a syringe-pump has been used throughout the procedure at the Karolinska University Hospital. No intubation has been required and oxygen supply has been ascertained via a nasal catheter. Usually a low peroral dose of some benzodiazepine compound has been used 1 h preoperatively. The legs should be wrapped or compression stockings used to counteract blood pressure drop in the semi-sitting position. General anesthesia with a short-acting barbiturate and endotracheal intubation is used only in particularly anxious patients. If used, it is important that the anesthesia be terminated with the patient in the sitting position with the
2431
2432
142
Retrogasserian glycerol injection for trigeminal neuralgia
. Figure 142‐1 Schematic picture illustrating the contents of the Meckelian cave: the Gasserian (semi-lunar) ganglion); the retroganglionic root fibers in the CSF-filled cistern and the meningeal coverings (modified from Ha˚kanson 1982). A needle is penetrating the ganglion, entering the cistern
patient’s head flexed according to the surgeon’s instructions. The skin at the point of needle insertion and the underlying soft tissue is infiltrated with local anesthetic (e.g., lidocaine 0.5% with adrenaline). Patient positioning. The procedure was earlier performed in a radiography suite with the patient seated in a rotating chair, but since the mid 1990s a slightly modified dentist’s operating chair in the surgical theater has been successfully used instead. A standard C-arm fluoroscopic image intensifier with image-storing capacity provides sufficient picture quality for the clinical procedure – but unfortunately not always for publication purposes. In most cases, fluoroscopy with lateral projection is used when the cistern is punctured. Further guidance is obtained by switching to the anteroposterior projection. In rare, difficult cases in which entering the proper part of the oval foramen is a problem, the patient’s head may be extended and rotated 15–20 away from the affected side, and the fluoroscopy arm tilted
to give an axial-oblique projection of the skull base including the foramen ovale. With old equipment, it might be difficult to readily identify the proper foramen on the fluoroscopy monitor, but exposing a film in most cases solves the problem. With new radiographic equipment it poses little problem. If one needle already has penetrated the foramen, the identification should be easy, and a new needle can readily be inserted in the desired (medial) part of the foramen. Surgical anatomy. The trigeminal cistern is punctured by the anterior percutaneous route through the foramen ovale, as described by Ha¨rtel [18] (> Figure 142‐1). After local anesthesia, a 22-gauge lumbar cannula (OD 0.7 mm; length 90 mm) is inserted from a point approximately 3–4 cm lateral to the corner of the mouth. The trajectory is aimed at a point that lies, in the lateral view, approximately 0.5 cm anterior to the anterior margin of the mandibular joint, and in the anteroposterior view, toward the medial margin of the pupil with the eyeball in the neutral position. There are several landmarks
Retrogasserian glycerol injection for trigeminal neuralgia
that may be used for reaching the foramen ovale [19,20], but in most cases these two simple coordinates are sufficient. Often it is wise to direct the needle even a little more medial, touching the medial wall of the foramen. The needle is then withdrawn a short distance, redirected a few millimeters more lateral, and introduced through the medial part of the foramen. Intermittent fluoroscopy is used during the entire procedure. When the needle penetrates the foramen, the patient- even when sedated- may signal a brief episode of pain. This is of course due to penetration of the third branch and the semilunar ganglion. As a rule, the cannula should not reach beyond the clival contour as seen on the orthogonal lateral projection. When the tip of the cannula in located inside the arachnoid of the trigeminal cistern, there should be a spontaneous exit of cerebrospinal fluid (CSF). Because the location of the trigeminal ganglion and cistern can vary in relation to the landmarks of the skull base, a contrast injection must be performed to ascertain the correct site for glycerol injection. However, as discussed later, spontaneous CSF drainage is not a sufficient requisite for accepting the location as intracisternal. Trigeminal cisternography. The technique we presently use is based on that described by Ha˚kanson [21], although estimation of the cisternal volume is of less importance and usually a standard volume is injected. The contrast medium must be water soluble, with high radiographic attenuation and low toxicity, and must have a higher specific gravity than the CSF. Nowadays, iohexol (300 mg iodine/ml) is the standard choice [22]. Approximately 0.3–0.6 ml is injected with the patient sitting with the head slightly flexed to retain as much of the medium in the cistern as possible. If intermittent fluoroscopy is used during injection, the position of the needle tip may be estimated immediately, but the exact position should always be confirmed
142
. Figure 142‐2 The typical pear-shaped appearance of the contrastfilled trigeminal cistern (arrow) in the lateral (a) and the almond-shaped medial tilt in the antero-posterior projections (b). Note the root fibers (in (a)) that give the cistern its stripexd texture. (modified from Linderoth and Ha˚kanson 2006)
by radiography in both the lateral and anteroposterior projections. The typical pear-shaped appearance of the trigeminal cistern is illustrated in > Figure 142‐2a. Ideally, the sensory root filaments should be visualized by lateral cisternography, leaving no doubt about the intracisternal
2433
2434
142
Retrogasserian glycerol injection for trigeminal neuralgia
. Figure 142‐3 Schematic picture of the different compartments in Meckel’s cave. The arachnoid is drawn with a wavy line, the dura with a thick line. Note the sites (*) where extra-cisternal deposits of contrast medium and glycerol may be performed (From Pauser G, Gerstenbrand F, Gross D [eds]: Gesichtsschmerz, Schmerzstudien Z. New York: Verlag, 1979.)
position of the tip. The typical 45 medial tilt of the almond-shaped cistern should be seen from the antero-posterior view (> Figure 142‐2b). The appearance of the cistern may vary considerably between patients. This is why it is essential that the surgeon is familiar with the regional anatomy. Furthermore, as is illustrated in > Figure 142‐3, there is a subdural-extracisternal compartment in Meckel’s cave that may be injected with contrast medium. This usually happens when contrast medium is injected without prior spontaneous CSF drainage, or if the needle is dislodged from its intracisternal position during the procedure (see below > Figure 142‐6). Specific difficulties. Spontaneous CSF drainage from the needle does not guarantee an intracisternal tip location. It is true that slow CSF drip from the cannula indicates a possible intracisternal needle tip position and this has recently been reported to correlate with a successful outcome [23] confirming earlier experiences. In fact, if the cannula is placed a few
millimeters too far laterally, the tip may be located in the subtemporal subarachnoid space, from where a brisk flow of CSF may occur. A subsequent cisternography solves this problem, as also illustrated in > Figure 142‐5a, a schematic A-P drawing, showing the proximity of the cisternal and subtemporal subarachnoid compartments in the anteroposterior projection. > Figure 142‐4a demonstrates a leakage from the cistern to the adjacent subtemporal space and > Figure 142‐4b,c illustrate purely subtemporal contrast injections. In > Figure 142‐5b the antero-posterior fluoroscopy demonstrates a too medial needle tip position, a case which is less common than the subtemporal placement. Our strategy in such cases has been to leave the first needle in place and to introduce a second needle using the first one for guidance (> Figure 142‐5b). It cannot be overemphasized that spontaneous CSF drainage is required before contrast injection. If, in such a case, the contrast medium is injected without fulfilling this requirement, the
Retrogasserian glycerol injection for trigeminal neuralgia
142
. Figure 142‐4 The antero-posterior roentgenogram (a) with both the trigeminal cistern and part of the subtemporal space injected with contrast medium, illustrating the proximity of the two compartments. Medially, the contrast-filled cistern is seen barely pierced in its lateral margin by the needle. Injection of contrast medium spilled over also to the subtemporal space (arrow). The lateral roentgenogram (b) shows a pure subtemporal contrast injection. (c) Subtemporal outflow of contrast medium in the antero-posterior projection is shown. No contrast filling of the cistern was obtained, although brisk exit of cerebrospinal fluid from the needle was obtained. (modified from Linderoth and Ha˚kanson 2006)
glycerol may be deposited in the extra-arachnoid, subdural compartment (> Figure 142‐6a) producing no or much less effect on the neuralgia [23]. Another problem may arise if the patient’s head is not adequately tilted forward during
the injection, because the contrast medium then flows out the porus trigemini and escapes to the posterior fossa, with insufficient cisternal filling to confirm the correct position of the needle tip (> Figure 142‐7). This situation
2435
2436
142
Retrogasserian glycerol injection for trigeminal neuralgia
. Figure 142‐5 The schematic figure (a) illustrates that because the oval foramen serves as a fulcrum point, the needle entry site determines the intracranial destination of the needle tip. A puncture site that is lateral produces a medial tip position (a). The adequate entry is shown by the track (b). (Adapted from Jho HD, Lunsford LD: Percutaneous retrogasserian glycerol rhizotomy. Neurosurg Clin North Am 8:63–74, 1997). A far lateral puncture site results in a medial tip position (5b). With an entry that is too medial the result may prove a placement of the needle tip lateral to the trigeminal cistern (resulting in a subtemporal contrast deposition: (4a-c.). 5c: The antero-posterior roentgenogram demonstrates the penetration of the oval foramen with three needles. Only contrast injection via the most medial one produced adequate cistern filling
may also jeopardize estimation of the volume of the cistern. Contrast evacuation. When the intracisternal position of the tip has been confirmed, the syringe
is removed from the cannula and the contrast medium permitted to flow out of it in a dropby-drop fashion. We usually evacuate the cistern by tilting the patient to the supine position. If it is
Retrogasserian glycerol injection for trigeminal neuralgia
142
. Figure 142‐6 When an adequate fluoroscopic needle position is obtained without spontaneous CSF exit the needle tip may well be in the extra-arachnoidal/subdural space as illustrated by these pictures showing the faulty injection in (a), the proper intracisternal deposit in (b), and a subdural injection (A–P projection) in (c)
difficult to evacuate all contrast medium from the bottom of the cistern to permit the glycerol to reach the lowest root fibers (see below), the patient can be placed in the Trendelenburg position for a few minutes, a maneuver that permits the contrast medium to drain to the posterior fossa. Furthermore, the cistern may be flushed with 2–3 ml of sterile saline until the control radiograph indicates that the cistern is emptied. This procedure
though not painful may induce some kind of unpleasant pressure sensation awakening the patient from sedation. This is a problem that may arise especially with reinjections of glycerol and may be due to fibrosis in the bottom of the cistern (see further below). Some of the less satisfactory results to be discussed below are probably due to failure to empty the cistern properly [11,17].
2437
2438
142
Retrogasserian glycerol injection for trigeminal neuralgia
. Figure 142‐7 During contrast injection, the patient’s head should be tilted slightly forward to prevent immediate contrast flow to the posterior fossa. As shown here, the positioning of the head was inadequate, and contrast immediately escaped (arrow) to the posterior fossa, resulting in a merely partial cistern filling (modified from Linderoth and Ha˚kanson 2006)
Glycerol injection. Glycerol injection should always be performed with the patient in the sitting position to minimize extracisternal spillover. The glycerol should be anhydrous (>99.5%) and sterile, and should be injected slowly from a 1-cc syringe. In general, 0.18–0.30 cc is sufficient, and the usual dose in our department is approximately 0.22–0.28 cc (present average 0.25 cc). When the neuralgia encompasses all three branches with multiple trigger points, a somewhat larger volume has to be used as compared to a case where only the third branch is involved. However, injection of volumes exceeding 0.35 ml is discouraged because of the risk of postoperative sensory deficits (see below). Furthermore, if a procedure fails to demonstrate an adequate intracisternal needle tip position a subsequent glycerol injection is contraindicated. Branch selectivity. If it is necessary to inject glycerol selectively into one or more trigeminal branches, one of the following four maneuvers can be used.
First, the volume of glycerol can be varied to fill more or less of the cistern. When the trigeminal cistern has been emptied entirely after contrast injection, the amount of glycerol injected partly determines which branches will be influenced. With the head of the patient only slightly flexed, a small volume of glycerol (i.e., 0.15–0.2 ml) is deposited at the bottom of the cistern and mainly affects the fibers of the third and second branches. Increasing the amount of glycerol causes additional rhizolysis in the two upper portions. Secondly, a small volume of contrast medium may be left in the bottom of the cistern to protect the third branch. Both contrast agents used so far have specific gravities exceeding that of pure glycerol (with 300 mg iodine/ml, metrizamide = 1.329g/l and iohexol = 1.345 g/l; compared with glycerol = 1.242 g/l and CSF = 1.007 g/l). This implies that these substances, when deposited in the same compartment, replace the original CSF contents and are layered with the contrast medium at the bottom and the glycerol on top. With the head slightly flexed and a little contrast medium remaining at the bottom of the cistern, thus protecting the fibers of the third branch, glycerol may be gently injected to form a layer floating on the contrast medium. Rhizolysis will then engage mainly the upper two branches. It also follows that when treating a patient with third-branch neuralgia, no contrast medium should be allowed to remain in the cistern for the glycerol injection (see above about saline rinsing). Thirdly, the position of the needle tip is partly indicative of which branch will be most affected. The tip of the cannula should preferentially be positioned in the part of the cistern traversed by the root fibers to be treated. This means that for treatment of third-branch neuralgia, the optimal position is in the lower part of the cistern, whereas a first or second-branch
Retrogasserian glycerol injection for trigeminal neuralgia
neuralgia is best treated using a tip site in the upper portion. During or immediately after injection of glycerol, the patient usually experiences strong paresthesias (a ‘‘pins-and-needles’’ sensation or sometimes even pain) in one or more of the divisions of the trigeminal nerve. This is an indication that the substance has been deposited inside the cistern and is affecting the closest root fibers. Usually this sensation persists for only 20–30 s. Proper placement of the needle seems especially important with reinjections, when adhesions inside the cistern may be present (> Figure 142‐9b) and partially prevent the drops of glycerol from sinking to the bottom to affect the fibers of the mandibular branch. Finally, the position of the patient’s head during and after the injection also produces some selectivity. The patient should be sitting in the upright position during the injection. For first-branch neuralgia, the head should be maximally flexed (i.e., approximately 40 ) and should be kept in approximately that position for 1 h after the injection. For second-branch treatment, flexion should be approximately 25 , and for the third branch, the head should be slightly tilted laterally toward the affected side, but kept upright in the anteroposterior plane [4]. The selectivity obtained by varying the head position has been systematically studied by Bergenheim and colleagues [24,25], who found a highly selective effect of this maneuver with regard to the ophthalmic branch, but less selectivity for the two lower branches. Immediately before glycerol injection, the free flow of CSF from the needle should be checked to ascertain a persisting intracisternal tip position. Dislocation is not unusual, especially if the full length of the needle had to be used and the patient moves his cheek. In certain cases, a longer needle has to be used to reach the cistern proper. Often, thin needles (OD 0.7 mm) of adequate length are not readily available in the hospital and have to
142
be ordered separately. It may be cumbersome to use such a long and thin cannula and if problems with needle guidance we now recommend switching to a needle with OD 0.9 120 mm since the use of propofol sedation in addition to LA takes care of the increase in pain during penetration of the ganglion. In full intubation anesthesia we regularly utilize larger diameter standard needles (OD 0.9 mm; length 90–120 mm) when needed in order to more easily navigate and to safely reach the cistern. Finally, when the tip of the needle is placed posteriorly in the cistern close to its clival passage, injection might cause pain projected to the eye, so such a position should be avoided. It is also possible that corneal complications are related to injections in this area. Permanent marking of cistern. A permanent marking of the cistern is recommended and can easily be obtained using the original mix of glycerol and sterile radiopaque tantalum dust (Merck, Haar, Germany; grain size <0.042 mm). Approximately 0.5 g of tantalum is mixed with 2 ml of glycerol and the suspension is then used for injection. Marking in this way usually outlines the bottom of the cistern (> Figure 142‐8), and is of considerable help as a guide if reinjections are needed. With certain requirements fulfilled, permanent marking may replace the need for renewed cisternography (see below). The tantalum marking is clearly visible on radiography many years after the injection. There are no indications that the metal dust per se causes inadvertent effects in the form of meningeal reactions, for example, or otherwise influences the results of the treatment. Repeat injections. Incomplete pain relief after the initial injection seems in most cases to be the result of technical problems during the procedure. The high initial success rate reported in many papers (80–97%) supports this view. Reinjections should not be performed within 3–4 weeks after the initial trial, a criterion set to exclude late responders. Before a
2439
2440
142
Retrogasserian glycerol injection for trigeminal neuralgia
. Figure 142‐8 The trigeminal cistern on the patient’s left side has previously been permanently marked by tantalum dust (arrow), outlining the arachnoid of the bottom and walls of the cistern (antero-posterior view). The right cistern has now been punctured and injected with contrast medium. After marking of the cistern with tantalum, the puncture and glycerol injection may be carried out without the intervening cisternography (see text for recommendations)
reinjection is made, facial sensibility should be carefully examined. If signs of sensory impairment are found, the reinjection should be meticulously performed using a minimal amount of glycerol. Alternatively, a non-destructive procedure should be considered in these patients. When trigeminal cisternography is performed after one or more intracisternal glycerol injections, less satisfactory outlining of the cistern sometimes is obtained (see > Figure 142‐9). The probable cause is the formation of fibrosis within the cistern induced by the glycerol and/or contrast medium [26]. If the cistern has been properly marked by tantalum dust during the initial injection and the marking is clearly visible during fluoroscopy, reinjection may be very simple and fast. When the position of the tip of the cannula is within the cistern according to radiography in the lateral and at least two antero-posterior projections at different angles, spontaneous drainage of CSF
. Figure 142‐9 When a case has been submitted to one or more previous cisternographies and glycerol injections (or other destructive procedures directed onto the Meckelian cave) and suffers from a recurrence of the neuralgia, a new cisternography may show irregularities in the cistern probably due to fibrosis. In (a) the first procedure is illustrated while (b) shows the cisternography preceeding the second treatment
Retrogasserian glycerol injection for trigeminal neuralgia
through the needle confirms an intracisternal tip position. However, if there is no exit of CSF through the cannula, cisternography with positive contrast should always be performed before glycerol injection. Evacuation of the contrast medium from the cistern is facilitated by rinsing with saline, as described above. Because of the cisternal adhesions, images on repeat cisternography may differ somewhat from the initial cisternography films (> Figure 142‐9a, b). This also points to the advantage of placing the needle tip in the part of the cistern housing the root fibers to be treated because the glycerol might become trapped by the fibrosis, preventing it from reaching the bottom of the cistern and the lowermost mandibular branch fibers. In contrast to our experience and the observations of Rappaport and Gomori [26] Lunsford [27] noted no abnormalities of the cistern in patients undergoing repeated glycerol injections. Postoperative management. After glycerol injection, the patient should remain seated upright in bed with his head in the selected position (usually slightly bent forward) for another hour. Failure to keep the head bent forward during the transport from the operation chair to the bed or later during the critical hour may jeopardize the outcome of the whole procedure. Some authors have tried other time periods, but there is no convincing evidence that extending the time of controlled head positioning increases the therapeutic effect. However, there are indications that shortening the period of retroganglionic fiber exposure to glycerol by actively evacuating it from the cistern may cause more extensive and less well controlled rhizolysis [9,24]. On the other hand, Sweet [9,28] reports that if first-branch sensory loss appears, leaving the glycerol in the cistern for 12 min or more may result in persistent corneal anesthesia. Hence, he recommends evacuation of the glycerol from the cistern within 10 min after the onset
142
of first-branch analgesia. However, the general methodology used by Sweet differs considerably from the procedure as described here. The latency for relief from neuralgia varies. Approximately half of the patients report that their paroxysms disappear immediately after the glycerol injection. In the remaining group, there may be occasional spells during the first postoperative day and sometimes for periods of up to 5 days, after which the neuralgia fades away. Even longer latencies of up to 3 weeks have been reported, especially after reinjections [27]. Hospital stay overnight is usually required because of the sedation in elderly or weak patients, or the use of general anesthesia. In younger patients, the procedure may be performed on an outpatient basis. Facial sensibility should be examined at least twice after the procedure, during the first postoperative day and after 3 months. Patients should be encouraged to report increased body temperature during the first postoperative week, ulcers in the vicinity of the mouth, rash, eye problems, or any adverse effects that might be due to the treatment. A telephone report to the surgeon about the treatment effects is due 10 days after surgery. The pharmaceutical regimen for the neuralgia is usually tapered off gradually during the weeks after a successful injection. If high doses of carbamazepine have been used, we usually decrease the daily intake by 100 mg every second day. If several drugs are used the patient should be provided with a written recommendation for gradual decrease of pharmacotherapy after discharge from the hospital.
Results The results from several series are shown with respect to initial and long-term relief from the
2441
2442
142
Retrogasserian glycerol injection for trigeminal neuralgia
paroxysmal pain, rate of recurrence, and various complications such as sensory disturbances recorded, herpes eruptions, and postoperative meningitis of the aseptic and bacterial types. Some of the major series are examined, and relevant data are given in tables and text. It should be noticed, however, that some of the material contain case series published previously and some series published in different format in other journals. Thus there is by necessity some degree of overlap in the total material. Problems with comparison of different series. A major problem when comparing the outcomes of different therapies for tic and also when studying the results from different centers is that there are no randomized controlled trials comparing the different minimally invasive techniques. For example, Lopez et al. [29] identified in total 175 studies on these techniques out of which only 9 could be used to evaluate rates of complete pain relief on an annual basis and 22 to list complications. When focusing on different series of glycerol therapy, a major problem is the variations in technique. Different methods have been used by different authors to determine an intracisternal needle position before glycerol injection. Strategies vary from that recommended – use of cisternography [5,8,27,30], drop-by-drop incremental glycerol injection and intraoperative recording of sensory response [4,6,31] – to the use of electric stimulation in the same way as is done in selective thermocoagulation [16,32]. In some series, cisternography is not used at all [6,31,33] in some only in conjunction with treatment of first-branch neuralgia, and in others only when there is no spontaneous CSF outflow through the cannula [34]. The extent to which cisternography is used in the series reviewed is noted in > Table 142‐1. In our experience, there is a definite relation between the quality of the cisternography, the adequacy of needle placement, and the treatment outcome. This is also supported by a recent publication by Jagia et al. [23].
When judging other series, of course, we have no information about the first two of these parameters. Another important factor is the amount of glycerol used. Sometimes the range of volume is not explicit, and in several series it far exceeds the original recommendations [6,11,35,36]. This has to be taken into account when interpreting the data (ranges of volumes of glycerol used are given in > Table 142‐3). A third factor of major importance for the outcome is whether the patient has had some other destructive procedure before (or after) the glycerol rhizolysis and if information about this is given in the reports. This is of critical importance with regard to the estimation of sensory disturbances after the procedure, and it also has an impact on the decision of the volume of glycerol to be injected. In some series, patients with diagnoses other than classic TN are included without notice. It is often impossible to determine this from the presentation of results. This is of course most important when judging the outcome, because the results, as described later, in other facial pain conditions often are inferior to those in classic tic doloureux. Short-term and late outcomes. The outcomes of 22 major series with long-term follow-up periods encompassing more than 4,000 patients are given in > Table 142‐1. As already pointed out above it should be kept in mind that some of these studies report on partly overlapping case materials. Cisternography was used routinely for all patients in only 16 of the series [5,11,17,19,27,30,34–44]. The percentage of patients enjoying immediate (within 2 weeks) relief from paroxysmal facial pain varies in the different series between 67 and 97%. There is no clear association between the mere use of cisternography and success rate for either immediate or long-term outcomes. This finding may seem remarkable in light of the experiences of the current authors,
Retrogasserian glycerol injection for trigeminal neuralgia
142
. Table 142‐1 Results of glycerol rhizolysisa Authors Ha˚kanson [30] (1983) Lunsford [27] (1985) Arias [4] (1986) Beck et al. [51] (1986) Dieckmann et al. [5] (1987) Saini [33] (1987) Burchiel [17] (1988) Young [6] (1988) Waltz et al. [85] (1989) Fujimaki et al. [11] (1990) North et al. [31] (1990) Ischia et al. [35] (1990) Steiger [34] (1991) Slettebo et al. [37] (1993) Bergenheim and Hariz [38] (1995) Jho and Lunsford [19] (1997) Blomstedt and Bergenheim [40] (2002) Febles et al. [39] (2003) Jagia et al. [23] (2004) Pollock et al. [42] (2005) Henson et al. [43] (2005) Kondziolka and Lunsford [44] (2005)
Number of patients
With cisternography (%)
Pain free after first injection (%)
Total pain free at follow-up (%)
100 100 50 31 100 0 100 Some 100 100 0 100 100 100 100 100 100
96 74 95 67 91 76b 80 90 73 80 >90 92 84 93 97 90 95
75 66 95 72 85 17 53 78 74 26 >50 71 59 50 76 55 50
–
66 71 50 60 60
100 62 100 58 252 550 46 162 200 122 85 112 122 60 99 523 139 30 100 98 79 1,174
100 0 100 100 100
90 73 80 90
a Outcomes of 22 major studies. Number of patients, use of cisternography for ascertaining intracisternal injection, percentage of patients with initial pain relief after first injection, and percentage of patients pain free (including reinjections) at follow-up are given Note the possible overlap of materials! b Most failures previously treated by destructive method; otherwise 96% – Not reported
but it may also point to large differences in the accuracy of the cisternographic procedure. The follow-up periods in these series range from a few months to more than 10 years. The recurrence rates are difficult to estimate correctly because of the differences in techniques of reporting, statistics, and so forth. Kaplan-Meier analysis has, however, been used by several authors [17,31,34,35], which facilitates the interpretation of recurrences. The recurrence rate in relation to length of follow-up is reviewed in > Table 142‐2. Roughly, within 2 years of treatment, between 2 and 50% of patients with a successful initial outcome experience recurrent pain. This large variation obviously casts doubt on both the technical
performance of the procedure and the follow-up methodology. It is difficult to evaluate how many of the patients have been reinjected, and if these cases are included in the final results. The original Stockholm series of 100 patients was followed between 5 and 10 years (mean, 5 years, 4 months). At the last followup, 53% were still pain free after the first injection. Twenty-two patients had at that time been reinjected with glycerol and, in total, 75% of patients from the entire series were pain free, and an additional 23% had only mild pain easily controlled by a pharmaceutical regimen [45]. It deserves to be mentioned that the average volume of glycerol injected in this series was only 0.21 ml.
2443
2444
142
Retrogasserian glycerol injection for trigeminal neuralgia
. Table 142‐2 Pain recurrence after glycerol rhizolysisa Authors Ha˚kanson [30] Lunsford [27] Arias [4] Beck et al. [51] Dieckmann et al. [5] Saini [33] Burchiel [17] Young [6] Waltz et al. [85] Fujimaki et al. [11] North et al. [31] Ischia et al. [35] Steiger [34] Slettebo et al. [37] Bergenheim and Hariz [38] Jho and Lunsford [19] Blomstedt and Bergenheim [40] Febles et al. [39] Jagia et al. [23] Pollock [42] Henson et al. [43] Kondziolka and Lunsford [44]
Number of patients 100 62 100 58 252 550 46 162 200 122 85 112 122 60 99 523 139
Early recurrence (<2 year, %)
Late recurrence (>2 year, %)
26 21 2 11 11 41 47 11 23 45 ~40 20 ~30 27 33 (1 year) 13 35
43 – 10 – 37 92 75 34 25 72 55 26 ~41 55 – – 45
– – 20 25 –
33 – 36 25 –
30 100 98 79 1,174
Range of follow-up 5–10 years 3–28 months 2–3 years 2–40 months 2–5 years 1–6 years 3–44 months 6–67 months 25–64 months 38–54 months 6–54 months 1–5 years 1–96 months 4.5–9 years 12 months 11 years Up to 11 years Median 33.5 months 6–36 months 3–52 months 4–100 months Up to 11 years
a
Percentage of patients in each series with early recurrence (within 2 years) after treatment and late (cumulative) recurrence (after 2 years). Follow-up range is also indicated. Note the possible overlap of materials! – Not reported
The average risk of recurrence within 2 years in well-controlled series appears to be approximately 20%, and the rate of late reappearance of symptoms (within 5–10 years) may approach 50%. Nevertheless, because injection can easily be repeated and carries little risk for the patient, most of the patients (>75%) may be maintained completely pain free by a small number of reinjections (see > Table 142‐1). By comparison, Jannetta [46] found an 80% rate of permanent pain relief in most series of microvascular decompression, 10% with some pain, and a 10% failure rate. Two short reports of major series of patients with TN treated by glycerol rhizolysis have been presented by Spaziante and colleagues [47] and Lunsford and Duma [48]. Because the data have been reported only in congress abstracts, they were not included in this chapter’s tables.
The two series comprise 191 and 480 patients, respectively. The follow-up periods were as long as 10 years. The percentage of patients pain free immediately after treatment was 93% in the series of Spaziante and associates [47]; long-term follow-up showed 77% [47] and 75% [48] of patients were pain free. No patients with anesthesia dolorosa were observed, but mild hypesthesia was found in 46% and 20%, respectively. Both authors conclude that glycerol rhizolysis is a mildly neurodestructive procedure indicated as the first choice [47] in elderly patients, those with MS, and in patients in whom other procedures, including microvascular decompression, have been unsuccessful [48]. A 1997 report of the Pittsburgh series by Jho and Lunsford [19] shows that of their large group of 523 patients, 90% were immediately pain free
Retrogasserian glycerol injection for trigeminal neuralgia
142
. Table 142‐3 Sensory side-effects after glycerol rhizolysisa Authors Ha˚kanson [30] Lunsford [27] Arias [4] Beck et al. [51] Dieckmann et al. [5] Saini [33] Burchiel [17] Young [6] Waltz et al. [85] Fujimaki et al. [11] North et al. [31] Ischia et al. [35] Steiger [34] Slettebo et al. [37] Bergenheim and Hariz [38] Jho and Lunsford [19] Blomstedt and Bergenheim [40] Febles et al. [39] Jagia et al. [23] Pollock [42] Henson et al. [43] Kondziolka and Lunsford [44]
Number of patients 100 62 100 58 252 550 46 162 200 122 85 112 122 60 99 523 139 30 100 98 79 1,174
Volume of glycerol (ml) 0.2–0.3 0.15–0.25b 0.1–0.4 0.2–0.4 0.15–0.4 0.2–0.3 0.15–? 0.15–0.55 0.2–0.6 0.3–0.5 0.3–0.4 0.4–0.5 0.2–0.35 0.15–0.70 0.20–0.35 0.20–0.50 0.13–0.35 –
0.1–0.3 0.18–0.45 ~0.3 –
Slight hypesthesia (%)
Severe hypesthesia (%)
60 21b 13 17 20
0 0 0 2 1 5d 7 12 7 29 2 0
–
72 72 37 63 4 32 53b 35 42 32 47.3
–
3d 6 6 45.5
Dysesthesia (%) 0 3c 0 0 2 11 13 3 2 26 4e 3 13 13 5b 2 22.7
53
13
–
–
–
45 52
–
34
40 8 5
–
–
–
a
Percentage of different types of sensory disturbance in each series. Range of glycerol volumes used is also indicated. In several series, it cannot readily be judged if sensory disturbances recorded after glycerol treatment were already present before the procedure Furthermore, other destructive procedures may have been subsequently used without specific notice b Many with previous or additional destructive procedures c After herpes reactivation d Only cases with previous destructive procedures e Transient – Not reported
and 55% stayed pain free at follow-up (which extended to more than 10 years). The highest figures for patients initially pain free are those of Ha˚kanson [3] (96%) and Bergenheim and Hariz [38] (97%). Side-effects and complications. In general, glycerol rhizolysis in the hands of the experienced and careful neurosurgeon should not result in any serious complications for the patient. Only one fatal outcome has been reported [19,48,49], in which a patient had a fatal myocardial infarction in the recovery room after the
procedure. Another complication reported recently was a cardio-respiratory arrest during the procedure, probably more related to the anesthesiologic conditions [50]. Since glycerol rhizolysis often is used in patients of advanced age (recently a female patient 98 years old was successfully treated in Stockholm) and sometimes with severe disabilities, the number of severe complications is actually astonishingly low. Other serious consequences of attempts to penetrate the foramen ovale with a needle
2445
2446
142
Retrogasserian glycerol injection for trigeminal neuralgia
(e.g., intracranial hemorrhage) have been described by Sweet [10], but these are related to the route of puncture and not to glycerol instillation per se. The most feared consequence of neurolytic procedures, postoperative anesthesia dolorosa, is rarely seen after glycerol rhizolysis. An exception to this is the rather heterogeneous series of 552 cases by Saini [33], in which rhizolysis was performed without cisternography or some other technique to confirm the localization of the needle tip before glycerol injection. In this series, the number of patients with anesthesia dolorosa amounted to an extreme 26 (5%). Furthermore, 16 patients (3%) had signs of disturbance of third-branch motor function. This malfunction, however, resolved within 3–4 months after injection. Transient masseter weakness has also been described by others [35,36]. Otherwise, only single cases with transitory cranial nerve dysfunction, after glycerol rhizolysis have been reported. Some of these are reviewed by Sweet [10]. A review of 260 consecutive procedures carried out in 139 patients between 1986 and 1987 reports a fairly high frequency of intraoperative side-effects [40]. In this series there occurred both a variety of intraoperative technical obstacles, severe vaso-vagal responses and even cardiac arrest. In total complications or side-effects occurred in 67.3% of procedures. The side effects and complications are detailed in the > Table 142‐3 and > Table 142‐4. This series evidently encompasses learning curves for several surgeons. Penetration of the foramina of Versalius (0.8%), spinosum (0.4%) occurred as well as buccal penetration (1.5%). Actually, three patients (1.2%) reported a postoperative decrease of hearing ability ipsilateral to the surgery. Arterial bleeding occurred in 0.4% while venous blood from the needle egressed in 3.5%. This type of honest and careful reports of technical difficulties and late problems are most important since such publications are rare.
Post-injection disturbance of facial sensibility. A disturbance of facial sensibility is not uncommon after the procedure, but usually lasts only a limited time – a few hours, days, or 1–2 weeks. This is not tabulated as a side effect. The following discussion focuses on more persistent alterations in sensory function or perception in the trigeminal area: (1) postoperative hypesthesia, slight or severe; and (2) the presence of dysesthesia or allodynia. Paresthesias that are not unpleasant are not included. Hypesthesia. As already stated, serious sensory disturbance (e.g., anesthesia dolorosa) is rare after glycerol rhizolysis, except in certain series (see earlier). However, transitory facial hypesthesia is a rather common phenomenon (up to 70%) after the procedure. Usually the complaints vanish during a 3-month period after the operation [45], and most of the rest within 6 months [6,36]. The frequency of slight hypesthesia persisting for longer periods is quite variable between different series, as can be seen from > Table 142‐3 In some series, only a small percentage of patients experience this [31] but in others more than two thirds present with hypesthesia at follow-up [6,17]. More severe hypesthesia and anesthesia fortunately is rare (see > Table 142‐3). There is only one series with a figure approaching 30% [11]. This is of course unacceptable, and the authors consequently abandoned the procedure. The figures from the series of Saini [33], Burchiel [17], Young [6], Waltz and associates [36], Bergenheim and Hariz [38] and Jho and Lunsford [19] also must be considered too high (5–12%). There are at least four possible explanations for this outcome: (1) some of these patients may have had other previous or subsequent destructive procedures; (2) there may have been technical difficulties during the procedure, with a suboptimal final needle position; (3) the volume of glycerol injected may have been too large; and (4) previous procedures may have produced
Retrogasserian glycerol injection for trigeminal neuralgia
142
. Table 142‐4 Infectious complications after glycerol rhizolysisa Authors Ha˚kanson [30] Lunsford [27] Arias [4] Beck et al. [51] Dieckmann et al. [5] Saini [33] Burchiel [17] Young [6] Waltz et al. [85] Fujimaki et al. [11] North et al. [31] Ischia et al. [35] Steiger [34] Slettebo et al. [37] Bergenheim and Hariz [38] Jho and Lunsford [19] Blomstedt and Bergenheim [40] Febles et al. [39] Jagia et al. [23] Pollock [42] Henson et al. [43] Kondziolka and Lunsford [44]
Number of patients
Herpes reactivation (%)
100 62 100 58 252 550 46 162 200 122 85 112 122 60 99 523 139 30 100 98 79 1,174
50 13 10 9 77 3 5 38 – – – – – – – 37 3.8 – 3.6 12 – ‘‘Common’’
Aseptic meningitis (%)
Bacterial meningitis (%)
0 3 2 4 1 – 2 0 7 – – – 0 1.6 – 0.6 1.5
0 0 0 0 1 – 2 1 1 – – – 0 – – – 1.5
– 0 1 –b 0.2
– 0.7 0 8b –
a
Percentage of cases in each series with herpes simplex reactivation, aseptic meningitis, and bacterial meningitis Three patients reported with meningitis, bacterial or aseptic not specified – Not reported b
cistern obliteration by fibrosis [19,26,27]. Thus, Sweet [10] describes a case in which 1.5 ml (SIC!) of glycerol was injected in 0.1-ml increments, not unexpectedly, resulting in anesthesia dolorosa. When properly performed, the incidence of severe sensory disturbances with glycerol rhizolysis should be low (<1%) [5,27,30,45,48]. Various kinds of dysesthesias. Dysesthesia, spontaneous or touch-evoked (allodynia) unpleasant sensations occur rarely and mostly transiently after glycerol rhizolysis. In the series reviewed here, the incidence in general remains between 0 and 4% (see > Table 142‐3). The results of Fujimaki and associates [11], Burchiel [17], Steiger [34], Saini [33] and Slettebo and colleagues [37] are extreme in this respect, with frequencies between 11 and 26%. There is one
prime suspect major cause of this side effect: a previous neurodestructive procedure with a sensory disturbance not specifically recorded before the glycerol rhizolysis. Some patients may also have had procedure-induced herpes eruptions resulting in this condition [27], although this has not been observed by us. Whatever the reason, these figures [17,33,34] are unacceptably high. Normally, dysesthesias should be present in at most 2% of previously untreated patients with TN treated by glycerol rhizolysis [4,5,19,27,30,36,51] and in this respect it compares well with other percutaneous methods. Infectious manifestations. Serious infectious complications after retrogasserian glycerol rhizolysis are rare. However, reactivation of latent viral infections of neural tissue (notably herpes
2447
2448
142
Retrogasserian glycerol injection for trigeminal neuralgia
simplex type 1) seems more common than might be expected. The infectious complications recorded in the aforementioned 22 major series are reviewed in > Table 142‐4. Herpes simplex activation. The activation of latent herpes simplex infections is a phenomenon encountered intermittently in neurologic surgery [52]. However, the viral eruption may be minor and thus easily overlooked by both the patient and physician. > Table 142‐4 indicates that the incidence of herpes simplex activation after glycerol injection varies considerably among different series. The range of 3–77% for postoperative viral infections probably largely reflects variations in accuracy of follow-up. It seems safe to state that perioral and gingival ulcers presenting during the first week after treatment are rather common, but more serious outbreaks are rare. The mouth ulcers usually require no specific therapy. Non-bacterial meningeal irritation. Symptoms of meningitis such as high fever, nuchal rigidity, and CSF pleocytosis presenting 24–36 h after the procedure may indicate a meningeal reaction to some agent introduced into the trigeminal cistern. CSF cultures usually are negative, which is why the condition is labeled aseptic meningitis. As a precaution, patients presenting with these symptoms are often treated by highdose intravenous antibiotics. When negative cultures are reported, the antibiotics are discontinued. These patients usually recover completely within a few days without any sequelae. If the neurosurgeon is convinced that the reaction is of the ‘‘aseptic meningismus’’ type, he may prescribe steroids, which can shorten the course of the condition [27]. Patients who have demonstrated such adverse reaction after a previous treatment have undergone repeated injections with steroid prophylaxis without any problems. Aseptic meningitis appears with a frequency varying between 0 and 7% (see > Table 142‐4). The etiology of this reaction is obscure. Contrast agents, the glycerol, tantalum dust [27] and
the extent of manipulation or number of ganglion penetrations have been suspected. Some authors report diminished frequency of reactions after reducing the amount of contrast medium [36]. The Stockholm experience is that the frequency was approximately 2% during the metrizamide era. After the change of contrast medium to iohexol in January 1986, we have had very few cases of aseptic meningitis. The plausibility of contrast medium as the causative agent is also supported by the observations by Arias [4] who presented a series of 100 patients of whom approximately 50% were treated without preliminary contrast medium injection. In the group subjected to contrast injection, two patients presented with aseptic meningitis, whereas no case was found in the other group. However, aseptic meningeal reactions are observed also after other manipulations of the trigeminal root where no contrast is used [46]. Bacterial meningitis. Bacterial meningitis after percutaneous puncture of Meckel’s cave is rare, regardless of the intended therapeutic procedure [45,53]. The most common bacterial agents are those typically colonizing the upper respiratory tract; infections by agents of low virulence also have been described [54]. The most likely cause seems to be inadvertent (and often unrecognized) penetration of the oral cavity mucosa. The frequency of bacterial meningitis reported after glycerol rhizolysis varies between 0 and 2% (see > Table 142‐4). The frequency in the Stockholm series remains steady at approximately 0.5%, and seems related to the extent of manipulation during the procedure. All patients have recovered without serious sequelae after adequate treatment with antibiotics.
Mechanisms of Glycerol in Trigeminal Rhizolysis Retrogasserian glycerol rhizolysis is a purely empirical method, the beneficial effects of glycerol
Retrogasserian glycerol injection for trigeminal neuralgia
having been discovered accidentally. There has been much debate about the putative mechanisms behind the effects of glycerol on paroxysmal pain. Glycerol is a trivalent alcohol normally present in human tissue, where it forms the skeleton of the triglycerides, among other functions [55,56]. Glycerol readily penetrates cell membranes and seems to possess distinct cryoprotective properties beneficial to cells. Its toxicity seems to be low, and comparatively high doses must be injected systemically or intrathecally to induce toxic effects [57,58]. To the clinician it is evident from the side effects (e.g., hypesthesia) that the substance is neurolytic in the concentration and dose used for injection. An important issue is whether the neurolytic effect is selective for a certain fiber spectrum. From clinical observations, it is clear that the trigger mechanism for the pain paroxysms is activated by tactile stimulation and impulse propagation in large myelinated fibers. Histological data. The neurolytic action of glycerol is considered to be due to its hypertonicity, a condition well known to injure nerve fibers, especially thin, unmyelinated as well as myelinated fibers [59]. Although the myelin sheath of the coarse fibers gives some transitory protection from this effect, length of exposure, neuron type, and the presence of previous demyelination may be important determinants of the vulnerability of individual fibers. For example, with longer exposures, Robertson [60] and Pal and colleagues [61] observed that myelinated fibers were particularly vulnerable, and the degree of damage positively correlated with fiber diameter. Studies on isolated animal nerve fibers showed morphologic changes after exposure to glycerol. These consisted of disruption of the tight junction between the Schwann cells and the axolemma, but without damage to the axon proper [62]. Thus, marked structural changes were observed with glycerol administration but, curiously enough, the conduction properties of the treated nerve axons remained intact [62].
142
After intraneural and perineural injection of glycerol, Ha˚kanson [63] and Rengachary and associates [64] observed axolysis with marked myelin sheath swelling. The coarse myelinated fibers sustained the most severe damage, whereas the small-diameter myelinated and unmyelinated fibers were relatively well preserved [61]. In contrast, Bremerich and Reisert [65] found only slight histomorphologic changes after glycerol injection in the region of the oval foramen of rats in a long-term (180 days) comparative study of axonal damage after injection of glycerol, phenol-glycerol, and saline. A more recent study in dogs submitted to glycerol injection into one trigeminal ganglion [66] demonstrated axonolysis both in myelinated as well as in nonmyelinated fibers. The damage following glycerol injection into a cavity with isotonic body fluid is probably considerably less severe. However, Lunsford and associates [7] observed extensive areas of myelin degradation and axonal swelling in cats subjected to retrogasserian glycerol injections 4–6 weeks earlier. The actual site of glycerol effects has been studied by Stajcic [67] who injected 3H-labeled glycerol into peripheral branches of the maxillary nerve and in the infraorbital canal of rats. The amount of radioactivity detected in the nerve distal to the foramen rotundum, as well as in the ipsilateral and contralateral gasserian ganglion, was less than 0.1% in all specimens. The conclusion was that a retrograde transport mechanism behind the effect is improbable and that the beneficial effect of glycerol occurs at the site of injection. To the best of our knowledge, there is as yet no publication of an autopsy series of patients with TN treated by retrogasserian glycerol rhizolysis. Sweet [10] provides an anecdotal description, mentioned earlier, of a patient undergoing a retrogasserian glycerol injection of the extreme volume of 1.5 ml, with subsequent development of anesthesia dolorosa. At a posterior fossa craniotomy
2449
2450
142
Retrogasserian glycerol injection for trigeminal neuralgia
‘‘many months’’ later, the trigeminal rootlets were markedly atrophic. Neurophysiological data. One hypothesis is that the damage to nerve axons is caused by the change in osmolarity in the surrounding body fluids, and the morphologic changes seem to be minimized by a gradual alteration in the osmolarity (e.g., by slowly instilling and removing glycerol from the compartment housing the axons). The functional consequences of glycerol application to normal and damaged nervous tissue are known only fragmentarily, but there are some few observations that might relate to the clinical use of the substance. Burchiel and Russel [68] studied the effect of glycerol on normal and damaged nerves in a rat neuroma model. The neuromas, produced by sectioning of the saphenous nerve, were mechano-sensitive and discharged both spontaneously and in response to light manipulation. These authors found evidence supporting the view that glycerol exerts its major action on the large-diameter fibers because exposure of the injured nerve to glycerol induced a short episode of increased spontaneous firing in the nerve, a response earlier shown to originate from the myelinated fibers. The observation by Rappaport and associates [69] that glycerol injected into neuromas was more effective than alcohol in decreasing autotomy in rats suggests that autotomy may be related to the presence of unpleasant ‘‘tic-like’’ paresthesias. The therapeutic mechanism, according to the authors, could be suppression of the ectopic impulse barrage from the neuroma. Sweet and coworkers [32] found that glycerol injected into the trigeminal cistern of patients abolished the late components (corresponding to A-delta and C-fibers) of trigeminal root potentials recorded with electric stimulation of the surface of the cheek. These recordings were made only minutes after the injection, and therefore do not permit conclusions concerning longterm effects.
Hellstrand and colleagues (unpublished data; see Ha˚kanson [63]) studied the effects of glycerol both on the isolated frog sciatic nerve and on trigeminal root fibers after cisternal injection in the cat. They observed a severe reduction of the evoked potentials with glycerol but a nearly total restoration after rinsing the compartment with saline. This recoverability probably relates to the clinical effects and must be taken into account when interpreting the short-term observations of Sweet and associates [32] referred to above. Evacuation of the glycerol from the cistern after a short time (e.g., 5–20 min 10, 13, 24) might, based on the knowledge that glycerol requires at least 30 min to equilibrate across a membrane of a living cell and according to the aforementioned experimental observations, induce more severe damage, especially to fine fiber systems, than a slow unloading by diffusion into the subarachnoid space. Longer-term observations of trigeminal evoked potentials have been reported by Bennett and Lunsford [70] who investigated patients before and 6 weeks after trigeminal glycerol rhizolysis. They confirmed the earlier findings of Bennett and Jannetta [71] that thresholds were elevated and evoked potentials had a markedly increased latency on the affected side compared with the healthy one. An additional, unexpected finding was that these aberrations were ‘‘normalized’’ after glycerol rhizolysis. Because partially demyelinated fibers are known to conduct with a slower velocity and at a lower rate [72,73] they interpreted this finding to indicate that glycerol selectively attacked partially damaged trigeminal axons and, after their elimination, the evoked trigeminal potentials appeared ‘‘normalized.’’ Further long-term observations were supplied by Lunsford and colleagues [7] who noted the clearest changes in trigeminal evoked potentials in cats in the large-diameter myelinated fibers, with additional changes noted as late as 6 weeks after the injection.
Retrogasserian glycerol injection for trigeminal neuralgia
Quantitative sensory testing using von Frey hairs, mechanical pulses, and the Marstock technique [74] also corroborates the notion that glycerol acts mainly on the large myelinated fiber spectrum [45]. Eide and Stubhaug [75] examined thresholds for tactile and temperature stimuli in patients with TN before and after glycerol rhizolysis. They found evidence that pain relief after glycerol treatment involved normalization of previously abnormal temporal summation phenomena with little accompanying sensory loss. Kumar and associates [76] found postinjection quantitative abnormalities of the blink reflex that correlated with sensory impairment. Thus, experimental and clinical observations indicate that the effects of glycerol may be due to its hyperosmolarity and that the rate of alteration of osmolarity is critical for the effect. Furthermore, there are indications that the effect is exerted through actions mainly on the large diameter myelinated fibers, notably those with previous damage to the myelin sheath, thereby possibly affecting the ‘‘trigger mechanism’’ for pain paroxysm. Glycerol also may downregulate central neuronal hyperexcitability, often without signs of significant additional nerve damage [75] (see above ignition hypothesis [13]).
Glycerol Rhizolysis in Other Types of Facial Pain Glycerol therapy for tic in MS patients. The paroxysmal facial pain sometimes seen in patients with MS is clinically identical to classic TN. Therefore, all methods used for symptomatic treatment of TN, except microvascular decompression, also may be used in MS. It has been estimated that TN is present in approximately 1–2% of patients with an established diagnosis of MS [77–79]. In different series of TN, the prevalence of MS seems to vary considerably. The prevalence most often ranges from 2 to 4%, but incidences of up
142
to 8% have been reported [80,81]. Furthermore, the occurrence of bilateral facial pain is not uncommon in MS, but is extremely rare in idiopathic TN. The initial success rate seems approximately equal for idiopathic TN and for TN in MS, whichever therapeutic method is used [81] but the late results differ considerably between the two. The recurrence rate is much higher in MS, possibly because the progressive demyelinating disease produces multiple lesions in the trigeminal system. Dieckmann and colleagues [5] and Linderoth and Ha˚kanson [81] reported a recurrence rate of more than 40% at a 2-year follow-up; the recurrence rate in classic TN was approximately 11% during the same period in Dieckmann and colleagues’ series. After longer follow-up periods (8–79 months after injection), Linderoth and Ha˚kanson [81] reported over 60% recurrent tics in the MS group, whereas the recurrence rate for the entire TN group (approximately 300 patients) was 38%. Elevated recurrence rates after treatment of TN in MS have also been observed with selective thermocoagulation. Broggi and Franzini [82] reported a 40% recurrence rate among patients with MS after thermorhizotomy, compared with 9% in a previous series of patients without MS. However, this finding could not be corroborated by Brisman [83] who found no significant difference in recurrence rates between 16 MS cases and 219 non-MS cases. The low tolerance of patients with MS to pharmacologic agents, both in form of carbamazepine and as anesthesia and sedation, merits specific comment. Linderoth and Ha˚kanson [81] reported that more than 90% of their patients with MS and TN complained of adverse medication effects. The symptoms in the MS group were more often an aggravation of pre-existing symptoms than the side effects usually mentioned. Thirty-eight percent of the patients with MS had to discontinue carbamazepine because of such symptoms. After glycerol treatment, 82%
2451
2452
142
Retrogasserian glycerol injection for trigeminal neuralgia
of the patients previously on carbamazepine were able to terminate it. The views on MS and glycerol therapy expressed above have more recently gained support from a study by Pickett et al. [84] who assessed 53 MS patients submitted to in total 97 glycerol procedures. Initial pain relief was obtained in 78% but after 24 months the recurrence rate was 60%. Given the low tolerance of MS patients to the medications used for tic, we therefore recommend that paroxysmal facial pain in MS should be more liberally treated by glycerol rhizolysis to save the patient from the incapacitating side effects of a drug regimen – especially if carbamazepine is used. Patients with MS are often disabled, weak, and considered at risk from general anesthesia; a method that does not require deep general anesthesia is therefore preferable. Glycerol therapy in other facial pain syndromes. In painful trigeminal neuropathy (previously sometimes called ‘‘atypical facial pain’’), a paroxysmal component may exist in addition to the well-known continuous neuropathic pain. The etiology varies (e.g., trauma, infection, postoperative, tumor, idiopathic) but includes previous treatment for TN. We agree with the observations of Dieckmann and colleagues [8], Waltz and associates [85], Lunsford and Apfelbaum [49], Rappaport [86] and Rappaport and Gomori [26] that glycerol rhizolysis is contraindicated in most cases and actually may aggravate a pre-existing neuropathy. The sole exception is when the paroxysmal component completely dominates, diminishing the quality of life for the patient, and where sensory disturbance is minimal. The diagnosis of classic TN may have been considered earlier. In such a case, a small amount of glycerol could be injected (0.15 ml). If complete pain relief is obtained (duration at least 3 months), a second injection may be performed using a somewhat larger glycerol volume.
Severe, intractable cluster headache (migraneous neuralgia; Horton’s syndrome) has also been treated by glycerol injection into the trigeminal cistern on the affected side [31,85,87] (G. Sundba¨rj, personal communication, 1988). In general, the outcome has been partial and transitory relief from symptoms in some of the patients, comparable with that obtained by other manipulations of the trigeminal system. However, one long-term follow-up study of 18 patients followed during in average 5.2 years demonstrates that 83% of the cases obtained immediate relief, the frequency of attacks decreased markedly, and only in 39% the condition recurred during the study period [87]. Thus, glycerol therapy in cluster headache may offer some relief in chronic cases of moderate severity, but in cases with extreme pain, periorbital edema etc., glycerol treatment injection does not provide symptom alleviation. Instead, chronic hypothalamic DBS has in recent years proven effective in selected cases [88–91]. Occipital nerve stimulation has also been used in the treatment of cluster headache [92]. Glycerol rhizotomy has, furthermore, been reported to provide relief from the paroxysmal pain component in selected cases of SUNCT [93,94] although also negative outcomes of surgical therapies in this condition have been published [95]. More recently a successful case of hypothalamic DBS for SUNCT has been reported by the Milano group [96].
The Place of Retrogasserian Glycerol Rhizolysis in the Treatment of TIC Glycerol rhizolysis is an inexpensive method that requires little special equipment. Although the method seems simple, the procedure must be meticulously performed by a surgeon with experience in the operation. In a comparison between glycerol injection, MVD and stereotactic
Retrogasserian glycerol injection for trigeminal neuralgia
radiosurgery for tic Pollock and Ecker [41] reported glycerol rhizolysis to be the most costeffective with 6,342 USD per quality adjusted painfree year although the difference to stereotactic radiosurgery (8,174 USD) proved not to be statistically significant. The major indication remains typical idiopathic TN, particularly in those who are elderly, weak, or have MS. If a young patient chooses to undergo glycerol rhizolysis in place of a microvascular decompression, he or she may do so. However, if this patient has a relatively rapid recurrence, he or she should be urged to accept the open procedure instead. In most cases, it should be possible to affect all the trigeminal branches with some selectivity; the most problematic might be the mandibular branch. With careful technique, using the suggestions given above, it should be possible to perform rhizolysis in that branch as well. In some patients submitted to several reinjections, adhesions inside the cistern may impede the glycerol from reaching the lowermost fibers. In such cases, selective thermocoagulation – or trigeminal nerve compression with a balloon – may be considered. In our opinion, microvascular decompression in young patients without MS and glycerol treatment in elderly or weak patients are the methods of choice in debutant cases of classic TN. At recurrence the repeat surgery might be different dependant on the branches involved.
Conclusions Trigeminal glycerol rhizolysis should be offered to healthy patients in their eighth decade or older with classic TN, as well as to somatically fragile patients and those with MS. These views are also supported by many previous writers in the field (e.g., Pollock and Ecker [41]; Henson et al. [43]; Spatz et al. [97]) especially when the patients suffer from extremely severe spells of pain, have difficulties drinking or eating – or even to
142
talk – and thus need prompt treatment. It also is an option in younger patients reluctant to undergo major surgery. With meticulous technique, the method is well tolerated by patients, can be performed with the patient in light sedation only, and carries little risk for severe side effects. The recurrence rate, although usually slightly higher than that with microvascular decompression, compares well with other percutaneous methods, particularly in view of the low risk of severe postoperative sensory disturbance.
Acknowledgements We are much indebted to Dr Sten Ha˚kanson MD, PhD., the originator of glycerol treatment for trigeminal neuralgia, who collaborated with the first author in earlier versions of this presentation.
References 1. Leksell L. [Trigeminal neuralgia. Some neurophysiologic aspects and a new method of therapy]. Lakartidningen 1971;68(45):5145-8. 2. Ha˚kanson S, Leksell L. Stereotactic radiosurgery in trigeminal neuralgia. In: Pauser G, Gerstenbrand F, Gross D, editors. Geschichtschmerz: Schmerzstudien 2. New York: Gustav Fischer Verlag; 1979. 3. Ha˚kanson S. Trigeminal neuralgia treated by the injection of glycerol into the trigeminal cistern. Neurosurgery 1981;9(6):638-46. 4. Arias MJ. Percutaneous retrogasserian glycerol rhizotomy for trigeminal neuralgia. A prospective study of 100 cases. J Neurosurg 1986;65(1):32-6. 5. Dieckmann G, Bockermann V, Heyer C, Henning J, Roesen M. Five-and-a-half years’ experience with percutaneous retrogasserian glycerol rhizotomy in treatment of trigeminal neuralgia. Appl Neurophysiol 1987;50 (1–6):401-13. 6. Young RF. Glycerol rhizolysis for treatment of trigeminal neuralgia. J Neurosurg 1988;69(1):39-45. 7. Lunsford LD, Bennett MH, Martinez AJ. Experimental trigeminal glycerol injection. Electrophysiologic and morphologic effects. Arch Neurol 1985;42(2):146-9. 8. Dieckmann G, Veras G, Sogabe K. Retrogasserian glycerol injection or percutaneous stimulation in the treatment of typical and atypical trigeminal pain. Neurol Res 1987;9 (1):48-9.
2453
2454
142
Retrogasserian glycerol injection for trigeminal neuralgia
9. Sweet WH. Glycerol rhizotomy. In: Youmans J, editor. Neurological surgery, 3rd ed. Philadelphia: WB Saunders; 1990. p. 3908-3921. 10. Sweet WH. Faciocephalic pain. In: Apuzzo MLJ, editor. Brain surgery: complication avoidance and management. New York: Churchill Livingstone; 1993. p. 2053-2083. 11. Fujimaki T, Fukushima T, Miyazaki S. Percutaneous retrogasserian glycerol injection in the management of trigeminal neuralgia: long-term follow-up results. J Neurosurg 1990;73(2):212-6. 12. Devor M, Govrin-Lippmann R, Rappaport ZH. Mechanism of trigeminal neuralgia: an ultrastructural analysis of trigeminal root specimens obtained during microvascular decompression surgery. J Neurosurg 2002;96 (3):532-43. 13. Devor M, Amir R, Rappaport ZH. Pathophysiology of trigeminal neuralgia: the ignition hypothesis. Clin J Pain 2002;18(1):4-13. 14. Rappaport ZH, Devor M. Trigeminal neuralgia: the role of self-sustaining discharge in the trigeminal ganglion. Pain 1994;56(2):127-38. 15. Devor M. Chronic pain in the aged: possible relation between neurogenesis, involution and pathophysiology in adult sensory ganglia. J Basic Clin Physiol Pharmacol 1991;2(1–2):1-15. 16. Sweet WH, Poletti CE. Problems with retrogasserian glycerol in the treatment of trigeminal neuralgia. Appl Neurophysiol 1985;48(1–6):252-7. 17. Burchiel KJ. Percutaneous retrogasserian glycerol rhizolysis in the management of trigeminal neuralgia. J Neurosurg 1988;69(3):361-6. 18. Ha¨rtel F. Die Leitungsana¨sthesie und Injectionsbehandlung des Ganglion Gasseri und der Trigeminussta¨mme. Arch Klin Chir 1912;100:193-292. 19. Jho HD, Lunsford LD. Percutaneous retrogasserian glycerol rhizotomy. Current technique and results. Neurosurg Clin N Am 1997;8(1):63-74. 20. Nugent GR, Berry B. Trigeminal neuralgia treated by differential percutaneous radiofrequency coagulation of the Gasserian ganglion. J Neurosurg 1974;40 (4):517-23. 21. Ha˚kanson S. Transoval trigeminal cisternography. Surg Neurol 1978;10(2):137-44. 22. Shaw DD, Bach-Gansmo T, Dahlstrom K. Iohexol: summary of North American and European clinical trials in adult lumbar, thoracic, and cervical myelography with a new nonionic contrast medium. Invest Radiol 1985;20 (1 Suppl):S44-50. 23. Jagia M, Bithal PK, Dash HH, Prabhakar H, Chaturvedi A, Chouhan RS. Effect of cerebrospinal fluid return on success rate of percutaneous retrogasserian glycerol rhizotomy. Reg Anesth Pain Med 2004;29(6):592-5. 24. Bergenheim AT, Hariz MI, Laitinen LV. Selectivity of retrogasserian glycerol rhizotomy in the treatment of trigeminal neuralgia. Stereotact Funct Neurosurg 1991;56 (3):159-65.
25. Bergenheim AT, Hariz MI, Laitinen LV. Retrogasserian glycerol rhizotomy and its selectivity in the treatment of trigeminal neuralgia. Acta Neurochir Suppl (Wien) 1993;58:174-7. 26. Rappaport ZH, Gomori JM. Recurrent trigeminal cistern glycerol injections for tic douloureux. Acta Neurochir (Wien) 1988;90(1–2):31-4. 27. Lunsford LD. Trigeminal neuralgia: Treatment by glycerol rhizotomy. In: Wilkins RH, Rengachary SS, editors. Neurosurgery. New York: McGraw-Hill; 1985. p. 2351-2356. 28. Sweet WH. Retrogasserian glycerol injection as treatment for trigeminal neuralgia. In: Schmidek HH, Sweet WH, editors. Operative Neurosurgical Techniques. 2nd ed. New York: Grune & Stratton; 1988. p. 1129-1137. 29. Lopez BC, Hamlyn PJ, Zakrzewska JM. Systematic review of ablative neurosurgical techniques for the treatment of trigeminal neuralgia. Neurosurgery 2004;54 (4):973-82; discussion 982‐3. 30. Ha˚kanson S. Retrogasserian glycerol injection as treatment of tic douloureux. Adv Pain Res Ther 1983;5:927-933. 31. North RB, Kidd DH, Piantadosi S, Carson BS. Percutaneous retrogasserian glycerol rhizotomy. Predictors of success and failure in treatment of trigeminal neuralgia. J Neurosurg 1990;72(6):851-6. 32. Sweet WH, Poletti CE, Macon JB. Treatment of trigeminal neuralgia and other facial pains by retrogasserian injection of glycerol. Neurosurgery 1981;9(6):647-53. 33. Saini SS. Retrogasserian anhydrous glycerol injection therapy in trigeminal neuralgia: observations in 552 patients. J Neurol Neurosurg Psychiatry 1987;50(11):1536-8. 34. Steiger HJ. Prognostic factors in the treatment of trigeminal neuralgia. Analysis of a differential therapeutic approach. Acta Neurochir (Wien) 1991;113(1–2):11-7. 35. Ischia S, Luzzani A, Polati E. Retrogasserian glycerol injection: a retrospective study of 112 patients. Clin J Pain 1990;6(4):291-6. 36. Waltz TA, Dalessio DJ, Ott KH, Copeland B, Abbott G. Trigeminal cistern glycerol injections for facial pain. Headache 1985;25(7):354-7. 37. Slettebo H, Hirschberg H, Lindegaard KF. Long-term results after percutaneous retrogasserian glycerol rhizotomy in patients with trigeminal neuralgia. Acta Neurochir (Wien) 1993;122(3–4):231-5. 38. Bergenheim AT, Hariz MI. Influence of previous treatment on outcome after glycerol rhizotomy for trigeminal neuralgia. Neurosurgery 1995;36(2):303-9; discussion 309‐10. 39. Febles C, Werner-Wasik M, Rosenwasser RH, Goldman WH, Curran WJ, Andrews DW. A comparison of treatment outcomes with Gamma Knife radiosurgery versus Glycerol Rhizotomy in the management of trigeminal neuralgia. Int J Radiat Oncol Biol Phys 2003:S253. 40. Blomstedt PC, Bergenheim AT. Technical difficulties and perioperative complications of retrogasserian glycerol rhizotomy for trigeminal neuralgia. Stereotact Funct Neurosurg 2002;79(3–4):168-81.
Retrogasserian glycerol injection for trigeminal neuralgia
41. Pollock BE, Ecker RD. A prospective cost-effectiveness study of trigeminal neuralgia surgery. Clin J Pain 2005;21(4):317-22. 42. Pollock BE. Percutaneous retrogasserian glycerol rhizotomy for patients with idiopathic trigeminal neuralgia: a prospective analysis of factors related to pain relief. J Neurosurg 2005;102(2):223-8. 43. Henson CF, Goldman HW, Rosenwasser RH, Downes MB, Bednarz G, Pequignot EC, et al. Glycerol rhizotomy versus gamma knife radiosurgery for the treatment of trigeminal neuralgia: an analysis of patients treated at one institution. Int J Radiat Oncol Biol Phys 2005;63(1):82-90. 44. Kondziolka D, Lunsford LD. Percutaneous retrogasserian glycerol rhizotomy for trigeminal neuralgia: technique and expectations. Neurosurg Focus 2005;18(5):E7. 45. Ha˚kanson S. Surgical treatment: retrogasserian glycerol injection. In: Fromm GH, Sessle BJ, editors. Trigeminal neuralgia: current concepts regarding pathogenesis and treatment. Boston: Butterworth-Heinemann; 1991. p. 185-204. 46. Jannetta PJ. Microvascular decompression. In: Fromm GH, Sessle BJ, editors. Current concepts regarding pathogenesis and treatment. Boston: Butterworth-Heinemann; 1991. p. 145-157. 47. Spaziante R, Cappabianca P, Graziussi G, Cacace G, Peca C, de Divitiis E. Percutaneous retrogasserian glycerol rhizolysis for treatment of trigeminal neuralgia. Results in 191 patients (abstract). Acta Neurochir (Wien) 1992;117:97. 48. Lunsford LD, Duma CH. Percutaneous retrogasserian glycerol rhizotomy:. a ten-year-experience (abstract)Acta Neurochir (Wien) 1992;117:97. 49. Lunsford LD, Apfelbaum RI. Choice of surgical therapeutic modalities for treatment of trigeminal neuralgia: microvascular decompression, percutaneous retrogasserian thermal, or glycerol rhizotomy. Clin Neurosurg 1985;32:319-33. 50. Rath GP, Dash HH, Prabhakar H, Pandia MP. Cardiorespiratory arrest during trigeminal rhizolysis. Anaesthesia 2007;62(9):971-2. 51. Beck DW, Olson JJ, Urig EJ. Percutaneous retrogasserian glycerol rhizotomy for treatment of trigeminal neuralgia. J Neurosurg 1986;65(1):28-31. 52. Nabors MW, Francis CK, Kobrine AI. Reactivation of herpesvirus in neurosurgical patients. Neurosurgery 1986;19(4):599-603. 53. Nugent GR. Surgical treatment: Radiofrequency gangliolysis and rhizotomy. In: Fromm GH, Sessle BJ, editors. Trigeminal neuralgia: current concepts regarding pathogenesis and treatment. Boston: Butterworth-Heinemann; 1991. p. 159-184. 54. Aspevall O, Hillebrant E, Linderoth B, Rylander M. Meningitis due to Gemella haemolysans. after neurosurgical treatment of trigeminal neuralgia. Scand J Infect Dis 1991;23(4):503-5. 55. Bunge RP. Myelin degeneration in tissue culture. Neurosci Res Prog Bull 1971;9(4):496-8.
142
56. Dulhunty AF, Gage PW. Differential effects of glycerol treatment on membrane capacity and excitationcontraction coupling in toad sartorius fibres. J Physiol 1973;234(2):373-408. 57. Diechmann D. Glycerol-effects upon rabbits and rats. Indust Med 1941;2:5-6. 58. Baxter DW, Schacherl U. Experimental studies on the morphological changes produced by intrathecal phenol. Can Med Assoc J 1962;86:1200-5. 59. King JS, Jewett DL, Sundberg HR. Differential blockade of cat dorsal root C fibers by various chloride solutions. J Neurosurg 1972;36(5):569-83. 60. Robertson JD. Structural alterations in nerve fibers produced by hypotonic and hypertonic solutions. J Biophys Biochem Cytol 1958;4(4):349-64. 61. Pal HK, Dinda AK, Roy S, Banerji AK. Acute effect of anhydrous glycerol on peripheral nerve: an experimental study. Br J Neurosurg 1989;3(4):463-9. 62. Freeman AR, Reuben JP, Brandt PW, Grundfest H. Osmometrically determined characteristics of the cell membrane of squid and lobster giant axons. J Gen Physiol 1966;50(2):423-45. 63. Ha˚kanson S. Trigeminal neuralgia treated by retrogasserian injection of glycerol. Published dissertation. Stockholm: Karolinska Institute; 1982. 64. Rengachary SS, Watanabe IS, Singer P, Bopp WJ. Effect of glycerol on peripheral nerve: an experimental study. Neurosurgery 1983;13(6):681-8. 65. Bremerich A, Reisert I. [Perineural application of glycerol and phenol-glycerol. Histomorphologic-morphometric study]. Dtsch Zahnarztl Z 1991;46(12):825-7. 66. Isik N, Pamir MN, Benli K, Erbengi A, Erbengi T, Ruacan S. Experimental trigeminal glycerol injection in dogs: histopathological evaluation by light and electron microscopy. Stereotact Funct Neurosurg 2002;79 (2):94-106. 67. Stajcic Z. Evidence that the site of action of glycerol in relieving tic douloureux is its actual site of application. Dtsch Zahnarztl Z 1990;45(1):44-6. 68. Burchiel KJ, Russell LC. Glycerol neurolysis: neurophysiological effects of topical glycerol application on rat saphenous nerve. J Neurosurg 1985;63(5):784-8. 69. Rappaport ZH, Seltzer Z, Zagzag D. The effect of glycerol on autotomy. An experimental model of neuralgia pain. Pain 1986;26(1):85-91. 70. Bennett MH, Lunsford LD. Percutaneous retrogasserian glycerol rhizotomy for tic douloureux: part 2. Results and implications of trigeminal evoked potential studies. Neurosurgery 1984;14(4):431-5. 71. Bennett MH, Jannetta PJ. Evoked potentials in trigeminal neuralgia. Neurosurgery 1983;13(3):242-7. 72. Rasminsky M, Sears TA. Internodal conduction in undissected demyelinated nerve fibres. J Physiol 1972;227 (2):323-50. 73. Waxman SG, Brill MH. Conduction through demyelinated plaques in multiple sclerosis: computer simulations
2455
2456
142 74.
75.
76.
77.
78.
79.
80. 81.
82.
83. 84.
85.
86.
Retrogasserian glycerol injection for trigeminal neuralgia
of facilitation by short internodes. J Neurol Neurosurg Psychiatry 1978;41(5):408-16. Fruhstorfer H, Lindblom U, Schmidt WC. Method for quantitative estimation of thermal thresholds in patients. J Neurol Neurosurg Psychiatry 1976;39 (11):1071-5. Eide PK, Stubhaug A. Sensory perception in patients with trigeminal neuralgia: effects of percutaneous retrogasserian glycerol rhizotomy. Stereotact Funct Neurosurg 1997;68(1–4 Pt 1):207-11. Kumar R, Mahapatra AK, Dash HH. The blink reflex before and after percutaneous glycerol rhizotomy in patients with trigeminal neuralgia – a prospective study of 28 patients. Acta Neurochir (Wien) 1995;137 (1–2):85-8. Rushton JG, Olafson RA. Trigeminal neuralgia associated with multiple sclerosis. A case report. Arch Neurol 1965;13(4):383-6. Brett DC, Ferguson GG, Ebers GC, Paty DW. Percutaneous trigeminal rhizotomy. Treatment of trigeminal neuralgia secondary to multiple sclerosis. Arch Neurol 1982;39(4):219-21. Jensen TS, Rasmussen P, Reske-Nielsen E. Association of trigeminal neuralgia with multiple sclerosis: clinical and pathological features. Acta Neurol Scand 1982;65 (3):182-9. Chakravorty BG. Association of trigeminal neuralgia with multiple sclerosis. Arch Neurol 1966;14(1):95-9. Linderoth B, Hakanson S. Paroxysmal facial pain in disseminated sclerosis treated by retrogasserian glycerol injection. Acta Neurol Scand 1989;80(4):341-6. Broggi G, Franzini A. Radiofrequency trigeminal rhizotomy in treatment of symptomatic non-neoplastic facial pain. J Neurosurg 1982;57(4):483-6. Brisman R. Trigeminal neuralgia and multiple sclerosis. Arch Neurol 1987;44(4):379-81. Pickett GE, Bisnaire D, Ferguson GG. Percutaneous retrogasserian glycerol rhizotomy in the treatment of tic douloureux associated with multiple sclerosis. Neurosurgery 2005;56(3):537-45; discussion 537‐45. Waltz TA, Dalessio DJ, Copeland B, Abbott G. Percutaneous injection of glycerol for the treatment of trigeminal neuralgia. Clin J Pain 1989;5(2):195-8. Rappaport ZH. Percutaneous retrogasserian glycerol injection for trigeminal neuralgia. One year follow-up. Pain Clin 1986;1(57–61).
87. Pieper DR, Dickerson J, Hassenbusch SJ. Percutaneous retrogasserian glycerol rhizolysis for treatment of chronic intractable cluster headaches: long-term results. Neurosurgery 2000;46(2):363-8; discussion 368‐70. 88. Franzini A, Ferroli P, Leone M, Broggi G. Stimulation of the posterior hypothalamus for treatment of chronic intractable cluster headaches: first reported series. Neurosurgery 2003;52(5):1095-9; discussion 1099‐101. 89. Leone M, Franzini A, Felisati G, Mea E, Curone M, Tullo V, et al. Deep brain stimulation and cluster headache. Neurol Sci 2005;26 Suppl 2:s138-9. 90. Franzini A, Marras C, Tringali G, Leone M, Ferroli P, Bussone G, et al. Chronic high frequency stimulation of the posteromedial hypothalamus in facial pain syndromes and behaviour disorders. Acta Neurochir Suppl 2007;97 (Pt 2):399-406. 91. Broggi G, Franzini A, Leone M, Bussone G. Update on neurosurgical treatment of chronic trigeminal autonomic cephalalgias and atypical facial pain with deep brain stimulation of posterior hypothalamus: results and comments. Neurol Sci 2007;28 Suppl 2: S138-45. 92. Burns B, Watkins L, Goadsby PJ. Treatment of medically intractable cluster headache by occipital nerve stimulation: long-term follow-up of eight patients. Lancet 2007;369(9567):1099-106. 93. Hannerz J, Linderoth B. Neurosurgical treatment of short-lasting, unilateral, neuralgiform hemicrania with conjunctival injection and tearing. Br J Neurosurg 2002;16(1):55-8. 94. Hannerz J, Linderoth B. Neurosurgical treatment of short-lasting, unilateral, neuralgiform hemicrania with conjunctival injection and tearing. Headache 2003;43 (4):429. 95. Black DF, Dodick DW. Two cases of medically and surgically intractable SUNCT: a reason for caution and an argument for a central mechanism. Cephalalgia 2002;22 (3):201-4. 96. Leone M, Franzini A, D’Andrea G, Broggi G, Casucci G, Bussone G. Deep brain stimulation to relieve drugresistant SUNCT. Ann Neurol 2005;57(6):924-7. 97. Spatz AL, Zakrzewska JM, Kay EJ. Decision analysis of medical and surgical treatments for trigeminal neuralgia: how patient evaluations of benefits and risks affect the utility of treatment decisions. Pain 2007;131 (3):302-10.
137 Spinal Cord Stimulation. Techniques, Indications and Outcome B. Linderoth . B. A. Meyerson
Background Spinal cord stimulation (SCS) as a direct clinical application of the gate-control theory is an exceptionally good example of translational pain research, i.e., when data derived from the laboratory bench are translated to bedside practice. In fact, in the classical publication in Science 1965, Melzack and Wall [1] explicitly stated that the theory could have therapeutic implications by means of selectively activating large diameter fiber systems for the control of pain. Later, Wall together with Sweet tested the possible applicability of the gate-control concept to pain in man by stimulating their own infra-orbital nerves by percutaneous needle electrodes and they observed that the pain sensitivity was decreased in the maxillar region after stimulation [2]. It should be noted, however, that the perception of pain being modified by an alteration of the interplay between large and fine fiber systems had been postulated and discussed by earlier researchers. Thus, already in 1906 Head and Thompson proposed that discriminative sensations, such as touch, normally exert an inhibitory influence on the impulses subserving pain. They further postulated that facilitation or inhibition of sensory impulses occur in the posterior horn before they are relayed to secondary neurons [3]. The notion that epicritic sensory system exerts an inhibitory influence over protopathic sensations was the base for trials with sensory thalamic stimulation already performed in 1962 in Paris by Mazars and colleagues [4]. They were thus the first to apply #
Springer-Verlag Berlin/Heidelberg 2009
electric stimulation of the sensory nervous system as treatment of severe neuropathic pain conditions. There is reason in this context also to refer to Noordenbos [5] who described the inhibitory influence of the fast on slow fibers as ‘‘fast blocks slow,’’ and, actually, Zotterman [6] earlier postulated that hyperalgesia may be due to absence of impulses rapidly conducted to the brain. In the early 1970s, the gate control theory was much criticized but it did represent a new conceptualization of pain. Its impact on modern research on pain cannot be overrated as recently also discussed in an editorial by Dickenson entitled ‘‘Gate Control Theory of Pain Stands the Test of Time’’ [7]. Based on experiments in cats, Shealy and colleagues [8] performed and reported the first trials with SCS as a clinical pain therapy. Soon, this method was enthusiastically applied to virtually all forms of chronic pain. It was estimated that already by 1978 a total of about 8,000 patients had been subjected to SCS in United States [9]. On the basis of the gate theory, SCS would be efficient in suppressing both acute and chronic pain of a nociceptive nature, and it is paradoxical that it is exclusively, or at least preferentially efficacious only for neuropathic forms of pain. It was not until around the mid-seventies that there was a general awareness of the distinction between nociceptive and neuropathic forms of pain, although it was anecdotally reported that SCS could not influence acute nociceptive pain [10] as well as when applied to experimentally induced nociceptive pain [11].
2306
137
Spinal cord stimulation. techniques, indications and outcome
Since the early 1990s successively more stringent selection criteria for the application of SCS have been applied and the method developed into an indispensable part of the armamentarium for the management of some forms of neuropathic pain. The growing interest in SCS and its practicing in numerous centers are reflected in the increasing number of publications, and in recent years, several meta-analyses and a few randomized controlled trials have been performed as well. Application of electric stimuli to the dorsal aspect of the spinal cord has been found to have different effects depending upon the level where the stimulation is applied and to affect functions in different organ systems. Thus, SCS may also influence pain due to ischemia by producing an increase in blood flow in diseased organs or territories. Furthermore, there is evidence indicating that the mechanisms involved when SCS is applied for pain in ischemic tissue and for neuropathic pain are fundamentally different. The beneficial effect on the former type of pain that is of a nociceptive type relates to the effect on local ischemia in the extremities or in angina pectoris (e.g., [12,13]. The technique of SCS application, indications and outcome have been surveyed in several recent reviews and textbooks [14–17]. It is estimated that more than 30,000 SCS pulse generators annually are implanted in North America, Europe and Australia; of these, about 18,000 are implantations of new systems. In many countries, restrictive reimbursement rules have prevented the general usage of SCS. A major concern is the cost of the device, and it is not until recently that cost-effectiveness analyses have demonstrated that the SCS treatment pays for itself within a relatively short period of time. To some extent, the dissemination of the SCS is hindered by a lack of knowledge, both of its clinical usefulness and its underlying mechanisms, and also by some skepticism towards its efficacy even among pain clinicians. Considering
that chronic neuropathic pain is relatively common and that it is often resistant to pharmacotherapy, SCS is still a much underused treatment modality.
SCS for Neuropathic Pain At present, various forms of neuropathic, and ‘‘mixed pain conditions’’ with a significant neuropathic component, are the main indications for SCS therapy. Since, as indicated above, the mechanisms activated by SCS in neuropathic pain and in ischemic pain conditions appear to be fundamentally different, there is good reason to deal with them separately also concerning selection principles, treatment regimens, etc. It should be evident from chapter 152 of this volume that merely moving the SCS lead along the neuroaxis will evoke stimulation-induced changes in different somatosensory and viscero-somatic reflexes, local microcirculation, etc. which may be used therapeutically.
Selection of Patients for Spinal Cord Stimulation Pain Analysis A thorough pain analysis is mandatory in order to establish a definite pain diagnosis and to identify and distinguish neuropathic and nociceptive pain components that often co-exist. In such cases the pain analysis also aims at an evaluation of which of these components contributes most to the patient’s suffering because alleviation of a nociceptive pain cannot be expected to occur. For obvious reasons, the chronicity of the pain must be certified and, in practice, this implies that the great majority of patients considered for SCS have a history of pain for at least
Spinal cord stimulation. techniques, indications and outcome
6 months. It is also the duty of the physician to make sure that the patient has been carefully examined by the organ specialists in order to confirm that no further etiological therapy is available. When considering interventional therapies, like SCS, it is generally required that the patient should have proven to be resistant to all available pharmacotherapy. However, there is reason to somewhat modify this requirement considering that in most cases the response to drugs effective for neuropathic pain is partial and often associated with troublesome side effects. This also applies to opioids that may in exceptional cases provide some relief also of neuropathic pain but carry a considerable risk of side effects as well. Therefore, SCS that has virtually no serious side effects is preferable to treatment with opioids, also when administered intrathecally, that can be anticipated to continue for many years. There are some special features of the pain that have to be taken into account because they may relate to the outcome of SCS:
Location of pain: pain located in the extremities is more likely to respond than pain in the trunk. In particular, pain located in axial midline structures is more difficult to influence. Pain related to posture and load is less likely to benefit than pain at rest. Paroxysmal and shooting pains are less likely to respond than steady pain. Disturbance of cutaneous sensibility associated with the pain should be carefully analysed. Preservation of some somatosensory functions in the painful area is a prerequisite for producing pain alleviation, and the presence of prominent hypesthesia may indicate loss of large-fiber dorsal column functions. The assessment of central conduction time of somatosensory evoked potentials has been advocated for ascertaining the integrity of the somatosensory system [18]. However, it has not been convincingly demonstrated
137
that this cannot be achieved by simply performing a thorough neurological examination, particularly with regard to cutaneous sensibility. Transcutaneous electric nerve stimulation (TENS) was originally developed in order to select patients for SCS, and it was not until later that it became a therapy by itself. That is the background why a positive response to TENS still has been regarded as indicating and predicting responsiveness to SCS. However, there are a few systematic studies addressing this issue and they have failed to demonstrate a clear relationship [19]. Nevertheless, it appears that the response to TENS is a useful predictor for the outcome of SCS when applied for angina pectoris. In most publications on the clinical usefulness of SCS it is strongly recommended that the selection of patients should include psychological evaluation, preferably performed by a psychologist or a pain-oriented psychiatrist. It is well documented that psychological factors correlate with the outcome of SCS [20–22]. However, such a relationship has been challenged, and insufficient evidence for selecting patients for SCS on the basis of psychological testing has been claimed [23,24]. Nonetheless, it is important to identify patients with major personality disorders, drug-seeking behavior and abuse since presence of any of these factors makes a positive outcome unlikely. An important aspect of the practical application of SCS is that, like other forms of electric stimulation in the treatment of pain, it requires some active participation of the patient. This implies that the patient must have some understanding of the technical aspects of the stimulation. The patient should also be able to communicate and describe the pain both from a quantitative and qualitative aspect. Therefore, language and culture barriers may render a successful SCS treatment almost impossible. It is crucial that the patient’s expectations on the outcome should
2307
2308
137
Spinal cord stimulation. techniques, indications and outcome
not be unrealistic because complete suppression of pain can generally not be expected, and often also partial pain alleviation is considered as a successful outcome. Another aspect of SCS treatment is that the responsible physician and his/her team must be prepared to supply an almost life-long continuing support.
Trial Stimulation The great majority of those practicing SCS consider trial stimulation, i.e., test stimulation via temporary, percutaneous extension cables, to be an indispensable step in deciding whether or not a complete stimulation system should be permanently implanted [25]. Moreover, in many countries a period of trial stimulation is required for reimbursement. In spite of the fact that 10–30 per cent of patients subjected to trial stimulation do not experience sufficient relief from pain to justify permanent implantation, there are those who maintain that such stimulation has no or very limited predictive value. However, it is true that there are, as yet, no systematic and well designed studies substantiating that trial stimulation improves the long-term outcome of SCS. Trial stimulation is generally performed for 1–2 weeks and it is then important that the patients spend at least part of the time at home in their natural environment that will help them to evaluate whether the treatment is of value also for their daily life activities. One would expect that the presence of percutaneous extension cables increases the risk for local infection but there are no data indicating that this is the case. (It should be noted that trial stimulation is rarely performed when SCS is applied for angina pectoris; most systems are here implanted in one session because of the high success rate in angina (see below)). The evaluation of the effect of trial stimulation is generally based on the records of the intensity of pain using the Visual Analogue Scale (VAS) and the patient is generally given a
form that can be used as part of a pain diary. The recommended stimulation regime varies among different centers. It is our experience that the assessment of the pain following periods of 30 min of stimulation is advantageous for distinguishing a ‘‘true’’ pain suppression from pain relief being the result of the masking effect of the paresthesiae that may occur if stimulation is on continuously. Therefore, we also pay special attention to the duration of the post-stimulatory pain suppressive effect, which we have found to relate to the efficacy of long-term stimulation. However, this issue has not yet been systematically investigated. Since SCS is often applied for ‘‘complex pain’’ conditions comprising different coexisting pain components that may be differentially affected by the stimulation, the possible relieving effects on these components should be assessed individually. In addition to the degree of pain alleviation the patient should also be instructed to evaluate the degree of global satisfaction with the treatment, improvement of function, etc. In order to further ascertain the existence of real pain suppression, and an attempt to reduce the placebo effect, the distribution of the stimulation-induced paresthesiae in relation to the painful area should be documented.
Technical Aspects The majority of electrode leads implanted have been, and still are, percutaneous, cable-like multipolar. Implantation is usually performed with the patients in the prone position assisted by frontal fluoroscopy. Usually some antibiotic prophylaxis is used. The procedure is performed under local anesthesia. The level of epidural puncture is determined by the desired position for the active electrodes and the length of the lead but it is usually situated some 20 cm below the target level. (> Figure 137-1). The approach is usually lateral–oblique, a technique which is mandatory for the thoracic spine (> Figure 137-2) using
Spinal cord stimulation. techniques, indications and outcome
137
. Figure 137-1 Schematic illustration of a percutaneously implantable SCS quadripolar electrode lead connected to a pulse generator (modified from Simpson et al. [16])
. Figure 137-2 Lateral-oblique approach for percutaneous implantation of a lead electrode using a Touhy needle for puncture of the epidural space at a shallow angle (this and > Figures 137-3–137-5 have been supplied by courtesy of Medtronic Inc. and Advanced Neuromodulation Systems (ANS), St Jude)
loss-of-resistance with air or saline. Fluoroscopy is used to guide the steering of the electrode lead and intraoperative test stimulation is used to finally place the lead on the target region where paresthesiae should cover the entire painful
region (or at least 80%). In some cases a sitting position is preferable because it may enable a more stable electrode position. Plate electrodes (‘‘surgical leads’’) are considered to be less likely to dislocate and are preferred as
2309
2310
137
Spinal cord stimulation. techniques, indications and outcome
first implants by some surgeons. The majority, however, uses plate implants when a percutaneous cable lead has been dislocated several times or when scar tissue prevents the passing of such an electrode to the target area. Plate implants can be performed under local or full anesthesia but also in spinal anesthesia that enables intraoperative test stimulation [26]. With the latest version of the surgical leads, double or triple columns of electrode poles enhance the possibility to steer the paresthesiae electrically (> Figures 137-3, > 137-4). There are many different electrode designs and configurations available in the market. > Figure 137-3 displays a variety of cable leads and plate electrodes. The latest generations of pulse generators are rechargeable, guaranteed to last for 9–15 years to motivate the high initial cost. Rechargeability is also desired when complex programming with several channels/programs/stimulation settings, consuming more
current, are used to cover a wide spread or otherwise difficult painful area (> Figure 137-5). There is a general trend to use more and more complex polar configurations and programming, often supplied by dual channel pulse generators, but there is little evidence supporting that the technically more advanced types of SCS systems are more effective than a simple quadripolar, percutaneously implantable electrode lead in experienced hands [103]. For a detailed description of the design and recommended clinical indications for the different kind of electrodes, the reader is referred to the extensive manuals supplied by the manufacturers.
Assessment of Outcome In many studies, the treatment result is evaluated only on the basis of the degree of pain relief
. Figure 137-3 Survey of different SCS electrodes supplied by various manufacturers
Spinal cord stimulation. techniques, indications and outcome
137
. Figure 137-4 (a). Radiograph (left) showing dual, multipolar SCS lead electrodes and (b). A schematic drawing (right) illustrating the extent and configuration of the electric field with a complex coupling of the stimulating poles
assessed by the use of VAS. As a rule, a relief of 50% is regarded as a satisfactory outcome and 75–90% as excellent. It is not until later years that the patients’ global evaluation of the usefulness of the treatment or their general satisfaction, health-related quality-of-life measures, analgesic medication and measures of functional ability have been more commonly employed. Such indices of efficacy of the treatment are in fact mandatory because evaluation of the intensity of pain using VAS may be misleading. There are many examples in the literature illustrating disaccordance between these two ways of assessing outcome. For example, in a recent study on the outcome of SCS in CRPS type 1 the VAS scores in the treated group did not differ significantly from those in a control group at 5 years’ follow-up. Nevertheless, 90% of the patients indicated that they benefitted from the SCS treatment [27]. In another study only 26% of the patients had 50% or greater relief from pain but 70% said that SCS helped them and that they would recommend it [28]. Such records
illustrate that it is often difficult to reach a definite conclusion regarding the clinical usefulness of a treatment like SCS. The reliability of outcomes of assessments in the great majority of earlier studies on SCS is doubtful due to the fact that the responsible physician has been involved also in the evaluating process. There are but a few studies where a ‘‘disinterested third party evaluator’’ has been employed. Furthermore, the outcome assessments have generally been based on telephone interviews or postal questionnaires, and ‘‘face-to-face’’ consultations have been rarely used. Therefore, information about changing distribution of paresthesiae, alteration of stimulation habits, etc., which may account for decreasing pain-relieving effects, may be missed. In the last decade there has been an increasing tendency towards applying evidence-based medicine principles for the evaluation of various forms of therapy, and this also relates to SCS. A number of evaluative reviews and meta-analyses are now available covering the SCS literature up
2311
2312
137
Spinal cord stimulation. techniques, indications and outcome
. Figure 137-5 Schematic representation of examples of paresthesiae distribution with different complex couplings of dual lead electrodes
to 2006. Somewhat different evidence classification systems have been used and although the rankings of quality of evidence are very similar, comparisons of the conclusions drawn in different reviewing publications may be somewhat confusing. A commonly used system, employed also by the Cochrane Collaboration, is the scale for quality assessment developed by Jadad et al. [29] using a score range between 0 (lowest quality) and 5 (high quality). Studies having a score of 3 or more are classified as high quality studies. Another system is that developed by the European Federation of Neurological Societies (EFNS) [30]. A third system is available within The Scottish Intercollegiate Guidelines Network supplying evidence-based clinical guidelines [31]. The base
for assigning a certain level of quality of evidence is a hierarchy of study types: 1. 2. 3. 4. 5.
Systematic reviews and meta-analyses of randomized controlled trials Randomized control trials (RCT) Non-randomized interventional studies Observational studies (case series) Expert opinion
The studies are graded according to, among other factors, bias on findings, confounding factors, and chance and degree of probability that a relationship is causal. There are four grades of recommendations approximately corresponding to the hierarchy of study type where the lowest level of
Spinal cord stimulation. techniques, indications and outcome
evidence corresponds to case reports, case series and expert opinion. The great majority of SCS publications represent very low level of evidence because they are observational and case series studies, most of them retrospective. There are just a few randomized controlled trials (RCT). However, it should be realized that the presence of perceived paresthesiae, which is a prerequisite for pain relief, precludes placebo or sham stimulation controlled studies. Therefore, because of this inherent feature of SCS it cannot be subjected to a truly blinded control.
Clinical Studies In the following sections, the clinical usefulness of SCS will be examined on the basis of some recent systematic reviews focusing on the most common indications for SCS when applied for neuropathic pain. Neuropathic and other pain indications for SCS and their likeliness to respond are summarized in > Table 137-1. It should be noted that, with proper selection, the highest success rates are obtained in vasculopathic pain, angina pectoris and with complex pain syndromes with autonomic components (CRPS 1 and 2) while mid-line neuropathic pain has a much lower probability to respond and pain conditions due to central lesions usually are totally unresponsive.
137
Lumbosacral Rhizopathy and Low Back Pain This is the most common indication for SCS. In the literature these forms of pain are generally referred to as Failed Back Surgery Syndrome, but this term is a misnomer because it is not a pain diagnosis and it only tells that the patient has been subjected to surgery of the low back. Besides, patients may present with the same type of pain without previous surgery. The precise term according to the IASP taxonomy is Lumbar Spinal or Radicular Pain after Failed Spinal Surgery. It should be recognized, however, that in most of the more recent publications the effect of SCS on the irradiating pain in a leg (or legs) is recorded separately from that on the low back pain component. There is common experience that the latter form of pain is much more difficult to influence, and this is at least partly due to the fact that it is difficult to produce paresthesiae that well cover the axial lumbar region. Moreover, this pain is presumably of a predominantly nociceptive nature and there is no evidence that SCS may directly alleviate such pain. There are, as yet, few studies that have specifically addressed the possibility of relieving lumbar axial pain (‘‘low back pain’’). It has been claimed that the use of dual, four-polar, eight-polar, or even 16-polar electrode leads, enhances the possibilities of producing paresthesiae that cover also axial structures [32], but this
. Table 137-1 Likely outcome of SCS relative to diagnosis (modified from Simpson et al. [16]) Success >> failure Angina pectoris. PVD: vasospastic
Success > failure
Success variable
Failure > success
Failure >>success
CRPS I and II. Peripheral nerve injury. Diabetic neuropathy. Brachial plexus. injury (partial). Lumbosacral and cervical rhizopathy. Cauda equina injury. Phantom limb and stump pain. PVD: occlusive
Intercostal neuralgia. Postherpetic neuralgia. Low back pain combined with leg neuropathy
Axial spinal pain. Perianal and genital pain. Partial spinal cord lesion
Central cerebral pain. Complete spinal cord lesion. Myelitis. Complete root avulsion
2313
2314
137
Spinal cord stimulation. techniques, indications and outcome
has not been confirmed in a later prospective, controlled trial [103]. Nevertheless, in these studies it appeared that about half of the patients, selected for permanent implantation after trial stimulation, enjoyed good relief of the pain component in the lower back [33,34]. For example, in one study on 41 patients no less than 69% were satisfied and reported fair to excellent relief of pain in the lower back [35]. There have also been some promising trials with a tripolar electrode design that was developed with the idea that it would facilitate the steering of the paresthesiae to better cover axial structures [36]. However, in an extensive, multicenter study it appeared that although this electrode configuration was useful for controlling the paresthesiae distribution the pain relieving effect was similar to that obtained by using a conventional type of electrode [37]. It should be noted that all these studies were case series and therefore representing a relatively low level of evidence. A Cochrane study on SCS for chronic pain was recently published, and it concluded that there was insufficient evidence to certify benefits of this form of therapy, mainly because of the very few controlled studies [38]. However, it should be noted that this analysis comprised publications from a data base search only up to September 2003. Taylor et al. [39] performed a systematic review of publications, most of them from North America and Europe, from 1995 to 2002 on SCS applied for ‘‘chronic back and leg pain.’’ Seventytwo case series from 1995 to 2002 were identified and subjected to a meta-analysis. These studies comprised 3,427 implanted patients and 62% of them had experienced >50% pain relief; relative to the total number subjected to trial stimulation 48% reported this benefit. Furthermore, it should be noted that 70% of the implanted patients expressed satisfaction with the treatment, and there were also significant improvement of health-related quality of life (HRQL). The maximal follow-up time was 10 years. The review
includes also one RCT study [40,41] where patients with radicular pain after lumbosacral spine surgery were randomized to SCS or re-operation. The out patient follow-up assessments were performed by ‘‘disinterested third-party interviewers.’’ A total of 45 patients were available for a mean follow-up of about 3 years and it appeared that SCS was more successful (in 47%; 9 out of 19 patients) than re-operation (in 12%; 3 out of 26 patients) (P < 0.01). Patients initially randomized to SCS were significantly less likely to cross over than those randomized to re-operation. This study was scored 4/5 on the Jadad scale. Recently, a prospective randomized controlled multicenter trial of the effectiveness of SCS (‘‘PROCESS-study’’) with the recruitment of 100 patients from a total of 12 centers has been reported [42]. The patients suffered from radiating leg pain (lumbosacral rhizopathy), predominant as compared to pain in the lower back. They were randomized to SCS combined with conventional medical management (CMM) or to CMM alone. At the initial, 6-months follow-up an intention-to-treat analysis showed that 48% of the SCS patients and 9% of the CMM patients (P < 0.001) achieved 50% or more relief from pain in the legs. Also with regard to relief from back pain quality of life, and functional capacity the SCS patients compared favorably to the CMM group. Interestingly, between 6 and 12 months 32 of the 48 CMM patients crossed over to SCS. The poor outcome of CMM treatment only is in accordance with the well-known fact that pharmacotherapy is effective only in at most 30–40% of patients with chronic neuropathic pain [43]. More recently, a 2 years follow-up has demonstrated a statistically significant higher pain relief as well as better quality of life and functional status in the SCS group [107]. This RCT study has been evaluated as class II (out of IV) representing evidence grading B according the EFNS guidelines referred to above [44]. Some of the recent major studies are briefly reviewed in > Table 137-2.
Spinal cord stimulation. techniques, indications and outcome
137
. Table 137-2 Selected studies of SCS for lumbar spinal and radicular pain
References
Study design
Follow-up/No patients and treatment 5 years/50 SCS
Burchiel et al. [106]
Prospective. Third party evaluation Prospective. Multicenter
Kumar et al. [102]
RCT consecutive
5 years/60 SCS. 44 CMM
North et al. [41]
RCT
3 years/19 SCS. 26 reoperation
Kumar et al. [42]
RCT multicenter
1 year/52 SCS. 48 CMM
Kumar et al. [107] (followup of study from 2007)
RCT multicenter
2 years/42 SCS. 41 CMM
North et al. [110]
1 year/70
Results and outcome measures
Comments
47% of patients >50% pain relief. 10 out of 40, preoperatively disabled, returned to work 55% of patients. >50% pain relief and significantly improved QoL measures. Analgesia consumption and work status unchanged QoL improved in 27% of SCS patients and 12 % CMM patients. Return to work: 15% in SCS and 0% in CMM patients SCS better pain relief (p < 0.01) and less opiod consumption (p > 0.025). Patients randomized to SCS significantly less likely to cross over 48% of SCS patients and 9% of CMM patients >50% pain relief. Significant improvement of QoL and satisfaction in SCS patients. 32 of CMM patients crossed over to SCS after 6 months Significantly higher pain relief, QoL and functional status in SCS patients
Impartial evaluation
Usage of a variety of pain and functional/QoL measures
SCS is cost-effective in the long term
The results obviate the need for reoperation in cases of recurrent radicular pain after spine surgery Variety of outcome measures. Still short follow-up
High level of evidence
RCT, randomized controlled trial; CMM, conventional medical management; QoL, quality of life
CRPS I The other principal diagnosis and indication for SCS that has been subjected to systematic reviews and analyses is the Complex Regional Pain Syndrome (CRPS) type I. This is the only indication where the efficacy of SCS has attained the highest degree of evidence level (A, Harbour-Miller scale; Jadad score 3/5 representing level 3 evidence). This is mainly due to a single, intention-to-treat RCT study described in a series of publications by
Kemler et al. [27,45–47] where originally 54 patients were assigned to SCS together with physical therapy (PT) (n = 36) or physical therapy alone (n = 18). After 6 months, patients in the PT group had the possibility to cross over to SCS. The patients were checked at follow-ups regularly on an annual base after the first year up to 5 years. At this end-point, the SCS + PT group contained 31 patients and the PT group 13. At 2 years’ follow-up, pain intensity, assessed by VAS, decreased by 3.6 cm in the SCS group as
2315
2316
137
Spinal cord stimulation. techniques, indications and outcome
compared to 0.2 cm in the PT group. A similar difference was recorded when the patients themselves assessed the degree of relief and there was also an improvement of HRQL in the SCS group. However, after 5 years’ assessment it appeared that there were no longer any significant differences between the 31 patients in the SCS + PT group compared with 13 patients in the PT group. Nevertheless, a comparison between patients who actually received SCS (irrespective of the original randomization) as compared to those with PT alone disclosed that the 20 SCS + PT patients available at the follow-up had a somewhat, albeit not a statistically significant, greater relief from pain (p = 0.06) [27]. Moreover, seven of these 20 patients with implants reported ‘‘much improvement’’ as compared to two of the 13 patients with PTonly, and no less than 90% of the patients with an SCS implant indicated that they had responded positively to the treatment. In spite of this, somewhat disappointing outcome, the authors conclude that they, for several reasons, remain confident that SCS treatment is worthwhile in chronic CRPS type I, and they refer to the fact that also after 5 years there was still a high degree of patient satisfaction. Taylor [48] and Taylor et al. [49] analyzed data from 25 case studies on SCS treatment of CRPS with a mean follow-up of 33 months. They found that 67% of the patients were reported to have a >50% relief from pain. A similar result was reported by the European Task Force for evaluation of neurostimulation (referred to above; [44]) but the evidence quality has been ranked as low, C. A literature review has been also performed by Grabow et al. [50]. They considered 15 studies, comprising one RCT (referred to above), two prospective observational and 12 retrospective observational studies worthwhile to be analyzed. They concluded that although the observational studies were of poor quality, available evidence from the examined literature suggests that SCS is effective in the management of CRPS. In a few, non-controlled studies it has been claimed
that a temporary relief from pain following a sympathetic nerve block is a reliable predictor for a beneficial effect of SCS in CRPS I [51,52]. Some of the recent major studies on CRPS are summarized in > Table 137-3. A systematic literature review of SCS for neuropathic pain was performed by the Medical Advisory Secreteriat, Ontario Ministry of Health [53]. It covered publications from 2000 to 2005 and the primary outcome for the review was pain relief and the secondary outcomes were functional status and QoL. Not more than two RCTs and two prospective non-randomized controlled trials were retrieved. It was concluded that good evidence exists to support the effectiveness of SCS to decrease pain with an evidence level of 2 (Jadad scale) for lumbosacral rhizopathy low back pain and CRPS I, and level 3a for postherpetic neuralgia [54]. There is also a similar Australian literature review [55]. The usefulness of SCS applied on other pain indications, as listed in > Table 137-1, is substantiated only by mostly retrospective case series and observational studies. However, several of these studies comprise a relatively large number of patients in each diagnostic group and follow-ups exceed 10 years. The reported outcome for some of these diagnoses are surprisingly concordant attaining a satisfactory relief from pain (>50%), a high degree of patient satisfaction and reduction in analgesic consumption also at long-term in 50–65% of the patients implanted (e.g., [56–58]. Even though the usefulness of SCS is poorly substantiated from a strict scientific point of view and from an evidence-based-medicine perspective, there is ample documentation of thousands of patients who have enjoyed persistent relief from pain that is otherwise resistant to various forms of treatment, and these data cannot be neglected. Moreover, the recently published outcome of the PROCESS-study substantially augments the quality of evidence for SCS applied for neuropathic leg and lower-back pain. Presumably, we can also anticipate in the near future further prospective
Spinal cord stimulation. techniques, indications and outcome
137
. Table 137-3 Selected studies of SCS for CRPS I Follow-up/No patients and treatment
References
Study design
Kemler et al. [46]
RCT
2 years/36 SCS + PT. 18 PT
Kemler et al. [27] (Follow-up of study from 2004)
RCT
5 years/36 SCS + PT. 18 PT
Grabow et al. [50] (including CRPS II)
Systematic literature review. One RCT, 2 prospective and 12 retrospective observational studies Prospective, noncontrolled
6–40 months/314 (IASP criteria for CRPS) SCS. Ten studies reported trial stimulation 12 months/29 SCS +PT, all with positive response to sympathetic block
Harke et al. [52]
Results and outcome measures 24 patients received permanent SCS. Intentionto-treat analysis: SCS better pain relief and global effect (both p < 0.001). Healthrelated QoL improved only in SCS patients Intention-to-treat analysis: Pain relief and other outcome measures were the same in both groups. Subgroup analysis of 20 SCS and 13 PT patients: SCS patients had better pain relief (p = 0.02) and global effect (p = 0.06) Successful results in 54–100% of patients. Some studies reported analgesic reduction, improved QoL and global perceived effect 28/29 reduction of pain (VAS 4) and allodynia in all (p < 0.01). Pain disability index significantly reduced (>50%) (p < 0.01)
Comments High level of evidence due to intentionto-treat analysis. Estimated NNT = 3.0
Despite diminishing effect of SCS, 95% of patients would repeat the treatment for the same result
High level of evidence for the review. Moderate level (B/C) of evidence for efficacy of SCS for CRPS Modest level of evidence. Biased patient selection
PT = physical therapy
and randomized controlled studies of SCS applied for other forms of neuropathic pain.
SCS for Ischemic Pain Conditions Background For more than 30 years spinal cord stimulation has been used for pain due to tissue ischemia in the extremities (peripheral vascular disease (PVD), or peripheral arterial occlusive disease (PAOD). The pioneer report by Cook et al. appeared in 1976 [59]. The spread of SCS for this indication was initially slow but after the presentation of a number of studies, most of them, the method
was in the late eighties and early nineties adopted in many centers, virtually all in Europe (review, see [60]. In the early period, criteria for patient selection were poorly defined and the enthusiasm over the favorable short-term results was soon replaced by disappointment over the long-term poor outcome. Meanwhile there was also many technical advancements in vascular surgery enabling longer by-pass grafting procedures, endovascular interventions, etc., as well as new more effective drugs, leaving the patient group, finally considered to be beyond benefits from surgery and pharmacotherapy, but eligible for SCS, small and in a relatively bad health condition. This group was often the elderly, who suffered from advanced
2317
2318
137
Spinal cord stimulation. techniques, indications and outcome
atherosclerosis in a stage of rapid progress towards critical, limb-threatening ischemia. In some countries, the use of SCS for PAOD was almost stopped and in many others it decreased considerably. Only in some centers did the method survive and finally, randomized controlled trials were initiated. However, stringent selection criteria were still not applied and these RCTs unfortunately encompassed heterogeneous patient materials that necessitated post-hoc stratification in order to explain for the variable outcomes. Electrostimulation treatment in pain due to coronary ischemia was started in the early 1980s with TENS introduced by Mannheimer and colleagues in Gothenburg, Sweden [61]. The outcome was remarkably promising. The electrodes were generally placed parasternally in order to produce paresthesiae covering the anterior chest. During the same period, SCS was in use for ischemic pain in the lower extremities, and it then seemed logical to move the spinal electrode rostrally (to the T1-T2 level) to induce paresthesiae projecting to the area where angina was experienced. The first reports on SCS in angina pectoris were published by an Australian and a Swedish group almost simultaneously in 1987–1988 [62,63] and the results were very promising. However, this new indication for SCS was first met with skepticism, and there was a fear that the stimulation would conceal important alarm signals of an imminent cardiac infarct. However, several studies clearly demonstrated that that was not the case. In the 1990s SCS for angina pectoris spread to many centers throughout Europe (reviews, see [60,64]) but was never approved by FDA for use in the United States. It is estimated that to date more than 5,000 SCS systems have been implanted worldwide on the indication ‘‘treatment-refractory angina.’’ Actually, angina has developed into the best indication for SCS with a general success rate above 80% of patients experiencing significant reduction in pain. Considering that SCS has since long been believed to be exclusively, or at least predominantly,
effective for neuropathic forms of pain (e.g., [65,66]) it appears contradictory that it may also relieve pain resulting from extremity ischemia as well as from cardiac disease which are mainly of a nociceptive nature. However, there are several indications that the mechanisms involved in the stimulation-induced alleviation of ischemic pain are fundamentally different from those active in neuropathic pain (cf. chapter 152).
SCS for Pain in PVD Patient selection. Peripheral arterial insufficiency is usually due to atherosclerosis and generally starts as intermittent claudication. This is a relatively common condition and affects about 5–10% of people above 65 years of age. In a fraction of these patients there is progress to critical ischemia producing severe pain at rest, ischemic ulcers and a threat to the extremity. The criteria for selection of cases suitable for SCS therapy have developed during the years and lately some ‘‘evidence-based’’ specific measures have been proposed to improve the selection: 1.
The principal indication is severe ischemic pain at rest (supine) and without tissue loss (Fontaine stage III; > Table 137-4). A careful evaluation of the preoperative pain condition is mandatory. Only the deep aching ischemic pain can be expected to respond to SCS. It should be noted that peripheral ischemia may also affect the peripheral nerves resulting in neuropathic pain, particularly in cases with diabetes. Pain and edema due to inadequate venous return are not remedied. Other nociceptive pain components such as those from ulcers and borders of gangrene, as well as those arising from ulcer care, should not be expected to be alleviated by the stimulation (review, see [67]). The patient must be
Spinal cord stimulation. techniques, indications and outcome
. Table 137-4 The Fontaine classification of symptoms in peripheral vascular disease Stage
Clinical features
I II
Arteriosclerosis with no symptoms Intermittent claudication with no symptoms at rest. IIa: Intermittent medium claudication. IIb: Intermittent severe claudication Claudication + rest and night pain without tissue involvement Grade III + tissue loss (ischemic ulceration; gangrene) IVa: with local inflammation IVb: with wide-spread inflammation
III IV
2.
3.
4.
5.
6.
aware of this before being admitted to surgery. No reconstructive vascular surgery is possible – or else it is contraindicated. Conservative therapeutic facilities are exhausted. If ischemic ulcers are present (Fontaine stage IV; > Table 137-4) they should be below 3 cm in diameter. In some cases arrest of tissue loss may be the primary goal, and SCS is then applied with the aim to permit a more distal amputation site. The preoperative transcutaneous oxygen pressure (TcpO2), measured apically on the diseased extremity (usually at the dorsum of the foot), should be between 10 and 30 mm Hg. Patients with a higher, as well as lower, microcirculatory index have lesser chance to benefit [68,69]. Gersbach et al. [70,71] have proposed comparative evaluation of TcpO2 in the supine and sitting positions. A gradient exceeding 15 mm Hg predicts a successful outcome in 88% of cases. Trial stimulation is recommended in order to assess not only alleviation of ischemic pain but also increase in TcpO2. Peripheral microcirculation can be further monitored by laser Doppler flowmetry and in a laboratory setting also by vital microscopy.
137
It should be noted that MR investigation with body coil is not permitted with an SCS device (absolute contraindication) and that the presence of an on-demand pacemaker may create problems (relative contraindication).
Technical Aspects The placement of the electrode lead follows the same principles as for neuropathic pain aiming at full paresthesiae coverage of the painful area. The SCS treatment is usually started within 2–3 days of continuous stimulation and subsequently the patient may use the stimulator freely. In cases of vasospastic ischemic pain (Raynaud’s phenomenon) the response latency is relatively short (20–30 min).
Clinical Outcome In the 1980s, a large number of case series with SCS applied for ischemic pain in the extremities were published. Although the reported outcomes with regard to pain alleviation and improved walking capacity were markedly concordant, the quality of these studies represented only a low level of evidence substantiating the usefulness and efficacy of SCS for this indication. Subsequently, several studies representing a good level of evidence were performed, but in these later publications interest has been focused more on a possible limb/tissue salvation effect of the SCS treatment than on the relief of the ischemic pain per se. In a Cochrane review [72,73] six randomized-controlled, or controlled, studies published after 1994, were identified. One of them was from Belgium, one from Germany, one from Sweden, two from The Netherlands and one was a multinational European study. The primary objective was to assess the SCS effect on limb salvage and the secondary on pain relief, wound healing
2319
2320
137
Spinal cord stimulation. techniques, indications and outcome
and quality of life. A meta-analysis of the pooled 444 patients showed an 11% lower amputation rate after 12 months in SCS patients as compared to optimal medical treatment alone. Moreover, the SCS group reported a lower demand of analgesics and more often reached Fontaine stage II than the conservatively treated group. However, the conclusion that SCS may reduce the risk of amputation has been challenged by Klomp et al. [68] based on the Dutch multicenter study and a later follow-up of that study [74]. Thus, there were no significant differences in amputation rates between the patients (N = 120) randomized to SCS vs. conservative treatment. These studies utilized the original intention-to-treat design, i.e., the ‘‘original randomization’’ was maintained and no post-hoc stratification was performed. The authors concluded that the NNT (Number Needed to Treat to save one patient from amputation) for SCS in leg ischemia might be as high as 13–14. However, when the patient material from the Dutch multicenter study was pos-hoc stratified according to the preoperative microcirculatory status, a clear tendency emerged that patients with a moderately compromised peripheral circulation benefited more from SCS than those with less or with more advanced PVD [69,71]. This finding was further supported by a later European, multicenter controlled, but not randomized, study (SCS-EPOS Study). It was demonstrated that if the patients were selected on the basis of TcpO2 and trial stimulation screening, those who presented with ‘‘PVD of moderate severity’’ (preoperative TcpO2 between 10 and 30 mm Hg) had significantly higher foot salvage than those with a lower value. However, patients implanted without meeting the inclusion criteria had an outcome comparable to that following conventional therapy only [75]. The intention-to-treat design has somewhat confounded the outcome also of the other RCTs, and it should be noted that this approach precluded the use of trial stimulation for screening
of the SCS candidates. A further problem is that the rate of cross-overs has been relatively high, making the conclusions based on the original intension-to-treat design invalid. Thus, in one Swedish study, 51 patients were randomized to SCS or conservative treatment, and the outcome indicated a clear, though not significant, trend for a limb saving effect of SCS [76]. However, 12% of patients randomized to SCS, for various reasons, never received implants. Furthermore, the authors observed a tendency for a less satisfactory result in cases with arterial hypertension, a condition that by other researchers has been reported to relate instead to a beneficial outcome [77]. If the hypertensive patients were excluded from the material the remaining SCS group demonstrated a significantly lower amputation rate. In a few studies the possible SCS specific effect on peripheral circulation has been examined. It was reported that improvement from critical leg ischemia (Fontaine grade III) to claudication (grade II) was significantly more common in an SCS treated group than in control patients. This corresponded to an increase of the TcpO2 value [78]. Three studies have reported significantly better relief from pain in the SCS groups than in those subjective to conventional therapy also including prostaglandin E1 [75–77]. Conversely, in the Dutch multicenter study there was no difference in the relief from pain, as assessed with VAS, between the two groups although the patients who had received SCS required significantly less analgesics, mostly opioids [78]. These contradictory data illustrate the methodological problems when using only VAS as a pain measure. It can be concluded that a relatively high level of evidence indicates that SCS may effectively alleviate pain resulting from limb ischemia although a few high quality studies fail to statistically demonstrate this effect. The general trend is that SCS exerts a positive effect on most of the measured parameters
Spinal cord stimulation. techniques, indications and outcome
such as relief from pain, limb salvage and various measures of the peripheral microcirculation but in unselected materials this often does not show up as statistically significant differences between SCS treated and control groups. However, as illustrated by several of the RCTs referred to the above, when patients are selected according to strict criteria related to preservation of peripheral microcirculatory dynamics [71], stage of disease progression (Fontaine), type of dominating pain (ischemic-ulcer, nociceptive-neuropathic), usage of trial stimulation for screening patients for permanent implantation and attentive controls to ensure optimal function of the stimulation, the SCS treatment can be expected to have
137
more predictably beneficial effects (discussion, see [79]). This needs further confirmation in future RCTs. SCS applied for vasospastic conditions has not been subjected to RCTs, and there are only a few retrospective case series (e.g., [80] (Raynaud’s disease); [81] (scleroderma)) demonstrating positive effects, both pain relief and improved peripheral circulation. Also in Buerger’s disease has SCS been reported to have a beneficial effect on microcirculation and to be protective for trophic lesions as substantiated in a retrospective case series of 29 patients [82]. Some recent major studies are briefly reviewed in > Table 137-5
. Table 137-5 Selected studies of SCS for ischemic limb pain Follow-up/No of patients and treatment
References
Study design
Jivegard et al. [76]
RCT
18 months/25 SCS. 26 analgesic treatment
Claeys and Horsch [77]
RCT. Designed to test TcPO2 as predictor
12 months/45 SCS + PGE1. 41 PGE1
Spincemaille et al. [78]
RCT. Dutch multicenter
18 months/60 SCS. 60 CMM
Amann et al. [75] (EPOS)
Prospective, controlled, European multicenter but not randomized
12 months/41 TcPO2 response to SCS. 32 SCS. 39 standard treatment
Petrakis and Sciacca [111]
Retrospective. Diabetic with limb ischemia corresponding to Fountaine III and IV. Test of TcPO2 as predictor
6 months/SCS: 32 without neuropathy and 28 with neuropathy
Results and Outcome measures Limb salvation in SCS group 62% and in control group 45% (NS). Pain relief better in SCS group Better healing of foot ulcer with SCS (p < 0.05). TcPO2 increased in SCS group (p < 0.0001). Better QoL with SCS. Transition from Fontaine IV to II: 40% in SCS versus 10% in PGE1 Intention-to-treat analysis: no major differences in pain relief and QoL between the groups, though SCS group used less analgesics Significantly better pain relief and increased TcPO2 with SCS. Fewer amputations (p < 0.05). SCS without TcPO2 response was no better than standard treatment 35/60 >75% pain relief. Amputation in 13/60. Significant increase in TcPO2 related to good outcome. Presence of neuropathy predicted poor outcome
Comments
High level of evidence
High level of evidence. However, no trial stimulation and TcPO2 monitoring Moderately high evidence
Modest level of evidence
2321
2322
137
Spinal cord stimulation. techniques, indications and outcome
SCS for Angina Pectoris Angina pectoris, which is a cardinal, but relatively late, symptom of heart disease, is the result of an imbalance between cardiac oxygen demand and the supply via the coronary arteries causing reversible ischemia in the myocardium. The great majority of patients suffering from angina are effectively helped by pharmacotherapy (short- and long-acting nitrates, beta-receptor blocking agents, calcium antagonists). In advanced cases with established occlusions of major coronary vessels percutaneous transluminal coronary angioplasty (PTCA) or by-pass surgery (CABG) may be indicated. However, many cases suffering from severe disabling angina (New York Heart Association (NYHA) class III-IV) are elderly and have concurrent diseases making them unsuitable for major invasive procedures. Sometimes the obliterations are too widespread or situated too far distally to permit a successful surgical intervention. It is estimated that about 10% of patients with coronary arterial disease are unsuitable for revascularization surgery. These patients, who suffer from what is termed ‘‘treatment refractory angina’’, have a low quality of life because of insufficient relief from pain, very limited physical capacity, and frequent periods of hospital admissions. In some cases with typical symptoms of angina, no obstructions of the cardiac circulation are demonstrable in the coronary angiogram. This is the so-called syndrome X, sometimes referred to as ‘‘small vessel disease’’, which is not suitable for surgical interventions. Refractory angina is a relatively common condition and it is estimated that at least 2 out of 3 patients seen by each cardiologist cannot get adequate relief from pain. For Europe, the prevalence has been estimated to between 30- and 50,000 patients. This means that a large group of patients sustain poor pain control, and they also represent large costs for society. For these patients classical medicine has little to offer beyond optimizing the
pharmacotherapy and SCS may therefore be a promising treatment option.
Patient Selection Inclusion criteria: 1.
2. 3.
4. 5.
Severe angina pectoris (NYHA class III-IV) refractory to conventional treatment. (In recent years the Canadian Cardiovascular Society classification (Class I-IV) has been more commonly used). A careful pain analysis is mandatory. Patients who have been subjected to thoracotomy may present with other forms of pain (post-thoracotomy syndrome, intercostal neuralgia) that must be distinguished from the angina due to reversible cardiac ischemia. Significant coronary artery disease. Examinations should have demonstrated reversible myocardial ischemia as a cause of the symptoms. Pain alleviation with TENS indicates a high likeliness for a positive response to SCS. Patients diagnosed as suffering from syndrome X may also benefit.
Exclusion criteria: 1. 2. 3. 4.
Acute myocardial infarction. Other on-going heart diseases (e.g., peri/ myocarditis). Patient has on-demand pace-maker (relative contraindication). Need of MRI study. (MRI investigation with body coil is not possible after implantation.)
Technical Aspects The placement of the electrode lead follows the same principles as for neuropathic pain aiming
Spinal cord stimulation. techniques, indications and outcome
at full paresthesia coverage of the painful area – usually the upper mid or left thoracic area. This is usually attained with a placement at T1-T2 slightly off midline. If the angina has a different distribution the paresthesiae should be steered accordingly, for example to encompass also the cheek, an arm, etc. Positive clinical outcome has also been reported with the electrode tip at as high as C2 [83]. Percutaneous testing is rarely necessary, and the lead and the pulse generator may be implanted in one session. The SCS intensity is usually set to produce mild paresthesiae during 4–8 h per day with the possibility to increase to maximum tolerance at times of angina attack or when exerting physical work, e.g., climbing a staircase.
Clinical Outcome Among the various indications for SCS, the application for refractory angina pectoris attains the highest level of evidence for its efficacy as shown by several, well designed RCTs. In general, 80–90% of patients report a marked relief from pain with a reduced frequency of angina attacks, a diminished need for short-acting nitrates, fewer visits to the emergency room and an enhanced quality of life [60,64,84–86]. Furthermore, there are many reports of stimulation-induced changes in various indices of coronary ischemia during work load such as reduction of the ST segment depression in the ECG [87], reversal of cardiac lactate production to extraction [87], and increased working capacity paralleled to increased time to angina during work load [87–89]. TenVaarwerk et al. [90] summarized results from 14 centers comprising a total of 517 patients subjected to SCS with a median followup of 23 months. The cardio-vascular mortality was around 5% per year. The mortality was related to factors as sex, cardiovascular history, age >71 and to the use of certain medications
137
probably indicating a more severe disease. As a result of SCS the NYHA class improved from 3.5 to 2.1. It was concluded, also on the basis of a literature search, that there were no indications that the clinical outcome of refractory angina is adversely affected by SCS treatment [91]. In a RCT, 13 SCS patients were compared to 12 controls during a 6 week’ period [89]. There were significant differences in favor of the SCS group with regard to exercise duration and time to angina, number of angina attacks, nitrate consumption, quality of life and degree of pain suppression. Similar favorable results have been reported in an Italian, multicenter prospective study comprising 104 patients with a mean follow-up of 13 months [92]. In a recent, placebocontrolled study in 12 responding SCS patients, the effects of conventional, subthreshold and lowoutput (0.1 V) stimulation during 4-weeks periods were compared [93]. It appeared that functional status and symptom relief were significantly improved only during active stimulation (including subthreshold) and absent with ‘‘placebo’’ output intensity. Thus, the presence of paresthesiae does not appear to be mandatory to obtain beneficial effects in angina. In another controlled study, the ESBY-project [94], 104 patients with angina were randomized to SCS or to coronary by-pass surgery (CABG). Both groups experienced significant symptom relief and there were no significant differences regarding angina control. In the CABG group, the exercise capacity increased significantly and also various indices of ischemia changed positively. However, with an intention-to-treat basis both the cardiovascular morbidity and the mortality rate were significantly lower in the SCS group. The conclusion was that SCS and CABG appear to provide comparable degrees of symptom relief, but considering the risk of complications associated with by-pass surgery, SCS is an attractive option, particularly for the elderly and fragile patients. Ekre et al. [95] reported parallel survival curves after the first 5 years of either CABG or SCS.
2323
2324
137
Spinal cord stimulation. techniques, indications and outcome
The beneficial effects of SCS on angina symptoms are also illustrated by the observation that when a pulse generator battery is depleted the patients are very eager to have it exchanged promptly. A review by Ekre et al. [96] clearly demonstrates the augmented frequency of angina attacks, the increased nitroglycerin intake as well as the return to the previous SCS condition when the battery wears out and is subsequently replaced. Recently Bo¨rjesson et al. (2008) [97] after a systematic survey of publications up to 2007 found 8 medium to high score studies according to the Jadad scale. These studies support the conclusion that SCS results in symptomatic and functional benefits and improved QoL in patients with severe angina pectoris. Some of the recent major studies are briefly reviewed in > Table 137-6.
reported that the cost pattern is reversed after about 2.5 years. However, the brake-even point is variable depending upon different costs for hospital stay, fees for physicians and physiotherapists, and drugs that differ considerably between countries [101,102,103]. For angina pectoris, [104] Rasmussen et al. already in 1992 calculated gains for society in the form of reduced hospital admissions to be 5,700 USD/ year/patient and for reduced home care, etc. to 2,300 USD/year/patient. More recent publications [95,105] report SCS for angina to be the more economic therapy after only 1.3–2.5 years. It should be noted, however, that the pulse generator has to be replaced after an average of about 3.5 years, causing an additional cost. The cost of rechargeable pulse generators is considerably higher than the conventional ones but may in selected cases be more cost-effective.
Safety Aspects As noted above it has been questioned, particularly among cardiologists, whether SCS may mask critical coronary ischemia. Andersen et al. [98] reviewed ten patients with SCS who had sustained acute coronary infarction. In 9/10 (for the tenth data are incomplete) the infarct was detected in spite of the ongoing stimulation therapy. This observation has been confirmed in subsequent studies [99]. Another concern is whether SCS may induce or aggravate arrhythmias. This has been investigated in two studies [89,100] that failed to demonstrate any such effects of SCS.
Cost-Effectiveness of SCS There are several studies that have examined the costs of SCS compared to the cost associated with health care without surgery for patients with neuropathic pain. There is solid evidence that although the initial cost of the SCS device together with hospital stay is much higher in the initial phase of the treatment, SCS treatment is over time potentially cost saving. Most studies have
Adverse Effects of SCS In virtually all studies on SCS the reported incidence of side effects is relatively high [104]. In spite of the on-going advancement of SCS technology the device related complications are still common. For example, in the recent PROCESS study no less than a third of the patients experienced such complications during the first 12 months of the study, and 24% of them required minor surgery to resolve the problems [42]. The principal complications were electrode migration. However, it should be noted that also the control group subjected to conventional pharmacotherapy experienced side effects in the form of drug adverse events. Also in other studies it has been reported that about a third of the patients has encountered technical problems. Infections, mostly occurring at the site of the pulse generator, are in most studies reported to occur in 3–5%, but it is very rare that they spread to the spinal canal. One can conclude that adverse events are common but the rate of serious complications is extremely low.
Spinal cord stimulation. techniques, indications and outcome
137
. Table 137-6 Selected studies of SCS for refractory angina pectoris
References
Study design
Follow-up/No patients and treatment
De Jongste et al. [84]
RCT
1 year/8 SCS 9 awaiting SCS for 8 weeks
Hautvast et al. [89] Mannheimer et al. [94] (ESBY study)
RCT
6 weeks/13 SCS 12 standard treatment 6 months/53 SCS. 51 CABG*
Andrell et al. [91] (ESBY study)
RCT
2 years/48 SCS. 39 CABG
Ekre et al. [96]
Prospective study of SCS dysfunction
65 months/32 with SCS battery dysfunction and after restitution of SCS function
McNab et al. [108] Di Pede et al. [92]
RCT
1 year/30 SCS. 30 PMR** 13 months/87 SCS
Eddicks et al. [93]
‘‘Placebo’’ controlled, randomized
RCT
Prospective, multicenter
4 weeks/12 SCS responding patients randomized to different stimulation regimes
Results and outcome measures At 8 weeks SCS patients had increased exercise capacity and time to angina, decreased signs of ischemia in ECG and decreased frequency of anginal attacks Same as De Jongste study. Also improved QoL Both groups had adequate symptom relief and no difference between SCS and CABG. Intention-to-treat analysis: mortality and cerebrovascular morbidity lower in SCS group SCS was less expensive than CABG and required fewer hospitalization days for the primary intervention and for cardiac events Anginal attacks and consumption of nitrates increased during dysfunction but were significantly diminished after restitution. No difference in angina-free exercise capacity and QoL >50% reduction anginal attacks in 73% of patients. Decreased hospital admissions and days of hospitalization Low intensity (placebo) stimulation resulted in reduction in walking distance, degradation in cardiovascular classification and increased frequency of angina attacks
Comments High level of evidence but few patients
High level of evidence but short follow-up High level of evidence. SCS seems preferable in patients with increased risk of surgical complication
Moderately high level of evidence
High level of evidence
Modest level of evidence Moderate level of evidence. First trial with placebo-like control First blinded SCS study
*Percutaneous myocardial laser revascularization **Coronary artery bypass grafting
Concluding Remarks SCS is presumably the most commonly practiced neuromodulation technique for pain and is presently an indispensable treatment modality for many pain clinicians. The main feature of SCS
is that it may be efficacious for some common pain conditions that otherwise are notoriously difficult to manage. Neuropathic pain in particular is known to respond to pharmacotherapy only in a minority of cases and the choice of drugs that may be helpful is very limited. For
2325
2326
137
Spinal cord stimulation. techniques, indications and outcome
some forms of neuropathic pain, other types of interventional treatment such as RF lesions of spinal nerves or ganglia as well as nerve blocks have been tried but with little or only temporary pain relief. Due to the paucity of RCTs, the usefulness of SCS, especially when applied for neuropathic pain, is not substantiated by a high level of evidence, and the great majority of publications reporting on the clinical long-term outcome of SCS are uncontrolled case series representing a relatively low study quality. An inherent feature of SCS is the presence of paresthesiae as a prerequisite for its pain suppressive effect and this precludes a truly blinded placebo-controlled study design. Only one study so far has employed a sort of placebo stimulation by using subliminal intensities [93]. Another problem when evaluating and comparing different studies is that the outcome assessment measures are quite varied. Nonetheless, SCS is presently fully accepted by the clinical pain community, and even by most heath care providing institutions – with the exception of countries with very limited budgets for public health care. The principle reason for its universal acceptance is that a positive costeffective balance is now supported by solid data. Further, it should be noted the results of SCS applied for neuropathic pain reported in a large number of publications are strikingly concordant. No doubt, considering that most patients with such pain fail to respond to pharmacotherapy it is impressive that more than half of them, sometimes up to about 65%, experience >50% pain alleviation and an even relatively higher degree of global satisfaction. One important aspect of SCS, not commented on above, is that the beneficial effect may persist for many years. There are several reports in the literature documenting that patients continue to experience good relief from pain for many years, even for decades (see also [107]). Conversely, there are no data indicating signs of ‘‘tolerance’’ or fatigue, and the majority of
reported late failures occur in the first year after implantation. The favorable experience of SCS as a unique treatment option has materialized in two consensus documents [25,44]. Furthermore, there is evidence that SCS may effectively ameliorate ischemic extremity pain, but in later years many studies have been designed to examine instead the possible effect on an objective, ‘‘easy to communicate,’’ result like limb salvage. The practicing of SCS for ischemic pain is presently performed only in a few centers. This is regrettable considering that, provided stringent selection criteria are applied, very favorable results can be obtained; it should be realized that some of the new specific criteria have been validated by robust data. The EPOS study, among others, indicates that application of SCS without such strict criteria does not produce better results than conventional treatment [75]. There is much evidence that SCS may provide very efficient alleviation of symptoms of cardiac ischemia due to coronary arteriosclerosis. In particular, relief of angina substantially reduces suffering, and the increased working and exercise capacity and less need for medication and hospital admissions contribute to improved healthrelated quality of life. Of particular importance is of course that SCS has no negative influence on the cardiac disease and does not mask signs of myocardial ischemia and infarction. Although it has not yet been convincingly demonstrated that SCS can replace surgical cardiac interventions, it is associated with less morbidity and it is an indispensable treatment option for the relatively large number of patients whose myocardial disease is not suitable for surgery. SCS is a form of treatment that is generally considered to represent a last resort therapy. This attitude has to be revised in view of the fact that multiple treatments that may have been tried, and failed, prior to SCS, often have prolonged the patient’s suffering. It should be emphasized that SCS is associated with virtually no serious complications and the relatively frequent
Spinal cord stimulation. techniques, indications and outcome
adverse, mostly device related effects are readily manageable. The implantation procedure is well tolerated by the patients and, even more important, it is a non-destructive, reversible intervention. For these reasons, SCS should in many cases be considered at an earlier phase of the disease, and not at an end-stage; it may well be tried before further attempts with drugs known to have troublesome side effects are initiated. Finally, it should be noted that in spite of the large number of publications on SCS there is still a lack of knowledge, or rather awareness, also among pain clinicians about what this treatment can offer. It can thus be concluded that SCS is still an underused form of neuromodulation.
References 1. Melzack R, Wall PD. Pain mechanisms: a new theory. Science 1965;150(699):971-9. 2. Wall PD, Sweet WH. Temporary abolition of pain in man. Science 1967;155:108-9. 3. Head H, Thompson T. The grouping of afferent impulses within the spinal cord. Brain 1906;29:537-741. 4. Mazars G, Roge R, et al. Results of the stimulation of the spinothalamic fasciculus and their bearing on the physiopathology of pain. Rev Prat 1960;103:136-8. 5. Noordenbos W. Pain. Amsterdam: Elsevier, (1959). 6. Zotterman Y. The nervous mechanism of touch and pain. Acta psychiat neurol 1939;14:91-7. 7. Dickenson AH. Gate control theory of pain stands the test of time. Br J Anaesth 2002;88(6):755-7. 8. Shealy CN, Mortimer JT, et al. Electrical inhibition of pain by stimulation of the dorsal columns: preliminary clinical report. Anesth Analg 1967;46:489-91. 9. Ray CD. Electrical stimulation: new methods for therapy and rehabilitation. Scand J Rehabil Med 1978;10(2):65-74. 10. Nashold BS, Somjen G, et al. Paresthesias and EEG potentials evoked by stimulation of the dorsal funiculi in man. Exp Neurol 1972;36:273-87. 11. Lindblom U, Meyerson BA. Influence on touch, vibration and cutaneous pain of dorsal column stimulation in man. Pain 1975;1:257-70. 12. Meyerson B, Linderoth B. Spinal cord stimulation: mechanisms of action in neuropathic and ischemic pain. In: Simpson B, editor. Electrical stimulation and relief of pain. New York: Elsevier, 2003. p. 161-82. 13. Linderoth B, Foreman R. Mechanisms of spinal cord stimulation in painful syndromes. Role of animal models. Pain Med 2006;7(Suppl 1):14-26.
137
14. Barolat G. Spinal cord stimulation for chronic pain management. Arch Med Res 2000;31(3):258-62. 15. Meyerson BA, Linderoth B. Spinal cord stimulation. In: Loeser JD, editor. Bonica’s management of pain. Philadelphia: Lippincott Williams & Wilkins, 2001. p. 1857-76. 16. Simpson B, Meyerson B, et al. Spinal cord and brain stimulation. In: McMahon S, Koltzenburg M, editors. Textbook of Pain. Amsterdam: Elsevier, 2005. p. 563-82. 17. Lee A Pititsis J. Spinal cord stimulation: indications and outcomes. Neurosurg Foucus 2006;21(6):1-6. 18. Sindou MP, Mertens P, et al. Predictive value of somatosensory evoked potentials for long-lasting pain relief after spinal cord stimulation: practical use for patient selection. Neurosurgery 2003;52(6):1374-83; discussion 1383-4. 19. Spiegelmann R, Friedman W. Spinal cord stimulation: a contemporary series. Neurosurgery 1991;28:65-71. 20. Kupers RC, Van den Oever R, et al. Spinal cord stimulation in Belgium: a nation-wide survey on the incidence, indications and therapeutic efficacy by the health insurer. Pain 1994;56:211-6. 21. Burchiel KL, Anderson VC, et al. Prognostic factors of spinal cord stimulation for chronic back and leg pain. Neurosurgery 1995;36:1101-11. 22. Dumoulin K, Devulder J, et al. A psychoanalytic investigation to improve the success rate of spinal cord stimulation as a treatment for chronic failed back surgery syndrome. Clin J Pain 1996;12(1):43-9. 23. North RB, Kidd DH, et al. Prognostic value of psychological testing in patients undergoing spinal cord stimulation: a prospective study. Neurosurgery 1996;39(2):301-10; discussion 310-1. 24. Simpson B, Bassett G, et al. Cervical spinal cord stimulation for pain: a report on 41 patients. Neuromodulation 2003;6:20-6. 25. Gybels J, et al. Neuromodulation of pain. A consensus statement. Europ J Pain 1998;2:203-11. 26. Lind G, Meyerson BA, et al. Implantation of laminotomy electrodes for spinal cord stimulation in spinal anesthesia with intraoperative dorsal column activation. Neurosurgery 2003;53(5):1150-3; discussion 1153-4. 27. Kemler MA, de Vet HC, et al. Effect of spinal cord stimulation for chronic complex regional pain syndrome Type I: five-year final follow-up of patients in a randomized controlled trial. J Neurosurg 2008;108(2):292-8. 28. Ohnmeiss DD, Rashbaum RF, et al. Prospective outcome evaluation of spinal cord stimulation in patients with intractable leg pain. Spine 1996;21(11):1344-50; discussion 1351. 29. Jadad AR, Cook DJ, et al. Methodology and reports of systematic reviews and meta-analyses: a comparison of cochrane reviews with articles published in paper-based journals. JAMA 1998;280(3):278-80. 30. Brainin M, Barnes M, et al. Guidance for the preparation of neurological management guidelines by EFNS scientific
2327
2328
137 31.
32.
33. 34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
Spinal cord stimulation. techniques, indications and outcome
task forces – revised recommendations 2004. Eur J Neurol 2004;11(9):577-81. Harbour R, Miller J. A new system for grading recommendations in evidence based guidelines. BMJ 2001;323 (7308):334-6. Barolat G, Oakley JC, et al. Epidural spinal cord stimulation with a multiple electrode paddle lead is effective in treating intractable low back pain. Neuromodulation 2001;4:59-66. De Andres J, Van Buyten JP. Neural modulation by stimulation. Pain Pract 2006;6(1):39-45. Van Buyten JP. Neurostimulation for chronic neuropathic back pain in failed back surgery syndrome. J Pain Symptom Manage 2006;31(4 Suppl):S25-9. Ohnmeiss DD, Rashbaum RF. Patient satisfaction with spinal cord stimulation for predominant complaints of chronic, intractable low back pain. Spine J 2001;1:358-63. Holsheimer J, Nuttin B, et al. Clinical evaluation of paresthesia steering with a new system for spinal cord stimulation. Neurosurgery 1998;42(3):541-7; discussion 547-9. Oakley J, Espinosa F, et al. Transverse tripolar spinal cord stimulation: results of an international multicenter study. Neuromodulation 2006;9:192-203. Mailis-Gagnon A, Furlan A, et al. Spinal cord stimulation for chronic pain (review). Cochrane Database Syst. Rev. 2004;3:CD003783. Taylor RS, Van Buyten JP, et al. Spinal cord stimulation for chronic back and leg pain and failed back surgery syndrome: a systematic review and analysis of prognostic factors. Spine 2005;30(1):152-60. North RB, Kidd DH, et al. A prospective, randomized study of spinal cord stimulation versus reoperation for failed back surgery syndrome: initial results. Stereotact Funct Neurosurg 1994;62:267-72. North RB, Kidd DH, et al. Spinal cord stimulation versus repeated lumbosacral spine surgery for chronic pain: a randomized, controlled trial. Neurosurgery 2005;56 (1):98-106; discussion 106-7. Kumar K, Taylor RS, et al. Spinal cord stimulation versus conventional medical management for neuropathic pain: a multicentre randomised controlled trial in patients with failed back surgery syndrome. Pain 2007;132(1–2):179-88. Attal N, Cruccu G, et al. EFNS guidelines on pharmacological treatment of neuropathic pain. Eur J Neurol 2006;13(11):1153-69. Cruccu G, Aziz TZ, et al. EFNS guidelines on neurostimulation therapy for neuropathic pain. Eur J Neurol 2007;14(9):952-70. Kemler MA, Barendse GAM, et al. Spinal cord stimulation in patients with reflex sympathetic dystrophy. N Eng J Med 2000;343:618-24. Kemler MA, De Vet HC, et al. The effect of spinal cord stimulation in patients with chronic reflex sympathetic dystrophy: two years’ follow-up of the randomized controlled trial. Ann Neurol 2004;55(1):13-8.
47. Kemler MA, de Vet HC, et al. Spinal cord stimulation for chronic reflex sympathetic dystrophy – five-year follow-up. N Engl J Med 2006;354(22):2394-6. 48. Taylor RS. Spinal cord stimulation in complex regional pain syndrome and refractory neuropathic back and leg pain/failed back surgery syndrome: results of a systematic review and meta-analysis. J Pain Symptom Manage 2006;31(4 Suppl):S13-9. 49. Taylor RS, Van Buyten JP, et al. Spinal cord stimulation for complex regional pain syndrome: a systematic review of the clinical and cost-effectiveness literature and assessment of prognostic factors. Eur J Pain 2006;10 (2):91-101. 50. Grabow SG, Tella PK, et al. Spinal cord stimulation for complex regional syndrome: an evidence-based medicine review of the literature. Clin J Pain 2003;19:371-83. 51. Kumar K, Nath RK, et al. Spinal cord stimulation is effective in the management of reflex sympathetic dystrophy. Neurosurgery 1997;40:503-8. 52. Harke H, Gretenkort P, et al. Spinal cord stimulation in sympathetically maintained complex regional pain syndrome type I with severe disability. A prospective clinical study. Eur J Pain 2005;9(4):363-73. 53. Medical Advisory Secreteriat, Ontario Ministry of Health. Spinal cord stimulation. Available at: http://www.health. gov.on.ca/english/providers/program/ohtac/tech/reviews/ sum_scs_030105.html, 2005. 54. Harke H, Gretenkort P, et al. Spinal cord stimulation in postherpetic neuralgia and in acute herpes zoster pain. Anesth Analg 2002;94(3):694-700. 55. Middleton P, Simpson B, et al. Spinal cord stimulation: an accelerated systematic review. Report no. 43. (AERSNIP-S), Australian safety and efficacy register of new interventional procedures, 2003. 56. North RB, Kidd DH, et al. Spinal cord stimulation for chronic, intractable pain: experience over two decades. Neurosurgery 1993;32:384-95. 57. Lazorthes Y, Siegfried J, et al. La stimulation me´dullaire chronique dans le traitement des douleurs neuroge`nes. Neurochirurgie 1995;41:73-88. 58. Kumar K, Hunter G, et al. Spinal cord stimulation in treatment of chronic benign pain: challenges in treatment planning and present status, a 22-year experience. Neurosurgery 2006;58(3):481-96; discussion 481-96. 59. Cook WW, Oygar A, et al. Vascular disease of extremities: electrical stimulation of spinal cord and posterior roots. N Y State J Med 1976;76:366-68. 60. Augustinsson LE, Linderoth B, et al. Spinal cord stimulation in cardiovascular disease. In: Gildenberg P editor. Neurosurgery clinics of North America. Vol. 6: 1995. p. 157–66. 61. Mannheimer C, Carlsson CA, et al. Transcutaneous electrical nerve stimulation in severe angina pectoris. Eur Heart J 1982;3:297-302. 62. Murphy DF, Giles KE. Dorsal stimulation for pain relief from intractable angina. Pain 1987;28:365-8.
Spinal cord stimulation. techniques, indications and outcome
63. Mannheimer C, Augustinsson LE, et al. Epidural spinal electrical stimulation in severe angina pectoris. Br Heart J 1988;59:56-61. 64. Eliasson T, Augustinsson LE, et al. Spinal cord stimulation in severe angina pectoris – presentation of current studies, indications and clinical experience. Pain 1996;65:169-79. 65. Gybels JM, Sweet WH. Neurosurgical treatment of persistent pain. Basel: Karger, (1989). 66. Meyerson BA. Electric stimulation of the spinal cord and brain. In: Bonica JJ editor. The management of pain. Philadelphia: Lea & Febiger, 1990. p. 1862-77. 67. Linderoth B. Spinal cord stimulation in ischemia and ischemic pain. In: Horsch S. and Claeys L., editors. Spinal Cord Stimulation: An Innovative Method in the Treatment of PVD and Angina. Steinkopff Verlag, Darmstadt: p. 19-35. 68. Klomp HM, Spincemaille GH, et al. Spinal-cord stimulation in critical limb ischaemia: a randomised trial. ESES study group. Lancet 1999;353(9158):1040-4. 69. Ubbink DT, Spincemaille GH, et al. Microcirculatory investigations to determine the effect of spinal cord stimulation for critical leg ischemia: the Dutch multicenter randomized controlled trial. J Vasc Surg 1999;30(2):236-44. 70. Gersbach P, Hasdemir MG, et al. Discriminative microcirculatory screening of patients with refractory limb ischaemia for dorsal column stimulation. Eur J Vasc Endovasc Surg 1997;13(5):464-71. 71. Ubbink DT, Gersbach PA, et al. The best TcpO(2) parameters to predict the efficacy of spinal cord stimulation to improve limb salvage in patients with inoperable critical leg ischemia. Int Angiol 2003;22(4):356-63. 72. Ubbink DT, Vermeulen H. Spinal cord stimulation for non-reconstructable chronic critical leg ischaemia. Cochrane Database Syst Rev 2005;3:CD004001. 73. Ubbink DT, Vermeulen H. Spinal cord stimulation for critical leg ischemia: a review of effectiveness and optimal patient selection. J Pain Symptom Manage 2006;31 (4 Suppl):S30-5. 74. Klomp HM, Steyerberg EW, et al. Spinal cord stimulation is not cost-effective for non-surgical management of critical limb ischaemia. Eur J Vasc Endovasc Surg 2006;31(5):500-8. 75. Amann W, Berg P, et al. Spinal cord stimulation in the treatment of non-reconstructable stable critical leg ischaemia: results of the European peripheral vascular disease outcome study (SCS-EPOS). Eur J Vasc Endovasc Surg 2003;26(3):280-6. 76. Jivegard LE, Augustinsson LE, et al. Effects of spinal cord stimulation (SCS) in patients with inoperable severe lower limb ischaemia; a prospective randomised controlled study. Eur J Vasc Endovasc Surg 1995;9:421-5. 77. Claeys L, Horsch S. Transcutaneous oxygen pressure as predictive parameter for ulcer healing in endstage vascular patients treated by spinal cord stimulation. Int Angiology 1996;15:344-9.
137
78. Spincemaille GH, Klomp HM, et al. Pain and quality of life in patients with critical limb ischaemia: results of a randomized controlled multicentre study on the effect of spinal cord stimulation. ESES study group. Eur J Pain 2000;4(2):173-84. 79. Linderoth B, Meyerson BA. Spinal cord stimulation in limb ischemia – time for revival? Eur J Pain 2000; 4(3):317-9. 80. Robaina FJ, Dominguez M, et al. Spinal cord stimulation for relief of chronic pain in vasospastic disorders of the upper limbs. Neurosurgery 1989;24:63-7. 81. Francaviglia N, Silvestro C, et al. Spinal cord stimulation in the treatment of progressive systemic sclerosis and Raynaud’s syndrome. Brit J Neurosurgery 1994;8:567-71. 82. Donas KP, Schulte S, et al. The role of epidural spinal cord stimulation in the treatment of Buerger’s disease. J Vasc Surg 2005;41(5):830-6. 83. Gonzales-Darder JM, Gonzales-Martinez V. High cervical spinal cord stimulation for unstable angina pectoris. Ster Funct Neurosurg 1991;56:20-7. 84. DeJongste MJL, Hautvast RWM, et al. Efficacy of spinal cord stimulation as adjuvant therapy for intractable angina pectoris: a prospective randomized study. J Am Coll Cardiol 1994;23:1592-7. 85. Jessurun GAJ, DeJongste MJL, et al. Current views on neurostimulation in the treatment of cardiac ischemic syndromes. Pain 1996;66:109-16. 86. Augustinsson LE, Linderoth B, et al. Spinal cord stimulation in peripheral vascular disease and angina pectoris. In: Gildenberg P, Tasker R, editors. Textbook of stereotactic and functional neurosurgery. New York: McGrawHill; 1997. p. 1973-8. 87. Mannheimer C, Eliasson T, et al. Effects of spinal cord stimulation in angina pectoris induced by pacing and possible mechanisms of action. Brit Med J 1993;307:477-80. 88. Sanderson JE, Brooksby P, et al. Epidural spinal electrical stimulation for severe angina: a study of its effects on symptoms, exercise and degree of ischaemia. Eur Heart J 1992;13:628-33. 89. Hautvast RW, Brouwer J, et al. Effect of spinal cord stimulation on heart rate variability and myocardial ischemia in patients with chronic intractable angina pectoris-a prospective ambulatory elecrocardiographic study. Clin Cardiol 1998;21:33-8. 90. TenVaarwerk IA, Jessurun GA, et al. Clinical outcome of patients treated with spinal cord stimulation for therapeutically refractory angina pectoris. The working group on neurocardiology. Heart 1999;82(1):82-8. 91. Andrell P, Ekre O, et al. Cost-effectiveness of spinal cord stimulation versus coronary artery bypass grafting in patients with severe angina pectoris–long-term results from the ESBY study. Cardiology (2003). 99(1): 20-4. 92. Di Pede F, Lanza GA, et al. Immediate and longterm clinical outcome after spinal cord stimulation for refractory stable angina pectoris. Am J Cardiol 2003;91 (8):951-5.
2329
2330
137
Spinal cord stimulation. techniques, indications and outcome
93. Eddicks S, Maier-Hauff K, et al. Thoracic spinal cord stimulation improves functional status and relieves symptoms in patients with refractory angina pectoris: the first placebo-controlled randomised study. Heart 2007;93 (5):585-90. 94. Mannheimer C, Eliasson T, et al. Electrical stimulation versus coronary artery bypass surgery in severe angina pectoris: the ESBY study. Circulation 1998;97:1157-63. 95. Ekre O, Eliasson T, et al. Long-term effects of spinal cord stimulation and coronary artery bypass grafting on quality of life and survival in the ESBY study. Eur Heart J 2002;23(24):1938-45. 96. Ekre O, Norrsell H, et al. Temporary cessation of spinal cord stimulation in angina pectoris-effects on symptoms and evaluation of long-term effect determinants. Coron Artery Dis 2003;14(4):323-7. 97. Bo¨rjesson M, Andrell P, et al. Spinal cord stimulation in severe angina pectoris - A systematic review based on the Swedish Council on Technology assessment in health care report on long-standing pain. Pain 2008;140 (3):501-8. 98. Andersen C, Hole P, et al. Does pain relief with spinal cord stimulation for angina pectoris conceal myocardial infarction? Br Heart J 1994;71:419-21. 99. Murray S, Carson KG, et al. Spinal cord stimulation significantly decreases the need for acute hospital admission for chest pain in patients with refractory angina pectoris. Heart 1999;82(1):89-92. 100. Eliasson T, Jern S, et al. Safety aspects of spinal cord stimulation in severe angina pectoris. Coron Artery Dis 1994;5:845-50. 101. Kemler MA, Furnee CA. Economic evaluation of spinal cord stimulation for chronic reflex sympathetic dystrophy. Neurology 2002;59(8):1203-9. 102. Kumar K, Malik S, et al. Treatment of chronic pain with spinal cord stimulation versus alternative therapies: costeffectiveness analysis. Neurosurgery 2002;51(1):106-15; discussion 115-6.
103. North RB, Kidd D, et al. Spinal cord stimulation versus reoperation for failed back surgery syndrome: a cost effectiveness and cost utility analysis based on a randomized, controlled trial. Neurosurgery 2007;61(2):361-8; discussion 368-9. 104. Cameron T. Safety and efficacy of spinal cord stimulation for the treatment of chronic pain: a 20-year literature review. J Neurosurg 2004;100(3 Suppl Spine):254-67. 105. Yu W, Maru F, et al. Spinal cord stimulation for refractory angina pectoris: a retrospective analysis of efficacy and cost-benefit. Coron Artery Dis 2004;15 (1):31-7. 106. Burchiel KJ, Anderson VC, et al. Prospective, multicenter study of spinal cord stimulation for relief of chronic back and extremity pain. Spine 1996;21:2786-2794. 107. Kumar K, Taylor RS, et al. The effects of spinal cord stimulation in neuropathic pain are sustained: a 24-month follow-up of the prospective randomized controlled multicenter trial of the effectiveness of spinal cord stimulation. Neurosurgery 2008;63(4):762-70; discussion 770. 108. McNab D, Khan SN, et al. An open label, single-centre, randomized trial of spinal cord stimulation vs. percutaneous myocardial laser revascularization in patients with refractory angina pectoris: the SPiRiT trial. Eur Heart J 2006;27(9):1048-53. 109. North RB, Kidd DH, et al. Spinal cord stimulation electrode design: a prospective, randomized, controlled trial comparing percutaneous with laminectomy electrodes: part II-clinical outcomes. Neurosurgery 2005;57 (5):990-6; discussion 990-6. 110. North RB, Ewend MG, et al. Failed back surgery syndrome: 5-year follow-up after spinal cord stimulator implantation. Neurosurgery 1991;28(5):692-9. 111. Petrakis IE, Sciacca V. Does autonomic neuropathy influence spinal cord stimulation therapy success in diabetic patients with critical lower limb ischemia? Surg Neurol 2000;53(2):182-8; discussion 188-9.
134 Surgical Dorsal Root Entry Zone Lesions for Pain M. P. Sindou
Introduction Until the sixties, pain pathways were defined as the pathways whose interruption produced analgesia: that is the sensory nerves, the dorsal roots and ganglia, the neo-and paleo-reticulo-thalamic tracts, the relay areas within the thalamus, and cortical representation(s) [1]. In the 1960s, the Gate Control theory [2] drew neurosurgeons’ attention to the dorsal horn as the first important level of modulation for pain sensation. This area was then considered a possible target for pain surgery through spinal cord stimulation [3] and destructive surgery in the Dorsal Root Entry Zone (DREZ) [4,5]. The DREZ was defined as an entity including the central portion of the dorsal rootlet, the most medial part of the tract of Lissauer, and the most dorsal layers (I–V) of the dorsal horn, where the afferent fibers synapse with the cells of the sensory spino-reticulo-thalamic ascending pathways (> Figure 134‐1 left). The first attempts at DREZ-lesioning were performed in March 1972 at the Neurological Institute Pierre Wertheimer in Lyon for localized malignant pain, namely the Pancoast-Tobias syndrome, using the microsurgical techniques to perform a destructive lesion in the ventro-lateral region of the DREZ. Because the first results were encouraging [4], the procedure was attempted in patients with neuropathic pain (associated with paraplegia in December 1972, secondary to amputation in July 1973 and brachial plexus avulsion in January 1974). Soon after (September 1974), Nashold and his group in Duke University started to develop DREZlesioning using RadioFrequency-thermocoagulation [6,7], especially for #
Springer-Verlag Berlin/Heidelberg 2009
pain related to brachial plexus avulsion. More recently, DREZ-procedures were performed with a Laser beam as the lesion-maker by Levy et al. [8] and Powers et al. [9], and with an ultrasound probe by Kandel et al. [10] and Dreval [11], also for pain caused by plexus brachial avulsion.
Anatomical Bases Each dorsal root divides into 4–10 rootlets according to metameres [12,13]. Each rootlet, of 0.25– 1.50 mm in diameter according to levels, can be considered an anatomical-functional entity, that is a root in miniature [4,5]. In the DREZ there is a spatial segregation of the afferent fibers according to their size and destination, with the fine fibers regrouping in the lateral region of the DREZ (> Figure 134‐1 right). The tract of Lissauer (TL) is situated dorso-laterally to the dorsal horn. It includes: (1) a medial part, which the small afferents enter and where they trifurcate to reach the dorsal horn, either directly or through a few metamere ascending or descending pathway. This part transmits the excitatory effects of each dorsal root to the adjacent segments [14,15] and (2) a lateral part through which a large number of longitudinal endogenous propriospinal fibers interconnect different levels of the substantia gelatinosa (SG). This part conveys the inhibitory influences of the SG into the neighboring metameres [15]. Most of the fine nociceptive afferents enter the dorsal horn through the TL medial part and the dorsal aspect of SG. Ramon y Cajal’s recurrent collaterals of large lemniscal fibers [16] approach the dorsal horn
2270
134
Surgical dorsal root entry zone lesions for pain
. Figure 134‐1 (left) Rexed’s lamination (I–IX). Transverse hemisection of the spinal cord (at the lower cervical level) with myelin stained by luxol-fuschine, showing the myelinated rootlet afferents that reach the dorsal column. The small arrow designates the pial ring of the dorsal rootlet (diameter, 1 mm). The large arrows show the MDT target. DC, dorsal column; P, pyramidal tract; Sg, substantia gelatinosa; tl, tract of Lissauer. (right) Schematic representation of the Dorsal Root Entry Zone (DREZ) area and the target of micro-DREZotomy (MDT). Upper part, each rootlet can be divided (owing to the transition of its glial support) into a peripheral and a central segment. The transition between the two segments is at the pial ring (PR), which is located approximately 1 mm outside the penetration of the rootlet into the dorsolateral sulcus. Peripherally, the fibers are mixed together. As they approach the PR, the fine fibers (considered nociceptive) more toward the rootlet surfaces. In the central segment, they group in the ventrolateral portion of the DREZ and enter the dorsal horn (DH) through the tract of Lissauer (TL). The large myotatic fibers (myot) are situated in the middle of the DREZ, whereas the large lemniscal fibers are located dorsomedially. Lower part, Schematic data on DH circuitry. Note the monosynaptic excitatory arc reflex, the lemniscal influence on a DH cell and an interneuron (IN), the fine fiber excitatory input onto DH cells, and the IN, the origins in layer I and Layers IV–VII of the anterolateral pathways (ALP), and the projection of the IN onto the motor neuron (MN). (DC, dorsal column). Rexed laminae are marked from I to VI. The MDT (arrowhead) cuts most of the fine and myotatic fibers and enters the medial (excitatory) portion of the LT and the apex of the dorsal horn. It should preserve most lemniscal presynaptic fibers, the lateral (inhibitory) portion of TL, and most of the DH (from [4])
through the ventro-medial aspect of SG [17]. Because a number of dendrites of the spino-reticulothalamic (SRT) cells make synaptic connections with the primary afferents inside the SG layers, SG exerts a strong segmental modulating effect on the
nociceptive input. When the large lemniscal afferents in peripheral nerves or dorsal roots are altered, there is a reduction in the inhibitory control of the dorsal horn [18]. This situation presumably results in excessive firing of the dorsal horn
Surgical dorsal root entry zone lesions for pain
134
. Figure 134‐2 Dorsal horn microelectrode recordings in humans. The electrode was a floating tungsten-glass microelectrode, implanted intraoperatively by free-hand under the operative microscope approximately 5 mm in depth (i.e., in laminae IV–VI). Upper trace, normal activity. Recordings in a non-deafferented dorsal horn at the lumbosacral level (spastic patient). Left: almost no spontaneous activity (three spikes at random). Middle: spike burst discharges (arrows) evoked by regular light tactile stimulation of the corresponding dermatome. Right: Spike burst discharges evoked by electrical stimulation of the corresponding peripheral nerve. Lower trace, deafferentation hyperactivity. Recordings in the L5 cord segment of a patient with pain due to a traumatic section of the hemicauda equina from root L4 to S4. Left: spontaneous activity of the recorded unit: continuous, regular, high frequency discharge. Middle: the unit during light tactile stimulation of the L4 to S1 dermatomas (arrow). Right: during electrical stimulation of the tibial nerve (the arrows are two consecutive stimuli). Note the continuous regular discharge that remains unaltered. The vertical bars are 50 mV; the horizontal bars are 100 ms) (from [19])
neurons (> Figure 134‐2). This phenomenon, thought to be at the origin of deafferentation pain, has been identified in patients by electrophysiological recordings [19–22] and reproduced in animal experiments [23–25]. Clinical and electrophysiological expression of experimental deafferentation pain can be alleviated by dorsal root entry zone lesions in rats [25]. Destruction of these hyperactive neurons should suppress the abnormal impulses generated in the spino-reticulo-thalamic pathways. Neurotransmitters generating pain should also be favorably modified by destruction of dorsal horn neurons [26].
Surgical Techniques of Microsurgical DREZotomy Working in the DREZ requires knowledge of the microsurgical gross anatomy of the spinal
cord and spinal roots [12,13] as well as of the internal morphologic anatomy of the spinal cord (> Figure 134‐3). Important is to avoid damaging the neighboring anatomical structures, and consequently prevent neurological deficits. There are several different options for performing DREZlesions. The one that we developed and named Microsurgical DREZotomy (MDT) has been described in detail in previous publications [12,13,27,28].
Principles of DREZotomy DREZotomy consists of a longitudinal opening incision of the dorsolateral sulcus, performed ventrolaterally at the entrance of the rootlets into the sulcus, and of microbipolar coagulations, performed inside the sulcus, down to the apex of the dorsal horn, continuously along all
2271
2272
134
Surgical dorsal root entry zone lesions for pain
. Figure 134‐3 Variations of shape, width and depth of the DREZ area, according to the spinal cord level (from top to bottom: Cervical n 7, Thoracic n 5, Lumbar n 4, Sacral n 3). The axis of the dorsal horn in relation to the sagittal plan crossing the dorso-lateral sulcus will condition the angulation of the DREZotomy. According to 82 measurements performed by Young (personal data), the mean DREZ angle is 30 degrees at C6, 26 at Th4, 37 at Th12 and 36 at L3. The site and extent of the DREZ lesion will also be conditioned by the shape, width and depth of Lissauer tract and dorsal horn. Note how, at the thoracic level, Lissauer’s tract is narrow and dorsal horn deep. It is easy to understand that DREZ lesions especially at this level can be dangerous for the corticospinal tract and the dorsal column
Surgical dorsal root entry zone lesions for pain
the spinal cord segments selected for surgery. The lesion, which penetrates the lateral part of the DREZ and the medial part of tract of Lissauer, extends to the apex of the dorsal horn, which can be recognized by its brown-grey color. The average lesion is 2–3 mm deep, is made at a 35-degree angle medially and ventrally, and is presumed to destroy preferentially the nociceptive fibers grouped in the lateral bundle of the dorsal rootlets and the excitatory medial part of the tract of Lissauer. The upper layers of the dorsal horn (I–V layers of Rexed’s classification) are also destroyed if microbipolar coagulations are performed inside the dorsal horn. The procedure is presumed to (partially) preserve the inhibitory structures of the DREZ (i.e., the lemniscal fibers reaching the dorsal horn and the substantia gelatinosa propriospinal interconnecting fibers running through the lateral part of the tract of Lissauer (> Figure 134-1). The method, named microsurgical DREZotomy (MDT), was conceived to prevent complete abolition of tactile and proprioceptive sensations and to avoid deafferentation phenomena [29]. The depth and extent of the lesion depend on the degree of the desired therapeutic effect and on the preoperative sensory and functional status of the patient.
Surgical Procedures Surgery is performed with the patient under general anesthesia, with an initial short-lasting curarization to allow intraoperative observation of motor responses to bipolar electrical stimulation of the nerve roots. Stimulated ventral roots have a motor threshold at least three times lower than the dorsal roots. Standard microsurgical techniques are used with 10–25 times magnification. Special microinstruments for MDT have been made by Leibinger-Fischer (Freiburg, West Germany).
134
Operative Procedure at the Cervical Level (When Roots are Intact) The prone position with the head and neck flexed in the ‘‘concorde’’ position has the advantage of avoiding brain collapse caused by cerebrospinal fluid depletion. The head is fixed with a three-pin head holder. The level of laminectomy is determined after identification of the prominent spinous process of C2 by palpation. For unilateral DREZ-surgery, a hemilaminectomy, generally from C4 to C7, with preservation of the spinous processes, allows sufficient exposure to the posterolateral aspect of the cervical spinal cord segments that correspond to the upper limb innervation, that is, the rootlets of C5 to T1 (> Figure 134‐4). When roots are present, dura and arachnoid are opened longitudinally. Then the exposed roots are dissected free by separating the tiny arachnoid filaments that bind them to each other, to the arachnoid sheath and to the cord pia mater. The radicular vessels are preserved. Each ventral and dorsal root from C4 to T1 is electrically stimulated at the level of its corresponding foramen to identify precisely its muscular innervation and its functional value. Stimulated ventral roots have a motor threshold at least 3 times lower than the dorsal roots. Responses are in the diaphragm for C4 (the response is palpable below the lower ribs), in the shoulder abductors for C5, in the elbow flexors for C6, in the elbow and wrist extensors for C7, and in the muscles intrinsic of the hand for C8 and T1. Microsurgical lesions are performed at the selected levels, that is, those that correspond to the pain territory. The technique is summarized and illustrated in > Figure 134-4. The incision is made with a microknife (razor blade in a blade-holder or ophtalmologic micro-scalpel). Then microcoagulations are made in a ‘‘chain’’ (i.e., dotted) manner. Each microcoagulation is
2273
2274
134
Surgical dorsal root entry zone lesions for pain
. Figure 134‐4 MDT technique at the cervical level. Left: Exposure of the right dorso-lateral aspect of the cervical cord at C6. The rootlets of the selected dorsal root(s) are displaced dorsally and medially with a hook or a microsucker to obtain access to the ventro-lateral aspect of the DREZ in the dorso-lateral sulcus. Center: Then an incision – 2 mm in depth, at 35 degrees ventrally and medially – is made with a microknife in the lateral border of the dorso-lateral sulcus. Right: Microcoagulations are then performed, down to the apex of the dorsal horn, in a dotted manner, using a sharp graduated bipolar microforceps
performed – under direct magnified vision – by short-duration (a few seconds), low intensity, bipolar electrocoagulation, with a specially designed sharp bipolar forceps incremented in millimeters. The depth and extent of the lesion depend on the desired therapeutic effect and the preoperative status of the limb. If the laxity of the root is sufficient, the incision is performed – continuously – in the dorsolateral sulcus, ventrolaterally along all of the rootlets of the targeted root, thus accomplishing a sulco-myelotomy. If not, a partial ventrolateral section is made successively on each rootlet of the root after the surgeon has isolated each one by separating the tiny arachnoid membranes that hold them together.
Operative Procedure at the Lumbo-sacral Level (When Spinal Cord and Roots are Intact) When roots are present, the patient is positioned prone on thoracic and iliac supports and the
head placed 20 cm lower than the level of the surgical wound to minimize the intraoperative loss of cerebrospinal fluid. The desired vertebral level is identified by palpation of the spinous processes or, if this is difficult, by lateral x-ray study that includes the S1 vertebra. Interspinous levels identified by a needle can then be marked with a droplet a nontoxic dye (methylene blue). A laminectomy, either bilateral or unilateral, according to pain topography, is performed from T11 to L1 (or L2). The dura and arachnoid are opened longitudinally, and the filum terminale is isolated. Identification of roots is performed by electrical stimulation (> Figure 134‐5). The L1 and L2 roots are easily identified at their penetration into their respective dural sheaths. Electrical stimulation of L2 produces a response of the iliopsoas and adductor muscles. Identification of L3–L5 is difficult for many reasons, (1) the exit through their respective dural sheaths is caudal to the exposure, (2) the dorsal rootlets enter the sulcus along an uninterrupted line, (3) the ventral roots are hidden in front of the dentate ligament, and (4) the motor
Surgical dorsal root entry zone lesions for pain
134
. Figure 134‐5 MDT technique at the lumbosacral level. Top drawings (upper left): exposure of the conus medullaris through a Th11 to L1 laminectomy. Upper right: approach of the dorso-lateral sulcus, (on the left side in this example). For doing so, the dorsal rootlets are displaced dorsally and medially to obtain proper access to the ventrolateral aspect of the DREZ. Operative view (lower left): the rootlets of the selected dorsal roots (on left side) are retracted dorso-medially and held with a (specially designed) ball-tip micro-sucker, used as a small hook, to gain access to the ventro-lateral part of the DREZ. After division of the fine arachnoidal filaments sticking the rootlets together with the pia-mater with curved sharp micro-scissors (not shown), the main arteries running along the dorso-lateral sulcus are dissected and preserved, whilst the smaller ones are coagulated with a sharp bipolar micro-forceps (not shown). Then, a continuous incision is performed using a micro-knife, made with a small piece of razor blade inserted within the striated jaws of a curved razor-blade-holder. The cut is – on average – at a 45 degree angle and to a depth of 2 mm. Operative view, same case (lower right): the surgical lesion is completed by doing micro-coagulations under direct magnified vision, at a low intensity, inside the postero-lateral sulcomyelotomy down to the apex of the dorsal horn. These microcoagulations are made all along the segments of the cord selected to be operated on by means of the special sharp bipolar forceps, insulated except at the tip over 5 mm and graduated every millimeter
responses in the leg to stimulation of the roots are difficult to observe intraoperatively because of the patients’ prone position. Stimulation of L3 produces a preferential response in the adductors and quadriceps, of L4 in quadriceps, and of L5 in the anterior tibialis. Stimulation of the S1 dorsal root produces a motor response of the gastrocnemiussoleus group that can be confirmed later, by repeatedly checking the Achilles ankle reflex before, during, and after MDT. Stimulation of the S2–S4 dorsal roots (or better, directly, the corresponding spinal cord segments at the DREZ) can be assessed by recording of the motor vesicle or anal response by use of cystomanometry, rectomanometry, or electromyography of the anal sphincter (or simply with a gloved finger into the rectum). Because neurophysiologic investigations
are time-consuming to perform in the operative room, we have found that measurements at the conus medullaris can be sufficient in the patients who already have severe preoperative impairment of their vesicoanal functions. These measurements, based on human postmortem anatomic studies, have shown that the landmark between the S1 and S2 segments is situated around 30 mm above the exit from the conus of the tiny coccygeal root [4,12,13]. MDT at the lumbosacral levels has the same principles as the ones at the cervical level. The technique is summarized and illustrated in > Figure 134‐5. At the lumbosacral level, MDT is difficult and possibly dangerous because of the rich vasculature of the conus. The dorsolateral spinal artery courses along the dorsolateral
2275
2276
134
Surgical dorsal root entry zone lesions for pain
sulcus. Its diameter is 0.1–0.5 mm, and it is fed by the posterior radicular arteries and joins caudally with the descending anterior branch of the Adamkiewicz artery through the conus medullaris anastomotic loop of Lazorthes. This artery has to be preserved by being freed from the sulcus.
Summary of Other DREZ Lesion Procedures In September 1974, Nashold and colleagues started to develop DREZ lesions using the radiofrequency (RF)-thermocoagulation as the lesion maker in the substantia gelatinosa of the dorsal horn first [6] and later on in the whole DREZ [7], especially for pain resulting from brachial plexus avulsion. More recently, DREZ procedures were performed with the use of the laser by Levy et al. [8] and Powers et al. [9], and with the use of an ultrasound probe by Kandel et al. [10] and Dreval [11], also for pain caused by plexus brachial avulsion. These various technical modalities, which are theoretically all directed to the DREZ, do not have exactly the same anatomic target and consequently the same sensory effects. (1) The RF-thermocoagulation procedure is performed with an electrode implanted through the pia mater into the dorsal horn; the coagulation involves the whole dorsal horn and has an ovoid shape. (2) DREZ lesions made with the laser, mostly the carbon dioxide laser, are more superficial and have a V shape; they are often accompanied by small infarcts because of the coagulation of the vessels located at the DREZ. (3) The ultrasonic DREZ procedure has been almost exclusively used for brachial plexus avulsion pain; it has the particularity to evacuate the spongy and gliotic tissue situated in the dorsal horn apex. All these methods destroy the entire DREZ and dorsal horn structures; they do not allow preservation of sensory functions in the operated areas if present preoperatively.
Pain After Brachial Plexus Injury Indications Brachial plexus injuries are diversely associated with secondary chronic pain according to the preganglionic or the post-ganglionic predominance of the lesions. The incidence of pain in literature review [30] was less than one-third for postganglionic locations, as opposed to 90% when location was predominantly preganglionic. Almost all of our patients who underwent DREZ surgery belonged to the last category [30]. All surgeons experienced in pain surgery and using DREZ procedures whatever their modality agree that DREZ operations are effective for pain developing after brachial plexus injury. We do not limit the DREZ lesion to the avulsed segments but extend it to the adjacent remaining roots, when they are atrophic or even of a grayish color, especially if their level corresponds to the painful territory. The long-term results in our series are concordant with those of the literature (> Table 134‐1). The group at Duke University reported a success rate of 54% in a series of 39 patients operated on with the RF-thermocoagulation technique [33]. The Queen Square group in London, in their series of 44 cases treated also with thermocoagulation, obtained a 68% success rate [36]. Rath et al. [37], in Germany reported a 61% success rate in their 13 cases, and Samii et al. [38], also from Germany, a success rate of 63% in their important 47 patient series also treated with RF-thermocoagulation. Dreval [11] reported a 87% rate of success in his series of 127 patients in whom the DREZ lesions were performed using a special small ultrasonic probe. Analysis of the literature concerning postoperative complications with RF or Laser DREZ procedures for brachial plexus avulsion revealed (more or less severe) corticospinal and/or dorsal column deficits in 0–10% of the patients with Laser and in as many as 50% of the patients with the
14
47
21 55
RF-Th
RF-Th
Microsurgery Microsurgery
0.5–27 y (7.7 y) 1–42 y (9 y)
0.5–40 y
1–30 y (7.3 y)
2–25 y (10.8 y)
1–24 y (5.8 y)
Pain durationrange-(mean)
Note: RF-Th, radiofrequency thermocoagulation; p, permanent
124 44
RF-Th, CO2Laser
UltraS RF-Th
18, 4
RF-Th RF-Th
7
19 39
RF-Th
RF-Th
18
RF-Th
Nashold and Ostdahl [7] Thiebault et al. [31] Ishijima [32] Friedman et al. [33] Young et al. [34]
Kumagai et al. [35] Dreval [11] Thomas and Kitchen [36] Rath et al. [37] Samii et al. [38] Prestor [39] Sindou [30]
Number of patients 18
Technique
Reference
2–10 y (5.6 y) 1–27 (6 y)
2–18 y (14 y)
3–12 y (6.2 y)
– (4 y) 1–12 y (5 y)
(4.2 y)
1–5 y (4 y)
– (1.7 y) 1–10 y
– (6 y)
Follow-uprange (mean) 1–4 y (1.8 y)
25%
17.8% 13%
17%
16.7%
47.6% 38%
<63%>
61%
87% 68%
28.5%
33.3% 31%
15%
11%
14.2%
RF 75%, Laser 50%
82.4% 54%
83%
55.6%
23%
21%
14.3% 4.8% <31%>
<37%>
13%
57.1%
RF 25%, Laser 50%
0% 33%
0%
27.7%
14% (5%)p 4.7% 1..8%
10% 22% (18%)p <43%>
RF 6%, Laser 10% 14.2%
62% 40%
5.6%
50%
14% 3.6%
15% 18% (9%)p
RF 3%, Laser 5% 71.4
40%
38.9%
72.2%
Sensory
Motor
50%
100%
75%
Complications
Results
. Table 134‐1 Review of the literature: surgery in the Dorsal Root Entry Zone for brachial plexus avulsion
Meningitis: 2 pts
2 subdural hematomas
1 death
New pain: 10%
New pain: 28.5%
Myelopathy: 1.3%, Urin dist.: 1.3%
New pain: 5.3%
CSF: 2 pts
Others
Surgical dorsal root entry zone lesions for pain
134 2277
2278
134
Surgical dorsal root entry zone lesions for pain
RF-technique, according to the reviewed series; there were fewer complications and side-effects with MDT [30]. This seems to be due to the better accuracy and selectivity in the lesioning process with the microsurgical technique; with this method, the lesion is performed under magnified direct vision of the microscope and through an opening of the dorso-lateral sulcus itself.
Anatomical Data Pre-operative Appraisal of Radicular Lesions At present, lesions can be preoperatively appraised through imaging with iodine computerized tomography myelography and/or T2-spinal MR Imaging, electroneuromyography, and SSEPs. During surgical repair, the location and type of lesion can be determined by recording SSEPs and nerve Action Potentials. Nevertheless, all of these studies do not prevent interpretation errors. Pseudomeningoceles of the radicular dural sleeve are considered to be classic indirect signs of root avulsion. In our series, pseudomeningoceles occurred in 31% of patients and almost always corresponded to total or partial avulsion of the contained root. Surprisingly, 50% of the partially or even totally avulsed roots were not associated with any observable pseudomeningocele on computerized tomography scans or MR images. Therefore we concluded that the identification of pseudomeningoceles through preoperative imaging is not very reliable. The risk of underestimating the number of avulsed roots should incite prudence before making a surgical decision based solely on imaging data.
Anatomical Lesions at Surgery In a series of patients referred to us for chronic pain, radicular lesions were wide and severe. All patients had at least two markedly affected root
levels. Altogether, 78.2% of dorsal roots were impaired 79.1% of which were totally avulsed, either partially avulsed or atrophic. The sensory deficit was larger that the ‘‘theoretical’’ territory of the avulsed roots in as many as 67.4% of patients. This effect, in part, could be related to the existence of additional atrophic and partially avulsed roots or extrarachidian lesions. Indeed, the extent of the sensory deficit corresponded to the dorsal root lesions at the intradural level in only 52% of the patients and was notably more severe in as many as 30% of cases, thus indicating that additional extrarachidian lesions coexisted with observed intradural damage in approximately one third of patients. This finding seems to be logical as the predominant mechanism in brachial plexus injuries is an intense stretching of the entire neural structure, resulting in multiple lesions located outside as well as inside the spinal canal. Topographies of the impaired dorsal roots and pain territory were concordant in almost all patients. A shift was observed in only 2% of cases. Nevertheless, the extent of that pain territory was smaller in as many as 44% of patients but was centered around the main deafferented spinal cord segments. In our patients, damage was predominant at the preganglionic level, results concordant with data published by Wynn-Parry [40] and Narakas [41] who found a high incidence of pain in patients with preganglionic lesions compared with that in patients with postganglionic lesions. In the group of 167 patients with lesions located distal to the dorsal root ganglion (postganlionic) in Wynn-Parry’s series [40], pain was most often transient and exceptionally unbearable. On the contrary, in the group of 108 patients with lesions situated proximal to the dorsal root ganglion (preganglionic), 90.7% of patients had persistent and severe pain, especially when avulsions were total and wide. All patients with a total avulsion extending from C5-Th1 experienced unbearable pain. In Narakas’s series [41], only
Surgical dorsal root entry zone lesions for pain
33% of the 47 patients with postganglionic lesions suffered pain, whereas 82% of the 123 patients with preganglionic lesions had severe pain. The percentage of patients with pain was different according to the number of avulsed roots: 71, 82, 85, and 90% for two, three, four, and five avulsed levels, respectively. These data as well as ours are consistent with the known mechanisms of deafferentation pain. In our series additional abnormalities were frequently found at the spinal cord level, (49% of patients), consisting mainly of marked deviation/ distortions of the cord, fixed by strong adhesive arachnoiditis and/or a more or less notable degree of cord atrophy on the same side as the avulsion. To identify the dorsolateral sulcus and to perform the incision that constitutes the first step of the MDT procedure accurately, a preliminary microsurgical freeing of the spinal cord and the roots is of prime importance. We also assert that freeing the spinal cord and roots contributes to relieve pain induced by neck movements. Focal gliosis and microcysts inside the DREZ and the dorsal horn were observed with the aid of high magnification of the surgical microscope in 36.4% of our patients. Such abnormalities were reported by others to be similarly important [11]. Formation of scar and gliotic tissue at the level of the avulsed root(s) is likely to play a role in the generation of pain by facilitating neural hyperactivity in the DH neurons.
Surgical Procedure Under general anesthesia with tracheal intubation and short-lasting curarization, the patient is placed in the prone position, the neck flexed in the so-called ‘‘concorde’’ position with the head maintained in a three pin head-holder. Through a median cutaneous-aponeurotic posterior incision and a unilateral paravertebral muscle division, a hemilaminectomy with preservation of the spinous processes is performed ipsilaterally to the avulsion and extended according to the
134
injured cord segments. As an example, for a total plexus avulsion (that is C5-T1), hemilaminectomy is performed from C3-C7. The dura mater is then longitudinally opened and sustained laterally with sutures. Arachnoı¨d opening is often difficult because of strong fibrotic adhesions to the cord. Pseudomeningoceles with fragile membranes, instead of dural sheaths, are frequently found at the level of the avulsed segment(s) (> Figure 134‐6). With the aid of the surgical microscope, the anatomical aspect (normal, grayish and atrophic, and partially or totally avulsed) of all roots – either ventral or dorsal – is carefully noted. The functional status of remaining ventral roots is checked by observing muscular responses to direct electrical stimulation at 1 mA (NIMBUS Stimulator, Newmedic/Hemodia, Toulouse, France). Dissection of the neural structures from the dura and the arachnoid may be hard to achieve due to adhesive scar-tissue. The cord is sometimes dramatically atrophic and/or distorted with an abnormal rotation. Identification of the dorso-lateral sulcus is often difficult, due to necrotic or gliotic changes in the cord at the level of the avulsed roots. To solve this problem the sulcus is isolated from the intact remaining rootlets, above and below the avulsed segment (s). The presence of tiny radicular vessels entering the cord helps to determine the site of the sulcus. Yellow areas corresponding to old hemorrhages on the cord surface, microcavities in the depth of the sulcus, and gliotic tissue within the dorsal horn provide guidance for tracing the dorso-lateral sulcotomy. Intra-operative monitoring of the dorsal column SSEPs evoked by stimulation of the homolateral tibial nerve may be helpful, especially when the sulcus is difficult to find [29]. The extent, in length, of the surgical lesioning is established on the basis of pain topography, which generally corresponds with the avulsed segments as well as the altered adjacent rootlets.
2279
2280
134
Surgical dorsal root entry zone lesions for pain
. Figure 134‐6 Micro-DREZotomy (MDT) technique at the cervical level for C6 to T1 Brachial plexus avulsion on left side. Upper and lower left: T2 weighted MRI showing pseudomeningoceles at the lower cervical spine on left side. Upper right: Operative view showing entire cervical spinal root freed from arachnoiditis by microsurgical dissection. The C5 dorsal root remains present but damaged with rootlets stuck together by focal arachnoı¨ditis. Segments from C6 down to T1 on the left side are absent due to total avulsion. Left dorso-lateral sulcus (DLS) can be identified. Lower right: Operative view showing MDT on the C6 avulsed segment; incision into the DLS has been made with the microknife and dotted microcoagulations inside the sulcus with the sharp bipolar microforceps
An incision, 2 mm in depth and oriented 35 medially and ventrally, is made in the dorsolateral sulcus by using a microknife in the axis of the cervical DH. With the aid of magnified vision and sharp graduated bipolar forceps (model 12 30179; Leibinger GmbH, Freiburg, Germany), dotted micro-coagulations are performed inside the DH (3 mm in depth from the surface of the cord). Each coagulation is performed with the aid of direct vision for 2 s at low intensity on the bipolar generator. Special care is taken to locate
these microcoagulations inside the limits of the dorsal horn, in between the cuneate fasciculus of the dorsal column, medially, and the cortico-spinal tract, laterally, to avoid impairing the sensory and motor pathways, respectively. Effects of microDREZotomy
After surgery, 65.9% of our 44 patients [30] who were followed for more than 1 year became pain-free with (nine patients) or without (20 patients) additional medical treatment (> Figure 134‐7). This percentage is well within
Surgical dorsal root entry zone lesions for pain
134
. Figure 134‐7 MicroDREZotomy for pain due to brachial plexus injury. Graph of cumulative percentage (Kaplan-Meier curve) of patients with pain relief (excellent or good) over time. Note that 59.8% of 44 patients, who were followed for more than 1 year (and up to 12 years) after surgery, showed persistent control of pain
the range of those reported in the literature whatever the DREZ lesioning process used, that is, microsurgical bipolar coagulations, radiofrequency thermocoagulation, laser-beam, and ultrasonic probe (> Table 134‐1). No clear-cut anamnestic, clinical, or therapeutic factors in a patient’s history appeared to play a significant role in the efficacy or nonefficacy of surgery. We observed that the less extended the pain, the better the relief. This observation was not statistically significant, however. The failure to relieve pain or the recurrence of pain was not statistically correlated to the time elapsed between the accident and the onset of pain. Surprisingly, we found no statistically significant correlation between duration of pain prior to surgery and outcome. Similarly, Samii et al. [38] found no significant negative influence of the time elapsed between trauma and the onset of pain, or of pain duration before DREZ surgery. DREZ-surgery appears to have more pronounced effect on paroxysmal pain. With regard to pain characteristics, 100% of patients (two of two) suffering from paroxysmal pain alone and 75% of patients (24 of 32 patients) with contin-
uous and paroxysmal pain combined had excellent outcomes as opposed to only 50% of patients (five of ten patients) suffering from continuous background pain alone. Nevertheless, the presence of continuous background pain should not be considered to be a negative factor, which is important when considering indications for surgery. The main neurological complications observed in our series (permanent ataxia (3.6%) and permanent motor weakness (3.6%) in the ipsilateral lower limb; genito-urinary disturbances (1.3%) [30]), as well as in the literature, were the consequence of long tract impairment. Motor weakness and sensory ataxia in the ipsilateral lower limb can be attributed to bad targeting of the dorsolateral sulcus, which is often hard to identify, and to the difficulty in accurately placing the lesion in relation to adjacent structures, namely the corticospinal tract ventrolaterally and the dorsal column cuneus dorsomedially. In conclusion, these results justify that MDT occupies a prominent place in the arsenal for intractable pain due to BPI, the more so as
2281
2282
134
Surgical dorsal root entry zone lesions for pain
genesis of this pain seems to be – predominantlyin the DREZ area and the DH neurons.
Pain After Spinal Cord/cauda Equina Lesions Indications Chronic pain after spinal cord and/or cauda equina injuries may be related to mechanical factors, such as bony instability, but also from central neuropathic pain in the affected spinal cord segments and/or in the territory caudal to the level of the lesion. As reported in our literature review [42], the incidence of central disabling pain after spinal cord and/or cauda equina injuries varies among the various reported and reviewed series, from 10 to 25% of patients. Neuropathic pain after spine injuries may be classified as radicular, segmental (at the level of the injured cord segments), infrasegmental (below the lesion) or visceral. Regarding pain that resides in dermatomes corresponding to the injury, some authors have postulated that it is merely radicular, resulting from nerve root contusion, entrapment or scarring at the level of injury. The development of central pain seems to be related to partial deafferentation of, or direct damage to, central neurons involved in the pain pathways. Multiple mechanisms might contribute to the development of central pain after spinal cord and/or cauda equina injuries, related either to the release of central neurons from their normal inhibitory impulses or to an increase in their intrinsic excitability [43]. Deafferented spinal neurons acquire abnormal spontaneous patterns of discharge in the dorsal horn [43]. Thus dorsal horn hyperactivities have been recorded during surgical procedures in patients with deafferented spinal cords [19,20,22]. A neighboring hypothesis, based on clinical observations, suggests that the origin of pain comes from the spinal cord segments just rostral to the site of injury. This theory is based
on the fact that cordectomy has been reported to improve neuropathic pain only if performed rostral to the level of the lesion [44,45]. The first step of management in patients developing pain after spinal injury is to verify the absence of persisting compressive factors which might be responsible for the pain, namely bony instability, persistent deformity and/ or bone fragments in the canal, which, if present, should be treated surgically. When medical therapies have failed, functional neurosurgery should be considered. In our experience, the microsurgical approach to the dorsal root entry zone (DREZotomy), introduced in the 1970s for pain hypothesized to be generated in the dorsal horn [4,27], has provided the most satisfactory results. Most patients who underwent DREZ-surgery for spinal cord or cauda equina lesions were patients with spine injury (> Table 134-2). According to author’s experience [42], MDT is effective only in patients whose pain has a ‘‘radiculo-metameric’’ distribution, that is, the pain corresponding with the level and extent of the spinal cord lesions. We name this pain: « segmental » pain. In contrast, pain in the territory below the lesion, especially in the perineosacral area, is not influenced even if DREZ surgery is performed at the lower medullary segments. This is particularly true when the pain consists of a permanent burning sensation and is located in a, infralesional, totally-anesthesic, area. Therefore MDT must be reserved for pain syndromes related to the injured medullary segments and the adjacent ones if modified by consecutive pathological processes (e.g. cavitation, gliosis, arachnoiditis). Of paramount importance, in patients with incomplete paraplegia, DREZ lesions have to be performed not too deeply and extensively, to avoid additional neurologic deficits. On the contrary, in patients with complete motor and sensory deficits below the lesion, MDT can be done extensively on the selected segments.
Surgical dorsal root entry zone lesions for pain
134
. Table 134‐2 Review of the literature: long-term results of surgery in the dorsal root entry zone for spinal/cauda equina injury Reference
Technique
Number of patients
Nashold and Bullit [46] Friedman and Nashold [47] Young [34]
Radiofrequency
13
54%
(5 mo–3 y)
Radiofrequency
56
50%
(6 mo–5 y)
Radiofrequency or CO2 laser Radiofrequency Radiofrequency Microsurgery Microsurgery
20
55%
(3–5 y)
Sampson et al. [48] Rath et al. [37] Spaic et al. [49] Sindou et al. [42]
39 22 6 44
The best indication for DREZ-surgery is traumatic lesions of the spine below T10 with complete functional interruption of the conus medullaris, especially when the pain is located in the legs rather than in the perineum. Pain caused by lesions of cauda equina can also be favorably influenced by MDT performed on the corresponding spinal cord segments.
Good results (=Pain relief 75%)
54% 55% 100% Segmental 68% Below lesion0%
(range) Mean follow-up
(1 wk–12 y) 3 y (10 mo–13 y) 5 y (7mo–1 y) 9 mo (1–20 y) 7y
larger necrotic cavities (which can be seen on the MRI-Scan). As a result of the spinal injury, not only the spinal cord is damaged, but also the adjacent tissues including the dorsal and ventral roots and the arachnoidal tissue. The arachnoidal scarring at the site of the spinal injury may be enough to tether the cord.’’
Surgical Procedure
Anatomical Data 1. A majority of patients with a conus medullaris lesion have plurisegmental damage; therefore, a large number of segments must be operated on. For cauda equina lesions, MDTmust be restricted to the medullary segments corresponding with the injured roots. The most frequent pathologic alterations due to spinal cord injuries have been well summarized by Nashold (personal communication): ‘‘Blunt injury to a segment of the spinal cord by a spinal dislocation results in a relatively localized spinal cord injury, whereas a gunshot wound may produce an injury that involves numerous segments above and below the injury. The initial insult is followed by central hemorrhage. After the hemorrhage resolves, small microcysts may form
2.
In most patients, the microsurgical procedure is preceded by a long dissection of the dura from the surrounding epidural fibrosis and a delicate dissection of the cord and the roots from adherent arachnoiditis. In patients with spinal fractures not previously completely operated, one must start with liberation of the neural structures from residual bone fragments occupying the intrarachidian space and even sometimes the intradural space. This preparatory approach may be long and bloody; in that eventuality, it is better to perform the first stage of the operation in a separate sitting, followed by MDT 2 weeks later. Then MDT is performed, as shown in > Figure 134‐8.
2283
2284
134
Surgical dorsal root entry zone lesions for pain
. Figure 134‐8 Upper drawings: Microsurgical DREZotomy performed bilaterally in a patient with a spinal cord injury in the conus medullaris, at the segments (lines with double arrows) corresponding with the ‘‘segmental pain’’ territories (hatched dermatomes). (L-L, level of the responsible lesion in the spinal cord (paraplegic patient)). Lower drawing: Patient with paraplegia with a spinal cord injury at the conus medullaris level (L1 fracture), with segmental pain in both legs (hatched dermatomes). Microsurgical DREZotomy (MDT) will be performed bilaterally in the corresponding conus medullaris segments (lines with double arrows). Drawing illustrates the intraoperative findings: necrotic cyst and gliosis in the segments of the conus medullaris corresponding with the vertebral fracture (L1 level). MDT was performed on both sides in the T12 to L4 spinal cord segments
Effects of microDREZtomy
In our published series [42], surgery was performed in 37 cases at the pathological spinal
cord levels that corresponded to the territory of the so-called ‘‘segmental pain,’’ and in seven cases, on the spinal cord levels below the lesion for ‘‘infralesional pain’’ syndromes. The
Surgical dorsal root entry zone lesions for pain
post-operative analgesic effect was considered to be ‘‘good’’ when the patient’s estimation of pain relief exceeded 75%, ‘‘fair’’ when pain was reduced by 25–75% and ‘‘poor’’ when the residual pain was more than 25% of preoperative estimation. An immediate good result was obtained in 70% of the patients and was long-lasting in 60% of the total series. The results varied essentially according to distribution of pain. Good long-term results were obtained in 68% of the patients with segmental pain distribution, compared to 0% in patients with predominant infralesional pain. As regard pain characteristics, a good result was obtained in 88% of the cases with predominantly paroxysmal pain, compared to 26% with pure continuous pain. There were no perioperative mortalities. Morbidity included cerebrospinal fluid leak (three patients), wound infection (two patients), subcutaneous hematoma (one patient) and bacteremia (in one patient). The above data justify the inclusion of DREZ lesioning surgery in the neurosurgical armamentarium for treating ‘‘segmental’’ pain due to spinal cord/cauda equina injuries. In conclusion, the above results justify inclusion of DREZ-lesioning surgery in the neurosurgical armamentarium for treating neuropathic pain due to spine injuries, provided pain being from segmental origin.
Pain Resulting from Peripheral Nerve Lesions When pain resulting from peripheral nerve injuries is not relieved by Transcutaneous Neurostimulation or Spinal Cord Stimulation, MDT may be considered. This group of patients consisted of 42 cases in our series. From this experience we conclude that MDT is indicated when the predominant component of pain is of the paroxysmal type (electrical flashing pain) and/or corresponds to allodynia/hyperalgesia or both. Good results can also be achieved in
134
severe post-traumatic causalgic syndromes with disabling hyperpathia (i.e., Complex Regional Pain Syndromes, type II). The pain and the vasomotor disturbances can be favorably influenced. In patients without neurologic deficit, DREZ lesion must not be too extensive in length and depth so that the tactile and proprioceptive sensory capacities can be (at least partially) retained, and uncomfortable paresthesias avoided. MDT can also be considered for severe occipital neuralgia. Surgery is performed at the C2–C3 spinal cord segments. At this level, the procedure is easy through a C2-hemilaminectomy. The results in the author’s three such cases were good. In the series of 11 cases published by Dubuisson [50] the effects of the operation were also good. After limb amputation, two main types of pain, which may coexist, can be encountered: pain in the phantom limb and pain in the stump. If Spinal Cord Stimulation fails, DREZ surgery may be considered. Phantom limb pain is generally relieved when rootlets are found avulsed. Pain in the stump is unconstantly influenced; better results are obtained when the pain is of the paroxysmal and/or allodynic type. The newly developed Precentral (motor) Cortex Stimulation (PCS) seems to be promising for pain after amputation, so that PCS shall be tried first.
Post-herpetic Pain Results of surgery in the DREZ for postherpetic pain have been reported by a few groups [37,51]. In our experience, only superficial pain located in the affected dermatomes was significantly improved, especially when of the allodynic type. The permanent (burning or aching) deep component is rather unrelieved; it can even be aggravated, with the patient complaining of additional constrictive sensations after operation.
2285
2286
134
Surgical dorsal root entry zone lesions for pain
At operation, identification of the roots corresponding to the herpetic lesions is difficult. The observation of an atrophy and a grayish color in the concerned root(s) can be helpful. When the thoracic spinal cord is the target, because at this particular level the dorsal horn is narrow and deeply situated as shown in > Figure 134‐3, encroachment of the corticospinal tract laterally and of the dorsal column medially might happen if DREZ surgery is not prudently performed. Before deciding on DREZotomy in patients with postherpetic neuralgia, one must be cautious. Although no death or postoperative neurologic complication occurred in our ten patient’s series, it is necessary to stress the possible vital risks in these patients, most patients being aged and psychologically impaired.
morphine is the technique of choice for advanced widespread pelvic cancers. In our personal series, a good result (i.e., withdrawal of narcotics) was obtained by MDT in 87% of 49 patients operated on at the cervical or the cervicothoracic level and in 78% of 38 patients who underwent surgery at the lumbar or sacral levels for well-localized cancers. Survival time ranged from 1 month to 4 years (average, 14 months). Postoperative infection occurred in two patients, and in two patients, surgery was considered to have precipitated death. Although restricted to a small number of patients among all those affected with pain due to malignancies, MDT is a valuable recourse for patients with topographically-limited painful cancers
Hyperspastic States with Pain Pain Due to Malignancies Good candidates are patients with long life expectancy, general conditions compatible with open surgery and topographically-limited pain caused by well-localized lesions. Pancoast-Tobias syndrome at the thoracic apex is typically a good indication for MDT [52]; the procedure is generally performed from C7 to T2. For more extended cervico-thoracic cancers, percutaneous CT-guided, open high cervical anterolateral cordotomy or stereotactic spino-thalamic tractomy is preferable. Other good indications for MDT are painful conditions caused by circumscribed malignant invasions of the thorax, abdomen wall, or perineal floor, and also pain due to limited neoplastic involvement of lumbo-sacral roots, plexuses, or both. Because extensive DREZ-operations at the lumbar and/or sacral segments would inevitably result in leg hypotonia and/or sphincter disturbances, for pain below the waist in patients who are able to walk, the procedure should be indicated only if it is limited. For perineal pain, midline myelotomy can be an alternative. Intrathecal
Because muscular tone was diminished in the operated territories after MDT performed for treatment of pain, [27] the procedure was applied as early as 1973 for harmful spasticity. [28,53] The antispastic effects can be explained by the fact that MDT interrupts the afferent components of the myotatic (monosynaptic) and the nociceptive (polysynaptic) reflexes, and so deprives the somatosensory relays of the dorsal horn from most of their excitatory inputs. Three groups of patients underwent MDT for hyperspasticity: (1) 45 hemiplegic patients underwent MDT at the cervical level for hyperspasticity in the upper limb; MDT was performed from C5 to T1 segments through C3–C7 hamilaminectomy. (2) 151 patients had MDT at the lumbosacral level for excessive spasticity complicating severe paraplegic states such as those observed in multiple sclerosis patients; MDT was performed bilaterally, through a T11 to L2 laminectomy, from L2 down to S2, and additionally down to S5 when there was a hyperactive neurogenic bladder with urine leakage around the catheter. (3) 15 patients underwent MDT at
Surgical dorsal root entry zone lesions for pain
the sacral (S2 to S3 or S4) level for hyperactive neurogenic bladder. Forty-two of the forty-five patients with harmful spasticity in the upper limb had associated pain. Eighty of the 151 paraplegic patients suffered from pain mostly as a result of spasms or contractures. The cord levels related to the undesirable spasticity were identified by studying the muscle responses to bipolar electrical stimulation of the ventral and dorsal roots. The motor threshold for stimulation of ventral roots was one-third that of the threshold for dorsal roots. The technical procedure was as follows. The ventro-lateral aspect of the DREZ was exposed so that the microsurgical lesions could be performed in the dorso-lateral sulcus, 2–3 mm deep and at a 35 degree angle (for cervical) or a 45 degree angle (for lumbo-sacral) levels, all along the selected segments of the spinal cord; see detailed description in the ‘‘surgical technique’’ section. Intraoperative neurophysiological monitoring was used to help in identifying cord levels, as well as quantifying the extent of MDT; see section on ‘‘intraoperative neurophysiologic monitoring.’’ Results have been detailed elsewhere [54–56]. Only a brief summary will be given here. Follow-up ranged from 2 to 25 years; average: 9 years). For paraplegic patients, a useful effect on lower limbs (i.e., a lasting decrease in tone allowing easy passive mobilization) was obtained in 87% of the patients. Bladder capacity was significantly improved in 85%; the patients who improved were those in whom detrusor was not irreversibly fibrotic. For hemiplegic patients with harmful spasticity in the upper limb, a good effect was obtained in 78%. The effect on the upper limb was significant and lasting, only at the level of the shoulder and elbow, allowing there reappearance of some voluntary movements when hidden behind hypertonia. Effect was much less beneficial at the level of wrist and fingers, so that additional peripheral neuro-
134
tomies, together with orthopedic surgery, were often required. MDT constantly produced a decrease in sensation in the operated territories: mild in 40%, marked in 40% and severe in 20%. When present, pain was durably relieved in 88% in both groups.
Conclusions Whatever pain syndrome is confronted, DREZsurgery must be considered within the framework of the entire armamentarium for pain surgery [57,58].
References 1. Gildenberg PH. History of pain management. In: A history of neurosurgery. Greenblatt SH, Dagi TF, Epstein MH, editors. Illinois: The American Association of Neurological Surgeons Park Ridge; 1997. p. 429-64. 2. Melzach R, Wall PD. Pain mechanism. A new theory. Science 1965;150:971-9. 3. Wall PD, Sweet WH. Temporary abolition of pain in man. Science 1967;155:108-9. 4. Sindou M. Study of the Dorsal Root Entry Zone. Implications for pain surgery. M.D.Thesis, University of Lyon Press, Lyon, 1972. 5. Sindou M, Quoex C, Baleydier C. Fiber organization at the posterior spinal cord-rootlet junction in man. J Comp Neurol 1974;153:15-26. 6. Nashold BS, Urban B, Zorub DS. Phantom pain relief by focal destruction of substantia gelatinosa of Rolando. In: Bonica JJ, Albe-Fessard D, editors. Advances in pain research and therapy, vol 1. New York: Raven Press; 1976. p. 959-63. 7. Nashold BS, Ostdahl PH. Dorsal root entry zone lesions for pain relief. J Neurosurg 1979;51:59-69. 8. Levy WJ, Nutkiewicz A, Ditmore M, Watts C. Laser induced dorsal root entry zone lesions for pain control. Report of three cases. J Neurosurg 1983;59:884-6. 9. Powers SK, Adams JE, Edwards SB, Boggan JE, Hosobuchi Y. Pain relief from dorsal root entry zone lesions made with argon and cardon dioxide microsurgical lasers. J Neurosurg 1984;61:841-7. 10. Kandel El, Ogleznev KJA, Dreval ON. Destruction of posterior root entry zone as a method for treating chronic pain in traumatic injury to the brachial plexus. Vopr Neurochir 1987;6:20-7.
2287
2288
134
Surgical dorsal root entry zone lesions for pain
11. Dreval ON. Ultrasonic DREZ-operations for treatment of pain due to brachial plexus avulsion. Acta Neurochir 1993;122:76-81. 12. Sindou M, Fischer G, Mansuy L. Posterior spinal rhizotomy and selective posterior rhizidiotomy. In: Krayenbu¨hl H, Maspes PE, Sweet WH, editors. Progress in neurological surgery, vol 7. Basel: Karger; 1993. p. 201-50. 13. Sindou M, Goutelle A. Surgical posterior rhizotomies for the treatment of pain. In: Krayenbu¨l H, editor. Advances and technical standards in neurosurgery, vol 10. Vienna: Springer-Verlag; 1983. p. 147-85. 14. Rand R. Further observations on Lissauer’s tractolysis. Neurochirurgica 1960;3:151-68. 15. Denny-Brown D, Kirk EJ, Yanagisawa N. The tract of Lissauer in relation to sensory transmission in the dorsal horn of spinal cord in the macaque monkey. J Comp Neurol 1973;151:175-200. 16. Ramon y Cajal S. Histologie du syste`me nerveux, vol 1. Paris: Maloine; 1901. p. 986. 17. Szentagothai J. Neuronal and synaptic arrangement in the substantia gelatinosa 1964;122:219-39. 18. Wall PD. Presynaptic control of impulses at the first central synapse in the cutaneous pathway. In: Eccles JC, Schade´ JP, editors. Physiology of spinal neurons. Amsterdam: Elsevier; 1964. p. 92-118. 19. Jeanmonod D, Sindou M, Magnin M, Baudet M. Intraoperative unit recordings in the human dorsal horn with a simplified floating microelectrode. Electroencephalogr Clin Neurophysiol 1989;72:450-4. 20. Loeser JD, Ward AA Jr, White LE Jr. Chronic deafferentation of human spinal cord neurons. J Neurosurg 1968;29:48-50. 21. Guenot M, Hupe JM, Mertens P, Ainsworth A, Bullier J, Sindou M. A new type of microelectrode for obtaining unitary recordings in the human spinal cord. J Neurosurg (Spine) 1999;91:25-32. 22. Guenot M, Bullier J, Rospars J. Lansky P, Mertens P, Sindou M. Single-Unit analysis of the spinal dorsal horn in patients with neuropathic pain. J Clin Neurophysiol 2003;20:142-50 23. Loeser JD, Ward AA Jr, White LE Jr. Some effects of deafferentation of neurons. J Neurosurg 1967;17:629-36. 24. Albe-Fessard D, Lombard MC. Use of an animal model to evaluate the origin of and protection against deafferentation pain. In: Bonica JJ, editors. et al. Advances research and therapy pain research and therapy, vol 5. New York: Raven Press; 1983. p. 691-700. 25. Guenot M, Bullier J, Sindou M. Clinical and electrophysiological expression of deafferentation pain alleviated by dorsal root entry zone lesions in rats. J Neurosurg 2002;97:1402-9. 26. Mertens P, Ghaemmaghami C, Bert L, Perret-Liaudet A, Guenot M, Naous H, Laganier L, Later R, Sindou M, Renaud B. Microdialysis study of amino-acid neurotransmitters in the spinal dorsal horn of patients undergoing
27.
28.
29.
30.
31.
32.
33.
34. 35.
36.
37.
38.
39.
40. 41.
42.
microsurgical dorsal root entry zone lesioning. J Neurosurg (Spine 1) 2001;94:165-73. Sindou M, Fischer G, Goutelle A, Mansuy L. La radicellotomie posterieure se´lective. Premiers re´sultats dans la chirurgie de la douleur. Neurochirurgie 1974;20:391-408. Sindou M, Fischer G, Goutelle A, Schott B, Mansuy L. La radicellotomie poste´rieure se´lective dans le traitement des spasticite´s. Rev Neurol 1974;130:201-15. Jeanmonod D, Sindou M. Somatosensory function following dorsal root entry zone lesions in patients with neurogenic pain or spasticity. J Neurosurg 1991;74:916-32. Sindou M, Blondet E, Emery E, Mertens P. Microsurgical lesioning in the dorsal root entry zone for pain due to brachial plexus avulsion: a prospective series of 55 patients. J Neurosurg 2005;102:1018-28. Thiebault JB, Bruxelle J, Thurel: C, et al. Pain in avulsion lesions of the brachial plexus: relief by dorsal horn lesions. In: Brunelli G, editor. Textbook of microsurgery. Paris: Masson; 1988. p. 817-19. Ishijima B, Shimoji K, Shimizu H, et al. Lesions of spinal and trigeminal dorsal root entry zone for deaffeentation pain. Appl Neurophysiol 1988;50:175-87. Friedman AH, Nashold BS, Bronec PR. Dorsal root entry zone lesions for the treatment of brachial plexus avulsion injuries: a follow-up study. J Neurosurg 1988;22:369-73. Young RF. Clinical experience with radio-frequency and laser DREZ lesions. J Neurosurg 1990;72:715-20. Kumagai Y, Shimoji K, Honma T, et al. Problems related to Dorsal Root Entry Zone lesions. Acta Neurochir 1992;115:71-8. Thomas DGT, Kitchen ND. Long-term follow-up of dorsal root entry zone lesions in brachial plexus avulsion. J Neurol Neurosurg Psychiatry 1994;57:737-8. Rath SA, Braun V, Soliman N, et al. Results of DREZcoagulations for pain related to plexus lesions, spinal cord injuries and post-herpetic neuralgia. Acta Neurochir 1996;138:364-9. Samii M, Bear-Henney S, Ludemann W, et al. Treatment of refractory pain after brachial plexus avulsion with doral root entry zone lesions. Neurosurgery 2001;48:1269-77. Prestor B. Microsurgical junctional DREZ-coagulation for treatment of deafferentation pain syndromes. Surg Neurol 2001;56:259-65. Wynn-Parry CB. Pain in avulsion of the brachial plexus. Neurosurgery 1984;15:960-5. Narakas A. Les syndromes douloureux dans les arrachements du plexus brachial. Douleur et Analge´sie 1992;3:83-101. Sindou M, Mertens P, Wael M. Microsurgical DREZotomy for pain due to spinal cord and/or cauda equina injuries. Long-term results in a series of 44 patients. Pain 2001;92:159-71.
Surgical dorsal root entry zone lesions for pain
43. William D, Willis JR. Central Neurogenic Pain: Possible Mechanisms. In: Nashold BS, Ovelmen-Levitt J, editors. Deafferentation pain syndromes; pathophysiology and treatment. Advances in pain research and therapy, vol 19. New York: Raven Press; 1991. p. 81-102. 44. Sindou M, Daher A. Spinal cord ablation procedures for pain. In: Dubner A, Gebhart GF, Bond MR, editors. Proceedings of the Fifth World Congress on Pain. Amsterdam: Elsevier; 1988. p. 477-95. 45. Gybels JM, Sweet WH. Neurosurgical treatment of persistant pain. In: Gildenberg PL, editor. Pain and headache, vol 2. Basel: Karger; 1989b. p. 293-302. 46. Nashold BS, Bullit E. Dorsal root entry zone lesions to control central pain in paraplegics. J Neurosurg 1986;65:465-9. 47. Friedman AH, Nashold BS. DREZ lesions for relief of pain related to spinal cord injury. J Neurosurg 1986;65:465-9. 48. Sampson JH, Chasman RE, Nashold BS, Friedman AH. Dorsal root entry zone lesions for intractable pain after trauma to the conus medullaris and cauda equina. J Neurosurg 1995;82:28-34. 49. Spaic M, Petkovic S, Tadic R, Minic L. Drez surgery on conus medullaris (after failed implantation of vascular omental graft) for treating chronic pain due
50.
51.
52.
53.
54.
55.
134
to spine (gunshot) injuries. Acta Neurochir 1999;141:1309-12. Dubuisson D. Treatment of occipital neuralgia by partial posterior rhizotomy at C1-3. J Neurosurg 1995;82:581-6. Friedman AH, Nashold BS, Ovelmen-Levitt J. Drezlesions for the treatment of post-herpetic neuralgia. J Neurosurg 1984;60:1258-61. Sindou M, Lapras C. Neurosurgical treatment of pain in the Pancoast-Tobias syndrome: selective posterior rhizotomy and open antero-lateral C2-cordotomy. In Bonica JJ, Ventafrida V, Pagni CA, editors. Advances in pain research and theraphy, vol 4. New York: Raven Press; 1982. p. 199-209. Sindou M, Millet MF, Mortamais J, et al. Results of selective posterior rhizotomy in the treatment of painful and spastic paraplegia secondary to multiple sclerosis. Appl Neurophysiol 1982;45:335-40. Sindou M, Abdennebi B, Sharkey P. Microsurgical selective procedures in the peripheral nerves and the posterior root-spinal cord junction for spasticity. Appl Neurophysiol 1985;48:97-104. Sindou M, Mifsud JJ, Boisson D, et al. Selective posterior rhizotomy in the dorsal root entry zone for treatment of hyperspasticity and pain in the hemiplegic upper limb. Neurosurgery 1986;18:587-95.
2289
56. Sindou M, Jeanmonod D: Microsurgical DREZ-tomy for the treatment of spasticity and pain in the lower limbs. Neurosurgery 1989;24:655-70.
57. Sindou M, Mertens P, Garcia-Larrea L. Surgical procedures for neuropathic pain. Neurosurgery Quarterly 2001;11:45-65. 58. Burchiel KJ. Surgical management of pain. New York: Thieme; 2002.
149 Surgical Treatment of Chronic Cluster Headache J. M. Castilla
Introduction Cluster headache (CH) is a rare type of headache, predominantly affecting young males, defined as ‘‘attacks of severe, strictly unilateral pain which is orbital, supraorbital, temporal or in any combination of these sites, lasting 15–180 minutes and occurring from once every other day to 8 times a day. The attacks are associated with one or more of the following, all of which are ipsilateral: conjunctival injection, lacrimation, nasal congestion, rhinorrhoea, forehead and facial sweating, miosis, ptosis, eyelid edema. Most patients are restless or agitated during an attack.’’ ‘‘Cluster headache attacks occurring for more than 1 year without remission or with remissions lasting less than 1 month’’ characterize the chronic type [1]. Previously used terms for cluster headache were ciliary neuralgia, erythro-melalgia of the head, erythroprosopalgia of Bing, hemicrania angioparalytica, hemicrania neuralgiformis chronica, histaminic cephalalgia, Horton’s headache, Harris-Horton’s disease, migrainous neuralgia (of Harris), petrosal neuralgia (of Gardner) [1]. The pathogenesis of CH still remains unclear but recent observations favored the hypothesis of a central brain generator more than a peripheral neurovascular trigger [1,2]. CH may be difficult to diagnose because of the variation of symptoms with time and from patient to patient, the diffusion of the pain and the associated psychological characteristics of the patient. Sometimes CH coexists with other headache types, as in cluster migraine or in cluster-tic syndrome [1,3,4]. #
Springer-Verlag Berlin/Heidelberg 2009
According to the European Federation of Neurological Societies guidelines, evidence-based level A recommendations for the treatment of acute attacks of cluster headache are the inhalation of 100% oxygen, subcutaneous or nasal spray sumatriptan or nasal zolmitriptan. For prophylaxis, verapamil is the treatment of choice and corticoids are recommended, with methysergide, lithium and topiramate as an alternative. Level B recommendations are intranasal lidocaine or subcutaneous octreotide for acute treatment or methysergide, lithium or topiramate for prophylaxis. Surgical procedures are not usually considered and are regarded only as a level C recommendation [5].
Surgical Therapy of Cluster Headache; Results and Complications In refractory cases with severe pain a variety of surgical procedures have been proposed.
Interruption of ‘‘Autonomic’’ Pathways Sphenopalatine Ganglion (SPG) Local anesthetic blocks
Because this ganglion carries sympathetic, parasympathetic and some somatic innervation for most of the mucous membranes of the pharynx, face and lachrymal gland, its blockade by the application of cocaine or lidocaine just posterior to the middle turbinate, where the ganglion lies, is
2526
149
Surgical treatment of chronic cluster headache
done in the hope of aborting an acute attack, serving both as treatment and to confirm diagnosis. Complications include epistaxis and occasionally temporary orthostatic hypotension [5–12]. Permanent destruction
Neurodestructive injection, radiofrequency lesioning or cryoneurolysis of the pterygopalatine fossa can be done repeatedly if necessary, though the nearby maxillary nerve, artery and sympathetic plexus can be injured [7,10,13,14]. Although the pain may recur after extirpation of the SPG, the recurrent pain is usually felt easier to manage [6,13,15]. Some consider it more effective in sphenopalatine neuralgia than in CH, differentiating sphenopalatine neuralgia from CH by its female predominance, longer lasting attacks of burning and aching pain, and commonly associated parasympathetic activity [7]. In 55 patients [10,13,15,16] treated with destructive procedures performed on the SPG, good results were reported in 29%; some procedures were repeated. Complications consisted of maxillary hyperpathia in 7%, significant hematoma in the cheek in 20%, epistaxis in 14%, transient palatal hypoesthesia in 30%, transient V2 paresthesia, oroantral fistula in 2%, xerophthalmia in 120 5%, corneal desiccation in 2%. One patient treated with radiosurgery was reported to have amelioration of pain and medication intake after a 12 month follow-up period [12]; another patient with cluster headache experienced lasting pain relief after radiosurgery but not others with different headaches [17].
ing these nerves, sometimes with intraoperative stimulation in the awake patient [19,20]. In 105 patients [3, 18–27] this operation gave good results in 43% of them. Complications consisted of facial weakness in 9% (some transient), keratitis in 2%, Horner’s syndrome in 1%, and crocodile tears in 1%.
Nervus Intermedius (NI) Section or Decompression It has been found that stimulation of the NI reproduced the pain of facial neuralgia patients who did very well after section of that nerve [28]. In five series [28–32] NI section yielded good results in 67%, complications including a 27% incidence of deafness, 27% of transient vertigo and nystagmus, 6% of wound infection, 6% of anosmia and 6% of facial nerve injury. NI decompression was inspired by the thinking that the causative mechanism of CH might be vascular compression of the nerve in a similar manner to that proposed in trigeminal neuralgia. Vascular cross compression of the seventh nerve/NI complex was found and relieved in some patients accompanied with other treatments [4,31,33,34].
Interruption of Trigeminal Pathway Peripheral Nerves
Greater and Lesser Superficial petrous Nerve (GSPN) Section It has been postulated that the parasympathetic and/or pain fibers passing through the GSPN may channel the periodic discharges which might account for CH symptoms [18,19] so that headache sufferers were treated by section-
Anesthetic blocks, alcoholization, avulsion or resection of the superficial vessels and/or nerves may be useful initial procedures avoiding major surgery [18,22,24,32,35]. Although easily performed, repeatable, and sometimes giving long lasting relief, pain, if relieved, usually recurs following nerve regeneration. In different series, short-lived benefit or no relief at all was reported following peripheral
Surgical treatment of chronic cluster headache
neurectomies and nerve blocks [10,18,25,28,34, 36–39]; sometimes reported details were insufficient to allow evaluation. Other peripheral targets historically reported on the assumption that the pain is generated in nociceptors in those structures had been the arteries to the face and scalp [13,18,19,23,40], the sympathetic chain [18,20,24,34,41], or otolaryngologic structures [42,43]. Interference with these structures, often associated with other procedures, led to uncertain or short-term results.
Trigeminal Ganglion and/or Roots Harris [35] used alcohol injection of the peripheral nerves serving the painful zone, but found more lasting effects after injecting the gasserian ganglion: of 29 patients so treated, 19 were completely relieved, and five others much improved; follow-up was not stated except for five patients, with one instance of transient keratitis reported. In four series of patients treated by glycerol injection into Meckels’ cave [44–47], 42% enjoyed good relief. Complications were: 37% dysesthesia without anesthesia dolorosa, 19% corneal hypoesthesia, 3% trigeminal paresis. Radiofrequency lesioning of trigeminal ganglion and/or roots in Meckel’s cave utilizes thermocoagulation of the pain-carrying fibers of the trigeminal nerve, and, guided by electrical stimulation, permits better control of the lesion than other percutaneous procedures. In eight reported series [21,22,34,37,48–51] 52% of patients enjoyed good relief with the following complications: perioperative arterial hypertension, corneal abrasion 2%, neuropathic pain 7%, transient diplopia 4%, transient masseter weakness 5%, hyperacusis 1%, aseptic meningitis, dermatosis, CSF rhinorrhea 1% each. Open surgery includes a variety of procedures from section of ophthalmic rootlets of
149
the first trigeminal division to complete section of the trigeminal root, via middle or posterior fossa approaches. Out of nine reported series of trigeminal section for cluster headache [18,20,22,24,29,36,40,49,52], 63% of patients enjoyed good pain relief. Complications included wound infection 4%, CSF leakage 4%, corneal abrasion 8%, hyperacusis 2%, neuropathic pain 4%, contralateral CH pain 2%, meningitis, subdural hematoma, tinnitus, middle ear effusion, trigeminal motor and IV nerve weakness 1% each. One out of 88 patients died. Complete section of the trigeminal sensory root produces better long-lasting pain relief than percutaneous procedures [20,36,49] but causes more extensive and severe sensory loss than the radiofrequency procedure, which can be more difficult to accept [6,36,40], and the procedure carries a greater surgical risk, but that morbidity may be acceptable [52]. Radiosurgery of the trigeminal nerve root entry zone: Its use as a treatment for refractory CH was first reported by Ford et al. in 1998 with 67% excellent results without significant side effects at 8–14 months of follow up [53]. In recent series [54–56] radiosurgery of the trigeminal nerve does not achieve long-term pain relief in cluster headache patients, and with 50–90% morbidity with follow-up of 5–88 months. As those authors treat trigeminal neuralgia in a similar way, with better results, a different structural vulnerability of the nerve or different processing of the sensory message in those different diseases is suspected [55].
Bulbar Procedures Bulbar procedures such as spinal descending trigeminal tract and caudalis nucleus lesions may help some patients, but risk significant morbidity [18,21,32,57].
2527
2528
149
Surgical treatment of chronic cluster headache
Combined Trigeminal and NI Section or Decompression Combined trigeminal and NI section or decompression have been reported in some patients with variable results [29,31,33,34,58,59]. Transient facial weakness, wound infection, postoperative headache or cerebrospinal fluid rhinorrhea were some of the complications.
Stimulation Procedures In some recently reported series [60–62], cluster headache patients have been treated with occipital nerve electrode stimulation. Although the mechanism is not clear, reported results were satisfactory for 60–75% of the patients so treated for a mean follow-up of 15–20 months, without significant side effects, but they needed frequent surgical revisions for electrode or battery reimplant. Often some weeks are needed before a significant clinical improvement may be reached although attacks returned in days when the device malfunctioned. Supraorbital nerve stimulation has been successfully tried in a cluster headache patient [63], and vagal nerve stimulation in another two [64]. Stimulation of hypothalamus: Based on findings suggesting hypothalamic hyperactivity related to CH bouts and morphometric changes in hypothalamus among patients with CH, a target for electrical stimulation in the posterior inferior homolateral hypothalamic grey mater was considered for the treatment of those patients. Thirteen out of sixty patients so treated were reported as being persistently pain-free to almost pain-free [65]. Parameters were unipolar stimulation with a frequency of 180 Hz, a pulse width of 60 ms and an amplitude of 0.7–3 V. Initial coordinates of the target were 6 mm posterior to the anterior commissure–posterior commissure midpoint, 2 mm left of the midline, and 8 mm below the commissural plane. Those
authors then changed the target to one 3 mm behind the midcommissural point, 5 mm below the midcommissural point, and 2 mm lateral to the midline [66,67]. It takes days to weeks to observe a change in the clinical picture. A common side effect was transient diplopia, which limited stimulation amplitude. In one patient, a small non-symptomatic hemorrhage into the third ventricle occurred following implant. Persistent side effects are absent except in one patient with bilateral stimulation, in whom stimulation was stopped to resolve vertigo and worsened bradycardia, but was resumed later without further problems [65]. Other groups [68,69] have treated more patients and confirm those initial promising results and the limitations, but even lethal complications have also been found. The brain target used does not correspond to a known specific anatomical entity, but is not far from the periventricular gray matter. Neurons in the target region showed low-frequency tonic discharge and are under study. Using H2 15O-positron emission tomography and alternately switching the hypothalamic stimulator on and off, the stimulation induced activation in the ipsilateral hypothalamic grey (the site of the stimulator tip), the ipsilateral thalamus, somatosensory cortex and praecuneus, the anterior cingulate cortex, and the ipsilateral trigeminal nucleus and ganglion. Additionally deactivation in the middle temporal gyrus, posterior cingulate cortex, and contralateral anterior insula was observed [2].
Comments The best surgical candidates are those with established strictly unilateral headache and those with a stable personality who are not prone to addiction [70]. Medical treatment should have been exhausted prior to consideration of surgery, and psychological screening or diagnostic local anesthetic block may help to improve the results.
Surgical treatment of chronic cluster headache
Complete analgesia or dense hypalgesia in the painful zone, is currently considered essential for success, but may not always be necessary [46,47,70,71]. Complications include corneal damage by corneal analgesia, worsening of any existing contralateral pain may occur [6,66,72]. Painful sequelae of deafferentation is seemingly more frequent in CH patients than in those with other diagnoses [21,55] but no quantitative study exists. Although the ideal therapy may be lacking, in a chronic medically intractable situation a less satisfactory outcome may be valuable, and appropriate surgery may be warranted and may be preferable to undesirably prolonged medical treatment [32]. Long-term pain relief may not be possible, even with the most rigid selection of patients, in part because of the behavioral changes produced by long lasting pain. Hopefully, the disease may remit or change to an episodic variety after several months or years, unrelated to any particular treatment [38]. The decision to resort to surgical treatment when confronted with a patient with unrelieved severe pain despite the use of all conservative treatment strategies must be tempered by the following realities: success with surgical (or other) treatment in chronic pain is limited; chronic pain tends to recur with time after its initial relief; many of the procedures used to treat chronic CH are postganglionic nerve sections inevitably followed by nerve regeneration. No better outcome should be expected after surgical treatment of cluster headache than in the treatment of chronic pain syndromes in general, which is in general less than 50% in most cases, and which result applies equally to nonsurgical strategies [72]. We may conclude that for the treatment of chronic cluster headache patients, surgery may have to be considered when the diagnosis is clear and the pain has failed to respond to nonsurgical therapy, although results are imperfect and the pain is rarely cured. Overall, in reported
149
series, some improvement occurs in about twothirds of chronic CH patients treated surgically, although long lasting good relief of pain is much more difficult to achieve. Most complications from the surgical procedures are transient or capable of medical treatment, but even patients with more severe complications may find the procedure worthwhile. Although the most prolonged pain relief followed section of the trigeminal or intermedius nerves, or both, these more aggressive procedures do not always ensure better results. New stimulation procedures appear promising and need further confirmation. Since the morbidity and mortality of intracranial procedures cannot be reduced to the levels of the percutaneous ones nor that of the percutaneous procedures to those of simple sections or avulsions, it may be wise to recommend the least aggressive treatment first which offers the possibility of repetition if needed.
References 1. Headache Classification Committee of the International Headache Society. The International classification of headache disorders. 2nd ed. Cephalalgia 2004;24 Suppl 1:1-160. 2. May A, Leone M, Boecker H, Sprenger T, Juergens T, Bussone G, Tolle TR. Hypothalamic deep brain stimulation in positron emission tomography. J Neurosci 2006;26:3589-93. 3. Ekbom K. A clinical comparison of cluster headache and migraine. Acta Neurol Scand 1970;46 Suppl 41:1-48. 4. Solomon S, Apfelbaum RI, Guglielmo KM. The clustertic syndrome and its surgical therapy. Cephalalgia 1985;5:83-9. 5. May A, Leone M, Afra J, Linde M, Sandor PS, Evers S, Goadsby PJ. EFNS guidelines on the treatment of cluster headache and other trigeminal-autonomic cephalalgias. Eur J Neurol 2006;13:1066-77. 6. Campbell JK, Onofrio BM. Surgical management of cluster headache. In: Tollison CD, Kunkel RS, editors. Headache, diagnosis and treatment. Baltimore: Williams and Wilkins; 1993. p. 205-10. 7. Cicala RS, Jernigan JR. Nerve blocks and invasive therapies. In: Tollison CD, Kunkel RS, editors. Headache, diagnosis and treatment. Baltimore: Williams and Wilkins; 1993. p. 357-68.
2529
2530
149
Surgical treatment of chronic cluster headache
8. Robbins L. Intranasal lidocaine for cluster headache. Headache 1995;35:83-4. 9. Felisati G, Arnone F, Lozza P, Leone M, Curone M, Bussone G. Sphenopalatine endoscopic ganglion block: a revision of a traditional technique for cluster headache. Laryngoscope 2006;116:1447-50. 10. Sanders M, Zuurmond WW. Efficacy of sphenopalatine ganglion blockade in 66 patients suffering from cluster headache: a 12- to 70-month follow-up evaluation. J Neurosurg 1997;87:876-80. 11. Yang Y, Oraee S. A novel approach to transnasal sphenopalatine ganglion injection. Pain Physician 2006;9:131-4. 12. Lad SP, Lipani JD, Gibbs IC, Chang SD, Adler JR Jr, Henderson JM. Cyberknife targeting the pterygopalatine ganglion for the treatment of chronic cluster headaches. Neurosurgery 2007;60:E580-E581. 13. Meyer JS, Binns PM, Ericsson AD, Vulpe M. Sphenopalatine ganglionectomy for cluster headache. Arch Otolaryngol 1970;92:475-84. 14. Cook N. Cryosurgery of headache. Res Clin Stud Headache 1978;5:86-101. 15. Cepero R, Miller RH, Bressler KL. Long-term results of sphenopalatine ganglioneurectomy for facial pain. Am J Otoraryngol 1987;8:171-4. 16. Brown LA. Mythical sphenopalatine ganglion neuralgia. South Med J 1962;55:670-2. 17. De Salles AAF, Gorgulho A, Golish SR, Medin PM, Malkasian D, Solberg TD, Selch MT. Technical and anatomical aspects of novalis stereotactic radio surgery sphenopalatine ganglionectomy. Int J Radiat Oncol Biol Phys 2006;66 Suppl 4:53-7. 18. White JC, Sweet WH, editors. Pain and the neurosurgeon, a forty-year experience. Springfield: Charles C Thomas; 1969. p. 359-80. 19. Gardner WJ, Stowell A, Dutlinger R. Resection of the greater superficial petrosal nerve in the treatment of unilateral headache. J Neurosurg 1947;4:105-14. 20. Wake M, Hitchcock E. A review of treatment modalities for periodic migrainous neuralgia. Pain 1987;31:345-52. 21. Sweet WH. Surgical treatment of chronic cluster headache. Headache 1988;28:669-70. 22. Watson CP, Morley TP, Richardson JC, Schutz H, Tasker RR. The surgical treatment of chronic cluster headache. Headache 1983;23:289-95. 23. Alvarez-Garijo JA, Bordes-Valls C, Vila-Mengual M. Surgical treatment of autonomic faciocephalalgia. Rev Esp Otoneurooftalm 1975;33:7-11. 24. Stowell A. Physiologic mechanisms and treatment of histaminic or petrosal neuralgia. Headache 1970;1:187-94. 25. Trowbridge WV, French JD, Bayless AE. Greater superficial petrosal neurectomy for orbitofacial pain. Neurology 1953;3:707-13. 26. Ponti L, Valerio M. Results of resection of the greater superficial petrosal nerve in several cases of post-otitis
27. 28. 29. 30. 31. 32.
33.
34.
35. 36.
37.
38.
39.
40. 41. 42.
43.
44.
45.
headache and of cranio-facial neuralgia. Clin Otorinolaringol 1960;12:409-18. Denecke HJ. Greater petrosal nerve surgery in long-term cluster headache. HNO 1977;25:48-50. Sachs E. The role of the nervus intermedius in facial neuralgia. J Neurosurg 1968;28:54-60. Morgenlander JC, Wilkins RH. Surgical treatment of cluster headache. J Neurosurg 1990;72:866-871. Sachs E. Further observations on surgery of the nervus intermedius. Headache 1969;9:159-61. Rowed DW. Chronic cluster headache managed by nervus intermedius section. Headache 1990;30:401-6. Gybels JM, Sweet WH. Periodic or chronic migrainous neuralgia-cluster headache. In: Gybels JM, Sweet WH, editors. Neurosurgical treatment of persistent pain. Basel: Karger; 1989. p. 70-87. Lovely TJ, Kotsiakis X, Jannetta PJ. The surgical management of chronic cluster headache. Headache 1998;38:590-4. Solomon S, Apfelbaum RI. Surgical decompression of the facial nerve in the treatment of chronic cluster headache. Arch Neurol 1986;43:479-82. Harris W. Alcohol injection of the gasserian ganglion for migrainous neuralgia. Lancet 1940;26:481-2. Kirkpatrick PJ, O’Brien MD, MacCabe JJ. Trigeminal nerve section for chronic migrainous neuralgia. Br J Neurosurg 1993;7:483-90. Taha JM, Tew JM. Long-term results of radiofrequency rhizotomy in the treatment of cluster headache. Headache 1995;35:193-6. Alberca R, Lo´pez JM, Aguilera JM, Casado JL, Arenas C, Serrano V, Moreno A. Chronic cluster headache: course and outcome of treatment. Neurologia 1994;9:379-86. Busch V, Jakob W, Juergens T, Schulte-Mattler W, Kaube H, May A. Occipital nerve blockade in chronic cluster headache patients and functional connectivity between trigeminal and occipital nerves. Cephalalgia 2007;11:1206-14. Mazars G, Merienne L Cioloca C. Present state of pain surgery. Neurochirurgie 1976;22 Suppl 1:128-9. Albertyn J, Barry R, Odendaal CL. Cluster headache and the sympathetic nerve. Headache 2004;44:183-5. Novak VJ, Malek M. Pathogenesis and surgical treatment of migraine and neurovascular headaches with rhinogenic trigger. Head Neck 1992;14:467-72. Schonsted-Madsen U. Long-term results in patients with headaches related to nasal obstruction. Ear Nose Throat J 1992;71:38-40. Waltz TA, Dalessio DJ, Ott KH, Copeland B, Abbott G. Trigeminal cistern glycerol injections for facial pain. Headache 1985;25:354-7. Ekbom K, Lindgren L, Nilsson BY, Hardebo JE, Waldenlind E. Retro-gasserian glycerol injection in the treatment of chronic cluster headache. Cephalalgia 1987;7:21-7.
Surgical treatment of chronic cluster headache
46. Hassenbusch SJ, Kunkel RS, Kosmorsky GS, Covington EC, Pillay PK. Trigeminal cisternal injection of glycerol for treatment of chronic intractable cluster headaches. Neurosurgery 1991;29:504-8. 47. Pieper DR, Dickerson J, Hassenbusch SJ. Percutanous retrogasserian glycerol rhizolysis for treatment of chronic intractable cluster headaches: long term results. Neurosurgery 2000;46:363-70. 48. Mathew NT, Hurt W. Percutaneous radiofrequency trigeminal gangliorhizolysis in intractable cluster headache. Headache 1988;28:328-31. 49. Onofrio BM, Campbell JK. Surgical treatment of chronic cluster headache. Mayo Clin Proc 1986;61:537-44. 50. Maxwell RE. Surgical control of chronic migrainous neuralgia by trigeminal ganglio-rhizolysis. J Neurosurg 1982;57:459-66. 51. Kanpolat Y, Savas A. Comments on Pieper DR, Dickerson J, Hassenbusch SJ. Percutanous retrogasserian glycerol rhizolysis for treatment of chronic intractable cluster headaches: long term results. Neurosurgery 2000;46:363-70. 52. Jarrar RG, Blak DF, Dodick DW, Davis DH. Outcome of trigeminal nerve section in the treatment of chronic cluster headache. Neurology 2003;60:1360-2. 53. Ford RG, Ford KT, Swaid S, Young P, Jennelle R. Gamma knife treatment of refractory cluster headache. Headache 1998;38:3-9. 54. McClelland S III, Tendulkar RD, Barnett GH, Neyman G, Suh JH. Long-term results of radiosurgery for refractory cluster headache. Neurosurgery 2006;59:1258-63. 55. Donnet A, Tamura M, Valade D, Re´gis J. Trigeminal nerve radiosurgical treatment in intractable chronic cluster headache: unexpected high toxicity. Neurosurgery 2006;59:1252-7. 56. McClelland S III, Barnett GH, Neyman G, Suh JH. Repeat trigeminal nerve radiosurgery for refractory cluster headache fails to provide long-term pain relief. Headache 2007;47:298-300. 57. Nashold BS, El-Naggar AO, Gorecki JP. The microsurgical trigeminal caudalis nucleus DREZ procedure. In: Nashold BS Jr, Pearlstein RD, editors. The DREZ operation. Parke Ridge Illinois: AANS Publications Committee; 1996. p. 159-88. 58. Kunkel RS, Dohn DF. Surgical treatment of chronic migrainous neuralgia. Cleve Clin Q 1974;41:189-92. 59. Wilkins RH. Neurovascular decompression procedures in disorders of cranial nerves V, VII, IX and X to treat pain. In: Schmideck HH, Sweet WH, editors. Operative
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
149
neurosurgical techniques. 3rd ed. Philadelphia: Saunders; 1995. p. 1457-68. Magis D, Allena M, Bolla M, De Pasqua V, Remacle JM, Schoenen J. Occipital nerve stimulation for drug-resistant chronic cluster headache: a prospective pilot study. Lancet Neurol 2007;6:314-21. Schwedt TJ, Dodick DW, Hentz J, Trentman TL, Zimmerman RS. Occipital nerve stimulation for chronic headache long-term safety and efficacy. Cephalalgia 2007;27:153-7. Burns B, Watkins L, Goadsby P. Treatment of medically intractable cluster headache by occipital nerve stimulation: long-term follow-up of eight patients. Lancet 2007;369:1099-106. Narouze SN, Kapural L. Supraorbital nerve electric stimulation for the treatment of intractable chronic cluster headache: a case report. Headache 2007;47:1100-2. Mauskop A. Vagus nerve stimulation relieves chronic refractory migraine and cluster headaches. Cephalalgia 2005;25:82-6. Bussone G, Franzini A, Proietti Cecchini A, Mea E, Curone M, Tullo V, Broggi G, Casucci G, Bonavita V, Leone M. Deep brain stimulation in craniofacial pain: seven year’s experience. Neurological Sciences 2007; 28 Suppl 1:146-9. Leone M, Franzini A, Broggi G, May A, Bussone G. Long-term follow-up of bilateral hypothalamic stimulation for intractable cluster headache. Brain 2004;127:2259-64. Franzini A, Ferroli P, Leone M, Broggi G. Stimulation of the posterior hypothalamus for treatment of chronic intractable cluster headaches: first reported series. Neurosurgery 2003;52:1095-101. Schoenen J, Di Clemente L, Vandenheede M, Fumal A, De Pasqua V, Mouchamps M, Remacle JM, de Noordhout AM. Hypothalamic stimulation in chronic cluster headache: a pilot study of efficacy and mode of action. Brain 2005;128:940-7. Starr PA, Barbaro NM, Raskin NH, Ostrem JL. Chronic stimulation of the posterior hypothalamic region for cluster headache: technique and 1-year results in four patients. J Neurosurg 2007;106:999-1005. Mathew NT. Treatment of headache. In: Gildenberg PL, Tasker RR, editors. Textbook of stereotactic and functional neurosurgery. New York: McGraw Hill; 1998. p. 1735-49. Matharu MS and Goadsby PJ. Persistence of attacks of cluster headache after trigeminal nerve root section. Brain 2002;125:976-84. Tasker RR. The recurrence of pain after neurosurgical procedures. Qual Life Res 1994;3 Suppl 1:S43-S49.
2531
136 Sympathectomy for Pain C. R. Telles-Ribeiro . L. F. de Oliveira
Several persistent pain syndromes are influenced by the sympathetic nervous system, including dysesthesias, reflex sympathetic dystrophy (complex regional pain syndrome II) induced by accidental or surgical trauma to nervous structures, postherpetic neuralgia, pain from the peripheral neuropathy of diabetes, and pain from the peripheral entrapment syndromes. These so-called sympathetically maintained pain syndromes have three common characteristics: (1) a history of noxious stimulation in the painful area; (2) a continuous burning pain with allodynia; and (3) relief by sympathetic blockade. Sympathetic procedures may treat neuropathic pain directly or provide vasodilatation (> Table 136-1). Classically, these procedureswere developed to alleviate ischemia and the pain induced by peripheral vascular diseases, thus reducing the risk of necrosis and amputation. Ablative procedures, such as lumbar sympathectomy, are used as a last resort for treating peripheral ischemic diseases such as diabetic angiopathy and Raynaud’s disease when clinical attempts to control the disease and the progression of ischemia are unsuccessful. Alternatively, less aggressive procedures, such as temporary local anesthetic block of the sympathetic ganglia, have been shown to be very useful not only as a predictor of the success of radical ablative procedures but also, when frequently or continuously applied, as an effective means of treating pain and ischemia resulting from peripheral ischemic disease and neuropathic pain. In several other pathological conditions, such as complex regional pain syndromes I and
#
Springer-Verlag Berlin/Heidelberg 2009
II (sympathetic reflex dystrophy and causalgia) and cluster headache, peripheral sympathetic fibers and/or sympathetic hyperactivity seem to be involved in the origin of the pain. In these extremely painful and disabling pathologies, the use of sympathetic blockade has been shown to be an effective alternative treatment. Other painful conditions in which there is no immediate evidence of sympathetic hyperactivity but in which afferent pain information emerging from the vasa and nervi nervorum of abdominal and thoracic viscera passes through sympathetic nerves and ganglia, as in postherpetic neuralgia, brachial plexus pain caused by the spread of lung and breast cancer (Pancoast’s syndrome) and upper abdominal cancer pain (through the celiac ganglion) can be totally or partially relieved by sympathetic procedures. Lately, repeated local anesthetic blocks have been preferred for the treatment of neuropathic pain over more radical ablative procedures, since the clinical results are almost the same and there is no risk of secondary hyperpathia with dysesthesia, as we sometimes observe after a somatic neurolytic block with alcohol or a major surgical intervention such as a lumbar sympathectomy. This chapter discusses the currently used sympathetic procedures, including those aimed at the treatment of nonischemic pain. The targets for sympathetic procedures used for the treatment of chronic pain are the lower cervical sympathetic ganglia (stellate ganglion), the celiac ganglion and the lumbar sympathetic ganglia (> Table 136-2).
2298
136
Sympathectomy for pain
. Table 136-1 Classification of pain syndromes with a sympathetic component Ischemic Diabetic angiopathy and neuropathy Complex regional pain syndrome I and II (Reflex sympathetic dystrophy and causalgia) Raynaud’s disease Nonischemic Postherpetic neuralgia Atypical facial neuralgia Cluster headache
. Table 136-2 Techniques of sympatholysis Percutaneous techniques Peripheral nerve blocks (intercostal blocks) Sympathetic ganglion blocks (stellate, celiac) Neurolytic sympathetic blocks Radiofrequency neurolytic blocks Continuous epidural sympathetic blocks Videothoracoscopic ganglionectomy Intravenous sympathetic block (Hannington-Kitf) IV regional blockade with guanethidine Surgical sympathectomy or ganglionectomy
Stellate Ganglion Block Anatomy The stellate, or inferior cervical sympathetic ganglion, is located just in front of the transverse process of the seventh cervical vertebra and receives fibers from the T1-T8 thoracic spinal segments. Although containing chiefly sympathetic fibers, the stellate ganglion is a mixed structure, since nociceptive thin fibers, emerging from the vasa und nervi nervorum of the upper limbs and head, cross the ganglion on their way to the spinal nociceptive system. Local anesthetic block of the ganglion ´ınterrupts not only the sympathetic outflow, inducing vasodilatation, but also nociception originating from the deep structures of the arms and head. This is the reason why pain in the arms, like that from neoplasic infiltration of the brachial plexus, can be controlled with stellate ganglion block.
Local Anesthetic Procedures
Stellate Ganglion Both ablative and nonablative procedures can be performed on the stellate ganglion to treat painful or ischemic diseases. By far the most important and frequently used procedure is the serial stellate ganglion blockade with local anesthetics for the treatment of craniofacial postherpetic neuralgia, upper limbs reflex sympathetic dystrophy, and, more rarely, pain caused by neoplasic infiltration of the brachial plexus. Stellate ganglion block is a safe and very effective means not only for alleviating pain of these very painful conditions but also, when started at the beginning of the disease, of aborting or retarding the evolution of the syndrome. Frequently, patients are not effectively treated because of the ignorance or fear of their doctors about the benefits of this procedure, although this is not a valid reason for denying patients access to the indicated treatment.
A detailed description of local anesthetic blocks can be found elsewhere. [1,2] The classical technique of stellate ganglion block is by the paratracheal approach. At the level of the cricoid cartilage the carotid artery is palpated, along with, when possible, the Chassaignac’s tubercle of the sixth cervical transverse process. Pulling the carotid artery laterally with the fingers of left hand and keeping it under the fingers tips, the skin is punctured with a 21-gauge needle attached to a 10-ml syringe filled with 0.20% ropivacaine or 0.25% bupivacaine, either plain or with methylprednisolone acetate. The needle is cautiously advanced until we feel the bony resistance (usually 15–25 mm deep) of the transverse process of the sixth cervical vertebra and then is pulled back 2–3 mm before the solution is injected. Careful aspiration is absolutely necessary before injecting to avoid accidental
Sympathectomy for pain
intravascular administration of the local anesthetic solution, and should be repeated at each 5 ml injected as well, to prevent accidental injection into vertebral artery. Although this is a serious and potentially fatal complication, in the hands of an experienced anesthesiologist this procedure is safe, and the fear of such a complication should not contraindicate the procedure unless the tcchnical conditions for it are not available (a trained anesthesiologist, oxygen, and equipment for cardiorespiratory ressuscitation). The accuracy of the blockade is verified by the appearance of an ipsilateral Claude Bernard– Horner syndrome. The procedure can be repeated daily or spaced 2–7 days apart, depending on the indication. When one is using a corticosteroid, at least 5 days should pass before each injection, and no more than four injections should be done in a row. In our experience patient compliance is good, as a result of the rewarding analgesic effect of the procedure (> Figure 136-1a and b).
Indications This procedure is quite effective for the treatment of several painful and potentially persistent pain syndromes, such as postherpetic neuralgia (PHN) of the face (Vl, V2 and V3 branches of the trigeminal nerve), PHN of the arm and scapula, . Figure 136-1 Stellate ganglion block a and b
136
intercostal PHN of upper thoracic segments (T1T3), reflex sympathetic dystrophy of the upper limb (including causalgia), and pain resulting from brachial plexus neoplasic infiltration by lung and breast cancer. Alternatively, when the procedure cannot be done (patient refusal, infection at the site of injection, etc.), placement of an epidural catheter at the C7-T l level followed by a continuous local anesthetic blockade with bupivacaine 0.15–0.20% (plus morphine in some cases) can be used for analgesia and sympathetic blockade with excellent results, in our experience.
Stellate Ganglion Ablation Neurolytic Block The stellate ganglion can be ablated by surgical or chemical means. Neurolysis of the stellate ganglion can be done with agents such as aqueous phenol (10%) or alcohol (50–70%). The procedure is similar to that of local anesthetic blockade but should be done under X-ray and fluoroscopic guidance. Before we attempt the neurolytic blockade, a local anesthetic blockade of the stellate ganglion should be carried out to evaluate the efficacy of the procedure and allow the patient to become acquainted with the Claude Bernard– Horner syndrome. After placement of the needle,
2299
2300
136
Sympathectomy for pain
l–2 ml of an ionic contrast medium should be injected to confirm the localization. The injection of alcohol (2–3 ml) is painful and should be done under sedation and consequently is seldom performed today. When ablation of the stellate ganglion is needed, the preference should be for the neurolytic block with aqueous phenol (10%, 2–3 ml) or for the surgical procedure that also includes the ablation of T2-T3 sympathetic ganglion for better results.
Surgical Ablation of the Stelate Ganglion – Cervicothoracic Sympathectomy The classical anterior approach to the stellate ganglion was described by Leriche, Gask and Ross [3]. More recently, a new percutaneous endoscopic transaxillar technique was introduced and is now the favorite procedure, usually performed by thoracic surgeons. The classical anterior cervical approach is performed under general anesthesia with the patient lying on his or her back and the neck slightly rotated to the contralateral side. A straight 6 cm vertical incision is made from the superior border of the clavicle following the internal border of the sternocleidomastoid muscle. Next, the medial cervical aponeurosis is sectioned to enter the prevertebral space and the neurovascular bundle is pushed aside. The anterior scalenus muscle and the phrenic nerve are pulled laterally, exposing the sixth cervical vertebra and the Chassaignac’s tubercle. The vertebral artery is identified, and the stellate ganglion is exposed. After the excision of the ganglion, as low as possible, the incision is closed in planes.
Celiac Ganglion Anatomy The celiac ganglion is located in front of and around the aorta at the level of the T12 and L1
vertebral bodies. It is a mixed sensory-sympathetic structure that innervates the abdominal viscera and receives nociceptive afferents from the supramesocolic structures of the abdomen (pancreas, liver, biliary tract, and stomach). Deep visceral pain arising from these structures is conducted to the spinal nociceptive system by visceral afferents that cross through the celiac ganglion, entering the spinal cord at the T5-T12 levels [4]. The celiac ganglion can be reached from the abdomen under direct vision when the patient undergoes abdominal surgery or through a percutaneous paravertebral approach.
Neurolytic Block Percutaneous neurolytic block of the celiac ganglion has already been described in detail elsewhere. [1,2] After establishing an intravenous access (16 or 18 G) and starting an intravenous drip with saline or lactate-Ringer’s solution, the patient is positioned in the prone position on an X-ray transparent operating table with a pillow under the abdomen. A Hudson or macro mask is fitted over the face of the patient and a 3 L/min oxygen flow is started. Usually only mild sedation is necessary before starting the procedure. We prefer 50–100 mg of IV fentanyl. The vertebral body of L1 should be localized under X-ray fluoroscopy, and the L1-L2 interspace should be marked on the skin. A straight line should be traced passing through this point from one floating rib (twelfth rib) to the other. The points where this line encounters the ribs (usually 7–10 cm from the midline) are the sites of insertion of the needles and should be marked on the skin. From this point, a line directed to the top of the L1 vertebral body is also drawn on the skin on both sides, forming a triangle with its base in the L1-L2 interspinous line and its vertex at the L1 body. Local anesthetic (l % lidocaine) is then injected at the points of insertion. The celiac ganglion needles (21-gauge, 15–18 cm length) are then inserted at an angle of 45 to the skin aiming at the L1 body, following
Sympathectomy for pain
the line traced between the insertion point and the L1 body, and gently advanced under fluoroscopic control. Local anesthetic should be administered as necessary, until contact is made with the L1 vertebral body. Lidocaine 1 ml should be injected at this point. The needle is then pulled back 4 or 5 cm and reinserted at a greater angle until its tip passes l–2 mm in front of the L1 body. The correct localization of the needle should be checked with a lateral X-ray. With both needles in place, 2 ml of an ionic contrast medium is injected and the propagation of the contrast along the anterior face of the vertebra is observed under fluoroscopy (> Figure 136-1). Care should be taken not to place the needle into the psoas muscle. In case of intravascular placement, the needle should be replaced or, in the case of aorta puncture, advanced until blood is no more aspirated. After sedation, usually with a low hypnotic dose of etomidate or propofol, 20–25 ml of 50% alcohol is injected through each needle. The needles are then withdrawn after washing the residue of alcohol with 1ml of either lidocaine or saline. The patient is then immediately turned on his or her back and respiratory and circulatory assistance is provided as necessary. Usually a drop in blood pressure is observed, although not immediately, and should be corrected with crystalloid and vasopressors as needed. Dehydrated patients may develop hypotension in the first 6–12 h and should remain in bed to avoid orthostatic hypotension. Diarrhea is common in the first 24–48 h. Celiac ganglion block is indicated for cancer pain originating from the supramesocolic viscera (stomach, liver, biliary tract and pancreas). The procedure is more effective when it is performed before the patient develops ascites or disseminated peritoneal disease. The pain of acute pancreatitis is better controlled with s.c. or i.v. opioids. Epidural morphine and/or 0.25% bupivacaine is an alternative therapy. In the case of chronic pancreatitis, when satisfactory pain control is not achieved with analgesics, either alcohol celiac block or epidural
136
opioids should be tried. Surgical interruption of the splanchnic nerves is seldom used.
Lumbar Sympathetic Ganglia Anesthetic blockade or surgical ablation of the lumbar sympathetic ganglia can be employed for the treatment of painful neuropathic and ischemic diseases of the lower limbs. In our opinion, continuous blockade of the lumbar sympathetic fibers with local anesthetics and opioids should always precede an attempt at surgical ablation. In postherpetic neuralgia of the lower limbs, repeated blockade of the L2-L3 ganglia with bupivacaine (0.15–0.20%) or continuous lumbar epidural blockade with morphine and bupivacaine [using a patient controlled analgesia (PCA) infusion device] is the treatment of choice and is quite effective when started within the first 3 months of the evolution of the disease.
Local Anesthetic Block The sympathetic innervation of the lower limbs originates in the lumbar segments of the spinal cord centering at the L2-L3 sympathetic ganglia. Percutaneous blockade of the lumbar sympathetic ganglia is similar to that described for the celiac ganglion. The L2 and L3 spinous processes are identified and a point 5 to 10 cm lateral to the midline is marked on the skin. After local infiltration with lidocaine 1%, the blocking needles, similar to those used for a celiac ganglion block, are inserted at a 45 angle in the direction of the L2 or L3 vertebral bodies until bone resistance is encountered. Lidocaine 1% (l–2 ml) is the injected at this point. The needles are then withdrawn 4–5 cm and reinserted at a wider angle and advanced to the anterior border of the vertebra. A volume of 10 ml of 0.2% bupivacaine is then injected at each side. A more simple approach is to make just one insertion at L2
2301
2302
136
Sympathectomy for pain
level and inject 20–25 ml of 0.15–0.2% bupivacaine on each side (> Figure 136-2). When repeated injections are needed, we prefer to implant an epidural catheter at the L2-L3 level and start a continuous block of the lumbar sympathetic fibers with a solution of 0.20% bupivacaine and morphine, using a PCA infusion device. This procedure, however, demands that the patient remains hospitalized during the period of the treatment, usually for 15 days.
Indications Lumbar sympathetic blockade is indicated for treatment of neuropathic and sympatheticmediated pain of the lower limbs (postherpetic neuralgia, postamputation pain, phantom limb pain, and reflex sympathetic dystrophy), ischemic pain (diabetic angiopathy and Raynaud’s disease), and preoperative pain control in patients with peripheral vascular diseases who will be submitted to amputation, with the objectives . Figure 136-2 Celiac ganglion block
of alleviating unnecessary suffering and reducing the chance of phantom limb pain.
Surgical Ablation Surgical ablation of the lumbar sympathetic ganglia consists of the excision of the second, third and fourth lumbar sympathetic ganglia and the nerve chain that connects them [3]. The anterolateral approach is preferred. With the patient under general anesthesia lying on his or her side with a pillow under the dependent lumbar region, a horizontal incision is made following the border of the eleventh rib. The transversalis fascia is opened and the peritoneal sac is entered by digital dissection to expose and excise the ganglionic chain. Care should be taken to avoid damage to the lumbar vein. This procedure is indicated for patients with sympathetic mediated pain of the lower limbs that show temporary relief after sympathetic blockade.
Personal Experience We now comment on our experience in the treatment of 384 patients with persistent neuropathic and sympathetically maintained pain characterized by dysesthesia and allodynia resulting from various causes, including surgical trauma, herpes zoster, and shoulder-hand syndrome, in the Pain Clinic of the University Hospital of the State University of Rio de Janeiro. Pain was measured with the visual analogue scale (VAS) before and after each treatment. We considered a good result when a decrease in the pain score >50% was obtained. When we analyzed the distribution of pain in 187 patients with complex regional pain syndrome or dysesthesia after injury, the most common localization was in the upper limbs (> Table 136-3).
Sympathectomy for pain
. Table 136-3 Distribution of complex regional pain syndrome (Reflex Sympathetic Dystrophy) Localization of pain
% of patients
Arms Legs Thorax Face Abdomen
55.6 33.6 3.2 6.4 1.2
Initially, all patients were medicated with neuroleptics, tricyclic antidepressants and analgesic drugs, chiefly codeine plus acetaminophen (paracetamol) or a nonsteroidal antiinflammatory drug. Stronger opioids were avoided because of the risk of addiction. Only 4.9% of these patients, however, achieved satisfactory control of the pain with pharmacological treatment. Most patients needed a neural blockade or surgical section for effective control of the pain. In the patients with reflex sympathetic dystrophy or dysesthesia of the arms or face, stellate ganglion blockade with 0.25% bupivacaine gave good results. Pain scores decreased more than 50% in 57.8% of these patients. On average, five blocks (from one to ten) were performed before maximum recovery was reached. In 12% of the patients, a cervical peridural catheter was placed and a continuous block of the sensory and sympathetic fibers with bupivacaine 0.20% and morphine (2 mg/12 h) was established and maintained for 12–18 days. Good results were obtained in 74% of these patients. In patients with complex regional pain syndrome of the legs, a series (an average of three) of lumbar sympathetic blockades or a series (an average of three) of lumbar peridural blockades with 0.25% bupivacaine were tried, with good results in 53.8 and 57.2% of the patients, respectively, so that both procedures were equally effective. Intravenous local guanethidine block [5] of the affected limb was tried on an average of 3 times (1–6 times) in 6.4% of the patients with complex regional pain syndrome of the arm.
136
In contrast to the results reported in the literature [6], however, only 30% of these patients showed good results. Other forms of treatment were tried, with no significant results or only episodic good results, such as transcutaneous electrical nerve stimulation (TENS), acupuncture, percutaneous cordotomy, and peripheral nerve exploration. No correlation was found between the cause of the pain and the response to treatment. The likelihood of good results obtained with the blocks did not vary significantly, regardless of the origin of the pain. We studied also, 166 patients with postherpetic neuralgia ranging in age from30 to 90 years, with the majority (45%) in the seventh and eigth decades. The patients were divided into three classes, depending on the time of evolution of the disease: (1) acute phase (less than l month): 10.4%; (2) subacute phase (between l and 6 months): 46.1% and (3) chronic phase (more than 6 months): 43.5%. The distribution of pain was most common in the thoracic region (48.8%), followed by the face (30.9%), the cervical region (15.4%), and lumbosacral region (4.9%). All the patients were treated initially with amitriptyline, sulpiride, and analgesics such as paracetamol plus codeine. Most of the patients, however, required a local anesthetic block to get sufficient pain relief. The most common procedure was the repeated intercostal nerve block for patients with thoracic PHN. An average of three blocks were performed in the first week with bupivacaine 0.25% added to methylprednisolone acetate (once a week in patients in the subacute phase), followed by a block each 5–7 days, until maximal relief. Greater than 50% improvement of the pain was observed in 60.7% of the patients. Stellate ganglion block was the standard treatment for most of the patients with facial or cervical PHN (33.7% of PHN patients). Good results (>50% decrease in the pain score) were observed in 76.7% of these patients. An average
2303
2304
136
Sympathectomy for pain
of five blocks were performed, ranging from one to ten procedures, at intervals from 48 h to l week. In 9% of the patients, either patients who refused the stellate block or in whom this procedure did not provide a satisfactory result, continuous peridural cervicothoracic (catheter placed at C7-T1 interspace) block with bupivacaine 0.25% plus morphine (2 mg q 12 h) was used. Among these patients, 76.9% had a >50% decrease in their pain scores at the end of the treatment. When we compared the results between patients in the acute, subacute, and chronic groups, we observed 76.4% of good results in the acute phase, 72.6% in the subacute phase and 56.8% in the chronic phase. The age of the patients did not interfere in the overall results, but we observed that patients older than 70 years had better results than did the patients in the acute phase of PHN (82.9% with >50% decrease in the pain scores). A total of 31 patients with intense abdominal cancer pain resistant to analgesic therapy with supramesocolic malignant disease, were also studied. In 30 of these patients, the cause was cancer of the pancreas, and one patient had cancer of the stomach invading the head of the pancreas. Most of these patients complained of thoracolumbar pain irradiating to the upper abdomen (20 patients), seven patients had pain only in the abdomen, three had referred pain only at the thoracolumbar region and one had
pain in the right hypochondrial region. The patients averaged 60-years old, and the duration of pain averaged 8 months. The treatment used was the neurolytic blockade of the celiac ganglion with 50% alcohol in 92% of the patients. Complete relief of the pain (a 100% decrease in the pain score) was observed in 80% of these patients. The very high index of success depends on the correct selection of patients. The procedure needs to be done before the patient develops ascites or peritoneal spreading of the disease. In case of a recurrence of the pain, the procedure can be repeated, although in our experience this has never been necessary.
References 1. Moore DC. Regional block. 4th ed. Springfield, IL: Charles C Thomas; 1965. 2. Raj PP. Pradical management of pain. 2nd ed. St. Louis, MO: Mosby; 1992. 3. Patel J, Leger L. Tratado de te´cnica quirurgica. lst ed. Barcelona: Toray-Masson; 1977. 4. Andersen S, Bond M, Mehta M, Swerdlow M. Chronic non-cancer pain. Lancaster: MTP Press; 1987. 5. Hannington-Kiff JG. Intravenous regional sympathetic block with guanethidine. Lancet. 1974;1:1019-1020. 6. Holland JR. The causalgia syndrome treated with regional intravenous guanethidine. Clin Exp Neurol. 1978;15:166-173. 7. Abram S. Incidence-hypothesis-epidemiology. In: StantonHicks M, editor. Pain and the sympathetic nervous system. Boston, MA: Kluwer; 1989.
124 Technique of Trigeminal Nucleotractotomy M. J. Teixeira . E. T. Fonoff
Introduction Management of certain craniofacial pain conditions, such as anesthesia dolorosa, atypical facial pain, postherpetic neuralgia, facial pain in Wallemberg’s syndrome, and pain caused by cranioorofacial malignancies represent a great therapeutic challenge. Despite the great number of medical and nonablative surgical therapies available for their treatment, ablative antialgic neurosurgical procedures are still necessary in several cases. However, the results of the many of these operations [1], when good, are only temporary, while others worsen the original clinical condition by inducing deafferentation pain syndromes or are not applicable in the treatment of neuropathic cranioorofacial pains [2]. The medulla carries the pain-transmitting nerve fibers and relay neurons related to the processing of cranioorofacial pain, being a key area of the central nervous system for pain surgery. Trigeminal tractotomy, trigeminal nucleotomy, and nucleus caudalis dorsal root entry zone (DREZ) lesioning have been described as percutaneous or open techniques to destroy centrally the nociceptive fibers from the fifth, seventh, ninth, and tenth cranial nerves or their nuclear relay neurons. For many years, surgical interruption of the trigeminal descending tract was performed using open techniques. Despite their efficacy, mainly in patients with nociceptive Electronic supplementary material Supplementary material is available in the online version of this chapter at http:// dx.doi.org/10.1007/978-3-540-69960-6_124 and is accessible for authorized users. #
Springer-Verlag Berlin/Heidelberg 2009
pains, the original trigeminal tractotomies present many disadvantages, because they are not efficient in cases deafferentation pains nor applicable in poor clinical status high-risk individuals, they are made under general anesthesia, which by itself precludes cooperation with the patients during the procedure or monitoring of their neurophysiologic functions, and present high cost and high complication rate. The new surgical methods, aiming the destruction of the trigeminal nuclear cells and not just the trigeminal tract, and the improvement of the classical procedures prompted neurosurgeons to develop less-invasive techniques using percutaneous approaches, especially for its application in patients with high risk of infection, hemorrhage, or neurological deficits [1–3].
Anatomophysiological Aspects of the Trigeminal Nuclear Structures Accordingly, to the studies about the functional organization of the trigeminal sensory complex, the spinal nucleus of the trigeminal sensory nuclear complex has an essential role as a relay for nociception of the cranioorofacial region [4]. Clinicopathological observations in patients with vascular injuries of the brain stem and syringobulbia affecting the trigeminal nuclei and the trigeminal pathways and the anatomic and electrophysiological studies made in animals and humans significantly improved the knowledge about the anatomy and physiology of the trigeminal system [1,5] allowing the development of the surgical procedures made in the brain stem
2098
124
Technique of trigeminal nucleotractotomy
and in the upper cervical cord for the treatment of the facial pains in patients [6]. The nociceptive fibers from the fifth, seventh, ninth, and tenth cranial nerves descend in the medulla, forming the descending trigeminal tract. The trigeminal root, on entering the ventral surface of the pons, penetrates 4–5 mm through the middle cerebellar peduncle and in the lateral region of the tegmentum, bifurcates into ascending branches that synapse within the main sensory nucleus and into descending branches that synapse mainly within the spinal trigeminal nucleus [7], and less densely, within the solitary tract nucleus [8,9]. Proprioceptive fibers of the trigeminal nerve synapse within the mesencephalic nucleus, send collaterals to the motor nucleus, and also give branches that enter the spinal trigeminal tract [9]. There is considerable overlap between the cervical and trigeminal afferent projections into the spinal trigeminal nucleus [9,10], which is the natural rostral extension of the dorsal horn [4,11]. There are also many connections between the neurons of the trigeminal nuclear complex through a polysynaptic bidirectional intranuclear pathway [12]. In humans, the subnucleus caudalis extends from the obex to the C4 spinal cord level [11] and is subdivided cytoarchitectonically into the subnucleus oralis, subnucleus interpolaris, and subnucleus caudalis [13]. Phylogenetically, the primitive subnucleus caudalis is analogous to the gray matter of the dorsal horn of the spinal cord [11]. Its cytoarchitecture and its electrophysiological characteristics also suggest that both structures are analogous with respect to their function [13]. It represents the substantia gelatinosa and their neurons synapse with the afferent A-delta and C fibers and presents circuit arrangements, and axoaxonal contacts, corresponding to the anatomic substrate of presynaptic inhibition [14,15]. Somesthetic inputs of the seventh, ninth, and tenth cranial nerves [16,17], and inputs from adjacent reticular formation, medial and lateral cineaste nuclei, contralateral subnucleus caudalis, cerebral cortex [18], and first-order
neurons of the upper levels of the cervical cord [19–21] also reach the subnucleus caudalis. Based on the anatomic and clinical data [5,12,22,23], it was concluded that the subnucleus caudalis is the most important locus subserving integration, processing, and conduction of nociception from the cranioorofacial structures [24]. Nociceptive mechanical stimulation in the face evokes potentials in its structure from very restricted ipsilateral fields [13,25]. Anatomic studies indicate that the axonal trajectories originating from the subnucleus caudalis differ from that of the more rostral neurons of the spinal trigeminal nucleus (subnucleus interpolaris and subnucleus oralis) and of the main sensory nucleus. Electrophysiologically, it was shown in cats, that there are fibers from the subnucleus caudalis projecting to the thalamus, reticular formation of the brain stem, and the more rostral neurons of the trigeminal nuclear complex [7,26]. Sectioning transverse the descending tract in monkeys, cats, rats [27], and humans [6,28,29], however, did not alter significantly dental and oral mucous membrane pain sensation. Lesions impinging the rostral region of the trigeminal sensory complex affecting the nucleus principalis, the subnucleus oralis and the subnucleus interpolaris but not the descending trigeminal tract significantly decreases the behavioral reactions to oral and dental noxious stimulation in animals [30]. This means that the more cephalic nuclei are also involved with pain transmission from the face and the related structures [31]. The importance of the rostral regions of the trigeminal nuclear complex in processing pain was also demonstrated by many other experiments. Nociceptive cells were identified not only in both superficial and deep area of the subnucleus caudalis [32,33], but also in the ventromedial region of the subnucleus interpolaris [34] and in the deep parts of the subnucleus oralis and nucleus principalis [30]. It was shown electrophysiologically that the pain sensation from the tooth pulp and oral mucous surface is relayed in the nucleus principalis and in the subnuclei oralis and
Technique of trigeminal nucleotractotomy
interpolaris, and that the nociceptive stimuli from the cutaneous surface of the face are mainly relayed in the subnucleus caudalis [27]. Horseradish peroxidase-labeled tooth pulp afferent fibers were found projecting to the entire ipsilateral trigeminal spinal nuclear complex, especially to the subnucleus oralis [35], and the pain fibers from the cornea were found projecting to the mid-pontine region [36]. The basal thalamic activity is preserved after the trigeminal tractotomy at the obex during the application of noxious stimuli in the oral region in animals [4]. In rats, projections from the main nucleus and subnucleus oralis signal information related to thermal pain [24]. In cats, a greater proportion of the subnucleus caudalis cells project within the rostral trigeminal nuclei that terminate itself directly in the contralateral thalamus [12]. These observations suggest that the rostral components of the trigeminal sensory complex are involved with other functions than the motor control or motivational affective mechanisms of facial pain [37]. They also explain the failure of the trigeminal medullary tractotomy in the abolition of neuralgia in many cases, despite the complete loss of facial nociception [38], and the distortion, but not abolition of pain and temperature sensation [39], even after anatomic evidence of complete sectioning of the trigeminal tract. Therefore, the classical concept of an exclusive role of the subnucleus caudalis in orofacial nociception should be abandoned. The evidences suggest that the entire spinal trigeminal nuclear complex is involved in the orofacial nociception [40]. The caudal trigeminal subnucleus may participate in other functions than through direct trigeminothalamic projections. It is possible that the subnucleus caudalis, through intranuclear ascending fibers [12], maintains a tonic facilitatory influence on relay neurons located rostrally in the subnucleus oralis and in the nucleus principalis and may determine whether the stimulus is perceived as noxious or not [4,24,41]. It was also shown that the activity of the subnucleus caudalis is modulated by descending projections from the
124
rostral trigeminal nuclei [42]. This means that the facial nociception may be encoded by central summation of impulses from many trigeminal nuclear afferent projections and that the cutaneous and oral mucous membrane nociception may depend on the spatial and temporal summation within the entire spinal trigeminal nucleus rather than activation on an exclusive nociceptive pathway relayed via the subnucleus caudalis [23,40]. In animals and in humans, strychnine reverses the facial analgesia caused by tractotomy [40,43]. There is a marked, but not maximal, elevation of the cutaneous pain thresholds after the medullary tractotomy. However, under special circumstances, the synaptic effectiveness of the input over a single afferent pathway is augmented and may be appreciated as painful [44] > Figure 124-1. It should also be emphasized that the neurons localized within the subnucleus caudalis subserve other functions than the perception of pain [4,5]. Despite the fact that the tactile afferents project primarily to the main sensory nucleus, a significant number of low-threshold non-nociceptive receptors also projects to the descending trigeminal tract [6,45]. This justifies the occurrence of partial reduction of touch sensation after the trigeminal tractotomy [28]. Studies based on evoked potentials revealed that the neurons of all three trigeminal branches in the spinal trigeminal nucleus of cats, despite the great overlap, tend to be laminated in a dorsoventral and caudorostral order; the ophthalmic branch-related neurons lie more ventrally and descend to the upper cervical segments, while those related to the mandibular branch are situated most dorsally and may not descend below the level of the obex and those related to the maxillary branch lie between them [46,47], findings were confirmed also in monkeys by Kerr [8]. The general sensory afferent fibers from the seventh, ninth, and tenth cranial nerves travel just medially to the third branch fibers, cross with them [48], and project in an area between the trigeminal tract and the fasciculus cuneatus
2099
2100
124
Technique of trigeminal nucleotractotomy
. Figure 124-1 (a) Schematic representation of the posterior aspect of lower brain stem. Observe the topographic relationship between the trigeminal tract and nuclei and the restiform body, right anteriorly (b) Fiber stained slice of the medulla on pyramid decussation level and out line of the neural structures in the vicinity of trigeminal tract and nucleus. The superior portion of the figure corresponds to the posterior aspect of the medulla
[45,49–52]. However, the caudal extension of this tract is unclear. The central projections of the ophthalmic branch fibers in the descending trigeminal tract is the most ventrolateral and descend lowest, as far as the C2-C3 level [48,50], while those from the central projections
of mandibular branch fibers are the most posteromedial and cross at higher levels at or about the obex [49]. Hosobuchi and Rutkin [22] used evoked potentials during trigeminal tractotomy in humans and observed that the central projectors of the ophthalmic branch fibers descend 8–10 mm below the obex to the level of the C2 posterior rootlets, the maxillary afferent fibers do not descend more than 10–12 mm below the obex, and the mandibullar fibers are located from 2 mm above to 2–3 mm below the obex. Others claim that all facial sensory pathways descend below the level of the foramen magnum [50,53]. In humans, it was shown by incremental coagulation of the nucleus caudalis during percutaneous trigeminal nucleotomies that the mucous surfaces are represented deeper than the skin at the same segmental level [25]. During stimulation, there is usually facilitation almost exclusively referred to the affected mucous membranes or skin [25]. The concept based on different rostrocaudal levels of termination of the trigeminal tract according to its origin in the specific branch of the trigeminal nerve can no longer be sustained. As proposed by Dejerine early this century [1], the topographic arrangement of the subnucleus caudalis of the trigeminal spinal nucleus has an ‘‘onion-skin’’ pattern in which the central or oral area of the face projects to the rostral part of the nucleus and the peripheral areas project to its caudal part [6,10,50,54]. Monopolar stimulation of the bulbocervical trigeminal nuclear neurons in conscious patients during percutaneous trigeminal tractotomies confirmed the onion-skin pattern of the functional areas of the spinal trigeminal nucleus in humans [25,55]. At high cervical levels, responses from all trigeminal nerve branches, including those involved in the innervation of the perioral regions, and from the seventh, ninth, and tenth cranial nerves are easily elicited [25]. However, to treat pain in the central area of the face and in the intraoral region, the lesion must be made far enough
Technique of trigeminal nucleotractotomy
upward to involve the most rostral part of the nucleus [31,54,55]. According to Kunc [51], to achieve complete analgesia of the central areas of the face, the tract must be interrupted 12–15 mm above the second dorsal root and according to Hosobuchi and Rutkin [22], this must be done at the level of the obex. The high level of the location of the central projection of the mandibular branch fibers in the spinal trigeminal nucleus is the explanation offered by some authors for the difficulty in inducing analgesia in the mandibular region and in the areas innervated by the seventh, ninth, and tenth cranial nerves by the trigeminal tractotomy [5,25,43,49,54,56]. However, Kunc [50] observed that the area in which the pain and temperature pathways of the trigeminal nerve lies extends rostrally just from the upper cervical cord 6–10 mm below the obex.
Trigeminal Tractotomies and Nucleotomies Many ablative methods aiming the interruption of the trigeminal descending tract and/or destruction of the spinal trigeminal nucleus were developed for the treatment of facial pains, including the open tractotomy [5], percutaneous stereotactic [53] or freehand nucleotractotomy [48], open caudalis DREZ lesions [57], and pontine stereotactic trigeminal nucleotractotomy [58]. In the following lines, these methods will be described according to the temporal sequence of their description.
Open Trigeminal Tractotomy In 1937, Sjoqvist [5] observed that the spinal tract of the trigeminal nerve could be safely and effectively sectioned by an incision through the posterolateral aspect of the medulla above the level of the obex. The procedure usually resulted
124
in ipsilateral thermoanalgesia of the face preserving other kinds of sensations. Animal experiments and operative results in humans published later confirmed these results [27,52,54]. This method became popular because caused less facial dysesthesia than did the trigeminal rhizotomies [49,52,54,59]. Sjoqvist [5] performed the procedure with the patient prone and under general anesthesia. After suboccipital craniotomy, the dura mater and the arachnoid membrane were opened, and the tonsil was elevated to allow exposure of the medulla. Having the olive as a reference, the trigeminal tract was sectioned few millimeters behind the last vagal rootlet [5] with the deepness of 3.5–4 mm, and 8–10 mm above the obex, in a region covered by the restiform body [60] (> Figure 124-2). The operation, however, was difficult and followed by many complications, such as . Figure 124-2 (a) Schematic representation of anterolateral aspects of the lower brain stem, showing the lesion site proposed by Sjoqvist and Grant and the relationship between the trigeminal tract and nucleus and the restiform body during an open trigeminal tractotomy
2101
2102
124
Technique of trigeminal nucleotractotomy
ipsilateral ataxia, lesion of the recurrent laryngeal nerve, lateropulsion, contralateral analgesia, paralysis of vocal cords, gait disturbances, and impairment of postural sensibility in the limbs [60]. Lesion of the restiform body and the nucleus ambigous was responsible for the most important complications of the trigeminal tractotomy. A few years later, Grant accidentally made a lesion below the level proposed by Sjoqvist and observed that it resulted in analgesia over all the three ipsilateral trigeminal nerve branches [28]. He concluded that lesions placed in areas distal to Sjoqvist’s technique resulted in fewer complications [60]. Therefore, the caudal part of the medulla and upper cervical cord became the preferred site for trigeminal tractotomy because the lesion of the spinocerebellar fibers could be avoided by sections made at or below the obex. Grant and Weiberger [60] recommended the incision be made 4–5 mm below the obex, where the tract occupies the dorsolateral border of the medulla and is covered by only a few sparse external arcuate fibers, and Olivecrona [54] recommended making the incision at the level of the obex to spare the restiform body and the fibers of the dorsal spinocerebellar tract, which run parallel to the incision. The most difficult technical problem in trigeminal tractotomy and nucleotomy was the determination of the rostral limit of the lesion. The elimination of pain sensation in the medial area of the face requires the destruction of the most rostral part of the nucleus near the obex [28,61]. The lesions at or below the obex produced incomplete and temporary facial analgesia especially in the region innervated by the third branch of the trigeminal nerve and in the territory of the seventh, ninth, and tenth cranial nerves [56], because the afferent fibers from these regions do not descend in sufficient numbers below the obex [5]. Mackenzie [28] noted that if the spinal tract was not sectioned high enough to produce analgesia in all areas supplied by the trigeminal nerve, the upper lip was most
commonly spared [62]. Even lesions at the obex often failed or produced only temporary analgesia in the region innervated by the third branch [2,5,45,49,54,63]. The analgesia mandibular branch territory was possible only with lesions at higher levels than those intended for ophthalmic branch analgesia. The mucous membranes may retain pain sensitivity even when analgesia is produced in the skin of the face [6]. Deeper lesions are necessary to achieve abolition of pain in the mucous membranes [6]. Many of the problems described with the modified procedure were due to difficulty in correctly placing the lesions in the medulla [28]. In many cases, the incision was made in the wrong place resulting in insufficient interruption of the spinal tract causing either no sensory loss or unsatisfactory improvement [49]. Reliance on anatomic landmarks was one of the reasons for these difficulties. Although most neurosurgeons used the obex as a landmark for trigeminal tractotomy, McKenzie [28] considered that its relation to the intrabulbar structures was not constant. Olivecrona [52] and Falconer [49] advised operating under local anesthesia to control the level and extent of the incision during the procedure by obtaining subjective responses from the patients [49,52]. Despite the possibility of operations under local anesthesia, direct stimulation of the trigeminal tract was painful because its fibers are primary neurons and its manipulation or section could be induce movement in the patients [28]. The operations sometimes were also taken too long to be performed under local anesthesia, making the cooperation of the patients unsatisfactory [2]. In 1977 Kunc [50,64] developed an original technique, the ‘‘vertical trigeminal partial nucleotomy’’ [64], that was more selective. Under local anesthesia, he made a vertical incision along the medial margin of the spinal trigeminal tract, at the border of the fibers of the third branch and the cuneatus nucleus, aiming at interruption of the fibers of the first-order neurons and also the
Technique of trigeminal nucleotractotomy
origin of the second-order neurons, as well as, both the extratrigeminal pathways terminating in the subnucleus caudalis and the intranuclear trigeminal fibers. The level of tractotomy was related to the onion-skin pattern of nuclear representation of facial nociceptive afferents. The more caudal in the tract the incision, the farther the margin of analgesia extended from the midline of the face, whereas with a more rostral incision, analgesia appeared in the center of the face [62]. Trigeminal tractotomy was performed for the treatment of trigeminal neuralgia, glossopharyngeal neuralgia, postherpetic neuralgic, anesthesia dolorosa of the face, other neuropathic facial pains, facial pain resulting from malignancies, and headaches1. According Guidetti [59], trigeminal tractotomy should be reserved for patients in whom the tactile sensibility of the face should be preserved, especially in young and middle-aged subjects and in those in whom pain is localized in the area of innervation of the ophthalmic and maxillary branches of the trigeminal nerve. Grant and Weinberger [60] recommended the procedure for the treatment of first branch neuralgia and for patients in whom neuralgia had developed on the second side after a previous trigeminal rhizotomy to prevent the risk of lesions in the motor root. The method had many advantages. It allowed the sacrifice of pain and thermal sensations without suppression of other sensory modalities in the territories of the trigeminal nerve and of the seventh, ninth, and tenth cranial nerves [62]. The residual tactile function was usually satisfactory for normal function, meaning that the operation could be performed bilaterally, reducing the frequency of the serious functional disturbances resulting from bilateral rhizotomies. It also is selective, allowing interruption of pain and thermal functions in the territory of an isolated trigeminal branch or in the territory of any seventh, ninth, and tenth cranial nerves [50].
124
According to some authors [49,52,61,64,65], trigeminal tractotomy resulted in immediate improvement in 90–100% of patients with trigeminal neuralgia. Others however, observed improvement in only 17–61% of them [54,55,62,66], even in the presence of complete analgesia. However, the recurrence rate was high, ranging from 29 to 37% [28,52,54,59,65], especially in cases of severe pain. The method was therefore usually considered unsatisfactory for the treatment of trigeminal neuralgia [56,67]. Selective tractotomy can help patients with glossopharyngeal neuralgia; all the patients operated on by Kunc [50] improved. Falconer [49] performed trigeminal tractotomy in two patients presenting anesthesia dolorosa of the face, and the result was good in both. No other paper was published about the use of this procedure for the treatment of this condition. The management of postherpetic neuralgia with this method, with rare exceptions [62], was frustrating [2]. Falconer [49] did not observe any improvement in all of the four patients he operated on. In 62% of the patients of Olivecrona [52] with migraine, the results of the trigeminal tractotomy was good. Tractotomy was very efficient and could be combined with high cervical rhizotomies inn patients with cancer in the head and neck and pain in the areas innervated by the fifth, seventh, ninth, and tenth cranial nerves [62]; the initial results were satisfactory in 50–100% of the cases [50,54,61,62] (> Table 124-1). Many complications resulted from the trigeminal tractotomy. Ataxia, resulting mainly from lesions of the spinocerebellar tract was described in more than 60% of the patients in the first series published in which the section of the descending tract was placed close to the obex [49,52,59,61]. After modifications of the original technique, the frequency of this complication fell to <5% [50,62,64,65]. Unwanted extension of analgesia is often presented after tractotomy. The area of thermoanalgesia usually extended
2103
2104
124
Technique of trigeminal nucleotractotomy
. Table 124-1 Results and complications of open trigeminal tractotomies Improvement, %
Authors
N, total
Etiology
Grant et al. [68]
12
Cancer
8
100
Trigeminal neuralgia
4
100
Grant and Weinberger [69]
Olivecrona [52]
19
Cancer
Trigeminal neuralgia Trigeminal neuralgia
N
12
83–89
7
100–83
34
90
6
17
13
70
Trigeminal neuralgia
39
13
Guidetti [59]
Trigeminal neuralgia Anesthesia dolorosa Postherpetic neuralgia Facial causalgia Trigeminal neuralgia
61 alleviation/ 39 improvement 100
Raney et al. [45]
Trigeminal neuralgia
59
Kunc [50]
Cancer Ninth nerve neuralgia
9 6
Grant [56] Hamby et al. [54]
Falconer [49]
Trigeminal neuralgia Cancer
2
100
4
0
1
0
127
9
100
Recurrence, %
24
Follow-up, months
Complications
1–13
Upper limb ataxia 25% Dysphagia (transient) 8%, Nystagmus 17% Neurological
Disturbances 60% Ataxia 66% (usually transient)
25
Herpes simplex 50% Ataxia Dysesthesia 43%, Mortality 17% Ataxia (permanent) 30% Ataxia (transient) many Herpes simplex 50% Mortality 5%
37
Ataxia (transient) 33% Ataxia (permanent) 12% Sensory ataxia 3%, recurrent nerve palsy 7% Contralateral thermoanalgesia 8% Paresthesias 7% Mortality 1.6% Herpes simplex (common) Dysarthria 2% Ataxia 2% Vagus nerve lesion 2% Mortality 1.4%
100
Technique of trigeminal nucleotractotomy
124
. Table 124-1 (Continued) Authors
Penzholz et al. [65]
Kunc [64] Plangger et al. [62]
N, total
Etiology Trigeminal and ninth nerve neuralgia Trigeminal neuralgia
Improvement, %
N
Recurrence, %
Follow-up, months
Complications
7
55
98
Trigeminal neuralgia Cancer
30
86
6
50
Trigeminal neuralgia Postherpetic neuralgia Traumatic neuropathy
12
58
1
100
29
Hemiparesis 5.5% Ataxia 1.8% Mortality 3.6% Ataxia 3% 72
1
beyond the trigeminal area to the areas innervated by the upper cervical roots and/or the seventh, ninth, and tenth cranial nerves [54]. Contralateral thermoanalgesia resulting from lesioning of the spinothalamic tract was described in a few patients after the trigeminal tractotomy [28,59]. In the past, it was believed that only pain and thermal fibers were present in the descending tract [5], so that touch sensation should be preserved and keratitis should be avoided after tractotomy. Indeed, touch sensation was preserved, at least partially, after the operation, but very often, tactile hypoesthesia was evident on clinical examination [61] but usually not noticed by the patients [6,45,59,61] and did not bother them [49]. Some patients, however, report a feeling of unpleasant sensations in the analgesic areas [28,49,59]. Paresthesia was mentioned by 25% of the patients of Grant and Weinberger [60], but fortunately, the denervation syndrome was not usually as disturbing as the original pain [49]. Corneal pain sensation was abolished, but the corneal touch was preserved and the blink reflex could usually be elicited [60] while the
corneal reflex became usually diminished or almost abolished by the tractotomy [22,49,61]. Fortunately, neuropathic keratitis developed only in few cases [49,59]. Kunc [51] hypothesized that the corneal reflex is preserved because the cornea has either tactile fibers and nonspecific fibers that are greatly dispersed through the subnuclei interpolaris and oralis so that tractotomy below the obex interrupted only a few of them. Lesions of the tenth nerve were described in 8–24% of the patients, especially when the lesions were made near the obex [61,63]. The lesion of the vagus nerve was less common when the section was made in the distal segment of the trigeminal tract [49,50,54,62,64,65]. Horner’s syndrome, headache [6,54,60], anxiety, and psychosis [59,60] were minor complications. The mortality ranged from 0 to 16.6% [45,54,59,64,65,70], being <5% in noncancer patients [45,49,59,65] and as high as 30% [50,54] in patients with cancer pains. However, despite the innovations [50,64], the problems associated with trigeminal tractotomy were very significant. The loss of pain sensation was often partial and patchy, and
2105
2106
124
Technique of trigeminal nucleotractotomy
sometimes no sensory change could be discerned [49,56]. Additionally, the high recurrence rate [67] and the frequency of morbidity and mortality have limited the use of Sjoqvist’s operation. Most neurosurgeons who have tried trigeminal tractotomy have been disappointed with the results, and the procedure has generally been discarded > Figure 124-3.
Stereotactic Trigeminal Nucleotractotomy The first application of a percutaneous procedure on the trigeminal descending tract was described by Crue and associates [71] in 1967. With the patient awake in the prone position, using a stereotactic frame and simple coordinates and electrophysiological control, and after a frustrating cisterna magna cisternography, they made a successful percutaneous radiofrequency lesion in the descending spinal trigeminal tract in a patient with facial pain caused by ethmoidal cancer, causing hypoalgesia in his ipsilateral forehead and occipital. In a nonrelated work,
. Figure 124-3 Schematic representation of posterior aspects of the lower brain stem, showing the multiple radiofrequency lesion sites during open trigeminal nucleotractotomy
Hitchcock in 1968 [53] described the full stereotactic trigeminal tractotomy. In 1970, he reported the results of the treatment of five patients with pain resulting from cancer, one with atypical facial pain, and one with trigeminal neuralgia. In 1972, Schvarcz [72] gave to the procedure the name trigeminal nucleotomy. This technique solved the problems described with the open procedures > Figure 124-4.
Trigeminal Nucleotractotomy with Cysternography The stereotactic trigeminal nucleotractotomy is performed after at least 8 h fast period and with the patient sedated with neuroleptics. Under local or general anesthesia with short-acting anesthetics the stereotactic frame is fixed to the head of the patient placed in a sitting [55], lateral decubitus [73], or prone position [74]. With the head kept flexed the cisterna magna is taped straight posteroanteriorly with a 18G needle, and positive contrast is injected for teleradiographic visualization of the craniocervical transition and delineation
. Figure 124-4 Schematic representation of posterior aspects of the lower brain stem, showing the radiofrequency lesion site during stereotactic trigeminal nucleotractotomy
Technique of trigeminal nucleotractotomy
of the anterior and posterior aspects of the spinal cord and of the medulla. The anteroposterior and lateral X-rays are taken, and the stereotactic coordinates are drawn [25,75]. The odontoid process is assumed to superimpose on the midline of the neuroaxis if there is no rotation of the head and the dorsal border of the cord is used as a reference for the target localization. The coordinates for the caudal dermatome (V3) are assumed to be 3 mm ventral to the dorsal aspect of the cord and 6 mm lateral to the midline. Immediately medially and dorsally lie the neurons receiving afferences from the seventh, ninth, and tenth cranial nerves. The rostral dermatome (V1) is located ventrolaterally, 5 mm in front of the dorsal border of the medulla and 7.5 mm laterally. The neurons related to the maxillary fibers lie between them > Figure 124-5.
Trigeminal Nucleotractotomy with Stereotomography In 1995, we described the method of target construction of the caudalis trigeminal nucleotractomy using the stereotomographic method. With the patient awake, supine and sedated with neuroleptics, the stereotactic fame is fixed to the skull under local anesthesia. Just after intravenous infusion of iodine contrast, 1-mm axial thickness stereotomographic slices of the
124
medulla and of the rostral cervical spinal cord are acquired and processed in a personal computer using a program for stereotaxis. The fusion of the computed tomography (CT) images with the magnetic resonance (MR) 1 mm T1 images of the craniovertebral region allows better delineation of the neural and surrounding structures, contemplating the better spatial resolution of the CT with the better anatomical resolution of the MR. This method makes possible the demonstration of the true shape and dimensions of the medulla and of the cervical spinal cord, and of the main arteries of the posterior fossa, steps important for the correct and safe positioning of the electrode. The calculus of the target point follows the same sequence as described for the trigeminal nucleotractotomy with cysternography.’’ In the operating room and in lateral decubitus with the ipsilateral hemiface placed in the right position and with the head slight flexed, the atantoaxial membrane is punctured with a 18G needle directed to the target point angled 30o lateral in the posteroanterior plane and 20o in the axial plane. In both procedures, and with the patient awake, a 0.5-mm diameter thermocoupleequipped electrode with a 2-mm bare tip is introduced through the guide needle and aimed at the target point (> Figure 124-6). Physiological corroboration of electrode placement is a
. Figure 124-5 (a) Intraoperative photograph of a patient in sitting position with the head fixed on a modified Hitchcock frame during stereotactic trigeminal nucleotractotomy guided by cysternography. (b) Lateral view of the intraoperative cysternography in the craniocervical region
2107
2108
124
Technique of trigeminal nucleotractotomy
. Figure 124-6 Representation of the computer templates of the stereotactic programming of stereoCT guided trigeminal nucleotractotomy. Illustration of a 3D reconstruction of the cranium and the probe trajectory through the craniocervical space
mandatory step because of cord mobility, individual anatomic variations, and proximity of important surrounding neuronal structures. Impedance recording indicates cord contact and penetration of nervous tissue, and monopolar stimulation (<1 V; 5–100 Hz) allows recognition of the position of the tip of the electrode. Stimulation of the target generates dysesthesias that are referred to the regions corresponding to the central projection of the primary afferent fibers. Ipsilateral sensory responses in the arm are produced when the electrode is placed medially and in contact with the cuneatus nucleus, contralateral sensory responses are elicited from stimulation of the spinothalamic tract that occurs when the electrode is placed anteriorly, and motor responses are produced when there is stimulation of the posteriorly located corticospinal tract. Trigeminal evoked potential recording [76] and neuroendoscopy [77] are useful adjuncts
to the precise localization of the electrode (see video) [Fonoff & Teixeira-personal communication]. When the electrode is appropriately placed, radiofrequency fractionated thermal electrocoagulations (70 C during 30 s) are performed, and the patients are tested for sensory loss [78]. At least three continuous lesions, are longitudinally placed in the subnucleus caudalis changing the axial angle of penetration of the needle from 10o to 30o in the axial plane (> Figure 124-6). As the lesioning procedure usually is very painful, short acting anesthetics can be safely used, when the patient is kept in lateral decubitus, or in sitting position. Trigeminal nucleotractotomy shares some of the features of open tractotomy but has other characteristics of its own that make it appropriate for the treatment of deafferentation pain. It differs from the open trigeminal tractotomy in that it targets not only the afferent
Technique of trigeminal nucleotractotomy
fiber tracts but also the second-order and intermuncial neurons and tracts in the spinal trigeminal complex [66] and eliminates abnormally discharging deafferented neurons [20,61,79,80] involved in the production of central or deafferentation pain in patients with trigeminal neuropathies [31,66]. Pain syndromes of extensive distribution are easily managed by a single lesion, as wide areas of analgesia, including C2-C3 dermatomes are obtained [78]. The circumferential analgesia produced with sparing of the central region of the face [53] is similar to that obtained by open section of the spinal tract [54]. Hyperactivity of deafferentated second-order neurons in the spinal trigeminal nucleus is one of the possible substrates for the constant orofacial paresthesias in patients with trigeminal neuropathy or mixed nociceptive and deafferentation facial pain [61]. Kerr [20] advocated the destruction of these neurons in humans to treat neuropathic facial pains. Lesioning of the primary pain afferent neurons and the hyperactive deafferentated secondorder neurons of the subnucleus caudalis [81] and elimination of the convergent inputs and of the intranuclear polysynaptic pathways [23,26] are the reasons for the disappearance of the hyperpathia and of the background pain in cases of trigeminal neuropathies after the nucleotractotomy [72,75,78], but not after medullary tractotomy [49]. This means that the trigeminal nucleotomy should be considered one of the most reliable method for treatment of the trigeminal, intermedius, glossopharyngeal, and vagal and cervical neuropathic pains, and also in the treatment of nociceptive craniorofacial and cervical pains [82,83,84]. Marked improvement of pain observed in 66–100% of the patients operated on [25,53,55,76,85]. Early improvement in trigeminal neuralgia and neuralgia of the ninth and tenth cranial nerves has been observed in almost all cases reported [25,53,55,76], but the recurrence rate is very high [76]. In contrast to the open trigeminal tractotomy, which has
124
consistently failed to relieve postherpetic facial pain [49], the results of the trigeminal nucleotractotomy are encouraging in these cases [75,82,83]. Allodynia is relieved and there is reduction or marked disappearance of the deep background pain in 76–100% of the patients with trigeminal, intermedius nerves and upper cervical roots postherpetic pain during follow-up periods ranging from 2 months to 13 years after a single lesion [25,55,75,82,85,86]. There is improvement of the suffering in 57–100% of the patients with anesthesia dolorosa of the face [25,55,76]. It is also very effective for the treatment of facial pain caused by multiple sclerosis and or traumatic trigeminal neuropathies [25,55], but not in cases of atypical facial pain [87]. Teixeira et al. observed improvement in 85.7% of seven patients with pain resulting from Wallenberg’s syndrome during a mean follow-up period lasting 24 months (> Table 124-2). As the procedure is performed under local anesthesia, sedation or short acting anesthesia, it is possible to monitor the result of the stimulation and the extent of the controlled selective lesioning of the nervous system. With patient cooperation in response to neurostimulation, the functional features of the region of the trigeminal nuclear complex and tract in that the electrode is placed are easily evaluated for correct positioning of the lesion by recording the patient’s reaction. The clinical examination just after each lesion allows very good documentation of the degree of the analgesia induced [75,78]. The method also is safe even in debilitated patients, because can be made under local anesthesia. Troublesome anesthesia is avoided, and both corneal reflex and facial touch sensation are partially preserved after the trigeminal nucleotomy [85,86]. Deafferentation pain was described in a few patients after trigeminal nucleotractotomy [85]. Contralateral thermoanalgesia usually is rare [85] and has been described in <33% of the patients [25,75,76,86]. Permanent upper limb ataxia is observed in <10% of the
2109
2110
124
Technique of trigeminal nucleotractotomy
. Table 124-2 Results and complications of stereotactic trigeminal nucleotractotomy
Reference Todd et al. [74]
Hitchcock [53]
Crue et al. [76]
Hitchcock and Schvarcz [75]
N, total 3
Etiology
N
Immediate improvement, %
Late Improvement, %
Follow-up, months
2
7
Cancer/ tumor Trigeminal neuralgia Cancer
5
80
2–12
1 1
0 100
Mean 7
12
Atypical Trigeminal neuralgia Cancer
8
100
75
Trigeminal neuralgia Anesthesia dolorosa Postherpetic neuralgia
3
33
33
1
100
100
3
100
66
3
100
1
Upper limb ataxia 8% (transient)
4–11
Mean 7 Schvarcz [25]
25
Hitchcock [85]
21
Schvarcz [86] Schvarcz [55]
6 100
Cancer/ tumor Multiple sclerosis Postherpetic neuralgia Anesthesia dolorosa Trigeminal neuralgia Ninth nerve neuralgia Postherpetic neuralgia + cancer Postherpetic neuralgia Cancer Multiple sclerosis Postherpetic neuralgia Anesthesia dolorosa Trigeminal neuralgia
Complications
8
100
1
100
1
100
3
67
11
100
1
100
21
12–16
Contralateral hypalgesia ?
47 66
4–18
6
100
>6
31
84
2–33
1 8
88
14
57
19
Contralateral hypoalgesia 17% Contralateral lower limb thermoanalgesia 33% Ipsilateral upper limb 33% ataxia Ataxia 8%
Ataxia or paresis 44%, Loss of corneal reflex 5%, dysesthesia 5% Contralateral some hypalgesia
Technique of trigeminal nucleotractotomy
124
. Table 124-2 (Continued)
Reference
Schvarcz [82]
Teixeira et al. [88]
N, total
32
58
Etiology Ninth and tenth nerve neuralgia Traumatic trigeminal neuropathy Trigeminal postherpetic neuralgia Postherpetic neuralgia C2-C3 postherpetic neuralgia Post-herpetic neuralgia Cancer
Wallenberg’s syndrome Atypical facial pain Trigeminal neuralgia
N
Immediate improvement, %
Late Improvement, %
Follow-up, months
Complications
2
25
Total 72
23
76
1–156
4 5
28
100.0
78.5
37.5
14
92.9
85.7
6
7
85.7
85.7
24
7
71.4
42.8
14
2
0
0
36
cases [25,76,89]. There is no documentation of mortality with stereotactic nucleotractotomy. This means that stereotactic trigeminal nucleotractotomy is safe and presents the additional advantage of minimal invasiveness, especially when the stereotomographic method is used which dispenses cysternography. As no important scar is induced, it can also be effective for repeated applications.
Percutaneous Freehand Trigeminal Tractotomy Free Hand Trigeminal Tractotomy In 1971, Fox [48] described the freehand percutaneous trigeminal tractotomy. However, very few authors used the original method [90].
Upper limb ataxia (transient) 79.3% Upper limb ataxia (permanent) 10.3%
With the patient awake and placed in the prone position and with the head immobilized by a head support, opaque contrast is injected into the cisterna magna to outline the floor of the fourth ventricle, the obex, and the dorsum of the brain stem. With the odontoid process used as a reference point for the determination of the midline in the posteroanterior film, a separate guide needle is passed through the skin approximately 4 cm from the midline under radiographic control and directed to the trigeminal tract at or just below the obex on the side of the pain. Further an electrode is inserted through the guide needle into the cerebrospinal fluid and into the medulla oblongata, aiming at a point 2–3 mm in front of the roof of the fourth ventricle. The monitoring of electrical impedance allows identification of the moment when the electrode penetrates the nervous structures, and
2111
2112
124
Technique of trigeminal nucleotractotomy
the electrical stimulation confirms its location. Finally, the lesion can be made with radiofrequency current [48,90].
. Figure 124-7 CT slice of the craniocervical region showing intraoperative programing of trigeminal nucleotractotomy based in odontoid process projection to the lower medulla
Free Hand Trigeminal Tractotomy Under Tomographic Control In 1989, Kanpolat and associates [91] described the method of freehand percutaneous trigeminal nucleotractotomy under CT control. The operation is performed with the patient awake. Before the procedure, neuroleptic anesthesia is administered at a dose that does not affect the patient’s cooperation. Radiopaque contrast material is administered into the subarachnoid space by lumbar puncture, 20 min before the procedure. With the patient placed in the prone position on the CT table, and with the head slight flexed with the head support of the CT scan table and fixed with a fixation band, a lateral scanogram and axial scans are obtained in the C1-occiput region. The cord diameters are measured and the target point is drawn 3 mm anterior to the posterior aspect of the spinal cord and 5–6 mm lateral to the midline at the first cervical segment. A 20G needle is then inserted in the C1-occiput region, 5–8 mm lateral to the midline and its tip is positioned into the subarachnoid space. Through it, a 2-mm exposed tip, 0.3 mm diameter, straight or curved electrode is inserted, aiming toward the lateral third of the transverse diameter (equator) of the hemicord. The final position of the tip is confirmed with a repeated 1 mm slice CT scans images, impedance measurements, and electrical stimulation, followed by 1–3 radiofrequency lesions made at 65–75 C for 60 s. Special electrodes were developed for this procedure aiming the reduction of the imaging artifacts [92] > Figure 124-7. Few papers about free hand trigeminal tractotomy were published [91,93,94], all suggesting that its effectiveness is high and the complication rate low [11,36]. Trigeminal tractotomy is very
effective in patients with craniofacial cancer pain, even when the pain is present in the sensory areas of the seventh, ninth, and tenth cranial nerves; the improvement rate expected with the procedure is of 80% [90,93]. The results of the treatment in cases of atypical neuralgia are disappointing [90]. The patient with trigeminal neuralgia operated by Fox did not improve [90]. Kanpolat et al. [95], by himself operated six patients with vagoglossopharyngeal neuralgia and three with intermedius nerve neuralgia. From them, one, in whom the pain relief was excellent during the early postoperative period, was lost to follow-up. Pain recurred in three during the follow-up period of the remaining patients that lasted 5–84 months. After repetition of the procedure, excellent (five patients) or good (three patients) permanent pain relief was obtained in each patient. Postoperative ipsilateral ataxia occurs frequently, but usually is transient [48,90]. Contralateral thermoanalgesia is observed in 25–33% of the patients [48,90]. Dysarthria and Homer’s syndrome are rare. Fever is very common [48]. The percutaneous method is free of mortality (> Table 124-3).
Technique of trigeminal nucleotractotomy
124
. Table 124-3 Results and complications of freehand trigeminal tractotomy Reference
N, total
Etiology
N
Fox [48]
13
Fox [90]
18
Cancer Postherpetic neuralgia Trigeminal neuralgia Anesthesia dolorosa Posttraumatic Cancer Postherpetic neuralgia Atypical neuralgia Trigeminal neuralgia
8 2 1 1 1 14 2 1 1
88 0 0 0
Kanpolat et al. [92]
19
5
80
Kanpolat et al. [95]
9
Trigeminal neuralgia Seventy nerve neuralgia
14 6
100
3
100
52
Ninth and tenth nerves neuralgia Atypical facial pain
16
44
Craniofacial cancer Trigeminal neuralgia Seventy nerve neuralgia Ninth nerve neuralgia Postherpetica neuralgia Anestesia dolorosa
11 3 4 13 4 1
82 67 65
Kanpolat et al. [96]
Cancer
Despite the possibility of the real time image control of the location of the electrode, the electrophysiological corroboration of the needle position still is critical, to improved anatomic localization by computed tomographic guidance [97]. One of the main disadvantages of the method is that it is time-consuming and its application may be difficult in many places unless a CT machine is available, and dedicated for use in neurosurgery. As the procedure is performed with the patient awake and in prone position, short-acting anesthetics and strong analgesic medications cannot be adequately administered. As a result, the operation may not be well-tolerated by elderly or apprehensive patients, especially during needle advancement and lesion making. A sudden movement of a patient experiencing pain or significant discomfort when the needle is being inserted into the spinal canal, or the electrode in
Results, %
Complications Horner’s syndrome 8% Dysarthria (transient) 8% Contralateral hypalgesia 28% Ataxia (transient) 100% Fever (almost all) Ataxia 6% Hiccoughs 16% Contralateral lower limb analgesia 30%
5–54 months (mean 49.5)
Ataxia 10%
50 0
the trigeminal tract or nucleus or during the lesioning may cause a complication [98].
Caudalis Drez Lesions In 1971, Hosobuchi and Rutkin [22] developed the method of open trigeminal tractotomy using radiofrequency current. More than 10 years later Nashold and associates [98] and Siqueira [57], improved this technique by placing multiple radiofrequency lesions into the subnucleus caudalis supposing that multiple lesions were required to relieve the pain [89]. Later, many authors published papers describing their experiences with this technique [66,70,84,99–101] under the name of nucleus caudalis DREZ lesions. The procedure is performed with the patient prone and under general anesthesia and with
2113
2114
124
Technique of trigeminal nucleotractotomy
magnification. Through a small suboccipital craniectomy and C1-C2 laminectomy, the tonsil is elevated and the obex is visualized. The caudal subnucleus of the trigeminal spinal nucleus occupies the triangular area between the dorsolateral sulcus and the emerging points of the accessory nerve. At the level of the C2 root, the subnucleus caudalis blends within the posterior horn of the spinal cord and is covered dorsolaterally by a thin layer of the descending trigeminal tract. Near the obex, it is covered by a layer of the external arcuate and dorsal spinocerebellar fibers. It is widest in the rostral area and tapers caudally until it joins the spinal DREZ of C2. Initially just one, and in operations performed latter, two series of rostrocaudal and mediolateral radiofrequency lesions at 75–80 C for 15 s are made using an electrode with a 1.5–2-mm tip exposure covering all of the subnucleus caudalis region [66]. Relief or significant improvement in suffering was observed in 53–94% of the patients with facial pain treated with open caudalis DREZ lesions [32] during a short postoperative follow-up period [66,102]. As with other pain procedures, the success rate is higher in the immediate postoperative period than after long-term follow-up [66]. The overall success falls to 58% after a mean follow-up of 9.8 months. According to Bernard and associates [66], patient’s description of the pain may be useful in predicting the postoperative results. Better results are observed in cases of sharp or burning pain, than in cases of dull pain. They observed that 41 and 55% of patients with sharp and burning pain, respectively, improved after the procedure, in contrast to 24% with dull pain, finding confirmed by others [73,100]. The correlation between the preoperative sensory deficit and the effectiveness of the operation is controversial. The results seem to be less satisfactory in patients with greater compromise of the afferent fibers. Better results were also observed when pain involved one or a few of the trigeminal
nerve and when it did not extended beyond the trigeminal territory. They also observed good results in 75% of patients with pain affecting the territory of within one trigeminal branch, 50% of patients with pain over two branches, and 38% of the patients when all trigeminal branches territories were involved. The duration of the pain complaints did not affect the final results [66]. Good results occurred in patients with brain stem lesions, and fair to poor results in patients with anesthesia dolorosa and traumatic injuries to the trigeminal nerve, particularly those with dull pain [89]. The number of patients with trigeminal neuralgia reported treated by caudalis DREZ lesions is small. Trigeminal neuralgia improved in just 20–66% of the patients of Bernard and associates [66,81] and all seven patients of Morita and Hosobuchi [103]. Caudalis DREZ lesions are also effective for the treatment of facial dysesthesias (anesthesia dolorosa) [102], but the number of cases reported is small [22,70,84]. Pain related to traumatic trigeminal neuropathies can also be alleviated [70,98,99]. Pain resulting from surgical procedures in the face or related structures was improved in 33.3% of the patients of Bernard and associates [66] Improvement occurred in 66–100% of patients with postherpetic neuralgia [31,32,66,70,84,102], being the best long-term results occurring in this group, finding not in agreement with others [22]. There are few papers about the results of the procedure in the treatment of central pain caused by vascular lesions of the brain stem. Sampson and Nashold [78] performed the procedure in two patients with facial pain resulting from infarction of the brain stem; one patient, presented complete alleviation of the pain, and in the other, the improvement was partial. Gorecki and Nashold [102] observed improvement in four patients with facial pain associated with cerebral stroke. This author also tried the procedure in one patient with facial pain associated with brain stem infarction without
Technique of trigeminal nucleotractotomy
success. The long-term success rate of the nucleus caudalis DREZ operation is low in cases of craniofacial pain resulting from malignancies; the initial improvement observed in almost 100% of the patients fell to 40–60% some months after the procedure [20,70,98,100] (> Table 124-4). The morbidity rate of the procedure is high [100]. Open caudalis DREZ lesions result in more complications than does stereotactic nucleotractotomy. The radiofrequency lesions may not only involve the subnucleus caudalis, the descending tract, and the ascending projections to other trigeminal nuclear structures and the midbrain and thalamic nuclei but also may extend medially and ventrally into nearby structures [104]. The complications are related to injury to the nearby structures of the spinal cord and medulla. Ipsilateral paresis is related to involvement of the corticospinal tract caused by lesions that are too deep, contralateral hypalgesia to lesions of the spinothalamic tract, and ataxia to damage to the spinocerebellar tracts and posterior columns [31,102]. Extension of analgesia into the area of the C1 root is very common [31]; contralateral thermoanalgesia occurs in 17–25% of these patients [22,84], ataxia affecting the upper and lower limbs in of 30–39% [66,70] and the upper limb in 17–61% [22,31,66,70,78,99,102,105]. The incidence of ipsilateral upper limb or upper and lower limb dysmetria fell from 74 to 39% [102] when a special angled electrode was used. Motor deficits were described in a few cases [84,104]. Brown–Sequard syndrome, eleventh nerve palsy, hemiparesis [104], cerebraospinal fluid leak, meningitis, pneumonia, myocardial infarction, and intraoperative stroke have also been described with caudalis DREZ lesions [66,70]. Deaths are rare [102,104]. The rostral upper lesions are placed 5 mm below the obex [5] and the caudal at the upper dorsal rootlets of C2. The caudalis subnucleus merges with the dorsal horn of the C2 root and extends rostrally for 20 mm like a continuation
124
of the dorsal horn to the subnucleus interpolaris at the level of obex and has a fusiform configuration with the diameter ranging from slightly more than 2 mm caudally to 1.1–1.4 mm at C1 and the obex. This means that a single row of lesions should be enough to destroy the subnucleus caudalis externally from the C2 rootlets rostrally in line with the DREZ [97], instead of two or more rows to induce significant more sensory deficits at the cost of other deficits as well as other complications [89] (> Figure 124-8). Based on an anatomopathological case [105], it is possible that the technique of just one line of radiofrequency lesions proposed by Siqueira [57] is safer than the proposed by Nashold and coworkers [98]. The subnucleus caudalis lies 1.2 mm beneath the surface of the medulla, and rostrally is covered by the spinal trigeminal and spinocerebellar tracts and caudally by the thin spinal tract. To optimize the results and minimize the complications of the unintended lesion in the superficial pathways, Nashold et al. [106] developed one right-angled electrode for lesions between C1 and C2 region and another for lesions between the obex and C1, with insulated surface in the areas of its contact with the spinocerebellar tract. The use these specially designed electrodes for nucleotractotomy [101,102,105] or of an ultrasound needle reduced the complication rate even when the lesions were placed at high levels [31] > Figure 124-9.
Pontine Stereotactic Trigeminal Nucleotractotomy In 1987, Hitchcock and Teixeira [58] described the pontine stereotactic trigeminal nucleotractotomy. Under stereotactic conditions, with the patient in a sitting position, the fourth ventricle is outlined by positive ventriculography or steretomography. Using the fastigial line as a
2115
8 9
Atypical facial pain Postherpetic neuralgia Anesthesia dolorosa Cancer
Postherpetic neuralgia Traumatic trigeminal neuropathy Cancer Postherpetic neuralgia Trigeminal neuralgia Dental operations Anesthesia dolorosa Glaucoma
Bernard et al. [70]
Ishijima et al. [84]
Bernard et al. [66]
Improved 25
1 4 5 9 3 3 1
Trigeminal neuralgia Postherpetic neuralgia Anesthesia dolorosa Postdental procedure Glaucoma
33
20 67
100
1
Postherpetic neuralgia + anesthesia dolorosa Anesthesia dolorosa Cancer 4
100
100 71 66 100 100 00
100 100 100 78
100
Result, %
2
19
18
6
N, total
Postherpetic neuralgia
3 7 3 3 21
1 1 1 1
Cancer
Hosobuchi and Rutkin [22]
Nashold et al. [98]
3
Etiology
Reference
Improved 52
0
100
100 100 0
100
Recurrence, %
60%-0
Repeat 15
2–52 Mean, 44
<48 Mean, 9.8
Mean, 8 12 2 24
4–15
Follow-up, months
Meningitis 4%
Upper + lower limb dysmetria (usually transient)
Upper limb dysmetria 44%
Axial ataxia 25%, upper limb hypoesthesia 25%
Sensorimotor (transient) 25%
Stroke 5% Myocardial infarction and pneumonia and meningitis 5%
Upper limb dysmetria 61%, upper and lower limb dysmetria 39%
+Contralateral analgesia (Transient) 17%
Upper limb ataxia
Complications
124
N
. Table 124-4 Results and complications of caudalis DREZ lesions
2116 Technique of trigeminal nucleotractotomy
Nashold et al. [105] Gorecki and Nashold [102]
Abdennebi et al. [99] Grigoryan et al. [31]
Rossitch et al. [100] Sampson and Nashold [78] Morita and Hosobuchi [103]
8 5
Atypical facial pain Anesthesia dolorosa stroke Facial trauma/surgery Headache Multiple sclerosis Trigeminal tumor Trigeminal actinic neuropathy 32
?71
8
Postherpetic neuralgia
4 4 1 1 1
53
16
4 1
Postherpetic neuralgia Cancer Many conditions
Excellent/good 74
100
85 50
100
89
100
100
9
7
A¨lleviation or partial improvement 50 100
Atypical facial pain
7
Trigeminal neuralgia
7
60 partial
100
2
2
Brain stem lesion
28
85
Postherpetic neuralgia
1 1 1 5
Trauma Postamygdalectomy Salivary calculus Cancer
Mean
9–180
Mean 26
4–48
+death 2%
Ataxia 39%
Ataxia 33%
Ipsilateral limb paresis 14%
Upper limb ataxia 22%
Anesthesia of ipsilateral shoulder (permanent) 29% Upper limb paresis (transient)14% Upper limb ataxia 50%
Upper limb ataxia (transient) 50%, (permanent) 50% Lateral pulsion (transient)43%
Cerebrospinal fluid leak 20%
Stroke 4% Myocardial infarction 4%
Technique of trigeminal nucleotractotomy
124 2117
2118
124
Technique of trigeminal nucleotractotomy
. Figure 124-8 Representation of the computer templates of the stereotactic programming of stereoCT/MRI fusion guided caudalis trigeminal nucleotractotomy. T1-weighted images show the safe angle and trajectory of the RF electrode in axial, sagittal, and coronal view
reference, the target point is located 5–10 mm from the midline, from 1 mm posterior to 6.6 mm anterior to the floor of the fourth ventricle, and from 10.5 mm rostral to 1.2 mm caudal to the fastigial line. Through a burr hole placed parasagittally to the target in the ipsilateral suboccipital region, an electrode is introduced and directed to the target point. Impedance monitoring helps the identification of the point of the contact of the electrode with tracts and nuclei and the electrical stimulation allows the identification of the neural structures. The stimulation of the target point (<1 V; 5–100 Hz) results in ipsilateral facial paresthesias; stimulation of the trigeminal quintothalamic tract causes paresthesias in the contralateral hemiface; stimulation of the trigeminal motor nucleus causes contraction of the ipsilateral masseter muscle; stimulation of the facial nucleus causes contraction of the ipsilateral facial muscles; and the stimulation of the vestibular nucleus results in ipsilateral
buzzing sensation. When the electrode is appropriately placed, a radiofrequency lesion is made, and the sensibility of the face evaluated (> Figure 124-10). Hitchcock and Teixeira [58] observed that three out of six patients with anesthesia dolorosa of the face improved after the procedure without complications. Facial analgesia induced by caudal nucleotractotomy is incomplete in distribution and degree [6,102,105,107]. This can be attributed to incomplete tract or nuclear lesion [40,56,107] or to the representation of the oral and perioral regions of the face rostrally within the trigeminal sensory complex, in areas placed rostrally to the subnucleus caudalis [107]. As the medullary tractotomy has a quantitative rather than a qualitative effect on pain sensation, it is possible that the reduction of the overall central summation of afferent inputs in the trigeminal nucleus results in hypoalgesia [40]. These are the reasons for lesioning the most rostral levels of the
Technique of trigeminal nucleotractotomy
124
. Figure 124-9 Postoperative MRI showing the lesion spot in the posterolateral aspescts of medulla, represented by a dot slightly hyperintense surrounded by hypointense edema in a larger area
. Figure 124-10 (a) Representation of the computer templates of the stereotactic programming of stereoCT/MRI fusion guided pontine trigeminal nucleotractotomy. After CT/MR fusion a corresponding fiber stained brainstem was merged into the image study for illustrative purposes. The fastigial line is an important anatomic landmark used in this procedure
2119
2120
124
Technique of trigeminal nucleotractotomy
trigeminal descending tract and nuclear complex, especially for the treatment of perioral, perinasal, and oral pain [58].
Conclusion Trigeminal nucleotractotomy is an additional option for the treatment of intractable facial pain. In addition to the classic indications, available data suggest novel ones, highlighting the special significance of lesioning or the subnucleus caudalis. The results are encouraging, especially in patients with cancer pain and deafferentation states in the areas of the fifth, seventh, ninth, and tenth nerves [91]. Postherpetic neuralgia, Wellemberg’s syndrome and malignant orofacial pain are the most common indications of the procedure [82,85]. Trigeminal nucleotractotomy can be performed by different surgical techniques. Currently, a single or few radiofrequency lesions made stereotactically or freehand (trigeminal nucleotomy) [22,25,74] or multiple radiofrequency lesions of the primary and secondary afferent neurons (trigeminal caudalis DREZotomy) [57,66] are most often used. The main difference between the two procedures is that destruction of the entire length of the subnucleus caudalis is possible only with the latter. However, the concept that the effect of DREZotomy parallels the completeness of destruction of neurons, as in DREZ of the spinal cord [84], is not applicable to the trigeminal area, since isolated lesions result in similar improvement [55,82]. Percutaneous, especially the stereotactic trigeminal tractotomy is a physiologically convincing method that has definite advantages over open procedures. Since it can be performed under local anesthesia, it is safer even in patients in very poor general condition. The location of the electrode can be determined precisely by evoking sensation during electrical stimulation and questioning the patient during the operation, and the extension of the lesion can be controlled by examining the patient awake [87].
The percutaneous procedure can also be repeated easily and without higher risk when necessary. The stereotomographic trigeminal nucleotractotomy is more comfortable and safer than the stereotactic or free hand procedure with the help of cysternography, because, do not involve the instillation of opaque contrast into the intracranial cysterna, and can be made with the patient in lateral decubitus, position that allows the use of short-acting anesthetics and aspiration of oropharyngeal secretion or vomits that can occur during the operation [73]. Open nucleotractotomies are indicated in uncooperative patients. They are performed under general anesthesia, which precludes the participation of the patient during the stimulation of the neural structures and did not allow the intraoperative control of the sensory abolition induced by the operation. These are, among others, reasons that it usually induces analgesia in all divisions of the trigeminal nerve and more neurological complications is higher than the percutaneous or stereotactic nucleotractotomy. As they are performed under general anesthesia and need opening and, consequently, of the structures that surrounds the craniovertebral region is also associated with many clinical complications. The use of the surgical microscope and of adequate equipments reduces the incidence of permanent complications seen with the earlier open trigeminal tractotomies [75]. It is possible that, as intraoperative recording becomes more sophisticated, the lesion will be made more selectively in the future. Stereotactic pontine nucleotractotomy is indicated in cases of deafferentiation extending through a long distance in the trigeminal nuclear complex. When pain cannot be alleviated by medullary tractotomy, additional lesions should be produced at higher levels in the descending nuclei, which are also involved in facial nociception [38]. Identification of the factors related to better outcomes is difficult. When pain is restricted to small areas and has a sharp or burning character, the prognosis seems to be better [66]. Failures are
Technique of trigeminal nucleotractotomy
attributable to inadequate lesioning of neuronal targets or to the existence of abnormally functioning neurons in more rostral locations [66].
References 1. Stookey B, Ransohoff F. Trigeminal neuralgia: its history and treatment. Springfield, IL: Charles C Thomas; 1959. 2. White JC, Sweet WH. Pain and the neurosurgeon: a forty-year experience. Springfield, IL: Charles C Thomas; 1969. 3. Pagni CA. Central pain and painful anesthesia. Prog Neurol Surg 1977;8:132-257. 4. Dallel R, Raboisson P, Auroy P, Woda A. The rostral part of the trigeminal sensory complex is involved in orofacial nociception. Brain Res 1988;448:7-19. 5. Sjoqvist O. Studies on pain conduction in the trigeminal nerve: contribution to surgical treatment of facial pain. Acta Psychiatr Scand Suppl 1938;17:1-139. 6. Weiberger LM. Experiences with intramedullary tractotomy. Arch Neurol Psychiatr 1942;48:355-81. 7. Stewart WA, Stoops WL, Pillone PR, King RB. An electrophysio logic study of ascending pathways from nucleus caudalis of the spinal trigeminal nuclear complex. J Neurosurg 1964;21:35-48. 8. Kerr FWL. The divisional organization of afferent fibers of the trigeminal nerve. Brain 1963;86:721-32. 9. Thelander HE. The course and distribution of the radix mesencephalica trigemini in the cat. J Comp Neurol 1924;37:207-20. 10. Taren JA, Kahn EA. Anatomic pathways related to pain in face and neck. J Neurosurg 1962;19:116-21. 11. Crosby EC, Yoss RE. The phylogenetic continuity of neural mechanisms as illustrated by the spinal tract of V and its nucleus. Res Publ Assoc Res Nerv Ment Dis 1954;33:174-208. 12. Sessle BJ. Trigeminal nociceptive and non-nociceptive neurones brain stem intranuclear projections modulation by orofacial periaquedutal grey and nucleus magnus stimuli. Brain Res 1979;170:547-52. 13. Olzewski J. On the anatomical and functional organization of the spinal trigeminal nucleus. J Comp Neurol 1950;92:401-13. 14. Azerad J, Woda A, Albe-Fessard D. Physiological properties of neurons in different parts of the cat trigeminal sensory complex. Brain Res 1982;246:7-21. 15. Darian-Smith I, Yokota T. Cortically evoked depolarization of trigeminal cutaneous afferent fibers in the cat. J Neurophysiol 1966;29:170-8. 16. Brodal A. Central course of afferent fibers for pain in facial, glossopharyngeal and vagus nerves. Arch Neurol Psychiatr 1947;57:292-306.
124
17. Kerr FWL. Facial, vagal and glossopharyngeal nerves in the cat: afferent connections. Arch Neurol 1962;6:264-81. 18. Brodal A, Szabo T, Torvik A. Corticofugal fibers to sensory trigeminal nuclei and nucleus of solitary tract: an experimental study in the cat. J Comp Neurol 1956;06:527-55. 19. Kerr FWL. Structural relation of the trigeminal spinal tract to upper cervical roots and the solitary nucleus in the cat. Exp Neurol 1961;4:134-48. 20. Kerr FWL. Spinal 5th nucleolysis: intractable craniofacial pain. Surg Forum 1966;17:419-21. 21. Kerr FWL, Olafson RA. Trigeminal and cervical volleys. Arch Neurol 1961;5:69-76. 22. Hosobuchi Y, Rutkin B. Descending trigeminal tractotomy: Neurophysiological approach. Arch Neurol 1971;25:115-25. 23. Wall PD, Taub A. Four aspects of trigeminal nucleus and a paradox. J Neurophysiol 1962;25:110-26. 24. Broton JG, Rosenfeld JP. Rostral trigeminal projection signal perioral facial pain. Brain Res 1982;243:395-400. 25. Schvarcz JR. Stereotactic trigeminal tractotomy. Confin Neurol 1975;37:73-7. 26. Stewart WA, King RB. Fiber projections from the nucleus caudalis of the spinal trigeminal nucleus. J Comp NeuroI 1963;121:271-86. 27. Young RF, Perryman KM. Neuronal responses in rostral trigeminal brain-stem nuclei of macaque monkeys after chronic trigeminal tractotomy. J Neurosurg 1986;65:508-16. 28. McKenzie KJ. Trigeminal tractotomy. Clin Neurosurg 1955;2:50-70. 29. Young RF. Effects of trigeminal tractotomy on dental sensation in humans. J Neurosurg 1982;56:812-18. 30. Young RF, Perryman KM. Pathways for orofacial pain sensation in the trigeminal brain-stem nuclear complex of the macaque monkey. J Neurosurg 1984;61:563-8. 31. Grigoryan YA, Slavin KV, Ogleznev KY. Ultrasonic lesion of the trigeminal nucleus caudalis for deafferentation facial pain. Acta Neurochir (Wien) 1994;131:229-35. 32. Mosso JA, Kruger L. Spinal trigeminal neurons excited by noxious thermal stimuli. Brain Res 1972;38:206-10. 33. Nord SG, Ross GS. Responses of trigeminal units in the monkey bulbar lateral reticular formation to noxious and non-noxious stimulation of the face: experimental and theoretical considerations. Brain Res 1973;58:385-99. 34. Gordon G, Landgren S, Seed WA. The functional characteristics of single cells in the caudal part of the spinal nucleus of the trigeminal nerve of the cat. J Physiol 1961;158:544-59. 35. Arvidson J, Gobel S. An HRP study of the central projections of primary trigeminal neurons which innervate both pulp in the cat. Brain Res 1981;208:1-16. 36. Gerard MW. Afferent impulses of the trigeminal nerve: the intramedullary course of the painful, thermal and tactile impulses. Arch Neurol Psychiatr 1923;9:306-38.
2121
2122
124
Technique of trigeminal nucleotractotomy
37. Mattews B, Baxter J, Watts S. Sensory and reflex responses to tooth pulp stimulation in man. Brain Res 1976;113:83-94. 38. Grant FC. Discussion on trigeminal tractotomy. Clin Neurosurg 1955;2:69-70. 39. Eskine CA, Rowbotham FG. Late changes in the brain stem following trigeminal tractotomy. Arch Neurol Psychiatr 1949;62:493-502. 40. Denny-Brown D, Yanagisawa N. The function of the descending root of the fifth nerve. Brain 1973;96:783-814. 41. Shigenaga Y, Chen IC, Suemune S, et al. Oral and facial representation within medullary and upper cervical dorsal horn in the cat. J Comp NeuroI 1986;243:388-408. 42. Hochfield S, Gobel S. An anatomical demonstration of projections to the medullar dorsal horn (trigeminal nucleus caudalis) from rostral trigeminal nuclei and the contralateral caudal medulla. Brain Res 1982;252:203-11. 43. Hodge CJ Jr, King RB. Medical modification of sensation. J Neurosurg 1976;44:21-8. 44. Sessle BJ, Greenwood LF. Inputs to trigeminal brain stem neurons from facial, oral tooth pulp and pharyngolaryngeal tissues. I. Responses to innocuous and noxious stimuli. Brain Res 1976;117:221-6. 45. Raney R, Raney AA, Hunter C. Treatment of major trigeminal neuralgia through section of the trigeminospinal tract in the medulla. Am J Surg 1950;80:11-17. 46. Harrison F, Corbin KB. Oscillographic studies on the spinal tract of the fifth cranial nerve. J Neurophysiol 1942;5:465-82. 47. MacKinley WA, Magoun HW. The bulbar projection of the trigeminal nerve. Am J Physiol 1942;137:217-24. 48. Fox JL. Intractable facial pain relieved by percutaneous trigeminal tractotomy. JAMA 1971;218:1940-1. 49. Falconer MA. Intratuedullary trigeminal tractotomy and its place in the treatment of facial pain. J Neurol Neurosurg Psychiatry 1949;12:297-311. 50. Kunc Z. Treatment of essential neuralgia of the 9th nerve by selective tractotomy. J Neurosurg 1965;23:494-500. 51. Kunc Z. Significance of fresh anatomic data on spinal trigeminal tract for possibility of selective tractotomies. In: Knigton RS, Dumke PR, editors. Pain: Henry Ford Hospital International Symposium. Boston: Little Brown; 1966. p. 351-63. 52. Olivecrona H. Tractotomy for relief of trigeminal neuralgia. Arch Neurol Psychiatr 1942;47:544-654. 53. Hitchcock E. Stereotactic trigeminal tractotomy. Ann Clin Res 1970;2:131-5. 54. Hamby WB, Shinners BM, Marsh IA. Trigeminal tractotomy. Arch Surg 1948;57:171-7. 55. Schvarcz JR. Spinal cord stereotaxic techniques for trigeminal nucleotomy and extralemniscal myelotomy. Appl Neurophysiol 1978;41:99-112. 56. Grant FC. Complications accompanying surgical relief of pain in trigeminal neuralgia. Am J Surg 1948;75:42-47. 57. Siqueira JM. A method for bulbospinal trigeminal nucleotomy in the treatment of facial deafferentation pain. Appl Neurophysiol 1985;48:277-80.
58. Hitchcock E, Teixeira MJ. Pontine stereotactic surgery and facial nociception. Neurol Res 1987;9:113-17. 59. Guidetti B. Tractotomy for relief of trigeminal neuralgia: observations in 124 cases. J Neurosurg 1950;7:499-508. 60. Grant FC, Weinberger LM. Experiences with intramedullary tractotomy. IV. Surgery of the brain stem and its operative complications. Arch Surg 1941;42:747-54. 61. Anderson L, Black RG, Abraham J, Ward AA Jr. Neuronal hyperactivity in experimental trigeminal deafferentation. J Neurosurg 1971;35:444-52. 62. Plangger CA, Fischer J, Grunert V, Moshsenipour I. Tractotomy and partial vertical nucleotomy for treatment of special forms of trigeminal neuralgia and cancer pain of face and neck. Acta Neurochirurg Suppl (Wien) 1987;39:147-50. 63. Grant FC. Surgical methods for relief of pain. Bull NY Acad Med 1943;19:373-85. 64. Kunc Z. Vertical trigeminal partial nucleotomy. Adv Pain Res Ther 1979;3:325-9. 65. Penzholz H, Menzel J, Hagenlocher HU. Results of surgical treatment of idiopathic trigeminal neuralgia using different operative techniques (a cooperative study). Adv Neurosurg 1975;3:320-7. 66. Bernard EJ, Nashold BS Jr, Caputti F. Clinical review of nucleus caudalis dorsal root entry zone lesions for facial pain. Appl Neurophysiol 1988;51:218-24. 67. Moffie D. Late results of bulbar trigeminal tractotomy: some remarks on recovery of sensibility. J Neurol Neurosurg Psychiatry 1971;34:270-4. 68. Grant FC, Groff RA, Lewy FH. Section of the descending spinal root of the fifth cranial nerve. Arch Neurol Psychiatr 1940;498-509. 69. Grant FC, Weinberger IM. Experiences with intramedullary tractotomy: relief of facial pain and summary of operative results. Arch Surg 1941;42:681-92. 70. Bernard EJ Jr, Nashold BS Jr, Caputti F, Mossy JJ. Nucleus caudalis DREZ lesions for facial pain. Br J Neurosurg 1987;1:81-92. 71. Crue BL, Todd EM, Carregal EJA, Kilham O. Percutaneous trigeminal tractotomy. Bull LA Neurol Soc 1967;32;86-92. 72. Schvarcz JR. Tractolisis trigeminal estereotaxica. Bol Asoc Arg Neurocir 1972;13:31. 73. Teixeira MJ. A lesa˜o do trato de Lissauer e do como posterior da substaˆncia cinzenta da medula espinal e a estimulac¸a˜o ele´trica do sistema nervoso central para o tratamento da dor por desaferentac¸a¨o. Ph.D. Thesis. Faculdade de Medicina of University of Sao Paulo; 1990. 74. Todd EM, Crue BL, Carregal EJA. Posterior percutaneous tractotomy and cordotomy. Confin Neurol 1969;31: 106-15. 75. Hitchcock ER, Schvarcz JR. Stereotaxic trigeminal tractotomy for post-herpetic facial pain. J Neurosurg 1972;37:412-17. 76. Crue BL, Carre gal EJA, Felsoory A. Percutaneous stereotactic radiofrequency trigeminal tractotomy with
Technique of trigeminal nucleotractotomy
77.
78.
79.
80.
81. 82.
83.
84.
85. 86.
87.
88.
89.
90. 91. 92.
93.
neurophysiological recordings. Clin Neurol 1972;34: 389-97. Crue BL, Lasby V, Kenton B, Felsoory A. Needle scope to stereotactic frame for inspection of cysterna magna during percutaneous radiofrequency trigeminal tractotomy. Appl Neurophysiol 1976/77;139:58-64. Sampson JH, Nashold BS. Facial pain due to vascular lesions of the brain stem relieved by dorsal root entry zone lesions in the nucleus caudalis: report of two cases. J Neurosurg 1992;77:473-5. Hitchcock ER, Tsukamoto Y. Physiological correlates in stereotactic spinal surgery. Acta Neurochir Suppl (Wien) 1974;21:119-23. Westrum LE. Changes in the synapses of the spinal trigeminal nucleus after ipsilateral rhizotomy. Brain Res 1968;11:706-9. Black R. Trigeminal pain. In: Crue BL, editor. Pain and suffering. Springfield, IL: Thomas; 1970. p. 119-37. Schvarcz JR. Craniofacial postherpetic neuralgia managed by stereotactic spinal trigeminal nucleotomy. Acta Neurochir Suppl (Wien) 1989;46:62-4. Schvarcz JR. Stereotactic trigeminal nucleotomy for dysesthesic facial pain. Adv Pain Res Ther 1979;3: 331-3. Ishijima B, Shimoji K, Shimizu H, et al. Lesions of spinal and trigeminal dorsal root entry zone for deafferentation pain. Appl NeurophysioI 1988;51:175-87. Hitchcock E. Stereotactic spinal surgery. Neurol Surg 1977;433:271-9. Schvarcz JR. Post-herpetic craniofacial dysesthesiae: their management by stereotactic trigeminal nucleotomy. Acta Neurochir (Wien) 1977;38:65-72. Hitchcock E. Electrophysiological exploration of the cervicomedullary region. In Proceedings of a Symposium held in Paris at the Faculte des Sciences, International Congress Series 253. Amsterdam: Excerpta Medica; 1971. p. 237-45. Teixeira MJ, Lepski G, Aguiar PH, Cescato VA, Rogano L, Alaminos AB. Bulbar trigeminal stereotactic nucleotractotomy for treatment of facial pain. Stereotact Funct Neurosurg 2003;81(1-4):37-42. Iskandar B, Nashold B. Spinal and trigeminal DREZ lesions. In: Gildenberg PL, Tasker RR, editors. Textbook of stereotactic and functional neurosurgery. New York: McGraw-Hill Co; 1998. p. 1573-83. Fox JL. Percutaneous trigeminal tractotomy for facial pain. Acta Neurochir (Wien) 1973;29:83-8. Kanpolat Y, Deda H, Akyar S, et al. CT-guided trigeminal tractotomy. Acta Neurochirurg (Wien) 1989;100: 112-14. Kanpolat Y, Cosman ER. Special radiofrequency electrode system for computed tomography-guided pain-relieving procedures. Neurosurgery 1996;38:600-3. Kanpolat y, Akyar S, Caglar S. Diametral measurements of the upper spinal cord for stereotactic pain procedures:
124
experimental and clinical study. Surg Neurol 1995;34:478-83. 94. Kanpolat Y, Caglar S, Akyar S, Temiz C. CT-guided pain procedures for intractable pain in malignancy. Acta Neurochir Suppl (Wien) 1995;64:88-91. 95. Kanpolat Y, Savas A, Batay F, Sinav A. Computed tomography-guided trigeminal tractotomy-nucleotomy in the management of vagoglossopharyngeal and geniculate neuralgias. Neurosurgery 1998;484-9. 96. Kanpolat Y, Savas A, Akyar S, Cosman E. Percutaneous computed tomography-guided spinal destructive procedures for pain control. Neurosurg Q 2004;14 (4):229-38. 97. Spiegelmann R, Friedman WA, Ballinger WE, Tedeschi H. Anatomic examination of a case of open trigeminal nucleotomy (nucleus caudalis dorsal root entry zone lesions) for facial pain. Stereotact Funct Neurosurg. 1991;56(3):166-78. 98. Nashold BS Jr, Caputi F, Bernard E. Trigeminal DREZ: Caudalis nuclear lesions for relief of facial pain. Neurosurgery 1984;19:150. 99. Abdennebi B, Bouatta F, Bougatene B. Nucle´otomie du noyau spinal do trijumeau: a propos de deux ne´vralgies trige´minales post-traumatiques ope´re´es. Neurochirurgie 1993;39:231-4. 100. Rossitch E Jr, Zeidman SM, Nashold BS Jr. Nucleus caudalis D.R.E.Z. for facial pain due to cancer. Br J Neurosurg 1989;3:45-9. 101. Young IN, Nashold BS, Cosman ER. A new insulated caudalis nucleus D.R.E.Z. electrode: technical note. J Neurosurg 1989;70:283-4. 102. Gorecki JP, Nashold BS. The Duke experience with the nucleus caudalis DREZ operation. Acta Neurochir Suppl (Wien) 1995;64:128-31. 103. Morita M, Hosobuchi Y. Desending trigeminal tractotomy for trigeminal neuralgia after surgical failure. Steretact Funct Neurosurg 1992;59:52-55 (DOI: 10. 1159/000098917). 104. Spiegelmann R, Friedman WA, Ballinger WE, Tedeshi H. Anatomic examination of a case of open trigeminal nucleotomy (nucleus caudalis dorsal root entry zone lesions) for facial pain. Stereotact Funct Neurosurg 1991;56:166-78. 105. Nashold BS Jr, EI-Naggar AO, Ovelman-Levitt J, Muwaffak A. A new design of radiofrequency lesion electrodes for use in caudalis DREZ operations. J Neurol 1994;80:1116-20. 106. Nashold BS Jr, El-Naggar AO, Ovelmen-Levitt J, et al. A new design or radiofrequency lesion electrodes for use in the caudalis nucleus DREZ operation. Technical note. J Neurosurg 1994;80:1116-20. 107. Young RF, Oleson M, Perryman KM. Effect of trigeminal tractotomy on behavioural response to dental pulp stimulation in the monkey. J Neurosurg 1981;55:420-30.
2123
123 The Central Lateral Thalamotomy for Neuropathic Pain D. Jeanmonod . A. Morel
Introduction Head and Holmes postulated in 1911 the existence of an ‘‘essential medial thalamic centre,’’ localized medial to a pain-generating lesion in the thalamic ventroposterior (VP) nucleus, and responsible for the pathogenesis of central pain [1]. This centre was thought to be exposed to a decreased inhibitory influence from thalamo-cortico-thalamic loops. A generation of abnormal impulses in VP and their amplification in a reverberating circuit between lateral and medial thalamic nuclei were also proposed in the seventies by Sano [2]. Furthermore, the medial thalamus has been known for years to be an amplifier/synchronizer for low electroencephalographic (EEG) frequencies [3]. From the beginning of stereotaxy in the fifties and in contrast to all other lesional surgeries, medial thalamotomies against neuropathic (synonym: neurogenic) pain were recognized as procedures with low complication rates and absence of risk for the development of iatrogenic pain manifestations. They were shown to bring pain relief to all body localizations, and that without producing somatosensory deficits. Although cases with total and stable pain relief were published, recurrence of the original pain, partial or complete, was frequent [2,4–10]. These observations were commonly reported, but many studies were relatively small and included inhomogeneous pain patient populations. These data provided us with the necessary basis and incentive to pursue the medial thalamic #
Springer-Verlag Berlin/Heidelberg 2009
path, with the goal to re-actualize this promising therapeutic option on the basis of newly developed anatomical, physiological and technical tools. Other reports of our experience in this field have been published elsewhere [11–16].
Anatomical Basis The role of the medial thalamus in pain, in particular the intralaminar nuclei, has long been recognized and related to motivational-affective aspects through its afferent connections with the spinothalamic (STT) and spino-reticulothalamic (SRTT) tracts, and efferent projections to pain-related areas in associative and paralimbic cortical domains. This so-called ‘‘medial pain system’’ has been in the past the target for surgical interventions in patients with chronic, therapy-resistant neuropathic pain. These targets were mainly located in the caudal intralaminar nuclei (Centre Me´dian/Parafascicular complex [CM/Pf]), central lateral nucleus (CL), posterior complex (POC) and in the medial pulvinar (PuM) [2,6,8,9,17]. The present account on the anatomy of the posterior part of the CL (CLp) as a surgical target for neuropathic pain is based on recent multiarchitectonic studies and integrates the nucleus in a large thalamocortical (TC) network responsible for the multiple sensory, cognitive and affective components of the neuropathic pain condition.
2082
123
The central lateral thalamotomy for neuropathic pain
Characteristics of CLp The CL has been considered as part of the anterior group of the intralaminar nuclei, but in non-human primates and in humans, there is a posterior extension of CL around the posterior pole of the mediodorsal (MD) nucleus, often included by others in the densocellular part of MD (MDdc). This region is characterized by populations of relatively large, darkly stained cells aggregated into clusters or islands within a ‘‘matrix’’ of smaller cells comparable to those of PuM and the parvocellular part of MD (MDpc). These darkly stained cells are in continuity with similar neurons in the limitans nucleus (Li) as well as in the adjoining part of the medullary lamina that corresponds to the CL proper identified by others, particularly in non-primate species (cats, rodents; [18,19]. The clusters in the human posterior CL were initially described by Hirai and Jones [20] as ‘‘. . . densely stained clumps coextensive with the clumps of acetylcholinesterase staining. . .’’ and subsequently characterized by a rich innervation of enkephalin and substance P fibers [21]. Using additional markers for a multiarchitectonic parcellation of the human thalamus, we demonstrated that these same clusters of CLp cells express high levels of immunoreactivity to the
calcium-binding proteins calbindin (CB) and calretinin (CR), in co-registration with those displaying dense acetylcholinesterase (AChE) staining [22,23]. These clusters are illustrated in > Figures 123‐1 and > 123‐2 in horizontal, frontal and sagittal sections of the human thalamus, in relation to postoperative magnetic resonance images (MRI) of the lesion localizations after central lateral thalamotomy (CLT) (see also ‘‘Stereotactic Technique’’ below). The dense cholinergic innervation of CL, as well as known inputs from the reticular formation (through SRTT), may be associated with its role in sleep/wakefulness and in attentional processes. The similarity in terms of cyto- and immunohistochemical staining characteristics between CLp clusters and neurons in the posteroventrally adjacent Li nucleus [24–26] indicate that they form part of the same anatomo-functional group. It is important to note that there is a continuity in terms of CB immunostaining between CL and Li and more lateral thalamic nuclei (posterior nucleus [Po], parvocellular division of the ventral posterior medial nucleus [VPMpc], ventral posterior inferior nucleus [VPI]) which are involved in pain processing [27–31]. The area identified as CLp and characterized by CB-/CR-immunoreactive clusters intermingled within a ‘‘matrix’’ of MD- and PuM-like lightly
. Figure 123‐1 Projection of atlas maps on to postoperative (5 years) axial (a) and frontal (c) 3D-IR MRI to show the localization of a CLT lesion in a patient with chronic trigeminal neuropathic pain. The blur of the MR image in (c) is due to a 3D reconstruction of the same lesion shown in (a). The lesion is well delimited by a black area. In (b) and (d), projections of the CLT lesion (grey outlines) onto corresponding levels of axial (5.4 mm dorsal to intercommissural level) and frontal (2 mm anterior to pc level) sections of the human thalamus stained for Nissl (b) and immunostained for CB (d). Clusters in the CLp are seen in both Nissl and CB stained sections. Dotted lines indicate pc level in (b) and intercommissural plane in (d). AM anteromedial nucleus; Cd caudate nucleus; CM centre me´dian; CL central lateral nucleus; CLp posterior part of CL; CLT central lateral thalamotomy; CP cerebral peduncle; GPe globus pallidus, external segment; Hb habenular nucleus; ic internal capsule; LD lateral dorsal nucleus; LGN lateral geniculate nucleus; Li limitans nucleus; LP lateral posterior nucleus; MD (pc, mc, pl), mediodorsal nucleus (parvocellular, magnocellular and paralamellar divisions); MGN medial geniculate nucleus; mtt mammillothalamic tract; Pf parafascicular nucleus; Po posterior nucleus; PuA anterior pulvinar; PuL lateral pulvinar; PuM medial pulvinar; PuT putamen; Pv paraventricular nuclei; R reticular thalamic nucleus; SNc substantia nigra, pars compacta; sm stria medullaris; VA (pc, mc) ventral anterior nucleus (parvocellular and magnocellular divisions); VLa ventral lateral anterior nucleus; VLp (pl, v), ventral lateral posterior nucleus (paralaminar and ventral divisions); VPL (p, a), ventral posterior lateral nucleus (posterior and anterior divisions); VPI ventral posterior inferior nucleus. Scales bars: 4 mm
The central lateral thalamotomy for neuropathic pain
. Figure 123‐1 (Continued)
123
2083
2084
123
The central lateral thalamotomy for neuropathic pain
. Figure 123‐2 Two-days postoperative sagittal T1W-MRI (a) and illustration of the AChE (b) and CR-immunostained (c) sections at similar M-L level (7 mm from midline) of the human thalamus. The outlines of the CL nucleus are shown in dotted line. The CLT lesion (grey outline), not including the surrounding edema, is projected onto AChE section in (b). The crosses in (a) and (b) indicate the position of the pc. The clusters in the CLp are clearly seen in both CR and AChE stained sections. VPMpc, ventral posterior medial nucleus, parvocellular division. For other abbreviations, see legend > Figures 123‐1. Scale bars: 4 mm in (a) and 2 mm in (b) and (c)
parvalbumin (PV) immunoreactive cells varies in size among subjects, in particular in dorsal extension [23]. In average, it extends 3–4 mm in antero-posterior (A-P), 5–6 mm in medio-lateral (M-L) and 11 to 13 mm in dorso-ventral (D-V) directions (see > Figures 123‐1 and > 123‐2). The targeting of CLp in the context of surgical CLT is discussed below.
Connectivity The CL nucleus as a whole is multimodal by its diverse sensory, motor and paralimbic connections, as also demonstrated by the diversity of responses evoked by stimulations [2]. The intralaminar nuclei, in particular CL, were long considered as mainly connected to the basal ganglia
The central lateral thalamotomy for neuropathic pain
(thalamostriatal system). However, particularly in non-human primates, the nucleus has important cortical projections that extend to large cortical domains through layer I projections [18,32]. In the context of pain, the CL is in a position
123
to transfer nociceptive informations, conveyed through the STT and SRTT, to relatively large domains of cortex including areas involved in nociception (primarily SII, insula, anterior cingulate cortex [ACC]) (> Figures 123‐3).
. Figure 123‐3 Upper part: schematic illustration of the differential extension of TC projections of the two major groups (medial in blue and lateral in orange) of pain-related thalamic nuclei in relation to their afferents from medial (m-STT) and lateral (l-STT) spinothalamic tracts. The large cortical domain innervated by CLp/Li comprises that dominantly innervated by TC from posterior and lateral thalamic nuclei (POC and VP complex) and corresponding to the ‘‘pain matrix.’’ Although we have grouped POC and VP together, there is evidence for a functional role of POC as intermediate between specific and non-specific. Lower part: frontal thalamic map illustrating the localization of the two major groups of pain-related thalamic nuclei (only Po of the POC is seen at this level). We have not considered PuA because of only scarce evidence for its role in pain processing. ml/stt, medial lemniscus/spinothalamic tracts. For other abbreviations, see legend > Figures 123‐1
2085
2086
123
The central lateral thalamotomy for neuropathic pain
Spinothalamic afferents to CL arise dominantly from deep dorsal horn laminae (V-VII) where large, often bilateral receptive fields are recorded (see [33,34] for reviews). However, lamina I STT afferents to the medial thalamus were also described [34,35], and some were shown to end in the ventrocaudal part of the MD nucleus (MDvc) in monkey [28]. This subregion of MD described as containing ‘‘. . . more darkly stained neurons of greater density and more varied size and shape than those in either MDpc or MDdc’’ seems to correspond closely to the area enclosing the ventral part of CLp and dorsal part of Li in the human thalamus (see > Figures 123‐1 and > 123‐2). It is important to note that a similar region was already described in the early sixties as receiving spinothalamic afferents in the human thalamus [30]. Other connections of CL are with the cerebellum, with dentato-thalamic afferents reaching the CLp in addition to more anterior part of CL [36–39]. Interestingly, among cerebellothalamic projections to CL, some appear to relay dentate afferents to the prefrontal cortex and are presumably involved in cognitive/mnestic processes [40]. Other afferents are from brainstem cholinergic, noradrenergic and serotoninergic groups, supporting the role of CL in general arousal. One important aspect of CL in the context of the pathophysiology of neuropathic pain is its profuse intrathalamic connections with the reticular nucleus (see paragraph below ‘‘Physiological Basis’’). The CLp, proposed as comprising part (or whole) of the MDdc and anterior part of the PuM, projects to a large part of the cortical mantle, from sensory to associative (prefrontal cortex [PFC], posterior parietal cortex [PPC]), premotor and paralimbic (insula, ACC) areas (see > Figure 123‐3). We would like to propose that (1) the CB-immunoreactive cells in the CLp islands represent, as in the rest of CL and other thalamic nuclei [41], a diffuse layer I projection, and that (2) the lightly PV-immunoreactive PuM- and MD-like cells project to layers III
and IV of a large, but yet more restricted domain in associative/paralimbic cortex (e.g., PFC, ACC, insula). The possibility to combine these two long-range TC systems within CLp may be considered as relevant in the context of emotional and cognitive aspects of pain, particularly in chronic neuropathic pain conditions (see below). Neurometabolic (fMRI, PET) and electrophysiological (EEG, MEG) studies indicate that acute nociceptive and chronic neuropathic pain conditions involve the ‘‘pain matrix’’ as well as large parts of the paralimbic/associative domain. However the distributions (within and between hemispheres) and degrees of activation or deactivation of cortical and thalamic areas may differ depending on the pain type [42–46] and study conditions. Interestingly, the decrease of regional cerebral blood flow observed by Hsieh et al.[44] in the thalamus in neuropathic pain (presumably associated with low-frequency production, see below) is located in its postero-dorsal part corresponding closely to the dorsal half of CLp.
CLp as a Surgical Target The advantages of the CLP target are following: 1.
2.
3.
There are known afferents from the STT, as also shown in the human brain [30]. No such convincing evidence exists for spinothalamic afferents to CM/Pf or to PuM, which were both targeted in the past [2,6,8,9,34,47]. There is a combination of ‘‘diffuse’’ (layer I) and ‘‘non-diffuse’’ (layers III-IV) TC projections to large cortical domains, including areas mediating discriminative (SI, SII), affective-motivational (ACC, insula), cognitive (PFC) and motor (premotor cortex) aspects of pain. This is not the case for the other medial thalamic targets. In comparison to CM thalamotomies, the CLp target is distant from primary
The central lateral thalamotomy for neuropathic pain
4.
somatosensory nuclei, in particular the VPM nucleus. Indeed, the frequent occurrence of sensory deficits provoked by CM lesions led Richardson to move the target gradually more posteriorly toward PuM, i.e., presumably close to CLp, with similar clinical results [6,8]. The interindividual variability of CLp is relatively low [23] as compared to PuM for example, in particular in the A-P axis. This small variation can be easily taken into account by MRI visualization of neighbouring landmark structures such as the posterior commissure, the habenula and the stria medullaris (see > Figure 123‐1 and the paragraph ‘‘Stereotactic Technique’’ below). For M-L and D-V variations, the visualization in MRI of the internal capsule and of the dorsal limit of the thalamus, respectively, allow the placement of CLT lesions so as to include most of the CLp. Because of the orientation of the penetration, some encroachment of the lesion onto adjacent structures, MD and PuM, is inevitable. Intrusion posteriorly into PuM has been unproblematic in our experience, and targets in PuM were shown to relieve pain, although not in the long-term [6]. The involvement of the posterior part of MD (e.g., > Figure 123‐2) never caused unwanted clinical effects. Considering the continuity of Li with the CLp in terms of cytoand immunohistochemical characteristics [21,22,25] and its projections to cortical areas involved in pain, this nucleus may participate in the beneficial effect of CLT. However, as discussed below, its closeness to the pretectum prevents its extensive lesioning.
Physiological Basis Using intra-operative microelectrode recordings to optimize medial thalamic targeting, we
123
observed a concentration of unit bursting activities in the posterior part of the medial thalamus which were identified as low threshold calcium spike (LTS) bursts [13]. These bursting activities have been described for the first time experimentally by Llinas and coworkers [48–50]. To our knowledge, the first evidence for the presence of LTS bursts in the human thalamus was presented by Modesti and Waszak in 1975 [51] at a time where they could not be identified, because their existence had not yet been claimed experimentally. The first identification of LTS bursts in the human thalamus (VP complex) was provided by Lenz and collaborators [52]. They found them confined in and around the portion of VP representing the deafferented and painful body part [53]. In contrast, in the medial thalamus, we found LTS bursts spread diffusely in and around the posterior part of the central lateral nucleus (CLp) [14]. This difference of distribution is in accordance with the well-known identification of two types of TC interactions, specific (addressing the ‘‘content’’) and non-specific (addressing the ‘‘context’’) [41,54–57]. The presence of LTS bursts in the same area (CLp) in patients suffering from neuropathic pain, tinnitus, uni- and multifocal epilepsy, movement and neuropsychiatric disorders allowed us to propose a common pathophysiological mechanism for these different symptomatologies [14]. The following observations were collected in CLp: (1) half of the recorded neurones displayed LTS bursting activity, (2) only a minority (<1%) responded to sensory or motor stimulations, (3) LTS bursts displayed an average interburst frequency of 4 Hz, and (4) there were no significant differences between recordings in patients suffering from peripheral and central neuropathic pain. Considering that thalamic and cortical partners are coupled in tight functional units, the so-called TC modules [54,55], the next step was to search for delta/theta activity increase at the cortical level. This was investigated first using magnetoencephalography (MEG), which
2087
2088
123
The central lateral thalamotomy for neuropathic pain
confirmed an increase in low frequency production [55,58]. A high frequency increase was also observed, and a conceptual framework named thalamocortical dysrhythmia (TCD) was proposed. Correlations between surface EEG and thalamic local field potentials demonstrated an increased production of theta rhythms in CLp and on the EEG, with high coherence between thalamus and cortex [59,60]. Gu¨cer and coworkers [61] had already shown a low frequency power increase in neuropathic pain patients in both surface and thalamic EEG recordings. Two detailed EEG studies [43,62] have confirmed a highly significant EEG spectral power increase in the delta, theta and beta domains and suggested an enhanced coupling between theta and beta frequencies. An increased interfrequency coupling was also found using a phase-independent measure of frequency correlation [55]. MEG [58] and EEG [43] studies have brought evidence for localization of the spectral frequency increases within the pain matrix, including the insular, cingulate and somatosensory areas. The sequence of events underlying the TCD process at the source of neuropathic pain may thus be proposed as follows: 1.
2.
A lesion leads to deafferentation of excitatory inputs on thalamic relay cells and initiates the neuropathic pain process. The lesion may be peripheral (nerves and roots) or central (spinothalamic tract and brainstem) leading to bottom-up deafferentation. A thalamic VP lesion induces deafferentation of neighbor TC modules, and a cortical or subcortical white matter lesion causes a top-down thalamic deafferentation. Deafferentation of excitatory inputs results in disfacilitation and thalamic cell membrane hyperpolarization. In the hyperpolarized state, deinactivation of calcium T- channels causes thalamic relay neurones to fire LTS bursts at 4 Hz. A tight functional TC coupling is sustained by thalamocortical and thalamo reticular as
3.
4.
well as by recurrent corticothalamic, reticulothalamic and cortico-reticulo-thalamic projections. This causes the affected TC modules to discharge at 4 Hz, and this situation is reinforced by the tendency of the TC network to maintain a given functional modality over time [63]. Divergent thalamocortical, corticothalamic and reticulothalamic projections provide the anatomical substrate for coherent diffusion of low frequency EEG activity to an increasing number of neighboring TC modules (theta cross-modular spread). This theta spread may extend variably over time and thus explain the often observed delay between the occurrence of the causal insult and the beginning of pain. Increased low frequency oscillations occur also during sleep [56] and cognitive activation [64,65], where they are part of normal processes. It is the continuous, widespread and stateindependent overproduction of slow rhythms in the awake brain that characterizes TCD. In slow wave sleep, widespread low frequency EEG activity correlates with absence of TC modular function. Thus, the low frequency overproduction in the TCD sequence may lead to the often underrated appearance of functional somatosensory deficits in or around the pain area. The final step toward production of neuropathic pain is related to the reciprocal corticocortical inhibition mediated by GABAergic interneurons, a general feature of cortical organization. The cortical pole of TC modules in the theta mode exerts less GABAergic cortico-cortical collateral inhibition on neighbouring high frequency cortical areas, which are thereby released and overshoot. This event has been termed edge effect [55]. Support for such an effect is given by (1) the spectral power increases in both low and high frequency domains, (2) the increased interfrequency coupling between theta and
The central lateral thalamotomy for neuropathic pain
beta activities measured by EEG or MEG, and (3) the colocalization in EEG LORETA source localization analysis of both increased frequency domains. Recently, experimental evidence for the existence of the edge effect (spatio-temporal increase in beta/gamma activation due to the presence of a conjoined low frequency activation) was provided in an in vitro preparation [66]. Thus, an increase of high frequency activity in cortical areas of the pain matrix can give rise to the pain symptom, explaining the paradox of a deactivation of the thalamus leading to a cortical activation.
Rationale for the Selective Regulatory TCD-Based Central Lateral Thalamotomy On the basis of clinical, physiological and anatomical data, a rationale for a selective regulatory TCD-based medial thalamic intervention is proposed. The goal, in the context of intracerebral interventions, is to rebalance or regulate, without reducing, the TC system. As a complex system, the TC pain network, like other TC networks, possesses specific, or content, topologically organized TC modules integrating the pain function, and non-specific, or context, more diffusely organized TC modules. The latter, especially the ones comprising CLp, have long-ranging thalamocortical, reticulothalamic and corticothalamic projections. They fulfill a regulatory role, interconnecting the different systems with each other. In neuropathic pain situations, the CLp reveals itself as a dysfunctional regulator, because (1) it sustains/amplifies a deleterious low frequency overproduction, and (2) it has lost over time its normal function (<1% receptive fields and absence of deficits after CLT). The non-topological organization of CLp explains the efficiency of lesions within it
123
against pain on any body part. The invasive indication for CLT is justified by the chronicity of the symptomatology (>1 year) and therapyresistance. These criteria strongly suggest that CLp does not have any relevant chance to regain a normal function. Given the extensive crossmodal TC plasticity (on-going study), other medial regulatory areas can take over the functions lost by CLp, probably long before the operation. We have collected the evidence, in a cancer patient who had to be operated upon less than 1 year after the appearance of pain, that the TCD dynamics can already be present 3 months after the causal insult. By selectively targeting a regulatory area having lost its normal function, the CLT leaves intact all other functional parts of the TC network, which is relevant for two reasons: (1) no postoperative reduction of hemispheric functions, including nociception and cognition (ongoing study), is to be experienced by the patient, and (2) there is no risk for rekindling of TCD components, as must be expected after any surgical additional deafferentation. Indeed, lesioning in any part of the specific pain system is known to produce deficits and to bring a high risk of ‘‘iatrogenic pain.’’ In consideration of all these elements, we advocate the regulatory approach provided by CLT. We have preliminary, not yet conclusive, evidence for a regulatory role of the CM/Pf complex. Our experience during initial explorations in different medial thalamic nuclei does not point to other candidates. The results obtained years ago in the medial thalamus by a few neurosurgical groups tend to support the primacy and possible exclusivity of CLp as regulatory medial thalamic target: (1) Sano [2], as an exception in his time, concentrated his efforts on the posterior part of the medial thalamus using a posterior approach, coming thus closer than anyone to CLp, which was most of the time not reached and explored by others, and (2) Hitchcock and Teixeira [9] as well as Young and col. [67] placed
2089
2090
123
The central lateral thalamotomy for neuropathic pain
relatively large lesions in the posterior part of CM/Pf, probably involving CLp. Our experience with a small group of cancer patients indicates a limited, if any, control of nociceptive pain by CLT.
Stereotactic Technique The CLT stereotactic intervention takes place under local anesthesia. A stereotactic frame (Cosman Robert Wells, Radionics, Burlington Massachusetts) compatible with magnetic resonance (MR) is used. Stereotactic MR images are taken to localize the target. A midsagittal section, on which the centers of the ac and pc are determined, allows the production of a stack of axial sections parallel to the intercommissural plane. The electrode reaches the computer-calculated target in the base of CLp through a prefrontal approach, under impedance monitoring. The coordinates of the CLT target are: (1) D-V: the intercommissural plane, (2) A-P: 2 mm posterior to the pc, and (3) M-L: 6 mm lateral to the thalamo-ventricular border. A safety measure for the pretectum can be introduced by the placement of the therapeutic lesion 1–2 mm above the intercommissural plane, which in our experience does not influence significantly the degree of the postoperative pain relief. The M-L angle of penetration is between 5 and 10 , the A-P angle between 60 and 65 , with the goal to fit at best the 3D shape of CLp (see > Figure 123‐2). To avoid penetrating through the premotor area, the entry point is placed anterior to the coronary suture. This keeps the antero-posterior angle below 60 in some patients. In such situations, a posterior correction of the antero-posterior target coordinate of 1 mm is added. The 3D shape of CLp makes it difficult to cover its total extent by a therapeutic lesion, unless it is enlarged, but with a risk for more encroachment onto adjacent structures such as MD and PuM.
In most patients however, a partial lesioning of CLp is enough to provide the system with the necessary de-amplification of low frequency production, as seen clinically and with quantitative EEG. The penetration track to CLp goes inevitably through the ventricle, and thus through the ependymal layer. This may offer a variable and unpredictable resistance to the electrode. We calculated the average absolute error of our CLp targets as compared with desired coordinates, and found a value within one millimeter for the A-P and M-L dimensions, but 1.9 mm in the D-V direction, along the penetration [68]. Since then, we have used an electrode tip with the form of a blunt pencil, developed for us by Radionics, in order to reduce ependymal resistance. Unit recordings were obtained using tungsten microelectrodes (impedance: 0.5–1 MΩ; bandpass filter: 300 Hz-3 kHz). Unit responses are sought with a protocol including tests of motor voluntary function, as well as tactile, nociceptive and proprioceptive stimuli. The microelectrode recordings are useful to first localize the ventral border of CLp, and second to confirm the clinical diagnosis of neuropathic pain by the lack of responding units and the presence of LTS bursts. At the end of the recording session, the macroelectrode (blunt pencil, active tip 1.1 mm diameter over 2 mm length) replaces the microelectrode in the common guide tube, and a macrostimulation session is performed analysing somatosensory, motor and cognitive/emotional responses. The typical stimulation profile (ongoing study) is characterized by paresthetic or dysesthetic manifestations, which can extend on the whole body but are often localized on and around the affected body part(s). A radiofrequency thermolesion (4 mm diameter over 10–12 mm length) is placed. This therapeutic lesion may encroach on the borders of neighbouring nuclei, such as the limitans, PuM (anterior part) or MD nuclei (posterior part), without detectable negative consequences.
The central lateral thalamotomy for neuropathic pain
Clinical Results In our largest published series [16], the MR- and microelectrode-guided stereotactic CLT was offered to 96 patients (40 females, 56 males) suffering from chronic peripheral or central neuropathic pain. Resistance to more conservative therapy was documented for all patients with proper antiepileptic and antidepressant treatments. The age of the patients by the time of CLT ranged between 19 and 84 years (mean 5615 years). The dates of injury and pain onset were known in 62 patients, and 56% of them presented a delay for pain onset ranging from 2 weeks to 5 years (mean 1014 months). The duration of the pain syndrome, i.e., the time between pain onset and CLT, varied between 1 month and 43 years (mean 7.58 years). Somatosensory deficits were found in 91.6% of the patients. Sixty patients (62.5%) had previous unsuccessful therapies including transcutaneous electrical nerve stimulation (31 cases), dorsal column or thalamic stimulation (26 cases) and interruptive (neurotomy, rhizotomy, cordotomy, sympathectomy) surgery (26 cases). Decompression was used in 14 cases and morphine reservoirs implanted in two patients. A majority of patients (58%) had lesions of primary afferents (peripheral nerve and/or dorsal root lesions). Among them, five had postherpetic neuralgia and seven an amputation. Twenty one percent of the patients had a pure central lesion (spinal cord, brainstem, thalamus and cortex), and the same percentage had a mixed, peripheral plus central, damage (root and spinal cord lesion, mainly plexus avulsions). In five patients, the neuropathic pain syndrome resulted from cancer. They were operated upon before the minimum 1 year of pain duration required for patients with normal life expectancy. Thirteen out of all the patients had bilateral lesions of the spinal cord leading to bilateral pain. The pain localization covered the face (23.9%), the head and neck (7.3%), the
123
upper and lower limb (69.8%), the thoracoabdominal and the perineal areas (20.8%) or the entire hemibody (8.3%), with some patients having more than one pain localization. Pre- and postoperative assessments comprised determining: (1) pain types, (2) pain qualities, (3) pain intensity on a visual analogue scale (VAS, 0–100), (4) the percentage of the global postoperative pain relief estimated by the patients, and (5) drug intake. Three pain types were recognized: continuous (C), intermittent (I) and allodynic. The I pain type was subdivided into paroxystic (less than 2 min) and episodic (more than 2 min). C pain type alone was present in 27%, I pain type alone in 10%, and the two pain types together in 63% of the patients. Allodynia was present in 85% of patients. It was triggered by tactile stimuli in 13.4%, by proprioceptive stimuli in 41.5%, and by both stimuli in 45.1% of the patients. It was encountered with either the C pain type alone (28%), the I pain type alone (5%), or both (67%). The pain qualities reported by the patients were subdivided into thermal (burning, in 79.1% of the patients), exteroceptive (pins and needles, in 86.4%), proprioceptive (compressive, tearing, in 91.6%) and electrical discharges (in 54.1%). Pain qualities were rarely seen alone (4%), the most frequent association was a combination of 3 or 4, seen in 71% of the patients. Patients were asked to estimate their maximal pain intensity (VASmax). The mean preoperative VASmax for all patients was 8510. Preoperatively, 92 patients received antiepileptics, 47 opiates, 37 benzodiazepines, 34 thymoleptics and 19 neuroleptics. For the most often used drugs, the mean dose of carbamazepine (66 patients) was 868427 mg, and of clonazepam (25 patients) 2.352.66 mg. Clomipramin was taken by 36 patients with a mean dose of 7644 mg. The most frequent drug combination was an antiepileptic plus a thymoleptic, encountered in 50 patients.
2091
2092
123
The central lateral thalamotomy for neuropathic pain
Follow ups ranged from 2 weeks (cancer patient) to 10 years 5 months (mean 3 years 9 months 2 years 9 months). The overall mean pain relief estimated by the patients reached 45.7 39.6%. Fifty three percent of the patients estimated their pain relief superior to 50%. There was complete pain relief for 18.7% of the patients. The C pain type was suppressed in 30.2% of the patients by CLT, the I pain type in 54.2%. The mean relief obtained in the patient group suffering from the C pain type, alone or with the I type, amounted to 20.4 25.8% and was significantly lower than in the patient group with the I type only (66 39.2%). An analysis of the VASmax showed only a significant postoperative decrease of 59.2% in the patient group with an I pain type alone, and confirmed a higher resistance of the C type to CLT. The mean duration of pain episodes and paroxysms was decreased significantly by 90 and 65%, respectively, and their mean VASmax reduced by 73 and 63%, respectively. Allodynia was suppressed in 57.3% of patients, with 66.6% relief of tactile and 73.5% of proprioceptive allodynia. After CLT, 66.2 and 63.4% decreases of the number of patients presenting exteroceptive and electrical discharge qualities were noted. Thermal and proprioceptive qualities were more resistant to surgery, with 43.4 and 37.5% decreases, respectively. No significant correlation could be found between pain relief and pain duration, and no significant influence of previous surgical therapies on the outcome after CLT was demonstrated. There was only a trend for better pain relief in patients with peripheral lesions as compared with central or mixed ones. In 28 patients suffering from unilateral continuous pain, the addition of an ipsilateral CLT provided a significant further pain relief (increase of the mean pain relief from 13.9 to 48.7%). A postoperative suppression of drug intake was observed in 31.6% of the patients. A postoperative improvement of the somatosensory examination (reduction of hypoesthesia and/or hypoalgesia) has
been described in a smaller series in 47% of patients [69]. We have not reanalysed this factor more recently, but can confirm the occurrence of this phenomenon in at least a third of the patients. Complications occurred in ten patients in the first years of our experience. They comprised one intraventricular bleeding resorbed without consequences in a few days, two cases of thalamic edema with short-lasting neurological manifestations, two thalamic bleedings with mainly short-term reversible pretectal and somatosensory deficits. In five other patients, partial and partly reversible pretectal deficits appeared, none of them causing a long-term disturbance. We attribute the fact that we have no longer experienced such complications over the last years to the following experiencebased developments: (1) the use of the blunt tip electrode mentioned above, to avoid problems due the ependymal passage, and (2) the final CLT coordinates [16], centered on the localization of the LTS bursting activity in CLp [14] and avoiding unnecessary more ventral (pretectum), lateral (lateral spinothalamic tract) or anterior medial thalamic penetrations. Two recent quantitative EEG studies [43,62] have shown that (1) the EEG power spectral profile evolves after CLT toward that of the control group, thus confirming the regulatory effect of CLTon EEG generators, (2) this normalization process takes time, at least a year, in accordance with progressive clinical improvements along time, and (3) the postoperative EEG power spectral reductions occurring in cortical areas of the pain matrix demonstrate the specificity of the CLT effect. A current study confirms these data on a larger group and adds the observation of a post-CLT improvement of a working memory task with normalization of different parameters of the EEG. Over the last years the fundamental influence of the cognitive/emotional factor in the genesis of chronic pain has been well recognized. Accumulating evidence underscores the fact that
The central lateral thalamotomy for neuropathic pain
mnestic [64,65] and emotional [70] activations increase hemispheric theta activity. Many cortical areas that are ascribed a function in pain processing, for example insular and cingulate areas, are integrated in the cognitive/emotional association and mesocortical, or paralimbic, domain. These two pieces of evidence provide an anatomofunctional substrate for a primary role of mental functions in the reactive modulation of neuropathic pain and the generation of psychogenic, or somatoform, pain. Related to this context, our experience with CLT, in principle similar to other therapeutic interventions, has shown various results after the same treatment: some patients do not get the expected relief that others do. The large extent and connectivity of the human association/paralimbic domain may explain these variations, leading to the conclusion that the mind of the patient is stronger than the therapeutic action, and may jeopardize its effect. A regulatory intervention like CLT may thus reduce the disease-related TCD but not a contraproductive mental dynamics. This one may then entertain either an anxio-depressive suffering and/or a psychogenic pain state, characterized typically, in our experience, by a deep, compressive and continuous pain quality, pain extension beyond the primarily affected body area, clinical inconsistencies such as repeated touching of a body part described as allodynic, pain intensity rating at or above 100/100 VAS, lack of behavioral evidence of suffering, undescribable or atypical pain qualities, high demands on the treating team, as well as negation of an emotional modulation of pain or negation of anxio-depressive elements. Such situations require early and intensive psychotherapeutic support. In our experience, the most resistant evolutions of this type correlate with frustration and non-acceptance of the disease-related health impairment. We have realized in our department the first steps toward an integrated (surgical and psychotherapeutic) support, and observe a global improvement (between 60 and 70% of pain relief) of the
123
patients, which we hope will manifest itself in our future studies. The conceptual and practical consequences of these considerations have a wide range of implications for the therapeutic support of neuropathic and psychogenic pain patients. These include the recognition of (1) a dual origin of chronic neuropathic and psychogenic pain (in body and mind respectively), and (2) a common mechanism for both, i.e., the TCD. Surgery and psychotherapy each address the problem selectively, on the neuropathic and psychogenic sides, respectively. They should be integrated in a therapeutic continuum, and combined whenever necessary. When deciding the most adequate treatment against chronic neuropathic pain, each patient situation must be evaluated for the primary (first in time) and dominant (first in intensity at treatment time), neuropathic or psychogenic, dimension. We have recently published a detailed case report of a patient suffering from both pain types and treated with success only after the application of both therapeutic options [71]. A quantitative EEG analysis showed before CLT a TCD anchored maximally in insular and cingulate areas, remaining diffusely distributed over the whole associative/paralimbic areas after CLT, and suppressed after 8 months of psychotherapy.
Conclusion CLT is a reactualization of one of the first stereotactic operations ever performed. A large and long-term clinical experience and detailed pathophysiological and anatomical data allow us to propose the following: 1.
2.
The precise CLT target position in CLp is based on both refined anatomical (see ‘‘characteristics of CLp’’) and pathophysiological (TCD) data. Its mode of action can be seen as selective regulatory, thanks to the targeting of a dysfunctional neuronal group, which has lost its
2093
2094
123 3.
4.
5.
6.
7.
The central lateral thalamotomy for neuropathic pain
normal regulatory function long before surgery and which distorts the thalamocortical dynamics. These characteristics of CLT explain both its efficiency and sparing quality. Because CLp is part of the diffuse, or context, network, which covers the whole body representation, CLT can be used against any pain localization. Because CLT does not touch functional executors of the pain network (and even not functional regulators in other areas of the medial thalamus), it does not bring the risk of activation of new iatrogenic TCD elements, so rightfully feared after any reduction of the somatosensory function. Our long-term results on a large, chronic and therapy-resistant neuropathic pain patient group demonstrate, to the contrary of the older literature, a stability of postoperative pain relief for more than 50% of the patients. This makes CLT a valuable modern surgical option against neuropathic pain. We provide a CLT target definition offering a low complication rate, in spite of the necessary passage of the ependymal layer. Like any other therapeutic option, its limitation in efficiency can and must be understood and integrated in the context of a powerful cognitive/emotional human dimension, which might amplify or jeopardize the therapeutic purpose of the operation.
References 1. Head H, Holmes G. Sensory disturbances from cerebral lesions. Brain 1911;34:102-254. 2. Sano K. Intralaminar thalamotomy (thalamolaminectomy) and postero-medial hypothalamotomy in the treatment of intractable pain. Prog Neurol Surg 1977;8:50-103. 3. Morison R, Dempsey E. A study of thalamo-cortical relations. Am J Physiol 1942;135:281-92. 4. He´caen H, Talairach J, David M, Dell M. Coagulations limite´es du thalamus dans les algies du syndrome thalamique. Rev Neurol 1949;81:917-31.
5. Fairman D. Unilateral thalamic tractotomy for the relief of bilateral pain in malignant tumors. Confin Neurol 1967;29(2):146-52. 6. Richardson DE. Thalamotomy for intractable pain. Confin Neurol 1967;29(2):139-45. 7. Laitinen LV. Mesencephalotomy and thalamotomy for chronic pain. In: Lunsford LD, editor. Modern stereotactic Neurosurgery. Boston, Martinus Nighoff; 1988. p. 269-76. 8. Gybels J, Sweet W. Neurosurgical treatment of persistent pain. Basel: Karger; 1989. 9. Hitchcock ER, Teixeira MJ. A comparison of results from center-median and basal thalamotomies for pain. Surg Neurol 1981;15(5):341-51. 10. Tasker R. Stereotactic surgery. In: Wall P, Melzack R, editors. Textbook of pain, 2nd ed. Edinburgh: Churchill Livingstone; 1989. p. 840-55. 11. Jeanmonod D, Sarnthein J, Magnin M, Stern F, Lebzeltes C, Aufenberg C, Morel A. Chronic neurogenic pain: thalamocortical dysrhythmic mechanisms and their surgical treatment. Thalamus Relat Syst 2005;3:63-70. 12. Jeanmonod D, Sarnthein J, Stern J, Magnin M, Aufenberg C, Morel A. Thalamotomy for human pain relief. In: Schmidt R, Willis WD, editors. Encyclopedia of pain. Berlin: Springer Verlag; 2007. p. 2432-5. 13. Jeanmonod D, Magnin M, Morel A. Thalamus and neurogenic pain: physiological, anatomical and clinical data. Neuroreport 1993;4(5):475-8. 14. Jeanmonod D, Magnin M, Morel A. Low-threshold calcium spike bursts in the human thalamus. Common physiopathology for sensory, motor and limbic positive symptoms. Brain 1996;119:363-75. 15. Jeanmonod D, Magnin M, Morel A, Siegemud M. Commentary to stereotactic medial thalamotomy for chronic pain: Is it an effective procedure? In: Burchiel K, editor. Surgical management of pain. New York: Thieme; 2002. p. 808-11. 16. Jeanmonod D, Magnin M, Morel A, Siegemund M. Surgical control of the human thalamocortical dysrhythmia: I. Central lateral thalamotomy in neurogenic pain. Thalamus Relat Syst 2001;1:71-9. 17. Cooper IS, Amin I, Chandra R, Waltz JM. A surgical investigation of the clinical physiology of the LP-pulvinar complex in man. J Neurol Sci 1973;18(1):89-110. 18. Jones EG, Leavitt RY. Retrograde axonal transport and the demonstration of non-specific projections to the cerebral cortex and striatum from thalamic intralaminar nuclei in the rat, cat and monkey. J Comp Neurol 1974;154(4):349-77. 19. Groenewegen HJ, Berendse HW. The specificity of the ‘nonspecific’ midline and intralaminar thalamic nuclei. Trends Neurosci 1994;17(2):52-7. 20. Hirai T, Jones EG. A new parcellation of the human thalamus on the basis of histochemical staining. Brain Res Rev 1989;14(1):1-34.
The central lateral thalamotomy for neuropathic pain
21. Hirai T, Jones EG. Distribution of tachykinin- and enkephalin-immunoreactive fibers in the human thalamus. Brain Res Rev 1989;14(1):35-52. 22. Morel A, Magnin M, Jeanmonod D. Multiarchitectonic and stereotactic atlas of the human thalamus. J Comp Neurol 1997;387(4):588-630. 23. Morel A. Stereotactic Atlas of the human thalamus and basal ganglia. New York: Informa Healthcare USA, Inc; 2007. 24. Munkle MC, Waldvogel HJ, Faull RL. Calcium-binding protein immunoreactivity delineates the intralaminar nuclei of the thalamus in the human brain. Neuroscience 1999;90(2):485-91. 25. Munkle MC, Waldvogel HJ, Faull RL. The distribution of calbindin, calretinin and parvalbumin immunoreactivity in the human thalamus. J Chem Neuroanat 2000;19(3):155-73. 26. Jones EG, Hendry SH. Differential calcium binding protein immunoreactivity distinguishes classes of relay neurons in monkey thalamic nuclei. Eur J Neurosci 1989;1(3):222-46. 27. Blomqvist A, Zhang ET, Craig AD. Cytoarchitectonic and immunohistochemical characterization of a specific pain and temperature relay, the posterior portion of the ventral medial nucleus, in the human thalamus. Brain 2000;123:601-19. 28. Craig AD. Distribution of trigeminothalamic and spinothalamic lamina I terminations in the macaque monkey. J Comp Neurol 2004;477(2):119-48. 29. Graziano A, Jones EG. Widespread thalamic terminations of fibers arising in the superficial medullary dorsal horn of monkeys and their relation to calbindin immunoreactivity. J Neurosci 2004;24(1):248-56. 30. Mehler WR. The posterior thalamic region in man. Confin Neurol 1966;27(1):18-29. 31. Craig AD, Bushnell MC, Zhang ET, Blomqvist A. A thalamic nucleus specific for pain and temperature sensation. Nature 1994;372(6508):770-3. 32. Jones EG. Viewpoint: the core and matrix of thalamic organization. Neuroscience 1998;85(2):331-45. 33. Willis WD. Nociceptive function of thalamic neurons. In: Steriade M, Jones EG, McCormick DA, editors. Thalamus. Amsterdam, Netherlands: Elsevier; 1997. p. 373-424. 34. Willis WD. The pain system: the neural basis of nociceptive transmission in the mammalian nervous system. In: Gildeberg G, editor. Pain and headache. Basel: Karger; 1985. 35. Hodge CJ, Jr., Apkarian AV. The spinothalamic tract. Crit Rev Neurobiol 1990;5(4):363-97. 36. Stepniewska I, Sakai ST, Qi HX, Kaas JH. Somatosensory input to the ventrolateral thalamic region in the macaque monkey: potential substrate for parkinsonian tremor. J Comp Neurol 2003;455(3):378-95. 37. Calzavara R, Zappala A, Rozzi S, Matelli M, Luppino G. Neurochemical characterization of the cerebellar-recipient
38.
39.
40.
41. 42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
123
motor thalamic territory in the macaque monkey. Eur J Neurosci 2005;21(7):1869-94. Sakai ST, Stepniewska I, Qi HX, Kaas JH. Pallidal and cerebellar afferents to pre-supplementary motor area thalamocortical neurons in the owl monkey: a multiple labeling study. J Comp Neurol 2000;417(2):164-80. Percheron G, Francois C, Talbi B, Yelnik J, Fenelon G. The primate motor thalamus. Brain Res Rev 1996;22 (2):93-181. Mason A, Ilinsky IA, Maldonado S, Kultas-Ilinsky K. Thalamic terminal fields of individual axons from the ventral part of the dentate nucleus of the cerebellum in Macaca mulatta. J Comp Neurol 2000;421 (3):412-28. Jones EG. The thalamic matrix and thalamocortical synchrony. Trends Neurosci 2001;24(10):595-601. Peyron R, Laurent B, Garcia-Larrea L. Functional imaging of brain responses to pain. A review and meta-analysis (2000). Neurophysiologie 2000;30(5):263-88. Stern J, Jeanmonod D, Sarnthein J. Persistent EEG overactivation in the cortical pain matrix of neurogenic pain patients. Neuroimage 2006;31(2):721-31. Hsieh JC, Belfrage M, Stone-Elander S, Hansson P, Ingvar M. Central representation of chronic ongoing neuropathic pain studied by positron emission tomography. Pain 1995;63(2):225-36. Casey KL, Minoshima S, Berger KL, Koeppe RA, Morrow TJ, Frey KA. Positron emission tomographic analysis of cerebral structures activated specifically by repetitive noxious heat stimuli. J Neurophysiol 1994;71(2):802-7. Apkarian AV, Bushnell MC, Treede RD, Zubieta JK. Human brain mechanisms of pain perception and regulation in health and disease. Eur J pain 2005;9(4):463-84. Apkarian AV, Hodge CJ. Primate spinothalamic pathways: III. Thalamic terminations of the dorsolateral and ventral spinothalamic pathways. J Comp Neurol 1989;288(3):493-511. Llinas R, Jahnsen H. Electrophysiology of mammalian thalamic neurones in vitro. Nature 1982;297(5865): 406-8. Jahnsen H, Llinas R. Ionic basis for the electroresponsiveness and oscillatory properties of guinea-pig thalamic neurones in vitro. J Physiol 1984;349:227-47. Jahnsen H, Llinas R. Electrophysiological properties of guinea-pig thalamic neurones: an in vitro study. J Physiol 1984;349:205-26. Modesti LM, Waszak M. Firing pattern of cells in human thalamus during dorsal column stimulation. Appl Neurophysiol 1975;38(4):251-8. Lenz FA, Kwan HC, Dostrovsky JO, Tasker RR. Characteristics of the bursting pattern of action potentials that occurs in the thalamus of patients with central pain. Brain Res 1989;496(1–2):357-60. Lenz FA, Kwan HC, Martin R, Tasker R, Richardson RT, Dostrovsky JO. Characteristics of somatotopic organization
2095
2096
123 54.
55.
56.
57. 58.
59.
60.
61.
62.
The central lateral thalamotomy for neuropathic pain
and spontaneous neuronal activity in the region of the thalamic principal sensory nucleus in patients with spinal cord transection. J Neurophysiol 1994;72(4):1570-87. Llinas R, Ribary U, Contreras D, Pedroarena C. The neuronal basis for consciousness. Philos Trans R Soc Lond B Biol Sci 1998;353(1377):1841-9. Llinas RR, Ribary U, Jeanmonod D, Kronberg E, Mitra PP. Thalamocortical dysrhythmia: A neurological and neuropsychiatric syndrome characterized by magnetoencephalography. Proc Natl Acad Sci USA 1999;96 (26):15222-7. Steriade M, Contreras D, Amzica F. The thalamocortical dialogue during wake, sleep and paroxysmal oscillations. In: Steriade M, Jones E, McCormick D, editors. Thalamus. Amsterdam: Elsevier; 1997. p. 213-94. Steriade M, Jones E, Llinas R. Thalamic oscillations and signaling. New York: Wiley; 1990. Schulman J, Ramirez RR, Zonenshayn M, Ribary U, Llinas R. Thalamocortical dysrhythmia syndrome: MEG imaging of neuropathic pain. Thalamus Relat Syst 2005;3:33-9. Sarnthein J, Jeanmonod D. High thalamocortical theta coherence in patients with neurogenic pain. Neuroimage 2008;39(4):1910-7. Sarnthein J, Morel A, von Stein A, Jeanmonod D. Thalamic theta field potentials and EEG: high thalamocortical coherence in patients with neurogenic pain, epilepsy and movement disorders. Thalamus Related Syst 2003;2:231-8. Gucer G, Niedermeyer E, Long DM. Thalamic EEG recordings in patients with chronic pain. J Neurol 1978;219(1):47-61. Sarnthein J, Stern J, Aufenberg C, Rousson V, Jeanmonod D. Increased EEG power and slowed dominant frequency in patients with neurogenic pain. Brain 2006;129:55-64.
63. Pedroarena C, Llinas R. Dendritic calcium conductances generate high-frequency oscillation in thalamocortical neurons. Proc Natl Acad Sci USA 1997;94(2):724-8. 64. Klimesch W. EEG alpha and theta oscillations reflect cognitive and memory performance: a review and analysis. Brain Res Rev 1999;29(2–3):169-95. 65. von Stein A, Sarnthein J. Different frequencies for different scales of cortical integration: from local gamma to long range alpha/theta synchronization. Int J Psychophysiol 2000;38(3):301-13. 66. Llinas R, Urbano FJ, Leznik E, Ramirez RR, van Marle HJ. Rhythmic and dysrhythmic thalamocortical dynamics: GABA systems and the edge effect. Trends Neurosci 2005;28(6):325-33. 67. Young RF, Jacques DS, Rand RW, Copcutt BC, Vermeulen SS, Posewitz AE. Technique of stereotactic medial thalamotomy with the Leksell Gamma Knife for treatment of chronic pain. Neurol Res 1995;17 (1):59-65. 68. Bourgeois G, Magnin M, Morel A, Sartoretti S, Huisman T, Tuncdogan E, Fleier D, Jeanmonod D. Accuracy of MRI-guided stereotactic thalamic functional neurosurgery. Neuroradiology 1999;41(9):636-45. 69. Jeanmonod D, Magnin M, Morel A. A thalamic concept of neurogenic pain. In: Gebhart G, Hammond D, Jensen T, editors. Progress in pain research and management. Seattle: IASP Press; 1994. p. 767-87. 70. Machleidt W, Gutjahr L, Muegge A. Grundgefu¨hle: Pha¨nomenologie. In: Psychodynamik und EEG Spektralanalytik. Berlin: Springer; 1989. 71. Stern J, Aufenberg C, Jeanmonod D. Der Widerhall von Nervenscha¨digungen und Schuldgefu¨hl im Elektroenzephalogramm. Eine neurochirugische Fallstudie zu GehirnOszillationen bei chronischen Schmerzen. In: Schmerz: Perspektiven auf eine menschliche Grunderfahrung. Zu¨rich: Chronos Verlag; 2007:131-58.
140 The Pathophysiology of Trigeminal Neuralgia R.W. Hurt
Introduction The first edition of this textbook was published 11 years ago. It contained Chapter 170, entitled The Pathophysiology of Trigeminal Neuralgia written by William H. Sweet. His chapter is an exceptionally valuable treatise on this subject. His remarkable knowledge, wisdom, and vast experience treating trigeminal neuralgia (TN) patients remain unmatched. I am humbled by the task of updating our understanding of possible causes and mechanisms involved in this most painful affliction. My interest in this topic was kindled by 7 years of training under Dr. Sweet during neurosurgical residency. This was followed by my 30-year experience surgically treating over 700 patients with TN. The late Dr. Sweet, one of the great pillars of neurosurgery, had enormous influence on my life. His scholarship overshadows my own, and it would be impossible for me to stray far from his teachings in writing this chapter.
Definitions for a Classification of Trigeminal Neuralgia The clinical syndrome of trigeminal neuralgia (TN) has been known for centuries, but Nicolaus Andre´, who introduced the term tic douloureux in 1756, has often been credited for one of the earliest descriptions of the symptoms characterizing this painful condition [1]. Since then, many astute physicians have created several classification schemes and a nomenclature system attempting to relate the clinical features of TN, #
Springer-Verlag Berlin/Heidelberg 2009
or tic douloureux, to its etiology (cause) or pathogenesis (mechanism by which the disease is caused). Our current knowledge about causes and mechanisms underlying TN magnifies the importance of establishing acceptable definitions of it and related facial pain conditions. These definitions are very useful in discussions about pathophysiology of TN and helpful in formulating appropriate treatments for this dreadful pain condition. Choices of terms and definitions that follow are based largely on published literature of the past century. The outline in > Table 140‐1 should enable us to focus on the features defining the various conditions that customarily are associated with the term trigeminal neuralgia. The syndrome of idiopathic typical TN has been clearly defined by five cardinal features: facial pain that is paroxysmal, provocable, unilateral, confined to the trigeminal territory, and with absence of sensory loss to light touch or pinprick by usual sensory testing. Idiopathic atypical TN has less clearly defined features and is characterized chiefly by slight departures from the classic symptoms, such as long periods of unprovoked milder pain with superimposed major paroxysms [2]. It is important to recognize this atypical variant because individuals in this category can frequently benefit from conventional medical or surgical treatment, even if the chance of success may be less than in typical cases. In practice, the syndrome of typical TN is quite stereotyped and is frequently encountered in clinics specializing in treatment of facial pain, whereas the atypical variant is seen much less often. Symptomatic (or secondary) TN can be viewed as another nosologic category in distinct
2360
140
The pathophysiology of trigeminal neuralgia
. Table 140‐1 A classification of trigeminal neuralgia A. Idiopathic trigeminal neuralgia (etiology unknown) 1. Typical trigeminal neuralgia (tic douloureux, classical, essential, cryptogenic) 2. Atypical trigeminal neuralgia B. Symptomatic trigeminal neuralgia (etiology recognized) – also called secondary trigeminal neuralgia 1. Clinical features same as idiopathic trigeminal neuralgia (typical) 2. Clinical features unlike idiopathic trigeminal neuralgia (atypical)
contrast to idiopathic TN. The symptomatic group consists of cases where the underlying cause of the trigeminal pain has been recognized. Clinical findings, neuroimaging and other diagnostic testing, or surgical observations that reveal the underlying cause help place these cases in this category. A host of various pathologic entities have been accepted as causes of the TN syndrome, though the underlying pathophysiological mechanisms remain unresolved. In the literature, clinical descriptions of neurological symptoms and findings are often inadequate for placing individual TN cases in their appropriate diagnostic category. This can create problems in the analysis of data from these cases. Identifying facial pain conditions distinct from TN is easier. These include post herpetic zoster neuralgia, postsurgical deafferentation pain, such as anesthesia dolorosa, nervus intermedius and vagoglossopharyngeal neuralgia, and several headache syndromes as well as other familiar clinical entities recognized in the differential diagnosis of craniofacial pain. In addition, there is an ill-defined category of facial pain commonly labeled atypical facial pain (neuralgia). Here, the clinical manifestations bear little or no resemblance to the syndrome of idiopathic TN. The unusual symptoms of this painful condition were described by Sir Charles Symonds in 1949 [3]. Sweet elaborated further on clinical characteristics of atypical facial pain. He admonished that one would be
“well advised to reserve the adjective atypical for those facial pains which are indeed typical of no nosological entity at the present stage of our ignorance.” He added that patients in this category present with pain not limited to the areas supplied by a single nerve or to one side of the face or head. The pain is often bilateral, constant, not paroxysmal, although severe exacerbations may be superimposed, and not provoked by external stimuli. The pain is often described as “gripping,” “drawing,” “pulling,” “boring,” and “bursting.” A tendency for drug addiction and neurotic personality is often present. Insofar as the diagnosis of atypical facial pain may be closely related to the term “psychogenic facial pain,” there is little doubt that psychological factors such as neuroticism or depression may be evident in patients with any chronic pain condition. Sweet eloquently stated: “with regard to the invariably underlying depression in patients with long-standing unalleviated pain, it is our impression that only a vacuous boob would not become disheartened if his medical attendants failed him constantly” [2]. The importance of having a conceptually clear system of classification for TN is evidenced in a classic neurosurgery textbook by Dandy [4] where he described two kinds of neuralgia in the trigeminal nerve: TN synonymous with tic douloureux and TN (not tic douloureux). The latter he described as steady, lasting for hours or days, and the pain not necessarily induced by sensory stimuli [5]. Unfortunately, he did not distinguish the two types of TN he proposed when describing surgical results in his patients. There are several published classifications of TN [6–8]. The one most widely used is perhaps the 1994 edition of The International Association for the Study of Pain Task Force on Taxonomy [8]. This group describes two categories: TN (tic douloureux) and secondary neuralgia (trigeminal) from central nervous system lesions. They make no distinction between idiopathic cases and those where the cause can be attributed to
The pathophysiology of trigeminal neuralgia
impingement on the nerve root by vessels. Jannetta considered vascular compression to be a “symptomatic” cause of TN [9]. Also, the presence or degree of sensory loss permitted in tic douloureux compared to secondary neuralgia is not clearly defined in this taxonomic scheme. In another classification for facial pain, Burchiel proposes two types of TN of spontaneous onset and five diagnostic pain categories due to traumatic injuries. Type I TN is defined as having greater than 50% episodic pain, whereas Type II has greater than 50% constant pain [10]. One might infer that Type I includes cases of idiopathic typical TN and Type II would include patients with idiopathic atypical TN. In practice, it may be difficult to draw a line between proportion of time for episodic and for constant pain. There remains a gray zone of symptoms separating typical and atypical idiopathic TN. Other conceptual schemes have been useful for classifying pain in general as nociceptive or neuropathic pain [11–14]. TN seems to be a unique form of neuropathic pain. A century ago Sir Henry Head described two types of pain, protopathic and epicritic [15]. Now there are efforts to develop mechanism-based classifications of pain [16,17]. New classification schemes of TN will be required as our knowledge of its pathophysiology grows.
Symptomatic Trigeminal Neuralgia In contrast to idiopathic TN, the designation symptomatic TN is customary whenever a cause for the symptoms of TN has been identified. In some cases the clinical syndrome is characteristic of idiopathic TN, but in other cases less typical facial pain symptoms are present. Occasionally, disorders associated with symptomatic TN are located within the central nervous system (intrinsic), but more often they are outside the central nervous system (extrinsic). The most
140
frequent intrinsic disorder causing symptomatic TN is MS, and the most common extrinsic cause is neurovascular compression. The classic syndrome of typical TN is often manifested in both of these conditions. Other symptomatic disorders tend to produce uncharacteristic features such as atypical prolonged pain and significant sensory loss or other neurologic deficits. Except for cases with neurovascular conflicts or MS, relatively few patients with a demonstrable pathologic process affecting peripheral or central trigeminal pathways fulfill the criteria for a diagnosis of typical TN. Lists of pathological conditions reported to cause symptomatic TN are often organized by location along the trigeminal pathway [2,18,19]. > Table 140‐2 is a listing of pathological conditions found in intrinsic and extrinsic locations that may cause symptomatic TN. The pathophysiology of TN in symptomatic cases remains poorly defined. In a study of 16 intracranial tumors in 2,000 patients with facial pain, Bullitt et al. observed that the more peripheral tumors (peripheral nerve and middle fossa) had more atypical features than tumors in the posterior fossa. Patients with typical symptoms had a better prognosis for pain relief after tumor removal [20]. When space-taking lesions, such as tumors, are present, mechanisms producing trigeminal pain have raised several questions. Dandy observed that acoustic tumors were unlikely to cause TN, and, when they did, the “nerve was barely reached by tumor and rarely much deformed” [21]. Mechanical factors, such as nerve stretching or compression, have been suggested as mechanisms [22]. With tumors contralateral to the side of facial pain, rotation of the brainstem causing angulation, or tensing of the nerve, or shifting of the basilar artery to bring an arterial loop into intimate relationship to the trigeminal root entry zone (REZ) have been postulated [23]. There are reports of cases of combined TN and ipsilateral hemifacial spasm (painful tic convulsif) due to posterior fossa tumors [24]. There are reports of tumors contralateral to the
2361
2362
140
The pathophysiology of trigeminal neuralgia
. Table 140‐2 Reported causes of symptomatic trigeminal neuralgia Intrinsic location Multiple sclerosis [315,316] Syringobulbia [37,318] Hydrocephalus [319,320] Lateral C1–2 neurinoma [323] Medullary compression by posterior inferior cerebellar artery [328–330] Pontine cyst [332] Extrinsic location Vascular lesions Normal vessels causing root compression Arteries [21,111] Veins [21,111] Abnormal vessels Aneurysm [337–339] Vertebrobasilar ectasia [21] Persistent primitive trigeminal artery [340,341] Cavernous angioma [21] Posterior fossa arteriovenous malformation [344] Intrinsic arteriovenous malformation of nerve [345] Microarteriovenous malformation (parenchymal/subpial) [346] Cryptic angioma [347] Carotid-cavernous sinus fistula [349] Dural arteriovenous fistula [351,352] Space-taking disorders Neoplasms Neurinoma (vestibular, trigeminal) [21,357,358] Epidermoid [21,357,362,363] Meningioma [357,364] Lipoma [371,372]
Pontine abscess [317] Infarct Medullary [19,321,322] Pontine [324–327] Viral rhombencephalitis [331]
Skull deformities Basilar impression (platybasia) Congenital [21,333] Paget’s disease [334,335] Osteogenesis imperfecta [336] Elevated petrous apex [335] Small posterior fossa [119] Arnold-Chiari type I malformation [342,343] Infections Nonviral
Sinusitis [19]
Gradenigo’s syndrome (petrositis) [19]
Syphilis [37,348] Lyme disease [350] Tuberculoma [353] Actinomycosis granuloma [354] Aspergillus granuloma [355,356] Cysticercosis [359–361] Viral Herpes simplex virus (unproven) [365–370] Other (almost always very atypical features)
. Table 140‐2 (Continued) Intrinsic location Hamartoma (neuronal, choristoma) [373,374] Embryonal rhabdomyosarcoma [376] Pituitary adenoma [377] Osteoma [21,378] Lymphoma [379] Chordoma [18] Chondrosarcoma [18] Malignant peripheral nerve sheath tumor [380] Malignant perineural invasion (basal cell, nasopharyngeal cancer) [381,382] Metastatic neoplasms [19] Benign tumors, cysts, adhesions Amyloidoma [383,384] Sphenoid mucocele [18]
Arachnoid cyst [387] Choroidal epithelial cyst [388] Arachnoid adhesions [119, 348]
Trauma (craniofacial) [375] Dental [19,43] Toxins [19] Stilbamidine Trichlorethyline Systemic disease Scleroderma [19] Dermatomyositis [19] Systemic lupus erythematosus [19]
Sjo¨gren’s syndrome [19] Sarcoidosis [19] Mixed connective tissue disease [385] Mitochondrial neurogastrointestinal encephalomyopathy [386] Miscellaneous Trigeminal neuritis [322] Tolosa-Hunt syndrome [18] Charcot–Marie–Tooth disease (hereditary peripheral neuropathy) [46]
side of painful tic convulsive [25] where trigeminal pain may persist after removal of contralateral tumors, but may respond to rhizotomy [26] or to microvascular decompression (MVD) [27]. With respect to central lesions causing TN, Sweet noted that there is relative preservation of axons and neurons in plaques of MS in contrast to brainstem infarcts that afflict with greater avidity the highly vascular clusters of neurons and tracts. Thus, in MS there is relative preservation of sensation, but the area of demyelinated axons may permit ephaptic jumping to the lightly myelinated pain fibers [28]. In five patients
The pathophysiology of trigeminal neuralgia
with both TN and MS Resnick et al. found MVD to be unreliable and recommended partial rhizotomy [29] as did Eldridge et al. [30], whereas Athanasiou et al. obtained pain relief in four out of five patients who underwent MVD [31]. Three patients with TN occurring in Charcot– Marie–Tooth disease underwent MVD with favorable response, but later one required a neurotomy [32]. Thus it appears that some consider neurovascular conflicts to be the cause of TN even in symptomatic cases due to tumor, MS, or peripheral demyelinating disease, as if the patient may harbor a “double-disease” associated with TN. The importance of a prompt, thorough diagnostic evaluation including appropriate imaging studies in all patients with TN, atypical or typical, is highlighted by the various treatable pathological conditions associated with symptomatic TN. In an extensive study at the Mayo Clinic, 5,058 patients with facial pain were seen from 1976 through 1990. TN was diagnosed in 2,972 patients, while tumors were the cause of facial pain in 296 patients. Neurologic deficits developed on follow-up evaluation in 47%. The most important observation was that there was an average delay of 6.3 years in tumor diagnosis after the patient presented with TN [33].
Trigeminal Neuralgia: Epidemiology and Genetics Epidemiologic studies. Like intracranial tumors, TN may be considered an uncommon disease. Nevertheless, pain conditions in general constitute a major public health burden. It is generally agreed that the exceptional pain in TN is the worst of all painful conditions afflicting mankind. Epidemiologic studies have shown the incidence rates and prevalence figures as well as the associated features and risk factors in TN. The most cited incidence rate of TN is 4.3/100,000/ year reported from the Mayo Clinic by Kurland
140
[34] and Katusic et al. [35] based on a survey of Rochester, Minnesota from 1945 to 1984. A lower incidence rate of 2.1/100,000/year was found in Carlisle, England [36]. Prevalence of TN has been estimated to be 155/1 million [37]. According to the U.S. Census Bureau in early 2008 the population of the United States was slightly over 300 million. Based on current incidence rates and estimated prevalence figures, there will be 21,000 new cases of TN per year. An estimated prevalence of about 50,000 individuals with TN in the United States in 2008 is probably an underestimate. Even if TN is not common, these numbers emphasize the importance of early diagnosis and prompt institution of known effective treatment in order to minimize the extreme pain and suffering experienced by the many individuals with TN. Among the factors associated with this disease, several that may reflect on its pathophysiology have been studied. The onset of TN is mostly in middle and older age groups, with a mean of about 51 years [38]. The incidence increases with age [35]. In patients with MS, mean age of onset of TN is about 5 years earlier than in the idiopathic variety. Onset of TN below age 20 is extremely uncommon. Jannetta has reported on 23 patients whose onset was below 18 years of age. When he performed MVD, he concluded that this pediatric cohort did not have the same beneficial response as adults [39]. He also has a case report of a child whose TN apparently began at age 13 months [40]. It was customarily reported that TN was significantly more common in women than men, but Rothman and Monson, in an epidemiologic study of 526 patients admitted to the Massachusetts General Hospital and Lahey Clinic in Boston, found the sex ratio to be 1.17:1.00 females:males. Their explanation for this finding was that all estimates had failed to consider the sex ratio in the population at risk: TN affects old people, among whom females predominate [41]. Rothman and Beckman examined the tendency for TN to occur on the right side of the face.
2363
2364
140
The pathophysiology of trigeminal neuralgia
They studied the specific trigeminal division affected and found the right side predominance occurred in patients with pain in the upper face. Of 500 cases studied, 245 had single division involvement: 40.4% mandibular, 50.2% maxillary, and 9.4% ophthalmic. The remaining 255 cases had two divisions involved: 27.8% ophthalmic and maxillary, and 72.2% maxillary and mandibular. Right-sided pain occurred in 53.1% when solely the mandibular division was involved in contrast to 69.2% with only maxillary involvement [42]. Harris treated over 2,500 patients with Gasserian alcohol injections and found that about 5.3% had a history of bilateral TN, sometimes with a delay of many years before the pain began on the initially unaffected side [43]. In a study of 526 patients with TN, Rothman and Wepsic found that about 3% had bilateral involvement. In the remaining 508 cases, the right side was affected in 62.2%, confirming the predominance of the right side of the face in TN. They also reviewed handedness and found no association with the side of facial pain [44]. Considering overall health risks, it was found that TN patients may survive longer than the general population [45]. Hypertension [35], MS [38], and Charcot–Marie–Tooth neuropathy [46] have an association with TN. In an epidemiologic study of vagoglossopharyngeal neuralgia there were no patients with MS [35]. The possible association of TN with vagoglossopharyngeal neuralgia has been reviewed [47]. Addressing suggestions that recurrent herpes simplex virus infections might cause TN, Rothman and Monson found no association between a history of frequent cold sores and TN [45]. Similarly, previous postulates that an elevated petrous apex could cause TN were contradicted in the radiological study of Rothman and Wepsic of 46 patients with TN. They found that only 30% of patients had an elevated petrous apex on the painful side. Conversely, 70% did not have such ipsilateral elevation [44].
The epidemiological features of the well known association of MS with TN have been described. Rushton and Olafson, in a Mayo Clinic study of 1,735 patients with TN and 3,880 with MS, found 35 cases of combined TN and MS. They found that 1% of patients with MS had TN and 2% with TN had MS. As noted previously, in patients with MS, TN started at a slightly earlier age than idiopathic cases and was more often bilateral (11%) [38]. In another study of 900 patients with TN, 22 (2.4%) had associated MS. Of these, 16 had typical and six had atypical TN. In three patients TN was the first manifestation of MS, and in 32% of cases the neuralgia was bilateral [48]. A study at a Canadian MS clinic followed 1,882 patients, of whom 35 (1.9%) developed TN. In five patients it was the first symptom, preceding the appearance of the next symptom of MS by 1–11 years; TN was bilateral in 14% of cases [49]. Hereditary Factors. The influence of heredity has been examined for its possible role in the etiology of TN. Lineages with multiple members of one family afflicted with TN are rarely reported. Recent publications have described six of seven siblings with TN [50], three non-twin sisters of seven siblings, and possibly a paternal aunt with TN [51]. Of those affected who chose surgical ablative procedures, all had favorable results, except for one who had a recurrence after percutaneous rhizotomy and then underwent MVD, but pain relief lasted only 3 weeks. In another report four family members in three generations were affected. Two of them underwent posterior fossa exploration with no vascular compression found [52]. Smyth et al. described four family members in three generations with TN. Two underwent successful MVD. Their review suggested autosomal dominant transmission leading them to consider a genetic locus such as an autosomal dominant vascular disorder or an autosomal dominant epileptic condition, but there was no evident connection in these cases [53].
The pathophysiology of trigeminal neuralgia
In his textbook, Adams discusses the observed familial tendency in patients with MS. In a first-degree relative of a patient with MS, the risk of developing MS is at least 15 times greater than for a member of the general population. The risk is greater for siblings than for parents, but this may simply reflect exposure to a common environmental factor [54]. In their paper on MS, Frohman et al. stated that there was substantial evidence to support the hypothesis that genetics has an important role in a person’s susceptibility to MS, probably in conjunction with environmental factors [55]. Another demyelinating disease, hereditary peripheral neuropathy, also called Charcot–Marie–Tooth disease, has been reported to have an association with TN. Testa et al. described a family with dominantly inherited Charcot–Marie–Tooth disease in eight members over three generations, and five of these had TN [56]. Coffey and Fromm reported on two different families with Charcot– Marie–Tooth disease. In one family, mother and son had Charcot–Marie–Tooth disease and the son had TN. In another family, in three generations eight individuals had Charcot–Marie– Tooth disease and four of these had TN [46]. Rarely, an individual may have both vagoglossopharyngeal and TN. A family of nine involving three generations had three members with the combined neuralgias [57]. The association of TN with other cranial nerve disorders, so called multiple cranial nerve dysfunction, such as hemifacial spasm, has been observed. Duff et al. reported a family of 23 members in three generations in which eight had TN (six of ten in one generation). One of these also had hemifacial spasm, but it was on the contralateral side [58]. Although TN almost always occurs sporadically, the rare familial cases deserve our attention with respect to the pathophysiological mechanisms in this disease. In contrast to genetic factors that may predispose to development of pain, in humans there are recognized conditions such as “congenital
140
insensitivity to pain.” A case of this unique condition was reported where serial neurological exams showed no definite abnormalities, and later, at autopsy, no abnormality of the nervous system was found [59]. A slightly similar recessively inherited disorder, called familial dysautonomia (Riley-Day disease), has a clinical feature of loss of pain and temperature sensation, and a diminution of small myelinated and unmyelinated nerve fibers on biopsy [54]. Sweet has reviewed this subject extensively and related it to experimental animal models of pain, including autotomy (self cannibalism), after spinal posterior rhizotomy or peripheral neurectomy. He described several sites where supratentorial surgical lesions may relieve relentless chronic pain, as may occur in cancer, yet this is accomplished without producing objective analgesia or thermoanesthesia. Such lesions may not affect the ability to feel acute pain [60]. Hurt and Ballantine [61] reported similar observations after anterior cingulate lesions were made in 68 patients with persistent pain from cancer or other severe chronic pain conditions. Devor and coworkers have studied predisposition to neuropathic pain in genetically different populations of rats and mice [62–64], and reviewed regulation of gene expression or “gene chips” correlated with changes in density and disposition of molecules of excitability in the neural membrane [65].
Normal and Pathological Anatomy of the Trigeminal System Human cadaver and biopsy studies of the anatomy of the trigeminal system in normal individuals and those with TN have yielded valuable information about structural lesions in the nerve that may be related to the pathophysiology of TN. These studies include trigeminal peripheral nerve branches and receptors, Gasserian (trigeminal, semilunar) ganglia with pseudounipolar sensory neurons, roots, central brainstem tracts
2365
2366
140
The pathophysiology of trigeminal neuralgia
and second order sensory nuclei, as well as their complex connections with other structures in the brain. With the advent of electron microscopy and growth of advanced neuroanatomical techniques, much has been learned about normal anatomy and ultrastructure in the trigeminal system [66,67]. Normal anatomy. Two components of the normal trigeminal system–the Gasserian ganglia and the motor roots – have been studied using electron microscopy. The fine structure of normal trigeminal ganglia was examined first in animals as a framework for similar studies in humans [68,69]. Based on previous animal studies, Beaver described the ultrastructure of neuronal cells, satellite cells, and axons in the trigeminal ganglia postmortem. He found he could distinguish autolysis artifact from intrinsic ganglion tissue [70]. Moses examined human ganglia from random autopsies, providing additional descriptions of fine structure in normal satellite and neuronal cells and myelinated and unmyelinated axons emanating from the neuronal ganglion cells [71]. In humans, Young examined the ultrastructure of the trigeminal motor roots for possible sensory fibers. He found that 20% of motor root fibers were unmyelinated, suggesting they were afferent fibers from the ganglion [72]. Comparable electron microscopic studies of normal trigeminal sensory roots and brainstem structures have not been done. However, the studies of normal tissue have provided a basis for similar investigation of specimens from individuals with TN obtained postmortem or from understandably limited biopsy material removed during surgery. Pathological anatomy – peripheral nerve. Fortunately, there are significant findings from several investigators as a result of electron microscopic studies of tissue taken from individuals with a history of TN. Earlier histologic studies with optical microscopy yielded little information about pathology in TN. Starting with
peripheral branches of the trigeminal nerve, Kumagami performed electron microscopic studies in two cases. The first had a 1-week history of “mental neuralgia.” The mental nerve showed hypertrophic myelin sheaths and undulation of axon cylinders giving the appearance of microneuromas. The second case had a 15-month history of “lingual neuralgia,” and a portion of the lingual nerve showed marked thickening and bizarre undulation of myelin sheaths with exposed axis cylinders due to segmental demyelination. In two cases of hemifacial spasm, facial motor nerve biopsies showed the same changes [73]. Pathological Anatomy – Ganglion. The fine structure of surgical biopsies of the Gasserian ganglion in patients with TN has been studied more extensively. Beaver et al. [74,75] examined specimens from 11 patients and found degenerative hypermyelination, segmental demyelination, hypertrophy of axis cylinders with microneuroma formation, and vacuolization of Nissl substance in the neuron. (> Figure 140‐1). Using electron microscopy, Kerr studied biopsy material from another 15 patients. In six of his cases with typical TN, uncomplicated by previous injections or surgery, he found prominent proliferative degenerative changes in all myelin sheaths, inconsistent with Wallerian degeneration because the axons were usually preserved instead of disintegrating and disappearing. He concluded that other changes observed, such as segmental demyelination, inclusions in Schwann cells, and moderate proliferation of myelin were due to normal aging processes [76,77]. Pathological anatomy – roots. Before discussing pathologic anatomy studies of trigeminal sensory roots, a brief review of the normal root anatomy in the middle and posterior fossa may be useful. Ferner has beautifully illustrated the sensory rootlets emerging from the Gasserian ganglion to form a plexus with anastomotic connections lying in Meckel’s cave in the middle cranial fossa. He calls this the pars triangularis.
The pathophysiology of trigeminal neuralgia
140
. Figure 140‐1 Tortuous hypertrophic axis cylinder giving the appearance of a “plexiform microneuroma.” Degenerating Schwann cell cytoplasm is discernible between the folds. Note also the extreme thinness of myelin sheath. X 9,000 (From Beaver et al. [75], with kind permission of Springer Science + Business Media)
The rootlets merge to pass through a meningeal opening at the petrous ridge (porus trigemini) and proceed through the posterior fossa within the pontine cistern to become a compact bundle (pars compacta radicis trigemini). The compact bundle of roots enters the midpons continuing as tracts to synapse with their respective brainstem nuclei [78]. Another name used for the pars compacta is portio major and for the motor root(s) is portio minor of the trigeminal nerve. There are small
root fascicles between the sensory and motor roots. These and the sensory REZ, 2–3 mm external to the pons, will be discussed in more detail later. In their description of ultrastructural findings in the trigeminal ganglia, Kerr and Beaver also studied sensory roots in the middle fossa arising from the Gasserian ganglia [74,77]. Other investigators have studied the fine structure of the sensory root near the pons including the REZ. The material examined has consisted of
2367
2368
140
The pathophysiology of trigeminal neuralgia
limited biopsy specimens taken from patients undergoing partial root sectioning for TN. Hilton et al. [79] reported on six rhizotomy specimens in which no vascular root compression was found during surgery. The biopsies were taken 1–2 mm from the REZ and electron microscopy showed minor nonspecific abnormalities with very occasional degenerating myelinated fibers. One of their specimens was from a patient with MS, and it showed demyelination, perivascular lymphocytes, scattered lipid filled macrophages, and numerous astrocytic processes. Another biopsy specimen was taken at the site of compression by a vein and small pontine arteriole. It revealed closely packed, demyelinated axons and occasional large, thickly myelinated axons they suggested could be remyelinated fibers [79]. Later Love et al. described their ultrastructural studies of two rhizotomy specimens. One had a dolichoectatic vertebral artery compressing the root and another had a small, unnamed artery between the origin of the basilar
artery and root, indenting the latter. By light microscopy there was a well-circumscribed zone of demyelination in the central part of the root and very occasional degenerating fibers with aberrantly myelinated axon clusters seen by electron microscopy [80]. In a review article, Love and Coakham [81] reported on several additional rhizotomy specimens taken from patients with vascular root compression. In the immediate vicinity of the point of vascular indentation foci of apposed demyelinated axons along with glial and inflammatory cells were consistent ultrastructural features (> Figure 140‐2). In a small number of rhizotomy specimens, obtained from patients with visible vascular compression, no ultrastructural abnormalities were seen. They believed this could reflect a sampling error [81]. Another very important contribution to our knowledge of ultrastructural abnormalities in the trigeminal sensory roots at the site of vascular compression has come from the collaborative
. Figure 140‐2 Electron microscopy shows large numbers of demyelinated nerve fibres in the region of compression, many of the fibres being in direct apposition; some of them are indicated by arrows (From Love and Coakham. [81], with permission of Oxford University Press)
The pathophysiology of trigeminal neuralgia
studies of Rappaport and Devor. They studied biopsy specimens in 12 consecutive cases taken as close as possible to the site of vascular compression, when present, during surgery for MVD. In four instances, no clear arterial compression was observed. Root compression, flattening, or grooving was noted by the surgeon in the other cases. They found extensive dysmyelination, demyelination, and axon loss (> Figure 140‐3). The most severe pathology noted was that the zone of demyelination was not in a single contiguous region but rather interdigitated with zones of dysmyelination where there was less severe disruption of the myelin sheaths and more surviving axons. In zones of demyelination with reduced axon population, the axolemma was directly exposed to the extracellular medium and neighboring denuded axons directly abutted other axons. Also, in the extracellular matrix there was striking overproduction of collagen fibrils and occasional massive, condensed collagen clumps. In a small proportion of myelinated fiber profiles there appeared to be individual large
140
and small axons in close apposition. This was interpreted to show numerous regenerating sprouts, presumably originating at a site of injury between the ganglion and the point of observation. Most specimens contained adjacent patches of fibers having both peripheral and central types of myelin. These two types of myelinated fibers showed both demyelination and dysmyelination changes [82,83]. The same investigators studied biopsy specimens from the glossopharyngeal and vagal nerves in a patient with vagoglossopharyngeal neuralgia and found large patches of demyelinated axons in close membrane to membrane apposition and zones of dysmyelination similar to the trigeminal root abnormalities from previous studies [84]. Pathological anatomy – brainstem. The normal anatomy of the tracts and nuclei of the trigeminal system in the brainstem has been studied extensively. Except for the mesencephalic nucleus (to be discussed later), all the trigeminal primary afferent axons, whose cell bodies reside in the Gasserian ganglion, enter the pons and
. Figure 140‐3 An electron microscopic preparation of the trigeminal root transition zone at the site of vascular compression: areas of myelin loss (upper left) and dysmyelination (center and bottom right) are visible. Swollen axons with a thin myelin sheath are prominently seen (cross-section of picture = 60 mm) (From Rappaport et al. [82], with permission of Karger, Basel)
2369
2370
140
The pathophysiology of trigeminal neuralgia
proceed as tracts to connect with the nucleus principalis in the pons, and/or continue caudally as the descending spinal tract to connect with nuclei in the pons, medulla, and upper cervical spinal cord: nucleus oralis, nucleus interpolaris, and nucleus caudalis in rostrocaudal order. The nucleus caudalis is in direct continuity with the dorsal horn of the upper spinal cord. As the trigeminal spinal tract descends and connects with its respective nuclei, fiber size tapers. This anatomic arrangement of the trigeminal system in the brainstem has been defined by several noted anatomists and is depicted in > Figure 140‐4 compiled by White and Sweet [2]. Based on clinical observations suggesting that afferent fibers for pain terminate in the nucleus caudalis in the lower medulla, Sjo¨qvist [89] reported on nine cases of TN in which he surgically sectioned the bulbar descending trigeminal tract. This resulted in several cases with successful pain relief accompanied by analgesia
but some preservation of touch sensation in the face [89]. There is one study of the pathological anatomy of the descending trigeminal tract in the region of the brainstem using electron microscopy of biopsy specimens taken during surgical medullary trigeminal tractotomy. Reporting on ultrastructure of biopsy specimens, Kunc et al. found mostly nonspecific changes in the myelin sheaths with degenerative hypermyelination (> Figure 140‐5) [90]. In summary, the central theme of these ultrastructural studies of the pathological anatomy in each region of the trigeminal system has been confirmation of abnormalities of myelination associated with close apposition of axons and occasional appearance of “microneuromas.” No clear abnormality has been found in the sensory neuronal cell bodies other than vacuolization of the Nissl substance. The number of cases studied and the description of the clinical syndrome in individual cases is limited. There is a need for
. Figure 140‐4 Central afferent pathways of trigeminal rootlets, redrawn from data of Ramo´n and Cajal [85], Windle [86], Olszewski [87], and Wall and Taub [88], illustrating (1) collaterals from same fibers passing to main sensory nucleus and to different nuclei of descending tract; (2) small descending fibers; (3) bifurcating ascending and descending fibers; (4) decrease in caliber of large fibers as they descend (From White et al. [2], courtesy of Charles C Thomas Publisher, Ltd., Sprigfield, Illinois)
The pathophysiology of trigeminal neuralgia
140
. Figure 140‐5 Irregularly waved and intermingled myelin sheaths and hypermyelinisation (degenerative hypermyelinisation, proliferative changes in the sheaths). X 5,000 (From Kunc et al. [90], with permission of Springer Science + Business Media)
continued effort to study rhizotomy specimens in cases with and without recognized vascular root compression undergoing partial root section. The correlation between specific abnormalities of ultrastructure and the clinical features in individual cases of TN remains a challenge (> Figure 140‐6).
Multiple Sclerosis and Trigeminal Neuralgia Multiple sclerosis (MS) is a neurological disease with a unique association with TN. This close association with MS (1% of patients with MS develop TN) is unlike any other neurological disease. The lesions of MS are within the central nervous system, affecting central trigeminal structures. In contrast, other causes of symptomatic TN, such as vascular root compression, affect the nerve peripherally. Adams and Kubik [92] described the neuropathology of MS in a classic study of about 15 cases. They classify MS as a demyelinating disease characterized by destruction of myelin sheaths
with relative sparing of axons, nerve cells, and supporting structures. The lesion may be multiple, disseminated throughout the brain (and spinal cord), or a single spreading focus. They state that these lesions have been referred to by the French as plaques, meaning patches, and may measure less than one millimeter to several centimeters in diameter. In the brainstem they found that myelin beyond the transition between glial and Schwann cells was normal. In chronic cases many fibers may be destroyed, but even in the oldest lesions a large proportion of axis cylinders persisted. The lesions showed no preference for nuclear masses or tracts, and the pons and medulla nearly always contained lesions (> Figure 140‐7) [92]. Subsequently, we have learned much more about the pathogenesis of the inflammatory and neurodegenerative elements of the MS plaque. An excellent review of this subject by Frohman et al. describes acute and chronic changes in plaques. Reflecting the dynamic nature of the disease process, old and new lesions are present in both acute and chronic cases of MS. They felt that axonal injury and cumulative loss of axons
2371
2372
140
The pathophysiology of trigeminal neuralgia
. Figure 140‐6 Honore´ Daumier, 1808–1879. Lithograph reproduced by the courtesy of the Museum of Fine Arts, Boston, Mass. Daumier’s legend to this cartoon reads: “M. Babinet pre´venu par sa portie`re de la visite de la come`te” (From Heimer [91], with permission of Springer Science + Business Media.) “(M. Babinet is apprised by his housekeeper of the arrival of the comet)”
resulted from inflammatory demyelination accompanied by some evidence of remyelination conspicuous in early lesions and in satellite areas of large lesions (shadow plaques). They stress the importance of using advanced MRI techniques to identify previously unrecognized discrete abnormalities and to follow evolution of the disease in MS [55]. Anatomically, it is clear that MS is a disseminated demyelinating disease affecting many regions of the central nervous system with plaques frequently distributed widely throughout the brainstem including the trigeminal system. The disease is confined to the central nervous system where myelin is formed by oligodendroglial cells, and it may extend into the central myelin portion of the trigeminal sensory root within 2–3 mm of the brainstem. The remainder of the nerve root, ganglion, and
peripheral branches of the trigeminal nerve are not involved in the disease process. The significance of the site of MS plaques with respect to trigeminal structures in the brainstem, and their correlation with clinical symptoms in cases with TN, has been reviewed extensively by Sweet. His detailed account of ten published autopsied cases revealed instances of plaques located in the pontine root zone, extending slightly into the central nervous system segment of the root, and some involving the brainstem tracts and nuclei. In four of the ten cases no REZ lesion was found on the painful side. Furthermore, there were two cases with REZ lesions, but no history of ipsilateral facial pain [28]. Lazar reported on ultrastructure studies of sensory root biopsy specimens taken adjacent to the pons in one patient with MS and TN. In this
The pathophysiology of trigeminal neuralgia
140
. Figure 140‐7 Multiple sclerosis. A, section of pons from a case in which trigeminal neuralgia was a symptom; myelin stain. B, pons from a case of acute multiple sclerosis, showing a region of demyelination made up of numerous smaller lesions, often separated by a zone of incomplete demyelination. A large proportion of the smaller lesions are perivascular (From Adams and Kubik [92], reprinted with permission of Elsevier Limited)
case, without vascular compression, fraying and fragmentation of myelin was seen [93]. Love et al. reported six cases with MS and TN who had undergone previous radiofrequency rhizotomies. Surgical biopsies obtained during open partial sensory rhizotomies, where no vascular compression was seen, showed foci of demyelination and gliosis. There was very occasional evidence of axon degeneration in the proximal CNS part of the root. Degenerating fibers present in the distal fragments of the root were attributed to previous radiofrequency rhizotomy. In places, clusters of axons were in close apposition. These findings were somewhat different from their previous studies in non MS cases where vascular compression was observed. In MS cases astrocytic processes were much more uncommon deep within the areas of demyelination, and inflammatory cells were more likely to be present [81]. The finding of demyelination in MS plaques, and demyelination in the trigeminal system in
cases without MS is a common theme in the pathologic studies of TN. It is intriguing that patients with widespread multifocal central nervous system plaques in MS and focal lesions attributed to vascular sensory root compression can often have longstanding severe pain of TN limited to only a single division, usually the maxillary division. Equally remarkable are the inexplicable remissions of long duration followed by recurrent pain in the same division.
Trigeminal REZ The trigeminal sensory REZ is a well-defined short segment of the sensory root situated 2–3 mm external to the pial surface of the pons. This anatomic structure has acquired special significance pathophysiologically because numerous investigators and clinicians have supported the hypothesis that injuries specifically to the REZ, especially by vascular compression from
2373
2374
140
The pathophysiology of trigeminal neuralgia
neighboring vessels, is the cause of TN, at least in cases where neurovascular compression is present. Detailed anatomy of the REZ has been studied for years by many investigators. Stated briefly, the REZ is the junctional zone where axons arising from the Gasserian ganglion change from being myelinated by Schwann cells of mesodermal origin, to being myelinated by oligodendroglial cells of neuroectodermal origin, before they enter the brainstem. This junction between peripheral and central myelin is equivalent to a node of Ranvier for individual myelinated axons. Similar REZs are found in other sensory cranial nerves and in the sensory roots of spinal nerves (dorsal REZ). History of the REZ. Historically, the REZ has also been called the Obersteiner-Redlich zone. Its discovery is attributed to a well-known Viennese neuroanatomist and pathologist, Heinrich Obersteiner, and his pupil, Emil Redlich. From their findings (1894) of the pathology of tabetic degeneration of the posterior columns, they postulated that the initial site of involvement was at the place in the spinal sensory root where the root becomes a central tract. In a biographical sketch by Spiegel, it was interesting to note that, in addition to discovering the REZ, as a hobby Obersteiner collected microtome sections of various sausages [94]. Clinicians have recognized that the unique paroxysmal pain characteristic of TN does not occur in disorders of the spinal sensory system even though the REZ (which was originally described in tabes dorsalis), is principally involved. Tabes dorsalis, a spirochetal infection, is rarely encountered today, but it may be worthwhile to review the pain symptoms of tabetic neurosyphilis and compare them to the pain in TN, since the REZ may be affected in both conditions. In his textbook of neurology, Adams [54] described pain present in over 90% of tabetic cases as lancinating or lightning pain, short, stabbing and brief like a flash. These pains are more frequent in the legs, but roam over the body from
face to feet sometimes playing persistently on one spot, “like the repeated twanging of a fiddle string,” coming in bouts lasting several hours or days. There may be impairment in tactile, pain, and thermal sensation. A pathology study revealed thinning and grayness of the posterior roots, with only a slight outfall of neurons in the dorsal root ganglion, and essentially normal peripheral nerves [54]. Anatomy of the REZ. There are detailed anatomical studies of the trigeminal REZ. Tarlov [95] reported an important histological examination of the REZ in human cranial nerves. He noted that, as rootlets leave the neural axis, they are surrounded by pia mater that is continuous peripherally beyond the transition zone or REZ with the endoneurium, the arachnoid with the perineurium, and the dura mater with epineurium of the peripheral segment. The central portion containing oligodendroglia, astrocytes, and microglial cells possessed the structure of a fiber tract in the brain, except that the neuroglial mesh was denser. The interstitial glial substance was replaced in the peripheral segment by a sheath of Schwann cells and endoneurium corresponding essentially to that of a peripheral nerve. As the nerve undergoes transition from central to peripheral, it passes through a double lamina cribrosa of dense neuroglial fibers and connective tissue. Peripheral to the transition zone, the axis cylinder and its myelin sheath was thicker than centrally. The transition zone was not always sharp and regular, and in some cases it was ragged presenting gaps that allowed tongues of glial tissue to project peripherally. Centrally, fibrous astrocytes produced an intricate weave binding the nerve fibers together. A single oligodendroglial cell may contribute to envelopment of several nerve fibers surrounding the myelin sheath closely. The pial ring reinforced by transverse and obliquely running fibers gave the nerve the appearance of great strength. Peripheral to the transition zone the pial ring was continuous with longitudinal endoneurial fibers. Central to
The pathophysiology of trigeminal neuralgia
the pial ring the pia was in continuity with connective tissue structures anchoring the root in the glial zone. There was attenuation of myelin in the transition zone, and the nerve fibers passing through the pial ring were tortuous giving the appearance of interrupted myelin. Various endoneural structures extended appreciable distances from the glial segment and supported the root at that point [95]. More recently, ultrastructural examinations of the REZ have been done. In a careful study of the trigeminal root central-peripheral transition zone in macaque monkeys, Maxwell et al. demonstrated by electron microscopy that the grossly visible “fibrous cone” of the REZ, previously described by Dandy [96] and Janetta and Rand [97], corresponded to a “glial dome” covered by a basal lamina demarcating a short transition from peripheral to central nervous system. This dome consisted of closely interwoven
140
astrocytic processes identical to the subpial astrocytic meshwork elsewhere in the central nervous system. The axonal transition from central to peripheral occurred at nodes of Ranvier where the basal lamina of the dome was continuous with the basal lamina of the Schwann cell at the last peripheral internode (> Figure 140‐8). Myelin in the transitional region commonly appeared poorly preserved with extensive splitting along the lamellae. Interrupted myelin in the dome was interpreted to be a likely artifact due to handling and shrinkage during fixation and other procedures [98]. These anatomical observations of the REZ structure in humans and macaque monkeys provide a useful basis for the interpretation of abnormalities found in biopsy specimens of the trigeminal root and the REZ. If a cause of TN is mechanical compression by a vessel at the REZ, the mechanism that makes this discrete region of
. Figure 140‐8 Entrance of a peripheral (upper right) myelinated fiber into the central nervous system (lower left) at the node of Ranvier (N). The extracellular space of the peripheral portion of the root (E) contains collagen fibers and deeply indents the dome. F indicates fibroblast processes common to peripheral nerves. The basal lamina of the dome and the Schwann cell of the first internode is continuous (arrows). Junctional complexes of the zonula occludens variety (O) and of the macula adherens type (a) are indicated. Astrocyte processes of the dome are labeled (A), and their fibrillar and glycogen granule content is evident. Note the diminution in thickness of the myelin sheath as it passes from peripheral to central. X 11,000 (From Maxwell et al. [98], reprinted with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc)
2375
2376
140
The pathophysiology of trigeminal neuralgia
the nerve root more susceptible to injury by mechanical forces than other regions of the nerve root remain speculative. Sunderland described several types of injury to peripheral and intracranial nerves. He discussed particular aspects of the supporting tissues of the nerve including the endoneurium, perineurium, and epineurium, and, in the case of intracranial nerves, he said each nerve fiber is ensheathed in endoneurium but perineurium and epineurium were absent. Sunderland concluded that intracranial nerves are more “vulnerable” to damage from any deforming forces because epineurium and perineurium offset the effects of stretch and compression. The undulation seen in peripheral nerves was not prominent in the intracranial segment, thereby reducing the tolerance of nerve fibers to stretch. He described some known features of the REZ, but he did not suggest that it was a point of vulnerability in an intracranial nerve [22]. Measurements. Several investigators have made anatomical measurements of the intracranial portion of the trigeminal nerve, including the distance of the REZ from the point of root entry into the pons. Gudmundsson et al. in 50 specimens found that the mean length of the trigeminal sensory root from ganglion to pons was 22 mm [99]. Soeira et al. in 20 specimens found that the mean length from ganglion to porus was 10 mm [100]. Lang found that the root length in the posterior fossa (cisternal portion) was 12–13 mm [101], and Peker et al. in 100 specimens found that the mean root length in the cisternal region was 12 mm [102]. Determining the distance of the REZ from the pons, de Ridder et al. found that the mean distance was 2.6 mm based on measurements from several investigators [103]. Ziyal in six specimens found that the distance of the REZ to pons was 2–2.5 mm [104]. In a more extensive study, Peker et al. examined 100 specimens in 50 cadavers and determined that the mean distance of the REZ to pons was 2.6 mm
(range 2.2–3.57 mm) [102]. On the other hand, with no actual measurements, McLaughlin and Jannetta et al., reporting on 4,400 cases of MVD postulated that central myelin could extend along the entire cisternal segment of the trigeminal root [105]. Based on his anatomical studies of six specimens, Ziyal et al. observed that the REZ may not always be sharply demarcated due to small cones of glial cells extending into the peripheral components, similar to “fjords of Scandinavia” [106]. Sindou has suggested that we refer to the REZ as the “transition zone” and call the point where the nerve root reaches the pons the “root entry/exit zone” [107]. This reminds us to be precise in describing the REZ located 2–3 mm from the pons. The length of the CNS segment of other cranial nerves has also been measured. In the study by Tarlov, the length of the CNS segment of the motor part of the facial nerve was 0.8 mm, for the cochlear part of the eighth nerve 8.3 mm, and for the sensory component of the ninth nerve 1.1 mm [95]. Tomii et al. recently reported on measurements of the facial nerve, using various definitions of the transition zone, and concluded that the transition from central to peripheral myelin occurred between 8.0 and 9.9 mm from the edge of the supraolivary fossette [108]. Regeneration. An important feature of the REZ is that regenerating axons after sensory root injury are unable to cross the barrier imposed at the REZ. In an experimental study it was determined that the barrier develops at an embryonic stage earlier in cranial nerve than dorsal root nerve. Once the mature phenotype had been reached, it was no longer possible, after crush injury, for axons regenerating to cross the peripheral-central interface [109]. This supports the clinical observation that functional recovery is unlikely after trigeminal rhizotomy. Mechanical injury to the trigeminal nerve appears to be etiologically related to TN, and it may be worthwhile to consider for a moment the processes of neural regeneration following nerve
The pathophysiology of trigeminal neuralgia
damage. Quoting from a monograph by the famous neuroanatomist, Ramo´n y Cajal, we have a vivid description of these processes: “If it were permissible to explain in terms of human behavior the action of things that are subject only to unconscious forces, we might compare the growth and progress of sprouts toward the boundary. . .to a race by night in which the runners were obliged to run across trackless fields strewn with many obstacles. The strongest among them, that is the winners, rapidly orient themselves, run without vacillation or flagging, and reach the goal. Others, less well endowed, vacillate, stop and turn, and lose themselves in the darkness. While some, the weakest, lacking in vigor and vitality, are arrested by the obstacles or laboriously return to the start” [110]. Today, studies of molecular factors that promote or inhibit nerve regrowth such as laminin, chondroitin sulphate proteoglycans, and other nerve growth factors may lead to new and useful measures to facilitate recovery of function and recovery from disease in the nervous system.
Vascular Anatomy and the Trigeminal Nerve Intracranial vascular anatomy in relation to the trigeminal nerve is an important topic for an examination of the pathophysiology of TN. Based on a number of anatomical and surgical observations, many clinicians and investigators have concluded that injury to the trigeminal nerve root in the posterior fossa by neurovascular compression can cause the painful syndrome of TN. Mechanisms involved in pain production are still being explored. The road to discovery of vascular root compression as a suggested cause of TN began with observations by Dandy [4] when he found indentations of the rootlets by blood vessels during surgery to perform posterior fossa rhizotomy for neuralgia [111]. Shortly afterward, he reported
140
on 215 posterior fossa rhizotomy cases. In 16, the superior cerebellar artery, or a loop of it, in some way affected the nerve, passing beneath or on the lateral surface of the root, indenting the nerve with advancing age. In another 30 cases a branch of the petrosal vein crossed or passed directly through the nerve. In this series, other abnormalities included vertebrobasilar ectasia (“aneurysm”), cavernous angioma, adhesions, and congenital skull base abnormalities. In 87 cases (40%) no gross finding was noted [21]. Later observations about trigeminal neurovascular compression were mentioned by Olivecrona [112] and Lazorthes et al. [113]. An anatomical study of neurovascular relationships and anomalies describing arteries and veins around the trigeminal nerve in cadaver specimens was presented by Sunderland [114]. The first report of vascular decompression for TN was by Gardner and Miklos [115]. In one patient with recurrent TN after a middle fossa decompression procedure, they explored the root in the posterior fossa and found an anomalous arterial loop lying against the nerve. The vessel was separated from the root by the interposition of a piece of absorbable gelatin sponge and the patient’s pain was relieved [115]. In a later paper, Gardner reported on exposure of the cerebellopontine angle in 18 patients. In six of them there was an arterial loop compressing, encircling, or transfixing the nerve, and in another six cases no explanation for the pain was found [116]. Jannetta and Rand [97] used a binocular operating microscope to perform a transtentorial subtemporal retrogasserian neurectomy for TN and noted small tortuous arteries compressing the root [97]. The next year Jannetta, reporting on five patients with TN, through the transtentorial approach found mildly to severely compressed nerves by one or more small tortuous arteries barely visible to the naked eye. He freed the artery from arachnoid membrane allowing it to assume a new position away from the nerve and then proceeded to carry out partial or total
2377
2378
140
The pathophysiology of trigeminal neuralgia
resection of the portio major (sensory root). In the same report, he described an anatomical study in 56 fresh cadavers and in none of these was there any evidence of nerve compression by a blood vessel at or near the pons. Based on these observations, he suggested that TN may eventually prove to be a “nerve entrapment syndrome” [9]. Since then, he and other surgeons have performed MVD procedures through a posterior fossa approach. By 1981 he had explored 411 patients and found arterial, venous, and mixed vascular compression at the REZ in 96% of his cases [117]. Jannetta has emphasized that the site of neurovascular compression causing TN is at the REZ. In addition, he reports that first division pain (V1) is caused by compression of the inferolateral aspect of the root, whereas lower facial pain (V3, or V2 and V3) is caused by compression of the rostral aspect of the root. He found that isolated V2 neuralgia was most commonly caused by compression of the lateral side of the root [118]. This finding may not be in accord with the somatotopic studies of the root at the pons by Gudmundsson et al., which showed that third division fibers are located in the caudal-lateral part of the root [99]. Other surgical observers have found that the site of vascular compression can be far from the pons near the lateral boundary of the root in the posterior fossa [119]. As described earlier, anatomical studies have shown the trigeminal REZ to be within 2–3 mm of the pons. Without anatomic corroboration, Jannetta has commented that the cone of central myelin may reach as far as Meckel’s cave [118]. Subsequently, in a paper by Jannetta and coworkers describing 4,400 MVD procedures, the assertion was made that the root entry or exit zone could vary in length from the brainstem to Meckel’s cave [105]. Anatomical Studies and Surgical Observation. In light of the surgical findings of the intimate neurovascular relationships in patients with TN, several investigators have sought to determine the naturally found vascular relationships with
the trigeminal nerve in cadaver studies of individuals with no history of facial pain. Hardy and Rhoton studied 50 trigeminal nerves in 25 cadavers and found arterial contact with 29 nerves (one with two contacts) at a point between the pons and entry into Meckel’s cave. Of these contacts, 26 were with a caudal loop of the superior cerebellar artery, and four with the anterior inferior cerebellar artery. In ten cases, contact was seen on both sides. The diameter of the superior cerebellar artery or its branch making contact varied from 0.5 to 2.6 mm (average 1.4 mm). There were no contacts in 60% of the nerves studied, in contrast to the studies by Jannetta, mentioned earlier, where no vascular contact was found in 56 cadavers examined. The point of contact of the superior cerebellar artery was usually on the superior aspect of the nerve and a few fascicles of the nerve were often distorted or indented. On average, the site of contact was 3.7 mm from the pons, and in six cases the contact was at the REZ. Referring to a comment by Dandy “just as one sees many cases of gallstones without pain, so one sees lesions attacking the sensory root in the angle productive of pain, but when the patient has pain and gallstones are present, the gallstones are unquestionably the cause” [21]. Hardy and Rhoton posit an equally reasonable conclusion in that the high incidence of neurovascular contacts in their specimens from a “non-tic” population may mean that the surgical findings of such neurovascular contact is coincidental. They concluded that the relief of pain in these cases must be achieved by some mechanism other than by the simple relief from vascular decompression [120,121]. Others have performed cadaver studies for neurovascular contact, including nerve compression and distortion. In 40 cadaver nerves in individuals with no pain history, Haines et al. found vascular contact or compression in 35%, in contrast to 40 surgical cases where vascular contact or compression was present in 85% [122]. In another autopsy study of 60 nerves, 13% were
The pathophysiology of trigeminal neuralgia
found to have arterial contact [123], and in 130 nerves of subjects without pain history, vascular contact or compression including branches of the basilar artery was found in 40% [124]. Nerve compression or distortion by veins, usually the superior petrosal vein, has been noted numerous times during surgery by Rhoton [121]. Neurovascular anatomic studies have raised questions about definitions of contact, compression, deformation, indentation, angulation, stretching, and flattening used to describe the appearance of the involved nerve root. In postmortem studies, issues have been raised about artifact due to shrinkage, en bloc specimen removal, and failure to accurately simulate conditions seen at surgery. In surgical cases, an issue had been raised about artifact caused by patient positioning and how this may affect neurovascular relations observed intra-operatively. In an attempt to address these issues, Hamlyn and King have tried to simulate operative conditions in 30 cadaver dissections. In 13% a vessel came in contact with a nerve, but no case of compression by their definition was found. However, in their 40 surgical cases, vascular contact was noted in 90%, and it was located at the REZ in 68% [125]. Reports from numerous series of surgical cases that underwent posterior fossa exploration for MVD for TN have described the frequency of neurovascular contacts observed. This has ranged from 96% according to Jannetta [126] to 11% according to Adams [127]. In a report by Piatt and Wilkins, in 140 cases with tic douloureux, arterial contact was present in 70% and distortion by a vein or venous angioma in 7% [128]. Like many others, when they found no vascular abnormality, they customarily performed a partial sensory rhizotomy. In a later paper, Young and Wilkins reviewed 83 patients in whom they performed a partial sensory rhizotomy; in 77% of these cases there was no evidence of vascular contact at operation [129]. Adams et al. at Oxford described 57 patients with
140
TN who underwent posterior fossa exploration. A vascular abnormality was found in only 11%. In the remaining cases a partial or total sensory rhizotomy was performed [127]. Careful reporting of outcomes of surgical procedures is very important. In the case of posterior fossa exploration for TN, it is still essential that reports clearly distinguish those cases with “pure” MVD from those in which another procedure was performed, such as partial or total sensory rhizotomy. A wealth of publications attest to the effectiveness of MVD in relieving pain in patients with TN, including also failure, recurrence, and major and minor complication rates. About 10 years ago it was estimated as many as 50,000 operative attempts had been made to mobilize an offending vessel away from the trigeminal nerve. In some series more than 90% of patients were relieved of pain [130]. Sindou remains a major contributor to our knowledge of many important aspects of MVD in patients with TN. He has proposed the term neurovascular conflict (“conflit”) for this anatomical finding. In an effort to reduce the likelihood of subsequent nerve trauma, after dislodging the vessel from the nerve in a series of 60 patients, he did not interpose any foreign material between the nerve and transposed artery. Instead, he used strings of Teflon, or a piece of Dacron, to keep the offending vessel separated from the nerve without touching the nerve. Comparing the results in this group with an “older” group of 60 patients who were treated with interposing foreign material between the nerve and vessel, pain was relieved without hypesthesia in 75% of the interposition group and in 83% of the group with no interposition. He concluded that MVD does not act by “neocompression” of the nerve but by a real decompressive mechanism [131]. In addition to his extensive experience and observations performing MVD, his large series of percutaneous thermorhizotomy for TN also has continued to grow. Reporting on 579 patients
2379
2380
140
The pathophysiology of trigeminal neuralgia
with idiopathic TN in whom he performed MVD, he made several important anatomical observations. No neurovascular conflict was found in only 3.3% of the cases, and in this group the nerve was usually globally atrophic. In the remaining 560 patients one or more conflicting vessels were found: superior cerebellar artery (alone or in association with other conflicting vessels) in 88% of the cases, anterior inferior cerebellar artery in 25% of them, basilar artery in 4%, and a vein embedded in the nerve in 28%. Several conflicting vessels were found in association in 38% of the cases. His thoughtful analysis showed the degree of severity of the conflict to consist of 18% of cases where the vessel was in contact, 49% where the vessel displaced and/or distorted the root, and 33% in which there was clear cut and marked indentation of the nerve root. Equally notable was the site of conflict along the root: 52% at the REZ (defined as less than 7 mm from entry to pons), 54% at mid-third (mid-cisternal segment), and 10% at the exit of the root from Meckel’s cave (juxta-petrous segment). He further noted the conflict site around the circumference of the root, and it correlated with the topography of pain. The conflict was supero-medial to the root in 54%, supero-lateral in 32%, and inferior in almost 15% of cases. Other observations included significant atrophy of the whole root (caliber reduced by one-third or more) in 47% of cases, and in another 18%, atrophy was more marked with ribbon shape and grayish color. Some of the patients had a previous failed thermorhizotomy. There was arachnoidal thickening with strong adhesions to the nerve in 18%, a shallow cisternal space (small posterior fossa) in 4%, and marked angulation of root crossing the petrous ridge in 13%. Such an angulation frequently coexisted with an elongated superior cerebellar artery in a superior position strongly pushing down the nerve, making it flat or even hammock-shaped and markedly atrophic. After reviewing 11 large series of MVD that reported
negative explorations ranging from 0% [132] to 28.5% [133], he felt his experience with 3.3% (19 cases) having no vascular conflict was not unusual. In contrast, he made another observation in 54 patients who had facial malignancies. When treated by juxtapontine sensory root section, no significant vascular contact was found. He concluded that his findings support a role of neurovascular conflict in the genesis of idiopathic TN, although it did not explain all of the pathogenesis of idiopathic TN [119]. Additional information of interest comes from a more recent study by Sindou et al. of the long-term effectiveness and prognostic factors in TN in 362 consecutive patients with clear cut neurovascular conflicts who underwent “pure” MVD with no lesioning of adjacent rootlets. Using statistical methods, he found that pure MVD in his series resulted in a 73.38% probability of long-term (15 years) cure of neuralgia. He noted that failure rate of 15–35% after MVD has been reported in the literature. Based on the classical text by his mentor, William Sweet, he classified his cases as typical, with purely paroxysmal pain, or atypical, in the sense that the paroxysmal pain (at least at the beginning) later had paroxysmal pain superimposed on a steady permanent deep aching or burning pain. He found the outcome after pure MVD to be the same in both the typical and atypical groups. None of the following patient-related factors played any significant role in the prognosis: sex, age, history of hypertension, duration of neuralgia before surgery, or history of failed trigeminal surgery (unless the patient had neuropathic pain following a lesioning procedure). Marked crosscompression of the root at surgery correlated with a high percentage of cure, but a low degree of compression or focal arachnoiditis had a negative effect on outcome as did involvement of all three divisions of the nerve. Neither the type of compressive vessel, nor its location along or around the root were prognostically significant [134]. In addition to his outstanding clinical
The pathophysiology of trigeminal neuralgia
accomplishments Sindou’s meticulous and salient observations described here are very valuable to our understanding and treatment of TN. Imaging studies. Current advances in magnetic resonance imaging (MRI) technology are providing newer methods to accurately assess trigeminal neurovascular relationships in individuals with no pain, as well as those with TN. These new imaging techniques are already complementing what we know about vascular anatomy from cadaver studies and intraoperative observations. They permit in vivo evaluation, circumventing the possible artifacts of postmortem studies. More important, they should provide better preoperative assessment of patients prior to undergoing MVD. Meaney et al. [135], using magnetic resonance tomographic angiography found agreement in 50 of 52 surgical explorations for TN [135]. Jawahar and his group in Pittsburgh, using contrast enhanced 1.5 T MRI, studied 275 patients for gamma knife radiosurgery. In 97 patients with no previous surgery they found that 67% had evidence of neurovascular compression. In the total series, 88 had undergone previous MVD and 21 (24%) had evidence of pontine infarction they attributed to sequelae of small venules coagulated during exposure. Eleven of the patients with pontine infarcts had facial sensory loss [136]. Using other advanced MRI techniques, more recent studies have revealed details of sensory and motor root neurovascular relationships [137–140]. Perhaps the most useful recent advance has been the combination of two MRI techniques: three-dimensional Fourier transform constructive interference in a steady state (CISS) imaging and complementary time-of-flight (TOF) angiography sequences. Using these two techniques on a 1.5 T unit in 60 cases with no history of facial pain, contact between nerve and artery was found in 25% of nerves [141]. A particularly significant imaging study was recently performed by Peker (in press) and his
140
colleagues in Istanbul. They used two types of MRI sequences: CISS and TOF with FLASH (“fast, low-angle shot”) on a higher fieldstrength 3.0 T magnet. This group examined 100 subjects with no history of TN. They found evidence of neurovascular compression (defined as no cerebrospinal fluid between nerve and vessel on MRI) in 92 of the 100 individuals. This was bilateral in 83 subjects and unilateral in nine. A total of 87.5% of 200 nerves examined showed neurovascular compression. In 50% of the cases the nerve was compressed at a site in the proximal third of the nerve. The compressing vessels were arteries in 86% and veins in 14%. They noted that the frequency of vascular compression of the trigeminal nerve was similar to rates that had been reported in large series in which MVD had been performed for TN. They concluded that their results strongly suggest that an MRI finding of a vessel compressing the trigeminal nerve is not necessarily pathological [142]. Given the frequent pain relief after MVD, there is widespread opinion that neurovascular conflict involving the trigeminal roots is a causative factor in the pain of TN. It has been noted that the degree of compression at the site of conflict along the nerve root varies substantially. Apfelbaum noted that the site of the pathologic condition is at the brainstem [143], while others, such as Sindou et al., have documented sites of compression along the entire length of the nerve up to Meckel’s cave [119]. Further confounding this issue are the numerous anatomic and imaging studies demonstrating frequent neurovascular contacts in individuals who have no history of TN. Equally important is the notable observation that in a substantial number of cases with typical TN, no vascular abnormality or other cause for the pain has been found, even at surgery. These cases, therefore, remain conceptually classified for the present as idiopathic TN. Adams has provided a thoughtful analysis of the controversies surrounding neurovascular compression as a cause for TN, and MVD as a
2381
2382
140
The pathophysiology of trigeminal neuralgia
cure for it [144]. Morley believed that a single case where a vessel did not touch the nerve would suffice to discredit the hypothesis of vascular compression as a cause for TN [145]. Adams considered this opinion too harsh. At the time, he described five surgical series, reported in the literature, with some cases where no vascular compression was found during surgery for MVD. In his own series, 51 out of 57 cases for MVD had no vascular compression [127]. Apfelbaum reported ten of 289 cases with no vascular compression [146], Zorman and Wilson reported 26 of 125 cases [147], Burchiel four of 41 cases [148], and Bederson and Wilson 30 of 252 cases without vascular compression [149]. As previously noted, Sindou et al. (2002) reported 11 large series with negative explorations at surgery ranging from 0 to 28.5%, not including Adams’s series [119]. Among the issues raised by Adams were how to reconcile the characteristic pain remissions in many patients with MVD and the observation, after successful treatment with MVD, that some had recurrences of pain. He is not alone in proposing that the efficacy in relieving pain by MVD may be due to some degree of trauma to the trigeminal root during surgery, recalling that other surgical procedures for TN have been performed with success by mechanically traumatizing or compressing various regions of the trigeminal nerve [150].
Functional Anatomy of the Trigeminal System Somatotopic organization. Studies of the trigeminal system have revealed that there is a general somatotopic organization from peripheral branches to the brainstem. There is no definite spatial separation within the nerve of fibers subserving different sensory modalities such as touch, pain, thermal sensation, mechanical pressure, or proprioceptive functions. Axons constituting the ophthalmic, maxillary, and mandibular branches
converge into their cell bodies in the Gasserian ganglion. Darian-Smith et al. found topographic arrangements when they recorded single neurons in the cat semilunar (Gasserian) ganglion. Cell bodies in the distribution of cutaneous receptive fields in the ophthalmic division were found anteromedially, those in the mandibular division were found posterolaterally, and those giving origin to the maxillary division were located between the other two in the ganglion [151]. Lende and Poulos [152] recorded extracellular unit responses in monkeys and found functional clusters of cell bodies with similar peripheral fields. This finding suggests that the ganglion is organized according to anatomical convenience rather than sensory modality. They noted that their observations did not support Dandy’s hypothesis that a functional reorganization of the trigeminal nerve occurred within the ganglion [4], but agreed with the proposal by Spiller and Frazier that the ganglion did not alter the anatomical distribution maintained by the three peripheral divisions throughout the ganglion and root [153]. Using horseradish peroxidase retrograde tracer technique in cats, Marfurt also demonstrated somatotopic organization in the dorsoventral axis of the ganglion. Corneal neurons were uniformly distributed in this axis with segregation of ophthalmic and maxillary divisions in overlap with maxillary and mandibular divisions [154]. The trigeminal sensory roots maintain a somatotopic organization as the component axons extend from the cell bodies in the Gasserian ganglion to their destination in the brainstem. This organization has been confirmed by anatomical and physiological studies as well as by clinical observations following surgical rhizotomies. A naming system has been developed for the components of the trigeminal roots in the posterior fossa as they approach the brainstem. There is a large compact bundle of most of the sensory roots called portio major or pars compacta. Nearby is the portio minor, several motor rootlets that consist of a mixture of efferent motor fibers and
The pathophysiology of trigeminal neuralgia
afferent sensory fibers. Between the main sensory root and the motor roots are several very small fascicles of fibers called “accessory roots” by Dandy [96], “intermediate fibers” by Jannetta and Rand [97], or “aberrant sensory rootlets” by Gudmundsson et al. [99]. Important early clinical observations after partial sensory trigeminal rhizotomy were made by Frazier, [155] and Spiller and Frazier [153], who found consistent anatomical arrangement in accord with their peripheral branches in the middle fossa rootlets after selective root sectioning. They believed the organization persisted within the roots up to the pons. On the other hand, using a posterior fossa approach, Dandy performed partial sectioning of the main sensory root near the pons. He concluded that there was a separation of function within the somatotopic organization of the root, because his partial rhizotomy appeared to abolish pain of TN with relative preservation of tactile sensation in the face [4,96]. Addressing the dichotomy of findings after surgery, Davis and Haven undertook an anatomical and physiological study in cats and humans and found a clear topographic anatomy of the sensory root from ganglion to entrance to pons. They observed numerous anastomoses between sensory root fibers near the hilus of the ganglion and definite rotation of the root between ganglion and pons. Experimental section of the ophthalmic division of the ganglion and inferior portion of the sensory root resulted in identical microscopic pictures of degeneration in the pons, and section in both instances was followed by loss of corneal sensation and keratitis. They were unable to find a separation of types of sensation in the roots [156]. Dandy had reported that sectioning the inferolateral portion of the root could cure pain in all three divisions. These findings were presented at a meeting in Boston, and in the published discussion that followed, Cushing commented “it is very extraordinary that the subjective sense of pain should be abolished by incomplete division of
140
the root near the pons and that complete preservation of sensitivity to pain should at the same time remain in the trigeminal skin field” [4]. Using anatomical degeneration techniques in cats and monkeys, Kerr found sharp lamination of fibers from each division in agreement with the findings of others. He also noted that Dandy’s use of the term “posterior part of the root” that he sectioned was confusing. The confusion arose because he referred to the part which first comes into view during his approach, which anatomically corresponds to the caudolateral part of the sensory root [157]. Additional studies have provided clear evidence of somatotopic organization in the sensory root, and no evidence has been found for Dandy’s original postulation regarding functional subdivisions with reference to separation of sensory modalities. As the number of patients with TN undergoing posterior fossa exploration for either MVD or partial rhizotomy grows, it becomes increasingly important to understand somatotopic organization of the sensory root. This is necessary to account for observed clinical results in terms of pain relief and sensory loss after rhizotomy, and to assess the significance of the site of vascular compression around the root as it relates to the specific divisions of pain involvement in the face. An elaborate autopsy study of the functional anatomy of 50 trigeminal roots in 25 adults was performed by Gudmundsson et al. [99]. They found that the cross section of the sensory root was elliptical between pons and petrous apex, with the greatest diameter of the ellipse at the pons measuring 3–4.5 mm. There was variability in the degree of rotation at the pons. The angle between the longest diameter of the cross section and the long axis of the body was 40–50 but could vary from 10 to 80 . An angle of 80 placed the third division fibers almost directly lateral to those of the first division, but an angle of 10 placed the third division fibers almost directly caudal to those of the first division. In the customary posterior fossa approach, the
2383
2384
140
The pathophysiology of trigeminal neuralgia
fifth nerve is approached caudally and laterally. With the most frequent rotation pattern, the surgeon would have the third division fibers nearest to him. This was consistent with Dandy’s observation that sensory loss was most pronounced in the third division when he cut through the caudal part of the nerve. In other cases, he found such a great variability in the difference in both quantity and quality of retained sensation “that one’s credibility might be tested” [158]. “Motor and accessory roots.” Gudmundsson et al. also examined the trigeminal motor root and found that it was composed of 4–14 filaments measuring 0.1–1 mm in diameter, most arising rostral to the sensory root. There were frequent anastomoses between the motor and sensory roots. In the interval between trigeminal sensory and motor roots, there were as many as 15 nerve rootlets that arose at the pons outside the main sensory cone. These were present in only 50% of nerves studied, and they called them “aberrant” or accessory sensory fibers. Some of these aberrant sensory roots joined the motor roots distally, and others joined the sensory root mostly in its ophthalmic portion [99]. Dandy used the term “accessory” sensory branches for those fibers that accompanied the motor root and later joined the sensory root. After complete sensory root section at the pons, he attributed the preservation of varying degrees of sensation in the face to the accessory fibers [96]. Using the operating microscope Jannetta and Rand observed fibers they called “intermediate roots” that apparently corresponded to the accessory roots seen by Dandy [97]. In another study of root anatomy, Saunders and Sachs [159] examined right and left fifth nerves in 20 autopsies. All specimens had 1 mm clusters of small rootlets rostral to the sensory root. These rootlets always arose in two separate groups that joined to form one root within 1 cm of the pons. They labeled these two rootlet groups “superior” and “inferior” rootlets.
The superior group, dorsal to the sensory root, consisted of three to six rootlets representing the classical origin of the motor root. In several specimens the fifth nerve was represented by a single extremely flattened root. Only by identifying the distal motor root and tracing it back to the pons could the motor root be definitely delineated from the remainder of the fifth nerve. In this case the motor root was always at the rostral edge of the conglomerate fifth nerve. There were more rootlets, often as many as ten, in the inferior motor rootlets ventral to the sensory root. The superior and inferior rootlet group conjoined within 1 cm of the pons and passed to the floor of Meckel’s cave beneath the Gasserian ganglion, and then to the foramen ovale. There were a few inconstant anastomoses between the conjoined motor roots and the main sensory root, and occasionally the motor root actually passed through the substance of the ganglion. These investigators concluded that the inferior group of motor rootlets corresponded to the “accessory” fibers described by Dandy. They gave tribute to Johannes Friedrich Meckel (1748) for his discovery of the same findings when he described the motor root arising from the pons as two separate, inferior and superior, groups of rootlets [159]. Pelletier et al. made a significant contribution to our knowledge of somatotopic and functional localization in the trigeminal root. Their studies involved a combination of electrophysiological recording and stimulation in macaques, evaluation of deficits in humans after posterior fossa root section, and microdissection in humans and macaques. They found that the roots in humans and monkeys were anatomically similar. Using sensory root recordings, they observed that the sensory root contained three spatially distinguishable but overlapping divisions which corresponded to the three peripheral divisions. Upon dissection, the motor root consisted of two bundles of nearly equal size, dorsal and ventral in relation to each other, and rostral to the
The pathophysiology of trigeminal neuralgia
sensory root at the pons. The bundles quickly joined to form a single motor root, curving around the sensory root, on their way to the ganglion. Between the motor and sensory roots there were a variable number (2–14) of smaller bundles entering the pons individually. Nearly all of these roots joined the motor root as it coursed beneath the root-ganglion complex. Occasionally, fibers crossed between sensory and motor roots or left the sensory root near the REZ to enter the pons separately. These “accessory” fibers were found to contain the same sensory and motor elements as the motor root which they joined. From motor root stimulation and recording they found that stimulation invariably resulted in jaw movement. Even the smallest branches that could be separated by microdissection contained not only motor functions but also sensory functions, activated by jaw opening or stretch of the appropriate muscle groups. All accessory fibers contained similar functions, and none exhibited other types of sensory activity other than proprioception. Their study also included evaluation of sensory deficits in ten patients who had undergone posterior fossa partial root sections. Patients were tested 1–11 years after surgery. In all patients the neurological deficits had remained stationary, and none had a recurrence of TN. The surgeon estimated three-fourths to nine-tenths of the root had been sectioned. Two of the ten patients had total sensory deficit throughout the trigeminal distribution with intact motor function. The other eight cases exhibited a wide spectrum of sensory loss to pinprick, light touch, temperature, two-point discrimination, and corneal reflexes. No patient exhibited a decrease or loss of any single sensory modality without a concomitant decrease in the other sensory modalities. In cases when an estimated nine-tenths of the root was sectioned, the sensory deficit was primarily in the second and third divisions. They noted, considering the estimated magnitude of root section, that these patients showed a
140
remarkable preservation of sensory function [160]. The evidence presented by these investigators does not support the earlier theories by Dandy [96], and Jannetta and Rand [97] of functional localization in the sensory root. Dandy had stated that partial root section was similar to a cordotomy in that only pain fibers were sacrificed and all other forms of sensation were retained. Pelletier et al. [160] made other valuable observations concerning clinical findings of relative preservation of sensory function after partial sensory rhizotomy. They suggested that if there was any reorganization with segregation of sensory modalities, it could happen in the subpial entry zone within the pons, since it did not occur in the root itself. These findings that the motor roots and accessory fibers were functionally the same, and that the only sensory afferents intermingling with the motor efferent fibers were exclusively proprioceptive, led them to two conclusions. First, these fibers played no role in the origin of pain in TN, and second, they were not involved in the occasional recurrence of pain after partial sensory rhizotomy. In their paper, they presented a personal communication in 1971 with an associate of Dandy’s, Dr. Frank Otenasak: “Although Dandy originally believed that the posterior fibers of the trigeminal root were those that carried pain sensation, he found, with increasing experience, that sectioning of this portion of the root did not always relieve pain. In fact, on one occasion in which I assisted him personally, he had left one little strand of sensory root, not more than 1–2 mm in diameter, and when the patient woke up she still has her old tic. The next day we took her back to the operating room and sectioned the remainder of the root. Following this he always did a total section and although he did not further comment upon his original ideas over the distribution of the pain fibers, his ideas, in fact changed.” Even after presumed total root section, Dandy attributed some preservation of sensation to the function
2385
2386
140
The pathophysiology of trigeminal neuralgia
of “accessory” fibers. Pelletier agreed with Davis and Haven [156], and Stookey and Ransohoff [1] who supposed that Dandy had performed subtotal rather than total root section in his cases with retained sensation. Pelletier et al. also provided a very plausible explanation for significant preservation of sensation after substantial (perhaps one-half or more) partial rhizotomy. They state that there appears to be sufficient reserve trigeminal innervation so that a considerable proportion of fibers, probably 50% or more, may be destroyed before symptoms of sensory loss appear, provided the fibers are decimated evenly and uniformly so that no single peripherally represented area is critically denervated. Based on this concept, the extent to which each division is interrupted would depend on the angle of approach, degree of rotation of the root, and depth of the cut. Other important findings about the functional anatomy of the trigeminal sensory root have been reported. Young and King [161] performed electron microscopic examinations of baboon trigeminal sensory roots to determine the ratio of myelinated and unmyelinated fibers in the trigeminal sensory root compared to fiber spectrum analysis of spinal dorsal roots in some vertebrates. They found that myelinated fibers outnumbered unmyelinated fibers in the trigeminal root by a ratio of 1.2:1. Their finding that 40% of trigeminal sensory root fibers were unmyelinated contrasted with findings of 80% unmyelinated fibers in spinal dorsal roots [161]. Studies of the organization of the ophthalmic division fibers have been important particularly for surgeons attempting to preserve corneal sensation during ablative procedures for TN. Wilkins and Sachs [162] reported on 26 cases of middle fossa root sectioning and concluded there was some interlacing between the ophthalmic division and the second and third divisions. They found the ophthalmic division was frequently not a discrete bundle, leading to unexpected sensory losses after surgery [162].
More recently, nerve supply to the cornea in cats was studied using horseradish peroxidase retrograde degeneration techniques. It was found that a majority of neurons innervating the cornea were located in the ophthalmic division, but 4% were in the maxillary division. Highly labeled axons were found in the corresponding first and second divisions of the root near the brainstem [163]. Stechison et al. performed intraoperative recordings in 15 patients undergoing MVD or partial rhizotomy by recording antidromic responses for mapping the root. Their results supported the previously known anatomic evidence that ophthalmic fibers were rostral and mandibular fibers caudal in the root. The authors noted that these findings were consistent with previously established somatotopic root organizations as reported by Rhoton and his group [99,121] but were incongruous with the findings of Jannetta [118] that suggested that rostral root compression produced lower facial pain and caudal root compression produced upper facial pain. They proposed a possible explanation for this incongruity could be that the vascular compression may not produce symptoms in the portion of the nerve in direct contact with the vessel, but, rather, may produce pain from a tenting-up phenomenon or distortion that may produce maximum dysfunction in the axons on the opposite surface of the nerve [164]. Still unresolved is the functional significance of sensory afferent fibers admixed with motor efferent fibers in the motor root (portio minor) of the trigeminal nerve. The excellent studies by Pelletier et al. showed that “accessory” and motor roots were functionally similar and their afferent component contained only modality-specific proprioceptive fibers from the muscles of mastication [160]. However, several experimental and clinical observations have suggested additional functional roles of the afferent fibers in the motor root. Failure to relieve trigeminal pain, or recurrence of TN after “total” sensory rhizotomy sparing motor function, has lead
The pathophysiology of trigeminal neuralgia
clinical observers to suspect that sensory fibers associated with the motor root may be involved in TN. Saunders and Sachs reported on 45 cases treated by retrogasserian neurectomy (Frazier procedure). There were nine recurrences in 15 patients when motor function was preserved and only six recurrences in 30 patients when motor function was lost after surgery [159]. According to the “Law of Bell and Magendie,” it was long accepted that spinal dorsal roots contained sensory axons and ventral roots contained motor axons. In an optical and electron microscopy study of spinal ventral roots in humans, Coggeshall et al. found large numbers of unmyelinated axons making up 27% of the total population of ventral root axons [165]. Young performed electron microscopic studies of five human trigeminal motor roots and found that 12–20% of the total fibers were unmyelinated. They concluded that these were sensory afferents except for possibly a few gamma motor neurons, and that they were not cranial visceral efferents that are confined to the third, seventh, ninth, and tenth cranial nerves [72]. In further studies in humans and cats, Young and Stevens concluded that the source of motor root unmyelinated fibers was unknown but might arise from trigeminal ganglion cells, even though their functional significance was also unknown. They credited the success of the technique of radiofrequency thermorhizotomy developed by Sweet, because it was directed more specifically at the fiber group functionally related to pain perception without regard to their mode of passage between ganglion and brainstem. They suggested that efforts to relieve pain by mechanical rhizotomy of the portio major sparing the portio minor would continue to result in relatively high failure rate because of the location of unmyelinated fibers in the trigeminal “motor root” [166]. To elaborate further on the anatomical significance of afferent fibers in the motor root, Young and Krueger employed neuroanatomical axon transport methods to trace connections of these
140
fibers in cats. Using tritiated amino acids for anterograde studies, they injected the trigeminal motor nucleus, whose fibers were traced solely into the portio minor. Similar injections into the ganglion showed, as expected, that the vast majority emitted axons to the portio major. For retrograde tracing, horseradish peroxidase labeling of distal cutaneous branches showed heavy labeling in the portio major, whereas the portio minor had only a small proportion of the total axon population labeled. Concomitant extensive labeling of the ganglion cells was observed. Their experiment did not confirm a cutaneous origin for the afferent fibers in the motor roots. In light of other studies showing proprioceptive muscle afferent fibers in the motor root, they concluded that the functional relationship of this with the mesencephalic nucleus was unknown [167]. Mesencephalic nucleus. The mesencephalic nucleus is unique in the trigeminal sensory system. It is composed of primary sensory neuron cell bodies grouped within the mid brain, rather than in the Gasserian ganglion, where all the other primary sensory cell bodies are located outside the brain. Studies have shown that the neurons of the mesencephalic nucleus subserve the sensory modality of proprioception in the innervated muscles of mastication (and possibly the ocular motor muscles). May and Horsley [168] were among the first to study the anatomy of this nucleus. Their experiments in cats and monkeys showed extensive chromatolysis in the neuronal cell bodies of the mesencephalic nucleus after lesions of the tract anterior to the trigeminal motor nucleus and after sectioning of the root between the ganglion and pons. Sectioning of the peripheral first and middle trigeminal divisions produced no chromatolysis, whereas section of the inferior division caused extensive chromatolysis in the mesencephalic nucleus [168]. Electrophysiological studies of the mesencephalic nucleus in cats demonstrated action potentials characteristic of proprioceptive impulses in response to jaw opening, and thence,
2387
2388
140
The pathophysiology of trigeminal neuralgia
stretch of the masticator muscles. In addition, action potentials were recorded in the caudal half of the mesencephalic nucleus after blunt pressure stimulation of the homolateral teeth and hard palate [169]. Ramo´n y Cajal had demonstrated axon collaterals from the mesencephalic nucleus to the motor nucleus provided a monosynaptic relay for the jaw jerk reflex. The functional role of the afferent fibers in the trigeminal motor roots and their relationship with the mesencephalic nucleus has not been fully established. Brainstem complex. All the trigeminal primary sensory axons project from the Gasserian ganglion, enter the brainstem at the mid-pontine level, and form specific tracts to make synaptic connections with brainstem nuclei composed of trigeminal second-order neurons. As the entering axons traverse the lateral pons, they either form tracts that lead to the main sensory nucleus in the pons, or they descend downward as the spinal tract to connect with the trigeminal nuclear complex, extending from pons caudally through the medulla as far as the C2–3 level of the cervical spinal cord. This assembly of tracts and nuclear masses of cells is called the trigeminal brainstem complex. The nuclear masses are formed by three cytoarchitectonically different nuclear groups, in descending order: nucleus oralis, nucleus interpolaris, and nucleus caudalis. Olszewski has described the cytoarchitecture and found that only the nucleus caudalis had the same fundamental cell arrangement as the posterior horn of the spinal cord. The nucleus caudalis consists of subnucleus zonalis (marginal zone) adjacent to the tract, subnucleus gelatinosa (substantia gelatinosa) with neurons of various sizes, and, medially, the subnucleus magnocellularis (nucleus proprius). At the level of the obex, the nucleus caudalis changes to form the nucleus interpolaris (> Figure 140‐4) [87]. Upon entering the pons, sensory root fibers of both large and small caliber were found by Windle [86] to consist of bifurcating and
non-bifurcating axons. Bifurcating axons proceeded directly to the main sensory nucleus and their other branch turned caudally to form the descending tract. Some fine caliber fibers bifurcated, but the majority did not bifurcate and made a downward bend to continue as the descending tract. A few fibers of large caliber turned upward into the main sensory nucleus giving off no descending branch. Bifurcating fibers of various calibers ascended to the main sensory nucleus and descended in the spinal tract to the spinal nuclear complex. He suggested that the course of painful afferent inputs was probably by way of the descending non-bifurcating fibers, and that the course of tactile impulses was probably by way of the ascending non-bifurcating fibers to the main sensory nucleus. In the descending tract, numerous axon collaterals project to all of the nuclei of the trigeminal nuclear complex, and second-order axons interconnect all of the brainstem trigeminal nuclei (> Figure 140‐4) [86]. Several investigators have confirmed a somatotopic organization within the spinal tract, and, similarly, in the main sensory and spinal nuclear complex. There is an upsidedown representation of the face, with fibers of the mandibular division most dorsal and ophthalmic fibers most ventral [88,170,171]. Wall and Taub also found a somatotopic arrangement so that the nucleus oralis, “in a rare coincidence of anatomic nomenclature,” was dominated by innervation of the mouth and tongue consistent with other findings of the onion-peel sensory loss that occurs when caudal lesions denervate the periphery of the face and rostral lesions denervate the central face and oral cavity. From their physiological studies of the brainstem, they concluded that small peripheral fibers failed to extend long distances caudally and that small fibers observed caudally were tapered continuations of large descending fibers (> Figure 140‐4). They found that cells rostral in the main sensory nucleus responded to all types of peripheral fibers converging onto them, while cells in the
The pathophysiology of trigeminal neuralgia
caudal part of the main sensory nucleus responded as if they were innervated mainly by large peripheral fibers. They found no cells in any of the brainstem nuclei that responded to intense skin pressure, leading them to conclude that failure to find “pain cells” in the nuclei was a paradox, because it was known that abolition of the afferent fibers to the nucleus caudalis by tractotomy resulted in facial analgesia in humans and cats. As an explanation for this paradox, they suggested that pain reactions are set off by massive spatial and temporal summation of impulses from primary nuclei on deeper cells rather than hypothetical “pain cells” [88]. The somatotopic upside-down organization in the brainstem was confirmed by Emmons and Rhoton by degeneration studies in monkeys. They showed that the rostral root fibers (predominantly ophthalmic division) projected ventrally and caudal root fibers (mandibular division) projected dorsally in the main sensory and spinal brainstem nuclei [172]. While demonstrating dorsal-ventral laminations in the root, main sensory nuclei, and spinal tract in cats and monkeys, Kerr established that all these divisions end in the same caudal level in the upper half of the C2 cord segment, but the number of ophthalmic division fibers was much less than the maxillary or mandibular divisions [171]. The descending spinal tract and the nucleus caudalis have been a special focus of attention by investigators interested in pain pathways of the trigeminal sensory system. Much of this interest has evolved following Sjo¨qvist’s report (1938) of the first surgical procedure to sever the descending trigeminal spinal tract rostral to the nucleus caudalis. Based on limited prior anatomical studies and observations of facial analgesia after lateral medullary infarct causing Wallenberg syndrome, Sjo¨qvist stated that Professor Olivecrona allowed him to perform this operation for the first time in May 1937. He reported on nine cases and the majority showed postoperative loss or impairment of sensation of pain, heat, and cold
140
with preservation of tactile sensation on the homolateral side of the face. He said the value of this new operation could not be judged at the present time [89]. Kerr’s ultrastructural studies of the nucleus caudalis region revealed small islets of neuropil within the spinal tract occasionally containing cells named interstitial neurons by Ramo´n y Cajal. Synaptic morphology and dendritic spines were examined in all the cytoarchitecturally characteristic cell groups in the nucleus caudalis [173]. Other studies have addressed the secondary efferent fiber projections from the trigeminal nuclear complex. Nauta and Kuypers found bilateral ascending fiber degeneration after lesioning of the descending nuclear complex [174]. In another study of fiber projections, Stewart and King found an ascending trigeminal intranuclear pathway originating in the nucleus caudalis that distributed pre-terminal fibers to all levels of the trigeminal nuclear complex in the medulla and pons, and to other adjacent brainstem nuclei. Other fibers in this pathway ascended to the mesencephalic tract of the trigeminal nerve and entered the central gray and tectum of the ipsilateral midbrain. They suggested these fibers may be part of a neural mechanism subserving the integration of noxious stimuli. Other efferent projections from the nucleus caudalis ascended bilaterally to midbrain and thalamus through reticular tegmentum, and to the midbrain and thalamus in lemniscal pathways. There were other fiber projections to intralaminar regions of thalamus providing anatomic support for the belief that these thalamic nuclei may be part of the neural mechanism subserving integration of noxious stimuli. They proposed that the nucleus caudalis was a primary relay nucleus for certain sensori-motor reflexes based on fiber projections between the nucleus caudalis and the ipsilateral motor nuclei of cranial nerves 12, 11, 10, 9, 7, and 5 [175]. Denny-Brown and Yanagisawa made another contribution to our understanding of the functional anatomy of the trigeminal
2389
2390
140
The pathophysiology of trigeminal neuralgia
brainstem complex. They correlated sensory testing with various lesions of the trigeminal brainstem system in monkeys. They concluded that the descending tract was a mechanism for convergence of stimuli taking origin in areas of overlapping innervation of the surface of the head by the 5th, 7th, and 10th cranial nerves and the 2nd and 3rd cervical sensory roots. An intrinsic system of small cells in the descending tract must receive collaterals from all entering sensory fibers and relay to all other parts of the trigeminal nuclear complex. They noted a ventrolateral inhibitory division and a dorsomedial facilitatory division in the tract. They concluded that the defect in sensation resulting from sectioning of the descending tract was the same that affects the border zone of sensory loss after sectioning of any spinal dorsal root. The sensory deficit results from loss of one pathway when two have normally converged. The outcome is a defect in central spatial summation which is evidently more necessary for the reception of pain than temperature, and of temperature than touch. This deficit is equally apparent in reflex and general reaction to stimuli, and is strong evidence in favor of consideration of pain as a quantitative aspect of common sensation [170]. Central Connections. The central projections from the trigeminal sensory nuclei in the brainstem described above have been elaborated on further by others. Young has reviewed extensively his and other findings related to projections and reflex connections involving the trigeminal system [176]. The main sensory nucleus is connected to the contralateral thalamic nucleus ventralis posteromedialis (VPM) by the trigeminal lemniscus and has an uncrossed connection to the ipsilateral thalamic nucleus VPM via the dorsal bundle of Wallenberg. The trigeminal brainstem nuclear complex is also connected with the thalamus by the trigemino-thalamic pathway lying medial to the spinothalamic tract near the central gray. Other projections from the brainstem complex include the superior
colliculus, cerebellum, and various cranial nerve nuclei involved in reflex response to oral and facial sensation. He also describes other advanced neuroanatomical studies of the trigeminal brainstem system. Horseradish peroxidase studies have revealed connections with thalamic interlaminar nuclei and the periaqueductal gray region of the midbrain. Immunocytochemical labeling showed projections to the medial thalamic nuclei indicating they contained enkephalins. Thalamic inputs to the somatosensory cortex were somatotopically organized, but little is known about cortical inputs related to facial pain. Connections have been demonstrated between the trigeminal brainstem system and the nucleus of the solitary tract, a major recipient of visceral efferent inputs from the vagus nerve, and the hypoglossal nucleus. Additional reflex connections have been suggested when the excitatory neurotransmitter L-glutamate was injected into the nucleus caudalis. This resulted in increased secretion of epinephrine and norepinephrine from the adrenal medulla, ipsilateral to the side of injection, with no such effect from contralateral injection. Interestingly, another study showed that calcitonin gene-related peptide (CGRP) immunoreactivity in the major cerebral arteries of the circle of Willis was derived from trigeminal ganglion cells via the ophthalmic division. When the division was transected, CGRP immunoreactivity of the ipsilateral cerebral arteries was eliminated causing a significant reduction in the diameter of the vessels as well. This suggested that the trigeminal system could be an important modulator of cerebral vascular diameter. Descending inputs into the spinal trigeminal complex have been found from sites such as periaqueductal gray, nucleus raphe magnus, and sensorimotor cortex. Centripetal terminals containing pain modulating chemicals such as serotonin, gamma amino butyric acid, and enkephalin have been demonstrated in the nucleus caudalis and the nucleus interpolaris, which provides further evidence that pain processing
The pathophysiology of trigeminal neuralgia
centers in the spinal trigeminal nuclear complex extend beyond the nucleus caudalis. In addition to this excellent review by Young, another source of further details about functional trigeminal brainstem anatomy has been provided by Shults [177].
Neurophysiology and Trigeminal Neuralgia Clinicians and basic neuroscientists are impressed by the excruciating, unbearable pain suffered by those unfortunate individuals with TN. Unlike other severely painful conditions, TN has a number of distinctive clinical features, namely exceptionally severe lancinating paroxysms of pain on just one side of the face. They are triggered like an electric shock by even the slightest touch in the affected region of the face. These symptoms, and their remarkable spontaneous remissions for months or years, provide a worthy challenge for understanding the pathophysiology of this unique condition. Except for very rare neuralgias of other cranial nerves, there is no painful disease in the nervous system resembling TN. Anatomical studies alone have not provided an answer to the question about etiology or pathogenesis of TN. However, information about structural abnormalities combined with careful clinical observations in patients with TN have special value to neurophysiologists working primarily in the experimental laboratory. From these studies we have new insights and important hypotheses about the pathophysiology of TN. Normal physiology. Fundamentally, the trigeminal system has been found to be very similar to the rest of the somatosensory system in the body. It consists of sensory receptors and axons nourished by their cell bodies that generate and transmit electrical impulses to second-order relay neurons in the brainstem. These, in turn, process and then relay signals to higher centers of
140
the brain for conscious perception of sensation, such as pain. For the trigeminal nerve, sensory receptors for pain are located in facial structures. They are free nerve endings called nociceptors, defined by Sherrington as “sense-organs that respond to agents of such intensity as threatens damage” [178]. Nerve fibers, or axons, are traditionally divided into A, B, and C fibers according to their diameter and physiological properties. A fibers are myelinated and fall into four partly overlapping groups with decreasing diameter: alpha, beta, gamma, and delta. B fibers (not to be confused with A-beta fibers) are lightly myelinated preganglionic fibers found only in the autonomic nervous system. C fibers are all small and unmyelinated. Sensory fibers are classified: A-alpha and A-beta (conduction rate 120–30 M/s), A-delta (conduction rate 30–4 M/s), or C fibers (conduction rate less than 2.5 M/s) [91]. Stimulation of large myelinated A fibers produces the sensation of touch. Stimulation of small A-delta fibers (diameter 3–5 mm) and C fibers (diameter 0.3–1.5 mm) produces prickly or burning pain. Collins et al. found that stimulation of large fibers was not painful, but stimulation of A-delta and C fibers could produce “unbearable” pain [179]. Other studies have shown that A-delta fibers may be classified as mechanical nociceptors, and C fibers may be classified as polymodal (multimodal) receptors because they can be activated by noxious, mechanical, thermal, and chemical stimuli. The neural basis for nociceptive transmission has been investigated and reviewed in detail in a monograph by Willis [14]. Abnormal mechanisms – primarily peripheral. The search for abnormal physiological mechanisms underlying TN and other neuropathic pain conditions has led to experimental laboratory studies of injured nerves as models of pain. Based on evidence from these physiological investigations, in combination with evidence from anatomical and clinical investigations, two schools of thought have emerged for describing
2391
2392
140
The pathophysiology of trigeminal neuralgia
the pathogenesis of pain in TN, one proposing primarily peripheral mechanisms and the other primarily central mechanisms. A line of investigation has been the demonstration of ectopic impulse generation after nerve injury in peripheral neuromas [180,181], in peripheral demyelinated nerves [182], in dorsal root ganglia [183,184], and in dorsal roots [185]. In second-order neurons of the central nervous system, ectopic impulses are generated by a process called central sensitization [186,187]. Thus, painful signals can be generated and transmitted in the absence of peripheral painful stimulation. In the presence of central sensitization, even a nonpainful peripheral tactile stimulus transmitted by A-beta fibers can generate (ectopically) a pain signal centrally (“A-beta pain”) [65]. Another route of studies has been directed at the concept of an “artificial synapse,” or ephapse, that permits long-lasting electrical cross-talk between neighboring axons in areas of demyelination after nerve damage [188,189]. For example, non-nociceptive input from large fibers could result in activity in nociceptive fibers. In situations where multiple axons are coupled, and one activates (by cross-talk) many others with instantaneous signal amplification, a lancinating or paroxysmal pain may result [190]. Perhaps more important than electrical ephaptic crosstalk, Lisney and Devor found another type of spread of excitation among injured axons resulting from diffusible chemical mediators they termed crossed-afterdischarge [191]. Amir and Devor later found that diffusible chemical mediators in injured axons also occurred in dorsal root ganglia [192]. Devor noted that with chemically mediated crossed-afterdischarge, unlike electrical ephapsis, single impulses have little effect. However, when repetitive activity in afferent fibers excites nonstimulated neighbors the concentration of paracrine chemical mediators build up. This progressively “winds up” discharge rates in the passive neighbors, sending them into a paroxysm of self-sustained firing [65].
Other abnormal physiological mechanisms have been invoked to account for the pain features of TN. One has been called the dorsal root reflex, which is an antidromic efferent impulse in customarily afferent dorsal root fibers. It is a consequence of presynaptic inhibition due to excessive presynaptic depolarization of A-delta (nociceptive) fiber terminals produced by synchronous activity in large myelinated afferents. Reverberatory activity in large fibers can cause enough presynaptic depolarization in the small fibers to generate a retrograde impulse [193]. Calvin et al. postulated that a trigeminal dorsal root reflex, coupled with retrograde reflected impulses, could allow a single impulse from the periphery to set up a reverberation, which in turn produces a burst of impulses arriving at the central axon terminals. This causes A-delta transmitter release, thereby creating a paroxysm of pain. This theory is based upon presynaptic inhibition and reflector sites due to focal change in axon diameter or myelination. Attempts to reduce pain by presynaptic inhibition should worsen the initial pain in a vicious cycle. If transmitter depletion at axo-axonic synapses terminated the attack, this would provide a refractory interval following an attack [194,195]. Various other peripheral factors have been proposed as responsible mechanisms for TN. These include mechanical pressure or traction on the root, infection-inflammatory lesions, and organic or functional vascular changes such as circulatory disturbances of the ganglion or root, hypertension with arteriosclerotic changes, and episodic ischemia. Kerr found variable changes in the roof of the carotid canal, including progressive decalcification with aging, and postulated that a mild pulsatile carotid artery contact with the adjacent trigeminal ganglion could be a significant factor in TN [157,196]. On critical analysis, List and Williams concluded that there was insufficient evidence for functional vascular changes as a primary etiologic factor for trigeminal pain. They noted the
The pathophysiology of trigeminal neuralgia
pathognomonic trigger mechanisms in some cases were located in the territory of a branch different from the pain, arguing against a peripheral origin. They concluded that a trigeminal pain attack represented a pathologic multineuronal reflex of the trigeminal system in the brainstem. The rationale of surgical treatment would be interruption of the peripheral pathway preventing physiological afferents from activating this abnormal mechanism [197]. Abnormal mechanisms – primarily central. Those who have espoused a central theory for the pathogenesis of pain in TN have relied heavily on evidence from clinical observations of patients to support their hypotheses. Trousseau [198] was one of the earliest to suggest that the painful paroxysms of TN resembled a convulsive seizure he called “epileptiforme neuralgia” [198]. Almost a century later, Wilson suggested the paroxysms of pain in TN could be a sensory “epileptiform” discharge regulated by some efferent sensory inhibitory mechanism to cause the paroxysm to cease [199]. Few have suggested that the cause of TN is in higher centers, such as the thalamus or sensory cortex, and most research on central mechanisms has been directed to the trigeminal brainstem structures where plaques of MS are an accepted cause of TN. Using semiquantitative methods, Kugelberg and Lindblom examined 50 patients with TN (48 with the common “cryptogenic” type) and observed that the neurophysiological state responsible for the tic syndrome was very unstable. They studied the relation between attacks of pain and stimuli applied to the skin trigger zone. A light touch stimulus or hair displacement, or in some a vibratory stimulus, were effective trigger mechanisms. Ineffective triggers, on the other hand, were electrical stimulation of the infraorbital nerve with a trigger area in its distribution, and a number of other stimuli such as temperature. Rapid displacement of a hair was sometimes sufficient to provoke an attack, but in most cases a larger spatial summation of afferent
140
impulses was necessary. Temporal summation of impulses over a considerable period of time was necessary to build up an excitatory state to evoke an attack. Occasionally, when the stimulus was prolonged, the pain would continue even after the stimulus was discontinued. Refractory phases were responsible for cutting the attack short. They considered that anticonvulsant drugs, such as hydantoin and lidocaine, raised the trigger threshold for the pain and thus favored a theory of central origin, probably in structures related to the trigeminal spinal nuclei, but did not prove a central origin, since both drugs also act on peripheral nerve [200]. In parallel with their clinical studies, King and Meagher showed in experimental animals a correlation between stimulation of large (touch) fibers and delayed activation of small (pain) fibers in the caudal part of the brainstem nuclei [201]. King’s group demonstrated the application of aluminum hydroxide gel or strychnine to the trigeminal spinal nucleus caused the animal to develop behavior consistent with trigeminal hyperesthesia [202,203]. A subsequent electrophysiological study showed complex (afferent and efferent) connections in the nucleus caudalis involving ascending trigeminal intranuclear pathways, with afferents from many nontrigeminal sources, and widespread efferent projections [204]. King summarized his findings by concluding we cannot exclude the possibility that multiple lesions may lead to secondary changes within the trigeminal complex, and that tic douloureux may become manifest because the physiological mechanisms have been triggered by any of a number of “etiologic” circumstances [205]. Evidence of a central confluence of extratrigeminal pathways in the trigeminal brainstem nuclear complex has provided further support to the hypothesis of a central origin for TN. As noted earlier, Denny-Brown and Yanagisawa studied the phenomenon of sensory overlap in dermatomes adjacent to the trigeminal territory [170].
2393
2394
140
The pathophysiology of trigeminal neuralgia
Wyburn-Mason [206] and Crue et al. [207] have found that blockade of the great auricular nerve or the greater occipital nerve, whose dermatomes abut the trigeminal territory, can sometimes relieve TN. Anatomically, the trigeminal nucleus caudalis is in direct continuity with the substantia gelatinosa of the upper cervical dorsal horn, providing a central region for interplay in TN. Crue recorded antidromic activity in peripheral trigeminal branches synchronous with consciously felt jabs of pain, which strengthened his conclusion that there is an underlying uncontrollable sensory epileptiform discharge responsible for TN [208]. Based on his analysis of the clinical characteristics of his 278 cases of TN, and on physiological data in the literature, Pagni postulated that an epileptic subliminal focus in the trigeminal nuclear group was “kindled” by chronic anomalous afferent barrage, making TN a sort of sensory reflex epilepsy [209]. Other observers have noted the similarity of the paroxysmal and provocable pain syndrome of TN with the much rarer neuralgias of the vagoglossopharyngeal and nervus intermedius cranial nerves, as well as their common central anatomic interconnections terminating in the nucleus caudalis. Sweet pointed out significant clinical differences in five different categories between the trigeminal prototype and the other two cranial nerve syndromes: variability in type and duration of pain; variability in location, with much more frequent radiation well beyond the aural and oropharyngeal zones; local tenderness; striking ancillary manifestations related to activity of the carotid sinus; and preoperative sensory loss in the zones of cranial nerves IX and X [28]. In addition to the disparate cranial nerve neuralgias that share features of paroxysmal and provocable pain, rarely TN may occur in one patient with concomitant dysfunction in another cranial nerve. The occasional association of TN with hemifacial spasm was first called “painful tic convulsif ” by Cushing [210]. In a study by
Rushton et al. of 217 cases of vagoglossopharyngeal neuralgia, 25 had TN, 24 ipsilateral and one contralateral [211]. The term hyperactive dysfunction syndrome has been employed recently, almost always for TN associated with hemifacial spasm, to imply a shared mechanism of vascular compression affecting each nerve. In a series of 1,472 patients considering surgery for vascular decompression, Kobata et al. had 41 cases with two cranial nerves involved, bilaterally in 19, forming a group they called combined hyperactive dysfunction syndrome. The patients who had involvement of two nerves were older than the rest of the group. There were no cases of combined vagoglossopharyngeal neuralgia and hemifacial spasm, and no cases of nervus intermedius neuralgia [212]. In contrast to hemifacial spasm and its recognized association with vascular compression of the seventh nerve, it is interesting that hemimasticatory spasm is almost unknown. Thompson and Carroll reported a single case. There was no trigeminal pain, and trigeminal function and cerebral angiography were normal [213]. In a review of cranial nerve dysfunction syndromes, Møller proposed that vascular compression itself did not cause these disorders, but rather that the symptoms and signs were produced by changes in their respective nuclei in association with vascular compression [214,215]. Adding to the insights gained from these clinical observations, other neurophysiological studies have led to further theories about mechanisms in trigeminal pain. Using experimental data, Dubner et al. have proposed that structural and functional changes in the trigeminal system result in an alteration in the receptive field organization of central wide-dynamic-range or multireceptor neurons. These provide an explanation for the pain triggered by innocuous stimuli, its radiation beyond the stimulus site, and its referral to distant sites. Noting that partial deafferentation can be present without detectable sensory loss, the hypothesis proposes that central
The pathophysiology of trigeminal neuralgia
changes could occur independently of any peripheral alteration; for example, in MS patients with TN, plaques can be located in the trigeminal nuclei. Touch stimulation could produce activity in wide-dynamic-range (WDR) neurons normally only produced by noxious stimuli [216]. Pharmacological agents that relieve pain in TN have been studied with respect to both their central and peripheral actions. Fromm et al. found that the anticonvulsants phenytoin and carbamazepine, and an antispasmodic, baclofen, can relieve pain in TN, whereas anticonvulsant barbiturates do not. In experimental studies, they found that the effective drugs have in common the ability to facilitate the segmental inhibition and depression of excitatory mechanisms in the spinal trigeminal nucleus. Their hypothesis was that in TN irritation of the trigeminal nerve caused failure of segmental inhibition along with excitation due to ectopic discharges. This was reversed by phenytoin, carbamazepine, and baclofen, or by surgical procedures that decreased the amount of afferent activity. They concluded that TN has a “peripheral etiology and a central pathogenesis” [217–219]. Burchiel showed that carbamazepine, in addition to its central pharmacologic action, inhibits spontaneous ectopic activity peripherally in experimental neuromas [220]. Sweet concluded that any minor tissue damage in the trigeminal system that permits breakdown of insulating mechanisms between axons without destroying them may permit the development of paroxysmal pain. Touch and proprioceptive stimuli provoke attacks because they cause the largest action potentials, and their large fibers have rich connections in the nucleus caudalis. He said that it was presently unknown why touch to or movement of the face does not normally cause pain by firing the caudal nuclear cells, nor is it known whether it was even necessary to hypothecate a primary abnormality at the level of the secondary afferent neurons, or their synapses, to account for the attack. The normal
140
central apparatus may be vulnerable solely on the basis of a lesion in the primary afferent fibers leading to a secondary abnormality in the central neurons [2]. Central responses to peripheral deafferentation have been the subject of pain studies by several investigators. If partial deafferentation due to peripheral injury to the trigeminal nerve is a mechanism for causing TN, then physiological studies of deafferentation are quite relevant. Loeser et al. studied neuronal activity in the injured spinal cord of a paraplegic patient. They found spontaneous hyperactivity in neurons of the spinal cord resembling the pattern found in experimental studies in cat and in primate cortical epileptic foci [221]. After retrogasserian rhizotomy in cats, progressive neuronal hyperactivity in the spinal trigeminal nucleus was found by Anderson et al. [222]. Westrum demonstrated, by ultrastructural studies in cats after trigeminal rhizotomy, that specific synaptic changes occurred in the spinal trigeminal nuclei. This suggested that the resultant effect may not be one of anticipated postsynaptic inhibition at the sites, but a different effect determined by deafferentation sites and affecting the total membrane properties of the neurons [223]. Gorecki et al. [224], and Rinaldi et al. [225] have found neuronal hyperactivity in thalamic nuclei in patients with chronic deafferentation pain. After reviewing his and other experimental results, Sessle concluded that deafferentation may result in morphological, chemical, and functional change within the central nervous system reflecting brain neuroplasticity [226]. Young has suggested that TN may be considered a form of partial deafferentation pain [176]. Clinical electrophysiology. Clinical electrophysiological studies in patients with TN have provided limited evidence of physiological abnormalities. Scalp recordings of somatosensory evoked potentials after mechanical or electrical stimulation of infraorbital and mental nerves demonstrated a delay in conduction time in
2395
2396
140
The pathophysiology of trigeminal neuralgia
some patients with TN as well as in some patients with posterior fossa tumors [227–229]. Intraoperative stimulation and recording from peripheral trigeminal nerve branches, ganglia, middle fossa, and posterior fossa roots during percutaneous or posterior fossa procedures thus far have yielded no consistent data about physiological abnormalities in TN [230–235]. In a report by Leandri et al. of ten patients undergoing MVD for TN, direct recordings from the REZ and scalp showed impaired conductance with immediate recovery in seven patients when the root was decompressed [236]. Using laser-generated radiant heat pulses as a stimulus to selectively activate A-delta and C nociceptive fibers and warmth receptors, Cruccu et al. recorded reflex and evoked responses in 47 patients with idiopathic TN. They showed that 51% had abnormal laser evoked potentials on the painful side. However, even on the nonpainful side, mean latency was longer than in controls. Carbamazepine markedly depressed the evoked potentials [237]. During percutaneous trigeminal rhizotomy, Sindou noted electrically evoked trigeminal motor responses in the masticatory muscles as well as in the facial muscles innervated by the facial nerve. He attributed this to a “trigemino-facial” reflex. He was mindful that TN was first named tic douloureux because of the motor reaction in the face accompanying a paroxysm of pain [238]. In a noninvasive laser evoked potential study, stimulation in any division of the trigeminal nerve would elicit a blink reflex in the orbicularis oculi [239]. While performing microneurographic recording of single and multi-unit action potential discharges in the trigeminal ganglion during percutaneous surgical procedures, Burchiel and Baumann found the technique useful for receptor field mapping and estimation of axonal conduction delays using electrical stimuli [240]. Others have measured trigeminal related reflex responses, specifically the blink or corneal reflex (orbicularis oculi reflex), and the masseter inhibitory reflex, comparing patients with TN
and those with atypical facial pain, yielding no consistent findings [241–243]. Adams has criticized many of the clinical electrophysiologic studies due to methodological variations in techniques [150]. Sensory loss in TN. There is only limited information regarding the possibility of facial sensory loss in patients with TN, with contradictory clinical and pathophysiological data available. It is widely recognized that customary, careful bedside techniques of testing perception of light touch and pinprick will reveal no impairment to sensation throughout the trigeminal territory in patients with idiopathic TN. On one hand, if even the slightest sensory defect is found, most clinicians agree that it would be prudent to conduct further diagnostic investigations to search for any underlying lesion causing symptomatic TN. In addition, patients with idiopathic TN, on careful questioning, rarely report subjective symptoms of facial numbness or paresthesias. On the other hand, several investigators, using elaborate semi-quantitative somatosensory testing methods, have reported evidence of subtle sensory changes in some modalities in patients with the classic symptoms of TN. One of the early reports of sensory loss in TN was by Lewy and Grant [244]. Using unspecified “electrical examination of sensibility” in 50 patients they found reduced number of “touch and pain points” in the face in 25% of cases [244]. In the last two decades somatosensory studies of patients with TN have employed a variable combination of methods to measure: tactile thresholds with von Frey filaments; vibratory sensation with a vibrometer; pinprick with weighted needles; pressure pain using forceps with a strain gauge; warm, cool, and heat pain with a thermode; warm and cold discrimination; two-point discrimination with a Weber compass; and skin temperature with a digital thermometer. The affected division(s) are typically tested, using unaffected ipsilateral adjacent division(s), and contralateral divisions, as controls.
The pathophysiology of trigeminal neuralgia
Hampf et al., in a study of 18 unoperated “virgin” TN patients, found no sensory deficits. However, he found a mean lower skin temperature of 0.45 C only in the affected division, signaling cutaneous vasoconstriction. Skin temperature returned to normal after successful radiofrequency rhizotomy [245]. Nurmikko performed similar studies in 26 patients with TN and found increased threshold to warm thermal stimuli and to tactile stimulation in the affected division. In the adjacent divisions he found increased thresholds to warm, hot pain, and two-point discrimination. When the results were combined, at least 15 of 26 patients showed at least one abnormal measure of sensation [246]. Bowsher et al. studied 26 patients with TN and found increased threshold to touch and thermal sensitivity, but not to pinpoint or heat pain. They also found a deficit to tactile, but not thermal, sensation in unaffected divisions on the side of pain [247,248]. Working in the same group, Miles et al. evaluated 19 patients who underwent MVD. Preoperative evaluation of touch and thermal modalities returned to normal, although a new deficit to pinprick appeared postoperatively, which returned to normal if the patient stayed pain-free [249]. Eide and Stubhaug studied 14 patients with idiopathic TN and found increased temperature and tactile thresholds [250]. Eide et al. studied 23 patients with trigeminal neuropathic pain described as continuous, dysesthetic pain. A subgroup had a history of previous nerve injury and the remainder had “spontaneous” neuropathic pain. They found an increase in touch and tactile thresholds in those with nerve injury, but not in those with spontaneous neuropathic pain [251]. In a more limited study by Sinay et al., cold and warm sensation thresholds were decreased, and cold and heat pain thresholds were increased, in the symptomatic division as well as in the two other divisions ipsilateral to the pain in nine patients with idiopathic TN. Tactile and pinprick thresholds were not tested [252]. The detection of a
140
consistent, even if limited, sensory loss in one or more modalities in patients with TN may be a significant clue for investigation of pathophysiologic mechanisms in TN. “Ignition hypothesis.” One of the most comprehensive hypotheses proposed to account for specific findings related to the mechanisms underlying the clinical pain syndrome of TN is the “ignition hypothesis,” first presented by Rappaport and Devor [253]. Their model was based on extensive laboratory research and clinical observations. It integrates the clinical phenomena of TN with the known histopathology, experimental pathophysiologic studies of injured sensory nerves, and the pharmacology of peripherallyacting membrane-stabilizing drugs. The hypothesis holds that abnormal impulse activity originates in peripheral aspects of the trigeminal system rather than in epileptic foci in the central nervous system, and addresses three neurophysiological problems: triggering, amplification, and a stop mechanism. They list 14 key features of the syndrome that call for an explanation, as shown in > Table 140‐3 [254]. The “ignition hypothesis” proposes that pain paroxysms result from synchronous recruitment of ectopic firing in a large population of hyperexcitable neurons in the trigeminal ganglion that innervate the territory or division(s) in which the pain is felt. The ectopic hyperexcitability of these neurons is due to an axonopathy caused by vascular compression, or, less frequently, to direct root or ganglion pathology, or to intracerebral damage of trigeminal root axons (e.g., tumor or MS). Synchronous recruitment results from nonsynaptic coupling mechanisms characteristic of injured peripheral neurons, chemically mediated cross-excitation, or electrical ephaptic cross-talk. In cross-excitation, neurotransmitter molecules, and probably also potassium ions, are released into the extracellular space from active neurons. These diffuse towards neighboring neurons and excite them. In ephaptic cross-talk, excitatory electrical current passes
2397
2398
140
The pathophysiology of trigeminal neuralgia
. Table 140‐3 Key clinical features of trigeminal neuralgia 1. Intense unilateral pain 2. Pain triggered by non-noxious touch stimuli 3. Temporal summation to repeated taps at the trigger site 4. Pain that outlasts the provoking stimulus 5. Pain that spreads beyond the (trigger) point stimulated 6. Brief pain paroxysms 7. Postattack refractoriness 8. Minor (but not dramatic) hypesthesia in the area of pain reference 9. Minor root, trigeminal ganglion, or CNS tract injury 10. Efficacy of nerve and root ablation 11. Efficacy of trigeminal ganglion manipulations (e.g., glycerol, balloon compression) 12. Efficacy of some anticonvulsants (e.g., carbamazepine), but not others (e.g., barbiturates) 13. Efficacy of surgery for MVD 14. Symptoms essentially unique to cranial nerves (e.g., no TN equivalent along the spine) From Devor et al. [83] With the exception of feature 14, all can be explained in terms of known abnormalities that develop in primary sensory neurons following injury
directly from the active neurons to closely apposed passive neighbors. According to the ignition hypothesis, paroxysms begin with discharge in a small cluster of trigeminal neurons, presumably low-threshold afferents, that begin to fire in response to cutaneous trigger-point stimulation, or even spontaneously. Cross-excitation and ephaptic coupling in the injured trigeminal root or ganglion then “ignites” activity in passive neighboring neurons, including nociceptors. This augmented activity ignites additional passive neighbors, which in turn ignite their passive neighbors. The resulting positive feedback chain reaction builds up rapidly into an intense explosive crescendo. Although triggered by peripheral afferent input, the actual ectopic impulse activity underlying the pain is generated within the injured trigeminal root axons or ganglion, and can persist for some time after the end of the stimulus itself (“afterdischarge”). Since afferent
neurons of all types become active simultaneously, something that otherwise occurs only with electrical stimulation, the sensation feels like an electric shock. After a few seconds of massive firing, activity evoked after-suppression develops, damping the afterdischarge paroxysms and establishing a period of refractoriness. Like central nervous system seizure activity, the ignition mechanism is expected to be sensitive to membrane-stabilizing drugs, such as anticonvulsants. Drug action, according to this hypothesis, is supposed to be in the peripheral nervous system rather than in the central nervous system [190,253–255]. As they and others have noted, trigeminal and other cranial nerve neuralgias respond to anticonvulsants, such as carbamazepine and hydantoin, that act as membrane stabilizers, whereas barbiturate anticonvulsants act synaptically and are ineffective in TN [65]. Recognizing that TN unquestionably occurs in some patients who do not appear to have compressive root pathology, it is proposed that vascular compression is a sufficient condition, but not a necessary condition, and that the primary disorder could be in or near the ganglion or spinal tract [254]. Devor has also noted that if a disease state causes sensory neurons to become hyperexcitable by virtue of faulty metabolic regulation, there could be ectopic firing and pain in the total absence of nerve injury [11]. The ignition hypothesis model for TN will be very valuable for future laboratory and clinical research in the pathophysiology of TN and for developing improved treatments for TN.
Natural History of Trigeminal Neuralgia Pre-trigeminal neuralgia. There are several clinical features in the natural history of TN that pose intriguing questions about the pathophysiology of the disease. The onset of severe paroxysmal pain is usually abrupt. However, very infrequently
The pathophysiology of trigeminal neuralgia
there may be prodromal symptoms uncharacteristic of typical TN that can precede the onset of typical paroxysmal pain by months or years. This condition has been called pre-TN, and its diagnosis awaits the appearance of the typical syndrome. First described by Symonds, there is dull, continuous aching pain in the upper or lower jaws and a much later addition of paroxysmal pain of the classic type [3]. Later, Mitchell coined the term, “pre-TN,” to describe 38 patients who reported dull, aching, or burning pain in an alveolar quadrant for weeks or months preceding symptoms of typical TN [256]. Fromm et al. reviewed records of 18 patients with typical TN who had a medical history of pre-TN, characterized by symptoms such as aching, gnawing, burning, toothache, or sinusitis-like pain sometimes triggered by chewing, brushing teeth, drinking hot or cold liquids, or yawning, until TN developed a few months to 12 years later. This affected one or more branches, sometimes the ophthalmic, and, when the pain of typical TN appeared, it always affected the same branch. Another six patients were evaluated while experiencing what appeared to pre-TN and had normal neurological, dental, and imaging examinations. They became pain-free after treatment with carbamazepine or baclofen. They concluded that recognition of pre-TN permits relieving the pain with appropriate medications and avoids unnecessary irreversible dental procedures [257]. Course-onset, distribution, radiation, migration. In large series of patients with TN, the mean age of onset has been noted to be in the early fifties. The pain seldom begins in the nineties or in teenage years, but rare cases have been reported with onset under the age of ten. Onset is usually spontaneous and acute. Kugelberg and Lindblom observed a distinct predilection for the pain and trigger zones to be in the maxillary and/or mandibular territories [200]. This led Henderson to suggest the term, “bigeminal neuralgia.” He noted the strong tendency for the pain to remain confined to the territory of
140
onset in 95% of cases, even after follow-up of more than 20 years. Except in cases associated with MS, pain was seldom bilateral, and in 5% of his cases with both sides affected, pain was frequently in symmetrical zones and did not begin on the opposite side until 2–10 years later [258]. Radiation of pain beyond the trigger zone, occasional migration of trigger zones, and the presence of trigger sites distant from the trigeminal territory have been noted in studies by Dubner et al. [216] and Adams [144]. Years after onset, even following successful treatment of the affected division(s), pain and trigger zones can migrate to, or develop in, adjacent division(s). Although it is common for paroxysms of pain to fluctuate in intensity and ease of provocation over days, weeks, and months, most patients report an inexorable worsening of the pain over time, concomitant with an apparent decline in effectiveness of medications. Spontaneous remissions. The most striking and inexplicable feature in the natural history of TN is the common occurrence of repeated spontaneous total remissions from pain lasting weeks, months, or even many years. Accounting for these remissions has posed a major problem for studies of the pathophysiology of this extremely painful disease. Rushton and MacDonald (1953) analyzed 155 cases of TN seen at the Mayo Clinic and found that half the cases had experienced one or more spontaneous remissions lasting 6 months or longer, and a fourth had similar remissions lasting 12 or more months. The patients with significant remissions considered them to be a “welcome oasis in a dreary desert of pain” [259]. Rasmussen performed a prospective study of 1,057 facial pain patients in Denmark over an 18 year period. In the group of 229 patients with typical TN, 58% had pain-free intervals (16% lasting weeks, 36% lasting months, 6% lasting years) [260]. Selby described total remissions in the course of untreated disease lasting a few weeks to 20 years or longer. He noted that the duration of pain-free
2399
2400
140
The pathophysiology of trigeminal neuralgia
intervals became shorter [19]. Terrence described characteristic remissions and cautioned that any study of medical or surgical treatment should incorporate some method to avoid bias due to the natural remission rate [261]. In a commentary about MVD, Loeser asked why there were pain-free intervals of days, weeks, months, or years, given that arterial compression is not thought to come and go [262]. There is no experimental evidence to show that remyelination, regeneration, or further loss of axons can account for the repeated spontaneous remissions so commonly observed in the course of TN. Theories about the pathogenesis of TN must address this phenomenon. Even in the group of patients with TN who never experienced temporary pain relief due to remission, a notable characteristic in the vast majority of TN patients, at least in my experience, is the stable, stoic personality maintained throughout their ordeal. This feature often stands in contrast to other patients with severe longstanding pain conditions whose understandable emotional lability and depression are often evident.
Lessons from Surgical Treatment The evolution of surgical treatment for TN has resulted from interplay between advances in scientific knowledge about the nervous system on the one hand, and lessons learned from clinical observations of surgical results by enlightened neurosurgeons on the other. On a personal note, my training under Dr. William Sweet was enriched by a fortunate opportunity to study and engage in research in brain science at the Massachusetts Institute of Technology. My gifted mentor at MIT, Professor Hans-Lukas Teuber, counseled me to view neurosurgical procedures as a form of “experiments in nature,” in order to learn more about nervous system functions. My predecessors and contemporaries have
made remarkable achievements in this manner, laying a foundation for our current ability to successfully treat many patients with often incapacitating TN. History. There are several excellent reviews about the historical development of surgery for TN by Meirowsky [263], Stookey and Ransohoff [1], Sweet [2, 264], and Gildenberg [265]. The landmarks in the surgical treatment of TN are summarized in > Table 140‐4, modified from Gybels and Sweet [266]. Safer and more effective surgical procedures have replaced some of the older ones. Currently, one group of procedures proven effective in selected cases is designed to ablate a component of the trigeminal system: peripheral branch neurectomy/alcohol neurolysis, percutaneous middle fossa rhizotomy/gangliolysis (radiofrequency thermocoagulation, glycerol injection, balloon compression), stereotactic radiosurgery, posterior fossa partial or total rhizotomy, stereotactic radiofrequency medullary trigeminal tractotomy/nucleotomy, and “dorsal REZ” (DREZ) radiofrequency lesioning of trigeminal tract and nucleus caudalis. The other procedures to relieve pain do not involve intentional ablation and consist primarily of posterior fossa MVD of trigeminal roots and stimulation of components of the trigeminal system. Detailed accounts of these operative procedures are presented in other chapters in this textbook. Compression versus decompression. From a historical perspective, some of the surgical procedures developed have been employed to compress the trigeminal roots or ganglia, while others have been used to decompress the roots, ganglia, or peripheral divisions. All have had a significant incidence of sustained pain relief with mild or no sensory impairment and relatively infrequent recurrences. Taarnhoj decompressed the ganglion and middle fossa roots by dividing the dural roof in ten patients. Later, in five patients, using a posterior fossa approach, he placed a hook into the porus in an attempt to enlarge it. Most of
The pathophysiology of trigeminal neuralgia
. Table 140‐4 Landmarks in the surgical treament of tic douloureux 1. First complete publication of a case 2. Unsuccessful peripheral section of a branch of the trigeminal nerve 3. Description of the disorder, coining of term, tic douloureux, chemical destruction of peripheral branch 4. Gasserian ganglion resection 5. Retrogasserian rhizotomy
6. Alcohol injection in a branch of the trigeminal nerve 7. Alcohol injection in Gasserian ganglion 8. Report of 298 consecutive Gasserian ganglionectomies without mortality 9. Partial section of V root at the pons 10. Electrocoagulation of Gasserian ganglion 11. Bulbospinal tractotomy 12. Stereotactic radiosurgery of trigeminal root 13. Decompression of the trigeminal root 14. Compression of the trigeminal root 15. Vascular decompression 16. Temperature-controlled coagulation of Gasserian ganglion and rootlets 17. Microvascular decompression 18. Retrogasserian glycerol injection 19. Percutaneous compression of Gasserian ganglion
Wepfer [389] Schlichting (according to Rose [390]) Andre´ [391]
Krause [392] Horsley et al. [393] Spiller and Frazier [394] Schloesser [395]
Harris [396] Cushing [210]
Dandy [96] Kirschner [397] Sjo¨qvist [89] Leksell [398] Taarnhøj [267] Shelden et al. [268] Gardner and Miklos [115] Sweet [399]
Jannetta [272] Ha˚kanson [400] Mullan and Lichtor [271]
Modified from Gybels and Sweet [266]
his patients experienced pain relief [133,267]. Shelden et al. performed decompression of peripheral divisions by enlarging the foramen ovale and/or foramen rotundum through a temporal
140
craniotomy with saline neurolysis and linear incision in the dural sheaths. Comparing their results with those reported by Taarnhoj, they concluded that the only apparent feature both procedures had in common was operative trauma. This led them to initiate a procedure to minimally expose, and then gently compress, the middle fossa roots. All 29 patients were completely relieved of pain, even though some reported minimal sensory loss [268]. Stender decompressed Meckel’s cave and ganglion by opening the dural roof. Of 16 patients with tic douloureux, all were relieved of pain for up to 13 months [269]. Malis decompressed the root at the petrous apex by dividing a dural band of fibers in 43 patients, all obtaining complete pain relief. There was only minimal sensory loss in six patients [270]. White and Sweet [2] summarized reports of 1,790 patients who had a useful degree of pain control attributed by the reporting surgeons to a compression procedure. The recurrence rates were from 8 to 28% in these reports [2]. Mullan and Lichtor developed a percutaneous compression procedure using a balloon inflated in Meckel’s cave, and this procedure remains in current use because of its success in relieving pain with little or no sensory loss [271]. The compression procedures appear to have outcomes similar to those of the increasingly popular posterior fossa MVD operation. Jannetta has observed that if the trigeminal nerve is traumatized at operation, the patient awakens painfree, but if it is not traumatized during the dissection, the patient will probably have tic douloureux for a few days to several weeks postoperatively [117,272]. He has emphasized that neurovascular compression in TN may be extremely subtle, with an arteriole as small as 50 mm [273] or a venule as small as 0.3 mm in diameter [274]. Loeser has commented that every trigeminal root that he has explored has capillaries and small veins and arteries within it bridging to the arachnoid [262].
2401
2402
140
The pathophysiology of trigeminal neuralgia
Jannetta maintains that neurovascular compression of the REZ can be clearly stated as causal of the vast majority of cases of TN [117]. It is difficult to reconcile how surgical compression or decompression procedures of the trigeminal nerve in the middle or posterior fossa both relieve pain of TN. Even more puzzling are anecdotal accounts of protracted pain relief after only exposure or minimal manipulation of the nerve. Parkinson described a case of planned subtemporal rhizotomy aborted because of bleeding. The patient awoke with perfect sensation in the face and never suffered trigeminal pain again [275]. Sweet described several similar cases he often liked to discuss with me. An example was a patient with a 6-year history of TN that had spread to the ophthalmic division. He chose a posterior fossa approach to perform MVD but found no offending vessel, using care to avoid any contact with the root. He then planned to perform his controlled radiofrequency thermocoagulation procedure. The patient, however, was pain-free after the aborted MVD and remained that way until he died eight and a half years later. Based on all the evidence from surgery, he concluded that the concept that root compression near the pons was the commonest cause of TN was untenable [28]. There is still debate about which surgical procedure is treating the “cause” of TN. For example, the Sweet procedure, percutaneous controlled radiofrequency thermocoagulation of the trigeminal roots, permits differential interruption of small A-delta and C pain fibers, allowing relative preservation of larger myelinated tactile fibers. This effect from controlled heating of the nerve has been demonstrated in the laboratory by Letcher and Goldring [276]. The Sweet procedure is the only one capable of selective reduction of transmission in nociceptive afferent fibers sparing larger myelinated afferents. At the same time, it has demonstrated a high success rate in producing prolonged pain relief. If the pathophysiology of TN is related to
abnormalities in distribution of afferent inputs from large and small fibers, then perhaps modification of the differential inputs produced by the Sweet procedure may play a role in treating the “cause” of TN. Stimulation. The application of stimulation techniques to treat pain, including TN, has received increasing interest from clinicians and investigators searching for improved methods to treat this condition. Inspired by the “gate-control theory of pain” proposed by Melzack and Wall [277], one of the first efforts to modulate pain by stimulation of the trigeminal system in humans was by Wall and Sweet [278]. They first electrically stimulated their own supraorbital nerves with needle electrodes and observed that it was not unpleasant. They noted that during the stimulation period, and for a few minutes thereafter, pinprick in the supraorbital territory did not feel sharp to either of them. They subsequently stimulated peripheral nerves in eight patients with severe cutaneous pain, including one with TN in the maxillary division. In that patient during a 5-min stimulation, and for 17 min afterward, it was not possible to provoke the usual stabs of pain. Later, transcutaneous electrical stimulation for pain was tried [279]. Others began to apply the principle of electrical stimulation for pain, stimulating the spinal cord, thalamus, and deep brain structures [280–282]. Pudenz et al. described three patients with intracranially implanted stimulators on the mandibular division of the trigeminal nerve. After several weeks of stimulation, one patient was pain-free for 7 years and recurrent pain was abolished by activation of the stimulating unit. Another patient was pain-free even though the unit was never activated during the period, until he was lost to follow-up several years later. The third patient was pain-free for 5 years. After the unit was re-energized, he had one more year of relief [283]. Percutaneous electrical implants in supraorbital and infraorbital nerves for trigeminal neuropathic pain were tried by Johnson and
The pathophysiology of trigeminal neuralgia
Burchiel [284]. Chronic electrical stimulation of the Gasserian ganglion was done by Meyerson and Ha˚kanson in five patients with atypical facial pain, obtaining good relief with a follow-up of 6–21 months [281]. Taub et al. also tried chronic Gasserian stimulation [285]. Thalamic stimulation and recording for deafferentation and central pain has been performed by Gorecki et al. [224], and motor cortex stimulation for thalamic and deafferentation pain was done by Tsubokawa et al. [286,287] and others. Outcomes. An important lesson from surgical treatment of TN is the lack of carefully executed outcome studies to compare safety and efficacy of the procedures currently being performed. Until more advanced outcome studies become available, examples of useful comparative analyses of reported outcomes after various surgical procedures for TN have been provided by Zakrzewska and Thomas [288] and Taha and Tew [289]. Based on personal experience and armed with the best outcome studies for the different operations, the neurosurgeon can then fully inform and advise prospective surgical candidates with TN about the benefits and risks expected from each procedure.
Lessons from Medical Treatment of Trigeminal Neuralgia In the modern era, several pharmacological agents have been introduced for treatment of TN, each with varying degrees of efficacy and adverse effects. What all these drugs have in common is their primary use as anticonvulsants. Opiates have been clearly ineffective in relieving trigeminal pain. Among the drugs in use today for TN are phenytoin, initially used in 1942, carbamazepine in 1962, baclofen in 1972, clonazepam in 1975, sodium valproate in 1980, and oxycarbazepine in 1991 [290]. Misoprostol, a prostaglandin E analogue, thought to possibly suppress inflammation in multiple sclerosis
140
plaques, has been tried with limited success in relieving TN associated with MS [291]. More recently, lamotrigine, gabapentin, and pregabalin are being offered to some patients with TN. There are no agreed upon guidelines on the conduct of randomized controlled trials for medical therapy. No survey has been done to ascertain what outcome the patient could expect, given that a complete cure is rarely achieved, and that most treatments have side effects [292]. Barker stated that reliable success and side effect rates are essentially unavailable for medical treatment of newly diagnosed TN [293]. Spontaneous remissions from pain further complicate the evaluation of treatment efficacy. As the most effective drug treatment, carbamazepine provides initially good responses in about 80% of patients with a gradual decline in efficacy over time. Adverse effects with this medication are frequent, especially as the dose often has to be adjusted upward. Consequently, many patients experience significant problems with cognitive function, with impaired intelligence and memory, as well as dizziness, problems with balance, and somnolence. On several occasions, Sweet emphasized that, being in a position of responsibility, he would be reluctant to take medications such as this because of these common side effects. The elderly and patients with motor weakness and incoordination due to MS are frequently intolerant of the adverse effects of carbamazepine, further limiting its usefulness. In my experience, one of the merits of this drug is to test its effectiveness in providing pain relief, at least on a short-term trial basis, in patients with atypical symptoms, when the diagnosis of TN is in doubt. Some insights into the pathophysiology of TN have come from pharmacological studies of the mechanisms of action of the drugs used to treat TN. Fromm showed that the antispastic agent, baclofen, resembles carbamazepine and phenytoin in that it facilitates segmental inhibition in the trigeminal brainstem nuclear complex [294]. Devor has reasoned that the most
2403
2404
140
The pathophysiology of trigeminal neuralgia
important action of therapeutic agents that act within the central nervous system is modulation of synaptic transmission. He finds it noteworthy that anticonvulsants that act synaptically, such as barbiturates, are largely ineffective in treating trigeminal pain in contrast to anticonvulsants like carbamazepine, phenytoin, and lamotrigine, which are effective analgesics and act as membrane stabilizers through peripheral sodium channel blockade. Also, he notes that baclofen is known to suppress dorsal root ganglion neurons and that gabapentin suppresses ectopic discharges, perhaps by suppressing calcium channel conductance [65,253]. A veterinarian recently gave me a report about an increasingly recognized idiopathic disease in horses called “head shaking.” Headshakers present with uncontrollable, spontaneous, frequently repetitive vertical, horizontal, or rotary head movements with such frequency and violence that it becomes dangerous to ride the horse, which appears to be distressed. It resembles the expected response to a fly or bee about the nose suggesting an abnormal painful sensation being felt by the horse. It was determined that the caudal nasal branch of the maxillary division of the trigeminal nerve could be anesthetized resulting in an 80–100% reduction in clinical signs. One side was more often predominantly affected. An occlusal nasal mask would stop air flow reducing stimulation of the presumed trigger zone in the nasal cavity. When the mask was removed the signs returned. Based on information about TN in humans, carbamazepine was tried and many headshakers demonstrated a positive response. According to the report, affected horses unsuccessfully treated were commonly sold or subject to euthanasia! [295]. As described, this condition in the equine species, sometimes responsive to carbamazepine, appears strangely similar to the early descriptions of tic douloureux in humans. Fortunately, when medical treatment fails, neurosurgeons can offer a variety of procedures
in the middle and posterior fossa with remarkable and mysterious high percentage of longsustained pain relief that, at least for some of us, defies rational explanation, according to Sweet [28]. The history of non-surgical treatment tried for TN is replete with unproven remedies and sometimes harmful tactics sadly delaying the suffering patient from receiving effective pain-relieving treatment. Such errors are not only compounded by temporary placebo effects, but also, in the case of TN, by common spontaneous remissions misleading the patient and therapist alike about the valid success of any given treatment modality. Unlike too many other severe persisting pain conditions, TN today is an imminently treatable disease. Even when pharmacologic therapy initially provides sustained freedom from pain, frequently the effectiveness of medication dwindles significantly within months, in spite of steadily escalating dosages and even resorting to polypharmacy. Often, in association with the use of multiple medications, patients with TN become significantly, sometimes seriously, impaired by the well-known adverse effects of the available drugs. Since this is a common problem, there is a great need for education of patients and physicians about the various effective surgical alternatives to ensure that patients obtain consultation with a neurosurgeon familiar with the treatment of TN. Patients should not have to ask later why they were not advised about surgical opportunities for effective treatment of their severe trigeminal pain.
Trigeminal Neuralgia: the Future Current and future studies will improve our knowledge of the pathophysiology and treatment of TN. Further ultrastructure studies of pathological anatomy in biopsy specimens of sensory roots, coupled with similar postmortem studies of the trigeminal system, will confirm and expand
The pathophysiology of trigeminal neuralgia
our understanding of morphological changes unique to TN. This will require programs with interest and expertise in neuroanatomy and neuropathology. Experimental studies in biology of neural degeneration and regeneration, including peripheral and central myelination, should add insights into abnormal mechanisms in TN. Advanced imaging techniques will provide new information about structural and functional abnormalities in the trigeminal system. New MRI techniques to define vascular relationships with the trigeminal sensory root, in normal and affected patients, should be valuable in surgical planning, as well as in defining the role of neurovascular conflicts in TN. Other imaging techniques, such as positron emission tomography (PET) scans, functional MRI (fMRI) scans, MRI spectroscopy, and diffusion tensor tractography will be important tools to investigate the trigeminal system, including the lesions of MS and their intriguing association with TN [296]. Improvement in somatosensory testing of trigeminal function should provide better instruments to measure the minimal sensory impairments that may be present in some individuals with TN. An example could be laser evoked potentials to assess small pain-fiber transmission in the trigeminal system [297]. Borsook et al., in a very promising line of investigation, studied fMRI activation in the trigeminal ganglion using brush strokes to activate A-beta fibers and painful thermal stimuli to activate A-delta and C fibers. This also permitted somatotopic mapping within the Gasserian ganglion [298]. They used similar technology to perform fMRI activation for the first time in a patient with tic douloureux during evoked and spontaneous tics. Activation maps in the central nervous system showed significant increased activities in the cortical region (frontal, parietal, and temporal cortices, and cingulate and insular cortices), as well as some central regions (thalamus, basal ganglia, and pontine nuclei) [299]. Investigations like these should help to elucidate mechanisms in TN, and other
140
new studies in this area should enable us to develop better treatment for trigeminal pain. Based on extensive neurophysiological studies, Devor has suggested new strategies for more effective pharmacological targeting of painrelieving drugs. Since anticonvulsants that relieve pain in certain neuralgias may succeed by acting peripherally, rather than in the central nervous system, he proposes the possibility of synthesizing a phenytoin-like compound that does not cross the blood-brain barrier, thereby avoiding the central depressor action that currently limits feasible doses of anticonvulsants [300]. Drugs that suppress ectopic discharges and act as membrane stabilizers, agents that suppress sodium conductance and open potassium channels, and their site and method of delivery should be investigated. Therapeutic agents that directly target the central sensitization process, rather than the peripheral nervous system factors that trigger and maintain pain, may prove to be useful [65]. Other novel suggestions for therapy still in experimental stages involve suicide retrograde axoplasmic transport. Ricin, a toxic lectin from castor beans, when applied to a peripheral nerve by retrograde transport, will kill the parent neuron [301]. When ricin was applied directly to the proximal stump of the mental and supraorbital nerves in the rat, within days the trigeminal ganglion cells developed diffuse chromatolysis and dissolution of cell bodies. Resultant Wallerian degeneration could be traced within the brainstem, suggesting the usefulness of this method in controlling various pain conditions without requiring direct attack on the target structures [302]. Newer neurosurgical strategies for making lesions in the trigeminal brainstem complex have evolved since the original Sjo¨qvist medullary tractotomy [89,303]. An open procedure with upper cervical laminectomy has been developed to permit radiofrequency ablation of the so-called DREZ. This technique, meant to destroy the spinal tract and the adjacent nucleus caudalis in the upper cervical spinal cord, was developed by
2405
2406
140
The pathophysiology of trigeminal neuralgia
Nashold and coworkers for treatment of neuropathic pain, including refractory TN and postsurgical anesthesia dolorosa [304]. Kanpolat and coworkers have developed a percutaneous CT-guided stereotactic procedure for trigeminal tractotomy-nucleotomy to accomplish radiofrequency ablation of the upper cervical spinal tract and oral pole of the caudal nucleus of the trigeminal nerve [305]. He has used this technique to treat vagoglossopharyngeal and geniculate (nervus intermedius) neuralgias [306] and for atypical facial pain [307]. These procedures might also be useful to treat postsurgical trigeminal deafferentation pain (anesthesia dolorosa). Further efforts are needed to develop effective treatments for deafferentation pain, fortunately uncommon after ablative surgery for TN. Neural transplantation is another promising field of investigation for pain treatment. Backlund et al. [308] performed the first human treatments with striatal implants of adrenal medullary autographs for parkinsonism [308]. More recent experimental studies in animals include Schwann cell cultures and embryonic neural grafts to accelerate axonal growth and myelination and endothelial implants of chromaffin cells from the adrenal medulla for chemical pain modulation [309,310]. Outcomes research employing principles of evidence-based medicine has become increasingly important to assess risks and benefits from current and new methods of medical and surgical treatment of TN. Accurate evaluation of the effectiveness of treatment methods will become reflected in our knowledge of the pathophysiology of TN. The need for this research, and suggested methods to accomplish it, have been published [311–313].
Conclusions In this chapter, an effort has been made to distinguish two different groups of patients
exhibiting the clinical syndrome of TN. The symptomatic group consists of patients with a demonstrable cause for their pain, such as MS, vascular root compression, or tumors; those in the idiopathic group have a yet unknown etiology for their pain. Epidemiological studies project at least 21,000 new cases will occur in 2008 in the United States. Pathologic evidence of structural nerve abnormalities in TN, in the limited number of cases studied, has shown that throughout the trigeminal system there are areas of disordered myelination, demyelinated axons in close apposition, and tortuous hypertrophic axons not characteristic of other diseases. Plaques of demyelination in MS located in the trigeminal brainstem complex and proximate REZ have been inconsistently observed in the few specimens examined. Neurophysiological studies, mostly experimental, have provided evidence of abnormalities in impulse generation, transmission, and central processing. These abnormalities include ectopic discharges, electrical ephaptic and chemical cross-talk among axons, central sensitization and other abnormal mechanisms leading to theories, such as the “ignition hypothesis,” to provide an explanation for the pathogenesis of TN. Clinical observations about the distinctive symptoms, natural history, and responses to pharmacologic agents and surgical procedures have complemented our knowledge about pathophysiology of TN. It is likely that both peripheral and central mechanisms are involved in the pathogenesis of TN, and the contributions of each to the trigeminal pain syndrome await further studies. As the search for a better understanding of the pathophysiology of TN and other painful conditions continues, the wisdom of Professor Wall paraphrased here may be pertinent. In a commentary he quotes Rutherford, who said “There are only two types of scientists: butterfly collectors and physicists.” In the challenging study of pain, Wall believes clinical and basic neuroscientists play an interdependent role and
The pathophysiology of trigeminal neuralgia
the job of the clinician is to collect and describe phenomena of the real world, not to pervert the facts to fit some popular theory. He proposes clinicians and basic scientists cultivate a mutual feeling and encouragement for the benefit of the patients, instead of attempting to sustain one theory over another [314]. From a clinical perspective, there are several features of TN that need further explanation. A major unsolved problem in pathophysiology is the common occurrence of repeated spontaneous remissions. Similarly, the frequently observed lifelong confinement of neuralgia to a single division, especially the maxillary division, remains a puzzle. Based on the functional anatomic evidence of somatotopic organization within the entire trigeminal system, it would be almost impossible strategically to create a lesion, except peripherally, affecting only the middle division of the nerve, whether by vascular compressive injury or some other mechanism. In the case of idiopathic typical TN, the presence, at least in some cases, of a very subtle sensory loss is poorly understood, as is the significance of this form of partial deafferentation in the TN syndrome. Partial or total surgical deafferentation often relieves pain of TN, but inexplicably the same maneuvers very infrequently can result in a new and different pain (anesthesia/analgesia dolorosa). Another related subject that continues to defy explanation is the fact that trauma to the trigeminal nerve can sometimes cause, and at other times relieve, pain of TN. Mechanical compression of the roots by vascular compression or tumor, stretching of the roots by arachnoidal adhesions, space-taking lesions, a high petrous bone, and sagging of the brain with aging are all putative causes of TN. The REZ has often been implicated as the site of root injury, but there is no study documenting that the REZ is more “vulnerable” to trauma. Even in the case of neurovascular conflicts, the locus of compression has sometimes been shown to lie a significant
140
distance from the REZ. TN pain can often be relieved by the tactic of MVD or removal of a compressive or nerve-stretching tumor; on the other hand, there is abundant evidence that mechanical trauma itself can relieve pain. For example, during MVD the root may be subjected to trauma. Observers have noted that pain relief may follow more rapidly if the nerve is gently traumatized during MVD, and somatosensory testing has demonstrated subtle sensory deficits not present before surgery. Sweet often reminded us of this paradoxical effect from surgery. He noted that just as many TN patients have been relieved of pain for long periods after decompression of the ganglion and roots in the middle (and posterior) fossa, equally important have been the long-term good results after surgical procedures to mildly compress or traumatize the ganglion and roots [28]. Current and future research in basic and clinical neurosciences will unravel many of the mysteries of TN, paving the way for improvements in treatment of this dreadful disease. At the same time, investigations concerned with TN should help us learn more about management of other disabling chronic pain conditions for which so little effective treatment is now available. With success in this endeavor, we should be able to eliminate the term idiopathic, and, most important, eliminate the pain and suffering of our patients with TN. To further illustrate the persisting enigma surrounding the pathophysiology of this disease, I would like to close this chapter with a personal anecdote about one of my patients who had idiopathic typical TN. Twelve years ago, I successfully treated her surgically using the Sweet procedure. At that time, she was a 65-year old, recently retired project manager for NASA during the Apollo Moon-Landing Program. She was referred to me after extensive consultations, imaging studies, and multiple dental procedures. The customary pharmacologic therapies with escalating doses of carbamazepine caused many
2407
2408
140
The pathophysiology of trigeminal neuralgia
unwanted side effects and eventually failed to give her significant pain relief. After 5 years of treatment by specialists, the referring physician, a family practitioner in suburban Houston, was the first to tell her that she had idiopathic TN. She asked him what the word idiopathic meant, and he told her that: “I am an idiot for how little I understand about it” and the second component meant “it is pathetic that you have this terrible pain.”
Acknowledgments The author wishes to express his appreciation to Andrew F. Hurt for his valuable assistance in the preparation of this manuscript and to the Neurosurgical Center for Pain Treatment and Research for its support.
References 1. Stookey B, Ransohoff J. Trigeminal neuralgia. Its history and treatment. Springfield: Charles C. Thomas; 1959. 2. White JC, Sweet WH. Pain and the neurosurgeon: a fortyyear experience. Springfield: Charles C. Thomas; 1969. 3. Symonds C. Facial pain. Ann R Coll Surg Engl 1949;4:206-12. 4. Dandy WE. Certain functions of the roots and ganglia of the cranial sensory nerves. Arch Neurol Psychiatry 1932;27:22-9. 5. Dandy WE. The brain. In: Lewis D, editor. Practice of surgery, vol. 12. Hagerstown: W.F. Prior; 1932. p. 1-682. 6. American Medical Association. International classification of diseases, 9th revision. Ingenix; 2006. 7. Headache Classification Subcommittee of the International Headache Society. The international classification of headache disorders. 2nd ed. Cephalalgia 2004;24 (Suppl. 1):190–207. 8. Merskey H, Bogduk N, editors. Classification of chronic pain: Description of chronic pain syndromes and definitions of pain terms. 2nd ed. IASP Press; Seattle: p. 59-68. 9. Jannetta PJ. Arterial compression of the trigeminal nerve at the pons in patients with trigeminal neuralgia. J Neurosurg 1967;26(Suppl.):159-62. 10. Burchiel KJ. A new classification for facial pain. Neurosurgery 2003;53:1164-6; discussion 1166‐7. 11. Devor, M. Commentary. In: Burchiel K, editor. Surgical management of pain. New York: Thieme; 2002. p. 38-41.
12. Gouda JJ, Brown JA. Atypical facial pain and other pain syndromes. Differential diagnosis and treatment. Neurosurg Clin N Am 1997;8:87-100. 13. Tasker RR. Deafferentiation. In: Wall PD, Melzack R, editors. Textbook of pain. New York: Churchill Livingstone; 1984. p. 119-32. 14. Willis WD. The pain system. The neural basis of nociceptive transmission in the mammalian nervous system. New York: Karger; 1985. 15. Head H, Rivers WHR, Sherren J. The afferent nervous system from a new aspect. Brain 1905;28:99-115. 16. Dworkin RH, Backonja M, Rowbotham MC, et al. Advances in neuropathic pain: diagnosis, mechanisms, and treatment recommendations. Arch Neurol 2003;60:1524-34. 17. Woolf CJ, Bennett GJ, Doherty M, et al. Towards a mechanism-based classification of pain? Pain 1998;77:227-9. 18. De Marco KJ, Hesselink JR. Trigeminal neuropathy. Neurosurg Clin N Am 1997;8:103-30. 19. Selby G. Diseases of the fifth cranial nerve. In: Dyck PJ, Thomas PK, Lambert EH, editors. Peripheral neuropathy, vol. 1. Philadelphia, PA: WB Saunders Company; 1975. p. 533-69. 20. Bullit E, Tew JM, Boyd J. Intracranial tumors in patients with facial pain. J Neurosurg 1986;64:865-71. 21. Dandy WE. Concerning the cause of trigeminal neuralgia. Am J Surg 1934;24:447-55. 22. Sunderland S. Cranial nerve injury. Structural and pathophysiological considerations and a classification of nerve injury. In: Samii M, Jannetta PJ, editors. The cranial nerves. Berlin: Springer; 1981. p. 16-23. 23. Haddad FS, Taha JM. An unusual cause for trigeminal neuralgia: contralateral meningioma of the posterior fossa. Neurosurgery 1990;26:1033-8. 24. Iwasaki K, Kondo A, Otsuka S, et al. Painful tic convulsif caused by a brain tumor: case report and review of the literature. Neurosurgery 1992;30:916-9. 25. Matsuura N, Kondo A. Trigeminal neuralgia and hemifacial spasm as false localizing signs in patients with a contralateral mass of the posterior cranial fossa. Report of three cases. J Neurosurg 1996;84:1067-71. 26. Grigoryan YA, Onopchenko CV. Persistent trigeminal neuralgia after removal of contralateral posterior cranial fossa tumor. Report of two cases. Surg Neurol 1999;52:56-60; discussion 60‐1. 27. Cappabianca P, Mariniello G, Alfieri A, et al. Trigeminal neuralgia and contralateral mass. J Neurosurg 1997;86:171-2. 28. Sweet WH. The pathophysiology of trigeminal neuralgia. In: Gildenberg PL, Tasker RR, editors. Textbook of stereotactic and functional neurosurgery. New York: McGraw-Hill; 1998. p. 1667-82. 29. Resnick DK, Jannetta PJ, Lunsford LD, et al. Microvascular decompression for trigeminal neuralgia
The pathophysiology of trigeminal neuralgia
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41. 42. 43.
44. 45. 46.
in patients with multiple sclerosis. Surg Neurol 1996;46:358-61; discussion 361‐2. Eldridge PR, Sinha AK, Javadpour M, et al. Microvascular decompression for trigeminal neuralgia in patients with multiple sclerosis. Stereotact Funct Neurosurg 2003;81:57-64. Athanasiou TC, Patel NK, Renowden SA, et al. Some patients with multiple sclerosis have neurovascular compression causing their trigeminal neuralgia and can be treated effectively with MVD: report of five cases. Br J Neurosurg 2005;19:463-8. De Matas M, Francis P, Miles JB. Microvascular decompression for trigeminal neuralgia in Charcot–Marie– Tooth disease. J Neurosurg 2000;92:715-7. Cheng TM, Cascino TL, Onofrio BM. Comprehensive study of diagnosis and treatment of trigeminal neuralgia secondary to tumors. Neurology 1993; 43:2298-302. Kurland LT. Descriptive epidemiology of selected neurologic and myopathic disorders with particular reference to a survey in Rochester, Minnesota. J Chronic Dis 1958;8:378-418. Katusic S, Beard CM, Bergstralh E, et al. Incidence and clinical features of trigeminal neuralgia, Rochester, Minnesota, 1945–1984. Ann Neurol 1990;27:89-95. Brewis M, Poskanzer DC, Rolland C, et al. Neurological disease in an English city. Acta Neurol Scand 1966;42 (Suppl. 24):1-89. Penman J. Trigeminal neuralgia. In: Vinken PJ, Bruyn GW, editors. Handbook of clinical neurology: headaches and clinical neuralgias, vol. 5. Amsterdam: NorthHolland;1968. p. 296-322. Rushton JG, Olafson RA. Trigeminal neuralgia associated with multiple sclerosis: report of 35 cases. Arch Neurol 1965;13:383-6. Resnick DK, Levy EI, Jannetta PJ. Microvascular decompression for pediatric onset trigeminal neuralgia. Neurosurgery 1998;43:804-7; discussion 807‐8. Mason WE, Kollros P, Jannetta PJ. Trigeminal neuralgia and its treatment in a 13-month-old child: a review and case report. J Craniomandib Dis 1991;5: 213-6. Rothman KJ, Monson RR. Epidemiology of trigeminal neuralgia. J Chronic Dis 1973;26:3-12. Rothman KJ, Beckman TM. Epidemiological evidence for two types of trigeminal neuralgia. Lancet 1974;1:7-9. Harris W. An analysis of 1,433 cases of paroxysmal trigeminal neuralgia (trigeminal-tic) and the end-results of gasserian alcohol injection. Brain 1940;63:209-24. Rothman KJ, Wepsic JG. Side of facial pain in trigeminal neuralgia. J Neurosurg 1974;40:514-23. Rothman KJ, Monson RR. Survival in trigeminal neuralgia. J Chronic Dis 1973;26:303-9. Coffey RJ, Fromm GH. Familial trigeminal neuralgia and Charcot–Marie–Tooth neuropathy. Report of two families and review. Surg Neurol 1991;35:49-53.
140
47. Manzoni GC, Torelli P. Epidemiology of typical and atypical craniofacial neuralgias. Neurol Sci 2005;26 (Suppl. 2):65-7. 48. Jensen TS, Rasmussen P, Reske-Nielsen E. Association of trigeminal neuralgia with multiple sclerosis: clinical and pathological features. Acta Neurol Scand 1982;65:182-9. 49. Hooge JP, Redekop WK. Trigeminal neuralgia in multiple sclerosis. Neurology 1995;45:1294-6. 50. Auld AW, Buermann A. Trigeminal neuralgia in six members of one generation. Arch Neurol 1965;13:194. 51. Kirkpatrick DB. Familial trigeminal neuralgia: case report. Neurosurgery 1989;24:758-61. 52. Fleetwood IG, Innes AM, Hansen SR, et al. Familial trigeminal neuralgia. Case report and review of the literature. J Neurosurg 2001;95:513-7. 53. Smyth P, Greenough G, Stommel E. Familial trigeminal neuralgia: case reports and review of the literature. Headache 2003;43:910-5. 54. Adams RD, Victor M. Principles of neurology. New York: McGraw-Hill; 1977. 55. Frohman EM, Racke MK, Raine CS. Multiple sclerosis – the plaque and its pathogenesis. N Engl J Med 2006;354:942-55. 56. Testa D, Milanese C, La Mantia L, et al. Familial trigeminal neuralgia in Charcot–Marie–Tooth disease. J Neurol 1981;225:283-7. 57. Knuckey NW, Gubbay SS. Familial trigeminal and glossopharyngeal neuralgia. Clin Exp Neurol 1979;16:315-9. 58. Duff JM, Spinner RJ, Lindor NM, et al. Familial trigeminal neuralgia and contralateral hemifacial spasm. Neurology 1999;53:216-8. 59. Baxter DW, Olszewski J. Congenital universal insensitivity to pain. Brain 1960;83:381-93. 60. Sweet WH. Animal models of chronic pain: their possible validation from human experience with posterior rhizotomy and congenital analgesia (Part I of the second John J. Bonica lecture). Pain 1981;10:275-95. 61. Hurt RW, Ballantine HT. Stereotactic anterior cingulate lesions for persistent pain: a report on 68 cases. Clin Neurosurg 1974;21:334-51. 62. Devor M, Raber P. Heritability of symptoms in an experimental model of neuropathic pain. Pain 1990;42:51-67. 63. Inbal R, Devor M, Tuchendler O, et al. Autotomy following nerve injury: genetic factors in the development of chronic pain. Pain 1980;9:327-37. 64. Mogil JS, Wilson SG, Bon K, et al. Heritability of nociception II. “Types” of nociception revealed by genetic correlation analysis. Pain 1999;80:83-93. 65. Devor M. Response of nerves to injury in relation to neuropathic pain. In: McMahon SB, Koltzenburg, editors. Wall and Melzack’s textbook of pain. 5th ed. New York: Churchill Livingstone; 2006. p. 905-27. 66. Nauta WJH, Ebbesson SOE. Contemporary Research Methods in Neuroanatomy. New York: Springer; 1970.
2409
2410
140
The pathophysiology of trigeminal neuralgia
67. Nauta WJH, Feirtag M, editors. Fundamental neuroanatomy. New York: WH Freeman and Co.; 1986. 68. Maxwell DS. Fine structure of the normal trigeminal ganglion in the cat and monkey. J Neurosurg 1967;26 (Suppl.):127-31. 69. Moses HL, Beaver DL, Ganote CE. Electron microscopy of the trigeminal ganglion. I. Comparative ultrastructure. Arch Pathol 1965;79:541-56. 70. Beaver DL, Moses HL, Ganote CE. Electron microscopy of the trigeminal ganglion. II. Autopsy study of human ganglia. Arch Pathol 1965;79:557-70. 71. Moses HL. Comparative fine structure of the trigeminal ganglia, including human autopsy studies. J Neurosurg 1967;26(Suppl.):112-26. 72. Young RF. Unmyelinated fibers in the trigeminal motor root. Possible relationship to the results of trigeminal rhizotomy. J Neurosurg 1978;49:538-43. 73. Kumagami H. Neuropathological findings of hemifacial spasm and trigeminal neuralgia. Arch Otolaryngol 1974;99:160-4. 74. Beaver DL. Electron microscopy of the gasserian ganglion in trigeminal neuralgia. J Neurosurg 1967;26 (Suppl.):138-50. 75. Beaver DL, Moses HL, Ganote CE. Electron microscopy of the trigeminal ganglion. III. Trigeminal neuralgia. Arch Pathol 1965;79:571-82. 76. Kerr FW. Pathology of trigeminal neuralgia: light and electron microscopic observations. J Neurosurg 1967;26 (Suppl.):151-6. 77. Kerr FW, Miller RH. The pathology of trigeminal neuralgia. Electron microscopic studies. Arch Neurol 1966; 15:308-19. 78. Ferner H. The anatomy of the trigeminal root and the gasserian ganglion and their relations to the cerebral meninges. In: Hassler R, Walker EA, editors. Trigeminal Neuralgia: pathogenesis and Pathophysiology. Stuttgart: Georg Thieme Verlag; 1970. p. 1-6. 79. Hilton DA, Love S, Gradidge T, et al. Pathological findings associated with trigeminal neuralgia caused by vascular compression. Neurosurgery 1994;35:299-303; discussion 303. 80. Love S, Hilton DA, Coakham HB. Central demyelination of the Vth nerve root in trigeminal neuralgia associated with vascular compression. Brain Pathol 1998;8:1-11; discussion 11-12. 81. Love S, Coakham HB. Trigeminal neuralgia: pathology and pathogenesis. Brain 2001;124:2347-60. 82. Rappaport ZH, Govrin-Lippmann R, Devor M. An electron-microscopic analysis of biopsy samples of the trigeminal root taken during microvascular decompressive surgery. Stereotact Funct Neurosurg 1997;68:182-6. 83. Devor M, Govrin-Lippmann R, Rappaport ZH. Mechanism of trigeminal neuralgia: an ultrastructural analysis of trigeminal root specimens obtained during microvascular decompression surgery. J Neurosurg 2002; 96:532-43.
84. Devor M, Govrin-Lippmann R, Rappaport ZH, et al. Cranial root injury in glossopharyngeal neuralgia: electron microscopic observations. Case report. J Neurosurg 2002;96:603-6. 85. Ramon Y, Cajal S. Histologie du syste`me nerveux de l’homme et des verte´bre´s. Paris: A. Maloine; 1909. 86. Windle W. Non-bifurcating nerve fibers of the trigeminal nerve. J Comp Neurol 1926;40:229-40. 87. Olszewski J. On the anatomical and functional organization of the spinal trigeminal nucleus. J Comp Neurol 1950;92:401-13. 88. Wall PD, Taub A. Four aspects of trigeminal nucleus and a paradox. J Neurophysiol 1962;25:110-26. 89. Sjo¨qvist O. Studies on pain conduction in the trigeminal nerve. Acta Psych Neurol Scand 1938;17: (Suppl.):1-139. 90. Kunc Z, Fuchsova´ M, Nova´k M. Excision of the spinal trigeminal tract. Electron microscopy. Acta Neurochir (Wien) 1978;41:233-41. 91. Heimer L. The human brain and spinal cord: functional neuroanatomy and dissection guide. 2nd ed. New York: Springer; 1995. 92. Adams RD, Kubik CS. The morbid anatomy of the demyelinative disease. Am J Med 1952;12:510-46. 93. Lazar ML, Kirkpatrick JB. Trigeminal neuralgia and multiple sclerosis: demonstration of the plaque in an operative case. Neurosurgery 1979;5:711-7. 94. Spiegel EA. Heinrich Obersteiner (1847–1922). In: Haymaker W, Schiller F, editors. The founders of neurology. Springfield: Charles C. Thomas; 1970. p. 355-8. 95. Tarlov IM. Structure of the nerve root: Part I. – Nature of the junction between the central and the peripheral nervous system. Arch Neurol Psychiatry 1937;37: 555-83. 96. Dandy WE. An operation for the cure of tic douloureux: partial section of the sensory root at the pons. Arch Surg 1929;18:687-734. 97. Jannetta PJ, Rand RW. Transtentorial retrogasserian rhizotomy in trigeminal neuralgia by microneurosurgical technique. Bull Los Angeles Neurol Soc 1966; 31:93-9. 98. Maxwell DS, Kruger L, Pineda A. The trigeminal nerve root with special reference to the central-peripheral transition zone: an electron microscopic study in the macaque. Anat Rec 1969;164:113-25. 99. Gudmundsson K, Rhoton AL, Rushton JG. Detailed anatomy of the intracranial portion of the trigeminal nerve. J Neurosurg 1971;35:592-600. 100. Soeira G, Abd El-Bary TH, Dujovny M, et al. Microsurgical anatomy of the trigeminal nerve. Neurol Res 1994;16:273-83. 101. Lang J, Reiter U. Intracisternal length of the trigeminal nerve. Neurochirurgia (Stuttg) 1984;27:159-61. 102. Peker S, Kurtkaya O, Uzu¨n I, et al. Microanatomy of the central myelin-peripheral myelin transition zone
The pathophysiology of trigeminal neuralgia
103.
104.
105.
106.
107. 108.
109.
110.
111. 112. 113.
114.
115.
116.
117.
118.
119.
of the trigeminal nerve. Neurosurgery 2006;59:354-9; discussion 354‐9. De Ridder D, Møller A, Verlooy J, et al. Is the root entry/ exit zone important in microvascular compression syndromes? Neurosurgery 2002;51:427-33; discussion 433‐4. Ziyal IM, Sekhar LN, Ozgen T, et al. The trigeminal nerve and ganglion: an anatomical, histological, and radiological study addressing the transtrigeminal approach. Surg Neurol 2004;61:564-73, discussion 573‐4. McLaughlin MR, Jannetta PJ, Clyde BL, et al. Microvascular decompression of cranial nerves: lessons learned after 4400 operations. J Neurosurg 1999;90:1-8. Ziyal IM, Ozgen T. Microanatomy of the central myelinperipheral myelin transition zone of the trigeminal nerve. Neurosurgery 2007;60:E582; author reply E582. Sindou M. Comments. Neurosurgery 2002;51:433. Tomii M, Onoue H, Yasue M, et al. Microscopic measurement of the facial nerve root exit zone from central glial myelin to peripheral Schwann cell myelin. J Neurosurg 2003;99:121-4. Toma JS, McPhail LT, Ramer MS. Comparative postnatal development of spinal, trigeminal and vagal sensory root entry zones. Int J Dev Neurosci 2006;24:373-88. Ramon Y, Cajal S. Degeneration and regeneration of the nervous system, vol. 1. London: Oxford University Press; 1928. Dandy WE. The treatment of trigeminal neuralgia by the cerebellar route. Ann Surg 1932;96:787-95. Olivecrona H. Die trigeminusneuralgie und Ihre behandlung. Der Nervenarzt 1941;14:49-57. Lazorthes G, Soujeole AC, Espagno J. Note sur les vaisseaux de l’angle ponto-ce´re´belleux: Variations et rapports avec la racine du trijumeau. Compte Rendues de l’Association des Anatomistes, 36e re´union. Lyon 1949;437–8. Sunderland S. Neurovascular relations and anomalies at the base of the brain. J Neurol Neurosurg Psychiatry 1948;11:243-57. Gardner WJ, Miklos MV. Response of trigeminal neuralgia to decompression of sensory root; discussion of cause of trigeminal neuralgia. J Am Med Assoc 1959;170:1773-6. Gardner WJ. Concerning the mechanism of trigeminal neuralgia and hemifacial spasm. J Neurosurg 1962; 19:947-58. Jannetta PJ. Vascular compression in trigeminal neuralgia. In: Samii M, Jannetta PJ, editors. The cranial nerves. Berlin: Springer; 1981. p. 331-40. Jannetta PJ. Microvascular decompression of the trigeminal nerve root entry zone. In: Rovit RL, Murali R, Jannetta PJ, editors. Trigeminal neuralgia. Baltimore: Williams & Wilkins; 1990. p. 201-22. Sindou M, Howeidy T, Acevedo G. Anatomical observations during microvascular decompression for idiopathic trigeminal neuralgia (with correlations between topography of pain and site of the neurovascular
120.
121.
122.
123. 124.
125.
126.
127.
128.
129.
130.
131.
132.
133. 134.
140
conflict). Prospective study in a series of 579 patients. Acta Neurochir (Wien) 1994;144:1‐12; discussion 12‐3. Hardy DG, Rhoton AL. Microsurgical relationships of the superior cerebellar artery and the trigeminal nerve. J Neurosurg 1978;49:669-78. Rhoton AL. Microsurgical anatomy of decompression operations on the trigeminal nerve. In: Rovit RL, Murali R, Jannetta PJ, editors. Trigeminal neuralgia. Baltimore: Williams and Wilkins; 1990. p. 165-200. Haines SJ, Jannetta PJ, Zorub DS. Microvascular relations of the trigeminal nerve. An anatomical study with clinical correlation. J Neurosurg 1980;52:381-6. Mehta L, Fatani J, Rao G. Superior cerebellar artery in the pontine zone. Saudi Med J 1981;2:213-6. Klun B, Prestor B. Microvascular relations of the trigeminal nerve: an anatomical study. Neurosurgery 1986; 19:535-9. Hamlyn PJ, King TT. Neurovascular compression in trigeminal neuralgia: a clinical and anatomical study. J Neurosurg 1992;76:948-54. Jannetta PJ. Treatment of trigeminal neuralgia by microoperative decompression. In: Youmans JR, editor. Neurological surgery, vol. 6. Philadelphia, PA: WB Saunders; 1982. p. 3589-603. Adams CBT, Kaye AH, Teddy PJ. The treatment of trigeminal neuralgia by posterior fossa microsurgery. J Neurol Neurosurg Psychiatry 1982;45:1020-6. Piatt JH, Wilkins RH. Treatment of tic douloureux and hemifacial spasm by posterior fossa exploration: therapeutic implications of various neurovascular relationships. Neurosurgery 1984;14:462-71. Young JN, Wilkins RH. Partial sensory trigeminal rhizotomy at the pons for trigeminal neuralgia. J Neurosurg 1993;79:680-7. Hamlyn PJ. Neurovascular relationships in the posterior cranial fossa, with special reference to trigeminal neuralgia. 1. Review of the literature and development of a new method of vascular injection-filling in cadaveric controls. Clin Anat 1997;10:371-9. Sindou M, Amrani F, Mertens P. Does microsurgical vascular decompression for trigeminal neuralgia work through a neo-compressive mechanism? Anatomicalsurgical evidence for a decompressive effect. Acta Neurochir Suppl (Wien) 1991;52:127-9. Barker FG, Jannetta PJ, Bissonette DJ, et al. The longterm outcome of microvascular decompression for trigeminal neuralgia. N Engl J Med 1996;334:1077-83. Taarnhøj P. Decompression of the trigeminal root. J Neurosurg 1954;11:299-305. Sindou M, Leston J, Decullier E, et al. Microvascular decompression for primary trigeminal neuralgia: longterm effectiveness and prognostic factors in a series of 362 consecutive patients with clear-cut neurovascular conflicts who underwent pure decompression. J Neurosurg 2007;107:1144-53.
2411
2412
140
The pathophysiology of trigeminal neuralgia
135. Meaney JF, Eldridge PR, Dunn LT, et al. Demonstration of neurovascular compression in trigeminal neuralgia with magnetic resonance imaging. Comparison with surgical findings in 52 consecutive operative cases. J Neurosurg 1995;83:799-805. 136. Jawahar A, Kondziolka D, Kanal E, et al. Imaging the trigeminal nerve and pons before and after surgical intervention for trigeminal neuralgia. Neurosurgery 2001;48:101-6; discussion 106‐7. 137. Cha´vez GD, De Salles AA, Solberg TD, et al. Threedimensional fast imaging employing steady-state acquisition magnetic resonance imaging for stereotactic radiosurgery of trigeminal neuralgia. Neurosurgery 2005;56:E628; discussion E628. 138. Erbay SH, Bhadelia RA, Riesenburger R, et al. Association between neurovascular contact on MRI and response to gamma knife radiosurgery in trigeminal neuralgia. Neuroradiology 2006;48:26-30. 139. Tanrikulu L, Hastreiter P, Troescher-Weber R, et al. Intraoperative three-dimensional visualization in microvascular decompression. J Neurosurg 2007; 107:1137-43. 140. Yousry I, Moriggl B, Holtmannspoetter M, et al. Detailed anatomy of the motor and sensory roots of the trigeminal nerve and their neurovascular relationships: a magnetic resonance imaging study. J Neurosurg 2004;101:427-34. 141. Adamczy KM, Bulski T, Sowin´ska J, et al. Trigeminal nerve – artery contact in people without trigeminal neuralgia – MR study. Med Sci Monit 2007;13 (Suppl. 1):38-43. 142. Peker S, Dincer A, Pamir NM. Vascular compression of the trigeminal nerve is a frequent finding in asymptomatic individuals: 3-Tesla MR imaging of 200 trigeminal nerves using 3D CISS sequences. Neurosurgery, in press. 143. Apfelbaum RI. Surgical management of disorders of the lower cranial nerves. In: Schmidek HH, Sweet WH, editors. Operative neurosurgical techniques. 2nd ed. New York: Grune & Stratton; 1988. p. 1097-109. 144. Adams CBT. Microvascular compression: an alternative view and hypothesis. J Neurosurg 1989;70:1-12. 145. Morley, TP. Current controversies in neurosurgery. Philadelphia, PA: WB Saunders; 1976. 146. Apfelbaum RI. Surgery for tic douloureux. Clin Neurosurg 1983;31:351-68. 147. Zorman G, Wilson CB. Outcome following microsurgical vascular decompression or partial sensory rhizotomy in 125 cases of trigeminal neuralgia. Neurology 1984;34:1362-5. 148. Burchiel KJ, Clarke H, Haglund M, et al. Long-term efficacy of microvascular decompression in trigeminal neuralgia. J Neurosurg 1988;69:35-8. 149. Bederson JB, Wilson CB. Evaluation of microvascular decompression and partial sensory rhizotomy in 252 cases of trigeminal neuralgia. J Neurosurg 1989; 71:359-67.
150. Adams CBT. The physiology and pathophysiology of the posterior fossa cranial nerve dysfunction syndromes: nonmicrovascular perspective. In: Bartlow D, editor. Surgery of the cranial nerves of the posterior fossa. Park Ridge: American Association of Neurological Surgeons; 1993. p. 131-54. 151. Darian-Smith I, Mutton P, Proctor R. Functional organization of tactile cutaneous afferents within the semilunar ganglion and trigeminal spinal tract of the cat. J Neurophysiol 1965;28:682-94. 152. Lende RA, Poulos DA. Functional localization in the trigeminal ganglion in the monkey. J Neurosurg 1970; 32:336-43. 153. Spiller WG, Frazier CH. Tic douloureux. Anatomic and clinical basis for subtotal section of sensory root of trigeminal nerve. Arch Neurol Psychiatry 1933;29:50-5. 154. Marfurt CF. The somatotopic organization of the cat trigeminal ganglion as determined by the horseradish peroxidase technique. Anat Rec 1981;201:105-18. 155. Frazier CH. Subtotal resection of sensory root for relief of major trigeminal neuralgia. Arch Neurol Psychiatry 1925;13:378-84. 156. Davis L, Haven HA. Surgical anatomy of the sensory root of the trigeminal nerve. Arch Neurol Psychiatry 1933;29:1-18. 157. Kerr FW. The etiology of trigeminal neuralgia. Arch Neurol 1963;8:15-25. 158. Troland CE, Otenasek FJ. Selected writings of Walter E. Dandy. Springfield: Charles C. Thomas; 1957. 159. Saunders RL, Sachs E. Relation of the accessory rootlets of the trigeminal nerve to its motor root. A microsurgical autopsy study. J Neurosurg 1970;33:317-24. 160. Pelletier VA, Poulos DA, Lende RA. Functional localization in the trigeminal root. J Neurosurg 1974; 40:504-13. 161. Young RF, King RB. Fiber spectrum of the trigeminal sensory root of the baboon determined by electron microscopy. J Neurosurg 1973;38:65-72. 162. Wilkins H, Sachs E. Variations in skin anesthesia following subtotal resection of the posterior root: with a report of twenty-six cases illustrating a series of variations. Arch Neurol Psychiatry 1933;29:19-49. 163. Morgan C, Jannetta PJ, deGroat WC. Organization of corneal afferent axons in the trigeminal nerve root entry zone in the cat. Exp Brain Res 1987;68:411-6. 164. Stechison MT, Møller A, Lovely TJ. Intraoperative mapping of the trigeminal nerve root: technique and application in the surgical management of facial pain. Neurosurgery 1996;38:76-81; discussion 81‐2. 165. Coggeshall RE, Applebaum ML, Fazen M, et al. Unmyelinated axons in human ventral roots, a possible explanation for the failure of dorsal rhizotomy to relieve pain. Brain 1975;98:157-66. 166. Young RF, Stevens R. Unmyelinated axons in the trigeminal motor root of human and cat. J Comp Neurol 1979;183:205-14.
The pathophysiology of trigeminal neuralgia
167. Young RF, Kruger L. Axonal transport studies of the trigeminal nerve roots of the cat. With special reference to afferent contributions to the portio minor. J Neurosurg 1981; 54:208-12. 168. May O, Horsley V. The mesencephalic root of the fifth nerve. Brain 1910;33:175-203. 169. Corbin KB, Harrison F. Function of mesencephalic root of fifth cranial nerve. J Neurophysiol 1940;3:423-35. 170. Denny-Brown D, Yanagisawa N. The function of the descending root of the fifth nerve. Brain 1973;96: 783-814. 171. Kerr FW. The divisional organization of afferent fibres of the trigeminal nerve. Brain 1963;86:721-32. 172. Emmons WF, Rhoton AJ. Subdivision of the trigeminal sensory root. Experimental study in the monkey. J Neurosurg 1971;35:585-91. 173. Kerr FW. The fine structure of the subnucleus caudalis of the trigeminal nerve. Brain Res 1970;23:129-45. 174. Nauta WJH, Kuypers HGJM. Some ascending pathways in the brainstem reticular formation. In: Jasper HH, editor. Reticular formation of the brain. Boston: Little, Brown and Co.; 1958. p. 3-30. 175. Stewart WA, King RB. Fiber projections from the nucleus caudalis of the spinal trigeminal nucleus. J Comp Neurol 1963;121:271-86. 176. Young RF. The trigeminal nerve and its central pathways. In: Rovit RL, Murali R, Jannetta PJ, editors. Trigeminal neuralgia. Baltimore: Williams and Wilkins; 1990. p. 27-51. 177. Shults RC. Nociceptive neural organization in the trigeminal nuclei. In: Light AR, editor. The initial processing of pain and its descending control: spinal and trigeminal systems. New York: Karger; 1992. p. 179-202. 178. Sherrington CS. The integrative action of the nervous system. New Haven: Yale University Press; 1906. 179. Collins WR, Nulsen FE, Randt CT. Relation of peripheral nerve fiber size and sensation in man. Arch Neurol 1960;3:381-5. 180. Devor M, Bernstein JJ. Abnormal impulse generation in neuromas: electrophysiology and ultrastructure. In: Culp WJ, Ochoa J, editors. Abnormal nerves and muscles as impulse generators. London: Oxford University Press; 1982. p. 363-80. 181. Wall PD, Gutnick M. Ongoing activity in peripheral nerves: the physiology and pharmacology of impulses originating from a neuroma. Exp Neurol 1974; 43:580-93. 182. Calvin WH, Devor M, Howe JF. Can neuralgias arise from minor demyelination? Spontaneous firing, mechanosensitivity, and afterdischarge from conducting axons. Exp Neurol 1982;75:755-63. 183. Howe JF, Loeser JD, Calvin WH. Mechanosensitivity of dorsal root ganglia and chronically injured axons: a physiological basis for the radicular pain of nerve root compression. Pain 1977;3:25-41.
140
184. Wall PD, Devor M. Sensory afferent impulses originate from dorsal root ganglia as well as from the periphery in normal and nerve injured rats. Pain 1983;17:321-39. 185. Burchiel KJ. Abnormal impulse generation in focally demyelinated trigeminal roots. J Neurosurg 1980; 53:674-83. 186. Hardy JD, Wolf HC, Goodell H. Pain sensations and reactions. New York: William and Wilkins; 1952. 187. Woolf CJ. Excitability changes in central neurons following peripheral damage: role of central sensitization in the pathogenesis of pain. In: Willis WD, editor. Hyperalgesia and allodynia. New York: Raven Press; 1992. p. 221-43. 188. Granit R, Leksell L, Skoglund CR. Fibre interactions in injured or compressed region of nerve. Brain 1944;67:125-40. 189. Seltzer Z, Devor M. Ephaptic transmission in chronically damaged peripheral nerves. Neurology 1979; 29:1061-4. 190. Amir R, Rappaport ZH, Devor M. Pain paroxysms. In: Schmidt RF, Willis WD, editors. Encyclopedia of pain. New York: Springer; 2007. p. 1728-32. 191. Lisney SJ, Devor M. Afterdischarge and interactions among fibers in damaged peripheral nerve in the rat. Brain Res 1987;415:122-36. 192. Amir R, Devor M. Chemically mediated cross-excitation in rat dorsal root ganglia. J Neurosci 1996;16:4733-41. 193. Turnbull IM, Black RG, Scott JW. Reflex efferent impulses in the trigeminal nerve. J Neurosurg 1961;18:746-52. 194. Calvin WH, Loeser JD, Howe JF. A neurophysiological theory for the pain mechanism of tic douloureux. Pain 1977;3:147-54. 195. Loeser JD, Calvin WH, Howe JF. Pathophysiology of trigeminal neuralgia. Clin Neurosurg 1977; 24:527-37. 196. Kerr FW. Trigeminal neuralgia: pathogenesis and description of a possible etiology for the “cryptogenic” variety. Trans Am Neurol Assoc 1962;87:118-23. 197. List CF, Williams JR. Pathogenesis of trigeminal neuralgia; a review. AMA Arch Neurol Psychiatry 1957;77:36-43. 198. Trousseau A. De la neuralgia epileptiforme. Arch Gen Med 1853;1:33-44. 199. Wilson SAK. Neurology, vol. 1. London: Edward Arnold & Co.; 1940. 200. Kugelberg E, Lindblom U. The mechanism of the pain in trigeminal neuralgia. J Neurol Neurosurg Psychiatry 1959;22:36-43. 201. King RB, Meagher JN. Studies of trigeminal nerve potentials. J Neurosurg 1955;12:393-402. 202. King RB, Barnett JC. Studies of trigeminal nerve potentials; overreaction to tactile facial stimulation in acute laboratory preparations. J Neurosurg 1957; 14:617-27.
2413
2414
140
The pathophysiology of trigeminal neuralgia
203. King RB, Meagher JN, Barnett JC. Studies of trigeminal nerve potentials in normal compared to abnormal experimental preparations. J Neurosurg 1956;13:176-83. 204. Stewart WA, Stoops WL, Pillone PR, et al. An electrophysiologic study of ascending pathways from nucleus caudalis of the spinal trigeminal nuclear complex. J Neurosurg 1964;21:35-48. 205. King RB. Evidence for a central etiology of tic douloureux. J Neurosurg 1967;26(Suppl.):175-80. 206. Wyburn-Mason R. The nature of tic douloureux: treatment by alcohol block or section of the great auricular nerve. BMJ 1953;2:119-22. 207. Crue BL, Shelden CH, Pudenz RH, et al. Observations on the pain and trigger mechanism in trigeminal neuralgia. Neurology 1956;6:196-207. 208. Crue BL, Carregal EJA. Postsynaptic repetitive neuron discharge in neuralgic pain. In: Bonica JJ, editor. Advances in neurology, vol. 4: international symposium on pain. New York: Raven Press; 1974. p. 643-9. 209. Pagni CA. The origin of tic douloureux: a unified view. J Neurosurg Sci 1993;37:185-94. 210. Cushing H. The major trigeminal neuralgias and their surgical treatment based on experiences with 332 Gasserian operations. I. Varieties of facial neuralgia. Am J Med Sci 1920;160:157-84. 211. Rushton JG, Stevens JC, Miller RH. Glossopharyngeal (vagoglossopharyngeal) neuralgia: a study of 217 cases. Arch Neurol 1981;38:201-5. 212. Kobata H, Kondo A, Iwasaki K, et al. Combined hyperactive dysfunction syndrome of the cranial nerves: trigeminal neuralgia, hemifacial spasm, and glossopharyngeal neuralgia: 11-year experience and review. Neurosurgery 1998;43:1351-61; discussion 1361‐2. 213. Thompson PD, Carroll WM. Hemimasticatory spasm – a peripheral paroxysmal cranial neuropathy? J Neurol Neurosurg Psychiatry 1983;46:274-6. 214. Møller AR. Cranial nerve dysfunction syndromes: pathophysiology of microvascular decompression. In: Bartlow D, editor. Surgery of the cranial nerves of the posterior fossa. Park Ridge: American Association of Neurological Surgeons; 1993. p. 105-29. 215. Møller AR. Vascular compression of cranial nerves: II: pathophysiology. Neurol Res 1999;21:439-43. 216. Dubner R, Sharav Y, Gracely RH, et al. Idiopathic trigeminal neuralgia: sensory features and pain mechanisms. Pain 1987;31:23-33. 217. Fromm GH, Chattha AS, Terrence CF, et al. Role of inhibitory mechanisms in trigeminal neuralgia. Neurology 1981;31:683-7. 218. Fromm GH, Sessle BJ. Trigeminal neuralgia: current concepts regarding pathogenesis and treatment. Boston, MA: Butterworth-Heinemann; 1991. 219. Fromm GH, Terrence CF, Maroon JC. Trigeminal neuralgia. Current concepts regarding etiology and pathogenesis. Arch Neurol 1984;41:1204-7.
220. Burchiel KJ. Carbamazepine inhibits spontaneous activity in experimental neuromas. Exp Neurol 1988l; 102:249-53. 221. Loeser JD, Ward AA, White LE. Chronic deafferentation of human spinal cord neurons. J Neurosurg 1968; 29:48-50. 222. Anderson LS, Black RG, Abraham J, et al. Neuronal hyperactivity in experimental trigeminal deafferentation. J Neurosurg 1971;35:444-52. 223. Westrum LE. Electron microscopy of deafferentation in the spinal trigeminal nucleus. In: Bonica JJ, editor. Advances in neurology, vol. 4: international symposium on pain. New York: Raven Press; 1974. p. 53-60. 224. Gorecki J, Hirayama T, Dostrovsky JO, et al. Thalamic stimulation and recording in patients with deafferentation and central pain. Stereotact Funct Neurosurg 1989;52:219-26. 225. Rinaldi PC, Young RF, Albe-Fessard D, et al. Spontaneous neuronal hyperactivity in the medial and intralaminar thalamic nuclei of patients with deafferentation pain. J Neurosurg 1991;74:415-21. 226. Sessle BJ. Effects of alterations in inputs to trigeminal brain stem neurons. In: Fromm GH, Sessle BJ, editors. Trigeminal neuralgia: current concepts regarding pathogenesis and treatment. Boston, MA: ButterworthHeinemann; 1991. p. 92-6. 227. Leandri M, Favale E. Diagnostic relevance of trigeminal evoked potentials following infraorbital nerve stimulation. J Neurosurg 1991;75:244-50. 228. Leandri M, Parodi CI, Favale E. Early trigeminal evoked potentials in tumours of the base of the skull and trigeminal neuralgia. Electroencephalogr Clin Neurophysiol 1988;71:114-24. 229. Singh N, Sachdev KK, Brisman R. Trigeminal nerve stimulation: short latency somatosensory evoked potentials. Neurology 1982;32:97-101. 230. Cruccu G, Inghilleri M, Manfredi M, et al. Intracranial stimulation of the trigeminal nerve in man. III. Sensory potentials. J Neurol Neurosurg Psychiatry 1987; 50:1323-30. 231. Karol EA, Sanz OP, Gonzalez La Riva FN, et al. A micrometric multiple electrode array for the exploration of gasserian and retrogasserian trigeminal fibers: preliminary report. Technical note. Neurosurgery 1993;33:154-8. 232. Leandri M, Gottlieb A. Trigeminal evoked potentialmonitored thermorhizotomy: a novel approach for relief of trigeminal pain. J Neurosurg 1996;84:929-39. 233. Soustiel JF, Chistyakov AV, Hafner H, et al. Intracranial recording from the brain-stem and the trigeminal nerve following upper lip stimulation. Electroencephalogr Clin Neurophysiol 1996;100:51-4. 234. Stechison MT. The trigeminal evoked potential: Part II. Intraoperative recording of short-latency responses. Neurosurgery 1993;33:639-43; discussion 643‐4.
The pathophysiology of trigeminal neuralgia
235. Stechison MT, Kralick FJ. The trigeminal evoked potential: Part I. Long-latency responses in awake or anesthetized subjects. Neurosurgery 1993;33:633-8. 236. Leandri M, Eldridge P, Miles J. Recovery of nerve conduction following microvascular decompression for trigeminal neuralgia. Neurology 1998;51:1641-6. 237. Cruccu G, Leandri M, Iannetti GD, et al. Small-fiber dysfunction in trigeminal neuralgia: carbamazepine effect on laser-evoked potentials. Neurology 2001;56:1722-6. 238. Sindou M, Fobe JL, Berthier E, et al. Facial motor responses evoked by direct electrical stimulation of the trigeminal root. Localizing value for radiofrequency thermorhizotomy. Acta Neurochir (Wien) 1994;128:57-67. 239. Romaniello A, Iannetti GD, Truini A, et al. Trigeminal responses to laser stimuli. Neurophysiol Clin 2003; 33:315-24. 240. Burchiel KJ, Baumann TK. Pathophysiology of trigeminal neuralgia: new evidence from a trigeminal ganglion intraoperative microneurographic recording. Case report. J Neurosurg 2004;101:872-3. 241. Ja¨a¨skela¨inen SK, Forssell H, Tenovuo O. Electrophysiological testing of the trigeminofacial system: aid in the diagnosis of atypical facial pain. Pain 1999; 80:191-200. 242. Manfredi M, Cruccu G. “Nociceptive” reflexes in the human trigeminal system. In: Lipton S, editor. Advances in pain research and therapy, vol. 13. New York: Raven Press, Ltd; 1990. p. 73-7. 243. Obermann M, Yoon MS, Ese D, et al. Impaired trigeminal nociceptive processing in patients with trigeminal neuralgia. Neurology 2007;69:835-41. 244. Lewy FH, Grant FC. Physiopathologic and pathoanatomic aspects of major trigeminal neuralgia. Arch Neurol Psychiatry 1938;40:1126-34. 245. Hampf G, Bowsher D, Wells C, et al. Sensory and autonomic measurements in idiopathic trigeminal neuralgia before and after radiofrequency thermocoagulation: differentiation from some other causes of facial pain. Pain 1990;40:241-8. 246. Nurmikko TJ. Altered cutaneous sensation in trigeminal neuralgia. Arch Neurol 1991;48:523-7. 247. Bowsher D. Trigeminal neuralgia: an anatomically oriented review. Clin Anat 1997;10:409-15. 248. Bowsher D, Miles JB, Haggett CE, et al. Trigeminal neuralgia: a quantitative sensory perception threshold study in patients who had not undergone previous invasive procedures. J Neurosurg 1997;86:190-2. 249. Miles JB, Eldridge PR, Haggett CE, et al. Sensory effects of microvascular decompression in trigeminal neuralgia. J Neurosurg 1997;86:193-6. 250. Eide PK, Stubhaug A. Relief of trigeminal neuralgia after percutaneous retrogasserian glycerol rhizolysis is dependent on normalization of abnormal temporal summation of pain, without general impairment of sensory perception. Neurosurgery 1998;43:462-72; discussion 472‐4.
140
251. Eide PK, Rabben T, Skjelbred P, et al. The effect of peripheral glycerol on trigeminal neuropathic pain examined by quantitative assessment of abnormal pain and sensory perception. Acta Neurochir (Wien) 1998;140:1271-7. 252. Sinay VJ, Bonamico LH, Dubrovsky A. Subclinical sensory abnormalities in trigeminal neuralgia. Cephalalgia 2003;23:541-4. 253. Rappaport ZH, Devor M. Trigeminal neuralgia: the role of self-sustaining discharge in the trigeminal ganglion. Pain 1994;56:127-38. 254. Devor M, Amir R, Rappaport ZH. Pathophysiology of trigeminal neuralgia: the ignition hypothesis. Clin J Pain 2002;18:4-13. 255. Rappaport ZH, Devor M. Tic and cranial neuralgias. In: Schmidt RF, Willis WD, editors. Encyclopedia of pain. New York: Springer; 2007. p. 2482-5. 256. Mitchell RG. Pre-trigeminal neuralgia. Br Dent J 1980;149:167-70. 257. Fromm GH, Graff-Radford SB, Terrence CF, et al. Pretrigeminal neuralgia. Neurology 1990;40:1493-5. 258. Henderson WR. Trigeminal neuralgia: the pain and its treatment. Br Med J 1967;1:7-15. 259. Rushton JG, Macdonald HN. Trigeminal neuralgia; special considerations of nonsurgical treatment. J Am Med Assoc 1957;165:437-40. 260. Rasmussen P. Facial pain. II. A prospective survey of 1052 patients with a view of: character of the attacks, onset, course, and character of pain. Acta Neurochir (Wien) 1990;107:121-8. 261. Terrence CF. Differential diagnosis of trigeminal neuralgia. In: Fromm GH, editor. The medical and surgical management of trigeminal neuralgia. Mt. Kisco: Futura Publishing; 1987. p. 43-60. 262. Loeser JD. True believer, with questions. APS J 1993;2:231-3. 263. Meirowsky AM, Pipito FF. Surgical history of trigeminal neuralgia. Arch Neurol Psychiat 1943;49:574-80. 264. Sweet WH. The history of the development of treatment for trigeminal neuralgia. Clin Neurosurg 1985; 32:294-318. 265. Gildenberg PL. History of pain management. In: Greenblatt SH, Dagi TF, Epstein MH, editors. A history of neurosurgery: in its scientific and professional context. Park Ridge: American Association of Neurological Surgeons; 1997. p. 439-64. 266. Gybels JM, Sweet WH. Neurosurgical treatment of persistent pain. Basel: Karger; 1989. 267. Taarnhøj P. Decompression of the trigeminal root and the posterior part of the ganglion as treatment in trigeminal neuralgia; preliminary communication. J Neurosurg 1952;9:288-90. 268. Shelden CH, Pudenz RH, Freshwater DB, et al. Compression rather than decompression for trigeminal neuralgia. J Neurosurg 1955;12:123-6.
2415
2416
140
The pathophysiology of trigeminal neuralgia
269. Stender A. Gangliolysis for the surgical treatment of trigeminal neuralgia. J Neurosurg 1954;11:333-6. 270. Malis LI. Petrous ridge compression and its surgical correction. J Neurosurg 1967;26(Suppl.):163-7. 271. Mullan S, Lichtor T. Percutaneous microcompression of the trigeminal ganglion for trigeminal neuralgia. J Neurosurg 1983;59:1007-12. 272. Jannetta PJ. Microsurgical approach to the trigeminal nerve for tic douloureux. In: Krayebu¨hl H, Maspes PE, Sweet WH, editors. Progress in neurological surgery: pain – its neurosurgical management, vol. 7. Basel: Karger; 1976. p. 180-200. 273. Jannetta PJ. Vascular compression is the cause of trigeminal neuralgia. APS J 1993;2:217-27. 274. Jannetta PJ, Møller MB, Møller AR, et al. Neurosurgical treatment of vertigo by microvascular decompression of the eighth cranial nerve. Clin Neurosurg 1986;33:645-65. 275. Parkinson D. Microvascular compression-decompression: a recollection. J Neurosurg 1989;70:819. 276. Letcher FS, Goldring S. The effect of radiofrequency current and heat on peripheral nerve action potential in the cat. J Neurosurg 1968;29:42-7. 277. Melzack R, Wall PD. Pain mechanisms: a new theory. Science 1965;150:971-9. 278. Wall PD, Sweet WH. Temporary abolition of pain in man. Science 1967;155:108-9. 279. Sweet WH. Pain modulation. The human experience. Neurosciences Res Prog Bull 1978;16:148-55. 280. Gybels J, Kupers R, Nuttin B. Therapeutic stereotactic procedures on the thalamus for pain. Acta Neurochir (Wien) 1993;124:19-22. 281. Meyerson BA, Ha˚kanson S. Alleviation of atypical trigeminal pain by stimulation of the Gasserian ganglion via an implanted electrode. Acta Neurochir Suppl (Wien) 1980;30:303-9. 282. Siegfried J. Long-term results of electrical stimulation in the treatment of pain by means of implanted electrodes (epidural spinal cord and deep brain stimulation). In: Rizzi R, Visentin M, editors. Pain Therapy: Proceedings of the International Postgraduate Practical Course on Pain Therapy, September 19th–October 2nd, 1982, Vicenza, Italy. Amsterdam: Elsevier; 1983. p. 463-75. 283. Pudenz RH. Neural stimulation: clinical and laboratory experiences. Surg Neurol 1993;39:235-42. 284. Johnson MD, Burchiel KJ. Peripheral stimulation for treatment of trigeminal postherpetic neuralgia and trigeminal posttraumatic neuropathic pain: a pilot study. Neurosurgery 2004;55:135-41; discussion 141‐2. 285. Taub E, Munz M, Tasker RR. Chronic electrical stimulation of the gasserian ganglion for the relief of pain in a series of 34 patients. J Neurosurg 1997;86:197-202. 286. Tsubokawa T. Motor cortex stimulation for relief of central deafferantation pain. In: Burchiel KJ, editor. Surgical management of pain. New York: Thieme; 2002. p. 555-64.
287. Tsubokawa T, Katayama Y, Yamamoto T, et al. Chronic motor cortex stimulation in patients with thalamic pain. J Neurosurg 1993;78:393-401. 288. Zakrzewska JM, Thomas DG. Patient’s assessment of outcome after three surgical procedures for the management of trigeminal neuralgia. Acta Neurochir (Wien) 1993;122:225-30. 289. Taha JM, Tew JM. Comparison of surgical treatments for trigeminal neuralgia: reevaluation of radiofrequency rhizotomy. Neurosurgery 1996;38:865-71. 290. Sheinberg MA, Sagher O. Medical management of trigeminal neuralgia. In: Burchiel KJ, editor. Surgical management of pain. New York: Thieme; 2002. p. 304-10. 291. Reder AT, Arnason BG. Trigeminal neuralgia in multiple sclerosis relieved by a prostaglandin E analogue. Neurology 1995;45:1097-100. 292. Zakrzewska JM. Trigeminal, eye and ear pain. In: Wall PD, Melzack R, editors. Textbook of pain. New York: Churhill Livingstone; 1999. p. 739-59. 293. Barker FG. Quality of life and individual treatment choice in trigeminal neuralgia. Pain 2007;131:234-6. 294. Fromm GH. The pharmacology of trigeminal neuralgia. Clin Neuropharmacol 1989;12:185-94. 295. Newton SA. Idiopathic headshaking in horses. Equine Vet Educ 2005;April:108–18. 296. Frohman EM, Goodin DS, Calabresi PA, et al. The utility of MRI in suspected MS: report of the therapeutics and technology assessment subcommittee of the American academy of neurology. Neurology 2003;61:602-11. 297. Cruccu G, Pennisi E, Truini A, et al. Unmyelinated trigeminal pathways as assessed by laser stimuli in humans. Brain 2003;126:2246-56. 298. Borsook D, DaSilva AF, Ploghaus A, et al. Specific and somatotopic functional magnetic resonance imaging activation in the trigeminal ganglion by brush and noxious heat. J Neurosci 2003;23:7897-903. 299. Borsook D, Moulton EA, Pendse G, et al. Comparison of evoked vs. spontaneous tics in a patient with trigeminal neuralgia (tic doloureux). Mol Pain 2007;3:34. 300. Devor M. Sources of variability in the sensation of pain. In: Dimitrijevic´ MR, Wall PD, Lindblom U, editors. Recent achievements in restorative neurology 3: altered sensation and pain. Karger; Basel: p. 189-96. 301. Wiley RG, Blessing WW, Reis DJ. Suicide transport: destruction of neurons by retrograde transport of ricin, abrin, and modeccin. Science 1982;216:889-90. 302. Yamamoto T, Iwasaki Y, Konno H. Experimental sensory ganglionectomy by way of suicide axoplasmic transport. J Neurosurg 1984;60:108-14. 303. McKenzie KG. Trigeminal tractotomy. Clin Neurosurg 1954;2:50-69; discussion 69‐70. 304. Cosman ER, Nashold BS, Ovelman-Levitt J. Theoretical aspects of radiofrequency lesions in the dorsal root entry zone. Neurosurgery 1984;15:945-50.
The pathophysiology of trigeminal neuralgia
305. Kanpolat Y, Deda H, Akyar S, et al. CT-guided trigeminal tractotomy. Acta Neurochir (Wien) 1989;100: 112-4. 306. Kanpolat Y, Savas A, Batay F, et al. Computed tomography-guided trigeminal tractotomy-nucleotomy in the management of vagoglossopharyngeal and geniculate neuralgias. Neurosurgery 1998;43:484-9; discussion 490. 307. Kanpolat Y, Savas A, Ugur HC, et al. The trigeminaltract and nucleus procedures in treatment of atypical facial pain. Surg Neurol 2005;64(Suppl.)2:S96-100; discussion S100‐1. 308. Backlund EO, Granberg PO, Hamberger B, et al. Transplantation of adrenal medullary tissue to striatum in parkinsonism. First clinical trials. J Neurosurg 1985; 62:169-73. 309. Bakay RA. Neural transplantation – what are the real possibilities? Clin Neurosurg 1991;37:179-92. 310. Vengsarkar US. Trigeminal neuralgia. Correspondence. Neurosurgery 1991;29:635. 311. Haines SJ, Walters BC, editors. Evidence-based neurosurgery. An introduction. New York: Thieme; 2006. 312. Harbaugh RE. Luddites, lemmings and outcome studies. AANS Bull (Guest Column) 1998;Fall:17. 313. Lopez BC, Hamlyn PJ, Zakrzewska JM. Stereotactic radiosurgery for primary trigeminal neuralgia: state of the evidence and recommendations for future reports. J Neurol Neurosurg Psychiatry 2004;75:1019-24. 314. Wall PD. The future of attacks on pain. In: Bonica JJ, editor. Advances in neurology, vol. 4: international symposium of pain. New York: Raven Press; 1974. p. 301-8. 315. Oppenheim H. Die peripherische Lahmung der Nerven an der Unterextremitat. Lehrbuch der Nervenkrankheiten: Fur Arzte und Studirende 1911;2:337. 316. Parker HL. trigeminal neuralgic pain associated with multiple sclerosis. Brain 1928;51:46-62. 317. Bekar A, Kocaeli H, Yilmaz E, et al. Trigeminal neuralgia caused by a pontine abscess: case report. Neurosurgery 2004;55:1434. 318. Foix, The´venard, Nicolesco. Algie faciale d’origine bulbo-trige´minale au cours de la Syringomye´lie. Troubles sympathiques concomitants. Douleur a` type cellulaire. Rev Neurol (Paris) 1922;38:990-8. 319. Maurice-Williams RS, Pilling J. Trigeminal sensory symptoms associated with hydrocephalus. J Neurol Neurosurg Psychiatry 1977;40:641-4. 320. Tucker WS, Fleming R, Taylor FA, et al. Trigeminal neuralgia in aqueduct stenosis. Can J Neurol Sci 1978; 5:331-3. 321. Fisher CM. Headache in cerebrovascular disease. In: Vinken PJ, Bruyn GW, editors. Handbook of clinical neurology: headaches and clinical neuralgias. Amsterdam: North Holland Publishing Co.; 1968. p. 124-56. 322. Harris W. Rare forms of paroxysmal trigeminal neuralgia, and their relation to disseminated sclerosis. Br Med J 1950;2:1015-9.
140
323. Conrad B, Mergner T. High cervical neurinoma (C1/ C2) diagnosed falsely as multiple sclerosis because of trigeminal neuralgia. Arch Psychiatr Nervenkr 1979; 227:33-7. 324. Balestrino M, Leandri M. Trigeminal neuralgia in pontine ischaemia. J Neurol Neurosurg Psychiatry 1997; 62:297-8. 325. Delitala A, Brunori A, Chiappetta F. Trigeminal neuralgia resulting from infarction of the root entry zone of the trigeminal nerve: case report. Neurosurgery 1999; 45:202. 326. Golby AJ, Norbash A, Silverberg GD. Trigeminal neuralgia resulting from infarction of the root entry zone of the trigeminal nerve: case report. Neurosurgery 1998;43:620-2; discussion 622‐3. 327. Peker S, Akansel G, Sun I, et al. Trigeminal neuralgia due to pontine infarction. Headache 2004;44:1043-5. 328. Mracek Z. Compression of the spinal trigeminal tract by pressure from the inferior posterior cerebellar artery as a cause of facial neuralgia. Rozhl Chir 19821;61:116-9 (Cze.) (English abstract). 329. Sunder-Plassmann M, Grunert V, Bo¨ck F. Vascular compression of the Tractus spinalis Nervi trigemini a possible cause of trigeminal neuralgia. Nervenarzt 1971;42:323-5. 330. Weber E. Die form der Schadelbasis in ihrer Beziehung zum Gesichtsschmerz, speziell zur Trigeminusneuralgie. Acta Neurochir Suppl (Wien) 1955;3:72-8. 331. Chang JW, Choi JY, Yoon Y, et al. Unusual causes of trigeminal neuralgia treated by gamma knife radiosurgery. Report of two cases. J Neurosurg 2002;97:533-5. 332. Mortara RH, Markesvery WR, Brooks WH. Pontine cyst presenting as trigeminal pain. Case report. J Neurosurg 1974;41:636-9. 333. Kanpolat Y, Tatli M, Ugur HC, et al. Evaluation of platybasia in patients with idiopathic trigeminal neuralgia. Surg Neurol 2007;67:78-81; discussion 81‐2. 334. Clarke CR, Harrison MJ. Neurological manifestations of Paget’s disease. J Neurol Sci 1978;38:171-8. 335. Gardner WJ, Todd EM, Pinto JP. Roentgenographic findings in trigeminal neuralgia. Am J Roentgen Radium Ther Nucl Med 1956;76:346-50. 336. Ibrahim AG, Crockard HA. Basilar impression and osteogenesis imperfecta: a 21-year retrospective review of outcomes in 20 patients. J Neurosurg Spine 2007; 7:594-600. 337. Ildan F, Go¨c¸er AI, Bag˘datog˘lu H, et al. Isolated trigeminal neuralgia secondary to distal anterior inferior cerebellar artery aneurysm. Neurosurg Rev 1996;19:43-6. 338. Leopold NA, Hirsh LF, Ray T. Paroxysmal facial neuralgia secondary to a posterior communicating artery aneurysm. Surg Neurol 1980;14:221-3. 339. Trotter MI, Choksey MS. Facial pain with intracranial aneurysm. J R Soc Med 2000;93:479-80.
2417
2418
140
The pathophysiology of trigeminal neuralgia
340. Morita A, Fukushima T, Miyazaki S, et al. Tic douloureux caused by primitive trigeminal artery or its variant. J Neurosurg 1989;70:415-9. 341. Tamura Y, Shimano H, Kuroiwa T, et al. Trigeminal neuralgia associated with a primitive trigeminal artery variant: case report. Neurosurgery 2003;52:1217-9; discussion 1219‐20. 342. Gnanalingham K, Joshi SM, Lopez B, et al. Trigeminal neuralgia secondary to Chiari’s malformation – treatment with ventriculoperitoneal shunt. Surg Neurol 2005;63:586-8; discussion 588‐9. 343. Rosetti P, Ben Taib NO, Brotchi J, et al. Arnold Chiari Type I malformation presenting as a trigeminal neuralgia: case report. Neurosurgery 1999;44:1122-3; discussion 1123‐4. 344. Garcı´a-Pastor C, Lo´pez-Gonza´lez F, Revuelta R, et al. Trigeminal neuralgia secondary to arteriovenous malformations of the posterior fossa. Surg Neurol 2006; 66:207-11; discussion 211. 345. Karibe H, Shirane R, Jokura H, et al. Intrinsic arteriovenous malformation of the trigeminal nerve in a patient with trigeminal neuralgia: case report. Neurosurgery 2004;55:1433. 346. Edwards RJ, Clarke Y, Renowden SA, et al. Trigeminal neuralgia caused by microarteriovenous malformations of the trigeminal nerve root entry zone: symptomatic relief following complete excision of the lesion with nerve root preservation. J Neurosurg 2002;97:874-80. 347. Tsubaki S, Fukushima T, Tamagawa T, et al. Parapontine trigeminal cryptic angiomas presenting as trigeminal neuralgia. J Neurosurg 1989;71:368-74. 348. Peet M, Schneider RC. Trigeminal neuralgia; a review of six hundred and eighty-nine cases with a follow-up study of sixty five per cent of the group. J Neurosurg 1952;9:367-77. 349. Bartlow B, Penn RD. Carotid-cavernous sinus fistula presenting as a posterior fossa mass. Case report. J Neurosurg 1975;42:585-8. 350. Fritz C, Ro¨sler A, Heyden B, et al. Trigeminal neuralgia as a clinical manifestation of Lyme neuroborreliosis. J Neurol 1996;243:367-8. 351. Du R, Binder DK, Halbach V, et al. Trigeminal neuralgia in a patient with a dural arteriovenous fistula in Meckel’s cave: case report. Neurosurgery 2003; 53:216-21; discussion 221. 352. Ito M, Sonokawa T, Mishina H, et al. Dural arteriovenous malformation manifesting as tic douloureux. Surg Neurol 1996;45:370-5. 353. Borne G. Trigeminal neuralgia as the presenting symptom of a tuberculoma of the cerebellopontine angle. Case report. J Neurosurg 1968;28:480-2. 354. Atri M, Robertson WD, Durity FA, et al. Actinomycotic granuloma of the trigeminal ganglion. AJNR Am J Neuroradiol 1987;8:167-9. 355. Kiya K, Sakoda K, Gen M, et al. A case of aspergillotic meningoencephalitis associated with trigeminal neuralgia. No Shinkei Geka 1982;10:861-6.
356. Wiles CM, Kocen RS, Symon L, et al. Aspergillus granuloma of the trigeminal ganglion. J Neurol Neurosurg Psychiatry 1981;44:451-5. 357. Revilla AG. Tic douloureux and its relationship to tumors of the posterior fossa: analysis of twenty-four cases. J Neurosurg 1947;4:233-9. 358. Tancioni F, Gaetani P, Villani L, et al. Neurinoma of the trigeminal root and atypical trigeminal neuralgia: case report and review of the literature. Surg Neurol 1995;44:36-42. 359. Revuelta R, Juambelz P, Balderrama J, et al. Contralateral trigeminal neuralgia: a new clinical manifestation of neurocysticercosis: case report. Neurosurgery 1995; 37:138-9; discussion 139‐40. 360. Revuelta R, Soto-Herna´ndez JL, Vales LO, et al. Cerebellopontine angle cysticercus and concurrent vascular compression in a case of trigeminal neuralgia. Clin Neurol Neurosurg 2003;106:19-22. 361. Tenuto RA, Canelas HM, Cruz OR, et al. Trigeminal neuralgia caused by cysticercosis of the cavum meckelii. Report of two cases. J Neurosurg 1963;20:169-71. 362. Kobata H, Kondo A, Iwasaki K. Cerebellopontine angle epidermoids presenting with cranial nerve hyperactive dysfunction: pathogenesis and long-term surgical results in 30 patients. Neurosurgery 2002;50:276-85; 363. Meng L, Yuguang L, Feng L, et al. Cerebellopontine angle epidermoids presenting with trigeminal neuralgia. J Clin Neurosci 2005;12:784-6. 364. Samii M, Carvalho GA, Tatagiba M, et al. Surgical management of meningiomas originating in Meckel’s cave. Neurosurgery 1997;41:767-74; discussion 774‐5. 365. Baringer JR, Swoveland P. Recovery of herpes-simplex virus from human trigeminal ganglions. N Engl J Med 1973;288:648-50. 366. Carton CA. Effect of previous sensory loss on the appearance of herpes simplex following trigeminal sensory root section. J Neurosurg 1953;10:463-8. 367. Grimm RJ. Activation of latent herpesvirus type I by trigeminal rhizotomy: a hypothesis. Trans Am Neurol Assoc 1973;98:170-4. 368. Knight G. Herpes simplex and trigeminal neuralgia. Proc R Soc Med 1954;47:788-90. 369. Pazin GJ, Ho M, Jannetta PJ. Reactivation of herpes simplex virus after decompression of the trigeminal nerve root. J Infect Dis 1978;138:405-9. 370. Wepsic JG. Tic douloureux: etiology, refined treatment. N Engl J Med 1973;288:680-1. 371. Celik SE, Kocaeli H, Cordan T, et al. Trigeminal neuralgia due to cerebellopontine angle lipoma. Case illustration. J Neurosurg 2000;92:889. 372. Kato T, Sawamura Y, Abe H. Trigeminal neuralgia caused by a cerebellopontine-angle lipoma: case report. Surg Neurol 1995;44:33-5. 373. Lena G, Dufour T, Gambarelli D, et al. Choristoma of the intracranial maxillary nerve in a child. Case report. J Neurosurg 1994;81:788-91.
The pathophysiology of trigeminal neuralgia
374. Patel N, Moss T, Coakham H. Neuronal hamartoma of the trigeminal sensory root associated with trigeminal neuralgia. J Neurosurg 2000;93:514. 375. Gregg JM. Studies of traumatic neuralgias in the maxillofacial region: surgical pathology and neural mechanisms. J Oral Maxillofac Surg 1990;48:228-37; discussion 238‐9. 376. Marshall PC, Rosman NP. Symptomatic trigeminal neuralgia in a 5-year-old child. Pediatrics 1977;60:331-3. 377. Ferrari MD, Haan J, van Seters AP. Bromocriptineinduced trigeminal neuralgia attacks in a patient with a pituitary tumor. Neurology 1988;38:1482-4. 378. Ruelle A, Datti R, Andrioli G. Cerebellopontine angle osteoma causing trigeminal neuralgia: case report. Neurosurgery 1994;35:1135-7. 379. Kinoshita M, Izumoto S, Oshino S, et al. Primary malignant lymphoma of the trigeminal region treated with rapid infusion of high-dose MTX and radiation: case report and review of the literature. Surg Neurol 2003;60:343-8; discussion 348. 380. Stark AM, Buhl R, Hugo HH, et al. Chronic recurrent subarachnoid hemorrhage from a trigeminal nerve malignant peripheral nerve sheath tumor: case report. Neurosurgery 2006;59:E425; discussion E425. 381. Chong VF. Trigeminal neuralgia in nasopharyngeal carcinoma. J Laryngol Otol 1996;110:394-6. 382. Eng CT, Vasconez LD. Facial pain due to perineural invasion by basal cell carcinoma. Ann Plast Surg 1984;12:374-7. 383. Bornemann A, Bohl J, Hey O, et al. Amyloidoma of the gasserian ganglion as a cause of symptomatic neuralgia of the trigeminal nerve: report of three cases. J Neurol 1993;241:10-4. 384. Love S, Bateman DE, Hirschowitz L. Bilateral lambda light chain amyloidomas of the trigeminal ganglia, nerves and roots. Neuropathol Appl Neurobiol 1997;23:512-5. 385. Hojaili B, Barland P. Trigeminal neuralgia as the first manifestation of mixed connective tissue disorder. J Clin Rheumatol 2006;12:145-7. 386. Peker S, Necmettin Pamir M. Trigeminal neuralgia in a patient with mitochondrial neurogastrointestinal encephalomyopathy (MNGIE). J Clin Neurosci 2005; 12:172-4.
140
387. Martuza RL, Ojemann RG, Shillito J, et al. Facial pain associated with a middle fossa arachnoid cyst. Neurosurgery 1981;8:712-6. 388. Urasaki E, Fukumura A, Ito Y, et al. Choroidal epithelial cyst in the cerebellopontine angle associated with trigeminal neuralgia – case report. Neurol Med Chir (Tokyo) 1989;29:424-8. 389. Wepfer J. Observationes medico-tracticae de affectibus capitis. Schaffhausen: Ziegler; 1727. 390. Rose W. Surgical treatment of trigeminal neuralgia (Abstract of the Lettsomian Lectures). Lancet 1892; January:9. 391. Andre´ NA. Observations pratiques sur les maladies de l’ure`tre et sur plusieurs faits convulsifs, et la gue´rison de plusieurs maladies chirurgicales. Imprimeur de Colle`ge et de l’Acad. Roy. de Chir. sParis: Olivier; 1756. p. 320-67. 392. Krause F. Die Neuralgie des Trigeminus nebst der Anatomie und Physiologie der Nerven. Leipzig: Vogel; 1896. 393. Horsley V, Taylor J, Colman WS. Remarks on the various surgical procedures devised for the relief or cure of trigeminal neuralgia (tic douloureux). Br Med J 1891;2:1139-43, 1191‐3, 1249‐52. 394. Spiller WG, Frazier CH. The division of the sensory root of the trigeminus for the relief of tic douloureux; an experimental, pathological and clinical study, with a preliminary report of one surgically successful case. Philad Med J 1901;8:1039-49. 395. Schloesser K. Erfahrungen in der Neuralgiebehandlung mit Alkoholeinspritzungen. Klin Wschr 1907;44:533-4. 396. Harris W. Alcohol injection of the gasserian ganglion for trigeminal neuralgia. Lancet 1912;1:218-21. 397. Kirschner M. Zur electrochirurgie. Arch Klin Chir 1931;161:761-8. 398. Leksell L. The stereotaxic method and radiosurgery of the brain. Acta Chir Scand 1951;102:316-9. 399. Sweet WH. Trigeminal neuralgias. In: Alling CC, editor. Facial pain. Philadelphia, PA: Lea & Febiger; 1968. p. 89-106. 400. Ha˚kanson S. Trigeminal neuralgia treated by the injection of glycerol into the trigeminal cistern. Neurosurgery 1981;9:638-46.
2419
146 Treatment of Headache N. T. Mathew
The International Headache Society classifies headache disorders into primary and secondary disorders [1]. > Table 146‐1 lists the primary headache disorders.
Migraine In an epidemiological study of migraine [2] using a self-administered questionnaire filled out by 23,611 individuals from 9,507 American households, 17.6% of women and 5.7% of men between ages 12 and 80 years had headaches that met a definition of migraine that was based on a modification of the International Headache Society’s criteria [1]. Projections from this study to the country as a whole indicate that 18 million women and 5.6 million men over age 12 suffer from severe migraine headaches. Diagnostic criteria for migraine with and without an aura are given in > Table 146‐2. Among these individuals, a projected 8.7 million women and 2.6 million men have moderate to severe disability from headache [2]. Fifty-five percent of men and seventy-two percent of women never consult a doctor for a headache problem [3]. Among those who seek medical attention, 44.5% of men and 46.3% of women consult family practitioners. With increasing awareness of headache as a biological problem, it is expected that a larger number of migraine patients will seek help for this condition. Migraine can occur with or without an aura (warning symptoms). The most common aura is visual in nature, although neurological auras such as hemisensory disturbances, hemiparesis, dysphasia, and changes in memory and state of consciousness occur occasionally. Migraine #
Springer-Verlag Berlin/Heidelberg 2009
without aura is far more common than migraine with aura; approximately 30% of migraine attacks are associated with aura. The same person can have migraine with aura and migraine without aura at different times. Migraine is predominantly a disease of females. Identification of trigger factors for attacks of migraine helps in making the diagnosis. The trigger factors are listed in > Table 146‐3. The severity and frequency of attacks vary over time. Cyclical exacerbations of migrainous episodes are possible during one’s lifetime. A migraine attack is typically episodic, occurring once or twice a month, and is manifested in many phases. The prodrome phase consisting of symptoms of excitation or inhibition of the central nervous system, including elation, excitability, irritability, increased appetite, craving for sweets, or excessive yawning, depression, sleepiness, and tiredness, occurs in 30% of patients. These symptoms may precede the attack by 12–24 h. The prodrome phase may be followed by the aura phase, which consists of specific visual or neurological symptoms. The headache phase is the most prominent part of the migraine attack. The headache is predominantly unilateral in at least 50% of patients, although it can be bilateral. It also may start on one side and switch to the other side. A pulsating quality of the head pain is seen in approximately 50% of these patients. Nonpulsating headache does not exclude migraine. The headache usually lasts from 4 to 72 h, and occasionally lasts longer. It is associated with gastrointestinal symptoms such as nausea and/or vomiting and diarrhea in 90% of patients. Heightened sensory perception, including phonophobia, photophobia, and increased sensitivity to smell, occurs during the
2484
146
Treatment of headache
. Table 146‐1 International Headache Society classification of primary headache disorders Migraine Migraine without aura Migraine with aura Migraine with typical aura Migraine with prolonged aura Familial hemiplegic migraine Basilar migraine Migraine aura without headache Migraine with acute-onset aura Ophthalmoplegic migraine Retinal migraine Childhood periodic syndromes that may be precursors to or associated with migraine Benign paroxysmal vertigo of childhood Alternating hemiplegia of childhood Complications of migraine Status migrainosus Migrainous infarction Migrainous disorder not fulfilling the above criteria Tension-type headache Episodic tension-type headache Episodic tension-type headache associated with disorder of pericranial muscles Episodic tension-type headache not associated with disorder of pericranial muscles Chronic tension-type headache Chronic tension-type headache associated with disorder of pericranial muscles Chronic tension-type headache not associated with disorder of pericranial muscles Headache of the tension type not fulfilling the above criteria Cluster headache and chronic paroxysmal hemicrania Cluster headache Cluster headache periodicity undetermined Episodic cluster headache Chronic cluster headache Unremitting from onset Evolved from episodic Chronic paroxysmal hemicrania Cluster headache-like disorder not fulfilling the above criteria Miscellaneous headaches not associated with a structural lesion Idiopathic stabbing headache External compression headache Cold stimulus headache External application of a cold stimulus Ingestion of a cold stimulus Benign cough headache Benign exertional headache
. Table 146‐1 (Continued) Headache associated with sexual activity Dull type Explosive type Postural type
attacks. Patients usually want to be left alone, and the attacks can be very disabling. The headache of migraine is of moderate or severe intensity (inhibits or prohibits daily activities) as opposed to episodic tension-type headache, in which the intensity is mild to moderate (may inhibit but does not prohibit daily activities). The headache of migraine is aggravated by any activity that increases stroke volume or intracranial pressure, such as climbing stairs, jogging, running, bending down, and coughing. During the headache, at least one of the following characteristics occurs: (1) nausea and/or vomiting and (2) photophobia and phonophobia. These symptoms are necessary for a diagnosis of migraine. Physical and neurological examinations should rule out any other structural or metabolic condition that can cause headache.
Menstrual Migraine Migraine without aura can occur almost exclusively at a particular time in the menstrual cycle. True menstrual migraine occurs between 2 days before menses and the last day of menses. Migraine also can occur as a part of the late luteal phase disphoric disorder (premenstrual tension). Migraine attacks are not uncommon during ovulation. Menstrual migraine is less responsive to prophylactic drug therapy than other types of migraine. Status migrainosus refers to a prolonged migraine attack that usually lasts for more than 72 h and is associated with nausea, vomiting, and dehydration. These patients usually are extremely sick and dehydrated and may have to be
Treatment of headache
146
. Table 146‐2 International Headache Society criteria for diagnosis of migraine 1.1 Migraine without aura Previously used term: common migraine Description: Idiopathic, recurring headache disorder manifesting in attacks lasting 4–72 h. Typical characteristics of headache are unilateral location, pulsating quality, moderate or severe intensity, aggravation by routine physical activity, association with nausea, and photo- and phonophobia. Diagnostic criteria: A. At least five attacks fulfilling B–D B. Headache attacks lasting 4–72 h (untreated or unsuccessfully treated) C. Headache has at least two of the following characteristics: 1. Unilateral location 2. Pulsating quality 3. Moderate or severe intensity (inhibits or prohibits daily activities) 4. Aggravation by walking stairs or similar routine physical activity D. During headache, at least one of the following: 1. Nausea and/or vomiting 2. Photophobia and phonophobia E. At least one of the following: 1. History and physical and neurological examinations do not suggest one of the disorders listed in groups 5–11 2. History and/or physical and/or neurological examinations do suggest such a disorder, but it is ruled out by appropriate investigations 3. Such disorder is present, but migraine attacks do not occur for the first time in close temporal relation to the disorder 1.2 Migraine with auraa Previously used term: classic migraine Description: Idiopathic, recurring disorder manifesting by attacks of neurological symptoms unequivocally localizable to cerebral cortex or brain stem, usually gradually developed over 5–20 min and usually lasting less than 60 min Headache, nausea, and/or photophobia usually follow neurological aura symptoms directly or after a free interval of less than an hour. The headache usually lasts 4–72 h but may be completely absent (1.2.5) Diagnostic criteria: A. At least two attacks fulfilling B B. At least three of the following four characteristics: 1. One or more fully reversible aura symptoms indicating focal cerebral cortical and/or brain stem dysfunction 2. At least one aura symptom develops gradually over more than 4 min or two or more symptoms occur in succession 3. No aura symptom lasts more than 60 min. If more than one aura symptom is present, the accepted duration is proportionally increased 4. Headache follows aura with a free interval of less than 60 min. (It may also begin before or simultaneously with the aura.) C. At least one of the following: 1. History and physical and neurological examinations do not suggest one of the disorders listed in groups 5–11 2. History and/or physical and/or neurological examinations do suggest such disorder, but it is ruled out by appropriate investigations 3. Such a disorder is present, but migraine attacks do not occur for the first time in close temporal relation to the disorder a
Aura as used here does not necessarily imply that it precedes the headache or imply a relationship with epilepsy
hospitalized. By the time they come to the emergency room, they may have already tried large quantities of analgesic medications and/or ergotamine with no benefit.
Cluster Headache Cluster headache is predominantly a disease of males. These headaches are almost always
2485
2486
146
Treatment of headache
. Table 146‐3 Triggers of migraine Common factors
Less common factors
Stress, worry, anxiety Menstruation Oral contraceptives Certain foodstuffs and alcohol
High humidity Excessive sleep High altitude Excessive vitamin A Drugs: nitroglycerine, reserpine, estrogens, hydralazine, ranitidine
Hunger Lack of sleep Glare, dazzle Weather or ambient temperature changes Physical exertion, fatigue Head trauma
Pungent odors Fluorescent lighting Allergic reactions Cold foods Refractory error
unilateral and short-lived, usually lasting about 45 min to an hour. Multiple episodes occur on a daily basis for periods of 2 or 3 months, with remissions lasting for a number of months to years, with headaches returning in a cluster fashion again for 2 or 3 months. Cluster pattern and remissions are characteristics of the disease, even though in approximately 10% of patients there are no remissions (chronic cluster headache). Associated with the pain are autonomic features such as watering from the eyes, redness of the eyes and congestion of the conjunctiva, and ipsilateral stopping up of the nostrils during the attack. The International Headache Society diagnostic criteria are given in > Table 146‐4.
Episodic Tension-Type Headache The most common type of headache is the episodic tension-type headache, for which patients rarely consult a doctor. This type of headache usually is pressing or tightening in quality, bilateral, mild to moderate in severity, and occasionally associated with very mild nausea, photophobia, or sonophobia. There is no vomiting, and the patients
. Table 146‐4 International Headache Society diagnostic criteria for cluster headache A. At least five attacks fulfilling B–D B. Severe unilateral orbital, supraorbital, and/or temporal pain lasting 15–180 min untreated C. Headache associated with at least one of the following signs, which have to be present on the pain side: 1. Conjunctival injection 2. Lacrimation 3. Nasal congestion 4. Rhinorrhea 5. Forehead and facial sweating 6. Miosis 7. Ptosis 8. Eyelid edema D. Frequency of attacks from 1 every other day to 8 per day E. At least one of the following: 1. History and physical and neurological examinations do not suggest one of the disorders listed in groups 5–11 2. History and/or physical and/or neurological examinations do suggest such a disorder, but it is ruled out by appropriate investigations 3. Such a disorder is present, but cluster headache does not occur for the first time in close temporal relation to the disorder
are able to carry on with their activities. The headache is not aggravated by physical activity.
Chronic Daily Headache Even though the term chronic daily headache is not included in the International Headache Society classification, from a practical point of view it is important. Chronic tension-type headaches are one of the types of chronic daily headache. The clinical features of chronic tensiontype headache are essentially the same as those of the episodic tension-type except that the headache occurs more than 180 days a year. The comorbid factors often seen in chronic tension-type headache include anxiety, depression, excessive intake of pain medications, abnormal personality profiles, inadequate personality, and repressed anger.
Treatment of headache
Both episodic and chronic tension-type headaches can be associated with pericranial muscle tenderness and a low pain threshold. Digital palpation of the pericranial muscles, including the neck muscles, reveals increased stiffness and tenderness. The migraine chronic tension-type headache complex (mixed headache) manifests as daily or nearly daily headaches that show features of migraine and chronic tension-type headache in a mixed form. Many patients have episodes of severe headache with migrainous features, with interictal tension-type headache occurring very frequently. Many of these patients have a history of episodic migraine that gradually evolves into daily headache (transformed migraine) [4]. It sometimes becomes difficult to identify the termination of one type of headache and the beginning of the other type. There are two distinguishable forms in this variety of headache: those associated with analgesic and ergotamine overuse and those not associated with drug overuse. It is well known to specialists in headache that daily or nearly daily use of analgesics and ergotamine in patients with migraine can lead to a chronic daily intractable headache condition that is referred to as an analgesic/ergotamine rebound headache. It is important to look for this disorder in any patient who presents with chronic headaches. Analgesic/ ergotamine rebound headache is refractory to regular treatments. The patients show many associated features, such as early morning awakening with severe headaches; sleep disturbances; tolerance to pain medications over a period of time, requiring larger quantity of medications; and manifestation of withdrawal symptoms when the medications are stopped. In addition, prophylactic antimigraine medications become ineffective as long as the patients are on daily pain medications or ergotamine.
Posttraumatic Headache Headache can follow relatively minor head and neck trauma, and previously dormant migraine
146
can be aggravated by trauma. Patients with posttraumatic headache usually manifest a mixed form of migraine and tension-type headache with considerable detectable neck muscle spasm and pericranial tenderness. Most headache specialists believe that the various clinical headache types are different manifestations of the same primary disorder, which presents as migraine with aura at one end of the spectrum, chronic tension-type headache at the other, and a combination of migraine without aura and tension-type headache in the middle [5]. Many patients who present with chronic daily headaches have what is described as ‘‘transformed migraine’’ that is, they report a history of clear-cut episodic migraine with increasing frequency of headache until they eventually end up with daily or nearly daily headaches, many of which retain features of migraine [4].
Biological Basis of Migraine Pharmacotherapy The two basic theories that have been postulated to explain the mechanisms of migraine are vascular and neurogenic, with considerable debate about whether migraine is primarily a cephalic vascular disorder or a disorder of the central nervous system (CNS). The vascular theory of migraine proposes that intracerebral vasoconstriction accounts for the aura or migraine, while intracranial and extracranial vasodilation accounts for the head pain. A lack of correlation between the observed changes in cerebral blood flow [6,7] and the occurrence of head pain in patients who have migraine with aura has led to the conclusion that vascular reactions may be associated with symptoms of headache but do not necessarily trigger an attack. Instead, recent clinical and experimental evidence strongly points to migraine as a disorder initiated in the brain and accompanied by secondary changes in the perivascular
2487
2488
146
Treatment of headache
nerve endings of the cephalic circulation that result in neurogenic inflammation.
The Trigeminal Vascular System A series of studies by Moskowitz and colleagues [8] established the trigeminal vascular system as the common final pathway for head pain. The perivascular C fibers of the trigeminal nerve in the cephalic circulation are the site of neurogenic inflammation that can be produced by antidromic stimulation of the trigeminal nerve [8]. Neurogenic inflammation is triggered by elaboration of vasoactive polypeptides such as substance P, neurokinin A, and calcitonin generelated polypeptide (CGRP). The neurogenic inflammation consists of platelet aggregation, protein extravasation, and vasodilatation. These changes initiate nociception through perivascular C fibers. Stimulation of the cranial vascular systems, such as the superior sagittal sinus, results in an increase in CGRP in the jugular blood in experimental animals [9]. Goadsby and Edvinsson [10] showed that during a migraine attack the level of CGRP in the external jugular vein increases. These observations undoubtedly prove that the trigeminal vascular system is involved in acute migraine attacks. The ultimate initiating trigger of a migraine attack may reside in the brain, which in turn activates the trigeminal vascular system.
Central Neuronal Hyperexcitability in Migraine Various observations suggest a CNS origin of migraine. A central neuronal hyperexcitability probably exists in migraine, as evidenced by prominent photic driving during electroencephalography, [11] augmented high-amplitude visually evoked responses, [12] increased contingent negative variation amplitude, [13] and
magnetoencephalographic abnormalities of largeamplitude waves with suppression in the direct current shifts [14]. Glutamate, an excitatory amino acid, has been implicated in the pathogenesis of migraine, since it may play a role in spreading depression experimentally [14]. Abnormalities in platelet glutamate levels have been reported in patients with migraine [15].
Brain Stem Migraine Generator Brainstem structures that are implicated in the pathogenesis of migraine and have relevance to pharmacotherapy include the dorsal raphe nuclei and periaqueductal gray matter. Raskin and coworkers [16] indicated that perturbation or dysfunction of the dorsal raphe nuclei, resulting in an increased firing rate of the raphe cells, may be one of the fundamental abnormalities in migraine. Sleep, which reduces the firing rate of dorsal raphe cells, is known to relieve migraine headache. Recently, using positron emission tomography (PET) scanning, increased perfusion of the upper brain stem area close to the dorsal raphe and periaqueductal gray matter was demonstrated in patients during migraine attacks [17]. These observations suggest the upper brain stem as a generator of migraine. The dorsal raphe nuclei are one of the major binding sites of dihydroergotamine mesylate (DHE), a very effective therapy for acute migraine [18]. The dorsal raphe nuclei contain large numbers of serotoninergic cells, suggesting that DHE and related medications may act through the serotoninergic system.
Serotonin and Migraine Serotonin [5-hydroxytryptamine (5-HT)] has long been implicated in the pathogenesis of migraine. Initial observations included the precipitation of
Treatment of headache
. Table 146‐5 Chronological Order in which 5-HT medications were introduced into migraine therapy Serotonin-related drug Ergotamine Dihydroergotamine Methysergide Cyproheptadine Pizotifen Amitriptyline Sumatriptan
. Table 146‐6 Antimigraine drug relative potencies at 5-HT1D receptor subtypes
Year 1928 1945 1959 1964 1968 1973 1991
migraine by reserpine, a serotonin-depleting agent, [19] and relief of migraine by the injection of serotonin even though severe side effects limit its clinical use [20]. Serotonin metabolites are increased in urine during an attack, [21] and plasma serotonin levels fall just before a migraine attack [22]. Medications that act on the serotonin system have long been used in the treatment of migraine. > Table 146‐5 shows the chronological order in which serotonin-related medications have been used in migraine. In recent years, there has been a great deal of interest in serotonin pharmacology. At lease three major classes of serotonergic receptors have been identified: 5-HT1, 5-HT2, and 5-HT3. 5-HT1D and 5-HT1A are probably the most relevant serotonin receptors in relation to the pharmacology of an acute migraine attack. In general, it appears that medications with an affinity for 5-HT1 receptors are effective in the treatment of acute migraine attacks and that medications that have an antagonist affinity for 5-HT2 are useful in migraine prophylaxis [23]. The relative affinities for 5-HT1 are shown in > Table 146‐6. It has been shown that pretreatment with sumatriptan and DHE, both of which are 5-HT1 agonists, can block the neurogenic inflammation that can be induced by antidromic stimulation of the trigeminal nerve [24,25].
146
Antimigraine drug Acute Sumatriptan Dihydroergotamine Prophylactic Methysergide Pizotifen, alprenolol, amitriptyline, cyproheptadine, nifedipine, pindolol, propranolol, verapamil, timolol, atenolol, or diltiazem
Relative affinity for 5-HT1D receptor 17 19 120 >1,000
Adapted from Peroutka SJ: Development of 5-hydroxytriptamine receptor pharmacology in migraine. Neurol Clin 8:831, 1990. With permission
In addition, these medications reduce or prevent the increase in CGRP in the jugular blood induced by craniovascular stimulation [9] and during migraine attacks [26]. Researchers do not fully agree on the mechanism of action of 5-HT1 agonists in migraine. On the basis of their experiments, Moskowitz and colleagues strongly believe that 5-HT1D receptors are located on the C fibers of the perivascular nerve endings. The agonistic action of sumatriptan and ergot alkaloids on 5-HT1D receptors blocks the neurogenic inflammation at the level of the nerve ending. They also have shown that 5-HT1D receptor agonists, in addition to blocking neurogenic inflammation, may reduce pain transmission through the trigeminal nerve; this conclusion was based on experiments using c-fos markers [27]. However, Humphrey and coworkers [28] believe that the effect of sumatriptan is primarily due to craniovascular vasoconstriction and closure of the arteriovenous anastomosis, which is postulated to be one of the mechanisms of migraine head pain [29]. It should be noted that sumatriptan has very little peripheral vasoconstrictive effect and is predominantly a selective craniovascular agent.
2489
2490
146
Treatment of headache
Treatment of Acute Migraine Attacks Factors that determine the choice of medications for acute migraine are time to reach maximum headache (time to peak), severity, and associated symptoms such as nausea and vomiting. The frequency of attacks combined with their severity determine prophylactic pharmacotherapy. The main principle of the treatment of acute migraine is the use of medications early in an attack. This is especially relevant for oral medications, which should be administered long before nausea and vomiting set in. For practical purposes, it may be worthwhile to divide acute migraine attacks into different treatment categories, depending on the severity of the attack (> Table 146‐7).
Mild to Moderate Attacks For mild to moderate attacks, simple analgesics such as aspirin and acetaminophen may be all that is necessary. There is an associated gastroparesis in migraine that results in poor . Table 146‐7 Aborting headache treatment options Time to peak, h <1
1–3
>3
Options Sumatriptan 6 mg subcutaneous DHE 1 mg intramuscular Nasal butorphanol 1–2 mg Sumatriptan tablets 50 mg; repeat 50 mg at 2 h Ergotamine suppository 2 mg (Wigraine, Cafergot) DHE nasal spray 2 mg Isometheptene compounds (Midrin); 2 capsules, repeat 1 in 1/2 h Ergotamine tablets 2 mg or ergotamine 1 mg with naproxen 550 mg or meclofenamate 200 mg Sumatriptan 50-mg tablets; repeat 1 in 2 h
absorption of aspirin from the large intestine [30]. Phenothiazines such as promethazine, which are commonly used as antiemetics, have a tendency to reduce gastric motility further, resulting in delayed absorption of aspirin. Therefore, metoclopramide, which increases gastric motility and enhances the absorption of aspirin, is the antiemetic of choice in patients with acute migraine attacks. Nonsteroidal anti-inflammatory drugs (NSAIDs) such as rapidly absorbed naproxen sodium are effective for the abortive treatment of mild to moderate cases of migraine. Isometheptene mucate (available in combination with dichloralphenazone and acetaminophen) is also a very useful agent for abortive migraine treatment. Isometheptene is a sympathomimetic vasoconstrictor, and its exact mechanism of action is unknown. Since it is well tolerated by most people, it is the drug of choice for mild to moderate attacks.
Moderate to Severe Attacks Until recently, ergotamine tartrate was the drug of choice for moderate to severe episodes of migraine. Ergotamine, alone or in combination with caffeine, is available in oral and rectal forms. If the patient is not nauseated, oral ergotamine is useful and can be combined with NSAIDs such as naproxen sodium and meclofenamate. Meclofenamate alone has been shown to have an antimigraine effect comparable to that of ergotamine, [31] and we find this combination useful. Injectable ketorolac has been very effective in some patients with moderate to severe migraine and has been recommended as a practical alternative to narcotic injections [32]. If the patient is nauseated, metoclopramide is often the antiemetic of choice. Since metoclopramide is a dopamine agonist, it may cause extrapyramidal reactions in the form of akathisia and restlessness.
Treatment of headache
Drugs Used to Treat Migraine Sumatriptan Sumatriptan is a 5-HT1D agonist that blocks neurogenic inflammation at the trigeminal vascular system and causes vasoconstriction. It is available in tablet and subcutaneous forms. Sumatriptan has become the drug of choice in the treatment of moderate to severe cases of migraine. Our knowledge of sumatriptan use in acute migraine is based on major clinical trials [33–35]. With subcutaneous injection, the onset of relief occurs in 10–15 min; 50% of patients get relief in 30 min. More than 80% show relief in less than 2 h, and 60 % become pain-free in 2 h. With tablets, the onset of relief occurs in 30 min. Sixty percent of these patients show improvement in 1 h, and 75% by 4 h. Nearly 50% are pain-free in 4 h. Sumatriptan is effective any time during an attack, unlike ergotamine, which is most effective in the early part of an attack. Another important advantage of sumatriptan over ergotamine is that the accompanying symptoms of migraine, such as nausea and vomiting, are relieved by sumatriptan, obviating the need for a separate antiemetic agent. With sumatriptan, patients are able to return to their normal activities very rapidly. Sumatriptan studies have shown that most of the adverse events are mild to moderate, occur early after the treatment, are of short duration, and resolve spontaneously. Electrocardiogram monitoring showed no more abnormalities with sumatriptan than with placebo. There are very few adverse reactions of any consequence, and as the data indicate, this is a relatively safe drug from a cardiovascular point of view. Typical adverse effects after subcutaneous injection of sumatriptan include tingling, a warm or hot feeling, heaviness of the upper part of the body, flushing, and a burning sensation of the head, but all these symptoms are mild and transient.
146
Transient or pressure symptoms in the chest occur in 3–5% of these patients. There is no evidence that the pressure is of cardiac origin. In the postmarketing surveillance, the incidence of cardiac ischemia has been extremely low, occurring in 1 in 1 million migraine attacks treated with sumatriptan. Some of the cardiac symptoms reported were related to the use of sumatriptan in patients with preexisting cardiac disease or concomitant risk factors. Misinterpretation of clinical symptomatology (e.g., misdiagnosing a stroke in evolution for a migraine attack) and inappropriate dosing were also factors in previously reported complications of therapy with sumatriptan. For subcutaneous injection, a 6-mg dose is the most effective and has the fewest side effects. For oral dosing, 25- or 50- mg tablets are used. Approximately 35% of patients have recurrence of the pain in 24 h, but data indicate that sumatriptan can be repeated in such cases with prompt relief of the recurrence. A practical way of using sumatriptan is as an injection followed by tablets. The injection gives immediate relief and the tablets continue to maintain adequate blood drug levels, and so pain does not recur within the first few hours.
Dihydroergotamine For the treatment of severe to very severe migraine, parenteral medications may have to be used, since these patients are extremely sick with nausea and vomiting. Intravenous DHE is the drug of choice in such a situation. Dihydroergotamine can be combined with intravenous (IV) metoclopramide, prochlorperazine, or chlorpromazine. Dihydroergotamine is a highly effective medication for acute migraine attacks but is underutilized. Its advantages include minimal arterial constriction and intravenous administration with far less nausea compared with ergotamine. One study comparing DHE with narcotics
2491
2492
146
Treatment of headache
showed that DHE is superior to meperidine and butorphanol tartrate in the acute treatment of migraine headache in the emergency room [36]. No physical dependence has been reported with DHE use. Peak plasma levels are attained in 15–45 min with subcutaneous injection, 30 min with intramuscular injection, 2–11 min with IV injection, and 30–60 min with intranasal administration. Self-injection by patients is possible. Dihydroergotamine has an affinity for 5-HT1A and 5-HT1D receptors, and this probably accounts for its antimigraine effects. It also has an affinity for 5-HT2 adrenergic receptors and dopaminergic receptors. Its affinity for dopaminergic receptors may account for the nausea that can occur as a side effect. The advantage of sumatriptan over DHE is that it has specific affinity for 5-HT1 receptors only and has no effects on adrenergic or dopaminergic receptors; this accounts for the lack of nausea and vomiting as a side effect. As was discussed earlier, DHE not only acts at the 5-HT1 receptor sites and the trigeminal vascular system but also is bound to dorsal raphe nuclei and other brain stem serotoninergic nuclei. Therefore, the action may also be central [18]. Status migrainosus is defined in the International Headache Society’s classification as a prolonged migraine attack that lasts more than 72 h; is associated with nausea, vomiting, and other gastrointestinal (GI) symptoms; and is totally incapacitating. The patient usually presents to the physician after having taken fairly large quantities of pain medications and usually is dehydrated. Hospitalization, IV fluids, and repeated injections of intravenous DHE for about 24–72 h may be necessary to relieve the headache. Various protocols are available for the use of repetitive injections of DHE [37,38]. In all of them, an initial test dose of 0.34 mg of DHE plus 5 mg of metoclopramide or prochlorperazine is given, followed by 0.5 mg of DHE with either of the two antiemetics every 6 h for 48–72 h.
Most patients are able to tolerate the medications used in these protocols. Intolerance to DHE can occur in a few patients because of severe nausea or vomiting in spite of antiemetics. Very rarely, its use may be limited by acute myalgia involving the lower extremities and numbness or paresthesia. Vasospastic reaction with angina has been reported but is rare. Repetitive IV DHE is the mainstay in the treatment of status migrainosus at present. The concomitant use of narcotics is not recommended. In fact, analgesics and narcotics have no place in the treatment of status migrainosus and very little place in the treatment of recurrent episodes of acute migraine. Frequent use of analgesics and narcotics may result in the transformation of episodic migraine into chronic daily headache [39].
Alternative Therapy for Acute Migraine Intravenous Prochlorperazine Patients who are not responsive to sumatriptan or IV or intramuscular DHE can be given IV prochlorperazine (Compazine) [40]. Compazine (5–10 mg) can be given intravenously in an emergency room setting. Dystonic reactions are possible in some patients who receive Compazine and can be counteracted by intramuscular injection of 1 mg of benztropine mesylate (Cogentin). Intravenous chlorpromazine (Thorazine) may also be worth trying in patients who do not respond satisfactorily to sumatriptan or DHE. From 12.5 to 25 mg (0.1 mg/kg) of chlorpromazine given in a piggy back IV is effective in many patients with acute migraine. Repeat dosing every 15 min up to a total of three doses may become necessary. Orthostatic hypotension is a distinct side effect, and patients have to be monitored for a while before they are allowed to get up and walk around or go home. They have
Treatment of headache
to be warned about the possibility of orthostatic hypotension.
Narcotics and Sedatives in Acute Migraine Most migraine attacks can be managed without narcotics, using the medications mentioned above. The reasons why narcotics are not preferred in migraines are as follows: 1.
2.
3.
Serotonin mechanisms are disturbed in migraine, and medications such as sumatriptan, ergotamine, and dihydroergotamine are 5-HT1D receptor agonists that reduce the neurogenic inflammation associated with migraine attacks in addition to their vasoconstrictive effect, whereas narcotics reduce pain without having any specific effects on the neurogenic inflammation or vasodilatation. Narcotics and analgesics with sedatives may in fact produce rebound headache phenomena and perpetuate the chronicity of migraine. With frequent use, habituation occurs.
146
Codeine, Esgic, Esgic with Codeine, Esgic Plus, and Phrenilin. While these preparations are useful for a person with occasional migraine, they are certainly not recommended for patients with frequent episodes of migraine or tension-type headache. Not only do these medications have a potential for abuse, they also invariably produce rebound headache phenomena when used frequently. The excessive use of medications containing butalbital results in lethargy, sleepiness, lack of concentration, and an overall sedated feeling. One of the other major dangers in the use of butalbital-containing medications is that abrupt discontinuation may result in withdrawal phenomena such as increased headache, nausea, irritability, sleeplessness, and even seizures. Because of these problems with the combination medications containing butalbital, the author does not recommend them for routine use. If one has to use a narcotic oral pain medication, a combination of acetaminophen with 30 mg of codeine is probably the least controversial.
Prophylaxis of Migraine Headaches Prophylactic Pharmacotherapy
For these reasons, narcotics are very rarely recommended for acute attacks of migraine. However, in a person who is totally nonresponsive to sumatriptan, ergotamine, DHE, and phenothiazines, one may use parenteral narcotics such as meperidine (Demerol) in a limited way; however, it should not be prescribed on a routine basis.
Combination of Analgesics with Sedatives A number of preparations are available that contain butalbital with acetaminophen or aspirin and caffeine with or without codeine, including Fiorinal, Fiorinal #3, Fioricet, Fioricet with
The decision to start prophylactic pharmacotherapy, which has to be made on a daily basis, depends totally on the impact of migraine on the patient. This impact depends on the frequency, the severity, the disability headache produces, and the comorbid factors. In general, the following are the broad indications for prophylactic pharmacotherapy: 1.
2. 3.
Two or more attacks per month that are disabling and result in inability to work or function Severe, prolonged, disabling attacks even if they occur less than twice a month Inability to cope with migraine episodes
2493
2494
146 4. 5. 6.
Treatment of headache
Failure of abortive therapy Serious side effects from abortive therapy Failure of nonpharmacological approaches
. Table 146‐8 Medications used in the prophylactic treatment of migraine Medication
Table 146‐8 lists the currently used prophylactic agents for migraine.
>
Steps Before Prophylactic Pharmacotherapy The following steps should precede prophylactic pharmacology: 1.
2. 3.
Women of childbearing age should practice contraception, preferably barrier contraception Vasodilators should be avoided as much as possible Concomitant daily analgesics should be avoided
One should be alert to the fact that many of these patients have comorbid disorders, making the treatment difficult and the prognosis less than satisfactory. The comorbidity includes depression, anxiety, panic episodes, bipolar illness, and neuroticism. Analgesic and ergotamine rebound must be recognized, and those who are on daily analgesics or ergotamine should be detoxified from those medications before prophylactic pharmacotherapy is instituted. It is difficult to assess the success of prophylactic therapy because of variability in migraine frequency and severity and the tendency for spontaneous improvement for prolonged periods. In addition, it is well known that migraine can come in cycles in an unpredictable fashion, and this also makes the assessment of prophylactic therapy difficult. In some situations, tachyphylaxis to medications is observed. It is important to give adequate time for prophylactic therapy to work. To judge the effectiveness of any medication used, one should treat
Dosage
Beta-adrenergic blocking agents Propranolol (Inderal) 40–160 mg/day in divided doses Propranolol long-acting 60–160 mg once daily (Inderal-LA) Nadolol (Corgard) 40–160 mg once daily Timolol (Blocadren) Up to 20 mg twice daily Metoprolol (Lopressor) 50–100 mg/day Pindolol (Visken) 10–30 mg/day Atenolol (Tenormin) 50–100 mg/day Antidepressants Tricyclic antidepressants Amitriptyline (Elavil, 25–200 mg at bedtime Endep) Doxepin (Sinequan, 10–100 mg at bedtime Adapin) Nortriptyline (Aventyl, 10–50 mg at bedtime Pamelor) Imipramine (Tofranil) 25–150 mg at bedtime Desipramine (Pertofrane, 25–50 mg, at bedtime Norpramin) Selective serotonin reuptake inhibitors Fluoxetine (Prozac) 20 mg daily in the morning Sertraline (Zoloft) 50 mg at bedtime Paroxetine (Paxil) 20 mg daily in the morning Monoamine oxidase inhibitors Phenelzine (Nardil) 15 mg three times daily Isocarboxazid (Marplan) 10 mg four times daily Calcium channel blockers Verapamil (Calan, Isoptin, 80–360 mg/day Verelan) Flunarizine (Sibelium)a 10–30 mg/day Diltiazem (Cardizem) 60–90 mg three times/day Nicardipine (Cardene) 20 mg three times/day Nimodipine (Nimotop) 30 mg three times/day Serotonin antagonists Methysergide (Sansert) 4–8 mg/day Cyproheptadine (Periactin) 8–16 mg/day Pizotifen (Sandomigran)a Anticonvulsants Divalproex sodium 500–1,500 mg/day (Depakote) Phenytoin (Dilantin) 100–300 mg/day Alpha-adrenergic agonist Clonidine (Catapres) 0.1–0.2 mg three times a day a
Not available in the United States
Treatment of headache
the patient at least for 2–3 months, particularly in the case of calcium channel blockers, which should be given for at least 3 months in adequate doses before they are judged to be ineffective. It is always good to start with small doses and gradually increase the dose in accordance with the tolerance of the patient. When patients are withdrawn from prophylactic therapy, this has to be done gradually.
Some Reasons for Prophylactic Treatment Failure There are several possible reasons for treatment failure: 1. 2. 3. 4. 5. 6.
Wrong diagnosis Not recognizing comorbidity Not recognizing analgesic rebound phenomena Inadequate dose Inadequate treatment period Unrealistic expectations on the part of the patient as well as the physician
Beta-adrenergic Blocking Agents Beta-adrenergic blockers are considered the first line of treatment for migraine prophylaxis at present. Propranolol and timolol have been approved by the U.S. Food and Drug Administration (FDA), whereas the other agents, such as nadolol, have not been specifically approved. It is better to start with small doses and increase the dose gradually. If a person is not adequately responsive to one particular beta blocker, another can be tried. There is no correlation between the efficacy of beta blockers and their ability to enter the CNS, their membrane stimulating properties, their 5-HT-blocking properties, or beta receptor
146
selectivity. Beta blockers are best suited for patients with migraine who are under stress and are anxious. They are suitable for patients with migraine and hypertension.
Contraindications and Adverse Effects of Beta Blockers Beta blockers are contraindicated in patients with active asthma, hypotension, congestive cardiac failure, and diabetes mellitus. Their main drawback is the side effects, which include weight gain, lethargy, extreme tiredness, and depression. Many patients have associated depression as a comorbid disorder, and beta blockers are not suitable for such patients.
Tricyclic Antidepressants Tricyclic antidepressants, particularly amitriptyline, are widely used for migraine. They are not as efficacious as beta blockers. The side effects, which include weight gain and sleepiness, anticholinergic effects such as dry mouth, blurred vision, and dysuria, may become a problem. In selecting a tricyclic, one should take into consideration the anticholinergic effects. If a person has many symptoms pointing to anticholinergic activity, the patient may be switched from amitriptyline to nortriptyline, which has less of an anticholinergic effect. Those who want sedation at night may be tried on doxepin. Tricyclic antidepressants, particularly amitriptyline, have a central analgesic effect that has been shown to reduce the firing rate of the trigeminal nucleus caudalis. The antidepressant effect helps patients with migraine, and the hypnotic effect is helpful in many patients. Tricyclic antidepressants are particularly useful in patients with frequent attacks of migraine, migraine with medication overuse, migraine with sleep disorders, migraine with tension-type headache, and migraine with depression.
2495
2496
146
Treatment of headache
Serotonin Reuptake Inhibitors
Serotonin Antagonists
Many patients prefer specific serotonin reuptake inhibitors (SSRIs) such as fluoxetine and sertraline. The less sedating effect of these medications make them attractive. They have not clearly been shown to have any antimigraine effect; however, they are good adjuncts in the treatment of migraine, especially in patients who cannot tolerate tricyclic antidepressants or experience unpleasant side effects. Occasionally there can be a complication referred to as serotonin syndrome in patients who are on SSRIs, lithium, or monoamine oxidases and also receive 5-HT1 agonists such as sumatriptan and DHE for acute attacks. We recently reported six patients with such a syndrome [41]. The manifestations of the syndrome consist of a combination of acute mental changes and neurological symptoms that include motor weakness, incoordination, myoclonus, and hyperreflexia. There may also be autonomic symptoms such as increased sweating, tachycardia, and fever. These are usually transient phenomena that occur in close proximity to the intake of the acute abortive agent. The recovery is complete. An increase in the available serotonin at the central synapses is thought to be the cause of this syndrome.
Methysergide (Sansert) is probably the most effective medication for the prophylaxis of migraine. Seventy percent of patients benefit from it, usually with doses of about 6–8 mg daily. The immediate side effects include muscle pains in the leg, water retention, swelling, discoloration, and telangiectasia of the ankle area. The most worrisome side effects are fibrotic reactions, which may occur in the retroperitoneal or pulmonary tissue or in the cardiac valves. The overall incidence of fibrotic reactions is very low. It appears that this is an idiosyncratic reaction and does not have any relation to the dose used or the length of treatment. However, the FDA has special instructions for using methysergide, which include a drug holiday for 2 months after 6 months of use. Patients who receive repeated courses of methysergide, have to be monitored carefully with chest x-ray, echocardiogram, and computed tomography (CT) of the abdomen to rule out fibrotic reactions. Very rarely, renal failure occurs without any warning; therefore, these patients have to be followed carefully.
Calcium Channel Blockers There are no good, adequately controlled studies on calcium channel blockers; however, many physicians with long-term experience find calcium channel blockers such as verapamil useful, especially in patients with complicated migraine (migraine with neurological symptoms such as basilar or hemiplegic migraine). Verapamil does cause water retention and constipation and should be used with care in patients with cardiac disorders. Verapamil is, of course, the most useful for the prophylaxis of cluster headache.
Cyproheptadine Cyproheptadine is useful for children, especially young children, with migraine. Four milligrams three or four times a day is the dose in children; weight gain is sometimes a problem, and drowsiness may occur.
Divalproex Sodium in the Prophylaxis of Migraine Divalproex sodium (valproic acid) is the latest addition to the armamentarium of drugs for the prophylaxis of migraine. Sodium valproate has been shown to reduce the neurogenic inflammation in Moskowitz’s experimental model [42].
Treatment of headache
It also has been shown to cause attenuation of c-fos activation in the trigeminal nucleus caudalis in Moskowitz’s experimental animals [43]. Valproate is known to increase gamma-aminobutyric acid (GABA) levels in the brain. In four separate double-blind studies, sodium valproate and divalproex sodium were shown to be superior to placebo [44–47]. Divalproex sodium also has been shown to be as effective as propranolol [48]. Divalproex sodium usually is given in small doses to start with, such as a 250-mg tablet twice a day, to be increased to a total of 1,000–1,500 mg per day. Starting with small doses and gradually increasing them in smaller increments will prevent the excess nausea and vomiting that may occur in some patients. The side effects of sodium valproate include asthenia, weight gain, hair loss, and tremor. Patients show an all-or-none response to valproate. Those who respond, respond very well and remain responsive for long periods. There is no point continuing the medication in those who do not show any response in a few weeks. Divalproex sodium is the only approved drug for migraine prophylaxis that has no direct cardiovascular effects. It can be used as a second-line drug in many patients and a firstline drug in the prophylactic treatment in the following situations: 1.
2. 3.
When beta blockers are contraindicated, as in asthma, congestive cardiac failure, low blood pressure, cardiac conduction defects, depression, patients with immunotherapy for allergies who cannot take beta blockers, patients who cannot tolerate excercise intolerance on beta blockers When the patient has comorbid migraine and epilepsy When the patient has comorbid migraine and bipolar illness
It should be noted that valproate has been approved for use in bipolar illness as well as epilepsy and will soon be available for migraine prophylaxis.
146
Prioritization of Prophylactic Pharmacotherapy Table 146‐9 is a summary that helps to prioritize prophylactic migraine therapy. Clinical efficacy, scientific proof of efficacy, and side effect potential are graded from + to ++++, with ++++ being the highest grade. This is an empirical grading based on a review of the literature and the experience of the authors and is modified from Tfelt-Hansen and Welch [49]. The beta blockers methysergide and valproate are the most effective. One has to choose between them on the basis of their side effect potentials. Because the side effect potential of methysergide is high, it is not considered a firstline drug. On the basis of this assessment, beta blockers and valproate are more or less equal in performance. >
Continuity of Care Patients with migraine need continuity of care. Tachyphylaxis to medications is possible; therefore, if the effectiveness diminishes, a change has to be made to another suitable medication. Side effects have to be monitored, and drug interactions must be kept in mind. Comorbid conditions such as depression, anxiety, and neurotic behavior have to be treated with medications as well as with nonpharmacological approaches. Stress management, biofeedback therapy, and individual counseling may help in some patients.
Treatment of Cluster Headache Abortive Treatment of Acute Attacks of Cluster Headache In the treatment of an acute attack of cluster headache, oxygen is the preferred agent [50].
2497
2498
146
Treatment of headache
. Table 146‐9 Clinical efficacy, scientific proof of efficacy, and potential for side effects
Drug
Clinical efficacya
Scientific proof of efficacyb
Side effect potentiala
Examples of side effects (examples of contraindications)
Beta blockers
++++
++++
++
Tiredness, cold extremities, vivid dreams, depression (asthma, brittle diabetes, atrioventricular conduction defects)
Propranolol, metoprolol, atenolol, nadolol, timolol Antiserotonin drugs Methysergide
++++
++
++++
+++
++
+++
Chronic use: fibrotic disorders (cardiovascular diseases) Weight gain, sedation (obesity)
+++
++++
+++
+
+
+
++ ++
+++ +++
++ ++
Dyspepsia, peptic ulcers (active peptic ulcers)
++ + ++
++ + +
++ + ++
Sedation, dry mouth, weight gain (glaucoma) Dry mouth Nausea, diarrhea (ischemic heart disease)
++++
++++
++
Nausea, vomiting, weight gain, tremor, hair loss
Pizotifen Calcium antagonists Flunarizine Verapamil NSAIDs Naproxen Tolfenamic acid Miscellaneous Amitriptyline Clonidine Dihydroergotamine Anticonvulsants Valproate
Sedation, weight gain, depression (depression, parkinsonism) Constipation (bradycardia, atrioventricular conduction defects)
a
The rating is based on a combination of the published literature and the author’s personal experience As judged by the authors (apparently conflicting with the overwhelming majority of comparative trials claiming equipotency of the two drugs; this claim of comparability is probably due to small trials) b
Oxygen inhalation at 8 L/min for 10 min using a mask will abort the attacks of cluster headache in approximately 70% of patients. Our patients with cluster headaches rent portable oxygen tanks. Oxygen may simply delay the headache in some patients; the headache will return after an hour or so. Oxygen inhalation can be combined with ergotamine in a form that is absorbed very rapidly. Ergotamine inhalation, which results in rapid plasma peak levels, is no longer available; therefore, one has to rely on sublingual, suppository, or oral preparations. Plasma peak levels after oral administration take longer time to achieve in acute attacks of cluster headache. Sublingual preparations are erratic in their
absorption pattern and thus are not very reliable. Suppositories are inconvenient for administration in cases of cluster headache, which comes on rapidly without any warning and ceases rapidly. In spite of these disadvantages, some patients respond to a combination of oxygen and ergotamine in an oral, sublingual, or rectal form. One-milligram ergotamine tablets, 2-mg sublingual tablets, or 2-mg suppositories may be tried. Sumatriptan is the drug of choice for acute episodes of cluster headache [51–53]. It is available in 100-mg tablets and 6-mg subcutaneous preparations. Subcutaneous sumatriptan produces a dramatic effect within 15 min of administration. It also can be combined with oxygen.
Treatment of headache
With this combination, the patient should get relief almost immediately and acute attacks should be aborted totally. Repeat administration of sumatriptan is possible, and the drug has not led to tolerance even after repeated use for more than approximately a year in patients with chronic cluster headache. Injectable sumatriptan is available in an autoinjector form and is very easy for patients to self-administer. The advantage of sumatriptan is its rapidity of action. Lack of nausea and vomiting is also a distinct advantage, as it is a specific 5-HTID agonist without any effects on other neurotransmitter receptors. Upper chest discomfort, a burning sensation at the site of injection, and a hot feeling in the body for a short period are the relatively minor side effects of sumatriptan. As the majority of patients with cluster headache are men and usually are heavy smokers, cardiac status has to be evaluated before drug therapy is started. Sumatriptan and ergotamine should not be used in patients with proven coronary artery disease and those who have multiple risk factors for coronary heart disease. Appropriate investigations to exclude ischemic heart disease have to be done before ergotamine, sumatriptan, and DHE are prescribed. DHE administered intramuscularly relieves cluster headache attacks effectively but acts more slowly than does sumatriptan. Since cluster headache occurs one to three times a day on an average, repeated intramuscular injections are painful and impractical. A nasal spray of DHE is under trial. Analgesics and narcotics have no real place in the treatment of cluster headache. It should be noted that the total period of pain from each cluster headache is approximately 45 min and that by the time an oral narcotic is absorbed and takes effect, the pain is usually over. The prescription of narcotic medications will simply lead to excessive use and habituation without any major benefit in terms of pain relief. Combination analgesics containing barbiturate and
146
caffeine (Fiorinal preparations) have no place in the treatment of acute cluster headaches.
Prophylactic Treatment of Cluster Headache Table 146‐10 lists the medications used for the prophylactic treatment of cluster headache. >
Verapamil Among all the medications used for the prophylaxis of cluster headache, verapamil (Calan, Isoptin, Verelan) appears to be the most effective and is the drug of choice. The usual dose is 120 mg three to four times a day, but the dose may have to be increased in some patients. Verapamil should be continued for at least 2–3 weeks after the patient becomes totally free of headaches of the episodic variety. In chronic cluster headache, the length of treatment has to be determined by trial-and-error. Most patients with chronic cluster headache require verapamil for an indefinite period. . Table 146‐10 Prophylactic pharmacotherapy of cluster headache Medication
Dosage
Verapamil (Calan, Isoptin, Verelan) Lithium carbonate (Lithobid, Eskalith, Lithane)a Methysergide (Sansert) Ergotamine (Wigraine, Cafergot) Prednisoneb
120–480 mg per day
Valproate (Depakote) Indomethacin (Indocin)c a
600–900 mg per day
4–8 mg per day 1–2 mg per day 40 mg per day to start with, 2- to 3-week course in decremental doses 500–1,500 mg per day 50–150 mg per day
Recommended for chronic cluster headache Recommended only in short courses to break the cycle if patient is unresponsive to other prophylactic agents c Useful only in chronic paroxysmal hemicrania b
2499
2500
146
Treatment of headache
Ergotamine Combinations of ergotamine and verapamil are known to produce very good results in patients with cluster headache. The dose of ergotamine is 1 mg twice a day, and unlike in migraine, its use in cluster headache does not appear to result in rebound phenomena. However, caution is necessary concerning daily ergotamine use in patients with risk factors for cardiovascular disease. Most cluster headache patients are heavy smokers, and some have hypertension; therefore, these patients have an increased risk for vascular disease.
Lithium Carbonate Lithium carbonate is useful for both episodic and chronic cluster headache prophylaxis. Lithium is administered in divided doses of 300 mg two to three times a day. Lithium becomes effective in less than a week. If it is to be continued, monitoring of the lithium level to keep it in the low therapeutic range of about 0.5–0.6 mEq per liter is necessary. The plasma level of lithium should never exceed 1.2 mEq per liter. Lithium is reasonably well tolerated by most patients. While on lithium, these patients should not take sodium-depleting diuretics, as hyponatremia leads to lithium toxicity. The common side effects of lithium include nausea, vomiting, tremor, and lethargy. Neurotoxicity occurs at higher plasma levels, resulting in ataxia, blurred vision, confusion, and altered consciousness. Lithium can be combined with verapamil or ergotamine tartrate. Combinations of lithium and verapamil are the drugs of choice in the treatment of chronic cluster headache.
Methysergide Methysergide is useful in patients with episodic cluster headache, whereas patients with chronic cluster headache are less responsive to it. One
tablet (2 mg) three to four times a day is the standard dose. The side effects of methysergide are described in the section on the treatment of migraine, above.
Corticosteroids Corticosteroids, particularly prednisone, have a definite place in the prophylactic treatment of cluster headache. The effect is usually dramatic, and these patients stop having cluster headache attacks within a day or two. However, when the corticosteroids are discontinued, the headache may recur with the original frequency. Because of exacerbation after the discontinuation of prednisone and the possibility of hypercorticism developing after frequent and prolonged use, prednisone should be reserved for short courses to break the cycle of headache when agents such as verapamil, ergotamine, lithium, and methysergide are not helpful. The usual dose of prednisone is 20 mg two to three times a day to start with, reduced gradually over a period of 2–3 weeks and then discontinued. The mechanism of action of corticosteroids in cluster headache is not clear; they may suppress the synthesis or release of humoral agents that mediate an attack of cluster headache or may influence neurotransmitters involved in the headache. Corticosteroids modulate serotoninergic pathways in the brain and may affect the hypothalamic biological clock that is disrupted in patients with cluster headache. Some headache specialists use prednisone at the onset of the cluster period along with verapamil. Then the prednisone is tapered off after 2 weeks, and the verapamil is continued for the duration of the cluster period. This is a reasonable alternative approach; however, as was mentioned above, exacerbation of the headache can occur after prednisone is discontinued, even though the chance of that happening is lower when the patients are continued on verapamil.
Treatment of headache
Indomethacin Indomethacin is specific for and always successful in the treatment of chronic paroxysmal hemicrania, which is a variant of cluster headache that occurs mostly in women. The attacks are shortlived, lasting on average 5–10 min, as opposed to cluster headache, which lasts for 45 min to an hour. Multiple attacks (15–20 per day) occur, and autonomic symptoms may accompany the headache. The headache is always unilateral, and there are no remissions, resembling the pattern seen in chronic cluster headache. The therapeutic response to indomethacin can be used as a diagnostic test for chronic paroxysmal hemicrania. The usual dose of indomethacin is 25–50 mg three times a day. As with other NSAIDs, gastric side effects are common with indomethacin. Misoprostol (Cytotec) may help protect the upper GI tract from the effects of indomethacin in patients with chronic paroxysmal hemicrania who need to continue indomethacin for an indefinite period. Those on long-term indomethacin (Indocin) should have renal function tests periodically.
Beta Blockers and Antidepressants Some of the medications that have been proved to be effective in the prophylaxis of migraine, such as beta-adrenergic blocking agents and tricyclic antidepressants, are not particularly useful in the treatment of cluster headache. However, an occasional patient may respond to these medications. In patients with chronic cluster headache who are also depressed, antidepressants may be of value as an adjunct.
146
that do not respond to regular prophylactic therapy. DHE given intravenously every 6 h will invariably break the cycle in 2–3 days. The remission obtained with DHE gives the physician an opportunity to adjust the prophylactic therapy. A course of DHE may put a patient into remission for a considerable period.
Surgical Treatment of Chronic Intractable Cluster Headache Approximately 10% of cluster headaches are chronic. By definition, chronic cluster headache patients have no remission of their headaches for at least a year. The headaches occur more frequently than in the episodic variety and are more difficult to treat medically. Prophylactic medical therapy includes combinations of agents such as verapamil, lithium, ergotamine, methysergide, and valproate and occasional short courses of corticosteroids. Triple therapy using any three of these agents may be the last pharmacotherapeutic strategy in some chronic cluster headache patients. When adequate trials of medical therapy fail completely, surgical treatment may be considered.
Indications for Surgery in Chronic Cluster Headache There are several indications for surgery in chronic cluster headache patients: 1. 2. 3.
Total resistance to medical treatment Strictly unilateral cases Stable psychological and personality profiles, including low proneness to addiction
Dihydroergotamine DHE is useful in breaking the cycle of headache in those with intractable cluster headache attacks
Over the last few decades, a number of procedures have been tried for the surgical treatment of cluster headache (> Table 146‐11).
2501
2502
146
Treatment of headache
. Table 146‐11 Surgical procedures for cluster headache Procedures directed toward sensory trigeminal nerve Alcohol injection into supraorbital and infraorbital nerves Alcohol injection into gasserian ganglion Avulsion of infraorbital, supraorbital, and supratrochlear nerves Retrogasserian glycerol injection Radiofrequency trigeminal gangliorhizolysis Trigeminal sensory root sections Procedures directed toward autonomic pathways Section of greater superficial petrosal nerve Section of nervus intermedius Section or cocainization of sphenopalatine ganglion
Radiofrequency Trigeminal Rhizotomy Among the procedures listed in > Table 146‐11, those directed toward the trigeminal nerve, particularly percutaneous radiofrequency trigeminal rhizotomy, have been the most effective [54–59]. Radiofrequency trigeminal rhizotomy utilizes thermocoagulation of the pain-carrying fibers of the trigeminal nerve. It is a stereotactic procedure. A needle is advanced through the foramen ovale under light general anesthesia. Once the needle is in place in the trigeminal ganglion region, the needle tip can be placed selectively, guided by electrical stimulation, in the individual V1, V2, or V3 roots of the sensory trigeminal nerve. Radiofrequency current is passed to the area of interest producing thermocoagulation and resulting in selective destruction of the pain fibers but maintaining touch sensation. Experience in many centers indicates that approximately 70–75% of patients benefit from radiofrequency trigeminal rhizotomy. In the majority, the cluster headache attacks stop. In a smaller percentage, there is substantial improvement with occasional mild episodes. The results are not completely satisfactory, however, with failure occurring in approximately 15% of patients, often for technical reasons.
Those with excellent and good results retain the improvement for a number of years, and even long-term follow-up for more than 20 years has shown continuing benefit. However, recurrence of pain occurs in approximately 20% of those who initially had excellent or very good results, in which case surgery can be repeated. The recurrence may occur on the opposite side of the head; in our experience, patients with a history of occasional headache on the opposite side may be the candidates to develop significant recurrence on the opposite side. Therefore, we recommend selecting patients with a strictly unilateral history of headache. A number of relatively minor complications can occur, especially in the immediate postoperative period, including transient diplopia, stabbing pain in the distribution of the trigeminal nerve, difficulty in chewing on the side of the lesion, and jaw deviation. These complications are usually transient, and complete recovery is the rule. A more troublesome complication is anesthesia dolorosa, although its incidence is very low. In our series of 98 patients with long-term followup, only 2 had moderately severe anesthesia dolorosa symptoms. Corneal analgesia may be produced by the radiofrequency lesion, in which case the patients have to be instructed to take particular care of their eyes after surgery and to consult an ophthalmologist if there is any sign of corneal infection. Untreated corneal infections can easily result in corneal opacification because of a lack of corneal sensation. The beneficial effects of this procedure, however, far outweigh the complications.
Some Observations Some of our observations over the last 13 years in 98 patients who have received radiofrequency lesions are as follows: 1.
Complete analgesia is necessary to ensure adequate beneficial effects. By comparison,
Treatment of headache
2.
in trigeminal neuralgia, partial analgesia is all that is required If the pain is confined to the orbital, retroorbital, infraorbital, or supraorbital area, a lesion involving the V1 and V2 divisions of the trigeminal nerve is adequate. If the pain also occurs in the temples and in the area of the ear, a lesion of the third division is necessary because the auricular branch of the mandibular nerve supplies the temples and the ear
Retrogasserian Glycerol Injection Retrogasserian glycerol injection was once a popular procedure [59,60]. The disadvantages of this technique, however, are as follows: 1. 2.
3.
Analgesia is, at most, not as complete as with radiofrequency lesioning It is difficult to control the glycerol lesion, whereas with radiofrequency lesions, selective destruction of V1, V2, or V3 is possible Glycerol may seep outside of Meckel’s cave and cause chemical meningitis
Many authors with experience with both radiofrequency lesions and glycerol injections prefer the former procedure. In summary, surgical treatment of cluster headache is a last resort and should be restricted to patients with medically resistant disabling chronic cluster headache. Radiofrequency trigeminal rhizotomy is the surgical treatment of choice. In view of the disability and suffering of chronic cluster headache patients, the benefits of this procedure far outweigh the complications.
References 1. Headache Classification Committee of the International Headache Society. Classification and diagnostic criteria for headache disorders, cranial neuralgias and facial pain. Cephalalgia 1988;8 Suppl 7:9-96.
146
2. Stewart WF, Lipton RB, Celentano DD, et al. The epidemiology of severe migraine headaches from a national survey: Implications of projections to the United States population. Cephalalgia 1991;11 Suppl 11:87-8. 3. Linet MS, Steward WF, Celentano DD, et al. An epidemiologic study of headache among adolescents and young adults. JAMA 1989;261:2211-16. 4. Mathew NT, Reuveni U, Perez FP. Transformed or evolutive migraine. Headache 1987;27:102-6. 5. Raskin NH. Headache. 2nd ed. New York: Churchill Livingstone; 1988. 6. Lauritzen M, Olesen J. Regional cerebral blood flow during migraine attacks by Xenon-133 inhalation and emission tomography. Brain 1984;107:447-61. 7. Olesen J. Cerebral and extracranial circulatory disturbances in migraine: Pathophysiological implications. Cerebrovasc Brain Metab Rev 1991;3:1-28. 8. Saito K, Markowitz S, Moskowitz MA. Ergot alkaloids block neurogenic extravasation in dura mater: proposed action in vascular headache. Ann Neurol 1988;24:732-7. 9. Goadsby PJ, Edvinsson L, Ekman R. Release of vasoactive peptides in the extracerebral circulation of humans and the cat during activation of the trigeminovascular system. Ann Neurol 1988;23:193-6. 10. Goadsby PJ, Edvinsson L. Sumatriptan reverses the changes in calcitonin gene-related peptide seen in the headache phase of migraine. Cephalalgia 1991;11 Suppl 11:3-4. 11. Slatter KH. Some clinical and EEG findings in patients with migraine. Brain 1968;91:85-98. 12. Gawel M, Connolly JF, Rose FC. Migraine patients exhibit abnormalities in the visual evoked potential. Headache 1983;23:49-52. 13. Schoenen J, Maertens de Noordhout A, TimsitBerthier M, et al. Contingent negative variation and efficacy of beta-blocking agents in migraine. Cephalalgia 1986;6:229-33. 14. Welch KM, D’Andrea G, Tepley N, et al. The concept of migraine as a state of central neuronal hyperexcitability. Neurol Clin 1990;8:817-28. 15. D’Andrea G, Cananzi AR, Joseph R, et al. Platelet glycine, glutamate and aspartate in primary headache. Cephalalgia 1991;11:197-200. 16. Raskin NH, Hosobuchi Y, Lamb S. Headache may arise from perturbation of brain. Headache 1987;27:416-20. 17. May A, Weiller C, Juptner M, et al. Brainstem activation in human migraine attacks: A PET study. Cephalalgia 1995;15 Suppl 14:122. 18. Goadsby PJ, Gundlach AL. Localization of 3H-dihydroergotamine-binding sites in the cat central nervous system: relevance to migraineAnn Neurol 1991;29:91-4. 19. Curzon G, Barrie M, Wilkinson MI. Relationship between headache and amine changes after administration of reserpine to migraineurs subjects. J Neurol Neurosurg Psychiatry 1969;32:555-61.
2503
2504
146
Treatment of headache
20. Kimball RW, Friedman AP, Vallego E. Effect of serotonin in migraine patients. Neurology 1960;10:107-11. 21. Sicuteri F, Testi H, Anselmi B. Biochemical investigations in headache: increase in hydroxyindolacetic acid excretion during migraine. Int Arch Allergy Appl Immunol 1961;19:55-8. 22. Curran DA, Hintenberg H, Lance JW. Total plasma serotonin, 5 hydroxyindolacetic acid and p-hydroxi-mmethoxymandelic acid excretion in normal and migrainous subjects. Brain 1965;88:997-1010. 23. Peroutka SJ. 5-hydroxytryptamine receptor subtypes. Annu Rev Neurosci 1988;11:45-60. 24. Buzzi MG, Moskowitz MA. The antimigraine drug, sumatriptan (GR43175), selectively blocks neurogenic plasma extravasation from blood vessels in dura mater. Br J Pharmacol 1990;99:202-6. 25. Markowitz S, Saito K, Moskowitz MA. Neurogenically mediated leakage of plasma protein occurs from blood vessels in dura mater but not brain. J Neurosci 1987;7:4129-36. 26. Goadsby PJ, Edvinsson L, Ekman R. Vasoactive peptide release in the extracerebral circulation of humans during migrain headache. Ann Neurol 1990;28:183-7. 27. Buzzi MG, Moskowitz MA. Evidence for 5-HTIB/ID. receptors mediating the antimigraine effect of sumatriptan and dihydroergotamine. Cephalalgia 1991; 11:165-8. 28. Humphrey PP, Fenink W, Perren MJ, et al. The pharmacology of the novel 5HT1-like receptor agonist, GR43175. Cephalalgia 1989;9 Suppl 9:23-33. 29. Heyck H. Pathogenesis of migraine. Res Clin Stud Headache 1969;2:1-28. 30. Volans GN. Research review: migraine and drug absorption. Clin Pharmacokinet 1978;3:313-18. 31. Hakkarainen H, Vapaatalo H, Gothoni G, et al. Tolfenamic acid is as effective as ergotamine during migraine attacks. Lancet 1979;2:326-8. 32. Klapper JA, Stanton JS. Ketorolac versus DHE and Metoclopramide in the treatment of migraine headaches. Headache 1991;31:523-4. 33. Doenicke A, Brand J, Perrin VL. Possible benefit of GR43175, novel 5-HT1-like receptor agonist, for the acute treatment of severe migraine. Lancet 1988; 1:1309-11. 34. Cady RK, Wendt JK, Kirchner JR, et al. Treatment of acute migraine with subcutaneous sumatriptan. JAMA 1991;265:2831-5. 35. The Subcutaneous Sumatriptan International Study Group. Treatment of migraine attacks with sumatriptan. N Engl J Med 1991;325:316–21. 36. Belgrade MJ, Ling LJ, Schleevogt MB, et al. Comparison of singledose meperidine, butorphanol and dihydroergotamine in the treatment of vascular headache. Neurology 1989;39:590-2. 37. Raskin NH. Repetitive intravenous dihydroergotamine as therapy for intractable migraine. Neurology 1986;36: 995-7.
38. Silberstein SD, Schulman EA, Hopkins MM. Repetitive intravenous DHE in the treatment of refractory headache. Headache 1990;30:334-9. 39. Mathew NT, Kurman R, Perez F. Drug induced refractory headache – clinical features and management. Headache 1990;30:634-8. 40. Jones J, Sklar D, Dougherty J, et al. Randomized doubleblind trial of intravenous prochlorperazine for the treatment of acute headache. JAMA 1989;261:1174-6. 41. Mathew NT, Tietjen GE, Lucker C. Serotonin syndrome complicating migraine pharmacotherapy. Cephalalgia (in press). 42. Lee WS, Limmroth V, Ayata C, et al. Peripheral GABAA receptor-mediated effects of sodium valproate on dural plasma protein extravasation to substance P and trigeminal stimulation. Br J Pharmacol 1995;116:1661-7. 43. Curter FM, Limmroth V, Ayata G, Moskowitz MA. Attenuation by valproate of C-FOS immunoreactivity in trigeminal nucleus caudalis induced by intracisternal capsaicin. Br J Pharmacol (in press). 44. Hering R, Kuritzky A. Sodium valproate in the prophylactic treatment of migraine: A double blind study versus placebo. Cephalalgia 1992;12:81-4. 45. Jensen R, Brinek T, Oleson J. Sodium valproate has a prophylactic effect in migraine without aura: A triple blind placebo cross over study. Neurology 1994;44:647-51. 46. Mathew NT, Saper JR, Silberstein SD, et al. Migraine prophylaxis with divalproex. Arch Neurol 1995;52:281-6. 47. Klapper J. Divalproex sodium in the prophylactic treatment of migraine (abstract). Headache 1995;35:290. 48. Kanieck RG. A comparison of sodium valproate to propranolol hydrochloride and placebo in the prophylaxis of migraine without aura (abstract). Headache 1995; 35:305. 49. Tfelt-Hansen P, Welch KMA. Prioritizing prophylactic treatment in the headaches. In: Olesen J, TfeltHansen P, Welch KMA, editors. The headaches. New York: Raven Press; 1993. p. 403-5. 50. Kudrow L. Response of cluster headache attacks to oxygen inhalation. Headache 1981;21:1-4. 51. The Sumatriptan Cluster Headache Study Group. Treatment of acute cluster headache with sumatriptan. N Engl J Med 1991;325:322–6. 52. Ekbom K, Cole JA. Subcutaneous sumatriptan in the acute treatment of cluster headache attacks. Can J Neurol Sci 1993:20 Suppl 4:F61. 53. Ekbom K, Monstad I, Prusinski A, et al. Subcutaneous sumatriptan in the acute treatment of cluster headache: a dose comparison study. Acta Neurol Scand 1993;88:63-9. 54. Maxwell RE. Surgical control of chronic migrainous neuralgia by trigeminal gangliorhizolysis. J Neurosurg 1982;57:459-66. 55. Onofrio BM, Campbell JK. Surgical treatment of chronic cluster headache. Mayo Clin Proc 1986;61:537-44. 56. Sweet WH, Wepsic JG. Controlled thermocoagulation of trigeminal ganglion and rootlets for differential
Treatment of headache
destruction of pain fibers: I. trigeminal neuralgia. J Neurosurg 1974;40:143-56. 57. Mathew NT, Hurt W. Percutaneous radiofrequency trigeminal gangliorhizolysis in intractable cluster headache. Headache 1988;28:328-31. 58. Taha JM, Tew JM. Long-term results of radiofrequency rhizotomy in the treatment of cluster headache. Headache 1995;35:193-6.
146
59. Waltz TA, Dalessio DJ, Ott KH, et al. Trigeminal cistern glycerol injections for facial pain. Headache 1985;25:354-357. 60. Ekbom K, Lindgren L, Nilsson BY, et al. Retro-Gasserian glycerol injection in the treatment of chronic cluster headache. Cephalalgia 1987;7:21-7.
2505
120 What Have PET Studies Taught Us About Cerebral Mechanisms Involved in Analgesic Effect of DBS? R. Kupers . J. Gybels
Introduction The first therapeutic applications of deep brain stimulation (DBS) in humans began in the late 1950s [1]. These early attempts targeted the septal areas and were inspired by the experiments of Olds and Milner [2] showing that electrical stimulation of the septal area has a rewarding effect in rats. Apart from these rare, precocious attempts, the real interest in DBS for the treatment of pain in humans arose only a decade later. The driving forces thereto were the discovery of stimulationproduced analgesia in the rat [3] and the proposal of the gate control theory [4]. Reynolds’ discovery that electrical midbrain stimulation can produce profound analgesia opened new perspectives for the treatment of pain in humans. The subsequent discovery of the intimate relationship between stimulation-produced analgesia and the endogenous opioid system offered a theoretical explanation for the underlying neurobiological mechanism. DBS for the treatment of chronic pain became relatively popular in the 1970s and 1980s. In sharp contrast with the early optimistic results, it became gradually clearer that the results were not that successful as was initially believed. The clinical results did not match well with the much more optimistic experimental findings in animals, and large discrepancies were noted between the results of different neurosurgical centers. In addition, the lack of controlled clinical #
Springer-Verlag Berlin/Heidelberg 2009
trials to document its therapeutic effectiveness and the absence of a viable theory to explain its mechanism of action hampered further progress. In the early 1990s, the successful initial results with motor cortex stimulation (MCS) in the treatment of certain forms of neuropathic pain [5–7] made many neurosurgeons to abandon DBS. However, inspired by the results of brain imaging findings [8], DBS recently regained interest for the treatment of some forms of headache [9]. In this chapter we will first give a short overview of brain imaging findings of pain. We will first briefly summarize brain imaging findings in acute experimental pain and clinical forms of pain. Next, we will discuss the results of brain imaging studies of DBS and MCS for chronic pain treatment.
Brain Imaging of Pain Imaging of Acute Experimental Pain The pain system is classically divided into a lateral and a medial system. The lateral pain system consists of spinothalamic tract neurons projecting from the ventrobasal nucleus of the thalamus to the primary (SI) and secondary (SII) somatosensory cortex, parietal operculum, and the insula. In contrast, the medial pain system includes the
2032
120
What have PET studies taught us about cerebral mechanisms involved in analgesic effect of DBS?
spinothalamic neurons projecting to the intralaminar and medial thalamic nuclei and further to the anterior cingulate cortex (ACC), the amygdala, the hippocampus and the hypothalamus, the spinoreticular projections to the parabrachial nucleus and the locus coeruleus, and the spinomesencephalic tract projections to the periaqueductal grey (PAG). Whereas the lateral pain system is mainly involved in the sensorydiscriminative aspects of pain processing, the medial pain system plays a crucial role in the motivational-affective and cognitive-evaluative aspects of pain processing, memory for pain, and the autonomic-neuroendocrine responses [10]. Using a wide variety of experimental paradigms, imaging techniques, and types of experimental pain, brain imaging studies have described with relatively high consistency a set of brain areas that are activated by painful stimuli [11]. Among these, the most commonly activated areas are the insula, the ACC, SI and SII, primary (MI) and premotor cortex, supplementary motor area (SMA), thalamus, prefrontal and posterior parietal cortices, basal ganglia, midbrain
(PAG), and cerebellum (> Figure 120-1). Together, this set of brain areas is often referred to as the ‘‘pain matrix.’’ Hence, the so-called pain matrix also includes areas that do not belong to either the lateral or medial pain system. The concept of the pain matrix is controversial and there is no real consensus as to which areas exactly form a part of it. Moreover, it is unclear as to what extent individual parts of the matrix are required for forming the conscious experience of pain. Finally, it should be borne in mind that parts of the so-called pain matrix are also activated by a variety of non-painful stimuli or conditions such as warmth, itch, negative emotions, stress, changes in blood pressure, thirst, and hunger.
Imaging of Clinical Pain Compared to the vast brain imaging literature related to the processing of acute experimental pain and the psychological processes involved in its top-down modulation, the literature on
. Figure 120-1 Summary of brain imaging findings for acute pain, migraine–cluster headache, and neuropathic pain. Brain areas show increased rCBF or BOLD response (indicated in red) in the three pain conditions. The red arrow in the acutepain condition represents supraspinal areas receiving spinal nociceptive input. Black circles refer to brain areas showing no pain-induced changes. Black lines show corticocortical connections between different areas within the pain-matrix (adapted after [12]). Abbreviations. SI: primary somatosensory cortex; SI: secondary somatosensory cortex; PPC: posterior parietal cortex; DLPF: dorsolateral prefrontal cortex; OFC: orbitofrontal cortex; SMA: supplementary motor cortex; ACC: anterior cingulate cortex; Ins: insula; Thal: thalamus; LN: lenticular nuclei; NA: nucleus accumbens; Amy: amygdala; Hyp: hypothalamus; PAG: periaqueductal grey; DP: dorsal pons; Cereb: cerebellum; MI: primary motor cortex
What have PET studies taught us about cerebral mechanisms involved in analgesic effect of DBS?
clinical pain processing is rather sparse. In addition, methodological problems make the results of most studies difficult to interpret or to generalize to a larger population [12]. Finding homogeneous patient populations has been proven difficult for most reports with the exception of studies in migraine and cluster headache, possibly explaining why the data in these conditions are the most converging and conclusive ones of the chronic pain literature.
120
reports on chronic nociceptive pain processing. A functional magnetic resonance imaging (fMRI) study showed that chronic low back pain mainly activates the medial prefrontal cortex, including the perigenual ACC [21]. A recent positron emission tomography (PET) study in patients with rheumatoid arthritis reported that arthritic pain is associated with increased activity in ACC, thalamus, and amygdala [22]. These are the areas that are involved in the processing of fear, emotions, and aversive conditioning.
Cluster Headache and Migraine Neuropathic Pain Brain imaging studies have shown that both spontaneous and evoked migraine headache are associated with an increase in regional cerebral blood flow (rCBF) in dorsal median brain stem structures [13,14,15]. These studies have further revealed that the dorsal pontine activation is lateralized to the side of the migraine attack [16]. Interestingly, most studies also reported a strong activation in the prefrontal cortex and cerebellum but not in other parts of the classical pain matrix. In contrast, cluster headache is associated with a prominent activation of the ipsilateral inferior hypothalamus [17–19]. Additional rCBF increases are observed in the insula, right ACC, right inferior frontal cortex, contralateral thalamus, and cerebellum but not in SI, SII, or the brain stem. Interestingly, patients with cluster headache also show structural changes in the hypothalamus [18], and DBS of the inferior hypothalamus offers clinical improvement in these patients [20].
Chronic Nociceptive Pain If we exclude the studies on nociceptive pain with an unknown pathophysiology (low back pain, fibromyalgia, irritable bowel syndrome, etc.), there are surprisingly very few brain imaging
Several studies have investigated the cerebral response pattern to painful, usually allodynic, stimulation in a nerve-injured territory. Limitations of these studies are that they suffer from small and heterogeneous patient samples and that they use stimulation of the homologous contralateral side for control stimulation [12]. An early PET study [23] showed that cold allodynia in patients with Wallenberg syndrome is associated with abnormal responses in the lateral pain pathways (contralateral ventroposterior thalamus and SI). A more recent study from the same group [24] assessed brain responses to cold and tactile allodynic stimulation in a large group of patients with neuropathic pain of mixed peripheral and central origin. The results differed from the earlier study and showed that both forms of allodynia recruit novel responses in ipsilateral SI, SII, and insula but are not associated with enhanced responses in thalamus and ACC. A PET study [25] in a small group of pain patients with peripheral neuropathic pain showed that brushing of the allodynic area provoked bilateral rCBF increases in the thalamus and SII, the contralateral SI, ACC, right anterior insula, the brainstem, and the cerebellum. A similar study in a larger group of patients with peripheral neuropathic pain [26] showed that brush-evoked allodynia activates the orbitofrontal cortex, a
2033
2034
120
What have PET studies taught us about cerebral mechanisms involved in analgesic effect of DBS?
structure with extensive connections with limbic and mesencephalic structures and which can induce analgesic effects upon electrical stimulation. This study also reported a preponderance of ipsilateral responses in the SII/insular region during allodynic stimulation. A recent study investigated hyperalgesic responses in patients with complex regional pain syndrome (CRPS)-type I [27]. In contrast with the above studies, CRPS-type I does not involve obvious deafferentation of primary afferent input and therefore the results cannot be affected by deafferentation-induced cortical reorganization. It was reported that punctuate mechanical hyperalgesia was associated with increased responses in SI, SII, contralateral anterior and ipsilateral posterior insula, and prefrontal and posterior parietal cortices [27].
the use of fMRI; so we have to rely on PET methodology. An inherent problem with brain imaging studies of neurostimulation procedures is the timing of the conditions. Because of the long stimulation after-effect, randomization of the scans over time is not possible. Once an analgesic effect is obtained, it may take several hours before the pain returns to baseline levels after stopping DBS. This explains why baseline (pain) scans must be acquired at the beginning of the experimental session. It can therefore be argued that the observed rCBF changes between the pain and pain-free states might reflect monotonic task-independent time effects rather than pain-related changes. For instance, time-related rCBF increases [29] and decreases [30] in the ACC have been reported.
Brain Imaging Studies of Therapeutic Effect of Deep Brain Stimulation
Deep Brain Stimulation for Cluster Headache
The possibility to use modern brain imaging techniques to study the mechanisms mediating the analgesic effects of DBS is very appealing. The uniqueness of the approach is that it allows the study in vivo of the functional role of the underlying cerebral networks by scanning patients with the stimulator switched on and off. Before the introduction of these techniques, it remained unclear by which of the following mechanisms DBS exerts its analgesic effect: (1) by a direct activation of descending pain-inhibitory pathways in the PAG and/or rostral ventromedial medulla (RVM), (2) by modulation of activity in the pain matrix, or (3) by local inhibition of the brain area stimulated. The first brain imaging studies to probe the mechanisms underlying the analgesic effects of DBS date back to the mid-1980s [28]. The marriage of modern brain imaging techniques with deep brain implants is, however, a complicated one. The presence of magnetic susceptible implanted material excludes
A recent PET study examined the changes in rCBF in 10 patients with cluster headache with a DBS system in the hypothalamus [31]. The authors used a block design in which hypothalamic DBS on and off scans were alternated. The results showed that hypothalamic stimulation led to significant rCBF increases in the hypothalamus, the ipsilateral thalamus and insular cortex, contralateral SI, and ACC (> Figure 120-2). Interestingly, DBS was not associated with either a reduction of activity within the pain matrix or with increased activity within key structures of the descending pain modulatory system (PAG, RVM), excluding a pure antinociceptive effect. These findings seem to suggest that DBS of the hypothalamus depolarizes this structure and modulates activity in key areas involved in the rostral transmission of noxious information. Although interesting, a few critical remarks need to be made. First, the scans were taken when patients were in their headache-free period. We therefore do not know what the results would
What have PET studies taught us about cerebral mechanisms involved in analgesic effect of DBS?
120
. Figure 120-2 Brain imaging findings in cluster headache. (a) Increased blood flow response in the hypothalamus during a cluster headache attack [17]. (b) Structural changes in the hypothalamus of patients with cluster headache [18]. (c) Hypothalamic stimulation for pain relief in cluster headache produces a significant increase in blood flow in the hypothalamus [31]
have been had the patients been scanned during a headache attack. Second, clinically it takes sometimes days or even weeks between turning on the hypothalamic stimulator and a change in pain perception. This raises the question as to what the relevance of these rapid rCBF changes is in explaining the much slower long-term clinical effects. The anatomical connectivity of the hypothalamus was recently investigated in vivo, using diffusion tensor imaging (DTI) techniques. There, the surgical target from the postoperative MRI scan of a patient with cluster headache with a hypothalamic implant was used as a seed for probabilistic tractography in a group of healthy control subjects [32]. The results showed highly consistent connections between the hypothalamus and the reticular nucleus, the orbitofrontal cortex and the cerebellum, suggesting a possible anatomical network in the genesis of cluster headache. Interestingly, no direct anatomical connections were demonstrated with areas that showed increased or decreased rCBF responses during hypothalamic stimulation as measured by PET [31]. There may be methodological explanations for this discrepancy such as the fact that tractography preferentially labels large fiber pathways, or the possibility that patients with cluster headache have altered anatomical connectivity of
the hypothalamus. Alternatively, the effect of hypothalamic stimulation on these areas is indirect.
Thalamic Stimulation for Neuropathic Pain Direct electrical stimulation of the somatosensory thalamus (ventroposterior lateral and medial; VPL/VPM) is exclusively used for treatment of neuropathic pain. The technique was pioneered in the early 1970s [33,34] and has been used with widely varying degrees of success during the following decades [35]. Nevertheless, despite numerous clinical studies reporting pain relief, the success of thalamic stimulation for the treatment of chronic pain has always remained poorly predictable [35]. Furthermore, evaluation of stimulation-produced pain relief is difficult because there can be large placebo effects [36]. The use of VPL/VPM stimulation in the treatment of neuropathic pain was largely inspired by the gate control theory, according to which the stimulation of large diameter fibers is capable of inhibiting nociceptive information. Since in certain neuropathic pain syndromes there are no primary afferent fibers in the peripheral nerve or spinal cord dorsal columns to serve
2035
2036
120
What have PET studies taught us about cerebral mechanisms involved in analgesic effect of DBS?
as substrate for stimulation, stimulation of the specific somatosensory relay nuclei of the thalamus is an alternative for activating the lemniscal system. Mazars and colleagues, however, practiced VPL-VPM stimulation already in the early 1960s, i.e., before the formulation of the gate control theory [33,34]. Their theoretical framework was the theory of Head and Holmes, according to which pain results from an imbalance between protopathic and epicritic sensory functioning [37]. Stimulation of the thalamic sensory relay nuclei would increase the epicritic component, thereby inhibiting the protopathic inflow. Electrophysiological studies in anesthetized animals confirmed that VPL thalamic stimulation inhibits the activity of both spinothalamic nociceptive neurons and thalamic parafascicular nociceptive neurons. A limited number of PET studies have been performed in patients treated with thalamic stimulation. As early as 1986, at the dawn of the era of noninvasive in vivo brain imaging in humans, Katayama and coworkers used PET to investigate the possible mechanisms of VPLinduced analgesia [28]. These investigators examined changes in cerebral activity in five patients with neuropathic pain treated with VPL stimulation. Brain responses measured during thalamic stimulation were significantly increased in frontal, postcentral, and thalamic regions, predominantly in the hemisphere ipsilateral to the stimulation site. These data therefore lend support to the original hypothesis that the effect of somatosensory thalamic stimulation is mediated by activation of tactile thalamocortical afferents. However, since few methodological details were provided, the data could not be interpreted unequivocally. We had to wait for more than a decade before a second PET study on thalamic stimulation was published. Duncan and colleagues [38] investigated rCBF changes in five patients with peripheral neuropathic pain for whom electrical stimulation of the somatosensory thalamus produced satisfactory long-term pain relief. Patients participated in two PET scanning
sessions, conducted on consecutive days. Each session consisted of four scanning conditions: a prestimulation baseline, an early stimulation scan (1 min after onset of the thalamic stimulator), a late stimulation scan (conducted 30 min after onset of the stimulation), and a scan which was conducted 5 min after thalamic stimulation was terminated. The most prominent rCBF change in response to thalamic stimulation was a large activation in the region approximating the thalamic stimulation site itself, contralateral to the patient’s clinical pain problem. This increase in blood flow was already present during the early stimulation scan but became much stronger in the late stimulation scan. The largest and only significant cortical activation site was localized within the rostral insula, ipsilateral to the thalamic stimulation (> Figure 120-3). A small and nonsignificant blood flow increase was measured in S1 ipsilateral to the site of thalamic stimulation and in the ACC. When rCBF measures before and after thalamic stimulation were compared, the only significant difference was a reduced activity in the posterior thalamus after stimulation. This possibly reflects a decrease in nociceptive sensory transmission underlying the therapeutic effects of thalamic stimulation. Analyses for patients with and without immediate pain reduction showed that the thalamic activation during DBS was more robust in the patients with immediate pain relief. In addition, the three patients with immediate pain relief also showed significant activation in the rostral insula, whereas those without immediate relief showed only a nonsignificant activation in this region. Although neither group activated S1, S2, or ACC, the subjects without immediate relief showed a tendency toward increased rCBF in S1 cortex. Interestingly, in addition to tactile paresthesiae, patients described the thalamic stimulation-induced sensation in terms of temperature (hot, warm, or cold). The data by Duncan and coworkers do not support the original hypothesis that thalamic
What have PET studies taught us about cerebral mechanisms involved in analgesic effect of DBS?
120
. Figure 120-3 (a) DBS of the VPL leading to a significant increase in rCBF in the anterior insular cortex [38]. (b) Increased blood flow response in the posterior ACC during thalamic stimulation [44]. (c) Amygdalar activation during VPM stimulation in a patient with facial neuropathic pain [45]
stimulation produces analgesia by activating tactile thalamocortical pathways that were inactivated by neuronal damage. The data therefore suggest that alternative mechanisms may be the basis of the analgesic effect of thalamic stimulation. One such mechanism may be the activation of temperature pathways. This is supported by the finding that the most prominent cortical increase in rCBF was found in the anterior insula and by the patients’ reports of stimulationinduced thermal sensations. The anterior insula is not only activated during experimental and neuropathic pain but also following application of warm and cool innocuous temperatures to the skin [39]. In monkeys, both nociceptive and thermoreceptive neurons were observed within the rostral insula [40], and this region receives a strong projection from the posterior portion of the ventral medial (VMpo) thalamic nucleus, which is subjacent to VPM and contains both nociceptive and thermoreceptive neurons. Stimulation in the region of human VMpo can evoke painful but more commonly thermal sensations [41]. It has been proposed that the temperature projection involving VMpo and anterior insula is an important pain-inhibitory pathway [42]. In conclusion, the results of this study suggest
that activation of temperature pathways is one of the candidate mechanisms of pain relief by thalamic stimulation. It may therefore seem paradoxical that most neurosurgeons select their stimulation target on the basis of stimulationinduced tactile paresthesiae and not of thermal sensations. However, microstimulation studies in patients during stereotactic procedures have revealed that the sites for evoking tactile and thermal sensations are very close to each other [43]. It is therefore likely that the bipolar or quadripolar stimulation electrodes that are often used stimulate neurons within both the VMpo-insular and VPL/VPM-S1 pathways. Davis and colleagues [44] used PET in five patients with neuropathic pain who had received a thalamic implant 4–18 months before entering the PET study. Although the patients initially reported therapeutic improvement with thalamic stimulation, at the point of the PET study only two of them still benefited from thalamic stimulation. A similar study design as in the previous study was used with brain scans taken before and during (early and late scan) thalamic stimulation and post stimulation. Thalamic stimulation led to rCBF increases in two separate regions within the ACC. The contralateral rostral ACC (BA32)
2037
2038
120
What have PET studies taught us about cerebral mechanisms involved in analgesic effect of DBS?
showed an increased rCBF response during and after thalamic stimulation. This rCBF response was already observed during the scan taken 2 min after switching on the stimulator and remained at the same level in the late scan and in the scans taken after stopping stimulation. A second rCBF response was measured in a more posterior part of the ipsilateral ACC (BA24). This response was present only during the late stimulation scan and recovered immediately after stopping thalamic stimulation. The ACC is a complex structure involved in multiple higher executive, cognitive, emotional, and motor functions. Whereas the rostral ACC is more involved in stimulus awareness and cognitive functioning, the posterior ACC is critically involved in pain intensity processing [45]. Taken together, the results suggest that whereas the response in the rostral ACC is likely related to cognitive factors related to the presence of paresthesiae, the increased posterior ACC response may indicate that thalamic stimulation modulates pain processing by altering activity within this region. There were no significant correlations between rCBF changes and pain relief scores in either ACC site. No further rCBF changes increases were measured in other cortical areas or in the thalamus around the stimulating electrode. A detailed case report of a patient with neuropathic pain who was successfully treated with VPM stimulation was presented by Kupers and coworkers [46]. These authors studied a patient (TG) who developed a sharp, stinging, and shooting pain and hypoesthesia to pinprick and temperature in the right side of the face (V2 area) following a surgical operation. TG was implanted with a stimulation electrode in the left VPM, with which he was able to completely suppress his pain. TG was scanned at rest after a 12-h stimulation-free period, during thalamic stimulation, and after switching the stimulator off again. TG became completely pain-free after switching on the stimulator, an effect which outlasted the actual DBS stimulation period and continued
throughout the scans taken after stopping stimulation. In line with the results by Duncan and coworkers, DBS led to a significant increase in rCF in the anterior insula and a subsignificant increase in the thalamus around the stimulation electrode. Since TG also reported a warmth component in his paresthesiae, these data further support the hypothesis that somatosensory thalamic stimulation activates thermal pathways in addition to tactile pathways. However, the most conspicuous rCBF increase during thalamic stimulation was found in the amygdala. The role of the amygdala in pain and in opioid analgesia is well established. Amygdalar activation during thalamic stimulation may be either via direct thalamo-amygdaloid projections or indirectly via the insular cortex. When comparing scans taken before and after stimulation, it was found that thalamic DBS led to significant decreases in rCBF in prefrontal cortex, anterior insula, and hypothalamus. Significant rCBF increases after DBS were measured in the substantia nigra/nucleus ruber, supplementary motor cortex and posterior thalamus, and/or pulvinar. No rCBF changes were measured in SI, SII, or in the posterior part of the ACC. Although there are important differences in the results of the above studies, a few general conclusions can be drawn. The first conclusion is that low-frequency stimulation as used in DBS for pain treatment activates rather than inhibits neural structures. This is in contrast with the effects of high-frequency stimulation of the thalamus in movement disorders, which produces an inhibition of the stimulated area [47]. Second, the results support only weakly the original hypothesis that the effect of VPL/VPM stimulation is mediated by activation of thalamo-cortical tactile pathways. Although the study by Katayama and coworkers reported activation of the postcentral region, depending on the intensity of stimulation, none of the other studies found evidence for SI activation. Instead, two studies suggest that thalamic stimulation involves activation of thermal
What have PET studies taught us about cerebral mechanisms involved in analgesic effect of DBS?
pathways in the insula. Third, there is little evidence that thalamic stimulation leads to a suppression of activity within the classical pain matrix. This may be explained by the fact that chronic spontaneous pain activates a different set of brain areas compared to acute pain [12,21,22]. Another explanation is that no reductions in activity in the pain matrix were observed because DBS in the PET studies produced at best a moderate decrease in pain. This latter hypothesis is supported by the observation that the only study reporting a powerful DBS-induced pain reduction showed indeed significant reductions in brain activity. Several methodological explanations can be put forward for the lack of congruence across the different studies such as differences in stimulation parameters, in the anatomical target, and in pain pathology, pain distribution and pain history.
PAG-PVG Stimulation No PET studies have been performed in patients with electrodes implanted in the PAG-PVG area. This may be due to the fact that PAG-PVG stimulation is more rarely performed compared to VPL-VPM stimulation. One case report of a single photon emission computerized tomography (SPECT) study in a patient with PVG stimulation reported that PVG DBS decreased perfusion in the sensori-motor cortex, the pons, and the midbrain region and increased perfusion in the lentiform nucleus and in the midbrain region between PVG and thalamus [48]. However, these data are difficult to interpret because of the poor resolution of SPECT data and because no statistical comparisons were made. More interesting data come from a study which measured field potentials in the thalamus during PVG stimulation in eight patients with central pain treated with PVG and/or stimulation [49]. Good pain relief was obtained in seven patients with lowfrequency (5–35 Hz) stimulation. PVG-induced pain relief was associated with a significant
120
reduction of low-frequency thalamic field potentials in the 0.2–0.4 Hz range. Higher-frequency PVG stimulation had no effect on thalamic field potentials and remained devoid of an analgesic effect. Another study used magnetoencephalography (MEG) to study the mechanisms underlying PVG/PAG-induced analgesia in a patient with severe phantom pain [50]. The patient experienced excellent pain relief with lowfrequency (7 Hz) stimulation in the right PVG/ PAG. He was first scanned at rest with the DBS switched on, followed by three scans with the stimulator off. Pain ratings were moderately higher during the last of the stimulator-off scans. This was associated with increased activity in the 10–20 Hz frequency band in SI, SII, insula, lateral orbitofrontal cortex, and rostral ACC. During the stimulation-on period, significant differences in activity were found in the 10–20 Hz band in regions involved in pain relief such as the mid-anterior orbitofrontal and subgenual cingulate cortices. In the 20–30 Hz band, differences were found in the motor and parietal cortices. Interestingly, whereas the mid-anterior orbitofrontal cortex has been shown to be involved also in placebo and opioid analgesia, the subgenual cingulate area seems to play an important role in depression.
Motor Cortex Stimulation for Neuropathic Pain MCS has largely replaced thalamic stimulation as the treatment option of choice for conventional therapy-resistant forms of neuropathic pain. It is a relatively recent technique pioneered in the 1990s. The therapeutic success of MCS outpaces that of classical thalamic and PAG/PVG stimulation. Whereas a theoretical model was present for explaining the analgesic effects of thalamic and PAG-PVG stimulation, a sound theoretical framework for explaining MCS was absent. This has certainly sparked interest in the use of brain
2039
2040
120
What have PET studies taught us about cerebral mechanisms involved in analgesic effect of DBS?
imaging techniques to study the purported mechanisms of MCS-induced analgesia. In addition, MCS was pioneered during the dawn of noninvasive brain mapping in humans. The PET studies on MCS are generally methodologically more sound than those of thalamic stimulation because of the fact that they are often planned in a prospective manner and include larger patient samples. However, brain imaging studies on MCS pose particular challenges owing to the long delay between the onset of stimulation and the analgesic effect. One of the first published series involved 10 patients undergoing MCS for drug-resistant unilateral neuropathic pain [51]. Four of the patients reported an average pain reduction of at least 50% by MCS at the time of investigation. Patients were scanned 15 min before onset of MCS (after the stimulator had been turned off for 18 h), after 5 and 20 min of continuous MCS, and 30 min after stopping MCS. It was found that MCS increased rCBF in the lateral thalamus ipsilateral to stimulation. When a more lenient statistical threshold was used, rCBF increases were also found in the rostral ACC (BA 32), the medial thalamus, and the anterior insula. These rCBF increases were already present in the scan taken 5 min after onset of MCS and remained at this level during the second scan. Blood flow returned to baseline 30 min after stopping MCS in all regions except the ACC. No rCBF changes were measured at the site of stimulation in MI or in the adjacent SI. When the data of the patients with good and poor analgesia were compared, no difference in rCBF increase in the thalamus was found, whereas patients with good analgesia had a significantly higher rCBF increase in the ACC. A more recent PET study from the same group focused in particular on MCS-induced rCBF changes over a longer time span [52]. A large group of 19 patients with unilateral neuropathic pain participated in this prospective study. The average long-term pain relief produced by MCS was 38%. In order to cope with the long, possible
after-effect of MCS, patients refrained from using their stimulator 4 weeks prior to the PET study. Patients first underwent four baseline scans with the stimulator off. Then, four scans were acquired during the 35-min period of MCS after which stimulation was discontinued, and five additional scans were acquired until 75 min after the stimulator had been switched off. Patients were left unaware as to whether the MCS was turned on or off. During MCS, correlations were found between pain relief by MCS and rCBF changes in mid-cingulate cortex, pregenual ACC, and lateral prefrontal cortex. No other activations were found during MCS. The most robust rCBF changes were found in the period after switching off the stimulator. During the off period, correlations with long-term pain relief were observed in the anterior and posterior part of the mid-cingulate cortex, pregenual ACC, orbitofrontal cortex, lateral prefrontal cortex, basal ganglia, supplementary motor area, cerebellum, and PAG. No correlations were found between rCBF changes and changes in pain during the scanning session. A functional connectivity analysis further revealed significant covariation between activity in pregenual ACC and PAG, lower pons, and basal ganglia. Interestingly, increased connectivity between the perigenual ACC, orbitofrontal cortex, and PAG has also been reported in other analgesic procedures such as placebo analgesia and opioid analgesia as well as during cognitive modulation of pain [53]. These data raise some interesting hypotheses with respect to the mechanisms mediating MCS-induced analgesia. First, the most prominent correlations between rCBF changes and long-term clinical pain relief were observed in the late scans when the stimulator had been switched off. This is in agreement with clinical observations that the analgesic effect of MCS becomes apparent only after a delay. Second, the areas showing a correlation between changes in rCBF and long-term relief (pACC, orbitofrontal cortex, PAG) in MCS have also been reported
What have PET studies taught us about cerebral mechanisms involved in analgesic effect of DBS?
to be implicated in many other analgesic procedures. For instance, both ablation and stimulation of the perigenual cingulate [54,55] relieve chronic pain, and stimulation of the PAG/PVG region is used to treat chronic pain in humans (see above). Like the perigeniculate ACC, the PAG/PVG is critically involved in many forms of analgesia such as placebo analgesia and opioid analgesia as well as in the cognitive modulation of pain [53]. Third, all MCS-induced effects involve increases in rCBF; no stimulation-induced rCBF decreases were reported in the pain matrix. Finally, both the perigeniculate ACC and the ACC form important relays in mediating the analgesic effect of opioids. This raises the question whether MCS activates the endogenous opioid system. Surprisingly, no pharmacological
120
studies have evaluated the effect of naloxone on the analgesic effect of MCS. However, Maarawi and coworkers [56] used PET to study MCS-induced changes in opioid binding. Eight patients with refractory neuropathic pain were scanned after the injection of the nonselective opioid tracer [11C] diprenorphine (> Figure 120-4). Patients were scanned before implantation of the MCS device and after 2 months of chronic stimulation. Comparing the results of the scans taken before and after MCS, significant decreases in [11C]diprenorphine binding potential were found in the mid portion of the ACC, prefrontal cortex, PAG, and cerebellum. Decreases in binding potential in PET studies indicate increased endogenous release because endogenous compounds occupy
. Figure 120-4 MCS-induced decreases in [11C] diprenorphine binding in posterior ACC (a) and PAG (b) [56]. MCS led to significant rCBF increases in the posterior part of the ACC (c) and in the motor cortex itself (d) [54]
2041
2042
120
What have PET studies taught us about cerebral mechanisms involved in analgesic effect of DBS?
more receptor sites, leaving fewer receptors available for the exogenously administered PET tracer. These results therefore suggest that MCS-induced analgesia is opioid-mediated. This hypothesis is further supported by the fact that the changes in binding potential in the mid-ACC and PAG showed significant correlations with MCS-induced pain relief as reported by the participants. Although these results fit nicely with the above-reported rCBF changes, it should be noted that whereas the rCBF study reported correlations between rCBF and long-term pain relief in the perigenual ACC, MCS-induced changes in opioid binding were found in the mid-cingulate but not in the perigeniculate cingulum. Kishima and coworkers [57] also used PET to investigate the mechanisms underlying MCS. Although the study only included six patients who had used MCS for at least 6 months, all patients suffered from pain in the same anatomical location. MCS was stopped 12 h before the PET study. Patients first underwent six scans with the stimulator off. Then, MCS was performed for 30 min, and another six scans were carried out with the stimulator switched off. Pain ratings already dropped during the active stimulation period and continued to decrease in the postMCS scans. The average pain relief was around 50%. A comparison of the scans taken before and after MCS revealed significant rCBF increases in the posterior thalamus (pulvinar) and posterior insula contralateral to the side of MCS. When the scans obtained in the early post-MCS period were analyzed separately, rCBF was significantly increased in left posterior insula and right orbitofrontal cortex. In contrast, when the scans in the late post-MCS period were compared with the pre-MCS scans, a significant rCBF increase was found in the posterior part of the midcingulate area (BA24). In contrast with the higher discussed studies, no rCBF increases were found either in perigenual ACC or in PAG. Also, in contrast with these studies is the fact that the activity in MI underlying the stimulation
electrode was significantly reduced in the postMCS scans, indicating that MCS locally inhibited neuronal activity. Since no scans were taken during active MCS in this study, nothing can be concluded as to which neural structures were directly activated by MCS. The same authors also reported a case report of a patient who suffered from neuropathic pain in the left upper limb following a stroke in the right thalamus [58]. This patient became nearly completely pain free with stimulation of the left motor cortex. Interestingly, the patient reported a sensation of warmth which invaded the painful limb during MCS. PET data showed increased rCBF in the left dorsolateral prefrontal cortex (BA9), left orbitofrontal cortex, left thalamus, and left perigeniculate ACC (BA32) in the pre-MCS scans compared to the post-MCS ones. Although these data partly agree with those reported by Kupers and coworkers [42] in a patient treated with thalamic stimulation, the lack of stimulation-induced rCBF increases in perigenual ACC and PAG is at odds with the results of the larger MCS studies.
Methodological Issues in Brain Imaging Studies of DBS As already mentioned at the beginning of this chapter, the use of brain imaging techniques in DBS is not straightforward, and many technological and methodological problems have to be dealt. First, the presence of a stimulation lead in the brain and an internal pulse generator in the body precludes the use of MRI-based techniques. It should be noted, however, that a few reports have been published in which MRI-based techniques were used in patients with implanted systems [59]. This basically leaves only PET as an option for investigation. Since PET uses radioactive tracers, only a limited number of scans can be performed in a patient, making follow-up studies very difficult or impossible. Another
What have PET studies taught us about cerebral mechanisms involved in analgesic effect of DBS?
disadvantage of PET compared to fMRI is that it has much poorer temporal and spatial resolution. It should, however, be acknowledged that one study used MEG in a patient with DBS systems. Further studies are needed to confirm whether MEG would represent a viable technique in DBS. A major advantage of MEG is its temporal resolution in the millisecond range, which allows study of the interactions between activities in different brain regions. Besides these technological issues for which a satisfactory solution might be found one day, a number of important methodological problems remain. One of these concerns the timing of the scans. Because of the often very long after-effects of DBS, in particular of MCS, randomization of the stimulation on and off scans cannot be done. Therefore, the ‘‘stimulator off ’’ scans always precede the ‘‘stimulator on’’ scans, which by definition precede the ‘‘after stimulation’’ scans. This makes it impossible to rule out with certainty whether any changes observed when comparing with the baseline state are due to genuine physiological effects of DBS or represent a timing effect. Some efforts have been made to deal with this problem. For instance, Peyron and coworkers [52] looked at linear changes over time in rCBF which occurred during the four initial baseline scans and used these as a mask for the further analysis. Although this may solve the problem partly, there is no guarantee that the spontaneous fluctuations in cerebral blood flow measured over a relatively short time span of around 30 min are similar to those which would occur over a much longer time span (up to 3 h). A way around this problem would be to scan a group of control patients over the same time span but without switching on the stimulator and look for rCBF increases and decreases over the whole time period. Another methodological problem is that most reported studies used small patient samples and included patients who varied greatly not only in pain pathology and history but also in terms of duration of DBS usage. This will make it
120
more difficult to detect relevant rCBF changes, particularly in areas with a known somatotopic organization. Another challenge is to dissociate stimulation effects from analgesic effects. This is particularly the case in studies of DBS in which the analgesic effect has a shorter delay onset than in MCS. If DBS results in a significant amount of pain relief during the scanning session, it is not be possible to figure out with certainty whether the observed changes in rCBF reflect direct downstream effects of stimulation of the DBS target or are a reflection of induced changes in pain perception. Again, several solutions have been proposed to this. One is to consider the results of the early brain scan immediately after switching on the stimulator separately since the early brain scan is only rarely associated with pain relief. A trade-off is that this reduces the number of scans and hence the difficulty to detect statistically significant changes in rCBF. An alternative approach has been to switch the stimulator on only within the relatively short time window of the PET data acquisition (usually around 60 s) and to leave it off during the 10-min break separating successive scans. The disadvantage of this approach is that it uses a form of DBS which is different from the one that produces analgesic effects in the patient’s daily life. The question is whether the rCBF changes reported in such a design are representative of what happens in the brain during uninterrupted stimulation. A final major methodological issue is how to dissociate genuine stimulation effects from placebo effects. Like in any other medical therapeutic intervention, placebo effects may largely contribute to the therapeutic outcome [36]. Although placebo procedures could be included in a brain imaging study, the problem is how to run a placebo-controlled study in DBS whereby both the patient and the examiner stay blind with respect to the treatment condition. The presence of distinct paresthesiae during DBS makes it very hard to come up with a convincing placebo condition. The situation is
2043
2044
120
What have PET studies taught us about cerebral mechanisms involved in analgesic effect of DBS?
easier for MCS where stimulation is devoid of conscious subjective experiences. It should be noted that some of the rCBF changes reported during, for instance, MCS remarkably resemble placebo-induced rCBF changes. For instance, placebo analgesia has been shown to be associated with increased rCBF in perigenual ACC and brainstem and with increased functional connectivity between these two areas. This is not to say that the rCBF changes observed in MCS should be interpreted as placebo effects but they warrant caution.
Concluding Remarks and Directions for Future Studies In this chapter we reviewed data on the use of brain imaging studies to explore the mechanisms mediating DBS-induced analgesia. Although such studies remain challenging, they offer a unique possibility to probe into the underlying biological mechanisms. Brain imaging studies in patients treated with thalamic stimulation are historically the oldest ones. They therefore suffer also the most from methodological shortcomings such as small and widely heterogeneous study samples, lack of placebo controls, etc. Notwithstanding, the results of these studies have revealed some new insights into the possible mechanisms mediating somatosensory thalamic stimulation-induced analgesia. For instance, they provided little support for the original hypothesis that the analgesic effect of thalamic stimulation is based upon activation of tactile thalamocortical pathways. Instead, these studies suggest that activation of thermal pathways in the insular cortex may contribute to the analgesic effect. In addition, activation of the rostral ACC was also reported. Although PAG/PVG stimulation is being practiced since the 1970s to treat some forms of chronic pain, it has attracted surprisingly little interest from people within the brain imaging community. With the exception
of one case report study using MEG, no brain imaging studies have been published in patients with PAG/PVG stimulation. Much therefore still needs to be done for exploring the mechanisms mediating this particular form of analgesia. MCS and hypothalamic stimulation have been pioneered much later, during the dawn of modern brain imaging investigations. Since these procedures lacked a firm hypothesis for their analgesic effect, brain imaging studies offered a good opportunity to gain deeper insight into the pathways through which stimulation of these areas produce their analgesic effect. These studies are methodologically more sound since they used larger and more homogeneous patient populations and were planned in a prospective rather than a retrospective manner. Although the number of published reports remains small, particularly for hypothalamic stimulation, the reported results fit nicely within a larger set of data related to the pathophysiology of the syndrome which all point in the direction of the hypothalamus as the generator of the symptoms. None of the brain imaging studies on MCS found evidence for the original theory that MCS produces its analgesic effects through orthodromic activation of neurons in SI. Instead, the available studies rather suggest that activation of limbic pathways, in particular the perigeniculate ACC and its connection with the PAG, plays a crucial role in mediating the analgesic effect of MCS. This hypothesis seems a viable one since brain imaging studies of a wide variety of analgesia-inducing procedures have shown activation of the same pathway. Since the perigenual ACC and the PAG are structures which are both rich in opioids, the data further suggest that endogenous opioids may also be involved in the analgesic effect of MCS. This was supported by a PET receptor binding study showing MCSinduced changes in opioid binding in the brain. However, not all pieces of the puzzle fit nicely. First, MCS is often used as a treatment of last resort, which means after all other pharmacological treatments (including high oral doses of
What have PET studies taught us about cerebral mechanisms involved in analgesic effect of DBS?
opioids) have failed. Second, the hypothesis that MCS exerts its effect through activation of opioidergic pathways is at odds with the results of pharmacological studies showing that the magnitude of MCS-induced pain relief correlated positively with the pre-MCS analgesic response to barbiturates and the N-methyl D-aspartate (NMDA) blocker ketamine but not with morphine [60]. Another caveat is that the results reported by different groups tend to show important differences. Whereas studies by one team showed a critical role played by the perigenual ACC and the PAG, other studies failed to show the functional implication of these structures in the therapeutic effects of MCS. Clearly, more studies are needed to resolve these controversies.
8.
9.
10.
11.
12.
13.
14.
Acknowledgments Ron Kupers is supported by grants from the Svend Andersen Foundation and the Lundbeck Foundation.
References 1. Heath RG, Mickle WA. Evaluation of seven years’ experience with depth electrodes in human patients. In: O’Doherty, editors. Electrical stimulation studies on the unanesthetized brain. New York: Hoerber; 1960. p. 214-47. 2. Olds J, Milner P. Positive reinforcement produced by electrical stimulation of the septal area and other regions of the rat brain. J Comp Physiol Psychol 1954;47:419-27. 3. Reynolds DV. Surgery in the rat during electrical analgesia induced by focal brain stimulation. Science 1969;164:444-5. 4. Melzack R, Wall PD. Pain mechanisms: a new theory. Science 1965;150:971-9. 5. Tsubokawa T, Katayama Y, Yamamoto T, Hirayama T, Koyama S. Chronic motor cortex stimulation for the treatment of central pain. Acta Neurochir Suppl (Wien) 1991;52:137-9. 6. Tsubokawa T, Katayama Y, Yamamoto T, Hirayama T, Koyama S. Chronic motor cortex stimulation in patients with thalamic pain. J Neurosurg 1993;78:393-401. 7. Meyerson BA, Lindblom U, Linderoth B, Lind G, Herregodts P. Motor cortex stimulation as treatment of
15.
16.
17.
18.
19.
20.
21.
22.
120
trigeminal neuropathic pain. Acta Neurochir Suppl (Wien) 1993;58:150-3. May A, Bahra A, Buchel C, Frackowiak RS, Goadsby PJ. Hypothalamic activation in cluster headache attacks. Lancet 1998;352:275-8. Leone M, Franzini A, Bussone G. Stereotactic stimulation of posterior hypothalamic gray matter in a patient with intractable cluster headache. N Engl J Med 2001;345:1428-9. Willis WD, Westlund KN. Neuroanatomy of the pain system and of the pathways that modulate pain. J Clin Neurophysiol 1997;14:2-31. Peyron R, Laurent B, Garcia-Larrea L. Functional imaging of brain responses to pain. A review and meta-analysis. Neurophysiol Clin 2000;30:263-88. Kupers R, Kehlet H. Brain imaging of clinical pain states: a critical review and strategies for future studies. Lancet Neurol 2006;5:1033-44. Weiller C, May A, Limmroth V, Juptner M, Kaube H, Schayck RV, et al. Brain stem activation in spontaneous human migraine attacks. Nat Med 1995;1:658-60. Afridi SK, Giffin NJ, Kaube H, Friston KJ, Ward NS, Frackowiak RS, et al. A positron emission tomographic study in spontaneous migraine. Arch Neurol 2005;62: 1270-5. Bahra A, Matharu MS, Buchel C, Frackowiak RS, Goadsby PJ. Brainstem activation specific to migraine headache. Lancet 2001;357:1016-7. Afridi SK, Matharu MS, Lee L, Kaube H, Friston KJ, Frackowiak RS, et al. A PET study exploring the laterality of brainstem activation in migraine using glyceryl trinitrate. Brain 2005;128:932-9. May A, Bahra A, Buchel C, Frackowiak RS, Goadsby PJ. PET and MRA findings in cluster headache and MRA in experimental pain. Neurology 2000;55:1328-35. May A, Ashburner J, Buchel C, McGonigle DJ, Friston KJ, Frackowiak RS, et al. Correlation between structural and functional changes in brain in an idiopathic headache syndrome. Nat Med 1999;5:836-8. Sprenger T, Boecker H, Tolle TR, Bussone G, May A, Leone M. Specific hypothalamic activation during a spontaneous cluster headache attack. Neurology 2004;62: 516-7. Leone M, Franzini A, Broggi G, May A, Bussone G. Long-term follow-up of bilateral hypothalamic stimulation for intractable cluster headache. Brain 2004;127: 2259-64. Baliki MN, Chialvo DR, Geha PY, Levy RM, Harden RN, Parrish TB, et al. Chronic pain and the emotional brain: specific brain activity associated with spontaneous fluctuations of intensity of chronic back pain. J Neurosci 2006;26:12165-73. Kulkarni B, Bentley DE, Elliott R, Julyan PJ, Boger E, Watson A, et al. Arthritic pain is processed in brain areas concerned with emotions and fear. Arthritis Rheum 2007;56:1345-54.
2045
2046
120
What have PET studies taught us about cerebral mechanisms involved in analgesic effect of DBS?
23. Peyron R, Laurent B, Garcia-Larrea L, Gregoire MC, Convers P, Lavenne F, et al. Allodynia after lateralmedullary (Wallenberg) infarct. A PET study. Brain 1998;121:1345-56. 24. Peyron R, Schneider F, Faillenot I, Convers P, Barral FG, Garcia-Larrea L, et al. An fMRI study of cortical representation of mechanical allodynia in patients with neuropathic pain. Neurology 2004;63:1838-46. 25. Petrovic P, Ingvar M, Stone-Elander S, Petersson KM, Hansson P. A PET activation study of dynamic mechanical allodynia in patients wih mononeuropathy. Pain 1999;83:459-70. 26. Witting N, Kupers RC, Svensson P, Jensen TS. A PET activation study of brush-evoked allodynia in patients with nerve injury pain. Pain 2006;120:145-54. 27. Maihofner C, Forster C, Birklein F, Neundorfer B, Handwerker HO. Brain processing during mechanical hyperalgesia in complex regional pain syndrome: a functional MRI study. Pain 2005;114:93-103. 28. Katayama Y, Tsubokawa T, Hirayama T, Kido G, Tsukiyama T, Iio M. Response of regional cerebral blood flow and oxygen metabolism to thalamic stimulation in humans as revealed by positron emission tomography. J Cereb Blood Flow Metab 1986;6:637-41. 29. Rajah NM, Hussey D, Houle S, Kapur S, McIntosh AR. Task-independent effect of time on rCBF. NeuroImage 1998;7:314-25. 30. Paus T, Zatorre RJ, Hofle N, Caramanos Z, Gotman J, Petrides M, et al. Time-related changes in neural systems underlying attention and arousal during the performance of an auditory vigilance test. J Cogn Neurosci 1997;9: 392-408. 31. May A, Leone M, Boecker H, Sprenger T, Juergens T, Bussone G, et al. Hypothalamic deep brain stimulation in positron emission tomography. J Neurosci 2006;26: 3589-93. 32. Owen SL, Green AL, Davies P, Stein JF, Aziz TZ, Behrens T, et al. Connectivity of an effective hypothalamic surgical target for cluster headache. J Clin Neurosci 2007;14:955-60. 33. Mazars G, Me´rienne L, Cioloca C. Stimulations thalamiques intermittentes antalgiques. Note Prelim Rev Neurol 1973:128:273-9. 34. Mazars G, Me´rienne L, Cioloca C. Traitement de certains types de douleurs par des stimulateurs thalamiques implantables. Neuro-Chirurgie 1974;2:117-24. 35. Gybels JM, Kupers R. Management of persistent pain by brain stimulation. In: Schmidek H, Sweet W, editors. Operative neurosurgical techniques. Philadelhia, PA: Saunders; 2000. p. 1639-51. 36. Marchand S, Kupers RC, Bushnell MC, Duncan GH. Analgesic and placebo effects of thalamic stimulation. Pain 2003;105:481-8. 37. Head H, Holmes G. Sensory disturbances from cerebral lesions. Brain 1911;34:102-254.
38. Duncan GH, Kupers RC, Marchand S, Villemure JG, Gybels JM, Bushnell MC. Stimulation of human thalamus for pain relief: possible modulatory circuits revealed by positron emission tomography. J Neurophysiol 1998;80:3326-30. 39. Craig AD, Chen K, Bandy D, Reiman EM. Thermosensory activation of insular cortex. Nat Neurosci 2000;3: 184-90. 40. Craig AD, Bushnell MC, Zhang ET, Blomqvist A. A thalamic nucleus specific for pain and temperature sensation. Nature 1994;372:770-3. 41. Davis KD, Lozano RM, Manduch M, Tasker RR, Kiss ZH, Dostrovsky JO. Thalamic relay site for cold perception in humans. J Neurophysiol 1999;81:1970-3. 42. Craig AD, Reiman EM, Evans A, Bushnell MC. Functional imaging of an illusion of pain. Nature 1996;384: 258-60. 43. Lenz FA, Seike M, Richardson RT, Lin YC, Baker FH, Khoja I, et al. Thermal and pain sensations evoked by microstimulation in the area of human ventrocaudal nucleus. J Neurophysiol 1993;70:200-12. 44. Davis KD, Taub E, Duffner F, Lozano AM, Tasker RR, Houle S, et al. Activation of the anterior cingulate cortex by thalamic stimulation in patients with chronic pain: a positron emission tomography study. J Neurosurg 2000;92:64-9. 45. Bu¨chel C, Bornhovd K, Quante M, Glauche V, Bromm B, Weiller C. Dissociable neural responses related to pain intensity,stimulus intensity,and stimulus awareness within the anterior cingulate cortex: a parametric single-trial laser functional magnetic resonance imaging study. J Neurosci 2002;22:970-6. 46. Kupers RC, Gybels JM, Gjedde A. Positron emission tomography study of a chronic pain patient successfully treated with somatosensory thalamic stimulation. Pain 2000;87:295-302. 47. Dostrovsky JO, Lozano AM. Mechanisms of deep brain stimulation. Mov Disord 2002;17 Suppl 3:S63-8. 48. Pereira EA, Green AL, Bradley KM, Soper N, Moir L, Stein JF, et al. Regional cerebral perfusion differences between periventricular grey, thalamic and dual target deep brain stimulation for chronic neuropathic pain. Stereotact Funct Neurosurg 2007;85:175-83. 49. Nandi D, Aziz T, Carter H, Stein J. Thalamic field potentials in chronic central pain treated by periventricular gray stimulation a series of eight cases. Pain 2003;101: 97-107. 50. Kringelbach ML, Jenkinson N, Green AL, Owen SL, Hansen PC, Cornelissen PL, et al. Deep brain stimulation for chronic pain investigated with magnetoencephalography. Neuroreport 2007;18:223-8. 51. Garcia-Larrea L, Peyron R, Mertens P, Gregoire MC, Lavenne F, Le Bars D, et al. Electrical stimulation of motor cortex for pain control: a combined PET-scan and electrophysiological study. Pain 1999;83:259-73.
What have PET studies taught us about cerebral mechanisms involved in analgesic effect of DBS?
52. Peyron R, Faillenot I, Mertens P, Laurent B, GarciaLarrea L. Motor cortex stimulation in neuropathic pain. Correlations between analgesic effect and hemodynamic changes in the brain. A PET study. Neuroimage 2007;34: 310-21. 53. Kupers R, Faymonville ME, Laureys S. The cognitive modulation of pain:hypnosis- and placebo-induced analgesia. Prog Brain Res 2005;150:251-69. 54. Ballantine HT Jr. Cassidy WL, Flanagan NB, Marino R Jr. Stereotaxic anterior cingulotomy for neuropsychiatric illness and intractable pain. J Neurosurg 1967; 26:488-95. 55. Spooner J, Yu H, Kao C, Sillay K, Konrad P. Neuromodulation of the cingulum for neuropathic pain after spinal cord injury. Case report. J Neurosurg 2007;107:169-72. 56. Maarrawi J, Peyron R, Mertens P, Costes N, Magnin M, Sindou M, et al. Motor cortex stimulation for pain control induces changes in the endogenous opioid system. Neurology 2007;69:827-34.
120
57. Kishima H, Saitoh Y, Osaki Y, Nishimura H, Kato A, Hatazawa J, et al. Motor cortex stimulation in patients with deafferentation pain: activation of the posterior insula and thalamus. J Neurosurg 2007;107: 43-8. 58. Saitoh Y, Osaki Y, Nishimura H, Hirano S, Kato A, Hashikawa K, et al. Increased regional cerebral blood flow in the contralateral thalamus after successful motor cortex stimulation in a patient with poststroke pain. J Neurosurg 2004;100:935-9. 59. Rezai AR, Baker KB, Tkach JA, Phillips M, Hrdlicka G. Is magnetic resonance imaging safe for patients with neurostimulation systems used for deep brain stimulation? Neurosurgery 2005;57:1056-62. 60. Yamamoto T, Katayama Y, Hirayama T, Tsubokawa T. Pharmacological classification of central post-stroke pain: comparison with the results of chronic motor cortex stimulation therapy. Pain 1997;72:5-12.
2047
166 Anterior Nucleus DBS in Epilepsy M. Hodaie . C. Hamani . R. Wennberg . W. Hutchison . J. Dostrovky . A. M. Lozano
Introduction Neuromodulation has emerged as an attractive strategy for the treatment of intractable epilepsy. A significant percentage of patients with medically intractable epilepsy are not good candidates for resective surgery. Neuromodulation offers a unique advantage to these patients. First, neuromodulation aims to address the neuroanatomical substrates thought to play a role in the propensity for seizure generation and propagation. Second, neuromodulation offers the ability to titrate the stimulation parameters, potentially providing improved benefit with minimal side effects. Third, the non-destructive effect of neuromodulation is important in minimizing the potential side-effects of the surgical procedure.
Neuroanatomical Rationale The anterior nucleus of the thalamus (AN) forms part of the limbic system and circuit of Papez. Afferent projections to the AN come from the hippocampus and mamillary bodies through the mamillothalamic tract (tract of Vicq D’Azyr). Efferent projections of the AN travel to the cingulate cortex and parahippocampal cortex, from there widely spreading to the neocortex. The anterior thalamic area is located in the anterior and medial part of the thalamus. It is commonly subdivided into several subnuclei. The principal nucleus or nucleus anterior principalis is the largest. The other nuclei are the anterior dorsal and anterior medial groups, also known as accessory nuclei [1,2]. Surrounding #
Springer-Verlag Berlin/Heidelberg 2009
structures include the large dorsomedian nucleus (DM) inferiorly and posteriorly. The DM has also been shown to have an important role in seizure maintenance and propagation, particularly for those seizures involving limbic structures [3,4]. Studies in the cat using retrograde horseradish peroxidase technique (HRP) show strong afferent input into the mamillary nuclei from subicular regions, frontocortical and cingulate projections as well as subcortical projections including the septum, the diagonal band of Broca and from the periaqueductal gray [5–7]. The projections of the mamillary bodies have been studied in further detail in the rat [8]. There appear to be topographic representation of the mamillary subnuclei to the anterior thalamus, with projections of the pars medialis centralis connecting almost exclusively to the anteromedial thalamic nucleus, while anteroventral thalamic nucleus receives primarily projections from the mamillary body pars medialis dorsalis and pars lateralis. Furthermore, nuclei of the dorsal half of the mamillary subdivisions project primarily to the medial pars of the anteroventral thalamus. With respect to cortical projections to the AN, cortical limbic areas send bilateral projections to the anteromedial nucleus as well as parvicellular anteroventral thalamic nucleus. Projections originate from layers V and VI of the limbic cortex. The reticular nucleus (RE), a thin, reticulated structure that allows crossing of all axons connecting the dorsal thalamus with the neocortex, is thought to be the pacemaker of spindle oscillations, with an important role in cortically
2794
166
Anterior nucleus DBS in epilepsy
generated spike-wave seizures. RE has strong connections to the anterior, lateral and ventral surfaces of dorsal thalamus [9,10]. The afferent input to the RE nucleus appears to arise from the brainstem and basal forebrain, with primary input from mesopontine cholinergic nuclei, locus ceruleous and dorsal raphe. It is thought that the generation of spindling activity in thalamic neurons is mediated through the RE, which has strong and reciprocal GABAergic connections with most thalamic nuclei, and is responsible for the EEG synchrony in spindle rhythms [11]. There are conflicting thoughts regarding the connections of the AN to the RE. Experiments in the cat suggest that the RE lacks connectivity to the AN, since the latter appear to lack spindling activity [12]. More recent work in the rat and monkey however suggest otherwise, with evidence of reciprocal connections of the AN with RE [13].
Animal Studies Based on the neuroanatomical rationale, a number of animal experiments have demonstrated a functional role of the AN in epilepsy. These experiments are primarily based on acute animal models of epilepsy, with targets being either the anterior thalamic area or its connections, such as the mamillary bodies (MB) or mamillothalamic tract (MTT). The initial animal experiments that illustrated the role of deep subcortical areas in the mediation of seizures were brought to the forefront by the work of Gale et al. who in the early 1980’s reported that lesioning of the substantia nigra or its pharmacological inhibition resulted in marked decrease in seizure activity, as demonstrated by decrease in hindlimb extension in rats, as part of maximal electroshock seizures [14–16]. Mirski and Ferrendelli reported initially in 1986 their work on bilateral interruption of the MTT in guinea pigs. Using an acute rat pentylenetetrazol
(PTZ) seizure model, animals received a GABA agonist into the anterior thalamus, resulting in protection from PTZ seizures, but not against maximal electroshock seizures (MES). Interestingly, injection into the SN resulted in protection against MES seizures but not PTZ seizures, suggesting separate neuronal circuits mediating each type [17]. In addition, injection of excitatory agents such as kainic acid (KA) into the AN resulted in facilitation of EEG convulsant action of PTZ. Protection against PTZ seizures was also observed when the AN was injected with muscimol, a GABA agonist. This effect was only seen when the injection was performed bilaterally. Unilateral injection did not prevent the hypersynchronous EEG discharges following PTZ, however it did result in delayed cortical hypersynchrony ipsilateral to the injected nucleus [18]. Using an ethosuximide-induced suppression of PTZ seizures with concurrent EEG recording and regional 14C-deoxyglucose in guinea pigs, selective uptake of label was observed in MB, MTT, AN, mamillary peduncles and the dorsal and ventral tegmental nuclei of the midbrain, suggesting the importance of these areas in seizure activity induced by PTZ as well as the anticonvulsant activity of ethosuximide [19]. In addition, bilateral electrolytic lesions of the MTT resulted in complete protection of the behavioural and EEG convulsant action and effect of PTZ [20]. The role of AN in other acute models of epilepsy, such as the pilocarpine model, has also been investigated. In an acute rat pilocarpine model, bilateral anterior thalamic lesions and high frequency stimulation appeared to be protective against both seizures and onset of status epilepticus. This effect was not observed with unilateral lesioning or with insertion of electrodes alone [21]. One important finding in animal studies relates to the effect of the frequency of stimulation. High frequency (100Hz) stimulation of the AN has been shown to be anticonvulsant in the PTZ rat model, whereas low frequency (8 Hz)
Anterior nucleus DBS in epilepsy
166
. Table 166-1 Clinical characteristics of patients with AN DBS [24]
Age (yr)
Age at epilepsy onset (yr)
Sex
Seizure type
Epilepsy classification
1 2 3
45 36 22
1 2 1
F F M
Symptomatic generalized Symptomatic generalized Multifocal vs. symptomatic generalized
4
30
5
F
5
19
1
M
GTC GTC Atonic drop attacks, complex partial vs. atypical absence, secondarily GTC Complex partial, secondarily GTC Partial motor, secondarily GTC
Patient
Partial with secondary generalization, right frontal maximum Multifocal/partial with secondary generalization, bifrontal, right > left
GTC generalized tonic–clonic seizure; DBS deep-brain stimulation. (Reproduced with permission, Blackwell.)
stimulation appears to be proconvulsant [22]. This is in keeping with the observation that high frequency stimulation of the AN or nonspecific intralaminar nuclei results in EEG desynchronization, which is thought to render the cortex less susceptible to seizures, whereas low frequency stimulation results in synchronization of the EEG and a classical EEG pattern of recruiting rhythm, which is believed to result in increased cortical susceptibility to seizures [11]. A recent report on the effect of patterns of deep brain stimulation in animals with DBS in the AN area has demonstrated, however, no significant difference in seizures with low or high frequency stimulation [23].
Clinical Studies DBS of AN in humans has focused primarily on patients with intractable epilepsy, having frequent disabling seizures and lacking a suitable response to multiple antiepileptic drugs (AEDs) [24–26]. All patients need to be investigated with scalp-EEG video monitoring, and found to have bilateral and/or non-localizing findings that preclude consideration of resective treatment for epilepsy, as well as adequate MR studies demonstrating no lateralizing structural abnormalities.
The availability of a caregiver that can assess and record daily seizure activities is also crucial for the assessment of the results. A number of clinical studies have been done to look at the effectiveness of AN DBS in epilepsy. In the initial clinical study reported [24], 5 patients were treated, with epilepsy classifications including symptomatic generalized; multifocal; partial with secondary generalization (> Table 166‐1). Patient ages ranged between 19 and 45. The study by Kerrigan et al. [26] reports a second set of five patients in whom four demonstrated decreased secondary generalization of seizures, while one of these patients demonstrated decreased seizure frequency. The large multicenter SANTE trial (Stimulation of the Anterior Nucleus of the Thalamus for Epilepsy) started in 2003 [27]. A blinded trial of the assessment of AN stimulation will be helpful in determining the degree of clinical benefit from this procedure. A recent long term study [25] suggests that the seizure benefit may last beyond 5 years.
Surgical Technique The basic technique is similar to most stereotactic neuromodulation procedures. Our current technique is described in the following stages:
2795
2796
166 1.
2.
3.
4.
Anterior nucleus DBS in epilepsy
Frame placement: We typically place the Leksell stereotactic frame under local anesthesia. The anterior pins are placed approximately one inch above the eyebrows, and using a local block at the area of the supraoptic nerve. The desired orientation of the frame is parallel to a line between the lateral canthus and the tragus of the ear, thereby aligning as much as possible the anterior and posterior commissures. Imaging with Leksell frame: The typical sequence characteristics include a full head FSPGR study performed in 1.5 mm slices, no gap (1.5T, GE) and T2 study through the area of the thalamus. No gadolinium contrast is given. The coordinates for the anterior (AC) and posterior commissures (PC) and midcommissural points are calculated based on the MRI images. The typical coordinates for the AN are: 6 mm lateral to the midcommissural point, 8 mm anterior to the posterior commissure, and 12 mm above the AC/PC line. The angle trajectory is approximately 60 in the anterior superior to posterior inferior direction. If a neuronavigation system is chosen to approach the target, consideration might be given of choosing the entry point also, preferably located over a gyrus and avoiding any deep sulci or venous structures. Once in the operating room the head is shaved and the area of the coronal suture widely prepped and draped. Two parasagittal burr holes are drilled. The dura is opened and the pia is coagulated. We typically place fibrin glue over the burr hole to minimize CSF loss, which may result in shifting of brain structures with resultant change in the relative position of the calculated coordinates. Note is made that fibrin glue is highly thrombogenic and not indicated for placement over the brain tissue itself. Ideally, once the electrodes are placed, these can
5.
6.
7.
8.
be surrounded with gelfoam and the fibrin glue placed over them, thereby minimizing the direct contact of the fibrin glue with the brain. Intraoperative microelectrode recordings are carried out. It should be noted that the trajectory is invariably through the lateral ventricle. Once the ventricular wall is crossed, the thalamus is encountered with characteristic action potentials. Our recordings typically start 10 mm above the intended target and extend between 5–10 mm beyond intended target. Electrophysiological findings during recordings are detailed below. After the microelectrode recordings and under fluoroscopic guidance quadripolar DBS electrodes (Medtronic model 3387, Medtronic, Minneapolis, MN, USA) are placed in the intended target. Given the discrepancy between the size of the anterior nucleus and the electrode, the bottom contact (Contact 0) typically rests in the dorsomedian nucleus. The anterior nucleus spans approximately 6 mm, while the contacts of the electrode used span 10.5 mm. Therefore at least one electrode must be either deep to the anterior nucleus or in the ventricle. After placement of the DBS electrodes, the wires are connected to extension cables and externalized. This completes the first stage of the operation. The externalized wires are marked for identification of the appropriate side and the patient is discharged to the epilepsy monitoring unit (EMU). The patient is continued on the same AEDs as in the preoperative period. Monitoring in the EMU allows for videoEEG recordings with scalp and thalamic depth electrodes. Stage 2 involves internalization of the DBS electrodes and connection to a pulse generator. This procedure is performed under general anesthesia.
Anterior nucleus DBS in epilepsy
Electrophysiological Findings of Recordings of the AN Upon entering the thalamus during microelectrode recordings, an increased background noise as well as presence of action potentials is noted. No specific electrophysiological boundaries could be distinguished between the AN and its surrounding nuclei, such as DM or NC (> Figure 166‐1). We commonly observed that many of the neurons were firing in short, high frequency bursts [28]. This potentially represents a normal neuronal property of cells in this area, however it may also be a consequence of either the patients’ epilepsy or an effect of AEDs. The majority of the bursting activity had characteristic calcium spikes with increasing interspace intervals for successive spikes within
166
a burst [29]. The finding of characteristic low threshold calcium bursting in the thalamus in awake individuals is thought to represent an abnormal condition possibly related to the patients’ seizure disorders. This finding has been reported elsewhere and related to pathological conditions, including epilepsy [29–33]. Alternatively, these characteristic bursts could be related to the use of anticonvulsant medication.
Effect of AN Stimulation on EEG Recordings Recording from the AN while the patients have externalized leads allows for the opportunity to carry out thalamic recordings and concurrent EEG recordings, providing unique
. Figure 166‐1 Schaltenbrand and Wahren atlas section 6.5 mm in the sagittal plane, showing a typical electrode trajectory through the anterior thalamic nucleus. Legend: AC: anterior commissure; DM: dorsomedian nucleus; Fx: fornix; H1, H2: fields of Forel; Hpth: hypothalamus; Icoll: inferior colliculus; Ll: lemniscus lateralis; MM: mammillary bodies; NC: nucleus cucularis; ON: optic nerve; PC: posterior commissure; Pulv: pulvinar; RN: red nucleus; Scoll: superior colliculus; SN: substantia nigra; STN: subthalamic nucleus; IIIVent: third ventricle. [28]. Reproduced with permission, Elsevier
2797
2798
166
Anterior nucleus DBS in epilepsy
possibilities for investigation of the EEG changes associated with seizures. During the patient’s stay in the EMU, combined scalp EEG and thalamic depth electrode recordings can be carried out using the implanted DBS electrodes. A number of seizures can be recorded during this period. In some instances a generalized alteration in scalp EEG background activity was seen, associated with the appearance of low amplitude muscle artifact, marking the earliest stages of a clinical seizure, preceding the first rhythmic ictal activity visible on the scalp EEG. Prior to both clinical onset and these earliest changes seen on scalp EEG, rhythmic ictal epileptiform activity may appear in the AN DBS electrode contacts, bilaterally in the case of generalized onset seizures and initially unilaterally in the case of partial onset seizures [34]. The influence of thalamic DBS on cortical activity was assessed with EEG. Trials of bipolar low frequency stimulation were carried out (2–10 Hz, 0.5–10 V, 330–450 ms pulse duration) using different contact combinations with concurrent scalp EEG recordings. Using these parameters, a recruiting rhythm recorded maximally over the ipsilateral frontal lobe region could be elicited in a majority of patients (> Figure 166‐2). This can be combined with source localization studies looking at the cerebral responses evoked by low frequency thalamic DBS. Examining the
initial cohort of AN patients, EEG evidence of time-locked cortical responses with latencies between 20 and 320 ms. were observed. This was seen primarily in ipsilateral cingulate gyrus, insular cortex and lateral neocortical mesial temporal structures, findings later confirmed in a patient with intracranial EEG recordings [35]. Interestingly, rhythmic cortical 5Hz EEG synchronization occurred in 75% of patients with clinical benefit with AN stimulation, whereas neither of the two patients without EEG synchronization had a significant reduction in seizure frequency [36]. Changes in motor excitability have been observed when these patients were assessed with transcranial magnetic stimulation (TMS). It was observed that these patients have increased short-interval intracortical inhibition with continuous deep brain stimulation, suggesting that DBS might drive cortical inhibitory circuits [37].
Effect on Seizure Frequency In the initial cohort of five patients reported, a decrease in seizure frequency was observed in the first month after implantation of the DBS electrodes, an improvement that continued after the stimulators were turned ON. In general, any of a number of alterations in stimulation parameters was associated a significant difference in the
. Figure 166‐2 Recruiting rhythm on EEG. Segments of left frontal leads (Fp1, F3, C3) are shown. Long downward arrow point to the artefact observed with the start of left thalamic stimulation. After a short lag phase, a rhythmic waxing and waning patters is visualized. Small arrows point to two examples of stimulus artefacts, which can be observed throughout the trace after the commencement of stimulation. Stimulation parameters: 10 Hz, 10 V, 330 ms pulse duration; electrode contacts, 1–2+. Average referential montage. Bar, 1 s. [24]. (Reproduced with permission, Blackwell.)
Anterior nucleus DBS in epilepsy
postimplantation seizure frequency. This benefit was sustained over 5 years of follow-up, with some of the patients continuing with a 90% seizure reduction compared to preimplantation baseline at last follow-up. Some of these patients have been described by caregivers to be more alert, cooperative and participatory in daily activities. A recent study by Lim et al. [38] also noted a strong implantation effect in a nonblinded trial of four patients.
Summary There is a strong neuroanatomical basis for the role of the thalamus in epilepsy. Clinical studies appear to suggest that bilateral implantation of the AN can, in selected patients, result in longterm benefit from intractable seizures. Understanding the clinical features of patients who have shown long-term benefit may potentially be helpful in determining the efficacy of this procedure. Postoperative concurrent thalamic and scalp EEG recordings appear to suggest that presence of a recruiting rhythm, elicited with low frequency stimulation, correlates with patients who show benefit from this procedure. Apart from eliciting a recruiting rhythm, there do not appear to be other electrophysiological correlates that allow for intraoperative identification of the AN, or boundaries that distinguish it from surrounding nuclei. A more conclusive assessment of the clinical effect of AN DBS will hopefully be available after the completion of the multicenter randomized stimulation trial.
References 1. Hassler R, Riechert T. Clinical and anatomical findings in stereotactic pain operations on the thalamus. Arch Psychiatr Nervenkr Z Gesamte Neurol Psychiatr. 1959;200:93-122. 2. Hirai T, Jones EG. A new parcellation of the human thalamus on the basis of histochemical staining. Brain Res Brain Res Rev. 1989;14:1-34.
166
3. Bertram EH, Zhang DX, Mangan P, Fountain N, Rempe D. Functional anatomy of limbic epilepsy: A proposal for central synchronization of a diffusely hyperexcitable network. Epilepsy Res. 1998;32:194-205. 4. Bertram EH, Mangan PS, Zhang D, Scott CA, Williamson JM. The midline thalamus: Alterations and a potential role in limbic epilepsy. Epilepsia. 2001;42:967-978. 5. Irle E, Markowitsch HJ. Single and combined lesions of the cats thalamic mediodorsal nucleus and the mamillary bodies lead to severe deficits in the acquisition of an alternation task. Behav Brain Res. 1982;6:147-165. 6. Irle E, Markowitsch HJ. Widespread cortical projections of the hippocampal formation in the cat. Neuroscience. 1982;7:2637-2647. 7. Irle E, Markowitsch HJ. Connections of the hippocampal formation, mamillary bodies, anterior thalamus and cingulate cortex. A retrograde study using horseradish peroxidase in the cat. Exp Brain Res. 1982;47:79-94. 8. Seki M, Zyo K. Anterior thalamic afferents from the mamillary body and the limbic cortex in the rat. J Comp Neurol. 1984;229:242-256. 9. Fuentealba P, Crochet S, Timofeev I, Bazhenov M, Sejnowski TJ, Steriade M. Experimental evidence and modeling studies support a synchronizing role for electrical coupling in the cat thalamic reticular neurons in vivo. Eur J Neurosci. 2004;20:111-119. 10. Fuentealba P, Steriade M. The reticular nucleus revisited: intrinsic and network properties of a thalamic pacemaker. Prog Neurobiol. 2005;75:125-141. 11. Steriade M. The thalamus. Amsterdam: Elsevier; 1997. 12. Pare D, Steriade M, Deschenes M, Oakson G. Physiological characteristics of anterior thalamic nuclei, a group devoid of inputs from reticular thalamic nucleus. J Neurophysiol. 1987;57:1669-1685. 13. Gonzalo-Ruiz A, Lieberman AR. GABAergic projections from the thalamic reticular nucleus to the anteroventral and anterodorsal thalamic nuclei of the rat. J Chem Neuroanat. 1995;9:165-174. 14. Gale K. Mechanisms of seizure control mediated by gamma-aminobutyric acid: role of the substantia nigra. Fed Proc. 1985;44:2414-2424. 15. Gale K. Role of the substantia nigra in GABA-mediated anticonvulsant actions. Adv Neurol. 1986;44:343-364. 16. Gale K. Progression and generalization of seizure discharge: Anatomical and neurochemical substrates. Epilepsia. 1988;29 Suppl 2:S15-S34. 17. Mirski MA, McKeon AC, Ferrendelli JA. Anterior thalamus and substantia nigra: two distinct structures mediating experimental generalized seizures. Brain Res. 1986;397:377-380. 18. Mirski MA, Ferrendelli JA. Anterior thalamic mediation of generalized pentylenetetrazol seizures. Brain Res. 1986;399:212-223. 19. Mirski MA, Ferrendelli JA. Selective metabolic activation of the mammillary bodies and their connections during
2799
2800
166 20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
Anterior nucleus DBS in epilepsy
ethosuximide-induced suppression of pentylenetetrazol seizures. Epilepsia. 1986;27:194-203. Mirski MA, Ferrendelli JA. Interruption of the mammillothalamic tract prevents seizures in guinea pigs. Science. 1984;226:72-74. Hamani C, Ewerton FI, Bonilha SM, Ballester G, Mello LE, Lozano AM. Bilateral anterior thalamic nucleus lesions and high-frequency stimulation are protective against pilocarpine-induced seizures and status epilepticus. Neurosurgery. 2004;54:191-195. Mirski MA, Rossell LA, Terry JB, Fisher RS. Anticonvulsant effect of anterior thalamic high frequency electrical stimulation in the rat. Epilepsy Res. 1997;28:89-100. Hamani C, Hodaie M, Chiang J, del Campo M, Andrade DM, Sherman D, Mirski M, Mello LE, Lozano AM. Deep brain stimulation of the anterior nucleus of the thalamus: effects of electrical stimulation on pilocarpine-induced seizures and status epilepticus. Epilepsy Res. 2008;78: 117-123. Hodaie M, Wennberg RA, Dostrovsky JO, Lozano AM. Chronic anterior thalamus stimulation for intractable epilepsy. Epilepsia. 2002;43:603-608. Andrade DM, Zumsteg D, Hamani C, Hodaie M, Sarkissian S, Lozano AM, Wennberg RA. Long-term follow-up of patients with thalamic deep brain stimulation for epilepsy. Neurology. 2006;66:1571-1573. Kerrigan JF, Litt B, Fisher RS, Cranstoun S, French JA, Blum DE, Dichter M, Shetter A, Baltuch G, Jaggi J, Krone S, Brodie M, Rise M, Graves N. Electrical stimulation of the anterior nucleus of the thalamus for the treatment of intractable epilepsy. Epilepsia. 2004;45:346-354. Graves NM, Lozano AM, Wennberg RA, Osorio I, Wilkinson S, Baltuch G, French JA, Kerrigan JF, Shetter A, Fisher RS. Brain stimulation for epilepsy: pilot patient results and implementation of a controlled clinical trial. Epilepsia. 2004;45 Suppl 7:148. Hodaie M, Cordella R, Lozano AM, Wennberg R, Dostrovsky JO. Bursting activity of neurons in the human anterior thalamic nucleus. Brain Res. 2006;1115:1-8 (abstract). Radhakrishnan V, Tsoukatos J, Davis KD, Tasker RR, Lozano AM, Dostrovsky JO. A comparison of the burst
30.
31.
32.
33.
34.
35.
36.
37.
38.
activity of lateral thalamic neurons in chronic pain and non-pain patients. Pain. 1999;80:567-575 (abstract) . Lenz FA, Kwan HC, Dostrovsky JO, Tasker RR. Characteristics of the bursting pattern of action potentials that occurs in the thalamus of patients with central pain. Brain Res. 1989;496:357-360 (abstract). Lenz FA, Kwan HC, Martin R, Tasker R, Richardson RT, Dostrovsky JO. Characteristics of somatotopic organization and spontaneous neuronal activity in the region of the thalamic principal sensory nucleus in patients with spinal cord transection. J Neurophysiol. 1994;72:1570-1587 (abstract). Jeanmonod D, Magnin M, Morel A. Thalamus and neurogenic pain: physiological, anatomical and clinical data. Neuroreport. 1993;4:475-478 (abstract). Buzsaki G, Smith A, Berger S, Fisher LJ, Gage FH. Petit mal epilepsy and parkinsonian tremor: hypothesis of a common pacemaker. Neuroscience. 1990;36:1-14 (abstract). Wennberg R, Pohlmann-Eden B, Chen R, Lozano A. Combined scalp-thalamic EEG recording in sleep and epilepsy. Clin Neurophysiol. 2002;113:1867-1869 (abstract). Zumsteg D, Lozano AM, Wennberg RA. Mesial temporal inhibition in a patient with deep brain stimulation of the anterior thalamus for epilepsy. Epilepsia. 2006;47:1958-1962 (abstract). Zumsteg D, Lozano AM, Wieser HG, Wennberg RA. Cortical activation with deep brain stimulation of the anterior thalamus for epilepsy. Clin Neurophysiol. 2006;117:192-207 (abstract). Molnar GF, Sailer A, Gunraj CA, Cunic DI, Wennberg RA, Lozano AM, Chen R. Changes in motor cortex excitability with stimulation of anterior thalamus in epilepsy. Neurology. 2006;66:566-571 (abstract). Lim SN, Lee ST, Tsai YT, Chen IA, Tu PH, Chen JL, Chang HW, Su YC, Wu T. Electrical stimulation of the anterior nucleus of the thalamus for intractable epilepsy: A long-term follow-up study. Epilepsia. 2007;48:342-347 (abstract).
165 Centromedian Thalamic Stimulation for Epilepsy F. Velasco . A. L. Velasco . M. Velasco . F. Jime´nez . J. D. Carrillo-Ruiz . G. Castro
Introduction Sudden onset of generalized seizures that includes loss of consciousness, involvement of global motor activity, and diffuse electroencephalographic (EEG) discharges suggests that convulsive activity is propagated through structures with widespread anatomical and physiological connections. The participation of a neural system with these characteristics in the genesis and propagation of epileptic attacks was proposed some time ago on the basis of clinical observations [1]. Midline and intralaminar thalamic nuclei corresponding to the so-called nonspecific thalamic system [2] meet anatomical and physiological criteria for consideration as an essential part of that neural network [3–9]. While high-frequency stimulation of this system induces cortical EEG desynchronization [10,11], low-frequency stimulation (6.8 Hz) induces cortical synchronization in the form of recruitment of monophasic, long latency, waxing, and waning potentials (recruiting responses, RR) [12–16]. Bilateral stimulation at 3 Hz in these thalamic nuclei or their fiber connections reproduce the clinical and EEG events of a typical absence in cats [17,18] and humans [19,20]. Other experiments have demonstrated the participation of brainstem structures, anatomically linked with the nonspecific thalamic system, in the onset of epileptic seizures in various epilepsy models [21–23]. Although the controversy on the cortical versus subcortical origin of epileptic attacks remains unsolved [11,21–30], there is agreement in that thalamocortical interactions are essential in the development of a variety of seizure types #
Springer-Verlag Berlin/Heidelberg 2009
and the propagation of the majority of these [15,24,25,27–30]. During the past 20 years, we have explored the effects on seizure control as it interferes with thalamocortical interactions using deep-brain electrical stimulation (DBS) of the centromedian thalamic nucleus (CM), which forms part of the nonspecific thalamic system. The CM was chosen as the target, considering its relatively large size (about 1 cm in diameter) and close relationship with the conventional landmarks employed for thalamic stereotactic surgery, i.e., the anterior commissure-posterior commissure (AC-PC) line and the vertical line that touches the anterior border of the posterior commissure (VPC) (> Figure 165-1). To date, we have worked on the following: standardizing surgical procedures and stimulation programs [32]; establishing the efficacy and safety of the procedure [33]; confirming correct electrode placement within the nonspecific thalamic system by evoking recruiting responses and EEG desynchronization as described in different animal species [16]; studying the effects of acute, subacute, and chronic electrical stimulation of centromedian nucleus (ESCM) on different seizure types, paroxysmal EEG discharges, and background activities [34,35]; identifying best surgical candidates, and determining predictors in case selection, electrode position, and confirmation, and stimulation parameters associated with favorable outcome of treated patients [36]. Finally, we have worked on describing the effects of ESCM on neuropsychological performance and the quality of life (QOL) patients attain during treatment [31]. Although emphasis has been placed on therapeutic outcome and complications derived from
2778
165
Centromedian thalamic stimulation for epilepsy
. Figure 165-1 Sagittal section of the thalamus, 10 mm from midline taken from the Schaltenbrand–Wahren anatomical atlas. Red lines indicate anterior commissure-posterior commissure (AC-PC) line and vertical line to posterior commissure (VPC). Centromedian thalamic nucleus (CM) contour is indicated by dashed line and parvocellular subdivision by the continuous line. Stereotactic coordinates provided in the text attempt to place at least two of the four electrode contacts within the parvocellular nucleus. Note that the target is only 1/5 of the total CM area and is located in its more anterior and basal position [31]
ESCM, in performing studies we have had the unique opportunity of exploring, through multiple contact electrodes, the epileptic activity occurring in CM and adjacent structures in hundreds of spontaneous seizures of several types and those that occur under different wakefulness and sleep conditions [19,30,37–41]. Epileptic activity in thalamic and upper mesencephalic areas could also be related in time and wave form with simultaneously recorded scalp EEG activities. In some cases, the relationship of CM paroxysmal activity with that of other subcortical structures could be determined through electrode implantation for localization of epileptic foci before deciding whether the case could be better treated by ESCM [39,40]. These experiences served to analyze and test different hypotheses on epileptogenesis advanced on experimental models of epilepsy and extrapolated to spontaneous seizures in humans [19,42]. They also confirmed the majority of experimental electrophysiological observations on the nonspecific thalamic system [16].
On the other hand, experience with ESCM has provided important information that served to design protocols for clinical trials utilizing electrical stimulation (ES) for epilepsy, and in particular double-blind protocols [36,43,45], parameters and modes of stimulation [31,36], confirmation of target localization [16,31], and plasticity of stimulated tissue [31,43].
Case Selection Patients selected for ESCM were not candidates for ablative procedures of the epileptic foci because they had evidence of bilateral or multifocal seizure onset, unilateral focus overlapping eloquent areas, or no evidence of focal onset of their epileptic attacks. All patients had a long history of seizures (from 4 to 33 years), with numerous to countless seizures that incapacitated them completely. Seizures were out of control despite adequate doses of the specific anticonvulsants corroborated by therapeutic blood levels. All treated cases had more than one seizure type; however, these have been grouped according the most prominent seizure and EEG pattern as follows: Group 1: Focal motor seizures (epilepsia partialis continua or EPC), with frequent generalization in the form of adversive tonic seizures and generalized tonic-clonic convulsions (GTC). They had focal interictal spikes in the central and parietal area, propagated to other ipsilateral areas, and occasionally generalized. Magnetic resonance imaging (MRI) revealed the presence of cortical dysplasia, focal encephalitis, or were normal [46,47]. Group 2: Complex partial seizures (CxP) with frequent generalization as GTC, with surface and/or intracranial bilateral-independent EEG ictal and interictal discharges, bilateral hippocampal sclerosis, or atrophy in MRI associated with memory deficits in the neuropsychological evaluation [36].
Centromedian thalamic stimulation for epilepsy
Group 3: Tonic seizures with fencing posture, frequently associated with propulsive and retropulsive atypical absences (AA) and GTC. Intracranial studies demonstrated bilateral-independent focal ictal or interictal EEG discharges in mesial frontal regions and no evidence of lesions or dysplasia in epileptic foci in MRI [49]. Group 4: AA and GTC associated with 2.0–2.5 cycles per sec (cps) spike-wave complexes (SKW) and in some cases, with bilateral frontal or temporal interictal spikes. All patients were mentally challenged, some of these since infancy, and others with progressive mental deterioration after onset of seizure history. They were considered as Lennox–Gastaut syndromes. MRI showed signs of regional or hemispheric atrophy, cortical dysplasia, and subependinal calcifications associated with skin stigmata of tuberous sclerosis (symptomatic Lennox–Gastaut syndrome); however, MRI was normal in 56% of cases (idiopathic Lennox–Gastaut syndrome) [31,36]. Patients were evaluated at the Epilepsy Clinic at least 6 months prior to surgery and all were referred from other epilepsy clinics or institutions where they had been treated for >3 years before being considered as refractory to medical treatment. Patients were placed in anticonvulsive regimes, which proved more efficient for seizure control when anticonvulsant blood levels fell within therapeutic range. These regimes were maintained during a 3–6-month preoperative observation (baseline) and throughout at least 1 year with ESCM. Patients and/or relatives were instructed to maintain a meticulous record of the frequency of each seizure type separately. They easily learned to differentiate among GTC, tonic or clonic (Type A), AA (Type B), and focally initiated (Type C) seizures.
Surgical Techniques Patients (many of these children) are operated on under general anesthesia. Electrodes are implanted
165
by means of a frontal parasagittal approach, and burr holes are centered at 13 mm at each side from midline and immediately behind the coronal suture. This allows to avoid the venous sinus and places the electrodes in a trajectory parallel to midline at a distance of 10–12 mm on each side and at a sagittal angle of 45–60 from the AC-PC line (> Figure 165-1). Most implantations have been guided by ventriculograms and electrode tips were aimed at the intersection of AC-PC and VPC lines. By using ventriculography, intraventricular symptom-associated blending has occurred very rarely [36], and X-ray films or fluoroscopy nicely confirm the correct position of the contacts after electrodes have been fixed to the skull. More recently, we have employed MRIcomputed tomography (CT) image fusion to target CM to avoid venous sinuses and brain and ventricular wall blood vessels in the trajectories of electrodes, particularly those that traverse the ventricular wall. MRI-CT electrodes are implanted with indirect targeting, because the conventional 1.5 T MRI does not permit visualizing the internal medullary laminae of the thalamus that surrounds CM as it unfolds in its posterior end. Coordinates are 11 mm lateral to third ventricle midline, measured from the anterior border of the posterior commissure and contact 0 of tetrapolar electrodes (numbered from 0 to 3) 2 mm above AC-PC level at an angle of 45–60 from AC-PC line. Sterile scalp electrodes are placed after preparing the skin in a conventional 10–20 system array to be used for electrophysiological confirmation of DBS electrode position. Sterile draping aids in maintaining scalp electrodes in place. Lowfrequency (6–8 Hz) bipolar stimulation is delivered through each pair of contacts (0–1, 1–2, 2–3) of DBS using a external electrical stimulator set at a 1.0-ms duration, increasing intensities from 0.2 to 1.0 mA in the search for scalp recruiting responses. Typical RRs are monophasic-negative, long-latency, waxing-waning potentials that may be better analyzed morphologically through
2779
2780
165
Centromedian thalamic stimulation for epilepsy
oscilloscopic recordings. These must be distinguished from augmenting and primary evoked potentials (> Figure 165-2a). Unilateral stimulation typically induced bilateral responses in parasagittal leads of scalp EEG, more prominent in frontopolar regions and ipsilateral to the stimulated site (> Figure 165-2b). On the other hand, high-frequency stimulation (60 Hz, 1.0 ms, 0.3–0.8 mA) of CM induces cortical desynchronization and a shift in EEG baseline level (DC-shift), with the same RR scalp distribution [16,36]. These electrocortical potentials are better recorded under awake-relaxed conditions; thus, we formerly performed postoperative electrophysiological confirmation during the first of a twostage operation with DBS temporally externalized.
Recently, we have conducted electrode position confirmation transoperatively; nonetheless, to elicit electrocortical responses the anesthesiologist is requested to situate the patient on a superficial anesthetic plane during the electrophysiological confirmation period and to utilize muscle relaxants and analgesics to maintain the patient relaxed and immobilized. The aforementioned CM stereotactic coordinates and electrocortical responses attempt to place at least two DBS electrode contacts within the nucleus’s parvocellular subdivision of the nucleus, that is, the most lateral, anterior, and ventral part. Outside this subdivision, ESCM is much less effective in controlling seizures [31,36,42] (> Figure 165-3). Thereafter, electrodes are connected to internalized pulse generators
. Figure 165-2 (a) Electrocortical potentials induced by centromedian thalamic nucleus (CM) stimulation and analyzed in oscillographic recordings: recruiting responses (RR) (top) are monophasic, 32–36-ms latency-negative potentials with maximal amplitude in frontal (F4) leads. They must be differentiated from augmenting responses (midline), which are positive-negative potentials elicited by stimulation of magnocellular CM subnucleus and better recorded in central (C4) leads. Also from primary (bottom), evoked potentials elicited by stimulation of specific sensory ventroposteromedial-ventroposterolateral (VPM-VPL) thalamic nucleus or posterior subdivision of CM and recorded in parietal (P4) leads. (b) Distribution of recruiting responses in scalp electroencephalogram (EEG): these are more prominent in ipsilateral frontopolar regions. RR on the left were elicited through externalized electrodes, and RR on the right were elicited through internalized pulse generators [36]
Centromedian thalamic stimulation for epilepsy
165
. Figure 165-3 Plotting of pairs of electrode contacts used for chronic bipolar electrical stimulation on sagittal (SI 9,0) and frontal (Fp 9,0) sections of the Schaltenbrand–Wahren anatomical atlas. (a) and (a0 ): plotting of cases with >80% decrease of seizures (good outcome). (b) and (b0 ): plotting of cases with <80% improvement (poor outcome). Note that while electrodes are grouped in the anterolateral part of centromedian thalamic nucleus (CM) (parvocellular area) in cases with good outcome, electrodes are dispersed toward more posterior, superior, and medial regions of CM (magnocellular, internal medullar laminae, and parafascicularis) in poor-outcome cases. AC, anterior commissure; PC, posterior commissure; VPC, vertical line posterior commissure; ML, midline; Ce, centromedian nucleus; mc, magnocellular; pc, parvocellular; Mfip, fibrosus posterior; Lam, lamina medullaris; Lm, medial lemniscus; M, territorium medialis thalami; Pf, parafascicularis; Pu, pulvinar; Raprl, prelemniscal radiations; Vci, ventralis caudalis internus; Vce, ventralis caudalis externus; Vcpc, ventralis caudalis parvocellularis; Vimi, ventralis intermedius; Vop, ventralis oralis posterior
(IPG) placed in subcutaneous subclavicular pockets through extension cables. Because the most frequent complications by far of the procedure comprise neurostimulator hardware-induced skin erosions, the following precautions are recommended: (1) Scalp
incision must be a skin flap at least 2 cm away from burr-hole edges and skull-fixation devices; (2) Low-profile extension cables should be placed and immobilized by a stitch under muscular fascia behind mastoid bone, and (3) Place the IPG away from subclavicular incision and
2781
2782
165
Centromedian thalamic stimulation for epilepsy
fixed to pectoralis muscle with stitches. In children and slender patients, it is better to place the IPG below muscular fascia at abdominal wall and ensure that the extension cable is loosened in gentle loops below IPG, as it will be required to straighten as the child grows.
A charge density ca 2–3 mC/cm2/phase corresponds to 50–80% of that necessary to induce RR or DC-shift [31].
thereafter. In view of the mental- and cognitivefunction deterioration of treated patients, neuropsychological evaluation consisted mainly of evaluating abilities by means of scales and QOL questionnaires standardized and validated for our population [31,35]. During the same EEG sessions, stimulation parameters delivered by IPGs and electrode impedance were revised. Also, electrocortical responses to test reactivity to ESCM of stimulated tissue were elicited by setting IPGs parameters at 6–8 Hz, 450 ms, 6–10 V for RR and 130 Hz, 450 ms, 7–10 V for DC-shift (> Figure 165-2). These precautions confirmed correct stimulation of CM, not only from the neurostimulation hardware standpoint, but also from the physiological response of stimulated tissue [36]. Some patients underwent a double-blind randomized trial of ESCM in which one half of these would have stimulation turned OFF between months 6 and 9 and the other one half, between 9 and 12 months. Because ESCM induced no objective and subjective sensations, patient and examiner remained unaware of the maneuver and the IPG stimulation-parameter code and reading was maintained in secret by a third party, the maneuver was considered valid. Significance in seizure reduction for total seizure number and for each seizure type was evaluated through Student t-test and for neuropsychological and QOL scales by Wilcoxon test.
Follow-Up
Results
Patients were interviewed every 1–3 months for the first year and every 6 months thereafter. Different seizure-type calendars were updated and stored in a database. After 1 year of ESCM, anticonvulsants could be adjusted according to response to ESCM and tolerance to medication. EEG and neuropsychological evaluation were repeated every 3 months for 1 year and every year
Effect of ESCM on Seizure Control
Chronic Stimulation A cycling mode of stimulation was initially adopted to avoid electrical current overcharge and damage to stimulated tissue [32]. With time, we have observed that cycling mode is as effective as continuous stimulation for seizure control. Therefore, we continue to use cycling mode that saves IPG battery to the point that it may last from 3 to 9 years. On the other hand, while adjusting the stimulation parameter to treat movement disorders is a relatively easy task, ES adjustment for treating epilepsy is more complicated. We have recently adjusted ES for seizure control not to exceed the 4 mC/cm2/phase, according to the following formula [44]: Charge density ðmC=cm2 =phaseÞ ¼ amplitude ðmAÞ pulse width ðmsÞ cathodal electrode area ðcm2 Þ
Group1: EPC included five children aged 3–7 years who were tested subacutely (for 3 months) with electrodes externalized through a connector cable away from skull burr hole. During the first month, a significant decrease (p < 0.01) in GTC and tonic adversive seizures was observed. By the end of
Centromedian thalamic stimulation for epilepsy
3 months, four patients were seizure-free and their electrodes were explanted. In the remaining patient, the stimulation system was internalized and seizures disappeared in 2 months and remained so for 8 months, when the stimulation system required explantation due to multiple skin erosions. Explantation was followed by progressive reappearance of seizures and the patient was finally treated with subpial transection. Group 2: Sixteen patients with CxP seizures who were not candidates for surgical ablation because they had evidence of bilateral independent foci were treated by ESCM. They had no significant decrease in CxP seizures during the first year, but both GTC and tonic seizures decreased or disappeared during that time (p < 0.001). However, an evaluation at 3 years showed significant decrease of even CxP seizures (p < 0.01) that persisted for periods up to 16 years [36,42]. Group 3: Six patients with tonic seizures and AA and evidence of bilateral interhemispheric foci had significant reduction of AA (p < 0.001). However, no patient in this group became seizure-free. We consider the results of ESCM for these three groups of patients satisfactory and certainly competitive with those reported in other trials involving electrical stimulation of different structures for seizure control [49–51]. Nevertheless, ESCM in children <7 years of age with EPC has been abandoned in view of frequent skin erosions along the neurostimulator-hardware trajectory and in older children and adults due to the promising results with ES of epileptic foci in eloquent areas. This is also true for patients in Groups 2 and 3; Group 2 cases are being studied and treated with stimulation of hippocampus [45], whereas persons in Group 3 are being administered treatment with cortical interhemispheric stimulation (See chapter K-13). During the past 7 years, we have used ESCM solely in cases with generalized seizures of Lennox–Gastaut syndrome and have treated 25 patients with long-term (>3 years) follow-up.
165
Lennox–Gastaut syndrome is one of the most severe forms of childhood epilepsy, characterized by drug-resistant GTC and AA, EEG generalized 2.0–2.5 per sec spike-wave complexes, and mental retardation. Over 80% of patients continue to experience seizures throughout adulthood [52]. Selected patients have been >8 years of age and weighed 25 kg. They experienced a severe to extreme epileptic condition and were completely disabled. Some of these had retardation in neuropsychological condition noticed in early childhood, while others became mentally retarded after seizure onset. ESCM in cases of individuals with Lennox– Gastaut syndrome resulted in global improvement of seizures of 83%, with 15.4% of patients becoming seizure-free. Most important was improvement in their QOL, particularly in cases of persons who experienced mental deterioration after seizure onset; these regained abilities and some are living a normal life, seizure-free and OFF medication [31]. Other surgical procedures have been proposed to treat Lennox–Gastaut syndrome, such as corpus callosum section (CCS) and vagal nerve stimulation (VNS). The degree of significant improvement in seizure occurrence has been reported for 60.9% [53], and for 72% [54] of patients in the most optimistic outcomes for CCS. The latter is particularly efficient in controlling drop attacks and some varieties of atypical absences, but not to date to a great extent for GTC [53,55]. VNS in Lennox–Gastaut syndrome has induced a global seizure reduction of 27–64% with no patient rendered seizure-free [56]. To our knowledge, there are no reports of DBS in other areas for treating this specific syndrome.
Predictors in Seizure Control with ESCM Although Lennox–Gastaut syndrome GTC and AA have been the best indications for ESCM,
2783
2784
165
Centromedian thalamic stimulation for epilepsy
certain cases responded better than others (global improvement range, 30–100%). Conversely, ESCM is less effective in CxP seizures: we continue to achieve seizure reduction from 52 to 98.6% in different patients and secondary GTC disappear in the majority of patients. Therefore, we have searched for other predictors related with favorable outcome, and have observed that stereotactic position corresponding to parvocellular area of CM and typical RR elicited through contacts used for therapeutic stimulation correlate with better outcome (p < 0.001). Other variables such as age, sex, length of seizure history, seizure frequency, stimulation parameters (amplitude and frequency, 60–130 Hz), or continuous versus cycling mode of stimulation did not significantly modify outcome [31,36].
ON/OFF Stimulation Condition Because epilepsy is a condition with frequent spontaneous fluctuations in seizure occurrence that are related with a number of factors (stress, drug intake, menstrual periods, and concomitant diseases, among others), evaluation of any therapy by means of a double-blind protocol is desirable. We have evaluated a group of patients in a doubleblind protocol designed to interrupt stimulation for 3 months beginning at day 60 (range, 60–90 days) in one half of these, and from 90 to 120 days in the remaining one half. Patients were assigned to each group by lottery number. A cross-sectional analysis of total seizure number and each seizure type separately was performed among baseline, the 3-month period preceding the double-blind maneuver, and the 3-month OFF stimulation period. Significant differences were found between baseline and ON and OFF period (p < 0.001) for total seizure number and each seizure type: GTC and AA (p < 0.001) and focal seizures (p < 0.05). Nonetheless, no significant differences were found between the period ON and OFF stimulation (> Figure 165-4a). We conclude that there was a residual effect of ESCM that outlasted the 3-month OFF stimulation period [36].
More recently, we analyzed the spontaneous OFF stimulation periods that occur either when the IPG battery charge depletes or when we have been required to explant neurostimulation systems because of skin erosions. In the former, we have observed that seizures reappear when the battery is depleted, although these do not reach baseline levels. Seizures decrease rapidly after the IPG is replaced and stimulation reinitiated. More interesting is that as the stimulation period is extended, recurrent battery depletions are accompanied by less and less recurrence in seizures and that these are controlled more rapidly after reinitiating stimulation (> Figure 165-4b and c). In cases of explantation, seizures slowly increase for months and eventually may – although not always – reach baseline levels. We have witnessed the case of a child with symptomatic Lennox– Gastaut syndrome who became seizure-free and OFF medication during 5 years with ESCM; this was accompanied by normalization of EEG recordings and the return of the child’s mental condition to normal after being completely deteriorated. This patient was explanted after 5 years due to skin erosions, and after 2 years only occasional 2.0–2.5 EEG spike-wave complexes without clinical manifestations have been found. This adolescent is living a normal life at present [31]. All these observations suggest that long-term ESCM induces plastic changes in the stimulated tissue that maintains a residual antiepileptic effect. Thus, new studies evaluating therapeutic trials of electrical stimulation in epilepsy place the doubleblind maneuver at the beginning of stimulation period, and one half the patients initiate stimulation immediately, while stimulation is delayed for 1–3 months in the other one half [43–45].
The Role of CM in the Genesis and Propagation of Epileptic Activity Simultaneous recording from scalp, electrode in CM, and other intracranial electrodes used for diagnostic purposes prior to ESCM was decided as treatment to provide information on the role of CM
Centromedian thalamic stimulation for epilepsy
165
. Figure 165-4 (a) Cross-section of 3-month follow-up epochs indicating baseline (BL), 3 months ON stimulation, and 3 months OFF stimulation (3–6 to 6–0 months after the onset of electrical stimulation of centromedian nucleus [ESCM]). Note that 3-month periods OFF stimulation are not accompanied by seizure return to basal level (BL). (b) Long-term follow-up in two cases in patients in whom implantable pulse generator (IPG) batteries depleted twice and were replaced. Note that initial improvement reached a plateau after 4–5 months, while subsequent reinitiation of ESCM arrows was followed by an immediate amelioration. (c) One case was explanted after 8 months of ESCM and followed up for several months. Seizure increased OFF stimulation but for months did not reach BL levels [31,33,36]
2785
2786
165
Centromedian thalamic stimulation for epilepsy
. Figure 165-5 EEG recordings from right and left centromedian thalamic nucleus through implanted deep-brain electrical stimulation and simultaneously from intracranial or scalp electrodes. (a) Spike-wave complexes recorded from scalp EEG and thalamic electrodes in a case with Lennox–Gastaut syndrome. Paroxysmal discharges appear to occur simultaneously. (b) Spike-wave complexes recorded during a typical absence. Paroxysmal discharges initiated in right CM (RCM) many seconds before left CM (LCM) and scalp [19]
in different seizure patterns. Intracranial electrodes were placed mainly along hippocampus amygdala axis, underneath temporal or frontal lobes, or at each side of interhemispheric frontal areas.
1.
EPC focal jerks were accompanied by cortical discharges that did not propagate to CM unless focal motor seizures involved the neck in a contraversive movement or at
Centromedian thalamic stimulation for epilepsy
2.
3.
GTC onset. Under these circumstances, repetitive bursts of polyspikes appeared in CM [41]. Cortical and CM spikes increased in amplitude, while muscular jerks decreased during slow-wave sleep. Both EEG and muscular spikes decreased to disappearance during paradoxical sleep. Cortical and temporal spikes increased during CxP without concomitant discharges in CM. When CxP evolved to GTC, delayed polyspikes and comb-like activity appeared in CM [47]. In cases of supplementary motor cortex foci, tonic seizures were accompanied by nearly simultaneous onset of cortical and CM paroxysmal discharges; also in AA, CM and cortical spike-wave complexes were simultaneous [40,48] (> Figure 165-5a).
4.
5.
165
In AA of the Lennox–Gastaut syndrome, spike and wave complexes appeared to occur simultaneously in CM and cortex. However, when the spike component of the EEG complexes exhibited a single negative spike, peak latency presented 35 ms before in CM and in upper mesencephalon than in cortex [40]. Typical absences occurring in young adults showed unilateral EEG spike-wave complexes anticipating (by 6–13 ms) the same complexes in contralateral CM and scalp [38] (> Figure 165-5b). Bilateral simultaneous high-amplitude 3-cps stimulation in CM induces generalized 3-cps spike-wave complexes accompanied by typical absences and motor arrest, as described many years ago in cats [17] (> Figure 165-6).
. Figure 165-6 Three-cycles per sec (3 cps) spike-wave complexes induced by simultaneous stimulation of right and left CM (RCMLCM) at 3 cps, 2.0 mA, 1.0 ms, while the patient was performing a task (pressing a button in response to a flash). During stimulation, the patient presented a typical absence-and-stop to respond to flashes that lasted precisely for the time of stimulation [16]
2787
. Figure 165-7 Effect of chronic ESCM on EEG generalized spike-wave complexes in a case of Lennox–Gastaut syndrome (left) and on secondary synchronous discharges (right) of a patient with complex partial seizure (CxP) and focal spikes on left temporal leads (arrows). After chronic ESCM, spike-wave complexes and secondary synchronous discharges disappeared, but focal spikes persisted
2788
165 Centromedian thalamic stimulation for epilepsy
Centromedian thalamic stimulation for epilepsy
6.
Myoclonic seizures frequently occur with no concomitant paroxysmal discharge in scalp, cortex, CM, or upper mesencephalon.
These observations correlate well with the effect of ESCM on different seizure types, with the exception of EPC. In fact, ESCM stimulation in CxP seizures interferes with generalization in the form of GTC seizures, but much less so in focal interictal spikes (> Figure 165-7). ESCM decreases adversive tonic and AA secondary to epileptic foci in supplementary motor cortex and AA of Lennox–Gastaut syndrome, while it exerts no effect on myoclonic epilepsy. CM recordings and acute and chronic stimulation provide evidence that the nucleus – and perhaps all nonspecific thalamic nuclei – may be the origin of spike-wave complexes associated with typical absences. They probably play an important role in the genesis of atypical absences associated with Lennox–Gastaut syndrome 2.0– 2.5 spike-wave complexes, most likely in a thalamic-cortical interplay. They participate in the propagation of epileptic activity focally initiated in the cortex, at least in frontal and temporal lobes. In contrast, they do not participate in the genesis or propagation of myoclonic jerks that may originate in brainstem [42].
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
References 17. 1. Penfield W, Jasper H. Epilepsy and the functional anatomy of the human brain. Little, Brown; Boston: p. 566-96. 2. Skinner JE, Lindsley DB. The non-specific mediothalamicfrontocortical systems: its influence on electrocortical activity and behavior. Psychophysiology on frontal lobes. New York: Academic Press; 1973. p. 185-236. 3. Dempsey EW, Morison RS. A study of thalamo cortical relations. Am J Physiol 1942;135:291-2. 4. Jasper HH. Non-specific thalamo cortical relations. In: Field J, et al., editors. Handbook of physiology: neurophysiology (Section I), vol 2. Washington, DC: American Physiological Society; 1960. p. 1307-21. 5. Jasper HH. Mechanisms of propagation. Extracellular studies. In: Jasper JJ, Ward AA, Pope A, editors. Basic
18.
19.
20.
165
mechanisms of the epilepsies. Boston: Little, Brown; 1969. p. 421-38. Monnier M, Kalberer M, Krupp P. Functional antagonism between diffuse reticular and intralaminar recruiting projections in the medial thalamus. Exp Neurol 1960;12:271-89. Mehler RW. Further notes on the center median nucleus of Luys. In: Purpura DP, Yahr MD, editors. The thalamus. New York: Columbia University Press, 1966. p. 109-27. Scheibel ME, Scheibel AB. Structural organization of non-specific nuclei and their projection toward cortex. Brain Res 1967;6:60-94. McLachlan RS, Gloor P, Avoli M. Differential participation of some ‘‘specific’’ and ‘‘non-specific’’ thalamic nuclei in generalized penicillin epilepsy. Brain Res 1984;307:277-8. Moruzzi G, Magoun HW. Brain stem reticular formation and activation of the EEG. Electroencephalogr Clin Neurophysiol 1949;1:455-73. Miller JW, Hall CM, Holland KD, Farandelli JA. Identification of a median thalamic system regulating seizures and arousal. Epilepsia 1989;30:493-500. Dempsey EW, Morison RS. The production of rhythmically recurrent cortical potentials after localized thalamic stimulation. Am J Physiol 1942;135:293-300. Dempsey EW, Morison RS. The mechanism of the thalamo cortical augmentation and repetition. Am J Physiol 1942;158:297-308. Hess WR. The diencephalic sleep center. In: Adrian ED, editor. Brain mechanisms and consciousness. Springfield, IL: Charles C. Thomas; 1954. p. 117-36. Pollen DA, Perot P, Reid KH. Experimental bilateral spike and wave from thalamic stimulation in relation to the level of arousal. Electroencephalogr Clin Neurophysiol 1963;15:459-73. Velasco M, Velasco F, Velasco AL, Brito F, Jime´nez F, Ma´rquez I, Rojas B. Electrocortical and behavioral responses produced by acute electrical stimulation of the human centromedian thalamic nucleus. Electroencephalogr Clin Neurophysiol 1996;102:461-71. Hunter J, Jasper HH. Effects of thalamic stimulation in anesthetized animals. The arrest reaction and petit mal like seizure activation patterns in generalized convulsions. Electroencephalogr Clin Neurophysiol 1949;1:305-24. Skinner JE. Abolition of several forms of cortical synchronization during blockade of the inferior thalamic peduncle. Electroencephalogr Clin Neurophysiol 1971;31:127-209. Velasco F, Velasco M, Ma´rquez I, Velasco G. Role of centromedian thalamic nucleus in the genesis, propagation and arrest of epileptic activity: an electrophysiological study in man. Acta Neurochir 1993;58(Suppl):201-5. Velasco M, Velasco F, Jime´nez F, Carrillo-Ruiz JD, Velasco AL, Salı´n-Pascual R. Electrocortical and behavioral responses elicited by acute electrical stimulation of
2789
2790
165 21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
Centromedian thalamic stimulation for epilepsy
inferior thalamic peduncle and nucleus reticularis thalami in a patient with major depression disorder. Clin Neurophysiol 2006;117:320-7. Miller JW, Farandelli JA. The central medial nucleus: thalamic site of seizure regulation. Brain Res 1990;508:297-300. Velasco F, Velasco M. Mesencephalic structures and tonic-clonic generalized seizures. In: Avoli M, Gloor P, Kustopoulus G, Naquet R, editors. Generalized epilepsy: neurobiological approaches. Boston: Birkhouser; 1990. p. 368-84. Jime´nez F, Velasco F, Carrillo-Ruiz JD, EstradaVillanueva F, Velasco M, Ponce H. Seizures induced by penicillin microinjections in the mesencephalic tegmentum 2000;38:33-44. Gloor P. Generalized epilepsy with spike and wave discharges: a reinterpretation of the electrographic and clinical manifestations. Epilepsia 1979;20:571-89. Gloor P, Quesney LF, Zumstein H. Pathophysiology of generalized penicillin seizures in the cat. The role of cortical and subcortical structures. II. Topical application of penicillin to the cerebral cortex and to subcortical structures. Electroencephalogr Clin Neurophysiol 1977;48:79-94. Harabaugh RE, Wilson DH. Telencephalic theory of generalized epilepsy. Observations in split brain patients. Neurosurgery 1982;10:725-32. Quesney F, Gloor P, Kratzemberg E, Zumstein H. Pathophysiology of generalized penicillin epilepsy in the cat the role of cortical and subcortical structures. I. Systemic application of penicillin. Electroencephalogr Clin Neurophysiol 1977;42:640-55. Avoli M, Gloor P, Kostopoulus G, Gutman J. An analysis of penicillin-induced generalized spike and wave discharges using simultaneous recordings of cortical and thalamic single neurons. J Neurophysiol 1983;50:819-37. Steriade M. Spindling, incremental thalamo-cortical responses and spike-wave like epilepsy. In: Avoli M, Gloor P, Kustopoulus G, Niquet R, editors. Generalized epilepsy: neurobiological approaches. Boston: Birkhouser; 1990. p. 161-80. Velasco M, Velasco F, Alcala´ H. Da´vila G, Dı´az de Leo´n AE. Epileptiform EEG activities in the centromedian thalamic nuclei in children with intractable generalized seizures of the Lennox–Gastaut syndrome. Epilepsia 1991;32:310-21. Velasco AL, Velasco F, Jime´nez F, Velasco M, Castro G, Carrillo-Ruiz JD, Fanghanel G, Boleaga B. Neuromodulation of the centromedian thalamic nuclei in the treatment of generalized seizures and the improvement of the quality of life in patients with Lennox–Gastaut syndrome. Epilepsia 2006;47(Suppl):1203-12. Velasco F, Velasco M, Ogarrio C, Fanghanel G. Electrical stimulation of the centromedian thalamic nucleus in the
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
treatment of convulsive seizures. A preliminary report. Epilepsia 1987;28:421-30. Velasco F, Velasco M, Velasco AL, Jime´nez F, Ma´rquez I, Rise M. Electrical stimulation of the centromedian thalamic nucleus in the control of intractable seizures: longterm studies. Epilepsia 1995;36:63-71. Velasco F, Velasco M, Velasco AL, Jime´nez F. Effect of chronic electrical stimulation of the centro median thalamic nuclei on various intractable seizure patterns. I. Seizure frequency and paroxysmal EEG activities. Epilepsia 1993;34(6):1052-64. Velasco M, Velasco F, Velasco AL, Velasco G, Jime´nez F. Effect of chronic electrical stimulation of the centro median thalamic nuclei on various intractable seizure patterns. II. Psychological scores and background EEG activities. Epilepsia 1993;34 (6):1064-74. Velasco F, Velasco M, Jime´nez F, Velasco AL, Brito F, Rise M, Carrillo-Ruiz JD. Predictors in the treatment of difficult to control seizures by electrical stimulation of the centromedian thalamic nucleus. Neurosurgery 2000;47:295-305. Velasco F, Velasco M, Alcala´ H. Electrical stimulation of the thalamus. In: Kutt H, Resor S, editors. Medical treatment of epilepsy: advances in neurology. New York: Marcel Dekker; 1991. p. 677-80. Velasco AL, Boleaga B, Santos N, Velasco F, Velasco M. Electroencephalographic and magnetic resonance correlation in children with intractable seizure of the Lennox– Gastaut syndrome and Epilepsia partialis continua. Epilepsia 1993;34:262-70. Velasco M, Dı´az de Leo´n AE, Brito F, Velasco AL, Velasco F. Sleep epilepsy interactions in patients with intractable generalized tonic seizures and depth electrodes in the centromedian thalamic nucleus. Arch Med Res 1995;26(Suppl):5117-25. Velasco M, Velasco F, Velasco AL. Temporo-spatial correlations between cortical and subcortical EEG spike-wave complexes of the idiopathic Lennox-Gastaut syndrome. Stereotact Funct Neurosurg 1997;69:216-20. Velasco M, Velasco F, Alcala´ H, Dı´az de Leo´n AE. Wakefulness-sleep modulation of EEG-EMG epileptiform activities: a quantitative study on child with intractable Epilepsia partialis continua. Int J Neurosci 1991;54:325-37. Velasco F, Velasco M, Jime´nez F, Velasco AL, Rojas B, Pe´rez ML. Centromedian nucleus stimulation for epilepsy: clinical, electroencephalographic and behavioral observations. Thalamus Relat Syst 2002;1:387-98. Velasco F, Carrillo-Ruiz JD, Brito F, Velasco M, Velasco AL, Ma´rquez I, Davis R. Double blind randomized controlled pilot study of bilateral cerebellar stimulation for treatment of intractable motor seizures. Epilepsia 2005;46(7):1071-81. Fisher RS. Anterior thalamic nucleus stimulation: issues in study design. In: Lu¨ders HO, editor. Deep brain
Centromedian thalamic stimulation for epilepsy
45.
46.
47.
48. 49.
50.
stimulation and epilepsy. London: Taylor and Francis; 2004. p. 305-19. Velasco AL, Velasco F, Velasco M, Trejo D, Castro G, Carrillo-Ruiz JD. Electrical stimulation of the hippocampal epileptic foci for seizure control: a double blind, long term follow up study. Epilepsia 2007;48:1895-903. Rasmussen T, Olsiewsky J, Lloyd-Smith D. Focal seizures due to chronic localized encephalitis. Neurology 1958;8:435-45. Velasco M, Velasco F, Velasco AL, Luja´n M, Va´zquez del Mercado J. Epileptiform EEG activities of the centromedian thalamic nuclei in patients with intractable partial motor, complex partial and generalized seizures. Epilepsia 1989;30:295-306. So KN. Mesial frontal epilepsy. Epilepsia 1998;39 Suppl 4:49-61. Amar AP, Apuzzo MLJ, Liu CY. Vagus nerve stimulation therapy after failed cranial surgery for intractable epilepsy: results from the vagus nerve stimulation therapy patient outcome registry. Neurosurgery 2004;55:1086-93. Davis R, Emmans SE. Cerebellar stimulation for seizure control: 17 year study. Stereotact Funct Neurosurg 1992;58:200-8.
165
51. Kerrigan JF, Litt B, Fisher RS, Cranstoun S, French JA, Blum DE, Tichter M, Shetter A, Baltuch G, Jaggi J, Krone S, Brodie M, Rise M, Graves N. Electrical stimulation of the anterior nucleus of the thalamus for the treatment of intractable epilepsy. Epilepsia 2004;45:346-54. 52. Kazuichi Y. Evolution of Lennox–Gastaut syndrome: a long term longitudinal study. Epilepsia 1996;37 Suppl 3:48-51. 53. Rossi GF, Colicchio G, Marchese E, Pompucci A. Callosotomy for severe epilepsies with generalized seizures: outcome and prognostic factors. Acta Neurochir (Wien) 1996;138:221-7. 54. Cukier A, Brattain JA, Mariana PP, Camera RB, Seda L, Baldauf CM, Argentoni M, Baise-Zung C, Forster CR, Mello VA. Extended, one stage callosal section for treatment of refractory generalized epilepsy in patients with Lennox–Gastaut and Lennox like syndromes. Epilepsia 2006;47(2):371-4. 55. Sass KJ, Spencer DD, Spencer SS. Corpus callosotomy for epilepsy. II. Neurologic and neuropsychological outcome. Neurology 1998;38:24-8. 56. Karceski S. VNS and LGS: a review of the literature and data for the VNS patient registry. CNS Spectr 2001;6:766-70.
2791
168 Cerebellar Stimulation for Seizure Control R. Davis
Abstract Studies of animal with epilepsy, since 1941, showed probable effectiveness of cerebellar stimulation (CS) in inhibiting seizures. Since 1973, 14 clinics have reported on 135 patients in whom chronic cerebellar stimulation (CCS) has been used to control intractable seizures; 116 were in 11 non-blinded studies, of which 88 (76%) benefited in terms of seizure reduction. Of three double-blind studies involving five, nine and five patients, the initial two series reported not to have seizure reduction with CCS. After critical reviews, however, four patients in each of the initial two studies appear to have benefited. The third double-blind study of five patients with generalized tonic-clonic seizures responded to CCS by the second month; and by 6 month, the mean seizure rate was reduced to 41% (14–75%) of the baseline level. Double-blind studies should not be done unless the stimulation is used for 3 months, also no cross-over studies should be attempted, because of the long carry-over effects following the use of CCS. In a detailed review of nine clinics reporting on 93 patients having a total of 174 seizures of seven types; using CCS, 79 patients (85%) benefited with 30% of their seizures stopped, 50% reduced, and 20% unaffected. Tonic-clonic seizures were the most common, being present in 71 (81%) of the 88 patients, of whom 29 (41%) became seizure-free, 28 (39%) benefited from reductions, and 14 (20%) were not affected. Histological studies of cerebellar cortex under the electrodes showed that the chronic use of charge densities stimulation (0.9–2.5 mC/cm2/phase) was safe. #
Springer-Verlag Berlin/Heidelberg 2009
CCS is non-ablative, removable and could be offered to patients with medically intractable seizures originating from bilateral or extratemporal foci.
Animal Studies The use of chronic cerebellar cortex stimulation (CCS) as a therapy for intractable human seizures has been based on studies of animals with seizures since 1941. Moruzzi [1] and Cooke and Snider [2] showed in acute experiments that seizure activity in various cerebral sites was abruptly terminated or modified by cerebellar cortex stimulation (CS). Iwata and Snider [3] extended the studies to the hippocampus and observed that CS could stop seizure patterns and prolonged after-discharges instituted by electrical stimulation. In the non-seizuring hippocampus, CS could induce slow waves while activation like patterns appeared simultaneously in the cerebrum. These observations were confirmed in unrestrained cats in 1963 by Fanardjian and Donhoffer [4]. In 1962, Dow and colleagues [5], working on a model of chronic epilepsy in awake unanesthetized rats, showed that CS could alter EEG activity and inhibit cobalt-induced frontal lobe seizures. Mutani and coworkers [6] showed that CS could reduce temporary cobalt-induced spiking in cats. In 1980, Laxer and coauthors [7] reviewed 22 animal epilepsy studies, which had used CS with a wide range of stimulation parameters, and found conflicting results. They drew two major conclusions: (1) Stimulation of the vermis and intermediate cerebellar cortex is more
2824
168
Cerebellar stimulation for seizure control
efficacious than is stimulation of the lateral hemisphere, and (2) CS is most effective in models of generalized or focal epilepsy of the limbic system and least effective in models of focal epilepsy of the sensorimotor cortex. In 1977, Brown and associates [8] studied the monkey cerebellar cortex with chronic CS, determined that a charge density (QD) of 7.4 mC/cm2/phase is safe and is four to five times the level required to evoke efferent activity [9]. They warned that without careful attention to the charge densities delivered to the cerebellar cortex, the conducting elements would be destroyed (above QD: 10 mC/cm2/phase), eventually rendering an implanted cerebellar stimulator ineffective.
Clinical Series Using a radio-frequency (RF) linked stimulator with electrodes placed bilaterally on the superiomedial cerebellar cortex, CCS was applied in 1973 for the first time in epileptic patients (EP) by Cooper et al. [10]. A reduction by at least 50% of seizure activity in 18 of 32 patients was reported [11]. Five patients died, one at operation and four later in their sleep. Of the remaining nine non-responding patients, two discontinued CCS because of headaches after 2–6 months, two used the stimulator very infrequently and then stopped stimulating altogether after 13 months, one had equipment failure after 4 months of excellent relief, one with a temporal glioma had no reduction after 13 months of CCS, and three had no change in their seizure patterns after 14–50 months. The seizure types of these 18 patients and those of the three non-responding patients were then analyzed in regard to the effects of CCS (> Table 168-1). By 1987, 12 clinical series were reported involving 95 patients [7,10–30]. Rossi and coworkers [31] critically reviewed this literature and indicated that the risks of using CCS were acceptable. Up to 2005,
a total of 135 seizure patients treated with CCS have been reported (> Table 168-1), 116 were in 11 non-blinded clinical studies, of whom 88 (76%) benefited by seizure reduction. The three double-blind studies involving 19 patients are discussed next.
Double-Blind Studies Using RF linked stimulators, two double-blind studies of five [29] and nine [30] patients were reported as having no reduction of seizures with CCS. Dow [32] reviewed the data of Van Buren and associates [29] and came to a conclusion different from their ‘‘no significant difference.’’ After correcting their percentage change equation to read [(post op freq – pre op freq)/pre op freq] X 100% and then converting their percentage changes to seizures per day, Dow found that although three of the five patients had increased seizure counts after implantation before stimulation, nevertheless the first four patients showed a decrease in seizure rates of 73%, 69%, 73%, and 85%, respectively, during the CCS ‘‘ON’’ phases compared with the ‘‘OFF’’ periods of their double-blind study. Davis also recalculated their data [29] and confirmed Dow’s findings that the first four of the five patients had the same numbers of seizures per day reduced with CCS as Dow had calculated. It was also noted that Van Buren’s [29] post-hospital family evaluations indicated that ‘‘increased alertness, interest in surroundings, and diminished intrafamily strife were most frequently cited. The seizures seemed less severe, of shorter duration, and less debilitating.’’ Realistically, the 1978 report of Van Buren and associates [29] had a negative effect on the further use of CCS for the control of intractable seizures. In 1984, Wright and coauthors [30] assessed 12 epilepsy patients in a double-blind study using RF linked stimulators, with the results in nine were capable of being quantified. They
168
Cerebellar stimulation for seizure control
. Table 168-1 Summary of clinical reports of 11 non-blinded and 3 doubled-blinded seizure patients using cerebellar stimulation Year clinics (References)
Number of patients
Non-blind clinical series 1973 Cooper 32 (10,11) 1977 Gilman 6 (22) 1977 Fenton 1 (21) 1977 Dow 3 (7) 1979 Levy 6 (25) 14 1981 Bidzinski (13) 32 1982/91 Davis (14–20) 1984 Heath 8 (23)
Seizurefree
Charge density (mC/ cm2/phase)
Stimulator mode
14*
Maximum 19
Radiofrequency
1
Maximum 25–30 Maximum 25–30 Maximum 7.6
Radiofrequency
Maximum 19–30 Unknown
Radiofrequency
10.9–2.5
Radiofrequency and totally implanted stimulator Radiofrequency and totally implanted stimulator Radiofrequency and totally implanted stimulator Radiofrequency and totally implanted stimulator Radiofrequency
Seizures reduced
No change
18 5
Increased
1 1
2
3
3
5
8
1
19
8
4
4
4
1.5–2.5
1*
1984 Madrazo (26) 1984 Amin (12)
3
3
1.5 and up
2
2
1.5–2.5
1987 Klun (24) Subtotal
9 116
Double-blind clinical series 1978 Van 5 Buren (29) 1984 Wright 9 (30) 9280 2005 5 Velasco (33) Subtotal 19 Total of all 135 series Percentage
3 31 27%
4
2
57 49%
27 23%
Unknown
5
5*
31
13(68%)* 69*
5(32%) 33*
23%
51%*
25%*
Radiofrequency
Direct
1 1%
4* 4*
Radiofrequency
Maximum 25–30 Maximum 19
Radiofrequency
2.0
Totally implanted stimulator
Radiofrequency
1%
*See text
concluded that ‘‘no reduction in seizure frequency occurred that could be attributed to stimulation, though 11 of the patients considered that the trial had helped them.’’ On examining their
table data, it can be seen that one of the nine patients had an infection needing five operations with removal of the stimulator, while four of the other eight required further surgery. Of those
2825
2826
168
Cerebellar stimulation for seizure control
eight patients, four had reduction of seizure count by 11%, 43%, 20%, and 97%, respectively, with continuous CCS. In 2005, Velasco et al. [33] reported on their double-blind study involving five adult patients with medically intractable generalized tonicclonic seizures (GTCS) using a fully implantable battery-powered, programable pulse generator (Itrel 3: IPG; Medtronic Inc., Minneapolis, MN). This IPG was connected to dual cerebellar electrodes placed on the superio-medial cerebellar surface (> Figure 168-1). This stimulator was a major improvement over the earlier RF-linked pulse generators and controllers, because many of the RF-linked systems were not calibrated, the external antennae would slip from the implanted receiver sites, and batteries were deleted in 2–3 days. The five subjects continued to be treated with specific drugs for their seizure types and were followed for an average at least 3 months of observation prior to implantation (Baseline Phase: BL). Serial determination of drugs levels in serum confirmed that medical treatments had been followed correctly. All patients had neither less than eight nor more than 22 different motor seizures per month that is GTCS in five patients, and tonic seizures (TS) in four patients, while two patients had drop-attacks (DA) and another
with myoclonic and atypical absences (AA). > Table 168-2 summarizes the double-blind data for each patient. The Study was done with an approval of the US Food and Drug Administration using an Investigational Device Exemption (# G910165). All patients and their relatives were informed of the protocol and signed written consents according to the International Research Committees and Ethical Regulations. After the implantation surgery, all patients had 1 month of no stimulation (Sham phase). There is no significant difference between the pre-implant BL phase and the 1-month of Sham Phase (p = 0.58). Thereafter for the next 3 months, a double-blind study was done where three patients received stimulation and two did not (> Table 168-2; > Figure 168-2a). Stimulation parameters were adjusted to deliver a given charge density of 2.0 mC/cm2/phase, 10 Hz, 4 min. ‘‘ON’’ and 4 min. ‘‘OFF’’; at this level no patient experienced the stimulation. The procedure used for randomization was to assign patients a lottery number. In the first doubleblind month, there was little or no effect; but the suppression of GTC seizures was apparent by the third month (> Figure 168-2a). After the fourth month post-implantation, all subjects’ stimulators were then ‘‘ON’’ until the end of
. Figure 168-1 Surgical anatomy of implanting dual cerebellar electrodes onto the superio-medial cerebellar cortex. (a) Bilateral sub-occipital burr-holes to expose the lateral sinuses. Insertion of the electrode pads into the posterior fossa through bilateral dural openings. Note the bilateral anchors to stop forward migration of the electrode pad. (b) Posterior view of the cerebellum with implanted dual cerebellar electrode pads on the superio-medial cerebellar cortex. (c) Sagittal view of the posterior fossa and cerebellum with the electrode pad in place; the posterior pad protects the edge of the cerebellum and stops the pad from pulling out of the fossa
GTC TS DA
GTC TS DA
GTC TS DA
GTC TS DA
2
3
4
5
10 3 4
10 17 None
5 7 None
6 None 2
2 5 1
2
15 10 9
4 6
5 5
3
5
6 5 15
1
8 0 2
7 6
1 13
4
4
8 8 1
0
2 3
0 1
1
0
6 9 0
3
2 3
0 11
1
1
3 11 1
4
1 5
0 7
0 Fever 2
3 16 2
5
I Stimulation: ON
11 12 26 11 0 3 0 6 1 0 0 1 KCB 5: Stim: OFF
1 7
Stim: ON 3 4 Stim: ON 7 1 2
2 11
1
1
4 12 1
2
2 Stim: ON 4 0
1 Stim: OFF 3 11 0 Stim: ON 2
Double blind
13 3 0
2 3
0 5
9 4 0
2 4
0 6
4
1
12* 2
3 9 0
7
3 10 1
6
II
10 3 5
3 2
2 2
2
1
2 6 0
8
16 9 7
1 5
2 1
0
0
4 8 0
9
10 13 6
1 4
0 4
0
2
1 11 0
10
III
4 9 2
2 5
0 5
1
0
0 9 0
11
1 9 3
3 2
0 4
Explanted
3 8 0
12
GTC general tonic-clonic seizures; TS tonic seizures; DA drop-attacks; MCS/AA: myoclonic seizures/atypical absences
10 5 4
5 6
3 5
3
3
5 0 3
GTC TS MCS/AA
1
KCB
Basal Months 3
Seizure type
Sham NO Stim
Implantation
3 2
2 6
5 12 0
14
/ 15 / 5 / 2 Electrode reposition
2 2
0 3
2 10 0
13
IV
0 2
0 10
2 14 0
16
2 4
0 1
2 6 0
17
12 9 5 11 12 14 10 8 4 Stim: ON
1 3
0 8
2 12 1
15
V
6 9 5
1 5
0 3
2 1 0
18
8 4 7
3 7
5 3
1 0 1
19
VI
10 10 8
0 9
0 6
3 1 0
20
5 7 3
2 12
0 0
1 1 0
21
1 2 0
0 5
0 12
2 1 0
22
VII
5 3 4
1 4
0 1
1 0 2
23
3 2 2
1 4
5 2
0 0 1
24
. Table 168-2 Double-blind study of Velasco et al. [33] Details of the monthly counts of the four different seizures for the five patients in the four phases (Basal: 3 months, Sham: 1 month, Double-Blind: 3 months and Stimulation ON: 21 months) over the 28 months of the study
Cerebellar stimulation for seizure control
168 2827
2828
168
Cerebellar stimulation for seizure control
their study (> Figure 168-2b). The patients remained on their medications throughout the study. During the double-blind phase, both patients and the evaluator were ‘‘blinded’’ with regards to whether the stimulator was ON or OFF, a different investigator manipulated the
stimulation code and checked the stimulator function monthly. Significant changes between the BL values and those of the later post-stimulation epochs (> Table 168-2, > Figure 168-2) were determined by using parametric dependent Student ‘‘T’’ tests
. Figure 168-2 Generalized Tonic-Clonic Seizures with the effects of Cerebellar Stimulation in five patients [33]. (a) Basal data (3 months) compared to the Double-Blind Phase (3 months). KCB2, 3 and 4: stimulator ON. KCB1 and 5: stimulator OFF. (b) Composite of the five patients in each phase. ‘‘Pt.2 Fever’’ indicate the result of the 12 seizures (average BL level of 4.6/month, occurring at the sixth month possibly linked with the infection and fever (Velasco et al. 2005 [33])
Cerebellar stimulation for seizure control
for the number of seizures. However, as time goes on the effects do become clearer. The three ‘‘pre’’ months (BL) versus the 6 months up to and including the ninth month show a significant reduction in GTCS rate (p = 0.027). Patient KCB2 developed an infection in the tissues around the CCS wires, which were all completely removed with the IPG at 12 month post-implant. By the 21–24 month, the three patients (KCB1, 3 & 4), who finished this series, ended with reduction of GTCS counts to mean of 24% of the BL levels (> Figure 168-2b). In regards to comparing seizures/month per Subject, there were 1.2 GTCS when compared with those of the BL phase where a mean of 5 GTC seizures per month per patient occurred. This represented a significant improvement (p < 0.001). Since Subject KCB5 was re-evaluated for the 3-month period ending in April 2004, which was approximately 3 years since the electrode pad was repositioned and CCS started, he was having 1–3 GTCS/month (Pre-operative Basal: 11.67 seizures/month). By adding the third year result for Subject KCB5 to the other three evaluated at the end of 2 years, the four Subjects were having a mean Basal rate of 9.57 GTC seizures/month; with CCS the mean seizure rate/month was 1.4 (14.6%). Therefore in the four Subjects with stimulators, the improvement of reduced GTC seizure frequency by CCS occurred and was sustained over a 2–4 year period, to a statistically significant degree. In contrast, decreases of Tonic seizures were only significantly reduced in the 7th epoch, decreasing from 6.2 TS/month/patient in the BL phases to 3.2 TS/month/patient which represents a decrease to 52% of the BL values (p < 0.05). Patients referred to changes in their Tonic seizures with CCS as being less intense and of shorter duration. Dropattacks data analysis: Subject KCB2 showed a reduction with CCS to 18% of the BL value in the 11 months of stimulation, while with Subject KCB5 the seizure count reduced to 35% by the 14th month of CCS (> Table 168-2). In Patient KCB1, Myoclonic/AA seizure counts were
168
reduced with CCS over the 24 months to 16% of the BL counts (> Table 168-2).
Discussion of the Three DoubleBlind Studies From the Velasco et al. study results [33], the major finding is that little if no effect occurs until the 2nd month of continuous stimulation (> Table 168-2; > Figure 168-2b). Both the Van Buren et al. [29] and the Wright et al. [30] studies show positive effects when Patients were using their stimulators over a longer term. This also applies to the results from non-blinded studies, where Patients are followed over long periods (> Tables 168-1– > 168-3). This indicates that CCS takes 3+ months to have its full effect on controlling seizures, and that double-blind studies should not be done unless the stimulation is used for 3 months. Also no cross-over studies should be attempted, because of the long carryover effects following the use of CCS [13,15]. The use of totally implantable batteryoperated IPGs were a major advancement in the use of CCS considering their accurate programmability, with complete control by the physicians and minimal if any interference by the patients and families. All five patients benefited from this device [33].
Surgical Procedure Under general anesthesia, Patients were operated in the lateral position with the head in a Mayfield head holder rotated 45–60 . A midline posterior incision from the inion to second cervical vertebra was opened. Below the inion, bilateral sub-occipital burr-holes were made 1.5 cm from the midline (> Figure 168-1a). The transverse sinuses were identified; the dura below each sinus was opened for 5 mm horizontally. The upper medial surface of each cerebellar hemisphere was identified and the silicone pad
2829
Amin Cooper Davis Fenton Gilman Health Klun Levy Madrazo Velasco Total
1 11
4 3 1
1 21
5
1 18 23 1 5 7 9 6 1 5 76
5
1 1
1 2
No
29
2 4 3 1
2 17
9
3 6
Total
2
1 1
SeizFree
Myoclonic
1 12 4 1 1 3 4 2 1 5 32
6
2 4
1
1
No
14
2
2 3
15
2
6 4
2
4 2
No
Reduced
Reduced
1 2 8
1
4
Total
Atonic
3
1 2
0
Selz.Free
Seiz-free
Total
Seiz-free
Total
Reduced
1 1
1 4
SelzFree
Benefit 1 18 28 1 5 8 7 3 3 5 79 (85%)
2 2
6 9 1
Total
Absences
Patients 1 21 33 1 6 8 9 6 3 5 93 7
4 1
2
2
1 1
No
1 2 7
1
3
Reduced
Reduced
1
1
No
40
8 2 3
4
6 17
3 6
3
Total
Tonic
Total
12
4
1 7
1
1
Seiz.Free
Seiz-free
Complex partial
3 4
1
Reduced
20
3
4
2
5 6
Reduced
1
1
No
8
2
2 8
4
No
168
Clinic Amin Cooper Davis Fenton Gilman Health Klun Levy Madrazo Velasco Total
Simple partial
Tonic-clonic seizures
. Table 168-3 Ten clinical studies reporting on 93 patients with 174 intractable seizures treated with CCS
2830 Cerebellar stimulation for seizure control
Cerebellar stimulation for seizure control
with its 4-button in-line electrodes was gently introduced onto the right and left superio-medial surface of the cerebellum (> Figure 168-1b,c). Fluoroscopic guidance was used to assist and verify the electrode placements. To avoid forward movement of the electrodes, each emerging lead wire with an attached silicone anchor was sutured to the adjacent periosteum and connective tissue (> Figure 168-1a). Once both electrodes were in place, the two leads connect in a ‘‘Y’’ join with its cable, which is tunneled to a subcutaneous pocket site on right chest, where it connects to the implanted battery-operated IPG.
Stimulation Parameters Stimulation parameters were adjusted to deliver a given charge density of 2.0 mC/cm2/phase, at this level patients do not experienced the stimulation. Charge density (mC/cm2/phase = amplitude (mAmp) pulse width (ms) cathodal electrode area (cm2) The stimulation frequency was set at ten pulses per second (pps) to follow the experience of Cooper et al.’s series [10,11], who defined that 10–20 pps was effective for seizure reduction, rather than using higher frequency (100–200 pps) which has been shown to be effective in reducing
168
spasticity. The stimulation was turned ‘‘ON’’ for 4 min alternating with 4 min ‘‘OFF’’ throughout the 24 h, has been used effectively by Davis [14–20]. Cooper [11] used 1 min of stimulation to the electrodes on one side of the superio-medial cerebellar surface, then the stimulation was electronically switched to the other side for 1 min; this pattern was switched back and forth. Both CCS systems used intermittent stimulation to each cortex that ran for a total time of 12 h throughout the 24 h, which is effective and, in this reported study [33], was a definite conservation for nonrechargeable batteries in implantable stimulators, particularly when running the frequency at 10 pps instead of 100–200 pps.
Seizure Types and CCS Results The results of CCS on 93 patients reported from 10 clinical series [11,12,20–26,33] had a total of 174 seizures of seven types were reviewed in > Tables 168-3 and > Tables 168-4, of whom 79 patients (85%) benefited. > Table 168-3 show details of the seven seizures types according to each Clinic. > Table 168-4 shows the details of each seizure type responding to CCS. Generalized tonic-clonic seizures were the most common, being present in 76 (82%) of the 93 patients, of whom 29 (38%) became seizure-free, 32 (42%)
. Table 168-4 The results of each of the seven seizure types responding to CCS Seizure types Tonic-clonic Simple partial Complex partial Absences Myclonic Atonic Tonic Total Seizure types
Seizure-free (%) 76 (44%) 14 (8%) 40 (23%) 21 (12%) 9 (5%) 8 (5%) 6 (3%) 174 7
29 (38) 3 (21) 12 (30) 5 (24) 2 (22) 1 (16) 52 (30) Improved: 80%
Seizure reduction (%)
No reduction (%)
32 (42) 7 (50) 20 (50) 11 (52) 6 (67) 7 (87) 4 (67) 87 (50)
15 (20) 4 (29) 8 (20) 5 (24) 1 (11) 1 (13) 1 (17) 35 (20)
2831
2832
168
Cerebellar stimulation for seizure control
benefited from reductions, and 15 (20%) were not affected. Complex partial seizures were the next most frequent, being present in 40 (43% of patients); 30% became seizure-free, another 50% experienced reductions in seizures, and 20% were not affected. Of interest was the success in stopping or reducing myoclonic seizures in eight of nine patients. After 1974, Davis and coworkers [16–20, 34,35] implanted cerebellar stimulators in 332 with spastic patients (SP), of whom 90% had cerebral palsy (CP); of these, 27 had intractable seizures. There were six who had epilepsy per se and were also implanted and studied. The results in these 33, were reported in 1982–1984 [16,18–20], where 27 (82%) had had seizure reduction. The stimulation charge densities used were 0.9–2.5 mC/cm2/phase delivered at 10–30 pps for the EP group and 150–180 pps for the spastic group, using bilateral electrode pads positioned on the superio-medial cerebellar cortex (> Figure 168-1). Among the 27 patients of Davis et al. [15] followed from 1974 through 1991, the mean age at implantation was 11.7 years (range 4–26 years), with 11 patients in the 4–10 year group, 12 in the 11–18 year group, and 4 in the 20–26 year group. Twenty-three (85%) have benefited from the long-term use of CCS. Eleven patients have continued with CCS for a mean of 13.9 years (9–17 years), with all benefiting (seizure-free, 7; reduced numbers of seizures, 4). Of the 16 who did use CCS for an average of 7.3 years (2–14 years) and who now have nonfunctioning units, 12 (75%) continued to benefit (seizure-free, 5; reduced seizures, 7). This carryover effect was observed by Bidzinski and coauthors [13] in 4 of the 12 epileptic patients undergoing only 10–12 days of stimulation with temporary leads; 1 year later, they reported the four patients as still significantly improved. Zeng and coworkers [37] and Dow [32] reported on Patient JG, who originally underwent bilateral implantation of Avery radiofrequency-coupled units (Avery, Farmingdale, NY) by Cooper et al. [11] in 1974 for
her complex partial (18–20/day) with generalized tonic-clonic seizures (2 per month). In February 1990, her two RF- receivers were replaced with a Cordis Stimucor, a totally implantable programmable battery-powered IPG (Cordis Corp., Miami) by Davis. She has been followed by Drs. Robt. Dow and Norman So (Portland, OR) who reported in 1997 [38] that she has one to several minor seizures at the time of her menses, and has been able to increase her work output at a sheltered workshop from 95 cents/h (1989) to $2.95/h (1990–1991). Her interictal spike frequencies were recorded and computer-analyzed by Zeng and coworkers [37]. With CCS ‘‘ON’’ 44 paroxysmal epileptiform discharges (PDs) were recorded in 73 min of random samples, while 119 PDs were recorded in similar sampling with the CCS ‘‘OFF’’. As of 2008, Patient JG is 49 years old and continuing to use a functioning Medtronic Itrel3 IPG for CCS, and she has only 3–4 complex partial seizures/month after 34 years of using CCS. Upton [28] reported significant reductions in PDs in three other EP patients during and after CCS. Using the positron emission tomography (PET) scanner, Cooper and Upton [39] reported the activation of glucose metabolism in hypometabolic areas of the cerebral cortex to normal with CCS in two cerebral palsy patients, whose spasticity was reduced by CCS. Upton (personal communication) found similar PET scan values in two epileptic patients using CCS. Wood and associates [40] reported a significant increase in nor-epinephrine (NE) in the cerebrospinal fluid of five epileptic patients after 7 days of CCS but not in three epileptic patients stimulated with subdural cerebral cortical leads. This raises the issue of whether some biochemical change is responsible for the effects of CCS. Dailey and colleagues [41] measured NE concentrations in 15 discrete areas of the central nervous system (CNS) in two strains: (1) genetically epilepsyprone rats and (2) control rats; and found markedly decreased NE levels in the epileptic rats.
Cerebellar stimulation for seizure control
They postulated that decreased levels of NE are important in the seizure predisposition of these rats and that this does not result from the seizure experience.
Safety Considerations The major concern in undertaking the management of patients with seizures using CCS, has been their long-term safety. There have been no reports of aggravation of seizures or of any deterioration in the patients’ neurological disabilities. Davis et al. studies [14–20,38] have indicated that CCS at the charge densities used (QD:1–2 mC/cm2/phase) has not initiated seizures, as seen particularly in 62 CP patients who had had a history of seizures prior to but not in the 3 years immediately before or the average of 4.3 years per patient after CCS [20]. Davis et al. reported that 33 seizure patients, CCS had not produced any detrimental effects in neurological function, in agreement with the results of other series [7,10–13,21–31,42,43]. Only one of Davis’ CP patients (12 years old) whose equipment was not functioning at the time had an increase in tonic-clonic seizures from one per week to 1 per day. In Davis’ series of 332 spastic patients from 1974 to 1982 undergoing CCS in an attempt to reduce spasticity and improve performance [34,35], none reported neurological deterioration resulting from this level of QD stimulation. There were no deaths during the peri-operative period or the post-operative 4 weeks. Infections had occurred in the tissues around the implants in 4.4% of the patients, representing 2.7% of the procedures performed [36]. Two CP patients (0.6%) had spontaneous cerebellar hemorrhages at post-implant of 5 and 12 years, respectively [44]. Three CP patients died from varied causes unrelated to CCS and had their cerebelli examined [45]. The two patients who had kept their CCS charge densities in the range
168
of 0.9–4 mC/q.cm/phase showed no histological change under the electrode buttons compared with adjacent control sites, though slight surface indentation was caused by the electrode pads. In the third patient, who independently adjusted his RF stimulator to a QD range of 14–25 mC/ cm2/phase which led to a return of spasticity; his cerebellum, at the electrode sites, showed focally severe Purkinje’s cell and granule cell loss and dense fibrillary gliosis, with the adjacent cerebellar cortex being normal. This is similar to the finding of Urich and colleagues [43] in a patient with amyotrophic lateral sclerosis who was stimulated with CCS for 33 months at a QD of 33 mC/cm2/phase. According to Davis’ literature review [34,36 17], clinics have implanted 676 patients, 90% of whom were spastic. There was one patient who died (0.1%) postoperatively from an extradural hematoma [10], CCS had not been started. Six epileptic patients undergoing CCS had died from unrelated causes and have been studied at autopsy [11,42,43,46]. Cooper and coauthors’ reported a 25-year-old male having CCS without benefit over 17 months with a QD of 33 mC/cm2/phase; the autopsy report revealed no electrode encapsulation and little or no damage to the underlying cortex, whose histology appeared to be similar to that of the biopsies taken at the time of implantation. In another of Cooper’s epileptic patients who had had 4 months of CCS (QD: 33 mC/cm2/phase), Gilman and coworkers [46] reported electrode encapsulation and moderate depletion of Purkinje’s cells throughout the cerebellum, particularly underneath both electrodes. Robertson and associates [42] reported autopsy findings in three epileptic patients who underwent CCS from 6.6 to 15 months (QD 10–17 mC/ cm2/phase. All these patients had capsule formation and exhibited extensive and severe damage of neural tissue at the caudal cerebellar border that was caused by the wire from the electrode assembly. A number of areas directly underneath the electrodes showed only minimal changes
2833
2834
168
Cerebellar stimulation for seizure control
similar to those found in cortical areas remote from the array; the most extensive damage was ‘‘usually associated with pieces of tissue that had adhered directly to the electrode capsule.’’ In the report of Wright et al. [30] on 12 epileptic patients, one patient died 16 months after implantation, ‘‘with Purkinje’s cell populations that were similar to those in a biopsy taken at implantation. There was deep grooving under the electrode pad assembly, but only a minor degree of neuronal loss and gliosis. The electrode pads were covered with a thick layer of fibrous tissue that did not adhere to the cortex’’. The two cerebellar electrode pads used in the above reported studies, were composed of one for the each cerebellar superio-medial surface, and were made of thin silicon sheets embedded with 2 rows of 4 button electrodes, one row for anodes and one for the cathodes. Each assembly was connected by 2 insulated leads to one each of two RF receivers located subcutaneously in the infraclavicular region. Davis’ narrower dual four in-line button electrode pads with posterior flexible expansions (> Figure 168-1) do protect the cerebellar edge from the emerging lead wires, and overall presents less bulk to the cerebellar cortex with less chance of breaking bridging veins. The bilateral total of 8 platinum electrodes were all cathodes, whereas the anodal plate was of a larger area placed infraclavicular and subcutaneously, either attached to the anodal side of the RF receiver or to the case of the battery-powered IPG. Both connecting wires from each electrode assembly join in the extra-cranial region and go to only one RF receiver or IPG. The 4-button electrode pad does have a risk of sliding forward on the superior cerebellar surface, so losing its target site. To mitigate this risk, the extra-cranial emerging lead from each of the dural opening, has an attached anchor, which is sutured securely to the extra-cranial tissues to stop this forward displacement. In the Velasco et al. study, 3 of the 5 patients had this complication and lost effectiveness
needing re-opening and re-positioning of the electrode with anchoring applied. Davis et al. [35] experienced this electrode pad migration problem in 4 patients early in their series of 332 spastic patients implanted, and then used the mitigating lead anchors to tie-down the connecting wires extra-cranially, without further problems. Davis et al. [17] have removed 60 such electrode pads from 30 patients after implantation durations of 1–46 months (broken leads, 22; infection, 8); only 2 electrode pads showed encapsulation with thin slightly reddish membranes. Access resistance was measured in 87 patients at implantation (mean: 374 O) and in 28 at 6 months to 4 years after implantation (mean: 334 O), showing about a 10% drop. This finding is opposite to the increase in access resistance and capsule formation found in monkeys with CCS as reported by Gilman and associates [46]. Dow, however, commented when discussing Gilman and associates report [46] that the human posterior fossa is 13 times larger than that of the monkey, thus allowing easier access. The most difficult problems in longer-term control of seizures have involved the RF-linked systems (> Table 168-3). Patients and families often find the equipment to be ‘‘too much trouble’’ with battery changes every 2–3 days and skin rashes from the adhesive tape used to hold the antenna to the skin. Against medical instructions, some patients and families would also adjust the external stimulator, leading to the serious consequences of permanent damage to the cerebellum and loss of benefits; four patients had to be re-operated and have their electrodes moved to a new adjacent site to recapture control. After 1977, Davis’ clinic calibrated and set, with a hidden screw, the pulse generators to deliver charge density (QD) pulses at 0.8–1.0 mC/cm2/ phase. If the QD dropped below 0.8, the clinical effects were lost [14–16,34,35] as confirmed by Wong and associates [48]. If the QD levels exceeded 4–5 mC/cm2/phase, the effects of CCS on spasticity and performance stopped
Cerebellar stimulation for seizure control
. Figure 168-3 The ‘‘window effect’’ of cerebellar cortex stimulation range according to Brown et al. [8] and Babbs et al. [9] from monkey data; and from clinical experience of Davis et al. [35]
[14–18,34,35], indicating that a ‘‘window effect’’ was present (> Figure 168-3). The use of the fully implantable units has been a major step forward in the safer management of patients with seizures and spasms complicated by behavioral problems [14–16,20], because the equipment cannot be damaged or the setting changed, though regular testing of the CCS equipment and leads is necessary.
Conclusions A detailed review in > Table 168-1 showed that 11 non-blinded CCS series involving 116 patients showed that the 88 patients (76%) benefited. The results of CCS on 93 patients where details of the seizure types were reported from 10 clinical series [11,12,20–26,33] had a total of 174 seizures of seven types which were detailed in > Tables 168-3 and > Tables 168-4, of whom 79 patients (85%) benefited. Although the earlier two doubleblinded clinical reports [29,30] casted doubts as to the benefits of CCS; in 2005 a double-blind study was published by Velasco et al. [33] indicating that the testing CCS phase must be not be less than 3 months with no cross-over, because of CCS carry-over effects. This CCS study on five patients with GTC seizures had reduction of seizures after 2+months. With CCS in the five patients did continue to decrease the seizure
168
counts over the first 6 months of stimulation, and then maintains this effectiveness over the study period of 2 years and beyond, expect for one patient who had an infection in the tissues at 11 months, when the system was removed. The statistical analysis showed a significant reduction in tonic-clonic seizures (p < 0.001) [33]. It is important to note that all five patients did responds to CCS, which was from a fully implanted battery-operated pulse generated with carefully controlled stimulation. This is a major equipment improvement since the early double-blind series [29,30] using RF-coupled systems. It should be noted that implanted battery operated IPG are now available with rechargeable battery which are designated to last probably 10 years. The superio-medial cerebellar cortex appears to be a significantly effective and safe target for limited electrical stimulation for decreasing motor seizure over the long-term. CCS is relatively safe, non-ablative and removable; and could be offered to patients with intractable seizures originating from bilateral or multiple extra-temporal foci.
References 1. Moruzzi G. Sui rapporti fra cervelletto e corteccia cerebrate. Arch Fisiol 1941;41:157-82. 2. Cooke PM, Snider RS. Some cerebellar influences on electrically-induced cerebral seizures. Epilepsia 1955;4: 19-28. 3. Iwata K, Snider RS. Cerebello-hippocampal influences on the electro-encephalogram. Electroencephalogr Clin Neurophysiol 1959;11:439-46. 4. Fanardjian VV, Donhoffer H. An electrophysiological study of cerebello-hippocampal relationships in the unrestrained cat. Acta Physiol Hung 1963;24:321-33. 5. Dow RS, Fernandez-Guardiola A, Manni E. The influence of the cerebellum on experimental epilepsy. Electroencephalogr Clin Neurophysiol 1962;14:383-98. 6. Mutani R, Bergaini L, Donguzzi T. Experimental evidence for the existence of an extrarhinencephalic control of the activity of the cobalt rhinencephalic epileptogenic focus: effects of the paleocerebellar stimulation. Epilepsia 1967;10:351-62.
2835
2836
168
Cerebellar stimulation for seizure control
7. Laxer KD, Robertson LT, Julien RM, Dow RS. Phenytoin: relationship between cerebellar function and epileptic discharges. In: Glaser GH, Penry J, Woodbury DM, editors. Antiepileptic drugs: mechanism of action. New York: Raven Press; 1980. p. 415-27. 8. Brown WJ, Babb TL, Soper HV. Tissue reactions to longterm electrical stimulation of the cerebellum in monkeys. J Neurosurg 1977;47:366-79. 9. Babb TL, Soper HV, Lieb JP. Electrophysiological studies of long-term electrical stimulation of the cerebellum in monkeys. J Neurosurg 1977;47:353-65. 10. Cooper IS, Amin I, Riklan M, et al. Chronic cerebellar stimulation in epilepsy: clinical and anatomical studies. Arch Neurol 1976;33:559-70. 11. Cooper IS, Ricklan M, Amin I, Cullinan T. A long-term follow-up study of cerebellar stimulation for the control of epilepsy. In: Cooper IS, editor. Cerebellar stimulation in man. New York: Raven Press; 1978. p. 19-38. 12. Amin IM. The reduction of spasticity, involuntary movements and seizures using a fully implantable cerebellar stimulator. In: Davis R, Bloedel JR, editors. Cerebellar stimulation for spasticity and seizures. Boca Raton, FL: CRC Press; 1984. p. 211-3. 13. Bidzinski J, Bacia T, Ostrowski K. Effect of cerebellar cortex electrostimulation on the frequency of seizures in drug-resistant epilepsy. Neurol Neurochir Pol 1981;31: 605-9. 14. Davis R, Emmons SE. Safety and efficacy of cerebellar stimulation for seizure control. Boll Lega Ital Epil 1988;64:105-15. 15. Davis R, Emmons SE. Cerebellar stimulation for seizure control: 17 year series. Stereotact Funct Neurosurg 1992;58:200-208. 16. Davis R, Engle H, Kudzma J, Dusnak A. Update of chronic cerebellar stimulation for spasticity and epilepsy. Appl Neurophysiol 1982;45:44-50. 17. Davis R, Gray EF. Technical problems and advances in cerebellar stimulating systems used for reduction of spasticity and seizures. Appl Neurophysiol 1980;43:230-43. 18. Davis R, Gray E, Engle H, Dusnak A. Reduction of intractable seizures using cerebellar stimulation. Appl Neurophysiol 1983;46:57-6. 19. Davis R, Gray E, Engle H, Dusnak A. The reduction of seizures cerebral palsy and epileptic patients using chronic cerebellar stimulation. Acta Neurochir Suppl (Wien) 1984;33:161-7. 20. Davis R, Gray E, Engle H, Dusnak A. The reduction of seizures cerebral palsy and epileptic patients using chronic cerebellar stimulation. In: Davis R, Bloedel JR, editors. Cerebellar stimulation for spasticity and seizures. Boca Raton, FL: CRC Press; 1984. p. 247-61. 21. Fenton GW, Fenwick PBC, Brindley GS, et al. Chronic cerebellar stimulation in the treatment of epilepsy: a preliminary report. In: Penry JK, editor. Epilepsy: the eighth international symposium. New York: Raven Press; 1977. p. 333-40.
22. Gilman S, Dauth GW, Tennyson VM, et al. Clinical, morphologic biochemical, and physiological effects of cerebellar stimulation. In: Hambrecht FT, Reswick JR, editors. Functional electrical stimulations. New York: Marcel Dekker; 1977. p. 191-226. 23. Heath RG. Cerebellar vermis stimulation: long-term response intractable behavioral disorders and epilepsy. In: Davis R, Bloedel JR, editors. Cerebellar stimulation for spasticity and seizures. Boca Raton, FL: CRC Press; 1984. p. 263-7. 24. Klun B, Stojanovic V, Strojnik P, et al. Chronic cerebellar stimulation in the treatment of epilepsy. In: Wullenweber R, Klinger M, Broc M, editors. Advances in neurosurgery, vol 15. Berlin: Springer-Verlag; 1987. p. 205-9. 25. Levy KF, Auchterlonie WC. Chronic cerebellar stimulation in the, treatment of epilepsy. Epilepsia 1979;20: 235-45. 26. Madrazo I, Rosas VH. Chronic cerebellar stimulation for reduction of grave behavioral changes and seizures in psychiatric patients. In: Davis R, Bloedel JR, editors. Cerebellar stimulation for spasticity and seizures. Boca Raton, FL, CRC Press; 1984. p. 273-80. 27. Riklan M, Halgin L, Schulman M, et al. Behavioral alterations following acute, short-term, and longer-term cerebellar stimulation in humans. In: Cooper IS, editor. Cerebellar stimulation in man. New York: Raven Press; 1978. p. 161-83. 28. Upton ARM. Neurophysiological mechanisms in modification of seizures. In: Cooper IS, editor. Cerebellar stimulation in man. New York: Raven Press; 1978. p. 39-57. 29. Van Buren JM, Wood JH, Oakley J, Hambrecht F. Preliminary evaluation of cerebellar stimulation by double-blind and stimulation and, biological criteria in the treatment of epilepsy. J Neurosurg 1978;48:407-16. 30. Wright GDS, McLellan DL, Brice JG. A double-blind trial of chronic cerebellar stimulation in twelve patients with severe epilepsy. J Neurol Neurosurg Psychiatry 1984;47:769-74. 31. Rossi GF, Colicchio G, Scerrati M. La stimolazione cerebellare cronica nel trattamento dell’epilessia farma coresistente. Boll Lega It 1987;60:5-12. 32. Dow RS. Personal communication, 1977. 33. Velasco F, Carrillo-Ruiz JD, Brito F, Velasco M, Velasco AL, Marquez I, Davis R. Double-blind, randomized controlled pilot study of bilateral cerebellar stimulation for treatment of intractable motor seizures. Epilepsia 2005;46:1071-81. 34. Davis R. Clinical experience with cerebellar stimulation in the treatment of spastic cerebral palsy. In: Lazorthes Y, Upton ARM, editors. Neurostimulation: an overview. Mt Kisco, NY: Futura; 1985. p. 213-30. 35. Davis R, Kudzma J, Gray E, Dusnak A. Graded clinical effects in spastic cerebral palsy groups following chronic cerebellar stimulation. In: Davis R, Bloedel JR, editors.
Cerebellar stimulation for seizure control
36.
37.
38.
39.
40.
41.
Cerebellar stimulation for spasticity and seizures. Boca Raton, FL: CRC Press; 1984, p. 223-43. Davis R, Kudzma J, Ratzan K. Management of infected cerebellar stimulation systems. Neurosurgery 1982;10: 340-43. Zeng J, Smith DB, DeToledo J, Dow RS. The effect of cerebellar stimulation on interictal spike frequency (abstract). American Epilepsy Society Meeting. Epilepsia, 1990. Davis, R. Cerebellar stimulation for seizure control. In: Gildenberg P, Tasker R, editors. Stereotactic and functional neurosurgery, Chapter 201. New York: McGraw-Hill; 1997. p. 1945-51. Cooper IS, Upton A. Therapeutic implications of modulation of metabolism and functional activity of cerebral cortex by chronic stimulation of cerebellum and thalamus. Biol Psychiatry 1985;20:811-3. Wood JH, Lake CR, Ziegler MG, et al. Cerebrospinal fluid norepinephrine alterations during electrical stimulation of cerebellar, cerebral surfaces in epileptic patients. Neurology 1977;27:716-24. Dailey JW, Mishra PK, Ko KH, et al. Noradrenergic abnormalities in the central nervous system of seizurenaive genetically epilepsy-prone rats. Epilepsia 1991;32: 168-73.
168
42. Robertson LT, Dow RS, Cooper IS, Levy LF. Morphological changes associated with chronic cerebellar stimulation in the human. J Neurosurg 1979;51:510-20. 43. Urich H, Watkins ES, Amin I, Cooper IS. Neuropathological observations on cerebellar cortical lesions in patients with epilepsy and motor disorders. In: Cooper IS, editor. Cerebellar stimulation in man. New York: Raven Press; 1978. p. 145-53. 44. Zuccarello M, Sawaya R, Lukin R, DeCourten-Myers G. Spontaneous cerebellar hematoma associated with chronic cerebellar stimulation. J Neurosurg 1986;65:860-62. 45. Gyori E, Davis R. Morphologic findings in cerebellar folia following chronic cerebellar stimulation in cerebral palsy (three autopsy reports). In: Davis R, Bloedel JR, editors. Cerebellar stimulation for spasticity and seizures. Boca Raton, FL: CRC Press; 1984, p. 109-24. 46. Gilman S, Dauth GW, Tennyson VM, et al. Morphological and biochemical effects of chronic cerebellar stimulation in monkey. Trans Am Neurol Assoc 1975;100:9-17. 47. Wright GDS, Weller RO. Biopsy and post-mortem findings in a patient receiving cerebellar stimulation for epilepsy. J Neural Neurosurg Psychiatry 1983;46:266-73. 48. Wong PKH, Hoffman RI, Froese AB, et al. Cerebellar stimulation in the management of cerebral palsy: clinical and physiological studies. Neurosurgery 1979;5:217-24.
2837
152 Classification of Epileptic Seizures and Epilepsies H. O. Lu¨ders . S. Noachtar
Systems used to classify signs, symptoms, syndromes, and diseases have been continuously evolving since the origin of clinical medicine. This was the consequence of a better understanding of the pathological conditions we were classifying and/or the availability of progressively better diagnostic procedures. The epilepsies and the main symptom associated with epilepsy— namely, the epileptic seizure—have been subject to innumerable classification attempts dating back to the early days of medical literature [1]. In 1970, the International Classification of Epileptic Seizures (ICES) provided, for the first time, a more generally accepted classification system, which facilitated communication among epileptologists [2]. The current version of this classification system (1981), which is mainly based on clinical semiology and electroencephalographic criteria, employs a double dichotomy that divides the seizures into generalized and partial seizures on one side and further subdivides the partial seizures into complex and simple partial seizures, depending on whether consciousness is altered or preserved during the ictal event [3]. This classification has been criticized by some authors [4], but compared to the older classification systems, which focused on the highly variable seizure symptomatology, the ICES system represented a major simplification that permitted even nonexperts to classify seizures correctly. For pharmacological treatment decisions, the dichotomy ‘‘generalized’’ versus ‘‘partial’’ actually provides the essential information needed to select the anticonvulsant most likely to be effective. Alteration of consciousness during an ictal event serves as a criterion to assess #
Springer-Verlag Berlin/Heidelberg 2009
the impact of seizures on the quality of life. Therefore, a better control of ‘‘complex partial seizures’’ is usually a good index of therapeutic efficacy. However, this classification system integrates very little localizing information, which is essential for epilepsy centers concerned with epilepsy surgery, as also for clinicians trying to define the location of a lesion by analyzing the clinical characteristics of seizures [5]. A recently published proposal for a seizure classification was developed to serve the special needs of epilepsy surgery centers and is currently in use in several selected epilepsy centers [5,6]. This classification system is exclusively based on the ictal seizure semiology and provides detailed localization information. The objective of this classification system is to provide the essential clinically relevant information about a patient with epileptic seizure in (1) a standardized and (2) the shortest possible format.
Theoretical Considerations Neurosurgical procedures have given us extensive material to define the effects of electrical stimulation of the human brain. Electrical stimulation can serve as a model to define epileptic seizure symptomatology. The seizure onset zone has been defined as the area of cortex from which seizures originate. It usually also includes the region of cortex which is involved earliest in the spread of seizures. However, it is important to recognize that an epileptic discharge will not be associated with any clinical symptomatology unless it spreads to eloquent cortical areas
2562
152
Classification of epileptic seizures and epilepsies
(> Figure 152‐1). The symptomatogenic zone is the area of cortex that, when activated by the epileptic discharge (or experimentally by electrical stimulation), produces the clinical symptomatology of the epileptic seizure [7].
Classification of Epileptic Seizures The classification of epileptic seizures presented here is based exclusively on the clinical semiology of an epileptic event. The electroencephalographic (EEG) data, if available, can be used to define the location of the epileptogenic zone (see classification of the epilepsies below) but for the purpose of seizure classification are used only to decide
. Figure 152‐1 This figure shows some eloquent cortex in humans as identified by electrical stimulation. Note that electrical stimulation in some cortical regions is not associated with any consistent clinical symptomatology. SNMA supplementary negative motor area; SSMA supplementary sensorimotor area; M1 primary motor area; S1 primary sensory area; PNMA primary negative motor area; BTL basal temporal language area. (modified from Lu¨ders and Nachter [6], with permission.)
whether we are dealing with an epileptic seizure or not [8]. As outlined above, the semiology of an epileptic seizure is an expression of the symptomatogenic zone activated by the epileptogenic discharge. Almost all focal and many generalized seizures are characterized by a seizure evolution. Focal ictal epileptic discharges tend to spread to contiguous cortical areas, and there are typical spread patterns that reflect common seizure semiology evolutions. The initial seizure symptomatology is usually the most important information for the localization of the epileptogenic zone. The following seizure types are distinguished (> Table 152‐1).
. Table 152‐1 Epileptic Seizures Auras Somatosensory aura Visual aura Auditory aura Gustatory aura Olfactory aura Psychic aura Autonomic aura Abdominal aura Autonomic seizures Dialeptic seizure Motor seizure Simple motor seizure Myoclonic Seizures Clonic Seizures Tonic Seizures Tonic-clonic Seizures Myoclonic Seizures Versive Seizures Epileptic Spasm Complex motor seizure Automotor seizure Hypermotor seizure Gelastic seizure Special epileptic seizure Atonic Seizures Akinetic Seizures Astatic Seizures Negative myoclonic seizures Hypomotor seizures Aphasic seizures
Classification of epileptic seizures and epilepsies
Auras Auras are defined as subjective symptoms that are not associated with any objective signs. Autonomic symptoms produced by epileptic discharges are also included as ‘‘auras’’ when the autonomic manifestations have not documented objectively. The expression aura should be used only if there is sufficient additional information (usually specific EEG abnormalities) to suggest that the manifestations are of epileptic nature. The existence of an aura reliably points to a focal seizure onset. Somatosensory aura. This aura is characterized by paresthesias with a clearly somatotopic distribution. The patients describe the paresthesias usually as a ‘‘numbness’’ or ‘‘vibration’’ but occasionally use nonspecific terms like an ‘‘unusual sensation.’’ Pain as an expression of a somatosensory aura is extremely rare [9]. Somatosensory auras, by definition, are always localized to a clearly defined somatosensory area. ‘‘Whole body sensation’’ or ‘‘sensations’’ that cannot be localized are classified as non-specific auras. Unilateral paresthesias, limited to a portion of the hand, face, arm, trunk, leg, or foot, are usually the expression of epileptic activation of the contralateral somatosensory region (Brodmann’s areas 1, 2, and 3) [10]. Epileptic activation of the supplementary sensorimotor region or the secondary sensory region frequently produces more extensive paresthesias, which can have a bilateral distribution [11]. Visual aura. Visual auras usually consist of simple phosphenes and may be described as ‘‘bright spots’’ of either white or colored light or, less frequently, ‘‘dark spots.’’ These phosphenes tend to blink and to ‘‘move around’’ but may also be steady. The visual impression frequently is restricted to a portion of the visual field. Visual auras are usually an expression of an ictal epileptic discharge in Brodmann’s area 17 or 18. Activation of the occipital and temporal association cortex is associated with more complex
152
hallucinations like seeing objects, animals, or human beings. These hallucinations frequently appear distorted. Visual illusions may also occur. Visual hallucinations and illusions frequently are associated with other illusions or hallucinations (smell, auditory, etc). These more complex auras are classified as psychic (‘‘experiential’’) auras. Auditory aura. Auditory auras consist of auditory hallucinations (hearing noises) and reflect epileptic activation of Heschl’s gyrus. More complex auditory hallucinations, like hearing voices or melodies, are extremely rare and point to an epileptic activation of the temporal association cortex. Frequently, these more complex auditory auras are associated with other hallucinations or illusions which are then classified as psychic auras. Gustatory aura. Gustatory auras consist of simple gustatory hallucinations or illusions. Usually the sensation is unpleasant. Most probably, they are an expression of epileptic activation of the insula. Olfactory aura. Olfactory auras consist of a smell, usually unpleasant, and are relatively rare. They usually are an expression of epileptic activation of the amygdala and less frequently of the orbitofrontal rectal gyrus. Psychic aura. This type of aura involves unusual sensations in which patients perceive the external or ‘‘internal’’ environment in a distorted manner. It includes the typical and frequent sensation of de´ja` vu and jamais vu. These experiences consist of a ‘‘feeling’’ that an object, environment or individual has never been seen before or is extremely familiar even when the patient knows that these feelings do not correspond to reality. Illusions, like the feeling that an object appears far away or smaller than it actually is, should be included here. Emotional experiences such as fear are also experiential auras of another type. Very frequently these auras consist in ‘‘complex experiences’’ in which a patient has illusions and hallucinations of different types (visual, auditory, olfactory, etc) occurring simultaneously.
2563
2564
152
Classification of epileptic seizures and epilepsies
Available data suggest that psychic auras are a reflection of epileptic activation of the temporal association cortex. Autonomic aura. Autonomic changes like tachycardia, respiratory changes, or sweating are included as auras because, unless special recordings are performed, it is usually impossible to document the autonomic change objectively; we only become aware of the autonomic change when the patient reports the symptoms experienced in relation to such a change. Tachycardia is observed with most epileptic seizures shortly after seizure onset and can be used effectively for automatic detection of epileptic seizures. However, often these tachycardias are caused by the initial seizure symptomatology, as opposed to being induced by the epileptic discharge itself (autonomic aura). The typical example is a patient who experiences an aura consisting of fear and reacts to it with a marked tachycardia, pupillary dilatation, and sweating. Similar reactions can be seen in patients who suffer other auras or focal motor seizures but are afraid of the secondary generalization that may follow such seizure symptomatology. These autonomic reactions should not be classified as autonomic auras or autonomic seizures (see below). Abdominal aura. This frequent aura type is described as a ‘‘funny feeling’’ in the stomach, which is usually unpleasant and tends to rise to the throat and head. It is set apart in a special category because of its frequent occurrence in patients with mesial temporal sclerosis. It is most probably produced by epileptic activation of the insular cortex and is frequently associated with autonomic symptoms like nausea. Occasionally actual ictal vomiting may occur.
Autonomic Seizures Autonomic seizures are seizures in which the main symptomatology is autonomic. The typical example is the patient who reports an aura of
palpitations and EEG and electrocardiographic (ECG) monitoring documents a tachycardia. There is evidence that epileptic activation of the frontal basal region and also of the anterior cingulate area may produce autonomic symptoms in the absence of any other aura or motor symptomatology. Autonomic symptomatology that is purely subjective without objective documentation (e.g., palpitations not documented as taquicardia by an ECG recording) is classified as autonomic auras. However, if the autonomic symptomatology is documented objectively (e.g., palpitations associated with taquicardia in the ECG) then it is classified as an autonomic seizure. Occasionally, an EEG seizure discharge (usually observed with invasive electrodes) may be associated with an ECG abnormality (e.g. taquicardia) even in the absence of any subjective manifestations. These episodes are also classified as autonomic seizures.
Dialeptic Seizure Dialeptic seizures consist of episodes of loss of consciousness during which the patient has limited responsiveness to external stimuli and for which the patient is partially or totally amnesic. These episodes are not associated with any significant loss of motor tone. Motor activity is usually reduced to a minimum, although occasionally some myoclonic jerking of the eyelids or poorly defined automatisms may occur. Patients may, however, continue doing the same activity they were doing before the seizure started, but invariably at a lower speed and with loss of motor skills. Cases in which the loss of muscle tone or the positive motor activity during the seizure overshadowing the loss of consciousness should not be classified as dialeptic seizures but as automotor, hypermotor, or one of the other types of motor seizures listed below (> Table 152‐1).
Classification of epileptic seizures and epilepsies
Motor Seizures Simple Motor Seizures Simple motor seizures consists of motor seizures with movements that are relatively simple, like myoclonic jerks, clonic or tonic movements. These movements do not imitate normal movements or postures because of their unnatural speed or temporal sequence or because, as the consequence of the movement, the patient adopts an unnatural posture. All these movements can be elicited by electrical stimulation either of the primary motor region or the supplementary sensorimotor area. These movements differ from most natural, voluntary movements because they are significantly more ‘‘complex’’, involving a large number of articulations in a ‘‘smooth’’ sequence. Note that in this definition simple and complex has absolutely nothing to do with ictal alteration of consciousness. It specifically addresses the ‘‘complexity’’ of the ictal movement. Myoclonic Seizures. These seizures consist of isolated, rapid jerks (usually of duration of less then 200 msec) that may have a generalized or focal distribution. They are produced by isolated epileptiform discharges that activate any of the cortical motor areas. The epileptiform discharges frequently consist of poly-spikes which, due to the temporal facilitation, are more likely to activate the motor cortex. The pathogenesis of myoclonic seizures is probably very similar to that of clonic seizures which essentially consist of repetitive myoclonic jerks. Clonic Seizures. Clonic seizures consist of intermittent, short contractions of variable groups of muscles that tend to recur at regular intervals. Focal clonic seizures most frequently affect the distal segments of the limbs, the face, and the tongue. They are usually an expression of epileptic activation of the primary motor region(Brodmann’s areas 4 and 6). Electrical stimulation of the supplementary sensorimotor area may also, though less frequently, elicit distal clonic movements.
152
However, it is not yet established whether epileptic activation of the supplementary sensorimotor cortex may result in focal clonic seizures. Generalized clonic seizures are usually associated with generalized epileptiform discharges. The cortical discharge usually has a one-to-one relationship to the muscle twitch. It is generally assumed that generalized clonic seizures are the result of an intermittent ‘‘generalized activation of the motor cortex’’ by the epileptic discharge. It is, however, not yet clear which region of area 4 or 6 gives rise to the type of movement commonly seen with generalized clonic seizures. Tonic Seizures. This seizure type is characterized by sustained involuntary contraction of one or more muscle groups, resulting in posturing. Tonic seizures due to focal brain activation are usually the result of epileptic activation of the supplementary sensorimotor area. These tonic seizures of focal brain origin affect mainly proximal limb muscles bilaterally, but usually in an asymmetrical fashion. The contralateral side tends to be more prominently involved, but ipsilateral involvement is not infrequent. Consciousness is consistently preserved at the initial phase of the seizure. Generalized bilateral symmetrical and synchronous tonic seizures are usually associated with generalized epileptiform discharges. The cortical structure whose activation may lead to generalized tonic seizures has not yet been elucidated. They are probably generated by activation of the cortical motor areas (Brodmann’s areas 4 and 6), which, in turn, excite brain-stem motor-activating centers. Tonic-clonic Seizures. This expression is used only for generalized tonic-clonic seizures. The expression refers to the typical ‘‘Grand Mal’’ seizures. These seizures have a typical temporal evolution. The initial event is a generalized tonic seizure (frequently associated with an ‘‘epileptic cry’’). The position of the extremities during the tonic phase varies but most frequently the arms and legs are adducted and extended at the elbows
2565
2566
152
Classification of epileptic seizures and epilepsies
and knees. The feet tend to be in dorsal extension whereas the wrist tend to in pronation and volar flexion. This initial tonic phase is slowly replaced by a ‘‘generalized tremor-like movement’’ which eventually gives rise to progressively slowing generalized clonic movements that affect primarily the elbows (rapid flexion followed by a slower relaxation) and the hip joins (again with rapid flexion followed by a slower relaxation). The clonic movements are usually relatively synchronous on both sides and in the upper and lower extremities. Consciousness is always lost at the beginning of the seizure and the patient is completely amnestic of the events occurring during the seizures. Postictally, the patient is in deep coma and it takes several minutes before he recovers consciousness. Versive seizure. These seizures are characterized by conjugate lateral deviation of the eyes. Usually, when the eyes reach the extreme lateral position, a deviation of the head and eventually of the trunk may occur. The lateral deviation of the eyes may be smooth (tonic) or saccadic. Lateral deviation of the head or trunk without preceding eye deviation is infrequent. Almost invariably these seizures are produced by epileptic activation of the frontal eye field (contralateral to the side to which the eyes deviate). Epileptic spasms. These seizures are seen most frequently in the first 2 years of age. They consist of a sudden flexion of the truncal muscles with abduction of the arms at the shoulders leading to a ‘‘salaam’’ position. Occasionally the patients relax immediately after such a myoclonic jerk. However, more frequently they tend to remain in the salaam position for 2–10 s. These spasms frequently occur in clusters with one seizure occurring every 5–20 s for 5–10 min.
Complex Motor Seizures The main feature of complex motor seizures are ‘‘complex’’ movements which resemble normal
movements but are involuntary and inappropriate for the situation. These complex movements are usually repetitive (‘‘automatisms’’). Automotor seizure. Automotor seizures are characterized by the occurrence of distal automatisms. Automatisms in this classification are defined as well-organized movements that are involuntary and are inappropriate to the environmental situation. Frequently these involuntary movements are repetitive. Typical examples of distal automatisms include oroalimentary automatisms such as chewing, lip smacking, or swallowing and manual automatisms like fumbling. The automatisms are usually associated with loss or alteration of consciousness, but there are well-documented exceptions [12]. Automotor seizures are distinguished from hypermotor seizures by the type of associated automatisms. In automotor seizures, the automatisms affect primarily the distal portions of the extremities and the lips or tongue. In hypermotor seizures (see below), the proximal segments of the limbs and trunk are involved in the automatisms, giving rise to more extensive movements such as body rocking, pedaling, or running automatisms. There is no clear agreement on which brain structure gives rise to automotor seizures. There are isolated reports that stimulation of the anterior cingulate will elicit ‘‘distal’’ automatisms [13]. It is possible, therefore, that the automatisms are an expression of seizure spread within the limbic system with activation of the anterior cingulate gyrus. Automotor seizures are most frequently seen in patients with temporal lobe epilepsy and less frequently in patients with frontal lobe epilepsy, particularly patients with lesions in the fronto-orbital region. In addition, automotor seizures may be produced by spread of the seizure discharge into the temporal region from the parietal or occipital lobe (> Figure 152-1). It is also interesting to notice that patients who have automotor seizures without loss of consciousness almost invariably suffer from epilepsy in the nondominant
Classification of epileptic seizures and epilepsies
temporal lobe [14]. This indicates that the alteration of consciousness is due to mechanisms other than the automatisms. Hypermotor seizure. Hypermotor seizures are characterized by automatisms affecting mainly the proximal portions of the limbs and trunk. The distinction from automotor seizures is not necessarily clear-cut because it is occasionally difficult to tell whether the automatisms affect predominantly proximal or distal limbs. As in automotor seizures, consciousness may be well preserved during the seizure but this feature has no known lateralizing significance. Hypermotor seizures are classified as a type separate from automotor seizures because there is sufficient evidence to suggest that hypermotor seizures are more frequently associated with frontal lobe epilepsy, whereas automotor seizures are seen most frequently with temporal lobe epilepsy. Gelastic Seizures. Gelastic seizures consist of ictal events during which the patient is laughing. Usually, the laughing is somewhat unusual and lacks mirth. This seizure type is distinguished as a special category because in close to 50% of the cases it is associated with a hypothalamic hamartoma. There is evidence that in those cases the symptomatogenic zone for the gelastic seizure is the hypothalamic hamartoma.
Special Seizures Atonic seizures. Atonic seizures are characterized by a sudden loss of muscle tone that leads to ‘‘drop attacks’’ or ‘‘head drops.’’ This seizure type is frequently associated with generalized epileptiform discharges. The pathogenesis is still unclear. Usually patients presenting with this seizure type also have tonic seizures, suggesting that atonic and tonic seizures may share a similar pathogenic mechanism. Frequently atonic seizures are preceded by a brief myoclonic jerk, which propulses or
152
retropulses the patient, causing falls with injuries rather than just ‘‘slumping down.’’ Akinetic seizures. This seizure type consists of the inability to perform voluntary movements. This inability to move may affect only selected muscle groups and lead to inability to move one hand or to speak, but it may also have a generalized distribution. Akinetic seizures, by definition, are not associated with alteration of consciousness. A typical example is a patient who, during a seizure, experiences a generalized inability to move. Focal akinetic seizures are probably due to epileptic activation of the negative motor areas [15,16]. However, a generalized inability to move may also be due to a focal epileptic activation of the negative motor area. A similar generalized inability to move can be elicited by focal electrical stimulation of the negative motor area. Astatic seizures. This expression is used in patients in whom the seizure produces ‘‘falls’’ but the exact mechanism of the fall is unknown. In other words, it is unknown if the ‘‘falls’’ (‘‘astatic seizures’’) are produced by a generalized muscle atonia, generalized proximal myoclonic jerks, or some other mechanisms. Negative myoclonic seizures. These seizures consist in a short, negative myoclonus. These seizures only become evident when the patient is performing a muscle contraction which is then interrupted by a short (less then 200 msec) disappearance of muscle contraction. The mechanisms involved in the generation of this seizure type are still unclear, but it is likely that the ‘‘silent period’’ elicited by cortical activation is responsible for the negative myoclonus seen in patients in whom the negative myoclonus is preceded by a focal cortical spike in the EEG [15]. Aphasic Seizures. These seizures are characterized by aphasia elicited by epileptic inactivation of a language center. The symptomatogenic zones for aphasic seizures include Broca’s area, Wernecke’s area, and the basal temporal language area.
2567
2568
152
Classification of epileptic seizures and epilepsies
Somatotopic Localization of Ictal Signs and Symptoms The clinical manifestation of an epileptic seizure may contain valuable lateralizing or somatotopic information. This information should be included in the seizure classification. The somatotopic information included with the seizure classification always refers to the ‘‘peripheral’’ expression of the seizure (i.e., not to the brain localization or brain lateralization of the epileptogenic zone). In visual and somatosensory auras and in motor seizures, the peripheral seizure manifestation will be included. In this classification system, a visual aura in the left visual field will be classified as left visual aura. Accordingly, a somatosensory aura in the left hand would be classified as left-hand somatosensory aura. The modifiers generalized, left, left face, left hand, left arm, left foot, left leg, right, right face, right hand, right arm, right foot, right leg, and bilateral asymmetric can be used with all the motor seizures except the versive seizures. For versive seizures, only the modifiers left or right should be used, indicating the side toward which the eyes and head move. The classification specifies the first localizing or lateralizing sign or symptom when the seizure has a spreading somatotopic evolution. For example:
Generalized clonic seizure refers to a seizure with bilateral synchronous clonic jerks. Left clonic seizure refers to a seizure with leftsided clonic jerks. Left visual aura refers to an aura with visual hallucinations or illusions in the left visual field.
Besides, seizures may include several ictal or postictal lateralizing signs [17]. A dystonia of one hand in the course of an automotor seizure points to a contralateral seizure origin [18]. Other lateralizing signs include ictal speech
[19], postictal aphasia, postictal Todd’s paralysis, the sign-of-four, and automotor seizures with preserved consciousness. These lateralizing signs should be listed after the seizure classification. Lateralizing signs that are already part of the seizure classification (e.g., right versive seizure [20]) should not be listed again as a lateralizing sign.
Seizure Evolution Epileptic seizures frequently evolve from one seizure type to another. In this seizure classification system, the seizure evolution is specified by linking different seizures by arrows. For example: Abdominal aura ! automotor seizure Right visual aura ! right clonic seizure Sequential motor seizures of different types should also be linked by arrows. For example, a poorly defined tonic seizure with bilateral, asynchronous involvement followed by a clonic seizure of the right hand should be classified as follows: Bilateral asymmetric tonic seizure ! right hand clonic seizure However, as mentioned above, if the same type of seizure shows a sequential change in somatotopic distribution, only the initial somatotopic distribution is specified. For example, a clonic seizure starting in the left hand and then evolving into a left face clonic seizure and eventually into a generalized tonic-clonic seizure would be classified as follows: Left hand clonic seizure ! generalizedrm tonic - clonic seizure In other words, the detailed ‘‘jacksonian march’’ is not specified to avoid excessive complexity. Also to avoid complexity, the frequently observed evolution from a motionless stare (‘‘dialeptic seizure’’) to an automotor seizure or hypermotor
Classification of epileptic seizures and epilepsies
seizure should be classified just as an automotor seizure or hypermotor seizure. In automotor seizures, we frequently may observe a versive eye and head deviation that reliably points to a contralateral seizure origin [20]. This seizure evolution will be classified as follows: Automotor seizure ! left ðor rightÞ versive seizure Generalized epileptic seizures may also demonstrate a seizure evolution. A typical example is the evolution from a generalized myoclonic seizure to a generalized tonic-clonic seizure, frequently seen in patients with juvenile myoclonic epilepsy: Generalized myoclonic seizure ! generalized tonic - clonic seizure
Classification of the Epilepsies Epileptic seizures are spells characterized by a constellation of symptoms and/or signs that are the consequence of an epileptic discharge activating the cortex and/or subcortical structures. A wide variety of signs and symptoms may occur in epileptic seizures (> Table 152‐1). As we discussed, it has advantages to define the epileptic seizure exclusively by the clinical signs and/or symptoms observed during the seizure [5]. Otherwise it is unclear what clinical information was used to classify the seizure, making it difficult to decide whether there is convergence of the diagnostic results (convergence means that the different test results point to the same anatomic region). This is especially critical in assessing the results of presurgical epilepsy testing. Terminology in which an epileptic seizure is identified by an anatomic region (like temporal lobe seizures) should not be used, because in these cases it is unclear whether the anatomic modifier defines the epileptogenic or symptomatogenic zone. For example, when epileptologists
152
refer to temporal lobe seizures, they usually refer to the epileptogenic zone (even if the symptomatogenic zone almost invariably is located in the extratemporal region). However, when epileptologists talk about supplementary sensorimotor area seizures, they refer to the symptomatogenic zone (even if the epileptogenic zone almost always tends to be outside the supplementary sensorimotor area) [21]. However, in defining the epilepsies, it is appropriate to define an anatomic region that defines the epileptogenic zone (area of cortex capable of generating seizures and whose surgical resection of which results in seizure freedom). To define the epileptogenic zone, all the available information (clinical data, ictal and interictal EEG, functional and anatomical imaging) are taken into account. Generalized epilepsies are those in which the epileptogenic zone involves a significant portion of the cerebral cortex of both hemispheres. This does not necessarily mean that a given seizure may not originate from a relatively localized cortical region (‘‘secondary generalization’’). It implies that seizures could also potentially originate from many other cortical areas on both hemispheres. From a practical point of view, there is a diffusely abnormal cortex, and resection of only a region of cortex would not eliminate the seizures, since other areas even on the opposite hemisphere would take over as ‘‘pacemakers.’’ Focal epilepsies are those in which the epileptogenic zone is limited to a restricted portion of the cortex. By definition, resection of these limited epileptogenic zones renders the patient seizurefree. Seizure spread occurs slowly and may remain restricted to a limited region of cortex for an indefinite time. The ictal onset zone is usually imbedded in a regional irritative zone, which is capable of generating interictal epileptiform discharges [7]. The tendency of restricted seizure spread is probably related to the fact that the adjacent cortex is normal and mechanisms to restrict the seizure spread are well preserved.
2569
2570
152
Classification of epileptic seizures and epilepsies
5-Tier Classification of the Epilepsies Epileptic seizures should be understood as a neurological symptom that occurs with a large variety of insults of the cerebral cortex. The ictal clinical manifestation will be primarily an expression of the cortical region from which the epileptic seizures originate from (‘‘epileptogenic zone). The management of patients with epileptic seizures will depend primarily on the etiology that produces the epileptic seizures (progressive vs. static, regionally restricted vs. diffuse, etc) and the clinical semiology of the seizures (generalized vs. focal seizures; myoclonic vs. dialeptic vs. clonic seizures, etc) which greatly influences the antiepileptics that are most effective in the medical treatment of the seizures. These considerations led us to a classification system of the epilepsies which for each patient defines the following 5 tiers [22]:
Epileptogenic zone Semiological Seizure Type Etiology Seizure Frequency Related Medical Condition(s)
In the previous paragraphs we outlined the semiological classification of the epileptic seizures. Below we will summarize the other 4 tiers of this epilepsy classification.
Tier #1: Epilepsy This tier lists primarily the epileptogenic zone. All the available clinical and test information is used to define the exact location and extend of the epileptogenic zone (clinical history, neurological examination, interictal and ictal EEG, MRI, PET, ictal SPECT, neuropsychological testing, etc). Therefore, the precision with which we can define
this tier will depend on the extend of the work-up. In some patients the clinician may be unable to define with any degree of certainty the location or extend of the epileptogenic zone. These patients may just be classified as ‘‘epilepsy’’. In other patients, the clinician might be able to conclude that the patient has focal epilepsy but he may be unable to further localize the process. At the other extreme, in some patients who had extensive testing, the clinician might define the location and extend of the epileptogenic zone with extreme precision. > Table 152‐2 describes the classification of the epileptogenic zones. Since the end of the nineteenth century, epileptologists have recognized that patients with epilepsy can be grouped (classified) according to certain constellations of symptoms and signs. These constellations are called ‘‘epileptic syndromes’’ and frequently carry a prognostic implication and also guide the physician in the choice of antiepileptic medication. Epilepsies arising from different cortical regions tend to show seizures of similar semiology and, therefore, are always classified by the region of brain from which the seizures originate (e.g., temporal lobe epilepsy, frontal lobe epilepsy, etc). Therefore, these epilepsies for a long time have already been classified by the location of the epileptogenic zone. On the other hand, generalized epilepsies may be characterized by different seizure types depending on the age of the patient and/or the cause of the epilepsy. For example, genetic epilepsies tend to present with epileptic seizures that are very different to symptomatic epilepsies. There are certain seizures that are never seen in children of less than 3 years of age (example, generalized tonic-clonic seizures) or in adult patients (typical epileptic spasms). This variety has led to the description of different ‘‘epileptic syndromes’’ in patients with generalized epilepsies. In most cases, the epileptic syndrome can be easily deduced when listing the 5-tier epilepsy classification discussed in this chapter. However,
Classification of epileptic seizures and epilepsies
. Table 152‐2 Classification of the Epileptogenic Zone Focal epileptogenic zone located within one cortical lobe Frontal Peri-rolandic Temporal Neocortical temporal Mesial Temporal Parietal Occipital Other Multilobar: epileptogenic zone affects more than one brain lobe Bilobar homotopic: epileptogeniz zone affects two homotopic brain lobes (bitemporal, bifrontal, etc) Other Generalized: epileptogenic zone is bilateral, diffusely distributed affecting most or all of the brain cortex
in some instances the direct listing of the ‘‘epilepsy syndrome’’ (which essentially is a shorthand for a constellation of clinical data) can be helpful. It is the reason, that we allow listing the epilepsy syndrome after describing the epileptogenic zone in the first tier of this classification (see examples below).
Tier # 2: Semiological Seizure Classification See detailed description above.
152
An example that illustrates this tendency very well, is the increased incidence of seizures in patients with brain tumors who also have a family history of epilepsy (‘‘genetic factor’’). This tendency to identify multiple etiologies for all patients with epilepsy will certainly increase in the near future, particularly with a future increased availability of genetic testing. In > Table 152‐3 the etiological classification is listed. Precise classification of the etiology (etiologies) that produces the epileptic seizures is essential for appropriate management of the patient. In many instances, removal of the etiological factor is as important as or even more important than the control of the seizures (example, in a patient with a progressive brain tumor, can also leading to epileptic seizures).
Tier # 4: Seizure Frequency Seizure frequency is an essential variable because it determines the severity of the epileptic condition. Management of the epileptic seizures consist primarily in a reduction and hopefully complete elimination of the epileptic seizures. In the 5-tier epilepsy classification we can just classify the seizure frequency in one of the 4 categories listed in > Table 152‐4 or actually list the precise number of seizures the patient had since the last visit.
Tier #3: Etiology Recent advances of diagnostic tests have established that all patients with epileptic seizures tend to have several etiologies for the seizure tendency. Each of these etiological factors contributes, by different degrees, to the generation of epileptic seizures. In some cases, one or two factors may be clearly predominant, in other cases the epilepsy indeed is multifactorial.
Tier #5: Related Medical Conditions In this tier we list additional information (free text) about major nonetiological medical conditions the patient is suffering form. This information may be useful in the management of the seizures. Examples: previous brain surgery for epilepsy, mental retardation, depression, etc.
2571
2572
152
Classification of epileptic seizures and epilepsies
. Table 152‐3 Etiology Hippocampal sclerosis Tumor Glioma Dysembrioplastic neuroepithelial tumor Ganglioglioma Other Malformation of cortical development (MCD) Focal MCD Hemimegalencephaly MCD with epidermal nevi Heterotopic grey matter Hypothalamic hamartoma Hypomelanosis of Ito Other Malformations of vascular development Cavernous angioma Arteriovenous malformation Sturge-Weber syndrome Other Central Nervous System Infections Meningitis Encephalitis Abscess Other Central Nervous System Inflammation Rasmussen encephalitis Vaculitis Other Hypoxic-Ischemic brain injury Focal ischemic infarction Diffuse hypoxic-ischemic injury Periventricular leukomalacia Hemorrhagic infarction Venous sinus thrombosis Other Head Trauma Head trauma with intracranial hemorrhage Penetrating injury Closed head injury Other Inheritable conditions Presumed genetic cause Tuberous sclerosis Progressive myoclonic epilepsy Metabolic syndrome Channelopathy Mitochondrila disorder Chromosomal aberration Other Structural brain abnormality of unknown cause Other Unknown Etiology
. Table 152‐4 Seizure Frequency Daily Seizures (one or more seizures per day) Persistent seizures (less than one seizure per day but at least one seizure/6 months) Rare or no seizures (Fewer than one seizure every 6 months Undefined (Seizure frequency can not be specified because of unknown seizure frequency, recent onset of seizures, or recent epilepsy surgery)
Examples of Epilepsy Classifications Right Mesial Temporal Lobe Epilepsy A patient with abdominal auras evolving into automotor seizures with a left-hand dystonia and eventually secondary generalized tonic-clonic seizures in which the MRI demonstrated a right mesial sclerosis will be classified as follows: Seizures: Abdominal aura ! automotor Seizure (LOA) ! generalized tonic-clonic seizure Lateralizing signs: Left hand dystonia Etiology: Mesial temporal sclerosis Seizure Frequency: Persistent Related medical Conditions: None
Left Frontal Lobe Epilepsy A patient with an unspecific aura that evolves either into a hypermotor or into a bilateral tonic seizure with preserved consciousness with a few clonic jerks in the right hand at the end of the seizure in whom MRI revealed a left frontal lesion suggestive of a low-grade astrocytoma would be classified as follows: Seizures: Aura ! hypermotor seizure Aura ! bilateral asymmetric tonic seizure ! right hand clonic seizure
Classification of epileptic seizures and epilepsies
152
Etiology: Tumor (left frontopolar low-grade astrocytoma?) Seizure frequency: daily Realted medical condition: right hand weakness
movements of the limbs either bilaterally or unilaterally involving the left or right side of the body. There were also episodes of blank staring and unresponsiveness Seizures:
Generalized Epilepsy (Juvenile Myoclonic Epilepsy)
1. 2. 3.
A patient had myoclonic seizures since the age of 14 years. His seizures were precipitated by sleep deprivation, and there was a positive family history of generalized tonic-clonic seizures. After a crescendo pattern, this patient’s myoclonic seizures culminate in generalized tonic-clonic seizures. The interictal EEG showed generalized polyspikes. Seizures: Generalized myoclonic seizure ! generalized tonic-clonic seizure Etiology: Presumed genetic cause Seizure Frequency: persistent Related Medical Condition: none
Left Fronto-Temporal Epilepsy A patient had seizures associated with unresponsiveness preceded by an unspecific sensation in the head. This patient’s interictal EEG demonstrated spikes in the left frontal and left temporal region. However, the neurological exam and neuroimaging were normal. Seizures: Aura ! Dialeptic Seizure Etiology: Unknown Seizure Frequency: Persistent Related medical Condition: None
Generalized Epilepsy A neurologically normal child had had seizures since infancy consisting of irregular clonic
Dialeptic seizures Generalized clonic seizures Left and right clonic seizures
Etiology: Unknown Seizure Frequency: Rare Related medical Conditions: retardation
Mild
mental
References 1. Temkin O. The Falling Sickness: A History of Epilepsy from the Greeks to the Beginning of Modern Neurology. 2d ed. Baltimore: Johns Hopkins Press; 1971. 2. Gastaut H. Clinical and electroencephalographical classification of epileptic seizures. Epilepsia 1970;11: 102-113. 3. Commission on Classification and Terminology of the International League against Epilepsy: Proposal for revised clinical and electroen-cephalographic classification of epileptic seizures. Epilepsia 1981;22:489-501. 4. Gloor P. Consciousness as a neurological concept in epileptology: A critical review. Epilepsia 1986;27:S14-S26. 5. Lu¨ders HO, Burgess RC, Noachtar S. Expanding the International Classification of Seizures to provide localization information. Neurology 1993;43:1650-1655. 6. Luders HO, Noachtar S. Alias und Videoband epileptischer Anfa¨lle und Syndrome. Wehr/Baden: Ciba Geigy Verlag; 1995. 7. Luders HO, Awad I. Conceptual considerations. HO Lu¨ders Epilepsy Surgery. New York: Raven Press; 1991. p. 51-62. 8. Lu¨ders HO, Noachtar S. Atlas und Klassifikation der Elektroenzephalographie. Wehr/Baden: Ciba Geigy Verlag; 1994. 9. Mauguiere F, Courjon J. Somatosensory epilepsy. Brain 1978;101:307-332. 10. Penfield W, Jasper H. Epilepsy and the Functional Anatomy of the Human Brain. Boston: Little Brown; 1954. 11. Kramer RE, Lu¨ders H, Goldstick LP, et al. Ictus emeticus: An electro-clinical analysis. Neurology 1988;38:1048-1052. 12. Noachtar S, Ebner A, Dinner DS. Das Auftreten von Automatismen bei erhaltenem Bewußtsein: Zur Frage
2573
2574
152 13.
14.
15.
16.
Classification of epileptic seizures and epilepsies
der Bewußtseinssto¨rung bei komplex-fokalen Anfa¨llen. Scheffner D, Epilepsie 91. Reinbek: Einhorn-Presse Verlag; 1992. p. 82-87. Talairach J, Bancaud J, Geier S, et al. The cingulate gyrus and the hu- man behaviour. Electroencephalogr Clin Neurophysiol 1973;34:45-52. Ebner A, Dinner DS, Noachtar S, Luders HO. Automatisms with preserved responsiveness (APR): A new lateralizing sign in psychomotor seizures. Neurology 1995;45:61-64. Noachtar S, Lu¨ders HO, Holthausen H, et al. Surface and EEG-video recordings of negative motor phenomena in epileptic seizures (abstr). Epilepsia 1994;35 (suppl. 8):123. Lu¨ders HO, Lesser RP, Morris HH, et al. Negative motor responses elicited by stimulation of the human cortex. Wolf P, Dam M, Janz D, editors. Advances in Epileptology. New York: Raven Press; 1987. p. 229-231.
17. Chee MWL, Kotagal P, Van Ness PC, et al. Lateralizing signs in intractable partial epilepsy: Blinded multiobserver analysis. Neurology 1993;43:2519-2525. 18. Kotagal P, Lu¨ders H, Mosssrris HH, et al. Dystonie posturing in complex partial seizures of temporal lobe onset: A new lateralizing sign. MWogv 1989;39:196-201. 19. Gabr M, Lu¨ders H, Dinner DS, et al. Speech manifestation in lateralization of temporal lobe seizures. Ann Neurol 1989;25:82-87. 20. Wyllie E, Lu¨ders H, Morris HH, et al. The significance of head and eye movement during epileptic seizures. Neurology 1986;36:606-611. 21. Lu¨ders HO, Supplementary sensorimotor area. Adv Neurol. Lippincott-Raven, Philadelphia, New York; vol 70. 1996. 22. Loddenkemper T, Kellinghaus C, Wyllie E, et al. A proposal for a five- dimensional patient-oriented epilepsy classification. Epileptic Disorders 2005;7:308-316.
162 Corpus Callosotomy R. E. Maxwell
Rationale The rationale for sectioning the corpus callosum to control generalized seizures rests on the tenets that seizures evolve in the cerebral cortex, secondarily spread through commissural pathways to the opposite cerebral hemisphere, and that therefore such generalization or bilateral synchronization can be reduced or eliminated by sectioning this main commissural bundle [1]. The corpus callosum, hippocampal commissure, and anterior commissure transmit interhemispheric discharges. The corpus callosum is by far the largest of the three, developing in proportion to the size and complexity of the neocortex and reaching maximal size in humans, where it contains somewhat on the order of 180 million axons [2]. The number of callosal connections varies from one cortical region to another. Definitive data on the topography of commissural fibers in humans are not available, though the issue has been investigated in the rhesus monkey by Pandya and others using degeneration and autoradiographic techniques. The vast majority of corpus callosum fibers connect homotypic regions of the two hemispheres, but heterotypic connections occur. Motor and association cortex mediating the head, trunk, and proximal limbs has more robust connections through the callosum than do the distal appendages. This may partially explain why secondarily generalized seizures with rapid bisynchronization often result in sudden drop attacks and traumatic falls. Pandya and Rosene have studied the topography of the interhemispheric connections using dark-field microscopy to trace the transport of radioactive amino acids along callosal fibers [3]. #
Springer-Verlag Berlin/Heidelberg 2009
There are relatively few callosal connections between the anterior and inferior temporal regions, but those fibers originating in the temporal lobes cross in the posterior part of the body and in the underlying hippocampal commisure. Fiber tracts originating in the frontal cingulate regions traverse the anterior half of the corpus callosum, including the rostrum, genu, and anterior body. Parietal connections cross to the opposite hemisphere through the posterior body and anterior splenium, and occipital connections occur through the most posterior and inferior portions of the splenium. The general rule of cortical excitability and spread of discharges is that the pattern of bilateral electrical discharge and spread is influenced by the density of callosal connections between corresponding areas of the two hemispheres. Therefore, regions with many callosal connections, such as the premotor and precentral cortices, will show wellsynchronized bilateral discharges more readily than those with fewer callosal connections, such as the posterior temporal and striate occipital cortices, which tend to show relatively independent discharges.
Experimental Rationale Electrical stimulation of the primate cortex evokes a biphasic wave in the contralateral cortex, with the largest potentials seen in homotopic areas, but heterotopic responses can also be recorded. These responses are obtained by stimulating almost any cortical region but are abolished by sectioning the corpus callosum [4]. It has clearly been shown that the corpus callosum
2724
162
Corpus callosotomy
transmits excitatory potentials between the cerebral hemispheres, but inhibitory effects are even more pronounced under some experimental conditions. Asanuma and Okuda found that transcallosal stimulation in cats was inhibitory of contralateral pyramidal tract activity over a much wider area than was the case with excitatory potentials [5]. A sentinel report was that of Erickson in 1940, who studied afterdischarges produced by electrical stimulation of the monkey cerebral cortex before and after sectioning the corpus callosum. He found that afterdischarges initially spread throughout the ipsilateral motor area, producing ipsilateral clonic movements before crossing to the opposite hemisphere when bilateral clonic movements occur. Complete section of the corpus callosum stopped the spread of discharges to the opposite hemisphere and clonic movements remained contralateral to the discharges [6]. Subsequent experimental studies in both feline and primate models have shown that sectioning the corpus callosum can disrupt bilaterally synchronous discharges induced by topical or systemic epileptogenic agents [7–10]. The corpus callosum is by far the most significant of the commissures connecting the cerebral hemispheres. Experimental studies of epilepsy in both feline and primate models have clearly demonstrated the role of the corpus callosum in mediating bilateral synchrony [11]. Complete section of the corpus callosum and contiguous hippocampal commissure produces disruption of bilateral synchrony and the reestablishment of an independent discharge in each hemisphere in acute cats whose seizures were induced with 0.5% strychnine [9]. Musgrave and Gloor demonstrated in the cat that they could abolish penicillin-induced bilaterally synchronous epileptic discharges by completely sectioning the corpus callosum and anterior commissure [10]. Incomplete callosal section did not completely disrupt synchrony. This helped confirm the important role of the corpus callosum in the propagation of 3-Hz
spike- and wave- discharges in animal models of generalized epilepsy. In the kindling model, Wada and his associates showed the anterior two-thirds of the corpus callosum to be important for the occurrence of generalized convulsive seizures after chronic amygdaloid stimulation [12]. Corpus callosotomy also disrupts the synchrony of spike-wave discharges in rodents and in the photosensitive baboon [13,14]. There have been acute and chronic studies using epileptogenic feline and primate models, suggesting increased epileptiform activity after corpus callosum sectioning. Kopeloff et al. found that chronic seizure foci induced by aluminum oxide cream increased in frequency and severity after the corpus callosum was sectioned [15]. Dividing the corpus callosum also increases cats’ susceptibility to seizures following acetylcholine and pentylenetetrazol administration [16]. Wada and Sato also found that division of the corpus callosum facilitated the progression of generalized seizures induced by daily electrical stimulation of the amygdala [17]. Experimental studies suggest alternate pathways to the corpus callosum that may account for the bilateral synchrony occurring with motor seizures. Wada and Sato reported evidence that afterdischarges occurred in the midbrain reticular formation before generalized discharges, and seizures were seen in the amygdala-kindled cat [18]. More recent work studying the cortical reactivity of the rat, monitored by thalamocortical responses, shows enhancement by repetitive transcallosal volleys at 5–20 Hz. This effect is inhibited by callosotomy. The facilitatory effect of the corpus callosum on the cerebral cortex suggests an explanation for interhemispheric epileptogenesis and the suppressive effect of corpus callosotomy on certain types of intractable epilepsy [19]. Seizures induced in rats by topical application of penicillin to a unilateral focus have been found to spread bilaterally by first increasing bilateral activation of intrathalamic pathways
Corpus callosotomy
until transsynaptic stimulation of medial and orbital frontal cortices occurred [20]. Kusske and Rush found that corpus callosotomy accelerated thalamic spread [21]. It has been proposed that the cortical thalamic pathway is perhaps as important or even more important for bilaterally synchronous seizure spread than the pathway through callosal fibers [22]. There is also evidence that the brain stem is important in the mediation of generalized motor seizures. Electrical activity is propagated to the substantia nigra during kindled seizures [17]. Increases in multiple unit activity of the midbrain and pontine reticular formations have been recorded at the onset of pentylenetetrazolinduced tonic seizures [23]. Experimental paradigms therefore suggest that the corpus callosum conducts both excitatory and inhibitory potentials and that although the spread and synchronization of epileptic discharges is mediated by the corpus callosum, there are other propagation pathways that involve the thalamus, substantia nigra, and brain-stem reticular formation.
Clinical Rationale The origin of the operation of corpus callosotomy for treating epilepsy and the articulation of a clinical rationale for the procedure can both be credited to the neurosurgeon William P. Van Wagenen. In 1940 Van Wagenen and Herren defined a rationale for corpus callosotomy based on their clinical observations and reported the clinical study of ten patients with a brief follow-up [24]. They observed that patients with malignant gliomas often had generalized seizures early in the course of the illness, but as the tumor enlarged and involved the corpus callosum, the seizures were often unilateral without loss of consciousness and were fewer in number. A patient with multiple meningiomas was found to have tumor projecting into the corpus callosum. Retrospective analysis of his medical history revealed frequent
162
seizures at the beginning of his symptomatology but fewer as the tumor progressed. Another observation was made at the time of a postmortem examination on a patient who had suffered seizures for 25 years and then experienced relief from the seizures following a cerebral hemorrhage. Examination of the brain revealed that the hemorrhage had destroyed most of the corpus callosum. Therefore, based on these clinical observations and the tenet that consciousness is not usually lost when the spread of the epileptic discharge is limited to one cerebral cortex, they decided to section the corpus callosum in an effort to confine spread of the seizure to one hemisphere. It has been pointed out that although significant experimental work regarding the role of the corpus callosum in epilepsy was being pursued at the time, there is no indication that Van Wagenen and Herren were aware of it, and it is not alluded to or referenced in their article [25]. After the initial ten patients were reported in 1940, Van Wagenen subsequently operated on 14 additional patients, sometimes including the anterior commissure in the commissurotomy. Over the subsequent three decades, several disconnection procedures were tried as variations on Van Wagenen’s original concept. Division of interhemispheric fiber tracts were performed that included various combinations of the corpus callosum, hippocampal commissure, anterior commissure, and fornix. Most of the reported series were small but provided substantial evidence that the frequency and severity of generalized seizures could be unequivocally improved after callostomy or commissurotomy [26–28]. The slow acceptance and reluctant rush to embrace such an otherwise promising technique for managing severe, intractable, generalized epilepsy was primarily due to the comparatively high morbidity and even mortality experienced at some centers attempting procedures where several forebrain commissures and even the massa intermedia were divided. Two factors made commissurotomy safer: the advent of microsurgical techniques as promoted
2725
2726
162
Corpus callosotomy
by Donald Wilson and the realization that division of the corpus callosum (with the hippocampal commissure) was as effective as sectioning multiple commissures for achieving control of secondarily generalized seizures [29,30]. The development of the operating microscope and improvements in presurgical evaluation and surgical techniques encouraged epilepsy centers to reevaluate callosotomy, with more emphasis on the neurophysiological and neuropsychological consequences of the procedure. Magnetic resonance imaging (MRI) has enhanced the preoperative and postoperative assessment of candidates for corpus callosotmy [31–36]. Frameless stereotactic surgery and radiosurgery are recent technological advancements that may further reduce the morbidity of callosotmy [37–41].
Pathophysiology The purpose of epilepsy surgery is of course, where possible, to achieve the cure or at least complete control of all seizures. It must be recognized, however, that there are important secondary reasons to evaluate patients for seizure surgery. One such purpose is to reduce the frequency and severity of particularly disabling and devastating seizures that cause falls and frequent injuries. Another purpose is to improve control of seizures so that the number and dosages of anticonvulsant medications can be reduced in order to limit or eliminate disabling side effects or toxicity. The goal of seizure surgery is to eliminate epileptogenic tissue where possible or confine the spread of seizure discharges so they do not become generalized with alterations of consciousness and muscle tone. Whereas temporal lobectomy for partial complex seizures has the potential, or even the likelihood in well-selected cases, for cure or complete control of epilepsy, commissurotomy can only confine or desynchronize the spread of the epileptic discharges. Corpus
callosotomy is therefore almost always palliative rather than curative and is not an appropriate choice when focal resection of an epileptogenic lesion is feasible. Corpus callosotomy is considered for patients with medically intractable epilepsy where the seizure focus is unapproachable, widespread, or multifocal. The secondarily generalized discharges result in tonic, atonic, tonic-atonic, or tonic-clonic seizures. The epileptic events must be well documented and significantly interfering with the patient’s health and wellbeing. Many of the patients are suffering repeated falls with repeated lacerations and fractures. Medical intractability is documented when intensive and methodical anticonvulsant therapy, verified by adequate drug levels, is proven inadequate to control the seizures and provide the patient with satisfactory home, school, and job adjustments. A variety of seizure types have been reported to benefit from callosotomy, but results are considerably better among patients where clinical and electroencephalographic evidence suggests rapid secondary generalization and bilateral synchronization of the seizure discharges by means of propagation through the corpus callosum. These patients will usually have asymmetrical abnormalities on neurological examination and focal lesions seen on neuroimaging studies [30,42]. Sudden head-drops and falls are associated with an atonic event, with loss of muscle tone in the axial or proximal limb musculature. Close observation often discloses a preceding tonic or myoclonic phase of the seizure, however, in Minnesota, only patients suffering frequent and disabling falls with repeated injuries have been selected for corpus callosotomy. Several authors have suggested that patients suffering generalized seizures associated with unilateral structural lesions seen by neuroimaging are better candidates for corpus callosotomy than those with no lateralizing features [43–47]. Corpus callosotomy has also been reported effective for stopping
Corpus callosotomy
prolonged, medically intractable, generalized or non-localizable status epilepticus [48]. Mental retardation is rarely a contraindication to corpus callosotomy. The humanitarian benefits of sparing the patient, family, and caregivers from the physical and psychological trauma of experiencing or witnessing repeated head-drops and falls is reason enough in some cases for considering the procedure even in the severely retarded. Furthermore, it is well recognized and documented that many retarded patients demonstrate gains in attentiveness, sociability, and behavior following callosotomy. This is not to say that mental retardation is not a factor in outcome. Clinical investigators who have studied this issue report that severe mental retardation is associated with a poor outcome [1,30,45,46]. Patients with lower IQ’s have less success with seizure reduction consistent with the widespread and multifocal nature of their cerebral pathology. Using the Minnesota criteria for patient selection – which include medical intractability, unresectability, and frequent generalized seizures causing falls and repeated injuries – it is unusual to have a candidate for callosotomy with an IQ greater than 75. The higher-functioning patients tend to have a smoother recovery, shorter convalescence and better seizure outcome after callosotomy. Section of the corpus callosum has been proposed as an alternative to hemispherectomy in patients with infantile hemiplegia [28,49,50]. This recommendation was primarily predicated on the higher immediate and delayed complication rates experienced with hemispherectomy Furthermore, hemispherectomy is not an ideal option in patients with intact visual fields or with lesser degrees of hemiplegia with preserved finger movements and a somewhat functional hand. In intractable cases where a widespread, unilateral epileptogenic process has resulted in an otherwise nonfunctional hemisphere with a useless contralateral hand, hemispherectomy is almost
162
always preferable to callosotomy and is one of the most effective procedures in the epilepsy surgeon’s repertoire. Investigators who have studied seizure control after callosotomy and hemispherectomy in comparable patient groups with lateralized epileptogenic lesions have reported hemispherectomy to be more effective and in some cases curative [51,52]. With the advent of modern surgical techniques and modified approaches to hemispherectomy, the rate of complications from superficial cerebral hemosiderosis with delayed bleeding and late-developing hydrocephalus has been reduced [53–55]. One of the vexing problems for clinical epileptologists is the identification of a seizure and EEG pattern that would better predict those patients likely to benefit from corpus callosotomy. Most patients considered for callosotomy have one of several interictal electroencephalographic (EEG) patterns: multifocal spikes or slow waves with secondary generalization, focal spikes with secondary bilateral synchrony, or generalized focal spikes with secondary bilateral synchrony, or generalized spike- and wave- discharges with or without a normal background. No interictal pattern has been identified that clearly portends a good outcome, but the presence of multifocal, independent interictal spike patterns apparently suggests a poor seizure outcome and an increased risk for developing more frequent and intense partial seizures after callostomy [46,56]. The identification of an ictal EEG pattern with unilateral onset and rapid secondary generalization and bilateral synchronization is critical for patient selection. The most favorable ictal pattern shows epileptiform fast activity at the onset of the seizure [57]. Patients with a generalized slow spike wave pattern, electrodecrement, and non-evolving low amplitude fast activity do well after callosotomy. The ictal EEG with these characteristics is able to identify a group of patients with a high likelihood of achieving total or near total resolution of seizures
2727
2728
162
Corpus callosotomy
accompanied by sudden falls [58]. Much poorer results are seen in patients with other EEG patterns and those who have seizures with a complex partial onset of frontal lobe origin with a strong tonic component and considerable asymmetry in the tonic activity. There is a consensus in the experimental and clinical literature that sectioning the corpus callosum disrupts EEG bisynchrony to some (albeit variable) degree and that this is correlated with a diminished generalization of clinical seizures in carefully and well-selected patients. The categories of patients considered for corpus callosotomy include those with multifocal or unresectable focal generalized epilepsy, progressive epileptic hemiplegic encephalitis (Rasmussen’s syndrome), Forme-Fruste infantile hemiplegia with a functional hand, and the Lennox-Gastaut syndrome. The pathological substrates for seizures amenable to palliation with corpus callosotomy include encephalotrigeminal angiomatosis (Sturge-Weber syndrome), cystic lesions in the middle cerebral artery distribution, and bihemispheric malformations of cortical development (lissencephaly, band heterotopia, perisylvian polymicrogyria, tuberous sclerosis, bihemispheric cortical dysplasias.)[49,59–66]
Technique Preoperative Considerations It is important that the preoperative evaluation clearly define the seizure frequency, type, and severity by video-EEG. The precise nature of the seizure(s) to be treated is documented by ictal video-EEG using safety harnesses when necessary to record and correlate the clinical seizure and the onset and pattern of discharge spread. The preoperative neuropsychological level of function and the existence of preexisting deficits must be recognized and documented. Magnetic resonance imaging (MRI) is done to define the
structural integrity and configuration of the corpus callosum and identify structural lesions and focal or diffuse pathology [34]. All patients considered for corpus callosotomy are classified according to seizure type(s) based upon clinical documentation of multiple ictal events and concomitant EEG recordings. This often requires sleep-deprived as well as routine EEG with multiple 4- to 6-h video-EEG or telemetered sessions. Other adjuncts include sphenoidal leads, hyperventilation, and photic stimulation. Carefully monitored and systematic anticonvulsant drug withdrawal is rarely necessary in this population because of the frequency of the seizures. Pharmacological activation is avoided because of the risk of provoking status epilepticus or nonstereotypical epileptic events. The preoperative evaluation occurs in three phases and may occur over a few weeks up to several years, depending upon the frequency and severity of the seizures and psychosocial issues. Phase 1 is concerned with categorizing the seizure types(s), establishing compliance with the anticonvulsant medical regimen, confirming intractability, and defining social and psychological issues. Phase 2 concerns the acquisition of data that support or contraindicate corpus callosotomy as a therapeutic option. Neuropsychometric studies, MRI, video-EEG and possibly positron emission tomography (PET) or single photon emission tomography (SPECT) studies are usually sufficient to rule out a surgically resectable lesion or seizure focus and define a secondarily generalized seizure pattern amenable to callosal section. The third-phase studies are invasive tests such as the sodium amytal (Wada) test to document language and memory dominance prior to making the final decision to proceed with callosotomy. The MRI is necessary not only for identifying cerebral pathology and congenital defects but for preoperative planning of the surgery. The configuration and dimension of the corpus callosum can be determined and, more importantly,
Corpus callosotomy
the location of cortical bridging veins in the region of the surgical approach determined and the position of the bone flap adjusted in order to avoid the risk of cortical venous infarction. A normal preoperative MRI has been reported to be the best predictor of good outcome in a series of patients undergoing anterior and complete callosotomies for the treatment of intractable generalized seizures [35]. The preoperative MRI is also important for subsequent comparison with the postoperative scan. Signal intensity changes are associated with significant side effects or complications from surgery [33]. Preoperative studies are important for deciding and planning the extent of callosotomy. Section of the anterior two-thirds of the corpus callosum carries little risk of permanent disconnection and is, therefore, the procedure of choice for patients in whom neuropsychometric studies indicate a relatively high IQ and level of performance. Many patients achieve significant improvement in their seizure control following anterior callosotomy, and the risk of total section is probably not warranted in the highly functioning patient. It is well recognized that a second-stage completion of the posterior corpus callosum transection can be performed at a later date if the seizure control after anterior callosotomy does not meet expectations. There are patients, however, in whom preoperative EEG and neuroimaging studies have demonstrated posterior cerebral pathology, where the propagation of seizure activity is likely occurring through the splenium and posterior body of the corpus callosum. In these cases the callosotomy should be completed as a primary procedure. Other factors promoting complete section of the corpus callosum in one stage include severe mental retardation and a low level of function that makes increased disability from disconnection unlikely, the presence of diffuse multifocal pathology extending beyond the boundaries of the frontal lobes, or a strong bias by the family, guardian, or caregivers against the possibility of
162
having to consider a second-stage procedure because of psychosocial issues. No attempt is made to spare the hippocampal commissure on the underside of the body of the corpus callosum. The anterior commissure and one limb of the fornix were sometimes included in the early series, but these more extensive commissurotomies did not result in noticeably better seizure outcome and were associated with increased morbidity. Anticonvulsant drug levels and a coagulation battery including a bleeding time are obtained preoperatively. Patients on valproate often have a prolonged bleeding time; this drug is discontinued and an appropriate anticonvulsant drug substituted if necessary at least 2 weeks before the scheduled surgery.
Surgical Considerations The operative technique for open section of the corpus callosum has varied little from institution to institution following introduction of the operating microscope in the early 1970s [29,50]. A recent reported refinement in the open craniotomy approach is the use of a frameless stereotactic image-guided system to guide the surgical approach and more accurately define the extent of callosotomy during the operation [39]. With the advent of stereotactic radiosurgery there have been several small series or case reports reporting the use of this noninvasive technology to ablate the corpus callosum [37,38,40,41]. The craniotomy is performed under general anesthesia. Intraoperative EEG monitoring is not necessary and probably of limited benefit, but if used to determine the extent of section up to the point where generalized epileptiform discharges become lateralized, the montages are placed prior to preparing and draping the field. The anesthetic dosages and end-tidal concentrations need to be closely monitored and controlled if intraoperative EEG is used.
2729
2730
162
Corpus callosotomy
At Minnesota the patient is placed in the supine position with the table slightly flexed and the skull secured by three-point skeletal fixation. This gives the surgeon a comfortable vantage, looking straight down on the vertex. Some surgeons prefer a lateral decubitus position in the belief that gravity acting on the dependent hemisphere aids exposure. A bicoronal incision is made with the right limb extending to the level of the top of the auricle and the left limb extending to the top of the insertion of the temporalis muscle. The skull is exposed by subperiosteal retraction of the scalp. The precise position of the bone flap is determined by preoperative assessment of the location of the bridging cortical veins. Usually the free bone flap is two-thirds in front and one-third behind the coronal suture. If a complete callosotomy is planned, it is better to have more exposure in the sagittal than in the coronal plane. The author prefers to trephine the skull on either side of the midline rather than directly over the sagittal sinus. The free bone flap is turned with a high-speed air drill or Gigli saw. The latter provides a beveled edge, allowing for a secure cosmetic closure in the frontal region. Beveled edges are not as necessary if titanium plates and burr hole covers are available that firmly secure the bone flap. It is important that the craniotomy cross the midline so that brain retraction is minimal while exposing the corpus callosum through the interhemispheric fissure. MRI with 3D reconstruction of the sagittal sinus and draining cerebral veins can provide anatomical guidance regarding optimum placement and extent of the craniotomy for the callosotomy [39]. The dura mater is opened far enough laterally so that there is little danger of injuring the large bridging veins as they approach the sagittal sinus. Large dural venous lakes and pacchionian bodies are also less conspicuous a few centimeters off the midline. The dura mater is opened to the edge of the sagittal sinus and dural retraction sutures are placed to assist the surgeon in obtaining exposure. It may be
necessary to tamponade venous ooze from the dura mater over the sagittal sinus with pledgets of Avitene, Surgicel, or Gelfoam. Bridging veins are spared, and this is usually possible if the bone flap is well placed and long enough in the sagittal plane. Mannitol is administered in a dosage of 1 g/kg at least 20 min prior to beginning the interhemisphere exposure. Hyperventilation to an end-tidal CO2 of 22 and the reverse Trendelenburg position may all assist with brain relaxation. If the ventricles are generous and the brain remains tight in spite of the above measures, the ipsilateral ventricle can be tapped as a last resort for relaxing the brain and avoiding excessive traction on the interhemispheric cortex. The surface of the cerebral cortex is protected with mosit pledgets, and a self-retaining retractor system is gently positioned once the brain is relaxed. Image-guided, frameless stereotactic surgical techniques have been reported to be effective in providing intraoperative feedback to the surgeon with regard to the location and extent of the callosotomy [39]. The operating microscope is brought into the field and standard microsurgical technique is used to take down interhemispheric adhesions. These adhesions may be quite dense in older patients who have had numerous falls and small bleeds or a prior episode of meningeal encephalitis. This dissection is usually much easier in young children. In cases where the cingulate gyri are densely adherent beneath the lower edge of the falx cerebri, some trauma to the pia mater and cingulate cortex may be unavoidable, but every effort should be made to limit such injury to one side. It is important to identify the callosal marginal and pericallosal arteries both for the purpose of avoiding injury to the vessels and for anatomical orientation. The pericallosal vessels lie immediately dorsal to the nacreous fibers of the corpus callosum. Division of the corpus callosum between the paired pericallosal arteries
Corpus callosotomy
assures the surgeon proximity to the midline and often exposes the cavum between the two leaves of the septum pellucidum once the corpus callosum is divided. Transection of the callosal fibers to either side of the paired pericallosal arteries will usually result in exposure of either the right or left lateral ventricle. The exposure is carried forward until the paired pericallosal arteries are seen merging with the anterior cerebral vessels where the frontopolar artery is given off and the anterior cerebral vessels dive down and around the genu and beneath the rostrum. Callosal fibers are divided and these vessels exposed until the anterior commissure is visualized to assure the surgeon that the genu and rostrum have been completely sectioned. Occasionally a single azygous anterior cerebral artery supplies both hemispheres. Various instruments can be used to section the callosal fibers, including microsuckers, microdissectors, the ultrasonic aspirator, and laser. Microsuction is safe and cost-effective. Exposure is facilitated by periodically adjusting the operating table into the reverse Trendelenburg position when working on the genu and the Trendelenburg positon when working on the splenium of the corpus callosum. In cases where there has been ventriculitis or ependymitis, the ependyma is quite firm and the integrity of the ventricles protected. Often, however, the ependyma is a gossamer structure that is easily fenestrated. With microsurgical technique and meticulous hemostasis, no complications or side effects are apparent when the ependyma is violated. Aseptic ventriculitis has not been a recognized problem in the Minnesota series. The extent of the ongoing section may be assessed in a number of ways and is, of course, not an issue when complete section is planned. It is helpful to study the midline sagittal cut of the MRI in order to appreciate the length and configuration of the corpus callosum. The inferior tip of the splenium can be a long reach in those cases where the corpus callosum takes the shape of an upside-down ‘‘U’’ and the splenium is quite bulbous. In such cases, it is necessary for the
162
surgeon to visualize the arachnoid membrane over the vein of Galen and internal cerebral veins before concluding that all the fibers of the splenium are sectioned. Distinctive anatomic features of the corpus callosum are recognizable on the MRI and guide the surgeon as to the extent of the section. MRI measurements can also be correlated with surgical measurements. One approach is to mark the posterior extent of the section and obtain an intraoperative plain skull film using the glabella, inion, and bregma as anatomic landmarks in conjunction with an MRI obtained before surgery [67]. A useful landmark is the isthmus of the corpus callosum that marks the boundary between the posterior body and the splenium. The experienced neurosurgeon can usually recognize this segment as the callosotomy progresses. If in doubt, however, a radiographic marker can be placed at the posterior point of the section and a lateral skull film obtained. A perpendicular line is then drawn from the marker to a line connecting the bregma and inion and compared with a similar line drawn from the isthmus on the preoperative MRI. The section can then be extended as necessary to assure a four-fifths partial callosotomy with sparing of the splenium [68]. Image-guided, frameless stereotactic surgical techniques have been reported to be effective in providing intraoperative feedback to the surgeon with regard to the location and extent of the callosotomy [39]. After irrigation of the wound and meticulous attention to hemostasis, the self-retaining retractors are removed and the anesthesiologist is asked to raise the patient’s venous pressure to see if this elicits any intradural bleeding. If not, the dura mater is closed with a running 4–0 nylon suture and dural tacking sutures are applied. The bone flap is replaced and secured with 0 gauge nonabsorbable sutures, wire, or titanium plates and screws. The galea is closed with interrupted inverted 2–0 Vicryl or similar suture. The dermis is closed with staples and a sterile dressing is applied.
2731
2732
162
Corpus callosotomy
Postoperative care is consistent with standard craniotomy precautions. The endotracheal tube is left in place until the patient is breathing well and has good protective reflexes. A nasogastric tube is often necessary during the early postoperative period for the administration of anticonvulsant medications. Children can usually be mobilized and discharged from the hospital more rapidly than adults.
Results The EEG changes are quite dramatic after complete section of the corpus callosum with desynchronization of the bilateral EEG abnormalities and the more clear-cut demonstration of focal abnormalities [69]. Cases have been reported where corpus callosotomy resulted during postoperative follow-up in the recognition of previously unapparent seizure foci that lead to subsequent successful resective surgery [70]. Corpus callosotomy can reduce the frequency and severity of seizures, however, without transforming the seizure pattern and ictal EEG activity from a generalized form into a strictly lateralized or partial one. This has been interpreted as suggesting a facilitatory role of the neurons participating in the corpus callosum that enables asymmetrical epileptogenicity of the two hemispheres to develop bisynchronous and symmetrical epileptiform discharges [71]. It is difficult if not futile to try to compare seizure outcomes between reported series of corpus callosotomy because of the differences in inclusion criteria among the epilepsy centers. Almost all reporting centers have found that partial or complete callosotomy is of significant benefit for patients with tonic/atonic seizures characterized by repeated falls (drop-attacks) and for generalized tonic-clonic seizures [72–89]. Maehara and Shimizu in a retrospective study of 52 patients with drop attacks concluded that total callosotomy is more effective for treatment
of drop-attacks than partial callosotomy [78]. Callosotomy for complex partial seizures is much more controversial, but a few investigators have reported successful outcomes for this seizure type [90,91]. There is considerable variation in the literature with regard to the effectiveness of partial callosotomy. Investigators agree that complete callosotomy is effective in reducing or eliminating bilaterally synchronous generalized discharges, but is this benefit worth the risk of neurological deficit? Comparing series of ‘‘anterior callosotomy’’ is difficult because of the variation in extent of callosotomy. At Minnesota, the extent of anterior callosotomy has evolved from a 50% section to two-thirds section and currently an 80% section. In children with a LennoxGastaut syndrome and severe mental retardation, a complete section is performed. An early age of seizure onset and callostomy during childhood correlate well with a favorable outcome [45,47,92]. The operation is better tolerated, seizures are more readily ameliorated, and the children avoid the risk of repeated falls and injuries as they grow. Education and training are also enhanced by the improved attentiveness, sociability, and behavior improvement seen after callosotomy. There is general agreement in the literature that a good seizure outcome is favored by a higher IQ [1,30,47]. Mental retardation is often associated with more severe and widespread brain damage, and the seizure disorder is accordingly more complex and refractory to both medical and surgical treatment. The majority of patients considered for corpus callosum section are mentally retarded, because this is the group of patients most in need of relief. Mental retardation is not a contraindication to corpus callosotomy. Purves et al. reported that the mentally retarded with diffuse cerebral pathology responded less well to anterior callosotomy than patients with unilateral pathology, but 40% of these patients were still classified as having an excellent seizure
Corpus callosotomy
outcome [46]. Cendes et al. reported a series of children undergoing corpus callosotomy, and 32 of the 34 patients were mentally retarded. Satisfactory seizure control was achieved in 73.5% of this population [93]. This incidence of mental retardation in the pediatric population considered for callosotomy is consistent with the Minnesota experience where there is a predominance of patients with Lennox-Gastaut syndrome and mental retardation. Patients with severe attentional disturbance of a frontal lobe type show improvement after either anterior or complete callosotomy. Attention disorders with frontal lobe epilepsy include incomplete retention of commands, easy distraction from tasks, and abrupt interruption of task in progress. The families and caregivers of patients undergoing corpus callosotomy are invariably enthusiastic about the improvement they witness in attentiveness, memory, sociability, verbalization, school performance, and integration with family, friends, and peers. They report that ‘‘we feel like we have our child back,’’ or ‘‘he seems more human.’’ This improvement has been ascribed to the relief of a frontal syndrome induced by frequent bilateral generalized seizures. The improvement in behavior is correlated with changes in frontal blood flow confirmed by single-photon emission computed tomography (SPECT) imaging [94]. In a study to evaluate functional outcome and parental satisfaction after corpus callosotomy, the parents of 15 of 17 patients undergoing anterior callosotomy for severe intractable generalized seizures reported satisfaction with the surgical outcome. The behavior change most closely associated with satisfaction was improved alertness and responsiveness. Parental satisfaction was not reported at a greater rate by the parents of nine patients who had more than 80% reduction in the targeted seizures than by the parents of six patients with 50–80% reduction in seizure frequency [95]. Seventy-six patients with LennoxGestaut syndrome and multiple severe seizures
162
occurring daily who had extended callosal sections back to the splenium were reported to have an increase in level of attention that was as useful as the improved seizure control in improving the quality of life of these patients [96]. Spencer et al. reported that total corpus callosotomy prevented secondarily generalized seizures in at least 75% of patients, which is consistent with other series [56]. They found that total callosotomy was more than twice as effective as partial section. They also found that the presence of two or more seizure types, a verbal IQ less than 80, and diffuse ictal EEG patterns were significantly more common in the patients who failed with anterior callosotomy and that these patients could benefit by extending and completing the callosotomy [97]. Fuiks et al. found that 70% of their 80 patients undergoing anterior callosotomy had significant improvement in their seizures, and 12.8% were cured. Eighty-six percent of patients with generalized tonic-clonic seizures and 83% of patients with atonic seizures were markedly better. Ten patients with either atonic, tonicclonic, or mixed seizures who failed to improve underwent subsequent completion of the callosotomy and in no case was the improvement sufficient to move the patient from one outcome category to another [90]. Other investigators have found worthwhile improvement in seizure outcome by completing a partial section [98]. Nei et al. compared anterior and complete corpus callosotomy with vagal nerve stimulation (VNS) in patients with refractory generalized seizures both with regard to seizure response and complications. Corpus callosotomy was reported to be more effective for seizure control but had a 3.8% permanent complication rate compared with no complications in the VNS treated group of patients [99]. Intraoperative EEG monitoring throughout the act of sectioning the corpus callosum showed that 78% of patients with generalized epileptiform discharges at the outset of the procedure
2733
2734
162
Corpus callosotomy
became lateralized when somewhere between two-thirds and all of the callosum was divided [100]. All of the 37 patients with tonic-atonic seizures with ‘‘drop’’ attacks had at least an 80% reduction in their seizure frequency. Those patients with the greatest decrease in generalized discharges had the greatest decrease in seizures (95.5%), but six patients with no lateralization of generalized discharges showed an 88% decrease in seizures. Patients with a mild or moderate decrease in generalized discharges had an 85% decrease in seizures. It therefore appears that although lateralization of generalized epileptiform discharges is evident in more than threefourths of patients, the degree of lateralization does not correlate well with the degree of tonicatonic seizures. Intraoperative surface EEG monitoring was therefore not considered particularly helpful as a guide to determining the extent of the callosal section. Kwan et al. retrospectively analyzed electrocorticograms (ECoGs) obtained during anterior callosotomies in 48 patients with Lennox-Gastaut syndrome. They concluded that changes in ECoGs during callosotomy do not predict postoperative seizure outcome in that insignificant blockage of bisynchronous epileptiform discharges in ECoGs during callosotomy did not predict a worse prognosis than that associated with significant intraoperative blockage [101]. Intraoperative electrophysiological observations in patients undergoing corpus callosotomy suggest more than a simple transference of cortical spike discharges through the corpus callosum but also a facilitatory or recruitment role in epileptogenesis [71,102].
Side Effects and Complications In 1936 Walter Dandy described his experience with sectioning the corpus callosum to approach third ventricular and pineal tumors. ‘‘The corpus callosum is split longitudinally from its posterior extremity to a point anteriorly where the third or
lateral ventricle comes into view; this incision is bloodless. Usually this incision takes most and sometimes all of this structure to its downward bend. No symptoms follow its division. This simple experiment at once disposes of the extravagant claims to function of the corpus callosum’’ [103]. Barely a year later, however, Trescher and Ford reported alexia in the left visual field in a patient of Dandy’s upon whom a third ventricular colloid cyst was approached transcallosally. They offered a number of explanations for this finding in the face of Dandy’s earlier disclaimer, suggesting that the section was perhaps more extensive for this lesion than that necessary for approaching pineal tumors. They also noted that ‘‘special methods of examination are required to demonstrate the essential symptoms’’ [104]. It is important to distinguish side effects from complications when discussing the results of corpus callosotomy and in presenting the risks of the procedures. For purposes of discussion in this context, a side effect is defined as a functional outcome that is expected as a result of the procedure and one that is acceptable in consideration of the potential benefits of surgery. It is well recognized that long-term, ageadjusted mortality rates are significantly higher in people with refractory epilepsy and the mortality is strongly related to seizure control. The excess mortality associated with refractory epilepsy is eliminated after epilepsy surgery when seizures are abolished. Patients undergoing palliative surgery such as corpus callosotomy, therefore, can be expected to and do have a higher subsequent long-term mortality than patients undergoing curative resective surgery. While it is true that sectioning the corpus callosum can result in side effects and complications, these are usually quite transient and do not interfere with the quality of the patient’s day-today activities. In well-selected patients, particularly those suffering severely from repeated falls and injuries, it is probably preferable to be aggressive rather than timid in considering the
Corpus callosotomy
extent of the section, because the disconnection syndrome is rarely significant in the daily life of these patients. Children tolerate callosal section better than adults, have fewer complications and side effects, and have a shorter convalescence. Neurological deficits following corpus callosotomy are best distinguished by whether they are transient or permanent, significant or only apparent by means of sophisticated neuropsychometric testing, expected or unexpected, a side effect of callosal disconnection or a complication of the operation such as infection, hemorrhage, or infarction. During the immediate postoperative period, which can be for as short a time as a few hours in a young child or for as long as 2–3 weeks in a severely mentally compromised adult, a number of transient neuropsychological phenomena are observed. The patient lacks spontaneity of speech and motion. If any speech occurs, it is often one or two primitive words or short phrases occurring only in response to verbal or physical stimulation. The patient shows a paucity of motion, and if asked to raise an arm or squeeze a hand, there is often a prolonged delay between the request or command and the response. There is a variable degree of left-sided neglect, more pronounced in the leg than in the arm. The inexperienced observer may interpret this as a paresis, but true weakness is unusual unless a cortical venous infarction or severe retraction edema in the posterior frontal lobe has occurred. Adult patients tend to sit slumped, with reduced postural tone; the head often deviates toward the left side; and the face wears a dull, apathetic expression. The etiology of these transient findings is not clearly established but may be caused by disruption of commissural fibers between the frontal lobes and supplementary motor cortices. Traction injury resulting in edema and vascular compromise to the medial frontal lobe may be a contributing factor in some cases. Mutism for example is almost always transient and resolution is complete [105].
162
This syndrome can be seen after anterior or total callosotomy and is distinct from the permanent disconnection syndrome documented by specific neuropsychometric stimulus testing after complete callosotomy. The left visual and tactile agnosia detected by specific and artificial testing paradigms is of no significance to the typical candidate for callosotomy. There are several specific situations, however, where callosotomy has a particular affinity for producing significant focal deficits. One such situation is where the early onset of hemispheric damage results in an acquired right hemispheric dominance for language in a right-handed individual who writes with the right hand. An analogous risk occurs when a genetically determined left-handed person is left hemispheric-dominant for speech. A disconnection deficit also occurs when hemianopsia or alexia exist in the visual field contralateral to the hemisphere dominant for language. If an early hemispheric injury causes speech and verbal memory to be located in contralateral hemispheres, corpus callosotomy produces a significant risk for language deficits [106]. New seizure types are sometimes first apparent or appreciated after corpus callosotomy [56,73,91,107]. These ‘‘new’’ seizures are not common and are usually described as being either complex partial, simple partial, or myoclonic. Simple partial seizures and focal myoclonic seizures may be recognized for the first time after corpus callosotomy [108]. These seizures may represent an active focus that is no longer generalizing or an irritable region of cortex no longer subjected to inhibitory influences. On occasion these ‘‘new’’ seizures are sufficiently frequent to concern the patient or family. The simple partial seizures tend to improve over time or with adjustments of medication. Analysis of patients from the Minnesota experience following partial callosotomies allows us to reach several conclusions relevant to permanent disconnection syndromes. Patients with two-thirds of the corpus callosum sectioned
2735
2736
162
Corpus callosotomy
demonstrate left ear suppression of verbal information under conditions of dichotic stimulation. Patients with only the splenium and most posterior portion of the body of the callosum intact are able to execute complex motor sequences involving the left hand and arm to verbal command and are capable of the interhemispheric transfer of visual, tactile, and kinesthetic messages. Patients with only the splenium intact exhibit a disconnection for all modalities tested except vision, which remains completely intact [109]. These findings underscore the importance of the posterior body and splenium of the corpus callosum for basic and sensorimotor interhemispheric integration. A study examining bimanual coordination during the drawing of symmetrical and asymmetrical figures confirmed that the posterior callosum is necessary to mediate the coordination of previously unlearned direction information between the hands during bimanual movements [110]. It has been observed that previously welllearned cooperative actions of the hands such as tying shoes remain intact after extensive callosotomy while novel actions are virtually impossible without visual guidance. This suggests that duplicate memory engrams of well-learned actions can be accessed by both cerebral hemispheres without callosal mediation, whereas callosal interactions are necessary for precise crossmatching of sensory information during spatial planning or motor learning [111]. The targeted population of intractable epilepsy patients whose surgical callosal sections have spared these posterior fibers do not demonstrate a permanent, significant, readily assessable disconnection syndrome or any measurable additional impairment of intellectual ability. This suggests that the practical significance of uncomplicated anterior callosum section is minimal. The topographic arrangement of callosal fibers accounts for the selective nature of the disconnection syndromes [112]. As patient selection criteria have been refined and modern neurosurgical and microsurgical
technology developed, the complications and risks of callosal sectioning have lessened considerably. Surgical complications include aseptic and septic ventriculitis that can result in delayed hydrocephalus. The risk of ventricular contamination by blood and tissue debris is less since Wilson introduced the operating microscope and the use of microsurgical techniques to callosotomy in the 1970s [50]. Wilson recommended that every effort be made to maintain the integrity of the ependyma in order to reduce ventricular soilage. Recognition that anterior commissurotomy, fornicotomy, and section of the massa intermedia are of little benefit also led to constructive advances in the avoidance of this complication. The risk of postoperative intracranial bleeding in the epidural or subdural spaces is reduced by preoperative attention to the coagulation studies and discontinuation of anticonvulsant drugs, such as divalproex (Depakote) known to prolong the bleeding time, well before the day of the surgery. Cortical venous infarction with secondary edema and hemorrhage is avoided by sparing the bridging cortical veins. Careful positioning of the craniotomy after analysis of the venous anatomy on preoperative scans or angiograms lessens this risk. The risk of air emboli is controlled by careful attention to the venous sinuses, waxing the bone edges, avoiding the sitting position, and utilizing the Doppler for detection of air throughout the procedure. Injury to the sagittal sinus is reduced by awareness of the skull thickness in a patient on chronic anticonvulsant therapy and by not placing burr holes directly over the sagittal sinus. The risk of corpus callosotomy at major epilepsy centers experienced with this procedure is comparable to that of resective surgery and is no longer a contraindication in patients meeting appropriate selection criteria. Evolving technologies such as frameless stereotactic surgery and stereotactic radiosurgery are currently being explored as means to further improve the effectiveness and reduce the morbidity of callosotomy [37–41].
Corpus callosotomy
References 1. Blume WT. Corpus callosum section for seizure control: rationale and review of experimental and clinical data. Cleve Clin Q 1984;51:319-32. 2. Tomasch J. Size, distribution, and number of fibres in the human corpus callosum. Anat Rec 1954;119:119-35. 3. Pandya DN, Rosene DL. Some observations on trajectories and topography of commissural fibers. In: Reeves AG, editor. Epilepsy and the corpus callosum. New York: Plenum Press; 1985. p. 21-39. 4. Curtis HJ. Intercortical connections of the corpus callosum as indicated by evoked potentials. J Neurophysiol 1940;3:404-13. 5. Asanuma H, Okuda O. Effects of transcallosal volleys on pyramidal tract cell activity of cat. J Neurophysiol 1962;25:198-208. 6. Erickson TC. Spread of the epileptic discharge; an experimental study of the afterdischarge induced by electrical stimulation of the cerebral cortex. Arch Neurol Psychiatry 1940;43:429-52. 7. Marcus EM, Watson CW. Bilateral synchronous spike wave electrographic patterns in the cat: interaction of bilateral cortical foci in the intact, the bilateral corticalcallosal, and diencephalic preparation. Arch Neurol 1966;14:601-10. 8. Marcus EM, Watson CW. Symmetrical epileptogenic foci in monkey cerebral cortex: mechanisms of interaction and regional variations in capacity for synchronous discharges. Arch Neurol 1968;19:99-116. 9. Marcus EM, Watson CW, Simon SA. An experimental model of some varieties of petit mal epilepsy: electricalbehavioral correlation of acute bilateral epileptogenic foci in cerebral cortex. Epilepsia 1968;9:233-48. 10. Musgrave J, Gloor P. The role of the corpus callosum in bilateral interhemispheric synchrony of spike and wave discharge in feline generalized penicillin epilepsy. Epilepsia 1980;21:369-78. 11. Marcus EM. Generalized seizure models and the corpus callosum. In: Reeves AG, editor. Epilepsy and the corpus callosum. New York: Plenum Press; 1985. p. 131-206. 12. Wada JA, Nakashima T, Kaneko Y. Forebrain bisection and feline amygdaloid kindling. Epilepsia 1982; 23:521-30. 13. Maquet R, Manini CH, Catier J. Photically induced epilepsy in Papio papio: the initiation of discharges and the role of the frontal cortex and of the corpus callosum. In: Petsche H, Brazier M, editors. Synchronization of the EEG in the epilepsies. Vienna: Springer; 1972. p. 347-67. 14. Vergnes M, Marescaux CL, Lannes B, DePaulis A, Micheletti G, Warter JM. Interhemispheric desynchronization of spontaneous spike-wave discharges by corpus callosum transaction in rats with petit mal-like epilepsy. Epilepsy Res 1989;4:8-13.
162
15. Kopeloff N, Kennard M, Pacella B, Kopeloff LM, Chusid JG. Section of corpus callosum in experimental epilepsy in the monkey. Arch Neurol Psychiatry 1950; 63:719-27. 16. Stavraky GW. Supersensitivity following lesions of the nervous system. Toronto: University of Toronto Press; 1961. p. 33-8. 17. Wada JA, Sato M. The generalized convulsive seizure state induced by daily electrical stimulation of the amygdala in split brain cats. Epilepsia 1975;16:417-30. 18. Wada JA, Sato M. Generalized convulsive seizures induced by daily electrical stimulation of the amygdala in cats: correlative electrographic and behavioral features. Neurology 1974;24:565-74. 19. Ono T, Fujimura K, Yoshida S, Ono K. Suppressive effect of callosotomy on epileptic seizures is due to the blockade of enhancement of cortical reactivity by transcallosal volleys. Epilepsy Res 2002;51:117-21. 20. Collins R, Kennedy C, Sokoloff L, Plum F. Metabolic anatomy of focal motor seizures. Arch Neurol 1976; 33:536-42. 21. Kusske J, Rush J. Corpus callosum and propagation of afterdischarge to contralateral cortex and thalamus. Neurology 1978;28:905-12. 22. Schwartzkroin P, Mutani R, Prince D. Orthodromic and antidromic effects of a cortical epileptiform focus on ventrolateral nucleus of the cat. J Neurophysiol 1975; 38:795-811. 23. Bremer F. Nouvelle recherches dans le me´canisme du sommeil. CR Soc Biol (Paris) 1936;122:460-4. 24. Van Wagenen WP, Herren RY. Surgical division of commissural pathways in the corpus callosum: relation to spread of an epileptic attack Arch Neurol Psychiatry 1940;44:740-59. 25. Joynt RJ. History of forebrain commissurotomy. In: Reeves AG, editor. Epilepsy and the corpus callosum. New York: Plenum Press; 1985. p. 237-41. 26. Bogen JE, Fisher ED, Vogel PJ. Cerebral commissurotomy: a second case report. Am J Med 1965;194:1328-9. 27. Bogen JE, Vogel PJ. Treatment of generalized seizures by cerebral commissurotomy. Surg Forum 1963;14:431-3. 28. Luessenhop AJ. Interhemispheric commissurotomy (the split brain operation) as an alternative to hemispherectomy for control of intractable seizures. Am Surg 1970; 36:265-8. 29. Wilson DH, Reeves AG, Gazzaniga MS, Culver C. Cerebral commissurotomy for control of intractable seizures. Neurology 1977;27:708-15. 30. Wilson DH, Reeves AG, Gazzaniga MS. ‘‘Central’’ commissurotomy for intractable generalized epilepsy: series two. Neurology 1982;32:687-97. 31. Gates JR, Mireles R, Maxwell R, Sharbrough F, Forbes G. Magnetic resonance imaging, electroencephalogram, and selected neuropsychological testing in staged corpus callosotomy. Arch Neurol 1986;43:1188-91.
2737
2738
162
Corpus callosotomy
32. Harris RD, Roberts DW, Cromwell LD. MR imaging of corpus callosotomy. Am J Neuroradiol 1989;10:677-80. 33. Khurana DS, Strawsburg RH, Robertson RL, Madsen JR, Helmers SL. MRI signal changes in the white matter after corpus callosotomy. Pediatr Neurol 1999;21:691-5. 34. Maxwell RE, Gates JR, McGeachie R. Magnetic resonance imaging in the assessment and surgical management of epilepsy and functional neurological disorders. Appl Neurophysiol 1987;50:369-73. 35. Sorenson JM, Wheless JW, Baumgartner JE, Thomas AB, Brookshire BL, Clifton GL, Willmore LJ. Corpus callosotomy for medically intractable seizures. Pediatr Neurosurg 1997;27:260-7. 36. Sussman NM, Scanlon M, Garfinkle W, Callanan M, O’Connor MJ, Harner RN. Magnetic resonance imaging after corpus callosotomy. Neurology 1987;37:350-4. 37. Celis MA, Moreno-Jime´nez S, La´rrago-Gutie´rrez JM, Alonso-Vanegas MA, Garcı´a-Gardun˜o OA, Martı´nezJua´rez IE, Ferna´ndez-Go´nzalez MC. Corpus callostomy using conformal stereotactic radiosurgery. Childs Nerv Syst 2007;23:917-20. 38. Feichtinger M, Schro¨ttner O, Eder H, Holthausen H, Pieper T, Unger F, Holl A, Gruber L, Ko¨rner E, Trinka E, Fazekas F, Ott E. Efficacy and safety of radiosurgical callosotomy: a retrospective analysis. Epilepsia 2006; 47:1184-91. 39. Hodaie M, Musharbash A, Otsubo H, Snead OC, Chitoku S, Ochi A, Holowka S, Hoffman HJ, Rutka JT. Image-guided, frameless stereotactic sectioning of the corpus callosum in children with intractable epilepsy. Pediatr Neurosurg 2001;34:286-94. 40. Pendl G, Eder HG, Schroettner O, Leber KA. Corpus callosotomy with radiosurgery. Neurosurgery 1999; 45:303-7. 41. Smyth MD, Klein EE, Dodson WE, Mansur DB. Radiosurgical posterior corpus callosotomy in a child with Lennox-Gastaut syndrome. Case report. J Neurosurg 2007;106:312-5. 42. Harbaugh RE, Wilson DH, Reeves AG, Gazzaniga MD. Forebrain commissurotomy for epilepsy: review of 20 consecutive cases. Acta neurochir 1983;68:263-75. 43. Garcia-Flores E. Corpus callosum section for patients with intractable epilepsy. Appl Neurophysiol 1987; 50:390-7. 44. Marino R, Jr, Ragazzo PC. Selective criteria and results of selective partial callosotomy. In: Reeves AG, editor. Epilepsy and the corpus callosum. New York: Plenum Press; 1985. p. 281–301. 45. Murro AM, Flanigan HF, Gallagher BB, King DW, Smith JR. Corpus callosotomy for the treatment of intractable epilepsy. Epilepsy Res 1988;2:44-50. 46. Purves SJ, Wada JA, Woodhurst WB, Kosaka B, Li D. Results of anterior corpus callosum section in 24 patients with medically intractable seizures. Neurology 1988; 38:1194-201.
47. Spencer SS, Spencer DD, Williamson PD, Sass KJ, Novelly RA, Mattson RH. Corpus callosotomy for epilepsy: I. Seizure effects. Neurology 1988;38:19-24. 48. Ma X, Liporace J, O’Connor MJ, Sperling MR. Neurosurgical treatment of medically intractable status epilepticus. Epilepsy Res 2001;46:33-8. 49. Avila JO, Radvany J, Huck FR, Pires de Camargo CH, Marino R, Jr, Ragazzo PC, Riva D. Anterior callosotomy as a substitute for hemispherectomy. Acta Neurochir Suppl (Wien) 1980;30:137-43. 50. Wilson DW, Culver C, Waddington M, Gazzaniga M. Disconnection of the cerebral hemispheres: an alternative to hemispherectomy for the control of intractable seizures. Neurology 1975;25:1149-53. 51. Goodman RN, Williamson PD, Reeves AG, Spencer SS, Spencer DD, Mattson RH, Roberts DW. Interhemisphere commissurotomy for congenital hemiplegics with intractable epilepsy. Neurology 1985;35:1351-4. 52. Tinuper P, Andermann F, Villemure JG, Rasmussen TB, Quesney LF. Functional hemispherectomy for treatment of epilepsy associated with hemiplegia: rationale, indications, results and comparisons with callosotomy. Ann Neurol 1988;24:27-34. 53. Davies KG, Maxwell RE, French LA. Hemispherectomy for intractable seizures: long-term results in 17 patients followed up to thirty-eight years. J Neurosurg 1993; 78:733-40. 54. McClelland S, III, Maxwell RE. Hemispherectomy for intractable epilepsy in adults. The first reported series. Ann Neurol 2007;61:372-6. 55. Rasmussen T. Hemispherectomy for seizures revisited. Can J Neurol Sci 1983;10:71-8. 56. Spencer SS, Spencer DD, Glaser GH, Williamson PD, Mattson RH. More intense focal seizure types after callosal section: the role of inhibition. Ann Neurol 1984; 16:686-93. 57. Fiol ME, Gates JR. EEG studies and corpus callosotomy results. EEG Clin Neurophysiol 1984;58:34. 58. Hanson RR, Risinger M, Maxwell R. The ictal EEG as a predictive factor for outcome following corpus callosum section in adults. Epilepsy Res 2002;49:89-97. 59. Connolly MB, Hendson G, Steinbok P. Tuberous sclerosis complex: a review of the management of epilepsy with emphasis on surgical aspects. Childs Nerv Syst 2006;22:896-908. 60. Guerreiro MM, Andermann F, Andermann E, Palmini A, Hwang P, Hoffman HJ, Otsubo H, Bastos A, Dubeau F, Snipes GJ, Olivier A, Rasmussen T. Surgical treatment of epilepsy in tuberous sclerosis: strategies and results in 18 patients. Neurology 1998;51:1263-9. 61. Kamida T, Maruyama T, Fujiki M, Kobayashi H, Izumi T, Baba H. Total callosotomy for a case of lissencephaly presenting with West syndrome and generalized seizures. Childs Nerv Syst 2005;21:1056-60.
Corpus callosotomy
62. Kawai K, Shimizu H, Yagishita A, Maehara T, Tamagawa K. Clinical outcomes after corpus callosotomy in patients with bihemispheric malformations of cortical development. J Neurosurg 2004;101:7-15. 63. Landy HJ, Curless RG, Ramsay RE, Slater J, AjmoneMarson C, Quencer RM. Corpus callosotomy for seizures associated with band heterotopia. Epilepsia 1993; 34:79-83. 64. Pallini R, Aglioti S, Tassinari G, Berlucchi G, Colosimo C, Rossi GF. Callosotomy for intractable epilepsy from bihemispheric cortical dysplasias. Acta Neurochir (Wien) 1995;132:79-86. 65. Stearns M, Wolf AL, Barry E, Bergey G, Gellad F. Corpus callosotomy for refractory seizures in a patient with cortical heterotopia: case report. Neurosurgery 1989; 25:633-5. 66. Vossler DG, Lee JK, Ko TS. Treatment of seizures in subcortical laminar heterotopia with corpus callosotomy and lamotrigine. J Child Neurol 1999;14:282-8. 67. Awad IA, Wyllie E, Luders H, Ahl J. Intraoperative determination of the extent of corpus callosotomy for epilepsy: two simple techniques. Neurosurgery 1990;26:102-6. 68. Shimizu H, Ohta Y, Suzuki I, Ishijima B, Sugishita M. Anterior extensive corpus callosotomy including resection of the isthmus. Jpn J Psychiatry Neurol 1993;47:264-6. 69. Gates JR, Maxwell R, Leppik, IE, Fiol M, Gumnit R. Electroencephalographic and clinical effects of total corpus callosotomy. In: Reeves AG, editor. Epilepsy and the corpus callosum. New York: Plenum Press; 1985. p. 315-28. 70. Clarke DF, Wheless JW, Chacon MM, Breier J, Koenig MK, McManis M, Castillo E, Baumgartner JE. Corpus callosotomy: a palliative therapeutic technique may help identify resectable epileptogenic foci. Seizure 2007; 16:545-53. 71. Matsuo A, Ono T, Baba H, Ono K. Callosal role in generation of epileptiform discharges: quantitative analysis of EEGs recorded in patients undergoing corpus callosotomy. Clin Neurophysiol 2003;114:2165-71. 72. Amacher AL. Midline commissurotomy for the treatment of some cases of intractable epilepsy. Childs Brain 1976;2:54-8. 73. Gates JR, Rosenfeld WE, Maxwell RE, Lyons RE. Response of multiple seizure types to corpus callosum sectioning. Epilepsia 1987;28:28-34. 74. Geoffroy G, Lassonde M, Delisle F. Corpus callosotomy for control of intractable epilepsy in children. Neurology 1983;33:891-7. 75. Jenssen S, Sperling MR, Tracy JI, Nei M, Joyce L, David G, O’Connor M. Corpus callosotomy in refractory idiopathic generalized epilepsy. Seizure 2006;15:621-9. 76. Kim DS, Yang KH, Kim TG, Chang JH, Chang JW, Choi JU, Lee BI. The surgical effect of callosotomy in the treatment of intractable seizure. Yonsei Med J 2004;45:233-40.
162
77. Kwan SY, Wong TT, Chang KP, Chi CS, Yang TF, Lee YC, Guo WY, Su MS. Seizure outcome after corpus callosotomy: the Taiwan experience. Childs Nerv Syst 2000; 16:87-92. 78. Maehara T, Shimizu H. Surgical outcome of corpus callosotomy in patients with drop attacks. Epilepsia 2001;42:67-71. 79. Mamelak AN, Barbaro NM, Walker JA, Laxer KD. Corpus callosotomy: a quantitative study of the extent of resection, seizure control, and neuropsychological outcome. J Neurosurg 1993;79:688-95. 80. McInerney J, Siegel AM, Nordgren RE, Williamson PD, Thadani VM, Jobst B, Reeves AG, Roberts DW. Long-term seizure outcome following corpus callosotomy in children. Stereotact Funct Neurosurg 1999; 73:79-83. 81. Papo I, Quattrini A, Ortenzi A, Paggi A, Rychlicki F, Provinciali L, Del Pesce M, Cesarano C, Fioravanti P. Predictive factors of callosotomy in drug-resistant epileptic patients with a long follow-up. J Neurosurg Sci 1997;41:31-6. 82. Phillips J, Sakas DE. Anterior callosotomy for intractable epilepsy: outcome in a series of twenty patients. Br J Neurosurg 1996;10:351-6. 83. Pinard JM, Delalande O, Chiron C, Soufflet C, Plouin P, Kim Y, Dulac O. Callosotomy for epilepsy after West syndrome. Epilepsia 1999;40:1727-34. 84. Rahimi SY, Park YD, Witcher MR, Lee KH, Marrufo M, Lee MR. Corpus callosotomy for treatment of pediatric epilepsy in the modern era. Pediatr Neurosurg 2007;43:202-8. 85. Rathore C, Abraham M, Rao RM, George A, Sankara Sarma P, Radhakrishnan K. Outcome after corpus callosotomy in children with injurious drop attacks and severe mental retardation. Brain Dev 2007; 29(9):577-85. 86. Rayport M, Ferguson SM, Corrie SW. Outcome and indications of corpus callosotomy section for intractable seizure control. Appl Neurophysiol 1983;46:47-51. 87. Rossi GF, Colicchio G, Marchese E, Pompucci A. Callosotomy for severe epilepsies with generalized seizures: outcome and prognostic factors. Acta Neurochir (Wien) 1996;138:221-7. 88. Rougier A, Claverie B, Pedespan JM, Marchal C, Loiseau P. Callosotomy for intractable epilepsy: overall outcome. J Neurosurg Sci 1997;41:51-7. 89. Shimizu H. Our experience with pediatric epilepsy surgery focusing on corpus callosotomy and hemispherotomy. Epilepsia 2005;46:30-1. 90. Fuiks KS, Wyler AR, Hermann BP, Somes G. Seizure outcome from anterior and complete corpus callosotomy. J Neurosurg 1991;74:573-8. 91. Roberts DW, Reeves AG. Effect of commissurotomy on complex partial epilepsy in patients without a resectable seizure focus. Appl Neurophysiol 1987;50:398-400.
2739
2740
162
Corpus callosotomy
92. Lassonde M, Sauerwein C. Neuropsychological outcome of corpus callosotomy in children and adolescents. J Neurosurg Sci 1997;41:67-73. 93. Cendes F, Ragazzo P, daCosta V, Martins L. Corpus callosotomy in treatment of medically resistant epilepsy: preliminary results in a pediatric population. Epilepsia 1993;34:910-7. 94. Septien L, Giroud M. Sautreaux J, Duma R. Corpus callosotomy for intractable seizures in the pediatric age group: influence on frontal syndrome. Childs Nerv Syst 1992;8:2-3. 95. Gilliam F, Wyllie E, Kotagal P, Geckler C, Rusyniak G. Parental assessment of functional outcome after corpus callosotomy. Epilepsia 1996;37:753-7. 96. Cukiert A, Burattini JA, Mariani PP, Caˆmara RB, Seda L, Baldauf CM, Argentoni M, Baise-Zung C, Forster CR, Mello VA. Extended, one-stage callosal section for treatment of refractory secondarily generalized epilepsy in patients with Lennox-Gastaut and Lennox-like syndromes. Epilepsia 2006;47:371-4. 97. Spencer S, Spencer D, Sass K, Westerveld M, Katz A, Mattson R. Anterior, total, and two-stage corpus callosum section: differential and incremental seizure responses. Epilepsia 1993;34:561-7. 98. Reutens D, Bye A, Hopkins I, Danks A, Somerville E, Walsh J, Bleasel A, Ouvier R, MacKenzie R, Manson J. Corpus callosotomy for intractable epilepsy: seizure outcome and prognostic factors. Epilepsia 1993;34:904-9. 99. Nei M, O’Connor M, Liporace J, Sperling MR. Refractory generalized seizures: response to corpus callosotomy and vagal nerve stimulation. Epilepsia 2006;47:115-22. 100. Fiol M, Gates J, Mireles R, Maxwell RE, Erickson DM. Value of intraoperative EEG changes during corpus callosotomy in predicting surgical results. Epilepsia 1993;34:74-8. 101. Kwan SY, Lin JH, Wong TT, Chang KP, Yiu CH. Prognostic value of electrocorticography findings during callosotomy in children with Lennox-Gastaut syndrome. Seizure 2005;14:470-5. 102. Ono T, Matsuo A, Baba H, Ono K. Is a cortical spike discharge ‘‘transferred’’ to the contralateral cortex via the corpus callosum?: an intraoperative observation of
103. 104.
105.
106.
107.
108.
109.
110.
111.
112.
electrocorticogram and callosal compound action potentials. Epilepsia 2002;43:1536-42. Dandy WE. Operative experiences in cases of pineal tumors. Arch Surg 1936;33:19-46. Trescher JH, Ford RR. Colloid cyst of the third ventricle: report of a case: operative removal with section of posterior half of the corpus callosum. Arch Neurol Psychiatry 1937;37:959-73. Quattrini A, Del Pesce M, Provinciali L, Cesarano R, Ortenzi A, Paggi A, Rychlicki F, Fioravanti P, Papo I. Mutism in 36 patients who underwent callosotomy for drug-resistant epilepsy. J Neurosurg Sci 1997; 41:93-6. Novelly R, Lifrak M. Forebrain commissurotomy reinstates effects of pre-existing hemisphere lesions: An examination of the hypothesis. In: Reeves AG, editor. Epilepsy and the corpus callosum. New York: Plenum Press; 1985. p. 467-500. Oguni H, Olivier A, Andermann F, Comair J. Anterior callosotomy in the treatment of medically intractable epilepsies: a study of 43 patients with a mean follow up of 39 months. Ann Neurol 1991;30:357-64. Nordgren R, Reeves A, Viguera A, Roberts D. Corpus callosotomy for intractable seizures in the pediatric age group. Arch Neurol 1991;48:364-72. Risse GL, Gates JR, Lund G, Maxwell R, Rubens A. Interhemispheric transfer in patients with incomplete section of the corpus callosum. Arch Neurol 1989; 46:437-43. Eliassen JC, Baynes K, Gazzaniga MS. Direction information coordinated via the posterior third of the corpus callosum during bimanual movements. Exp Brain Res 1999;128:573-7. Franz EA, Waldie KE, Smith MJ. The effect of callosotomy on novel versus familiar bimanual actions: a neural dissociation between controlled and automatic processes? Psychol Sci 2000;11:82-5. Caille´ S, Sauerwein HC, Schiavetto A, Villemure JG, Lassonde M. Sensory and motor interhemispheric integration after section of different portions of the anterior corpus callosum in nonepileptic patients. Neurosurgery 2005;57:50-9.
153 EEG in Epilepsy M. Hoppe . R. Wennberg . P. Tai . B. Pohlmann-Eden
Introduction Electroencephalography (EEG) has for decades been a crucial method in the diagnostic process of epilepsy. Even today, EEG is still the only procedure that allows for the bed-side detection of epileptiform activity, providing both diagnosis as well as non-invasive insights into the pathophysiological dynamics of epilepsy. Other conditions or situations where EEG has proven useful and often diagnostic include encephalopathies, slow virus infections, sleep disorders, coma staging (including so-called brain death), and screening for structural focal lesions. However, EEG remains most prominent as a tool for the investigation of epilepsy. Within the epileptic disorders, EEG allows one to address the following issues (in particular when combined with video recording).
Differentiation of epileptic and psychogenic non-epileptic seizures Seizure classification (focal vs. generalized) Classification of epilepsy syndromes Detection of photosensitivity Detection of underlying encephalopathies or focal disturbances Monitoring of antiepileptic drug therapy (generalized epilepsies particularly) Identification of candidates for epilepsy surgery
Definition of Epileptiform Activity Most pathological abnormalities recorded with EEG are not specific with respect to etiology. This is true, for example, in the case of focal intermittent #
Springer-Verlag Berlin/Heidelberg 2009
slow wave activity and other asymmetrical findings, which simply reflect focal or lateralized disturbances of brain function that may be caused by diverse pathologies [1]. Intermittent rhythmic slow wave activity in the delta (<4 Hz) frequency band, when localized to the temporal lobe, is something of an exception to this rule. This pattern, first described by [2] as TIRDA (temporal intermittent rhythmic delta activity), is highly associated with temporal lobe epilepsy [3,4]. However, it is the transient, sharp ‘‘epileptiform’’ potentials, which often stand out strikingly from the cerebral background activity, that are the hallmark of epilepsy in EEG. Epileptiform activity may be focal or generalized. > Table 153-1 gives an overview of the terminology of interictal (between seizures) epileptiform potentials [5]. A detailed discussion of the morphological features necessary to classify sharp transients as epileptiform or non-epileptiform in EEG can be found in [6]. An example of a focal temporal lobe spike and wave complex is shown in > Figure 153-1. The neuronal correlate of epileptiform activity recorded with EEG is thought to be a paroxysmal depolarization of the cell membrane potential, the so-called paroxysmal depolarization shift [7]. The spike or sharp wave component of epileptiform activity recorded with EEG is dependent for its generation on a critical mass of synchronously firing neurons. The subsequent slow wave component is the result of an inhibitory process thought to reflect delayed hyperpolarization. The transition from interictal states to ictal (seizure) states is a poorly understood phenomenon, which has been conceptualized to occur, in most cases, via a process of either disinhibition or hypersynchronization. In the case of disinhibition the delayed hyperpolarization
2576
153
EEG in epilepsy
might evolve stepwise to a prolonged depolarization generating fast repetitive action potentials (recorded by EEG as low amplitude high frequency field potential activity). In the alternate case, ictal transition from hypersynchronization is conceived to result in repetitive spike and wave discharges and subsequently clinical symptoms [7].
Sensitivity and Specifity of Interictal EEG ‘‘Routine’’ interictal EEG (typically a 20–30 min recording) performed in a patient with clinical . Table 153-1 Terminology of interictal epileptiform potentials in epilepsy Spikes (duration <70 ms) Sharp waves (duration <200 ms) Spike and wave complexes Polyspikes Polyspike and wave complexes
epilepsy frequently shows no epileptiform activity; conversely, the presence of interictal epileptiform potentials on EEG is not proof of clinical epilepsy. Epileptiform potentials were reported in 2.2–3.5% of children and 0.2–0.5% of adults from a healthy population with no clinical epilepsy, and in 2.0–2.6% of individuals with a psychiatric disorder [8]. Approximately 50% of patients with previously diagnosed epilepsy had epileptiform activity on their initial EEG recording when first investigated in an epilepsy clinic [9,10]. The percentage of patients with epileptiform activity on EEG increased to 92% if repeated recordings (up to four, including a sleep tracing) were performed [10]. In patients presenting with a first seizure epileptiform potentials were found in 12–50%; 43–74% had normal EEGs [8,11]. The timepoint of EEG after a first seizure may be of importance with respect to sensitivity, which has been reported to be highest within 48 h of any seizure [12]. In a large Australian study by [13], EEG
. Figure 153-1 Focal right temporal (RT) interictal spike and wave complexes in a patient with right temporal lobe epilepsy. Anterior-posterior longitudinal bipolar montage. LFF 0.5 Hz, HFF 70 Hz, this and all other figures
EEG in epilepsy
performed within 24 h after a first seizure detected epileptiform abnormalities in 51% of patients, compared with 34% of the patients who had later recording. Two other studies with EEG performed within 48 h of a first seizure reported abnormalities in up to 70% of patients [14,15]. Serial EEG and long-term recordings as well as different activation maneuvers such as hyperventilation and photic stimulation can increase the sensitivity for detection of epileptiform activity and other abnormalities. It is well known that in certain epilepsy syndromes sleep recordings are particularly valuable for increasing sensitivity (for overview see [8,16]), particularly for the detection of 3 Hz spike and wave complexes in primary generalized (idiopathic) epilepsies, focal epileptiform potentials in benign childhood epilepsies (e.g., benign epilepsy of childhood with rolandic spikes or BECRS), and hypsarrhythmia in infants/children [12]. Pathological findings that may be seen during or after hyperventilation include focal slow wave activity, transient asymmetries, and activation of focal and, especially, generalized spike and wave epileptiform abnormalities [17]. The effect of hyperventilation is especially strong in idiopathic generalized epilepsy, and much less so in focal seizure disorders. Intermittent photic stimulation may produce characteristic physiological (photic driving, orbitofrontal myoclonus) or pathological (photoparoxysmal) reactions on EEG [18]. The prevalence of photoparoxysmal responses in the general population is low (0.5–5%; [19]). The risk for developing epilepsy was calculated to be as high as 20% if a generalized photoparoxysmal response was found on EEG [20]. Five percent of all epilepsy patients are photosensitive; in idiopathic generalized epilepsy the percentage of patients with photosensitivity is in the range of 25–35% [21].
Benign Epileptiform Variants It is important to distinguish a number of benign variants that have epileptiform morphologies but
153
no proven association with clinical epilepsy. > Table 153-2 presents the benign patterns that have been identified to date [21,22]. Apart from BETS/SSS and wicket spikes, which are not particularly uncommon, the rest of these patterns are extremely rare.
Periodic Epileptiform Patterns Periodic epileptiform potentials are most commonly seen in association with acute or subacute cerebral structural damage, although they may occasionally be seen in epilepsy, particularly at the end of an episode of status epilepticus, or as the electrographic correlate of epilepsia partialis continua. Focal areas of highly active periodic, repetitive or rhythmic epileptiform activity, which may have a variety of different morphologies, in otherwise well patients with epilepsy, are suggestive of underlying cortical dysplasias [23]. The most common periodic pattern is referred to as PLEDs (periodic lateralized epileptiform discharges), which may be seen in a wide range of brain disorders, usually involving a structural lesion (stroke, encephalitis, abscess, tumor), often in conjunction with metabolic encephalopathies, though occasionally seen after status epilepticus in patients with epilepsy and no new structural lesion or metabolic derangement (> Figure 153-2). PLEDs may persist over a wide range of durations (from seconds to days, weeks and, rarely, even years) and may reflect a continuum of ictal-interictal transitions not always clinically associated with seizures (for overview see [24]). PLEDs preceded by spikes or polyspikes, a pattern known as ‘‘PLEDs-plus,’’ have been associated with an increased likelihood of imminent ictal transition [25]. Other periodic patterns that have been described, rarely seen in epilepsy, include bilateral independent PLEDs (BIPLEDs) and generalized periodic epileptiform discharges (GPEDs), usually related to severe multifocal or diffuse brain
2577
Adults, adolescents
Drowsiness, sleep stages I-II Generalized or lateralized
Brief spikes, diphasic, <50 ms
Main component negative
Low Sporadic
Age
Sleep/Wake states Topography
Morphology
Polarity
Amplitude Time course
Benign Epileptiform Transients of Sleep (BETS)/Small Sharp Spikes (SSS) Adults, adolescents Wakefulness, drowsiness Generalized posterior or anterior maximum Spike and wave-like complexes Main component negative Low Sequence of 1–2 s
6 Hz (‘‘phantom’’) Spike and wave complexes
Spikes, 13–16 Hz and/ or 6–7 Hz
Rhythmic, sharp configuration, 4–7 Hz
Medium Duration of 20 s to few min
Low to medium Sequences of 1–2 s, rarely sporadic
Positive
Posterior unilateral or bilaterally independent
Generalized or lateralized or rarely focal
Wakefulness
Children and adolescents > > adults Drowsiness, sleep
14 and 6 Hz positive spikes
Adults >50 years old
Subclinical Rhythmic Electrographic Discharges of Adults (SREDA) Adults >30 years old Drowsiness, light sleep Temporal unilateral or bilaterally independent Sharp waves, arcade-like, 4–7 Hz Main component negative Medium Brief sequences, rarely sporadic
Wicket spikes
Medium Sequences of few sec
Negative
Rhythmic, sharp configuration, 5–7 Hz
Temporal bilaterally synchronous or independent
Drowsiness > > wakefulness
Adults, adolescents
Psychomotor variant
153
. Table 153-2 Benign epileptiform EEG variants, not associated with epilepsy
2578 EEG in epilepsy
EEG in epilepsy
153
. Figure 153-2 Left hemispheric PLEDs, maximal over the mid-temporal region, in a patient with temporal lobe epilepsy recovering from an episode of focal non-convulsive status epilepticus
damage, as well as periodic sharp waves or triphasic waves in subacute sclerosing panencephalitis and Creutzfeldt-Jakob disease [26].
Video-EEG Simultaneous video-EEG recordings have become a crucial methodology for correlating ictal seizure patterns with clinical seizure semiology, and also for facilitating the accurate identification of electrographic artifacts [27]. Video-EEG is often essential for making the diagnosis of psychogenic non-epileptic seizures (pseudoseizures). Clinical epileptic seizures documented with video recording may have no electrographic correlate on EEG, particularly in the setting of simple partial seizures, whether manifested as isolated auras or brief tonic or clonic events. It has been reported that in simple partial seizures EEG may reveal epileptiform changes in only 15% of events [28]. Ictal epileptiform activity on surface EEG
can be missed in cases of short lasting, deeply located seizures arising in the mesial temporal, frontobasal, frontoparietal or cingulum regions. Prolonged seizures from the same areas may lead to more diffuse, scalp-recordable patterns as seizures propagate to involve larger regions of adjacent and distant cortex. In the case of short lasting, non-propagated seizures without EEG correlates, video evidence of stereotyped clinical manifestations can be invaluable for diagnosis.
Ictal EEG: Generalized EEG Seizure Patterns Characteristically, generalized seizure patterns occur most frequently in idiopathic generalized seizure disorders. One example within this syndrome group is the sterotypic 3 Hz spike and wave complex pattern of absence seizures (> Figure 153-3). However, other generalized seizure patterns occur in the so-called symptomatic
2579
2580
153
EEG in epilepsy
. Figure 153-3 Generalized, bilaterally synchronous, 3 Hz spike and wave activity in a patient with primary (idiopathic) generalized epilepsy. Referential montage; reference = linked ears
generalized epilepsies, such as those associated with cerebral storage diseases, or, classically, in the Lennox-Gastaut syndrome. Generalized ‘‘slow spike and wave complexes’’ (<2 Hz) are typical, often associated with clinical ‘‘atypical absence’’ seizures, with tonic or atonic seizures associated with generalized polyspikes or a diffuse generalized attenuation followed by low amplitude high frequency patterns (for overview see [29]). Occasionally focal seizure disorders may show diffuse, generalized ictal EEG patterns, part of a phenomenon known as secondary bilateral synchrony, most common in cases where the focal seizure generator is situated distant from the cortical surface, especially in the mesial frontal or mesial parietal areas, or the basal frontal region.
Ictal EEG: Focal EEG Seizure Patterns Focal seizure patterns are recorded in cases of cryptogenic or symptomatic focal epilepsy
(i.e., epilepsies with assumed or proven structural lesions), but are also seen in MRI negative idiopathic focal epilepsies, for example BECRS. > Figure 153-4 shows a right temporal focal seizure in a patient with a symptomatic (right mesial temporal sclerosis) focal epilepsy. In drug resistant cases of focal epilepsy the identification of a well-delineated focal seizure generator is a crucial part of the multidisciplinary investigations performed to determine candidacy for resective epilepsy surgery [30]. This is of course most commonly performed for mesial temporal lobe epilepsy, in which a rhythmic seizure pattern in the theta frequency band is most commonly found over the affected temporal lobe [31], however, focal ictal EEG seizure patterns of other morphologies and localizations are equally used to direct epilepsy surgery in other lobes of the brain. So-called bitemporal epilepsy, wherein patients are found to have independent interictal epileptiform potentials arising from both temporal lobes and, occasionally, subclinical
EEG in epilepsy
153
. Figure 153-4 Focal rhythmic seizure pattern localized to right temporal region, same patient as > Figure 153-1
or clinical seizures arising independently from both temporal lobes, is a not uncommon clinical condition. Despite the bilateral involvement, intracranial EEG performed with subdural or depth electrodes can often determine one temporal lobe to be primarily responsible for a patient’s clinical seizures, and a resective surgery thus directed can often be beneficial [32–34]. Complex and widespread propagation of ictal patterns is frequently found in extratemporal lobe epilepsy. In frontal lobe epilepsy, rapid interhemispheric propagation times between 5 and 20 s have been reported [35]. In focal epilepsies with posterior localizations seizure propagation can occur in many ways, primarily ipsilaterally, either above the Sylvian fissure to lateral central, frontal, or supplementary sensorimotor areas or below the Sylvian fissure to temporal structures, but also contralaterally to parietooccipital, temporal and frontal regions. Ictal EEG in these situations of rapid
seizure propagation can be at times misguiding and confusing [36].
Ictal EEG: Mulitfocal EEG Seizure Patterns Mulitfocal seizure patterns on EEG may be the result of diffuse and widespread propagation pathways in patients with seizures generated by just one circumscribed brain region, as mentioned above. Alternatively, they may reflect multiple independent sites of seizure generation, such as may occur in brain disorders with multiple structural lesions such as tuberosis sclerosis or Rasmussen encephalitis, as well as in postencephalitic, posthypoxic, and posttraumatic epilepsies. The development of ‘‘mirror foci,’’ or the process of secondary epileptogenesis, may underlie multifocal seizure patterns in patients originally presenting with unifocal epilepsy [37,38].
2581
2582
153
EEG in epilepsy
. Figure 153-5 Comparison of intracranial interictal and ictal epileptiform activity recorded during sleep with simultaneous scalp EEG. (a) Focal spikes in left and right hippocampus (LH and RH), electrode contacts LHD1 and RHD1, show no scalp EEG correlates; more diffuse right temporal spike and wave complexes (RT) apparent at multiple contacts of right temporal depth electrode (RHD1–4) are associated with visible epileptiform potentials on scalp EEG (channels F8, T4). (b) Focal electrographic seizure in right hippocampus (rhythmic activity at intracranial depth electrode contact RHD1) has no scalp EEG correlate. Referential montage; reference = common average 10–20 electrodes. Top 16 channels = scalp EEG. Channels 17–20 and 21–24 = left and right, respectively, temporal depth electrode recordings. Sensitivity = 15 mV/mm for scalp EEG, 50 mV/mm for intracranial recordings. See text for more details
EEG in epilepsy
Ictal and Interictal EEG Patterns: Comparison with Field Generator Distributions Derived from Intracranial Recordings As alluded to above, intracranial recordings performed through surgically implanted subdural or depth electrodes are often necessary to fully investigate certain patients for potential epilepsy surgery. Other chapters in this book are dedicated specifically to intracranial EEG recording, or electrocorticography, both acute (intraoperative) and chronic. As a link to those chapters, it is worth considering for a moment why intracranial recording may be needed. This question calls for an understanding of the cortical generators underlying the electrical potentials recorded with scalp EEG. The voltage fluctuations in time that are recordable with EEG are generated primarily by summed excitatory and inhibitory post-synaptic potentials present in the superficial dendritic layers of cortical pyramidal cells: summed activity that must occur synchronously across neuronal populations large enough to generate an electrical field detectable with EEG [39]. It has been shown that synchronous activity across a cortical region of at least 6 cm2 is necessary for detection with scalp EEG [39,40]. Indeed, most interictal epileptiform potentials seen with EEG show an intracranial cortical source closer to 10 cm2 [41]. On the other hand, the vast majority of intracranially-evident epileptiform activity, which is frequently confined to small, relatively restricted brain areas, often situated far from the cortical convexities, is not detected with scalp EEG [39,41,42]. > Figure 153-5 provides some ictal and interictal examples of epileptiform activity recorded using simultaneous EEG and intracranial depth electrodes in a patient with bitemporal epilepsy. The two intracranial depth recordings shown were obtained from 4-contact electrodes implanted orthogonally through the second temporal gyrus, bilaterally, with the deepest contact
153
of each electrode situated in the anterior hippocampus and the most superficial contact situated in the lateral temporal neocortex. As can be seen in the figure, focal interictal and ictal epileptiform abnormalities restricted to the deep mesial structures of the temporal lobes have no scalp EEG correlate, whereas the focal spike and wave complexes seen with EEG are associated with widespread intracranial electrical fields. Hence, it must be accepted that scalp EEG will not necessarily ‘‘see’’ small foci of epileptiform activity, even if superficially located, and that deep seated epileptiform activity will remain hidden from scalp EEG. In cases where there is a clinical imperative to localize small or deep seated generators of epileptiform activity, intracranial recording is required.
References 1. Zifkin BG, Cracco RQ. An orderly approach to the abnormal EEG. In: Daly DD, Pedley TA editors. Current practice of clinical electroencephalography. New York: Raven Press; 1990. p. 253-67. 2. Reiher J, Beaudry M, Leduc CP. Temporal intermittent rhythmic delta activity (TIRDA) in the diagnosis of complex partial epilepsy: sensitivity, specificity and predictive value. Can J Neurol Sci 1989;16:398-401. 3. Normand MM, Wszolek ZK, Klass DW. Temporal intermittent rhythmic delta activity in electroencephalograms. J Clin Neurophysiol 1995;12:280-4. 4. Di Gennaro G, Quarato PP, Onorati P, Colazza GB, Mari F, Grammaldo LG, Ciccarelli O, Meldolesi NG, Sebastiano F, Manfredi M, Esposito V. Localizing significance of temporal intermittent rhythmic delta activity (TIRDA) in drug-resistant focal epilepsy. Clin Neurophysiol 2003;114:70-8. 5. Noachtar S, Binnie C, Ebersole J, Mauguie`re F, Sakamoto A, Westmoreland B. A glossary of terms most commonly used by clinical electroencephalographers and proposal for the report form for the EEG findings International Federation of Clinical Neurophysiology. Electroencephalogr Clin Neurophysiol Suppl 1999;52:21-41. 6. Gloor P. Contributions of electroencephalography and electrocorticography to the neurosurgical treatment of the epilepsies. In: Purpura D, Penry J, Walter R, editors, Neurosurgical management of the epilepsies. New York: Raven Press; 1975. p. 59-105.
2583
2584
153
EEG in epilepsy
7. Engel JJ. Intracerebral recordings: organization of the human epileptogenic region. J Clin Neurophysiol 1993;10:90-8. 8. Walczak TS, Jayakar P. Interictal EEG. In: Engel JJ, Pedley TA, editors. Epilepsy: a comprehensive textbook. Philadelphia, PA: Lippincott-Raven; 1997. p. 831-48. 9. Goodin DS, Aminoff MJ. Does the interictal EEG have a role in the diagnosis of epilepsy? Lancet 1984;14:837-9. 10. Salinsky M, Kanter R, Dasheiff RM. Effectiveness of multiple EEGs in supporting the diagnosis of epilepsy: an operational curve. Epilepsia 1987;28:331-4. 11. van Donselaar CA, Schimsheimer RJ, Geerts AT, Declerck AC. Value of the electroencephalogram in adult patients with untreated idiopathic first seizures. Arch Neurol 1992;49:231-7. 12. Sundaram M, Hogan T, Hiscock M, Pillay N. Factors affecting interictal spike discharges in adults with epilepsy. Electroencephalogr Clin Neurophysiol 1990;75:358-60. 13. King MA, Newton MR, Jackson GD, Fitt GJ, Mitchell LA, Silvapulle MJ, Berkovic SF. Epileptology of the firstseizure presentation: a clinical, electroencephalographic, and magnetic resonace imaging study of 300 consecutive patients. Lancet 1998;352:1007-11. 14. Neufeld MY, Chistik V, Vishne TH, Korczyn AD. The diagnostic aid of routine EEG findings in patients with a presumed first-ever provoked seizure. Epilepsy Res 42:197–202. 15. Schreiner A, Pohlmann-Eden B. Value of the early electroencephalogram after a first unprovoked seizure. Clin Electroencephalogr 2003;34:140-4. 16. Dinner DS. Effect of sleep on epilepsy. J Clin Neurophysiol 2002;19:504-13. 17. Kellaway P. An orderly approach to visual analysis: characteristics of the normal EEG of adults and children. In: Daly DD, Pedley TA, editors. Current practice of clinical electroencephalography. New York: Raven Press; 1990. p. 139-99. 18. Kasteleijn-Nolst Trenite´ DG, Guerrini R, Binnie CD, Genton P. Visual sensitivity and epilepsy: a proposed terminology and classification for clinical and EEG phenomenology. Epilepsia 2001;42:692-701. 19. Gregory RP, Smith PT, Rudge P. Electroencephalogram epileptiform abnormalities in candidates for aircrew training. Electroencephalogr Clin Neurophysiol 1993;86:75-7. 20. Kasteleijn-Nolst Trenite´ DG, Binnie CD, Harding GF, Wilkins A. Photic stimulation: standardization of screening methods. Epilepsia 1999;40 Suppl 4:75-9. 21. MacDonald DB. Normal electroencephalogram and benign variants. Neurosciences 2003;8 Suppl 2:110-18. 22. Westmoreland BF, Klass DW. Unusual EEG patterns. J Clin Neurophysiol 1990;7:209-28. 23. Gambardella A, Palmini A, Andermann F, Dubeau F, Da Costa JC, Quesney LF, Andermann E, Olivier A. Usefulness of focal rhythmic discharges on scalp EEG of patients
24.
25.
26.
27.
28.
29.
30. 31.
32.
33.
34.
35.
36.
37. 38.
39.
40.
with focal cortical dysplasia and intractable epilepsy. Electroencephalogr Clin Neurophysiol 1996;98:243-9. Pohlmann-Eden B, Hoch DB, Cochius JI, Chiappa KH. Periodic lateralized epileptiform discharges – a critical review. J Clin Neurophysiol 1996;13:519-30. Reiher J, Rivest J, Grand’Maison F, Leduc CP. Periodic lateralized epileptiform discharges with transitional rhythmic discharges: association with seizures. Electroencephalogr Clin Neurophysiol 1991;78:12-17. Brenner RP, Schaul N. Periodic EEG patterns: classification, clinical correlation, and pathophysiology. J Clin Neurophysiol 1990;7:249-67. Schulz R, Ebner A. EEG-Anfallsmustern a¨hnelnde Artefakte im pra¨chirurgischen Video-EEG-Monitoring. Z Epileptol 2003;16:42-7. Devinsky O, Kelley K, Porter RJ, Theodore WH. Clinical and electrographic features of simple partial seizures. Neurology 1988;38:1347-52. Roger J, Bureau M, Dravet Ch, Dreifuss FE, Wolf P. Epileptic syndromes in infancy, childhood and adolescence. 2nd edn. London: John Libbey; 1992. Hoppe M, Schulz R. Der Stellenwert des EEG in der Epilepsiediagnostik. Klin Neurophysiol 2003;34:147-55. Risinger MW, Engel J Jr, Van Ness PC, Henry TR, Crandall PH. Ictal localization of temporal lobe seizures with scalp/sphenoidal recordings. Neurology 1989;39:1288-93. So N, Gloor P, Quesney LF, Jones-Gotman M, Olivier A, Andermann F. Depth electrode investigations in patients with bitemporal epileptiform abnormalities. Ann Neurol 1989a;25:423-31. So N, Olivier A, Andermann F, Gloor P, Quesney LF. Results of surgical treatment in patients with bitemporal epileptiform abnormalities. Ann Neurol 1989b;25:432-9. Schulz R, Lu¨ders HO, Hoppe M, Tuxhorn I, May T, Ebner A. Interictal EEG and ictal EEG propagation are highly predictive of surgical outcome in mesial temporal lobe epilepsy. Epilepsia 2000;41:564-70. Blume WT, Ociepa D, Kander V. Frontal lobe seizure propagation: scalp and subdural EEG studies. Epilepsia 2001;42:491-503. Boesebeck F, Schulz R, May T, Ebner A. Lateralizing semiology predicts the seizure outcome after epilepsy surgery in the posterior cortex. Brain 2002;125:2320-31. Morrell F. The role of secondary epileptogenesis in human epilepsy. Arch Neurol 1991;48:1221–4. Morrell F, de Toledo-Morrell L. From mirror focus to secondary epileptogenesis in man: an historical review. Adv Neurol 1999;81:11-23. Gloor P. Neuronal generators and the problem of localization in electroencephalography: application of volume conductor theory to electroencephalography. J Clin Neurophysiol 1985;2:327-354. Cooper R, Winter AL, Crow HJ, Walter. Comparison of subcortical, cortical and scalp activity using chronically
EEG in epilepsy
indwelling electrodes in man. Electroencephalogr Clin Neurophysiol 1965;18:217-28. 41. Tao JX, Ray A, Hawes-Ebersole S, Ebersole JS. Intracranial EEG substrates of scalp EEG interictal spikes. Epilepsia 2005;46:669-76.
153
42. Wennberg RA, Lozano AM. Intracranial volume conduction of cortical spikes and sleep potentials recorded with deep brain stimulating electrodes. Clin Neurophysiol 2003;114:1403-18.
2585
163 Hemispherectomy J.- G. Villemure . R. T. Daniel
Historical Background Dandy was the first surgeon to carry out cerebral hemispherectomy in humans [1]. Between 1923 and 1928, he proceeded to hemispherectomy in five patients suffering from a right hemisphere glioma, the objective of the operation being to eradicate the tumor. In his preliminary report, published in 1928, Dandy relates the patients’ histories and physical findings pre- and postoperatively and describes the technique of anatomic hemispherectomy. Dandy insists that mentation is not impaired postoperatively. The hemiplegia, while complete in the arm and leg, is only partial in the face. He reports a patient surviving 3.5 years postoperatively who regained some flexion and extension of the knee and thigh. He goes on to describe that patients develop mild rigidity of the paralyzed limbs following an initial period of flaccidity; deep sensation is preserved, while epicritic and protopathic sensations are lost. Hemianopsia is the rule, and Dandy noted little if any impairment of the cranial nerves. The technique utilized required, as a first step, ligation of the middle and anterior cerebral arteries at the carotid bifurcation, which ‘‘causes a considerable lessening of the cerebral bulk’’; the main veins are ligated. Dandy removed the hemisphere in fragments; first, the frontal lobe was sectioned, requiring ‘‘only a sweep of the scalpel.’’ The corpus callosum was divided; the incision through the internal capsule was then carried through the depth of the temporal lobe. Dandy’s patients expired within 48 h and upto 3.5 years postoperatively from hemorrhage, pneumonia, or tumor progression. Throughout
#
Springer-Verlag Berlin/Heidelberg 2009
his report, he insists that the operation did not create any mental impairment. While hemispherectomy for glioma has been abandoned, Dandy pioneered the technique and demonstrated that half the brain could be removed without intellectual changes and in those of his patients who were already hemiplegic, without physical deterioration. Contrary to what has been reported by Rasmussen [2], Goodman [3] and Davies [4], Jean Lhermitte never did a hemispherectomy. In his 1928 paper on hemispherectomy published in L’Encephale (the same year as Dandy’s report), Lhermitte discusses Dandy’s clinical observations from an anatomophysiological point of view and points out their significance, taking stands which in some instances were completely opposed to some of the current theories [5]. In 1933, W. J. Gardner [6] reported his experience with hemispherectomy for tumor in three cases; two patients are reported to have died of hyperthermia within 36 h following the operation and the other was alive 21 months postoperatively at time of the report. Gardner emphasizes the early postoperative return of function in the affected leg of one patient, so that the patient could flex and extend it, allowing her to walk without assistance within 2 months following surgery. Contrary to Dandy, Gardner did not excise the basal ganglia; the early motor recovery in his patient led him to conclude that the ‘‘function present in the leg is due to basal ganglion innervation.’’ This same patient was discussed again by J. D. O’Brien [7] when, in 1936, he reported
2742
163
Hemispherectomy
deterioration occurring after a fall. The clinical evolution (days or weeks) led to trephination and drainage of a subdural hematoma on the side opposite to hemispherectomy and repeated drainage of bloody spinal fluid from the hemispherectomy cavity; initially these punctures improved the patient’s condition, but this therapy eventually failed and she expired 4 months following the onset of deterioration. This represents the first report of the late complications that may be encountered following anatomic hemispherectomy, in this case, the presence of a subdural hematoma following a fall; it is also possibly the first clinical case of superficial cerebral hemosiderosis (SCH). No autopsy was obtained to document the pathological findings further. We owe to Kenneth McKenzie [8] the credit for having carried out the first hemispherectomy for control of seizures as presented at the American Medical Association Annual Session in 1938. The patient, a female aged 16 years at time of surgery, had sustained a head injury at age 3 weeks and developed a left hemiplegia and seizures. After the ‘‘en bloc’’ hemispherectomy, she showed no clinical aggravation and the seizures stopped. By 1949, a total of 13 cases of hemispherectomy for tumor had been reported in the literature, according to Bell and Karnosh [9] whose paper reviews the major postoperative clinical features encountered in these patients and report a case with a 10-year follow-up. In 1950, R. A. Krynauw reported a series of 12 patients with infantile hemiplegia treated by hemispherectomy [10] for seizures and/or behavior disturbances; two patients without a history of seizures underwent hemispherectomy. In his paper, Krynauw discusses all his 12 cases individually in details relating the history as well as the preoperative, operative, and postoperative findings. The operative technique consisted of dividing the hemisphere into four fragments and dissecting them from within the ventricle.
The results of these operations were dramatic, with complete seizure control in all cases and improved personality and behavior. This paper by Krynauw led many neurosurgical centers in Europe and America to carry out hemispherectomy as well, so that, by 1961, White found 261 cases recorded in the literature [11]. His paper discusses the clinical syndrome of infantile hemiplegia, gives a review of the literature, and presents two cases of his own. During the 1950s, many papers from multiple surgeons were published describing small series of patients operated upon or describing the technique utilized [12–16]. While hemispherectomy consisted of the anatomic removal of the hemisphere, this was done either en bloc or in fragments (> Table 163-1). This choice of technique had no physiological implications, but the issue of leaving or excising the basal ganglia was more controversial. Gardner [6] had already postulated that preservation of the basal ganglia in his cases was responsible for their better motor performance postoperatively, in contradiction to Dandy’s patients, who had these structures removed and did not seem to regain much motor function [1]. Both French [14] in the United States and Laine [17] in France noted no long-term effects of either the ablation or the preservation of the basal ganglia on their patients’ motor performance. Complications such as hemorrhage, infection, and hydrocephalus were reported in the . Table 163-1 Hemispherectomy ‘‘en bloc’’ or in fragments "en bloc’’
Fragments
French McKenzie Obrador Rasmussen White
Cairns Dandy Feld Griffith Krynauw McKissoc Penfield Ransohoff
Hemispherectomy
series published in the 1950s [18] but it was only in the mid-1960s [19], following the recognition of superficial cerebral hemosiderosis as a late complication of anatomic hemispherectomy that this operation lost its popularity. The first clinical report of the condition with pathological correlates was published in the Revue Neurologique by Ulrich in 1965 [20].The same year, Griffith delivered the Hunterian Lecture at the Royal College of Surgeons of England entitled ‘‘Cerebral Hemispherectomy for Infantile Hemiplegia in the Light of the Late Results’’ [21]. Griffith reviewed the Oxford experience of 18 cases operated upon between 1950 and 1961: after a general discussion of the syndrome of infantile hemiplegia, mention of the technique and results, Griffith went on to discuss ‘‘The three cases,’’ who followed a stereotyped clinical pattern after hemispherectomy ‘‘with an initial period of well being, succeeded by a steady deterioration over some years, and at postmortem all have presented a rare picture – that of hydrocephalus and hemosiderosis of the central nervous system.’’ Brief mention of the pathological findings follows, which was the object of a more extensive paper published a year later by Oppenheimer and Griffith [19]. Griffith concluded his report by stating that ‘‘the act of removing the hemisphere may be the important one beginning the Fatal sequence.’’
163
He went on by saying ‘‘I suggest that we are now presented with the opportunity, and indeed the obligation, to try other manoeuvers and modifications of hemispherectomy for these unfortunate children.’’ The concept of disconnection rather than excision began to emerge and Griffith suggested some techniques of undercutting tracts and sparing of the posterior (less epileptogenic) portion of the brain. The paper ‘‘Persistent Intracranial Bleeding as a Complication of Hemispherectomy’’ published in 1966 by Oppenheimer and Griffith [19], describes in great details the clinical and pathological (autopsy and histology) findings encountered in the three patients who died of superficial cerebral hemosiderosis (SCH) (> Figure 163-1). The similarities in these three cases were striking and are summarized as follows by the authors: 1. 2. 3.
4.
Infantile hemiplegia, treated in childhood by hemispherectomy. A trouble-free period lasting for some years. A period of deterioration, extending over several years and ending in death. During this period there was evidence of bleeding into the cerebrospinal fluid pathways and later of obstructive hydrocephalus. Postmortem findings of superficial hemosiderosis of the central nervous system:
. Figure 163-1 Superficial cerebral hemosiderosis following anatomical hemispherectomy demonstrating gliosis and hemosiderin deposits, photomicrographs (courtesy K Meagher-Villemure). (a) Section at the level of the aqueduct (b) Section at level of fourth ventricle
2743
2744
163
Hemispherectomy
chronic granular ependymitis leading to obstruction of cerebrospinal fluid pathways and evidence of multiple bleeding points in the membrane that had replaced the missing hemisphere and in the extension of this membrane onto the lining of the ventricular system.
. Table 163-2 Superficial cerebral hemosiderosis Surgeon
Incidence
Griffith Falconer Ransohoff Rasmussen
3/18 (16.6%) 4/26 (25%) 5/18 (27.7%) 9/27 (33.3%) 21/79 (26.6%)
The authors then discuss the physiopathology of SCH and conclude that repeated spontaneous small hemorrhages from the membranes are responsible for the siderosis and the hydrocephalus, both of which lead to neurological deterioration. These observations led the authors to propose a modification of the hemispherectomy technique by which ‘‘the abnormal hemisphere could be disconnected from the rest of the brain, leaving it in place, with an intact ventricular lining,’’ or to ‘‘attempt to close off the cavity from the ventricles, taking up the space in the cavity with a biologically inert prosthesis’’[19]. Following the description of the clinical pattern and pathological findings of SCH other surgeons reported the occurrence of this late complication following anatomic hemispherectomy. In 1967, Griffith [21] reported that Falconer and Scoville had indicated that they had also encountered the condition. Four years later Falconer and Wilson [18] reported ‘‘Complications related to delayed hemorrhage after hemispherectomy’’: 4 of their 18 patients presented the complication, which they attempted to control by evacuation of the membrane, lavage of the cavity, and shunting procedures. Ransohoff [22] and Rasmussen [23] also reported their experience with SCH. This complication has been found in 15–35% of cases, associated with a high incidence of clinical deterioration and death. The median interval between anatomic hemispherectomy and SCH has been 8 years (> Table 163-2). The ensuing years saw fewer anatomical and more subtotal hemispherectomies, in an attempt
to provide the benefit of the surgical removal of the most epileptogenic brain tissue to control seizures while avoiding the late complications. In keeping with this more conservative surgical approach, new methods of hemispherectomy started emerging in the 1970s, and they are still the object of debate today. These techniques, discussed in another section of this review, are described briefly below. In 1967, Ignelzi and Bucy described the surgical technique of hemidecortication and reported four cases [24].The principle of this operation consists of avoiding the opening of the ventricle but still completely removing the cortex of the affected hemisphere. In 1968, Gibbs and Wilson [25] introduced the concept of modified hemispherectomy popularized by Adams [26], where the hemispherectomy cavity is isolated from the ventricular system by obstructing the foramen of Monro with muscle and made smaller by stripping the dura and stitching it to the falx, tentorium, and basal dura at the expense of creating a large extradural cavity. It was after analyzing the results of subtotal hemispherectomies, done in patients who might have been candidates for hemispherectomy, that Rasmussen noted the absence of the late complication of SCH. This led him to consider subtotal anatomic hemispheric removal, but with complete hemispheric disconnection, which are the principles of functional hemispherectomy [2]. Modifications to these basic techniques of hemispherectomy have been proposed over the past 40 years, but the underlying surgical
Hemispherectomy
principles of each of these operations have been preserved and elaborated. Current techniques of hemispherectomy are discussed in another section of this review.
Rationale and Pathophysiology The use of cerebral hemispherectomy for control of seizures implies that the pathological processes of the epileptogenic brain, the seizure foci, are lateralized to one hemisphere, and that the other hemisphere has preserved its anatomical and physiological integrity. In that respect, hemispherectomy may be considered as the most radical and extensive focal (unilateral) brain excision. While the pathological process responsible for the seizures will have classically produced a contralateral neurological deficit characterized by complete hemiplegia and hemianopsia, one may consider that, from a physiological aspect, the hemispherectomy has already been achieved by the disease process. However, in progressive conditions such as progressive chronic encephalitis, extensive Sturge-Weber syndrome, and in deteriorating condition secondary to infantile spasms, in which we may invariably anticipate a clinical deterioration that will lead to a maximum contralateral neurological deficit or an epileptic encephalopathy, early hemispherectomy may cause an aggravation of the neurological deficit in exchange for complete or improved seizure control, which in turn is accompanied by an important psychosocial improvement. In more than half the patients who are candidates for hemispherectomy, electroencephalography (EEG) demonstrates epileptic activity originating from the good hemisphere; this is either secondary or independent. It is possible that this phenomenon represents an intermediate form of secondary epileptogenesis [27]. In many instances where it appears independent, it disappears following hemispherectomy and the patient
163
may remain seizure-free, off medication [28,29]. However, the presence of independent contralateral hemispheric epileptic activity bears a less favorable prognosis [30]. In most instances of hemispherectomy, the EEG investigation reveals not only one hemispheric focus but multiple independent hemispheric foci. It is for that reason that focal cortical excision is usually not successful in controlling seizures in these instances. The mechanisms by which hemispherectomy controls seizures are of two types: excision and disconnection. Surgical removal of the neurological tissue responsible for the seizures, the cerebral cortex, can be accomplished by anatomic hemispherectomy or hemidecortication. The mechanical eradication of the epileptogenic tissue should be accompanied by a cessation of the seizures. The same objective may be reached by disconnecting the epileptogenic tissue from the effectors, in this instance, the diseased hemisphere, from the rest of the brain i.e., the contralateral hemisphere and the brain stem. This disconnection can be achieved according to the principles of functional hemispherectomy. In this instance, neurons can still generate epileptic potentials but they have nowhere to go because of the disconnection, so that patients remain seizure-free. These epileptic potentials are recorded on posthemispherectomy corticography and also on posthemispherectomy conventional EEG done even years later [28,29]. In cases where seizures with the same or a different clinical pattern persist postoperatively, we may suspect that either the good hemisphere is not as normal as expected and that it is responsible for the seizures, or that the hemispherectomy, either by excision or disconnection, is incomplete [31].
Preoperative Evaluation The decision to proceed to hemispherectomy should be based on a critical preoperative
2745
2746
163
Hemispherectomy
evaluation of the intractability and types of seizures, the etiology of the underlying condition, the neurological examination, the EEG, and the imaging studies [32]. When a patient is being considered for hemispherectomy, usually seizures have not been controlled pharmacologically despite multiple drug trials or the use of combinations of drugs. Seizure frequency will vary from a few per day to over 100 per day in some cases. Seizures are not only refractory to pharmacological treatment but also incapacitating, interfering with normal life, having repercussions on social and psychological development. Most patients who are candidates for hemispherectomy have a focal motor component to their seizure pattern even though they may exhibit many seizure types, i.e., generalized, focal motor, drop attacks, etc. Epilepsia partialis continua is usually characteristic of chronic encephalitis. The etiological factor responsible for the seizures and the neurological deficits are classified in two major groups: congenital or acquired. Congenital etiologies regroup conditions such as infantile hemiplegia from prenatal vascular insult, hemimegalencephaly, diffuse non hypertrophic dysplasia and Sturge-Weber disease, while acquired conditions such as cerebrovascular accident (hemorrhagic or embolic), head injury, cerebral infection, or chronic encephalitis of Rasmussen occur after early normal development. More than 50% of patients suffer from an acquired condition responsible for their seizures (> Table 163-3). The neurological findings on examination are of primary importance in the decision to carry out hemispherectomy. In most instances, the neurological deficit is fixed, static, non-evolving and, generally speaking, characterized by complete hemiplegia and hemianopsia. In these instances, where the patient is unable to do foot tapping or to use the fingers individually, the hemispherectomy will not worsen the motor performance except possibly for temporary hypotonia. In other instances, possible
. Table 163-3 Anatomical substrate in 99 cases of functional hemispehrectmies Substrate Anoxic Rasmussen’s encephalitis Diffuse non hypertrophic dysplasia Hemimegalencephaly Infantile hemiplegia seizure syndrome Infection Sturge Weber Trauma Vascular nb
No. of cases 1 31 6 10 35 1 4 6 5 99
neurological dysfunction is traded for seizure control with the accompanying benefit of psychosocial development. In conditions known to be progressing and where a progressive hemispheric deficit will undoubtedly develop, such as extensive Sturge-Weber disease, active chronic encephalitis and infantile spasms, serious consideration to early hemispherectomy, prior to the development of maximal deficit, should be given. In these cases, especially in the young age group, early hemispherectomy with aggravation of the neurological dysfunction is justifiable for seizure control and improved psychosocial development; the hemiplegia and hemianopsia are easily compensated for. The tone of the affected limbs is usually increased preoperatively secondary to the underlying brain pathology. Postoperatively, tone is either unchanged or markedly reduced with loss of movements for a few days to 1 week, or chronically improved, so that some movements become easier. Voluntary movements are present at proximal joints in the upper and lower extremities in hemiplegic patients pre and postoperatively. Patients who can walk preoperatively but who are unable to perform repeated foot tapping usually walk within a week postoperatively. The sensory examination, which is impaired preoperatively if there is significant hemispheric involvement, remains unchanged postoperatively, with impaired proprioception, stereognosis, and
Hemispherectomy
two-point discrimination. Pain and light touch sensations are present but altered. Mentally, most patients present some degree of mental retardation preoperatively, usually proportional to whether the good hemisphere is involved or not, the frequency and severity of the seizures, and whether the social milieu in which they develop is favorable. The EEG gives evidence of diffuse hemispheric damage characterized by low voltage, slow waves, and multifocal independent epileptic spikes. Depending upon whether there are focal anatomic changes within the damaged hemisphere, these changes may be more pronounced in some areas. Not infrequently (50% of cases) epileptic abnormalities are recorded from the good hemisphere; these may be either secondary or independent. The presence of independent spikes should raise some concerns about the integrity of the good hemisphere but do not represent an absolute contraindication to hemispherectomy. They may represent evidence of secondary epileptogenesis or of diseased brain from the primary condition; posthemispherectomy, they often disappear, remain asymptomatic, but are nevertheless indicative of a less favorable seizure outcome [27–30].
163
Imaging (skull x-ray, computed tomography (CT), or magnetic resonance imaging (MRI) usually demonstrates atrophy characterized by a thick, flattened skull, widened sulci, small gyri, and enlarged ventricles. According to the underlying pathology responsible for the seizures, the radiological findings may be more specific. Superficial calcifications may be seen in cases of SturgeWeber; an enlarged hemisphere and an abnormal gyral pattern may be seen in cases of hemimegalencephaly (> Figure 163-2) and non hypertrophic diffuse hemispheric dysplasia; marked porencephaly is encountered in infantile hemiplegia secondary to prenatal cerebrovascular problems (> Figure 163-3); progressive diffuse hemispheric atrophy may be documented in the active phase of chronic encephalitis with specific pattern of signal and atrophic distribution demonstrated on sequential MRI [33] (> Figure 163-4). The pathology is usually strikingly unilateral and the atrophy, even though it predominates on one side, may also involve the other hemispheres, but to a lesser degree [34]. The preoperative investigation should indicate clearly if the candidate for hemispherectomy is likely to benefit from the operation. Assessment
. Figure 163-2 Hemimegalencephaly. (a) MRI, Coronal T2 demonstrating the enlarged hemisphere, signal abnormality and abnormal gyral pattern (b) MRI, Sagittal T1 demonstrating irregularities of the ventricular wall and abnormal gyral pattern
2747
2748
163
Hemispherectomy
. Figure 163-3 MCA infarct (infantile hemiplegia). MRI, axial T2 (a) and sagittal T1 (b) demonstrating the signal abnormalities corresponding to the vascular territory of MCA, and enlarged ipsilateral ventricle
. Figure 163-4 Rasmussen’s chronic encephalitis, Axial MRI T2. (a) Rt fronto-temporo-insular signal abnormalities in early case (b) Lt hemispheric signal abnormalities and diffuse atrophy in late case
of the criteria discussed above (i.e., seizures, neurological function, etiology. EEC and radiology) should lead to a firm decision without much hesitation.
Ideally, the operation and the choice of the candidate should aim at obtaining a cure of the seizures. There are instances where the investigation predicts improvement in seizure control
Hemispherectomy
without cure: patients with more severe psychomotor retardation, indicating a more diffuse brain involvement, and those where the EEG or the imaging indicates bilateral involvement. We have learned that although these patients may not become seizure-free, they may benefit from a major improvement (over 80%) in their seizure tendency.
Surgical Techniques Historically, the surgical techniques of hemispherectomy are classified in two main categories: the anatomical hemispherectomy with its variations, as practiced from the time of Dandy to that of Griffith, and the ‘‘modifications’’ proposed since the mid-1960s that aim at decreasing the complications rate, namely SCH. These hemispherectomy techniques are of three types: ‘‘hemidecortication,’’ proposed by Ignelzi and Bucy [24] ‘‘modified hemispherectomy,’’ described by Wilson and Adams [26] and ‘‘functional hemispherectomy,’’ described by Rasmussen [2]. The evolution of ‘‘functional hemispherectomy’’ has evolved to emphasize disconnection rather than excision which led to the techniques of hemispherotomy. The surgical methodology of the current hemispherectomy techniques are based on two different principles: excision or disconnection of the epileptogenic tissue. The hemispherectomy techniques consisting in excision are the anatomical hemispherectomy, the hemidecortication and the ‘‘modified’’ hemispherectomy. Hemispherectomy techniques based on disconnection are qualified as ‘‘functional’’ and consist in the classical functional hemispherectomy and the hemispherotomies where the ratio disconnection to excision is the greatest (> Table 163-4). Anatomical hemispherectomy (> Figure 163-5) consists in the anatomic removal of one cerebral hemisphere with or without the basal ganglia: the end result is the creation of a large subdural cavity. The procedure may be done with
163
. Table 163-4 Hemispherectomy techniques according to surgical principles Excision Anatomical hemispherectomy Hemidecortication "Modified hemispherectomy’’ Disconnection Classical fiunctional hemispherectomy Hemispherotomy Vertical Peri-insular (lateral)
resection ‘‘en bloc’’ or in fragments, depending on the surgeon’s preference. All vascular input and output must be interrupted, complete callosotomy carried out, and the corona radiata sectioned rostral to the thalamus. The posteroinferior frontal cortex as well as the medial temporal structures are excised. Complications encountered following this method of hemispherectomy have led many centers to abandon it in its original form [35]. Hemidecortication (> Figure 163-6) consists in the removal of the whole cerebral cortex with sparing of the white matter, thus avoiding opening of the lateral ventricle; this reduces the size of the hemispherectomy cavity created and reduces the mixing of bloody material and debris from the surgery with the ventricular cerebrospinal fluid (CSF). This method of hemispherectomy has been proposed by Ignelzi and Bucy [24] and principally by Hoffman [57] and Kossoff [36]. Some technical amendments to the original technique have been introduced by Hoffman [57] who added plication of the dura and morcellation of the skull in view of reducing the cavity; in the technique used by Welch and reported by Winston and coworkers [37], rather than proceeding piecemeal resection of the cortex with suction or ultrasonic aspirator, large slabs of cortex are undermined and removed. Modified hemispherectomy (> Figure 163-7) as developed by Adams [26] consists of anatomical hemispherectomy followed by occlusion of
2749
2750
163
Hemispherectomy
. Figure 163-5 Anatomical hemispherectomy. (a) intracranial cavity following anatomical hemispherectomy, intra-operative photograph (Courtesy T Rasmussen) (b) ‘‘en bloc’’ hemispherectomy, anatomical specimen (Courtesy T Rasmussen)
. Figure 163-6 Hemidecortication. Schematic representation illustrating the removed cortex (hatched). (Adapted from Ignelzi and Bucy [24])
the ipsilateral foramen of Monro with muscle to prevent communication between ventricular CSF and the hemispherectomy cavity, adding the reduction of the volume of the hemispherectomy cavity by tacking the convexity dura to the falx, the basal dura, and the tentorium, thus creating a large extradural space.
. Figure 163-7 Schematic representation of modified hemispherectomy (From Adams [26], with permission)
Functional hemispherectomy (> Figure 163-8) consists of an anatomical subtotal but physiologically complete hemispherectomy [2,38].The operation is based on principles of disconnection rather than excision. The classic form requires the excision of the central frontoparietal cortex including
Hemispherectomy
. Figure 163-8 Artistic representation of Classical functional hemispherectomy
the parasagittal tissue from the level corresponding to the genu of the corpus callosum to the splenium. A temporal lobectomy with excision of amygdala and hippocampus is carried out. The residual frontal and parieto-occipital lobes are disconnected medially by interrupting all fibers entering the corpus callosum, and proceeding to interruption of all ipsilateral connections by aspirating from within the ventricle all tissue down to the orbital surface of the sphenoid wing in the frontal region and to the tentorium in the posterior part of the hemisphere. Hemispherotomy consists in disconnecting the hemisphere with minimal brain tissue removal. Two approaches have been described. Delalande and colleagues [39–41] proposed a vertical approach where the hemisphere is disconnected through a posterior frontal transcortical approach to the lateral ventricle. From within the ventricle, the callosotomy is carried out, the fornix sectioned, and an incision made through the basal ganglia, reaching the temporal horn (> Figure 163-9). The lateral approach (Peri-insular hemispherotomy) is part of a continuum with the highest disconnection-versus-excision ratio in the technical variations of functional hemispherectomy [42,43] (> Figure 163-10). The hemisphere is disconnected through a peri-insular transventricular approach; the surgical technique
163
has been described in details [44,45] and is here summarized, with the clinical data presented (> Tables 163-5 – > 163-7). The procedure is carried out in three major steps: the suprasylvian window, the infrasylvian window, the insula. The supra-insular window step will allow to disconnect the fronto-parieto-occipital cortices from the ipsilateral structures; this is achieved by removal of the fronto-parieto-temporal opercular cortices and transection of the corona radiata. Once the lateral ventricle is entered, a paramedian callosotomy is carried out from the rostrum to the splenium; this step will disconnect the hemisphere from the contralateral one. Some technical details may here be useful; once in the body of the lateral ventricle, the surgeon will be facing the medial superior wall of the ventricle; the aim is then to visualize the pericallosal cistern; this is done by aspirating the medial tissue corresponding to the cingulate gyrus, in a rostro-caudal manner, till the medial pia is identified; the dissection is carried out caudally to visualize the peri-callosal vessels and underneath, the corpus callosum (CC). Once the CC is identified, the medial ventricular wall tissue is transected toward the splenium posteriorly and the genu anteriorly in a subpial fashion, keeping the vessels and the CC as landmarks. Posteriorly, the callosotomy will go around the splenium; at that point, one is able to identify the choroidal fissure within the trigone. The medial ventricular wall incision is then prolonged to reach the choroidal fissure thus transecting the fimbriafornix and deefferenting part of the hippocampal outflow (posterior hippocampotomy). Anteriorly, the callosal incision at the genu will curve back and caudally to reach the rostrum. Once the paramedian transventricular callosotomy has been completed, one proceed to a coronal incision from within the anterior portion of the lateral ventricle, to reach the posterior orbito-frontal surface at a level corresponding to the sphenoid wing. The callosotomy step
2751
2752
163
Hemispherectomy
. Figure 163-9 Vertical hemispherotomy (Reproduced with permission [39]). Reproduction of the surgical incisions on MRI pictures T1
. Figure 163-10 Peri-insular hemispherotomy – anatomical specimens. Coronal section (a), lateral view (b) demonstrating the supra and infrainsular windows
Hemispherectomy
163
. Table 163-5 Disconnective hemispherectomy – Author’s experience
. Table 163-6 Clinical data in 99 cases of functional hemispherectomies CFH 31, PIH 68 Females 45: Males 54 Age: 4 months to 44 years (mean 12 year, median 8 year) Duration of seizures: 6 months to 43 years (mean 8.6 year, median 4.5 year) Side: left 56, right 43 Follow-up: 6 months to 19 years (mean 7 year, median 6 year)
. Table 163-7 Outcome in 99 cases of functional hemispherectmies Seizure
Engels’ 1 2 3 4
Death Post-op hemorrhage Brain abcess Hydrocephalus
4 1 1 CFH PIH
>6 months follow up 78 3 0 5 nb 86
5/31 (16%) 2/68 (4%) 7/99 (7%)
isolates the hemisphere from the contralateral one; the frontal vertical incision disconnect the frontal lobe from ipsilateral diencephalon. The infrainsular window consists in resection of the temporal opercular cortex, section of the temporal stem from the trigone to the anterior portion of the temporal horn, entering the ventricle; the temporal disconnection is completed by removal of the uncus, including as much of the superomedial portion of the amygdala and the anterior hippocampus up to the point where the choroidal fissure is reached. The hippocampus may either be excised or its output interrupted by the posterior hippocampotomy. We prefer to interrupt the hippocampus at the fimbria-fornix level immediately opposite the inferior portion of the splenium at time of callosotomy. This step isolates the whole temporal cortex from any contralateral or ipsilateral connections. The insular removal has in the past been a matter of question relative to its benefit in seizure control. In analyzing a series of 55 hemispherectomy patients who had the insular cortex removed or spared, we found that there were more seizure-free patients in the group with the insula preserved; this led to think that insular removal was not absolutely necessary [46].
2753
2754
163
Hemispherectomy
At that point in our practice we removed the insula only when corticography done after hemispherectomy demonstrated spiking activity from this cortex. However, after we and others witnessed clinical seizures originating from the insular cortex left after hemispherectomy, insular cortex removal or disconnection has been carried out routinely [47,48]. It has since been demonstrated with a Cox regression analysis of multiple variables (pathology, age at seizure onset, age at surgery, duration of epilepsy, type of surgery, surgeon, residual insular cortex) that the presence of residual insular cortex correlated positively with persistent postoperative seizures [49]. The insular cortex may be removed by aspirating the tissue or by isolating it through an incision carried out at the level of the claustrumputamen from back to front; this is done using bipolar coagulation and aspiration, being cautious of the MCA. During peri-insular hemispherotomy, as many arteries and veins are preserved, and surgery is thus carried out through working channels. This will decrease the risk of infarct, hemorrhagic infarct and brain swelling that could occur post-operatively, taking into account the large volume of tissue not removed. Variations to the PIH technique have been proposed by Schramm [50] who reaches the circular sulcus through a transsulcal approach; Shimizu [51] carries the suprasylvian window as described above, and directs the dissection under the insular cortex to reach and resect the medial temporal structures. In principle, for the same indications, any of the method of hemispherectomy utilized should have an identical impact on seizure control, since these techniques completely eliminate the epileptogenic influence of the diseased hemisphere either by removal or disconnection. When surgery is carried out with a curative objective, the persistence of seizures post-hemispherectomy should be of concern. However, the surgical methodology of hemispherectomy is only one important element that
contributes to seizure control, the other important one being the pathological substrates. This has been documented in analyzing results of pooled series as well as in our own experience [31]; actually, patients whose pathological substrate is clearly unilateral such as infantile hemiplegia, extensive Sturge-Weber, chronic encephalitis will exhibit better seizure control than those suffering from hemispheric migrational disorders [31,36,52–54], trauma or infection. In the latters, some bilateral involvement by the underlying pathology or the development of a secondary encephalopathy may explain that seizure control is not as good.
Complications and Avoidance Complications associated with hemispherectomy can be classified according to the time of occurrence in relation to the surgery, i.e., perioperative, early postoperative, and late postoperative (more than 30 days), their pathological nature or the method of hemispherectmy.
Perioperative Complications The skin incision and bone flap should be planned to avoid the superior sagittal sinus: a bone flap slightly away from the midline allows easy access to the parasagittal region and falx and prevents opening over and damaging the superior sagittal sinus. Bleeding from or thrombosis of the superior sagittal sinus may be dramatic and possibly fatal. The skin incision and bone flap should be as small as possible, varying according to the method of hemispherectomy utilized and influencing anesthesia time and blood loss. Few papers discuss perioperative complications. Brian [55] reported the anesthetic management of 10 hemispherectomy children operated upon over a 5-year period and pointed out that ‘‘massive and sudden blood loss was a common finding. . . Fluid recuscitation frequently was an
Hemispherectomy
ongoing process.’’ These hemorrhages requiring blood replacement were associated with coagulopathy, hypokalemia, and hypothermia. Technically, subpial aspiration is recommended in the medial regions – i.e., in the neighborhood of the optic apparatus and the medial temporal structures as well as at the time of callosotomy to avoid damage to the pericallosal arteries. These risks are further reduced with the utilization of magnification. In hemispherectomy techniques where brain is left in the hemispherectomy cavity, as many arteries and veins as possible should be preserved to assure viability of the tissue and avoid early brain swelling and further brain atrophy [56]. Avoidance of free communication between the ventricle of the operated side with the rest of the ventricular system may prevent some early hydrocephalus. This is accomplished by filling the lateral ventricle with wet cottons and frequent irrigation. Leaving a ventricular drain for 48 h post-operatively helps in washing out blood debris. The best way to avoid perioperative complications is for the surgeon to have a perfect understanding of the three-dimensional surgical anatomy necessary for the method of hemispherectomy utilized; this may be even more important in techniques where the hemisphere is not removed but disconnected. Anticipation of complications remains the key to their avoidance.
Early Postoperative Complications The syndrome of aseptic meningitis is invariably present in the early postoperative period and is discussed under this heading, even though it is not a complication but rather a phenomenon expected to occur. The clinical picture is characterized by lethargy, irritability, and low-grade fever, usually seen following excisions where the ventricle is opened. It is suspected to be secondary to blood products mixing with CSF and creating meningeal irritation. It rarely lasts
163
more than 1 week–10 days and appears to be less severe with surgical techniques with smaller brain volume removal. We do not consider this as an indication for the use of steroids. Failure to control seizures should not be considered a complication of the operation, since this is more the result of selection of patients than of the technique used. Early postoperative seizures occurring within hours do not preclude a good prognosis and may be related to chemical and pharmacological changes associated with the surgery and anesthesia; they may also reflect some degree of epileptogenicity of the opposite hemisphere. The occurrence of early postoperative seizures is however troublesome [31]. Early postoperative wound infection and hemorrhage in the hemispherectomv cavity have been recorded [10]. Considering the extent of the operation, these complications are relatively rare. Meningitis (CSF infection) occurred in up to 5 and 17% of patients respectively in some published series [35,55,57]. Eighteen percent incidence of bone flap infection has been reported with ‘‘modified hemispherectomy’’ [26]. Villemure [57] reported one case of a brain abscess (1%) in the residual brain following functional hemispherectomy successfully treated with antibiotics. Early postoperative brain shift, herniation and death have been reported following anatomical hemispherectomy. Cabiese [58] postulated that this complication was secondary to the development of hydrocephalus of the good hemisphere, combined with displacement of the residual hemisphere toward the hemispherectomy cavity. The creation of a large cavity following anatomical hemispherectomy may predispose to such mechanical shift. Early hydrocephalus following hemispherectomy is manifested by deterioration of neurological function days to a few weeks after the surgery; lethargy and intellectual changes characterize this complication. It has been recorded in 33% of patients undergoing hemispherectomy
2755
2756
163
Hemispherectomy
for Rasmussen’s encephalitis in one series of hemidecortication [59]. Other series of hemidecortication [60] and classical functional hemispherectomy report an incidence of 20 and 16% (> Table 163-7) respectively. In hemispherotomy, the reported incidence of hydrocephalus varies according to the technique utilized. It occurred in 15% of a series of 83 children following vertical hemispherotomy [41], and in 4% of a series of 68 patients who underwent PIH (> Table 163-7). Early hydrocephalus may develop because of the surgical method utilized, which involves, in modified hemispherectomy and hemidecortication, removal of half the supratentorial subarachnoid space and in functional hemispherectomy as much as 35% or as little as 5%, according to technical variations. The possible mixture of blood and blood debris from the hemispherectomy cavity with CSF may contribute to the development of hydrocephalus. It may also result from the underlying cause of the seizures, such as head injury or infection, which may have altered the processes of CSF circulation and absorption.
Late Postoperative Complications Infection, spontaneous or posttraumatic hemorrhages have been reported to occur late following hemispherectomy. The presence of a large empty space following anatomic hemispherectomy has been suspected to favor the development of hematoma after even minor head injury [18]. The main late complication has been SCH discussed earlier in this review [19]. Its occurrence at a median interval of 8 years postoperatively in 15–33% of cases has raised serious concerns about anatomic hemispherectomy. No such cases have been reported with the techniques of hemidecortication, modified hemispherectomy, functional hemispherectomy or hemispherotomy. No cases of SCH have been reported with anatomical hemispherectomy since the early ’70s [61]. What these methods of hemispherectomy have in common as
opposed to anatomic hemispherectomy may be summarized as follows: reduction of ventricular exposure and of the volume of the hemispherectomy cavity.
Mortality Mortality rate following hemispherectomy was reported by White in 1961 to be about 6% [11]. Wilson in 1970 [25] reported a mortality of 39% following anatomic hemispherectomy, taking into account all the deaths secondary to hemispherectomy whether in the early or late postoperative period. In a large series of hemidecortication, a 3% mortality was reported [60] while a 4% incidence is reported in a series of 99 disconnective hemispherectomies (> Table 163-7). Hemispherectomy represents a major surgical procedure and the perioperative period remains critical.
Results For the same indications, the benefit of hemispherectomy should be identical and independent of the choice of technique. For hemispherectomy either by excision (anatomic or hemidecortication) or disconnection (functional hemispherectomy), the variations reside not in the surgical benefit but rather in the incidence of complications. The outcome of the operation should be measured according to its primary objective, i.e., benefits to the patient measured by seizure control and lowest rate of adverse events. Outcome figures must consider the different surgical techniques, the underlying pathology responsible for the seizures, variability in the selection criteria for surgery, types and rate of complications. Long-term improvement in seizure control following hemispherectomy is anticipated in 90–95% of patients. This benefit can be further divided into two categories: those patients who become and remain seizure-free (70–85%) and those who continue to have some seizures but
Hemispherectomy
benefit from at least an 80% reduction in seizure frequency (10–20%). These figures reflect the experience of different surgeons, using different techniques, for different pathologies and are useful for discussion with patients and their families [36,41,44,49–54,61]. Depending on the underlying pathology, selection criteria and surgical experience with a particular method of hemipsherectomy, these results may vary slightly within the general range described. While hemispherectomy is ideally carried out only in the presence of unilateral cerebral seizure onset, it is obvious that in many instances it might not be easy to strictly lateralize the epileptogenicity. For instance, the EEG may show bilateral epileptic abnormalities in over 50% of patients who are otherwise candidates for hemispherectomy; there might be independent contralateral epileptic activity, which has been shown to disappear after hemispherectomy; the condition responsible for the seizures may affect one hemisphere only or both hemispheres but to different degrees [28–30]. Considering these features, it is not surprising to observe that some patients do not become seizure-free. Another reason for persistent seizures is related to the surgeon’s experience with a specific technique and the completeness of the hemispherectomy from a physiological point of view, eliminating by excision or disconnection the influence of the diseased hemisphere [31]. While the control of seizures represents the first and ultimate objective of hemispherectomy, there are secondary benefits from improved seizure control in the sphere of psychosocial development, including improvement in behavior. These are documented by formal measurements of psychosocial functioning as well as by observations by health care professionals and patients’ families. In hemispherectomy patients, Lindsay et al. [62] have been among the first to clearly demonstrate the ill effects of repeated seizures which are accompanied by intellectual deterioration in patients candidates
163
for hemispherectomy. Beardsworth and Adams [63] have documented not only intellectual stabilization following hemispherectomy but also continued intellectual improvement. Recent literature confirm the benefit of early hemispherectomy on intellectual and social development [53,54,64– 66]. These benefits are predominantly secondary to better seizure control, and to a much lesser degree, to the reduction or elimination of anticonvulsant medication. The epileptic encephalopathy that may result from repeated frequent seizure can not only be stabilized but may reverse to different degree following hemispherectomy [67]. The social gains are characterized by better and easier integration in the family and at school. Control of seizures allows patients to develop confidence, interact with others, communicate more easily, and make friends. A minority of patients with strictly unilateral brain damage and a supportive family may, after hemispherectomy be able to earn a living and be autonomous or exceptionally raise a family.
Conclusion Hemispherectomy is a very effective method of controlling pharmacologically refractory seizures originating from a diffusely damaged hemisphere. Its indications depend on the evaluation of the seizures, neurological function, etiological factor responsible for the seizures, imaging, and the EEG investigation. The preoperative evaluation should establish as clearly as possible the degree of damage of the diseased hemisphere and the degree of integrity of the good hemisphere. In well-selected cases, hemispherectomy is among the most effective procedures for the control of refractory seizures. It provides complete seizure control in some 70–85% of patients and almost complete control in 90–95%. These results are independent of the surgical method used which, whether by excision or disconnection, eliminates the epileptic foci of one
2757
2758
163
Hemispherectomy
hemisphere, thus achieving the same end result. The various hemispherectomy techniques differ in the technical strategies used to reduce complications. All techniques except anatomic hemispherectomy share the objective of reducing the volume of the hemispherectomy cavity. For seizure control and improved psychosocial integration, hemispherectomy should be performed as early as possible once a patient is shown to meet the necessary selection criteria. In progressive neurological conditions with intractable epilepsy, whose natural course will lead eventually to a hemispheric syndrome hemispherectomy should be considered before the patient reaches maximal hemispheric deficit or develop an epileptic encephalopathy. Hemispherectomy should be considered early since psychosocial development occurs predominantly in the first decade of life. The surgical technique chosen should expose the patient to the lowest possible complications rate.
References 1. Dandy W. Removal of right cerebral hemisphere for certain tumors with hemiplegia. JAMA 1928;90:823-5. 2. Rasmussen T. Hemispherectomy for seizures revisited. Can J Neurol Sci 1983;10:71-8. 3. Goodman R. Hemispherectomy and its alternatives in the treatment of intractable epilepsy in patients with infantile hemiplegia. Dev Med Child Neurol 1986;28:251-8. 4. Davies KG, Maxwell RE, French LA. Hemispherectomy for intractable seizures: long-term results in 17 patients followed for up to 38 years. J Neurosurg 1993;78:733-40. 5. Lhermitte J. L’ablation comple`te de l’he´misphe`re droit. Ence´phale 1928;23:314-23. 6. Gardner WJ. Removal of the right cerebral hemisphere for infiltrating glioma. JAMA 1933;12:154-64. 7. O’Brien JD. Further report on case of removal of right cerebral hemisphere. JAMA 1936;107:657. 8. McKenzie KC. The present status of a patient who had the right cerebral hemisphere removed. JAMA 1938;111:168. 9. Bell E, Karnosh LJ. Cerebral hemispherectomy. J Neurosurg 1949;6:285-93. 10. Krynauw RA. Infantile hemiplegia treated by removing one cerebral hemisphere. J Neurol Neurosurg Psychiatry 1950;13:243-67.
11. White HH. Cerebral hemipsherectomy in the treatment of infantile hemiplegia. Conf Neurol 1961; 21:1-50. 12. Cairns H. Hemispherectomy in the treatment of infantile hemiplegia. Lancet 1951;2:411-5. 13. Feld M. L’he´misphe´rectomie totale et subtotale: conside´rations de technique ope´ratoire. Rev Neurol 1952;87:525-32. 14. French LA, Johnson DR, Brown IA, Van Bergen FB. Cerebral hemispherectomy for control of intractable convulsive seizures. J Neurosurg 1955;12:154-64. 15. McKissock W. Infantile hemiplegia. Proc R Soc Med 1956;46:431-4. 16. Obrador A. About the surgical technique of hemispherectomy in cases of cerebral hemiatrophy. Acta Neurochir 1952;3:57-63. 17. Laine E, Pruvot P, Osson D. Re´sultats ´eloigne´s de l’he´misphe´rectomie dans les cas d’he´miatrophie ce´re´brale infantile ge´ne´ratrice d’e´pilepsie. Neurochirurgie 1964; 10:507-22. 18. Falconer MA, Wilson PJE. Complications related to delayed hemorrhage after hemispherectomy. J Neurosurg 1969;30:413-26. 19. Oppenheimer DR, Griffith HB. Persistent intracranial bleeding as a complication of hemispherectomy. J Neurol Neurosurg Psychiatry 1966;9:229-40. 20. Ulrich J, Isler W, Vassalli L. L’effet d’he´morragie leptome´ninge´e re´pe´te´es sue le syste`me nerveux. Rev Neurol 1965;112:466-71. 21. Griffith HB. Cerebral hemispherectomy for infantile hemiplegia in the light of late results. Ann R Coll Surg Engl 1967;41:183-201. 22. Ransohoff J, Hess W. Discussion, in Rasmussen T (ed). Post-operative superficial hemosiderosis of the brain: its diagnosis, treatment and prevention. Am Neurol Assoc 1973;98:133–7. 23. Rasmussen T. Post-operative superficial hemosiderosis of the brain: its diagnosis, treatment and prevention. Am Neurol Assoc 1973;98:133-7. 24. Ignelzi RJ, Bucy PC. Cerebral hemidecortication in the treatment of infantile cerebral hemiatrophy. J Nerv Ment Dis 1968;147:14-30. 25. Wilson PJE. Cerebral hemispherectomy for infantile hemiplegia. Brain 1970;93:147-80. 26. Adams CBT. Hemispherectomy: a modification. J Neurol Neurosurg Psychiatry 1983;46:617-9. 27. Morrell F. Varieties of human secondary epileptogenesis. J Clin Neurophysiol 1989;6:227-75. 28. Smith SJM, Andermann F, Villemure JG, Rasmussen T, Quesney LF. Functional hemispherectomy: EEG findings, spiking from isolated brain postoperatively, and prediction of outcome. Neurology 1991;41:1790-4. 29. Wennberg R, Quesney LF, Villemure JG. ECoG findings in hemispherectomy. Electroencephalogr Clin Neurophy 1998;48:132-9.
Hemispherectomy
30. Carmant L, Kramer U, Riviello JJ, Helmers SL, Mikati MA, Madsen JR, Black PM, Lombroso CT, Holmes GL. EEG prior to hemispherectomy: correlation with outcome and pathology. Electroencephalogr Clin Neurophy 1995;94:265-70. 31. Holthausen H, May TW, Adams CBT, Andermann F, Comair Y, Delalande O, Duchowny M, Freeman JM, Hoffman H, May P, Oppel F, Oxbury JM, Peacock WJ, Polkey C, Resnick T, Schramm J, Shewmon DA, Tuxhorn I, Vigevano F, Villemure JG, Wyllie E, Zaiwalla Z. Seizures post-hemispherectomy. In: Tuxhorn I, Holthausen H, Boenigk H, editors. Paediatric epilepsy syndrome and their surgical treatment. London: John Libbey; 1997. p. 749-73. 32. Villemure JG. Hemispherectorhy. In: Resor SR, Kutt H, editors. The medical treatment of epilepsy. New York: Marcel Dekker; 1992. p. 243-9. 33. Granata T, Gobbi G, Spreafico R, Vigevano F, Capovilla G, Ragona F, et al. Rasmussen’s encephalitis: early characteristics allow diagnosis. Neurology 2003;60:422-5. 34. Bien CG, Granata T, Antozzi C, Cross JH, Dulac O, Kurthen M, Lassmann H, Mantagazza R, Villemure JG, Spreafico R, Elger CE. Pathogenesis, diagnosis and treatment of Rasmussen encephalitis: a European consensus statement. Brain 2005;128:454-71. 35. Villemure JG. Hemispherectomy: techniques and complications. In: Wyllie E, editor. The treatment of epilepsy: principles and practice. Philadelphia: Lea & Febiger; 1993. p. 1116-9. 36. Kossoff EH, Vining EP, Pillas DJ, Pyzik PL, Avellino AM, Carson BS, Freeman JM. Hemispherectomy for intractable unihemispheric epilepsy: etiology and outcome. Neurology 2003;61:887-90. 37. Winston KR, Welch K, Adler JR, Erba G. Cerebral hemicorticec-tomy for epilepsy. J Neurosurg 1992;77:889-95. 38. Villemure JG, Rasmussen T. Functional hemispherectomy: methodology. J Epilepsy 1990;3:177-82. 39. Villemure JG, Vernet O, Delalande O. Hemispheric disconnection: callosotomy and hemispherotomy. In Cohadon A, editor. Advances and technical standards in neurosurgery (vol. 26). New York/Wien: Springer; 2000. p. 26–69. 40. Delalande O, Pinard JM, Basdevant C, et al. Hemispherotomy: a new procedure for central disconnection. Epilepsia 1992;33 Suppl 3:99-100. 41. Delalande O, Bulteau C, Dellatolas G, Fohlen M, Jalin C, Buret V, Viguier D, Dorfmuller G, Jambaque L. Vertical parasagittal hemispherotomy: surgical procedures and clinical long-term outcomes in a population of 83 children. Neurosurgery 2007;60:19-32. 42. Villemure JG, Mascott C. Hemispherotomy: the periinsular approach. Technical aspects. Epilepsia 1993;34:48. 43. Villemure J-G, Mascott CR. Peri-insular hemispherotomy: surgical principles and anatomy. Neurosurgery 1995;37:975-81.
163
44. Villemure JG, Daniel RT. Peri-Insular hemispherotomy in paediatric epilepsy. Childs Nerv Syst 2006;22:967-81. 45. Villemure JG, Daniel RT. Functional hemispherectomy and peri-insular hemispherotomy In operative techniques in epilepsy surgery. New York: Thieme; 2009 eds Baltuch 6H, Villemure JG. p. 138-145. 46. Villemure JG, Mascott C, Andermann F, Rasmussen T. Hemispherectomy and the insula. Epilepsia 1989; 34:639. 47. Mascot C, Villemure JG, Andermann F, Rasmussen T. Hemispherectomy and the Insula. Can J Neurol Sci 1990;17:236. 48. Freeman JM, Arroyo S, Vining EP, Breiter SN, Barker PB, Pardo CA, Carson BS, Zuckerberg AL. Insular seizures: a study in Sutton’s law. Epilepsia 1995;35 Suppl 8:49. 49. Cats EA, Kho KH, Van Nieuwenhuizen O, Van Veelen CW, Gosselaar PH, Vasn Rijen PC. Seizure freedom after functional hemispherectomy and a possible role for the insular cortex: the Dutch experience. J Neurosurg 2007;107 Suppl 4:275-80. 50. Schramm J, Kral T, Clusmann H. Transsylvian keyhole functional hemispherectomy. Neurosurgery 2001;49:891-901. 51. Shimizu H, Maehara T. Modification of peri-insular hemispherotomy and surgical results. Neurosurgery 2000;47:367-72. 52. Villemure JG, Meagher-Villemure K, Montes JL, Farmer JP, Broggi G. Disconnective hemispherectomy for hemispheric dysplasia. Epileptic Disord 2003;5 Suppl 2:125-30. 53. Jonas R, Nguyen S, Hu B, Asarnow RF, LoPresti C, Curtiss S, de Bode S, Yudovin S, Shields WD, Vinters HV, Mathern GW. Cerebral hemispherectomy: hospital course, seizure, developmental, language and motor outcome. Neurology 2004;62:1712-1721. 54. Lettori D, Battaglia D, Sacco A, Veredice C, Chieffo D, Massimi L, Tartaglione T, Chiricozzi F, Staccioli S, Mittica A, Di Rocco C, Guzzetta F. Early hemispherectomy in catastrophic epilepsy: a neuro-cognitive and epileptic long-term follow-up. Seizure 2008;17:49-63. 55. Brian JE, Deshpande JK, McPherson RW. Management of cerebral hemispherectomy in children. J Clin Anesth 1990;2:91-5. 56. Daniel RT, Villemure JG. Peri-insular hemispherotomy: potential pitfalls and avoidance of complications. Stereotact Funct Neurosurg 2003;80:22-7. 57. Villemure JG, Adams CBT, Hoffman HJ, Peacock WJ. Hemispherectomy. In: Engel J Jr, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1993. p. 511-8. 58. Cabiese F, Jeni R, Landa R. Fatal brain-stem shift following hemispherectomy. J Neurosurg 1957;14:74-91. 59. Vining EPG, Freeman JM, Brandt J, et al. Progressive unilateral encephalopathy of childhood (Rasmussen’s syndrome): a reappraisal. Epilepsia 1993;34:639-50.
2759
2760
163
Hemispherectomy
60. Kossoff EH, Vining EP, Pyzik PL, Kriegler S, Min KS, Carson BS, Avellino AM, Freeman JM. The post-operative course and management of 106 hemidecortications. Pediatr Neurosurg 2002;37:298-303. 61. O’Brien DF, Basu F, Williams DH, May PL. Anatomical hemispherectomy for intractable seizures: excellent seizure control, low morbidity and no superficial cerebral hemosiderosis. Childs Nerv Syst 2006;22:489-98. 62. Lindsay J, Ounsted C, Richards P. Hemispherectomy for childhood epilepsy: a 36 years study. Dev Med Child Neurol 1987;29:592-600. 63. Beardsworth ED, Adams CBT. Modified hemispherectomy for epilepsy: early results in 10 cases. Br J Neurosurg 1988;2:73-84.
64. Hamiwka L, Duchowny M, Alfonso I, Liu E. Hemispherectomy in early infantile epileptic encephalopathy. J Child Neurol 2007;22:41-4. 65. Basheer SN, Connolly MB, Lautzenhiser S, Sherman EM, Hendson G, Steinbok P. Hemispheric surgery in children with refractory epilepsy: seizure outcome, complications and adaptive function. Epilepsia 2007; 48:133-40. 66. Pulsifer MB, Brandt J, Salorio CF, Vining EP, Carson BS, Freeman JM. The cognitive outcome of hemispherectomy in 71 children. Epilepsia 2004;45:243-5. 67. Duchowny M. Hemispherectomy and epileptic encephalopathy. Epilepsy Curr 2004;4:233-5.
156 Image Guided Epilepsy Surgery Y. G. Comair . R. B. Chamoun
Introduction The history of extratemporal epilepsy, epilepsy surgery, and neurosurgery are intimately related. The intractability of posttraumatic epilepsy led to the early use of trephination as a treatment modality. In the eighteenth century, Percivall Pott, a British neurosurgeon, established that the brain rather than the skull was responsible for the seizures. Posttraumatic scars were often focal and offered guidance for early neurosurgeons into the functional anatomy of the brain. The era of modern neurosurgery was ushered in by Sir Victor Horsley, who introduced asepsis, anesthesia, and the new science of cerebral localization [1]. He also described the technique of subpial cortical dissection and resection, a standard technique used by epilepsy surgeons today. Whereas most temporal lobe epilepsy (TLE) is characterized by hippocampal pathology [2–4], neocortical epilepsy lacks a common pathologic substrate. A wide range of structural anomalies has been associated with chronic partial neocortical epilepsy. These anomalies can be classified into five large categories: malformative, tumoral, ischemic, traumatic, and infectious. In a small number of patients, the nonspecific substrate of gliosis is found. Most of these lesions are not believed to be epileptogenic per se; rather, the epileptogenicity is due to excitability of the surrounding cortex. Numerous mechanisms have been implicated [5]. These include neurochemical changes, pressure effect, and ischemia. In the congenital malformative diseases, intrinsic epileptogenicity has been established with recording of continuous ictal epileptogenic discharges on electrocorticography [6]. Dual or multiple #
Springer-Verlag Berlin/Heidelberg 2009
pathologic substrates are increasingly detected, owing to advances in imaging techniques and careful pathologic studies from large surgical series. The occurrence of hippocampal volume loss in association with pathologies outside of the archipallium are examples of distant but coexistent disease [7,8]. Furthermore, the coexistence of dysplastic and neoplastic disease has been established in developmental tumors, e.g., gangliogliomas and dysembryoplastic neuroepithelial tumors (DNET) [9]. This bears important consequences for the surgical strategies, which are generally designed to resect all identified pathologies for best results.
Indications Principal Candidates Three major groups are identified: The first group consists of lesional, nondevelopmental pathology causing partial epilepsy. A majority of these patients continue to have intractable epilepsy despite antiepileptic drugs (AEDs). A favorable outcome following resective surgery is expected, often with lesionectomy alone. The second group consists of developmental abnormalities causing partial epilepsy and is characterized by a more diffuse pathology and difficulty in precisely localizing the ictal onset zone. Patients in this category are considered for surgical treatment only following failure of AEDs. The third group has partial epilepsy without evidence of a lesion on MRI or metabolic imaging. This is the most difficult subgroup of patients who have less favorable results following
2634
156
Image guided epilepsy surgery
resective surgery. These patients should be investigated in an experienced epilepsy surgery center.
Goals of Surgery The goal of the surgery is to render the patient seizure-free without disturbances in cognitive function, and off or with decreased antiepileptic medications. This goal can be reasonably achieved in patients with localized pathology that is concordant with the electroclinical syndrome. In the less than ideal candidate, when a seizure-free outcome is less likely, surgery is undertaken when there are significant chances of seizure improvement and psychosocial gain.
Presurgical Evaluation The purpose of the presurgical work-up is to identify the epileptogenic zone, which is the cortical region from which seizures arise and for which removal results in cessation of seizures. An accurate history and physical exam, high-resolution MRI, video-EEG monitoring and neuropsychological evaluation are essential components of the initial evaluation. If concordant localization results from this initial step (presence of a single lesion on the MRI that is compatible with the patient’s semiology and interictal & ictal EEG findings) then resective surgery can be planned. In case of non-concordant localization, dual pathology or negative MRI, then further evaluations should be done, including non-invasive (PET, SPECT, PEG. . .) as well as invasive studies (subdural grid, depth electrodes. . .). If a localized epileptogenic focus results from this second step, then resective surgery can be planned; if not, then other therapeutic options should be discussed (vagal nerve stimulation. . .). Another essential part of the work-up is to identify the eloquent cortical areas that might be
included or localized in close proximity to the epileptogenic focus. Multiple techniques are available for this purpose including non-invasive (fMRI. . .) or invasive methods (cortical stimulation intra or extra-operatively).
History, Neurologic and Cognitive Examination In establishing the electroclinical syndrome, the history and physical examination are of paramount importance. The familial and prenatal history yield clues to genetic susceptibility or early insults. Information concerning the natural evolution of the seizure pattern is gathered, particularly the presence of an aura, as well as findings of the various seizure types and response to medication for each type. The semiology of the ictal phenomenon can yield important clues to locate the generator. In some patients, subtle neurologic deficits are noted, which help in the localization process. It is essential to establish the lateralization of speech in neocortical epilepsy because there is a high incidence of unusual speech lateralization in this patient population, particularly with an early age of seizure onset.
Scalp and Invasive Electroencephalography Scalp EEG with interictal and ictal recordings are essential in the localization process of focal epilepsy. Since in neocortical epilepsy the epileptogenic zone is usually larger than in temporal lobe epilepsy, the interictal discharges tend to be widely distributed. In addition, the availability of several pathways of propagation results in a poor localization of the ictal discharges. Generalized ictal onsets are noted in mesial frontal and occipital lobe epilepsy. Lateral frontal and parietal lobe seizures have frequently a localized onset. To further define the ictal onset area, several classes of invasive electrodes have been
Image guided epilepsy surgery
introduced. These consist of electrodes of intermediate invasiveness such as epidural pegs, foramen ovale and more invasive electrodes such as depth, subdural grid, and strip electrodes.
Functional Mapping When the lesion or the epileptogenic zone involves or is adjacent to eloquent cortex, functional mapping should be performed so a safe and complete resection is possible. Mapping is usually performed when resections are contemplated in the central area, dominant inferior frontal cortex, dominant temporal lobe posterior to the precentral sulcus, and the dominant parietal lobe and occipital lobe. A variety of mapping methods is currently available.
Awake Craniotomy Patient cooperation and an anesthesiologist familiar with the technique are essential [10]. This technique provides continuous feedback while the patient is maintained awake during resections adjacent to language areas. An awake craniotomy is difficult to perform in children who, therefore, may require an implanted grid and extraoperative language mapping [11].
Acute Mapping Under General Anesthesia Cortical stimulation can be performed in the primary motor cortex with judicious use of short-acting muscular blockade agents, low concentrations of inhalation agents, and the substitution or addition of propofol, with adequate responses. This may be supplemented by somatosensory evoked potentials to map the pre and post-central gyri [12].
156
Evaluation and Outcome in Relation to Specific Substrates Developmental Malformations of cortical development (MCDs) consist of a group of brain developmental anomalies (For a comprehensive discussion and classification scheme, see [13]). These disorders were originally named according to their gross anatomical appearances: polymicrogyria (PMG), schizencephaly (SZ), lissencephaly (LIS), hemimegalencephaly (HM), gray-matter heterotopia (GH), and focal cortical dysplasia (FCD). Various classifications were introduced with different emphasis. For example, the classification by Barkovich et al. in 1996 subdivided these disorders based mostly on stages of development: disorders of neuronal and glial proliferation, abnormal neuronal migration, and/or cortical organization [14]. Further categorization could be made based on location, extend of disease, and histological findings. The classification by Palmini et al. in 2004 was exclusively based on histology [15]. More recently, information from genetics, molecular biology and neuroimaging were now included in some of the newer classification schemes [13].
Epidemiology The prevalence of MCD in general population and in various patient groups is not precisely known, the reason being that the presentation of MCD could vary widely. Patients with mild forms of MCD (i.e., FCD) might have no symptoms at all or mild symptoms that were undiagnosed. In an autopsy series of 7,374 brains of healthy individuals published in the early 1990s, FCD was found in 1.7% of subjects [16]. In the era of modern imaging, however, more MCD disorders were identified in asymptomatic subjects. In a report of 100 consecutive patients with
2635
2636
156
Image guided epilepsy surgery
a neuroimaging diagnosis of MCD, 33 had normal neurological examination, 61 had partial epileptic syndromes, and13 had secondary generalized syndrome [17]. In patients who had severe signs and symptoms, i.e., developmental delay and/or epilepsy, severe MCD diagnosis of increased severity and at a higher frequency might be expected. For example, in a series of 98 mentally retarded children with epilepsy, 60% were found to have generalized MCD on MRI [18]. The prevalence of MCD in epilepsy patients is estimated via surgical series or imaging. In a study published in 1995, 12% of 341 patients with refractory epilepsy had MRI diagnoses of MCD [19]. Prior to the era of imaging, 3% of patients who underwent surgery at Montreal Neurological Institute for epilepsy were found to have some type of MCD (senior author’s experience). A case series of 133 consecutive resections for extratemporal intractable epilepsy at Cleveland Clinic established MCD as histological diagnosis in 37.6% of cases [20]. MCD is particularly common in the pediatric epilepsy surgery population. In the pediatric UCLA surgical cohort from 1986 to 2005, MCD was identified in 45.5% of operated patients from age 2 months to 19 years [21]. In another study published in 1998 based on patients less than 20 years old, MCD was found in 31 of 120 patients (26%) [22]. Moreover, the diagnosis of MCD was found to be an independent prognostic factor for seizure intractability [23], and patients with MCD often have drug-resistant epilepsies [24].
Presurgical Evaluation: to Determine the Extent of the Epileptogenic zone (EZ) The goal of presurgical evaluation is to determine the extent of the epileptogenic zone (EZ). The determination of the EZ relies on data obtained from clinical examination, seizure semiology, imaging, and electrophysiology. An
essential first step is a demonstration of a MCD on MRI. Although the EZ frequently extends beyond the MRI abnormality when demonstrated, the finding of an abnormality on the MRI greatly facilitates the presurgical evaluation and is considered an essential first step in the delineation of the extent of the epileptogenic zone. However, in a significant number of cases, MRI is negative despite the acquisition of sophisticated sequences and techniques described below. These patients are still considered for surgery with a heavier reliance on invasive electrodes.
Role of Imaging i. MRI
The critical importance of MRI in the diagnosis of MCD and treatment of epilepsy due to MCD has already been noted. It has a direct impact on the establishment of diagnosis, which in turn influences the surgeon’s and clinician’s decision on treatment selections. Furthermore, a recent systematic review of the literature confirmed that an abnormal MRI was among the strongest predictors of seizure outcome after surgery [25]. The MRI evaluation is age-dependent and is modified depending on the extent of the brain myelination. Prior to 10 months of age, heavily T2 weighted spin-echo images are obtained. In older individuals, thin cuts, 1–2 mm, T1 and T2 weighted images are necessary. There are expert recommendations on MRI acquisition techniques and sequences for MCD in general, with a particular emphasis on the importance of FLAIR and T2-weighted sequences [26]. Many reports have described typical MR imaging features of FCD: gyration anomalies, focal thickenings of the cortex, blurring of the grey-white junction, and abnormal cortical and subcortical signal intensity [27,28]. MCDs lead frequently to abnormal gyral and sulcal development of the brain. Therefore, knowledge of the normal sulcal anatomy is essen-
Image guided epilepsy surgery
tial for the detection of subtle anomalies. If the initial MRI investigation is negative, subtle anomalies can be detected by using advanced techniques that include phase array surface coils, 3-D curvilinear reconstruction with post processing morphometry [29,30]. 3-Tesla MRI has demonstrated an increased capability in detection of dysplasia, particularly of the transmantle type. ii. Magnetic Resonance Spectroscopy
The sensitivity and specificity of MR spectroscopy in the detection of MCD has not been assessed in a systematic fashion. Previously, spectroscopy has suffered from a limited spatial resolution and limited sampling of the region of interest. Several studies have demonstrated a decrease in NAA/Cho ratio and NAA/Cr ratio in FCD [31–33]. These findings are nonspecific for cortical dysplasia and have been reported in temporal lobe epilepsy [34,35]. In general the extent of spectroscopic abnormality is larger than the structural imaging abnormality and the utility of MR Spectroscopy in MCD is being assessed in patients whose initial MRI yielded unremarkable results. iii. Tractography and Diffusion Tensor Imaging (DTI)
Abnormalities of proliferation, migration, and layering theoretically should result in anomalies of the white matter tracts of the brain. DTI techniques have demonstrated increased diffusion in a high proportion of MCDs and tractography has shown decreased and aberrant fiber connections with brain areas outside the structural abnormality. This technique is promising and contributes to a better assessment of the extent of the lesion. The abnormal white matter tracts associated with MCD could provide an explanation of the unusual propagation patterns and unexpected interactions between noncontiguous cortical and subcortical sites that has been reported by Duchowny using subdural electrodes [36,37]. iv. PET Scanning
PET scanning has been used in the detection of MCDs prior to the development of high resolu-
156
tion MRI [38]. The FD6-PET hypometabolism demonstrated reflects the functional deficit zones, therefore it includes but is wider than EZ [39]. PET scanning can be very useful in MRI- negative temporal and extratemporal cases [40–41]. Several new ligands are expanding the applications of PET scanning. [11C]-Flumazenil (FMZ) PET binding with the GABA receptor and [11C]-a-methyl-L-tryptophan (AMT) improves the visualization of neocortical epilepsy in MRI negative patients [42–49]. Decreased flumazenil binding appears to correlate better with the EZ than glucose hypometabolism. FMZ may demonstrate secondary epileptogenic foci remote from the primary EZ. Other ligands are being developed (for review see [50]). v. SPECT Scanning
Ictal SPECT scanning with substraction techniques appears as a highly sensitive method in the presurgical evaluation of a patient with cortical dysplasias. It is especially useful in MRI-negative cases; this ranges from 56% [51] to 80% [52]. Its yield depends on early administration of the radiotracer at the onset of the seizure followed by sophisticated analysis methods [53–55]. These requirements have limited the use of ictal SPECT to centers who have developed a specific expertise in its administration and data analysis. vi. EEG fMRI
EEG fMRI is a new investigation modality that combines the high temporal resolution of EEG with exquisite anatomical resolution of MRI. This allows a precise localization of the cortical and subcortical sources of paroxysmal EEG changes. Although reports of this technique are preliminary and have not been applied on a large scale to the study of cortical dysplasia, recent reports have provided important insights into the epileptogenicity of MCDs and interaction between MCD and surrounding cerebral structures in the genesis of epileptic activity [56–61]. These findings are discussed in conjunction with the role of invasive electrodes.
2637
2638
156
Image guided epilepsy surgery
Role of Electrophysiology i. Long-Term EEG Monitoring
The value of surface EEG in the presurgical evaluation of MCD has not been investigated. Dubeau and Palmini reported that two-thirds of patients with FCD have lateralized interictal findings on EEG corresponding to the localization of the malformation [62]. However, bilateral independent interictal activity is not infrequent in children and should not lead to exclusion from presurgical evaluation. Widdess-Walsh, in a series of 48 patients, showed that 64% of patients had a regional surface ictal EEG pattern, 25% had a lateralized onset, and 11% had a nonlocalizable ictal pattern [63]. Foldvary demonstrated that false localization occurs in high frequency in mesial frontal (75%) and mesial parieto-occipital locations (36%) [64]. Ictal b rhythms, when present, are an excellent prognostic factor in frontal lobe epilepsy [65]. ii. Invasive Recordings
The value of invasive EEG recordings in the delineation of the extent of the epileptogenic zone has not been completely settled. In most centers, some form of EEG recording is used to delineate the extent of resection. We will discuss the benefits and limitations of each method. a.
ECOG Palmini in 1991 described distinctive ECOG patterns over dysplastic cortex consisting of frequent ictal trains of continuous spikes [66]. This pattern was present in 67% of patients and correlated well with the anatomical extent of the lesion. This EEG pattern was not present in controlled patients and was more restricted than diffuse interictal discharges recorded in 82% of patients. These findings were confirmed by several authors. Ferrier demonstrated that the continuous ECOG spiking activity of Palmini was also noted in 12% of patients with glioneural tumors [67]. In their series, only 55% of patients with FCD demonstrated continuous spiking. The discharge
b.
pattern did not correlate with the presence or absence of balloon cells on pathology. These studies and others demonstrate that, when present, epileptiform discharges can delineate the extent of the epileptogenic zone and provide the surgeon with an excellent tool in determining the extent of resection. Their persistence post resection would indicate residual dysplastic epileptogenic tissue and, therefore, a poor outcome. However, it has been our experience that ECOG is non-informative in approximately 50% of patients with FCD. Invasive electrodes Two types of invasive electrodes are used in the investigation of MCD: Subdural grid electrodes and stereo EEG using depth electrodes. These both allow mapping of cortical function, therefore correlating ictal discharges with adjacent cortical function. The choice of electrodes is influenced by the center preference and the location of the malformation. Periventricular nodular dysplasias, given their location, are more amenable to depth electrode investigation. At times, a combination of subdural and depth electrodes is indicated.
Subdural Electrodes: The efficacy of subdural electrodes in detecting an ictal onset area in MCD is still controversial. Despite the wide use of these electrodes in MCD, results have not been adequately reported. A study from the Cleveland Clinic on 48 patients who received subdural electrodes demonstrated a diffuse ictal onset zone in 35% of the patients despite wide coverage. An ictal onset involving the edge of the subdural electrode was noted 49% of the time, and two or more separate ictal onset areas were detected in 41% of patients. Complete resection of the ictal onset zone led to a 65% seizure free outcome. Incomplete resection of the ictal onset zone was associated with an 8% seizure free outcome [63].
Image guided epilepsy surgery
Recent advances in EEG are providing clues that could improve the ability of invasive recordings to demonstrate the ictal onset zone in some patients with MCD. Classical recording techniques have used a 200 Hz sampling rate and filtered EEG signal to around 70 Hz. High frequency oscillation (HFO) above 70 Hz, although reported in animal models [68] and in patients with FCDs were not initially perceived as clinically relevant [69]. Several centers have now demonstrated that HFO can be recorded using subdural and depth electrodes and that their depiction appears to approximate closely with the ictal onset area [70]. Ochi, et al., (2007) reported that resection of areas associated with ictal HFO resulted in good surgical seizure free outcomes [71]. Depth Electrodes and SEEG: The results of depth electrode investigations of the major types of MCD have been reported and have improved our understanding of the epileptogenicity of these malformations [57,72–79]. Depth electrodes appear to be superior to subdural electrodes in detecting a focal ictal onset area with 93% efficacy in a large series [72]. This is explained by the ability of depth electrodes to sample tissue that is inaccessible to subdural electrode investigation. Intralesional SEEG recordings from focal cortical dysplasia, in Taylor type and also nonTaylor type, have demonstrated a characteristic interictal pattern of repetitive and rhythmic spikes and polyspike waves [73]. This therefore confirms the high intrinsic epileptogenicity of FCDs noted in Palmini’s reports using ECOG. Depth electrode investigations are also particularly informative in periventricular nodular heterotopias, and several authors consider them to be necessary prior to resective surgery [74]. Epileptogenicity in these lesions involves a complex network that includes the ectopic greymatter, the overlying cortex, and not infrequently the hippocampal formation. Identifying a focal generator is a good predictor for resective surgery in this MCD subtype.
156
Electrode recordings from nodular heterotopias have shown independent epileptiform discharges occurring in the heterotopia and the overlying cortex. In the temporal lobe, epileptiform activity can come from heterotopia, the overlying cortex, and/or mesial temporal structures independently. In some cases, activations are recorded from the heterotopia and surrounding cortex with concomitant activation of a distant cortical area that bears no clear relationship to the lesion. In band heterotopias, wide activation from both the lesion and surrounding cortex were noted confirming the diffuse nature of this lesion. Interestingly, fMRI activation rarely involved the entire heterotopia but was confined to restricted areas within the lesion [76]. These data appear to contradict findings using invasive recordings using SEEG [74]. Focal polymicrogyria have also been investigated using depth electrodes. In this pathology, the ictal onset zone is frequently outside the polymicrogyric cortex and involves the mesial temporal structures. As previously described by Silbergeld [80], wide resections that include the epileptogenic zone and the polymicrogyria cortex, when possible, is recommended [75]. fMRI EEG has clarified the epileptogenicity of polymicrogyria. Previously, invasive recordings had failed to confirm intrinsic epileptogenicity within PMG. fMRI activation was noted within the polymicrogyria in 65% of patients. It also involved a focal area within the BMC in 61% of patients [76]. MEG: MEG offers several advantages over EEG in the determination of the extent of the EZ and its relationship to eloquent cortex in MCD. Since MCD are frequently deep and assume complex configurations, MEG-source modeling allows better determination of the location, time, activities of the dysplastic neurons as compared to EEG. It also allows the display of these activities on MRI imaging facilitation the surgical planning [81]. Several studies using MEG have confirmed intrinsic epileptogenicity of FCD previously
2639
2640
156
Image guided epilepsy surgery
reported by surface EEG recordings [82–84]. In one study comparing MEG with invasive EEG recordings in patients with CD it was revealed that MEG and EEG are complementary as EEG is mostly to superficial currents and MEG detects preferentially tangential current within fissures [85]. In MRI negative patients with CD, MEG investigation reveals subtle abnormalities that are verified pathologically post surgery [86]. MEG plays a role in the localization of essential cortex. Localization of the central sulcus is possible with an accuracy of a few millimeters [87]. In addition, various language paradigms allow mapping of speech areas [86,88–90]. MEG therefore can add critical information to the decision making in the presurgical assessment of patients with MCD. Surgical Outcome in Focal Cortical Dysplasia: FCD is localized by definition and therefore affords the possibility of surgical resection as a treatment option. As a result many more patients with FCD have undergone surgery compared to other types of MCD. A greater availability of surgical specimens enables histological, molecular and genetic studies, which then make correlations with clinical and imaging findings possible. Surgical Outcome: Focal cortical dysplasia is by far the most common MCD identified in surgical specimens. The focal nature of the disease means that patients are more often selected for surgery than those with more diffuse types of MCD. We have reviewed surgical series based on patients who underwent resective surgery and whose histological diagnosis was focal cortical dysplasia [78,91–107]. All 18 series were published since 2000 and all with a follow-up period of at least 1 year. All patients had pre-operative MRI with 1.5-T magnet or better. Studies were included only if seizure-free outcome was reported explicitly. A total of 469 patients and their surgical outcome were available for review. Of 469
patients (59.7%), 280 achieved seizure-free status at 1 year post-operative. This rate of success does not differ between patients who were older than 18 years (60.3%) or younger (59.8%). Of 36 patients (89.4%), 20 of the patients had a lesion that was detectable on MRI, but a considerable percentage of patients who had a normal pre-operative MRI were seizure-free after the surgery. The one variable that most significantly correlates with a seizure-free result was the extent of lesion resection. Of 87 patients (81.6%), 71 were seizure-free after complete resection compared to 23 of 94 patients (24.5%) whose resection was incomplete. The difference was statistically significant (P < 0.001). Temporal lobe location was more favorable than extratemporal location. Of 82 patients (68.3%), 56 who underwent temporal resection within the temporal lobe were seizure-free in contrast to 102 of 204 patients (50.0%) whose resection was extratemporal. The difference was relatively small but statistically significant (P = 0.019). Whether or not invasive monitoring and intraoperative electrocorticography were employed did not affect the surgical results. The most visible and easily classified lesion was the balloon cell. As a result a large number of the studies focused on whether the presence of balloon cells predicts surgical outcome. Conflicting results were found in these surgical series. While some had found the presence of balloon cells was correlated to better outcome [94,95,107], others found the opposite relation [98], or no relation at all (Kloss, Alexandre) [93,103]. As a whole, data from a pool of 210 patients indicate that the presence of balloon cells was not a reliable prognosticator of seizure-free surgical outcome.
Image guided epilepsy surgery
Tumors Brain tumors are found in 15–30% of patients undergoing surgical resection for neocortical epilepsy [108,109]. These tumors can be classified into two main categories: glial (mainly astrocytomas, oligodendrogliomas and oligoastrocytomas) and neuronoglial (mainly gangliogliomas and dysembryoplastic neuroepithelial tumors [DNET].). In the published experience at Cleveland Clinic Foundation over a 17-year period [109], 133 patients underwent extratemporal resections for epilepsy: tumors were found in 27.8% of cases; of these, 27% were low grade astrocytomas, 18.9% gangliogliomas, 16.2% DNET, 16.2% glioneuronal hamartomas, 10.8% oligodendrogliomas, 8.1% anaplastic astrocytomas and 2.7% mixed oligo-astrocytomas. Seizures are more frequently associated with low grade than with high grade tumors, and their occurrence appear to be a favorable prognostic factor [110]. The pathogenesis underlying these seizures remains poorly understood, however multiple mechanisms have been proposed: traditional explanations included impaired vascularisation and ischemic changes in the surrounding cortex [111] and local peritumoral ischemia induced by the mass effect [112]. More recent studies invoke peritumoral disturbance of compounds that alter the membrane potential of neurons [110]: mainly amino acid neurotransmitters (GABA, taurine, aspartate, and glutamate) [113], magnesium [114], and iron ions [115]. Furthermore the pH which modulates neuronal excitability is significantly more alkaline in the peritumoral cortex [116,117]. On the other hand, glioneuronal tumors (gangliogliomas and DNET) may have intrinsic epileptogenic activity. The histopathological hallmark of these tumors is a combination of neuronal and glial cell elements [118].The neuronal component of the tumor itself may contribute to epileptic activity[119].
156
Compared to other lesions, tumors were found to have a higher rate of seizure control after surgery [110]. Several authors have reported a seizure-freedom rate ranging between 66 and 82% [108,111]. Laoprasert et al. [112] reported a better seizure outcome with total compared to partial resection (89% seizure-free with total vs. 63% with partial resection). Given their developmental nature, neuronoglial tumors are frequently associated with cortical dysplasia [113]. This could explain the persistent seizures following some lesionectomies. The role of electrocorticography has not been established as the interictal discharges are frequently diffuse or not prominent at all. In dealing with this pathology, wide resection of the gyrus involved is recommended rather than a pure lesionectomy. If seizures persist, an invasive electrode study is indicated.
Gliotic Substrates Trauma Seizures as a complication of head injury were first reported by Hippocrates. Subsequently, landmarks in understanding the pathophysiology of epilepsy and cerebral localization resulted from the careful studies of meningocerebral scars by pioneers of epilepsy surgery [114,115]. Recent data showed that posttraumatic epilepsy can complicate 25–30% of cases of severe head injury and 5–10% of cases of mild to moderate injury [116]. It is a common cause of epilepsy, accounting for approximately 4% of focal epilepsy in the general population, and is the leading cause of epilepsy with onset in young adults (15–24 years of age) [117]. Two types of head injury are distinguished – penetrating and nonpenetrating. A variety of pathologic changes is noted following blunt or nonpenetrating injury to the brain. These include axonal damage, intracerebral hematomas,
2641
2642
156
Image guided epilepsy surgery
ischemic parenchymal changes, and contusion frequently involving the orbitofrontal cortex and the basal and anterior temporal lobes. The incidence of seizures following blunt injury is variable [118]. Due to the diffuse nature of closed head injury, localization of the epileptogenic cortex is frequently difficult. Dual pathologies can occur with the presence of orbitofrontal and associated mesial temporal epilepsy [7]. Patients are considered for surgery when there is a good correlation between the electroclinical syndrome and pathologic changes on MR. Wide cortical resections are recommended to ensure a good surgical outcome. Invasive electrodes are frequently used in these cases.
Vascular Malformations Vascular malformations are increasingly being detected in patients with localization-related epilepsy. These usually fall into three categories: cavernous angiomas, arteriovenous malformations (AVMs), and venous angiomas. Cavernous angiomas could be considered as one of the more focal models of epilepsy in humans. Seizures are their most common presenting symptom, observed in 38–100% in different reported series [119,120], and they probably result from the toxic effects of iron deposition in the form of hemosiderin [115,121]. Although the pathophysiologic mechanism appears to be very focal, several series have recorded suboptimal seizure outcome following lesionectomies. This could be due to incomplete resection of the hemosiderin-impregnated area or the existence of dual pathology, specifically when located in the mesial temporal lobe. An accepted management strategy is to resect the surrounding damaged cortex until normal cortex is identified. In primary cortex or speech areas, surgery is generally performed under local anesthesia.
The role of radiosurgery remains questionable and its results regarding seizure control are inferior to those of surgery: Regis et al. reported a seizure freedom rate of 53% [122], and Shih et al. in a comparative study between craniotomy and radiosurgery found better results with craniotomy (79% seizure free with craniotomy vs. 25% with radiosurgery) [123]. Furthermore many authors reported no effect of radiosurgery on the bleeding risk associated with cavernous malformations [124,125]. In the case of AVMs, several mechanisms for the pathogenesis of epilepsy have been advanced, including: previous hemorrhage resulting in gliosis or hemosiderin, perilesional neurochemical changes and focal cerebral ischemia due to a steal phenomenon [126–128]. A number of earlier studies showed a poor seizure outcome following surgery with some even reporting an increase in seizure frequency [129–132], however more recent series documented postoperative seizures in less than 40% of patients with a history of preoperative seizures and less than 10% in those without a history preoperative seizures [133,134]. A recently published large series [135] found that there was more than 50% reduction in seizure frequency from the preop incidence. Furthermore, 75% and 83% of patients were seizurefree postop in two recent surgical series [133,136]. On the other hand, seizure freedom rates ranging between 51% and 80% are reported in recent series of gamma knife radiosurgery [137–139]; the mechanism of the antiepileptic effect of radiosurgery being unknown. Venous angiomas are rarely associated with epilepsy. They are at times discovered during the surgical workup and are frequently unrelated to the seizures. Resection of these lesions can yield to hemorrhagic infarcts; thus, surgical excision is not recommended.
Image guided epilepsy surgery
Evaluation and Outcomes in Relation to Localization Related Epilepsy Frontal Lobe Resections The frontal lobe encompasses one-third of the cerebral cortex volume, yet, despite its large size and its high epileptogenicity, the frequency of surgically treated frontal lobe epilepsy is small as compared with temporal lobe epilepsy. This varies between 18% [140,141] and 5.5% in large reported series. The manifestations of frontal lobe epilepsy are protean [142,143] and can be divided into three major groups [144]: supplementary motor type, complex partial seizures, and focal motor seizures. Five electroclinical syndromes [140] are distinguished: frontopolar, orbitofrontal, dorsolateral, supplementary motor area, and cingulate. The localization of an ictal onset area is difficult in frontal lobe epilepsy and is facilitated when a lesion is present on imaging [140]. In nonlesional cases, chronic recordings are recommended using subdural electrodes.
Frontal Lobectomy When the epileptogenic zone is diffuse, a complete frontal lobectomy anterior to the precentral sulcus can be performed under general anesthesia. The patient is positioned supine, with the head slightly rotated to the contralateral side. A large C-shaped skin incision is fashioned, followed by a large craniotomy extending to the midline. The location of the central sulcus is determined using somatosensory evoked potentials or cortical stimulation. The resection initially is carried one gyrus anterior to the precentral sulcus. This is subsequently removed in a subpial fashion, taking care not to undercut the white matter tracts coursing from the precentral gyrus; the ascending venous system draining the central
156
area is spared. On the dominant hemisphere, the Broca’s area should be identified and preserved. This is best performed under local anesthesia or with the use of a subdural grid. The complications of a complete frontal lobectomy are rare. Hemiplegia occurs in 0.5% of reported cases. There are no significant neurophysiologic deficits. In the Montreal Neurological Institute series reported in 1975 [145], 55% of patients benefited from a frontal lobe resection, with 23% becoming completely seizure free. In a more recent series [140], 68% became seizure free following frontal lobe resection. In this series, chronic EEG recordings were found to be helpful in planning surgery but had limited sensitivity and specificity.
Supplementary Motor Area The supplementary motor area (SMA), as defined by Penfield and Welch [146], is located in the mesial superior frontal cortex anterior to the primary motor cortex of the lower extremity and superior to the cingulate gyrus. Functional studies have shown that this area is activated during initiation of movement and vocalization. Stimulation of this area leads to a fencing posture with bilateral motor movement. Unilateral responses are rare. Resection of this area leads to transient contralateral weakness and apraxia. The SMA is extensively and somatotopically connected through the corpus callosum; this results in quick spread of the ictal discharges to the contralateral side, making lateralization of the ictal onset zone difficult. When the ictal onset area is not clearly defined, bilateral subdural electrodes are placed within the interhemispheric fissure. These are also used to delineate the primary motor cortex. There is considerable anatomic variation in the precentral sulcus and its relationship with the marginal ramus. On long-term follow-up, no gross motor deficits are noted with this resection.
2643
2644
156
Image guided epilepsy surgery
Orbitofrontal Resections The orbitofrontal area is limited laterally by the orbitofrontal sulcus, medially by the olfactory sulcus, anteriorly and superiorly by the frontomarginal sulcus, and posteriorly by the anterior perforated area. The orbitofrontal cortex is extensively connected with the anterior and mesial temporal lobe, cingulum, and opercular area. For this reason, orbitofrontal seizures are frequently misdiagnosed as anterior temporal [147]. Adequate sampling of these structures using invasive electrodes is recommended. On the nondominant side, extensive resection of the orbitofrontal cortex can be performed. The intersection of the optic nerve and olfactory nerve is used as the posterior limit of the resection. On the dominant side, mapping of Broca’s area should be performed.
Resections in the Primary Motor and Sensory Cortex Central type epilepsies are rare, and their surgical management remains a real challenge because of the risk of a permanent neurologic deficit. However, we recently reported our series of 24 patients operated at the American University of Beirut between 1997 and 2003, where we showed that selective resection in the PSMC can be performed and the neurologic deficit observed in the immediate postop status is transient [148]. Following selective motor resections, severe motor deficit occurred in the immediate postoperative period, gradual improvement started 1 month postop on average and substantial improvement was found at 6 months of follow-up. Long term sequelae included impaired fine hand movements after motor hand area resection, and no recovery of toe movements after motor leg area resection. Following sensory resections, severe sensory deficit occurred in the immediate postoperative
period but progressively improved over 2–3 months. Long term sequelae included impaired position sense after sensory hand area resection. The average period of follow-up was 4.6 years, and 18 of 24 patients remained seizure-free at last follow-up. Functional recovery following injury to the PSMC has been studied in experimental animal studies [149], and in humans following strokes (mainly using fMRI [150] and PET [151]). The main proposed mechanisms for recovery involve reorganization within the PSMC [149,150], recruitment of non-primary motor areas [149,151] and possibly recruitment of the contralateral PSMC [152,153].
Parietal Lobe Resections Seizures originating in the parietal lobe account for up to 6% of reported series. A large proportion of patients exhibit an aura, most commonly somatosensory [154]. Pain, vertiginous sensations, aphasia, or disturbance of body image are suggestive of parietal origin. The ictal manifestations are varied and reflect the quick spread to the frontal lobe in superior parietal epilepsies and to the temporal lobe in inferior parietal cases [155]. Interictal and ictal scalp EEG recordings were not reliable markers for parietal lobe epilepsy.
Occipital Lobe Epilepsy Occipital lobe seizures are rare, representing 1% of the MNI series [156]. Early clinical manifestations of elementary visual hallucinations, ictal amaurosis, eye movement sensations, and blinking are highly suggestive of an occipital origin [157]. In infracalcarine cases, quick spread to the temporal lobe can produce symptomatology typical of mesial temporal lobe epilepsy.
Image guided epilepsy surgery
Various spread patterns have been described and can make diagnosis difficult. The presence of a superior quadranopsia or posterior temporal interictal and ictal discharges should be suggestive of occipital lobe epilepsy. On the dominant hemisphere, the speech-related cortex should be identified and spared. The management of patients with intact vision is challenging. When a circumscribed lesion is found, lesionectomy can yield satisfactory results. In nonlesional cases, the ictal onset area should be precisely localized using invasive electrodes. These are used in addition to mapping of the calcarine cortex and speech-related cortex. With this strategy, visual deficits can be minimized. Resections of the dominant basal temporal lobe should be carefully planned as this can yield to an alexia without agraphia deficit. Good results have been reported in 65–80% of patients with occipital lobe epilepsy in recent series. In the MNI series of 37 patients, 46% became seizure free, 21% had rare seizures, and 10% had a worthwhile improvement.
Summary and Conclusions The management of neocortical epilepsy relies on the identification of a pathologic substrate, which will guide the electrophysiologic evaluation. Circumscribed cortical resections are performed when the pathologic substrate is focal and concur with an electroclinical syndrome. When the pathologic substrate is diffuse or nonconcordant, electrophysiologic invasive electrodes should be used to determine the volume of resection. The outcome following neocortical resections has significantly improved with the introduction of high-resolution MRI. However, it is still not optimal. Advances in metabolic imaging could yield to a more precise definition of the epileptogenic cortex.
156
References 1. Horsley V. Brain surgery. Br Med Assoc 54th Annual Meeting 1886;2:670-75. 2. Berkovic SF, Andermann F, Olivier A, et al. Hippocampal sclerosis in temporal lobe epilepsy demonstrated by magnetic resonance imaging. Ann Neurol 1991;29:175-82. 3. Kuzniecky R, Burgard S, Faught E, Morawetz R, Bartolucci A. Predictive value of magnetic resonance imaging in temporal lobe epilepsy surgery. Arch Neurol 1993;50:65-9. 4. Kuzniecky R, de la Sayette V, Ethier R, et al. Magnetic resonance imaging in temporal lobe epilepsy: pathological correlations. Ann Neurol 1987;22:341-7. 5. Robitaille Y, Rasmussen T, Dubeau F, Tampieri D, Kemball K. Histopathology of nonneoplastic lesions in frontal lobe epilepsy. Review of 180 cases with recent MRI and PET correlations. Adv Neurol 1992;57:499-513. 6. Palmini A, Gambardella A, Andermann F, et al. Intrinsic epileptogenicity of human dysplastic cortex as suggested by corticography and surgical results. Ann Neurol 1995;37:476-87. 7. Mathern GW, Pretorius JK, Babb TL. Quantified patterns of mossy fiber sprouting and neuron densities in hippocampal and lesional seizures. J Neurosurg 1995;82:211-9. 8. Raymond AA, Fish DR, Sisodiya SM, Alsanjari N, Stevens JM, Shorvon SD. Abnormalities of gyration, heterotopias, tuberous sclerosis, focal cortical dysplasia, microdysgenesis, dysembryoplastic neuroepithelial tumour and dysgenesis of the archicortex in epilepsy. Clinical, EEG and neuroimaging features in 100 adult patients. Brain 1995;118 (Pt 3): 629-60. 9. Khajavi K, Comair YG, Prayson RA, et al. Childhood ganglioglioma and medically intractable epilepsy. A clinicopathological study of 15 patients and a review of the literature. Pediatr Neurosurg 1995;22:181-8. 10. Archer DP, McKenna JM, Morin L, Ravussin P. Conscious-sedation analgesia during craniotomy for intractable epilepsy: a review of 354 consecutive cases. Can J Anaesth 1988;35:338-44. 11. Alvarez L, Jayakar P. cortical stimulation with subdural electrodes: special consideration in infancy and childhood. In: Duchowny M, Resnick T, Alvarez L, editors. Pediatric epilepsy surgery. New York: Demos; 1990. p. 125-30. 12. Wood CC, Spencer DD, Allison T, McCarthy G, Williamson PD, Goff WR. Localization of human sensorimotor cortex during surgery by cortical surface recording of somatosensory evoked potentials. J Neurosurg 1988;68:99-111. 13. Barkovich AJ, Kuzniecky RI, Jackson GD, et al. A developmental and genetic classification for malformations of cortical development. Neurology 2005;65:1873-87.
2645
2646
156
Image guided epilepsy surgery
14. Barkovich AJ, Kuzniecky RI, Jackson GD, et al. A classification scheme for malformations of cortical development. Neuropediatrics 1996;27:59-63. 15. Palmini A, Najm I, Avanzini G, et al. Terminology and classification of the cortical dysplasias. Neurology 2004;62:S2-S8. 16. Mencke HJ. Molecular neurobiology of epilepsy. Amsterdam: Elsevier; 1992. 17. Montenegro MA, Cendes F, Lopes-Cendes, I et al. The clinical spectrum of malformations of cortical development. Arq Neuropsiquiatr 2007:65;196-201. 18. Steffenburg U, Hedstrom A, Lindroth A, et al. Intractable epilepsy in a population-based series of mentally retarded children. Epilepsia 1998:39;767-75. 19. Li LM, Fish DR, Sisodiya SM, et al. High resolution magnetic resonance imaging in adults with partial or secondary generalized epilepsy attending a tertiary referral unit. J Neurol Neurosurg Psychiat 1995:59;384-7. 20. Frater JL, Prayson RA, Morris HH, et al. Surgical pathologic findings of extratemporal-based intractable epilepsy: a study of 133 consecutive resections. Arch Pathol Lab Med 2000;124:545-9. 21. Cepeda C, Andre VM, Levine MS, et al. Epileptogenesis in pediatric cortical dysplasia: the dysmature cerebral developmental hypothesis. Epilepsy Behav 2006;9:219-35. 22. Wylie E, Comair YG, Kotagal P, et al. Seizure outcome after epilepsy surgery in children and adolescents. Ann Neurol 1998;44:740-8. 23. Semah F, Picot MC, Adam C, et al. Is the underlying cause of epilepsy a major prognostic factor for recurrence? Neurology 1998;51:1256-62. 24. Raymond AA, Fish DR, Sisodiya SM, et al. Abnormalities of gyration, heterotopias, tuberous sclerosis, focal cortical dysplasia, microdysgenesis, dysembryoplastic neuroepithelial tumour and dysgenesis of the archicortex in epilepsy: clinical, EEG and neuroimaging features in 100 adult patients. Brain 1995;118:629-60. 25. Tonini C, Beghi E, Berg AT, et al. Predictors of epilepsy surgery outcome: a meta-analysis. Epilepsy Res 2004;62:75-87. 26. Ruggieri PM, Najm I, Bronen R, et al. Neuroimaging of the cortical dysplasias. Neurology 2004;62:S27-S29. 27. Raybaud C, Shroff M, Rutka JT, et al. Imaging surgical epilepsy in children. Childs Nerv Syst 2006;22:786-809. 28. Widdess-Walsh P, Diehl B, Najm I. Neuroimaging of focal cortical dysplasia. J Neuroimaging 2006;16:185-96. 29. Wilke M, Kassubek J, Ziyeh S, et al. Automated detection of gray matter malformations using optimized voxel-based morphometry: a systematic approach. Neuroimage 2003;20:330-43. 30. Kassubek J, Huppertz HJ, Spreer J, et al. Detection and localization of focal cortical dysplasia by voxel-based 3-D MRI analysis. Epilepsia 2002;43:596-602. 31. Kuzniecky R, Hetherington H, Pan J, et al. Proton spectroscopic imaging at 4.1 tesla in patients with
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
malformations of cortical development and epilepsy. Neurology 1997;48:1018-24. Munakata M, Haginoya K, Soga T, et al. Metabolic properties of band heterotopia differ from those of other cortical dysplasias: a proton magnetic resonance spectroscopy study. Epilepsia 2003;44:366-71. Kaminaga T, Kobayashi M, Abe T. Proton magnetic resonance spectroscopy in disturbances of cortical development. Neuroradiology 2001;43:575-80. Ng TC, Comair YG, Xue M, et al. Temporal lobe epilepsy: presurgical localization with proton chemical shift imaging. Radiology 1994;193:465-72. Kuzniecky R, Palmer C, Hugg J, Martin R, Sawrie S, Morawetz R, et al. Magnetic resonance spectroscopic imaging in temporal lobe epilepsy: neuronal dysfunction or cell loss? Arch Neurol 2001;58:2048-53. Turkdogan D, Duchowny M, Resnick T, Jayakar P. Subdural EEG patterns in children with taylor-type cortical dysplasia: comparison with nondysplastic lesions. J Clin Neurophysiol 2005;22:37-42. Duchowny M, Jayakar P, Levin B. Aberrant neural circuits in malformations of cortical development and focalepilepsy. Neurology 2000;55:423-8. Chugani HT, Shields WD, Shewmon DA, et al. Infantile spasms: I. PET identifies focal cortical dysgenesis in cryptogenic cases for surgical treatment. Ann Neurol 1990;27:406-13. Lee N, Radtke RA, Gray L, et al. Neuronal migration disorders: positron emission tomography correlations. Ann Neurol 1994;35:290-7. Sood S, Chugani HT. Functional neuroimaging in the preoperative evaluation of children with drug-resistant epilepsy. Childs Nerv Syst 2006;22:810-20. da Silva EA, Chugani DC, Muzik O, Chugani HT. Identification of frontal lobe epileptic foci in children using positron emission tomography. Epilepsia 1997;38:1198-208. Savic I, Persson A, Roland P, Pauli S, Sedvall G, Wide´n L. In-vivo demonstration of reduced benzodiazepine receptor binding in human epileptic foci. Lancet 1988;2:863-6. Henry TR, Frey KA, Sackellares JC, et al. In vivo cerebral metabolism and central benzodiazepine-receptor binding in temporal lobe epilepsy. Neurology 1993;43: 1998-2006. Juha´sz C, Chugani DC, Muzik O, et al. Relationship of flumazenil and glucose PET abnormalities to neocortical epilepsy surgery outcome. Neurology 2001;56: 1650-8. Asano E, Chugani DC, Muzik O, et al. Multimodality imaging for improved detection of epileptogenic foci in tuberous sclerosis complex. Neurology 2000;54:1976-84. Juha´sz C, Chugani HT Imaging the epileptic brain with positron emission tomography. Neuroimaging Clin N Am 2003;13:705-16.
Image guided epilepsy surgery
47. Fedi M, Reutens D, Okazawa H, et al. Localizing value of alpha-methyl-L-tryptophan PET in intractable epilepsy of neocortical origin. Neurology 2001;57:1629-36. 48. Juha´sz C, Chugani DC, Muzik O, et al. Electroclinical correlates of flumazenil and fluorodeoxyglucose PET abnormalities in lesional epilepsy. Neurology 2000;55:825-35. 49. Muzik O, da Silva EA, Juhasz C, et al. Intracranial EEG versus flumazenil and glucose PET in children with extratemporal lobe epilepsy. Neurology 2000;54:171-9. 50. Luat AF, Chugani HT. Molecular and diffusion tensor imaging of epileptic networks. Epilepsia 2008;49 Suppl 3:15-22. 51. Won HJ, Chang KH, Cheon JE, et al. Comparison of MR imaging with PET and ictal SPECT in 118 patients with intractable epilepsy. AJNR Am J Neuroradiol 1999;20:593-9. 52. O’Brien TJ, So EL, Cascino GD, Hauser MF, Marsh WR, Meyer FB, et al. Subtraction SPECT coregistered to MRI in focal malformations of corticaldevelopment: localization of the epileptogenic zone in epilepsy surgery candidates. Epilepsia 2004;45:367-76. 53. Cascino GD, Buchhalter JR, Sirven JI, et al. Peri-ictal SPECT and surgical treatment for intractable epilepsy related to schizencephaly. Neurology 2004;63:2426-8. 54. Wetjen NM, Cascino GD, Fessler AJ, et al. Subtraction ictal single-photon emission computed tomography coregistered tomagnetic resonance imaging in evaluating the need for repeated epilepsy surgery. J Neurosurg 2006;105:71-6. 55. Cascino GD, Buchhalter JR, Mullan BP, So EL. Ictal SPECT in nonlesional extratemporal epilepsy. Epilepsia 2004;45 Suppl 4:32-4. 56. Detre JA, Crino PB. A multilayered approach to studying cortical malformations: EEG-fMRI. Neurology 2005; 64:1108-10. 57. Kobayashi E, Bagshaw AP, Jansen A, et al. Intrinsic epileptogenicity in polymicrogyric cortex suggested by EEG-fMRI BOLD responses. Neurology 2005;64:1263-6. 58. Kobayashi E, Bagshaw AP, Grova C, Gotman J, Dubeau F. Grey matter heterotopia: what EEG-fMRI can tell us about epileptogenicity of neuronal migration disorders. Brain 2006;129:366-74. 59. Salek-Haddadi A, Friston KJ, Lemieux L, Fish DR. Studying spontaneous EEG activity with fMRI. Brain Res Rev 2003;43:110-33. 60. Diehl B, Salek-haddadi A, Fish DR, Lemieux L. Mapping of spikes, slow waves, and motor tasks in a patient with malformation of cortical development using simultaneous EEG and fMRI. Magn Reson Imaging 2003;21:1167-73. 61. Gotman J, Be´nar CG, Dubeau F. Combining EEG and FMRI in epilepsy: methodological challenges and clinical results. J Clin Neurophysiol 2004;21:229-40. 62. Dubeau F, Palmini A, Fish D, et al. The significance of electrocorticographic findings in focal cortical dysplasia: a review of their clinical, electrophysiological and
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
156
neurochemical characteristics. Electroencephalogr Clin Neurophysiol Suppl 1998;48:77-96. Widdess-Walsh P, Jeha L, Nair D, Kotagal P, Bingaman W, Najm I. Subdural electrode analysis in focal cortical dysplasia: predictors of surgical outcome. Neurology 2007;69:660-7. Foldvary N, Klem G, Hammel J, Bingaman W, Najm I, Lu¨ders H. The localizing value of ictal EEG in focal epilepsy. Neurology 2001;57:2022-8. Worrell GA, So EL, Kazemi J, et al. Focal ictal beta discharge on scalp EEG predicts excellent outcome of frontallobe epilepsy surgery. Epilepsia 2002; 43:277-82. Palmini A, Andermann F, Olivier A, et al. Focal neuronal migration disorders and intractable partial epilepsy: a study of 30 patients. Ann Neurology 1991;30:741-9. Ferrier CH, Aronica E, Leijten FS, et al. Electrocorticographic discharge patterns in glioneuronal tumors and focalcortical dysplasia. Epilepsia 2006;47:1477-86. Bragin A, Wilson CL, Engel J, Jr. Chronic epileptogenesis requires development of a network of pathologically interconnected neuron clusters: a hypothesis. Epilepsia 2000;41:S144-52. Fisher RS, Webber WR, Lesser RP, Arroyo S, Uematsu S. High-frequency EEG activity at the start of seizures. J Clin Neurophysiol 1992;9:441-8. Jirsch JD, Urrestarazu E, LeVan P, Olivier A, Dubeau F, Gotman J. High-frequency oscillations during human focal seizures. Brain 2006;129:1593-608. Ochi A, Otsubo H, Donner EJ, et al. Dynamic changes of ictal high-frequency oscillations in neocortical epilepsy: using multiple band frequency analysis. Epilepsia 2007;48:286-96. Chassoux F. Stereo-EEG: the Sainte-anne experience in focal cortical dysplasias. Epileptic Disord 2003; 5:95-103. Francione S, Nobili L, Cardinale F, et al. Intra-lesional stereo-EEG activity in Taylor’s focal cortical dysplasia. Epileptic Disord 2003;5(S2):S105-S114. Aghakhani Y, Kinay D, Gotman J, et al. The role of periventricular nodular heterotopia in epileptogenesis. Brain 2005;128:641-51. Chassoux F, Landre E, Rodrigo S, et al. Intralesional recordings and epileptogenic zone in focal polymicrogyria. Epilepsia 2008;49:51-64. Kobayashi E, Bagshaw AP, Grova C, et al. Grey matter heterotopia: what EEG-fMRI can tell us about epileptogenicity of neuronal migration disorders. Brain 2006; 129:366-74. Battaglia G, Chiapparini L, Franceschetti S, et al. Periventricular nodular heterotopia: classification, epileptic history and genesis of epileptic activity. Epilepsia 2006; 47:86-97. Chassoux F, Devaux B, Landre E, et al. Stereoelectroencephalography in focal cortical dysplasia. A 3D approach
2647
2648
156
Image guided epilepsy surgery
to delineating the dysplastic cortex. Brain 2000;123: 1733-51. 79. McGonigal A, Bartolomei F, Re´gis J, et al. Stereoelectroencephalography in presurgical assessment of MRInegative epilepsy. Brain 2007;130:3169-83. 80. Silbergeld DL, Miller JW. Resective surgery for medically intractable epilepsy associated with schizencephaly. J Neurosurg 1994;80:820-5. 81. Baumgartner C, Pataraia E. Revisiting the role of magnetoencephalography in epilepsy. Curr Opin Neurol 2006;19:181-6. 82. Otsubo H, Ochi A, Elliott I, et al. MEG predicts epileptic zone in lesional extrahippocampal epilepsy: 12 pediatric surgery cases. Epilepsia 2001;42:1523-30. 83. Morioka T, Nishio S, Ishibashi H, et al. Intrinsic epileptogenicity of focal cortical dysplasia as revealed by magnetoencephalography and electrocorticography. Epilepsy Res 1991;33:177-87. 84. Ishibashi H, Simos PG, Wheless JW, et al. Localization of ictal and interictal bursting epileptogenic activity in focalcortical dysplasia: agreement of magnetoencephalography and electrocorticography. Neurol Res 2002;24:525-30. 85. Bast T, Oezkan O, Rona S, et al. EEG and MEG source analysis of single and averaged interictal spikes revealsintrinsic epileptogenicity in focal cortical dysplasia. Epilepsia 2004;45:621-31. 86. Zhang W, Simos PG, Ishibashi H, et al. Multimodality neuroimaging evaluation improves the detection of subtle cortical dysplasia in seizure patients. Neurol Res 2003;25:53-7. 87. Gallen CC, Sobel DF, Waltz T, et al. Noninvasive presurgical neuromagnetic mapping of somatosensory cortex. Neurosurgery 1993;33:260-8. 88. Papanicolaou AC, Simos PG, Castillo EM, et al. Magnetocephalography: a noninvasive alternative to the Wada procedure. J Neurosurg 2004;100:867-76. 89. Hirata M, Kato A, Taniguchi M, et al. Determination of language dominance with synthetic aperture magnetometry: comparison with the Wada test. Neuroimage 2004;23:46-53. 90. Bowyer SM, Moran JE, Weiland BJ, et al. Language laterality determined by MEG mapping with MRFOCUSS. Epilepsy Behav 2005;6:235-41. 91. Hong SC, Kang KS, Seo DW, et al. Surgical treatment of intractable epilepsy accompanying cortical dysplasia. J Neurosurg 2000;93:766-73. 92. Edwards JC, Wyllie E, Ruggeri PM, et al. Seizure outcome after surgery for epilepsy due to malformation of cortical development. Neurology 2000;55:1110-14. 93. Kloss S, Pieper T, Pannek H, et al. Epilepsy surgery in children with focal cortical dysplasia (FCD): results of longterm seizure outcome. Neuropediatrics 2002;33:21-6. 94. Tassi L, Colombo N, Garbelli R, et al. Focal cortical dysplasia: neuropathological subtypes, EEG, neuroimaging and surgical outcome. Brain 2002;125:1719-32.
95. Kral T, Clusmann H, Blumcke I, et al. Outcome of epilepsy surgery in focal cortical dysplasia. J Neurol Neurosurg Psychiatry 2003;74:183-8. 96. Bautista JF, Foldvary-Schaefer N, Bingaman WE, et al. Focal cortical dysplasia and intractable epilepsy in adults: clinical, EEG, imaging, and surgical features. Epilepsy Res 2003;55:131-6. 97. Francione S, Vigliano P, Tassi L, et al. Surgery for drug resistant partial epilepsy in children with focal cortical dysplasia: anatomical-clinical correlations and neurophysiological data in 10 patients. J Neurol Neurosurg Psychiatry 2003;74:1493-501. 98. Fauser S, Schulze-Bonhage A, Honegger J, et al. Focal cortical dysplasias: surgical outcome in 67 patients in relation to histological subtypes and dual pathology. Brain 2004;127:2406-18. 99. Hader WJ, MacKay M, Otsubo H, et al. Cortical dysplastic lesions in children with intractable epilepsy: role of complete resection. J Neurosurg (Pediatrics 2) 2004;100:110-17. 100. Fountas KN, King DW, Meador KJ, et al. Epilepsy in cortical dysplasia: factors affecting surgical outcome. Stereotact Funct Neurosurg 2004;82:26-30. 101. Hudgins RJ, Flamini JR, Palasis S, et al. Surgical treatment of epilepsy in children caused by focal cortical dysplasia. Pediatr Neursurg 2005;41:70-6. 102. Cohen-Gadol AA, Ozduman K, Bronen RA, et al. Long-term outcome after epilepsy surgery for focal cortical dysplasia. J Neurosurg 2004;101:55-65. 103. Alexandre V, Walz R, Bianchin MM, et al. Seizure outcome after surgery for epilepsy due to focal cortical dysplastic lesions. Seizure 2006;15:420-7. 104. Siegel AM, Cascino GD, Meyer FB, et al. Surgical outcome and predictive factors in adult patients with intractable epilepsy and focal cortical dysplasia. Acta Neurol Scand 2006;113:65-71. 105. Cossu M, Russo GL, Francione S, et al. Epilepsy surgery in children: results and predictors of outcome in seizures. Epilepsia 2008;49:65-72. 106. Urbach H, Scheffler B, Heinrichsmeier T et al. Focal cortical dysplasia of Taylor’s balloon cell type: a clinicopathological entity with characteristic neuroimaging and histopathological features, and favorable postsurgical outcome. Epilepsia 2002;43:33-40. 107. Tassi L, Pasquier B, Minotti L et al. Cortical dysplasia: electroclinical, imaging, and neuropathologic study of 13 patients. Epilepsia 2001;42:1112-23. 108. Zentner J, Hufnagel A, Ostertun B, et al. Surgical treatment of extratemporal epilepsy: clinical, radiologic, and histopathologic findings in 60 patients. Epilepsia 1996;37:1072-80. 109. Frater JL, Prayson RA, Morris IH, Bingaman WE. Surgical pathologic findings of extratemporal-based intractable epilepsy: a study of 133 consecutive resections. Arch Pathol Lab Med 2000;124:545-9.
Image guided epilepsy surgery
110. Beaumont A, Whittle IR. The pathogenesis of tumour associated epilepsy. Acta Neurochir (Wien) 2000;142: 1-15. 111. Penfield W, Erickson T, Tarlov I. Relation of intracranial tumors and symptomatic apilepsy. Arch Neurol Psychiatry 1940;44:300-15. 112. Gonzalez D, Elvidge AR. On the occurrence of epilepsy caused by astrocytoma of the cerebral hemispheres. J Neurosurg 1962;19:470-82. 113. Sherwin AL, Vernet O, Dubeau F, Olivier A. Biochemical markers of excitability in human neocortex. Can J Neurol Sci 1991;18:640-4. 114. Avoli M, Drapeau C, Louvel J, Pumain R, Olivier A, Villemure JG. Epileptiform activity induced by low extracellular magnesium in the human cortex maintained in vitro. Ann Neurol 1991;30:589-96. 115. Singh R, Pathak DN. Lipid peroxidation and glutathione peroxidase, glutathione reductase, superoxide dismutase, catalase, and glucose-6-phosphate dehydrogenase activities in FeCl3-induced epileptogenic foci in the rat brain. Epilepsia 1990;31:15-26. 116. Linn F, Seo K, Hossmann KA. Experimental transplantation gliomas in the adult cat brain. 3. Regional biochemistry. Acta Neurochir (Wien) 1989;99:85-93. 117. Okada Y, Kloiber O, Hossmann KA. Regional metabolism in experimental brain tumors in cats: relationship with acid/base, water, and electrolyte homeostasis. J Neurosurg 1992;77:917-26. 118. Wolf HK, Muller MB, Spanle M, Zentner J, Schramm J, Wiestler OD. Ganglioglioma: a detailed histopathological and immunohistochemical analysis of 61 cases. Acta Neuropathol 1994;88:166-73. 119. Blumcke I, Wiestler OD. Gangliogliomas: an intriguing tumor entity associated with focal epilepsies. J Neuropathol Exp Neurol 2002;61:575-84. 120. Tonini C, Beghi E, Berg AT, et al. Predictors of epilepsy surgery outcome: a meta-analysis. Epilepsy Res 2004; 62:75-87. 121. Britton JW, Cascino GD, Sharbrough FW, Kelly PJ. Low-grade glial neoplasms and intractable partial epilepsy: efficacy of surgical treatment. Epilepsia 1994; 35:1130-5. 122. Laoprasert P, Black P, Bromfield E, Wen P. Seizure outcome in patients with low-grade glioma resected using 3D magnetic resonnance imaging navigation or intraoperative MRI. Epilepsia 1998;39:90. 123. Prayson RA, Khajavi K, Comair YG. Cortical architectural abnormalities and MIB1 immunoreactivity in gangliogliomas: a study of 60 patients with intracranial tumors. J Neuropathol Exp Neurol 1995;54:513-20. 124. Penfield W. The mechanism of critical contraction in the brain. Brain 1927;50:499-517. 125. Penfield W. The radical treatment of traumatic epilepsy and its rationale. Can Med Assoc J 1930;23:189-97. 126. Asikainen I, Kaste M, Sarna S. Early and late posttraumatic seizures in traumatic brain injury rehabilitation
127.
128. 129.
130. 131.
132.
133.
134.
135.
136.
137.
138.
139.
140.
141.
142.
143.
156
patients: brain injury factors causing late seizures and influence of seizures on long-term outcome. Epilepsia 1999;40:584-9. Annegers J. The epidemiology of epilepsy. In: Wyllie E, editor. The treatment of epilepsy: principles and practice. 2nd ed. Williams & Wilkins; Baltimore: p. 165-72. Caveness WF. Epilepsy, a product of trauma in our time. Epilepsia 1976;17:207-15. Kim DS, Park YG, Choi JU, Chung SS, Lee KC. An analysis of the natural history of cavernous malformations. Surg Neurol 1997;48:9-17; discussion 17-8. Robinson JR, Awad IA, Little JR. Natural history of the cavernous angioma. J Neurosurg 1991;75:709-14. Vives K, Awad I. Vascular causes of epilepsy. In: Kotagal P, Luders H, editors. The epilepsies: etiologies and prevention. New York: Academic Press; 1999. p. 371-83. Regis J, Bartolomei F, Kida Y, et al. Radiosurgery for epilepsy associated with cavernous malformation: retrospective study in 49 patients. Neurosurgery. 2000;47:1091-7. Shih YH, Pan DH. Management of supratentorial cavernous malformations: craniotomy versus gammaknife radiosurgery. Clin Neurol Neurosurg 2005;107:108-12. Karlsson B, Kihlstrom L, Lindquist C, Ericson K, Steiner L. Radiosurgery for cavernous malformations. J Neurosurg 1998;88:293-7. Seo Y, Fukuoka S, Takanashi M, et al. Gamma Knife surgery for angiographically occult vascular malformations. Stereotact Funct Neurosurg 1995;64 Suppl 1: 98-109. Kraemer DL, Awad IA. Vascular malformations and epilepsy: clinical considerations and basic mechanisms. Epilepsia 1994;35 Suppl 6:S30-43. Wolf HK, Roos D, Blumcke I, Pietsch T, Wiestler OD. Perilesional neurochemical changes in focal epilepsies. Acta Neuropathol 1996;91:376-84. Yeh HS, Privitera MD. Secondary epileptogenesis in cerebral arteriovenous malformations. Arch Neurol 1991;48:1122-4. Crawford PM, West CR, Shaw MD, Chadwick DW. Cerebral arteriovenous malformations and epilepsy: factors in the development of epilepsy. Epilepsia 1986; 27:270-5. Foy PM, Copeland GP, Shaw MD. The incidence of postoperative seizures. Acta Neurochir (Wien) 1981;55:253-64. Murphy MJ. Long-term follow-up of seizures associated with cerebral arteriovenous malformations. Results of therapy. Arch Neurol 1985;42:477-9. Parkinson D, Bachers G. Arteriovenous malformations. Summary of 100 consecutive supratentorial cases. J Neurosurg 1980;53:285-99. Piepgras DG, Sundt TM, Jr, Ragoowansi AT, Stevens L. Seizure outcome in patients with surgically treated cerebral arteriovenous malformations. J Neurosurg 1993;78:5-11.
2649
2650
156
Image guided epilepsy surgery
144. Yeh HS, Tew JM, Jr, Gartner M. Seizure control after surgery on cerebral arteriovenous malformations. J Neurosurg 1993;78:12-8. 145. Thorpe ML, Cordato DJ, Morgan MK, Herkes GK. Postoperative seizure outcome in a series of 114 patients with supratentorial arteriovenous malformations. J Clin Neurosci 2000;7:107-11. 146. Dodick DW, Cascino GD, Meyer FB. Vascular malformations and intractable epilepsy: outcome after surgical treatment. Mayo Clin Proc 1994;69:741-5. 147. Kida Y, Kobayashi T, Tanaka T, Mori Y, Hasegawa T, Kondoh T. Seizure control after radiosurgery on cerebral arteriovenous malformations. J Clin Neurosci 2000;7 Suppl 1:6-9. 148. Kurita H, Kawamoto S, Suzuki I, et al. Control of epilepsy associated with cerebral arteriovenous malformations after radiosurgery. J Neurol Neurosurg Psychiatry 1998;65:648-55. 149. Schauble B, Cascino GD, Pollock BE, et al. Seizure outcomes after stereotactic radiosurgery for cerebral arteriovenous malformations. Neurology 2004;63: 683-7. 150. Laskowitz DT, Sperling MR, French JA, O’Connor MJ. The syndrome of frontal lobe epilepsy: characteristics and surgical management. Neurology 1995;45:780-7. 151. Olivier A, Awad I. Extratemporal resections. In: Engel JJ, editor. Surgical treatment of the epilepsies. 2nd ed. New York: Raven Press; 1993. p. 489-500. 152. Bancaud J, Talairach J. Clinical semiology of frontal lobe seizures. Adv Neurol 1992;57:3-58. 153. Williamson PD, Spencer SS. Clinical and EEG features of complex partial seizures of extratemporal origin. Epilepsia 1986;27 Suppl 2:S46-63. 154. Salanova V, Morris HH, Van Ness P, Kotagal P, Wyllie E, Luders H. Frontal lobe seizures: electroclinical syndromes. Epilepsia 1995;36:16-24. 155. Rasmussen T. Surgery of frontal lobe epilepsy. In: Purpusa D, Perry J, Walter R, editors. Advances in neurology. New York: Raven Press; 1975. p. 207-26. 156. Penfield W, Welch K. The supplementary motor area of the cerebral cortex; a clinical and experimental study. AMA Arch Neurol Psychiatry 1951;66:289-317. 157. Tharp BR. Orbital frontial seizures. An unique electroencephalographic and clinical syndrome. Epilepsia 1972;13:627-42.
158. Chamoun R, Alaraj A, Bouclaous C, Mikati M, Comair Y. Functional and seizure outcome following resections in the primary sensory-motor cortex [abstract]. Epilepsia 2005;46:55. 159. Rouiller EM, Yu XH, Moret V, Tempini A, Wiesendanger M, Liang F. Dexterity in adult monkeys following early lesion of the motor cortical hand area: the role of cortex adjacent to the lesion. Eur J Neurosci 1998;10:729-40. 160. Jaillard A, Martin CD, Garambois K, Lebas JF, Hommel M. Vicarious function within the human primary motor cortex? A longitudinal fMRI stroke study. Brain 2005;128:1122-38. 161. Seitz RJ, Hoflich P, Binkofski F, Tellmann L, Herzog H, Freund HJ. Role of the premotor cortex in recovery from middle cerebral artery infarction. Arch Neurol 1998;55:1081-8. 162. Baciu M, Le Bas JF, Segebarth C, Benabid AL. Presurgical fMRI evaluation of cerebral reorganization and motor deficit in patients with tumors and vascular malformations. Eur J Radiol 2003;46:139-46. 163. Cao Y, D’Olhaberriague L, Vikingstad EM, Levine SR, Welch KM. Pilot study of functional MRI to assess cerebral activation of motor function after poststroke hemiparesis. Stroke 1998;29:112-22. 164. Salanova V, Andermann F, Rasmussen T, Olivier A, Quesney LF. Parietal lobe epilepsy. Clinical manifestations and outcome in 82 patients treated surgically between 1929 and 1988. Brain 1995;118 (Pt 3): 607-27. 165. Williamson PD, Boon PA, Thadani VM, et al. Parietal lobe epilepsy: diagnostic considerations and results of surgery. Ann Neurol 1992;31:193-201. 166. Salanova V, Andermann F, Olivier A, Rasmussen T, Quesney LF. Occipital lobe epilepsy: electroclinical manifestations, electrocorticography, cortical stimulation and outcome in 42 patients treated between 1930 and 1991. Surgery of occipital lobe epilepsy. Brain 1992;115(Pt 6): 1655-80. 167. Williamson PD, Thadani VM, Darcey TM, Spencer DD, Spencer SS, Mattson RH. Occipital lobe epilepsy: clinical characteristics, seizure spread patterns, and results of surgery. Ann Neurol 1992;31:3-13.
155 Imaging Evaluation of Epilepsy D. Madhavan . R. Kuzniecky
Introduction Recent advancements in the field of neuroimaging have resulted in a broadened ability to characterize and manage epilepsy [1,2]. The improved ability to image structural abnormalities in the brain has allowed clinicians to dramatically impact the management and prognosis of patients with epilepsy. Perhaps this is most strikingly demonstrated in the diagnosis and treatment of mesial temporal sclerosis, where in vivo identification of this lesion using Magnetic Resonance Imaging (MRI) can lead to surgical therapy, often with excellent results [3,4]. Additionally, modern imaging techniques offer the ability to visualize specific functional activities in brain, including hemodynamic perfusion, neuronal metabolism, and electromagnetic fields. These can then be coregistered to a structural image, yielding a detailed combination of anatomy and function that allows the opportunity to view pathophysiological and molecular processes associated with different types of epilepsy. The goal of this chapter is to describe some of these imaging modalities and their role in the diagnosis and management of the epileptic patient.
Imaging Modalities Computed Tomography Computed Tomography (CT) uses beams of ionizing radiation to generate a series of twodimensional X-ray scans that can be reconstructed into a three-dimensional image. This
#
Springer-Verlag Berlin/Heidelberg 2009
generates excellent imaging of hard tissue with moderate soft tissue resolution. The main advantage of CT is its relative low cost, speed, and accessibility when compared to other imaging techniques. Specifically, newer generation CT scanners can generate a brain image in a matter of seconds [5–9]. With the advent of MRI, the role of CT has greatly diminished in the diagnosis and management of epilepsy. However, CT remains the imaging procedure of choice in some selected instances. Due to its speed and sensitivity in detecting bleeding, CT is often used in the emergency room or the perioperative period to visualize gross structural changes, hydrocephalus, and intracranial hemorrhage [10]. In addition, CT can identify calcified lesions and subacute/chronic infarctions. Despite the above uses, CT has significant limitations in imaging epilepsy patients, with a 30% sensitivity in identifying abnormalities in unselected epilepsy populations [11]. A major weakness of CT is its inability to visualize the temporal fossa with high resolution, severely limiting its use in detecting mesial temporal sclerosis (MTS), the most common lesional pathology in intractable mesial temporal epilepsy [12]. In addition, CT may fail to detect the presence of other epileptogenic structural abnormalities, i.e., small tumors and arteriovenous maformations (AVMs) in up to 50% of patients. Current International League of Epilepsy (ILAE) guidelines for neuroimaging identify CT as the imaging modality of choice if and MRI is not available. However, because of the above imaging limitations of CT, the ILAE recommends an MRI study if the CT is normal.
2618
155
Imaging evaluation of epilepsy
Magnetic Resonance Imaging (MRI) MRI produces images by visualizing the relaxation properties of hydrogen nuclei in tissue subjected to a powerful magnetic field. This magnetic field induces a particular spin of each hydrogen nucleus, and aligns them relative to the field, which produces the image. MRI has a number of advantages over CT including using non-ionizing radiation to produce images, ability to produce multiplanar images, and significantly improved anatomic resolution and contrast of soft and hard tissue. MRI is the imaging procedure of choice for patients with epilepsy. It is useful in detecting several cerebral abnormalities related to epilepsy in adults and children, including MTS, tumors, and developmental malformations [13–19]. The ability of MRI to properly visualize and characterize these lesions is dictated by the underlying pathology, the type of MRI sequences employed, and the experience of the interpreting physician. MRI is of particular use in visualizing MTS, which is the most common pathologic etiology seen in patients with intractable temporal lobe epilepsy [15]. MTS is grossly characterized by the presence of a firm, atrophic hippocampus, and histologically noted to have neuronal loss and gliosis in CA1, CA3, and CA4 of the hippocampal subfields. MTS has been recently suggested to be due to a complex interplay of acquired and genetic factors, rather than a homogenous disease process. This interplay may explain the association of MTS with febrile seizures, traumatic injury, or status epilepticus. In addition, MTS can often be detected in conjunction with lateral temporal or extratemporal pathology, such as porencephaly, tumors, or cortical dysplasias, phenomenon known as dual pathology. In fact, the development of dual pathology has been noted in patients with temporal cortical dysplasia and MTS, a condition which is more likely in patients experiencing febrile seizures [20].
MRI is capable of detecting MTS with high sensitivity and specificity when hippocampal neuronal loss is at least 50%. The MRI features of hippocampal sclerosis include (1) hippocampal atrophy; (2) increased signal on T2-weighted images; and (3) decreased signal on inversion recovery sequences [4,21,22]. The detection of these abnormalities should be carried out with optimized imaging techniques, which include angulated coronal sections obtained perpendicular to the long axis of the hippocampal structures. It has been shown that a combination of sequences, particularly FLAIR and T1-weighted images can correctly identify MTS in over 90% of patients [4]. Volumetric measurements may also aid in diagnosis, but the gains in sensitivity are at best 5–10% over qualitative diagnosis [23]. Developmental malformations are commonly associated with childhood epilepsies, and should be investigated in children with epilepsy [24–28]. MRI can accurately define diffuse malformations such as lissencephaly, band heterotopia, and periventricular nodular heterotopia. It can also define hemimegalencephaly, schizencephaly, and focal subcortical heterotopias. Focal cortical dysplasias (FCD) are the most common developmental pathology in children with extratemporal lobe seizures [27,29,30], and can also be an etiologic factor in the adult population. The MRI features of FCDs are variable, but are often due to abnormal migration patterns of neurons and glia from the deep ventricular zones to the cortical surface, which results in an abnormal cortical mantle with disturbed gray-white matter architecture and thickening of the cortex. T2-weighted abnormalities in the underlying white matter correlate with abnormal cells and remnants of radial/glia fibers. Since these lesions can be small in size, the MRI examination should be targeted to clinically suspected regions using both 3D volume techniques and inversion recovery sequences and surface coils. Newer techniques such as phased array surface coils, Magnetic Resonance Spectroscopy (MRS), and Diffusion
Imaging evaluation of epilepsy
Tensor Imaging (DTI) can assist in the localization of the dysplasia (see below). Additional lesional childhood epilepsies that can be visualized by MRI include Sturge-Weber syndrome and Tuberous Sclerosis Complex. In Sturge-Weber syndrome, cortical hemangiomas and calcifications are often detected with conventional MRI sequences, with more subtle changes in brain metabolism and underlying white matter imaged using various advanced MRI techniques. In Tuberous Sclerosis Complex, MRI is valuable in identifying cortical hamartomas and Subependymal Giant cell Astrocytomas (SEGAs). In both conditions, MRI imaging can be critical in presurgical planning for the identification of lesions amenable to resection. Early destructive injuries constitute another major group of pathologies underlying diffuse or focal injury in patients with seizures. The MRI appearance is dependent on the type and time of injury to the brain. Early injuries (first 6 months of gestation) will result in porencephaly. Late gestational, perinatal or postnatal injuries will result in encephalomalacia or ulegyria. MRI can distinguish both conditions based on imaging features [31]. Of importance however, is the fact that porencephaly can be associated with ipsilateral MTS (dual pathology). Encephalomalacia can be diffuse such as in anoxic injuries or can be localized to the distribution of a cerebral artery branch. Other pathologies commonly associated with epilepsy include developmental tumors (ganglioglial), vascular lesions (> Figures 155-1–> 155-3) and less common neurocutaneus syndromes (incotinentia pigmenti, Neurofibromatosis Type I, etc). In general, these disorders are diagnosed on the basis of other features and will not be discussed here.
Single Photon Emission Computed Tomography (SPECT) SPECT scanning is a nuclear tomographic study that utilizes gamma radiation to visualize an
155
injected radionuclide tracer, usually ECD (Neurolite; DuPont Pharma) and HMPAO (Ceretec; Nycomed Amersham), attached to 99mTechnicium. This tracer is injected into the venous . Figure 155-1 Mesial temporal Sclerosis. (A) T1-W coronal MRI showing right hippocampal atrophy. Note abnormal signal from hippocampus and severe atrophy
. Figure 155-2 Laminar heterotopia. The coronal T-1W image shows double cortex. This is a mutation of DCX or doublecortin gene or LIS1 gene
2619
2620
155
Imaging evaluation of epilepsy
. Figure 155-3 Peri-rolandic vascular lesion. Young patient with intractable seizures since age 11. Abnormal signal with hemosiderin deposition is observed around sulcus
pool and eventually trapped in the brain during the first pass. Once injected and trapped, it will allow scanning up to 6 h later without major degradation. SPECT provides a snapshot of cerebral perfusion at a given point in time, which can utilized in the presurgical evaluation of patients with intractable epilepsy [32–37]. Several studies have previously investigated the utility of SPECT scanning in the interictal state. It has been demonstrated that only about 50% of patients with well-documented temporal lobe epilepsy demonstrate interictal temporal hypoperfusion in the corresponding epileptogenic temporal region. Therefore, interictal SPECT studies do not show perfusion abnormalities in the other half of this patient group. In addition, 5–10% of patients may demonstrate hypoperfusion in the contralateral temporal region, raising the possibility of false lateralization [38]. Based on these results, it appears that interictal SPECT scanning may identify abnormalities in patients with well defined MRI and electroencephalogram (EEG) abnormalities, but it is not sensitive enough to be a reliable or cost-effective
study. However, they may be of use in subtraction SPECT studies (see below). The major utility of SPECT in epilepsy emerges during the ictal state. Several studies have shown focal increased perfusion in the epileptogenic focus with ictal SPECT in 70–90% of patients, with more accurate localization in patients with temporal lobe epilepsy [34,39–41]. These ictal patterns are variable and depend on the localization of the epileptogenic focus and the time of injection. As there can be rapid changes in cerebral blood flow during the initial seconds of a seizure, delayed injections may not reveal changes in perfusion or may display erroneous results. The yield of ictal studies diminishes if the injection is given in the late ictal or postictal phases. Therefore, the tracer injection should be given within the first seconds of ictal onset, which requires ictal SPECT studies being performed in the setting of inpatient video-EEG monitoring. Additionally, personnel that are specialized and experienced in performing ictal SPECT studies enhance the likelihood of obtaining an accurate scan. Recently, computational analysis techniques have been employed to produce SPECT images where the interictal baseline SPECT is subtracted from the ictal SPECT, and coregistered to a structural MRI scan [42–46]. Statistical Paramteric Mapping (SPM) is a quantitative algorithm that has been utilized as a methodology to perform this comparison, or to compare SPECT with other functional imaging modalities, in order to visualize areas of ictal abnormalities [47]. These subtraction scans have shown a substantially higher yield of localization of the epileptogenic zone (> Figure 155-4).
Positron Emission Tomography (PET) PET scanning is another nuclear tomographic technique that has been used for the investigation
Imaging evaluation of epilepsy
. Figure 155-4 Ictal SPECT/MRI with statistical parametric maps of blood flow show increases corresponding to the epileptic focus. This SPECT technique is useful for localization for surgical intervention
of epilepsy patients for more than a decade [48– 52]. These studies have traditionally used 2-deoxy-2(18F)fluoro-D-glucose (FDG) to measure neuronal uptake of glucose in the interictal state, thus providing a visualization of cellular metabolism. In temporal lobe epilepsy, interictal studies show hypometabolic areas in the epileptogenic regions. These findings have been observed in approximately 80% of patients with temporal lobe epilepsy [53]. The changes however, are more extensive than the structural and EEG abnormalities and may involve the ipsilateral suprasylvian and parietal regions. The high sensitivity of MRI in detecting MTS and other pathologies in temporal lobe surgical candidates has diminished the role of PET in the presurgical investigation of such patients. Nevertheless, when MRI is normal, PET may be indicated to aid in localization. In extra-temporal lobe epilepsy, interictal PET-FDG studies are less useful, especially if the MRI is normal and the scalp EEG is non-focal. However, PET has been reported to be more sensitive in neonates and infants with focal epilepsy,
155
as it is likely that a developmental malformation is present in those cases that would contribute to focal metabolic derangements. This is particularly the case in patients with infantile spasms and focal features on EEG. When a single region of metabolic abnormality corresponding to the EEG abnormality is detected, surgical treatment is effective in controlling seizures and improving developmental outcome. PET has also improved the understanding of the pathophysiology of infantile spasms by demonstrating activation of cortical regions, brainstem and lenticular nuclei. In contrast to FDG PET, ligand/neuroreceptor PET studies can improve sensitivity and specificity in detecting temporal vs. extratemporal lobe epilepsy and for particular situations. Unlike FDG, which characteristically shows decreased uptake in the region of the seizure focus interictally, neuroreceptor tracers show increased uptake in epileptogenic brain regions even in the interictal state. This represents a clear advantage over other imaging studies, especially when multiple structural lesions are present and potentially confound detection of the epileptogenic zone. AMT (a[11C]methyl-L-tryptophan) PET is a particular example of this specialized type of PET. The first successful clinical application of AMT PET in intractable epilepsy was demonstrated in children with the tuberous sclerosis complex [54–59]. AMT PET has been demonstrated to aid in differentiating between epileptogenic and non-epileptogenic tubers in the interictal state [55]. While the specificity of focally increased AMT uptake for the epileptic focus appears to be very high, its sensitivity is low (40–80%), and depends on the underlying pathology as well as the analytic approach applied. Other receptor types with respective PET tracers include opiate, benzodiazepine and NMDA. Clinical experience with these neuroligands have not been well established, but a number of preliminary findings have been reported in patients [60–64]. Opiate receptor studies with 11 C-carfentanil (mu-opiate receptor) have shown
2621
2622
155
Imaging evaluation of epilepsy
increased binding in the lateral neocortex of temporal lobe epilepsy patients without mesial temporal involvement. Other opiate receptor studies have confirmed the specificity of mu-opiate bindings in temporal lobe epilepsy. In addition, benzodiazepine-labeling studies with 11C-flumazenil, a central benzodiazepine receptor antagonist, have shown reduced binding in the epileptogenic focus, and may bind more specifically than FDG studies.
Magnetic Source Imaging (MSI) MSI involves utilizing magnetoencephalography (MEG) to non-invasively record intraneuronal electromagnetic currents and co-registering the MEG source localization to an anatomical image, usually an MRI. MSI is currently FDA approved for two purposes; The presurgical localization of epileptogenic activity, and the mapping of normal neuronal function for localization of somatosensory and motor function. This latter indication for MSI is mainly used for the presurgical planning of a lesional resection, such as a tumor. MSI is also currently being investigated in the mapping of language and cognitive function, with promising results. The main advantage of MSI is its ability to map electromagnetic function with extremely high spatial and temporal sensitivity. Unlike electrical potentials measured with EEG, which are significantly attenuated in strength and spatially blurred by tissues between brain and scalp surface, magnetic fields are not significantly affected by intervening tissue layers. Another distinction between MEG and EEG is that MEG can only record the magnetic fields generated from currents oriented tangentially to the surface of the brain while EEG records the electrical potentials from both tangentially and radially oriented neuronal populations [65–70]. To a variable degree this is another potential advantage for MEG because it can decrease some of the assumptions in the modeling required
for computing source localization. It is thus possible to measure magnetic fields emerging from brain tissue in real time, thereby providing an advantage when compared to perfusion or metabolism based functional imaging, such as PET/ SPECT or functional MRI (fMRI). The main challenge of MEG is that the magnetic fields emerging from the skull are extremely weak, and require highly sensitive sensors that are susceptible to noise or magnetic interference. However, due to new computational techniques and improved shielding of the MEG apparatus, making it possible to simultaneously record the whole brain in a clinical population. In the evaluation of epilepsy patients, MEG can be used to visualize and identify sources of abnormal epileptogenic activity, similar to the spikes and sharp waves seen with EEG. MEG is reliable for localization of spike sources in the following patients: patients with no lesion visible on MRI, cystic lesions (post-traumatic b encephalomalacia with prior surgical resection) in which MRI localization is ambiguous, and in patients with lesions of undetermined significance. MEG may also be of use in patients with bilateral synchronous spikes or those with multifocal spike discharges [71–74]. MEG is often concurrently performed with EEG [75,76], which can enhance localization of discharges. For example, it has been reported that the spatial orientation and trajectory of the MEG spike in patients with temporal lobe epilepsy appears to be predictive of mesial or lateral localization of the seizure onset zone [77]. MEG appears to also be of additional use in patients with bilateral synchronous spikes or multifocal epileptiform discharges, as these tend to be obscured with EEG analysis alone. In fact, one study reported the positive predictive value in MSI in localizing the seizure onset zone to be between 82–90%, which suggested that enough clinical validity existed for MSI to potentially replace intracranial EEG for localization of the seizure onset zone [78] (> Figures 155-5 and > 155-6).
Imaging evaluation of epilepsy
. Figure 155-5 Magnetic Source Imaging (MSI). Magnetoencephalography combined with MRI showing a cluster of dipoles deep in the frontal lobe in patient with focal dysplasia
Functional MRI (fMRI) fMRI measures hemodynamic changes related to neuronal activity, and has been applied to the study of patients with epilepsy [79]. When neurons are metabolically active, blood oxygen consumption is increased in the region of activity. As the magnetic signal of blood is different depending on its level of oxygenation, fMRI uses specialized MRI pulse sequences to measure this differential oxygenation, a process known as Blood Oxygen Level Dependent (BOLD) imaging. Like MSI, one of the major uses of fMRI is the presurgical mapping of somatosensory, motor, or visual cortices. A large number of fMRI studies have demonstrated primary sensorimotor cortex activation along the central sulcus during movement [80–85], including demonstration of the somatotopic organization of this region. Finger movements are used most commonly, since face or large limb movements increase the likelihood of unacceptable movement artifacts. A simple and commonly used
155
procedure is to have the patient oppose the fingers sequentially to the thumb as quickly as possible. Such tasks appear to reliably activate sensorimotor cortex in the central sulcus, and have been used in a number of patient studies. Motor cortex localization with fMRI has generally been highly concordant with intraoperative electrocortical stimulation mapping [86–88]. This is of especially high clinical utility in lesional or seizure resections where the surgical focus is in proximity to the central sulcus. It might also be possible to minimize any resulting deficit by purposefully sparing activated and immediately surrounding regions, though no quantitative studies have verified the effectiveness of such an approach. FMRI information is perhaps particularly useful when anatomical structures are distorted by mass effects, making it difficult to ascertain the location of the central sulcus with certainty. fMRI has been recently FDA approved for the localization of language functions preoperatively to minimize postoperative language deficits [89–95]. Recent experience suggests that with simple paradigms and training, even children as young as 5–7 years are able to perform reliable task for fMRI. A number of studies have compared fMRI language maps with language maps obtained using cortical stimulation mapping. The published studies comparing fMRI and cortical stimulation report encouraging results, but they involved relatively small samples (<15 patients). One study with 11 patients showed an average sensitivity of 81% and specificity of 53%, but another reported an average sensitivity of 92% and specificity of 61%. A more meaningful measure of the validity of fMRI language maps is how well they predict post-operative language deficits. Although very few studies have been carried out to answer this issue, available data appears to suggest that fMRI may have a higher positive predictive value than Wada testing for predicting significant language decline.
2623
2624
155
Imaging evaluation of epilepsy
. Figure 155-6 EEG of a left temporal spike (A, left), followed by corresponding MEG of the same spike (A, right). The MEG spike can be viewed as a contour map (B, note the increased concentration of field lines in the left temporal region), and later modeled as an Equivalent Current Dipole (ECD) co-registered to the patient’s MRI image (C) (received from Chad Carlson, M.D., New York University Comprehensive Epilepsy Center, personal communication)
More recently, fMRI has been applied to study memory using improved stimulating paradigms and better imaging techniques [96–99]. Complex cognitive functions such as memory typically activate a distributed network of brain regions, not all of which are critical for task performance. Conversely, some critical regions may not be activated due to limitations in sensitivity or task design. Initial fMRI studies suggest reductions in memory activation lateralize to the seizure foci, and results also suggest that post-surgical amnesia correlates with fMRI activation ipsilateral to the resection. As with any new diagnostic or therapeutic modality, fMRI memory localization should be carefully validated in a prospective fashion before using it in routine clinical management. A recent development in fMRI research addresses the change in BOLD imaging in the
interictal and ictal states. Investigators have demonstrated ictal focal blood flow changes in patients with partial seizures using fMRI techniques. Detection of interictal changes using BOLD for detection of the epileptic focus in conjunction with EEG is possible, with several laboratories have demonstrating the feasibility of performing spike-triggered or continuous fMRI-EEG studies [100–102]. This is still an experimental modality, however. It appears that the regions of EEG and fMRI activations have a limited coconcordance, with the fMRI activations appearing to be biased towards the lower frequency components of the intracranial EEG [103]. As these could potentially only represent slow wave components, this modality may be better suited as a complementary technique with scalp EEG to delineate regions related to the generation of interictal discharges. With further
Imaging evaluation of epilepsy
155
. Figure 155-7 Distributed language function averaged from four subjects visualized with fMRI (left) and MEG (right). The cortical image is ‘‘inflated’’ for the viewing of sulcal activations [104]
study and characterization, these techniques may increase spatial and temporal resolution for interictal localization in the future (> Figure 155-7).
Recent and Experimental Imaging Techniques As imaging technology has continued to progress, several new methods have emerged that can be applied to the study of epilepsy. These include functional studies such as Magnetic Resonance Spectroscopy (MRS), or advanced structural scanning such as Diffusion Tensor Imaging (DTI). MRS can provide noninvasive biochemical measurements of specific brain metabolites. In epilepsy, two major techniques have been applied. 31 P, or phosphorous spectroscopy is designed
to measure phospholipid metabolism and highenergy phosphate compounds. Studies have demonstrated a consistent abnormality in the epileptogenic region characterized by abnormal phosphocreatine/inorganic phosphate ratios. 31P, however, is difficult to obtain and less sensitive than proton studies [104]. 1 H spectroscopy in epilepsy patients has demonstrated abnormalities of N-acetyl-aspartate (NAA), a mitochondrial neuronal compound, and creatine (Cr) and choline (Cho). Several groups have found a consistent abnormality in NAA/Cr ratios in correspondence with the epileptogenic focus in temporal and extra-temporal lobe epilepsy [105–107]. The abnormal ratio is likely due to low NAA. Reductions In NAA are not correlated to neuronal cell loss and probably represent dysfunctional neurons [108–110]. However, MRS is more sensitive than MRI in
2625
2626
155
Imaging evaluation of epilepsy
detecting abnormalities and thus, it may be useful in patients with normal MRI studies. 1H has also demonstrated bitemporal abnormalities in up to 40% of patients but the significance of these findings remains unclear. Studies have also demonstrated increase lactate concentrations in the postictal state in the epileptogenic temporal lobe [111]. As MRS becomes better at high field (3 T), its use in epilepsy will increase. Changes on standard Diffusion Weighted MRI (DWI) are well established in the ictal and postictal states [112], with restricted diffusion profiles seen in the region of the ictal onset zone. This is thought to be related to cytotoxic edema being produced secondary to cellular membrane dysfunction and localized cerebral ischemia. DTI is a specialized DWI technique which visualizes the movement of water molecules along white matter fibers, thereby allowing visualization of these tracts. It is also possible to retrieve information about the directionality and magnitude of the diffusing water molecules, yielding additional information about the strength of white matter connectivity between brain regions. DTI has been shown in small samples to longitudinally visualize reorganization of language and visual pathways following epilepsy surgery, in order to gain a fuller understanding of postoperative deficits in conjunction with neuropsychological evaluation. In a study of a small series of patients who underwent temporal lobectomy for unilateral MTS, DTI revealed abnormal diffusion parameters of the fornix, cingulum, and external capsule bilaterally, with evidence of irreversible wallerian degeneration in the fornix and cingulum ipsilateral to the resection [113]. Another study revealed extensive extratemporal DTI abnormalities in patients with unilateral mesial temporal sclerosis, although the surgical implications for this are not clearly understood [114]. As with MRS, the advent of more powerful MRI scanners will increase the use of DTI, and will allow the opportunity to further characterize it in the setting of epilepsy (> Figure 155-8).
. Figure 155-8 Diffusion Tensor Imaging (DTI) of left hemisphere arcuate fasciculus in a right handed subject (received from Matthew White M.D., University of Nebraska Medical Center, personal communication)
A Practical Approach to Imaging in Epilepsy It is not necessarily required to image every epilepsy patient. For example, children with uncomplicated single febrile seizures and a normal neurological examination do not require an imaging study. Additionally, patients with a well-defined epilepsy syndrome such as idiopathic generalized epilepsy (such as childhood absence epilepsy) or benign epilepsy with centro-temporal spikes (Rolandic Epilepsy) may not require imaging. However, cases of patients with these apparently ‘‘benign’’ epilepsies and structural MRI abnormalities have been published. Thus, it is best to maintain a low threshold to image all patients with epilepsy. Largely, those who absolutely require imaging will declare themselves by an atypical presentation or clinical course. However, all patients with a symptomatic generalized epilepsy or a partial epilepsy that do
Imaging evaluation of epilepsy
not fit into the above ‘‘benign’’ categories should unequivocally be imaged. Also, patients with epilepsy who have an uncharacterized or unidentified syndrome should be imaged as well. Because of the aforementioned superiority of MRI over CT to visualize small lesions, MRI is the imaging procedure of choice in the evaluation of the seizure patient, especially in the presence of a focal neurologic exam finding or abnormal EEG. MRI is also indicated in the presence of a normal CT scan or with progressive neurologic symptomatology. A repeated MRI is also indicated at 2–5 years in the context of a previously normal MRI in a patient with persistent seizures. PET and SPECT studies are considered for epilepsy evaluation only in the presurgical setting. PET may provide localizing information when the MRI is normal in any patient considered for focal resection, especially in the setting of nonlesional temporal lobe epilepsy, and in children with refractory infantile spasms with normal MRI and when surgical treatment is contemplated. Ictal SPECT is extremely useful in the localization of extra-temporal foci particularly if the MRI is normal or when the area of epileptogenesis is not well localized. MSI has recently been proven to be a robust method of uncovering epileptiform activity, particularly in the presurgical setting. It can be of great use in guiding intracranial electrode placement, or to narrow the suspected seizure onset zone in nonlesional cases. Along with fMRI, it can also be used for somatosensory and motor cortex localization prior to surgery. fMRI can additionally be used to reliably map language function in the presurgical setting. Lastly, experimental techniques like MRS and DTI will likely be of increased use in the future, but should currently be interpreted with caution. MRS can be useful in surgical candidates, particularly in the setting of temporal lobe epilepsy with normal MRI or those with unusual features. However, normative data is needed for correct interpretation of the MRS studies. The high rate of bilateral
155
abnormalities makes the technique difficult to use for localization alone. However, MRS may serve as an independent surrogate marker of neuronal dysfunction. DTI studies have the potential to uncover underlying networks of epileptogenicity and cognitive function, and appear to be a promising method of visualizing postsurgical changes after resection. However, further validation is required for this modality as well.
Summary Recent developments in imaging technology have allowed the opportunity to visualize brain structure and function at levels that were unimagined only a few years ago. As this field advances, the ability to comprehensively image the brain in patients with epilepsy will allow a greater understanding of their underlying disease processes, and will ensure that those undergoing surgery will experience improved outcomes with sparing of functional cortex.
References 1. Kuzniecky R, Jackson G. Neuroimaging in epilepsy. In: Magnetic resonance in epilepsy. Raven Press; New York: p. 27-48. 2. Kuzniecky R. Structural imaging. In: Engel JJ, editor. Surgical treatment of the epilepsies. 2nd ed. New York: Raven Press; 1993. p. 197-200. 3. Jackson GD, Berkovic SF, Duncan JS, Connelly A. Optimizing the diagnosis of hippocampal sclerosis using magnetic resonance imaging. AJNR Am J Neuroradiol 1993;14:753-62. 4. Kuzniecky R, Bilir E, Gilliam F, Faught E, Palmer C, Morawetz R, Jackson G. Mutimodality MRI in mesial temporal sclerosis: relative sensitivity and specificity. Neurology 1997;49:774-8. 5. Lee Y, Van Tassel P, Bruner JM, Moser RP, Share JC. Juvenile pilocytic astrocytomas: CT and MR characteristics. AJNR Am J Neuroradiol 1989;10:363-70. 6. Ulivelli M, Rocchi R, Vatti G, Giannini F, D’Andrea P, Batani B, Passero S, Battistini N. CAT and MRI in the study of partial epilepsy: comparison of the 2 methods and correlations with EEG. Riv Neurol 1991;61(5):161-5.
2627
2628
155
Imaging evaluation of epilepsy
7. Bauer J, Stefan H, Huk WJ, Feistel H, Hilz MJ, Brinkmann HG, Drushky KF, Neundorfer B. CT, MRI and SPECT neuroimaging in status epilepticus with simple partial and complex partial seizures: case report. J Neurol 1989;236(5):296-9. 8. Duncan R, Patterson J, Hadley, DM, Macpherson P, Brodie MJ, Bone I, McGeorge AP, Wyper DJ. CT, MR and SPECT imaging in temporal lobe epilepsy. J Neurol Neurosurg Psychiatry 1990;53(1):11-15. 9. Zimmerman RD, Leeds, NE, Danziger A. Subdural empyema: CT findings. Radiology 1984;150:417-22. 10. Butcher K, Laidlaw J. Current intracerebral haemorrhage management. J Clin Neurosci 2003;10(2):158-67. 11. Gastaut H, Gastaut JL. Computerized transverse axial tomography in epilepsy. Epilepsia 1976;17:325-36. 12. Wyler AR, Bolender NF. Preoperative CT diagnosis of mesial temporal sclerosis for surgical treatment of epilepsy. Ann Neurol 1983;13(1):59-64. 13. Jack C, Rydberg CH, Krecke KN, Trenerry MR, Parisi JE, Rydberg JN, Cascino GD, Riederer SJ. Mesial temporal sclerosis: diagnosis with FLAIR versus spin-echo MR imaging. Radiology 1996;199:367-73. 14. Mathieson G. Pathology of temporal lobe foci. In: Penry JK, Daly DD, editors. Complex partial seizures and their treatment. New York: Raven Press; 1975. p. 163-85. 15. Falconer MA. Mesial temporal (Ammon’s horn) sclerosis as a common cause of epilepsy: aetiology, treatment and prevention. Lancet 1974;2:767-70. 16. Babb TL, Pretorius JK. Pathological substrates of epilepsy. In: Wylie E, editor. The treatment of epilepsy: principles and practice. Philadelphia: Lea and Febiger; 1993. p. 55-70. 17. Kuzniecky R, Murro A, King D, Morawetz R, Smith J, Powers R, Yaghmai F, Faught E, Gallagher B, Snead OC. Magnetic resonance imaging in childhood intractable partial epilepsies: pathologic correlations. Neurology 1993;43:681-7. 18. Kuzniecky R. Magnetic resonance imaging in developmental disorders of the cerebral cortex. Epilepsia 1994;35(6):S44-S56. 19. Kuzniecky R, Garcia JH, Faught E, Morawetz RB. Cortical dysplasia in TLE: MRI correlations. Ann Neurol 1991;29:293-8. 20. Fauser S, Huppertz HJ, Bast T, Strobl K, Pantazis G, Altenmueller DM, Feil B, Rona S, Kurth C, Rating D, Korinthenberg R, Steinhoff BJ, Volk B, Schulze-Bonhage A. Clinical characteristics in focal cortical dysplasia: a retrospective evaluation in a series of 120 patients. Brain 2006;129(Pt 7):1907-16. 21. Cascino GD, Jack CR Jr, Parisi JE, Sharbrough FW, Hirschorn KA, Meyer FB, Marsh WR, O’Brien PC. Magnetic resonance imaging-based volume studies in temporal lobe epilepsy: pathological correlations. Ann Neurol 1991;30:31-6. 22. Cendes F, Andermann F, Gloor P, Evans A, JonesGotman M, Watson C, Melanson D, Olivier A,
23. 24.
25. 26.
27.
28.
29. 30.
31.
32.
33.
34.
35. 36.
37.
Peters T, Lopes-Cendes I. MRI volumetric measurement of amygdala and hippocampus in temporal lobe epilepsy. Neurology 1993;43:719-25. Jack C. MRI-based hippocampal volume measuremnts in epilepsy. Epilepsia 1994;35 Suppl 6:14-19. Guerrini R. Focal anomalies of the cortical development and epilepsy: electroclinical features in the bilateral opercular malformations. Boll Lega Ital Epilessia 1990;71:109-11. Norman M, et al. Congenital malformations of the brain. New York: Oxford University Press; 1995. Kuzniecky R, Gilliam F, Morawetz R, Faught E, Palmer C, Black L. Occipital lobe developmental malformations and Epilepsy: clinical spectrum, treatment and outcome. Epilepsia 1997;38(2):175-81. Barkovich A, Kuzniecky R. Neuroimaging of focal malformations of cortical development. J Clin Neurophys 1996;13:(6):481-94. Ferna´ndez G, Effenberger O, Vinz B, Steinlein O, Elger CE, Do¨hring W, Heinze HJ. Hippocampal malformations as a cause of familial febrile convulsions and subsequent hippocampal sclerosis. Neurology 1998; 50:909-17. Barkovich, AJ. Pediatric Neuroimaging. 2nd ed. New York: Raven Press; 1995. Barkovich AJ, Guerrini R, Battaglia G, Kalifa G, N’Guyen T, Parmeggiani A, Santucci M, GiovanardiRossi P, Granata T, D’Incerti L. Band heterotopia: correlation of outcome with MRI parameters. Ann Neurol 1994;36:609-17. Barkovich AJ, Kjos BO, Jackson DE Jr, Norman D. Normal maturation of the neonatal and infant brain: MR imaging at 1.5 T. Radiology 1988;166:173-80. Andersen AR, Waldemar G, Dam M, FuglsangFrederiksen A, Herning M, Kruse-Larsen C. SPECT in the presurgical evaluation of patients with temporal lobe epilepsy – a preliminary report. Acta Neurochir Suppl (Wien) 1990;50:80-3. Gru¨nwald F, Durwen HF, Bockisch A, Hotze A, Kersjes W, Elger CE, Biersack HJ. Technetium-99mHMPAO brain SPECT in medically intractable temporal lobe epilepsy: a postoperative evaluation. J Nucl Med 1991;32(3):388-94. Bauer J, Stefan H, Feistel H, Schu¨ler P, Platsch G, Neubauer U, Neundo¨rfer B. Ictal and interictal SPECT measurements using (99m)Tc-HMPAO in patients suffering from temproal lobe epilepsies. Nervenarzt 1991;62(12):745-9. PET and SPECT in epilepsy [Editorial]. Lancet 1989;1:135‐7. Spencer S. MRI, SPECT, and PET imaging in epilepsy: their relative contributions. Epilepsia 1994;35 Suppl 6: S72-S89. Mountz J. Quantification of the SPECT brain scan. In: Freeman LM, editor. Nuclear medicine annual. New York: Raven Press; 1991. p. 67-98.
Imaging evaluation of epilepsy
38. Krausz Y, Cohen D, Konstantini S, Meiner Z, Yaffe S, Atlan H. Brain SPECT imaging in temporal lobe epilepsy. Neuroradiology 1991;33(3):274-6. 39. Marks DA, Katz A, Hoffer P, Spencer SS. Localization of extra temporal epileptic foci during ictal SPECT. Ann Neurol 1992;31:250-5. 40. Ho SS, Berkovic SF, Newton MR, Austin MC, McKay WJ, Bladin PF. Parietal lobe epilepsy: clinical features and seizure localization by ictal SPECT. Neurology 1994;44:2277-84. 41. Laich E, Kuzniecky R, Mountz J, Liu HG, Gilliam F, Bebin M, Faught E, Morawetz R. Supplementary sensorimotor area epilepsy: seizure localization, cortical propagation and subcortical activation pathways using ictal SPECT. Brain 1997;120:855-64. 42. Harvey AS, Hopkins IJ, Bowe JM, Cook DJ, Shield LK, Berkovic SF. Frontal lobe epilepsy: clinical seizure characteristics and localization with ictal 99mTcHMPAO SPECT. Neurology 1993;43:1966-80. 43. Hogan RE, Cook MJ, Binns DW, Desmond PM, Kilpatrick CJ, Murrie VL, Morris KF. Perfusion patterns in postictal 99mTc-HMPAO SPECT after coregistration with MRI in patients with mesial temporal lobe epilepsy. J Neurol Neurosurg Psychiatry 1997;63 (2):235-9. 44. Chiron C, Ve´ra P, Kaminska A, Cieuta C, Hollo A, Ville D, Gardin I, Stie´venart JL, Dulac O. Ictal SPECT in the epileptic child. Contribution of subtraction interictal images and superposition of with MRI. Rev Neurol (Paris) 1999;155(6–7):477-81. 45. Spanaki MV, Spencer SS, Corsi M, MacMullan J, Seibyl J, Zubal IG. Sensitivity and specificity of quantitative difference SPECT analysis in seizure localization. J Nucl Med 1999;40(5):730-6. 46. O’Brien TJ, So EL, Mullan BP, Hauser MF, Brinkmann BH, Bohnen NI, Hanson D, Cascino GD, Jack CR Jr, Sharbrough FW. Subtraction ictal SPECT co-registered to MRI improves clinical usefulness of SPECT in localizing the surgical seizure focus. Neurology 1998;50(2):445-54. 47. McNally KA, Paige AL, Varghese G, Zhang H, Novotny EJ Jr, Spencer SS, Zubal IG, Blumenfeld H. Localizing value of ictal-interictal SPECT analyzed by SPM (ISAS). Epilepsia 2005;46(9):1450-64. 48. Mauguiere F, Ryvlin P. Intercitical hypometabolism assessed by positron emission tomography (PET) in adult patients with temporal lobe epilepsy. Epilepsies 1991;3(2):2-3. 49. Ryvlin P, Philippon B, Cinotti L, Froment JC, Le Bars D, Mauguie`re F. Functional neuroimaging strategy in temporal lobe epilepsy: a comparative study of 18FDG-PET and 99mTc-HMPAO-SPECT. Ann Neurol 1992;31(6):650-6. 50. Radtke RA, Hanson MW, Hoffman JM, Crain BJ, Walczak TS, Lewis DV, Beam C, Coleman RE, Friedman AH. Temporal PET hypometabolism: predictor of positive outcome after temporal lobectomy. Epilepsia 1993;31:626.
155
51. Theodore WH, Katz D, Kufta C, Sato S, Patronas N, Smothers P, Bromfield E. Pathology of temporal lobe foci: correlation with CT, MRI, and PET. Neurology 1990;40 (5):797-803. 52. Chugani H. The role of PET in childhood epilepsy. J Child Neurol 1994;9:S82-S88. 53. Engel J Jr, Brown WJ, Kuhl DE, Phelps ME, Mazziotta JC, Crandall PH. Pathological findings underlying focal temporal lobe hypometabolism in partial epilepsy. Ann Neurol 1982;12(6):518-28. 54. Toczek MT, Carson RE, Lang L, Ma Y, Spanaki MV, Der MG, Fazilat S, Kopylev L, Herscovitch P, Eckelman WC, Theodore WH. PET imaging of 5-HT1A receptor binding in patients with temporal lobe epilepsy. Neurology 2003;60(5):749-56. 55. Juha´sz C, Chugani DC, Muzik O, Shah A, Asano E, Mangner TJ, Chakraborty PK, Sood S, Chugani HT. Alpha-methyl-L-tryptophan PET detects epileptogenic cortex in children with intractable epilepsy. Neurology 2003;60(6):960-8. 56. Szelies B, Sobesky J, Pawlik G, Mielke R, Bauer B, Herholz K, Heiss WD. Impaired benzodiazepine receptor binding in peri-lesional cortex of patients with symptomatic epilepsies studied by [(11)C]-flumazenil PET. Eur J Neurol 2002;9(2):137-42. 57. Lee SK, Lee DS, Yeo JS, Lee JS, Kim YK, Jang MJ, Kim KK, Kim SK, Oh JB, Chung CK. FDG-PET images quantified by probabilistic atlas of brain and surgical prognosis of temporal lobe epilepsy. Epilepsia 2002;43(9):1032-8. 58. Hammers A, Koepp MJ, Hurlemann R, Thom M, Richardson MP, Brooks DJ, Duncan JS. Abnormalities of grey and white matter [11C]flumazenil binding in temporal lobe epilepsy with normal MRI. Brain 2002;125(Pt 10):2257-71. 59. Hammers A, Koepp MJ, Labbe´ C, Brooks DJ, Thom M, Cunningham VJ, Duncan JS. Neocortical abnormalities of [11C]-flumazenil PET in mesial temporal lobe epilepsy. Neurology 2001;56(7):897-906. 60. Koepp MJ, Richardson MP, Labbe´ C, Brooks DJ, Cunningham VJ, Ashburner J, Van Paesschen W, Revesz T, Duncan JS. 11C-flumazenil PET, volumetric MRI, and quantitative pathology in mesial temporal lobe epilepsy. Neurology 1997;49(3):764-73. 61. Fedi M, Reutens D, Okazawa H, Andermann F, Boling W, Dubeau F, White C, Nakai A, Gross DW, Andermann E, Diksic M. Localizing value of a-methylL-tryptophan PET in intractable epilepsy of neocortical origin. Neurology 2001;57(9):1629-36. 62. Juha´sz C, Nagy F, Muzik O, Watson C, Shah J, Chugani HT. [11C]Flumazenil PET in patients with epilepsy with dual pathology. Epilepsia 1999;40(5):566-74. 63. Lamusuo S, Pitka¨nen A, Jutila L, Ylinen A, Partanen K, Ka¨lvia¨inen R, Ruottinen HM, Oikonen V, Na˚gren K, Lehikoinen P, Vapalahti M, Vainio P, Rinne JO. [11C] Flumazenil binding in the medial temporal lobe in patients with temporal lobe epilepsy: correlation with
2629
2630
155 64.
65.
66.
67.
68.
69. 70.
71.
72.
73.
74.
75.
76. 77.
Imaging evaluation of epilepsy
hippocampal MR volumetry, T2 relaxometry, and neuropathology. Neurology 2000;54(12):2252-60. Savic I, Thorell JO, Roland P. [11C]flumazenil positron emission tomography visualizes frontal epileptogenic regions. Epilepsia 1995;36(12):1225-32. Hammers A, Koepp MJ, Richardson MP, Hurlemann R, Brooks DJ, Duncan JS. Grey and white matter flumazenil binding in neocortical epilepsy with normal MRI. A PET study of 44 patients. Brain 2003;126(Pt 6):1300-18. Tanaka N, Shiga T, Kamada K. Combined imaging of epileptic foci and brain metabolism using MEG and FDG-PET. No To Shinkei 2003;55(3):247-50. Yoshinaga H, Nakahori T, Ohtsuka Y, Oka E, Kitamura Y, Kiriyama H, Kinugasa K, Miyamoto K, Hoshida T. Benefit of simultaneous recording of EEG and MEG in dipole localization. Epilepsia 2002;43(8):924-8. Moore KR, Funke ME, Constantino T, Katzman GL, Lewine JD. Magnetoencephalographically directed review of high-spatial-resolution surface-coil MR images improves lesion detection in patients with extratemporal epilepsy. Radiology 2002;225(3):880-7. Yumoto, M. Recent advances of MEG. Rinsho Byori 2000;48(1):38-41. Baumgartner C, Pataraia E, Lindinger G, Deecke L. Neuromagnetic recordings in temporal lobe epilepsy. J Clin Neurophysiol 2000;17(2):177-89. Patt S, Steenbeck J, Hochstetter A, Kraft R, Huonker R, Haueisen J, Haberland N, Ebmeier K, Hliscs R, Fiehler J, Nowak H, Kalff R. Source localization and possible causes of interictal epileptic activity in tumor-associated epilepsy. Neurobiol Dis 2000;7(4):260-9. Ganslandt O, Fahlbusch R, Kober H, Gralla J, Nimsky C. Use of magnetoencephalography and functional neuronavigation in planning and surgery of brain tumors. Nervenarzt 2002;73(2):155-61. Amo C, Saldan˜a C, Hidalgo MG, Maestu´ F, Ferna´ndez A, Arrazola J, Ortiz T. Magnetoencephalographic localization of peritumoral temporal epileptic focus previous surgical resection. Seizure 2003;12(1):19-22. Stefan H, Schneider S, Feistel H, Pawlik G, Schu¨ler P, Abraham-Fuchs K, Schlegel T, Neubauer U, Huk WJ. Ictal and interictal activity in partial epilepsy recorded with multichannel magnetoelectroencephalography: correlation of electroencephalography/electrocorticography, magnetic resonance imaging, single photon emission computed tomography, and positron emission tomography findings. Epilepsia 1992;33(5):874-87. Crisp D, Weinberg H, Podrouzek KW. Imaging techniques in the localization of epileptiform abnormalities. Int J Neurosci 1991;60(1–2):33-57. Imai K. Magnetoencephalography analysis of epileptic foci. No To Hattatsu 2001;33(2):146-52. Shiraishi H, Watanabe Y, Watanabe M, Inoue Y, Fujiwara T, Yagi K. Interictal and ictal magnetoencephalographic study in patients with medial frontal lobe epilepsy. Epilepsia 2001;42(7):875-82.
78. Assaf BA, Karkar KM, Laxer KD, Garcia PA, Austin EJ, Barbaro NM, Aminoff MJ. Magnetoencephalography source localization and surgical outcome in temporal lobe epilepsy. Clin Neurophysiol 2004;115:2066-76. 79. Knowlton RC, Elgavish R, Howell J, Blount J, Burneo JG, Faught E, Kankirawatana P, Riley K, Morawetz R, Worthington J, Kuzniecky RI. Magnetic source imaging versus intracranial electroencephalogram in epilepsy surgery: a prospective study. Ann Neurol 2006;59(5):835-42. 80. Detre JA, Sirven JI, Alsop DC, O’Connor MJ, French JA. Localization of subclinical ictal activity by functional magnetic resonance imaging: correlation with invasive monitoring. Ann Neurol 1995;38:618-24. 81. Lantz G, Spinelli L, Menendez RG, Seeck M, Michel CM. Localization of distributed sources and comparison with functional MRI. Epileptic Disord 2001;Special Issue:45‐58. 82. Detre JA, Floyd TF. Functional MRI and its applications to the clinical neurosciences. Neuroscientist 2001; 7(1):64-79. 83. Turner R. fMRI: methodology – sensorimotor function mapping. Adv Neurol 2000;83:213-20. 84. Holloway V, Gadian DG, Vargha-Khadem F, Porter DA, Boyd SG, Connelly A. The reorganization of sensorimotor function in children after hemispherectomy. A functional MRI and somatosensory evoked potential study. Brain 2000;123(Pt 12):2432-44. 85. Gaillard, WD. Structural and functional imaging in children with partial epilepsy. Ment Retard Dev Disabil Res Rev 2000;6(3):220-6. 86. Matthews PM, Clare S, Adcock J. Functional magnetic resonance imaging: clinical applications and potential. J Inherit Metab Dis 1999;22(4):337-52. 87. Masuoka LK, Anderson AW, Gore JC, McCarthy G, Spencer DD, Novotny EJ. Functional magnetic resonance imaging identifies abnormal visual cortical function in patients with occipital lobe epilepsy. Epilepsia 1999;40(9):1248-53. 88. Fahlbusch R, Ganslandt O, Nimsky C. Intraoperative imaging with open magnetic resonance imaging and neuronavigation. Childs Nerv Syst 2000;16(10–11):829-31. 89. Morris GL III, Mueller WM, Yetkin FZ, Haughton VM, Hammeke TA, Swanson S, Rao SM, Jesmanowicz A, Estkowski LD, Bandettini PA. Functional magnetic resonance imaging in partial epilepsy. Epilepsia 1994; 35(6):1194-8. 90. Gaillard WD, Hertz-Pannier L, Mott SH, Barnett AS, LeBihan D, Theodore WH. Functional anatomy of cognitive development: fMRI of verbal fluency in children and adults. Neurology 2000;54(1):180-5. 91. Hammeke TA, Bellgowan PS, Binder JR. fMRI: methodology–cognitive function mapping. Adv Neurol 2000;83:221-33. 92. Lehe´ricy S, Cohen L, Bazin B, Samson S, Giacomini E, Rougetet R, Hertz-Pannier L, Le Bihan D, Marsault C, Baulac M. Functional MR evaluation of temporal and frontal language dominance compared with the Wada test. Neurology 2000;54(8):1625-33.
Imaging evaluation of epilepsy
93. Ferna´ndez G, de Greiff A, von Oertzen J, Reuber M, Lun S, Klaver P, Ruhlmann J, Reul J, Elger CE. Language mapping in less than 15 minutes: real-time functional MRI during routine clinical investigation. Neuroimage 2001;14(3):585-94. 94. Hanakawa T, Ikeda A, Sadato N, Okada T, Fukuyama H, Nagamine T, Honda M, Sawamoto N, Yazawa S, KuniedaT, Ohara S, Taki W, HashimotoN, Yonekura Y, Konishi J, Shibasaki H. Functional mapping of human medial frontal motor areas. The combined use of functional magnetic resonance imaging and cortical stimulation. Exp Brain Res 2001;138(4):403-9. 95. Gaillard WD, Balsamo L, Xu B, Grandin CB, Braniecki SH, Papero PH, Weinstein S, Conry J, Pearl PL, Sachs B, Sato S, Jabbari B, Vezina LG, Frattali C, Theodore WH. Language dominance in partial epilepsy patients identified with an fMRI reading task. Neurology 2002; 59(2):256-65. 96. Binder JR, Swanson SJ, Hammeke TA, Morris GL, Mueller WM, Fischer M, Benbadis S, Frost JA, Rao SM, Haughton VM. Determination of language dominance using functional MRI: a comparison with the Wada test. Neurology 1996;46(4):978-84. 97. Detre JA, Maccotta L, King D, Alsop DC, Glosser G, D’Esposito M, Zarahn E, Aguirre GK, French JA. Functional MRI lateralization of memory in temporal lobe epilepsy. Neurology 1998;50(4):926-32. 98. Dupont S, Van de Moortele PF, Samson S, Hasboun D, Poline JB, Adam C, Lehe´ricy S, Le Bihan D, Samson Y, Baulac M. Episodic memory in left temporal lobe epilepsy: a functional MRI study. Brain 2000;123(Pt 8): 1722-32. 99. Dupont S, Samson Y, Van de Moortele PF, Samson S, Poline JB, Adam C, Lehe´ricy S, Le Bihan D, Baulac M. Delayed verbal memory retrieval: a functional MRI study in epileptic patients with structural lesions of the left medial temporal lobe. Neuroimage 2001;14 (5):995-1003. 100. Golby AJ, Poldrack RA, Illes J, Chen D, Desmond JE, Gabrieli JD. Memory lateralization in medial temporal lobe epilepsy assessed by functional MRI. Epilepsia 2002;43(8):855-63. 101. Jackson GD, Connelly A, Cross JH, Gordon I, Gadian DG. Functional magnetic resonance imaging of focal seizures. Neurology 1994;44:850-6. 102. Krakow K, Woermann FG, Symms MR, Allen PJ, Lemieux L, Barker GJ, Duncan JS, Fish DR. EEGtriggered functional MRI of interictal epileptiform activity in patients with partial seizures. Brain 1999;122(Pt 9):1679-88. 103. Krakow K, Allen PJ, Lemieux L, Symms MR, Fish DR. Methodology: EEG-correlated fMRI. Adv Neurol 2000;83:187-201.
155
104. Thesen T., Carlson C.E., McDonald C.M., Kuzniecky R.I., Hagler D.J., Stout J.D., Nearing K.I., Dale A.M., Barr W.B., Devinsky O., & Halgren E. Comapring functional MRI and MEG in the study of language processing. American Epilepsy Society, Philadelphia, USA 2007. 105. Be´nar CG, Grova C, Kobayashi E, Bagshaw AP, Aghakhani Y, Dubeau F, Gotman J. EEG-fMRI of epileptic spikes: concordance with EEG source localization and intracranial EEG. Neuroimage 2006;30 (4):1161-70 106. Kuzniecky R, Elgavish GA, Hetherington HP, Evanochko WT, Pohost GM. In vivo 31P nuclear magnetic resonance spectroscopy of human temporal lobe epilepsy. Neurology 1992;42(8):1586-90. 107. Gadian DG, Connelly A, Duncan JS. 1H magnetic resonance spectroscopy in the investigation of intractable epilepsy. Acta Neurol Scand Suppl 1994;152:116-21. 108. Connelly A, Jackson GD, Duncan JS, King MD, Gadian DG. Proton MRS in temporal lobe epilepsy. Neurology 1994;44(8):1411-17. 109. Cendes F, Stanley JA, Dubeau F, Andermann F, Arnold DL. Proton magnetic resonance spectroscopic imaging for discrimination of absence and complex partial seizures. Ann Neurol 1997;41:74-81. 110. Hetherington H, Kuzniecky R, Pan J, Mason G, Morawetz R, Harris C, Faught E, Vaughan T, Pohost G. Proton MRS in human temporal lobe epilepsy at 4.1 T. Ann Neurol 1995;38(3):396-404. 111. Kuzniecky R, Hugg JW, Hetherington H, Butterworth E, Bilir E, Faught E, Gilliam F. Relative utility of proton MRS and volumetry in the lateralization of mesial temporal sclerosis. Neurology 1998;51(1): 66-71. 112. Kuzniecky R, Palmer C, Hugg J, Martin R, Sawrie S, Morawetz R, Faught E, Knowlton R. Magnetic resonance spectroscopic imaging in temporal lobe epilepsy: neuronal dysfunction or cell loss? Arch Neurol 2001;58 (12):2048-53. 113. Ng TC, Comair YG, Xue M, So N, Majors A, Kolem H, Luders H, Modic M. Temporal lobe epilepsy: presurgical localization with proton chemical shift imaging. Radiology 1994;193:465-71. 114. Yogarajah M, Duncan JS. Diffusion based magnetic resonance imaging and tractography in epilepsy. Epilepsia 2008;49(2):189-200. 115. Gross DW, Concha L, Beaulieu C. Extratemporal white matter abnormalities in mesial temporal epilepsy demonstrated with diffusion tensor imaging. Epilepsia 2006;47(8):1360-3. 116. Concha L, Beaulieu C, Wheatley BM, Gross DW. Bilateral white matter diffusion changes persist after epilepsy surgery. Epilepsia 2007;48(5):931-40.
2631
Management of Epilepsy
151
Indications for Surgical Management of Epilepsy H. G. Wieser . D. Zumsteg
General Principles of Epilepsy Surgery Macewen [1] and Horsley [2] are frequently considered to be the forerunners of modern epilepsy surgery. One could say that Horsley integrated Hughling Jackson’s achievements into surgical action [3]. Foerster [4] and Krause and Schum [5] were the pioneers in epilepsy surgery in Germany, providing remarkable results emphasizing the importance of the excision of the ‘‘primary convulsing center,’’ that is, the ‘‘discharging lesion.’’ At that time, the ‘‘primary convulsing center’’ was basically defined by the macroscopic location of the cortical lesion and by clinical ictal semiology. In the 1950s, Bailey and Gibbs [6] and the Montreal pioneers Penfield and Jasper [7]; Penfield and Jasper [23] were the first to use electroencephalographic recordings in order to plan and perform epilepsy surgery. The advance of stereo-electroencephalography, established in St. Anne, Paris, by Talairach [8] and Bancaud [9], has contributed enormously to our understanding of the origin and spread of seizure discharges, thus revolutionizing the concept of the primary epileptogenic zone. Today, many authors proffer a more pragmatic definition to determine the extent of resection. A simplified definition of the epileptogenic zone is ‘‘the minimum amount of cortex that must be resected, inactivated or completely disconnected in order to produce seizure-freedom.’’ The concept of the primary epileptogenic zone will be discussed below in greater detail. A number of attempts have been undertaken to categorize epilepsy surgery procedures. To this aim, ‘‘curative’’ procedures have been #
Springer-Verlag Berlin/Heidelberg 2009
distinguished from ‘‘palliative’’ procedures. Curative epilepsy surgery aims at total seizure control by complete resection of the seizure generating area, whereas the primary aim of palliative surgery is the amelioration of seizure frequency or severity rather than seizure freedom (although this may occasionally occur). ‘‘Palliative’’ procedures such as callosotomy or multiple subpial transection interrupt pathways that are important for seizure spread, or excise secondary ‘‘amplifier structures,’’ as with palliative amygdalohippocampectomy. Furthermore, lesionectomy or lesion-oriented epilepsy surgery has been differentiated from epilepsy-oriented nonlesional surgery, that is epilepsy surgery sensu stricto (often being used synonymously with epilepsy surgery in magnetic resonance imaging (MRI) negative patients). Finally, acknowledged epilepsy surgery techniques have been differentiated from procedures in evaluation or experimental procedures, and standardized techniques have been distinguished from so-called tailored resections. It is obvious that so-called standardized resections, such as selective amygdalohippocampectomy or anterior temporal lobe resection, may frequently turn into tailored resections, based on preoperative or intraoperative investigations such as electrocorticography or functional brain mapping. Individually tailored resections comprise topectomies and some larger resections. From the point of view of the strategy, a ‘‘stepwise’’ approach might be considered with repeated surgery if the first therapeutic intervention was not successful. The main criteria for epilepsy surgery have been formulated by Walker [10] as early as 1974.
2544
151
Indications for surgical management of epilepsy
According to his suggestions, the following criteria have to be met to qualify for resective ‘‘curative’’ epilepsy surgery: (1) focal or regional seizure onset, (2) drug intractability, (3) seizures represent a severe handicap, (4) seizures exist for at least 2 years without tendency for remission and despite adequate medical treatment, (5) sufficient general and mental health state of the patient who is sufficiently motivated and compliant in order to collaborate preoperatively, intraoperatively (if necessary) and postoperatively. Some modifications of Walker’s criteria concern (1) the demand for early surgery (at least in certain epilepsy syndromes such as mesial temporal lobe epilepsy, MTLE), (2) indications for ‘‘palliative’’ surgery, and (3) a more liberal indication in children. With the advent of modern non-invasive structural and functional imaging technique and the improved microsurgical techniques the surgical treatment of epilepsies has multiplied worldwide, but still remains underutilized to date. Taking into consideration the amount of difficult-to-treat or pharmacotherapy-resistant epilepsy patients, the question ‘‘when a patient should be considered for epilepsy surgery and undergo presurgical evaluation’’ is of major importance.
When Should Epilepsy Surgery Be Envisaged? Based on studies such as that of Kwan and Brodie [11] it has been estimated that about 40% of newly diagnosed epilepsy patients are potential candidates for presurgical evaluation of some kind (> Figure 151-1). The authors prospectively studied the response to antiepileptic drug (AED) treatment in 525 patients aged 9–93 years with newly diagnosed epilepsy. The patients were grouped into idiopathic, symptomatic and cryptogenic epilepsy. Seizure-freedom was defined as having no seizure during one year. As expected,
. Figure 151-1 About 40% of newly diagnosed epilepsies are difficultto-treat (‘‘pharmacotherapy resistant’’ = ‘‘refractory’’) and may be considered potential candidates for epilepsy surgery (after [11])
seizures persisted in a higher percentage in the symptomatic and cryptogenic group compared to the idiopathic group. Patients with a history of 20 seizures before initiation of AED therapy were less likely to become seizure-free. There was no difference in response with respect to ‘‘old’’ versus ‘‘new’’ AEDs. Factors that influenced response to AED treatment were age, gender, weight, co-morbidity, history of psychiatric illnesses and life style. The prediction of the response to AED treatment is key to the decision as to whether a patient is a candidate for epilepsy surgery. While there is no doubt that the response to newly administered AED highly depends on the past treatment history, there remains a controversy as to how many AEDs and in which combinations need to be tried to determine drug-resistance. In a study including 478 patients, Schiller and Najjar [12] found that seizure-free rates decreased from 61.8% for the first AED to 41.7% for the second AED. After failure of two to five past AED, the percent of patients rendered seizure-free by the newly administered AED dropped to 16.6%, and to 0% after failure of six or seven past AEDs. This response curve corresponded to
Indications for surgical management of epilepsy
a mono-exponential function with a maximal response of 61.8% and half-decay constant of 1.5 AEDs. The response curve describing a 50% reduction in seizure frequency corresponded to a mono-exponential function with a maximal response of 85.3% and half-decay constant of 2 AEDs. The authors identified three additional independent prognostic factors predicting the response to AEDs: type of epilepsy, duration of epilepsy, and number of seizures in the 3 months prior to AED initiation. Although relative drug-resistance may be diagnosed after failure of two past AEDs, the results of this study suggest that absolute drug resistance would require failure of 6 AEDs, considering that a moderate but significant minority of patients (16.6%) is rendered seizure-free by addition of another AED even after failure of 2–5 past AEDs. It is also pertinent to note that the prognosis of AED resistance clearly depends on the type of epilepsy. In patients with mesial temporal sclerosis as an underlying pathology of temporal lobe epilepsy (TLE), for example, the prognosis is known to be especially poor, with incomplete seizure control in 58–90% [13]; [14]. Therefore, in certain surgically amenable epilepsy syndromes, a history of seizures of 2 years might be sufficient to proceed to epilepsy surgery, if no satisfactory response to adequate AED trials could be obtained. In other syndromes such as Rasmussen’s encephalitis, on the other hand, clear neurological deficits are mandatory before functional hemispherectomy can be suggested. The burden of suffering epileptic seizures despite high-dose AEDs, the chance of benefit from epilepsy surgery and the risks of side effects have to be carefully and individually balanced by taking into consideration all findings of the presurgical evaluation. Many other factors including age, co-morbidity, history of psychiatric illnesses, life style, need to drive, professional and psychosocial variables need also to be considered.
151
With respect to patients with TLE, there is evidence from a recent randomized controlled trial of surgical versus medical therapy that anteromesial temporal resection is safe and more effective than medical therapy [15]. Wiebe [16] concluded that the number of patients needed-totreat for one patient to become free of disabling seizures is two; a number that is clearly superior to most other therapeutic interventions in neurology. This view is further substantiated by the results of other non-randomized trials on temporal lobe surgery showing that about two-thirds of patients become seizure-free (compared to 8% with medical therapy) and that quality of life improves early after epilepsy surgery [17]. These improvements sustain on the long term and are both statistically and clinically significant. Surgical morbidity with permanent sequelae of clinical relevance, on the other hand, is only 2%. The results have been shown to be remarkably similar across studies from different parts of the world. Yet, epilepsy surgery still remains underutilized in developed countries, and does not exist in all but a few developing countries. Current randomized trials are underway to explore the effect of early surgery versus optimum medical therapy on the prevention of disability in patients with mesial temporal lobe epilepsy (MTLE), and to examine the effectiveness of novel interventions, such as minimally invasive surgery and deep brain stimulation.
The Epileptic Focus Concept – Focal Seizure Onset Resective curative epilepsy surgery relies on the concept of the ‘‘epileptic focus,’’ which consists of several zones. Talairach and Bancaud [18] described three zones, namely a lesional, an irritative and an epileptogenic zone. Lu¨ders enriched this concept in 1992 by adding a functional deficit zone and a symptomatogenic zone (> Figure 151-2). Furthermore, many authors
2545
2546
151
Indications for surgical management of epilepsy
. Figure 151-2 The concept of the epileptic focus with its various zones
distinguished a primary epileptogenic zone from a secondary epileptogenic zone. In TLE, the secondary epileptogenic zone has been discussed in relation to the ‘‘mirror focus’’ and ‘‘kindling.’’ Lu¨ders [19] distinguished between the actual seizure onset zone and the potential seizure onset zone in order to better account for the theoretical difference between the epileptogenic zone and the seizure onset zone (> Figure 151-3). Lu¨ders suggested that incomplete resection of both the actual or the potential seizure onset zone may result in incomplete seizure control. The experimental focal seizure model of Wyler and Ward [20] provides a theoretical basis for some possible mechanisms of seizure onset and is particularly applicable to seizure provoking and seizure preventing factors (> Figure 151-4). In this model, intrinsically abnormal Class I epileptic neurons are distinguished from labile Class II neurons and the precipitation of seizures by specific (and probably to a lesser degree unspecific) stimuli depends on the recruitment of Class II neurons. It is our belief that specific seizure precipitating mechanisms and specific
. Figure 151-3 The ‘‘actual’’ versus ‘‘potential’’ seizure onset zone in relation to seizure outcome following resection (after [19])
seizure preventing strategies share common features, insofar as both may act on Class II neurons. With the recruitment of Class II neurons into the discharge behavior of Class I neurons, a ‘‘critical mass’’ is reached. With the physiological occupation of Class II neurons by relevant stimulation, in turn, the critical mass is not reached and the impending seizure can thus be aborted. In a patient with ictal musical hallucinations characterized by long-lasting seizure discharges in the Heschl’ gyrus in stereo-EEG recordings, Wieser [21] showed that the physiological occupation of Class II neurons by appropriate stimulation (presentation of music similar to that hallucinated) may indeed modify the seizure discharges.
How to Define the Seizure Onset Zone and the Extent of Cortical Excision The presurgical evaluation of a potential candidate for epilepsy surgery aims at the precise characterization of the zones of the epileptic focus. Even if one thinks more in terms of
Indications for surgical management of epilepsy
151
. Figure 151-4 The ‘‘epileptic focus’’ model with three classes of neurons categorized according to their burst index [20]. Note that Class I neurons are intrinsically abnormal, and characterized by the paroxysmal depolarization shift (PDS)
distributed networks than of a ‘‘focus,’’ the characterization of the various zones including the seizure spread remains essential for the success of epilepsy surgery.
The Lesional and Functional Deficit Zone It is generally accepted that the epileptogenic zone lies within or in close spatial neighborhood of the macroscopic lesion (if present) in the majority of patients. However, there are cases in which microscopic abnormalities that are not necessarily visible in the MRI may substantially extend from the macroscopic lesion. A typical example would be malformations of cortical development. The MRI-visible lesion even might be less epileptogenic than the MRI-invisible area surrounding the lesion. There is ample evidence from tumor surgery that, in addition to the lesion, the surrounding structures (the area of ‘‘early spread’’) has also to be resected in order to assure post-surgical seizure control, at least in
certain histopathological conditions. Moreover, it is obvious that a lesion per se must not necessarily be epileptogenic (in this respect one should remember that resective epilepsy surgery creates a lesion itself). We have performed successful epilepsy surgery in several patients in whom an old and presumably ‘‘innocent’’ (‘‘burned out?’’) lesion was left unchanged. The variability of the epileptogenic zone with respect to time and medical treatment has also to be taken into account. In the following, we shall summarize the advent of modern imaging methods providing information about the lesion and functional deficit zone, before discussing the problem of the localization and delineation of the seizure onset zone. Several modern imaging modalities provide reliable information about structural or functional abnormalities in localization-related epilepsy. We shall give some reference to the findings in the prototype of surgically amenable epilepsy syndromes: the mesial temporal lobe epilepsy with hippocampal sclerosis (MTLEHS) [22]. These modern imaging methods include: a) interictal MRI to study abnormalities of
2547
2548
151
Indications for surgical management of epilepsy
brain structure, b) interictal proton magnetic resonance spectroscopy (proton-MRS) to study reductions of N-acetyl-aspartate (NAA) as a measure of neuronal integrity, c) ictal/interictal single-photon emission computed tomography (SPECT) with [Tc-99 m]technetium-hexamethylpropyleneamine oxime (HMPAO) or with [Tc-99 m]technetium-ethylene cysteine dimer (ECD) to study ictal perfusion changes, d) interictal positron emission tomography (PET) with 2-[F-18] fluoro-2-deoxyglucose (FDG) to study glucose metabolic dysfunction, and e) interictal PET with [C-11]flumazenil to study altered densities of central benzodiazepine receptors (cBZRs). There are other modern imaging methods that may provide valuable data, but their role has not yet been fully elucidated. These methods include ictal functional MRI (fMRI) to study ictal blood flow, postictal MRI with T2 or diffusion weighted sequences to study the effects of single or repeated seizures on local water redistribution, interictal proton-MRS with gamma-aminobutyrate and homocarnosine spectra to study inhibitory neurotransmitter concentrations, interictal proton-MRS with glutamate-glutamine spectra to study excitatory neurotransmitter concentrations, interictal phosphorus-MRS with adenosine triphosphate and inorganic phosphorus spectra to study pH and cellular energy metabolism, interictal SPECT with [I-123]iomazenil to study cBZR density, interictal PET with (S)-[N-methyl-C-11]ketamine to study N-methyl-D-aspartate receptor density, interictal PET with [C-11] alpha-methyl-L-tryptophan to study serotonin synthesis, and interictal PET with [C-11]deuterium-L-deprenyl to study glial density by measuring monoamine oxidase-B concentration. How far these modern functional imaging techniques will improve the accuracy of seizure focus delineation remains to be seen. Interictal MRI is highly sensitive and specific for the detection of neoplasia, dysplasia, vascular malformations and other lesions such as hippocampal sclerosis (HS) in patients with MTLE.
Qualitative MRI interpretation and quantification of hippocampal volume (volumetry) and water density (T2 relaxometry) is gold standard for diagnosing HS in vivo, but mild HS may be missed. Hippocampal MRI abnormalities in patients with MTLE-HS typically are unilateral or asymmetric, but may also occur bilaterally symmetric. Moreover, brain MRI frequently shows abnormalities of volume (atrophy) and signal (T2 increase or T1 decrease) in structures outside the hippocampus, though usually ipsilateral to the sclerotic hippocampi. The MRI hippocampal abnormalities most typically found in patients with MTLE-HS are: Atrophy (detected in 90–95% of cases in which HS is found in resected tissue), loss of internal architecture (in 60–95%), T2 increase (in 80–85%), and T1 decrease (in 10–95%). The most typical extrahippocampal abnormalities are atrophy-signal alterations of the ipsilateral amygdala, temporal neocortex, temporal lobe white matter, fornix, mammillary body, insula, thalamus and basal frontal cortex, as well as atrophy-signal alterations of the contralateral hippocampus (which is usually less severe than the ipsilateral hippocampal alterations). A diffuse hemispheric atrophy may occur ipsilaterally to the hippocampal atrophy but is rare. Interictal proton-MRS assesses neuronal integrity by measurement of NAA, usually by comparing the concentrations of this neuron-specific chemical with concentrations of choline or creatine. The typical proton-MRS finding in patients with MTLE-HS is a decreased NAA to choline/ creatine ratio. However, proton-MRS measurements are usually limited to a small number of relatively large voxels, thus not encompassing the entire brain like other brain mapping techniques such as MRI, SPECT and PET. Subsequent determination of spectra in many voxels is rendered impractical by the relatively long time required to obtain the spectra. The limited signal to noise ratio of proton-MRS impedes sampling of small brain tissue volumes.
Indications for surgical management of epilepsy
Ictal/interictal SPECT illustrates cerebral perfusion during and shortly after a single seizure (it is mostly carried out by comparing ictal perfusion with interictal baseline perfusion). In patients with unilateral MTLE-HS, interictal [Tc-99 m]HMPAO or [Tc-99 m]ECD retention is commonly mildly or moderately decreased in the ipsilateral anterior temporal lobe, but false lateralization may occur in about 10% of patients. However, with injection of the radioligand during a complex partial seizure, the temporal lobe of ictal onset usually shows a great rise in radioligand retention, and false lateralization is quite rare when using co-registered ictal versus interictal temporal lobe perfusion maps. It is not surprising that ictal SPECT commonly depicts areas of both ictal onset and most intense seizure propagation. Peri-ictal SPECT studies of TLE seizures have shown a characteristic evolution in regional perfusion, the region of ictal hyperperfusion declining to severe hypoperfusion for several minutes postictally, then rising back to a milder degree of interictal hypoperfusion within about 20 min after seizure onset. This phenomenon has been called the ‘‘postictal switch,’’ since it may cause false lateralization when early postictal scans, erroneously taken for ictal scans, are compared with interictal scans. Ictal/interictal SPECT studies require considerable expertise in timing of radioligand injection. It is obvious that radioligand injection must be performed under surveillance of continuous electroencephalography. Interictal FDG PET assesses the distribution of interictal glucose metabolism, based on the deoxyglucose model. Glucose metabolism and cerebral blood flow appear to be uncoupled in the interictal state of TLE. False lateralization of temporal lobe hypometabolism in interictal FDG PET scans is rare in patients with MTLE-HS (unlike with SPECT). The typical interictal FDG-PET findings in patients with MTLE-HS are anterior temporal lobe hypometabolism (ipsilateral to HS, or bilateral but greater on the
151
side of HS) and extratemporal hypometabolism ipsilateral to the predominant anterior temporal hypometabolism, usually including the thalamus, basal ganglia, insula, inferior frontal cortex and lateral parietal cortical areas. Research and clinical application of FDG PET is facilitated by co-registration to the subject’s MRI and by quantitative analysis (which of course also holds true for ictal/interictal SPECT and most other functional imaging techniques). Interictal FMZ PET illustrates the distribution of cBZR density, thereby mapping the density of the GABAA receptor-chloride ionophore complex, which is the principal inhibitory site of the mammalian cortex. Severe decrease of cBZR density in sclerotic hippocampi is associated with the loss of principal CA1 neurons and other mesial temporal subregions. However, based on autoradiographic studies of human surgical specimens it has been shown that the severity of cBZR loss may exceed the severity of neuronal loss in HS. Mapping of cBZR density with FMZ PET is considered to be more useful than cBZR mapping with IMZ SPECT. The typical interictal FMZ-PET findings in patients with MTLE-HS are decreased cBZR densities in anterior mesial temporal regions (ipsilateral to HS) and extratemporal cBZR density decreases ipsilateral to predominant anterior mesial temporal cBZR density decreases, occasionally including the thalamus and the insula. Reduction in cBZR densities appears to be more focal than ictal hyperperfusion (SPECT) or interictal glucose hypometabolism (FDG PET) in patients with MTLE-HS.
The Symptomatogenic Zone – Seizure Semiology Very reliable information on the location of the seizure onset zone can be derived from the ictal semiology, in particular the aura [21,23,24]. The symptomatogenic zone has been defined as
2549
2550
151
Indications for surgical management of epilepsy
the area of cortex that, when activated, produces the initial ictal symptoms or signs. It might include cortical areas a` distance to the actual seizure onset zone that become activated (in the case of ‘‘positive’’ symptoms) or deactivated (in the case of ‘‘negative’’ symptoms) due to seizure spread. If the seizure onset zone happens to be located in a ‘‘silent’’ or ‘‘non-eloquent’’ brain region, the early seizure discharge might be subclinical, and the first signs and symptoms might indicate early seizure spread. Several studies on the anatomo-electroclinical seizure characteristics have been undertaken and culminated in the Bancaud and Talairach’s approach of stereo-electroencephalography [8,9, 21,25]. It is important to note that the Talairach and Bancaud concept included ‘‘early seizure spread’’ (without defining precisely what ‘‘early’’ means) in the definition of the epileptogenic zone (as a part of the potential seizure onset), and the authors reasoned that it should be surgically removed to obtain better surgical results. The study of the signs and symptoms of a seizure with the primictal (aura-)symptoms and the facultative ‘‘march of symptoms’’ in the time domain reveals often a typical gestalt of a seizure in a given syndrome (such as MTLE), which might be even more characteristic as a single sign and symptom. Localizing and lateralizing signs and symptoms have been demonstrated in various correlation analyses [22].
The Irritative Zone The irritative zone has been defined as the area of cortex that generates interictal spikes. It can be measured by non-invasive or invasive EEG, MEG and fMRI. The irritative zone is usually more extended than the seizure onset zone. Since the pioneering days of Penfield and Jasper [23] it is clear that the cortex outside the borders of a successful surgery can produce interictal spikes, irrespective of the precision of the methodology
used to measure the irritative zone with spikes. This is maybe best illustrated in patients with bilateral independent mesial temporal spikes who can be rendered seizure-free by unilateral mesial temporal resection. In Penfield and Jasper’s terminology, spikes originating in areas that necessarily have to be resected are labeled red spikes, as opposed to green spikes originating from cortical areas that need not to be resected (because they constitute a part of the irritative zone that does not overlap with the epileptogenic zone). The problem of differentiation of green and red spikes is aggravated by the fact that the extent of the irritative zone is regularly underestimated in scalp EEG recordings, given the relative insensitivity of scalp EEG signals in comparison to direct intracranial EEG recordings. Invasive recordings, on the other hand, generally cover only a small fraction of the brain, a problem which is known as the sampling problem. Moreover, there are no generally accepted rules to define epileptiform activity in invasive EEG recordings.
Defining the Seizure Onset Zone (i.e., the Cortical Area That Initiates Clinical Seizures) The seizure onset zone is mainly determined by invasive and non-invasive ictal EEG recordings, but may also be identified by ictal SPECT or fMRI and MEG. Video/EEG-long-term monitoring is the preferred method to determine the seizure onset zone of habitual seizures. Due to the limitations of scalp-EEG recordings with regard to sensitivity and artifacts, seizure analysis is usually obtained with the use of several invasive and semi-invasive EEG recording techniques. Scalp-EEG, stereo-EEG, subdural grids and strips and foramen ovale electrodes [26] can be combined according to the needs in a given patient. Each of these techniques has advantages and disadvantages, and their optimal
Indications for surgical management of epilepsy
use requires an experienced team and solid hypotheses on the seizure onset zone(s). Confirmation and exclusion of alternative hypotheses on seizure onset must be taken into account. Occasionally, it may be justified to restrict the invasive techniques to the detection of a surgically amenable region. > Figures 151-5 –151-7 illustrate examples of the EEG findings in patients with invasive and semi-invasive presurgical evaluation, as well as with intraoperative ECoG. Various seizure onset patterns may be observed and have been studied with regard to their value for characterizing the epileptogenic zone. The presence of high-frequency oscillations,
151
often referred to as ‘‘rapid discharges,’’ the highfrequency low-amplitude built-up of seizure discharges, slower recruiting rhythm, ictal decrement, and the hypersynchronous seizure onset pattern are considered characteristic electrophysiological patterns that might have different meaning and prediction (> Figure 151-8). A key question in epilepsy surgery is the amount of tissue that has to be removed in order to obtain seizure control. This decision generally depends on the epileptic syndrome, the underlying histopathology and the preoperative findings regarding the actual and potential seizure onset zone. > Figure 151-9 illustrates the
. Figure 151-5 Stereo-electroencephalographic recordings before and at onset of a seizure characterized by an aura of detachment. Note the special montage consisting of uninterrupted bipolar long-distance derivations in the parasagittal plane, connecting the inner (channels 4–7) and outer contacts (channels 8–11) of the depth electrodes 7,1,2,3 and 4, as well as the interrupted short-distance derivations along depth electrodes 1–8 (channels 12–28). The pre-ictal epileptiform activity consists of fairly regular 2/s spike-slow-waves complexes with a clear predominance in the anterior neocortical lateral neocortex (left), whereas the low-voltage high-frequency ictal onset discharge is localized in the anterior hippocampus (electrode 2, contact 2). The positions of the depth electrodes are indicated in the brain map (right) based on the Talairach space. The ten contacts of a depth electrode are numbered from inside outwards (small numbers). The electrodermal activity of the right fingers (EDA R), the electrocardiogram (ECG) and one scalp EEG channel (Cz-Pz) are shown in addition to the depth EEG – Modified from [27]
2551
2552
151
Indications for surgical management of epilepsy
. Figure 151-6 Recording of a short right mesiotemporal seizure by foramen ovale electrodes (FO). Note that the seizure discharge is barely detected in the scalp EEG recording. The upper panels illustrate the free-hand insertion technique and the position of the FO-electrodes (intraoperative x-ray - lateral and ap and post-insertion CT with the electrodes at the level of foramen ovale and deep in the ambient cisterns). The montage for recording from both 10-contact FO-electrodes is a closed chain as depicted at the top right. R, right, L, left
spectrum of resective epilepsy surgery, ranging from hemispherectomy in Rasmussen encephalitis to selective amygdalohippocampectomy in patients with MTLE. As illustrated in a representative example of intraoperative ECoG recording (> Figure 151-7), a individually tailored approach is preferred over standardized resections. We have argued that a selectice amygdalohippocampectomy might be a better approach than standard anterior temporal lobe resection in patients with MTLE and
hippocampal sclerosis or with other lesions confined to this region [29]. The Palm Desert Survey in 1992 and or own data show that seizure outcome with sAHE is at least as good as with TLresection (> Table 151-1) [17,30,31]. With respect to sAHE, a critical question is as to how much of the parahippocampal gyrus should be removed. In a volumetric MRI study on patients with sAHE, we determined a critical volume of about 8 cm, and the removal of at least the anterior part of the parahippocampal gyrus
. Figure 151-7 Intraoperative electrocorticography (ECoG) recorded with a subdural 32-contact-grid (G) from the lateral and inferior temporal lobe and with an intraventricular 4-contact strip electrode (S) from the hippocampus (see positioning on the model to the right). The interictally spiking contacts are marked with full circles in the schema
Indications for surgical management of epilepsy
151 2553
2554
151
Indications for surgical management of epilepsy
. Figure 151‐8 Typical seizure onset patterns recorded with depth electrodes (stereo-EEG) and foramen ovale (FO) electrodes. From top to bottom: High-frequency oscillations (often referred to as ‘‘rapid discharges’’); ictal decrement; 19/s low-amplitude built-up of seizure discharges and 7/s repetitive spike trains with higher frequency discharges in neighbouring contacts; ‘‘hypersynchronous seizure onset pattern’’ with 3/s sharp and slow wave like pattern; so-called ‘‘paroxysmal pattern change’’ (paroxysmale Umschaltung)
was considered necessary in order to obtain good results (> Figure 151-10). This finding has important implications considering that the more radical resection of both the hippocampus and the parahippocampal gyrus bears the danger
of greater postoperative verbal memory impairment, particularly if carried out in the language-dominant hemisphere. In order to predict the postoperative memory performance following unilateral sAHE, we have
151
Indications for surgical management of epilepsy
. Figure 151-9 Types of resective (curative) epilepsy surgery – modified from [28]
. Table 151-1 Results of temporal lobe and neocortical epilepsy surgery (1992 Second Palm Desert Survey) [30]. Note that before 1986, anterior temporal lobe resection (ATL) and selective amygdalohippocampectomy (AHE) were not differentiated (top), and that within the neocortical resections (bottom) no differentiation was done between non-lesional extratemporal resections and lesionectomies Temporal Lobe
1986–1990
Before 1985 Seizur-free Improved Not improved Total (n,%)
ATL 1,296 648 392 2,336
55.5% 27.7% 16.8% 100%
AHE
2,429 860 290 3,579
67.9% 24.0% 8.1% 100%
Neocortical Resections
68.8% 22.3% 9.0% 100%
1986–1990 Extratemporal resections
Before 1985 Seizur-free Improved Not improved Total (n,%)
284 92 37 413
356 229 240 825
43.2% 27.8% 29.1% 100%
363 283 159 805
45.1% 35.2% 19.8% 100%
Lesionectomies 195 63 35 293
66.6% 21.5% 11.9% 100%
2555
2556
151
Indications for surgical management of epilepsy
. Figure 151‐10 Removed volume and percents of resected subcompartements in 30 patients who underwent selective AHE (All, a), grouped into non-lesional (nl) and lesional (l) cases and correlated with seizure outcome (Engel Classes I-IV) – modified from [32]
the co-injection and subsequent SPECT imaging allow for a precise determination of the inactivated structures. In patients at risk for postoperative memory decline, the STLAMT allows for a reliable prediction of postoperative memory performance following sAHE. The refinements enable us to interpret individual test data with high confidence, and the prediction for memory outcome is generally good. Recent research projects study if and to which degree the STLAMT can be substituted by fMRI and special PET activation studies [33–35].
Indications for Palliative Epilepsy Surgery Procedures
developed and steadily improved the so-called Selective Temporal Lobe Amobarbital Memory Test (STLAMT) with short-term inactivation of the to-be-resected brain structures and neuropsychological testing of the effects of this inactivation. The superselective injection of amobarbital into the territory of the anterior choroidal artery (acha) appears to be most suitable. During the last years, a co-injection of amobarbital and SPECT-tracer ([Tc-99 m]HMPAO or [Tc-99 m]ECD) was realized. Together with behavioral and EEG-monitoring (preferentially with foramen ovale- and/or depth electrodes),
Palliative epilepsy surgery can be offered if curative surgery with the aim of postoperative seizure freedom is not possible. Callosotomy (section of the corpus callosum, CCT) is one of the accepted forms of palliative surgery. It can be performed in epilepsy patients with severe falls who are not amenable to resective surgery. Presurgical evaluation seeks to find evidence that the ‘‘generalized’’ seizure discharges are actually generated in one hemisphere, before affecting the contralateral hemisphere by fast propagation via the corpus callosum. Sophisticated EEG analysis techniques such as phase and coherence analysis may occasionally allow for the detection of a short delay between hemispheres. Another technique that has been used to* lateralize ‘‘generalized’’ discharges is the intracarotid Wada test, with the goal to demonstrate disappearance of discharges after inactivation of the epileptically leading hemisphere. CCT has lately attracted attention because it may be performed by means of Gamma Knife radiosurgery [36] (> Figure 151‐11). Stereotactic lesions have been performed with the goal to prevent seizure spread along anatomically pre-existing and preferred seizure propagating pathways. The most often used targets were fornix, thalamic nuclei,
Indications for surgical management of epilepsy
151
. Figure 151‐11 Anterior callosotomy using an open microsurgical approach (fecit Prof. Y. Yonekawa) (top) or by Gamma Knife (bottom) [36] – modified from [26]
Forel-H field, amygdala, hippocampus, and a variety of ‘‘combined’’ lesions in the limbic system [37]. In general, the expectations in stereotactic epilepsy surgery with lesions in these various targets could not be substantiated by the results. Therefore, stereotactic epilepsy surgery with definitive lesioning is no longer recommended, with the exception maybe of rare amygdalar epilepsy [38]. Alternative and experimental treatment options including vagus nerve stimulation (VNS), deep brain stimulation (DBS) and radiotherapy for certain epilepsies have enriched
the armamentarium of epilepsy surgery during the last decade.
References 1. Macewen W. Tumour of the dura mater removed during life in a person affected with epilepsy. Glas Med J 1879;12:210. 2. Horsley V. Brain surgery. Brit Med J 1886;2:670-75. 3. Jackson H. Selected writings of John Hughlings Jackson. In: Taylor J, editor. 2 vols. London: Hodder and Stoughton; 1931‐32.
2557
2558
151
Indications for surgical management of epilepsy
4. Foerster O. Zur operativen Behandlung der Epilepsie. Dtsch Zschr Nervenheilk 1926;89:137-47. 5. Krause F, Schum H. Die spezielle Chirurgie der Gehirnkrankheiten. In: Die epileptischen Erkrankungen, vol. 2. Stuttgart: Enke; 1931 6. Bailey P, Gibbs FA. The surgical treatment of psychomotor epilepsy. JAMA 1951;145:365-70. 7. Jasper H, Pertuisset B, Flanigin H. EEG and cortical electrograms in patients with temporal lobe seizures. Arch Neurol Psychiatr 1951;65:272-90. 8. Talairach J, Bancaud J, Szikla G, Bonis A, Geier S, Vedrenne C. Approche nouvelle de la neurochirurgie de l’e´pilepsie. Me´thodologie ste´re´otaxique et re´sultats the´rapeutiques. Neurochirurgie 1974;20 Suppl 1:1-240. 9. Bancaud J. Apport de l’exploration fonctionelle par voie ste´re´otaxique a` la chirurgie de l’e´pilepsie. Neurochirurgie 1959;5:55-112. 10. Walker AE. Surgery for epilepsy. In: Vinken PJ, Bruyn GW, editors. Handbook of clinical neurology. vol. 15. Amsterdam: North-Holland; 1974. p. 739-57. 11. Kwan P, Brodie MJ. Early identification of refractory epilepsy. N Engl J Med 2000;342:314-9. 12. Schiller Y, Najjar Y. Quantifying the response to antiepileptic drugs: effect of past treatment history. Neurology 2008;70:54-65. 13. Semah F, Picot MC, Adam C, Broglin D, Arzimanoglou A, Bazin B, Cavalcanti D, Baulac M. Is the underlying cause of epilepsy a major prognostic factor for recurrence? Neurology 1998;51:1256-62. 14. Stephen LJ, Kwan P, Brodie MJ. Does the cause of localisation-related epilepsy influence the response to antiepileptic drug treatment? Epilepsia 2001;42:357-62. 15. Wiebe S, Blume WT, Girvin JP, Eliasziw M. Effectiveness and efficiency of surgery for temporal lobe epilepsy study group. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med 2001;345:311-8. 16. Wiebe S. Effectiveness and safety of epilepsy surgery: what is the evidence? CNS Spectr 2004;9:120-2, 126‐32. 17. Wieser HG, Ortega M, Friedman A, Yonekawa Y. Longterm seizure outcome following amygdalohippocampectomy. J Neurosurg 2003;98:751-63. 18. Talairach J, Bancaud J. Lesion, ‘‘irritative’’ zone, and epileptogenic focus. Confin Neurol 1966;27:91-4. 19. Lu¨ders HO, Najm I, Nair D, Widdess-Walsh P, Bingman W. The epileptogenic zone: general principles. Epileptic Disord 2006;8 Suppl 2:S1-9. 20. Wyler AR, Ward AA Jr. Epileptic neurons. In: Lockard JS, Ward AA Jr, editors. Epilepsy, a window to the brain mechanisms. NY: Raven; 1980. p. 51-68. 21. Wieser HG. Electroclinical features of the psychomotor seizure: a stereoelectroencephalographic study of ictal symptoms and chronotopographical seizure patterns including clinical effects of intracerebral stimulation. Stuttgart/London: Gustav Fischer/Butterworths; 1983.
22. Wieser HG. For the ILAE Commission on Neurosurgery of Epilepsy. Mesial temporal lobe epilepsy with hippocampal sclerosis (ILAE Commission Report). Epilepsia 2004;45:695-714. 23. Penfield W, Jasper H. Epilepsy and the functional anatomy of the human brain. London: Churchill; 1954. 24. Wieser HG, Williamson PD. Ictal semiology. In: Engel J Jr, editor. Surgical treatment of the epilepsies. 2nd ed. New York: Raven; 1993. p. 161-71. 25. Kahane P, Landre´ E, Monotti L, Francione S, Ryvlin P. The Bancaud and Talairach view on the epileptogenic zone: a working hypothesis. Epileptic Disord 2006;8 Suppl 2:S16-26. 26. Wieser HG. Foramen ovale and peg electrodes. In: Engel J Jr, Pedley TA, editors chief. Epilepsy: a comprehensive textbook. 2nd ed. PA: Lippincott Williams & Wilkins; 2008. p. 1779-89. 27. Wieser HG. Stereo-electroencephalography. In: Wieser HG, Elger CE, editors. Presurgical evaluation of epileptics. Berlin: Springer; 1987. p. 192-204. 28. Wieser HG, Siegel AM. Operative Therapie. In: Fro¨scher W, Vasella F, editors. Die Epilepsien: Grundlagen, Klinik, Behandlung. Berlin: Walter de Gruyter; 1994. p. 615-28. 29. Wieser HG, Yasargil MG. Selective amygdalohippocampectomy as a surgical treatment of mesiobasal limbic epilepsy. Surg Neurol 1982;17:445-57. 30. Engel J Jr, Van Ness PC, Rasmussen TB, Ojemann LM. Outcome with respect to epileptic seizures. In: Engel J Jr, editor. Surgical treatment of the epilepsies. 2nd ed. NY: Raven; 1993. p. 609-21. 31. Wieser HG, Ha¨ne P. Antiepileptic drug treatment before and after selective amygdalohippocampectomy. Epilepsy Res 2003;55:211-23. 32. Siegel AM, Wieser HG, Wichmann W, Yasargil GM. Relationships between MR-imaged total amount of tissue removed, resection scores of specific mediobasal limbic subcompartments and clinical outcome following selective amygdalohippocampectomy. Epilepsy Res 1990;6:56-65. 33. Wieser HG, Mu¨ller S, Schiess R, Khan N, Regard M, Landis T, Bjeljac M, Buck A, Valavanis A, Yasargil G, Yonekawa Y. The anterior and posterior selective temporal lobe amobarbital tests: angiographical, clinical, electroencephalographical, PET and SPECT findings, and memory performance. Brain Cogn 1997;33:71-97. 34. Henke K, Treyer V, Weber B, Nitsch RM, Hock CH, Wieser HG, Buck A. Functional neuroimaging predicts individual memory outcome after amygdalohippocampectomy. NeuroReport 2003;14:1197-202. 35. Wieser HG, Jones-Gotman M, Smith ML, Zumsteg D, Mueller S, Buck A, Regard M, Yonekawa Y, Valavanis A. Selective temporal lobe amobarbital memory test. Epileptologia 2007;15:85-106.
Indications for surgical management of epilepsy
36. Pendl G, Eder H, Schro¨ttner O, Leber KA. Corpus callosotomy with radiosurgery. Neurosurgery 1999;45:303-7. 37. Wieser HG, Zumsteg D. Subthalamic and thalamic stereotactic recordings and stimulations in patients with intractable epilepsy. Paris: John Libbey Eurotext; 2008.
151
38. Siegfried J, Wieser HG. The actual role of stereotactic operations on deep brain structures in the treatment of medically refractory epilepsies. In: Broggi G, editor. The rational basis of the surgical treatment of epilepsies. John Libbey; London-Paris: p. 113-9.
2559
157 Intraoperative Monitoring in Epilepsy G. Ojemann
Intraoperative monitoring in epilepsy surgery includes two aspects: monitoring to aid in identifying the epileptogenic zone (focus), and the tissue that must be resected to control the seizures, and monitoring to identify the functionally important cortex, so that it can be spared in the resection. Intraoperative monitoring to aid identification of the epileptogenic zone is most often based on electrocorticogram (ECoG) recordings. Intraoperative monitoring to identify the functionally important (eloquent) cortex most often utilizes electrical stimulation mapping. These techniques are discussed in this chapter. There are also several experimental techniques for intraoperative monitoring of eloquent cortex utilizing imaging, including optical imaging of the ‘‘intrinsic’’ signal [1,2], infrared signal [3] or intraoperative functional magnetic resonance imaging [4] that are beyond the scope of this chapter.
Electrocorticographic Monitoring to Aid in Identifying the Epileptogenic Zone Intraoperative electrocorticography as a technique for planning resective surgery for epilepsy was introduced over half a century ago, not long after the development of EEG. Its use was part of the transition of epilepsy surgery from lesionbased localization, usually a meningovascular cicatrix, to electrophysiologically based localization, and resections from predominately sensorymotor cortex to temporal lobe. Jasper and Penfield at the Montreal Neurologic Institute #
Springer-Verlag Berlin/Heidelberg 2009
were its best-known advocates [5]. Intraoperative recordings very rarely capture a seizure, so that interictal spikes (IIS) are the information used to identify the resection requiring apparent focus. The value of corticography, then, is the value of IIS in providing this information. This is particularly an issue for temporal lobe resections for refractory epilepsy, as most patients thought to have extratemporal foci are evaluated with chronic intracranial electrode recordings that provide both ictal and interictal information, whereas patients with clear unilateral temporal foci, particularly if there is imaging evidence of medial temporal lesions such as mesial temporal sclerosis (MTS) usually do not require chronic intracranial recording [6]. In experimental models of partial seizures, the location of IIS and ictal onsets is generally similar. However, early in the use of intraoperative electrocorticography it was recognized that not all IIS indicated the epileptogenic zone, what Rasmussen called ‘‘red’’ and ‘‘green’’ spikes [7]. Specifically, IIS recorded from the insula [7], and those that first appeared following a resection, especially on the resection margins [5], were not considered to indicate tissue requiring resection. The predictive value of lateral cortical IIS that persist after temporal resections has been controversial. Several studies found higher rates of seizure-free patients when post-resection recordings were free of IIS [8–12]. For example, Fiol et al. reported that 72% of patients without residual spikes on post-resection ECoG were seizure-free, compared to only 47% with residual spikes [9]. Others have reported no relation between presence or absence of lateral or basal
2652
157
Intraoperative monitoring in epilepsy
temporal cortical post-resection IIS and their outcome [13–24]. Because of these disparate findings, many epilepsy surgeons have abandoned the use of ECoG in planning the resection, instead basing the extent of resection either on findings from extraoperative chronic intracranial recordings or, for temporal lobe epilepsy, performing standardized resections based only on anatomy, most often a standard measured anterior temporal resection or an amygdalohippocampectomy [25]. Those who continue to use ECoG findings to plan resections consider the location and timing of IIS valuable in predicting outcome. Posterior temporal spikes have been implicated as a predictor of poor outcome [13], as also the failure to include all areas of earliest appearing spikes in the resection [26]. In most published reports, the ECoG has been analyzed visually. Several recent reports suggest that online computer analysis of the preresection ECoG may provide information from neocortical IIS not evident visually that may be useful in planning a resection with a more favorable outcome. Alarcon and coworkers [27,28] investigated subtle (<200 ms) timing differences between IIS and reported that resections that included the earliest spikes had a much more favorable outcome than those that did not. Ortega et al. [29] identified clusters of synchronous neocortical IIS activity, using linear correlation and phase synchronization between spikes, and reported that temporal lobe resections that included these clusters had a more favorable outcome than those that did not, independent of the extent of resection. The pattern of IIS in the ECoG has also been related to the underlying pathology, with continuous spiking and bursts of spikes suggestive of an area of dysplasia [30]. Most reports on the value of IIS in temporal lobe resections for epilepsy have dealt with the spikes recorded from lateral or basal cortex. There is evidence that IIS recorded directly from the hippocampus intraoperatively are useful in
planning the extent of medial temporal resections. Jooma et al. [31] reported that limited mesial resections were just as effective as larger ones, when the extent of the resection was based on removing all IIS recorded from mesial structures. McKhann et al. [32] recorded IIS directly from hippocampus, from a strip electrode inserted in the temporal horn of the lateral ventricle. The posterior extent of these discharges was found to vary, and the medial extent of the resections was generally tailored to include only the portion of hippocampus with discharges, with the exception of a few cases, where memory concerns indicated a smaller hippocampal resection. With this technique, in the initial series of 140 patients, the outcome was similar in those with large or small hippocampal resection, varying from 5 to 46 mm [32]. Moreover, there was a clear relation between the presence or absence of IIS in the residual hippocampus, with 73% of those without discharges seizure-free, and only 29% of those with discharges in the residual hippocampus. The relation held for both patients with and without mesial temporal sclerosis in the resected hippocampus. Similar results were found in a second series of 46 patients [33]. In both these series, the presence or absence of IIS after lateral cortical resections had no significant relation to the outcome. The extent of a mesial temporal lobe resection becomes an important issue in reducing the risk to verbal memory with dominant hemisphere operations, the major morbidity of these operations for medically refractory epilepsy. There are a variety of risk factors for this deficit [34]; one is failure to become seizure-free postoperatively. Thus any strategy to reduce the verbal memory deficit after temporal resections should not reduce the seizure-free rate. Miles et al. [33] evaluated verbal memory performance in a series of 26 dominant hemisphere resections with the mesial extent tailored to the extent of hippocampal IIS. They found a significant correlation between the medial extent of the resection
Intraoperative monitoring in epilepsy
and the change in verbal memory performance on the Wechsler Memory Scale Form I from pre-to postoperative as assessed by combining changes in the immediate and 30-min-delayed logical memory test and the easy and hard paired associate measures. This correlation remained significant for the 15 cases with pathologically normal hippocampi and intact preoperative verbal memory (additional findings indicating a high risk for postoperative memory loss). This study indicates a partial strategy for minimizing memory loss in high-risk cases that are likely to become seizure-free with smaller hippocampal resections based on tailoring to the hippocampal IIS. Using this technique to tailor the dominant temporal resections in the 15 patients of that series at high risk for memory loss, a Class 1 outcome was achieved in seven with minimal (<10%) or no verbal memory loss. Thus, this approach provides a technique for achieving seizure control with little or no verbal memory loss in some patients with medically refractory temporal lobe epilepsy, dominant hemisphere foci, and intact memory with a high risk for postoperative loss, although it does not solve the problem of memory loss for those patients with interictal epileptic activity that involves most or all of the hippocampus. The conditions of intraoperative ECoG recording may also be important in determining its value. In many reports suggesting little value to intraoperative ECoG, recordings were made under general anesthesia, often low concentrations of isoflurane, a short-acting narcotic such as fentanyl, and nitrous oxide. Isoflurane suppresses IIS in a concentration-dependent manner [35–37] while nitrous oxide has variable effects [38–40]. Comparison of IIS recorded intraoperatively under general anesthesia without nitrous oxide to the awake IIS pattern recorded extraoperatively through chronic subdural electrodes showed similar distribution as long as intraoperative IIS frequency exceeded 1 per min [41]. Recordings showing a relation between hippocampal IIS and outcome were obtained
157
under the combination of local anesthesia field block with varying levels of intravenous propofol, so that recordings could be obtained in both an awake state and with different levels of sedation. Hippocampal IIS tends to be more prominent with propofol sedation, although there are a few patients with a reverse pattern and more prominent hippocampal IIS in the awake state. The effect of this difference, if any, on the outcome of surgery has not been investigated. Several different techniques have been used to record intraoperative ECoG. In the area of cortical exposure, carbon tip electrodes have largely supplanted the older silver ball or saline wick electrodes. These multiple-surface electrodes are kept in contact with the pial surface with gentle pressure from an electrode holder attached to the skull. An alternative technique is to record from the exposed cortex through a grid of stainless steel or platinum-iridium electrodes fixed to a silastic sheet, similar to that used for chronic extraoperative recording. Recordings from the cortex that is not exposed by the craniotomy use small grids or multicontact strips. For medial temporal recordings, two or three strips are placed from lateral to medial in the subdural space over the basal temporal cortex to the tentorial edge, avoiding bridging veins. Occasionally, depth electrodes are also placed acutely to record from medial structures. Recordings are usually referential to a linked neck reference.
Identification of Functional Cortex with Electrical Stimulation Mapping Electrical stimulation mapping is most often used intraoperatively to identify cortex related to motor and language function. The different effects of stimulation are used to identify these different areas. Stimulation of the motor cortex, as well as primary somatosensory, auditory and visual cortices, evokes responses. Presumably the
2653
2654
157
Intraoperative monitoring in epilepsy
primary effect of stimulation is the activation of neurons. Applying similar currents, below the threshold for evoking a seizure, to the remaining association cortex rarely evokes any responses. However, if the patient is engaged in an ongoing task, stimulation of localized areas of the cortex will interfere with the task, the location of these areas varying with the nature of the task. Presumably this effect represents depolarization blockade of neural activity by the current [42]; an alternative possibility is selective activation of inhibitory networks. This is the effect used in language mapping, using tasks such as auditory or visual naming or reading.
Identification of Rolandic Cortex The electrical excitability of motor cortex has been known since the experimental animal work of Fritsch and Hitzig [43]. The first stimulation of human motor cortex is attributed to Barthalow [44]. Stimulation to identify Rolandic cortex intraoperatively has been part of neurosurgery from its earliest days [45], with extensive experiences reported by Foerster (1936) [5,46] and Penfield and Jasper [5]. Based on the location of evoked movements, the classic homocular pattern of Rolandic motor and sensory responses was identified in human cortex, best known from the diagram of Penfield, with contralateral face representation lateral-inferior, upper extremity mid-superior, and lower extremity on the medial face of the hemisphere, the sensory representation displaced slightly superiorly from motor. Despite the extensive experience with mapping of Rolandic cortex, refinements in this homunculus are still being identified [47,48]. Motor cortex can be identified with stimulation in a patient awake under local anesthesia, or under general anesthesia. The threshold for evoking observable movement is usually much lower in an awake patient. In that setting, the author commonly begins stimulation in the
expected location of precentral cortex with brief trains of 60 Hz, biphasic square wave pulses, 1 ms per phase, from a constant current stimulator, beginning at 1 mA and increasing the current at 1 mA increments until movement is observed, commonly at 2–3 mA (measured between pulse peaks). Sensory effects are similarly evoked from the post central cortex. Under general anesthesia much larger currents may be required, up to 15 mA and of course, only motor effects can be obtained. Using electromyographic (EMG) changes rather than observed movement as evidence of a motor evoked response has several advantages when mapping under general anesthesia, as EMG responses are evoked at smaller currents than movements, and changes involving the face are more easily identified [49]. An alternative method for intraoperative identification of Rolandic cortex under general anesthesia is the use of somatosensory evoked potentials, most often to medial nerve stimulation, identifying upper extremity sensory representation. The central sulcus is identified by phase reversal of the earliest components of those potentials, though occasionally there has been a difference in one sulcus in the identification of the central sulcus with this technique, compared to the sulcus separating motor and sensory effects with stimulation [50]. Intraoperative mapping of Rolandic cortex has only limited application in epilepsy surgery now, as most patients with presumed Rolandic foci will be evaluated with chronic intracranial electrodes. However, it is extensively used in resections of intraparenchymal brain tumors near the Rolandic cortex. In that situation, stimulation is used not only for cortical localization but also to identify subcortical motor pathways [51].
Identification of Language Cortex Spontaneous speech is rarely evoked by cortical stimulation. However, if the patient is engaged in
Intraoperative monitoring in epilepsy
an ongoing language task such as naming objects, stimulation of some cortical sites, outside of primary motor and sensory areas, will interfere with the task. This effect was first reported by Penfield [5,52]. In those studies, most of the sites where this effect was evoked were in the perisylvian cortex of the language dominant hemisphere, in a distribution that was similar to, though more widespread than, the classic Broca and Wernicke language areas derived from the location of brain damage that resulted in aphasia. Penfield concluded that the sites of this interference effect were the cortical representation of language. Subsequent studies demonstrated that when a resection encroached on these perisylvian sites, a postoperative language deficit was much more likely than when there was a 1–2 cm margin between resection and the site of interference [53,54], providing evidence that these sites where cortical surface stimulation interfered with a language task predicted the language effects of a resection, including both the surface cortex and that buried in the sulci. These surface sites are often quite focal in an individual patient, often one in the posterior inferior frontal gyrus and one or more separate sites in mid-posterior superior temporal gyrus, but with considerable variation in their exact location between patients [55,56]. It is the combination of the focal nature of essential perisylvian language cortex in an individual patient with the variability in the location of these areas across patients, so that anatomic landmarks do not reliably indicate what cannot be resected, that makes stimulation mapping of language a useful tool in planning cortical resections. Penfield also found that stimulation in the supplementary motor area interfered with the language task. However, in contrast to perisylvian language sites, resection of supplementary motor cortex sites results in only a transient language disturbance, the patient usually mute for a few days, then quickly recovering. Stimulation of anterior medial basal temporal cortex has
157
also been shown to interfere with language [57] but permanent language deficits rarely follow resection of those sites. Stimulation mapping of language requires an awake, performing patient. For most adolescents and adults, there are two settings where it can be performed: extraoperatively through chronically implanted subdural electrode grids, or intraoperatively when the procedure is done with a technique where the patient is awake under local anesthesia for part of the operation. The choice of approaches has been discussed in detail elsewhere [58]. Briefly, extraoperative mapping requires two craniotomies, with increased surgical risk and investment of resources, but provides more time for mapping and may be more comfortable for the patient and surgeon. It is the author’s view that the choice of approaches depends on the need for intracranial ictal recording; if that is needed to identify the epileptogenic zone, then the mapping should be extraoperative through those electrodes. However, if all that is needed is language mapping, as with mapping to guide resection of a lesion, or of an epileptic focus identified with preoperative interictal criteria, it can usually be accomplished with less risk and greater accuracy by intraoperative mapping. Indeed, if a resection is to be done close to essential language sites based on extraoperative mapping, the author will usually do that resection with the patient awake, using intraoperative language testing to identify the essential language areas with a finer resolution than can be accomplished with extraoperative mapping, and to assess language during the resection. Although language can be localized in young children with stimulation mapping [59], it requires the extraoperative approach. There are a number of other areas of controversy with regard to language mapping. There is currently little or no controversy about using language mapping in posterior temporal resections in the dominant hemisphere. Indeed with that technique it is often possible to resect an
2655
2656
157
Intraoperative monitoring in epilepsy
epileptogenic zone that, based solely on anatomic criteria, would be considered to be in Wernicke’s area and thus unresectable. It is in anterior temporal resections where there is controversy concerning the usefulness of language mapping. Classically, dominant hemisphere anterior temporal cortex in middle and inferior temporal gyri is not considered essential for language. Exactly how far posterior one can safely go, in this classical view, varies: the line of the central fissure, or the vein of Labbe, or 4–4.5 cm from the temporal tip, has been proposed as the posterior limit of the ‘‘safe’’ zone. However, stimulation mapping has identified sites where the current interferes with language within all of these ‘‘safe’’ zones. Such sites were present in the early Penfield and Roberts [52] experience (though not commented on) and confirmed in later series [55,60]. The Ojemann et al. [55] series of language mapping in 117 dominant hemisphere patients provided estimates of finding essential language sites in various zones of the temporal cortex. In the superior temporal gyrus, 15% of these patients had such sites anterior to the line of the central sulcus. In some patients these sites have been within 3 cm of the temporal tip. In the middle temporal gyrus, 5% of patients had essential language sites. These would be the patients at risk from an anterior temporal resection without language mapping. Indeed in a comparative study of mapped and not mapped anterior temporal resections that spared the superior temporal gyrus, this is about the proportion of patients with major postoperative language deficits, although with this small a proportion, with the numbers of patients in that study, there were no differences in postoperative language performance between the total populations of patients managed either way [61,62]. Whether to use language mapping with anterior temporal resections, then, depends on what risk of a postoperative language disturbance is considered acceptable.
Another area of controversy is whether one can safely proceed with a resection in the absence of having identified any sites where stimulation interfered with language. It has been the author’s view that identifying such ‘‘positive’’ sites was necessary in order to be certain that an adequate current level had been used to map the ‘‘negative’’ sites in the cortex that was to be resected [63]. This requirement has the disadvantage that stimulation effects must often be determined for a wider area of cortex than that required for the resection. A recent study of language mapping in tumor resections found that resections based only on ‘‘negative’’ findings could be safely done, with the current set only by the threshold for after discharge [64]. The choice of language tasks to use with mapping has also been controversial. Penfield used naming of visually presented objects [52]. However, a recent study [65] suggests that sites with interference in auditory naming provided a better indicator of what portion of dominant anterior temporal lobe could be resected without a postoperative language deficit than did sites with interference in naming of visually presented objects. In multilingual patients, mapping has often shown separation of sites where stimulation interferes with naming of the same objects in one language or the other, particularly in the temporal lobe [66,67]. Reading of phrases or single words has also been used with mapping [57]. There is substantial separation between sites where stimulation interferes with object naming or reading [57,68]. In patients dependent on multiple languages or reading, it is desirable to map these different modalities. There is little evidence that the nature of the errors made with stimulation during naming or reading has predictive value in planning resections, although different types of errors localize to different parts of language cortex [69–71]. All tasks used with intraoperative mapping need to be sufficiently simple so that baseline error rates in the absence
Intraoperative monitoring in epilepsy
of stimulation do not exceed 25% or so, since each site is sampled only a few (often three) times so that with any higher baseline error rate significant stimulation effects cannot be distinguished from random errors. This limits the usefulness of intraoperative stimulation mapping in patients with preexisting aphasias. Stimulation-evoked interference has also been used to localize other functions in the human cortex. Lateral temporal sites where stimulation interferes with verbal memory encoding or storage have been identified [70,72,73] and found to be separate from those for naming of the same material. Resection of these sites was associated with an increased chance of a postoperative verbal memory deficit [74]. Cortical stimulation effects on verbal learning [75] and visuospatial functions [76] have also been investigated. During tumor resections, stimulation interference effects have been used to map subcortical language pathways [77,78]. Intraoperative stimulation mapping for language requires an awake craniotomy technique [63]. The author uses propofol intravenous anesthesia for placement of a scalp field block of 0.5% lidocaine and 0.25% bupivacaine local anesthesia, the craniotomy, and blocking of dural sensation with an intradural injection of the local anesthetic around the middle meningeal artery. The craniotomy is planned to expose areas where language is likely to be located as well as that needed for the planned resection. Once the craniotomy is completed the patient is awakened from the propofol anesthesia. Electrocorticography and motor mapping precede language mapping. Mapping use 3–4 s trains of 60 Hz biphasic square wave pulses, 1 ms per phase, from a constant current stimulator, delivered in a bipolar manner across electrodes separated by 5 mm. The current level for language mapping is set as the largest current that does not evoke after discharges in the exposed area of cortex with an arbitrary upper limit of 10 mA (measured between pulse peaks). Commonly, current is
157
above the threshold for evoked effects from the Rolandic cortex. Cortical sites are identified by numbers, the patient begins the language task commonly at the rate of one item every 3–4 s. Stimulation is applied on every third or fourth item, continuing the current until there is a response or the next item. All sites are sampled before repeating this at least three times. Stimulation effects at each site are compared to performance on the interspersed control items. In planning the resection, sites with significant evoked interference are spared with a 2–3 cm margin along a continuous gyrus. If the resection is to be closer to such sites, it is advisable to test language during the resection and stop as soon as errors develop.
References 1. Haglund MM, Ojemann GA, Hochman DW. Optical imaging of epileptiform and functional activity in human cerebral cortex. Nature 1992;358:668-71. 2. Toga AW, Ojemann GA, Ojemann J, Cannestra A. Intraoperative brain mapping. In: Mazziota J, Toga AW, Frackowiak RS, editors. Brain mapping: the disorders. San Diego: Academic Press; 2000. p. 77-106. 3. Gorbach AM, Heiss J, Kufta C, Sato S, Fedio P, Kammerer WA, Solomon J, Oldfield EH. Intraoperative infrared functional imaging of human brain. Ann Neurol 2003;54:297-09. 4. Hall WA, Liu H, Maxwell RE, Truwit CL. Influence of 1.5-tesla intraoperative MR imaging on surgical decision making. Acta Neurochir Suppl 2003;85:29-37. 5. Penfield W, Jasper H. Epilepsy and the functional anatomy of the human brain. Boston: Little, Brown; 1954. 6. Holmes MD, Kutsy RL, Ojemann GA, Wilensky AJ, Ojemann LM. Interictal, unifocal spikes in refractory extratemporal epilepsy predict ictal origin and postsurgical outcome. Clin Neurophysiol 2000;111:1802-8. 7. Rasmussen T. Surgical treatment of patients with complex partial seizures. In: Penry JK, Daly DD, editors. Complex partial seizures and their treatment. Raven Press; New York: p. 415-450. 8. Bengzon A, Rasmussen T, Gloor P. Prognostic factors in the surgical treatment of temporal lobe epileptics. Neurology 1968;18:717-31. 9. Fiol ME, Gates JR, Torres F, Maxwell RE. The prognostic value of residual spikes in the postexcision electrocorticogram after temporal lobectomy. Neurology 1991;41: 512-6.
2657
2658
157
Intraoperative monitoring in epilepsy
10. Jasper H, Pertussiet B, Flanigan H. EEG and cortical electrograms in patients with temporal lobe seizures. Arch Neurol Psychiatr 1951;65:272-90. 11. So N, Olivier A, Andermann F, Gloor P, Quesney LF. Results of surgical treatment in patients with bitemporal epileptiform abnormalities. Ann Neurol 1989;25:432-9. 12. Stefan H, Quesney LF, Abou-Khalil B, Olivier A. Electrocorticography in temporal lobe epilepsy surgery. Acta Neurol Scand 1991;83:65-72. 13. Binnie CD, Alarcon G, Elwes RD, Garcia Seoane JJ, Juler J, Polkey CE. Role of ECoG in ‘en bloc’ temporal lobe resection: the Maudsley experience. Electroencephalogr Clin Neurophysiol Suppl 1998;48:17-23. 14. Chatrian GE, Tsai ML, Temkin NR, Holmes MD, Pauri F, Ojemann GA. Role of the ECoG in tailored temporal lobe resection: the University of Washington experience. Electroencephalogr Clin Neurophysiol Suppl 1998;48:24-43. 15. Devinsky O, Canevini M, Sato S. Quantitative electrocorticography in patients undergoing temporal lobectomy. J Epilepsy 1992;5:178-85. 16. Engel J Jr, Driver MV, Falconer MA. Electrophysiological correlates of pathology and surgical results in temporal lobe epilepsy. Brain 1975;98:129-56. 17. Fenyes I, Zoltan I, Fenyes G. Temporal lobe epilepsies with deep seated epileptogenic foci. Arch Neurol 1961;4: 103-15. 18. Graf M, Niedermeyer E, Schiemann J. Electrocorticography: information derived from intraoperative recordings during seizure surgery. Clin Electroencephalogr 1984;15: 83-91. 19. Kanazawa O, Blume WT, Girvin JP. Significance of spikes at temporal lobe electrocorticography. Epilepsia 1996;37: 50-5. 20. Rasmussen TB. Surgical treatment of complex partial seizures: results, lessons, and problems. Epilepsia 1983; 24 Suppl 1:S65-76. 21. Schwartz TH, Bazil CW, Walczak TS, Chan S, Pedley TA, Goodman RR. The predictive value of intraoperative electrocorticography in resections for limbic epilepsy associated with mesial temporal sclerosis. Neurosurgery 1997;40:302-09; discussion 309–11. 22. Tran TA, Spencer SS, Marks D, Javidan M, Pacia S, Spencer DD. Significance of spikes recorded on electrocorticography in nonlesional medial temporal lobe epilepsy. Ann Neurol 1995;38:763-70. 23. Tuunainen A, Nousiainen U, Mervaala E, Pilke A, Vapalahti M, Leinonen E, Paljarvi L, Riekkinen P. Postoperative EEG and electrocorticography: relation to clinical outcome in patients with temporal lobe surgery. Epilepsia 1994;35:1165-73. 24. Wyllie E, Luders H, Morris HH, III, Lesser RP, Dinner DS, Hahn J, Estes ML, Rothner AD, Erenberg G, Cruse R, et al. Clinical outcome after complete or partial cortical resection for intractable epilepsy. Neurology 1987;37: 1634-41.
25. Wieser HG, Yasargil MG. Selective amygdalohippocampectomy as a surgical treatment of mesiobasal limbic epilepsy. Surg Neurol 1982;17:445-57. 26. Engel J Jr. Outcome with respect to epileptic seizures. In: Engel J Jr., editor. Surgical treatment of the epilepsies. New York: Raven Press; 1987. p. 553-571. 27. Alarcon G, Garcia Seoane JJ, Binnie CD, Martin Miguel MC, Juler J, Polkey CE, Elwes RD, Ortiz Blasco JM. Origin and propagation of interictal discharges in the acute electrocorticogram. Implications for pathophysiology and surgical treatment of temporal lobe epilepsy. Brain 1997;120(Pt 12):2259-82. 28. Alarcon G, Guy CN, Binnie CD, Walker SR, Elwes RD, Polkey CE. Intracerebral propagation of interictal activity in partial epilepsy: implications for source localisation. J Neurol Neurosurg Psychiatry 1994;57:435-49. 29. Ortega GJ, Menendez de la Prida L, Sola RG, Pastor J. Synchronization clusters of interictal activity in the lateral temporal cortex of epileptic patients: intraoperative electrocorticographic analysis. Epilepsia 2008;49:269-80. 30. Ferrier CH, Aronica E, Leijten FS, Spliet WG, van Huffelen AC, van Rijen PC, Binnie CD. Electrocorticographic discharge patterns in glioneuronal tumors and focal cortical dysplasia. Epilepsia 2006;47:1477-86. 31. Jooma R, Yeh HS, Privitera MD, Rigrish D, Gartner M. Seizure control and extent of mesial temporal resection. Acta Neurochir (Wien) 1995;133:44-9. 32. McKhann GM, II, Schoenfeld-McNeill J, Born DE, Haglund MM, Ojemann GA. Intraoperative hippocampal electrocorticography to predict the extent of hippocampal resection in temporal lobe epilepsy surgery. J Neurosurg 2000;93:44-52. 33. Miles AN, Dodrill C, Ojemann GA. Postoperative decline in measures of verbal memory after dominant temporal lobe resections is proportional to the extent of medical resection. Philadelphia PA, Presented at AES Annual Meeting American Epilepsy Society, 2001. 34. Dodrill CB, Ojemann GA. An exploratory comparison of three methods of memory assessment with the intracarotid amobarbital procedure. Brain Cogn 1997;33:210-23. 35. Dworacek B, De Vlieger M. Absence of electroencephalographic excitation pattern under isoflurane anesthesia. Acta Anaesthesiol Belg 1984;35:211-7. 36. Fiol ME, Boening JA, Cruz-Rodriguez R, Maxwell R. Effect of isoflurane (forane) on intraoperative electrocorticogram. Epilepsia 1993;34:897-900. 37. Ito BM, Sato S, Kufta CV, Tran D. Effect of isoflurane and enflurane on the electrocorticogram of epileptic patients. Neurology 1988;38:924-8. 38. Artru AA, Lettich E, Colley PS, Ojemann GA. Nitrous oxide: suppression of focal epileptiform activity during inhalation, and spreading of seizure activity following withdrawal. J Neurosurgical Anesthesiology 1990;2: 189-93. 39. Babb TL, Ferrer-Brechner T, Brechner VL, Crandall PH. Limbic neuronal firing rates in man during administration
Intraoperative monitoring in epilepsy
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52. 53.
54.
of nitrous oxide – oxygen or sodium thiopental. Anesthesiology 1975;43:402-9. Ferrer-Allado T, Brechner VL, Dymond A, Cozen H, Crandall P. Ketamine-induced electroconvulsive phenomena in the human limbic and thalamic regions. Anesthesiology 1973;38:333-44. Asano E, Benedek K, Shah A, Juhasz C, Shah J, Chugani DC, Muzik O, Sood S, Chugani HT. Is intraoperative electrocorticography reliable in children with intractable neocortical epilepsy? Epilepsia 2004;45:1091-9. Ranck JB Jr. Which elements are excited in electrical stimulation of mammalian central nervous system: a review. Brain Res 1975;98:417-40. Fritsch GT, Hitzig E. Uber die elektrishe Erregbarkeit des Grosshirns. Arch Anat Physiol Wiss Med 1870;300‐14. Bartholow R. Experimental investigations into the functions of the human brain. Am J Med Sci 1874;139: 305-13. Cushing H. A note upon the faradic stimulation of the postcentral gyrus in conscious patients. Brain 1909; 32:44-53. Foerster O. Motorische Felder and Bahnen, sensible cortical Felder. In: Bumke O, Foerster O, editors. Handbuch der Neurologie. Berlin: Julius Springer; 1936. p. 1-448. Bradley WE, Farrell DF, Ojemann GA. Human cerebrocortical potentials evoked by stimulation of the dorsal nerve of the penis. Somatosens Mot Res 1998;15:118-27. Farrell DF, Burbank N, Lettich E, Ojemann GA. Individual variation in human motor-sensory (rolandic) cortex. J Clin Neurophysiol 2007;24:286-93. Yingling CD, Ojemann S, Dodson B, Harrington MJ, Berger MS. Identification of motor pathways during tumor surgery facilitated by multichannel electromyographic recording. J Neurosurg 1999;91:922-7. Wood CC, Spencer DD, Allison T, McCarthy G, Williamson PD, Goff WR. Localization of human sensorimotor cortex during surgery by cortical surface recording of somatosensory evoked potentials. J Neurosurg 1988;68: 99-111. Keles GE, Lundin DA, Lamborn KR, Chang EF, Ojemann G, Berger MS. Intraoperative subcortical stimulation mapping for hemispherical perirolandic gliomas located within or adjacent to the descending motor pathways: evaluation of morbidity and assessment of functional outcome in 294 patients. J Neurosurg 2004;100: 369-75. Penfield W, Roberts L. Speech and brain mechanisms. Princeton, NJ: Princeton University Press; 1959. Haglund MM, Berger MS, Shamseldin M, Lettich E, Ojemann GA. Cortical localization of temporal lobe language sites in patients with gliomas. Neurosurgery 1994;34:567-76; discussion 576. Ojemann GA. Eletrical stimulation and the neurobiology of language. Behav Brain Sci 1983;2:221-6.
157
55. Ojemann GA, Ojemann JG, Lettich E, Berger M. Cortical language localization in left, dominant hemisphere. An electrical stimulation mapping investigation in 117 patients. J Neurosurg 1989;71:316-26. 56. Ojemann GA, Whitaker HA. Language localization and variability. Brain Lang 1978;6:239-60. 57. Luders H, Lesser RP, Hahn J, Dinner DS, Morris H, Resor S, Harrison M. Basal temporal language area demonstrated by electrical stimulation. Neurology 1986;36:505-10. 58. Ojemann GA, Sutherling W, Lesser RP, Dinner DS, Jayakar P, Saint-Hilaire J-M. Cortical stimulation. In: Engel J Jr., editor. Surgical treatment of the epilepsies. 2nd ed. Raven Press; New York: p. 339-414. 59. Ojemann SG, Berger MS, Lettich E, Ojemann GA. Localization of language function in children: results of electrical stimulation mapping. J Neurosurg 2003;98:465-70. 60. Devinsky O, Perrine K, Llinas R, Luciano DJ, Dogali M. Anterior temporal language areas in patients with early onset of temporal lobe epilepsy. Ann Neurol 1993;34: 727-32. 61. Barbaro NM, Walker JA, Laxer KD. Temporal lobectomy and language function. J Neurosurg 1991;75:830-1. 62. Hermann BP, Wyler AR, Somes G. Language function following anterior temporal lobectomy. J Neurosurg 1991;74:560-6. 63. Ojemann GA. Awake operations with mapping in epilepsy. In: Schmidek H, Sweet W, editors. Operative neurosurgical techniques. Philadelphia: Saunders; 1995. p. 1317-22. 64. Sanai N, Mirzadeh Z, Berger MS. Functional outcome after language mapping for glioma resection. N Engl J Med 2008;358:18-27. 65. Hamberger MJ, Seidel WT, McKhann GM, II, Perrine K, Goodman RR. Brain stimulation reveals critical auditory naming cortex. Brain 2005;128:2742-49. 66. Lucas TH, McKhann GM, Ojemann GA. Functional separation of languages in the bilingual brain: a comparison of electrical stimulation language mapping in 25 bilingual patients and 117 monolingual control volunteers. J Neurosurg 2004;101:449-57. 67. Ojemann GA, Whitaker HA. The bilingual brain. Arch Neurol 1978;35:409-12. 68. Ojemann GA. Some brain mechanisms for reading. In: Von Euler C, Lundberg I, Lennerstrand G, editors. Brain and reading. Hampshire, London: MacMillan Press; 1989. p. 47-59. 69. Corina D, Gibson E, Martin RF, Poliakov A, Brinkley JF, Ojemann GA. Dissociation of action and object naming: evidence from cortical stimulation mapping. Human Brain Mapping 2005;24:1-10. 70. Ojemann GA. Organization of short-term verbal memory in language areas of human cortex: evidence from electrical stimulation. Brain Lang 1978;5:331-40.
2659
2660
157
Intraoperative monitoring in epilepsy
71. Ojemann GA. Brain organization for language from the perspective of eletrical stimulation mapping. Behav Brain Sci 1983;6:189-206. 72. Ojemann GA, Dodrill CB. Verbal memory deficits after left temporal lobectomy for epilepsy. Mechanism and intraoperative prediction. J Neurosurg 1985;62:101-7. 73. Perrine K, Devinsky O, Uysal S, Luciano DJ, Dogali M. Left temporal neocortex mediation of verbal memory: evidence from functional mapping with cortical stimulation. Neurology 1994;44:1845-50. 74. Ojemann GA, Dodrill CB. Intraoperative techniques for reducing language and memory deficits with left temporal lobectomy. Adv Epileptol 1987;16:327-30. 75. Ojemann JG, Ojemann GA, Lettich E. Cortical stimulation mapping of language cortex by using a verb generation
task: effects of learning and comparison to mapping based on object naming. J Neurosurg 2002;97:33-38. 76. Fried I, Mateer C, Ojemann G, Wohns R, Fedio P. Organization of visuospatial functions in human cortex. Evidence from electrical stimulation. Brain 1982;105: 349-71. 77. Duffau H, Capelle L, Sichez N, Denvil D, Lopes M, Sichez JP, Bitar A, Fohanno D. Intraoperative mapping of the subcortical language pathways using direct stimulations. An anatomo-functional study. Brain 2002;125: 199-214. 78. Mandonnet E, Nouet A, Gatignol P, Capelle L, Duffau H. Does the left inferior longitudinal fasciculus play a role in language? A brain stimulation study. Brain 2007; 130:623-9.
159 Medical Management of Epilepsy M. E. Newmark
Introduction Epilepsy is perhaps the most prevalent of the neurological diseases common in all age groups that require on-going treatment, and can have significant and even devastating consequences for the patient who must face this problem [1]. Loss of employment, automobile driving, or basic independence may develop and become major issues, especially if the seizures are incompletely controlled. The socio-economic consequences of having seizures are profound, and the anti-epileptic drugs themselves, the cornerstone of epilepsy treatment, often have substantial medical and neurological untoward effects. Relevant to the practice of the neurosurgeon, many patients with primary brain tumors present with or have epilepsy, and epilepsy is commonly associated with conditions that a neurosurgeon may face, including stroke, hemorrhage, and brain trauma [2,3]. An understanding of the use of anti-epileptic drugs (AED’s) is important for the efficient care of the surgical patient.
The Question of Whom to Treat Not all epileptic seizures require immediate or chronic treatment with anti-epileptic agents. There are numerous provoking stimuli or conditions which do not lead to further seizures or an enduring epileptic state, if the provoking stimulus or condition is removed or treated. Seizures presenting immediately after an acute head trauma do not necessarily portend the development of epilepsy, and seizures may be similarly #
Springer-Verlag Berlin/Heidelberg 2009
provoked by metabolic/anoxic changes, by some prescribed medications, or by some recreational agents [4,5]. When these conditions or drugs are corrected and removed, the seizures frequently remit without specific anti-epileptic medication. > Table 159-1 summarizes some of the more common provoking causes of seizures, which if corrected, will not necessarily predispose the patient to develop epilepsy. A more controversial question is whether one should treat a single unprovoked seizure. Basically, the decision to treat is determined by weighing the risk of a second event against the risk of medication toxicity [5–8]. If there are data that suggest an on-going neurological condition which is associated with an enduring epileptic state, then epilepsy can be diagnosed, and treatment is recommended. For example, epileptiform discharges on the EEG, a structural abnormality associated with epilepsy identified on brain imaging, or an abnormal neurological exam increase the likelihood of a second event, and treatment should be initiated in most instances. If all examinations are normal, one may still want to treat if there are significant socio-economic or other reasons to avoid a seizure, as the rate of relapse may approach or even surpass 50% if patients are followed for a sufficiently long period [5,9,10]. On the other hand, if the patient is a child or a woman planning to become pregnant, a more conservative approach may be warranted as either driving is a non-issue or possible teratogenic effects of AED’s become more important. Another controversial issue is whether it is possible to prevent epilepsy in patients known to have undergone trauma, cerebral hemorrhage, or
2670
159
Medical management of epilepsy
. Table 159-1 Some provoked causes of seizures Acute injuries or metabolic changes Acute head trauma Hypoglycemia Hyperglycemia Uremia Hepatic failure Hypercalcemia Hyponatremia Prescribed medications Tramadol Bupropion Ciprofloxacin Intoxications Alcohol-related Psychostimulants
brain surgery. Physicians vary in their practices, but the data are fairly clear that AED prophylaxis after the first week, at least with phenytoin, does not prevent the development of further unprovoked seizures for the patient who has sustained head trauma, and is usually not recommended [11]. Nevertheless, whereas chronic treatment is usually avoided, AED treatment is occasionally given in the first week to reduce potential complications from a seizure in an acute, significantly ill patient. Similar to the patients with head trauma, patients with subarachnoid or intracerebral hemorrhage are at risk for having seizures, at least in the acute phase [12–14]. Again, there are no data supporting the use of prophylaxis after the acute event for the patient who has not suffered a seizure as chronic AED treatment does not seem to reduce the epilepsy risk. As with the trauma patients, treatment may also be considered during the acutely ill phase, but even here there are scanty data recommending this practice. Unfortunately, sometimes it is difficult to identify whether a patient has had, or is having seizures, at least in someone who has suffered an intracerebral hemorrhage. As reported recently, subclinical electrographic seizures may occur
frequently after intracerebral hemorrhage, and may be identified only after continuous electroencephalographic monitoring [14]. If seizures have not been identified, then the chronic use of antiepileptic agents is not recommended. In fact, they may even increase the risk of an unfavorable outcome among patients after a subarachnoid hemorrhage, as have been noted recently with phenytoin, which may increase morbidity in a dose-dependent manner [15]. The picture is also blurred for patients undergoing surgery for a supratentorial brain tumor who have not yet had a seizure. Although patients often present with a clinical seizure as the first symptom of a brain tumor, AED prophylaxis to prevent epilepsy in a seizure-free tumor patient has not been proven effective. Untoward reactions to AED’s among patients with brain tumors are not rare, and skin reactions may be particularly common [16]. If the tumor treatment ends successfully with surgery, most patients will be weaned successfully from AED’s after the first week or so. However, if the treatment includes post-surgical brain radiation, the risk of epilepsy climbs substantially, and AED prophylaxis is suggested for this sub-group of patients [3,17].
Medical management of epilepsy
An equally challenging is when to discontinue medication. Most studies involve children, an age group in which the socio-economic factors of driving a motor vehicle and maintaining employment are not so important. Arbitrarily, except for the children with juvenile myoclonic epilepsy who generally relapse upon withdrawal of AED’s, a trial of discontinuation of medicine is often attempted after two years of seizure freedom. Relapse rates are highest in children with relatively late onset epilepsy, epileptiform features on the electroencephalogram, or a known etiology for the seizure disorder [18]. Discontinuation of AED’s in the adult age group is rarely accomplished because of the fear of a relapse which would significantly affect employment or life-style, with the exception of some women contemplating pregnancy. Pregnant women will often stop medication because of worries over AED’s teratogenic effects, but almost all women given pre-pregnancy counseling will choose to remain on medicine. Even after apparent successful surgery for intractable epilepsy, most patients will opt to remain on medication as relapse rates are high without medication [19].
159
Status Epilepticus Status epilepticus, defined as a continuing epileptic state when one seizure follows another without return of consciousness or when a single seizure lasts greater than 10 min, is a medical emergency with morbidity and mortality high in the adult patient. There are many causes, but status epilepticus can present in patients with cerebral hemorrhage, stroke, head trauma, or supratentorial tumor. The best treatment results occur when a standardized protocol is used consisting of measures for general support of the patient, pharmacological treatment of the seizures, and treatment of the underlying cause, if possible. > Table 159-2 is an example of one such approach combining features of several available protocols [20,21]. An important factor for all AED treatment protocols for status epilepticus is the administration of adequate amounts of medication, as outlined in > Table 159-2. Lesser doses are often unsuccessful, and multiple small doses of a combination of agents increase toxicity to the patient but do not treat the seizures adequately.
. Table 159-2 A protocol for the treatment of status epilepticus Management upon presentation in status epilepticus Monitor and maintain ventilation Monitor and maintain blood pressure EKG Establish venous line Obtain blood samples for CBC, CPK, hepatic and renal function, serum electrolytes, serum calcium and magnesium, blood glucose, serum AED levels, and toxicological blood and urine screen, and begin treatment of abnormalities Administer 50 ml i.v. 50% dextrose preceded by 100 mg i.v. thiamine in patients likely to have chronic alcohol abuse Assess and treat metabolic acidosis Pharmacological treatment of status epilepticus Lorazepam 0.05–0.1 mg/kg i.v. (maximum infusion rate of 2 mg/min) to a maximum of 0.2 mg/kg or diazepam 0.1 mg/kg i.v (in 120 s), to a maximum of 0.2 mg/kg Fosphenytoin 20 mg/kg phenytoin equivalents at a maximum infusion rate of 150 mg/min with BP and EKG monitoring If seizures continue, sodium valproate 40–60 mg/kg in 10 min or more, or levetiracetam 20 mg/kg in 20 min If seizures continue, switch to another choice in #3, or establish airway and administer propofol 1–2 mg/kg i.v. bolus, repeated q 5 min to a maximum until seizures stop or 10 mg/kg is reached, followed by continuous infusion (up to 5 mg/kg/h)
2671
2672
159
Medical management of epilepsy
Lorazepam or other benzodiazepines such as diazepam, require concurrent treatment with another anti-epileptic drug, as they are ineffective after the first several minutes because of medication re-distribution within tissues, and cannot be used for on-going, chronic management. On the other hand, valproate, levetiracetam, and phenytoin can be used by themselves as they will prevent subsequent seizures once the status epilepticus is brought under control. Once a decision has been made to begin an AED in the neurosurgical patient, there are almost a bewildering number of choices now available [22]. Agents are best selected by considering their efficacy against the seizure type, their side-effect profile, and their pharmacodynamics. Despite the apparent complexity, it is fairly easy to formulate an educated decision based on a few principles, some of which are summarized in > Table 159-3. There have been many attempts to compare AED’s for efficacy against seizure types, but basically no drug has been shown to be convincingly superior to any other if the agent is used to treat appropriate seizure types [22–25]. The choice of an efficacious drug is actually pretty straightforward, as partial and secondarily generalized seizures, the most common of the seizure types faced by the surgeon, respond to almost all the available AED’s except for ethosuximide, whose
efficacy is fairly restricted to absence attacks. Perhaps still the most convincing of the studies confirming this parity among agents are the two VA cooperative reports summarized by Mattson et al., which showed that all the older antiepileptic drugs were about equally effective for the treatment of partial or secondarily generalized seizures [23,24]. Since the newer agents are tested for parity against the older ones, one would expect that all agents, new or old, are basically equally effective [25,26]. The surgeon may occasionally be faced with the treatment of a patient with primary generalized seizures, and in these circumstances, the situation is a little more complex. Some of the agents, including carbamazepine, gabapentin, lamotrigine, and possibly pregabalin and oxcarbazepine, can exacerbate some primary generalized seizures, particularly myoclonic and absence seizures, and should be avoided [22,27,28]. In these instances, levetiracetam, valproic acid, and possibly zonisamide may be better choices. Since most of the drugs are equally efficacious in most instances for the seizure types that a surgeon typically faces, other factors become important, with side-effects the most prominent one. It is impossible for one person to know all the toxicity of all the agents, but some of them have characteristic features that may sway the physician’s choice. Some drugs are cleared only
. Table 159-3 AED selection factors The efficacy for the proposed agent against the specific seizure type is the determining factor Once efficacy is established, side-effects are the second major factor in the choice of an AED The availability of fast and reliable therapeutic blood levels may be important (currently available only for the older, established drugs in most institutions) Older drugs are often equally effective compared with newer ones and may be less expensive Drug-drug interaction may be important The availability of intravenous preparation may be significant for some patients Some agents can be quickly loaded to an efficacious dose and blood level, whereas others may require days or weeks to reach this goal Financial cost of treatment may be a determining factor
Medical management of epilepsy
by the liver or may have prominent hepatic toxicity [29]. Valproic acid and felbamate have been found to cause hepatic toxicity and should be avoided in patients with possible hepatic dysfunction. Those agents, such as phenytoin, metabolized primarily through the liver may not be the wisest choice for a patient with marginal hepatic function. Other agents, such as carbamazepine, may have hematological toxicity, and one may want to avoid them in a patient with a low white count if other agents are available. Levetiracetam, gabapentin, pregabalin, zonisamide, and lamotrigine are cleared essentially totally through the kidney; either adjustment must be made with dose or another drug selected in patients with prominent renal disease. Other drugs may be associated with hyponatremia, some affect body weight, and a number of agents cause drowsiness or sedation as a side-effect [30]. Frequently encountered side-effects or metabolic considerations that are faced in the use of AED’s are summarized in > Table 159-4. For the acutely ill patient or for the epilepsy patient with toxicity or incomplete control of seizures, the availability of accurate, meaningful blood levels is often important. Established and easily obtainable levels are not typically available for all drugs but can be obtained for phenytoin, valproic acid, phenobarbital, and carbamazepine. The other agents either have levels which
159
are not universally available on a rapid basis, or have levels that are not tightly correlated with toxicity and efficacy. The metabolism of the drug can occasionally be the significant factor for the selection of treatment. Only phenytoin, valproic acid, phenobarbital, and levetiracetam have intravenous preparations and can be given immediately and titrated to a therapeutic level on an emergency basis. Carbamazepine and oxcarbazepine cannot be loaded to achieve a therapeutic level acutely; they can be initiated at therapeutic doses. On the other hand, topiramate, lamotrigine, gabapentin, pregabalin, zonisamide, and primidone must be started at lower than therapeutic doses because of either intolerance to the usual therapeutic dose if started acutely or because of the risk of allergic reaction (lamotrigine.) For topiramate and lamotrigine especially, a therapeutic dose may not be reached for weeks after the initiation of the agent. The newer drugs often have fewer drug-drug interactions. Zonisamide, levetiracetam, lamotrigine, gabapentin, and pregabalin are not heavily protein-bound, are excreted through renal mechanisms, and/or are metabolized through hepatic non-cytochrome P450 mechanisms, which have little effect on other agents [31]. On the other hand, use of phenytoin and carbamazepine may bring many drug-drug interactions, especially
. Table 159-4 Metabolic or side-effect considerations for AED choices Hepatic clearance or hepatic toxicity: phenytoin, valproic acid, carbamazepine, oxcarbazepine, felbamate Renal clearance: levetiracetam, gabapentin, pregabalin, zonisamide, lamotrigine Hyponatremia: oxcarbazepine, carbamazepine Drowsiness: phenobarbital, primidone, gabapentin, pregabalin, levetiracetam Agitation, anxiety: zonisamide, felbamate, levetiracetam Renal stone: zonisamide, topiramate Weight gain: valproic acid, gabapentin Weight loss: zonisamide, topiramate Decreased neutrophils and/or platelets: carbamazepine, valproic acid Teratogenicity: valproic acid
2673
2674
159
Medical management of epilepsy
with such important agents as warfarin and the oral contraceptives. Valproic acid is heavily proteinbound, and therefore will compete with other agents that are similarly protein-bound. Even when an apparently appropriate AED is given at adequate doses, the patient may continue to have seizures. In these circumstances, often a second agent is added to the first one, avoiding replication of the first one’s potential toxicity or mechanisms of action. As examples, levetiracetam might be considered for a patient who has incompletely controlled attacks with phenytoin, rather than carbamazepine or oxcarbazepine, drugs which tend to share mechanisms of actions with phenytoin. If topiramate incompletely controls seizure activity, then zonisamide, a similar agent, would be a poor first alternative choice. Similarly, primidone and phenobarbital are both barbiturates, and gabapentin and pregabalin, which share many similar properties, would not be primary substitutions for each other. If seizure control is achieved, then discontinuation of the first agent is often considered after several months. Occasionally rescue medication may be needed in patients who have a history of serial seizures. Rectal diazepam gel has been shown to be an effective and well tolerated treatment for small children with repetitive seizures, but can be awkward to use in the adolescent or adult patient [32]. For the larger child and adult, oral concentrated diazepam can be safely delivered and is often an acceptable alternative [33].
Conclusion We now have so many choices for the pharmacological treatment of seizures that at times the decision can be overwhelming [34,35]. Perhaps the best course would be to select a few of the drugs and become familiar with their pharmacological characteristics and side-effect profile. Phenytoin or carbamazepine are often used for
the new surgical patient because of the ease of obtaining blood levels, the many years of experience with these agents, and the rapid onset of seizure protection. If these agents are unwise because of metabolic or other factors, levetiracetam can reach therapeutic levels quickly. There are many equally valid choices. It is important to realize that we now have the use of a variety of effective AED’s, and stand a good chance of prescribing an agent that will control seizures and can be well tolerated by the patient.
References 1. Hirtz D, Thurman DJ, Gwin-Hardy K, Mohamed M, Chaudhuri AR, Zalutsky R. How common are the ‘‘common’’ neurological disorders? Neurology 2007;68:326-37. 2. Michelucci R. Optimizing therapy of seizures in neurosurgery. Neurology 2006;67 Suppl 4:S14-S18. 3. Vecht CJ, van Breemen M. Optimizing therapy of seizures in patients with brain tumors. Neurology 2006;67 Suppl 4:S10-S13. 4. Brust JCM. Seizures and substance abuse. Neurology 2006;67:S45-S48. 5. Pohlmann-Eden B, Beghi E, Camfield C, Camfield P. The first seizure and its management in adults and children. BMJ 2006;332:339-42. 6. Hirtz D, Berg A, Bettis D, Camfield C, Camfield P, Crumrine P, Gaillard WD, Schneider S, Shinnar S. Practice parameter: treatment of the child with a first unprovoked seizure: report of the Quality Standard Subcommittee of the American Academy of Neurology and the Practice Committee of the Child Neurology Society. Neurology 2003;60 (2):166-75. 7. Engel J. International League against Epilepsy (ILAE). A proposed diagnostic scheme for people with epileptic seizures and with epilepsy: report of the ILAE Task Force on Classification and Terminology. Epilepsia 2001;42(6):796-803. 8. Fisher RS, van Emde Boas W, Blume W, Elger C, Genton P, Lee P, Engel J. Epileptic Seizures and Epilepsy: definitions proposed by the International League against Epilepsy (ILAE) and the International Bureau for Epilepsy (IBE). Epilepsia 2005;46:470-2. 9. Annegers JF, Shirts SB, Hauser WA, Kurland LT. Risk of recurrence after an initial unprovoked seizure. Epilepsia 1986;27(1):43-50. 10. Hauser WA, Rich S, Annegers JF, Anderson VE. Seizure recurrence after a 1st unprovoked seizure: an extended follow-up. Neurology 1990;40:1163-70.
Medical management of epilepsy
11. Temkin NR, Dikmen SS, Wilensky AJ, Keihm J, Chabal S, Winn HR. A randomized, double-blind study of phenytoin for the prevention of post-traumatic seizures. N Engl J Med 1990;323:497-502. 12. Bladin CF, Alexandrov AV, Bellavance A, Bornstein N, Chambers B, Cote R, LeBrun L, Pirisi A, Norris JW, for the Stroke Study Group. Seizures after stroke: a prospective multicenter study. Arch Neurol 2000;57:1617-22. 13. Rosengart AJ, Schultheiss KE, Tolentino J, Macdonald RL. Prognostic factors for outcome in patients with aneurysmal subarachnoid hemorrhage. Stroke 2007;38:2315-21. 14. Classen J, Jette N, Chum F, Green R, Schmidt M, Choi H, Jirsch J, Frontera JA, Connolly ES, Emerson RG, Mayer SA, Hirsch LJ. Electrographic seizures and periodic discharges after intracerebral hemorrhage. Neurology 2007;69:1356-65. 15. Naidech AM, Krieter KT, Janjua N, Ostapkovich N, Para A, Commichau C, Connolly ES, Jayer SA, Fitsimmons BF. Phenytoin exposure is associated with functional and cognitive disability after subarachnoid hemorrhage. Stroke 2005;36:583-7. 16. Dubinsky SG, Newmark E. Skin rashes after antiepileptic drug use. Epilepsia 1991;32 Suppl 3:17. 17. Glantz MJ, Cole BF, Forsyth PA, Recht LD, Wen PY, Chamberlain MC, Grossman SA, Cairncross JG. Practice parameter: anticonvulsant prophylaxis in patients with newly diagnosed brain tumors. Report of the Quality standards Subcommittee of the American Academy of Neurology. Neurology 2000;54:1886-93. 18. Peters ACB, Brouwer AF, Geerts AT, Arts WFM, Stroink H, van Donselaar CA. Randomized prospective study of early discontinuation of antiepileptic drugs in children. Neurology 1998;50:724-30. 19. Schiller Y, Cascino GD, So EL, Marsh WR. Discontinuation of antiepileptic drugs after successful epilepsy surgery. Neurology 2000;54:346-9. 20. Minicucci F, Muscas G, Perucca E, Capovilla G, Vigevano F, Tinuper P. Treatment of status epilepticus in adults: guidelines of the Italian League against Epilepsy. Epilepsia 2006;47 Suppl 5:9-15. 21. Varelas PN, Mirski MA. Seizures in the adult intensive care unit. J Neurosurg Anaesthesiol 2001;13:163-75. 22. Nadkarni S, Lajoie J, Devinsky O. Current treatments of epilepsy. Neurology 2005;64 Suppl 3:S2-S11. 23. Mattson RH, Cramer JA, Collins JF, Smith DB, Delgado-Escueda AV, Browne TR, Williamson PD, Treiman DM, McNamara JO, McCutchen CB et al. Comparison of carbamazepine, phenytoin, and primidone
24.
25.
26.
27. 28.
29.
30. 31.
32.
33.
34.
35.
159
in partial and secondarily generalized tonic clonic seizures. N Engl J Med 1985;313:145-51. Mattson RH, Cramer JA, Collins JF, and the Department of Veterans Affairs Epilepsy Cooperative Study No. 264 Group. A comparison of valproate with carbamazepine for the treatment of partial seizures and secondarily generalized seizures in adults. N Engl J Med 1992;327:765-71. Brodie MJ, Perucca E, Ryvlin P, Ben-Menachem E, Meencke HJ, Levetiracetam Monotherapy Study Group. Comparison of Levetiracetam and controlled-release carbamazepine in newly diagnosed epilepsy. Neurology 2007;68(6):402-8. French JA, Kanner AM, Bautista J, Abou-Khalil B, Browne T, Harden CL, Theodore WH, Bazil C, Stern J, Schachter SC, Bergen D, Hirtz D, Montouris GD, Nespeca M, Gidal B, et al. Efficacy and tolerability of the new antiepileptic drugs. I. Treatment of new onset epilepsy: report of the Therapeutics and Technology Assessment Subcommittee and Quality Standards Subcommittee of the American Academy of Neurology and the American Epilepsy Society. Neurology 2004;62:1252-60. Chaves J, Sander JW. Seizure aggravation in idiopathic generalized epilepsies. Epilepsia 2005;46 Suppl 9:133-9. Kaddurah AK, Holmes GL. Possible precipitation of myoclonic seizures with oxcarbazepine. Epilepsy Behav 2006;8(1):289-93. Lacerda G, Krummel T, Sabourdy C, Ryvlin P, Hirsch E. Optimizing therapy of seizures in patients with renal or hepatic dysfunction. Neurology 2006;67 Suppl 4:S28-S33. Sheth RD. Metabolic concerns associated with antiepileptic medications. Neurology 2004;63 Suppl 4:S24-S29. Bourgeois BFD. Pharmacokinetic properties of current antiepileptic drugs: what improvements are needed? Neurology 2000;55 Suppl 3: S11-S16. Dreifuss FE, Rosman P, Cloyd JC, Pellock JM, Kuzniecky RI, Lo WD, Matsuo F, Sharp GB, Conry JA, Bergen DC, Bell WE. A comparison of rectal diazepam gel and placebo for acute repetitive seizures. N Engl J Med 1998;338:1869-75. Newmark ME, Dubinsky SG. Concentrated oral diazepam as a rescue medicine for patients with epilepsy. Epilepsia 2004;45 Suppl 7:142-3. Chadwick D, Marson T. Choosing a first drug treatment for epilepsy after SANAD: randomized controlled trials, systemic reviews, guidelines, and treating patients. Epilepsia 2007:48(7):1259-63. French JA. Can evidence-based guidelines and clinical trials tell us how to treat patients? Epilepsia 2007;48(7):1264-7.
2675
158 MEG in Epilepsy A. Fujimoto . T. Akiyama . H. Otsubo
Introduction The field of magnetoencephalography (MEG) began in 1968 when Cohen used a 1-millionturn single induction coil to measure human alpha activity [1]. MEG was first applied to epilepsy by Barth et al. recording interictal spike discharges using a one-channel MEG sensor in two patients with partial seizures [2]. By 1989, multi-channel MEG sensors (37-channels) covering a relatively large region of the scalp were commercially available, making the clinical application of MEG efficient for patients with epilepsy. Currently all MEG machines are whole-head style and effectively cover most of the brain, making detection of intracerebral epileptic discharges feasible on a routine basis. This chapter describes basic physiology of MEG, characteristics of MEG spike sources, diagnosis of epilepsy, and epilepsy surgery.
Basic Physiology of MEG MEG is a technique that measures magnetic fields associated with intracellular current flow within neurons; unlike electroencephalography (EEG), which measures extracellular volume currents. Magnetic source imaging (MSI) is the combination of functional source-localization results derived from magnetoencephalographic recording, often using an equivalent current dipole (ECD) model as a source of intracranial electrical activity, coregistered with structural magnetic resonance imaging (MRI). MEG sensors utilize superconducting pickup coils to detect extracranial magnetic fields #
Springer-Verlag Berlin/Heidelberg 2009
generated by intraneuronal electric currents and Superconducting QUantum Interference Devices (SQUIDs) to amplify the fields [3,4], which are smaller than one millionth of environmental magnetic field. The neural currents that give rise to both MEG and EEG signals are caused by a flow of ions through postsynaptic dendritic membranes. The intracellular current flow within an individual neuron is quite small (10 14 nA), with a proportionately small magnetic field. It has been suggested that approximate one million synchronously activated neurons can produce a magnetic signal which is detectable extracranially [5]. It has been estimated that a typical dipole moment of 5 nA of somatosensory evoked responses on MEG, would correspond to the signals simultaneously arising from 100 to 250 mm2 of the cerebral cortex [6,7]. Studies using simultaneous MEG and subdural EEG recordings have shown that a finite area of synchronized epileptic activity is necessary to produce a detectable MEG signal [8–11]. When epileptic spikes in the convex cortical surface of the brain extend over 3 cm2 across a fissure, MEG spikes can be detected with a probability of greater than 50% and their localization correlate with the spatial extent and amplitude of spikes recorded simultaneously by electrocorticography (ECoG) [10]. In contrast to MEG, simultaneous recordings of scalp EEG and intracranial electrodes indicated that larger areas of 6–20 cm2 of cortex synchronously discharged to generate a measurable scalp-EEG spike [12–14]. According to Ørsted’s right-hand rule, which explains the relation between electrical current and magnetic field, current flows must
2662
158
MEG in epilepsy
be parallel to the surface of the skull in order for the magnetic field to penetrate the skull and be detected by sensors oriented perpendicular to the scalp. On the other hand, EEG mainly detects signals projecting radially towards the surface of brain. Thus, MEG is relatively more sensitive than the EEG to fissural sources, whose current flows are tangential to the surface of the scalp and which exist in two thirds of the surface of the human brain [15]. While, current flows in the gyral crests that are perpendicular to the scalp are only minimally detected by MEG [16,17]. Perhaps the most significant advantage of MEG compared to EEG is that, unlike the electrical fields, magnetic fields are not attenuated or distorted by intervening brain, dura, cerebrospinal fluid, bone, and skin [2,18]. Therefore, the sources of evoked magnetic fields and spontaneous magnetic discharges can be precisely localized and projected onto MRI with only a 2–3 mm localization error [19].
Characteristics of MEG Spike Sources MEG interictal epileptic events consist of spikes and sharp waves and are similar to EEG epileptic discharges. The ECD model can be used to localize the sources of such interictal MEG spikes on MSI. Because of higher resolutions of spatial and temporal analyses on MEG than EEG, MEG provides source localization of early components of complex of spikes. The areas and distribution characteristics of MEG spike sources have been correlated with the epileptogenic zones that have been demonstrated by intracranial video EEG and epilepsy surgery [20–26]. The distributional characteristics of MEG spike sources have been defined by number and density [21]. A spikes source ‘‘cluster’’ consists of six or more spike sources with 1 cm or less between adjacent sources. Clustered MEG spike
sources have been correlated with the area of cortical excision determined by intracranial video-EEG monitoring. A ‘‘scatter’’ pattern of spikes sources consists of fewer than six spike sources regardless of the distance between sources, or spike sources with more than 1 cm between sources regardless of the number of sources in a group. Coexisting scatters remote from clusters are considered to be non-epileptogenic and do not require excision. A scatter alone should be examined by intracranial VEEG monitoring because an epileptic zone may exist within this distribution.
Diagnosis of Epilepsy Epilepsies and epileptic syndromes are classified into three categories and three etiologies [27]. Their categories are localization-related epilepsy, generalized epilepsy and undetermined. The etiologies are idiopathic, symptomatic and cryptogenic. The localization-related epilepsy originates from cortex and is represented by partial seizures with and without secondary generalization. Idiopathic generalized epilepsies such as childhood absence epilepsy with generalized 3 Hz spike and wave discharges, are hypothesized to have a common central (midline/subcortical) pacemaker with centrencephalic origin [28]. In symptomatic generalized epilepsies both subcortical areas and cortices are considered to be affected. For example, Lennox-Gastaut syndrome presents atypical absences, sudden and symmetrical tonic, atonic and tonic-clonic seizures, and West syndrome presents epileptic spasms in infancy. Epilepsies and syndromes undetermined are associated with both generalized and focal seizures or without unequivocal generalized or focal features. Clustered MEG spike sources are frequently seen in localization-related epilepsy. A focal epileptic activity was rarely seen on MEG in children with Lennox-Gastaut syndrome [29]. Diverse
MEG in epilepsy
MEG dipole patterns corresponded to seizure types in West syndrome [30]. The relative contributions of thalamus and cortex to the pathophysiology of typical absence seizures in idiopathic generalized epilepsy are still a matter of debate (generalized or not) among clinicians and experimental researchers [31]. EEG frequently displays abrupt onset and cessation of bilaterally synchronous 3 Hz spike and slow waves over homologous cortical areas [32]. In a subset of frontal lobe epilepsy patients, EEGs show bilateral or diffuse epileptiform discharges mimicking the bilaterally synchronous 3 Hz spike and slow waves, while MEG using ECD modeling and high temporal resolution can localize focal MEG spike sources consistent with localization-related epilepsy [33,34]. Thus, MEG can identify the origin of diffuse or generalized spike discharges to differentiate localizationrelated epilepsy from generalized epilepsy in a subset of patients with intractable epilepsy with apparently generalized epileptiform discharges.
MEG for Epilepsy Surgery MEG for Lesional Epilepsy In patients with intractable localization related epilepsy, those with lesional epilepsy are the best surgical candidates, since there is high probability of seizure control or seizure freedom after the surgical resection of the epileptogenic lesion and zone. MSI is a powerful tool to delineate the epileptogenic zone correlating with MRI lesion [35]. For example, in lesional epilepsy secondary to brain tumor, the MEG spike sources have been shown to be asymmetrically localized to the lesion [36]. The maximum resection of tumor and complete resection of marginal extrinsic epileptogenic cortex resulted in a favorable seizure outcome in patients regardless of residual postexcisional extramarginal spikes [36,37].
158
Neuronal migration disorders are pathologically distinct and frequently associated with epilepsy (> Figure 158-1). Because of its intrinsic epileptogenicity, focal cortical dysplasias of neuronal migration disorders frequently cause intractable epilepsy [38]. Focal cortical dysplasias frequently show intrinsic clusters of MEG spike sources both within and extending from the MRI lesion [36,39]. Various patterns of MEG spike sources occurred in conjunction with multiple lesions and complex networks secondary to tuberous sclerosis complex with which patients frequently present intractable epilepsy [40–43]. The spike sources revealed surrounding the presumed epileptogenic tuber [42,44]. Cortical tubers themselves on MRI presented slow wave components on MEG frequency analysis [44].
MEG for Non-lesional Epilepsy MRI often does not aid in the presurgical evaluation, as shown in nearly 29% of intractable partial epilepsy patients in whom it was normal or showed nonspecific findings [45]. The outcome following epilepsy surgery in patients with normal brain MRI depends on the case selection criteria and expertise of the epilepsy center. In nonlesional intractable epilepsy, surgical candidates are selected based on concordant interictal and ictal scalp EEG findings, seizure semiology, SPECT [46], PET findings [47] and MEG [26]. The more concordant the results are, the greater the rationale for intracranial VEEG is for resective surgery. We studied 22 pediatric patients with normal MRI and clustered MEG spike sources for resective surgery [26]. Good postsurgical seizure outcome was attained in 17 (77%) children, including the eight patients who were seizure free. The pathology showed hidden neuronal migration disorders in nine (41%) patients. A restricted ictal onset zone within the clustered MEG spike sources was associated with postsurgical seizure freedom.
2663
2664
158
MEG in epilepsy
. Figure 158-1 MRI and MEG spike sources. This is a 3-year-old right-handed girl with a history of infantile spasms and partial seizures since 3 months old. (a) Axial T2 MRI shows reduced white matter (open circle) in the right inferior frontal lobe. (b) Axial T1 MRI shows 2 of clustered 15 MEG spike sources in the right inferior frontal region. She underwent intracranial video EEG to resect the inferior frontal and temporal lobe. The pathology was reported microdysgenesis
Children with multiple seizure types failed to achieve postsurgical seizure freedom, as did those in whom the MEG spike sources were either bilateral or diffuse. Thus, guided by MEG, good postsurgical outcome can be obtained in a majority of children with intractable epilepsy who have a normal or nonfocal MRI.
Clusterectomy, Resection of Clustered MEG Spike Sources A cluster of interictal MEG spike sources delineates the irritative zone, and the combination of frameless stereotactic method and MEG data promises to provide reliable and accurate information for performing lesionectomies around functional cortex [21]. Furthermore, the focal densely clustered MEG spike sources have been useful in guiding subdural grid placement to cover the entire epileptogenic zone even in the non-lesional epilepsy. A subset of MEG spike source clusters is indicative of the epileptogenic
zone [23,25]. To develop non-invasive methods for localizing and understanding the epileptic network in epilepsy patients with especially extratemporal lobe epilepsy, MEG has played important roles. In contrast to this concept of clusters or scatters using analysis of individual MEG spikes, averaged MEG spike sources point to a center of focal cortical dysplasia [48]. The center point should be removed to control seizures. The sources of epileptic discharges located neither symmetrically nor circularly from the core of epileptogenecity. In cases with intractable epilepsy, the pattern and distribution of the sources were more complicated and not simply classified compared to those of benign epilepsy in childhood [49]. The single ECD method is ideal to localize a simplified neuronal current at a single time point. However, when the ECD is applied to the epileptic discharges, ECD provides only an abstraction or a center of gravity in the epileptic areas at one time point [50–52]. Indeed, epileptic
MEG in epilepsy
discharges often rapidly propagate within the complicated epileptic networks to generate asynchronous interictal discharges [53,54]. When many different sources simultaneously activate and affect each other, the single ECD method cannot explain all of the exciting epileptic activities. Gradient magnetic field topography [55], dynamic statistical parametric maps [33,34] and synthetic apparatus magnetoencephalography [44] methods have been developed to improve the clinical application of MEG in the analysis of these complicated neuronal networks that are involved in epileptogenesis.
MEG for Mesial Temporal Lobe Epilepsy The network associated with the most common human intractable epilepsy is a mesial temporal/ limbic network [54]. This network includes
158
amygdala, hippocampus, entorhinal cortex, lateral temporal cortices, and extratemporal components of the mesial thalamus and inferior frontal lobe. Three MEG dipole models representing a part of the temporal discharges in this mesial temporal/limbic network have been described [56,57]. The three patterns based on orientation and location, were anterior temporal vertical (ATV), sources originating from basal temporal cortex [57]; posterior temporal vertical (PTV), sources originating from lateral or basal temporal cortex in nonlesional temporal-lobe epilepsy [57,58]; and anterior temporal horizontal (ATH). Pataraia et al. [58] suggested different limbic networks in patients with mesial temporal lobe epilepsy because independent anterior mesial horizontally and vertically oriented dipole patterns were observed on MEG. They suggested that anterior mesial horizontal dipoles involved the temporal pole and the anterior parts of the lateral temporal
. Figure 158-2 MRI and MEG spike sources. This is a 16-year-old left-handed girl with complex partial seizures with sensory aura, unresponsiveness, automatism and postictal confusion since 11 years old. (a) FLAIR MRI shows a multilocular lesion with a number of small cystic changes in the right anterior mesial temporal region. (b) Two MEG spike sources with horizontal orientation are localized in the anterior mesial region, indicating mesial temporal discharges to project the temporal pole and anterior parts of the lateral temporal lobe. This morphology and characteristics are consistent with a ganglioglioma
2665
2666
158
MEG in epilepsy
lobe (> Figure 158-2). Iwasaki et al. [59] reported that a subset of PTV sources was related to mesial temporal structures, since in some patients who became seizure free after anterior temporal lobectomy, PTV discharges disappeared postoperatively; though in other seizure-free cases, PTV sources remained on postoperative MEG, even after removal of mesial temporal structures. This dipole modeling, when used for temporal lobe epilepsy, shows that equivalent current dipoles of MEG localize to the mesial-temporal network rather than an exact location [57]. MEG dipole source modeling is less exact for mesial-temporal discharges. There are five reasons; (1) mesial-temporal areas are farther from MEG sensors [9] and magnetic fields attenuate in proportion to the squared distance from the source [16]; hence, MEG spikes are less prominent with mesial temporal discharges; (2) the cylindrical architecture of hippocampal neurons cancels the generated excitatory postsynaptic potentials (closed circuit), in contrast to the linear and laminar architecture of neocortical neurons (open circuit) [9]; (3) insufficient coverage of the subtemporal magnetic fields by some whole-head MEG sensor arrays increases errors for dipole estimation; (4) propagation to surrounding temporal structures through the limbic network is not suitable for application of single dipole analysis [53]; (5) magnetic fields from lateral and superior temporal cortices overwhelm those from mesial temporal structures. We reported an MEG study of a pediatric patient with intractable epilepsy secondary to a left temporo-parieto-occipital porencephalic cyst [60]. The patient’s left temporal lobe contained mesial, anterior, and basal structures but lacked superior and lateral cortex. The magnetic fields generated by the epileptic discharges in the mesio-basal temporal areas were detected by MEG without interference from the lateral and superior temporal cortices. MEG can detect the discharges of deep parahippocampal and fusiform gyri probably projected from hippocampus.
Summary MEG provides excellent spatio-temporal resolution of intracranial epileptic and functional activities. MEG and EEG appear to complement each other in the diagnosis of epilepsy and localization of interictal epileptiform discharges for epilepsy surgery. The combined analysis of whole-head MEG and intracranial video EEG promises to provide highly accurate localization of the epileptogenic zone and a more precise characterization of epileptic networks. In a subset of lesional epilepsy, MEG with neuronavigation system provides maximum lesionectomy and complete resection of adjacent epileptogenic zone of clustered MEG spike sources for seizure control. In the non-lesional epilepsy, MEG explores the surgical candidate with focal single epileptogenic zone for the intracranial video EEG. In the mesial temporal lobe epilepsy, the individual epileptic network should be considered from both MEG dipole patterns and the ictal video EEG data in each patient. This noninvasive technology will be a better diagnostic tool to improve seizure control for all patients, children and adults with intractable epilepsy.
References 1. Cohen D. Magnetoencephalography: evidence of magnetic fields produced by alpha rhythm currents. Science 1968;161:784-6. 2. Barth DS, Suthering W, Engel J, Jr, Beatty J. Neuromagnetic localization of epileptiform spike activity in the human brain. Science 1982;218:891-4. 3. Cohen D. Magnetoencephalography: detection of the brain’s electrical activity with a superconducting magnetometer. Science 1972;175:664-6. 4. Zimmerman JE, Thiene P, Harding JT. Design and operation of stable rf-biased superconducting quantum interference devices and a note on the properties of perfectly clean metal contacts. J Appl Phys 1970;41:1572-80. 5. Lewine JD, Orrison WW. Magnetoencephalography and magnetic source imaging. In: Orrison WW, Lewine JA, Hartshorne MF, editors. Functional brain imaging. St. Louis, MO: Mosby Year Book, Inc; 1995. p. 369-417.
MEG in epilepsy
6. Hari R. The neuromagnetic method in the study of the human auditory cortex. Adv Audiol 1990;6:222-82. 7. Ha¨ma¨la¨inen MS. Magnetoencephalography-thery, instrumentation, and applications to noninvasive studies of the working human brain. Rev Mod Phys 1993;65:413-97. 8. Baumgartner C, Barth DS, Levesque MF, Sutherling WW. Detection of epileptiform discharges on magnetoencephalography in comparison to invasive measurements. In: Hoke M, Erne SN, Okada YC, Romani GL, editors. Biomagnetism: clinical aspects. Amsterdam: Elsevier; 1992. p. 67-71. 9. Mikuni N, Nagamine T, Ikeda A, Terada K, Taki W, Kimura J, Kikuchi H, Shibasaki H. Simultaneous recording of epileptiform discharges by MEG and subdural electrodes in temporal lobe epilepsy. Neuroimage 1997;5:298-306. 10. Oishi M, Otsubo H, Kameyama S, Morota N, Masuda H, Kitayama M, Tanaka R. Epileptic spikes: magnetoencephalography versus simultaneous electrocorticography. Epilepsia 2002;43:1390-5. 11. Shigeto H, morioka T, hisada K, Nishio S, Ishibashi H, Kira D, Tobimatsu S, Kato M. Feasibility and limitations of magnetoencephalographic detection of epileptic discharges: simultaneous recordeing of magnetic fields and electrocorticography. Neurol Res 2002;24:531-6. 12. Abraham K, Ajmone Marsan C. Patterns of cortical discharges and their relation to routine scalp electroencephalography. Electroencephalogr Clin Neurophysiol 1958;10:447-61. 13. Cooper R, Winter AL, Crow HJ, Walter WG. Comparison of subcortical, cortical and scalp activity using chronically indwelling electrodes in man. Electroencephalogr Clin Neurophysiol 1965;18:217-28. 14. Tao JX, Ray A, Hawes-Ebersole S, Ebersole JS. Intracranial EEG substrates of scalp EEG interictal spikes. Epilepsia 2005;46:669-76. 15. Brodmann K. Vergleichende Localisationslehre der Grosshirnrinde in ihren Prinzipien dargestellt auf Grund des Zellenbaues. Leipzig, Barth JA reprinted 1925 1909; XII:324. 16. Sato S, Balish M, Muratore R. Principles of magnetoencephalography. J Clin Neurophysiol 1991;8:144-56. 17. Merlet I, Paetau R, Garcia-Larrea L, Uutela K, Granstro¨ ML, Mauguie`re. Apparent asynchrony between interictal electric and magnetic spikes. Neuroreport 1997; 8:1071-6. 18. Gallen CC, Hirschkoff EC, Buchanan DC. Magnetoencephalography and magnetic source imaging. Neuroimaging Clin N Am 1995:5:227-49. 19. Gallen CC, Sobel DF, Waltz T, Aung M, Copeland B, Schwartz BJ, Hirschkoff EC, Bloom FE. Noninvasive presurgical neuromagnetic mapping of somatosensory cortex. Neurosurgery 1993;33:260-8. 20. Pataraia E, Simos PG, Castillo EM, Billingsley RL, Sarkari S, Wheless JW, Maggio V, Maggio W, Baumgartner JE, Swank PR, Breier JI, Papanicolaou AC. Does magnetoencephalography add to scalp video-EEG as a
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
158
diagnostic tool in epilepsy surgery? Neurology 2004;23(62):943-8. Iida K, Otsubo H, Matsumoto Y, Ochi A, Oishi M, Holowka S, Pang E, Elliott I, Weiss SK, Chuang SH, Snead OC, III, Rutka JT. Characterizing magnetic spike sources by using magnetoencephalography-guided neuronavigation in epilepsy surgery in pediatric patients. J Neurosurg (Pediatrics 2) 2005;2S:187-96. Fischer MJ, Scheler G, Stefan H. Utilization of magnetoencephalography results to obtain favourable outcomes in epilepsy surgery. Brain 2005;128:153-7. Oishi M, Kameyama S, Masuda H, Tohyama J, Kanazawa O, Sasagawa M, Otsubo H. Single and multiple clusters of magnetoencephalographic dipoles in neocortical epilepsy: significance in characterizing the epileptogenic zone. Epilepsia 2006;47:355-64. Knowlton RC, Elgavish R, Howell J, Blount J, Burneo JG, Faught E, Kankirawatana P, Riley K, Morawetz R, Worthington J, Kuzniecky RI. Magnetic source imaging versus intracranial electroencephalogram in epilepsy surgery: a prospective study. Ann Neurol 2006;59:835-42. Mohamed IS, Otsubo H, Ochi A, Elliott I, Donner E, Chuang S, Sharma R, Holowka S, Rutka J, Snead OC, III. Utility of magnetoencephalography in the evaluation of recurrent seizures after epilepsy surgery. Epilepsia 2007;48:2150-9. RamachandranNair R, Otsubo H, Shroff MM, Ochi A, Weiss SK, Rutka JT, Snead OC, III. MEG predicts outcome following surgery for intractable epilepsy in children with normal or nonfocal MRI findings. Epilepsia 2007;48:149-57. Commission on classification and terminology of the international league against epilepsy. Proposal for revised classification of epilepsies and epileptic syndromes. Epilepsia 1989;30:389-99. Jasper HH, Kershman J. Electroencephalographic classification of the epilepsies. Arch Neurol Psychiat (Chicago) 1941;45:903-43. Sakurai K, Tanaka N, Kamada K, Takeuchi F, Takeda Y, Koyama T. Magnetoencephalographic studies of focal epileptic activity in three patients with epilepsy suggestive of Lennox-Gastaut syndrome. Epileptic Disord 2007;9:158-63. Hattori H, Yamano T, Tsutada T, Tsuyuguchi N, Kawawaki H, Shimogawara M. Magnetoencephalography in the detection of focal lesions in West syndrome. Brain Dev 2001;23:528-32. Meeren HK, Pijn JP, Van Luijtelaar EL, Coenen AM, Lopes da Silva FH. Cortical focus drives widespread corticothalamic networks during spontaneous absence seizures in rats. J neurosci 2002;22:1480-95. Niedermyer E, Lopes da Silva FH. Electroencephalography: basic principle, clinical applications and related fields. 5th ed. Philadelphia, PA: Lippincott, Williams and Wilkins; 2005. Shiraishi H, Ahlfors SP, Stufflebeam SM, Takano K, Okajima M, Knake S, Hatanaka K, Kohsaka S, Saitoh S,
2667
2668
158 34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
MEG in epilepsy
Dale AM, Halgren E. Application of magnetoencephalography in epilepsy patients with widespread spike or slowwave activity. Epilepsia 2005;46:1264-72. Hara K, Lin FH, Camposano S, Foxe DM, Grant PE, Bourgeois BF, Ahlfors SP, Stufflebeam SM. Magnetoencephalographic mapping of interictal spike propagation: a technical and clinical report. Am J Neuroradiol 2007;28:1486-8. Morioka T, Nishio S, Ishibashi H, Muranishi M, Hisada K, Shigeto H, Yamamoto T, Fukui M. Intrinsic epileptogenicity of focal cortical dysplasia as revealed by magnetoencephalography and electrocorticography. Epilepsy Res 1999;33:177-87. Otsubo H, Ochi A, Elliott I, Chuang SH, Rutka JT, Jay V, Aung M, Sobel DF, Snead OC. MEG predicts epileptic zone in lesional extrahippocampal epilepsy: 12 pediatric surgery cases. Epilepsia 2001;42:1523-30. Awad IA, Rosenfeld J, Ahl J, Hahn JF, Lu¨ders H. Intractable epilepsy and structural lesions of the brain: mapping, resection strategies, and seizure outcome. Epilepsia 1991;32:179-86. Palmini A, Gambardella A, Andermann F, Dubeau F, da Costa JC, Olivier A, Tampieri D, Gloor P, Quesney F, Andermann E, Paglioli E, Paglioli-Neto E, Coutinho L, Leblanc K, Kim HI. Intrinsic epileptogenicity of human dysplastic cortex as suggested by corticography and surgical results. Ann Neurol 1995;37:476-87. Otsubo H, Iida K, Okuda C, Ochi A, Pang E, Weiss SK, Chuang S, Rutka JT, Snead OC, III. Neurophysioligic findings of neuronal migration disorders: intrinsic epileptogenicity of focal cortical dysplasia on electroencephalography, electrocorticography, and magnetoencephalography. J Child Neurol 2005;20:357-63. Iida K, Otsubo H, Mohamed IS, Okuda C, Ochi A, Weiss SK, Chuang SH, Snead OC, III. Characterizing magnetoencephalographic spike sources in children with tuberous sclerosis complex. Epilepsia 2005;46:1510-7. Wu JY, Sutherling WW, Koh S, Salamon N, Jonas R, Yudovin S, Sankar R, Shields WD, Mathern GW. Magnetic source imaging localizes epileptogenic zone in children with tuberous sclerosis complex. Neurology 2006;66:1270-2. Jansen FE, Huiskamp G, van Huffelen AC, BourexSwart M, Boere E, Gebbink T, Vincken KL, van Nieuwenhuizen O. Identification of the epileptogenic tuber in patients with tuberous sclerosis: a comparison of high-resolution EEG and MEG. Epilepsia 2006; 47:108-14. Jansen FE, Van Huffelen AC, Van Rijen PC, Leijten FS, Jennekens-Schinkel A, Gosselaar P, Van Nieuwenhuizen O. Dutch collaborative epilepsy programme. Epilepsy surgery in tuberous sclerosis: the Dutch experience. Seizure 2007;16:445-53. Xiao Z, Xiang J, Holowka S, Hunjan A, Sharma R, Otsubo H, Chuang S. Volumetric localization of epileptic activities in tuberous sclerosis using sysnthetic aperture magnetometry. Pediatr Radiol 2006;36:16-21.
45. Semah F, Picot MC, Adam C, Broglin D, Arzimanoglou A, Bazin B, Cavalcanti D, Baulac M., Is the underlying cause of epilepsy a major prognostic factor for recurrence? Neurology 1998;51:1256-62. 46. Hong KS, Lee SK, Kim JY, Lee DS, Chung CK. Pre-surgical evaluation and surgical outcome of 41 patients with non-lesional neocortical epilepsy. Seizure 2002;11:184-92. 47. Lee SK, Lee SY, Kim KK, Hong KS, Lee DS, Chung CK. Surgical outcome and prognostic factors of cryptogenic neocortical epilepsy. Ann Neurol 2005;58:525-32. 48. Bast T, Oezkan O, Rona S, Stippich C, Seitz A, Rupp A, Fauser S, Zentner J, Rating D, Scherg M. EEG and MEG source analysis of single and averaged interictal spikes reveals intrinsic epileptogenicity in focal cortical dysplasia. Epilepsia 2004;45:621-31. 49. Chitoku S, Otsubo H, Ichimura T, Saigusa T, Ochi A, Shirasawa A, Kamijo K, Yamazaki T, Pang E, Rutka JT, Snead OC, III. Characteristics of dipoles in clustered individual spikes and averaged spikes. Brain and Development 2003;25:14-21. 50. Romani GL, Rossini P. Neuromagnetic functional localization: principles, state of the art, and perspectives. Brain Topogr 1988;1:5-21. 51. Scherg M. Functional imaging and localization of electromagnetic brain activity. Brain Topogr 1992;5:103-11. 52. Scherg M, Ebersole JS. Models of brain sources. Brain Topogr 1993;5:419-23. 53. Alarcon G, Guy CN, Binnie CD, Walker SR, Elwes RDC, Polkey CE. Intracerebral propagation of interictal activity in partial epilepsy: implications for source localization. J Neurol Neurosurg Psychiatry 1994;57:435-49. 54. Spencer SS. Neural networks in human epilepsy: evidence and implications for treatment. Epilepsia 2002;43:219-27. 55. Hashizume A, Iida K, Shirozu H, Hanaya R, Kiura Y, Kurisu K, Otsubo H. Gradient magnetic-field topography for dynamic changes of epileptic discharges. Brain Res 2007;1144:175-9. 56. Ebersole JS. Defining epileptogenic foci: past, present, future. J Clin Neuophysiol 1997;14:470-83. 57. Ebersole JS. Classification of MEG spikes in temporal lobe epilepsy. In: Yoshimoto T, Kotani M, Kuriki S,Karibe H, Nakasato N, editors. Recent advances in biomagnetism. Sendai: Tohoku University Press; 1999. p. 758-61. 58. Pataraia E, Lindinger G, Deecke L, Mayer D, Baumgartner C. Combined MEG/EEG analysis of the interictal spike complex in mesial temporal lobe epilepsy. Neuroimage 2005;24:607-14. 59. Iwasaki M, Nakasato N, Shamoto H, Nagamatsu K, Kanno A, Hatanaka K, Yoshimoto T. Surgical implications of neuromagnetic spike localization in temporal lobe epilepsy. Epilepsia 2002;43:415-24. 60. Imai K, Otsubo H, Sell E, Mohamed I, Ochi A, RamachandranNair R, Snead OC, III. MEG source estimation from mesio-basal temporal areas in a child with a porencephalic cyst. Acta Neurol Scand 2007;116:263-7.
164 Radiosurgery in Epilepsy I. Yang . N. M. Barbaro
Introduction Radiosurgery is a treatment technique for the precise application of focused radiation with stereotactic guidance to a targeted volume area within the brain identified on magnetic resonance imaging (MRI) [1]. Conceptualized by Leksell for use in functional neurosurgery, radiosurgical treatment for neurologic disorders has progressively widened its utility and is now also a treatment modality option for several neoplastic [2–51] and vascular indications [47,52–85]. Differing from standard dose fractionated radiotherapy, radiosurgery allows the neurosurgeon to deliver effective, precise, and accurate radiation to a smaller volume without effecting large portions of normal parenchyma allowing for a powerful radiobiologic effect on the chosen targeted volume [1,86–88]. Epilepsy is one of the most common serious neurological diseases with a prevalence of 0.5– 1.0% of the population in the U.S [89,90]. Approximately, 20% of patients with epilepsy have refractory seizures that fail to respond to medications. Despite recent and modern advances in new antiepileptic medications, the percentage of patients with medically refractory epilepsy has not improved. Patients with medically refractory seizures may be referred for possible surgical therapy and management, and approximately half of these patients are found to be suitable candidates for open neurosurgical resection of their seizure focus [90,91]. Focal partial epilepsies are typically responsive to open surgical resection treatments and are increasingly being treated using ‘‘structural’’ treatment modalities [84,92]. #
Springer-Verlag Berlin/Heidelberg 2009
The most common type of open surgery performed for epilepsy is an anterior temporal lobectomy, which is a resection of the anterior portion of the temporal lobe [90,92–95]. With modern advances in surgical and anesthetic techniques, microsurgical resection of mesial temporal lobe structures can be performed with low morbidity and even lower mortality [88]. Open surgical procedures though have inherent risks including damage to the brain (either directly or indirectly by injury to important blood vessels), bleeding (which can require re-operation), blood loss (which can require transfusion), infection (which can require re-operation and/or antibiotics), and general anesthetic risks [96–99]. In addition, significant postoperative pain can result from surgical incisions and scars. Several clinical studies evaluating the morbidity of open microsurgery for temporal lobectomy report that approximately 5–23% of epilepsy patients undergoing open microsurgery had a symptomatic neurologic deficit or complication post-operatively [93,96,97,100,101]. Furthermore, open surgical procedures require several days of in-hospital care typically including admission into the intensive care unit which contributes to the economic costs of open microsurgical treatment of medically refractory epilepsy [84]. There also exists a significant population of patients with medically intractable epilepsy that is unsuitable for conventional open microsurgery [84]. These patients may have their epileptic focus in difficult to access or eloquent regions of the brain where surgical resection could result in a neurological deficit such as irreversible language, motor, or visual impairment [84,85].
2762
164
Radiosurgery in epilepsy
Radiosurgery is now being evaluated as an alternative treatment modality to open resective microsurgery for medically intractable epilepsy [102]. High dose radiation is toxic to all living cells, but the highly focused and precise nature of stereotactic radiosurgery allows stereotactic guidance and the sparing of adjacent neurologic tissues from the damaging effects of radiation. Although performed in a hospital setting, stereotactic radiosurgery is relatively non-invasive using placement pins that only penetrate the skin in order to firmly fix the stereotactic frame to the skull. Typically, patients can return to full activity within 1–2 days after treatment as an outpatient procedure. Currently, stereotactic radiosurgery is under investigations as a treatment modality for epilepsy associated with vascular malformations, gelastic epilepsy associated with hypothalamic hamartomas, and medial temporal lobe epilepsy (MTLE) associated with mesial temporal sclerosis (MTS)[1,47,53,58,61,64,85,87,88,103–122].
Preclinical Evidence Pre-clinical studies investigating focused high dose radiosurgery in animal models of epilepsy have demonstrated the potential utility of radiosurgical treatment applied to non-human epilepsy models. Early animal experiments indicated the efficacy of focused radiation in a feline model of epilepsy to reduce seizure activity [84,103,104]. Using doses between 10 and 20 Gy (one gray, Gy, is equivalent to one joule of energy per kilogram of tissue), cats with epileptic foci treated with an implanted cobalt radiation source for focused radiation treatment had reduced seizure activity. Histologic analysis of these treated animal tissue specimens revealed ‘‘neuronal reafferentation’’ as a proposed potential mechanism for seizure amelioration due to focused radiation [103,104]. Recently, Sun and colleagues report that focused radiosurgery successfully reduces seizure activity in a pre-clinical animal epilepsy model
[85]. In this preclinical investigation, a linear accelerator was used to treat at doses of 10 or 40 Gy at the 90% isodose line. These investigators report that seizure thresholds to external electrical stimulation in these rats were significantly increased in the group treated with 40 Gy. These anti-epileptic effects were observed 1 week after radiosurgery treatment and persisted even at the 3 month follow-up period [85]. Other animal studies from the University of Virginia evaluated the effects of stereotactic radiosurgery treatment on a chronic spontaneous limbic epilepsy model in rodents [123]. In this preclinical investigation, hippocampal electrodes were implanted utilizing a single prolonged period of stimulation to produce a pre-clinical animal epilepsy model. Ten weeks later, Gamma Knife1 radiosurgery with doses between 10 and 40 Gy, was applied as the therapeutic modality. While the group receiving the lowest dose (10 Gy) showed no improvement in seizure activity, the 20 Gy group though did exhibit a gradual and continuous reduction in seizure occurrences between 2 and 6 months after radiosurgical treatment. Lastly, the 40 Gy treated rodent group displayed a dramatic reduction in seizures by the 2nd month after radiosurgery. Histologic analysis of targets treated with radiosurgery in this animal study revealed no necrosis in the radiation treated tissue specimens. Functionally, synaptically driven neuronal firing was found to be intact in these treated rodent brain slices, suggesting that neuronal death was associated with the identified seizure reduction [123]. Recent pre-clinical stereotactic radiosurgery experiments at the University of Pittsburgh were designed to determine the dose of radiation that was necessary to reduce seizure activity in a rat kainic-acid induced epilepsy model [113]. In this animal investigation, rats underwent stereotactic injection of kainic acid into the hippocampus to induce seizures, and then 10 days after the injections, the epileptic foci injection site was treated with Gamma Knife1 radiosurgery using a range
Radiosurgery in epilepsy
of 20–100 Gy. The animals treated with the lowest dose of 20 Gy demonstrated a progressive reduction in the number of daily seizures during each week of observation after radiosurgery treatment. Furthermore, 3 weeks after radiosurgery, all treated animals at each radiosurgery dose – 20, 40, 60 and 100 Gy – showed a statistically significant reduction in seizure activity confirmed with EEG evaluations. The authors report that histologic tissue evaluation only revealed radiation-induced necrosis at the 100 Gy radiosurgery dose. However, the injection of kainic acid induces a loss of CA3 neurons in all animals, and for this reason, the interpretation of histological findings especially for radiation induced necrosis may be difficult and problematic in this animal epilepsy model. Small areas of kainic acid induced necrosis though were seen in two out of twenty control animals and in 14 out of 37 radiosurgically treated animals, but only in the animals treated with the 100 Gy dose of radiosurgery did the observed histologic necrosis match the collimator size [113]. A second pre-clinical study using the same kainic acid induced epilepsy model was undertaken to further investigate the pathological and behavioral effects of ‘‘subnecrotic’’ radiosurgery doses [124]. Stereotactic hippocampal kainic acid injections were subsequently followed by Gamma Knife1 radiosurgery with doses of either 30 or 60 Gy. A statistically significant reduction in seizures was noted in all treated animals, but this effect was observed earlier in the rats treated with the higher radiosurgery dose of 60 Gy (weeks 5–9 compared to weeks 7–9). Furthermore, in this preclinical investigation, no animals treated with radiosurgery demonstrated a deficit in new memory attainment tasks with water maze testing compared to control animals only injected with kainic acid, but both groups with kainic acid induced epilepsy showing ‘‘cognitive’’ impairment when compared to controls without kainic acid injection. For the histologic tissue analysis in this study, two blinded, experienced
164
observers evaluated the histologic specimens from all animals at 13 weeks after radiosurgery treatment. Typical changes for kainic acid injections were seen in all injected animals including a loss of pyramidal cells in CA3–4. Furthermore, in 25 out of 46 injected animals, unilateral hippocampal atrophy with cell loss extending into CA1 and CA2 was also noted. Although histologic assessment is difficult to assess given the use of the kainic acid induced model as noted earlier, radiation induced necrosis matching the target volume of radiation was not observed in any of the animals treated with stereotactic radiosurgery [124]. These preclinical animal findings suggest that reduction of seizure activity following radiosurgery is not associated with necrosis or concomitant loss of neurons [124]. With the suggestion that necrosis is not associated with seizure reduction, more recently, the radiosurgery group in Prague report on their characterization of a ‘‘subnecrotic’’ radiation dose using radiosurgery in a rat animal model [125,126]. This preclinical investigation evaluated radiosurgery doses of 25, 50, 75 or 100 Gy delivered bilaterally to rat hippocampus, and then characterized memory function tests, MRI and histological examinations at 1, 3, 6 and 12 months following stereotactic radiosurgery treatment. A progressive time-dependent and dose-dependent response curve was observed in memory function, T2 weighted edema on MRI, and necrotic histopathology. Animals radiosurgically treated with the 100 Gy dose died by 6 months following radiation and all histopathology tissue specimens had radiation induced necrotic lesions. All animals treated with 75 Gy displayed memory functional impairments, T2 weighted edema on MRI, and radiation associated necrotic lesions, whereas only one of the animals treated with 50 Gy were noted to have edema and necrosis. Animals treated with radiosurgery doses of 25 and 50 Gy did not demonstrate any functional impairments after radiosurgery treatment [125]. This observation of a potential
2763
2764
164
Radiosurgery in epilepsy
subnecrotic radiosurgery dose associated with reduced seizure activity prompted a second preclinical animal study where a 35 Gy radiosurgery dose was used with a follow up period of 16 months after radiosurgery treatment [126]. In this long term pre-clinical study, after 6 months post-irradiation, T2 weighted edema was commonly observed on MR imaging, and this edema was most pronounced at 9 months after stereotactic radiosurgery. After 17 months, two of six treated animals had radiation associated necrotic cavities after treatment with a 35 Gy dose of radiosurgery. The four remaining treated animals without obvious necrotic cavities had other notable structural findings such as severe atrophy of the corpus callosum, loss of thickness of somatosensory cortex and damage to the striatum oriens hippocampi [126]. These preclinical animal studies suggest that the full radiobiologic and histological effect of stereotactic radiosurgery may only be manifested after several months following radiosurgery treatment. These preclinical animal studies report the amelioration of seizures as well as histologic neuronal tissue changes associated with radiosurgical treatment in different animal epilepsy models. These animal studies suggest that the antiepileptic efficacy of radiosurgery is time dependant and dose dependant [88,113,123,124]. Most of these studies suggest that a radiosurgery dose of approximately 25 Gy is required to see a therapeutic antiepileptic effect, and that the full histological and biological toxicity may require several months to fully develop and manifest [84,85,113,123– 126]. Animal models though may be poor predictors of radiation effects for translation to human biologic responses.
Clinical Evidence The first radiosurgical application for epilepsy surgery was utilized by Talairach in the 1950s with the implantation of radioactive yttrium in patients with MTLE without an obvious lesion [84,85,88].
Further clinical experiences with Gamma Knife1 radiosurgery and linear accelerator (LINAC) based radiosurgery for the treatment of arteriovenous malformations and low grade tumors have also reported the incidental antiepileptic effects of radiosurgery [58,64,88,107,111,112,122]. Using Gamma Knife(R) and LINAC to treat arteriovenous malformations (AVMs), several groups have also reported an incidental concomitant improvement in seizure control [47,53,120]. Although it is not clear whether lesion resolution itself may reduce seizure activity, these clinical reports of seizure improvement with radiosurgery provided the impetus for investigating stereotactic radiosurgery as an alternative structural treatment for medically intractable epilepsy.
Medial Temporal Lobe Epilepsy Medial temporal lobe epilepsy (MTLE) associated with mesial temporal sclerosis (MTS) is perhaps the most well defined epilepsy syndrome responsive to structural intervention. Mesial temporal sclerosis is an idiopathic process associated with an extensive loss of neurons and an associated increase in astrocytes in the mesial temporal structures including the amygdala and hippocampus. When temporal lobe epilepsy is due to underlying mesial temporal sclerosis, seizure improvements with open microsurgical structural resections can be expected in between 65 and 90% of patients. [84,90,92,95,127–132]. This form of intractable epilepsy is particularly amenable to structural interventions such as radiosurgery because 80–90% of these cases show detectable structural changes on magnetic resonance imaging (MRI) [87,129]. Recently, stereotactic radiosurgery has been explored as an alternative to open resective microsurgery for MTS associated medial temporal lobe epilepsy. In a small series of patients with MTLE treated with Gamma Knife(R) radiosurgery, Regis et al. reported improvement of
Radiosurgery in epilepsy
seizures with minimal morbidity and mortality [114,115]. A recent, prospective, multi-center European study evaluating Gamma Knife(R) radiosurgery for MTLE associated with MTS showed comparable efficacy rates (65%) for seizure reduction by conventional microsurgery or radiosurgery after 2 years of follow up [88]. Using a marginal dose of 24 Gy, Regis et al. demonstrate that radiosurgery can be effectively used as an alternative to conventional resective microsurgery to treat MTLE associated with MTS and also improve quality of life with comparable rates of morbidity and mortality [118]. In the United States, a multi-center Pilot Trial is currently being conducted with preliminary results showing that 85% of patients treated
164
with 24 Gy (to the 50% isodose line) to the medial temporal lobe, including the amygdala, anterior hippocampus and nearby cortex with 2 years of follow up are seizure-free with minimal morbidity (Barbaro, et al. unpublished). This study group is also planning a larger, Phase 3 multi-center trial comparing open microsurgery with radiosurgery for patients with clinically and radiographically defined MTS associated MTLE. Although radiosurgery has proven effective and safe in ameliorating MTS associated MTLE, the beneficial effects of radiosurgery are not demonstrated immediately. Typically, patients with MTLE treated with radiosurgery can achieve seizure reduction at 9–12 months and complete cessation of seizures between 18 and 24 months
. Figure 164-1 MRI changes at 12 months and 24 months after radiosurgery
2765
2766
164
Radiosurgery in epilepsy
after radiosurgery treatment. A transient increase in partial seizures (auras) can be noted at approximately the same time when complex seizures decrease in frequency [88]. Greater than half of all treated patients may require corticosteroids to treat the radiation-induced edema associated with the radiosurgical effect commonly most severe at 10–15 months post-treatment (> Figure 164-1) [88, Barbaro, personal observation]. One of the difficulties in applying stereotactic radiosurgery broadly as an application for intractable MTS associated MTLE is the definition of the radiosurgical target (> Figure 164-2). Because the MTS associated with MTLE is not clearly defined anatomically, the precise boundaries and structures for radiosurgical targeting are not well characterized. Hence standardization amongst different radiosurgery treatment centers is not yet implemented and may be difficult
to achieve. Successful radiosurgical treatment however has been shown to be associated with the structural targets treated with radiosurgery. For example, in recent reports, Regis et al. radiosurgically targeted the mesial temporal lobe structures in their series whereas Kawai et al. defined their treatment to the amygdala or hippocampus structures, and each series reported contrasting rates of successful amelioration of medial temporal lobe seizures with radiosurgery [88,109,114,115]. Although target definition may be variable amongst different practitioners, radiosurgery for MTS associated MTLE is an attractive therapeutic modality because of its low morbidity and mortality and the consistent presentation of this disease with identifiable imaging characteristics on MR imaging. Furthermore, conventional open anterior temporal lobectomy surgery can also be pursued if the initial radiosurgical treatment is
. Figure 164-2 Radiosurgery treatment planning session with Gamma Knife
Radiosurgery in epilepsy
ineffective and after sufficient time has been permitted for the delayed radiosurgical antiepileptic effect (after approximately 3 years) [88]. Furthermore, recent clinical dose studies have also suggested that a lower marginal dose of 20 Gy may be less effective in reducing seizures. Cmelak et al. report unsuccessful seizure reduction with stereotactic radiosurgery using a 15 Gy marginal dose [133]. Kawai et al. also report two cases of radiosurgery treatment with unsuccessful anti epileptic effect with a radiation dose of 18 Gy [109]. Finally, Srikijvilaikul et al. from the Cleveland Clinic also report ineffective seizure control with radiosurgical treatment with a 20 Gy radiation dose [134].
Histologic Evaluation After Radiosurgical Treatment for MTLE Histological examination of radiosurgically treated human mesial temporal tissue for MTLE has been limited due to the effective nature of stereotactic radiosurgery for MTS associated MTLE. However there are some histologic analysis of radiation treated tissues involving patients who underwent post radiosurgery resection due to ineffective seizure control after radiosurgery treatment [109,133,134]. Using a sub-therapeutic dose, Cmelak et al. report no radiationinduced histopathologic changes in tissues treated with 15 Gy of radiosurgery [133]. In another report with two patients treated with 18 Gy, one patient was noted to have a necrotic focus with some prominent vascular changes consisting of vessel-wall thickening, fibrinoid and hyaline degeneration, while the other patient treated with this subtherapeutic dose showed no necrosis or vascular histopathologic changes [109]. Treated with a higher, yet sub-therapeutic, dose of 20 Gy, all five patients from a series reported from the Cleveland Clinic demonstrated histopathologic necrosis, perivascular sclerosis, and macrophage infiltration upon resection and evaluation despite
164
the ineffective seizure control in this series [134]. These reports suggest that with clinical utilization of stereotactic radiosurgery, significant identifiable histologic tissue changes may only be present and observed in radiosurgical doses greater than or equal to 20 Gy. These radiobiologic and histological markers such as necrosis and vascular changes may be required for an effective antiseizure effect to manifest clinically. Thus, a dose that produces some tissue necrosis and histopathologic tissue effects without producing an excessive biologic response such as edema (perhaps 24 Gy) may be the optimal effective dose in the radiosurgical treatment of MTLE [88,114,115]. Currently, the radiobiology of stereotactic radiosurgery in the setting of MTS associated MTLE is not yet completely understood. While some preclinical studies have suggested an anti-epileptic effect of radiation with sub-necrotic doses [124], human clinical studies have suggested that a certain amount of tissue necrosis and histopathologic changes may be required to see a significant clinical amelioration of MTS associated seizures. The importance of this issue on radio biologic effect is that radiosurgical treatment of eloquent brain regions would be possible if an effective subnecrotic dose could be found.
Hypothalamic Hamartomas Associated Gelastic Epilepsy Hypothalamic hamartomas are uncommon lesions with a prevalence of 1–2 in 100,000 and is commonly associated with precocious puberty, developmental cognitive delay, and gelastic epilepsy [116,135]. Typically seizures associated with hypothalamic hamartomas are gelastic and medically refractory [116,135,136]. These hypothalamic hamartomas are ectopic tissue consisting of glia, neurons and other neuronal fiber bundles [116,135,136]. Microsurgical resection of hypothalamic hamartomas has been reported to improve
2767
2768
164
Radiosurgery in epilepsy
control of gelastic seizure activity, but due to the technical and surgical difficulties of reaching these lesion in a deep and critical area, open microsurgical resection is often complicated, incomplete and associated with a high risk of neurologic complications such as motor, visual and hypothalamic deficits [23,116,135–137]. Unger et al. report two patients with hypothalamic hamartomas associated gelastic seizures treated with low dose radiosurgery who demonstrated a significant seizure improvement after 36 and 54 months after radiosurgery treatment respectively [121]. This therapeutic delay in this report is consistent with the delayed radiobiological effects and manifestations of radiosurgery treatment. In a recent retrospective multi-center study, Regis et al. report 10 patients treated with 18 Gy radiosurgery that had improvement in their refractory gelastic seizures after radiosurgical treatment of hypothalamic hamartomas [116]. In a larger series from the same institution, 19 out of 30 patients were shown to have short-term improvements in gelastic seizure control after 6 months of follow-up, but further long-term follow-up data is still being evaluated [118]. This alternative treatment of stereotactic radiosurgery for hypothalamic hamartoma associated gelastic epilepsy holds great potential given the significant surgical morbidity associated with open microsurgical dissection of hypothalamic hamartomas. Further investigations with larger series, longer follow-up in a prospective manner must be conducted to establish the true safety, risk profile, and impact of this treatment option on refractory gelastic epilepsy associated with hypothalamic hamartoma.
Cavernous Malformation Associated Epilepsy The most common presentation of patients with cavernous malformations is seizure. These
congenital vascular malformations can cause hemorrhage and repetitive neurologic deficit, but more frequently manifest as repetitive seizures [117,138]. The incidence of medically intractable epilepsy associated with cavernous malformations is not yet established [117]. Radiosurgical treatment for cavernous malformations is controversial because clear evidence for protection from hemorrhage and the risk of re-hemorrhage has yet to be established [61,110,117]. Although resective, open microsurgical treatment of cavernous malformations remains the standard treatment modality, a recent series by Regis et al. suggest a role for radiosurgery in the treatment of seizures associated with challenging cavernous malformations near eloquent and ‘‘highly functional cortex’’ that may preclude open microsurgical resection [117]. Using a mean dose of 19 Gy in a series of 49 patients with cavernous malformation and refractory seizures, 59% of radiosurgically treated patients became seizure-free and 20% were significantly improved at 2 years follow up after radiosurgery treatment [117], demonstrating that epilepsy associated with cavernous malformations near eloquent cortex may be effectively treated with radiosurgery to reduce seizure frequency. Given the low bleeding risk of cavernous malformations in cortical regions and and the frequent symptoms of seizures [139], patients seeking an alternative to microsurgical open resection with decreased morbidity may opt for radiosurgical treatment of medically intractable seizures associated with these cavernous malformations. In addition to cavernous malformations near eloquent cortex, deep seated cavernous malformations near critical brain structures which are not easily accessible with open surgical procedures may also potentially be amenable to anti-epileptic therapy with radiosurgery treatment. Unfortunately these deep seated lesions have a higher risk of clinical bleeding and significant neurologic sequelae from hemorrhage [139], and the effect of stereotactic
Radiosurgery in epilepsy
radiosurgery on this hemorrhage risk is still unclear [61,110,117]. Without clear evidence of the effect of stereotactic radiosurgery on bleeding risk, microsurgical open resection remains the gold standard therapy for cavernous malformations, but the promising potential for radiosurgical treatment for cavernous malformation associated epilepsy is currently under investigation.
Long Term Radiosurgical Complications Although the long-term complications of radiosurgery are not yet fully characterized, it appears that these delayed risks are extremely minimal. There are several reported cases of radiosurgery associated ‘‘radiation-induced’’ malignancies, but these reported cases are rare [88,140–143]. Much longer periods of follow-up must be investigated to fully characterize the possible long term complications and risk of radiation induced neoplasms, as the development of new radiosurgery associated neoplasms may require decades to develop [17]. A conservative estimate of this chronic delayed risk of neoplasm after radiosurgery suggests this rate at less than 1% at 20 years post-treatment which may be a negligible risk of increased malignancy with stereotactic radiosurgery. Nonetheless, this may be an important consideration in younger treated patients who have a longer opportunity for oncogenesis.
The Antiepileptic Radiosurgery Mechanism Although radiosurgery has been shown to reduce seizures in various forms of medically intractable epilepsies, the mechanism by which this abatement occurs is not yet well understood. It has been suggested that radiation itself has a direct anti-epileptic effect that may operate through
164
several mechanisms. As glial cells are more radiosensitive than neurons, Barcia-Salorio proposed that low-dose radiosurgery may induce glial scar formation allowing increased dendritic sprouting and improved cortical reorganization resulting in fewer seizures [106]. Elomaa et al. have theorized that the anti-epileptic effect of radiation is further mediated through effects of somatostatin [144]. Although the clinical results of the most recent human studies suggest that the therapeutic efficacy of radiosurgery is linked to histopathologic tissue changes and identifiable necrosis of mesial temporal structures, proof for this theory would need to come from direct observation and histologic evaluation of tissue samples in patients where radiosurgery has effectively controlled seizures. This is unlikely to occur, as only patients with persistent seizures after stereotactic radiosurgery are likely to undergo further open resective microsurgery. Surrogate markers of radiation effect and radiobiology such as imaging changes on MRI have thus far shown variable results. Radiationinduced edema typically becomes evident in most patients 9–15 months following radiosurgery treatment (> Figure 164-1). These imaging findings though are usually time-limited and are often followed by focal atrophic changes. Thus, MRI changes may not be diagnostic or indicative of true radiation necrosis. Furthermore, our pilot clinical trials have shown that MRI changes and peak MRI effects were poorly correlated with post treatment symptoms or improvement in seizures. The actual bio-mechanism by which high-dose radiation and radiosurgery reduces neuronal hyper-excitability to ameliorate seizures will not likely be found or elucidated from human studies. Although preclinical animal evidence and the results from early clinical human trials suggest control of seizures might be possible with doses of radiosurgery that were lower than those typically applied to tumors [105,108], recent case reports also demonstrate the failure
2769
2770
164
Radiosurgery in epilepsy
of low dose radiosurgery to control seizures [109,133,134]. While failure of seizure control is easy to identify, it is a much more difficult task to determine that lack of seizure control is caused by an insufficient radiation dose. The time dependence of radiosurgical effects is also a confounding factor that has not been fully elucidated, and a consensus among different treating radiosurgical centers for when radiosurgical treatment has ‘‘failed’’ has not yet been agreed upon [88]. Furthermore, radiosurgery patients reported with inadequate seizure reduction commonly had radiation doses of 20 Gy or less, and these patients showed little evidence of radiation-induced necrosis or histopathologic changes in their tissue specimens [109,133,134]. Thus, the best evidence to date from human and animal preclinical experiments suggests that there is a steep dose-response curve for seizure reduction and that some neuronal necrosis is required to produce seizure abatement. This suggests that the radiosurgery radiation dose required to reduce seizures is very close to the absolute tolerability threshold of human brain tissue.
parenchyma. However, the available clinical human data suggests that it is necessary to produce changes on MRI consistent with tissue necrosis and histopathologic tissue changes in order to eliminate seizures. Recent prospective trials suggest that radiosurgery may be an effective and safe treatment modality for medically intractable epilepsy associated with MTS. Prospective trials with larger numbers of patients in multi-center studies will be required to establish radiosurgery as a standard alternative therapy for MTS associated MTLE. Further potential is being investigated in expanding the promising utility of radiosurgery for seizure control in medically intractable epilepsy associated with cavernous malformations and hypothalamic hamartomas. Radiosurgery may prove to be especially appealing in treating lesions near functional cortex or deep seated lesions when open microsurgical resection may not be feasible without significant morbidity.
References Conclusions Recent data suggests radiosurgery is an effective and safe alternative treatment modality for reducing epileptiform activity and seizures in several forms of medically intractable epilepsy. In preclinical studies, the low doses of radiation required to be therapeutic have not been shown to cause histologic tissue changes or significant learning deficits. When animals are observed over longer time periods, the patterns of changes seen on MRI closely mimic those observed in human trials, and associated histological analysis indicates that structural lesions are created. Animal studies have not yet proven if the antiepileptic effects of radiosurgery are due to tissue necrosis and functional ablation or if seizure activity has been eliminated in still functional
1. Nguyen DK, Spencer SS. Recent advances in the treatment of epilepsy. Arch Neurol 2003;60(7):929-35. 2. Adler JR, Cox RS, Kaplan I, Martin DP. Stereotactic radiosurgical treatment of brain metastases. J Neurosurg 1992;76(3):444-9. 3. Alexander E, III, Loeffler JS. Radiosurgery for primary malignant brain tumors. Semin Surg Oncol 1998;14 (1):43-52. 4. Andrews DW, Suarez O, Goldman HW, Downes MB, Bednarz G, Corn BW, et al. Stereotactic radiosurgery and fractionated stereotactic radiotherapy for the treatment of acoustic schwannomas: comparative observations of 125 patients treated at one institution. Int J Radiat Oncol Biol Phys 2001;50(5):1265-78. 5. Aoyama H, Shirato H, Tago M, Nakagawa K, Toyoda T, Hatano K, et al. Stereotactic radiosurgery plus wholebrain radiation therapy vs stereotactic radiosurgery alone for treatment of brain metastases: a randomized controlled trial. Jama 2006;295(21):2483-91. 6. Bakay RA. Stereotactic radiosurgery in the treatment of brain tumors. Clin Neurosurg 1992;39:292-313. 7. Becker G, Duffner F, Kortmann R, Weinmann M, Grote EH, Bamberg M. Radiosurgery for the treatment
Radiosurgery in epilepsy
8.
9.
10. 11. 12.
13. 14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
of brain metastases in renal cell carcinoma. Anticancer Res 1999;19(2C):1611-7. Black PM. Solitary brain metastases. Radiation, resection, or radiosurgery? Chest 1993;103 (4 Suppl): 367S-369S. Boyd TS, Mehta MP. Stereotactic radiosurgery for brain metastases. Oncology (Williston Park), 1999;13 (10):1397-409; discussion 1409–10, 13. Boyd TS, Mehta MP. Radiosurgery for brain metastases. Neurosurg Clin N Am 1999;10(2):337-50. Brada M. Radiosurgery for brain tumours. Eur J Cancer 1991;27(12):1545-8. Brada M, Laing R. Radiosurgery/stereotactic external beam radiotherapy for malignant brain tumours: the Royal Marsden Hospital experience. Recent Results Cancer Res 1994;135:91-104. Brophy BP. Acoustic neuroma–surgery or radiosurgery? Stereotact Funct Neurosurg 2000;74(3–4):121-8. Buatti JM, Friedman WA, Bova FJ, Mendenhall WM. Treatment selection factors for stereotactic radiosurgery of intracranial metastases. Int J Radiat Oncol Biol Phys 1995;32(4):1161-6. Coffey RJ, Lunsford LD, Flickinger JC. The role of radiosurgery in the treatment of malignant brain tumors. Neurosurg Clin N Am 1992;3(1):231-44. Combs SE, Thilmann C, Debus J, Schulz-Ertner D. Long-term outcome of stereotactic radiosurgery (SRS) in patients with acoustic neuromas. Int J Radiat Oncol Biol Phys 2006;64(5):1341-7. de Ipolyi AR, Yang I, Buckley A, Barbaro NM, Cheung SW, Parsa AT. Fluctuating response of a cystic vestibular schwannoma to radiosurgery: case report. Neurosurgery 2008;62(5):E1164-5; discussion E1165. DiBiase SJ, Kwok Y, Yovino S, Arena C, Naqvi S, Temple R, et al. Factors predicting local tumor control after gamma knife stereotactic radiosurgery for benign intracranial meningiomas. Int J Radiat Oncol Biol Phys 2004;60(5):1515-9. Engenhart R, Kimmig BN, Hover KH, Wowra B, Romahn J, Lorenz WJ, et al. Long-term follow-up for brain metastases treated by percutaneous stereotactic single high-dose irradiation. Cancer 1993;71(4):1353-61. Ewend MG, Carey LA, Brem H. Treatment of melanoma metastases in the brain. Semin Surg Oncol 1996;12 (6):429-35. Fedorcsak I, Sipos L, Horvath A, Kontra G, Bognar L, Osztie E. Multiple intracranial melanoma metastases treated with surgery and radiosurgery with long term control. A case report. J Neurooncol 1993;16(2):173-6. Flickinger JC, Kondziolka D, Lunsford LD, Coffey RJ, Goodman ML, Shaw EG, et al. A multi-institutional experience with stereotactic radiosurgery for solitary brain metastasis. Int J Radiat Oncol Biol Phys 1994;28 (4):797-802. Fohlen M, Lellouch A, Delalande O. Hypothalamic hamartoma with refractory epilepsy: surgical procedures
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
164
and results in 18 patients. Epileptic Disord 2003;5 (4):267-73. Foote KD, Friedman WA, Buatti JM, Meeks SL, Bova FJ, Kubilis PS. Analysis of risk factors associated with radiosurgery for vestibular schwannoma. J Neurosurg 2001;95 (3):440-9. Forster DM, Kemeny AA, Pathak A, Walton L. Radiosurgery: a minimally interventional alternative to microsurgery in the management of acoustic neuroma. Br J Neurosurg 1996;10(2):169-74. Friedman RA, Brackmann DE, Hitselberger WE, Schwartz MS, Iqbal Z, Berliner KI. Surgical salvage after failed irradiation for vestibular schwannoma. Laryngoscope 2005;115(10):1827-32. Gerosa M, Nicolato A, Foroni R. The role of gamma knife radiosurgery in the treatment of primary and metastatic brain tumors. Curr Opin Oncol 2003; 15(3):188-96. Gieger M, Wu JK, Ling MN, Wazer D, Tsai JS, Engler MJ. Response of intracranial melanoma metastases to stereotactic radiosurgery. Radiat Oncol Investig 1997;5 (2):72-80. Goyal S, Prasad D, Harrell F, Jr, Matsumoto J, Rich T, Steiner L. Gamma knife surgery for the treatment of intracranial metastases from breast cancer. J Neurosurg 2005;103(2):218-23. Hakim R, Alexander E, III, Loeffler JS, Shrieve DC, Wen P, Fallon MP, et al. Results of linear acceleratorbased radiosurgery for intracranial meningiomas. Neurosurgery 1998;42(3):446-53; discussion 453–4. Harris AE, Lee JY, Omalu B, Flickinger JC, Kondziolka D, Lunsford LD. The effect of radiosurgery during management of aggressive meningiomas. Surg Neurol 2003;60 (4):298-305; discussion 305. Hayhurst C, Dhir J, Dias PS. Stereotactic radiosurgery and vestibular schwannoma: hydrocephalus associated with the development of a secondary arachnoid cyst: a report of two cases and review of the literature. Br J Neurosurg 2005;19(2):178-81. Huang CF, Tu HT, Lo HK, Wang KL, Liu WS. Radiosurgery for vestibular schwannomas. J Chin Med Assoc 2005;68(7):315-20. Hudgins WR, Barker JL, Schwartz DE, Nichols TD. Gamma Knife treatment of 100 consecutive meningiomas. Stereotact Funct Neurosurg 1996;66 Suppl 1:121-8. Iwai Y, Yamanaka K. Gamma Knife radiosurgery for skull base metastasis and invasion. Stereotact Funct Neurosurg 1999;72 Suppl 1:81-7. Iwai Y, Yamanaka K, Ishiguro T. Gamma knife radiosurgery for the treatment of cavernous sinus meningiomas. Neurosurgery 2003;52(3):517-24; discussion 523–4. Jawahar A, Shaya M, Campbell P, Ampil F, Willis BK, Smith D, et al. Role of stereotactic radiosurgery as a primary treatment option in the management of newly diagnosed multiple (3–6) intracranial metastases. Surg Neurol 2005;64(3):207-12.
2771
2772
164
Radiosurgery in epilepsy
38. Kamath R, Ryken TC, Meeks SL, Pennington EC, Ritchie J, Buatti JM. Initial clinical experience with frameless radiosurgery for patients with intracranial metastases. Int J Radiat Oncol Biol Phys 2005; 61(5):1467-72. 39. Kobayashi T, Kida Y, Mori Y. Long-term results of stereotactic gamma radiosurgery of meningiomas. Surg Neurol 2001;55(6):325-31. 40. Koc M, McGregor J, Grecula J, Bauer CJ, Gupta N, Gahbauer RA. Gamma Knife radiosurgery for intracranial metastatic melanoma: an analysis of survival and prognostic factors. J Neurooncol 2005;71(3): 307-13. 41. Kocher M. [Radiosurgery for brain metastases of acoustic neurinoma]. Strahlenther Onkol 2001;177(6):313-4. 42. Koehler M, Merlo A. [Treatment of brain metastases]. Ther Umsch 2001;58(12):732-7. 43. Kondziolka D, Lunsford LD, Coffey RJ, Flickinger JC. Stereotactic radiosurgery of meningiomas. J Neurosurg 1991;74(4):552-9. 44. Kreil W, Luggin J, Fuchs I, Weigl V, Eustacchio S, Papaefthymiou G. Long term experience of gamma knife radiosurgery for benign skull base meningiomas. J Neurol Neurosurg Psychiatry 2005;76(10):1425-30. 45. Lee JY, Niranjan A, McInerney J, Kondziolka D, Flickinger JC, Lunsford LD. Stereotactic radiosurgery providing long-term tumor control of cavernous sinus meningiomas. J Neurosurg 2002;97(1):65-72. 46. Linskey ME, Lunsford LD, Flickinger JC. Tumor control after stereotactic radiosurgery in neurofibromatosis patients with bilateral acoustic tumors. Neurosurgery 1992;31(5):829-38; discussion 838–9. 47. Lunsford LD, Kondziolka D, Flickinger JC, Bissonette DJ, Jungreis CA, Maitz AH, et al. Stereotactic radiosurgery for arteriovenous malformations of the brain. J Neurosurg 1991;75(4):512-24. 48. Lunsford LD, Kondziolka DS, Flickinger JC. Radiosurgery of tumors of the cerebellopontine angle. Clin Neurosurg 1994;41:168-84. 49. Nicolato A, Ferraresi P, Foroni R, Pasqualin A, Piovan E, Severi F, et al. Gamma Knife radiosurgery in skull base meningiomas. Preliminary experience with 50 cases. Stereotact Funct Neurosurg 1996;66 Suppl 1:112-20. 50. Noren G. Gamma knife radiosurgery of acoustic neurinomas. A historic perspective. Neurochirurgie 2004;50(2–3) (pt 2) 253-6. 51. Nwokedi EC, DiBiase SJ, Jabbour S, Herman J, Amin P, Chin LS. Gamma knife stereotactic radiosurgery for patients with glioblastoma multiforme. Neurosurgery 2002;50(1):41-6; discussion 46–7. 52. Barker FG, II, Butler WE, Lyons S, Cascio E, Ogilvy CS, Loeffler JS, et al. Dose-volume prediction of radiationrelated complications after proton beam radiosurgery for cerebral arteriovenous malformations. J Neurosurg 2003;99(2):254-63.
53. Betti OO, Munari C, Rosler R. Stereotactic radiosurgery with the linear accelerator: treatment of arteriovenous malformations. Neurosurgery 1989;24(3): 311-21. 54. Bollet MA, Anxionnat R, Buchheit I, Bey P, Cordebar A, Jay N, et al. Efficacy and morbidity of arc-therapy radiosurgery for cerebral arteriovenous malformations: a comparison with the natural history. Int J Radiat Oncol Biol Phys 2004;58(5):1353-63. 55. Chang TC, Shirato H, Aoyama H, Ushikoshi S, Kato N, Kuroda S, et al. Stereotactic irradiation for intracranial arteriovenous malformation using stereotactic radiosurgery or hypofractionated stereotactic radiotherapy. Int J Radiat Oncol Biol Phys 2004;60(3):861-70. 56. de Tribolet N. Radiosurgery of deep arteriovenous malformations. J Neurosurg 2004;100(2):205-6; discussion 206-7. 57. Duffner F, Freudenstein D, Becker G, Ernemann U, Grote EH. Combined treatment effects after embolization and radiosurgery in high-grade arteriovenous malformations. Case report and review of the literature. Stereotact Funct Neurosurg 2000;75(1):27-34. 58. Eisenschenk S, Gilmore RL, Friedman WA, Henchey RA. The effect of LINAC stereotactic radiosurgery on epilepsy associated with arteriovenous malformations. Stereotact Funct Neurosurg 1998;71(2):51-61. 59. Flickinger JC, Kondziolka D, Lunsford LD, Pollock BE, Yamamoto M, Gorman DA, et al. A multi-institutional analysis of complication outcomes after arteriovenous malformation radiosurgery. Int J Radiat Oncol Biol Phys 1999;44(1):67-74. 60. Friedman WA, Bova FJ, Bollampally S, Bradshaw P. Analysis of factors predictive of success or complications in arteriovenous malformation radiosurgery. Neurosurgery 2003;52(2):296-307; discussion 307–8. 61. Karlsson B, Kihlstrom L, Lindquist C, Ericson K, Steiner L. Radiosurgery for cavernous malformations. J Neurosurg 1998;88(2):293-7. 62. Karlsson B, Lax I, Soderman M. Risk for hemorrhage during the 2-year latency period following gamma knife radiosurgery for arteriovenous malformations. Int J Radiat Oncol Biol Phys 2001;49(4):1045-51. 63. Karlsson B, Lax I, Soderman M, Kihlstrom L, Lindquist C. Prediction of results following Gamma Knife surgery for brain stem and other centrally located arteriovenous malformations: relation to natural course. Stereotact Funct Neurosurg 1996;66 Suppl 1:260-8. 64. Kida Y, Kobayashi T, Tanaka T, Mori Y, Hasegawa T, Kondoh T. Seizure control after radiosurgery on cerebral arteriovenous malformations. J Clin Neurosci 2000;7 Suppl 1:6-9. 65. Kurita H, Ueki K, Shin M, Kawamoto S, Sasaki T, Tago M, et al. Headaches in patients with radiosurgically treated occipital arteriovenous malformations. J Neurosurg 2000;93(2):224-8.
Radiosurgery in epilepsy
66. Kwon Y, Jeon SR, Kim JH, Lee JK, Ra DS, Lee DJ, et al. Analysis of the causes of treatment failure in gamma knife radiosurgery for intracranial arteriovenous malformations. J Neurosurg 2000;93 Suppl 3:104-6. 67. Levegrun S, Hof H, Essig M, Schlegel W, Debus J. Radiation-induced changes of brain tissue after radiosurgery in patients with arteriovenous malformations: correlation with dose distribution parameters. Int J Radiat Oncol Biol Phys 2004;59(3):796-808. 68. Levy EI, Niranjan A, Thompson TP, Scarrow AM, Kondziolka D, Flickinger JC, et al. Radiosurgery for childhood intracranial arteriovenous malformations. Neurosurgery 2000;47(4):834-41; discussion 841–2. 69. Maesawa S, Flickinger JC, Kondziolka D, Lunsford LD. Repeated radiosurgery for incompletely obliterated arteriovenous malformations. J Neurosurg 2000;92(6):961-70. 70. Maruyama K, Kawahara N, Shin M, Tago M, Kishimoto J, Kurita H, et al. The risk of hemorrhage after radiosurgery for cerebral arteriovenous malformations. N Engl J Med 2005;352(2):146-53. 71. Miyawaki L, Dowd C, Wara W, Goldsmith B, Albright N, Gutin P, et al. Five year results of LINAC radiosurgery for arteriovenous malformations: outcome for large AVMS. Int J Radiat Oncol Biol Phys 1999;44 (5):1089-106. 72. Nataf F, Ghossoub M, Schlienger M, Moussa R, Meder JF, Roux FX. Bleeding after radiosurgery for cerebral arteriovenous malformations. Neurosurgery 2004;55 (2):298-305; discussion 305–6. 73. Nicolato A, Lupidi F, Sandri MF, Foroni R, Zampieri P, Mazza C, et al. Gamma Knife radiosurgery for cerebral arteriovenous malformations in children/adolescents and adults. Part I: Differences in epidemiologic, morphologic, and clinical characteristics, permanent complications, and bleeding in the latency period. Int J Radiat Oncol Biol Phys 2006;64(3):904-13. 74. Pan DH, Guo WY, Chung WY, Shiau CY, Chang YC, Wang LW. Gamma knife radiosurgery as a single treatment modality for large cerebral arteriovenous malformations. J Neurosurg 2000;93 Suppl 3:113-9. 75. Pollock BE, Flickinger JC, Lunsford LD, Maitz A, Kondziolka D. Factors associated with successful arteriovenous malformation radiosurgery. Neurosurgery 1998;42(6):1239-44; discussion 1244–7. 76. Pollock BE, Gorman DA, Coffey RJ. Patient outcomes after arteriovenous malformation radiosurgical management: results based on a 5- to 14-year follow-up study. Neurosurgery 2003;52(6):1291-6; discussion 1296–7. 77. Schlienger M, Atlan D, Lefkopoulos D, Merienne L, Touboul E, Missir O, et al. Linac radiosurgery for cerebral arteriovenous malformations: results in 169 patients. Int J Radiat Oncol Biol Phys 2000;46(5):1135-42. 78. Shin M, Maruyama K, Kurita H, Kawamoto S, Tago M, Terahara A, et al. Analysis of nidus obliteration rates after gamma knife surgery for arteriovenous
164
malformations based on long-term follow-up data: the University of Tokyo experience. J Neurosurg 2004; 101(1):18-24. 79. Smyth MD, Sneed PK, Ciricillo SF, Edwards MS, Wara WM, Larson DA, et al. Stereotactic radiosurgery for pediatric intracranial arteriovenous malformations: the University of California at San Francisco experience. J Neurosurg 2002;97(1):48-55. 80. Tanaka T, Kobayashi T, Kida Y, Oyama H, Niwa M. Comparison between adult and pediatric arteriovenous malformations treated by Gamma Knife radiosurgery. Stereotact Funct Neurosurg 1996;66 Suppl 1:288-95. 81. Vernimmen FJ, Slabbert JP, Wilson JA, Fredericks S, Melvill R. Stereotactic proton beam therapy for intracranial arteriovenous malformations. Int J Radiat Oncol Biol Phys 2005;62(1):44-52. 82. Zabel-du Bois A, Milker-Zabel S, Huber P, Schlegel W, Debus J. Stereotactic linac-based radiosurgery in the treatment of cerebral arteriovenous malformations located deep, involving corpus callosum, motor cortex, or brainstem. Int J Radiat Oncol Biol Phys 2006;64(4):1044-8. 83. Zipfel GJ, Bradshaw P, Bova FJ, Friedman WA. Do the morphological characteristics of arteriovenous malformations affect the results of radiosurgery? J Neurosurg 2004;101(3):393-401. 84. Kitchen N. Experimental and clinical studies on the putative therapeutic efficacy of cerebral irradiation (radiotherapy) in epilepsy. Epilepsy Res 1995;20(1):1-10. 85. Sun B, DeSalles AA, Medin PM, Solberg TD, Hoebel B, Felder-Allen M, et al. Reduction of hippocampal-kindled seizure activity in rats by stereotactic radiosurgery. Exp Neurol 1998;154(2):691-5. 86. Kondziolka D, Lunsford LD, Witt TC, Flickinger JC. The future of radiosurgery: radiobiology, technology, and applications. Surg Neurol 2000;54(6):406-14. 87. Dillon WP, Barbaro N. Noninvasive surgery for epilepsy: the era of image guidance. AJNR Am J Neuroradiol 1999;20(2):185. 88. Regis J, Rey M, Bartolomei F, Vladyka V, Liscak R, Schrottner O, et al. Gamma knife surgery in mesial temporal lobe epilepsy: a prospective multicenter study. Epilepsia 2004;45(5):504-15. 89. Cascino GD. Epilepsy: contemporary perspectives on evaluation and treatment. Mayo Clin Proc 1994;69 (12):1199-211. 90. Engel J, Jr. Surgery for seizures. N Engl J Med 1996;334 (10):647-52. 91. Cohen-Gadol AA, Wilhelmi BG, Collignon F, White JB, Britton JW, Cambier DM, et al. Long-term outcome of epilepsy surgery among 399 patients with nonlesional seizure foci including mesial temporal lobe sclerosis. J Neurosurg 2006;104(4):513-24. 92. Wiebe S, Blume WT, Girvin JP, Eliasziw M. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med 2001;345(5):311-8.
2773
2774
164
Radiosurgery in epilepsy
93. Spencer SS. Long-term outcome after epilepsy surgery. Epilepsia 1996;37(9):807-13. 94. Spencer SS, Berg AT, Vickrey BG, Sperling MR, Bazil CW, Shinnar S, et al. Predicting long-term seizure outcome after resective epilepsy surgery: the multicenter study. Neurology 2005;65(6):912-8. 95. Spencer SS, Berg AT, Vickrey BG, Sperling MR, Bazil CW, Shinnar S, et al. Initial outcomes in the Multicenter Study of Epilepsy Surgery. Neurology 2003;61(12): 1680- 5. 96. Behrens E, Schramm J, Zentner J, Konig R. Surgical and neurological complications in a series of 708 epilepsy surgery procedures. Neurosurgery 1997;41(1):1-9; discussion 9–10. 97. Rydenhag B, Silander HC. Complications of epilepsy surgery after 654 procedures in Sweden, September 1990–1995: a multicenter study based on the Swedish National Epilepsy Surgery Register. Neurosurgery 2001;49(1):51-6; discussion 56–7. 98. Siegel AM, Cascino GD, Meyer FB, Marsh WR, Scheithauer BW, Sharbrough FW. Surgical outcome and predictive factors in adult patients with intractable epilepsy and focal cortical dysplasia. Acta Neurol Scand 2006;113(2):65-71. 99. Sperling MR, Feldman H, Kinman J, Liporace JD, O’Connor MJ. Seizure control and mortality in epilepsy. Ann Neurol 1999;46(1):45-50. 100. Guldvog B, Loyning Y, Hauglie-Hanssen E, Flood S, Bjornaes H. Surgical versus medical treatment for epilepsy. I. Outcome related to survival, seizures, and neurologic deficit. Epilepsia 1991;32(3):375-88. 101. Guldvog B, Loyning Y, Hauglie-Hanssen E, Flood S, Bjornaes H. Surgical versus medical treatment for epilepsy. II. Outcome related to social areas. Epilepsia 1991;32(4):477-86. 102. Yang I, Barbaro NM. Advances in the radiosurgical treatment of epilepsy. Epilepsy Curr 2007;7(2):31-5. 103. Barcia Salorio JL, Roldan P, Hernandez G, Lopez Gomez L. Radiosurgical treatment of epilepsy. Appl Neurophysiol 1985;48(1–6):400-3. 104. Barcia-Salorio JL, Vanaclocha V, Cerda M, Ciudad J, Lopez-Gomez L. Response of experimental epileptic focus to focal ionizing radiation. Appl Neurophysiol 1987;50(1–6):359-64. 105. Barcia-Salorio JL, Barcia JA, Hernandez G, LopezGomez L. Radiosurgery of epilepsy. Long-term results. Acta Neurochir Suppl 1994;62:111-3. 106. Barcia-Salorio JL. Radiosurgery in epilepsy and neuronal plasticity. Adv Neurol 1999;81:299-305. 107. Heikkinen ER, Konnov B, Melnikov L, Yalynych N, Zubkov Yu N, Garmashov Yu A, et al. Relief of epilepsy by radiosurgery of cerebral arteriovenous malformations. Stereotact Funct Neurosurg 1989;53(3):157-66. 108. Heikkinen ER, Heikkinen MI, Sotaniemi K. Stereotactic radiotherapy instead of conventional epilepsy surgery.
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
121.
122.
A case report. Acta Neurochir (Wien) 1992;119(1– 4):159-60. Kawai K, Suzuki I, Kurita H, Shin M, Arai N, Kirino T. Failure of low-dose radiosurgery to control temporal lobe epilepsy. J Neurosurg 2001;95(5):883-7. Kondziolka D, Lunsford LD, Flickinger JC, Kestle JR. Reduction of hemorrhage risk after stereotactic radiosurgery for cavernous malformations. J Neurosurg 1995;83(5):825-31. Kurita H, Kawamoto S, Suzuki I, Sasaki T, Tago M, Terahara A, et al. Control of epilepsy associated with cerebral arteriovenous malformations after radiosurgery. J Neurol Neurosurg Psychiatry 1998;65(5):648-55. Kurita H, Suzuki I, Shin M, Kawai K, Tago M, Momose T, et al. Successful radiosurgical treatment of lesional epilepsy of mesial temporal origin. Minim Invasive Neurosurg 2001;44(1):43-6. Mori Y, Kondziolka D, Balzer J, Fellows W, Flickinger JC, Lunsford LD, et al. Effects of stereotactic radiosurgery on an animal model of hippocampal epilepsy. Neurosurgery 2000;46(1):157-65; discussion 165–8. Regis J, Peragui JC, Rey M, Samson Y, Levrier O, Porcheron D, et al. First selective amygdalohippocampal radiosurgery for ‘mesial temporal lobe epilepsy’. Stereotact Funct Neurosurg 1995;64 Suppl 1:193-201. Regis J, Bartolomei F, Rey M, Genton P, Dravet C, Semah F, et al. Gamma knife surgery for mesial temporal lobe epilepsy. Epilepsia 1999;40(11):1551-6. Regis J, Bartolomei F, de Toffol B, Genton P, Kobayashi T, Mori Y, et al. Gamma knife surgery for epilepsy related to hypothalamic hamartomas. Neurosurgery 2000;47 (6):1343-51; discussion 1351–2. Regis J, Bartolomei F, Kida Y, Kobayashi T, Vladyka V, Liscak R, et al. Radiosurgery for epilepsy associated with cavernous malformation: retrospective study in 49 patients. Neurosurgery 2000;47(5):1091-7. Regis J, Hayashi M, Eupierre LP, Villeneuve N, Bartolomei F, Brue T, et al. Gamma knife surgery for epilepsy related to hypothalamic hamartomas. Acta Neurochir Suppl 2004;91:33-50. Regis J, Bartolomei F. Comment on: failure of Gamma Knife radiosurgery for mesial temporal lobe epilepsy: report of five cases. Neurosurgery 2004;54(6):1404. Sutcliffe JC, Forster DM, Walton L, Dias PS, Kemeny AA. Untoward clinical effects after stereotactic radiosurgery for intracranial arteriovenous malformations. Br J Neurosurg 1992;6(3):177-85. Unger F, Schrottner O, Haselsberger K, Korner E, Ploier R, Pendl G. Gamma knife radiosurgery for hypothalamic hamartomas in patients with medically intractable epilepsy and precocious puberty. Report of two cases. J Neurosurg 2000;92(4):726-31. Whang CJ, Kwon Y. Long-term follow-up of stereotactic Gamma Knife radiosurgery in epilepsy. Stereotact Funct Neurosurg 1996;66 Suppl 1:349-56.
Radiosurgery in epilepsy
123. Chen ZF, Kamiryo T, Henson SL, Yamamoto H, Bertram EH, Schottler F, et al. Anticonvulsant effects of gamma surgery in a model of chronic spontaneous limbic epilepsy in rats. J Neurosurg 2001;94(2):270-80. 124. Maesawa S, Kondziolka D, Dixon CE, Balzer J, Fellows W, Lunsford LD. Subnecrotic stereotactic radiosurgery controlling epilepsy produced by kainic acid injection in rats. J Neurosurg 2000;93(6):1033-40. 125. Liscak R, Vladyka V, Novotny J, Jr, Brozek G, Namestkova K, Mares V, et al. Leksell gamma knife lesioning of the rat hippocampus: the relationship between radiation dose and functional and structural damage. J Neurosurg 2002;97 (5 Suppl):666-73. 126. Herynek V, Burian M, Jirak D, Liscak R, Namestkova K, Hajek M, et al. Metabolite and diffusion changes in the rat brain after Leksell Gamma Knife irradiation. Magn Reson Med 2004;52(2):397-402. 127. Bien CG, Kurthen M, Baron K, Lux S, Helmstaedter C, Schramm J, et al. Long-term seizure outcome and antiepileptic drug treatment in surgically treated temporal lobe epilepsy patients: a controlled study. Epilepsia 2001;42(11):1416-21. 128. Engel J, Jr. Finally, a randomized, controlled trial of epilepsy surgery. N Engl J Med 2001;345(5):365-7. 129. Engel J, Jr. Update on surgical treatment of the epilepsies. Summary of the Second International Palm Desert Conference on the Surgical Treatment of the Epilepsies (1992). Neurology 1993;43(8):1612-7. 130. Cascino GD. Structural neuroimaging in partial epilepsy. Magnetic resonance imaging. Neurosurg Clin N Am 1995;6(3):455-64. 131. Cascino GD. Clinical correlations with hippocampal atrophy. Magn Reson Imaging 1995;13(8):1133-6. 132. Garcia PA, Laxer KD, Barbaro NM, Dillon WP. Prognostic value of qualitative magnetic resonance imaging hippocampal abnormalities in patients undergoing temporal lobectomy for medically refractory seizures. Epilepsia 1994;35(3):520-4. 133. Cmelak AJ, Abou-Khalil B, Konrad PE, Duggan D, Maciunas RJ. Low-dose stereotactic radiosurgery is inadequate for medically intractable mesial temporal lobe epilepsy: a case report. Seizure 2001;10(6):442-6.
164
134. Srikijvilaikul T, Najm I, Foldvary-Schaefer N, Lineweaver T, Suh JH, Bingaman WE. Failure of gamma knife radiosurgery for mesial temporal lobe epilepsy: report of five cases. Neurosurgery 2004;54(6):1395-402; discussion 1402–4. 135. Nguyen D, Singh S, Zaatreh M, Novotny E, Levy S, Testa F, et al. Hypothalamic hamartomas: seven cases and review of the literature. Epilepsy Behav 2003; 4(3):246-58. 136. Berkovic SF, Arzimanoglou A, Kuzniecky R, Harvey AS, Palmini A, Andermann F. Hypothalamic hamartoma and seizures: a treatable epileptic encephalopathy. Epilepsia 2003;44(7):969-73. 137. Delalande O, Fohlen M. Disconnecting surgical treatment of hypothalamic hamartoma in children and adults with refractory epilepsy and proposal of a new classification. Neurol Med Chir (Tokyo) 2003;43(2):61-8. 138. Maraire JN, Awad IA. Intracranial cavernous malformations: lesion behavior and management strategies. Neurosurgery 1995;37(4):591-605. 139. Porter PJ, Willinsky RA, Harper W, Wallace MC. Cerebral cavernous malformations: natural history and prognosis after clinical deterioration with or without hemorrhage. J Neurosurg 1997;87(2):190-7. 140. Yu JS, Yong WH, Wilson D, Black KL. Glioblastoma induction after radiosurgery for meningioma. Lancet 2000;356(9241):1576-7. 141. Shamisa A, Bance M, Nag S, Tator C, Wong S, Noren G, et al. Glioblastoma multiforme occurring in a patient treated with gamma knife surgery. Case report and review of the literature. J Neurosurg 2001;94 (5):816-21. 142. Kaido T, Hoshida T, Uranishi R, Akita N, Kotani A, Nishi N, et al. Radiosurgery-induced brain tumor. Case report. J Neurosurg 2001;95(4):710-3. 143. Ganz JC. Gamma knife radiosurgery and its possible relationship to malignancy: a review. J Neurosurg 2002;97 (5 Suppl):644-52. 144. Elomaa E. Focal irradiation of the brain: an alternative to temporal lobe resection in intractable focal epilepsy? Med Hypotheses 1980;6(5):501-3.
2775
169 Stimulation of the Hippocampus and the Seizure Focus A. L. Velasco . F. Velasco . M. Velasco . G. Castro . J. D. Carrillo-Ruiz . J. M. Nu´n˜ez D. Trejo
Introduction Neuromodulation has been used as an alternative nonlesional surgical procedure for patients with intractable epilepsy. Several targets for such modulation have been proposed. Cerebellar [1–3], centromedian thalamic [4–6], vagal nerve [7–9], anterior nucleus of the thalamus [10] and subthalamic nucleus [11] stimulation have all shown variable degrees of efficacy in decreasing either primary or secondary generalized seizures. Regardless of the chosen target, neuromodulation has proved to be very well tolerated by patients. If adverse effects occur, there are options in changing stimulation parameters such as decreasing intensity, frequency, or changing stimulated contacts. Another advantage is that there is no deterioration of neurologic functions. On the contrary, improvement in the performance of ability scales of patients with Lennox-Gastaut Syndrome who have undergone chronic stimulation of the thalamic centromedian nuclei have been reported [12]. The above mentioned targets are stimulated with the purpose of influencing seizure propagation or the excitability of large cortical or subcortical areas; they are not directed to the epileptic focus itself. As a result, their effect on partial seizures has been limited. Our experience, as well as other Epilepsy surgery clinics [13,14] is that 70% of referrals for epilepsy surgery are patients with focal onset of epileptic discharges in mesial temporal lobe. Even though ablative surgery in these cases has demonstrated to be #
Springer-Verlag Berlin/Heidelberg 2009
very efficient in reducing or eliminating seizures [15–19], there are a number of patients in whom there is a high risk of neurologic sequelae, for example patients with epileptic focus localized in dominant hippocampus have high risk of memory deficit [20,21] and [22]; which is even higher in resection of bilateral hippocampal foci can produce amnesia [23]. In these patients, a reversible nonlesional method such as neuromodulation is a very attractive option.
Rationale for Using Hippocampal Stimulation Susan Weiss and her group [24] observed that low level direct current inhibits amygdala kindling in rats. The rats were implanted with depth electrodes in the temporal lobe amygdala and were intended to be used for electrical stimulation in a kindling paradigm to produce seizures and afterdischarchges. She applied simultaneous continuous 1–5 mA DC current delivered through the same depth electrode and observed that this stimulation disrupted epileptogenesis. In fact, these rats could not be kindled and instead, they showed an increase in the threshold to induce afterdischarges. This antiepileptic effect was named quenching. We performed initial study based on Weiss’s observations an in ten patients with intractable mesial temporal lobe epilepsy, candidates for temporal lobectomy, in whom diagnostic hippocampal electrodes or basotemporal grids were implanted
2840
169
Stimulation of the hippocampus and the seizure focus
for epileptic foci localization. Once the epileptic focus was detected, patients underwent a 3 week trial of continuous 130 Hz, 300 mA, 450 ms pulse width, electrical stimulation. Thereafter, patients underwent temporal lobectomy. In seven out of ten patients, seizures stopped after 6 days of stimulation and EEG showed a significant reduction of interictal spikes (> Figure 169-1). Our conclusion was that electrical stimulation of epileptic focus arrested epileptic seizures within a period of days [25]. Those patients had no MRI evidence of mesial temporal sclerosis, whereas
those areas that did not improve have. This was not realized in the first report and failure in improvement was attributed to other factors such as failure to reach the precise stimulation target. To have been able to perform a temporal lobectomy after the stimulation trial permitted us to study the epileptic tissue under light microscopy. The surgical specimens of stimulated temporal lobes were compared with specimens from other ones obtained from epileptic patients who underwent diagnostic electrode implantation,
. Figure 169-1 The effect of subacute stimulation on electroencephalogram background activity. Two 10 s samples of maximal paroxysmal EEG activities in records performed on day 1 and day 9 of stimulation. Recordings were made from surface right and left fronto-temporal (FP2, F8, FP1, F7), central (C4, C3) and left subdural anterior (ANT), medial (MED) and posterior (POST) parahippocampal (PHC), fusiform (FUS), and inferior temporal gyri (IT). All EEG recordings were referred to ipsilateral ear lobe electrodes (A2 and A1). On day 1 there was a large number of interictal spikes and slow waves in all subdural recordings, which were more prominent in the anterior parahippocampal gyrus where the seizures initiated. After 9 days of subacute stimulation of the epileptic site, both spikes and slow waves disappeared. In addition a monomorphic delta activity appeared in the medial parahippocampal region (immediately posterior to the stimulated region)
Stimulation of the hippocampus and the seizure focus
but were not stimulated. The pathologist was asked to report differences between the adjacent brain tissue to electrode contacts used for stimulation and those that were not stimulated in the same patient [26]. The histopathology abnormalities consisted in diffuse, moderate gliosis and cell loss of cortical layers I and II, increase in mononuclear inflammatory cells in the subarachnoid space and meningeal thickening of the cerebral tissue attached to the electrode grid. Similar abnormalities were found in depth electrode trajectories, most likely in relation to body reaction to the presence of the silastic sheet of the electrodes. The pathologist was unable to determine which contacts of the grid or area of the trajectory of the hippocampal electrodes had been stimulated. Therefore, seizure reduction in the stimulated patients was related to electrical stimulation and not to microlesions induced by the implanted electrodes. We performed neurophysiologic tests such as producing epileptogenic afterdischarges, as Weiss et al. used [26], recovery cycle tests, as well as SPECT studies to try to explain the mechanisms through which the stimulation produced its antiepileptic effect [25,27,28,29)]. We compared basal conditions with 3 weeks stimulation conditions (previous to lobectomy). Producing hippocampal afterdischarges by using acute local electrical stimulation is a technique used to evaluate the susceptibility of cerebral tissue to present clinical and electroencephalographic epileptic responses. For this purpose, we applied short (10 s) trains of high frequency (130/s) 1.0 ms duration square pulses and increasing intensities every 5 V to the site of the EEG epileptic foci. The threshold (mA) and duration of afterdischarges were measured. > Figure 169-2 shows the afterdischarge obtained in basal condition (previous to subacute hippocampal stimulation) in the upper record. An 88 s afterdischarge consisting of fast-frequency recruiting EEG spikes initiated at the contiguous hippocampal amygdaloidal region, which propagated to the other parasagittal
169
and lateral regions bilaterally, was elicited by acute 8 s stimulation at 560 mA. This afterdischarge was accompanied by a clinical complex partial seizure (epigastric sensation, behavioral arrest, right adversion of the head, left hand exploratory automatisms). The lower record shows the response obtained in the same patient after 560 h of subacute hippocampal stimulation. We had to increase stimulation intensity to 5,300 mA to elicit only a few spikes in the area, with no propagation and no accompanying symptoms. The recovery cycle test (> Figure 169-3) is an electrophysiological technique that evaluates changes in neuronal excitability. It consists of the application of a pair of pulses with identical physical characteristics to evoke a response, applied at different interstimulus time intervals. As shown in > Figure 169-3a, when we stimulated the amygdala and recorded the hippocampus in basal condition, the paired pulses had similar amplitudes when the interstimulus interval between the first (conditioning) and the second (test) response was over 100 ms. As the stimulus interval shortened, the amplitude of the test response decreased (refractory period). When compared to the recovery test obtained from the same patient after sub acute hippocampal stimulation, the test response never reached the amplitude of conditioning response suggesting an inhibitory effect (> Figure 169-3b). Single photon emission computed tomography (SPECT) is used to evaluate the regional cerebral blood flow (rCBF) perfusion as and indirect evidence of hyper or hypo metabolic neuronal dysfunction. > Figure 169-4a shows the basal SPECT in a patient with left temporal epilepsy demonstrating hypoperfusion in the corresponding area. > Figure 169-4b in same patient after sub acute hippocampal stimulation shows a further decrease in (rCBF) in the stimulated hippocampal area. Compared to basal conditions, the results after the hippocampal stimulation showed that afterdischarges were blocked, paired pulses diminished
2841
2842
169
Stimulation of the hippocampus and the seizure focus
. Figure 169-2 Effect of subacute stimulation on afterdischarges in patient KG 111: 2A shows the threshold and duration of the afterdischarge produced by acute stimulation of the right anterior parahippocampal gyrus with 8 s trains of rectangular pulses (130/s and 1.0 ms) before initiating subacute stimulation. Threshold stimulation of 600 mA produced an 80 s afterdischarge initiated in the amygdala PHC region close to the stimulation site (indicated by the star), which spread 10 s later to the right and left fronto-temporal scalp regions. The afterdischarge was accompanied by symptoms of a spontaneous complex partial seizure. 2B shows that after 560 hrs of subacute stimulation, the threshold for the afterdischarge increased to 5,300 mA and its duration decreased from 80 to 8 s, with no clinical symptoms. (Arrows indicate ON and OFF for the 8 s trains to obtain afterdischarges)
their amplitude and SPECT showed decreased rCBF in the stimulated area. Moreover, determination of benzodiazepine receptor binding measured by autoradiography was used to evaluate the activity of GABA system receptors, indicative of neuronal inhibition of the stimulated hippocampal region. There was an increase of hippocampal benzodiazepine receptors density in those patients who had undergone hippocampal stimulation compared to non stimulated tissue, obtained from temporal lobectomies in epileptic patients who had not undergone subacute stimulation [29].
Chronic Hippocampal Stimulation Patient Selection In our experience, 30% of intractable temporal lobe epilepsy patients undergo bilateral hippocampal electrode implantation for diagnostic purposes before performing a temporal lobectomy. From this group of patients we selected nine for chronic electrical stimulation of the hippocampal foci [30]. We settled the criteria for selection according to the following parameters:
Stimulation of the hippocampus and the seizure focus
169
. Figure 169-3 Effect of subacute hippocampal stimulation on recovery cycles of the amygdala hippocampal evoked responses. 3a shows basal condition, note that paired pulses had similar amplitudes when the interstimulus interval between the first (conditioning) and the second (test) response is 100 ms. As the stimulus interval shortened, the amplitude of the test response decreased. 3b shows the recovery test obtained from the same patient after sub acute hippocampal stimulation, the test response never reached the amplitude of conditioning response
. Figure 169-4 SPECT studies of same patient before (a) and after (b) 360 hrs of subacute electrical stimulation. Note that there was a relative hypoperfusion of the left epileptic hippocampus compared to that of the right hippocampus in basal conditions. Arrow shows that the hypoperfusion is more evident in the left hippocampus after 360 hrs of subacute stimulation
2843
2844
169 1.
2.
3.
Stimulation of the hippocampus and the seizure focus
Four patients with bilateral hippocampal epileptic foci which were confirmed with depth EEG recording of several seizures arising independently from both hippocampuses. These patients would have been rejected for bilateral ablative surgery because of the resulting severe amnesia [22]. Epileptic focus localized on dominant hippocampus. It has been established by several authors that temporal lobectomies affect verbal memory when ablative surgery is performed in dominant hemispheres [21,22], so as a result, either they are rejected as surgical candidates or limited resections are performed with the consequent seizure persistence and elevated risk of memory loss. Three patients with left hippocampal foci with neuropsychological tests showing left dominance were included. Two patients with right hippocampal foci were also included, one of them because she had bilateral hippocampal sclerosis in MRI imaging and the other because there was independent interictal activity in the left hippocampus.
The group consisted of 6 males and 3 females, ages ranging from 14 to 43 years, with seizure onset between 3 and 16 years of age, seizure number varied from 15 to 70 seizures per month (average 28). The study protocol for epilepsy surgery was followed [5]. All of them had complex partial seizures; seven had secondary tonic clonic generalized seizures. All were right handed. According to the neuropsychological tests battery that was applied (> Table 169-1) [31–35] and [36], six patients had mild memory impairment in neuropsychological tests and three of them had severe abnormalities. Serial EEGs were abnormal with bitemporal paroxystic epileptic activity and secondary bilateral synchrony. The magnetic resonance studies were normal in five patients, three had left hippocampal
. Table 169-1 Neuropsychological tests battery used to evaluate patients. Notice that special interest was centered in memory and language dominance. All tests used were standardized for Spanish speaking patients Test
Function
Dichotic listening test Neuropsi Attention and Memory Battery Rey verbal learning Digit Counting Logic memory Visual reproduction Wind Mill visual spatial Bezarez Test
Language dominance
Verbal memory Verbal memory Verbal memory Non verbal memory Non verbal memory
sclerosis and one had bilateral hippocampal sclerosis. All patients had undergone several trials of antiepileptic drugs without obtaining seizure control and were willing to participate and sign the informed consents. The study was reviewed and approved by the Scientific and Ethical Committee of the General Hospital of Mexico. The Committee agreed on an aleatory (randomized by lottery number) double-blind maneuver with an initial 1 month OFF period in one half of the subjects; the other half initiated stimulation immediately after stimulation system internalization.
Surgical Procedure The selected patients had a 3 month seizure baseline for reliable seizure calendars and afterwards were hospitalized to undergo diagnostic depth electrode implantation. The surgical procedure was performed in two stages. In the first stage an eight contact, spaced 0.7 mm from center to center, were implanted from an occipital approach. The stereotactic frame was fixed under general anesthesia and the patient was placed in ventral decubitus to have a double contrast enhanced CT scan. Two and a half mm CT axial sections were
Stimulation of the hippocampus and the seizure focus
taken to be fused with preoperative MRI sections taken the day before surgery. On the fused image, virtual trajectories that traversed the entire hippocampus, avoiding blood vessels and ending in the basal part of the temporal lobe amygdala were planned. Those trajectories had a lateromedial angle of 10–15 , starting 26.0–30.0 mm in the skull entrance and ending 22.0–25.0 mm lateral to the midline. It is important to mention that the estimated center of the planned burr holes averages 3.0 cm lateral to the midline. Therefore, about 10.0 cm distance between the pins that will fix the posterior part of the skull is required. This has to be taken into consideration when placing the stereotactic frame. An occipital horseshoe shape incision is performed, leaving a distance of 2 cm between the estimated edge of the 14 mm diameter burr holes, the electrodes fixation device and the edge of the incision. This is to avoid skin erosions over the implanted electrodes. Diagnostic electrodes are left externalized for EEG localization of the epileptic foci. A post operative MRI will confirm the exact position of each contact of the electrodes (> Figure 169-5a).
169
Once the electrode position was verified, antiepileptic drugs were tapered and continuous recording for seizure detection was performed. After localizing the epileptic focus, we performed the neurophysiologic testing. This testing consisted in stimulating the pair of selected contacts for chronic stimulation at low (6 Hz) and high (60 Hz) frequencies to elicit electrocortical responses. This test had two purposes, first, to set the amplitude parameter for long term stimulation (50% of the amplitude needed to obtain electrocortical responses) and to record scalp responses in the ipsilateral temporal leads to monitor long-term stimulation (see below). Antiepileptic medication was reinitiated and the patients underwent the second surgical stage. The stereotactic frame was replaced under general anesthesia and CT scan was repeated. This CT study was fused with the post operative MRI with the diagnostic electrodes in place. Four contact electrodes (for deep brain stimulation) with a 3 mm distance from center to center of adjacent contacts were implanted for long term therapeutic stimulation. They were directed to the previously identified epileptic focus with at
. Figure 169-5 MRI axial images with bilateral diagnostic 8 contact hippocampal electrodes position (5a) and final position of the permanent therapeutic 4 contact electrode (5b) placed where the epileptic focus was localized. Contacts 2 and 3 are currently being stimulated
2845
2846
169
Stimulation of the hippocampus and the seizure focus
least two contacts within the epileptic site (> Figure 169-5b). Afterwards the pulse generator was implanted and connected to the depth stimulation electrode.
Double Blind Maneuver Five patients had an initial 1 month OFF according to the aleatory selection and four patients initiated stimulation immediately. Patients and medical personnel that collected seizure calendars were unaware whether pulse generator was ON or OFF as previously explained. Bipolar stimulation was performed choosing the pair of contiguous contacts which covered the area where the epileptic focus was localized. In case of bilateral foci, bilateral hippocampal stimulation was used. The parameters for chronic stimulation of the hippocampus were the following: Cyclic stimulation: 1 min trains of square pulses with 4 min interstimulus interval Charge density adjusted to 2–4 mC/sq cm/ phase High frequency: 130 Hz Pulse width: 450 ms Amplitude of 300 mA which equals 50% of the amplitude needed to obtain electrocortical responses. According to the formula referred by Velasco et al. [3], the charge density for these parameters is <3.0 mC/cm2/phase. In patients with bilateral foci, parameters were the same but alternating 1 min stimulation on one side with a 4 min interval between right and left sides. The stimulation parameters used are within a safe nonlesional range [3,37].
Follow Up Protocol Follow up in all patients was at least 18 months (18–84 months) after initiating hippocampal
stimulation. Seizure calendars were collected once a month; EEG and neuropsychological tests were performed on months 6, 12 and 18 after stimulation onset. Neurophysiologic tests to assess the viability of the stimulated tissue to electrical stimuli were carried out every 6 months. The internalized pulse generators were programmed with a transcutaneous computer to stimulate through the selected contacts for therapeutic stimulation using 8 Hz pulses, 6–8 V 450 ms to induce electrocortical responses, EEG recording referred to ipsilateral ear; responses are seen in the ipsilateral temporal leads.
Results The postsurgical control MRIs showed no evidence of hemorrhage or edema, all patients tolerated the surgical procedure well. None of the patients had side effects with the stimulation parameters employed, as a matter of fact; they were unaware of device activation which permitted us to have a double blind ON OFF protocol. Impact on seizure number: > Figure 169-6 shows the seizure graphs of two patients as examples of the response to stimulation. Significance of seizure reduction was determined by Student’s T test. Although all patients improved, there were two types of responses. One of them is shown in > Figure 169-6a. The graph shows the seizure number per month (individual bars), the first three bars show the baseline seizure count followed by the OFF bar which indicates the first month of the double blind protocol during which this particular patient had the stimulator OFF. The following 18 bars show the seizure counts during 18 months follow-up. Observe the immediate decrease in seizures as soon as stimulation was initiated (ON). It is also outstanding how the patient was seizure free after 2 months of stimulation and remained so during the whole follow-up. A total of five
Stimulation of the hippocampus and the seizure focus
169
. Figure 169-6 Seizure counts per month of patients KG111 (a) and patient KG 112 (b). First three bars correspond to the 3 months basal condition. OFF indicates the initial double blind month with implanted electrodes but pulse generator turned off; ON indicates when chronic therapeutic stimulation is initiated. Note that both patients had a seizure decrease but there is a difference in their response. ‘‘A’’ had no hippocampal sclerosis in her MRI and had an immediate seizure decrease. She remained seizure free from month 3 on. ‘‘B’’ had a slower response which was not significant till month 8 of therapeutic stimulation. He remained with a 60% seizure reduction throughout follow-up
patients had this type of response, four of them remained seizure free and one had occasional brief complex partial seizures. The seizure reduction for these patients beyond month 3 was highly significant (p < 0.0005). > Figure 169-6b shows another type of response. Note that the difference in the number of seizures per month between 3 months baseline, 1 month OFF stimulation, the stimulation onset and first 8 months follow-up is not significant.
After this time, there is a seizure reduction but not as significant as the previous group of patients (p < 0.05), this reduction was maintained, but none of the patients became seizure free. A retrospective review of all patients showed that the difference between the two groups of patients was that those who had a better and faster response, had normal MRIs and those who had a modest response, had evidence
2847
2848
169
Stimulation of the hippocampus and the seizure focus
of hippocampal sclerosis shown in the MRI studies. Stimulation parameters in both groups were similar. Conceivably, sclerotic tissue may have higher impedance and therefore, those patients with mesial temporal sclerosis should have been stimulated with higher density charge. Or maybe, the smaller response is due to the disruption of the normal histology of the tissue. ON-OFF protocol: According to the aleatory double-blind maneuver with an initial 1 month OFF period that was authorized, five patients underwent an OFF period after deep brain stimulation system implantation. The other four patients initiated stimulation immediately after neurostimulator implantation. This study design was based in the preliminary subacute studies where we had observed a significant seizure reduction after day 6 of therapeutic stimulation. As already mentioned, patients who responded to subacute stimulation in the study had no evidence of hippocampal sclerosis by MRI. In contrast, the double blind protocol included both patients with normal MRI and with evidence of mesial temporal sclerosis, so we did not realize that there would be a retarded response in patients with mesial temporal sclerosis. For this reason, the results of double blind protocol are difficult to analyze. We can say that all patients who underwent an initial OFF period, showed either no changes in seizure number when compared with baseline in four patients and seizure increase in one patient. From four patients who went ON electrical stimulation immediately, three had a modest non significant seizure decrease (two of them had mesial temporal sclerosis); one patient had a 90% seizure decrease. Regardless of having or not mesial temporal lobe sclerosis in MRI, all patients had a significant lower seizure count by the end of the 18 months follow-up compared with baseline. Neuropsychological impact: Our main concern and reason for including these patients was the high risk of memory deficit with temporal
lobectomy due to either bilateral hippocampal foci or left side (dominant hemisphere) localization. All patients had some degree of memory loss in the baseline stage, probably due to the long seizure history and poor medical treatment response. When stimulation started, no patient had a memory decline and after 18 months stimulation, and there was a trend to improve in both verbal and non verbal memory evaluations. The small number of patients studied does not permit statistical analysis. After the initial publication in 2000 of our first report on hippocampal stimulation, a series of studies using hippocampal stimulation for the control of mesial temporal lobe epilepsy [38,39] and [40] have been published. Results have varied for a number of reasons: considering evidence of hippocampal sclerosis in MRI an inclusion criteria, when we have described above that these patients do not have the same seizure reduction than patients with no sclerosis. Skipping the first surgery where diagnostic eight contact electrodes are implanted and thus possibly missing the exact location of the hippocampal focus could also explain different results. Regardless of the differences in seizure reduction, all studies agree that stimulation of the hippocampus is a safe method with no evidence of tissue damage, all studies use stimulation parameters within a safe nonlesional range, and more importantly, all authors agree that there is no deterioration in memory function. We can conclude that stimulation of the hippocampus is a safe, nonlesional alternative for patients with complex partial seizures, with or without secondary generalization who are not candidates for resective surgery. Other inclusion criteria might be considered, for example, patients with previous temporal lobectomy who develop or have residual contralateral intractable seizures. More studies have to be performed to clarify a number of questions. What if we use other stimulation parameters for the sclerotic hippocampus? What happens if, instead of
Stimulation of the hippocampus and the seizure focus
stimulating a hippocampal sclerotic tissue, we stimulate the parahippocampus to avoid seizure propagation? We should also conduct multicenter studies to validate the neuropsychological findings. Even though patients with mesial temporal lobe epilepsy are the most frequently referred ones for surgery, could other types of partial seizures be treated with stimulation? We will address the last question in the next part of this chapter.
Stimulation of the Motor Cortex for the Treatment of Supplementary and Primary Motor Cortex Seizures Ablative surgery of epileptic foci located in the supplementary motor or the primary motor cortices has been performed in several epilepsy surgery centers [41–44,45]. Though results vary within each center, the outcome in seizure reduction varies from 65 to 100%. Most of the cases are patients who have lesions such as cortical dysplasia, cavernomas and gliosis; very few non lesional cases are included. The main problem with these surgeries is that there are a number of neurological sequelae, i.e., paralysis, paresis, apraxia, aphasia and mutism. There were also complications due to the surgical procedure itself. No wonder the epileptologists have a great concern when they have to operate these patients. If the patient has no evidence of a lesion in the MRI, this concern is even greater. With all this considered, we decided to evaluate the possible anticonvulsive effect of stimulating the epileptic foci located in the motor area in two patients with non lesional intractable epilepsy, one of them in the supplementary motor area and the other in the primary motor area. Both had severe seizures despite multiple antiepileptic drugs and were candidates for intracranial grids for foci detection. The two patients were studied following the Epilepsy Surgery
169
Protocol of the General Hospital of Me´xico. The study was reviewed and approved by the Scientific and Ethical Committee of the General Hospital of Mexico. Patients were willing to participate and signed the informed consents. The selected patients had a 3 month seizure baseline to collect reliable seizure calendars. Neuropsychological evaluation was performed and QOL scales were applied. Afterwards patients were hospitalized to undergo diagnostic depth electrode implantation. Two diagnostic 20 contact grids were implanted through frontal craniotomies. Patient I: 17 year old male with refractory supplementary motor seizures. Perseverance and verbal aggressiveness were present. Surface EEG showed frontal parasagittal epileptic activity. MRI was normal. Bilateral 20 contact grids were implanted in right and left SMA (> Figure 169-7a). Ictal depth EEG showed spontaneous seizure onset located at contacts 3, 2 and 9 of right grid. Patient II: 24 year old female with left primary motor seizures and progressive loss of motility of the left side of the body and face. Surface EEG showed slowing in right frontal area, MRI was normal. Two 20 contact grids were implanted in the upper and lower right motor cortex. Ictal EEG showed seizure starting in contact 10 of the upper grid (> Figure 169-8b). Daily depth recording was performed without AEDs and ictal EEG activity was obtained. Once the epileptic focus was detected, patients reinitiated AEDs. Grids were explanted and replaced by a four contact, 1cm diameter, plate electrode localized over the epileptic focus (> Figure 169-7b) for chronic stimulation. The position of the electrodes was fixed by suturing them to the dura matter using nylon stitches in each end of the electrode. In the case of primary motor cortex, electrode was fixed to the dura in the convexity; in the case of supplementary motor cortex focus, the electrode was fixed to the cerebral falx. Thereafter, electrode was connected to a DBS system. Stimulation was started with
2849
2850
169
Stimulation of the hippocampus and the seizure focus
. Figure 169-7 Diagram of diagnostic 20 contact grid on right interhemispheric cortex (7a) and sagittal MRI showing definite 4 contact electrode position. Contacts 2 and 3 where epileptic focus was located are currently being stimulated
. Figure 169-8 Graphs seizure reduction for patient 1 who had supplementary motor area seizures (a) and for patient 2 who had primary motor area seizures (b). Both had an immediate seizure decrease; patient I became seizure free from month 8 on and patient 2 became seizure free from month 7 on
Stimulation of the hippocampus and the seizure focus
the following parameters: bipolar continuous stimulation between contacts that covered the epileptogenic zone, 130 Hz, 350 mA and 450 ms. Both patients had a seizure reduction which was observed from the beginning of the stimulation of the epileptic focus. Patient I became seizure free from month 8 on and has remained so for 14 months and patient II from month 7 with no seizures for 3 months. Motor function was preserved, no adverse effects were observed and patients are unaware of stimulation. QOL scales improved in both patients. We present two cases where results are dramatic and encourage us to continue attempting to stimulate the motor cortex too. We are sure we will face other challenges, as for example, having a huge epileptic area with multiple foci for which there are no grid type electrodes available for cortical stimulation. Would we need several electrodes, could they be connected to a single pulse generator, would we need a specially designed electrode? All these questions point to the need of a near collaboration between clinicians and biomedical engineers to design better hardware for stimulation that can be well tolerated by the patient and avoid skin erosions or other problems such as lead breakage [30,46] Another goal to achieve is to decrease costs so more patients have the opportunity to try this alternative surgical method. Many questions and challenges arise, and to be able to answer them, a multidisciplinary group is mandatory with collaboration of basic scientists, neuropsychologists, epileptologists and neurosurgeons.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
References 14. 1. Cooper IS, Amin I, Ricklan M. Chronic cerebellar stimulation in epilepsy: clinical and anatomical studies. Arch Neurol 1976;33:559-70. 2. Davis R, Emmans SE. Cerebellar stimulation for seizure control 17 year study. Stereotact Funct Neurosurg 1992;58:200-8. 3. Velasco F, Carrillo-Ruiz JD, Brito F, Velasco M, Velasco AL, Ma´rquez I, Davis R. Double-blind randomized
15.
169
controlled pilot study of bilateral cerebellar stimulation for treatment of intractable motor seizures. Epilepsia 2005;46:1-11. Velasco F, Velasco M, Velasco AL. Effect of chronic electrical stimulation of the centromedian thalamic nuclei on various intractable seizure patterns. I. Clinical seizures and paroxysmal EEG activity. Epilepsia 1993;34:1052-64. Velasco F, Velasco M, Jime´nez F. Predictors in the treatment of difficult to control seizures by electrical stimulation of the centromedian thalamic nucleus. Neurosurgery 2000;47:295-305. Velasco M, Velasco F, Velasco AL. Acute and chronic electrical stimulation of the centromedian thalamic nucleus: modulation of reticulo-cortical systems and predictor factors for generalized seizure control. Arch Med Res 2000;31:304-15. Morris G, Mueller W. Long-term treatment with vagus nerve stimulation in patients with refractory epilepsy. Neurology 1999;53:1731-5. Frost M, Gates J, Helmers SL. Vagus nerve stimulation in children with refractory seizures associated with Lennox– Gastaut syndrome. Epilepsia 2001;42:1148-52. Amar AP, Apuzzo MLJ, Liu CY. Vagus nerve stimulation therapy alters failed cranial surgery for intractable epilepsy: results from the vagus nerve stimulation therapy patient outcome registry. Neurosurgery 2004;55: 1086-93. Kerrigan JF, Litt B, Fisher RS, Cranstoun S, French JA, Blum DE, Dichter M, Shetter A, Baltuch G, Jaggi J, Krone S, Brodie M, Rise M, Graves N. Electrical stimulation of the anterior nucleus of the thalamus for the treatment of intractable epilepsy. Epilepsia 2004; 45:346-54. VesperJ, Steinhoff B, Rona S, Wille C, Bilic S, Nikkhah G, Ostertag C. Chronic high frequency deep brain stimulation of the STN/SNr for progressive myoclonic epilepsy. Epilepsia 2007;48:1984-9. Velasco AL, Velasco F, Jime´nez F, Velasco M, Castro G, Carrillo J-D, Fangha¨nel G, Boleaga B. Neuromodulation of the centromedian thalamic nuclei in the treatment of generalized seizures and the improvement of the quality of life in patients with Lennox–Gastaut syndrome. Epilepsia 2006;47: 1203-12. Wieser HG, Engel J Jr, Williamson PD, et al. Surgically remediable temporal lobe syndromes. In: Engel J Jr, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1993. p. 49-63. Williamson PD, Wiesser HG, Delgado Escueta, AV. Clinical characteristics of partial seizures. In: Engel J Jr, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1993. p. 387-97. Velasco AL, Boleaga B, Brito F, Jime´nez F, Gordillo JL, Velasco F, Velasco M. Absolute and relative predictor values of some non-invasive and invasive studies for the outcome of anterior temporal lobectomy. Arch Med Res 2000;31:62-74.
2851
2852
169
Stimulation of the hippocampus and the seizure focus
16. Primrose DC, Ojeman GA. Outcome of resective surgery for temporal lobe epilepsy. In: Lu¨ders H, editor. Epilepsy surgery. New York: Raven Press; 1961. p. 601-18. 17. Cahan LD, Sutherling W, McCullough MA. Review of the 20-year UCLA experience with surgery of epilepsy. Cleve Clin J Med 1984;51:313-23. 18. Engel J Jr. Outcome with respect to epileptic seizures. In: Engel J Jr, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1987. p. 553-69. 19. Radhakrishnan K, So EL, Silbert PL, Jack CR Jr, Cascino G, Sharborough FW, O’Brien PC. Predictors of outcome of anterior temporal lobectomy for intractable epilepsy. Neurology 1998;51:465-71. 20. Helmstaedter C, Kurthen M, Lux S, Reuber M, Elger CE. Chronic epilepsy and cognition: a longitudinal study in temporal lobe epilepsy. Ann Neurol 2003;54:425-32. 21. Kapur N, Prevett M. Unexpected amnesia: are there lessons to be learned from cases of amnesia following unilateral temporal lobe surgery? Brain 2003;126 (Pt 12):2573-85. 22. Trenerry MR, Jack CR Jr., Ivnik RJ, Sharbrough FW, Cascino GD, Hirschorn KA, Marsh WR, Kelly PJ, Meyer FB. MRI hippocampal volumes and memory function before and after temporal lobectomy. Neurology 1993;43:1800-5. 23. Scoville WB and Milner B. Loss of recent memory after bilateral hippocampal Lesions. J. Neurol 1957;20:11-21. 24. Weiss SR, Eidsath A, Li XL, Heynen T, Post RM. Quenching revisited: low level direct current inhibits amı´gdala kindling. Exp Neurol 1898;154:185-92. 25. Velasco AL, Velasco M, Velasco F, Me´nes D, Gordon F, Rocha L, Briones M, Ma´rquez I. Subacute and chronic electrical stimulation of the hippocampus on intractable temporal lobe seizures. Arch Med Res 2000;31:316-28. 26. Velasco M, Velasco F, VELASCO AL, Boleaga B, Jime´nez F, Brito F, Ma´rquez I. Subacute electrical stimulation of the hippocampus blocks intractable temporal lobe seizures and paroxysmal EEG activities. Epilepsia 2000; 41:158-69. 27. Velasco F, Velasco M, Velasco AL, Me´nes D, Rocha L. Electrical stimulation for epilepsy 1. Stimulation of hippocampal foci. Stereotact Funct Neurosurg 2001;77:223-7. 28. Velasco M, Velasco F, Velasco AL. Centromedian thalamic and hippocampal electrical stimulation for the control of intractable epileptic seizures. Clin Neurophys 2001; 18:1-15. 29. Cuellar-Herrera M, Velasco M, Velasco F, Velasco AL, Jime´nez F, Orozco S, Briones M, Rocha L. Evaluation of GABA system and cell damage in parahippocampus of patients with temporal lobe epilepsy showing antiepileptic effects alter subacute electrical stimulation. Epilepsia 2004;45:459-66. 30. Velasco AL, Velasco F, Velasco M, Trejo D, Castro G, Carrillo-Ruiz JD. Electrical stimulation of the hippocampal epileptic foci for seizure control: a double blind, long-term follow-up study. Epilepsia 2007;48:1895-903.
31. Kimura D. Cerebral dominance and the perception of verbal stimuli. Can J Neuropsych 1961;15:166-71. 32. Voyer D. Reliability and magnitude of perceptual asymmetries in a dichotic word recognition task. Neuropsychology 2003;17:393-401. 33. Azan˜o´n-Gracia E, Sebastia´n-Gale´s N. Test de escucha dico´tica en espan˜ol: pares de palabras bisila´bicas. Rev Neurol 2005;41:657-63. 34. Rey A. Solicitation de la memoire de fixation par des mots et des objects presentes simultanement. Arch Psychol 1959;37:126-37. 35. Deutsch Lezac M. Memory I: tests in neurophsychological assessment. 3rd ed. New York: Oxford University Press; 1995. p. 429-98. 36. Ostrosky-Solı´s F, Ardila A, Roselli M. Neuropsi a brief neuropsychological test battery in Spanish with norms by age and educational level. J Int Neuropsychol Soc 1999;5:413-33. 37. McCreery DB, Agnew WF, Yuen TG, Bullara L. Charge density and charge per phase as cofactors in neural injury induced by electrical stimulation. IEEE Trans Biomed Eng 1990;37:996-1001. 38. Vonck K, Boon P, Achten E, De Reuck J, Caemaert J. Long term amygdalohippocampal stimulation for refractory temporal lobe epilepsy. Ann Neurol 2002; 52:556-65. 39. Tellez-Zenteno JF, McLachlan RS, Parrent A, Kubu CS, Wiebe S. Hippocampal electrical stimulation in mesial temporal lobe epilepsy. Neurology 2006;66:1-5. 40. Boon P, Vonck K, De Herdt V, Van Dycke A, Goethals M, Goosens L, Van Zandijcke M, De Smedt T, Dewaele I, Achten R, Wadman W, Dewaele F, Caemaert J, Van Roost D. Deep brain stimulation in patients with refractory temporal lobe epilepsy. Epilepsia 2007;48: 1551-60. 41. Mihara T, Tottori T, Inoue Y, Seino M. Surgical strategies for patients with supplementary sensorimotor area epilepsy. The Japanese experience. Adv Neurol 1996; 70:405-14. 42. Smith JR and King DW. Surgical strategies for patients with supplementary sensorimotor area epilepsy. The Medical College of Georgia experience. Adv Neurol 1996;70:415-427. 43. Engel J, Van Ness PC, Rasmussen TB, Ojeman LM. Outcome with respect to epileptic seizures. In: Jerome Engel J, editor. Surgical treatment of the epilepsies. 2nd ed. New York: Raven Press; 1993. p. 609-21. 44. Olivier A. Surgical strategies for patients with supplementary sensorimotor area epilepsy. The Montreal experience. Adv Neurol 1996;70:429-43. 45. Spencer DD, Schumacher J. Surgical strategies for patients with supplementary sensorimotor area epilepsy. The Yale experience. Adv Neurol 1996;70:415-27. 46. Oh MY, Abosch A, Kim SH. Long-term hardware-related complications of deep brain stimulation. Neurosurgery 2002;50:1268-74.
160 Selective AmygdaloHypocampectomy T. A. Valiante
Historical Aspects The foundations for diagnosis and management of epilepsy were laid by two fundamental advances in neurophysiology: cerebral localization, and electroencephalography (EEG). Of equal importance in the context of surgical interventions for epilepsy were the development of anesthesia, antisepsis, surgical technique, and surgical instrumentation. In regards to cerebral localization, it was the clinical observations of Broca [1], and Jackson [2] and the landmark observations of Fritsch and Hitzig [3], that firmly established not only the electrical excitability of the human brain, but the discrete localization of brain functions. Through experiments of electrical stimulation carried out on narcotized dogs Fritsch and Hitzig concluded that ‘‘One part of the convexity of the cerebrum of the dog is motor another part is not motor’’ [3]. Three years later Ferrier extended these finding through electrical stimulation mapping, identifying 15 different centers in the monkey cortex [4], and was later credited by Sherrington as the main figure to prove the concept of cerebral localization [5]. Ferrier however credited Fritsch and Hitzig in this regard stating that ‘‘The whole aspect of cerebral physiology and pathology was revolutionized by the discovery, first made by Fritsch and Hitzig in 1870’’ [6]. Ferrier’s deduction that this functional localization could be used for the diagnosis and treatment of brain disorders was not overlooked by Bennet and Godlee [7]. Utilizing Ferrier’s localizations, they correctly direct the resection of a tumor localized to the ‘‘middle part of the fissure of Rolando’’ [7]. Their widely publicized work caught the #
Springer-Verlag Berlin/Heidelberg 2009
attention of Sir Victor Horsley arguably the father of modern day neurosurgery [8]. Horsley among many of firsts to his credit [8,9], likely performed the first craniotomy specifically for the amelioration of epilepsy in a patient suffering from post-traumatic seizures [10]. Developments in the recording of electrical activity from brain tissue paralleled brain stimulation experiments mentioned above, albeit somewhat in a delayed fashion. In 1875 Richard Caton demonstrated that electrical activity could be measured from the cat brain [11], and Berger reproduced these results in humans publishing the first paper on human electroencephalography (EEG) in 1929 [12]. Frederic and Erna Gibbs were the first to begin classifying epilepsy syndromes both clinically and electrophysiologically, and in 1938 described three different patterns that distinguished what then were referred to as grand mal, petit mal, and the psychomotor epilepsies [13]. Around this time, Jasper and Kershmann made a fundamental observation that the EEG manifestations of the epilepsies could be classified according to location and not by the pattern of the paroxysmal discharge. Through the use of bipolar chain montages they were able to reliably recognize the temporal lobe (TL) origin of many epileptic discharges [14]. Gibbs utilized this information and was able to convince Percival Bailey a contemporary of Wilder Penfield, to operate solely on EEG findings in those individuals with psychomotor epilepsy with anterior temporal spikes on the EEG [15]. The first reported series of surgical interventions for temporal lobe epilepsy (TLE) was published by Wilder Penfield, in which he discussed
2678
160
Selective amygdalo-hypocampectomy
seizure outcome in the context of extent of TL resection [16]. Penfield further observed that an individual that failed surgery involving resection of limited amounts of temporal neocortex, could benefit from resection of the medial TL structures, namely the hippocampus and the amygdale [17,18]. This was largely driven by Penfield’s extensive use of electrical stimulation mapping that he had learned from Foerster and the ever increasing EEG evidence implicating the mesial TL structures such as the hippocampus and amygdala in the genesis of TL seizures. Thus was emerging the concept of mesial TLE (mTLE) as a distinct clinical pathological entity as a result of: advancements in scalp EEG recordings and interpretation; Penfield’s observations on clinical outcomes, intra-operative brain stimulation, and electrocorticography (ECoG); and, histopathological descriptions of tissues removed during TL surgery [19,20]. Indeed Murray Falconer, contributed much to understanding the pathological correlates, specifically mesial temporal sclerosis (MTS) within the clinical syndrome of mTLE through the introduction of the en bloc temporal lobectomy and the unadulterated pathological specimens that this surgical approach afforded [20]. Penfield being well attuned to the potential functional ramifications of surgery on the human brain and specifically memory, solicited help from one of the student’s of Donald Hebb, a psychologist, to test his surgical patients. Penfield’s description of memory deficits in some of his post-operative patient’s [21,22] peaked the interest of William Scoville, another neurosurgeon who had observed severe memory deficits in a patient H.M. [23] in whom he had removed both hippocampi. Brenda Milner, a student of Hebb was dispatched to examine H.M amongst others [24], and went on to not only contribute significantly to the understand of the various types of memory processes, but as well as to establish the fundamental role of the hippocampus in these processes.
Many new developments have occurred since these early days, but the contributions mentioned above remain the fundamental building blocks of the modern practice of epilepsy surgery. The multidisciplinary approach involving neurology, neurosurgery, and neuropscyhology, emphasizes the complexity of treating functional brain disorders, and the significant impact that epilepsy can have on the individual both physiologically and psychologically.
Introduction Epidemiology The detailed epidemiology of epilepsy will not be covered in this chapter as it has been done elsewhere. However as they relate specifically to epilepsy surgery some numbers are certainly in order. Epilepsy, according to the World Health Organization is the most serious neurological disorder in the world [25] affecting up to 1% of the world population, and approximately 0.6% of those in developed countries [26]. It is generally accepted that up to 70% of those with newly diagnosed epilepsy will find remission with medical management alone [27]. The remaining 30% will be refractory to medications, and up to half of these individuals will benefit from either a diagnostic or therapeutic surgical intervention [28]. Worldwide, despite the staggering number of individuals that could benefit from some form of surgical intervention, surgery for epilepsy remains significantly underutilized [29]. In the United States, it has been estimated that although there are over 200,000 individuals that could benefit from epilepsy surgery, only 1,500 epilepsy operations are performed each year [30]. In Canada it has been estimated that 352 individuals out of approximately 20,000 individuals with surgically amenable epilepsy received surgery [31]. In this author’s locale the Ministry of Health has estimated that in Ontario (population
Selective amygdalo-hypocampectomy
of 12 million) of the 10,000 individuals that could potentially benefit from some form of surgical intervention less than 2% of these individuals obtain surgery [28].
When to Assess for Surgery Any individual who has a focal seizures disorder [32], who continues to have seizures after two first line anti-seizure medications (ASM) have been administered sequentially (not together) should be referred for a surgical evaluation [32–35]. This basic rule however, should be applied to all individuals with epilepsy, whether from a semiological perspective their seizures appear to be either focal or generalized. This medication ‘‘trial’’ may take anywhere from 3 months to 2 years, and it must be emphasized that continued seizures are not a ‘‘necessary evil’’ [36], and should not be tolerated even if one per year [36]. It is well documented that focal seizures especially those of the complex partial variety [32] (CPS) are more likely to be resistant to ASMs than are generalized seizures [37]. Furthermore, the TL is the most common site of generation of CPS [38]. Therefore if an individual suffering from CPS has evidence of a TL structural lesion or mesial temporal sclerosis (MTS), then it is quite likely that they will fail their medication trial [39]. Although the distinction of a generalized seizure from a CPS may be obvious from an academic perspective [32], for those who are not epileptologists (including neurosurgeons) in practice the distinction can be anything but obvious [40]. It is on this premise that it has been suggested that individuals who are suspected to have epilepsy should be referred to an epileptologist for diagnosis and treatment of their epilepsy [33,34]. Although this might seem as an affront to the generalist (neurologist, family practitioner) this should be understood as a collaborative effort with the initial diagnosis and management plan
160
developed in consultation with an epileptologist, and subsequent management fundamentally involving surveillance by the individuals neurologist or their family practitioner [41,42]. As with other chronic conditions like diabetes and heart disease, we are increasingly able to forecast the pre-operative and post-operative path an individual will take with their epilepsy given that accurate seizure classification and neuroimaging has been obtained [39,43]. TLE should be viewed as a surgically remediable form of epilepsy at its onset [44], and thus any individual with CPS who has failed medical management should be referred for a surgical evaluation.
Goals of Surgery In the context of TLE, the primary goal of surgical intervention is to effect cure, and all interventions should be assessed according to this standard. Although attaining seizure freedom for those with epilepsy would appear to be an intuitive concept, it is borne out by evidence than suggests the greatest benefit to an individual with epilepsy is through complete cessation of seizure activity [45]. Complete cessation of seizures results in: improved psychosocial adjustment [46], reduction in AEDs taken [47], increased employment and educational status [48], marked improvements in QOL [47,49], reduced mortality rates as epilepsy is associated with increased mortality among its sufferers [50], increases in full scale IQ (FIQ) [51], and significant cost saving to society [52]. Furthermore, although greater and more permanent benefits are obtained the earlier one obtains surgery in their disease course [53–56], it usually takes decades for an individual to obtain surgery [57–59]. Thus as mentioned above, and it can not be overemphasized, TLE should be viewed as a surgically remediable form of epilepsy from its outset. From an intention to treat perspective for all those with suspected TLE, Wiebe et al. have
2679
2680
160
Selective amygdalo-hypocampectomy
shown that surgery can obtain an almost 60% chance of complete seizure remission [60] for those with TLE. Futhermore, this rate of remission is likely higher in an individual who has undergone their pre-surgical assessment and is found to have MTS, the most common pathological finding in TLE [61,62]. Weibe’s study highlights the need to recognize the predictable outcome of patients with TLE and confirms decades of research attesting to the efficacy of temporal lobectomy [38]. Indeed the study by Weibe et al. [63] comparing temporal lobectomy to medical management, concluded that the number needed to treat to effect complete seizure remission at 1 year was two [60]! Unfortunately complete seizure freedom can not always be obtained with surgery. Although as mentioned above the ultimate goal in those with TLE is complete remission, reduction in seizure frequency does benefit the individual to some extent in the spheres of assessment listed above, albeit paling in comparison to the benefits of seizure cessation [45].
Subtypes of TLE From a electrical and clinical perspective as well as one that adheres to the anatomy of the TL, two primary types of epilepsy ‘‘syndromes’’ can be distinguished: one being mesial TLE (mTLE), and the other that of neocortical (lateral temporal [32)] TLE (nTLE) [38]. This electroclinical classification follows from the fact that these two entities (although there is indeed overlap) can be distinguished by the pathologies that give rise to the seizures, their electrophysiological differences on EEG and electrocorticography (ECoG), different neuropsychological findings, and how amenable they are to surgery [38]. Although these two syndromes differ in many ways they may semiologically look quite similar, as there can be rapid spread of abnormal neuronal synchronization from neocortical generators to limbic structures
within the medial TL [64–67]. The two syndromes are compared and contrasted in > Table 160-1. Other categorizations and thus terminologies applied to TLE separate the temporal lobe epilepsies along pathological, and the absence of imaging abnormalities. From a pathological perspective TLE secondary to MTS has unfortunately and rather confusingly fallen under the rubric of non-lesional TLE along side non-specific gliosis and truly negative histopathology. It thus follows from this that TLE due to various foreign–tissue lesions are referred to as lesional TLE (> Table 160-2). In the context of imaging, if one were to categorizing TLE along MRI findings, then a patient with an MRI that does not reveal an obvious lesion, would be referred to as having ‘‘MRI normal’’ TLE. In general a normal MRI portends a poorer prognosis with respect to seizure outcome following surgical interventions [38,43,70]. At the other end of the spectrum an entity entitled ‘‘dual-pathology’’ refers to the presence of both MTS and an associated extrahippocampal lesion [75,76]. Clinically these various subtypes of TLE may be rather difficult to distinguish. Nonetheless, MRI normal TLE and dual pathology will be discussed separately given the unique diagnostic challenge, surgical approach, and seizure outcome these entities engender.
mTLE mTLE is the prototype of the TL epilepsies, and the pathology that is most often associated with mTLE is MTS. Being observed as early as 1825 that half of institutionalized patient with epilepsy dying form natural causes have pathological changes within the mesial TL [77], it still remains that the majority of mTLE that is treated surgically is associated with MTS [20,78–80]. In regards to MTS, it is important to recognize that the pathological changes noted in the medial TL may not be causative, as it has been noted for some time that many different lesions within
Selective amygdalo-hypocampectomy
160
. Table 160-1 Comparison of the electroclincal and diagnostic aspects of mTLE and nTLE
Clinical aspects [32,38]
Pre-operative testing
mTLE
nTLE
Auras (simple partial seizure) Visceral sensation, fear or both De´ja` vu Illusions, hallucinations Seizures (CPS) Autonomic changes (heart rate, respiration, piloerection, borborygmi) Arrest of behaviour and motionless stare Loss of awareness (‘‘dysconscious’’) Oroalimentary automatisms Contralateral dystonic posturing Post-ictal Nose-rubbing [68] Dysphasia if dominant hemisphere involvement MRI MTS, other structural pathologies, both (dual pathology), or none (‘‘MRI normal’’ TLE) Neuropsychological testing
Similar to MTLE
Lateralized memory impairment. Scalp EEG ‘‘Classical’’ anterior temporal inter-ictal spikes(IIS) [70,71]
Intracranial recordings
Seizures originate from medial TL (amygdala, hippocampus, parahippocampal gyrus) [73]
the medial TL can cause mTLE [20] (see > Table 160-2). Initially described by Sommer [81], the classical pathological description of MTS includes neuronal loss in Sommer sector (CA1) (see > Figure 160-1), CA4 and the dentate gyrus, gliosis, and with more severe cases showing involvement of the uncus, fusiform gyrus, and amygdale [81–83]. A more complete description includes, dispersion or duplication of the granule cell layer [84], sprouting of mossy fiber system in the dentate gyrus [85], loss of pyramidal and polymorphic
MRI – same as MTLE Neuropsychological testing
Less likely to have lateralized memory impairment More likely to have naming problems [69] Intracarotid sodium amobarbital testing Less likely to have lateralized memory impairment on side of seizure onset than MTLE [72] Scalp EEG No clearly unique EEG pattern, although there may be absence of classical IIS or multiple types of IIS Variable, with wide spread electrophysiological changes [74]
neurons in CA4 [86], and preservation of neurons in the CA2 region. How isolated MTS develops and how it causes seizures still remains somewhat enigmatic [87]. Falconer posited that prolonged febrile seizures may lead to the development of MTS [88], consistent with epidemiological data suggesting that prolonged febrile convulsions increase the probability of developing TLE [89]. Other pathologies associated with mTLE include tumors either malignant or benign (i.e., gliomas, and meningiomas), infectious causes (i.e., herpes, tuberculosis, and cysticercoisis),
2681
2682
160
Selective amygdalo-hypocampectomy
. Table 160-2 Pathological entities associated with lesional TLE Pathological entities associated with lesional TLE Congenital: Migrational disorder Cortical dysplasia Hamartoma Acquired: Tumors: Oligodendroglioma Astrocytoma (various WHO grades) Ganglioglioma Dysembryoplastic neuroepithelial tumor (DNET) Mixed glial tumor Meningioma Vascular malformations and their sequelae: Cavernous malformation Arterio-venous malformation Previous hemorrhagic or ischemic stroke Infections: Herpes Tuberculosis Neurocysticercosis Trauma: Encephalomalacia and gliosis
vascular malformations (arteriovenous malformations, and cavernomas), trauma, and hamartomas (see > Table 160-2). Interestingly a specific gene mapping to chromosome 4q13.2-q21.3 [90] has been associated with an abundance of simple partial seizures (auras), with a paucity of CPS entitled Familial Temporal Lobe Epilepsy [91]. mTLE has a relatively distinct inter-ictal and ictal electrophysiological signatures on both scalp recordings and intracranial recordings [73,92,93]. Although these electrophysiological features will not be reviewed here it is important to mention that these ‘‘signatures’’ can occur bilaterally either synchronously, asynchronously, or independently [73,92,93].
Neocortical TLE (nTLE) As eluded to above, the semiology of nTLE may be similar to that of TLE, making them at times clinically indistinguishable [38,70]. Various features
of nTLE and mTLE are compared and contrasted in > Table 160-1. The various pathologies that can give rise to nTLE epilepsies include those pathologies that give rise to TLE in general excluding MTS, and are listed in > Table 160-2.
MRI Normal TLE Approximately 30% of individuals undergoing surgery for TLE have a normal appearing MRI scan [94]. Apparently a normal MRI however does not always imply the absence of pathology. In a recent series 6 out of 17 (35%) patients with a normal pre-operative 1.5T MRI displayed MTS from pathological specimens with the remaining patients confirmed to not have any detectable pathology on re-review [95]. This raises the question of whether or not MRI normal TLE is a subset of mTLE arising from radiologically undetectable MTS or a distinct pathological entity. Unlike the findings in the above study [95], it has been argued by others that MRI normal TLE is a distinct entity, since only one patient out of ten displayed MTS in light of a normal pre-operative MRI [94]. Further complicating this picture is that in a series of 13 patients with normal pre-operative MRI, only one did not display any pathological findings, 12 displayed nonspecific finds which included gliosis, heterotopia, chronic inflammation, and microinfarcts, and there was one case with MTS [96]. The discrepancies noted above can be explained to some extent by the limited sensitivity of MRI to disclose small lesions, inability to detect T2 signal changes in the absence of atrophy, and bilateral hippocampal sclerosis that escapes subjective review and that can only be objectively manifest through volumetric analysis [97]. Higher field strengths and improved pulse sequences may further improve the sensitivity of MRI to detect limbic and neocortical pathologies [98], possibly reducing the number of normal pre-operative MRIs in those with TLE.
Selective amygdalo-hypocampectomy
160
. Figure 160-1 3 Tesla (3T) MRI imaging, gross pathology, and histology of mesial temporal sclerosis (MTS): (a1) 3T coronal inversion and recovery (IR) sequence in a patient with right sided MTS. Note the marked atrophy and distortion of internal architecture of the right hippocampus. (a2) Coronal FLAIR imaging revealing high signal with in the atrophic right hippocampus. (b) En bloc resection of hippocampus displaying gross anatomy and pathological changes of atrophy, particularly noted within the body of the hippocampus. At the time of surgery one can appreciate the rather firm consistency which is rather unlike normal brain. (c) Luxol Fast Blue stain at 2.5X magnification of hippocampal section from the same patient as in A1, marked neuronal drop out in the CA1 region. Abbreviations: PHG, parahippocampal gyrus; H, head of hippocampus; B, body of hippocampus; HS, hippocampal sulcus; DG, dentate gyrus, and; CA1, Sommer sector
Dual Pathology MTS associated with lesions outside of limbic structures including extra-temporal sites has been estimated from MRI volumetric analysis to occur in up to 15% of individuals presenting with partial epilepsy [99]. Histologically it has been shown that the majority of specimens from individuals with the electroclinical diagnosis of
mTLE and dual pathology almost uniformly display some amount of hippocampal cell loss [100]. However, the amount of hippocampal cell loss appears to be a function of the type of extra-hippocampal pathology present with gliomas and hamartomas [100,101] to be least likely to be associated with MTS, and the converse for heterotopias [100]. In another study a similar pattern was found from the MRI volumetric study
2683
2684
160
Selective amygdalo-hypocampectomy
mentioned above, but in a addition to gliomas and hamartomas, a paucity of MTS was also found in association with vascular lesions [99]. oWhether it is the extra-hippocampal pathology that is the epileptogenic lesion in those with dual pathology, the hippocampus alone, or some combination remains a matter of speculation [100]. Regardless, it would appear that removal of both lesions affords the greatest chance of seizure freedom [102,103] provided there is concordant localization from pre-operative testing.
Assessments for Surgical Candidacy Routine Assessments Before a patient is evaluated for surgery they have typically undergone some form of structural imaging either CT and rarely MRI, as well as scalp recordings either routine or sleep deprived. The MRI is typically ‘‘normal’’ as most MRIs not specifically tuned to the temporal lobes will miss MTS [97]. MRI is the imaging of choice in the context of newly diagnosed and medically intractable epilepsy [104], with coronal FLAIR and inversion and recovery (IR) pulse sequences of particular utility in the imaging of the temporal lobes [97]. As discussed above CPS are harbingers of TL involvement and thus the imaging requisition should guide the radiologist coding the scan to the temporal lobes and the pulse sequences listed above (in addition to the usual MRI sequences usually ordered). With respect to routine electrophysiological testing, a single 2-h awake-sleep deprived scalp EEG greatly enhances the sensitivity of routine outpatient scalp EEG monitoring in TLE [105], and should be done in preference to repeated routine EEGs following a normal initial EEG result [104]. These ‘‘routine’’ tests need not wait until referral to a surgical epilepsy program. The finding of a lesion on MRI in an individual with CPS is an important management clue that they
will likely fail their ASM medication [39] and a strong predictor that surgery will be of benefit to them [43]. In light of continued seizures these individuals should be referred to a surgical epilepsy center early in their disease course, rather than being subjected to a protracted and potentially deleterious trial of poly-pharmacy [50]. The ultimately goal of assessment for surgical candidacy is identification of the epileptogenic zone, which is that area of the brain that when removed is sufficient to abolish seizures [106,107] (see > Figure 160-2). The ictal onset zone is rarely if ever identified with certainty and may be contained within, or may surround the epileptogenic lesion if one exists. The irritative zone is thought to be the most expansive of the zones which when identified may be useful in directing one to the epileptogenic zone, but need not be removed in its entirety to obtain seizure freedom [108].
Neuroimaging Structural As mentioned above MRI is not only the imaging modality of choice, but is becoming increasingly available to general neurologists and family doctors. Improvements in MRI technology are increasingly providing better spatial resolution and tissue contrast (see > Figure 160-1), improving the detection of more subtle cortical abnormalities especially malformations of cortical development [97]. In the context of MTS and its radiological determination, subjective interpretation has been shown to be highly sensitive and specific for MTS [109]. However, a more quantitative approach to determining atrophy is volume measurements obtained from MRI which have been shown to correlate well with the degree of neuronal loss in post-surgical specimens [110]. Another quantitative approach involves determining the tissue hydration, known as T2 relaxometry
Selective amygdalo-hypocampectomy
160
. Figure 160-2 Anatomical and conceptual considerations. Various anatomical components of the TL are labeled on the right TL. On the left TL are identified theoretical zone involved in the generation, electrophysiological manifestations and surgery of TLE (adapted from [106,107]). The epileptogenic lesion is the structural cause of the seizures; the ictal onset zone corresponds to the region where seizures are initiated; the epileptogenic zone is the region that when resected results in abatement of seizures, and; the irritative zone is the cortex that generates inter-ictal spikes. Abbreviations: T1, superior temporal gyrus; T2, inferior temporal gyrus; T3, inferior temporal gyrus; FUS, fusiform gyrus; PHG, parahippocampal gyrus, and; H, hippocampus
[111], as MTS is distinguished not only by decreased volumes, but as well increases in tissue hydration. This is a result of expansion of the extracellular space resulting from neuronal loss which is then reflected in increased signal on T2 weighted MRI scans [109] (see > Figure 160-1).
Tests of Cortical Function Functional MRI (fMRI) is showing great promise and may ultimately supplant more invasive tests like the intracarotid sodium amobarbital test [112] (Wada test), and its recent variant the etomidate speech and memory test [113] (eSAM) in the determination of hemispheric dominance for
language [114]. As well fMRI has utility in identifying motor areas both primary and supplementary, as well as primary somatosenory area, but localization of these functions is rarely required in the assessment of an individual with TLE. Positron Emission Tomography (PET) has been utilized for functional localization [115] and appears to be increasingly available throughout the world [108].
Localization of the Irritative and Epileptogenic Zones Ictal and inter-ictal radionuclide based imaging techniques to determine regional metabolic (PET)
2685
2686
160
Selective amygdalo-hypocampectomy
or blood flow changes (single-photon emission computed tomography; SPECT) have been used to guide surgical procedures for sometime now [116] (Bonte, 1983). Ictal-SPECT appears to have its greatest utility in TLE [117], with subtraction ictal-SPECT coregistered to MRI (SISCOM) (Lewis, 2000) apparently further increasing localization of seizure foci as compared to side-by-side comparison [118]. Presently ictal and interictal PET appears to show greater promise that SPECT in the work-up of those with TLE [117]. fMRI has also been shown to relatively reliably map ictal and interictal phenomenon [119], with further confirmation being obtained with concurrent MRI and EEG recordings [120]. The utility of magnetoencephalography (MEG) remains to be proven in TLE due to the rather deep anatomical location and orientation of the medial TL neuronal laminae which creates difficulty in detecting and orienting dipoles [121]. MEG appears to have greater utility in the neocortical epilepsies including nTLE, adding much improved spatial information to inter-ictal phenomena [121]. Like fMRI, MEG has utility in functional mapping including motor and sensory mapping [121], and as well has shown promise in the determination of hemispheric dominance for language [122].
Neuropsychological Testing Pre- and post-operative neuropsychological testing comprises an important aspect of the management of individuals being considered for, and those who have undergone surgery [123]. Preoperative testing seeks to identify deficits in the cognitive domains of: attention, memory, language, visuospatial ability, and executive functioning [123]. In the context of TLE the two primary spheres of cognitive testing that help not only to lateralize and localize the epileptogenic zone, but as well to predict post-operative neuropsychological status and seizure outcome are memory and language [123].
In regards to ascribing structure to function, it has been well established that the hippocampus is vitally important in memory and learning [24], whereas the temporal neocortex of the dominant hemisphere plays a role in word-retrieval [69]. Thus tests of memory [124] both verbal and visuospatial in the form of the Weschler Memory Scale, and tests of language[69] (Boston Naming Test) can be used to assess medial and neocortical TL function respectively. Deficits in these various domains can then be used to infer whether the seizure focus is medial or lateral, and whether it is dominant or nondominant. Discordance between neuropsychological testing and other diagnostic tests (EEG, neuroimaging) will often require further testing to lateralize and localize function as well as the epileptogenic zone (i.e., intracranial recordings) within the TL. For example, although a patient with left sided MTS with visuospatial memory deficits may be assumed to have atypical speech, this would need to be confirmed either by fMRI or Wada testing. In regards to predicting outcome, given that pre-operative verbal memory scores are predictive of post-operative memory status following resection of the dominant TL [125] (in addition to other factors), results of such testing can be used to inform the patient as well as the surgical decision making.
Intracarotid Sodium Amobarbital Test (IAT) The IAT also known as the Wada Test [126] (named after its founder Juhn Wada) was initially developed to determine hemispheric dominance for language. Brenda Milner at the Montreal Neurological Institiute (MNI) further extended its use to the assess the integrity of the medial TL with their prominent role in memory [127]. There is a plethora of literature attesting to the ability of the Wada Test to predict post-surgical
Selective amygdalo-hypocampectomy
memory outcome, and as well a significant literature proclaiming its limitations (for discussion of these issue see Ojemann & Kelly, 2002 [128], and more recently Baxendale et al. (2008) [129] with additional enlightening commentaries). Despite this on going debate [129,130] it is important to highlight that the Wada Test still remains the test of choice in the few selected patient’s in whom there is concern for significant post-operative amnesia [131]. However, given the improvements in MRI technology this may not be the case for too much longer [132,133].
160
demonstrated that the constellation of unilateral ‘‘classical’’ medial TL spikes with concordant imaging (MTS ipsilateral to inter-ictal spikes) and neuropsychological testing has the same predictive value in regards to seizure outcome than does localizing information obtained from ictal recordings [105,137,138]. Given the large number of individuals who can’t access surgery, this may be a safe and effective way of lessening the bottle-neck for those with the clearly defined syndrome of mTLE [139]. A detailed description of the use of EEG in the context of epilepsy can be found in the chapter by Wennberg et al.
Video EEG EEG remains the undergird of surgical interventions for TLE. Continuous EEG monitoring can capture ictal and inter-ictal phenomena, and when correlated to behavior through video monitoring is a fundamental tool in seizure classification and localization of the epileptogenic zone. Patients admitted to the Epilepsy Monitoring Unit (EMU) have either electrodes glued to their scalp or implanted intracranially, and by the passage of time and through provocative measures (medication reduction, and sleep deprivation) seizures and their clinical manifestation are captured digitally and reviewed off-line. Although there had been a initial interest in pharmacological activation in the identification of the epileptogenic zone, this approach has fallen into disfavor [134]. Clinical testing by the attending nurse or nurse practitioner during a seizure or afterwards in the post-ictal state, can further aid in defining the cortical areas involved during seizure activity. Although there has been a move away from reliance on inter-ictal activity [135] to ictal data as the primary guide for surgical intervention [108], surgery based on inter-ictal data alone may be appropriate in some cases of mTLE given recent advancements in imaging and a better understanding of the variables that predict seizure freedom post-surgery [136]. It has been
Diagnostic Surgical Interventions As many as one third of the individuals with TLE can not be lateralized by scalp recordings alone [28,140] and up to 80% of those with what appears to be bi-temporal seizure onsets from scalp electrodes can be shown to have unilateral seizure onsets from intra-cranial recordings [141], Although the number of individuals requiring invasive testing may be declining due to the introduction and improvements of some of the pre-surgical tests mentioned above, invasive recordings are likely to remain an important aspect of the diagnostic work-up in those with TLE. This is largely due to the fact that the non-invasive tests mentioned above save for scalp recordings, are inferential tests, and unlike invasive electrophysiological recordings can not prove epileptogenicity [142] which is the basis of surgical interventions for epilepsy. Surgical interventions to delineate the epileptogenic zone can be either semi-invasive or invasive. The placement of these electrodes seeks to lateralize and localize seizure onsets that will then be used to guide surgical ablation of these sites tempered by avoiding areas of eloquence. In the context of localization, the typical question to be answered is whether the seizures arise from the medial structures, the temporal
2687
2688
160
Selective amygdalo-hypocampectomy
neocortex, extratemporal sites, or some combination of the later. If the recordings suggest they arise from the neocortex then the question of where they arise neocortically must be answered. From a technical perspective these procedures seek as their endpoint the placement of electrodes on or within the brain in order to obviate the cranial vault that poses anatomical and electrical constraints on electrophysiological recordings. Although placing electrodes closer to potential sources has many advantages including direct sampling from putative epileptogenic areas, there is an obvious sampling bias, as only a limited amount of the brain surface and volume can be recorded from. Nonetheless, a hypothesis driven by non-invasive testing [143] and primarily guided by EEG and increasingly by MEG as to where the possible seizure generators are, guides the placement of these electrodes. Two broad categories of these types of recordings can be distinguished based upon the duration of the observation period. Chronic recordings require admission to the EMU following implantation of electrodes and last on average 10 days but may be up to as long as 4 weeks. Conversely, acute recordings from the brain termed electrocorticography (ECoG) are performed in the operating room and are thus restricted to the duration of the surgery and therefore typically last not more than 30 min. ECoG requires reasonably good pre-operative localization of seizure onsets as the area of recording is limited to the region afforded to one by a craniotomy.
Chronic Recordings Semi-invasive recordings utilize foramen ovale (FO) electrodes and peg electrodes, while invasive monitoring utilizes depth electrodes, subdural strips, and grid arrays. FO ovale electrodes are placed percutaneously via the same approach utilized for the treatment of trigeminal neuralgia
[144], and are then clamped to the skin. Their primary utility is for recording from both mesiobasal TLs and are thus helpful is lateralizing seizure onsets, and localizing onsets to the mesial structures of the TL [145]. Peg electrodes are as the name implies are essentially metal plugs, inserted into holes made in the skull, with their distribution on the calvarium predetermined from the pre-operative assessments. These are epidural recordings and their utility is as sentinels when it is unclear to proceed with more conventional invasive recordings. They have often been placed contralateral to conventional intracranial electrode implantations to rule out any contralateral ictal activity [145]. Depth electrodes, subdural strip electrodes and grids, are constructed by embedding metal electrodes that are MRI compatible within a matrix which is typically constructed from Silastic. As the name implies, depth electrodes are used to record from regions within the brain, and are thus placed through the substance of the brain towards or within a defined target. These electrodes are cylindrical in shape and measure not much more than 1 mm in diameter. In the context of TLE these electrodes are usually placed with stereotactic guidance within the amygdala and the hippocampus. The hippocampus can be approached either orthogonally via temporal burr holes (more common approach), or along the long axis of the hippocampus via an occipital burr hole [146]. Subdural strip electrodes are malleable arrays of electrodes usually a single contact wide, and are placed on the surface of the brain in the subdural space. Their malleability allows them conform to the shape of the TL contours ensuring that they are closely approximated to the cortex. These electrodes come in a variety of lengths, number of contacts, and widths, that can be exploited depending on the clinical question. This being said, in cases of bi-temporal epilepsy the electrode configuration utilized in Toronto is relatively standardized as the question posed
Selective amygdalo-hypocampectomy
in this situation is rather stereotypical (see > Figure 160-3). For placement of these electrodes a burr hole is placed above the root of the zygoma just anterior to the tragus of ear, which is often enlarged to some extent with a bone punch to facilitate passage of three subdural strip electrodes and a depth electrode. Frameless stereotaxy can be useful in guiding these electrodes to desired locations on the TL surface including anterior, mid- or basal, and posterior temporal neocortical locations. The anterior temporal electrode is guided around the tip of the temporal, the mid-temporal such that the most distal contact lies on the parahippocampal gyrus, and the posterior temporal reaches back to the end of Sylvia fissue along the middle temporal gyrus.
160
Post-operatively electrode locations can be confirmed by visual inspection of the electrode artifacts, or three dimensionally through various methods of image reconstruction [147,148]. The term grid refers to parallel arrays of electrodes that essentially form a sheet or matrix of electrode contacts. This malleable grid must be placed via a craniotomy and can cover up to 10 cm2 of cortex. When utilized in the context of TLE, their physical structure is ideally suited to localize neocortical generators. Grids can be placed over any combination of lateral temporal neocortex and frontal, parietal, or occipital cortices, but are not easily placed on the basal aspects of the TL, a location generally reserved for strip electrodes. Unique issues arise with
. Figure 160-3 Chronic intracranial recordings for bi-temporal epilepsy: For some clinical situations rather stereotypical electrode placement schemes can be devised. For example in an individual with bi-temporal abnormalities on scalp EEG (without evidence of posterior temporal or frontal lobe involvement) coverage could include the mid-, basal, mesial, and anterior TL neocortex with surface electrodes with or without hippocampal depth electrodes. (a) Schematic of electrode configuration displaying three subdural strip electrodes, and one hippocampal depth electrode. (b) Post-operative electrode locations. Skull stripped three dimensional reconstruction with overlaid electrode positions, inferred from the artifacts generated by the metal contacts. (c) Location of depth electrodes
2689
2690
160
Selective amygdalo-hypocampectomy
implanting such large amounts of foreign material not the least of which is infection (see below), but as well mass effect. Strategies to reduce these effects are mannitol prior to implantation, fashioning of a duraplasty, and loosely affixing the craniotomy with suture (rather than rigidly with metal plates and screws) to accommodate the added intracranial volume. Regardless of electrode configuration, the electrode leads must be externalized, and this is accomplished by tunneling the electrodes subcutaneously as far from the implantation site as possible. The leads which can be rather numerous at times are anchored to the skin with a purse string that ensures a tight seal around the exiting electrode lead minimizing leakage of cerebrospinal fluid (CSF). Tunneling and anchoring are simplified by shaving the patients head in its entirety, which also helps with post-operative hygiene as these electrodes may remain in for several weeks. The most significant intra-operative complications is that of vascular injury. Placement of depth electrodes is associated with a 1% risk of hemorrhage and a an overall neurological risk of similar magnitude [149]. Lower hemorrhage rates have been reported for subdural strip electrodes [150], but unlike depth electrodes, no deaths have been reported following placement of subdural strips [150,151]. Other complications include infections, CSF leak, brain swelling, aseptic meningitis, status epilepticus, and transient neurological deficits [152]. Infection occurs in approximately 1% of patient’s implanted with subdural strip electrodes [150], and this rate might be somewhat higher for grid electrodes [152,153]. Continued antibiotics following strip electrode implantation does not reduce the infection rate [150] and thus beyond the single dose given routinely pre-operatively [154] antibiotics need not be continued. If there is evidence of infection the recording period is terminated, the electrodes removed and the infection treated with antibiotics.
Overall, despite the invasiveness of these procedures, and their total attendant surgical risk of 3% [152] they comprise a relative safe and often necessary component of the diagnostic assessment of those with TLE.
Acute Recordings As mentioned above acute recordings from the brain fall under the rubric of ECoG, and are performed as an adjunct to an already planned resection involving a craniotomy. ECoG has a long history in surgery for TLE being used to define potential areas of seizure onset during TL surgery, and as well in the mapping of eloquent cortex [155]. ECoG has as well been utilized to guide the extent of hippocampal resections, and as a prognostic tool that utilizes evidence of residual inter-ictal spikes as a predictor of seizure outcome [156]. The term ‘‘tailoring’’ has been applied to a number of the activities listed above in order to distinguish this type of surgical approach from a standard or anatomical resection [20,157,158]. ECoG is performed from the exposed brain utilizing special electrode arrays (see > Figure 160-4), by placing strips, grid, and depth electrodes on the surface or within the substance of the brain. The use of grids in the acute setting is somewhat problematic as their malleability often makes it difficult to maintain uninterrupted contact with the exposed cortex. Due to their short duration, acute recordings preclude recordings seizures. Thus the utility of intra-operative ECoG relies on inferring the limits of the epileptogenic zone through identification of the irritative zone, as inter-ictal activity is likely the only abnormality to be recorded in this setting. Although the ultimate utility of ECoG in TLE continues to engendes debate, there is a plethora of data to suggest that tailoring of TL resections in non-lesional TLE to pre-resection ECoG has little bearing on ultimate seizure control [62,159–162], and that post-resection
Selective amygdalo-hypocampectomy
160
. Figure 160-4 Electrode arrays for acute and chronic ECoG: (a) Electrode array utilized for acute ECoG (Grass Technologies, West Warwick, RI). This array has been firmly affixed to head frame, with grounding and reference electrodes placed in the temporalis muscle (red a green wires). The electrodes are spring loaded to afford continuous contact with the brain. This array affords 16 electrode contacts, which can be increased through application of subdural strip electrodes. (b) Intermediate stage of implantation of a subdural grid and subdural strip (SS) electrodes (Ad-Tech Medical Instruments Corp., Racine, WI) via a craniotomy for chronic recording. For orientation purposes the temporal lobe (TL) can be seen well demarcated from the frontal lobe (FL) by the sylvian fissure (SF). In this specific case the final grid location was guided by identifying the motor strip (MS) through stimulation mapping (paper ticket under grid). Concordant data was obtained in regards to the location of the MS via recording of the phase reversal of the median nerve somatosensory evoked potential. Note how the grid does not remain applied to the brain, which limits its utility in acute ECoG recordings
2691
2692
160
Selective amygdalo-hypocampectomy
spiking is an equally unhelpful prognosticator for ultimate seizure outcome [160–162]. The other utility of ECoG is in the context of electrical stimulation mapping, a technique used to identify and preserve areas of eloquence [155]. In this capacity ECoG is used not only to record after discharges as a positive control and establishment of appropriate stimulation currents [163], but as well for identifying not an uncommon consequence of electrical stimulation of the brain, that being induced focal seizure activity
(see > Figure 160-5), which can be abated by the application of cold TissuSol before its potential generalization.
Therapeutic Surgical Interventions Before delving into the surgical procedures utilized in the treatment of TLE some of the relevant functional anatomy will be reviewed as it is these considerations that determine the surgical target
. Figure 160-5 Intra-operative ECoG during stimulation mapping: (a) Upper panel. Unfiltered trace from electrode nearest to stimulation site (E) during dominant hemisphere neocortical TL resection near epileptogenic zone identified from previous intracranial implantation. Stimulation parameters: 60 Hz, 13 mA, 100 ms pulse width. The time of stimulus application (Stim) is shown as a red bar, with the ictal region identified with a yellow time bar, and application of cold TissuSol as a blue time bar. A preceding stimulus artifact is reflected at this electrode from a distal site. Lower panel. Same data as in the upper panel albeit filtered (1 Hz high pass filer, 250 Hz low pass filter). The ictal activity begins approximately 3 s after the stimulus is removed, and does not appear to wane until the application of cold TissuSol. Given abatement of the seizure, mapping could continue to identify naming areas at numbered tickets 1 and 2. (b) Intra-operative photo taken post-stimulation mapping (after removal of some of the electrodes for visualization) indicating areas of anomia during picture presentation (areas 1 & 2), and right facial contraction (area A). Patient is supine, with the head turned to the right. Black and white photos are presented via a heads up display [164] attached to a laptop computer
Selective amygdalo-hypocampectomy
and the extent to which resective temporal procedures can be undertaken. For excellent reviews of the surgical anatomy, surgical approaches, and operative complications the reader is referred to Pilcher and Ojemann [165], Wen et al. [166], and Olivier [167]. The detail contained in these works will not be reproduced here and thus it is recommended that these primary articles be reviewed for a more complete treatment of surgical anatomy, surgical technique, and complication avoidance.
Functional Anatomy Lateral Neocortex Anatomically the TL is considered the most heterogeneous of the four lobes, having a typical six layered isocortical mantle that hides a complex convolution of mesocortex and allocortex medially, known as the limbic core, consisting of the hippocampal and prepiriform cortex and the amygdala. On the lateral surface of the brain the isocortical margin is indistinct posteriorly, ending at an imaginary perpendicular drawn from the end of the sylvian fissure to a line joining the parieto-occipital sulcus to the temporo-occipital notch. This lateral isocortical mantle is highly variable except for the superior temporal gyrus which is delineated by the sylvian fissure dorsally and the superior temporal sulcus ventrally. The superior temporal sulcus is usually deep and travels in a unbroken line running parallel to the sylvian fissure usually ending at the angular gyrus. The middle and inferior temporal sulci show considerable variation, are usually shallow and often discontinuous. On the ventral surface the fusiform or temporo-occipital gyrus of the TL is demarcated by the inferior temporal sulcus laterally and the collateral sulcus medially. The collateral sulcus is invariant like the superior temporal sulcus, and separates the fusiform gyrus from the parahippocampal gyrus.
160
Language
Penfield identified three brain regions that when electrically stimulated produced speech arrest and were designated as anterior, posterior, and superior language areas [168]. The superior language area is considered to be the supplementary motor area and is generally not considered to be an ‘‘essential’’ language area. The anterior and posterior areas correspond to Wernicke’s and Broca’s area respectively, and have conceptually represented rather discrete and immutable brain regions dedicated to language in the dominant hemisphere [1,169]. With respect to posterior language areas, although these areas may be discrete in any one individual there is clearly variability from person to person [155,163,168]. Through detailed stimulation mapping during epilepsy surgery language areas have been identified in the perisylvian region including the posterior portion of the superior temporal gyrus, the supramarginal gyrus, the inferior portion of the inferior parietal lobule, and some rather anterior portions of the middle and dorsal portion of the inferior temporal gyrus [155,163,168]. Ojemann et al. (1989) found that in a series of 117 patients who underwent stimulation mapping of dominant perisylvian cortex, that there was both interand intra-site variability for essential language areas during object naming [163]. With respect to inter- and intra-site variability, the majority of sites displayed discrete boundaries, but some had well defined transition zones, and their composite areas were quite variable [163], and being considerably smaller than originally reported by Penfield [168]. Essential language areas have also been demonstrated as far ventrally as the basal temporal fusiform gyrus [170–173]. Lu¨ders et al. (1986) reported on a 38-year-old patient with left temporal seizures who developed complete receptive and expressive aphasia during extraoperative stimulation mapping of the basal TL [170]. Furthermore the same group performed stimulation mapping in 22 dominant and 7 non-dominant
2693
2694
160
Selective amygdalo-hypocampectomy
sided extraoperative localizations for TLE. Of the 22 dominant sided mappings, eight patients developed receptive and expressive aphasia when stimulation occurred over the fusiform gyrus within 3–7 cm from the temporal tip [171]. Resection of basal temporal language sites may cause long term impairments of object naming in those undergoing temporal lobectomy for epilepsy, although the decrement appears rather small, compared to those who have had sparing of this region [174]. Verbal Memory
Neuropsychological tests linking dominant lateral temporal isocortex to verbal memory have demonstrated decrements in verbal memory postoperatively in patients undergoing anterior temporal lobectomies for the relief of seizures [175]. More recent evidence has included verbal memory deficits following temporal lesions [176–178], and recent verbal memory changes during temporal cortical stimulation [176,179–181]. It has been suggested that the magnitude of the memory deficits that occur post-operatively correlate with the extent of the lateral resection and not the medial resection [124,176,182]. Ojemann and Dodrill (1985), used stimulation mapping during a verbal memory paradigm in patients undergoing limited medial resections dependent on the results of preoperative intracarotid sodium amobarbital testing. They found a significant correlation between the verbal memory scores and the extent of lateral resections, but not medial resections, and this correlation was independent of the degree of seizure control [176].
The Limbic Lobe This synthetic lobe includes a ring (limbus) of structures on the most medial margins of the frontal, parietal and temporal lobes surrounding the corpus callosum. It consists of the hippocampal formation, the dentate gyrus, the amygdala, plus the subcallosal, cingulate, and parahippo-
campal gyri. The parahippocampal gyrus is continuous with the cingulate gyrus through the isthmus of the cingulate gyrus, being bound laterally by the rhinal and collateral sulci, and superiorly and medially by the hippocampal sulcus. The parahippocampal gyrus wraps around the hippocampal sulcus to form the uncus. Olfactory Lobe
This area of the medial TL is separated from the mesocortex of the parahippocampal gyrus, by the rhinal sulcus, the posterior portion of which develops into olfactory area (anterior perforated substance) and the pyriform lobe. The pyriform lobe includes the pyriform, periamygdaloid and the entorhinal areas. The entorhinal area (area 28 of Broadmann) comprises the majority of the anterior parahippocampal gyrus, and constitutes the major afferent supply to the hippocampus [183]. The pyriform lobe is the caudal extension of the lateral olfactory stria, and is often referred to as the lateral olfactory gyrus. The peri-amygdaloid area is a small area dorsal and rostral to the amydaloid nuclear complex. The pyriform cortex plus the peri-amygdaloid constitute the primary olfactory cortex, constituting a unique sensory area in which information is not relayed through the thalamus. Parahippocampal Region
This region includes the entorhinal, perirhinal, and parahippocampal cortices. The entorhinal cortex appears to have a pivotal position receiving most of its input from the perirhinal and postrhinal corticies and in turn providing the majority of the input to various sectors of the hippocampus, the dentate gyrus, and the subiculum. Furthermore the majority of the output from the hippocampus (primarily CA1) and the subiculum is then relayed back to the entorhinal cortex, and then back to the perirhinal and postrhinal cortices [184] a well as to cortical regions via the fornix. The parahippocampal region receives unimodal and polymodal neocortical inputs form a variety of association areas, including visual, auditory
Selective amygdalo-hypocampectomy
and somatosensory areas [185], with reciprocal connections to neocortical association areas. The entorhinal cortex is known to project to the prepiriform, prelimbic, infralimbic, agranular insular, cingulate, retrosplenial, posterior oribitofrontal and temporal corticies [186–188]. The perirhinal cortex projects to the prefrontal, parietal, temporal, and occipital cortices [189,190]. The parahippocampal gyrus has a significant role in memory processes as an import way station for hippocampal afferents [191], and an important component of a network of brain regions mediating familiarity, one form of declarative memory [192]. Hippocampus
This structure can be divided into three parts, precommisural, supracommisural, and retrocommisural, with respect to their positions relative to the corpus callosum. The retrocommisural hippocampus forms the main portion of the hippocampal formation, and is differentiated into three longitudinally arranged allocortical structures, the hippocampal and dentate gyri, and the subiculum which abuts a mesocortical transition zone adjacent to the parahippocampal gyrus. The pyramidal layer of the hippocampal gyrus is subdivided into four regions designated CA1 to CA4 inclusively, which forms a ‘‘C’’ shaped structure interlocked with the ‘‘C’’ shaped granule cell layer of the denate gyrus. The flow of information through the hippocampal formation follows the well known trisynaptic circuit. Input to the hippocampal formation as mention previously originates primarily from the entorhinal cortex entering via two pathways: the perforant path which terminates on granule cells of the dentate gyrus, and the alvear path. The trisynaptic circuit is formed by synaptic connections of entorhinal to dentate gyrus via the perforant path, dentate gyrus to CA3 via the mossy fiber projections, CA3 to CA1 via the Schaffer collaterals, and then back to the entorhinal cortex. In addition to the entorhinal input, the hippocampus receives corticohippocampal projections from orbitofrontal,
160
infralimbic, cingulate and temporopolar cortices [190,193]. The main efferent system of the hippocampal formation is the fornix, which is composed of axons arising form the pyramidal cells of hippocampus and subiculum. The fimbria (see > Figure 160-1), a convergence of axons that begin in the alveus, continue as the fornix, approaches the contralateral fornix in the midline under the corpus callosum forming a rather poorly developed fornical (hippocampal) commisure. Although generally thought of as a purely efferent fiber bundle, the fornix relays hippocampal and subicular output from one hippocampal formation to the other. The last point is a topic of some debate but there is evidence to suggest that the hippocampal commissure may be functional in humans [194]. The role of the hippocampus in memory can not be over stated, with some of the most compelling information regarding its role in this process being derived from patients who had undergone surgery for epilepsy or psychiatric disorders with resultant bilateral hippocampal damage [24,195]. The study of the index case H.M. along with a plethora of other clinical cases and animal experiments has provided a wealth of information regarding to the role of the hippocampus in long term storage (consolidation) of explicit forms of memory [196]. In regards to explicit material specific memory the hippocampus and related structures within the language domiant hemisphere appear to be involved in verbal memory, whereas in the non-dominant hemisphere they appear to involved in visuo-spatial memory [197,198]. Furthermore, in addition to its role in consolidation and material specific memory, it as well appears to be fundamentally involved in retrieval of personal experience, so called autobiographical memory [199–201]. Amygdala
The amygdala is a large nuclear group located in the dorsomedial portion of the TL, forming the rostromedial and rostrodorsal portion of
2695
2696
160
Selective amygdalo-hypocampectomy
the inferior horn of the lateral ventricle. It is ontogenetically related to the ganglionic eminence, part of which gives rise to the corpus striatum. It is generally divided into a corticomedial group which blends with the cortex of the uncus, forming the gyrus semilunaris of the uncus, and a basolateral group. The basolateral group unlike the corticomedial group does not receive direct olfactory inputs. This structure is markedly interconnected with the hippocampal and parahippocampal regions [202], as well as the basomedial telencephalon and hypothalamus, the brain stem, and non-olfactory areas of the cerebral cortex [203]. Projection to the septopreopticohypothalamic area occur via the major efferent system of the amygdala, the stria terminalis, as well as by the ventral amygdalofugal pathway [204]. The ventral amygdalofugal pathway also transmits fibers to brainstem regions including the peripeduncular nucleus, substantia nigra pars compacta, ventral tegmental area, cuneiform nucleus, periaqueducal grey matter, raphe nuclei, parabrachial nuclei, locus ceruleus, dorsal motor nucleus of the vagus, and the nucleus of the solitary tract [203]. The thalamus and striatum receives unidirectional projections via the inferior thalamic peduncle, originating mainly from the basal nuclear group [205]. From a functional point of view the amygdala appears to receive input from all sensory modalities [206,207] and integrates them into ‘‘visceral, autonomic, somatosensory, somatomotor components of affective behavior’’ [203]. Furthermore, through its powerful reciprocal connections with the hippocampus and parahippocampal regions it may be fundamentally involved in enhancing memory, especially through arousal, for emotionally stimulating events [208].
Epileptogenic Substrates mTLE Some general mechanisms of hyperexcitability are applicable when discussing involvement of the mesial TL and temporal neocortex in TLE [209]. With respect to the cellular neurophysiology of neurons and glia these mechanism include, increased NMDA receptor function, decreased effectiveness of GABAergic transmission, excess release of excitatory amino acids, decreased uptake of extracellular neurotransmitters, activation of metabotropic glutamate receptors, disruption of intracellular Ca2+ homeostasis, and generation of paroxysmal depolarizing shifts [209].
Hippocampus In the context of MTS the hippocampus displays selective loss of excitatory neurons which presents somewhat of a paradoxical situation when considering potential mechanisms of hyperexcitability. An interesting model proposed by Sloviter [210] suggests that in some cases, end folium sclerosis [82] being at times the only pathology evident at autopsy [82,211], is sufficient in of itself to produce TL seizures. End folium sclerosis, in contrast to hippocampal sclerosis, displays neuronal loss and gliosis confined to the end folium or CA4 region [82]. The perforant path stimulation model of epilepsy suggests that with excessive stimulation of the dentate gyrus, it is the hilar cells that are the most easily damaged which include the mossy cells [212]. Mossy cells appear to innervate the g-amino butyric acid (GABA) containing basket cells which mediate feedback and feed-forward inhibition [212]. There is a similar loss of basket cell innervation through loss of CA3 (via Schaffer collaterals) input to the CA1 region [213,214]. Thus the cellular loss that is incurred during some for of insult, results in deafferentation of inhibitory basket cells, and a consequence there is less
Selective amygdalo-hypocampectomy
inhibitory drive to dampen excitatory input through the hippocampal-entorhinal loop [210]. This excess excitation at the level of the denate gyrus and hilus, has been suggested by others as well, to result in what has been termed maximal dentate activation [215]. Maximal dentate activation is characterized by large multiple field potentials, with an associated rise in extracellular potassium and a negative shift in DC potential [215]. The suggestion that this is a reverberatory phenomenon occurring between the hippocampal and parahippocampal regions, was suggested by complete abolition or marked reduction in maximal dentate activation following lesioning of the entorhinal cortex [216]. Mossy fiber sprouting in the human brain [217,218] has been proposed as another contributing mechanism in excessive excitation within the hippocampal-parahippocampal circuit, that is thought to be induced by neuronal injury [219] or excessive activity [220]. The normal dentate gyrus is considered a relatively refractory region with respect to excitability possibly due to its paucity of recurrent excitatory synapses and strong inhibitory local circuits [221]. These functional recurrent synapses [222], may confer upon the dentate gyrus a propensity for synchronous population allowing initiation or propagation within the hippocampal-parahippocampal circuit.
Amygdala and Parahippocampal Region Involvement of the amydala co-existing with hippocampal sclerosis has been well documented since the early 1950 [17,83,223]. In some series almost eighty percent of the specimens analyzed have revealed amygdaloid damage accompanying hippocampal sclerosis [211]. Furthermore, it has been shown that amgdalar sclerosis can exist independently of hippocampal pathology in patients with TLE [224]. These patients appear
160
to differ with respect to their clinical presentation, preoperative imaging, and outcome following surgical intervention [225]. In a large series for the Montreal Neurological Institute 100 patients undergoing resection largely restricted to the amydala, with sparing of the hippocampus, obtained seizure control similar to those undergoing hippocampal removal plus removal of the amygdale [226,227]. Furthermore, in a series of patients described by Jooma et al. (1995) those patients displaying interictal discharges confined to the amydala who underwent selective resection of the abnormally active amygdala were all seizure-free [228]. As has been pointed out by Gloor [229], these results suggest that in some patients the amygdala may play a more important role in the generation of TL seizures than the hippocampus. From an experimental perspective the rate at which a structure kindles, suggests how important that structure is to the generation of seizures: the amygdala initially being shown as the fastest kindled structure in the forebrain [230]. It is not only more rapidly kindled than the hippocampus [231] but there is evidence from animal models that along with neurons in the dentate hilus, it may be one of the structures displaying the earliest and most profound neuronal damage [232,233]. It has been further shown that the olfactory and piriform corticies can be kindled even faster than the amygdala, suggesting that they may also be critical in the generalization of limbic seizures [234]. There is clinical data that would suggest that in light of the accepted involvement of the hippocampus in seizure generation and propagation, the amygdala and parahippocampal regions have been neglected as key players in mTLE. Some of this stems from the potential technical limitations of recording from these structure [229]. A greater role of extrahippocampal structures including the amygdala and the parahippocampal region has been suggested from depth electrode recordings
2697
2698
160
Selective amygdalo-hypocampectomy
performed at the Montreal Neurological Institute [235]. Eighteen patients who had undergone invasive recordings from the amygdala, anterior hippocampus, and parahippocampal gyrus for intractable epilepsy of presumed TL origin were reviewed. In this group of patients although the majority of seizure onsets were from the hippocampus, they remained subclincal and did not spread beyond the hippocampus. However, those arising in the amygdala, and parahippocampal gyrus were much more likely to spread electrographically, and subsequently manifest clinically [235]. Resective data also speaks to the potentially greater role of the parahippocampus in seizure generation and propagation. Thirty patients of the 204 that underwent selective amygdalohippocampectomy at the University Hospital of Zurich, obtained special pre- and post-operative MRI to correlate the extent of resection with seizure outcome [236]. In this report 92% of the hippocampus, uncus, and amydala were removed with only 45% of the dentate gyrus, and 32% of the parahippocampal gyrus. As has been demonstrated in other studies [237], it was concluded that in general the more tissue removed the better the outcome. However, it was the extent of resection of the parahippocampal gyrus that correlated with seizure outcome, and not the extent of resection of the amygdala or hippocampus [236]. A rather novel suppressive role for the hippocampus has been proposed [238]. It has been suggested that interictal activity originating form the hippocampus may actually have suppressive effect on seizures originating in the entorhinal/parahippocampal region [238,239].
Surgical Strategies Extent of Lateral Resection The extent of the lateral neocortical resection is variable, and depends largely on strategies that seek to avoid post-operative language
disturbances and whether or not one is dealing with mTLE or nTLE. mTLE
In the context of mTLE, one approach to determining the extent of the neocortical resection is to remove a fixed amount of neocortex according to language dominance: 4.5 cm along the sylvian fissure on the dominant side, 5 cm within the non-dominant hemisphere [16,18,240]. Indeed it was the early experience of Penfield and Jasper that resections beyond 5–6 cm in the dominant hemisphere lead to post-operative aphasia [155]. A modification of this which spares a greater amount of superior temporal gyrus, encorporates 4.5 cm of the middle temporal gyrus and only a minor amount of superior temporal gyrus [241]. Others remove an even smaller amount of middle temporal gyrus, not extending beyond 3.5 cm along the middle temporal gyrus, and sparing the entire superior temporal gyrus in the dominant hemisphere [242,243]. An alternative approach is to tailor the lateral resection depending on stimulation mapping, identifying essential language sites and avoiding resections that encroach within 2 cm of these sites [163]. Indeed it has been shown by stimulation mapping that areas considered safe in the context of anatomical criteria, may contain essential language sites [163]. A further reduction in the amount of lateral temporal neocortex removed for TLE, can be obtained via selective procedures (see Chapter by Snyder and Valiante). These approaches seek to target only mesial structures which in general are thought to be the substrates underlying TLE. First introduced by P. Niemeyer in the mid-1950s, this approach exposes the hippocampus and amygdala through a trans-ventricular approach via an opening in the second temporal gyrus [244,245]. This opening which measures approximately 2–3 cm in length, 4–5 mm in height, and is situated anterior to the central sulcus along the upper border of the second temporal gyrus obtains the ependyma of the
Selective amygdalo-hypocampectomy
ventricle beyond the depth of the superior temporal sulcus. Within the ventricle the key anatomical landmarks are the choroidal fissure with its attached choroids plexus, the hippocampus lateral to the choroidal fissure, the collateral eminence lateral to the hippocampus, the uncal recess, and the inferior choroidal point. Three disconnections then ensue, anterior, posterior and medial, which have recently been well illustrated and described [166]. In the limit Yas¸argil devised a trans-sylvian route that approaches the mesial structures, albeit transventricularly, through the temporal peduncle, sparing all lateral temporal neocortex [246]. This approach designed to minimize the amount of ‘‘normal’’ brain removed, creates a corridor of entry through the sylvian fissure, through the mesiobasal portion of the superior temporal gyrus adjacent to the limen isulae. The ventricle is entered via this approach, and the amygdala is removed piece-meal, whereas the hippocampus, and parahippocampal gyrus are removed en bloc [246]. It was felt that this may result in fewer neuropsychological deficits postoperatively [246]. nTLE
In the context of nTLE the amount of neocortex to be removed should include the epileptogenic zone determined from pre-operative testing and possibly from intra-operative ECoG (see discussion of the utility of ECoG) tailored to avoid language areas in the dominant hemisphere [163].
Extent of Mesial Resection Jasper can be largely credited as the first individual to clearly identify the abnormal electrophysiological manifestations of partial epilepsy arising form the TL [247]. However it was Gibbs who was able to convince Percival Baliey to operate on individuals with TLE solely based on the presence of anterior temporal spikes recorded on scalp EEG
160
[15]. In both Penfield’s and Bailey’s initial series, resections were generally limited to the temporal neocortex, sparing mesial temporal structures. Penfield’s resection were largely lesion directed, and in only two patients was there resection of the mesial structures [16], whereas Bailey would often remove two or three of the temporal gyri [15]. As EEG became more firmly established as the prime instrument for the localization of epileptic regions, evidence implicating the hippocampus began to grow and in Montreal, re-operations to remove mesial structures were often beneficial with respect to control of seizures [18]. The fact that resection of the mesial structures was associated with improved outcomes, especially if performed at the original resection [19,248,249], prompted Falconer 1 year later to introduced the standard en bloc anterior temporal lobectomy [19]. Since the introduction of the en bloc anterior temporal lobectomy, and the subsequent introduction of selective procedures much debate has been generated regarding the identity of those critical structures that should be removed during a temporal resection, and to what extent should they be removed, to achieve the greatest likelihood of seizure freedom. Indeed it is not surprising that there is such variability in practice with seemingly similar outcomes, when one considers the possible permutations resulting from different combinations of: (1) removal or sparing part or all of the amygdala, (2) variability in the extent of hippocampal and parahippocampal gyrus resection, (3) variability in the extent of lateral temporal neocortical removal, and (4) most importantly, the variability of circuitry abnormalities associated with each individuals seizure disorder. Hippocampus
Historically the role of limbic structures in the generation and manifestation of TLE, became evident as improved outcomes were being obtained with their routine removal either during the patients first operation or during re-operation [17–19,223,248,249] for TLE. Although the gen-
2699
2700
160
Selective amygdalo-hypocampectomy
eral consensus is that the hippocampus should be included in resective procedures for TLE, there is certainly controversy as to how much should be removed, and weather or not intraoperative ECoG can help select optimal volumes of resection. Randomized studies in the field of epilepsy are few and far between, amongst the few is a study by Wyler et al., that looked at the impact of the extent of hippocampal removal on seizure and neuropsychological outcome [250]. In this study 70 patients were randomized to two groups, one group undergoing resection of the hippocampus to the lateral edge of the cerebral peduncle (partial hippocampectomy), and the other back to the superior colliculus (total hippocampectomy), with the extent of the lateral neocortical resection being held constant (4.5 cm of all three temporal gyri). There was a minimum of 1 year follow-up (mean 3.94 years), and the two primary endpoints were seizure outcome (seizure free or not seizure free), and neuropsychological outcome. Hippocampi were graded according to severity of sclerosis, and this served as a variable for post-hoc analysis. There was a statistically significant difference between the two groups, with 69% of the patients undergoing total hippocampectomy being seizure free at 1 year, as opposed to 38% seizure free in the partial hippocampectomy group. Not unexpectedly, for those undergoing left sided resection, only the degree of sclerosis correlated with verbal memory outcome [250], not extent of resection, a finding consistent with other published reports [251–257]. Thus those individuals who were least likely to decline neuropsychologically following a total hippocampectomy were those with the greatest amount of hippocampal sclerosis [250]. If it appears that those with clear cut mesial temporal sclerosis are less likely to suffer a decrement in material-specific memory during resective procedures what about those without mesial temporal sclerosis. In those without mesial temporal sclerosis there appears to be a greater
chance of post-operative decline in verbal memory following dominant resections [250– 257]. Furthermore some studies have shown that larger hippocampal resections are associated with a greater decline in material specific memory [198,258–260]. It thus may be argued that those that run the highest risk to verbal memory should undergo resections that attempt to spare as much ‘‘normal’’ hippocampus, while maximizing the chances of seizure freedom (largest removal possible). It is this philosophy has prompted some to perform intra-operative ECoG to ‘‘tailor’’ the extent of the hippocampal resection to each individual patient. It is assumed that the irritative zone as identified by ECoG, also encompasses the epileptogenic zone, since some studies suggest that post resection interictal spikes on ECoG correlate with seizure outcome [247,261,262]. However generally it is accepted that for TLE intertictal activity is not as reliable as ictal onsets for identifying the epileptogenic zone [263] and specifically that post-resection spikes may not correlate with seizure outcome [62,264–268]. More recently however McKhann et al., have demonstrated that post-resection hippocampal interictal spikes predict a poorer seizure-free outcome regardless of the presence or absence of associated hippocampal sclerosis, and that the extent of surgical resection did not correlate with seizure free outcome [269]. Although it has been suggested that tailoring the hippocampal resection according to intraoperative ECoG may improve memory outcomes [228], this was not specifically addressed in McKhann’s study [269]. Thus given that patients without mesial temporal sclerosis are more likely to have poorer memory outcomes [250], and that intraoperative hippocampal interictal activity may be predictive of a poorer seizure-free outcome [247,261,262,269], it may be reasonable to consider tailored resections in patients at high risk of verbal memory decline; this however has not been directly addressed. Parahippocampal Gyrus
Selective amygdalo-hypocampectomy
From a surgical perspective the PHG does not generally warrant a separate discussion as it is generally removed along with the hippocampus [158,165,167]. As mentioned there is evidence from depth electrode data studies to suggest that epileptiform activity originating within the PHG and amygdale is more likely to manifest clinically than activity originating within the hippocampus [73]. Furthermore in a restrospective review of the selective amygdalohippocampectomy cases performed via the trans-sylvian route by Yasargil [158], volumetric analysis suggested that it was only the volume of resected PHG that correlated with seizure outcome, not resected volumes of the other mesial structures [236]. Amygdala
With respect to the amygdala with its intricate connections to both limbic structures as well as neocortical structures, and a great propensity to generate seizures following kindling experiments [270], it is not surprising that it has been suggested as a target for removal. The combination of focal epileptic discharges from the periamygdaloid region and stimulation mapping able to reproduce automatisms and amnesia in this area, suggested the importance of including the amygdala in resections for TLE [271,272]. Indeed it was probably the realization that pathological changes in mesial structures are probably the most common association with TLE, that suggested development of selective surgical approaches aimed solely at the removal of the amygdala and the hippocampus [244]. Although sclerosis within the hippocampi of epileptics had been described for some time [77,82,83], associated amygdalar sclerosis was first described by Falconer in 1955 in a series of patients who had undergone temporal lobectomy for epilepsy [223]. More recently, in patients presenting with TLE, amygdalar sclerosis may be an isolated finding, with the hippocampus at times seemingly uninvolved [224]. From a clinical and electroencephalographic correlation, Wieser described five psychomotor seizure types: mesio-
160
basal limbic, temporal pole, temporal neocortical posterior, opercular (insular), and frontobasalcingulate. [273,274]. It was opined that the mesiobasal limbic type was the most amenable to selective procedures that seeked to remove only the offending mesial structures (amygdala, hippocampus, uncus, parahippocampal gyrus), while preserving temporal neocortex [246]. Thus one might infer from the above information, that a further subcategorization of mesiobasal limbic epilepsy can be made: those with purely amygdalar pathology termed amygdaloid seizures [227], and others with some combination of amygdalar and hippocampal sclerosis. Interestingly Jooma et al. [228] performed resection that were limited to corticoamygdalectomies in patients who manifested epileptiform activity arising from only the amygdala, or the amygdala and the lateral temporal neocortex. In this small series of patients, with follow-up of at least 2 years, seizure-freedom was obtained in all 8 patients undergoing such limited resections [228]. Is thus might appear that there is a subset of patients that may benefit from a very selective procedure targeted at the amygdala alone. There is however, evidence that the amygdala does not play such a central role in temporal lobe seizures, although admittedly, many of these reports do not subcategorize patients electrophysiologically to make such a claim. Evidence to the contrary has been presented by Goldring, who performed 70 temporal lobectomies, which included a 4–4.5 cm lateral resection, 2.5 cm of hippocampus and associated parahippocampal gyrus, with sparing of the amydala [275]. Patients were selected primarily on the epidural ECoG results, although MRI, scalp EEG and PET scans were used as ancillary test. In a recent review regarding seizure outcomes following temporal lobectomy mean seizure freedom rates calculated from a total of 99 studies, was 67%, with a interquartile range of 59–77%, with a mean follow- up period of 2.9 years [136]. Considering a 4.1 year followup, only 23 of 70 (33%) patients were seizure
2701
2702
160
Selective amygdalo-hypocampectomy
free in Goldring’s study [276], which is lower than most studies with a 5 year follow-up [277,278]. Seizure free rates in 100 patients operated on at the MNI were similar weather a major or minor hippocampectomy was performed in conjunction with an amygdalectomy, with approximately 39% of patients being seizure free during a follow-up period that spanned almost two decades [226,227]. Given that 8 out of 70 patients in the study by Jooma et al., had seizure activity restricted to the amygdala and adjacent neocortex, it is interesting to speculate that if these patients were ‘‘missed’’ during operations that spare the amydala then a higher seizure free rate might be have been obtained [275]. It is however likely that for the majority of patients with mesiobasal limbic type epilepsy that resection of some component of the limbic circuitry is all that is necessary to alter seizure frequency, whether the amygdala is included or not [136].
Leisonectomy Seizures resulting from focal application of alumina to the motor cortex of monkeys generates epileptiform activity in surrounding cortex that continues despite the removal of the inciting event [279]. It is not until this areas of electrically abnormal tissue, or ‘‘margin’’ of tissue, is removed that seizure activity ceases [279]. Such data would argue that removing the lesion plus any electrophysiologically abnormal surrounding tissue would be the most effective approach to lesional TLE, and that incomplete resection of the lesion would be suboptimal. Indeed although the former still remains controversial there is a significant clinical literature to suggest that the later indeed holds true [103,280]. In the study by Awad et al. (1991) 17 of the 18 patients that underwent complete resection of their lesion were rendered seizure despite the extent of resection of the epileptogenic region [280]. Conversely only 11 out of 29 individuals who had a
partial resection or no resection of the lesion were seizure free post-operatively [280]. It is important however to recognize that in this study lesionectomy included a 2 cm margin surrounding the lesion. The conceptual oversimplification presented in > Figure 160-3 has exceptions in the context of the lesional and non-lesional TLE. The side of MTS or location of a lesion may not always represent the site from which seizures originate. The Seattle group reported on a retrospective cohort of 20 patients who had discordance between a single structural abnormality (MTS or otherwise) and the epileptogenic zone determined from intracranial recordings [281]. Interestingly of this 20, 12 had unilateral MTS on MRI of which 6 underwent a contralateral TL resection including resection of the mesial structures resulting in half being seizure with at least 3 year follow-up [281]. Of the total cohort of 20, over 85% of the patient were either seizure-free or had >75% reduction in their seizure frequency [281]. It is important to note that although this group only represented 5% of all the surgical cases performed at this center, they highlight the need for appropriate pre-surgical assessment and the complexity that underlies the focal epilepsies. When approaching a lesionectomy, one must consider the margin of tissue surrounding the lesion that should be removed. As reviewed by Fried and Cascino [282] various types of margins can be defined including: (1) Radiological margins; (2) gross surgical margins; (3) histological margins, and; (4) electrophysiological margins. In practice one typically removes as much abnormal tissue that is grossly visible and that can be safely removed (gross surgical margin) plus an unavoidable amount of normal appearing brain tissue at the lesion brain interface. Further resection then might then depend on available resources which may include intra-operative MRI (radiological margin), electrocorticography (electrophysiological margins), or repeated quick sections (histological margins). Cavernous mal-
Selective amygdalo-hypocampectomy
formations (CMs) present a convenient example of a discrete lesion that is clearly demarcated from a surrounding region of gliotic and hemosiderin leaden brain (akin to the gliotic brian surround the alumina lesion mentioned above), which is then further demarcated from grossly normal appearing brain tissue. Like other clinical and experimental lesions [283] that don’t include neural tissue CMs are not epileptogenic in of themselves, but structural and functional changes induced in the surrounding brain tissue render the hemisoderin stained margin epileptogenic [284]. Consistent with these observations, the extent to which the hemisoderin stained margin is removed has been shown to correlate to seizure outcome [285], with a more complete removal affording greater seizure control. Unlike CMs, some lesions may not have such distinct demarcation from the surrounding brain tissue and thus the intra-operative strategy may involve delineating the margin of resection through ECoG [286], stereotactic lesionectomy [287], and histological margins [288]. Indeed ECoG may be applied to any setting, and may provide additional seizure control during the resection of CMs and arteriovenous malformations (AVMs) [289].
Dual Pathology As mentioned above dual pathology refers to the coexistence of MTS with an extrahippocampal lesion that is either temporal or extra-temporal. We will focus on extra-hippocampal lesions that remain within the confines of the TL. The basic problem is in determining the full extent of the epileptogenic zone give the presence of two pathological substrates that can independently cause TLE and whose removal in the setting of single pathology can effect abatement of seizures with relatively high probability [63,290]. When this condition is present it would appear that further resection of the sclerotic hippocampus and adjacent neocortex can transform a failed
160
lesionectomy operation into a successful operation [291]. It is unclear in this report as to whether or not the pre-surgical evaluation suggested multiple generators, although it appeared that ictal onsets were either ‘‘focal’’ or ‘‘lobar’’ being restricted to a single contact, or single lobe respectively [291]. Retrospective analysis of a much larger cohort found that seizure freedom was achieved in 73% of those who underwent mesial temporal resection in addition to lesionectomy [102]. The seizure freedom rate was considerable poorer when only one of the two pathologies present was resected. Resection of mesial temporal structures alone obtained a seizure-free rate of 20%, whereas lesionectomy alone resulted in a seizure-free rate of 12% [102].
MRI Normal TLE It is generally accepted that individuals with normal pre-operative MRIs present a unique set of challenges with a concomitantly poorer postsurgical prognosis as compared to those with MTS or foreign tissue lesions [43,70,136]. These individuals will often have a more extensive work-up which will often exploit the utility of intracranial recordings to determine the origin, either mesial, neocortical or both, and the extent of the epileptogenic region [70]. In those individuals undergoing dominant hemispheric surgery the absence of hippocampal pathology is a strong predictor of post-operative neuropsychological decline [250] and hence pre-surgical assessments of memory function and reserve are important in surgical planning. Furthermore if seizures are ultimately determined to be of neocortical origin then the posterior extent of the neocortical resection will often limited by eloquent neocortex. To maximize the neocortical resection and hence to include as much of the epileptogenic zone as possible, intra-operative stimulation mapping is of great utility in determining
2703
2704
160
Selective amygdalo-hypocampectomy
areas of eloquence to permit a more aggressive and safer resection [165]. The absence of a localizing lesion on MRI is considered a poor prognostic sign [43]. Individuals in this category are unlikely to obtain the same rates of seizure freedom post TL surgery as those with MTS despite the added investigations these individuals often undergo [70]. Nonetheless it is estimated that despite the challenges that a normal MRI presents in an individual with TLE, between 40 and 60% can expect seizure freedom following surgery [43,74].
Other Surgical Interventions In this chapter we have focused on the most effective treatment for medically refractory TLE, that being open surgical ablation of the putative epileptogenic zone. Other existing and emerging technologies exist that seek to either ablate areas non-invasively, or functionally modify the brain through electrical stimulation to effect control of seizure activity.
Stereotactic Radiosurgery (SRS)
Outcomes Following Resective Surgery As has been mentioned the primary goal of resective surgery is seizure freedom, and its attendant benefits (see section ‘‘Goals of surgery’’). However these benefits do not come without their risks. Surgical risk associated with invasive diagnostic tests has been discussed, however those associated with resective procedures themselves that have not been reviewed in this chapter, but can be found elsewhere [165]. The two most fundamental outcomes of resective procedures which are seizure rates post surgery, and neuropsychological status have been compared and contrasted by type of surgical approach herein, Valiante [292]. Other variables that should be considered in a thorough discussion of outcomes include associated costs to achieve a desired outcome, as well as the overall utility of resective surgery which falls under the rubric of quality of life. The former will not be dealt with here, whereas the later has been discussed briefly in a previous section and further detail can be found in selected references [293–297].
Somewhat serendipitously it was observed during treatment of AVMs with SRS that in addition to ablation of the AVM a high rate of seizure freedom was obtained in those individuals who initially had presented with seizures [298]. Specific application to lesional epilepsy soon followed with similar successes [299–302]. This was extended to non-leisonal mTLE is a series of four patients who had undergone the usual pre-surgical evaluation and deemed to be candidates for resection of mesial TL structures [303]. More recently a prospective study reported a seizure-free rate of 65% in a cohort of 21 patients with a minimum of 2 years of follow-up [304]. However unlike open ablative surgery, seizurefreedom was achieved in a gradual fashion waning over an approximately 2 year period [304]. This period has been characterized by an increase in seizures and auras, and steroid responsive signs and symptoms of increased intracranial pressure most likely as a result of the rather significant edema that is often visualized on the post-radiosurgery MRI scan [304,305]. The purported benefits of less neuropsychological decline, and similar seizure-free rates as compared to open resective procedures has yet to be definitively addressed [306], although given the variability of practice as it relates to the surgical management of epilepsy this might ultimately prove to be a rather formidable goal.
Selective amygdalo-hypocampectomy
Neuromodulation Various open-loop, and semi-closed loop approaches have been developed in an attempt to modulate pathological brain activity. This generally involves implantation of electrodes into or on structures that are either thought to mediate or modulate seizure activity. The electrodes are then attached to a battery operated device that generates a stimulus train that is delivered to the brain region of interest. Open-loop devices have no sensing capability and thus the stimulus they deliver, which is usually a continuous pulse train, has no bearing on going neural activity. Electrodes attached to this type of device have been attached or implanted on the vagal nerve [307], hippocampus [308], and anterior nucleus of the thalamus [309]. A semi-closed loop device modulates a fixed stimulus, again usually a pulse train, in an on off fashion in response to identification of abnormal brain activity. One such device involves implantation of cortical electrodes and a stimulation pack that is recessed into the calvarium [310,311]. The above mentioned devices are indicated when resective surgery is not an option, with their ultimate effectiveness and cost-utility still yet to be determined.
Summary TLE is the most common cause of medically refractory epilepsy. The most common manifestation of TLE are CPSs which are usually associated with a past history of febrile seizures, and more often than not occur in association with MTS. Appropriate imaging will often disclose MTS or a foreign tissue lesion, whereas a normal MRI is a less likely but not uncommon finding in those with TLE. Failure to control CPS following sequential administration of two first line ASM should prompt early referral to a specialized epilepsy center for further investigations. The
160
ultimate determination of surgical candidacy can be made by non-invasive test in the majority of those with TLE. Nonetheless, intracranial recordings may be required in those with bitemporal abnormalities, discordance amongst the various pre-operative tests, nTLE, and those with normal MRI scans. Those individuals with CPS and suspected TLE who are ultimately deemed to be resective surgical candidates can expect a 60% chance of seizure remission in the first year following surgery [63], and this number is considerably higher if they are found to have mTLE from MTS or lesional TLE [61,62]. From a societal perspective surgery for TLE is cost saving [52]. From an individual perspective greater biological, psychological, and social benefits can be derived from earlier intervention in the course of their disease [53–56] with seizure-freedom, the ultimate goal, affording the individual with epilepsy the ability to partake in the ‘‘activities of daily living, and derive the satisfaction of accomplishment which is the ultimate outcome of life. . .’’ [312].
References 1. Broca P. Remarques sur le sie`ge de la faculte´ du language articule´, suives d’une observation d’aphe´mie (perte del la parole). Bull Soc Anat 1861;280:834-43. 2. Jackson JH, Colman WS. Case of epilepsy with tasting movements and ‘‘dreamy state’’ with very small patch of softening in the left uncinate gyrus. Brain 1898;21:580-90. 3. Fristsch G, Hitzig E. Ueber die elektrische Erregbarkeit des Grosshirns. Arch Anat Physiol wiss Med 1870;300-32. 4. Ferrier D. The functions of the brain. London: Smith, Elder; 1876. 5. Sherrington CS. The inntegrative actions of the nervous system. NY: Scribner’s; 1906. 6. Ferrier D. The localization of function in the brain (MS) communicated by J. B. Sanderson 5 March 1874. Archives of the Royal Society, AP.56.2; abstract in Proc R Soc 1874;2. 7. Bennnett AH, Godlee RJ. Case of cerebral tumor. MedChir Trans 1885;68:243-275. 8. Tan TC, Black PM. Sir Victor Horsley (1857–1916): pioneer of neurological surgery. Neurosurgery 2002;50 (3):607-11; discussion 611–2.
2705
2706
160
Selective amygdalo-hypocampectomy
9. Vilensky JA, Gilman S. Horsley was the first to use electrical stimulation of the human cerebral cortex intraoperatively. Surg Neurol 2002;58(6):425-6. 10. Horsley V. Brain surgery. BMJ 1886;2:670-5. 11. Caton R. The electrical currents of the brain. BMJ 1875;2:278. 12. Berger H. Uber das Elektrenkephlogram des Menshen. Arch Psychiatr Nervenkr 1929;87:527-70. 13. Gibbs FA, Gibbs EL. Cerebral dysrhythmias of epilepsy. Arch Neurol Psychiatry 1938;39:298. 14. Jasper H, Kershman J. Electroencephalographic classification of the epilepsies. Arch Neurol Psychiatry 1941;45:903. 15. Bailey P, Gibbs F. The surgical treatment of psychomotor epilepsy. JAMA 1951;145:365. 16. Penfield W, Flanigin H. Surgical therapy of temporal lobe seizures. Arch Neurol Psychiatry 1950;64:491-500. 17. Earle K, Baldwin M, Penfield W. Incisural sclerosis and temporal lobe seizures produced by hippocampal herniation at birth. Arch Neurol Psychiatry 1953;69:27-42. 18. Penfield W, Baldwin M. Temporal lobe seizures and the technique of subtotal temporal lobectomy. Ann Surg 1952;136:625-34. 19. Falconer M. Discussion on the surgery of temporal lobe epielpsy: surgical and pathological aspects. Proc R Soc Med 1953;46:971-4. 20. Falconer MA, et al. Treatment of temporal-lobe epilepsy by temporal lobectomy; a survey of findings and results. Lancet 1955;268(6869):827-35. 21. Milner B. Psychological defects produced by temporal lobe excision. Res Publ Assoc Res Nerv Ment Dis 1958;36:244-57. 22. Milner B, Penfield W. The effect of hippocampal lesions on recent memory. Trans Am Neurol Assoc 1955(80th Meeting):42–8. 23. Scoville WB. The limbic lobe in man. J Neurosurg 1954;11(1):64-6. 24. Scoville WB, Milner B. Loss of recent memory after bilateral hippocampal lesions. J Neurol Neurosurg Psychiatry 1957;20(1):11-21. 25. Reynolds EH. The ILAE/IBE/WHO epilepsy global campaign history. International League Against Epilepsy. International Bureau for Epilepsy. Epilepsia 2002;43 Suppl 6:9-11. 26. Theodore WH, et al. Epilepsy in North America: a report prepared under the auspices of the global campaign against epilepsy, the International Bureau for Epilepsy, the International League Against Epilepsy, and the World Health Organization. Epilepsia 2006;47(10):1700-22. 27. Cockerell OC, et al. Prognosis of epilepsy: a review and further analysis of the first nine years of the British National General Practice Study of Epilepsy, a prospective population-based study. Epilepsia 1997;38(1):31-46. 28. Functional Brain Imaging (pre-edit proof), Medical Advisory Secretariat, Ministry of Health and Long Term Care for Ontario Health Technology Advisory Committee, http://www.health.gov.on.ca/english/providers/
29. 30. 31. 32.
33. 34.
35.
36.
37.
38.
39. 40. 41.
42.
43.
44.
45. 46.
program/ohtac/tech/recommend/rec_fbi_012507.pdf, Last accessed Nov 2007. WHO, Atlas – epilepsy care in the world 2005. 2005, World Health Organization: Geneva. Engel J Jr. Finally, a randomized, controlled trial of epilepsy surgery. N Engl J Med 2001;345(5):365-7. McLachlan RS. Commentary on epilepsy surgery in Canada. Can J Neurol Sci 2001;28(1):4-5. Proposal for revised classification of epilepsies and epileptic syndromes. Commission on Classification and Terminology of the International League Against Epilepsy. Epilepsia 1989;30(4):389–99. SIGN, Diagnosis and management of epilepsy in adults, in SIGN Guideline No 70, revised version. 2003. NICE, The epilepsies: diagnosis and management of the epilepsies in children and young people in primary and secondary care, in (NICE) Clinical Guideline 20. 2004, National Institute for Clinical Excellence. Engel J Jr, et al. Practice parameter: temporal lobe and localized neocortical resections for epilepsy: report of the Quality Standards Subcommittee of the American Academy of Neurology, in association with the American Epilepsy Society and the American Association of Neurological Surgeons. Neurology 2003;60(4):538-47. Engel JJ, Shewmon DA. Who should be considered a surgical candidate? In: Engel JJ, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1991. Mattson RH, Cramer JA, Collins JF. Prognosis for total control of complex partial and secondarily generalized tonic clonic seizures. Department of Veterans Affairs Epilepsy Cooperative Studies No. 118 and No. 264 Group. Neurology 1996;47(1):68-76. Wieser HG, et al. Surgically remediable temporal lobe syndromes. In: Engel JJ, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1991. Kwan P, Brodie MJ. Early identification of refractory epilepsy. N Engl J Med 2000;342(5):314-9. Fisher RS. Imitators of epilepsy. New York: Demos Publication; 1994. Wagner EH, Austin BT, Von Korff M. Organizing care for patients with chronic illness. Milbank Q 1996;74 (4):511-44. Gruman J, et al. Organizing health care for people with seizures and epilepsy. J Ambul Care Manage 1998;21 (2):1-13; discussion 14–7. Berkovic SF, et al. Preoperative MRI predicts outcome of temporal lobectomy: an actuarial analysis. Neurology 1995;45(7):1358-63. Engel J Jr. Recent advances in surgical treatment of temporal lobe epilepsy. Acta Neurol Scand Suppl 1992;140:71-80. Ojemann GA. Temporal lobe epilepsy. In: Clinical neurosurgery. Montreal, QC: Williams & Wilkins; 1996. Jacoby A. Epilepsy and the quality of everyday life. Findings from a study of people with well-controlled epilepsy. Soc Sci Med 1992;34(6):657-66.
Selective amygdalo-hypocampectomy
47. Bien CG, et al. Assessment of the long-term effects of epilepsy surgery with three different reference groups. Epilepsia 2006;47(11):1865-9. 48. Sperling MR, et al. Occupational outcome after temporal lobectomy for refractory epilepsy. Neurology 1995;45 (5):970-7. 49. Gilliam F, et al. Patient-oriented outcome assessment after temporal lobectomy for refractory epilepsy. Neurology 1999;53(4):687-94. 50. Sperling MR, et al. Seizure control and mortality in epilepsy. Ann Neurol 1999;46(1):45-50. 51. Rausch R, Crandall PH. Psychological status related to surgical control of temporal lobe seizures. Epilepsia 1982;23(2):191-202. 52. Wiebe S, et al. An economic evaluation of surgery for temporal lobe epilpesy. J Epilepsy 1995;8:227-35. 53. Dlugos DJ. The early identification of candidates for epilepsy surgery. Arch Neurol 2001;58(10):1543-6. 54. Jeong SW, et al. Prognostic factors for the surgery for mesial temporal lobe epilepsy: longitudinal analysis. Epilepsia 2005;46(8):1273-9. 55. Janszky J, et al. Temporal lobe epilepsy with hippocampal sclerosis: predictors for long-term surgical outcome. Brain 2005;128(Pt 2):395-404. 56. Hennessy MJ, et al. Predictors of outcome and pathological considerations in the surgical treatment of intractable epilepsy associated with temporal lobe lesions. J Neurol Neurosurg Psychiatry 2001;70(4):450-8. 57. Berg AT, et al. How long does it take for partial epilepsy to become intractable? Neurology 2003;60(2):186-90. 58. Trevathan E, Gilliam F. Lost years: delayed referral for surgically treatable epilepsy. Neurology 2003;61 (4):432-3. 59. Yoon HH, et al. Long-term seizure outcome in patients initially seizure-free after resective epilepsy surgery. Neurology 2003;61(4):445-50. 60. Wiebe S, et al. A randomized, controlled trial of surgery for temporal lobe epilepsy. N Eng J Med 2001;345:311-18. 61. Jack CR Jr, et al. Magnetic resonance image-based hippocampal volumetry: correlation with outcome after temporal lobectomy. Ann Neurol 1992;31(2):138-46. 62. Cascino GD, et al. Electrocorticography and temporal lobe epilepsy: relationship to quantitative MRI and operative outcome. Epilepsia 1995;36(7):692-6. 63. Wiebe S, et al. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med 2001;345(5):311-8. 64. Fish D, Andermann F, Olivier A. Complex partial seizures and small posterior temporal or extratemporal structural lesions: surgical management. Neurology 1991;41 (11):1781-4. 65. Williamson PD, et al. Parietal lobe epilepsy: diagnostic considerations and results of surgery. Ann Neurol 1992;31 (2):193-201. 66. Williamson PD, et al. Occipital lobe epilepsy: clinical characteristics, seizure spread patterns, and results of surgery. Ann Neurol 1992;31(1):3-13.
160
67. Blume WT, Girvin JP, Stenerson P. Temporal neocortical role in ictal experiential phenomena. Ann Neurol 1993;33 (1):105-7. 68. Geyer JD, et al. Postictal nose-rubbing in the diagnosis, lateralization, and localization of seizures. Neurology 1999;52(4):743-5. 69. Jones-Gotman, M. Clinical neuropsychology and neocortical epilepsies. Adv Neurol 2000;84:457-62. 70. Madhavan D, Kuzniecky R. Temporal lobe surgery in patients with normal MRI. Curr Opin Neurol 2007;20(2):203-7. 71. Hamer HM, et al. Interictal epileptiform discharges in temporal lobe epilepsy due to hippocampal sclerosis versus medial temporal lobe tumors. Epilepsia 1999;40(9):1261-8. 72. Burgerman RS, et al. Comparison of mesial versus neocortical onset temporal lobe seizures: neurodiagnostic findings and surgical outcome. Epilepsia 1995;36(7):662-70. 73. Wennberg R, et al. Preeminence of extrahippocampal structures in the generation of mesial temporal seizures: evidence from human depth electrode recordings. Epilepsia 2002;43(7):716-26. 74. Jung WY, Pacia SV, Devinsky R. Neocortical temporal lobe epilepsy: intracranial EEG features and surgical outcome. J Clin Neurophysiol 1999;16(5):419-25. 75. Falconer MA, Driver MV, Serafetinides EA. Temporal lobe epilepsy due to distant lesions: two cases relieved by operation. Brain 1962;85:521-39. 76. Cavanagh JB. On certain small tumours encountered in the temporal lobe. Brain 1958;81(3):389-405. 77. Bouchet and Cazauvieilh, De l’e´pilepsie considere´e dans ses rapports avec l’alie´nation mentale. Arch Gen Med 1826;9:510‐42. 78. Plate KH, et al. Neuropathological findings in 224 patients with temporal lobe epilepsy. Acta Neuropathol 1993;86(5):433-8. 79. Wolf HK, et al. Surgical pathology of temporal lobe epilepsy. Experience with 216 cases. J Neuropathol Exp Neurol 1993;52(5):499-506. 80. Babb TL. Pathological findings in epilepsy. In: Engel JJ, editor. Surgical treatment of the epilepsies. New York: Raven Press, Ltd.; 1987. p. 511-40. 81. Sommer W. Erkrankung des Ammonshorns als aetiologisches Moment der epilepsie. Arch Psychiatr Nervenkr 1880;10:631-75. 82. Margerison JH, Corsellis JA. Epilepsy and the temporal lobes. A clinical, electroencephalographic and neuropathological study of the brain in epilepsy, with particular reference to the temporal lobes. Brain 1966;89 (3):499-530. 83. Sano K, Malamud N. Clincal significance of sclerosis of the Cornu Ammonis. Arch Neurol Psychiatry 1953;70:40-53. 84. Houser CR. Granule cell dispersion in the dentate gyrus of humans with temporal lobe epilepsy. Brain Res 1990;535 (2):195-204.
2707
2708
160
Selective amygdalo-hypocampectomy
85. Cotman CW, Nadler JV. Reactive synaptogenesis in the hippocampus. In: Cotman CW, editor. Neuronal plastcity. New York: Raven Press; 1978. p. 227-71. 86. Dam AM. Epilepsy and neuron loss in the hippocampus. Epilepsia 1980;21(6):617-29. 87. Dube CM, et al. Fever, febrile seizures and epilepsy. Trends Neurosci 2007;30(10):490-6. 88. Falconer MA, Serafetinides EA, Corsellis JA. Etiology and pathogenesis of temporal lobe epilepsy. Arch Neurol 1964;10:233-48. 89. Cendes F, Andermann E. Do febrile seizures: promote temporal lobe epilepsy? In: Baram TZ, Shinnar S, editors. Febrile seizures. New York, Academic Press; 2002. 90. Hedera P, et al. Familial mesial temporal lobe epilepsy maps to chromosome 4q13.2-q21.3. Neurology 2007;68 (24):2107-12. 91. Berkovic SF, et al. Familial temporal lobe epilepsy: a common disorder identified in twins. Ann Neurol 1996;40(2):227-35. 92. Quesney LF, Risinger MW, Shewmon DA. Extracranial EEG evaluation. In: Engel JJ, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1991. p. 173-95. 93. Wieser HG, et al. Surgically remediable temporal lobe syndromes. In: Engel JJ, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1993. 94. Carne RP, et al. MRI-negative PET-positive temporal lobe epilepsy: a distinct surgically remediable syndrome. Brain 2004;127(Pt 10):2276-85. 95. Sylaja PN, et al. Seizure outcome after anterior temporal lobectomy and its predictors in patients with apparent temporal lobe epilepsy and normal MRI. Epilepsia 2004;45(7):803-8. 96. Chapman K, et al. Seizure outcome after epilepsy surgery in patients with normal preoperative MRI. J Neurol Neurosurg Psychiatry 2005;76(5):710-3. 97. Duncan JS. Imaging and epilepsy. Brain 1997;120 (Pt 2):339-77. 98. Alvarez-Linera Prado J. 3-Tesla MRI and temporal lobe epilepsy. Semin Ultrasound CT MR 2007;28(6):451-61. 99. Cendes F, et al. Frequency and characteristics of dual pathology in patients with lesional epilepsy. Neurology 1995;45(11):2058-64. 100. Levesque MF, et al. Surgical treatment of limbic epilepsy associated with extrahippocampal lesions: the problem of dual pathology. J Neurosurg 1991;75(3):364-70. 101. Kim JH, et al. Hippocampal neuronal density in temporal lobe epilepsy with and without gliomas. Acta Neuropathol 1990;80(1):41-5. 102. Li LM, et al. Surgical outcome in patients with epilepsy and dual pathology. Brain 1999;122(Pt 5):799-805. 103. Li LM, et al. Surgical treatment of patients with single and dual pathology: relevance of lesion and of hippocampal atrophy to seizure outcome. Neurology 1997;48 (2):437-44. 104. Stokes T, et al. Clinical guidelines and evidence review for the epilepsies: diagnosis and management in adults
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
and children in primary and secondary care. London: Royal College of General Practitioners; 2004. Cascino GD, et al. Routine EEG and temporal lobe epilepsy: relation to long-term EEG monitoring, quantitative MRI, and operative outcome. Epilepsia 1996;37 (7):651-6. Lu¨ders HO, Awad I. Conceptual considerations. In: Lu¨ders HO, editor. Epilepsy surgery. New York: Raven Press; 1992. p. 51-62. Dinner DS, Lu¨ders H, Klem G. Chronic electroencephalography: Cleveland clinic experience. Electroencephalogr Clin Neurophysiol Suppl 1998;48:58-69. Lu¨ders H, Engel JJ, Munari C. General Principles. In: Engel JJ, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1993. p. 137-53. Berkovic SF, et al. Hippocampal sclerosis in temporal lobe epilepsy demonstrated by magnetic resonance imaging. Ann Neurol 1991;29(2):175-82. Cascino GD, et al. Magnetic resonance imaging-based volume studies in temporal lobe epilepsy: pathological correlations. Ann Neurol 1991;30(1):31-6. Jackson GD, et al. MRI detection of hippocampal pathology in temporal lobe epilepsy: increased sensitivity using quantitative T2 relaxometry. Neurology 1993;43:1793-99. Wada JA. A new method for the determination of the side of cerebral speech dominance. Igaku to Seibutsugaki, Tokyo 1949;14:221-2. Jones-Gotman M, et al. Etomidate speech and memory test (eSAM): a new drug and improved intracarotid procedure. Neurology 2005;65(11):1723-9. Medina LS, Bernal B, Ruiz J. Role of functional MR in determining language dominance in epilepsy and nonepilepsy populations: a Bayesian analysis. Radiology 2007;242(1):94-100. Tharin S, Golby A. Functional brain mapping and its applications to neurosurgery. Neurosurgery 2007;60 (4 Suppl 2):185-201; discussion 201–2. Engel J Jr, Kuhl DE, Phelps ME. Patterns of human local cerebral glucose metabolism during epileptic seizures. Science 1982;218(4567):64-6. Whiting P, et al. A systematic review of the effectiveness and cost-effectiveness of neuroimaging assessments used to visualise the seizure focus in people with refractory epilepsy being considered for surgery. Health Technol Assess 2006;10(4):1-250, iii-iv. O’Brien TJ, et al. Subtraction ictal SPECT co-registered to MRI improves clinical usefulness of SPECT in localizing the surgical seizure focus. Neurology 1998;50(2):445-54. Detre JA, et al. Localization of subclinical ictal activity by functional magnetic resonance imaging: correlation with invasive monitoring. Ann Neurol 1995;38(4):618-24. Hoffmann A, et al. Electroencephalography during functional echo-planar imaging: detection of epileptic spikes using post-processing methods. Magn Reson Med 2000;44(5):791-8.
Selective amygdalo-hypocampectomy
121. Knowlton RC, Shih J. Magnetoencephalography in epilepsy. Epilepsia 2004;45:Suppl 4:61-71. 122. Papanicolaou AC, et al. Magnetocephalography: a noninvasive alternative to the Wada procedure. J Neurosurg 2004;100(5):867-76. 123. Jones-Gotman M. Neuropsychological testing for localizing and lateralizing the epileptogenic region. In: Engel J, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1993. p. 203-11. 124. Jones-Gotman M. Commentary: psychological evaluation-testing hippocampal function. In: Engel J, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1987. p. 203-11. 125. Helmstaedter C, Elger CE. Cognitive consequences of two-thirds anterior temporal lobectomy on verbal memory in 144 patients: a three-month follow-up study. Epilepsia 1996;37(2):171-80. 126. Wada JA. A new method for determination of the side of cerebral dominance: a preliminary report on the intra-carotid injection of sodium amytal in man. Igaku Seibutsugaki(Medicine and Biology) 1949;14:221-2. 127. Milner B, Branch C, Rasmussen T. Study of short-term memory after intracarotid injection of sodium Amytal. Trans Am Neurol Assoc 1962;87:224-6. 128. Ojemann JG, Kelley WM. The frontal lobe role in memory: a review of convergent evidence and implications for the Wada memory test. Epilepsy Behav 2002;3 (4):309-15. 129. Baxendale S, Thompson PJ, Duncan JS. The role of the Wada test in the surgical treatment of temporal lobe epilepsy: an international survey. Epilepsia 2008;49 (4):715-20. 130. Helmstaedter C, Kurthen M. Validity of the WADA test. Epilepsy Behav 2002;3(6):562-3. 131. Abou-Khalil B. An update on determination of language dominance in screening for epilepsy surgery: the Wada test and newer noninvasive alternatives. Epilepsia 2007;48(3):442-55. 132. Abou-Khalil BW. How close is fMRI to providing the memory component of the wada test? Epilepsy Curr 2005;5(5):184-6. 133. Richardson MP, et al. Pre-operative verbal memory fMRI predicts post-operative memory decline after left temporal lobe resection. Brain 2004;127(Pt 11):2419-26. 134. Wieser HG, et al. Comparative value of spontaneous and chemically and electrically induced seizures in establishing the lateralization of temporal lobe seizures. Epilepsia 1979;20(1):47-59. 135. Engel JJ. Approaches to localization of the epileptogenic lesion. In: Engel JJ, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1987. p. 75-95. 136. McIntosh AM, Wilson SJ, Berkovic SF. Seizure outcome after temporal lobectomy: current research practice and findings. Epilepsia 2001;42(10):1288-307.
160
137. Chang BS, et al. Outpatient EEG monitoring in the presurgical evaluation of patients with refractory temporal lobe epilepsy. J Clin Neurophysiol 2002;19(2):152-6. 138. Cendes F, et al. Is ictal recording mandatory in temporal lobe epilepsy? Not when the interictal electroencephalogram and hippocampal atrophy coincide. Arch Neurol 2000;57(4):497-500. 139. Guerreiro CA, et al. Daytime outpatient versus inpatient video-EEG monitoring for presurgical evaluation in temporal lobe epilepsy. J Clin Neurophysiol 2002;19 (3):204-8. 140. Spencer SS, et al. Reliability and accuracy of localization by scalp ictal EEG. Neurology 1985;35(11):1567-75. 141. So N, et al. Depth electrode investigations in patients with bitemporal epileptiform abnormalities. Ann Neurol 1989;25(5):423-31. 142. Engel J, Jr, et al. Correlation of criteria used for localizing epileptic foci in patients considered for surgical therapy of epilepsy. Ann Neurol 1981;9(3):215-24. 143. Ajmone-Marsan C, Postscript. When are noninvasive tests enough? In:Engel JJ, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1991. p. 313. 144. Kirschner M. Electrocoagulation des Ganglion Gasseri. Zentralbl Chir 1932;47:2841-3. 145. Wieser HG, Quesney F, Morris HHI. Foramen Ovale and Peg Electrodes. In: Engel JJ, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1991. p. 331-9. 146. Spencer DD. Depth electrode implantation at Yale University. In: Engel JJ, editor. Surgical treatment of the epilepsies. New York: Raven Press, Ltd.; 1987. p. 603-8. 147. Kovalev D, et al. Rapid and fully automated visualization of subdural electrodes in the presurgical evaluation of epilepsy patients. AJNR Am J Neuroradiol 2005;26 (5):1078-83. 148. Atamai, Atamai Interactive Visualization, http://www. atamai.com/, Last accessed. 149. Sansur CA, et al. Incidence of symptomatic hemorrhage after stereotactic electrode placement. J Neurosurg 2007;107(5):998-1003. 150. Wyler AR, Walker G, Somes G. The morbidity of longterm seizure monitoring using subdural strip electrodes. J Neurosurg 1991;74(5):734-7. 151. Spencer SS. Depth versus subdural electrode studies for unlocalized epilepsy. J Epilepsy 1989;2:123-7. 152. Burneo JG, et al. Morbidity associated with the use of intracranial electrodes for epilepsy surgery. Can J Neurol Sci 2006;33(2):223-7. 153. Wiggins GC, Elisevich K, Smith BJ. Morbidity and infection in combined subdural grid and strip electrode investigation for intractable epilepsy. Epilepsy Res 1999;37(1):73-80. 154. Korinek AM, et al. Risk factors for neurosurgical site infections after craniotomy: a critical reappraisal of antibiotic prophylaxis on 4,578 patients. Br J Neurosurg 2005;19(2):155-62.
2709
2710
160
Selective amygdalo-hypocampectomy
155. Penfield W, Jasper H. Epilepsy and the functional anatomy of the human brain. Boston, MA: Little, Brown & Co.; 1954. 156. Ojemann GA. Intraoperative tailoring of temporal lobe resections. In: Engel J, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1993. p. 481-8. 157. Niemeyer P. The transventricular amygdalo-hippocampectomy in temporal lobe epilepsy. In: Baldwin M, Bailey P, editors. Temporal lobe epilepsy. Springfield, IL: Charles C Thomas; 1958. 158. Yasargil GM, Teddy PJ, Roth P. Selective amygdalohippocampectomy: operative anatomy and surgical technique. In: Symon L, editor. Advances and technical standards in Neurosurgery. Austria: Springer; 1985. 159. Fenyes I, Zoltan L, Fenyes G. Temporal epilepsies with deep-seated epileptogenic foci. Postoperative course. Arch Neurol 1961;4:559-71. 160. McBride MC, et al. Predictive value of intraoperative electrocorticograms in resective epilepsy surgery. Ann Neurol 1991;30(4):526-32. 161. Binnie CD, et al. Epilepsy Surgery: non-invasive versus invasive focus localization. In: Dam AM, et al., editors. Electrocorticoggraphy and stimulation. Coopenhagen: Munksgaard; 1994 (Acta Neurol Scand Suppl). 162. Chatrian GE, et al. Role of the ECoG in tailored temporal lobe resection: the University of Washington experience. Electroencephalogr Clin Neurophysiol Suppl 1998;48:24-43. 163. Ojemann G, et al. Cortical language localization in left, dominant hemisphere. An electrical stimulation mapping investigation in 117 patients. J Neurosurg 1989;71(3):316-26. 164. MyVu. http://www.myvu.com/, Last accessed April 2008. 165. Pilcher WH, Ojemann G. Presurgical evaluation and Epilepsy Surgery. In: Apuzzo MLJ, editor. Brain surgery: complication avoidance and management. New York: Churchill Livingstone; 1993. p. 1525-55. 166. Wen HT, et al. Microsurgical anatomy of the temporal lobe: part 1: mesial temporal lobe anatomy and its vascular relationships as applied to amygdalohippocampectomy. Neurosurgery 1999;45(3):549-91; discussion 591–2. 167. Olivier A. Surgical techniques in temporal lobe epilepsy. In: Congress of Neurological Surgeons. Montreal, QC: Williams & Wilkins; 1996. 168. Penfield W, Roberts L. Speech and brain mechanisms. Princeton, N.J.: Princeton University Press; 1959. 169. Wernicke C. Der aphasische Symptomenkomplex. Breslau: Cohen and Weigart; 1874. 170. Lu¨ders H, et al. Basal temporal language area demonstrated by electrical stimulation. Neurology 1986;36 (4):505-10. 171. Lu¨ders H, et al. Basal temporal language area. Brain 1991;114(Pt 2):743-54. 172. Schaffler L, Luders HO, Beck GJ. Quantitative comparison of language deficits produced by extraoperative
173.
174. 175.
176.
177.
178. 179.
180.
181.
182.
183.
184. 185.
186.
187.
188.
electrical stimulation of Broca’s, Wernicke’s, and basal temporal language areas. Epilepsia 1996;37(5):463-75. Schaffler L, et al. Anatomic distribution of cortical language sites in the basal temporal language area in patients with left temporal lobe epilepsy. Epilepsia 1994;35(3):525-8. Krauss GL, et al. Cognitive effects of resecting basal temporal language areas. Epilepsia 1996;37(5):476-83. Meyer V, Yates A. Intellectual changes following temporal lobectomy for psychomotor epilepsy: preliminary communication. J Neurol Neurosurg Psychiatry 1955;18:44-52. Ojemann GA, Dodrill CB. Verbal memory deficits after left temporal lobectomy for epilepsy. Mechanism and intraoperative prediction. J Neurosurg 1985;62 (1):101-7. Milner B. Brain mechanisms suggested by studies of temporal lobes. In: Darley F, editor. Brain mechanisms underlying speech and language. New York: Grune and Stratton; 1967. p. 122-45. Albert ML. Short-term memory and aphasia. Brain Lang 1976;3(1):28-33. Fedio P, Van Buren JM. Memory and perceptual deficits during electrical stimulation in the left and right thalamus and parietal subcortex. Brain Lang 1975;2 (1):78-100. Ojemann GA. Brain organization for language form the perspective of electrical stimulation mapping. Behav Brain Sci 1983;6:189-230. Ojemann GA. Organization of short-term verbal memory in language areas of human cortex: evidence from electrical stimulation. Brain Lang 1978;5(3):331-40. Frisk V, Milner B. The relationship of working memory to the immediate recall of stories following unilateral temporal or frontal lobectomy. Neuropsychologia 1990;28(2):121-35. Suzuki WA, Amaral DG. Topographic organization of the reciprocal connections between the monkey entorhinal cortex and the perirhinal and parahippocampal cortices. J Neurosci 1994;14(3 Pt 2):1856-77. Burwell RD. The parahippocampal region: corticocortical connectivity. Ann N Y Acad Sci 2000;911:25-42. Suzuki WA, Amaral DG. Perirhinal and parahippocampal cortices of the macaque monkey: cortical afferents. J Comp Neurol 1994;350(4):497-533. Sorensen KE, Shipley MT. Projections from the subiculum to the deep layers of the ipsilateral presubicular and entorhinal cortices in the guinea pig. J Comp Neurol 1979;188(2):313-33. Kosel KC, Van Hoesen GW, Rosene DL. Nonhippocampal cortical projections from the entorhinal cortex in the rat and rhesus monkey. Brain Res 1982;244(2):201-13. Goldman-Rakic PS, Selemon LD, Schwartz ML. Dual pathways connecting the dorsolateral prefrontal cortex with the hippocampal formation and parahippocampal
Selective amygdalo-hypocampectomy
189.
190.
191. 192.
193.
194.
195.
196. 197.
198.
199.
200.
201.
202.
203.
204.
205.
cortex in the rhesus monkey. Neuroscience 1984;12 (3):719-43. Witter MP, Groenewegen HJ. Connections of the parahippocampal cortex in the cat. III. Cortical and thalamic efferents. J Comp Neurol 1986;252(1):1-31. Van Hoesen GW. The parahippocampal gyrus. New observations regarding its cortical connections in the monkey. Trends Neurosci 1982;5:345-50. Suzuki WA, Eichenbaum H. The neurophysiology of memory. Ann N Y Acad Sci 2000;911:175-91. Yonelinas AP. The nature of recollection and familiarity – a review of 30 years of research. J Mem Lang 2002;46:441-517. Van Hoesen G, Pandya DN. Some connections of the entorhinal (area 28) and perirhinal (area 35) cortices of the rhesus monkey. I. Temporal lobe afferents. Brain Res 1975;95(1):1-24. Spencer SS, et al. Human hippocampal seizure spread studied by depth and subdural recording: the hippocampal commissure. Epilepsia 1987;28(5):479-89. Penfield W, Milner B. Memory deficit produced by bilateral lesions in the hippocampal zone. Arch Neurol Psyciatry 1958;79:475-97. Squire LR, Zola-Morgan S. The medial temporal lobe memory system. Science 1991;253(5026):1380-6. Helmstaedter C, et al. Right hemisphere restitution of language and memory functions in right hemisphere language-dominant patients with left temporal lobe epilepsy. Brain 1994;117(Pt 4):729-37. Milner B. Disorders of learning and memory after temporal lobe lesions in man. Clin Neurosurg 1972;19:421-46. Viskontas IV, McAndrews MP, Moscovitch M. Remote episodic memory deficits in patients with unilateral temporal lobe epilepsy and excisions. J Neurosci 2000;20 (15):5853-7. Barr WB, et al. Retrograde amnesia following unilateral temporal lobectomy. Neuropsychologia 1990;28 (3):243-55. Nadel L, Moscovitch M. Memory consolidation, retrograde amnesia and the hippocampal complex. Curr Opin Neurobiol 1997;7(2):217-27. Pitkanen A, et al. Reciprocal connections between the amygdala and the hippocampal formation, perirhinal cortex, and postrhinal cortex in rat. A review. Ann N Y Acad Sci 2000;911:369-91. Nieuwenhuys R, Voogd J, van Huijzen C. The human central nervous system: a synopsis and atlas. 3rd ed. New York: Springer; 1988. Krettek JE, Price JL. Amygdaloid projections to subcortical structures within the basal forebrain and brainstem in the rat and cat. J Comp Neurol 1978;178 (2):225-54. Aggleton JP, Mishkin M. Projections of the amygdala to the thalamus in the cynomolgus monkey. J Comp Neurol 1984;222(1):56-68.
160
206. Sanghera MK, Rolls ET, Roper-Hall A. Visual responses of neurons in the dorsolateral amygdala of the alert monkey. Exp Neurol 1979;63(3):610-26. 207. Ben-Ari Y, Le Gal le Salle G, Champagnat JC. Lateral amygdala unit acitvity: I. Relationship between spontaneous and evoked activity. Electroencephalogr Clin Neurophysiol 1974;37(5):449-61. 208. Cahill L, McGaugh JL. Mechanisms of emotional arousal and lasting declarative memory. Trends Neurosci 1998;21(7):294-9. 209. Ure JA, Perassolo M. Update on the pathophysiology of the epilepsies. J Neurol Sci 2000;177(1):1-17. 210. Sloviter RS. The functional organization of the hippocampal dentate gyrus and its relevance to the pathogenesis of temporal lobe epilepsy. Ann Neurol 1994;35 (6):640-54. 211. Bruton C. The neuropathology of temporal lobe epilepsy. New York: Oxford University Press; 1988. 212. Sloviter RS. Decreased hippocampal inhibition and a selective loss of interneurons in experimental epilepsy. Science 1987;235(4784):73-6. 213. Bekenstein JW, Lothman EW. Dormancy of inhibitory interneurons in a model of temporal lobe epilepsy. Science 1993;259(5091):97-100. 214. Sloviter RS. Permanently altered hippocampal structure, excitability, and inhibition after experimental status epilepticus in the rat: the ‘‘dormant basket cell’’ hypothesis and its possible relevance to temporal lobe epilepsy. Hippocampus 1991;1(1):41-66. 215. Stringer JL, Lothman EW. Maximal dentate gyrus activation: characteristics and alterations after repeated seizures. J Neurophysiol 1989;62(1):136-43. 216. Stringer JL, Lothman EW. Reverberatory seizure discharges in hippocampal-parahippocampal circuits. Exp Neurol 1992;116(2):198-203. 217. Sutula T, et al. Mossy fiber synaptic reorganization in the epileptic human temporal lobe. Ann Neurol 1989;26(3):321-30. 218. Houser CR, et al. Altered patterns of dynorphin immunoreactivity suggest mossy fiber reorganization in human hippocampal epilepsy. J Neurosci 1990;10(1):267-82. 219. Ramo´n y Cajal S. Degeneration and regeneration of the nervous system. editor. London: Oxford University Press; 1928. p. 656-92 [Trans and edited: May RM]. 220. Sutula T, et al. Synaptic reorganization in the hippocampus induced by abnormal functional activity. Science 1988;239(4844):1147-50. 221. Collins RC. Functional anatomy of epilepsy. Ann Neurol 1983;13(2):223-4. 222. Wuarin JP, Dudek FE. Electrographic seizures and new recurrent excitatory circuits in the dentate gyrus of hippocampal slices from kainate-treated epileptic rats. J Neurosci 1996;16(14):4438-48. 223. Falconer M, et al. Treatment of temporal lobe epilepsy by temporal lobectomy. A survey of findings and results. Lancet 1955;1:827-35.
2711
2712
160
Selective amygdalo-hypocampectomy
224. Hudson LP, et al. Amygdaloid sclerosis in temporal lobe epilepsy. Ann Neurol 1993;33(6):622-31. 225. Miller LA, et al. Amygdalar sclerosis: preoperative indicators and outcome after temporal lobectomy. J Neurol Neurosurg Psychiatry 1994;57(9):1099-105. 226. Rasmussen T, Feindel W. Temporal lobectomy: review of 100 cases with major hippocampectomy. Can J Neurol Sci 1991;18 (4 Suppl):601-2. 227. Feindel W, Rasmussen T. Temporal lobectomy with amygdalectomy and minimal hippocampal resection: review of 100 cases. Can J Neurol Sci 1991;18 (4 Suppl):603-5. 228. Jooma R, et al. Seizure control and extent of mesial temporal resection. Acta Neurochir 1995;133 (1–2):44-9. 229. Gloor P. The temporal lobe and limbic system. New York: Oxford University Press; 1997. p. 865. 230. Goddard GV, McIntyre DC, Leech CK. A permanent change in brain function resulting from daily electrical stimulation. Exp Neurol 1969;25(3):295-330. 231. Racine RJ, Burnham WM, Glibert M. Kindling mechanisms. In: Kindling JA, III, Wada, editor. New York: Raven Press; 1986. p. 263-82. 232. Tuunanen J, Halonen T, Pitkanen A. Status epilepticus causes selective regional damage and loss of GABAergic neurons in the rat amygdaloid complex. Eur J Neurosci 1996;8(12):2711-25. 233. Nitecka L, et al. Maturation of kainic acid seizure-brain damage syndrome in the rat. II. Histopathological sequelae. Neuroscience 1984;13(4):1073-94. 234. McIntyre DC, Racine RJ. Kindling mechanisms: current progress on an experimental epilepsy model. Prog Neurobiol 1986;27(1):1-12. 235. Wennberg R, et al. Preeminence of extrahippocampal structures in the generation of mesial temporal seizures: evidence from human depth electrode recordings. Epilepsia 2002;43(7):716-26. 236. Siegel AM, et al. Relationships between MR-imaged total amount of tissue removed, resection scores of specific mediobasal limbic subcompartments and clinical outcome following selective amygdalohippocampectomy. Epilepsy Res 1990;6(1):56-65. 237. Awad IA, et al. Extent of resection in temporal lobectomy for epilepsy. I. Interobserver analysis and correlation with seizure outcome. Epilepsia 1989;30(6): 756-62. 238. Avoli, M. Do interictal discharges promote or control seizures? Experimental evidence from an in vitro model of epileptiform discharge. Epilepsia 2001;42 Suppl 3: 2-4. 239. Barbarosie M, Avoli M. CA3-driven hippocampalentorhinal loop controls rather than sustains in vitro limbic seizures. J Neurosci 1997;17(23):9308-14. 240. Olivier A. Surgery of epilepsy: methods. Acta Neurol Scand Suppl 1988;117:103-13. 241. Crandall PH. Standard en bloc anterior temporal lobectomy. In: Spencer DD, Spencer SS, editors. Surgery for
242.
243.
244.
245.
246.
247.
248.
249.
250.
251.
252.
253.
254.
255.
256.
epilepsy. Boston: Blackwell Scientific Publication; 1991. p. 118. Spencer DD, et al. Access to the posterior medial temporal lobe structures in the surgical treatment of temporal lobe epilepsy. Neurosurgery 1984;15(5):667-71. Spencer DD. Anteromedial temporal lobectomy: Directing the surgical approach to the pathological substrate. In: Spencer DD, Spencer SS, editors. Surgery for epilepsy. Boston, MA: Blackwell Scientific Publication; 1991. Niemeyer P. The transventricular amygdalohippocampectomy in temporal lobe epilepsy. In: Baldwin M, Bailey P, editors. Temporal lobe epilepsy. Springfield, IL: Charles C. Thomas; 1958. p. 461-82. Niemeyer P, Bello H. Amygdalo-hippocampectomy in temporal lobe epilepsy. Microsurgical technique. Excerpta Medica 1973;293(20). Wieser HG, Yasargil MG. Selective amygdalohippocampectomy as a surgical treatment of mesiobasal limbic epilepsy. Surg Neurol 1982;17(6):445-57. Jasper H, Pertuisset B, Flanigin HF. EEG and cortical electrograms in patients with temporal lobe seizures. Arch Neurol Psychiatry 1951;65:272-90. Morris AA. Temporal lobectomy with removal of uncus, hippocampus, and amygdala; results for psychomotor epilepsy three to nine years after operation. AMA Arch Neurol Psychiatry 1956;76(5):479-96. Green JR, Duisberg RE, Mc GW. Focal epilepsy of psychomotor type; a preliminary report of observations on effect of surgical therapy. J Neurosurg 1951;8 (2):157-72. Wyler AR, Hermann BP, Somes G. Extent of medial temporal resection on outcome from anterior temporal lobectomy: a randomized prospective study. Neurosurgery 1995;37(5):982-90.; discussion 990–1. Sass KJ, et al. Specificity in the correlation of verbal memory and hippocampal neuron loss: dissociation of memory, language, and verbal intellectual ability. J Clin Exp Neuropsychol 1992;14(5):662-72. Sass KJ, et al. Degree of hippocampal neuron loss determines severity of verbal memory decrease after left anteromesiotemporal lobectomy. Epilepsia 1994;35 (6):1179-86. Sass KJ, et al. Verbal memory impairment correlates with hippocampal pyramidal cell density. Neurology 1990;40(11):1694-7. Martin RC, et al. Individual memory change after anterior temporal lobectomy: a base rate analysis using regression-based outcome methodology. Epilepsia 1998;39(10):1075-82. Davies KG, et al. Language function after temporal lobectomy without stimulation mapping of cortical function. Epilepsia 1995;36(2):130-6. Baxendale SA, et al. Hippocampal cell loss and gliosis: relationship to preoperative and postoperative memory function. Neuropsychiatry Neuropsychol Behav Neurol 1998;11(1):12-21.
Selective amygdalo-hypocampectomy
257. Bell BD, Davies KG. Anterior temporal lobectomy, hippocampal sclerosis, and memory: recent neuropsychological findings. Neuropsychol Rev 1998;8(1):25-41. 258. Helmstaedter C, Elger CE. Functional plasticity after left anterior temporal lobectomy: reconstitution and compensation of verbal memory functions. Epilepsia 1998;39(4):399-406. 259. Katz A, et al. Extent of resection in temporal lobectomy for epilepsy. II. Memory changes and neurologic complications. Epilepsia 1989;30(6):763-71. 260. Nunn JA, et al. Differential spatial memory impairment after right temporal lobectomy demonstrated using temporal titration. Brain 1999;122 (Pt 1):47-59. 261. Ajmone-Marsan C, Van Buren JM. Epileptiform activity in cortical and subcortical structures in the temporal lobe of man. In: Baldwin M, Bailey P, editors. Temporal lobe epilepsy. Springfield, IL: Charles C Thomas; 1958. p. 368-95. 262. Bengzon AR, et al. Prognostic factors in the surgical treatment of temporal lobe epileptics. Neurology 1968;18(8):717-31. 263. Engel JJ, Ojemann GA. The next step. In: Engel JJ, editor. Surgical treatment of the epilepsies. New York: Raven Press, Ltd.; 1993. p. 319-29. 264. Binnie CD, et al. Role of ECoG in ‘en bloc’ temporal lobe resection: the Maudsley experience. Electroencephalogr Clin Neurophysiol Suppl 1998;48:17-23. 265. Schwartz TH, et al. The predictive value of intraoperative electrocorticography in resections for limbic epilepsy associated with mesial temporal sclerosis. Neurosurgery 1997;40(2):302-9; discussion 309–11. 266. Kanazawa O, Blume WT, Girvin JP. Significance of spikes at temporal lobe electrocorticography. Epilepsia 1996;37(1):50-5. 267. Tran TA, et al. Significance of spikes recorded on electrocorticography in nonlesional medial temporal lobe epilepsy. Ann Neurol 1995;38(5):763-70. 268. Tuunainen A, et al. Postoperative EEG and electrocorticography: relation to clinical outcome in patients with temporal lobe surgery. Epilepsia 1994;35(6):1165-73. 269. McKhann GM, II, et al. Intraoperative hippocampal electrocorticography to predict the extent of hippocampal resection in temporal lobe epilepsy surgery. J Neurosurg 2000;93(1):44-52. 270. Goddard GV. Development of epileptic seizures through brain stimulation at low intensity. Nature 1967;214 (92):1020-1. 271. Feindel W, Penfield W. Localization of discharge in temporal lobe automatism. AMA Arch Neurol Psychiatry 1954;72(5):603-30. 272. Feindel W, Penfield W, Jasper H. Localization of epileptic discharge in temporal lobe automatism . Trans Am Neurol Assoc 1952;56:14-7. 273. Wieser HG, Meles HP. Clinical and chronotopographic psychomotor seizure patterns (SEEG study with reference to postoperative results). Acta Neurochir Suppl 1980;30:103-12.
160
274. Wieser HG, Meles HP. Limbic seizures: intracortical EEG activity and clinical signs. In: Gingris M, Kiloh LG, editors. Limbic epilepsy and the dyscontrol syndrome. Amsterdam: Elsevier; 1980. p. 195-206. 275. Goldring S, et al. Results of anterior temporal lobectomy that spares the amygdala in patients with complex partial seizures. J Neurosurg 1992;77(2):185-93. 276. Goldring S, et al. Temporal lobectomy that spares the amygdala for temporal lobe epilepsy. Neurosurg Clin N Am 1993;4(2):263-72. 277. Sperling MR, et al. Temporal lobectomy for refractory epilepsy. Jama 1996;276(6):470-5. 278. Foldvary N, et al. Seizure outcome after temporal lobectomy for temporal lobe epilepsy: a Kaplan-Meier survival analysis. Neurology 2000;54(3):630-4. 279. Harris AB, Lockard JS. Absence of seizures or mirror foci in experimental epilepsy after excision of alumina and astrogliotic scar. Epilepsia 1981;22(1):107-22. 280. Awad IA, et al. Intractable epilepsy and structural lesions of the brain: mapping, resection strategies, and seizure outcome. Epilepsia 1991;32(2):179-86. 281. Holmes MD, et al. Hippocampal or neocortical lesions on magnetic resonance imaging do not necessarily indicate site of ictal onsets in partial epilepsy. Ann Neurol 1999;45(4):461-5. 282. Fried I, Cascino G. Lesional surgery. In: Engel JJ, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1993. p. 501-9. 283. Ajmone-Marsan C. Acute effects of topical epileptogenic agents. In: Jasper HH, Ward AA, Pope A, editors. Basic mechanisms of the epilepsies. Boston, MA: Little Brown and Company; 1969. p. 299-319. 284. Awad I, Jabbour P. Cerebral cavernous malformations and epilepsy. Neurosurg Focus 2006;21(1):e7. 285. Hammen T, et al. Prediction of postoperative outcome with special respect to removal of hemosiderin fringe: a study in patients with cavernous haemangiomas associated with symptomatic epilepsy. Seizure 2007;16(3):248-53. 286. Rasmussen TB. Surgical treatment of complex partial seizures: results, lessons, and problems. Epilepsia 1983;24 (Suppl 1):S65-76. 287. Cascino GD, et al. Stereotactic resection of intra-axial cerebral lesions in partial epilepsy. Mayo Clin Proc 1990;65(8):1053-60. 288. Cohen-Gadol AA, et al. Long-term outcome after epilepsy surgery for focal cortical dysplasia. J Neurosurg 2004;101(1):55-65. 289. Yeh HS, et al. Surgical management of epilepsy associated with cerebral arteriovenous malformations. J Neurosurg 1990;72(2):216-23. 290. Cascino G, Boon PAJM, Fish D. Surgically Remediable Lesional Syndromes. In: Engel JJ, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1993. p. 77-86. 291. Cascino GD, et al. Operative strategy in patients with MRI-identified dual pathology and temporal lobe epilepsy. Epilepsy Res 1993;14(2):175-82.
2713
2714
160
Selective amygdalo-hypocampectomy
292. Valiante TA. Selective amygdalohippocampectomy. In: Lozano AM, Gildenberg PL, Tasker R, editors. Textbook of stereotactic and functional neurosurgery. Germany: Springer, 2009. 293. Vickrey BG, et al. A health-related quality of life instrument for patients evaluated for epilepsy surgery. Med Care 1992;30(4):299-319. 294. Trimble MR. Epilepsy and quality of life. New York: Raven Press; 1994. 295. Hermann BP. The evolution of health-related quality of life assessment in epilepsy. Qual Life Res 1995;4(2): 87-100. 296. Cramer JA. A clinimetric approach to assessing quality of life in epilepsy. Epilepsia 1993;34 Suppl 4:S8-13. 297. Vickrey BG, et al. Outcome assessment for epilepsy surgery: the impact of measuring health-related quality of life. Ann Neurol 1995;37(2):158-66. 298. Kjellberg RN, et al. Bragg-peak proton-beam therapy for arteriovenous malformations of the brain. N Engl J Med 1983;309(5):269-74. 299. Barcia-Salorio JL, et al. Radiosurgery of epilepsy. Acta Neurochir Suppl (Wien) 1993;58:195-7. 300. Barcia JA, et al. Stereotactic radiosurgery may be effective in the treatment of idiopathic epilepsy: report on the methods and results in a series of eleven cases. Stereotact Funct Neurosurg 1994;63(1–4):271-9. 301. Whang CJ, Kwon Y. Long-term follow-up of stereotactic Gamma Knife radiosurgery in epilepsy. Stereotact Funct Neurosurg 1996;66 Suppl 1:349-56. 302. Whang CJ, Kim CJ. Short-term follow-up of stereotactic Gamma Knife radiosurgery in epilepsy. Stereotact Funct Neurosurg 1995;64 Suppl 1:202-8. 303. Re´gis J, et al. First selective amygdalohippocampal radiosurgery for ‘mesial temporal lobe epilepsy. Stereotact Funct Neurosurg 1995;64 Suppl 1:193-201.
304. Re´gis J, et al. Gamma knife surgery in mesial temporal lobe epilepsy: a prospective multicenter study. Epilepsia 2004;45(5):504-15. 305. Hoggard N, et al. The clinical course after stereotactic radiosurgical amygdalohippocampectomy with neuroradiological correlates. Neurosurgery 2008;62(2):336-44; discussion 344–6. 306. Quigg M, Barbaro NM. Stereotactic radiosurgery for treatment of epilepsy. Arch Neurol 2008;65(2):177-83. 307. Rutecki P. Anatomical, physiological, and theoretical basis for the antiepileptic effect of vagus nerve stimulation. Epilepsia 1990;31 Suppl 2:S1-6. 308. Velasco AL, et al. Subacute and chronic electrical stimulation of the hippocampus on intractable temporal lobe seizures: preliminary report. Arch Med Res 2000;31 (3):316-28. 309. Kerrigan JF, et al. Electrical stimulation of the anterior nucleus of the thalamus for the treatment of intractable epilepsy. Epilepsia 2004;45(4):346-54. 310. Kossoff EH, et al. Effect of an external responsive neurostimulator on seizures and electrographic discharges during subdural electrode monitoring. Epilepsia 2004;45 (12):1560-7. 311. Sun FT, Morrell MJ, Wharen RE, Jr. Responsive cortical stimulation for the treatment of epilepsy. Neurotherapeutics 2008;5(1):68-74. 312. Commonwealth Department of Human Services and Health. Guidelines for the pharmaceutical industry on preparation of submissions to the Pharmaceutical Benefits Advisory Committee, including major subissions involving economic analysis. Australian Government Publishing Service; 1995.
161 Subpial Transection Z. S. Tovar-Spinoza . J. T. Rutka
Its aim is to impair the capacity of cortical tissue to generate sufficient neuronal synchrony to produce epileptiform discharges, without interfering with its capability to mediate normal physiologic transactions. (Frank Morrell 1989; > Figure 161‐1)
Principles of Multiple Subpial Transections a.
Background The technique of Multiple Subpial Transections (MST) was described by Morrell [1,2] to treat patients with intractable focal epilepsy who are not candidates for excisional surgery because their epileptogenic lesions lie in eloquent cortical territories responsible for speech, movement, primary sensation or memory. Resective surgery in these areas would result in unacceptable functional deficits – hemiplegia, hemianesthesia or aphasia – [3] in addition to the morbidity caused by the seizure disorder itself.
b.
Rationale The MST procedure consists of transections arranged in linear and parallel cuts 5 mm apart across the region defined as the epileptogenic zone. The principle upon which the procedure is based is the selective destruction of the short horizontal fiber connections with preservation of vertically oriented neuronal elements (> Figure 161‐2). The technique also permits preservation of the entire vascular supply and a virtually intact pial bank over the area through which the transection is made. From a functional point of view the advantage of this technique is that epileptogenic activity is eliminated while the normal functional role of the tissue is virtually unaffected [4]. #
Springer-Verlag Berlin/Heidelberg 2009
c.
Columnar Organization: In the 1960s, when the principle of MST was conceived, the vertical neuronal column was considered a master organizational unit for the normal cerebral cortex. The MST procedure is based on experimental evidence showing that the basic functional unit of cortical physiology is the vertically oriented neuronal elements and the incoming and outgoing fibers which are also vertically oriented. The axonal trajectory of thalamocortical afferents and primary efferent pathways run perpendicular to cortical surface. The intracortical interneurons relay vertically to apical dendrites of pyramidal cells which discharges radially [5–9]. Synchrony of cell discharge in epilepsy: In contrast to the requirements for normal neuronal propagation for functional output, an epileptogenic focus requires a critical volume of neurons generating a paroxysmal hypersynchronous epileptiform discharge having side-to-side or tangentialhorizontal linkages [10–13]. The critical mass of cerebral tissue: The minimum contiguous volume necessary to sustain synchronous spiking is 12.5 sq mm. Cortical islands greater than 5 mm in width, or tangential connections greater than 5 mm are essential for generation of paroxysmal discharges [14–16]. When two experimental foci are placed 4 mm apart, the paroxysmal discharges generated tend to become synchronous. If the two foci
2716
161
Subpial transection
. Figure 161‐1 Picture of Frank Morrell
d.
. Figure 161‐2 Rationale of MST consists of linear and parallel cuts 5 mm apart across the region defined as the epileptogenic zone. The principle is based on the selective destruction of the short horizontal fiber connections with preservation of vertically oriented neuronal elements
e.
are 6.7 mm apart, the activity in the two regions remains independent [10]. This provided the basis for the recommendation of a transection interval of 5 mm perpendicular to the long axis of the gyrus. In this way, the lesions made by MST should limit seizure propagation while sparing neuronal function [1]. Spread of Seizure Discharge: The most common pattern of spread of an evolving ictal focus is that of neighborhood propagation across the cortex in a non-uniform spatial pattern [17,18]. The spread of seizures depends on transverse connections in all cortical layers, but mainly in layers IV and V of the cerebral cortex [18,19]. Layer V might also be involved in the initiation of many epileptiform events [20]. Thus, disrupted horizontal connections might eliminate the capacity for epileptogenic discharges to spread [10,12]. Blood Supply Preservation: The blood supply to the gyrus enters perpendicular to the surface, with cortical arterial flow and venous drainage patterns following a parallel trajectory with axonal fibers. Based on this knowledge, MSTs should preserve the vascular supply to the cortex [1], and the subpial bank integrity should not be affected.
Indications 1.
Focal seizures arising in eloquent cortex (with or without resective surgery of the adjoining area when the epileptogenic zone extends away from the eloquent cortex). Published series include patients where MSTs were combined with lesionectomy or cortical resection. Therefore, it is difficult to assess the effectiveness of MSTs, per se, as an
Subpial transection
individual surgical procedure. The use of MST as stand alone therapy has been described by Schramm et al. [21] (20 patients with pure MST), Smith [22] (32 pure MST cases which 16 were patients with epilepsy), Lui et al. [23] (50 cases with less than 2 years follow up) and Whisler [24] (13 pure MST cases) series. Schramm obtained 45% and 50% satisfying results according to Engel’s and Spencer’s classifications. Seizure free outcome was obtained in 5% of Schramm’s series, 37.5% of Smith’s series, and 63% of Whisler’s series. As expected, MSTs performed on large areas of cortex were associated with poor outcome. MSTs performed on small areas of cortex had a better prognosis likely because they were associated with focal epileptic syndromes. In their meta-analysis of several series of patients with MSTs undergoing resective and nonresective surgery followed for 5 or more years, Tellez-Zenteno et al. [25] reported MST as the procedure having the lowest long-term seizure free rate outcome (16%) of all surgical procedures. However, this analysis combined the results of pure MST (23 patients) with MST plus resective surgery (51 patients). Previously, Spencer et al. [26] reported their meta-analysis from six series and reviewed the data of 55 patients who underwent pure MST. Here, the seizure outcome was available for 27 patients. They found that the overall seizure outcome was similar in the patients who had pure MST and MST with cortical resection (62–71% for patients with partial seizures had >95% seizure reduction, and 68–87% for patients with generalized seizures had >95% seizure reduction). In their analysis, better results were obtained in patients with generalized seizures having MST and cortical resection. However, this study did not specified the post surgical follow-up period for outcome analysis. This report and others [22,27–32]
2.
3.
161
concluded that MST might be an effective alternative to subtotal resection of the epileptogenic zone in critical brain areas. In children, as in adults, MST have a role mainly when performed in conjunction with cortical resection or lesionectomy [27,33,34]. Landau-Kleffner Syndrome. Laundau-Kleffner Syndrome (LKS) has been traditionally one of the main indications for MST in children [3]. LKS is defined as an acute or progressive, acquired epileptic aphasia (AEA) or verbal auditory agnosia in previously normal children associated with the presence of epileptiform discharges over the central and superior temporal regions that become more frequent during sleep. The condition resembles the epileptic syndromes of continuous-spike-and-wave disturbances in slow-wave-sleep (CSWS) [35] or Electrical Status Epilepticus during Sleep (ESES). Antiepileptic drugs (AED) can improve clinical seizures but EEG findings can be resistant to treatment [36]. Morrell et al. [3] reviewed their experience with 14 patients with LKS who underwent MSTs. Seven of 14 patients recovered age-appropriate speech and no longer needed speech therapy or special education classes. Another four (29%) of 14 had marked improvement of speech and understanding of instructions given verbally but still required speech therapy. Three had no changes in their preoperative conditions. Sawhney et al. [32] reported improvement in all three of their patients with AEA who underwent MSTs. Neville et al. [37] described one case with significant postoperative improvement who developed a word reading vocabulary, sign language and scored within the average on two non verbal subtest. Nass et al. [38] described their experience of MST in seven patients with atypical forms of LKS that demonstrated modest improvement in
2717
2718
161
4.
5.
Subpial transection
receptive more so than expressive language. Irving et al. [39] reported five children with LKS who underwent MST. Language improvement occurred in all children. None improved to an age appropriate level but seizures and behavior disturbances were immediately controlled in all of their patients. In summary, it has been reported that MST for LKS can result in improvements in language, but it may take years before these gains are seen [40]. Use of MSTs for severe autistic regression in childhood epilepsy is not always associated with improvement in cognitive and behavioral function. At best, gains observed are temporary. Use of MST over the pacemaker zone, where epileptogenic activity impairs the development of normal function, has proven successful, in whole or in part, in improving language, reducing autistic features, and controlling seizures [38]. Malignant Rolandic-Sylvian Epilepsy Syndrome. Malignant Rolandic-Sylvian Epilepsy (MRSE) syndrome was described by Otsubo et al. [41] in children presenting with intractable sensorimotor partial seizures that progress to secondary generalization. Their disorder is further characterized by frontocentro-temporal spikes on EEG, absence of lesions on MRI, localized spike sources in the rolandic-sylvian regions by MEG, and neurocognitive problems. MRSE requires surgery for maximal excision in nonfunctional cortex and MSTs throughout eloquent cortex. The results seen in seven patients suggest that this combination of techniques can reduce seizure activity from the epileptogenic zone and eliminate seizures or sharply reduce their incidence without postoperative permanent motor deficits or further language deficits. Cortical Dysplasia, Epilepsia Partialis Continua (EPC) and Rasmussen’s syndrome. Molineux et al. described a successful treatment of one patient with epilepsia partialis
continua by MSTs with epileptogenic foci over the left central cortex, normal MRI findings and diagnosis of cortical dysplasia by biopsy [42]. MST has also been successfully used to improve EPC caused by Rasmussen’s encephalitis [32,43].
Defining the Focus Presurgical evaluation for patients requiring MSTs is similar to patients with intractable epilepsy being considered for cortical resection. Ictal and interictal scalp EEG, ictal and interictal intracranial EEG, positron emission tomography (PET), single-proton emission computed tomography (SPECT), and magnetic resonance (MRI) with special sequences (sagittal T1weighted images, axial and coronal dual-echo T2-weighted images, coronal fluid-attenuated inversion recovery and coronal volumetric 3-D Fourier transform gradient-echo times on proton density and T2-weighted images to evaluate myelination and the use of Gadolinium-contrast to rule out neoplasms) [33] should all be performed. In addition, formal neurological and neuropsychological evaluations are done to determine preoperative verbal and memory performance and lateralization. Language dominance is determined by WADA testing, functional MRI or MEG. MEG spikes source localization is also used to demarcate the epileptogenic zone and can be overlaid onto MR images generating a magnetic source image (MSI) [44]. The type and frequency of seizures should be recorded for postoperative outcome analysis.
Surgical Technique The technique of MST is not new, as it was initially described by Morrell and colleagues in 1969 [4,45]. After the surgical focus has been
Subpial transection
defined, the bipolar electrocautery is used to cauterize a small pial point. The pia mater is then sharply incised. The MST knives – small, blunt, right angle hooks – (> Figure 161‐3) are introduced through the incised pial point and swept forward to the sulcal margin in an arc like fashion to a depth of approximately 1.5–2 cm, making a right-angled cut to the long axis of the gyrus. The blade of the MST knife is maintained in a strictly vertical orientation to avoid undercutting the cortex (> Figure 161‐4). The tip of the MST knife is visualized through the pia mater as it is drawn back along the subpial space completing the transection. Care should be taken to avoid disrupting the pia matter or injuring sulcal vessels during the transection procedure. There is an option for the surgeon to make the pial insertion point at the center of the gyrus in which case half the gyrus is transacted as described above before the instrument is removed and reinserted aiming in the opposite direction. Sulcal cortex transection is sometimes technically difficult to perform but should be attempted for completion o the MST. Another variation on the technique of MST is to point the MST knife downwards and to use a sharpened rather than a blunt knife to minimize the damage of using blunt instruments [35]. After removing the MST knife there may be small capillary bleeding at the pial entrance point which is easy to stop and is also a useful marker to measure the 5 mm distance to the next parallel transection. In cases where MSTs are performed . Figure 161‐3 MST knives
161
in multiple contiguous gyri in one or more lobes of the brain, intraoperative ultrasonography can be useful at the end of the procedure to ensure that no underlying intracerebral hematoma has been inadvertently created [33]. Electrocorticography (ECoG) is usually performed before and after MSTs, to evaluate the interictal activity and response to surgery. In cases such as LKS, MSTs are stopped when the ECoG shows a significant reduction in the epileptiform discharge. This is of course assuming that intraoperative ECoG correlates well with postoperative seizure control. As previously reported, the role of ECoG in predicting seizure outcome has been the subject of some controversy [33,34,46].
Postoperative Complications MSTs have generally been associated with low morbidity. Bleeding from capillaries at the sulcal . Figure 161‐4 The artist represents the technique of MST knife insertion on the cortex of the epileptogenic zone
2719
2720
161
Subpial transection
entry points is easily controlled with gentle pressure and hemostatic agents. Formation of small subarachnoid hemorrhages and 2–3 mm intracerebral hematomas [24,47] and rarely large and symptomatic ones [21,24] have been described. Some degree of brain swelling is expected but excessive brain swelling not related to hematomas formation has been reported only once [21]. Immediate transient neurological deficits (dysphasia, hemiparesis, memory disturbances, sensory disturbances) are common in a high proportion of patients [24,29,31,32,34,48] however, permanent deficits defined by Schramm as any deficit lasting longer than 3 months, are rare events (propioception disturbance [30], hemiparesis [22,24,32,49], dysphasia [24], adiadochokinesis [48]) even in cases where large regions of cortex were treated with MSTs [21].
Reasons for MST Failure 1.
2.
Efficacy for partial rather than generalized seizures: Hashizume and Tanaka [50] examined the effect of MST on the EEG of the epilepsy rat model. They found that the epileptic activity remain propagated to the contralateral cortex. They hypothesized that even if the horizontal fibers involved in cortical seizures evolution are disrupted preventing partial seizures from spreading, vertical connections from the epileptogenic cortex to subcortical structures involved on the seizure generalization could persist [26]. Failure of transecting the entire width of the cortex: Some technical aspects could account for the failure of MSTs. There may be failure to keep the strict verticality of the transaction at the gyral surface, resulting in sparing of sulcal cortical tissue [30]. There may be differences in the length of the
3.
curved MST knives which play a factor in the completion of the transection [51]. Preserved, non-transected tissue bridges in layers IV and V or V and VI could allow for seizure propagation. [18] Kaufman et al. [52] performed a histological study in humans who underwent MST. They showed that the spacing and orientation of the MSTs macroscopically were perpendicular to the gyral axis and to the correct depth. However, in microscopic examination, they found that a significant proportion of the transections were not selective, due to inappropriate depth (too deep or not deep enough) or inappropriate, such as oblique, orientation. Lack of homogeneous seizure outcome analysis. Postoperative outcome for epilepsy surgery has been usually reported at 2 years after the surgical procedure [53]. However, seizure outcome is dynamic over this time and may change up to 5 years post surgery [21,45,54]. Different MST studies presented their outcome analysis based on different follow up periods. Whether there is initial seizure improvement at 1–2 years post op which declines incrementally over subsequent years is a question worth asking in many cases. Another issue is that outcomes have been reported using different scales (Engel [53] or modified Engel [54] or Spencer classifications [21]), making comparative analysis difficult. Thus, effectiveness of MSTs in the long term has been difficult to assess.
Conclusions MSTs offer a relatively safe surgical alternative to direct cortical excision for patients with
Subpial transection
intractable epilepsy arising from within functional brain elements. As stand alone therapy, MSTs are probably not as effective as when combined with cortical resection. Patients with partial seizures may have a better outcome than those with secondary generalization based on experimental and clinical data. Morbidity is acceptable with the expectation that function will improve in the early post-operative period. In the future, seizure outcomes should be measured prospectively and over longer periods of time to fully define the utility of MSTs in patients with intractable epilepsy [1].
References 1. Morrell F, Whisler WW, Bleck TP. Multiple subpial transection: a new approach to the surgical treatment of focal epilepsy. J Neurosurg 1989;70:231-39. 2. Morrell F, Hanbery JW. A new surgical technique for the treatment of focal cortical epilepsy. Electroencephalogr Clin Neurophysiol 1969;26:120. 3. Morrell F, Whisler WW, Smith M, et al. Landau-Kleffner syndrome. Treatment with subpial intracortical transection. Brain 1995;118:1529-46. 4. Morrell F, Hanberry JW. A new surgical technique for the treatment of focal cortical epilepsy. Electroencephalogr Clin Neurophysiol 1969;26:120. 5. Asanuma H, Sakata H. Functional organization of a cortical efferent system examined with focal depth stimulation in cats. J Neurophysiol 1967;30:35-54. 6. Asanuma H, Stoney SD, Jr, Abzug C. Relationship between afferent input and motor outflow in car motorsensory cortex. J Neurophysiol 1968;31:670-81. 7. Hubel DH, Wiesel TN. Receptive fields, binocular interaction and functional architecture in the cat’s visual cortex. J Physiol 1962;160(1):106-54. 8. Mountcastle V. Modality and topographic properties of single neurons of cat’s somatic sensory cortex. J Neurophysiol 1957;20:408-34. 9. Powell T, Mountcastle V. Some aspects of the functional organization of the cortex of the post-central gyrus of the monkey: a correlation of findings obtained in a single unit analysis with cytoarchitecture. Bull Johns Hopkings Hosp 1959;105:133-62. 10. Lueders H, Bustamante L, Zablow L, et al. The independence of closely spaced discrete experimental spike foci. Neurology 1981;31:846-51.
161
11. Lueders H, Bustamante L, Zablow L, et al. Quantitative studies of spike foci induced by minimal concentrations of penicillin. Electroencephalogr Clin Neurophysiol 1980;48:80-9. 12. Reichental E, Hocherman S. The critical cortical area for the development of penicillin-induced epilepsy. Electroencephalogr Clin Neurophysiol 1977;42:248-51. 13. Tharp B. The penicillin focus: a study of field characteristics using cross-correlation analysis. Electroencephalogr Clin Neurophysiol 1971;31:45-55. 14. Dichter M, Spencer W. Penicillin-induced interictal discharges from cat hippocampus. II. Mechanisms underlying origin and restriction. J Neurophysiol 1969;32:663-87. 15. Dichter M, Spencer W. Penicillin-induced interictal discharges from the cat hippocampus. I. Characteristics and topographical features. J Neurophysiol 1969;32:649-62. 16. Goldensohn E, Zablow L, Salazar A. The penicillin focus I: distribution of potential at the cortical surface. Electroencephalogr Clin Neurophysiol 1977;42:480-92. 17. Chervin R, Pierce P, Connors B. Periodicity and directionality in the propagation of epileptiform discharges across neocortex. J Neurophysiol 1988;60:1695-713. 18. Telfeian A, Connors B. Layer-specific pathways for the horizontal propagation of epileptiform discharges in the neocortex. Epilepsia 1998;39:700-8. 19. Ebersole J, Chatt A. The laminar susceptibility of cat visual cortex to penicillin induced epileptogenesis. Neurology 1980;30. 20. Connors B. Initiation of synchronized neuronal bursting in neocortex. Nature 1984;310:685-7. 21. Schramm J, Aliashkevich AF, Grunwald T. Multiple subpial transections: outcome and complications in 20 patients who did not undergo resection. J Neurosurg 2002;97:39-47. 22. Smith M. Multiple subpial transection in patients with extratemporal epilepsy. Epilepsia 1998;39:S81-9. 23. Lui L, Zhao Q, Li S, et al. Multiple subpial transection for treatment of intractable epilepsy. Chin Med J 1995;108:539-41. 24. Whisler WW. Multiple subpial transection. Tech Neurosurg 1995;1:40-4. 25. Tellez-Zenteno JF, Dhar R, Wiebe S. Long-term seizure outcomes following epilepsy surgery: a systematic review and meta-analysis. Brain 2005;128:1188-98. 26. Spencer SS, Schramm J, Wyler A, et al. Multiple subpial transection for intractable partial epilepsy: an international meta-analysis. Epilepsia 2002;43:141-5. 27. Guenot M. [Surgical treatment of epilepsy: outcome of various surgical procedures in adults and children]. Rev Neurol (Paris) 2004;160 Spec No 1:5S241-50.
2721
2722
161
Subpial transection
28. Liu Z, Zhao Q, Tian Z, et al. Multiple subpial transection for treatment of intractable epilepsy. Chin Med J 1995;108:539-541. 29. Mulligan LP, Spencer DD, Spencer SS. Multiple subpial transections: the Yale experience. Epilepsia 2001;42:226-9. 30. Pacia SV, Devinsky O, Perrine K, et al. Multiple subpial transections for intractable partial seizure: seizures outcome. J Epilepsy 1997;10:86-91. 31. Rougier A, Sundstrom L, Claverie B, et al. Multiple subpial transection: report of 7 cases. Epilepsy Res 1996;24:57-63. 32. Sawhney IM, Robertson IJ, Polkey CE, Binnie CD, Elwes RD. Multiple subpial transection: a review of 21 cases. J Neurol Neurosurg Psychiatry 1995;58:344-9. 33. Benifla M, Otsubo H, Ochi A, et al. Multiple subpial transections in pediatric epilepsy: indications and outcomes. Childs Nerv Syst 2006;22:992-8. 34. Blount JP, Langburt W, Otsubo H, et al. Multiple subpial transections in the treatment of pediatric epilepsy. J Neurosurg 2004;100:118-24. 35. Wyler AR. Multiple subpial transections in neocortical epilepsy: Part II. Adv Neurol 2000;84:635-42. 36. Buelow J, Aydelott P, Pierz D, et al. Multiple subpial transections for Landau-Kleffner Syndrome. AORN J 1996;63:727-39. 37. Neville BG, Harkness WF, Cross JH, et al. Surgical treatment of severe autistic regression in childhood epilepsy. Pediatr Neurol 1997;16:137-40. 38. Nass R, Gross A, Wisoff J, et al. Outcome of multiple subpial transections for autistic epileptiform regression. Pediatr Neurol 1999;21:464-70. 39. Irwin K, Birch V, Lees J, et al. Multiple subpial transections in Landau-Kleffner syndrome. Dev Med Child Neurol 2001;43:248-52. 40. Grote CL, Van Slyke P, Hoeppner JA. Language outcome following multiple subpial transections for LandauKleffner syndrome. Brain 1999;122:561-6. 41. Otsubo H, Chitoku S, Ochi A, et al. Malignant rolandicsylvian epilepsy in children: Diagnosis, treatment and outcomes. Neurology 2001;57:590-6.
42. Molyneux PD, Barker RA, Thom M, et al. Successful treatment of intractable epilepsia partialis continua with multiple subpial transections. J Neurol Neurosurg Psychiatry 1998;65:137-8. 43. Nakken KO, Eriksson AS, Kostov H, et al. [Epilepsia partialis continua (Kojevnikov’s syndrome)]. Tidsskr Nor Laegeforen 2005;125:746-9. 44. Otsubo H, Oishi M, Snead OC, III. Magnetoencephalography. In: Miller J, Silbergeld D, editors. Epilepsy surgery: principles and controversies neurological disease and theraphy. New York: Mercel Decker; 2007. p. 752-67. 45. Morrell F, Whisler W, Bleck T. Multiple subpial transection: a new approach to the surgical treatment of focal epilepsy. J Neurosurg 1989;70:231-9. 46. Wennberg R, Quesney L, Lozano A, et al. Role of electrocorticography at surgery for lesion-related frontal lobe epilepsy. Can J Neurol Sci 1999;26:33-9. 47. Shimizu H, Suzuki I, Ishijima B, et al. Multiple subpial transections (MST) for the control of seizures that originated in unresectable cortical foci. Jpn J Psychiatry Neurol 1991;45:354-6. 48. Hufnagel A, Zentner J, Fernandez G, et al. Multiple subpial transections for control of epileptic seizures: effectiveness and safety. Epilepsia 1997;38:678-88. 49. Patil AA, Andrews R, Torkelson R. Isolation of dominant seizure foci by multiple subpial transections. Stereotact Funct Neurosurg 1997;69:210-5. 50. Hashizume K, Tanaka T. Multiple subpial transection in kainic acid-induced focal cortical seizure. Epilepsy Res 1998;32:389-99. 51. Patil AA, Andrews RV, Torkelson R. Surgical treatment of intractable seizures with multilobar or bihemispheric seizure foci (MLBHSF). Surg Neurol 1997;47:72-7; 52. Kaufmann WE, Krauss GL, Uematsu S, et al. Treatment of epilepsy with multiple subpial transections: an acute histologic analysis in human subjects. Epilepsia 1996;37:342-52. 53. Engel J, Van Ness P, Rasmussen T, et al. Outcome with respect to epileptic seizures. In: J E editor. Surgical treatment of the epilepsies. 2nd ed. New York: Raven Press; 1993. p. 609-21. 54. Orbach D, Romanelli P, Devinsky O, et al. Late seizure recurrence after multiple subpial transections. Epilepsia 2001;42:1316-19.
154 The Wada Test-60th Year Anniversary Update-In Epilepsy Surgery J. A. Wada . B. Kosaka
Prologue This is the 60th and 53rd Anniversary year, respectively, of the first Carotid Amytal Injection performed in Japan in 1948 [1] and its introduction to North America via the Montreal Neurological Institute in 1955 [2,3]. During this time, the prevailing neuro-investigative landscape of over a half century ago when these events had taken place were all invasive in nature except for 8 channel EEG and skull X-ray. Neuropsychology was a budding research tool. Prior to the development of the Wada Test, the understanding of the brain mechanism of cognitive function was elusive and based largely on a chance encounter with a localized lesion with postmortem clinical–pathological correlation. The introduction of the Wada Test added an entirely new dimension since it enabled us to electively assess each cerebral hemispheric function through reversible deactivation in a behavioral state. The rapidity with which the Wada test was accepted and disseminated worldwide may be viewed as a reflection of the then prevailing unmet need for the prevention of postsurgical cognitive side effects. Such need was obviously greatest for an elective procedure especially for epilepsy surgery. The publication in 1954 of Penfield and Jasper’s [4] ‘‘Epilepsy and the Functional Anatomy of the Human Brain’’ had significantly invigorated interest in epilepsy surgery and therefore a need was most acutely felt in an ever expanding community of neurologists and neurosurgeons involved in the management of #
Springer-Verlag Berlin/Heidelberg 2009
medically refractory epilepsy as an increasing number of epilepsy centers were created during the latter half of the twentieth century. During this period, the Wada Test not only provided an unambiguous result on the pattern of language representation as it was originally concei ved, but also lead to further speculation and investigations into hemispheric specialization for example, the melodic/prosodic expression which we observed during Wada testing at UBC as well as attentional asymmetry [6,221]. There has also been verification that the mechanism underlying sign language is the same as that for spoken language [227,229]. It has also provided a hypothesis that later learned languages may be more bilaterally represented than the first language learned [9]. Most importantly, it has given us invaluable insight into the plastic reorganization of language function in patients with epilepsy. The memory component that was later added to hopefully predict rare but devastating global amnesia, had produced challenging results. Indeed severe postoperative amnesia has been rare since the inception of the Wada Test but debate continues on the prediction and the nature of postoperative ‘‘poor memory.’’ However, the outcome of the memory component so far appears to have added a complementary component along with other diagnostic measures pointing to epileptogenic laterality. In the mean time, unexpected discontinuation of Amytal production by Eli Lilly had a profound effect. Although the supply issue was eventually resolved by the resumption of amytal
2588
154
The Wada test-60th year anniversary update-in epilepsy surgery
production by Ranbaxy, the crisis prompted many epilepsy centers to search for drugs that can possibly be employed as a substitute to amytal. As a result, the use of sodium secobarbitol [10], methohexital [11], propofol [12–14], and etomidate [15] were reported. In the latter, serial infusions are made to maintain a level of anesthesia at an added risk due to prolonged catheterization. With the amytal shortage, expectation was raised for brain imaging, a research tool, to become a valid and dependable clinical tool for the presurgical assessment of language and memory function in individual patient with epilepsy. In 1993, 45 years after the Wada Test was born, a worldwide survey response of 71/102 epilepsy centers indicated that 95% of them were using the Wada test, and 85% very frequently as a tool for presurgical investigation [16]. Three/71 centers responded that they were not using the Wada test due to non-medical reasons. Fifteen years later in 2008, another international survey response of 92/207 epilepsy centers indicated that 87% of responding epilepsy centers were using the Wada test; nearly 40% frequently and 47% less frequently. Thirteen percent were reported not using it for medical reasons. Concentrations of using centers in North America were assumed to be related to medical-legal implications [17]. An informal survey conducted on the David Loring’s NPSYCH listserve [18], which is largely subscribed to by neuropsychologists, revealed that many sites continue to do the Wada test now. This was a very limited informal survey as there are epilepsy centers where the neurologist conducts the testing and they would not have been aware of this survey. Twenty nine of 31 sites identified that the Wada test is being used. Two of 31 sites were using Methohexital instead of amytal. Eight of 31 sites were doing selective Wada protocols, for example just on suspected cases for atypical language dominance or patients with a left temporal focus. Optical imaging had
replaced Wada testing at one site. Another site relied on PET/SPECT/MRI and CCTV-EEG monitoring. Again, many active epilepsy centers were not approached during this informal survey. In Europe, 16/23 sites surveyed there during 2000–2005 continued to do a Wada protocol with a relatively low risk (1.02%) of morbidity (personal communication Frank Oltmans). These results appear to represent a transitional state in the back drop of a remarkable transformation of a neuro-diagnostic landscape surrounding epilepsy surgery during the past decades including the maturation of neuropsychology as a clinically relevant subspecialty, the development of long term video/EEG monitoring (electro-clinical correlation leading to better definition of semiology and objective peri-ictal and ictal cognitive assessment), and the development and proliferation of brain imaging approaches particularly high definition structural MRI. They all contributed to the gradual phasing out of earlier invasive procedures except for occasional use of intracranial EEG monitoring. Ten years ago in 1998, 50 years after the Wada test was developed we wrote a chapter on this topic. This updating chapter (1) highlights information derived by the Wada test with emphasis on dynamic reorganization of language representation and memory related issues, (2) a selective review on the evolving state of brain imaging with some emphasis on commonly used ones as a complementary tool for presurgical cognitive evaluation, and (3) the Wada Test beyond 2008.
The Assessment of Language Via Wada Protocol It has been recognized that severe epilepsy can affect language function and lead to atypical cerebral dominance. Intractable, frequent generalized tonic–clonic seizures and status epilepticus are associated with language problems, particularly if
The Wada test-60th year anniversary update-in epilepsy surgery
the seizure focus is lateralized to the language dominant hemisphere. The neurosurgeon and epileptologist can decide if more detailed localization is required via intraoperative mapping techniques once cerebral hemispheric dominance for language has been established. When chronic seizures begin at an early age, prior to age six, with or without a structural lesion, there is a higher risk for the presence of cerebral reorganization. More recently, this phenomenon has been demonstrated using pediatric populations [226] and by using different techniques such as MEG [20]. The patterns of hemispheric language dominance which we discussed in our first chapter are seen in > Table 154-1. These ranges come from studies completed at epilepsy centers. The diversity in Wada protocols has contributed to the range of incidence in atypical patterns of language. Each center has established protocols where the emphasis has been on speech production including naming (i.e., name the object being shown to you), automatic or sequenced speech (i.e., counting) as well as repetition. Over time, many protocols began to incorporate auditory comprehension including following simple commands (i.e., stick out your tongue), repetition of short phrases, and the reading of simple statements. However a more critical question is ‘‘What does the data reflect?’’ Is it the language impairment of the injected hemisphere or the language capacity of the uninjected hemisphere? This becomes most challenging when one is determining the possible bilateral hemispheric representation of language. It also supports the recommendation of having an EEG recording
154
simultaneously throughout the test to chronologically assess the cerebral effects of the amytal and the recording of behavioral changes over time. Our clinical experience at UBC was that patients with bilateral dominance would name objects, follow commands, repeat a nursery rhyme, and read simple cards with injection to either hemisphere. We also observed that some patients had speech production transfer to the right hemisphere but language comprehension appeared to reside in the left hemisphere. These patients were identified as having bilateral dissociated or asymmetrical representation. Risse and her group accurately point out that certain tasks such as simple rote speech in the right hemisphere may not be a valid example of bilateral language. More importantly, she suggests that not all patterns of bilateral language are a reflection of insult and language reorganization. It may in fact be representing a normal developmental variant explaining that there may be some genetic predisposition that allows the development of language in both hemispheres [22].
The Assessment of Memory Via Wada Protocol Milner [23] noted that there was an urgent clinical need in the early 1950s to assess potential surgical risk for lasting anterograde amnesia following a unilateral temporal lobectomy. The ipsilateral deactivation by amytal of the healthy temporal lobe was expected to create a brief amnestic state if the contralateral hippocampus is compromised. The clinical challenges at this
. Table 154-1 Patterns of hemisphere dominance for language [21]
Right-handers Left handers Mixed handedness
Left Hemisphere dominance
Bilateral Hemisphere dominance
Right Hemisphere dominance
92–96% 50–70% 50–70%
7% 19% 15–20%
4–15% 15–20% 15–20%
2589
2590
154
The Wada test-60th year anniversary update-in epilepsy surgery
time included that the injection was still being made into the common carotid artery and without angiography. There was no assurance of actual drug distribution. The memory protocol had to assess memory capacity but not be confounded by the presence of aphasia. Successful recognition testing of the test stimuli from a multiple-choice array would be accepted as a ‘‘pass’’ as amnestic patients often failed both free recall and recognition memory tasks. Behaviorally, it had been observed that patients could often recognize objects that they could not name during drug effect and that some individuals could also not recognize objects that they had named correctly. More importantly, the original purpose of memory testing during a Wada test was to attempt to predict only the risk for global amnesia [24,25]. As mentioned in our original chapter, the Wada technique has provided an opportunity to investigate material-specific memory problems. However the material-specific/dual-encodability factor of test stimuli has created challenges. Most nonverbal material can have verbal labels attached to it. There are very few stimuli or tasks that are so material-specific that only the unilateral, nondominant hemisphere is involved in the processing. Meador and Loring [26] pointed out that Wada protocols typically have not relied on word memory tasks due to the aphasia confound when the language dominant hemisphere is injected. They cautioned that these results could not be generalized to all other Wada procedures due to the aphasia confound. More importantly, the results from Wada testing may have limited correlation to the variable of seizure outcome postoperatively. Secondly, the relationship between Wada test results to hippocampal function is one of controversies. Variability in the memory test results correlating with different hippocampal subfields may in part be reflecting histological technical differences. Meador and Loring [5] discuss how the Wada test does not typically perfuse
the posterior two-thirds of the hippocampus. However, no significant correlation between filling of the posterior cerebral artery and Wada memory score was found while injecting contrast agent before the Wada to detect posterior cerebral arterial filling [27]. The likely explanation would be, therefore, deafferentation of the posterior hippocampus from the cortical areas is more relevant than direct anesthesia of the structure. Results of EEG observation during Wada test are consistent with this view [28,29]. Memory scores from Wada protocols may be affected by the timing of the injection. Grote et al [30] demonstrated that Wada memory scores were lower for the second injection when performed on the same day but were similar for the first and second injections if they were conducted on separate days. Various epilepsy centers have compared the left and right memory scores as a percentage of the total possible score while also balancing the injection order between the epileptic focus and nonfocus side of injection. This finding highlights some of the potential challenges when comparing studies from different epilepsy centers. The general approach during a Wada protocol is to present stimuli while one hemisphere has been anesthetized. Recall and recognition are done when the amytal effects have dissipated. It is reported to have a 48% predictive value for postoperative amnesia [31]. It is more challenging when one has anesthetized the language dominant hemisphere but allowing the patient to have a multisensory experience including touching the object, hearing the name of the object and ensuring that it is in the correct visual field is essential for optimizing test administration. Many seasoned Wada assessors have realized that information is often encoded by what appears to be an obtunded patient. Another approach (Dodrill Seattle protocol) involves a memory protocol that is reported to have a 76% predictive value for postoperative memory states, and shows the test stimuli prior
The Wada test-60th year anniversary update-in epilepsy surgery
to injection and continuously shows the items following injection. The subject names the object, then experiences a brief interference task, and then is asked to recall the object just before the card. High risk for postoperative amnesia is determined when a patient cannot recall the named item and that the hemisphere injected is the one needed for memory functioning [31–33]. Again it can be controversial about what is being tested. If items are remembered correctly and a passing score is obtained, is the nonanesthetized lobe intact to handle memory? If there is poor performance, then does the nonanesthetized hemisphere have poor capacity to handle memory? There has also been the interpretation that if recognition does not occur, then memory function relies heavily on the hemispheric injected. Our UBC experience has been to view that the test scores obtained have been a reflection of the non-anesthetized hemisphere. It has been reported that muteness, dysphasia and the arterial distribution of amytal may not significantly impact on the interpretation of Wada memory testing [34].
Aberrant Organization of Cognitive Function Epileptic activity has been known to not only exert noxious/adverse effects on cerebral function in general but also activate plastic and compensatory hemispheric processes. Potential significance of the latter is suggested by increased cognitive risk with later- but not early-onset seizure group postoperatively [7]. Differences found are small and relative but this ‘‘age of onset effect’’ may reflect age dependent cognitive plasticity that diminishes with age in response to focal epileptogenesis [35]. Some comments relevant to organization/reorganization of hemispheric language and memory function are given as follows in the light of the results of Wada test and some imaging technologies.
154
Language Several risk factors that had been underscored for promotion of atypical language representation were (1) greater tissue loss in the left hemisphere, (2) early seizure onset, (3) handedness, and (4) proximity of epileptic lesion to language area [36]. The significance of early seizure onset has been repeatedly confirmed [37–40] though the upper age limit appears to extend beyond 9–16 years of age depending on the circumstances [41–44]. The third issue of handedness is intriguing and the precise relationship between handedness and cerebral language dominance remains unclear. It was stated that ‘‘an early lesion that does not modify hand preference is on the whole unlikely to change speech representation’’ [36]. Although handedness is a useful clinical marker, change of laterality of handedness and language dominance can occur independently. There is also a possibility of naturally endowed right language dominance in which a right cerebral lesion may cause laterality change in handedness/language dominance either alone or together [45]. Finally, the significance of lesion location has to be viewed with a new perspective based on our experience that the functional consequence of localization related epilepsy is often not localization related. For example, an extensive cyst formation next to language areas may not cause shift of language function [46]. Partial seizures associated with lesions near the frontal and temporal language areas maintain left hemisphere language in children. In adults epileptogenic lesions in the vicinity of language area may cause perilesional language reorganization [47]. Later onset disease in childhood and leftsided congenital lesions tended to maintain left representation [37,48,49]. Reorganization of cortical function is known to result from abnormal, neuronal/glial proliferation (cortical dysplasia) but not abnormal cortical organization (polymicrogyria) [50]. Transhemispheric transfer of language function occurs frequently in mesial temporal lobe epilepsy [20,48].
2591
2592
154
The Wada test-60th year anniversary update-in epilepsy surgery
On the other hand, Wada tested cases of left mesial temporal lobe epilepsy has f MRI evidence of more bi-hemispheric distribution of language organization including contralateral homologous areas both pre- and postoperatively [51–53] though evidence of precise homology is lacking [54]. In addition to age factor, some functional elements appear to play an important role for reorganization of language function: atypical representation cases verified by Wada test with left mesial temporal lobe epilepsy were found to have higher interictal discharge frequency and lateral temporal-related sensory aura (presumably due to seizure spread) than those left language dominant cases [55]. This finding poses an important question as to potential long-term consequence of interictal discharge since it is conceived to have only a transient disruptive effect on cortical cognitive function [56]. The general pattern of language representation found by the Wada test in epilepsy population indicates a continuum from left to right, i.e., L, L > R, L ¼ R, L < R and R. Our observation suggests that hemispheric language representation is not a clear cut dichotomous but rather continuous variable which appears more compatible with many neuroimaging results. Indeed combined quantitative–qualitative evaluation of Wada test results yields a more graded account of the varieties of language representation [57–59]. Among the atypical language patterns, bilateral L > R representation appears to be the majority. L ¼ R is rare but displays expressive and receptive difficulty bilaterally and equally. Historically, there have been suggestions that only a unilateral injection was needed during a Wada investigation but this data would support that both hemispheres need to be assessed for language. Unilateral injection judged to be appropriate may lead to erroneous conclusion [60]. Also, there are patients with no linguistic disruption despite hemiplegia on either side. Two of the three such patients had no language impairment following left and right temporal lobectomy,
respectively [61]. The absence of linguistic impairment with the Wada test does not exclude ipsilateral language representation [62] and no cortical mapping was done in the reported cases. It is noteworthy that one of our UBC series with no language disruption bilaterally despite hemiplegia was ambidextrous as were the reported three cases. Some bilateral cases have asymmetrical dissociated representation with the Broca on one side and the Wernicke on the other [59,63– 65]. The finding suggests that anterior frontal and posterior temporal functions may be differentially affected [29] and that reorganization of expressive and receptive language capabilities can occur independently [57,58,66]. Atypical language representation with right dominance or bilateral representation has been discussed largely on the basis of left hemisphere epilepsy. However, suggestive evidence of transfer of language function from a ‘‘natural’’ right dominant hemisphere to the left exists in some right hemisphere epilepsy. Among right hemisphere epilepsy, atypical representation occurs in later- rather than early-onset. The latter observation raises a potential scenario of early right hemisphere epilepsy interfering with the establishment of ‘‘natural’’ right language dominance and causing it to shift to the left hemisphere in a person predisposed to develop right dominance [67]. Likely existence of ‘‘natural’’ bilateral representation must be kept in mind [58,68]. In this regard, structural basis of transhemispheric language representation and more specifically how white matter pathways are affected in aberrant processes of brain reorganization remains unknown. A recent MRI study revealed maturational changes including increased axonal fiber organization especially in the arcuate fasiculus (AF), during late adolescence (age 16) suggesting that fiber organization rather than myelination may play a greater role in the AF development [69]. The AF is the important anatomical language pathway in classical aphasiology. One way of evaluating morphological correlates of
The Wada test-60th year anniversary update-in epilepsy surgery
language reorganization would be to use diffusion tensor imaging (DTMRI – see later). The presence of a ‘‘natural’’ right hemisphere dominance and its potential reorganization particularly in right-sided epilepsy is obviously an intriguing area of future enquiry since recent brain imaging studies of right-handed healthy population suggest 4–6% bilateral representation [70,71] and up to 7.5% right dominance [218] though a f MRI study of a 100-right-handed healthy population failed to identify a single person with right dominance [71]. Experience with hemispherectomy in children with severe epilepsy taught us that extensive left hemisphere damage can cause contralateral transfer of language with good language capability though with a crowding effect on right hemisphere functions. Through Wada test results, we now know that a well-circumscribed left epileptogenic focus remote from eloquent cortical area can also cause atypical representation. Hippocampal focus appears to be particularly potent in activating ‘‘reorganizational force’’ [73–77]. Cortical mapping studies in left mesial temporal lobe epilepsy found aberration of the left temporal language area either more anteriorly placed [228] or having a noncontiguous pattern [78]. The latter was more frequent among those with Wada verified cases with bilateral dominance [79]. This finding needs to be replicated but suggests a pervasive nature of reorganizational force in patients with mesial temporal lobe epilepsy. Altogether, early onset epileptic activity particularly that involving the hippocampus within the language dominant hemisphere appears to play a powerful role in initiating reorganization of language function.
Memory Aberrant language organization has claimed a lion’s share of our attention for the cognitive consequence of epilepsy. However, a similar
154
process appears to take place in hemispheric memory mechanism. Patients with early onset seizure disorder tend to have less cognitive decline in the domain of memory after left temporal lobectomy [80–82]. Among temporal lobe epilepsy, Wada-verified atypical language organization was found more frequently in early seizure onset group (70%) than late onset group (30%). Those with atypical organization showed relatively better postsurgical memory outcome in several neuropsychological measures than typical organization group. This ‘‘age of onset effect,’’ presumably protective in nature, was conceived to be due to presurgical functional reorganization of memory mechanism [35]. A Wada memory study on left temporal lobe epilepsy found a significant correlation between right memory function and age at onset with onset after 5 years of age being associated with poor memory function, suggesting a ‘‘critical period’’ for right hemisphere reorganization of verbal memory before that age [83]. Results of recent brain imaging studies support this notion. fMRI studies have shown that in left temporal lobe epilepsy verbal memory tasks involve the right temporal lobe, a pattern that is not found in healthy normal population [84,85]. This is consistent with the reported predictive value of memory associated event-related potentials recorded from the right hippocampus in left temporal lobe epilepsy [86]. Similarly, a magnetoencephalographic study showed activation of right mesial temporal lobe with a trend of less left mesial temporal neuromagnetic activity during verbal memory task in left temporal lobe epilepsy with hippocampal dysfunction than in normal control subjects. The finding suggests that the patients with left hippocampal dysfunction are more likely to recruit the right mesial temporal lobe for verbal memory tasks [87]. The effectiveness of such reorganization is not well understood and how it interacts with atypical language representation remains to be explored [88]. Different patterns of reorganization for language and memory function was also found in right language
2593
2594
154
The Wada test-60th year anniversary update-in epilepsy surgery
dominant patients with left temporal lobe epilepsy when compared with left language dominant patients with left or right temporal lobe epilepsy [89]. Differential reorganization mechanisms between language and memory function, i.e., a shift of verbal memory to the non-dominant right hemisphere can occur independently without concomitant change of hemispheric language organization in left temporal lobe epilepsy is highly suggestive [90,91]. It is obvious that the outcome of verbal memory function must depend on the extent and the quality of language/memory reorganization involving both dominant and nondominant hemispheres and, therefore, how to precisely define the nature and pattern of reorganization is the challenge facing us now. The hippocampus has been regarded as a crucial component in the process of episodic memory but it appears to contribute to other cognitive capabilities such as semantic memory [223]. A recent neurophysiological study suggested a possibility that the left hemisphere language may be linked to a process of learning and memory trace formation with putative advantage of building up memory traces for any language elements [93]. The nature of such advantage remains unknown. However, since hippocampal focus/lesion in many intractable temporal lobe epilepsy is assumed to have occurred during first years of life, it implies that the hippocampus plays a substantial role not only in the memory process but also in establishing language dominance.
Alternate Techniques for Determining Language Dominance & Memory Capacity Historically, noninvasive approaches such as tachistoscopic presentations or dichotic listening have attempted to determine dominance. However, no method has been able to unequivocally measure cerebral dominance. Visual half-field approaches must consider several methodological
factors [94,95] to maximize the effectiveness of this approach. Bourne also points out that divided visual field tasks can be complex and fatiguing. Dichotic listening [96] has not reached reliability for atypical language dominance such as bilateral representation and fMRI imaging has suggested that dichotic listening may involve frontotemporal and not just temporal networks [97]. The lack of reliability of ear advantage in dichotic listening was revealed by Strauss et al. [98] who showed that although the right ear advantage was consistent in 86% left hemisphere dominant – Wada tested subjects; a right ear advantage was also evident for 50% of the right and 71% of the bilateral dominance, respectively. A more recent study aiming to examine the magnitude of ear advantage on an individual basis, the fused rhymed dichotic words test (FDWT) in 61 Wada tested patients indicated extreme asymmetries are associated with contralateral speech dominance while lesser asymmetries are more often associated with bilateral speech representation [99]. Another FDWT study reported concordance but the number of patients involved is very small, i.e., 5 Wada tested children with three left and two bilateral representation [100]. Electrophysiological study of Wada tested patients with spectral analysis of photic and click evoked response showed over 90% of accuracy with miscalculation in 1/16 left dominance and 1/6 right dominance [101]. Event related potential analysis in 12 Wada tested patients including 4 bilateral dominance cases yielded unreliable and conflicting results [102]. The past two decades have witnessed spectacular advances in invivo imaging technologies for morphological to functional assessment of the human brain in health and disease. In the epilepsy population, the evolution of MRI scanning has provided a solid window to the anatomical structure, asymmetries, pathologies, and has changed the vista of clinical epilepsy. Based on well established prenatal peri-Sylvian asymmetry [103,104], a structural MRI study found
The Wada test-60th year anniversary update-in epilepsy surgery
concordance between planum temporale (PT) asymmetry and Wada tested patients (10 left dominance with PT L > R; 1 right dominance with PT R > L) [105]. However, another study with larger number of Wada tested patients (20 left, 11 right and 13 bilateral representation) failed to correlate laterality between PT and language. Interestingly, this study found more white matter in the dominant Broca’s area in either left or right dominant group but there was no reliable difference to identify individual language laterality [106]. In this regard, the development of functional diffusion tensor imaging (DTMRI) opens a new vista. It is dependent on the tendency of water molecules to diffuse in the direction of myelinated fiber bundles, enabling diffusion-weighed MRI to yield tensor maps of fiber orientation [107]. Thus DTMRI allows the reconstruction of white matter pathways leading to virtual in vivo dissection of the human brain [108] and has so far disclosed details of peri-Sylvian language networks [109] and arcuate fasiculus (AF) asymmetry with higher fractional anisotropy (FA), i.e., measure of magnitude and direction of fiber tracts, in the left hemisphere [110] of healthy subjects. The significance of FA in the left temporal lobe and reading ability in children (8–18 years) has been shown [111]. An image from Arjan-Hillebrand’s work also visually demonstrates the role of the left temporal lobe in letter fluency for ‘‘groupaveraged’’ data. In refractory epilepsy, evidence of reorganized language function has been identified by combined DTMRI and fMRI [112,113]. In addition, a recent DTMRI study with a small number of patients suggested that integrity of AF is related to memory performance in the left but not right temporal lobe epilepsy [114]. Although replication with larger sample sizes will be required, this study not only indicates a promising future of DTMRI for further exploration of structure–function correlation in language reorganization but also has potential for contributing insight into epilepsy-induced functional alteration such as secondary epileptogenesis.
154
It has been estimated that MRI scans can be interpreted as being ‘‘normal’’ in 30% of the cases identified as nonlesional epilepsy [115]. However, structural MRI has also significantly contributed not only to quantitative analysis but also to the identification of pathology such as mesial temporal sclerosis. Although structural MRI does not directly reveal functional information of the structures imaged, it can help identify hippocampal asymmetry, for example, that may provide a sensitive index for their functional state and surgical outcome [116]. Voxel-based morphometry yields significant group difference between the brains of healthy and epilepsy population but it appears to have little clinical utility for individual comparison [117]. How much more clinically relevant information can be extracted from quantitative/volumetric analysis of structural MRI in the individual patient remains a challenge. Many other brain imaging approaches began to be applied in epilepsy as a clinical research tool primarily for the cognitive evaluation and validation purpose with Wada tested patients. Data obtained by brain imaging based on activation and the Wada Test based on deactivation are complementary while discordance between the two continues to be debated. For the purpose of this chapter, the emphasis will be on more commonly used techniques for the investigation of language dominance and memory performance in the context of Wada test. > Table 154-2 provides a synopsis of techniques available for presurgical evaluation with respective limitation.
TMS – Transcranial Magnetic Stimulation TMS is a noninvasive method of ‘‘deactivation’’ by using electromagnetic induction. Repetitive transcranial magnetic stimulation (rTMS) can induce longer lasting effects, which may lend itself well to assessing cognition by the induction
2595
2596
154
The Wada test-60th year anniversary update-in epilepsy surgery
. Table 154-2 Utility and limitations of presurgical cognitive assessment techniques Procedure
Principle
Language
Memory
Comments
Wada
Deactivation
+
+
rTMS
Deactivation
+
MRI
Structure
+
DTMRI
White matter pathways
+
fMRI
Activation hemodynamic change
+
MEG
Magnetic flux directly associated with event-related activation Activation hemodynamic change Activation hemodynamic change Activation hemodynamic change Activation hemodynamic change
+
Invasive, efficient utility for lateralization of language dominance Speech arrest can assist in determining language dominance, however one needs to know where to stimulate Expensive, structural asymmetries and/or mesial temporal sclerosis may imply dysfunction, expensive Expensive, inference from arcuate fasciclus asymmetry Expensive, difficult with young children and the cognitively challenged, utility may be in localization Expensive, poor spatial resolution, only receptive language has been assessed in controls
+
+
+
+
+
f TCD NIRS/OT SPECT PET
of speech arrest for example. This technique has been reported to correlate reasonably well with Wada test results [118–120]. Wassermann et al’s study found that rTMS of language areas produces an increase in errors of visual naming but not word reading. It was the first study to show rTMS effects on spoken language without induction of speech arrest. On the other hand, in Jennum’s study, dysphasia secondary to the contralateral facial and laryngeal muscle contraction was difficult to differentiate from aphasia. The validity of cognitive evaluation by rTMS depends on reliable elicitation of speech disruption unencumbered by the accompanying involuntary muscle activity. EMG analysis indicated that rTMS applied over a posterior site, lateral to the motor hand area of both left and right hemisphere caused speech disruption accompanied
Lack of temporal bone window in 15–20% general population, utility is in lateralization Superficial brain areas only, inability to access deep brain structures, surface data only Minimally invasive, poor spatial/temporal resolution, utility in lateralization Expensive, minimally invasive, poor spatial/temporal resolution. Utility in lateralization, inferred dysfunction if hypometabolism present
by mentalis muscle activation while rTMS applied over an anterior site on the left but not the right hemisphere resulted in the speech disruption without mentalis muscle activation [121]. This nonmotor class of speech disruption is the one which suggests potential utility of rTMS in clinical setting. Another rTMS study of Wadatested 17 epilepsy surgery candidates, reported a lack of reliability for the presurgical evaluation of language dominance [122]. Theodore [123] also reports that rTMS appears to reveal significant bilateral language function in patients with uncontrolled partial seizures and it remains unclear if this is an artifact of the technique or some other physiological phenomenon. At this time, he did not feel that rTMS could be used clinically for the determination of language dominance or memory lateralization. His paper provides an
The Wada test-60th year anniversary update-in epilepsy surgery
overview of some of the parameters of this technique. For memory assessment by rTMS, a study found no quantitative but qualitative changes only in the left temporal lobe epilepsy patients [124]. In addition to facial and laryngeal muscle contraction interfering with speech performance, rTMS carries a small but definite risk for seizures, particularly in epileptic patients [8]. For potential benefit, rTMS could play a complementary role in an individual through deactivation to characterize behavioral effects in fMRI assessment [125] since, as is discussed later, task dependent fMRI activation does not independently confirm that such activation is necessary or critical. More recent reviews are by Devlin and Watkins [126] and Robertson, Theoret and Pascual-Leone [22].
fTCD-Functional Transcranial Doppler Sonography f TCD measures cerebral perfusion changes related to neuronal activation in a way comparable to fMRI and 15O PET. fTCD makes it possible to compare perfusion changes (by measuring blood flow velocities) during repeated performance of a task (word generation by letter) within the territories of the 2 middle cerebral arteries that comprise the potential language zones. This quantitative technique provides an operational laterality index which resembles the one obtained by Wada [3,127,128]. It has been validated by direct comparison of results with Wada testing [129] and fMRI results [130]. In Knecht et al’s study [129], 4/19 Wada tested patients could not be examined due to the absence of acoustic temporal bone window for insonation while of the remaining 11 patients, 8 left and 3 bilateral dominance representation were concordant. Another study assessed 12 Wada tested patients but an acoustic temporal window was missing in 1 [131]. Among the
154
remaining 11 patients, 9 left, 1 right and 1 bilateral representation, fTCD and Wada results were concordant [131]. Similarly, 2/13 patients lacked temporal window but the remaining 11 patients, 9 left, 1 each right and bilateral representation, were concordant [132]. Finally, in 17 surgical candidates including 14 epilepsy patients, language lateralization either left or right representation (but no bilateral case) the results between fTCD and SPECT was concordant [133]. fTCD allows the determination of hemispheric language dominance in individual subjects including children and cognitively challenged people [132]. This is significant since young age and cognitive impairment have been identified as predictive of inconclusive Wada test results [134]. Considerable amount of normative data on the fTCD patterns of language representation, including left, right and bilateral dominance in healthy population are available [68,72,135,136]. fTCD is an effective, reliable, inexpensive and noninvasive procedure for language lateralization with excellent temporal but limited spatial resolution. As for all functional imaging techniques, fTCD cannot determine whether an activated region during a particular task is a critical region that, when damaged, will result in a loss of that particular function. An additional limitation is that fTCD cannot be used in a minority (15–20%) of patients who lack an acoustically penetrable bone window. However, auxiliary techniques are becoming available that allow fTCD assessment even in the presence of thick bones [137]. The most serious drawback is the inability to define memory function particularly for the verbal modality.
NIRS – Near-infrared Spectroscopy and OT – Optic Topography NIRS is a relatively new noninvasive technique, which enables measurement of hemodynamic changes associated with neural activity. The
2597
2598
154
The Wada test-60th year anniversary update-in epilepsy surgery
different light absorption spectra of oxyhemoglobin (oHb) and deoxyhemoglobin (dHb) within the near infra-red spectra allow for in-vivo measurement of concentration of these substances. Near-infrared light of two wave lengths between 680 and 1,000 nm is directed through optic fibers to the head of the patient. The amount of detected light reflects the amount of absorption of the two wave lengths in targeted cerebral regions a few centimeters below the scalp, informing on concentration changes of oHb and dHb in the region. Cerebral activation usually results in an increase in blood flow that is disproportionate to the metabolic rate of oxygen. Regions that are activated will have a greater oxygenation and a greater ratio of oHb to dHb. This would allow for the assessment of blood flow change in the superficial brain areas. NIRS was used for language laterality assessment using a word generation task and was concordant in Wada tested 4 left and 2 right dominant patients [138]. Similarly, a group of 16 epilepsy, Wada tested patients was studied with a complex design by Watson et al. [139] preoperatively while 6/16 and 10/16 had NIRS pre- and postoperatively, respectively. Eleven patients that included two atypical representation were concordant (5/6 preoperative and 6/10 postoperative cases). Among 5 discordant patients, 4 and 1 were pre- and postoperative NIRS cases, respectively. NIRS appears reasonably accurate for lateralizing language in normally developed individuals with presumed left hemisphere dominance. However there may be limitations for atypical language dominance. More recently, NIRS was concordant in 5 Wada tested epilepsy patients: one adult (left dominant) and 4 pediatric (ages 9–15 years old: 3 left dominance, 1 bilateral) [140]. Same group of investigators reported that the combined and prolonged NIRS-EEG monitoring in a 10-year-old epileptic boy found its utility not only for language lateralization but also for the identification of ictal onset zone [141]. Optical topography (OT) is based on the same principle as NIRS. OT functionally maps
the human cerebral cortex. The major difference between OT and NIRS is that OT measures spectroscopic reflection and scattering simultaneously from multiple measurement points whereas NIRS measures them with one or a few pairs of a light emitter and detector. OT has been successfully used for the identification of activation in the temporal [222] and inferior frontal cortex [142] during speech recognition and syntactic processing, respectively in healthy adults. In another study of 54 Wada tested patients, concordance rate for 43 left dominance was 74.5% (32/43), 6 right dominance 83% (5/6) and 3 bilateral representation 60%(3/5) [10]. The advantage of this method is that it is a relatively inexpensive, portable, with no major restriction on movement or verbalization during recording and it can be used in young children and babies as well as the cognitively challenged person. It is also clear that NIRS cannot assess deep structures because of the shallow penetration of the photons (between 3 and 5 cm), that renders it difficult to collect reliable information from subcortical structures. Therefore it will be limited for language assessment only. Another area of the NIRS utility is the assessment of differential hemodynamic pattern according to seizure type and lateralization [143–145,225].
fMRI – Functional Magnetic Resonance Imaging In the early 1990s, fMRI a MRI innovation, allowed images weighted by blood oxygenation level to be obtained in as little as a few tens of milliseconds. Since bold oxygenation is finely controlled and responds promptly to local energy demand, voxel-by –voxel estimation of changes in blood oxygenation represents maps of brain activation. Thus, it is based on hemodynamic changes, i.e., blood oxygenation level dependent (BOLD) signal determined by cerebral activity. The increase in cerebral blood flow in
The Wada test-60th year anniversary update-in epilepsy surgery
activated cerebral areas surpasses the increase in metabolic rate for oxygen, which induces increased capillary and venous oxygenation level while reducing the relative concentration of deoxyhemoglobin which is paramagnetic. This increase and decrease of oxygenated and decreased deoxyhemoglobin result in an increased BOLD signal. fMRI has been used to lateralize and localize language and memory functions for prediction and prevention of cognitive complication of epilepsy surgery. Many epilepsy centers have attempted to compare fMRI (activation) with the Wada test (deactivation) for validation purpose. Not surprisingly, comparison of fMRI and the Wada language assessment is challenging despite general agreement because of different fMRI language tasks, analysis, scoring and classification of laterality as well as varying Wada techniques and scoring involved. Lesser [146] points out that reproducibility of language results may be variable dependant on what aspects of language dominance (frontal vs. temporal) one is assessing. Concordant results reported for fMRI and Wada language determination are: using visually presented words in 6 left and 3 right dominant patients [147]; semantic decision task (presentation of animal names from which subject was to pick domesticated ones)in 18 left and 4 atypical representation [148]; with semantic language activation task in 39 left, 11 atypical (3 right and 8 bilateral) representations [71]; with word generation task in 8 left, 1 right and 4 bilateral representation [19]; 8 left, 3 right and 2 bilateral representation [149]; and 15 left, 1 right and 3 bilateral representation [51]. Concordance was also seen with three language tasks (object naming, single word reading, verb generation) in 9 left and 3 atypical representations [150], and another three language tasks (semantic verbal fluency, story listening,covert sentence repetition) in 10 patients [127,146] respectively. In the latter study, the former two tasks were concordant with frontal asymmetry while temporal asymmetry using covert sentence repetition and story listening
154
had no correlation with Wada asymmetry. Partial concordance is reported using read response naming in 3/21 patients, i.e., one Wada tested left dominant-fMRI bilateral, two Wada tested bilateral dominance-fMRI left [151], 1/20 Wada rightfMRI bilateral [152], covert word generation in 9/100 patients [153] and visually presented object naming in 2/21 patients [154]. A battery of tasks was found to yield better lateralization in 26 patients but bilateral representation remained a challenge [154]. The use of 4 tasks and statistical manipulation produced more robust result but one each of 11 left, three right and four bilateral dominance cases remained discordant [59]. Similarly, the use of semantic decision task (presentation of nouns designating inexpensive common object and pressing a ‘‘yes’’ button) in 68 patients showed mostly asymmetrical frontal and temporal activation with 21% discordance [155]. It is noted that in the latter study, while discordance rate was 11% in right temporal lobe epilepsy, it was 27.5% in left temporal lobe epilepsy. It is also noted that fMRI correctly classified all right dominance cases with less sensitivity to bilateral representation while it missed 17% of left dominance. In one recent fMRI study of a diverse group of 35 pediatric patients, three were found to be discordant to Wada test results, electrical cortical stimulation or other measures of lateralization [156]. Finally, it is rare but false fMRI language lateralization can occur [157,158]. The validity of fMRI in defining language lateralization may be tested by prediction of language deficit subsequent to dominant hemisphere surgery. In this regard, fMRI prediction with the Boston naming test was favorable in 24 Wada tested patients having greater sensitivity in predicting verbal naming difficulty following left anterior temporal lobectomy [159]. Another approach for validation involves correlation of language fMRI (activation) with direct cortical stimulation (inactivation). A reasonably good correlation is reported by several studies [160–164] but contrasting results are reported
2599
2600
154
The Wada test-60th year anniversary update-in epilepsy surgery
by others indicating fMRI by itself cannot be used to decide which cortical region can be resected [59,165,166]. It is noted that the result of extraoperative cortical stimulation language mapping in 10 pediatric and 14 adult Wada tested patients indicated the superiority of the Wada test in younger age group [92]. This confounding factor is not limited to only fMRI since the question of whether every speech arrest site by cortical stimulation is responsible for speech generation remains unanswered [167]. Clearly, much development is needed to standardize paradigms and procedures in fMRI. fMRI has been seen as an imaging technique that would provide more detailed information regarding language reorganization. Atypical and particularly bilateral representations are challenges not only for the Wada test but also for fMRI since bilateral activation in the latter is also dependent on task difficulty, for example [146,168]. Furthermore it may not tell the complete story. Hertz-Pannier et al. [42] describe a young patient with seizure onset at 5.5 years of age and Rasmussen’s encephalitis. He had normal language and his first fMRI indicated left hemisphere language dominance using a word fluency task. Left hemispherectomy at age nine was preceded by two prior partial resections at age seven. Immediately after surgery he became mute and was unable to read. Eighteen months later, a fMRI revealed a right-sided homologous language network that was not evident on the initial fMRI. There was also neuropsychological evidence of bilateral receptive but not expressive language. It appears that there may be a longer critical period for language plasticity due to the nature and extent of the injury (i.e., in this case, serial injury ending with hemispherectomy), the cognitive domain involved (i.e., the receptive functions of language) as well as the nature of the task (silent word fluency and sentence generation). They also acknowledge the possible artifact in design or data analysis, which would include picking a particular cognitive function
and an appropriate experimental task as well as the need for an appropriate baseline task to control for brain activation not related to the experimental task. This particular case highlights the test–retest advantage of fMRI. A similar Rasmussen Encephalitis case with delayed language reorganization has been reported [220]. Atypical language representation in epilepsy as a reflection of injury-induced reorganization of the brain has been discussed in terms of differences in methodology and determination of language dominance. Some fMRI studies have correlated their findings with those of the Wada test: the correlations are not perfect, possibly reflecting differences in the language tasks and the criteria for establishing language dominance. Even when there is an indication of asymmetry or a ‘‘strong lateralization,’’ fMRI may indicate bilateral activation [170]. Complex cognition such as language may often incorporate bilateral activation. Not all activated areas may be critical for the task. It is difficult to decipher the extent and nature of the right hemisphere’s contribution and whether bilateral fMRI activation implies that either side can mediate language function independently or that both hemispheres are necessary. In a right hemiparetic patient with an early onset left temporal lobe epilepsy, the result of Wada testing (125 mg amytal) was judged to indicate right language dominance with no change in speech or right hemiparesis following left injection but left hemiplegia and a loss of consciousness with right injection. Motor functional fMRI showed strong left sided activation with right paretic hand movement. This strikingly contrasting result between Wada deactivation and fMRI activation illustrates potential complementary contributions of each procedures [219]. fMRI assessment of memory has yielded promising results and has revealed presurgical memory lateralization and prediction of postoperative memory [171–173]. However, the theoretical framework of developing appropriate and reliable paradigms appears more challenging than for language since one has to consider different
The Wada test-60th year anniversary update-in epilepsy surgery
aspects of memory such as encoding, retrieval, recognition and possible material specific influences such as verbal and nonverbal stimuli and different patterns of presentation. For predictive value of fMRI for postsurgical memory outcome, high correlation was found in 16 right mesial temporal lobe epilepsy not tested by the Wada test on a single subject level [171]. Another study also suggested that fMRI testing of visual memory correlates well with Wada test results and predicted postoperative visual memory deficit in patients going on to temporal lobe resection [172], the gold standard for fMRI and Wada. This study involved 30/35 Wada tested epilepsy patients. All underwent both fMRI using a complex scene encoding task and Wada testing. Encoding performance was assessed by the follow-up recognition test. Twenty-three patients who subsequently underwent temporal lobe resection completed the same task outside of the scanner in an average of 6.9 months postresection. Asymmetry ratio (AR) was calculated for activities in the region of interest: Hippocampus alone (H) and hippocampus, parahippocampus and fusiform gyrus (HPF). Healthy controls showed symmetrical AR indicating bihemispheric verbal and visuospatial involvement while patient’s AR correlated with Wada for HPF but not H. In addition, ipsilateral HPF and AR showed a significant inverse correlation with good post-operative memory outcome (i.e., the lower ipsilateral fMRI absolute activation, the better the memory outcome). This finding is encouraging since it demonstrates that fMRI can predict post excisional visual memory deficit in intractable temporal lobe epilepsy patients. However, no normative data comparing the two methods have been established. Most intriguingly, AR differences in either H or HPF did not show the same correlation between seizure outcome and Wada memory results as previously reported [169,174]. Recently, prediction of postoperative verbal memory decline was also reported by using fMRI language lateralization measure in 60 left temporal lobe epilepsy [175].
154
Although fMRI represents a major advance, since it is noninvasive and fast, it still relies on indirect measures of neuronal activity. Increases and decreases in measured activity are difficult to interpret because the processes that regulate oxygen supply are complex and not fully understood. Ultimately, the measurements still rely on blood supply. Despite promising results, fMRI exploration of memory function remains a major challenge.
MEG – Magnetoencephalography Magnetoencephalography (MEG) measures the magnetic fields produced by electrical activity in the brain via extremely sensitive devices such as superconducting quantum interference devices (SQUIDs). These measurements are commonly used in both research and clinical settings. MEG has the advantage of measuring brain activity over time as a cascade of extremely fast events and the unfolding of specialized processes, segregated in space and time and organized into well defined stages of processing. It has been useful in mapping and localizing epileptic activity in both children [176] and adults. This technique can measure the activation of language-associated cortex. The dewars (helmet shaped device that the patient inserts their head into) can contain hundreds of sensors measuring brain activity. MEG relies on the averaging of brain responses evoked by a given stimulus. This technique has been advantageous when studying primary sensory and motor regions. However when studying a complex process such as language, it is reflecting a network of activation in primary sensory cortex followed by activation in the association cortex such as language. Beamforming techniques analyze oscillatory changes across cortex unlike traditional analysis and may be more useful in studying cortical networks involving sensory and higher level cognitive processes [177].
2601
2602
154
The Wada test-60th year anniversary update-in epilepsy surgery
With a word recognition task, MEG has yielded robust language lateralization concordant with the Wada test results: 87.5% in 16 patients [27], 100% in 11 patients [178], 87% in both 19 [9] and 15 pediatric patients [179], 100% in 15-right-handed patients [180], 87% in 85 patients [181], 95% in 20 patients [182], and 89% in 27 patients [183]. Similarly robust test–retest reliability was obtained in 21 patients [184]. One Spanish validation study with 8 Wada tested patients is worth quoting [185]. Seven patients with left dominance were concordant though typically MSI (magnetic source imaging) activity sources were found bilaterally in language specific areas. Wada test results of the remaining right handed, right mesial temporal lobe epilepsy patient were ‘‘inconclusive’’ with extremely rapid postinjection recovery within 40 s suggesting an inadequate deactivation. MEG repeated twice indicated bilateral dominance and then right dominance. This patient had a secondarily generalized convulsion after right anterior temporal lobectomy and with a transient postictal ‘‘global aphasia’’ lasting for 15 min. It is likely that this patient had bilateral asymmetrical representation. MEG use in determining language function has been largely confined to the area of receptive language which may be a disadvantage since, as already discussed, reorganization of receptive and expressive function can occur independently. Some have addressed this issue using picture [180] and action [186,224] naming but largely in healthy populations. MEG evidence of transhemispheric and intrahemispheric receptive language reorganization was reported in mesial temporal and lesional epilepsy patients, respectively [20]. Pre- and post operative MEG evaluation of receptive language in Wada tested mesial temporal lobe epilepsy patients showed more right hemisphere participation after surgery in bilateral representation cases while left dominant cases were more likely to show intrahemispheric changes with a slight inferior shift of the putative location of Wernicke’s area
even when the resection did not impinge on Wernicke’s area [187]. No study has been published that has addressed evaluating memory function. However, in addition to relatively robust neuromagnetic activity in the left peri-Sylvian and temporoparietal structures, the verbal memory task for language lateralization was reported to evoke less robust activity in the ipsilateral mesial temporal structures [181]. This observation was recently investigated in mesial temporal lobe epilepsy patients and healthy subjects [188]. The study not only affirmed mesial temporal activity but also demonstrated some evidence of reorganization of memory function in left temporal lobe epilepsy. It suggests a potential role for MEG in memory assessment to become a clinically relevant measure, but further development of paradigms that can reliably generate robust neuromagnetic mesial temporal activity will be required. MEG has a great potential for measuring very early reaction time in auditory and linguistic processing including speech perception, and contributes significantly to the characterization of the successive stages in language processing. In addition, it has the potential advantage of being able to localize neurophysiological processes associated with cognitive function within the whole brain and can provide high temporal and moderate spatial resolution suited for detailed spatial/time analyses of cognitive activation but the impact of age and education remains unclear. MEG is expensive and considerable technical support is required for both the administration and data analyses.
PET – Positron Emission Tomography PET started the modern era of functional brain imaging in the early 1980s. It relies on the detection of emission from radioactivity labeled molecules introduced as a bolus into the bloodstream.
The Wada test-60th year anniversary update-in epilepsy surgery
With each bolus, the concentration of radioactivity labeled molecules increases for a few minutes, particularly in metabolically active area. Injections of a bolus are given in separate sessions, some just before the subject executes a task (the active sessions), and some while the subject rests (the baseline sessions). Comparing the difference or some statistical measure of the detected radioactivity emanating from each brain voxel provides a map of the activated areas. Thus, it requires a radiotracer for measurement of glucose metabolism with 18 FDG, cerebral blood flow with 15O-labeled water or quantification of Benzodiazepine receptors with [11C] Flumazenil. PET is somewhat invasive since it requires radioactive materials. It offers a measure of hemispheric topographic organization and potential identification of eloquent language area. A European multicenter evaluation of PET in language lateralization disclosed significant differences across centers [189]. Many tasks and paradigms have been used including verb generation, story listening, object naming, and verbal fluency. It has been useful in establishing language dominance with significant concordance between regional PET data and Wada test results in children and adolescents: the activation of inferior frontal lobe with listening to and repetition of sentences and Wernicke’s area for repetition [190]. Factors contributing to discordance between PET and Wada test results include: simplistic and limited language tasks/the sensitivity of the PET, a higher number of Wada tested atypical representation, and motion artifacts in their pediatric cases. The inability to distinguish critical primary regional activations for a task versus secondary activations may have also contributed to these results. Successful use of 15O water has been reported in identifying the language area in a partial epilepsy patient with angiomatous malformation [191]. A PET study with auditory and visual confrontational naming activation involving 12 Wada tested patients(10 Left and 2 Right
154
dominance) reported a considerable interrater variation of lateralization. There was a complete discordance in a left temporal lobe epilepsy patient with Wada right dominance and PET left dominance. The patient had language difficulty following left anterior temporal lobectomy. The Wada test had used 75 mg amytal and bilateral R > L representation could have been missed [192]. Another PET study with verb generation from words and/or pictures in Wada tested patients had concordance in 23/24 (20 left and 3 right dominance) but a mixed handed patient with Wada left dominance was PET right dominant. This patient had decline of verbal function following right anterior temporal lobectomy [193]. Again, possible bilateral representation could not be excluded. For the purpose of validating PET language activation with both visual and auditory naming tasks, 7 Wada tested patients, 6 left and 1 right dominant patients underwent extraoperative cortical stimulation mapping. Not all the activated area stimulated caused language disruption but cortical regions that showed activation during the tasks were located in the same regions as electrodes that caused language disruption during stimulation [194]. The study suggests that PET can direct to likely eloquent areas that should be tested for more precise localization within the Wada identified hemisphere. PET has also been reported to be useful for the evaluation of memory function in temporal lobe epilepsy [195,196] and detected PET hypometabolism can infer lateralized memory impairment for postsurgical prediction of memory outcome [197,198]. While memory alone is a complicated process [199], Krause et al. have shown differences based on the PET data analyses technique used. Covariance analysis-based data analysis allows for functional interactions between brain regions of a neuronal network in comparison with a subtraction strategy data analysis. These authors again note that functional imaging is impacted by age related changes,
2603
2604
154
The Wada test-60th year anniversary update-in epilepsy surgery
decrease in cortical volume in age, gender differences and handedness of volunteers. The presence of atrophy may adversely impact on metabolic changes and may artificially lower regional metabolism and the interpretation of PET images [200]. Additional correlation with structural abnormalities on CTor MRI needs to be done in such situations. Silverman also notes that changes in cerebral metabolism due to normal development and aging are present in PET. Other potential confounding variables include sex, handedness, sensory environment, level of alertness, mood, drug effects, serum glucose levels and head fraction (the portion of administered tracer that passes into the brain). There can also be right hemisphere activation in language comprehension and production tasks in PET [201]. Again it remains unclear if the subjects had bilateral language dominance or the right hemisphere activation was related to a network of attentional processes. The authors acknowledge the right hemisphere’s role in language including prosody, automatic idiom processing as well as vocabulary storage and the possible affective content of language. Apart from the technical challenges of producing and handling of short-lived radioactive material, PET has two additional limitations. It is too slow, and safety issues limit the number of times a person can be a subject as well as the number of bolus injections given per session (hence the independent task and baseline sessions). Thus, there must be considerable infrastructure present. In turn, more investigations are needed so that variance due to scanning protocols, etc. can be accounted for different outcomes.
by capturing single photons emitted by injected radioactive material. Regional blood flow is expected to increase with cerebral activation secondary to a given cognitive task performance. SPECT has been used to evaluate language lateralization in epilepsy patients with fluent verbal activation task with a reasonable success [133] while inconclusive data were reported with auditory verbal stimulation [202]. Using this technique, there has been a suggestion that injection of sodium amytal had greater functional effect on the dominant hemisphere in comparison with the nondominant hemisphere. Wada tested patients with bilateral speech showed no consistent trends. The dosage of amytal did not appear to be a factor. They suggested that their SPECT study may be revealing greater physiological effects following injection of the dominant hemisphere and in turn, why memory failures during Wada testing may be affected by the effect of dominance [203]. SPECT has been appealing due to its affordability and accessibility. However, spatial and temporal resolution is relatively poor. Scanning is time-consuming and again there can be no movement during scanning thus limiting its application to certain populations. Artifacts can occur due to uneven distribution of the radiotracer. There can also be underestimation of activity in deep structures. Coregistration with MRI images enhances anatomic accuracy. Based on a complete concordance of results between SPECT and fTCD in the assessment of language dominance in 14 and 3 patients with epilepsy and AV malformation, respectively, a suggestion was made that language mapping using SPECT might be considered when fMRI or PET are not available or contraindicated [133].
SPECT – Single Photon Emission Tomography
Current Status and Challenges
Another somewhat invasive nuclear medicine technique, SPECT measures cerebral blood flow
The Wada test has its’ own challenges including universal definitions of atypical dominance,
The Wada test-60th year anniversary update-in epilepsy surgery
particularly bilateral dominance. One operational definition of when bilateral dominance is present is when the patient is speaking or naming an object with either the right hemisphere or left hemisphere. Some have included performance on receptive language tasks as well. We have seen individual differences such as what appears to be the transfer of speech production to the other hemisphere but not the posterior comprehension language skills. The Wada test has stimulated interest and provided a new insight in to cerebral organization of language and conceptual frameworks about reorganization of cerebral functioning. It has been reported that patients will occasionally have catastrophic or emotional reactions that can interfere with testing. Detailed evaluation of 5/81 Wada tested patients who developed bizarre behavior failed to identify predictive factors and that premorbid personality factors did not appear to play a role. The authors suggested that thorough explanation and simulation might be helpful to lessen this risk [204]. At UBC, as a part of the preprocedure education, we would explain the procedure and would do a simulated, mock procedure. Patients are encouraged to ask questions. Anxiety, which usually contributes to the catastrophic reactions, was greatly reduced due to this effort. Introducing and orienting them to the angiogram suite and staff have also helped. Similarly concerns about unavoidable drowsiness at times is dealt with by preassessing important factors and by adjusting amytal dose accordingly with great success. Risk of morbidity from the invasive aspects of Wada Test has always been a legitimate concern. Minor and major complications up to 11.6% in 79 Wada tested patients over 4-year period have been reported [60]. Review of relevant data specifically related to femoral artery puncture and internal carotid catheterization that are involved in the Wada test may be helpful. Recent large series of studies on the topic, (i.e., neurological complication rate of cerebral angiography) involving 2,524 [205] and 2,899 [206]
154
procedures report 0.34 and 1.3% respectively. Two retrospective studies of neurological complication involving 569 [207] and 19,626 [208] procedures report morbidity rate of 0.5 and 2.63% respectively. The latter study involved many patients with acute severe cerebrovascular disease and had mortality of 0.06%. Common risk factors identified by these studies are advancing age (>55 years of age), acute cerebrovascular disease, duration of procedure (>10 min) and the degree of skill/expertise of the operator involved. As mentioned earlier, a recent survey of 16 Wada testing European epilepsy centers had 1.02% morbidity during a period of 5 years (personal communication Frank Oltmans). Obviously, risk and benefit have to be carefully weighed while considering the Wada test for older patients with or without cerebrovascular risk factors. With our UBC protocol, testing was performed on separate days at least 24 h apart. Following the initial injection, catheter was prompty withdrawn to aorta as soon as desired information was judged to have been obtained. Rarely a second small bolus of amytal injection was given only when a sufficient neurological deficit did not occur. We have been fortunate not to have experienced any neurological morbidity or mortality since 1960 when the test began at UBC. The advantage of the traditional Wada protocol is that the results are evident rather quickly for most cases where establishing language dominance is crucial prior to surgery. When atypical dominance may be suspected, such as bilateral hemispheric dominance for language, more rigorous testing may be considered both during Wada testing and in utilizing other techniques such as comprehensive neuropsychological testing or other neuroimaging techniques. Wada testing does not require considerable manpower and is typically done in existing angiogram suites. Our approach has always been to run simultaneous EEG recordings during testing as an adjunct objective measure of drug effect.
2605
2606
154
The Wada test-60th year anniversary update-in epilepsy surgery
Following a hand injected angiogram, the typical time needed for testing and interpretation of test results can be done within an hour appealing to the added cost-effectiveness of the technique. New techniques have attempted to validate by comparing them with results of the Wada test. fMRI studies correlated with Wada test results have shown the correlation with activation asymmetry in the frontal but not temporal lobes [209] and that laterality activation can differ between the frontal and temporal regions. Is it possible that fMRI may not be able to pick up brain asymmetries for all tasks, in all regions of the brain? fMRI as an activation technique may reveal areas that are involved in processing a task but may not reveal all areas of the brain involved in completing the task [210]. The nature and degree of difficulty of the task may impact on how the subject approaches the task. Some subjects may visualize the stimuli rather than focus on the linguistic aspects of the task, thus activating different areas and hemispheres of the brain. In patients with epilepsy, does the epileptogenic focus impact on the activation/inactivation of certain neural networks? What does the task become with repetition? Does novelty to the task impact on the patterns of activation? Is it the same task? Does familiarity and/or learning alter the sites of activation? Intriguing results from repeat fMRI studies on the same individuals [210] suggest that laterality clearly remained the same but localization was more consistent for frontal sites but more variability occurred in the temporal sites. These results were interpreted as suggesting that cortical mapping may still be required to localize all critical language zones. Major techniques such as fMRI, MEG and PET require a model of brain functioning in terms of knowing what stimuli or task should be chosen and speculating what areas of the brain (ROI) should be activated. However in behavioral neurosciences, the understanding of activation may be limited as we infer brain functioning from normal controls to people who may have longstanding neurological conditions (i.e., epilepsy
from birth) or atypical dominance, which could impact on cortical and subcortical activation. There are normal, developmental changes underlying the neural basis of cognitive functioning such as the final development of white matter pathways/connections continuing through the first two decades of life [211]. Also, different neural networks may be activated based on an individual’s experience, familiarity and the duration of training with the task which has been demonstrated in fMRI and MEG images in the processing of music [212,213]. Other limitations of neuroimaging techniques include claustrophobia, the presence of metal (e.g., orthopedic screws) and the inability to stay still are typically the behavioral limitations leading to a patient’s inability to endure techniques involving neuroimaging. Some centers will only do fMRI scanning for research purposes. The challenge for the fMRI approach in language and memory testing is what and how will the numerous confounding variables such as compromised intellectual functioning and/or pre or perinatal insult, impact on the activation patterns on fMRI? Determining what the control state will be is important. Typically it is a resting state but one wonders what a ‘‘neutral control’’ state is. The subject could very well be thinking about her grocery-shopping list that she must complete on her way home while she is lying in the gantry. The other important question is ‘‘are the neural networks being activated during a task essential or primary?’’ For example, during a naming task, cells 1, 2 and 3 are activated but in fact, cell 2 is activated more intensely when there has been injury to cell 1 and 3. It remains unclear how plasticity and neural networks are impacted when there is structural versus an epileptogenic, nonstructural seizure focus in the temporal lobe at an early age versus later onset? Imaging results may differ between normal populations and clinical populations and future studies must be responsible for providing clarification regarding these differences.
The Wada test-60th year anniversary update-in epilepsy surgery
In memory processing, MEG has shown bilateral hemispheric activation for some tasks such as tone recognition. Despite the correlation of fMRI and Wada test results, there are still limitations in the assessment of atypical and bilateral hemispheric language dominant cases. The bilateral hemispheric activation and the activation of nonlanguage areas continue to exist and protocols that are sensitive to asymmetrical activation must be devised. fMRI studies of reliable individual predictions of postoperative memory changes are sparse. As Meador and Loring [26] pointed out that while group studies looked promising, fMRI on an individual basis had been less consistent with respect to the relationship between fMRI memory activation and memory outcomes following temporal lobectomy. Furthermore, chronological change in memory activation pattern may occur postoperatively as mentioned in the previous case study of postoperative changes in language activation pattern [42]. It remains unclear how stable the postoperative memory activation patterns are over time when one is investigating the relationship between pre-to postoperative memory imaging. It has been shown with a more basic motor deficit, a strong prediction from preoperative fMRI data to immediate post-operative motor deficit following a resection of medial frontal cortex but the correlations were absent following a few months [214]. It remains unclear how time impacts on pre to postoperative memory fMRI patterns. The basis for establishing cognitive tasks for both behavioral neurosciences and clinicians utilizing newer neuroimaging techniques will be the accumulation of normative data as it remains unclear how the dynamic changes of a developing brain and critical periods for learning languages impacts on neuroimaging studies. Part of the challenge in trying to critically evaluate the clinical utility and reliability of any alternate technique are the small N’s, span of age groups, diversity of cognitive tasks and paradigms used, differences in data analysis, etc. Our review of the
154
literature is that there is much promise with many of these new techniques but the current state of the art does not allow one yet to unequivocally state that the ‘‘ABC technique’’ provides valid and reliable data, particularly for memory functioning for the majority of presurgical epilepsy patients. It will take years of data collection to determine if in fact memory networks are activated in a language paradigm. At this time, it is also too early to dismiss all ‘‘nonlanguage activated areas’’ as methodological or statistical artifacts. Prospective studies with samples that consider age and education effects as well as improvements in technical methodology will advance our understanding. Has the Wada test been replaced? Not completely. Many of the newer neuroimaging tests are not easily accessible and affordable to some clinicians or can yield the test results within the test session due to the statistical processing and data analysis. The drowsiness and obtundation concerns from the Wada test may be no different from the concerns about claustrophobia or inability to stay still in a fMRI gantry. There are also some limitations regarding administration of test stimuli when using MEG technique (i.e., patients being unable to wear glasses, eye movement contributing artifact). There are some complex patient presentations that the Wada test may be the most appropriate technique (i.e., cerebral palsy patients with intractable seizures). We have also very limited understanding of how the alternative techniques are impacted by ESL issues or gender issues. Also requests for unique testing situations (i.e., assessing language dominance in someone who is mute/deaf but uses American Sign Language to communicate) can occur in presurgical investigations. Functional neuroimaging would not have been able to accommodate this situation due to the confound of motor movement from signing. Our own clinical experience has revealed that the presurgical evaluation of ESL patients using translators or patients with motor difficulties such as cerebral palsy can be investigated using the traditional Wada technique.
2607
2608
154
The Wada test-60th year anniversary update-in epilepsy surgery
Neuroimaging approaches need to travel the same journey of concerns as the Wada test has regarding validity and reliability across age groups, the possible impact of education effects, novelty/familiarity with the task, establishing operational definitions for atypical patterns of cerebral language dominance, and how to handle the variability among centers with their test protocols. One might argue that the variability in protocols or the lack of standardization for a universal protocol has actually promoted further understanding about the subtleties and influences of test administration and test stimuli. Until there are numerous studies with larger Ns for many of the newer techniques, the data from the Wada test continue to provide valuable clinical information that is well understood in terms of its contributions and limitations particularly for individual cases. Future imaging studies need to be clear about their apriori purpose as the field evolves. The memory testing of the Wada test initially was for the sole purpose of attempting to identify at risk patients for postoperative global amnesia. Over time, the memory scores from Wada testing have been utilized to confirm seizure focus laterality, predict postoperative memory functioning, and material specificity and sensitivity of the temporal lobes. How much of the academic insight gained from neuroimaging regarding language and memory activation patterns lends itself to clinical utility remains unknown at this time. Neuroimaging is plagued with the same issues as the Wada test including differences in protocols, data analyses, interpretation, and reliability issues for between groups as well as within the individual. Abou-Khalil [215] provides an overview of alternative techniques for determining language dominance and notes that repetitive magnetic stimulation (rTMS) does provide a methodology for deactivation of language cortex. Although fMRI and MEG have potential for determining language dominance, he is clear that these techniques have not yet proved to be a
replacement for memory testing via the Wada approach. For many centers without easy access to fMRI, the Wada test results along with other clinical information such as the patient’s semiology, peri and postictal cognitive state, neuropsychological test results, patient’s history and neuroimaging can provide the necessary information to make the decision about risk for significant postoperative language or memory disturbance. Killgore, et al. [216] have suggested that fMRI and Wada test data provide complementary data that may not correlate, but both accounted for nonover lapping unique variance in predicting surgical outcome. Finally, advantages of non- or minimally invasive brain imaging technologies over Wada test in clinical setting are well recognized and the availability of multiple approaches will add to our knowledge of localization of function and the nature of reorganization. However, from both the clinical and scientific point of view they are all handicapped because they are necessarily based on highly artificial experimental protocols of limited scope. We believe that the true potential of brain imaging will be realized only when the technology allows one not only (1) to assess ecological validity of the subject-oriented approach and findings, (2) to extend beyond our current preoccupation of answering the question of ‘‘where’’ to ‘‘how’’ regarding the process of language and memory, but also (3) to address quintessential issues of natural language in health and disease, i.e., human communication from multidimensional interactive perspective including symbols, memories and emotions.
Epilogue Conceptualization of the carotid amytal injection 60 years ago was a reflection of a life that was saved by a fluke from an act of the Second World War and then engaged in youthful
The Wada test-60th year anniversary update-in epilepsy surgery
exploration that has ultimately matured to stand the test of time. For new cognitive evaluation technologies, issues of activation/deactivation, temporal/spatial resolution, standardization/validation, ecological validity and the reason for discordance between them and the Wada test, are all under debate while developing countries face the issue of affordability and availability. With surgical outcome as the gold standard, our accumulated experience has led to progressive sharpening and sophistication of clinical perception. Ultimately, there may be a smaller percentage of pre-surgical candidates requiring indepth pre-surgical evaluation of language and memory which is an integral part of cohesive multimodal approaches to ensure patient’s safety and success of surgical intervention. Clinical acumen reminds us to consider such factors to define specific investigational strategy as incongruent handedness between the patient and family history, history of early brain insult, peri/post-ictal phenomena, unclear semiology or/and semiology that is inconsistent with EEG localization, ‘‘normal’’ or non-focal EEG/MRI in partial epilepsy, and a left temporal lobe epilepsy focus with the absent past history of post-ictal language disturbance. The Wada test was never meant to be a stand-alone test or an unequivocal evaluation of determination of language and memory functioning. Any investigation for epilepsy surgery is only a complementary component, the use of which depends on specific circumstances, clinical wisdom and judgment. We fully expect that noninvasive neuroimaging approach to become the new standard one day [2]. As expressed 11 years ago [217] while we await the arrival of validated safe alternatives, judicious and innovative use of the carotid amytal deactivation by a skilled, experienced clinician, when justified, cannot only continue to help patients but also create new information and hypotheses on the mechanism of function and dysfunction of the human brain in the behavioral state.
154
References 1. Wada JA. A new method for the determination of the side of cerebral speech dominance: a preliminary report on the intracarotid injection of sodium amytal in man. Igaku to Seibutsugaku 1949;14:221-2. 2. Wada JA. Fateful encounter: sixty years later – reflection on the Wada test. Epilepsia 2008;49:726-7. 3. Wada JA, Rasmussen T. Intracarotid injection of Sodium amytal for the lateralization of cerebral speech dominance. J Neurosurg 1960;17:266-82. 4. Penfield W, Jasper H. Epilepsy and the functional anatomy of the human brain. 2nd ed. Boston: W. Little, Brown and Co; 1954. 5. Meador K, Loring D. The Wada test: controversies, concerns, and insights. Neurology 1999;52:1535-6. 6. Meador K, Loring D, Lee G, et al. Right cerebral specialization for tactile attention as evidenced by intracarotid sodium amytal. Neurology 1998;38:1763-6. 7. Davies K, Risse G, Gates J. Naming ability after tailored left temporal resection with extraoperative language mapping: increased risk of decline with later epilepsy onset age. Epilepsy Behav 2005;7:273-8. 8. Homberg V, Netz J. Generalized seizures induced by transcranial magnetic stimulation of the motor cortex. Lancet 1989;334: 1223 Letter. 9. Breier J, Simos P, Zouridakis G. Language dominance determined by magnetic source imaging: comparison with the Wada procedure. Neurology 1999;53:938-45. 10. Oguro K, Yokota H, Watanabe E. Non-invasive evaluation of language function by optical topography (Japanese). Rinsho-Noha (Clinical Electroencephalography) 2008;50:110-17. 11. Buchtel H, Passaro E, Selwa L, et al. Sodium Methohexital (Brevital) as an anesthetic in the Wada test. Epilepsia 2002;43:1056-61. 12. Loddenkemper T, Moeddel G, Morris H. Complications during the intracarotid amobarbitol test. Neurology 2004;62:A248-9. 13. Masters L, Perrine K, Devinsky O, et al. Wada testing in pediatric patients by use of propofol anesthesia. Am J Neuroradiol 2000;21:1302-5. 14. Mikuni N, Satow T, Hayashi N, et al. Evaluation of adverse effects in intracarotid propofol injection for Wada test. Neurology 2005;65:1813-6. 15. Jones-Gotman M, Sziklas V, Djordjevic J, et al. Etomidate speech and memory test (eSAM). Neurology 2005;65:1723-9. 16. Rausch R, Silfvenius HG, Dodrill C, et al. Intrarterial amobarbital procedures. In: Engel P, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1983. p. 341-57. 17. Baxendale S, Thompson P, Duncan J. The role of the Wada test in the surgical treatment of temporal lobe epilepsy: an international survey. Epilepsia 2008;49:715-9.
2609
2610
154
The Wada test-60th year anniversary update-in epilepsy surgery
18. Kosaka B. Wada Testing. Online posting. May 9, 2007. www.npsych.com. 19. Yetkin F, Swanson S, Fischer M, et al. Functional MR of frontal lobe activation comparison with Wada language results. AJNR Am J Neuroradiol 1998;19:1095-8. 20. Patarala E, Simos P, Castillo E, et al. Reorganization of language specific cortex in patients with lesions or mesial temporal epilepsy. Neurology 2004;63:1825-32. 21. Kosaka B, Wada JA, Gildenberg P, et al. Textbook of stereotactic and functional neurosurgery. New York: McGraw-Hill; 1998. p. 1801-7. 22. Robertson E, Theoret H, Pascual-Leone A. Studies in cognition: the problems solved and created by transcranial magnetic stimulation. J Cogn Neurosci 2003; 15:948-60. 23. Milner B. Amobarbital memory testing: some personal reflections. Brain Cogn 1997;33:14-17. 24. Jones-Gotman M. Commentary: psychological evaluation – testing hippocampal function. In: Engel J Jr, editor. Surgical treatment of epilepsy. New York: Raven Press; 1987. p. 203-11. 25. Milner B, Branch C, Rasmussen T. Study of short-term memory after intracarotid injection of sodium amytal. Trans Am Neurol Assoc 1963;87:224-226. 26. Meador K, Loring D. The Wada test for language and memory lateralization. Neurology 2005;65:659. 27. Simos PG, Breier JH, Zouridakis G, et al. Identification of language specific brain activity during magnetoencephalography. Exp Neuropsychol 1998;20:706-22. 28. Gotman J, Bouwer M, Jone-Gotman M, et al. Intracranial EEG study of brain structures affected by internal carotid injection of amobarbital. Neurology 1992; 42:2136-43. 29. Thivard L, Hombrouck J, Tezenas S, et al. Productive and perceptive language reorganization in temporal lobe epilepsy. NeuroImage 2005;24:841-51. 30. Grote C, Wierenga C, Smith M, et al. Wada difference a day makes. Interpretative cautions regarding same-day injections. Neurology 1999;52:1577-82. 31. Ojemann J, Kelley W. The frontal lobe role in memory: a review of convergent evidence and implications for the Wada memory test. Epilepsy Behav 2002;3:309-15. 32. Dodrill CB, Ojemann G. An exploratory comparison of three methods of memory assessment with the intracarotid amobarbital procedure. Brain Cogn 1997;33:210-23. 33. Dodrill CB. Preoperative criteria for identifying eloquent brain. Intracarotid amytal for language and memory testing. Neurosurg clin N Am 1993;4:211-16. 34. Smith I, McGlone J, Fox A. Intracarotid amobarbital memory protocol: muteness, dysphasia, and variations in arterial distribution of the drug do not affect recognition results. J Epilepsy 1993;6:75-84. 35. Griffin S, Tranel DJ. Age of seizure onset, functional reorganization, and neuropsychological outcome in temporal lobectomy. J Clin Exp Neuropsychol 2007;29:13-24.
36. Rasmussen T, Milner B. The role of early left-brain injury in determining lateralization of cerebral speech functions. In: Diamond S, Blizzard D, editor. Evolution and lateralization of the brain, vol. 299. New York: New York Academy of Sciences; 1977. p. 355-69. 37. Saltzman-Benaiah J, Scott K, Smith ML. Factors associated with atypical speech representation in children with intractable epilepsy. Neuropsychologia 2003;41:1967-74. 38. Satz P, Strauss E, Wada J. Some correlates of intra- and inter-hemispheric speech organization after left focal injury. Neuropsychologia 1998;26:345-50. 39. Strauss E, Satz P, Wada J. An examination of the crowding hypothesis in epileptic patients who have undergone the carotid amytal test. Neuropsychologia 1990;28:1221-7. 40. Woods RP, Dodrill CR, Ojemann GA. Brain injury, handedness, and speech lateralization in a series of amobarbital studies. Ann Neurol 1988;23:510-8. 41. Boatmann D, Freeman J, Vining E, et al. Language recovery after left hemispherectomy in children with lateonset seizures. Ann Neurol 1999;46:579-86. 42. Hertz-Pannier L, Chiron C, Jambaque I, et al. Late plasticity for language in a child’s non-dominant hemisphere. A pre and post-surgery fMRI study. Brain 2002;125:361-72. 43. Loddenkemper T, Wyllie E, Lardizabal D, et al. Late language transfer in patients with Rasmussen Encephalitis. Epilepsia 2003;44:870-1. 44. Telfeian A, Berqvist G, Danielack C. Recovery of language after left hemispherectomy in a 16-year old girl with late onset seizures. Pediatr Neurosurg 2002;37:19-21. 45. Rey M, Dellatolas G, Bancaud J, et al. Hemispheric lateralization of motor and speech functions after early brain lesion: study of 73 epileptic patients with intracarotid Amytal test. Neuropsychologia 1988;26:167-72. 46. Wada JA, Davies AE. Fundamental nature of human infant’s brain asymmetry. Can J Neurol Sci 1977;4:203-7. 47. Liegeois F, Connelly A, Gross J. Language reorganization in children with early-onset lesions of the left hemisphere: an fMRI study. Brain 2004;127:1299-36. 48. DeVos KJ, Wylie E, Geckler C, et al. Language dominance in patients with early childhood tumours near left hemisphere language areas. Neurology 1995;45:349-56. 49. Duchowny M, Jayakar P, Harvey AS, et al. A Language cortex representation: effects of development versus acquired pathology. Ann Neurol 1996;140:31-8. 50. Burneo J, Kuzinelky R, Babib D, et al. Cortical reorganization in malformation of cortical development: a magnetoencephalographic study. Neurology 2003;63:1818-24. 51. Adcock J, Wise R, Oxbury J, et al. Quantitative fMRI assessment of the differences in lateralization of languagerelated brain activation in patients with temporal lobe epilepsy. NeuroImage 2003;18:423-38. 52. Backes W, Deblaere K, Vonck K, et al. Language activation distributions revealed by fMRI in post-operative epilepsy patients: differences between left- and right-sided resections. Epilepsy Res 2005;66:1-12.
The Wada test-60th year anniversary update-in epilepsy surgery
53. Staudt M, Lidzba K, Grodd W, et al. Right hemisphere organization of language following early left sided brain lesions: fMRI topography. NeuroImage 2002; 16:954-67. 54. Voets N, Adcock E, Flitney D, et al. Distinct right frontal lobe activation in language processing following left hemisphere injury. Brain 2006;129:754-66. 55. Janszky J, Martens M, Janszky I, et al. Left-sided interictal epileptic activity induces shift of language lateralization in temporal lobe epilepsy: an fMRI study. Epilepsia 2006;47:921-7. 56. Holmes GI, Lenck-Santini PP. Role of interictal epileptiform abnormalities in cognitive impairment. Epilepsy Behav 2006;6:504-15. 57. Kurthen M, Helmstaedter C, Linke D, et al. Quantitative and qualitative evaluation of patterns of cerebral language dominance. Brain Lang 1994;46:536-64. 58. Risse G, Gates J, Fangman M. A reconsideration of bilateral language representation based on the intracarotid amobarbital procedure. Brain Cogn 1997;33:118-32. 59. Rutten G, Ramsey M, Van Rijen P, et al. fMRI determined language lateralization in patients with unilateral or mixed language dominance according to the Wada test. NeuroImage 2002;17:447-60. 60. Loddenkemper T, Dinner D, Kubu C, et al. Aphasia after hemispherectomy in an adult with early epilepsy and hemiplegia. J Neurol Neurosurg Psychiatry 2004;75:149-51. 61. Inoue T, Watanabe Y, Funakoshi D, et al. Relation of hand preference and cerebral speech representation as determined by the sodium amobarbital procedure (Wada Test). Jpn J Epilepsy Res 1994;12:125-31. 62. Wylie E, Lueders H, Murphy D, et al. Intracarotid amobarbital (Wada) test for language dominance: correlation with cortical stimulation. Epilepsia 1990;31:156-61. 63. Kamada K, Takeuchi F, Kuriki S, et al. Dissociated expressive and receptive language functions on magnetoencephalography, functional magnetic resonance imaging, and amobarbital studies. J Neurosurg 2006;104:598-608. 64. Kurthen M, Helmstaedter G, Linke DB, et al. Interhemispheric dissociation of expressive and receptive language function in patients with complex partial seizures: an amobarbital study. Brain Lang 1992;43:694-712. 65. Ries M, Boop F, Griebel M, et al. Functional MRI and Wada determination of language lateralization: a case of crossed dominance. Epilepsia 2004;45:85-9. 66. Loring D, Meadow K, Lee G, et al. Crossed aphasia in a patient with complex partial seizures: evidence from intracarotid amobarbital testing, functional cortical mapping, and neuropsychological assessment. J Clin Exp Neuropsych 1990;12:340-54. 67. Helmstaedter C, Kurthen M, Gleissnetu, et al. Natural atypical language dominance and language shift from right to left hemisphere in right hemisphere pathology. Naturwissenshaften 1997;84:250-2.
154
68. Knecht S, Draeger A, Floel L, et al. Behavioural relevance of atypical language lateralization in healthy subjects. Brain 2001;124:1657-65. 69. Ashtari M, Cervellione K, Hasan K, et al. White matter development during late adolescence in healthy males: a cross-sectional diffusion tensor imaging study. Neuroimage 2007;35:501-10. 70. Pujol J, Deus J, Lossila J, et al. Cerebral lateralization of language in normal left handed people studied by fMRI. Neurology 1999;52:1038-43. 71. Springer J, Binder J, Hammeke T, et al. Language dominance in neurologically normal and epilepsy patients subjects: a fMRI study. Brain 1999;122:2933-46. 72. Knecht S, Deppe B, Draeger B, et al. Language lateralization in healthy right-handers. Brain 2000;123:74-81. 73. Berl M, Balsamo L, Xu B, et al. Seizure focus affects regional language networks assessed by fMRI. Neurology 2005;65:1604-11. 74. Bradzil M, Chlebus P, Mikl M, et al. Reorganization of language related neuronal networks in patients with left temporal lobe epilepsy-an fMRI study. Eur J Neurol 2005;12:268-75. 75. Noppeney U, Price C, Duncan J, et al. Reading skills after left temporal lobe resection: an fMRI study. Brain 2005;128:1377-85. 76. Satz P, Strauss E, Hunter M, et al. Re-examination of the crowding hypothesis. Neuropsychologia 1994; 34:255-62. 77. Weber B, Wellmer J, Reuber M, et al. Left Hippocampal pathology is associated with atypical language lateralization in patients with focal epilepsy. Brain 2006;129:346-51. 78. Schwartz T, Devinsky O, Doyle W, et al. Preoperative predictors of anterior temporal language areas. J Neurosurg 1998;89:962-70. 79. Jabbour RA, Penovich PE, Risse GL, et al. Atypical language cortex in the left temporal lobe: relationship to bilateral language. Neurology 2004;68:1833-7. 80. Gleissner U, Helmstaedter C, Schrammm J, et al. Memory outcome after selective amygdalohippocampectomy. A study in 140 patients with temporal lobe epilepsy. Epilepsia 2002;43:87-95. 81. Hermannn B, Seidenvberg M, Haltiner A, et al. Relationship of age at onset, chronological age, and adequacy of preoperative performance to verbal memory change after anterior temporal lobectomy. Epilepsia 1995;36:137-45. 82. Saykin P, Gur R, Sussman N, et al. Memory deficits before and after temporal lobectomy. Effect of laterality and age of onset. Brain Cogn 1989;9:191-200. 83. Jokeit H, Ebner A. Long term effects of refractory temporal lobe epilepsy on cognitive abilities: a cross sectional study. J Neurol Neurosurg & Psychiat 1999;67:44-50. 84. Golby A, Poldrack R, Illes J. Memory lateralization in medial temporal lobe epilepsy assessed by functional MRI. Epilepsia 2002;43:855-63.
2611
2612
154
The Wada test-60th year anniversary update-in epilepsy surgery
85. Richardson M, Strange B, Duncand J, et al. Preserved memory function in left medial temporal pathology involves reorganization of function to right medial temporal lobe. NeuroImage 2003;20:5112-9. 86. Grunwald T, Lehnerz K, Helmstaedter C. Limbic ERPs predict verbal memory after left-sided hippocampectomy. Neuroreport 1998;9:3375-8. 87. Urbach H, Kurthen M, Klemm E, et al. Amobarbital effects on the posterior hippocampus during the intracarotid amobarbital test. Neurology 1999;52:1596-602. 88. Powell H, Richardson M, Symms M, et al. Reorganization of verbal and non-verbal memory in temporal lobe epilepsy due to unilateral hippocampal sclerosis. Epilepsia 2007;48:1512-25. 89. Helmstaedter C, Kurthen DB, Elger CE. Right hemisphere restitution of language and memory functions in right hemisphere language dominant patients with left temporal lobe epilepsy. Brain 1994;117:729-37. 90. Jokeit H, Ebner A, Holthausen H, et al. Reorganization of memory functions after human temporal lobe damage. Neuroreport 1996;7:1627-30. 91. Wood A, Saling M, O’Shea M, et al. Reorganization of verbal memory and language: a case of dissociation. J Int Neuropsychol Soc 1999;5:3375-8. 92. Shevron C, Carlson C, Zaroff C, Weiner H, Doyle W, et al. Pediatric language mapping: sensitivity of neurostimulation and Wada testing in epilepsy surgery. Epilepsia 2007;47:1-7. 93. Shtyrov Y, Pihko E, Pulvermueller F. Determinants of dominance: language laterality explained by physical or linguistic features of speech? NeuroImage 2005;27:37-47. 94. Bourne V. The divided visual field paradigm: methodological considerations. Laterality 2006;11:373-93. 95. Hunter Z, Brysbaert M. Visual half-field experiments are a good measure of cerebral language dominance if used properly: evidence from fMRI. Neuropsychologia 2008;46:316-25. 96. Kimura D. Functional asymmetry of the brain in dichotic listening. Cortex 1967;3:163-78. 97. Jancke L, Shah N. Does dichotic listening probe temporal lobe functions? Neurology 2002;58:736-43. 98. Strauss E, Gaddes E, Wada J. Performance on a freerecall verbal dichotic listening task and cerebral dominance determined by the carotid amytal test. Neuropsychologia 1987;25:747-53. 99. Zattore R. Perceptual asymmetry on the dichotic fused words test and cerebral speech lateralization determined by the carotid sodium amytal test. Neuropsychologia 1989;27:1207-19. 100. Fernandes M, Logan W, Crawley A, et al. Comparing language lateralization determined by dichotic listening and fMRI activation in frontal and temporal lobes in children with epilepsy. Brain Lang 2006;96:106-14. 101. Davis A, Wada J. Lateralization of speech dominance by spectral analysis of evoked potentials. J Neurol Neurosurg Psychiat 1994;40:1-4.
102. Gerschlager W, Lalouschek W, Lehrner J, et al. Language related hemispheric asymmetry in healthy and patients with temporal lobe epilepsy as studied by event related potentials and intracarotid amobarbital test. Electroencephalogr clin Neurophysiol 1998; 108:274-82. 103. Wada JA. Pre-language and fundamental asymmetry of the infant brain. Ann NY Acad Sci 1977;299:370-9. 104. Wada JA, Clark R, Hamm A. Cerebral hemispheric asymmetry in humans: cortical speech zones in 100 adults and 100 infants brains. Arch Neurol 1975;32:239-46. 105. Foundas A, Leonard C, Gilmore R, et al. Planum temporale asymmetry and language dominance. Neuropsychologia 1994;32:1225-3. 106. Dorsaint-Pierre R, Penhune V, Watkins K, et al. Asymmetries of planum temporale and Hescl’s gyrus: relationship to language lateralization. Brain 2006;129:1164-76. 107. Basser PJ, Pajevic S, Pierpaoli C, et al. In vivo fiber tractography in human brain using diffusion tensor MRI data. Magn Reson Med 2000;44:625-32. 108. Catani M, Howard R, Pajevic S, et al. Virtual in vivo dissection of white matter fascicule in the human brain. NeuroImage 2002;17:77-94. 109. Catani M, Jones D, Fytche H. Perisylvian language networks of the human brain. Ann Neurol 2005;57:8-16. 110. Buechel T, Raedler T, Sommer M, et al. White matter asymmetry in the human brain. A diffusion tensor MRI study. Cerebral Cortex 2004;14:945-51. 111. Nagy Z, Westerberg H, Klinberg T. Maturation of white matter is associated with the development of cognitive functions during childhood. J Cogn Neurosci 2004;16:1227-33. 112. Briellmann RS, Mitchell LA, Waites AB, et al. Correlation between language organization and diffusion tensor abnormalities in refractory partial epilepsy. Epilepsia 2003;44:1541-5. 113. Heller S, Heier L, Watts R, et al. Evidence of cerebral re following perinatal stroke demonstrated with fMRI and DTI tractography organization. J Clin Imaging 2005;29:283-7. 114. Diehl B, Busch R, Duncan J, et al. Abnormalities in diffusion tensor imaging of the uncinate fasciculus related to reduced memory in temporal lobe epilepsy. Epilepsia 2008; [Epub ahead of print]. 115. Carne R, O’Brien T, Kilpatrick C, et al. MRI-negative PET-positive temporal lobe epilepsy: a distinct surgically remediable syndrome. Brain 2004;127:2276-85. 116. Lineweaver T, Morris H, Naugle R, et al. Evaluating the contribution of the state of the art assessment techniques to predicting memory outcome after unilateral anterior temporal lobectomy. Epilepsia 2006; 48:1895-03. 117. Keller S, Roberts N. Voxel-based morphometry of temporal lobe epilepsy: an introduction and review of the literature. Epilepsia 2008;49:741-57.
The Wada test-60th year anniversary update-in epilepsy surgery
118. Jennum P, Friberg L, Fukulasang-Frederiksen A, et al. Speech localization using repetitive transcranial magnetic stimulation. Neurology 1994;44:269-73. 119. Pascual-Leone A, Gates J, Dhuna A. Induction of speech arrest and counting errors with rapid-rate transcranial magnetic stimulation. Neurology 1991; 41:697-702. 120. Wassermann E, Blaxton T, Hoffman E, et al. Repetitive transcranial magnetic stimulation of the dominant hemisphere can disrupt visual naming in temporal lobe epilepsy patients. Neuropsychologia 1999;37:537-44. 121. Stewart L, Walsh V, Frith U, et al. TMS produces two dissociable types of speech disruption. NeuroImage 2001;13:472-8. 122. Epstein C, Woodard J, Stringer A, et al. Repetitive transcranial magnetic stimulation does not replicate the Wada test. Neurology 2000;55:1025-7. 123. Theodore W. Transcranial magnetic stimulation in epilepsy. Epilepsy Curr 2003;3:191-7. 124. Duzel E, Hufnagel A, Helmstaedter C, et al. Verbal working memory components can be selectively influenced by transcranial magnetic stimulation in patients with left temporal lobe epilepsy. Neuropsychologia 1996; 34:775-83. 125. Matthews PM, Adcock J, Chen Y, et al. Towards understanding language organization in the brain using fMRI. Hum Brain Mapp 2003;18:230-47. 126. Devlin J, Watkins K. Stimulating language: insights from TMS. Brain 2007;130:610-23. 127. Aaslid R. Visually evoked dynamic blood flow response of the human cerebral circulation. Stroke 1987;18: 771-5. 128. Silverstini M, Cupini L, Matteis M, et al. Bilateral simultaneous assessment of cerebral flow velocity during mental activity. J Cereb Blood Flow Metab 1994;14:643-8. 129. Knecht S, Deppe M, Ebner A, et al. Noninvasive determination of language lateralization by functional transcranial Doppler sonography. Stroke 1998;2:82-6. 130. Deppe M, Knecht S, Papke R, et al. Correlation of cerebral blood flow velocity and regional cerebral blood flow during word generation. NeuroImage 1998;7:S448. 131. Rihs R, Sturzenegger M, Gutbrod K, et al. Determination of language dominance Wada test confirms functional transcranial dopper sonography. Neurology 1999;52:1591-6. 132. Knake S, Haag A, Hamer H, et al. Language lateralization in patients with temporal lobe epilepsy: a comparison of functional transcranial Doppler sonography and the Wada test. NeuroImage 2003;19:1228-32. 133. Borbely K, Gjedde A, Nyary I, et al. Speech activation of language dominant hemisphere: a single photon computed tomography study. NeuroImage 2003; 20:987-94. 134. Hamer H, Wylie E, Stanford L, et al. Risk factors for unsuccessful testing during the intracarotid amobarbitol procedure in preadolescent children. Epilepsia 2000;41:554-63.
154
135. Floel A, Jansen A, Deppe M, et al. Atypical hemisphere dominance for attention: functional MRI topography. J Cereb Blood Flow Metab 2005;25:1197-208. 136. Knecht S, Jansen A, Frank A, et al. How atypical is atypical language? NeuroImage 2003;18:917-27. 137. Kaps M, Schaffer M, Beller K, et al. Phase 1: transcranial echo contrast studies in healthy volunteers. Stroke 1995;26:2648-52. 138. Watanabe E, Maki A, Kawaguchi F, et al. Non-invasive assessment of language dominance with near-infrared spectroscopic mapping. Neurosci Lett 1998; 256:49-52. 139. Watson N, Dodrill C, Farrel D, et al. Determination of language dominance with near-infrared spectroscopy: comparison with the intracarotid amobarbital procedure. Seizure 2004;13:399-402. 140. Gallagher A, Theriault M, Mclin E, et al. Near-infrared spectroscopy as an alternative to the Wada test for language mapping in children, adults and special populations. Epileptic Disord 2007;9:241-55. 141. Gallagher A, Lassonde M, Bastien D, et al. Noninvasive pre-surgical investigation of a 10 year-old epileptic boy using simultaneous EEG-NIRS. Seizure 2008;17(6):576-82. 142. Noguchi Y, Takeuchi T, Sakai K. Lateralized activation in the inferior frontal cortex during syntactic processing: event related optical topography study. Hum Brain Mapp 2002;17:89-99. 143. Sokol D, Markand O, Daley E, et al. Near infrared spectroscopy distinguishes seizure types. Seizure 2000;9:323-7. 144. Watanabe E, Maki A, Kawaguchi F, et al. Non-invasive cerebral blood volume measurement during seizures using multi-channel near infrared spectroscopic topography. J Biomed Opt 2000;5:287-90. 145. Watanabe E, Nagahori Y, Mayanagi Y. Focus diagnosis of epilepsy using near-infrared spectroscopy. Epilepsia 2002;43 Suppl 9:50-5. 146. Lesser R. Can fMRI substitute for the Wada test? Epilepsy Curr 2003;3:210-11. 147. Desmond J, Sun J, Wagner A, et al. fMRI measurement of language lateralization in Wada tested patients. Brain 1995;118:1411-9. 148. Binder J, Swanson S, Hammeke T. Determination of language dominance using fMRI: a comparison with the Wada test. Neurology 1996;46:978-84. 149. Spreer J, Arnold S, Quiske A, et al. Determination of hemisphere dominance for language: comparison of frontal and temporal fMRI activation with intracarotid amytal testing. Neuroradiology 2002;22:467-74. 150. Benson R, Fitzgerald D, LeSuer L, et al. Language dominance determined by whole brain fMRI in patients with brain lesions. Neurology 1999;52:798-809. 151. Gaillard WD, Balsamo L, Xu B, et al. Language dominance in partial epilepsy patients identified with fMRI reading task. Neurology 2002;59:256-65.
2613
2614
154
The Wada test-60th year anniversary update-in epilepsy surgery
152. Sabbah P, Chassoux F, Leveque C, et al. fMRI in assessment of language dominance in epileptic patients. NeuroImage 2003;18:460-7. 153. Woermann F, Jokeit H, Luerding R. Language lateralization by Wada test and fMRI in 100 patients with epilepsy. Neurology 2003;61:699-701. 154. Gaillard W, Balsamo L, Xu B, et al. fMRI language task panel improves determination of language dominance. Neurology 2004;63:1403-08. 155. Benke T, Koeylue B, Visani P, et al. Language lateralization in temporal lobe epilepsy: a comparison between fMRI and Wada Test. Epilepsia 2006;47:1308-19. 156. Anderson D, Harvey A, Saling M, et al. fMRI lateralization of expressive language in children with cerebral lesions. Epilepsia 2006;47:988-1008. 157. Jayakar P, Bernal B, Medina S, et al. False lateralization of language cortex on functional MRI after a cluster of seizures. Neurology 2002;59:490-2. 158. Westerveld M, Stoddard K, Spencer D, et al. Case report of false lateralization using fMRI language localization, Wada testing, and cortical stimulation. Archiv Clin Neurol 1999;14:162-63. 159. Sabsevitz D, Swanson S, Hammeke T. Use of preoperative functional neuroimaging to predict language deficits from epilepsy surgery. Neurology 2003;60: 1788-92. 160. Carpentier A, Pugh K, Westerveld M, et al. fMRI of language processing: dependence on input modality and temporal lobe epilepsy. Epilepsia 2001;42: 1241-54. 161. Fitzgerald D, Cosgove G, Ronner S, et al. Location of language in the cortex: comparison between fMRI and electrocortical stimulation. Am J Neuroradiol 1997;18:1529-39. 162. Lurito J, Lowe M, Sartorius C, et al. Comparison of fMRI and intraoperative cortical stimulation in localization of receptive areas. J Comput Assist Tomogr 2000;24:99-105. 163. Pouratian N, Bookheimer S, Rex D, et al. Utility of preoperative fMRI for identifying language cortices in patients with vascular malformations. Neurosurg Focus 2002;13:21-32. 164. Ruge M, Victor J, Hosain S, et al. Concordance between fMRI and intraoperative language mapping. Stereotact Funct Neurosurg 1999;72:95-102. 165. Ko K, Leijten S, Rutten G, et al. Discrepant findings for Wada test and fMRI with regards to language function: use of electrocortical mapping to confirm results. J Neurosurg 2005;102:169-73. 166. Roux FE, Boulanouar K, Mejidoubi M, et al. Language fMRI in preoperative assessment of language areas: correlation with direct cortical stimulation. Neurosurgery 2003;52:1335-47. 167. Seeck M, Pegna A, Ortigue S, et al. Speech arrest with cortical stimulation may not reliably predict language deficit after epilepsy surgery. Neurology 2006;66:592-4.
168. Just MA, Carpenter PA, Keller TA. Brain activation modulated by sentence comprehension. Science 1996;274:114-16. 169. Loring D, Meador K, Lee G, et al. Wada memory performance predicts seizure outcome following anterior temporal lobectomy. Neurology 2004;44:2322-4. 170. Goldmann R, Golby A. Atypical language representation in epilepsy: implications for injury-induced reorganization of brain function. Epilepsy Behav 2005;6:473-87. 171. Janszky J, Jokeit H, Kontopoulou K. Functional MRI predicts memory performance after right mesiotemporal epilepsy. Epilepsia 2005;46:244-50. 172. Rabin M, Narayan V, Kimberg D, et al. Functional MRI predicts post-surgical memory following temporal lobectomy. Brain 2004;127:2286-98. 173. Richardson MP, Strange BA, Thompson PJ. Preoperative verbal memory fMRI predicts post-operative memory decline after left temporal lobe resection. Brain 2004;127:2419-26. 174. Perrine K, Westerveld M, Sass K, et al. Wada memory disparities predict seizure laterality and postoperative seizure control. Epilepsia 1995;36:851-6. 175. Binder JR, Sabsevitz D, Swanson SJ, et al. Use of preoperative functional MRI to predict verbal memory decline after temporal lobe epilepsy surgery. Epilepsia 2008;49:1-18. 176. Verrotti A, Pizzella V, Trotta D, et al. Magnetoencephalography in pediatric neurology and in epileptic syndromes. Pediatric Neurol 2003;28:253-61. 177. Arjan-Hillebrand K, Singh I, Furlong P, et al. New approach to neuroimaging with magnetoencephalography. Hum Brain Mapp 2005;25:199-211. 178. Zouridakis G, Simos PG, Breier JI, et al. Functional hemispheric asymmetry assessment in a visual language task using MRG. Brain Topogr 1999;11:57-65. 179. Breier J, Simos P, Wheless B, et al. Language dominance in children as determined by magnetic source imaging and the intracarotid amobarbitol procedure: a comparison. J Child Neurol 2001;16:124-30. 180. Kober H, Mueller M, Ninsky C, et al. New approach to localize speech relevant brain areas and hemispheric dominance using spatially filtered magnetoencephalography. Hum Brain Mapp 2001;14:236-50. 181. Papanicolau A, Simos P, Castilo E. Magnetoencephalography: non-invasive alternative to the Wada procedure. J Neurosurg 2004;100:867-76. 182. Hirata M, Kato A, Tanigucji M, et al. Determination of language dominance with synthetic aperture magnetometry: a comparison with the Wada test. NeuroImage 2004;23:46-53. 183. Boyer JI, Moran J, Weiland BJ. Language laterality determined by MEG mapping with MR- FOCUSS. Epilepsy Behav 2005;6:235. 184. Lee D, Sawrie S, Simos P, et al. Reliability of language mapping with magnetic source imaging in epilepsy surgery candidates. Epilepsy Behav 2006;8:742-9.
The Wada test-60th year anniversary update-in epilepsy surgery
185. Maetu F, Oritz T, Fernandez A. Spanish language mapping using MEG: a validation study. Neuroimage 2002;17:1579-86. 186. Breier J, Papanicolaou A. Spatiotemporal patterns of brain activation during an action naming task using magnetoecephalography. J Clin Neurophysiol 2008;25:7-12. 187. Pataraia E. The role of magnetoencephalography in presurgical evaluation of epilepsy patients. Epilepsia 2007;48:s3, P1-6. 188. Ver Hoef L, Sawrie S, Killen J, et al. Left mesial temporal sclerosis and verbal memory: a magnetoencephalographic study. J Clin Neurophysiol 2008;25:1-6. 189. Poline J, Vandenberghe R, Holmes AP. Reproducibility of PET activation studies: lessons from a multi-center European experiment. NeuroImage 1996;4:34-54. 190. Muller R, Rothermel R, Muzik O, et al. Determination of language dominance by [15O]-water PET in children and adolescents: a comparison with the Wada test. J Epilepsy 1998;11:152-61. 191. Ohta Y, Nariai KT, Ishii K. Meningo-angiomayosis with focal epilepsy: value of PET in diagnosis and preoperative planning of surgery. Acta Neurochir 2003;145:587-91. 192. Hunter KE, Braxton TA, Bookheimer SY. (15) O Water positron emission tomography in language localization: a study comparing visual and computerized region of interest analysis with the Wada test. Ann Neurol 1999;45:662-5. 193. Tatlidil R, Luther S, West A, et al. Comparison of fluorine-18 deoxyglucose and O-15 water PET in temporal lobe epilepsy. Acta Neurol Belg 2000;100:214-20. 194. Bookheimer SY, Zeffiro TA, Blaxton T, et al. A direct comparison of PET activation and electrical cortical stimulation mapping for language localization. Neurology 1997;48:1956-65. 195. Griffith HR, Pizalski RW, Seidenberg M, et al. Memory relationships between MRI volumes and resting PET metabolism of medial temporal lobe structure. Epilepsy Behav 2004;5:669-76. 196. Hong SH, Roh SY, Kim SM, et al. Correlation of temporal lobe glucose metabolism with the Wada memory test. Epilepsia 2000;41:1554-9. 197. Akanuma N, Koutroumanidis M, Adachi N, et al. Presurgical assessment of memory-related brain structures: the Wada test and functional imaging. Seizure 2003;12:346-58. 198. Salanova V, Markand O, Worth R, et al. Presurgical evaluation and surgical out come of temporal lobe epilepsy. Pediatr Neurol 1999;20:179-84. 199. Krause B, Hautzel H, Schmidt D, et al. Learning related interactions among neuronal systems involved in memory processes. J Physiol 2006;99:318-32. 200. Silverman D, Alavi A. PET imaging in the assessment of normal and impaired cognitive function. Radiol Clin North Am 2005;43:67-77.
154
201. Papathanassiou D, Etard O, Mellet L, et al. A common language network for comprehension and production: a contribution to the definition of language epicentres with PET. NeuroImage 2000;11:347-57. 202. Beversdorf D, Metzger S, Nelson D, et al. Single word auditory stimulation and regional cerebral blood flow as studied by SPECT. Psychiatry Res 1995;61:181-9. 203. McMakin D, Jones-Gotman M, Dubeaur F, et al. Regional cerebral blood flow and language dominance SPECT during intracarotid amobarbital testing. Neurology 1998;50:943-50. 204. De Paola L, Maeder M, Germiniani F, et al. Bizarre behaviour during intracarotid sodium amytal testing (Wada Test): are they predictable? Arq Neuropsychiatry 2004;62:444-8. 205. Dawkins A, Evans A, Wallam J, et al. Complications of cerebral angiography: prospective analysis of 2924 consecutive procedures. Neuroradiology 2007;49:1217-23. 206. Willinsky R, Taylor S, TerBrugge K, et al. Neurologic complications of cerebral angiography: prospective analysis of 2899 procedures and review of the literature. Radiology 2003;227:522-8. 207. Johnston D, Chapman K, Goldstein L. Low rate of complications of cerebral angiography in routine clinical practice. Neurology 2001;57:2012-4. 208. Kaufmann T, Huston J, Mandrekar J, et al. Complications of diagnostic cerebralangiography evaluation of 19,826 consecutive procedures. Radiology 2007;243:812-9. 209. Lehericy S, Cohen L, Bazin B, et al. Functional MR evaluation of temporal and frontal language dominance compared with the Wada test. Neurology 2000;54:1625-33. 210. Fernandez G, Specht K, Tendolkar I, et al. Intrasubject reproducibility of presurgical lateralization and mapping using fMRI. Neurology 2003;60:969-75. 211. Miller J, McKinstry R, Philip J, et al. Diffusion-tensor MR imaging of normal brain maturation: a guide to structural development and myelination. Am J Roentol 2003;180:851-9. 212. Fujioka T, Ross B, Kakigi R, et al. One year of musical training affects development of auditory cortical-evoked fields in young children. Brain 2006;129:2593-608. 213. Ohnishi T, Matsuda H, Asada T, et al. Functional anatomy of musical perception in musicians. Cereb Cortex 2001;11:754-60. 214. Krainik A, Lehericy S, Duffau H, et al. Role of the supplementary motor area in motor deficit following medial frontal lobe surgery. Neurology 2001; 57:871-8. 215. Abou-Khalil B. An update on determination of language dominance in screening for epilepsy surgery: the Wada test and newer noninvasive alternatives. Epilepsia 2007;48:442-55. 216. Killgore W, Glosser G, Casasanto D, et al. Functional MRI and the Wada test provide complementary
2615
2616
154 217. 218.
219.
220.
221. 222.
223.
The Wada test-60th year anniversary update-in epilepsy surgery
information for predicting post-operative seizure control. Seizure 1999;8:450-5. Wada JA. Youthful season revisited. Brain Cogn 1997;33:7-10. Knecht S, Draeger M, Deppe M, et al. Handedness and language dominance in healthy humans. Brain 2000;123:2512-8. Lanzenberger R, Wiest G, Geissler A, et al. fMRI reveals functional cortex in a case of inconclusive Wada testing. Clin Neurol Neurosurg 2005;107:147-51. Loddenkemper T, Moddel G, Schuele S, et al. Seizures during intracarotid methohexital and amobarbital testing. Epilepsy Behav 2007;10:49-54. Meador K, Loring D, Lee G, et al. Hemispheric asymmetry for eye gaze mechanism. Brain 1989;112:103-111. Sato H, Takeuchi T, Sakai K. Temporal cortex activation during speech recognition: an optical topography study. Cognition 1999;77:B55-66. Seidenberg M, Geary E, Hermann B. Investigating temporal lobe contribution to confrontation naming using MRI quantitative volumetrics. J Int Neuropsychol Soc 2005;11:358-9.
224. Soros P, Cornelissen K, Lane M, et al. Naming actionsand objects. Cortical dynamics in healthy adults and in an anomic patient with a dissociation in action/object naming. NeuroImage 2003;4:1787-801. 225. Watanabe E, Yamashita Y, Maki A, et al. Non-invasive functional mapping with multi-channel near-infrared spectroscopic topography in humans. Neurosci Lett 1996;205:41-4. 226. Yuan W, Szaflarski J, Schmithorst V, et al. fMRI shows atypical language lateralization in pediatric epilepsy patients. Epilepsia 2006;47:1528-167. 227. Damasio A, Belllugi U, Damasio H, et al. Sign language aphasia during left-hemisphere Amytal injection. Nature 1986;322:363-5. 228. Devinsky G, Perrine K, Luciano D, et al. Anterior temporal language areas in patients with early onset of temporal lobe epilepsy. Ann Neurol 1993;134: 727-32. 229. Homan RW, Criswell E, Wada JA, et al. Hemispheric contributions to manual communication (signing and finger-spelling). Neurology 1982;32:1020-3.
167 Vagal Nerve Stimulation for Seizures A. P. Amar . J. B. Elder . M. L. J. Apuzzo
The implantable vagus nerve stimulation (VNS) device from Cyberonics, Inc. (Houston, TX) delivers intermittent electrical stimulation to the left cervical vagus nerve trunk, which secondarily transmits rostral impulses to exert widespread but poorly-defined effects on neuronal excitability throughout the central nervous system. VNS therapy has gained increasing popularity and credibility as a treatment option for patients with intractable epilepsy. It has also emerged as a novel adjunct in the management of patients with refractory depression and potentially other neurological disorders. Clinical experience with VNS began in 1988 with the first human implantation of the VNS device. Since then, more than 50,000 patients worldwide have received VNS therapy, and more than 100,000 patient-years of experience have accrued (Cyberonics, data on file). This chapter reviews the theoretical rationale, practical background, and clinical applications pertinent to VNS therapy. Other papers detail the use of VNS for the treatment of epilepsy [1–5] and depression or other conditions [6,7], as well as the operative procedure for inserting the VNS device [1,8,9].
Theoretical Basis of VNS As with many other anticonvulsant therapies, information about the neural mechanisms underlying VNS lags behind the appreciation of its clinical efficacy. The exact means by which VNS modulates seizure activity and its locus of action in the brain remain uncertain despite investigations of electroencephalography (EEG), #
Springer-Verlag Berlin/Heidelberg 2009
functional imaging, neurotransmitter analysis, and other approaches [1]. The suggestion that afferent stimulation may modulate seizure activity dates back at least 2,000 years to the teachings of Pelops, the master of Galen. He described a technique using ligatures applied to the limb in which partial seizures began as a means of aborting the progression of a focal seizure or preventing its generalization. Subsequent studies have confirmed that stimulation of cutaneous afferent fibers and other sensory pathways, including direct stimulation of the cervical vagus nerve, can affect EEG synchronization and sleep cycles. Because highly synchronized patterns are characteristic of electrographic seizures, these studies of EEG rhythmicity form the neuroanatomic and neurophysiologic foundations for the hypothesis that appropriately timed stimulation of the vagus nerve might prevent or abort paroxysmal epileptiform activity. Although the vagus nerve carries efferent projections that innervate the striated muscle of the larynx and provide parasympathetic control of the heart, lungs, and gastrointestinal tract, over 80% of its fibers are special visceral and general somatic afferents leading towards the brain [10]. While it was initially suspected that VNS works by recruiting afferent C-fibers and A-delta fibers, this contention has been recently challenged by observations that VNS retains its antiepileptic effects even after selective destruction of these small unmyelinated fibers by capsaicin treatment [11]. Visceral afferent fibers of the vagus nerve originate from receptors in the lungs, aorta, heart, and gastrointestinal tract, including the
2802
167
Vagal nerve stimulation for seizures
esophagus [10]. Somatic sensory afferents transmit sensation from the concha of the ear, but these fibers represent a small component of the nerve [10]. Vagal afferents project to diffuse areas of the central nervous system, many of which are potential sites of epileptogenesis. These include the cerebellum, diencephalon, amygdala, hippocampus, insular cortex, and multiple brainstem centers. Most of these projections relay through the nucleus tractus solitarius, while a smaller proportion forms direct, monosynaptic connections with their targets in the medial reticular formation of the medulla, the dorsal motor nucleus of the vagus, the area postrema, and the nucleus cunneatus [10,12]. Although it remains unclear which of these pathways underlie the mechanism of VNS action, the locus coeruleus (LC) and dorsal raphe nucleus (DRN) appear to be key intermediaries. The basal firing rates of neurons within the LC and DRN are significantly increased after long-term treatment with VNS [13]. LC firing rates increase earlier than those of the DRN, and since the LC has an excitatory influence on DRN, it is possible that the increased firing rate in the latter is secondary to the initial increase in LC firing rates [13]. Furthermore, bilateral chemical lesions of these centers abolish the seizure-suppressing effects of VNS therapy in animal models [14]. These results imply that norepinephrine and serotonin, which are diffusely released by the LC and DRN, respectively, may mediate the anticonvulsant actions of VNS. Indeed, these two neurotransmitters are known to modulate seizure threshold in some parts of the brain by inducing interneurons to release gamma-amino butyric acid (GABA), leading to widespread inhibition of neuronal excitability throughout the brain. However, the levels of GABA and serotonin metabolites in the cerebrospinal fluid of patients undergoing VNS appears to be inversely correlated with the efficacy of treatment, and the neurotransmitter systems that mediate the antiepileptic actions of VNS remain uncertain [1].
Since noradrenergic and seritonergic pathways are also implicated in the pathogenesis of depression, the effects of VNS on these neurotransmitters may underlie its proposed utility as an antidepressant treatment as well. At the stimulation parameters typically used for human application, VNS has no effect on background EEG rhythms [15–17]. Most patients have a decrease in the number and duration of interictal epileptiform discharges (IED) [18]. No correlation between IED and VNS efficacy has been reported, however. Vagal stimulation induces evoked responses from regions as disparate as the cerebral cortex, hippocampus, brainstem, thalamus, and cerebellum, and many authors have proposed that its antiepileptic actions relate to effects on the brainstem reticular activating system, which then projects to these forebrain structures [2]. However, positron emission tomography (PET) experiments measuring regional cerebral blood flow (rCBF) in response to VNS reveal changes confined to more circumscribed regions, such as the ipsilateral anterior thalamus and cingulate gyrus, contralateral thalamus and ipsilateral cerebellum, or bilateral activation of the hypothalami and insular cortices [1,2]. The reasons for this disparity in activated rCBF patterns from study to study are not immediately apparent, but may relate to differences in stimulation parameters, individual patient variation, and other factors. Inconsistencies between PET studies acquired during the acute versus long-term phases of VNS may reflect chronic adaptation to central processing, which attenuates responses to individual trains of VNS. Furthermore, PET studies have been confounded by multiple methodological limitations, such as seizures occurring during PET acquisition, the effects of prior cranial surgery, etc [1]. In any case, the central consequences of VNS on rCBF are not as diffuse as might be expected were its effects mediated through the brainstem reticular substance [2]. Right-versus left-sided vagal stimulation is equally effective in controlling seizures in animal
Vagal nerve stimulation for seizures
models, and bilateral stimulation produces no measurably greater effect than unilateral stimulation [1]. Using techniques such as EEG and immunolabeling against fos, a nuclear protein expressed under conditions of high neuronal activity, studies suggest that unilateral afferent vagal impulses generate bilaterally symmetric responses in the cerebral cortex and subcortical structures [19]. In contrast, vagal efferent innervation appears asymmetric [1]. In some species, the right vagus nerve innervates the sinoatrial node while the left one preferentially supplies the atrioventricular node. Canine studies have shown that stimulation of the right vagus nerve produces greater cardiac slowing than similar stimulation of the left vagus. For these reasons, the VNS system is generally inserted on the left side [9]. Experience with right-sided VNS in humans is anecdotal. In patients who developed device infections that precluded repeat left-sided VNS, for instance, the device has been reinserted on the right vagus nerve. In these patients, no cardiac side effects occurred, though some did experience reactive airway disease and respiratory compromise [20]. In some patients, a differential benefit occurred when the left-sided VNS was replaced by right-sided VNS, raising the possibility that those who fail left-sided VNS may benefit from a trial of contralateral stimulation. Some animal studies have shown that cardiac and respiratory function are adversely affected by VNS while others have not, depending on the species used, the stimulation parameters applied, and other variables [1]. Such side effects generally do not occur in humans because stimulation can be performed distant from the site at which the cardiac branches originate from the cervical vagus trunk [9]. In fact, VNS exerts central cardioinhibition through its afferent effects without vasodepression or negative chronotropic effects of efferent vagal stimulation, leading some to propose VNS as a potential cardioprotective strategy in high risk patients with ischemic heart disease
167
due to unrestrained sympathoexcitation [21]. Rarely, however, vagally-induced bradyarrhythmia, perfectly correlated with periods of stimulation, has been described even years after uneventful VNS therapy in some patients [22].
VNS Device Components > Figure 167‐1 depicts a schematic representation of VNS therapy. A pulse generator inserted in the subcutaneous tissues of the upper left chest delivers intermittent electrical stimulation to the cervical vagus nerve trunk via a bifurcated helical lead. In addition to the implantable lead and pulse generator, the Cyberonics system includes a number of peripheral components, such as a telemetry wand that interrogates and programs the pulse generator noninvasively. This programming wand is battery- powered and is interfaced with a Dell Axim handheld that runs a proprietary
. Figure 167‐1 Schematic representation of VNS therapy. A pulse generator inserted in the subcutaneous tissues of the upper left chest delivers intermittent electrical stimulation to the cervical vagus nerve trunk via a bifurcated helical lead. Reproduced with permission from Cyberonics, Inc
2803
2804
167
Vagal nerve stimulation for seizures
. Figure 167‐2 NCP Programming wand interfaced to Dell Axim1 Handheld. Reproduced with permission from Cyberonics, Inc
menu-based software package (> Figure 167‐2). The system also includes a portable magnet that patients may carry with them in order to alter the character of stimulation that the generator delivers. The NCP pulse generator has approximately the same size and shape as a cardiac pacemaker. It contains an epoxy resin header with a receptacle that accepts the connector pin extending from the bifurcated lead (> Figure 167‐3). The generator is powered by a single lithium battery encased in a hermetically sealed titanium module. The projected battery life of the generator varies with the stimulus parameters but can be as long as 6–10 years under normal conditions. Once it has expired, the generator can be replaced with the patient under local anesthesia during a simple outpatient procedure. The generator contains an internal antenna that receives radiofrequency signals emitted from the telemetry wand and transfers them to a microprocessor that regulates the electrical output of the pulse generator. The generator delivers a charge-balanced waveform characterized by five programmable parameters: output current, signal frequency, pulse width, signal on-time,
. Figure 167‐3 NCP Pulse generator and lead. Reproduced with permission from Cyberonics, Inc
and signal off-time. These variables are titrated empirically in the outpatient setting, according to individual patient tolerance and seizure frequency. Altering the parameters of stimulation will have various consequences on VNS efficacy, side effects, and battery life. In clinical application, the most common parameters are 0.25–2.0 mA current (titrated to effect and tolerance), 30 Hz frequency, 500 ms pulse width, and 30-s on/5-min off duty cycle. Other paradigms, such as 7 s on and 18 s off (‘‘rapid cycling’’) have also been utilized. Regardless, the duration of VNS action is obviously longer than the periods of intermittent stimulation. The bipolar lead is insulated by a silicone elastomer and can be safely implanted in patients with latex allergies. One end of the lead contains a connector pin that inserts directly into the generator, while the opposite end contains an electrode array consisting of three discrete helical coils that wrap around the vagus nerve. The middle and distal coils represent the positive and negative electrodes, respectively, while
Vagal nerve stimulation for seizures
the most proximal one serves as an integral anchoring tether that prevents excessive force from being transmitted to the electrodes when the patient turns his neck. The leads come in two sizes, measured by the internal diameter of each helix. Although the majority of patients can be fitted with the 2 mm coil, it’s desirable to have the 3 mm one available in the operating room as well. Each electrode helix contains three loops (> Figure 167‐4). Embedded inside the middle turn is a platinum ribbon coil that is welded to the lead wire. This shape permits the platinum ribbon to maintain optimum mechanical contact with the nerve. Suture tails extending from either end of the helix permit manipulation of the coils without injuring these platinum contacts. The electrode is intended to fit snugly around the nerve while avoiding compression, thus allowing the electrode to shift with the nerve and minimizing abrasion from relative movement of the nerve against the electrode. Damage to the nerve is greatly reduced by the self-sizing, open helical design of the VNS electrode array, which permits blood flow and interstitial fluid exchange with the nerve. Thus, compared with cuff electrodes, mechanical trauma and ischemia to the nerve are minimized. Histological examination of the vagus nerve following VNS has revealed no . Figure 167‐4 NCP Helical lead array. Reproduced with permission from Cyberonics, Inc
167
axonal loss, demyelination, lymphocytic infiltration, or other evidence of permanent damage resulting from electrical stimulation. Other observations have confirmed the safety of chronic nerve stimulation when the duty cycle (the fraction of time the nerve undergoes stimulation) is less than 50% [23,24]. The hand-held magnet performs several functions. When briefly passed across the chest pocket where the generator resides, it manually triggers a train of stimulation superimposed upon the baseline output. Such on-demand stimulation can be initiated by the patient or a companion at the onset of an aura, in an effort to diminish or even abort an impending seizure. The parameters of this magnet-induced stimulation may differ from those of the prescheduled activation. Alternatively, if the device appears to be malfunctioning or if the patient desires to terminate all stimulation for any other reason, the system can be indefinitely inactivated by applying the magnet over the generator site continuously. Finally, patients are instructed to test the function of their device periodically by performing magnet-induced activation and verifying that stimulation occurs; most patients can perceive the stimulation as a slight tingling sensation in the throat. Patients with a programmable ventriculoperitoneal shunt valve must be cautious, however, since the magnet can affect its calibration. Symptomatic underdrainage of cerebrospinal fluid has been reported after unintended adjustment of the valve by the VNS magnet [25].
Surgical Insertion Insertion of the VNS device takes less than two hours and is typically performed under general anesthesia; however, regional cervical blocks have also been used in awake patients. While it can be performed as an outpatient procedure, it may be desirable to observe patients overnight for vocal cord dysfunction, dysphagia, respiratory
2805
2806
167
Vagal nerve stimulation for seizures
compromise, or seizures induced by anesthesia, even though these complications are rare. Prophylactic antibiotics are administered preoperatively and maintained for 24 h postoperatively. The implantation procedure is conceptually straightforward. The first step involves creation of a chest pocket that accommodates the pulse generator. Next, through a separate incision, the carotid sheath is opened, the internal jugular vein mobilized, and the vagus nerve trunk isolated. The lead is tunneled within a subcutaneous tract between the two incisions. The helical electrodes are applied to the vagus nerve, and the lead connector pin is attached to the generator. After additional electrodiagnostic testing, the lead and generator are secured to adjacent tissue, and the wounds are closed in standard, multilayer fashion. Others have described access to both the cervical and chest regions through a single supraclavicular incision [26].
Operative Procedure: Relevant Anatomy Planning for placement of the VNS system requires thorough anatomic understanding of the relevant neural, vascular, and muscular components of the anterior cervical triangle in order to minimize hazard to the ansa cervicalis, recurrent laryngeal nerve, tributaries to the internal jugular vein, and other structures. A full understanding of the anatomy of the superior and inferior cardiac branches will also reduce the rare occurrence of intraoperative bradycardia during the lead test. Several branches of the vagus nerve arise cephalad to the midcervical trunk, where the VNS electrodes are applied [9]. These include projections to the pharynx and carotid sinus, as well as superior and inferior cervical cardiac branches leading to the cardiac plexus. The diameter, appearance, and location of the cardiac branches may approximate those of the nerve
trunk itself, and care must be taken to avoid mistaking the two. If the cardiac branches are stimulated directly, small currents as low as 0.8 mA may produce significant bradycardia [27]. The superior laryngeal nerve arises rostral to the carotid bifurcation before descending towards the larynx, and high currents applied to the midcervical vagus nerve trunk may recruit these fibers, leading to tightness or pain in the pharynx or larynx. The recurrent laryngeal nerve travels with the main trunk and branches caudally at the level of the aortic arch before ascending in the tracheo-esophageal groove. As a result, hoarseness is a common occurrence during periods of stimulation or after VNS implantation. In addition to branches of the vagus nerve trunk, several other nerves in the vicinity of the carotid sheath risk hazard from the implantation procedure itself or from subsequent stimulation. The hypoglossal nerve arises cephalad to the midcervical region, making unilateral tongue weakness an infrequent complication of VNS implantation. The phrenic nerve lies deep to a fascial plane beneath the carotid sheath, and hemiparalysis of the diaphragm has been reported with stimulation at high output currents, though not as an operative complication. The sympathetic trunk lies deep and medial to the common carotid artery. It gives off fibers that ascend with the internal carotid artery (ICA) towards the intracranial contents. Weakness to the muscles of the lower face may result from injury to branches of the facial nerve, which ramify through caudal aspect of the parotid gland. In general, hypoglossal and facial nerve injury are more common sequelae of carotid endarterectomy incisions, which tend to be higher than those used for placement of the VNS device.
Operative Procedure in Detail The operating room should be organized to ease the surgeon’s access and to minimize traffic
Vagal nerve stimulation for seizures
within the area. Following endotracheal intubation, we rotate the table 90 clockwise from the anesthesia setup, which thus lies alongside the patient’s right foot. This permits the surgeon to stand at the patient’s left while his assistant stands at the patient’s right. The scrub technician is positioned at the patient’s head, affording ready access to each surgeon on either side. The electrophysiology staff remains behind the assistant’s back but within reach of the scrub technician, in order to conduct pre-implant diagnostic testing once the generator has been placed within the sterile field. The patient is positioned supine with a shoulder roll beneath the scapulae in order to provide mild neck extension. This facilitates passage of the tunneling tool that connects the two incisions. The head is rotated 30–45 towards the right, bringing the left sternocleidomastoid muscle into prominence. Many options exist for placement of the skin incisions. Often, a 5-cm transverse chest incision is made approximately 8 cm below the clavicle, centered above the nipple. The underlying fat is dissected to the level of the pectoralis fascia, and a subcutaneous pocket is fashioned superiorly. Although others have suggested an incision in the deltopectoral groove with inferior dissection to create the pocket, we believe that the scar tissue formed beneath the pectoral incision helps prevent caudal migration of the generator. Recently, we have employed a lateral incision along the anterior fold of the axilla, which affords better cosmetic results, especially among women. Implantation of the device deep to the pectoralis muscle has also been described [28]. Next, a 5-cm longitudinal incision is made along the anterior border of the sternocleidomastoid muscle, centered over its midpoint. Generally, this incision is a little lower than that for an endarterectomy. Alternatively, a transverse skin incision at C5/6, similar to the approach for an anterior cervical discectomy, can be made. For the inexperienced surgeon, the longitudinal incision
167
permits a wider exposure, which facilitates electrode placement through this aperture. The platysma muscle is divided vertically, and the investing layer of deep cervical fascia is opened along the anterior border of the sternocleidomastoid, allowing it to be mobilized laterally. Following palpation of the carotid pulse, the neurovascular bundle is identified and sharply incised to reveal its contents. Self-retaining retractors with blunt blades expedite this stage of the procedure. Care is taken to limit the exposure between the omohyoid muscle and the common facial vein complex, thus minimizing potential hazard to adjacent neurovascular structures. Within the carotid sheath at the level of the thyroid cartilage, the vagus nerve is generally encountered deep and medial to the internal jugular vein, encased in firm areolar tissue lateral to the common carotid artery. Great variability exists in the relative position of these structures, however, and the strategy by which the nerve is isolated from the remainder of the neurovascular bundle must account for such individual diversity. We attempt to minimize direct manipulation of the nerve itself. Instead, we prefer to mobilize the vessels away from the nerve. Dissection generally commences with isolation and retraction of the internal jugular vein using vessel loops. Next, the nerve trunk is identified and dissected with the aid of the operating microscope or surgical loupes. At least 3–4 cm of the nerve must be completely freed from its surrounding tissues. At this stage, we’ve found that the insertion of a blue background plastic sheet between the nerve and the underlying vessels greatly facilitates the subsequent steps of the procedure. The technique of mobilizing the vessels away from the nerve usually preserves the vasa nervosum. This nuance may reduce the incidence of postoperative complications such as hoarseness. A tunneling tool is then used to create a subcutaneous tract between the two incisions. The tool is directed from the cervical to pectoral
2807
2808
167
Vagal nerve stimulation for seizures
sites, in order to minimize potential injury to the vascular structures of the neck. Depending on the relative size of the exposed nerve, either a small or large helical electrode is then selected for insertion. The lead connector pin is passed through the tunnel and emerges from the chest incision, while the helical electrodes remain exposed in the cervical region. Before applying the electrodes, the lead wire should be directed parallel and lateral to the nerve, with the coils occupying the gap between them. Each coil is applied by grasping the suture tail at either end and stretching the coil until its convolutions are eliminated. The central turn of this unfurled coil is applied either obliquely or perpendicularly across or beneath the vagus trunk and wrapped around the surface of the nerve. The coil is then redirected parallel to the nerve as the remainder of its loops is applied proximal and distal to this midpoint (> Figure 167‐5). The memory within the elongated coil will cause it to reassume its helical configuration and conform to the nerve snugly. Either the positive or negative terminal may be applied first, but the anchoring tether is generally applied last. While all these maneuvers are taking place, additional electrodiagnostic testing of the generator is simultaneously carried out by the scrub technician. The telemetry wand interrogates the device from within a sterile sheath to measure its internal impedance. Once the generator passes this pre-implant test, it is ready for insertion. The lead connector pin is connected to the pulse generator and secured to its receptacle. Additional electrodiagnostic examination is then performed in order to assess the coupling of all connections and to verify the integrity of the overall system. Then, a 1-min lead test is performed at a frequency of 20 Hz and a pulse width of 500 ms. The current should start at 0.25 mA and then ramp up in small increments to 1 mA. During this test stimulation, the response of the patient’s vital signs and electrocardiogram are monitored.
. Figure 167-5 Strategy for applying the helical electrodes to the vagus nerve trunk. Reproduced with permission from Cyberonics, Inc
Rarely, profound bradycardia will result, necessitating the use of atropine. The incidence of this event is thought to be about 1 in 1,000. If it occurs, attention should be directed to the lead to assure that the electrodes encircle the vagus nerve trunk itself rather than one of its cardiac branches. Following the test stimulation, the generator is restored to its inactive status until 1–2 weeks postoperatively. This waiting period allows for resolution of postoperative edema and proper fixation of the electrode to the nerve. The redundant portion of the lead between the generator and electrode is secured to several areas of the cervical fascia with Silastic tie-downs. The objective is to form superficial and deep restraint configurations that help prevent excessive traction from being transmitted to the electrodes during repetitive neck motion. First, a U-shaped strain relief bend is made inferior to the anchoring tether, and the distal lead is secured
Vagal nerve stimulation for seizures
to the fascia of the carotid sheath. Next, a strain relief loop is established by securing the lead to the superficial cervical fascia between the sternocleidomastoid and platysma muscles. Care is taken not to sew the lead directly to the muscle. Finally, the generator is retracted into the subcutaneous pocket and secured to the pectoralis fascia with O-Prolene or similar nonabsorbable suture, using the suture hole contained within the epoxy resin header. Any excess lead is positioned in a separate pocket at the side of the generator. To prevent abrasion of the lead, however, it should not be placed behind the pulse generator. Wound closure then proceeds in standard multilayer fashion, using a subcuticular stitch for the skin. The cosmetic results are generally very good.
Lead Removal or Revision In some circumstances, it may become necessary to remove and/or replace the electrodes that encircle the vagus nerve trunk. Although fibrosis and adhesions may develop in the vicinity of the vagus nerve, Espinosa [29] has demonstrated that the spiral electrodes may be safely removed from the nerve, even years after they were implanted. The extent of scarring does not appear related to the duration of implant.
Surgical Complication Avoidance and Management In a meta-analysis of the five clinical trials leading to approval of the VNS device by the United States Food and Drug Administration (FDA), the most commonly observed surgical complication was infection [30]. The site of the infection was either at the generator in the chest or near the leads in the neck. The overall infection rate was 2.86%, but more than half of these patients were successfully treated with antibiotic therapy alone, while only
167
about 1.1% required explantation of the device. The causative organisms have not been reported. In many cases, the device has been replaced successfully after removal for infection. Transient vocal cord paralysis is the second most common surgical complication of VNS implantation. The incidence of this event in the collective study experience was only 0.7% [30]. However, since video stroboscopy and formal swallowing assessments are rarely performed after surgery, it is possible that more cases went undetected, and the true prevalence of vocal cord paresis is not known. Fortunately, most reported cases resolve clinically. Vocal cord dysfunction should be minimized by careful manipulation of the vagus nerve, with preservation of its rete vascular supply and avoidance of excessive traction on the nerve. Temporary lower facial hypesthesia or paralysis occurred in another 0.7% of patients in the meta-analysis [30]. As stated above, excessively high surgical incisions could have been a cause. To our knowledge, only one case of Horner’s syndrome has occurred. This complication is more commonly reported after carotid endarterectomy and may be due to injury or manipulation of the sympathetic plexus immediately below the carotid sheath, or from traction on the sympathetic fibers on the internal carotid. Lead breakage occurred commonly with earlier versions of the VNS system but has only rarely been described since device modification. Data from the manufacturer indicates that there have been a total of 6 lead breaks in the first 5,000 implants since FDA approval (lead breakage rate = 0.12%). Suturing directly to the lead body was a possible cause in one extremely early case, and generator movement that caused excessive forces on the lead electrode may have been the cause in two others (in both cases the generator was placed in breast tissue in women and the suture loop in the generator may not have been used). In another case, no strain relief loop was used.
2809
2810
167
Vagal nerve stimulation for seizures
In the first 10,000 implantation procedures, only 9 cases of intraoperative bradycardia or asystole have been reported, accounting for an incidence less than 0.1%. All events occurred during the lead test. Asconape [27] has analyzed the factors that potentially contribute to this event and the means of their prevention. As mentioned, the superior or inferior cervical cardiac branches might be mistaken for the vagus trunk itself, and correct positioning of the electrodes on the intended nerve must be verified. Proper placement of the skin incision, centered over the midcervical portion of the nerve, will also help avert this complication. Current spread to the cardiac nerves can be minimized by measures that insulate them from the midcervical vagus trunk during the lead test, such as placement of a plastic dam beneath the nerve trunk and removal of pooled blood or saline from the vicinity. Finally, the current should be ramped up in small increments during the lead test, starting with 0.25 mA. Variants of the surgical procedure described above have been described for certain high risk populations. For instance, patients with cognitive delay are prone to wound tampering, leading to breakdown of the incision and secondary infection. In such patients, placement of the pulse generator between the scapulae may reduce the frequency of this event. In one study of cognitively delayed children, no infections occurred in the 9 who underwent interscapular pulse generator placement, in contrast to 2 of 14 (14%) who required device externalization after infection of their subclavicular wound [31].
Clinical Utility of VNS Since 1988, more than 1,000 patients have participated in seven corporate-sponsored clinical trials throughout 26 countries, and greater than 3,000 patient-years of data have accrued. Five of these studies (E01-E05) have been conducted in
the United States. The E01 and E02 trials were non-randomized pilot studies in small groups of patients with partial onset seizures designed to investigate the safety and feasibility of VNS therapy. An important observation was the fact that patients can perceive the onset of stimulation as a tingling sensation in the throat. In order to compensate for this phenomenon, subsequent controlled, randomized, double-blinded studies (E03 and E05) had to incorporate an ‘‘active control’’ group rather than a true placebo. In the latter studies, this was accomplished by comparing high (presumably therapeutic) versus low (presumably less therapeutic) parameters of stimulation. The E04 trial was an open label compassionate use study that examined the potential efficacy of VNS in patient populations that weren’t studied in the other four trials, namely those with generalized onset epilepsy, patients <12 years old, patients with fewer than six seizures per month, etc. This study was important for confirming the potential efficacy of VNS in these other patient groups, but the scientific validity of its results cannot attain to those of the two randomized trials. E05 was the largest controlled clinical trial and included comprehensive safety monitoring; this study was pivotal in obtaining FDA approval 6 months after its completion [32]. Collectively, these studies confirm the longterm safety, efficacy, feasibility, and tolerability of VNS, as well as the durability of the device [1,3,33]. In 1997, the FDA approved the VNS device ‘‘as an adjunctive therapy in reducing the frequency of seizures in adults and adolescents over 12 years of age with partial onset seizures which are refractory to antiepileptic medications.’’ Post-marketing experience validates the earlier clinical trials, and in 1999, the Therapeutics and Technology subcommittee of the American Association of Neurology declared VNS ‘‘safe and effective,’’ based on a preponderance of class I evidence [34]. Although VNS requires a large initial investment due to the price of the device
Vagal nerve stimulation for seizures
itself as well as its surgical insertion, cost-benefit analysis suggests that the expense of VNS is recovered within 2 years of follow-up [35]. Positive effects have been documented for both the utilization of health care services and the time spent on epilepsy-related tasks. In one study, statistically significant reductions in numbers of emergency department visits, hospitalizations, and hospital lengths of stay occur within the first fiscal quarter after VNS implantation, and the average numbers of days on which patients could not work because of health-related concerns also decreased after initiation of VNS therapy [36]. No accepted standard exists for reporting the outcome of VNS in clinical trials. A novel classification scheme has recently been proposed, which incorporates measures of both seizure frequency and ictal or postictal severity, as well as considerations unique to VNS therapy such as the efficacy of magnet usage [37]. However, most papers merely report the percent reduction in seizure frequency, expressed as the average for the entire cohort of patients. The response to VNS is not normally distributed. Usually, the histogram depicting response rates is skewed to the left, reflecting the disproportionate influence of the few patients who derive no benefit from therapy. Thus, the most valid summary statistic of central tendency for this non-parametric data is the median reduction in baseline seizure frequency rather than the mean. Patients who enter clinical trials of new antiepileptic therapies such as VNS generally have the most intractable form of the disease. Because they are often pharmacoresistant, such patients are not expected to become completely seizure free by the addition of a new investigational agent. Furthermore, many are predicted to fail completely. Therefore, the primary outcome measure of most antiepileptic medication trials has been the 50% responder rate (the proportion of patients who achieve a 50% or greater reduction in seizure frequency). Although complete eradication of seizure activity always remains the goal of therapy,
167
even 50% reductions can dramatically improve the quality of patients’ lives. In addition to seizure control, quality of life also depends on the side effects and toxicity of the treatment being rendered. Improvements in cognitive function and mood not related to seizure frequency per se are also reflected in these latter indices. A meta-analysis was performed of the 454 patients enrolled in one of five controlled, multicenter clinical trials (two double blind and three open label studies) conducted in the United States [33]. For the study population as a whole, the median reduction in seizure frequency was 35% at 1 year, 43% at 2 years, and 44% at 3 years. These results were obtained using a ‘‘last visit carried forward’’ analysis, which minimizes selection bias by extrapolating data from non-responders who exit the trial and thus tends to underestimate the efficacy among responders. For patients persisting in the trial (declining N analysis), sustained efficacy was even greater. An important observation is that the response to VNS is maintained during prolonged stimulation, and unlike the case with chronic medication therapy, seizure control actually improves with time. The response of individual patients to VNS varies widely. While rare subjects enjoy complete seizure cessation, others derive no benefit. The remainder experience intermediate results. In the collective study experience, the proportion of patients who sustained a 50% reduction in baseline seizure frequency was approximately 23% at 3 months [33]. Although this figure is similar to the initial results of many new drug trials, the 50% responder rate also showed substantial increases with time, reaching 37, 43, and 43% after 1, 2, and 3 years, respectively [33]. These improvements occurred in a highly refractory population of patients who typically had an average of 1.7 seizures per day despite administration of more than 2 antiepileptic medications. Clinical results since FDA approval have generally surpassed those in the company-sponsored
2811
2812
167
Vagal nerve stimulation for seizures
trials. In some cases, this may reflect more aggressive titration of stimulation parameters. In other instances, the more favorable results reflect the trend of improved seizure control with time. For instance, in one study, the median reduction in seizure frequency at 3 months was 45% and that at 12 months 58% [38]. In another, mean seizure reduction at 1–6 years was, respectively, 14, 25, 29, 29, 43, and 50% [39]. In yet another, the median reduction in seizure frequency improved from 28% at 12 months to 72% at follow-up 5–7 years after implantation. Some patients in the latter study whose seizure frequency was not reduced during the initial 12 months of VNS therapy experienced significant reduction during the follow-up period [40]. The 50% responder rates of nearly 60% and seizure-freedom rates of up to 10% have been reported in some post-market trials [41]. A potential confounding effect in the interpretation of this data is the fact that many published trials of VNS for epilepsy have permitted simultaneous alterations in antiepileptic drug regimens, making it difficult to know which treatment variable was most relevant in any given patient. In spite of the well known functions of the vagus nerve as the principal efferent component of the parasympathetic nervous system, VNS has not been shown to adversely effect any aspect of physiological function in a consistent fashion, including cardiac rhythm (as assessed by EKG and ambulatory Holter monitoring), pulmonary function, gastrointestinal motility and secretion [32,33,42]. Moreover, unlike many antiepileptic medications, VNS therapy does not impair cognition, balance, or emotion during extensive testing. Plasma concentrations of antiepileptic medications remain unchanged. Some adverse effects do occur with VNS, however. At three months of therapy during the acute phase studies, hoarseness, cough, paresthesia, and other symptoms were common, occurring in up to half of patients. These effects were rated as
mild or moderate 99% of the time [42]. They tend to occur concomitant with stimulus delivery and not throughout the day, unlike the side effects of antiepileptic medications. Furthermore, the side effects of VNS are generally transient, and their long-term incidence is much lower. The most common complaints after 1 year of treatment were hoarseness (28%) and paresthesias (12%) [17]. At 2 years, they were hoarseness (19.8%) and headache (4.5%), and after 3 years shortness of breath (3.2%) was the principal side effect [33]. The true incidence of recurrent laryngeal nerve injury is unknown, as most reports of transient vocal changes are based on perceptual observations by the patients only. Using a strict protocol employing laryngeal electromyography (EMG), video laryngoscopy, and quantitative assessments of voice, one study demonstrated that perioperative vocal cord paresis occurs in almost half of patients within 2 weeks of surgery [43]. Laryngeal EMG performed before implantation of the VNS device strongly predicted which patients were at risk for more prolonged vocal cord abnormalities [43]. Such information should be considered before patients undergo anterior cervical discectomy, right-sided VNS, or other procedures that could result in bilateral vocal cord paresis. More serious adverse events are rare. Although some deaths have occurred among the 454 study patients receiving VNS, none were definitely attributed to VNS therapy itself [33]. In fact, some studies suggest that the incidence of sudden unexplained death in epilepsy patients (SUDEP) is actually lower after treatment with VNS [44]. Patient satisfaction with VNS therapy is generally high. One way to quantify this parameter is to measure the percentage of patients who continue their therapy after completing the acute phase of a clinical trial. Continuation rates in the collective study experience were 97, 85, and 72% after 1, 2, and 3 years of therapy, respectively [33]. A related measure of patient satisfaction is
Vagal nerve stimulation for seizures
the percentage of patients who opt to undergo replacement of the generator after the battery has expired. With a previous model of the VNS device, battery expiration typically occurred 4–5 years after initiating therapy, and about threefourths of patients elected to change it at that time [33]. Long-term continuation rates reflect the unique profile of safety, efficacy, and tolerability that VNS provides. In the E05 study, for instance, global assessments of well being were recorded on a 10 cm line at each follow up visit. The center of the line represented no change relative to baseline, while deviations to the left (up to 50 mm) reflected worsening and those to the right (up to 50 mm) signified improvement. The mean improvement in patient ratings ranged from 10 to 15 mm at various time intervals after initiating VNS therapy [32]. An additional measure of patient satisfaction is assessment of overall quality of life (QOL). In randomized controlled trials, improvements in QOL were independently documented by the patient, the blinded physician, and the patient’s companion using a visual analog scale (VAS) [33,42]. Some patients who don’t experience any reduction in seizure frequency still report improvements of overall well-being. Even among populations with low intelligence quotient who live in long-term care facilities, the improved quality of life during VNS therapy can enhance attention, language, balance, and activities of daily living [45]. The basis of the enhanced QOL parameters, especially among patients without any improvements in seizure control, remains uncertain and reinforces the enigma of how VNS actually works. Most studies documenting improved QOL have relied on relatively crude measures such as VAS rather than comprehensive neuropsychological testing. In summary, VNS appears to offer several advantages over pharmacotherapy and other surgical modalities. VNS avoids cerebral toxicity and the attendant impairments of cognition,
167
emotion, and coordination that often complicate antiepileptic medication. The pre-programmed, computer-controlled characteristic of the system permits complete and involuntary treatment compliance. VNS is potentially reversible, unlike cerebral surgery. Unlike the case with many medications, the effectiveness of VNS is maintained during prolonged therapy and, in fact, overall seizure frequency diminishes with time. Furthermore, there are no adverse drug interactions. The improved quality of life and cognitive function perceived by patients during VNS trials is a testimony to this unique combination of efficacy and favorable side effect profile. In addition, the ability to initiate stimulation during an aura restores an element of sovereignty to patients’ lives, which are severely disrupted by the unpredictability of epilepsy. Thus, VNS is both a preventive and abortive therapy. This attests to a mechanism of action that is of longer duration than the periods of intermittent stimulation.
Patient Selection Epilepsy affects up to 1% of the general population and is the second most common neurological disorder overall. Despite recent advances in our understanding of the molecular and cellular basis of epilepsy and the development of several new medications targeted against these mechanisms, satisfactory seizure control remains elusive in 30–40% of patients. In the United States alone, there are at least 300,000 people with medically refractory seizures of partial onset. Although there is disagreement as to which of these patients should undergo cerebral surgery, it is estimated that only 30,000–100,000 patients are appropriate candidates for temporal lobectomy, focal cortical resection, callosotomy, hemispherectomy, subpial transection, and other extant procedures [1]. The selection criteria for insertion of the VNS system remain in evolution and reflect current
2813
2814
167
Vagal nerve stimulation for seizures
governmental standards as well as institutional biases and general guidelines from prior clinical trials. As noted, the VNS device is indicated ‘‘as an adjunctive therapy in reducing the frequency of seizures in adults and adolescents over 12 years of age with partial onset seizures which are refractory to antiepileptic medications.’’ Off-label use of VNS to treat children less than 12 or those with primarily generalized epilepsy has been rewarding, though these patient groups were not studied in randomized, controlled trials. Patients with both idiopathic epilepsy and seizures of structural etiology are considered appropriate candidates. The definition of medical intractability varies from center to center. Standards from previous studies commonly required a frequency of at least six seizures per month and a seizure free interval of no longer than 2–3 weeks despite therapy with multiple medications. However, seizure frequency, seizure type, severity of attacks, drug toxicity, and overall impact on quality of life must all be considered before a patient is deemed refractory to pharmacotherapy. As noted above, the response to VNS is highly variable, and previous clinical trials have failed to characterize the demographic factors that predict a favorable outcome. Furthermore, VNS is rarely curative (<10%). Although reductions in seizure frequency can dramatically improve patients’ quality of life, residual seizures may still preclude them from driving a car, maintaining employment, or other basic functions. Therefore, we do not consider the VNS device an alternative to conventional methods of epilepsy surgery that may offer a higher likelihood of seizure cessation, and we generally reserve VNS for patients in whom such operations are not indicated. These include those patients whose seizure focus is bilateral, not associated with a structural abnormality, or cannot be completely resected due to overlap with functional cortex. Extensive workup, including invasive monitoring when indicated, should precede the decision to opt for VNS before resective surgery is excluded.
For obvious reasons, the VNS system cannot be inserted in patients who have undergone a prior left cervical vagotomy. Furthermore, the safety of VNS has not been tested in several conditions in which impairment of vagus nerve function might produce deleterious effects. Thus, relative contraindications include progressive neurologic or systemic diseases, pregnancy, cardiac arrhythmia, asthma, chronic obstructive pulmonary disease, active peptic ulcer disease, and insulin dependent diabetes mellitus. Patients with prior anterior cervical discectomy or other neck surgery should undergo video laryngoscopy or EMG to assess the possibility of preexisting vocal cord paresis.
Off-Label Use of VNS for Epilepsy Subgroup analysis of the children and adolescents treated in one of the 5 multicenter prospective VNS trials conducted prior to FDA approval suggests that they derived substantial benefit from VNS, achieving median reductions in seizure frequency and 50% responder rates at least as favorable as those in adults. A total of 60 pediatric patients were treated in these studies [46]. Children 12–18 years old were included in the double-blinded, controlled trials, while patients as young as 3½ years old were studied in the open-label compassionate use protocol. After 3 months of stimulation, the median reduction in seizure frequency among these 60 patients was 23%. Using a last visit carried forward analysis, this figure improved to 31% at 6 months, 37% at 12 months, and 44% at 18 months. At 12 months, the 50% responder rate was 29%. These results are similar to those achieved by adults in the same trials. Since FDA approval, small uncontrolled studies exclusively studying pediatric patients have been confirmed the safety and efficacy of VNS in children. The results of these latter trials appear even more salutary than those in the older
Vagal nerve stimulation for seizures
populations [4]. Infants less than 1 year old have been successfully treated [47]. As in adults, the response to VNS improves with time. In one study, the mean reduction in total seizures was 39% at 3 months, 38% at 6 months, 49% at 12 months, 61% at 24 months, and 71% at 36 months [48]. In another, the reduction was 56% at 3 months, 50% at 6, 63% at 12, 83% at 24, and 74% at 36 months [49]. Nearly 40% of children achieve at least 90% reductions in seizure frequency [50,51], and up to 10% are seizure free [49]. As in the adult population, improvements in quality of life parameters such as memory, mood, behavior, alertness, achievement, and verbal skills have been documented in children [52,53]. Analysis of seizures by type failed to identify any classification that is consistently more responsive to VNS than others in children. Stratification into symptomatic versus idiopathic epilepsy was likewise unrevealing, since children with both types appeared to benefit from VNS in some cases [46,53]. Atonic seizures generally respond well [47], whereas myoclonic seizures, which frequently accompany mitochondrion electron transport chain disorders, tend to respond poorly [54]. Adverse events in children were also similar to those in adults in the company-sponsored trials, and none of them necessitated termination of therapy [46,55]. Serious complications included aspiration pneumonia and necrosis of the skin overlying the generator site, each occurring in one child. Post-market approval studies have revealed rare instances of both central and obstructive patterns of sleep apnea [56–59]. Although approved only for partial-onset seizures, VNS has been successfully applied for the off-label treatment of patients suffering from generalized-onset seizures, such as those with Lennox Gastaut syndrome [4]. In some studies, VNS is an even more effective therapy for idiopathic generalized epilepsy than for partial epilepsy [60]. Even the most refractory patient
167
populations, such as those with persistent seizures after failed cranial surgery, derive significant benefit from VNS [5].
Alternative Uses of VNS In the course of studying VNS for the treatment of epilepsy, a number of serendipitous observations have occurred. Many patients report an improvement in mood, cognition, and wellbeing not related to seizure control per se [1,33]. Stimulation of the vagus nerve has been shown to enhance retention in verbal learning tasks, confirming the hypothesis that vagus nerve activation modulates memory formation similarly to arousal. In addition, VNS has been shown to exert an antinociceptive effect [61]. As a result of these fortuities, VNS has been proposed as a possible treatment of a number of diverse neurologic conditions [62]. One of the potential applications that has received much attention, both within the medical community and among patient groups and the lay press, is depression. Several lines of evidence support this practice [63]. First is the clinical observation of substantial improvements in mood during VNS trials for epilepsy that weren’t attributable to seizure control alone. Second, neuroanatomic studies of vagal afferent connections suggest that the NTS and LC project to the amygdala, stria terminalis, and other limbic structures involved in mood regulation [10]. In VNS trials for epilepsy, for instance, PET studies have shown decreased blood flow to the hippocampus, amygdala, and cingulate gyrus reminiscent of the effects of selective serotonin reuptake inhibitors and other antidepressant drugs [63]. In addition, many anticonvulsant medications have mood-stabilizing effects and are useful treatments for the depressive phase of bipolar affective disorder [63]. Conversely, electroconvulsive therapy – the most effective antidepressant therapy currently available – has potent
2815
2816
167
Vagal nerve stimulation for seizures
anticonvulsant effects. Furthermore, VNS alters the CNS concentrations of norepinephrine, serotonin, glutamate, and other monoamine neurotransmitters implicated in the pathogenesis of major depression. Finally, it is well established that depressed patients have autonomic system dysfunction that is mediated by the vagus nerve. If depressed patients have abnormalities in brain regions that control the vagus nerve from the top down, then perhaps stimulating the vagus nerve might engage this dysfunctional circuit from the bottom up [63]. A corporate-sponsored, nonrandomized clinical trial of VNS for depression was conducted [64]. In this open-label pilot study, 30 patients with treatment-resistant depression were enrolled. All had failed at least two pharmacological trials, and more than half had failed ECT as well. Following a baseline period with stable medication regimens, patients underwent insertion of the VNS device. A 2-week single-blind recovery period was followed by a 10-week period of active stimulation, using parameters similar to those employed for epilepsy. Functional status was assessed by several scales, with response defined by a 50% or greater reduction in baseline scores. For both the 28-item Hamilton Depression Rating Scale and the Clinical Global ImpressionsImprovement index, the response rate was 40%. For the Montgomery-Asberg Depression Rating Scale, the response rate was 50%, 17% of patients had complete remission. Symptomatic responses and functional improvements have been sustained during follow up as long as 9 months [64]. The promising results of the pilot study have been replicated in larger company-sponsored studies, including the pivotal randomized, controlled acute phase trial that enrolled 235 patients [65]. Furthermore, the durability of the treatment effect has been confirmed. In followup of the patients in the pilot trial, for instance, 72 and 61% of the early responders (those with a 50% reduction in symptom scores by 3 months
of therapy) were still responders at 12 and 24 months, respectively, and nearly 80% of the later responders (those achieving a 50% reduction by 12 months) were still responders at 24 months [66]. For the pivotal trial, analogous numbers were 63, 77, and 61% [66]. As with VNS for epilepsy, a trend towards greater efficacy was observed over time. In the pilot study, for instance, Hamilton-D-28 response rates were 31% at 3 months, 44% after 1 year, and 42% after 2 years based on last visit carried forward analysis [67]. Remission rates for similar time periods were 15, 27, and 22% [67]. Based on these and other studies, the VNS device gained FDA approval in July 2005 ‘‘for the adjunctive long-term treatment of chronic or recurrent depression for patients 18 years of age or older who are experiencing a major depressive episode and have not had an adequate response to four or more adequate antidepressant treatments.’’ As with VNS for epilepsy, cost analysis suggests that, compared to other treatment modalities such as maintenance electroconvulsive therapy (M-ECT) at frequent intervals, VNS for depression is associated with a cost savings, The cost of VNS is estimated at $3,900 annually ($31,000 cumulative total for surgery, device, and office visits, prorated over 8 years) as opposed to M-ECT at $1,000 per treatment [68]. Despite the suggestion of clinical efficacy and the potential cost savings, however, the Centers for Medicare and Medicaid Services issued a non-coverage decision (NCD) applying to VNS therapy for refractory depression. Since third party insurers generally issue similar policies, the future of VNS for depression is questionable. Furthermore, the validity of the scientific data in support of VNS for the treatment of depression has recently been challenged. Some patient advocacy groups, and even some physicians, have requested that the FDA repeat its review of VNS for depression and repeal its approval for this indication.
Vagal nerve stimulation for seizures
As with VNS for epilepsy, the locus of effect and mode of action of VNS for depression remains uncertain. PET studies have shown VNS-induced increases in rCBF in the bilateral orbitofrontal cortex, bilateral anterior cingulated cortex, and right superior and medial frontal cortex. Decreases were found in the bilateral temporal cortex and right parietal regions. These areas of change are consistent with brain structures associated with depression and the afferent pathways of the vagus nerve [69]. More recently, functional MRI using blood oxygenation level dependent (BOLD) signal as the dependent variable has revealed that, over time, VNS therapy is associated with ventro-medial prefrontal cortex deactivation, an effect similar to that of other antidepressant treatments [70]. Also, as with VNS for epilepsy, it is difficult to predict which patients will respond to VNS for depression and which won’t. In one study, however, auditory event-related potentials (ERP) demonstrated some prognostic value, as enhancement of the P300 peak distinguished VNS responders from non-responders 10 weeks after therapy onset [71]. The P300 peak also correlated with Hamilton-D scores. Other hints suggest that VNS may have utility for additional neuropsychiatric illnesses. For instance, several theories of anxiety purport faulty or erratic interpretation of peripheral information that flows into the CNS [63]. By affecting the flux of this information, VNS might have therapeutic potential in treating anxiety disorders. Similarly, the vagus nerve is known to transmit signals pertaining to hunger, satiety, and pain. For those reasons, potential applications for obesity, addiction, and pain syndromes seem plausible. The effects of VNS on feeding behavior were investigated in a canine model [Reddy, unpublished observations, 1999]. Six dogs underwent bilateral VNS at parameters similar to those used for epilepsy. Feeding times, amount consumed, and weight were serially monitored and
167
compared with baseline. In response to VNS, feeding behavior changed following a variable period of latency. Both the rate of consumption and the amount consumed decreased, leading to weight loss. When stimulation was suspended, eating returned to baseline in 3–5 days, but resuming the stimulation reproduced the initial dietary changes. Similarly, in a pig model, low frequency vagus nerve pacing produced continuous decreases in food intake and body weight gain over 8 weeks of experimentation [72]. In humans treated with VNS for refractory depression significant, gradual weight loss despite the lack of any intentional change in diet or exercise has been another serendipitous observation [73]. Weight loss did not correlate with changes in mood symptoms. In another study of patients undergoing VNS for depression, acute device activation resulted in a significant change in cravings ratings for sweet foods [74]. A phase I clinical trial of the effects of VNS on obesity is currently in progress. In rodent models, VNS has been shown to enhance long-term potentiation, and human studies suggest a favorable impact on recognition memory [75]. Based on these observations, a pilot study for the treatment of Alzheimer’s disease been conducted [76]. In this trial of 17 patient, 7 (42%) and 12 (70%) demonstrated improved or stable Alzheimer’s Disease Assessment Scalecognitive subscale (ADAS-cog) and Mini-Mental State Examination (MMSE) scores, respectively, after 1 year of stimulation. Furthermore, there was a median reduction of cerebrospinal fluid tau protein, a marker of Alzheimer’s disease severity, by 4.8% (p = 0.057) after 1 year [76]. Seizures are a common comorbidity of the autism spectrum disorders and occur in as many as 30% of patients [77]. Preliminary results in patients suffering from both epilepsy and autism or Asperger’s syndrome suggest that VNS may exert beneficial effects in treatment of the latter conditions alone [77,78].
2817
2818
167
Vagal nerve stimulation for seizures
Moreover, the NTS sends fibers to the dorsal raphe and other areas of the reticular formation known to control levels of consciousness [10]. Thus, VNS has been considered as potential treatment for disorders of sleep or alertness such as narcolepsy and coma. VNS is also a possible treatment for additional conditions such as movement disorders,. For instance, a patient with medically refractory epilepsy and concurrent Tourette’s syndrome underwent VNS and experienced improvements of his tics [79]. VNS does not appear to be helpful in the treatment of essential tremor, however [80]. Electrical stimulation of vagal afferents inhibits spinal nociceptive reflexes and transmission, resulting in altered pain perception and thresholds. Changes in the activity of spinal trigeminal nucleus neurons may underlie this effect [61]. These findings provide the impetus for studying VNS in the treatment of chronic headache, intractable migraines, and cluster headache. Through mechanisms that remain to be clarified, VNS following experimental traumatic brain injury enhances recovery of motor and cognitive function. Attenuation of cerebral edema may underlie this effect [81], and further study may disclose additional applications for VNS in the treatment of this condition. Finally, it is important to recall that the VNS device permits delivery of stimulation at different amplitudes, frequencies, pulse widths, and duty cycles [1]. At present, these settings are titrated empirically according to desired effect and tolerability. Although stimulation paradigm (standard vs. rapid cycle) and output current do not correlate with the effectiveness of VNS for seizure control [38], it is possible that varying these parameters in different combinations may affect different regions of the brain, thereby influencing distinct pathologic conditions and producing pleiotropic effects. As more becomes known about the physiology of afferent autonomic stimulation, the utility of VNS is likely to broaden.
Future Directions VNS remains a promising therapy for a population of patients who have failed prior medication trials and have no other surgical option. However, several limitations pertain to this therapy. These include the low likelihood of complete seizure eradication and the inability to predict who will respond favorably. Such modest gains will curtail more widespread application of VNS. An area of active investigation is the prognostication of possible benefit from VNS therapy. As reported above, atonic seizures tend to respond favorably whereas myoclonic seizures do not. Some studies suggest better outcomes among patients where the onset of seizure activity occurs in the temporal area, while patients with frontal or frontocentral seizures have the poorest outcomes [82]. Seizure semiology does not seem to have predictive value. As stressed above, it is important to consider that the stimulation parameters for VNS are empiric. Some studies have shown no differences between patients who were placed on standard versus rapid cycling, or changed from standard to rapid paradigms during crossover phase [38]. Likewise, in a randomized trial of three unique modes of VNS, which varied by duty cycle, all three were equally effective [83]. However, individual patients who fail to demonstrate benefit from one paradigm may improve substantially as stimulation parameters are titrated over time. The success of directly stimulating the vagus nerve via an implantable device has prompted interest in less invasive methods of afferent cranial nerve activation. Building upon early experiments of transcutaneous electrical stimulation of the recurrent laryngeal nerve as a diagnostic aid in laryngoscopy and as a therapeutic tool in controlling the glottic aperture [84], more recent studies have focused on the potential of transcutaneous stimulation of the vagus nerve for control of partial onset seizures [85] and potentially
Vagal nerve stimulation for seizures
other conditions. For instance, transcutaneous electrical stimulation of the sensory auricular branch of the vagus nerve innervating parts of the outer ear results in sensory evoked potentials originating in vagus nuclei within the brainstem [86]. In one application of this approach, vagal branches of the left outer auditory canal were simulated, followed by functional MRI and psychometric assessments of mood. Transcutaneous VNS produced brain activation patterns in limbic and other areas reminiscent of invasive VNS and also caused significant improvements in overall feelings of well-being [87]. Similarly, the trigeminal nerve (the fifth cranial nerve) is known to converge upon many of the same brainstem nuclei as the vagus nerve. The safety and preliminary efficacy of afferent trigeminal nerve stimulation (TNS) for epilepsy was recently evaluated in a pilot feasibility study of transcutaneous stimulation of the infraorbital and supraorbital branches of the trigeminal nerve [88]. In this study, TNS was well tolerated, and four (57%) of seven subjects who completed 3 or more months of stimulation experienced a 50% or greater reduction in seizure frequency. It is therefore conceivable that, in the future, direct (invasive) vagus nerve stimulation for the treatment of epilepsy or other disorders will be supplanted by transcutaneous or less invasive methods of afferent cranial stimulation.
4.
5.
6.
7. 8.
9.
10.
11.
12.
13.
14.
15.
References 1. Amar AP, Heck CN, Levy ML, Smith T, DeGiorgio CM, Oviedo S, Apuzzo ML. An institutional experience with cervical vagus nerve trunk stimulation for medically refractory epilepsy: rationale, technique, and outcome. Neurosurgery 1998;43:1265-80. 2. Amar AP, Heck CN, DeGiorgio CM, Apuzzo MLJ. Experience with vagus nerve stimulation for intractable epilepsy: some questions and answers. Neuro med-chir 1999;39:489-95. 3. Amar AP, DeGiorgio CM, Tarver WB, Apuzzo MLJ, E05 Study Group. Long-term multicenters experience with vagus nerve stimulation for intractable partial seizures:
16.
17.
18.
167
results of the XE5 trial. Stereotact Funct Neurosurg 1999;73:104-8. Amar AP, Levy ML, McComb JG, Apuzzo MLJ. Vagus nerve stimulation for control of intractable seizures in childhood. Pediatr Neurosurg 2001;34:218-23. Amar AP, Apuzzo MLJ, Liu CY. Vagus nerve stimulation (VNS) therapy after failed cranial surgery for intractable epilepsy: results from the VNS therapy patient outcome registry. Neurosurgery 2004;55:1086-93. Amar AP. Vagus nerve stimulation for the treatment of intractable epilepsy. Expert Rev Neurother 2007;7:1763-73. Amar AP, Levy ML, Liu CY, Apuzzo MLJ. Vagus nerve stimulation. Proc IEEE 2008;96:1142-1151. Amar AP, Levy ML, Apuzzo MLJ. Vagus nerve stimulation for intractable epilepsy. In: Rengechary S, editor. Neurosurgical operative atlas (vol. 9). Chicago: American Association of Neurological Surgeons; 2000. p. 179-88. DeGiorgio CM, Amar A, Apuzzo MLJ. Surgical anatomy, implantation technique, and operative complications. In: Schachter SC, editor. Vagus nerve stimulation. London, UK: Dunitz; 2001. p. 31-50. Rutecki P. Anatomical, physiological, and theoretical basis for the antiepileptic effect of vagus nerve stimulation. Epilepsia 1990;31 Suppl 2:S1-S6. Krahl SE, Senanayake SS, Handforth A. Destruction of peripheral C-fibers does not alter subsequent vagus nerve stimulation-induced seizure suppression in rats. Epilepsia 2001;42:586-9. Cheng G, Zhou X, Qu J, Ashwell KW, Paxinos G. Central vagal sensory and motor connections: human embryonic and fetal development. Auton Neurosci 2004;114:83-96. Dorr AE, Debonnel G. Effect of vagus nerve stimulation on serotonergic and noradrenergic transmission. J Pharmacol Exp Ther 2006;318:890-8. Krahl SE, Clark KB, Smith DC, Browning RA. Locus coeruleus lesions suppress the seizure-attenuating effects of vagus nerve stimulation. Epilepsia 1998;39 Suppl 7:709-14. Hammond EJ, Uthman BM, Reid SA, Wilder BJ. Electrophysiological studies of cervical vagus nerve stimulation in humans: I. EEG effects. Epilepsia 1992;33:1013-20. Hammond EJ, Uthman BM, Reid SA, Wilder BJ. Electrophysiological studies of cervical vagus nerve stimulation in humans: II. Evoked potentials. Epilepsia 1992;33:1021-8. Salinsky MC, Burchiel KJ. Vagus nerve stimulation has no effect on awake EEG rhythms in humans. Epilepsia 1993;34:299-304. Santiago-Rodrı´guez E, Alonso-Vanegas M, Ca´rdenasMorales L, Harmony T, Bernardino M, Ferna´ndezBouzas A. Effects of two different cycles of vagus nerve stimulation on interictal epileptiform discharges. Seizure 2006;15:615-20.
2819
2820
167
Vagal nerve stimulation for seizures
19. Naritoku DK, Terry WJ, Helfert RH. Regional induction of fos immunoreactivity in the brain by anticonvulsant stimulation of the vagus nerve. Epilepsy Res 1995; 22:53-62. 20. McGregor A, Wheless J, Baumgartner J, Bettis D. Rightsided vagus nerve stimulation as a treatment for refractory epilepsy in humans. Epilepsia 2005;46:91-6. 21. Fallen EL. Vagal afferent stimulation as a cardioprotective strategy? Introducing the concept. Ann Noninvasive Electrocardiol 2005;10:441-6. 22. Amark P, Sto¨dberg T, Wallstedt L. Late onset bradyarrhythmia during vagus nerve stimulation. Epilepsia 2007;48:1023-4. 23. Agnew WF, McCreery DB. Considerations for safety with chronically implanted nerve electrodes. Epilepsia 1990;31 Suppl 2:S27-S32. 24. Tougas G, Fitzpatrick D, Huboda P, Talalla A, Shine G, Hunt R, Upton AR. Effects of chronic left vagal stimulation on visceral vagal function in man. Pacing Clin Electrophysiol 1992;15:1588-96. 25. Guilfoyle MR, Fernandes H, Price S. In vivo alteration of strata valve setting by vagus nerve stimulator-activating magnet. Br J Neurosurg 2007;21:41-2. 26. Patil A, Chand A, Andrews R. Single incision for implanting a vagal nerve stimulator system (VNSS): technical note. Surg Neurol 2001;55:103-5. 27. Asconape JJ, Moore DD, Zipes DP, Hartman LM, Duffel WH. Bradycardia and asystole with the use of vagus nerve stimulation for the treatment of epilepsy: a rare complication of intraoperative device testing. Epilepsia 1999;40:1452-4. 28. Bauman JA, Ridgway EB, Devinsky O, Doyle WK. Subpectoral implantation of the vagus nerve stimulator. Neurosurgery 2006;58:322-6. 29. Espinosa J, Aiello MT, Naritoku DK. Revision and removal of stimulating electrodes following long-term therapy with the vagus nerve stimulator. Surg Neurol 1999;51:659-64. 30. Bruce D, Li M, Fraser R, Alksne J. The Neuro-Cybernetic prosthesis (NCP) system for the treatment of refractory partial seizures: surgical technique and outcomes. Epilepsia 1998;39 Suppl 6:92-3 (abstract). 31. Le H, Chico M, Hecox K, Frim D. Interscapular placement of a vagal nerve stimulator pulse generator for prevention of wound tampering. Technical note. Pediatr Neurosurg 2002;36:164-6. 32. Handforth A, DeGiorgio CM, Schachter SC, Uthman BM, Naritoku DK, Tecoma ES, Henry TR, Collins SD, Vaughn BV, Gilmartin RC, Labar DR, Morris GL, Salinsky MC, Osorio I, Ristanovic RK, Labiner DM, Jones JC, Murphy JV, Ney GC, Wheless JW. Vagus nerve stimulation for partial-onset seizures: a randomized active-control trial. Neurology 1998;51:48-55. 33. Morris GL, Mueller WM, E01-E05 VNS Study Group. Long-term treatment with vagus nerve stimulation
34. 35.
36.
37.
38.
39.
40.
41.
42. 43.
44.
45.
46. 47.
48.
in patients with refractory epilepsy. Neurology 1999;53:1731-5. Fisher RS, Handforth A. Reassessment: vagus nerve stimulation for epilepsy. Neurology 1999;53:666-9. Boon P, Vonck K, Vandekerckhove T, D’have M, Nieuwenhuis L, Michielsen G, Vanbelleghem H, Goethals I, Caemaert J, Calliauw L, De Reuck J. Vagus nerve stimulation for medically refractory epilepsy: efficacy and cost-benefit analysis. Acta. Neurochir. 1999;141:447-53. Bernstein AL, Barkan H, Hess T. Vagus nerve stimulation therapy for pharmacoresistant epilepsy: effect on health care utilization. Epilepsy Behav 2007;10:134-7. McHugh JC, Singh HW, Phillips J, Murphy K, Doherty CP, Delanty N. Outcome measurement after vagal nerve stimulation therapy: proposal of a new classification. Epilepsia 2007;48:375-8. Labar D. Vagus nerve stimulation for 1 year in 269 patients on unchanged antiepileptic drugs. Seizure 2004;13:392-8. Ardesch JJ, Buschman HP, Wagener-Schimmel LJ, van der Aa HE, Hageman G. Vagus nerve stimulation for medically refractory epilepsy: a long-term follow-up study. Seizure 2007;16:579-85. Spanaki MV, Allen LS, Mueller WM, Morris GL. Vagus nerve stimulation therapy: 5-year or greater outcome at a university-based epilepsy center. Seizure 2004;13:587-90. De Herdt V, Boon P, Ceulemans B, Hauman H, Lagae L, Legros B, Sadzot B, Van Bogaert P, van Rijckevorsel K, Verhelst H, Vonck K. Vagus nerve stimulation for refractory epilepsy: a Belgian multicenter study. Eur J Paediatr Neurol 2007;11:261-9. Schachter SC, Saper CB. Progress in epilepsy research: vagus nerve stimulation. Epilepsia 1998;39:677-86. Shaw GY, Sechtem P, Searl J, Dowdy ES. Predictors of laryngeal complications in patients implanted with the cyberonics vagal nerve stimulator. Ann Otol Rhinol Laryngol 2006;115:260-7. Annegers JF, Coan SP, Hauser WA, Leetsma J, Duffell W, Tarver B. Epilepsy, vagal nerve stimulation by the NCP system, mortality, and sudden, unexplained death. Epilepsia 1998;39:206-12. Huf RL, Mamelak A, Kneedy-Cayem K. Vagus nerve stimulation therapy: 2-year prospective open-label study of 40 subjects with refractory epilepsy and low IQ who are living in long-term care facilities. Epilepsy Behav 2005;6:417-23. Murphy JV. Left vagal nerve stimulation in children with medically refractory epilepsy. J. Pediatr. 1999;134:563-6. Blount JP, Tubbs RS, Kankirawatana P, et al. Vagus nerve stimulation in children less than 5 years old. Childs Nerv Syst 2006;22:1167-9. Rychlicki F, Zamponi N, Trignani R, Ricciuti RA, Iacoangeli M, Scerrati M. Vagus nerve stimulation:
Vagal nerve stimulation for seizures
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60. 61.
62.
63.
clinical experience in drug-resistant pediatric epileptic patients. Seizure 2006;15:483-90. Alexopoulos AV, Kotagal P, Loddenkemper T, Hammel J, Bingaman WE. Long-term results with vagus nerve stimulation in children with pharmacoresistant epilepsy. Seizure 2006;15:491-503. Benifla M, Rutka JT, Logan W, Donner EJ. Vagal nerve stimulation for refractory epilepsy in children: indications and experience at The Hospital for Sick Children. Childs Nerv Syst 2006;22:1018-26. Saneto RP, Sotero de Menezes MA, et al. Vagus nerve stimulation for intractable seizures in children. Pediatr Neurol 2006;35:323-6. Kang HC, Hwang YS, Kim DS, Kim HD. Vagus nerve stimulation in pediatric intractable epilepsy: a Korean bicentric study. Acta Neurochir Suppl 2006;99:93-6. You SJ, Kang HC, Kim HD, et al. Vagus nerve stimulation in intractable childhood epilepsy: a Korean multicenter experience. J Korean Med Sci. 2007;22:442-5. Arthur TM, Saneto RP, de Menezes MS, et al. Vagus nerve stimulation in children with mitochondrial electron transport chain deficiencies. 1. Mitochondrion 2007;7:279-83. Murphy JV, Hornig GW, Schallert GS, Tilton CL. Adverse events in children receiving intermittent left vagal nerve stimulation. Pediatr Neurol 1998;19:42-4. Khurana DS, Reumann M, Hobdell EF, Neff S, Valencia I, Legido A, Kothare SV. Vagus nerve stimulation in children with refractory epilepsy: unusual complications and relationship to sleep-disordered breathing. Childs Nerv Syst 2007;11:1309-12. Papacostas SS, Myrianthopoulou P, Dietis A, Papathanasiou ES. Induction of central-type sleep apnea by vagus nerve stimulation. Electromyogr Clin Neurophysiol 2007;47:61-3. Rychlicki F, Zamponi N, Cesaroni E, Corpaci L, Trignani R, Ducati A, Scerrati MRychlicki F, Zamponi N, Cesaroni E, . et al. Complications of vagal nerve stimulation for epilepsy in children. Neurosurg Rev 2006;29:103-7. Zaaimi B, Grebe R, Berquin P, Wallois F. Vagus nerve stimulation therapy induces changes in heart rate of children during sleep. Epilepsia 2007;48:923-30. Ng M, Devinsky O. Vagus nerve stimulation for refractory idiopathic generalised epilepsy. Seizure 2004;13:176-8. Multon S, Schoenen J. Pain control by vagus nerve stimulation: from animal to man. . .and back. Acta Neurol Belg 2005;105:62-7. Ansari S, Chaudhri K, Al Moutaery KA. Vagus nerve stimulation: indications and limitations. Acta Neurochir Suppl 2007;97 (Pt 2):281-6. George MS, Sackeim HA, Rush AJ, Marangell LB, Nahas Z, Husain MM, Lisanby S, Burt T, Goldman J, Ballenger JC. Vagus nerve stimulation: a new tool for
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
167
brain research and therapy. Biol. Psychiatry 2000;47:287-95. Rush AJ, George MS, Sackeim HA, Marangell LB, Husain MM, Giller C, Nahas Z, Haines S, Simpson RK Jr, Goodman R. Vagus nerve stimulation (VNS) for treatment-resistant depressions: a multicenter study. Biol. Psychiatry 2000;47:276-86. Rush AJ, Marangell LB, Sackeim HA, George MS, Brannan SK, Davis SM, Howland R, Kling MA, Rittberg BR, Burke WJ, Rapaport MH, Zajecka J, Nierenberg AA, Husain MM, Ginsberg D, Cooke RG. Vagus nerve stimulation for treatment-resistant depression: a randomized, controlled acute phase trial. Biol. Psychiatry 2005;58:347-54. Sackeim HA, Brannan SK, John Rush A, George MS, Marangell LB, Allen J. Durability of antidepressant response to vagus nerve stimulation (VNSTM). Int J Neuropsychopharmacol 2007;10:817-26. Nahas Z, Marangell LB, Husain MM, Rush AJ, Sackeim HA, Lisanby SH, Martinez JM, George MS. Two-year outcome of vagus nerve stimulation (VNS) for treatment of major depressive episodes. J Clin Psychiatry 2005;66:1097-104. Warnell RL, Elahi N. Introduction of vagus nerve stimulation into a maintenance electroconvulsive therapy regimen: a case study and cost analysis. J ECT 2007;23:114-9. Conway CR, Sheline YI, Chibnall JT, George MS, Fletcher JW, Mintun MA. Cerebral blood flow changes during vagus nerve stimulation for depression. Psychiatry Res 2006;146:179-84. Nahas Z, Teneback C, Chae JH, Mu Q, Molnar C, Kozel FA, Walker J, Anderson B, Koola J, Kose S, Lomarev M, Bohning DE, George MS. Serial vagus nerve stimulation functional MRI in treatment-resistant depression. Neuropsychopharmacology 2007;32:1649-60. Neuhaus AH, Luborzewski A, Rentzsch J, Brakemeier EL, Opgen-Rhein C, Gallinat J, Bajbouj M. P300 is enhanced in responders to vagus nerve stimulation for treatment of major depressive disorder. J Affect Disord 2007;100:123-8. Matyja A, Thor PJ, Sobocki J, Laskiewicz J, Kekus J, Tuz R, Koczanowski J, Zaraska W. Effects of vagal pacing on food intake and body mass in pigs. Folia Med Cracov 2004;45:55-62. Pardo JV, Sheikh SA, Kuskowski MA, SurerusJohnson C, Hagen MC, Lee JT, Rittberg BR, Adson DE. Weight loss during chronic, cervical vagus nerve stimulation in depressed patients with obesity: an observation. Int J Obes 2007;31:1756-9. Bodenlos JS, Kose S, Borckardt JJ, Nahas Z, Shaw D, O’Neil PM, George MS. Vagus nerve stimulation acutely alters food craving in adults with depression. Appetite 2006;48:145-53. Clark KB, Naritoku DK, Smith DC, Browning RA, Jensen RA. Enhanced recognition memory following
2821
2822
167 76.
77.
78.
79.
80.
81.
Vagal nerve stimulation for seizures
vagus nerve stimulation in human subjects. Nat. Neuroscience 1999;2:94-8. Merrill CA, Jonsson MA, Minthon L, Ejnell H, C-son Silander H, Blennow K, Karlsson M, Nordlund A, Rolstad S, Warkentin S, Ben-Menachem E, Sjo¨gren MJ. Vagus nerve stimulation in patients with Alzheimer’s disease: additional follow-up results of a pilot study through 1 year. J Clin Psychiatry 2006;67:1171-8. Warwick TC, Griffith J, Reyes B, Legesse B, Evans M. Effects of vagus nerve stimulation in a patient with temporal lobe epilepsy and Asperger syndrome: case report and review of the literature. Epilepsy Behav 2007; 10:344-7. Berta S, Park MS, Meltzer HS, Amar AP, Apuzzo MLJ, Levy KM, Levy ML. Vagus nerve stimulation therapy in patients with autism spectrum disorder: results from the vagus nerve stimulation therapy patient outcome registry. Neurosurgery 2005;57:417-8 (abstract). Diamond A, Kenney C, Jankovic J. Effect of vagal nerve stimulation in a case of Tourette’s syndrome and complex partial epilepsy. Mov Disord 2006;21:1273-5. Handforth A, Ondo WG, Tatter S, Mathern GW, Simpson RK Jr, Walker F, Sutton JP, Hubble JP, Jankovic J. Vagus nerve stimulation for essential tremor: a pilot efficacy and safety trial. Neurology 2003;61:1401-5. Clough RW, Neese SL, Sherill LK, Tan AA, Duke A, Roosevelt RW, Browning RA, Smith DC. Cortical edema in moderate fluid percussion brain injury is attenuated by vagus nerve stimulation. Neuroscience 2007;147:286-93.
82. Casazza M, Avanzini G, Ferroli P, Villani F, Broggi G. Vagal nerve stimulation: relationship between outcome and electroclinical seizure pattern. Seizure 2006; 15:198-207. 83. DeGiorgio C, Heck C, Bunch S, Britton J, Green P, Lancman M, Murphy J, Olejniczak P, Shih J, Arrambide S, Soss J. Vagus nerve stimulation for epilepsy: randomized comparison of three stimulation paradigms. 1. Neurology 2005;65:317-9. 84. Sanders I, Aviv J, Biller HF. Transcutaneous electrical stimulation of the recurrent laryngeal nerve: a method of controlling vocal cord position. Otolaryngol Head Neck Surg 1986;95:152-7. 85. Ventureyra EC. Transcutaneous vagus nerve stimulation for partial onset seizure therapy. A new concept. Childs Nerv Syst 2000;16:101-2. 86. Fallgatter AJ, Neuhauser B, Herrmann MJ, Ehlis AC, Wagener A, Scheuerpflug P, Reiners K, Riederer P. Far field potentials from the brain stem after transcutaneous vagus nerve stimulation. J Neural Transm 2003;110:1437-43. 87. Kraus T, Ho¨sl K, Kiess O, Schanze A, Kornhuber J, Forster C. BOLD fMRI deactivation of limbic and temporal brain structures and mood enhancing effect by transcutaneous vagus nerve stimulation. J Neural Transm 2007;114:1485-93. 88. DeGiorgio CM, Shewmon A, Murray D, Whitehurst T. Pilot study of trigeminal nerve stimulation (TNS) for epilepsy: a proof-of-concept trial. Epilepsia 2006; 47:1213-5.
175 Ablative Procedures for Depression V. A. Coenen . C. R. Honey
Depression Depression is the commonest psychiatric disorder in North America. Approximately 11 million patients were affected in the USA in 1990 [1]. Depression is eight times more frequent than schizophrenia and 16 times more frequent than Parkinson’s disease. Angst et al. estimated life time-prevalence in all age groups to be 9.5% [2]. The economical burden of depression on society is higher than that of other chronic diseases such as hypertension, asthma, osteoporosis or rheumatoid arthritis [3,4]. Greenberg estimated this cost in 1990 was as high as US$43.7 billion for the USA alone [1]. The majority of patients who suffer from depression, however, respond well to a combination of medication and psychotherapy. Up to 20%, however, remain refractory and require a combination of medications or electroconvulsive therapy [5]. Within this group there are some patients who remain in a state of severe depression despite all medical efforts. For this small cohort, Neurosurgery needs to be considered.
Surgery for Depression The Swiss psychiatrist Burckhard [6] was the first to formally report a neurosurgical attempt to improve a psychiatric condition. He performed bilateral cortical incisions (topectomies) on six demented, aggressive and institutionalized patients. Burckhard stopped this surgery after the publication of his discouraging results in 1891 and was later criticized because he lacked a rationale for his approach [7,8]. #
Springer-Verlag Berlin/Heidelberg 2009
In 1935, Portuguese neurologist Egas Moniz convinced his neurosurgical colleague Almeida Lima to perform alcohol injections into the frontal lobes of institutionalized psychiatric patients. He coined the term “psychosurgery” and received the Nobel Prize for his work [9]. Walter Freeman, an American psychiatrist, with the initial help of his neurosurgical colleague James Watts, popularized “prefrontal lobotomy” in North America [7,9,10]. Freeman modified the technique from frontal burr holes to what would come to be called the “ice pick lobotomy” with an entry made up through the orbital roof. The procedure’s popularity grew widely, largely through Freeman’s zeal, despite its high complication rate and questionable indications. Ultimately the combination of new medications able to control schizophrenia and the realization that “psychosurgery” had severe side effects, led to the demise of widespread neurosurgical procedures for psychiatric conditions. Many countries banned the operations. A few countries, however, continued their work on what would become “limbic surgery.” In 1937, Papez published his idea of a reverberating circuit for emotion, memory and anxiety. The idea of this circuit was revisited by Nauta [11], who added additional relays in the brain stem and other brain regions to it. The limbic lobe – as already denominated by Broca [12] – became the target structure for surgery in otherwise untreatable affective disorders such as depression, obsessive-compulsive disorder (OCD), and anxiety disorders. The open surgical approaches to the frontal lobes were abandoned but a few neurosurgeons pursued less invasive surgical strategies.
2944
175
Ablative procedures for depression
Stereotaxy had evolved in the late 1940s [13], and allowed more precision with less complications. A review of the literature of ablative surgery for depression reveals that there are three target sites and four different operations as shown in > Table 175-1: subcaudate tractotomy (SCT),
anterior cingulotomy (ACT), stereotactic limbic leucotomy (SLL) and, to a certain extent, anterior capsulotomy (AC) (> Figure 175-1).
. Table 175‐1 Stereotactic ablative procedures for the treatment of refractory depression
(Stereotactic) Subcaudate Tractotomy (SCT)
Surgery type
Anatomical target
Stereotactic subcaudate tractotomy (SCT) Anterior cingulotomy (ACT)
Frontobasal subcortical white matter Anterior cingulum (Papez circuit) Frontobasal subcortical white matter and anterior cingulum (ACT & SCT) Anterior limb of internal capsule, fronto-thalamic projections
Stereotactic limbic leucotomy (SLL)
Anterior capsulotomy (AC)
Four Different Ablative Surgical Procedures for Depression
A post mortem evaluation of the extent of lesions in patients who had received successful orbital undercutting procedures [7] suggested that the subcaudate region might be the ideal target. Stereotactic subcaudate tractotomy was reported by Knight [14,15] and dramatically reduced the size of the surgical lesion from the prefrontal leucotomy to a stereotactic target [7,16]. The procedure was more common in Great Britain than in North America where cingulotomy was preferred [16]. Newcombe described the SCT
. Figure 175‐1 Four stereotactic ablative procedures for refractory depression (schematic). Left, coronal; right, sagittal. ACT ¼ anterior cingulotomy, SCT ¼ subcaudate tractotomy, SLL ¼ stereotactic limbic leucotomy (essentially a combination of ACT þ SCT), AC ¼ anterior capsulotomy
Ablative procedures for depression
lesions as located just posterior to the orbitofrontal cortex. Although some authors reported the lesion included substantia inominata [16,17], most authors felt it did not [18,19].
175
. Figure 175‐2 Postoperative axial MRI following a subcaudate tractotomy (SCT). (image courtesy of R. Cosgrove, MGH, Boston)
Technique of SCT Knight’s minimally invasive technique used (90Y) Yttrium seeds of 7 mm length and 1 mm diameter. The operation was performed through bilateral burr holes and utilized a ventriculogram. The target area was typically 15 mm from the midline, 10 mm above the planum sphenoidale and at the most anterior aspect of the sella. Six rods were implanted on each side in three lines with an inter-rod distance of 3 mm and an interline distance of 5 mm. 90Y emits a strong betaradiation and decays to stable Zirconium with a half life of approximately 60 h. The lesions were therefore matured within a few weeks and typically measured 20 mm length, 10–20 mm width and 5 mm thickness [19]. The lesions were shown to be more uniform than those done with orbital undercutting [20]. In further developments of the technique and particularly in the later limbic leucotomies (see below), lesions were created with cryo-probes or diathermy instead of radioactive seeds [7,21]. An example is shown in (> Figure 175-2).
Outcome After SCT The outcome after SCT was reviewed over the years by various authors [17,18,21–23]. Probably the most extensive review came from the group of Go¨ktepe et al. [18] who evaluated 208 patients at least 2½ years after surgery. Amongst these patients, 78 suffered from intractable depression. The outcome was graded according to an earlier proposal by Kerr [24] into groups of patients who did well (categories I and II), and those patients who showed a poor outcome (categories
III and IV). According to this evaluation, almost 68% of patients showed a clear improvement of their condition. Amongst patients treated for depression, OCD and anxiety, those suffering from depression appeared to respond the best to SCT. Poynton and co-workers [23] prospectively followed 23 patients who underwent stereotactic subcaudate tractotomy (SST), 16 of those had a major depression. The authors performed serial assessments pre-operatively, 2 weeks and 6 months and 1 year postoperatively, using the Hamilton rating scale for depression [25], Beck Depression inventory [26], and Taylor Manifest Anxiety Scale [27]. A significant decrease in depression scores at 2 weeks was documented and maintained at 6 months and 1 year post surgery. A second study from the same institution [18] was published by Hodgkiss et al. [22] in 1995. It reported treatment between 1979 and 1991. The results were less favorable with only 34% improvement (category I and II) at 12 months follow up. Two reasons for the poorer improvement in the more recent study
2945
2946
175
Ablative procedures for depression
were proposed; first, the follow-up time of 12 months may have been too short to capture those patients with more delayed benefits [18]. Second, the later patient population was likely more refractory since new more aggressive pharmacologic treatments (unavailable during the first series) may have eliminated less severe patients from surgical consideration [22]. The lesions created with the Knight technique typically did not involve the gyral surface (unlike orbital undercutting), and therefore the reported incidence of postoperative epilepsy for the SCT was much lower [20,28].
stereotactic lesions. Burr holes 9.5 cm above the nasion and 1.5 cm to either side of the midline allowed ventricular needles to produce a pneumoencephalogram. Guided by orthogonal fluoroscopy, needles with uninsulated tips were placed 2–4 m posterior the anterior tip of the lateral ventricle, 7 mm lateral to the midline, and 1 mm dorsal (above) the ventricular shadow. The lesion was created with radio-frequency current at 85˚C for 75 s. The resultant lesions were typically 1 cm in diameter. In a later MR-guided technique published by Spangler et al. [31], the lesions were placed in a comparable location with some patients receiving additional lesions 1 cm above the first. An example is shown in (> Figure 175-3).
Anterior Cingulotomy (ACT) According to Bailey et al. [29], the anterior cingulotomy (ACT), or cingulotractotomy, has an important role in the treatment of severe medically refractory depression. Fulton originally proposed that the key to every affective illness lies in the limbic (cingulate) circuitry. In subsequent years reproducible techniques of ACT were developed [10,29–31]. ACT has been the most common neurosurgical treatment for depression in North America for the last 30 years [7,10,30,31]. Other current indications include chronic anxiety and OCD.
Technique of ACT Baily et al. [29] reported an open surgical approach. Bilateral craniotomies were performed 13 cm above the glabella and 2.5 cm to the sides. A dissection was performed through the cortex in the plane of the anterior horn of the lateral ventricle down to the anterior cingulate gyrus. The inferior aspect of cingulate gryus was visualized and then the dissection continued to the medial side of the gyrus. A lesion, less than 1.5 cm, was performed bilaterally [29]. Ballantine [10,30] reported a less invasive technique for ACT using
Outcome After ACT Early reports from Bailey et al. [29] claimed success rates of 85% following ACT. The patients included depressed as well as obsessed and phobic patients. More recent reviews, however, claim much more modest success with only 33% improving.
. Figure 175‐3 Postoperative coronal MRI following an anterior cingulotomy (ACT). (image courtesy of A. Lozano, TWH, Toronto)
Ablative procedures for depression
The reduced success rate in recent reviews may reflect a more treatment refractory cohort of patients since more potent psycho-pharmacological drugs are available now [7,31]. Regarding adverse events related to ACT, an early study reported low morbidity [29]. Ballantine published a review of 886 patients from a multi-centric study [10] with a mortality of 0.11%. Severe complications (hemiplegia, coma) occurred in 0.45% [10]. An evaluation of the results of a single institute performing ACT between 1962 and 2004 showed that the procedure was safe with no mortality, 2 infections in >800 patients treated [7]. In their MR-guided approach, Spangler et al.[31] described that 22% of patients developed transient urinary incontinence lasting for days as well as headaches, dizziness, low-grade fever and confusion lasting <48 h.
Stereotactic Limbic Leucotomy (SLL) According to Kim et al. [17], limbic leucotomy is essentially a combination of bilateral ACT and SCT, although there are slight variations in the surgical approaches and in lesion size and location. The method was introduced in 1973 by Kelly and Richardson [18,32,33]. The basis of this technique was the idea that the dual lesion would produce better results than either single lesion. The subcaudate lesion was thought to interrupt fronto-limbic connections and the cingulate lesion was thought to disconnect Papez circuit [11,33,34].
Technique of SLL Kelly describes the surgical approach in detail [32,35]. The operation was done bilaterally through burr holes on each side just behind the coronal suture. The medial lower quadrant frontal lobe lesions (representing the SCT) usually fell on
175
a line between the base of the anterior clinoid process and a point 1 cm posterior of the tip of the frontal horn of the lateral ventricle on a lateral pneumoencephalogram. Two lesions were placed 1–1.5 cm above the floor of the anterior cranial fossa, 6 mm and 14 mm lateral to the midline. For the cingulotomy, two lesions were made. The first lesion was made on a vertical line connecting to the base of the anterior clinoid process, 5 mm above the roof of the ventricle at variable distances from the midline. A second lesion was placed 12–15 mm behind the first one. Before lesioning with a cryoprobe at −70˚C, bipolar stimulation was performed to assess for autonomic responses (changes in respiration, heart rate, blood pressure) which had previously been shown to correlate with stimulation of the limbic circuitry [32,35]. The revised technique described by Cosgrove et al. [36] using magnetic resonance imaging guidance, defined the target points differently and only one cingulotomy lesion was performed on each side. Radio-frequency lesions were placed approximately 2 cm posterior to the tip of the frontal horn of the lateral ventricle within the cingulated gyrus. The posteromedial frontal cortex lesions were placed bilaterally 7 and 15 mm from the midline beneath the head of the caudate nucleus [36]. An example is shown in (> Figure 175-4).
Outcome After SLL In the initial series of 40 patients treated with SLL [32,35], 61% of the patients with depression showed an improvement. In their review, Kim and co-workers report a recovery rate of 78% for the depressed [17,37]. According to Cosgrove [7], 78% of patients with refractory depression improved. Persistent complications are rare with this technique, although in the early postoperative course urinary incontinence, drowsiness and confusion are reported. More recent results from the magnetic resonance imaging guided approach [7,36] showed up to a 50%
2947
2948
175
Ablative procedures for depression
. Figure 175‐4 Postoperative saggital MRI following a stereotactic limbic leucotomy (SLL). (image courtesy of R. Cosgrove, MGH, Boston)
Furthermore, AC has been performed in a subgroup of depressed patients in a combination with ACT [42,43]. The physiological idea behind the procedure was the interruption of the frontothalamic connections. Over time it was discovered that the ideal lesion has to be placed in the ventral portion of the anterior third of the anterior limb of the internal capsule. This target has therefore moved closer to the lesion site for SCT [44].
Technique of AC
improvement for the subgroup of depressed patients (total n = 21, depression n = 6; 29%). Kelly et al. [35] surprisingly report no persistent adverse effects after the SLL, although urinary incontinence appeared to be an early and transient side effect of the anterior cingulate lesion. Montoya and co-workers, on the contrary, reported a significant number of long-term side effects for the whole group. They found 14% of persistent incontinence, 5% of complex partial seizures, and 9% of short-term memory deficits [36].
Stereotactic Anterior Capsulotomy (AC) Most neurosurgeons have reserved AC for the treatment of severe anxiety disorders or obsessive-compulsive disorder (OCD) [38]. When the procedure was first reported, however, Talairach [39] and Leksell [40], who performed cryo- and Gamma-knife lesions [41] to the anterior third of the anterior limb of the internal capsule, reported good results in refractory depression amongst a variety of other psychiatric diseases.
Leksell describes the exact target localization in the anterior third of the anterior limb of the internal capsule. The target is located 20 mm lateral to the midline, 5 mm behind the tip of the frontal horns of the lateral ventricles at the level of the inter-commissural plane. For the Gamma-Knife lesion, a minimum dose of 110 Gy to the edge of the above described target volume is necessary with the 4 mm collimator to create a sufficient lesion [41]. In a review of 29 patients that were treated for OCD with either radiosurgery or thermo-coagulation AC, Lippitz and co-workers found, that effective lesions were in the middle of the anterior limb of the internal capsule, approximately 4 mm above the level of the mid-commissural plane. If thermocoagulation is used, the probe has to follow the 25–30 degree coronal angulation, to create a lesion located within the capsule. For OCD these authors report a relative importance of the rightsided lesion, which might be sufficient if the lesion is done to the described extent [44]. Whether this information can be extrapolated to patients with depression is unclear. An example is shown in (> Figure 175-5).
Outcome After AC In the first 116 patients, who were operated by Leksell, 48% of patient with severe depression
Ablative procedures for depression
. Figure 175‐5 Postoperative axial MRI following an anterior Capsulotomy (AC)
175
brain stimulation (DBS) [47]. This appears to be happening in the neurosurgical treatment of depression [5,45,46,48]. Neuromodulation has the benefit of being titrateable and reversible. Patients and their referring psychiatrists may be more comfortable considering surgery if no permanent lesion is made. Neurosurgeons may be more comfortable targeting new structures if the procedure is reversible [49,50]. There will be one advantage of lesional surgery over neuromodulation and that is the result will be more permanent by its very nature. Neuromodulation will always have the risk of sudden loss of benefit precipitating lethal suicidal crisis when the battery fails.
References
showed a remarkable recovery after AC. Only patients who were free of symptoms were regarded as successfully treated. Of the patients who were regarded as being worse after surgery, 9 (7%) of them belonged to the group of depressed patients.
Conclusion Over a period of 50 years, neurosurgeons have performed a variety of limited ablative surgical procedures for the treatment of medically refractory depression. Ablative procedures have been confirmed as a valid treatment option for this disease. If new, less invasive treatment options like vagus nerve stimulation or repetitive transcranial magnetic stimulation (rTKMS) [45,46] turn out to give effective and eduring results, lesioning surgery for depression will decline in use. In the neurosurgical treatment of movement disorders there has been an evolution from lesioning to neuromodulation with deep
1. Greenberg PE, Stiglin LE, Finkelstein SN, Berndt ER. The economic burden of depression in 1990. J Clin Psychiatry 1993;54:425-6. 2. Angst J. The epidemiology of depressive disorders. Eur Neuropsychopharamcol 1995;5Suppl:95-8. 3. Berto P, D’Illario D, Ruffo P, Di Virgilio R, Rizzo F. Depression: cost-of-illness studies in the international literature, a review. J Ment Health Policy Econ 2000;3:3-10. 4. World Health Organization. Chapter 2: Burden of mental and behavioral disorders. In: The WHO Report 2001: Mental Health, New Understanding, New Hope, 2001. http://www.who.int/whr/2001 5. Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminovicz D, Hamani C, Schwalb JM, Kennedy SH. Deep brain stimulation for treatment-resistant depression. Neuron 2005;45:651-60. ¨ ber Rindenexcisionen, als Beitrag 6. Burckhard G. U zur operativen Therapie der Psychosen. Z Psychiatrie 1891;47:463-548. 7. Cosgrove GR. Neurosurgery for psychiatric disorders. In: Winn HR, editor. Youmans neurological surgery. Philadelphia: Elsevier; 2004. p. 2853-62. 8. Henschen C, Klingler J, Riechert T. Kraniocerebrale Korrelationstopographie thalamofrontaler Bahnen und gezielte Hirnoperationen. Langenbecks Arch U Dtsch Z Chir 1953;273:548-65. 9. Moniz E. Prefrontal leucotomy in the treatment of mental disorders. Am J Psychiatry 1937;93:1379-85. 10. Ballantine HT. Neurosurgery for behavioural disorders. In: Wilkins RH, Rengachary SS, editors. Neurosurgery. NY: McGraw-Hill; 1985. p. 2527-37.
2949
2950
175
Ablative procedures for depression
11. Nauta WJ. Hippocampal projections and related neural pathways to the midbrain in the cat. Brain 1958;81:319-40. 12. Broca P. Anatomie compare´ e des circonvolutions ce´ re´ brales. Le grand lobe limbique et la scissure limbique dans la se´ rie des mammife´ res. Rev Antrhrop 1878;1:385-498. 13. Spiegel EA, Wycis HT, Marks M, Lee AJ. Stereotaxic apparatus for operations on the human brain. Science 1947;106:349-50. 14. Knight G. The orbital cortex as an objective in the surgical treatment of mental illness. The results of 450 cases of open operation and the development of the stereotactic approach. Br J Surgery 1964;51:114-24. 15. Knight G. Stereotactic tractotomy in the surgical treatment of mental illness. J Neurol Neurosurg Psychiatry 1965;28:304-10. 16. Mashtour GA, Walker EE, Martuza RL. Psychosurgery: past present and future. Brain Res Rev 2005;48:409-19. 17. Kim MC, Lee TK, Choi CR. Review of long-term results of stereotactic psychosurgery. Neurol Med Chir (Tokyo) 2002;42:365-71. 18. Go¨ktepe EO, Young LB, Bridges PK. A further review of the results of stereotactic subcaudate tractotomy. Br J Psychiatry 1975;126:270-80. 19. Newcombe R. The lesion in stereotactic subcaudate tractotomy. Br J Psychiatry 1975;126:478-81. 20. Corsellis JAN, Jack AB. Neuropathological observations on yttrium implants and on undercutting in the orbital frontal areas of the brain. In: Laitinen LV, Livingstone KE, editors. Surgical approaches to psychiatry, Ch. 52. Lancaster: Medical and Technical Publishing Co; 1973. p. 90-9. 21. Kalyanaraman S, Ramamurthi B. Stereotaxic basofrontal tractotomy. Neurol India 1973;21:113-8. 22. Hodgkiss AD, Malizia AL, Bartlett JR, Bridges PK. Outcome after the psychosurgical operation of stereotactic subcaudate tractotomy, 1979–1991. J Neuropsychiatry Clin Neurosci 1995;7:230-4. 23. Poynton AM, Kartsounis LD, Bridges PK. A prospective clinical study of stereotactic subcaudate tractotomy. Psychol Med 1995;25:763-70. 24. Kerr TA, Roth M, Shapira K, Gurney C. The assessment and prediction of outcome in affective disorders. Br J Psychiatry 1972;121:167-74. 25. Hamilton M. Development of a rating scale for primary depressive illness. Br J Soc Clin Psychol 1967;6:278-96. 26. Beck AT, Ward CHH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Gen Psychiatry 1961;4:561-71. 27. Taylor JA. A personality scale of manifest anxiety. J Abnorm Soc Psychiatry 1955;48:285-90. 28. Stro¨m-Olson R, Carlisle S. Bi-frontal stereotactic tractotomy: a follow-up study of its effects on 210 patients. Br J Psychiatry 1971;118:141-54.
29. Bailey HR, Dowling JL, Davies E. Studies in depression, III* the control of affective illness by cingulotractotomy: a review of 150 cases. Med J Aust 1973;2:366-71. 30. Ballantine HT Jr, Cassidy WL, Flanagan NB, Marino R Jr. Stereotactic anterior cingulotomy for neuropsychiatric illness and intractable pain. J Neurosurg 1967;26:488-95. 31. Spangler WJ, Cosgrove GR, Ballantine HT, Cassem EH, Rauch SL, Nierenberg A, Price BH. Magnetic resonance image-guided stereotactic cingulotomy for intractable psychiatric disease. Neurosurgery 1996;38:1071-78. 32. Kelly D, Richardson A, Mitchell-Heggs N. Stereotactic limbic leucotomy: neurophysiological aspects and operative technique. Br J Psychiatry 1973;123:133-40. 33. Richardson A. Stereotactic limbic leucotomy: surgical technique. Postgrad Med J 1973;48:860-64. 34. Papez JW. A proposed mechanism of emotion. Arch Neurol Psychiatry 1937;38:725-43. 35. Kelly D, Richardson A, Michell-Heggs N, Greenup J, Chen C, Hafner RJ. Stereotactic limbic leucotomy: a preliminary report on forty patients. Br J Psychiatry 1973;123:141-8. 36. Montoya A, Weiss AP, Price BH, Cassem EH, Daugherty DD, Nierenberg AA, Rauch SL, Cosgrove GR. Magnetic resonance imaging-guided stereotactic limbic leukotomy for treatment of intractable psychiatric disease. Neurosurgery 2002;50:1043-49. 37. Mitchell-Heggs N, Kelly D, Richardson A. Stereotactic limbic leucotomy – a follow-up at 16 months. Br J Psychiatry 1976;128:226-40. 38. Mindus P. Present-day indications for capsulotomy. Acta Neurochirurgica Suppl (Wien) 1993;58:29-33. 39. Talairach J, Hecaen H, David M. Lobotomie prefrontale limitee par electrocoagulation des fibres thalamofrontalis leur emergence du bras anterior de la capsule interne. In Proceedings of the Fourth Congress Neurologique Internationale, Pris, Masson, 141, 1949. 40. Meyerson BA, Mindus P. The role of anterior internal capsulotomy in psychiatric surgery. In: Lunsford L, editor. Modern stereotactic neurosurgery. Boston: Martinus Nijhoff Publishing; 1988. p. 353-64. 41. Kihlstro¨m L, Hindmarsh T, Lax I, Lippitz B, Midnus P, Lindquist C. Radiosurgical lesions in the normal human brain 17 years after gamma knife capsulotomy. Neurosurgery 1997;41:396-402. 42. Hurwitz TA, Mandat T, Forster B, Honey CR. Tract identification by novel MRI signal changes following stereotactic anterior capsulotomy. Stereotactic Funct Neurosurg 2006;84:228-35. 43. Ridout N, O’Caroll RE, Dritschel B, Christmas D, Eljamel M, Matthews K. Emotion recognition from dynamic emotional displays following anterior cingulotomy and anterior capsulotomy for chronic depression. Neuropsychologia 2007;45:1735-43.
Ablative procedures for depression
44. Lippitz BE, Mindus P, Meyerson BA, Kihlstro¨m L, Lindquist C. Lesion topography and outcome after thermocapsulotomy or gamma-knife capsulotomy for obsessive-compulsive disorder: relevance of the right hemisphere. Neurosurgery 1999;44:452-8. 45. Schlaepfer TE, Cohen MX, Frick C, Kosel M, Brodesser C, Axmacher N, You Joe A, Kreft M, Lenartz D, Sturm V. Deep brain stimulation to reward circuitry alleviates anhedonia in refractory major depression. Neuropsychopharmacology 2004; 33:368-77. 46. Schlaepfer TE, Lieb K. Deep brain stimulation for treatment of refractory depression. Lancet 2005;366:1420-22.
175
47. Benabid AL, Beazzouz A, Hoffmann D, Limousin P, Krack P, Pollak P. Long-term electrical inhibition of deep brain targets in movement disorders. Mov Disord 1998;13 Suppl 3:119-25. 48. Sachdev PS, Sachdev J. Long-term outcome of neurosurgery for the treatment of resistant depression. J Neuropsychiatry Clin Neurosci 2005;17:478-85. 49. Sachdev P. Is deep brain stimulation a form of psychosurgery? Australas Psychiatry 2007;15:97-9. 50. Sartorius A, Henn FA. Deep brain stimulation of the lateral habenula in treatment resistant major depression. Med Hypothesis 2007;69:1305-08.
2951
172 Cingulotomy for Depression and OCD G. R. Cosgrove
Introduction Cingulotomy has been performed for many years on patients with intractable major depression (MD) and obsessive compulsive disorder (OCD) but it has not been without controversy. Fulton was the first to suggest that the anterior cingulum would be an appropriate target for psychosurgical intervention and cingulotomy was initially carried out as an open procedure [1]. Foltz and White reported their experience with stereotactic cingulotomy for intractable pain and noted the best results were in those patients with concurrent anxiety and depression [2]. Ballantine subsequently demonstrated the safety and effectiveness of stereotactic cingulotomy in a large number of patients and it has been the surgical procedure of choice in North America over the last 60 years [3]. In this chapter, we will explore the neurobiological basis for cingulotomy and guidelines for the appropriate selection of surgical candidates along with details of operative technique and post-operative management. The long term surgical results and complications of modern cingulotomy will also be presented.
Neurobiological Basis for Cingulotomy In 1937, Papez published his paper entitled ‘‘A Proposed Mechanism of Emotion’’ in which he postulated that a reverberating circuit in the brain might be responsible for emotion, anxiety and memory [4]. The anatomic components of #
Springer-Verlag Berlin/Heidelberg 2009
this circuit consisted of the hypothalamus, septal area, hippocampus, mamillary bodies, anterior thalamic nuclei, cingulate gyri and their interconnections. This rudimentary limbic system was subsequently expanded to include paralimbic structures including orbito-frontal, insular and anterior temporal cortices, the amygdala and dorso-medial thalamic nuclei [5]. The hypothalamus, a central component in this system, controls autonomic function and stimulation of the hypothalamus in animals can produce autonomic, endocrine and complex motor effects that support the role of the hypothalamus in the behavioral expression of emotional states [6]. But the neural outflow from the hypothalamus can be modulated by cortical and brainstem inputs and the limbic system represents a direct conduit to the hypothalamus. Neocortical areas are also connected to the limbic system proper by paralimbic structures [7]. We now think of MD and OCD as ‘‘neural network problems,’’ as opposed to dysfunction in any particular neural structure or any single neurotransmitter. This neural network is comprised of neocortex, the limbic and paralimbic systems, the basal ganglia with ventral striatum and all their interconnections. The nucleus accumbens (NAc) is located in the ventral striatum, and is heavily connected to limbic areas of the brain [8]. NAc receives projections from orbitofrontal cortex, medial prefrontal cortex, and amygdala, all regions involved in the processing of emotion [9]. NAc also receives projections from motor control centers such as globus pallidum, and from regions involved in memory, such as hippocampus [10]. NAc sends projections to midbrain
2888
172
Cingulotomy for depression and OCD
dopaminergic neurons as well as medial prefrontal cortex, ventral pallidum, lateral hypothalamus, amygdala and Cg25 [11]. The NAc is divided into an outer shell and a central core which surrounds the anterior commissure. The core projects primarily to motor-related regions of the basal ganglia, and the shell projects primarily to limbic structures [8]. The anterior cingulate cortex (ACC) is central to this neural network and is composed of subdivisions that play critical roles in emotional processing, cognition, and motor function [12]. ACC has direct limbic and paralimbic connections, and includes Brodmann’s area (BA) 24, 32, and 25 [13]. The subgenual cingulate or area Cg25, which underlies the genu of the corpus callosum (along with the caudal aspects of BA 32 and BA 24), has projections to medial caudate nucleus, amygdala, insula, hypothalamus, midline and mediodorsal thalamic nuclei, as well as widespread projections to subcortical structures that in turn project to other cortical areas including orbitofrontal, medial prefrontal and cingulate cortex [14]. These projections suggest a role for Cg25 in the control of visceromotor function and in the circadian regulation disturbances (sleep, libido, appetite) and in the behavioral disturbances (of motivation, reward, learning, and memory) of depression. Additional evidence implicates the limbic system and its interconnections with the basal ganglia and forebrain in the pathophysiology of these psychiatric disorders. Electrical stimulation of the ACC, has been shown in humans to alter both autonomic responses and anxiety levels [15,16] and can cause movements in awake patients that resemble compulsive actions [17]. The clinical observation that many patients with Sydenhams Chorea and Tourette’s Syndrome have OCD and that 20% of patients with OCD have motor tics has prompted some authors to implicate a common pathophysiological mechanism for movement disorders and OCD [18]. The frontal-striatal-pallidothalamic connections which have been so well characterized for the control of motor function
might explain some features of OCD. Modell postulates that this network has two components; an orbitofrontal-thalamic loop mediated by the excitatory neurotransmitter glutamate and an orbitofrontal-striatal-thalamic loop mediated by various transmitters including glutamate, dopamine, seretonin, GABA and glutamate [19]. In this model, overactivity of the orbitofrontal-thalamic loop would give rise to obsessive-compulsive behavior. Neuroimaging studies also implicate a circuit composed of orbitofrontal cortex, striatum, thalamus and anterior cingulate cortex in the pathophysiology of OCD and MD. PET studies have provided perhaps the most compelling evidence for implicating orbitofrontal cortex and basal ganglia dysfunction in OCD. Abnormalities of glucose metabolism have been found in the caudate nucleus, orbitofrontal cortex and cingulum in patients with OCD on positron emission tomography (PET) [20,21]. Symptom provocation studies have shown increased metabolism in the orbital gyri, caudate and thalamus among other areas [22]. Treatment studies (pre-post) using various modalities such as clomipramine [23], selective serotonin reuptake inhibitors [24,25] and behavioral therapy [26] have demonstrated reduction in metabolism in corresponding areas with successful symptom attenuation. In some studies increased pre-treatment local metabolic rates in the orbital gyri (OG) predicted a poorer response to fluoxetine while decreased OG metabolism predicted better response to behavioral therapy [27]. Similarly, PET studies have shown reduced glucose metabolism in the lateral frontal cortex is a correlate of the depressive state in certain patients [28]. Decreased lateral prefrontal metabolism and increased metabolism in the medial prefrontal and subgenual cingulate regions have also been noted in subjects with depression [29,30]. The severity of depression has been shown to correlate with the degree of hypermetabolism in the subgenual cingulate region [31].
Cingulotomy for depression and OCD
172
Moreover, the increased subgenual cingulate metabolism normalizes in MDD patients who respond to treatment [32,33]. Also, induced sadness in non-depressed control subjects has been demonstrated to result in increased cerebral perfusion in the subgenual cingulate region (BA25) [34]. Finally, Dougherty and colleagues performed preoperative resting-state FDG-PET imaging on 13 patients who then underwent anterior cingulotomy for the treatment of MDD [35]. The authors found that the degree of preoperative hypermetabolism noted in the left subgenual prefrontal cortex and in the left thalamus correlated with greater percentage improvements on the Beck Depression Inventory (BDI) at 12 months following surgery. Various investigators have noted increased resting-state activity within anterior cingulate cortex (ACC) in imaging studies of patients with major depression [36,37]. The increased ACC activity has been noted to normalize with treatment response. Mayberg found that metabolic activity on PET in the rostral anterior cingulate (BA 24a/b) could differentiate eventual drug treatment responders from non-responders, in that responders were hypermetabolic, and nonresponders were hypometabolic compared to control subjects [38]. Hyperactivity in the resting state has also been noted in the OFC of patients with MDD during depressive episodes [39]. Although the exact neuroanatomical and neurochemical mechanisms underlying MD and OCD are still being delineated, it is likely that these psychiatric disorders reflect a final common pathway of limbic dysregulation. Contemporary neurobiological models of anxiety and affective disorders have all emphasized the fundamental role of the limbic system and its related structures in the pathophysiology of these disorders.
psychiatrist, guided by the informed input of the other members of the psychosurgical team. There must be unanimous agreement that the patient satisfies the selection criteria, that the surgery is indicated and that the requirements for informed consent are fulfilled. Cingulotomy is only considered for patients with severe, treatment refractory MD (unipolar or bipolar depression) or OCD that interferes significantly with normal functioning. Many OCD patients, because of the severity of their disorder, have co-existent major depression and they remain candidates for surgery. Schizophrenia is not currently considered an indication for cingulotomy. All conventional therapies including psychotropic medication, psychotherapy, electroconvulsive therapy (ECT), and behavior modification therapy, must have been tried without success before considering surgery. The severity of the patient’s illness can be estimated using validated clinical research instruments such as a Yale-Brown Obsessive Compulsive Scale (YBOCS) score of >20 or a Beck Depression Inventory (BDI) score >30. Disability may be reflected, by a Global Assessment of Function (GAF) score of <50. The duration of illness is not as important as its severity although symptoms should generally have been present and unremitting for several years. Contraindications to surgery include hysterical or sociopathic personalities, or other axis II pathology. Impaired cognitive function and organic brain lesions demonstrated on imaging may increase the risk of complications. Advanced age and serious medical illness can also increase the risk of perioperative complications and postoperative confusion.
Patient Selection
Surgical Technique
The appropriate selection of patients for surgery remains the primary responsibility of the
Cingulotomy was initially performed using ventriculography but this has been replaced by
2889
2890
172
Cingulotomy for depression and OCD
modern MR-guided stereotactic techniques. This allows for more accurate placement of the lesions and direct visualization of individual differences in cingulate and ventricular anatomy. Briefly, the surgery is performed under local anesthesia with intravenous sedation. A MRI compatible frame is attached to the cranial vault and T1-weighted mid-sagittal and oblique coronal scans are obtained parallel to the proposed electrode trajectory, spanning the entire anterior cingulate gyri and the frontal horns of the lateral ventricles. Target coordinates are calculated for a point in the anterior cingulate gyrus 2–2.5 cm posterior to the tip of the frontal horn, 7 mm from the midline, and 1–2 mm above the corpus callosum bilaterally. Burr holes are placed bilaterally just anterior to the coronal suture and 1.5 cm from the midline. A standard thermocoagulation electrode (Radionics, Inc., Burlington MA) with a 10-mm uninsulated tip is inserted to the target coordinates and heated to 85 C for 90 s. This results in a lesion of approximately 1 cm in vertical height and 8–10 mm in diameter in the anterior cingulum (> Figure 172-1). In recent years, we have begun placing 3 lesions simultaneously in
the anterior cingulate gyrus 7, 14 and 21 mm posterior to the frontal horn to reduce the need for re-operation (> Figure 172‐2). The procedure is performed bilaterally.
Postoperative Care The initial postoperative care for cingulotomy patients is similar to that of any patient undergoing a stereotactic procedure. The patient is observed closely for the early detection and management of complications. A postoperative MRI scan is obtained to document lesion placement. Mild oral analgesics are all that is generally required. Occasionally, the acute effects of cingulotomy in combination with high doses of psychotropic medication may cause mild drowsiness but most patients are maintained on their preoperative drug regimen. Minor symptoms of headache, low-grade fever and nausea are common after cingulotomy but generally last <24–48 h. Transient unsteady gait, dizziness, confusion, urinary retention and isolated seizures can occur but are generally mild and self-limited.
. Figure 172-1 T1 – weighted MR images of the cingulotomy lesions seen 24 h postoperatively in the sagittal (left) and axial (right) views
Cingulotomy for depression and OCD
172
. Figure 172-2 T1 – weighted MR images of a patient who has undergone an extended cingulotomy (3 contiguous lesions) seen 24 h postoperatively in the sagittal (left) and axial (right) views
Psychiatric patients require careful, longterm psychiatric management to readjust their medications (often to lower doses) and proceed with other conventional therapies like psychotherapy and ECT which are often more effective postoperatively. Cingulotomy is not considered as curative but rather as an adjunct to optimal psychiatric treatment. There is also rarely any acute benefit from cingulotomy. The improvement in mood or OCD symptoms is generally not appreciated until 3–6 months after surgery Initially, about half of the patients required a repeat cingulotomy. Reoperation and enlargement of the cingulotomy lesion was considered if there had been no worthwhile improvement after 6 months. Repeat lesions were made anterior to the initial lesion to avoid injury to the premotor area. More recently, we have performed 3 simultaneous lesions on each side at the initial surgery ablating ffi2.5 cm of the ACC in the hopes of reducing the need for re-operation. If the patient does not respond to this ‘‘extended’’ cingulotomy, we generally recommend converting the cingulotomy to a limbic leucotomy by placing lesions in the subcaudate region bilaterally.
Surgical Results Although many patients experience an immediate reduction in anxiety after cingulotomy, there is generally a delay to the onset of beneficial effect on depression and obsessive compulsive symptoms. This latency has been observed by many centers and suggests that the improvement may be related not only to interruption of neural pathways but also to reorganization of neural pathways after injury. This latency may be as long as 3–6 months and must be clearly explained to the patient and referring psychiatrist. In patients with co-existent OCD and MD, the depressive symptoms generally improve before OCD symptomatology. The results of bilateral cingulotomy in 198 patients suffering from a variety of psychiatric disorders were reported retrospectively by Ballantine et al in 1987 [40]. Outcomes were rated by referring clinicians using a subjective rating scale. With a mean follow-up of 8.6 years, 64% of patients with severe affective disorder and 56% of patients with OCD were found to have undergone worthwhile improvement. Overall, 62% of
2891
2892
172
Cingulotomy for depression and OCD
all patients operated upon had a successful outcome. No improvement or patients being worse after surgery was observed in 12% but the majority of these cases were in the schizophrenic, personality disorders or miscellaneous diagnostic categories. The great majority of the depressed patients (83%) were suicidal and 1–2% of patients committed suicide during the follow up period. However, this annual suicide rate post-cingulotomy is dramatically less than the expected average annual suicide rate for severely depressed patients [41]. Using the identical outcome measures employed by Ballantine, a more modern retrospective study of MRI-guided stereotactic cingulotomy yielded similar success rates underscoring the positive bias that occurs with subjective outcome rating scales. Those patients with MD seemed to respond better than those with OCD and approximately 40% of patients required more than one procedure [42]. A more detailed retrospective study evaluating cingulotomy in 33 patients with refractory OCD using unbiased observers has also been reported. Outcome was rated using validated clinical research tools including the MGH Visual Analog Assessment scale, Maudsley Obsessional Compulsive Inventory, YBOCS, Clinical Global Improvement (CGI) and BDI by unbiased observers. This study demonstrated that using more rigorous criteria for defining a successful outcome, at least 25–30% of patients were clear cut responders to cingulotomy and that an additional 10–15% of patients were partial responders [43]. The first prospective long term follow-up study of 18 patients who underwent cingulotomy for intractable OCD was reported in 2002 and utilized unbiased observers, a detailed preoperative evaluation and standardized outcome rating scales (YBOCS, YBOCS symptoms checklist, CGI and BDI) in all patients [44]. Follow up evaluation was carried out every 6 months by personal or telephone interview with administration of a
YBOCS, MGH Visual Analog Assessment scale, CGI, BDI, Sickness Impact Profile (SIP) and SAFTEE, a tool used in psychopharmacologic medication studies which identifies side effects and adverse outcome. A successful outcome (responder) was only acknowledged if the patient had a >35% improvement on his YBOCS and a CGI of 1 or 2. Patients who had a >35% improvement on his YBOCS or a CGI of 1 or 2 were considered possible responders. Five patients (28%) met these very conservative criteria as treatment responders and three others (17%) were considered possible responders for an overall response rate of ffi45%. Overall, the entire group improved significantly in terms of functional status and no significant adverse effects were encountered. Dougherty et al published a larger experience using the same rigorous outcome criteria in 44 patients with intractable OCD and found remarkably similar results. After an average follow up of nearly 3 years, 32% of patients were clear cut responders and 14% were partial responders for an overall response rate of ffi46% [45]. The value of these two studies is that they were the first to demonstrate in a prospective manner with adequate long term follow up, that cingulotomy is effective for OCD as measured by standard psychiatric rating scales and unbiased independent observers. In terms of MD, cingulotomy is generally even more effective. In a modern retrospective series of 15 patients with severe treatment refractory MD, 60% of patients had a >50% improvement in their BDI and were responders while 12% were considered partial responders [42]. These results were supported by a larger study of prospectively gathered data on 33 patients with intractable MD who underwent the extended triple lesion cingulotomy followed by subsequent subcaudate tractotomy if there was no improvement at 1 year. This latter approach effectively created a staged limbic leucotomy. Responders were defined by a >50%
Cingulotomy for depression and OCD
reduction in their BDI and a CGI of 1 or 2 (very much improved or much improved). Approximately 75% of patients were found to have had substantial benefit from surgery. Almost half of the patients however had to undergo a staged procedure with placement of a subcaudate tractotomy [46].
Complications While cingulotomy appears effective in a proportion of severely affected patients, it must also be proven to be a safe procedure with minimal morbidity and few side effects. In nearly 1,000 cingulotomies performed at the MGH since 1962, there have been no deaths and only two infections. Two acute subdural hematomas occurred early on in the series secondary to laceration of a cortical artery at the time of introduction of ventricular needles but only one patient suffered permanent neurologic impairment [3]. Early postoperative seizures have been seen in 1% of patients and are generally easily controlled with anticonvulsant medication. Headache and mild temperature elevation is common and temporary urinary dysfunction can occur occasionally. An independent analysis of 34 patients who underwent cingulotomy demonstrated no significant behavioral or intellectual deficits as a result of the cingulate lesions themselves. The only clear-cut neurologic deficit demonstrated was a deterioration of performance on the Taylor Complex Figure Test in patients over 40 years of age. A comparison of preoperative and postoperative Weschler IQ scores demonstrated significant gains postoperatively. This improvement was greatest in patients with chronic pain and depression but negligible in those with the diagnosis of schizophrenia [47]. The absence of any significant cognitive or intellectual change was confirmed by detailed neuropsychological evaluations in 17 patients who had undergone cingulotomy for OCD [48].
172
Discussion A variety of neurosurgical interventions have been performed for intractable MD and OCD throughout the last 60–70 years including cingulotomy, anterior capsulotomy, subcaudate tractotomy and limbic leucotomy. Cingulotomy has been the preferred intervention in the United States while the other procedures have been more prevelant in Europe. All procedures have been reported to yield favorable outcomes and it is unclear whether there is one optimal surgical technique or strategy. A direct comparison of results across centers is difficult because of diagnostic inaccuracies, non-standardized presurgical evaluation tools, center bias and varied outcome assessment scales. While the clinical superiority of any one procedure is not convincing, cingulotomy appears to be the safest of all procedures currently performed. Anterior cingulotomy using modern MRI guided stereotactic techniques is a safe and relatively effective procedure. It is most effective in the treatment of intractable MD with ffi45–75% of patients achieving substantial benefit. OCD is a more difficult psychiatric illness to treat but cingulotomy significantly improves ffi30–45% of previously intractable patients. In both groups of patients, there is typically no sign of improvement until 3–6 months after surgery which argues strongly against any kind of placebo effect. Today, all forms of ablative surgery are being replaced by deep brain stimulation (DBS) techniques using many of the same targets that have been used over the past 60–70 years. The early results of DBS for MD and OCD have been promising. Using a target in the subgenual cingulate cortex, DBS has improved ffi2/3 of patients with MD [49]. DBS of the anterior capsule/ventral striatum has also shown substantial improvement in ffi53% of patients with treatment refractory depression [50]. In patients with OCD, DBS seems to be effective in ffi60% of patients at least in short term follow-up [51]. The advantages of
2893
2894
172
Cingulotomy for depression and OCD
DBS include a reversible and adjustable ‘‘lesion’’ and the ability to change stimulation parameters over time. However, it is extremely expensive in terms of both equipment and physician time. Currently, in order to maintain benefit, the batteries have to be changed very frequently, sometimes every 9–12 months, which makes it impractical for widespread use. DBS has the advantage however of being able to be used to explore new targets within the limbic system and may allow neurosurgeons and researchers to discover new, more effective targets that might be appropriate for ablation. While controversy exists regarding the best surgical procedure to be employed, there is unanimous agreement that the presurgical evaluation be performed by committed multidisciplinary teams with expertise and experience in the surgical treatment of psychiatric illness. Accurate diagnosis based upon the DSM-III-R classification scheme is encouraged and prospective trials employing standardized clinical instruments with long term follow-up are needed. Comparisons of preoperative and postoperative functional status in addition to target psychiatric symptoms remain important parameters in characterizing outcome. All centers with experience emphasize the importance of rehabilitation postoperatively and the need for ongoing psychiatric follow-up. Cingulotomy is not a panacea and should be considered as only one aspect in the overall management of these patients. While the role of cingulotomy in the treatment of psychiatric disease has been questioned, responsible centers report good outcomes in the 45–75% range. This kind of success might be considered less than impressive, but if an experimental add-on pharmacotherapeutic agent was introduced that produced a 45–75% response rate in a group of treatment refractory OCD or MD patients, it would be embraced as a wonderful new therapy. Certainly, the patients undergoing surgery for psychiatric indications are desperately ill and their lives are severely impaired. Under these circumstances, it may be
acceptable to offer surgery in the hopes of salvaging a life, even if the intervention might risk subtle neuropsychological or personality changes.
Conclusions Cingulotomy can be helpful in selected patients with severe, disabling and treatment refractory psychiatric disease including major affective disorders and obsessive compulsive disorder. It should only be carried out by an expert multidisciplinary team with experience in these disorders. Cingulotomy should be considered as one part of an entire treatment plan and must be followed by an appropriate psychiatric rehabilitation program. Many patients are greatly improved after surgery and the complications or side effects are few. Cingulotomy remains an important therapeutic option for disabling psychiatric disease and is probably underutilized.
References 1. Fulton JE. Frontal lobotomy and affective behavior: a neurophysiological analysis. New York: WW Norton; 1951. 2. Foltz EL, White LE, Jr. Pain relief by frontal cingulotomy. J Neurosurg 1962;19:89-94. 3. Ballantine HT, Giriunas IE. Treatment of intractable psychiatric illness and chronic pain by stereotactic cingulotomy. In: Schmidek HH, Sweet WH, editors. Operative neurosurgical techniques. New York: Grune & Stratton; 1982. p. 1069-75. 4. Papez JW. A proposed mechanism of emotion. Arch Neurol Psychiatry 1937;38:725-43. 5. McLean PD. Some psychiatric implications of physiologic studies on the frontotemporal portion of limbic system. Electroenceph Clin Neurophysiol 1952;4:407-18. 6. Ranson SW. The hypothalamus: its significance for visceral innervation and emotional expression. Trans Coll Physicians Phila [Series IV] 1934;2:222-42. 7. Mesulam M-M. Patterns in behavioral neuroanatomy: association areas, the limbic system, and hemispheric specialization. In: Mesulam M-M, editor. Principles of behavioral neurology. Philadelphia, PA: F.A. Davis Co.; 1985. p. 1-70.
Cingulotomy for depression and OCD
8. Heimer L, Zahm DS, Churchill L, et al. Specificity in the projection patterns of accumbal core and shell in the rat. Neuroscience 1991;41:89-125. 9. Haber SN, Kim KS, Mailly P, et al. Reward-related cortical inputs define a large striatal region in primates that interface with associative cortical connections, providing a substrate for incentive-based learning. J Neurosci 2006;26:8368-76. 10. Friedman DP, Aggleton JP, Saunders RC. Comparison of hippocampal, amygdala, and perirhinal projections to the nucleus accumbens: combined anterograde and retrograde tracing study in the Macaque brain. J Comp Neurol 2002;450:345-65. 11. Nicola SM. The nucleus accumbens as part of a basal ganglia action selection circuit. Psychopharmacology (Berl) 2007;191:521-50. 12. Bush G, Luu P, Posner MI. Cognitive and emotional influences in anterior cingulate cortex. Trends Cogn Sci 2000;4:215-22. 13. Haber SN, Kim KS, Mailly P, et al. Reward-related cortical inputs define a large striatal region in primates that interface with associative cortical connections, providing a substrate for incentive-based learning. J Neurosci 2006;26:8368-76. 14. Freedman LJ, Insel TR, Smith Y. Subcortical projections of area 25 (subgenual cortex) of the macaque monkey. J Comp Neurol 2000;421:172-88. 15. Kelly D. The limbic system, sex and emotions. In: Anxiety and emotions: physiologic basis and treatment. Springield, IL: Charles C. Thomas publisher; 1980. p. 197-300. 16. Laitinen LV. Emotional responses to subcortical electrical stimulation in psychiatric patients. Clin Neurol Neurosurg 1979;81:148-57. 17. Talairach J, Bancaud J, Geier S. The cingulate gyrus and human behaviour. Electroencephalogr Clin Neurophysiol 1973;34:45-52. 18. Green RC, Pitman RK. Tourette syndrome and obsessivecompulsive disorder. In: Jenike MA, Baer L, Minichiello WE, editors. Obsessive-compulsive disorders: theory and management. Littleton, MA: PSG Publishing Co; 1986. p. 147-64. 19. Modell J, Mountz J, Curtis G, et al. Neurophysiologic dysfunction in basal ganglia/limbic striatal and thalamocortical circuits as a pathogenetic mechanism of obsessive compulsive disorder. J Neuropsychiatry 1989;1:27-36. 20. Weilburg JB, Mesulam MM, Weintraub S, et al. Focal striatal abnormalities in a patient with obsessive compulsive disorder. Arch Neurol 1989;46:233-6. 21. Luxenberg JS, Swedo SE, Flament MF, et al. Neuroanatomical abnormalities in obsessive compulsive disorder detected with a quantitative x-ray computed tomography. Am J Psychiatry 1988;145:1089-93. 22. Rauch SL, Jenike MA, Alpert NM, et al. Regional cerebral blood flow measured during symptom provocation in obsessive-compulsive disorder using 15-O-labelled CO2
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
172
and positron emission tomography. Arch Gen Psychiatry 1994;51:62-70. Swedo SE, Pietrii P, Leonard HL, et al. Cerebral glucose metabolism in childhood-onset obsessive-compulsive disorder. Revisualization during pharmacotherapy. Arch Gen Psychiatry 1992;49(9):690-4. Baxter LR, Jr, Schwartz JM, Bergman KS, et al. Caudate glucose metabolic rate changes with both drug and behavior therapy for obsessive compulsive disorder. Arch Gen Psychiatry 1992;49:681-9. Saxena S, Brody AL, Maidment KM, et al. Localized orbitofrontal and subcortical metabolic changes and predictors of response to paroxetine treatment in obsessive-compulsive disorder. Neuropsychopharmacol 1999;21:683-93. Schwartz JM, Stoessel PW, Baxter LR, Jr, Martin KM, Phelps ME. Systematic changes in cerebral glucose metabolic rate after successful behavior modification treatment of obsessive-compulsive disorder. Arch Gen Psychiatry 1996;53(2):109-13. Baxter LR, Schwartz JM, Guze BH, et al. Neuroimaging in obsessive compulsive disorder: Seeking the mediating neuroanatomy. In: Jenike MA, Baer L, Minichiello WE, editors. Obsessive compulsive disorder: theory and management. 2nd ed. Chicago, IL: Year Book Medical Publishers; 1990. p. 167-88. Baxter LR, Schwartz JM, Phelps ME, et al. Reduction of prefrontal glucose metabolism common to three types of depression. Arch Gen Psychiatry 1989;46:243-50. Kennedy SH, Evans KR, Kruger S, et al. Changes in regional brain glucose metabolism measured with positron emission tomography after paroxetine treatment of major depression. Am J Psychiatry 2000;158:899-905. Mayberg HS. Positron emission tomography imaging in depression: a neural systems perspective. Neuroimaging Clin N Am 2003;13:805-15. Osuch EA, Ketter TA, Kimbrell TA, et al. Regional cerebral metabolism associated with anxiety symptoms in affective disorder patients. Biol Psychiatry 2000;48:1020-3. Mayberg HS, Brannan SK, Tekell JL, et al. Regional metabolic effects of fluoxetine in major depression: serial changes and relationship to clinical response. Biol Psychiatry 2000;48:830-43. Mayberg HS, Brannan SK, Mahurin RK, et al. Cingulate function in depression: a potential predictor of treatment response. Neuroreport 1997;8:1057-61. Liotti M, Mayberg HS, McGinnis S, et al. Unmasking disease-specific cerebral blood flow abnormalities: mood challenge in patients with remitted unipolar depression. Am J Psychiatry 2002;159:1830-40. Dougherty DD, Weiss AP, Cosgrove GR, et al. Cerebral metabolic correlates as potential predictors of response to anterior cingulotomy for treatment of major depression. J Neurosurg 2003;99:1010-17.
2895
2896
172
Cingulotomy for depression and OCD
36. Brody AL, Saxena S, Silverman DH, et al. Brain metabolic changes in major depressive disorder from pre- to post-treatment with paroxetine. Psychiatry Res 1999;91:127-39. 37. Pizzagalli D, Pascual-Marqui RD, Nitschke JB, et al. Anterior cingulate activity as a predictor of degree of treatment response in major depression: evidence from brain electrical tomography analysis. Am J Psychiatry 2001;158:405-15. 38. Mayberg HS, Brannan SK, Mahurin RK, et al. Cingulate function in depression: a potential predictor of treatment response. Neuroreport 1997;8:1057-61. 39. Drevets WC. Orbitofrontal cortex function and structure in depression. Ann NY Acad Sci 2007;1121:499-527. 40. Ballantine HT, Bouckoms AJ, Thomas EK, et al. Treatment of psychiatric illness by stereotactic cingulotomy. Biol Psychiatry 1987;22:807-19. 41. Pokorney AD. Prediction of suicide in psychiatric patients. Arch Gen Psychiatry 1983;40:249-57. 42. Spangler W, Cosgrove GR, Ballantine HT, Cassem N, Rauch SR, Neirenberg A, et al. MR-guided stereotactic cingulotomy for intractable psychiatric disease. Neurosurgery 1996;38:1071-8. 43. Jenike MA, Baer L, Ballantine HT, et al. Cingulotomy for refractory obsessive compulsive disorder. A long term follow-up of 33 patients. Arch Gen Psychiatry 1991;48:548-55. 44. Baer L, Rauch SL, Ballantine HT, Martuza R, Cosgrove GR, Cassem E, et al. Cingulotomy for intractable obsessive-compulsive disorder: a prospective long-term follow-up of 18 patients. Arch Gen Psych 1995;52:384-92.
45. Dougherty DD, Baer L, Cosgrove GR, Cassem EH, Price BH, Nierenberg AA, et al. Update on cingulotomy for intractable obsessive-compulsive disorder: prospective long-term follow-up of 44 patients. Am J Psychiatry 2002;159:269-75. 46. Shields DC, Asaad W, Eskandar EN, Jain FA, Cosgrove GR, Flaherty AW, et al. Prospective assessment of stereotactic ablative surgery for intractable major depression. Biol Psych 2008;64(6):449-54. 47. Corkin S, Twitchell TE, Sullivan EV. Safety and efficacy of cingulotomy for pain and psychiatric disorders. In: Hitchcock ER, Ballantine HT, Myerson BA, editors. Modern concepts in psychiatric surgery. Amsterdam: Elsevier; 1979. p. 253-72. 48. Jung HH, Kim CH, Chang JH, Park YG, Chung SS, Chang JW. Bilateral anterior cingulotomy for refractory obsessive compulsive disorder: Long term follow-up results. Stereotact Funct Neurosurg 2006;84:184-9. 49. Lozano AM, Mayberg HS, Giacobbe P, Hamani C, Craddock RC, Kennedy SH. Subcallosal cingulate gyrus deep brain stimulation for treatment-resistant depression. Biol Psychiatry 2008;64(6):461-7. 50. Malone DA, Jr, Dougherty DD, Rezai AR, Carpenter LL, Friehs GM, Eskandar EN, et al. Deep brain stimulation of the ventral capsule/ventral striatum for treatment-resistant depression. Biol Psychiatry 2008. 51. Greenberg BD, Gabriels LA, Malone DA, Jr, Rezai AR, Friehs GM, Okun MS, et al. Deep brain stimulation of the ventral internal capsule/ventral striatum for obsessive-compulsive disorder: worldwide experience. Mol Psychiatry 2008.
173 DBS for OCD L. Gabrie¨ls . P. Cosyns . K. van Kuyck . B. Nuttin
Obsessive-Compulsive Disorder Clinical Aspects of Obsessive Compulsive Disorder Obsessive-compulsive disorder (OCD) affects approximately 2% of the general population [1], with remarkably consistent lifetime prevalence in Western countries [2]. The cardinal symptoms of OCD are intrusive thoughts (obsessions) and/or repetitive behaviors (compulsions) that persist against the patient’s attempts to eliminate them. Common themes are checking, washing and cleaning, excessive need for order and symmetry, unwanted aggressive thoughts, unwanted sexual thoughts, counting, the need to ask or confess, and hoarding. The obsessions and compulsions are accompanied by overwhelming anxiety and are distressing and time-consuming. They lead to serious impairment in occupational, scholastic, and/or social functioning. Patients go to great lengths to avoid objects or situations that provoke obsessions or compulsions and severely restrict general functioning. OCD symptoms are both egodystonic and associated with relatively intact reality testing: the patient recognizes the obsessive thoughts as a product of his own mind. Patients with OCD neither want nor enjoy the obsessive thoughts or time-consuming compulsions. OCD patients exhibit heightened arousal, over-focused attention, and chronic vigilance concerning salient stimuli that trigger obsessive thoughts or compulsive actions. The embarrassment and shame causes them to hide their symptoms even from their close relatives and often deter them from seeking professional help.
#
Springer-Verlag Berlin/Heidelberg 2009
Patients with OCD display several cognitive distortions and biases: they misinterpret their intrusive thoughts as being overwhelmingly important and significant, revealing, and extremely threatening or even catastrophic. They are convinced that they should be fully able to exert complete control over their thoughts. They regard their thoughts as being psychologically equivalent to the corresponding action (thought–action fusion) and believe that having intrusive thoughts about misfortunes actually increase the risk that these misfortunes will occur [3]. Patients with OCD tend to think that they are personally responsible for events over which they have no control at all and overestimate the consequences of being responsible for a misfortune. Another specific cognitive characteristic in OCD is chronic doubt, with feelings of uncertainty regarding their own behavior, resulting in repetitive checking [4]. Comorbid disorders are common and include major depression, personality disorders, and other anxiety disorders, including panic and social phobia. This comorbidity often complicates and aggravates the clinical picture. In some studies, up to 60–90% of participants received one or more additional psychiatric diagnoses [5]. Comorbidity of OCD with depression is considerable; up to 67% of patients with primary OCD have a lifetime history positive for major depressive disorder [6]. Most OCD patients view their depressive symptoms as occurring secondary to the demoralization and hopelessness accompanying their OCD symptoms. Research suggests that when OCD improves longitudinally with effective treatment, depressive symptoms disappear as well [7].
2898
173
DBS for OCD
Quality of Life in OCD Clinical experience with patients suggests that OCD affects quality of life (QoL) in many ways (e.g., distraction from work, disturbance in relationships and recreational activities by fear, worry and obsession). The bizarre and exaggerated aspects of the obsessive thoughts and compulsive rituals result in a deep sense of shame and may lead to social isolation and depression. Available data support the substantial adverse effect of OCD on health-related QoL of both patients and their families. More than half of OCD patients reports moderate or severe interference with socializing, family relationships, and ability to study, while 30% report moderate to severe interference with ability to work [8]. Among patients who were previously employed, approximately 40% is temporally unable to work because of OCD symptoms, 38% for more than 1 year. The average career achievement level remains below their educational attainment. About 75–90% report a decrease in self-esteem and 50% reported having frequent thoughts about suicide. The more severe the subject’s OCD, the poorer health-related QoL in the domain of social functioning. OCD diminishes the quality of family relationships with lower rates of marriage and more divorces than non-OCD individuals [9,10]. Nearly half of married OCD sufferers reports marital maladjustment or dissatisfaction [11]. Individuals with OCD often force family members to participate in compulsive rituals. They may demand their children or spouse to avoid exposure to ‘‘contaminated’’ objects or areas, enforce them to wash for hours or ask for help in repetitive checking routines. Close peers are asked over and over again to provide reassurance. Failure to comply with the OCD sufferers’ demands often incites angry outbursts and verbal aggression [12–14]. Caregivers of OCD patients experience a higher degree of burden in spouse-related areas such as poor support from spouse in family
responsibilities, inadequate satisfaction of emotional and sexual needs, and deteriorated marital relationship than key relatives of schizophrenia patients [15].
Neurobiological Model of OCD Neuroimaging research has helped to provide a progressively sound basis for constructing neurobiological models of OCD, with a focus on cortico– striato–thalamo–cortical circuitry (CSTC). There is a convergence of evidence implicating dysfunction in the CSTC loops involving orbitofrontal cortex (OFC), anterior cingulate cortex (ACC) and basal ganglia as central to the pathophysiology of OCD [16–21]. In general, the various cortical regions send their specific point-to-point excitatory projections to the striatum consisting of the putamen, caudate nucleus, and nucleus accumbens. Two distinct routes are conceptualized from the striatum to the thalamus; the so-called ‘‘direct’’ and ‘‘indirect’’ pathways. The direct pathway projects from the cortex to the striatum to the internal segment of the globus pallidus and substantia nigra, pars reticulata to the thalamus, and then back to the cortex. The indirect pathway is similar from the cortex to the striatum, but then projects to the external segment of the globus pallidus to the subthalamic nucleus, before returning to the internal segment of the globus pallidus/substantia nigra, joining the direct pathway to the thalamus and projecting back to the cortex. Taking the nature of the predominant neurotransmitters of these direct and indirect routes into account, impulses transmitted via the direct pathway disinhibit the thalamus, presumably resulting in a release of behaviors as necessary for an adaptive function. Activity in the indirect pathway inhibits the thalamus, resulting in the cessation of ongoing behavioral routine. The prevailing theory on OCD suggests that a hitherto unknown primary striatal pathologic process underlies a relative imbalance favoring the direct over indirect pathways within these circuits.
DBS for OCD
The resulting failing striatothalamic inhibition leads to hyperactivity within the OFC and ACC, the caudate nucleus, and the thalamus. Overactivation of this direct pathway has also been linked by some researchers to the development of the abnormal and intrusive thoughts and behaviors that are associated with OCD. Most of the studies using positron emission tomography (PET) consistently revealed increased metabolic rates in the OFC, ACC, and subcortical areas such as the caudate nucleus both in a resting state [22] and during periods of provoked OCD symptoms [23]. Such studies are confounded by nonspecific anxiety since symptom provocation studies in specific phobias [24] and posttraumatic stress disorder [25] also have found activation of ACC and other paralimbic areas. Increased metabolism of the caudate nucleus and the anterolateral OFC seem to be more specific to the symptomatic state in OCD. Neuroimaging studies before and after treatment with an SSRI as well as with behavioral therapy demonstrate a reduction of the activity in OFC, ACC, and caudate nucleus after successful treatment of OCD [26,27]. Parietocerebellar involvement in OCD is present as well [28].
Treatment Options in OCD Serotonergic antidepressants and cognitive behavioral therapy (CBT) represent the first-line treatment for OCD [29]. CBT holds a central place in psychotherapeutic treatment of OCD, especially the technique of exposure and response prevention (ERP) [30]. After drawing up a systematic inventory of individual symptoms and environmental triggers, the patient engages in deliberate controlled exposure to the triggers (either directly or by imagination). At the same time he is strongly encouraged to refrain from ritualizing, with support and structure provided by the therapist, and possibly by others whom the patient recruits for assistance.
173
A recent compilation of outcome studies indicated that an average of 76% who completed the full course of ERP treatment showed clinically significant relief. Pharmacological interventions with selective serotonin reuptake inhibitors (SSRIs) are the pharmacologic treatment of choice [31]. A single SSRI trial generally provides clinically meaningful relief for 40–60% of OCD patients [32]. Medications, however, rarely bring about remission. Response in OCD research is generally defined as a reduction of 25–35% in OCD symptoms. All SSRIs seem to be equally effective. Clomipramine, a tricyclic antidepressant, equals the SSRIs in therapeutic effect and remains the second-line agent due to the somewhat more unfavorable side effect profile [33]. Patients with an inadequate response to serotonergic monotherapy can benefit from medication augmentation strategies. Augmentation of SSRIs with atypical antipsychotics (olanzapine, risperidone, and quetiapine) demonstrated efficacy in further reduction of OCD symptoms [34–36]. As a rule, treatment adherence is necessary for lasting symptom relief, but even when effective, side effects substantially hamper compliance. Up to half of the patients who benefit from an adequate trial of SSRI abort the medication within 2 years of treatment due to sexual dysfunction, weight gain, and sedation [37]. Augmentation strategies enhance the therapeutic effect but the burden of the side effects increases as well [38]. Notwithstanding the important advances in the efficacy, safety, and tolerability of treatments for OCD made over the last decades, some OCD patients show persistent disabling symptoms in spite of combined pharmacological and CBT treatment. Up to 7.1% of the patients remain refractory and run a chronic deteriorating course of OCD despite all available treatment [7]. There is little evidence for spontaneous remission in severe, intractable, and longstanding OCD [39]. In a follow-up of patients who are eligible for intervention but never undergo surgery for
2899
2900
173
DBS for OCD
different reasons, their condition remained the same and some of them eventually committed suicide [40].
Stereotactic Procedures for OCD For some of the treatment-refractory OCD patients who, despite all the currently available pharmacological and behavioral treatment options, continue to have severe incapacitating symptoms and are extremely disabled, a stereotactic neurosurgical intervention has proved to be an option. The surgical procedures most often used for treating treatment-resistant OCD are capsulotomy, cingulotomy, limbic leucotomy, and subcaudate tractotomy.
Subcaudate Tractotomy Geoffrey Knight introduced subcaudate tractotomy in Great Britain as one of the first attempts to limit adverse effects by restricting lesion size [41]. It has been used extensively in the United Kingdom since 1963 as a treatment for major mood disorders, OCD, severe chronic anxiety states, and a variety of other psychiatric disorders. Clinical improvement is reported in approximately 50% of OCD patients after subcaudate tractotomy [42,43]. Complications seen in a series of 208 patients with depression and OCD with mean follow-up of 2.5 years included transient postoperative confusion (10%), postoperative seizures (2%), and mild undesirable personality traits (7%). Transient disinhibition syndromes were common. Recently a new, frameless method for subcaudate tractotomy procedure with promising initial results for intractable OCD was introduced [44].
Anterior Cingulotomy Ballantine and colleagues from Massachusetts General Hospital demonstrated the safety of
anterior cingulotomy. While the first cingulotomies were performed for intractable pain, the best results were achieved in patients with comorbid anxiety or depressive conditions. Anterior cingulotomy targets the ACC and the fibers of the cingulum. The anxiety component is mediated through the Papez circuit, which includes the cingulum bundle, the target of cingulotomy [45]. Minor symptoms of headache, low-grade fever, and nausea are common after cingulotomy but generally last <24–48 h. Transient unsteady gait, dizziness, confusion, memory dysfunction, urinary retention, and isolated seizures can occur and may last up to several weeks. Insomnia and weight changes have been observed. Significant behavioral or cognitive decline has not been reported after cingulotomy [46]. The incidence of seizure is from 1 to 5%, most often occurring in patients with a preexisting seizure history. There is generally a delay in onset of any beneficial effect following anterior cingulotomy, with the latency being as long as 3–6 months. Studies mostly report on small patient series and yield 25–50% success rates [47–52].
Limbic Leucotomy Kelly et al. [53] introduced limbic leucotomy in England in 1973 as a combination of anterior cingulotomy and subcaudate tractotomy. He reasoned that two lesions might lead to a better result than either lesion alone. Indications for this procedure have included OCD, chronic anxiety states, and major depression, along with a variety of other psychiatric diagnoses. Kelly’s group has published several reports [53,54], including an initial series of 66 patients studied prospectively with a mean follow-up of 16 months. Postoperative improvement followed a fluctuating course with progressive reduction of symptoms over the first postoperative year. They found significant improvement in 89% of patients with OCD. In a separate report regarding 49 OCD patients, they noted that 84% were
DBS for OCD
improved at 20 months follow-up. Complications include transient confusion and urinary incontinence in the early postoperative period, persistent complaints of lethargy (12%), mild personality changes (7%), and one case of severe permanent memory loss due to inaccurate lesion placement. Significant cognitive impairment was not observed, and standard measurements of intelligence showed slight improvement postoperatively [55].
Anterior Capsulotomy Thermocapsulotomy Although Talairach et al. first described anterior capsulotomy in France [56], it was Leksell [57] who popularized the procedure in Sweden. Anterior capsulotomy interrupts frontothalamic connections where they converge in the anterior limb of the internal capsule, between the head of the caudate nucleus and putamen. Clinical indications for capsulotomy initially included schizophrenia, depression, chronic anxiety states, and OCD, but schizophrenia is not longer considered an indication. In a large series of capsulotomies, 64% of patients were judged to have a satisfactory result [58]. In a recent open series, 52.9% of the patients showed a 33% decrease in symptoms, 29.4% of the cases showed a 50% decrease, and 17% showed a 66% decrease in symptoms as measured on the Yale–Brown obsessive compulsive scale (YBOCS), No cognitive deficit was disclosed by neuropsychological screening tests [59]. Adverse effects include transient headache or urinary incontinence and postoperative confusion. Common complications include transient confusion (86%), nocturnal incontinence lasting several days (27%), and persistent fatigue (32%). Rare complications (about 1%) are intracranial hemorrhage, seizures, and transient hallucinations. Weight gain is common, with an overall mean weight gain of 10% [60]. An MRI study correlated the degree of symptomatic improvement with
173
lesion size [61]. PET scans after anterior capsulotomy show significant decreases in glucose utilization in the orbital gyrus and caudate nucleus [62].
Gamma Knife Leksell and coworkers [57] also developed a radiosurgical or gamma knife capsulotomy. The lesions are made without craniotomy or shaving using a radiosurgical device known as the gamma knife. The relative advantages and disadvantages of this noninvasive stereotactic radiosurgical approach remain undetermined. Recovery from gamma knife capsulotomy may be swifter and cause less discomfort, but adverse effects from radiation necrosis may be delayed for up to 8–12 months. Follow up data are scarce, but MRI assessment of gamma knife lesions found that lesion size and location were more variable than the thermocapsulotomy lesions [63]. In an open prospective study on 15 OCD patients, a small bilateral gamma knife lesion with a 4-mm collimator was produced in the part of the anterior capsule that contained the maximal number of orbitomedial frontal thalamic fibers [64]. Only one patient was improved. Thirteen patients received a second gamma knife lesion immediately ventral to the initial lesion, with improvement in an additional 40% of patients. In another series of the same authors, 23 OCD patients received two gamma knife lesions at each side on the same day with improvement in 40%. Delayed transient local edema was seen in 15% of the patients. No adverse effect on personality measures was observed.
Outcome Across Contemporary Stereotactic Neurosurgical Procedures for OCD Recent data suggest that between 30 and 40% of OCD patients experience significant symptom relief and a further proportion shows some improvement after stereotactic neurosurgery [65].
2901
2902
173
DBS for OCD
Although the target symptoms, compulsions and anxiety, are successfully reduced, side effects remain a major concern. These include weight gain, memory problems, attentional slowing, lower performance intelligence quotient, loss of initiative/ psychasthenia, diminution of inhibition, elevated mood occasionally with an overshoot toward carelessness, emotional shallowness, and in some cases severe alcoholism, aggressive tendencies, poor impulse control, sexual assaults, and even suicide [66–72]. Frontal syndromes, confusion, or subtle cognitive deficits are influenced by lesion size as well as choice of surgical approach, and they are minimal in comparison with past freehand procedures. Notably, overall cognitive function as indicated by the standard intelligence quotient is generally enhanced, which is attributed to the overriding beneficial effects of symptomatic improvement.
Ethical Considerations The history of psychosurgery in the early twentieth century must incite us to prudence when we consider surgical intervention in psychiatric patients. In fact, only after strong ethical consideration we can decide to perform neurosurgery for psychiatric disorders. According to the World Health Organization [73] it is a patient’s right to be offered a treatment that can alleviate suffering and permit improvement in QoL. Despite adherence to therapeutic guidelines and conscientious compliance, some OCD patients remain refractory to conventional treatment with medication and empirically proven psychotherapies; they are severely incapacitated and have a very low QoL. Approximately 10% of all OCD patients show an unrelenting downward course despite available treatments [74,75]. Some of these patients can be helped by neurosurgical treatment and it seems unethical to refrain from helping them, knowing that a possible treatment exists. Curing or relieving uncontrolled
pathological obsessions, compulsions, anxiety and mood through neurosurgical procedure is clearly an important therapeutic achievement. Nevertheless, the permanent nature of the damage to the brain circuits and adverse psychological effects of this damage cannot be ignored. It is precisely for this reason that psychosurgery remains an intervention of last resort. In order to demonstrate the efficacy of neurosurgery for psychiatric disorders, beyond dispute it is argued that double blind controlled experiments should be done using sham operations. It is difficult to see how such experimental procedures involving the use of ‘‘placebo operations’’ could be ethically and acceptably undertaken [76]. Much of the discussions on stereotactic neurosurgery as a last resort treatment option are based upon unwillingness to view psychiatric illness in the same way as physical illness [77]. But increasingly the line between mind and brain becomes blurred and the mind–brain dichotomy is less tenable than the ethical demarcation to justify treatment for one disorder and prohibiting it for another one [78]. The argument that our knowledge of psychopathological processes underlying OCD is still incomplete and thus neurosurgical treatment should be delayed or not undertaken at all does not hold. After all, this argument counts for any intervention aimed at reducing the severity of OCD and nobody questions the use of evidence-based pharmacological treatment of OCD. Development of electrical brain stimulation opens a new avenue for research and treatment in psychiatric disorders. A major advantage of stimulation compared to conventional ablative neurosurgery is that it is reversible. The implantation of electrodes in the brain does not significantly damage brain tissue and the stimulation itself can be modified or discontinued in the event of side effects. Electrical brain stimulation enables double blind research as the electrical current may be switched on and off in a blinded fashion.
DBS for OCD
An important issue in neurosurgery for psychiatric disorders is the ability to give informed consent. For several reasons, a careful psychological evaluation of the patient must be part of the selection of candidates for surgery. A family member or other person who knows the patient well should be part of the consent process, together with the patient. With severe OCD, consent appears in the same light as for any other medical discipline. Although some cognitive distortions and biases are present, these patients are able to exercise judgment and are fully conscious of their sufferings. They often initiate the plea for intervention of themselves and the sometimes very strong and demanding claim is in fact closer to desire than to a statement of consent stricto sensu. It is the paradoxical ease of securing consent that could become dangerous in ethical terms, so that some degree of control is required. The patient must fully understand that in the initial phases of development of electrical brain stimulation, research is an important component; since it is not yet an established therapy for treatment refractory OCD, it must be given the status of a therapeutic innovation.
Deep Brain Stimulation With deep brain stimulation (DBS) a nondestructive and reversible technique for modulation of neuronal function appeared that does not require the creation of a permanent lesion. This neuromodulation technique involves implantation of electrode-leads in a specific subcortical target location of the brain. Extension wires are tunneled under the skin and connect the leads to implanted pulse generators (IPGs). DBS instigates no destruction of brain tissue and does not preclude other newer treatments should they become available. Moreover, the patient’s autonomy is guaranteed. If he wants to continue this treatment, or if severe side effects should appear that contraindicate further
173
stimulation, DBS can be switched off or even the electrodes can be removed. DBS is now the standard of therapy for medically refractory Parkinson’s disease (PD), essential tremor and intentional tremor in multiple sclerosis. Investigational uses for DBS further include epilepsy [79,80], chronic pain [81], dystonia [82,83], cluster headache [84,85], epilepsy [86], and persistent vegetative states [87]. The positioning of the chronic DBS electrode or the lesioning probe in PD is in an exactly the same area. In a randomized prospective study clinicians confirmed the inherent advantage of DBS over their lesioning counterparts [88]. DBS had similar therapeutic benefits, fewer side effects, and was superior in overall improvement of daily functioning. The exact mechanisms of neurostimulation are unknown. There are several prevailing theories explaining why electrical stimulation is effective in alleviating symptoms in various neural disorders. One theory suggests that stimulation acts like a reversible ablative lesion, inactivating cells either by a depolarization blockade or by release of inhibitory neurotransmitters. Electrical stimulation could also activate cells or axons by depolarization, directly influencing activity in a neural circuit [89]. A third possibility involves the tonic influence of electrical stimulation on the resting potentials of target neurons. Such neurons, according to intrinsic voltage gate properties, would begin to fire at different frequencies than when they are free of stimulation influence. This in turn would alter the activity of the neural circuitry involving these targets. The mechanism of action is likely to be complex; associating cellfiring inhibition, neurotransmitter depletion, jamming and excitation of inhibitory pathways that lead to functional inhibition, mimicking the effects of lesioning of the stimulated structures. High-frequency stimulation of the subthalamic nucleus induces neuroprotection in animal models but has not yet been demonstrated in human patients suffering from Parkinson’s disease [90].
2903
2904
173
DBS for OCD
DBS for OCD The development of DBS as an effective treatment in PD incited to consider research into stimulation in the anterior limbs of the internal capsules as a therapeutic option in treatment refractory OCD. In the event patients would not benefit from DBS, removing the electrodes and eventually performing a capsulotomy remained a therapeutic option. The anatomical extent of stimulation is adjustable because each implanted lead has four independently programmable electrode contact sites. Depending on the placement of the lead and the contacts used, a variable number of afferent and efferent fibers are influenced. The amplitude, frequency, and pulse width are programmable within safety limits restricting the density of the electrical charge induced. Such large ‘‘parameter space’’ provides flexibility and the opportunity to optimize a therapeutic response and to minimize adverse effects. The rationale for the selection of the anterior limbs of the internal capsules as targets for electrical stimulation in OCD was parallel to the choice of target for treatment in tremor and PD, where the identification of surgical lesions with therapeutic benefits was followed by DBS applied with high frequencies to the same structures. The anterior limb of the internal capsule is a large and complex array of fibre tracts. It contains the anterior thalamic radiation (or peduncle) as well as the prefrontal corticopontine tract and fibers connecting the caudate nucleus with the putamen [91]. The anterior thalamic peduncle forms a reciprocal connection between the dorsomedial thalamic nuclei and the dorsolateral and medial prefrontal cortex, and between the anterior thalamic nuclei and the cingulate gyrus. Other important anatomical structures immediately adjacent to the internal capsule [92] that could be influenced by selection of the optimal parameters are the stria terminalis and the bed nucleus of the stria terminalis, both part of the extended amygdala concept, and the nucleus accumbens.
The ventral striatopallidal complex is believed to be the key structure linking motivation and action, at the interface of the limbic system with motor mechanisms. The cortico-subcortical loop through the ventral striatopallidal complex and the mediodorsal thalamus parallels the more classic loop through the main dorsal part of the basal ganglia and the ventrolateral thalamus. The input of the ventral striatopallidal complex is largely related to the olfactory and limbic systems, while the rest of the striatopallidal system is more closely linked to the sensorimotor and association cortices. The dorsal striatopallidal system is thought to play a prominent role in initiating motor behavior that stems from cognitive activities, while the ventral striatopallidal system is involved in initiating movements in response to stimuli that are motivationally or emotionally significant [93]. The input from the limbic cortices to the striatum suggests a role for striatum and basal ganglia in non-sensorimotor functions associated with motivational, emotional and cognitive types of behaviors [94,95]. Within the cortico-subcortical loops, the neural activity in a direct pathway that disinhibits the thalamus and leads to the initiation of a motor action is normally in dynamic balance with the neural activity in an indirect pathway that inhibits the thalamus, resulting in the discontinuation of the action after the goal has been accomplished. A model of OCD, with a disequilibrium between the direct and indirect pathways and predominance of the direct pathway, is proposed as a model of how the brain mediates the expression of OCD symptoms [96]. In most mammals, the extended amygdala presents themselves as a ring of neurons encircling the internal capsule and basal ganglia. Remnants from an embryonic continuous structure connecting the bed nucleus of the stria terminalis and central medial amygdala form interrupted cell columns within the stria terminalis as it takes a semicircular detour above and behind the internal capsule and thalamus [97]. The extended amygdala are directly continuous with
DBS for OCD
the caudomedial shell of the nucleus accumbens and together they establish specific neuronal circuits with the medial prefrontal-OFC [98,99]. They project significantly to many areas in the hypothalamus and the brainstem, including the ventrolateral part of the periaqueductal gray, which has received considerable attention as a prominent staging area for the coordination of somatomotor and autonomic responses in affective–defensive behavior [100]. The amygdala and the bed nucleus of the stria terminalis are critically involved in the mediation of stimulusspecific fear and anxiety and both structures receive highly processed sensory information from the basolateral nucleus of the amygdala and hence are in the position to respond to emotionally significant stimuli [101].
Experimental Design We chose for a prospective, explorative single case (N = 1) study. It allows combining clinical practice with prospective research on treatment efficacy using scientific methods. In this withinpatient design the patient serves as his own control. An independent variable (i.e., capsular stimulation) is manipulated and systematic changes in behavioral measures and assessment tools are subsequently evaluated. The use of capsular stimulation in our patients has primarily a therapeutic aim: it is implemented to improve the incapacitating symptoms and substantial suffering of severe, treatment refractory OCD patients. Nevertheless, we intend to gather as much information as possible from each patient to optimize the procedure for the next subjects.
Patient Selection All patients suffer from long-standing, severe, highly disabling OCD and fulfil the criteria for OCD (300.30) according to the Diagnostic and Statistical
173
Manual of Psychiatric Disorders, 4th edition (DSM-IV) [102]. Hospital Ethics Review Boards of the University Hospital of Antwerpen and the University hospitals of Leuven approved the study protocol. A multidisciplinary ‘‘committee for neurosurgical treatment for psychiatric disorders’’ advised on the suitability of surgical treatment according to strict criteria [103,104]. Patient selection and inclusion and exclusion criteria have been described elsewhere [105–107]. Inclusion criteria required a diagnosis by Structural Clinical Interview for DSM-IV (SCID-IV) [108] of OCD with a Yale-Brown Obsessive-Compulsive Scale (YBOCS) score [109,110] of at least 30/40 and a Global Assessment of Functioning (GAF) score [111] of 45 or less. Patients were treatment resistant to adequate trials of at least three SSRIs and clomipramine, augmentation strategies with antipsychotics, and CBT. Patients were between 18 and 60 years. The patient and a close family member were repeatedly and fully informed on both procedures (capsulotomy and capsular stimulation). Exclusion criteria were a current or past psychotic disorder, any clinically significant disorder or medical illness affecting brain function or structure (other than motor tics or Gilles de la Tourette syndrome), or current or unstably remitted substance abuse. Baseline observations describe the individual’s behavior prior to intervention, after the medication is tapered off to a stable level.
Surgical Procedure Neurosurgical intervention on all patients was performed by the same neurosurgeon (Bart Nuttin). We described this procedure elsewhere [105–107]. In all patients discussed here, quadripolar electrodes Model 3887 Pisces Quad Compact (4-mm contact spacing, 3-mm contact length; Medtronic Inc., Minnesota, USA) were stereotactically implanted into both anterior limbs of the internal capsules.
2905
2906
173
DBS for OCD
The stimulation targets in the internal capsules were similar to those aimed for in the anterior capsulotomy [103,112]. The stereotactic coordinates of the middle of the electrode tip in the first patient that turned out to have excellent results were 3.5 mm anterior to the posterior border of the anterior commissure, 13 mm lateral on the right side, 14 mm lateral on the left side and at the level of the horizontal AC-PC plane. The operations were performed under local anesthetics. The electrodes were connected to two IPGs (Itrell II, SynergyTM or KinetraTM, Medtronic Inc.) with subcutaneously connecting wires. These IPGs are subcutaneously implanted in the subclavicular or abdominal region.
Current Results of Capsular Stimulation DBS was implemented as a nondestructive alternative to capsulotomy in OCD in a series of ten patients and therapeutic benefits were reported [105,107,113,114]. With varying stimulation-parameters, a whole range of acute stimulation effects were observed. Some of them have been reported elsewhere [113,115,116]. Changes in affect were most prominent under bilateral stimulation. All patients reported sudden happiness, joy, and a good feeling some seconds after stimulation was switched on with particular contact combinations. They became more communicative and talked in a louder voice when happy feelings were induced. In several patients, unilateral stimulation (both left and right) with the deepest contacts produced transient contralateral contraction of facial muscles resulting in a typical demi-smile with higher amplitudes. This has been reproduced by others [117]. In some patients certain combinations led to a worsening mood, depressive feelings, and more anxiety. Switching stimulation off or changing to other contact combinations reversed those feelings.
A report on the first six patients described an average reduction in YBOCS scores from 32.3 3.9 to 19.8 8.0 following electrical stimulation for periods ranging from 3 to 31 months in this cohort of patients [106]. Eleven patients have been included in our protocol and have received bilateral electrode implants in the anterior limbs of the internal capsules. Eight patients participated in a double blind assessment of YBOCS scores during stimulation ‘‘on’’ and stimulation ‘‘off.’’ Mean Y-BOCS (SD) at baseline before surgery was 33.8 (3.2). At the end of the crossover branch, during which electrical stimulation was switched off, mean Y-BOCS (SD) was 33.1 (2.7), while at the end of the crossover branch during which patients received electrical stimulation mean Y-BOCS (SD) was 17.0 (7.8). In six of these eight patients the severity of OCD, as measured by the Y-BOCS, decreased with more than 35% and thus were considered responders [116]. Side effects include changes in weight and sleep pattern. The changes in sleep pattern were reversible with stimulation, but we could not find effective parameters that could avoid this changed sleep pattern and patients preferred those side effects to their OCD symptoms. Other side effects (disinhibition, overconfidence, and inaccurate risk assessment) are amplitude-dependent and disappear when the amplitude is lowered, but this sometimes comes at the cost of the therapeutic effect on OCD and requires careful balancing between therapeutic aim and undesirable secondary effects. Due to the high current densities necessary to obtain optimal therapeutic benefit, battery life is currently restricted. Since obsessions and compulsive rituals typically return within hours to days after failure of the batteries, patients require regular replacements. Obsessions and symptoms return with former intensity and often patients become severely depressed as well. Anderson and Ahmed [118] have also reported on single case studies of DBS used in this manner, and two similar investigations at European centers have been published [119,120].
DBS for OCD
Recently, follow-up results from other centers were published [121]. Ten adult OCD patients meeting stringent criteria for severity and treatment resistance received DBS in a target in the internal capsule immediately rostral to the anterior commissure extending into adjacent ventral capsule/ventral striatum. This open study found promising long-term effects of DBS in highly treatment-resistant OCD. Overall, approximately 75% of the patients treated with DBS at these centers have shown clinically significant improvements in OCD symptoms: 50% had a reduction in YBOCS of at least 35, and another 25% showed a reduction between 25 and 35% in YBOCS after a follow-up time of 36 months.
Impact of Frequency and Pulse Width To study the impact of frequency and pulse width (FQPW) on therapeutic effects, we set up a double blind experimental protocol. This subprotocol was planned either in the initial chronic phase or in the open continuation phase. Three different pulse widths (60, 210, and 450 ms) were combined with three different frequencies (10, 100, and 130 Hz) resulting in nine different combinations. The nine FQPW were tested successively in randomized order on the same day. The session was planned while patients were stimulated with stable parameters for at least 3 months. The amplitude was kept constant during all FQPW combinations at the level of the stimulation of the preceding months. Baseline data Y-BOCS, Hamilton Rating Scale for Depression (HAM-D) [122–124], Hamilton Anxiety Scale (HAM-A) [123,125], and the Beck Depression Inventory (BDI) [126,127] were collected at the start of the tests, with stimulation on. Patients completed ten consecutive test batches: a ‘‘dry run’’ with baseline parameters and nine FQPW.
173
Each batch contained three tasks.
Task 1: short free report. Patients were given a double page with a collection of photographs with neutral to pleasant value (nature, gardens, flowers, etc.). If they found some of the pictures unpleasant or threatening, they could choose another set. The instruction was to tell anything they wanted during 90 s. The photographs merely served as a guide. They could describe and elaborate on them or associate freely or tell anything that crossed their mind. At the end of their talk they were asked for a subjective units of distress score (SUDS). This score is used as a measure of concern, of psychological pain, or other distress that from the patient’s perspective is associated with OCD. The patient rates verbally SUDS on a scale from 0 (no distress at all) to 10 (maximally distressed – cannot bear it any longer). Task 2: expression of emotions. Patients were instructed to express and act out emotional states; for 5 s they were asked to mimic someone who feels angry, happy, startled, sad, and anxious. In between, they were asked to return to their neutral, normal expression. At the end of this task, they completed the Profile of Mood States (POMS) and Visual Analogue Scales (VAS) for obsessions, compulsions, avoidance, selfconfidence, and well-being. The POMS in short version [128] assesses transient, fluctuating affective mood states. This selfreport measures five dimensions of mood: depression, fatigue, anger, vigor, and tension and is totaled in a Mood Disturbance Score. VAS are ‘‘here and now’’ assessments of severity of symptoms and qualities. The patient puts a mark on an analog 100 mm line anchored at both ends (left = 0 = minimal level; right = 100 = maximal level). Task 3: Behavioral Avoidance Test (BAT). The BAT is a gradual exposure task. We used this
2907
2908
173
DBS for OCD
BAT to measure the impact of stimulation parameters on OCD symptoms in acute tests and devised them based on the BAT for OCD [129]. The task is designed to assess observable behavior and self reported severity of OCD symptoms that occur while the patient is attempting to carry out obsession/compulsion provoking tasks. After the exposure, the patient was asked for a SUDS and he completed again the POMS and VAS. To design the BAT, we obtained an inventory of obsessions, compulsions, trigger situations, and avoidance behavior for each patient. Trigger situations that evoked obsessive thoughts were hierarchically ordered from ‘‘easy to confront’’ to ‘‘most difficult, certainly not feasible.’’ We worked out at least five levels of increasing potency to trigger obsessive thoughts and drew up a stepwise script of individually tailored gradual exposure sessions. Patient and evaluator agreed upon the detailed script of exposure to trigger situations. Patients were instructed to expose themselves as far as they could. They were consecutively invited to advance to the next step, but they had full freedom to renounce to do so. The evaluator paid special attention to indirect avoidance strategies (e.g., passing responsibility upon the evaluator, mental checking compulsions). During sessions, patients were not allowed to perform the usual compulsive response. In between sessions, they had permission to give in to rituals if they felt the urge to do so.
The BAT requires strict instructions since performance varies with the perceived demands to perform the task. If the assessor strongly encourages the patient to carry out the feared task, this may not provide an accurate measure of naturally occurring avoidance. Tasks are sometimes difficult to construct [129]. Five patients (C2, C3, C4, C5, and C6) consented to participate in a study of the impact of FQPW on therapeutic effects. A separate informed consent was obtained. Follow-up time at the moment of this test session was 39 months (C2), 25 months (C3), 20 months (C4), 16 months (C5), and 13 months (C6). Baseline data (Y-BOCS, HAM-D, HAM-A, and the BDI) were collected at the start of the tests, with the parameters the patient was used to (> Table 173-1). The nine FQPW combinations were fully randomized and the order was changed between patients as to balance time effects (fatigue, habituation). We numbered the batches according to their sequence in time (1 = first batch to 9 = last batch). Thus we grouped them in ‘‘early’’ (time order 1–4) and ‘‘late’’ (time order 5–9). The time order of the batches between patients was not significantly correlated. We grouped the FQPW combinations in ‘‘low’’ FQPW (frequency = 10 Hz or pulse width = 60 ms) and ‘‘high’’ FQPW (frequency >10 Hz and pulse width >60 ms). The w2 test revealed that the ‘‘early’’ and ‘‘late’’ time group was not statistically significantly associated with ‘‘low’’ or ‘‘high’’ FQPW (w2 = 0.00; p = 0.9465).
. Table 173-1 Stimulation parameters and Y-BOCS, HAM-A, HAM-D, and BDI of five patients collected at the start of the FQPW test session, with stimulation on C2 C3 C4 C5 C6
Contacts (bilateral) 0+1–2+ 0–1–2–3+ 0–1+ 0–1–2+ 0–1+
Amplitude (V) 10.5 8.5 10.5 7.0 10.5
Frequency (Hz) 100 100 100 100 100
PW (s) 450 300 450 450 450
Y-BOCS 19 14 22 28 12
HAM-A 20 4 12 13 4
HAM-D 17 3 7 12 4
BDI 30 7 18 22 9
DBS for OCD
The amplitude was programmed at a constant level for all FQPW combinations, equal to the level with which the patient was stimulated on entering the FQPW protocol. However, we received technical information from Medtronic after analysis of the FQPW data. This information stipulated that certain combinations of high amplitude, pulse width, and frequency settings are not achievable with Synergy neurostimulators. The Synergy program device allows to set these combinations, but when the maximal performance of Synergy is exceeded, the amplitude is lowered automatically (without external notice or notification of error) to the maximal limit for the given pulse width and frequency. In > Table 173-2, the actual amplitudes for the given FQPW combinations are represented for each patient. The exposure triggers for the BAT in each FQPW combination, with the successive steps are listed in > Table 173-3. The exposure sessions induced distress and increased SUDS in all FQPW conditions(> Table 173-4). The POMS and VAS were scored twice in each batch, first time after the patient expressed emotional states and second time after the BAT. The mean scores on the POMS after expression of emotions and after the BAT are presented in > Table 173-5. There was considerable
173
interindividual variability in the scores of the five patients. The decrease in mean score on the POMS for mood disturbance and on POMS subscales for depression and tension in ‘‘high FQPW’’ versus ‘‘low FQPW’’ was statistically significant (p < 0.05) both after expression of emotional states and after exposure. After expression of emotional states, POMS mood disturbance decreased by a mean of 13.1, from 81.6 (low FQPW) to 68.5 (high FQPW) (95% CIdiff = 8.7–17.5). After exposure, POMS mood disturbance showed a mean decrease of 13.6, from 84.8 (low FQPW) to 71.2 (high FQPW) (95% CIdiff = 6.1–21.0). After expression of emotional states, POMS depression decreased by a mean of 3.8, from 28.0 (low FQPW) to 24.2 (high FQPW) (95% CIdiff = 2.2–5.6). After exposure, POMS depression showed a mean decrease of 3.6, from 28.7 (low FQPW) to 25.1 (high FQPW) (95% CIdiff = 1.7–5.5). After expression of emotional states, POMS tension decreased by a mean of 4.0, from 23.8 (low FQPW) to 19.8 (high FQPW) (95% CIdiff = 1.8–6.2). After exposure, POMS tension showed a mean decrease of 3.8, from 24.4 (low FQPW) to 20.6 (high FQPW) (95% CIdiff = 2.0–5.8).
. Table 173-2 Parameters used in the FQPW protocol, with amplitude limited by the electronic properties of the Synergy neurostimulators Frequency (Hz) 10 10 10 100 130 100 100 130 130
PW (s) 60 210 450 60 60 210 450 210 450
Amplitude C2 (V) 10.5 10.5 10.5 10.5 10.5 9.5 7.1 8.6 6.0
Amplitude C3 (V) 8.5 8.5 8.5 8.5 8.5 8.5 7.1 8.5 6.0
Amplitude C4 (V) 10.5 10.5 10.5 10.5 10.5 9.5 7.1 8.6 6.0
Amplitude C5 (V) 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 6.0
Amplitude C6 (V) 10.5 10.5 10.5 10.5 10.5 9.5 7.1 8.6 6.0
Mean (SD) (V) 9.4 (1.6) 9.4 (1.6) 9.4 (1.6) 9.4 (1.6) 9.4 (1.6) 8.8 (1.1) 7.1 (0) 8.3 (0.7) 6.0 (0)
2909
2910
173
DBS for OCD
. Table 173-3 Overview of the steps for gradual exposure used in BAT for frequency-pulse width tests Exposure General theme
C2 Contamination plants
Step a
Plant on table
Step b
Touch the plant with full hand Pull off/remove some leaves
Step c
Step d
Crush the leaves in your hands
Step e
Touch your face and mouth with dirty hands
C3 Nonexistence Terrifying objects Three objects (easy, medium, difficult) on table Take the easy object (pencil) and confront it Take the medium object (post card) and confront it Take the difficult object (pillow) and confront it Put your head on the pillow
C4 Contamination Nicotine, ashes
C5 Contamination towels
Mixture from tobacco and ashes on plate
Touch clean towel
C6 Aggressive thoughts about poisoning others Strip of pills on table
Touch the mixture
Touch towel with bread crumbs
Press one pill out of the strip
Take some of the mixture and rub it in your hands
Touch towel soiled with earth
Closed jar with pills
Rub the mixture in a hospital sheet
Touch towel with brown (chocolate) and red (ketchup) stains Touch a dirty towel used to clean the toilet
Open jar and take many in your hands
Contaminate a drinking glass, cup and saucer of the hospital
Keep hands with pills under tabletop, out of reach from the camera
. Table 173-4 Mean SUDS scores (n = 5) after free report (SUDS A) and after BAT exposure (SUDS B) in nine frequency-pulse width combinations Frequency (Hz) 10 10 10 100 130 100 100 130 130
PW (s) 60 210 450 60 60 210 450 210 450
Free report SUDS A Mean (SD) 6.2 (2.6) 5.6 (2.9) 5.8 (2.6) 6.0 (2.6) 5.4 (2.9) 5.0 (1.6) 5.2 (2.8) 4.6 (2.3) 4.2 (2.0)
Exposure SUDS B Mean (SD) 7.6 (2.8) 7.6 (3.0) 7.6 (2.5) 8.2 (1.6) 6.8 (3.3) 6.8 (2.6) 5.8 (2.8) 6.6 (2.5) 6.4 (2.5)
Mean SUDS B–SUDS A 1.4* 2.0* 1.8* 2.2* 1.4* 1.8* 0.6 2.0* 2.2*
95% CIdiff SUDS B–SUDS A 0.1–3.1 0.8–3.5 0.3–3.6 0.6–4.2 0.1–3.1 0.3–3.6 0.9–2.5 0.1–4.3 0.9–3.8
*p < 0.05
The mean score on the POMS fatigue and POMS anger also decreased in ‘‘high FQPW’’ versus ‘‘low FQPW’’ both after expression of emotional states and after exposure but the difference did not reach statistical significance.
POMS vigor (a quality with positive valence) increased in ‘‘high FQPW’’ versus ‘‘low FQPW,’’ both after expression of emotional states and after exposure but the difference did not reach statistical significance (p < 0.05).
Frequency (Hz) 10 10 10 100 130 100 100 130 130
FQPW group Low
PW (ms) 60 210 450 60 60 210 450 210 450
PW (ms) 60 210 450 60 60 210 450 210 450 Mn amplitude (V) 9.4 9.4 9.4 9.4 9.4 8.8 7.1 8.3 6.0
Fatigue, Mn (SD) 20.2 (4.8) 21.4 (5.6) 23.6 (4.0) 20.8 (4.2) 18.8 (4.7) 21.0 (4.5) 18.4 (4.0) 18.6 (5.0) 19.8 (4.6)
Depression, Mn (SD) 29.2 (4.6) 28.0 (6.7) 28.2 (6.8) 30.6 (5.0) 27.6 (6.7) 25.8 (10.1) 23.8 (9.4) 25.6 (6.3) 25.2 (10.4)
Fatigue, Mn (SD) 22.0 (5.0) 21.6 (5.1) 22.4 (4.7) 23.6 (3.5) 20.2 (3.7) 21.8 (4.7) 18.2 (5.0) 18.2 (4.8) 20.2 (4.9)
After BAT exposure to triggers
Depression, Mn (SD) 29.6 (4.0) 27.0 (7.8) 26.6 (6.6) 30.0 (6.1) 27.0 (6.6) 25.6 (9.8) 22.6 (7.8) 23.8 (7.7) 24.6 (9.7)
After expression of emotions
Anger, Mn (SD) 24.2 (6.9) 20.8 (8.3) 20.2 (9.5) 23.2 (6.3) 21.2 (7.4) 19.0 (9.1) 19.8 (6.6) 20.0 (6.8) 19.2 (9.8)
Anger, Mn (SD) 22.4 (5.9) 20.8 (7.7) 18.2 (9.3) 23.0 (6.9) 20.0 (8.4) 18.8 (9.7) 18.4 (7.9) 17.8 (8.7) 18.0 (8.2)
Note: Mean scores (Mn) and standard deviation (SD) for five patients, in nine frequency-pulse width combination
High
High
Frequency (Hz) 10 10 10 100 130 100 100 130 130
FQPW group Low
Mn amplitude (V) 9.4 9.4 9.4 9.4 9.4 8.8 7.1 8.3 6.0
Profile Of Mood State (POMS)
Vigor, Mn (SD) 11.8 (5.1) 11.8 (5.4) 12.8 (4.1) 11.6 (5.5) 13.0 (3.9) 13.8 (3.6) 14.0 (3.8) 12.4 (2.9) 14.0 (4.1)
Vigor, Mn (SD) 11.4 (4.1) 11.6 (4.7) 12.6 (4.3) 12.2 (4.8) 12.8 (4.0) 14.0 (3.5) 12.8 (2.8) 12.2 (3.6) 13.8 (3.6)
25.8 (4.1) 24.0 (4.2) 23.4 (4.4) 25.6 (3.2) 23.4 (4.6) 21.4 (8.3) 19.2 (7.9) 21.0 (6.1) 20.6 (9.0)
Tension, Mn (SD)
23.6 (4.9) 23.8 (4.4) 24.0 (4.3) 24.2 (4.3) 23.4 (5.1) 21.2 (7.5) 18.0 (6.2) 19.6 (6.3) 20.4 (8.3)
Tension, Mn (SD)
Mood disturbance, Mn (SD) 89.4 (21.4) 82.6 (17.6) 81.2 (20.4) 91.4 (15.9) 79.4 (19.1) 74.2 (28.2) 67.2 (26.5) 72.4 (21.4) 71.2 (30.5)
Mood disturbance, Mn (SD) 84.4 (21.1) 81.4 (17.8) 79.8 (21.3) 85.8 (19.6) 76.4 (22.0) 72.6 (28.4) 64.6 (23.5) 67.6 (23.0) 69.0 (28.9)
. Table 173-5 Profile Of Mood State (POMS) subscale scores for depression, fatigue, anger, vigor, and tension and total score for mood disturbance after expression of emotions (upper part) and after exposure in the BAT (bottom part)
DBS for OCD
173 2911
2912
173
DBS for OCD
On the individual level we analyzed the data after z-transformation. > Figures 173-1–173-6 represent the individual z-transformed scores of the five patients on the POMS after exposure for the nine frequency-pulse width combinations. The VAS scales for obsessions, compulsions and avoidance increased and VAS scales for selfconfidence and well-being decreased after the BAT
exposures, compared with the scores obtained after the emotional expressions, but interindividual variation was high (> Table 173-6). After expression of emotional states, VAS obsessions decreased by a mean of 24, from 63 (low FQPW) to 39 (high FQPW) (95% CIdiff = 11–37). After exposure, VAS obsessions showed a mean decrease of 22, from 71 (low FQPW) to 49 (high FQPW) (95% CIdiff = 15–30).
. Figure 173-1 Stacked column chart of the z-transformed scores of five patients on Profile of Mood States (POMS) depression after Behavioral Avoidance Test (BAT) exposure in nine frequency-pulse width combinations
. Figure 173-2 Stacked column chart of the z-transformed scores of five patients on POMS fatigue after BAT exposure in nine frequency-pulse width combinations
DBS for OCD
173
. Figure 173-3 Stacked column chart of the z-transformed scores of five patients on POMS anger after BAT exposure in nine frequency-pulse width combinations
. Figure 173-4 Stacked column chart of the z-transformed scores of five patients on POMS vigor after BAT exposure in nine frequency-pulse width combinations
After expression of emotional states, VAS compulsions decreased with a mean 16, from 53 (low FQPW) to 37 (high FQPW) (95% CIdiff = 13–20). After exposure, VAS compulsions showed a mean decrease of 18, from 62 (low FQPW) to 44 (high FQPW) (95% CIdiff = 10–37). After expression of emotional states, VAS avoidance decreased with a mean 20, from 59 (low FQPW) to 39 (high FQPW) (95% CIdiff = 12–29). After exposure, VAS avoidance showed a mean
decrease of 17, from 62 (low FQPW) to 45 (high FQPW) (95% CIdiff = 3–30). After expression of emotional states, VAS selfconfidence increased with a mean 10, from 27 (low FQPW) to 37 (high FQPW) (95% CIdiff = 2–23). After exposure VAS self-confidence showed a mean increase of 15, from 21 (low FQPW) to 36 (high FQPW) (95% CIdiff = 10–28). After expression of emotional states, VAS well-being increased with a mean 14, from 28
2913
2914
173
DBS for OCD
. Figure 173-5 Stacked column chart of the z-transformed scores of five patients on POMS tension after BAT exposure in nine frequency-pulse width combinations
. Figure 173-6 Stacked column chart of the z-transformed scores of five patients on POMS mood disturbance after BAT exposure in nine frequency-pulse width combinations
(low FQPW) to 42 (high FQPW) (95% CIdiff = 3–27). After exposure, VAS well-being showed a mean increase of 20, from 21 (low FQPW) to 41 (high FQPW) (95% CIdiff = 9–30). On the individual level, we analyzed the data after z-transformation. > Figures 173-7– 173-11 represent the individual z-transformed scores of the five patients on the VAS after exposure for the nine frequency-pulse width combinations.
Impact of Stimulation Parameters Few indications are available about the role of the different parameters of stimulation on the effectiveness of DBS. The fundamental purpose of DBS is to modulate neural activity with applied electric fields, but a quantitative understanding of the effects of manipulating the various stimulation parameters on the neural response to the stimulation is lacking. The selection of therapeutic
Frequency (Hz) 10 10 10 100 130 100 100 130 130
FQPW group Low
PW (ms) 60 210 450 60 60 210 450 210 450
PW (ms) 60 210 450 60 60 210 450 210 450
Mn amplitude (V) 9.4 9.4 9.4 9.4 9.4 8.8 7.1 8.3 6.0
Mn amplitude (V) 9.4 9.4 9.4 9.4 9.4 8.8 7.1 8.3 6.0 Compulsions, Mn (SD) 59.6 (15.8) 56.4 (23.3) 52.6 (13.4) 48.2 (24.2) 49.6 (25.6) 42.2 (23.6) 37.2 (26.4) 35.0 (27.0) 34.4 (22.0)
Obsessions, Mn (SD) 74.2 (13.7) 70.2 (17.7) 62.6 (25.0) 73.2 (23.0) 76.2 (23.0) 56.6 (26.1) 44.2 (31.4) 46.2 (32.5) 48.0 (27.8)
Compulsions, Mn (SD) 69.4 (16.8) 66.0 (20.6) 58.8 (27.5) 61.8 (23.0) 54.2 (29.7) 51.8 (33.1) 37.6 (33.8) 41.6 (31.1) 43.0 (33.2)
After exposure and BAT
Obsessions, Mn (SD) 67.8 (15.9) 64.4 (17.6) 50.2 (21.5) 65.6 (31.2) 67.2 (22.2) 44.0 (22.5) 33.2 (27.3) 38.0 (24.1) 41.0 (21.6)
Note: Mean scores (Mn) and standard deviation (SD) for five patients, in nine frequency-pulse width combinations
High
High
Frequency (Hz) 10 10 10 100 130 100 100 130 130
FQPW group Low
After expression of emotions
Visual Analog Scales (VAS)
Avoidance, Mn (SD) 80.2 (18.8) 60.6 (24.2) 58.6 (29.3) 56.8 (29.2) 55.0 (30.6) 52.8 (30.7) 36.4 (34.3) 48.8 (33.7) 44.4 (31.5)
Avoidance, Mn (SD) 74.8 (19.6) 55.0 (28.2) 58.4 (25.5) 60.6 (27.6) 47.8 (26.1) 47.2 (30.2) 32.8 (25.4) 41.6 (25.7) 32.8 (16.8)
. Table173-6 Visual Analogue Scale (VAS) after expression of emotions (upper part) and after BAT (bottom part)
Self-confidence, Mn (SD) 16.6 (14.2) 21.6 (19.7) 27.0 (23.6) 18.4 (10.5) 24.2 (13.4) 37.2 (28.6) 38.6 (22.6) 30.0 (12.7) 38.8 (26.3)
Self-confidence, Mn (SD) 32.4 (28.0) 20.0 (12.6) 32.8 (26.1) 17.6 (2.9) 30.6 (23.2) 34.2 (27.5) 39.4 (27.7) 41.2 (30.0) 34.4 (24.1)
Well-being, Mn (SD) 14.6 (15.4) 28.4 (18.9) 21.0 (11.6) 17.6 (2.4) 25.2 (11.3) 43.0 (27.7) 43.0 (16.7) 36.2 (15.4) 41.6 (31.3)
Well-being, Mn (SD) 21.6 (15.9) 25.8 (13.5) 32.4 (17.9) 22.4 (6.3) 35.8 (21.1) 39.0 (25.1) 40.0 (24.6) 41.2 (21.6) 49.0 (29.7)
DBS for OCD
173 2915
2916
173
DBS for OCD
. Figure 173-7 Stacked column chart of the z-scores of five patients on Visual Analogue Scale (VAS) obsessions after BAT exposure in nine frequency-pulse width combinations
. Figure 173-8 Stacked column chart of the z-scores of five patients on VAS compulsions after BAT exposure in nine frequencypulse width combinations
stimulation parameters for DBS can be a difficult and time-consuming process. A recent clinical study found that the total time spent programming the stimulator and assessing DBS patients ranged from 18–36 h per patient [130]. Nevertheless, the
therapeutic benefit achieved with DBS is strongly dependent on the contact combination and the other parameters programmed. We already reported on the dose-relationship between stimulation amplitude and
DBS for OCD
173
. Figure 173-9 Stacked column chart of the z-scores of five patients on VAS avoidance after BAT exposure in nine frequency-pulse width combinations
. Figure 173-10 Stacked column chart of the z-scores of five patients on VAS self-confidence after BAT exposure in nine frequency-pulse width combinations
therapeutic benefit in OCD patients, when an effective contact combination is used [105]. We investigated the impact of two other stimulation parameters, FQPW, in a separate protocol carried out after more than 1 year of DBS for treatment-refractory OCD. This time lag was sufficient to eliminate any clinical improvement due to microlesion effects of the surgical procedure, to
determine an effective contact combination and obtain a stabilization of the clinical and mental state of the patients. Within a given beneficial contact combination stimulation with high amplitude (up to 10.5 V) did not induce mood improvement or decrease in OCD symptoms when low frequency (10 Hz) or narrow pulse width (60 ms) was used. Higher frequencies (100, 130 Hz) and
2917
2918
173
DBS for OCD
. Figure 173-11 Stacked column chart of the z-scores of five patients on VAS feeling of well-being after BAT exposure in nine frequency-pulse width combinations
larger pulse widths (210, 450 ms) induced improvements in self-rated mood disturbance, depression, and tension, with important decrease in obsessive thoughts, in the urge to perform compulsive rituals and to avoid trigger situations during a behavioral assessment. Patients felt more self-confident and their general feeling of well-being increased as well. Whether the beneficial effects for FQPW are dose-dependent, or whether a certain minimum threshold level of FQPW is required for effectiveness, in an ‘‘all or nothing’’ fashion, needs further elucidation. Another study reported on the importance of frequency in elucidating positive clinical responses (decrease in anxiety and tension, feeling of well-being, and relaxation) and strange sensations (stiffness of neck, head movement; headaches, tremor of the head) with periodic (left sided) capsular stimulation prior to capsulotomy in patients with anxiety disorders and OCD [131]. For DBS in PD as well, an increase in clinical effectiveness with increased pulse width was noted, but considerable modifications of pulse width were required to obtain significant differences, and this coincided with considerable narrowing of the therapeutic window [132].
The relationship between various stimulus parameters and the resultant electric field around the electrodes has been studied mostly in a homogeneous and isotropic medium. Such models are not directly applicable to the internal capsule, a morphological fibrous, anisotropic, and asymmetrical structure. It is surrounded by a range of gray and white matter structures, resulting in an inhomogeneous and anisotropic environment that distorts the shape of the DBS electric field and subsequent neural response to stimulation. In an homogeneous medium, the stimulus amplitude directly affects the extend of the spatial spread around the active contacts and as such exerts a primary control of the spatial control of DBS therapy. Whether a neuronal structure at a specific spatial position within this electrical field is activated or not depends on the current intensity and the pulse width it is exposed to. This can be described in the strength–duration relationship curve. The characteristics of such a curve vary depending upon the neuronal substructure being activated (axon vs. dendrite vs. cell body) and the size of the structure (e.g., larger axons can be excited at narrower pulse widths than smaller axons) [133–135]. Differences in pulse
DBS for OCD
width, at a given amplitude, can therefore have a significant influence on the activation of neuronal structures within the electrical field, with increasing pulse widths leading to an expanded region of activation within the overall electrical field (personal communication, P. Stypulkowsky, Medtronic). The influence of FQPW on the electrophysiology in the volume influenced by DBS depends on the electrode–tissue interface at the target location. The electric field generated by DBS is dependent on the stimulus waveform, electrode capacitance, electrode impedance, electrode geometry (monopolar/bipolar), and electrical conductivity of the surrounding tissue medium. Computational models that accurately estimate the volume of tissue activated by DBS as a function of the stimulation parameters (contact, impedance, voltage, pulse width, frequency) and electrode location in the brain [136,137] are developed. However, the computational power and computer science skills necessary to effectively implement such models are not available to most clinicians. Software packages that could aid the postoperative programming of DBS are underway [138]. The model system consists of three fundamental components: a 3D anatomical model of the subcortical nuclei and DBS electrode position in the brain, each derived from magnetic resonance imaging, a finite element model of the DBS electrode and electric field transmitted to the brain, with tissue conductivity properties, and the prediction of the volume of tissue activated derived from the response of myelinated axons to the applied electric field, which is a function of the stimulation parameters (contact, impedance, voltage, pulse width, frequency). The combined analysis of data obtained from neuroimaging (MRI and DTI), neuroanatomy, neurostimulation in the individual patient, and behavioral data observed or reported, has the potential to shed new light on many scientific and clinical questions related to DBS and to the pathophysiology of the disorders in which it is used.
173
Conclusion DBS was implemented as a nondestructive alternative to capsulotomy in patients with severe and treatment refractory OCD and therapeutic benefits were reported in 75% of the patients, but the current experience is only limited, with <50 patients reported in the literature worldwide. With this experiment, we could demonstrate that even with high amplitudes (up to 10.5 V) a beneficial contact combination stimulation did not induce mood improvement or decrease in OCD symptoms when frequency was low (10 Hz) or pulse width was narrow (60 ms). Higher frequencies (100, 130 Hz) and larger pulse widths (210 ms, 450 ms) induced improvements in self-rated mood disturbance, depression and tension, with important decrease in obsessive thoughts, in the urge to perform compulsive rituals and to avoid trigger situations during a behavioral assessment.
References 1. Rasmussen SA, Eisen JL. The epidemiology and clinical features of obsessive-compulsive disorder. Psychiatr Clin North Am 1992;15:743-58. 2. Weissman MM. Cross-national epidemiology of obsessive-compulsive disorder. CNS Spectrums 1998;3(5):6-9. 3. Rachman S, Shafran R. Cognitive distortions: thoughtaction fusion. Clin Psychol Psychother 1999;6:80-5. 4. Salkovskis PM, Wroe AL, Gledhill A, et al. Responsibility attitudes and interpretations are characteristic of obsessive compulsive disorder. Behav Res Ther 2000;38:347-72. 5. LaSalle V, Cromer K, Nelson K, Kazuba D, Justement L, Murphy D. Diagnostic interview assessed neuropsychiatric disorder comorbidity in 334 individuals with obsessive-compulsive disorder. Depress Anxiety 2004; 19:163-73. 6. Rasmussen SA, Eisen JL. Clinical and epidemiological findings of significance to neuropharmacologic trials in OCD. Psychopharmacol Bull 1988;24:466-70. 7. Zitterl W, Demai U, Aigner M, et al. Naturalistic course of obsessive compulsive disorder and comorbid depression. Psychopathology 2000;33:75-80. 8. Hollander E, Stein DJ, Kwon JH, et al. Psychosocial function and economic costs of obsessive-compulsive disorder. CNS Spectrums 1998;3(5):48-58.
2919
2920
173
DBS for OCD
9. Steketee G. Disability and family burden in obsessivecompulsive disorder. Can J Psychiatry 1997;42(9):919-28. 10. Hollander E, Stein D, Broatch J, Himelein C, Rowland C. A pharmacoeconomic and quality of life study of obsessive-compulsive disorder. CNS Spectrums 1997;2:16-25. 11. Emmelkamp PMG, de Haan E, Hoogduin CAL. Marital adjustment and obsessive-compulsive disorder. Br J Psychiatry 1990;156:55-9. 12. Calvocoressi L, Libman D, Vegso SJ, et al. Global functioning of inpatients with obsessive compulsive disorder, schizophrenia and major depression. Psychiatr Serv 1998;49:379-381. 13. Amir N, Freshman M, Foa EB. Family distress and involvement in relatives of obsessive-compulsive disorder patients. J Anxiety Disord 2000;14:209-17. 14. Koran LM. Quality of life in obsessive-compulsive disorder. Psychiatry Clin N Am 2000;23:505-17. 15. Jayakumar C, Jagadheesan K, Verma A. Disability in obsessive-compulsive disorder: a comparison with schizophrenia. Int J Rehabil Res 2002;25(2):147-51. 16. Alexander GE, DeLong MR, Strick PL. Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annu Rev Neurosci 1986;9:357-81. 17. Aouizerate B, Guehl D, Cuny E, Rougier A, Bioulac B, Tignol J, Burbaud P. Pathophysiology of obsessivecompulsive disorder: a necessary link between phenomenology, neuropsychology, imagery and physiology. Prog Neurobiol 2004;72:195-221. 18. Rauch SL, Jenike MA. Neurobiological models of obsessive-compulsive disorder. Psychosomatics 1993;34 (1):20-32. 19. Rauch SL, Savage CR, Alpert NM, et al. Probing striatal function in obsessive-compulsive disorder: a PET study of implicit sequence learning. J Neuropsychiatr Clin Neurosci 1997;9:568-73. 20. Saxena S, Brody AL, Schwartz JM, Baxter LR. Neuroimaging and frontal-subcortical circuitry in obsessivecompulsive disorder. Br J Psychiatry 1998;173:26-37. 21. Rauch SL, Benkelfat C, Dager SR, et al. Neuroimaging research and neurocircuitry models of obsessivecompulsive disorder: proceedings of the third IOCDC. CNS Spectrums 1999;4:25-34. 22. Baxter LR, Schwartz JM, Mazziotta JC, et al. Cerebral glucose metabolic rates in nondepressed patients with obsessive-compulsive disorder. Am J Psychiatry 1988;145:1560-3. 23. Rauch SL, Jenike MA, Alpert NM, et al. Regional cerebral blood flow measured during symptom provocation in obsessive-compulsive disorder using 15-O-labelled CO2 and positron emission tomography. Arch Gen Psychiatry 1994;51:62-70. 24. Rauch SL, Savage CR, Alpert NM, et al. A positron emission tomographic study of simple phobic symptom provocation. Arch Gen Psychiatry 1995;52:20-28. 25. Rauch SL, van der Kolk BA, Fisler RE, et al. A symptom provocation study of posttraumatic stress disorder using
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
positron emission tomography and script-driven imagery. Arch Gen Psychiatry 1996;53:380-7. Brody AL, Saxena S, Schwartz JM, et al. FDG-PET predictors of response to behavioral therapy and pharmacotherapy in obsessive-compulsive disorder. Psychiatry Res 1998;84:1-6. Schwartz JM. Neuroanatomical aspects of cognitivebehavioural therapy response in obsessive-compulsive disorder. Br J Psychiatry 1998;173:38-44. Kang D, Kwon J, Kim J, et al. Brain glucose metabolic changes associated with neuropsychological improvements after 4 months of treatment in patients with obsessive-compulsive disorder. Acta Psychiatr Scand 2003;107:291-7. Dell’Osso B, Altamura A, Mundo E, Marazziti D, Hollander E. Diagnosis and treatment of obsessivecompulsive disorder and related disorders. Int J Clin Pract 2007;61(1):98-104. Foa EB, Liebowitz MR, Kozak MJ, et al. Randomized, placebo controlled trail of exposure and ritual prevention, clomipramine and their combination in the treatment of obsessive-compulsive disorder. Am J Psychaitry 2005;162:151-61. Zohar J, Chopra M, Sasson Y, et al. Psychopharmacology of obsessive-compulsive disorder? World J Biol Psychiatry 2000;1:92-100. Pallanti S, Hollander E, Bienstock C, et al. Treatment non response in OCD: methodological issues and operational definitions. Int J Neuropsychopharmacol 2002;5:181-91. Dell’Osso B, Nestadt G, Allen A, Hollander E. Serotoninnorepinephrine reuptake inhibitors in the treatment of obsessive-compulsive disorder: a critical review. J Clin Psychiatry 2006;67:600-10. McDougle CJ, Epperson CN, Pelton GH, et al. A double blind, placebo-controlled study of risperidone addition in serotonin reuptake inhibitor-refractory obsessive-compulsive disorder. Arch Gen Psychiatry 2000;57:794-801. Francobandiera G. Olanzapine augmentation of serotonin uptake inhibitors in obsessive-compulsive disorder: an open study. Can J Psychiatry 2001;46:356-8. Kaplan A, Hollander E. A review of pharmacologic treatments for obsessive-compulsive disorder. Psychiatric Services 2003;54:1111-18. Eisen JL, Goodman WK, Keller MB, et al. Patterns of remission and relapse in obsessive-compulsive disorder: a 2-year prospective study. J Clin Psychiatry 1999;60:346-52. Hollander E, Bienstock CA, Koran LM, Pallanti S, Marazziti D, Rasmussen SA, Ravizza L, Benkelfat C, Saxena S, Greenberg BD, Sasson Y, Zohar J. ‘‘Refractory obsessive-compulsive disorder: state-of-the-art treatment.’’ J Clin Psychiatry 2002; 63 Suppl 6:20-9. Rasmussen SA, Tsuang MT. The epidemiology of obsessive-compulsive disorder. J Clin Psychiatry 1984;45:450-7.
DBS for OCD
40. Mindus P. Present-day indications for capsulotomy. Acta Neurochir 1993;58(Suppl):29-33. 41. Knight GC. The orbital cortex as an objective in the surgical treatment of mental illness. The development of the stereotactic approach. Br J Surgery 1964;51:114-24. 42. Go¨ktepe EO, Young LB, Bridges PK. A further review of the results of stereotactic subcaudate tractotomy. Br J Psychiatry 1975;126:270-80. 43. Hodgkiss AD, Malizia AL, Bartlett JR, Bridges PK. Outcome after the psychosurgical operation of stereotactic subcaudate tractotomy, 1979–1991. J Neuropsychiatry 1995;7:230-4. 44. Woerdeman PA, Willems PW, Noordmans HJ, Berkelbach van der Sprenkel JW, van Rijen PC. Frameless stereotactic subcaudate tractotomy for intractable obsessive-compulsive disorder. Acta Neurochir Wien 2006;148(6):633-7. 45. Martuza RL, Chiocca EA, Jenike MA, Jiriunas IE, Ballantine HT Jr. Stereotactic radiofrequency thermal cingulotomy for obsessive compulsive disorder. J Neuropsychiatry Clin Neurosci 1990;2:331-6. 46. Kim C, Chang J, Koo M, Kim J, Suh H, Park I, Lee H. Anterior cingulotomy for refractory obsessive-compulsive disorder. Acta Psychiatr Scand 2003:107:283-90. 47. Ballantine HT, Giriunas IE. Treatment of intractable psychiatric illness and chronic pain by stereotactic cingulotomy. In: Schmidek HH, Sweet WH, editors. Operative neurosurgical techniques. New York: Grune and Stratton; 1982. p. 1069-75. 48. Ballantine HT, Bouckoms AJ, Thomas EK, Giriunas IE. Treatment of psychiatric illness by stereotactic cingulotomy. Biol Psychiatry 1987;22:807-19. 49. Baer L, Rauch SL, Ballantine HT, et al. Cingulotomy for intractable obsessive-compulsive disorder: prospective long-term follow-up of 18 patients. Arch Gen Psychiatry 1995;52:384-92. 50. Dougherty DD, Baer L, Cosgrove GR, et al. Prospective long-term follow-up of 44 patients who received cingulotomy for treatment refractory obsessive-compulsive disorder. Am J Psychiatry 2002;15(2):269-75. 51. Spangler WJ, Cosgrove GR, Ballantine HT, et al. Magnetic resonance image-guided stereotactic cingulotomy for intractable psychiatric disease. Neurosurgery 1996;38:1071-6. 52. Jenike M, Baer L, Ballantine T, et al. Cingulotomy for refractory obsessive-compulsive disorder: a long-term followup of 33 patients. Arch Gen Psychiatry 1991;48:548-55. 53. Kelly D, Richardson A, Mitchell-Heggs N, Greenup J, Chen C, Hafner RJ. Stereotactic limbic leucotomy: a preliminary report on forty patients. Br J Psychiatry 1973;123:141-8. 54. Mitchell-Heggs N, Kelly D, Richardson A. Stereotactic limbic leucotomy – a follow-up at 16 months. Br J Psychiatry 1976;128:226-40. 55. Mitchell-Heggs N, Kelly D, Richardson A, McLeigh J. Further exploration of limbic leucotomy. In: Hitchcock
56.
57. 58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
173
ER, Ballantine HT Jr, Meyerson BA, editors. Modern concepts in psychosurgery. Amsterdam: Elsevier/NorthHolland Biomedical Press; 1979. Talairach J, Hecaen H, David M. Lobotomie prefrontale limite´e par electrocoagulation des fibres thalamo-frontalis a` leur emergence du bras anterior de la capsule interne. Proceedings of the 4th Congress Neurologique Internationale. Paris: Masson; 1949. p. 141. Leksell L. A stereotaxic apparatus for intracerebral surgery. Acta Chir Scand 1949;99:229-33. Mindus P, Rasmussen SA, Lindquist C. Neurosurgical treatment for refractory obsessive-compulsive disorder: implications for understanding frontal lobe functions. J Neuropsychiatry Clin Neurosci 1994;6(4):467-77. Oliver B, Gasco´n J, Aparicio A, Ayats E, Rodriguez R, Maestro De Leo´n JL, Garcı´a-Bach M, Soler PA. Bilateral anterior capsulotomy for refractory obsessive-compulsive disorders. Stereotact Funct Neurosurg 2003;81(1–4): 90-5. Herner T. Treatment of mental disorders with frontal stereotactic thermal lesions. A follow-up study of 116 cases. Acta Psychiatr Neurolog Scand 1961;158:36-42. Mindus P, Bergstrom K, Levander SE, et al. Magnetic resonance images related to clinical outcome after psychosurgical intervention in severe anxiety disorders. J Neurol Neurosurg Psychiatry 1987;50:1288-93. Mindus P, Nyman H, Mogard J, et al. Frontal lobe and basal ganglia metabolism studies in patients with incapacitating obsessive-compulsive disorder undergoing capsulotomy. Nord J Psychiatry 1990; 44:309-12. Kihlstrom L, Guo W-L, Lindquist C, et al. Radiobiology of radiosurgery for refractory anxiety disorders. Neurosurgery 1995;36:294-302. Rasmussen S, Greenberg B, Mindus P, et al. Neurosurgical approaches to intractable obsessive-compulsive disorder. CNS Spectrums 2000;5(11):23-34. Royal College of Psychiatrists, London. Neurosurgery for mental disorders. Report of the Neurosurgery Working Group of the Royal College of Psychiatrists. Council report CR89, 2000. Tan E, Marks IM, Marset P. Bi-medial leukotomy in obsessive-compulsive neurosis: a controlled serial enquiry. Br J Psychiatry 1971;118:155-64. Kullberg G. Differences in effect of capsulotomy and cingulotomy. In: Sweet WH, Obrador S, Martin-Rodriguez JG, editors. Neurosurgical treatment in psychiatry, pain, and epilepsy. Baltimore, Maryland: University Park Press; 1977. p. 301-8. Hay P, Sachdev P, Cumming S, et al. Treatment of obsessive-compulsive disorder by psychosurgery. Acta Psychiatr Scand 1993;87:197-207. Sachdev P, Hay P. Does neurosurgery for obsessivecompulsive disorder produce personality change? J Nerv Ment Dis 1995;183:408-13. Irle E, Exner C, Thielen K, et al. Obsessive-compulsive disorder and ventromedial frontal lesions: clinical and
2921
2922
173 71. 72.
73.
74.
75. 76.
77. 78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
DBS for OCD
neuropsychological findings. Am J Psychiatry 1998;155:255-63. Jenike MA. Neurosurgical treatment of obsessive-compulsive disorder. Br J Psychiatry 1998;173:79-90. Albucher R, Curtis G, Pitts K. Neurosurgery for obsessivecompulsive disorder: problem with comorbidity. Lett Am J Psychiatry 1999;156:495-6. World Health Organization. Health for all by the Year 2000: strategies. WHO Official document 173. Switzerland; Geneva, 1980. Rasmussen SA, Eisen J. Treatment strategies for chronic and refractory obsessive-compulsive disorder. J Clin Psychiatry 1997;58 Suppl 13:9-13. Jenike MA. An update on obsessive-compulsive disorder. Bull Menninger Clin 2001;65:4-25. Earp JD. Psychosurgery: the position of the Canadian Psychiatric Association. Can Psychiatr Assoc J 1979;24:353-5. Merskey H. Psychosurgery: ethical aspects. Encyclopedia of Bioethics 1995; 2150–3. Fins J. From psychosurgery to neuromodulation and palliation: history’s lesson for the ethical conduct and regulation of neuropsychiatric research. Neurosurg Clin N Am 2003;14:303-19. Loddenkemper T, Pan A, Neme S, et al. Deep brain stimulation in epilepsy. J Clin Neurophysiol 2001;18 (6):514-32. Hodaie M, Wennberg RA, Dostrovsky JO, Lozano AM. Chronic anterior thalamus stimulation for intractable epilepsy. Epilepsia 2002;43(6):603-8. Kumar K, Toth C, Nath RK. Deep brain stimulation for intractable pain: a 15-year experience. Neurosurgery 1997;40(4):736-46. Vercueil L, Krack P, Pollak P. Results of deep brain stimulation for dystonia: a critical reappraisal. Mov Disord 2002;17 Suppl 3:S89-S93. Volkmann J, Benecke R. Deep brain stimulation for dystonia: patient selection and evaluation. Mov Disord 2002;17 Suppl 3:S112-S115. Cortelli P, Guaraldi P, Leone M, Pierangeli G, Barletta G, Grimaldi D, Cevoli S, Bussone G, Baruzzi A, Montagna P. Effect of deep brain stimulation of the posterior hypothalamic area on the cardiovascular system in chronic cluster headache patients. Eur J Neurol 2007;14 (9):1008-15. Starr PA, Barbaro NM, Raskin NH, Ostrem JL. Chronic stimulation of the posterior hypothalamic region for cluster headache: technique and 1-year results in four patients. J Neurosurg 2007;106(6):999-1005 Vonck K, Boon P, Van Roost D. Anatomical and physiological basis and mechanism of action of neurostimulation for epilepsy. Acta Neurochir Suppl 2007;97(Pt 2):321-8. Yamamoto T, Katayama Y, Oshima H, Fukaya C, Kawamata T, Tsubokawa T. Deep brain stimulation therapy for a persistent vegetative state. Acta Neurochir Suppl 2002;79:79-82.
88. Schuurman PR, Bosch DA, Bossuyt PNM, et al. A comparison of continuous thalamic stimulation and thalamotomy for suppression of severe tremor. N Engl J Med 2000;342:461-8. 89. Starr PA, Vitek JL, Bakay RAE. Deep brain stimulation for movement disorders. Neurosurg Clin North Am 1998;9:381-400. 90. Benabid A. What the future holds for deep brain stimulation. Expert Rev Med Devices 2007;4(6):895-903. 91. Axer H, von Keyserlingk D. Mapping of fiber orientation in human internal capsule by means of polarized light and confocal scanning laser microscopy. J Neurosci Methods 2000;94:165-75. 92. Mai J, Assheuer J, Paxinos G. Atlas of the human brain. Orlando: Acadamic Press; 1998. p. 162-72. 93. Heimer L, Switzer RC, Van Hoesen GW. Ventral striatum and ventral pallidum. Compnents of the motor system? Trends Neurosci 1982;5:83-7. 94. Parent A. Comparative neurobiology of the basal ganglia, New York, Toronto: Wiley; 1986. p. 227-52. 95. Graybiel A. The basal ganglia and chunking of action repertoires. Neurobiol Learn Mem 1998;70:119-36. 96. Baxter LR, Clark EC, Iqbal M, Ackerman RE. Corticalsubcortical systems in the mediation of obsessive-compulsive disorder. In: Lichter GL, Cummings JL, editors. Frontal-subcortical circuits in psychiatric and neurological disorders. New York: The Guildford Press; 2001. p. 207-30. 97. Martin LJ, Powers RE, Dellovade TL, Price DL. The bed nucleus-amygdala continuum in human and monkey. J Comp Neurol 1991;309:445-85. 98. Heimer L, Harlan RE, Alheid GF, et al. Substantia innominata: a notion which impedes clinical-anatomical correlations in neuropsychiatric disorders. Neuroscience 1997;76(4):957-1006. 99. Heimer L. A new anatomical framework for neuropsychiatric disorders and drug abuse. Am J Psychiatry 2003;160:1726-39. 100. Carrive P. The periaqueductal gray and defensive behavior: functional representation and neuronal organization. Behav Brain Res 1993;58:27-47. 101. Davis M, Shi S. The extended amygdala: are the central nucleus of the amygdale and the bed nucleus of the stria terminalis differentially involved in fear versus anxiety? Ann N Y Acad Sci 1999;877: 281-91. 102. American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th ed. Washington DC; 1994. 103. Cosyns P, Caemaert J, Haaijman W, et al. Functional stereotactic neurosurgery for psychiatric disorders: an experience in Belgium and The Netherlands. In: Symons L, et al, editors. Advances and technical standards in neurosurgery. New York: Springer Verlag; 1994. p. 242-79. 104. Gybels J, Cosyns P. Cerebral lesions for psychiatric disorders and pain. In: Schmidek H, Sweet WH, editors.
DBS for OCD
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
Operative neurosurgical techniques. 4th ed. Pennsylvania: WB Saunders; 2000. p. 2(126):1660-9. Gabrie¨ls L, Cosyns P, Nuttin B, et al. Deep brain stimulation for treatment-refractory obsessive-compulsive disorder: psychopathological and neuropsychological outcome in 3 cases. Acta Psychiatr Scand 2003:107:1-8. Nuttin B, Gabrie¨ls LA, Cosyns PR, et al. Long-term electrical capsular stimulation in patients with obsessive-compulsive disorder. Neurosurgery 2003;52 (6):1263-74. Gabrie¨ls L, van Kuyck K, Welkenhuysen M, Cosyns P, Nuttin B. Electrical brain stimulation in obsessive-compulsive disorder. In: Tarsy D, Okun M, Starr P, Vitek J, editors. Current clinical neurology, deep brain stimulation in neurological and psychiatric disorders 2008, Springer-Verlag. First MB, Spitzer RL, Gibbon M, et al. Structured clinical interview DSM-IV axis II (SCID-II), version 2. New York: Biometrics Research Department, New York State Psychiatric Institute 1997. Goodman WK, Price LH, Rasmussen SA, et al. The Yale-Brown obsessive-compulsive scale. I. Development, use and reliability. Arch Gen Psychiatry 1989;46:1006-11. Goodman WK, Price LH, Rasmussen SA, et al. The Yale-Brown obsessive-compulsive scale. II. Validity. Arch Gen Psychiatry 1989;46:1012-16. Frances A, Pincus HA, First MB. The Global Assessment of Functioning Scale (GAF). Diagnostic and Statistical Manual of Mental Disorder-IV. Washington DC: American Psychiatric Association; 1994. Lippitz B, Mindus P, Meyerson B, et al. Lesion topography and outcome after thermocapsulotomy and gamma knife capsulotomy for obsessive-compulsive disorder: relevance of the right hemisphere. Neurosurgery 1999;44:452-60. Gabrie¨ls L. Deep brain stimulation in treatment refractory obsessive compulsive disorder. Doctoral dissertation 2004, University of Antwerpen. Nuttin B, Cosyns P, Demeulemeester H, Gybels J, Meyerson B. Electrical stimulation in anterior limbs of internal capsules in patients with obsessive-compulsive disorder. Lancet 1999;30354:(9189):1526. Nuttin B, Gabriels L, van Kuyck K, Cosyns P. Electrical stimulation of the anterior limbs of the internal capsules in patients with severe obsessive-compulsive disorder: anecdotal reports. Neurosurgery Clin North Am 2003;14(2):267-74, Gabrie¨ls L, Nuttin B, Cosyns P. DBS and the treatment of obsessive compulsive disorder. Prog Neurosci Ther Neurostimulation 2007;1(15):57-61. Okun M, Bowers D, Springer U, Shapira N, Malone D, Rezai A, Nuttin B, Heilman K, Morecraft R, Rasmussen S, Greenberg B, Foote K, Goodman WK. What’s in a ‘‘smile?’’ Intra-operative observations of contralateral smiles induced by deep brain stimulation. Neurocase 2004;10(4):271-9.
173
118. Anderson D, Ahmed A. Treatment of patients with intractable obsessive-compulsive disorder with anterior capsular stimulation. Case report. J Neurosurg 2003;98(5):1104-8. 119. Aouizerate B, Cuny E, Martin-Guehl C, Guehl D, Amieva H, Benazzouz A, Fabrigoule C, Allard M, Rougier A, Bioulac B, Tignol J, Burbaud P. Deep brain stimulation of the ventral caudate nucleus in the treatment of obsessive-compulsive disorder and major depression. J Neurosurg 2004;101(4):682-6. 120. Sturm V, Lenartz D, Koulousakis A, Treuer H, Herholz K, Klein JC, Klosterkotter J. The nucleus accumbens: a target for deep brain stimulation in obsessive-compulsive- and anxiety-disorders. J Chem Neuroanat 2003;26(4):293-9. 121. Greenberg B, Malone D, Friehs G, Rezai A, Kubu C, Malloy P, Salloway S, Okun M, Goodman W, Rasmussen S. Three-year outcomes in deep brain stimulation for highly resistant obsessive-compulsive disorder. Neuropsychopharmacology 2006;31(11):2384-93. 122. Hamilton M. A rating scale for depression. J Neurol Neursurg Psychiatry 1960;23:56-62. 123. Guy W. ECDEU assessment manual for psychopharmacology. Rockville, MD: US National Institute of Health, Psychopharmacology research Brand; 1976. 124. Williams J. A structural interview guide for the Hamilton Depression Rating Scale. Arch Gen Psychiatry 1988;45(8):742-7. 125. Hamilton M. The assessment of anxiety states by rating. Br J Med Psychol 1959;32:50-5. 126. Beck AT, Ward CH, Mendelson M, et al. An inventory for measuring depression. Arch Gen Psychiatry 1961;4:53-63. 127. Beck AT, Steer RA, Garbin MG. Psychometric properties of the beck depression inventory: twenty-five years of evaluation. Clin Psychol Rev 1988;8:77-100. 128. Shacham S. A shortened version of the Profile of Mood States. J Pers Assess 1983;47(3):305-6. 129. Steketee G, Chambless D, Tran G, Worden H, Gillis M. Behavioral avoidance test for obsessive compulsive disorder. Behav Res Ther 1996;34:73-84. 130. Hunka K, Suchowersky O, Wood S, Derwent L, Kiss ZH. Nursing time to program and assess deep brain stimulators in movement disorder patients. J Neurosci Nurs 2005;37(4):204-10. 131. Laitinen LV. Emotional Responses to subcortical electrical stimulation in psychiatric patients. Clin Neurol Neurosurg 1979;81(3):148-57. 132. Rizzone M, Lanotte M, Tavella A, et al. Deep brain stimulation of the subthalamic nucleus in Parkinson’s disease: effects of variation in stimulation parameters. J Neurol Neurosurg Psychiatry 2001;71:215-19. 133. Holsheimer J, Dijkstra EA, Demeulemeester H, Nuttin B. Chronaxie calculated from current-duration and voltage-duration data. J Neurosc Meth 2000;97:45-50.
2923
2924
173
DBS for OCD
134. Holsheimer J, Demeulemeester H, Nuttin B, de Sutter P. Identification of the target neuronal elements in deep brain stimulation. Eur J Neurosci 2000;12:4573-7. 135. Ranck J. Which elements are excited in electrical stimulation of mammalian central nervous system: a review. Brain Res 1975;98:417-40. 136. Butson CR, McIntyre CC. Tissue and electrode capacitance reduce neural activation volumes during deep brain stimulation. Clin Neurophysiol 2005;116:2490-500.
137. Butson CR, Maks CB, McIntyre CC. Sources and effects of electrode impedance during deep brain stimulation. Clin Neurophysiol 2006;117(2):447-54. 138. Mcintyre C, Butson C, Maks C, Noecker A. Optimizing deep brain stimulation parameter selection with detailed models of the electrode-tissue interface. 28th Annual International Conference of the IEEE. EMBS’ 06. Engineering in Medicine and Biology Society; 2006.
176 Deep Brain Stimulation for Depression C. Hamani . B. Snyder . A. Laxton . A. Lozano
Introduction
Deep Brain Stimulation
Major depression (MDD) is a common psychiatric disorder with a 6 months prevalence of approximately 5% [1]. The initial therapeutic approach for patients with depression comprises the use of medication and/or psychotherapy. In patients who continue to be disabled despite these measures, suitable alternatives are the use of other medications from the same or different drug classes, augmentative pharmacological regimes, the combination of antidepressant agents from different classes and eventually electroconvulsive therapy (ECT). The main objectives of treatment are the remission of symptoms, restoration of daily function and the prevention of relapses and recurrences. While the majority of patients respond favorably to treatment, 10–20% will have chronic and treatment resistant forms of the disease. For this refractory population, surgical treatment has been used as part of the therapeutic armamentarium. Modern neurosurgical management of depression has mainly taken three forms: Ablative procedures designed to disrupt presumed pathologically active circuits, deep brain stimulation (DBS) and vagal nerve stimulation (VNS), with cortical stimulation also being investigated. In this chapter, we will review the clinical outcome of DBS. The diagnostic criteria of major depression, indications for surgical procedures, outcome with stereotactic lesions and VNS will described in detail in other chapters of this book.
Deep brain stimulation involves the delivery of electrical current directly into the brain parenchyma through implanted electrodes. This enables the modulation of pathological neural circuits and avoids the need for tissue destruction (as is the case for stereotactic radiofrequency lesions). The main advantages of the technique are that the therapy is titratable and also potentially reversible. This is relevant if one envisages that newer treatments may become available in the future and that most of the side effects of DBS can be managed by adjusting stimulation parameters (e.g., decreasing the intensity of current to avoid adverse effects). Within the last decades, DBS has been extensively used for the treatment of movement disorders and pain [2–9]. In addition, DBS is currently under investigation for a number of other disorders including: epilepsy [10–17], minimally conscious states [18], aggressiveness [19], cluster headache [20,21], obesity [22], memory modulation [22], and psychiatric disorders, including obsessive compulsion disorder (OCD) [23–27] and major depression as discussed here (see below).
#
Springer-Verlag Berlin/Heidelberg 2009
DBS Targets for Depression DBS targets proposed to date for the surgical treatment of depression are the subcallosal cingulate gyrus (including Brodmann area 25), the
2954
176
Deep brain stimulation for depression
inferior thalamic peduncle, the nucleus accumbens and the anterior capsule.
Subcallosal Cingulate Gyrus (SCG) The rationale for deep brain stimulation in the SCG to treat major depression was based upon observations derived from imaging studies. BA 25 cerebral blood flow increases when healthy subjects are asked to rehearse autobiographic sad scripts or upon rapid tryptophane depletion [28–32]. In patients with depression, baseline metabolic activity is increased in BA25 and decreased in BA 46/9 [30,33,34]. After treatment with antidepressants, behavioral therapy, or ECT, this pattern is reversed with a reduction in BA25 and an increase BA46/9 activity [30,33–35]. In this context, SCG DBS was proposed to modulate the increased activity observed in the subgenual cingulum of patients with depression. To be included in the initial trial, patients had to have major depressive disorder diagnosed according to DSMIV criteria. Their last major depressive episode had to be of at least 1-year duration. Hamilton Depression Rating Scale (HAMD)-17 scores had to be >20 and Global Assessment of Function scores (GAF) 50. Also very important, patients had to be treatmentresistant. In our trial, this was defined as a failure to respond to four different anti-depressants treatments, including medications and evidence-based psychotherapy. Though most patients had previous ECT, failure to this therapy was not a requirement for defining treatment resistance. Due to the risk of memory deficits, patients sometimes refused to undergo ECT and did not consent with this therapy. Also excluded were patients with overt manic features, suicidal plans during the recruitment phase, other Axis I or II disorders, or neurological or clinical conditions that could interfere with the technique or safety of the surgical procedure (i.e., coagulopathy).
The surgical procedure for implanting SCG electrodes in our institution is similar to that used for other Functional Neurosurgical conditions. A Leksell frame is applied under local anesthesia, after which patients undergo a stereotactic MRI scan. For targeting, we initially identify the cingulate region lying ventral to the genu of the corpus callosum on reconstructed sagittal images. This often corresponds to a coronal section in which the initial aspect of the anterior horns of the lateral ventricles can be visualized. In the mediolateral plane, the selected target is the transition between grey and white matter of the SCG. Bilateral burr holes are then made under local anesthesia anterior to the coronal suture approximately 2 cm lateral to the midline. Microelectrode recording were not essential for the procedure. However, this technique helped us localizing the junction of grey and white matter of the superior and inferior banks of the SCG cortex. While in grey matter we can record activity of isolated cell and local field potentials, the white matter is often electrophysiologically silent. Once the physiological target is determined, DBS quadripolar electrodes are implanted in this region (> Figure 176-1). We have begun to shy away from intra-operative macrostimlation. Our early experience with this demonstrated a sub-set of patients who responded with rather dramatic effects to intraoperatively stimulation (approximately 70%). However, there remained a cohort of patients that failed to display any effect but improved with chronic stimulation. In our experience the emotional setback to the patients that did not show an acute response during intra-operative trials warrant the use of macrostimulation only in specific circumstances. After removal of the head-frame we implant the IPG in the right subclavicular region on the same day under general anesthesia. Placement of the electrodes is confirmed by post-operative MRI imaging (> Figure 176-1). Patients are discharged 1–3 days post-operatively.
Deep brain stimulation for depression
176
. Figure 176-1 Location of DBS electrodes implanted in the subcallosal cingulate gyrus (SCG). Sagittal (a) and coronal (b) views of the SCG (circles) represented on a stereotactic neurosurgical atlas. Sagittal (c) and coronal (d) images of the SCG DBS mapped on T2 MRI images. Sagittal (e) and coronal (f) MRI sections of DBS electrodes implanted into the SCG. sgCg, subgenual cingulate; cc, corpus callosum; g, genu of the corpus callosum; ac, anterior commissure (from [35] reprinted with permission)
We consider a 50% improvement in HAMD scores as a clinically significant response. Initially, 4/6 of patients (66%) met the criteria of response at 6 months. We have now operated on 30 patients and these overall results have been sustained at long-term. At 12 months, 11 out of 20 patients (55%) responded to the procedure and 35% achieved or were within one point of remission (scoring 8 or less on the HAMD-17 scale) [36]. Though the effects of stimulation tended to build to with time, at
1 month postoperatively most long-term responders have already shown significant reductions in HAMD scores [36]. In our 20 patients followed for 1 year or more, the effects of stimulation reached a plateau at 6 months, with no major changes in outcome thereafter [36]. Twelve months after DBS, patients had a significant improvement in all subscores of the HAMD scale, including mood, anxiety, sleep and somatic [36]. The most common parameters used during chronic stimulation are 3–4 V, 60 ms of pulse
2955
2956
176
Deep brain stimulation for depression
width and 130 Hz [35,36]. Side effects related to the surgical procedure or hardware implants were similar to those with DBS in other targets (e.g., hardware infections, pain in the site of the pulse generator, etc) [36]. Comprehensive neuropsychological assessment 12 months after stimulation onset did not reveal any adverse effects in patients treated with SCG DBS [37]. Pre-treatment PET scanning in our series revealed an increased activity in BA25 and a decreased activity in prefrontal and premotor cortices, the dorsal anterior cingulate gyrus and anterior insula. This pattern was reversed as early as 8 weeks after stimulation onset in patients that improved with DBS [35]. Now that we have experience with 30 patients, the idea is to characterize clinical or imaging features capable of predicting surgical outcome. In a recent study using tractography, Johansen-Berg and colleagues investigated the potential connectivity of cingulate regions targeted for stimulation in our patients [38]. Two main regions were defined: The pregenual (pACC) and subgenual anterior cingulate cortices (sACC). Common areas of connectivity were the midcingulate cortex, the frontal pole, hypothalamus, and nucleus accumbens. Major differences between pACC and sACC were that the former connected more strongly with the frontal pole, whereas the later had more strong connections with the orbitofrontal cortex and medial temporal lobe structures [38]. Based on these findings, we investigated whether differences in placement of the electrodes along the anteroposterior axis could lead to a variable in outcome after SCG DBS. We found no correlation between location of contacts used for stimulation and postoperative HAMD scores and no meaningful differences in the location of electrodes in patients that responded or did not respond to the procedure (unpublished data). These are early days in the field of DBS for depression. We expect that with increased experience and additional centers conducting clinical
trials, prognostic factors for the use of SCG DBS will be characterized.
Inferior Thalamic Peduncle Only one patient with treatment resistant depression treated with inferior thalamic peduncle (ITP) DBS has been reported to date [39,40]. The rationale for targeting the ITP is that this system of fibers conveys projections from intralaminar and midline thalamic nuclei to the orbitofrontal cortex [41]. Initially, the authors implanted eight contact electrodes into the target in such a way that contacts 1–2 would be in the ventromedial hypothalamus, contacts 3–4 close to the fornix, contacts 5–6 in the vicinity of the ITP, and contacts 7–8 within the region of the nucleus reticularis polaris. Bipolar stimulation through the ventral contacts [2,10,23] induced a series of adverse effects, including vertical nystagmus, anxiety and autonomic features (e.g., an increase in heart rate and blood pressure). No stimulation-induced effects were observed when the other electrode contacts were activated. After testing, the initially implanted electrodes were replaced by quadripolar electrodes for chronic stimulation (> Figure 176-2). The patient had an insertional effect for 1 week, with a significant reduction in the symptomatology of depression without any stimulation being delivered. This was demonstrated by the decrease in HAMD scores from 42 at baseline to 3 immediately after the placement of the electrodes into target. After the first week, depression partially relapsed and the electrodes were then turned ‘‘on’’ at 2.5 V, 450 ms and 130 Hz. The patient was assessed in a series of followup visits during the first 8 months of stimulation. During that time, a significant clinical improvement was noticed with HAMD scores varying from 2 to 8. Thereafter, the patient underwent a double-blinded evaluation in which the electrodes were turned ‘‘off’’ for 12 months. This led to
Deep brain stimulation for depression
176
. Figure 176-2 Location of DBS electrodes implanted in the inferior thalamic peduncle (ITP). Axial (a) and coronal (b) MRI sections with DBS electrodes implanted in the ITP. Figures on the right are schematic representations of the location of the contacts on a stereotactic neurosurgical atlas. Cma, anterior commissure; Fx, fornix; Put, putamen; Pl, lateral pallidus; Pm, medial pallidus; Pme, medial pallidus externus; Cpip, posterior branch of internal capsule; Zi, zona incerta; Raprl, prelemniscal radiations; Pu, pulvinar; Cd, caudate nucleus; Cpig, genu of internal capsule; ITP, inferior thalamic peduncle (Pdthif); Vm, nucleus ventralis hypothalami; An Pd, ansa lenticularis; B, Meynert’s basal nucleus; II, optic tract; R, right electrode; L, left electrode (from [39] reprinted with permission)
significant fluctuations in HAMD scores (from 2 to almost 20) and a deterioration of Global Assessment of Function scores. Stimulation was recommenced and carried on for an additional 4 months with recapture of the initial benefits.
Nucleus Accumbens According to the authors, DBS in the nucleus accumbens was based on the following lines of reasoning [41]: The structure is implicated in
mechanisms of reward. It acts as a ‘‘motivation gateway’’ between systems involved in emotion and motor control. Finally, the ventral striatum is in an anatomical location to modulate activity of other brain regions involved in mechanisms of depression [41]. To date, only the short-term follow-up of three patients has been reported. All had severe MDD (HAMD 24 scores of 38, 31 and 32), and were refractory to multiple drugs, augmentation therapy, and ECT. The electrodes were implanted in such a way that the distal contacts were in the
2957
2958
176
Deep brain stimulation for depression
. Figure 176-3 Location of DBS electrodes implanted in the nucleus accumbens. (Upper row) Lowest contacts of DBS electrodes in the horizontal and coronal planes. (Lower row) Actual position of the electrode leads in post-operative control X-rays (from [41] reprinted with permission)
shell and core of the accumbens (> Figure 176-3). Patients were unaware as to whether the electrodes were ‘‘on’’ or ‘‘off’’ but some reported an increased reward-seeking motivation during stimulation (i.e., wanted to travel to another city). Outcome measurements were acquired in a double-blinded fashion. The authors reported that all three patients had some improvement with stimulation. Only 1 of the 3 patients however, had a >50% reduction in HAMD scores at the last follow up (6, 8 and 23 weeks postoperatively). In that study, stimulation parameters were 145 Hz, 90 fY´s of pulse width and voltages of up to 4–5 V. In addition to clinical outcome, the circuitry modulated by DBS in the nucleus accumbens
was also investigated with PET at baseline and after 1 week of stimulation. The authors reported an increased metabolic activity within the nucleus accumbens, dorsolateral prefrontal cortex, and amygdala, and a decreased activity in the medial prefrontal cortex and caudate nucleus during stimulation.
Anterior Capsule The rational for anterior capsule stimulation to treat depression was based on the improvements in mood observed in patients with obsessive compulsive disorder treated with DBS in this same region [25–27]. In addition, the anterior
Deep brain stimulation for depression
capsule has been used as a target for lesions (capsulotomy) for decades. Despite of the promising results and anecdotal reports in meetings, no studies investigating the effects of anterior capsule DBS for depression have been published so far. Results of trials that are currently on the way are much awaited.
Conclusions The recent employment of neuromodulation techniques in clinical practice, including deep brain stimulation and vagal nerve stimulation, has opened up a new era in the neurosurgical management of psychiatric illness. We are now reaching a point where science and imaging technology are allowing for a safe, and measured approach to the surgical treatment of these diseases. The most appealing aspect of these procedures is that they are both titratable and potentially reversible. With that in mind, patients might be assured an added measure of safety. With the renewed interest in surgery for psychiatric disorders, it is important to stress that for an appropriate surgical management of psychiatric conditions the following is necessary: Patients have to have a clear diagnosis of the condition to be treated. Be refractory to all conventional reasonable therapies. Clinical management has to be multidisciplinary, including psychiatrists, neurosurgeons and psychologists. Adequate scales and quantitative measures should be applied before and after surgery to clearly establish the outcome of the procedures. Surgery for psychiatric disorders still suffers from the stigma it acquired in the era of the lobotomies. While we have crawled out from those days, the future of psychosurgery largely depends on the manner in which the trials are conducted. In the nearby future, studies with a higher number of patients conducted on a double-blinded fashion with sham and/or placebo
176
arms will be necessary. This will help to characterize the role of neuromodulation therapies in the treatment of major depression.
References 1. Depression Guideline Panel, (ed). Depression in primary care: detection and diagnosis, vol. 1. Rockville, MD: US Department of Health and Human Services, Public Health Service, Agency for Health Care Policy and Research; 1993. 2. Deep-brain stimulation for Parkinson’s disease study group. Deep-brain stimulation of the subthalamic nucleus or the pars interna of the globus pallidus in Parkinson’s disease. N Engl J Med 2001;345:956-63. 3. Deuschl G, Schade-Brittinger C, Krack P, Volkmann J, Schafer H, Botzel K, Daniels C, Deutschlander A, Dillmann U, Eisner W, Gruber D, Hamel W, Herzog J, Hilker R, Klebe S, Kloss M, Koy J, Krause M, Kupsch A, Lorenz D, Lorenzl S, Mehdorn HM, Moringlane JR, Oertel W, Pinsker MO, Reichmann H, Reuss A, Schneider GH, Schnitzler A, Steude U, Sturm V, Timmermann L, Tronnier V, Trottenberg T, Wojtecki L, Wolf E, Poewe W, Voges J. A randomized trial of deepbrain stimulation for Parkinson’s disease. N Engl J Med 2006;355:896-908. 4. Kupsch A, Benecke R, Muller J, Trottenberg T, Schneider GH, Poewe W, Eisner W, Wolters A, Muller JU, Deuschl G, Pinsker MO, Skogseid IM, Roeste GK, VollmerHaase J, Brentrup A, Krause M, Tronnier V, Schnitzler A, Voges J, Nikkhah G, Vesper J, Naumann M, Volkmann J. Pallidal deep-brain stimulation in primary generalized or segmental dystonia. N Engl J Med 2006;355:1978-90. 5. Levy RM. Deep brain stimulation for the treatment of intractable pain. Neurosurg Clin N Am 2003;14: 389-99, vi. 6. Levy RM, Lamb S, Adams JE. Treatment of chronic pain by deep brain stimulation: long term follow-up and review of the literature. Neurosurgery 1987;21:885-93. 7. Schuurman PR, Bosch DA, Bossuyt PM, Bonsel GJ, van Someren EJ, de Bie RM, Merkus MP, Speelman JD. A comparison of continuous thalamic stimulation and thalamotomy for suppression of severe tremor. N Engl J Med 2000;342:461-8. 8. Vidailhet M, Vercueil L, Houeto JL, Krystkowiak P, Benabid AL, Cornu P, Lagrange C, Tezenas du Montcel S, Dormont D, Grand S, Blond S, Detante O, Pillon B, Ardouin C, Agid Y, Destee A, Pollak P. Bilateral deepbrain stimulation of the globus pallidus in primary generalized dystonia. N Engl J Med 2005;352:459-67.
2959
2960
176
Deep brain stimulation for depression
9. Wallace BA, Ashkan K, Benabid AL. Deep brain stimulation for the treatment of chronic, intractable pain. Neurosurg Clin N Am 2004;15:343-57, vii. 10. Andrade DM, Zumsteg D, Hamani C, Hodaie M, Sarkissian S, Lozano AM, Wennberg RA. Long-term followup of patients with thalamic deep brain stimulation for epilepsy. Neurology 2006;66:1571-3. 11. Benabid AL, Koudsie A, Benazzouz A, Vercueil L, Fraix V, Chabardes S, Lebas JF, Pollak P. Deep brain stimulation of the corpus luysi (subthalamic nucleus) and other targets in Parkinson’s disease. Extension to new indications such as dystonia and epilepsy. J neurol 2001;248 Suppl 3:III37-47. 12. Chabardes S, Kahane P, Minotti L, Koudsie A, Hirsch E, Benabid AL. Deep brain stimulation in epilepsy with particular reference to the subthalamic nucleus. Epileptic Disord 2002;4 Suppl 3:S83-93. 13. Fisher RS, Uematsu S, Krauss GL, Cysyk BJ, McPherson R, Lesser RP, Gordon B, Schwerdt P, Rise M. Placebocontrolled pilot study of centromedian thalamic stimulation in treatment of intractable seizures. Epilepsia 1992;33:841-51. 14. Hodaie M, Wennberg RA, Dostrovsky JO, Lozano AM. Chronic anterior thalamus stimulation for intractable epilepsy. Epilepsia 2002;43:603-8. 15. Kerrigan JF, Litt B, Fisher RS, Cranstoun S, French JA, Blum DE, Dichter M, Shetter A, Baltuch G, Jaggi J, Krone S, Brodie M, Rise M, Graves N. Electrical stimulation of the anterior nucleus of the thalamus for the treatment of intractable epilepsy. Epilepsia 2004;45:346-54. 16. Velasco AL, Velasco F, Velasco M, Trejo D, Castro G, Carrillo-Ruiz JD. Electrical stimulation of the hippocampal epileptic foci for seizure control: a doubleblind, long-term follow-up study. Epilepsia 2007;48: 1895-903. 17. Velasco F, Velasco M, Velasco AL, Jimenez F, Marquez I, Rise M. Electrical stimulation of the centromedian thalamic nucleus in control of seizures: long-term studies. Epilepsia 1995;36:63-71. 18. Schiff ND, Giacino JT, Kalmar K, Victor JD, Baker K, Gerber M, Fritz B, Eisenberg B, O’Connor J, Kobylarz EJ, Farris S, Machado A, McCagg C, Plum F, Fins JJ, Rezai AR. Behavioural improvements with thalamic stimulation after severe traumatic brain injury. Nature 2007;448:600-3. 19. Franzini A, Marras C, Ferroli P, Bugiani O, Broggi G. Stimulation of the posterior hypothalamus for medically intractable impulsive and violent behavior. Stereotact Funct Neurosurg 2005;83:63-6. 20. Franzini A, Ferroli P, Leone M, Broggi G. Stimulation of the posterior hypothalamus for treatment of chronic intractable cluster headaches: first reported series. Neurosurgery 2003;52:1095-9; discussion 1099-101.
21. Leone M, Franzini A, Broggi G, May A, Bussone G. Long-term follow-up of bilateral hypothalamic stimulation for intractable cluster headache. Brain 2004; 127:2259-64. 22. Hamani C, McAndrews MP, Cohn M, Oh M, Zumsteg D, Shapiro CM, Wennberg RA, Lozano AM. Memory enhancement induced by hypothalamic/fornix deep brain stimulation. Ann Neurol 2008;63:119-23. 23. Anderson D, Ahmed A. Treatment of patients with intractable obsessive-compulsive disorder with anterior capsular stimulation. Case report. J Neurosurg 2003; 98:1104-8. 24. Aouizerate B, Martin-Guehl C, Cuny E, Guehl D, Amieva H, Benazzouz A, Fabrigoule C, Bioulac B, Tignol J, Burbaud P. Deep brain stimulation for OCD and major depression. Am J Psychiatry 2005; 162:2192. 25. Greenberg BD, Malone DA, Friehs GM, Rezai AR, Kubu CS, Malloy PF, Salloway SP, Okun MS, Goodman WK, Rasmussen SA. Three-year outcomes in deep brain stimulation for highly resistant obsessive-compulsive disorder. Neuropsychopharmacology 2006;31:2384-93. 26. Nuttin B, Cosyns P, Demeulemeester H, Gybels J, Meyerson B. Electrical stimulation in anterior limbs of internal capsules in patients with obsessive-compulsive disorder. Lancet 1999;354:1526. 27. Nuttin BJ, Gabriels LA, Cosyns PR, Meyerson BA, Andreewitch S, Sunaert SG, Maes AF, Dupont PJ, Gybels JM, Gielen F, Demeulemeester HG. Long-term electrical capsular stimulation in patients with obsessivecompulsive disorder. Neurosurgery 2003;52:1263-72; discussion 1272-4. 28. Damasio AR, Grabowski TJ, Bechara A, Damasio H, Ponto LL, Parvizi J, Hichwa RD. Subcortical and cortical brain activity during the feeling of self-generated emotions. Nat Neurosci 2000;3:1049-56. 29. George MS, Ketter TA, Parekh PI, Horwitz B, Herscovitch P, Post RM. Brain activity during transient sadness and happiness in healthy women. Am J Psychiatry 1995;152:341-51. 30. Mayberg HS, Liotti M, Brannan SK, McGinnis S, Mahurin RK, Jerabek PA, Silva JA, Tekell JL, Martin CC, Lancaster JL, Fox PT. Reciprocal limbic-cortical function and negative mood: converging PET findings in depression and normal sadness. Am J Psychiatry 1999;156:675-82. 31. Pardo JV, Pardo PJ, Raichle ME. Neural correlates of selfinduced dysphoria. Am J Psychiatry 1993;150:713-19. 32. Talbot PS, Cooper SJ. Anterior cingulate and subgenual prefrontal blood flow changes following tryptophan depletion in healthy males. Neuropsychopharmacology 2006;31:1757-67. 33. Mayberg HS. Modulating dysfunctional limbic-cortical circuits in depression: towards development of brain-
Deep brain stimulation for depression
34.
35.
36.
37.
38.
based algorithms for diagnosis and optimised treatment. Br Med Bull 2003;65:193-207. Seminowicz DA, Mayberg HS, McIntosh AR, Goldapple K, Kennedy S, Segal Z, Rafi-Tari S. Limbic-frontal circuitry in major depression: a path modeling metanalysis. Neuroimage 2004;22:409-18. Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, Schwalb JM, Kennedy SH. Deep brain stimulation for treatment-resistant depression. Neuron 2005;45:651-60. Lozano AM, Mayberg HS, Giacobbe P, Hamani C, Craddock RC, Kennedy SH. Subcallosal cingulate gyrus deep brain stimulation for treatment resistant depression. Biol Psychiatry 2008; (in press). McNeely HE, Mayberg HS, Lozano AM, Kennedy SH. Neuropsychological impact of Cg25 deep brain stimulation for treatment-resistant depression: preliminary results over 12 months. J Nerv Ment Dis 2008;196:405-10. Johansen-Berg H, Gutman DA, Behrens TE, Matthews PM, Rushworth MF, Katz E, Lozano AM, Mayberg HS.
176
Anatomical connectivity of the subgenual cingulate region targeted with deep brain stimulation for treatment-resistant depression. Cereb Cortex 2008;18:1374-83. 39. Jimenez F, Velasco F, Salin-Pascual R, Hernandez JA, Velasco M, Criales JL, Nicolini H. A patient with a resistant major depression disorder treated with deep brain stimulation in the inferior thalamic peduncle. Neurosurgery 2005;57:585-93; 40. Jimenez F, Velasco F, Salin-Pascual R, Velasco M, Nicolini H, Velasco AL, Castro G. Neuromodulation of the inferior thalamic peduncle for major depression and obsessive compulsive disorder. Acta Neurochirurgica 2007;97:393-8. 41. Schlaepfer TE, Cohen MX, Frick C, Kosel M, Brodesser D, Axmacher N, Joe AY, Kreft M, Lenartz D, Sturm V. Deep brain stimulation to reward circuitry alleviates anhedonia in refractory major depression. Neuropsychopharmacology 2008;33:368-77.
2961
Psychiatric Surgery
170 Ethical Considerations in Psychiatric Surgery B. S. Appleby . P. V. Rabins
The History of Neurosurgery for Psychiatric Illnesses Surgical treatment for mental illnesses most likely began when burr holes were placed in the skulls of individuals over a millennium ago. Skulls with trephination lesions have been uncovered at multiple sites by archeologists, suggesting that this practice arose independently several times throughout history. In the late nineteenth century, experimental intracranial ablative surgery was performed in Germany, but there was little evidence of its efficacy and the practice did not become widespread. In the mid 1930s, a Portuguese neurologist, Egas Moniz, learned of a report by Yale physiologists Jacobsen and Fulton, that destruction of the primate frontal lobe had a calming effect on chimpanzee behavior, a finding that paralleled reports of decreased anxiety following injuries to the frontal lobes in humans. Walter Freeman, the chair of neurology at George Washington University, reviewed Fulton’s research and thought that the results of the surgery were due to the destruction of bilateral association areas. He devised techniques to perform this operation in humans and popularized the procedure in the 1940s with neurosurgeon James Watts. The first patients to undergo the operation suffered from severe depression and improvement was observed soon after patients recovered from the surgery. However, schizophrenia soon became the main indication and by the early 1950s more than 20,000 procedures had been performed. With the exception of a few specialist referral centers throughout the world, the procedure was #
Springer-Verlag Berlin/Heidelberg 2009
essentially abandoned for the treatment of psychiatric illnesses in the early 1950s mainly because efficacy could not be shown for schizophrenia. Many individuals had been operated on either against their will or without consent procedures in place at the time. Additionally, the methods later advocated and used by Freeman appeared to be used in an unsafe and cavalier manner. In 1950, psychiatrist Robert Heath placed recording electrodes in the brains of patients with schizophrenia and eventually performed similar procedures in other patients with psychiatric and behavioral disorders. Acting on the basis of reports by Olds and Milner that stimulation of the median forebrain bundle induced pleasure in rats, Heath found that patients with schizophrenia and depression reported pleasurable mood states during stimulation that was not experienced when the stimulator was deactivated. This research continued until 1970 when it was abandoned because of lack of funding and psychosocial reasons.
Psychiatric Aspects of DBS for Movement Disorders Ablative basal ganglia surgery for the treatment of Parkinson’s disease (PD) began in the 1940s and has continued into the current century. As the electrophysiology of this region was delineated though the experimental induction of parkinsonism in rats, clinical trials demonstrated improvements in parkinsonian tremor with deep brain stimulation (DBS) of the subthalamic nucleus (STN) in humans. By the 1980s, DBS
2856
170
Ethical considerations in psychiatric surgery
was used experimentally to treat PD and other movement disorders (e.g., essential tremor and primary dystonia). DBS has since demonstrated efficacy and safety in the treatment of several movement disorders and many researchers and clinicians have observed unintended neuropsychiatric effects of this treatment. Initial case reports and case series noted alterations in patients’ mood while being treated with DBS for movement disorders. These reports described abrupt changes in mood states that resulted in depression and mania. A widely publicized report in the New England Journal of Medicine described a woman with PD who displayed acute tearfulness and hopelessness following the implantation and activation of bilateral electrodes in the STN that abruptly ceased upon the electrodes’ deactivation [1]. Several reports describing similar phenomenology followed and data obtained from this body of literature suggest that affective changes such as depression, mania, hypomania, and anxiety, occur in 3.2–6.2% of patients treated with DBS and does not appear to differ significantly by treatment indication or surgical site [2]. Besides the mood effects of DBS, other neuropsychiatric adverse events were also observed including delirium (4–8%), psychosis (0.6–1.2%), and apathy (0.3–0.6%). Cognitive decline was also studied in patients receiving DBS and it does not appear to contribute to global cognitive impairment. Several studies have reported a decline in verbal fluency but this is not thought to be clinically significant and do not appear to vary significantly by surgical site or indication. One troubling adverse event is the possible increased prevalence of suicide in patients receiving DBS. Eleven suicides have been reported in the literature up until 2006 and suicidal ideation and/or attempts have been reported in 0.3–0.7% of patients following DBS surgery. Completed suicides have been reported for all of the common indications and implantation sites, but most have been reported in patients with PD
with STN implants, most likely because this group comprises the majority of reported DBS cases. The aforementioned suicide rate in a population consisting mostly of patients with PD is a cause for concern because patients with PD typically have an annual suicide rate that is ten times lower than the average population. Although cases are limited, the majority of individuals who committed suicide did so on an average of 2.4 years following surgery and had been judged to have had a favorable neurological outcome by clinicians. The prevalence of delayed suicidal behavior in patients receiving DBS despite positive motor outcomes has prompted many clinicians to consider the psychosocial ramifications of DBS. DBS has revolutionized the management of PD and it often has a profound effect on individuals and their families. Often having to rely on others for basic needs, many patients regain much of their independence following DBS surgery, which may require significant psychosocial adjustment. Many patients find it difficult to view themselves outside of the sick role as many aspects of their life had been devoted to coping with their disease. Despite regaining occupational capacity, several patients do not choose to return to work as they still regard themselves as disabled or continue to experience neuropsychiatric impairments despite an improvement in their motor function. These important adjustments have received little attention and but are currently being researched to ascertain the follow-up needs of patients who experience life-altering changes following DBS surgery. As researchers and clinicians began to acknowledge the neuropsychiatric manifestations of DBS, studies began to measure these effects using standardized psychiatric instruments. Many studies included depression, cognition, anxiety, and OCD symptoms as secondary outcome measures and with the exception of cognition, which largely remained unchanged, neuropsychiatric symptoms generally improved with DBS treatment. Because
Ethical considerations in psychiatric surgery
these outcome measures were recorded in individuals with concomitant neurological diseases that often display neuropsychiatric manifestations, it remained unclear whether improvements in neuropsychiatric symptoms were due to the treatment of the underlying disease or some other factor. Nevertheless, adverse events and secondary outcome data from studies of DBS in movement disorders were used in the evaluation of DBS for the possible treatment of several psychiatric disorders.
Scientific Investigation of Ablative Surgery Prior to1980 One of the major concerns in the field of psychiatric neurosurgery has been a lack of scientifically valid clinical research studies demonstrating safety and efficacy of this alternative treatment. In part, this is because modern clinical trial methods were not developed until the 1950s and the diseases being treated were so severe that randomized trials were not possible. The nature of the surgery also makes blinded evaluation impossible. However, case series published in the 1950s–1980s reported on several different surgical sites but with few comparison studies because the use of cingulotomy and capsulotomy procedures is largely determined by country. Research has also been confounded by the use of outcome raters who also members of the treatment team, introducing the likelihood of rater bias. Furthermore, longterm outcome studies have been lacking until recently, and are crucial for determining the long-term efficacy and complications of these procedures. Reports of clinical efficacy are particularly difficult to assess prior to standardized diagnostic and outcome criteria. Several mid-century studies report neurosurgical results of patients with various psychiatric symptoms but most do not describe their diagnostic criteria. Many of
170
these concerns were addressed in the 1960s and 1970s by committees such as the United States’ National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research due to concerns of efficacy and safety measures as well as its abuse potential. The introduction of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III) and the International Statistical Classification of Diseases and Related Health Problems (ICD) in 1980 provided standardized diagnostic criteria with established reliability.
Current Psychiatric Indications for Neurosurgery Although psychotropic medications have improved our ability to treat psychiatric illnesses, these disorders continue to have a large impact on public health. In 1997, the Global Burden of Disease Study included five psychiatric illnesses in the top thirty conditions that cause the most Disability Adjusted Life Years (DALY’s), defined as the amount of life lost due to premature mortality and disability. Unipolar depression was the fourth leading cause of DALY’s worldwide regardless of gender. Although many individuals maintain functional ability with a concomitant psychiatric illness, a large proportion of patients are significantly impaired in functional, psychosocial, and occupational domains. Neurosurgical treatment of psychiatric illnesses was partly initiated for treatment-refractory cases, though there were limited therapeutic options at the time of its initiation and many psychosocial issues contributed to its widespread utilization in the 1940s and 1950s. A strong body of neuropathological, radiological, laboratory, and treatment response data demonstrates that major mental illnesses such as depression, schizophrenia, and OCD, are caused by a functional and/or structural impairment of the brain and thus have a strong biological etiology.
2857
2858
170
Ethical considerations in psychiatric surgery
Because of the permanence of surgical treatment, the possibility of spontaneous recovery over time in some mental illnesses, and the usually long projected lifespan of patients, neurosurgical procedures for psychiatric disorders are only considered in chronic, treatment-refractory, and extremely disabling cases. The concept of ‘‘treatment-refractoriness’’ typically refers to the failure of an adequate amount of trials of non-surgical therapies including medications, psychotherapy, and other somatic treatments such as light therapy, electroconvulsive therapy (ECT), repetitive transmagnetic stimulation (rTMS), and vagal nerve stimulation (VNS). There are no consensus criteria for the duration, type, or number of trials that are required prior to receiving surgery. Severity is often assessed by examining the patient’s level of personal distress and decline in functional abilities. A multi-disciplinary team consisting of psychiatrists, neurologists, and neurosurgeons evaluates patients to ensure the adequacy of prior treatments and the appropriateness of the patient for the procedure. Some centers require consent from both the patient and a close family member prior to the procedure. Patients must also agree to continue therapy following the operation.
. Table 170-1 DSM-IV criteria for a major depressive episode
Depression
adjustment disorder with depressed features, bereavement, and demoralization. Neurosurgery is reserved for patients with major depressive episodes who have experienced chronic debilitating symptoms and conventional treatment resistance. Thirty to forty percent of patients with major depression will fail a trial of an antidepressant medication. A large proportion of these patients are inadequately treated, have an underlying medical condition, are taking depression-inducing substances, or are non-compliant with treatment. There is no standard definition for treatmentresistant depression, though various operational and staging criteria have been suggested in the literature. Many experts establish the diagnosis of treatment-resistant depression following adequate
The lifetime prevalence of affective disorders is 19.3% and major depressive episodes comprise the majority of cases (17%). Depression is one of the leading causes of DALYs and is one of the leading causes of disability in general. A contributing factor to its morbidity is its frequent recurrence in affected individuals > Table 170-1. Major depressive episodes can occur in the setting of major depressive disorder (MDD) and bipolar affective disorder, types I and II. Depressive illnesses that do not meet the criteria for a major depressive episode include dysthymic disorder, substance induced mood disorder, mood disorder due to a general medical condition,
i) Five or more of the following symptoms present during the same 2-week time period that represents a change from prior functioning (one symptom must be either #1 or #2): Depressed mood for most of the day; nearly every day as reported by the patient or others Diminished interest and capacity to experience pleasure in all or most daily activities nearly every day as reported by the patient or others Significant unintentional weight loss or change in appetite nearly every day Change in sleeping patterns nearly every day Psychomotor changes nearly every day as reported by others Fatigue or low energy levels nearly every day Feelings of worthlessness or guilt that are inappropriate and excessive nearly every day Poor concentration or indecisiveness nearly every day as reported by the patient or others Recurrent thoughts of death, suicidal ideation, or plans to commit suicide ii) Symptoms do not meet criteria for a Mixed Episode (i.e., features of depression and mania) iii) Symptoms cause significant distress and impairment in social, occupational, and other important areas of functioning iv) Symptoms are not due to the direct effects of a substance or medical condition v) Symptoms are not better accounted for by bereavement
Ethical considerations in psychiatric surgery
trials of antidepressants from different drug classes, augmentation trials with lithium, antipsychotics, or levothyroxine, appropriate forms of psychotherapy (e.g., cognitive behavioral therapy (CBT), psychodynamic psychotherapy, and dialectical behavioral therapy), and failure to respond to ECT. As depression is often undertreated, Harold Sackeim suggests that clinicians evaluate four factors when determining treatment-resistance: (1) treatment adherence, (2) adequate dosing of medication, (3) adequate duration of treatment, and (4) critical evaluation of treatment outcomes. The rationale for discontinuing a medication should also be considered as this may clarify the adequacy of a treatment trial. There are several predisposing factors to treatment-resistant depression. The history of an affective illness in first-degree relatives was shown to predict treatment-resistant depression in one study. Early and late age at onset of a depressive illness has been suggested to be a risk factor. However, this finding may be complicated by co-morbid personality disorders and substance abuse in younger patients as well as concomitant dementia in elderly patients. A subgroup of patients with treatment-resistant depression experience greater functional impairment, prolonged illness duration, low rates of spontaneous remission, and elevated relapse rates. Chronic depression, defined as depression that lasts longer than two years, is also a risk factor for treatment-resistance. According to one study, chronic depression occurs in 20% of patients with MDD. Treatment-refractoriness can exacerbate many aspects of depressive illness, affecting both individuals and society. There is a higher rate of suicide in patients with treatment-resistant depression compared to those who are responsive to treatment. Treatment-resistant depression is a global economic burden as individuals spend an estimated $14,990 per year in medical, pharmaceutical, and work-loss expenditures. Patients with chronic treatment-refractory depression often display
170
greater impairment than those afflicted with other general medical disorders. Standardized instruments are employed to measure the severity and disability associated with a depressive episode in research studies and the clinical setting. A Beck Depression Inventory (BDI) score greater than 30 is often used to determine illness appropriate neurosurgical treatment. Likewise, functional rating scales are used to quantify functional disability (i.e., Global Assessment of Function (GAF) less than 50). These instruments are useful not only for the initial assessment of patients, but are also important for monitoring their outcomes.
Obsessive-Compulsive Disorder OCD is another psychiatric disorder that is sometimes treated by neurosurgery. OCD is characterized by the presence of obsessions or compulsions (> Table 170-2), the realization that they are excessive or illogical, marked distress and dysfunction in multiple domains, the spending of over 1 h a day in OCD activities, and by the fact that they cannot be the result of another psychiatric or medical disorder or substance. At present, neurosurgical treatment is reserved for patients with treatment-refractory OCD thatisassociatedwithchronicillnessand/orextreme disability. A diagnosis of treatment-refractoriness typically requires adequate treatment trials attempted over a 5-year period. Illness severity is quantified using standardized instrument scales, the most common of which is the Yale-Brown Obsessive Compulsive Scale (Y-BOCS). A Y-BOCS score greater than 20 is usually required for a center to consider neurosurgical interventions in individuals with OCD. It is not uncommon for disability to be a major consideration for surgery as the compulsive drive can consume a patient and they may spend their entire day performing rituals, restricting their ability to work or even care for themselves or family members. OCDs can have a
2859
2860
170
Ethical considerations in psychiatric surgery
. Table 170-2 DSM-IV definitions for obsessions and compulsions Obsessions: Recurrent or persistent thoughts, impulses, or images that are intrusive and cause distress The above are not excessive worries about real-life problems The individual attempts to ignore, suppress, or distract themselves from the above The above are recognized as coming from the individual’s own mind Compulsions: Repetitive behaviors that the individual feels driven to perform because of obsessions or in accordance to rules The above actions are performed to relieve distress or prevent an unwanted event in an unrealistic manner
prolonged course, but symptoms sometimes improve over the lifespan. Nevertheless, one study reports an average illness duration of 30 years in a sample of OCD patients. The onset of illness under 20 years of age predicted a poor prognosis in this group.
Tourette’s Syndrome Gilles de Tourette’s syndrome (TS) is a movement disorder with prominent neuropsychiatric manifestations. TS is characterized by the childhood onset of motor or vocal tics that frequently remit with age (> Table 170-3). Psychiatric comorbidities are common in TS and patients often display symptoms of OCD, attention deficit hyperactivity disorder (ADHD), depression, and impulse control disorders. Pharmacological management of TS and its associated behavioral manifestations include dopamine receptor blockers (i.e., antipsychotics), dopamine depleting medications (i.e., tetrabenazine), and serotonergic medications for OCD and depressive symptoms (i.e., SSRI’s and clomipramine). Although TS typically resolves by adulthood, several cases persist beyond adolescence. Neurosurgical treatment for a
. Table 170-3 Characteristics of Tourette’s syndrome i) Tics Motor (intermittent, sudden, brief, jerking movements of limbs, mouth, nose, etc.) Vocal (sniffing, coughing, shouting, barking, coprolalia, echolalia, palilalia) ii) Neuropsychiatric disorders Obsessive-compulsive disorder Attention deficit hyperactivity disorder Affective disorders Learning disorders Sleep disorders Behavioral disorders Impulse control disorders Oppositional behavior Self-injurious behavior Anger outbursts
disorder that is often self-limited has been somewhat controversial. TS can result in self-injurious behaviors and severe debilitation due to the severity of tics, neuropsychiatric symptoms, or a combination of both. This particularly disabling form of TS is termed ‘‘malignant TS,’’ which is potentially life threatening. Neurosurgical treatment is considered a viable treatment alternate in malignant TS, treatment-refractory cases, and in individuals with a prolonged course of TS.
Ablative Surgery (Since 1980) Because of the controversy surrounding psychiatric neurosurgery, few medical centers currently perform these procedures. The international centers that do perform ablative surgeries for psychiatric indications lesion several different sites that often vary between centers. The most frequently used and studied procedures are the anterior cingulotomy, subcaudate tractotomy, limbic leukotomy, and capsulotomy (> Table 170-4). Many of these procedures are used for the same indications and different disorders are often treated with the same procedures.
Ethical considerations in psychiatric surgery
. Table 170-4 Common neurosurgical procedures for psychiatric disorders Neurosurgical procedure
Indications
Anterior cingulotomy
Affective disorders, OCD, Schizophrenia, Personality disorders Affective disorders and OCD Affective disorders and OCD Affective disorders, OCD, Panic disorder
Subcaudate tractotomy Limbic leucotomy Capsulotomy (anterior limb of the internal capsule)
In 1953, Lars Leksell, the developer of the Gamma Knife, was the first to use radiosurgery in the treatment of psychiatric disorders. The Gamma Knife changed the field of neurosurgery, especially in regard to treating psychiatric disorders. By using localized g-radiation, neurosurgical treatments became much less invasive. In addition, using focal sites of radiation spared the rest of the brain from consequent irradiation. Radiosurgery allowed for sham-controlled studies that could examine the effects of psychiatric neurosurgery without subjecting patients to undue surgical complications. Despite the minimal amount of tissue destruction, radiosurgery is still an ablative procedure that results in a cerebral lesion, a prospect that is distasteful to many patients and clinicians. Neurosurgical ablation is currently employed for the treatment of psychiatric disorders in select academic centers with fewer than 25 cases performed annually in the United States and England combined. Patients are initially evaluated by a multi-disciplinary team that has expertise in the field of psychosurgery to determine their appropriateness for the procedure. Virtually all surgeries are part of a research protocol with Institutional Review Board (IRB) approval and longitudinal follow-up procedures. The patient is expected to continue with follow-up care as it pertains to their disorder.
170
Deep Brain Stimulation (Since 1980) The United States’ Federal Drug Administration has approved the use of DBS for the treatment of PD, essential tremor, and primary dystonia, but the number of experimental indications continues to rise annually. Among these investigational trials are primary psychiatric disorders, specifically depression, OCD, and TS. Neuropsychiatric data derived from prior studies of DBS for the treatment of movement disorders demonstrated improvements of depression, anxiety, and OCD symptoms and prompted its use in recalcitrant psychiatric disorders. Subsequently, Gabrie¨ls et al. (2003) reported on three patients with treatment-resistant OCD who received DBS surgery [3]. Two patients experienced sustained improvement one year following surgery but became anxious when the stimulator was deactivated. This case report generated excitement in the field of psychiatry and several studies examining the use of DBS for psychiatric illnesses followed. Since 2003, six studies containing data on 23 patients have reported the clinical effects of DBS for the treatment of OCD. All of the patients had bilateral electrodes implanted into the anterior limb of the internal capsule (ALIC), similar to an anterior capsulotomy. All studies used DSM-IV criteria for a diagnosis of OCD, noted co-morbid psychiatric diagnoses, and used standard clinical rating scales (e.g., Y-BOCS) to measure treatment outcomes. A multi-disciplinary team evaluated patients for treatment-refractoriness and disability in order to determine their appropriateness for the procedure. Hospital ethic committees were consulted in at least three studies. The stimulator was experimentally turned off in several studies, resulting in a rapid recurrence of symptoms. Several trials performed a double-blinded, ‘‘on-off’’ designed study; however, there was evidence that patients subjectively experienced effects of the stimulator
2861
2862
170
Ethical considerations in psychiatric surgery
when it was activated. Over half of the patients reported in these studies sustained clinically significant improvement with the longest followup period of 3 years. Despite its experimental use in the treatment of OCD, it is estimated that over 100 cases have been performed for this indication, many of which were done in community hospital settings. These treatment practices raise several ethical concerns that will be described later in further detail. Despite data supporting an improvement of depressive symptoms in patients with movement disorders who are treated with DBS, the treatment of idiopathic depression (i.e., MDD) with DBS was not studied until Helen Mayberg and Andres Lozano published a report on six patients in 2005 [4]. After extensive collaboration with three ethical committees, patients with treatment-refractory major depression had electrodes implanted in the bilateral subgenual cingulate region, Brodmann’s area 25 (Cg25WM). This surgical target was selected because of Mayberg’s imaging studies that suggested the Cg25WM region is heavily involved in depressive symptoms. Four of the six patients continued to show significant improvement six months following surgery. Three other case reports described significant clinical improvement in depressive symptoms in all five cases. These reports were double-blinded, ‘‘onoff’’ studies with a maximum follow-up duration of 20 months. Two other studies were performed in patients with bilateral inferior thalamic peduncle electrodes, while the patients included in an additional study had bilateral electrodes implanted within the nucleus accumbens. Although the same standardized measurement instruments and diagnostic criteria were used in all of the studies, the surgical locations varied in virtually all of the reports making comparison studies impossible. Psychiatric co-morbidities with Axis II (i.e., personality disorders) diagnoses also complicated the interpretation of such studies. Initial studies have demonstrated efficacy of DBS for the treatment of depression but further research requires
standardization of the surgical site to determine efficacy and safety. The majority of studies of neurosurgical treatment for psychiatric disorders have been in patients with TS. Nine studies have investigated the treatment of motor and psychiatric symptoms via DBS implantation in patients with TS. All of the studies were either case reports or case series except for one that followed 18 patients with TS for up to 19 months. Most studies were either single or double-blinded in design. Surgical sites differed among studies and included the globus pallidus interna (GPi), the centromedianparafasicular complex of the thalamus (CM-Pfc) with and without leads in the GPi or ventral oralis complex of the thalamus, the ALIC, and the nucleus accumbens. The majority of patients received DBS implantations in the CM-Pfc, which appears to have fewer psychiatric side effects compared to other sites (e.g., ALIC). In general, reports have noted improvements in motor symptoms, but DBS’s effects on neuropsychiatric symptoms have varied by study. Two studies reported surgeries in patients under 18 years of age, a subject that is somewhat contentious as TS often improves or remits with age. In addition to depression, OCD, and TS, neurosurgery has also been used to treat patients with treatment-refractory chronic anxiety and behavioral disorders. Its use in patients with behavioral disorders, such as substance dependence, has been controversial as the etiologic mechanism of behavioral disorders has remained unclear and is likely multi-factorial in nature. For example, there is a biological drive in alcohol dependence; however, a volitional and significant psychosocial component exists as well. Historical neurosurgical procedures for aggression, violence, homosexuality, and criminal behavior are interventions directed at ameliorating motivated behaviors, not psychiatric disorders partitioned by a disease-based etiology. Although all behaviors are governed by the brain, there is no clear
Ethical considerations in psychiatric surgery
evidence of a ‘‘broken part’’ or dysfunctional biological mechanism that is primary to these disorders. There are also several case reports of DBS for the treatment of other psychiatric disorders including impulsivity following traumatic brain injury, alcohol dependency, and pathological gambling.
Ethical Considerations It is easy to be critical of prior experiences with psychiatric neurosurgery, but one must realize that standards of care, research, and ethics were different in the past. Nonetheless, the wide acceptance of the frontal lobotomy in the late 1940s was in large part due to the lack of effective treatments for chronic mental illnesses such as schizophrenia, the championing of the procedure by a select few, and a lack of critical review of outcomes. Furthermore, many practitioners, especially in the U.S., did not carefully distinguish among the different diagnoses; so the occasional success of a leucotomy in a patient with severe depression was extrapolated to include all chronic mental illnesses, especially schizophrenia. The lack of current standards of research (the randomized control trial was not established as a standard until the 1950s) made scientific assessment of treatment outcomes less likely, but the combination of clinicians’ desire for efficacious treatments and Freeman’s enthusiastic proselytizing led to the performance of 20,000 procedures between 1946 and 1951. The lesson to be gained from this experience is that careful study is needed before establishing new therapies, especially those with meaningful degrees of potential harm and expense, and study results should be made widely available. Furthermore, as a group the mentally ill have been subject to abuse by society and professionals for hundreds of years necessitating the careful scrutiny of any invasive, potentially harmful, or radical therapy.
170
In spite of these concerns, the preliminary data on a few subjects with specific psychiatric illnesses suggests potential benefit. However, many questions remain unanswered. Are various forms of psychosurgery including DBS, effective and safe for disorders of thought, behavior, and mood? For which specific disorders do they meet efficacy and safety standards? Do the benefits persist? For DBS, how long is stimulation necessary? What are the long-term side effects, even if short-term efficacy and safety are established? Should individuals who are unable to give consent because of the severity of their illness undergo any form of psychosurgery if it has established efficacy? Many of these concerns are both scientific and ethical, a point that speaks in favor of the continued interchange between clinicians, researchers, and ethicists. Psychiatric neurosurgery has been shown to be efficacious in several patients, which has lead to a more scientific examination using modern day stereotactic neurosurgery, which has also reported efficacy in patients who have not responded to other treatments. As an efficacious treatment for those who are recalcitrant to non-surgical interventions, psychiatric neurosurgery remains a valuable treatment option for the extremely mentally ill. In light of the reservations surrounding psychiatric neurosurgery, several groups of expert clinicians and researchers have developed consensus criteria for selecting and treating patients using this treatment modality. Three major consensus statements regarding the use of DBS for the treatment of depression, OCD, and TS have been published since 2003 [5–7]. Key points of these statements include inclusion and exclusion criteria with a particular emphasis on the patient’s appropriateness for neurosurgical treatment. Strict criteria for determining a patient’s illness duration and severity have been proposed and the disabling features of the disease have been elucidated. The requirement of an established and multidisciplinary team with expertise
2863
2864
170
Ethical considerations in psychiatric surgery
in the evaluation and treatment of the indicated illness and familiarity with the implantation and maintenance of DBS hardware is universally agreed. The most recent conference that has addressed the ethical issues of psychiatric neurosurgery was held in September 2007 in Baltimore, Maryland. An international group of leading researchers actively involved in the study and treatment of depression, OCD, and TS using ablative or DBS neurosurgery met with key ethicists, historians, neuroanatomists, and patients to design a consensus statement regarding the use of DBS in the treatment of psychiatric disorders. Several important ethical tenets arose from the aforementioned consensus statements; these are outlined below:
Secondary outcome measures of neuropsychiatric effects of DBS in patients with movement disorders in addition to several case studies that have used DBS for the treatment of primary psychiatric disorders have demonstrated improvements of neuropsychiatric symptoms in several patients. These findings support the need for further research of psychiatric neurosurgery using primary and secondary outcome measures. The use of DBS for the treatment of psychiatric disorders is investigational and no standards exist for the selection or treatment of patients using this modality. Further techniques and surgical sites should be explored and studies should be designed to allow for comparison analyses. Because the long-term consequences of psychiatric neurosurgery are unknown, longterm follow-up of patients is mandatory to establish efficacy and record adverse event data, and should be included in an international registry. We also recommend that further longitudinal data be collected in patients with movement disorders who are
treated with DBS given the psychosocial effects and elevated suicide rate in this population. Likewise, we recommend that these patients receive long-term follow-up with attention to psychiatric symptoms and psychosocial adjustment. Comparison studies between ablative and neurostimulatory treatments are needed. Although DBS is reversible, in that electrodes can be removed from the patient, there are several drawbacks to this procedure including long-term maintenance (e.g., battery and lead replacements), possibility of electrode migration, and the likelihood of needing replacement electrodes in the case of prolonged treatment. All of the above complications make DBS an expensive treatment option that may not be accessible to all patients, especially in third world countries. In comparison, ablative surgery does not require maintenance care, with the exception of the rare possibility of needing additional lesions. The cost associated with ablative surgery is much less than DBS, establishing it as an affordable alternative. However, ablative surgery does cause irreversible destruction to neuronal tissue, an issue that is usually cited as its major drawback. However, focal radiosurgery may become more acceptable depending on comparison studies with DBS. The neurosurgical treatment of psychiatric disorders should only occur in a university setting given its experimental use. All surgeries should be performed following the approval of hospital ethics committees and be part of an approved research protocol. We agree with the prior criteria that necessitates patients be evaluated by a comprehensive team comprised of neurosurgeons and neurologists with experience in DBS therapy and psychiatrists with expertise in diagnosing and treating the psychiatric condition for which the surgery is indicated.
Ethical considerations in psychiatric surgery
The team should likewise include case management and a neuropsychologist. Standard inclusion and exclusion criteria should be established by experts in each of the psychiatric disorders being studied (e.g., depression, OCD, TS). The definitions of ‘‘treatment-refractory,’’ ‘‘chronic,’’ and ‘‘disabling’’ should be determined for each disorder with standardized procedures for measuring these evaluative and outcome variables. Subjects who are selected for psychiatric neurosurgery should be 18 years or older due to the propensity for several disorders to improve or remit over time. Extensive data and informant interviews should be collected on each individual patient pertaining to the above criteria in order to establish their appropriateness for the procedure. Although all procedures should be part of an established and approved research protocol, neurosurgical procedures performed for humanitarian reasons should always require review by an independent ethics committee. The consent process should clearly delineate the known and possible risks, benefits, and alternatives to psychiatric neurosurgery. This treatment modality should be described as part of a comprehensive psychiatric rehabilitation program that requires long-term care and follow-up, which will likely not ameliorate all aspects of the disorder. Patients who are involved in investigational studies of psychiatric neurosurgery should not bear the financial responsibilities of treatment or complications relating to the surgery. Financial responsibility should be appropriated to entities that are involved in the treatment process prior to the procedure. The ideal study for investigating neurosurgical treatments of psychiatric disorders would be a randomized, double-blinded, prospective study that compares surgery to a sham operation. There are many ethical
170
barriers to conducting such a study such as performing a sham procedure on individuals ill enough to receive such an intervention. The most practical way to perform such a study would be via Gamma Knife surgery, where a control group would not be exposed to undue surgical risks and not receive the possibly therapeutic effects from the actual procedure. Such studies should be discussed among the research community. We believe that receiving informed consent from a competent patient is sufficient to proceed with neurosurgery. There is no evidence that individuals with psychiatric disorders differ from other groups of patients with chronic medical illnesses in that they do not lack the capacity to consent to a procedure on the sole basis of having a psychiatric disorder. As data from research studies accumulate, further ethical conferences should convene to re-evaluate the established criteria and to determine the need for any additional actions.
In conclusion, the history of psychiatric neurosurgery is vast and complex [8]. Scientific and social immaturity has made society cautious about the use of neurosurgery for the treatment of psychiatric disorders. However, studies with scientific merit demonstrate clinical improvement in several individuals receiving neurosurgery for depression, OCD, and TS. Psychiatric neurosurgery can be effective in treating patients with treatment-refractory, chronic, and debilitating psychiatric illnesses, but its study must be properly regulated. Further investigation by appropriately designed research protocols that are reviewed by ethics committees are needed to elaborate psychiatric neurosurgery’s efficacy and safety. Clinicians should discourage the use of neurosurgery as a treatment for psychiatric disorders outside of previously approved centers that do not meet the aforementioned investigational standards.
2865
2866
170
Ethical considerations in psychiatric surgery
References 1. Bejjani B, Damier P, Arnulf I, Thivard L, Bonnet A, Dormont D, Cornu P, Pidoux B, Samson Y, Agid Y. Transient acute depression induced by high-frequency deep-brain stimulation. N Engl J Med 1999;340 (19):1476-80. 2. Appleby BS, Duggan PS, Regenberg A, Rabins PV. Psychiatric and neuropsychiatric adverse events associated with deep brain stimulation: a meta-analysis of ten years’ experience. Mov Disord 2007;22(12):1722-8. 3. Gabrie¨ls L, Cosyns P, Nuttin B, Demeulemeester H, Gybels J. Deep brain stimulation for treatment-refractory obsessive-compulsive disorder: psychopathological and neuropsychological outcome in three cases. Acta Psychiatrica Scandinavica 2003;107(4):275-82. 4. Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, Schwalb JM, Kennedy SH. Deep brain stimulation for treatment-resistant depression. Neuron 2005;45(5):651-60. 5. Mink JW, Walkup J, Frey KA, Como P, Cath D, Delong MR, Erenberg G, Jankovic J, Juncos J,
6.
7.
8.
9.
Leckman JF, Swerdlow N, Visser-Vandewalle V, Vitek JL, Tourette Syndrome Association, Inc. Patient selection and assessment recommendations for deep brain stimulation in Tourette syndrome. Mov Disord 2006;21(11):1831-8. Greenberg BD, Price LH, Rauch SL, Friehs G, Noren G, Malone D, Carpenter LL, Rezai AR, Rasmussen SA. Neurosurgery for intractable obsessive-compulsive disorder and depression: critical issues. Neurosurgical Clinics of North America 2003;14(2):199-212. Scientific and Ethical Issues Related to Deep Brain Stimulation for Disorders of Mood, Behavior, and Thought. Baltimore, MD, September 2007. Arch Gen Psychiatry [In Press] Valenstein ES. Great and desperate cures: the rise and decline of psychosurgery and other radical treatments for mental illness. New York, NY: Basic Books; 1986. Pressman, Jack D. Last Resort. Psychosurgery and the Limits of Medicine. Cambridge University Press, Cambridge, 1998.
174 Medical Management and Indications for Surgery in Depression P. Giacobbe . S. Kennedy
Major Depressive Disorder: Definitions and Prevalence Major Depressive Disorder (MDD) is a major public health problem throughout the world and is also referred to as Unipolar Depression. It is characterized by a single major depressive episode (MDE), or more often recurrent episodes. An MDE may also occur in patients who have a diagnosis of Bipolar Disorder (BD). The association between depression and various medical illnesses including diabetes and coronary artery disease has also been recognized and partially accounts for the high mortality rates of depressed patients. First-line treatments for MDD are mainly antidepressant medications and brief psychotherapies, such as cognitive behavioral therapy (CBT) or interpersonal psychotherapy (IPT). Despite the refinement of antidepressant medications for more than 50 years: monoamine oxidase inhibitors and tricyclic antidepressants were the first generation, followed by selective serotonin reuptake inhibitor (SSRI) and dual serotonin and norepinephrine reuptake inhibitor (SNRI) antidepressants; up to 30% of depressed patients fail to respond to any of these treatments. Reduction in symptom severity of >50% on a classical scale such as the Hamilton Depression Rating Scale -17 Item (HDRS-17 [1]) or the Montgomery-Asberg Depression Rating Scale (MADRS-[2]) is the criterion for ‘‘response.’’ However, many patients who meet this standard
#
Springer-Verlag Berlin/Heidelberg 2009
are left with a considerable symptom burden and are at a high risk of relapse and recurrence. Recent evidence suggests that patients who achieve ‘‘remission,’’ typically defined as achieving a score of 7 or less on the HDRS-17 or a score of 10 or less on the MADRS, have a greater likelihood of achieving a better quality of life and a functionally productive outcome. Pooled analyses of data from relatively homogenous clinical trials suggest that approximately 40% of depressed patients achieve remission after antidepressant treatment [3,4] and remission rates as high as 70–80% have been reported following extended trials [5]. However, results from a large sequenced effectiveness trial carried out in the United States were less encouraging. After two sequential trials involving antidepressants from different classes or CBT only 50% of patients achieved remission and the cumulative gain from additional treatments was minimal [6]. In summary, despite significant advances in both pharmacotherapy and psychotherapy options, 20–30% of patients remain in a state of chronic or treatment resistant depression. By definition, chronic depression requires a duration of active depressive symptomatology for 2 years or more, while treatment resistant depression is defined as failure to respond to at least two different classes of antidepressants at adequate doses for adequate duration. In reality, many patients remain more or less constantly depressed for years despite multiple trials of antidepressant monotherapy and combination therapy.
2926
174
Medical management and indications for surgery in depression
Current Concepts of the Pathophysiology of Depressive Illness Developments in the field of neurogenetics, neuroimaging, and other fields of neuroscience support a neurobiological basis for depression, characterized by an interaction between external stress factors and innate biological vulnerabilities. For example, individuals with a genetic polymorphism of the serotonin transporter (short allele) are more likely to experience depression following adverse life events than those with the long allele [7]. It is likely that other neurobiological differences distinguish sub-populations with MDD, and if properly characterized, could help to predict optimal treatment. Converging lines of evidence from neuroimaging studies of the brain as well as replicated preclinical studies that antidepressant modalities promote neuroplasticity and neurogenesis have led to a paradigmatic switch in the conceptualization of depression [8]. Rather than a ‘‘monoamine neurotransmitter deficit’’ model, current neurocircuitry models of depression hypothesize that depressive symptoms are reflective of a failure of innate homeostatic processes in the brain to compensate for endogenous or exogenous stressors [9,10]. This inability to set into motion changes at a molecular level aimed at re-establishing a euthymic mood state, results in failure to ‘‘turn off ’’ the depressive state and re-establish a normal mood state. An influential neurocircuitry model of depression has been advanced by Mayberg and colleagues, based on functional neuroimaging studies of regional brain metabolic activity in the depressive mood state and following successful antidepressant response [11]. Successful antidepressant treatment is associated with a normalization of metabolic activity in both the underactive dorsal (cortical) and overactive ventral (limbic) compartments of the neurocircuits involved in the regulation of emotion [10,12,13].
Situated between the dorsal and ventral circuits, the subgenual cingulate cortex (SCC), which includes Brodmann Area 25 (BA25), is in an anatomically strategic position to modulate the activity of both the frontal cortex and limbic system. In a PET study to examine the metabolic effects of transient sadness, healthy volunteers demonstrated increased metabolic activity in BA 25, while reductions in the activity of BA25 were observed in depressed patients successfully treated with the antidepressant fluoxetine [14]. Similar findings of increased baseline activity of BA25, and subsequent reduction in activity of BA25 following successful antidepressant treatment with various treatment modalities including CBT, pharmacotherapy [13], sleep deprivation [15] and ablative neurosurgical techniques [16] have been reported. The success of various methods to treat depression, such as psychotherapy, pharmacotherapy and various forms of neuromodulation can be understood in an anatomical framework, whereby different treatment modalities modulate distinct neural targets in both a ‘‘top-down’’ and a ‘‘bottom-up’’ fashion to restore homeostasis. Patients with TRD may represent a subset of depressed individuals for whom the antidepressant and psychotherapeutic interventions are unable to augment the brain’s innate homeostatic mechanisms to turn off a depressive state. These patients are candidates for somatic treatments such as electroconvulsive therapy (ECT), repetitive transcranial magnetic stimulation (rTMS), vagus nerve stimulation (VNS) or deep brain stimulation (DBS), order to exogenously ‘‘kick start’’ the cascade of adaptive processes necessary to compensate for a depressive mood state [17]. These other neurostimulation therapies are not the focus of this chapter, but evidence about their proposed mode of action including preclinical studies of neurogenesis and neural circuitry studies is concordant with the emerging finding about DBS for TRD.
Medical management and indications for surgery in depression
Selection of Anatomical Sites for DBS Several factors may be important in guiding the selection of an anatomical target for DBS in depression. First, given the heterogeneity in the symptomatology of depression, various symptoms may have distinct patterns of localization and dysfunction within the anatomically relevant neurocircuitry of mood regulation subserving mood regulations [18,19]. Therefore, it is possible stimulation at different target sites may differentially improve various symptom profiles. Second, the selection of neuroanatomical sites for implantation of DBS electrodes in neurological conditions has been directed at grey matter nuclei that are considered to have a ‘‘pacemaker’’ or ‘‘gating’’ function in the transmission of neural impulses, such as the thalamus [20], or alternatively densely packed white matter fiber bundles which transmit axon potentials to a more widely distributed anatomical network. Both of these strategies allow the discrete area being stimulated by the DBS electrode to potentially affect the activity of a much larger area in the brain. A third consideration is the selection of a site which can ameliorate depressive symptoms without requiring high voltage stimulation. The need for high-amplitude stimulation calls into question the notion that the effect is due to the precise anatomical location of the electrodes, as the physiological effects of current spreading beyond the target, increase in importance with increasing voltage [21]. An equally effective site of stimulation that requires lower voltages is to be preferred to prolong battery life and minimize the number of operative procedures required for battery replacement. Although there is no clear consensus on one optimal target for stimulation, evidence from functional neuroimaging studies supports the selection of subgenual cingulate cortex (BA25), the anterior limb of the internal capsule, the nucleus accumbens and other anatomical locations
174
as targets for pilot studies. See > Table 174-1 for a summary of published reports to date.
Subgenual Cingulate Cortex (SCC) To date, the largest reported case series of DBS for depression is a report from the University of Toronto on 20 individuals with TRD who received BA25 DBS [25,27,29]. The initial report of this open-label study described the 6 month outcomes of 6 patients who received implantation of DBS electrodes in the white matter tracts adjacent to BA25 in the SCC [25]. Inclusion and exclusion criteria for the study are presented in > Table 174-2. Response was defined as a decrease in the HDRS17 score of 50% or greater from the pretreatment baseline, and remission as an absolute HDRS-17 score 7. Two months after surgery, five patients met criteria for response, which was maintained in four patients after 6 months. A significant decrease in the average HDRS-17 score from baseline was seen after 1 month of stimulation and improvements in depressive symptomatology remained stable for the 12-month period of observation [27]. The follow up of these patients at 1 year has demonstrated a similar rate of clinical response (13/20) [27] as reported in the initial report [25]. Baseline abnormalities in cerebral blood flow seen on PET, namely increased activity in the SCC and decreased activity in dorsolateral prefrontal, ventrolateral prefrontal and anterior cingulate cortices were normalized following 3 months of DBS and were maintained at 6 months [25]. Extensive neuropsychological testing demonstrated that there is no evidence of neuropsychological impairment [30].
Anterior Limb of Internal Capsule Neuromodulation of the anterior limb of the internal capsule, with both ablative and stimulation
2927
2928
174
Medical management and indications for surgery in depression
. Table 174-1 Treatment studies of brain stimulation for mood disorders and related conditions Reference Heath et al. [22]
Number of patients
Site of stimulation
Stimulation parameters
Follow-up period
6
Cerebellar vermis
100 Hz
5–30 months
150–250 s
Aouizerate et al. [23]
1
Ventral caudate
2V
27 months
130 Hz 90 ms Jimenez et al. [24]
1
ITP
3.0 V
24 months
130 Hz
Mayberg et al. [25]
Kuhn et al. [26]
6
1
Subgenual Cingulate Cortex
NAcc
450 ms 4.0 V
6 months
130 Hz 60 ms 4.5 V
12 months
130 Hz 90 ms Lozano et al. [27]
20
Schlaepfer et al. [28]
3
Subgenual Cingulate Cortex Ventral striatum
12 months
0–5 V
6–23 weeks
Notes Lack of standardized diagnostic criteria and scales Voltage ‘‘determined by sensory evoked potentials recorded from the scalp and by the patient’s clinical response’’ Comorbid OCD Remission of depressive symptoms at 6 months Improvement in OCD symptoms following 15 months of stimulation Comorbid bulimia nervosa and borderline personality disorder No relapse in depressive symptoms during 12 months following discontinuation of stimulation
Comorbid anxiety disorder and Alcohol Dependence No improvement in anxiety or depressive symptoms Marked reduction in alcohol consumption Extension of sample from Mayberg et al. [25] Double-blind changes to stimulation parameters
145 HZ 90 ms
techniques, has been employed in the treatment of refractory obsessive-compulsive disorder (OCD). Recently the antidepressant effect of stimulation in the anterior limb of the internal capsule has been explored for patients with TRD. At the time of writing, there are no published studies on the antidepressant efficacy of DBS in the internal capsule for individuals with a primary diagnosis of depression. Support for the mood elevating effects from stimulating the internal capsule emanates from reports that depressive symptoms have in
OCD and Tourette’s syndrome (TS) patients been improved during DBS. In an open-label trial of DBS in the internal capsule, Greenberg and colleagues reported that in a sample of ten OCD patients, of whom eight met criteria for comorbid MDD, half of the patients demonstrated rapid changes in mood and affect when DBS was initiated or interrupted [31]. Additional reports of DBS in the internal capsule for OCD [32–34] and TS [35] have also documented improvements in depressive symptoms.
Medical management and indications for surgery in depression
. Table 174-2 Inclusion and exclusion criteria Inclusion criteria: Age 30–75 Diagnosis: major depressive episode (unipolar) by DSM-IV derived from the SCID Recurrent disease; minimum four major depressive episodes Chronic illness with current episode ~12 months duration Response failure to multiple treatment regimens. Resistance or intolerance to at least four treatments from different categories (SSRI, SNRI, TCA, HCA, MAOI, atypical, Lithium, anticonvulsants, ECT, VNS, CBT/IPT). Documentation of adequate dose and duration of each treatment Hamilton Rating Scale for Depression (HRSD-24) score >20 Global assessment of function. score ~50 No neurological disease; no other Axis 1 or Axis II diagnosis; no substance abuse Stable on current antidepressant medication regimen or medication free ~4 weeks Able to give informed consent in accordance with institutional policies Able to comply with all testing and follow-up visit requirements defined by the Study Protocol. Premenopausal women must agree to use acceptable methods of birth control (radiation risk of PET) Exclusion criteria: Pregnancy Psychotic subtype of major depressive disorder Alcohol or substance dependence within 12 months Alcohol or substance abuse within 6 months, excluding nicotine Current suicidal ideations, plan or intent for selfharm In past 1 year, repeated suicide attempts, resulting in emergency room or inpatient care Major medical illness, cardiac pacemaker/ defibrillator, and other implanted stimulator Likely to relocate or move to a location distant from the study site within 1 year of enrollment
Ventral Striatum/Nucleus Accumbens There have been several reports to date on the effectiveness of stimulating of the ventral striatum, a region which includes the nucleus accumbens (NAcc), for both TRD and treatment-refractory anxiety disorders [23,26,28,36,37]. Based on
174
differences in cytoarchitecture, neuronal projections and neurotransmitter receptors, the NAcc can be divided into core and shell subregions. The NAcc shell projects to the limbic brain areas, including the subgenual cingulate cortex (BA25), while the NAcc core is connected to the extrapyramidal motor system (reviewed in [38]). Due to its rich dopaminergic projections, the NAcc has become a focus of both human and animal studies linking dysfunction in this brain region with impaired reward processing seen in depression [39,40]. The nucleus accumbens is uniquely situated anatomically to mediate the emotional and behavioral aspects of emotion. In animal models, stimulation of the NAcc has hedonic properties, resulting in increased exploratory behavior, food intake, and animals readily learn operant tasks to increase the rate of electrical pulses delivered to the NAcc [41]. The effectiveness of DBS in the ventral caudate nucleus in ameliorating mood and anxiety symptoms has been the subject of a case report [23]. Bilateral quadripolar electrodes were guided through the head of the caudate nucleus in a manner resulting in the two most ventral contacts being located within the nucleus accumbens, while the two most dorsal contacts remained in the ventromedial portion of the caudate nucleus, in a single patient with intractable OCD and concomitant MDD. Stimulation was delivered to the two deepest contacts located in the nucleus accumbens during the first month, resulting in little improvement in depressive or anxiety symptoms. With the addition of stimulation via the two upper contacts located within the caudate nucleus, substantial clinical improvements in depressive, but not obsessivecompulsive, symptoms were observed. The patient achieved remission of his depressive symptoms after 6 months of stimulation, and this effect was sustained 15 months after the surgery. A delayed improvement in obsessive-compulsive symptoms was seen in this individual, and remission of OCD (defined as a Y-BOCS score less than or equal to 16) was only seen after 12 months of DBS.
2929
2930
174
Medical management and indications for surgery in depression
The site of action of stimulation responsible for clinical benefits in this case study remains unclear, as stimulation was provided to both the ventromedial head of the caudate nucleus and the nucleus accumbens concurrently. Rather than specifically stimulating white matter tracts, the electrodes were placed in grey matter. If replicated, this result may provide the foundation for a potentially novel site for stimulation. Recently, the results of a pilot study of DBS in the nucleus accumbens for depression have been published [28]. Three patients with unipolar depression, refractory to antidepressant medications, psychotherapy and electroconvulsive therapy, received bilateral electrode implantation in the ventral striatum. The electrode was positioned such that each of the four contacts was able to stimulate distinct anatomical locations. The two most ventral contacts were placed in the shell and core regions of the nucleus accumbens, while the two most dorsal contacts were situated in the ventral and medial internal capsule. Changes to the stimulation parameters were performed in a double-blinded fashion. Schlaepfer and colleagues reported that acute improvements in depressive symptoms, as measured by the HDRS24-item and the MADRS were noted when the stimulator was turned on and that these changes dissipated when the stimulator was turned off. They also reported that increased voltages were associated with greater improvements in depressive symptomatology. Results from FDG-PET following 1 week of stimulation revealed increased bilateral metabolic activity in the ventral striatum, dorsolateral and dorsomedial prefrontal cortex, cingulate cortex and amygdala, while decreased activity was seen bilaterally in the ventrolateral and ventromedial prefrontal cortex, caudate and thalamus. Kuhn reported a case study of DBS in the nucleus accumbens in an individual with depression, comorbid anxiety and alcohol dependence [26]. Upon initiation of stimulation, an acute sense of ‘‘appeasement’’ was described by this
individual, although no improvement was noted in his depressive or anxiety symptoms over the course of 12 months. However, within 1 month of beginning stimulation, the patient reported losing the desire to consume alcohol. This change was sustained as he decreased his alcohol consumption from more than 10 drinks per day to occasional use during a year of DBS follow up. This case study suggests stimulation of the nucleus accumbens may have promise in the treatment of addictions and impulse control disorders, where aberrant regulation of brain reward systems may play a role in etiological role in the development of these conditions.
Cerebellum Heath and colleagues (1979) [22] reported results of stimulation of the vermis of the cerebellum in a heterogeneous group of 38 psychiatric patients, including 6 patients with depression. The rationale for selecting this site of stimulation was based on preclinical and clinical observations that stimulation of the rostral cerebellar vermis was associated with increased neuronal firing in the anterior septal region and inhibition in hippocampal activity [42]. During the followup period, which lasted up to 30 months, five of the six patients with depression were reported to be ‘‘significantly’’ improved and to be free of medication. The lack of response in the sixth individual was considered to be secondary to a lack of compliance in using the stimulator. It is difficult to compare this early report to more recent findings. The lack of diagnostic clarity, and absence of validated clinical rating scales to quantify the degree of improvement in depressive symptomatology are major limiting factors. Nevertheless, these data demonstrate that focal brain stimulation is able to effect changes in neuronal activity over broadly distributed brain regions, foreshadowing the emergence of a neurocircuitry model of depression.
Medical management and indications for surgery in depression
174
Inferior Thalamic Peduncle
Patient Selection
The effects of DBS in the inferior thalamic peduncle (ITP) in a 49-year-old woman with a 23-year history of recurrent unipolar depression, bulimia nervosa and comorbid borderline personality disorder have been reported [24]. This site was selected on the basis of preclinical studies by this group, which suggested that stimulation of this white matter bundle exerts an inhibitory effect on the orbitofrontal cortex [43]. The patient experienced a dramatic reduction in depression scores (HDRS decreased from 42 to 3), after bilateral implantation of electrodes but before beginning acute stimulation. The HDRS score was between 2 and 8 during 8 months of active DBS. Following this active stimulation phase, the patient entered a double-blind protocol with stimulators ‘‘OFF.’’ During 12 months without stimulation (or psychotropic medications), no relapse of depression was noted. Given the psychiatric comorbidity, the specificity of antidepressant effects of ITP stimulation are unclear, and these findings require replication.
As with DBS for neurological conditions, choosing suitable TRD patients for DBS is one of the most important steps in this procedure [46]. Although there are no universally accepted guidelines to assist the clinical team in selecting the ideal TRD candidate, we now have more than 5 years’ experience in the pre-operative selection process and post-operative management of a cohort of TRD patients who have received DBS to BA25. The strategies for psychiatric management that follow are based on the clinical experience at our site. With the increasing use of DBS to treat neuropsychiatric conditions in centers across the world, a challenge remains in creating a central database to evaluate demographic, neuroimaging, neurosurgical and psychiatric factors associated with clinical outcomes. This process will be facilitated by the adoption of standardized diagnostic systems, and the use of wellvalidated psychiatric and other scales to quantify change in mood and other psychiatric symptomatology. Incorporating ‘‘lessons learned’’ from patients who have received DBS for movement disorders who have been evaluated for more than a decade, together with the growing psychiatric experience with TRD individuals, holds the promise of assisting in the better selection of candidates for psychiatric surgery and improving patient outcomes. Ideally the selection of candidates for DBS for TRD will be driven by knowledge of the biological or neuroimaging endophenotypes which predict response to this treatment. However, until there are clear predictors of response to DBS for TRD, the selection of appropriate candidates assumes great importance. The inclusion and exclusion criteria for the trial of BA25 DBS for TRD at the University of Toronto are presented in > Table 174-2. Given the complexity of this undertaking, DBS for TRD is ideally performed under the
DBS for TRD – the Role of the Psychiatrist It has been known for decades, that the attainment of superior clinical outcomes following neurosurgical procedures for psychiatric illness is dependent on the role of the psychiatrist in selecting suitable candidates and in providing quality post-operative psychiatric care to this population [44,45]. Given the emerging role of the psychiatrist in providing ‘‘triple therapy’’ to patients undergoing DBS for psychiatric indications, that is, delivering evidence-based psychotherapeutic and pharmacological treatments, as well as managing the electrical stimulation parameters to maximize benefit and minimize adverse effects of stimulation, the role of the psychiatrist in ensuring optimal post-surgical outcomes has never been more crucial [29].
2931
2932
174
Medical management and indications for surgery in depression
auspices of a multidisciplinary team comprised of neurosurgeons with an expertise in functional and stereotactic surgery, psychiatrists with an expertise in the assessment and the global management of mood disorders, neuropsychologists, imaging specialists, ethicists, nurses and allied health professionals. For these reasons, embarking on DBS for TRD is only feasible in specialized academic centers with the resources and expertise outlined above. At the University of Toronto, in order to facilitate the assessment process we have adopted the ‘‘Traffic Light’’ system of evaluating the suitability of candidates for this procedure. Patients referred to our center for evaluation are screened before their visit for the presence of an obvious exclusion criterion. Those deemed ineligible receive a ‘‘red light,’’ and are excluded from further consideration for the DBS protocol. Those who appear eligible are seen by a psychiatrist (PG, SK) on the DBS team. The initial consultation is to meet the individual and assess current medical and psychiatric status, as well as the previous treatment history, to determine the adequacy of previous pharmacological, psychological and somatic antidepressant treatments. Collateral information from family members and medical providers is essential to complete a comprehensive assessment. One of the primary goals of this phase of the psychiatric assessment is to strive towards achieving clarity in the psychiatric diagnosis. Given the lengthy medical history of most individuals seeking DBS for TRD, the use of a standardized diagnostic tool such as the Structured Clinical Interview for DSM Disorders (SCID) [47] as an adjunct to the clinical interview, is invaluable in reaching diagnostic clarity. An important goal is also to assess responses to all prior treatments, including evidence of at least a partial or temporary response to one or several treatments. That is, are the symptoms due to a unipolar Major Depressive Disorder, or are they occurring in the context of another Axis I or Axis II condition, such as Bipolar Affective Disorder or a personality
disorder, for which the efficacy of antidepressant therapies are not clearly established? This approach is similar to distinguishing between clinical conditions which are levodopa-responsive and those that are levodopa-unresponsive in a patient with parkinsonian symptoms, a distinction with prognostic implications for symptomatic improvement with DBS [48]. During the course of the assessment, if any potential candidates are deemed ineligible due to exclusion criteria, such as concurrent substance abuse or recent suicide attempt, do not have a primary psychiatric diagnosis of MDD or have not demonstrated sufficient degree of resistance to antidepressant treatments, these individuals are also classified as a ‘‘red light’’ and excluded. All others proceed to have an assessment with the other psychiatrist on the DBS team. A ‘‘yellow light’’ may be given in cases where further information or assessment is deemed necessary. This can occur when the patient’s self report or corroborating reports from previous treatments are suggestive of possible mania or hypomania; where an individual scores below the minimum severity score at assessment; or where further collateral information is required to assess the adequacy of prior antidepressant treatment. Patients with ‘‘yellow light’’ status may require several additional visits to complete their psychiatric assessment. Individuals with ‘‘green light’’ status are then referred to the neurosurgical team for assessment of their surgical suitability for the procedure. Monthly multidisciplinary DBS rounds provide a forum to discuss the status of patients in various stages of the assessment process. Typically it takes several months to complete the assessment phase of the protocol. This longitudinal assessment also provides some insight into patient motivation to receive the treatment, reliability of attending appointments and allows the patient and their family to meet with the DBS team and have their questions about the treatment answered.
Medical management and indications for surgery in depression
At the present time, a limited number of medical centers have the ability to provide DBS for patients with TRD. As a result, patients and their families may be required to travel some distance in order to attend the assessment and follow-up visits. A practical requirement for including patients in a DBS protocol is that they are available for at least 3 months for follow-up adjustment visits. Patients who are not located in close proximity to the academic center providing DBS should have existing medical supports in place in their local region.
The Role of Education The emergence of DBS and other neurostimulation techniques as potential treatments for psychiatric illness has garnered widespread attention in both the scientific and lay press. An analysis of the number of articles in the print media suggests that coverage of neurostimulation procedures has skyrocketed in the last decade [49]. It is estimated that 40–100 individuals worldwide have received DBS for depression or OCD, compared to over 30,000 patients with a primary diagnosis of Parkinson’s disease [50]. Despite this relatively limited experience for psychiatric indications, Racine and colleagues reported that 21% of neurostimulation articles in the popular press dealt with this technology for mental disorders, with the vast majority focusing on mood disorders [49]. Media reports of immediate improvement experienced by some individuals once stimulation was initiated in the operating room may engender unrealistic expectations in patients who TRD. As a result, an important part of the screening process is to assess the patient’s expectations and to provide education about the procedure, expected time course of the response, and the role of concurrent pharmacotherapy and ongoing psychiatric follow-up. Patients should be informed that as in DBS for Parkinson’s disease, sustained improvement in depressive symptomatology often requires months
174
and may necessitate multiple changes to their stimulation parameters [51]. An analogy that has proven to be helpful in explaining the process of recovery to our patients has been to liken the post-operative course to that of someone who has received a hip replacement after years of chronic arthritis. In the case of the patient undergoing hip replacement, the chronicity and severity of the arthritis may result in the development of compensatory strategies such as an antalgic gait with hypertrophy of contralateral muscles. For the patient with TRD, maladaptive processes such as dependency on others and catastrophic thinking may similarly become overutilized. In both cases, while the surgery may have corrected the underlying disease process, the patient requires ongoing rehabilitation to learn to ‘‘walk again’’ in the absence of the previous learned adaptations. This formulation emphasizes that the implantation of the DBS electrodes is not the end of the treatment process, but rather the beginning. Explaining this to patients during the assessment period, sets the stage for a process of psychiatric rehabilitation whereby it is understood that the patient plays an active role in their recovery.
Management of Psychiatric Symptoms Post-DBS – General Considerations The emergence of symptoms of mania or psychosis, dramatic worsening of depressive symptoms, including the expression of suicidal ideation in an individual receiving DBS should be considered behavioral emergencies, that prompt psychiatric attention. If such symptoms emerge soon after changes to stimulation parameters, then returning the parameters to a previously tolerated setting should be the first response. Following a change in the stimulation parameters, the patient should be observed for a period of time, lasting minutes to hours. In the absence of a robust observable improvement in mental status within this time,
2933
2934
174
Medical management and indications for surgery in depression
hospitalization for observation and safety should be considered. In the absence of improvement in the above symptoms following a change in stimulation parameters, the pharmacotherapy regimen should be reviewed and altered in the same way as it would be for any other treatment resistant patient. The optimal use of psychotropic medications in individuals receiving DBS for TRD has not been studied. Historical reports of patients who received ablative psychosurgery have suggested that patient sensitivity to previously ineffective medications may change, leading to increased effectiveness [52], as well as to the emergence of more dose-dependent medication side-effects [53]. Similarly, we observed that patients who received SCC DBS for TRD were able to decrease and in some cases discontinue medications used in the treatment of their depression [25] with no loss of response. An untested hypothesis is that surgical treatments for depression may in part exert their salutary effects due to a ‘‘priming’’ of the brain to respond to other antidepressant treatments. As larger cohorts of patients receive DBS for TRD, with the inclusion of some patients who are medication-free, it will be possible to explore this hypothesis.
Managing Stimulation Parameters for TRD In contrast to the rapidly observable changes elicited in PD patients once they receive DBS, it can take weeks to obtain optimal effects when treating TRD with DBS [27]. In addition to the delayed clinical effect in the treatment of depression, to date there are no clear guidelines to define the optimal stimulation parameters in the programming of DBS for TRD, as exists for PD [54]. This results in added complexity in determining the optimal DBS stimulation parameters for TRD.
In the absence of data to guide electrode location and stimulation settings, we have been influenced by the much larger PD experience at our center, and the phenotypic similarities between melancholic depression and PD. Without further insights into the mechanisms of action of stimulation of the brain, DBS remains an empirical therapy, and out of necessity, further refinement of stimulation parameters will only emerge from greater clinical experience with DBS for TRD [29,55]. The mean stimulation parameters used in our group for BA25 DBS in TRD are 4.0 V, 60 ms pulse widths, and a frequency of 130 Hz [25]. Although Schlaepfer reported that increased voltages were associated with greater symptomatic improvement in depressive symptoms in his cohort [28], the optimal dose range has not yet been established. In order to adequately assess the effect of a selected stimulator setting on improving the individual’s depression, our group aims not to make an adjustment more frequently than every 4 weeks and to avoid concomitant changes in pharmacotherapy. Once the appropriate electrode and stimulation parameters are achieved, the antidepressant effect is usually sustained without requiring alteration in DBS or pharmacotherapy.
Management of Worsening Depressive Symptoms Post-DBS Major Depressive Disorder is a characterized by symptoms which may wax and wane in the context of various endogenous and exogenous stressors [56,57]. Concordant with historical reports of ablative neurosurgery for various psychiatric disorders [58], and recent publications of DBS for TS [59], epilepsy [60] and OCD [33,61], a significant percentage of the patients in our TRD cohort have experienced a rapid improvement in affect during the intraoperative period and in the days that follow the surgery. It is not clear, however, if this rapid but frequently
Medical management and indications for surgery in depression
unsustained change in emotional experience and expression represents a microlesion effect from the insertion of the electrodes, as has been suggested in the PD literature [62,63], the paroxysmal effect of initiating stimulation of neural tissue, or a placebo effect. Patients should be informed that these early symptomatic changes, are part of the expected course of events during the first month post-surgery. In our experience, this represents an important time to re-explore the patient’s expectations for recovery, and if necessary challenge catastrophic thinking, but these fluctuation have not required changes in stimulation parameters. There is limited experience to guide the treatment of individuals who have not responded to DBS for TRD. The presence of implanted DBS electrodes does not appear to be a contraindication for receiving ECT [64].
Management of Suicidal Ideation Post-DBS The worsening of suicidal ideation and/or the emergence of a suicidal plan or intent is a psychiatric emergency. While it must be accepted that suicide is an inevitable outcome for a minority of TRD patients with and without DBS, there are potential lessons to be learned from the larger post-surgical experience of patients who have undergone this procedure for the treatment of movement disorders. In two longitudinal cohort studies of individuals receiving DBS for movement disorders, it was reported that 12.5% of patients expressed suicidal ideation during the first year post-implantation [65] and the rate of completed suicide was reported to be 4.3% [66]. In several reports of completed suicide in this population, the authors noted that the deaths occurred in individuals who demonstrated marked improvements in the motor symptoms [66–70]. There has also been a report of completed suicide in
174
a patient receiving DBS for OCD, despite marked reduction in OCD symptomatology [33]. As with the pharmacological management of depression with suicidal ideation, a decrease in psychomotor retardation may be associated with increased risk of suicide attempts or completion. As a result, asking about the presence of suicidal ideation should be a routine element of the functional inquiry during ongoing DBS. The relationship between DBS and suicide is likely to be complex, and multifactorial. Risk factors associated with suicide in movement disorder patients receiving DBS include being male [66,71], younger age of onset [66,71], the need for more than one DBS surgery [66], pre-DBS suicide attempts [71], worsening depressive symptoms [65,72] and recent stimulator changes [69]. The average time from surgery to completed suicide was approximately 3 years in movement disorder patients [66,71]. Of note, the routine involvement of psychiatric services was not part of the followup procedures reported by some groups [66,69]. These findings reinforce the need for long-term psychiatric care, both in the pre- and postoperative phases of this treatment, by psychiatrists who are part of the DBS team, to assist the patients and their families to adapt to the changes that emerge from this treatment. Several investigators have also explored the role of psychological readaptation to one’s interpersonal and professional circumstances as a potential contributor to suicide in the post-DBS period. In a longitudinal study of quality of life of Parkinson’s patients undergoing DBS, Houeto and colleagues reported that improvements in the motor symptoms of PD were not associated with patient or physician-rated improvements in indices of social support, marital and familial relations, and interpersonal communication [73]. The same group also described that the regaining of personal autonomy after improvements in parkinsonian symptoms with DBS, was associated with the deterioration in mental relationships [74]. As has been noted by patients following successful
2935
2936
174
Medical management and indications for surgery in depression
epilepsy surgery [75], others have had expectations of patients with respect to relationships, work, and other aspects of daily life that they cannot meet. A similar ‘‘rebalancing’’ of family dynamics may occur in patients who experience symptomatic improvement of their chronic depressive symptoms, disrupting previously established patterns of behavior.
Management of Other Psychiatric Symptoms Post-DBS The risk of adverse psychiatric and behavioral sequelae following DBS for PD has been recognized and there is growing clinical expertise in the assessment and management of these symptoms [70,71,76]. As with other treatments for depression, the clinician must be on the lookout for significant mood changes including emergent manic or hypomanic symptoms, psychotic symptoms or worsening anxiety [77]. Symptoms of mania include euphoric or irritable mood, racing thoughts, hypersexuality, decreased need for sleep, and grandiosity. Signs of mania include rapid or pressured speech, distractibility and increased psychomotor agitation. Mania can also be accompanied by psychotic symptoms, particularly mood congruent delusional beliefs as well as disorganized speech. A diagnosis of manic episode is made when three of more symptoms of mania in addition to a distinct period of abnormally and persistently elevated, expansive, or, irritable mood is present which last for 1 week, or irrespective of its duration requires hospitalization. Hypomania is defined as a period of persistently elevated, expansive or irritable mood lasting at least 4 days, which is not associated with the presence of psychotic symptoms, and neither require hospitalization, nor equivocally interferes with the individual’s level of functioning. Rates of emergence of mania in patients undergoing DBS for movement disorders have been estimated to between 0.9 and 4.0% by two
groups [70,71]. ‘‘Switching’’ from a depressed to manic state can occur with any antidepressant treatment [78,79], which requires the clinician to be vigilant to its presence. The emergence of manic or hypomanic symptoms should be considered a psychiatric emergency and should prompt investigation into exploring factors such as drug use, change in medications, new or worsening medical illness and recent changes in the DBS parameters, which may be contributing to the clinical presentation. Manic symptoms in the presence of DBS treatment should be initially managed with a decrease or cessation in stimulation until the factors contributing to the worsening mood state are clarified. Turning off the stimulation resulted in the disappearance of manic symptoms in a patient receiving internal capsule DBS for TS [35]. If a change in stimulation parameters does not produce immediate amelioration of the patient’s symptoms, then hospitalization with the goal of initiating mood stabilizing medications is a priority. A review of evidence-based treatment strategies for the management of mania is available [80]. Symptoms of anxiety are frequently present in patients with MDD. Epidemiological studies of community-based individuals estimate that over 40% of patients with a depression have a lifetime history of an anxiety disorder, and are at almost four times an increased risk of developing an anxiety disorder compared to those without depression. [81]. The presence of comorbid anxiety symptoms significantly decreases the likelihood of achieving remission of depressive symptoms with pharmacotherapy [82]. Anxiety disorders have been observed in less than 2% of patients receiving STN DBS for movements disorders [70]. Most reports in movement disorder patients have found that anxiety improves with DBS treatment [73,83,84], although some have reported a worsening in anxiety [74]. Improvements in anxiety symptoms with DBS have usually been correlated with improvements in the underlying condition being treated, although an improvement in anxiety was found to
Medical management and indications for surgery in depression
be independent of PD symptom amelioration in one study [83]. This finding suggests that the development of anxiety symptoms may be subserved by neural circuits which are anatomically distinct from those underlying the motor improvements seen in DBS for PD. It is also possible that anxiety may represent the manifestation of excessive non-specific stimulation of the brain. Studies of the relationship between stimulation parameters, anatomical location and the magnitude of anxiety symptoms are required to distinguish between these two possible mechanisms. The presence of anxiety symptoms and its impact on treatment response in patients receiving DBS for TRD is underexplored. In our cohort of 20 patients receiving SCC DBS for TRD, anxiety as measured by the Beck Anxiety Inventory was positively correlated (r = 0.58, p < 0.01) with HDRS-17 score at 12 months [27]. The management of treatment-emergent anxiety symptoms includes identifying and minimizing exacerbating factors, such as insomnia and caffeine use. In our experience, these symptoms have been amenable to reassurance, and if necessary, a decrease in the voltage administered.
174
placebo trials with a reasonable sample size have so far been reported. The neurosurgery-psychiatry team at University of Toronto has extensive experience with DBS to the subgenual cingulate cortex since 2003. From a psychiatric perspective, careful preliminary assessment, patient education and family support are essential. Post-surgical rehabilitation involves management of antidepressant medication, often at reduced doses, combined with cognitive behavioral strategies to reestablish work and leisure activities as well as achieve symptomatic remission. Psychiatric complications post-surgery can include relapse into depression with or without suicidal intent or rarely a switch into mania or hypomanic. Adjustments of stimulator settings, medication, or admission to hospital may be required. Importantly neuropsychological evaluation of patients before and after SCC DBS show no cognitive deterioration and in certain domains cognitive benefits. There is an urgent need for randomized controlled trials in this previously treatment resistant population of depressed patients.
References Conclusion While psychiatric disorders lack neurobiological markers for diagnostic confirmation, there are established criteria for severity, chronicity, and treatment resistance among patients with MDD. Advances in neurosciences and particularly functional neuroimaging have facilitated the emergence of neurocircuitry models for MDD and provide a rationale for stereotactic surgical interventions. There is more evidence to support DBS to subgenual cingulate cortex than other brain regions, although the anterior limb of the internal capsule and the ventral striatum/nucleus accumbens are also therapeutic targets. All three sites are currently under investigation in patients with TRD, although no randomized controlled
1. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry 1960;23:56-62. 2. Montgomery SA, Asberg M. A new depression scale designed to be sensitive to change. Br J Psychiatry 1979;134:382-9. 3. Nemeroff CB, Entsuah R, Benattia I, Demitrack M, Sloan DM, Thase ME. Comprehensive analysis of remission (COMPARE) with venlafaxine versus SSRIs. Biol Psychiatry 2008;63:424-34. 4. Kennedy SH, Andersen HF, Lam RW. Efficacy of escitalopram in the treatment of major depressive disorder compared with conventional selective serotonin reuptake inhibitors and venlafaxine XR: a meta-analysis. J Psychiatry Neurosci 2006;31:122-31. 5. Boulenger JP, Huusom AK, Florea I, Baekdal T, Sarchiapone M. A comparative study of the efficacy of long-term treatment with escitalopram and paroxetine in severely depressed patients. Curr Med Res Opin 2006;22:1331-41. 6. Rush AJ, Trivedi MH, Wisniewski SR, Nierenberg AA, Stewart JW, Warden D, Niederehe G, Thase ME, Lavori
2937
2938
174 7.
8.
9. 10. 11.
12.
13.
14.
15.
16.
17.
18.
Medical management and indications for surgery in depression
PW, Lebowitz BD, McGrath PJ, Rosenbaum JF, Sackeim HA, Kupfer DJ, Luther J, Fava M. Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: a STAR*D report. Am J Psychiatry 2006;163:1905-17. Caspi A, Sugden K, Moffitt TE, Taylor A, Craig IW, Harrington H, McClay J, Mill J, Martin J, Braithwaite A, Poulton R. Influence of life stress on depression: moderation by a polymorphism in the 5-HTT gene. Science 2003;301:386-9. Pittenger C, Duman RS. Stress, Depression, and neuroplasticity: a convergence of mechanisms. Neuropsychopharmacology 2008;33:88-109. McEwen BS. Mood disorders and allostatic load. Biol Psychiatry 2003;54:200-7. Mayberg HS. Defining neurocircuits in depression. Psychiatric Annals 2006;36:259-68. Mayberg HS. Modulating dysfunctional limbic-cortical circuits in depression: towards development of brainbased algorithms for diagnosis and optimised treatment. Br Med Bull 2003;65:193-207. Kennedy SH, Evans KR, Kru¨ger S, Mayberg HS, Meyer JH, McCann S, Arifuzzman AI, Houle S, Vaccarino FJ. Changes in regional brain glucose metabolism measured with positron emission tomography after paroxetine treatment of major depression. Am J Psychiatry 2001a;158:899-905. Kennedy SH, Konarski JZ, Segal ZV, Lau MA, Bieling PJ, McIntyre RS, Mayberg HS. Differences in brain glucose metabolism between responders to CBTand venlafaxine in a 16-week randomized controlled trial. Am J Psychiatry 2007;164:778-88. Liotti M, Mayberg HS, McGinnis S, Brannan SL, Jerabek P. Unmasking disease-specific cerebral blood flow abnormalities: mood challenge in patients with remitted unipolar depression. Am J Psychiatry 2002; 159:1830-40 Wu J, Buchsbaum MS, Gillin JC, Tang C, Cadwell S, Wiegand M, Najafi A, Klein E, Hazen K, Bunney WE Jr, Fallon JH, Keator D. Prediction of antidepressant effects of sleep deprivation by metabolic rates in the ventral anterior cingulate and medial prefrontal cortex. Am J Psychiatry 1999;156:1149-58. Dougherty DD, Weiss AP, Cosgrove GR, Alpert NM, Cassem EH, Nierenberg AA, Price BH, Mayberg HS, Fischman AJ, Rauch SL. Cerebral metabolic correlates as potential predictors of response to anterior cingulotomy for treatment of major depression. J Neurosurgery 2003;99:1010-7. Kennedy SH, Giacobbe P. Treatment resistant depression – advances in somatic therapies. Ann Clin Psychiatry 2007;19(4):279-87. Mataix-Cols D, Wooderson S, Lawrence N, Brammer MJ, Speckens A, Phillips ML. Distinct neural correlates of washing, checking, and hoarding symptom dimensions in
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
obsessive-compulsive disorder. Arch Gen Psychiatry 2004;61:564-76. Milak MS, Parsey RV, Keilp J, Oquendo MA, Malone KM, Mann JJ. Neuroanatomic correlates of psychopathologic components of major depressive disorder. Arch Gen Psychiatry 2005;62:397-408. Boon P, Vonck K, De Herdt V, Van Dycke A, Goethals M, Goossens L, Van Zandijcke M, De Smedt T, Dewaele I, Achten R, Wadman W, Dewaele F, Caemaert J, Van Roost D. Deep brain stimulation in patients with refractory temporal lobe epilepsy. Epilepsia 2007;48(8):1551-60. Lipsman N, Neimat JS, Lozano AM. Deep brain stimulation for treatment refractory obsessive-compulsive disorder: the search for a valid target. Neurosurgery 2007;61:1-13. Heath RG, Llewellyn RC, Rouchell AM. Brain mechanisms in psychaitric illness: rationale for and results of treatment with cerebellar stimulation. In: Hitchcock ER, Ballantine HT, Meyerson BA, editors. Modern concepts in psychiatric surgery. Amsterdam: Elsevier/NorthHolland Biomedical Press; p. 77-82. Aouizerate B, Martin-Guehl C, Cuny E, et al. 1979. Deep brain stimulation for OCD and major depression. Am J Psychiatry 2005;162:2192. Jime´nez F, Velasco F, Salin-Pascual R, Herna´ndez JA, Velasco M, Criales JL, Nicolini H. A patient with a resistant major depression disorder treated with deep brain stimulation in the inferior thalamic peduncle. Neurosurgery 2005;57(3):585-93. Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, Schwalb JM, Kennedy SH. Deep brain stimulation for treatment-resistant depression. Neuron 2005;45:651-60. Kuhn J, Lenartz D, Huff W, Lee S, Koulousakis A, Klosterkoetter J, Sturm V. Remission of alcohol dependency following deep brain stimulation of the nucleus accumbens: valuable therapeutic implications? J Neurol Neurosurg Psychiatry 2007;78(10):1152-3. Lozano AM, Mayberg HS, Giacobbe P, Hamani C, Craddock RC, Kennedy SH. Subcallosal cingulate gyrus deep brain stimulation for treatment-resistant depression. Biol Psychiatry 2008 Jul 16. [Epub ahead of print]. Schlaepfer TE, Cohen MX, Frick C, Kosel M, Brodesser D, Axmacher N, Joe AY, Kreft M, Lenartz D, Sturm V. Deep brain stimulation to reward circuitry alleviates anhedonia in refractory major depression. Neuropsychopharmacology 2008;33(2):368-77. Giacobbe P, Kennedy SH. Deep brain stimulation for treatment-resistant depression: a psychiatric perspective. Curr Psychiatry Rep 2006;8:437-44. McNeely HE, Mayberg HS, Lozano AM, Kennedy SH. Neuropsychological impact of Cg25 deep brain stimulation for treatment-resistant depression: preliminary results over 12 months. J Nerv Ment Dis 2008;196: 405-410.
Medical management and indications for surgery in depression
31. Greenberg BD, Malone DA, Friehs GM, Rezai AR, Kubu CS, Malloy PF, Salloway SP, Okun MS, Goodman WK, Rasmussen SA. Three-year outcomes in deep brain stimulation for highly resistant obsessive-compulsive disorder. Neuropsychopharmacology 2006;31(11): 2384-93. 32. Gabriels L, Cosyns P, Nuttin B, Demeulemeester H, Gybels J. Deep brain stimulation for treatment refractory obsessive-compulsive disorder: psychopathological and neuropsychological outcome in three cases. Acta Psychiatr Scand 2003;107:275-82. 33. Abelson JL, Curtis GC, Sagher O, et al. Deep brain stimulation for refractory obsessive-compulsive disorder. Biol Psychiatry 2005;57:510-6. 34. Van Laere K, Nuttin B, Gabriels L, Dupont P, Rasmussen S, Greenberg BD, Cosyns P. Metabolic imaging of anterior capsular stimulation in refractory obsessive-compulsive disorder: a key role for the subgenual anterior cingulate and ventral striatum. J Nucl Med 2006;47:740-7. 35. Flaherty AW, Williams ZM, Amirnovin R, et al. Deep brain stimulation of the anterior internal capsule for the treatment of Tourette syndrome: technical case report. Neurosurgery 2005;57(4 Suppl):E403. 36. Sturm V, Lenartz D, Koulousakis A, et al. The nucleus accumbens: a target for deep brain stimulation in obsessive– compulsive- and anxiety disorders. J Chem Neuroanat 2003;26:293-99. 37. Schlaepfer TE, Lieb K. Deep brain stimulation for treatment of refractory depression. Lancet 2005;366:1420-2. 38. Kopell BH, Greenberg BD. Anatomy and physiology of the basal ganglia: implications for DBS in psychiatry. Neurosci Biobehav Rev 2007. doi:10.1016/j.neubiorev.2007.07.004. 39. Tremblay LK, Naranjo CA, Graham SJ, Herrmann N, Mayberg HS, Hevenor S, et al. Functional neuroanatomical substrates of altered reward processing in major depressive disorder revealed by a dopaminergic probe. Arch Gen Psychiatry 2005;62:1228-36. 40. Nestler EJ, Carlezon WA Jr. The mesolimbic dopamine reward circuit in depression. Biol Psychiatry 2006;59:1151-9. 41. van Kuyck K, Gabrie¨ls L, Cosyns P, Arckens L, Sturm V, Rasmussen S, Nuttin B. Behavioural and physiological effects of electrical stimulation in the nucleus accumbens: a review. Acta Neurochir Suppl 2007;97(Pt 2):375-91. 42. Heath RG. Brain function and behavior. I. Emotion and sensory phenomena in psychotic patients and in experimental animals. J Nerv Ment Dis 1975;160 (3):159-75. 43. Velasco F, Velasco M, Jime´nez F, Velasco AL, SalinPascual R. Neurobiological background for performing surgical intervention in the inferior thalamic peduncle for treatment of major depression disorders. Neurosurgery 2005;57(3):439-48.
174
44. Poppen JL. Technic of prefrontal lobotomy. J Neurosurg 1948;5:514-20. 45. Ballantine HT Jr, Levy BS, Dagi TF, Girunas IB. Cingulotomy for psychiatric illness. In: Sweet WH, et al. editor. Neurosurgical treatment in psychiatry, pain and epilepsy. Baltimore, MD: University Park Press; 1977 p. 333-53. 46. Rodriguez RL, Fernandez HH, Haq I, Okun MS. Pearls in patient selection for deep brain stimulation. Neurologist 2007;13:253-60. 47. First MB, Spitzer RL, Gibbon M, Williams JBW. Structured Clinical Interview for DSM-IV Axis I Disorders, Clinician Version (SCID-CV). Washington, DC: American Psychiatric Press; 1996. 48. Krack P, Limousin Dowsey P, Benabid AL, et al. Ineffective subthalamic nucleus stimulation in levodopa-resistant postischemic parkinsonism. Neurology 2000;54:2182-4. 49. Racine E, Waldman S, Palmour N, Risse D, Illes J. ‘‘Currents of hope’’: neurostimulation techniques in U.S. and U.K. print media. Camb Q Healthc Ethics 2007;16:312-6. 50. Pereira EA, Green AL, Nandi D, Aziz TZ. Deep brain stimulation: indications and evidence. Expert Rev Med Devices 2007;4:591-603. 51. Ondo WG, Bronte-Stewart H, Group DS. The North American survey of placement and adjustment strategies for deep brain stimulation. Stereotact Funct Neurosurg. 2005;83:142-7. 52. Rylander G. Stereotactic radiosurgery in anxiety and obsessive-compulsive states: psychiatric aspects. In: Hitchcock ER, et al., editors. Modern concepts in psychiatric surgery. Amsterdam: Elsevier; 1979 p. 235-240. 53. Cosyns P, Gybels J. Psychiatric process analysis of obsessive compulsive behaviour modification by psychiatric surgery. In: Hitchcock ER, et al., editors. Modern concepts in psychiatric surgery. Amsterdam: Elsevier; 1979 p. 225-234. 54. Volkmann J, Moro E, Pahwa R. Basic algorithms for the programming of deep brain stimulation in Parkinson’s disease. Mov Disord 2006;21 Suppl 14:S284-9. 55. Lozano AM, Hamani C. The future of deep brain stimulation. J Clin Neurophysiol 2004;21:68-9. 56. Judd LL, Paulus MJ, Schettler PJ, Akiskal HS, Endicott J, Leon AC, Maser JD, Mueller T, Solomon DA, Keller MB. Does incomplete recovery from first lifetime major depressive episode herald a chronic course of illness? Am J Psychiatry 2000;157:1501-4. 57. Kendler KS, Thornton LM, Gardner CO. Genetic risk, number of previous depressive episodes, and stressful life events in predicting onset of major depression. Am J Psychiatry 2001;158:582-6. 58. Sachdev P, Sachdev J. Sixty years of psychosurgery. Aust NZ J Psychiatry 1997; 31:457-464. 59. Bajwa RJ, de Lotbinie`re AJ, King RA, Jabbari B, Quatrano S, Kunze K, Scahill L, Leckman JF. Deep
2939
2940
174 60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
Medical management and indications for surgery in depression
brain stimulation in Tourette’s syndrome. Mov Disord 2007;22:1346-50. Andrade DM, Zumsteg D, Hamani C, Hodaie M, Sarkissian S, Lozano AM, Wennberg RA. Long-term follow-up of patients with thalamic deep brain stimulation for epilepsy. Neurology 2006;66:1571-3. Nuttin BJ, Gabriels L, van Kuyck K, Cosyns P. Electrical stimulation of the anterior limbs of the internal capsules in patients with severe obsessive-compulsive disorder: anecdotal reports. Neurosurg Clin N Am 2003; 14:267-74. Wester K, Hauglie-Hanssen E. The prognostic value of intra-operative observations during thalamotomy for parkinsonian tremor. Clin Neurol Neurosurg 1992;94:25-30. Benabid AL, Koudsie´ A, Benazzouz A, Fraix V, Ashraf A, Le Bas JF, Chabardes S, Pollak P. Subthalamic stimulation for Parkinson’s disease. Arch Med Res 2000;31:282-9. Chou KL, Hurtig HI, Jaggi JL, et al. Electroconvulsive therapy for depression in a PD patient with bilateral subthalamic nucleus deep brain stimulators. Parkinsonism Relat Disord 2005;11:403-6. Berney, et al. Effect on mood of subthalamic DBS for Parkinson’s disease. A consecutive series of 24 patients. Neurology 2002;59:1427-9. Burkhard, et al. Suicide after successful deep brain stimulation for movement disorders. Neurology 2004; 63:2170-2. Doshi PK, Chhaya NA, Bhatt MH. Depression leading to attempted suicide following bilateral subthalamic nucleus stimulation for Parkinson’s disease. Movement Disorders 2002;17:1084-5. Albanese A, Piacentini S, Romito LM, Leone M, Franzini A, Broggi G, Bussone G. Suicide after successful deep brain stimulation for movement disorders. Neurology 2005;65:499-500. Foncke, et al. Suicide after deep brain stimulation of the internal globus pallidus for dystonia. Neurology 2006;66 (1):142-3. Temel Y, et al. Behavioural changes after bilateral subthalamic stimulation in advanced Parkinson disease: a systematic review. Parkinsonism Relat Disord 2006;12:265-72. Appleby BS, et al. Psychiatric and neuropsychiatric adverse events associated with deep brain stimulation: a meta-analysis of ten years’ experience. Mov Disord 2007;22(12):1722-8. Tir, et al. Exhaustive, one-year follow-up of subthalamic nucleus deep brain stimulation in a large, single-center cohort of parkinsonian patients. Neurosurgery 2007;61:297-305. Houeto, et al. Subthalamic stimulation in Parkinson disease: behavior and social adaptation. Arch Neurol 2006;63:1090-5.
74. Houeto, et al. Behavioural disorders, Parkinson’s disease and subthalamic stimulation. J Neurol Neurosurg Psychiatry 2002;72:701-7. 75. Ring, et al. A prospective study of the early postsurgical psychiatric associations of epilepsy surgery. J Neurol Neurosurg Psychiatry 1998;64:601-4. 76. Miyasaki JM, Shannon K, Voon V, et al. Practice parameter: evaluation and treatment of depression, psychosis, and dementia in Parkinson disease (an evidence-based review): report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 2006;66:996-1002. 77. Kennedy SH, Lam RW, Cohen NL, Ravindran AV. CANMAT depression work group: clinical guidelines for the treatment of depressive disorders. IV medications and other biological treatments. Can J Psychiatry 2001b;46 Suppl 1:38S-58S. 78. Post RM, Altshuler LL, Frye MA, Suppes T, Rush AJ, Keck PE Jr, McElroy SL, Denicoff KD, Leverich GS, Kupka R, Nolen WA. Rate of switch in bipolar patients prospectively treated with second-generation antidepressants as augmentation to mood stabilizers. Bipolar Disord 2001;3:259-65. 79. Vieta E, Martinez-Ara´n A, Goikolea JM, Torrent C, Colom F, Benabarre A, Reinares M. A randomized trial comparing paroxetine and venlafaxine in the treatment of bipolar depressed patients taking mood stabilizers. J Clin Psychiatry 2002;63:508-12. 80. Yatham LN, Kennedy SH, O’Donovan C, Parikh S, MacQueen G, McIntyre R, Sharma V, Silverstone P, Alda M, Baruch P, Beaulieu S, Daigneault A, Milev R, Young LT, Ravindran A, Schaffer A, Connolly M, Gorman CP, Canadian Network for Mood and Anxiety Treatments (CANMAT) guidelines for the management of patients with bipolar disorder: consensus and controversies. Bipolar Disord 2005;7 Suppl 3:5-69. 81. Hasin DS, Goodwin RD, Stinson FS, Grant BF. Epidemiology of major depressive disorder: results from the National Epidemiologic Survey on alcoholism and related conditions. Arch Gen Psychiatry 2005;62:1097-106. 82. Fava M, Rush AJ, Alpert JE, Balasubramani GK, Wisniewski SR, Carmin CN, Biggs MM, Zisook S, Leuchter A, Howland R, Warden D, Trivedi MH. Difference in treatment outcome in outpatients with anxious versus nonanxious depression: a STAR*D report. Am J Psychiatry 2008;165:342-51. 83. Higginson CI, Fields JA, Troster AI. Which symptoms of anxiety diminish after surgical interventions for Parkinson disease? Neuropsychiatry Neuropsychol Behav Neurol 2001;14:117-21. 84. Woods SP, Fields JA, Troster AI. Neuropsychological sequelae of subthalamic nucleus deep brain stimulation in Parkinson’s disease: a critical review. Neuropsychol Rev 2002;12:111-26.
Medical management and indications for surgery in depression
85. Aouizerate B, Cuny E, Martin-Guehl C, et al. Deep brain stimulation of the ventral caudate nucleus in the treatment of obsessive-compulsive disorder and major depression. Case report. J Neurosurg 2004;101:682-6. 86. Nemeroff CB. The burden of severe depression: a review of diagnostic challenges and treatment alternatives. J Psychiatr Res 2007;41:189-206. 87. Okun MS, Tagliati M, Pourfar M, Fernandez HH, Rodriguez RL, Alterman RL, Foote KD. Management of referred deep brain stimulation failures: a retrospective analysis from 2 movement disorders centers. Arch Neurol 2005;62:1250-5.
174
2941
171 Psychosurgery – A Historical Perspective C. R. Bjarkam . J. C. Sørensen
Summary This chapter describes the use of psychosurgery in a historical context demonstrating that surgical treatments of psychiatric symptoms have occurred throughout human history. The number of performed procedures, the used surgical techniques, and the public opinion towards these procedures has, however, been closely interconnected with concurrent developments in neurosurgery, neuroscience and psychopharmacology. We illustrate how modern psychosurgery, introduced by Egas Moniz and Almeira Lima in 1935, after a euphoric beginning, crowned with Egas Moniz receiving the 1949 Nobel Prize, rapidly fell into demise in the 1950s, due to uncritical and inappropriate use combined with the advent of electro-convulsive therapy and effective psychopharmacology. During the last half of the twentieth century psychosurgery was, accordingly, primarily performed in low numbers by few specialized neurosurgical units who stereotaxically targeted small well defined brain areas, demonstrating that such an approach was reasonably safe and effective on severe otherwise treatment resistant psychiatric cases. Recent efforts in neuroscience combined with the advent of advanced high resolution brain imagers and stereotaxic neuromodulation with deep brain stimulation, stem cells and gene therapy, however, hold promise for the future use of psychosurgery. A successful re-introduction of psychosurgery demand that such procedures are based on a continuous integration of basic neuroscience and technical developments in functional neurosurgery, evaluated by long-term #
Springer-Verlag Berlin/Heidelberg 2009
blinded prospective studies, and performed by experienced neurosurgeons on carefully selected patients in a multi-disciplinary setting.
Introduction Psychosurgery, e.g., surgical procedures performed on the brain in order to treat psychiatric diseases and symptoms, has led a turbulent life during the last century causing fierce debate within the medical profession, among lawmakers and in the general public [1–9]. Much of this debate is based on fundamental questions such as the neurobiology of psychiatry, the role of the free will, and the degree of biology underlying our behaviors, while misuse and malpractice of psychosurgery are other issues that regularly cause public anxiety [1–12]. A proper view on psychosurgery, however, demand that one considers it in a historical perspective and in context with concurrent developments in neuroscience, psychiatry, psychopharmacology and neurosurgery, because these concurrent events often laid the foundation for, changed, and in some cases even discredited the use of psychosurgery (> Figure 171-1). In the remaining part of this chapter we aim to provide a chronological description of the most influential psychosurgical procedures and place these in their proper historical context (> Figure 171-1). We demonstrate in that way, that psychosurgery newer was abandoned totally and that current neurobiological insight combined with the advent of advanced high resolution brain imagers and the widespread use of
2868
171
Psychosurgery – a historical perspective
. Figure 171-1 Timeline depicting the introduction of influential psychosurgical procedures and their relation to major advances in neuroscience, psychopharmacology and surgical/imaging techniques. Note how the public opinion of psychosurgery has changed forth and back during the last century
stereotaxic neuromodulation with deep brain stimulation, stem cells and gene therapy, may lead to a general revitalization of psychosurgery [2–4,13–15]. The past history of psychosurgery implies that a successful re-introduction of psychosurgery must be based on a continuous integration of basic neuroscience and psychiatry with technical developments in functional neurosurgery, evaluated by long-term blinded prospective studies, and performed by experienced neurosurgeons on carefully selected patients in a multidisciplinary setting with ethical, neurosurgical, psychiatric and psychological expertise.
Psychosurgery in Ancient Times and Primitive Cultures Psychosurgery probably dates back a long time through human history, as several skulls dating
back to Neolithic time shows recovery after successful trephination [16,17]. One may well imagine that some of these trephinations were performed by the tribe shaman on persons displaying psychiatric or epileptic symptoms and in that way considered sick or possessed by demons or evil spirits. Anecdotal knowledge on the relation between brain/head trauma and loss of cognitive abilities and or development of psychiatric symptoms, likewise, occur in religious writings, literature and paintings dating back to the earliest times (> Figure 171-2) [2,3,8,9,18–20]. The first widely known case to illustrate the close connection between brain function and cognitive and behavioral abilities was, described by John Martyn Harlow who was the physician of an American rail road worker named Phineas Gage. Phineas Gage was on September 13, 1848, while using a tamping iron to pack explosive
Psychosurgery – a historical perspective
. Figure 171-2 Painting by Hieronymus Bosch (1450–1516) demonstrating that the concept of psychosurgery not was unknown to the fifteenth century man. The painting is entitled ‘‘The cure of folly or the operation for the stone’’ the latter referring to the contemporary belief that madness was caused by brain stones [18,19]. The painting can be seen at the Prado National Museum, Madrid, Spain
powder into a rock, triggering an uncontrolled explosion causing the 109 cm long, 3 cm thick and 13 pound heavy iron bar to pass through his left check, eye and frontal cranium resulting in severe left frontal lobe damage (> Figure 171-3) [2,21–23]. Gage was only unconscious for a short period after the accident and apart from left eye amaurosis suffered no additional sensory or motor dysfunction. John M. Harlow noted however in his writings that Gage’s personality was so radically changed that his friends and acquaintances said ‘‘he was no longer Gage’’ [20].
171
. Figure 171-3 Digitally remastered illustration of Phineas Gage’s dramatic incident causing severe and permanent personality changes. This case elegantly described by Gage’s physician John M. Harlow [20] is probably the first widely known description on the relation between frontal damage and personality changes indicating that certain brain areas have a major role in cognitive and mental abilities. The picture is taken from Ratiu and Talos beautiful digitally remastered illustration of Gages incident [21]
Gage who before the accident was a capable foreman for his colleagues, responsible and altruistic became after the accident immature, aggressive and childishly selfish [2,21,23]. The case of Phineas Gage was widely propagated within the medical community and consolidated together with the concurrent cases presented by Paul Broca [24,25] and Carl Wernicke [26] the importance of the brain in higher mental and cognitive functioning. It would therefore be reasonable to assume that the late nineteenth century doctor knowing these cases would begin to consider medical and surgical interventions that directly targeted ‘‘unordered’’ brain areas for the correction of psychiatric symptoms and disturbed mental abilities [2].
2869
2870
171
Psychosurgery – a historical perspective
Gottlieb Burckhardt Introduces Surgical Treatment of Psychosis in 1888 Being under the influence of the cases described above, the Swiss psychiatrist Gottlieb Burckhardt, who was director of the Pre´fargier Asylum in Switzerland [2], proposed after initial studies on dogs in the late 1880s that bilateral cerebral excision at multiple foci in the frontal, parietal and temporal lobes could be a useful therapy against psychosis [27]. Burckhardt tried his procedure on six psychotic patients starting in 1888 and presented his results to the medical society in the ¨ ber rindexcisionen, als bei1891 paper entitled: ‘‘U trag zur operativen therapie der psychosen’’ [27]. Burckhardt claimed that three patients had a successful outcome, two were partial successes and one a failure (died) [2,27]. His work was, however, not favorably received by the Swiss medical community and this combined with neuroscience and neurosurgery still being in its infancy caused psychosurgery to be paused for the next 40 years, and the work of Burckhardt largely to be forgotten by his successors.
Egas Moniz Introduces the Prefrontal Leucotomy in 1935 In 1935 Carlyle Jacobsen and John Farquhar Fulton presented a primate study on the role of the frontal lobe in short-term memory at the Second International Congress of Neurology held in London [28]. In their study two monkeys were tested for their ability to remember different cards, an ability that was considerably diminished after resection of the frontal lobes. Jacobsen and Fulton, however, made the peculiar observation that one of the monkeys (Becky) before the frontal lobe resection was anxious towards attending the card game and showed prominent aggression when she failed the game, whereas the same animal after the frontal lobe resection
seemed undisturbed by the game setting and furthermore even though she played the game more poorly after the surgical procedure, she did not display signs of aggression when loosing (> Figure 171-4) [28]. The study of Jacobsen and Fulton emphasized the role of the frontal lobes in anxiety and aggressive behavior, and probably, together with the nineteenth century cases described above, provided the clinical rationale for the Portuguese neurologist Egas Moniz, who attended the Fulton lecture in London to consider surgery on the frontal lobes for the treatment of certain psychoses [2–5,8,9,29,30]. Egas Moniz (1874–1955) returned to Lisbon and together with the neurosurgeon Almeida Lima, he introduced a procedure named the prefrontal leucotomy that were performed on a few selected psychiatric patients
. Figure 171-4 Image displaying John Fulton together with one of his primate research objects. Fulton’s research on the primate brain was of great importance for the introduction of psychosurgery in the beginning of the twentieth century. The original photograph is with courtesy of Yale University, Harvey Cushing/John Hay Withney Medical library presented in a paper commemorating the fiftieth anniversary of Moniz’s Nobel Prize by Ann Jane Tierney [5]
Psychosurgery – a historical perspective
171
. Figure 171-5 Goggled photographs of Egas Moniz (left) and Almeida Lima who introduced the prefrontal leucotomy in 1935 [29]
with prominent psychiatric symptoms such as depressive or psychotic anxiety and psychomotor agitation (> Figure 171-5) [29]. The prefrontal leucotomy (> Figure 171-6) was performed in order to disconnect the prefrontal part of the frontal lobes from the remaining part of the brain and thereby release the latter from a putative harmful (anxiety generating) prefrontal influence. The procedure was initially performed by destroying the white mater in centrum semiovale with injection of absolute alcohol into the subcortical white matter of the prefrontal cortex and later on (after the first seven cases) performed with a special made instrument, a leucotome (> Figure 171-7), introduced into the brain through one or more precoronal burr holes placed on each side of the skull (> Figure 171-6) [2–5,8,9,29,30]. Egas Moniz noted in his first monograph based on 20 patients that seven were cured, seven improved and six patients unchanged [29]. In a later paper on the prefrontal leucotomy [30] he noted that using the described technique he only saw minor and temporary disturbances referable to injury of the frontal lobes. He, furthermore, stated that the procedure could
result in prominent recovery although deteriorated patients only obtained slight or no recovery from the treatment [30]. "
‘‘Prefrontal leucotomy is a simple operation, always safe, which may prove to be an effective surgical treatment in certain cases of mental disorder.’’ Egas Moniz, 1936 [30]
Walther Freeman and James Watts Introduce the Prefrontal Lobotomy The number of prefrontal leucotomies performed by the Lisbon group newer exceeded more than 100 cases in total [5]. Moniz’ work had, however, a great impact on the neurologist Walther Freeman who together with the neurosurgeon James Watts rapidly and with great enthusiasm introduced psychosurgery to the United States in the late 1930s [2,3,5,8,9,32]. Freeman and Watts modified soon Moniz prefrontal leucotomy procedure in order to make larger lesions and in that way obtain a larger effect on the psychiatric symptoms [2,33–35]. The procedure was performed by the
2871
2872
171
Psychosurgery – a historical perspective
. Figure 171-6 Schematic illustration from E Hitchcock, L Laitinen and K Vaernet, Psychosurgery: Proceedings of the second International Conference on Psychosurgery. Charles C. Thomas. Springfield, 1972, demonstrating how to perform a prefrontal leucotomy. Precoronal burr holes were placed 1–1.5 cm in front of a vertical line passing through the base of the tragus, and 3 cm to either side of the midsaggital line. The leucotome was then introduced to a depth of 4.5 cm in an anterolateral direction, the loop opened and the instrument turned in order to cut a core of 1 cm in diameter in the white matter. The leucotome was then retracted 1 cm and the procedure repeated, followed by another 1 cm retraction and procedure repetition. The leucotome was then withdrawn entirely and re-introduced in an anteromesial direction to a depth of 4 cm where the first core was cut, and the leucotome withdrawn 1 cm in order to cut a second core at 3 cm before a final core was cut at 2 cm [30]
insertion of a calibrated blade with rounded edges and movable sidearm (named a precision leucotome) into the cerebral prefrontal white mater through more lateral placed burr holes [33–35]. The white matter was then severed by movements of the precision leucotome in the coronal plane (> Figure 171-8). The procedure was named the minimal, standard or radical prefrontal lobotomy, respectively, depending on the volume of severed brain tissue [2,33–35]. The minimal prefrontal lobotomy was primarily performed on
. Figure 171-7 Drawing of a leucotome reproduced from [31]. When introduced to its proper place in the white matter the stylet handle is pressed down, whereby a steel loop or a steel band protrudes from the instrument for resection of white matter cores when the instrument is rotated around a vertical axis [30]. The steel loop version was initially used, but Moniz noted that it tended to crush the white matter and it was therefore replaced by the steel band version that cuts rather than compresses the white matter [30]
. Figure 171-8 Schematic drawing of the prefrontal lobotomy procedure performed with a precision leucotome, invented and propagated by Freeman and Watts [33–35]. This procedure was also known as the minimal, standard or radical lobotomy, respectively, depending on the amount of brain tissue severed [2]. The illustration is taken from W Freeman and J Watts [35, p. 42]
patients with affective psychoneurotic disorders, while the larger scale radical prefrontal lobotomy was for schizophrenic patients or patients requiring reoperation due to lacking effect from their first lobotomy [2]. It is notable that Freeman and Watts in their initial writings on their procedure stated that ‘‘every patient probably loses something by this operation, some spontaneity, sparkle and flavor of the personality,’’ and that ‘‘indiscriminate use of prefrontal lobotomy could result in vast harm’’ and the procedure should be reserved for a small
Psychosurgery – a historical perspective
group of specially selected cases in which conservative methods of treatment have not yielded satisfactory results [33]. They noted, furthermore, like Moniz [30] that deteriorated patients probably not would benefit of the procedure and that best results would be obtained with patients showing symptoms of agitated depression and anxiety [33,34].
171
. Figure 171-9 Photograph from the Museum of the History of Medicine of the University of Zu¨rich showing cell belt, straitjacket, chains and covered bath tub (72 150 69 cm) used to restrain agitated psychiatric patients at Burgho¨lzli Hospital, Zu¨rich
Psychosurgery is Initially Meet with Enthusiasm The introduction of psychosurgery in the late 1930s was, however, greeted with enthusiasm within the medical community and in the public due to the lack of other ‘‘effective’’ treatments towards severe psychiatric symptoms [2–5,8,9,31,36]. People suffering from chronic psychosis and severe affective disturbances were kept in over crowed asylums for years, where the hygienic and nursing standard in many cases were kept at a minimum resulting in frequent outbreaks of infectious diseases such as e.g. tuberculosis [2,3,8,9,31]. Aggressive patients were at best treated with barbiturates but more often locked up for long periods in locked bath tubes, chains or strait-jackets (> Figure 171-9). Many professionals and patient relatives therefore saw psychosurgery as a possible way out of an institutionalized life or a least offering a possibility to dampen aggressive and anxious patients, so that they were manageable in the asylum without needing more or less permanent restraint in cellars and chains [2,3,8,9,31]. The initial conservative use of psychosurgery as a last resort for a small group of patients displaying severe psychiatric symptoms was therefore slowly but gradually changed in the early 1940s towards a more aggressive and indiscriminate use of the procedure [2,3,8,9,31]. The use of prefrontal lobotomy was now in contrast to Moniz’s and Freeman’s earlier findings described to be beneficial even in the treatment of more deteriorated patients such as severe chronic
schizophrenics [37]. The need for psychosurgery, accordingly, increased during the second world war due to expanded indications for its use and because the already over crowded state hospitals at that time received a steadily increased number of war-traumatized patients [2,3,8,9,31,36].
Freeman Introduces the Transorbital Lobotomy in 1946 Walther Freeman seems to have lost his caution towards psychosurgery under these devastating circumstances and, probably under the inspiration of the work by the Italian psychiatrist Amarro Fiamberti [38] invented the technically simpler transorbital lobotomy in 1946. This procedure was performed under electroshock induced unconsciousness by inserting an ice pick
2873
2874
171
Psychosurgery – a historical perspective
like instrument (named a steel orbitoclast or a transorbital leucotome) bilaterally through the thin bony orbit above the eye into the white matter of the frontal lobes hereby severing it by consecutive movements of the instrument in the coronal plane (> Figure 171-10) [35]. This procedure took only 15–20 min and required in
. Figure 171-10 Schematic drawing of the transorbital lobotomy procedure invented by Walther Freeman in 1946. The illustration is taken from W Freeman and J Watts [35, p. 52]
Freemans opinion not necessarily the attendance of neurosurgeons or anesthetists as well as the use of sterile technique and proper surgical settings often were omitted [2–4,8,9]. Freeman who now was abandoned by Watts traveled across the United States in his Winnebago camper called the ‘‘lobotomobile’’ visiting state hospitals, asylums, and even motel rooms with his orbitoclast in order to fulfill, what in his eyes was an unanswered need for psychosurgery (> Figure 171-11) [2–4,8,9]. The ease whereby psychosurgery now could be performed combined with the demand for effective treatments toward severe psychiatric symptoms led to a tremendous rise in the number of performed lobotomies. It has been estimated that in United States fewer than 300 lobotomies were performed in 1945 whereas more than 5,000 lobotomies each year where performed in 1949, 1950, and 1951, respectively [5,39]. The increased use of psychosurgery was done in full publicity and crowned with the award of the 1949 Nobel Prize in medicine or physiology to be shared between Egas Moniz ‘‘for his discovery of the therapeutic value of leucotomy in certain psychoses,’’ and Walter Rudolf
. Figure 171-11 Walther Freeman ‘‘on tour’’ demonstrates his transorbital lobotomy procedure for an interested public. Note the apparent lack of sterile technique and proper surgical setting
Psychosurgery – a historical perspective
Hess ‘‘for his discovery of the functional organization of the interbrain as a coordinator of the activities of the internal organs’’ [40].
Psychosurgery Comes in Disrepute Psychosurgery was no longer reserved for the selected few but instead performed in high numbers outside the neurosurgical ward and often performed by non-neurosurgeons e.g., orthopedics, neurologists or psychiatrists [2–5,8,9,41]. The lacking caution towards psychosurgery is superbly illustrated by the following highlighted citation from the Norwegian psychiatrist Ørnulf Ødega˚rd, who for 34 years was the director of Norway’s main mental hospital Gaustad in Oslo. "
‘‘psychosurgery can be easily performed by the psychiatrist himself with the tools he might have in his pocket, and strangely enough it may be harmless and effective’’ Ørnulf Ødega˚rd, 1953 [12,42].
It is clear that psychosurgery in it self often being performed as a closed procedure through superficial burr holes or through the roof of the orbita could lead to damage of cerebral vessels resulting in severe or fatal intracerebral hemorrhage, even for the neurosurgeons, but these incidents were most probably increased together with the risk of cerebral infections and unwanted neuronal damage when these procedures were taken over by non-neurosurgeons (> Figure 171-11). Morbidity and mortality associated with psychosurgery, accordingly, raised to two-digit numbers at some hospitals during this period [10]. The expanded indications for psychosurgery, likewise, meant that these procedures in some instances was used on unwilling patients without their consent and in some cases also used on prisoners or youngsters in correctional institutions in order to correct dysfunctional behavior [2,10].
171
This practice naturally caused some public concern, which was further aggravated by the high morbidity and mortality numbers which was becoming more and more evident to the public as the number of performed procedures increased [2,3,8,9]. The final blow to the generalized use of psychosurgery was, however, caused by the introduction of lithium in the late 1940s [43,44] and the first effective antipsychotic drug chlorpromazine (> Figure 171-12) in the mid-1950s [2–5,8,9,31,41]. The introduction of these drugs combined with the general acceptance of electroconvulsive therapy (ECT, introduced in 1938) [45] as an effective treatment of severe depression meant that the psychiatrists suddenly had effective non-surgical treatments towards severe psychotic and affective symptoms. The beneficial and adverse effects of psychosurgery were, accordingly, suddenly compared to treatment regimes based on antipsychotic drugs and ECT resulting in favoritism towards the latter. This change was furthermore supported by the irreversible nature of the introduced psychosurgical procedures and the raised ethical concerns associated with its practice [2–5,8,9,31,41]. The number of performed psychosurgical procedures decreased with great haste toward . Figure 171-12 The molecular structure of chlorpromazine, the first effective anti-psychotic to be used in 1952 [41]. The invention of effective antipsychotics together with lithium introduced in the late 1940s [43,44] and ECT introduced in 1938 [45] meant that psychosurgery was evaluated in the context of the former treatment modalities. This evaluation was generally unfavorable
2875
2876
171
Psychosurgery – a historical perspective
. Figure 171-13 Picture from the 1975 movie One Flew over the Cuckoo’s Nest based on a 1962 novel with the same title by Ken Kesey. The novel and subsequent movie, which gained a wide audience, raised grave criticism against the psychiatric system and its use of psychosurgery to pacify (punish) maladjusted subjects, exemplified by the rebel Randle Patrick McMurphy played excellently by the American actor Jack Nicholson [11]
United Kingdom with 15,000–20,000 cases [36] and the Scandinavian counties (Denmark, Sweden, and Norway) with 9,000–10,000 cases [10,12]. The high number of performed cases emphasize that at least a reasonable number of these cases, by the contemporaries of that period, must have been considered to have a satisfying outcome and acceptable adverse effects.
The Introduction of Stereotaxic Psychosurgery
the end of the 1950s well supported by a negative publicity as exemplified by Ken Kesey’s novel One Flew over the Cuckoo’s Nest published in 1962 and adapted for the screen by Milos Forman in 1975 (> Figure 171-13).
The Heyday of Mass Psychosurgery Came to an End During the 1950s For the reasons mentioned above psychosurgery was generally abandoned by the end of the 1950s. It is estimated from the prevailing literature that the number of lobotomies performed in total during this period worldwide averaged 70,000– 90,000 cases with USA accounting for 50,000– 60,000 lobotomy cases [2,3,8,9,36], followed by
The high morbidity and mortality rates associated with the traditional closed psychosurgical procedures mentioned above had, already during the 1940s raised considerable concern among some neurosurgeons although they found the idea of psychosurgery scientifically sound. They maintained that smaller and more specific brain structures should be targeted in a proper neurosurgical setting only, with optimized surgical techniques in order to cause minimal surgical trauma to neighboring brain structures. Some of these pioneers such as William Beecher Scoville and Hugh Cairns abandoned, accordingly, the closed procedures and introduced instead open psychosurgical techniques targeting more restricted brain areas by selective undercutting of the orbitofrontal cortex [46] and the cingulum bundle in the anterior cingulate gyrus [47], respectively. Many of these open techniques were, during the next decades, replaced by procedures using the human stereotaxic technique, developed by Ernest A. Spiegel and Henry T. Wycis in 1947 (> Figure 171-14) [48–50]. Spiegel and Wycis based initially their stereotaxic procedure on internal cerebral landmarks visualized by pneumoencephalography, and later on in the 1950s when ventricular contrast agents were developed on ventriculography [48,49,52]. The later method soon became the general standard method to calculate target
Psychosurgery – a historical perspective
. Figure 171-14 Photograph of the frame system developed by Spiegel and Wycis in 1947 for human stereotaxy [48–50]. This technique, initially described to be used in the primate cerebellum by Horsley and Clarke in 1908 [51], enabled the neurosurgeon to reach deeply situated brain structures with minimal damage to neighboring areas and represented for some neurosurgeons a way to obtain the beneficial effects of psychosurgery without the high mortality and morbidity rates associated with the early lobotomy techniques [50]. The image of the original Spiegel-Wycis apparatus which now is in the Smithsonian Institute is taken from Philip L. Gildenbergs enlightening paper on the early days of human stereotaxy [48]
coordinates for stereotaxy until CT and shortly thereafter MRI and use of surgical planning computers revolutionized the field of stereotaxy in the last decade of the former millennium. Psychosurgically Spiegel and Wycis used their new procedure to target the dorsomedian nucleus of the thalamus in order to treat agitation and psychosis [48–50,53]. Other stereotaxic procedures in psychosurgery [2] were the bilateral amygdalotomy [54], hypothalotomy for the treatment of sexual violence [55] or
171
posteromedial hypothalotomy for schizophrenia complicated by agitation and destructiveness [56,57], anterior cingulotomy [58–61], anterior capsulotomy [62–64], subcaudate tractotomy [65,66] and limbic leucotomy [32,67]. Among these only the later four procedures have been in regular use at few highly specialized centers and these procedures will, accordingly, be presented chronologically further below.
Hugh Cairns Introduces the Anterior Cingulotomy in 1948 James W. Papez had in 1937 introduced his proposed circuit of emotion stating that neural information concerning emotions was propagated between the cingulate and parahippocampal gyrus, hypothalamus and thalamus by specific fiber bundles such as the cingulum, the fornix, the mammillothalamic tract, and the thalamocortical radiation [68]. This influential concept which later evolved into the theories of the limbic system [69–71] probably formed the theoretical rationale for Sir Hugh Cairns (> Figure 171-15) when he tried to disconnect the frontal/anterior part of the ‘‘unorderly working’’ cingulate gyrus from the remaining limbic structures described above by severing the cingulum bundle [47]. This procedure was in 1967 converted into a stereotaxic procedure by H. T. Ballantine [58] (> Figure 171-15) and is today the most reported neurosurgical procedure for psychiatric disease in the United States and Canada [4]. The anterior cingulotomy procedure is performed by MRI guided stereotaxic placement of thermocoagulative electrodes in the cingulum bundle between the anterior and mid third of the cingulate gyrus located above the corpus callosum [60,72,73]. Thermocoagulative lesions large enough to disconnect the cingulum are then made at the electrode location [60,72,73] (> Figure 171-16). Anterior cingulotomy is used
2877
2878
171
Psychosurgery – a historical perspective
. Figure 171-15 Goggled images of Sir Hugh Cairns the inventor of the anterior cingulotomy, and H.T. Ballantine who converted the operation into a stereotaxic procedure in 1967
. Figure 171-16 Sagittal (a) and coronal (b) postoperative MR-images illustrating the placement and the extent of the electrolytic lesions made in order to perform an anterior cingulotomy. The images are from Spangler et al. [73]
against chronic pain [58], treatment resistant major depressive disorder [59,61] and obsessive-compulsive disorder [60,61]. It is reported that up to 40–60% of these patients undergoing anterior cingulotomy may be improved [59,61]. Side effects have been reported to be few [58–61], consisting mainly of seizures (1%) and urinary incontinence [59,60,72]. A notable number of suicides (1–9%) [59,60] has, however, been reported among the treated
patients, although a part of these cases probably can be attributed to the underlying psychiatric disease.
Jean Talairach Introduces the Anterior Capsulotomy in 1949 It was probably autopsy finding and their correlation with clinical effects after frontal lobotomy that
Psychosurgery – a historical perspective
171
. Figure 171-17 Goggled photographs of Jean Talairach the inventor of the anterior cingulotomy and Lars Leksell who propagated the technique in Sweden were it also was performed with the gamma knife in some instances. The image of Jean Talairach is taken from www.rneurocirugia.com/media/pagina18.jpg, while the image of Lars Leksell is taken from www.gammaknife.gazi.edu.tr
gave Jean Talairach (> Figure 171-17) the idea to disconnect frontothalamic fibers e.g. fibers running between the subgenual anterior cingulate cortex and the orbitofrontal cortex, and the medial, anterior and dorsomedial thalamic nuclei [4,56,64]. The procedure represented in that way a restricted lobotomy procedure and was performed stereotaxically by electrocoagulation placed between the anterior and mid third of the anterior limb of the internal capsule at the level of the interventricular foramen [64]. Lars Leksell (> Figure 171-17) propagated this procedure in Sweden were the procedure also have been performed with the gamma knife [56,63]. The anterior capsulotomy has been reported to be especially effective against obsessivecompulsive disorder with a long lasting improvement among 48–78% [4,56,62,63,74]. Side effects are generally few and transient consisting of headache, confusion, urinary incontinence and weight gain. The most prevalent side effect is tiredness and lack of initiative, this lethargia, although
generally transient, has been reported to last for several weeks or months and at least in one case resulting in a persistent mild apathy dominant frontal lobe syndrome [4,56,61–63,74].
Geoffrey Knight Introduces the Subcaudate Tractotomy in 1961 A similar restricted frontal lobotomy aiming at severing the connections between the orbitofrontal cortex and related thalamic and limbic structures was performed by Scoville as an open procedure in the late 1940s [46]. Geoffrey Knight modified this procedure initially by restricting the surgical undercutting to the last 2 cm of the original lesion where it entered the subcudate region, and latter in 1961 by converting the open hand procedure to a stereotaxic one [46,65,66,75,76]. Stereotactic subcaudate tractotomy was initially performed by bilateral stereotaxic insertion of 1 7 mm long yttrium
2879
2880
171
Psychosurgery – a historical perspective
(Y90) rods in the frontal white matter beneath and just anterior to head of the caudate nucleus (> Figure 171-18) [65,66,75]. Short lived beta radiation emitted from the yttrium rods would then destroy neighboring tissue up to 2 mm from the surface of the rods resulting in bilateral lesions of 20 20 5 mm in the frontal white matter (> Figure 171-18) [65,66,75]. Currently, lesions are, however, made in a similar way by electrocoagulation due to discontinued production of yttrium in 1995 [65,76]. Stereotactic subcaudate tractotomy has primarily been used towards treatment resistant affective disorders, obsessive-compulsive disorder and anxiety [65,77–79] and response rates between 40–70% have been reported for this procedure against the diseases listed above [65,76–79]. More than 1,300 patients have been treated with stereotactic subcaudate tractotomy in the UK and the reports on adverse effects from these procedures reveal that one death could be related directly to the procedure whereas more common adverse
effects were transient postoperative disorientation (10%) and seizures (1.8%) [65,76–79].
Desmond Kelly and Alan Richardson Introduce the Limbic Leucotomy in 1973 It was often noted that the former mentioned stereotaxic restricted lobotomy procedures and especially the anterior cingulotomy were too small to be effective. Kelly and Richardson proposed on that background a procedure, the limbic leucotomy, that essentially combined the subcaudate tractotomy with the anterior cingulotomy performed by stereotaxic thermocoagulative lesioning at both target areas (> Figure 171-19) [3,4,13,32,67,80,81]. The limbic leucotomy would in that way sever orbitofrontal fibers as well as disrupt fibers in Papez’s proposed circuit of emotion [13,68]. Limbic leucotomy has like the previously described stereotaxic procedures
. Figure 171-18 Illustration of stereotactic subcaudate tractotomy as it initially was performed by insertion of two rows of 3–5 Yttrium rods bilaterally into the white matter of the frontal lobe just beneath and anterior to the head of the caudate. The schematic illustration on the left is taken from a 1998 paper by Malhi and Bartlett [76], while the radiograph demonstrating the bilateral placement of the two rows of yttrium (Y90) rods in the subcaudate white matter is taken from a 1969 paper by Knight [66]
Psychosurgery – a historical perspective
171
. Figure 171-19 MRIs illustrating the lesion location in the limbic leucotomy procedure which essentially is a combined cingulotomy and subcaudate tractotomy. Axial T2-weighted MRIs demonstrating lesions in the anterior cingulate gyrus (a) and the subcaudate white matter (b). Sagital T1-weighted MRI displaying both lesions are at the same time (c). The MRIs are taken from the 2002 paper by Montoya et al. [80]
primarily been used against treatment resistant depression, and obsessive-compulsive disorder with response rates around 36–50% [4,67,82]. Side effects are likewise seen in low to moderate numbers and consist of transient headaches, lethargy, urinary incontinence and seizures [4,32,80,82].
Stereotaxic Psychosurgery on Restricted Brain Areas is Reasonable Effective and Safe The four stereotaxic procedures described above have during the last 50 years been performed on more than 5,000 patients suffering from severe and otherwise treatment resistant depression, anxiety and obsessive-compulsive disorder. Although these patients were severely ill, the instituted procedures have generally been effective in more than half of the cases and the side effects been kept to a reasonably low level with transient headaches and confusion being the most prevalent, while permanent deficits such as seizures, apathy and urinary incontinence was seen among less than 5%. One may thus conclude, that even though the public
opinion in the same period generally have been most unfavorable towards psychosurgery, including the before mentioned stereotaxic procedures, the latter has when it comes to efficacy and safety definitively been most justified. This standpoint is underscored by the examination of the practice of psychosurgery by the United States Congress after an intense public debate on psychosurgery in 1974 [3,81]. The Congressional Committee issued in 1977, most surprisingly to psychosurgery antagonists, a favorable report supporting the use of psychosurgery against otherwise treatment refractory psychiatric disease under appropriate ethical and technical control [1,3,6,7,81]. "
‘‘We looked at the data and saw that they did not support our prejudices. I, for one, did not expect to come out in favor of psychosurgery, but we saw that some very sick people had been helped by it and that it did not destroy their intelligence or rob them of their feelings. Their marriages were intact. They were able to work. The operation should not be banned.’’ J. Kenneth Ryan (Harvard Physician and Chairman of the 1974–1976 US Congress Commission evaluating the use of psychosurgery) [3]
2881
2882
171
Psychosurgery – a historical perspective
Modern Neuromodulation Techniques Form the Basis for Current and Future Neurosurgical Treatments of Severe Psychiatric Disorders The psychosurgical procedures described so far have all been based on ablative techniques lesioning nerve fibers connecting the frontal lobe with limbic structures located more caudally in the cerebrum. New ways of neuromodulation e.g., techniques performed in order to change the activity or function of specific brain areas and circuits have, however, successfully emerged with great speed during the last 20 years [14,15]. These developments have, furthermore, been aided by a similar optimization of the stereotaxic techniques based on the evolution of high resolution MR and CT brain imaging combined with the use of sophisticated computer programs for stereotaxic planning and targeting (> Figure 171-20). Modern techniques of neuromodulation, accordingly, encompass the use of stem-cells [92–97], gene-therapy [95,98–101], and deep brain stem
stimulation (DBS) [83–87,89–91] within specific brain areas using a modern stereotaxic approach (> Figure 171-20) [14,15]. High frequency deep brain stimulation performed by stereotaxic implanted electrodes placed in the subthalamic nucleus or the internal part of the globus pallidus have been used with great success in more than 15,000 Parkinson patients worldwide elegantly demonstrating that disturbed brain function may be effectively and safely corrected resulting in prominent lessening of parkinsonian core symptoms [14,15,83–87]. Most DBS-electrodes have several leads along their axis enabling stimulation with numerous combinations of different stimulation parameters over a wide and/or selected area without changing the placement of the electrode. The DBS-electrode is connected to a stimulator placed subcutaneously in the thorax region (> Figure 171-20), which enables the therapist to choose the stimulation lead and to modify the frequency, voltage and pulse width of the electrical current during and after implantation. Deep brain stimulation hereby represents a more
. Figure 171-20 High frequency deep brain stimulation performed by electrodes (left) inserted into the subthalamic nucleus by a stereotaxic approach based on high resolution brain imaging and subsequent computerized targeting calculation (right) has proven an effective and safe method of neuromodulation in Parkinson disease [14,15,83–87]. The same technique is currently being adapted to the treatment of several psychiatric diseases such as OCD and depression in Europe and North America [88–91]
Psychosurgery – a historical perspective
flexible (reversible) and adjustable method for the modulation of diseased brain areas than ablative surgery. The use of deep brain stimulation techniques instead of irreversible surgical ablations and tract lesions have, more over, been expanded to the treatment of psychiatric disorders like severe obsessive compulsive disorder with electrodes placed in the anterior limb of the internal capsule [89,91] or ventral caudate [88] and to treatment resistant depression with the electrodes placed in the subgenual cortical area [90]. Although the number of these procedures are still small, and more follow-up on the operated patients are required, these techniques hold promise for the future use of neuromodulatory psychosurgery as they in contrast to psychopharmacological treatment represent several possibilities for local modification of a diseased nervous system by replacement of lost neurons or neurotransmitters, and improvement of dysfunctional brain circuits [15]. These techniques might therefore represent a valuable treatment modality for neuropsychiatric patients with no or minor benefit from medical treatment [15]. The past history of psychosurgery combined with up to date knowledge on the limitations of the employed neurosurgical technique must be kept in mind to ensure a brighter future for functional neurosurgery directed towards neuropsychiatric disorders: All the neurosurgical neuromodulatory procedures accounted for above carry the risk of causing a potentially life-threatening hemorrhage or introducing infectious agents into the brain tissue. These complications are fortunately rare in most published materials [83–87,89–91], but underscore that the use of neurosurgery always must be based on a careful patient selection and pathophysiological models depicting reliable points of intervention. Furthermore, the newly introduced psychosurgical procedures should at first only be performed in neurosurgical and psychiatric centers of excellence, enabling careful evaluation of inclusion criteria, surgical procedures, and short- and long-term postsurgical
171
outcome. Procedures using DBS equipment should, likewise, be submitted to blinded cross over evaluations as the technique offer the possibility to turn the electrodes on or off. We must also recognize that psychiatric patients may have impaired capacity to make informed decisions. The performance of psychosurgery on psychiatric patients will, accordingly, often be complicated by ethical considerations as well as the public opinion, which due to the enormous impact of neuroscience on society during the last 20 years seem to be more acceptable towards a neurobiological understanding of psychiatric disease, although psychosurgery as a hole still is condemned in many places. A complete change in favor of psychosurgery demand, however, that we learn from the past history of psychosurgery and ensure that such procedures are based on a continuous integration of basic neuroscience data and technical developments in functional neurosurgery.
References 1. Cullington BJ. Psychosurgery: national commission issues surprisingly favorable report. Science 1976;194:299-301. 2. Feldman RP, Goodrich JT. Psychosurgery: a historical overview. Neurosurgery 2001;48:647-59. 3. Heller AC, Amar AP, Liu CY, Apuzzo MLJ. Surgery of the mind and mood: a mosaic of issues in time and evolution. Neurosurgery 2006;59:720-39. 4. Kopell BH, Machado AG, Rezai AR. Not your father’s lobotomy: psychiatric surgery revisited. Clin Neurosurg 2005;52:315-30. 5. Tierney AJ. Egas Moniz and the origins of psychosurgery: a review commemorating the 50th anniversary of Moniz’s Nobel Prize. J Hist Neurosci 2000;9(1):22-36. 6. U.S. Department of Health, Education, and Welfare. Protection of human subjects. Use of psychosurgery in practice and research: report and recommendations of National Commission for the Protection of Human Subjects. Fed Regist 1977;42(99):26318-32. 7. Valenstein ES. The practice of psychosurgery: a survey of the literature, 1971–1976: report to the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, United States Government Printing Office, Washington DC, Publication (05) 77–0002, 1977. 8. Valenstein ES. Great and desperate cures. New York: Basic Books; 1986.
2883
2884
171
Psychosurgery – a historical perspective
9. Valenstein ES. History of psychosurgery. In: Greenblatt SH, editor. The history of neurosurgery. Park Ridge: AANS; 1997. ¨ gren K, Sandlund M. Psychosurgery in Sweden 10. O 1944–1964. J Hist Neurosci 2005;14:353-67. 11. One Flew Over the Cukoo’s Nest. The Internet Movie Database. Available at: http://www.imdb.com/title/ tt0073486/. 12. Tranøy J, Blomberg W. Lobotomy in Norwegian psychiatry. Hist Psychiatr, 2005;16(1):107-10. 13. Binder D, Iskandar B. Modern neurosurgery for psychiatric disorders. Neurosurgery 2000;47(1):9-23. 14. Bjarkam CR, Sørensen JC, Sunde NA˚, Geneser FA, Østergaard K. New strategies for the treatment of Parkinson’s disease hold considerable promise for the future management of neurodegenerative disorders. Biogerontology 2001;2:193-207. 15. Bjarkam CR, Sørensen JC. Therapeutic strategies for neurodegenerative disorders: emerging clues from Parkinson’s disease. Biol Psychiatry 2004;56:213-16. 16. Piek J, Lidke G, Terberger T, von Smekal U, Gaab MR. Stone age skull surgery in Mecklenburg-Vorpommern: a systematic study. Neurosurgery 1999;45(1):147-51. 17. Stone JL, Miles ML. Skull trepanation among the early Indians of Canada and the United States. Neurosurgery 1990;26(6):1015-19. 18. Gross CG. Psychosurgery in renaissance art. Trends Neurosci 1999;22:429-31. 19. Salcman M. The cure of folly or the operation for the stone by Hieronymus Bosch (c. 1450–1516). Neurosurgery 2006;59:935-7. 20. Harlov JM. Recovery from the passage of an iron bar through the head. Proc Mass Med Soc 2:327–346 (original report published in a letter to Boston Med Surg J later renamed N Engl J Med 1868;39:389–392, 1848;19). 21. Ratiu P, Talos I-F. The tale of Phineas Gage, digitally remastered. N Eng J Med 2004;351:23. 22. Demasio H, Grabowski T, Frank R, Galaburda A, Damasio A. The return of Phineas Gage: clues about the brain from the skull of a famous patient. Science 1994;264:1102-5. 23. Haas LF. Phineas Gage and the science of brain localization. J Neurol Neurosurg Psychiatry 2001;71(6):716. 24. Broca P. Remarques sur le siege de la faculte´ du langage articule´; suivies d’une observation d’aphe´mie (perte de parole). Bulletins de la Socie´te´ Anatomique (Paris), 2e se´rie 6. 1861;6:330-357; Broca P. Nouvelle observation d’aphe´mie du langage articule´, suivies d’une observation d’aphe´mie (perte de la parole). Bulletins de la Socie´te´ Anatomique (Paris), 2e se´rie 6. 1861;6:398–407. 25. Finger S. Paul Broca (1824–1880). J Neurol 2004;251:769-70. 26. Wernicke C. Der aphasische symptomen complex. Breslau: Max Cohn and Weigert; 1874.
¨ ber rindexcisionen, als beitrag zur oper27. Burckhardt G. U ativen therapie der psychosen, Allg Z Psychiatr Psych Med 1891;47:463-548. 28. Fulton JF, Jacobsen CF. The functions of the frontal lobes: a comparative study in monkeys, chimpanzees, and man. In Abstracts of the Second International Neurological Congress, London, 1935. pp. 70-71. 29. Moniz E. Tentatives operatoires dans le traitement de certaines psychoses. Paris: Masson; 1936. 30. Moniz E. Prefrontal leucotomy in the treatment of mental disorders. Am J Psychiatry 1937;93:1379-85. 31. Jansson B. Controversial psychosurgery resulted in a Nobel Prize. Available at: http://nobelprize.org/nobelprizes/medicine/articles/moniz/index.html, 1998. 32. Kelly D, Richardson A, Mitchell-Heggs N, Greenup J, Chen C, Hafner RJ. Stereotactic limbic leucotomy: a preliminary report on forty patients. Br J Psychiatry 1973;123(573):141-8. 33. Freeman W, Watts JW. Prefrontal lobotomy in the treatment of mental disorders. South Med J 1937;30: 23-31. 34. Freeman W, Watts JW. Psychosurgery: intelligence, emotion, and social behavior following prefrontal lobotomy for mental disorders. Springfield, IL: Charles C. Thomas; 1942. 35. Freeman W, Watts JW. Psychosurgery: in the treatment of mental disorders and intractable pain, 2nd ed. Springfield, IL: Charles C. Thomas; 1950. 36. Swayze II, VW. Frontal leucotomy and related psychosurgical procedures in the era before antipsychotics (1935–1954): a historical overview. Am J Psychiatry 1995;152(4):505-15. 37. Streckler EA, Palmer HD, Grant FC. A study of frontal lobotomy: neurosurgical features and results in 22 cases with a detailed report on 5 chronic schizophrenics. Am J Psychiatry 1942;98:524-32. 38. Fiamberti A. Proposta di una tecnica operatoria modificata e semplificata per gli interventi alla Moniz sui lobi frontali in malati di mente. Raas Studi Psichiat 1937;26:797. 39. Pressman JD. Last resort, psychosurgery and the limits of medicine. Cambridge: Cambridge University Press; 1998. 40. Nobel Foundation. The Nobel Prize in Physiology or Medicine 1949. http://nobelprize.org/nobel_prizes/ medicine/laureates/1949/index.html, 2007 41. Lo´pez-Mun˜oz F, Alamo C, Cuenca E, Shen WW, Clervoy P, Rubio G. History of the discovery and clinical introduction of chlorpromazine. Ann Clin Psychiatry 2005;17(3):113-35. 42. Ødega˚rd Ø. Nye framsteg i psykiatrien. Tidsskrift for den Norske Lægeforening 1953;123:411-14. 43. Cade JFJ. Lithium salts in treatment of psychotic excitement. Med J Aust 1949;36:349-52. 44. Schou M, Juel-Nielsen N, Stro¨mgren E, Voldby H. The treatment of manic psychoses by the administration
Psychosurgery – a historical perspective
45.
46.
47.
48. 49.
50.
51.
52. 53.
54.
55.
56. 57.
58.
59.
60.
61.
of lithium salts. J Neurol Neurosurg Psychiatry 1954;17:250-60. Bini L. Experimental researches on epileptic attacks induced by the electric current. The treatment of schizophrenia: insulin shock, cardiozol, sleep treatment. Am J Psychiatry 1938;94:172-4. Scoville WB. Selective cortical undercutting as a means of modifying and studying frontal lobe function in man. J Neurosurg 1949;6(1):65-73. Whitty CM, Duffield JE, Tow PM, Cairns H. Anterior cingulectomy in the treatment of mental disease. Lancet 1952;1:475-81. Gildenberg PL. History repeats itself. Stereotact Funct Neurosurg 2003;80:61-75. Spiegel EA, Wycis HT, Marks M, Lee AS. Stereotactic apparatus for operations on the human brain. Science 1947;106:349. Spiegel EA, Wycis HT, Orchinik C. Thalamotomy and hypothalamotomy for the treatment of psychoses. Res Publ Assoc Res Nerv Ment Dis 1953;31:379-91. Horsley V, Clarke RH. The structure and functions of the cerebellum examined by a new method. Brain 1908;31:45-124. Gildenberg PL. Spiegel and Wycis – the early years. Stereotact Funct Neurosurg 2001;77:11-16. Spiegel EA, Wycis HT, Freed H. Thalamotomy in mental disorders. In Procedings of the First International Congress of Psychosurgery. Lisbon, Livraria Luso-Espanhola, 1949. p. 91-9. Narabayashi H, Nagao T, Saito Y, Yoshida M, Nagahata M. Stereotactic amygdalotomy for behavior disorders. Arch Neurol 1963;9:1-16. Diermann NG, Hassler R. Treatment of sexual violence by stereotactic hypothalamotomy. In Sweet WH, Obrador S, Martin-Rodriguez JG, editors. Neurosurgical treatment in psychiatry, pain, and epilepsy. University Park Press; Baltimore: p. 451-62. Laitinen LV. Psychosurgery. Stereotact Funct Neurosurg 2001;76:239-42. Sano K. Sedative neurosurgery with special reference to postero-medial hypothalatomy. Neurol Med Chir (Tokyo) 1962;4:112-42. Ballantine HT, Jr, Cassidy WL, Flanagan NB, Marino R, Jr. Stereotaxic anterior cingulotomy for neuropsychiatric illness and intractable pain. J Neurosurg 1967;26: 488-95. Ballantine HT, Jr, Bouckoms AJ, Thomas EK, Giriunas IE. Treatment of psychiatric illness by stereotactic cingulotomy. Biol Psychiatry 1987;22:807-19. Dougherty DD, Baer L, Cosgrove CR, Cassem EH, Price BH, Nierenberg AA, Jenike MA, Rauch SL. Prospective long-term follow-up of 44 patients who received cingulotomy for treatment-refractory obsessive-compulsive disorder. Am J Psychiatry 2002;159:269-75. Greenberg BD, Price LH, Rauch SL, Friehs G, Noren G, Malone D, Carpenter LL, Rezai AR, Rasmussen SA.
62.
63.
64.
65.
66.
67.
68. 69. 70.
71. 72.
73.
74.
75. 76.
77.
171
Neurosurgery for intractable obsessive-compulsive disorder and depression: critical issues. Neurosurg Clin N Am 2003;14:199-212. Bingley T, Leksell L, Meyerson BA. Long-term results of stereotactic capsulotomy in chronic obsessive compulsive neurosis. In: Sweet WH, Obrador S, Martin-Rodriguez JG, editors. Neurosurgical treatment in psychiatry, pain, and epilepsy. Baltimore: University Park Press; 1977. p. 287-9. Herner T. Treatment of mental disorders with frontal stereotaxic thermo-lesions. Acta Psychiat Neurol Scand 1961;36 Suppl 158:1-140. Talairach J, He´caen H, David M. Lobotomie pre´frontale limitee´ par e´lectrocoagulation des fibres thalamo-frontales a` leur emergence du bras antere´rieur de la capsule interne. Rev Neurol 1949;83:59. Knight G. Stereotactic tractotomy in the surgical treatment of mental illness. J Neurol Neurosurg Psychiatry 1965;28:304-10. Knight GC. Bifrontal stereotactic tractotomy: an atraumatic operation of value in the treatment of psychoneurosis. Br J Psychiatry 1969;115:257-66. Kelly D, Richardson A, Mitchell-Heggs N. Stereotactic limbic leucotomy: neurophysiological aspects and operative technique. Br J Psychiatry 1973;123(573):133-40. Papez JW. A proposed mechanism of emotion. Arch Neurol Psychiatry 1937;38:725-43. MacLean PD. Psychosomatic disease and the ‘‘visceral brain.’’ Psychosom Med 1949;11:338-53. MacLean PD. Some psychiatric implications of physiological studies on front temporal portion of limbic system (visceral brain). EEG Clin Neurophysiol 1952;4: 397-406. MacLean PD. The limbic system and its hippocampal formation. J Neurosurg 1954;11:29-44. Jung HH, Kim C-H, Chang JH, Park YG, Chung SS, Chang JW. Bilateral anterior cingulotomy for refractory obsessive-compulsive disorder: long-term follow-up results. Stereotact Funct Neurosurg 2006;84:184-9. Spangler WJ, Cosgrove GR, Ballantine HT, Cassem EH, Rauch SL, Nierenberg A, Price BH. Magnetic resonance image-guided stereotactic cingulotomy for intractable psychiatric disease. Neurosurgery 1996;38:1071-6. Oliver B, Gasco´n J, Aparicio A, Ayats E, Rodriguez R, de Leon JLM, Garcia-Bach M, Soler PA. Bilateral anterior capsulotomy for refractory obsessive-compulsive disorders. Stereotact Funct Neurosurg 2003;81:90-5. Knight G. The orbital cortex as an objective in the surgical treatment of mental illness. Br J Surg 1964;51:119-24. Malhi GS, Bartlett JR. A new lesion for the psychosurgical operation of stereotactic subcaudate tractotomy (SST). Br J Neurosurg 1998;12(4):335-9. Bridges PK, Bartlett JR, Hale AM, Poynton AM, Malizia AL, Hodgkiss AD. Psychosurgery: stereotactic subcaudate tractotomy an indispensible treatment. Br J Psychiatry 1994;165:599-611.
2885
2886
171
Psychosurgery – a historical perspective
78. Goktepe EO, Young LB, Bridges PK. A further review of the results of stereotactic subcaudate tractotomy. Br J Psychiatry 1975;126:270-80. 79. Strom-Olsen R, Carlisle S. Bifrontal stereotactic tractotomy: a follow-up of its effects on 210 patients. Br J Psychiatry 1971;118:141-54. 80. Montoya A, Weiss AP, Price BH, Cassem EH, Dougherty DD, Nierenberg AA, Rauch SL, Cosgrove GR. Magnetic resonance imaging-guided stereotactic limbic leucotomy for treatment of intractable psychiatric disease. Neurosurgery 2002;50:1043-52. 81. Mashour GA, Walker EE, Martuza RL. Psychosurgery: past, present, and future. Brain Res Rev 2005;48:409-19. 82. Mitchell-Heggs N, Kelly D, Richardson A. stereotactic limbic leucotomy: a follow-up at 16 months. Br J Psychiatry 1976;128:226-40. 83. Hammerstad J, Hogarth P. Parkinson’s disease: surgical options. Curr Neurol Neurosci Rep 2001;1:313-19. 84. Krack P, Poepping M, Weinert D, Schrader B, Deuschl G. Thalamic, pallidal or subthalamic surgery for Parkinson’s disease. J Neurol 2000;247 Suppl 2:127-34. 85. Lang AE. Surgery for Parkinson’s disease: a critical evaluation of the state of the art. Arch Neurol 2000;57:1118-25. 86. Limousin P, Krack P, Pollak P, Benazzouz A, Ardouin C, Hoffman D, Benabid AL. Electrical stimulation of the subthalamic nucleus in advanced Parkinson’s disease. N Engl J Med 1998;339:1105-11. 87. Tronnier VM, Fogel W, Krause M, Bonsanto MM, Tronnier J, Heck A, Mu¨nkel K, Kunze S. High frequency stimulation of the basal ganglia for the treatment of movement disorders: current status and clinical results. Minim Invasive Neurosurg 2002;45(2):91-6. 88. Aouizerate B, Martin-Guehl C, Cuny E, Guehl D, Amieva H, Benazzouz A, Fabriqoule C, Bloulac B, Tignol J, Burbaud P. Deep brain stimulation for OCD and major depression. Am J Psychiatry 2005;162:2192. 89. Gabrie¨ls L, Cosyns P, Nuttin B, Demeulemeester H, Gybels J. Deep brain stimulation for treatment-refractory obsessive-compulsive disorder: psychopathological and neuropsychological outcome in three cases. Acta Psychiatr Scand 2003;107:275-82. 90. Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, Schwalb JM, Kennedy SH. Deep brain stimulation for treatment-resistant depression. Neuron 2005;45:651-60. 91. Nuttin B, Cosyns P, Demeulemeester H, Gybels J, Meyerson B. Electrical stimulation in anterior limbs of internal capsules in patients with obsessive-compulsive disorder. Lancet 1999;354:1526.
92. Bjo¨rklund A, Dunnett SB, Brundin P, Stoessl AJ, Freed CR, Breeze RE, Levivier M, Peschanski M, Studer L, Barker R. Neural transplantation for the treatment of Parkinson’s disease. Lancet Neurol 2003;2:437-45. 93. Kordower JH, Freeman TB, Snow BJ, Vingerhoets FJ, Mufson EJ, Sanberg PR, Hauser RA, Smith DA, Nauert GM, Perl DP, Olanow CW. Neuropathological evidence of graft survival and striatal reinnervation after the transplantation of fetal mesencephalic tissue in a patient with Parkinson’s disease. N Engl J Med 1995;332 (17):1118-24. 94. Lindvall O. Cerebral implantation in movement disorders: state of the art. Mov Disord 1999;14(2):201-5. 95. Martinez-Serrano A, Bjo¨rklund A. Immortalized neural progenitor cells for CNS gene transfer and repair. Trends Neurosci 1997;20:530-8. 96. Piccini P, Brooks DJ, Bjo¨rklund A, Gunn RN, Grasby PM, Rimoldi O, Brundin P, Hagell P, Rehncrona S, Widner H, Lindvall O. Dopamine release from nigral transplants visualized in vivo in a Parkinson’s patient. Nat Neurosci 1999;2(12):1137-40. 97. Piccini P, Lindvall O, Bjorklund A, Brundin P, Hagell P, Ceravolo R, Oertel W, Quinn N, Samuel M, Rehncrona S, Widner H, Brooks DJ. Delayed recovery of movement-related cortical function in Parkinson’s disease after striatal dopaminergic grafts. Ann Neurol 2000;48 (5):689-95. 98. During MJ, Kaplitt MG, Stern MB, Eidelberg D. Subthalamic GAD gene transfer in Parkinson disease patients who are candidates for deep brain stimulation. Hum Gene Ther 2001 12:1589–91. 99. Gill SS, Patel NK, Hotton GR, O’Sullivan K, McCarter R, Bunnage M, Brooks DJ, Svendsen CN, Heywood P. Direct brain infusion of glial cell line-derived neurotrophic factor in Parkinson disease. Nat Med 2003;9:589-95. 100. Kordower JH, Emborg ME, Bloch J, Ma SY, Chu Y, Leventhal L, McBride J, Chen EY, Palfi S, Roitberg BZ, Brown WD, Holden JE, Pyzalski R, Taylor MD, Carvey P, Ling Z, Trono D, Hantraye P, De´glon N, Aebischer P. Neurodegeneration prevented by lentiviral vector delivery of GDNF in primate models of Parkinson’s disease. Science 2000;290:767-72. 101. Luo J, Kaplitt MG, Fitzsimons HL, Zuzga DS, Liu Y, Oshinsky ML, During MJ. Subthalamic GAD gene therapy in a Parkinson’s disease rat model. Science 2002;298:425-9.
177 Surgical Procedures for Tourette’s Syndrome V. Visser-Vandewalle
Introduction In 1885, Georges Gilles de la Tourette described a specific syndrome as a nervous affection characterized by lack of motor coordination accompanied by echolalia and coprolalia. Later Charcot named the condition Tourette’s Syndrome (TS) [1]. TS is a complex neuropsychiatric disorder characterized by tics. Tics are sudden, repetitive, stereotyped muscle contraction (motor tics) sounds [2]. Tics can be seen as fragments of normal motor action or vocal productions that are misplaced in context and at times confused with goal-directed behavior [3]. Tics may be abrupt in onset, fast and brief (clonic tics) or may be slow and sustained (dystonic or tonic tics) [4]. The motor patterns of tics may involve individual muscles or small groups of muscles with discrete contractions (simple tics) like eye blinking, facial grimacing, sniffing or throat clearing. When more muscles are acting in a coordinated pattern to produce more complicated movements that may resemble purposeful voluntary movements, we call them complex tics [2]. Complex tics include head shaking, scratching, throwing, touching or uttering phrases. Coprolalia, or uttering obscene words, one of the most distressing and recognisable symptoms, occurs in only 10% of patients. Tics are often more frequent and forceful when the patient is alone [5]. They can be temporarily suppressed by an effort of will or concentration, but may rebound afterwards [6]. Tics usually begin in the first decade of life, with a mean age of 7 years [3]. #
Springer-Verlag Berlin/Heidelberg 2009
The first tics are usually motor tics, with brief bouts of transient tics involving the face or head. The severity of tics typically increases during the prepubescent years, and the majority of patients improve spontaneously as they reach adulthood. An important feature of TS is its association with a wide range of co-morbid behavioral abnormalities. More than half of all children and adolescents with Tourette Syndrome show evidence of psychiatric comorbidity, exhibiting symptoms of attention-deficit hyperactivity disorder (ADHD), obsessive-compulsive behavior (OCB), and other anxiety and mood disorders [3]. When present these coexisting conditions can add greatly to morbidity associated with Tourette Syndrome and have a negative impact on the overall quality of life [7]. The occurrence of ADHD in TS patients ranges from 21 to 90% of clinical populations [3]. Symptoms consist of inattention and distractibility with or without behavioral hyperactivity. OCB may occur in up to 50% of TS patients. The more severe obsessions in TS may involve sexual, violent, religious, aggressive and symmetrical themes; the compulsions may manifest with symptoms such as checking, counting, forced touching, and self-damage. Like tics, OCB-symptoms often wax and wane during the course of the illness. Robertson [4] reported that over onethird of clinical TS patients suffered from self injuring behavior (SIB). The most frequent type of SIB was head banging. While once thought to be rare, TS is now recognized as a relatively common disorder with an estimated worldwide prevalence of 4–5/10,000.
2964
177
Surgical procedures for tourette’s syndrome
It occurs three to four times more commonly in males [8]. There is a considerable variation among studies reporting on the prevalence of TS which is most likely due to variations in sex, age, diagnostic criteria, and assessment methods [9].
Treatment For many patients, especially those with mild symptomatology, psychobehavioral strategies are sufficient. Pharmacological treatment, however, may be considered when symptoms begin to interfere with social interactions, academic or job performance, or with activities of daily living. At present, pharmacotherapy, mainly involving neuroleptic medications, A2-adrenergic agonists, and dopamine agonists as well as local injections of botulinum toxin is the mainstay of treatment for tics [3]. Selective serotonin reuptake inhibitors are recommended for the treatment of obsessivecompulsive behavior. Psychostimulants, such as methylphenidate, are the treatment of choice for attention deficit hyperactivity disorder [3]. For patients refractory to any medical treatment, surgery may be the treatment of last resort. Although no precise numbers are available, this seems to represent a very small percentage of patients with TS.
History of Neurosurgical Treatment of TS Despite increasing evidence of specific basal ganglia involvement in TS, there has not been consensus on the best surgical approach. Since 1960, there have been more than 35 reports of surgical therapy for Tourette Syndrome [10]. Surgical procedures have included prefrontal leucotomy, limbic leucotomy, cingulotomy, lesions of the medial and intralaminar thalamic nuclei, lesions of the zona incerta and red nucleus, and dentatotomies. In most of these studies, patient selection was not
standardized, assessments typically were not blinded, and outcome was not quantified. The results were often unsatisfactory or major sideeffects occurred such as hemiplegia or dystonia. Hassler and Dieckmann [11] reported the effects of ablation of the intralaminar and medial thalamic nuclei in nine patients with Tourette Syndrome. Three of the patients treated bilaterally were reported to have tic reductions of 90–100%. Deep brain stimulation (DBS) was first introduced as a new surgical technique for the treatment of intractable TS in 1999 [12]. Vandewalle et al. selected a trajectory that included the centromedian (CM) and ventral oral internal thalamic nuclei (Voi) and the substantia periventricularis (Spv). This single trajectory was thought to most accurately mimic the lesions performed by Hassler and Dieckmann, whose procedures often required 10 ablations per side. The same group described the promising effects of bilateral thalamic DBS in three patients in greater detail in 2003 [13]. With a follow-up period of 5 years, 1 year and 8 months respectively, there was an improvement in both tic reduction (with 72–90%) and in associated behavioral disorders. Stimulation induced side-effects consisted of sensation of reduced energy and changes in sexual functioning [13,14].
Neuroanatomical Basis for DBS in TS The pathophysiology of TS is still a matter of considerable debate. A growing body of evidence indicates that an abnormality in corticostriatothalamocortical circuits and their neurotransmitter systems is likely to underlie tics and coexisting problems in Tourette Syndrome. Evidence supporting a dopaminergic abnormality in Tourette Syndrome comes from therapeutic responses to neuroleptics, preliminary from postmortem studies, and a variety of nuclear imaging protocols [15].
Surgical procedures for tourette’s syndrome
Within the brain, there are anatomically segregated, parallel circuits representing different functions (motor, oculomotor, cognitive and limbic). These basal ganglia circuits traverse the cortex, striatum, globus pallidus, and thalamus. Each circuit includes a direct and an indirect pathway. It is hypothesized that disinhibition of excitatory neurons in the thalamus results in hyperexcitability of cortical motor areas in the release of tics. If Tourette Syndrome is associated with excessive dopaminergic activity, whether via supersensitive dopamine receptors, dopamine hyperinnervation or abnormal presynaptic terminal function, it inhibits the indirect pathway, leading to an overactivity of thalamocortical drive [16]. In addition to short term effects, dopamine can modulate cortico-striatal transmission by the mechanism of long term depression or potentation. Dopamine inducing fluctuating abnormalities in the resting potential of striatal neurons have been hypothesized to influence waxing and waning of tics. This could be an explanation of the lack of identifiable abnormalities in dopamine transmission. Alterations in striatal function have also been demonstrated in Tourette patients on fMRI during active tic suppression, with a decreased activity in putamen, ventral pallidum and thalamus bilaterally, and an increased activity of the head of the right caudate nucleus, and frontal and temporal cortices [17]. Other cortical-subcortical loops may also be implicated in TS pathophysiology. The excitatory feedback loops from the thalamus towards the striatum, originating from the centromedianparafascicular complex (CM-Pf), towards the motor part of the striatum, and the midline thalamic nuclei Spv, towards the limbic part of the striatum are considered to be circuits that may be affected in TS and explain the action of DBS in this location. Several studies have suggested that both the sensorimotor and the limbic-innervated parts of the basal ganglia, including the dorsal and ventral striatum, are involved in the pathophysiology of TS [18–21].
177
This may also explain both motor and nonmotor symptoms.
Targets After the initiation of thalamic DBS as a potential treatment for patients with refractory TS, several other targets have been used. Up until now, five targets have been used for DBS for TS in 32 patients: the medial portion of thalamus, at the cross point of CM-Spv-Voi [12,13,22–26]; medial portion of thalamus, CM-Pf [27]; the globus pallidus internus (GPi), posteroventrolateral part [28–30]; GPi, anteromedial part [27]; the nucleus accumbens (NAC) and anterior limb of internal capsule (IC) [31–32]. Servello et al. [23] reported on the beneficial effects of bilateral stimulation of the CM-Pf and (Voi) complex of the thalamus, in 18 patients with TS (age varying between 17 and 47 years), with the duration of the follow-up assessments ranging from 3 to 18 months. In this report there was an apparently, but not statistically significant, better response of motor tics when compared to phonic tics. These authors also reported positive effects on behavioral disorders, with no serious permanent adverse effects. Bajwa et al. [25] described the beneficial effects of the same thalamic target in a 48-year-old patient, with relentless, violent head jerks resulting in progressive neurological impairment. DBS resulted in 66% tic reduction on the Yale Global Tic Severity Scale (YGTSS) at 24 months follow up. The patient’s self assessment indicated 95% improvement. Also a positive effect on mood and obsessive-compulsive symptoms was reported. Recently, a randomized double blind trial of thalamic DBS in five adult patients with Tourette Syndrome is reported by Maciunas et al. [26]. After 4 months of deep brain stimulation they describe a good effect on motor tics (40%) and an increase of vocal tics (21%) based on the modified Rush Video Based Rating Scale [33]. To date, 27 patients have been reported who have undergone bilateral
2965
2966
177
Surgical procedures for tourette’s syndrome
thalamic stimulation with promising results on tics and associated behavioral disorders. The effects of bilateral DBS of the posteroventral (motor) part of the internal segment of the globus pallidus (GPi) in a single patient were firstly described by Van der Linden et al. in 2002 [28]. This target was chosen on the basis of the positive effects of GPi DBS in hyperkinesias such as in advanced PD or dystonia. At six months follow-up, a tic reduction of 95% was noted. In 2004, Diederich et al. [29] described the beneficial effects of chronic stimulation of the same target in a 27-year-old Tourette patient, with a follow-up of 14 months. Tic frequency per minute decreased by 73% in the postoperative phase and in particular the vocal tics became less intense. The patient noticed a decrease of the internal urge to produce tics. However, there was no change in the patient’s ‘‘very mild compulsive tendencies.’’ The effects of GPi DBS in a 16-year-old boy were described by Shahed et al. [30] The authors reported a significant effect on tics (84%) and behavior at six months follow-up. However, a body shield was needed for 4 weeks because the patient compulsively pushed on the IPG’s. Houeto et al. [27] described the effects of bilateral pallidal and thalamic stimulation in a 38-year-old female patient. The thalamic target was located in the centromedian-parafascicular complex, and the pallidal target in the anteromedial (limbic) part of the GPi. This latter was based on the assumption that TS is more a limbic than a motor disorder. Both thalamic and pallidal stimulation had 65% reduction in tics, but thalamic stimulation was superior for treatment of the associated behavior disturbance, after 24 months. Recently Flaherty et al. [31] described the effects of bilateral stimulation of the anterior portion of the internal capsule in a single 37-year-old Tourette patient who suffered from severe tics without associated behavioral disorder. After 18 months, there was a 25% reduction in tics. In this patient, the lowest electrode contacts produced mild depression while the highest contacts
caused hypomania. The effects of DBS of the nucleus accumbens, with two poles of the electrode at the level of the anterior capsule, were described by Kuhn et al. [32] in a 26-year-old male patient suffering from severe TS, OCB and SIB. The best effect on tics, with a 40–50% tic reduction on the YGTSS after 2,5 years, were obtained by monopolar stimulation of all poles of the quadripolar electrode. Also a clear amelioration of obsessive and compulsive symptoms was noticed. A small hematoma around the tip of the electrode, have been reported to date in the literature in two cases [29,34]. Unexpected stimulation induced side-effects such as drowsiness, reduced energy, changes in sexual behavior and mild dysarthria, seem to be emerging in the majority of reported cases [10,12,14,23]. One patient with bilateral thalamic and bilateral anteromedial GPi DBS appeared to be more depressed with pallidal stimulation [27] (> Table 177-1).
Clinical and Surgical Evaluation Patient Selection As mentioned in the first section, in most cases TS symptoms wane before or at onset of adolescence. Not all patients require therapy and, of those who do, only a minority fail to respond to any medical treatment. Suitable candidates for DBS will be adults who have received careful trials of standard therapies without adequate benefit. Candidates for DBS in TS should be evaluated by a multidisciplinary team including at least a neurologist, a psychiatrist, and a doctoral level psychologist or neuropsychologist with expertise in TS and comorbid conditions. The Dutch-Flemish Tourette Surgery Study Group has established guidelines for DBS in TS [35], and recently the Movement Disorders Society has published a position statement [36]. These statements include the following selection criteria.
Thalamus (med.) Thalamus (med.)/Gpi vpl Thalamus (med.)
Vandewalle [12]
Thalamus (Cm-Pf)/ Gpi am Thalamus (med.) Internal capsule Thalamus (med.) NAC GPi
Houeto [27]
Thalamus (med.)
Servello [23]
18
5
1 1
1
1
3–17m
4m
30m 6m
24m
18m
6m
24m
1
1
14m
5y, 1y, 8m
Immed. postop / 6m
4m
F-U
1
3
1
1
No of pt
40% motor, increased vocal tics 21% Good
40–50% 84%
66%
25%
78%
70% (both)
66%
90%, 72%, 83%
80%/95%
90–100%
Good
Very good No effect on automutilation Good
Good
–
Not mentioned
Very good(both)
No effect on compulsions
Very good
Not mentioned
Not mentioned
Effect on behavioral disorders
Wound dehiscence, abdominal subcutaneous hematoma.
None Acute psychosis, spontaneous recurrence of tics, Reduced energy
Not mentioned Not mentioned
None
None
None
Small haematoma around right electrode tip None
None
None
None
Complications
None None
Hypomania and depression None
Mild dysarthria
With Gpi DBS more depressed
Drowsiness, changes in sexual behaviour (2 pt) Impairment of left rapidly alternating movements
None
Not mentioned
Side-effects
Med. = medial part (Cm-Spv-Voi); Cm = centromedian nucleus; Spv = substantia periventricularis; Voi = nucleus ventro-oralis internus; Pf = Parafascicular nucleus; Gpi = globus pallidus internus; vpl = ventroposterolateral part; am = anteromedial part; NAC = nucleus accumbens; F-U = follow-up period; y = year(s); m = month(s); immed.postop = immediately postoperatively
Thalamus (med.)
Maciunas [26]
Kuhn [32] Shahed [30]
Bawja [25]
Flaherty [31]
Egidi [24]
Gpi vpl
Diederich [29]
Visser-Vandewalle [35]
van der Linden [28]
Target
Reference
Tic reduction
. Table 177-1 Reports on deep brain stimulation in patients with Tourette syndrome
Surgical procedures for tourette’s syndrome
177 2967
2968
177
Surgical procedures for tourette’s syndrome
Inclusion of Patients
Surgical Procedure
1.
The technique of DBS applied to TS is similar to that used for more classical indications like Parkinson disease. Targets for TS such as the nuclei of the medial portion of the thalamus, are not visible with current imaging techniques. Moreover, TS patients may pull themselves out of the stereotactic frame because of the frequent motor tics which occur in the head region. One solution would be to operate with the patient under general anaesthesia [27,29]. Because of the uncertainty of the ideal target and the importance of intra-operative findings, it is preferable if the patient is able to be awake and cooperative during surgery. To avoid general anaesthesia, patients may be sedated with a combination of lormetazepam and clonidine [13] or with a Propofol Target Controlled Infusion [28], which reduces tics sufficiently to improve safety and efficacy of the stereotactic procedure. With the patient awake their symptoms can be assessed so that acute negative stimulation-induced sideeffects can be detected and the position of the electrode adjusted as needed.
2. 3.
4.
5.
The patient has definite Tourette’s syndrome, established by two independent clinicians, preferably a psychiatrist and neurologist. The diagnosis is established according to DSMIV-TR criteria [37] and with the aid of the Diagnostic Confidence Index (DCI) [38]. The patient has severe and incapacitating tics as his primary problem. The patient is treatment refractory. This means that the patient either has not or very partially responded to 3 different medication regimes, each for at least 12 weeks, and in adequate doses, or has been proven not to tolerate medications due to sideeffects. Three different groups of neuroleptics should have been tried: i. ‘‘classic’’ Dopamine-2 antagonists (haloperidol, pimozide or clonidine) ii. modern anti-psychotic medications (e.g., risperidone, olanzapine, clozapine, sulpiride, aripiprazole) iii. experimental drugs (e.g., pergolide) Finally, a trial of at least 10 sessions of behavioral therapy for tics, such as habit reversal or exposure in vivo, may be attempted. The patient should be over 25 years of age.
Exclusion of Patients Patients should be excluded from neurosurgical treatment if they have a tic disorder other than TS, severe psychiatric co-morbid conditions (other than associated behavioral disorders), or mental deficiency. Contra-indications for surgical treatment for DBS in TS are severe cardiovascular, pulmonary or haematological disorders and structural MRI-abnormalities as well as active suicidal ideation.
Perioperative Evaluation It is of paramount importance that in TS patients treated with DBS the exact location of the electrode and position of the stimulating contact is precisely determined and all effects are meticulously described. A more comprehensive survey of guidelines for the perioperative assessment of the effects of DBS in TS is available elsewhere [36].
Post-operative Evaluation For the assessment of clinical effects, a careful and detailed description of the effect of DBS on tics and
Surgical procedures for tourette’s syndrome
associated behavioral disorders and stimulationinduced side-effects are mandatory. The most commonly used scale for tic rating is the Yale Global Tic Severity Scale (YGTSS) [39]. The modified Rush Video-Based Rating Scale is also commonly used [33,40]. For a more objective evaluation, the patient should also be recorded on video with and without stimulation. The tics should be rated on video by two independent investigators. Ideally, the patient and investigator should be blinded to the status of the stimulation. A careful psychiatric and neuropsychological evaluation should be performed at regular intervals, for example 3, 6 and 12 months. It is essential that any investigation of the clinical effects of DBS for TS include accurate post-operative imaging to identify the exact electrode placement. In a region like the medial thalamus that encompasses many small vessels, the localization is especially important. The most prudent approach may be to perform a CT-scan postoperatively and fuse these images with preoperative MR-images, although many centers successful employ other imaging approaches. Only if these prerequisites are fulfilled and a maximum amount of data is exchanged between centers, the optimal target can be established.
Programming According to our experience with DBS in the medial portion of the thalamus, the best effect in the majority of patients is obtained with a frequency between 75 and 100 Hz and a pulse width of 210 ms. From day one postoperatively bipolar stimulation is started (to obtain the most selective effect), with each pole made active during four consecutive days (e.g., day 1: pole 0 -, pole 1þ; day 2: pole 1 -, pole 2þ, etc). During programming, the voltage is progressively increased until unwanted side-effects occur. Thereafter, the combination of electrodes may
177
be altered (for example 2 electrodes negative), or monopolar stimulation may be chosen, as suggested by clinical effects. As for other DBS indications, programming is a matter of ‘‘trial and error’’ as directed by the best clinical effects and fewest adverse effects.
References 1. Kra¨mer H, Daniels C. Pioneers of movement disorders: Georges Gilles de la Tourette. J Neural Transm 2004; 111:691-701. 2. Mink JW. Basal ganglia dysfunction in Tourette’s syndrome: a new hypothesis. Pediatr Neurol 2001;25:190-8. 3. Leckman JF, Bloch MH, Scahill L, King RA. Tourette Syndrome: the self under siege. J Child Neurol 2006;21:642-9. 4. Robertson MM. Tourette syndrome, associated conditions and the complexities of treatment. Brain 2000;123:425-62. 5. Lombroso PJ, Mack G, Scahill L, et al. Exacerbation of Tourette’s syndrome associated with thermal stress: a family study. Neurology 1991;41:1984-87. 6. Berardelli A, Curra A, Fabbrini G, Gilio F, Manfredi M. Pathophysiology of tics and Tourette syndrome. J. Neurol 2003;250:781-7. 7. Hoekstra PJ, Anderson GM, Limburg PC, Korf J, Kallenberg CG, Minderaa RB. Neurobiology and neuroimmunology of Tourette’s syndrome: an update. Cell Mol Life Sci 2004;61:886-98. 8. Riederer F, Stamenkovic M, Schindler SD, Kasper S. Tourette’s syndrome – a review. Nervenarzt 2002;73: 805-19. 9. Leckman JF. Tourette’s syndrome. Lancet 2002;360: 1577-86. 10. Temel Y, Visser-Vandewalle V. Surgery in Tourette syndrome. Mov Disord 2004;19:3-14. 11. Hassler R, Dieckmann G. Traitement stereotaxique des tics et cris inarticule´s ou coprolaliques conside´re´s comme phe´nome`ne d’obsession motrice au cours de la maladie de Gilles de la Tourette. Rev Neurol (Paris) 1970;123:89-100. 12. Vandewalle V, van der Linden C, Groenewegen HJ, Caemaert J. Stereotactic treatment of Gilles de la Tourette syndrome by high frequency stimulation of thalamus. Lancet 1999;353:724. 13. Visser-Vandewalle V, Temel Y, Boon P, et al. Chronic bilateral thalamic stimulation: a new therapeutic approach in intractable Tourette syndrome. J Neurosurg 2003;99:1094-100.
2969
2970
177
Surgical procedures for tourette’s syndrome
14. Temel Y, van Lankveld JJ, Boon P, Spincemaille GH, van der Linden C, Visser-Vandewalle V. Deep brain stimulation of the thalamus can influence penile erection. Int J Impot Res 2004;16:91-4. 15. Harris K, Singer HS. Tic disorders: neural circuits, neurochemistry, and neuroimmunology. J Child Neurol 2006;21:678-89. 16. Singer HS, Minzer K. Neurobiology of Tourette’s syndrome: concepts of neuroanatomic localization and neurochemical abnormalities. Brain Rev 2003;25:S70-84. 17. Peterson BS, Skudlarski P, Anderson AW, et al. A functional magnetic resonance imaging study of tic suppression in Tourette syndrome. Arch Gen Psychiatry 1998;55: 326-33. 18. Graybiel AM. The basal ganglia. Curr Biol 2000;10: R509-11. 19. Groenewegen HJ, van den Heuvel OA, Cath DC, Voorn P, Veltman DJ. Does an imbalance between the dorsal and ventral striatopallidal systems play a role in Tourette’s syndrome? A neuronal circuit approach. Brain Dev 2003;25:S3-14. 20. Peterson BS, Thomas P, Kane MJ, et al. Basal ganglia volumes in patients with Gilles de la Tourette syndrome. Arch Gen Psychiatry 2003;60:415-24. 21. Stern E, Silbersweig DA, Chee KY, et al. A functional neuroanatomy of tics in Tourette syndrome. Arch Gen Psychiatry 2000;57:741-8. 22. Servello D, Sassi M, Geremia L, Porta M. Bilateral thalamic stimulation for intractable Tourette syndrome. Paper presented at the 14th meeting of the WSSFN, Rome, 2005. 23. Servello D, Porta M, Sassi M, Brambilla A, Robertson MM. Deep brain stimulation in 18 patients with severe Gilles de la Tourette Syndrome refractory to treatment; the surgery and stimulation. J Neurol Neurosurg Psychiatry 2007 (in press). 24. Egidi M, Carrabba G, Priori A, et al. Thalamic DBS in Tourette’s syndrome: case report. Paper presented at the 14th meeting of the WSSFN, Rome, 2005 25. Bawja RJ, de Lotbiniere AJ, King RA, et al. Deep brain stimulation in Tourette’s syndrome. Mov Disord 2007;22:1346-50. 26. Maciunas RJ, Maddux BN, Riley DE, et al. Prospective randomized double-blind trial of bilateral thalamic deep brain stimulation in adults with Tourette syndrome. J Neurosurg 2007;107:1004-14. 27. Houeto JL, Karachi C, Mallet L, et al. Tourette’s syndrome and deep brain stimulation. J Neurol Neurosurg Psychiatry 2005;76:992-5. 28. Van der Linden C, Colle H, Vandewalle V, Alessi G, Rijckaert D, De Waele L. Successful treatment of tics
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
with bilateral internal pallidum (GPi) stimulation in a 27-year-old male patient with Gilles de la Tourette’s syndrome. Mov Disord 2002;17:S341. Diederich NJ, Bumb A, Mertens E, Kalteis K, Stamenkovic M, Alesch F. Efficient internal segment pallidal stimulation in Gilles de la Tourette syndrome: a case report. Mov Disord 2004;19:S440. Shahed J, Poysky J, Kennedy C, Simpson K, Jankovic J. GPi deep brain stimulation for Tourette syndrome improves tics and psychiatric comorbidities. Neurology 2007;68:159-60. Flaherty AW, Williams ZM, Amimovin R, et al. Deep brain stimulation of the internal capsule for the treatment of Tourette syndrome: technical case report. Neurosurgery 2005;57:E403. Kuhn J, Lenartz D, Mai JK, et al. Deep brain stimulation of the nucleus accumbens and the internal capsule in therapeutically refractory Tourette-syndrome. J Neurol 2007;254:963-5. Goetz CG, Pappert EJ, Louis ED, Raman R, Leurgans S. Advantages of a modified scoring method for the Rush Video-Based Tic Rating Scale. Mov Disord 1999;14:502-6. Ackermans L, Temel Y, Bauer NJ, Visser-Vandewalle V. Vertical gaze palsy after thalamic DBS. Neurosurgery (in press). Visser-Vandewalle V, Van der Linden C, Ackermans L, . et al. Deep brain stimulation in Gilles de la Tourette’s syndrome. Guidelines of the Dutch-Flemish Tourette surgery study group. Neurosurgery 2006;58:E590. Mink JW, Walkup J, Frey KA, et al. for the Tourette Syndrome Association, Inc. Recommended guidelines for deep brain stimulation in Tourette Syndrome. Mov Disord 2006;21:1831-8. American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th ed. (Text revision). American Psychiatric Association, Arlington, VA; 2000. Robertson MM, Banerjee S, Kurlan R, et al. The Tourette syndrome diagnostic confidence index: development and clinical associations. Neurology 1999;53:2108-12. Leckman JF, Riddle MA, Hardin MT, et al. The Yale Global Tic Severity Scale: initial testing of a clinicianrated scale of tic severity. J Am Acad Child Adolesc Psychiatry 1989;28:566-73. Goetz CG, Leurgans S, Chmura TA. Home alone: methods to maximize tic expression for objective videotape assessments in Gilles de la Tourette syndrome. Mov Disord 2001;16:693-7.
178 Treatment of Aggressive Behavior G. Broggi . A. Franzini
Introduction From the following data and considerations came up the former thought to deliver chronic high frequency stimulation (HFS) within the postero-medial hypothalamus (pHyp) to treat patients affected by aggressive and impulsive behavior refractory to any conservative therapy: 1.
2.
3.
4.
The previous experience of Sano and Mayanagi [1], Arjona [2], Schwarcz et al. [3] and Ramamurthy [4] who performed stereotactic lesions within the pHyp to treat aggression. The more recent report of disruptive behavior induced by acute electrical stimulation within the so-called ‘‘triangle of Sano’’ in a parkinsonian patient [5]. The experience of pHyp HFS in cluster headache patients, in fact during pain bouts these patients may develop selfaggression and violent behavior suggesting a common anatomic involvement of the pHyp in the etipathogenesis of cluster symptomatology and impulsive uncontrolled behavior [1,3,4,6–12]. The experimental data about the connections between the pHyp, the amygdala, and the Papez circuit [11].
Finally we strongly considered the need to offer to these severely impaired patients the option of a more conservative and reversible treatment to substitute the lesional procedures, which were the only extreme alternative to major contentive measures and complete social isolation. #
Springer-Verlag Berlin/Heidelberg 2009
Our experience is based on a small series of cases, but it still represents the only neuromodulation procedure available to treat disruptive behavior in mentally retarded patients [7].
Patients Since 2002, we treated six patients (aged 21–68 years). Only one patient was female. All patients were of subaverage IQ. Two patients had refractory generalized multifocal epilepsy. The aetiology of the disease was: 1.
2. 3.
Posttraumatic in one case with bilateral damage of the temporomesial structures (> Figure 178-1) Congenital (unknown origin) in three cases (normal MRI) Cardiac arrest in one case (MRI demonstrated only diffuse damage of frontal cortex)
All patients needed major contentive measures and two were chronically hospitalized. Patients’ data are summarized in > Table 178-1.
Surgery Informant consense for surgery was obtained by the parents and/or legal guardians. The stereotactic implantation was performed with the Leksell frame (Eleckta, Stockholm, Sweden) under general anesthesia. Preoperative antibiotics were administrated to all patients. A preoperative
2972
178
Treatment of aggressive behavior
. Figure 178-1 Fusion between preoperative MRI spin echo inversion recovery T2 images and postoperative CT in stereotactic conditions. The electrodes are correctly placed within the postero-medial hypothalamus (pHyp). Note the posttraumatic lesions within both temporal lobes
. Table 178-1 Aetiology
Age
IQ
Epilepsy
Comorbidity
MRI
26 34
Not evaluable Not evaluable
Multifocal No
No No
Normal Normal
3 P.M. 4 C.A.
Idiopathic Perinatal toxoplasmosis Idiopathic Postanoxia
21 64
40 30
No No
5 D.C.
Posttraumatic
37
Not evaluable
No
No Insomnia Severe arterial hypertension No
6 C.C.
Idiopathic
20
30
Multifocal
No
Normal Bilateral frontal cortical atrophy Bilateral temporal porencephaly Normal
1 P.G. 2 B.A.
MRI (brain axial volumetric fast spin echo inversion recovery and T2 images) was used to obtain high-definition images for the precise determination of both anterior and posterior commissures and midbrain structures below the commissural plane, such as the mammillary bodies and the red nucleus. MR images were fused with 2-mm thick CT slices that were obtained under stereotactic conditions by using an automated technique that is based on a mutual-information algorithm (Frame-link 4.0, Sofamor Danek Steathstation, Medtronic, Minneapolis, MN). The work-station also provided stereotactic coordinates of the target: 3 mm behind the mid-commissural point,
5 mm below this point, and 2 mm lateral from the midline. The target planning based exclusively on the midcommissural point caused electrode misplacement in one patient as previously reported [13]. This kind of error is due to the anatomical individual variability of the angle between the brainstem and the commissural plane. To correct this possible error, we introduced a third anatomical landmark, which allowed final target registration. We called this landmark ‘‘interpeduncular nucleus’’ or ‘‘interpeduncular point’’ and it is placed in the apex of the interpeduncular cistern 8 mm below the commissural plane at the level of the maximum diameter of the
Treatment of aggressive behavior
mammillary bodies [8]. The Y value of the definitive target (anteroposterior coordinate to the mid-commissural point in the classical mid commissural reference system) was corrected in our patients and the definitive target coordinate was chosen 2 mm posterior to the interpeduncular point instead of 3 mm posterior to the midcommissural point. A dedicated program and atlas has been developed and is freely available on the internet to get the proper coordinates of the target (www.angelofranzini.com/BRAIN.htm). A rigid cannula was inserted through a 3 mm, coronal, paramedian twist-drill hole and placed up to 10 mm from the target. This cannula was used as both a guide for microrecording [14] (> Figure 178-2) and for the placement of the definitive electrode (Quad 3389; Medtronic). No autonomic responses or cardiovascular effects were evoked by intraoperative macrostimulation (185 Hz, 80 ms, 3 V). When side effects were ruled out at the standard parameters stimulation, the guiding cannula was removed and the electrode secured to the skull with microplates. Postoperative stereotactic CT was performed to rule out complications and was merged with the
178
preoperative MRI to confirm the correct electrode placement [15] (> Figure 178-1; > Figure 178-3). Bilateral implantable pulse generator (IPG) (Soletra, Medtronic, Inc.) was placed in the subclavicular pocket and connected to the brain electrode for chronic continuous electrical stimulation. The parameters of chronically delivered electrical currents were: 185 Hz, 60–90 ms, and 1–3 V in unipolar configuration with casepositive. The current amplitude was progressively increased till to the threshold of collateral side effects consisting in ocular movements impairment in all cases.
Results Case 1 had prompt disappearance of selfaggression and burst of uncontrolled violence became less frequent and disappeared completely within 3 weeks. The patient was reinserted in the family and started to attend a therapeutical community for mentally impaired patients as outpatient. Generalized epileptic seizures disappeared
. Figure 178-2 Microrecording from the target (pHyp). Two action potential of targeted neurons are magnified in the upper portion of the figure
2973
2974
178
Treatment of aggressive behavior
. Figure 178-3 Fusion between preoperative MRI spin echo inversion recovery T2 images and postoperative CT in stereotactic conditions. The electrodes are correctly placed within the pHyp
and partial seizures and absences were reduced to 50%. The antiepileptic drug therapy was consistently reconsidered and reduced to 30%. The follow-up is for 4 years. Case 2 had immediate disappearance of violence bursts and could be discharged from Hospital within 3 months. Discharge from hospital was possible and he was able to live in a therapeutic community for mentally disabled patients. Three years later, after the IPG was temporarily turned off for knee surgery, the violent behavior relapsed after the end of general anesthesia. When the chronic stimulation was restored, the therapeutic effect resulted considerably reducted in
spite of the increase of the current amplitude, which could not be set higher than 2 V due to the appearance of side effects. Psychiatrists, which had the patient in charge, suggested a possible evolution of the original disease to explain the loss of the therapeutic effect. Anyway, with the IPG turned on the burst of violence are still less frequent and less intense than in absence of stimulation. The follow-up is for 4 years. Case 3 had a marked reduction of the rate and duration of the violence attacks only when the amplitude of stimulation was set to 1.8 V few months after surgery. This patient is still quiet
Treatment of aggressive behavior
178
. Figure 178-4 Intraoperative scalp EEG recorded from temporal and frontal areas. Left: interictal EEG activity before insertion of the left electrode at the target. Right: interictal EEG activity after the left electrode insertion. Note the disappearance of spikes possibly due to the mechanical effect of the electrode penetrating the posterior hypothalamus. T3, left temporal activity; T4, right temporal activity; Fp1, left frontal activity; Fp2, right frontal activity
and the social activities improved consistently. Now she is able to attend dedicated community and the family integration is good. Violence bursts may appear only if the patient is provoked by adverse events. The follow-up is of 28 months. Case 4 had only improvement in sleep (before surgery he slept only 2 h per night and after surgery he sleeps more than 6 h per night). The behavior was not affected by the stimulation in spite of electrical current raised to 2 V amplitude. Two years after surgery the stimulator was turned off but the improvement of sleep was not reverted to the preoperative condition and at 3 years follow-up he still sleeps more than 6 h per night. The same patient had a stable decrease of arterial pressure and all antipertensive drugs could be withdrawan. Also, this effect is still present in spite of the IPG turned off. The follow-up is of 3 years. Case 5 had a prompt marked improvement of aggressive behavior and the family care became consistently easier. The therapeutic effect was stable at 1-year follow-up, but when both IPG are turned off the violent behavior reappears
within few hours. The left IPG has been recently removed due to skin erosion and the therapeutic effects seems to be sustained only by the right side stimulation of the pHyp, but a replacement of the IPG is planned. The follow-up is of 1 years. Case 6 had only 6 months follow-up but it was impressive that the rate of epileptic seizures decreased to 50% of the preoperative condition just in the early postoperative weeks. In this patient, the insertion of the second electrode at the target was immediately followed by the disappearance of interictal epileptic activity from the scalp EEG (> Figure 178-4). The aggressive behavior is still present but the amplitude of the delivered current is still low (1 V).
Conclusion and Discussion From these series of patients the following remarks may be stressed: IQ subaverage patients in which violent and aggressive behavior is associated with pathologic conditions of different etiologies were benefited by
2975
2976
178
Treatment of aggressive behavior
HFS of the pHyp. The benefits ranged from a major stable improvement of behavior with considerable positive effects on social activity (Case 1, Case 4, and Case 5) to a less impressive improvement (Case 2, Case 3). No patient worsened after surgery and no patient developed new adverse symptoms. Possible cognitive modification should be investigated on long-term follow-up. In all patients medical treatment was significantly reduced. Patient 3 had a significant improvement of sleep and control of arterial pressure confirming published data concerning the effect of hypothalamic stimulation on sleep and cardiovascular system [1,3,4,6–13,15–17]. The patient with bilateral lesion of temporomesial structures (Case 5) responded well to therapeutic stimulation while the patient with diffuse frontal cortical damage (Case 4) had the worst outcome as regards the control of disruptive behavior. Patient 1 and 6 had a marked decrease of epileptic seizures, which were refractory to drug treatment before surgery; in both cases the pharmacological antiepileptic therapy was consistently reduced. This observation was also reported by Spinosa et al. who treated by pHyp HFS a patient with aggressive behavior and epileptic seizures (personal communication and poster presentation at the meeting of the AASFN held in Boston, June 2006). Experimental data are also available on this topic [10]. In conclusion, the reversibility and the positive effects of pHyp chronic stimulation made this procedure ethically acceptable in mentally retarded patients with violent aggressive behavior. The integrity of the frontal cortex seems to be a relevant factor affecting positively the therapeutic response. The possible adjunctive benefits of stimulation may include the control of refractory epilepsy, which sometimes is associated to this complex syndrome. In any case, the reported methodology is the only neuromodulation procedure available to treat disruptive and aggressive behavior [6,7,13] and it is still the only alternative to classical lesional surgery. The benefits on
sleep [17] and arterial pressure [6–9,13,15,16] control will be matter of further studies and may further wide the field of DBS application.
References 1. Sano K, Mayanagi Y. Posteromedial hypothalamotomy in the treatment of violent, aggressive behaviour. Acta Neurochir Suppl (Wien) 1988;44:145-51. 2. Arjona VE. Stereotactic hypothalamotomy in heretic children Acta Neurochir Suppl 1974;21:185-91. 3. Schvarcz JR, Driollet R, Rios E, Betti O. Stereotactic hypothalamotomy for behaviour disorders. J Neurol Neurosurg Psychiatry 1972;35:356-9. 4. Ramamurthy B. Stereotactic operation in behaviour disorders. Amygdalotomy and hypothalamotomy. Acta Neurochir Suppl (Wien) 1988;44:152-7. 5. Bejjani BP, Houeto JL, Hariz M, Yelnik J, Mesnage V, Bonnet AM, Pidoux B, Dormont D, Cornu P, Agid Y. Aggressive behavior induced by intraoperative stimulation in the triangle of Sano. Neurology 2002;59(9):1425-7. 6. Franzini A, Ferroli P, Leone M, Broggi G. Stimulation of the posterior hypothalamus for treatment of chronic intractable cluster headaches: first reported series. Neurosurgery 2003;52(5):1095-9. 7. Franzini A, Marras C, Ferroli P, Bugiani O, Broggi G. Stimulation of the posterior hypothalamus for medically intractable impulsive and violent behavior. Stereotact Funct Neurosurg 2005;83(2–3):63-6. 8. Franzini A, Marras C, Tringali G, Leone M, Ferroli P, Bussone G, Bugiani O, Broggi G. Chronic high frequency stimulation of the posteromedial hypothalamus in facial pain syndromes and behaviour disorders. Acta Neurochir Suppl 2007;97:399-406. 9. Takeda K, Bun˜ag RD. Sympathetic hyperactivity during hypothalamic stimulation in spontaneously hypertensive rats. J Clin Invest 1978;62(3):642–8. 10. Nishida N, Huang ZL, Mikuni N, Miura Y, Urade Y, Hashimoto N. Deep brain stimulation of the posterior hypothalamus activates the histaminergic system to exert antiepileptic effect in rat pentylenetetrazol model. Exp Neurol 2007;205:132–44. 11. Tarnecki R, Mempel E, Fonberg E, Lagowska J. Some electrophysiological characteristics of the spontaneous activity of the amygdala and effect of hypothalamic stimulation on the amygdalar units responses. Acta Neurochir Suppl 1976;23:135-40. 12. Torelli P, Manzoni GC. Pain and behaviour in cluster headache. A prospective study and review of the literature. Funct Neurol 2003;18:205-10. 13. Franzini A, Ferroli P, Leone M, Bussone G, Broggi G. Hypothalamic Deep Brain Stimulation for the treatment of Chronic Cluster headaches: a series report Neuromodulation 2004;7(1):1-8.
Treatment of aggressive behavior
14. Cordella R, Carella F, Leone M, Franzini A, Broggi G, Bussone G, Albanese A. Spontaneous neuronal activity of the posterior hypothalamus in trigeminal autonomic cephalalgias. Neurol Sci 2007;28(2):93-5. 15. Ferroli P, Franzini A, Marras C, Maccagnano E, D’Incerti L, Broggi GA. simple method to assess accuracy of deep brain stimulation electrode placement: preoperative stereotactic CT postoperative MR image fusion. Stereotact Funct Neurosurg 2004;82:14-19.
178
16. Cortelli P, Guaraldi P, Leone M, Pierangeli G, Barletta G, Grimaldi D, Cevoli S, Bussone G, Baruzzi A, Montagna P. Effect of deep brain stimulation of the posterior hypothalamic area on the cardiovascular system in chronic cluster headache patients. Eur J Neurol 2007;14(9):1008-15. 17. Vetrugno R, Pierangeli G, Leone M, Bussone G, Franzini A, Brogli G, D’Angelo R, Cortelli P, Montagna P. Effect on sleep of posterior hypothalamus stimulation in cluster headache. Headache 2007;47(7):1085-90.
2977
180 Apnea: Phrenic Nerve Stimulation* S. Rehncrona . G. Sedin . H. Fodstad
Historical Notes Already more than a hundred years ago electric stimulation of the phrenic nerves was discussed as a possibility for reanimation of breathing and resuscitation [1–4]. At the Harvard School of Public Health, JS Sarnoff and collaborators in the late 1940s, after a series of animal experiments, were able to show that ‘‘electrophrenic respiration’’ could maintain adequate breathing in man for considerable time periods [5,6]. In 1968 the first patient with chronic phrenic nerve stimulation for primary alveolar hypoventilation was published [7]. Subsequently, in collaboration with the Avery Laboratory (Avery Biomedical Devices, Inc., Commack, NY, USA), William Glenn developed a commercially available diaphragm pacing system based upon permanently implanted phrenic nerve electrodes. These electrodes are connected to a subcutaneously placed radio frequency receiver controlled by an external transmitter supplying the electrical energy. Later other stimulators like Atrostim1 (Atrotech Oy., Tampere, Finland) and MedImplant (Vienna, Austria) were constructed and introduced into the market [8,9]. The different types of equipment have now been available for the treatment of patients for more than 30 years. Despite the relatively few number of patients needing diaphragm pacing, it can be estimated that at present more than 2,500 operations have been
*Also termed ‘‘electrophrenic ventilation’’, while ‘‘diaphragm pacing’’ depicts both phrenic nerve stimulation and muscle stimulation using electrodes directly implanted in the diaphragm. The latter is not in the scope of the present overview. #
Springer-Verlag Berlin/Heidelberg 2009
performed worldwide, the majority of which with the Avery system. Essentially the devices consist of an implanted electrode connected to an implanted radiofrequency receiver in contact transcutaneously, via an antenna, with an external transmitter (> Figure 180‐1). Repetitive stimuli activate the nerve resulting in contractions of the diaphragm muscle (> Figure 180‐2). These rhythmic contractions will cause downward movements of the diaphragm and thereby mimic normal breaths. Respiratory frequency, progressive increase in amplitude (slope of the curve), pulse width and interval, inspiratory and expiratory times can be varied and optimized for each patient by programing the device.
Physiology and Anatomy Normal breathing is accomplished by active inspiration of air by increase of the thoracic volume creating a negative intrathoracic pressure for filling the lungs with air. The expansion of the chest needs activation of the breathing muscles, the most important of which being the intercostal muscles increasing the horizontal and the diaphragm muscle increasing the vertical chest volume. Expiration is passive and implies a relaxation of these muscles. While the intercostal muscles receive innervations from the thoracic spinal nerves, the phrenic nerves supply the diaphragm muscle. The phrenic nerves emanate from cell bodies in the anterior horn of the cervical spinal cord segments C3–C5 and descend in close vicinity of the truncs of the cervical plexus, frontal to the
2992
180
Apnea: phrenic nerve stimulation
anterior scalenus muscle and posterior to the jugular vein before entering the thoracic aperture. The nerves then pass through the mediastinum close to the pleura and pericardium on either side of the heart for finally splitting up in a number of branches innervating the hemidiaphragms. The respiratory nerves are controlled mainly from the autonomic respiratory centre situated in the brainstem, but also from cortical neurons enabling a voluntary control of breathing. The most important of the neuronal control is the autonomic one efficacious also during sleep and at lower degrees of unconsciousness. The brainstem respiratory centre is mainly responsive to changes in blood PCO2 –levels and hypercapnia will stimulate while hypocapnia will inhibit respiration but to . Figure 180‐1 Phrenic nerve stimulator. (Avery device) (Ill. Roy Hallgard)
a lesser degree by the arterial PO2. Any block in the neuronal circuitry connecting the central nervous control system with the respiratory muscles will cause impaired breathing, hypoxia and, if total, lead to apnoea and death.
Indications and Patient Selection Diaphragm pacing by stimulation of the phrenic nerves is more a physiologic alternative for artificial respiration than positive pressure insufflations of air by means of a mechanical respirator in patients with chronic neurogenic ventilatory failure. Main candidates for diaphragm pacing are patients suffering from an insensitive respiratory centre, either congenital or acquired, as well as high spinal cord lesions (above the C3 –level). Absolute conditions for the method to be applicable include intact phrenic nerve function and absence of significant atrophy of the diaphragm. Another important condition is that the patient and his/ her closely related are willing and motivated for taking active part in the pacing treatment.
Central Alveolar Hypoventilation (CAHV, Ondine’s Curse) CAHV is caused by insensitivity of the respiratory center to increases in blood PCO2 resulting . Figure 180-2 Respiratory cycle with phrenic nerve stimulated inspiration and passive (not stimulated) expiration phases (Avery device) (Ill. Roy Hallgard)
Apnea: phrenic nerve stimulation
in progressive hypercapnia to levels inducing a state of general anesthesia, apnoea, hypoxia, and finally death. Voluntary respiration is not affected and, therefore, this deleterious development only arises during sleep, in fact, at deeper sleep than REM-sleep since the cortical activity during REMsleep will stimulate voluntary respiration mechanisms. The name Ondine’s curse emanates from mythology. There are several versions in folklore about a mermaid called Ondine (or Undine) who, because of her husbands’ unfaithfulness, delivered him a curse that he should stop breathing and die if he ever fell asleep. The most widely known version of this tale was written in 1811 by Friedrich de la Motte Fouquet and stimulated Severinghaus and Mitchell to use her name for CAHV [10]. The congenital form of CAHV is assumed to be caused by a sporadic gene mutation [11] and may be one cause of the sudden infant death syndrome. Acquired CAHV may arise as a consequence of vascular brain stem lesion, tumor, and encephalitis. Correct diagnosis implies a thorough examination at a sleep laboratory with EEG, PCO2, and PO2 analyses during different sleep levels as well as tests of respiratory responses to increased PCO2.
Spinal Cord Lesions Spinal cord lesions at the C1–C2 levels may typically be results of traumatic injuries to the spine after diving-, traffic- and fall accidents. If severe, this may cause a functionally complete or closeto-complete transversal section of the medulla and block of the nervous communication between brain and body with quadriplegia and loss of voluntary as well as automatic respiration. These patients will permanently be dependant on artificial respiration but retain the function of the peripheral phrenic nerves, which can be used for electric pacing of the diaphragm. Severe spinal cord lesions at the levels C3–C5 in most cases also lose the majority of cell bodies of the phrenic
180
nerves, which will rapidly degenerate and are impossible to stimulate. This problem may however be overcome by a combined surgical technique anastomosing intercostal nerves to the distal portion of the phrenic nerve and implanting a nerve stimulator [12]. In lesions below C5 the patients retain adequate phrenic nerve function and their ability for both voluntary and automatic diaphragmal respiration, which eliminates the need for artificial respiration.
Presurgical Evaluations Before surgery it should be first ensured that pulmonary function with gas exchange is adequate and that no primary lung disease exists. Second the function of the phrenic nerves must be evaluated. This is done with EMG measurement of the nerve conduction time combined with fluoroscopic measurements of diaphragmal contraction upon percutaneous nerve stimulation. The phrenic nerves are easily accessed for test stimulation at 2–3 cm above the clavicles at the posterior border of the sternocleidomastoideus muscles. As measured fluoroscopically the dome of the diaphragm muscle should descend at least 4 cm upon stimulation of the nerve if permanent pacing should be considered [4,13,14]. Simultaneous spirometric measurement of the tidal volume is valuable, but may not be necessary. During fluoroscopy, a possible pulling of the mediastinum to the side ipsilateral to stimulation may diminish the space for filling the lung, should be observed. This is not unusual in small children and, if so, bilateral phrenic nerve stimulation activated simultaneously must be considered.
Surgical Techniques There a few fundamental differences between different models of phrenic nerve stimulators concerning implantation procedures. We, therefore,
2993
2994
180
Apnea: phrenic nerve stimulation
here summarize the original techniques as developed by Glenn [15] and used at our own departments. Two different approaches can be recommended: the cervical (often preferred by neurosurgeons) and the thoracic (often preferred by thoracic surgeons) approaches. General anesthesia is used but, muscle relaxants should be omitted to allow intraoperative nerve test stimulation and spirometry during stimulation.
. Figure 180-3 Surgical implantation of cervical electrode for phrenic nerve stimulation (Avery device) (Ill. Roy Hallgard)
The Cervical Approach With the patient in prone position, head slightly turned backwards and rotated contralaterally, skin is opened by a five cm incision 2–4 cm above and in parallel to the clavicle passing over the posterior border of the sternocleidomastoid muscle. Platysma is opened and the sternocleidomastoid muscle retracted frontally and medially, exposing the scalenus anterior muscle and, frontal-medial, to the jugular vein. A small fat pad is usually located over the muscle and has to be retracted for exposing the phrenic nerve lying closely under the muscle facia and crossing the muscle from superior-lateral to inferior-medial (> Figure 180-3a). For localizing the nerve a bipolar nerve stimulator may be used for electrical testing and observing the muscle response from the diaphragm. After 5–6 mm incisions through the scalenus fascia on both sides of and in parallel to the nerve, the electrode is placed under the nerve so that the nerve will be located into the electrode semicircular groove (> Figure 180-3b). Finally the electrode wings are sutured to the muscle fascia on both sides (> Figure 180-3c). The extension wire from the electrode is passed subcutaneously to a small new thoracic skin incision some distance below. A subcutaneous pocket is formed for implanting the transmitter, which will be connected to the lead extension wire. For the monopolar mode the electrode connects with the cathode while the anode contact is built-in into the transmitter.
Bipolar electrodes are implanted similarly and they differ only with respect to that both the negative and positive contacts are placed in the phrenic nerve electrode for a narrow electric field. The advantage with this is that it can be used in patients with cardiac pacemakers.
The Thoracic Approach This is essentially similar to what has been seen earlier and was also originally described by Glenn et al [15], and modified by Miller et al [16]. The thoracic approach was advocated mainly because of three reasons. First, in some (roughly 25%) the phrenic nerve may receive an accessory branch joining the main trunk below the clavicle [14,17]. Second, to avoid a risk of undesired sensory effects and muscular twitches in the arm because of the stimulation of the lower nerve trunks of the cervical plexus. Third, the great mobility of the head and neck could contribute to electrode dislocation or break. With this approach the skin is incised (about 10–15 cm) on the anterior-lateral thorax between the second and third ribs. The ribs are spread and
Apnea: phrenic nerve stimulation
the lung retracted for access to the anterior mediastinum and the phrenic nerves, situated at a depth of around 5 cm. This technique may have advantages, but the disadvantage is that a thoracotomy with some risk for complications may be detrimental in a patient with already impaired respiratory function. The surgical complications reported have, however, been few and recently, minor traumatic approaches were suggested by using endoscopic and robotic techniques [18,19]. The timing of surgery is important and must be individualized especially for patients requiring full time mechanical ventilatory support. After a traumatic high cervical cord lesion with initial quadriplegia, some neuronal recovery may occur after several months with a possibility of partly return of ability of spontaneous breathing. In the early course of events, it is impossible to judge to which extent this capacity will be functionally sufficient. Therefore, a period of waiting before arriving at a definite decision is motivated. On the other hand due to disuse, atrophy of the diaphragm starts very early and may by time reach levels difficult to recondition [20]. It is not unlikely that early surgery should promote an optimal result of the treatment.
180
any or only little diaphragma atrophy and the muscle does not need to be trained. Phrenic nerve stimulation can, therefore, be started with long time periods almost directly. However, it is recommended that the patients in the beginning use pacing in 1–3 h periods while awake for habituation before using continuous pacing for maximal 12 h during sleep. Usually these patients only need a unilateral phrenic nerve stimulator, since only pacing is necessary during the sleeping hours and the risk of muscular fatigue is little. In patients totally dependent of artificial respiration, there is always atrophy of the diaphragm which needs training. For each patient an individual conditioning program should be made up starting pacing in 3–5 min periods per hour and prolonging these periods successively to the point when continuous pacing can be used. Usually this conditioning period will take 3–6 months. In quadriplegic patients bilateral implants are necessary and in order to prevent muscle fatigue the patient should stimulate unilaterally maximally for 12 h and then switch to stimulate the other side. The alternative is alternate stimulation of the two sides in shorter intervals (2–3 h) in accordance with individual preferences.
Postimplantation Management After surgery the patient should be allowed a 12–14 day period of rest for wound healing and postoperative tissue edema to resolve, before starting the stimulation. During fluoroscopic observation of the diaphragm contractions, the amplitude threshold for muscle response is recorded, with spirometry the amplitude is now increased until maximal tidal volume is obtained. The transmitter is set at the lowest value for obtaining maximum tidal volume. The slope of the amplitude as well as the respiratory rate is set (usually 12–14 per min). Since patients with central hypoventilation syndromes retain their capacity of voluntary breathing, there are seldom
Outcomes Several papers published during the 1970s verified the efficacy of phrenic nerve stimulation for the treatment of neurogenic respiratory failure in central hypoventilation syndromes and in quadriplegic patients [21–24]. In 1988 Glenn et al [25] reviewed the records of 165 patients treated in six centers included in a cooperative study. Sixty-five patients received unilateral stimulators and 100 bilateral, 37% were men and 64% women. Diagnoses were: cervical cord and brain stem lesions (64%); hypoventilation syndromes (35%), and 1% peripheral lesions.
2995
2996
180
Apnea: phrenic nerve stimulation
Wound infections were seen in 4.5% of all implantation procedures. In seven cases removal of the device was necessary, but pacing could be reinstituted after reimplantations. Surgical risk for nerve injury was lower with the monopolar electrode than with bipolar and the lowest risk of injury to the nerve was found with monopolar electrodes implanted with the thoracic approach. At follow up close to 80% of the patients were either independent of ventilation or needed minimal periods of support with mechanical ventilation. The majority (65%) lived at home, while 23% were hospitalized and 13% stayed in rehabilitation units. Fodstad [26] reported postoperative results follow-up ranging from 2 months to 10 years in thirty-five patients with either central hypoventilation or spinal cord lesions treated with phrenic nerve mediated diaphragm pacing. Seven patients died because of reasons not related to stimulation and five stopped pacing. Fifteen patients were totally independent while eight were partly dependent on mechanical ventilation. In another long-term follow-up study of 12 patients with complete respiratory paralysis from high cervical spinal cord lesions, all treated with stimulation of the thoracic phrenic nerves, full time pacing was accomplished from 0.5 to 16 years [27]. Nine of the patients could live at home; one of them married and was able to work full time; two completed collage studies. A recently published study [28] prospectively collected data from 64 patients with high spinal cord lesions and compared the outcome of phrenic nerve stimulation with the outcome of mechanical ventilation support. Data from the study period of more than 20 years show a statistically significant lower incidence of respiratory infections; significantly better quality of speech; and more patients managing employments in the 32 patients with stimulator controlled breathing than the 32 with mechanical ventilators. Subjective evaluations of quality of life by the patients and their doctors also favored the stimulation treatment and SCIM (the Spinal
Cord Independence Measure) increased by a factor four as compared to mechanical respiratory treatment. The incidence of long-term pacer complications (intermittent or absent pacing) was similar in a pediatric patient group (n = 35) and an adult group (n = 29), but higher in active children with Ondine’s curse than in adult or children with quadriplegia [29]. After interventions successful pacing was reached in 94% of the children and 86% of the adult patients.
Concluding Remarks In highly selected patients with central hypoventilation syndromes and spinal cord lesions above the C3 level, support of respiration by permanently implanted phrenic nerve stimulators may be of great value. The technique affords a possibility for more independency, reduces the incidence respiratory infections, may reduce the need for permanent tracheotomy, and reduces nursing care. Despite a significant financial investment for each system to be implanted, calculations indicate that initial costs may be paid off within 1–3 years [28,30]. Furthermore, for the individual sufferer benefits to costs ratio may well be favorable. Future improvements of the technique should take measures to develop fully implantable system including built-in energy source in similarity to deep brain and dorsal column stimulator devices. This would avoid risks for unintentional decoupling of antennas connecting the external transmitter with the internal receiver. Closed loop systems that could respond to changes in the patients’ actual metabolic needs and automatically adapt to the ventilation accordingly would also be a target for future research efforts. For patients with some retained capacity of voluntary breathing, a system that could synchronize stimulation-initiated breaths with voluntary initiation of breaths is warranted.
Apnea: phrenic nerve stimulation
References 1. Hufeland CW. Usum uis electriciae in asphyxia experimentis illustratum. Go¨ttingen, Germany: Dissertatio Inauguralis Medica; 1783. 2. Duchenne GBA. De l’Electrisation Localisee et de son Application a la Pathologie et a le Therapeutique par Courant Induits et apr Courants Galvanique Inerrompus et Continus par le Dr. Duchenne. Paris, France: Bailliere; 1872. 3. Beard GM, Rockwell AD. A practical treatise on the medical and surgical uses of electricity. New York: William Wood; 1886. p. 664. 4. Creasey G, Elefteriades J, DiMarco A, Talonen P, Bijak M, Girsch W, et al. Electrical stimulation to restore respiration. J Rehabil Res Dev. 1996;33:123-32. 5. Sarnoff SJ, Hardenbergh E, Whittenberger JL. Electrophrenic respiration. Science 1948;108:482. 6. Whittenberger JL, Sarnoff SJ, Hardenbergh E. Electrophrenic respiration. II. Its use in man. J Clin Invest. 1949;28:124-28. 7. Judson JP, Glenn WW. Radio-frequency electrophrenic respiration. Long-term application to a patient with primary hypoventilation. JAMA 1968;203:1033-37. 8. Talonen PP, Baer GA, Ha¨kkinen V, Ojala JK. Neurophysiological and technical considerations for the design of an implantable phrenic nerve stimulator. Med Biol Eng Comput. 1990;28:31-37. 9. Holle J, Mortiz E, Thoma H, Lischka A. Karusselstimulation, a new metod of electrophrenic long-term nerve stimulation. Wien Klin Wochenschr 1974;86:23-27. 10. Severinghaus JW, Mitchell RA. Ondine’s curse – failure of respiratory center automaticity while awake. Clin Res. 1962;10:122. 11. Dubreuil V, Ramanantsoa N, Trochet D, Vaubourg V, Amiel J, Gallego J, et al. A human mutation in Phox2b causes lack of CO2 chemosensitivity, fatal apnea, and specific loss of parafacial neurons. Proc Natl Acad Sci. 2008;105:1067-72. 12. Krieger LM, Krieger AJ. The intercostals to phrenic nerve transfer: An effective means of reanimating the diaphragm in patients with high cervical spine injury. Plast Reconstr Surg. 2000;105:1255-61. 13. Devine A, Watt JWH. Anaesthesia and diaphragmatic pacing in patients with tetraplegia. A review of perioperative management in patients over a 10-year period. Eur J Anaesthesiol. 1966;13:553-61. 14. Glenn WWL, Sairenji H. Diaphragm pacing in the treatment of chronic ventilatory insufficiency. In: Roussos C, Macklem PT, editors, The thorax: lung biology in health and disease, Marcel Dekker; New York, NY: p. 1407-440. 15. Glenn WWL, Hogan JF, Phelps ML. Ventilatory support of the quadriplegic patient with respiratory paralysis by diaphragm pacing. Surg Clin North Am. 1980;60:1055-78.
180
16. Miller JI, Farmer JA, Stuart W, Apple D. Phrenic nerve pacing of the quadriplegic patient. J Thoracic Cardiovasc Surg. 1990;99:35-40. 17. Oda T, Glenn WWL, Fukuda Y, Hogan JF, Gorfien J. Evaluation of electrical parameters for diaphragm pacing: an experimental study. J Surg Res. 1981;30:142-53. 18. Shaul DB, Danielson PD, McComb JG, Keens TG. Thoracoscopic placement of phrenic nerve electrodes for diaphragmatic pacing in children. J Pediatr Surg. 2002;37:974-78. 19. Morgan JA, Morales DL, John R, Ginsburg ME, Kherani AR, Vigilance DW, et al. Endoscopic, robotically assisted implantation of phrenic pacemakers. J Thorac Cardiovasc Surg. 2003;126:582-83. 20. Levine S, Nguyen T, Taylor N, Friscia ME, Budak MT, Rothenberg P, et al. Rapid disuse atrophy of diaphragm of fibers in mechanically ventilated humans. N Engl J Med. 2008;358:1327-35. 21. Glenn WW, Holcomb WG, Gee JB, Rath R. Central hypoventilation: long-term ventilation assistance by radiofrequency electrophrenic respiration. Ann Surg. 1970;172:755-73. 22. Glenn WW, Holcomb WG, McLaughlin AJ, O’Hare JM, Hogan JF. Yasuda R. Total ventilatory support in a quadriplegic patient with radiofrequency electrophrenic respiration. N Engl J Med. 1972;286:513-16. 23. Glenn WW, Holcomb WG, Shaw RK, Hogan JF, Holschuh KR. Long-term ventilation support by diaphragm pacing in quadriplegia. Ann Surg. 1976;183: 566-77. 24. Fodstad H, Blom S, Linderholm H. Artificial respiration by phrenic nerve stimulation (diaphragm pacing) in patients with cervical cord and brain stem lesions. Scand J Rehabil Med. 1983;15:173-81. 25. Glenn WWL, Brouillette RT, Dentz B, Fodstad H, Hunt CE, Keens TG, et al. Fundamental considerations in pacing of the diaphragm for chronic ventilatory insufficiency: A multi-center study. PACE 1988;11:2121-27. 26. Fodstad H. Pacing of the diaphragm to control breathing in patients with paralysis of central nervous system origin. Stereotact Funct Neurosurg. 1989;53:209-22. 27. Elefteriades JA, Quin JA, Hogan JF, Holcoomb WG, Letsou GV, Chlosta WF, et al. Long-term follow up of pacing of the conditioned diaphragm in quadriplegia. PACE 2002;25:897-906. 28. Hirschfeld S, Exner G, Luukkaala T, Baer GA. Mechanical ventilation or phrenic nerve stimulation for trearment of spinal cord injury-induced respiratory insufficiency. Spinal Cord, advance online publ, 13 May 2008;doi:10.1038/sc.2008.43. 29. Weese-Mayer DE, Silvestri JM, Kenny AS, Ilbawi MN, Hauptman SA, Lipton JW, et al. Diaphragm pacing with a quadripolar phrenic nerve electrode: an international study. PACE 1996;19:1311-19. 30. Similowski T, Derenne JP. Stimulation phre´nique implante´e. Medicine et Terapeutique 2001;7:457-695.
2997
Special and Emerging Applications
179 DBS Disorders of Consciousness N. D. Schiff
Introduction Deep brain stimulation (DBS) methods have been applied in the thalamus, upper brainstem, spinal and allied targets in the basal ganglia in attempts to restore consciousness in chronically unconscious patients following severe brain injuries [1–10]. However, the use of DBS technology for the treatment of patients with disorders of consciousness not has advanced rapidly for several related reasons. Importantly, initial studies which have focused on patients in the vegetative state (VS) have shown limited evidence of effects and lacked statistical rigor in behavioral characterization and analysis of outcomes [10]. The increasing use of DBS as a mode of treatment for neuropsychiatric disorders raises the possibility that related applications for use in cognitive impairment following non-progressive brain injuries will now be revisited. Recent proposals have considered the specific application of central thalamic DBS to improve cognitive function in conscious patients with severe cognitive disabilities [11–13]. The proposed mechanism of action differs conceptually from earlier efforts. Patients with disorders of consciousness and severe disabilities who nonetheless retain either elements of goal-directed behavior and appropriate language production, or other higher cognitive functions show prima facie evidence of integrative processing within the forebrain. The application of central thalamic DBS in such patients ties directly to both the basic neurophysiological functions of these thalamic neurons in forebrain arousal regulation mechanisms and the underlying pathology associated with chronically
#
Springer-Verlag Berlin/Heidelberg 2009
impaired cognitive function following severe brain injury [13,14]. This chapter briefly reviews the history of earlier clinical and scientific studies of DBS for disorders of consciousness and recent developments. A review of related clinical and experimental data are also provided to indicate future research directions for further development of central thalamic stimulation as an investigational therapeutic method. Particular emphasis is given to a recent study that provided detailed and quantitative evidence of behavioral improvements with central thalamic stimulation in a single human subject who had remained in the minimally conscious state (MCS) for 6 years following a severe traumatic brain injury [15]. The application of DBS for human disorders of consciousness will be complex and involve solving difficult problems of patient selection criteria, study design and ethical frameworks to assess the proportionality of clinical goals and the proper development of the technique [9,16–19]. Morruzi and Magoun [20] first demonstrated that electrical stimulation of the brainstem reticular formation and midline thalamus in anesthetized animals could produce desynchronization of low frequency disorganized activity in the EEG and generated a background activity similar to the patterns seen in wakeful states. Together with other related experimental findings these experiments helped to foster the concept of an ‘‘ascending reticular activating system’’ that controlled sleep-wake cycling and was situated within the central midbrain and intralaminar regions of the thalamus. Although a direct anatomical pathway from the midbrain reticular formation (MRF) to
2982
179
DBS disorders of consciousness
the intralaminar nuclei of the thalamus (ILN) was eventually demonstrated [21] a modern view of arousal systems controlling the overall patterns of sleep and wake cycling of the forebrain no longer assigns a key role to the central thalamus (intralaminar and paralaminar regions) [14,22,23]. Ascending arousal is now seen to depend on brainstem and basal forebrain cholinergic neuronal populations and noradrenergic afferents from the locus ceruleus and serotoninergic afferents from the medial raphe most of which have direct cortical innervation [22,23]. The central thalamus, however, does receive very heavy innervation from these regions and acts as a key gating structure that can respond to variations in arousal regulation during the wakeful state [see ref. 14 for review].
DBS in Vegetative State Patients Beginning in the late 1960s and into the early 1980s a few published studies evaluated electrical stimulation of the tegmental midbrain and posterior components of the central thalamus (posterior intralaminar nuclei – centromedian parafasicularis complex, Cm-Pf), and basal ganglia (globus pallidus interna) as a method for restoring patterned arousal and consciousness in chronically unconscious patients (mostly in the vegetative state by current classification methods) [1–4]. In these initial studies nearly all patients showed eye opening and some fragmentary movements when receiving electrical stimulation consistent with a generalized arousal effect. These acute arousal responses nonetheless did not predict further improvement. Importantly, arousal responses, including wide eye opening, changes in autonomic function and shifts to higher frequency content (‘‘desynchronization’’) of the EEG represent a basic and broad activation of forebrain, brainstem, and spinal cord systems [14,22]. The presence of apparent wakefulness and fragmentary movements are not alone dispositive of an potential effect on outcome and specifically
they do not indicate a that recovery of higher integrative brain function will follow. In these early studies no examples of recovery of sustained interactive behavior and communication were reported. Of note, a single subject study from Sturm et al. [3] described a patient with focal injuries in the midbrain and thalamus after a posterior circulation stroke described as in ‘‘some kind of unconsciousness which was neither a manifest coma nor a typical apallic syndrome (older term for vegetative state).’’ In this patient, electrical stimulation of the anterior central thalamus produced brief recovery of simple command following (a behavioral feature above the level of VS and consistent with minimally conscious state, see below) but became ineffective over a few weeks of application. Circa 1990, a large multi-center trial involving neurosurgeons in France, Japan, and the United States was initiated by Medtronic, Inc in a series of VS patients who received implanted neurostimulator systems. [5–8]. In these studies a group of ~50 VS patients had DBS electrode placed in the centromedian thalamus or cervical spinal cord stimulation in the dorsal columns. The majority of the patients in these studies had suffered traumatic brain injuries although other etiologies, including anoxic encephalopathy were included [7]; most famously the VS patient, Terri Schiavo who later became the center of controversy in the US in 2005 was included in the trial. As in previous studies, patients in this trial were shown to have an acute behavioral arousal response with DBS with consistent physiological responses including shifting of the frequency content of the EEG to higher frequencies and marked increases in cerebral metabolic rates measured using positron emission tomography [5] (notable for demonstrating the retained functional integrity of the connections from the central thalamus in eliciting cerebral activation). However, substantive clinical improvements were not identified in the VS patients treated with DBS [5–7]. Two centers involved in the trial reported that a small number of patients
DBS disorders of consciousness
with traumatic brain injury showed significant functional improvement including recovery of independence [7,8]. However, these patients received DBS within the known timeframes for spontaneous recovery from VS after traumatic brain injury, beginning electrical stimulation 3–6 months into their recovery course. The probability of recovery of consciousness for VS patients who have suffered traumatic brain injury and remain in VS at these time points range from 35–16% [10]. More recently, one of the investigative teams has reported that the few patients in their cohort that improved did not meet international diagnostic criteria for VS [8] but had shown evidence of non-reflexive behavior consistent with the minimally conscious state (MCS) [24]. Unlike VS patients, MCS patients have a high probability of further recovery past 6 month time points including outcomes above severe disability [25]. Recent prospective studies of MCS patients show that significant spontaneous recovery may occur after 1 year [25] and quite rarely even after decades [26]. These observations place stringent limitations on the formal assessment of DBS interventions in patients with disorders of consciousness and require that careful diagnostic evaluations and a structured approach of assessment of DBS effects are employed, including formal blinding when collecting behavioral data [10]. The lack of evidence supporting the use DBS to improve function in VS patients can be understood in light of the underlying pathology of chronic VS. Patients remaining in VS until death (1 month after anoxic brain injury, or 3 months after traumatic brain injury) show consistent findings in neuropathologic studies [27]. Both post-traumatic and post-anoxic brain injuries of VS were associated with widespread neuronal death in the thalami with relatively fewer patients demonstrating widespread neocortical cell death (64% in anoxia, only 13% in traumatic brain injury [27]). While the DBS studies in VS patients were motivated by the concept
179
that DBS might support an overall state change, restoring desynchronized EEG and an accompanying wakeful brain state, the overwhelming structural disconnection of the typical VS brain (after severe injury and months of unchanging clinical features) was not considered in the studies. The majority of patients most likely had suffered widespread cerebral injury (as later seen in the Schiavo autopsy which confirmed placement of the DBS lead in the right posterior intralaminar thalamus [28]). Activation of the thalamic afferents notably could still drive the neocortex to produce an increase in overall cerebral metabolic rates and in some patients a general shift to higher frequencies in the EEG [5]. However, the widely disconnected and damaged forebrain in these patients lacked the capacity to restore integrative function. At present, the available data do not appear to support the application of DBS in VS patients.
DBS for Impaired Cognitive Function in Conscious Patients An important difference comparing the proposed application of central thalamic DBS for disorders of consciousness and other DBS interventions is that small lesions in the central thalamus can produce disorders of consciousness, not ameliorate the condition [29,30]. Recovery from MCS and related outcomes of severe brain injuries are most often approached through a statistical relationship to time after injury [29]. Given that the underlying mechanisms are poorly understood and not currently able to guide an approach to an individual patient the potentially significant risks of injury to critical structures within the central thalamus with implantation of DBS electrodes must be balanced by developing carefully designed trials and proportionately scaled risk/ benefit ratios (Fins) [17]. Proposed application of DBS techniques in conscious patients with chronically impaired
2983
2984
179
DBS disorders of consciousness
cognitive function [11–13] thus require that DBS intervention only be considered after spontaneous recovery is statistically unlikely, available less invasive options have been tested, and formal quantitative behavioral assessments demonstrate an unchanging baseline [10,15, and see further ethical considerations below]. The foremost consideration is that patient selection emphasize clinical evidence of consciousness and specifically elements of wakeful behavior that provide a potential basis for further recovery of interactive behavior. Patients fulfilling the diagnostic criteria for MCS represent the first level of re covery where clear and consistent demonstration of elements of purpose behavior are identified [24]. MCS patients typically show substantial evidence of fluctuations in behavioral performance during their wakeful states including in some cases inconsistent communication. To assess a causal influence of DBS on recovery, formal behavioral assessments must establish the baseline diagnosis and assure that natural recovery has plateaued [10,15]. Without a clear and established baseline of behavioral data it is not possible to track emergence of cognitively-mediated behaviors induced by DBS. All further recovery after application of DBS may be potentially seen as incidental to the intervention particularly if effects of DBS are mixed with both acute and enduring changes. A recent study reported findings from a study of DBS electrodes implanted bilaterally in the central thalamus (targeting the anterior intralaminar regions) as part of a pilot clinical trial in a 38-yrold male who remained in a minimally conscious state for 6 years following a severe traumatic brain injury [15]. Although the patient was unable to communicate reliably, prior characterization of brain function using fMRI showed preservation of bi-hemispheric large-scale cerebral language networks [30]. > Figure 179‐1 shows functional magnetic resonance imaging (fMRI) studies of the patient’s brain activation pattern in response to passive language stimulation. In this paradigm auditory stimuli were presented to the patient
and normal volunteers, each stimulus consistent of a recorded narrative which was played either forward in time or time-reversed (allowing no decoding of the speech content). As seen in the first panel of > Figure 179‐1, the patient showed a marked, bilateral network activation in response to the forward presentation of the narrative but not the time-reversed presentation. In contrast, normal subjects in the study show robust responses to both narrative presentations as illustrated by one normal subject’s data in the second panel of > Figure 179‐1. These fMRI findings suggest that the patient retained a latent capacity to broadly co-activate cerebral networks involved in language processing but may have less background activity of these networks at rest. This interpretation is strengthened by positron emission tomography imaging in the patient that revealed severe hypometabolism across cerebral structures in the wakeful state [30]. Importantly, eligibility criteria for the DBS study allowed only MCS patients with relatively widely preserved brain structure and clear evidence of interactive behavior with elements of language function (command following, verbalization, or inconsistent communication). The conceptual rational for DBS in this patient group differs from the earlier model of restoring wakeful patterns of global cerebral activation which were demonstrated (as reviewed above) not to correlate with recovery of higher integrative cerebral functions. In conscious patients with clear fragments of language responsiveness or goal-directed behavior, improving arousal regulation within functionallyconnected but inconsistently active large-scale cerebral networks may be possible with DBS with the goals of restoring communication and consistent behavioral responsiveness. In the DBS study a 4-month quantitative behavioral assessment was completed prior to insertion of the DBS electrodes and followed by a 2-month period with the electrodes remaining off to reassess the behavioral baseline which did not change [15]. After a 5-month titration phase of
DBS disorders of consciousness
179
. Figure 179‐1 Functional magnetic resonance imaging of passive language stimuli in a minimally conscious state patient and a normal subject. Yellow color indicates brain activation in response to spoken narrative, blue color indicates response when narrative play backwards (time-reversed), and red color indicate areas of overlapping activation. See reference 30 for further details
testing tolerance to DBS and varying stimulation parameters and duration of stimulation the patient entered into a 6-month double-blind alternating crossover study. Evaluation of the blinded crossover data showed that bilateral DBS of the central thalamus improved behavioral responsiveness with a significant increase in the frequency of attentive responsiveness, functional limb control and oral feeding during periods in which DBS was on as compared with periods in which it was off [15]. A detailed logistic regression modeling of the behavioral data demonstrated statistical linkage between the observed functional improvements
and recent stimulation history for both the crossover data and effects seen during the titration phase. > Figure 179‐2a shows a logistic regression analysis of object naming performance during the titration phase of testing (see timeline in > Figure 179‐2b). The top tracing shows the varying time periods of stimulation during this part of the trial during which increasing durations of stimulation (number of days) were tested to establish safety and tolerance. Over the same time period the intensity of stimulation also increased (i.e., progressively higher voltages) effectively producing a linear trend in the DBS
2985
2986
179
DBS disorders of consciousness
. Figure 179‐2 Central thalamic deep brain stimulation in the minimally conscious state. a. Logistic regression modeling of titration phase testing of object naming (see text and reference [15] for further details). b. Trial timeline and coronal MRI image of electrodes in situ
DBS disorders of consciousness
stimulation history across this time period. The behavioral time series (second tracing) indicates the number of intelligible verbalizations (labeled as 1) when prompted against other types of response (labeled as 0). These raw data show a clear increasing frequency of intelligible verbalization as time goes on following the initial onset of this behavior with the beginning of the electrical stimulation titration phase. The third tracing shows a best fitting probability model obtained by logistic regression for the observed behavioral data given the stimulation history and time as potential additional variables [see 15 for details]. The result of the modeling shows that both a linear dependence on time (here ambiguous in terms of contribution of DBS stimulation history which contains a linear component of increasing exposure and intensity and elapsed time alone) and a very strong dependence on the detailed time history of stimulation (blue distribution indicates bootstrap resampling of data and robustness of a coefficient, C, in the logistic regression model accounting for dependence on stimulation history as well as time represented by the ‘‘B’’ coefficient); red distributions indicate effects of random time shuffling of data which removes contributions of both the B and C coefficients).
Potential Mechanisms Underlying Central Thalamic DBS Effects on Chronic Brain Injury An important observation in the data shown in > Figure 179‐2a is the demonstration of carryover effects with a shifting baseline of function in the DBS OFF period arising after initiation of stimulation. The changing baseline be the result of strengthening activated synapses or other mechanisms underlying neuronal plasticity. In a rodent model of the DBS stimulation parameters used in the human study, continuous unilateral electrical stimulation of the central lateral nucleus
179
at 100 Hz with comparable current intensities (scaled for the rat brain) showed facilitation of untrained goal-directed seeking behavior and object recognition memory [31]. > Figure 179‐3 shows the typical location of stimulating electrodes in these experiments in the central lateral nucleus of the thalamus (3A,B) and the results of performance of an objection recognition behavior across three days comparing animals receiving electrical stimulation to sham-stimulated controls. Of note, the carry-over effects that are noted in these experiments arose with only exposing the animals to 30 min of stimulation a day over 3 days. During stimulation the stimulated rats also showed a notably increased level of exploratory motor behaviors and grooming activity consistent with an increased level of arousal [22]. A parallel study in other rats demonstrated that continuous stimulation with same stimulation parameters produced up-regulation of memory-related immediate early genes in the anterior cingulate cortex, motor cortex and hippocampus suggesting a mechanism for the observed carry-over effects through activation of long-term potentiation mechanisms [31]. Similar effects on behavioral responsiveness have been demonstrated in primate studies including central thalamic DBS effects on vigilance [32] and an early and detailed study of electrical stimulation of the mesencephalic reticular formation (with direct connections to anterior intralaminar regions) that produced improved perceptual discrimination and shortened reaction times [33]. These findings indicate that adjustments in arousal regulation may also be elicited in undamaged brains through central thalamic DBS and can be systematically studied to aid further development of the technique as applied to the injured brain. Another consideration is the specificity of connections between components of the central thalamus and cerebral cortex, basal ganglia and other subcortical structures. In contrast to earlier human neurosurgical studies [1–8] that targeted the centromedian nucleus of the posterior
2987
2988
179
DBS disorders of consciousness
. Figure 179‐3 Central thalamic electrical stimulation in the rat. a. Gross histological specimen. b. Corresponding atlas section illustration of location of electrode tip during experiment. c. Behavioral testing results of object recognition memory during electrical stimulation. Ordinate indicates ratio of time spent with a novel object compared to previously explored object (new), a standard method of testing recognition memory (see text and reference [31] for further details)
intralaminar thalamic nuclei, spinal cord and globus pallidus, as shown in > Figure 179‐2b, the study described above targeted the anterior components of the intralaminar system. These neurons receive greater innervation from the brainstem arousal systems [34] and are the principal thalamic targets of the midbrain reticular projection [21]. They are strongly and reciprocally connected with medial frontal cortical systems that regulate arousal level [35–38]. These regions of the thalamus are enriched with high concentration of calbindin staining neurons that project most strongly to supragranular cortical regions allowing for a parallel role in cerebral activation comparable to projections from the brainstem arousal systems [39]. The calbindin enriched regions of the central thalamus include
the central lateral nucleus, paracentral nucleus, parafasicularis nucleus and paralaminar regions of surrounding association nuclei. In addition to strong supragranular cortical projections, these neurons have projections to the striatum which likely play a key role in arousal regulation during goal directed behaviors that support their proposed use as DBS targets for cognitive neuromodulation [13,14].
Ethical Considerations and the Way Ahead Application of deep brain stimulation (DBS) for disorders of consciousness demands a careful assessment of risk/benefit ratios and development
DBS disorders of consciousness
of safeguards for the highly vulnerable research subject population. Fins has developed detailed framework outlining the ethical challenges in this area of clinical neuroscience innovation [17–19]. An important theme of Fins’ analysis is that the potential risks of harm to subjects be carefully balanced by providing the patient access to therapies that have a potential for the benefit of possibly restoring cognitive function [17–19]. Because severely brain–injured subjects cannot provide their own consent for investigational procedures these studies all require surrogate consent mechanisms. Study authorization in the study described above was developed through a staged surrogate authorization and the included a provision to obtain subject consent should decisional capacity be restored [15,19]. As this area of research evolves it will require considerable attention to the further development of frameworks for consent and formulation of readjusted risk/benefits analysis based on evolving understanding of the potentials and limitations of technique.
References 1. McLardy T, Ervin F, Mark V, Scoville W, Sweet W. Attempted inset-electrodes-arousal from traumatic coma: neuropathological findings. Trans Am Neurol Assoc 1968;93:25-30. 2. Hassler R, Ore GD, Dieckmann G, Bricolo A, Dolce G. Behavioural and EEG arousal induced by stimulation of unspecific projection systems in a patient with posttraumatic apallic syndrome. Electroencephalogr Clin Neurophysiol. 1969;27(3):306-10. 3. Sturm V, Kuhner A, Schmitt HP, Assmus H, Stock G. Chronic electrical stimulation of the thalamic unspecific activating system in a patient with coma due to midbrain and upper brain stem infarction. Acta Neurochir 1979;47: 235-44. 4. Cohadon F, et al. Deep brain stimulation in cases of prolonged traumatic unconsciousness. In: Lazorthes Y, Upton ARM, editors. Neurostimulation: an overview. Mt Kisco, NY: Futura Publishers; 1985. 5. Tsubokawa T, et al. Deep-brain stimulation in a persistent vegetative state: follow-up results and criteria for selection of candidates. Brain Inj 1990;4(4):315-27.
179
6. Cohadon F, Richer E. Deep cerebral stimulation in patients with post-traumatic vegetative state: 25 cases. Neurochirurgie 1993;39(5):281-92. 7. Hosobuchi Y, Yingling C. The Treatment of prolonged coma with neurostimulation. In: Devinsky O, Beric A, Dogali M, editors. Electrical and magnetic stimulation of the brain and spinal cord. New York: Raven Press; 1993. p. 247-52 8. Yamamoto T, Katayama Y. Deep brain stimulation therapy for the vegetative state. Neuropsychol Rehabil 2005;15(3–4):406-13. 9. Schiff ND, Fins JJ. Deep brain stimulation and cognition: moving from animal to patient. Curr Opin Neurol 2007;20:638-42. 10. Schiff ND, Giacino JT, Kalmar K, Victor JD, Baker K, Gerber M, Fritz B, Eisenberg B, O’Connor J, Kobylarz EJ, Farris S, Machado A, McCagg C, Plum F, Fins JJ, and Rezai AR. Reply: Arousal by stimulation of deep brain nuclei. Nature 2008;452:E1-2. 11. Schiff ND, Rezai A, Plum F. A neuromodulation strategy for rational therapy of complex brain injury states. Neurol Res 2000;22 (3):267-72. 12. Schiff ND, Plum F, Rezai AR. Developing prosthetics to treat cognitive disabilities resulting from acquired brain injuries. Neurol Res 2002;24:116-24. 13. Schiff ND, Purpura KP. Towards a neurophysiological basis for cognitive neuromodulation. Thalamus Relat Syst 2002;2(1):55-69. 14. Schiff ND. Central thalamic contributions to arousal regulation and neurological disorders of consciousness. Annals of New York Academy of Sciences 2008;1129:105–118. 15. Schiff ND, . et al. Behavioral improvements with thalamic stimulation after severe traumatic brain injury. Nature 2007;448:600-3. 16. Fins JJ. A Proposed ethical framework for interventional cognitive neuroscience: a consideration of deep brain stimulation in impaired consciousness. Neurol Res 2002;22:273‐8. 17. Fins JJ. Constructing an ethical stereotaxy for severe brain injury: balancing risks, benefits and access. Nat Rev Neurosci 2003;4(4):323-7. 18. Fins JJ. Clinical pragmatism and the care of brain damaged patients: toward a palliative neuroethics for disorders of consciousness. Prog Brain Res 2005;150:565-82. 19. Fins J, Giacino J, Rezai A, Schiff N. Ethical insights from a neuromodulation clinical trial to restore function in the minimally conscious state (MCS). In: The 36th annual meeting of the society for neuroscience, Atlanta, GA; 2006, Abstract 182.3. 20. Moruzzi G, Magoun HW. Brainstem reticular formation and activation of the EEG. Electroencephalogr Clin Neurophysiol 1949;1:455-73. 21. Steriade M, Glenn LL. Neocortical and caudate projections of intralaminar thalamic neurons and their synaptic
2989
2990
179 22. 23.
24.
25.
26.
27.
28. 29.
30.
DBS disorders of consciousness
excitation from midbrain reticular core. J Neurophysiol 1982;48:352-71. Pfaff D. Brain arousal and information processing. Harvard University Press; Cambridge, MA: Saper CB, Scammell TE, Lu J. Hypothalamic regulation of sleep and circadian rhythms. Nature 2005;437 (7063):1257-63. Giacino JT, et al. The minimally conscious state: definition and diagnostic criteria. Neurology 2002;58: 349-53. Lammi MH, Smith VH, Tate RL, Taylor CM. The minimally conscious state and recovery potential: a follow-up study 2 to 5 years after traumatic brain injury. Arch Phys Med Rehabil 2005;86(4):746-54. Voss HU, et al. Possible axonal regrowth in late recovery from minimally conscious state. J Clin Invest 2006;116:2005-11. Adams JH, Graham DI, Jennett B. The neuropathology of the vegetative state after acute insult. Brain 2000;123:1327-38. Fins JJ, Schiff ND. The afterlife of Terri Schiavo. Hastings Center Report 2005;34(4):8. Posner J, Saper C, Schiff N, Plum F. Plum and Posner’s diagnosis of stupor and coma. 4th ed. Oxford: Oxford University Press; 2007. Schiff ND, et al. fMRI reveals large-scale network activation in minimally conscious patients. Neurology 2005;64:514-23.
31. Shirvalkar P, et al. Cognitive enhancement with central thalamic electrical stimulation. Proc Natl Acad Sci 2006;103(45):17007-12. 32. Schiff ND, Hudson AE, Purpura KP. Modeling wakeful unresponsiveness: characterization and microstimulation of the central thalamus. In: The 31st annual meeting of the society for neuroscience, San Diego, CA; 2002, Abstract 62.12. 33. Fuster J. Effects of stimulation of the brain stem on taschistoscopic perception. Science 1958;127:150. 34. Steriade M, Jones E, McCormick D. Thalamus. Oxford: Elsevier; 1997. 35. Nagai Y, et al. Brain activity relating to the contingent negative variation: an fMRI investigation. Neuroimage 2004;21(4):1232-41. 36. Paus T, et al. Time-related changes in neural systems underlying attention and arousal during the performance of an auditory vigilance task. J Cognit Neurosci 1997;9:392-408. 37. Vogt BA, Rosene DL, Pandya DN. Thalamic and cortical afferents differentiate anterior from posterior cingulate cortex in the monkey. Science 1979;204(4389):205-7. 38. Morel A, et al. Divergence and convergence of thalamocortical projections to premotor and supplementary motor cortex: a multiple tracing study in the macaque monkey. Eur J Neurosci 2005;21(4):1007-29. 39. Jones EJ. The thalamus. 2nd ed. Cambridge: Cambridge University Press; 2007.
181 DBS for Bladder Dysfunction R. Almusa . M. M. Hassouna
Introduction The urinary bladder function is perturbed in patients with movement disorders. The following chapter presents the etiology of bladder dysfunction in patients with movement disorders and the effect of deep brain stimulation on the bladder function.
Physiology of Micturition Micturition is a complex event involving muscular and neural components. The muscular part is the smooth muscle of the bladder and urethra which is called the vesicourethral muscularis. The neural component comprises both central and peripheral, afferent (sensory) and efferent (motor) autonomic pathways which are coordinated by cephalic control centers, spinal cord nuclei and infraspinal relay stations in peripheral ganglia [1]. Alterations in any of these components may result in abnormalities in micturition. Peripheral pathways (> Figure 181-1) consist of: (1) Parasympathetic (pelvic) nerves which arise at the sacral level of the spinal cord and cause bladder contraction and urethral relaxation. (2) Sympathetic (hypogastric) lumbar nerves which inhibit bladder body contraction and enhance bladder base and urethral contraction. (3) Pudendal (Somatic) nerves cause external urethral sphincter contraction. Central pathways: Central pathways (> Figure 181-2) play a very important role in micturition control although still very little is known about the
#
Springer-Verlag Berlin/Heidelberg 2009
areas involved. There have been considerable recent advances in understanding how the central neurons coordinate the mechanisms involved in voiding and storage of urine. In cats, stimulation of specific nuclei in the brain stem, including anterior cingulate gyrus, preoptic area of the hypothalamus, amygdala, bed nucleus of the stria terminalis and septal nuclei was found to trigger bladder contractions [2,3]. It was also found that the medial preoptic area (MPO), which is the rostral part of the hypothalamus, projects directly to the Pontine Micturition center (PMC) and may play a role in the initiation of micturition [4]. Electrical or chemical stimulation in the PMC was found to produce bladder contractions [5]. PMC (or as it is called, Barrington’s area or M- Region) which is in the dorsal pontine tegmentum projects to the sacral cord intermediolateral cell column is excitatory in nature and contacts dendrites and somata of parasympathetic preganglionic bladder motor neurons [6]. Destruction of the PMC bilaterally might lead to chronic urinary retention [7]. PMC also plays a major role in promoting bladder sphincter synergy by providing direct synaptic inputs to the neurons that carry the excitatory outflow to the bladder (Sacral preganglionic neurons and GABAergic neurons in the sacral dorsal region) and those involved in mediating the inhibitory outflow to the external urethral sphincter [6] (> Figure 181-3). Those findings could explain why patients and/or animals with spinal cord injuries above the sacral level have a condition called detrusorsphincter dyssynergia where they experience great
3000
181
DBS for bladder dysfunction
. Figure 181-1 Sympathetic, parasympathetic, and somatic innervation of the urogenital tract (Campbell-Walsh Urology, Ninth edition: page 1937. Fig. 56-12)
difficulty emptying the bladder because when their bladder contracts, their urethral sphincter also contracts [9]. In clinical situations, Block et al. [6] performed a study on 17 male patients to evaluate the brain structures involved during micturition. Patients were scanned by PET scans 15 min before micturition, during micturition, 15 min after micturition and 30 min after micturition. The result showed increased blood flow during micturition in the right dorsomedial pontine tegmentum, the periaqueductal grey, the hypothalamus and the right inferior frontal gyrus in 10 patients who were able to void during the scanning. In the
other 7 patients who couldn’t void, scanning also showed increased blood flow during a trial of micturition in the right ventral pontine tegmentum which is similar to the result obtained in the cat studies. The study results suggest that those pontine and cortical micturition sites are predominantly on the right side. A similar study was performed again by Block et al. [8] but on 18 right handed women this time. It was performed using almost the same method and parameters for the male study. The result of this study showed significant increased blood flow during micturition in the right dorsal pontine tegmentum and right
DBS for bladder dysfunction
181
. Figure 181-2 Central pathways (Campbell-Walsh Urology, Ninth edition: page 1941. Fig. 56-16)
inferior frontal gyrus which is almost identical to the previous study on the male group. Another important pontine area which was found to have a role in the maintenance of continence is called the L-region which is located more laterally and ventrally than PMC (> Figure 181-3) and gives direct projections to the nucleus of Onuf in the sacral cord. This nucleus contains motor neurons innervating pelvic floor including urethral sphincter [6,10,11]. Bilateral lesions in the L-region in cats cause severe urge incontinence [5,7]. The thalamus plays a role in micturition through the ventral intermediate nucleus(VIM) which is considered as a relay station for the sensomotor pathways connecting cerebellar and somatosensory afferents with the motor and
premotor cortex. High frequency electrical stimulation of this nucleus enhanced the sensation of bladder filling during cystometry [12]. Matsurra et al. [13] found, in a study on 11 male volunteers, that PET (Positron emission tomography) of the brain in humans detected thalamic activation when the bladder was distended to its maximal capacity. Specifically, the brain activation was found in the pons, midbrain periaqueductal gray, anterior insula, putamen, thalamus and anterior cingulate gyrus. In the same study, six other male volunteers were instilled intravesically with ice water. This significantly activated several regions in frontal and parietal lobes, the amygdala-hippocampus area, and the crus cerebri ventral border. These data conclude that brain regions involved in bladder filling are
3001
3002
181
DBS for bladder dysfunction
. Figure 181-3 Overview of pathways between spinal and supraspinal structures involved in the control of micturition (Fig. 1 in [8])
functionally separated from those associated with bladder cold perception in healthy individuals. Many studies showed also that the basal ganglia, including subthalamic nucleus, seem to have an effect on voiding control but the role is still unclear. Presence of detrusor overactivity in Parkinson’s disease patients suggested an inhibitory role of the basal ganglia on the pontine micturition center (PMC) [14]. The hypothesis behind this is that increased inhibition of damaged basal ganglia in post stroke and Parkinson’s
disease patients may cause detrusor overactivity. Another explanation is that basal ganglia exert control of the voiding cycle via GABAergic neurons that descend directly to the PMC [15,16].
Mechanism of Storage and Voiding Reflexes Storage reflexes: During the storage of urine, distention of the bladder produces low-level bladder
DBS for bladder dysfunction
afferent firing. Afferent firing in turn stimulates (1) the sympathetic outflow to the bladder outlet (base and urethra) and (2) Pudendal outflow to the external urethral sphincter. These responses occur by spinal reflex pathways and represent ‘‘guarding reflexes,’’ which promote continence. Sympathetic firing also inhibits detrusor muscle and transmission in bladder ganglia. Voiding reflexes: At the initiation of micturition, intense vesical afferent activity activates the brain stem micturition center, which inhibits the spinal guarding reflexes (sympathetic and pudendal outflow to the urethra). The pontine micturition center also stimulates the parasympathetic outflow to the bladder and internal sphincter smooth muscle. Maintenance of the voiding reflex is through ascending afferent input from the spinal cord, which may pass through the periaqueductal gray matter (PAG) before reaching the pontine micturition center. (> Figure 181-2).
Parkinson’s Disease and Voiding Symptoms Parkinson’s disease (PD) is a neurological condition characterized by neural degeneration of substantia nigra pars compacta, a mesencephalic nucleus related to the cortical subcortical motor pathway. Lower urinary tract symptoms are common in PD cases, 60% of these patients having disturbances of bladder storage/emptying [17]. The most common symptoms are urgency, frequency, and urge incontinence (50–75%) while dysuria and urinary retention have also been reported less commonly [18]. The most frequent urodynamic finding is neurological detrusor overactivity in 45–93% [17,18]. The degree of bladder function deterioration appears to be linked with advancing disease severity with detrusor hyperreflexia occurring in 79% of patients and hyporeflexia in 16% [19]. Also, the degree of lower urinary tract symptoms
181
associated with this condition in general correlates with the severity and duration of disease [20]. The PD severity was staged in the previous studies [19,20] according to the Hoehn and Yahr stage of disability where stage 1 is mild and stage 5 is the most severe form of PD [21]. Detrusor-sphincter dyssynergia and detrusor overactivity with impaired contractile function were seen only at advanced stages (stage 3–5) whereas no significant bladder abnormalities were observed at mild stages (stage 1–2) [20].
Deep Brain Stimulation and Movement Disorders During the last two decades, Deep Brain Stimulation (DBS) has re-emerged as one of the most effective treatments for advanced movement disorders. It is an FDA approved treatment for PD. The DBS system consists of a lead that is implanted into the targeted brain structure (thalamus, GPi or STN). The lead is connected to an implantable pulse generator (IPG), which represents the battery of the system. Among the three different implantation sites, Subthalamic Nucleus Stimulation (DBS-STN) is currently the most common surgical procedure for PD with remarkable improvement of motor function ranging from 45 to70% (in a study of 49 Patients) [22]. Studies of thalamic stimulation reported rates of significant improvement between 85% of PD patients and 89% of essential tremor at a follow-up of 12 months [23] with maintenance of good result for motor symptoms for more than 6 years [24].
Deep Brain Stimulation and Voiding Function One of the most recent studies on this subject were conducted by Kessler TM, et al. [25] to
3003
3004
181
DBS for bladder dysfunction
assess the effect of Thalamic Deep Brain Stimulation (DBS) on lower urinary tract function in patients with essential tremor (ET). They evaluated prospectively seven patients with ET and a median age of 66 years who had DBS implanted into the VIM (ventral intermediate) nucleus of the thalamus. Two patients had unilateral DBS; the remaining patients had bilateral thalamic DBS, the tremor being controlled in all patients. The patients underwent a full urological evaluation including medical history, neurourological examination, urinalysis, and urine culture. They had to complete the validated International Prostate Symptom Score (IPSS) questionnaire, which is used to assess LUTS in both women and men, immediately before the urodynamic study with the stimulator turned on. Patients then underwent two urodynamic investigations during chronic stimulation (ON state) and two urodynamic investigations 30 min after turning off the stimulator (OFF state). Urodynamic assessment included the following parameters: bladder volume at first desire to void, bladder volume at strong desire to void, maximum cystometric capacity, change in detrusor pressure during filling cystometry, bladder compliance, detrusor overactivity, maximum detrusor pressure, detrusor pressure at maximum
flow rate, maximum flow rate, voided volume, post void residual and pelvic floor electromyographic activity. The study followed the standards recommended by the International Continence Society regarding methods, definitions and units [26]. The result showed a significant decrease in bladder volume at first desire and at strong desire to void, and at maximum cystometric capacity in all the patients. There was no significant difference in all other parameters assessed (> Table 181-1). Overall, the study results suggest a regulatory role of the thalamus in lower urinary tract function (which was manifested as an increase in the bladder tonus in this study) and therefore the thalamus could be a very promising target for the development of new therapies for lower urinary tract dysfunction. Some limitations of this study are that the results are derived from a small number of patients and the time between DBS lead implant and the urodynamic assessment varied widely which could affect the result. A study was conducted by Herzog et al. [27] aimed at assessing changes in brain activity during bladder filling in Parkinson’s Disease patients with an STN-DBS implant. The study included a
. Table 181-1 Comparison of urodynamic parameters between ON and OFF thalamic deep brain stimulation state using the Wilcoxon signed rank test
Storage phase of the bladder Bladder volume at first desire to void Bladder volume at strong desire to void Maximum cystometric capacity (ml) Change in detrusor pressure during filling cystometry Bladder compliance (ml/cm H2O) Voiding phase of the bladder Maximum detrusor pressure (cm H2O) Detrusor pressure at maximum flow rate (cm H2O) Maximum flow rate (ml/s) Voided volume (ml) Postvoid residual (ml) *Significant result
Median on
Median off
P value
218 305 345 4 74
365 435 460 4 103
0.031* 0.031* 0.016* 0.99 0.31
53 45 11 278 15
57 44 11 440 10
0.38 0.99 0.31 0.06 0.09
DBS for bladder dysfunction
total of 11 patients (six females and five males) with Idiopathic Parkinson’s Disease with a mean age of 57.7 10.9 years and mean disease duration of 15.2 3.6 years. Patients had bilateral STN-DBS implants. In all patients, STNDBS led to significant improvement of their motor dysfunction. Using the PET scan (positron emission tomography) to measure changes in regional cerebral blood flow (rCBF); all patients were studied during bladder filling in STN-DBS on and off status. STN-DBS led to significant changes in first desire to void and increasing bladder capacity. They summarized that the result of the study could explain the influence of the STN-DBS on bladder function in PD patients and showed that STN-DBS modulates activation of Lateral Frontal Cortex and Anterior Cingulate Cortex which exert inhibitory influence on the PMC allowing maintenance of continence. In a study conducted by Finazzi-Agro et al. [28] five patients (three males, two females) who had Parkinson’s Disease with a mean age of 63 3.7 years, had their voiding function evaluated. They had bilateral subthalamic nucleus electrode implantation (DBS-STN) for 4–9 months before the urodynamic study. Patients underwent urodynamic evaluation during chronic stimulation and 30 min after turning off the stimulators. The UDS included filling cystometry and pressure flow study with striated pelvic floor muscle electromyography. Urodynamic parameters included hyperreflexic detrusor contraction threshold and amplitude,
181
bladder capacity, detrusor- sphincter pseudodyssynergia, maximum urinary flow and detrusor pressure at maximum flow. The results showed that all patients showed increased reflex volume and bladder capacity, and decreased hyperreflexic detrusor contraction amplitude during the activation of the DBS. Subjectively, 3 of the 5 patients reported urinary symptoms improvement after the implantation of the stimulator by few weeks. However, there were no significant changes for the other urodynamic parameters (> Table 181-2). They concluded in this study that subthalamic nucleus stimulation seems to be effective for decreasing detrusor hyperreflexia in Parkinson’s Disease cases and thus confirms the importance of the basal ganglia for voiding control and the potential therapeutic role of modulating their action not only on motor, but also on urinary symptoms. On the other hand, a different study came up with different results from the previously mentioned ones. Winge et al. [29] performed a prospective study to evaluate lower urinary tract symptoms and bladder control in patients with Parkinson’s disease before and after implantation of electrodes in the subthalamic nucleus (STN). All patients completed two sets of validated questionnaires for the voiding symptoms (IPSS and Dan-PSS) before the surgery, then urodynamic assessment (UDS) after discontinuation of medical treatment for 12 h. After implantation of the STN-DBS, patients were asked to complete the same questionnaires 3 and 6 months after
. Table 181-2 Urodynamic parameters for ON and OFF subthalamic nucleus status
Hyperreflexic detrusor contraction threshold(ml) Hyperreflexic detrusor contraction amplitude (cm H2O) Bladder capacity (ml) Maximum urine flow (ml/sec) Detrusor pressure at maximum flow (cm H2O)
Median on
Median off
P value
250 23 320 10 45
110 37 130 12 45
0.043 0.223 0.043 0.655 0.480
3005
3006
181
DBS for bladder dysfunction
the procedure and the urodynamic evaluation was performed 6 months after the implantation procedure. A total of 16 patients completed the study out of 21. The median age of the patients was 58 years. The median times between performing the UDS and surgery were 2.5 months, and between implantation of electrodes and post operative investigation 6 months. They found that total number of mixed voiding symptoms and the severity of mixed lower urinary tract symptoms which were measured in the study by International Prostate Symptoms Score (IPSS) and Danish Prostate Symptoms Score (DanPSS) respectively did not change significantly after implantation of electrodes in the STN. The obstructed voiding symptoms were not affected by the DBS-STN implant but patients reported a significant decrease in irritative bladder symptoms after the implantation. In regard to UDS, none of the primarily measured UD parameters changed significantly after surgery.
Summary In conclusion, we think from all the previous studies that we mentioned that Deep Brain Stimulation is an effective therapy for treating movement disorders and it seems to have some effect on voiding dysfunction symptoms. It may decrease detrusor hyperreflexia in some patients, may affect the urodynamic parameters in other patients in unpredictable manner. On the other hand, other patients did not have any changes in their voiding symptoms. The contradiction in the results most likely is related to the relatively limited number of patients in all the studies. The studies supported a known fact that the basal ganglia plays a role in lower urinary tract behavior. Yet, further investigations and more studies with larger numbers of patients are needed to determine the potential therapeutic role of modulating their action on voiding
and whether DBS could be a management option for voiding dysfunction patients.
References 1. Elbadawi A. Anatomy and innervation of the vesicourethral muscular unit of micturition. In: Krane RJ, Siroky MB, editors. Clinical neuro-urology, 2nd ed. Boston: Little, Brown and Co; 1991. p. 5-23. 2. Gjone R, Setekleiv J. Excitatory and inhibitory bladder responses to stimulation of the cerebral cortex in the cat. Acta Physiol Scand 1963;59:337-48. 3. Gjone R. Excitatory and inhibitory bladder responses to stimulation of ‘limbic’, diencephalic and mesencephalic structures in the cat. Acta Physiol Scand 1966;66:91-102. 4. Holstege G. Some anatomical observations on the projections from the hypothalamus to brainstem and spinal cord: an HRP and autoradiographic tracing study in the cat. J Comp Neurol 1987;260:98-126. 5. Holstege G, Griffiths D, de wall H, Dalm E. Anatomical and physiological observations on supraspinal control of bladder and urethral sphincter muscles in the cat. J Comp Neurol 1986;250:449-61. 6. Blok BF, Willemsen AT, Holstege G. A PET study on brain control of micturition in humans. Brain 1997;120:111-21. 7. Griffiths D, Holstege G, de Wall, Dalm E. Control and coordination of bladder and urethral function in the brain stem of the cat. Neurourol Urodyn 1990;9:63-82. 8. Blok BF, Sturms LM, Holstege G. Brain activation during micturition in women. Brain 1998;121:2033-42. 9. Blaivas JG. The neurophysiology of micturition: a clinical study of 550 patients. J Urol 1982;127:958-63. 10. Holstege G, Kuypers HG, Boer RC. Anatomical evidence for direct brain stem projections to the somatic motoneuronal cell groups and autonomic preganglionic cell groups in cat spinal cord. Brain Res 1979;171:329-33. 11. Sato M, Mizuno N, Konishi A. Localization of motoneurons innervating perineal muscles: a HRP study in cat. Brain Res 1978;140:149-54. 12. Kessler TM, et al. Effect of thalamic deep brain stimulation on lower urinary tract function. Eur Urol 2007; doi:10.1016/j.eururo.2007.07.015. 13. Matsuura S, et al. Human brain region response to distention or cold stimulation of the bladder: a positron emission tomography study. J Urol 2002;168:2035-9. 14. Sakakibara R, Fowler CJ, Hattori T. Voiding and MRI analysis of the brain. Int Urogynecol J pelvic Floor Dysfunct 1999;10:192. 15. Sakakibara R, et al. SPECT imaging of the dopamine transporter with [123I]-b-CIT reveals marked decline of nigrostriatal dopaminergic function in Parkinson’s disease with urinary dysfunction. J Neurol Sci 2001;187:55.
DBS for bladder dysfunction
16. Limousine P, Greene J, Pollak P, Rothwell J, Benabid AL, Frackowiak R. Changes in cerebral activity pattern due to subthalamic nucleus or internal pallidum stimulation in Parkinson’s disease. Ann Neurol 1997;42:283. 17. Hattori T, et al. Voiding dysfunction in Parkinson’s disease. Jpn J Psychiatr Neurol. 1992;46:181. 18. Fowler CJ. Urinary disorders in Parkinson’s disease and multiple system atrophy. Functional Neurol 2001;16:277. 19. Araki I, Kuno S. Assessment of voiding dysfunction in Parkinson’s disease by the international prostate symptom score. J Neurol Neurosurg Psychiatr 2000;68:429. 20. Araki I, et al. Voiding dysfunction and Parkinson’s disease: urodynamic abnormalities and urinary symptoms. J. Urol 2000;164:1640-43. 21. Hoehn MM, Yahr MD. Parkinsonism: onset, progression and mortality. Neurology 1967;17:427. 22. Krack P, Batir A, Van Blercom N, Chabardes S, Fraix V, Ardouin C, et al. Five-year follow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinson’s disease. N Engl J Med 2003;349:1925-34. 23. Limousin P, et al. Multicentre European study of thalamic stimulation in Parkinsonian and essential tremor. J Neurol Neurosurg Psychiatr 1999;66(3):289-96.
181
24. Sydow O, Thobois S, Alesch F, Speelman JD. Multicentre Euroean study of thalamic stimulation in essential tremor: six years follow up. J Neurol Neurosurg Psychiatr 2003;74(10):1387-91. 25. Kessler TM, et al. Effect of thalamic deep brain stimulation on lower urinary tract function. Eur Urol 2007; doi:10.1016/j.eururo.2007.07.015. 26. Abrams P, Cardozo L, Fall M, et al. The standardisation of terminology of lower urinary tract function: report from the Standardisation Sub-committee of the International Continence Society. Neurourol Urodyn 2002;21:167-78. 27. Herzog J, Weiss PH, Assmus A, et al. Subthalamic stimulation modulates cortical control of urinary bladder in Parkinson’s disease. Brain 2006;129:3366-75. 28. Finazzi-Agro E, Peppe A, D’Amico A, et al. Effects of subthalamic nucleus stimulation on urodynamic findings in patients with Parkinson’s disease. J Urol 2003;169:1388-91. 29. Winge K, et al. Lower urinary tract symptoms and bladder control in advanced Parkinson’s disease: effects of deep brain stimulation in the subthalamic nucleus. Mov Disord 2007;22:220-5.
3007
186 Gene Therapy for Brain Tumors M. L. M. Lamfers . E. A. Chiocca
Introduction Despite advances in surgical techniques as well as radiation- and chemotherapies, the prognosis for patients diagnosed with malignant gliomas still remains dismal with a mean survival of only 14.4 months for those with a diagnosis of grade IV astrocytoma (WHO criteria) [1]. Development of alternative treatment modalities, preferably with mechanisms of action dissimilar to that of standard therapies is therefore urgently needed. Advances in our understanding of tumor biology and development of molecular techniques have led to extensive research in the field of gene therapy, immunotherapy, targeted toxins, and virotherapy. In this context, therapeutic approaches based on biological agents, including a large variety of viruses, have gained widespread attention the past two decades. In this chapter we will focus on preclinical and clinical developments of gene therapy and virotherapy for the treatment of malignant glioma.
Preclinical Gene Therapy Studies for Glioma Gene therapy encompasses an approach based on delivery of therapeutic genes to the target cell. Initially developed to correct genetic disorders based on single gene defects, gene therapeutic strategies were soon implemented in oncology research. For transfer of genetic material to the target cell, direct injection of naked DNA plasmids, or in complex with various types of lipid layers, has been applied. A more efficient technique involves
#
Springer-Verlag Berlin/Heidelberg 2009
the use of vectors derived from viruses. A wide variety of viruses have been genetically engineered for this purpose including retrovirus, human adenovirus, human herpes virus, and adeno associated virus. The vectors are designed to deliver a therapeutic gene (transgene) to the host cell without the normally inherent cytolytic activity of the virus, i.e., they are made replication defective. This is achieved by removing viral genes which are essential for the replication of the virus and replacing these regions with therapeutic anti-cancer genes. A great number of therapeutic genes have been studied in the context of malignant glioma over the past two decades (> Table 186-1).The gene therapeutic approach most widely applied thus far relies on transfer of the herpes simplex virus thymidine kinase (HSV-Tk) gene in combination with systemic delivery of its substrate ganciclovir (GCV). Expression of HSV-Tk by the infected cell leads to conversion of this non-toxic prodrug to its toxic form, thereby killing the cell. Other enzyme/prodrug systems include Cytosine Deaminase (CD) with 5-fluorocytosine (5-FC), which is converted to the toxic 5-fluorouracil (5-FU), Carboxyl Esterase (CE) with CPT-11 which is converted to SN-38, and Cytochrome P4502B1 with cyclophosphamide which is ultimately converted to phosphoramide mustard. A second gene therapeutic approach applied for glioma involves delivery of genes which induce apoptosis either directly (e.g., the apoptosisinducing ligand TRAIL), or indirectly by gene repair (e.g., p53). Alternatively, genes have been delivered which stimulate an anti-tumor immune response (e.g., interleukin-2) or which inhibit angiogenesis (e.g., Angiostatin). A more recent
Gene repair Gene repair Gene repair Growth inhibition
Apoptosis induction Apoptosis induction
P21 P27 PTEN EGF(R) inhibitors
TRAIL
Fas (Ligand)
Gene repair
P16
P53
HSV-Tk + connexin 43
Enzyme/prodrug
Cytochrome P450 (+ CPA) Deoxycytidine kinase (+ cytosine arabinoside) HSV-Tk + CD
Zhang [65]
Kock [60]
Wang [41]
Wang [41]
Wang [41]
Rogulski [34]
Manome [31], Chen [32]
Chen [14], Perez-Cruet [15], Vincent [12], Maron [16], Greco [17] Ichikawa [26], Miller [27]
Lee [61], Naumann [62], Rubinchik [63], Kim [64] Ambar [66], Rubinchik [63]
Badie [42], Kock [43], Li [44], Cirelli [45], Abe [46], Kim [47] Fueyo [49], Lee [50], Harada [51], Kim [47] Chen [52] Chen [53], Park [54] You [55]
Wang [35], Chang [36]
Enzyme/prodrug
CD (+5-FC)
Ezzeddine [8], Culver [9], Ram [10], Izquirdo [11], Vincent [12], Pizzato [13] Ge [22], Wang [23], Tai [24], Wang [25] Wei [28], Wei [29], Jounaidi [30]
Boviatsis [3], Badie [4]
Dual enzyme/ prodrug Enzyme/ prodrug + gap junctions component Gene repair
Enzyme/prodrug
HSV-Tk (+GCV)
Short [2], Boviatsis [3]
Adenoviral vectors
Manome [33]
Marker gene
b-Galatosidase
Retroviral vectors
Enzyme/prodrug
Therapeutic strategy
Ho [67]
Saydam [56]
Moriuchi [37] Marconi [39], Niranjan [40]
Niranjan [18]
Boviatsis [3]
Herpes vectors Okada [5], Enger [6] Okada [5], Mizumo [19]
AAV vectors
Pu [57], Kang [58], Zhang [59]
Hsiao [48]
Aghi [38]
Zerrouqi [20], von Eckardstein [21]
Kofler [7]
Non-viral vectors
186
Transgene
. Table 186-1 Overview of published preclinical gene therapy studies for malignant glioma
3084 Gene therapy for brain tumors
Apoptosis induction Apoptosis induction Apoptosis induction Apoptosis induction Apoptosis induction Angiogenesis inhibition Angiogenesis inhibition Angiogenesis inhibition Angiogenesis inhibition Angiogenesis inhibition Angiogenesis inhibition Angiogenesis inhibition Angiogenesis and invasion inhibition
Bcl-2 inhibitors
Angiogenesis and invasion inhibition Immunestimulation
Immunestimulation
Anti-HGF + anti-cMET ribozymes IL-2
IL-4
FGF-R1 (truncated form) Plasminogen or MMP inhibitors
Tissue factor pathway inhibitor-2 BAI-1
Platelet factor-4
VEGF inhibitors
Endostatin
Angiostatin
IL-1b convert. enzyme E2F-1
FADD
Bax
Apoptosis induction
Caspases
Ram [97], Vincent [12], Tseng [98], Sampson [99], Pizzato [13] Wei [101], Tseng [98], Sampson [99], Saleh [102], Okada [103]
Tanaka [85]
Sasaki [81], Niola [82]
Tanaka [78]
Yu [76]
Kondo [75]
Kock [60]
Wei [104], Lumniczky [105]
Donson [100]
Abounader [96]
Lakka [89], Lakka [90], Gondi [91]
Saiki [88]
Kang [87]
Ciafre [83]
Tanaka [78]
Fueyo [77], Mitlianga [72]
Mitlianga [72], Arafat [73]
Shinoura [68], Shinoura [69], Shinoura [70], Komata [71] Kock [60]
Ho [67]
Lee [74]
Okada [5]
Yanamandra [86]
Harding [84]
Ma [79]
Gondi [92], Lakka [93], Lakka [94], Gondi [95] Abounader [96]
Barnett [80]
Gene therapy for brain tumors
186 3085
Therapeutic strategy
Immunestimulation Immunestimulation
Immunestimulation
Immunestimulation
Immunestimulation Immunestimulation
Transgene
IL-12 TNFa
Interferons
GM-CSF
GM-CSF + B7-2 Flt3 ligand
Tseng [98], Sampson [99], LeFranc [118], Herrlinger [119] Parney [121]
Sampson [99], Saleh [112]
Sampson [99]
Retroviral vectors
Ali [122], Ali [123]
Lumniczky [105]
Ehtesham [107]
Donson [100], Liu [106] Ehtesham [107], Yamini [108], Yamini [109]
Adenoviral vectors
Herrlinger [120]
Niranjan [18], Moriuchi [110] Kanno [113]
Herpes vectors
Yoshida [114], Streck [115]
AAV vectors
Mizuno [111], Natsume [116], Natsume [117]
Mizuno [111]
Non-viral vectors
186
. Table 186-1 (Continued)
3086 Gene therapy for brain tumors
Gene therapy for brain tumors
development in gene therapy for glioma is the implementation of RNA interference technology in which small interfering RNAs (siRNAs) are introduced to silence specific (sets of) genes.
Retroviral Vectors The advancing field of gene therapy during the 1980s was first implemented in experimental glioma studies using retroviral vectors to deliver a (marker) gene to the tumor [2]. Retroviral vectors are derived from retroviruses, which comprise a variety of enveloped RNA viruses, of which the genome is converted to DNA by the viral enzyme reverse transcriptase prior to random integration into the host genome. To gain access to the host genetic material this process requires a cell that is actively dividing. This selectivity for mitotic cells was considered to make retroviruses particularly applicable to brain tumors which are surrounded by predominantly quiescent normal tissue. Most retroviral vectors used for gene therapy are derived from the Moloney Murine Leukemia Virus (MMLV).
Retroviral Gene Therapy Studies in Experimental Models of Malignant Glioma The first retroviral gene therapy studies were based on the ex vivo delivery of the HSV-Tk gene followed by in vivo GCV treatment [8]. Soon thereafter, direct intratumoral (i.t.) injection of these vectors was attempted but found to be highly ineffective, and led to the introduction of strategies in which vector-producing cells were inoculated into the tumor. Using this approach the antitumor effect of HSV-Tk + GCV was demonstrated in various intracerebral glioma models [9–12]. Other enzyme-prodrug strategies that demonstrated efficacy in animal models of malignant
186
glioma include CD with 5-FC [22] and cytochrome P450 with CPA [28–30]. Simultaneous treatment of glioma cells with two suicide paradigms HSVTk + GCV and CD + 5-FC, resulted in synergistic glioma cell kill compared to the single agents [38,124]. Interestingly, in subcutaneous (s.c.) glioma models, it was seen that immunocompromised mice were less responsive to HSV-Tk therapy than immunocompetent animals, suggesting a role for the immune system in the process of tumor eradication [97]. This finding initiated a great number of studies in various glioma models, assessing the effects of local pro-inflammatory cytokine (over) expression by tumor cells, alone or combined with HSV-Tk + GCV therapy, including IL-2, IL-4, TNFa and INFg [12,97,99,101,102,112,125]. Interestingly, s.c. injection of glioma cells transduced with genes encoding granulocyte-macrophage colony-stimulating factor (GM-CSF) and the T cell co-stimulatory molecule B7-2, markedly inhibited growth of wild-type tumors at distant sites [121]. These reports led to further investigations into the use of retroviral gene delivery in tumor vaccination studies for glioma. In immune-competent animals, s.c. injection of GM-CSF- or IL-4-transduced glioma cells induced a potent immune response to intracranial gliomas, both as a vaccination against subsequent intracranial glioma cell implantation, as well as for treatment of established tumors [103,118,119]. Alternative gene therapy strategies employing retroviral vectors and showing anti-tumor activity in glioma models include delivery of the tumor suppressor genes p16, p21, and p53 [41], delivery of inhibitors of angiogenesis, such as antisense or siRNAs to VEGF, angiostatin or platelet factor 4 cDNA [78,81,82,85], and the delivery of apoptosis inducing genes such as FADD, IL-1beta-converting enzyme, or siRNAs to cyclin E1 [75,76,126]. An overview of retrovirusbased gene therapy approaches for glioma is included in > Table 186-1.
3087
3088
186
Gene therapy for brain tumors
Improving Retroviral Gene Delivery to Glioma Cells Limitations to the use of MMLV-based retroviruses or retrovirus-producing cells include their inability to infect cells that do not express the retroviral receptor and their inability to infect non-dividing cells within the tumor population. The latter limitation can be circumvented by using vectors based on the retrovirus subclass lentivirus, which is capable of infecting both dividing and non-dividing cells. Lentiviral vectors based on the human immunodeficiency virus have been most broadly applied, including in glioma models [60,127]. Approaches to circumvent the limitation of retrovirus’ inability to infect cells that do not express the retroviral receptor have also been embarked upon. Retrovectors pseudotyped with proteins from other viruses, such as the vesicular stomatitis virus G protein, have demonstrated broader tropism or superior selectivity for glioma versus normal brain cells [128–130]. Moreover, modified retroviruses displaying a single-chain antibody fragment directed against the c-Met receptor, localized the virus at the glioma cell surface [131]. A different approach to improving retroviral gene transfer is the use of replication competent retroviruses (RCR). Contrary to other replicationcompetent viruses, replicating retroviruses do not kill the cell during replication. Instead these vectors have been employed to achieve long term transgene expression and drastically increase transduction efficiency in solid tumors [23]. RCR encoding CD demonstrated impressive anti-glioma activity without neurotoxic effects or signs of inflammation [24,25]. Taken together, during the nineties, retroviral vector gene delivery strategies for malignant glioma yielded such promising preclinical data that rapid translation to clinical phase I and II trials appeared warranted. However, seeing that poor gene transfer efficiency to tumor cells and
lack of tumor specificity are major impediments to therapeutic utility, the search for more efficacious and more tumor-selective gene delivery vehicles continued.
Adenoviral Vectors Adenoviral vectors are derived from adenoviruses which are non-enveloped DNA viruses that cause upper respiratory tract infections in humans. The wildtype adenoviral genome contains 36 kb of double-stranded DNA which encodes early (E) and late (L) genes. The E regions (E1-E4) of the adenovirus genome are expressed first, followed by expression of adenovirus proteins encoded by the L regions. Deletion of the E1 region renders the virus replication-deficient and provides space for insertion of transgenes. Adenoviral vectors offer the important advantages of their efficient infection of both dividing and non-dividing cells, and the feasibility of producing high titers, which facilitates delivery of large viral doses to tumors. In addition, extensive knowledge of their genome, permitting genetic modification to improve their safety and efficacy, has contributed to their forming the widest assortment of vectors developed thus far.
Adenoviral Gene Therapy Studies in Experimental Models of Malignant Glioma The first published adenoviral gene transfer studies for malignant glioma originate from 1994 [3,4]. Stereotactic delivery of a recombinant adenovirus encoding b-Galactosidase to intracerebral experimental gliomas demonstrated efficient infection of the tumor. These studies were soon followed by reports on impressive anti-tumor effects of adenoviral gene therapy of experimental glioma using enzyme/prodrug strategies including
Gene therapy for brain tumors
HSV-Tk + GCV, CD + 5-FC, cytochrome P450 2B1 + CPA, or combinations of the above [12,14–16,26,27,31,32,35,36]. Adenoviral delivery of various tumor suppressor genes and apoptosisinducing genes also demonstrated significant antitumor activity in experimental glioma models. The most widely studied is this field is the tumor suppressor protein p53 [42,43,45,46,132,133], but also Rb pathway regulators [49,50,52,54,77], as well as proteins involved in apoptosis induction [61,62,64,66,68,70] have demonstrated efficacy. Simultaneous delivery of various combinations of these molecules improved tumor kill compared to the individual gene transfer strategies [47,63,69,134,135]. As with retroviruses, adenoviral delivery of immunostimulating genes has also been described and shown effective in obtaining anti-glioma responses. IL-4, IL-12, IFNg and TNFa all induced significant inhibition of tumor growth and prolonged survival of tumorbearing rodents [100,104,106,107]. Impressive results were also obtained with the delivery of human soluble fms-like tyrosine kinase ligand (hsFlt3L). Treatment of animals bearing syngeneic gliomas with hsFlt3L led to the recruitment dendritic cells and to tumor regressions, even in a model for large established tumors [122,123]. Adenoviral vectors have also been engineered to deliver genes that inhibit tumor angiogenesis. Delivery of angiostatin inhibits tumor-associated angiogenesis resulting in increased apoptotic tumor cell death [78]. Examples of other molecules that demonstrated anti-angiogenic activity after adenoviral delivery in glioma models include a truncated form of the FGF receptor, anti-VEGF ribozymes, antisense constructs targeting proteolytic enzymes, and the brain-specific angiogenesis inhibitor BAI-1 [83,87,88,90,91]. A more complete overview of adenovirus-mediated gene therapy approaches applied in glioma research is provided in > Table 186-1. Although impressive anti-glioma activity has been reported using various adenoviral gene
186
transfer strategies, the ability of the adenovirus to disseminate beyond the local injection site and to transduce adjacent normal brain was shown to induce significant toxicity, in particular after HSV-Tk + GCV treatment [136,137]. These and other reports led to the search for strategies to redirect the adenoviral vector or its cytotoxic transgene expression specifically to the tumor cells.
Strategies to Achieve Tumor-Selective Gene Transfer to Glioma Cells A wide range of strategies have been pursued to improve glioma-selective gene transfer and which can be divided in two approaches: transductional targeting and transcriptional targeting. The first limitation for efficient adenoviral infection is the low levels of native adenovirus receptor CAR on their cell membrane [138,139]. To overcome this barrier, adenovirus can be retargeted towards alternative molecules highly expressed on the glioma cell membranes (transductional targeting). This concept was demonstrated using bispecific antibodies directed towards EGFR on one end and the adenovirus fiber knob on its other end which improved infection of primary glioma cell cultures obtained from tumor material [138,139]. Another successful approach involved the insertion of an integrin binding peptide, arg-gly-asp (RGD), into the fiber of the virus, allowing the virus to make its primary attachment to integrins [139,140]. Also, the coupling of receptor ligands to the fiber knob to redirect viral entry to highlyexpressed cognate receptors was demonstrated for basic fibroblast growth factor-2 (FGF2) and for IL-13 [141,142]. A very different targeting approach involves the substitution of Ad5 wildtype fibers, with the fiber shafts and knob domains of other adenoviral serotypes or even of non-human adenoviruses. Chimeric vectors based on human Ad3 or Ad35 serotype fibers as well as
3089
3090
186
Gene therapy for brain tumors
canine adenovirus type I fiber knobs have shown superior infectivity of primary glioma samples [143–145], although such capsid modifications do not appear to be the sole determinant of in vivo gene transfer [146]. For achieving true tumor selective infection it is essential to eliminate binding of the virus to non-target cells by abolishing the native tropism of adenovirus. Such ablated vectors have shown a dramatic reduction in infection of various organ tissues [147]. In vitro studies in which the EGFR targeting was combined with ablated adenoviral vectors exhibited up to 38-fold-improved tumorto-normal brain targeting index compared to native vectors [148]. The second approach to achieve tumorselective transgene delivery, involves incorporation of a tissue- or tumor-specific promoter (TSP) to drive the expression of the inserted transgene (transcriptional targeting). The tissue specific glial fibrillary acidic protein and nestin promoters were reported to efficiently drive transgenes in glioma whereas expression was nearly undetectable in non-glioma cells [149–152]. Tumorselectivity in relation to normal brain tissue, was demonstrated for the promoters of telomerase, of E2F-1, a transcription factor regulated by the Rb pathway, of survivin, a member of the inhibitor of apoptosis protein family, of CXCR4, a chemokine receptor involved in tumor cell migration, and of midkine, a growth factor implicated in the development and repair of various tissues [71,153–156]. Moreover, therapeutic effects of an adenoviral vectors encoding TNF-a or HSV-Tk under the control of radio/ chemo-inducible elements of the early growth response-1 gene promoter demonstrated impressive anti-glioma activity [17,108]. Although a vast amount of adenoviral vectors have been developed and found effective in preclinical glioma models, the vectors tested in clinical setting have thus far been limited to E1-deleted vectors carrying HSV-Tk or p53 transgenes (see section ‘‘Clinical gene therapy trials for GBM’’).
Herpes Simplex Viral Vectors Herpes simplex viral vectors are derived from the human herpes simplex virus 1 (HSV-1) strain, a member of the herpesvirus family, and can give rise to both a lytic or latent infection in humans. Its high infectivity, natural neurotropism and ability to infect both dividing and non-dividing cells, make HSV-1 an attractive gene therapy vector for treating tumors of the CNS. It carries a large, double-stranded DNA genome which encodes more than 80 genes of which half are non-essential for growth and production in tissue culture, but support the viral life cycle within the host cell. By removing these dispensable genes, multiple or large therapeutic transgenes can be accommodated. The encoded genes of the herpes simplex genome are classified as immediate early (IE/a), early (E/b) or late (Lg), based on their temporal pattern of expression during the lytic cycle of the virus. Neurotoxicity of HSV-1 represents a challenge to the development of these virus-based vectors, since the wild-type virus is capable of propagating in neurons and glia, resulting in necrotizing and potentially life-threatening encephalitis. Nevertheless, attempts have been made to construct gene transfer vehicles based on herpes simplex virus. Replication-deficient HSV-1 vectors have been constructed by deletion of one or other of the essential IE genes termed ICP4 and ICP27 [157–160]. These gene products are essential for viral replication, and need to be provided in trans for in vitro virus production. Transduction with such replication-defective vectors causes a latentlike infection in both neural and non-neural tissue. The vectors may be further refined to prevent vector-associated cytotoxicity by deleting additional nonessential IE genes such as ICP0, ICP22, ICP47 [161,162]. Another type of HSV-1 based vector system is known as the HSV amplicon. These gene transfer vehicles are replication-incompetent bacterial plasmids that contain specific HSV DNA
Gene therapy for brain tumors
sequences, allowing them to be replicated and packaged into viral particles by supplying almost the entire viral genome in trans [163–165]. Both types of replication-incompetent herpes virusbased viral vectors have been applied in glioma research, although their use in treating nonmalignant diseases of the peripheral and central nervous system has received much more attention by virtue of their neurotropism and longterm gene expression profiles. Conversely, herpes vectors with tumor-selective replication capacity have been applied more extensively in neurooncology research (see paragraph ‘‘Oncolytic herpes simplex viruses’’).
HSV-1 Vector-Based Gene Therapy Studies in Experimental Models of Malignant Glioma The unique property of herpes vectors to accommodate multiple transgenes led to the development of vectors combining various gene therapy approaches by simultaneous delivery of multiple genes designed to induce toxicity to tumor cells. Niranjan et al. demonstrated effective treatment of experimental glioblastoma by HSV vectormediated TNFa and HSV-tk gene transfer in combination with irradiation and GCV administration [18]. Addition of the gene encoding connexin 43, the major component of gap junctions between cells facilitating intercellular spread of activated GCV, further enhanced the antiglioma activity of this vector [40]. A replication-defective HSV vector expressing both HSV-Tk and a mutant form of a NF-kappaB inhibitor also showed improved antiglioma activity compared to single agent controls [166]. Successful gene therapy experiments using herpes-based amplicons have also been described. Intratumoral injection of amplicons encoding the murine IFNg gene caused tumor necrosis while ex vivo IFNg or GM-CSF gene transfer to the cells prior to implantation had the effect of tumor
186
vaccination [113,120]. Amplicon-mediated delivery of double-stranded hairpin RNA directed against EGFR resulted in the growth inhibition of human glioblastoma in vivo [56]. A transcriptional targeting approach for amplicons was reported using cyclin A promoter elements and the GFAP enhancer, resulting in a glioma-specific and cell cycle-regulated HSV-I amplicon. Incorporation of apoptosis-inducing genes into this vector resulted in the induction of apoptosis in target glioma cells both in vitro and in vivo [67,167]. One of difficulties in application of this technology to human clinical trials has been the inability to produce high-titer of virus as well as the incapacity to obtain, helper virus-free batches of replication-defective herpes vectors. While the former challenge still exists, several methods are now in use which have effectively eliminated the contamination by helper virus [168].
Adeno-Associated Viral Vectors Adeno-associated viruses are small singlestranded DNA viruses of the parvovirus family and non-pathogenic for humans. AAVs are nonautonomous, requiring another virus to replicate such as adenovirus or herpes simplex virus. They are capable of efficiently infecting a wide variety of both dividing and non-dividing cells, and can integrate into the host cell genome. The small size of AAV particles has given them the advantage of improved penetration capacity in solid tumor structures, including those of malignant glioma [6]. Moreover, delivery of AAV particles by convection-enhanced delivery into rat brain revealed a clear dose response with a volume of 300 mm3 of brain tissue partially transduced in the high-dose group [169]. Although the use of AAV-vectors, predominantly serotype-2, in glioma gene therapy studies is still at the level of preclinical evaluation, promising results using various transgenes have been reported.
3091
3092
186
Gene therapy for brain tumors
Okada et al. reported that the AAV-mediated delivery of the HSV-Tk and IL-2 genes with GCV treatment led to a 35-fold reduction in intracerebral tumor growth without inducing to normal brain toxicity [5]. Complete regression of intracerebral glioma could be obtained when multiple injections of AAV-Tk followed by GCV treatment were given [19]. Similarly, 6 injections of AAV encoding the interferon-b gene were required to achieve complete growth inhibition of orthotopic gliomas in mice [114]. Systemic delivery of this vector was also capable of inducing regression of s.c. gliomas [115]. Anti-glioma strategies directed at inhibiting angiogenesis using AAV vectors include delivery of tissue factor protein inhibitor-2, angiostatin, and the soluble VEGF receptor. All three approaches led to significant inhibition of tumor vascular structures and tumor growth [79,86,170]. Apart from AAV-2, other AAV serotypes have been shown to represent highly efficient gene transfer vehicles to glioma cells including AAV-1, AAV-4, AAV-5, AAV-6, AAV-7, and AAV-8, the latter three showing superior infection efficiency to AAV-2 [84,171,172]. However, the serotypes AAV-7, AAV-8 and AAV-5 also demonstrated enhanced transduction efficiency in the murine CNS, revealing a marked tropism for neuronal cells [84]. These studies suggest that tumor-selective targeting strategies are also a necessity in AAV vector-mediated gene therapy studies for glioblastoma. Reports on targeting strategies to restrict AAV-mediated transgene expression to tumor cells is thus far limited to one study describing hypoxia-induced transgene expression in malignant glioma cells using hypoxia-responsive elements from the erythropoietin gene [173].
Non-viral Vectors The DNA of choice can be also delivered to the cell nucleus by using non-viral methods.
The simplest method to deliver DNA is by plasmid DNA expression vectors driven by eukaryotic promoters. However, the efficiency of gene delivery by this approach is very poor. To increase the transfection efficiency, the DNA can be associated with polycationic polymers and liposomes. Liposomes typically enter the cell cytoplasm by endocytosis or transient membrane disruption and have been shown to be successful in transfecting glioma cell lines [7]. Liposomemediated intratumoral delivery of HSV-Tk or p53 inhibited the growth of established gliomas [20,21,48]. Likewise, antisense strategies directed at glial cell derived neurotrophic factor or EGFR [57,58,174] both demonstrated effective inhibition of glioma growth. In comparison to viral vectors, non-viral gene transfer approaches are associated with relatively low toxicity and immunogenicity [21,175], which offers the important advantage of being able to deliver these therapeutics systemically. This was demonstrated by Barnett et al. by intra-arterial delivery of a plasmid encoding the anti-angiogenic endostatin gene resulting in significant inhibition of intracerebral tumor vascularization and growth in rats [80]. Intravenous liposomal delivery of short interfering RNAs aimed at EGFR [59] also showed anti-glioma activity. The RNA interference technology has also been employed to inhibit members of the plasminogen activator and matrixmetalloprotease families of proteolytic enzymes, known to play a role in invasion, apoptosis and angiogenesis. Plasmids encoding these inhibitors were delivered both locally and intraperitoneally and induced impressive anti-tumor activity in intracerebral glioma-bearing mice [92–95]. Because gene transfer by non-viral vectors is relatively inefficient, various approaches have been undertaken to enhance transduction efficiency such as incorporation of fusion proteins or viral capsid proteins. Hemagglutinating virus of Japan (HVJ)-liposomes are coated by the Sendai virus envelope protein and have high potential
Gene therapy for brain tumors
to deliver foreign genes to glioma cells [176]. More selective targeting of liposomes to glioma cells has also been demonstrated by coupling monoclonal antibodies to them. Such so called immunoliposomes directed against CD44, or the truncated deletion mutant of EGFR (EGFRVIII), induced markedly higher transgene levels compared to the untargeted DNA/liposome [177]. Similarly, immunoliposomes targeted to the transferrin receptor or the insulin receptor [178,179] have also been applied. Cationic liposomes have also been directly associated with the transferrin protein as a targeting moiety and demonstrate reduced toxicity and enhanced specific gene activity [180]. Alternatively, transcriptional targeting has been illustrated in the context of non-viral vectors by placing the transgene under control of the multi-drug resistance gene promoter [181]. Although non-viral vector technology has significantly improved during recent years, the gene transfer efficiency is still an order of magnitude lower than most virus-based delivery systems. This significant draw-back delayed its development toward clinical application.
Clinical Gene Therapy Trials for GBM The earliest application of gene therapy to human glioma therapy involved the use of retroviral vector delivered at the time of surgery for a recurrent malignant glioma. Because of the concerns that retroviral vector stability could be impaired in the tumor milieu, trials were conducted by direct intratumoral injection of the murine packaging cell line (vector producer cell or VPC) into the tumor with the hope that release of progeny retroviruses would occur into the tumor. The approach underwent a phase I, phase II and phase III test where the HSV Tk/ganciclovir approach was attempted [182]. Unfortunately, the study did not show evidence of efficacy, providing a sobering dose of reality regarding the need for
186
further laboratory studies to optimize the technique. Other small gene therapy trials using VPCs to deliver HSV-Tk, or IL2 + HSV-Tk were reported and while the technique appeared safe, for the most part, evidence of efficacy was not forthcoming [183–191]. One likely reason for the absence of evidence of efficacy relates to the amount of gene that was actually transferred into the tumor. Studies by Ram [192] showed by in situ hydridization HSV-Tk mRNA to be located immediately within the confines of injected vector producer cells. An immunohistochemical study by Harsh et al. [193], went on to show that the number of cells expressing HSV-Tk and the amount of converted ganciclovir within tumor was minimal. The overall conclusion from this work was that injected VPCs did not efficiently transduce surrounding tumor cells once injected into the human neoplastic mass. An additional confirmation of this result came from the work of Puumalainen et al. [194]. They were able to compare delivery of a reporter gene (the E. coli lacZ gene) in humans whose tumor cavity was injected with VPCs that produced a retroviral vector versus an adenoviral vector. They found that the latter was able to transduce much larger volume of tumor compared to the former, setting up the stage for studies where adenoviral vector were employed to deliver HSV tk (Ad-Tk) Several such studies have now been published [195,196]. In a phase I, dose-escalation of stereotactic administration of Ad-tk into recurrent malignant gliomas, Trask et al. [187] were actually able for the first time ever to determine a maximum tolerated dose (MTD). Two patients treated at a dose of 2 1012 vector particles (vp), suffered dose-limiting toxicities and thus the authors recommended the lower dose (2 1011 vp) as the appropriate dose for the next phase of their trial. Sandmair et al. [197] compared the overall median time to survival between patients treated with Ad-Tk and historical controls of patients treated with the VPCs
3093
3094
186
Gene therapy for brain tumors
that delivered retrovirus-Tk. The former was over 15 months while the latter was about 7 months. The same group then published a phase II randomized trial in patients with newly diagnosed and recurrent malignant gliomas where they were randomized to receive Ad-Tk into the cavity of the tumor after resection or no gene therapy. The newly diagnosed group still was treated with standard radiation. The authors reported an overall significant increase in survival for the gene therapy group. This data was confounded by the inclusion of newly diagnosed with recurrent glioblastoma as well as with anaplastic astrocytomas. However, when results were stratified to only look at newly diagnosed glioblastoma, mean survival was reported to be still significantly different for the gene therapy group (p = 0.04). While this was encouraging, more meaningful than overall survival (which could be affected by additional therapies, such as more surgery, provided to patients after recurrence) would have been determination of progression-free survival for the newly diagnosed glioblastoma group treated with gene therapy compared to control [198]. Nevertheless, these results were considered significant enough to warrant a large phase III multicenter study in Europe to compare Ad-Tk + ganciclovir after resection of newly diagnosed malignant glioma followed by standard radiation and temozolomide to standard radiation and temozolomide alone[199]. HSV–Tk has not been the only transgene used for gene therapy in gliomas. Knowledge related to the action of the p53 tumor suppressor protein in producing apoptosis in p53-deficient gliomas or growth inhibition in p53-expressing tumors provided a basis for a phase I clinical trial in recurrent malignant gliomas [200]. Although this trial was safe and, in fact, did not reach a MTD, the authors were able to obtain data related to the extent of gene transfer from injected tumors by harvesting these a few days after stereotactic injections. They reported that p53 gene transfer occurred primarily in direct
vicinity (within a millimeter) of the needle, thus questioning efficiency of biodistribution of this modality. More recently, a phase I multi-institutional dose-escalation trial of stereotactic intratumoral injection of an adenoviral vector expressing interferon-beta, followed a few days later by resection of injected tumor and additional vector injection into cavity was reported. In this trial, dose-escalation also appeared to correlate with increased apoptosis and increased inflammation of injected tumor tissue, providing evidence of a pharmacologic effect by the gene transfer procedure (Chiocca et al., in press) [201]. One patient at the highest dose level did suffer a severe doselimiting toxicity, thus causing the trial to close. However, dose-related correlative apoptosis and inflammation in resected specimens did provide suggestive evidence that the gene transfer procedure was actually producing a phenotypic alteration in injected tumor. There have been other gene therapy trials using liposomes to deliver HSV-Tk (by CED) [202] or IFN-b [203]. In addition, imaging of gene transfer was also reported by Jacobs et al. [204] However, it is evident that the gene transfer approach will become effective in humans only when the issue of intratumoral biodistribution will be resolved. More detailed overviews on clinical gene therapy trials for malignant glioma were recently published [205,206].
Preclinical Oncolytic Virus Therapy for Glioma The disappointing results of the glioma gene therapy trials and the growing awareness of the crucial factors responsible for their failure, i.e., the limited spread of the vector within the tumor mass and the inefficient infection of the individual glioma cells, contributed to a new concept in cancer therapy: viral oncolysis, or virotherapy, using replication-competent viruses. This approach
Gene therapy for brain tumors
utilizes the inherent cytopathic effects which result from the normal life cycle of the virus. Once inside the cell, the virus expresses proteins that interfere with cellular processes e.g., to prevent apoptosis or induce cell-cycle progression. This ensures viral replication and viral protein production, which eventually cause lysis of the infected cell and release of progeny virus that can infect neighboring cells, upon which the cycle is repeated. In this way, the virus is capable of disseminating into a solid tumor mass. The terminology ‘‘oncolytic virus’’ refers to viruses which preferentially replicate in tumor cells as compared to normal cells. Some naturally attenuated oncolytic viral strains exist that appear to replicate more efficiently in cancer cells than in normal tissue (such as reovirus), whereas others acquire oncolytic qualities through genetic engineering (such as herpes simplex virus and adenovirus). The tumor selectivity of oncolytic viruses occurs either during infection or replication. Oncolytic viral vectors based on herpes virus, adenovirus, Newcastle disease virus, and reovirus have already been incorporated in phase I/II clinical studies for malignant glioma, whereas various other DNA and RNA viruses are showing impressive therapeutic efficacy in preclinical setting.
Oncolytic Herpes Simplex Viruses The first oncolytic viruses to be developed, already in the early nineties, were based on the herpes simplex virus. By now these vectors are furthest along in their development and testing for virotherapy. HSV-1 vectors were designed to target cells which have altered signal-transduction pathways that promote tumorigenesis. Mutations or deletions in the viral genome were introduced torender the vector dependent on the tumor cells for viral propagation and hampering their ability to replicate in quiescent cells; a general approach later applied to various other virus strains as
186
well. Different single- or multiple-mutated replication-conditional HSV-1 vectors have been created. The first attenuated HSV-1 vectors were designed by deleting the viral TK gene, making viral replication dependent on the presence of cellular TK [207,208]. These replication conditional HSV-1 vectors prolonged survival in animal brain tumor models. However, neurotoxicity was retained and combined with the loss of susceptibility to the anti-viral agents acyclovir or gancyclovir, these mutants were unsuitable for clinical application. In the next generation, HSV-1 vectors with intact Tk genes but harboring single mutations in the ribonucleotide reductase (RR)/ICP6 gene or the neurotoxicity gene g34.5 were designed, which exhibited therapeutic efficacy and decreased neurovirulence in animal glioma models [209–213]. The g34.5-deleted mutant (designated 1716) has progressed to clinical trial evaluation (see paragraph ‘‘Oncolytic virus trials for GBM’’). The use of single mutant HSV-1 vectors are potentially associated with the risk of restoring a wild-type phenotype [214], and concern about this risk led to the design of vectors with dual mutations. The most extensively studied dual mutated virus, known as G207, is an HSV-1 derivative with deletions of both the g34.5 gene and the RR gene [215]. G207 demonstrated effective anti-tumor activity in various models of malignant brain tumors [215–217] and, moreover, elicited a systemic antitumor immune response causing regression of remote tumor nodules in the brain and providing protection against tumor rechallenge [218]. Safety of G207 injection in the brains of mice and primates was also demonstrated [219,220] and rapidly followed by translation to clinical trial (see paragraph ‘‘Oncolytic virus trials for GBM’’). Neurotoxicity of HSV-1 vectors represents a concern since the wild-type virus is capable of propagating in neurons and glia, resulting in necrotizing and potentially fatal encephalitis.
3095
3096
186
Gene therapy for brain tumors
Although anti-herpetic agents exists which can terminate uncontrolled viral replication [221], approaches to further improve the tumor selectivity of these agents have been undertaken.
Improving Selectivity of Oncolytic Herpes Viruses Strategies to further restrict replication of herpes vector to the target cells have been developed by linking viral gene transcription to promoters being expressed only in dividing or malignant cells. As deletion of the g34.5 gene markedly reduces cytotoxicity mediated by HSV-1, Chung et al. reintroduced this gene in the G207 backbone under the control of the promoter of B-Myb, a cell cycle regulatory protein which is overexpressed in cancer cells with mutations in the p16/Rb pathway. Virulence was targeted to the tumor cells while the neuroattenuated phenotype of the virus was maintained [222]. Similarly, transcriptional regulation of the viral g34.5 gene by the promoters of nestin, an intermediate filament expressed in glioma and not in astrocytes, and Musashi-1, a neural RNA binding protein expressed by a variety of tumors, demonstrated enhanced therapeutic efficacy compared to the parental vector G207 [223–225]. As many intracerebral tumors are glial and predominantly astrocytic in origin, McKie et al. confirmed the ability of the g34.5-deleted mutant 1716 to deliver a reporter gene specifically to astrocytes using the glial fibrillary acidic protein (GFAP) promoter [226].
Improving the Potency of Oncolytic Herpes Viruses As the results of the first oncolytic virus trials became available (see paragraph ‘‘Oncolytic virus trials for GBM’’), it also became evident
that while attenuated vectors demonstrated safety in humans, they had concomitantly lost a great part of their cytotoxic activity. Therefore researchers turned their focus on potentiating the viral anti-tumor effect, whilst retaining tumor selectivity. One approach to achieving this is by incorporating therapeutic transgenes that had proven their value in gene therapy research in the context of replication-defective vectors. The feasibility of this approach was confirmed by the addition of ganciclovir to treatment with oncolytic HSV mutants that already express HSV-Tk, which enhanced the regression of syngeneic rat gliosarcomas [209]. Other enzyme-prodrug strategies that have been shown to improve HSV-1induced oncolysis in the context of glioma include the p450 gene with CPA or the CD gene with 5-FC [227,228]. Double enzyme approaches such as p450 and CE in combination with CPA and irinotecan, or the combination of HSV-TK and p450 gene transfer with GCV and CPA treatment, provided anti-tumor effects that were more significant than control treatment combinations [229,230]. Other transgenes demonstrating enhanced efficacy of oncolytic HSV treatment of glioma include vectors encoding a dominant negative fibroblast growth factor or platelet factor 4, by simultaneously inhibiting tumor growth and angiogenesis [231,232]. The involvement of the host immune response in the anti-tumor effect of HSV-mediated virotherapy has also been investigated and exploited to enhance therapeutic efficacy. Oncolytic HSV mutants expressing IL-4, IL-12, IL-2, or the soluble B7-1 immunomodulatory molecule were found to produce survival benefit compared to control viruses in various tumor models including glioma and neuroblastoma in immunocompetent mice [233–237] Moreover, insertion of the potent immune stimulator GM-CSF improved shrinkage or clearance of tumors and protected against rechallenge with tumor cells [238], suggesting not only that expression of
Gene therapy for brain tumors
immunomodulatory molecules can potentiate oncolysis but may also induce a level of antitumor immunity. Interestingly, the anti-viral immune response has also been shown to contribute to rapid elimination of the virus thereby counteracting the therapeutic effect. For this reason, several strategies were developed to circumvent early inactivation of the virus by the immune system. The non-specific early occurring immune response contributes the largest effect to elimination of the viruses and includes the activation of the complement cascade and the recruitment and activation of immune cells which kill the infected cells [239,240]. Combined treatment of oncolytic HSV with cobra venom factor, a potent complement inhibitor, and the immunosuppressive agent cyclophosphamide (CPA), inhibited both the innate and anti-HSV neutralizing antibody response, and their concerted action prolonged survival of rodents bearing intracerebral tumors [241,242]. Fulci et al. observed that the effect of CPA treatment involves reduced infiltration of tumor-associated phagocytic cells into the virusinfected tumor [243]. Whether the immune response to virus is more helpful or harmful to the therapeutic outcome of oncolytic virus therapy requires additional investigations into the underlying mechanisms and especially into the precise time course of events. It is not inconceivable that initial immune suppression to allow viral infection and propagation to take place, should be followed by immune boosting to evoke an optimal vaccination response.
Oncolytic Adenoviruses Two approaches have been described to render adenovirus propagation selective for tumor cells: 1) deletions or mutations in viral genes essential for replication in non-malignant cells, 2) insertion of tumor-specific promoters driving viral replication. As with tumor-selective herpes viruses, the
186
first replication-conditional adenoviruses that were constructed targeted signal transduction pathways operating in tumor cells. The first oncolytic adenovirus constructed in this manner, termed dl1520 or ONYX-015, lacked the E1B p53-binding viral protein, enabling selective replication of this virus in cells with a disrupted p53 pathway [244], although this mechanism of action was later shown to be untrue and to depend on differential export of viral RNA from nucleus [245]. This oncolytic adenovirus proved to be very effective in a wide range of preclinical models for various tumor types, including malignant glioma, with reports on complete regression of xenograft tumors and long-term survival [244,246–248]. On a similar theoretical basis, adenoviruses have been engineered to replicate selectively in tumor cells with lesions in the retinoblastoma tumor repressor (pRb) pathway. This was accomplished by deleting 24 bp from the E1A gene which abolishes the pRb binding capacity of the E1A protein [249,250]. This adenovirus mutant, known as AdD24 or Addl 922–947, demonstrated reduced replication in non-proliferating normal cells relative to other gene-deleted or wild-type adenoviruses and demonstrated potent anti-glioma activity in vivo [249,251]. Strict tumor-selectivity of E1A and E1B mutants has been challenged [252–254] and additional mutations may be needed to enhance selectivity. Indeed, E1A-E1B double mutants have been developed and demonstrate improved therapeutic profiles as compared to single mutants in various tumor models including glioma [255–257]. The second approach to constructing conditionally replication-competent adenoviruses is to place the adenovirus E1A region under the control of tissue/tumor-specific promoters. The rationale behind these vectors is that expression of the early viral gene E1A, and therefore the whole adenoviral transcription program, will depend on the activity of these TSPs. In the context of glioma, tumor selectivity of various TSPs had
3097
3098
186
Gene therapy for brain tumors
already been demonstrated using replicationdeficient adenoviral vectors (see section ‘‘Strategies to achieve tumor-selective gene transfer to glioma cells’’) and these approaches were subsequently incorporated into the replicationcompetent setting. Adenoviruses with E1A driven by the midkine, survivin, and human telomerase reverse transcriptase promoters have displayed glioma-specific lysis and in vivo therapeutic efficacy [154,258,259]. Likewise, a hypoxia dependent replicative adenovirus using the hypoxiainducible factor-1 (HIF-1) promoter induced cytolysis in a hypoxia-dependent manner and effective anti-glioma efficacy in combination with chemotherapy [260,261]. A combination of the two approaches for attaining tumor selectivity was recently described by Alonso et al. who constructed a vector termed ICOVIR-5 encompassing both the deletion in the E1A Rb protein-binding region and the substitution of the E1A promoter for E2F-responsive elements [262]. ICOVIR-5 efficient lysed cancer cells and inhibited glioma growth but replication was dramatically impaired in cells with restored Rb function.
Improving Selectivity of Oncolytic Adenoviruses Transductional targeting of oncolytic adenoviruses by modification of viral tropism can further improve the specificity and/or efficiency of these agents. Various targeting approaches used in adenoviral gene therapy studies have been translated to oncolytic adenoviruses. To employ bispecific antibodies for adenovirus retargeting, the expression cassette encoding the bispecific antibody is inserted into the genome of the adenovirus, thereby allowing the progeny viruses to become retargeted. With this technique, EGFRtargeted AdD24 efficiently killed primary human CAR-deficient brain tumor specimens that were refractory to the parent control virus [263]. With
the incorporation of the integrin-binding peptide RGD into the fiber knob of the AdD24 virus, termed Delta24-RGD [264], it was demonstrated that improved infection translates to markedly enhanced oncolysis in (primary) glioma cells [265,266]. The highly potent anti-glioma activity of Delta24-RGD in s.c. and intracerebral xenografts as well as its capacity to destroy glioma stem cells is now well-established [265–269]. The impressive anti-tumor activity of Delta24-RGD has resulted in further development of this vector towards clinical application and it is scheduled to enter phase I testing for (recurrent) glioma in 2008 (Charles Conrad, personal communication). Retargeting of E1B mutant adenoviruses to glioma cells has also been studied. An increased cytopathic effect on glioma cells and in s.c. xenografts was found when the ONYX-015 fiber was modified with a stretch of 20 lysine residues [270]. Targeting of the survivin promoter-driven oncolytic adenovirus to heparan sulfate proteoglycans provided minimal viral replication and toxicity in normal brain explants and effective anti-glioma efficacy in vivo [155]. Other effective targeting approaches applied on this viral backbone, include fiber replacement with serotype 3 fibers and RGD insertion in the fiber knob [156,271].
Improving Potency of Oncolytic Adenoviruses Based on the results of the first clinical trials of ONYX-015, which underscored the safety of this agent, investigators pursued approaches to bolster the therapeutic potential of these agents. One such approach involves the incorporation of therapeutic genes into the oncolytic virus genome. A large and still expanding collection of such ‘‘armed therapeutic viruses’’ has been constructed, although their application in glioma research is still quite limited. The first armed adenoviruses constructed incorporated the HSV-Tk/GCV
Gene therapy for brain tumors
enzyme/prodrug strategy. This resulted in a striking improvement of treatment efficacy in various human cancer models, including human glioma xenografts [272–274]. Likewise, Conrad et al. inserted the CD gene in the AdD24 backbone. Combined treatment with systemic 5-FC significantly improved survival of mice bearing intracranial gliomas compared to controls [275]. A different approach using the same adenoviral backbone involved the insertion of an expression cassette for tissue inhibitor of metalloproteinase-3, which is reported to inhibit angiogenesis and tumor cell infiltration and induce apoptosis [276]. The above-described strategies were aimed at inducing a therapeutic bystander effect of the transgene in conjunction with the lytic actions of the virus. However, insertion of specific transgenes can improve the oncolytic potential of the virus in the infected cell itself. Oncolysis of cancer cells may be suboptimal due to cancer cell specific genetic alterations. Interference in these processes concomitant with the viral replicative cycle was hypothesized to affect the anti-tumor potential of these agents [277]. This concept was confirmed using AdD24 engineered to express p53 during late stages of viral replication and which exhibited up to >100-fold enhanced oncolytic potency on human cancer cell lines of various tissue origins [278]. AdD24-p53 also caused more frequent regression and more delayed growth of GBM xenografts in vivo [279]. Similarly, expression of the apoptosis-inducing TRAIL by an oncolytic adenovirus killed GBM cells more efficiently than control virus [280]. Another approach is to insert transgenes which boost the immune response to the infected tumor cells by stimulating localized inflammatory and/or immune responses. The efficacy of this approach was established for a variety of cytokines and immune stimulatory molecules in multiple tumor types. For glioma, only one such armed oncolytic adenovirus has been tested thus far, an IL-4-expressing HIF-1 promoter-driven oncolytic adenovirus, which induced hypoxia-dependent
186
IL-4 expression and viral replication, and maintained tumor regression [281]. Of note, as with oncolytic HSV vectors, suppression of the immune response using CPA prior to adenovirus inoculation resulted in prolonged intratumoral viral replication and concomitant transgene expression after local injection in glioma-bearing mice [282]. This indicates that immune-boosting approaches for oncolytic adenoviruses may potentially be further improved by initial immune suppression. Finally, improvement of oncolytic adenoviral therapy for malignant glioma has been endeavoured by combining with conventional therapies. Combinations of various oncolytic adenoviruses with radiation therapy, irinotecan and temozolomide have been described [248,265,267,283–286].
Other Oncolytic DNA Viruses Although HSV-1 and adenoviruses account for the great majority of DNA viruses being developed as novel viral therapeutic anti-cancer agents, other DNA viruses are also under investigation and showing promising results. Vaccinia virus (VV), a double stranded DNA virus, is a member of the poxviridae family and has been used as a live vaccine in the smallpox vaccination program. It can infect a wide range of human tissues but does not cause any known human disease. VV appears a promising oncolytic agent due to its rapid lifecycle, its efficient cell-to-cell spreading and its wide range of immune evasion strategies. Moreover, it possesses a strong natural tumor tropism. After i.v. injection in tumor-bearing mice, the greater part of the viral load is found in the tumor [287,288]. To enhance the tumor specificity of this virus, various mutants have been developed. A Tk-deleted VV mutant demonstrated tumor selectivity over other tissues after both i.v. and i.p. injection [289,290]. Another approach makes use of the ability of cancer cells to evade apoptosis.
3099
3100
186
Gene therapy for brain tumors
Mutant VV lacking anti-apoptotic genes would therefore survive better in malignant compared to normal cells. A vaccinia mutant deleted of the anti-apoptosis genes SPI-1 and SPI-2, demonstrated enhanced tumor selectivity and improved safety [291]. Armed VV have also been developed encoding CD, GM-CSF, and p53 [292–294]. The latter has been extensively studied in preclinical models of malignant glioma [294–297]. Another DNA virus, myxoma virus, a poxvirus previously considered rabbit specific, has been shown to replicate productively in a variety of human tumor cells [298] and is being studied in preclinical models of glioma and medulloblastoma [299,300]. Lun et al. confirmed that a single intratumoral injection of myxoma virus dramatically prolonged survival of glioma-bearing mice and induced [299]. The last of the DNA viruses to be discussed here, the parvoviruses, are small, icosahedral viruses containing a single strand DNA genome of approximately 5 kb. Several members of the rodent parvovirus group, including parvovirus H1 of rats and the minute virus of mice (MVM) demonstrate tropism for and oncolytic activity in human transformed cells [301,302], including glioma cell lines and low-passage cultures of patient-derived glioblastomas [303,304] In addition, H-1 virus is capable of inducing apoptosis in TRAIL- and cisplatin-resistant glioma cells in vitro and in vivo [305]. These advantageous characteristics of parvoviruses, combined with their reported low toxicity in normal astrocytes and microglial cells [306], make them promising candidates for glioma oncolytic therapy.
Oncolytic RNA Viruses Reoviruses Oncolytic RNA viruses are gaining interest as oncolytic agents. Reoviruses are naturally occurring, cytoplasmically replicating viruses found
in many species including humans and are considered benign. While attenuated in healthy tissue, reoviruses are inherently oncolytic and demonstrate tumor specificity. First investigations into the mechanism behind this selectivity revealed that reovirus replicates preferentially in the presence of an active Ras signaling pathway, a common characteristic of cancer cells, including malignant glioma cells [307]. Indeed, impressive effects of reovirus therapy in glioma cell cultures and xenografts in mice have been reported [308,309]. Yang et al. confirmed the anti-glioma efficacy of reovirus in rodent glioma and medulloblastoma models, and, importantly, showed that inoculation of reovirus into the brains of nonhuman primates did not produce significant toxicities [310,311]. The lack of toxicity and impressive efficacy of this virus is generating excitement for this novel potential therapeutic for neurological tumors and contributed to the design of a phase I trial for glioma (see section ‘‘Oncolytic virus trials for GBM’’).
Newcastle Disease Virus The avian Newcastle disease virus (NDV) is a non human pathogen. This RNA virus is a member of the Paramyxoviridae family and is closely related to the infectious agent of human mumps. Although non-pathogenic to humans, the virus has been found to harbor effective replication capacity in human malignant cells. This natural tumor selectivity is thought to reside in the inability of tumor cells to respond to type I interferons and mount an anti-viral response [312] Virus-mediated killing of tumor cell cultures, including glioma, was shown to occur by p53independent apoptotic cell death [313]. Preclinical studies evaluating the oncolytic potential of these agents demonstrated highly effective tumor regression in xenograft models of fibrosarcoma and neuroblastoma after local or systemic delivery [314,315]. Although virtually no preclinical
Gene therapy for brain tumors
data using NDV in glioma models has been published thus far, a number of small-scale clinical trials in GBM patients have been described (see section ‘‘Oncolytic virus trials for GBM’’).
186
so-called replicons, poliovirus-based vectors which are genetically incapable of producing infectious virus but which do demonstrate direct oncolytic effects on tumor cells and in vivo anti-tumor activity in mice bearing glioma xenografts [330].
Measles Virus Measles virus is an enveloped, negative-stranded RNA virus of the Paramyxoviridae family. Its use as oncolytic agent was inspired by an early case report describing the spontaneous regression of Burkitt’s lymphoma in a patient experiencing a natural measles infection [316]. Both the parental virus and the recombinant vaccine (Edmonston B) strain have shown anti-tumor efficacy in xenografts of various tumor types [317–319]. Malignant brain tumor cells, including astrocytoma and oligodendroglioma were also shown to be susceptible to measles virus infection [320–322] and a recombinant vector based on the Edmonston strain caused marked regression of intracerebral U-87 xenografts [323]. Targeting strategies to improve selectivity of measles virus have been described, including EGFR or EGFRVIII targeting in glioma models [324,325], as well as novel delivery techniques such as the use of endothelial cells as vehicles to deliver measles virus to orthotopic gliomas in mice [326].
Poliovirus Poliovirus is a nonenveloped positive-strand RNA virus belonging to Picornaviridae and is the causative agent of paralytic poliomyelitis. Exchange of the poliovirus internal ribosomal entry site with its counterpart from human rhinovirus type 2 results in attenuation of neurovirulence. This attenuated virus possesses potent oncolytic activity in cell lines derived from malignant gliomas and led to tumor elimination in glioma models [327–329]. Another poliovirus-based approach was described using
Other RNA Viruses A number of less well-known RNA viruses are under investigation as novel oncolytic agents. Vesicular stomatitis virus (VSV) is an essentially nonpathogenic negative-stranded RNA virus, the replication of which is extremely sensitive to the antiviral effects of IFN. As IFN signaling is often defective in malignant cells, VSV shows replication specificity for neoplasms inducing apoptosis-mediated cell death [331]. VSV also caused significant inhibition of tumor growth when administered to mice bearing subcutaneous gliomas [332]. Tumor selectivity was improved by pretreatment with IFNb and inoculation with a replication-restricted vector with its glycoprotein gene deleted [333]. I.t. or systemic delivery of the attenuated VSV strain deltaM51 produced marked regression of intracerebral gliomas prolonging survival of mice with unilateral or bilateral gliomas [334]. Another less well-known RNA virus being investigated is the Sendai virus, also known as haemagluttinating virus of Japan, a murine parainfluenza virus. An engineered derivative was shown to infect and kill several different cancer cell types in vitro and human fibrosarcomas and colorectal adenocarcinomas in xenograft models [335]. Published studies on oncolytic effects of Sendai virus in preclinical glioma models have yet to be reported. One more virus which was reported to obtain oncolytic potency is the Sindbis virus, a positive strand RNA virus that belongs to the Togaviridae family. In humans, Sindbis virus infection is considered to induce none or mild symptoms.
3101
3102
186
Gene therapy for brain tumors
Sindbis infection has been shown to induce tumor-specific cytopathic effects and apoptosis in cervical and ovarian cancer cells in vitro and in vivo [336]. Interestingly, in a report by Wollmann et al. describing a comparative study of nine oncolytic viruses in glioma models, Sindbis virus and Vesicular Stomatitis virus showed the strongest cytolytic actions and highest rates of replication and spread in glioblastoma cells, as well as in vivo efficacy in glioma xenografts [303].
Oncolytic Virus Trials for GBM The inefficiency of gene transfer observed with gene therapy vectors such as replication-defective retroviruses and adenoviruses have led to an interest in the use of replication-conditional (oncolytic) viruses to improve biodistribution of these agents in tumors [205,206]. Two oncolytic HSV1s were employed in two different trials in patients with recurrent malignant gliomas. As discussed above, G207 is a recombinant HSV1 where both of the viral ICP34.5 genes and ICP6 genes are defective severely curtailing growth of the virus in normal cells. However, in tumor cells viral replication can still occur. The published trial involved direct stereotactic administration of the OV into the glioma mass. There were no serious adverse events reported and a MTD was not reached [337]. HSV 1716 is derived from a viral strain that is different than that of G207 and it possesses only the ICP34.5 defects but still maintains intact ICP6 expression. Two clinical trials have been published with this virus [338,339]. Again, there was no evidence of serious adverse events attributable to the agent either when administered as a stereotactic bolus or when injected directly into the tumor cavity at the time of surgical resection. The oncolytic adenovirus, ONYX-015, was also evaluated in a multi-institutional doseescalation phase I clinical trial in humans with recurrent malignant glioma. The trial also showed
the absence of serious adverse events attributable to the oncolytic virus when injected into brain adjacent to resected tumor [340]. In some patients, tissue was available after injection demonstrating evidence for a tumoral inflammatory response which is likely due to the host response against the injected oncolytic virus. Results from the first dose-escalating phase I trial of reovirus in recurrent malignant showed that the treatment was well tolerated and provide the basis for additional trials (Forsyth et al., in press) Three cases of patients whose gliomas were injected with the MTH-68/H strain of NDV [341–343] and two vaccination studies using NDV-infected tumor lysates [344,345] also appear to show that RNA viruses seem to be fairly well tolerated and may also have potential as agents against gliomas.
Future Perspectives The field of gene therapy (including oncolytic viruses) has been maturing since there has been a realization that additional laboratory work is needed in order to generate agents that will be both safe and effective. In a way, it was important to test a number of these viruses and vectors in the clinical arena to determine if a safety profile existed. Now that we have reassured ourselves that these agents can be delivered into malignant glial tumors or brain surrounding such tumors in a relatively safe manner, laboratory studies are focused on improving the anticancer efficacy of the current generation of vectors. One of the most important issues revolves around the role of the immune and of the host response to the virus, vector, or transgene product since this response will contribute to the presence or absence of efficacy and toxicity. Another issue relates to biodistribution of the vector or virus and to the capacity to deliver this agent like other pharmacologic antitumor agents (i.e., orally or intravenously). Finally, the engineering of vectors
Gene therapy for brain tumors
and/or viruses which exhibit increased selectivity for glioma remains the object of intense laboratory effort. The convergence of discoveries in these three primary areas will certainly result in gene therapy products with proven anti-glioma efficacy.
References 1. Stupp R, Mason WP, Bent van den MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005;352:987-96. 2. Short MP, Choi BC, Lee JK, Malick A, Breakefield XO, Martuza RL. Gene delivery to glioma cells in rat brain by grafting of a retrovirus packaging cell line. J Neurosci Res 1990;27:427-39. 3. Boviatsis EJ, Chase M, Wei MX, et al. Gene transfer into experimental brain tumors mediated by adenovirus, herpes simplex virus, and retrovirus vectors. Hum Gene Ther 1994;5:183-91. 4. Badie B, Hunt K, Economou JS, Black KL. Stereotactic delivery of a recombinant adenovirus into a C6 glioma cell line in a rat brain tumor model. Neurosurgery 1994;35:910-5; discussion 5-6. 5. Okada H, Miyamura K, Itoh T, et al. Gene therapy against an experimental glioma using adeno-associated virus vectors. Gene Ther 1996;3:957-64. 6. Enger PO, Thorsen F, Lonning PE, Bjerkvig R, Hoover F. Adeno-associated viral vectors penetrate human solid tumor tissue in vivo more effectively than adenoviral vectors. Hum Gene Ther 2002;13:1115-25. 7. Kofler P, Wiesenhofer B, Rehrl C, Baier G, Stockhammer G, Humpel C. Liposome-mediated gene transfer into established CNS cell lines, primary glial cells, and in vivo. Cell Transplant 1998;7:175-85. 8. Ezzeddine ZD, Martuza RL, Platika D, Short MP, Malick A, Choi B, Breakefield XO. Selective killing of glioma cells in culture and in vivo by retrovirus transfer of the herpes simplex virus thymidine kinase gene. New Biol 1991;3:608-14. 9. Culver KW, Ram Z, Wallbridge S, Ishii H, Oldfield EH, Blaese RM. In vivo gene transfer with retroviral vectorproducer cells for treatment of experimental brain tumors. Science 1992;256:1550-2. 10. Ram Z, Culver KW, Walbridge S, Blaese RM, Oldfield EH. In situ retroviral-mediated gene transfer for the treatment of brain tumors in rats. Cancer Res 1993;53: 83-8. 11. Izquierdo M, Cortes M, de Felipe P, Martin V, DiezGuerra J, Talavera A, Perez-Higueras A. Long-term rat survival after malignant brain tumor regression by retroviral gene therapy. Gene Ther 1995;2:66-9.
186
12. Vincent AJ, Vogels R, Someren GV, et al. Herpes simplex virus thymidine kinase gene therapy for rat malignant brain tumors. Hum Gene Ther 1996;7:197-205. 13. Pizzato M, Franchin E, Calvi P, Boschetto R, Colombo M, Ferrini S, Palu G. Production and characterization of a bicistronic Moloney-based retroviral vector expressing human interleukin 2 and herpes simplex virus thymidine kinase for gene therapy of cancer. Gene Ther 1998;5:1003-7. 14. Chen SH, Shine HD, Goodman JC, Grossman RG, Woo SL. Gene therapy for brain tumors: regression of experimental gliomas by adenovirus-mediated gene transfer in vivo. Proc Natl Acad Sci USA 1994;91:3054-7. 15. Perez-Cruet MJ, Trask TW, Chen SH, Goodman JC, Woo SL, Grossman RG, Shine HD. Adenovirus-mediated gene therapy of experimental gliomas. J Neurosci Res 1994;39:506-11. 16. Maron A, Gustin T, Le Roux A, et al. Gene therapy of rat C6 glioma using adenovirus-mediated transfer of the herpes simplex virus thymidine kinase gene: long-term follow-up by magnetic resonance imaging. Gene Ther 1996;3:315-22. 17. Greco O, Joiner MC, Doleh A, Powell AD, Hillman GG, Scott SD. Hypoxia- and radiation-activated Cre/loxP ‘molecular switch’ vectors for gene therapy of cancer. Gene Ther 2006;13:206-15. 18. Niranjan A, Moriuchi S, Lunsford LD, et al. Effective treatment of experimental glioblastoma by HSV vectormediated TNF alpha and HSV-tk gene transfer in combination with radiosurgery and ganciclovir administration. Mol Ther 2000;2:114-20. 19. Mizuno M, Yoshida J, Colosi P, Kurtzman G. Adenoassociated virus vector containing the herpes simplex virus thymidine kinase gene causes complete regression of intracerebrally implanted human gliomas in mice, in conjunction with ganciclovir administration. Jpn J Cancer Res 1998;89:76-80. 20. Zerrouqi A, Rixe O, Ghoumari AM, Yarovoi SV, Mouawad R, Khayat D, Soubrane C. Liposomal delivery of the herpes simplex virus thymidine kinase gene in glioma: improvement of cell sensitization to ganciclovir. Cancer Gene Ther 1996;3:385-92. 21. von Eckardstein KL, Patt S, Zhu J, Zhang L, CervosNavarro J, Reszka R. Short-term neuropathological aspects of in vivo suicide gene transfer to the F98 rat glioblastoma using liposomal and viral vectors. Histology and histopathology 2001;16:735-44. 22. Ge K, Xu L, Zheng Z, Xu D, Sun L, Liu X. Transduction of cytosine deaminase gene makes rat glioma cells highly sensitive to 5-fluorocytosine. Int J Cancer 1997;71:675-9. 23. Wang WJ, Tai CK, Kasahara N, Chen TC. Highly efficient and tumor-restricted gene transfer to malignant gliomas by replication-competent retroviral vectors. Hum Gene Ther 2003;14:117-27. 24. Tai CK, Wang WJ, Chen TC, Kasahara N. Single-shot, multicycle suicide gene therapy by replication-competent
3103
3104
186 25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
Gene therapy for brain tumors
retrovirus vectors achieves long-term survival benefit in experimental glioma. Mol Ther 2005;12:842-51. Wang W, Tai CK, Kershaw AD, Solly SK, Klatzmann D, Kasahara N, Chen TC. Use of replication-competent retroviral vectors in an immunocompetent intracranial glioma model. Neurosurg Focus 2006;20:E25. Ichikawa T, Tamiya T, Adachi Y, et al. In vivo efficacy and toxicity of 5-fluorocytosine/cytosine deaminase gene therapy for malignant gliomas mediated by adenovirus. Cancer Gene Ther 2000;7:74-82. Miller CR, Williams CR, Buchsbaum DJ, Gillespie GY. Intratumoral 5-fluorouracil produced by cytosine deaminase/5-fluorocytosine gene therapy is effective for experimental human glioblastomas. Cancer Res 2002; 62:773-80. Wei MX, Tamiya T, Chase M, et al. Experimental tumor therapy in mice using the cyclophosphamide-activating cytochrome P450 2B1 gene. Hum Gene Ther 1994; 5:969-78. Wei MX, Tamiya T, Rhee RJ, Breakefield XO, Chiocca EA. Diffusible cytotoxic metabolites contribute to the in vitro bystander effect associated with the cyclophosphamide/cytochrome P450 2B1 cancer gene therapy paradigm. Clin Cancer Res 1995;1:1171-7. Jounaidi Y, Waxman DJ. Combination of the bioreductive drug tirapazamine with the chemotherapeutic prodrug cyclophosphamide for P450/P450-reductase-based cancer gene therapy. Cancer Res 2000;60:3761-9. Manome Y, Wen PY, Chen L, et al. Gene therapy for malignant gliomas using replication incompetent retroviral and adenoviral vectors encoding the cytochrome P450 2B1 gene together with cyclophosphamide. Gene Ther 1996;3:513-20. Chen L, Yu LJ, Waxman DJ. Potentiation of cytochrome P450/cyclophosphamide-based cancer gene therapy by coexpression of the P450 reductase gene. Cancer Res 1997;57:4830-7. Manome Y, Wen PY, Dong Y, Tanaka T, Mitchell BS, Kufe DW, Fine HA. Viral vector transduction of the human deoxycytidine kinase cDNA sensitizes glioma cells to the cytotoxic effects of cytosine arabinoside in vitro and in vivo. Nat Med 1996;2:567-73. Rogulski KR, Kim JH, Kim SH, Freytag SO. Glioma cells transduced with an Escherichia coli CD/HSV-1 TK fusion gene exhibit enhanced metabolic suicide and radiosensitivity. Hum Gene Ther 1997;8:73-85. Wang ZH, Zagzag D, Zeng B, Kolodny EH. In vivo and in vitro glioma cell killing induced by an adenovirus expressing both cytosine deaminase and thymidine kinase and its association with interferon-alpha. J Neuropathol Exp Neurol 1999;58:847-58. Chang JW, Lee H, Kim E, Lee Y, Chung SS, Kim JH. Combined antitumor effects of an adenoviral cytosine deaminase/thymidine kinase fusion gene in rat C6 glioma. Neurosurgery 2000;47:931-8; discussion 8-9.
37. Moriuchi S, Wolfe D, Tamura M, Yoshimine T, Miura F, Cohen JB, Glorioso JC. Double suicide gene therapy using a replication defective herpes simplex virus vector reveals reciprocal interference in a malignant glioma model. Gene Ther 2002;9:584-91. 38. Aghi M, Kramm CM, Chou TC, Breakefield XO, Chiocca EA. Synergistic anticancer effects of ganciclovir/thymidine kinase and 5-fluorocytosine/cytosine deaminase gene therapies. J Natl Cancer Inst 1998;90:370-80. 39. Marconi P, Tamura M, Moriuchi S, et al. Connexin 43enhanced suicide gene therapy using herpesviral vectors. Mol Ther 2000;1:71-81. 40. Niranjan A, Wolfe D, Tamura M, et al. Treatment of rat gliosarcoma brain tumors by HSV-based multigene therapy combined with radiosurgery. Mol Ther 2003; 8:530-42. 41. Wang TJ, Huang MS, Hong CY, Tse V, Silverberg GD, Hsiao M. Comparisons of tumor suppressor p53, p21, and p16 gene therapy effects on glioblastoma tumorigenicity in situ. Biochem Biophys Res Commun 2001;287:173-80. 42. Badie B, Drazan KE, Kramar MH, Shaked A, Black KL. Adenovirus-mediated p53 gene delivery inhibits 9L glioma growth in rats. Neurological research 1995;17: 209-16. 43. Kock H, Harris MP, Anderson SC, et al. Adenovirusmediated p53 gene transfer suppresses growth of human glioblastoma cells in vitro and in vivo. Int J Cancer 1996;67:808-15. 44. Li H, Alonso-Vanegas M, Colicos MA, et al. Intracerebral adenovirus-mediated p53 tumor suppressor gene therapy for experimental human glioma. Clin Cancer Res 1999;5:637-42. 45. Cirielli C, Inyaku K, Capogrossi MC, Yuan X, Williams JA. Adenovirus-mediated wild-type p53 expression induces apoptosis and suppresses tumorigenesis of experimental intracranial human malignant glioma. J Neurooncol 1999;43:99-108. 46. Abe T, Wakimoto H, Bookstein R, Maneval DC, Chiocca EA, Basilion JP. Intra-arterial delivery of p53containing adenoviral vector into experimental brain tumors. Cancer Gene Ther 2002;9:228-35. 47. Kim SK, Wang KC, Cho BK, et al. Interaction between p53 and p16 expressed by adenoviral vectors in human malignant glioma cell lines. J Neurosurg 2002;97:143-50. 48. Hsiao M, Tse V, Carmel J, Tsai Y, Felgner PL, Haas M, Silverberg GD. Intracavitary liposome-mediated p53 gene transfer into glioblastoma with endogenous wildtype p53 in vivo results in tumor suppression and longterm survival. Biochem Biophys Res Commun 1997;233:359-64. 49. Fueyo J, Gomez-Manzano C, Yung WK, et al. Adenovirusmediated p16/CDKN2 gene transfer induces growth arrest and modifies the transformed phenotype of glioma cells. Oncogene 1996;12:103-10.
Gene therapy for brain tumors
50. Lee SH, Kim MS, Kwon HC, et al. Growth inhibitory effect on glioma cells of adenovirus-mediated p16/INK4a gene transfer in vitro and in vivo. Int J Mol Med 2000;6:559-63. 51. Harada H, Nakagawa K, Iwata S, et al. Restoration of wild-type p16 down-regulates vascular endothelial growth factor expression and inhibits angiogenesis in human gliomas. Cancer Res 1999;59:3783-9. 52. Chen J, Willingham T, Shuford M, Bruce D, Rushing E, Smith Y, Nisen PD. Effects of ectopic overexpression of p21 (WAF1/CIP1) on aneuploidy and the malignant phenotype of human brain tumor cells. Oncogene 1996;13: 1395-403. 53. Chen J, Willingham T, Shuford M, Nisen PD. Tumor suppression and inhibition of aneuploid cell accumulation in human brain tumor cells by ectopic overexpression of the cyclin-dependent kinase inhibitor p27KIP1. J Clin Invest 1996;97:1983-8. 54. Park KH, Lee J, Yoo CG, et al. Application of p27 gene therapy for human malignant glioma potentiated by using mutant p27. J Neurosurg 2004;101:505-10. 55. You Y, Geng X, Zhao P, et al. Evaluation of combination gene therapy with PTEN and antisense hTERT for malignant glioma in vitro and xenografts. Cell Mol Life Sci 2007;64:621-31. 56. Saydam O, Glauser DL, Heid I, et al. Herpes simplex virus 1 amplicon vector-mediated siRNA targeting epidermal growth factor receptor inhibits growth of human glioma cells in vivo. Mol Ther 2005;12:803-12. 57. Pu P, Liu X, Liu A, Cui J, Zhang Y. Inhibitory effect of antisense epidermal growth factor receptor RNA on the proliferation of rat C6 glioma cells in vitro and in vivo. J Neurosurg 2000;92:132-9. 58. Kang CS, Zhang ZY, Jia ZF, et al. Suppression of EGFR expression by antisense or small interference RNA inhibits U251 glioma cell growth in vitro and in vivo. Cancer Gene Ther 2006;13:530-8. 59. Zhang Y, Zhang YF, Bryant J, Charles A, Boado RJ, Pardridge WM. Intravenous RNA interference gene therapy targeting the human epidermal growth factor receptor prolongs survival in intracranial brain cancer. Clin Cancer Res 2004;10:3667-77. 60. Kock N, Kasmieh R, Weissleder R, Shah K. Tumor therapy mediated by lentiviral expression of shBcl-2 and S-TRAIL. Neoplasia 2007;9:435-42. 61. Lee J, Hampl M, Albert P, Fine HA. Antitumor activity and prolonged expression from a TRAIL-expressing adenoviral vector. Neoplasia 2002;4:312-23. 62. Naumann U, Waltereit R, Schulz JB, Weller M. Adenoviral (full-length) Apo2L/TRAIL gene transfer is an ineffective treatment strategy for malignant glioma. J Neurooncol 2003;61:7-15. 63. Rubinchik S, Yu H, Woraratanadharm J, VoelkelJohnson C, Norris JS, Dong JY. Enhanced apoptosis of glioma cell lines is achieved by co-delivering FasL-GFP and
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
186
TRAIL with a complex Ad5 vector. Cancer Gene Ther 2003;10:814-22. Kim CY, Jeong M, Mushiake H, et al. Cancer gene therapy using a novel secretable trimeric TRAIL. Gene Ther 2006;13:330-8. Zhang Y, Wang C, Zhang Y, Sun M. C6 glioma cells retrovirally engineered to express IL-18 and Fas exert FasL-dependent cytotoxicity against glioma formation. Biochemical and biophysical research communications 2004;325:1240-5. Ambar BB, Frei K, Malipiero U, Morelli AE, Castro MG, Lowenstein PR, Fontana A. Treatment of experimental glioma by administration of adenoviral vectors expressing Fas ligand. Hum Gene Ther 1999;10:1641-8. Ho IA, Hui KM, Lam PY. Targeting proliferating tumor cells via the transcriptional control of therapeutic genes. Cancer Gene Ther 2006;13:44-52. Shinoura N, Koike H, Furitu T, Hashimoto M, Asai A, Kirino T, Hamada H. Adenovirus-mediated transfer of caspase-8 augments cell death in gliomas: implication for gene therapy. Hum Gene Ther 2000;11:1123-37. Shinoura N, Saito K, Yoshida Y, Hashimoto M, Asai A, Kirino T, Hamada H. Adenovirus-mediated transfer of bax with caspase-8 controlled by myelin basic protein promoter exerts an enhanced cytotoxic effect in gliomas. Cancer Gene Ther 2000;7:739-48. Shinoura N, Sakurai S, Asai A, Kirino T, Hamada H. Caspase-9 transduction overrides the resistance mechanism against p53-mediated apoptosis in U-87MG glioma cells. Neurosurgery 2001;49:177-86; discussion 86–7. Komata T, Kondo Y, Kanzawa T, et al. Treatment of malignant glioma cells with the transfer of constitutively active caspase-6 using the human telomerase catalytic subunit (human telomerase reverse transcriptase) gene promoter. Cancer Res 2001;61:5796-802. Mitlianga PG, Gomez-Manzano C, Kyritsis AP, Fueyo J. Overexpression of E2F-1 leads to bax-independent cell death in human glioma cells. Int J Oncol 2002;21: 1015-20. Arafat WO, Buchsbaum DJ, Gomez-Navarro J, et al. An adenovirus encoding proapoptotic Bax synergistically radiosensitizes malignant glioma. Int J Radiat Oncol Biol Phys 2003;55:1037-50. Lee A, DeJong G, Guo J, Bu X, Jia WW. Bax expressed from a herpes viral vector enhances the efficacy of N,N’-bis (2-hydroxyethyl)-N-nitrosourea treatment in a rat glioma model. Cancer Gene Ther 2000;7:1113-9. Kondo S, Ishizaka Y, Okada T, et al. FADD gene therapy for malignant gliomas in vitro and in vivo. Hum Gene Ther 1998;9:1599-608. Yu JS, Sena-Esteves M, Paulus W, Breakefield XO, Reeves SA. Retroviral delivery and tetracycline-dependent expression of IL-1beta-converting enzyme (ICE) in a rat glioma model provides controlled induction of apoptotic death in tumor cells. Cancer Res 1996;56:5423-7.
3105
3106
186
Gene therapy for brain tumors
77. Fueyo J, Gomez-Manzano C, Yung WK, et al. Overexpression of E2F-1 in glioma triggers apoptosis and suppresses tumor growth in vitro and in vivo. Nat Med 1998;4:685-90. 78. Tanaka T, Cao Y, Folkman J, Fine HA. Viral vectortargeted antiangiogenic gene therapy utilizing an angiostatin complementary DNA. Cancer Res 1998;58:3362-9. 79. Ma HI, Lin SZ, Chiang YH, Li J, Chen SL, Tsao YP, Xiao X. Intratumoral gene therapy of malignant brain tumor in a rat model with angiostatin delivered by adenoassociated viral (AAV) vector. Gene Ther 2002;9:2-11. 80. Barnett FH, Scharer-Schuksz M, Wood M, Yu X, Wagner TE, Friedlander M. Intra-arterial delivery of endostatin gene to brain tumors prolongs survival and alters tumor vessel ultrastructure. Gene Ther 2004;11: 1283-9. 81. Sasaki M, Wizigmann-Voos S, Risau W, Plate KH. Retrovirus producer cells encoding antisense VEGF prolong survival of rats with intracranial GS9L gliomas. Int J Dev Neurosci 1999;17:579-91. 82. Niola F, Evangelisti C, Campagnolo L, et al. A plasmidencoded VEGF siRNA reduces glioblastoma angiogenesis and its combination with interleukin-4 blocks tumor growth in a xenograft mouse model. Cancer Biol Ther 2006;5:174-9. 83. Ciafre SA, Niola F, Wannenes F, Farace MG. An antiVEGF ribozyme embedded within the adenoviral VAI sequence inhibits glioblastoma cell angiogenic potential in vitro. J Vasc Res 2004;41:220-8. 84. Harding TC, Dickinson PJ, Roberts BN, Yendluri S, Gonzalez-Edick M, Lecouteur RA, Jooss KU. Enhanced gene transfer efficiency in the murine striatum and an orthotopic glioblastoma tumor model, using AAV-7- and AAV-8-pseudotyped vectors. Hum Gene Ther 2006;17: 807-20. 85. Tanaka T, Manome Y, Wen P, Kufe DW, Fine HA. Viral vector-mediated transduction of a modified platelet factor 4 cDNA inhibits angiogenesis and tumor growth. Nat Med 1997;3:437-42. 86. Yanamandra N, Kondraganti S, Gondi CS, Gujrati M, Olivero WC, Dinh DH, Rao JS. Recombinant adenoassociated virus (rAAV) expressing TFPI-2 inhibits invasion, angiogenesis and tumor growth in a human glioblastoma cell line. Int J Cancer 2005;115:998-1005. 87. Kang X, Xiao X, Harata M, et al. Antiangiogenic activity of BAI1 in vivo: implications for gene therapy of human glioblastomas. Cancer Gene Ther 2006;13:385-92. 88. Saiki M, Mima T, Takahashi JC, et al. Adenovirusmediated gene transfer of a truncated form of fibroblast growth factor receptor inhibits growth of glioma cells both in vitro and in vivo. J Neurooncol 1999;44:195-203. 89. Lakka SS, Rajan M, Gondi C, et al. Adenovirus-mediated expression of antisense MMP-9 in glioma cells inhibits tumor growth and invasion. Oncogene 2002;21:8011-9. 90. Lakka SS, Gondi CS, Yanamandra N, Dinh DH, Olivero WC, Gujrati M, Rao JS. Synergistic down-regulation of
urokinase plasminogen activator receptor and matrix metalloproteinase-9 in SNB19 glioblastoma cells efficiently inhibits glioma cell invasion, angiogenesis, and tumor growth. Cancer Res 2003;63:2454-61. 91. Gondi CS, Lakka SS, Yanamandra N, et al. Expression of antisense uPAR and antisense uPA from a bicistronic adenoviral construct inhibits glioma cell invasion, tumor growth, and angiogenesis. Oncogene 2003;22:5967-75. 92. Gondi CS, Lakka SS, Dinh DH, Olivero WC, Gujrati M, Rao JS. Downregulation of uPA, uPAR and MMP-9 using small, interfering, hairpin RNA (siRNA) inhibits glioma cell invasion, angiogenesis and tumor growth. Neuron Glia Biol 2004;1:165-76. 93. Lakka SS, Gondi CS, Dinh DH, et al. Specific interference of urokinase-type plasminogen activator receptor and matrix metalloproteinase-9 gene expression induced by double-stranded RNA results in decreased invasion, tumor growth, and angiogenesis in gliomas. J Biol Chem 2005;280:21882-92. 94. Lakka SS, Gondi CS, Yanamandra N, Olivero WC, Dinh DH, Gujrati M, Rao JS. Inhibition of cathepsin B and MMP-9 gene expression in glioblastoma cell line via RNA interference reduces tumor cell invasion, tumor growth and angiogenesis. Oncogene 2004;23:4681-9. 95. Gondi CS, Lakka SS, Dinh DH, Olivero WC, Gujrati M, Rao JS. Intraperitoneal injection of a hairpin RNAexpressing plasmid targeting urokinase-type plasminogen activator (uPA) receptor and uPA retards angiogenesis and inhibits intracranial tumor growth in nude mice. Clin Cancer Res 2007;13:4051-60. 96. Abounader R, Lal B, Luddy C, Koe G, Davidson B, Rosen EM, Laterra J. In vivo targeting of SF/HGF and cmet expression via U1snRNA/ribozymes inhibits glioma growth and angiogenesis and promotes apoptosis. Faseb J 2002;16:108-10. 97. Ram Z, Walbridge S, Heiss JD, Culver KW, Blaese RM, Oldfield EH. In vivo transfer of the human interleukin2 gene: negative tumoricidal results in experimental brain tumors. J Neurosurg 1994;80:535-40. 98. Tseng SH, Hwang LH, Lin SM. Induction of antitumor immunity by intracerebrally implanted rat C6 glioma cells genetically engineered to secrete cytokines. J Immunother (1997) 1997;20:334-42. 99. Sampson JH, Ashley DM, Archer GE, Fuchs HE, Dranoff G, Hale LP, Bigner DD. Characterization of a spontaneous murine astrocytoma and abrogation of its tumorigenicity by cytokine secretion. Neurosurgery 1997;41:1365-72; discussion 72–3. 100. Donson AM, Foreman NK. Adenovirus mediated gene therapy in a glioblastoma vaccine model; specific antitumor immunity and abrogation of immunosuppression. J Neurooncol 1998;40:205-14. 101. Wei MX, Tamiya T, Hurford RK, Jr, Boviatsis EJ, Tepper RI, Chiocca EA. Enhancement of interleukin4-mediated tumor regression in athymic mice by in situ retroviral gene transfer. Hum Gene Ther 1995;6:437-43.
Gene therapy for brain tumors
102. Saleh M, Wiegmans A, Malone Q, Stylli SS, Kaye AH. Effect of in situ retroviral interleukin-4 transfer on established intracranial tumors. J Natl Cancer Inst 1999;91:438-45. 103. Okada H, Giezeman-Smits KM, Tahara H, et al. Effective cytokine gene therapy against an intracranial glioma using a retrovirally transduced IL-4 plus HSVtk tumor vaccine. Gene Ther 1999;6:219-26. 104. Wei MX, Li F, Ono Y, Gauldie J, Chiocca EA. Effects on brain tumor cell proliferation by an adenovirus vector that bears the interleukin-4 gene. J Neurovirol 1998;4:237-41. 105. Lumniczky K, Desaknai S, Mangel L, Szende B, Hamada H, Hidvegi EJ, Safrany G. Local tumor irradiation augments the antitumor effect of cytokine-producing autologous cancer cell vaccines in a murine glioma model. Cancer Gene Ther 2002;9:44-52. 106. Liu Y, Ehtesham M, Samoto K, et al. In situ adenoviral interleukin 12 gene transfer confers potent and longlasting cytotoxic immunity in glioma. Cancer Gene Ther 2002;9:9-15. 107. Ehtesham M, Samoto K, Kabos P, Acosta FL, Gutierrez MA, Black KL, Yu JS. Treatment of intracranial glioma with in situ interferon-gamma and tumor necrosis factor-alpha gene transfer. Cancer Gene Ther 2002;9:925-34. 108. Yamini B, Yu X, Gillespie GY, Kufe DW, Weichselbaum RR. Transcriptional targeting of adenovirally delivered tumor necrosis factor alpha by temozolomide in experimental glioblastoma. Cancer Res 2004;64:6381-4. 109. Yamini B, Yu X, Pytel P, et al. Adenovirally delivered tumor necrosis factor-alpha improves the antiglioma efficacy of concomitant radiation and temozolomide therapy. Clin Cancer Res 2007;13:6217-23. 110. Moriuchi S, Oligino T, Krisky D, Marconi P, Fink D, Cohen J, Glorioso JC. Enhanced tumor cell killing in the presence of ganciclovir by herpes simplex virus type 1 vector-directed coexpression of human tumor necrosis factor-alpha and herpes simplex virus thymidine kinase. Cancer Res 1998;58:5731-7. 111. Mizuno M, Yoshida J, Oyama H, Sugita K. Growth inhibition of glioma cells by liposome-mediated cell transfection with tumor necrosis factor-alpha gene– its enhancement by prior gamma-interferon treatment. Neurologia medico-chirurgica 1992;32:873-6. 112. Saleh M, Jonas NK, Wiegmans A, Stylli SS. The treatment of established intracranial tumors by in situ retroviral IFN-gamma transfer. Gene Ther 2000;7: 1715-24. 113. Kanno H, Hattori S, Sato H, et al. Experimental gene therapy against subcutaneously implanted glioma with a herpes simplex virus-defective vector expressing interferon-gamma. Cancer Gene Ther 1999; 6:147-54. 114. Yoshida J, Mizuno M, Nakahara N, Colosi P. Antitumor effect of an adeno-associated virus vector containing the
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
126.
127.
186
human interferon-beta gene on experimental intracranial human glioma. Jpn J Cancer Res 2002;93:223-8. Streck CJ, Dickson PV, Ng CY, et al. Antitumor efficacy of AAV-mediated systemic delivery of interferon-beta. Cancer Gene Ther 2006;13:99-106. Natsume A, Mizuno M, Ryuke Y, Yoshida J. Antitumor effect and cellular immunity activation by murine interferon-beta gene transfer against intracerebral glioma in mouse. Gene Ther 1999;6:1626-33. Natsume A, Tsujimura K, Mizuno M, Takahashi T, Yoshida J. IFN-beta gene therapy induces systemic antitumor immunity against malignant glioma. J Neurooncol 2000;47:117-24. Lefranc F, Cool V, Velu T, Brotchi J, De Witte O. Granulocyte macrophage-colony stimulating factor gene transfer to induce a protective anti-tumoral immune response against the 9L rat gliosarcoma model. Int J Oncol 2002;20:1077-85. Herrlinger U, Kramm CM, Johnston KM, et al. Vaccination for experimental gliomas using GM-CSF-transduced glioma cells. Cancer Gene Ther 1997;4:345-52. Herrlinger U, Jacobs A, Quinones A, et al. Helper virusfree herpes simplex virus type 1 amplicon vectors for granulocyte-macrophage colony-stimulating factorenhanced vaccination therapy for experimental glioma. Hum Gene Ther 2000;11:1429-38. Parney IF, Petruk KC, Zhang C, Farr-Jones M, Sykes DB, Chang LJ. Granulocyte-macrophage colonystimulating factor and B7–2 combination immunogene therapy in an allogeneic Hu-PBL-SCID/beige mousehuman glioblastoma multiforme model. Hum Gene Ther 1997;8:1073-85. Ali S, Curtin JF, Zirger JM, et al. Inflammatory and anti-glioma effects of an adenovirus expressing human soluble Fms-like tyrosine kinase 3 ligand (hsFlt3L): treatment with hsFlt3L inhibits intracranial glioma progression. Mol Ther 2004;10:1071-84. Ali S, King GD, Curtin JF, et al. Combined immunostimulation and conditional cytotoxic gene therapy provide long-term survival in a large glioma model. Cancer Res 2005;65:7194-204. Rogulski KR, Zhang K, Kolozsvary A, Kim JH, Freytag SO. Pronounced antitumor effects and tumor radiosensitization of double suicide gene therapy. Clin Cancer Res 1997;3:2081-8. Benedetti S, Dimeco F, Pollo B, et al. Limited efficacy of the HSV-TK/GCV system for gene therapy of malignant gliomas and perspectives for the combined transduction of the interleukin-4 gene. Hum Gene Ther 1997;8: 1345-53. Gurzov EN, Izquierdo M. Cyclin E1 knockdown induces apoptosis in cancer cells. Neurological Res 2006;28: 493-9. Hong JS, Waud WR, Levasseur DN, et al. Excellent in vivo bystander activity of fludarabine phosphate against human glioma xenografts that express the escherichia coli
3107
3108
186 128.
129.
130.
131.
132.
133.
134.
135.
136.
137.
138.
139.
Gene therapy for brain tumors
purine nucleoside phosphorylase gene. Cancer Res 2004;64:6610-5. Galipeau J, Li H, Paquin A, Sicilia F, Karpati G, Nalbantoglu J. Vesicular stomatitis virus G pseudotyped retrovector mediates effective in vivo suicide gene delivery in experimental brain cancer. Cancer Res 1999;59:2384-94. Miletic H, Fischer YH, Neumann H, et al. Selective transduction of malignant glioma by lentiviral vectors pseudotyped with lymphocytic choriomeningitis virus glycoproteins. Hum Gene Ther 2004;15:1091-100. Steffens S, Tebbets J, Kramm CM, Lindemann D, Flake A, Sena-Esteves M. Transduction of human glial and neuronal tumor cells with different lentivirus vector pseudotypes. J Neurooncol 2004;70:281-8. Solly SK, Nguyen TH, Weber A, Horellou P. Targeting of c-Met and urokinase expressing human glioma cell lines by retrovirus vector displaying single-chain variable fragment antibody. Cancer Biol Ther 2005;4:987-92. Li H, Lochmuller H, Yong VW, Karpati G, Nalbantoglu J. Adenovirus-mediated wild-type p53 gene transfer and overexpression induces apoptosis of human glioma cells independent of endogenous p53 status. J Neuropathol Exp Neurol 1997;56:872-8. Gomez-Manzano C, Fueyo J, Kyritsis AP, et al. Adenovirus-mediated transfer of the p53 gene produces rapid and generalized death of human glioma cells via apoptosis. Cancer Res 1996;56:694-9. Shinoura N, Furitsu T, Asai A, Kirino T, Hamada H. Co-transduction of p27Kip1 strongly augments Fas ligand- and caspase-8-mediated apoptosis in U-373MG glioma cells. Anticancer Res 2001;21:3261-8. Shinoura N, Sakurai S, Shibasaki F, Asai A, Kirino T, Hamada H. Co-transduction of Apaf-1 and caspase-9 highly enhances p53-mediated apoptosis in gliomas. Br J Cancer 2002;86:587-95. Parr MJ, Wen PY, Schaub M, Khoury SJ, Sayegh MH, Fine HA. Immune parameters affecting adenoviral vector gene therapy in the brain. J Neurovirol 1998;4:194-203. Goodman JC, Trask TW, Chen SH, et al. Adenoviralmediated thymidine kinase gene transfer into the primate brain followed by systemic ganciclovir: pathologic, radiologic, and molecular studies. Hum Gene Ther 1996;7:1241-50. Miller CR, Buchsbaum DJ, Reynolds PN, et al. Differential susceptibility of primary and established human glioma cells to adenovirus infection: targeting via the epidermal growth factor receptor achieves fiber receptor-independent gene transfer. Cancer Res 1998;58:5738-48. Grill J, Van Beusechem VW, Valk Van Der P, et al. Combined targeting of adenoviruses to integrins and epidermal growth factor receptors increases gene transfer into primary glioma cells and spheroids. Clin Cancer Res 2001;7:641-50.
140. Staba MJ, Wickham TJ, Kovesdi I, Hallahan DE. Modifications of the fiber in adenovirus vectors increase tropism for malignant glioma models. Cancer Gene Ther 2000;7:13-19. 141. Wang W, Zhu NL, Chua J, et al. Retargeting of adenoviral vector using basic fibroblast growth factor ligand for malignant glioma gene therapy. J Neurosurg 2005;103:1058-66. 142. Ulasov IV, Tyler MA, Han Y, Glasgow JN, Lesniak MS. Novel recombinant adenoviral vector that targets the interleukin-13 receptor alpha2 chain permits effective gene transfer to malignant glioma. Hum Gene Ther 2007;18:118-29. 143. Hoffmann D, Meyer B, Wildner O. Improved glioblastoma treatment with Ad5/35 fiber chimeric conditionally replicating adenoviruses. J Gene Med 2007; 9:764-78. 144. Stoff-Khalili MA, Rivera AA, Glasgow JN, et al. A human adenoviral vector with a chimeric fiber from canine adenovirus type 1 results in novel expanded tropism for cancer gene therapy. Gene Ther 2005;12: 1696-706. 145. Zheng S, Ulasov IV, Han Y, Tyler MA, Zhu ZB, Lesniak MS. Fiber-knob modifications enhance adenoviral tropism and gene transfer in malignant glioma. J Gene Med 2007;9:151-60. 146. Van Houdt WJ, Wu H, Glasgow JN, et al. Gene delivery into malignant glioma by infectivity-enhanced adenovirus: in vivo versus in vitro models. Neuro Oncol 2007;9:280-90. 147. Einfeld DA, Schroeder R, Roelvink PW, Lizonova A, King CR, Kovesdi I, Wickham TJ. Reducing the native tropism of adenovirus vectors requires removal of both CAR and integrin interactions. J Virol 2001;75: 11284-91. 148. van Beusechem VW, Grill J, Mastenbroek DC, et al. Efficient and selective gene transfer into primary human brain tumors by using single-chain antibody-targeted adenoviral vectors with native tropism abolished. J Virol 2002;76:2753-62. 149. Vandier D, Rixe O, Besnard F, et al. Inhibition of glioma cells in vitro and in vivo using a recombinant adenoviral vector containing an astrocyte-specific promoter. Cancer Gene Ther 2000;7:1120-6. 150. de Leeuw B, Su M, ter Horst M, et al. Increased gliaspecific transgene expression with glial fibrillary acidic protein promoters containing multiple enhancer elements. J Neurosci Res 2006;83:744-53. 151. Chen J, Bezdek T, Chang J, Kherzai AW, Willingham T, Azzara M, Nisen PD. A glial-specific, repressible, adenovirus vector for brain tumor gene therapy. Cancer Res 1998;58:3504-7. 152. Kurihara H, Zama A, Tamura M, Takeda J, Sasaki T, Takeuchi T. Glioma/glioblastoma-specific adenoviral gene expression using the nestin gene regulator. Gene Ther 2000;7:686-93.
Gene therapy for brain tumors
153. Parr MJ, Manome Y, Tanaka T, Wen P, Kufe DW, Kaelin WG, Jr, Fine HA. Tumor-selective transgene expression in vivo mediated by an E2F-responsive adenoviral vector. Nat Med 1997;3:1145-9. 154. Van Houdt WJ, Haviv YS, Lu B, et al. The human survivin promoter: a novel transcriptional targeting strategy for treatment of glioma. J Neurosurg 2006; 104:583-92. 155. Ulasov IV, Zhu ZB, Tyler MA, et al. Survivin-driven and fiber-modified oncolytic adenovirus exhibits potent antitumor activity in established intracranial glioma. Hum Gene Ther 2007;18:589-602. 156. Ulasov IV, Rivera AA, Sonabend AM, Rivera LB, Wang M, Zhu ZB, Lesniak MS. Comparative evaluation of survivin, midkine and CXCR4 promoters for transcriptional targeting of glioma gene therapy. Cancer Biol Ther 2007;6:679-85. 157. DeLuca NA, McCarthy AM, Schaffer PA. Isolation and characterization of deletion mutants of herpes simplex virus type 1 in the gene encoding immediate-early regulatory protein ICP4. J Virol 1985;56: 558-70. 158. McCarthy AM, McMahan L, Schaffer PA. Herpes simplex virus type 1 ICP27 deletion mutants exhibit altered patterns of transcription and are DNA deficient. J Virol 1989;63:18-27. 159. Samaniego LA, Webb AL, DeLuca NA. Functional interactions between herpes simplex virus immediateearly proteins during infection: gene expression as a consequence of ICP27 and different domains of ICP4. J Virol 1995;69:5705-15. 160. Wu N, Watkins SC, Schaffer PA, DeLuca NA. Prolonged gene expression and cell survival after infection by a herpes simplex virus mutant defective in the immediate-early genes encoding ICP4, ICP27, and ICP22. J Virol 1996;70:6358-69. 161. Samaniego LA, Neiderhiser L, DeLuca NA. Persistence and expression of the herpes simplex virus genome in the absence of immediate-early proteins. J Virol 1998;72: 3307-20. 162. Krisky DM, Wolfe D, Goins WF, et al. Deletion of multiple immediate-early genes from herpes simplex virus reduces cytotoxicity and permits long-term gene expression in neurons. Gene Ther 1998;5:1593-603. 163. Spaete RR, Frenkel N. The herpes simplex virus amplicon: a new eucaryotic defective-virus cloning-amplifying vector. Cell 1982;30:295-304. 164. Freese A, Geller A. Infection of cultured striatal neurons with a defective HSV-1 vector: implications for gene therapy. Nucleic Acids Res 1991;19:7219-23. 165. Fraefel C, Song S, Lim F, et al. Helper virus-free transfer of herpes simplex virus type 1 plasmid vectors into neural cells. J Virol 1996;70:7190-7. 166. Moriuchi S, Glorioso JC, Maruno M, et al. Combination gene therapy for glioblastoma involving herpes simplex virus vector-mediated codelivery of mutant
167.
168.
169.
170.
171.
172.
173.
174.
175. 176.
177.
178.
179.
180.
186
IkappaBalpha and HSV thymidine kinase. Cancer Gene Ther 2005;12:487-96. Ho IA, Hui KM, Lam PY. Glioma-specific and cell cycleregulated herpes simplex virus type 1 amplicon viral vector. Hum Gene Ther 2004;15:495-508. Saeki Y, Fraefel C, Ichikawa T, Breakefield XO, Chiocca EA. Improved helper virus-free packaging system for HSV amplicon vectors using an ICP27-deleted, oversized HSV-1 DNA in a bacterial artificial chromosome. Mol Ther 2001;3:591-601. Cunningham J, Oiwa Y, Nagy D, Podsakoff G, Colosi P, Bankiewicz KS. Distribution of AAV-TK following intracranial convection-enhanced delivery into rats. Cell Transplant 2000;9:585-94. Harding TC, Lalani AS, Roberts BN, et al. AAV serotype 8-mediated gene delivery of a soluble VEGF receptor to the CNS for the treatment of glioblastoma. Mol Ther 2006;13:956-66. Huszthy PC, Svendsen A, Wilson JM, Kotin RM, Lonning PE, Bjerkvig R, Hoover F. Widespread dispersion of adeno-associated virus serotype 1 and adenoassociated virus serotype 6 vectors in the rat central nervous system and in human glioblastoma multiforme xenografts. Hum Gene Ther 2005;16:381-92. Thorsen F, Afione S, Huszthy PC, et al. Adenoassociated virus (AAV) serotypes 2, 4 and 5 display similar transduction profiles and penetrate solid tumor tissue in models of human glioma. J Gene Med 2006;8:1131-40. Ruan H, Su H, Hu L, Lamborn KR, Kan YW, Deen DF. A hypoxia-regulated adeno-associated virus vector for cancer-specific gene therapy. Neoplasia 2001;3:255-63. Wiesenhofer B, Weis C, Humpel C. Glial cell linederived neurotrophic factor (GDNF) is a proliferation factor for rat C6 glioma cells: evidence from antisense experiments. Antisense Nucleic Acid Drug Dev 2000;10:311-21. Hirko AC, Buethe DD, Meyer EM, Hughes JA. Plasmid delivery in the rat brain. Biosci Rep 2002;22:297-308. Mabuchi E, Shimizu K, Miyao Y, et al. Gene delivery by HVJ-liposome in the experimental gene therapy of murine glioma. Gene Ther 1997;4:768-72. Yoshida J, Mizuno M. Simple preparation and characterization of cationic liposomes associated with a monoclonal antibody against glioma-associated antigen (immunoliposomes). J Liposome Res 1995;5:981-95. Zhang Y, Boado RJ, Pardridge WM. In vivo knockdown of gene expression in brain cancer with intravenous RNAi in adult rats. J Gene Med 2003;5:1039-45. Zhang Y, Boado RJ, Pardridge WM. Marked enhancement in gene expression by targeting the human insulin receptor. J Gene Med 2003;5:157-63. Cardoso AL, Simoes S, de Almeida LP, Pelisek J, Culmsee C, Wagner E, Pedroso de Lima MC. siRNA delivery by a transferrin-associated lipid-based vector: a non-viral strategy to mediate gene silencing. J Gene Med 2007;9:170-83.
3109
3110
186
Gene therapy for brain tumors
181. Zhang Y, Wang CW, Wang ZG, et al. Construction of double suicide genes system controlled by MDR1 promoter with targeted expression in drug-resistant glioma cells. J Neurooncol 2008;86:3-11. 182. Rainov NG. A phase III clinical evaluation of herpes simplex virus type 1 thymidine kinase and ganciclovir gene therapy as an adjuvant to surgical resection and radiation in adults with previously untreated glioblastoma multiforme. Hum Gene Ther 2000;11: 2389-401. 183. Sobol RE, Fakhrai H, Shawler D, et al. Interleukin2 gene therapy in a patient with glioblastoma. Gene Ther 1995;2:164-7. 184. Palu G, Cavaggioni A, Calvi P, et al. Gene therapy of glioblastoma multiforme via combined expression of suicide and cytokine genes: a pilot study in humans. Gene Ther 1999;6:330-7. 185. Packer RJ, Raffel C, Villablanca JG, et al. Treatment of progressive or recurrent pediatric malignant supratentorial brain tumors with herpes simplex virus thymidine kinase gene vector-producer cells followed by intravenous ganciclovir administration. J Neurosurg 2000;92:249-54. 186. Shand N, Weber F, Mariani L, et al. A phase 1–2 clinical trial of gene therapy for recurrent glioblastoma multiforme by tumor transduction with the herpes simplex thymidine kinase gene followed by ganciclovir. GLI328 European-Canadian Study Group. Hum Gene Ther 1999;10:2325-35. 187. Trask TW, Trask RP, Aguilar-Cordova E, et al. Phase I study of adenoviral delivery of the HSV-tk gene and ganciclovir administration in patients with current malignant brain tumors. Mol Ther 2000;1:195-203. 188. Colombo F, Barzon L, Franchin E, et al. Combined HSV-TK/IL-2 gene therapy in patients with recurrent glioblastoma multiforme: biological and clinical results. Cancer Gene Ther 2005;12:835-48. 189. Parney IF, Chang LJ, Farr-Jones MA, Hao C, Smylie M, Petruk KC. Technical hurdles in a pilot clinical trial of combined B7–2 and GM-CSF immunogene therapy for glioblastomas and melanomas. J Neurooncol 2006;78: 71-80. 190. Klatzmann D, Valery CA, Bensimon G, et al. A phase I/ II study of herpes simplex virus type 1 thymidine kinase ‘‘suicide’’ gene therapy for recurrent glioblastoma. Study Group on Gene Therapy for Glioblastoma. Hum Gene Ther 1998;9:2595-604. 191. Izquierdo M, Martin V, de Felipe P, et al. Human malignant brain tumor response to herpes simplex thymidine kinase (HSVtk)/ganciclovir gene therapy. Gene Ther 1996;3:491-5. 192. Ram Z, Culver KW, Oshiro EM, et al. Therapy of malignant brain tumors by intratumoral implantation of retroviral vector-producing cells. Nat Med 1997;3: 1354-61. 193. Harsh GR, Deisboeck TS, Louis DN, et al. Thymidine kinase activation of ganciclovir in recurrent malignant
194.
195.
196.
197.
198.
199.
200.
201.
202.
203.
204.
205.
206.
gliomas: a gene-marking and neuropathological study. J Neurosurg 2000;92:804-11. Puumalainen AM, Vapalahti M, Agrawal RS, et al. Beta-galactosidase gene transfer to human malignant glioma in vivo using replication-deficient retroviruses and adenoviruses. Hum Gene Ther 1998;9:1769-74. Smitt PS, Driesse M, Wolbers J, Kros M, Avezaat C. Treatment of relapsed malignant glioma with an adenoviral vector containing the herpes simplex thymidine kinase gene followed by ganciclovir. Mol Ther 2003;7:851-8. Germano IM, Fable J, Gultekin SH, Silvers A. Adenovirus/herpes simplex-thymidine kinase/ganciclovir complex: preliminary results of a phase I trial in patients with recurrent malignant gliomas. J Neurooncol 2003;65:279-89. Sandmair AM, Loimas S, Puranen P, et al. Thymidine kinase gene therapy for human malignant glioma, using replication-deficient retroviruses or adenoviruses. Hum Gene Ther 2000;11:2197-205. Immonen A, Vapalahti M, Tyynela K, et al. AdvHSV-tk gene therapy with intravenous ganciclovir improves survival in human malignant glioma: a randomised, controlled study. Mol Ther 2004;10:967-72. Westphal M. A controlled, randomized, parallel group, multicenter study of the efficiacy and safety of Hsv tk gene therapy (Cerepro) with subsequent ganciclovir for the treatment of patients with operable highgrade glioma. Abstract ST-02, Society for Neurooncology, 12th Annual Scientific meeting, Nov 15–18, 2007, Dallas, TX. Lang FF, Bruner JM, Fuller GN, et al. Phase I trial of adenovirus-mediated p53 gene therapy for recurrent glioma: biological and clinical results. J Clin Oncol 2003;21:2508-18. Chiocca EA, Smith KM, McKinney B, et al. A Phase I, dose-escalation, multi-institutional trial of Ad.hIFN-b gene therapy for recurrent malignant glioma: safety and correlative molecular results. Mol Ther 2008;3:618–26 . Voges J, Reszka R, Gossmann A, et al. Imaging-guided convection-enhanced delivery and gene therapy of glioblastoma. Ann Neurol 2003;54:479-87. Yoshida J, Mizuno M, Fujii M, et al. Human gene therapy for malignant gliomas (glioblastoma multiforme and anaplastic astrocytoma) by in vivo transduction with human interferon beta gene using cationic liposomes. Hum Gene Ther 2004;15:77-86. Jacobs A, Voges J, Reszka R, et al. Positron-emission tomography of vector-mediated gene expression in gene therapy for gliomas. Lancet 2001;358:727-9. Lawler SE, Peruzzi PP, Chiocca EA. Genetic strategies for brain tumor therapy. Cancer Gene Ther 2006;13: 225-33. Fulci G, Chiocca EA. The status of gene therapy for brain tumors. Expert Opin Biol Ther 2007;7:197-208.
Gene therapy for brain tumors
207. Martuza RL, Malick A, Markert JM, Ruffner KL, Coen DM. Experimental therapy of human glioma by means of a genetically engineered virus mutant. Science 1991;252:854-6. 208. Jia WW, McDermott M, Goldie J, Cynader M, Tan J, Tufaro F. Selective destruction of gliomas in immunocompetent rats by thymidine kinase-defective herpes simplex virus type 1. J Natl Cancer Inst 1994;86:1209-15. 209. Boviatsis EJ, Park JS, Sena-Esteves M, et al. Long-term survival of rats harboring brain neoplasms treated with ganciclovir and a herpes simplex virus vector that retains an intact thymidine kinase gene. Cancer Res 1994; 54:5745-51. 210. Mineta T, Rabkin SD, Martuza RL. Treatment of malignant gliomas using ganciclovir-hypersensitive, ribonucleotide reductase-deficient herpes simplex viral mutant. Cancer Res 1994;54:3963-6. 211. Markert JM, Malick A, Coen DM, Martuza RL. Reduction and elimination of encephalitis in an experimental glioma therapy model with attenuated herpes simplex mutants that retain susceptibility to acyclovir. Neurosurgery 1993;32:597-603. 212. McKie EA, MacLean AR, Lewis AD, et al. Selective in vitro replication of herpes simplex virus type 1 (HSV-1) ICP34.5 null mutants in primary human CNS tumours – evaluation of a potentially effective clinical therapy. Br J Cancer 1996;74:745-52. 213. Lasner TM, Kesari S, Brown SM, Lee VM, Fraser NW, Trojanowski JQ. Therapy of a murine model of pediatric brain tumors using a herpes simplex virus type-1 ICP34.5 mutant and demonstration of viral replication within the CNS. J Neuropathol Exp Neurol 1996;55: 1259-69. 214. Mohr I, Gluzman Y. A herpesvirus genetic element which affects translation in the absence of the viral GADD34 function. Embo J 1996;15:4759-66. 215. Mineta T, Rabkin SD, Yazaki T, Hunter WD, Martuza RL. Attenuated multi-mutated herpes simplex virus-1 for the treatment of malignant gliomas. Nat Med 1995;1:938-43. 216. Yazaki T, Manz HJ, Rabkin SD, Martuza RL. Treatment of human malignant meningiomas by G207, a replicationcompetent multimutated herpes simplex virus 1. Cancer Res 1995;55:4752-6. 217. Todo T, Rabkin SD, Martuza RL. Evaluation of ganciclovir-mediated enhancement of the antitumoral effect in oncolytic, multimutated herpes simplex virus type 1 (G207) therapy of brain tumors. Cancer Gene Ther 2000;7:939-46. 218. Todo T, Rabkin SD, Sundaresan P, Wu A, Meehan KR, Herscowitz HB, Martuza RL. Systemic antitumor immunity in experimental brain tumor therapy using a multimutated, replication-competent herpes simplex virus. Hum Gene Ther 1999;10:2741-55. 219. Hunter WD, Martuza RL, Feigenbaum F, et al. Attenuated, replication-competent herpes simplex virus type 1
220.
221. 222.
223.
224.
225.
226.
227.
228.
229.
230.
231.
186
mutant G207: safety evaluation of intracerebral injection in nonhuman primates. J Virol 1999;73:6319-26. Sundaresan P, Hunter WD, Martuza RL, Rabkin SD. Attenuated, replication-competent herpes simplex virus type 1 mutant G207: safety evaluation in mice. J Virol 2000;74:3832-41. Balfour HH, Jr. Antiviral drugs. N Engl J Med 1999;340:1255-68. Chung RY, Saeki Y, Chiocca EA. B-myb promoter retargeting of herpes simplex virus gamma34.5 gene-mediated virulence toward tumor and cycling cells. J Virol 1999;73:7556-64. Kambara H, Okano H, Chiocca EA, Saeki Y. An oncolytic HSV-1 mutant expressing ICP34.5 under control of a nestin promoter increases survival of animals even when symptomatic from a brain tumor. Cancer Res 2005;65:2832-9. Kambara H, Saeki Y, Chiocca EA. Cyclophosphamide allows for in vivo dose reduction of a potent oncolytic virus. Cancer Res 2005;65:11255-8. Kanai R, Tomita H, Hirose Y, et al. Augmented therapeutic efficacy of an oncolytic herpes simplex virus type 1 mutant expressing ICP34.5 under the transcriptional control of musashi1 promoter in the treatment of malignant glioma. Hum Gene Ther 2007;18:63-73. McKie EA, Graham DI, Brown SM. Selective astrocytic transgene expression in vitro and in vivo from the GFAP promoter in a HSV RL1 null mutant vector— potential glioblastoma targeting. Gene Ther 1998; 5:440-50. Chase M, Chung RY, Chiocca EA. An oncolytic viral mutant that delivers the CYP2B1 transgene and augments cyclophosphamide chemotherapy. Nat Biotechnol 1998;16:444-8. Guffey MB, Parker JN, Luckett WS, Jr, Gillespie GY, Meleth S, Whitley RJ, Markert JM. Engineered herpes simplex virus expressing bacterial cytosine deaminase for experimental therapy of brain tumors. Cancer Gene Ther 2007;14:45-56. Tyminski E, Leroy S, Terada K, et al. Brain tumor oncolysis with replication-conditional herpes simplex virus type 1 expressing the prodrug-activating genes, CYP2B1 and secreted human intestinal carboxylesterase, in combination with cyclophosphamide and irinotecan. Cancer Res 2005;65:6850-7. Aghi M, Chou TC, Suling K, Breakefield XO, Chiocca EA. Multimodal cancer treatment mediated by a replicating oncolytic virus that delivers the oxazaphosphorine/rat cytochrome P450 2B1 and ganciclovir/ herpes simplex virus thymidine kinase gene therapies. Cancer Res 1999;59:3861-5. Liu TC, Zhang T, Fukuhara H, et al. Dominantnegative fibroblast growth factor receptor expression enhances antitumoral potency of oncolytic herpes simplex virus in neural tumors. Clin Cancer Res 2006; 12:6791-9.
3111
3112
186
Gene therapy for brain tumors
232. Liu TC, Zhang T, Fukuhara H, et al. Oncolytic HSV armed with platelet factor 4, an antiangiogenic agent, shows enhanced efficacy. Mol Ther 2006;14:789-97. 233. Andreansky S, He B, van Cott J, et al. Treatment of intracranial gliomas in immunocompetent mice using herpes simplex viruses that express murine interleukins. Gene Ther 1998;5:121-30. 234. Parker JN, Gillespie GY, Love CE, Randall S, Whitley RJ, Markert JM. Engineered herpes simplex virus expressing IL-12 in the treatment of experimental murine brain tumors. Proc Natl Acad Sci USA 2000;97:2208-13. 235. Carew JF, Kooby DA, Halterman MW, Kim SH, Federoff HJ, Fong Y. A novel approach to cancer therapy using an oncolytic herpes virus to package amplicons containing cytokine genes. Mol Ther 2001;4:250-6. 236. Wong RJ, Patel SG, Kim S, et al. Cytokine gene transfer enhances herpes oncolytic therapy in murine squamous cell carcinoma. Hum Gene Ther 2001;12:253-65. 237. Todo T, Martuza RL, Dallman MJ, Rabkin SD. In situ expression of soluble B7–1 in the context of oncolytic herpes simplex virus induces potent antitumor immunity. Cancer Res 2001;61:153-61. 238. Liu BL, Robinson M, Han ZQ, et al. ICP34.5 deleted herpes simplex virus with enhanced oncolytic, immune stimulating, and anti-tumour properties. Gene Ther 2003;10:292-303. 239. Da Costa XJ, Brockman MA, Alicot E, et al. Humoral response to herpes simplex virus is complementdependent. Proc Natl Acad Sci USA 1999;96: 12708-12. 240. Wakimoto H, Johnson PR, Knipe DM, Chiocca EA. Effects of innate immunity on herpes simplex virus and its ability to kill tumor cells. Gene Ther 2003;10: 983-90. 241. Ikeda K, Ichikawa T, Wakimoto H, et al. Oncolytic virus therapy of multiple tumors in the brain requires suppression of innate and elicited antiviral responses. Nat Med 1999;5:881-7. 242. Ikeda K, Wakimoto H, Ichikawa T, Jhung S, Hochberg FH, Louis DN, Chiocca EA. Complement depletion facilitates the infection of multiple brain tumors by an intravascular, replication-conditional herpes simplex virus mutant. J Virol 2000;74:4765-75. 243. Fulci G, Breymann L, Gianni D, et al. Cyclophosphamide enhances glioma virotherapy by inhibiting innate immune responses. Proc Natl Acad Sci USA 2006;103: 12873-8. 244. Bischoff JR, Kirn DH, Williams A, et al. An adenovirus mutant that replicates selectively in p53-deficient human tumor cells. Science 1996;274:373-6. 245. O’Shea CC, Soria C, Bagus B, McCormick F. Heat shock phenocopies E1B-55K late functions and selectively sensitizes refractory tumor cells to ONYX-015 oncolytic viral therapy. Cancer Cell 2005;8:61-74. 246. Heise C, Sampson-Johannes A, Williams A, McCormick F, Von Hoff DD, Kirn DH. ONYX-015, an E1B gene-attenuated adenovirus, causes tumor-specific
247.
248.
249.
250.
251.
252.
253.
254.
255.
256.
257.
258.
259.
260.
261.
cytolysis and antitumoral efficacy that can be augmented by standard chemotherapeutic agents. Nat Med 1997;3: 639-45. Geoerger B, Grill J, Opolon P, et al. Oncolytic activity of the E1B-55 kDa-deleted adenovirus ONYX-015 is independent of cellular p53 status in human malignant glioma xenografts. Cancer Res 2002;62:764-72. Geoerger B, Grill J, Opolon P, et al. Potentiation of radiation therapy by the oncolytic adenovirus dl1520 (ONYX-015) in human malignant glioma xenografts. Br J Cancer 2003;89:577-84. Fueyo J, Gomez-Manzano C, Alemany R, et al. A mutant oncolytic adenovirus targeting the Rb pathway produces anti-glioma effect in vivo. Oncogene 2000;19: 2-12. Heise C, Hermiston T, Johnson L, et al. An adenovirus E1A mutant that demonstrates potent and selective systemic anti-tumoral efficacy. Nat Med 2000;6:1134-9. Jiang H, Gomez-Manzano C, Alemany R, et al. Comparative effect of oncolytic adenoviruses with E1A-55 kDa or E1B-55 kDa deletions in malignant gliomas. Neoplasia 2005;7:48-56. Johnson L, Shen A, Boyle L, et al. Selectively replicating adenoviruses targeting deregulated E2F activity are potent, systemic antitumor agents. Cancer Cell 2002; 1:325-37. Hall AR, Dix BR, O’Carroll SJ, Braithwaite AW. p53dependent cell death/apoptosis is required for a productive adenovirus infection. Nat Med 1998;4:1068-72. Goodrum FD, Ornelles DA. p53 status does not determine outcome of E1B 55-kilodalton mutant adenovirus lytic infection. J Virol 1998;72:9479-90. Gomez-Manzano C, Balague C, Alemany R, et al. A novel E1A-E1B mutant adenovirus induces glioma regression in vivo. Oncogene 2004;23:1821-8. Fukuda K, Abei M, Ugai H, et al. E1A, E1B doublerestricted adenovirus for oncolytic gene therapy of gallbladder cancer. Cancer Res 2003;63:4434-40. Kim J, Kim JH, Choi KJ, Kim PH, Yun CO. E1A- and E1B-Double mutant replicating adenovirus elicits enhanced oncolytic and antitumor effects. Hum Gene Ther 2007;18:773-86. Kohno S, Nakagawa K, Hamada K, et al. Midkine promoter-based conditionally replicative adenovirus for malignant glioma therapy. Oncol Rep 2004;12:73-8. Ito H, Aoki H, Kuhnel F, et al. Autophagic cell death of malignant glioma cells induced by a conditionally replicating adenovirus. J Natl Cancer Inst 2006;98: 625-36. Post DE, Van Meir EG. A novel hypoxia-inducible factor (HIF) activated oncolytic adenovirus for cancer therapy. Oncogene 2003;22:2065-72. Post DE, Devi NS, Li Z, et al. Cancer therapy with a replicating oncolytic adenovirus targeting the hypoxic microenvironment of tumors. Clin Cancer Res 2004;10:8603-12.
Gene therapy for brain tumors
262. Alonso MM, Cascallo M, Gomez-Manzano C, et al. ICOVIR-5 shows E2F1 addiction and potent antiglioma effect in vivo. Cancer Res 2007;67:8255-63. 263. van Beusechem VW, Mastenbroek DC, Doel van den P, Lamfers MLM, Grill J, Wu¨rdinger T, Haisma H, Pinedo HM, Gerritsen WR. Conditionally replicative adenovirus expressing a targeting adapter molecule exhibits enhanced oncolytic potency on CAR-deficient tumors. Gene Ther 2003;10:1982-91. 264. Suzuki K, Fueyo J, Krasnykh V, Reynolds PN, Curiel DT, Alemany R. A conditionally replicative adenovirus with enhanced infectivity shows improved oncolytic potency. Clin Cancer Res 2001;7:120-6. 265. Lamfers ML, Grill J, Dirven CM, et al. Potential of the conditionally replicative adenovirus Ad5-Delta24RGD in the treatment of malignant gliomas and its enhanced effect with radiotherapy. Cancer Res 2002;62: 5736-42. 266. Fueyo J, Alemany R, Gomez-Manzano C, et al. Preclinical characterization of the antiglioma activity of a tropism-enhanced adenovirus targeted to the retinoblastoma pathway. J Natl Cancer Inst 2003;95:652-60. 267. Lamfers ML, Idema S, Bosscher L, et al. Differential Effects of Combined Ad5-{Delta}24RGD and Radiation Therapy in In vitro versus In vivo Models of Malignant Glioma. Clin Cancer Res 2007;13:7451-8. 268. Jiang H, Gomez-Manzano C, Aoki H, et al. Examination of the therapeutic potential of Delta-24-RGD in brain tumor stem cells: role of autophagic cell death. J Natl Cancer Inst 2007;99:1410-4. 269. Alonso MM, Gomez-Manzano C, Bekele BN, Yung WK, Fueyo J. Adenovirus-based strategies overcome temozolomide resistance by silencing the O6-methylguanineDNA methyltransferase promoter. Cancer Res 2007; 67:11499-504. 270. Shinoura N, Yoshida Y, Tsunoda R, et al. Highly augmented cytopathic effect of a fiber-mutant E1B-defective adenovirus for gene therapy of gliomas. Cancer Res 1999;59:3411-6. 271. Zhu ZB, Makhija SK, Lu B, et al. Incorporating the survivin promoter in an infectivity enhanced CRAdanalysis of oncolysis and anti-tumor effects in vitro and in vivo. Int J Oncol 2005;27:237-46. 272. Wildner O, Morris JC, Vahanian NN, Ford H, Jr, Ramsey WJ, Blaese RM. Adenoviral vectors capable of replication improve the efficacy of HSVtk/GCV suicide gene therapy of cancer. Gene Ther 1999;6:57-62. 273. Morris JC, Wildner O. Therapy of head and neck squamous cell carcinoma with an oncolytic adenovirus expressing HSV-tk. Mol Ther 2000;1:56-62. 274. Nanda D, Vogels R, Havenga M, Avezaat CJ, Bout A, Smitt PS. Treatment of malignant gliomas with a replicating adenoviral vector expressing herpes simplex virusthymidine kinase. Cancer Res 2001;61:8743-50. 275. Conrad C, Miller CR, Ji Y, et al. Delta24-hyCD adenovirus suppresses glioma growth in vivo by combining
276.
277.
278.
279.
280.
281.
282.
283.
284.
285.
286.
287.
288.
186
oncolysis and chemosensitization. Cancer Gene Ther 2005;12:284-94. Lamfers ML, Gianni D, Tung CH, et al. Tissue inhibitor of metalloproteinase-3 expression from an oncolytic adenovirus inhibits matrix metalloproteinase activity in vivo without affecting anti-tumor efficacy in malignant glioma. Cancer Res 2005;65:9398-405. Kruyt FA, Curiel DT. Toward a new generation of conditionally replicating adenoviruses: pairing tumor selectivity with maximal oncolysis. Hum Gene Ther 2002;13:485-95. van Beusechem VW, Doel van den PB, Grill J, Pinedo HM, Gerritsen WR. Conditionally replicative adenovirus expressing p53 exhibits enhanced oncolytic potency. Cancer Res 2002;62:6165-71. Geoerger B, Vassal G, Opolon P, et al. Oncolytic activity of p53-expressing conditionally replicative adenovirus AdDelta24-p53 against human malignant glioma. Cancer Res 2004;64:5753-9. Wohlfahrt ME, Beard BC, Lieber A, Kiem HP. A capsidmodified, conditionally replicating oncolytic adenovirus vector expressing TRAIL Leads to enhanced cancer cell killing in human glioblastoma models. Cancer Res 2007;67:8783-90. Post DE, Sandberg EM, Kyle MM, et al. Targeted cancer gene therapy using a hypoxia inducible factor dependent oncolytic adenovirus armed with interleukin-4. Cancer Res 2007;67:6872-81. Lamfers ML, Fulci G, Gianni D, et al. Cyclophosphamide increases transgene expression mediated by an oncolytic adenovirus in glioma-bearing mice monitored by bioluminescence imaging. Mol Ther 2006;14:779-88. Bieler A, Mantwill K, Holzmuller R, et al. Impact of radiation therapy on the oncolytic adenovirus dl520: implications on the treatment of glioblastoma. Radiother Oncol 2007. Idema S, Lamfers ML, van Beusechem VW, et al. AdDelta24 and the p53-expressing variant AdDelta24p53 achieve potent anti-tumor activity in glioma when combined with radiotherapy. J Gene Med 2007; 9:1046-56. Gomez-Manzano C, Alonso MM, Yung WK, et al. Delta-24 increases the expression and activity of topoisomerase I and enhances the antiglioma effect of irinotecan. Clin Cancer Res 2006;12:556-62. Alonso MM, Gomez-Manzano C, Jiang H, et al. Combination of the oncolytic adenovirus ICOVIR-5 with chemotherapy provides enhanced anti-glioma effect in vivo. Cancer Gene Ther 2007;14:756-61. Whitman ED, Tsung K, Paxson J, Norton JA. In vitro and in vivo kinetics of recombinant vaccinia virus cancer-gene therapy. Surgery 1994;116:183-8. Zeh HJ, Bartlett DL. Development of a replicationselective, oncolytic poxvirus for the treatment of human cancers. Cancer Gene Ther 2002;9:1001-12.
3113
3114
186
Gene therapy for brain tumors
289. Gnant MF, Noll LA, Irvine KR, Puhlmann M, Terrill RE, Alexander HR, Jr, Bartlett DL. Tumor-specific gene delivery using recombinant vaccinia virus in a rabbit model of liver metastases. J Natl Cancer Inst 1999; 91:1744-50. 290. Puhlmann M, Brown CK, Gnant M, Huang J, Libutti SK, Alexander HR, Bartlett DL. Vaccinia as a vector for tumor-directed gene therapy: biodistribution of a thymidine kinase-deleted mutant. Cancer Gene Ther 2000; 7:66-73. 291. Guo ZS, Naik A, O’Malley ME, et al. The enhanced tumor selectivity of an oncolytic vaccinia lacking the host range and antiapoptosis genes SPI-1 and SPI-2. Cancer Res 2005;65:9991-8. 292. McCart JA, Puhlmann M, Lee J, Hu Y, Libutti SK, Alexander HR, Bartlett DL. Complex interactions between the replicating oncolytic effect and the enzyme/ prodrug effect of vaccinia-mediated tumor regression. Gene Ther 2000;7:1217-23. 293. Kim JH, Oh JY, Park BH, et al. Systemic armed oncolytic and immunologic therapy for cancer with JX-594, a targeted poxvirus expressing GM-CSF. Mol Ther 2006;14:361-70. 294. Timiryasova TM, Chen B, Haghighat P, Fodor I. Vaccinia virus-mediated expression of wild-type p53 suppresses glioma cell growth and induces apoptosis. Int J Oncol 1999;14:845-54. 295. Timiryasova TM, Li J, Chen B, Chong D, Langridge WH, Gridley DS, Fodor I. Antitumor effect of vaccinia virus in glioma model. Oncol Res 1999; 11:133-44. 296. Gridley DS, Andres ML, Li J, Timiryasova T, Chen B, Fodor I. Evaluation of radiation effects against C6 glioma in combination with vaccinia virus-p53 gene therapy. Int J Oncol 1998;13:1093-8. 297. Chen B, Timiryasova TM, Andres ML, Kajioka EH, Dutta-Roy R, Gridley DS, Fodor I. Evaluation of combined vaccinia virus-mediated antitumor gene therapy with p53, IL-2, and IL-12 in a glioma model. Cancer Gene Ther 2000;7:1437-47. 298. Sypula J, Wang F, Ma Y, Bell J, McFadden G. Myxoma virus tropism in human tumor cells. Gene Ther Mol Biol 2004;8:103-14. 299. Lun X, Yang W, Alain T, et al. Myxoma virus is a novel oncolytic virus with significant antitumor activity against experimental human gliomas. Cancer Res 2005;65:9982-90. 300. Lun XQ, Zhou H, Alain T, et al. Targeting human medulloblastoma: oncolytic virotherapy with myxoma virus is enhanced by rapamycin. Cancer Res 2007;67:8818-27. 301. Moehler M, Blechacz B, Weiskopf N, et al. Effective infection, apoptotic cell killing and gene transfer of human hepatoma cells but not primary hepatocytes by parvovirus H1 and derived vectors. Cancer Gene Ther 2001;8:158-67.
302. Cornelis JJ, Salome N, Dinsart C, Rommelaere J. Vectors based on autonomous parvoviruses: novel tools to treat cancer? J Gene Med 2004;6 Suppl 1:S193-202. 303. Wollmann G, Tattersall P, Pol van den AN. Targeting human glioblastoma cells: comparison of nine viruses with oncolytic potential. J Virol 2005;79:6005-22. 304. Herrero YCM, Cornelis JJ, Herold-Mende C, Rommelaere J, Schlehofer JR, Geletneky K. Parvovirus H-1 infection of human glioma cells leads to complete viral replication and efficient cell killing. Int J Cancer 2004;109:76-84. 305. Di Piazza M, Mader C, Geletneky K, et al. Cytosolic activation of cathepsins mediates parvovirus H-1induced killing of cisplatin and TRAIL-resistant glioma cells. J Virol 2007;81:4186-98. 306. Abschuetz A, Kehl T, Geibig R, Leuchs B, Rommelaere J, Regnier-Vigouroux A. Oncolytic murine autonomous parvovirus, a candidate vector for glioma gene therapy, is innocuous to normal and immunocompetent mouse glial cells. Cell Tissue Res 2006;325:423-36. 307. Strong JE, Coffey MC, Tang D, Sabinin P, Lee PW. The molecular basis of viral oncolysis: usurpation of the Ras signaling pathway by reovirus. Embo J 1998;17:3351-62. 308. Coffey MC, Strong JE, Forsyth PA, Lee PW. Reovirus therapy of tumors with activated Ras pathway. Science 1998;282:1332-4. 309. Wilcox ME, Yang W, Senger D, et al. Reovirus as an oncolytic agent against experimental human malignant gliomas. J Natl Cancer Inst 2001;93:903-12. 310. Yang WQ, Lun X, Palmer CA, et al. Efficacy and safety evaluation of human reovirus type 3 in immunocompetent animals: racine and nonhuman primates. Clin Cancer Res 2004;10:8561-76. 311. Yang WQ, Senger D, Muzik H, et al. Reovirus prolongs survival and reduces the frequency of spinal and leptomeningeal metastases from medulloblastoma. Cancer Res 2003;63:3162-72. 312. Krishnamurthy S, Takimoto T, Scroggs RA, Portner A. Differentially regulated interferon response determines the outcome of Newcastle disease virus infection in normal and tumor cell lines. J Virol 2006; 80:5145-55. 313. Fabian Z, Csatary CM, Szeberenyi J, Csatary LK. p53independent endoplasmic reticulum stress-mediated cytotoxicity of a Newcastle disease virus strain in tumor cell lines. J Virol 2007;81:2817-30. 314. Phuangsab A, Lorence RM, Reichard KW, Peeples ME, Walter RJ. Newcastle disease virus therapy of human tumor xenografts: antitumor effects of local or systemic administration. Cancer Lett 2001;172: 27-36. 315. Lorence RM, Katubig BB, Reichard KW, et al. Complete regression of human fibrosarcoma xenografts after local Newcastle disease virus therapy. Cancer Res 1994; 54:6017-21.
Gene therapy for brain tumors
316. Bluming AZ, Ziegler JL. Regression of Burkitt’s lymphoma in association with measles infection. Lancet 1971;2:105-6. 317. Grote D, Russell SJ, Cornu TI, Cattaneo R, Vile R, Poland GA, Fielding AK. Live attenuated measles virus induces regression of human lymphoma xenografts in immunodeficient mice. Blood 2001;97:3746-54. 318. Peng KW, TenEyck CJ, Galanis E, Kalli KR, Hartmann LC, Russell SJ. Intraperitoneal therapy of ovarian cancer using an engineered measles virus. Cancer Res 2002;62:4656-62. 319. Blechacz B, Splinter PL, Greiner S, et al. Engineered measles virus as a novel oncolytic viral therapy system for hepatocellular carcinoma. Hepatology (Baltimore, MD) 2006;44:1465-77. 320. Ghali M, Schneider-Schaulies J. Receptor (CD46)- and replication-mediated interleukin-6 induction by measles virus in human astrocytoma cells. J Neurovirol 1998;4:521-30. 321. Duprex WP, McQuaid S, Hangartner L, Billeter MA, Rima BK. Observation of measles virus cell-to-cell spread in astrocytoma cells by using a green fluorescent protein-expressing recombinant virus. J Virol 1999; 73:9568-75. 322. Plumb J, Duprex WP, Cameron CH, RichterLandsberg C, Talbot P, McQuaid S. Infection of human oligodendroglioma cells by a recombinant measles virus expressing enhanced green fluorescent protein. J Neurovirol 2002;8:24-34. 323. Phuong LK, Allen C, Peng KW, et al. Use of a vaccine strain of measles virus genetically engineered to produce carcinoembryonic antigen as a novel therapeutic agent against glioblastoma multiforme. Cancer Res 2003;63:2462-9. 324. Allen C, Vongpunsawad S, Nakamura T, et al. Retargeted oncolytic measles strains entering via the EGFRvIII receptor maintain significant antitumor activity against gliomas with increased tumor specificity. Cancer Res 2006;66:11840-50. 325. Paraskevakou G, Allen C, Nakamura T, et al. Epidermal growth factor receptor (EGFR)-retargeted measles virus strains effectively target EGFR- or EGFRvIII expressing gliomas. Mol Ther 2007;15:677-86. 326. Wei J, Wahl J, Nakamura T, Stiller D, Mertens T, Debatin KM, Beltinger C. Targeted release of oncolytic measles virus by blood outgrowth endothelial cells in situ inhibits orthotopic gliomas. Gene Ther 2007;14: 1573-86. 327. Gromeier M, Lachmann S, Rosenfeld MR, Gutin PH, Wimmer E. Intergeneric poliovirus recombinants for the treatment of malignant glioma. Proc Natl Acad Sci USA 2000;97:6803-8. 328. Gromeier M, Wimmer E. Viruses for the treatment of malignant glioma. Curr Opin Mol Ther 2001;3:503-8. 329. Merrill MK, Bernhardt G, Sampson JH, Wikstrand CJ, Bigner DD, Gromeier M. Poliovirus receptor
330.
331. 332.
333.
334.
335.
336.
337.
338.
339.
340.
341.
342.
186
CD155-targeted oncolysis of glioma. Neuro Oncol 2004;6: 208-17. Ansardi DC, Porter DC, Jackson CA, Gillespie GY, Morrow CD. RNA replicons derived from poliovirus are directly oncolytic for human tumor cells of diverse origins. Cancer Res 2001;61:8470-9. Balachandran S, Barber GN. Vesicular stomatitis virus (VSV) therapy of tumors. IUBMB life 2000;50:135-8. Balachandran S, Porosnicu M, Barber GN. Oncolytic activity of vesicular stomatitis virus is effective against tumors exhibiting aberrant p53, Ras, or myc function and involves the induction of apoptosis. J Virol 2001;75:3474-9. Duntsch CD, Zhou Q, Jayakar HR, et al. Recombinant vesicular stomatitis virus vectors as oncolytic agents in the treatment of high-grade gliomas in an organotypic brain tissue slice-glioma coculture model. J Neurosurg 2004;100:1049-59. Lun X, Senger DL, Alain T, et al. Effects of intravenously administered recombinant vesicular stomatitis virus (VSV(deltaM51)) on multifocal and invasive gliomas. J Natl Cancer Inst 2006;98:1546-57. Kinoh H, Inoue M, Washizawa K, et al. Generation of a recombinant Sendai virus that is selectively activated and lyses human tumor cells expressing matrix metalloproteinases. Gene Ther 2004;11:1137-45. Unno Y, Shino Y, Kondo F, et al. Oncolytic viral therapy for cervical and ovarian cancer cells by Sindbis virus AR339 strain. Clin Cancer Res 2005;11:4553-60. Markert JM, Medlock MD, Rabkin SD, et al. Conditionally replicating herpes simplex virus mutant, G207 for the treatment of malignant glioma: results of a phase I trial. Gene Ther 2000;7:867-74. Rampling R, Cruickshank G, Papanastassiou V, et al. Toxicity evaluation of replication-competent herpes simplex virus (ICP 34.5 null mutant 1716) in patients with recurrent malignant glioma. Gene Ther 2000;7: 859-66. Papanastassiou V, Rampling R, Fraser M, et al. The potential for efficacy of the modified (ICP 34.5(-)) herpes simplex virus HSV1716 following intratumoural injection into human malignant glioma: a proof of principle study. Gene Ther 2002;9:398-406. Chiocca EA, Abbed KM, Tatter S, et al. A phase I openlabel, dose-escalation, multi-institutional trial of injection with an E1B-Attenuated adenovirus, ONYX-015, into the peritumoral region of recurrent malignant gliomas, in the adjuvant setting. Mol Ther 2004;10: 958-66. Csatary LK, Bakacs T. Use of Newcastle disease virus vaccine (MTH-68/H) in a patient with high-grade glioblastoma. Jama 1999;281:1588-9. Csatary LK, Gosztonyi G, Szeberenyi J, Fabian Z, Liszka V, Bodey B, Csatary CM. MTH-68/H oncolytic viral treatment in human high-grade gliomas. J Neurooncol 2004;67:83-93.
3115
3116
186
Gene therapy for brain tumors
343. Wagner S, Csatary CM, Gosztonyi G, et al. Combined treatment of pediatric high-grade glioma with the oncolytic viral strain MTH-68/H and oral valproic acid. Apmis 2006;114:731-43. 344. Schneider T, Gerhards R, Kirches E, Firsching R. Preliminary results of active specific immunization with
modified tumor cell vaccine in glioblastoma multiforme. J Neurooncol 2001;53:39-46. 345. Steiner HH, Bonsanto MM, Beckhove P, et al. Antitumor vaccination of patients with glioblastoma multiforme: a pilot study to assess feasibility, safety, and clinical benefit. J Clin Oncol 2004;22:4272-81.
185 Gene Therapy for Neurological Disorders (Except Oncology) M. G. Kaplitt
Introduction Gene therapy is among the most direct methods for capitalizing upon advances in understanding the biological basis of disease in order to create improved human therapies. This led to premature exuberance for this technology, which was severely dampened following some celebrated difficulties in the overall field. Nonetheless, sober and incremental advances in gene therapy have led to a resurgence in recent years, and this has been led by applications of gene therapy to neurological disorders. The past 7 years has seen numerous trials of gene therapy for neurodegenerative and neurogenetic diseases, including Parkinson’s disease, Alzheimer’s disease, Canavan and Batten diseases. The are in addition to the longer history of applying gene and viral therapies to central nervous system malignancies, which continue to be explored. Nearly all of these studies have involved surgical infusion of the gene therapy agent into specific brain targets, using often home-made infusion devices and a variety of stereotaxic targeting methods. Gene therapy agents do not efficiently cross the blood-brain barrier, and even if this were achieved there remain advantages to restricting gene transfer to one or more isolated brain regions by selective infusion. Therefore, for the foreseeable future, advancing gene therapy in the brain will require active participation of stereotactic and functional neurosurgeons. Here we will review the basic science of gene therapy along with regulatory issues which should be familiar to anyone participating
#
Springer-Verlag Berlin/Heidelberg 2009
in gene therapy clinical trials. Finally, the current state of gene therapy applications which have reached human trials will be reviewed with the expectation that these will only be the beginning of a field that has the potential to revolutionize the practice of stereotactic and functional neurosurgery.
Basic Science of Gene Therapy There are two general types of gene therapy methods, referred to as ‘‘ex vivo’’ and ‘‘in vivo’’ gene therapy. Ex vivo gene therapy involves genetic alteration of cells in culture, followed by transplantation into a target organ in the patient. These cells can be used as in situ factories to produce a therapeutic protein such as a growth factor. In such cases, the cells are chosen to be relatively inert relative to the target organ. This may also be a method to alter the properties of cells which are themselves intended to be therapeutic, such as altering stem cells so that they may increase survival or better interact with the host brain. In vivo gene therapy involves introduction of genetic material directly into cells in the body to modify their function or influence survival. Given the limited efficacy of direct transfer of naked or lipid-encapsulated DNA or RNA, by far the most common method of gene delivery for in vivo gene therapy has been the use of genetically modified viruses as gene transfer vehicles (vectors). Viruses are naturally extremely efficient at transferring their own
3062
185
Gene therapy for neurological disorders (except oncology)
genetic material into host cells, since this is critical to complete the viral lifecycle and allow spread of progeny virions to surrounding cells. Translation of in vivo gene therapy into human clinical trials has only become possible following the development of technologies which retain the property of viral particles for efficient gene transfer while limiting or eliminating any toxicity due to viral replication or inflammatory reactions to viral proteins.
Viral Vectors The first viral vectors used in human CNS gene therapy were retroviral vectors for malignant brain tumors (see below). Retroviruses are RNA viruses which must synthesize a copy of their own DNA upon entry into a cell using a reverse transcriptase enzyme which is brought in with the viral genome [1]. These agents must also integrate within the host cell chromosome for full functionality. Both of these properties require active cell division, so retroviral vector applications in the CNS have largely been limited to tumor therapy (see below) [2]. Since these are very simple viruses with small genomes, cell lines can be created which continuously produce and shed packaged viruses while remaining viable. Therefore, both purified retroviral vectors as well as producer cell lines can be implanted into a target tissue [2–5]. Retroviral vectors have also been used to modify cells in culture prior to transplantation as part of ex vivo gene therapy strategies [6,7]. For human applications in the brain, this has been primarily used in Alzheimer’s disease (see below). Finally, lentiviral vectors have more recently emerged as novel agents for both ex vivo and in vivo gene transfer in the brain. Lentiviruses are RNA-based retroviruses with a fairly unique biology which permits gene transfer into nondividing as well as dividing cells [8–10]. This family of viruses includes the human immunodeficiency virus (HIV), which
raises some safety concerns. However, most lentiviral vectors being used in the brain for pre-clinical studies are based on nonhuman and/or non-HIV lentiviruses, which are then further altered to prevent proliferation and recombination with more dangerous viruses [11,12]. While lentiviral vectors have not yet been used for gene therapy in the human central nervous system, the safety and efficacy profile appears to be sufficiently promising in animals that this is likely to enter human clinical trials in the near future. The major alternative to retroviral vectors are DNA viral vectors. These derive from viruses which use DNA as their genome, which does not require active cell division since the DNA enters the nucleus of the cell and can begin directing expression of a therapeutic gene without further DNA synthesis. Perhaps the earliest DNA virus used for gene transfer in the brain was herpes simplex virus (HSV). HSV type 1 (HSV1) naturally infects the nervous system, and causes disease by infection of peripheral nerves and eventual spread to the central nervous system [13,14]. It is this natural tropism that created early enthusiasm for HSV1 vectors. The first vectors were recombinant vectors, in which one or more genes that are essential for viral replication, spread and/or pathogenesis were removed [13,14]. While these modifications block viral spread and reduce toxicity, these vectors nonetheless retain many viral genes which can be expressed and possibly promote toxicity. However, advances in this technology have led to vectors which are sufficiently improved that they are being moved into clinical application. These applications have primarily focused upon diseases of peripheral nerves, including diabetic neuropathy and pain [15,16]. Alternative forms of recombinant HSV vectors retain the ability to replicate, but remove genes which permit replication in nondividing cells. These ‘‘conditionally replicating’’ vectors proliferate in dividing cells but they cannot complete a life cycle in
Gene therapy for neurological disorders (except oncology)
nondividing cells [17]. As a result, they have been very popular as a therapy for brain tumors, which are a population of dividing cells in a background of nondividing neurons and glial cells (see below) [17–19]. Finally, a variation on the HSV vector is called the amplicon. This vector contains HSV replication and packaging signals but absolutely no HSV genes. When produced in cell culture, HSV gene products are provided by another source, which can then package the amplicon into an HSV coat [20]. The resulting vector has no HSV genes, unlike a recombinant virus. Early in the development of gene therapy for the brain, these were shown to efficiently transfer genes into neurons in cell culture and in living animal brains [21–24]. While these continue to be studied, production remains a bit cumbersome and therefore these have not yet been proposed for human clinical trials [25]. DNA vectors based upon adenovirus have also been used for human gene transfer, again primarily in brain tumor clinical trials. Adenovirus is far smaller than HSV and therefore has fewer genes requiring manipulation [26]. While adenoviral vectors were developed for gene transfer into neurons soon after the evolution of HSV vectors, they also retained viral gene products which could limit safety and longevity of gene expression [27–30]. More recent systems have been developed similar to the HSV amplicon which allows packaging of larger pieces of DNA into an adenovirus package while eliminating adenovirus genes [26,31]. Again this is a bit of a difficult system to use from a technical standpoint, but improvements to this technology over time could make this attractive for clinical use. However, as with HSV vectors, conditionally replicating adenovirus vectors have been created which selectively replicate in dividing cells but not in nondividing cells. These have also been used in clinical trials for brain tumor gene therapy (see below). The final vector system which is the most widely used for gene transfer in the brain for
185
non-neoplastic diseases is the adeno-associated virus (AAV). AAV is by far the smallest of the DNA viruses used as vectors, and AAV has not been associated with human disease [32,33]. As with the HSV amplicon system, AAV vectors contain no native AAV genes but rather package the therapeutic gene of interest into an AAV coat [34–37]. Because of the small size and limited number of AAV genes, systems have been developed which allow efficient production of packaged AAV vectors without generating contaminating wild-type viruses. Further development of this technology has led to generation of vectors of sufficiently high concentration and purity as to facilitate use in human clinical trials. The original serotype of AAV used as a gene therapy vector, AAV2, appeared to be highly efficient for gene transfer into neurons in the brain and yielded long-term stability, safety and improvement in an animal model of Parkinson’s disease [35]. This same serotype has now proven to be safe and stable in primate brain for many years, further supporting the utility of this system for gene therapy of disorders such as neurodegenerative diseases which will require ongoing therapy for many years [38]. Based upon these factors, AAV2 is the only viral vector which has now been used in human clinical trials of gene therapy for neurological diseases other than brain tumors (see below). In recent years, numerous other serotypes and variants of AAV vectors have evolved which have different profiles of affinity and efficiency in various cell types, and these may provide greater flexibility for human gene therapy in the future [37].
Regulatory Approval Gene therapy is among the most highly regulated areas of human research worldwide. In the United States, any new gene therapy study (as with any other type of new drug or device) must be approved and regulated by the U.S. Food and
3063
3064
185
Gene therapy for neurological disorders (except oncology)
Drug Administration (FDA). In addition, prior to FDA approval any study involving introduction of recombinant genetic material (DNA or RNA) into humans for any purpose must first be reviewed by the Recombinant DNA Advisory Committee (RAC), which is an advisory committee of the National Institutes of Health (NIH) Office of Biotechnology Activities (OBA). This must be completed even if the protocol is funded by NIH, although as with any federally-funded clinical trials, additional oversight is usually required by the relevant Institute funding the trial. At the institutional level, gene therapy protocols must be approved by both the local Institutional Review Board (IRB) and by a separate group responsible for oversight of biological agents and also regulated by NIH, called the Institutional Biosafety Committee (IBC). Even for the seasoned investigator, navigating the regulatory process to obtain approval and satisfy oversight for human gene therapy can be daunting. Although some issues in this process have not been fully specified in federal guidelines, the requirements for gene therapy clinical trial approval and suggestions for a standard approach to this process are outlined below. Although ethical considerations surrounding gene therapy have been widely studied and published in peerreviewed journals, there are few published references which provide the comprehensive guidance through the regulatory approval process provided here. Since nearly every trial to date of gene therapy in the human brain has involved neurosurgical infusion of the gene therapy agent, neurosurgeons have been directly involved in these studies and in many cases were either the principal investigators or key site investigators for multi-center studies. Therefore, a review of the current regulatory process is provided as a guide for both new investigators who may wish to initiate a study as well as those who may wish to gain greater knowledge prior to participating in a multi-center study. Most of the regulatory information contained in this section is based upon guidelines published on federal
government websites which will be enumerated and/or on personal experience of the author.
RAC Review For any new gene therapy protocol, the most appropriate first step is usually submission of a protocol for RAC review. Details of guidelines for this process are available at http://www4.od.nih. gov/oba/. There is no specifically required order of protocol submission, and therefore neither RAC review nor FDA approval must precede any of the institutional reviews. However, given the diverse number of experts and possible general public input which is likely to occur with RAC review of a new submission, it is unlikely that the protocol emerging from this process will be sufficiently similar to the original submission to avoid a repeat review by institutional oversight committees. Therefore, in general the best use of time and effort is to usually submit a protocol for RAC review first, then proceed with FDA review and approval, followed by institutional review. It has been our experience that few significant changes are requested by institutional reviewers once a protocol has emerged from the extensive RAC and FDA processes. The RAC process is referred to as review, because it is the only one of the agencies with oversight of gene therapy studies which does not have statutory authority to approve or deny a new protocol. The most recent revision of the NIH guidelines governing recombinant DNA research specifies that new protocols involving transfer of recombinant DNA (or RNA deriving from recombinant DNA) into humans must be reviewed by RAC and that review process must be completed before enrolling a new patient. However, the authority to approve gene transfer protocols was removed from this group many years ago. The current requirement is that a review process be completed but the results of that process are simply information available to the public and to other regulatory agencies which
Gene therapy for neurological disorders (except oncology)
do have the authority to approve or deny a new protocol. The RAC review process involves submission of a new protocol to the committee. Within 15 days a notification will be sent indicating whether full, public review will be completed. The RAC review process is public, with meetings open to the public and broadcast on the web. Protocols are available on the web, and journals have even published protocols submitted to the RAC before discussion or any revision has occurred. The RAC even specifically prohibits submission of trade secrets or other information which cannot be divulged without specific approval, since this would obviate the intent of a comprehensive public review. Therefore, it must be recognized when submitting to the RAC that, unlike an NIH grant application which is considered confidential during the review process, any information within the protocol should be considered public once submitted. While wellintentioned, this can also create confusion among the public and potential patients, particularly if a new protocol is made public but is then extensively changed in response to RAC and FDA review. Not all protocols are required to be discussed at a public meeting. If the committee feels that the protocol does not present sufficiently novel gene therapy issues (such as new disease, new gene transfer vector, new gene, potential safety concerns based upon data presented, etc. . .), then the protocol may be reviewed without a public discussion. This usually expedites the process, since a lead time of at least 2 months is generally required prior to a scheduled RAC meeting for protocols which require public discussion. The RAC review process is considered complete after one of two possible letters is received by the Principal Investigator. If no public review is required, then a letter to that effect completes the process. If public review is required, then the protocol is discussed at a meeting, where the investigator(s) usually have a brief initial opportunity to clarify points in the protocol or provide
185
additional information (simply summarizing the protocol is usually neither required nor desirable). Questions may then be asked of the investigators, and reviewers will also provide comments. This meeting is open to the public, so there is finally an opportunity for anyone in attendance to ask questions or comment. After this meeting, receipt of a letter from the RAC summarizing the review and the comments from the meeting completes the public review process. One of these letters must be obtained before enrolling a subject in a new gene therapy protocol. Following RAC review, there are ongoing oversight and reporting responsibilities to the RAC once the trial has been initiated. Documentation to NIH OBA of relevant IRB and IBC approvals, along with the FDA Investigational New Drug (IND) application number and any protocol changes required by FDA (and NIH grant number if federally-funded) must also all be provided within 3 weeks of enrolling the first patient in the study, regardless of whether or not the patient has been treated. An annual report summarizing the protocol to date, including enrollment, adverse events and any available results must be submitted. An expedited reporting is required for any serious adverse events which is fatal or life-threatening (regardless of association to the treatment), or is both unanticipated and related directly to the gene therapy agent. This report must be provided within 7 days of occurrence. While other types of serious adverse events must be reported rapidly to local IRBs, data safety monitoring boards and FDA, only these types of serious adverse events require expedited reporting to RAC outside of the regular annual report.
FDA Approval For any new gene therapy study in the United States, approval of even a phase I protocol by the FDA is required before enrolling and treating patients. Therefore, FDA has statutory authority
3065
3066
185
Gene therapy for neurological disorders (except oncology)
to approve or prevent a gene therapy study, in addition to oversight responsibilities. Most investigators and/or sponsors usually complete the RAC review process before submission to FDA. An FDA representative usually attends RAC meetings and they consider the ensuing public discussion among a multitude of scientific and medical experts carefully. The likelihood of timely review and approval of a new gene therapy protocol by FDA is usually greatly enhanced by completion of RAC review and modification of the protocol as appropriate based upon the review process. It should be noted that there is no requirement to incorporate every issue raised at the RAC review into a protocol modification prior to submission to FDA, particularly if investigators believe that a reasonable argument can be made for disagreeing with a suggestion from the RAC. In most cases, however, if there is consensus among the RAC on a particular point after open discussion (usually with comments and responses at the time from the investigator), then it is likely that the issue will be taken seriously by FDA as well. It is technically possible but usually ill-advised to attempt FDA review prior to RAC review. Responsibility for review, approval and oversight of cellular and gene therapy clinical trials within FDA lies with the Office of Cellular, Tissue and Gene Therapies (OCTGT) in the Center for Biologics Evaluation and Research (CBER). CBER has responsibility for a wide-variety of biologicals, including vaccines and devices related to biological agents (more on this below) but OCTGT is the group which specifically evaluates and regulates gene therapy protocols. The details of FDA submission, review and approval requirements for gene therapy clinical trials can be found at http://www.fda.gov/CBER/gene.htm. To date, no gene therapy agent of any kind has been approved for sale in the United States for any disease, so this process has still not matured to the level of other regulatory processes within the FDA. Therefore, certain assumptions and unknowns remain
regarding the disposition of late phase gene therapy trials as well as final approval procedures, since modifications of the current regulations may become necessary when one or more agents eventually reach this milestone. To date, however, several gene therapy approaches to neurological disease have passed initial phase I studies, so therefore these issues will become pressing in the near future. The review of new human gene therapy protocols by CBER usually involves three components performed by three different reviewers. The first is a review of the preclinical data supporting the trial. This is a fairly standard review which is similar to review of a manuscript or a research grant, but in some ways is far more rigorous than any of those review processes. The goal is to determine if the preclinical data supports the claims made in the protocol. Although some compelling case supporting the potential efficacy of the proposed agent must certainly be made using this data, the major focus is upon safety considerations which could result in harm to patients. Therefore, a detailed accounting of each individual animal used in supporting preclinical studies must be provided and not simply summary data. This is often not anticipated by the investigator who has not been through this process previously, even if they are very accomplished in publication and grantwriting. If an animal study is performed for a potential new gene therapeutic, therefore, this should be considered early, since it can be very costly and time-consuming to repeat one or more animal studies to obtain detailed data on individual animals that may be absent from otherwise very compelling efficacy studies. Data which is often requested includes regular recording of animal weights, temperatures, any unexpected deaths and necropsy data to determine tissue distribution of any gene therapeutic (out of concern for possible oncogenicity, teratogenicity or transmission to other people or to an unborn fetus). If unexpected deaths or disease
Gene therapy for neurological disorders (except oncology)
occur in animals in an individual study, these must be explained and cannot simply be ignored even if a second study failed to identify such issues. Detailed immunological data on some subjects are also usually required, although this is not usually necessary for every animal in every study. Finally, of particular relevance to neurological gene therapy studies is the requirement for primate or other large-animal studies. Even if there is no primate model available for a particular disease, it is generally important (although not necessarily required) to perform at least a safety study in primates using the same gene therapy formulation planned for a human study, usually at a dose far greater than the top dose proposed in the human protocol. The second component is a review of the clinical protocol, usually by an FDA physician with a background in neurological disease. This review extensively evaluates the proposed clinical protocol in all aspects, such as inclusion and exclusion criteria, indications for the trial, possible adverse events and methods for responding to such events, treatment and follow-up plans and adequacy of planned enrollment. This is a straightforward but sometimes time-consuming process and it is usually advisable to maintain flexibility and consider all issues carefully if the goal is to proceed with a reasonable trial. For anyone who has been through an FDA review of any other type of new therapy, this review would probably the most familiar since it is a component of almost any FDA review process. Finally, there is a review of the product specification. This focuses upon the procedures for creating, purifying and certifying the gene therapy agent. The requirements for creation of a gene therapeutic for human use are quite different than those for generation of a gene therapy agent for use in a laboratory study. For example, production of a gene therapeutic for late-phase trials (and eventually for sale if this is eventually achieved) required use of a facility that is certified for Good Manufacturing Practices (GMP).
185
Creation of such a facility can be very costly, since there are various requirements for air filtration, barriers to contamination, oversight and certification. Many such facilities exist, however, and capacity is currently well beyond need so a protocol can usually be followed at a pre-existing facility. For phase I studies, however, the gene therapy agent is not required to be generated within a GMP facility. GMP practices must still be followed, however, so that the procedures must be ‘‘GMP-like.’’ This permits a non-GMP lab to generate a reagent for human use in a phase I trial without the burden and expense of either creating a GMP facility or adapting protocols to an existing GMP facility. Other aspects of the product specification to be reviewed include assays used to determine and certify the potency of the agent (usually a titer of a viral vector for gene therapy) and criteria and procedures for sterility testing and ultimate release of the agent for human use. This often requires detailed technical data that are not necessarily the type of information that most basic scientists are enthusiastic about (and is often difficult to publish in a journal), yet this is an absolutely critical step to obtaining approval of a gene therapy agent. Problems with the product specification can delay initial approval of a phase I study, but this can be particularly difficult as later stages of the approval process proceed. While a single lot is usually used for phase I and most phase II studies, for eventual sale it must be recognized that only a very narrow potency range can be tolerated between lots, and very robust criteria, protocols and assays are necessary for lot testing to ensure the stability of the therapy between patients. Therefore the product specification requires particular attention as gene therapy protocols emerge from a successful phase I study and move toward later stages and eventually possible sale. Initially, all FDA protocols are considered on ‘‘hold’’ until review is complete. To begin a phase I study, the protocol must be taken off ‘‘hold.’’
3067
3068
185
Gene therapy for neurological disorders (except oncology)
Even after this, the FDA may still request additional information or modification, but those issues are not considered of sufficient risk or concern to further delay initiation of the study. Any serious adverse event, regardless of anticipation and possible relationship to the study agents, must be reported to the FDA in a timely fashion. The study can continue unless a stopping criteria has been reached, or unless notification is received from the FDA that the study has been placed back on ‘‘hold.’’ At that point, no further patients can be treated or enrolled until the study is taken off ‘‘hold’’ again. However, the obligation to continue to follow, treat and report on patients already treated in the study remains. Following completion of a phase I study, subsequent phases are submitted as amendments to the original IND application and remain under the original IND application number. A study is automatically considered to be on ‘‘hold’’ again once all of the patients proposed in the original protocol have been treated. Additional patients, whether as an extension of the phase I study or as part of a later-phase study, can only be treated once an addendum has been submitted and the study is again considered to be off ‘‘hold.’’
Standard IRB approval, oversight and reporting is necessary for all gene therapy protocols. IBCs are constituted as a local, institutional arm of the NIH OBA, which has oversight over these committees. General information on the role of IBCs and NIH oversight is available at http://www4.od.nih.gov/oba/IBC/IBCrole.htm, although it is encouraged to contact the institutional IBC for details of the local review and approval process. Unlike RAC review, however, the IBC must also review and approve the gene therapy protocol, as they must review any research protocol of any kind within the institution which involves recombinant DNA. Since the expertise and role of the IBC is in the area of recombinant DNA and biological agents, this review generally focuses upon the gene therapy agent (usually a viral vector), although the IBC can raise any issue of concern regarding any component of the protocol. Again, if an extensive RAC review and FDA review has occurred and concerns been thoughtfully addressed, substantive new concerns at the IBC level are unusual. As indicated above, documentation of IRB and IBC approval must be submitted to NIH OBA within 20 days of enrolling the first patient in the study.
IRB and IBC As with any experimental therapy, all gene therapy studies must be reviewed and approved by the local IRB. As indicated earlier, in most cases it is most efficient and effective to initiate this process after RAC review and FDA approval. If the IRB raises significant substantive concerns which were not raised in either process, and these must be addressed with changes in the protocol and/or consent form, then submission of these changes to the FDA as a protocol addendum is usually necessary. However, given the detailed nature of the two processes outlined above (which are usually not part of most standard IRB submissions), this is very unusual.
Human Clinical Trials of Gene Therapy for CNS Diseases Brain Tumors Malignant gliomas are among the most lethal diseases known. While upfront resection and radiation can significantly increase lifespan and quality of life, 5-year survival rates remain negligible [39]. Given the great need for novel therapies to treat malignant brain tumors, it is not surprising that gene therapy was first studied for neoplastic brain disorders. The unique features of brain tumors which limit efficacy of current therapies include invasion of tumor cells into
Gene therapy for neurological disorders (except oncology)
normal brain with an intact blood-brain barrier limiting access to some therapies and the inability to sacrifice substantial normal brain tissue in an attempt to eradicate these diseases. Malignant glioma cells are also often relatively slowly dividing, so intermittent radiation or chemotherapy treatments which target processes involved in cellular replication may be less effective in these tumors at any given moment [40,41]. Gene therapy thus presents an opportunity to continuously expose tumor cells to anti-neoplastic agents within the tumor or surrounding normal brain, potentially bypassing these limiting factors [42–44]. One approach to gene therapy expands upon more traditional chemotherapy strategies by delivering pro-drug activating genes to tumors. These convert otherwise inert molecules into anti-neoplastic cytotoxic agents within the tumor. The first and most widely studied gene in this class was the herpes simplex virus (HSV) thymidine kinase (TK) gene [45,46]. This is the target of antiviral therapies such as acyclovir and gancyclovir. Once modified by HSV TK, these nucleotide analogs can incorporate into growing viral DNA chains and block elongation. They are not specific to viral DNA, however, and can incorporate into replicating cellular DNA. For actively dividing cells, such as neoplastic cells, this can be lethal. This toxicity is substantially reduced in nondividing or terminally differentiated cells, since there is no active DNA replication. Thus, HSV TK gene therapy should lead to selective toxicity of gancyclovir for dividing tumor cells while sparing nondividing normal cells, such as neurons and quiescent glial cells. A ‘‘bystander effect’’ has also been observed when prodrug activated in one cell can diffuse to other cells which may not have received the gene transfer, potentially amplifying the antitumor efficacy of this treatment [47,48]. Initial trials of HSV-TK involved transplantation of retrovirus producer cells into tumors. The belief was that this would provide a higher
185
level of vector production and gene transfer in tumor cells compared with infusion of purified retroviral vectors expressing HSV-TK. Outcomes of these studies were not particularly promising, with reasonable safety profiles but little convincing evidence of reproducible efficacy [49–54]. A similar lack of efficacy was seen in a phase III study as well, but in this case there did appear to be an increase in the rate of intracranial hemorrhage and other complications in the gene therapy patients compared with controls [55]. In order to improve efficiency of HSV-TK gene transfer to tumor cells, subsequent studies used intratumoral infusion of adenoviral vectors as gene transfer agents [56–59]. Animal studies and human clinical trials for malignancies outside of the brain have suggested that the transduction efficiency of adenoviral vectors is far superior for neoplastic cells compared with retroviral vectors, particularly since these will function in cells at any stage in the cell cycle while retroviral vectors will only transduce the actively dividing population. Results of isolated studies as well as a head-to-head comparison with retroviral vectors have suggested that the adenoviral gene transfer approach has greater promise. In addition, a randomized study has indicated that adenoviral HSV-TK gene therapy can increase survival following tumor resection compared with control patients undergoing resection followed by conventional therapy [42,44,56–59]. Studies to further examine the potential utility of this approach remain ongoing. Another strategy to selectively destroy dividing tumor cells against a background of largely nondividing cells in the brain is through oncolytic virus therapy [60,61]. Many viruses destroy cells following replication, while spread of replicated virions to surrounding susceptible cells can amplify the cytotoxic effect. To complete their lifecycle, however, many viruses harbor genes which are required for full virulence in certain cell types. Therefore, removal of these genes will block reproduction and cytotoxicity
3069
3070
185
Gene therapy for neurological disorders (except oncology)
within these cells, while retaining these abilities within other cells. Such viruses can therefore be engineered to replicate in and destroy tumor cells, while sparing surrounding normal brain cells, yielding a biologically-effective, self-amplifying anti-tumor agent. Strictly this is not gene therapy but viral therapy, since the goal is not to transfer a particular therapeutic gene into tumors but rather to use the selective replication of viruses specifically within tumor cells to destroy them while sparing normal tissue. Herpes simplex virus (HSV) was the first virus to be modified and used as an oncolytic virus specifically designed for malignant brain tumors [62]. HSV contains two genes which help produce nucleotides for incorporation into the new viral genomes, thymidine kinase (TK) and ribonucleotide reductase (RR). TK is the same gene used for many of the prodrug activating strategies, and early pre-clinical studies utilized viruses with TK mutations. However, retention of the TK gene seemed to retain greater potency within the virus, and it also provided the opportunity to combine prodrug activation therapy with oncolytic virus therapy. Therefore, the clinical trials using conditionally-replicating HSV viruses have used agents harboring deletions in the RR gene. The first such trial used the G207 virus, which also contained a mutation in a second gene called g34.5[63]. This is an unusual gene product which is fairly unique to HSV, and appears to be essential for full HSV virulence within neurons. Unlike RR or TK mutants which select for dividing cells as compared with nondividing cells, deletion of g34.5 retains full toxicity within non-neuronal cells but reduced virulence in neurons. A second virus, HSV1716, has also been tested in clinical trials, but this harbors deletions in the two copies of the g34.5 gene but retains wild-type RR [64]. Theoretically this variant should be more virulent and more potent as an anti-tumor agent than G207, since it should efficiently infect and replicate in nondividing or slowly dividing
glioma cells as well as rapidly dividing cells. But this also potentially increases toxicity and the likelihood that this would replicate in normal non-neuronal cells. Results with both agents were fairly encouraging, with good safety profiles, and reduced tumor volume radiographically and/or extended lifespan in isolated patients [63,64]. Ongoing development of oncolytic HSV vectors to add new genes which may minimize toxicity and/or improve efficacy should help to further advance the therpeutic potential of this approach. A variety of viruses in addition to HSV have also been used as oncolytic agents in human brain tumor trials. ONYX-015 is an adenovirus variant, haboring a deletion in the E1B-55K gene. Since E1B-55K normally blocks activity of the p53 tumor suppressor gene, ONYX-015 can only replicate in tumor cells which do not have a functional p53 gene, but cannot complete its lifecycle in normal cells expressing functional p53 [65]. This has been extensively studied in a variety of cancers, with particular promise in head and neck cancers. Clinical testing of ONYX015 in brain tumors revealed a good safety profile but little evidence of therapeutic efficacy [65]. Further modifications of this virus, such as addition of other therapeutic genes as is being studied for HSV, may improve the potential of this approach for the future. A novel approach to oncolytic virus therapy for CNS tumors uses the avian Newcastle Disease Virus (NDV). Unlike other gene therapy studies for brain tumors, this oncolytic virus is administered via systemic intravenous infusion, and the virus then appears to enter tumors efficiently from the bloodstream and selectively replicate in and destroy neoplastic cells within the brain. Recently this has been tested in a phase I/II study in 11 patients with glioblastoma [66]. No significant toxicity was encountered and no maximum tolerated dose was identified, despite evidence of profound immune responses to the virus within 1 month of treatment. There was evidence of a good
Gene therapy for neurological disorders (except oncology)
response to therapy in several patients, with one patient having complete regression of tumor. If this were ratified in more rigorous follow-up studies, it would be a shift in traditional approach to brain tumor gene therapy, providing a nonsurgical option for patients. The final approach to gene therapy for brain tumors is more traditional viral vector-mediated gene transfer. This is similar to the prodrug activating strategies, but here genes are transferred into tumor cells which on their own are expected to be therapeutic without need for adjuvant drug therapy. These have generally expressed either a tumor suppressor gene or cytokine gene to modulate the activity of either the tumor cell or the immune response to tumors. These have been used in a variety of tumors, and have been tested in mostly early phase studies for brain tumors as well. Gene therapy using interleukins and GM-CSF have all been tested with generally good safety profiles, but only limited evidence of efficacy [42,43,67–70]. However, these have also been tested in combination with either oncolytic virus therapy or adjuvant chemotherapy or radiation and may show greater promise. Introduction of a normal copy of the p53 gene into tumor cells has also been tested in a clinical trial, based upon the belief outlined above that most high grade gliomas have p53 deletions [71]. While well tolerated, this has not shown substantial efficacy and has not been aggressively pursued. In all likelihood, limited diffusion of vectors into normal brain prevents transduction of cells invading normal tissue, and some tumor cells are malignant despite the presence of wild-type p53, and both of these features can limit the efficacy of this approach. Of historical interest, while no gene therapeutic of any kind has yet been approved for general use in any Western country, adenovirus-p53 has been approved in China for use as an anti-cancer agent for a variety of tumors, making this the first gene therapeutic ever approved for sale in any country [72].
185
Parkinson’s Disease Following neoplastic disease, there have been more human clinical trials of gene therapy for Parkinson’s disease (PD) than for any other disorder. Three different approaches to gene therapy for Parkinson’s disease have already been tested in early clinical trials, and other approaches are entering the clinic as well. The loss of dopaminergic neurons in the substantia nigra pars compacta in PD and resulting reduction in striatal dopamine transmission leads to dysregulation of the downstream basal ganglia circuitry which regulate movement [73,74]. While the cause of this cell loss is under active investigation, the anatomical and functional consequences of this pathology is better understood than most other neurological diseases. This has led to a particular enthusiasm for developing rationally designed therapies for PD, including biological, cell and gene-based therapies. In fact, the first demonstration that AAV vectors can be safe and effective for long-term gene transfer in the brain was performed in an animal model of PD nearly 15 years ago [35]. Several earlier clinical trials of biological treatments other than gene therapy have provided additional information to support the development of gene therapy. Cell transplantation has received strong interest among the general public based upon the hope that this may ultimately repair the damage caused by PD. While stem cell therapy remains a somewhat distant hope, multiple trials of fetal cell transplantation into the human striatum have been published [75,76]. Overall these have been considered failures, although this was in part due to how the investigators defined the primary endpoints. In fact, certain endpoints in these trials did reveal significant improvements in some motor parameters. Furthermore, subgroups of patients appeared to respond better to these therapies, such as younger patients. These studies also highlighted the importance of functional
3071
3072
185
Gene therapy for neurological disorders (except oncology)
imaging, since PET scans demonstrated that the fetal dopamine cells in most cases did survive and continue to produce dopamine in the striatum [75–78]. These scans also suggested that the location of transplantation within the striatum may have influenced outcome. The development of ‘‘off-period’’ or ‘‘runaway’’ dyskinesias, which are involuntary dopaminerelated hyperkinetic movements in the absence of medication, also highlighted the potential concern with unregulated continuous high-level dopamine production in some patients regardless of the mechanism of dopamine production in the striatum. Another concern is a general perception in the neurology community of a possible placebo effect in PD clinical trials, although this has not been observed in all controlled surgical trials of cell or growth factor infusion [75,76,79]. Placebo patients in a surgical trial can be particularly challenging and controversial, but again the transplant studies demonstrated successful and ethical mechanisms for incorporating such groups to enhance the confidence in resulting data among clinicians. In particular, the use of partial-thickness burr holes minimizes the risk of any intracranial injury in control patients since the inner table of the skull is not violated, so that patients and evaluators cannot distinguish placebo from treatment yet the risks of being a control patient are thus substantially minimized. As an alternative to cell transplantation, multiple trials of infusing the powerful dopaminergic trophic factor glial derived neurotrophic factor (GDNF) have been completed. The goal is to prevent further neuronal loss while encouraging re-innervation of the striatum from the survival intact dopaminergic terminals within the striatum. Intraventricular infusion of the recombinant GDNF were stopped due to excessive complications largely related to toxicity from exposure of periventricular structures [80]. This led to direct infusion of GDNF into the striatum, with very encouraging safety and efficacy results from an open-label phase I study [81]. This
unfortunately did not translate into success in a phase II double-blind, placebo controlled study [79]. Interestingly, there was not a significant placebo effect in this study but rather a minimal therapeutic effect which lead to failure. Along with minimal placebo effects in some transplant studies, this raises the question as to whether such confounds truly influence surgical studies to the same degree as drug studies. Nonetheless, concerns regarding possible placebo effects remain in the neurology community, so it is anticipated that most later-phase gene therapy studies will still incorporate a placebo control design. There is also controversy regarding the reason for failure in the phase II study, with some concern that a change in the type of infusion catheter and mechanism of infusion may have adversely influenced the therapy and led to a false negative result. While the objective evidence to support this hypothesis remains scant, nonetheless this highlights the importance of careful study design particularly in surgical trials which can be difficult to repeat due to limited patient access and greater risk than drug trials. It is essential to match as closely as possible the design of early phase studies in later trials, since important changes in design such as method of surgical delivery can not only confound the study but may also make it impossible to determine the ultimate reason for negative data if a late phase trial fails. Currently, the company controlling the rights to GDNF has stopped all studies of recombinant GDNF infusion and has not supported development of gene therapies using this growth factor as well. With all of this in mind, three approaches to gene therapy for PD are currently in clinical testing. Infusion of an AAV vector with the glutamic acid decarboxylase (GAD) gene into the subthalamic nucleus (STN) was the first in vivo gene therapy approach to be tested in adult human patients for any neurodegenerative disorder [82]. GAD is the rate-limiting step in the synthesis of GABA, the major inhibitory neurotransmitter in the brain [83]. Loss of striatal
Gene therapy for neurological disorders (except oncology)
dopamine leads to a reduction of GABAergic transmission into the STN as well as into STN targets such as the globus pallidus interna (GPi) and substantia nigra reticulata (SNr) [84–87]. The hyperactivity of the STN in PD leads to increased glutamatergic activation of these same targets, which further exacerbates this dysregulation. AAV-GAD gene therapy should thus allowing transduced STN neurons to begin producing GABA, which would then be released locally as well as through efferent connection to targets such as GPi and SNr. This was designed therefore to be not only a local therapy restoring GABA levels in the STN, but as a treatment which would take advantage of the efferent STN connection to re-establish GABAergic tone to distal areas within the basal ganglia circuit which are pathologically active. In rodent models, the validity of this theory was confirmed by using microdialysis to demonstrate an increase in evoked GABA release into the SNr following STN AAV-GAD gene therapy [82]. Also, since glutamate is the substrate for GABA production by GAD, this would simultaneously reduce glutamate transmission from the STN by shunting this to GABA production. This approach attempts to capitalize on successful STN surgery in human PD, such as deep brain stimulation (DBS) and lesioning, in order to potentially reduce the risk that a promising pre-clinical therapy would not effectively translate into humans [84,88,89]. It is largely a symptomatic therapy, since it does not aim to repair damage caused by neurodegeneration. However, this may have a neuroprotective function in addition to the symptomatic effect which could slow further loss of nigral dopamine neurons after treatment. There is evidence that excessive glutamatergic input to the nigra in PD can accelerate dopaminergic neurodegeneration through an excitotoxicity mechanism, and preclinical studies do in fact suggest that AAV-GAD can block neurodegeneration in animal models of PD [82]. A phase I study of AAV-GAD gene therapy for PD has recently been completed and detailed
185
results have been reported [90,91]. This study involved 12 patients with advanced PD who would normally have met criteria for deep brain stimulation. Since this was the first time that in vivo gene therapy was attempted for an adult neurological disorder, only unilateral therapy was approved based upon the belief that an unanticipated adverse event might be more devastating if an injury occurred to the same structure bilaterally. All patients had bilateral disease, but most were asymmetric with one body half more symptomatic than the other side so the more symptomatic brain hemisphere was treated. Although unilateral therapy was not originally proposed but was required following regulatory review, the result was that the untreated symptomatic hemisphere was available in each patient for both clinical and functional imaging comparison. Therefore, although traditional phase I studies are designed purely to measure safety and have no controls, this study was in fact controlled as the untreated hemisphere in each patient was available for both clinical and functional imaging comparison with the treated hemisphere. Infusion of AAV-GAD into the STN was safe at the doses tested in this study, which was the primary endpoint of this phase I study [90]. There were no significant adverse events related to the therapy throughout the 1-year follow-up. Aside from safety, however, there was also a significant functional improvement as measured by part III of the standard and widely-used Unified Parkinson’s Disease Rating Scale (UPDRS). The total body score significantly improved beginning at 3 months after surgery and continued for 1 year, both ‘‘off ’’ and ‘‘on’’ medication. When analyzed by body side, the effect was largely restricted to the hemibody opposite the treated hemisphere in both conditions over the same time period. Interestingly, there was no significant change at 1 month following surgery. While it has been suggested that any effect could be due to surgical intervention, lesioning usually results in immediate improvement, while delayed
3073
3074
185
Gene therapy for neurological disorders (except oncology)
improvement from 1 to 3 months is more consistent with the known biology of AAV, which usually requires several weeks to maximize gene expression. Functional imaging with flourodeoxyglucose (FDG) PET scans provided additional information on the physiology of motor circuits in these patients over time [91]. FDG measures neuronal metabolism, primarily at axonal terminals, and has been used to understand changes in brain function following accepted drug and surgical therapies [88,89,92–95]. This further demonstrated significant improvements in abnormal metabolism over time in several key structures, including the thalamus and cerebral cortex, in a manner consistent with results from more standard therapies and restricted only to the treated hemisphere [90]. To further address questions of mechanism, a subsequent study compared FDG PET from AAV-GAD gene therapy patients with patients who underwent therapeutic subthalamic nucleus lesioning [91]. While the effect on the thalamus was similar, the effect on the GPi (which intervenes between the STN and thalamus) were completely opposite, in a manner consistent with the different potential mechanism of action of gene therapy and lesioning. Thus, while the result on downstream structures such as the thalamus were similar, the mechanism for achieving this result differed, further indicating that outcomes following gene therapy were not due to a lesioning effect. These studies not only demonstrate the safety and potential efficacy of gene therapy, but highlight the value of careful functional imaging as adjunctive outcome measures to help ratify clinical findings and suggest mechanisms of action which may address outcomes questions. This approach recently initiated a phase II trial which involves a blinded, control group with a partial-thickness burr hole to investigate possible placebo effects which were not addressed in the phase I study. The second gene therapy approach to have entered clinical trials for PD uses a gene for a
growth factor to try to slow disease progression and promote plasticity of dopamine neurons within the striatum. Despite substantial preclinical evidence supporting gene therapy using GDNF, development of GDNF gene therapy for human use was complicated following the difficulties with the recombinant GDNF trials since the company controlling the intellectual property for this gene chose not to support further GDNF therapies. Neurturin, however, is a growth factor in the GDNF family, which can activate GDNF receptors and have similar trophic effects [96–98]. As with GDNF, neurturin gene therapy has resulted in impressive efficacy and safety profiles in both rodent and primate pre-clinical studies [96–98]. Using again an AAV vector as the gene transfer vehicle, neurturin gene therapy has now been tested in human clinical trials. In the phase I study, patients with advanced PD were divided into two groups, receiving bilateral striatal infusion of either low or high concentration of AAVneurturin. As with AAV-GAD, this appeared to be safe with no clear adverse events related to the therapy. Results have not yet been published, but the study has been completed and results presented at scientific meetings have suggested significant improvements in numerous clinical ratings in treated patients, including ‘‘off ’’ medication UPDRS as well as improvements in total ‘‘on’’ time. This involved infusion at Six sites within the striatum through three surgical tracts, which was also well tolerated. Although the major promise of growth factor gene therapy is as a disease modifying agent, these data and ongoing studies are primarily designed as therapeutic efficacy studies rather than neuroprotection studies, since proof of neuroprotection will require considerably more time and far larger numbers of patients. However, given the strong safety and encouraging efficacy profile, a phase II study is ongoing and was the first gene therapy approach for Parkinson’s disease to enter phase II. This involves a 2:1 design, with two treated patients
Gene therapy for neurological disorders (except oncology)
for each control, with controls again receiving only partial-thickness burr holes. At present, enrollment in this study has been completed and follow-up is planned for 1 year, after which results will be unblinded and analyzed. The final gene therapy approach currently in human testing involves expression of dopamine biosynthetic genes locally within the striatum. Although early studies in animals explored unregulated synthesis of dopamine (and very recently a clinical trial in Europe began to explore such a concept in humans), results of the fetal transplantation studies along with complications of excessive dopamine therapy in advanced patients led to a desire to use gene therapy in a manner which would permit regulated control of dopamine synthesis. To achieve this, transfer of the aromatic acid decarboxylase (AADC) gene into the human striatum has been under investigation. AADC converts the L-dopa to dopamine, and is the enzyme which is imaged with F-dopa PET [99]. While this is not the rate-limiting enzyme in the synthesis of dopamine (this is the property of tyrosine hydroxylase, which synthesizes L-dopa), increasing striatal AADC levels has been shown in animal studies to improve both the magnitude and longevity of response to L-dopa drug therapy [38,100–102]. This also has the advantage of being the only approach which is somewhat regulated. While gene expression is unregulated in all three studies, here the gene product is only effective in the presence of L-dopa. Therefore, if there were an adverse event from overproduction of dopamine, theoretically this could be addressed by reducing L-dopa therapy. Pre-clinical studies have been promising, particularly in primates, which led to initiation of a human phase I clinical trial using AAV again as a gene transfer vehicle [38,100–102]. Although this too involved bilateral striatal infusion of AADC, only a single injection was used in each striatum as compared with the AAV-neurturin study. This study is ongoing but some impressive preliminary functional imaging data has been
185
presented at scientific meetings. Again F-dopa PET was used, but in this case it is a direct measure of gene transfer as F-dopa is retained only in cells expressing AADC, resulting in conversion to radioactive dopamine which does not freely diffuse out from cells. Therefore, unlike the other two approaches which use functional imaging as indirect measures of efficacy by analyzing changes in target cell function, this actually images the gene being expressed. This concept was validated in primates, where in vivo imaging of gene expression was clearly demonstrated [38,102]. So far several patients appear to have substantial increases in striatal F-dopa uptake, particularly in areas targeted for gene therapy, suggesting ongoing gene expression and increased striatal dopamine production.
Alzheimer’s Disease Alzheimer’s disease (AD) is a dementing neurodegenerative disorder usually associated with advanced age. Although the cause of AD remains unknown, pathologically this disorder is characterized by accumulation of extracellular plaques composed largely of fragments of the betaamyloid protein as well as dystrophic neurites with accumulation of hyperphosphorylated tau protein [103]. There are many drug and gene therapy strategies being pursued to potentially alter or prevent these abnormalities, but these have not yet reached the level of clinical trials [104]. Drug therapies currently marketed for AD focus upon increasing cholinergic transmission in the brain, since the loss of basal forebrain cholinergic neurons in AD reduces cholinergic transmission in key regions responsible for learning and memory [105,106]. These influence the entire brain, however, and therefore the therapeutic window is fairly low. There may also be additional benefit to retaining or promoting synaptic connections between cholinergic projection fibers from this region and target areas
3075
3076
185
Gene therapy for neurological disorders (except oncology)
responsible for learning and memory. Thus an alternative approach to gene therapy is to provide growth factors which will help maintain survival and possibly promote plasticity of these basal forebrain cholinergic neurons. Nerve growth factor (NGF) is primarily trophic for peripheral neurons, but its basal forebrain cholinergic neurons are among the population of central nervous system neurons which is also particularly responsive to this agent. Based upon extensive pre-clinical data supporting a role for NGF gene therapy in AD models, two NGF-based gene therapy approaches have reached human clinical trials. The first trial involved ex vivo gene therapy and was the first trial of any type of gene therapy for an adult neurodegenerative disorder [107,108]. In this phase I study, skin fibroblasts were obtained from eight patients were altered to express NGF. Patients were divided into three groups of two, with escalating doses of cells injected into each group. Patients initially underwent stereotactic injections awake, but two patients developed intracranial hemorrhages which were thought to be due to movement during surgery. One of these patients died several weeks later from a cardiopulmonary event, and subsequent patients were sedated or anesthetized during surgery with no further complications. There were no complications referable to the cell transplantation or the gene. In the one patient who died several weeks after surgery, gene expression was confirmed histologically with some suggestion of axonal growth towards the graft expressing NGF. There was also clinical data which suggested a slowing of the expected rate of decline over 2 years in standard cognitive assessments in these patients, and FDG-PET data supported an increase in cortical glucose metabolism. Ex vivo gene therapy is technically difficult and expensive, and it is difficult to create a uniform or ‘‘off the shelf ’’ therapy which cells must be obtained and altered following skin biopsies from each patient. Since inception of this trial, AAV gene
therapy technology advanced substantially over time, and the Canavan’s disease study in children (see below) and the AAV-GAD study in adult Parkinson’s patients had both already been initiated. A second phase I study was thus undertaken using AAV-mediated gene delivery of NGF to basal forebrain cholinergic neurons [109]. Results of this study have not yet been published but appear to be encouraging, and with later phase studies anticipated. Unlike PD, AD is not currently a neurosurgical disorder, so this is a clear example of the potential for biologicallybased therapies such as gene therapy to expand the role of the stereotaxic neurosurgeon in treatment of devastating neurological and psychiatric diseases.
Neurogenetic Disorders Among the earliest conceptions of gene therapy was the belief that this would primarily be an excellent method for correcting genetic defects. This seemed to be a fairly obvious application, since a normal copy of a defective gene could be inserted into a cell and thereby correct the problem. However, as evident from the above discussion, genetic diseases have not predominated among clinical trials of gene therapy in the central nervous system. This is in part due to unique difficulties with applying gene therapy to genetic diseases of the brain. In the periphery, diseases such as hemophilia or immunological disorders theoretically can be corrected at almost any age and result in a therapeutic benefit. Many central nervous system genetic defects cause profound developmental problems, however, and often the resulting damage is so severe that subsequently inserting a normal gene into the brain following birth may not provide adequate benefit. In addition, the diseases outlined above were all based upon surgical infusion into a focal brain region to influence a disease process. For neurogenetic diseases, the defect influences the biology of
Gene therapy for neurological disorders (except oncology)
every neuron throughout the brain, and even correction of a limited number of key structures still requires far more widespread gene delivery than is necessary for the adult diseases studied to date. This adds a significant technical problem, since at present there is no clinically acceptable method for widespread gene delivery throughout the brain from a peripheral injection. Nonetheless, these diseases are also usually lethal and have no reasonable therapeutic option, which substantially lowers the threshold for translation of a rationale gene therapy into the clinic. With these issues in mind, two neurogenetic disorders have been studied in phase I trials of AAVmediated gene therapy. The first trial of gene therapy in the human brain for a non-neoplastic disorder focused upon children with Canavan disease. This is caused by a defect in the oligodendrocyte-specific aspartoacylase gene, which results in a severe hypomyelination during development and spongiform degeneration of the brain. As a result, affected children are profoundly developmentally delayed, with severe psychomotor retardation, and most children die in the first decade of life. The first clinical trial of gene therapy to correct this defect delivered a normal copy of the aspartoacylase gene using a plasmid-based DNA vector in a liposomal, nonviral system [110]. This appeared to be safe but any potential therapeutic benefit was unclear. Subsequently, AAV was used as a more efficient method at gene delivery, and that trial is still ongoing [78]. Although outcome data have not been published, the immunological responses of these children have been reported [111]. There were no severe adverse reactions to the vector in any of the children, and no profound inflammatory responses. However, 30% of the children studied developed or increased neutralizing antibodies to the AAV vector following treatment compared with preoperative assessments. This is in contrast to the AAV-GAD study in adult PD patients, where no patient developed a change in neutralizing antibodies
185
following surgery. This could reflect a diseasespecific difference or a difference between adults and children. However, in this study, a much larger amount of virus was injected into children with severely atrophic brains, delivered at 12 sites from six catheters inserted at various places throughout the brain. Since the AAV-GAD trial involved a single injection deep in the brain, it is possible that the greater vector load delivered to a larger area of the brain (and possibly greater leakage into the CSF) could also explain this difference. Despite this mild immune response, there have been no reported clinical complications in any of the human patients tested and this was a pioneering trial which helped advance development of gene therapy in the human brain. Outcome data from this study have not yet been reported, but some information provided at national meetings has suggested some encouraging clinical and functional imaging findings. Batten disease is a second neurogenetic disorder which has used AAV-mediated gene therapy in a phase I study. This is also known as late infantile neuronal ceroid lipofuscinosis (LINCL), which is a neuronal lysosomal storage disease caused by a genetic defect in the CLN2 gene encoding the tripeptidyl peptidase I enzyme. Again children develop a progressively severe neurodengerative disorder, and can have varying levels of psychomotor retardation and often a seizure disorder, with death usually occurring between 8 and 12 years of age [112]. In the first few years of life, however, many patients have a more mild, early form of the disease which raises the possibility of intervening with gene therapy to at least attempt to arrest the progression of disease and possibly improve both quality of life and longevity. Studies in mouse models suggested that AAV2 gene therapy could cause both histological and functional improvements and primate safety testing supported further clinical development [113–115]. A similar protocol to the Canavan disease study was followed in a series of patients with
3077
3078
185
Gene therapy for neurological disorders (except oncology)
very severe, and subsequently more moderate forms of Batten disease [116]. Again 12 injections were performed in frontal, pre-motor and posterior parietal cortex through six burr holes. As with Canavan patients, Batten patients have severe brain atrophy with extremely large subarachnoid spaces. In order to maximize delivery to the brain parenchyma and minimize the risk of infusion into the CSF, this study added the use of frameless navigation. This is a method of identifying targets within the brain using volumetric MRI images which can be registered without a stereotactic frame to the patient’s head. Preoperative trajectory planning was performed using these images, to identify entry points which would facilitate entry of the catheters and subsequent infusion in brain parenchyma. Both clinical ratings and functional imaging (MR spectroscopy) were used as outcome measures in this study. The phase I study was completed and publication of the data is expected soon. A more recent study in the rodent model has demonstrated even greater efficacy with a primate strain of AAV which seems to be more efficient at widespread gene delivery [117]. Therefore, it is possible that such technological improvements may facilitate further development of gene therapy for such severe, global neurogenetic diseases.
Summary It is often surprising to most that there are so many clinical trials of gene therapy in the human brain which have been completed or remain ongoing. Following some difficulties in public perception of gene therapy at the end of the last decade, the resurgence in human gene therapy in recent years is clearly being led by investigations into neurological diseases. With the exception of the Newcastle disease virus treatment for brain tumors, every one of these trials required neurosurgical application. Most of the techniques used
in these studies have derived from more traditional stereotaxic neurosurgical methods. While new methods and techniques are actively being developed which are specific to the field of gene and cell-based therapies, these have benefited enormously from the long experience of stereotaxic and functional neurosurgeons with experimental therapies in challenging brain regions. The unusually strong safety record of gene therapy in the brain is in part due to the active involvement of neurosurgeons in the planning and execution of most of these trials. In addition, as with the expansion of indications for experimental deep brain stimulation, gene therapy now opens new opportunities for stereotaxic neurosurgeons to begin to intervene in devastating neurological and psychiatric disorders which have not traditional been viewed as neurosurgical diseases. The encouraging results from the human studies reviewed here along with the increasing number of studies moving into later phase trials raises the realistic hope that the promise of gene therapy may finally be realized to some degree through stereotaxic and functional neurosurgery.
References 1. Varmus H. Retroviruses. Science 1988;240(4858): 1427-35. 2. Barzon L, et al. Clinical trials of gene therapy, virotherapy, and immunotherapy for malignant gliomas. Cancer Gene Ther. 2006;13(6):539-54. 3. Aghi M, Rabkin S. Viral vectors as therapeutic agents for glioblastoma. Curr Opin Mol Ther. 2005;7(5):419-30. 4. Aronoff R, Petersen CC. Controlled and localized genetic manipulation in the brain. J Cell Mol Med. 2006; 10(2):333-52. 5. Yin LH, et al. Results of retroviral and adenoviral approaches to cancer gene therapy. Stem Cells 1998; 16(Suppl 1):247-50. 6. Behrstock S, Svendsen CN. Combining growth factors, stem cells, and gene therapy for the aging brain. Ann N Y Acad Sci. 2004;1019:5-14. 7. Blesch A, Tuszynski M. Ex vivo gene therapy for Alzheimer’s disease and spinal cord injury. Clin Neurosci. 1995;3(5):268-74.
Gene therapy for neurological disorders (except oncology)
8. Naldini L. Lentiviruses as gene transfer agents for delivery to non-dividing cells. Curr Opin Biotechnol. 1998;9(5):457-63. 9. Naldini L, et al. Efficient transfer, integration, and sustained long-term expression of the transgene in adult rat brains injected with a lentiviral vector. Proc Natl Acad Sci USA 1996;93(21):11382-8. 10. Naldini L, et al. In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector. Science 1996;272(5259):263-7. 11. Cockrell AS, Kafri T. Gene delivery by lentivirus vectors. Mol Biotechnol. 2007;36(3):184-204. 12. Jakobsson J, Lundberg C. Lentiviral vectors for use in the central nervous system. Mol Ther. 2006;13(3): 484-93. 13. Burton EA, Fink DJ, Glorioso JC. Replication-defective genomic HSV gene therapy vectors: design, production and CNS applications. Curr Opin Mol Ther. 2005;7(4): 326-36. 14. Glorioso JC, Fink DJ. Herpes vector-mediated gene transfer in treatment of diseases of the nervous system. Annu Rev Microbiol. 2004;58:253-71. 15. Chattopadhyay M, et al. HSV-mediated gene transfer of vascular endothelial growth factor to dorsal root ganglia prevents diabetic neuropathy. Gene Ther. 2005;12(18):1377-84. 16. Hao S, et al. Gene transfer to interfere with TNFalpha signaling in neuropathic pain. Gene Ther. 2007;14(13): 1010-6. 17. Martuza RL, et al. Experimental therapy of human glioma by means of a genetically engineered virus mutant. Science 1991;252(5007):854-6. 18. Mineta T, Rabkin SD, Martuza RL. Treatment of malignant gliomas using ganciclovir-hypersensitive, ribonucleotide reductase-deficient herpes simplex viral mutant. Cancer Res. 1994;54(15):3963-6. 19. Mineta T, et al. Attenuated multi-mutated herpes simplex virus-1 for the treatment of malignant gliomas. Nat Med. 1995;1(9):938-43. 20. Frenkel N. The history of the HSV amplicon: from naturally occurring defective genomes to engineered amplicon vectors. Curr Gene Ther. 2006;6(3):277-301. 21. Federoff HJ, et al. Expression of nerve growth factor in vivo from a defective herpes simplex virus 1 vector prevents effects of axotomy on sympathetic ganglia. Proc Natl Acad Sci USA. 1992;89(5):1636-40. 22. Geller AI, Breakefield XO. A defective HSV-1 vector expresses Escherichia coli beta-galactosidase in cultured peripheral neurons. Science 1988;241(4873):1667-9. 23. Geller AI, Freese A. Infection of cultured central nervous system neurons with a defective herpes simplex virus 1 vector results in stable expression of Escherichia coli betagalactosidase. Proc Natl Acad Sci USA. 1990;87(3): 1149-53. 24. Kaplitt MG, et al. Expression of a functional foreign gene in adult mammalian brain following in vivo gene transfer
25. 26. 27. 28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40. 41.
42.
185
via a herpes simplex virus type 1 defective viral vector. Mol Cell Neurosci. 1991;2:320-30. Neve RL, et al. Use of herpes virus amplicon vectors to study brain disorders. Biotechniques 2005;39(3):381-91. Douglas JT. Adenoviral vectors for gene therapy. Mol Biotechnol. 2007;36(1):71-80. Akli S, et al. Transfer of a foreign gene into the brain using adenovirus vectors. Nat Genet. 1993;3(3):224-8. Bajocchi G, et al. Direct in vivo gene transfer to ependymal cells in the central nervous system using recombinant adenovirus vectors. Nat Genet. 1993;3(3):229-34. Davidson BL, et al. A model system for in vivo gene transfer into the central nervous system using an adenoviral vector. Nat Genet. 1993;3(3):219-23. Le Gal La Salle G, et al. An adenovirus vector for gene transfer into neurons and glia in the brain. Science 1993;259(5097):988-90. Jager L, Ehrhardt A. Emerging adenoviral vectors for stable correction of genetic disorders. Curr Gene Ther. 2007;7(4):272-83. Hermonat PL, Muzyczka N. Use of adeno-associated virus as a mammalian DNA cloning vector: transduction of neomycin resistance into mammalian tissue culture cells. Proc Natl Acad Sci USA. 1984;81(20):6466-70. Tratschin JD, et al. A human parvovirus, adeno-associated virus, as a eucaryotic vector: transient expression and encapsidation of the procaryotic gene for chloramphenicol acetyltransferase. Mol Cell Biol. 1984;4(10): 2072-81. Flotte TR, et al. Stable in vivo expression of the cystic fibrosis transmembrane conductance regulator with an adeno-associated virus vector. Proc Natl Acad Sci USA. 1993;90(22):10613-7. Kaplitt MG, et al. Long-term gene expression and phenotypic correction using adeno-associated virus vectors in the mammalian brain. Nat Genet. 1994;8(2):148-54. Samulski RJ, Chang LS, Shenk T. A recombinant plasmid from which an infectious adeno-associated virus genome can be excised in vitro and its use to study viral replication. J Virol. 1987;61(10):3096-101. Wu Z, Asokan A, Samulski RJ. Adeno-associated virus serotypes: vector toolkit for human gene therapy. Mol Ther. 2006;14(3):316-27. Bankiewicz KS, et al. Long-term clinical improvement in MPTP-lesioned primates after gene therapy with AAVhAADC. Mol Ther. 2006;14(4):564-70. Sathornsumetee S, Rich JN, Reardon DA. Diagnosis and treatment of high-grade astrocytoma. Neurol Clin. 2007;25(4):1111-39, x. Louis DN. Molecular pathology of malignant gliomas. Annu Rev Pathol. 2006;1:97-117. Lassman AB, Holland EC. Incorporating molecular tools into clinical trials and treatment for gliomas? Curr Opin Neurol. 2007;20(6):708-11. Fulci G, Chiocca EA. The status of gene therapy for brain tumors. Expert Opin Biol Ther. 2007;7(2):197-208.
3079
3080
185
Gene therapy for neurological disorders (except oncology)
43. King GD, et al. Gene therapy and targeted toxins for glioma. Curr Gene Ther. 2005;5(6):535-57. 44. Pulkkanen KJ, Yla-Herttuala S. Gene therapy for malignant glioma: current clinical status. Mol Ther. 2005; 12(4):585-98. 45. Culver KW, et al. Gene therapy for the treatment of malignant brain tumors with in vivo tumor transduction with the herpes simplex thymidine kinase gene/ganciclovir system. Hum Gene Ther. 1994;5(3):343-79. 46. Oldfield EH, et al. Gene therapy for the treatment of brain tumors using intra-tumoral transduction with the thymidine kinase gene and intravenous ganciclovir. Hum Gene Ther. 1993;4(1):39-69. 47. van Dillen IJ, et al. Influence of the bystander effect on HSV-tk/GCV gene therapy. A review. Curr Gene Ther. 2002;2(3):307-22. 48. Pope IM, Poston GJ, Kinsella AR. The role of the bystander effect in suicide gene therapy. Eur J Cancer. 1997;33(7):1005-16. 49. Kun LE, et al. Stereotactic injection of herpes simplex thymidine kinase vector producer cells (PA317G1Tk1SvNa.7) and intravenous ganciclovir for the treatment of progressive or recurrent primary supratentorial pediatric malignant brain tumors. Hum Gene Ther. 1995;6(9):1231-55. 50. Packer RJ, et al. Treatment of progressive or recurrent pediatric malignant supratentorial brain tumors with herpes simplex virus thymidine kinase gene vector-producer cells followed by intravenous ganciclovir administration. J Neurosurg. 2000;92(2):249-54. 51. Prados MD, et al. Treatment of progressive or recurrent glioblastoma multiforme in adults with herpes simplex virus thymidine kinase gene vector-producer cells followed by intravenous ganciclovir administration: a phase I/II multiinstitutional trial. J Neurooncol. 2003;65(3): 269-78. 52. Ram Z, et al. Therapy of malignant brain tumors by intratumoral implantation of retroviral vector-producing cells. Nat Med. 1997;3(12):1354-61. 53. Shand N, et al. A phase 1-2 clinical trial of gene therapy for recurrent glioblastoma multiforme by tumor transduction with the herpes simplex thymidine kinase gene followed by ganciclovir. GLI328 European-Canadian Study Group. Hum Gene Ther. 1999;10(14):2325-35. 54. Klatzmann D, et al. A phase I/II study of herpes simplex virus type 1 thymidine kinase ‘‘suicide’’ gene therapy for recurrent glioblastoma. Study Group on Gene Therapy for Glioblastoma. Hum Gene Ther. 1998;9(17):2595-604. 55. Rainov NG. A phase III clinical evaluation of herpes simplex virus type 1 thymidine kinase and ganciclovir gene therapy as an adjuvant to surgical resection and radiation in adults with previously untreated glioblastoma multiforme. Hum Gene Ther. 2000;11(17):2389-401. 56. Eck SL, et al. Treatment of advanced CNS malignancies with the recombinant adenovirus H5.010RSVTK: a phase I trial. Hum Gene Ther. 1996;7(12):1465-82.
57. Germano IM, et al. Adenovirus/herpes simplex-thymidine kinase/ganciclovir complex: preliminary results of a phase I trial in patients with recurrent malignant gliomas. J Neurooncol. 2003;65(3):279-89. 58. Immonen A, et al. AdvHSV-tk gene therapy with intravenous ganciclovir improves survival in human malignant glioma: a randomised, controlled study. Mol Ther. 2004;10(5):967-72. 59. Trask TW, et al. Phase I study of adenoviral delivery of the HSV-tk gene and ganciclovir administration in patients with current malignant brain tumors. Mol Ther. 2000; 1(2):195-203. 60. Russell SJ, Peng KW. Viruses as anticancer drugs. Trends Pharmacol Sci. 2007;28(7):326-33. 61. Vaha-Koskela MJ, Heikkila JE, Hinkkanen AE. Oncolytic viruses in cancer therapy. Cancer Lett. 2007;254(2): 178-216. 62. Markert JM, et al. Oncolytic HSV-1 for the treatment of brain tumours. Herpes 2006;13(3):66-71. 63. Markert JM, et al. Conditionally replicating herpes simplex virus mutant, G207 for the treatment of malignant glioma: results of a phase I trial. Gene Ther. 2000;7(10): 867-74. 64. Harrow S, et al. HSV1716 injection into the brain adjacent to tumour following surgical resection of high-grade glioma: safety data and long-term survival. Gene Ther. 2004;11(22):1648-58. 65. Chiocca EA, et al. A phase I open-label, dose-escalation, multi-institutional trial of injection with an E1B-Attenuated adenovirus, ONYX-015, into the peritumoral region of recurrent malignant gliomas, in the adjuvant setting. Mol Ther. 2004;10(5):958-66. 66. Freeman AI, et al. Phase I/II trial of intravenous NDVHUJ oncolytic virus in recurrent glioblastoma multiforme. Mol Ther. 2006;13(1):221-8. 67. Colombo F, et al. Combined HSV-TK/IL-2 gene therapy in patients with recurrent glioblastoma multiforme: biological and clinical results. Cancer Gene Ther. 2005;12(10):835-48. 68. Eck SL, et al. Treatment of recurrent or progressive malignant glioma with a recombinant adenovirus expressing human interferon-beta (H5.010CMVhIFN-beta): a phase I trial. Hum Gene Ther. 2001;12(1):97-113. 69. Okada H, et al. Gene therapy of malignant gliomas: a phase I study of IL-4-HSV-TK gene-modified autologous tumor to elicit an immune response. Hum Gene Ther. 2000;11(4):637-53. 70. Ren H, et al. Immunogene therapy of recurrent glioblastoma multiforme with a liposomally encapsulated replication-incompetent Semliki forest virus vector carrying the human interleukin-12 gene–a phase I/II clinical protocol. J Neurooncol. 2003;64(1–2):147-54. 71. Lang FF, et al. Phase I trial of adenovirus-mediated p53 gene therapy for recurrent glioma: biological and clinical results. J Clin Oncol. 2003;21(13):2508-18.
Gene therapy for neurological disorders (except oncology)
72. Peng Z. Current status of gendicine in China: recombinant human Ad-p53 agent for treatment of cancers. Hum Gene Ther. 2005;16(9):1016-27. 73. Lang AE, Lozano AM. Parkinson’s disease. First of two parts. N Engl J Med. 1998;339(15):1044-53. 74. Samii A, Nutt JG, Ransom BR. Parkinson’s disease. Lancet 2004;363(9423):1783-93. 75. Freed CR, et al. Transplantation of embryonic dopamine neurons for severe Parkinson’s disease. N Engl J Med. 2001;344(10):710-9. 76. Olanow CW, et al. A double-blind controlled trial of bilateral fetal nigral transplantation in Parkinson’s disease. Ann Neurol. 2003;54(3):403-14. 77. Piccini P, et al. Factors affecting the clinical outcome after neural transplantation in Parkinson’s disease. Brain 2005;128(Pt 12):2977-86. 78. Janson C, et al. Clinical protocol. Gene therapy of Canavan disease: AAV-2 vector for neurosurgical delivery of aspartoacylase gene (ASPA) to the human brain. Hum Gene Ther. 2002;13(11):1391-412. 79. Hovland DN, Jr, et al. Six-month continuous intraputamenal infusion toxicity study of recombinant methionyl human glial cell line-derived neurotrophic factor (r-metHuGDNF) in rhesus monkeys. Toxicol Pathol. 2007;35(5):676-92. 80. Nutt JG, et al. Randomized, double-blind trial of glial cell line-derived neurotrophic factor (GDNF) in PD. Neurology 2003;60(1):69-73. 81. Gill SS, et al. Direct brain infusion of glial cell line-derived neurotrophic factor in Parkinson disease. Nat Med. 2003;9(5):589-95. 82. Luo J, et al. Subthalamic GAD gene therapy in a Parkinson’s disease rat model. Science 2002;298(5592): 425-9. 83. Erlander MG, et al. Two genes encode distinct glutamate decarboxylases. Neuron 1991;7(1):91-100. 84. Obeso JA, et al. Pathophysiologic basis of surgery for Parkinson’s disease. Neurology 2000;55(12 Suppl 6): S7-12. 85. Parent A, Cote PY, Lavoie B. Chemical anatomy of primate basal ganglia. Prog Neurobiol. 1995;46(2–3): 131-97. 86. Smith Y, et al. Ionotropic and metabotropic GABA and glutamate receptors in primate basal ganglia. J Chem Neuroanat. 2001;22(1–2):13-42. 87. Wichmann T, DeLong MR. Pathophysiology of Parkinson’s disease: the MPTP primate model of the human disorder. Ann N Y Acad Sci. 2003;991:199-213. 88. Carbon M, Eidelberg D. Modulation of regional brain function by deep brain stimulation: studies with positron emission tomography. Curr Opin Neurol. 2002;15 (4):451-5. 89. Fukuda M, Edwards C, Eidelberg D. Functional brain networks in Parkinson’s disease. Parkinsonism Relat Disord. 2001;8(2):91-4.
185
90. Kaplitt MG, et al. Safety and tolerability of gene therapy with an adeno-associated virus (AAV) borne GAD gene for Parkinson’s disease: an open label, phase I trial. Lancet 2007;369(9579):2097-105. 91. Feigin A, et al. Modulation of metabolic brain networks after subthalamic gene therapy for Parkinson’s disease. Proc Natl Acad Sci USA. 2007;104(49):19559-64. 92. Feigin A, et al. Metabolic correlates of levodopa response in Parkinson’s disease. Neurology 2001;57 (11): 2083-8. 93. Moeller JR, et al. Reproducibility of regional metabolic covariance patterns: comparison of four populations. J Nucl Med. 1999;40(8):1264-9. 94. Su PC, et al. Metabolic changes following subthalamotomy for advanced Parkinson’s disease. Ann Neurol. 2001;50(4):514-20. 95. Trost M, et al. Network modulation by the subthalamic nucleus in the treatment of Parkinson’s disease. Neuroimage 2006;129(10):2667-78. 96. Fjord-Larsen L, et al. Efficient in vivo protection of nigral dopaminergic neurons by lentiviral gene transfer of a modified Neurturin construct. Exp Neurol. 2005;195(1):49-60. 97. Gasmi M, et al. AAV2-mediated delivery of human neurturin to the rat nigrostriatal system: long-term efficacy and tolerability of CERE-120 for Parkinson’s disease. Neurobiol Dis. 2007;27(1):67-76. 98. Kordower JH, et al. Delivery of neurturin by AAV2 (CERE-120)-mediated gene transfer provides structural and functional neuroprotection and neurorestoration in MPTP-treated monkeys. Ann Neurol. 2006;60(6): 706-15. 99. Sanchez-Pernaute R, et al. Functional effect of adenoassociated virus mediated gene transfer of aromatic L-amino acid decarboxylase into the striatum of 6OHDA-lesioned rats. Mol Ther. 2001;4(4):324-30. 100. Forsayeth JR, et al. A dose-ranging study of AAVhAADC therapy in Parkinsonian monkeys. Mol Ther. 2006;14(4):571-7. 101. Sanftner LM, et al. AAV2-mediated gene delivery to monkey putamen: evaluation of an infusion device and delivery parameters. Exp Neurol. 2005;194(2): 476-83. 102. Bankiewicz KS, et al. Convection-enhanced delivery of AAV vector in parkinsonian monkeys; in vivo detection of gene expression and restoration of dopaminergic function using pro-drug approach. Exp Neurol. 2000;164(1): 2-14. 103. Swerdlow RH. Pathogenesis of Alzheimer’s disease. Clin Interv Aging. 2007;2(3):347-59. 104. Spencer B, et al. Novel strategies for Alzheimer’s disease treatment. Expert Opin Biol Ther. 2007;7(12):1853-67. 105. Shah S, Reichman WE. Treatment of Alzheimer’s disease across the spectrum of severity. Clin Interv Aging. 2006;1(2):131-42.
3081
3082
185
Gene therapy for neurological disorders (except oncology)
106. Herrmann N, Lanctot KL. Pharmacologic management of neuropsychiatric symptoms of Alzheimer disease. Can J Psychiatry. 2007;52(10):630-46. 107. Tuszynski MH, et al. Gene therapy in the adult primate brain: intraparenchymal grafts of cells genetically modified to produce nerve growth factor prevent cholinergic neuronal degeneration. Gene Ther. 1996;3(4):305-14. 108. Tuszynski MH, et al. A phase 1 clinical trial of nerve growth factor gene therapy for Alzheimer disease. Nat Med. 2005;11(5):551-5. 109. Tuszynski MH. Nerve growth factor gene delivery: animal models to clinical trials. Dev Neurobiol. 2007; 67(9):1204-15. 110. Leone P, et al. Aspartoacylase gene transfer to the mammalian central nervous system with therapeutic implications for Canavan disease. Ann Neurol. 2000;48(1): 27-38. 111. McPhee SW, et al. Immune responses to AAV in a phase I study for Canavan disease. J Gene Med. 2006;8(5): 577-88. 112. Worgall S, et al. Neurological deterioration in late infantile neuronal ceroid lipofuscinosis. Neurology 2007; 69(6):521-35.
113. Griffey MA, et al. CNS-directed AAV2-mediated gene therapy ameliorates functional deficits in a murine model of infantile neuronal ceroid lipofuscinosis. Mol Ther. 2006;13(3):538-47. 114. Hackett NR, et al. Safety of direct administration of AAV2(CU)hCLN2, a candidate treatment for the central nervous system manifestations of late infantile neuronal ceroid lipofuscinosis, to the brain of rats and nonhuman primates. Hum Gene Ther. 2005;16(12): 1484-503. 115. Sondhi D, et al. AAV2-mediated CLN2 gene transfer to rodent and non-human primate brain results in longterm TPP-I expression compatible with therapy for LINCL. Gene Ther. 2005;12(22):1618-32. 116. Crystal RG, et al. Clinical protocol. Administration of a replication-deficient adeno-associated virus gene transfer vector expressing the human CLN2 cDNA to the brain of children with late infantile neuronal ceroid lipofuscinosis. Hum Gene Ther. 2004;15(11):1131-54. 117. Sondhi D, et al. Enhanced Survival of the LINCL mouse following CLN2 gene transfer using the rh.10 rhesus macaque-derived adeno-associated virus vector. Mol Ther. 2007;15(3):481-91.
183 Impaired Hearing: Auditory Prosthesis J. K. Niparko . A. Marlowe . H. W. Francis
Introduction Medical approaches to deafness were motivated when a startling level of hearing was restored by a simple wire placed on the auditory nerve of a 50-year-old man during mastoid surgery for cholesteatoma [1]. Since then, clinicians have applied emerging understanding of auditory neurobiology and digital technologies of growing sophistication to enable the perception of sound in deafness. A cochlear implant is best characterized as a system that provides a patient with access to voiced, musical and environmental sounds. The device provides informational cues from the listener’s environment including speech that may escape visual detection. There are multiple effects of profound hearing loss in infants and toddlers and thus cochlear implants have been applied to ever younger children in an attempt to mitigate delays in developmental learning. For more than two decades, implantable devices have been applied to deaf children as an increasingly large proportion of all cochlear implants placed. Now, more than half of all newly implanted devices are placed in children, and most of these in children under 5 years of age. Auditory thresholds of cochlear-implanted children allow access to auditory information beyond that available to deaf children who routinely use conventional amplification (hearing aids), offering a critical substrate for developmental learning. In adults with acquired deafness, the reintroduction of sound enables impressive improvements in both sensitivity to, and recognition #
Springer-Verlag Berlin/Heidelberg 2009
of environmental, voiced and musical sounds. Through faster rates of stimulation, auditory stations within the brain appear capable of encoding information from the implant to enable speech comprehension and, increasingly, access to musical sounds. Candidates for a cochlear implant are those identified without material benefit associated with the use of powerful hearing aids. The benefit provided by implants may vary with a number of conditions including: hearing history, age of deafness onset, age at implantation, etiology of deafness, linguistic abilities, and the presence of a motivated system of support of oral language development. A patient’s health and environmental circumstances should be given individual consideration in determining candidacy for a cochlear implant. For children, planning rehabilitative and education services after surgery and activation of the device are critical considerations. The evolution of the cochlear implant is considered monumental for several reasons. The cochlear implant represents one of many innovative technologies that enable the rapid transfer of processed information. A unique feature of implant technology, however, is that it represents the effective merging of strategies that process information through both humanengineered and natural, neural circuits. To the extent that a cochlear implant can encode the sounds of speech with precision, the device can provide impressive opportunities to address a sensory abnormality with pervasive and longitudinal effects that has, in the past, exerted a marked impact on life choices.
3022
183
Impaired hearing: auditory prosthesis
Candidacy Cochlear implants convert acoustic signals into codes that preserve those features critical to speech understanding in normal listeners. In a seminal review of the neural substrates required to achieve effective, prosthetic stimulation of the auditory pathway, Parkins [2] described four principal considerations: 1.
2.
A substrate for peripheral auditory processing: There must first be sufficient initial processing of the acoustic signal. That is, appropriately designed patterns of an encoded stimulus that represents acoustic waveforms are needed to transfer information that is physiologically useful to the auditory system. Preserved auditory nerve fibers with retained responsivity: The majority of temporal bone studies of cases of profound sensorineural hearing loss (SNHL) reveal substantial histologic evidence of preserved spiral ganglion cells within the canal of Rosenthal. Although neuronal survival varies somewhat with etiology (> Figure 183‐1), histologic studies reveal surviving spiral ganglion counts that typically range from 1,000 to 30,000 cells of the normal complement of over 35,000 neurons. Retained neural
. Figure 183‐1 Neuronal populations in profound deafness
elements are typically disbursed throughout the cochlea outside of the proximal basal turn [3,4]. For many etiologies neuronal survival varies widely, but it is predictably high in cases of deafness induced by ototoxicity and low in deafness due to bacterial meningitis. In subjects without clinical SNHL, neuronal loss usually occurs at a rate of 2,000 per decade as a result of senescent changes alone [5]. In both older age groups and in longer duration of profound hearing loss the spiral ganglion cell populations are more attenuated [6]. The effect of the size of retained spiral ganglion populations on implant performance is unclear. For instance, clinical surveys have shown no clear relationship between the etiology of deafness and success in speech recognition with a cochlear implant with early designs of multichannel cochlear implants [7]. This suggests that although light microscopy may confirm spiral ganglion cell presence, it doesn’t confirm the functionality of auditory afferents. The requisite number of functional neurons required to effectively encode speech has been estimated in studies that correlate speech audiometry with neuronal reserves. In moderate-to-severe SNHL, histologic correlative studies indicate that approximately
Impaired hearing: auditory prosthesis
3.
20–35% of the normal neuronal complement is required to subserve socially useful speech recognition. Kerr and Schuknecht [8] suggested that neurons in the region of the upper basal and second turn of the cochlea are most critical in predicting preserved capabilities for speech recognition. Otte et al. [9] concluded that at least 10,000 spiral ganglion cells – with 3,000 or more in the apical 10 mm of the cochlea – are required for speech discrimination in cases of severe sensorineural hearing loss. The minimal number of auditory neurons needed to facilitate speech recognition with a cochlear implant is less certain. However, the number is likely quite small given observations of speech understanding in cases with only a modest number of residual neurons, with less than 10% of the normal complement of auditory neurons [10,11]. Established patterns of central neuronal pathways that enable processing of transmitted signals: Studies of employing morphometry of neurons in animal models [12,13] and in humans [14] reveal the altered appearance of neurons within the auditory brain stem as a consequence of auditory deafferentation (> Figure 183‐1). The greatest degree of degenerative changes is associated with deafness that is profound, of early onset, and of long duration wherein changes extend across synapses to affect neuron size. Nonetheless, there is a high degree of nerve survival within the auditory brain stem even under these conditions. Auditory plasticity may account for observed trends in favorable or unfavorable results with selected implant populations, respectively, when a cochlear implant is introduced early, or later on after a substantially prolonged period of deafness. Tyler and Summerfield [15] found evidence suggesting that central auditory factors may affect implant performance even
4.
183
in postlinguistic adults. Speech perception ability after implantation showed negative correlations with the duration of profound/ total deafness before implantation. Outcomes also showed acclimatization in the form of significant improvements in performance over time after implantation. For adult patients, the level of performance measured shortly after implantation was, on average, about half the level measured eventually. On average, performance reached asymptote after 30–40 months of implant use, although individual differences in the rate and amount of improvement were large. The accuracy of speech perception with implants by adults related to preoperative measures in three principal domains: (1) the number and physiological responsiveness of auditory ganglion cells and nerve fibers, indexed by measures of hearing sensitivity, duration of deafness, and age; (2) the responsiveness of the central nervous system, indexed by measures of cognitive and linguistic ability, and possibly also by age and duration of deafness; and (3) psychological motivation in learning to use the implant. These observations, paired with clinical results with cochlear implantation, indicate that an established, durable pathway for auditory processing is often present even in profound SNHL. Though there are important negative correlations between duration of deafness and performance [16–18], such correlations do not appear to be determinative. Even a prolonged period of deafness does not rule out the prospect for open-set speech understanding with a cochlear implant. Controllable interactions between an applied electrical field and responses of target neurons: Parkins [2] postulated that the relationship between the pattern of electrical stimulation and auditory neuronal response dictates the effectiveness of information
3023
3024
183
Impaired hearing: auditory prosthesis
transfer from an implant system. Ultimately, response patterns within the auditory pathway depend on individual auditory nerve fiber excitability. Neuronal shrinkage and demyelinization produce elevated thresholds and limit the maximal rate at which auditory nerve fibers can be driven. This change can affect not only the threshold for activation, but also a fiber’s adaptation (fatigability) and dynamic range of responsivity across, for example, the level of the stimulus as correlated with sound intensity. Spiral ganglion cell loss and shrinkage of remaining neurons likely account for reduced dynamic ranges [19] of both individual neuronal and psychophysical (behavioral) responses. Whereas acoustic stimulation is associated with behavioral and individual fiber dynamic ranges of 120 and 20–40 dB, respectively, electrical stimulation is associated with dynamic ranges of behavioral responses of 10–40 dB [20] and individual fiber responses of 7–10 dB, respectively [21,22]. These relatively constrained dynamic ranges require that compression be introduced into processing schemes. This design feature simulates normal cochlear transduction in order to accommodate the wide variance in sound intensities that are normally rendered both detectable and discriminable by the ear (> Figure 183‐2). The need for controllable interactions between the implant and target neurons underscores the importance of assessing surgical suitability. Clinical trials bear out the requirement for a stable interaction between the applied stimulus and target neurons. Substantially reduced benefit has been reported in association with compromised cochlear anatomy in clinical trials of adults [23] and children [24]. With normal cochlear anatomy the scala tympani offers a robust, shielded site to ensure electrode stability.
. Figure 183‐2 Artist’s rendition of a cochlear implant system demonstrating ear level processor and internal device. The processor receives sound and encodes acoustic waveforms into pulsatile stimuli which are transmitted from the externally worn headpiece to the internal receiver/ stimulator. The implantable stimulator delivers current to the electrode array implanted within the cochlea. (used with permission from Cochlear Corporation)
The Assessment of Implant Candidacy To ensure complete assessment of candidacy, clinicians consider myriad factors likely to affect daily use and performance of the device. The importance of comprehensive assessment is underscored by several factors.
The cochlear implant is a communication tool and does not cure disease of the cochlear transducers responsible for profound SNHL. Expectations, often defined by a patient’s [25] or family’s [26] psychological set, provide a critical framework for postoperative satisfaction with auditory rehabilitation [27]. The multifaceted nature of communication disorders requires parallel rehabilitative strategies, particularly in children whose
Impaired hearing: auditory prosthesis
prior communicative competence varies and deficits in auditory processing, speech production, cognitive ability, and sustainable attention often need to be addressed. Modifiers that affect cochlear implant performance prompt the need for candidacy assessment by a multidisciplinary team – one that can address the possible challenges to a successful implant experience. Candidates for cochlear implantation present with a unique set of baseline capabilities, needs, and objectives. Although advanced levels of sensorineural hearing loss are common in this group, implant candidates differ in virtually every other descriptor: Candidate age, age of hearing loss onset, etiology, and pattern of progression of deafness, cognitive and educational level, communication mode and competence within that mode, family and environment and personal motivation. These factors influence both candidacy decisions and the outcome of implantation. A team approach to candidacy assessment optimizes information on which to render broadly based, conclusive recommendations. A team approach also offers the means to guide intervention or counseling, if needed, and an informational base on which to formulate a plan of (re)habilitation after device activation.
183
High-resolution computed tomography (CT) of the temporal bone depicts the auditory and vestibular labyrinthine anatomy and provides information about cochlear abnormalities that guide counseling and surgical planning. Temporal bone CT scans survey temporal bone anatomy with attention to the caliber of the internal auditory canal, and labyrinthine anatomy [28] (> Figure 183‐3). Magnetic resonance imaging (MRI) offers a useful adjunct to CT for assessment of implant candidacy [29–31]. Whereas CT is the procedure of choice for detailing bony anatomy, MR imaging of soft tissues provides detailed assessment of nerves within the internal auditory canal, and soft tissue within heralding cochlear ossification after meningitis. A patient’s need for future assessment with MRI should be addressed preoperatively. The magnet in the internal device, used to retain the headset, may be contraindicated in these patients. A nonmagnetic modification of one commercially available device is available for patients whose medical or neurological condition mandates future MR studies [32]. Baumgartner [33] however, found that MRI applied to cochlear implant patients using different devices, imaged at 1 Tesla, did not cause implant malfunction or patient injury (> Figure 183‐4).
. Figure 183‐3 Coronal CT images of the right temporal bone demonstrating the patency of the cochlear apex (single arrow) and base (double arrow)
3025
3026
183
Impaired hearing: auditory prosthesis
. Figure 183‐4 Coronal MRI of left internal auditory canal and cochlea demonstrating innervation of the cochlear apex in a Mondini defect
Surgery Electrode carriers placed within the scala tympani lie proximate to target peripheral dendrites and cell bodies of auditory nerve afferent fibers (> Figure 183‐5). The design of the electrode array must be biocompatible and mechanically stable. It must allow for practical fabrication and facilitate atraumatic insertion. From a surgical perspective, the trauma of inserting the device is minimized through both effective design of the array and surgical technique. Multichannel arrays are typically placed within scala tympani along the first turn and a varying portion of the second turn of the cochlea (> Figure 183‐6). The depth of insertion is determined by several variables: surgical technique, obstructing tissue within the cochlea, and electrode design (e.g., surface features and longitudinal stiffness imparted by the carrier and connectors within the core of the implanted array). Electrode design, particularly the longitudinal (curved vs. straight) and cross-sectional
. Figure 183‐5 Electrode array placed in the scala tympani of the right cochlea
shape, and the stiffness of the carrier influence bending and the trajectory of the array tip during insertion. These properties influence the depth of insertion and the position of electrode contacts relative to neuronal fibers housed within the modiolar core of the cochlea. Current electrode carriers are typically inserted over a distance of
Impaired hearing: auditory prosthesis
183
. Figure 183‐6 Internal device showing magnified views of the electrode array to be inserted into scala tympani
20–30 mm. Although the potential advantages of insertion beyond 20 mm into the cochlea are as yet uncertain, at a minimum, deeper scala tympani insertion provides the opportunity to differentially stimulate frequency sites, particularly the low-frequency sites, and, depending on electrode design, may enable wider electrode spacing with the goal of providing distinct pitch perception via individual electrodes. Cochlear implant surgery is performed in the conventional otologic position using routine aseptic techniques. Implant systems currently in use are placed via the transmastoid, facial recess approach to the round window and scala tympani. The mastoid is exposed using a pedicled flap. Intraoperative flap design and plans for device positioning are aided by use of a mock implant and mock behind-the-ear processor. The flap is elevated to expose landmarks of the mastoid cortex – the spine of Henle, linea temporalis, and the mastoid tip – and at least 3 cm of bone above and beyond the mastoid. A simple mastoidectomy is performed. Bone at the margins of the cavity can provide
protection for connecting leads, and a platform for stabilizing the receiver-stimulator. The facial recess is approached using strategies that maximize visualization: adequate thinning of the posterior canal wall and full exposure of the horizontal semicircular canal, fossa incudis, and chorda tympani n.-facial n. angle. The facial recess is opened to visualize the incudostapedial joint and cochlear windows. If the facial recess is small or if the particular device implanted requires a generous facial recess exposure [34], the nerve may be sacrificed. In this case, care should be taken not to damage the tympanic membrane, as the chorda enters the middle ear at the level of the annulus. Bone on the anterior aspect of the vertical portion of the facial nerve should be removed in order to maximize visualization of the round window niche [35]. Misinterpretation of the anatomy may lead the surgeon to insert the electrode into a hypotympanic air cell or the petrous apex. Concern that the insertion may be suboptimal may prompt a skull radiograph obtained before leaving the operating room to confirm electrode position.
3027
3028
183
Impaired hearing: auditory prosthesis
The scala tympani is opened either directly, through the round window membrane, or, indirectly, through the promontory. The preferred approach is to enter the scala tympani through a cochleostomy that is formed at the anteroinferior margin of the round window (> Figure 183‐2). A small diamond burr is used to create a fenestra slightly larger than the electrode to be implanted. The electrode array is inserted into the scala under direct visualization, using methods designed to minimize trauma to the membranous components of the cochlea [36]. If resistance to insertion is encountered, the electrode can be withdrawn slightly, rotated medially (counterclockwise for the right cochlea and clockwise for the left) and carefully advanced [37]. Since buckling of the implant can produce spiral ligament, basilar membrane, and localized neural injury, aggressive insertion attempts should be avoided. Full insertion of the array within the basal turn of the cochlea may require extended drilling as occurs with labyrinthitis ossificans due to pressingitis or otosclerosis (> Figure 183‐7). Those electrodes placed deepest (most apical) in the cochlea approach spiral ganglion cells subserving the lower frequency regions, and those electrodes in the more proximal, or basal region, stimulate neurons subserving the higher frequency ranges of hearing. . Figure 183‐7 Coronal CT demonstrating areas of new bone growth partially obstructing the cochlea. An extended drilling procedure is generally required for a full insertion
After the array is inserted, the cochleostomy is sealed around the electrode with fibrous tissue. The array can be stabilized in a variety of ways. The electrode lead is positioned within the mastoid cavity such that the natural spring of the electrode lead does not advance the electrode out of the cochleostomy. The lead can be tucked medial to the short process of the incus after removal of the medial portion of the ‘‘incus bridge’’ at the superior aspect of the facial recess [38]. Prior to insertion of the electrode array, a depression is created in the bone behind the mastoid to accommodate the receiver-stimulator portion of the internal device. The receiver-stimulator should be placed to minimize protrusion, thereby reducing vulnerability to external trauma, and to restrict device movement which can shear connecting leads or displace the cochlear array. A growing percentage of all cochlear implants are placed in young children and implantation of the young child requires specific knowledge of the unique anatomy of the temporal bone in this age group and of the impact of skull growth on the implanted device. Although temporal bone growth has been shown to continue through adolescence, anatomy of the facial recess is fully developed at birth [39]. The most significant developmental changes are in the size and configuration of the mastoid cavity, which has been shown to expand in width, length, and depth from birth until at least the teenage years. Growth of the mastoid during this time parallels the growth patterns of the skull, with two periods of rapid development: one starts at birth and continues through early childhood, and the other occurs at puberty. From 1 year of age to adulthood, the average mastoid can be expected to grow 2.6 cm in length, 1.7 cm in width, and 0.9 cm in depth for males and 2.0 cm in length, 1.7 cm in width, and 0.8 cm in depth for females. Investigation in the young primate has demonstrated that cochlear implantation had no adverse effects on skull growth [40]. Moreover, the electrode appears to remain in a stable position
Impaired hearing: auditory prosthesis
with no migration over time. These observations strongly suggest that lateral skull base development occurs in a pattern that circumscribes the implanted device, with soft-tissue anchoring of connecting leads (> Figure 183‐2). In younger children, in whom the skull is much thinner, the bone is often drilled to the level of the dura, or a mobile island of thin bone can be created over the dura in the center of the well for protection. Alternatively, the surgeon can to thin the skull to approximate, but not reach, the level of the dura. Retention sutures are placed between the bone and dura. Electrode insertion and closure are similar to the procedures in the adult. Cochlear implantation entails risks inherent in extended mastoid surgery and those associated with the implanted device. Hoffman and Cohen [41] characterized implant-related complications as major if they required revision surgery, and minor if they resolved with minimal or no treatment. Hoffman and Cohen [41] noted that in later follow-up 220 (8%) major, and 119 (4.3%) minor complications among 2,751 implantations. Facer et al. [42] reported major and minor complication rates of 5.5 and 12.7%, respectively. Longitudinal tracking indicates a substantial reduction in the incidence of major complications in the past 10 years [41]. Major complications include facial nerve paralysis and implant exposure due to flap loss. Facial nerve injury is uncommon and, when recognized promptly, unlikely to produce permanent, complete paralysis. Loss of flap viability can lead to wound infection and device extrusion, necessitating scalp flap revision and, when intractable infection is present, device removal with or without replacement.
Implant Infection and Meningitis The risk of bacterial infection of an implanted device producing labyrinthitis or meningitis is
183
low, but represents significant morbidity and potential mortality. In 2002, the number of post-implantation meningitis cases was noted to have increased anecdotally. Initial reports suggested a higher risk of meningitis in patients implanted with a particular device design and particularly when placed with an electrode positioner; the manufacturer ultimately recalled unimplanted devices utilizing the positioner. Since then, continued studies have revealed a higher risk for the disease in patients with all cochlear implants compared to the general population. Children appear to be particularly affected. Of 52 cases originally reported by the FDA, 33 (63%) were under the age of 7. Reefhuis et al. [43] evaluated 4,264 children implanted between 1997 and 2002 and discovered 29 cases of bacterial meningitis in 26 children. Case-control analysis found identified variables that conferred heightened risk: a history of placement of a ventriculoperitoneal cerebrospinal fluid (CSF) shunt, prior otitis media, the presence of CSF leaks alone or inner-ear malformations with CSF leak, the use of a positioner-implant, incomplete insertion of the electrode, signs of middle-ear inflammation at the time of implantation, and cigarette smoking in the household. Effective vaccination against the strep pneumococcal and hemophilus influenza bacteria appear to confer substantial protection against postcochlear implantation messingitis with very few cases of bacterial meningitis known to have occurred in fully immunized implant users. Major complications that are strictly devicerelated involve partial or complete device failure. As the materials used to fabricate the internal device are expected to last more than 100 years, use-related failure per se is not expected; failure is attributed to either manufacturing flaws or trauma. Electrode array compression can occur with aggressive array insertion and may simply reduce the number of functional channels, produce noxious sensations with stimulation, and even complete device malfunction. The
3029
3030
183
Impaired hearing: auditory prosthesis
connecting lead between the receiver-stimulator and the electrode array is vulnerable to shearing if the device is not properly secured. Accordingly, embedding the device in bone and fixation with suture material is strongly advised. Device reliability is better than that of cardiac pacemakers, likely owing to design differences and the cochlear implant’s position in a relatively immobile site. Device failures requiring replacement with a new device do not appear to lead to compromised performance with the second implant [44].
. Figure 183‐8 External worn ear level process for speech encoding and cochlear implant activation
Results Factors Related to the Measurement of Auditory Performance Auditory Performance is measured at pre- and post- implant intervals. Pre-implant assessments are used to determine a patient’s candidacy. Postimplant measures serve to chart progress of the individual after activation of the device with an external processor (> Figure 183‐8), while allowing the clinician to monitor progress and device integrity. Measurement variables associated with auditory testing should be standardized as much as possible. Evaluation can be based on closed-set tests (e.g., forced choice of one answer from a list of four or six) and open-set tests of words and/or sentences in auditory- only condition. Closed-set tests and sentence tests (as scored by words correct) typically produce substantially higher percentages correct than do open-set tests. This difference reflects the contextual information available when closed-set and sentence materials are presented. The method of presentation can also affect speech perception scores [45]. Live presentation will typically produce higher rates of correct responses than will recorded presentations. Trends towards higher rates of open-set speech recognition with newer implant technology
mandate more stringent assessments of receptive capability. Increasing the difficulty of a speech perception test by adding background noise and more difficult words/sentences has the effect of limiting the ‘‘ceiling effects’’ that result from testing with simple, everyday phrases. For the purpose of generating more meaningful comparative data, increasing test difficulty normalizes distributions across populations, permitting the use of more powerful statistical designs in searching for differences between groups. Testing under conditions of auditory (implant)-only input reveals significant openset speech understanding capability (without visual cues) in more than 75% of patients after 3 years of device use. Normal listening is subserved by three parameters of sound. Cochlear implants are programmed to represent these characteristics of sound required for speech understanding. Cochlear implant programming:
Intensity (loudness) Intensity is encoded by current level
–
Impaired hearing: auditory prosthesis
Timing (temporal-onset & duration) Latency between stimuli is determined by rates & patterns Frequency (pitch) – Frequency is encoded by place: Site of stimulation (> Figure 183‐9)
–
Results in Adults Improved speech perception is the primary goal of cochlear implantation. Whereas initial clinical series judged implant efficacy mostly on environmental sound perception and performance on closed-set tests, greater emphasis is now placed on measures of open-set speech comprehension. Speech-perception results from early clinical trials have served to guide the evolution of cochlear implantation. Gantz et al. [7] provide early comparative data in their assessment of environmental sound and speech perception in a large cohort of nonrandomized single- and multiple-channel implant users. Multiple-channel implants provided significantly higher levels of performance on all measures. Cohen and colleagues [46] performed the first prospective, randomized trial of cochlear implants
183
through the U.S. Veterans’ Administration hospital system. The trial provided high-level, quality clinical data and convincing evidence of open-set speech recognition and the clear superiority of multiple- over single-channel designs. Moreover, the study was the first to report a low complication rate and high device reliability in postlingually deafened individuals, despite the fact that implantations were performed and monitored across multiple centers and with varying levels of prior implant experience. Cohen et al. [47], Wilson et al. [48], and Skinner et al. [49] compared the performance of subjects who were fitted with (upgraded) multiple-channel processors. The goal of these upgrades is to increase the rate of information transfer in association with higher rates of stimulation. Each study assessed a different processor; study design and the changes in processing strategy also differed between studies. The use of a speech processor of greater sophistication led to significantly better performance in open-set speech recognition in each study. Clinical observations in patients with current processors indicate that for patients with implant experience beyond 6 months, the mean score on open-set word testing approximates 25–40%, with a range of 0–100% [16–18,46]. Results achieved
. Figure 183‐9 Place coding enables pitch perception. Tonotopic layout of the cochlea is utilized to enable perception of lower pitches in more apical locations and higher pitch perceptions in more basal locations of the cochlea
3031
3032
183
Impaired hearing: auditory prosthesis
with the most recently developed speech processing strategies reveal mean scores above 75% correct on words-in-sentence testing, again with a wide range of 0–100%. Though subjects perform substantially less well on single-word testing, these mean scores continue to improve as the speechprocessing strategy evolves [50]. After implantation, speech recognition by telephone [51] and music appreciation are often observed. Again, this benefit seems to be achieved in larger percentages of patients with the use of more recently developed processing strategies.
Predictors of Benefit in Adults Evaluation of the benefit of cochlear implantation in adults has largely focused on measuring gains in speech perception. Assessments of speech recognition in implanted adults offer the opportunity to develop models of benefit prediction. As investigators identify the salient predictive factors, choices regarding candidacy, device and processing strategy, and possibly the need for postoperative auditory rehabilitation can be informed. Multivariate analysis, a statistical technique that determines the role of individual factors contributing to variation in performance, is the most commonly employed methodology [16–18,46,52,53]. The following are among the factors that have been evaluated:
Subject variables: age of onset, age of implantation, deafness duration, etiology, preoperative hearing, survival and location of spiral ganglion cells, patency of the scala tympani, cognitive skills, personality, visual attention, motivation, engagement, communication mode, and auditory memory. Device variables: processor, implant, electrode geometry, electrode number, duration and pattern of implant use, and the strategy employed by the speech processing unit.
Although the factors identified as most determinative have varied with different study populations, multiple regression analysis used in these studies has identified several variables with high predictive value for speech comprehension. Length of implant use accounts for a high degree of variance on speech perception measures. When length of use is controlled across implanted population, preoperative hearing, particularly with respect to speech recognition, also accounts for a high degree of variance. Variables of duration of deafness, age of implantation, cochlear patency, subject engagement with the therapeutic regimen, and processor type also carry relatively higher predictive values for speech understanding. Another variable that influences speech perception is the generation of technology incorporated in the cochlear implant system. Improvements in speech perception have been associated with improvements in signal processing strategies, speech processors, and electrode arrays [54] and current steering [53], but may reflect clinical trends as well as technological advances [18]. As a consequence, cochlear implant candidacy criterion are constantly expanding, and the trend towards the implantation of patients with more residual hearing has resulted in more promising performance levels with the cochlear implant (> Figure 183‐10). Identification of monosyllabic words for individuals using early generations of the Nucleus multichannel cochlear implant averaged 16% [7], whereas performance averages over 60% on similar measures with current technology [53]. Prior studies have contrasted performance across different manufacturers of devices. However, differences in performance appear more closely related to general demographic differences across patient populations, tests used, and length of experience with the devices than to differences in characteristics of like-generation devices per se [54,55].
Impaired hearing: auditory prosthesis
183
. Figure 183-10 Cochlear implant performance. With faster rates and relaxed criteria, mean performance levels have improved
Results in Children Auditory performance in children is assessed with a battery of audiological tests that can address the range of perceptual skills exhibited by children with severe-to-profound sensorineural hearing loss, before and after implantation. Although substantial auditory gains are apparent in implanted children, the range of quantifiable improvement varies widely between children and depends heavily upon the duration of use of the device, as well as preoperative variables. For this reason testing should survey a levels of speech recognition through a hierarchical approach – from simple awareness of sound, to pattern perception (discrimination of time and stress differences of utterance), to closed-set (multiple choice) and open-set (auditory only) speech recognition. Methodological challenges and developmental considerations inherent in pediatric implant assessment have been examined in detail by Kirk et al. [45] who categorized variables as relating to:
the child’s age and level of language and cognitive development (internal variables),
the child’s ability and willingness to respond as influenced by reinforcement and required memory task (external variables), and the procedure of voice presentation, the test administered, and the available options from which to choose a response (methodological variables).
The era of pediatric cochlear implantation began with House-3M, single-channel implants in 1980. Investigational trials with multiplechannel cochlear implants began with adolescents (aged 10 through 17 years) in 1985 and with children (aged 2 through 9 years) in November 1986. Implantation of infants and toddlers younger than 2 years of age began within 1995. Although clinical experience with cochlear implantation is considerably shorter in children than in adults, a large body of clinical reports is now available [56]. Investigators have evaluated the hearing and speech receptive benefit of cochlear implants in children for nearly 2 decades. Several observations have served as key findings in the field of pediatric cochlear implantation. A seminal report on pediatric cochlear implantation was provided by Osberger and
3033
3034
183
Impaired hearing: auditory prosthesis
colleagues [57] who described scale for stratifying children candidates using hearing-aids according to pure-tone detection thresholds. Gold-category, Silver-category, and Bronze-category hearing-aid users are classified according to unaided thresholds at.5, 1, and 2 kHz: Hearing-aid User Class Gold Silver Bronze
Unaided [email protected]/1kHz/ 2kHz (HL) 90–100 dB @ 2 of 3 frequencies (mean = 94 dB) 101–110 dB @ 2 of 3 frequencies (mean = 104 dB) >110 dB @ 2 of 3 frequencies (mean > 110 dB)
Hearing-aid users in the Gold category generally fall into Boothroyd’s [58] category of ‘‘considerable residual hearing.’’ In many such cases, hearing-aid users that function at this level demonstrate substantial capabilities for acquiring speech and oral language. This particular group of profoundly hearing-impaired children, therefore, sets a standard for speech perception performance and provides a useful cohort for comparison with children who have received cochlear implants. Silver-category hearing-aid users receive few spectral cues and rely heavily on timing aspects for speech perception. Bronzecategory users are considered ‘‘totally deaf ’’ and receive only low frequency percepts and restricted or no temporal cues to enable meaningful speech perception. In the 1990’s, most hearing-impaired children who have received cochlear implants were always in or eventually moved into the Bronze category. In comparison with their counterparts using amplification, cochlear-implanted children have demonstrated the following.
After 2 years of implant experience, mean speech intelligibility scores of implanted children surpass those of Silver-category hearing-aid users and approximate within
10% scores of Gold-category users, particularly on more stringent tests of speech comprehension [59,60]. Importantly, longitudinal tracking of implanted children shows no evidence of a plateau effect in mean scores of speech comprehension. This observation, as well as evidence based on multivariate analysis, suggests that length of implant use is the primary determinant of receptive benefit. Trends toward continued improvement in speech reception are also associated with improved intelligibility of speech [61]. Although the measured proficiency of speech produced remains low through the first 2 years after implantation, thereafter intelligibility scores show dramatic gains and surpass those of Silver-category hearing-aid users [57]. These observations are consistent with observations of gains in speech intelligibility in other childhood cohorts [62].
Over the past 15 years reports document further gains in speech recognition in young deaf children using multichannel cochlear implants [17,63–65]. Miyamoto et al. [66] noted that in 29 children with 1–4 years of experience with a cochlear implant, roughly half achieved open-set speech recognition. Waltzman et al. [65] found that children implanted before the age of 3 years attained high levels of speech perception performance with only 2 years of experience. Fryauf-Bertschy et al. [67] examined speech perception in cochlear-implanted children over 4 years and noted significant increases in pattern perception and results of closed-set speech perception tests. Variability in speech-perception performance across subjects is widely recognized. Factors implicated in speech recognition variability include the amount of residual hearing [68,69], age of implantation [70], mode of communication [24,65], family support [67], and length of
Impaired hearing: auditory prosthesis
deafness [69,71]. Miyamoto and colleagues [64] found that duration of deafness, communication mode, age at onset of deafness, and processor used accounted for roughly 35% of variance in closed-set testing. The length of implant use alone accounted for a larger percentage of variance on all speech perception measures. O’Donoghue and colleagues [70] found in a 5 year follow-up that age at implantation was the most significant covariate and that mode of communication was the most significant between-individuals factor. The latter study demonstrated that young age at implantation and oral communication mode are the most important known determinants of later speech perception in young children after cochlear implantation. Waltzman et al. [72] and Brackett et al. [73] reported improved speech reception in children implanted at age 2–3 years compared with children implanted at an older age. Data reported by Osberger [74], McConkey Robbins [75] and Manrique et al. [76] indicate that the performance of children implanted under the age of 2 years is significantly better than that of children implanted between age 2 and 3 years. More recent assessments of implantation in infants indicates that implantation at this stage is feasible and provides options for earlier access to speech stimuli, with potentially valuable effects stemming from language access early on in critical periods of development [77,78]. However, Osberger [74] also identified an important confound that exists in the children who receive a cochlear implant at a younger age: they are more likely to utilize an oral mode of communication. This, by itself, may be a predictor of higher implant performance – an observation borne out in early studies of a national childhood cohort assembled by Geers and colleagues [24]. Such observations underscore the importance of ongoing prospective studies. Osberger [74] also found that children with more residual hearing were undergoing
183
implantation relative to earlier cohorts. Gantz and colleagues [79] compiled data from across centers that indicate children with some degree of preoperative open-set speech recognition obtain substantially higher levels of speech comprehension. Taken together, these studies suggest the strongest potential for benefit exists with implantation at a young age, when intervention is provided early or, in the case of a progressive loss, before audition is lost completely. Cheng et al. [56] performed a meta-analysis that surveyed peer-reviewed published reports on pediatric cochlear implant results. Of 1,916 reports on cochlear implants published between 1966 and 1991, 44 provided sufficient patient data to compare speech recognition results between published (n = 1,904 children) and unpublished (n = 261) trials. Pooling results of these studies was hampered by the diversity of tests required to address the full spectrum of speech reception in implanted children. However, study results could be compared and the impact of selected variables (e.g., age at implantation, duration of use, etiology, and age at onset of deafness) could be determined in larger populations. The main conclusions of this meta-analysis was that earlier implantation is associated with a greater trajectory of gain in speech recognition, differences in performance diminish in time between congenital and acquired etiologies, and there is a distinct absence of a plateau of speech-recognition benefits over time. More than 75% of the children with cochlear implants reported in peer-reviewed publications have achieved substantial openset speech recognition after 3 years of implant use.
Bilateral Cochlear Implantation In an effort to expand the benefits obtained with unilateral cochlear implantation, bilateral
3035
3036
183
Impaired hearing: auditory prosthesis
implantation has been pursued to increasing numbers of patients, either during simultaneous or sequential surgeries. As of this writing, approximately 20% of all implant recipients utilize bilateral devices. The potential benefits of bilateral implantation relate to expansion of the soundfield. Ideally, the ability to integrate the signals from each ear would yield true, binaural benefit. Possible binaural advantages may include
Lowered tonal thresholds as a result of bilateral summation of inputs Improved localization and Improved speech recognition in noise
Tyler and colleagues [80] note that bilateral cochlear implantation generally enables head shadow effects in discerning speech in noise. However, a minority of patients exhibit true summation and squelch effects. Most bilateral recipients demonstrate improved horizontal plane localization and right v. left judgments are highly accurate. Litovsky et al. [81] and Schoen et al. [82] have observed general trends towards additional benefit with the use of a second implant. However, different binaural benefits were obtained in different subjects, and there are patients who demonstrate no additional benefit with bilateral implantation. Though the ‘‘era of bilateral implantation’’ has arrived, there are few controlled trials to date. There is evidence of an expanded sound field and loudness summation, and some level of sound localization ability to the majority of bilateral recipients [83]. In a minority of patients, benefits attributable to an effective squelch of background noise have been noted. Such findings suggest central integration of electrical stimulation from the two ears. However, systems that enable integrated speech processing of inputs from the bilateral sound fields have yet to be introduced clinically. It is uncertain whether
potential benefits of bilateral implantation could be expanded through such processing.
Language Acquisition in Children with Cochlear Implants The above-mentioned studies have helped to characterize gains in speech recognition. However, the primary goal of implantation in children is to facilitate comprehension and expression through the use of spoken language. Thus assessments of general communicative competence through the effective use of language have emerged as the measures of greatest importance in assessing early cochlear implantation. Language is defined as a vehicle for shaping and relating abstractions for communication [84]. The meaning of language-mediated abstraction is independent of the immediate situation. Practical use of language is based on the assignment of a single name to various appearances and situations under varying conditions. Information exchange via spoken language involves a conversion of thought into speech, a conversion that relies on mental representations of phonological (sound) structure and syntactic (phrase) structure. Preverbal communication behaviors underpin verbal language learning. Substantial gains in pre-linguistic behaviors, including eye contact and turn-taking [85], and in verbal spontaneity are observed as early correlates of benefit, developing within 6 months of implantation in young children. Tait et al. [86] found that preverbal measures obtained 12 months after implantation are predictive of late performance on speech perception tasks. They also observed a significant association between the preverbal measure of ‘‘autonomy’’ obtained before implantation and later speech perception performance. This latter finding carries theoretical implications for understanding of language development and
Impaired hearing: auditory prosthesis
suggests that intervention that promotes autonomy in adult-child interaction may lead to improved outcomes. Such intervention can be introduced as soon as deafness is discovered. Given improved auditory access, sound and phrase structure may be augmented by the cochlear implant. Although challenging to characterize, effects on receptive language skills and language production after implantation should be considered a crucial measure of the effectiveness of implants applied to young children. One approach is to assess language performance on standardized tests. The Reynell Developmental Language Scale evaluates both receptive and expressive skills independently [87]. These scales have been normalized on the basis of performance levels of hearing children over an age range of 1–8 years and have been used extensively in populations of deaf children. Whereas deaf children without cochlear implants achieved language competence at half the rate of normal-hearing peers, implanted subjects exhibited language-learning rates that matched, on average, those of their normal-hearing peers [87,88]. Though an improved rate of language learning was achieved with implantation, a gap in language level between cochlear-implanted children and their normal-hearing peers persisted due to delays in language acquisition noted at pre implantation baseline. The average age of implantation in this cohort was approximately 4 years. It remains to be seen whether earlier implantation might limit early language delays to enable more age-appropriate language acquisition. In evaluating the role of communication methodology, Robbins and colleagues [87] noted that implantation improved languagelearning rates for children in both oral- and total-communication settings based on Reynall Developmental Language Scale. Geers and colleagues [24], also assessing language skills in implanted children enrolled in oral and total communication (TC) settings, found that the groups did not differ in language level, though
183
the oral group demonstrated significantly better intelligibility in their speech production. Determinants of the level of verbal language attained after cochlear implantation likely relate to CNS processing. Pisoni et al. [89] assessed performance in two groups of pediatric cochlear implant users: ‘‘Stars’’ were children whose PB-K scores placed them in the top 20%, and children were placed in the control group if their scores placed them in the bottom 20%. Scores from behavioral tests of speech perception, spoken word recognition, speech intelligibility and language level were compared between the two groups at preimplantation assessments, and yearly postimplant assessments. ‘‘Stars’’ did not score consistently better than children in the control group on all behavioral measures. Differences between the groups were manifest only on specific tests and task demands. Children with superior implant performance were consistently better on measures of speech perception (i.e., vowel and consonant recognition), spoken word recognition, comprehension, language development and speech intelligibility than control children. However, the two groups did not differ on measures of vocabulary knowledge, non-verbal intelligence, visual-motor integration or visual attention. ‘‘Star’’ performance on measures of spoken language processing and speech intelligibility appears not due to global differences in overall performance, but to differences in the specific task of processing of auditory information provided by the cochlear implant. A strong relationship between spoken word recognition and speech intelligibility for the superior implant performers suggests that (1) knowledge between perception and production are transferred, and (2) perception and production share a common system of internal representation. Exceptional performance of the ‘‘Stars’’ appears to be due to their superior abilities to perceive, encode and retrieve information about spoken words from lexical memory. The concept of a ‘‘working memory’’ is important
3037
3038
183
Impaired hearing: auditory prosthesis
here, referring to the capacity to access information needed for processing tasks that propel the manipulation and transformation of the phonological representations of spoken words. Pisoni and Geers [90] investigated the impact of working memory on measures of speech perception, speech intelligibility, language processing and reading in implanted children with prelingual deafness. The assessed correlations between children’s ability to recall lists of digits presented in the auditory-only modality and their performance on these measures. Moderate to high correlations were found between auditory memory and performance in each outcome area, suggest that working memory plays an important role in mediating performance across these higher communication tasks. There appears to be commonality in the perceptual processes employed in these tasks. The ability to formally register speech sounds, coupled with rehearsal, can be used to encode and retrieve the representations of spoken words from lexical memory.
Educational Placement and Support of Implanted Children The hearing-impaired child is at substantial risk for educational underachievement [91,92]. Educational achievement by the hearing-impaired child can be enhanced by verbal communication, and traditionally teaching children to talk is more successful with children who have enough residual hearing to benefit from early devices of hearing rehabilitation [93]. Improved speech perception and production provided by cochlear implants offer the possibility of increased access to oral-based education and enhanced educational independence. Koch et al. [94] and Francis et al. [95] tracked the educational progress of implanted children by using an educational resource matrix to map educational and rehabilitative resource utilization. The matrix was developed on the
basis of observations that changes in classroom settings (e.g., into a mainstream classroom) are often compensated by an initial increase in interpreter and speech-language therapy. Follow-up of 35 school-aged children with implants indicated that relative to age-matched hearing-aid users with similar levels of baseline hearing, implanted students are mainstreamed at a substantially higher rate but this effect is not immediate and appears to require added rehabilitative support to be achieved. Within 5 years after implantation, the rate of full-time assignment to a mainstream classroom increases from 12 to 75%. The educational resource matrix also offers a basis for assessing overall cost-benefit ratios of the cochlear implant in children in the United States. Although educational costs for all implanted students remained static or actually increased initially, ultimate achievement of educational independence for the majority of implanted children produced net savings that ranged from $30,000–$100,000 per child, including the costs associated with initial cochlear implantation and postoperative rehabilitation. Language- and education-related outcomes in children with cochlear implants have been supplemented with parental perspectives of quality of life effects to yield cost-utility ratings [96,97] as described below. Despite cost- and utilitybases that made conservative assumptions, both studies supported the view that pediatric cochlear implantation is, relative to other medical and surgical interventions, highly costeffective in profoundly hearing-impaired young children.
Quality of Life and Cost-Effectiveness Assessment Studies of the cost-utility of cochlear implants in adults have assessed quality of life and health status to determine the utility gained from the multichannel cochlear implant [97–99]. Utility is
Impaired hearing: auditory prosthesis
a concept basic to commerce that reflects the true value of a good or service. Cost-utility methods determine the ratio of monetary expenditure to change in utility as defined by a change in quality of life over a given period. The assessment of cost-utility is based on the following: Cost Utility ¼ Costsðin $Þ=ðQuality Adjusted Life YearsÞ ¼ Costsðin $Þ=ðLife Years Health UtilityÞ
Life-Years is the mean anticipated number of years of implant experience based on a lifeexpectancy analysis of the participating cohort. The change in health utility reflects the preimplant and postimplant difference in scores on survey instruments designed and validated to accurately reflect quality of life. In the United States England, and Canada, health interventions with a cost-utility ratio under US$20,000are generally considered to represent acceptable value for money expended, i.e., they are ‘‘cost-effective’’ [100–102]. Costs per quality-adjusted life-year (QALY) for the cochlear implant in adult users were determined with use of cost data that account for the pre-, post-, and operative phases of cochlear implantation [99–101]. Benefits were determined on the basis of functional status and quality of life. The precise cost-utility results varied between studies, likely owing to differences in methods used to value benefit, the level of benefit obtained, and differences in costs associated with the intervention. Nonetheless, these appraisals consistently indicated that the multiple-channel cochlear implant in adult populations is associated with costutility ratios in the range of $14,000–$18,000/ QALY, indicating a highly favorable position in terms of its cost-effectiveness relative to other medical and surgical interventions employed within the United States. Hearing impairment is one of the most common clinical conditions affecting elderly people in the United States [103]. Hearing loss is so profound in 10% of the aged hearing-impaired
183
population that little or no benefit is gained with conventional amplification [104]. Assessing the effectiveness of cochlear implants in the elderly requires consideration of both audiological and psychosocial factors. The social isolation associated with acquired hearing loss in the elderly is accompanied by a significant decline in quality of life and an increase in emotional handicap [105]. The rehabilitation of hearing loss is therefore an important goal in this vulnerable population, providing both functional and psychological contributions to quality of life. But age-related degeneration of the spiral ganglion [106] and progressive central auditory dysfunction [107] raises concerns about the efficacy of cochlear prostheses in the elderly population. Comparable gains in speech understanding have been reported for elderly and younger groups of implant recipients [108,109], but the implications of these functional gains on the quality of life of older adults have not been well characterized. The determination of both auditory efficacy and quality of life is critical to any cost-benefit analysis in elderly hard of hearing and deaf adults, and it may be helpful in guiding clinical and resource decisions, particularly in light of the high costs associated with cochlear implantation as a non–life-saving intervention. Reports of quality of life gains in elderly patients with cochlear implants have been favorable [110], but are based on questionnaires that are difficult to correlate with function and costutility. Francis and colleagues [111] evaluated 47 patients, aged 50–80 years with multiple-channel cochlear implants, who completed the Ontario Health Utilities Index Mark 3 (HUI 3) survey and a quality of life survey. Responses to questions related to device use and quality of life changes were assessed and health utility scores before and after cochlear implantation were measured. There was a significant mean gain in health utility of 0.24 (s.d. 0.33) associated with cochlear implantation (p < 0.0001). Improvements in hearing and emotional health
3039
3040
183
Impaired hearing: auditory prosthesis
attributes were primarily responsible for this increase in the health-related quality of life measure. There was a significant increase in speech perception scores at 6 months after surgery (p < 0.0001 for both CID sentence and monosyllabic word tests) and a strong correlation between the magnitude of health utility gains and the postoperative enhancement of speech perception (r = 0.45, p < 0.05). Speech perception gain also correlated with improvements in emotional status and the number of hours that the implant is used daily. Thus cochlear implantation has a significant impact on the quality of life of older deaf patients and is a cost-effective intervention in this population. Improvements in speech perception after cochlear implantation are predictive of gains in health-related quality of life and have proven emotional benefits. Published cost-utility analyses of the cochlear implant in children have been limited by using either health utilities obtained from adult patients [96,98,101] or hypothetically estimated utilities of a deaf child [112–115]. These studies yielded costutility ratios that fell out over a wide range ($3,141–$25,450/QALY). Utilities derived from adult patient surveys may not capture the impact of issues unique to childhood deafness [116]. Cheng et al. [97] surveyed parents of a cohort of 78 children (average age 7.4 years, with 1.9 years of cochlear implant use) who received multichannel implants at the Johns Hopkins Hospital to determine direct and total cost to society per QALY. Parents of profoundly deaf candidate children (n = 48) awaiting cochlear implantation served as a comparison group to assess the validity of recall. Parents rated their child’s health state ‘‘now,’’ ‘‘immediately before,’’ and ‘‘one year before’’ the cochlear implant using the Time TradeOff (TTO), Visual Analog Scale (VAS) and Health Utilities Index – Mark III (HUI). Mean VAS scores increased 0.27 on a scale of 0–1 (from 0.59 to 0.86), TTO scores increased 0.22 (from 0.75 to 0.97), and HUI scores increased 0.39 (from 0.25 to 0.64). Discounted direct medical costs were
$60,228, yielding cost-utility ratios of $9,029/ QALY using the TTO, $7,500/QALY using the VAS, and $5,197/QALY using the HUI. Including indirect costs, such as reduced educational expenses, the cochlear implant yielded a net savings of $53,198 per child. Based on assessments of this cohort based in a single center, childhood cochlear implantation produces a positive impact on quality of life at reasonable direct costs and results in societal savings.
Auditory Rehabilitation After Cochlear Implantation The uniqueness of the listening experience enabled by a cochlear implant is underscored by qualitative differences in how sound perception is elicited as compared with other strategies of auditory rehabilitation. For example, a hearing aid filters, amplifies, and compresses the acoustic signal, thereby delivering a processed signal to the cochlea for transduction. In contrast, a cochlear implant receives, processes, and transmits acoustic information by generating electrical fields. Electrical injection bypasses nonfunctional cochlear transducers and directly depolarizes auditory nerve fibers. Implant systems convey an electrical code based in those selected features of speech that are critical to phoneme and word understanding in normal listeners, without the advantages of signal preparation provided by cochlear mechanisms of sound processing that render complex sounds listenable and discriminable. The cochlear implant listening experience also differs from normal audition in the timing of when that experience begins. It is either through a process of learning in the early, formative years in children or by virtue of ‘‘auditory memory’’ in adults that capabilities for meaningful listening develops for the majority of implant recipients. A central clinical question relates to the potential benefit to be derived from
Impaired hearing: auditory prosthesis
dedicated sensory training, commonly referred to as ‘‘auditory training.’’ Training appears to enable the cochlear implant user to apply both latent skills and adaptive strategies to listening tasks that offset at least partially the limitations of cochlear implant listening and prior deprivation and capitalize on the device capabilities. Much of the question of the need for and potential benefit of auditory training relates to the ability to promote the acquisition of perceptual and productive skills to yield effective spoken language. In the case of an adult who had little or no auditory experience in childhood, the ‘‘critical period’’ of auditory language learning is past. In such cases, adaptive strategies may be emphasized. In contrast, children who receive cochlear implants early in life can utilize adaptive capabilities to process sound information. In such cases, employing strategies that enable an implanted child to fully experience the stimuli that normally subserve development would seem on empiric grounds to offer benefit. Classic descriptions of auditory training strategy [117] stress the importance of training skills in auditory discrimination. Discrimination is based on acoustic characteristics that furnish the critical cues within an auditory experience – cues based in pitch, loudness, overtones, and the way in which these parameters change from one instant to another. Auditory discriminations required in everyday life vary in the ease with which distinctions can be made, ranging from gross discriminations (e.g., between environmental and voiced sounds), to difficult speech discriminations that demand that every phonetic element be heard with precision (e.g., speech discrimination in noise or discriminations of personal names or technical terms). To address the impairment of auditory discrimination as a consequence of hearing loss, Carhart [117] classified hearing impairments, with each having its own set of training requirements depending on the onset and severity of the loss. Those having early-onset and severe hearing
183
loss across the frequency spectrum require early training in sound awareness, beginning with crude discrimination tasks For late-onset hearing losses where discrimination ability may have been adequate previously, training in adaptive ‘‘habits of listening’’ are key in avoiding confusions in the use of spoken language. Classic approaches to auditory training [117] also stress the importance of training discrete auditory skills within a hierarchical construct (bottom-up approach). Auditory discrimination is based on acoustic characteristics that furnish the critical cues within an auditory experience – cues based in pitch, loudness, overtones, and the way in which such parameters change from one instant to another. In contrast, language-based approaches to auditory learning stress the importance of integrating auditory perception and language into the context of everyday routines. Language-based approaches reflect a continual interaction between top-down (integration of information for comprehension) and bottom-up (discrete) processing and are supported by the literature on child language development to more likely result in learning that is both generalized and durable.
Central Auditory Brain Stem Prostheses Syndromic bilateral neurofibromas of the VIIIth nerve (Neurofibromatosis type II (NF-2) induce profound hearing loss when function of the auditory division of the VIIIth is ablated by pathology or surgery. Affected patients are typically in the early adult stage of life. Cochlear implantation is an option in selected patients, but inadequate auditory nerve survival due to tumor progression or tumor removal can exclude this option for auditory rehabilitation. As an alternative, electrical hearing can be achieved with a multi-electrode array implanted on the surface of the anteroventral cochlear
3041
3042
183
Impaired hearing: auditory prosthesis
nucleus of the pontomedullary brain stem. Level of benefit so far has not reached open-set speech recognition without visual or contextual cues for the average user of brainstem implants [118,119]. However, newer designs employing multi-tined penetrating microelectrodes may permit access to afferents within the cochlear nucleus, offering an improved dynamic range of perceived inputs and better spectral representation through the stimulation of more discrete neuronal subpopulations [120]. Early studies of implantation of the auditory midbrain offer the prospect of stimulating the central auditory pathway at a site not vulnerable to tumor effacement [121] (> Figure 183‐11).
Summary As the pace of implanting circuits into the auditory pathway quickens, with more than 100,000 users worldwide, empirical evidence indicates
that the majority of deaf children and adults with implants gain meaningful benefit in important life experiences related to hearing, language, speech and environmental contact. Even when language- and hearing-specific gains are difficult to measure, cochlear implants provide greater access to environmental sound than is possible with hearing aids in well selected patients, although issues such as hearing in noise, transmitted frequencies and sound localization remain. Nonetheless, implant research addressing the impact of implants on literacy, academic achievement, and social and emotional growth continues to show positive benefits for the majority of recipients. Clinical evidence also reveals considerable variability in levels of benefit. Continued evaluation of the impact of cochlear implantation will require ongoing consideration of the effects of candidacy criteria, patient and anatomic characteristics, the impact of associated clinical interventions and the continued evolution of implant technology.
. Figure 183‐11 Design of implant arrays for direct stimulation of cochlear nucleus within the Brain stem
Impaired hearing: auditory prosthesis
References 1. Djourno A, Eyries C. Auditory Prosthesis by means of a distant electrical stimulation of the sensory nerve with the use of an indwelt coiling. Presse Med 1957;65 (63):1417. 2. Parkins C. The bionic ear: principles and current status of cochlear prostheses. Neurosurg 1985;16:853-65. 3. Hinojosa R, Marion M.R. Histopathology of profound sensorineural deafness. Ann. N.Y. Acad. Sci 1983;405:459-84. 4. Nadol JB. Incidence of reciprocal synapses on outer hair cells of the human organ of Corti. Ann Otol Rhinol Laryngol 1984;93247-50. 5. Schuknecht HF. Pathology of presbycusis. In: Goldstein J, Kashima C, Koopmann C, editors. Geriatric otorhinolaryngology 1989; p. 40–5. 6. Nadol JB, Young YS, Glynn RJ. Survival of spiral ganglion cells in profound sensorineural hearing loss: implications for cochlear implantation, Ann Otol Rhinol Laryngol 1989;98:411-16. 7. Gantz BJ, Tyler RR, Knutson JF, Woodworth G, et al. Evaluation of five different cochlear implant designs: audiologic assessment and predictors of performance. Laryngoscope 1988;98:1100-06. 8. Kerr A, Schuknecht HF. The spiral ganglion in profound deafness. Acta Otolaryngol 1968;65(6):586-98. 9. Otte J, Schuknecht HF, Kerr A. Ganglion cell populations in normal and pathological human cochleae: implications for cochlear implantation. Laryngoscope 1978;88:1231-46. 10. Linthicum F, Galey F. Histologic evaluation of temporal bones with cochlear implants. Ann Otol Rhinol Laryngol 1983;92:610-13. 11. Fayad J, Linthicum FH, Otto SR, et al. Cochlear implants: histopathologic findings related to performance in 16 human temporal bones. Ann Otol Rhinol Laryngol 1991;100:807-11. 12. Ryugo DK, Pongstapom T, Huchton DM, Niparko JK. Ultrastructural analysis of primary endings in deaf white cats: morphologic alterations in endbulbs of Held. J Comp Neurol 1997;385:230-44. 13. Niparko J. Activity influences on neuronal connectivity within the auditory pathway. Laryngoscope 1999;109:1721-30. 14. Moore J, Niparko J, Miller M, Linthicum F. Effect of profound hearing loss on a central auditory nucleus. Am J Otol 1997;15:588-95. 15. Tyler R, Summerfield Q. Cochlear implantation: relationships with research on auditory deprivation and acclimatization. Ear Hear 1996;17:38S-52S. 16. Gantz B, Woodworth G, Knutson J, Abbas P, Tyler R. Multivariate predictors of audiological success with multichannel cochlear implants. Ann Otol Rhinol Laryngol 1993;102:909-16.
183
17. Waltzman S, Fisher S, Niparko J, Cohen N. Predictors of postoperative performance with cochlear implants. Ann Otol Rhinol Laryngol 1995;104 supp 165:15-18. 18. Rubinstein J, Parkinson W, Tyler R, Gantz B. Residual speech recognition and cochlear implant performance: effects of implantation criteria. Am J Otol 1999;20:445-52. 19. Kiang N, Moxon E. Physiological considerations in artificial stimulation of the inner ear. Ann Otol Rhinol Laryngol 1972;81:714-30. 20. Zeng FG, Grant G, Niparko J, Galvin JJ, Shannon RV, Opie J, Segel P. Speech dynamic range and its effects on cochlear implant performance. J Acoust Soc Am 2002;111:377-86. 21. Parkins CW, Colombo J. Auditory-nerve single-neuron thresholds to electrical stimulation from scala tympani electrodes. Hear Res 1987;31:267-85. 22. Javel E, Tong Y, Shepherd R, Clark G. Responses of cat auditory nerve fibers to biphasic electrical current pulses. Ann Otol Rhinol Laryngol 1987;96 suppl 128: 26-30. 23. Cohen NL, Hoffman RA. Complications of cochlear implant surgery. In: Eisele DW, editors. Complications in Head and Neck Surgery. Baltimore: Mosby, 1993. 24. Geers A, Nicholas J, Tye-Murray N, Uchanski R, et al. Effects of communication mode on skills of long-term cochlear implant users. Ann Rhinol Otol Laryngol 2000;109:89-92. 25. Watson C. Auditory perceptual learning and the cochlear implant. Am J Otol 1991;12 suppl: 73-9. 26. Beadle E, Shores A, Wood E. Parental perceptions of the impact upon the family of cochlear implantation in children. Ann Otol Rhinol Laryngol 2000;Suppl 185:111–14. 27. Ross M, Levitt H. Consumer satisfaction is not enough: hearing aids are still about hearing. Sem Hear 1997;18:7-11. 28. Woolley AL, Oser AB, Lusk RP, et al. Pre-operative temporal bone computed tomography scan and its use in evaluating the pediatric cochlear implant candidate. Laryngoscope 1997;108:1100-6. 29. Harnsberger HR, Dart DJ, Parkin JL, Smoker WR, Osborn AG. Cochlear implant candidates: assessment with CT and MR imaging. Radiology 1987;164(1):53-7. 30. Casselman JW, Kuhweide R, Deimling M, Ampe W, Dehaene I, Meeus L. Constructive interference in steady state-3DFT MR imaging of the inner ear and cerebellopontine angle. Am J Neuroradiol 1993;14:47-7. 31. Bettman R, Beek E, Van Olphen A, Zonneveld F, Huizing E. MRI versus CT in assessment of cochlear patency in cochlear implant candidates. Acta Otolaryngol 2004;124 (5):577-81. 32. Heller JW, Brackmann DE, Tucci DL, Nyenhuis JA, Chou CK. Evaluation of MRI compatibility of the modified nucleus multichannel auditory brainstem and cochlear implants. Am J Otol 1996;17(5):724-9.
3043
3044
183
Impaired hearing: auditory prosthesis
33. Baumgartner WD, Youssefzadeh S, Hamzavi J, Czerny C, Gstoettner W. Clinical application of magnetic resonance imaging in 30 cochlear implant patients. Otol Neurotol 2001;22:818-22. 34. Lalwani AK, Larky JB, Wareing MJ, Kwast K, et al. The Clarion multi-strategy cochlear implant surgical techniques, complications, and results: a single institutional experience. Am J Otol 1998;19:66-70. 35. Proctor B, Bollobas B, Niparko J. Anatomy of the round window niche. Ann Otol Rhinol Laryngol 1986;95:444-6. 36. Cohen NL. Cochlear implant soft surgery: fact or fantasy? Otolaryngol Head Neck Surg 1997;117:214-6. 37. Kennedy D. Multichannel intracochlear electrodes: mechanism of insertion trauma. Laryngoscope 1987;97: 42-9. 38. Balkany T, Telischi FF. Fixation of the electrode cable during cochlear implantation: the split bridge technique. Laryngoscope 1995;105(2):217-18. 39. Bielamowicz SA, Coker NJ, Jenkins HA, Igarashi M. Surgical dimensions of the facial recess in adults and children. Arch Otolaryngol Head Neck Surg 1988;114 (5):534-7. 40. Burton MJ, Shepherd RK, Xu SA, Xu J, Franz BK, Clark GM. Cochlear implantation in young children: histological studies on head growth, leadwire design, and electrode fixation in the monkey model. Laryngoscope 1994;104 (2):167-75. 41. Hoffman RA, Cohen NL. Surgical pitfalls in cochlear implantation. Laryngoscope 1993;103:741. 42. Facer GW, Peterson A, Brey RH, Marion M, Cevette M, Balko K, Green JD, Rose D, Pool A. The mayo clinic experience with the cochlear implant. Ear Nose Throat J 1994;73(3):149-52, 43. Reefhuis J, Honein MA, Whitney CG, et al. Risk of bacterial meningitis in children with cochlear implants. N Engl J Med 2003;349:435-45. 44. Rubinstein JT, Parkinson WS, Lowder MW, Gantz BJ, et al. Single-channel to multichannel conversions in adult cochlear implant subjects. Am J Otolaryngol 1997;19:461-6. 45. Kirk KI, Diefendorf A, Pisoni D, Robbins A. Assessing speech perception in children. In: Mendel L, Danhauer J, editors. Audiologic evaluation and mangement and speech perception assessment. Singular; San Diego: p. 101-32. 46. Cohen NL, Waltzman S, Fisher S. The department of veterans affairs cochlear implant study group. A prospective, randomized study of cochlear Implants. N Engl J Med 1993;328:233-7. 47. Cohen NL, Waltzman SB, Fischer SG. Prospective randomized clinical trial of advanced cochlear implants: preliminary results of a department of veterans affairs cooperative study. Ann. otol. rhinol. Laryngol 1991;100 (10):823-9.
48. Wilson BS, Finley CC, Lawson DT, Wolford RD, Eddington DK, Rabinowitz WM. Better speech recognition with cochlear implants Nature 1991;352:236-8. 49. Skinner MW, Fourakis MS, Holden TA, Holden LK, Demorest ME. Identification of speech by cochlear implant recipients with the multipeak (MPEAK) and spectral peak (SPEAK) Speech Coding Strategies I: vowels. Ear Hear 1996;17(3):182-97. 50. Skinner MW, Fourakis MS, Holden TA, Holden LK, et al. Identification of speech by cochlear implant recipients with the multipeak (MPEAK) and spectral peak (SPEAK) speech coding strategies II: consonants. Ear Hear 1999;20:443-60. 51. Cohen NL, Waltzman SB, Roland JT Jr, Bromberg B, Cambron N, Gibbs L, Parkinson W, Snead C. Results of speech processor upgrade in a population of veterans affairs cochlear implant recipients. Am J Otol 1997;18 (4):462-5. 52. Kileny P, Zimmerman-Phillips S, Kemink J, Schmaltz S. Effects of preoperative electrical stimulability and historical factors on performance with multichannel cochlear implants. Ann Otol Rhinol Laryngol 1991;100: 563-8. 53. Buechner A, Brendel M, Kru¨eger B, Frohne-Bu¨chner C, Nogueira W, Edler B, Lenarz T. Current steering and results from novel speech coding strategies. Otol Neurotol 2008;29(2):203-7. 54. Wilson BS. Cochlear implant technology. Cochlear implants: principles & practices. Lippincott Williams & Wilkins; Philadelphia: p. 109-19. 55. Kiefer J, Gall V, Desloovere C, Knecht R, Mikowski A, von Ilberg C. A follow-up study of long-term results after cochlear implantation in children and adolescents. Eur Arch Otorhinolaryngol 1996;253(3):158-66. 56. Cheng A, Grant G, Niparko J. A meta-analysis of the pediatric cochlear implantation. Ann Otol Laryngol Rhinol 1999;177:124-8. 57. Osberger MJ, Maso M, Sam LK. Speech intelligibility of children with cochlear implants, tactile aids, or hearing aids. J Speech Hear Res 1993;36:186-203. 58. Boothroyd A. Hearing aids, cochlear implants, and profoundly deaf children. In: Owens E, Kessler D, editors. Cochlear implants in young deaf children. Boston: College Hill Press; 1989. p. 81-99. 59. Miyamoto R, Osberger M, Robbins A, Myres WA, Kessler K, Pope ML. Longitudinal evaluation of communication skills of children with single- or multichannel cochlear implants. Am J Otol 1992;13:215-22. 60. Miyamoto R, Osberger M, Todd S, Robbins A, et al. Variables affecting implant performance in children. Laryngoscope 1994;104:1120-24. 61. Osberger MJ, Todd SL, Berry SW, Robbins AM, Miyamoto RT. Effect of age at onset of deafness on children’s speech perception abilities with a cochlear implant. Ann Otol Rhinol Laryngol 1991;100:883-8.
Impaired hearing: auditory prosthesis
62. Tye-Murray N. Cochlear implants and children: a handbook for parents, teachers and speech & hearing professionals. Washington, DC: Alexander Graham Bell Publishing; 1994. 63. Miyamoto RT, Kirk KI, Robbins AM, Todd S, Riley A. Speech perception and speech production skills of children with multichannel cochlear implants. Acta Otolaryngol 1996;116(2):240-3. 64. Staller SJ, Beiter AL, Brimacombe JA, et al. Pediatric performance with the Nucleus 22-channel cochlear implant system. Am J Otol 1991;12 Suppl:126-36. 65. Waltzman SB, Cohen NL, Gomolin RH, Shapiro WH, Ozdamar SR, Hoffman RA. Long-term results of early cochlear implantation in congenitally and prelingually deafened children. Am J Otol 1994;15 Suppl 2:9-13. 66. Miyamoto RT, Osberger MJ, Robbins AM, Myres WA, Kessler K. Prelingually deafened children’s performance with the nucleus multichannel cochlear implant. Am J Otol 1993;14(5):437-45. 67. Fryauf-Berschy H, Tyler R, Kelsay D, Gantz BJ, Woodworth GG. Cochlear implant use by prelingually deafened children: the influences of age at implant and length of device use. J Speech Hearing Res 1997;40:183-99. 68. Zwolan TA, Zimmerman-Phillips S, Ashbaugh CJ, Hieber SJ, Kileny PR, Telian SA. Cochlear implantation of children with minimal open-set speech recognition skills. Ear Hear 1997;18(3):240-51. 69. Meyer TA, Svirsky MA, Kirk KI. Improvements in speech perception by children with profound prelingual hearing loss effects of device, communication mode, and chronological age. J Speech Lang Hearing Res 1998;41:846-58. 70. O’Donoghue G, Nikolopoulos T, Archbold S. Determinants of speech perception in children after cochlear implantation. Lancet 2000;5(356):466-8. 71. Vieu A, Mondain M, Blanchard K, Sillon M, ReuillardArtieres F, Tobey E, Uziel A, Piron JP. Influence of communication mode on speech intelligibility and syntactic structure of sentences in profoundly hearing impaired French children implanted between 5 and 9 years of age. Int J Pediatr Otorhinolaryngol 1998;44 (1):15-22. 72. Waltzman S, Cohen N, Shapiro W. Use of a multichannel cochlear implant in the congenitally and prelingually deaf population. Laryngoscope 1992;102:395-9. 73. Brackett D, Zara CV. Communication outcomes related to early implantation. Am J Otol 1998;19:453-60. 74. Osberger M. Candidacy and performance trends in children. Presented at eighth symposium on cochlear implants in children, Los Angeles, CA, 2001. 75. McConkey Robbins A, Koch DB, Osberger MJ, Zimmerman-Phillips S, Kishon-Rabin L. Effect of age at cochlear implantation on auditory skill development in infants and toddlers. Arch Otolaryngol Head Neck Surg 2004;130(5):570-4.
183
76. Manrique MJ, Huarte A, Amor JC, Baptista P, GarciaTapia R. Results in patients with congenital profound hearing loss with intracochlear multichannel implants. Adv Otorhinolaryngol 1993;48:222-30. 77. Waltzman SB, Roland JT, Jr. Cochlear implantation in children younger than 12 months. Pediatrics 2005;116(4): e487-93. 78. Dettman SJ, Pinder D, Briggs RJ, Dowell RC, Leigh JR. Communication development in children who receive the cochlear implant younger than 12 months: risks versus benefits. Ear Hear 2007;28 (2 Suppl):11S-18S. 79. Gantz B, Rubinstein J, Tyler R, Teagle H, Cohen N, Waltzman SB, Miyamoto RT, Kirk K. Long-term results of cochlear implants in children with residual hearing. Ann Otol Rhinol Laryngol Suppl 2000;185:33-6. 80. Tyler RS, Dunn CC, Witt SA, Preece JP. Residual speech perception and cochlear implant performance in postlingually deafened adults. Ear Hear 2003;24(6):539-44. 81. Litovsky RY, Parkinson A, Arcaroli J, et al. Bilateral cochlear implants in adults and children. Arch Otolaryngol Head Neck Surg 2004;130(5):648-55. 82. Schoen F, Mueller J, Helms J, Nopp P. Sound localization and sensitivity to interaural cues in bilateral users of the Med-El Combi 40/40 + cochlear implant system. Otol Neurotol 2005;26(3):429-37. 83. Murphy J, O’Donoghue G. Bilateral cochlear implantation: an evidence-based medicine evaluation. Laryngoscope 2007;117(8):1412-8. 84. Jackendoff R. Phonological structure. In: Patterns in the mind: language and Human nature. New York: Basic Books; 1994. p. 53-65. 85. Tait M, Lutman M. Comparison of early communicative behavior in young children with cochlear implants and with hearing aids. Ear Hear 1994;15:352-62. 86. Tait M, Lutman ME, Nikolopoulos TP. Communication development in young deaf children: review of the video analysis method. Int J Pediatr Otorhinolaryngol 2001;61:105-12. 87. Robbins AM, Svirsky M, Kirk KI. Children with implants can speak, but can they communicate? Otolaryngol Head Neck Surg 1997;117155-60. 88. Svirsky M, Robbins A, Kirk K, Pisoni D, et al. Language development in profoundly deaf children with cochlear implants. Psychol Sci 2000;11:153-8. 89. Pisoni DB, Svirsky MA, Kirk KI, Miyamoto RT. Looking at the ‘‘Stars’’: a first report on the intercorrelations among measures of speech perception, intelligibility and language development in pediatric cochlear implant users. Bloomington, IN: Indiana University; 1997. p. 51-91. 90. Pisoni DB, Geers AE. Working memory in deaf children with cochlear implants: correlations between digit span and measures of spoken language processing. Ann Otol Rhinol Laryngol 2000;Suppl 185:92–3.
3045
3046
183
Impaired hearing: auditory prosthesis
91. Trybus R, Karchmer M. School achievement scores of hearing impaired children: national data on achievement status and growth patterns. Am Ann Deaf 1977;122:62-9. 92. Holt J. Stanford achievement test–8th ed. Am Ann Deaf 1993;138:172-5. 93. Geers A, Moog J. Evaluating the benefits of cochlear implants in an education setting. Am J Otol 1991;12 suppl:116-25. 94. Koch M, Wyatt JR, Francis H, Niparko J. A model of educational resource use by children with cochlear implants. Otolaryngol Head Neck Surg 1997;117:174-9. 95. Francis H, Koch M, Wyatt J, Niparko J. Trends in educational placement and cost-benefit considerations in children with cochlear implants. Arch Otolaryngol Head Neck Surg 1999;125:499-505. 96. O’Neill C, O’Donoghue GM, Archbold SM, Normand C. A cost-utility analysis of pediatric cochlear implantation. Laryngoscope 2000;110:156-60. 97. Cheng A, Rubin H, Powe N, Mellon N, Francis H, Niparko J. A cost-utility analysis of the cochlear implant in children. JAMA 2000;284:850-6. 98. Summerfield A, Marshall D. Cochlear implantation in the U.K. 1990–1994. Nottingham: Medical Research Council Institute of Hearing Research; 1995 p. 199-236. 99. Wyatt JR, Niparko JK, Rothman ML, de Lissovoy GV. Cost utility of the multichannel cochlear implants in 258 profoundly deaf individuals. Laryngoscope 1996;106:816-21. 100. Azimi N, Welch H. The effectiveness of cost-effectiveness analysis in containing costs. J Gen Intern Med 1998;13:664-9. 101. Summerfield A, Marshall D, Archbold S. Cost-effectiveness considerations in pediatric cochlear implantation. Am J Otol 1997;18 suppl 6:S166-8. 102. Kind P, Gudex C. The role of QALYs in assessing priorities between health-care interventions. In: Drummond MF, Maynard A, editors. Purchasing and providing costeffective health care. Edinburgh, Scotland: ChurchillLivingstone; 1993. p. 94-108. 103. Campbell VA, Crews JE, Moriarty DG, Zack MM, et al. Surveillance for sensory impairment, activity limitation, and health-related quality of life among older adults – United States, 1993–1997. MMWR CDC Surveill Summ 1999;48:131-56. 104. Havlik R. Aging in the eighties: impaired senses for sound and light in persons age 65 years and over. Preliminary data from the supplement on aging to the National Health Interview Survey: United States, January-June 1984. National Center for Health Statistics, DHHS; 1986.
105. Mulrow CD, Aguilar C, Endicott JE, Velez R, et al. Association between hearing impairment and the quality of life of elderly individuals. J Am Geriatr Soc 1990;38:45-50. 106. Ng M, Niparko JK, Nager GT. Inner ear pathology in severe to profound sensorineural hearing loss. In: Niparko JK, Kirk KI, Mellon NK, Robbins AM, et al. editors. Cochlear implants: principles and practices. Philadelphia: Lippincott Williams & Wilkins; 2000. p. 57-100. 107. Stach BA, Spretnjak ML, Jerger J. The prevalence of central presbycusis in a clinical population. J Am Acad Audiol 1990;1:109-15. 108. Kelsall DC, Shallop JK, Burnelli, T. Cochlear implantation in the elderly. Am J Otol 1995;16:609-15. 109. Shin YJ, Fraysse B, Deguine O, et al. Benefits of cochlear implantation in elderly patients. Otolaryngol Head Neck Surg 2000;122:602-6. 110. Facer G, Peterson A, Brey R. Cochlear implantation in the senior citizen age group using the Nucleus 22channel device. Ann Otol Rhinol Laryngol 1995;166 187-90. 111. Francis HW, Chee N, Yeagle J, Cheng A, Niparko JK. Impact of cochlear implants on the functional health status of older adults. Laryngoscope 2002;112:1482-8. 112. Lea A, Hailey D. The cochlear implant. A technology for the profoundly deaf. Med Prog Technol 1995;21:47-52. 113. Lea A. Cochlear implants. Health technology series, No. 6. Canberra: Australian Institute of Health; 1991. 114. Hutton J, Politi C, Seeger T. Cost-effectiveness of cochlear implantation of children. A preliminary model for the UK. Adv Otorhinolaryngol 1995;50:201-6. 115. Carter R, Hailey D. Economic evaluation of the cochlear implant. Int J Tech Assess 1999;15:520-30. 116. Cheng A, Niparko J. Cost utility of the cochlear implant in adults. Arch Otolaryngol Head Neck Surg 1999;125:1214-8. 117. Carhart R. Auditory training. In: Davis H, editors. Hearing and deafness. New York: Rinehart & Co.; 1947. p. 276-99. 118. Schwartz MS, Otto SR, Brackmann DE, Hitselberger WE, Shannon RV. Use of a multichannel auditory brainstem implant for neurofibromatosis type 2. Stereotact Funct Neurosurg 2003;81(1–4):110-4. 119. Otto SR, Brackmann DE, Hitselberger W. Auditory brainstem implantation in 12- to 18-year-olds. Arch Otolaryngol Head Neck Surg 2004;130(5):656-9. 120. Rauschecker JP, Shannon RV. Sending sound to the brain. Science 2002;295(5557):1025-9. 121. Lenarz T, Lim HH, Reuter G, Patrick JF, Lenarz M. The auditory midbrain implant: a new auditory prosthesis for neural deafness-concept and device description. Otol Neurotol 2006;27(6):838-43.
184 Impaired Motor Function: Functional Electrical Stimulation R. B. Stein . A. Prochazka
Introduction Electrical stimulators to restore the rhythmicity of the heart (cardiac pacemakers) are currently being implanted in more than 400,000 people/ year [1]. These stimulators provide patterned activation of the heart muscle to replace the natural pacemaker function. Similarly, patterned stimulation of cochlear neurons can provide remarkable restoration of hearing in completely deaf individuals and these cochlear stimulators have been implanted in more than 80,000 individuals worldwide [2]. Finally, thousands of stimulators have been implanted to restore micturition (bladder stimulators) [3,4]and respiration (phrenic nerve stimulators) [5]. These are all examples of ‘‘functional electrical stimulation (FES)’’ in that the stimulus provides a pattern of stimulation that directly generates a functional movement or sensation. This is in contrast, for example, to a deep brain stimulator where a constant frequency of stimulation is applied that modulates the activity in particular brain regions. The stimulation may enable functional movements to occur or block unwanted movements from occurring, but the applied stimulus pattern has no direct relation to the movements. This chapter reviews the application of FES to restore limb movements in people who have a variety of CNS disorders (e.g., stroke, spinal cord injury (SCI), brain injury, multiple sclerosis). The number of people who could benefit from such an approach is potentially enormous. For example, there are more than 2 million stroke survivors in the U.S. alone and this disease constitutes the largest cause of disability in all #
Springer-Verlag Berlin/Heidelberg 2009
developed nations [6]. Yet, the total number of FES stimulators applied worldwide is on the order of 10,000 [7]. In this Chapter we will discuss some of the issues that have limited FES applications to restore limb movements. We will also argue that many of these issues are soluble and that the numbers of FES applications to limbs may increase rapidly in the next few years. The following sections will consider basic issues related to electrodes, stimulating and recording methods. The final sections will discuss application of FES to restore movements of the upper and lower extremities (arms and legs).
Electrodes Electrodes can be divided into three categories: (1) noninvasive surface electrodes, (2) minimally invasive electrodes and (3) fully implanted electrodes. Noninvasive electrodes are used widely for therapy, for example to strengthen muscles that have been weakened following disease or surgery. These electrodes typically consist of a flat conductor made of metal, carbon or carbonized rubber ranging from 1 to 25 cm2 in area with an interface of hydrogel or moistened material that delivers electrical current from the stimulator and its attached wires to the skin overlying motor points. The gel or absorbent material interface traps water and ions that are dissolved in the water. The current has both capacitive and resistive components. In electronics a capacitor is a device that has the ability or capacity to store charge. Electric charge delivered to the interface by the
3048
184
Impaired motor function: functional electrical stimulation
stimulator attracts equal and opposite charge to the dermis from deeper lying structures. The capacitive component of current corresponds to the charge stored on the interface and dermis as the voltage changes. The other, resistive, component of current comprises charge that actually crosses from the interface through the dermis and epidermis to the underlying tissues. Resistive currents have a greater chance of producing allergic reactions to the substances crossing the boundary or causing skin breakdown, if the currents applied are sufficiently large. Simple water-filled electrodes that pass ions dissolved in tap water were used for over 2,000 patients with stimulators made in the former Yugoslavia [8]. Water-filled electrodes are less likely to cause an allergic reaction, but may harbor bacteria or other organisms. If they dry out, they can develop a higher resistance to passage of current than a hydrogel. If the area through which water flows is limited, this may increase the chance of skin breakdown. Finally, a standard electrode gel, such as that used in EEG or ECG measurements can be applied to lower the resistance. However, these gels often contain a highly concentrated salt solution that will irritate the skin if used for several h/day as is typical in FES applications. All commercial, surface FES systems currently use hydrogels to our knowledge. A motor point is empirically found as a low threshold point for activating a particular muscle or nerve and is often near the neuromuscular junctions of a muscle. Even though the electrode is over a muscle all FES applications involve stimulating nerves. Muscle cells have a much higher electrical threshold. Though denervated muscles can be stimulated in experimental animals with very large, long-lasting pulses, these stimuli have not produced functional movements that could be used clinically [9]. Surface electrodes. The obvious advantages of surface electrodes compared to implanted electrodes are: (1) no surgery is required, thus reducing the associated costs, risks of infection,
postoperative pain, etc.; (2) a variety of placements and combinations can be easily tried to see which is most effective and (3) if the subject decides he/ she doesn’t want the stimulation or he/she recovers function through remission of the disease or regeneration, the electrodes are easily removed. However, surface electrodes have many disadvantages: (1) the greater the number of electrodes, the more tedious it is to apply them on a daily basis and the more variable the responses from day to day; (2) deep-lying muscles or nerves are difficult to activate; (3) passage of current through the skin may activate pain fibers in the skin and limit tolerable stimulus levels. Incomplete activation of an already weakened muscle may lead to inadequate functional movement. Minimally invasive electrodes, such as percutaneous wires, are an attempt to combine some of the good features of surface and fully implanted electrodes. The simplest have a hook or barb (‘‘tine’’) at the end and are commonly used for recording EMG or stimulating muscles selectively in a single experiment [10]. If implanted for longer periods of time in moving muscles, these electrodes will bend back and forth and eventually break. Forming the wire into a helix or providing other means of strain relief [11,12], as well as using improved materials, have increased longevity and the average time to breakage can be on the orders of months or even years [13]. In addition, since the wires emerge through the skin, a small, but significant risk of infection exists and care of the entry sites is an issue. Nonetheless, percutaneous electrodes offer a flexible method for trying out various potential FES approaches that could not be easily implemented if at all with surface electrodes. Successful approaches can then be implemented in systems that would be fully implanted at a later date. An interesting example of a minimally invasive approach is the use of BIONs (> Figure 184-1). A BION is a microstimulator encapsulated in a glass or ceramic case that can be implanted via a
Impaired motor function: functional electrical stimulation
184
. Figure 184-1 Devices under development. (a) Original design of the BION microstimulator designed to be inserted through a hypodermic needle into a muscle or near a nerve of interest. The electrodes are at the two ends of the device and a coil inside the glass package is used to receive power and control signals from an external coil (not shown). American and Canadian pennies are shown to indicate the overall size. (b) Stimulus Router System (SRS) comprising implanted leads that pick up some of the current delivered by an external stimulator through surface electrodes. (c) Schematic of a complete SRS system for eliciting hand opening and closing, triggered by an earpiece sensor that detects small voluntary tooth clicks
hypodermic needle [14,15]. Once the stimulation site is found the hypodermic needle is removed. The skin will seal around the puncture site so in effect the BION becomes a completely implanted device. However, we have included them in the section on minimally invasive approaches, since they do not require open surgery beyond that required to insert a hypodermic needle. BIONs have been used in FES and other applications [16,17]. BIONs contain a coil that receives radio frequency signals that are decoded to produce a pattern of stimulation. One transmitter can send signals to a number of BIONs so in principle they can be used for quite complex
FES systems. In the original design power as well as control signals were transmitted, and the efficiency of coupling is quite limited. The complexity of possible systems is limited by the number and placement of coils needed to communicate with and power the implanted BIONS. BIONs containing a rechargeable battery are being developed [18]. Without the need for transmitting power continuously, communication over longer distances is possible and a number of BIONs placed in various parts of the leg can be controlled by one central, external transmitter. However, the batteries would still need to be recharged periodically and the technical problems have not been fully overcome.
3049
3050
184
Impaired motor function: functional electrical stimulation
Fully implanted systems containing from 1 to 24 channels have been tested in human applications [19,20]. The simplest systems have been used for example to prevent foot drop, which can result from several CNS disorders and will be discussed in more detail below [21–23]. The most complex systems have been designed to restore walking and other functions in people with a complete, thoracic SCI. The 24-channel stimulator was a modified cochlear stimulator. However, in contrast to the cochlear application, where all the stimulation leads are localized in one place (the cochlea), wires were led from a pacemaker-like unit implanted in the chest to a variety of muscle and nerve locations in both legs. This involved three long surgeries and the project was discontinued after implantation in three subjects [20]. Also, the large volume and extent of the implant will increase the cost and risk of infection. Alternative, fully implanted systems have been considered. For example, the area of the spinal cord controlling the leg (the lumbar enlargement) is small (about 5 cm in length in humans), compared to the lengths discussed above which may be a meter or more from a chest-based stimulator to the distal parts of both legs. In addition, motor pools (groups of motor neurons innervating a particular muscle) are arranged quite systematically in the ventral horn of the spinal cord [24]. Stimulating through one electrode may activate a synergy involving muscles spanning the hip, knee and ankle. As few as four electrodes can produce alternating flexion and extension of the lower limbs in a cat with low current intensities [25], but unwanted co-contractions often occur, particularly after some weeks [26,27]. Although promising, this approach, known as intra-spinal micro-stimulation (ISMS), has only been studied at present in animal experiments and numerous obstacles must be overcome before it can be considered as a clinical modality. A novel approach to stimulating the nerves innervating muscles is the ‘‘Stimulus Router
System’’ (SRS). It comprises an implanted lead that picks up some of the current delivered through the skin by a surface stimulator and delivers it to a target nerve via a nerve cuff. There are no implanted electronic components. Animal data have shown that the SRS can activate target nerves and muscles without stimulating local nerves under the surface electrodes [28]. A recent test during human peripheral nerve surgery showed that the SRS works similarly in humans [29]. The SRS has the advantages of an implanted stimulator: selectivity, reproducibility and convenience, at a lower cost, since only passive leads are implanted, the stimulator remaining external.
Stimulating and Recording Methods Some of the issues associated with surface electrodes were discussed above. Implanted electrodes for FES applications have some of the same problems as those used for the other implanted stimulators discussed in the Introduction, but there are additional problems that will be discussed below. Classically, with reversible Ag/AgCl electrodes the reaction will be Ag þ Cl ¼ AgCl þ e Supplying electrons from the negative pole of a battery will force the reaction to the left and Cl will come off that electrode. At the other, positive electrode AgCl will be formed. As long as there is still a coating of AgCl at both electrodes the reaction is reversible but if the coating is dissolved, the Ag will go into solution and the electrode itself will be dissolved. Thus, stimulation pulses should be completely chargebalanced to prevent the coating at one electrode and eventually the electrode itself from being destroyed. In practice, the electrodes used in commonly implanted stimulators such as cardiac pacemakers,
Impaired motor function: functional electrical stimulation
cochlear stimulators, vagal nerve stimulators and phrenic nerve stimulators are made of stainless steel and/or platinum-iridium alloys. These metals are biologically inert, have surface oxide layers that resist corrosion and have a relatively high capacitive storage capability. Provided the current density (charge per unit area of metaltissue interface) and the charge per pulse are kept below specific limits, and biphasic pulses are used to minimize the net charge transferred per pulse, the electrochemical reactions that occur at these interfaces are reversible [30–33]. Reversible reactions are desirable because they are less likely to cause damage to the stimulated neurons, and because they avoid metal dissolution. The electrochemistry involved is complex and much effort has gone into developing equivalent circuit models and electrochemical models of the commonly used materials. Other materials such as sintered iridium, iridum oxide and tantalum oxide that all increase the charge storage capacity of surfaces have been evaluated [34] and more recently conductive polymer nanostructures have been developed with the aim of enlarging the effective contact areas at the electrode-tissue interface [35,36]. Regulatory agencies such as Health Canada and the Food and Drugs Administration in the USA have stringent requirements on the materials used, the quality of manufacture and documentation, the reporting of adverse events and risk factors such as nerve damage, postimplant infection and hazards such as the heating of implanted wires that may occur when diathermy or magnetic resonance imaging is used. Recording electrodes have considerably more problems than stimulating electrodes. In stereotactic surgery for deep brain stimulation the stimulating electrodes are also used for recording at the time of surgery to verify that the electrode is in the right location. However, the electrode is subsequently used for stimulation only. Over time several layers of tissue build up around the electrodes. This phenomenon is shown elegantly in > Figure 184-2 using immunohistochemistry of
184
an electrode implanted in a rat brain for 4 weeks [37]. Around the electrode is staining for ED1 in red, which is a marker for inflammation. Outside of this is a staining in green for GFAP, a marker for astrocytes and further out are stains for NeuN and NF which are markers for neuronal cell bodies and neurofilaments. The extent of these layers differs with different electrode types (e.g., silicon arrays or metal wires), whether the electrodes are tethered by lead wires or free to move, the amount of movement (e.g., electrodes in the spinal cord may be subjected to more relative movement than in the brain), the size and density of the array of electrodes. The build up of these tissue layers and the reduction in neuronal cell numbers will clearly affect the ability to record from relevant neurons. It may not compromise stimulation, because the stimulus level can be increased to stimulate more distal neurons, but a loss of specificity may occur. Despite these potential problems arrays of 100 electrodes have been implanted in the cortex of human subjects [38] who have very limited motor function. Useful recordings have been made for a number of months. As will be described below, the aim of the experiments was to record neural activity in motor cortex and associated areas related to intended movement. These signals can then be decoded to produce movement of a cursor on a screen or to control movements of a robot or the person’s own muscles through FES [38–40].
Leg Movements Liberson et al. [41] first proposed the use of electrical stimulation to treat the condition of foot drop that occurs after stroke and other central nervous system conditions. In able bodied people cortical control of dorsiflexor muscles is relatively strong, so flexion of the ankle is often compromised when the cortex or its connections to the spinal cord are damaged. During the swing
3051
3052
184
Impaired motor function: functional electrical stimulation
. Figure 184-2 Immunohistochemical staining of the area around an implanted electrode. (a) Shows combinations of the individual stains for ED1 (an antibody that recognizes some leukocyte-associated molecules and is a marker of inflammation), GFAP (glial fibrillary acid protein, an astrocyte specific cell marker), NeuN (a stain for neuronal cell bodies) and NF (neurofilaments) that are displayed in (b) Note that neurons are displaced 100 mm or more from the site of the electrode (indicated schematically as an orange oval at distance 0), which will affect the ability to record from neurons, but may not prevent the ability to stimulate neurons chronically. From Polikov et al. [37]
phase of the gait cycle the foot drops and may drag on the ground. Stimulation of the dorsiflexor muscles during swing will lift the foot and assist gait. Recently, the Food and Drug Administration (FDA) in the U.S. has approved several foot drop stimulators and these are available commercially in a number of countries: WalkAide (http://www. WalkAide.com), the Odstock Dropped foot stimulator (http://www.ODFS.com) and the L300 (http://www.Bioness.com). These systems all use surface stimulation, so no surgery is required. The WalkAide and Bioness L300 are shown in > Figure 184-3. Liberson et al. [41] used a heel switch for control of the foot-drop stimulator. Stimulation was turned on when the heel lifted off the
ground and turned off when the heel touched the ground again. This system is still used in most foot-drop stimulators, but requires the heel switch to be placed in a shoe and the presence of wires (ODFS) or telemetry (L300) to send the signals to the stimulator. The WalkAide uses a tilt sensor that measures the orientation of the leg with respect to gravity [42]. When the leg is tilted back behind the body at the end of stance the stimulator is turned on and when the leg is tilted forward at the end of swing, the stimulator is turned off. The tilt sensor is incorporated in the stimulator package without the need for external wires or telemetry and can be used with any type of footwear including no footwear (bare feet).
Impaired motor function: functional electrical stimulation
184
. Figure 184-3 The WalkAide foot drop stimulator (left) is about the size of an iPod unit (the size can be judged from the grey, bottom compartment that holds an AA 1.5 V battery). All the electronics including a tilt sensor for control are included in the upper compartment. The L300 foot drop stimulator (right) contains an electronics package and cuff on the leg, like the WalkAide, but has in addition an in-shoe wireless foot sensor and a remote control unit (not shown) that fits in a pocket or on a belt. Both devices have a remote clinician interface that allows parameters to be adjusted, and usage data and patient records to be stored, analyzed and printed
Implanted stimulators using open surgery [22,23,43] or BIONS [44] are also under development. The implanted devices offer the potential to have a more reproducible, balanced dorsiflexion during daily use. However, so far little or no additional functional benefit has been demonstrated in terms of speed or effort to walk with the implanted devices compared to the surface stimulation. Recently two types of implantable peroneal nerve stimulator have become available commercially in Europe, the Finetech STIMuSTEP (www.finetech-medical. co.uk) and the Neurodan ActiGait (www.neurodan.com). In a recent pilot study 15 individuals with footdrop due to stroke were implanted with the ActiGait system and showed improvements in gait [21]. Technical problems occurred with the stimulators, but these were resolved at follow-ups [21]. Like the original Liberson device, the STIMuSTEP and ActiGait stimulators are
triggered from a heel sensor. A new innovation that is still at the experimental stage is to implant a nerve cuff around a sensory nerve and use processed nerve signals to decide when to turn the stimulator on and off [22,45]. The nerve cuff can be implanted together with the stimulating leads of an implanted foot-drop stimulator. A nerve cuff will record signals from all the large fibers in a nerve and is less sensitive to the growth of connective tissue, so it should provide a small, but more stable signal. Careful comparisons will be required from large numbers of subjects using various surface and implanted devices. Indeed, a cost-benefit analysis is required, not only between implanted and surface stimulators, but between these two classes of device and an ankle-foot orthosis (AFO), a plastic brace that is most commonly prescribed for foot drop. An AFO passively holds the ankle in a neutral position, but has a number
3053
3054
184
Impaired motor function: functional electrical stimulation
of drawbacks. By bracing the ankle any residual dorsiflexion will be ineffective and the muscles may atrophy. In contrast, a recent study showed [46] that regular use of a foot-drop stimulator over a period of months increases the maximum voluntary contraction in the dorsiflexor muscles as well as the motor evoked potential (MEP) generated by stimulating the foot region of the motor cortex with transcranial magnetic stimulation (TMS). These results are in agreement with other recent studies [47–49] showing remarkable plasticity in cortical function produced by stimulation in adult humans. The mechanism of this plasticity is unknown, but appears to have much in common with long-term potentiation of sensory-motor circuits. Many injuries and disease processes will affect a variety of muscles in addition to the ankle dorsiflexors. Training programs using modified treadmills that are able to support part of the body’s weight [50–52] and robotic manipulators [53] are also effective in patterning the activity of various affected muscle groups and thereby improving walking after more extensive CNS damage. The training is labor-intensive and not effective for people with a complete SCI. Field-Fote et al. used several of these techniques [54] and they are now comparing their relative benefits for walking. FES systems have been developed that provide limited stepping capabilities for people with complete paraplegia at a thoracic level. The simplest system involves stimulating the quadriceps muscles to lock the knee in extension during stance and stimulating the common peroneal nerve to produce a flexion reflex that brings the leg forward during swing. Four channels of stimulation can therefore produce a basic bipedal gait [55]. Adding additional channels of stimulation such as those activating the gluteal or paraspinal muscle can enhance upright stance [56]. One system has received FDA approval (Parastep; www.sigmedics.com). Fatigue is the limiting factor because of the high level of stimulation to the
quadriceps muscle needed to prevent the knee from buckling. The walking is slow and limited to the order of 10m because of the very high energy consumption. Limited walking is also possible using existing braces such as the reciprocal gait orthosis (RGO). The RGO braces the ankle, knee and hip so a stifflegged gait results. A cable links the two legs so forward movement of the trunk on one side leads to the forward movement of the leg on that side with respect to the other. The system is timeconsuming to put on and take off and standing up is difficult. However, once the standing position is reached it can be maintained with little energy. An RGO also allows walking in people with complete thoracic spinal lesions with less energy than FES systems. Popovic et al. [57] first suggested the concept of a hybrid system that uses bracing to maintain an upright posture and FES for propulsion. Combining FES with an RGO can enhance the performance, compared to either system on its own [19,58]. New types of braces allow the knee joint to be locked and unlocked (stance control) and only control the joints from the knee down (knee-ankle-foot orthosis, KAFO). A stance control KAFO can be donned and doffed while sitting in a wheel chair and FES can be added to allow easier standing and propulsion for walking [59]. Since the hips are not controlled, trunk stability must be maintained by using the arms on a walker. About half the body weight is typically borne by the arms and fatigue of arm muscles becomes the limiting factor [59]. Several groups have developed systems using implanted electrodes to allow stimulation of more muscles and better control of the stepping movements (e.g., [11,20]). None of these systems has been commercialized and published data are based on a few intensively trained individuals. No studies directly compare the best implanted and the best surface systems, so the relative costs and benefits of surgery are still unknown. As well as energy cost, control is a limiting factor in more complex systems. The greater the
Impaired motor function: functional electrical stimulation
level of disability the greater the need for sensory feedback, but many of the complex systems incorporate little or no feedback control. For example, control of the Parastep stimulator is by hand switches so control depends on the voluntary activity of muscles above the SCI. As well as initiating each step, the arms are maintaining balance with a walker or forearm crutches, since the legs are operating without feedback (open loop) and this contributes in part to the high energy cost to the upper body. Various types of sensors, in addition to those mentioned above have been proposed including force sensors, accelerometers, goniometers, and gyroscopes [60–62]. However, if the sensors are external, they have to be placed on a daily basis which adds variability and time to put the system on. In a hybrid system sensors can be mounted on the braces, as can the stimulating electrodes and this reduces the donning time and improves reliability. Still another approach is to place recording arrays in the dorsal root ganglia supplying the legs [63]. This provides access to a large variety of sensors and could be implanted together with an intraspinal microstimulation system. However, the system is not yet feasible for human trials due to problems in long-term signal viability (> Figure 184-2). Future work is clearly needed on appropriate surface or implanted systems to provide enhanced movements and some feedback control to respond to fatigue, external obstacles and other challenges.
Arm Movements Vodovnik and colleagues in Ljubljana, Slovenia were the first to explore FES control of the upper limb [64,65]. In the late 1970s a therapeutic program for restoring hand function in stroke subjects was implemented at the Rancho Los Amigos Hospital in Los Angeles. Groups of participants performed FES-assisted biofeedback exercises daily [66]. In the mid-1990s, two
184
surface FES devices were developed for quadriplegic people, the Handmaster [67] and the Bionic Glove [68,69]. A similar device, the ETHZ Paracare, was developed at the Swiss Federal Institute of Technology Zurich, Switzerland [70,71]. The Handmaster was marketed in Holland for several years and recently became available in the USA as the Bioness H200. It consists of a hinged splint with in-built electrodes and a separate stimulator. Stimulation is triggered by a push-button switch. The Bionic Glove is a flexible garment with an integral stimulator and electrodes. It is triggered by moving the hand into flexion or extension, boosting tenodesis grasp and release. Both devices have been shown to help restore hand function in quadriplegic people [68,72,73] when used for therapeutic exercise and training. The Bionic Glove was also used as an aid in activities of daily life by some people. A new version of the Bionic Glove is being tested as part of a study involving in-home tele-rehabilitation in quadriplegic subjects in Edmonton [74]. The device, provisionally called the Hand-E-Stim, is a flexible garment with an integral stimulator the size of an iPod Nano. The Hand-E-Stim is triggered by a wireless sensor like a hearing aid that detects vibrations in the tissues in front of the ear. The user sequentially activates stimulation for hand opening and grasp with small tooth clicks [75]. The device will likely be available commercially in North America in 2009. The Hand-E-Stim has been designed to be used as an orthosis in activities of daily living, as well as for therapeutic training. Coordinated FES of several muscles of the forearm and upper arm has been tested experimentally in individuals with SCI who have paralyzed elbow, wrist and hand muscles [76]. A programmable multi-channel surface stimulator was used to activate muscles in a sequence that allowed forward reach, grasp and flexion. One of the problems with surface stimulation of large muscles such as biceps and triceps brachii is that
3055
3056
184
Impaired motor function: functional electrical stimulation
during activity the motor points of these muscles can move several centimeters under the skin. This movement changes the relationship between the stimulating electrode and the motor nerve. Thus, the amount of muscle activation changes as the elbow flexes and extends, which results in problems of control. Nonetheless, encouraging therapeutic results were reported in this study. An implanted stimulator, the Freehand System, was developed and tested in the late 1980s and 1990s at Case Western Reserve University (CWRU) [77]. The FDA approved it for commercial sale by Neurocontrol Corp in 1997. Freehand systems had been implanted in over 200 individuals by the year 2001. An external radiofrequency control unit activated an implanted device the size of a cardiac pacemaker, which then generated pulse trains and delivered them through epimysial electrodes to the targeted muscles. Signals from transducers monitoring voluntary shoulder or wrist movements were used to control the stimulation of the implanted muscles to produce a variety of hand movements. A multicentre study on 50 of the recipients showed that their hand function improved considerably while using the device. Unfortunately, the Freehand System was withdrawn from the market in 2002 for a variety of reasons that have been analyzed in detail in a Princeton University thesis [78]. A successor to the Freehand System, also developed at CWRU, has been implanted in seven SCI individuals [79]. This device is controlled by electromyographic signals picked up from muscles still under the user’s voluntary control. The biceps muscle is activated as well as the muscles eliciting prehension. In a more recent report six subjects were implanted with a second-generation neuroprosthesis consisting of 12 stimulating electrodes, two EMG signal recording electrodes, an implanted stimulatortelemeter device, an external control unit and a transmit/receive coil [80]. Three of the subjects were monitored for at least 2 years. EMG signals
could be recorded from voluntary muscles in the presence of electrical stimulation of nearby muscles. All three subjects had significantly increased pinch force and grasp. At least five tasks in the Activities of Daily Living Abilities Test improved. Each subject was able to use the device at home. Given the ability to stimulate several muscles in the arm and hand in future systems, the control problem becomes significant. Musculoskeletal models of the whole upper extremity are being developed that will allow the synthesis of movements with electrical stimulation to be tested and optimized [39]. Another new development is the Finetech STIMuGRIP [81] which has been implanted in a number of hemiparetic people and the SRS described above, which was implanted in the first human recipient in June 2008. One of the important issues from a reimbursement point of view is the relative efficacy of implanted systems compared to surface stimulators. Implanted systems are more selective in stimulating the desired muscles, and have the potential to be more reproducible in their action from one day to the next, but they are also at least an order of magnitude more expensive. Therefore, their advantages must be analyzed in carefully designed comparative studies, preferably using quantified outcome measures. An important step in this direction is the recent analysis of cost savings of bladder, bowel and upper extremity neural prostheses [82,83]. The cost of the most expensive hand grasp neuroprosthesis would be recovered over the lifetime of the user if the time a personal caregiver was needed was reduced by just 2 h/ day [83]. It will be interesting to analyze the time to cost recovery of simpler systems such as the StimuGRIP and SRS once more experience is gained with them. As the level of a lesion becomes higher, more muscles need to be controlled, but the available voluntary control sites become fewer. One potential solution to this problem is to record neural activity from the motor cortex and use these
Impaired motor function: functional electrical stimulation
signals to control external devices or the person’s own muscles through FES. These neural signals may be available even in people who can activate very few if any skeletal muscles. Donoghue and his colleagues [38] have used a 10 10 array of microelectrodes inserted in the motor cortex of such patients. Single and multi-unit neural signals could be discriminated from noise on many of the electrodes for a period of several months. Further, these signals could be decoded on-line and used to move a cursor on a screen to operate environmental controls, do email, etc. This is an interesting proof of principle, but the data rates are so slow that the system is unlikely to be accepted in its current form. Velliste et al. [40] reported on experiments using an array of electrodes in motor cortex of monkeys. The signals again were processed on-line and the monkey could use the signals to manipulate a robot to feed itself. Although still slower than the normal feeding movements in these intact animals, the speeds were fast enough that they would be practical for feeding or other tasks. Why better control was obtained in monkeys, compared to humans, remains uncertain. These developments are exciting, but many questions remain about the reliability, practicality and durability of such systems. The number of patients who can not use external systems is limited and the relative benefits of external versus implanted systems in relation to cost remains to be determined. With the advent of more affordable and convenient implanted systems for less severely disabled individuals the future looks bright for implantable neural prostheses to restore motor function in the arms and hands. Similarly, as described in an earlier section, a range of options has recently become available for neural prostheses for leg function. Thus, although the field has developed more slowly than for heart pacemakers or cochlear prostheses, we are hopeful that widespread clinical acceptance of affordable neural prosthesis implants will occur in the next few years. These successes can then serve
184
as a springboard for development of more sophisticated and effective devices to assist with a range of arm and leg functions.
References 1. Brunner M, Olschewski M, Geibel A, et al. Long-term survival after pacemaker implantation. Prognostic importance of gender and baseline patient characteristics. Eur Heart J 2004;25:88-95. 2. Gubbels SP, McMenomey SO. Safety study of the Cochlear nucleus 24 device with internal magnet in the 1.5 tesla magnetic resonance imaging scanner. Laryngoscope 2006;116:865-71. 3. Gaunt RA, Prochazka A. Control of urinary bladder function with devices: successes and failures. Prog Brain Res 2006;152:163-94. 4. Rijkhoff NJ. Neuroprostheses to treat neurogenic bladder dysfunction: current status and future perspectives. Childs Nerv Syst 2004;20:75-86. 5. Elefteriades JA, Quin JA. Diaphragm pacing. Chest Surg Clinics North Am 1998;8:331-57. 6. Wolf P. An overview of the epidemiology of stroke. Stroke 1990;21:4-6. 7. Stein RB, Mushahwar V. Reanimating limbs after injury or disease. Trends Neurosci 2005;28:518-24. 8. Kralj A, Acimovic R, Stanic U. Enhancement of hemiplegic patient rehabilitation by means of functional electrical stimulation. Prosthet Orthot Int 1993;17:107-14. 9. Modlin M, Forstner C, Hofer C, et al. Electrical stimulation of denervated muscles: first results of a clinical study. Artif Organs 2005;29:203-6. 10. Basmajian JV. Muscles alive: their functions revealed by electromyography. 2nd edn. Baltimore, MD, Williams & Wilkens; 1967. 11. Marsolais EB, Kobetic R. Functional electrical stimulation for walking in paraplegia. J Bone Joint Surg 1987;69A:728-33. 12. Peckham PH, Marsolais EB, Mortimer JT. Restoration of key grip and release in the C6 tetraplegic patient through functional electrical stimulation. J Hand Surg [Am] 1980;5:462-9. 13. Handa Y, Hoshimiya N, Iguchi Y, et al. Development of percutaneous intramuscular electrode for multichannel FES system. IEEE Trans Biomed Eng 1989;36:705-10. 14. Loeb GE, Zamin CJ, Schulman JH, et al. Injectable microstimulator for functional electrical stimulation. Med Biol Eng Comp 1991;NS13–9. 15. Schulman JH, Loeb GE, Gord JC, et al. Structure and method of manufacture of an implantable microstimulator, U.S. Patent #5193540;1993. 16. Dupont A-C, Bagg SD, Creasey JL, et al. First patients with BION implants for therapeutic electrical stimulation. Neuromodulation 2004;7:38-47.
3057
3058
184
Impaired motor function: functional electrical stimulation
17. Weber DJ, Stein RB, Chan KM, et al. Functional electrical stimulation using microstimulators to correct foot drop: a case study. Can J Physiol Pharmacol 2004;82:784-92. 18. Schulman JH, Mobley JP, Wolfe J, et al. Battery powered BION FES network. Conf Proc IEEE Eng Med Biol Soc 2004;6:4283-6. 19. Marsolais EB, Kobetic R, Polando G, et al. The Case Western Reserve University hybrid gait orthosis. J Spinal Cord Med 2000;23:100-8. 20. Smith BT, Johnston TE, Betz RR, et al. Implanted FES for upright mobility in pediatric spinal cord injury: collective experience with two multi-channel systems. In: Tenth International FES Society Meeting, Montreal; 2005. p. 306–8. 21. Burridge JH, Haugland M, Larsen B, et al. Phase II trial to evaluate the ActiGait implanted drop-foot stimulator in established hemiplegia. J Rehabil Med 2007;39:212-8. 22. Hoffer JA. Initial results with fully implanted Neurostep FES system for foot drop. In: Tenth Annual IFESS Meeting, Montreal; 2005. p. 53–5. 23. Kottink AIR, Buschman HPJ, Kenney LPJ, et al. The sensitivity and selectivity of an implantable two-channel peroneal nerve stimulator system for restoration of dropped foot. Neuromodulation 2004;7:277-83. 24. Yakovenko S, Mushahwar V, VanderHorst V, et al. Spatiotemporal activation of lumbosacral motoneurons in the locomotor step cycle. J Neurophysiol 2002;87:1542-53. 25. Mushahwar VK, Gillard DM, Gauthier MJ, et al. Intraspinal micro stimulation generates locomotor-like and feedback-controlled movements. IEEE Trans Neural Syst Rehabil Eng 2002;10:68-81. 26. Moritz CT, Lucas TH, Perlmutter SI, et al. Forelimb movements and muscle responses evoked by microstimulation of cervical spinal cord in sedated monkeys. J Neurophysiol 2007;97:110-20. 27. Prochazka A. Neuroprosthetics. In: Field-Fote E, editor. Spinal cord injury rehabilitation. Philadephia, PA: FA Davis; 2008. 28. Gan LS, Prochazka A, Bornes TD, et al. A new means of transcutaneous coupling for neural prostheses. IEEE Trans Biomed Eng 2007;54:509-17. 29. Prochazka A, Gan L, Olson J, et al. The stimulus router system (SRS): first human intra-operative testing of a novel neural prosthesis. In: Neural Interfaces, Cleveland OH; 2008. 30. McCreery DB, Agnew WF, Yuen TG, et al. Charge density and charge per phase as cofactors in neural injury induced by electrical stimulation. IEEE Trans Biomed Eng 1990;37:996-1001. 31. Mortimer JT. Motor prostheses. In: Brooks VB, editor. Handbook of physiology section 1: the nervous system. Bethesda MD: American Physiological Society; 1981. p. 155-87. 32. Peterson DK, Nochomovitz ML, Stellato TA, et al. Longterm intramuscular electrical activation of the phrenic
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
nerve: safety and reliability. IEEE Trans Biomed Eng 1994;41:1115-26. Scheiner A, Mortimer JT, Roessmann U. Imbalanced biphasic electrical stimulation: muscle tissue damage. Ann Biomed Eng 1990;18:407-25. Troyk PR, Detlefsen DE, Cogan SF, et al. ‘‘Safe’’ chargeinjection waveforms for iridium oxide (AIROF) microelectrodes. Conf Proc IEEE Eng Med Biol Soc 2004;6:4141-4. Abidian MR, Martin DC. Experimental and theoretical characterization of implantable neural microelectrodes modified with conducting polymer nanotubes. Biomaterials 2008;29:1273-83. Kim DH, Abidian M, Martin DC. Conducting polymers grown in hydrogel scaffolds coated on neural prosthetic devices. J Biomed Mater Res A 2004;71:577-85. Polikov VS, Tresco PA, Reichert WM. Response of brain tissue to chronically implanted neural electrodes. J Neurosci Methods 2005;148:1-18. Hochberg LR, Serruya MD, Friehs GM, et al. Neuronal ensemble control of prosthetic devices by a human with tetraplegia. Nature 2006;442:164-71. Blana D, Hincapie JG, Chadwick EK, et al. A musculoskeletal model of the upper extremity for use in the development of neuroprosthetic systems. J Biomech 2008;41:1714-21. Velliste M, Perel S, Spalding MC, et al. Cortical control of a prosthetic arm for self-feeding. Nature 2008;453 (7198):1098-101. Liberson WT, Holmquest HJ, Scott D, et al. Functional electrotherapy, stimulation of the peroneal nerve synchronized with the swing phase of the gait of hemiplegic patients. Arch Phys Med 1961;42:101-5. Stein RB. Assembly for functional electrical stimulation during movement. Continuation in part, U.S. Patent #5814093;1998. O’Halloran T, Haugland M, Lyons GM, et al. Modified implanted drop foot stimulator system with graphical user interface for customised stimulation pulse-width profiles. Med Biol Eng Comput 2003;41:701-9. Weber DJ, Stein RB, Chan KM, et al. BIONic WalkAide for correcting foot drop. IEEE Trans Neural Syst Rehabil Eng 2005;13:242-6. Haugland M, Sinkjaer T. Cutaneous whole nerve recordings used for correction of footdrop in hemiplegic man. IEEE Trans Rehabil Eng 1995;3:307:317. Stein RB, Everaert DG, Chong S, et al. Using FES for foot drop strengthens cortico-spinal connections. In: 12th Annual Conference International FES Society, Philadelphia; 2007. Khaslavskaia S, Sinkjaer T. Motor cortex excitability following repetitive electrical stimulation of the common peroneal nerve depends on the voluntary drive. Exp Brain Res 2005;162:497-502. Knash ME, Kido A, Gorassini M, et al. Electrical stimulation of the human common peroneal nerve elicits lasting
Impaired motor function: functional electrical stimulation
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
facilitation of motor-evoked potentials. Exp Brain Res 2003;221:366-77. Roy FD, Norton JA, Gorassini MA. Role of sustained excitability of the leg motor cortex after transcranial magnetic stimulation in associative plasticity. J Neurophysiol 2007;98:657-67. Barbeau H, Blunt RA. A novel interactive locomotor approach using body weight support to retrain gait in spastic paretic subjects. In: Wernig A, editor. Plasticity of motoneuronal connections. Elsevier; Amsterdam: p. 461-74. Harkema SJ, Hurley SL, Patel UK, et al. Human lumbosacral spinal cord interprets loading during stepping. J Neurophysiol 1997;77:797-811. Wernig A, Muller S. Laufband locomotion with body weight support improved walking in persons with severe spinal cord injuries. Paraplegia 1992;30:229-38. Hesse S, Schmidt H, Werner C, et al. Upper and lower extremity robotic devices for rehabilitation and for studying motor control. Curr Opin Neurol 2003;16:705-10. Field-Fote EC. Combined use of body weight support, functional electric stimulation, and treadmill training to improve walking ability in individuals with chronic incomplete spinal cord injury. Arch Phys Med Rehabil 2001;82:818-24. Kralj A, Bajd T. Functional electrical stimulation, standing and walking after spinal cord injury. Boca Raton, FL: CRC Press; 1989. Graupe D, Kohn KH. Functional Electrical stimulation for ambulation by paraplegics: twelve years of clinical observations and system studies. Malabar, FL: Krieger; 1994. Popovic D, Tomovic R, Schwirtlich L. Hybrid assistive system: neuroprosthesis for motion. IEEE Trans Biomed Eng 1989;BME37:729-38. Solomonow M, Aguilar E, Reisin E, et al. Reciprocating gait orthosis powered with electrical muscle stimulation (RGO II). Part I: performance evaluation of 70 paraplegic patients. Orthopedics 1997;20:315-24. Stein RB, Hayday F, Chong SL, et al. Speed and efficiency in walking and wheeling with novel stimulation and bracing systems after spinal cord injury: a case study. Neuromodulation 2005;8:264-71. Luinge HJ, Veltink PH. Inclination measurement of human movement using a 3-D accelerometer with autocalibration. IEEE Trans Neural Syst Rehabil Eng 2004;12:112-21. Mayagoitia RE, Nene AV, Veltink PH. Accelerometer and rate gyroscope measurement of kinematics: an inexpensive alternative to optical motion analysis systems. J Biomech 2002;35:537-42. Williamson R, Andrews BJ. Sensor systems for lower limb functional electrical stimulation (FES) control. Med Eng Phys 2000;22:313-25. Weber DJ, Stein RB, Everaert DG, et al. Limb-state feedback from ensembles of simultaneously recorded
64.
65.
66.
67. 68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
184
dorsal root ganglion neurons. J Neural Eng 2007;4: S168-80. Rebersek S, Vodovnik L. Proportionally controlled functional electrical stimulation of hand. Arch Phys Med Rehabil 1973;54:378-82. Vodovnik L, Crochetiere WJ, Reswick JB. Control of a skeletal joint by electrical stimulation of antagonists. Med Biol Eng 1967;5:97-109. Waters R, Bowman B, Baker L, et al. Treatment of hemiplegic upper extremity using electrical stimulation and biofeedback training. In: Popovic DJ, editor. Advances in external control of human extremities, Yugoslav Committee for Electronics and Automation; Belgrade: p. 251-66. Nathan RH. Device for generating hand function, US Patent #5,330,516;1994. Prochazka A, Gauthier M, Wieler M, et al. The bionic glove: an electrical stimulator garment that provides controlled grasp and hand opening in quadriplegia. Arch Phys Med Rehabil 1997;78:608-14. Prochazka A, Wieler M, Kenwell Z. Garment for applying controlled electrical stimulation to restore motor function, U.S. Patent #5,562,707;1996. Mangold S, Keller T, Curt A, et al. Transcutaneous functional electrical stimulation for grasping in subjects with cervical spinal cord injury. Spinal Cord 2005;43:1-13. Popovic MR, Thrasher TA, Zivanovic V, et al. Neuroprosthesis for retraining reaching and grasping functions in severe hemiplegic patients. Neuromodulation 2005;8:58-72. Alon G, McBride K. Persons with C5 or C6 tetraplegia achieve selected functional gains using a neuroprosthesis. Arch Phys Med Rehabil 2003;84:119-24. Popovic D, Stojanovic A, Pjanovic A, et al. Clinical evaluation of the bionic glove. Arch Phys Med Rehabil 1999;80:299-304. Kowalczewski JA, Chong S, Prochazka A. Home-based upper extremity rehabilitation in spinal cord injured patients, Vol 33. San Diego, CA: Society for neuroscience; 2007. p. 75.71/GG71. Prochazka A. Method and apparatus for controlling a device or process with vibrations generated by tooth clicks, US Patent # 6961623;2005. Popovic MR, Thrasher TA, Adams ME, et al. Functional electrical therapy: retraining grasping in spinal cord injury. Spinal Cord 2006;44:143-51. Peckham PH, Keith MW, Kilgore KL, et al. Efficacy of an implanted neuroprosthesis for restoring hand grasp in tetraplegia: a multicenter study. Arch Phys Med Rehabil 2001;82:1380-88. Hall SW. Commercializing neuroprostheses: the business of putting the brain back in business. Princeton NJ: Princeton University; 2003. p. 185. Kilgore KL, Hart RL, Montague FW, et al. An implanted myoelectrically-controlled neuroprosthesis for upper
3059
3060
184
Impaired motor function: functional electrical stimulation
extremity function in spinal cord injury. Conf Proc IEEE Eng Med Biol Soc 2006;1:1630-33. 80. Kilgore KL, Hoyen HA, Bryden AM, et al. An implanted upper-extremity neuroprosthesis using myoelectric control. J Hand Surg [Am] 2008;33:539-50. 81. Spensley J. STIMuGRIP(R); a new Hand Control Implant. Conf Proc IEEE Eng Med Biol Soc 2007;1:513.
82. Creasey GH, Dahlberg JE. Economic consequences of an implanted neuroprosthesis for bladder and bowel management. Arch Phys Med Rehabil 2001;82:1520-5. 83. Creasey GH, Kilgore KL, Brown-Triolo DL, et al. Reduction of costs of disability using neuroprostheses. Assist Technol 2000;12:67-75.
1 8 2 Impaired Vision: Visual Prosthesis J. P. Girvin . A. G. Martins
The Idea In the eighteenth century, Benjamin Franklin, from his observations and experiments with electricity, appreciated that application of electrical stimulation to the brain could result in some type of visual sensation, and this formed part of a scientific address in 1751[55]. In 1953, Krieg outlined ‘‘two unexplored paths,’’ the exploitation of which might result in aids for the blind – i.e., electrically induced stimuli applied to the skin (sensory prosthesis) or directly to the visual cortex (electroneuroprosthesis). He described the ways in which light signals from the environment might be transduced into electrical stimuli, which might be patterned in order to provide orientation of visual patterns. In 1955, Shaw proposed the use of photoelectric reception of light from the periphery to charge condensers, which would, in turn, ignite gas triode tubes. Through coupling, these could result in application of electrical signals to the visual cortex through stimulating electrodes. He considered that with a number of photoelectric cells, patterns of visual sensation might be created. Van den Bosch [54] proposed ‘‘giving the blind a sensation of light and shadow’’ and eventually ‘‘a picture of the outside world by transducing the electroretinogram, recorded in response to visual stimulation, into an electrical stimulus applied to the visual cortex.’’ He considered that a reproduction of the exact electroretinographic wave form, if conveyed to the visual cortex, ‘‘would no doubt be interpreted by the brain as a picture.’’ #
Springer-Verlag Berlin/Heidelberg 2009
In 1962, Button and Putnam demonstrated that light signals, picked up by a photoelectric cell, could be transduced into electrical signals, which, when applied through small wires to the visual cortex, allowed blind volunteers to identify the light source by the appropriate orientation of the hand-held photoelectric cell bank. The most successful of the three cases was a 48-year-old man who was able to follow a flashlight carried by an attendant 15 ft away. Thus, by the early 1960s, the improvement in photoelectric cell technology, transduction of biological signals, and stimulating electrode arrays led numerous investigators to suggest that the blind might be provided with devices that would not only allow them to perceive levels of environmental illumination but might also enable them to identify bright objects in the environment and perhaps even to appreciate patterns of visual perception.
Early-Twentieth-Century Neurosurgical Observations The early observations of German neurosurgeons operating upon patients under local anesthesia disclosed clearly that punctate stimulation of the primary visual cortex gave rise to motionless dots of light in the contralateral visual field. [36,25,37] Further, it was shown that these dots of light formed, at least in a rudimentary fashion, a visuotopic map [25] not dissimilar to what would have been predicted on the basis of Holmes’s clinical observations (1918) regarding the visual field deficits produced in soldiers during
3010
182
Impaired vision: visual prosthesis
the First World War who suffered occipital lobe injuries. Penfield and colleagues in the early 1950s described stimulation of the visual cortex, noting that resulting sensations could take a number of forms (stars, wheels, disks, spots, lines, etc.), which could be moving and which were often reported as being colored [44,43].
Late-Twentieth-Century Experimentation The modern era of artificial vision research by electrocortical stimulation was ushered in by Brindley and colleagues, taking advantage of these neurosurgical observations and the advancing technology of biological stimulation. He hypothesized that 50 channels of stimulation, appropriately placed, should allow the reading of ten letters in a single fixational pause and that perhaps as few as 600 channels would allow normal reading[7]. In 1968, Brindley [8,9,39] reported the use of a prosthetic implant of 80 platinum electrodes applied to the right occipital cortex of a blind patient. These were connected to a series of radio receivers placed beneath her scalp. These, in turn, could be stimulated radiotelemetrically by an array of transmitters incorporated into a helmet placed externally to the scalp over the receivers. These experiments gave rise to a number of important observations regarding the punctate sensations of light, called phosphenes. They disclosed that (1) very small punctate phosphenes could be produced; (2) the resulting phosphenes corresponded roughly to Holmes’s visuotopic, or retinotopic, map with varying unexplained discontinuities; (3) the size of the phosphene varied directly with distance from the foveal point of fixation; (4) the phosphenes were colorless; (5) brightness modulation of the phosphenes could be produced by varying, within limits, any stimulus parameter; (6) two phosphenes could be distinguished from one another when induced by an electrode pair
separated by as little as 2.4 mm on the cortical surface; (7) more than one phosphene might be seen from a single electrode at stimulation strengths well above threshold; (8) phosphenes move in the direction of eye movement; (9) simultaneous stimulation of two closely applied electrodes might give rise to complex interactions; (10) simultaneous stimulation of several electrodes allows the recognition of simple patterns; (11) phosphenes usually flickered regardless of stimulation frequency; (12) there was absence of any definite flicker fusion frequency of stimulation; (13) stimulation less than 1.5 times threshold current resulted in the immediate cessation of the phosphene at the end of the stimulation train; and (14) persistence of phosphenes could occur for as long as 2 min with the absence of evidence of enlargement or change of position in the visual field. Although of no practical benefit, this first prototype visual device remained functional 3 years later[10]. Brindley considered that many more cortical electrodes were required in order to provide sufficient phosphenes for delineation of all the letters of the alphabet and thus to enable the patient to read. A second patient was operated upon by Brindley and colleagues in 1972 [14,11,12,48]. This 64-year-old man had been blind for 30 years from retinitis pigmentosa. The phosphenes in this patient were in general large. There were a number of discontinuities such that stimulationinduced effects could not be predicted on the basis of the Holmes’s visuotopic map. Further, electrodes located close together on the cortex might give phosphenes far apart in the visual field, and vice versa. There appeared to be no difference in the character of phosphenes produced from areas 17 and 18. Brindley was responsible for a third implantation – a second implant in the first patient fitted with 151 electrodes. Other than the report that stimulation of 120 of these electrodes resulted in the production of phosphenes [13] no further reports have appeared.
Impaired vision: visual prosthesis
Following the early observations of Brindley on his first patient, there was rekindling of interest in a true visual prosthesis based upon electrocortical stimulation. An international conference was held, with representation from most interested parties, in an attempt to outline plans for future visual prosthetic research [51]. Three clearly defined independent courses of research emerged from this. Brindley planned experiments with further implants, attempting to increase the sophistication of pattern presentation. The National Institutes of Health established the Neural Prosthesis Program under the directorship of Frank and Hambrecht with the primary purpose of examining the ideal biocompatible electrode materials for chronic stimulation, the identification and minimization of electrochemical injury during the course of such stimulation, and animal experiments aimed at identifying any pathological effects of chronic stimulation. The reader is referred to various publications recording the very important contributions made through this Program [2,3,15,30,31,46,47,57]. Dobelle and colleagues, working in Salt Lake City at the University of Utah as part of the Neural Prosthesis Program, attempted to determine feasibility issues with respect to an eventual electrocortical prosthesis by observations of acute stimulation of the visual cortex in North American patients undergoing operations under local anesthesia for treatment of various lesions (e.g., epilepsy, arteriovenous malformations, tumors, etc.). In the early 1960s, the acute stimulation experiments involving the visual cortex in sighted individuals by Dobelle and Mladejovsky [19] and their colleagues disclosed a number of points; (1) suitable phosphenes could be produced only from the primary visual cortex; (2) electrode size, and thus current density, had little effect on the threshold for phosphene production; (3) with the exception of some change in threshold, stimulus parameters had little effect on the induced subjective sensation; (4) constant current thresholds ranged from 1 to 5 mA;
182
(5) brightness modulation could be achieved using pulse durations less than 100 ms; (6) suprathreshold stimulation could produce a complex variety of multiple phosphenes; (7) most of the phosphenes were colored; (8) thresholds increased and phosphenes faded with continuous stimulation of greater than 10–15 s; (9) phosphenes moved with eye movements; and (10) the phosphenes appeared to be coplanar. Following this work, it was concluded that the point of diminishing returns for scientific observations with such patients had been reached. In 1973, Dobelle and colleagues implanted two blind patients who had been training in the Visual Prosthesis Program at the University of Utah for the previous 4 years. They were implanted with arrays of 64 platinum disk electrodes, applied to the right visual cortex. Many of the observations in Brindley’s first patient were replicated in these experiments. The size of the phosphenes ranged from ‘‘a grain of rice’’ to a ‘‘coin’’ at arm’s length, with the smaller phosphenes appearing nearer the occipital pole (central vision), as had been found by Brindley. Some phosphenes flickered while others did not. The larger peripheral phosphenes had an orange hue, whereas the smaller phosphenes were colorless. Brightness modulation could be achieved by changes in pulse amplitude and both patients reported that the phosphene faded when stimulation train lengths exceeded 10–15 s. All phosphenes were coplanar and phosphenes moved uniformly with eye movement. Thresholds in the patient who had been blind for 7 years ranged from 0.6 to 2.8 mA (average 1.3 mA) and those for the patient blind for 28 years from 2.1 to 8.1 mA (average 4.5 mA). Using appropriately placed phosphenes in ‘‘phosphene space,’’ interlaced stimulation of multiple electrodes (up to seven) allowed the recognition of simple patterns and letters. Finally, a number of both positive and negative interactions were seen when multiple electrodes were stimulated simultaneously; that is to say, the stimulation of a number of
3011
3012
182
Impaired vision: visual prosthesis
electrodes resulted in a greater number of phosphenes being seen (positive interaction) or a lesser number (negative interaction) [19]. Since these were the initial operations that were carried out (by JPG) in normal individuals and since the rather bulky electrodes were brought out through the scalp directly, it was declared pre-operatively that they would not be left in place for more than 72 h. Given the commitment of the patients the experimental team felt that there had to be some way of increasing the experimental time. The following year, 1974, a third patient underwent a further planned acute implantation with a modified electrode and connecting cable, which was tunneled beneath the scalp over a distance of 15 cm. It was thought that as much as ten days of experimentation might be realized. The difficult implantation was associated with more retraction (of the brain) than usual and the subsequent transient post-operative swelling precluded any real useful observations being achieved before a week. Thus, a clear commitment. . . . was made to chronic implantation in the ensuing experimental efforts. Chronic implantations were then carried out in two patients in 1975 . . . in which the electrode array connector was brought out through a postauricular pyrolytic carbon pedestal. Experiments on the second of these patients over a number of years have led to further observations: (1) the average threshold was 1.8 mA zero-to-peak (range 0.8–5.2 mA); (2) the average variability in threshold over a 7-month period was 25%; [29] (3) over a period of 10 years there was no further variation (approximately 10%) in threshold than is seen over hours or days (unpublished observations); (4) temporal integration of visual cortical stimulation was critically related to the train duration. Indeed temporal integrative factors possibly accounted for the psychophysical functions assessed when thresholds were determined in response to different stimulus parameters; [29] (5) the plotting of phosphenes in the visual field disclosed that they were clustered
around the point of fixation and the vertical meridian, in keeping with what might have been predicted from surface stimulation of the visual cortex on the basis of Holmes’s observations; (6) he was easily able to learn ‘‘phosphene braille’’ by using an appropriately constructed array of six phosphenes [20]. He was a poor tactile braille reader, but in the initial session was able to recognize letters three times more accurately with ‘‘cortical braille’’ than tactile braille. With the phosphene braille he was able to read simple sentences in a minute; [20] and (7) with the use of a TV camera on a joystick, operated by the subject, he was able to recognize the orientation of strips of white (adhesive tape) placed on a black (felt) background [20,23]. In 1978 two further patients were implanted with right occipital visual cortical arrays of 64 electrodes. One of the patients, who was blind from very shortly after birth and had no memory of visual imagery, never ‘‘saw’’ anything from stimulation over the course of years. From time to time he seemed to have some type of perceptual alteration, but an understanding of this was never achieved. The increased power of small batteries, the improved quality of edge detection software and the miniaturization of hardware in the latter part of the 1990s allowed the hundreds of kilograms of equipment of the 1970s to be replaced by a belt pack, weighing a few kilograms, which could be utilized independently in walking. The increased processing power of the computers was also a very important advancement as in the 1970s, it would take considerable time to process one image, whereas currently at least ten frames per second are easily achievable (logically, with the edge detection included). If the frame rate was too slow, it would render the system impractical. The other patient, known as ‘‘Jerry,’’ was able to walk and follow a wide strip of black tape on the floor or to follow a child down a hall. He eventually satisfactorily achieved a task involving the removal of a dark hat from a peg on a white
Impaired vision: visual prosthesis
wall, of then turning around and identifying a mannequin, on the other side of the room to which he walked over and upon which he placed the hat! (This achievement of this task led him to being placed in the 2000 edition of Ripley’s Believe It or Not!). Bak and colleagues have shown that stimulation currents of as little as units to tens of microamperes delivered intracortically in human occipital cortex can result in the production of phosphenes [5,41]. Further, in these preliminary experiments, phosphene interactions seemed less likely to occur. More recently, Hambrecht and colleagues have implanted an individual chronically with intracortical electrodes and carried out a number of experiments over a 4-month period [32]. Threshold ranged from 2 to 20 mA (biphasic, cathodic first, capacitor-coupled pulses; frequency 200-Hz pulse width 200 ms, train length 125 m). Other potentially important findings relating to cortical prosthetic research included the fact that intracortical electrodes in this patient separated by as little as 500 mm gave rise to phosphenes that could be resolved by the patient. Another interesting finding was the first observation to date of absence of coplanarity of the phosphenes in the stimulation of the occipital cortex.
Twenty-First Century Observations In April, 2002 Dobelle and colleagues implanted eight patients with bilateral occipital implants, each array consisting of 68–72 stimulating contacts. The majority of the patients ‘‘saw’’ something of the order of 90–100 phosphenes. A further eight patients, over the ensuing 2 years, were similarly implanted, three with arrays consisting of 242 electrodes. There were a number of differences between these and those used in the volunteers of the late twentieth century. Firstly, the substrate of the array was a thin, more flexible silastic. Secondly,
182
the wires of the connector were stainless steel, not platinum. Thirdly, the return electrode consisted of a large subgaleal platinum ground plane, in the majority of patients. Fourthly, the percutaneous pedestal consisted of titanium, not pyrolytic carbon. There were some modifications of some of the electrodes, but they all were of the same basic design as those used in the volunteers. The current authors were not directly involved in the scientific evaluation of these prostheses. However, there is little doubt that the resulting ‘‘vision,’’ experienced by the patients, was greatly enhanced over any previous observations. Small objects, such as pencils and screwdrivers, could be ‘‘seen’’ and accurately identified on surfaces in which there was sufficient contrast between the surface and the object. Some saw outlines of pictures on the walls of their homes, trees on their lots, doorways, windows, curbs, telephone posts, etc, never having seen these previously. One individual was able to drive an automobile in a vacant parking lot, stopping in front of walls and avoiding obstacles. Without belaboring the many additional similar observations, which rather astounded the involved patients, two other simple examples will be noted. One involved a young woman who was able to walk outside her house, identify her garbage can at the curb and bring it back into her yard. Another was able to get of his car, identify a front door light, and walk straight to the entrance of the house (a house, with which he had no familiarity), without any guidance. There seemed no doubt that these patients had artificial vision that was far superior to any previous attempts, certainly justifying the pioneering experiments of Brindley and his colleagues. However, the extreme optimism arising from the foregoing observations lasted only for a period of months, giving away to a series observations that clearly heralded deterioration in the quality of the achieved vision. These included very gradual elevations in the phosophene thresholds, a loss of the numbers of phosphenes, and an alteration in some of the qualities of the
3013
3014
182
Impaired vision: visual prosthesis
phosphenes. There are absolutely no clear cut reasons for this deterioration, which as of this date (December 2007) has resulted in the loss of any usefulness of any of the prostheses! Dr. Dobelle died in October of 2004. The authors of this chapter were not involved in following these patients, but nevertheless are now attempting to explore the potential causes of these losses. Unfortunately there has not been good documentation of many of the potential variables, which would be of importance in the understanding of the visual deterioration and loss. A thorough discussion of the possibilities for it is well beyond the scope of this paper. Quite briefly, the possibilities include: (1) some type of neuronal death as a result of over stimulation (which might have been the cause in a minority of patients); (2) toxic electrolytic\electrochemical (current) effects associated with the stimulating apparatus as a result of the alteration in the metals and\or their insulation (given the fact that some patients lost their phosphenes without having used their prostheses for any significant length of time, thus precluding a phenomenon of over stimulation); and (3) some type of malfunction of the hardware, leading to one of the foregoing.
5.
6.
7. 8.
9.
10.
Brightness modulation: Can be achieved theoretically from varying all stimulus parameters. Phosphene interactions: Both positive and negative interactions may occur at higher levels of stimulation and with pattern (multiple electrode) stimulation. Multiple phosphenes: Produced with suprathreshold stimulation in most instances [28]. Cortical ‘‘two-point discrimination’’: When stimulation of the cortical surface is applied to two sites as little as 2–3 mm apart, the two resulting phosphenes can be distinguished from one another. However, this may not apply with planar arrays because of nonlinear summation. Pattern presentation of phosphenes: Simple patterns and letters can be built up by stimulating a number of appropriately spaced electrodes. Intracortical stimulation: The threshold for the production of phosphenes by intracortical stimulation is much less than that of surface cortical stimulation [5].
What is Probable? Observations to Date 1.
What is Known 1.
2.
3.
4.
Threshold: Typical surface electrode zeroto-peak thresholds are in the range of 0.5–8 mA. Visuotopic map: In general, the visuotopic map predicted on the basis of Holmes’s observations exists in the blind. Size of phosphenes: The size of phosphenes is sufficiently small that enough densely packed phosphenes would allow satisfactory resolution in an eventual prosthesis. ‘‘Depth’’ perception: Phosphenes are coplanar [9,19].
2. 3.
4.
Phosphene color: Phosphenes are usually colored in sighted patients [19,52,53] and colorless in blind patients. Phosphene flicker: Phosphene flicker is less prominent in unsighted patients [45]. Influence of duration of blindness on phosphene size: It would appear that the longer the period of blindness, the larger the phosphenes and the higher their thresholds (Brindley’s second patient) [19]. Influence of age of onset of blindness: If blindness occurs early enough in life that the patient has no ‘‘memory’’ of visual imagery, then perhaps no phosphenes can be produced (Button’s and Putnam’s second case) [27].
Impaired vision: visual prosthesis
5.
Relationship of phosphene size to stimulation in visual pathways: The literature suggests that stimulation in pathways proximal to the cortex – e.g., in the optic nerve [42] or the optic (geniculocalcarine) radiations [1,17] – gives larger phosphenes.
What is Unknown (Unexplored)? 6.
7.
Effects of retinal stimulation: There has been no systematic study to determine the potential discreteness of retinal stimulation. Area of ‘‘perception’’ of phosphene: There is no knowledge as to where a phosphene is perceived; phosphenes can certainly be produced in hemianopic patients, but how much visual pathway must be intact for the perception of a phosphene is unknown.
Current Expectations of a Visual Prosthesis There are probably about a quarter of a million North Americans who are legally blind. Fewer than 20% of these individuals can read braille and fewer than 10% have achieved mobility. Eighty percent of visually impaired individuals have some residual vision, enhancement of which is perhaps the most appropriate way to help such individuals. Seventy-five percent of the visually impaired are over the age 65 and, in Britain, 68% of such individuals have at least one other disability. Sixty-five percent of newly blinded individuals are over the age of 65 years. Whereas achieving the ability to read was the most important goal of a visual prosthesis at one time, when radio was the only means of public communication available to blind individuals, the growth in mass communication with access to multichannel radio and television,
182
along with the use of magnetic tapes upon which most books, etc., have been transcribed, has drastically changed the priorities in the last two to three decades. Computers that recognize printed text (optical character recognition) and produce synthetic speech are increasing in quality and falling rapidly in price. In their broadest sense, they may be used even to convert other analog data such as temperature, pressure, etc., into ‘‘readable’’ material. ‘‘Speaking appliances’’ such as stoves are already available. Thus, reading is no longer the prime requirement for the blind, but rather the mobility, which can provide independence. The long cane has been the time-honored device for achieving mobility and still remains the most useful. The trained dog plays an extremely important role for a small segment (1–2%) of the blind community. The laser cane, which seemed to have such promise, has remained heavy, bulky, costly, and unwieldy; it was never really accepted by the blind and has provided most blind subjects with little or no improvement over the long cane. Devices on glasses that produce auditory signals distract patients, impairing the perception of environmental sounds, which may be extremely important in enhancing mobility. Infrared information (e.g., ‘‘talking signs’’) helps, but, unfortunately, such infrared systems are too expensive for widespread use, so that they are relegated to fixed environments. Rather ironically, it is the fixed environment (e.g., worksite, home, familiar streets, etc.) with which the blind become very quickly familiar. There is clearly no substitute for independent mobility and for esthetics to the unsighted individual, e.g., the recognition of facial expression, bodily contours, etc. The discussion in this communication has been centered on the development of a visual prosthesis through electrocortical stimulation. This should not be taken as denying the very important and ingenious attempts at transcutaneous information transfer, both mechanically [6,4,18] and electrically. [26] Most of these ‘‘noncortical’’ devices
3015
3016
182
Impaired vision: visual prosthesis
were created for the purpose of reading, but when considered for the production of mobility and esthetic ‘‘sight,’’ the skin has insufficient frequency response to convey complex visual material [35].
to identify the thresholds above which multiple phosphenes appear, so that they could be prevented by the control system in the memory of the individualized integrated chip within the eventual electronics package.
The Ideal Prosthesis The ideal visual prosthesis is one that grants the unsighted individual independent mobility, the appreciation of form, and halftone resolution and that interferes as little as possible with any other intact sensory pathway. The basis of such a prosthesis is the ability to recognize halftones, e.g., edges, shadows, facial expression, dimness of object, etc., by providing information regarding brightness. As already shown, brightness modulation can be achieved theoretically by variation in a number of stimulus parameters, perhaps by some better than by others – options that have been only very briefly explored [24]. The most probable reason for this neglect is the enormous amount of psychophysical experimentation required to provide good discrimination of levels of gray. However, unless a break-through occurs in some other technology, exploring this issue seems unavoidable. Finally, ideally some form of telemetric stimulation, thus avoiding percutaneous (direct) penetration with a foreign body, and its potential complications.
The Avoidance\Abolition of Phosphene Interactions Interaction of phosphenes would similarly impair resolution. Positive interactions, i.e., the appearance of greater numbers of phosphenes than the number of electrodes stimulated, would seem more likely to impair resolution than would negative (i.e., the loss of one or more phosphenes with multiple electrode stimulation) interactions. Although this is a more complex problem because of the time that would be required to identify such interactions with multiple electrode stimulation, common, gross, or frequent interactions could be identified and precluded by registration in the computer memory.
The Achievement of Depth in the Phosphene Field
Problems That Must Be Overcome in Order to Achieve the Ideal Electrocortical Visual Prosthesis
Phosphenes are coplanar in both sighted and unsighted individuals; hence there would have to be some mechanism instituted for the appreciation of depth (e.g., possibly zoom apparatus on camera). In fact, most depth cues are contextual (texture, foreshortening, parallax) or oculomotor (accommodation, convergence) and not related to binocular steropsis from image disparity.
Avoidance of Multiple Phosphenes from Single Electrode Stimulation
The Achievement of Access to the Buried Visual Cortex
Production of multiple phosphenes with the stimulation of a single electrode would markedly impair resolution. Experimentation would have
The discontinuities seen in phosphene mapping in all the patients of Brindley and Dobelle and their colleagues are predictable to varying
Impaired vision: visual prosthesis
degrees on the basis of Holmes’s visuotopic map. In order to fill out the phosphene field completely (which would be necessary for optimizing resolution) one would have to stimulate the buried cortex (i.e., sulci), for the majority (67%) of striate cortex is buried [50]. Because of the very low thresholds for phosphene production from intracortical stimulation and the fact that interactions may be less likely from intracortical stimulation, [5,32] it would appear that newer electrode technology should be aimed at the fabrication of very fine intracortical stimulating electrodes with a configuration that would allow access not only to the surface but also to the buried occipital cortex [56].
Telemetric Stimulation Brindley used radiotelemetry for stimulation of the visual cortex. Dobelle and colleagues used a biocompatible, pyrolytic carbon pedestal [20] as a hard-wired device for stimulation in their experiments. Such a hard-wired connection would, of course, preclude the need for complex transmitters to transfer the enormous amount of information in real time required for the ideal prosthesis through the scalp. Experience over the last 20 years with such percutaneous pedestals, however, has provided clear evidence that even the use of biocompatible materials does not preclude the occurrence of repeated secondary infections around the pedestal base. In one patient, the electrode had to be removed because of the intracranial spread of such an infection. This has enhanced the impetus to take advantage of newer, more advanced technology for telemetric. . . transfer of information. Using surface stimulation the high requirements for current may make it difficult, if not impossible, to transmit all power through RF. The use of intracortical microstimulation (ICMS) [5,32], which is being studied now in a number of centers, has disclosed the significantly reduced current requirements
182
for obtaining satisfactory elicitation of phosphenes. Thus, a full array utilizing ICMS would greatly enhance the possibility of telemetric information transfer in a visual prosthesis.
The Achievement of Half-Tone Resolution Brightness modulation is an absolute necessity for the achievement of halftone phosphene ‘‘imagery.’’ Adequate brightness modulation would probably require something of the order of four to eight levels of gray. This would seem to be most readily achieved through modulation of each individual phosphene. As already noted, attempts to achieve this will be hampered by the appearance of multiple phosphenes and interactions with higher amplitudes of stimulation. It is the authors’ view that it may not be possible to achieve such modulation through manipulation of stimulus parameters per se. An alternative method of brightness modulation is that which is so commonly utilized in newspaper print and low-grain photographs. That is to say, gray levels are altered by changing the density of either black or white units (pixels). If a sufficient number of phosphenes existed (e.g., from surface and intracortical stimulation) and they were sufficiently small and densely packed, brightness modulation could be achieved by simply modulating the ‘‘on’’ or ‘‘off ’’ status of the individual phosphenes in question. In this case, then, the halftone pictures would be the result of the density of the ‘‘on’’ phosphenes rather than of the alteration in brightness of each individual phosphene (but at a high cost in terms of numbers of electrodes and channels needed!).
Summary Thus, while meaningful observations concerning phosphene production have been collected over
3017
3018
182
Impaired vision: visual prosthesis
three-quarters of a century and meaningful experiments undertaken over the latter part of the century, nevertheless, the feasibility of a visual prosthesis from electrocortical stimulation must still remain an open question. This may seem surprising, given the fact that it at first appeared feasible to create such a prosthetic device within 10 or 15 years after Brindley’s initial experiments [40]. The very large multidisciplinary effort required to determine such feasibility is expensive. Building a functional clinical system (if it proves feasible) would be a very large undertaking for the medical device industry. These issues particularly explain why progress has been somewhat slow. It is hoped that restructuring of the corporate sector and education of the governments of the western industrialized nations with respect to the importance of research for future health care will result in acceleration of research for the disabled. The question is always asked, as it should be, whether or not the expense is justified on the basis of the final achievement. Certainly, when one considers that the care of the blind in the United States alone costs well in excess of $1 billion per year, in 1960s dollars [34], then it certainly is not difficult to justify the expenditure of enough money (in the range of millions of dollars) to determine the feasibility of a visual prosthesis using electrocortical stimulation. If, indeed, such a prosthetic device is shown to be feasible, it will be the reduction in cost of the individualization, or customization, of each prosthesis which will be the most important determinant of the eventual price of such a device. It is the latter that will ultimately determine whether or not such a device is economically reasonable. To make predications on how much biological and electronic experimentation might be required and on what the cost might be when technology is changing so rapidly, would be little more than idle speculation.
References 1. Adams JE, Rutkin BB. Visual responses to subcortical stimulation in the visual and limbic system. In: Sterling TD, Bering EA, Jr, Pollack SV, Vaughan HJ, Jr, editors. Visual prosthesis: the interdisciplinary dialogue. New York: Academic Press; 1971. p. 49-55. 2. Agnew WF, Yuen TGH, Pudenz RH, Bullara LA. Electrical stimulation of the brain: IV. Ultrastructural studies. Surg Neurol 1975;4:438-8. 3. Agnew WF, Yuen TGH, McCreery DB. Morphological changes after prolonged electrical stimulation of the cat’s cortex at defined charged densities. Exp Neurol 1983;79: 397-411. 4. Bach-y-Rita P. Visual information through the skin – A tactile vision substitution system. Trans Am Acad Ophthalmol Otolaryngol 1974;78:OP729-40. 5. Bak M, Girvin JP, Hambrecht FT, et al. Visual sensations produced by intracortical microstimulation of the human occipital cortex. Med Biol Eng Comput 1990; 28:257-9. 6. Bliss JC. Now the totally blind can read the materials of the sighted. Trans Am Acad Ophthalmol Otolaryngol 1978;78:OP723-28. 7. Brindley GS. The number of information channels needed for efficient reading. J Physiol 1964;177:44. 8. Brindley GS, Lewin WS. The visual sensations produced by electrical stimulation of the medial occipital cortex. J Physiol 1968;194:54-5. 9. Brindley GS, Lewin WS. The sensations produced by electrical stimulation of the visual cortex. J Physiol 1968;196:479-93. 10. Brindley GS. Sensations produced by electrical stimulation of the occipital poles of the cerebral hemispheres and their use in constructing visual prostheses. Ann R Coll Surg 1970;47:106-8. 11. Brindley GS. Sensory effects of electrical stimulation of the visual and paravisual cortex in man. In: Jung R, editor. Handbook of sensory physiology. vol II\3B, , Berlin: Springer; 1973. p. 583-94. 12. Brindley GS, Rushton DN. Implanted stimulators of the visual cortex as visual prosthetic devices. Trans Am Acad Ophthalmol Otolaryng 1974;78:OP741-5. 13. Brindley GS. Visual prostheses based on electrical stimulation of the human visual cortex. In 17th Neural Prosthesis Workshop, NIMCDS, NTH, 1986. 14. Brindley GS, Donaldson PEK, Falconer MA, Rushton DN. The extent of the region of occipital cortex that when stimulated gives phosphenes fixed in the visual field. J Physiol 1972;225:57-8. 15. Bullara LA, Agnew WF, Yuen TGH, et al. Evaluation of electrode array material for neural prostheses. Neurosurgery 1979;5:681-6. 16. Button J, Putnam T. Visual responses to cortical stimulation in the blind. J lowa Med Soc 1962;52:17-21.
Impaired vision: visual prosthesis
17. Chapanis N, Uematsu S, Konigsmark B, Walker AE. Central phosphenes in man: A report of three cases. Neuropsychologia 1973;11:1-19. 18. Craig JC. Vibrotactile pattern recognition and masking. In: Gordon G (ed). Active Touch—The Mechanism of Recognition of objects by Manipulation: A Multidisciplinary Approach. Oxford, England: Pergamon Press; 1978. p. 229-42. 19. Dobelle WH, Mladejovsky M. Phosphenes produced by electrical stimulation of human occipital cortex, and their application to the development of prosthesis for the blind. J Physiol 1974;243:553-76. 20. Dobelle WH, Mladejovsky MG, Evans JR, et al. “Braille” reading by a blind volunteer by visual cortex stimulation. Nature 1976;259:111-2. 21. Dobelle WH, Mladejovsky MG, Girvin JP. Artificial vision for the blind: Electrical stimulation of visual cortex offers hope for a functional prosthesis. Science 1974;183: 440-4. 22. Dobelle WH, Mladejovsky MG, Roberts TS, Girvin JP. A percutaneous connector for chronic electrodes in man (abstr). Amer Soc Art Int Organs, 1976. 23. Dobelle WH, Quest DO, Antunes JL, et al. Artificial vision for the blind by electrical stimulation of the visual cortex. Neurosurgery 1979;5:521-7. 24. Evans JR, Gordon J, Abramov l, et al. Brightness of phosphenes elicited by electrical stimulation of human visual cortex. Sens Proc 1979;3:82-94. 25. Foester O. Beitra¨ge zur Pathophysiologie der Sehbahn und der Sehspha¨re. J Psychol Neurol Lpz 1929;39:463-85. 26. Geldard FA. Cutaneous Communication Systems and Devices. Austin, TX: The Psychonomic Society, 1974. 27. Givin JP. Personal observation, 1976. 28. Girvin JP, Dobelle WH. Changes in phosphene multiples with increased strengths of electrical stimulation of human visual cortex. In preparation. 29. Girvin JP, Evans JR, Dobelle WH, et al. Electrical stimulation of human visual cortex: The effect of stimulus parameters on phosphene threshold. Sens Proc 1979; 3:66-81. 30. Hambrecht FT. Visual prostheses: Theoretical objectives, present status and future possibilities. In: Fields WS, Leavitt LA (editors). Neural Organization and Its Relevance to Prosthetics. New York: Inter-continental Medical Book; 1973. p. 281-91. 31. Hambrecht FT. Neuroprostheses. Annu Rev Biophys Bioeng 1979;8:239-67. 32. Hambrecht FT, Abc M, Kuftac V, et al. Feasibility of a visual prosthesis for the blind utilizing intracortical microstimulation. Proceedings of the 4th Vienna International Workshop on Functional Electrostimulation, Baden, Austria, September 24-27, 1992. 33. Holmes G. Disturbances of vision by cerebral lesions. Br J Ophthalmol 1918;2:253-384. 34. Hoover RE. Introduction to symposium: prosthetic aids for the blind. Trans Am Acad Ophthalmol Otolaryngol 1974;78:OP711-2.
182
35. Hoover RE, Freiberger H, Bliss JC, et al. Prosthetic aids for the blind. Trans Am Acad Ophthalmol Otolaryngol 1974;78:711-46. 36. Krause F. Die Sehbahnen in chirurgischer Beziehung und die faradische Reizung des Sehzentrums. Klin Wochenschr 1924;3:1260-65. 37. Krause F, Schum H. Die epileptischen Erkrankungen. In: kattner H (ed). Neue Deutsche Chirugie. vol 49, Stuttgart: Enke, 1931;482-6. 38. Krieg WJS. Functional Neuroanatomy, 2nd ed. Evanston, IL: Blakiston, 1953. p. 207-208. 39. Lewin W. Towards a visual prosthesis. Clin Neurosurg 1971;18:155-65. 40. Loeb GE. Whatever happened to the visual prosthesis? Andrade JD, et al. (ed). Artificial Organs. New York: VCH Publishers; 1987. p. 457-46. 41. Loeb GE. Neural prosthetic interfaces with the nervous system. Trends Neurosci 1989;12:195-201. 42. Nakagawa J. Experimental study on visual sensation by electrical stimulation of the optic nerve in man. Br J Ophthalmol 1962;46:592-96. 43. Penfield W, Jasper H. Epilepsy and Functional Anatomy of the Human Brain. London: Churchill, 1954. p. 116-26, 404-6. 44. Penfield, W, Rasmussen T. The Cerebral cortex of Man. New York: Macmillan; 1952. p. 135-47, 165-6. 45. Pollen DA. Some perceptual effects of electrical stimulation of the visual cortex in man. In: Tower DB editors. The Nervous System. New York: Raven Press; 1975. p. 519-28. 46. Pudenz RH, Bullara LA, Dru V, Talalla A. Electrostimulation of the brain: II. Effects on the bolld-brain barrier. Surg Neurol 1975;4:265-70. 47. Pudenz RH, Bullara LA, Jacques S, Hambrecht FT. Electrical stimulation of the brain: III. The neural damage model. Surg Neurol 1975;4:389-400. 48. Rushton DN, Brindley GS. Short- and long-term stability of cortical electrical phosphenes. In: Rose FC (ed). Physiological Aspects of Clinical Neurology. New York: Blackwell; 1977. p. 123-153. 49. Shaw JD. Method and mean for aiding the blind. United States Patent Number 2,721,316, 1955. 50. Stenass SS, Eddington DK, Dobelle WH. The topography and variability of the primary visual cortex in man. J Neurosurg 1974;40:747-55. 51. Sterling TD, Bering EA, Pollack SV, Vaughan HG Jr. Visual Prosthesis: The Interdisciplinary Dialogue. New York: Academic Press; 1971. 52. Talalla A, Bullara L, Pudenz R. Electrical stimulation of the human visual cortex-A Preliminary report. Can J Neurol Sci 1974;1:236-8. 53. Uematsu S, Chapanis N, Gucer G, et al. Electrical stimulation of the cerebral visual system in man. Confin Neurol 1974; 36:113-24. 54. van den Bosch FJG. Read before the Sigma Delta Epsilon, Kappa chapter, New York, New York Academy of Science; March 20, 1959.
3019
3020
182
Impaired vision: visual prosthesis
55. Watson W. An account of Mr. Benjamin Franklin’s treatis, lately published, entitled, “Experiments and Observations on Electricity,” made at Philadelphia in America. Phil Trends R Soc Lond 1951;47:201-11. 56. Wise KD, Najafi K. Microfabrication techniques for integrated sensors and Microsystems. Science 1991; 254:1335-42.
57. Yuen TGH, Agnew WF, Bullara LA. Histological evaluation of neural damage from electrical stimulation: Considerations for the selection of parameters for clinical application. Neurosurgery 1981;9:292.
187 Microdialysis M. J. Smith . M. G. Kaplitt
Background
Principles of Microdialysis
The recent advent of in-vivo electrical and chemical monitoring in the human brain provides opportunities to study markers for individual disease processes and responses to therapy in functional disorders. Cerebral microdialysis is one such method for monitoring the local neurochemical environment of a target brain region of interest. The first description of cerebral microdialysis in the literature is by Bito et al. in 1966. He implanted a dialysis membrane filled with dextran solution into canine cortex and found that amino acid levels in brain extracellular fluid were greater than in cerebrospinal fluid [1]. Capitalizing on and refining this method, Ungerstedt et al. implanted a small loop of dialysis tubing in rat brain and continuously perfused it with a fluid medium. High performance liquid chromatography with electrochemical detection was used to identify and quantify amino acid concentrations in the dialysate [2]. Interestingly, when cerebral microdialysis was eventually translated to use in human patients, one of the first applications was in functional neurosurgery for Parkinson’s disease patients undergoing thalamotomy [3]. Additionally, commercially available probes, dialysis fluids, collectors and analyzers have been created, improving access to the technique and making the use of this potentially complicated bedside monitoring more convenient and feasible. In essence, this methodology provides a window into the dynamic microenvironment of cortical and deep brain structures, as any extracellular molecule smaller than the dialysis membrane can be selectively sampled.
Definition
#
Springer-Verlag Berlin/Heidelberg 2009
Cerebral microdialysis is a tool to identify and measure concentrations of extracellular neurochemicals within cerebral structures, locally and in vivo. A double lumen probe containing an inlet and an outlet port (generally separate small tubes) are surrounded by a semi-permeable membrane which is sealed above and below a point at which the two tubes are inserted into the membrane lumen. This dialysis probe is then placed through a burr hole to a preset depth within the brain parenchyma. A perfusion solution with an electrochemical concentration similar to the local environment is propelled through the inlet tube and into the semi-permeable membrane (> Table 187-1). Molecules in the extracellular space, smaller than the permeable membrane, diffuse across it into the perfusion solution. The solution then flows through the outlet port and is collected in vials for further processing and analysis (> Figures 187-1 and > 187-2). The most frequently measured substances and their published reference values are listed in > Table 187-2. The novelty of microdialysis compared to prior open ended ‘‘push-pull’’ perfusion techniques is the addition of the semi-permeable membrane. This feature maintains the integrity of the adjacent interstitial tissue and prevents inaccuracy due to perturbation of the local environment from fluid leakage [7,18]. To date, cerebral microdialysis is the most commonly used technique for sampling neurochemicals in deep brain structures in vivo for both pre-clinical and clinical research purposes.
3118
187
Microdialysis
. Table 187-1 Components of microdialysis perfusion solutions Chemical
Artificial CSF (mM/L)a
Artificial CSF (mM/L)b
Ringer’s solution (mM/L)b
Normal saline (mM/L)c
NaCl KCl CaCl2 MgCl2 Na2HPO4 NaH2PO4 Glucose
Distilled water 140 2 1.2–2.4 1 1.2 0.27 7.2 pH 7.4
Distilled water 145 3 1.4 1 2 pH 7.4
Ringer’s solution 147 4 2.3 pH 6–6.3
Distilled water 154 -
a
Stamford et al., 1992 Amara et al., 1998 c Bullock et al., 1995 b
. Figure 187-1
Techniques – Methods of Analysis After collection, the dialysate vials may be stored at 20 to 70 C or immediately analyzed using a variety of techniques [4,19]. There are commercially available assays for the most common markers including glucose, lactate, pyruvate, and glycerol. Biogenic monoamine analysis requires more complex techniques and equipment. High performance liquid chromatography with electrochemical detection (HPLC EC) provides a high sensitivity and specificity method for measuring biogenic amines, in particular noradrenaline
(NA), dopamine (DA), and serotonin (5-HT) [7,18,20]. For amino acids including glutamate, aspartate, GABA, serine, and taurine, HPLC with fluorometric analysis may be used according to published techniques [20].
Technical Considerations and Limitations Microdialysis is a particularly attractive tool for in vivo monitoring of extracellular molecules in the brain because a substance of any type which
Microdialysis
187
. Figure 187-2
is smaller than the pore size of the semi-permeable membrane can be measured as long as some detection method is available. As with any technique, however, the unique limitations and issues which may influence either the use or interpretation of data generated from microdialysis must be carefully considered before proceeding with a particular application. First, the invasiveness of the technique presents risks for hemorrhage, neuronal damage, or infection, which is why this method has generally been applied in humans only when invasion of the brain is already being performed toward a therapeutic end. There have been few reports of significant complications from microdialysis in the literature, but there remains to be a systematic review of the overall safety profile of this still experimental technique. Therefore, patients and/or families must be made aware and consented for the possibility that microdialysis could slightly increase these risks beyond those already present for the therapeutic intervention. The accuracy of microdialysis can also be affected by multiple factors. For example, the absolute concentration of a substance is difficult to determine because it is dependent on the flow
rate. If the flow rate is too rapid, a steady state concentration of a chemical between the dialysate and extracellular fluid will not be achieved, leading to an underestimation of concentration. Some researchers have increased the probe length in order to improve this limitation; however, a larger probe size leads to more interstitial tissue damage and decreased spatial resolution [18,21]. This is of particular importance in functional neurosurgical applications as compared with trauma, since deep brain targets are often relatively small, so a longer length of dialysis membrane may not restrict sampling to only the desired target. A second factor affecting accuracy is the local tissue damage that occurs at the time of insertion. Studies have shown extremely high initial levels of monoamines immediately post-implantation, with subsequent normalization of values. These authors postulated that neurotransmitters may be released en masse from damaged cells and values do not accurately reflect the normal extracellular environment until they plateau at lower levels, reportedly between 2–24 h [4]. Third, a local fibrotic reaction has been shown to develop around the probe tip during chronic microdialysis experiments. This fibrosis correlates
3119
3120
187
Microdialysis
. Table 187-2 Published microdialysis markers in normal and diseased states Marker
Normal value per liter
Observed trend
Disease state or process
Diagnosis
GABA
35–52 nM
↑
Seizure
Epilepsy, PD
Glutamate Glutamate
240–480 nM 16 16 mM
↑
Hypoxia/Ischemia, Excitiotoxicity, Neuronal death
PD TBI, SAH, Stroke, Tumor, Necrosis, Seizure
DA
14 3.3 mM -
↑
Epilepsy, PD
NO
-
↓
Glucose
1.7 0.9 mM
↓
Working memory, L-DOPA treatment Vasospasm, Ischemia Hypoxia/Ischemia, Hypoglycemia, Cerebral hypermetabolism
Glycerol
2.12 0.15mM 82 44 mM
↑
Lactate
81 12.4 mM 29 0.9 mM
L/Pa
3.06 0.32mM 166 47 mM 151 12 mM 23 4
L/G
19.3 1.7 1.62 0.18
Pyruvate
Location of probe Hippocampus Basal Ganglia Basal Ganglia Cortex
References (4–6) (4) (7–11)
(3,12)
SAH
Amygdala Thalmamus Cortex
TBI, SAH
Cortex
(7,9,10)
Hypoxia/Ischemia, Cell membrane breakdown
TBI, Seizure
Cortex
(5,7,9,10,14)
↑
Hypoxia/Ischemia, Cellular redox states, Decreased glucose supply, Mitochondrial dysfunction
TBI, SAH, Epilsepsy, Intracranial hypertension
Cortex
(7,9,10,14–16)
-
-
-
-
(9,10)
↑
Hypoxia/Ischemia, Cellular redox states, Decreased glucose supply, Mitochondrial dysfunction
TBI, SAH, Epilsepsy, Intracranial hypertension
Cortex
(7,9,10,14–16)
↑
Cerebral hypermetabolism
Epilepsy
-
(9,17)
(13)
GABA: Gamma-aminobutyric acid; PD: Parkinson’s Disease; TBI: Traumatic brain injury; SAH: Subarachnoid hemorrhage; DA: Dopamine; L-DOPA: Levodopa; NO: Nitric oxide; L/P: Lactate to pyruvate ratio; L/G: Lactate to glucose ratio a most reliable marker of ischemia; – not available
with decreased basal concentrations of neurochemicals sampled after approximately 5–7 days, thus bringing the accuracy of values obtained after extended periods into question [18]. Finally,
placement of the catheter may not be exact and should be confirmed if it is to be concluded that the data reflects the physiology of a particular target structure. Early methods of microdialysis
Microdialysis
did not provide a method for confirming the accuracy of probe placement; however, recent studies and commercially available probes contain a radio opaque marker at the tip for confirmation of placement by CTscan or X-ray [4]. Again this is particularly important when microdialysis is used for functional neurosurgical procedures, such as Parkinson’s disease and epilepsy, since conclusions based upon the data generated are only valid if the samples are obtained from the brain region being targeted. While the local nature of the microdialysis is usually desirable to specifically analyze a target brain region, in other circumstances this can be a limitation. Unlike jugular venous oxygen sampling and positron emission tomography which are hemispheric or global investigative techniques, the information obtained from microdialysis represents the environment of approximately 1 cm around the probe tip. This information is usually not representative of the entire brain and must be used in conjunction with other monitoring modalities in order to safely direct patient care. This is more important when microdialysis data will be used to influence therapy, such as in trauma or ischemia situations, since focal data is used to direct global therapy. When microdialysis is used to better understand the physiology of a brain target in functional diseases, however, this focal measurement is far more desirable than a less relevant global or hemispheric analysis. Finally, one must always remember what neuronal functions are in fact being measured by microdialysis. The neurochemicals sampled are a product not only of synaptic release, but also of local metabolism, capillary delivery and neuronal release and uptake from groups of cells [5,19]. The diameter of most microdialysis probes is fairly large from a cellular standpoint (usually several hundred microns), and this is difficult to overcome since the membrane diameter must be larger than the combined diameters of the inlet and outlet tubes which are inserted into the membrane. This is far too large for
187
sampling synaptic activity in isolation (which would require a probe in the single digit micron range), so all microdialysis probes used to date measure tissue levels of neurochemicals rather than synaptic levels. This is important to understand when interpreting data, since a negative result with microdialysis does not necessarily mean that there is, in fact, no change in synaptic neurotransmitter levels. Conversely, the implications of positive results must be considered within the context of the limitations of the probe as a measure of local tissue rather than synaptic levels. Nonetheless, there are numerous applications of microdialysis which provide unprecedented opportunities to understand living human brain physiology even when these issues are taken into account, and several of these applications are reviewed below.
Uses in Human Patients Despite its limitations, microdialysis has been successfully used as an adjunct to traditional research and therapy in many human patients. Recent years have seen an increase in the use of microdialysis as a clinical tool in a variety of neurosurgical disease processes, particularly traumatic brain injury and subarachnoid hemorrhage, where an opportunity exists to detect and prevent secondary cerebral injury [5,7]. Other neurological diseases with focal origins of pathology such as movement disorders and epilepsy have been studied using microdialysis as a research tool. Electrical recording in the human brain has provided a wealth of data regarding neuronal firing patterns in functional disorders, and functional imaging provides a non-invasive window into certain physiological changes in the living human brain. However, microdialysis is a minimally invasive method with a proven safety record to directly analyze nearly any substance within brain tissue with much higher resolution than functional imaging. This provides
3121
3122
187
Microdialysis
unprecedented opportunities to sample local neurochemical environments of deep brain structures which can then be correlated with other data in order to better understand the physiological basis for human neurological disease. An overview of the most recent utilization of microdialysis reveals its widespread applicability.
Traumatic Brain Injury Traumatic brain injury causes immediate cell death, in part from mechanical forces that result in diffuse axonal injury and vascular damage [8,22,23]. This primary injury also initiates a subsequent heterogeneous cascade, including disruption of ion channel pumps, leading to extreme shifts in intra and extracellular chemicals, massive increases in excitatory amino acids, and the generation of oxygen free radicals. The simultaneous activation of these pathways significantly increases cellular metabolic demands and ultimately result in neuronal injury and death [8]. While primary cell death after brain trauma may not be a reasonable target for intervention, secondary cerebral injury may benefit from early detection and neuroprotective strategies. Cerebral microdialysis is one tool that can provide early detection of factors leading to delayed neurologic insults. Changes in several neurochemical markers are indicative of neuronal risk (> Table 187-2). For example, increased lactate, lactate to pyruvate ratio, lactate to glucose ratio, and hypoxanthine levels have been associated with increased metabolic demand or decreased nutrient supply in brain injured patients [24,25]. Furthermore, a corresponding decline in pH may warn of the shift from efficient aerobic to inefficient anaerobic metabolism [26]. These findings have been correlated with a decrease in local brain tissue oxygen partial pressures, ischemia, infarction and ultimately poor outcome [5,25,27–30]. As part of a multimodality monitoring system, microdialysis may detect these changes in
advance of permanent cerebral injury, opening a treatment window to prevent these secondary insults and improve patient outcome.
Subarachnoid Hemorrhage A similar paradigm exists for the detection, prevention and reversal of secondary cerebral insults for the treatment of subarachnoid hemorrhage. Vasospasm and its resultant cerebral ischemia is a major cause of morbidity and mortality after aneurysmal subarachnoid hemorrhage. Several studies have examined the patterns of neurochemical changes using cerebral microdialysis in this patient population in order to better understand the mechanisms of vasospasm and define markers of early detection. Increased levels of lactate, lactate to pyruvate ratio, glutamate and glycerol, as well as decreased glucose concentrations have all been observed in patients with clinical vasospasm [31]. Furthermore, microdialysis levels of nitric oxide, a proposed etiologic agent of vasospasm, have been shown to be significantly lower in patients during severe clinical vasospasm. These depressed levels also correlated with decreased brain tissue oxygen partial pressure values [13]. Most recently, cerebral microdialysis in conjunction with proteome-wide screening has been used to define two protein markers, heat-shock cognate 71kDa protein (HSP7C) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH), as possible early markers of vasospasm. Levels in patients experiencing vasospasm were significantly different than in patients without vasospasm. Furthermore, this difference was detected approximately 4 days prior to the onset of clinical spasm [32].
Ischemic Stroke Another avenue for clinical investigation using microdialysis has been ischemic stroke. Although
Microdialysis
only a few studies have been conducted on this topic, they have focused on detecting elevations in excitatory amino acids. Bullock et al. demonstrated massively increased and sustained levels of both aspartate and glutamate in the infarcted tissue of a patient with a large hemispheric stroke. They postulated that the delayed release of excitatory amino acids may contribute to further infarction within the penumbra leading to malignant edema [33]. A subsequent study found that significant increases in microdialysis levels of glutamate and lactate to pyruvate ratio were associated with malignant edema, thus acting as potential indicators for decompressive hemicraniectomy [34]. Future investigations may consider focusing on the association between neuronal microdialysis markers after reperfusion using chemical thrombolytics or mechanical thrombectomy, although the safety of maintaining an implanted intraparenchymal probe during active systemic thrombolysis remains unclear and must seriously be considered before combining microdiaylsis with such therapies. Additionally, the relevance of any of these results to therapeutic outcome remains unclear in the absence of correlation with long-term outcome measures.
Epilepsy An imbalance of neurochemical factors has long been proposed to underlie the onset of seizures and neuronal damage associated with epilepsy. A combination of excess production or decreased reuptake of excitatory neurotransmitters such as glutamate or conversely, decreased production or increased reuptake of inhibitory neurotransmitters such as g-aminobutyric acid (GABA) may contribute to seizure foci [35]. During and Spencer were the first authors to document the use of microdialysis to study human epilepsy patients, addressing the hypothesis that increases in extracellular glutamate may trigger spontaneous
187
seizures [35]. In patients with complex partial epilepsy refractory to pharmacologic treatment undergoing planned depth electrode placement, they combined the depth electrode with a microdialysis probe. This combined probe was named a dialytrode [35]. The microdialysis probe was attached to a polyurethane/silastic flexible depth electrode with nichrome contacts. A sterilized, pyrogen-free, artificial extracellular fluid was perfused through the dialysis probe at a rate of 2.5 mL/min. Given the small dead space of the outlet tubing, a temporal delay of only ten minutes occurred between sampling and collection. After confirmation of placement of the combined electrodes by MRI, patients were transferred to the neurosurgical ICU where continuous video EEG monitoring and collection of dialysate every 30 min was performed using an automated sampler (CMA200, Carnegie Medicin, Stockholm, Sweden). Frequency of sampling was increased during a seizure. Analysis of amino acids in the dialysate was performed using HPLC (Bioanalytical Systems Inc., West Lafayette, In, USA) [35]. Of the six patients studied, results demonstrated that extracellular glutamate significantly increased prior to and during a seizure. Moreover, concentration of extracellular GABA increased during a seizure, most notably in the non-epileptogenic hippocampus. In addition to this study, a similar experimental setup was used by these authors to evaluate the extracellular levels of lactate within the hippocampus of partial complex epilepsy patients undergoing bilateral depth electrode placement. An increase in lactate concentration of 91 32% for 60–90 min following a seizure was observed. Of note, in patients without secondary generalization, this finding lateralized to the side of the seizure. Furthermore, interictal spikes were also associated with a significant increase in extracellular lactate levels. As a result of these findings, the authors postulated that seizures and interictal excitatory events increase the local, non-oxidative glucose metabolism of neurons. In addition, the increase
3123
3124
187
Microdialysis
in local lactate levels directly decreases extracellular pH, a condition known to suppress seizures. This phenomenon may be responsible for the arrest of ongoing seizures and the postictal refractory period [15]. This group also utilized cerebral microdialysis to evaluate local, extracellular levels of antiepiletic drugs, namely phenytoin and carbamezapime. After systemic administration of these medications, simultaneous extracellular fluid (ECF) levels via cerebral microdialysis and unbound serum levels were obtained and analyzed. The ratio of ECF to serum concentrations were between 0.84 and 0.87 suggesting a significant correlation between local cerebral and systemic levels. Given the indirect mechanisms of newer antiepileptic drugs, the authors discuss the potential of cerebral microdialysis for studying their complex, local effects [36–38]. An extension of the combined microdialysis and depth electrode, containing single-unit recording microelectrodes, was created by Fried et al. By placing probes trans-occipitally into the hippocampus, amygdala, entorhinal cortex, posterior parahippocampal gyrus, orbitofrontal cortex, and cingulate cortex they gathered samples from 86 probes in 42 patients. Sampling occurred routinely at 30 min intervals, or 5 min intervals during events, over a 6 day period. The proposed temporal delay given the length of the outflow tubing was reported to be 20 min. Overall, the authors found an increase in the excitatory amino acids glutamate and aspartate as well as the inhibitory amino acids tuarine and GABA in the ipsilateral amygdala during seizure events [20].
Learning and Memory Pilot studies conducted by Fried et al. in seizure patients with combined depth electrode, single-unit recording microelectrode and microdialysis probes noted changes in dopamine levels during cognitive tasks [20]. More recently,
however, this group has used this technique to perform detailed experiments assessing the role of dopamine in specific learning and memory tasks. In ten patients with 16 probes placed in the central and basolateral nuclei of the amygdala, they analyzed samples of dialysate at 5 min intervals before, during and after either working memory, reading or word-paired associates learning tasks. As predicted, there was a significant increase in extracellular dopamine concentration throughout the presentation of various tasks. Interestingly, the magnitude of the change was dependent on task order, where the greatest increase above baseline was achieved when a memory task followed a reading task. The authors concluded that the human mesolibmic dopaminergic system has sustained activation during the performance of cognitive tasks. Moreover, as a higher level of dopamine was observed during working memory tasks when presented second, there is a relationship between amygdala dopamine release and novelty of stimulus [12]. This study highlights the potential power of microdialysis combined with other monitoring modalities and functional neurological assays to provide insights into human brain functioning which can be difficult to extrapolate from preclinical animal experiments.
Movement Disorders Another field that has capitalized on the use of surgically placed deep brain probes to study the local neurochemical environment with microdialysis is functional neurosurgery for movement disorders, in particular Parkinson’s disease (PD). Stefani et al. published studies describing intraoperative microdialysis during bilateral deep brain stimulation (DBS) in six patients undergoing surgery for refractory PD. In the acute period immediately before, during and one hour after placing DBS in the subthalamic nucleus (STN), they measured GABA, cyclic
Microdialysis
guanosine 30 ,50 -monophosphaste (cGMP) and glutamate. After the initiation of DBS, significant increases in cGMP of 200 and 481% above baseline were noted in the putamen and internal globus pallidus (GPi), respectively, while a decrease in GABA concentration of 25% below baseline was noted in the anteroventral thalamus (VA). Additionally, they found a significant increase in cGMP in the substantia nigra reticulata that coincided with increased neuronal firing during electrophysiological recordings. They postulated that these findings represent mechanisms that differentially affect fibers crossing the STN area. The STN-GPi pathway demonstrates activation while the GPi-VA pathway demonstrates inhibition, both contributing to a restoration of physiologic putamen activity [39–43]. Based upon the earlier epilepsy work described above, we developed an in vivo microdialysis method for continuous chronic sampling of local neurotransmitter concentrations in the subthalamic nucleus (STN) and substantia nigra reticulate (SNr) of patients for several days following implantation of STN deep brain stimulation (DBS) electrodes [4]. Six patients undergoing bilateral DBS for medically refractory Parkinson’s Disease also received unilateral implantation of a microdialysis probe, inserted through a twist-drill hole adjacent to the burr hole for the permanent DBS electrode. In four patients, the probe was placed in the STN (1 mm away from the DBS electrode) and in two the probe was placed in the SNr at a depth of 3–4 mm ventral and 1 mm lateral to the DBS probe. Artificial CSF propelled by a portable CMA 107 pump (CMA Microdialysis, Stockholm, Sweden) was used to perfuse the membrane. Sampling occurred daily up to postoperative day 4 and during collection periods, samples were obtained at 3–10 min intervals. In two patients, the probe ceased to function on day 3. The samples were manually collected and analyzed using HPLC. Several novel technical methods were used in this investigation due to the unique nature
187
of the patient population under study. At the time, commercially-available microdialysis probes were not impregnated with any material which could be visualized radiographically (this has since changed), so published studies to that point in trauma and subarachnoid hemorrhage did not document the location of the probe. Given the large area influenced by those diseases, and the cortical position of the probes in most cases, this was not believed to be limiting. However, given the extremely small size of the STN and the SNr, and their deep locations, it would be difficult to draw conclusions regarding the physiology of these structures from the neurochemical data if there was no documentation that the probe was actually in the desired location. For the epilepsy studies, this was not an issue since the depth electrode fused to the microdialysis probe could be visualized. However, those electrodes were eventually removed prior to surgical resection, whereas the DBS electrode is a permanent implant. Therefore, it was not possible to fuse the microdialysis probe to the DBS electrode without either leaving the probe permanently in the brain or replacing the system after microdialysis with a new DBS electrode, both of which were unacceptable. Therefore, we chose to fuse a microelectrode wire to the microdialysis probe up to the beginning of the exposed membrane, but the membrane was not fused similar to what was done with the epilepsy dialytrode. With this, the probe could be visualized on intraoperative X-ray and on post-operative CT scans. This was critical, since a CT scan was performed on each patient following completion of the microdialysis sampling on day 3 or 4 after surgery, and the location relative to the DBS electrode confirmed that the probe location remained unchanged throughout the entire sampling period, so that all of the generated data reflected the neurochemical environment of the planned target in these patients. In addition, commercial probes at that time (and for the most part today as well) had a minimum exposed membrane length of
3125
3126
187
Microdialysis
10 mm in the dorsal-ventral direction. While this is reasonable for large cortical areas subject to injury, for a structure such as the STN or SNr, this would be too long as this would sample not only the structure in question but several mm above and/or below that structure. Therefore, again based upon the prior epilepsy experience, we manufactured probes in the laboratory which had a diameter of just over 300 mm and a length of exposed membrane of only 3 mm. The inlet and outlet tubings had inner diameters of only 75 and 100 mm respectively, so that the deadspace was roughly 3 ml/m of tubing. Therefore, this probe not only sampled specifically and only within the target structures, but also had such a small deadspace that the lag time between sampling in the brain and collection at the end of the tubing was only a few minutes. Overall, steady state levels of glutamate and GABA averaged 240 nM and 18–29 nM, respectively. GABA concentration was significantly higher in samples taken from the SNr compared to the STN; however, this difference was not appreciated for glutamate. A few conclusions were drawn from these studies regarding intrinsic steady state neurotransmission in these deep brain structures. The STN is mainly glutamatergic and the glutamate levels were on average 4–5 times higher than GABA levels in each patient. Conversely, the SNr is largely GABAergic and the levels of GABA in were 5- to 10-fold higher than in the STN. As a result, the authors concluded that the steady state neurotransmitter levels were largely a result of intrinsic neurons, as the less abundant neurotransmitter is released by synaptic inputs [4]. This also suggests the possibility that a neurochemical probe modeled on the microdialysis method could in the future be used either in addition to or in place of electrophysiological recordings for targeting these structures for DBS or other surgical interventions. Furthermore, although the relative difference in neurotransmitter between nuclei in this study was similar to results published during intraoperative sampling, the absolute amounts were different
[40]. Though this could be explained by patient variability, it may also reflect treatment effect on local neurochemical environment as the former study’s patients discontinued PD medications 2 weeks prior to surgery, while the latter discontinued medications on the day of surgery. Another possibility is the difference between acute and chronic sampling. In our study, the glutamate levels in the STN were substantially higher 2 h after surgery compared with 18–24 h later. Steady state was only reliably achieved by 36 h following surgery, at which point the daily levels were relatively consistent within patients. Therefore, it is possible that acute effects of insertional trauma may influence tissue neurotransmitter levels to a greater degree than previously appreciated. However, our study patients also had a DBS electrode inserted just prior to placement of the microdialysis probe, so the tissue trauma is likely to be far greater than when the microdialysis probe alone is inserted. Therefore, our early data immediately following surgery and for the ensuing 36 h may in fact be a neurochemical reflection of the microlesioning effect that is often seen for several days following DBS implantation. All six patients in this study had no complications at any point, and the microdialysis probes were completely removed without disrupting the DBS electrode location. Additionally, all patients had excellent clinical responses to their bilateral STN DBS. These findings highlight the potential for microdialysis to both define the normal physiologic environment within a deep brain nucleus and better elucidate the basis for pathologic disease states and mechanisms for treatment effects. It also suggests that such methods may be safely used to monitor brain neurochemistry even in situations where no brain tissue is to be removed or at risk and a device will be left permanently in place.
Less Common Uses Although its implementation is not universal in neurosurgical centers, microdialysis use as a
Microdialysis
research and clinical tool is becoming increasingly widespread, especially as commercially available setups such as CMA Microdialysis have received FDA approval. Other uses not mentioned in detail in this chapter include examining neurochemical marker profiles in patients with hepatic encephalopathy, hydrocephalus, malignant gliomas, and hyperventilation. Intraoperative studies have been conducted during lobectomy for TBI, aneurysm surgery, extracranial to intracranial carotid bypass, and spinal cord surgery in the dorsal root entry zone. Additionally, in the field of neuropharmacology, it is being used to measure extracellular concentrations of drug delivery to the brain and reverse microdialysis is being utilized for drug delivery [5]. Along with the more detailed studies outlined above, each of these applications reflects the versatility of microdialysis as both a research and clinical tool for analyzing human brain neurochemistry.
References 1. Bito L, Davson H, Levin E, Murray M, Snider N. The concentrations of free amino acids and other electrolytes in cerebrospinal fluid, in vivo dialysate of brain, and blood plasma of the dog. J Neurochem 1966;13:1057-67. 2. Zetterstrom T, Sharp T, Marsden CA, Ungerstedt U. In vivo measurement of dopamine and its metabolites by intracerebral dialysis: changes after d-amphetamine. J Neurochem 1983;41:1769-73. 3. Meyerson BA, Linderoth B, Karlsson H, Ungerstedt U. Microdialysis in the human brain: extracellular measurements in the thalamus of parkinsonian patients. Life Sci 1990;46:301-8. 4. Kaplitt MG, During MJ, Luo J, St.-Cyr J, Hutchinson WD, Dostrovsky JO, Ashby P, Lozano AM. Chronic in vivo microdialysis analysis of neurochemical responses to subthalamic nucleus deep brain stimulation in patients with Parkinson’s disease. In: Abstract Presented at the Society for Neuroscience, San Diego, CA, November 15, 2001. 5. Hillered L, Vespa PM, Hovda DA. Translational neurochemical research in acute human brain injury: the current status and potential future for cerebral microdialysis. J Neurotrauma 2005;22:3-41. 6. Wilson CL, Maidment NT, Shomer MH, Behnke EJ, Ackerson L, Fried I, Engel J Jr. Comparison of seizure related amino acid release in human epileptic hippocampus versus a chronic, kainate rat model of hippocampal epilepsy. Epilepsy Res 1996;26:245-54.
187
7. Tisdall MM, Smith M. Cerebral microdialysis: research technique or clinical tool. Br J Anaesth 2006;97:18-25. 8. Kermer P, Klocker N, Bahr M. Neuronal death after brain injury: models, mechanisms, and therapeutic strategies in vivo. Cell Tissue Res 1999;298:383-95. 9. Reinstrup P, Stahl N, Mellergard P, Uski T, Ungerstedt U, Nordstrom CH. Intracerebral microdialysis in clinical practice: baseline values for chemical markers during wakefulness, anesthesia, and neurosurgery. Neurosurgery (2000);47:701-9; discussion 709-10. 10. Schulz MK, Wang LP, Tange M, Bjerre P. Cerebral microdialysis monitoring: determination of normal and ischemic cerebral metabolisms in patients with aneurysmal subarachnoid hemorrhage. J Neurosurg 2000;93:808-14. 11. Roslin M, Henriksson R, Bergstrom P, Ungerstedt U, Bergenheim AT. Baseline levels of glucose metabolites, glutamate and glycerol in malignant glioma assessed by stereotactic microdialysis. J Neuro Oncol 2003; 61:151-60. 12. Fried I, Wilson CL, Morrow JW, Cameron KA, Behnke ED, Ackerson LC, Maidment NT. Increased dopamine release in the human amygdala during performance of cognitive tasks. Nat Neurosci 2001;4:201-6. 13. Khaldi A, Zauner A, Reinert M, Woodward JJ, Bullock MR. Measurement of nitric oxide and brain tissue oxygen tension in patients after severe subarachnoid hemorrhage. Neurosurgery 2001;49:33-8; discussion 38–40. 14. Bellander BM, Cantais E, Enblad P, Hutchinson P, Nordstrom CH, Robertson C, Sahuquillo J, Smith M, Stocchetti N, Ungerstedt U, Unterberg A, Olsen NV. Consensus meeting on microdialysis in neurointensive care. Intensive Care Med 2004;30:2166-9. 15. During MJ, Fried I, Leone P, Katz A, Spencer DD. Direct measurement of extracellular lactate in the human hippocampus during spontaneous seizures. J Neurochem 1994;62:2356-61. 16. Tofteng F, Jorgensen L, Hansen BA, Ott P, Kondrup J, Larsen FS. Cerebral microdialysis in patients with fulminant hepatic failure. Hepatology 2002;36:1333-40. 17. Cornford EM, Shamsa K, Zeitzer JM, Enriquez CM, Wilson CL, Behnke EJ, Fried I, Engel J. Regional analyses of CNS microdialysate glucose and lactate in seizure patients. Epilepsia 2002;43:1360-71. 18. Stamford JA. Monitoring neuronal activity. Oxford: Oxford University Press; 1992. 19. Amara SG. Methods in enzymology: neurotransmitter transporters, San Diego, CA: Academic Press; vol. 296. 1998. 20. Fried I, Wilson CL, Maidment NT, Engel J Jr, Behnke E, Fields TA, MacDonald KA, Morrow JW, Ackerson L. Cerebral microdialysis combined with single-neuron and electroencephalographic recording in neurosurgical patients. Technical note. J Neurosurg 1999;91:697-705. 21. Bungay PM, Morrison PF, Dedrick RL. Steady-state theory for quantitative microdialysis of solutes and water in vivo and in vitro. Life Sci 1990;46:105-19.
3127
3128
187
Microdialysis
22. Doppenberg EM, Choi SC, Bullock R. Clinical trials in traumatic brain injury. What can we learn from previous studies? Ann NY Acad Sci 1997;825:305-22. 23. Teasdale GM, Graham DI. Craniocerebral trauma: protection and retrieval of the neuronal population after injury. Neurosurgery 1998;43:723-37; discussion 737–8. 24. Persson L, Hillered L. Chemical monitoring of neurosurgical intensive care patients using intracerebral microdialysis. J Neurosurg 1992;76:72-80. 25. Valadka AB, Goodman JC, Gopinath SP, Uzura M, Robertson CS. Comparison of brain tissue oxygen tension to microdialysis-based measures of cerebral ischemia in fatally head-injured humans. J Neurotrauma 1998;15:509-19. 26. Landolt H, Langemann H, Gratzl O. On-line monitoring of cerebral pH by microdialysis. Neurosurgery 1993;32:1000-4; discussion 1004. 27. Goodman JC, Valadka AB, Gopinath SP, Uzura M, Robertson CS. Extracellular lactate and glucose alterations in the brain after head injury measured by microdialysis. Crit Care Med 1999;27:1965-73. 28. Robertson CS, Gopinath SP, Uzura M, Valadka AB, Goodman JC. Metabolic changes in the brain during transient ischemia measured with microdialysis. Neurol Res 1998;1(20 Suppl):S91-4. 29. Vespa PM, McArthur D, O’Phelan K, Glenn T, Etchepare M, Kelly D, Bergsneider M, Martin NA, Hovda DA. Persistently low extracellular glucose correlates with poor outcome 6 months after human traumatic brain injury despite a lack of increased lactate: a microdialysis study. J Cerebral Blood Flow Metab 2003;23:865-77. 30. Zauner A, Doppenberg EM, Woodward JJ, Choi SC, Young HF, Bullock R. Continuous monitoring of cerebral substrate delivery and clearance: initial experience in 24 patients with severe acute brain injuries. Neurosurgery 1997;41:1082-91; discussion 1091–3. 31. Peerdeman SM, van Tulder MW, Vandertop WP. Cerebral microdialysis as a monitoring method in subarachnoid hemorrhage patients, and correlation with clinical events – a systematic review. J Neurol 2003; 250:797-805. 32. Maurer MH, Haux D, Sakowitz OW, Unterberg AW, Kuschinsky W. Identification of early markers for symptomatic vasospasm in human cerebral microdialysate after subarachnoid hemorrhage: preliminary results of a proteome-wide screening. J Cerebral Blood Flow Metab 2007;27:1675-83. 33. Bullock R, Zauner A, Woodward J, Young HF. Massive persistent release of excitatory amino acids following human occlusive stroke. Stroke 1995;26:2187-9.
34. Schneweis S, Grond M, Staub F, Brinker G, Neveling M, Dohmen C, Graf R, Heiss WD. Predictive value of neurochemical monitoring in large middle cerebral artery infarction. [erratum appears in Stroke 2001;32(12):2961] Stroke 2001;32:1863-7. 35. During MJ, Spencer DD. Extracellular hippocampal glutamate and spontaneous seizure in the conscious human brain. Lancet 1993;341:1607-10. 36. Mattson RH, Scheyer RD, Petroff OA, During MJ, Collins TL, Spencer D. Novel methods for studying new antiepileptic drug pharmacology. Adv Neurol 1998;76:105-12. 37. Scheyer RD, During MJ, Hochholzer JM, Spencer DD, Cramer JA, Mattson RH. Phenytoin concentrations in the human brain: an in vivo microdialysis study. Epilepsy Res 1994;18:227-32. 38. Scheyer RD, During MJ, Spencer DD, Cramer JA, Mattson RH. Measurement of carbamazepine and carbamazepine epoxide in the human brain using in vivo microdialysis. Neurology 1994;44:1469-72. 39. Fedele E, Mazzone P, Stefani A, Bassi A, Ansaldo MA, Raiteri M, Altibrandi MG, Pierantozzi M, Giacomini P, Bernardi G, Stanzione P. Microdialysis in parkinsonian patient basal ganglia: acute apomorphine-induced clinical and electrophysiological effects not paralleled by changes in the release of neuroactive amino acids. Exp Neurol 2001;167:356-65. 40. Fedele E, Stefani A, Bassi A, Pepicelli O, Altibrandi MG, Frasca S, Giacomini P, Stanzione P, Mazzone P. Clinical and electrophysiological effects of apomorphine in Parkinson’s disease patients are not paralleled by amino acid release changes: a microdialysis study. Funct Neurol 2001;16:57-66. 41. Galati S, Mazzone P, Fedele E, Pisani A, Peppe A, Pierantozzi M, Brusa L, Tropepi D, Moschella V, Raiteri M, Stanzione P, Bernardi G, Stefani A. Biochemical and electrophysiological changes of substantia nigra pars reticulata driven by subthalamic stimulation in patients with Parkinson’s disease. Eur J Neurosci 2006;23:2923-8. 42. Stefani A, Fedele E, Galati S, Pepicelli O, Frasca S, Pierantozzi M, Peppe A, Brusa L, Orlacchio A, Hainsworth AH, Gattoni G, Stanzione P, Bernardi G, Raiteri M, Mazzone P. Subthalamic stimulation activates internal pallidus: evidence from cGMP microdialysis in PD patients. Ann Neurol 2005;57:448-52. 43. Stefani A, Fedele E, Galati S, Raiteri M, Pepicelli O, Brusa L, Pierantozzi M, Peppe A, Pisani A, Gattoni G, Hainsworth AH, Bernardi G, Stanzione P, Mazzone P. Deep brain stimulation in Parkinson’s disease patients: biochemical evidence. J Neural Transm Suppl 2006; 40:401-8.
195 Future Ethical Challenges in Neurosurgery N. Lipsman . M. Bernstein
As the practice of neurosurgery keeps pace with neuroscience research in the twenty-first century, it will be fraught with new ethical challenges. As technology continues to advance, the onus will be on the practicing neurosurgeon to justify the application of novel technology to new and evolving patient populations. The purpose of this chapter is to examine the present, and to anticipate the future ethical challenges that will face functional neurosurgery and to discuss how those challenges can be approached, and ultimately best managed, by building an ethical framework into clinical and research practice.
Introduction Given the emphasis placed in modern medicine, both by researchers and the public, on patient autonomy, identity and free will, it is not surprising that research in the neurosciences gives rise to numerous issues prone to ethical discussion and debate. Indeed, the term neuroethics has been applied to the discussion of bioethical issues in a neuroscientific context [1–3]. The last decade has seen a flurry of progress in neuroscience, and many view the development of the field of neuroethics as a step that will provide the necessary checks and balances required for responsible practice. As with any question in the medical sciences, it is important to approach ethical issues in neurosurgery systematically. Ethical questions, as they apply to neurosurgical practice, come primarily from two sources, namely from within as well as from outside the field of medicine. Within #
Springer-Verlag Berlin/Heidelberg 2009
the practice of medicine, physicians and researchers are asking questions like how to best approach target selection for surgical intervention, and once selected, whether sham surgeries are ethical to control for the patient’s, and researcher’s, expectations [4,5]. Furthermore, issues such as the proper acquisition of informed consent prior to experimental procedures, as well as discussion surrounding the precise nature and definition of a pathologic disease, remain hot topics of debate [6,7]. From outside the field, represented largely by public interest and the popular media, there are questions surrounding the direction of neurosurgical technology and whether progress is being made at the expense of patient autonomy. Again, issues of patient, and disease, selection continue to arise, as well as the notion that something of humanity may be compromised should technology manipulate so intimately the human brain. Given the diversity of the questions identified, we suggest a framework to adequately discuss and interpret the present and future ethical issues facing neurosurgery. Ethical questions can primarily be divided into three broad categories: those surrounding the disease, those affecting the patient, and those related to issues of technology (> Figure 195‐1). In this chapter we will attempt to focus on the issues within each category that we believe to be most pertinent and timely, namely those surrounding neuropsychiatric disease, personal identity and self, as well as enhancement technology. We will then turn our attention to the future and anticipate the challenges that await neuroscience and neurosurgery in the years to come.
3230
195
Future ethical challenges in neurosurgery
. Figure 195-1 Framework for ethical issues in functional neurosurgery
Neuropsychiatric Disease The application of neurosurgery to psychiatric disease is not a novel idea [8]. However, there has been a recent paradigm shift in the neurosurgical world, whereby reversible, and relatively safe technology such as deep brain stimulation (DBS) is being used to alleviate psychiatric complaints. Ironically, the relative safety and efficacy of DBS could also raise several ethical questions. One can anticipate the concern that some may view DBS as a panacea and cast a wide net when considering which patients and diseases to select for operation. Since DBS is reversible, and doesn’t expose patients to the same risks associated with craniotomy or permanent brain lesioning, regulations need to be established to ensure that this powerful technology is wielded responsibly.
Disease Selection As psychiatric disease is a dynamic substrate for neuroscience research, attention will be turned
towards precisely which diseases could, and should, be amenable to surgical intervention. Traditionally the mood and anxiety disorders have been offered up as good candidates, largely because their presumed dysfunctional underlying neurocircuitry has been linked to that of the movement disorders [9,10]. The neuroanatomy and physiology of voluntary movement, involving direct and indirect pathways subserved by cortical and subcortical structures, has been the substrate of successful neurosurgical intervention with DBS and lesioning procedures [11]. The assumption that these pathways are functionally linked in a parallel fashion with the dysfunctional circuits underlying obsessivecompulsive disorder (OCD), for example, has led to lesioning and stimulation based procedures being explored as therapeutic alternatives in treatment refractory patients. Although still in the experimental stage, DBS for OCD has met with some success; a success that must be moderated, however, by the variability in the targets selected and the quality of studies performed [12]. Additional
Future ethical challenges in neurosurgery
progress has been made in the realm of treatment refractory depression where targets were selected based on functional neuroimaging and animal studies. Results are promising, pointing to a successful resumption of normal activities in a majority of patients previously rendered incapacitated by their psychiatric symptoms, which were refractory to more conventional therapeutic options [13]. Although far from perfect results, the trials of DBS in psychiatric disease and the history of relative success of lesion based procedures for some mood and anxiety disorders do provide valuable information. First is the idea, often taken for granted, that mind, rather than body, can be a substrate for surgical intervention; although we are operating on a physical structure, brain matter, our end goal is not the physical manipulation of that structure, as it is in, for example, brain tumor surgery. It is instead the manipulation of a state of mind, of a way of thinking. Secondly, neuroimaging is becoming an increasingly powerful tool that is allowing clinicians to evaluate not only neuroanatomic but also neurophysiologic changes within their patients. When applied to ‘‘normal, healthy’’ patients, the implications for fields such as lie detection and criminology become clear [14]. As researchers begin to recognize patterns of brain activation in areas subserving specific functions, this allows them to investigate avenues such as pre-disposition to certain disease and possibly the once far-flung idea of ‘‘reading’’ thoughts. Underlying the discussion of which diseases are appropriate for surgical intervention is the basic idea that anatomy and physiology are inextricably linked to function. It is not yet known whether it is function that fashions anatomy, or anatomy that determines the function. Nevertheless, when determining which, particularly neuropsychiatric, illnesses should be approached surgically, we must first ask what threshold of plausible evidence has been surpassed to consider surgery a viable alternative to standard care.
195
For OCD, depression, and now Tourette’s syndrome, a specific level of disability has to be demonstrated and a wealth of genetic and imaging evidence has accumulated that supports the role of certain structures in the manifestation and etiology of these diseases. This threshold is necessary to avoid exposing patients to unnecessary risk associated with surgery. Importantly, the more vulnerable the patient, in terms of functional impairment, reality testing, and cognitive functioning, the higher that threshold must become. The selection of a disease for surgical intervention is a multidisciplinary effort, and must be approached carefully in a psychiatric context. In addition to being treatment refractory, the right disease should be one that is linked on an anatomic level to existing circuits as demonstrated by neuroimaging and/or animal models (macroscopic), it should have sufficient physiologic and genetic evidence supporting a somatic intervention (microscopic), and it shouldn’t expose the patient to unnecessary risk. After selecting an appropriate disease, researchers can turn their attention to safely selecting patients for a targeted and focused intervention that provides a viable alternative to the widely distributed actions of psychopharmacologic agents.
Patient Selection Surgery for psychiatric disease, perhaps more so than other illnesses, cannot exist in a vacuum [15]. Although psychosurgery has existed for decades, only recently has a multidisciplinary approach been applied to the practice. It is not unusual, and indeed it is necessary, for neurosurgeons to work together with psychiatrists, neuropsychologists, neurophysiologists, social workers and community care workers to ensure proper care and selection of patients for surgery. Among the areas most relevant to ethicists is the notion of informed consent and to
3231
3232
195
Future ethical challenges in neurosurgery
what extent it can be achieved prior to complex surgery in general, and prior to functional neurosurgery for psychiatric disease in particular. Some have posited that given this complexity, as well as the inability to predict every risk and the power differential existing between physician and patient, that truly informed consent is nearly impossible to obtain [7]. This idea is heightened in the context of surgery for psychiatric disease where the refractory nature of the illness, itself a prerequisite for surgery, can make some patients desperate and more willing to go along with an as-yet unproven procedure. When one considers the prospect of operating on patients with impaired reality testing, or severe paranoid delusions as is the case with schizophrenics with positive symptoms, the importance of informed consent and its validity become much more important. In these instances, it is protecting the patient not only from the potential physical harm of the procedure that is important, but also protecting them from the misguided use of a new technology in the name of progress. Patient selection is arguably the most important component of any procedure. It is the responsibility of the neurosurgeon and the multidisciplinary team, practicing in an ethical manner, to match the patient to the procedure, to obtain as close as possible to fully informed consent, and to ensure that alternatives to the present course of action have been disclosed and discussed. As neurosurgery evolves, as the diseases amenable to surgical intervention increase in number, and as advanced technology becomes a crucial component of clinical practice, this responsibility will increase proportionately.
Personal Identity and Self Diseases affecting the brain are unique in that both their natural course and treatment may potentially alter a patient’s sense of personal
identity. Although changes in personality and cognitive deterioration may be the initial harbingers of neurological disease, it is often the case that the treatment, namely neurosurgical intervention, presents an entirely unique set of risks that places an individual’s sense of self in harm’s way. Furthermore, the use of functional neurosurgical procedures with the explicit intent of altering personality traits, leads to new challenges in the interpretation of identity change. The question then arises as to whether personal identity can, and should, be used as a substrate for surgical intervention. Issues of identity are not unique to neurosurgery, and indeed are common in plastic and general surgery where various operations can have an impact on an individual’s self-perception. Neurosurgery is unique, however, in primarily three respects. First, is that neurosurgery has the potential of affecting not only the components of one’s identity, such as memory or cognition, but it also risks the more fundamental sense of self that is separate from identity. Philosophers and neuroscientists frequently make this distinction between self and identity. The Self can be defined as the personal feeling that one’s thoughts and actions are one’s own. Personal identity can be defined as the sum total of the experiences and characteristics that one attributes to oneself. Identity, therefore, is largely dependent on the normal functioning of memory, as many will support the claims about themselves by referring to their autobiographical past. So although mastectomy can in some ways alter one’s identity, by affecting some womens’ perception of their femininity, only neurosurgery has the potential to alter, or change the Self. Second, is the notion that, related to identity, crucial parts of the human psyche reside in the brain; functions such as memory, emotion, and cognition. Finally, and perhaps most relevant to discussions of the ethical use of neurosurgical technology, is the idea that some neurosurgical operations, such as DBS for depression, make it
Future ethical challenges in neurosurgery
their explicit objective to alter personal identity. This point is behind the paradigm shift in functional neurosurgery, whereby minimally invasive surgical intervention could be used to alter the natural course of not only physical disability, but mental disability, and as a result, personality, as well. Although there is no consensus in the literature regarding the neuroanatomy of self-referential processes, the search for the neurological correlates of self remains an area of intense and fascinating research. One study, utilizing functional magnetic resonance imaging (fMRI) techniques in five subjects, found significant activation in right superior, middle and inferior frontal gyri, when participants viewed pictures of themselves [16]. The anterior right hemisphere was also implicated in the processing of self-information in a study that demonstrated several subjects’ failure to recognize their own face as familiar during an intracarotid amobarbitol (Wada) test for determination of hemispheric dominance [17]. Other studies point to cortical midline structures, such as orbital and medial prefrontal cortex as well as anterior and posterior cingulate regions, as having major roles in the processing of selfreferential information [18]. This position has been strengthened recently by a meta-analysis that echoed the importance of cortical midline structures for the processing of self-relevant stimuli across several domains [19]. The authors further concluded that owing to complex reciprocal connections involving the densely populated cortical midline, a more broad solution may be a ‘‘cortical-subcortical midline system underlying human self.’’ In this way, the search for the self, and personal identity, moves away from a search for single neuroanatomic structures, and towards a distributive model. The implications for neurosurgery are clear, as procedures exist that may either indirectly harm structures important for the selfconcept, or that may directly attempt to influence those same structures in an effort to modify personality. Indeed, the anterior cingulate region, a
195
midline cortical structure, has already been targeted with significant success by DBS procedures for treatment of refractory depression [13]. It is known that personality and cognitive alterations are at the very least potential consequences of surgery. Several authors have found both short and long term cognitive effects with temporal lobe resection for epilepsy [20,21]. Another study demonstrated the presence of increased apathy, without depression, in patients who underwent DBS for Parkinson’s Disease [22]. Such studies point at the intimate relationship between motor and affective neuroanatomic circuits, and illustrate that diverse neurosurgical interventions can indeed affect components of personal identity. The negative effects of whole brain radiation therapy (WBRT) on neurocognitive functioning are known [23,24]. Gamma Knife Radio Surgery can reduce the deleterious sequelae of WBRT while preserving clinical efficacy. As cognition is intimately linked to identity, the impact of GK procedures on cognitive capacity is important to explore. One study, utilizing interviews with 47 GK patients, found that 13% experienced feelings of depression, and aggressive behavior in the post-operative period while an additional 10% found increased prevalence of ‘‘spacing out’’ and ‘‘fuzzy’’ thoughts [25]. Other studies examining GK surgery for the management of brain metastases have found no significant cognitive consequences, as well as effective improvements in quality of life [23,26]. The literature on the short and long-term complications of GKRS remains sparse as trials continue to explore the clinical efficacy of the procedure across a wide range of diagnoses. The possibility of identity and/or personality alteration is a reality of neurosurgical practice. In addition to these alterations being a risk of surgery, they are increasingly becoming the central objective of surgery as well, namely the alteration of maladaptive or dysfunctional personality traits such as amotivation and anhedonia. The ethical
3233
3234
195
Future ethical challenges in neurosurgery
challenges will come not only when determining which patients are sufficiently sick to proceed to surgery, and whether the risk of surgery is justified, but also where the pathological line is drawn in the sand. The possibility of surgery to alter maladaptive personality traits as they exist within a recognized psychiatric disease, will lead to discussions of whether dysfunctional and maladaptive traits that exist outside of recognized pathological criteria should be amenable to surgical intervention as well. It is from this corner that discussions about cognitive enhancement, or the alteration of ostensibly normal human characteristics, arise, and where neuroethics must tackle some of the most serious and potentially troubling challenges.
Neuroenhancement The term cosmetic neurology has been applied to the field of enhancement, by pharmacologic means, of normal human characteristics and personality traits. Although largely theoretical at this point, some have speculated that the advent of technology that aims to assist or enhance mental functioning will radically change the face of medicine in the twenty-first century [27,28]. Issues such as who will dispense the technology, who should have access to it, and even whether research into such technology should be conducted are presently being hotly debated. Suffice it to say, that no clear answers exist yet, but a relative consensus is emerging that enhancement technology will be a reality. The question can thus be raised whether the era of cosmetic neurosurgery, can be that far off. Enhancement technology exemplifies the ethical challenges that await neurosurgeons and neuroscientists in the years to come. There is, and will continue to be, a justifiable and important tension between progress and the elucidation of the mechanisms of thought and behavior, and respect
for patient autonomy, safety and good responsible science. As more is learned about the anatomy and physiology of normal cognition and behavior, and as the circuitry of psychiatric and psychologic disturbance becomes clearer, it will be time to consider taking the additional step of safely altering characteristics that are maladaptive, or even potentially dangerous, but that don’t necessarily meet the criteria for a pathologic condition. Several pre-requisites must first be achieved, which themselves remain in relative infancy. A clear line must be drawn from anatomy and physiology to behavior. Further, any intervention must have demonstrated specificity, such that local disturbances are not extended to adjacent, and unrelated areas in the brain. Also, in addition to being effective and safe, any intervention must surpass a significant enough threshold of efficacy that permits the exposure of patients to the risks associated with surgery. Most discussions of enhancement assume that the technology will necessarily be applied to ostensibly normal human traits, namely those that fall into the extremes of normal variation. Although this is largely true, enhancement technology has additional implications to the world of the disabled, who at present must either live with relatively low-tech enhancements such as artificial limbs, or learn to adapt to a life of some impediment. Surgery for neuroenhancement may offer these patients an additional alternative and nowhere is this more clear than in research dealing with brain-machine interface. Several institutions are making significant progress in allowing paralyzed patients to interact with, and alter, their surroundings by merely thinking about, literally ‘‘willing,’’ a specific movement [29]. To some this represents a genuine revolution in neuroscience, whereby technology is bridging mind and body in severely handicapped individuals. The question that neuroscientists and neurosurgeons should be asking, however, is to what
Future ethical challenges in neurosurgery
extent they are comfortable with implanting artificial devices into the brains of patients with the explicit aim of cognitive and/or functional enhancement. Asked in another way, as we learn more about the brain’s various functions, how comfortable are we with machinery becoming a permanent resident of the human brain. Although it is a lofty ambition to improve the lives of the disabled, and the research is welljustified, are we willing to deal with this type of technology being used one-day in normal subjects, and if not, what mechanisms can we establish to prevent that from happening. One of the concerns with the cloning debate, was that active research, and active media and public attention, would somehow fuel research in ‘‘rogue’’ laboratories. This led to the worry that cloning technology would be ultimately usurped and applied to less than noble purposes. This possibility applies to the neuroenhancement debate, as one wonders whether continued research in human enhancement will one day lead to the development of technology allowing, for example, extreme vision, fine listening, and perhaps heightened intelligence in otherwise ‘‘normal’’ or ‘‘average’’ individuals. Another possibility is that of taking a ‘‘normal,’’ happy individual and by enhancement, making them ‘‘super-happy.’’ Neuroenhancement is perhaps one of the most exciting areas of neuroscience and neurosurgery research, and offers the promise of not only relieving significant impairment in some patients, but also of revealing important questions surrounding the limits of responsible and ethical science and clinical intervention.
Future Directions New avenues of potential exploration for functional neurosurgery are truly limited only by the imagination of the research team involved. As outlined in this chapter, fundamental questions surrounding human consciousness, personal
195
identity, cognitive and perceptual functioning and the roots of mood and behavior are now being explored within a neuroscientific context. These questions not only provide insight into the human experience but offer surgeons the opportunity to better understand the diseases they are confronted with and to provide viable and effective therapeutic alternatives. In order for their work to continue unimpeded, however, researchers must actively build an ethical framework into their research. By formulating, and then addressing, questions surrounding the disease they are trying to treat, the patients they want to recruit, and the technology they hope to develop, researchers are striving to immunize their research from questionable ethical and moral conduct. An example for how this could be done for enhancement technology is shown in > Figure 195‐2. Although much of this chapter dealt with present issues in the functional neurosurgical world, we can begin to anticipate some of the challenges that await the field in the future. Two factors will ultimately contribute to the wider interest in using neurosurgical technology for more diverse applications. The first will be an improved understanding of the underlying mechanism of behavior and mood on both a macro, i.e., neuroanatomic level, and micro, i.e., neurophysiologic and genetic level. Increasingly sophisticated imaging technology, coupled with, for example, innovations in genetic modeling of disease, will open up new avenues for potential intervention. Neuroscientists, are just now understanding the complex mechanics underlying addiction behavior. It is possible to envision a targeted intervention aimed at breaking the destructive behavioral cycles that are at the root of addiction, given sufficiently accurate localization that is informed by solid neuroscientific evidence. The second factor that will contribute to the exponential rise of functional neurosurgery is the shift towards minimally invasive procedures, where attempts are being made to minimize both the risk to patients and damage to normal, healthy
3235
3236
195
Future ethical challenges in neurosurgery
. Figure 195‐2 Enhancement technology in an ethical context. In order to adequately assess enhancement technology in an ethical context, the question needs to be framed within an ethical framework, using the three categories described in the text. For example, there needs to be an agreed upon definition of disease, we must prove that the technology is safe, and informed consent must be achieved. The three categories are inter-related, such that the right technology must be matched to the right patient with the right disease
tissue. The increasing use of DBS is a testament to this fact and exemplifies the move towards technology that is effective, and safe. In the coming decades, as neurosurgical technology evolves and matures, we believe that functional neurosurgery will take a central position in the field of neuroscience. The treatment of psychiatric disease will more frequently include surgical alternatives for patients deemed refractory to more traditional therapies. In addition to depression, obsessive-compulsive disorder, and Tourette’s syndrome, disorders affecting reality testing such as bipolar disorder and schizophrenia as well as disorders of impulse such as severe obesity, drug addiction and alcoholism will have neurosurgical options for treatment. We also believe that the coming generation will see a fusion of machine and brain, that will not only give hope to patients living with paralysis, but will also blur the line between the body’s natural ‘‘hardware,’’ and the artificial ‘‘hardware’’ implanted in these patients. Significant strides will be made in the twenty-first century in discovering the neural correlates of consciousness, and functional neurosurgery will be in a prime position to contribute to that most elusive of goals. The mapping of the human brain, its innumerable functions and complex underlying circuitry, will provide neurosurgeons with a
clearer map for exploration, that will ultimately lead to new and effective treatments for patients. Each innovation, however, must be tempered with warnings that protect the very patients we are attempting to treat. The ethical challenges awaiting neurosurgeons in the coming years are significant, however these are challenges that are necessary to meet for the continued development and growth of our field.
Acknowledgment This work was supported in part by CIHR Grant NNF 80045.
References 1. Fuchs T. Ethical issues in neuroscience. Curr Opin Psychiatry 2006;19:600-7. 2. Glannon W. Neuroethics. Bioethics 2006;20:37-52. 3. Illes J, Bird SJ. Neuroethics: a modern context for ethics in neuroscience. Trends Neurosci. 2006;29(9):511-7. 4. Frank SA, Wilson R, Holloway RG, Zimmerman C, Peterson DR, Kieburtz K, Kim SY. Ethics of sham surgery: perspective of patients. Mov Disord 2008;23 (1):63-8. 5. Angelos P. Sham surgery in clinical trials. JAMA 2007;297:1545-6. 6. De Santis A. Reflections on informed consent. J Neurosurg Sci 2007;51:39-43.
Future ethical challenges in neurosurgery
7. Bernstein M. Fully informed consent is impossible in surgical clinical trials. Can J Surg 2005;48:271-2. 8. Sakas DE, Panourias IG, Singounas E, Simpson BA. Neurosurgery for psychiatric disorders: from the excision of brain tissue to the chronic electrical stimulation of neural networks. Acta Neurochir Suppl 2004;97:365-74. 9. Alexander GE, DeLong MR, Strick PL. Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annu Rev Neurosci 1986;9:357-81. 10. Kopell BH, Greenberg B, Rezai AR. Deep brain stimulation for psychiatric disorders. J Clin Neurophysiol 2004;21:51-67. 11. Hamani C, Richter E, Schwalb JM, Lozano AM. Bilateral subthalamic nucleus stimulation for Parkinson’s disease: a systematic review of the clinical literature. Neurosurgery 2005;56:1313-21. 12. Lipsman N, Neimat JS, Lozano AM. Deep brain stimulation for treatment-refractory obsessive-compulsive disorder: the search for a valid target. Neurosurgery 2007;61:1-11. 13. Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, Schwalb JM, Kennedy SH. Deep brain stimulation for treatment-resistant depression. Neuron 2005;45:651-60. 14. Illes J, Racine E. Imaging or imagining? A neuroethics challenge informed by genetics. Am J Bioeth 2005;5:5-18. 15. Fins JJ, Rezai AR, Greenberg BD. Psychosurgery: avoiding an ethical redux while advancing a therapeutic future. Neurosurgery 2006;59:713-6. 16. Platek SM, Keenan JP, Gallup GG, Mohamed FB. Where am I? The neurological correlates of self and other. Brain Res Cogn Brain Res 2004;19:114-22. 17. Keenan JP, Nelson A, O’Connor M, Pascual-Leone A. Self-recognition and the right hemisphere. Nature 2004;409:305. 18. Northoff G, Bermpohl F. Cortical midline structures and the self. Trends Cogn Sci 2004;8:102-7.
195
19. Northoff G, Heinzel A, de Greck M, Bermpohl F, Dobrowolny H, Panksepp J. Self-referential processing in our brain–a meta-analysis of imaging studies on the self. J Neuroimage 2006;31:440-57. 20. Rausch R, Kraemer S, Pietras CJ, Le M, Vickrey BG, Passaro EA. Early and late cognitive changes following temporal lobe surgery for epilepsy. Neurology 2003;60:951-9. 21. Hermann BP, Seidenberg M, Haltiner A, Wyler AR. Adequacy of language function and verbal memory performance in unilateral temporal lobe epilepsy. Cortex 1992;28:423-33. 22. Drapier D, Drapier S, Sauleau P, Haegelen C, Raoul S, Biseul I, Peron J, Lallement F, Rivier I, Reymann JM, Edan G, Verin M, Millet B. Does subthalamic nucleus stimulation induce apathy in Parkinson’s disease? J Neurol 2006;253:1083-91. 23. Levin KJ, Youssef EF, Sloan AE, Patel R, Zabad RK, Zamorano L. Gamma knife radiosurgery in patients with advanced breast cancer undergoing bone marrow transplant. J Neurosurg 2002;97:663-5. 24. Crossen JR, Garwood D, Glatstein E, Neuwelt EA. Neurobehavioural sequelae of cranial irradiation in adults: a review of radiation-induced encephalopathy. J Clin Oncol 1994;12:627-42. 25. St. George EJ, Kudhail J, Perks J, Plowman PN. Acute symptoms after gamma knife radiosurgery. J Neurosurgery 2002;97 Suppl 5:631-4. 26. Vesagas TS, Aguiler JA, Mercado ER, Mariano MM. Gamma knife radiosurgery and brain metastases: local control, survival and quality of life. J Neurosurg 2002;97:507-10. 27. Chatterjee A. The promise and predicament of cosmetic neurology. J Med Ethics 2006;32:110-13. 28. Dees RH. Cosmetic neurology: the controversy over enhancing movement, mentation, and mood. Neurology 2005;64:1320. 29. Birbaumer N, Cohen LG. Brain-computer interfaces: communication and restoration of movement in paralysis. J Physiol 2007;579:621-36.
3237
192 The Future of Cell Transplantation M. B. Newman . R. A. E. Bakay
Introduction Ramon y Cajal (1852–1934) established the dogma that once neural development ended the axons and dendrites were fixed and the nerve paths immutable. He observed no neural regeneration. Similarly, Giulio Bizzozero (1846–1901) classified the tissues of the human body into ‘‘labile, stable, and everlasting.’’ Neurons were ‘‘everlasting formed by post-mitotic cells’’ and believed to be unable to regenerate in the postnatal brain. This central dogma of neurobiology was taught up until the last decade, to generations of medical students and neuroscientists all over the world. This concept has dominated neuroscience and influenced the clinical approach and experimentation of neurological disorders. The demonstration of functional restoration with neural transplantation in the 1970s provided the hope for neural repair. The initial effort was to structurally replace site-specific focal cell loss with neuroblasts. From this work came the appreciation of the power of neurotrophic factors in the adult brain, and the potential for adult neural tissue to replace and repair itself. However, not until the 1990s and the discovery of neural stem cells (NSCs) in the adult brain was adult neurogenesis considered undeniable [1,2]. Originally, the field of cell transplantation was limited to a cell specific replacement for loss of cells in neurological disease or injury. However, now both specific cell and multiple cell types, along with widespread non-specific beneficial effects that might attenuate detrimental inflammation, protect against neurodegeneration, and enhance the endogenous reparative #
Springer-Verlag Berlin/Heidelberg 2009
and/or replacement process are considered state of the art cellular therapeutic strategies. In this chapter, we present a brief review of the history of neural transplantation and discuss the future prospects of NSC therapy for neurological diseases. The focus will mostly be on Parkinson’s disease (PD), as this is where the greatest insight can be achieved. Understanding the clinical trials of cell transplantation to treat PD along with the limitations and caveats that these trials brought out and those that need to be surmounted is essential before stem cells are readily available as a therapeutic agent to treat CNS diseases and disorders.
Transplantation in Parkinson’s Disease PD is a neurodegenerative disorder in which the striatum becomes deficient in dopamine (DA) secondary to the degeneration of DA-producing neurons in the substantia nigra pars compacta [3]. This results in a progressive movement disorder characterized by muscle rigidity, tremors, and a slowing of physical movement. The primary symptoms are the result of decreased stimulation of the motor cortex by the basal ganglia, which is caused directly by the insufficient availability of DA and the loss of DA producing cells [4]. The treatment of PD has focused on dopamine replacement via the pro-drug levodopa (L-dopa), which is converted to dopamine. L-dopa continues to be the most efficacious available treatment for PD, by providing substantial clinical benefit, and modifying cardinal symptoms for almost all
3162
192
The future of cell transplantation
PD patients. However, several limitations prevent its long-term clinical effectiveness [5]. First, motor complications from L-dopa such as dyskinesias and motor fluctuations developed in up to 80% of PD patients 5–7 years after starting treatment [6]. Second, PD patients develop symptoms unresponsive to L-dopa therapy including freezing episodes, postural instability, falls, autonomic dysfunction, constipation, depression, and dementia. Third, despite the early symptomatic value of L-dopa, the underlying neurodegenerative process in PD continues unabated, with most patients experiencing permanent disability. Altogether, chronic treatment with L-dopa coupled with advancing disease eventually results in severe disability for patients. Over time, it becomes increasingly difficult to deliver a dose of L-dopa that both controls parkinsonian motor features and avoids dyskinesias. Patients with advanced PD may cycle between disabling ‘‘on’’ (time when the patients feel that the medication is effective in controlling their symptoms and have their typical degree of benefit) and ‘‘off’’ (time when the patients feel that the medication is ineffective to control their symptoms) states for the majority of each day. Current medical approaches for the treatment of L-dopa-induced motor complications include manipulation of the dose and frequency, addition of DA agonists, long-acting formulations of L-dopa, catechol-O-methyl transferase inhibitors, monoamine oxidase-B inhibitors, and N-methyl-D-aspartic acid receptor antagonists. However, disabling side effects continue to be a major problem [7]. These treatments can provide benefit to individual patients, particularly when motor complications are mild, but are ineffective for patients with established motor fluctuations and dyskinesias. Continuous mechanical systemic administration of L-dopa can prevent the occurrence of these disabling side effects [6,8,9]. However, most methods of administration are impractical and unacceptable to patients with PD, and none of these methods
have been approved by the FDA. The problems discussed here limit the long-term utility of L-dopa and have resulted in a search for more effective treatment strategies. This has allowed for the development of novel neuroprotective and restorative therapies, which have the potential to greatly improve the quality of life for PD patients. Among neurological disorders, PD is one of the most amenable for cell transplantation because of the strong relationship between the cardinal symptoms and the striatal dopamine insufficiency. Consequently, the transplantation of DA secreting cells into the striatum is anticipated to assist in re-establishing striatal function and restore movement in PD patients. Neural transplantation, as a clinical therapy for PD, started in 1985, with autologous transplants of chromaffin cells of the adrenal gland. Transplanted chromaffin cells can produce dopamine, norepinephrine, and epinephrine. The first clinical trials of cell transplants were performed in Sweden in a unique series of clinical investigations by Backlund et al., which consisted of stereotaxic transplantation of adrenal medullary chromaffin cells into the striatum of PD patients [10]. However, the results of these surgeries were disappointing. Then a microsurgical intraventricular approach was initially reported to be successful, but subsequent results were again disappointing with patients experiencing very modest and short-lived (1–2 years) improvement in symptoms [11]. Additionally, 40% of the 126 patients, who underwent surgery in the USA, suffered from side effects of the procedure [12]. Patients’ data from autopsy revealed that no DAproducing chromaffin cells had survived, even though the host showed robust re-innervations around the graft [13]. Interestingly, Bakay et al. [14] found a direct correlation between the acute improvements and the number of TH immunoreactive boutons in the caudate of patients removed at the time of surgery, but not for the amount of medullary tissue grafted (> Figure 192-1). Thus,
The future of cell transplantation
192
. Figure 192-1 (a) The photomicrograph shows the paucity adrenal medullary chromaffin graft survival, but luxurious dopaminergic fiber response to the implant. The extent of the neurotrophic effect related mostly to the host and not as much to graft. (b) This photomicrograph shows the caudate removed from a patient with >40% improvement in which a large number of dopaminergic processes were retained, whereas in (c) the patient had <10% improvement and very few dopaminergic processes. There was a statistically significant correlation between clinical improvement and dopaminergic profiles
the improvement in some of the patients’ symptoms was likely the result of neurotrophic effects on the host surviving DA neurons [13–16]. Consequently, the transplantation of chromaffin cells was halted because of the poor success rate [16]. Nevertheless, this opened the way for other clinical studies of neural transplantation.
Clinical Outcomes of Fetal Tissue/ Cell Transplants in Parkinson’s Disease In PD the objective of neural transplantation has been the replacement of loss DA neurons, which makes the use of fetal ventral mesencephalon (VM) tissue very advantageous. Previous studies have shown the targeted delivery of fetal tissue to the desired site can result in restoring long-term function with the integration of fetal neurons into the host brain tissue [16]. Transplantation of fetal neuronal tissue and cells, in PD, has the advantage of potentially providing synaptic
replenishment of DA processes that are capable of regulating the DA output. Theoretically such transplantation avoids or ameliorates the complications of chronic intermittent L-dopa therapy. The disadvantages of fetal tissue for transplantation include supply shortage, storage, and expansion problems, along with the need for immunosuppression therapy and the potential contamination in culture and within the brain. Moreover, such studies are extremely expensive and raise numerous ethical concerns [17–21]. In 1988, human fetal DAergic transplants replaced chromaffin cell transplants. During the following decade, more than 200 patients received fetal DA neuron grafts worldwide [16]. The results of these transplantation trials were highly variable in terms of transplantation methods, improvement in symptoms, and development of side-effects. Initial transplantation studies performed in the 1990s, which used allogenic human fetal VM tissue transplantation, were un-blind and employed a small number of patients, but reported greatest clinical benefit from DAergic
3163
3164
192
The future of cell transplantation
cell transplantation [22–29]. In an attempt to compare fetal VM tissue transplantation with other surgical procedures used in the treatment of advanced PD, 19 research papers published from 1991 to 1998 were assessed by experts in this field [30]. They concluded, ‘‘Human fetal transplantation into the striatum in PD appears to be an encouraging procedure when performed on severely advanced PD patients. It is promising, because its efficacy appears to be good in the published open trial reports, and its associated morbidity and mortality are low [30].’’ In order to obtain objective efficacy and safety data, the National Institutes of Health (NIH) funded two multicenter placebo controlled prospective double blind studies. The purpose of these studies was to objectively determine whether striatal FVM transplantation provides clinically meaningful benefit to PD patients. The design of each study included a ‘‘sham’’ surgery arm in which one half of enrolled patients were randomized to undergo an imitation procedure involving general anesthesia, cranial burr holes and identical post operative care. The first NIH funded double blinded, placebo controlled, collaborative clinical trial combined teams of investigators from Columbia University and the University of Colorado [31,32]. In this study 40 PD patients from the ages of 34 and 75 were randomized to receive either fetal VM implants or sham surgery. Grafted patients received solid tissue grafts derived from a total of four embryos. The tissue was in shape of strands that had been cultured for up to 4 weeks and then implanted in the putamen using a frontal approach. In the sham transplanted patients, holes were drilled through the skull, but the dura was never penetrated. Results at 1 year showed that for the primary endpoint (an ill-advised subjective global rating score) there was not a significant difference between sham and fetal VM transplants. A secondary endpoint was significant, but with a modest motor improvements in ‘‘off ’’ Unified Parkinson’s Disease Rating Scale
(UPDRS) scores (p = 0.01) in patients younger than age 60. The greatest of improvements were in rigidity and bradykinesia. The older patients did not show any significant improvements. There were acute adverse events that were significantly higher in fetal transplanted patients, but not related to the procedure. One patient died due to accidental injury in a storm. Autopsy of the brain in this patient showed survival of ~40,000 transplanted DAergic cells. However, the most surprising finding of the study was the late emergence of unexpected, disabling ‘‘off ’’ period of dyskinesias in 15% of the transplanted patients [32]. These dyskinesias occured in patients with positive Positron emission tomography (PET) scans that indicated growth of the graft in vivo. The mechanism mediating these ‘‘runaway’’ or off state dyskinesias is believed to be the non-homogeneous fiber innervations derived from fetal nigral grafts producing hyperDAergic ‘‘hot spots’’ causing excessive local release of DA (> Figure 192-2). These patients have shown arm and leg dyskinesias that interfere with activities of daily living and walking, as well as facial dystonia. In one patient, symptoms were severe enough to warrant the placement of a feeding tube for nutritional support. In some patients, deep brain stimulation has given symptomatic relief [33]. This NIH study was the first prospective randomized double blind placebo controlled trial for any neurosurgical therapy for PD. The results from this very important study provided valuable objective information on the clinical benefits of fetal transplantation and raised some troubling questions. This study objectively proved that striatal allogenic fetal VM transplantation, in PD patients <60 years of age, improves motor function particularly for dyskinesias and rigidity. These findings validate the results from previously published open label human and the preclinical animal studies, which led up to these clinical trials. The prominent placebo effect on the global rating score of the ‘‘sham’’ surgery patients raises
The future of cell transplantation
192
. Figure 192-2 (a) The photomicrograph represents a high power image of a fetal ventral mesencephalic graft in a PD-like MPTP monkey. (b) Represents a high power image of a fetal ventral mesencephalic graft at autopsy from a human case (courtesy of Dr. Jeffery Kordower). The cytoarchitecture is remarkably similar, in both hosts, the number of cells that survived was small, and reinnervation was limited thereby resulting in ‘‘hot spots’’ and ‘‘cold spots.’’ The immunocytochemistry staining is with the tyrosine hydroxylase antibody
the question whether all surgical interventions for PD should be scrutinized using randomized prospective placebo controlled clinical trials. The second NIH funded placebo controlled collaborative clinical trial, involving teams of investigators from the University of South Florida, Mount Sinai Medical Center, University of British Columbia, and Rush University Medical Center, was different in design from the previous study and paralleled that which was utilized in the successful animal models [34]. Thirtyfour patients with severe PD from the ages of 30 to 75 were randomized to receive bilateral grafts with one or four donors per side in the post-commissural putamen or a sham surgery. Patients receiving transplants had two surgical
procedures separated by 1 week. Sham-treated patients received burr holes that only partially penetrated the skull [34]. Unlike the method used by Freed et al. [32], tissue used for transplantation was stored for a maximum of 2 days before transplantation. Patients were treated with cyclosporine for 2 weeks before the surgical procedure and for 6 months thereafter. Transplanted patients exhibited increased striatal fluorodopa uptake, suggestive of surviving DA neurons. The primary endpoint of the study was a change in the UPDRS scores during off phases. Unfortunately, results of the study revealed no difference in the primary endpoint measure between placebo and transplant groups. The patients in the four-donor group had a tendency, albeit not statistically significant, to have
3165
3166
192
The future of cell transplantation
a greater symptomatic improvement than patients in the one-donor group. Perhaps in a manner analogous to the Freed study, patients with milder disease (although not necessarily younger) had a significant treatment effect compared to shamtreated patients (p = 0.006). Like the first study there was the late emergence of unexpected, disabling ‘‘off’’ period dyskinesias in the transplanted patients. After the surgery, a large proportion of grafted patients (56%) developed dyskinesias during practically defined off periods despite amelioration of dyskinesias during ‘‘on’’ periods in many patients. The mechanisms mediating the postoperative worsening in some aspects of dyskinesias are unclear. There is reason to suspect that a number of factors are causative: graft-derived patterns of re-innervation, immune response to grafted cells, L-dopa priming and others. Clinical trials suggest that fetal nigral transplantations are beneficial for a certain population of PD patients. However, many patients do not show symptomatic improvements after surgery and many develop side-effects, such as debilitating dyskinesias. In addition, the survival of implanted neurons is relatively poor, making it necessary to transplant large amounts of tissue per patient. In an autopsy of a patient 14 years after transplantation, grafted nigral neurons were found to have Lewy body-like inclusions that stained positively for alpha-synuclein and ubiquitin, and to have reduced immunostaining for DA transporter [35]. These pathological changes suggest that PD is an ongoing process that can affect grafted cells in a manner similar to host substantia nigra. Fetal dopaminergic tissue is difficult to obtain due to lack of availability and ethical considerations. Attempts to avoid ethical restrictions in the use of aborted human fetal tissue have lead to an exploration for alternative sources of tissue for grafting, which are discussed later in this chapter. Due to these problems, it is not likely that fetal VM grafting is feasible for generalized therapy to treat PD patients. However, all research performed in the field of nigral
grafting was and continues to be an important learning tool and sets the foundation for future transplantation strategies using other cell-based therapies.
Clinical Outcomes of Non-neural Cells and Alternative Techniques Human retinal pigment epithelial (hRPE) cells are an excellent candidate as an alternative tissue that can be obtained easily in large quantities from the postmortem human eyes and are available from eye and tissue banks. This tissue secretes L-dopa, possibly DA as well, survives well after transplantation, and does not require systemic immunosuppression [36,37]. Human RPE cells are found in the inner layer of the neural retina located between the photoreceptors and the choriocapillaris [38,39]. These epithelial cells contribute an important role in maintaining the blood-retinal barrier, and may have nutritive, phagocytic, and trophic functions endogenously and possibly after transplantation. hRPE cells produce L-dopa, as a precursor to the formation of the characteristic brown-black eumelanin pigment [40,41], and they have been reported to secrete dopamine [41,42], as well as expressing D2 receptors [43]. Another unique characteristic of hRPE cells is while they express the vesicular monoamine transporter (VMAT) they do not express the dopamine transporter (DAT). The expression of the D2 type of DAergic receptors on the surface of hRPE cells probably serves as autoreceptors to regulate the DAergic functioning. The hRPE cells secretion of growth factors has also been characterized and include plateletderived growth factor (PDGF), epidermal growth factor (EGF), fibroblast growth factor (FGF), insulin-like growth factor (IGF), transforming growth factor (TGF), vascular endothelia growth factor (VEGF), pigment epithelial-derived factor (PEDF), and the Fas-ligand [44–46]. Under the FDA’s good manufacturing practices (GMP)
The future of cell transplantation
regulations hRPE cells can be cultured, expanded, stored for an extended period, and extensively tested prior to transplantation. Cells derived from a single donor eye could potentially treat several hundred patients (> Figure 192-3). Accordingly, hRPE cells are an attractive candidate for transplantation as a therapeutic agent to deliver L-dopa and neurotrophic factors. While there are many facets contributing to ocular immune privilege [47], hRPE cells express the cytokine CD95L and release tumor growth factor beta (TGF-b), both of which may create local immunosuppression [48–52]. However, fetal RPE cells do express Major Histocompatibility Complex (MHC) I, many minor histocompatibility antigens [53,54], and when in culture, RPE cells can be induced to express MHC class II [55,56]. Allogeneic neonatal RPE cells can sensitize to the recipient host when grafted into non-immune privileged sites, but not when grafted into immune privileged sites [48]. Thus, systemic immunosuppression does not appear to be necessary. In addition, when hRPE cells are attached to microcarriers, an increase in survival and immune privilege is seen after transplantation.
192
Microcarriers made of gelatin and other materials have traditionally been employed in cell cultures to enhance the cell viability. Human RPE cells are anchorage dependent cells and undergo apoptosis in vivo or in vitro in the absence of a support matrix [57]. Similarly, cells passively attached to biocompatible microcarriers, and when transplanted into the brain of rodents and nonhuman primates display prolonged and enhanced survival in vivo, even in the absence of immunosuppression [42,58,59]. This is known as cell coated microcarrier (CCMTM) technology that appears to be broadly applicable in terms of cell type and the microcarrier composition that may represent a solution for enhancing cell survival [59–61]. This technology is the basis for hRPE cells attaching to the gelatin microcarriers or Spheramine1 (Bayer Health Care) (> Figure 192-4).
Preclinical Study with Microcarriers The preclinical studies were initially performed in rodents using a hemiparkinsonian (HP) model [36,62] for efficiency and safety. These studies were
. Figure 192-3 hRPE cells are found in the inner layer of the neural retina (a) and are readily isolated from eyes obtained from eye/ tissue banks. Under GMP these cells can be grown easily, expanded in culture, stored for prolonged periods, and extensively tested prior to transplantation. Therefore, cells derived from a single donor eye could potentially treat several hundred patients. (b) hRPE cells produce L-dopa as a precursor to the formation of their characteristic brown-black eumelanin pigment as indicated in the immunocytochemistry staining for L-dopa. These cells also secrete a number of growth factors. The hRPE cells are a cell type that is a readily available transplantable source for DA and its precursors, as well as for potential therapeutic neurotrophic factors
3167
3168
192
The future of cell transplantation
. Figure 192-4 The diagram presents the properties of Spheramine. Spheramine consists of hRPE cells that passively attach to a microcarrier support matrix (MSM) composed of gelatin. These cells are readily available, unmodified human cells, which have the advantage of being from a non-fetal tissue source. In addition, with these cells immunosuppression is not required, they secrete cytokines that inhibit rejection, and the microcarrier enhances survival and prevents migration. Spheramine provides a standardized product, produced under GMP conditions that produces L-dopa and is believed to stabilize the delivery of L-dopa to the brain
followed by a blinded placebo controlled nonhuman primate study [63], in which 16 monkeys were administered 1-methyl 4-phenyl 1,2,3, 6-tetrahydropyridine (MPTP) and brought to a stable right HP state for 3 months. Animals were then assessed for their responsiveness to optimal doses of oral combination L-dopa & carbidopa, then randomized into four equally balanced groups to receive either: low dose of Spheramine1 (approximately 12,000 cells/target), high dose of Spheramine1 (approximately 60,000 cells/target), microcarriers only, or sham transplanted injection into the striatum (two caudate targets and three putamen targets). The blinded behavioral assessments of the monkeys at 3 months showed significant improvements of condition (p = 0.01), compared to only 16% change in control animals (microcarriers alone and sham injections). The high dose of Spheramine1 produced robust improvements of 50–60% from baseline in HP monkeys that remained constant throughout the 12-month study, and was significantly different from surgical controls (p < 0.05). The low dose group of Spheramine1 and the
microcarrier only group did not show significant improvements from baseline, nor were there any differences in the low dose group and the sham injection control group, at any of the time points. Histological examination, of the brains, revealed cells consistent with hRPE cells attached to beads at the implantation sites [64]. A minimal inflammatory response was seen in and around the injection tracts at the time the brains were taken. In a different study, 11C-raclopride PET imaging 1 month after transplantation, into bilaterally lesioned MPTP monkeys, revealed decreased binding in the precise location of the implantation site of Spheramine1 .Thus suggesting enhanced dopamine levels [65]. The results from this blind, placebo controlled study indicated that xeno-transplanted human RPE cells with microcarriers improved PD-like behavior and were well tolerated in non-immunosuppressed MPTP treated monkeys up to 12 months. In addition, the results indicate that stereotaxic intrastriatal hRPE microcarrier transplantation may be a potential therapy for advanced PD. Furthermore, the toxicology studies have showed
The future of cell transplantation
no adverse effects with injections of over 450,000 cells with no migration of cells [65].
Clinical Study with Spheramine in Parkinson’s Disease Patients Based on the very promising preclinical data above, the FDA approved an open label pilot clinical trial of intrastriatal transplantation of Spheramine1 in patients with advanced PD [66]. Assessment of the safety and tolerability of Spheramine1 demonstrated that all patients tolerated surgery well and no major adverse events occurred. All six patients showed significant improvements at 12 months post-implant both in the primary outcome measure (p = 0.006) and the UPDRSmotor score in the practically defined off state. These improvements were sustained 24 months after treatment with a 41% average reduction of the disability score (p = 0.03). The overall change in the UPDRS motor off-state score was most
192
robust contralateral to the implanted striatum. The UPDRS total score revealed significant improvements from baseline to 12 months after implantation (p = 0.03), and this was sustained through the 24-month follow-up evaluation (p = 0.03). The on-state time, measured using patient diaries, increased significantly (p = 0.04) from an average of 44% of waking time at baseline to 65% at 24 months after treatment, whereas average off-state time decreased from 41 to 28% during the same period. Noteworthy, the increase of on-state time that occurred in the setting of decreased or unchanged ‘‘on time with dyskinesias’’ (14% at baseline to 15% at 12 months and 7% at 24 months after implantation). Long term follow-up suggests that Spheramine is well tolerated through 4 years postoperatively [67]. Efficacy results indicate continued improvement in UPDRS motor scores off medication of 44% (> Figure 192-5). Secondary efficacy variables also showed improvement at this
. Figure 192-5 In an open-label pilot study, investigators implanted approximately 325,000 RPE cells attached to gelatin microcarriers into the putamen, contralateral to the more affected side of the body in six young patients with advanced idiopathic PD. (a) All six patients showed significant improvements at 12 months post-implant in the primary outcome measure, the UPDRS-motor score in the practically defined off state. This was sustained through 48 months. (a) The mean UPDRS motor scores in off are plotted. (b) The mean percentage of improvement is shown
3169
3170
192
The future of cell transplantation
time. The study has been extended through 10 years for follow-up to capture any delayed effects. A Phase II study was initiated to further evaluate the safety and efficacy of Spheramine implantation (STEPS trial), which is a double-blind sham surgery controlled trial. The primary efficacy data is expected in late 2008. The gold standard for neural transplantation for PD has been fetal tissue. However, the results from clinical trials have not demonstrated dramatic improvements overall, then the questions become why would Spheramine1 be expected to produce improvements and why is there no dyskinesias produced? Unlike fetal tissue, hRPE cells are not neuronal cells and they do not differentiate to form axons or make synaptic connections with the host after transplantation. At first this might suggest a disadvantage, but in fact this may be a very important advantage. The hRPE cells may be acting as DA pumps that selfregulate and do not overly innervate the host tissue. Multiple clinical studies have demonstrated that a continuous supply of DA or its precursors can decrease the motor fluctuations [6,8,9]. The inability of the advanced PD patient to produce, store, and regulate DA is believed to be responsible for motor fluctuations and dyskinesias [68–70]. The hRPE cells may serve to replace these functions and the expression of the D2 receptors on the surface of hRPE cells may serve as autoreceptors to regulate the DAergic function of hRPE cells. Although DA does not diffuse far in the extracellular fluid of normal brain tissue, DA does disperse farther in denervated tissue, and the precursors of DA will spread even farther [3,71]. The decrease in raclopride (an D2 antagonist) binding as determined by PET in the preclinical non-human primate study suggests this degree of diffusion is sufficient to prevent receptor super-sensitivity, thus improving symptoms and lowering adverse side effects of this therapy [65]. The functional improvements shown in the clinical study, along with the lack of new dyskinesias, also
suggest that the patients’ DA function is better regulated after Spheramine implantation.
Stem Cell Therapy for CNS Disorders The above portrayal of the treatment of PD with fetal tissue emphasizes the importance and need to establish stable stem cell lines to be used in cellular transplantation therapies. The most realistic approach for PD would be to culture human stem cell lines that can be directed toward a DAergic phenotype, which then may be immortalized, propagated in vitro, and prepared as needed. In addition to PD, several other diseases or disorders, such as Multiple Sclerosis, stroke, and traumatic brain injury (TBI) would benefit from neural stem cell line(s), but would require multiple different phenotypes. The possibilities of utilizing stem cells for replacing dead or damaged cells is progressing into new realms of potential therapeutics to manage the diseases or disorders of the human central nervous system. Preclinical research over the last few years, has produced models that regenerate tissue and cells through transplantation of stem cells or by their differentiated progeny; that deliver genes and neurotrophic factors to aid in the repair by utilizing stem or neural stem cells as the delivery vehicle; that use encapsulation of xenogenic cells for delivery of bioactive agents while restricting the cytotoxic or immune reactive agents release and thus suppressing the host reaction; use transplanted neural grafts to serve as neuroprotective agents; and lastly by the stimulation of endogenous stem cells and neurotrophic factors to aid in repairing degenerating tissue [72,73]. These areas are continuing to evolve with each new model or application adding to the techniques that are available now to the researchers and in the near future to the physician. However, there are practical issues that need to be addressed and
The future of cell transplantation
resolved with each technique and with the maintaining and expansion of the cell population that will ultimately be used.
Sources of Human Stem Cells, Neural Stem Cells and Alternative Human Cells In addition to the progress of stem cells in general, there have been tremendous advancements in understanding and controlling the culturing, expansion, and directing of human embryonic stem cells (hESCs) to specific cell types (phenotypes). The achievement of creating defined cell populations has led to further development in the testing of pharmacological agents, as well as expansion of well-characterized cell transplantation techniques. We now know hESC are pluripotent – meaning they will give rise to derivatives from each of the three primary germ layers [74], can establish a stem cell line without the use of immortalizing agents, can be propagated as a homogeneous culture, symmetrically amplified for an extended period [75], will remain stable [76], may be cultured for long periods while maintaining the character of the founder cells origin [77], will develop normally when returned to the developing organism [78,79], and when transplanted into an ectopic site will form the appropriate cell type [80]indicating further these cells have the intrinsic property for prolonged self-renewal [81]. In addition, hESC can be genetically manipulated by conventional infection or transfection procedures [82] together with long term modification, such as site directed deletion, point mutation, reporter insertion, and coding sequence replacement [83– 85]. Achievements have also been made in the culturing and expanding hESCs including cultures free of feeder layers and with no animal serum [86] in the media, which are one of pressing importance because of the desire to develop
192
regenerative therapies that are FDA approved (> Figure 192-6). The next most studied cells in regenerative medicine are human NSCs (hNSCs), which are considered multipotent stem cells, and may be cultivated from the brain at the embryonic stage adult stage or derived from hESCs. True NSCs are a population of undifferentiated cells that in the embryonic and adult mammalian reside within the tissue lining of the ventricular system, are selfrenewing, will differentiate into one of the three neural cell types, and can be expanded to produce a clinically significant number of cells for transplantation [2,73,87]. Studies have shown that upon transplantation into the adult CNS the NSCs can integrate seamlessly within the host tissue and differentiate according to the microenvironment [88]. When obtained from the brain during development NSCs are most abundant just after neural tube closure and prior to the onset of neurogenesis. NSC numbers decline over subsequent stages of development [89]. However, throughout development and in the mature brain there are a significant number of NSCs that may be propagated from the cerebral cortex, subependymal zone (SEZ), subgranular layer of the hippocampus and the spinal cord [90–92]. These are typically cultured as monolayers of substrate anchored cells [91,93] or as suspended, spherical structures called neurospheres [94,95]. The motivation in the use of both hESCs and hNSCs for regenerative medicine is the previous use of human fetal ventral mesencephalic tissue (dopaminergic neurons) in PD patients. While there continue to be considerable ethical concerns in using fetal tissue for PD transplantation, there are also ethical concerns in the use of hESCs that are isolated from the inner cell mass of developing blastocysts and of hNSCs derived from fetal tissue or ESCs [96]. The possible alternatives to either fetal tissue or embryonic cells for cell-based therapies are the procurement of hNSCs from the postmortem brain [97] or the precursor of NSCs that may be derived from the pluripotent stem
3171
3172
192
The future of cell transplantation
. Figure 192-6 Microphotographs represent various stages of hESCs while in culture. (a) hESCs showing the formation of a embryoid body (EB) after the fourth passage and 15 days in culture (DIV). The EB is rather large and has been in culture long enough for cells to attach to the plate substrate, migrate away from the EB, and then spontaneously differentiate. (b) Shows a higher magnification (20 PH) of hESCs migrating away from the EB with cells starting to differentiate. (c) The microphotograph is from part of a study on the expansion of hESCs. Initially 10 cells were plated in a 35-mm dish with defined proliferation media, and culture from 1 to 30 DIV. The cells here are at 21 DIV and the dish is full of multiple layers of cells (cells that appear darker) still having the round morphology of early stem cells (5). (d) This microphotograph represents a neurosphere derived from the hESCs after 30 DIV, and culture in neural induction media (low magnification). In this study, the effects of long-term culture on the neurospheres were determined. The spherical shape was lost, cells were differentiating while still attached to the neurosphere, and there were long processes extending from the cells. (e)Shows hNSCs, derived from hESCs, after 45 DIV, in which a network of neural morphological cells have formed with the processes extending and contacting the other cell groups. In all culture conditions discussed here, feeder layers were not used nor were any animal by products such as serum used. Therefore, these particular human ESCs may meet the criteria for stem cell use in humans, in that they are easily cultured using only defined media and the use of human recombinant proteins for neural induction thus, making these hESCs (a gift from William Freed at NIH) an excellent source for neural induced cells
cells of human embryonic carcinoma (EC) [98,99] or embryonic germ (EG) [100], or the hematopoietic stem cells from human bone marrow [101] and human umbilical cord blood [102– 104]. Stem cells from the developing embryonic brain and adult brain from animal species have been isolated to produce NSCs [2,105,106]. However, xenografted neural tissue itself generates a number of practical, ethical, safety, and immunological issues that have to be solved prior to any clinical application. The likelihood of animal cells
being used clinically is minimal and will not be discussed in this chapter. Each cell source has advantages and disadvantages especially when considered for use as a therapeutic treatment.
Practical Issues to Overcome in Cell Transplantation in CNS In the perfect world, there would be the ultimate cell that meets and exceeds the requirements one
The future of cell transplantation
needs for neural transplantation in regenerative medicine. The requirements of this ‘‘prefect cell’’ would be obtained easily with no ethical concerns, maintained and expanded in culture to meet the needs of adult patients while keeping a stable genome, induced in culture or once transplanted into all three neural lineages (neurons, astrocytes, oligodendrocytes), selectively enhanced for particular genotypic to phenotypic properties, survive and integrate into the host brain while not being recognized as foreign to the host (no graft rejection), and have no properties or potential for tumorigenicity. Even with our ‘‘perfect cell,’’ there would be practical issues and limitations that would need to be addressed before being ready for transplantation.
Critical Windows and Cell Age Depending on the source of the cells, there may be critical windows in which to obtain or transplant the cells. When cells are coming from a donor source (not an established cell line, NSC line or hESC), then the donor age must be considered and compared to a developmental chart that is species specific. More specifically, we know that when transplanting grafts of primary CNS tissue, they will only survive well when taken from embryonic donors and not from old donors [107]. We also know there are critical windows during embryonic development, which corresponds to the birth of neurons and neurotransmitters, or when individual populations of cells are born during development [108–110]. Identifying these birthdays can be calculated by mapping the appearance of neurotransmitters or by labeling explicitly dividing cells (tag the DNA) [111]. The donor cells are typically obtained at or around their birthday, because at this time the cells are ideal for transplantation. This is also the time when the cells’ fate are determined, have undergone the final differentiation, are still considered immature, and are still small and round
192
(making cells less susceptible to trauma) and are able to undergo longer periods of anoxia than more mature cells [112]. The more mature cells have neurite and axon growth, are under the cells own ontogenetic programs and are less likely to survive isolation and transplantation. These factors may be due to their susceptibility to axotomy and anoxia in addition to having less plasticity. when more mature cells are taken, e.g., the progenitor cell, they show limited capacity to differentiate when transplanted into a mature brain [2,113]. This suggests that signals during embryonic development are not carried through to the mature brain. During tissue processing for transplantation and the injection procedure into the host brain, fetal neurons are faced with numerous challenges that can negatively influence their survival, such as oxidative stress, ischemia, lack of trophic factors, and host immune rejection of the transplanted cells. As a result, only 5–20% of fetal cell/tissue transplants survive the grafting procedure with most of the cells dying within the first 4 days [114]. The most empirical determinate for the donor source’s critical window is an experiment; however, this is not always practical and the information may already be determined. Previous transplantation studies along with tissue and neuronal differentiation studies, and the brain atlas are valuable sources for birthdays of cell populations and for identifying the critical windows. In relation to fetal tissue, the ‘‘birth’’ of neuron populations are of critical importance for successful transplant treatment [115,116]. Unfortunately, when unfamiliar types of donor cells are used for transplants, there is little or no empirical evidence to indicate at what passage or at what day of neural induction that these cells should be utilized. The only source for information is from previously published studies; however, in some instances the information will be unavailable. In this case, in vitro assays to determine cell cycle, proliferation rate, neural induction and other important properties
3173
3174
192
The future of cell transplantation
will be necessary. When using cultured hESCs or hNSCs often we rely on or account for the number of passages, inherent properties of the cell cycle, and how long the cells have been in culture. The exact age of a cell is almost impossible to identify unless the time of copulation that leads to conception is known. The gestational age of the embryo is not always absolute and usually this is a best estimate measure. There are a few different in vitro assays available for determining the age of cell populations. One such assay is PCR-based called Telomeric Repeat Amplification Protocol (TRAP). Telomeres are nucleoprotein structures that prevent chromosomes from recombining (coming unraveled) by capping the chromosome. The length of the telomeres allows the immature cells to be identified from the mature cells [117]. Another method is real-time polymerase chain reaction (RT-PCR), which allows for the identification of copies of the mitochondrial genome per mitochondrion (mtDNA copy number) [118]. The most critical elements with either hESCs or hNSCs for transplant therapy are that karyotype and epigenetic stability are maintained along with homozygosity in long term cultures and expansion culture. These can be analyzed by single genome-wide single-nucleotide polymorphism analysis, DNA fingerprinting, and comprehensive imprinting analysis [119–123].
Identification of Transplanted Cells Identifying the cells once transplanted is often the most technically challenging area in regenerative medicine. The time to determine how these transplanted cells will be observed and their phenotype identified in the host tissue is before the cells are transplanted. However, these measures involve overcoming technical limitations and there may be extreme difficulties with indisputably identifying the transplanted cells once they integrate into the host brain. Cells that have a gene inserted for a coded protein marker, are pre-labeled with a
tracer, or labeled with a proliferation marker, such as Bromodeoxyuridine (BrdU) may all be down regulated over time as the cells either proliferates or differentiate. This will continue until the markers are diluted beyond detectable levels in the progeny [124–126]. The advent of newer species-specific antibodies that recognize cells which have been cross species transplanted has made the identification of donor cells somewhat easier [125,127]. However, the antibodies must be absolutely species specific. In situ hybridization of probes to specific genes of the grafted cells in the host is a well established and reliable method of identifying the donor cells. A probe is coded for a gene on the donor cells; this overcomes the problem of down regulated protein markers [128] because neither expression of mRNA nor protein markers are required. However, when the gene has a low copy number, this may be difficult if not impossible to visualize due to a low signal and high noise ratio, in which case a probe against repeated sequences would be a better choice. In addition, identifying the differentiated donor cells in the host tissue may require the use of at least two different methods. The combination of fluorescence in situ hybridization (FISH) [129,130] and fluorescent immunohistochemistry can be used to identify the fate of the original transplanted cells [131]. Protocols for both in situ hybridization and immunohistochemistry are readily available in scientific articles, textbooks, manufacturers protocols, and from authentic web sites. In addition, commercially available kits usually contain all necessary reagents for carrying out both immunohistochemistry and in situ hybridization. Often they have excellent protocols that are easy to follow and are tissue/cell specific.
Delivery System(s) One of the most fundamental items to consider in administering biologically active molecules or
The future of cell transplantation
cells to treat a disease or disorder of the CNS is the route of delivery. While systemic delivery is preferred over an intracranial to the specific target area, this method presents problems. The most apparent problems are the inability of the cells to cross the blood brain barrier (BBB), global reactions, and metabolism of molecules by peripheral organs or premature cell death due to a hostile environment thus inducing phagocytosis of the cells. However, direct delivery of either biological agents or cells into the brain also has limitations especially with the need for long-term or chronic delivery. Some other shortcomings with direct delivery include limited diffusions, limited extracellular space, cellular metabolism, and the turnover of cerebral spinal fluid (CSF) along with other substance problems [123,132]. Scientists have developed innovative means to overcome some of these problems. Especially for regenerative medicine’ mini-pumps (implanted or external) allow constant flow of reagents directly to the areas of interest [133,134]. Genetically engineered cells are also appropriate for delivery of therapeutic substance(s) to a well-defined area acting as a pump, and these cells may integrate into the host environment allowing for continued delivery of the biological agents [82,135–137]. The discovery of neural stem cells (NSC) in the subependymal layer [138] within the lining of the ventricular system could allow for the manipulation of these endogenous NSCs or other endogenous factors to assist in recovery [2,139–141]. However, most of these methods require further pre-clinical and clinical studies before they will be approved for use by the FDA.
Immunosuppression and Immune Response Allografts and xenografts can present molecules on the cell surface that may be recognized by the host immune system as foreign to the body after transplantation, causing the rejection of the grafts. Graft rejection can be prevented by the use
192
of immunosuppressive therapy during the few days which the blood brain barrier is open. However, in clinical trials (discussed above) employing allogeneic fetal VM tissue, the duration of immunosuppression has varied extensively, from no immunosuppression therapy use [142] to up to 6 months after transplantations [143]. The immunology of transplantation in the CNS is complicated and rarely studied. Published work suggests that allografts in humans and non-human primates are unlikely to be rejected [144], and there are no recent studies, using newer techniques, in which immune response was the primary focus. However, this is referring to the use of fetal VM tissue and not hESCs or hNSCs. In future and current transplantation studies using hESCs or hNSCs, the inclusion of the immune system response will be beneficial to this entire field of research. A key factor in the recipient’s immune response is the source of the donor cells. Are the grafts are xenogeneic, allogeneic, and autogeneic to that of the host? There have been studies in which the host initially treated with immunosuppressant therapy followed by immunosuppressant withdrawal where no rejection of the grafts were observed [22,143,145–148]. In one study, using fetal DA cells, no immunosuppression therapy was used and rejection of grafted cells was not observed [142]. These findings suggest the use of immunosuppressants, such as cyclosporine-A, may not be necessary under all circumstances. However, in allograft animal models there has been graft rejection, which can cause the production of unwanted cytokines. In addition, as previously discussed, withdrawal of immunosuppressant therapy has resulted in a loss of benefits to the patient [149]. With non-immunosuppressed human recipients there may be concerns for safety and efficacy as this has not been addressed and should be carefully evaluated. To what extent immunosuppressant agents could compromise the health of individual recipients is unknown at this time. In one study, a PD patient developed renal complications because
3175
3176
192
The future of cell transplantation
he could not tolerate cyclosporine after initial transplantation [150]. The autopsy tissue from two PD patients who received multiple allografts with no immunosuppression for the last 12 months before they died [146,147,153] showed immune markers for macrophages, Tand B cells, and microglia within the grafted sites [147]. Although no rejection of the actual grafts were found, it is not clear whether these findings were due to an ineffective immune or nonspecific response from trauma of the surgical procedure, or an early sign of possible graft rejection [150]. The issue of non-use of immunosuppressant theraphy may not be critical unless the drug is also serving in some protective capacity [151,152]. The most widely used immunosuppressant, cyclosporine-A, may have beneficial effects. When cyclosporine-A is administered both the inflammatory reaction and autoimmune mechanisms are suppressed. Both immune and inflammatory reactions have been shown to contribute to the progression of diseases like PD and HD [154,155]. The use of cyclosporine-A in animal models resulted in an increase in xenograft survival [156,157]. Human fetal VM cells showed a six fold increase in the number of surviving DA neurons when grafted into immunosuppressed rats [158]. Other studies have indicated immunosuppressants may have neurotrophic and neuroprotective properties [156,157,159]. Several studies have observed an increase in pro-inflammatory cytokines in both animal models of PD [160,161] and in human brains of PD patients studied upon autopsy [162]. Pro-inflammatory cytokines can be harmful and may add to the progression of PD [163]. These cytokines have been shown to be decreased in animal models of PD, in which the animals were treated with cyclosporine. Therefore, immunosuppressant adjunct therapy may not only decrease the possibility of graft rejection; this treatment may be beneficial in other indications to the patient. Another concern revolves around the likelihood that hESCs will form teratomas once
implanted into the host tissues. Although these cells have been shown to proliferate and differentiate upon transplantation into the appropriate phenotype cell, such as a DA cell in the PD animal model, they have also been shown to develop rapidly into teratomas upon transplantation [164], thus retaining their mitotic ability after transplantation. Consequently, choosing stem cells that have been further differentiated in vitro prior to transplantation (in order for the graft to contain only tissue-specific precursor cells) may be the better choice. Any cell that retains its mitotic ability after transplantation should be used with caution and fully investigated before human studies are performed.
Functional Assessment True functional recovery assessment for any disease depends not only on clinical rating scales, but also on neurophysiological measures that may be correlated to symptomatic recovery. Additionally, in order for comparisons to be made across clinical centers (trials/outcomes) a unified clinical rating scale for each disorder or disease as well as standardization of neurophysiological measures must be employed. Without these parameters in place, the ability to analyze data accumulated from different centers is difficult, if not impossible, and this lack of concession hinders the progression of the neural transplantation field. In the case of PD, several standardized rating scales have been used to evaluate the results of grafts in patients, including the UPDRS and the Schwab-England Disability Scores, as well as the Core Assessment Program for Intracerebral Transplantation (CAPIT), which incorporates the UPDRS. However, until recently there has been no agreement between centers on which test should be used. The CAPIT was developed to encourage uniformity and to standardize the rating scale to allow for congruent comparisons across clinical transplant studies and to serve as a
The future of cell transplantation
registry [150]. The CAPIT includes patients’ inclusion criteria, rating scales for motor and dyskinesia, and testing time to be used for motor behaviors in relation to pharmacological challenge, brain imaging, and for baseline status and post-graft effects over time. This type of assessment program would allow for both intrapatient and inter-patient study design. Unfortunately, consensus has yet to be reached among clinical centers for the type of assessment program on all neural transplantation protocols. This is most likely due to the laborious nature and cost of such programs. A surrogate marker for graft survival is striatal F-DOPA uptake measured by PET. Both striatal F-DOPA uptake and cerebral blood flow of the supplementary motor area (SMA) and of the dorso-lateral prefrontal cortex (DLPFC) can be measured and correlated to the rating scale functional recovery. PET scans have demonstrated an increase in F-DOPA occurs earlier than an increase in blood flow to the cerebral area after transplantation of fetal dopaminergic neurons in PD patients. This indicates that grafted DA neurons need to be integrated fully into these areas and have established efferent and afferent connections [24,165]. However, whether the increase in F-DOPA is due to dopaminergic neurons of the graft or terminal sprouting of host DA neurons cannot be determined by PET analysis [145,166,167]. The use of numerous measures allows for a clearer understanding of the patient’s progress, the progression of treatment over time, and should be employed whenever possible. Unlike the pharmaceutical trials, in which the regulation of new medicine is extremely strict, the area of neural transplantation is less regulated. Besides the CAPIT-PD, CAPIT for HD and the updated CAPSIT-PD, there are no approved assessment programs for other human transplantation studies. Although open-trial analysis allows for quick access to critical information, if the data obtained is not interpretable,
192
then there is information and vital knowledge is wasted. It is the belief of these authors that a program such as CAPIT would serve a useful function and that no transplantation study should be performed without the forethought of obtaining an interpretable data system and allowing for possible comparison across studies.
Future Applications of Stem Cells and Transplantation in CNS Engineered Stem Cell The potential of grafted cells working as vectors to deliver protein (or other molecules) may be more easily realized if one applies a plausible circumstance. For instance, using engineered cells to produce and release molecules with antitumor actions once transplanted into the targeted area of the brain has been shown in pre-clinical models [168], but has not been tested clinically. Before clinical trials may occur, gene therapy needs to be proven safe and innocuous. This includes testing to ensure other brain functions and areas are unimpaired with no systemic or local side effects. In the late 1980s and early 1990s fibroblast cells were initially examined for ex vivo gene therapy studies. In landmark papers by Rosenberg et al. [169] and Fisher et al. [170], fibroblast cells were genetically engineered to secrete NGF [169] and L-dopa [170], respectively, once transplanted into the striatum. While both of these studies are important because they establish ‘‘proof of principle,’’ there are inherent problems. The engineered fibroblasts once transplanted did not migrate, thus the distribution of the trophic factor was limited. Further, because there was no immunosuppression therapy, the cells were recognized as foreign to the host’s brain tissue, leading to an inflammatory response by the host system. While the inflammatory response can be overcome by encapsulating the cells, there is still
3177
3178
192
The future of cell transplantation
limited distribution of the therapeutic agent and no chance of cellular or anatomical correction to the disease. Perhaps cells of neuronal origin would be more efficient in treating CNS diseases/disorders. In fact, both engineered astrocytes and NSCs have been shown to graft well into the host, with normal inflammatory levels, and with a wider and more stable distribution of the therapeutic agent [171]. In this context, NSCs are considered an attractive choice for therapeutic applications to the damaged brain because of their important attributes for gene therapy. These cells are capable of extensive in vitro expansion and, in some cases, a particular tropism toward pathological brain areas. In vitro models for gene therapy are extremely useful when first identifying the pathology of a disease, screening for a therapeutic compound, or for the genetic defect underlying a disease. After the genetic defect is determined, a transgenic model can reproduce this disease, allowing further study. Development of an in vitro model should select the gene and neuronal cells closest to representing the disease. One way of accomplishing this is by selecting primary cells from relevant area(s) of the brain. However, inherent problems of primary neuronal cultures include the quick loss of tissue characteristics, loss of neurons, and difficulty delivering a transgene(s) to the culture in a stable manner. These difficulties may be overcome with the use of custom NSC lines that are isolated from select areas of the cerebrum, and with the neurospheres of these cultures exposed to the retroviral vectors that carry the pathological transgene. An in vitro model may also be isolated from transgenic animals that carry the pathology of the disease. This method may be preferred because there is no separate engineering of the cells and the isolated cells are of a stable phenotype that expresses the disease. Clearly, the challenges for future clinical development of utilizing NSCs with gene therapy is to define the most appropriate NSCs for a given application along with what genes or
molecules can be delivered, and what diseases are suitable targets. Depending on the disease, the NSC lines could be either homogeneous or heterogeneous, but in either case, they should be well characterized in terms of their in vitro stability and grafting capacities.
Conclusion In summary, there are a multitude of problems that may occur when using human fetal tissues, mostly because each is unique. While some of these problems may be overcome, attempts to standardize studies will face drawbacks linked to the use of fetal VM tissue such as: (1) lack of sufficient amounts of tissue for transplantation in a large number of patients, (2) variable age, consistency and viability that results in an inconsistency in functional outcomes with some patients showing major improvement and others modest if any clinical benefit, and (3) occurrence of troublesome dyskinesias in a significant proportion of patients after transplantation. The intrinsic problems with the fetal VM tissue and the outcomes of the clinical trials has motivated researchers to find more reliable cell sources with less inconsistencies, better graft survival, and increased ability to re-innervate the host to prevent dyskinesias [22,23,142–144,172]. The key points to remember are the techniques and tissues used for grafting will need to be tested in the appropriate model that reflects the clinical conditions. Thus, the MPTP model alone is inadequate as it doesn’t reflect the state of the transplant population, which will have L-dopa induced dyskinesias and therefore the MPTP dyskinetic model should be used. Clinical trials should not be initiated until the evident problems from the fetal VM transplants are resolved. These include improved cell survival, greater reinnervation of the host tissue, and absence of graft-induced dyskinesias. The clinical studies should be required to examine patients with the
The future of cell transplantation
best available instruments for evaluation of PD, and these methods should be standardized so that other physicians and researchers can follow the same protocol. Finally, the ultimate investigation requirement is a double blinded, placebo controlled, collaborative clinical trial. The past decade has shown that hESCs and hNSCs can be isolated, immortalized, and used for cellular repair in animal models of injury/disease. The problems that remain to be solved are identifying the best source of cells, determining the optimal route of delivery of the cells, and the best method to promote the innervation of cells within the host to help ensure functional recovery. Each disease may require individualized therapeutic considerations. Thus, oligodendrocyte progenitors may treat myelin disorders, but a multi-progenitor or mixed individual progenitor pool may be needed for a diffuse phenotype disorder such as trauma. Therefore, cells for both site-specific and diffuse repair are required. This can be accomplished by anatomical integration with cell replacement by the grafted cells or through induction of host repair mechanisms to provide structural rebuilding of the damaged neural architecture. Alternatively, the cell may be pharmacologically integrated (minipump) releasing enzymatic, trophic cytokine, angiogenic or other factors that alter the host or host environment to induce repair. This may require genetic engineering of the cells; and the combination of stem cells with gene therapy is most likely treatment choice in the near future: Parkinson’s disease [72,173], Huntington’s disease [174,175], stroke [176,177], and spinal cord injury [132,178,179] may require individualized special cells, multiple cell types, the combination of stem cells with gene therapy, or of the combination of technologies. The urgency to find therapies for neurodegenerative disease must be tapered with the practice of good science. Hopefully, we will apply the knowledge obtained from previous clinical transplantation trials to the utilization of hESCs and hNSCs in new treatments for patients with PD and other devastating diseases.
192
Acknowledgements We would like to give our deepest appreciation to Mr. Leo Kelly for his thoughtful and critical help in preparing this manuscript.
References 1. Gage FH. Neurogenesis in the adult brain. J Neurosci 2002;22(3):612-3. 2. Gage FH. Mammalian neural stem cells. Science 2000;287(5457):1433-8. 3. Hornykiewicz O, Kish SJ. Biochemical pathophysiology of Parkinson’s disease. Adv Neurol 1987;45:19-34. 4. Wichmann T, Delong MR, Vitek JL, editors. Pathophysiological considerations in basal ganglia surgery: role of the basal ganglia in hypokinetic and hyperkinetic movement disorders progress. In: Lunsford LD, editor. Neurological surgery (Movement disorder surgery), vol. 15. New York: Karger; 2002. P. 31-57. 5. Jankovic J, Marsden CD. Therapeutic strategies in Parkinson’s disease. In: Jankovic J, Tolosa E, editors. Parkinson’s disease and movement disorders. 4th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2002. p. 116-51. 6. Chase TN, Engber TM, Mouradian MM. Palliative and prophylactic benefits of continuously administered dopaminomimetics in Parkinson’s disease. Neurology 1994;44 (7 Suppl 6):S15-8. 7. Marsden CD. Problems with long-term levodopa therapy for Parkinson’s disease. Clin Neuropharmacol 1994;17 Suppl 2:S32-44. 8. Chase TN, et al. Fluctuation in response to chronic levodopa therapy: pathogenetic and therapeutic considerations. Adv Neurol 1987;45:477-80. 9. Fabbrini G, et al. Levodopa pharmacokinetic mechanisms and motor fluctuations in Parkinson’s disease. Ann Neurol 1987;21(4):370-6. 10. Backlund EO, et al. Transplantation of adrenal medullary tissue to striatum in parkinsonism. First clinical trials. J Neurosurg 1985;62(2):169-73. 11. Goetz CG, et al. United Parkinson Foundation Neurotransplantation Registry on adrenal medullary transplants: presurgical, and 1- and 2-year follow-up. Neurology 1991;41(11):1719-22. 12. Bakay RAE. Neurotransplantation and protective therapy for movement disorders. Semin Neurosurg 2001;12(2): 245-52. 13. Kordower JH, et al. Putative chromaffin cell survival and enhanced host-derived TH-fiber innervation following a functional adrenal medulla autograft for Parkinson’s disease. Ann Neurol 1991;29(4):405-12.
3179
3180
192
The future of cell transplantation
14. Bakay RA, et al. Preliminary report on adrenal medullary grafting from the American Association of Neurological Surgeons Graft Project. Prog Brain Res 1990;82:603-10. 15. Freed WJ, Poltorak M, Becker JB. Intracerebral adrenal medulla grafts: a review. Exp Neurol 1990;110(2):139-66. 16. Bakay RAE, Kordower JH, Starr P, editors. Restorative surgical therapies for Parkinson’s disease. In: Tuszynski M, Kordower J, editors. CNS regeneration: basic science and clinical applications. San Diego, CA: Academic Press; 1999. p. 389-418. 17. Freed CR, Breeze RE, Fahn S. Placebo surgery in trials of therapy for Parkinson’s disease. N Engl J Med 2000;342 (5):353-4; author reply 354–5. 18. Freeman T.B, et al. Use of placebo surgery in controlled trials of a cellular-based therapy for Parkinson’s disease. N Engl J Med 1999;341(13):988-92. 19. Horisberger JD. Placebo surgery in trials of therapy for Parkinson’s disease. N Engl J Med 2000;342(5):354-5. 20. Slattery J. Placebo surgery in trials of therapy for Parkinson’s disease. N Engl J Med 2000;342(5):354; author reply 355. 21. Weiner WJ. Placebo surgery in trials of therapy for Parkinson’s disease. N Engl J Med 2000;342(5):353; author reply 354–5. 22. Spencer DD, et al. Unilateral transplantation of human fetal mesencephalic tissue into the caudate nucleus of patients with Parkinson’s disease. N Engl J Med 1992; 327(22):1541-8. 23. Henderson BT, et al. Implantation of human fetal ventral mesencephalon to the right caudate nucleus in advanced Parkinson’s disease. Arch Neurol 1991;48(8):822-7. 24. Lindvall O. Update on fetal transplantation: the Swedish experience. Mov Disord 1998;13 Suppl 1:83-7. 25. Widner H, et al. Bilateral fetal mesencephalic grafting in two patients with parkinsonism induced by 1-methyl-4phenyl-1,2,3,6-tetrahydropyridine (MPTP). N Engl J Med 1992;327(22):1556-63. 26. Kopyov OV, et al. Clinical study of fetal mesencephalic intracerebral transplants for the treatment of Parkinson’s disease. Cell Transplant 1996;5(2):327-37. 27. Lopez-Lozano JJ, et al. Long-term improvement in patients with severe Parkinson’s disease after implantation of fetal ventral mesencephalic tissue in a cavity of the caudate nucleus: 5-year follow up in 10 patients. Clinica Puerta de Hierro Neural Transplantation Group. J Neurosurg 1997;86(6):931-42. 28. Peschanski M, et al. Bilateral motor improvement and alteration of L-dopa effect in two patients with Parkinson’s disease following intrastriatal transplantation of foetal ventral mesencephalon. Brain 1994;117 (Pt 3):487-99. 29. Mendez I, et al. Simultaneous intrastriatal and intranigral fetal dopaminergic grafts in patients with Parkinson disease: a pilot study. Report of three cases. J Neurosurg 2002;96(3):589-96.
30. Hallett M, Litvan I. Evaluation of surgery for Parkinson’s disease: a report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology. The Task Force on Surgery for Parkinson’s Disease. Neurology 1999;53(9):1910-21. 31. Fahn S, Greene PE, Tsai WY, Eidelberg D, Winfield H, Dillon S. Double-blind controlled trial of human embryonic dopaminergic tissue transplants in advanced Parkinson’s disease; clinical outcomes. Neurology 1999;5(A405). 32. Freed CR, et al. Transplantation of embryonic dopamine neurons for severe Parkinson’s disease. N Engl J Med 2001;344(10):710-9. 33. Bakay RA. Is transplantation to treat Parkinson’s disease dead? Neurosurgery 2001;49(3):576-80. 34. Olanow CW, et al. A double-blind controlled trial of bilateral fetal nigral transplantation in Parkinson’s disease. Ann Neurol 2003;54(3):403-14. 35. Kordower JH, et al. Lewy body-like pathology in longterm embryonic nigral transplants in Parkinson’s disease. Nat Med 2008;14(5):504-6. 36. Subramanian T, et al. Striatal xenotransplantation of human retinal pigment epithelial cells attached to microcarriers in hemiparkinsonian rats ameliorates behavioral deficits without provoking a host immune response. Cell Transplant 2002;11(3):207-14. 37. Pfeffer B. Improved methodology for cell culture of human and monkey retinal pigment epithelium. Prog Retina Res 1991;10:251-91. 38. Marmor MF, Wolfensberger TJ. The Retinal pigment epithelium: function and disease. New York, NY: Oxford University Press; 1998. p. 521. 39. Schraermeyer U, Heimann K. Current understanding on the role of retinal pigment epithelium and its pigmentation. Pigment Cell Res 1999;12(4):219-36. 40. Pawelek JM, Korner AM. The biosynthesis of mammalian melanin. Am Sci 1982;70(2):136-45. 41. Boulton M. Melanin and the retinal pigment epithelium. In: Marmor MF, Wolfensberger TJ, editors. The retinal pigment epithelium: function and disease. New York, NY: Oxford University Press; 1998. p. 65-78. 42. Cherksey BD. Method of increasing viability of cells which are administered to the brain or spinal cord, U.S Patent 618,531. 1997. 43. Burnside B, Bost-Usinger L. The retinal pigment epithelial cytoskeleton. In: Marmor MF, Wolfensberger TJ, editors. The retinal pigment epithelium: function and disease. New York, NY: Oxford University Press; 1998. p. 459-77. 44. Campochiaro PA. Growth factors in the retinal pigment epithelium and retina. In: Marmor MF, Wolfensberger TJ, editors. The retinal pigment epithelium: function and disease. New York, NY: Oxford University Press; 1998. p. 478-85.
The future of cell transplantation
45. Tombran-Tink J, Chader GG, Johnson LV. PEDF: a pigment epithelium-derived factor with potent neuronal differentiative activity. Exp Eye Res 1991;53(3):411-4. 46. Jorgensen A, et al. Human retinal pigment epithelial cellinduced apoptosis in activated T cells. Invest Ophthalmol Vis Sci 1998;39(9):1590-9. 47. Wenkel H, Streilein JW. Analysis of immune deviation elicited by antigens injected into the subretinal space. Invest Ophthalmol Vis Sci 1998;39(10):1823-34. 48. Wenkel H, Streilein JW. Evidence that retinal pigment epithelium functions as an immune-privileged tissue. Invest Ophthalmol Vis Sci 2000;41(11):3467-73. 49. Liversidege J, Forrester JV. Regulation of immune responses by the retinal pigment epithelium. In: Marmor MF, Wolfensberger TJ, editors. The retinal pigment epithelium: function and disease. New York, NY: Oxford University Press; 1998. p. 511-27. 50. Tanihara H, et al. Identification of transforming growth factor-beta expressed in cultured human retinal pigment epithelial cells. Invest Ophthalmol Vis Sci 1993;34 (2):413-9. 51. Holtkamp GM, et al. Expression of multiple forms of IL-1 receptor antagonist (IL-1ra) by human retinal pigment epithelial cells: identification of a new IL-1ra exon. Eur J Immunol 1999;29(1):215-24. 52. Campochiaro PA. Cytokine production by retinal pigmented epithelial cells. Int Rev Cytol 1993;146:75-82. 53. Dutt K, et al. In vitro phenotypic and functional characterization of human pigment epithelial cell lines. Curr Eye Res 1989;8(5):435-40. 54. Rezai KA, et al. The immunogenic potential of human fetal retinal pigment epithelium and its relation to transplantation. Invest Ophthalmol Vis Sci 1997;38(12):2662-71. 55. Percopo CM, et al. Cytokine-mediated activation of a neuronal retinal resident cell provokes antigen presentation. J Immunol 1990;145(12):4101-7. 56. Liversidge J, et al. Lymphokine-induced MHC class II antigen expression on cultured retinal pigment epithelial cells and the influence of cyclosporin A. Immunology 1988;63(2):313-7. 57. Tezel TH, Del Priore LV. Reattachment to a substrate prevents apoptosis of human retinal pigment epithelium. Graefes Arch Clin Exp Ophthalmol 1997;235(1):41-7. 58. Cherksey BD. Method for transplanting cells into the brain and therapeutic uses therefore, US Patent 5750103. 2000. 59. Cherksey BD, Sapirstein VS, Geraci AL. Adrenal chromaffin cells on microcarriers exhibit enhanced long-term functional effects when implanted into the mammalian brain. Neuroscience 1996;75(2):657-64. 60. Borlongan CV, Saporta S, Sanberg PR. Intrastriatal transplantation of rat adrenal chromaffin cells seeded on microcarrier beads promote long-term functional recovery in hemiparkinsonian rats. Exp Neurol 1998; 151(2):203-14.
192
61. Saporta S, et al. Microcarrier enhanced survival of human and rat fetal ventral mesencephalon cells implanted in the rat striatum. Cell Transplant 1997;6(6):579-84. 62. Potter BM, Kidwell W, Cornfeldt ML. Functional effects of intrastriatal hRPE grafts in hemiparkinsonian rats is enhanced by adhering to microcarriers. Presented at the 27th Annual Meeting of the Society for Neuroscience, New Orleans, LA, 1994. 63. Subramanian T, Bakay RAE, Cornfeldt ME. Blinded placebo-controlled trial to assess the effects of striatal transplantation of human retinal pigmented epithelial cells attached to microcarriers (hRPE-M) in parkinsonian monkeys. Parkinsonism Rel Dis 1999;5:S111. 64. Watts RL, et al. Stereotaxic intrastriatal implantation of human retinal pigment epithelial (hRPE) cells attached to gelatin microcarriers: a potential new cell therapy for Parkinson’s disease. J Neural Transm Suppl 2003 (65):215–27. 65. Doudet DJ, et al. PET imaging of implanted human retinal pigment epithelial cells in the MPTP-induced primate model of Parkinson’s disease. Exp Neurol 2004;189 (2):361-8. 66. Stover NP, et al. Intrastriatal implantation of human retinal pigment epithelial cells attached to microcarriers in advanced Parkinson disease. Arch Neurol 2005;62 (12):1833-7. 67. Bakay RA, et al. Implantation of Spheramine in advanced Parkinson’s disease (PD). Front Biosci 2004;9:592-602. 68. Fabbrini G, et al. Motor fluctuations in Parkinson’s disease: central pathophysiological mechanisms, Part I. Ann Neurol 1988;24(3):366-71. 69. Colosimo C, et al. Motor response to acute dopaminergic challenge with apomorphine and levodopa in Parkinson’s disease: implications for the pathogenesis of the on-off phenomenon. J Neurol Neurosurg Psychiatry 1996;60(6):634-7. 70. Leenders KL, et al. Inhibition of L-[18F]fluorodopa uptake into human brain by amino acids demonstrated by positron emission tomography. Ann Neurol 1986;20(2):258-62. 71. Hornykiewicz O. Dopamine and Parkinson’s disease. A personal view of the past, the present, and the future. Adv Neurol 2001;86:1-11. 72. Newman MB, Bakay RA. Therapeutic potentials of human embryonic stem cells in Parkinson’s disease. Neurotherapeutics 2008;5(2):237-51. 73. Newman MB, et al. Neural stem cells for cellular therapy in humans. In: Bottenstein JE, editor. Neural stem cells: development and transplantation. New York: Kluwer; 2003. p. 379-412. 74. Clark AT. Establishment and differentiation of human embryonic stem cell derived germ cells. Soc Reprod Fertil Suppl 2007;63:77-86. 75. Moon SY, et al. Generation, culture, and differentiation of human embryonic stem cells for therapeutic applications. Mol Ther 2006;13(1):5-14.
3181
3182
192
The future of cell transplantation
76. Hoffman LM, Carpenter MK. Human embryonic stem cell stability. Stem Cell Rev 2005;1(2):139-44. 77. Hoffman LM, Carpenter MK. Characterization and culture of human embryonic stem cells. Nat Biotechnol 2005;23(6):699-708. 78. Zhang X, et al. Derivation of human embryonic stem cells from developing and arrested embryos. Stem Cells 2006;24(12):2669-76. 79. Goldstein RS. Transplantation of human embryonic stem cells to the chick embryo. Methods Mol Biol 2006; 331:137-51. 80. Draper JS, et al. Culture and characterization of human embryonic stem cells. Stem Cells Dev 2004;13(4):325-36. 81. Plaia TW, et al. Characterization of a new NIH-registered variant human embryonic stem cell line, BG01V: a tool for human embryonic stem cell research. Stem Cells 2006;24 (3):531-46. 82. Eiges R. Genetic manipulation of human embryonic stem cells by transfection. Methods Mol Biol 2006;331:221-39. 83. Davis RP, et al. Targeting a GFP reporter gene to the MIXL1 locus of human embryonic stem cells identifies human primitive streak-like cells and enables isolation of primitive hematopoietic precursors. Blood 2008;111(4):1876-84. 84. Pickering SJ, et al. Generation of a human embryonic stem cell line encoding the cystic fibrosis mutation deltaF508, using preimplantation genetic diagnosis. Reprod Biomed Online 2005;10(3):390-7. 85. Park S, et al. Genetically modified human embryonic stem cells relieve symptomatic motor behavior in a rat model of Parkinson’s disease. Neurosci Lett 2003;353(2):91-4. 86. Klimanskaya I, et al. Human embryonic stem cells derived without feeder cells. Lancet 2005;365(9471):1636-41. 87. Gritti A, Vescovi AL, Galli R. Adult neural stem cells: plasticity and developmental potential. J Physiol Paris 2002;96(1-2):81-90. 88. Flax JD, et al. Engraftable human neural stem cells respond to developmental cues, replace neurons, and express foreign genes. Nat Biotechnol 1998;16(11):1033-9. 89. Zigova T, Snyder EY, Sanberg PR. Neural stem cells for brain repair. Contemporary neuroscience. Totowa, NJ: Humana Press; 2002. 90. Kukekov VG, et al. Multipotent stem/progenitor cells with similar properties arise from two neurogenic regions of adult human brain. Exp Neurol 1999;156(2):333-44. 91. Palmer TD, et al. Fibroblast growth factor-2 activates a latent neurogenic program in neural stem cells from diverse regions of the adult CNS. J Neurosci 1999;19(19):8487-97. 92. Scheffler B, et al. Marrow-mindedness: a perspective on neuropoiesis. Trends Neurosci 1999;22(8):348-57. 93. Richards LJ, Kilpatrick TJ, Bartlett PF. De novo generation of neuronal cells from the adult mouse brain. Proc Natl Acad Sci USA 1992;89(18):8591-5. 94. Reynolds BA, Tetzlaff W, Weiss S. A multipotent EGFresponsive striatal embryonic progenitor cell produces neurons and astrocytes. J Neurosci 1992;12(11):4565-74.
95. Reynolds BA, Weiss S. Generation of neurons and astrocytes from isolated cells of the adult mammalian central nervous system. Science 1992;255(5052):1707-10. 96. Joannides AJ, et al. A scaleable and defined system for generating neural stem cells from human embryonic stem cells. Stem Cells 2007;25(3):731-7. 97. Palmer TD, et al. Cell culture. Progenitor cells from human brain after death. Nature 2001;411 (6833):42-3. 98. Iacovitti L, Stull ND, Jin H. Differentiation of human dopamine neurons from an embryonic carcinomal stem cell line. Brain Res 2001;912(1):99-104. 99. Matin MM, et al. Specific knockdown of Oct4 and beta2microglobulin expression by RNA interference in human embryonic stem cells and embryonic carcinoma cells. Stem Cells 2004;22(5):659-68. 100. Turnpenny L, et al. Derivation of human embryonic germ cells: an alternative source of pluripotent stem cells. Stem Cells 2003;21(5):598-609. 101. Zhao LR, et al. Human bone marrow stem cells exhibit neural phenotypes and ameliorate neurological deficits after grafting into the ischemic brain of rats. Exp Neurol 2002;174(1):11-20. 102. Ha Y, et al. Neural phenotype expression of cultured human cord blood cells in vitro. Neuroreport 2001;12 (16):3523-7. 103. Newman MB, et al. Transplantation of human umbilical cord blood cells in the repair of CNS diseases. Expert Opin Biol Ther 2004;4(2):121-30. 104. Newman MB, et al. Human umbilical cord blood (HUCB) cells for central nervous system repair. Neurotox Res 2003;5(5):355-68. 105. Palmer TD, Takahashi J, Gage FH. The adult rat hippocampus contains primordial neural stem cells. Mol Cell Neurosci 1997;8(6):389-404. 106. Tropepe V, et al. Distinct neural stem cells proliferate in response to EGF and FGF in the developing mouse telencephalon. Dev Biol 1999;208(1):166-88. 107. Olson TA, et al. Newborn polymorphonuclear leukocyte aggregation: a study of physical properties and ultrastructure using chemotactic peptides. Pediatr Res 1983; 17(12):993-7. 108. Martens DJ, Tropepe V, van Der Kooy D. Separate proliferation kinetics of fibroblast growth factor-responsive and epidermal growth factor-responsive neural stem cells within the embryonic forebrain germinal zone. J Neurosci 2000;20(3):1085-95. 109. Song HJ, Stevens CF, Gage FH. Neural stem cells from adult hippocampus develop essential properties of functional CNS neurons. Nat Neurosci 2002;5 (5):438-45. 110. Sommer L, Rao M. Neural stem cells and regulation of cell number. Prog Neurobiol 2002;66(1):1-18. 111. Shin S, et al. Whole genome analysis of human neural stem cells derived from embryonic stem cells and stem
The future of cell transplantation
112.
113.
114.
115.
116.
117.
118.
119. 120.
121.
122.
123.
124. 125.
126.
127.
128.
and progenitor cells isolated from fetal tissue. Stem Cells 2007;25(5):1298-306. Ourednik V, et al. Segregation of human neural stem cells in the developing primate forebrain. Science 2001;293(5536):1820-4. Joannides AJ, et al. Environmental signals regulate lineage choice and temporal maturation of neural stem cells from human embryonic stem cells. Brain 2007;130 (Pt 5):1263-75. Sortwell CE, Pitzer MR, Collier TJ. Time course of apoptotic cell death within mesencephalic cell suspension grafts: implications for improving grafted dopamine neuron survival. Exp Neurol 2000;165(2):268-77. Freeman TB, et al. The influence of donor age on the survival of solid and suspension intraparenchymal human embryonic nigral grafts. Cell Transplant 1995; 4(1):141-54. Freeman TB, Sanberg PR, Isacson O. Development of the human striatum: implications for fetal striatal transplantation in the treatment of Huntington’s disease. Cell Transplant 1995;4(6):539-45. Saretzki G, et al. Down-regulation of multiple stress defence mechanisms during differentiation of human embryonic stem cells. Stem Cells 2008;26(2):455-64. St John JC, et al. The analysis of mitochondria and mitochondrial DNA in human embryonic stem cells. Methods Mol Biol 2006;331:347-74. Ellerstrom C, et al. Derivation of a xeno-free human embryonic stem cell line. Stem Cells 2006;24(10):2170-6. Li SS, et al. Characterization and gene expression profiling of five new human embryonic stem cell lines derived in Taiwan. Stem Cells Dev 2006;15(4):532-55. Stephenson EL, Braude PR, Mason C. International community consensus standard for reporting derivation of human embryonic stem cell lines. Regen Med 2007;2 (4):349-62. Kim KP, et al. Gene-specific vulnerability to imprinting variability in human embryonic stem cell lines. Genome Res 2007;17(12):1731-42. Zeng X, Rao MS. Human embryonic stem cells: long term stability, absence of senescence and a potential cell source for neural replacement. Neuroscience 2007;145(4):1348-58. Cattaneo E, McKay R. Identifying and manipulating neuronal stem cells. Trends Neurosci 1991;14(8):338-40. Fenderson BA, et al. Staining embryonic stem cells using monoclonal antibodies to stage-specific embryonic antigens. Methods Mol Biol 2006;325:207-24. Pruszak J, et al. Markers and methods for cell sorting of human embryonic stem cell-derived neural cell populations. Stem Cells 2007;25(9):2257-68. Poltavtseva RA, et al. Evaluation of progenitor cell cultures from human embryos for neurotransplantation. Brain Res Dev Brain Res 2002;134(1–2):149-54. Seigel GM, et al. Human embryonic and neuronal stem cell markers in retinoblastoma. Mol Vis 2007;13:823-32.
192
129. Caisander G, et al. Chromosomal integrity maintained in five human embryonic stem cell lines after prolonged in vitro culture. Chromosome Res 2006;14(2):131-7. 130. Hanson C, Caisander G. Human embryonic stem cells and chromosome stability. Apmis 2005;113 (11–12):751-5. 131. Grinnemo KH, et al. Human embryonic stem cells are immunogenic in allogeneic and xenogeneic settings. Reprod Biomed Online 2006;13(5):712-24. 132. Bambakidis NC, et al. Stem cell biology and its therapeutic applications in the setting of spinal cord injury. Neurosurg Focus 2008;24(3-4):E20. 133. Newman MB, Shytle RD, Sanberg PR. Locomotor behavioral effects of prenatal and postnatal nicotine exposure in rat offspring. Behav Pharmacol 1999;10 (6-7):699-706. 134. Harbaugh RE, et al. Intracerebroventricular bethanechol chloride infusion in Alzheimer’s disease. Results of a collaborative double-blind study. J Neurosurg 1989;71(4):481-6. 135. Bankiewicz KS, et al. Practical aspects of the development of ex vivo and in vivo gene therapy for Parkinson’s disease. Exp Neurol 1997;144(1):147-56. 136. Gropp M, Reubinoff B. Lentiviral vector-mediated gene delivery into human embryonic stem cells. Methods Enzymol 2006;420:64-81. 137. Kim JH, et al. Efficient gene delivery in differentiated human embryonic stem cells. Exp Mol Med 2005;37 (1):36-44. 138. Altman J, Bayer SA. Atlas of prenatal rat brain development. Boca Raton, FL: CRC Press; 1995. p. 589. 139. Redmond DE Jr, et al. Behavioral improvement in a primate Parkinson’s model is associated with multiple homeostatic effects of human neural stem cells. Proc Natl Acad Sci USA 2007;104(29):12175-80. 140. Craig CG, et al. In vivo growth factor expansion of endogenous subependymal neural precursor cell populations in the adult mouse brain. J Neurosci 1996;16(8):2649-58. 141. Kuhn HG, et al. Epidermal growth factor and fibroblast growth factor-2 have different effects on neural progenitors in the adult rat brain. J Neurosci 1997;17 (15):5820-9. 142. Freed CR, et al. Survival of implanted fetal dopamine cells and neurologic improvement 12 to 46 months after transplantation for Parkinson’s disease. N Engl J Med 1992;327(22):1549-55. 143. Hauser RA, et al. Long-term evaluation of bilateral fetal nigral transplantation in Parkinson disease. Arch Neurol 1999;56(2):179-87. 144. Freed CR, et al. Fetal neural transplantation for Parkinson’s disease. In: Rich RR, editor. Clinical immunology: principles and practice. St. Louis: Mosby; 1995. p. 1677-87. 145. Kordower JH, et al. Neuropathological evidence of graft survival and striatal reinnervation after the transplantation of fetal mesencephalic tissue in a patient with
3183
3184
192 146.
147. 148.
149.
150.
151.
152.
153.
154.
155. 156.
157.
158.
159.
160.
161.
162.
The future of cell transplantation
Parkinson’s disease [see comments]. N Engl J Med 1995;332(17):1118-24. Kordower JH, et al. Dopaminergic transplants in patients with Parkinson’s disease: neuroanatomical correlates of clinical recovery. Exp Neurol 1997;144(1):41-6. Kordower JH, et al. Treatment with fetal allografts. Neurology 1997;48(6):1737-8. Freeman TB, et al. Transplantation of human fetal striatal tissue in Huntington’s disease: rationale for clinical studies. Novartis Found Symp 2000;231:129-38; discussion 139-47. Freeman TB, et al. Double blind controlled trial of bilateral fetal nigral transplantation in Parkinson’s disease. Annuals of Neurology 2004;54(3):403-14. Freeman TB, et al. Human fetal tissue transplantation. In: Germano IM, editor. Neurosurgical treatment of movement disorders. Park Ridge, IL: The American Association of Neurological Surgeons; 199. p. 177-92. Okoye GS, Freed WJ, Geller HM. Short-term immunosuppression enhances the survival of intracerebral grafts of A7-immortalized glial cells. Exp Neurol 1994;128 (2):191-201. Bakay RA, et al. Immunological responses to injury and grafting in the central nervous system of nonhuman primates. Cell Transplant 1998;7(2):109-20. Kordower JH, et al. Functional fetal nigral grafts in a patient with Parkinson’s disease: chemoanatomic, ultrastructural, and metabolic studies. J Comp Neurol 1996;370(2):203-30. Borlongan CV, Sanberg PR. Neural transplantation for treatment of Parkinson’s disease. Drug Discov Today 2002;7(12):674-82. Borlongan CV, Sanberg PR. Neural transplantation in the new millenium. Cell Transplant 2002;11(6):615-8. Borlongan CV, et al. Cyclosporine A-induced hyperactivity in rats: is it mediated by immunosuppression, neurotrophism, or both? Cell Transplant 1999;8(1):153-9. Borlongan CV, et al. CNS immunological modulation of neural graft rejection and survival. Neurol Res 1996;18 (4):297-304. Brundin P, et al. Human fetal dopamine neurons grafted in a rat model of Parkinson’s disease: immunological aspects, spontaneous and drug-induced behaviour, and dopamine release. Exp Brain Res 1988;70 (1):192-208. Borlongan CV, Sanberg PR, Freeman TB. Neural transplantation for neurodegenerative disorders. Lancet 1999;353 Suppl 1:SI29-30. Nagatsu T, Sawada M. Inflammatory process in Parkinson’s disease: role for cytokines. Curr Pharm Des 2005;11 (8):999-1016. Nagatsu T, et al. Changes in cytokines and neurotrophins in Parkinson’s disease. J Neural Transm Suppl 2000(60):277-90. Sawada M, Imamura K, Nagatsu T. Role of cytokines in inflammatory process in Parkinson’s disease. J Neural Transm Suppl 2006(70):373-81.
163. Carvey PM, Punati A, Newman MB. Progressive dopamine neuron loss in Parkinson’s disease: the multiple hit hypothesis. Cell Transplant 2006;15(3):239-50. 164. Bjorklund LM, et al. Embryonic stem cells develop into functional dopaminergic neurons after transplantation in a Parkinson rat model. Proc Natl Acad Sci USA 2002;99(4):2344-9. 165. Lindvall O, Hagell P. Cell replacement therapy in human neurodegenerative disorders. Clin Neurosci Res 2002; 2:86-92. 166. Lindvall O, et al. Grafts of fetal dopamine neurons survive and improve motor function in Parkinson’s disease. Science 1990;247(4942):574-7. 167. Borlongan CV. Transplantation therapy for Parkinson’s disease. Expert Opin Investig Drugs 2000;9(10):2319-30. 168. Phillips MI, Tang YL. Genetic modification of stem cells for transplantation. Adv Drug Deliv Rev 2008;60 (2):160-72. 169. Rosenberg MB, et al. Grafting genetically modified cells to the damaged brain: restorative effects of NGF expression. Science 1988;242(4885):1575-8. 170. Fisher LJ, et al. Survival and function of intrastriatally grafted primary fibroblasts genetically modified to produce L-dopa. Neuron 1991;6(3):371-80. 171. Ridet JL, Deglon N, Aebischer P. Gene transfer techniques for the delivery of GDNF in Parkinson’s disease. Novartis Found Symp 2000;231:202-15; discussion 215-9, 302-6. 172. Freed CR, et al. Fetal substantia nigra transplants lead to DA cell replacement and behavioral improvement in bonnet monkeys with MPTP induced parkinsonism. In: Beart PM, Woodruff G, Jackson DM, editors. Pharmacology and functional regulation of dopaminergic neurons. London: Macmillan; 1988. p. 353-60. 173. Hall VJ, Li JY, Brundin P. Restorative cell therapy for Parkinson’s disease: a quest for the perfect cell. Semin Cell Dev Biol 2007;18(6):859-69. 174. Dunnett SB, Rosser AE. Stem cell transplantation for Huntington’s disease. Exp Neurol 2007;203(2):279-92. 175. Lim HC, et al. Neuroprotective effect of neural stem cellconditioned media in in vitro model of Huntington’s disease. Neurosci Lett 2008;435(3):175-80. 176. Levy R, et al. Cortical stimulation for the rehabilitation of patients with hemiparetic stroke: a multicenter feasibility study of safety and efficacy. J Neurosurg 2008;108 (4):707-14. 177. Kondziolka D, Wechsler L. Stroke repair with cell transplantation: neuronal cells, neuroprogenitor cells, and stem cells. Neurosurg Focus 2008;24(3-4):E13. 178. Louro J, Pearse DD. Stem and progenitor cell therapies: recent progress for spinal cord injury repair. Neurol Res 2008;30(1):5-16. 179. Takeuchi H, et al. Intravenously transplanted human neural stem cells migrate to the injured spinal cord in adult mice in an SDF-1- and HGF-dependent manner. Neurosci Lett 2007;426(2):69-74.
The Future of Stereotactic and Functional Neurosurgery
188 The Future of Computers and Imaging B. A. Kall
Historical Perspectives Computer and imaging technologies have had a major impact on the field of stereotactic neurosurgery in the last 30 years. The first known human stereotactic procedure was performed by Spiegel and Wycis in 1947 using manually calculated methods of determining a target for coagulating the dorsal median nucleus of the thalamus [1]. Through the 1960s, many centers performed human stereotactic procedures primarily for the treatment of movement disorders and chronic pain using manual calculations. The advent of L-dopa in 1968 caused a dramatic decrease in stereotactic procedures until the development of computed tomography (CT) in 1971. CT ushered in the era of image-guided stereotactic procedures. The first production class CT scanner, known as the EMI-scanner, acquired two adjacent slices in about 4 min and then was postprocessed on a Data General Nova minicomputer in another 7 min into 80 pixel by 80 pixel slices. Each new generation of CT scanner generated higher resolution three-dimensional images over a larger field of view in a much shorter amount of time. With the advent of CT, a small number of centers began developing computer software and systems to automate manual calculations and to perform more sophisticated image-guided interventions. Our group began using a Data General mini-computer attached to a Tektronix vector display and a sonic digitizer in order to calculate coordinate target points and transpose imagederived volumes on a modified CT-compatible #
Springer-Verlag Berlin/Heidelberg 2009
Todd-Wells stereotactic frame in the late 1970s at the Erie County Medical Center in Buffalo, New York [2]. The Data General S140 computer, configured with a mere 128 kilobytes of memory was the size of a large refrigerator. The nine platter 192 megabyte (MB) external disk drive was the size of a washing machine. An Apple II personal computer was used as a remote display in the operating room to perform the first CT-based, computer-assisted laser resections of deep-seated brain tumors [3–5]. This whole computer configuration cost about $100,000. Our group later moved to Sisters of Charity Hospital in Buffalo and incorporated another Data General minicomputer-based system: an Independent Physician Diagnostic Console (IPDC, General Electric Medical Systems, New Berlin, Wisconsin). This system contained a 320 320 raster display device which allowed us to develop software to directly display and manipulate CT, MR and digital angiographic images and write software for the simulation and performance of a variety of image-guided stereotactic procedures including biopsy, thalamotomy, interstitial implants and volumetric craniotomies [3–13]. Your cell phone has more computing power and storage than either of the computer systems our group started with years ago.
Computing Perspectives Gordon Moore, a founder of computer chip manufacturer Intel, predicted in 1965 that the
3132
188
The future of computers and imaging
number of transistors on integrated circuits would double nearly every year, thereby approximately doubling the speed, memory access and capacity comparably in the same timeframe [14]. Processors and memory technology have generally followed this prediction. Engineering-class and personal computers used in image-guided surgery systems have traditionally contained a single central processing unit (CPU) connected to internal random access memory (RAM) and external, but slower data storage on external (hard drive) peripherals. The focus had been on making single CPUs process data even faster by making larger and more capable processor chips on a single silicon wafer referred to as wafer scale integration. Highend, ‘‘supercomputers’’ have, for years, integrated multiple processors within the same system. This technology has recently trickled down to advanced engineering workstations and personal computing systems over the last few years. These systems are defined by separate, but interconnected, processor chips contained on a single integrated circuit board. More recently, multiple core processors are making their way into commodity-type personal computers. Multiple core systems package multiple processors into a single component on an integrated circuit board. Both of these multiprocessing technologies attempt to speed up computer calculations by techniques known as pipelining and multithreading. Software does not automatically execute linearly or exponentially faster using multiple processor or multi-core systems. For example, a software program will unlikely execute two times faster on a two-processor or two-core computer. The software has to be separated into pieces that can be efficiently executed in parallel on the multiple processors and this is not a simple task [15]. Amdahl’s Law for processor speedup postulates that the theoretical maximum performance increase using parallel computing versus single CPU computing would be about twenty times no matter how many processors/cores are used if 95% of a program can be parallelized [16]. Therefore, at the current time, it appears
like there is a theoretical limit on the numbers of processors/cores that can be utilized to optimize the speed of an algorithm even through there is no current hardware limit on the number of processors/cores that can be built assuming you could provide the power and dissipate the heat generated by such systems. Furthermore, software engineering tools for developing, optimizing and debugging multi-processor software are a generation behind hardware developments, but the benefit of general purpose multiprocessing technology is huge. Increasing individual processors calculation speeds and combining these processors into multiple processors and core systems is not the only way to increase computer performance. Bottlenecks in the communication between processors slow down the overall benefit of multiprocessor systems. Current technology uses small wires to facilitate communication between processors and there is a fundamental limit in the speed by which data can move over wire technology. Additionally, wire technology utilizes extra electrical current and worst of all, generates heat that must be dissipated. Novel research is currently underway to develop new technologies to increase the computational speed while decreasing the electrical current and the production of heat in multiprocessor computer systems. Silicon Photonics is a evolving field within computer science to provide a more efficient manner to interconnect chips at high speeds so that newly developed computer systems are perhaps a thousand times faster, more energy efficient and produce less heat. Sun Microsystems has recently been awarded a multiyear grant from the Pentagon to explore options for replacing wires used to communicate between chips with laser technology. Each chip would interconnect and communicate at extremely high speeds with every other chip using laser light that could carry tens of billions of bits of information every second. NEC, a Japanese maker of supercomputers also recently announced advances in optical chip interconnections for supercomputers that too,
The future of computers and imaging
will trickle down into the engineering and personal computer technology used in image guided surgery.
Imaging Perspectives Advances in medical imaging have come a long way since the early days of image-guided surgery. The typical first-generation image-guided dataset was a relatively small number (10–30 slices) of rather thick (5–10 mm) preoperatively collected 256 256 or 320 320 pixel computed tomography (CT) images. Digital angiographic and MRI images were integrated into the imageguided repertoire in the early 1980s as soon as hardware registration methods and image transfer software were developed. One center incorporated a CT scanner directly in an operating room in the early 1980s [17]. New imaging modalities as well as other digital input sources of data will continue to be integrated into image-guided databases. Currently the majority of image-guided procedures use one, or perhaps two, preoperatively collected imaging databases such as computed tomography (CT) or variations of images collected from magnetic resonance (MR) scanners. Some image-guided procedures are now being performed in an operating room adjacent to a scanner and the patient moved back and forth when an updated scan may be necessary. Other systems are available to move a scanning device to the patient such as a track-mounted MR or a mobile C-arm system that creates near-computed tomographic images intraoperatively. Many of these portable systems are bulky or hard to move around, may be limited to being used in only one operating room or do not fit well in a relatively smaller sized operating room that is available at many institutions. Some centers are performing procedures within the imaging unit, but not all image-guided procedures are candidates for being performed in the confines of the scanner. A variety of advances in imaging technology are discussed elsewhere in this textbook. From a
188
technical perspective, advances in imaging technology will impact image-guided surgery by delivering a wider variety of more densely collected datasets. Four-dimensional imaging will be the norm in the future. This higher variety and density of imaging data will require more capable computing systems to manipulate these data, many of which will be produced and registered to the patient during the procedure. Medical image registration and registration of images to treatment delivery systems by sensor based and robotic technologies are described elsewhere in this textbook. Briefly, images are created in a coordinate system defined by the scanning device. Treatment delivery devices generally define their own coordinate system by incorporation of stereotactic headframes, articulated arms, magnetic field and optical digitizers and Cartesian or other robotic devices. Image-guided registration involves spatially relating the imaging data to the treatment device. If more than one image dataset is utilized for a procedure or intervention, they must be individually registered to the patient [18] or correlated or fused to each other and then registered to the patient using image correlation or image fusion techniques [19]. New methods of image to patient spatial registration will need to be developed so that, rapidly scanned, densely collected, intraoperative datasets may be rapidly and accurately integrated and registered into the system. These methods will need to be less obtrusive than current ones, spatially correct for patient repositioning and movement between acquisitions as well as correct for potential geometric distortions inherent to the imaging unit and work even while a patient is underneath surgical drapes.
Tying It All Together First-generation image-guided systems were developed mainly at academic institutions and combined high performance minicomputers or engineering workstations with stereotactic frames,
3133
3134
188
The future of computers and imaging
custom-built Cartesian robotic technology [20] and commercially available articulated arms. Second generation image-guided systems integrated newer sensor technology including magnetic field and optical technology and more sophisticated software. Current generation image-guided systems combine high end personal computers and similar second-generation sensor technology that are often contained in a large cart that needs to be moved into and out of the operating room. A few image-guided systems involve fixed or ceiling mounted systems that work in one or two operating or procedure rooms. Some institutions and manufacturers have focused on developing instrumentation and software to perform procedures directly in an imaging device while others have focused on developing technology to temporarily move the imaging device into and out of the surgical field. Other centers have built operating facilities adjacent to the imaging device allowing the patient to be moved in and out of the scanner when updated imaging is required. The primary deliverable of an image-guided system is to three-dimensionally direct instruments or guide treatment to specific locations as defined by imaging databases. Imaging data needs to be spatially registered to the patient as positioned during the procedure regardless of the system or method of treatment delivery. Registration is accomplished by a number of current methods that include stereotactic frames, external stick-on and implantable fiducials and surface matching. Frames are considered cumbersome and patients don’t necessarily like them, but they do produce the most accurate, reliable and most reproducible accuracy, because they are rigidly attached to the patient and their imaging reference system deposits geometrically well-defined markings on every image leading to a very direct and intuitive spatial mapping between the entire image volume and the delivery mechanism. External stick-on markers are widely used in image-guided surgery, but can move on the skin or even fall off. Externalized implantable fiducials
have been available for number years, but these require a minor surgical procedure to implant them and there are potential issues of infection. Both of these forms of external point-based registration techniques involve depositing only a small number of markings in the imaging data to spatially register the images to the delivery system using point matching/least-squares fit registration algorithms. The limited number of external points utilized in these registration methods are not the most optimal, especially when used with potentially geometrically distorted image data that may be provided by, for example, magnetic resonance imaging [21]. Surface registration methods have been available for years, but are not widely used and have limited use when the patient ‘‘surface’’ is hidden under surgical drapes. Newer image registration methods will be developed that are easy to use, comfortable for the patient, enable highly accurate spatial registration even with potentially distorted imaging data, are less obtrusive and are accessible even under surgical drapes. Wireless radio frequency identification tags (RFID) are being widely adopted to monitor the location of inventory in stores. It is very likely that very small, wireless, multimodality compatible fiducials will be integrated with newer intraoperative sensor and robotic technology in future generations of image-guided systems. Telepresence technology will also continue to be incorporated in image-guided systems. Computing devices will becoming smaller, faster and contain much more capacity to store and manipulate larger and denser imaging datasets. Newer software methods will also be refined to automatically divide image-guided software algorithms into internal computing code that can take advantage of multiprocessors systems without the software engineer having to worry about how to separate their code into separate segments that can be processed in parallel. The days of the large, portable equipment rack with a large, optical sensing bar, cables
The future of computers and imaging
sometime strung across the floor with a monitor 10 feet or more away from the surgeon will be a thing of the past. Computer and imaging systems integrated into current generation image-guided system sits idle during even much of the procedure. Many of these types of systems may remain powered on in an adjacent room waiting for data to arrive from radiology which is not very energy efficient. Furthermore, the average lifetime of the hardware in an image-guided computer and imaging system is generally out of date in a 3–5 year timeframe necessitating large expenditures for the ‘‘next generation’’ of an image-guided system every few years. Computing power will be delivered similar to how electricity is delivered from an electric utility: you only use what you need and you do not need to provide and support (or buy) your own dedicated system. This type of utility computing is commonly referred to as cloud computing. Future image-guided system will utilize cloud computing and will also likely incorporate display technology similar to interactive internet browser-like technology which will only require a small wireless touch screen interface. Speech and voice recognition technology will also be more fully integrated into these types of systems. The image-guided facility of the future, whether in an operating room or in an imaging scanning facility will be configured as an indoor global positioning system-like (GPS) system [22,23]. An indoor, high fidelity GPS systems will define the work envelop as the entire room eliminating line-of sight or other second-generation sensor-based limitations enabling the patient, wireless registration fiducials, every instrument and the location and directional view of the surgeon to be tracked. The system will automatically incorporate and register intraoperatively collected, dense imaging data, provide the surgeon various methods to view and interact with the imaging data superimposed on the surgical field while optimally enabling them to simulate and deliver instruments or
188
treatments with much higher magnitudes of accuracy than present systems. Computing and medical image technology have had a significant impact over the past 30 years and will continue to positively impact the field of image-guided surgery well into the future.
References 1. Spiegal EA, Wycis HT, Marks M. Stereotaxic apparatus for operations on the Human Brain. Science 1947;106:349-50. 2. Goerss S, Kelly PJ, Kall B, Alker GJ, Jr. A computed tomographic stereotactic adaptation system. Neurosurgery 1982;10:375-9. 3. Kelly PJ, Kall B, Goerss S, Alker GJ, Jr. Stereotactic surgery for CNS tumors. CT clinical symposium 1982; Vol. 5, #5. 4. Kelly PJ, Kall B, Goerss S, Alker GJ, Jr. Precision resection of intra-axial CNS lesions by CT-based stereotactic craniotomy and computer monitored CO2 laser. Acta Neurochirurgica 1983;69:1-9. 5. Alker GJ, Kelly PJ, Kall B, Goerss S. Stereotaxic laser ablation of intracranial lesions. Am J Neuroradiol 1983;4:727-30. 6. Kelly PJ, Kall B, Goerss S, Alker GJ, Jr. A method of CTbased stereotactic biopsy with arteriographic control. Neurosurgery 1984;14:172-7. 7. Kelly PJ, Kall B, Goerss S. Transposition of volumetric information derived from CT scanning into stereotactic space. Surg Neurol 1984;21:465-71. 8. Kelly PJ, Kall BA, Goerss S. Computer simulation for the stereotactic placement of interstitial radionuclide sources into CT defined tumor volumes: technical note. Neurosurgery 1984;14:442-8. 9. Kelly PJ, Kall BA, Goerss S. Computer-assisted stereotactic biopsy utilizing CT and digitized arteriographic data. Acta Neurochirurgica 1984;Suppl 33:233-235. 10. Kelly PJ, Kall B, Goerss S. Stereotactic CT scanning for the biopsy of intracranial lesions and functional neurosurgery, Appl Neurophysiol 1984;46:193-9. 11. Kelly PJ, Kall B, Goerss S. Functional stereotactic surgery utilizing CT data and computer generated stereotactic atlas. Acta Neurochirurgica 1984;33:577-83. 12. Kelly PJ, Kall B. Pre-operative computer determination of interstitial iridium192 source placement into CNS tumor volumes. Acta Neurochirurgica 1984;33:377-83. 13. Kall BA, Kelly PJ, Goerss S. Interactive stereotactic surgical system for the removal of intracranial tumors utilizing the CO2 laser and CT derived data base. IEEE Trans Biomed Eng 1985;32:2, 112-16.
3135
3136
188
The future of computers and imaging
14. Moore Gordon E. Cramming more components onto integrated circuits. Electronic Magazine 1965; 38:8. 15. Kuck DJ. High performance computing. Oxford: Oxford University Press. 16. Horoi M, Enbody RJ. Amdahl’s law as an instrumentation tool for building efficient parallel code, http://www. phy.cmuch.edu/horoi/ijsca00.ps 17. Lundsford LD, Parrish R, Albright L. Intraoperative imaging with a therapeutic computed tomography scanner. Neurosurgery 1984;15:559-61. 18. Kall BA, Kelly PJ, Goerss SJ, Earnest F, IV. Crossregistration of points and lesion volumes from MR and CT. In: Proceedings of the seventh annual meetings of frontiers of engineering and computing in health care, Chicago, 1985. p. 692-5.
19. Kall B. Image reconstruction, registration and fusion. In: Gildenberg PL, Tasker RR, editors. Textbook of stereotactic neurosurgery. New York: McGraw-Hill; 2009. 20. Goerss SJ, Kelly PJ, Kall BA. An automated stereotactic positioning system. Appl Neurophysiol 1987;50:100-6. 21. Wang D, Strungell W, Cowin G, Doddrell DM, Slaughter R. Geometric distortion in clinical MRI systems. Part I: evaluation using a 3D phantom. Magn Reson Imaging 2004;22:9, 1211-21. 22. http://www.gpsworld.com/gpsworld/article/articleDetail. jsp?id = 3086, 2001. 23. Maurer CR, Jr, Fitzpatrick M. A review of medical image registration. In: Maciunas RJ, editor. Interactive image-guided neurosurgery. Park Ridge, IL: American Association of Neurological Surgeon; 1993. p 17-44.
191 The Future of Infusion Systems in Neurosurgery R. D. Penn
Drug infusion into the CSF by implanted drug pump systems represents a solution to the general problem of site specific delivery of medications to the brain. The success of selective perfusion of the spinal cord for pain and spasticity illustrates how powerful this technique can be if applied to the appropriate problem. Progress has been rapid for spinal applications because the anatomy is straightforward, the pharmacokinetics are simple and the devices for delivery are available and reliable. Other applications which require intraparenchymal delivery are much more complex, such as delivery of antibodies to control tumor growth, inhibitory drugs to control movement disorders, anti-seizure medications for epilepsy, or neurotrophins for Parkinson’s disease. These possible treatments are much more difficult and are still experimental. Spinal delivery of medications takes advantage of convective flow of CSF along the spinal cord. This rapid movement of fluid carries any medication from the site of infusion up and down the canal. Two forces are at work; the to and fro oscillations due to pulsations of the CSF driven by the cardiac pulse and the regular slow production and absorption of CSF. Medication released into the spinal subarachnoid space from a catheter will immediately mix with CSF due to pulsations and then be carried caudad and cephalad slowly with the general CSF bulk flow. The turnover of CSF results in the half-life of water soluble molecules of about 90 min as the old fluid is replaced by new. Proteins as large as neurotrophins and molecules as small as morphine and baclofen have almost the same half-life in CSF. Lipid soluble molecules and metabolized #
Springer-Verlag Berlin/Heidelberg 2009
substances are lost from the CSF more rapidly by uptake or conversion. The flow of CSF has another important effect. Newly produced CSF is always mixing with the old and diluting any medication present. The CSF coming down into the spinal canal from the basilar cisterns dilutes the spinal fluid containing drug. The further away from the source of the drug at the catheter tip the lower the concentration will be. This effectively lowers the concentration in the higher cervical area from lumbar infusion by a factor of four on average [1]. Thus, for a steady state infusion, the distribution of drug is maintained at a high concentration in the lumbar area and much lower in the cervical region. This regional distribution decreases supratentorial effects. The high levels in the cord region extend the therapeutic window and this is the key advantage of slow spinal infusion by drug pumps. Bolus infusion is different. A single large bolus given into the lumbar theca is carried up the neuraxis in relatively high concentration. It reaches the brainstem in 3–4 h and can produce major side effects. The same dose given slowly over hours will not produce a problem. The major danger of over-dosing a patient and producing respiratory depression occurs most frequently with bolus administration, either by errors in programming or in surgery with accidental injection of the drug containing fluid in the catheter. To be effective drug has to get to its specific site of action in the spinal cord from the CSF. For most medications this is by diffusion through the tortuous extracellular space of the spinal cord. Such diffusion is slow. The 2–3 mm of cord tissue baclofen has to go from the CSF to its site of
3156
191
The future of infusion systems in neurosurgery
action in Rexed areas II and III takes 30–60 min. Diffusion is driven by concentration differences. The amount of drug diffusing into the cord is directly proportional to the CSF level and its CSF level in turn is directly proportional to the infusion rate into the CSF by the pump [2]. These simple pharmacokinetic principles mean that doubling the infusion rate will double the delivery of the drug to its site of action once the steady state is reached. The time to a steady state in the CSF is approximately seven times the half-life plus the time that it takes for diffusion. As a rule of thumb, half a day is needed to get the full effect of a change in infusion rate. Once a drug has diffused into the spinal or brain tissue, it takes many hours for it to diffuse out. This delay means that any side effects will last for a long time. An overdose of morphine or baclofen can last for 24–48 h after the drug is stopped. Ziconotide which binds to the cord can accumulate and last even longer [3]. The fundamental fact of drug delivery to the spinal subarachnoid space that it is very slow to reach its target and is long-lasting once it does so. In contrast to I.V. or oral medications, IT therapy kinetics are much longer. The challenge in designing implantable systems for IT infusion has been how to deliver precise doses of medications over many years of therapy with high reliability. Fortunately, the first implanted drug pumps were developed for insulin administration. Although they failed because of the viscosity of insulin, the systems were engineered with precision and reliability. The first totally implanted drug pump was a constant rate device which was gas pressure driven. This evolved into a programmable pump that provided dosing flexibility but required a battery operated pump mechanism and an external control device. Currently, these two types of systems are the only ones available. To understand their advantages and drawbacks, the design principles of each type needs to be outlined. The constant infusion pumps are all based on simple mechanical parts. The Codman-3000
device, since it is most widely available, will be used as an example. It has a compressible reservoir that holds the medication. This is surrounded by Freon gas which at body temperature exerts a steady pressure on the reservoir regardless of the reservoir volume. The reservoir connects to a bacteriostatic outflow filter, and then to a long flow restrictive pathway (a small diameter channel) which slows flow to a milliliter or less per day. The outport is attached to a catheter which goes to the subarachnoid space. To provide access to the reservoir for refilling, the pump has a central silastic port which is penetrated by a special non-coring needle. For access to the catheter for giving a bolus or diagnostic test, another needle with a side hole is utilized (See > Figure 191-1a,b). Pumps using these design features have been used for almost 30 years. They are reliable and if they fail it is in a safe mode of gradually slowing down as the flow restrictive system becomes narrower. The clear advantages are that they do not need a battery which will deplete or have a control system or a motor which will be subject to electrical or mechanical failure. The Codman pumps range in weight from 98 to 173 g unfilled and are the size of a hockey puck they have rounded edges and an easily palpated fill-port in the middle. The reservoir can hold 16, 30 or 50 ml and flow at constant rate from a third to 1 cc per day. The most obvious disadvantage is that the rates cannot be varied, so the only way to provide a dose change is removal of all the fluid in the reservoir and replacement with a new concentration of drug. If the patient requires frequent changes, this is an inconvenient and expensive process. The pump’s output is subject to the laws of physics. Higher body temperatures or lower barometric pressure will increase the rate of flow. In clinical practice, these differences of 5–15% are rarely important. Once the pump is in a particular patient, the flow rate remains relatively constant and can be calculated from the residual volume at the time of refill.
The future of infusion systems in neurosurgery
191
. Figure 191-1 Diagram of the Codman 3000 implantable drug pump. The reservoir is filled by a non-coring needle that goes through a silastic self sealing septum (left). The bolus path way is reached by a special needle with a side hole (right). The reservoir is surrounded by pressurized gas to propel fluid through the flow restrictor (Courtesy of Codman Inc.)
. Figure 191-2 Diagram of the InSet Prometra programable implanted pump which is in clinical trials. The key feature is a fluid accumulator which is controlled by a simple electronic system to release programmed mimi- boluses of fluid to the out port (Courtesy of InSet Technologies Inc.)
To control the rate of flow from the pressurized reservoir, a number of mechanical solutions can be employed. > Figure 191-2 shows a simple control system that accumulates a small amount of fluid from the reservoir and then releases it to the outport. The number of releases per minute is set electronically so a series of mini-boluses can be given and the desired rate of infusion controlled. The system takes advantage of the kinetics of CSF delivery. The mini-boluses blend together during the mixing and convection of
CSF, so, in effect, a slow constant rate is achieved. The IN-SET Company has made such a device (Prometra Programmable Implanted Pump) which is in clinical trials in the United States. The potential advantages are active control with few moving parts, and its lightweight and energy efficiency. A more complex and versatile solution has been provided by Medtronics peristaltic pump mechanism. Its basic design has remained the same since it was introduced over 25 years ago.
3157
3158
191
The future of infusion systems in neurosurgery
. Figure 191-3 The Medtronic SynchroMed II system. (a) Picture of the pump with the top removed to see the parts of the system, (b) a diagram of the inside of the pump showing the fluid flow pathways and reservoir, (c) picture of the peristolic pump, and (d) x-ray view, (e) hand held physician programmer (N’ Vision) and a sample display page (Medtronic web site)
The future of infusion systems in neurosurgery
The newest member in this line is the SynchroMed II and the cutaway and > Figure 191-3 shows its parts. A rotary (peristaltic) pump is interposed between the reservoir and the outport. A clock-like mechanical system moves the rotor. An electronic module and battery powers the motor and drives it according to a specified program. The battery last 4–7 years depending on the rates of flow. The programming is performed using a hand-held physician programmer than can set flow rates, bolus dosing, and variations of rate during a 24-h cycle. The programmer can also adjust rates to accommodate changes made in the drug concentration. As it has evolved over the years, additional features have been added to prevent over filling of the reservoir and inadvertent injection of the side port. It is important to note that the pump at present does not measure the volume of fluid in the reservoir. Instead, it counts the amount of fluid which should have been pumped. It calculates when the refill is due but is not able to determine if the medicine was actually given. The electronic system will warn that the battery is running down, that a refilled is due or if an electronic failure occurred. In spite of the complexity of the system, the Medtronic pump has a relatively low failure rate of 1–2% per year. When it fails it stops rather than giving an overdose. On rare occasions, it can stop or stop and restart due to mechanical problems. A present all the available pumps share certain problems. They are large, weighty and cosmetically unacceptable to some patients. In children and thin adults, they can be difficult to implant in the abdominal wall. Because they continue to use a gas-driven reservoir, they all slow down significantly as the last milliliters of fluid are reached. The slow down can become clinically evident in sensitive patients. A report of serious drug withdrawal symptoms occurring when a Medtronic pump had 2 ml of fluid left in the reservoir is a reminder of what can occur with this type of pressurized design [4]. Patients
191
who complain of more pain or spasticity as the refill time approaches can be managed by earlier refills. While the pump is the most complex element in the infusion system, the part that fails most frequently is the catheter. The first catheters were not anchored and slipped out from the lumbar subarachnoid space with movement. The initial Medtronic catheters were light and flexible and weakened by impregnation with radio-opaque material. Slowly better catheters, anchors and implant techniques have been engineered and catheter complications have been decreasing. Catheters can be a site of sterile granuloma formation. Granulomas have been described almost exclusively in patients receiving high concentration of morphine and morphinelike drugs [5]. However recently, three patients with baclofen infusion had been reported to have granulomas [6–8]. Increased pain and signs and symptoms suggesting cord compression should alert patients and physicians to the problem. An MRI can identify the mass. The most common cause of drug withdrawal in patients with implanted drug pumps is failure to perform the refill on time. This points out that the infusion of medicine is part of a complex treatment modality that requires appropriate support systems run by nurses and doctors who have educated their patients. Patients must be told how the pump works and the consequences of sudden withdrawal. They need to be able to recognize early withdrawal symptoms and know where to get rapid treatment. On rare occasions withdrawal can lead to fatal complications [5]. While disruption of infusion is often a patient/medical system problem, it can also be a mechanical problem with the catheter or flow in the subarachnoid space. Chapter 117 describes the steps to take to diagnose an obstruction or a delivery problem. In contrast to drug underdosing, overdoses are almost always iatrogenic. An error in programming, for example using milligrams instead of micrograms, or forgetting that drug is in the
3159
3160
191
The future of infusion systems in neurosurgery
catheter system, can lead to a sudden bolus overdose. Unfortunately, the Medtronic programmer software does not provide a warning about potential overdosing. It only asks if it is okay to give an instruction to the pump. After a quarter century of use these simple precautions have still not been incorporated into the software, and unnecessary programming errors continue to be made. The use of the side port is also a place for error. The dead spaces in the side port and in the catheter contain drug which can be infused suddenly as a bolus if not withdrawn first. The slow kinetics of IT delivery mean that the overdose does not arrive at the brainstem level until hours later. The patient may have returned home by the time respiratory depression occurs. Although drug infusion systems have some drawbacks and complications, they have been used to treat spasticity and pain in over 100,000 patients with results that could not have been achieved by systemic medication. The systems remain expensive, large, bulky and not particularly user-friendly, but they are reliable and provide the most accurate, long term delivery of medicine ever devised. The future of these systems depends on our ability to find new applications and new medications for delivery. While an implanted insulin pump remains a dream, the
technology used for intrathecal drug delivery has resulted in great benefits for many patients around the world.
References 1. Kroin JS, Ali A, et al. The distribution of medication along the spinal canal after chronic intrathecal administration. Neurosurgery 1993; 33(2):226-30; discussion 230. 2. Muller ZK, Penn R. Local-spinal therapy of spasticity. Berlin: Springer; 1988. p. 270. 3. Penn RD, Paice JA. Adverse effects associated with the intrathecal administration of ziconotide. Pain 2000;85 (1–2):291-6. 4. Rigoli G, Terrini G, et al. Intrathecal baclofen withdrawal syndrome caused by low residual volume in the pump reservoir: a report of 2 cases. Arch Phys Med Rehabil 2004;85(12):2064-6. 5. Coffey RJ, Burchiel K. Inflammatory mass lesions associated with intrathecal drug infusion catheters: report and observations on 41 patients. Neurosurgery 2002;50(1): 78-86; discussion 86-7. 6. Deer TR, Raso LJ, et al. Inflammatory mass of an intrathecal catheter in patients receiving baclofen as a sole agent: a report of two cases and a review of the identification and treatment of the complication. Pain Med 2007;8 (3):259-62. 7. Murphy PM, Skouvaklis DE, et al. Intrathecal catheter granuloma associated with isolated baclofen infusion. Anesth Analg 2006;102(3):848-52. 8. Narouze SN, Mekhail NA. Intrathecal catheter granuloma with baclofen infusion. Anesth Analg 2007;104 (1):209; author reply 209–10.
194 The Future of Molecular Neuro-Oncology J. A. J. King . M. D. Taylor
Introduction Knowledge of the molecular mechanisms responsible for the development of tumors in general and within the central nervous system in particular, is the cornerstone around which to develop targeted molecular therapies and ultimately to improve outcomes for patients in the field of neuro-oncology. From the point of view of the clinician, an understanding of the genetic aberrations involved in the individual patient’s tumor will likely aid us in our decisions regarding the most appropriate treatment, the likely response to that treatment and our ability to provide this patient with information regarding prognosis. In this chapter, a review of the more common mechanisms of tumor development will be presented followed by comments on the important inherited cancer syndromes with relevance to the central nervous system, and finally a summary of the known genetic and epigenetic events underlying the development of specific brain tumors.
The Hallmarks of Cancer Cancers arising from different tissues in the body are vastly different in the biology and behavior of their cell population, but despite this heterogeneity there are some common themes in all cancer development.
#
Springer-Verlag Berlin/Heidelberg 2009
Hanahan and Weinberg have described the hallmarks of cancer as the following six essential characteristics of the transformed cell population, (1) self-sufficiency in growth signals, (2) insensitivity to anti-growth signals, (3) evasion of apoptosis, (4) limitless replicative potential, (5) tissue invasion and metastasis and (6) sustained angiogenesis [1]. At the origin of the development of these characteristic traits, the transforming cell acquires a series of mutations to its genomic DNA [2] resulting in altered function of crucial regulatory genes. The specific types of mutational events that can occur within the human genome in oncogenesis are many and varied, resulting from insults of a chemical, physical (thermal, ionizing and UV radiation) and viral nature or by defective internal systems of DNA replication and repair [3]. There is a small group of inherited disorders (1% of human cancers [4]) where vertical transmission of a genetic mutation is responsible for an increased cancer risk. Neurofibromatosis type 2 (NF2) is one such disease, one of a number of inherited disorders that will be reviewed, which also provide critical insights into the molecular mechanisms responsible for sporadic tumors. Regulating these hallmark processes that become disordered in the development of cancer, are a series of important genes that exert either an inhibitory or activating effect on the process. A gene in which mutation leads to overactivity of that protein or system (a stimulator of proliferation, angiogenesis, migration etc.) is called an
3202
194
The future of molecular neuro-oncology
oncogene, whereas an altered gene that has subsequently lost its inhibitory control of the process (proliferation etc.), is termed a tumor suppressor gene.
Control of Cell Growth and Proliferation Cell proliferation can be regulated directly through the mechanisms of mitogenic stimulation that determine whether a cell passes through the restriction points of the cell cycle, or indirectly through regulation of the commitment to terminal differentiation, or programmed cell death (apoptosis).
Mitogenic Stimuli, RTKs, the RAS/MAP Kinase Pathway Normal cell growth is dependent on mitogenic stimuli being transmitted from transmembrane receptors to the nucleus, through complex intracellular signaling cascades. Extracellular matrix components, cell-cell interaction molecules and perhaps most importantly, locally produced growth factors can all provide stimulatory signals for growth. Growth factors (GFs) in serum have been shown to be essential for proliferation of cells in culture, and these factors exert their action through a diverse array of receptor tyrosine kinases (RTKs) at the cell membrane. Downstream signaling following RTK activation via the Ras/MAP kinase pathway activates amongst others, the transcription factor Elk-1 resulting in altered gene transcription. Stimulation of this pathway initiates a chain of events that leads to progression of the cell through the ‘‘restriction’’ point of the cell cycle and ultimately to mitosis and expansion of the cell population. This can occur as a part of the normal processes of growth through infancy, childhood and adolescence, as a part of the replacement of normal cells in regions of the body with high
cell turnover such as the gut epithelium, or in an abnormal fashion in benign and malignant tumor formation.
Independence from Growth Signals The transformed or deregulated cell becomes independent of the environment for its cues to grow and can do this through a number of mechanisms. It is well documented that tumor cells can overproduce growth factors to which they can respond (i.e., PDGF-BB in glioma [5,6]), and over expression and modification of the RTKs themselves also serves to increase mitogenic stimulation of the abnormal cell. Over-expression can be a consequence of gene amplification resulting in enhanced or constitutive kinase activity. Thirty-one percent of breast carcinomas over express Erb-B2 and this correlates with a poor prognosis [7]. The D2–7 epidermal growth factor receptor (EGFR) mutant which is constitutively active, is found in up to 40% of glioblastomas [8–11]. Modification of RTKs via gain of function mutations as well as the formation of oncogenic fusion proteins between the catalytic domain of an RTK and an unrelated protein (TrkA, Ret in papillary thyroid carcinoma [12], FGFR 1 in AML [4]), are further mechanisms via which transformation can take place. Integrins expressed on the cell surface are capable of interacting with the extracellular matrix and sending proliferative signals via the Ras/ MAPK pathway to the nucleus. Alteration of integrin expression can be utilized by the transformed cell in order to maximize mitogenic stimuli. In the intracellular compartment, alteration of any of the activators in the mitogenic cascade can have an oncogenic effect. The Ras/MAP kinase pathway is central to this and Ras mutations are seen in over 15% of human cancers [13,14]. Ras mutations are seldom seen in brain tumors however the pathway is commonly activated upstream of Ras. BRAF mutations have been
The future of molecular neuro-oncology
detected in 66% of human malignant melanomas [14]. Indeed all members of the Ras/MAP kinase signaling cascade can cause uncontrolled proliferation when over expressed.
Regulation of the Cell Cycle The cell cycle comprises all the fundamental processes a cell requires to undergo division and the creation of two identical cells. There is an elaborate system of cell cycle control that regulates cell division and may be disturbed in the origins of cancer, resulting in both excessive proliferation and the loss of terminal differentiation [15]. The standard eukaryotic cell cycle is characterized by two fundamental phases. The first being DNA replication (synthesis, S-phase), followed by segregation of the two copies of the chromosomal DNA (mitosis, M-phase), separated from one another by the gap phases, G1 and G2. Withdrawal from the cell cycle, the resting or quiescent state, is known as G0. The two key periods in the determination of progression through the cell cycle are the checkpoints within the gap phases (G1 and G2).
G1/S Transition The most significant checkpoint in the cell cycle occurs in late G1, approximately four hours prior to the cell’s entry into S-phase, and represents the last restraint after which the cell is irrevocably programmed to begin DNA synthesis, regardless of mitogenic or anti-proliferative signaling. In order for the cell to pass through the G1 checkpoint (or ‘‘restriction point’’), it requires the signals that a favorable environment for growth exists. Extracellular signals acting via receptors at the cell surface set in train a series of protein interactions and ultimately alterations in gene expression. Essentially, cell cycle progression is determined by the presence of cyclin-dependent kinases
194
(CDKs), made up of both cyclin (regulatory) and cdk (catalytic) subunits, in particular cyclin D- and E-dependent kinases and their subsequent effects on the retinoblastoma protein, Rb. The antiproliferative signals acting at the G1 checkpoint are concentrated on the phosphorylation state of the Rb protein, and its related proteins p107 and p130. Dephosphorylated Rb binds and inactivates proteins (the transcriptional activating E2F family) that are pro-proliferative, thereby inhibiting proliferation [16].
G2 Checkpoint The G2 checkpoint provides a point of review of the process of DNA synthesis and confirmation that the phases of the cycle have been conducted correctly prior to mitosis. This is principally determined by the activation status of the mitosis promoting factor (MPF), consisting of the cyclin dependent kinase, Cdc2 and a B-type cyclin, which is under the influence of a variety of regulators [17]. p53 is thought to play a role in the regulation of the G2 checkpoint. Exit from the cell cycle is determined by the cessation of mitogenic stimuli and a reversal of the activating processes. Cyclin D1 production is halted, cyclin D1/Cdk4 complexes are degraded and the CDK inhibitors block cyclin E-Cdk2 activity, ultimately leading to hypo-phosphorylation of Rb. Rb is not hypo-phosphorylated until mitosis is completed and cells re-enter G1.
Disorders of Cell Cycle and Cell Cycle Control That Lead to Cancer Determination of the signal to divide is the outcome of a conflict between proliferation (oncogene) and anti-proliferation (tumor suppressor) gene expression. There are a series of safety mechanisms that exist in normal cells to ablate
3203
3204
194
The future of molecular neuro-oncology
the activity of oncogenic stimuli. Mutations that modify or disrupt these mechanisms that control G1 phase progression in particular, are seen with a high frequency in human cancers. Tumor cells thus develop the capacity to override or ignore antiproliferative signals. The Rb pathway, critical to regulation of the cell cycle, can be disrupted by direct inactivation of Rb [18], by cyclin D-dependent kinase (cyclinD1/Cdk4) over expression or by loss of the CDK inhibitors p16INK4a [19] and p27KIP1 [20]. Abnormal expression of the genes that regulate Rb phosphorylation status can result in loss of cell cycle control despite the presence of a normal Rb protein. Mutations occur in a mutually exclusive fashion in the Rb system in human cancer, as functional alteration of one member is sufficient to disrupt the pathway and provide that essential contribution to uncontrolled proliferation. It is of interest that cells with mutations affecting the Rb pathway do not divide at an accelerated rate, with the mutation contributing to carcinogenesis by providing a survival advantage in the prevention of differentiation or senescence. The addition of further mutations, for example in the p53 pathway, then enables the abnormal cell to remain in cycle.
Physiological Apoptosis Apoptosis is a critical physiological process observed in a wide variety of tissues in the living organism that controls normal cell numbers during development, disease and ageing. In addition to this physiological role of apoptosis in development, programmed cell death represents a safety mechanism via which abnormal cells can be removed from a given cell population. In order to survive, abnormal malignant cells must acquire the ability to evade or develop resistance to normal apoptotic signals. A complex series of competing events regulate the balance between cell proliferation and
apoptosis in mammalian tissues in order to maintain the normal tissue structure and function [21]. Subservient to this control system is a series of effector molecules, principally the caspases, which can be activated by both the so-called ‘‘intrinsic’’ or mitochondrial pathway (chemotherapy, radiation) and the ‘‘extrinsic’’ apoptotic pathway (death receptors of the tumor necrosis factor receptor (TNFR) family). Cleavage of these molecules and their substrates coordinate morphologically recognizable apoptotic cell death [22]. Once initiated, the process of apoptosis is manifest as a characteristic progression starting with disruption of the cell membrane, disintegration of the cytoskeleton, extrusion of the cytosol culminating in nuclear and DNA fragmentation [23]. Early biochemical evidence of apoptosis is external exposure of the normally inward facing phosphatidyl serine residues of the cell surface, the recognition of which, by macrophages, stimulates phagocytosis. The DNA binding transcription factor, p53, is one of the key regulators of apoptosis. p53 has a wide array of functions, principally in coordinating the cellular response to stress. p53 is critically involved in the cell cycle at both G1S and G2M transitions, inducing cell cycle arrest upon sensing DNA damage in order to allow for repair [24]. It exerts this effect at least in part by inducing the transcription of another regulatory gene, p21Waf1 [25]. p53 is also involved in regulating senescence, inhibition of angiogenesis and metastasis [26], DNA repair and in response to hypoxia.
Deranged Apoptosis in Cancer Mechanisms by which the transformed cell can evade apoptosis are multiple and include selection of mutations abrogating the function of pro-apoptotic regulatory elements such as death receptors and up regulation and aberrant expression of inhibitors of apoptosis, both at the control and effector level.
The future of molecular neuro-oncology
Alterations in important members of the apoptotic cascade have been described in human tumors, from the Bcl 2 family of proteins (Bax, Bak, Bid, Bim, Bcl-2, Bcl-XL, Bcl-W) [1,27–29], to mutations of fas ligand [30], and over-expression of apoptosis inhibitors, LFG [31], c-FLIP [32] and Fas associated protein, FAP-1 [33]. Finally, mutations of the proapoptotic regulator, p53 are the most common genetic lesion in human cancer, present in more than 50% of all cases of the disease [34,35] and will be discussed subsequently.
Differentiation The development of a complex multicellular organism with a vast array of differentiated cells performing a variety of structural and functional roles, arising from a single fertilized egg, requires an intricate system of controls. During embryogenesis the pattern of the body is determined, cell types are determined and then growth is required for full development. Differentiated cells can be renewed by simple duplication or by the differentiation of relatively undifferentiated stem cells.
Deranged Cellular Differentiation in Cancer In general, changes that block differentiated, post mitotic states and result in evasion of differentiation-inducing signals [36] whether they be transmitted via RTKs or other systems, may play a role in cancer formation. Leukemias are seen to arise from disruption of the normal program of differentiation, such that committed progenitors continue to divide indefinitely instead of terminally differentiating. Angiogenesis is ultimately a specialized form of migration, proliferation and differentiation and an abnormality of the differentiation process is implicated in the abnormal immature blood vessels seen in solid tumors.
194
Attempts to utilize this knowledge in the identification of novel therapies has led to the use of retinoic acid to induce differentiation in human promyelocytic leukemia (PML), a terminal differentiation therapy that achieved some success [37].
Limitless Replicative Potential The maintenance of telomeres is critical to the development of a limitless replicative potential of a cell. A telomere is a region of highly repetitive DNA (sequence TTAGGG) at the end of a linear chromosome that functions to protect that region. Replication of chromosomal DNA during cell division results in loss of genetic material at the telomeric end. Loss of the telomeric DNA prevents loss of important genetic information. Telomeric shortening results in chromosomal unfolding which is interpreted as DNA damage and the cell will enter cellular senescence, growth arrest or apoptosis depending on its genetic background (i.e., p53 status). Telomerase is a ribonucleoprotein that catalyzes the synthesis and elongation of telomeric repeats at chromosomal ends. Human somatic cells do not express telomerase and the failure to acquire telomerase activity in successive cultures confers senescence. Ninety percent of tumors show up regulation of the enzyme telomerase. Another putative mechanism whereby tumor cells evade senescence includes use of alternative lengthening of telomeres (ALT) [38]. Telomerase activity has been documented in CNS tumors such as GBM, AA, oligdendroglioma and meningioma [39,40]. Malignant tumors have a higher rate of telomerase activity than benign tumors [41].
Invasion/Metastasis Invasion, more so than metastasis, is critical to the malignant phenotype of CNS tumors. Cellular invasion through normal brain is thought to be mediated by cell-surface adhesion molecules,
3205
3206
194
The future of molecular neuro-oncology
proteinases and other enzymes that have proteolytic capacity to degrade the extracellular matrix (ECM). Tumor cells have been described to follow myelinated fiber tracts (corpus callosum) and/or along vascular structures. Other potential routes of tumor dissemination include along the leptomeninges and via CSF pathways. Overexpression of cell adhesion molecules (CAM) such as integrins [42], aberrant production of ECM components such as tenascin and vitronectin, disordered production of cell motility regulators, the Rho family of GTP binding proteins by tumor cells are all thought to promote invasiveness in the setting of proteinase secretion [43–46] to degrade the ECM and create space for movement.
Angiogenesis Angiogenesis is a physiological process which coordinates growth and maintenance of blood vessels within tissues and organs. Stimulatory (VEGF, PDGF) and inhibitory (TSP-1, angiostatin, endostatin) signaling pathways for angiogenesis co-exist to reach a delicate balance in normal tissues. In the CNS this process is tightly regulated and limited to the growing brain and within the endothelium of established vasculature.
Pathological Angiogenesis Critical to the continued viability of a tumor cell mass is the supply of oxygen and nutrients. To achieve this and in response to stimuli such as hypoxia, an established tumor mass will initiate the pathological growth and development of new blood vessels. This process is principally thought to be upregulated by increased expression of vascular endothelial growth factor (VEGF) and its receptor (VEGFR) in addition to stimulation of the angiopoietin system, incorporating family members Ang-1 and -2 and
the angiopoietin receptor TIE2. Simultaneously angiogenesis inhibitors (endostatin, angiostatin) are down regulated. With an understanding of the basic biological mechanisms underlying cancer formation, a review of the genetics behind inherited cancer syndromes and sporadic tumor types in the CNS is presented.
Inherited Cancer Syndromes Inherited cancer syndromes contribute 1–2% to the overall incidence of cancer and represent rare/unusual cancers or combinations of these sometimes associated with developmental defects or non-neoplastic phenotype. In the majority of cases inherited cancer syndromes are secondary to germline loss of function mutations in tumor suppressor genes. The inheritance is generally autosomal dominant, with a variable degree of penetrance. The study of these conditions has contributed significantly to the understanding of the molecular mechanisms of cancer formation in general.
Li Fraumeni Syndrome Individuals with Li Fraumeni syndrome are predisposed to developing several malignancies, including brain tumors, sarcomas, breast cancer, leukemia and adrenal cortical carcinoma [47]. Li Fraumeni is a diagnosis based on clinical criteria, being defined as a patient under the age of 45 with a sarcoma who has a first degree relative aged 45 or under with any cancer and an additional first or second degree relative under 45 years in the same lineage with any cancer or sarcoma. In LFS, about 12% of tumors are in the CNS, including astrocytoma, medulloblastoma, primitive neuroectodermal tumor, choroid plexus carcinoma and ependymoma.
The future of molecular neuro-oncology
Germline mutations in the p53 gene have been identified in 70% of patients with the clinical diagnosis of Li Fraumeni syndrome [48]. Most mutations are located in exons encoding the DNA binding region of the p53 protein although there is no clear genotype: phenotype relationship with regards to CNS tumors. p53 is a transcription factor that controls several biological processes important for tumor growth control including regulation of the cell cycle, angiogenesis and apoptosis. Loss of p53 is the most common mutation in human cancer and leads to unregulated growth and further mutational events. Mouse models have been developed in support of this hypothesis. p53 +/ and p53/ mice are viable and develop a variety of tumors, principally hematological malignancies and sarcomas. Brain tumor development in these mice is rare [49]. Adenoviral p53 gene therapy resulting in over expression of p53 in tumor cells with the induction of growth arrest and apoptosis, is reported in the treatment of Li Fraumeni related tumors although large scale studies have not been published [50] (> Figure 194-1).
Neurofibromatosis Type 1 (NF1) NF1 is a common (1/3,000) autosomal dominant inherited cancer syndrome. Clinical diagnostic criteria are well described in the literature, with the key features including cafe´ au lait spots, intertriginous freckling, Lisch nodules of the iris, subcutaneous neurofibromas and a positive family history. These patients are susceptible to the development of a number of CNS tumors. NF1 patients (15–20%) develop low grade astrocytomas involving the visual pathways (optic pathway glioma (OPG)). Plexiform neurofibromas and malignant nerve sheath tumors are seen in NF1 with ependymoma, medulloblastoma and glioblastoma also reported but seen much less frequently than OPG. NF1 patients also have
194
. Figure 194‐1 T1 weighted, gadolinium enhanced axial MRI of a left frontal glioblastoma multiforme (GBM) in a 15–yearold child with Li Fraumeni Syndrome and a previous adrenocortical carcinoma
a higher incidence of developmental delay, behavioral problems and learning difficulties. The NF1 gene is on chromosome 17q and is a large gene, prone to mutation with a 50% incidence of de novo mutation and 50% with a family history. The gene product, neurofibromin, a Ras GTPase activating protein (GAP), is responsible for inactivating RAS by accelerating its conversion from the active to the inactive state. The protein reduces Ras mediated mitogenic signaling and cell proliferation thereby functioning as a tumor suppressor protein. High grade glioma is rarely seen in NF1 and it is felt that additional mutations in critical regulatory genes such as P53 and PTEN are needed for the development of these malignancies. It has been postulated that radiation can precipitate these events and 50% of patients receiving radiotherapy for ONG early in life develop second tumors [51]. Heterozygote mice for nf1 mutation do not develop neurofibromas or pigmentation defects
3207
3208
194
The future of molecular neuro-oncology
but are at an increased risk for tumor development, in particular phaeochromocytomas and myeloid leukemias [52]. In a neural-specific NF1 knockout mouse model, Zhu and colleagues have shown that NF1 plays a key role in the development of the CNS glial cell lineage and that sustained increased proliferation in glial progenitor cells may underlie tumor formation in the optic nerve [53]. With regards to novel therapies, phase one trials of treatment of NF1 related tumors with farnesyl transferase inhibitors, which block posttranslational isoprenylation of Ras and inhibit Ras activity, are reported [54] (> Figure 194-2 and > 194-3).
Neurofibromatosis Type 2 (NF2) Neurofibromatosis type 2 is an autosomal dominant disorder characterized by the development of tumors of the central nervous system [55].
. Figure 194-2 FLAIR, axial MRI of a left occipital anaplastic astrocytoma (AA) in a 12-year-old child with a germline homozygous mismatch repair gene (MLH1) mutation and mosaic NF1
The hallmark feature of NF2 is bilateral vestibular schwannoma (VS), and is diagnostic of the condition. Diagnostic guidelines have been defined to further clarify the varying presentations of this disorder [56]. Individuals with this disease classically develop symptoms within the second and third decades of life. Patients can develop a variety of different tumors in addition to vestibular schwannomas including cranial and spinal meningiomas (fibroblastic subtype), central and peripheral schwannomas, and rarely spinal ependymomas and astrocytomas. Juvenile posterior subcapsular lens opacity are seen in 85% of NF2 patients [57]. The condition is the result of an inherited mutation in one allele of the NF2 gene located on chromosome 22q11.2 [58,59] which encodes the 595 amino acid protein named merlin. Approximately 50% of patients with this condition give no family history of the disease implying a 50% spontaneous mutation rate. Whilst this is a rare condition with an overall incidence of 1/40,000 . Figure 194-3 T2 weighted coronal MRI of multiple neurofibromas resulting in spinal cord compression in a 7-year-old child with NF1
The future of molecular neuro-oncology
individuals [60], patients with this condition are encumbered with significant morbidity and mortality, and current treatment which is primarily surgery, is largely unsatisfactory. Within the cohort of individuals with NF2, there is a group that present at an earlier age with multiple tumors in addition to the VSs with a more aggressive phenotype. This has been described as the Wishart variant of NF2. Those presenting later in life with a less aggressive clinical course are referred to as the Gardner variant [61,62]. Such phenotypic variations have been shown to be correlated with the nature of the inactivating mutation in the NF2 gene. The NF2 gene encodes the protein, Merlin, which is categorized as a member of the 4.1 superfamily of proteins, on the basis of it having an amino-terminal membrane-binding domain (known as the FERM (4.1-ezrin-radixin-moesin) domain). This protein and its relatives have been found in many cell types often concentrated in the nucleus as well as in regions of cell to cell contact. The proposed function of the FERM domain is to facilitate protein-protein interactions. In this way merlin contributes to multimolecular complexes with transmembrane and membrane associated proteins, and these complexes are likely to be important for cellular stability, migration, growth and signal transduction at sites of cell interaction. nf2 mutation studies in the mouse have been informative in regards to the tumor suppressor function of merlin and the conditional knockout has provided an excellent mammalian model for the study of this disease. Mice heterozygous for an nf2 mutation are at risk of developing a variety of malignant tumors later in life (10–30 months) including osteosarcomas, fibrosarcomas and hepatocellular carcinomas of which almost all showed loss of the wild-type allele of NF2 on Southern blotting [63]. Interestingly, most tumors arising in nf2 +/ mice metastasize, which is in contrast to the human phenotype [63]. Metastasis is a rare
194
phenomenon in the mouse. This is likely secondary to the fact that nf2 is close to tp53 in the mouse chromosome 17 and is likely lost in a second mutational event. By comparison, in p53 heterozygote animals only 13% of osteosarcomas metastasized. Also of interest is that no schwannomas, meningiomas or ependymomas were detected in these nf2 +/ animals. Expression of an N-terminal mutant of merlin in transgenic mice under the control of the Schwann cell specific promoter (P0) results in Schwann cell hyperplasia and tumors [64]. Conditional NF2 knockout mice with restricted biallelic NF2 mutation in Schwann cells as determined by the P0 promoter, confirmed that biallelic NF2 inactivation is the rate-limiting step in the formation of murine Schwann cell tumors [65]. No novel therapies have been reported for use in this disease at present.
Tuberous Sclerosis Complex (TSC) Tuberous Sclerosis is a multisystem autosomal dominant syndrome where affected individuals develop hamartomatous changes of the nervous system, skin, eye, kidneys and heart. It occurs with an incidence of 1/6,000 and the CNS lesions include tubers in cerebral cortex, subependymal nodules and subependymal giant cell astrocytomas (SEGA) classically occurring at the foramen of Munro with extension into the ventricular system. The cutaneous manifestations of the disease include hypopigmented ashleaf spots, facial angiofibromas, subungual fibromas and shagreen patches [66]. The clinical presentation of children with this disorder is classically with a seizure disorder and/ or developmental delay or with hydrocephalus secondary to subependymal giant cell astrocytoma (SEGA) although the disease has a broad spectrum of clinical severity. The radiologic findings of multiple subependymal nodules ‘‘candle guttering’’ between the
3209
3210
194
The future of molecular neuro-oncology
thalamus and the caudate are classical on computed tomography (CT). The role for surgery in TSC is in evolution with a suggestion of improved seizure and developmental outcomes in children undergoing tuberectomy at a young age [67–69]. Only a minority of SEGA require excision and most commonly in the setting of obstructive hydrocephalus. The molecular genetics of TSC have been characterized with two genes identified as being responsible in equal significance for this condition. The genes are the TSC1 gene Chr 9q34 [70,71], which encodes a 1,164 amino acid protein named hamartin and the TSC2 gene Chr 16p13 [72,73], which encodes a 1,807 amino acid protein, tuberin. Sixty percent of cases are spontaneous de novo mutations, and of those with familial TSC, 50% of families show linkage to TSC1 and 50% to TSC2. TSC1 and TSC2 mutations manifest nearly identical clinical disease, and it has been postulated that they are components of a single cellular pathway. It is believed that both TSC1 and TSC2 gene products act as tumor suppressor proteins as tumors in this condition show loss of heterozygosity at the TSC locus. Indeed the two proteins,
tuberin and hamartin have been shown to associate and stabilize each others expression. Tuberin normally switches Rap1/Rheb from the active GTP-bound state to its inactive GDPbound state. Tuberin loss of function results in constitutive activation of the protein kinase mTOR and ultimately cell cycle progression. TSC1 and TSC2 mutation mouse models exhibit a similar phenotype to humans with TSC [74,75], developing tubers and both renal and brain tumors. The use of rapamycin in the treatment of SEGA is an excellent example of a novel targeted molecular therapy [76]. Rapamycin binds to and inhibits the ability of mTOR to phosphorylate downstream signaling targets, such as S6Ks, thereby inhibiting cell growth and protein synthesis. There is limited experience of this treatment at present but initial reports are promising (> Figure 194-4a,b).
Von Hippel Lindau Disease Autosomal dominant inheritance is characteristic of VHL, with the phenotypic CNS lesion being hemangioblastoma, seen in association
. Figure 194-4 (a) Intraoperative photograph of an interhemispheric transcallosal approach to a subependymal giant cell astrocytoma (SEGA) in a 5-year-old child with TS (b) T1 weighted, gadolinium enhanced MRI of a SEGA in TS
The future of molecular neuro-oncology
with retinal angiomatosis, pancreatic and epididymal cysts, renal cell carcinoma and pheochromocytoma. VHL is a rare condition with an incidence of 1 in 36,000. Of interest to the clinician, up to 30% of patients with hemangioblastoma will have VHL and 75% of VHL patients will develop a hemangioblastoma [77]. The VHL gene was identified in 1993 [78] and germline mutation at this locus at chromosome 3q25 will result in the syndrome. The presence (type 2; 7–20% of families) or absence (type 1) of pheochromocytomas has been used to further sub classify VHL families. This is reflected in the type of mutation, with most type 2 families having missense mutations, whereas most type 1 families are affected by deletions or premature termination mutations. Prognostic information regarding the risk of pheochromocytoma can be determined by analysis of the underlying mutation in patients without family histories of VHL. CNS hemangioblastoma is a tumor made up of two cell populations, stromal and vascular, with the stromal type thought to represent the neoplasm with the vascular differentiation a reactive process. The VHL protein is involved in the cell’s ability to detect hypoxia. Mutation of the VHL gene results in constitutive activation of this hypoxia detection pathway and hence the excessive vascularity seen in hemangioblastoma. The VHL gene product binds two transcription factors, elongin B and elongin C. Homozygous inactivation of the VHL gene in mice results in embryonic lethality, and the heterozygote animal displayed no phenotype [79]. A conditional VHL knockout mouse has been produced and these mice do develop a number of vascular tumors particularly in the liver [80].
Cowden’s Disease Cowden’s disease is an autosomal dominant condition where individual are at risk of the
194
development of thyroid malignancy, breast cancer, hamartomas and Lhermitte-Duclos disease (LDD). LDD is notable for the presence of dysplastic gangliocytoma of the cerebellum, a hamartomatous lesion that can be responsible for the clinical features of mass effect in the posterior fossa, resulting in hydrocephalus and/ or brainstem compression. The MRI appearance of the lesion is of non-enhancing enlargement of the cerebellar folia and is usually diagnostic. The genetic basis for this condition is a germline mutation in the tumor suppressor gene PTEN on Chromosome 10q23. PTEN gene encodes a lipid phosphatase in the phosphatidylinositol 3-kinase pathway. PTEN indirectly inhibits phosphorylation of Akt, thereby increasing apoptosis. PTEN loss has been identified in malignant gliomas, however patients with Cowden’s do not develop glioblastoma. It is apparent that PTEN loss is a progression mutation in glial carcinogenesis rather than an initiating mutation. The implication of a diagnosis of Cowden’s syndrome is of close clinical and radiologic monitoring for the development of breast malignancy. pten heterozygote mice are predisposed to cancer and develop T cell lymphomas, leukemias, adenocarcinoma of the colon and germ cell tumors [81,82].
Turcot’s Syndrome The coexistence of colonic and CNS neoplasia is the key feature of Turcot’s syndrome which can be the result of a number of different genetic mutations. A germline mutation in the APC tumor suppressor gene on chromosome 5q21, results in the development of multiple (thousands of) colonic polyps and a diagnosis of familial polyposis coli. There is an associated risk of CNS malignancy in this condition, including medulloblastoma, malignant glioma and ependymoma. The APC protein negatively
3211
3212
194
The future of molecular neuro-oncology
regulates Wnt signaling, and its loss results in up regulation of this pathway. Conditional APC knockout mice have been shown to develop colonic adenomas in 4 weeks from the commencement of adenoviral Cre recombinase [83]. Both colonic and CNS malignancy are also seen with germline mutations in DNA mismatch repair genes such as hMLH1, hMSH2, hPMS1 and hPMS2. The loss of these genes results in unstable DNA that accumulates characteristic mutations. Malignant glioma in the setting of DNA mismatch repair mutation is often seen in young adults or children and appears to carry a more favorable prognosis [84].
Gorlin Syndrome/Nevoid Basal Cell Carcinoma Syndrome Gorlin Syndrome is an autosomal dominantly inherited condition where both developmental anomalies (jaw cysts, macrocephaly, rib anomalies and dural calcification) in addition to a predisposition to cancer are seen (multiple basal cell carcinomas, medulloblastoma and meningioma). The underlying mutation is in the PATCHED gene on chromosome 9q22. PATCHED is the receptor for the mitogen Sonic hedgehog. The cells of the external granular cell layer of the cerebellum are responsive to sonic hedgehog mitogenic signals, the putative cells of origin for medulloblastoma. Ptch +/- knockout mice develop cerebellar tumors identical to medulloblastoma in 30% of cases and have many of the features of Gorlin Syndrome including skeletal abnormalities. HSUFU is another inhibitor of the sonic hedgehog pathway and has been documented to be mutated in the germline of a subset of children with medulloblastoma [85]. Novel therapies that inhibit sonic hedgehog signaling have shown some success in the treatment of medulloblastoma in mouse models [86].
Trilateral Retinoblastoma Trilateral retinoblastoma refers to the coexistence of unilateral or bilateral retinoblastoma with an intracranial primitive neuroectodermal tumor (pineoblastoma) in the pineal or suprasellar region [87]. Median survival following diagnosis is 6 months despite maximal therapy which includes chemotherapy and craniospinal irradiation [88]. Retinoblastoma is seen at an early age in children with a germline mutation in the Rb1 gene on chromosome 13q14. Spontaneous mutation in the remaining normal allele of the Rb gene results in disordered regulation of the cell cycle. It is of interest that cells with mutations affecting the Rb pathway do not divide at an accelerated rate, with the mutation contributing to carcinogenesis by providing a survival advantage in the prevention of differentiation or senescence. Individuals with germline Rb mutations are at risk for the development of astrocytomas and sarcomas later in life [89] (> Figure 194-5).
Rubinstein-Taybi Syndrome In this condition individuals display a phenotype of severe developmental delay, abnormal facies with broad thumbs and toes. Associated malignancies include medulloblastoma, oligodendroglioma and leukemias [90]. Germline mutations in Crebs Binding Protein (CBP) on chromosome 16p are responsible, with this large docking protein thought to play a role in Sonic hedgehog signaling, Wnt signaling and p53 signaling.
Melanoma-Astrocytoma Syndrome There are families in which melanoma is seen to occur in addition to astrocytoma and this is
The future of molecular neuro-oncology
. Figure 194-5 T1 weighted, gadolinium enhanced axial MRI of trilateral retinoblastoma in a 2-month-old child presenting with a failure to fix and follow
described as the melanoma-astrocytoma syndrome. A percentage of these families will have germline mutations in the INK4A locus on Chromosome 9.33. This locus encodes both p16 (functions in cell cycle control with Rb) and p14ARF (in p53 pathway). The INK4A locus is frequently mutated in sporadic glioblastoma.
Carney’s Complex In Carney’s complex, an autosomal dominant syndrome, affected individuals are at risk for pituitary tumors, skin pigmentation, cardiac myxomas, and nerve sheath tumors. The peripheral nerve lesions have a characteristic pathology because they are often melanotic. The pituitary tumors often secrete growth hormone. Germline mutations in the PRKAR1A gene on chromosome 17 have been identified in some patients with Carney’s complex. This gene encodes a subunit of the protein kinase A complex [91].
194
. Figure 194-6 T1 weighted, gadolinium enhanced axial MRI of a 3-year-old girl with developmental delay, seizures and Aicardi syndrome showing a choroid plexus papilloma in the occipital horn of the right lateral ventricle
Aicardi Syndrome Aicardi syndrome is a rare X-linked dominant syndrome almost exclusively affecting females. The characteristic features are infantile spasms, corpus callosum agenesis, calcification of the basal ganglia and white matter, demyelination chorioretinal lacunes, and lymphocytosis in the CSF. It has clinical and imaging features that may mimic in utero viral infection. There are now a number of reports of choroid plexus papilloma occurring in patients with documented Aicardi syndrome. These tumors are classically unilateral and have slow insidious growth [92]. Aicardi syndrome can result from mutations in the TREX1 gene, a major mammalian exonuclease and in genes encoding ribonuclease H2 subunits [93,94] (> Figure 194-6).
3213
3214
194
The future of molecular neuro-oncology
Specific Tumor Types Diffuse Cerebral Glioma Glioblastoma Multiforme (WHO grade 4) Glioblastoma (GBM) is the most common malignant primary tumor of the adult CNS. Glioblastoma is seen in a number of inherited syndromes including Li Fraumeni (p53), NF1 (NF1) and Turcot’s (APC), highlighting the complexity of the genetic basis of this disease. Cytogenetic alterations in glioblastoma are many and varied with the most common findings being gain of chromosome 7 and losses of 1p, 9p (INK4A/p16) and 10q (PTEN) [95,96]. Whilst the clinical entities are identical, glioblastoma has been shown to be molecularly distinct on the basis of whether they arise de novo (primary GBM, >90%) or are thought to arise from within a pre-existing low grade glioma, WHO grade II and III (the so-called secondary GBM, 5–10%) [97]. We shall review what is known of the molecular pathogenesis of both primary and secondary GBM with a concentration of the signaling pathways involved in conferring the so-called ‘‘hallmarks’’ of cancer.
Primary and Secondary Glioblastoma With regards to the key growth signaling pathways, direct mutations of Ras have not been identified in glioblastoma with any great frequency. The Ras-Mitogen Activated Protein Kinase (MAPK) pathway is however excessively active secondary to RTK over-expression/stimulation and other regulators. Primary GBM is genetically characterized by amplification of the epidermal growth factor receptor (EGFR, 7p12), seen in up to 50% of tumors [11] and overexpression of EGFR in >60%. GBMs (40%)
express a constitutively active EGFR mutant (D2–7). In secondary glioblastoma, EGFR mutations are rarely seen [98–100] and overexpression is seen in less than 10% [99]. PDGFR amplification and over-expression is also evident in 30% of malignant gliomas [101]. Frequent alteration in the PI3K/AKT-PKB pathway in glioblastoma has been linked to the finding of loss of loci on chromosome 10q seen in malignant gliomas. One copy of chromosome 10 is lost in 70% of tumors and up to 90% of gliomas show heterozygous deletions involving at least part of chromosome 10 [102,103]. PTEN (MMAC (mutated in multiple advanced cancers)) has been identified as a candidate gene and its loss is seen in 15–40% of primary GBMs but is rarely seen in secondary and low grade glioma. PTEN acts as an inhibitor of the PI3Kinase-AKT-PKB cell survival pathway and increased AKT activity is seen in PTEN deficient tumors and cell lines [104]. The p53 pathway plays a crucial role in the development of secondary glioblastoma. Conferring less restricted passage through the cell cycle, genomic instability and evasion of apoptosis, alterations of P53 are seen in two thirds of secondary GBM [105,106] and the importance of this gene in glioma genesis is underlined by the occurrence of this disease in the Li Fraumeni syndrome. The frequency of p53 mutation in primary glioblastoma is less than 30% [107]. In GBM, mutated forms of the P53 gene are found in a region corresponding to the DNA binding domain of the protein. P53 has been shown to be mutated with about equal incidence in low grade (WHO Grade 2) and secondary glioblastoma suggesting that it plays an early role in oncogenesis. Loss of normal p53 function can be the result not only of altered expression of p53 itself but also of the MDM2 or p14ARF genes [108]. The p53 protein is inhibited by the MDM2 gene product, thereby targeting it for destruction. MDM2 amplification and over expression is
The future of molecular neuro-oncology
seen in a subset of astrocytomas that do not have p53 mutations [109]. MDM2 over expression is seen more commonly in primary GBMs as compared with secondary GBMs, the converse of the incidence of p53 mutations. p14ARF is a tumor suppressor gene that is mutated or deleted in a subset of human astrocytomas (>70% [110]). The p14ARF protein blocks the association between MDM2 and p53, thus enhancing the activity of p53. Mutations of p53 and p14ARF appear to be mutually exclusive in human glioblastomas. Mice that have had both alleles of p19arf (the mouse homologue of p14ARF) knocked out spontaneously develop low-grade gliomas in a small percentage of cases. In both primary and secondary glioblastoma, expression of cell cycle regulators (Rb, p16(INK4A) or CDK4) are frequently disordered. The Rb gene is mutated in approximately 25% of high grade glioma thereby allowing unrestricted passage through from G1 to S phase of the cell cycle [111]. Additionally the tumor suppressor p16(INK4A), a cell cycle regulator that inhibits G1 to S phase transition, is mutated with a high incidence (up to 70%) in primary GBM with a similar consequence to Rb loss [112]. p16(INK4A) mutation is rare in secondary GBM (4%) although there was no significant difference in the overall frequency of alterations when promoter methylation is taken into account [110]. CDK4 likewise is amplified in a subgroup of glioblastoma, also promoting G1 to S phase progression. Telomerase expression, conferring limitless replicative potential is seen in the majority of glioblastomas, although rarely in WHO Grade 2 and 3 glioma [40]. Angiogenesis is a further hallmark of glioblastoma development and it has been documented that tumor cells secrete diffusible angiogenic factors that stimulate proliferation and differentiation of endothelial cells. Expression levels of FGF, VEGF, PDGF and angiopoetin have all been shown to be increased and may correlate with higher grade of tumor [113,114].
194
With regards to glioblastoma migration and invasion, components of the extracellular matrix have been found to be modified. For example, tenascin was shown to be over expressed in invasive gliomas in vivo and in a higher concentration around vessels [115,116]. Proteinases such as the MMPs facilitate migration by disrupting the ECM and have also been shown to be up regulated in glioblastoma [117].
Anaplastic Astrocytoma (grade 3 WHO) P53 mutation is seen in 53% of WHO Grade 3 anaplastic astrocytoma, in contrast to EGFR amplification, which is seen in only 15% of these tumors. A mouse model of anaplastic astrocytoma has been developed. nf1 trp 53 mutant mice develop astrocytoma at 100% penetrance [118].
Low Grade Glioma (Grade 1 and 2 WHO) Low grade glioma and pilocytic astrocytoma is seen with a high incidence in NF1. The majority of pediatric pilocytic astrocytomas studied have been shown to have a normal karyotype. Sporadic pilocytic astrocytoma occasionally shows LOH on chromosome 17q but this has been shown not to be loss of NF1 gene. Trisomy of chromosomes 7 and 8 are the most common aberrations seen in pediatric pilocytic astrocytomas whereas adult PA has a variety of cytogenetic abnormalities. The most common cytogenetic abnormality in adult LGG is LOH on chromosome 17p in two thirds [119] and loss of chromosome 22q13.3 in 30% [120]. P53 mutations are absent or infrequently seen in pediatric pilocytic astrocytoma [121] but are seen in >60% of cases of WHO grade
3215
3216
194
The future of molecular neuro-oncology
2 astrocytoma [122,123], similar to rates seen in anaplastic astrocytoma and GBM. Low-grade astrocytomas with p53 mutations may be more likely to progress to glioblastoma than low grade astrocytomas without p53 mutations. This can be explained by the fact that loss of p53 impairs DNA repair resulting in genomic instability. The Ras/MAPK pathway is up regulated in WHO grade 2 astrocytoma by over expression of the platelet derived growth factor (PDGF) and its receptor (PDGFR-alpha) in 50% of cases, representing an autocrine growth loop [119].
Oligodendroglioma Oligodendroglioma is rarely associated with inherited cancer syndromes or reported to occur in families. Chromosomal loss in oligodendroglioma has been reported for 1p, 19q, 4, 14, 15 and 18. Of these, loss of chromosome 1p and 19q is seen in approximately 83% of pure oligodendroglioma [124,125]. This is significant for the fact that allelic loss of chromosome 1p and 19q is a significant predictor of prognosis in oligodendroglioma [126] with median survival of greater than 6 years with LOH 1p/19q as compared to 3–3.5 years without LOH and even poorer in the setting of associated P53 mutation. Epigenetic alterations, such as gene silencing (via promoter methylation of the methyl-guanine methyltransferase (MGMT))gene have been shown to be important in predicting response to alkylating agents in high grade glioma. MGMT expression is decreased in up to 86–93% of oligodendrogliomas and 88% of oligodendrogliomas had MGMT promoter hypermethylation, being more frequent in tumors with 1p/19q LOH. This may explain the link between loss of 1p/19q and good prognosis. Mutation/deletion of the INK4A locus on chromosome 9p, encoding both p16(INK4A) and p14ARF is seen in a subset of oligodendrogliomas and may be more common in anaplastic
oligodendroglioma [127–129]. This may also be associated with poor prognosis and lack of response to chemotherapy [130]. P53 mutations are seldom seen in oligodendroglioma and PTEN loss is described in about 10–20% of cases [131]. Overexpression of EGFR, PDGFR and PDGF is demonstrated in WHO grade 2 oligodendroglioma [125] whereas Grade 3 tumors more frequently have CDKN2A and CDKN2C mutations or deletions, CDK 4, EGFR and MYC amplifications, LOH for 9p and 10q and overexpression of VEGF. A mouse model of the disease has been developed. When mice over-expressing PDGF-B are bred with those with homozygous deletions of INK4a-ARF, the progeny develop malignant oligodendrogliomas [132,133].
Ependymoma Spinal intramedullary ependymomas are frequently seen in NF2, and NF2 gene mutations are seen in sporadic intramedullary spinal ependymoma, however not in intracranial or myxopapillary ependymoma. Ependymoma is also described in Li Fraumeni syndrome, Turcot’s and in the setting of MEN 1. There are several reports of clustering of ependymomas in families [134,135]. The most frequent genetic change in sporadic intracranial ependymoma is loss of chromosome 22q (50–60% of adult patients), suggesting the presence of an ependymoma tumor suppressor gene (other than NF2) on that chromosome. This also appears to correlate with adult disease and spinal location. Losses on chromosomes 6q, 17p, and 16 are frequently reported. Gains of chromosome 7 have been reported in anaplastic ependymomas. Pediatric intracranial ependymoma appears to have a differing cytogenetic profile than the adult intracranial disease with loss of 17p seen in approximately 50% [136] and gain of 1q in 26% [137].
The future of molecular neuro-oncology
There is little evidence for specific genes to be implicated in ependymoma oncogenesis. Microarray experiments of pediatric ependymoma have shown increased expression of the oncogenes, WNT5A and p63, with reduced expression of the NF2-interacting gene SCHIP-1 and APC. ERB-B2 and ERB-B4 expression is seen in 75% of ependymomas. A study demonstrating that ependymoma gene expression recapitulates that of radial glia cells has added insight to the origins of these tumors. For supratentorial tumors, CDK4 and Notch signaling pathway genes were over expressed, whereas for infratentorial tumors, IFG-1 and several HOX homeobox genes were over expressed, and for the spinal tumors, the ID genes and the aquaporins were over expressed [138].
Medulloblastoma Of interest, patients with medulloblastoma are at a 5.4-fold increased risk of secondary cancers including astrocytomas and oligodendrogliomas. Medulloblastoma is seen in association with inherited cancer syndromes such as Gorlin’s (PATCHED), Li Fraumeni (p53) and Turcot’s (APC). There are also reports of monozygotic twins developing medulloblastomas simultaneously. A subset of patients with desmoplastic medulloblastoma will have germline mutations of the Sonic Hedgehog pathway inhibitor Human Suppressor of Fused [85]. This results in up regulation of the sonic hedgehog signaling pathway. Loss of heterozygosity studies reveal allelic loss in 17p 13.3–13.2 in 50% (9/18) of tumors [139]. This region excludes the p53 and ABR (active BCR related) gene. Loss of heterozygosity of chromosome 17p is associated with a significantly worse prognosis in this condition. Gain of chromosome 7 is reported in addition to loss of 1p, 9q, 10q, 11 and 16p [140].
194
Medulloblastoma has been shown to be associated with abnormalities in the Wingless (WNT), Sonic Hedgehog (SHH), Receptor Tyrosine Kinase (ERBB and TrkC), NOTCH, and BMP cellular signaling pathways. The wingless (WNT) pathway is involved in growth of neural progenitors. Mutations in this pathway have been identified in 10% of sporadic medulloblastoma including B-Catenin and Axin1/Axin2, mostly in the classic subtype. APC gene is involved in this pathway, hence the identification of medulloblastoma in Turcot syndrome. PTCH1 is a component of the sonic hedgehog signaling pathway (SHH) and encodes a transmembrane surface receptor that interacts with a similar protein smoothened (SMOH). This interaction results in translocation of Gli proteins into the nucleus where they induce transcription of target genes. Mutations of PTCH have been identified in sporadic medulloblastoma [141–143]. Within the same pathway, mutations of SHH, PTCH2, SMOH and SUFUH are also described. Mouse models are informative with patched +/ mice developing cerebellar tumors very similar to medulloblastoma. The receptor tyrosine kinases are a family of transmembrane receptors that are crucial in the regulation of proliferation, migration and differentiation. The ERB B2 receptor has been shown to be highly expressed in medulloblastoma and is thought to correlate with a higher risk of metastasis and poor prognosis [144]. TRKC, another receptor tyrosine kinase and involved in neurotrophin signaling has been identified as a positive prognostic factor for progression free and overall survival in medulloblastoma [145]. Amplification or over expression of the MYC proto-oncogenes (c-myc and n-myc) has been demonstrated in a subset of medulloblastomas, in particular in the large cell subgroup and is of prognostic significance [146]. HIC-1 (hypermethylated in cancer) encodes a zinc finger transcription factor in the region
3217
3218
194
The future of molecular neuro-oncology
of chromosome 17p highlighted in LOH studies. It is expressed ubiquitously in normal tissues but complete methylation and resultant silencing of this putative tumor suppressor is reported in medulloblastoma [147]. The molecular profiles of medulloblastoma were thought to be distinct for classic (LOH 17p, HIC-1, MYC amplification) and desmoplastic (LOH9q22, PTCH inactivation) tumor subtypes but more recent evidence suggests the presence of at least five subgroups [148].
Atypical Rhabdoid Teratoid Tumor (ATRT) Atypical rhabdoid teratoid tumor (ATRT) is a very aggressive embryonic malignancy that can occur throughout the body but is most commonly seen in the CNS or kidney. It can be differentiated from other primitive pediatric tumors on molecular grounds. Germline mutations in the INI1 gene are identified in a significant percentage of children with multiple or single ATRT, thus representing a familial tumor syndrome. Loss or deletion of long arm of chromosome 22 (22q11) was described in early reports of this tumor raising the possibility of the presence of a tumor suppressor gene at this location (Biegel JA 1990). Deletions on chromosome 11 have also been reported in ATRT. On chromosome 22q11, the Human Sucrose Non-Fermenting 5 gene (hSNF5/INI1) was subsequently identified [149] and 80% of ATRT have genomic mutations of this gene and it is thought to be the key tumor suppressor gene for this tumor. Mutation in the hSNF5 gene has also been identified in choroid plexus carcinomas. The protein functions to regulate the structure of DNA (chromatin) to either allow or deny access of transcription factors to their respective promoters. How this results in ATRT formation
has not been elucidated. Evidence of mutation of hSNF5 alone is not sufficient to make the diagnosis but does appear to be related to outcome and is an important part of the diagnostic workup of these children. Heterozygous SNF5/INI1 +/ mice do develop tumors similar to rhabdoid tumors.
Meningioma Meningiomas may occur in a sporadic fashion, in association with an inherited cancer syndrome such as NF2 or in a delayed fashion following scalp radiation. There are rare reports of meningioma occurring in the setting of Gorlin syndrome and with germline mutation of human Suppressor of Fused, although this is confounded by the administration of radiation to a subset of these patients. The most frequent genetic alteration in sporadic meningioma is loss of heterozygosity (LOH) on chromosome 22, underlying which is the loss of the known tumor suppressor gene NF2. In sporadic meningiomas, loss of NF2 is in the vicinity of 40–60% [150,151], whereas meningiomas found in the setting of NF2 have a detectable incidence of NF2 (chromosome 22q) mutations of greater than 95%. It has been shown that the NF2 status of meningiomas is also associated with tumor localization and histology [152] but not prognosis. Intact NF2 and a meningothelial subtype were shown to correlate with an anterior skull base localization [152]. Loss of 1p, 3p and 14q are also reported in meningioma and associated with more aggressive phenotype. There are other genes of interest on chromosome 22, where meningiomas show loss of heterozygosity for this chromosome and display an intact NF2 gene. Candidates include the SIS oncogene, LARGE, BAM22 and MN1 [153]. Loss of heterozygosity studies have defined chromosome 1p as the second most commonly deleted region
The future of molecular neuro-oncology
in meningiomas. A study of p18 revealed only 1 of 40 meningiomas had evidence of p18 mutation [154]. Loss of heterozygosity on chromosome 10q in 41% of 34 meningiomas is suggestive of the residence of a tumor suppressor gene here but the complete absence of PTEN mutations in this series has raised the question of another TSG in this region [155]. Gene expression profiling using highdensity oligonucleotide microarrays on human meningioma specimens have identified growth hormone receptor, IGFBP-7, endothelin receptor A and IGF2 as having altered expression profiles between tumor and normal controls [156]. Malignancy in meningiomas is rare and the genetic alterations associated with malignant progression have not been elucidated. Recent reports have demonstrated chromosomal aberrations involving 1p, 6q, 10p, 10q, 14q and 18q with particular interest in 1p and the Cdk inhibitors, CDKN2 C, CDKN2 A (p16) and CDKN2 B proteins on chromosome 9 [157,158] as well as p14ARF [159] mutations, which would result in inactivation of the G1/S phase cell cycle checkpoint. Activity of telomerase is seen in most atypical and malignant meningiomas but only rarely in benign lesions. Loss of expression of the 4.1B protein, a member of the protein 4.1 family of proteins including merlin(NF2), ezrin, radixin and moesin, has been described to occur in 60% of sporadic meningiomas but is thought to be important for progression rather than initiation [160,161]. The 4.1B mouse knockout has not been shown to have a pathological phenotype [162]. The absence of NF2 gene mutations in radiation induced meningioma in comparison with the occurrence of such mutations in 50% of sporadic tumors is of interest [163]. Other chromosomal lesions including loss of 1p13 may be more important in this subset of meningiomas. In human meningioma, the role of hormones, growth factors and growth factor receptors is thought to be important in oncogenesis [164].
194
The increased incidence of meningioma in women (2–4 times more common [165]), has suggested a role for sex hormones and their receptors in the etiology of meningioma and numerous studies have analyzed the expression profiles of gonadal steroid hormones and receptors [166]. Progesterone and progesterone receptor amplification may play a role in the regulation of meningioma growth with progesterone receptors identified in 50–100% of tumors and correlating inversely with tumor grade [164,167–171], similarly there are studies suggesting important roles for estrogen and androgen receptors [172]. Of the growth factor receptors, PDGFRb and not a has been detected in meningioma tissue specimens [164]. Dopamine D1, somatostatin and growth hormone receptors have all been analysed in meningioma specimens, with therapeutic interventions under assessment [173–175]. VEGF expression levels have been shown to be highly correlated with the extent of peritumoral edema around a meningioma and the existence of a pial blood supply [176]. A conditional mouse NF2 knockout system (Cre-mediated excision of NF2 exon 2 by adenoviral injection into the CSF) has confirmed the importance of NF2 in meningioma formation showing that from 4 months of age, thirty percent of mice developed a range of meningioma subtypes histologically related to the human tumors [155,177].
Vestibular Schwannoma (sporadic) Acoustic neuroma occurring outside the setting of germline NF2 mutation has a very high incidence of spontaneous mutation in the NF2 gene. Close to 100% of sporadic schwannoma and 50–60% of spontaneous meningiomas [178] have absent merlin expression and this is clearly the key event in the genesis of these lesions [179].
3219
3220
194
The future of molecular neuro-oncology
In addition the Rb-cyclin dependent pathway is deregulated in a high percentage of sporadic tumors [180,181]. There is evidence to suggest that constitutive activation of the neuregulin-1/ErbB signaling pathway is contributory to ongoing proliferation in these tumors [182]. Vestibular schwannoma is not seen in the mice as in the human disease however there are considerable anatomical differences between the species and in particular in the presence or absence of a myelin sheath.
Over expression of c-myc has been reported in approximately 30% of all pituitary adenomas [183]. FGF-4 is overexpresed in some prolactinomas [184]. P53 loss is reported in invasive nonfunctioning adenomas and in the majority of ACTH secreting adenomas [185]. Heterozygote MEN1 mice recapitulate the human MEN-1 syndrome with the development of pituitary tumors in 26% of animals by 16 months [186].
Pituitary Tumors Pituitary tumors make up 10% of all brain tumors although they are rarely seen in hereditary tumor syndromes. Multiple Endocrine Neoplasia type 1 (MEN1) is an autosomal dominantly inherited syndrome where patients develop pituitary tumors, predominantly prolactinomas, in addition to tumors of the parathyroid, pancreatic islets, duodenal endocrine cells. The MEN1 gene has been located on chromosome 11q13 and encodes a 610 AA protein (menin) which is thought to be localized to the nucleus. Evidence of loss of menin expression in sporadic pituitary tumors is minimal. In both Carney complex (CNC1/PRKAR1A, discussed above) and McCune-Albright syndrome the pituitary pathology is hyperplasia rather than adenoma. Individuals with McCuneAlbright Syndrome develop skeletal dysplasias, precocious puberty, thyrotoxicosis, acromegaly, gigantism or Cushing’s syndrome and the gene responsible is GNAS1, the loss of which results in gsp activation and constitutive hormone production. Guanine nucleotide-binding protein (GNAS1) is a known oncogene for adenoma development. Mutations of the GNAS1 gene have been identified in 40% of GH adenomas, 10% of non-functioning adenomas and 6% of ACTH adenomas.
Examples of Molecular Therapies Molecular targeted therapies in clinical use include those that inhibit both the ligand and receptor for growth factors, inhibitors of intracellular signaling cascades, cell migration and invasion, deacetylase and proteosome function. An example of a ligand blocker is bevacizumab. Bevacizumab (Avastin) is a humanized neutralizing monoclonal antibody to VEGF which has shown to result in radiographic response in malignant glioma in 63% of patients when used in combination with irinotecan [187,188]. Inhibition of the VEGF receptor is being studied with a kinase inhibitor, vatalanib with modest results thus far. The tyrosine kinase inhibitor imatanib mesylate (Gleevec) was originally developed for use in the treatment of bcr-abl driven leukemia, although it has activity against PDGFR. PDGFR is known to be over expressed in both high and low grade glioma although clinical trials have shown no effect in monotherapy but perhaps modest benefit when given in association with hydroxyurea in high grade glioma [189,190]. EGFR tyrosine kinase inhibitors with activity against gliomas include gefitinib (Iressa) and erlotinib (Tarceva) but trials reveal only modest impact on progression free or overall survival rates [191].
The future of molecular neuro-oncology
Intracellular growth signaling, in particular the Ras-MAP kinase mitogenic pathway can modified by farnesyl transferase inhibitors such as tipifarnib and lonafarnib. These agents are undergoing investigation against overactive Ras signaling pathways associated with high grade glioma and in optic pathway gliomas in NF1 [192], as sole agents and in combination with chemotherapeutic regimens. The PI3K/AKT/mTOR pathway can also be targeted. A study of the use of mTOR inhibitors such as rapamycin for SEGA in tuberous sclerosis has shown moderate results with disease control in all four patients treated [76]. Temsirolimus, another mTOR inhibitor has been studied in high grade glioma but has not shown a survival benefit [193]. The limited success of single target therapies has led to the investigation of multi-targeted kinase inhibitors such as AEE788, active against EGFR and VEGFR-2, and the use of multiple agents in combination, modifying different signaling pathways. EGFR kinase inhibitors and mTOR inhibitors are undergoing investigation in Phase two trials [194–197]. Targeting glioma cell invasion, cilengitide, an integrin inhibitor, has been shown to be safe in phase one studies and is currently under investigation in phase two studies in high grade glioma [198]. The authors of a recent double blinded, randomized controlled trial of maramistat, a matrix metalloproteinase inhibitor, used in treatment of glioblastoma following surgery and radiotherapy, concluded that there was no improvement in survival or quality of life with this agent [199]. Whilst initial results are modest, improvements should be made with better molecular markers for specific tumor types, rapid identification of the pathways involved in the genesis of these tumors, the development of drugs with less systemic toxicity, the use of multiple agents and the development of better trial protocols to test these interventions.
194
Markers of Response to Treatment and Prognostic Markers There are a number of examples in current clinical practice where genetic analysis of tumors is providing information for clinicians as to the likely response to treatment and prognosis for the patient. Whilst the presence of LOH of Chr 1p and 19q does not appear to significantly influence the response to treatment of oligodendroglioma, the prognosis for patients with these mutations if significantly better than those patients where it is not seen [200]. The presence of epigenetic silencing of the MGMT (O6-methylguanine-DNA methyltransferase) gene by promoter methylation in GBM has been shown to correlate with response to temozolomide chemotherapy [201]. In a study of 220 primary glioblastomas, the most significant factors affecting prognosis were age and performance status, although MDM2 amplification was found to correlate with poor outcome and EGFR amplification with good prognosis [202]. Prognostic markers in medulloblastoma (TrkC – favorable, ERB-B2, C-MYC, LOH 17punfavorable) are becoming increasingly well characterized and may shortly be used in risk stratification around which extent and intensity of therapy can be determined. The large cell anaplastic medulloblastoma may correlate with over expression of ERB-B2 or C-MYC and carry a worse prognosis [203,204]. The presence of B-catenin in the nucleus of medulloblastoma cells correlated with a better prognosis than those patients with cytoplasmic or no B-catenin staining. This finding was shown to correlate with CTNNB1 mutation [205]. Expression of hTERT, conferring limitless replicative potential, in pediatric ependymoma has been shown to correlate with poor 5-year survival (41% with hTERT positive v. 84% with hTERT negative tumors) [206].
3221
3222
194
The future of molecular neuro-oncology
Conclusions This is a rapidly developing field. Many trials of molecular targeted therapies have been commenced in addition to the small but increasing number of proven therapies currently in use. It is anticipated that in the future molecular sub-typing of tumors will be commonplace with treatments tailored to the underlying mutations and the deregulated systems that they are contributing to. The future is brighter than it has ever been in neuro-oncology.
12. 13.
14. 15. 16.
17. 18.
References 1. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000;100(1):57-70. 2. Cho KR, Vogelstein B. Genetic alterations in the adenoma–carcinoma sequence. Cancer 1992;70 (6 Suppl):1727-31. 3. Lehmann AR, et al. Repair of ultraviolet light damage in a variety of human fibroblast cell strains. Cancer Res 1977;37(3):904-10. 4. Blume-Jensen P, Hunter T. Oncogenic kinase signalling. Nature 2001;411(6835):355-65. 5. Kumabe T, et al. Amplification of alpha-platelet-derived growth factor receptor gene lacking an exon coding for a portion of the extracellular region in a primary brain tumor of glial origin. Oncogene 1992;7(4):627-33. 6. Westermark B, Heldin CH, Nister M. Platelet-derived growth factor in human glioma. Glia 1995;15(3):257-63. 7. Braud AC, et al. Overexpression of erb B2 remains a major risk factor in non-metastatic breast cancers treated with high-dose alkylating agents and autologous stem cell transplantation. Bone Marrow Transplant 2002;29 (9):753-7. 8. Malden LT, et al. Selective amplification of the cytoplasmic domain of the epidermal growth factor receptor gene in glioblastoma multiforme. Cancer Res 1988;48 (10):2711-4. 9. Ekstrand AJ, et al. Genes for epidermal growth factor receptor, transforming growth factor alpha, and epidermal growth factor and their expression in human gliomas in vivo. Cancer Res 1991;51(8):2164-72. 10. Nishikawa R, et al. A mutant epidermal growth factor receptor common in human glioma confers enhanced tumorigenicity. Proc Natl Acad Sci USA 1994;91 (16):7727-31. 11. Ekstrand AJ, et al. Amplified and rearranged epidermal growth factor receptor genes in human glioblastomas reveal deletions of sequences encoding portions of the
19.
20.
21. 22.
23. 24. 25.
26.
27.
28.
29.
30.
N- and/or C-terminal tails. Proc Natl Acad Sci USA 1992;89(10):4309-13. Jhiang SM. The RET proto-oncogene in human cancers. Oncogene 2000;19(49):5590-7. Medema RH, Bos JL. The role of p21ras in receptor tyrosine kinase signaling. Crit Rev Oncog 1993;4 (6):615-61. Davies H, et al. Mutations of the BRAF gene in human cancer. Nature 2002;417(6892):949-54. Sherr CJ. Cancer cell cycles. Science 1996;274 (5293):1672-7. Weintraub SJ, et al. Mechanism of active transcriptional repression by the retinoblastoma protein. Nature 1995;375(6534):812-5. Ohi R, Gould KL. Regulating the onset of mitosis. Curr Opin Cell Biol 1999;11(2):267-73. Knudson AG. Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci USA 1971;68 (4):820-3. Kamb A, et al. Analysis of the p16 gene (CDKN2) as a candidate for the chromosome 9p melanoma susceptibility locus. Nat Genet 1994;8(1):23-6. Hannon GJ, Beach D. p15INK4B is a potential effector of TGF-beta-induced cell cycle arrest. Nature 1994;371 (6494):257-61. Vaux DL, Strasser A. The molecular biology of apoptosis. Proc Natl Acad Sci USA 1996;93(6):2239-44. Kaufmann SH. Induction of endonucleolytic DNA cleavage in human acute myelogenous leukemia cells by etoposide, camptothecin, and other cytotoxic anticancer drugs: a cautionary note. Cancer Res 1989;49 (21):5870-8. Wyllie AH, Kerr JF, Currie AR. Cell death: the significance of apoptosis. Int Rev Cytol 1980;68:251-306. Lakin ND, Jackson SP. Regulation of p53 in response to DNA damage. Oncogene 1999;18(53):7644-55. Harper JW, et al. The p21 Cdk-interacting protein Cip1 is a potent inhibitor of G1 cyclin-dependent kinases. Cell 1993;75(4):805-16. Nicol CJ, et al. A teratologic suppressor role for p53 in benzo[a]pyrene-treated transgenic p53-deficient mice. Nat Genet 1995;10(2):181-7. Adachi M, et al. Preferential linkage of bcl-2 to immunoglobulin light chain gene in chronic lymphocytic leukemia. J Exp Med 1990;171(2):559-64. Rampino N, et al. Somatic frameshift mutations in the BAX gene in colon cancers of the microsatellite mutator phenotype. Science 1997;275(5302):967-9. Beerheide W, et al. Downregulation of proapoptotic proteins Bax and Bcl-X(S) in p53 overexpressing hepatocellular carcinomas. Biochem Biophys Res Commun 2000; 273(1):54-61. Landowski TH, et al. Mutations in the Fas antigen in patients with multiple myeloma. Blood 1997;90 (11):4266-70.
The future of molecular neuro-oncology
31. Somia NV, et al. LFG: an anti-apoptotic gene that provides protection from Fas-mediated cell death. Proc Natl Acad Sci USA 1999;96(22):12667-72. 32. Djerbi M, et al. The inhibitor of death receptor signaling, FLICE-inhibitory protein defines a new class of tumor progression factors. J Exp Med 1999;190(7):1025-32. 33. Sato T, et al. FAP-1: a protein tyrosine phosphatase that associates with Fas. Science 1995;268(5209):411-5. 34. Hollstein M, et al. p53 mutations in human cancers. Science 1991;253(5015):49-53. 35. Hollstein M, et al. Database of p53 gene somatic mutations in human tumors and cell lines. Nucleic Acids Res 1994;22(17):3551-5. 36. Foley KP, Eisenman RN. Two MAD tails: what the recent knockouts of Mad1 and Mxi1 tell us about the MYC/ MAX/MAD network. Biochim Biophys Acta 1999;1423 (3):M37-47. 37. Huang ME, et al. Use of all-trans retinoic acid in the treatment of acute promyelocytic leukemia. Blood 1988;72(2):567-72. 38. Bryan TM, et al. Evidence for an alternative mechanism for maintaining telomere length in human tumors and tumor-derived cell lines. Nat Med 1997;3(11):1271-4. 39. Nakatani K, et al. The significant role of telomerase activity in human brain tumors. Cancer 1997;80(3):471-6. 40. Langford LA, et al. Telomerase activity in human brain tumours. Lancet 1995;346(8985):1267-8. 41. Sallinen P, et al. Increased expression of telomerase RNA component is associated with increased cell proliferation in human astrocytomas. Am J Pathol 1997;150 (4):1159-64. 42. Kanamori M, et al. Integrin beta3 overexpression suppresses tumor growth in a human model of gliomagenesis: implications for the role of beta3 overexpression in glioblastoma multiforme. Cancer Res 2004;64 (8):2751-8. 43. Giese A, Westphal M. Glioma invasion in the central nervous system. Neurosurgery 1996;39(2):235-50; discussion 250‐2. 44. Giese A, et al. Cost of migration: invasion of malignant gliomas and implications for treatment. J Clin Oncol 2003;21(8):1624-36. 45. Uhm JH, et al. Mechanisms of glioma invasion: role of matrix-metalloproteinases. Can J Neurol Sci 1997;24 (1):3-15. 46. Fillmore HL, VanMeter TE, Broaddus WC. Membranetype matrix metalloproteinases (MT-MMPs): expression and function during glioma invasion. J Neurooncol 2001;53(2):187-202. 47. Varley JM, et al. Germ-line mutations of TP53 in LiFraumeni families: an extended study of 39 families. Cancer Res 1997;57(15):3245-52. 48. Malkin D, et al. Germ line p53 mutations in a familial syndrome of breast cancer, sarcomas, and other neoplasms. Science 1990;250(4985):1233-8.
194
49. Iwakuma T, Lozano G, Flores ER. Li-Fraumeni syndrome: a p53 family affair. Cell Cycle 2005;4(7):865-7. 50. Gabrilovich DI. INGN 201 (Advexin): adenoviral p53 gene therapy for cancer. Expert Opin Biol Ther 2006;6 (8):823-32. 51. Sharif S, et al. Second primary tumors in neurofibromatosis 1 patients treated for optic glioma: substantial risks after radiotherapy. J Clin Oncol 2006;24(16):2570-5. 52. Jacks T, et al. Tumour predisposition in mice heterozygous for a targeted mutation in Nf1. Nat Genet 1994;7 (3):353-61. 53. Zhu Y, et al. Inactivation of NF1 in CNS causes increased glial progenitor proliferation and optic glioma formation. Development 2005;132(24):5577-88. 54. Widemann BC, et al. Phase I trial and pharmacokinetic study of the farnesyltransferase inhibitor tipifarnib in children with refractory solid tumors or neurofibromatosis type I and plexiform neurofibromas. J Clin Oncol 2006;24 (3):507-16. 55. Gutmann DH. Molecular insights into neurofibromatosis 2. Neurobiol Dis 1997;3(4):247-61. 56. Baser ME, et al. Evaluation of clinical diagnostic criteria for neurofibromatosis 2. Neurology 2002;59 (11):1759-65. 57. Mulvihill JJ, et al. NIH conference. Neurofibromatosis 1 (Recklinghausen disease) and neurofibromatosis 2 (bilateral acoustic neurofibromatosis). An update. Ann Intern Med 1990;113(1):39-52. 58. Trofatter JA, et al. A novel moesin-, ezrin-, radixin-like gene is a candidate for the neurofibromatosis 2 tumor suppressor. Cell 1993;75(4):826. 59. Rouleau GA, et al. Alteration in a new gene encoding a putative membrane-organizing protein causes neurofibromatosis type 2. Nature 1993;363(6429):515-21. 60. Evans DG, et al. A genetic study of type 2 neurofibromatosis in the United Kingdom. I. Prevalence, mutation rate, fitness, and confirmation of maternal transmission effect on severity. J Med Genet 1992;29(12):841-6. 61. Watson CJ, et al. A disease-associated germline deletion maps the type 2 neurofibromatosis (NF2) gene between the Ewing sarcoma region and the leukaemia inhibitory factor locus. Hum Mol Genet 1993;2(6):701-4. 62. Bourn D, et al. A mutation in the neurofibromatosis type 2 tumor-suppressor gene, giving rise to widely different clinical phenotypes in two unrelated individuals. Am J Hum Genet 1994;55(1):69-73. 63. McClatchey AI, et al. Mice heterozygous for a mutation at the Nf2 tumor suppressor locus develop a range of highly metastatic tumors. Genes Dev 1998;12(8):1121-33. 64. Giovannini M, et al. Schwann cell hyperplasia and tumors in transgenic mice expressing a naturally occurring mutant NF2 protein. Genes Dev 1999;13(8):978-86. 65. Giovannini M, et al. Conditional biallelic Nf2 mutation in the mouse promotes manifestations of human neurofibromatosis type 2. Genes Dev 2000;14(13):1617-30.
3223
3224
194
The future of molecular neuro-oncology
66. Weiner DM, et al. The tuberous sclerosis complex: a comprehensive review. J Am Coll Surg 1998;187 (5):548-61. 67. Jansen FE, et al. Epilepsy surgery in tuberous sclerosis: a systematic review. Epilepsia 2007;48(8):1477-84. 68. Jansen FE, et al. Epilepsy surgery in tuberous sclerosis: the Dutch experience. Seizure 2007;16(5):445-53. 69. Madhavan D, et al. Surgical outcome in tuberous sclerosis complex: a multicenter survey. Epilepsia 2007;48 (8):1625-8. 70. Green AJ, Johnson PH, Yates JR. The tuberous sclerosis gene on chromosome 9q34 acts as a growth suppressor. Hum Mol Genet 1994;3(10):1833-4. 71. van Slegtenhorst M, et al. Identification of the tuberous sclerosis gene TSC1 on chromosome 9q34. Science 1997;277(5327):805-8. 72. Kandt RS, et al. Linkage of an important gene locus for tuberous sclerosis to a chromosome 16 marker for polycystic kidney disease. Nat Genet 1992;2(1):37-41. 73. Identification and characterization of the tuberous sclerosis gene on chromosome 16. Cell 1993;75(7): 1305–15. 74. Wilson C, et al. A mouse model of tuberous sclerosis 1 showing background specific early post-natal mortality and metastatic renal cell carcinoma. Hum Mol Genet 2005;14(13):1839-50. 75. Scheidenhelm DK, Gutmann DH. Mouse models of tuberous sclerosis complex. J Child Neurol 2004;19(9): 726-33. 76. Franz DN, et al. Rapamycin causes regression of astrocytomas in tuberous sclerosis complex. Ann Neurol 2006;59(3):490-8. 77. Kaelin WG, Jr, Maher ER. The VHL tumour-suppressor gene paradigm. Trends Genet 1998;14(10):423-6. 78. Latif F, et al. Identification of the von Hippel-Lindau disease tumor suppressor gene. Science 1993;260 (5112):1317-20. 79. Gnarra JR, et al. Defective placental vasculogenesis causes embryonic lethality in VHL-deficient mice. Proc Natl Acad Sci USA 1997;94(17):9102-7. 80. Ma W, et al. Hepatic vascular tumors, angiectasis in multiple organs, and impaired spermatogenesis in mice with conditional inactivation of the VHL gene. Cancer Res 2003;63(17):5320-8. 81. Stambolic V, et al. Negative regulation of PKB/Akt-dependent cell survival by the tumor suppressor PTEN. Cell 1998;95(1):29-39. 82. Di Cristofano A, et al. Pten is essential for embryonic development and tumour suppression. Nat Genet 1998;19 (4):348-55. 83. Shibata H, et al. Rapid colorectal adenoma formation initiated by conditional targeting of the Apc gene. Science 1997;278(5335):120-3. 84. Taylor MD, et al. The hPMS2 exon 5 mutation and malignant glioma. Case report. J Neurosurg 1999;90 (5):946-50.
85. Taylor MD, et al. Mutations in SUFU predispose to medulloblastoma. Nat Genet 2002;31(3):306-10. 86. Sasai K, et al. Medulloblastomas derived from Cxcr6 mutant mice respond to treatment with a smoothened inhibitor. Cancer Res 2007;67(8):3871-7. 87. Paulino AC. Trilateral retinoblastoma: is the location of the intracranial tumor important? Cancer 1999;86 (1):135-41. 88. Kivela T. Trilateral retinoblastoma: a meta-analysis of hereditary retinoblastoma associated with primary ectopic intracranial retinoblastoma. J Clin Oncol 1999;17 (6):1829-37. 89. Strong LC. Genetic implications for long-term survivors of childhood cancer. Cancer 1993;71(10 Suppl): 3435-40. 90. Miller RW, Rubinstein JH. Tumors in Rubinstein-Taybi syndrome. Am J Med Genet 1995;56(1):112-5. 91. Kirschner LS, et al. Mutations of the gene encoding the protein kinase A type I-alpha regulatory subunit in patients with the Carney complex. Nat Genet 2000;26 (1):89-92. 92. Frye RE, Polling JS, Ma LC. Choroid plexus papilloma expansion over 7 years in Aicardi syndrome. J Child Neurol 2007;22(4):484-7. 93. Crow YJ, et al. Mutations in the gene encoding the 30 -50 DNA exonuclease TREX1 cause Aicardi-Goutieres syndrome at the AGS1 locus. Nat Genet 2006;38 (8):917-20. 94. Crow YJ, et al. Mutations in genes encoding ribonuclease H2 subunits cause Aicardi-Goutieres syndrome and mimic congenital viral brain infection. Nat Genet 2006;38(8):910-6. 95. Mohapatra G, et al. Genetic analysis of glioblastoma multiforme provides evidence for subgroups within the grade. Genes Chromosomes Cancer 1998;21(3): 195-206. 96. Mao X, Hamoudi RA. Molecular and cytogenetic analysis of glioblastoma multiforme. Cancer Genet Cytogenet 2000;122(2):87-92. 97. Dropcho EJ, Soong SJ. The prognostic impact of prior low grade histology in patients with anaplastic gliomas: a case-control study. Neurology 1996;47(3): 684-90. 98. Biernat W, et al. TP53 mutations in malignant astrocytomas. Pol J Pathol 1997;48(4):221-4. 99. Watanabe K, et al. Overexpression of the EGF receptor and p53 mutations are mutually exclusive in the evolution of primary and secondary glioblastomas. Brain Pathol 1996;6(3):217-23; discussion 23–4. 100. Ivanchuk SM, et al. The INK4A/ARF locus: role in cell cycle control and apoptosis and implications for glioma growth. J Neurooncol 2001;51(3):219-29. 101. Joensuu H, et al. Amplification of genes encoding KIT, PDGFRalpha and VEGFR2 receptor tyrosine kinases is frequent in glioblastoma multiforme. J Pathol 2005;207 (2):224-31.
The future of molecular neuro-oncology
102. Bigner SH, Vogelstein B. Cytogenetics and molecular genetics of malignant gliomas and medulloblastoma. Brain Pathol 1990;1(1):12-8. 103. Somerville RP, et al. Molecular analysis of two putative tumour suppressor genes, PTEN and DMBT, which have been implicated in glioblastoma multiforme disease progression. Oncogene 1998;17(13):1755-7. 104. Cantley LC, Neel BG. New insights into tumor suppression: PTEN suppresses tumor formation by restraining the phosphoinositide 3-kinase/AKT pathway. Proc Natl Acad Sci USA 1999;96(8):4240-5. 105. Srivastava S, et al. Germ-line transmission of a mutated p53 gene in a cancer-prone family with Li-Fraumeni syndrome. Nature 1990;348(6303):747-9. 106. Rutka JT, et al. Alterations of the p53 and pRB pathways in human astrocytoma. Brain Tumor Pathol 2000;17 (2):65-70. 107. Ohgaki H, et al. Genetic pathways to glioblastoma: a population-based study. Cancer Res 2004;64 (19):6892-9. 108. Ohgaki H, Kleihues P. Genetic pathways to primary and secondary glioblastoma. Am J Pathol 2007;170(5): 1445-53. 109. Reifenberger G, et al. Amplification and overexpression of the MDM2 gene in a subset of human malignant gliomas without p53 mutations. Cancer Res 1993;53 (12):2736-9. 110. Nakamura M, et al. p14ARF deletion and methylation in genetic pathways to glioblastomas. Brain Pathol 2001;11 (2):159-68. 111. Behin A, et al. Primary brain tumours in adults. Lancet 2003;361(9354):323-31. 112. Jen J, et al. Deletion of p16 and p15 genes in brain tumors. Cancer Res 1994;54(24):6353-8. 113. Zadeh G, et al. Role of Ang1 and its interaction with VEGF-A in astrocytomas. J Neuropathol Exp Neurol 2004;63(9):978-89. 114. Zadeh G, et al. Targeting the Tie2/Tek receptor in astrocytomas. Am J Pathol 2004;164(2):467-76. 115. Leins A, et al. Expression of tenascin-C in various human brain tumors and its relevance for survival in patients with astrocytoma. Cancer 2003;98(11): 2430-9. 116. Zagzag D, et al. Tenascin expression in astrocytomas correlates with angiogenesis. Cancer Res 1995;55(4): 907-14. 117. Demuth T, Berens ME. Molecular mechanisms of glioma cell migration and invasion. J Neurooncol 2004;70 (2):217-28. 118. Reilly KM, et al. Nf1;Trp53 mutant mice develop glioblastoma with evidence of strain-specific effects. Nat Genet 2000;26(1):109-13. 119. Hermanson M, et al. Association of loss of heterozygosity on chromosome 17p with high platelet-derived growth factor alpha receptor expression in human malignant gliomas. Cancer Res 1996;56(1):164-71.
194
120. Oskam NT, Bijleveld EH, Hulsebos TJ. A region of common deletion in 22q13.3 in human glioma associated with astrocytoma progression. Int J Cancer 2000;85(3):336-9. 121. Cheng Y, et al. Pilocytic astrocytomas do not show most of the genetic changes commonly seen in diffuse astrocytomas. Histopathology 2000;37(5):437-44. 122. Huang H, et al. Gene expression profiling of low-grade diffuse astrocytomas by cDNA arrays. Cancer Res 2000;60(24):6868-74. 123. Reifenberger J, et al. Analysis of p53 mutation and epidermal growth factor receptor amplification in recurrent gliomas with malignant progression. J Neuropathol Exp Neurol 1996;55(7):822-31. 124. Smith JS, et al. Localization of common deletion regions on 1p and 19q in human gliomas and their association with histological subtype. Oncogene 1999;18 (28):4144-52. 125. Reifenberger J, et al. Molecular genetic analysis of oligodendroglial tumors shows preferential allelic deletions on 19q and 1p. Am J Pathol 1994;145(5):1175-90. 126. Cairncross JG, et al. Specific genetic predictors of chemotherapeutic response and survival in patients with anaplastic oligodendrogliomas. J Natl Cancer Inst 1998;90(19):1473-9. 127. Bigner SH, et al. Molecular genetic aspects of oligodendrogliomas including analysis by comparative genomic hybridization. Am J Pathol 1999;155(2):375-86. 128. Bigner SH, et al. Morphologic and molecular genetic aspects of oligodendroglial neoplasms. Neuro Oncol 1999;1(1):52-60. 129. Bortolotto S, et al. CDKN2A/p16 inactivation in the prognosis of oligodendrogliomas. Int J Cancer 2000;88 (4):554-7. 130. Miettinen H, et al. CDKN2/p16 predicts survival in oligodendrogliomas: comparison with astrocytomas. J Neurooncol 1999;41(3):205-11. 131. Sasaki H, et al. PTEN is a target of chromosome 10q loss in anaplastic oligodendrogliomas and PTEN alterations are associated with poor prognosis. Am J Pathol 2001;159(1):359-67. 132. Dai C, et al. PDGF autocrine stimulation dedifferentiates cultured astrocytes and induces oligodendrogliomas and oligoastrocytomas from neural progenitors and astrocytes in vivo. Genes Dev 2001;15 (15):1913-25. 133. Tchougounova E, et al. Loss of Arf causes tumor progression of PDGFB-induced oligodendroglioma. Oncogene 2007;26(43):6289-96. 134. Honan WP, et al. Familial subependymomas. Br J Neurosurg 1987;1(3):317-21. 135. Dimopoulos VG, Fountas KN, Robinson JS. Familial intracranial ependymomas. Report of three cases in a family and review of the literature. Neurosurg Focus 2006;20(1):E8. 136. von Haken MS, et al. Molecular genetic analysis of chromosome arm 17p and chromosome arm 22q DNA
3225
3226
194 137.
138. 139.
140.
141.
142.
143. 144.
145.
146.
147.
148.
149.
150.
151.
152.
The future of molecular neuro-oncology
sequences in sporadic pediatric ependymomas. Genes Chromosomes Cancer 1996;17(1):37-44. Dyer S, et al. Genomic imbalances in pediatric intracranial ependymomas define clinically relevant groups. Am J Pathol 2002;161(6):2133-41. Taylor MD, et al. Radial glia cells are candidate stem cells of ependymoma. Cancer Cell 2005;8(4):323-35. Steichen-Gersdorf E, et al. Deletion mapping on chromosome 17p in medulloblastoma. Br J Cancer 1997;76 (10):1284-7. Bayani J, et al. Molecular cytogenetic analysis of medulloblastomas and supratentorial primitive neuroectodermal tumors by using conventional banding, comparative genomic hybridization, and spectral karyotyping. J Neurosurg 2000;93(3):437-48. Vorechovsky I, et al. Somatic mutations in the human homologue of Drosophila patched in primitive neuroectodermal tumours. Oncogene 1997;15(3):361-6. Wolter M, et al. Mutations in the human homologue of the Drosophila segment polarity gene patched (PTCH) in sporadic basal cell carcinomas of the skin and primitive neuroectodermal tumors of the central nervous system. Cancer Res 1997;57(13):2581-5. Raffel C, et al. Sporadic medulloblastomas contain PTCH mutations. Cancer Res 1997;57(5):842-5. Gilbertson RJ, et al. Expression of the ErbB-neuregulin signaling network during human cerebellar development: implications for the biology of medulloblastoma. Cancer Res 1998;58(17):3932-41. Grotzer MA, et al. TrkC expression predicts good clinical outcome in primitive neuroectodermal brain tumors. J Clin Oncol 2000;18(5):1027-35. Eberhart CG, et al. Histopathological and molecular prognostic markers in medulloblastoma: c-myc, N-myc, TrkC, and anaplasia. J Neuropathol Exp Neurol 2004; 63(5):441-9. Rood BR, et al. Hypermethylation of HIC-1 and 17p allelic loss in medulloblastoma. Cancer Res 2002;62 (13):3794-7. Thompson MC, et al. Genomics identifies medulloblastoma subgroups that are enriched for specific genetic alterations. J Clin Oncol 2006;24(12):1924-31. Versteege I, et al. Truncating mutations of hSNF5/INI1 in aggressive paediatric cancer. Nature 1998;394(6689): 203-6. Ng HK, et al. Combined molecular genetic studies of chromosome 22q and the neurofibromatosis type 2 gene in central nervous system tumors. Neurosurgery 1995;37 (4):764-73. Gutmann DH, et al. Loss of merlin expression in sporadic meningiomas, ependymomas and schwannomas. Neurology 1997;49(1):267-70. Kros J, et al. NF2 status of meningiomas is associated with tumour localization and histology. J Pathol 2001; 194(3):367-72.
153. Lekanne Deprez RH, et al. Cloning and characterization of MN1, a gene from chromosome 22q11, which is disrupted by a balanced translocation in a meningioma. Oncogene 1995;10(8):1521-8. 154. Santarius T, et al. Molecular analysis of alterations of the p18INK4c gene in human meningiomas. Neuropathol Appl Neurobiol 2000;26(1):67-75. 155. Bostrom J, et al. Mutation of the PTEN (MMAC1) tumor suppressor gene in a subset of glioblastomas but not in meningiomas with loss of chromosome arm 10q. Cancer Res 1998;58(1):29-33. 156. Watson MA, et al. Molecular characterization of human meningiomas by gene expression profiling using highdensity oligonucleotide microarrays. Am J Pathol 2002;161(2):665-72. 157. Perry A, et al. A role for chromosome 9p21 deletions in the malignant progression of meningiomas and the prognosis of anaplastic meningiomas. Brain Pathol 2002;12 (2):183-90. 158. Simon M, et al. Alterations of INK4a(p16-p14ARF)/ INK4b(p15) expression and telomerase activation in meningioma progression. J Neurooncol 2001;55(3): 149-58. 159. Bostrom J, et al. Alterations of the tumor suppressor genes CDKN2A (p16(INK4a)), p14(ARF), CDKN2B (p15(INK4b)), and CDKN2C (p18(INK4c)) in atypical and anaplastic meningiomas. Am J Pathol 2001;159 (2):661-9. 160. Gutmann DH, et al. Loss of DAL-1, a protein 4.1-related tumor suppressor, is an important early event in the pathogenesis of meningiomas. Hum Mol Genet 2000;9 (10):1495-500. 161. Nunes F, et al. Inactivation patterns of NF2 and DAL-1/ 4.1B (EPB41L3) in sporadic meningioma. Cancer Genet Cytogenet 2005;162(2):135-9. 162. Yi C, et al. Loss of the putative tumor suppressor band 4.1B/Dal1 gene is dispensable for normal development and does not predispose to cancer. Mol Cell Biol 2005;25 (22):10052-9. 163. Shoshan Y, et al. Radiation-induced meningioma: a distinct molecular genetic pattern? J Neuropathol Exp Neurol 2000;59(7):614-20. 164. Black P, Carroll R, Zhang J. The molecular biology of hormone and growth factor receptors in meningiomas. Acta Neurochir Suppl (Wien) 1996;65:50-3. 165. Kleihues P, Sobin LH. World Health Organization classification of tumors. Cancer 2000;88(12):2887. 166. Black PM, Meningiomas. Neurosurgery 1993;32: 643-657. 167. Inoue T, et al. Progesterone production and actions in the human central nervous system and neurogenic tumors. J Clin Endocrinol Metab 2002;87(11): 5325-31. 168. Fewings PE, Battersby RD, Timperley WR. Long-term follow up of progesterone receptor status in benign
The future of molecular neuro-oncology
169.
170.
171.
172. 173.
174.
175.
176.
177.
178.
179.
180.
181. 182.
183.
184. 185.
meningioma: a prognostic indicator of recurrence? J Neurosurg 2000;92(3):401-5. Hsu DW, Efird JT, Hedley-Whyte ET. Progesterone and estrogen receptors in meningiomas: prognostic considerations. J Neurosurg 1997;86(1):113-20. Whittle IR, Hawkins RA, Miller JD. Sex hormone receptors in intracranial tumours and normal brain. Eur J Surg Oncol 1987;13(4):303-7. Whittle IR, et al. Progesterone and oestrogen receptors in meningiomas: biochemical and clinicopathological considerations. Aust N Z J Surg 1984;54(4):325-30. Smith DA, Cahill DW. The biology of meningiomas. Neurosurg Clin N Am 1994;5(2):201-15. Schrell UM, Nomikos P, Fahlbusch R. Presence of dopamine D1 receptors and absence of dopamine D2 receptors in human cerebral meningioma tissue. J Neurosurg 1992;77(2):288-94. Carroll RS, et al. Dopamine D1, dopamine D2, and prolactin receptor messenger ribonucleic acid expression by the polymerase chain reaction in human meningiomas. Neurosurgery 1996;38(2):367-75. Friend KE, Radinsky R, McCutcheon IE. Growth hormone receptor expression and function in meningiomas: effect of a specific receptor antagonist. J Neurosurg 1999;91(1):93-9. Otsuka S, et al. The relationship between peritumoral brain edema and the expression of vascular endothelial growth factor and its receptors in intracranial meningiomas. J Neurooncol 2004;70(3):349-57. Kalamarides M, et al. Nf2 gene inactivation in arachnoidal cells is rate-limiting for meningioma development in the mouse. Genes Dev 2002;16(9):1060-5. Gutmann DH, Hirbe AC, Haipek CA. Functional analysis of neurofibromatosis 2 (NF2) missense mutations. Hum Mol Genet 2001;10(14):1519-29. Sainz J, et al. Mutations of the neurofibromatosis type 2 gene and lack of the gene product in vestibular schwannomas. Hum Mol Genet 1994;3(6):885-91. Lasak JM, et al. Retinoblastoma-cyclin-dependent kinase pathway deregulation in vestibular schwannomas. Laryngoscope 2002;112(9):1555-61. Welling DB, et al. cDNA microarray analysis of vestibular schwannomas. Otol Neurotol 2002;23(5):736-48. Hansen MR, et al. Constitutive neuregulin-1/ErbB signaling contributes to human vestibular schwannoma proliferation. Glia 2006;53(6):593-600. Ezzat S, Yu S, Asa SL. The zinc finger Ikaros transcription factor regulates pituitary growth hormone and prolactin gene expression through distinct effects on chromatin accessibility. Mol Endocrinol 2005;19(4):1004-11. Asa SL, Ezzat S. The cytogenesis and pathogenesis of pituitary adenomas. Endocr Rev 1998;19(6):798-827. Vallar L, Spada A, Giannattasio G. Altered Gs and adenylate cyclase activity in human GH-secreting pituitary adenomas. Nature 1987;330(6148):566-8.
194
186. Williamson EA, et al. G-protein mutations in human pituitary adrenocorticotrophic hormone-secreting adenomas. Eur J Clin Invest 1995;25(2):128-31. 187. Vredenburgh JJ, et al. Phase II trial of bevacizumab and irinotecan in recurrent malignant glioma. Clin Cancer Res 2007;13(4):1253-9. 188. Vredenburgh JJ, et al. Bevacizumab plus irinotecan in recurrent glioblastoma multiforme. J Clin Oncol 2007;25(30):4722-9. 189. Dresemann G. Imatinib and hydroxyurea in pretreated progressive glioblastoma multiforme: a patient series. Ann Oncol 2005;16(10):1702-8. 190. Reardon DA, et al. Phase II study of imatinib mesylate plus hydroxyurea in adults with recurrent glioblastoma multiforme. J Clin Oncol 2005;23(36):9359-68. 191. Rich JN, et al. Phase II trial of gefitinib in recurrent glioblastoma. J Clin Oncol 2004;22(1):133-42. 192. Kesari S, et al. Targeted molecular therapy of malignant gliomas. Curr Neurol Neurosci Rep 2005;5(3):186-97. 193. Galanis E, et al. Phase II trial of temsirolimus (CCI-779) in recurrent glioblastoma multiforme: a North Central Cancer Treatment Group Study. J Clin Oncol 2005;23 (23):5294-304. 194. Reardon DA, et al. Phase 1 trial of gefitinib plus sirolimus in adults with recurrent malignant glioma. Clin Cancer Res 2006;12(3)(Pt 1):860-8. 195. Reardon DA, et al. Recent advances in the treatment of malignant astrocytoma. J Clin Oncol 2006;24(8): 1253-65. 196. Reardon DA, Wen PY. Therapeutic advances in the treatment of glioblastoma: rationale and potential role of targeted agents. Oncologist 2006;11(2):152-64. 197. Doherty L, et al. Pilot study of the combination of EGFR and mTOR inhibitors in recurrent malignant gliomas. Neurology 2006;67(1):156-8. 198. Nabors LB, Targeted molecular therapy for malignant gliomas. Curr Treat Options Oncol 2004;5(6):519-26. 199. Levin VA, et al. Randomized, double-blind, placebocontrolled trial of marimastat in glioblastoma multiforme patients following surgery and irradiation. J Neurooncol 2006;78(3):295-302. 200. Cairncross G, et al. Phase III trial of chemotherapy plus radiotherapy compared with radiotherapy alone for pure and mixed anaplastic oligodendroglioma: Intergroup Radiation Therapy Oncology Group Trial 9402. J Clin Oncol 2006;24(18):2707-14. 201. Hegi ME, et al. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med 2005;352 (10):997-1003. 202. Houillier C, et al. Prognostic impact of molecular markers in a series of 220 primary glioblastomas. Cancer, 2006. 106(10):2218-23. 203. Eberhart CG, et al. Histopathologic grading of medulloblastomas: a Pediatric Oncology Group study. Cancer 2002;94(2):552-60.
3227
3228
194
The future of molecular neuro-oncology
204. Eberhart CG, et al. Comparative genomic hybridization detects an increased number of chromosomal alterations in large cell/anaplastic medulloblastomas. Brain Pathol 2002;12(1):36-44. 205. Ellison DW, et al. Beta-catenin status predicts a favorable outcome in childhood medulloblastoma: the United
Kingdom Children’s Cancer Study Group Brain Tumour Committee. J Clin Oncol 2005;23(31):7951-7. 206. Tabori U, et al. Human telomere reverse transcriptase expression predicts progression and survival in pediatric intracranial ependymoma. J Clin Oncol 2006;24(10): 1522-8.
193 The Future of Neural Interface Technology M. C. Park . M. A. Goldman . T. W. Belknap . G. M. Friehs
Introduction The concept of an interface between brain and the external world, i.e., connecting a brain to a machine or a computer, is not new and often has been the subject of popular science fiction. As early as the 1950s, electrodes were regularly implanted into the brains of humans or animals for electrical recording or stimulation [1] to influence brain function or treat neurological disorders [2,3]. For many decades, neurophysiologists have also been recording neural electrical potentials and stimulating through similar types of acutely and chronically implanted electrodes, as tools for understanding brain function. However, the specific goal of interfacing brain directly to devices, bypassing normal routes to the muscles, has recently experienced a resurgence in popularity and received renewed interest as seen in both research publications and in the popular media. Such rise in interest may be due to the recent advances in neuroscience as well as rapid developments in computers and electronics, as predicted by Moore in 1965 [4,5], allowing large amounts of information to be processed and converted into neurallyderived control signals in real-time. The concept of transforming thought into action and sensation into perception for those lacking normal pathways is now becoming feasible. Central nervous system prostheses, known as brain-machine interfaces (BMIs), brain-computer interfaces (BCIs), or, more generally, neural interface (NI) systems, designed to establish a new communication link between a functioning brain and the outside world [6], can be usefully divided into three categories. The first category is #
Springer-Verlag Berlin/Heidelberg 2009
the input NI; those devices that are used to replace or restore lost communication to the brain, i.e., lost sensory input, by replacing the physical stimuli transduction of primary sensory cells with electrical stimulation of either sensory nerve fibers or a central pathway directly. The second category is the output NI; those devices that are used to replace or restore lost communication from the brain, for example, to the muscles when cerebral connections to the muscles are damaged. Output NI could, for instance, extract cortical motor command signals that could be used to control limb prostheses, robotic equipment, or a computer cursor. An extension of this concept would use motor signals to reanimate the muscles. One manner of restoring brain-tomuscle communication being explored is the use of cortical signals to control a functional electrical stimulation (FES) system. FES is an approved technology that stimulates muscles through a fully implanted stimulator, coupled by wires to the muscles. The third category, which is a more speculative and a very early stage NI system, can be described as a cognitive or bypass NI. Cognitive NIs are unique in that they are intended to replace communication between disconnected neural structures. One current example is the hippocampal-cortical prosthesis attempting to reestablish temporal lobe memory pathways [7]. More than 2 million people in the United States suffer from movement disorders [8], and approximately 12,000 spinal cord injuries occur each year [9], resulting in a significant loss of productivity, and in high health care and living expenses, for individuals and society [9,10]. Also, many members of the United States Armed
3186
193
The future of neural interface technology
Forces are now returning from the conflicts abroad with traumatic amputations and a wide range of neurological injuries, resulting in movement impairments [11]. Thus, many people suffer from forms of motor paralysis in which cortical motor areas are intact, but cannot generate movement [12]. The goal of an output NI system is to provide a neuromotor prosthesis (NMP) thus restoring the lost communication between brain and the external world, providing control and communication, and hence restoring independence. These signals may connect directly to assistive technologies or back to muscles, to provide a form of replacement motor system. As far reaching as these concepts may appear, the initial proof of concept both in animal models and in humans with tetraplegia has already been established. Although there are many hurdles remaining until this technology is commonly available, it is on a path where a succession of increasingly sophisticated devices will emerge. In order to describe the future of this promising neurotechnological advance, this chapter will first describe the concept of NI, its history, its current status and application in humans, and then will provide a view of what may appear in the coming decades.
Background and Review of Literature Developments in Electrophysiology Once felt to only exist in the realm of science fiction, NI systems have become a plausible reality in the latter part of the last century. The essential platform for an NI may have begun as early as 1863 when John Hughlings Jackson first proposed that cerebral cortex contained localized regions that controlled body musculature to produce coordinated movements, contrary to the commonly accepted contemporary belief that
cerebrum was reserved for cognition [13,14]. Subsequently, Fritsch and Hitzig in 1870 demonstrated the electrical excitability of the cerebral cortex when they evoked movements by stimulating the cerebral cortex of a dog with galvanic currents [15,16]. This was later followed by the landmark works on the localization of motor cortex by Ferrier [17] and Sherrington [18], who demonstrated the motor map in nonhuman primates, and by Penfield and Boldrey [19] in humans. Penfield’s map came to be the famous – albeit, as they warned, grossly oversimplified – human motor homunculus figure in virtually every neuroscience textbook [14,20]. In 1952, Delgado advanced the concept of permanently placed single or multiple electrodes in the brains of animals and then humans, to study the physiological basis for psychosurgical procedures [1], leading to predictions of NIs for therapy and restoration of function. Jasper recorded from neurons in awake monkeys performing tasks [21], which was later utilized by Evarts in studying the properties of neurons in awake monkeys performing voluntary movements [14,22,23]. This heralded a groundbreaking move towards direct sensing of the details of motor signaling at the single cell level, rather than at the macro scale, field potential, electroencephalogram (EEG) scale. Since then, numerous single and multiple microelectrode electrophysiological techniques have been developed for both animal and human research to better understand neural activity at the single cell and cell-ensemble level of motor cortical areas [24–32]. After a period of little activity, there was an upsurge in interest and advances in moving towards useful NI systems based on directly sampling motor signals from motor areas that generate movement command signals [6,33,34], which is the emphasis of this chapter. A number of studies, by using signals recorded from a population of neurons, showed that neuroprosthetic devices could be controlled and used to perform tasks, directly from neural spiking activity in
The future of neural interface technology
motor cortical areas. The work by Chapin et al. [35] demonstrated that rats could learn to use motor cortex population activity to control a single joint mechanical arm. Major steps in NI development came with the demonstration that it was possible to perform two and three dimensional, goal-directed control of a computer cursor using neurons in primary motor cortex (M1) in monkeys [36–48] and shortly afterwards in humans with paralysis [39–41]. The potential to use these pilot stage systems to control robotics arms and assistive technologies such as a TV remote control was also shown. The potential for other areas to provide control signals was also demonstrated [42].
Developments in Technology These advances demanded a new type of multielectrode array technology, which faced substantial challenges of remaining capable of safely recording very small electrical spiking potentials for many years. Several designs are being evaluated (see [6] for discussion and references), but two have been successfully applied for initial human testing. These include a 44 mm platform of 100 microelectrodes for intracortical placement at a cortical surface location [27] and cone electrodes [40]. Animal evaluation over many months (in one case >1 year [43]) indicated that many electrodes in the platform system can successfully record over long periods of time, as did cone electrodes, which will be discussed later in detail in the sections to follow. The realization and implementation of NI also benefited from the rapid development in computer technology. Processing of large number of channels of spiking data quickly enough to be used in real-time requires very fast computers, which are now available. Assistive technology currently available to patients have also benefited from technological and manufacturing advances, i.e., motorized wheelchairs, wireless devices that
193
communicate to virtually any electrically operated device, and light-weight composite material artificial limbs.
Concept of Neural Interface Terms and Definitions With increase and resurgence of interest in NI research, as evidenced by their growing presence in scientific journals, numerous terms have emerged to describe the integration between brain and devices, both physical and biological. However, there is yet to be a consensus among researchers on a common nomenclature, although efforts are underway [44]. The following list of terms is neither definitive nor complete but rather is commonly found in recently published literature. Neural interface system (NIS) ‘‘couples the nervous system to a device that may either stimulate tissue or record neural activity, or perform both in a closed loop system’’ and relies on ‘‘successful sensing of neural activity to provide a command signal to control computers, machines, or any of a range of prosthetic devices that span from physical to biological elements’’ [45]. NIS devices that ‘‘transform a neurally based motor intention into a command signal that can operate physical systems’’ are often called BCI or BMI [45]. Neuroprosthesis (NP) or NMP is a type of NIS, intended to provide movement commands from the brain, that can ‘‘guide movement by harnessing the existing neural substrate for that action – that is, neuronal activity patterns in motor areas’’ [46]. NMP acts to interpret brain signals and drive the appropriate effectors such as muscles or a robotic arm. However, it is also important to note that NIs do not activate alternate pathways producing anatomical compensation or truly restore the structural lesions to its original state or full anatomical recovery, but rather help in restoring the lost function, thus achieving functional recovery [47].
3187
3188
193
The future of neural interface technology
Input NI Input NI describes a type of neural interface in which a device coupled to a nervous system is used to stimulate the tissue. These devices may be used to replace lost sensory input to the brain or to modulate brain function to modify brain in abnormal functional states [6]. Devices which reestablish sensory communication to the brain, by replacing the transduction of physical stimuli by primary sensory cells with electrical stimulation of either sensory nerve fibers or sensory cortex directly, already exist and such input NI devices are widely accepted and in use today. The most successful input NI device to date is arguably the cochlear implant (CI), which is use by more than 100,000 people to restore hearing. The CI converts the sound waves into electrical signals interpreted by the brain thus addressing the loss of hearing [48,49]. Devices that bypass the cochlea for hearing include CI, auditory brainstem implant (ABI) and auditory midbrain implant (AMI) [50]. However, unlike the commercially available CI, these more central devices are at a much earlier stage of testing. The AMI, a recent advance, utilizes a penetrating electrode array to stimulate auditory structures in the midbrain, that reportedly has more success in restoring hearing by providing loudness, pitch, temporal, and directional cues [51]. Stimulation along visual pathways, at the retina, nerve or cortex, is being used to restore sight, especially for those with macular degeneration and retinitis pigmentosa, the leading causes of vision loss [52–54]. In 1968, Brindley and Lewin published their findings on their attempt of ‘‘making a useful visual prosthesis’’ by implanting a stimulator array into the medial occipital cortex of a 52-year-old female who had ‘‘little more than perception of light’’ [55]. The platinum electrodes, arranged in an array of 80, when stimulated with short pulses of radio waves, elicited phosphenes, small spots of light seen at a constant position in the visual field [55].
In 2000, Dobelle reported on a visual prosthesis designed to provide useful ‘‘artificial vision’’ to a blind patient with a microstimulation electrode array implanted into the occipital cortex. The patient was connected to a digital video camera mounted on sunglasses and a computer [56]. Veraart and colleagues reported on the feasibility and safety of chronically implanted optic nerve visual prosthesis [57]. FES above or below the retina is in development [58] and one system is beyond the pilot trial stage [53]. Normal motor control is heavily dependent on sensory feedback, and the somatic sensory system plays a major role in this function. Thus a fully functional output NI would also be able to convey somatic sensory information to the cortex that is indistinguishable from a natural stimulus [6]. Recent studies have shown that such goals are attainable by a local microstimulation within the somatosensory cortex. Romo and colleagues demonstrated that during a perceptual task requiring frequency discrimination based on either skin or electrical stimulation, microstimulation of the somatosensory cortex could replace skin vibration [59]. Talwar et al. also demonstrated that the rats could use direct electrical stimulation of their cortical whisker areas as a directional cue for left-right motion [60]. These results suggest that future NIs will be able to deliver sensory feedback to allow more normal movement control using neural output signals. Other input NIS such as deep brain stimulation (DBS), vagus nerve stimulation (VNS) and spinal cord stimulation (SCS) are considered to be in the realm of neuromodulation. These types of stimulating devices aim to alter the brain function, by modulating existing neural circuitry into activity states that overcome imbalances or disruptions produced by disease or injury. DBS has been in wide clinical use for relief of tremor, rigidity and bradykinesia of Parkinson’s disease by manipulating basal ganglia activity through stimulation [61]. Other examples are VNS for intractable seizures and more recently medically
The future of neural interface technology
refractory depression [62,63]. Stimulation of central or peripheral structures is also being evaluated as a treatment for other conditions such as essential tremor, cognitive deficits in Alzheimer’s disease, anxiety disorders, bulimia, resistant obesity, addictions, sleep disorders, narcolepsy, coma and memory and learning deficits [62]. SCS provides pain relief by modulating the chronic pain signals in the dorsal columns of the spinal cord with a programmable low-frequency electrical current from a percutaneously implanted electrode arrays in the epidural space [64,65]. It is likely that many new applications of input NIs will appear in the coming years, adding a large set of new methods to address neurological and psychiatric disorders.
Output NI Output NI describes an NI in which a device coupled with a nervous system is used to provide a command signal from the cortex, serving as a: new functional output to control disabled body parts or physical devices (computers or robotic limbs). Such devices re-establish the communication from the brain, mainly focusing on the motor command or a neural signal that can take the place of a missing motor command. In its simplest form, an output NI does not activate any alternate pathways or restore a structure to an original state (that is, it only detects ongoing activity, it does not try to affect it). Instead, it provides the command output that best substitute for the intended communication to the limbs, a motor command. Thus, an output NI can be considered as a ‘‘motor NI’’ or NMP. An output NI consists of three distinct modules working in unison to restore aspects of lost motor function: the sensory module, data interpretation module, and the data output module. The challenges to the implementation of output NI has been the lack of adequate physical neural interface and technological limitations in processing large amounts of
193
data, and the implemental mathematical tools that can convert complex neural signals into a useful command. The three modules are discussed in detail below.
Sensor Module The sensor module refers that part of an NI system that acquires the neural signal; it is a sensor and all the processing needed to derive signals of interest, such as field potentials (FP) or action potentials (AP), known in experimental neurophysiology as spikes. Indirect NIs use FP, indirect measures of brain processes that can be recorded inside or outside the brain, and direct NIs rely on spiking signals, the information-rich signaling mechanism produced by a single neuron. Direct NIs, at least for the present, require invasive microelectrodes to sense them. Indirect NI
An NI that senses FP is termed an indirect NI because they are based on signals that are not, per se, directly related to movement. The indirect NI records spatially and temporally summed electrical rhythms, such as EEG recorded noninvasively on the scalp, the electrocorticogram (ECoG) recorded invasively under the scalp near the cortical surface, or the local field potential (LFP) recorded invasively within the parenchyma. FP can either be recorded as ongoing rhythms containing different bands, or timelocked, event related potentials (ERPs), which can be recorded in the EEG, ECoG, or LFP in response to various types of stimuli [47]. Standard surface electrodes placed onto the scalp record the signal reflecting the synchronous changes in postsynaptic potentials of thousands of cortical pyramidal cells, which is reported as EEG [66]. EEG electrodes, single or multiple, can be mounted on the head or assembled in a headgear [67–73], or even placed on the dura or the cortical surface for better signal quality, in case of ECoG. Microelectrodes placed in the brain can
3189
3190
193
The future of neural interface technology
record more LFP signal. The spatially and temporally summed synchronous electrical activity are low-pass filtered and volume-averaged based on the type and size of electrodes used, the local filtering effects of tissues, i.e., cerebrospinal fluid and skull, neuronal orientation with respect to the cortical surface, and the site of recording. Consequently, ECoG, recorded directly from the cortical surface, contains higher frequency potentials, wider spectrum and spatial characteristics, and information content than EEG, since the low-pass filtering of extracortical tissues such as skull and skin is eliminated thus improving the signal-to-noise ratio [74,75]. At a larger scale, functional magnetic resonance imaging (fMRI), near-infrared imaging (NIR) and magnetoencephalography (MEG) have been used as non-invasive signal transducers for NI [76]. Blood-oxygen-level-dependent (BOLD) fMRI which measures local hemodynamic alterations in response to changes in LFP [77], often used to study the brain structures involved in voluntary movement and motor learning [78–81], has been used to construct a real-time NI based on fMRI-based feedback related to differential activation of various cortical areas [82]. For example, You et al. demonstrated an fMRI-based NI where subjects were trained to utilize four different imaginary mental tasks observable on BOLD fMRI as a control for cursor movement in four directions [83]. NIR can record cortical hemodynamic and oxygenation changes related to neural activity via lightemitting diodes attached to the scalp [84,85]. These light-emitting diodes, or optodes, can be used to distinguish between two states of motor: intent [86] which could be useful in the development of NI. However, these devices are impractical for their size and cost. They are being used for feedback based therapy and control where they can be implemented in a more restrictive setting. Electrical activity recorded with indirect NI can provide useful command signals, but in the case of rhythms, the command signals must be
derived from a relationship between a learned brain state and the modulation of EEG frequency bands [45]. Wolpaw and McFarland demonstrated that different EEG frequency bands, mu (8–12 Hz) and beta (18–26 Hz) rhythm frequency bands, carry different information that can be independently modulated to provide a multidimensional control signal [87,88]. Individuals can also learn to self-regulate slow cortical potentials [89], or to voluntarily control a specific frequency component of EEG [90–92]. Evoked potentials can also provide information for use in NI with P300 wave, a scalp-measured response evoked to an ‘‘oddball’’ or a series of character sets, which has been used effectively to identify screen location or letters of interest [93,94]. Indirect NI based on an EEG sensor is advantageous in that it carries minimal clinical risks to the user because of its non-invasive nature. ECoG electrodes are invasive and their risk as a long-term implant is unknown, because they are not approved in the US for this purpose, nor are they in clinical trials. Indirect NI has disadvantages in that its use requires considerable training (for EEG systems) and focused attention to maintain control. The method for obtaining signals extracortically has shortcomings for daily use because the electrodes are cumbersome to attach, must be attached daily, may cause discomfort, and has movement-related artifacts. Indirect NIs also require "unnatural" mappings in that one might need to learn to imagine to move the tongue in order to achieve up and down motion of a cursor and the arm to achieve leftright cursor motion. Use requires practice at combining the learned modulation of both of these signals. As such, indirect NIs are already in use, providing means of communication for those with severe paralysis, i.e., in a locked-in state [92,95]. The temporal and spatial resolution of the EEG signal is limited because output signal often depend on multiple repeated samples and may take several seconds for the user to modify the signals [6]. By contrast, using invasive
The future of neural interface technology
methods, the LFP recorded intracortically carry twice as much information about the arm movement direction as those recorded extracortically [96]. The limited information is adequate for applications such as interaction with a computer cursor but may fall short in providing more complex signals such as detailed and continuous multiple degree of freedom actions, without heavy attentional demands. Despite its shortcomings, indirect NI has laid the important foundation in the development of NI by providing important test-bed for the development of mathematical methods and instrumentation to derive command signal from the recorded brain activity, the development of multi-channel signal processing, and the development of variety of computer interfaces [6]. Furthermore, they have demonstrated some practical value for persons with paralysis and provide a choice among a growing set of potential alternatives for those who might desire NI systems to restore lost control and independence. Direct NI
NI, which uses spikes, the msec-long AP from individual neurons, is categorized as a direct NI. The important requirement for direct NI is the ability to record from multiple neurons for long period of time. Assembly of small wires and ‘‘microwires’’ have been successfully used for many years for chronic cortical recordings in the experimental animal setting [97,98], though the foreign bodies such as dental acrylic used to attach the assembly to the skull is a strong impediment to their use in human cortical applications [36,37,99], but are effective for short term use in deep brain structures [100]. For neural prosthesis application, it is necessary to have more information than that provided by one spiking neuron, thus necessitating arrays of recording electrodes [101]. Two systems are in human use. One type is a multi-electrode array consisting of a mm-scale platform with numbers of microelectrodes arranged in a regular grid emanating from below and is in animal [29,102,103]
193
and human [46] use. The second type tested in humans is the neurotropic bioactive extracellular recording glass cone electrode. The cone electrode is a microwire encased in a glass pipette tip, filled with substances that promote nerve in-growth to establish and anchor long-term connections. The cone electrode allows long-term extracellular recording from the cerebral motor cortex of patients [40,41,104–106]. Information within the spike activity is often expressed in terms of number of spikes within a defined interval (rate or a related measure) [101], or relative timing or synchronization between neuron pairs [45]. Modulations in the spike rates often reflect various aspects of movement such as speed, hand position, direction and force. Use of multi-unit activity (MUA) is also being evaluated as a possible neural control signal source. MUA consists of a collection of local spiking neurons that are potentially more robust to record because they are less sensitive to electrode movement and do not necessarily require the same intensive signal processing as the spikes recorded and sorted from each other. LFP can be recorded simultaneously with spikes via intraparechymal recordings. The LFP appears to reflect the local regional neuronal activity, although the nature and source of this local form of FP is under investigation [107]. LFPs have the possibility of providing supplementary, complementary or redundant information related to spiking or MUA and are thus an interesting command signal choice, although this has not been extensively investigated [108]. Thus, electrical activity of invasive, direct NI can provide signals derived from single and multi-neuron spike activity and LFP [45]. Multiple electrodes provide ensemble neuronal activity in which population activity can also be used to derive higher levels of information contained in NIs [109,110]. Intraparenchymal electrodes present challenges in obtaining stable, long-lasting recordings. Slight movements, tissue reaction or local damage could change the population of recorded cells
3191
3192
193
The future of neural interface technology
or eliminate recordings altogether. In addition, systems will need to be developed that can process signals inside the body to remove the need for percutaneous connectors. While these are formidable challenges, initial human recordings have been rather successful over many months [40,41,45,46]. More advanced multiple electrode arrays are in development, relying on advanced designs and manufacturing techniques [102,111,112], but these are not yet translated into human testing. Thus, sensor technology is showing promising results, but additional advances as well as testing of existing sensors are critical in establishing a full understanding of what is necessary for this part of an NIS. Many central nervous system areas carry movement related signals. For an NI that will likely sample from a limited area, it is necessary to select a site that will have useful signals. The M1 arm-hand area is one area of focus [6,113,114]. Electrophysiological studies have shown that features of voluntary motor movement, including dynamics, kinematics, and goals are encoded in M1 neuronal activity [115,116]. Although M1 has traditionally been believed to be an ideal cortical location to extract motor command signals, other cortical areas may provide additional or even more useful signals for some desired applications [34,47]. Single neuron studies demonstrate neurons related to movement learning, planning and performance in more than a dozen fronto-parietal areas: supplementary motor area; cingulate motor areas; and premotor areas [117–125], as well as parietal cortex [38,99,126–128]. Neuronal signals suitable for NI should be clear and reproducible in its relationship to movement [34]. Subcortical structures such as thalamus and basal ganglia may also provide the suitable signals for NI [100] but the long-term access to these deeper structures poses some unique technical challenges compared to surface structures [34].
Data Interpretation Module The data interpretation module refers to the signal processor and the neural decoder, also know as a filter, which is a mathematical approach that transforms digitized brain signal into a control signal that can be used to perform intended actions. Decoded signals may be designed to control computer cursor movements, button activations and complex time-varying movement trajectories such as reaching for an object by a robot. Ideally, transfer of neural signals to processors should occur wirelessly through telemetric communication [41] but most systems now require physical connections [36]. For single neuron decoders, recorded and digitized APs must be discriminated from each other and from noise. Neural firing rates or other measures of a single neuron, MUA or FP activity of interest that carry information must be processed in real-time into a command signal, which is achievable but often an intensive task for modern computers [109,129,130]. It is now possible to predict hand trajectory in real-time in about the same time it takes to execute a volitional act in an able bodied person [36,37,99]. Mathematical decoding methods such as linear regression, best fit model, population vector and neural network model can be used to predict or recreate the hand movement in space with reasonable accuracy [90,91,99,131–134]. Various measures of successful decoding, which is a measure of the overall system operation (i.e., signal plus decoding reliability) have been used [36,132,135], but there is no general agreement about how best to evaluate the quality of the command signal. Correlation and mean squared error, as well as successful target capture and bit rate are all currently used metrics of performance [45,136].
Data Output Module The data output module refers to the output devices, also known as the actuator, assistive
The future of neural interface technology
technology or the ‘‘effectors,’’ that perform a desired action using the neural command. The actuators or the ‘‘effected organs’’ can range from a computer cursor, a motorized wheelchair, a semi-autonomous robot, a robotic or prosthetic arm, to the FES device that reanimates a paralyzed limb. The purpose of a data output module, and in effect the NIS, is to restore function, and subsequently independence to disabled patients. As demonstrations of successful NI operation, neural commands have most often been used to control a computer cursor [46,67,69,72,123,137,138]. Cursor control, such as computer mouse-like point-andclick function, could enable a person with movement limitations to re-establish communication with the world by providing access and interface with the Internet, e-mail or virtual keyboard for text communication. Cursor control may also enable the access to other assistive devices with preprogrammed or automated protocols, limited in flexibility mainly by the degrees of freedom in the control signal. Output signals can also be used to control robotic devices or robotic limbs [33,35,139,140], which may be mounted on a wheelchair [141] or be worn [142]. One of the more exciting types of output module is the activation and movement of patient’s own limb through FES driven by the cortical command signal, a goal that moves towards the recreation of a physical nervous system re-linking brain to body. FES, in which implanted electrode assemblies are used to stimulate motor nerves as they enter the muscles, has been tested in patients implanted with NP in proximal arm and leg [143,144], but not under direct neural control. The muscles are stimulated with varying strength and in orchestrated sequence to achieve basic movement such as handgrip, i.e., in low cervical spinal cord injury (SCI) patients, to give more independence and control. Current FES system utilizes mechanical control such as switches or joystick worn on the contralateral shoulder [143,145] or automated control signals such as pattern generators or EMG-based switches. With NI, the mechanical or automated control signals
193
can be substituted by direct neural control. Lauer et al. has demonstrated that subjects can be trained to control EEG rhythms to trigger the opening and closure of the hand fitted with FES [67].
Neuromotor Prosthesis Thought into Action When volitional motor signal pathways are lost, the information generated in the motor cortical areas cannot reach the muscles, the executing organ. For the last four decades, various approaches have been used to re-establish the means for intentions to become useful actions. One strategy to accomplish this is the direct output NI. The interest in this concept is high due to the implication in treatment of variety of medical conditions previously discussed. In effect, the NI will provide an alternative to a pathway that has been disrupted by trauma or disease. Successful and functional NMP will combine the three core components, described in the previous section, which must function together [47]. Recently, an NI, consisting of a chronically implanted multi-electrode array and a signal processor that can control a computer cursor, has been developed and applied in humans with tetraplegia as part of a pilot human clinical trial [36].
Clinical Experience and Application Researchers at Brown University, Massachusetts General Hospital and the Department of Veterans Affairs, in conjunction with the sponsor company of the clinical trial, Cyberkinetics Neurotechnology Systems, Inc. have been evaluating a pilot NIS, called BrainGate. It is noteworthy that one of the authors of this chapter (GMF) is a founder of this company (see Acknowledgment section, conflict of interest statement). In June of 2004, as part of the BrainGate trial, a pilot safety and feasibility trial for the development of
3193
3194
193
The future of neural interface technology
NI/BCI for persons with paralysis, 25-year-old male with traumatic SCI and complete tetraplegia (C4 ASIA A) [146], was implanted with a multi-electrode platform array that consisted of a 44 mm platform and 100 Si electrodes of 1 mm length. The BrainGate Neural Interface System includes the above-described implantable sensor with a percutaneous pedestal plug and cables providing signal access, external components for signal processing, signal decoding, and external device control, as well as a computer to evaluate signal features and control using visual feedback. The multi-array electrode was implanted into the precentral ‘‘knob’’ (arm/hand area) of the M1, as identified by a pre-operative MRI. The pedestal was secured to the skull and externalized percutanteously, allowing a cable to be connected to external components. Neural recordings were obtained approximately weekly. As part of the study, the participant was asked to imagine limb movements, and attempts were made to decode the real-time neural activity into a control signal for an external device (e.g., a computer cursor). Recordings demonstrated the presence of neurons and the patient’s ability to modulate the recorded activity with movement intention, despite long-standing disconnection of the motor cortex from effectors. This was a landmark observation because it showed that movement related activity remained years after SCI, and that movement intention activated the spiking in the absence of an actual performed arm movement, which are essential steps in creating a useful NI system. While this sensor was used to record neuronal spiking for NI applications, the ability to detect ensembles of spiking neurons may be useful in a variety of clinical settings for recording this type of neuronal activity (e.g., epilepsy monitoring at a new, single neuron scale). The array was removed after the planned initial trial period of 1 year. Upon removal, no grossly notable tissue reaction was observed other than slight indentation of the cortex at the site of each microelectrode penetration.
Consistent with what was observed in preclinical monkey studies, this preliminary observation in one participant indicates that a small array of microelectrodes can be successfully implanted and removed from the human motor cortex.
Cognitive or Bypass NI The third category of NI can be defined as a cognitive or bypass NI. These are unique in that they restore the lost communication between brain regions that results from damage to a particular neural structure. The only current example of this is a hippocampal-cortical prosthesis.
Hippocampal-Cortical Prosthesis The role of the hippocampus in processing and establishing long-term declarative or fact-based memory has long been recognized. Damage to the hippocampus that disrupts the propagation of spatio-temporal patterns of activity through the intricate internal circuitry of the hippocampus results in severe anterograde amnesia. This can occur as the result of head injury, anoxic injury, stroke, dementia, mesial temporal lobe epilepsy, or any combination of these. Development of a neural prosthesis for a damaged hippocampus presents unique challenges. This is essentially a neural prosthesis to replace lost cognitive function. It is quite different than an input prosthesis for sensory transduction or an output prosthesis for translation of thought into motor activity. It involves replacing communication between brain regions. It requires the restoration of the non-linear multi-input, multi-output transformation of spatio-temporal patterns of activity [7]. Currently there are a few ongoing endeavors to investigate and develop a bypass for nonfunctioning hippocampus circuitry [147,148]. Berger et al. have published the results of using a hippocampal slice preparation as an initial step
The future of neural interface technology
in the development of a hippocampal prosthesis [147]. This model involves surgically transecting CA3 axons, replacing them with a hardware integration model of the nonlinear dynamics of CA3. Then, through a multi-site electrode array, it transmits output of the dentate gyrus directly to the CA1synaptic inputs, thus bypassing the CA3 region. Their results showed close agreement between data from intact slices and transected slices with the hardware-substituted CA3 [147]. This may be the first step toward the development of an in vivo hippocampal prosthesis that may one day benefit millions of people suffering from many crippling disorders that damage the hippocampus.
Future of Neural Interface Systems The accomplishments of NI research mark the beginning of what appears to be a promise of a range of input and output NI that has the potential to restore normal function through neuromodulation of neural circuits that are imbalanced, through prostheses that could restore any lost sensory function from hearing to movement, and through an output NI that could restore communication and control for those with movement limitations, going as far as to return voluntary control over muscles. This is a very long-term and somewhat speculative view, but the recent success in demonstrating the ability to detect signals at the single neuron level as well as the ability for persons with long-standing tetraplegia to engage these neurons and then for decoders to make this neural activity into useful control signals, suggests that major essential steps towards a meaningful new way to approach disorders of the nervous system have been accomplished. From a device or system design standpoint, NIs must now be affirmed to be effective in a larger number of people with paralysis to show the generality of findings to date. Many of the next steps involve challenging, but not
193
unreasonably difficult steps: useful output systems need to become fully implantable and run without technical oversight. This will require the development and testing of sophisticated microsystems that transcend the most complex machines now implanted in humans, such as cardiac pacemaker/defibrillators. Other big challenges for a successful NI system in the future are cost and availability. Cost of a complex device will likely appear to be high, but given the potential benefit, this could be justified. NI systems are expensive to develop and produce, as they must be adequately tested to ensure long-term efficacy and, most significantly, safety. The large number of devices now implanted in humans with relatively low complication rates suggests that risk is low for current devices. Ultimately, NI should perform increasingly complex functions driven by the brain signals, and feedback increasingly complex inputs into the brain, with minimal interference to other cortical functions. In terms of system requirement specifically for a NMP, an ideal NI should be able to provide a control signal that restores natural movement of paralyzed body parts without extensive learning. Control should emerge from the voluntary intent to carry out an action without effort and significant attentional demands. The demonstrations of direct NIs suggest that this is achievable, as long as signals are maintained. These also appear to be tractable issues that will allow development of devices capable of more complex functions. We could conceive of a system of multiple sensors in arm and leg regions of motor cortex that would allow independent control of both arms and both legs, ultimately to stand, walk and perform bimanual tasks. Again, these are very ambitious goals with many intervening steps before such a goal is reached. An ideal NI should also deliver useful sensory signals beyond vision, such as feedback of limb position, and this too is both achievable and, we speculate, likely to be demonstrated in humans within the next decade. Input signals of this
3195
3196
193
The future of neural interface technology
bidirectional system should be perceived as a natural input, without disturbing behavior more than the arrival of a typical sensory signal in the intact nervous system. Feedback of limb position and touch will also be essential to mimicking the natural state of the closed-loop system. A future NI should be a seamless and intimately integrated assistive device that restores function and ability to communicate and interact with the outside world to those who have lost the ability. This demand will require further miniaturization of processors, electronics and assistive devices. The devices should perform complex tasks as it interacts with our brain yet should be as simple as putting on a pair of glasses, but still retain flexibility. The far future set of NIs could restore our vision, hearing, mobility and cognitive function. Future input NIs, perhaps in the nearer term, will restore function in terms of improved control of neuropsychiatric disorders refractory to medical treatment, i.e., obsessive compulsive disorder, Parkinson’s disease and depression.
Acknowledgment Conflict of interest statement: G. M. Friehs is one of the founders of Cyberkinetics Neurotechnology Systems, Inc. (CYKN), a medical device company that is running the BrainGate pilot clinical trial. M. C. Park, M. A. Goldman and T. W. Belknap have no conflict of interest.
References 1. Delgado JM, Hamlin H, Chapman WP. Technique of intracranial electrode implacement for recording and stimulation and its possible therapeutic value in psychotic patients. Confin Neurol 1952;12:315-9. 2. Delgado JM. Physical control of the Mind. New York: Harper and Rowe; 1969. 3. Cooper IS. Twenty-five years of experience with physiological neurosurgery. Neurosurgery 1981;9:190-200. 4. Moore GE. Cramming more components onto integrated circuits. Electronics 1965;38:114-7.
5. Peercy PS. The drive to miniaturization. Nature 2000; 406:1023-6. 6. Donoghue JP. Connecting cortex to machines: recent advances in brain interfaces. Nat Neurosci 2002;5 Suppl:1085-8. 7. Song D, Chan RH, Marmarelis VZ, Hampson RE, Deadwyler SA, Berger TW. Nonlinear dynamic modeling of spike train transformations for hippocampal-cortical prostheses. IEEE Trans Biomed Eng 2007;54:1053-66. 8. Murray CJL, Lopez AD. The global burden of disease: a comprehensive assessment of mortality and disability from diseases, injuries and risk factors in 1990 projected to 2020. Boston: Harvard University Press; 1996. 9. National Spinal Cord Injury Statistical Center. Spinal cord injury. Facts and figures at a glance. University of Alabama at Birmingham. http://www.spinalcord.uab.edu/show.asp? durki¼116979&site¼1021&return¼19775 Accessed 1 May 2008. 10. Trafton PG. Spinal cord injuries. Surg Clin North Am 1982;62:61-72. 11. Aaron RK, Herr HM, Ciombor DM, Hochberg LR, Donoghue JP, Briant CL, et al. Horizons in prosthesis development for the restoration of limb function. J Am Acad Orthop Surg 2006;14(10):S198-S204. 12. Jackson AB, Dijkers M, Devivo MJ, Poczatek RB. A demographic profile of new traumatic spinal cord injuries: change and stability over 30 years. Arch Phys Med Rehabil 2004;85:1740-8. 13. Humphrey DR. Representation of movements and muscles within the primate precentral motor cortex: historical and current perspectives. Fed Proc 1986;45:2687-99. 14. Cheney PD. Electrophysiological methods for mapping brain motor and sensory circuits. In: Toga AW, Mazziotta JC, editors. Brain mapping: the methods. San Diego, CA: Academic Press; 1996. p. 277-309. ¨ ber die elektrische Erregbarkeit des 15. Fritsch G, Hitzig E. U Grosshirns. Arch Anat Physiol 1870;37:300-32. 16. Fritsch G, Hitzig E. On the electrical excitability of the cerebrum. In: Some papers on the cerebral cortex (trans: von Bonin G). Springfield, IL: Charles C Thomas; 1960. p. 73-96. 17. Ferrier D. Experiments in the brain of monkeys. Proc R Soc Lond B Biol Sci 1875;23:409-30. 18. Sherrington CS. On the motor area of the cerebral cortex. In: Denny-Brown D, editor. Selected writings of Sir Charles Sherrington. London: Hamish Hamilton Medical Books; 1939. p. 397-439. 19. Penfield W, Boldrey E. Somatic motor and sensory representation in the cerebral cortex of man as studied by electrical stimulation. Brain 1937;60:389-443. 20. Schott GD. Penfield’s homunculus: a note on cerebral cartography. J Neurol Neurosurg Psychiatry 1993;56: 329-33. 21. Jasper H, Ricci GF, Doane B. Patterns of cortical neuronal discharge during conditioned responses in monkeys. In: Wolstenholme GEW, O’Connor CM, editors.
The future of neural interface technology
22.
23.
24. 25.
26.
27.
28.
29.
30.
31.
32.
33. 34.
35.
36.
37.
38.
Neurological basis of behaviour. Boston, MA: Little Brown; 1958. p. 277-94. Evarts EV. Methods for recording activity of individual neurons in moving animals. Methods Med Res 1966;11:241-61. Evarts EV. Relation of pyramidal tract activity to force exerted during voluntary movement. J Neurophysiol 1968;31:14-27. Salcman M, Bak MJ. A new chronic recording intracortical microelectrode. Med Bio Eng 1976;14:42-50. Kru¨ger J. A 12-fold microelectrode for recording from vertically aligned cortical neurons. J Neurosci Methods 1982;6:347-50. Verloop AJ, Holsheimer J. Simple method for the construction of electrode arrays. J Neurosci Methods 1984;11:173-8. Rousche PJ, Normann RA. A method for pneumatically inserting an array of penetrating electrodes into cortical tissue. Ann Biomed Eng 1992;20:413-22. Howard MA, Volkov IO, Noh MD, Granner MA, Mirsky R, Garell PC. Chronic microelectrode investigations of normal human brain physiology using a hybrid depth electrode. Stereotact Funct Neurosurg 1997;68: 236-42. Rousche PJ, Normann RA. Chronic recording capability of the Utah intracortical electrode array in cat sensory cortex. J Neurosci Methods 1998;82:1-15. Fried I, Wilson CL, Maidment NT, Engle J Jr, Behnke E, Fields TA, et al. Cerebral microdialysis combined with single-neuron and electroencephalographic recording in neurosurgical patients. J Neurosurg 1999;91:697-705. Ojemann GA, Schoenfield-McNeill J. Activity of neurons in human temporal cortex during identification and memory for names and words. J Neurosci 1999;19: 5674-82. Ulbert I, Halgren E, Heit G, Karmos G. Multiple microelectrode-recording system for human intracortical applications. J Neurosci Methods 2001;106:69-79. Fetz EE. Real-time control of a robotic arm by neuronal ensembles. Nat Neurosci 1999;2:583-4. Cheney PD, Hill-Karrer J, Belhaj-Saı¨f A, McKiernan BJ, Park MC, Marcario JK. Cortical motor areas and their properties: implications for neuroprosthetics. Prog Brain Res 2000;128:135-60. Chapin JK, Moxon KA, Markowitz RS, Nicolelis MA. Real-time control of a robot arm using simultaneously recorded neurons in the motor cortex. Nat Neurosci 1999;2:664-70. Serruya MD, Hatsopoulos NG, Paninski L, Fellows MR, Donoghue JP. Instant neural control of a movement signal. Nature 2002;416:141-2. Taylor DM, Tillery SI, Schwartz AB. Direct cortical control of 3D neuroprosthetic devices. Science 2002;296: 1829-32. Carmena JM, Lebedev MA, Crist RE, O’Doherty JE, Santucci DM, Dimitrov DF, et al. Learning to control a
39.
40.
41.
42.
43.
44.
45.
46.
47.
48. 49.
50.
51.
52. 53. 54.
193
brain-machine interface for reaching and grasping by primates. PLoS Biol 2003;1:E42. Goldring S, Ratcheson R. Human motor cortex; sensory input data from single neuron recordings. Science 1972;175:1493-5. Kennedy PR, Bakay RA. Restoration of neural output from a paralyzed patient by a direct brain connection. Neuroreport 1998;9:1707-11. Kennedy PR, Bakay RE, Moore MM, Adams K, Goldwaithe J. Direct control of a computer from the human central nervous system. IEEE Trans Rehab Eng 2000;8: 198-202. Musallam S, Corneil BD, Greger B, Scherberger H, Andersen RA. Cognitive control signals for neural prosthetics. Science 2004;305:258-62. Suner S, Fellows MR, Vargas-Irwin C, Nakata GK, Donoghue JP. Reliability of signals from a chronically implanted, silicon-based electrode array in non-human primate primary motor cortex. IEEE Trans Neural Syst Rehabil Eng 2005;13:524-41. Ku¨bler A, Mushahwar VK, Hochberg LR, Donoghue JP. BCI meeting 2005 – workshop on clinical issues and applications. IEEE Trans Neural Syst Rehabil Eng 2006;14:131-4. Donoghue JP, Nurmikko A, Black M, Hochberg LR. Assistive technology and robotic control using motor cortex ensemble-based neural interface systems in humans with tetraplegia. J Physiol 2007;579:603-11. Hochberg LR, Serruya MD, Friehs GM, Mukand JA, Saleh M, Caplan AH, et al. Neuronal ensemble control of prosthetic devices by a human with tetraplegia. Nature 2006;442:164-71. Friehs GM, Zerris VA, Ojakangas CL, Fellows MR, Donoghue JP. Brain-machine and brain-computer interfaces. Stroke 2004;35(11 Suppl 1):2702-5. Parmet S, Lynm C, Glass RM. JAMA patient page. Cochlear implants. JAMA 2004;291:2398. Spahr AJ, Dorman MF. Performance of subjects fit with the Advanced Bionics CII and Nucleus 3G cochlear implant devices. Arch Otolaryngol Head Neck Surg 2004;130:624-8. Kuchta, J. Twenty-five years of auditory brainstem implants: perspectives. Acta Neurochir Suppl 2007;97 (Pt 2): 443-9. Lim HH, Lenarz T, Joseph G, Battmer RD, Samii A, Samii M, et al. Electrical stimulation of the midbrain for hearing restoration: insight into the functional organization of the human central auditory system. J Neurosci 2007;27:13541-51. Maynard EM. Visual prostheses. Annu Rev Biomed Eng 2001;3:145-68. Weiland JD, Liu W, Humayun MS. Retinal prosthesis. Annu Rev Biomed Eng 2005;7:361-401. Winter JO, Cogan SF, Rizzo JF, III. Retinal prostheses: current challenges and future outlook. J Biomater Sci Polym Ed 2007;18:1031-55.
3197
3198
193
The future of neural interface technology
55. Brindley GS, Lewin WS. The sensations produced by electrical stimulation of the visual cortex. J Physiol 1968;196:479-93. 56. Dobelle WH. Artificial vision for the blind by connecting a television camera to the visual cortex. ASAIO J 2000;46:3-9. 57. Veraart C, Wanet-Defalque MC, Ge´rard B, Vanlierde A, Delbeke J. Pattern recognition with the optic nerve visual prosthesis. Artif Organs 2003;27:996-1004. 58. Hetling JR, Baig-Silva MS. Neural prostheses for vision: designing a functional with retinal neurons. Neurol Res 2004;26:21-34. 59. Romo R, Herna´ndez A, Zainos A, Brody CD, Lemus L. Sensing without touching: psychophysical performance based on cortical microstimulation. Neuron 2000;26:273-8. 60. Talwar SK, Xu S, Hawley ES, Weiss SA, Moxon KA, Chapin JK. Rat navigation guided by remote control. Nature 2002;417:37-8. 61. Benabid AL, Koudsie A, Benazzouz A, Piallat B, Krack P, Limousin-Dowsey P, et al. Deep brain stimulation for Parkinson’s disease. Adv Neurol 2001;86:405-12. 62. Ansari S, Chaudhri K, Al Moutaery KA. Vagus nerve stimulation: indications and limitations. Acta Neurochir Suppl 2007;97(Pt 2):281-6. 63. Park MC, Goldman MA, Carpenter LL, Price LH, Friehs GM. Vagus nerve stimulation for depression: rationale, anatomical and physiological basis of efficacy and future prospects. Acta Neurochir Suppl 2007;97 (Pt 2):407-16. 64. Lanner G, Spendel MC. Spinal cord stimulation for the treatment of chronic non-malignant pain. Acta Neurochir Suppl 2007;97(Pt 1):79-84. 65. Nicholson CL, Korfias S, Jenkins A. Spinal cord stimulation for failed back surgery syndrome and other disorders. Acta Neurochir Suppl 2007;97(Pt 1):71-7. 66. Thakor NV, Tong S. Advances in quantitative electroencephalogram analysis methods. Annu Rev Biomed Eng 2004;6:453-95. 67. Lauer RT, Peckham PH, Kilgore KL. EEG-based control of a hand grasp neuroprosthesis. Neuroreport 1999;10: 1767-71. 68. Birbaumer N, Ku¨bler A, Ghanayim N, Hinterberger T, Perelmouter J, Kaiser J, et al. The thought translation device (TTD) for completely paralyzed patients. IEEE Trans Rehabil Eng 2000;8:190-3. 69. Kostov A, Polak M. Parallel man-machine training in development of EEG-based cursor control. IEEE Trans Rehabil Eng 2000;8:203-5. 70. Obermaier B, Neuper C, Guger C, Pfurtscheller G. Information transfer rate in a five-classes brain-computer interface. IEEE Trans Neural Syst Rehabil Eng 2001;9: 283-8. 71. Guger C, Edlinger G, Harkam W, Niedermayer I, Pfurtscheller G. How many people are able to operate an EEG-based brain-computer interface (BCI)? IEEE Trans Neural Syst Rehabil Eng 2003;11:145-7.
72. Obermaier B, Muller GR, Pfurtscheller G. ‘‘Virtual keyboard’’ controlled by spontaneous EEG activity. IEEE Trans Neural Syst Rehabil Eng 2003;11:422-6. 73. Sheikh H, McFarland DJ, Sarnacki WA, Wolpaw JR. Electroencephalographic (EEG)-based communication: EEG control versus system performance in humans. Neurosci Lett 2003;345:89-92. 74. Leuthardt EC, Schalk G, Wolpaw JW, Ojemann JG, Moran DW. A brain-computer interface using electrocorticographic signals in humans. J Neural Eng 2004;1: 63-71. 75. Graimann B, Huggins JE, Levine SP, Pfurtscheller G. Toward a direct brain interface based on human subdural recordings and wavelet-packet analysis. IEEE Trans Biomed Eng 2004;51:954-62. 76. Mellinger J, Schalk G, Braun C, Preissl H, Rosenstiel W, Birbaumer N, et al. An MEG-based brain-computer interface (BCI). Neuroimage 2007;36:581-93. 77. Logothetis NK. The underpinnings of the BOLD functional magnetic resonance imaging signal. J Neurosci 2003;23:3963-71. 78. Mattay VS, Weinberger DR. Organization of the human motor system as studied by functional magnetic resonance imaging. Eur J Radiol 1999;30:105-14. 79. Baker JT, Donoghue JP, Sanes JN. Gaze direction modulates finger movement activation patterns in human cerebral cortex. J Neurosci 1999;19:10044-52. 80. Sanes JN, Donoghue JP. Plasticity and primary motor cortex. Annu Rev Neurosci 2000;23:393-415. 81. Ungerleider LG, Doyon J, Karni A. Imaging brain plasticity during motor skill learning. Neurobiol Learn Mem 2002;78:553-64. 82. Weiskopf N, Mathiak K, Bock SW, Scharnowski F, Veit R, Grodd W, et al. Principles of a brain-computer interface (BCI) based on real-time functional magnetic resonance imaging (fMRI). IEEE Trans Biomed Eng 2004;51:966-70. 83. Yoo SS, Fairneny T, Chen NK, Choo SE, Panych LP, Park H, et al. Brain-computer interface using fMRI: spatial navigation by thoughts. Neuroreport 2004;15:1591-5. 84. Benaron DA, Hintz SR, villringer A, Boas D, Kleinschmidt A, Frahm J, et al. Noninvasive functional imaging of human brain using light. J Cerebral Blood Flow Metab 2000;20:469-77. 85. Pouratian N, Sheth SAS, Martin NA, Toga AW. Shedding light on brain mapping: advances in human optic imaging. Trends Neurosci 2003;26:277-82. 86. Coyle S, Ward T, Markham C, McDarby G. On the suitability of near-infrared (NIR) systems for next-generation brain-computer interfaces. Physiol Meas 2004;25:815-22. 87. Wolpaw JR, McFarland DJ. Control of a two-dimensional movement signal by a noninvasive brain-computer interface in humans. Proc Natl Acad Sci USA 2004;101: 17849-54. 88. Wolpaw JR. Brain-computer interfaces as new brain output pathways. J Physiol 2007;579:613-19.
The future of neural interface technology
89. Ku¨bler A, Neumann N, Kaiser J, Kotchoubey B, Hinterberger T, Birbaumer NP. Brain-computer communication: self-regulation of slow cortical potentials for verbal communication. Arch Phys Med Rehabil 2001;82:1533-9. 90. McFarland DJ, Sarnacki WA, Wolpaw JR. Braincomputer interface (BCI) operation: optimizing information transfer rates. Biol Psychol 2003;63:237-51. 91. McFarland DJ, Wolpaw JR. EEG-based communication and control: speed-accuracy relationships. Appl Psychophysiol Biofeedback 2003;28:217-31. 92. Pfurtscheller G, Neuper C, Mu¨ller GR, Obermaier B, Krausz G, Schlo¨gl A, et al. Graz-BCI: state of the art and clinical applications. IEEE Trans Neural Syst Rehabil Eng 2003;11:177-80. 93. Krusienski DJ, Sellers EW, Cabestaing F, Bayoudh S, McFarland DJ, Vaughan TM, et al. A comparison of classification techniques for the P300 Speller. J Neural Eng 2006;3:299-305. 94. Sellers EW, Krusienski DJ, McFarland DJ, Vaughan TM, Wolpaw JR. A P300 event-related potential braincomputer interface (BCI): the effects of matrix size and inter stimulus interval on performance. Biol Psychol 2006;73:242-52. 95. Birbaumer N, Ghanayim N, Hinterberger T, Iversen I, Kotchoubey B, Ku¨bler A, et al. A spelling device for the paralysed. Nature 1999;398:297-8. 96. Mehring C, Nawrot MP, de Oliveira SC, Vaadia E, Schulze-Bonhage A, Aertsen A, et al. Comparing information about arm movement direction in single channels of local and epicortical field potentials from monkey and human motor cortex. J Physiol Paris 2004;98: 498-506. 97. Marg E, Adams JE. Indwelling multiple micro-electrodes in the brain. Electroencephalogr Clin Neurophysiol 1967;23:277-80. 98. Palmer C. A microwire technique for recording single neurons in unrestrained animals. Brain Res Bull 1978;3:285-9. 99. Wessberg J, Stambaugh CR, Kralik JD, Beck PD, Laubach M, Chapin JK, et al. Real-time prediction of hand trajectory by ensembles of cortical neurons in primates. Nature 2000;408:361-5. 100. Patil PG, Carmena JM, Nicolelis MA, Turner DA. Ensemble recordings of human subcortical neurons as a source of motor control signals for a brain-machine interface. Neurosurgery 2004;55:1-10. 101. Stevens CF, Zador A. Neural coding: the enigma of the brain. Curr Biol 1995;5:1370-1. 102. Maynard EM, Nordhausen CT, Normann RA. The Utah intracortical electrode array: a recording structure for potential brain-computer interfaces. Electroencephalogr Clin Neurophysiol 1997;102:228-39. 103. Musallam S, Bak MJ, Troyk PR, Andersen RA. A floating metal microelectrode array for chronic implantation. J Neurosci Methods 2007;160:122-7.
193
104. Kennedy PR. The cone electrode: a long-term electrode that records from neuritis grown onto its recording surface. J Neurosci Methods 1989;29:181-93. 105. Kennedy PR, Mirra SS, Bakay RA. The cone electrode: ultrastructural studies following long-term recording in rat and monkey cortex. Neurosci Lett 1992;142:89-94. 106. Kennedy PR, Bakay RA. Activity of single action potentials in monkey motor cortex during long-term task learning. Brain Res 1997;760:251-4. 107. Belitski A, Gretton A, Magri C, Murayama Y, Montemurro MA, Logothetis NK, et al. Low-frequency local field potentials and spikes in primary visual cortex convey independent visual information. J Neurosci 2008;28: 5696-709. 108. Andersen RA, Musallam S, Pesaran B. Selecting the signals for a brain-machine interface. Curr Opin Neurobiol 2004;14:720-6. 109. Georgopoulos AP. Population activity in the control of movement. Int Rev Neurobiol 1994;37:103-19. 110. Maynard EM, Hatsopoulos NG, Ojakangas CL, Acuna BD, Sanes JN, Normann RA, et al. Neuronal interactions improve cortical population coding of movement direction. J Neurosci 1999;19: 8083-93. 111. Bai Q, Wise KD. Single-unit neural recording with active microelectrode arrays. IEEE Trans Biomed Eng 2001;48:911-20. 112. Rousche PJ, Pellinen DS, Pivin DP, Jr, Williams JC, Vetter RJ, Kipke DR. Flexible polyimide-based intracortical electrode arrays with bioactive capabilities. IEEE Trans Biomed Eng 2001;48:361-71. 113. Schwartz AB. Cortical neural prosthetics. Annu Rev Neurosci 2004;27:487-507. 114. Serruya M, Donoghue J. Design principles of a neuromotor prosthetic device. In: Horch KW, Dhillon GS, editors. Neuroprosthetics: theory and practice. River Edge, NJ: World Scientific; 2004. p. 1158-96. 115. Humphrey DR, Tanji J. What features of voluntary motor control are encoded in the neuronal discharge of different cortical motor areas? In: Humphrey DR, Freund HJ, editors. Motor control: concepts and issues. New York: Wiley; 1991. p. 413-43. 116. Scott SH. The role of primary motor cortex in goaldirected movements: insights from neurophysiological studies on non-human primates. Curr Opin Neurobiol 2003;13:671-7. 117. Dum RP, Strick PL. The origin of corticospinal projections from the premotor areas in the frontal lobe. J Neurosci 1991;11:667-87. 118. He SQ, Dum RP, Strick PL. Topographic organization of corticospinal projections from the frontal lobe: motor areas on the lateral surface of the hemisphere. J Neurosci 1993;13:952-80. 119. Galea MP, Darian-Smith I. Multiple corticospinal neuron populations in the macaque monkey are specified by
3199
3200
193 120.
121.
122.
123.
124.
125.
126.
127.
128.
129.
130.
131.
132.
133.
The future of neural interface technology
their unique cortical origins, spinal terminations, and connections. Cereb Cortex 1994;4:166-94. He SQ, Dum RP, Strick PL. Topographic organization of corticospinal projections from the frontal lobe: motor areas on the medial surface of the hemisphere. J Neurosci 1995;15:3284-306. Dum RP, Strick PL. Spinal cord terminations of the medial wall motor areas in macaque monkeys. J Neurosci 1996;16:6513-25. Picard N, Strick PL. Motor areas of the medial wall: a review of their location and functional activation. Cereb Cortex 1996;6:342-53. Donoghue JP, Saleh M, Caplan A, Serruya MD, Morris DJ, Ramchandani S, et al. Direct control of a computer cursor by frontal cortical ensembles in humans: prospects for neural prosthetic control. Program No. 607.9. 2003 Abstract Viewer/Itinerary Planner. Washington, DC: Society for Neuroscience; 2003 Online. Ojakangas CL, Caplan A, Serruya M, Ramchandani S, Donoghue JP, Friehs GM. Properties of human frontal cortex neurons during visuomotor tasks. Program No. 919.14. 2003 Abstract Viewer/Itinerary Planner. Washington, DC: Society for Neuroscience; 2003 Online. Ojakangas CL, Shaikhouni A, Friehs GM, Caplan AH, Serruya MD, Saleh M, et al. Decoding movement intent from human premotor cortex neurons for neural prosthetic applications. J Clin Neurophysiol 2006;23:577-84. Cohen YE, Batista AP, Andersen RA. Comparison of neural activity preceding reaches to auditory and visual stimuli in the parietal reach region. Neuroreport 2002;13:891-4. Pesaran B, Pezaris JS, Sahani M, Mitra PP, Andersen RA. Temporal structure in neuronal activity during working memory in macaque parietal cortex. Nat Neurosci 2002;5:805-11. Shenoy KV, Meeker D, Cao S, Kureshi SA, Pesaran B, Buneo CA, et al. Neural prosthetic control signals from plan activity. Neuroreport 2003;14:591-6. Evarts EV. Pyrimidal tract activity associated with a conditioned hand movement in monkey. J Neurophysiol 1966;29:1011-27. Humphrey DR, Schmidt EM, Thompson WD. Predicting measures of motor performance from multiple cortical spike trains. Science 1970;170:758-62. Schwartz AB, Taylor DM, Tillery SI. Extraction algorithms for cortical control of arm prosthetics. Curr Opin Neurobiol 2001;11:701-7. Helms Tillery SI, Taylor DM, Schwartz AB. Training in cortical control of neuroprosthetic devices improves signal extraction from small neuronal ensembles. Rev Neurosci 2003;14:107-19. Serruya M, Hatsopoulos N, Fellows M, Paninski L, Donoghue J. Robustness of neuroprosthetic decoding algorithms. Biol Cybern 2003;88:219-28.
134. Paninski L, Fellows MR, Hatsopoulos NG, Donoghue JP. Spatiotemporal tuning of motor cortical neurons for hand position and velocity. J Neurophysiol 2004;91: 515-32. 135. Isaacs RE, Weber DJ, Schwartz AB. Work toward realtime control of a cortical neural prosthesis. IEEE Trans Rehabil Eng 2000;8:196-8. 136. Wu W, Gao Y, Bienenstock E, Donoghue JP, Black MJ. Bayesian population decoding of motor cortical activity using a Kalman filter. Neural Comput 2006;18:80-118. 137. Neuper C, Muller GR, Kubler A, Birbaumer N, Pfurtscheller G. Clinical application of an EEG-based brain-computer interface: a case study in a patient with severe motor impairment. Clin Neurophysiol 2003;114: 399-409. 138. Pfurtscheller G, Muller GR, Pfurtscheller J, Gerner HJ, Rupp R. ‘‘Thougth’’-control of functional electrical stimulation to restore hand grasp in a patient with tetraplegia. Neurosci Lett 2003;351:33-6. 139. Craelius W. The bionic man: restoring mobility. Science 2002;295:1018-21. 140. Taylor DM, Tillery SI, Schwartz AB. Information conveyed through brain-control: cursor versus robot. IEEE Trans Neural Syst Rehabil Eng 2003;11:195-9. 141. Driessen BJ, Evers HG, van Woerden JA. MANUS – a wheelchair-mounted rehabilitation robot. Proc Inst Mech Eng 2001;215:285-90. 142. Bonivento C, Davalli A, Fantuzzi C. Tuning of myoelectric prostheses using fuzzy logic. Artif Intell Med 2001;21:221-6. 143. Kilgore KL, Peckham PH, Keith MW, Thrope GB, Wuolle KS, Bryden AM, et al. An implanted upperextremity neuroprosthesis. Follow-up of five patients. J Bone Joint Surg Am 1997;79:533-41. 144. Triolo RJ, Liu MQ, Kobetic R, Uhlir JP. Selectivity of intramuscular stimulating electrodes in the lower limbs. J Rehabil Res Dev 2001;38:533-44. 145. Hart RL, Kilgore KL, Peckham PH. A comparison between control methods for implanted FES hand-grasp systems. IEEE Trans Rehabil Eng 1998;6:208-18. 146. Maynard FM, Jr, Bracken MB, Creasey G, Ditunno JF, Jr, Donovan WH, Ducker TB, et al. International standards for neurological and functional classification of spinal cord injury. American Spinal Injury Association. Spinal Cord 1997;35:266-74. 147. Hsiao MC, Chan CH, Srinivasan V, Ahuja A, Erinjippurath G, Zanos TP, et al. VLSI implementation of a nonlinear neuronal model: a ‘‘neural prosthesis’’ to restore hippocampal trisynaptic dynamics. Conf Proc IEEE Eng Med Biol Soc 2006;1:4396-9. 148. Brijesh R, Ravindran G. A spiking neural network of the CA3 of the hippocampus can be a neural prosthesis for lost cognitive functions. Conf Proc IEEE Eng Med Biol Soc 2007;1:4755-8.
189 The Future of Neuronavigation D. W. Roberts
As the concept of frameless stereotaxy, imageguided surgery or neuronavigation has moved from early development to general commercial distribution and rapidly approaches a mature stage, consideration of its course over the next decade warrants discussion of the likely refinement of its underlying technology, expansion of its clinical application, and its implications, including the effect on the training of tomorrow’s surgeons. As one might have anticipated, various implementations and strategies over the past two decades have proven of variable utility, but the overarching methodology, by which the operative field and additional information in the form of imaging or atlases can be co-registered to provide accurate and generally planned surgical intervention, has gradually become integrated into our general procedures. Trying to predict the future is often a fool’s errand, but some consideration of where we might be headed may facilitate informed planning at both the developmental and clinical levels.
Technology Digitizers The earliest image-guidance digitizing technologies had included sonic-based range-finders and articulated arms, and although these are occasionally still used, they have been largely replaced by optical, and to a lesser extent electromagnetic, digitizers. The accuracy and robustness of these methods have been satisfactory for practical use in the operating room environment, and the respective line-of-sight
#
Springer-Verlag Berlin/Heidelberg 2009
and non-ferromagnetic constraints have been generally accepted. The most widely used systems today employ camera-based systems tracking either light-emitting diodes (LEDs) or reflective spheres [1–3]; a smaller number of systems utilize electromagnetic field detection, accepting potential interference by ferromagnetic material near the operative field for increasing miniaturization, low-cost and elimination of the line-ofsight requirement [4,5]. There has not been great or immediate pressure to move beyond these technologies, but there are emerging methodologies offering potential improvements in being less intrusive and more automated. Laser scanners digitizing scalp and facial surfaces have been demonstrated to be capable of initial registration through matching with surface contours derived from imaging [6,7]. Without reliance on designated points or fiducial markers, such methods could become fully automated. Potentially they could track the brain or resection surface intraoperatively as well and play a role in updating registration during a procedure. Optical systems incorporating more advanced image processing are capable of digitizing features of the surgical field as well as instruments themselves, and can be used for both registration and instrument tracking. Incorporation of machine-vision methods was recognized early on [8] but overshadowed by LED-based systems. Improving resolution, better algorithms, and decreasing costs will likely tempt revisiting this technology. Both room-based as well as operating microscope camera-based systems have appeal, and the potential of this approach has been
3138
189
The future of neuronavigation
demonstrated in a microscope-mounted system developed by Sun et al. [9]. In this implementation, two high-resolution cameras are attached to stereoscopic ports of the operating microscope, and image-processing algorithms properly align the images and autodetect features on the cortical surface. In this manner, the cortical surface can be tracked real-time during surgery in a fully automated manner. Other technologies have been used in the role of digitizers, including X-ray [10,11] and ultrasound [12,13], and these can effect co-registration. As their information content increases, of course, they may provide guidance to information within their own image, supplanting the need for coregistration to preoperative imaging, although that may be accomplished as well. The move toward intraoperative MRI (and CT) represents such a development, and this will be discussed later. Lastly, the integration of more than one digitizer into a procedure will likely become more common and offer advantages of redundancy and complimentary abilities. Just as biological development has seen the evolution of multisensate organisms, intraoperative tools will evolve similarly. The SurgiScope operating microscope system (ISIS, Grenoble), for example, early on employed both encoders in the joints of its robotic links as well as an LED-based optical system [14]. Back-up digitizing systems to function when constraints (e.g., line-of-sight) preclude utilization of a primary system are presently expensive but obviously could be helpful. Alternatively, sequential registration – either for refinement of an initial registration or its updating – might well utilize different technologies.
Registration Paired point registration methods have dominated image-guidance systems to present, although alternatives that match surfaces, combinations of points and surfaces, and volumes have been
explored [15–17]. Paired point registration methods, whether based on scalp or bone fiducials or natural landmarks, are easily implemented and reasonably robust [18]. As presently implemented, however, they do have significant disadvantages, including a need for prospectively planned (meaning often additional) imaging, in the case of fiducial-based registration, and a manual participation in the registration process. It can be reasonably anticipated that the inefficiency of the latter will drive automation of the process. Laser-scanners, fluoroscopy, or camerabased systems, either autodetecting fiducial markers or employing feature/object recognition capabilities, can accomplish this task. A second area of development, already inherent in the use of intraoperative imaging devices, will be in updated or iterative registration to account for intraoperative changes in the operative field. Addressing such issues as shift or deformation, these measures have seen preliminary implementation [19–21] but will become more widespread. Further, methods imaging objects of interest directly will provide additional information regarding extent of resection. Lastly, such intraoperative information sources will extend their role to include in some instances intraoperative monitoring of normal brain function.
Intraoperative Imaging It is fairly obvious that the neuroimaging revolution that has fueled so much of investigational, diagnostic and interventional neuroscience represents one of the major developmental forces in image-guided surgery. Imaging both enriches the preoperative database co-registered with the operative field and provides in many instances intraoperative information directly. Imaging systems continue to decrease in cost, size, and portability, and for all of the reasons alluded to previously, the appeal to integrate their function
The future of neuronavigation
with therapeutic intervention will not lessen [22–24]. Conceptually, integration of more and more capable imaging into the operating room will have to balance the expediency of simply placing technology in the surgeon’s environment with cost-effective co-registration strategies that may in some instances accomplish similar ends.
Robotics For certain tasks, such as setting instrumentation along specified trajectories, robotics offers considerable advantage over manual manipulation. The accuracy, efficiency, and likely decreased probability of major error offered by robotic instrumentation make this attractive. The cost, size, and accessibility of this technology continue to trend favorably, and their incorporation will eventually be a logical, uncontested development. Numerous implementations in the operating room now exist, many but not all of them stereotactically registered [25–29]. Quantitative, non-iterative, efficient, stable platforms for instrument, probe, electrode or other effector positioning are difficult to argue with. The potential to bring enhanced capabilities – readily functioning at humanly impossible scale or speed – will initially improve upon present neurosurgical procedures but in the long-run will open up interventions not yet envisioned.
Telecommunications The ability to share data seamlessly across large distances could be employed to advantage in some situations. Access to additional information, whether anatomic, physiological, or normative databases could prove desirable intraoperatively, either routinely or in unanticipated circumstances. Relatedly, the ability to bring remote expertise for surgical assistance could be readily implemented [30]. For the young neurosurgeon learning a procedure or the more senior neurosurgeon
189
performing a rare operation, guidance from expertise at a different site may be invaluable. Coupled with robotic instrumentation, such links could provide actual technical intervention.
Applications It is fairly safe to anticipate widespread extension of frameless stereotactic methods to many additional surgical procedures, both neurosurgical and otherwise. Early adaptation in intracranial tumor surgery was logical given its ability to find small lesions and define extent of larger ones. If not already a standard, this utilization will almost certainly become one in the very near future. Its utility in this setting will only grow as all of the above technological advances are incorporated. Similarly, application in functional procedures, already facilitated by the familiarity of functional neurosurgeons with stereotactic methodology, will only continue. Traditional frame-based procedures such as biopsy needle or electrode placement have been transitioning to frameless methods, and refined instrumentation will continue this development. Functional procedures requiring craniotomy – notably resections for epilepsy – already benefit from these techniques, and incorporation of functional imaging and other physiological information will push further adoption of these methods. In cerebrovascular, the combination of more stereotypical, extra-axial pathology in which guidance is less essential, as in aneurysm surgery, and surgeons less familiar with or interested in traditional stereotactic methods has accounted for less application to date. But utilization in selected settings, such as approaches to more peripheral and atypical aneurysms or to vasculature feeding or draining vascular malformations has already been appreciated. Integration of finer detail or perspective at a complex aneurysm neck may also have appeal. Spine’s integration of these methods has paralleled its growth in
3139
3140
189
The future of neuronavigation
utilization of implanted instrumentation. Improvement in segmental registration as well as efficiency of use will facilitate further use. The greatest growth in application, however, will likely almost certainly be in applications outside neurosurgery. Registration methodologies applicable to extracranial soft tissue entail additional challenges, but extensive work has been done in non-rigid transformations and related fields that will one day enable other surgeons to derive benefits neurosurgeons have enjoyed [31,32]. Otolaryngology has been involved with image-guided surgery for many decades, particularly for procedures involving the cranial sinuses [33,34]. Plastic surgery has seen applications weighted toward the planning of craniofacial reconstructions and similar procedures [35], with subsequent development of intraoperative monitoring of the procedure. The latter, illustrative of the power of presurgical planning, clearly extends beyond plastic surgery, with orthopedics [36] and vascular immediately brought to mind. Surgical oncology ultimately may share procedural benefits with neurosurgery. Radiation oncology, of course, has been linked to stereotactic principles since the time of Leksell, and the development of radiosurgery has paralleled that of other stereotactic surgery. With the Cyberknife’s image-intensifier registration methods and robotically manipulated linear accelerator, radiosurgery has long been a frameless procedure [10]. Of greater note, however, is the adoption of stereotactic principles and associated improvement in treatment planning and delivery by the field of radiation oncology in other applications using fractionated therapy.
Implications for Treatment Planning and for Training The gap between surgical planning and its execution has been greatly narrowed by the widespread adoption of stereotaxy. This can be
relatively straightforward, as in tumor resection, or more elaborate as in a functional procedure for epilepsy. In the latter, multi-disciplinary teams utilizing information from structural imaging, functional imaging, electrophysiology, as well as neuropsychological considerations may jointly develop a treatment strategy; execution of that strategy with millimetric accuracy is replacing the standardized anatomic resection. Relatedly, much of the infrastructure underlying image-guided surgery – the platform to incorporate imaging studies, segment and otherwise process for better visualization and analysis, manipulate perspective, and in some instances actually remove tissue – also well serves planning and surgical visualization [37]. As a tool for surgical rehearsal or training, its power has not been fully appreciated. Surgical simulation has required more powerful and affordable computational resources to take-off, but stereotactic applications are well positioned to be at the forefront of this development.
References 1. Germano IM, Villalobos H, Silvers A, Post KD. Clinical use of the optical digitizer for intracranial neuronavigation. Neurosurgery 1999;45:261-9; discussion 269-70. 2. Barnett GH, Miller DW, Weisenberger J. Frameless stereotaxy with scalp-applied fiducial markers for brain biopsy procedures: experience in 218 cases. J Neurosurg 1999;91:569-76. 3. Smith KR, Frank KJ, Bucholz RD. The NeuroStation – a highly accurate, minimally invasive solution to frameless stereotactic neurosurgery. Comput Med Imaging Graph 1994;18:247-56. 4. Zaaroor M, Bejerano Y, Weinfeld Z, Ben-Haim S. Novel magnetic technology for intraoperative intracranial frameless navigation: in vivo and in vitro results. Neurosurgery 2001;48:1100-07; discussion 1107-8. 5. Mascott CR. Comparison of magnetic tracking and optical tracking by simultaneous use of two independent frameless stereotactic systems. Neurosurgery 2005;57: 295-301; discussion 295-301. 6. Cao A, Thompson RC, Dumpuri P, et al. Laser range scanning for image-guided neurosurgery: investigation of image-to-physical space registrations. Med Phys 2008; 35:1593-605.
The future of neuronavigation
7. Marmulla R, Eggers G, Muhling J. Laser surface registration for lateral skull base surgery. Minim Invasive Neurosurg 2005;48:181-5. 8. Heilbrun MP, McDonald P, Wiker C, Koehler S, Peters W. Stereotactic localization and guidance using a machine vision technique. Stereotact Funct Neurosurg 1992;58:94-8. 9. Sun H, Farid H, Roberts DW, Rick K, Hartov A, Paulsen KD. A noncontacting 3-D digitizer for use in image-guided neurosurgery. Stereotact Funct Neurosurg 2003;80:120-4. 10. Adler JR, Jr., Chang SD, Murphy MJ, Doty J, Geis P, Hancock SL. The Cyberknife: a frameless robotic system for radiosurgery. Stereotact Funct Neurosurg 1997; 69:124-8. 11. Henderson JM, Hill BC. Fluoroscopic registration and localization for image-guided cranial neurosurgical procedures: a feasibility study. Stereotact Funct Neurosurg 2008;86:271-77. 12. Amstutz C, Caversaccio M, Kowal J, et al. A-mode ultrasound-based registration in computer-aided surgery of the skull. Arch Otolaryngol Head Neck Surg 2003; 129:1310-16. 13. Muratore DM, Russ JH, Dawant BM, Galloway RL, Jr. Three-dimensional image registration of phantom vertebrae for image-guided surgery: a preliminary study. Comput Aided Surg 2002;7:342-52. 14. Haase J. Neurosurgical tools and techniques – modern image-guided surgery. Neurol Med Chir (Tokyo) 1998; 38 Suppl 303-07. 15. Friets EM, Strohbehn JW, Roberts DW. Curvature-based nonfiducial registration for the frameless stereotactic operating microscope. IEEE Trans Biomed Eng 1995; 42:867-78. 16. Bucholz R, Macneil W, Fewings P, Ravindra A, McDurmont L, Baumann C. Automated rejection of contaminated surface measurements for improved surface registration in image guided neurosurgery. Stud Health Technol Inform 2000;70:39-45. 17. Henderson JM, Smith KR, Bucholz RD. An accurate and ergonomic method of registration for image-guided neurosurgery. Comput Med Imaging Graph 1994;18: 273-77. 18. Heilbrun MP, Koehler S, MacDonald P, Siemionow V, Peters W. Preliminary experience using an optimized three-point transformation algorithm for spatial registration of coordinate systems: a method of noninvasive localization using frame-based stereotactic guidance systems. J Neurosurg 1994;81:676-82. 19. Comeau RM, Sadikot AF, Fenster A, Peters TM. Intraoperative ultrasound for guidance and tissue shift correction in image-guided neurosurgery. Med Phys 2000; 27:787-800. 20. Trobaugh JW, Richard WD, Smith KR, Bucholz RD. Frameless stereotactic ultrasonography: method and applications. Comput Med Imaging Graph 1994;18:235-46.
189
21. Roberts DW, Miga MI, Hartov A, et al. Intraoperatively updated neuroimaging using brain modeling and sparse data. Neurosurgery 1999;45:1199-206; discussion 1206-197. 22. Nimsky C, Fujita A, Ganslandt O, von Keller B, Kohmura E, Fahlbusch R. Frameless stereotactic surgery using intraoperative high-field magnetic resonance imaging. Neurol Med Chir (Tokyo) 2004;44:522-33; discussion 534. 23. Hall WA, Truwit CL. Intraoperative MR-guided neurosurgery. J Magn Reson Imaging 2008;27:368-75. 24. Schulder M, Sernas TJ, Carmel PW. Cranial surgery and navigation with a compact intraoperative MRI system. Acta Neurochir Suppl 2003;85:79-86. 25. Sutherland GR, Latour I, Greer AD, Fielding T, Feil G, Newhook P. An image-guided magnetic resonancecompatible surgical robot. Neurosurgery 2008;62:286-92; discussion 292-83. 26. DiMaio SP, Pieper S, Chinzei K, et al. Robot-assisted needle placement in open MRI: system architecture, integration and validation. Comput Aided Surg 2007;12: 15-24. 27. Varma TR, Eldridge P. Use of the NeuroMate stereotactic robot in a frameless mode for functional neurosurgery. Int J Med Robot 2006;2:107-13. 28. Eljamel MS. Robotic application in epilepsy surgery. Int J Med Robot 2006;2:233-7. 29. McBeth PB, Louw DF, Rizun PR, Sutherland GR. Robotics in neurosurgery. Am J Surg 2004;188:68S-75S. 30. Tian Z, Lu W, Wang T, Ma B, Zhao Q, Zhang G. Application of a robotic telemanipulation system in stereotactic surgery. Stereotact Funct Neurosurg 2008; 86:54-61. 31. Rohlfing T, Maurer CR, Jr. Nonrigid image registration in shared-memory multiprocessor environments with application to brains, breasts, and bees. IEEE Trans Inf Technol Biomed 2003;7:16-25. 32. Warfield SK, Haker SJ, Talos IF, et al. Capturing intraoperative deformations: research experience at Brigham and Women’s Hospital. Med Image Anal 2005;9:145-62. 33. Metson R, Cosenza M, Gliklich RE, Montgomery WW. The role of image-guidance systems for head and neck surgery. Arch Otolaryngol Head Neck Surg 1999; 125:1100-04. 34. Olson G, Citardi MJ. Image-guided functional endoscopic sinus surgery. Otolaryngol Head Neck Surg 2000; 123:188-94. 35. Demianczuk AN, Antonyshyn OM. Application of a three-dimensional intraoperative navigational system in craniofacial surgery. J Craniofac Surg 1997;8:290-7. 36. Tria AJ, Jr. The evolving role of navigation in minimally invasive total knee arthroplasty. Am J Orthop 2006; 35:18-22. 37. Stadie AT, Kockro RA, Reisch R, et al. Virtual reality system for planning minimally invasive neurosurgery. Technical note. J Neurosurg 2008;108:382-94.
3141
190 The Future of Radiosurgery and Radiotherapy L. Ma . P. K. Sneed
Radiotherapy and radiosurgery will continue to play an important role in stereotactic and functional neurosurgery in the future.
#
There have been tremendous technological advances in radiosurgery and radiotherapy treatment machines over the past few decades; this progress will continue, allowing improved targeting of radiation, improved sparing of normal tissues, and broader applicability of radiosurgery and highly focused radiotherapy. Integrated imaging with fast feedback to the treatment equipment and, in some cases, to the treatment planning system will assure more accurate delivery of radiosurgery and radiotherapy treatment. Advances in technology will allow complex problems to be solved more efficiently, so that radiosurgery and radiotherapy planning and delivery techniques of the future will provide straightforward and comfortable treatments to the patient without being excessively labor-intensive for medical personnel. The practice of radiosurgery and radiotherapy will also be enhanced by further improvements in anatomic, functional, and metabolic/ biological imaging techniques, with gains in types of information obtainable as well as in spatial and temporal resolution. Radiosurgery for functional indications such as movement disorders may become more widespread as functional imaging allows very precise noninvasive localization of targets. We will gain a better understanding of how to appropriately target infiltrating gliomas. Various imaging Springer-Verlag Berlin/Heidelberg 2009
datasets will be easily correlated with each other and readily integrated into the radiation treatment planning process. Imaging will more routinely become ‘‘four-dimensional,’’ taking into account the variable of time, to allow precise targeting of mobile, shrinking, or biologically changing targets. There will be continued gains in knowledge to allow refinements in case selection, target delineation, and selection of appropriate combinations of radiation with surgery, chemotherapy, molecular targeted therapy, gene therapy, and other new modalities, for both intracranial and extracranial indications. In some cases, new applications for radiosurgery will be developed, such as its use for refractory temporal lobe epilepsy, which is being compared with temporal lobectomy in a prospective randomized trial. There has been a long interest finding drugs to act as radiosensitizers to render tumors more sensitive to radiation or radioprotectors to render normal tissues less sensitive to radiation. One radioprotector, amifostine, is being using to a limited extent currently, but more agents are likely to become available in the future to protect against, treat, or mitigate radiation effects [1,2]. Perhaps neural progenitor or stem cells or molecular targeted therapy will be used to mitigate or treat brain and spinal cord radiation injury and edema [3]. Multidisciplinary collaboration and interaction will expand, maximizing benefits to patients from the advances in all of these areas.
3144
190
The future of radiosurgery and radiotherapy
Many of the topics mentioned earlier are discussed in other chapters in this section; this chapter will focus specifically on the future of radiotherapy and radiosurgery in terms of equipment, integrated imaging, motion management, and treatment planning systems. Frame-based radiosurgery equipment. Although frameless radiosurgery is becoming increasingly popular, frame-based radiosurgery will continue to be used for many years to come. A stereotactic frame provides rigid immobilization of the area to be treated and defines a threedimensional coordinate system to apply an action such as externally applied radiation to an internal target. The advantage of the frame is that once the coordinate system has been rigidly established, the isocenter of the radiation beam can be easily placed anywhere inside the target by merely shifting the patient in the frame along x-, y-, and z-directions without the need for any rotational corrections, because the system is by default Cartesian. Most commonly used frames are nonrelocatable, such as the Leksell G frame for Gamma Knife radiosurgery or the BRW (Brown–Roberts–Wells) head ring for linear accelerator based delivery; generally the frame is directly fixated onto the patient skull via four metal screws or pins. Relocatable frames commonly use a combination of a dentition mold, Velcro straps, and thermoplastic mask for immobilization, such as the GTC (Gill–Thomas– Cosman) frame. They facilitate dose fractionation and are less invasive, but also less precise than nonrelocatable frames. Frame-based radiosurgery can be performed using any high-energy radiation source (photons produced by a linear accelerator, gamma rays, or protons). Linear accelerator (linac) radiosurgical units commonly use noncoplanar converging arc beams aimed at the target or dynamic rotation with simultaneous couch and gantry rotation. Linac radiosurgery systems have tended to move away from multiple arcs with circular tertiary collimator cones toward the use of mini- or micromultileaf collimators (MLCs) (> Figure 190-1) to
. Figure 190-1 Micro MLC (Image courtesy of BrainLAB AG)
continuously change beam shaping during the course of arc travel to conform dose distributions to the target contour. Examples of such systems include the Novalis system from BrainLAB AG (Feldkirchen, Germany) (> Figure 190-2) and the XKnife™ system from Integra Radionics (Burlington, Massachusetts). The use of MLC shaped-beam radiosurgery tends to allow more uniform dose distributions and time-efficient treatment with few isocenter shifts for large or complex targets. In addition, the MLC allows intensity modulation of the radiation beams directed from multiple fixed angles. Traditional arc delivery has generally been limited to constantspeed gantry rotation with constant dose rate, i.e., constant dose per degree of rotation. Variable gantry speed with adjustable dose rate was recently made possible (RapidArc™, Varian Medical Systems, Inc., Palo Alto, California). The capability of varied dose rate delivery will significantly improve the efficiency and quality of traditional arc beam delivery in the near future. Most systems allow a combination of fixed intensity-modulated beams with arc beam delivery.
The future of radiosurgery and radiotherapy
190
. Figure 190-2 Novalis system (Image courtesy of BrainLAB AG)
Future systems will have refined delivery techniques allowing highly conformal treatment to be performed efficiently and accurately. While other manufacturers are moving away from frame-based radiosurgery, Elekta (Stockholm, Sweden) has intentionally maintained the use of a stereotactic frame with the Leksell Gamma Knife, setting the gold-standard for intracranial radiosurgical accuracy. The Gamma Knife uses 192 or 201 cobalt-60 sources in a quasi hemispherical distribution to produce 1.25 MeV gamma ray beams aimed at a fixed isocenter with a dose rate on the order of 300 cGy/min at the isocenter. The patient is moved to align the target with the fixed isocenter. Elekta recently completely redesigned the Gamma Knife with the ambitious aim of producing an ultimate radiosurgery tool with excellent dosimetry performance and radiation protection, unlimited cranial reach, fully automated one-push-button
treatment, outstanding patient comfort and safety, and superb reliability and serviceability. The resulting PERFEXION™ Gamma Knife (> Figure 190-3) is a prime example of how complex technology and elegant engineering solutions can simplify workflow and increase efficiency [4]. The large tungsten collimator body with built-in collimators and the couch patient positioning system allow comfortable, efficient treatment of one or more targets any place in the head. Previously, targets in extreme locations sometimes required uncomfortable patient positioning, stereotactic frame repositioning, or, rarely, abandonment of treatment. These problems have been virtually entirely eliminated. In addition, the treatment delivery has become much more efficient. The in-room time is now only a few minutes longer than the total beam-on time, whereas complex treatments on older Gamma Knife models required up to twice the
3145
3146
190
The future of radiosurgery and radiotherapy
. Figure 190-3 Perfexion Gamma Knife
beam-on time. This is the sort of advancement that is needed for future radiosurgery and radiotherapy equipment, making complex treatment more widely applicable, easier, and efficient. One notable innovation in Gamma Knife PERFEXION™ is its first-time introduction of independent source group motions; in all previous Gamma Knife models, all of the sources are fixed in position. Since linear accelerators predominantly use a single beam transporting line with a single radiation source for radiation production, the delivery of linac radiosurgery will most likely continue to favor a single isocenter or a small number of isocenters with enhanced beam arcing or modulation capabilities. In contrast, the Gamma Knife will continue to extend itself on the delivery paradigm of superposition of a large number of isocenters to create conformal dose distributions. With the introduction and future refinement of independent source group motions, the Gamma Knife system will gain substantial beam-shaping capability while maintaining its classic scheme of multi-isocenter delivery. Future development will further explore this new beam-shaping
capability for clinical applications such as improving the dose gradient for better sparing of critical structures.
Nonframe-Based Radiosurgery and Radiotherapy Equipment With the exception of the Gamma Knife, the distinction has blurred between equipment used for radiosurgery versus radiotherapy; an increasing number of systems allow delivery of one, a few, or many fractions of highly focused radiation without a stereotactic frame. To achieve high focal precision, the majority of nonframeless systems employ special or dedicated linear accelerators that possess submillimeter isocenter tolerance, milli- or micro-MLC (e.g., 3–5 mm leaf width), a high output dose-rate of 800–1000 cGy/min or even more, as well as add-on or on-board image guidance systems. The Novalis system from BrainLAB AG (Feldkirchen, Germany) (> Figure 190-2) is an example of a versatile linear accelerator system for frame-based cranial radiosurgery, or cranial
The future of radiosurgery and radiotherapy
or exracranial frameless radiosurgery, or hypofractionated fractionated radiotherapy [5,6]. The system is comprised of a 6 MV Varian linac with a built-in micro-MLC that has 14 pairs of 3-mm leaves at the center flanked on each side by 3 pairs of 4.5-mm leaves and 3 pairs of 5.5-mm leaves. As a result, the treatment fields range from 0.3 0.3 cm2 to 10 10 cm2. The current system allows radiation delivery via fixed open fields, intensity-modulated fields, conventional fixedfield arcs, or dynamic conformal arcs with the MLC leaves moving while the gantry rotates. Future systems will likely allow finer MLC resolution for larger-sized fields and incorporate variable speed gantry rotation to enhance overall treatment delivery accuracy and efficiency. The CyberKnife1 (> Figure 190-4) is a current and evolving robotic single-dose or fractionated radiosurgery system produced by Accuray, Inc. (Sunnyvale, California) [7,8]. Instead of a conventional linear accelerator that has a rotating gantry, a compact x-band 6 MV linear accelerator is mounted on an industrial robot arm with six degrees-of-freedom. During the course of . Figure 190-4 CyberKnife
190
treatment, the patient lies on a treatment couch and the robot arm aims at the target from hundreds of directions, using complex, nonisocentric beam patterns. For target localization, dual kV X-ray imagers are mounted on the ceiling (vs. beneath the floor for the Novalis system). Treatment proceeds in a repetitive stop, image, align, and shoot sequence. For targets within the head, a thermoplastic mask is typically used, but perfect immobilization is not required. The system is able to treat anyplace in the body, virtually free of patient size and positioning constraints. The imaging that makes this possible is described below in the section on ‘‘Integrated imaging.’’ Future systems will allow more beam directions but also choose beam directions more efficiently to decrease delivery time, improve treatment clearance and patient safety with regard to the moving robot, and reduce scattered dose and integral radiation dose outside of the target. Future systems will also automatically adjust for rotational setup errors, further improve the target tracking accuracy, perform corrections in true real time, and allow deformable target corrections.
3147
3148
190
The future of radiosurgery and radiotherapy
In contrast to Novalis shaped-beam delivery, the CyberKnife system still uses a limited number of tertiary cones to define beam diameters ranging from 5 to 60 mm. Recently, Accuray has introduced several mechanisms for automatic cone switching plus an integrated system that automatically adjust the cone diameters. With increasing research on shaped-beam delivery, we envision that the future system will merge flexible beam-shaping, perhaps with the addition of micro-MLC, with the sixdegrees-of-freedom robotic arm with unrestricted beam angles and source-to-target distance. The Tomotherapy unit (Tomotherapy, Inc., Madison, Wisconsin) is another futuristic radiosurgery/radiotherapy system (> Figure 190-5) [9]. A compact 6 MV linac accelerator mounted on a rotating slip ring delivers treatment in an axial fan-beam arrangement through 64 open or shut slit apertures as the patient couch moves through the bore of the machine, creating a beam passing through the patient in a spiral trajectory. Complex dose distributions can be created via the helical beam path around the patient. Currently,
. Figure 190-5 Tomotherapy
the system couch does not permit active or real-time six-degrees-of-freedom patient positioning correction, particularly in pitch, yaw, or roll angular rotations. Therefore, the initial treatment setup position needs to match the reference planning position as close as possible. With a 2p ring-and-detector configuration, the future system may overcome this limitation via treatment planning optimization on the fly, constant dose verification, and adaptive manipulation of the dose delivered to the patient. In addition to the above specialized systems, major linear accelerator manufactures have begun to introduce combination systems in attempt to integrate a wide-range of applications into one machine with high-definition, multiple beam energies, and dual-modalities (electrons and photons), such as the Trilogy™ System from Varian Medical Systems, Inc. (Palo Alto, California). The name Trilogy implies the ambitious combination of three modern radiotherapy technologies: intensity modulated therapy, image guided therapy, and stereotactic single-dose or multi-fraction
The future of radiosurgery and radiotherapy
radiosurgery. This system has a versatile 4–23 MV linear accelerator tube used with either tertiary cones or a built-in MLC. The system’s on-board imaging capability is described later. Other creative treatment delivery systems have been developed and will be developed in the future with the aim of providing accurate radiosurgery/radiotherapy to one or more targets with increasingly complex dose constraints and yet increasing ease of use compared with past systems. Ongoing advances in accelerator engineering are making protons and heavier ions, such as carbon, more affordable for use in clinical radiosurgery and radiotherapy. The depth dose profile of protons and heavy particles consists of increasing dose with depth of penetration up to a sharp maximum near the end of their range, called the Bragg peak, with essentially zero dose deposited in tissue distal to the sharp edge of the Bragg peak. This offers potential for improved dose distributions, and heavy particles have the additional potential advantage of increased radiobiological effectiveness [10]. Whereas gamma-ray and photon-based radiosurgery systems rely on a large number of convergent beams to create rapid fall-off of dose outside of the target, proton and heavy particle radiation units use a small number of beams with extremely sharp fall-off of dose at the distal edge of the target. This could be advantageous for tumors requiring high dose in close proximity to critical structures such as the optic chiasm or spinal cord. Particle machines are already becoming more widely available. Future particle systems will create more compact and cost-effective systems applying concepts from photon/electron beam radiotherapy such as beam intensity modulation, real-time feed-back adaptation, and improved range modulation capability to maximize the therapeutic ratio of charged particle delivery [11]. With wider availability of particle therapy, critical evaluation of clinical experience and clinical trials will be needed to determine appropriate case selection.
190
Integrated Imaging Nonframe-based radiosurgery/radiotherapy is more versatile but of course has the tradeoff of more uncertainty in terms of targeting accuracy. Such systems require the use of imaging to confirm accurate positioning prior to starting treatment and, in some cases, during or after each treatment fraction. The CyberKnife takes the futuristic approach of adjusting the treatment to small variations in the patient’s positioning rather than requiring the patient to be precisely positioned for treatment and rigidly maintained in that position throughout treatment. Orthogonal X-ray images captured by flat-panel detectors are obtained prior to commencement of treatment and every minute or so during the treatment. Within tens to hundreds of milliseconds, these images are compared with a pre-calculated library of digitally reconstructed radiographs based on thin-cut treatment planning computed tomography (CT) scan to detect small deviations in positioning. The comparison is based on the skull for intracranial targets, spine bony anatomy for spinal or paraspinal targets, soft tissue targets visible by X-ray, or implanted radio-opaque fiducial markers. The first set of images is used to establish patient positioning within an acceptable range prior to commencement of treatment. Repeat imaging during the course of treatment allows for the robot to alter treatment delivery as needed to adjust for measured small patient movements in x-, y-, and z-directions. Future systems will become more responsive to realtime changes, adjust for rotational setup errors, and further reduce the reliance on implanted fiducial markers. During frameless radiosurgical treatment with the Novalis system, a pair of X-ray images are taken with kV X-ray imagers mounted beneath the floor of the treatment room. The images are compared with the digitally reconstructed radiographs generated from the CT
3149
3150
190
The future of radiosurgery and radiotherapy
studies used for treatment planning. Because the patient is immobilized without a frame, the patient setup position generally differs from the reference treatment planning positions in both translational and rotational variables. To correct for this, the Novalis system uses a robotic couch that moves with full six degrees-of-freedom for any shifts along x-, y-, and z-axis and also any rotations around these three axes, i.e., roll, pitch, and yaw. To provide full three-dimensional imaging information, CT imaging can be built into radiosurgery or radiotherapy treatment machines. This yields much more information than a pair of twodimensional images, and has the potential for allowing more accurate treatment as well as adjustment of treatment to tumor shrinkage or other anatomic changes during the course of treatment. The Tomotherapy unit incorporates a kV X-ray tube and xenon detectors into the rotating ring that also delivers treatment, and a CT scan can be performed daily prior to and after treatment to ensure accurate daily setup. Cone-beam CT acquires imaging data on the actual treatment machine with a large rotating field rather than conventional CT which acquires data slice by slice. Cone beam CT scans can be performed with kVenergy used for conventional diagnostic imaging or with MV energy used for radiation treatment. MV cone-beam CT may be advantageous to minimize metal artifact from spine stabilization hardware (> Figure 190-6) or hip prostheses [12]. The Varian Trilogy System and Elekta Synergy System have a side-mounted kV X-ray unit and flat panel imager enabling conventional port films or conebeam CT imaging. Recently, an on-board imaging device was added to the side of the Novalis linear accelerator to allow for in-room cone-beam CT acquisitions for three-dimensional soft-tissue targeting. Future systems will allow for full integration of rapid two-dimensional planar radiographs with relatively slow three-dimensional cone-beam CT imaging acquisition. Another setup confirmation approach called digital tomosynthesis creates
. Figure 190-6 MV cone beam CT for spine
tomographic images based on limited angles of rotation, potentially decreasing time and dose exposure, yet providing sufficient three-dimensional information [13]. Radiosurgery and radiotherapy systems of the future will employ adaptive image-guidance,
The future of radiosurgery and radiotherapy
allowing imaging information to be used to automatically correct patient positioning with an adjustable couch or adapt the treatment plan and treatment delivery based on setup variations or changes that occur during the course of treatment [14]. Another important area of development will be in the application of nonionizing radiation such as magnetic resonance imaging or ultrasound for real-time on-line imaging or delivery verification. This will eliminate increasing concerns about radiation exposure to the patient when more frequent imaging studies are needed for tracking or managing target changes throughout the treatment course.
Motion Management Various approaches have been taken to apply radiosurgery to mobile targets such as small lung or liver tumors that move with respiration [15]. These movements can be large and somewhat complex, and the radiation tolerance of normal lung and liver tissue is quite limited, so that it is often unacceptable to simply treat with a large enough margin to encompass moving targets with static fields. With passive tracking delivery, the radiation beam may turn on only when the tumor movement is restrained (e.g., with assisted breathing control) or when the tumor moves into a favorable state (e.g., with respiratory-gated treatment). Recent integration of ‘‘four-dimensional’’ CT imaging capability (taking into account time) will refine the treatment planning process for traditional tracking delivery by better modeling the delivered dose and motion patterns. The CyberKnife was the first system to implement active tracking – beam movement in synchrony with the target movement. The flexible and responsive sixdegrees-of-freedom robotic arm of the CyberKnife and the active tracking method created a happy marriage since the robotic arm can easily ‘‘breathe’’ together with the patient. This has
190
dramatically shortened treatment time and improved the overall dose accuracy [16]. Other approaches for treating moving, deforming targets will become available in the future, such as adaptive MLC shaping, so that such targets will be treated easily and accurately.
Treatment Planning Future treatment planning will become simpler and more automated and yet also more powerful, with added capability of adapting a plan as needed during the course of treatment in response to changes in anatomy, organ motion, tumor shrinkage, and detected setup variations. As radiation treatment planning has become increasingly sophisticated over the past decade, it has also become much more labor intensive with the advent of ‘‘inverse planning’’ necessitated by treatment machines capable of creating very complex dose distributions. Inverse planning typically requires the outlining of many normal structures of interest as well as one or more targets and specification of dose/volume constraints for each structure. The computer then develops a complex treatment plan to accomplish those goals. Plans often require multiple iterations, with the planner adjusting dose constraints or adding factitious ‘‘tuning structures’’ to try to force the inverse planning algorithm to come up with a desired solution. Future software will automate many of these steps to make this process much less labor intensive, and it will incorporate algorithms to conform high-dose areas to targets and lessen integral dose to normal tissues. The planning software will also automatically optimize treatment time without sacrificing target coverage or normal tissue sparing. With the many degrees-of-freedom of complicated systems, intelligent optimization of a large number of free parameters will be necessary to ensure that treatment planning and delivery will not be unnecessarily complex or lengthy.
3151
3152
190
The future of radiosurgery and radiotherapy
Currently, treatment plans are evaluated chiefly by examining dose-volume histograms for various structures; future software will estimate probabilities of tumor control, acute effects, and late normal tissue complications based on generic patient data, with the capability of entering individual patient-specific parameters. Treatment planning of the future will also incorporate biological information and feedback from imaging obtained during the course of radiation to optimize and adapt the treatment plan as needed [17,18]. Imaging obtained during the course of treatment will allow treatment plans to be adapted automatically or semiautomatically to various factors such as detected setup variations, daily organ variation, tumor shrinkage, patient weight loss, and biological response to treatment. Biologicallyguided and adaptive image-guided radiotherapy will become routine practice.
Conclusions The use of radiosurgery has burgeoned, first for intracranial targets and more recently for extracranial targets. Radiosurgery and radiotherapy equipment will advance rapidly to allow accurate and efficient delivery of complex dose distributions to fixed, mobile, deformable, or changing targets in a wide variety of body sites. Frameless systems will incorporate more advanced integrated or on-board imaging to help ensure targeting accuracy. Treatment planning systems will become much more sophisticated and more automated, and they will integrate imaging and other data to allow image-guided, biologically guided, and image-adaptive radiotherapy. Many developments-in-progress have come at a cost of increased labor intensiveness in the short run, but further technologic advances are now yielding significant efficiency gains. This trend will continue, as evidenced by machines such as the Gamma Knife PERFEXION™. Equipment and treatment planning software for the delivery of radiosurgery
and radiotherapy will continue to become more sophisticated but also more integrated, automated, efficient to use, and widely available, benefiting patients, medical personnel, and healthcare systems.
References 1. Brizel DM. Pharmacologic approaches to radiation protection. J Clin Oncol 2007;25:4084-9. 2. Moulder JE, Cohen EP. Future strategies for mitigation and treatment of chronic radiation-induced normal tissue injury. Semin Radiat Oncol 2007;17:141-8. 3. Oh BC, Liu CY, Wang MY, et al. Stereotactic radiosurgery: adjacent tissue injury and response after high-dose single fraction radiation. Part II: Strategies for therapeutic enhancement, brain injury mitigation, and brain injury repair. Neurosurgery 2007;60:799-814; 4. Lindquist C, Paddick I. The Leksell Gamma Knife Perfexion and comparisons with its predecessors. Neurosurgery 2007;61:130-40; 5. Chen JC, Rahimian J, Girvigian MR, et al. Contemporary methods of radiosurgery treatment with the Novalis linear accelerator system. Neurosurg Focus 2007;23:E4. 6. Teh BS, Paulino AC, Lu HH, et al. Versatility of the Novalis system to deliver image-guided stereotactic body radiation therapy (SBRT) for various anatomical sites. Technol Cancer Res Treat 2007;6:347-54. 7. Adler JR Jr, Chang SD, Murphy MJ, et al. The Cyberknife: a frameless robotic system for radiosurgery. Stereotact Funct Neurosurg 1997;69:124-8. 8. Hara W, Soltys SG, Gibbs IC. CyberKnife robotic radiosurgery system for tumor treatment. Expert Rev Anticancer Ther 2007;7:1507-15. 9. Welsh JS, Patel RR, Ritter MA, et al. Helical tomotherapy: an innovative technology and approach to radiation therapy. Technol Cancer Res Treat 2002;1:311-6. 10. Schulz-Ertner D, Tsujii H. Particle radiation therapy using proton and heavier ion beams. J Clin Oncol 2007;25:953-64. 11. Karger CP, Jakel O. Current status and new developments in ion therapy. Strahlenther Onkol 2007;183:295-300. 12. Morin O, Gillis A, Chen J, et al. Megavoltage cone-beam CT: system description and clinical applications. Med Dosim 2006;31:51-61. 13. Wu QJ, Godfrey DJ, Wang Z, et al. On-board patient positioning for head-and-neck IMRT: comparing digital tomosynthesis to kilovoltage radiography and cone-beam computed tomography. Int J Radiat Oncol Biol Phys 2007;69:598-606. 14. Jaffray D, Kupelian P, Djemil T, et al. Review of imageguided radiation therapy. Expert Rev Anticancer Ther 2007;7:89-103.
The future of radiosurgery and radiotherapy
15. Giraud P, Yorke E, Jiang S, et al. Reduction of organ motion effects in IMRT and conformal 3D radiation delivery by using gating and tracking techniques. Cancer Radiother 2006;10:269-82. 16. Seppenwoolde Y, Berbeco RI, Nishioka S, et al. Accuracy of tumor motion compensation algorithm from a robotic respiratory tracking system: a simulation study. Med Phys 2007;34:2774-84.
190
17. Mageras GS, Mechalakos J. Planning in the IGRT context: closing the loop. Semin Radiat Oncol 2007;17:268-77. 18. Stewart RD, Li XA. BGRT: biologically guided radiation therapy-the future is fast approaching. Med Phys 2007;34:3739-51.
3153
Subject Index A AAV2 vector genome 1720 abdominal aura 2564 ablative spinal cord procedures 2159 ablative stereotactic surgery 1810 ablative surgery 1447, 2860 – basal ganglia 2855 – procedures 2125 – scientific investigation 2857 abscess management 623 AccuPoint 525 – Targeting Sphere 526 accuracy 443 – radiological 924 acetaminophen 2490, 2493 – paracetomol 2071 acetylcholine (ACh) 310 Ackerman, Abraham 203 acoustic nerve tumor, focused and conventional radiation 1151 acoustic neurinoma, non-spherical 1105 acoustic neuroma 932, 1014 acoustic tumor 1015 AC-PC plane 253, 573, 1241, 1671 acromegaly 1045 – radiosurgery 1177, 1178 action potential 3189 action tremor 1757, 1758 acute recording 2690 Addiction 146, 165, 166, 2360 adeno-associated virus (AAV) 3064, 3083 – vector 3091 adenocarcinoma 667 – breast 969 adenoma – nonsecretory 1042 – secretory 1043 adenoviral vector 3088 adenovirus 1720 adenovirus 3083 – oncolytic 3097 – oncolytic, improving potency 3098 – oncolytic, improving selectivity 3098 – mediated p53 gene therapy 760 adrenal medullary tissue, autologous transplantation 14 adverse effects, CNS tissue, treatment-related 1135 AESOP 3000 593 afferent 1992 Afshar 354
#
Springer-Verlag Berlin/Heidelberg 2009
aggressive behavior 166 – surgery 2971 – surgery, microrecording 2973 – treatment 2971 Aicardi syndrome 3213 AIDS 657 akinesia 1497 Albe-Fessard, D. 100 alcohol abuse 1532 alcohol injection 8, 11, 2527 alfentanil 1339 algorithm for the probabilistic segmentation 382 Allen, Horace N. 171, 172 allergic reactions 284 allodynia 2024, 2109 alpha cell kill 1115 alpha/beta ratio 1115 alprazolam 1744 Alzheimer’s disease, gene therapy 3075 amantadine 1516 American Association of Neurological Surgeons (AANS) 135 American Society for Stereotactic and Functional Neurosurgery (ASSFN) 37,40 – meetings 40 American Society of Testing and Manufacturing 621 amino acid uptake 316 amplicon 1720 amplifier design 1259 amplitude 1258, 1391 amygdala 2696, 2697, 2701 amygdalo-hypocampectomy, selective 2677 analgesia 2109 – mechanisms 2027 analgesic therapy 2069 analgesics 2493 anatomic structures, segmentation 707 anatomic targeting 1586 anatomy, venous 707 Andrew 354 anesthesia – cardiovascular parameters 1335 – complications 1335 – dolorosa 19, 2109–33, 2427 – functional neurosurgery 1331 – general 1334 – intraoperative considerations 1333 – local 1333 – neurological complications 1336 – pediatric patients 1344
3240
Subject index
– postoperative care 1344 – preoperative assessment 1332 – procedure 1332 – respiratory complications 1335 aneurysm 299 – arterial 1071 Angina Arachnoiditis angiogenesis 996, 3096, 3206 – inhibition 746 – pathways 997 angiogram – postoperative 302 – preoperative 302 – stereotactic 1063 – vertebral 1079 angiographic localizer 473 angiographically occult vascular malformations (AOVM) 1133 angiography 253, 299, 823, 909, 1060 – indication 299 – limitations 304 – plane film 1095 – rotational 303 – techniques 299 angioma, venous 1070 angiostatin 3083 anisotropy 1739 ansotomy 1472 anterior capsule 2958 anterior limb of internal capsule 2927 anterior nucleus – deep brain stimulation 2793 – deep brain stimulation, EEG recording 2797 – deep brain stimulation, effect on seizure frequency 2798 – deep brain stimulation, electrophysiological findings of recordings 2797 – deep brain stimulation, neuroanatomical rationale 2793 – deep brain stimulation, surgical technique 2795 anterolateral system 2150 anticholinergics 1516, 1776 anticonvulsants 774, 1744 antidepressants 2207, 2501 anti-epileptic drug 800, 2071, 2669, 2672 – side effects 2673 antigen-presenting cell (APC) 756, 758 antimicrobial agents 771, 773 antineoplastic therapy 2068 anti-seizure medication (ASM) 2679 antisense therapy 761 anxiety 1520, 1744 – treatment-refractory chronic 2862 apathy 1492 apnea, phrenic nerve stimulation 2991 Apokyn 1512
apomorphine 1512 apoptosis 3204 – inducing gene 3087 Arau´jo Barros, Jose´ de 217, 218 Arcadis Orbic 713 arc-centered principle 469 Archimedes screw 478 arcquadrant principle 88 arm – articulated 702 – movement 3055 aromatic amino acid decarboxylase (AADC) 1726 Artane 1516 artery, small 304 arteriovenous malformation (AVM) 263, 515, 693, 887, 931, 941, 943, 944, 969, 1007, 1059, 1131, 2642 – carotid angiograms 1010 – cysts 1078 – dose selection 942 – dural, total obliteration 1069 – midbrain, total obliteration 1061 – patial obliteration 1062 – radiosurgery, adverse radiation effects 1011 – radiosurgery, probability of obliteration 1010 – radiosurgery, risk of hemorrhage 1009 – spinal 1218 – staged volume radiosurgery 1011 – stereotactic image acquisition 942 – stereotactic radiosurgery 980 – subtotal obliteration 1063, 1065 – subtotal obliteration, clinical outcomes 1065 – subtotal obliteration, imaging 1065 – subtotal obliteration, parameters 1065 – surgery 299, 301 – Sylvian fissure 1062 – unobliterated 1064 – vessel after Gamma Knife surgery 1041 aryl hydrocarbon receptor nuclear translocator (ARNT) 997 Ashworth scale 1974 Asian Society for Stereotactic, Functional and Computer Assisted Neurosurgery (ASSFCN) 44, 135, 177 asleep awake asleep technique 1340 asleep craniotomy 325 – indications 325 – surgical technique 327 aspirin 2490 assessment, atlas-assisted postoperative 410, 425 ASSFN 37, 38, 40–43, 120, 244, 628 astrocytes 666 astrocytoma 666, 669, 1021, 1058 – anaplastic 717, 3215 – brain stem, radiosurgery 785 – high grade 1055 – low grade 689, 712, 1054 – pilocytic 689 – thalamic pilocytic 691
Subject index
Atlantic Research System microdrive 1637 atlas – anatomical and probabilistic 395 – anatomical, limitations 409 – benefits 422 – combined anatomical-functional 420, 425 – computerized 399 – 2-D printed 377 – 3-D electronic 379 – functional population-based 385 – future stereotactic 430 – PFA-centric combined 420 – planning with multiple 409 – SW-centric combined 420 – use 422, 423 – use benefits 426 atlas-assisted quantification 398 atlas-based applications 427 atlas construction, community-centric 425 atlas deformation 396 atlas labeling 396 atlas representation 397 atlas segmentation 396 atlas structure 396 atlas-to-data registration 432 atlas-to-scan registration 421 atlas-to-patient deformation, validation 389 Atlas View 573 atrophy, multiple systemic 1609 attention deficit hyperactivity disorder (ADHD) 2860, 2963 atypical facial pain 1919, 2097 atypical rhabdoid teratoid tumor (ATRT) 3218 audiometry 1153, 1156, 1161 audiotactic surgery 732 auditory aura 2563 auditory brainstem implant 3188 auditory midbrain implant 3188 auditory pathways 1298 auditory performance 3030 auditory prosthesis 3021 – surgery 3026 auditory rehabilitation aura 2563 Austin and Lee three-piece ball-and-socket stereotaxic instrument 522 Austin-Lee device 529 Austrege´silo Rodrigues Lima, Anto˜nio 198 automatic endoscopic system for optimal positioning (AESOP) 593 automatic positioning system (APS) 903, 1228 automatic tube currentmodulation (ATCM) 280 automotor 2566 autonomic aura 2564 autonomic dysfunction 1518 autonomic pathway, interruption 2525 Avastin 754, 3220
average cerebral directional line 349 Avison, Oliver R. 173 awake craniotomy 327, 1338, 2635 – indications 327 – mapping, internal capsule 329 – preoperative functional imaging 330 – preoperative neurological evaluation 329 – surgical technique 330 – tumor surgery 1342 – tumor surgery, anesthetic agents 1343 – tumor surgery, complications 1343 – tumor surgery, preoperative assessment 1342 – tumor surgery, techniques 1343 Azilect 1515
B Backlund biopsy set 477, 478 Backlund hematoma evacuator 478 Backlund, Erik-Olof 69 baclofen 1776, 1866 – administration 1952 – intrathecal 1973, 1975 – intrathecal, complications 1978 – intrathecal, side effects 1977 – intrathecal, withdrawal, differential diagnosis 1979 Bacteroides 769 Bailey, P. 351 ball-and-socket plate 524 ball-and-socket probe holder, burr-hole mounted 521 Ballantine, H.T. 2878 ballism 1455 – differential diagnosis 1457 balloon compression 2461 – trigeminal neuralgia 2457 – trigeminal neuralgia, complications 2462 – trigeminal neuralgia, operative procedure 2458 – trigeminal neuralgia, post-operative management 2462 – trigeminal neuralgia, pre-operative preparation 2458 Barbosa, Renato Tavares 201 Barcia-Salorio, J.L. 186 basal ganglia 369 – circuit anatomy 1497 – function, simplified model 1475 – Meyers’ open surgical approach 8 – neuronal activity changes 1498 – neurotrophic factor distribution 1741 – sensorimotor circuit, functional organization 1780 – thalamocortical motor circuit 1780 basal ganglia-thalamo-cortical circuitry 1580 basal ganglia-thalamo-cortical motor circuit, Parkinsonism-related changes 1499 base ring 471, 474, 475, 480 BCNU 743 beam channel 921
3241
3242
Subject index
behavioral avoidance test 2907 behavioral disorders 2862 Beijing Neurosurgical Institute 128 Beijing San-Bo Brain Science Institute 144 Bell-Magendie law 1959 Benabid, A. 21, 40, 50, 88, 97, 495, 1603 benign tumors 1042 benzodiazepine 1334, 1744, 1776 benztropine 1516, 1776, 1865 Bergmann, Ernst von 53 Bertrand, C. 48, 113, 116, 117, 209, 217 Bertrand procedure 1887 Bertrand, G. 115 Bertrand’s technique 1896, 1898 best target point – coordinates 423 – atlas-derived 422 beta blocker 2495, 2501 – contraindications 2495 beta cell kill 1115 beta-amyloid protein 3075 beta-endorphin tract 2058 Betti, Osvaldo 102 bevacizumab 754, 998, 3220 BG operation 181 bilateral percutaneous cordotomy 2145 binary collimator system 970, 971 biocompatibility, neurostimulation implant 1409 bioheat transfer equation 1429 bioimpedance 631 – electrical 631 – historical aspects 632 biologically effective (equivalent) dose (BED) 857, 864, 1115, 1117, 1162 BION microstimulator 3049 Bionic Glove 3055 biopsy – high grade neoplasm 665 – image-guided 645 – inflammation, handling 675 – instrument 577 – kit 511, 513, 514 – low grade glioma 669 – mass lesions 521, 665, 671 – mass lesions, HIV 672 – neoplasms 671 – procedure 514 – recurrent lesion after radiation for glioma 675 – stereotactic 645 – target 648 – types 663 biplanar magnet design 821 bipolar electrode stimulation 331 bisphosphonate 2071 bitmap representation 398 bladder dysfunction, deep brain stimulation 2999
blending 340, 341 blepharospasm 1771, 1872 blood loss, sudden 2754 blood oxygen level dependent (BOLD) imaging 2623 blood oxygen level-dependent (BOLD) changes 288 blood supply preservation 2716 blood brain barrier (BBB) 739, 749, 991, 1508, 3175 Blundell, John 115 bone fiducial 577 bone flap 2754 bone pain 1111 Borke 357 Bosch, Hieronymus 2869 botulinum toxin 1455, 1865 – injection 1892, 1940, 1952 brachytherapy 102, 740, 791 – stereotactically guided, brain metastases bradykinesia 1334, 1497, 1531 Bragg Peak 957, 978 – curve 854 brain abscess 624 – antimicrobial agents 773 – aspiration 773 – clinical presentation 770 – conventional stereotactic aspiration 772 – differential diagnosis 770 – etiology 769 – excision 771 – freehand aspiration 772 – image-guided management 769 – incidence 769 – methicillin-resistant Staphylococcus aureus (MRSA) 773 – multiple 775 – open evacuation of PUS 772 – operative technique 772 – pathogenesis 769 – reaspiration 774 – sources 770 – treatment, nonstereotactic methods 770, 771 – treatment, nonsurgical 771 – treatment, stereotactic methods 772 brain activity, unspecific 308 brain atlas 395 – anatomical 396 – anatomical, computerized 396 – anatomical, printed 396 – probabilistic functional 413 brain atlas for functional imaging (BAFI) 429 brain atrophy 1532 brain biopsy 620 – CT/MRI image guidance, impedance monitoring 633 – stereotactic image-guided, contraindications 647 – stereotactic image-guided, indications 647
Subject index
brain collapse 1964 brain cyst – management 621 – stereotactic management 622 brain disorders 634 brain extracellular fluid (ECF) 740 brain imaging 461 – studies 2034 brain impedance 1825 brain injury – parenchymal, radiosurgical induced 1184 – traumatic 3122 brain mapping 114 brain metastases 708, 833, 987, 1016, 1022, 1124, 1135 – prognosis 1139 – radiosurgery 1141 – solitary, surgical extirpation 1055 – stereotactic radiosurgery 982 brain microrecording 635 brain neoplasm, benign 1015 brain pace maker 261 brain shift 304, 683, 701, 731, 2755 brain stem 779, 355, 656, 2125 – abscess, stereotactic aspiration 792 – activity, changes 1501 – auditory evoked potential (BAEP) 2467 – frameless image-guidance 785 – glioma 880, 1054 – lesions, focal 1759 – lesions, image-guided management 779 – lesions, interstitial brachytherapy 791 – lesions, MR images 790, 791 – lesions, providing histological diagnosis 789 – lesions, radiosurgery 786 – lesions, stereotactic biopsy 780 – macrostimulation 1308 – migraine generator 2488 – neuronavigation for surgery 783 – pyogenic abscess 792 – stereotactic approaches 789 – tumors, imaging 779 – tumors, management strategy 779 – tumors, open surgical approaches 784 brain surgery, therapeutic CT scanner 620 brain tissue impedance 1325 brain tumor 314, 634, 635, 699, 2641 – gene therapy 3068, 3083 – histologic diagnosis 735 – localization 633 – malignant 1054 – management 735 – novel therapies 749 – resection 592 – resection, image guided 23 – retrovirus 3062
– surgery 103 – surgery, conventional image guided 726 – surgery, virtual reality image guided 727 BrainBench 428 brain-computer interface (BCI) 3185 brain-derived neurotrophic factor (BDNF) 1697 brain-function visualization 256 BrainLab image guided system 567 – applications 571 – history 568 – technology 568 BrainLab stereotactic biopsy, components 578 BrainLab vectorvision system 703 brain-machine interface (BMI) 3185 Brainsterium 433 BrainSUITE 569, 579, 705, 712 – intraoperative MRI 715 – intraoperative MRI, head-holder 716 Bravo, G 184, 185 Brazilian League Against Epilepsy 203 Brazilian neurosurgeons 239 Brazilian Society for Stereotactic and Functional Neurosurgery 210, 237 Brazilian Society of Neurosurgery 200, 236 breast cancer detection 634 breast metastasis, conventional fractionated irradiation 1213 breathing 2991 bremsstrahlung 853 brightness modulation 3014 Broca’s area 263 Broggi, G. 193, 774, 2971 6-bromodeoxyuridine (BUdR) 989 Brown, Russell 458 Brown-Roberts-Wells (BRW) – apparatus 23 – base ring 459 – CT localizer 1091 – frame 459, 816 – frame, prototype 458 – headring 1090 – localizing box 255 – phantom base 459, 547 – stereotactic frame 453, 456, 462 – system 648, 704, 1089 bubble device 910 Budde Halo retractor system 525, 526 Bucy, P. 10, 45, 53, 203, 1256, 1467, 1665, 2744, 2749 bulbar DREZ Procedures 2125 bupivacaine 1339 Burckhardt, Gottlieb 2870 buried visual cortex 3016 Burkitt’s lymphoma 3101 burr holes 477, 1245 Burster cells 1590
3243
3244
Subject index
Burzaco, J. A. 187 butalbital 2493
C C1/2 nerves, microsurgical anatomy 1897 C1/2 rhizotomy 1899 cabergoline 1511 Cairns, Hugh 2877, 2878 Calan 2499 calcium channel blocker 2496 calculations 272 camera system 819 campotomy 13 camptothecin 754, 993 Canada 113 Canadian Neurosurgical Society 121 cancer – infiltration of bone 2065 – infiltration or compression of nerve tissue 2065 – invasive 2061 cancer therapy 2063 Cao Cao 125 capacitance 1325 capsaicin 1996 capsule, internal 2058 capsulotomy 69 – anterior 2878, 2901 – gamma knife radiosurgery 1197 – obsessive compulsive disorder 146 – stereotactic anterior (AC) 2948 carbamazepine 1776, 2673 carbidopa 1508, 1510, 1513 carbon ion radiotherapy 890 carbon nanofiber array 639, 640 carbon nanotube 638 – electrodes, advantages 638 carboplatin 994 carcinoma, metastatic 667 cardiac pacemaker 3047, 3056 cardinal reference planes 352 carmustine wafer 1249 Carney’s complex 3213 Cartesian coordinate system 3, 251, 252, 443, 499, 604 Cartesian frames 98 Cartesian planes 4, 7 Cartesian robotics system 681 CAS-BH5 robotic system 593 CASMIL 428 catechol-O-methyl transferase inhibitor 3162 catheter design 1738 catheter placement 521 caudalis drez lesions 2113 – complications 2115–2117 – results 2114, 2116
caudalis trigeminal nucleotractotomy, computer templates 2118 caudate confusion hypothesis 1614 caveats 720 cavernous angioma 2642 cavernous malformation 1012, 1068, 2768 cavernous sinus meningioma 1125 cavernous sinus surgery, stereotactic neuronavigation and doppler 825 CCBA-Plan 417, 430 CCNU 743 CD-ROM application 429 celiac ganglion 2300 – anatomy 2300 – neurolytic block 2300 cell cycle 859 – control 3203 – disorders 3203 – regulation 3203 cell growth, control 3202 cell migration, inhibition 746 cell proliferation, control 3202 cell transfer, immunological 758 cell transplantation, future 3161 cell, primate STT 1994 cellular differentiation 3205 cellular proliferation, inhibition 745 center-of-arc principle 488 center-of-arch system 498 central alveolar hypoventilation 2992 central auditory brain stem prostheses 3041 central body 900 central lateral thalamotomy 2081 – rationale for the selective central nervous system (CNS) 2375 – disorder, stem cell therapy 3170 – electrochemistry 640 – electrodes, limitations 637 – lesions, benign 1131 – lesions, malignant 1135 – prostheses 3185 – surgery, impedance recording 631 – tissue identification 631 – tissue identification, impedance monitoring 632 central thalamic electrical stimulation 2988 centroid coordinates 423 centromedian thalamic stimulation 2777 – electrocortical potentials 2780 cerebellar circuit 1791 cerebellar cortex stimulation (CCS) 2823 – safety considerations 2833 – stimulation parameters 2831 – studies 2830 – surgical procedure 2829 cerebellar cortex, implanting dual cerebellar electrodes 2826
Subject index
cerebellar nuclei 355 cerebellar stimulation (CS) 2823 – clinical reports 2827 – studies 2825 cerebellum 1790, 2930 cerebral artery aneurysm 303 cerebral blood flow (rCBF) perfusion 2841 cerebral cortex 1307, 2004 cerebral localization 2677 cerebral mechanism 2031 cerebral metastases 831, 37 – image guided management 831 – multiple, stereotactic radiosurgery (SRS) 842 – multiple, surgical resection 835 – stereotactic radiosurgery (SRS) 839 – surgery vs radiosurgery 842 – surgical adjuncts 836 – surgical resection 833 – treatment algorithm 845 – whole brain radiotherapy (WBRT) 833, 834, 840, 841 cerebral palsy 1847, 1926 cerebral peduncle 326 cerebral tissue, critical mass 2715 cerebrospinal fluid (CSF) – depletion 1964 – drainage 799 – impedance 1326 Cerefy anatomical brain atlas database 402 Cerefy atlas 400 – commercial neurosurgical workstations 415 – geometrical models 404 Cerefy clinical brain atlas 429 Cerefy Electronic Brain Atlas Library 404 Cerefy Neuroradiology Atlas 428 Cerefy Schaltenbrand and Wahren atlas 408 Cerefy Talairach and Tournoux atlas 406 CereTom mobile CT unit 713 cervical dystonia 13 cervical muscles – anesthesia 1894 – denervation 1895 – innervation 1889, 1892, 1893 characterization, histopathologic 646 charge density 1412, 2782 charge monitoring, electrical 637 Chejungwon 171, 172 chemopallidectomy 12, 48, 161, 180 chemothalamectomy 12, 180 chemothalamotomy 174 chemotherapeutic agents, direct delivery 750 chemotherapy 743 – wafer 740 – tailored 743 China 125
Chinese Society of Stereotactic and Functional Neurosurgery 134 Chinese Stereotactic Neurosurgery Institute 133 Chintan Nambiar 156 chlorhexidine 808 chlorpromazine 2875 cholinergic system 310 chondroma 1059 chondrosarcoma 1026, 1059 – proton therapy 961 chord electrode 492 chordoma 1026, 1058 – proton therapy 961 chorea 1455 – differential diagnosis 1457 choreoathetosis 1847 choroidal artery ligation 1471 Christian Medical College (CMC), Vellore 156 chronic brain injury, central thalamic deep brain stimulation effects 2987 chronic recording 2688 chronic stimulation 2782 Chu Kul Lee 171 Cicerone 429 cilengitide 999 cingulotomy 2878 – anterior 2900, 2946 – complications 2893 – major depression 2887 – neurobiological basis 2887 – obsessive compulsive disorder 2887 – patient selection 2889 – postoperative care 2890 – surgical technique 2889 cingulum 1307 cintredekin exotoxin (PE) 751 circuit, equivalent 1262 cisternography 2440 Clarke and Horsley’s primate stereotaxic apparatus 79 Clarke, Robert Henry 3, 77, 78 Clarke’s stereoscopic instrument 77 clearance check tool 912 clinical pain situation 2023 clinical pain, types 2198 clinical target volume (CTV) 1206 clivus 1058 clomipramine 2899 clonazepam 1744, 1866 cluster headache 106, 2033, 2478, 2485, 2517 – chronic, surgery 2501 – chronic, surgery, indications 2501 – deep brain stimulation 2517 – diagnosis 2517 – diagnostic criteria 2486
3245
3246
Subject index
– hypothalamic stimulation 2517 – macrostimulation 2501 – microrecording 2520 – open surgery 2527 – pain, neural network 2523 – prophylactic treatment 2499 – stereotactic implantation 2519 – surgical procedures 2502 – surgical treatment 2525 – treatment 2497 clusterectomy 2664 coagulation, computer modeling 1428 coating, polypyrrole (PPy) 639 coating, type IV collagen 639 cobalt–60 decay 897, 898 cobalt–60 sources 900 cochlear apex, innervation 3026 cochlear implant 3021, 3188 – activation 3030 – children, educational placement 3038 – children, language acquisition 3036 – cost-effectiveness assessment 3038 – infection 3029 – programming 3030 – quality of life 3038 – system 3024 cochlear implantation – auditory rehabilitation 3040 – bilateral 3035 – pediatric 3033 cochlear nucleus, direct stimulation 3042 Codman 3000 implantable drug pump 3157 Cogentin 1516 cognitive assessment technique 2596 cognitive behavioral therapy (CBT) 2899, 2925 cognitive function 1573 – aberrant organization 2591 cognitive impairment 1456 collagen, type IV 639 collateral sprouting 1928 collimation system 897, 900, 901, 969 collimator – binary 971 – body 1008 – helmet 481, 902 – multileaf 971 – Perfexion system 1227 colloid cyst 625 colon metastasis, conventional fractionated irradiation 1212 color-coding 396 columnar organization 2715 coma 107 commissioning 920 commissural myelotomy 2164 communication theory 449
COMPASS – headring 1637 – stereotactic frame 681 – stereotactic system 24, 685, 700, 726 Compazine 2492 compensator, physical 968, 970 complex partial seizure 2778 complex regional pain syndrome 2303 complications of intrathecal drug delivery system – bleeding 2176 – cerebrospinal fluid leaks 2177 – infections 2178 – necrosis 2178 – neurologic injury 2177 – seroma 2178 – skin perforations 2178 composite map 362 composite shot 922 compressing vessel 2470 compulsion 2898 – DSM-IV definition 2860 computer 3131 computer display 704 computer tomography (CT) 86, 254, 279, 475, 504, 651, 2617 – advantages 568, 713 – angiogram 300 – angiography 301 – angiography neuronavigation 299 – basic principles 269 – cisternography 610 – contrast enhanced 648 – enhanced scans 281 – estimations of radiation-related cancer risk 279 – fluoroscopy, real time 627 – image guided surgery 619 – imaging 1206 – intraoperative imaging, limitations 822 – limitations 2617 – localization 1096 – MRI fusion technique 825 – percutaneous cervical cordotomy 2149 – preoperative volumetric 807 – radiation parameters 279 – safety in functional neurosurgery 279 – safety of ionizing radiation 279 – scanning 22 – scanning, functional stereotaxis 272 – stereotactic principles 255 – strategies for reducing radiation dose 280 computer work station 819 computer-generated slice image 682 COMT (catechol-O-methyltransferase) inhibitor 1512 COMTAN 1512 Concorde position 1964 conductivity 1429
Subject index
cone beam – CT scanner 823 – imaging 1209 Cone, William 113 conformal dose planning, techniques 1233 conformality 1116 conformity index 867, 1116 – new 1116 connectivity 2084 connexin 3091 conscious sedation 1334, 1339 conscious state, central thalamic deep brain stimulation 2986 consciousness 1308 console 904 constipation 1518 contouring 396 – tool 905 contralateral thermoanalgesia 2109, 2112 contrast administration 284 Contremoulins 100 control procedure, pre-implantation 507 control, moment-to-moment 589 convection-enhanced delivery (CED) 740, 751, 1249 Cooper, I. S. 182 – balloon 156, 161, 180 – instrument 70 – Irving 11, 180 coordinate frame, application 473 coordinate system of the Leksell model G stereotactic base ring 472 coordinate system, intracerebral 349 coordinate x-ray indicator 473 co-planar arc 978 cordotomy 457, 1468, 2162 – cervical 16 – high cervical, procedures 1468 – percutaneous, injection of the contrast material 2153 – percutaneous, insertion of the needle electrode system 2153 – percutaneous, lesion 2155 – percutaneous, physiologic localization 2155 – percutaneous, positioning 2153 – percutaneous, preoperative preparation 2153 co-registration – point based 444 – surface based 447 – volume based 449 corneal pain sensation 2105 coronal plane 355 corpus callosotomy 2723 – complications 2734 – pathophysiology 2726 – preoperative considerations 2728 – rationale 2723
– side effects 2734 – surgical considerations 2729 corpus callosum 2723 Correˆa, Cla´udio Fernandes 214, 215 cortex 1606 – hypermetabolism 1792 cortical ablation 1467 cortical activation, fMRI images 293 cortical activity, changes 1501 cortical depolarization 331 cortical function, test 2685 cortical localization 1274 cortical mapping technique 294 cortical resection 203 cortical stimulation 105, 1625 corticosteroid 2071, 2500 cortico-striato-thalamo-cortical circuitry (CSTC) 2898 Cosman calculations 460 Cosman Model G4 1361 Cosman Model RFG–1A 1361 Cosman Spinal Kit 1372 Cosman, Bernard J. 1360 Cosman, Eric 460, 462 Cosman-Roberts-Wells (CRW) – arc system 461 – base frame 648 – stereotactic frame 453, 461, 462, 600, 816, 1584 – stereotactic frame, original drawings 461 – system 648 – system, current, phantom base 464 – system, software 465 – system, versatility 463 cost-utility 3039 coverage 1116 – index 867 Cowden’s disease 3211 CP-PO line 349 cranial landmark 351 cranial nerve 2103 – sensitivity 1039 cranial neuropathy following radiosurgery 1183 cranial planning unit, mobile 569 craniocerebral lesions 1848 craniopharyngioma 670, 1020, 1046, 1047 – proton therapy 959 – suprasellar 818 craniotomy 679, 800, 2245 – flaps, planning 521 – image-guided 623, 699 – open 647 Crue, B. 16, 2106, 2110, 2133, 2394 Cryoprobe 12 C-SW atlas 401, 403 – contour representation 408 – coronal microseries 407 C-TT atlas 397, 399
3247
3248
Subject index
cumulative dose 1159 cuneatus nucleus 2108 Cushing’s disease 70, 1043 – Gamma Knife radiosurgery 1182 – radiosurgery 1175, 1176 cutaneous stimulation device (CSD) CyberKnife 586, 610, 879, 880, 949, 970, 3147 – clinical aspects 1111 – configuration 950 – definition 1111 – efficacy 882 – illustrative cases 1118 – neuro-surgical considerations 1117 – performance features 881 – quality 882 – registration system 881 – robotic radiosurgery system 973 – radiosurgical system 1208 – stereotactic radiosurgery 1111 – stereotactic radiosurgery, brain metastases 1124 – stereotactic radiosurgery, cavernous sinus meningioma 1125 – stereotactic radiosurgery, complications 1124 – stereotactic radiosurgery, non-neurosurgical tumors 1127 – stereotactic radiosurgery, pain 1122, 1123 – stereotactic radiosurgery, plan 610 – stereotactic radiosurgery, quality of life 1122, 1123 – stereotactic radiosurgery, survival 1124 – stereotactic radiosurgery, treatment of intracranial lesions 1124 – stereotactic radiosurgery, vestibular schwannoma 1125 – team 954 – technical Aspects 949 – treatment plan 884, 1121 cyclophosphamide 3097 cyclotome 78 cyclotron 68 Cygnus frameless stereotactic system 535 Cygnus software 539 cyproheptadine 2496 cyst fenestration 811 cyst formation 1011 – delayed 1075, 1076 cyst localization 633 cyst puncture needle 478 cystic lesion, aspiration 521 cytokine 757 – modulator therapy 757 cytotoxin, hypoxic 987
D 3-D (three-dimensional) radiological imaging 335 3-D (three-dimensional) coordinates 1243 3-D (three-dimensional) C-TT atlas 400
3-D (three-dimensional) frame tracking 257 3-D (three-dimensional) model 300 3-D (three-dimensional) rendering 1098 3-D (three-dimensional) rotational angiography 303 3-D (three-dimensional) ultrasound scan 301 4-D (four-dimensional) imaging 820 6-D fiducial tracking 951 6-D skull tracking 951 D2 receptor expression 1784 da Vinci Surgical Robot 589 Da-Jie Jiang’s frame 130 Dandy, W. 10, 54, 251, 373, 1080, 1885, 2382, 2401, 2465, 2734, 2741 Dandy’s ventriculography 254 Dastur 163 data interpretation module 3192 data output module 3192 DaVinci robotic surgery 732 deafferentation pain 2109 deafferented spinal cord, physiological observations 2254 deafness 3021 deep brain microrecording, intraoperative 634 deep brain stimulation (DBS) 19, 20, 105, 194, 240, 319, 792, 1247, 1308, 1447, 1840, 2031, 2053, 2059, 2227, 2861, 2903, 2953, 3188 – anatomical target localization and surgical planning 2230 – anatomy 1383 – anterior nucleus 2793 – cluster headache 2034, 2517 – complications 2234 – depression 2953 – disorders of consciousness 2981 – electrode 1403 – electrode, clinical testing of the patient 1435 – electrode, initial placement 1632 – electrode, permanent 1258 – electrode, placer 214 – electrode, practical lesioning 1436 – electrode, stereotactic implantation 1435 – electrode, therapeutic lesion 1427 – essential tremor 1745 – impaired cognitive function 2983 – in the human 2055 – leads 1406 – mechanisms of action 1385 – multifocal 1844 – non-cancer pain 2233 – obsessive compulsive disorder 2897 – PAG/PVG 2227 – pallidal 1843 – Parkinson’s disease 572 – physiological target localization 2231 – present indications and use 2058 – prognostic factors 2234
Subject index
– pulse generator 1329 – selection criteria/surgical targets 2229 – sensory thalamus 2228 – subthalamic 1845 – subthalamic nucleus 1569 – surgery 139 – surgery, patients selection 1530 – system 1404 – test stimulation trial 2232 – thalamic 1843 – vegetative state patients 2982 – voiding function 3003 deep brain structures 376 deep brain target 1242 deformation – automatic methods 388 – manual linear 387 – semiautomatic linear 387 – visual 386 Delgadoc, J.M.R. 184 delivering intensity map 972, 974 delivery, convection-enhanced 1722, 1724 dementia 314, 1521, 1532, 1608 – PET images 315 Demerol 2493 dendritic cell 758 denervation pain 2537 density 1432 depression 1489, 1520, 1532, 2858, 2943 – ablative procedures 2943 – cingulotomy 2887 – deep brain stimulation 2953 – deep brain stimulation, selection of anatomical sites 2927 – medical management 2925 – sleep disturbance 2207 – suicide risk 2207 – surgery 2943 – surgery, indications 2925 – vicious cycle 2207 depressive illness, pathophysiology 2926 depth dose curve 854 depth electrode 2688 deSalles, A. 51, 1206 DesqView 547 detrusor overactivity 3003 detrusor-sphincter dyssynergia 3003 device, mechanical, non-robotic 585 dexamethasone 773 dexmedetomidine 1340, 1343 diagnostic surgical intervention 2687 diaphragm pacing 2991 dichloralphenazone 2490 Dierssen, G. 182 difference imaging 634 diffuse cerebral glioma 3214
diffusion 1722 diffusion tensor 1739 diffusion tensor imaging (DTI) 326, 570, 718, 1250, 2625, 2626 diffusion tensor tractography (DTT) 837 diffusion weighted MRI 2626 Digital Imaging and Communications in Medicine (DICOM) 336, 554 – image 571 – image import 905 digital radiological images 505 digital radiology 505 digital subtraction angiography (DSA) 301, 472, 505, 909 digitally reconstructed radiograph (DRR) 951, 1206 digitization in navigation 545 digitizer – electromagnetic 535, 545 – future 3137 – mechanical 545 dihydroergotamine 2491, 2501 dimenhydrinate 1340 direct epidural motor stimulation evoked response mapping 1274 Disability Adjusted Life Year (DALY) 2857 disabling hyperspasticity, algorithms for treating 1951 disabling spasticity, cerebral palsy, algorithms for treatment 1955 diskynesia 183 disorders of consciousness, deep brain stimulation 2981 distortion 1243 distortion control 1819 divalproex sodium 2496 DNA base analogue 989 DNA injury 1115 DNA plasmid 3083 DNA topoisomerase 754 DNA virus 3062 – oncolytic 3099 dopa decarboxylase inhibitor (DDCI) 1508 dopamine 1393, 1448, 1699, 1732, 3161 – agonist 1510, 1732 – cells 1734 – depletor 1776 – detection 642 – receptor blocker 1776 – transporter 3166 dopaminergic activity 1390 dopaminergic innervation 1698 dopaminergic system 308 dorsal column 1998, 2000, 2001 dorsal column stimulator (DCS) 1402 dorsal column-medial lemniscal system 2150 dorsal horn 1961, 1963, 1990, 1993 – microelectrode recordings in humans 2271
3249
3250
Subject index
dorsal root entry zone (DREZ) 1368, 1949, 1959, 1962, 1989, 2125, 2405 – advantage of radiofrequency lesioning 2265 – anatomical bases 2269 – anatomical lesions at surgery 2278 – anatomy 2251 – area, schematic representation1960 – area, variations of shape 2272 – cancer pain 2258 – complications 2262 – exposure of appropriate segments 2259 – lesion along the spinal cord 2261 – lesion location and technique 2260 – lesion procedures 2276 – long-term results of surgery for spinal/cauda equina injury 2283 – outcomes of pain relief 2264 – pathophysiology of pain 2253 – physiology 2251 – post-amputation (phantom limb) pain 2257 – post-herpetic pain 2258 – post-operative care 2262 – pre-operative appraisal of radicular lesions 2278 – preoperative measures 2259 – prognosis of pain relief 2263 – radiofrequency 2251 – selection and assessment of patients 1963 – surgery 1953 – surgery, brachial plexus avulsion 2277 – surgery, cervical level, operative procedure 1964 – surgery, indications 1970 – surgery, intraoperative neurophysiologic monitoring 1967 – surgery, lumbo-sacral level, operative procedure 1965 – surgery, postoperative care 1968 – surgery, rehabilitation program 1968 – surgery, spasticity 1959 – surgical 2269 – surgical anatomy 1962 – surgical procedure 2279 – surgical technique 1964 – traumatic injury of the cervical plexus 2255 – traumatic injury of the lumbar plexus and cauda equina 2256 – traumatic injury of the spinal cord 2257 dorsal rootlet 1959 dose – calculation model 920 – cochlea 1163 – contribution 920 – delivery 1206 – distribution 915, 922, 969, 1106 – distribution, cochlear apparatus 1165 – distribution, cochlear nerve 1164 – fractionation 856
– matrice 914 – normalization 920 – per fraction 860, 861 – rate 859 – rate (output) test 921 – response 862 – selection 1192 – tolerance 1114 dose-volume histogram analysis 864 dosimetry 867, 918 double donut 821 – configuration 821 double-event double-strand break 855 drainage catheter 774 DREZotomy, operative procedure at the cervical level 2273 DREZotomy, operative procedure at the lumbo-sacral level 2274 DREZotomy, principles 1962, 2271 DREZotomy, surgical techniques 2271 droperidol 1339 drug – addiction 165 – distribution imaging 1251 – distribution surrogate marker 1251 – intrathecal 1973 – non-steroidal anti-inflammatory 2070 – pump 1977 – pump implanted 1925 drug delivery – computer-assisted brain analysis 1740 – convection-enhanced, role of imaging 1250 – intracranial, MRI guided 1249 – intracranial, role of imaging 1250 – intraparenchymal, Parkinson’s disease 1731 – system, problem 1977 drug-producing cell, injection 1249 duodopa 1625 dura mater, opening 1915 dural arteriovenous fistula (DAVF) 1012 dural malformation 1068 dysarthria 2112 dysembryoplastic neuroepithelial tumor (DNET) 2633, 2641 dyskinesia 1531 – graft-induced 1699 – levodopa-induced 1572 dyskinetic movements, clinical examination 1821 dysplasia – cortical 2718 – focal cortical, surgical outcome 2640 dystonia 106, 1451, 1767 – abnormal sensory processing 1792 – abnormal striatal activity 1783 – abnormal thalamic activity 1789 – altered cerebellar activity 1791 – associated with Parkinson’s disease 1851
Subject index
– – – – – – – – – – – – – –
brachial 1453 causes 1452 cerebellum, imaging abnormalities 1791 cerebellum, structural abnormalities 1790 classification 1452, 1768 clinical features 1767 cortex, hypermetabolism 1792 diagnosis 1767 doparesponsive (DRD) 1774 DYT1 1782 genetic forms 1772 GPi activity 1787 impaired intracortical inhibition 1792 integration of cerebellar and pallidal circuits 1791 – medical management 1767 – monogenic forms 1773 – multifocal 1769 – musculorum deformans 1802 – neuroimaging 1775 – neuron model 1793 – neurophysiology 1782 – Oppenheim 1773 – oromandibular (OMD) 1771 – paroxysmal 1454 – pathology 1775 – pathophysiology 1779 – pharmacologic treatment 1777 – physiology 1775 – plus syndromes 1453, 1836, 1949 – prevalence 1775 – thalamic lesions 1789 – treatment 1454, 1776 dystonia, cervical 1769 – bilateral complications 1874 – bilateral pallidal stimulation 1878 – central procedures 1871 – classification of symptoms 1889 – clinical features 1862, 1871 – diagnosis 1857 – electromyography pattern 1890 – epidemiology 1871 – features 1888 – involved muscles 1860, 1890 – medical management 1857, 1864 – microvascular decompression 1905 – natural history 1871 – outcomes 1874 – pallidal stimulation 1876 – pallidal stimulation, complications 1880 – pallidal stimulation, duration of effect 1878 – pallidal stimulation, predictors of success 1878 – pallidal stimulation, stimulation parameters 1877 – pallidal stimulation, time course 1878 – peripheral procedures 1885
– peripheral surgical procedures 1895, 1905 – predictors of outcome 1875 – pre-operative screening 1873 – psychiatric considerations 1493 – relevant complications 1874 – secondary 1863 – secondary, outcomes 1875 – secondary, red flags 1864 – sensory tricks 1861 – spinal cord stimulation 1906 – surgical indications 1873 – surgical procedures 1874 – surgical treatment, history 1885 – surgical treatment, indication 1892 – thalamic stimulation 1875 – thalamic targets 1874 – thalamotomy 1874 – time course 1875 – types 1888 – unilateral complications 1874 dystonia, cranial 1453 – cervical segmental 1872 – cervical, pallidal stimulation, outcomes 1880 – cervical, outcomes 1875 dystonia, focal 1769, 1783 dystonia, generalized 1769 dystonia, heredodegenerative 1774 dystonia, idiopathic cervical 1860 – clinical features 1859 – diagnosis 1859 – natural history 1863 dystonia, primary 1452 – focal 1453 – central procedures 1801 – clinical features 1801 – recommendations for surgical treatment 1805 – thalamocortical network model 1795 dystonia, secondary 1453, 1774, 1872 – causes 1775, 1837 – classification 1835 – clinical signs 1839 – deep brain stimulation 1840 – deep brain stimulation, multifocal 1844 – deep brain stimulation, pallidal 1843 – deep brain stimulation, subthalamic 1845 – deep brain stimulation, thalamic 1843 – differential diagnosis 1835 – pallidotomy 1840, 1842 – surgical procedures 1839 – thalamotomy 1840, 1841 – treatment, functional stereotactic procedures 1835 dystonia, segmental 1769 dystonia, tardive 1838, 1845 dystonic storm 1848
3251
3252
Subject index
dystono-dyskinetic syndrome (DDS) 1801 – ablative stereotactic surgery 1810 – classification 1823 – deep brain stimulation 1811 – Dopa-responsive 1802 – electrophysiology 1810 – globus pallidus internus stimulation 1814 – globus pallidus internus, deep brain stimulation 1812 – heredodegenerative 1803 – heredodegenerative, surgical treatment 1805 – intraoperative clinical observation 1810 – neuromodulation 1811 – pallidotomy 1811 – primary 1802 – primary, surgical treatment 1805 – secondary 1803 – secondary, surgical treatment 1805 – stereotactic MRI-based target localization 1810 – target localization methods 1808 – thalamotomy 1810 – ventriculography 1809 DYT1 1802 – dystonia 1782 – non- 1802 DZY-A 131
E ear level processor and internal device 3024 early responding tissue 858 Easy-Guide Neurosystem 703 edema 1011 education 2933 efaproxaril 987, 988 effect, analgesic 2031 effector 557, 562 Eldepryl 1514 Electreat 17 electric field 1362 – line pattern 1363 electric ray 1402 electrical conductivity 1433 electrical impedance tomography (EIT) 633 – applications 634 electrical neurostimulation, basic physiology 1383, 1384 electrical settings 1820 electrical stimulation 1284, 2075 – centromedian nucleus (ESCM) 2777 – mapping 2653 – nervous system 17 – posterior subthalamic target 1668 electrical torpedo fish 1402 electricity 1401 electrocautery, bipolar 2719 electroconvulsive therapy (ECT) 2953
electrocortical stimulation (ECS) 287 electrocorticogram 114, 2651 electrocorticography 143, 1342, 2651, 2719 – epilepsy surgery 144 – equipment 331 electrode 477, 3047 – array 2691 – damage 1411 – design 1258 – geometry 1365 – grid 2245 – heating 1817 – implantation 1613 – position, control 1820 – position, radioscopic monitoring 1817 – positioning, CT stereotactic control 493 – system 2152 – system, calibration 2152 – depth 2639 – implantation, subthalamic nucleus, surgical procedure 1610 – invasive 2638 – quadripolar 2796 – subdural 2638 – tunneling in the neck 1591 electroencephalogram background activity 2840 electroencephalography (EEG) 2661, 2677 – epilepsy 2575 – epileptiform pattern, periodic 2577 – epileptiform variant, benign 2577 – ictal 2579 – interictal, sensitivity 2576 – interictal, specifity 2576 – invasive 2634 – recording 103 – scalp 2634 – scalp electrode 638 – -type electrode 633 – video 2579 electrogoniometer 702 electromagnetic field 535 electromagnetic localization, engineering aspects 535 electromagnetic stereotactic system, advantages 540 electromagnetic stereotaxis 536 electrometer/ionization chamber 923 electromyography (EMG) 1889 electron microscopy 668 electronic clinical brain atlas (ECBA) 429 electrophrenic ventilation 2991 electrophysiology, developments Elekta 137, 257, 429, 599, 623, 648, 702, 905, 1326, 1584, 3145 elephant 577 Elephant God 156 El-Naggar-Nashold Nuclear Caudalis Electrode tip geometry 1370
Subject index
emboli detection, pulmonary 634 embolization 1064 embouchure dystonia 1772 EMI scanner 87, 3131 Emilio, Paolo 193 encapsulated cells 1249, 1625 encephalometer of Zernov 98 endocrine remission rates 1044 endoport 625 endoscope 726 endoscopic surgery 732 endoscopic third ventriculostomy 808, 811 entacapone 1510, 1512, 1513 enzyme replacement 1725 ependymoma 1021, 1058, 3216 epidermal growth factor (EGF) 752, 3166 epidermal growth factor receptor (EGFR) 745 – pathway signaling 995 – human 752 – inhibitor 753 – tyrosine kinase inhibitor 3220 epidural quadripolar electrode paddle, surgical planning for implantation 1682 epilepsia partialis continua (EPC) 2718 epilepsy 10, 114, 317, 1029, 1073, 3123 – anterior nucleus deep brain stimulation 2793 – bi-temporal, chronic intracranial recording 2689 – cavernous malformation 2768 – centromedian thalamic stimulation 2777 – centromedian thalamic stimulation, surgical techniques 2779 – classification 2561, 2569 – computed tomography 2617 – diagnosis 2662 – electroencephalography 2575 – experimental imaging techniques 2625 – functional magnetic resonance imaging 2623 – Gamma knife radiosurgery plan 1198 – gelastic 2767 – generalized 2573 – imaging, evaluation 2617 – imaging, practical approach 2626 – intraoperative monitoring 2651 – juvenile myoclonic 2573 – left frontal lobe 2572 – left fronto-temporal 2573 – magnetic resonance imaging 2618 – magnetic source imaging 2622 – magnetoencephalography 2661 – medical management 2669 – positron emission tomography 2620 – presurgical Evaluation 2634 – protocol 175 – radiosurgery 1197, 2761 – radiosurgery, long term complications 2769 – right mesial temporal lobe 2572
– single photon emission computed tomography 2619 – surgery 21, 103, 113, 141, 159, 175, 1337, 2544 – surgery, anesthetic agents 1339 – surgery, complications 1340 – surgery, general anesthesia 1341 – surgery, goal 2634 – surgery, image guided 2633 – surgery, image guided, indications 2633 – surgery, intraoperative anesthetic considerations 1339 – surgery, intraoperative recordings and activation 1342 – surgery, palliative, indications 2556 – surgery, preoperative assessment 1338 – surgery, resective, types 2555 – surgery, techniques 1339 – synchrony of cell discharge 2715 – 5-tier classification 2570 epileptic activity – genesis 2784 – propagation 2784 epileptic focus concept 2545–2547 epileptiform activity, definition 2575 epileptogenic afterdischarges 2841 epileptogenic focus, preoperative localization 1337 epileptogenic zone 2545, 2636, 2685 – electrocorticographic monitoring 2651 – slassification 2571 epinephrine 1339 equianalgesic opioid conversion 2180 equivalent current dipole (ECD) model 2661 ergotamine 2490, 2500 – inhalation 2498 erlotinib 753, 3220 Esgic 2493 essential tremor 106, 1028 – deep brain stimulation 1745 – deep brain stimulation, clinical outcomes 1749 – gamma knife radiosurgery 1195 – management 1743 – pharmacologic treatment 1743 – surgical technique 1747 – surgical treatment 1745 – thalamotomy 1749 ESSFN 25, 37, 38, 41–44, 187 etanidazole 987 Ethics 106, 2863–2865 ETHZ Paracare 3055 etomidate speech and memory test (eSAM) 2685 Etopalin 12 euphoria 1490 European Society for Stereotactic and Functional Neurosurgery (ESSFN) 37, 42, 135, 187 – congresses 43
3253
3254
Subject index
evaluating the chronic pain patient – interview 2213 – questions 2214–2217 evoked potential 1260, 1287 – equipment for stimulation/generation 1258 – functional neurosurgery 1255 – functional neurosurgical procedure 1261 – median nerve 1274 – recording, bipolar 1265 – recording, referential 1265 evoked response – wrist kinesthetic excitatory 1271 – wrist kinesthetic inhibitory 1272 Evolution 1 594 Ewing’s sarcoma 1120 excessive daytime sleepiness (EDS) 1519 expiration phases 2992 extracellular electricity 1418 extracranial dose 926 eye-tracking system 545
F facet denervation 2291 – anatomy 2291 – complications 2294 – electrodes 2292 – indications 2291 – results 2293 – technique 2291 facial nerve preservation 1013 facial pain 2466 – glycerol rhizolysis 2451 Faraday cage 6 Fasano, Victor Aldo 193 Fasano-Sguazzi frame 194 fasciculus cuneatus 2099 Fas-ligand 3166 fast-scan cyclic voltammetry (FSCV) 641 FDG-PET 315 feedback during surgery 704 fentanyl 1334, 1339, 1343 ferromagnetic instrument 715 fetal cells, mesencephalic 1625 fetal ventral mesencephalon 3163 fiber tracking 257, 570 fibres en passage 8 fibroblast growth factor (FGF) 3166 fibromyalgia 2026 fiducial implantation 1206 fiducial marker 909, 1097 fiducial skin marker, position 706 fiducial system, distortion 1243 fiducial-free system, advantages 953 fiduciary system, axial representation 255 field potential 3189 field size, limitations 1099
Figueiredo, Djacir 199 figurine map 361 Filho, Branda˜o 198 finger tip stimulation 1794 Fioricet 2493 Fiorinal 2493 first branch neuralgia 2103 First National Stereotactic and Functional Neurosurgery Conference (Hefei City, Anhui Province) 134 Fischer 382 fistula, carotid-cavernous 1070 FLAIR image 712 Fleishmann, Elizabeth 22 Flock of Birds system 535, 539 fluid recuscitation 2754 fluorescence 564 Fluorodopa-PET 258 FluoroNav 558 fluoroscopic guidance 2796 fluoroscopic image 713 fluoroscopy 822 fluropyrimidine 994 5-fluoruracil 994 focal motor seizure 2778 focal seizure onset 2545 Foltz, E. 16, 2887 foot drop 3051 foramen 2436 foramen ovale 2460 forceps system 477 Forel’s field 13, 1257 fornix 2695 Fo¨rster, O. 53 fossa posterior 656 fossa surgery 826 – stereotactic neuronavigation 827 fractionated dose-effect curve 856 fractionated stereotactic treatment technique 1155 fractionation 1116, 1117 – conventional 1157 – factor 857 frame – adaptor 1232 – application 466 – Brent Robert Wells (BRW) 98 – cap 912 – cap fitting check 1232 – Cosman Robert Wells (CRW) 98 – goniometric 98 – localizer 576 – placement 905, 2796 – setup 466 frame-based image-guided stereotactic biopsy 836 frame-based stereotactic procedures, complications 479 frame-based surgery 136
Subject index
frame-based systems 816 – disadvantages 816 frameless IGS, basic components 818 frameless image-guidance systems 836 frameless localization 551 frameless stereotaxy 816 France 97 Franc¸ois Cohadon 107 free hand trigeminal tractotomy under tomographic control 2112 Freehand System 3056 freehand trigeminal tractotomy, results and complications 2113 Freeman, Walther 2871, 2873, 2874 frontal lobe resection 2643 frontal lobectomy 2643 Fulton, John 2870 functional anatomy, trigeminal system – brain stem complex 2388 – central connections 2390 – mesencephalic nucleus 2387 – motor and accessory roots 2384 – somatotopic organization 2382 functional brain mapping 318 functional cortex, electrical stimulation mapping 2653 functional cortical localization 292 functional deficit zone 2547 functional brain disorder 1071 – radiosurgery 1027 – stereotactic radiosurgery 983 functional electrical stimulation 3047, 3185 functional imaging 1311 – preoperative 325 – radiosurgery 1200 functional magnetic resonance imaging (fMRI) 263, 287, 311, 559, 717, 1088, 1312, 2598, 2623, 2685 – acquisition 287 – activations 291 – advantages 294 – application to cortical resections 289 – disadvantages 295 – ECS 292 – illustrative cases 289 – interpretation 291 – physiological basis 287 – sensitivity 291 – sensorimotor foot activation 289 – sensorimotor hand activation 289 – signal changes 290 – technological requirements 288 functional neuroimaging 610 functional neuronavigation 837 functional neurosurgery 113, 605, 624, 792, 1191 – atlas-assisted 421 – ethical issues 3230 – evoked potential 1255
– evoked potential, history 1256 – history 1192 – image guided 1239 – impedance recording 1325 – microelectrode recording 1283 – pain 146 – stimulation physiology 1383 – stimulation technology 1401 functional radiosurgery (FRS) – imaging 1193 – parenchymal, dose-selection 1192 functional stereotactic radiosurgery, MRI 611 functional stereotaxis, clinical applications 272 functional surgery – history in Brazil 197 – history in Canada 113 functional targets on MR-images, defining 1239 functional transcranial doppler sonography 2597 fusion 335 – technique 627 FVM transplant 1695
G GABAergic system 310 GABAergic transmission 2696 gabapentin 1744, 2673 gadolinium based magnetic contrast agent (GBCMA) 284 gadolinium-based contrast-induced Nephrogenic Systemic Fibrosis (NSF) 284 gadolinium enhanced image 1155, 1166 gait analysis 1944 gait disorder – hypokinetic-rigid 1656 – parkinsonian-like 1656 gait disturbance 1609 gait syndrome, major neurologic 1655 gamma amino butyric acid (GABA) 310, 1388, 1973, 2696, 3123 Gamma Knife 68, 187, 839, 8861113, 1172 – basic treatment process 905 – central body 900 – clearance checks 911 – clinical experience 1007 – collimation system 900 – console 904 – CT 909 – docking the patient 911 – dosimetry 918 – frame placement 905 – imaging 907 – irradiation 1561 – isocenter profile 924 – models 1225 – models, first 978 – MR fiducial box 908
3255
3256
Subject index
– MT 907 – output test 921 – pallidotomy 1029 – plan evaluation 916 – planning system 905 – primary components 900 – quality assurance program 923, 1230 – radiation body 899, 900 – radiosurgery treatment planning 2766 – stereotactic frame attachment 902 – stereotactic localization 899 – technical aspects 897 – thalamotomy 138, 1028, 1194 – treatment execution 913 – treatment parameters 910 – treatment planning 909, 910, 913, 914 – treatment planning, tricks of the trade 916 – treatment process 906 – unit 176 – XA 909 Gamma Knife Perfexion 480, 898, 920, 970, 974, 1185 – treatment volume 925 Gamma Knife radiosurgery – benign brain neoplasms, Indications 1015 – brain disorders treated world-wide 1226 – dose plan 1014 – functional brain disorders 1028 – glial neoplasms 1021 – indications 1008, 1054 – meningioma 1049 – metastatic neoplasms 1022 – miscellaneous neoplasms 1026 – plan 864 – procedure 1229 – technical issues 1225 – procedures 472 Gamma Knife surgery 911 – adenoma, nonsecretory, goal 1042 – adenoma, secretory 1043 – arteriovenous malformation, decision making 1060 – arteriovenous malformation, dose-response curve 1064 – brain metastases 1057 – clinical outcomes 1042 – imaging 1042 – indications 1042 – pain, facial 2475 – pituitary adenoma 1044 – pituitary adenoma, goals 1042 – repeat, arteriovenous malformation 1066 – repeat, arteriovenous malformation parameters 1066 – repeat, arteriovenous malformation, large 1067 – repeat, clinical outcome 1066
– repeat, imaging 1066 – repeat, treatment parameters 1066 – treatment protocol 1041 – trigeminal neuralgia 2475 – undue effects 1074 gamma radiation emitter 978 gamma rays 853 GammaPlan 905, 913 – functionality 905 gammathalamotomy 1039, 1071 ganglioglioma 2641 ganglionectomy 15 Ganser 388 gap junction 3091 gasserian ganglion 15 gastric emptying time 634 gastroparesis 2490 gate control theory of pain 2162 gefitinib 753, 3220 gene expression 761 gene technique-based immunotherapy 759 gene therapy 739, 1625, 3061 – Alzheimer’s disease 3075 – basic science 3061 – brain tumors 3068, 3083 – neurogenetic disorders 3076 – neurological disorders 3061 – Parkinson’s disease 3071 – regulatory approval 3063 gene transfer 1719 – infusion methods 1722 – Parkinson’s disease 1719 – Parkinson’s disease, clinical trials 1725 – Parkinson’s disease, risks 1727 – stereotactic surgical approaches 1721 – surgical planning 1722, 1723 – tumor-selective 3089 generalized tonic-clonic convulsion (GTC) 2778 geometric contours 398 geometric convergence 1102 Germany 53 giant colloid cyst 695 Gildenberg, P. L. 3, 35, 45, 583, 725, 2159, 2197, 2533 Gildenberg technology scale, phases 587 Gilles de Tourette’s syndrome 2860 Gillingham, Francis John 80, 81 Gleevec 754, 3220 Gliadel 750 glial derived neurotrophic factor (GDNF) 1625, 1697, 1724, 3072 – infusion 1737, 1738 glial tumor 1020 – high-grade 687 glioblastoma 666, 667, 969 – discrete 738 – multiforme 744, 3214
Subject index
– primary 3214 – secondary 3214 – stereotactic radiosurgery 982 glioma 601, 710 – progression 752 – surgery 325 – brain stem, MR PET-guided targeting 781 – diffuse 736 – low-grade 711, 790, 1021, 3215 – low-grade, proton therapy 960 – malignant 940, 1020 – malignant, adenoviral gene therapy studies 3088 – malignant, clinical gene therapy trials 3093 – malignant, gene therapy studies 3083, 3084 – malignant, HSV–1 vector-based gene therapy studies 3091 – malignant, oncolytic virus trials 3102 – malignant, preclinical oncolytic virus therapy 3094 – malignant, retroviral gene therapy studies 3087 – mixed 667 glioma cell – retroviral gene delivery 3088 – interaction with immune system 756 Glioma Outcomes Project 720 gliosis 2743 gliotic substrate 2641 globus pallidus 11, 1267, 1303, 1304, 1383, 1606, 1667, 1785 – external (GPe) 1582 – firing patterns 1305 – internus (GPi) 269, 1488, 1569, 1577, 1582 – internus (GPi) stimulation, dystono-dyskinetic syndrome 1814 – internus (GPi), activity of single cells 1500 – internus (GPi), deep brain stimulation 1569 – internus (GPi), deep brain stimulation, complications 1595 – internus (GPi), deep brain stimulation, indications 1581 – internus (GPi), deep brain stimulation, programming parameters 1592 – internus (GPi), deep brain stimulation, surgical technique 1584 – internus (GPi), patient selection 1581 – internus (GPi), procedure 1266 – internus (GPi), role in Parkinson’s disease 1579 – internus (GPi), sensorimotor territory 1584 – internus (GPi), sensorimotor, characteristics of neurons 1590 glomus tumor 1024 glucose 308 glutamate 1389 glutamatergic system 310 glutamic acid decarboxylase gene 3072 glycerol injection 2438 – Meckels’ cave 2527
glycerol rhizolysis, in other types of facial pain 2451 glycerol therapy in other facial pain syndromes 2452 Gorlin syndrome 3212 grafting methods 1625 granisetron 1340 granulocyte/macrophage-colony stimulating factor (GMCSF) 758 graphical overlay 340 gray matter 1299 – impedance 1326 Great Britain 77 greater occipital nerve 2507 – neurolysis 2509 Grenoble setup 497, 499, 505 grid 2689 grid system, proportional 500, 501 gross motor function measure 1954 gross tumor volume (GTV) 1206 growth – elivery 1724 – infusion 1625 growth signal 3202 guide tube 1240, 1246, 1247 Guidetti, Beniamino 193 Guiot, G. 100 Guiot’s geometric scheme 1633 Guiot-Gillingham frame 131 Guiot-Gillingham stereotactic apparatus 82 Gusma˜o, Sebastia˜o 197 gustatory aura 2563 gustatory effect 1308
H H spectroscopy 2625 hair transplantation, robotic 590 Hallervorden-Spatz disease 139, 1837 hallucination 1522 hamartoma 1198 – hypothalamic 2767 Hammilton depression rating scale (HDRS) 532 hamstring neurotom, skin incision 1939 Handmaster 3055 Hanyang University 176 hardware related complications 1595 Hassler’s parcellation 396 HC stereotactic apparatus 209 Hc/Hb complications 181 head techniques, high-resolution 476 headache 1061 – chronic daily 2486 – episodic tension-type 2486 – posttraumatic 2487 – primary, classification 2484 – tension-type 2484 – treatment 2483 headframe application 1632
3257
3258
Subject index
head-frame placement 1584 heads-up display device 682 hearing 3021 – -aid 3034 – loss 1160 – outcome 1160 – preservation 1013, 1163 – ret formula 1162, 1163 heating 283 heavy particle radiotherapy 887 – evidence-based studies 888 helical tomotherapy 877, 878 hemagglutinating virus of Japan (HVJ) 3092 hemangioblastoma 1025, 1053 hemangioma 1025 hemangiopericytoma 1059 hematoma, aspiration 521 hemibrainstem 356 hemichorea-ballism 1570 – lesion-induced (HCB) 1572 hemicrania, chronic paroxysmal 2484 hemidecortication 2749, 2750 hemidystonia 1769, 1848 hemilaminectomy 1898 hemimegalencephaly 2747 hemiplegia, infantile 2748 hemispherectomy 2741 – anatomical 2749, 2750 – disconnective 2753 – early postoperative complications 2755 – functional 2746, 2750, 2751, 2753 – late postoperative complications 2756 – modified 2749, 2750 – mortality 2756 – pathophysiology 2745 – peri-insular 2752 – perioperative complications 2754 – preoperative evaluation 2745 – rationale 2745 – surgical techniques 2749 – vertical 2752 hemorrhage 654, 693, 1061, 1064, 2742, 2755 – arteriovenous malformation 1009 – intracerebral 736 – intracranial 1335, 1595 – secondary 1053 – subarachnoid 3122 hemosiderin deposit 2743 hemosiderosis, superficial cerebral 2743 Hepburn, Howard H. 120 heredodegenerative disorders 1836 herpes simplex virus (HSV) 1720, 3062, 3083 – gene therapy 760 – oncolytic 3095 – oncolytic, improving potency 3096 – oncolytic, improving selectivity 3096
– vector 3090 – thymidine kinase (HSV-Tk) 739, 759, 760, 3087 – thymidine kinase (HSV-Tk) gene 3083 Hess, Walter Rudolf 73 heterotopia, laminar 2619 high frequency single pulse microstimulation 1392 high frequency stimulation (HFS) 105, 106, 1385, 1623 hippocampal electrodes position 2845 hippocampal stimulation – chronic 2842, 2882 – chronic, double blind maneuver 2846 – chronic, surgical procedure 2844 – subacute 2843 hippocampal-cortical prosthesis 3194 hippocampus 2695, 2696, 2699 – stimulation 2839 – stimulation, rationale 2839 histology 376, 382 histopathology 635 Hitchcock, Ted 83 Hitchock’s stereotactic apparatus 83 Hitzig, E. 199, 1284, 2654, 2677, 3186 Holmes’ tremor 1450 Homer’s syndrome 2112 homogeneity index 867 Horner’s syndrome 2809 Horsley and Clark stereotactic apparatus 252 Horsley, Sir Victor 3, 17, 77, 78, 89, 141 Horsley-Clarke – animal stereotaxic apparatus 35 – apparatus 4, 7, 9, 77 hospitalization 1558 hot flash 1378 hot spot 1038 Hounsfield, Sir Godfrey 87 H-reflex 1257 Hua Tuo (Hua Lun) 125, 126 Huang Ti Nei Jing 125 human brain tissue, impedance 1430 human embryonic stem cell (hESC) 3171 human immunodeficiency virus (HIV) 657, 3062 human retinal pigment epithelial (hRPE) cell 3166 Hun Jae Lee 174 Huntington’s chorea 10, 47 – pallidotomy 11 Huntington’s disease 1456, 1804, 1851 hydrocephalus 811, 2742 5-hydroxytryptamine (5-HT) 309, 2488 hydroxyurea 754 hyperbaric oxygen 988 hyperexcitability, central neuronal 2488 hyperspasticity 1952 hypertension 797, 1335 hypofractionation 1157 hypopituitarism, radiosurgical induced 1185
Subject index
hypothalamus – posteromedial, high frequency stimulation 2517 – stimulation 2528 hypoxia response element (HRE) 997 hypoxia-inducible factor (HIF) 753, 997
I IASP 2313, 2317 ignition hypothesis 2397 illumination 564 image acquisition 648 image co-registration 905 image fusion 259, 335, 905 – errors 340 image guidance – accuracy 443 – conventional 726 – techniques 1207 – virtual reality 726 image guided biopsy, pathology techniques 663 image guided craniotomy 679, 699 – history 699 – methodology 699 – methods 680 image guided functional neurosurgery 1239 – clinical methods 1239 – post-operative complications 1248 – rationale 1239 image guided neurosurgery 299, 307, 533 – electromagnetic localization, engineering aspects 535 – functional MRI 287 image guided stereotactic trajectory, frameless 801 Image guided surgery 22 – computer tomography (CT) 619 – magnetic resonance imaging (MRI) 599 image quality 1240 image reconstruction 335 – verification techniques 341 image registration 554, 560, 905, 913 – errors 340 – surface-based 338 image resolution 1240 image volume 340 image-guidance, types 651 image-guided biopsy 645 – complications 653 – indications 646 – instruments 648 – locations 655 – pathology 654 – technique 480, 648 image-guided frame-based stereotactic procedures 479 image-guided neuroendoscopy – applications 810 – surgical technique 807 – technical aspects 807
image-guided radiological image, typical 335 image-guided surgery 259, 524 imaging 1586, 1632 – brain function 559 – cell-based therapy 1250 – integrated 3149 – intraoperative 557, 564, 711 – intraoperative, future 3138 – Leksell frame 2796 – modalities 251 – perspectives 3133 – preoperative 554, 559 – selection 707 – structural imaging 1310 – studies 2042 – studies, remote postoperative 518 – tumor type 705 – tumor type, recommendations 707 imatanib mesylate 3220 imatinib mesylate 754 immobilization 1206 – patient 270 immuno-histochemical technique 2057 immunoperoxidase staining 668, 674 immunophilin 1720 immunosuppression 3175 immunotherapeutic approach, passive 757 immunotherapy 755 impaired cognitive function, deep brain stimulation 2983 impaired hearing, auditory prosthesis 3021 impaired motor function, functional electrical stimulation 3047 – recording methods 3050 impaired vision, visual prosthesis 3009 impedance 631 – electrical 1325 – monitoring 632, 1286 – monitoring, advantages 1328 – monitoring, functional procedures 1327 – monitoring, limitations 1329 – monopolar 1325 – recording, functional neurosurgery 1325 – spectroscopy, electrochemical 638 – tomography, electrical 633 implant candidacy, assessment 3024 implantable pulse generator (IPG) 1405 implanted stimulator, functional neurosurgery 1349 impulse control disorder (ICD) 1491, 1521 inadequate assessment 2062 incision, subcortical, method 684 India 155 Indian mythology 155 indomethacin 2501 inducible promoter 1720 inductance 1325 infection 1595, 2742
3259
3260
Subject index
infectious complications after glycerol rhizolysis 2447 inferior thalamic peduncle 2931, 2956 – location of DBS electrodes 2957 inflammation 2006 – granulomatous 791 – nonspecific 676 infrared light-emitting diode (IRLED) 875 infusion systems, future 3155 Inherited cancer syndrome 3206 inhibitory neurotransmission, enhancement 1726 injection 1249 – cannula, design 1723 injury repair paradigm 1040 InSet Prometra programable implanted pump 3157 insomnia 1519 inspection window 1100, 1101 Institutional Biosafety Committee (IBC) 3064 instrument calibration device 809 insulin receptor 3093 insulin-like growth factor (IGF) 3166 integrating imaging 543 integrin 746, 999 intensity map 972 intensity modulated radiation therapy (IMRT) 1209 – technical and clinical aspects 965 – treatment delivery 968 – treatment plan 966 intensity modulated radiosurgery (IMRS) 965 intensity modulating technology 970 intensity profile 1102 intensity, single isocenter 1103 intercommissural line 352 interference, magnetic 538 interference warning indicator 539 interferon 740 interictal epileptiform discharges (IED) 2802 interleukin 757, 3083 internal capsule 326, 1267 – anterior, impedance monitoring 1328 International Society for Research in Stereoencephalotomy 9, 35 International Workshop on Functional Neurosurgery for Movement Disorders and Mental Illness & Commemoration 89 interneuron 2005 interpersonal psychotherapy (IPT) 2925 interruption of pain pathways, within spinal cord 2167 intra-axial 735 intracarotid sodium amobarbital test (IAT) 2685, 2686 intracerebral hemorrhage (ICH) – CT imaging studies 804 – endoscopic technique for removal 803 – etiology 797 – evacuation, surgical 800 – evacuation, techniques 800 – image guided management 797
– medical management 798 – medical management, goals and study benefits 799 – minimally invasive/stereotactic evacuation 802 – spontaneous 797 – spontaneous, incidence 797 – spontaneous, management 797 – surgical technique for removal 801 – timing of surgery 801 intracortical stimulation 3014 intracranial cyst 810 intracranial lesion, treatment 1124 intracranial pressure (ICP) 797 – identification and location 797 intradural benign tumors 1216 intraoperative examination 2056 intraoperative magnetic resonance imaging (iMRI) 821 – biplanar magnet design 821 – cylindrical superconducting magnet 821 – double donut configuration 821 intraspinal opioids in cancer pain, efficacy 2183 intrathecal and intracerebroventricular morphine administration, indications 2187 intra-thecal baclofen therapy (ITB) 1951 intrathecal drug delivery system 2173 – access-port system 2174 – externalized systems 2173 – surgical complications 2176 – surgical technique 2175 – totally implanted system 2174 intrathecal morphine 2183 intrathecal opiates – cancer pain 2171 – cancer pain, patients selection 2171 intraventricular catheter (IVC) 799 intraventricular lesion 810 intrusive thoughts 2898 invasion 3205 invasive stereotactic fixation 261 inverse planning 868 involuntary movements 181 iodinated contrast media (ICM) 281 – adverse reactions 281 – safety studies 281 – types of radiographic 281 iodobenzovesamicol (IBVM) 1793 iododeoxyuridine (IUdR) 989 ionizing radiation, types 853 iPlan 415 – Flow 1250 – Net 570 – stereotaxy 571, 572 ipsilateral ataxia 2112 Iressa 3220
Subject index
irinotecan 754, 993 Iris Collimator 973 irradiation – effects 1113 – late CNS effects 1114 irritative zone 2545, 2550, 2685 ischemic changes 1532 ischemic stroke 3122 ISG viewing wand system 702 isocenter 917, 1092, 1113 – placing 1106 – profile 924 – testing 1093 isocentrical beam 952 isodose 1104 – distribution 1105 – line 611, 864 – rings 1560 – volume 1116 isoeffective dose 858, 861 isometheptene 2490 Isoptin 2499 isotherm surfaces 1363 Italy 193 I-Th cell and gene therapy 2188 IVS Voxim software 506
J Japan 59 Jasper and Hunter stereotactic instrument 116 Jasper, Herbert 114, 115 Jernberg, Bengt 69 Jian-Ping Xu 131
K Kanpolat Cordotomy Electrode 1368, 1369 Kaolin 1987 Kelly, Desmond 2880 Kelly’s Compass system 545 Kelly, P. 535, 679 KEM thalamotomy 163 keratitis 2105 Ki Sup Lee 172 kinesthetic area 1291 knee-ankle-foot orthosis 3054 Knight, Geoffrey Cureton 84, 2879 Knight’s stereotactic subcaudate tractotomy 84 Korea 171 Korean Sterotactic and Functional Neurosurgery Society 176 Krause, Fedor 53 Krauss, J. K. 3, 35, 53, 487, 1835 Krayenbu¨hl, Hugo 73 Kuka industrial robot 586 Kullback-Leibler distance 450 Kumar, Krishna (Kris) 120
L lack of familiarity with options 2062 lack of training 2062 Laitinen stereoguide 513 – apparatus 511 – apparatus, radiological applications 514 – apparatus, surgical applications 514 Laitinen, Lauri 70, 71 lamellar organization 1554 laminectomy 262 – syndrome 2056 laminotomy 1406 lamotrigine 2673 landmarks 424 language 1338, 2591, 2693 – assessment 2588 – cortex, identification 2654 – function mapping 327 – hemisphere dominance 2589 – mapping 318 – stimuli, functional magnetic resonance imaging 2985 Laplace’s equation 1428 Larson, Borje 479 laser beam 480 late infantile neuronal ceroid lipofuscinosis (LINCL) 3077 lateral nucleus, posterior part 2081, 2082 – surgical target 2086 late responding tissue 858 laterocollis 1860 Latin American Society for Stereotactic and Functional Neurosurgery (SLANFE) 223 Laundau-Kleffner syndrome (LKS) 2717 Law of Bell and Magendie 2387 l-dopa 14, 20 lead 1406 – breakage 2809 – computer modeling 1428 learning 3124 Leber’s disease 1804 LED tracking array 553 leg movements 3051 leisonectomy 2702 Leksell and Riechert-Mundinger systems 10 Leksell arc-centered design 470 Leksell coordinate system 899 Leksell frame 131, 575 – magnetic compatible 1815 Leksell Gamma Knife 471, 479 – 4C 1227 Leksell Gamma Knife Perfexion 482, 886, 1008, 1226, 1228 – collimator system 1229 – radiation unit 1229 – treatment plan 888
3261
3262
Subject index
– efficacy 887 – performance features 886 – quality 887 – treatment plan 1230 – technical specifications 1231 – model U 479 Leksell GammaPlan PFX (LGP PFX) 1228 Leksell model B Gamma Knife 481 Leksell model G expanded stereotactic arc 474 Leksell stereotactic apparatus 469 – accessories 477 – versatility 477 Leksell stereotactic coordinate frame, original 470 Leksell stereotactic frame 117, 634, 1230, 1241, 1585 Leksell stereotactic G-frame 899, 906, 907 Leksell stereotactic instrument 66, 116 – model G 471 Leksell SurgiPlan 480, 483 Leksell system 272 Leksell, L. 9, 65, 67, 469, 479, 1038, 2879 Leksell’s disciples 69 Lennox-Gastaut syndrome 2787 lentivirus 1720, 3062 Lesch-Nyhan disease (LND) 1804, 1851 lesion – chronic pain 2052 – cranial, histopathologic diagnosis 645 – cystic 657 – deafferenting 1297 – electrode 1362 – generator 477, 1326 – human brain, practical guidelines 1440 – intracranial 646 – intracranial, in children 657 – intraventricular 693 – non-neoplastic 658 – pallidal 1785 – selective 1556 – specific, types 687 – stimulator, functional neurosurgery 1349 – thalamic 1789 lesional zone 2545 lesionectomy 193 lesioning procedures 1353 leucotome 183, 2872 leucotomy – limbic 2880, 2881, 2900 – prefrontal 2870 – stereotactic limbic (SLL) 2947 leukoencephalopathy, progressive multifocal 674 leukotome 8 – Claude Bertrand 116 levator scapulae denervation – operative view 1904 – skin incision 1903
levator scapulae muscle 1901 – anatomy 1902 – contraction 1902 – microsurgical anatomy of innervations 1903 levetiracetam 2672, 2673 Levin cordotomy electrode 1368 levodopa 104, 183, 1508, 1510, 1513, 1530, 1625, 1626, 3161 – responsiveness 1608 Li fraumeni syndrome 3206 Li Pan 130 lidocaine 1339 – infusion 1733 ligand blocker 3220 light emitting diode (LED) 3137 – system 703 Lima, Almeida 2871 Lima, Joffre Moreira 234 limbic lobe 2694 limbic system 16 limited myelotomy 2165 linear accelerator (LINAC) 548, 839, 929, 930, 978, 1112, 1154 – accuracy 931 – adapted 869 – arteriovenous malformation (AVM), radiosurgery 945 – dedicated isocentric 871 – radiosurgery 929, 979, 1172 – radiosurgery, equipment certification 1093 – radiosurgery, non-spherical target 1092 – radiosurgery, photon radiosurgery paradigm 1099 – radiosurgery, radiation testing of isocenter 1093 – radiosurgery, technique 931, 1087 – radiosurgery, treatment 1107 – radiosurgery, treatment plan optimization 1099 – radiosurgery, treatments plans 1092 – robotic 879 – setup 1093 – stereotactic mounting system 1094 linear energy transfer (LET) 859 linear quadratic formulation 1115, 1116 liposome 3092 Lissauer’s tract 1960, 1989, 1990, 2252, 2254 lithium 1866 – carbonate 2500 lobotomy 6 – prefrontal 2871, 2872 – transorbital 2873, 2874 local delivery, enhancing 740 localization – invasive physiological 1283 – invasive physiological, techniques 1284 – magnetic field sensor-based 816 – subcortical physiological 1283 localizer, intraoperative 701 localizing information 726
Subject index
Lord Ganesha 156 low frequency single pulse microstimulation 1392 low frequency stimulation 1391 lower brain stem 2100, 2101, 2106 lower cervical percutaneous cordotomy 2166 lower facial hypesthesia 2809 lower facial paralysis 2809 Lozano, A. 769, 1283, 1383, 1649, 2227, 2793, 2953 Lubag’s disease 1851 lumbar sympathetic ganglia – indications 2302 – local anesthetic block 2301 – surgical ablation 2302 lung cancer, non-small cell 692
M Macaca fascicularis monkey 1395 machine vision 704 macroadenoma 701 macroelectrode 637, 643 – stimulation 1239, 1271 macrostimulation 1285, 1286 – cluster headache 2501 – sensorimotor effects 1292 Madras Medical College and Government Hospital 156 magnetic field 713 – strength, static, risks 283 – field system 703 magnetic induction tomography (MIT) 636 magnetic interference 538 magnetic resonance, advantages 568 magnetic resonance angiography (MRA) 299 – neuronavigation 300 magnetic resonance electrical impedance tomography (MREIT) 636 magnetic resonance imaging (MRI) 475, 504, 652, 1232, 2618 – contrast-enhanced 688 – equipment, primary hazards 282 – feasibility 1816 – fiber tracking 263 – fiducial system 472, 908 – gadolinium-enhanced 690 – image guided surgery 599 – image quality 1240 – intraoperative 577, 558, 606, 714, 719, 821 – intraoperative, advantage 719 – localization 1098 – localizer 505 – management of metal implants and foreign bodies 282 – preoperative volumetric 807 – principles 255 – safety issues 282 – safety in functional neurosurgery 279 – scan 302
– scan, contrast administration, safety issues 284 – scan, parameters 1240 – scan, preoperative 332 – scanner 1240 – scanning for functional stereotaxis 273 – stereotactic localization 476 – technology, basic principles 269 magnetic resonance spectroscopy (MRS) 2625 magnetic source imaging (MSI) 326, 1313, 2622, 2661 magnetoencephalography (MEG) 132, 837, 2601, 2622, 2661 – basic physiology 2661 – clusterectomy 2664 – epilepsy surgery 2663 – lesional epilepsy 2663 – mesial temporal lobe epilepsy 2665 – non-lesional epilepsy 2663 – spikes 2662, 2664 magnification 501 Mai 367 major depression 2887, 2925, 2953 – prevalence 2925 major depressive episode, DSM-IV criteria 2858 malformations of cortical development (MCD) 2635 – diffusion tensor imaging 2637 – EEG fMRI 2637 – electrophysiology 2638 – epidemiology 2635 – invasive recordings 2638 – long-term EEG monitoring 2638 – magnetic resonance imaging 2636 – magnetic resonance spectroscopy 2637 – PET scanning 2637 – presurgical evaluation 2636 – SPECT scanning 2637 – tractography 2637 malignant rolandic-sylvian epilepsy (MRSE) syndrome 2718 mamma carcinoma 1142 management – acute pain 2200 – chronic/cancer pain, ablative procedures 2221 – chronic/cancer pain, comparison 2220 – chronic/cancer pain, narcotics 2221 – chronic/cancer pain, neurosurgical pain procedure 2223 – chronic/cancer pain, stereotactic procedure 2223 – postoperative 657 mania 1490, 2933 mannitol 2730 Mao-ShanWang’s stereotactic device 137 mapping 1937 – acute 2635 – functional 2635 – language 318 – language function 327
3263
3264
Subject index
– motor 318 – motor function 325 – neurophysiologic 325 Marino Jr, Raul 208, 209 Marossero, Franco 193 Martins, Luiz Fernando 220 mass spectrometer 564 master-slave control 589 master-slave system 593 Mattos Pimenta, Aloysio 206 Maure´lio Temponi, Gianni 204 maximum tolerated dose (MTD) 862 Mayfield ACCISS stereotactic system 525, 526 Mayfield Headrest System 525 Measles virus 3101 measurements 272 mechanoreceptor 1990 Meckel’s cave 2434, 2461 Meckelian cave 2432 meclofenamate 2490 medial temporal lobe epilepsy (MTLE) 2764 – radiosurgical treatment, histologic evaluation 2767 medial thalamus – physiological basis of ist role in pain 2087 – role in pain 2081 medications, classes 2070 Medtronic 813 Medtronic cardiac stimulator 18 Medtronic SynchroMed II system 3158 medulla 2102 medulloblastoma 959, 1021, 1058, 3217 – proton therapy 960 Mehrkoordinaten Manipulator (MKM) robotic navigation system 591 melanoma 1056 – metastatic 667 melanoma-astrocytoma syndrome 3212 MELAS 1804 Melzack, Ron 1403 Memorial Stereotactic Body Frame (MSBF) 1207 memory 1338, 2593, 3124 – assessment 2589 meningeal stimulation 1307 meningioma 70, 611, 692, 709, 935, 958, 1016, 1048, 1077, 1134, 3218 – cavernous sinus 935, 936 – clinoidal 1050 – ethmoidal 969 – multiple 925 – proton therapy 959 – spinal 1217 – stereotactic radiosurgery 981 – treatment plan 1119 meningitis 3029 – aseptic 2755
mental functions 2093 meperidine 2493 MERRF 1804 mesencephalic locomotor region (MLR) 1656 mesencephalic region, electrophysiological exploration 1666 mesencephalon, anatomy 2534 mesencephalotomy, cancer pain 2533 mesial temporal sclerosis (MTS) 2683, 2764 metastases 708, 1055, 3205 – cerebral 70, 831, 937 – intracranial, radiosurgery 1141 – leptomeningeal 709 – radiosurgery 1139 – spinal 969 – stereotactic radiosurgery 982 – stereotactic radiosurgery, survival 982 metastatic lesions, median survival 1124 methicillin-resistant Staphylococcus aureus (MRSA) 773 methyltriazeno-imidazole-arboxamide (MTIC) 991 methysergide 2500 metoclopramide 1340 metronidazole 987 mexiletine 1866 Meyers, Russel 8 MGMT 744 Micro MLC 3144 microcarrier, preclinical study 3167 microdialysis 3117 – accuracy 3119 – cerebral 3117 – epilepsy 3123 – ischemic stroke 3122 – learning 3124 – marker 3120 – memory 3124 – movement disorders 3124 – perfusion solution, components 3118 – subarachnoid hemorrhage 3122 – traumatic brain injury 3122 micro-DREZotomy (MDT) 1950, 1959 – effects 2280 – instruments 1965 – pain due to brachial plexus injury 2281 – target 1960 – technique 1966, 2280 – upper limb 1968 microdrive, mechanical 492 microelectrode 477, 637, 1258 – construction method 1287 – construction setup 1287 – mapping 119, 1588 microelectrode recording 13, 138, 466, 491, 1239, 1541, 1587 – electrical stimulation 1284 – functional neurosurgery 1283
Subject index
– – – – –
intraoperative 2796 invasive physiological localization 1283 subcortical mapping 1288 VC nucleus of the thalamus 1269 vs stimulation for physiological localization in stereotactic surgery 1285 micro-electrophysiological recording, intraoperative 100 microinjection, test substance 1286 micromanipulator control 589 Micromar – electrode placer 215 – radiofrequency generator 239 – stereotactic system 212 micromulitleaf collimators (MMLC) 868 microphone array 548 microrecording 1551 – cluster headache 2520 – intraoperative 1615 – neuronal activity 1614 microscope 553 microstimulation 1285 microsurgery 1064 microtransplantation 1701 microvascular decompression (MVD) 1905, 1911, 2465 – complications 2472 – dystonia, cervical 1905 – hemifacial spasm 175 – trigeminal neuralgia 1911, 2465 microvascular Doppler sonography 300 microvascular proliferation 666 micturition 2999 midazolam 1334, 1343 midbrain 360 midcommissural plane 352, 354 mid-sagittal plane 352 mid-thalamic plane 354 migraine 2033, 2103, 2483 – attack, treatment 2490 – criteria for diagnosis 2485 – menstrual 2484 – pharmacotherapy, biological basis 2487 – prophylaxis 2493, 2494 – therapy, efficacy, scientific proof of efficacy, and potential for side effects 2498 – triggers 2486 Minerva robot 592 Mini Multileaf Collimators (MMLC) 971 miniframe devices 524, 529 – intraoperative imaging 527 – mechanical, first burr-hole mounted 522 miniframe frameless stereotaxis 525 miniframe stereotactic apparatus 521 – image-guide surgery 524 minimally conscious states 107 minimally invasive electrode 3048 minute virus of mice (MVM) 3100
Mirapex 1510 misonidazole 987 misregistration 340 – small angular 342 Mitochondrial disease 1804 mitogenic stimuli 3202 MIT-spectroscopy (MITS) 636 MobileSCAN CT 528 model G instrument 471 Model One 544 Model V apparatus 6 model, 3-D geometric surface (polygonal) 398 model, volumetric 398 molecular neuro-oncology, future 3201 molecularly targeted therapy 751 Mondini defect 3026 Moniz, Egas 2870, 2871 monoamine oxidase (MAO)-B inhibitor 1514, 2925, 3162 monocrystalline iron oxide nanoparticle (MION) 827 monosyllabic words 3032 Montreal Neurological Institute 113 mood 1489 mood disorders, brain stimulation, treatment studies 2928 Morel 369 morphine 2054 – intrathecal, pharmacologic side effects, constipation 2185 – intrathecal, pharmacologic side effects, fluid retention 2185 – intrathecal, pharmacologic side effects, hyperalgesia 2186 – intrathecal, pharmacologic side effects, nausea and vomiting 2185 – intrathecal, pharmacologic side effects, neurotoxicity 2187 – intrathecal, pharmacologic side effects, pruritus 2184 – intrathecal, pharmacologic side effects, respiratory depression 2186 – intrathecal, pharmacologic side effects, sedation and cognitive symptoms 2186 – intrathecal, pharmacologic side effects, technical (delivery system) complications 2187 – intrathecal, pharmacologic side effects, urinary retention 2185 Morrell, Frank 2716 motexafin gadolinium (MGd) 989 motion management, future 3151 motor block 1936 motor branches fascicles, anatomical identification 1937 motor circuit 1780 motor cortex 326 – intraoperative mapping 328 – location 2244 – Parkinson’s disease 1665
3265
3266
Subject index
– seizures, primary 2849 – stimulation 2849 motor cortex stimulation (MCS) 240, 1674, 1679, 2041 – frequency 1687 – burr hole versus craniotomy 2243 – complications 2247 – externalized trial 2242 – historical background and outcomes 2240 – implantation 2246 – intensity 1687 – Parkinson’s disease 1679 – Parkinson’s disease, clinical improvement 1684 – patient selection and trial period 2241 – persistent non-cancer pain 2239 – planning of surgical technique 2243 – surgery 1680 – use of dual electrodes 2246 motor fluctuation 1531, 1608, 1725 motor function mapping 325 motor mapping 318, 1277 motor pathway 1300 motor point 3048 motor symptom 1508 – treatment strategy 1517 motor tics 1458 motor-sensory homunculus 113 movement disorders 10, 181, 1072, 1427, 3124 – deep brain stimulation 3003 – definition 1445 – history of surgery 1467 – management, psychiatric considerations 1487 – motor cortex stimulation 1674 – psychiatric aspects of deep brain stimulation 2855 – radiosurgery 1028 – rhythmical, differential diagnosis 1449 – surgery 20, 161, 1445 movement-sensing area 1291 MPTP 1447, 1500, 1569 mTor pathway inhibition 997 Muggiati, Renato de 227 multi-frequency electrical impedance tomography (MFEIT) 634, 638 multi-isocenter circular cone plan 1105 multileaf collimator system 969–971 multiple sclerosis 1450, 1758 – related tremor 1028 – trigeminal neuralgia 2522 multiple subpial transection (MST) 2715 – defining the focus 2718 – failure 2720 – indications 2716 – knife 2719 – postoperative complications 2719 – principles 2715
– rationale 2715 – surgical technique 2718 – rationale 2716 multisynaptic pain pathways 2160 multisynaptic pathway 15 multivane collimator system 969 Mundinger, F. 56 muscimol – infusion 1733 – injection 1732 muscle stiffness 1497, 1950 Mussen’s human stereotactic instrument 80 mutual information registration 339 myelin-stained sagittal section 353 myelitis 1205 myelogram effect 1118 Myelostat Dorsal Column Stimulator 1402, 1403 myoclonus 1460 – causes 1461 – classification 1461 myoclonus dystonia syndrome 1803, 1850 myofascial 2211 myxoma virus 3100
N nanoarray – carbon nanofiber 642 – macro-size 643 nanoelectrode 637 – array 642 nanofibril 640 nanotechnology 585 naproxen 2490 Narabayashi, Hirotaro 59, 119 Narabayashi’s system 9 narcotics 2492 – drug abuse 2205 – inappropriate use 2203 – withdrawal 2204 Nashold, B. 15, 39, 40 Nashold DREZ Electrode tip geometry 1370 natural killer (NK) cell 758 navigation – error 819 – system 300 – system, frameless stereotactic optic 817 – unit, mobile 569 – with CT 602 – with MRI 602 Navigus trajectory guide 529 NCD – complications 2130, 2132, 2133 – electrophysiological monitoring 2129, 2131 – facial pain 2129, 2134 – general anesthesia 2129
Subject index
NCP – Helical lead array 2805 – pulse generator 2804 NDY stereotactic frame 137 neck muscle – denervation 1887 – layers 1897 necrosis 860 Nelson’s syndrome 1045 – radiosurgery 1181, 1182 neocortex, lateral 2693 neoplasia, radiosurgery induced 1075 neoplasms, secondary, radiosurgical associated 1185 neovascularization 753 nephrogenic systemic fibrosis (NSF) 284 nerve growth factor (NGF) 3076 nervus intermedius – decompression 2526 – section 2526 network dysfunction 1782 Neupro 1511 neural electromodulation 2073 neural interface 3185, 3187 – bypass 3194 – cognitive 3194 – direct 3191 – indirect 3189 – input 3185, 3188 – output 3185, 3189 – technology, future 3185 neural transplantation 166 neural-electrical interface (NEI) 637 neuralgia – glossopharyngeal 2477 – glossopharyngeal, diagnostic criteria 2477 – occipital 2507 – vagoglossopharyngeal 2477 neuraxial drug delivery 2073 neuret 861 neurinoma, acoustic 70 neuroanatomy 568 NeuroArm 592, 593 neuroaugmentative procedures 240 NeuRobot 593 neurocircuitry model 2926 neurocytoma 1052, 1053 Neurodan ActiGait 3053 neurodegenerative disorders 1949 neuroendoscopy 807 – complications 811 – pitfalls 811 – ultrasound directed 812 neurofibroma 1217 neurofibromatosis type 1 (NF1) 3207 neurofibromatosis type 2 (NF2) 1015, 3041, 3208 neurogenesis 1395
neurogenetic disorders, gene therapy 3076 neuroimaging 2019, 2684 neurolept anesthesia 1339 neurological disorders, gene therapy 3061 neurolytic therapy 2073, 2074 neuroma, acoustic 872, 874, 1133 – clinical presentation 1152 – diagnostic testing 1153 – epidemiology 1151 – fractionated stereotactic treatment technique 1155 – natural history 1152 – pathology 1151 – patient fixation 1154 – proposed treatment guidelines 1164 – radiation biology 1154 – radiation physics 1154 – treatment goals 1153 NeuroMate (Integrated Surgical Systems, ISS) 591 neuromodulation 16, 637, 642, 1811, 2705, 2793, 2839 – stimulation 17 neuromonitoring 642 neuromotor prosthesis 3193 neuron – catecholamine-containing 1999 – dorsal horn 1991, 1992 – nociceptive dorsal horn 2006 – nociceptive dorsal horn projection 1993 – nociceptive post-synaptic dorsal column 2006 – primate spinothalamic 1994 – spinohypothalamic tract 1997 – spinomesencephalic tract 1997 – spinoparabrachial tract 1997 – spinoreticular tract 1997 – spinothalamic 2032 – STT 1994 – thalamic 2006 – trigeminal primary sensory 2126 neuronal activity, pattern 1392 neuronavigation 24, 106, 136, 567, 703 – CT angiography 299 – frameless 700 – functional MRI 137 – MRA 300 – multi-axial 810 – repeat surgery 818 – rotational angiography 303 – software 579 – window 506 neuronavigator 702 neuropathic pain 2145 – conditions, allodynia 2024 – conditions, basal state 2023 neurophysiologic mapping 325 NeuroPlanner 427
3267
3268
Subject index
neuroprotection 1395, 1623 neuropsychological testing 2686, 2844 Neuro-Station 24, 553 neurostimulation – conceptual model 1415 – implants, electrical safety issues 1411 – implants, electro-biocompatibility 1409 – lead 1421 – system 1405 Neurosurgery Association of the Chinese Medical Doctors Association (CMDA) 134 Neurosurgery Specialty Society of the Chinese Medical Association (CMA) 133 neurosurgery – development in Germany 53 – psychiatric indications 2857 – robotic 583 – robot 591 neurotomy 1937 – femoral 1940 – femoral, skin incision 1941 – hamstring 1938 – median 1941 – median, skin incision 1943 – musculocutaneous 1941 – musculocutaneous, skin incision 1942 – obturator 1938 – pectoralis major 1941 – peripheral 1952 – teres major 1941 – tibial 1939, 1940 – ulnar 1943 – ulnar, skin incision 1943 neurotransmission 1388, 2024 neurotransmitter 2254 – monitoring, micro-level 640 – monitoring, nano-level 640 – system, specific 308 neurotrophin 1732 neurovascular compression, removal 1921 Neurturin 1448, 1724 nevoid basal cell carcinoma syndrome 3212 Newcastle disease virus (NDV) 3100 Nexframe stereotactic device 529, 530 Niemeyer, Paulo 202 nimorazole 987 nitroimadazole 987 nitrosourea 743 NMDA receptor 2696 N-methyl-D-aspartic acid receptor antagonist 3162 nociceptor 1985, 1990, 2005 – group C 1987 – group IV 1987 – primary afferent 2005 – silent 1985, 2005
nomenclature 396 Nomos stereotactic apparatus non-cancer pain, persistant 2227 non-coplanar arcs, multiple 1101 non-motor symptoms (NMS) 1517 non-steroidal anti-inflammatory drug (NSAID) 2490 non-viral vector 3092 nordic countries 65 normal tissue complication probability (NTCP) 1115 Novalis shaped-beam surgery 1209 Novalis stereotactic radiosurgery 873 – efficacy 873 – quality 873 Novalis system 3145 Novalis treatment plan 874 Novalis Tx system 871, 1211 Nowinski, Wieslaw 380 noxious input 2054 Nuclear Regulatory Commission (NRC) 1230 nuclei of the thalamus 1299 nucleotomy, vertical trigeminal partial 2102 nucleus accumbens 2929, 2957 – location of DBS electrodes 2958 nucleus caudalis 2126 – DREZ 2127 nucleus ventralis lateralis (NVL) lesion 1072 nucleus, intralaminar 2004 nucleus, medial thalamic 2004 nucleus, ventral posterior 2004 nucleus, VPI 2003 nucleus, VPL 2003 nucleus, VPM 2003
O O-Arm Intra-operative Imaging System 557, 559 objective hand measurement 1642 obliteration – partial, definition 1060 – subtotal, definition 1060 – total 1069 – total, definition 1060 Obrador, S. 180 obsession, DSM-IV definition 2860 obsessive compulsive disorder (OCD) 106, 147, 1073, 2859, 2887, 2897 – anterior cingulotomy 2900 – capsulotomy, anterior 2901 – cingulotomy 2887 – clinical aspects 2898 – deep brain stimulation 2897, 2904 – deep brain stimulation, patient selection 2905 – deep brain stimulation, stimulation parameters 2908, 2914 – deep brain stimulation, surgical procedure 2905 – limbic leucotomy 2900
Subject index
– neurobiological model 2898 – quality of life 2898 – stereotactic procedures 2900 – subcaudate tractotomy 2900 – treatment options 2899 obstructive hydrocephalus 810 – memory disturbance 695 obstructive sleep apnea (OSA) 1520 obturator neurotomy 1938 – skin incision 1939 occipital lobe epilepsy 2644 occipital nerve – electrode stimulation 2528 – stimulation 2511 – stimulator, electrode placement 2512, 2513 occipital neuralgia 2507 – ablative surgical treatment 2509 – clinical features 2508 – decompressive surgical treatment 2509 – non-surgical treatment 2508 odansteron 1340 Ohm’s Law 631, 1414 Ohye, C. 59, 1549 oil-procaine – injection 8 – wax injection 8 olanzapine 2899 olfactory aura 2563 olfactory effects 1308 olfactory groove meningioma 824 olfactory lobe 2694 oligodendroglioma 705, 3216 – anaplastic 667 – anaplastic 744 Oliveira Jr, Jose´ Oswaldo 214, 215 Oliver’s apparatus 164 Ommaya reservoirs 709 OmniSight Excel 464 ON/OFF stimulation 2784 On-Board Imager (OBI) 875 ONYX 760 open nucleotractotomy 2120 open subcortical procedures 1470 open surgery 941 – spinal metastases 1216 open trigeminal tractotomy 2101 operating microscope 681, 700, 726 operating room, virtual reality 725 operative microscope, LED tracking array 553 opiates – level 2057 – weak 2071 opioids 1334, 1339 – intrathecal, pharmacology 2179 – intrathecal, pharmacology, intraspinal agents 2181
– intrathecal, pharmacology, intraspinal opioids 2180 – intrathecal, pharmacology,mechanism of action 2179 – strong 2072 Oppenheim dystonia 1773, 1802 optic ret 861 optical camera 550 OrthoDoc Preoperative Planning Workstation 590 Orthoplast cap 69 orthostatic hypotension (OH) 1518 orthovoltage X-ray tube 1191 Oulu Neuronavigator System/Leksell Index System 592 OUR-XGD rotating gamma system 140 output test 921 OWL cordotomy system 2141 oxcarbazepine 2673 oxygen inhalation 2498
P p53 tumor suppressor 759, 760 paclitaxel 994 PACS system 571 Pagura, Jorge Roberto 213 pain 10, 1214, 2019 pain, acute 2198 pain after brachial plexus injury, indications 2276 – acute 2032 – acute, experimental 2031 – after spinal cord/cauda equina lesions 2282 – associated component 2022 – barriers to treatment 2061 – bilateral 2000 – brain imaging 2031 – brain response 2025 – causes 2068 – central 2066 – character 2128 – chronic 15, 120, 2041, 2056 – chronic nociceptive 2033 – chronic trigeminal neuropathic 2082 – control 2049, 2058 – clinical 2032 – component 2020 – condition 2020 – discriminative component 2020 – due to benign pathology 2127 – due to malignancies 2286 – duration 2063 – empathy 2022 – facial 15, 2120 – facial neuropathic 2037 – facial, atypical 2522 – facial, distribution 2467 – facial, Gamma Knife surgery 2475 – factors that make it worse 2202
3269
3270
Subject index
– – – – – – – – – – – – – – –
functional neurosurgery 1983 hyperspastic states 2286 hysteric 2026 intensity 2063 interactions between components 2023 intractable 14 location 2064 management 15, 2197 management, general rules 2199 management plan 2218 matrix 2086 mechanism, in cancer patient 2064 migraine 2032 modulation 2053 neuropathic 2032, 2033, 2035, 2038–2041, 2066, 2081, 2522 – neuropathic conditions 2023 – nociceptive 2064 – of benign origin 2197 – of cancer origin 2197 – paroxysmal stabbing 1912 – pathological factors 2201 – pathway 16, 1295, 2160 – pathway, spinothalamic 2050 – perception 2050 – persistent 14 – physiological 2020 – physiological factors 2201 – post-herpetic 2285 – psychological factors 2201, 2202 – quality – radiosurgical 1195 – recurrence after glycerol rhizolysis 2444 – resulting from peripheral nerve lesions 2285 – social history – stepladder 2070 – surgical procedures 2073 – symptoms 2068 – syndromes, directly caused by cancer 2065 – syndromes, psychological factors 2067 – syndromes, with a sympathetic component 2298 – system 1986, 2005 – terms 2066 – thalamic 15 – thalamic nuclei 2085 – thalamotomy 1308 – thalamus during 2004 – types, allodynic 2091 – types, continuous 2091 – types, intermittent 2091 pain, cancer 105, 1028, 1071, 1985, 2067, 2149, 2159 – evaluation 2219 – management 2061 – mesencephalotomy 2533 – neurosurgical indications 2138 – psychogenic aspects 2067
pain, chronic 2198 – depression 2205 – evaluating 2212 – intolerance to stress 2209 – physical examination 2218 – physical regression 2208 – psychological regression 2208 – syndrome 2202, 2211, 2063 – syndrome, components 2224 – syndrome, five components 2211, 2212 palliation 982 pallidal circuit 1791 pallidal electrophysiological map 1589 pallidal oscillation 1786 pallidal spike burst 1786 pallidal spike rate 1786 pallidal stimulation 1812, 1876 pallidal tremor cells 1305 pallidotome without stereotactic armamentarium 180 pallidotomy 13, 20, 119, 181, 1029, 1310, 1472, 1811, 1840, 1842 – efficacy 611 – Laitinen’s reapplication 88 – Parkinson’s disease 1539 – Parkinson’s disease, indications 1540 – Parkinson’s disease, postoperative imaging 1542 – Parkinson’s disease, preoperative imaging 1541 – Parkinson’s disease, surgery 1541 – posteroventral 1349 – posteroventral 1474 – radiosurgical 1195 – safety 611 – side effects 1543 pallidum 1327 – reorganization of functional receptive fields 1788 palsy, progressive supranuclear 1609 pantothenate kinase-associated neurodegeneration (PKAN) 1804, 1836, 1850 parahippocampal gyrus 2701 parahippocampal region 2694, 2697 parallax errors, correction 501 Paramyxoviridae 3100 Parastep stimulator 3055 Parcopa 1509 parietal lobe resection 2644 Parkinson’s disease 10, 11, 115, 312, 1294, 1335, 1349, 1387, 1445, 1471, 2049 – advanced disease with motor complications 1517 – atypical 312, 1445, 1532, 1609 – cognitive outcomes 1488 – differential diagnoses 1447 – early disease 1517 – electrode placement 1487 – gene therapy 3071 – gene transfer 1719
Subject index
– globus pallidus internus, deep brain stimulation, improvement 1593 – globus pallidus stimulation 1577 – idiopathic 312, 1608 – intraparenchymal drug delivery 1731 – medical management 1507 – motor cortex 1665 – motor cortex stimulation 1679 – motor symptom 1508 – neuropsychiatric complications 1487 – non-motor symptoms 1517 – pallidotomy 1539 – pathophysiology 1497 – pedunculopontine nucleus stimulation 1649 – PET images 313 – posterior subthalamic target 1665, 1667 – posterior subthalamic target, electrical stimulation 1668 – of life 1492 – social functioning 1492 – spheramine 3169 – subthalamic nucleus stimulation 1603 – subthalamotomy 1569 – surgery, exclusion criteria 1532 – surgery, patient selection 1529 – surgery, practical considerations 1534 – surgery, specific considerations 1534 – targets to treat 1665 – thalamic stimulation 1631 – thalamotomy 1549 – tissue transplantation 1691 – transplantation 3161 – tremor 1028 – voiding symptom 3003 parkinsonian signs 1571 passive stereoscopic video 704 pathology 654 – techniques 663 patient – assessment 2062 – comfort 926 – fixation 1154 – positioning system 903, 922, 1228 – registration 554, 560 Pauser cells 1590 PC12 neuron 640 PDGF receptor 746 Peacock system 869 pedunculopontine nucleus (PPN) 100, 1649 – afferents 1651 – anatomy 1649, 1650 – connections 1651 – connectivity, basal ganglia circuitry 1652 – efferents 1652 – electrophysiology 1653 – inhibition, evidence 1657
– – – – – – – –
neuronal pharmacology 1653 neurotransmitter effects 1653 stimulation 1649 stimulation, evidence 1656 role in gait disfunction 1658 role in locomotion 1654 role in tone 1654 surgical target in Parkinson’s disease 1658 pedunculopontine tegmental nucleus 1607 pedunculotomy 1469 peel-away introducer 803 Pelorus ball-and-socket device 525 Pelorus stereotactic system 522–524 Penfield, Wilder 113, 114 percutaneous cordotomy 2137 – cervical 2165 – complications 2143, 2144, 2156 – contraindications 2137 – corticospinal tract 2139 – diaphragmatic reticulospinal tract 2139 – electrode 2140, 2154 – electrode positions 2142 – head holder 2140 – indications 2137, 2151 – mechanism 2137 – published success 2143 – spinothalamic tract 2139 – target 2152 – technique 2153 percutaneous freehand trigeminal tractotomy 2111 percutaneous lower cervical cordotomy 2161 percutaneous radiofrequency rhizotomy 2423 – buccal hematoma 2427 – complications 2426 – corneal anesthesia 2427 – cranial nerve palsies 2427 – dysesthesia 2427 – image of needle placement 2425 – late recurrence of pain 2426 – lesioning 2426 – needle placement 2423 – physiological localization 2425 – postoperative care 2426 – relief of pain 2426 – results 2426 – technique 2423 – transient reactivation of herpes simplex lesions 2427 – weakness of the muscles of mastication 2427 Perfexion 886 – collimator helmet 886 – frame cap 912 – unit 903 periaqueductal grey (PAG) 120 – stimulation 105
3271
3272
Subject index
peripheral nerve stimulating (PNS) – lead 1408 – neuropathic extremity pain 1405 – neuropathic pain 2349 – neuropathic pain, complications 2353 – neuropathic pain, implanted nerves 2354 – neuropathic pain, mechanism of action 2349 – neuropathic pain, nerve electrode 2354 – neuropathic pain, results 2350 – neuropathic pain, surgical techniques 2353 – patient selection criteria 2355 peripheral neurotomy 1936 periventricular lesion 810 persistent pain 2199 personal identity 3232 pertinent anatomy 2149 PET scanning 257, 314, 319, 1735, 2036, 2637 Positron emission tomography (PET) 132, 256, 307 Peters, Tony 386 PFA-STN 404, 405 – histogram 406 PFA-VIM 405, 406, 418 phantom base 466, 514 phantom probe 507 phantom ring 488 phase 1258 phase reversal mapping 1276 phenobarbital 2673 phenothiazine 2490 phenytoin 2672 phonic tics 1458 phosphatase and tensin homolog (PTEN) 755 – signaling pathway 755 phosphatidylinositol–3-kinase (PI3K) pathway 755 phosphene interaction, avoidance\abolition 3016 phosphenes 3014 – multiple 3016 photo documentation, intraoperative 635 photodynamic therapy 739 photon radiosurgery paradigm 1099 photons 1154 photosensitizing porphyrin-based dye 739 phrenic nerve stimulation – apnea 2991 – cervical approach 2994 – postimplantation management 2995 – surgical implantation of cervical electrode 2994 – surgical techniques 2993 – thoracic approach 2994 phrenic nerve stimulator 2992 phrenilin 2493 physical dose 864 physician-related factors 2062 physiological localization, applications 1308 pigment epithelial-derived factor (PEDF) 3166 pineal gland 1058
pineal region 655 – tumor 1023 – tumor, endoscopy 711 pinealoma 70 pineocytoma 1058 pins 474, 475 PISCES Spinal Cord Stimulator 1407 pituitary adenoma 70, 825, 981, 1018, 1042, 1134 – definition 1171 – nonfunctioning (nonsecretory), radiosurgery 1174 – proton therapy 960 pituitary lesion 917 pituitary macroadenoma 1046 – nonsecretory 1043 pituitary tumor 884, 3219 – treatments plans 1092 – complications following radiosurgery 1183 – endocrine improvement 1182 – radiosurgery 1171 – radiosurgical goals 1173 – radiosurgical techniques 1172 placebo 2043 placement, depth electrode 579 plane film angiography 1095 planning – atlas-assisted preoperative 410, 425 – neurosurgical 569 – PFA-based 425 – surgical 622, 708 – system 905 – with multiple atlases 425 plasmid DNA 759 – expression vector 3092 platelet-derived growth factor (PDGF) 751, 3166 – receptor (PDGFR) 752, 754 platinum agents 994 plugging 917 pneumoencephalogram 8 pneumoencephalography 142, 193 pneumotaxic guide 117 point matching registration 337 point-based transformation 817 pointing device 819 PointMerge 556 PoleStar iMRI System 557 Poliovirus 3101 poly[bis(p-carboxyphenoxy)propane-sebacic acid] 750 polycationic polymer 3092 polymerically-controlled release 750 polymers, biodegradable 1249 polypyrrole (PPy) 638 polystyrene phantom 921 polytomography 609 pontine lesion 939 pontine micturition center (PMC) 2999
Subject index
pontine stereotactic trigeminal nucleotractotomy 2115 pontine trigeminal nucleotractotomy 2119 positron emission tomography (PET) 132, 256, 307, 1088, 1311, 2602, 2620 – 11C (Carbon) methionine 256 – 18F-FDG (fluoro-deoxy-glucose) 256 – brain tumor 314 – characteristics of brain anatomy 257 – diagnostic 307 – differential diagnostic tool 311 – epilepsy 317 – functional brain mapping 318 – in morphological stereotactic surgery 256 – limits 311 – preoperative assessment tool 311 – radiation necrosis 317 – recurrent tumor 317 – scans in functional neurosurgery 257 posterior subthalamic target 1667 – anatomic-physiologic correlations 1672 posteroventral pallidum 274 postherpetic neuralgia 2127 potential, evoked 1255 poxviridae 3099 pramipexole 1510 precision diode tool 924 prednisone 2500 prefrontal leucotomy 165 pregabalin 1744, 2673 pre-trigeminal neuralgia 2398 primary motor cortex, resection 2644 primidone 1743, 2673 printing functionality 905 probabilistic functional atlas (PFA) 413 – advantages 418 – concept and algorithm 414 – limitations 419 – STN 414 probability of normal tissue complication (NTCP) 859, 860, 862 probability of tumor control (TCP) 858, 859 probe 1245 Probe View 573 procarbazine 743 prochlorperazine 1340, 2492 Profile Of Mood State (POMS) subscale scores 2911 profiling, molecular 745 progressive multifocal leukoencephalopathy (PML) 674 projectile effect 283 prolactinoma 1045 – radiosurgery 1179, 1180 promethazine 2490 promoter 3092 propofol 1334, 1340, 1343, 2653 propranolol 1743
prosthesis, ideal visual 3015 protein kinase C (PKC) 752, 754 – pathway 754 proton 957 proton beam radiosurgery 1172 – angiographically occult vascular malformations 1133 – arteriovenous malformation 1131 – brain metastases 1135 – clinical experience 1131 – meningioma 1134 – pituitary adenoma 1134 – vestibular schwannomas (acoustic neuromas) 1133 – technical and clinical aspects 957 proton beam 68 – therapy 889, 1059 – unit 980 proton density stereotactic MRI scan 275, 276 proton therapy – clinical experience 958 – history 957 – physics 957 proton 853 – dosimetric advantange 958 – production 957 proto-oncogene 751 pseudocyst 673 Pseudomonas endotoxin 757 psychiatric abnormality 10 psychiatric disorders – common neurosurgical procedures 2861 – radiosurgery 1196 – surgery 165 psychiatric illness 1532 – history of neurosurgery 2855 psychiatric surgery 21 – ethical considerations 2855, 2863 psychiatric symptoms, post-DBS, management 2933 psychic aura 2563 psychosis 1522, 2870, 2933 psychosurgery 10, 22, 106, 145, 1073 – historical perspective 2867 – stereotaxic 2876, 2881 pulse generator 1616, 1820, 2803 – implanted (IPG) 2903 pulse width 1391 pulsed radiofrequency technique 1374 pulvinarotomy 1299 PUMA 592 puncture of craniopharyngioma cysts 66 pure tone average (PTA) 1167 putamen 1695, 1698 pyramidotomy 203, 1468 pyrimidine 994 – halogenated 989
3273
3274
Subject index
Q quadrantanopsias 1183 quadrigeminal plate 1058 quality assurance 920, 1230 quetiapine 2899
R radiation – administration 1234 – biology 1154 – body 899, 900 – boost treatment 1215 – changes 1075, 1076 – conventional 741, 1151 – damage, long-term recovery 863 – dose intensity 1102 – dose planning 1206 – effects 854 – focused 1151 – fractionated 1038 – hearing loss 1160 – induced tumor 1011 – injury 1039 – mammalian cell survival curves 854, 855 – necrosis 258, 317, 1011 – physics 1154 – physics, principles 979 – physics, toxicity 979 – risk 1115 – single dose 855 – testing of isocenter 1093 – therapy 741 – tolerance 1039 – treatment, primary 1215 – volume on spinal cord tolerance 862 radio opaque scale 472 radio opaque system 473 radiobiology 853, 1038, 2769 – basic principles 854 – clinical aspects 1038 – CNS tumors 858 – effects at cellular level 1038 – effects at tissue level 1038 – effects on AVMs 1040 – effects on normal brain vasculature 1040 – fractionated radiotherapy 855 – normal tissue 858, 860 – normal tissue, complications 860 radiofrequency 1359 – circuit 1362 – coagulation 492 – current patterns 1362 – electrode 1427 – electrode configuration 1367 – electrode, E-field pattern 1376 – electrode, intracranial lesioning 1374
– electrode, stereotactic 1373 – electrode, temperature falloff 1364 – heat lesion generation, physical principles 1361 – heating, irregularities 1365 – lesion 1359 – lesion, advantages 1360 – lesion, electrode tip diameter 1367 – lesion, generator 174, 1361 – lesion, heating parameters 1366 – lesion, making, practical tips 1379 – lesion, making, rules 1365 – lesion, postmortem sizes 1366 – lesion, sizes 1365 – lesion, spine, neck and back pain 1371 – lesion, trigeminal ganglion 1368 – lesioning 2141, 2527 – rhizotomy 2421 – stimulation, fresh egg whites, effects 1429 – stimulation, human cadavers, effects 1430 – trigeminal rhizotomy 2502 – waveforms 1375 radiograph, digitally reconstructed (DRR) 556 radiological focus point (RFP) 922 radiology, limitations of conventional 269 radio-opaque stylette 1247 radiopharmaceuticals 308 radioresistance 1124 radio-sensitive structures 1111 radiosensitivity 860, 1039 – of tumors, factors affecting 859 radiosensitizer 987 – chemotherapeutic agents 990 – contemporary 989 – future treatment strategy 999 – molecularly targeted agents 994 – traditional 987 radiosurgery 88, 102, 1087 – adverse radiation effects 1011 – amygdalohippocampal 1199 – antiepileptic, mechanism 2769 – boost with whole brain radiotherapy (WBRT) vs WBRT alone 1143 – boost, vs whole brain radiotherapy 1147 – epilepsy 1029 – epilepsy, clinical evidence 2764 – epilepsy, preclinical evidence 2762 – equipment 3146 – extracranial 1208, 1210 – for arteriovenous malformations 941 – for benign tumors 932 – for malignant tumors 937 – functional neurosurgery 1191 – future 3143 – injury, adjacent vascular structures 1184 – Linac 929 – metastases 1139
Subject index
– nonframe-based 3146 – of the spine 261 – pituitary tumors 1171 – quality assurance 1093 – robotic 949 – spinal 952 – stereotactic 742 – targets, classification 864 – technology 867 – treatment planning 1096 radiotherapy – conventional 930 – fractionated 855 – future 3143 – heavy particle 887 – proton beam 957 radium plaque therapy 1059 Ra¨hn, Tiit 70 Ramamurthy, B. 157, 158 Ramos, Jony Soares 225 rapamycin 3210 – mTor 997 rapamycin-dependent expression system 1720 rapid eye movement (REM) behavior disorder (RBD) 1520 rapid onset dystonia-parkinsonism 1453 – DYT12 1803 – RDP 1949 RAS/MAP kinase pathway 3202 rasagiline Rasmussen’s chronic encephalitis 2748 Rasmussen’s encephalitis 2756 Rasmussen’s syndrome 2718 Ray Rhizotomy Electrode 1373 real-time 704 – intraoperative imaging 820 receiver, electromagnetic 536, 537 receptor – antigen targeting 757 – on receptor 1987 – pharmacological 1988 – substance P 1988, 1991 – transmitter 1989 reciprocal gait orthosis 3054 Recombinant DNA Advisory Committee (RAC) 3064 reconstruction, multiplanar 713 recovery cycle test 2841 rectal carcinoma 843 rectified image 820 recursive partitioning analysis (RPA) 1139 reference frame 819 reference planes 357 referential recording 1258 reflective markers 819 refractory depression, stereotactic ablative procedures 2944
registration 335, 336, 396, 701, 817, 949 – future 3138 – methods 336 – mutual information 339 – point matching 337 – sources 819 – stereotactic frame 337 – surface/edge matching 337 – surface-based 701 – unit, mobile 577 – voxel-based similarity 339 regulatory TCD-based 2089 Regulus navigator 24 relative biological effectiveness 859 reluctance – patient 2061 – prescribe opioids 2062 remifentanil 1334, 1339, 1343 Remy 100 Rennes setup 499 reovirus 3100 repetitive behaviors 2898 repetitive transcranial magnetic stimulation (rTMS) 1685 replication-competent virus 3094 Requip 1511 research prototypes 427 resection – limits 713 – stereotactic endoscopic 625 – stereotactic, method 683 – surgical navigation 623 reserpine 1776 respiratory cycle 2992 resting tremor 1758 restless legs syndrome (RLS) 1520 reticular nucleus 2793 reticulospinothalamic projection 2535 retinal epithelial pigmented cells 1625 retinoblastoma (Rb) protein 760 retinoblastoma, trilateral 3212 retractor, cylindrical stereotactic 682 retractor, stereotactic cylindrical 685 retractor-dilator system 682 retrogasserian glycerol injection 2429, 2503 – anesthesia 2431 – bacterialmeningitis 2448 – branch selectivity 2438 – complications 2445 – contrast evacuation 2436 – herpes simplex activation 2448 – hypesthesia 2446 – infectious complications 2447 – infectious manifestations 2447 – intracisternal deposit 2437 – late outcomes 2442
3275
3276
Subject index
– – – – – – – –
mechanisms of glycerol 2448 non-bacterial meningeal irritation 2448 pain recurrence 2444 patient positioning 2432 permanent marking of cistern 2439 pharmaceutical regimen 2441 positioning of the head 2438 post-injection disturbance of facial sensibility 2446 – postoperative management 2441 – problems with comparison of different series 2442 – repeat injections 2439 – results 2441 – sensory side-effects 2445 – short-term outcomes 2442 – specific difficulties 2434 – surgical anatomy 2432 – trigeminal cistern 2440 – trigeminal cisternography 2433 – various kinds of dysesthesias 2447 – treatment of TIC 2452 retroviral vector 3087 retrovirus 761, 3062, 3083 REZ – distance from the pons 2376 – regeneration 2376 rhizotomy – dorsal 1926 – dorsal, techniques 1947 – intradural 1885 – lumbo-sacral dorsal 1948 rhombencephalon 356 rhythmicity, heart 3047 Richardson, Alan 2880 Riechert, Traugott 54, 56 Riechert-Mundinger – frame 70 – stereotactic apparatus 487 – stereotactic apparatus, classical 488 – stereotactic apparatus, translational system 488 – system 489 – system, instrument holder 490 risperidone 2899 RNA virus, oncolytic 3100 Roberts, D. 462 Roberts, T. 456 RoboDoc 590 robot 583 – neurosurgical 591 – surgical 587 robot-assisted microsurgery system (RAMS) 594 robotic device 563 robotic motion compensation 954 robotic neurosurgery 583, 949
robotics 106, 136 – future 3139 Roentgen 22 rolandic cortex, identification 2654 ropinorole 1511 ropivacaine 1339 Rosenthal fiber 670 RosoMoff, H. 50, 1367, 2143, 2154, 2161 Rossi, Gianfranco 193 rostral limit of the lesion 2102 rotation angle 444 rotigotine transdermal system 1511 Rubinstein-taybi syndrome 3212
S S7 system 559 Safety 279–285 sagittal section 1999 Sainte Anne setup 498 Salles, Antonio Afonso Ferreira De 235 sampling, techniques 649 Sang Sup Chung 174 sarcoma 1121, 1122 scala tympani 3026 scan thickness 271 scanner slice thickness 819 scanning electron microscopy 640 scanning plane 270 Scerrati’s goniometer 498 Schaltenbrand 351, 363 – reference system 423 Schaltenbrand-Wahren atlas 573 schizophrenia 22, 2049 schwannoma 610, 1126, 1217 – non-vestibular 1024 – trigeminal 1052 – vestibular 888, 932–934, 1012, 1048, 1125, 1133, 3219 – vestibular, stereotactic radiosurgery 981 sciatic nerve 2051, 2054 sclerosis – hippocampal 2844 – mesial temporal 1199, 2619 sculpture, homuncular composite 362 Sedan-type needle aspiration biopsy system 477 sedation 1340 sedative 2493 Segawa’s disease 1802 segmentation 817 seizure 1061, 1073, 2702 – akinetic 2567 – aphasic 2567 – astatic 2567 – atonic 2567 – autonomic 2564 – causes 2670
Subject index
– clonic 2565 – control, cerebellar stimulation 2823 – dialeptic 2564 – discharge, spread 2716 – epileptic 2562 – epileptic, classification 2561, 2562 – evolution 2568 – focal, eloquent cortex 2716 – focus, stimulation 2839 – frequency 2571 – Gamma knife radiosurgery plan 1198 – generalized tonic-clonic – mesiotemporal, recording 2552 – motor, complex 2566 – motor, simple 2565 – myoclonic 2565 – negative myoclonic 2567 – onset pattern 2554 – onset zone 2546, 2550 – semiology 2549 – surgery, goal 2726 – tonic 2565 – tonic-clonic 2565 – types 2831 – vagal nerve stimulation 2801 – versive selective peripheral neurotomy (SPN) 1936 selective serotonin reuptake inhibitor (SSRI) 2925 selective temporal lobe amobarbital memory test (STLAMT) 2556 selegiline 1514 – oral disintegrating semiological seizure classification 2571 Sendai virus 3101 sensitization 1991, 1992, 1996 – central 1991 – windup 1991 sensitizer – hypoxic 987 – non-hypoxic 989 sensor module 3189 sensorimotor effects, macrostimulation-induced 1292 sensorimotor network 1781 sensory cortex, resection 2644 sensory system, anterolateral 2150 sensory tricks 1861 septostomy 808 – endoscope view 810 septum pellucidotomy 811 serotonergic antidepressants 2899 serotonin 309 – antagonist 2496 – migraine 2488 – reuptake inhibitor 2496 serotoninergic system 309 Severance Hospital 173
sexual dysfunction 1519 SGS-I stereospecific gamma ray whole body treatment system 140 Shaltenbrand-Wahren coronal plate 260 Shanghai Huashan Institute of Neurological Surgery (SHIN) 135 Shealy, N. 18, 19, 1371–1373, 1402, 2050 Shealy Rhizotomy Kit 1372 shielding 916 – automated 926 Shiva Purana 155 shunt insertion 808 shunt placement 811 shuttle time 926 side effects of intrathecal morphine, pharmacologic Siegfried, J. 73 Siegfried side-outlet stereotactic electrode tip geometry 1375 Siegfried stereotactic electrode 1374 Signa SP double doughnut unit 714 signal transduction pathway 994 signaling pathway 752 Silicon Photonics 3132 Sindbis virus 3101 Sindou, M. 1935, 1959, 2269 Sinemet 1508 single dose radiosurgery, biologically effective dose 857 single isocenter plan 1108 single lung cancer (NSCC) metastasis 1142 single modality image augmented fusion 822 single photon emission computed tomography (SPECT) 132, 2604, 2619, 2841 single-beam measurement system 919 single-dose effect curve 856 single-event double-strand break 855 skin burn 1362 skin fiducials 819 skin incision 1895, 2754 skull base 656, 823 – cancer, invasive 1027 – surgery 783, 823 – surgery, extended endoscopic anterior 824 – tumor 1024 – tumor, imaging, preoperative 815 – tumor, intraoperative image guidance 815 skull data 905 sleep 1491 – apnea 1520 – disturbances 1519 small-angle double-incidence angiogram (SADIA) 502 SMART trial 990 smear – of anaplastic astrocytoma 666 – of ependymoma 672 – of glioblastoma 666, 667 – of low grade astrocytoma 669
3277
3278
Subject index
– of malignant lymphoma 668 – of medulloblastoma 672 – of meningioma 673 – of metastatic adenocarcinoma 667 – of metastatic melanoma 668 – of normal cerebrum 669 – of oligodendroglioma 670 – of psammoma bodies 673 – of subependymal giant-cell astrocytoma 671 – recurrent astrocytoma 675 – technique 663 Snapper-Stereoguide 527 SNN-Olivier FreeGuide 525 social barrier 2062 Sociedad Latinoamericana de Neurocirugia Funcional y Estereotaxia (SLANFE) 44 Society for Stereotactic and Functional Neurosurgery 214 somatosensory aura 2563 somatosensory cortex 326 somatosensory evoked phase-reversal potential mapping 1274 somatosensory evoked potential (SSEP) 1255 somatosensory testing 1275 somatostatin 1020 somposite shot 926 sonic base ring 546, 547 sonic digitizers 702 sonic forceps 547 sonic workstation 548 SonoNav 558 Souza, Paris Ferreira 229 spacing tool 1106 Spain 179 spasm, epileptic 2566 spasticity 1925 – anesthesia 1937 – decision-making in adults 1950 – decision-making in children 1953 – destructive neurosurgical procedures 1935 – EMG recordings 1929 – global functional score 1971 – intrathecal baclofen therapy 1951 – intrathecal drugs 1973 – lower limb 1955 – mapping 1937 – methods for managing 1936 – neuro-destructive procedures 1952 – neurosurgical management 1925 – operative techniques 1938 – orthopedic surgery 1949 – pre-operative motor blocks 1936 – spasm rating scales – surgery for the lower limb 1938 – surgery for the upper limb 1941 – surgery for the upper limb, causes of recurrence 1944
– surgery for the upper limb, complications 1944 – surgery for the upper limb, side-effects 1944 – surgery in dorsal root entry zone 1945 – surgery in spinal cord 1945 – surgery in spinal roots 1945 – upper limb 1956 spatial cross-correlation 398 spatial exploration 398 specific absorption rate (SAR) 1816 spectroscopy, near-infrared 2597 speech 1303 – center 717 – encoding 3030 – problems 1609 sphenopalatine ganglion 2525 spheramine 1448, 3168 spherical coordinates 499 Spiegel, Ernest A. 5, 35, 36, 46, 348 Spiegel-Wycis Award 40 Spiegel-Wycis procedure 8 Spiegel-Wycis stereotactic apparatus 6, 252 spike-wave complex 2787, 2788 spinal circuits 1931 spinal cord 1999, 2001 – C2 level 2151 – cauda equina lesions, anatomical data 2283 – cauda equina lesions, surgical procedure 2283 – cross section diagram 2252 – electrode 1367 – injury, radiation-induced 1205 – lesions 2993 – localization of opioid receptors in the substantia gelatinosa 2255 – longitudinal diagram 2253 – tolerance, radiation 1205 – tolerance, radiosurgery 1205 – transverse hemisection 2270 spinal cord stimulation (SCS) 240, 1906, 2305, 3188 – adverse effects 2324 – angina pectoris 2322, 2338 – angina pectoris, clinical outcome 2323 – angina pectoris, diagram illustrating effects 2339 – angina pectoris, patient selection 2322 – angina pectoris, safety aspects 2324 – angina pectoris, technical aspects 2322 – assessment of outcome 2310 – clinical outcome 2319 – clinical pain states associated with dysautonomia 2336 – clinical studies 2313 – complex regional pain syndrome type I (CRPS I) 2315 – cost-effectiveness 2324 – cutaneous hypersensitivity 2334 – cylindrical (percutaneous) electrode 1407
Subject index
– – – – – – – – – – – – – – – – – – – – –
dorsal horn and spinal circuitry 2333 dual lead electrodes 2312 effects on cardiac function 2340 electrode designs and configurations 2310 failed back syndrome 1405 features of the pain 2307 hierarchy of study types 2312 irritable bowel syndrome 2341 ischemic pain 2317, 2337 lead electrodes 2311 lumbar spinal and radicular pain 2315 lumbosacral rhizopathy and low back pain 2313 mechanisms of action 2331 neuropathic extremity pain 1405 neuropathic pain 2306, 2332 organ dysfunctional syndromes 2342 paddle-type lead 1408 pain analysis 2306 pain in peripheral vascular disease (PVD) 2318 participation of the patient 2307 percutaneous implantation of a lead electrode 2309 – possible mode of action 2336 – possible transmitter mechanisms 2335 – selection of patients 2306 – studies for ischemic limb pain 2321 – system 1403 – technical aspects 2308, 2319 – stimulation 2308 spinal delivery of medication 3155 spinal lesion, stereotactic radiosurgery 983 spinal metastases 1203 spinal radiosurgery 1203 – advantages 1219 – after open surgery 1216 – arteriovenous malformation (AVM) 1218 – meningioma 1217 – neurofibroma 1217 – pain control 1214 – procedure 1206 – progressive neurological deficit 1215 – quality of life improvement 1214 – radiographic tumor progression 1215 – schwannoma 1217 spinal tracking without fiducials 952 spinal tumor, pain 1214 spine 658 – lesions, malignant, radiosurgery 1211 – radiosurgery, background 1204 – radiosurgery, indications 1214 – radiourgery, candidate lesions 1214 – stereotactic surgery 261 spinocervical tract 2003 spinohypothalamic tract 1998 spinomesencephalic tract 1995 spinoreticular tract 1995, 2051, 2052
spinothalamic tract 1295, 1993, 1995, 2051, 2052, 2535 – interruption 2162 – primate 1996 – surgical anatomy 2159 spiral biopsy system 478 spiral scanner 476 spondylosis, moderate cervical 1862 spread-out Bragg peak (SOBP) 887, 889 SSEP pathway, median nerve 1273 staging 1116 Stalevo 1510, 1513 Staphylococcus 769 STASSIS 1571 status epilepticus 2671 – treatment 2671 Steady Hand System 590 Stealth 703 Stealth Station 24 StealthMerge 554 StealthStation 415, 605 – images 602 – treatment guidance system 543 – treatment guidance system, current status 554 – treatment guidance system, frameless localization 551 – treatment guidance system, history 543 – treatment guidance system, prototype 544 – treatment guidance system, second generation 551 Steiner, Ladislau 70 stellate ganglion 2298 – ablation, cervicothoracic sympathectomy 2300 – ablation, neurolytic block 2299 – block, anatomy 2298 – block, indications 2299 – block, local anesthetic procedures 2298 stem cell 1625, 1702 – differentiation 1702 – engineered 3177 – future applications 3177 – host neural circuits 1704 – human, sources 3171 – neural 3171 – therapy 3170 – therapy, delivery system 3174 – transplantation 1702 – transplantation caveats 1704 – trophic effects, therapeutic applications 1704 stereoadapter 511–516 – Cartesian reference structures 512 – noninvasive multipurpose 511 Stereocalc 464 stereo-electro-encephalography 103, 2021 stereoencephalatome 80
3279
3280
Subject index
stereoencephalography 74 stereoencephalotomy 7, 35, 348, 350 stereoguide 511, 513, 515 stereotactic accuracy 603 stereotactic amygdalotomy 166 stereotactic angiography 104, 680 stereotactic apparatus, miniframe 521 stereotactic aspiration 802 – endoscopic 774 – frameless 774 – interactive MRI-guided 775 – ultrasound-guided 775 stereotactic atlas 85 – contemporary 367 – electronic 373 – orientation of slices 7 – printed 347 – traditional 363 stereotactic biopsy 101, 645, 711 – comparison frame-based vs frameless 651 – complication rate 736 – frame-based 621, 648 – frameless 649 – frameless stereotaxis 525 – ideal lesion 736 – iMRI 609 – magnetic resonance imaging 599 – pathology techniques 663 – success rate 736 – survival 737 stereotactic brain surgery, functional 624 stereotactic coordinate base ring 473 stereotactic craniotomy 679 – methods 680 – patient selection 686 stereotactic CT scan 515, 680 stereotactic device 455 – of Jiang (left) and Wu SL 132 – of Wu SL 132 stereotactic electrocoagulation 180 stereotactic electrode-guiding device, installation 1816 stereotactic endoscopic surgery, CT-based 624 stereotactic environments, future 432 stereotactic frame 270 – application 474, 1231 – attachment 902 – computerized imaging system 504 – history of the development 97 – of Barros 218 – registration 337 stereotactic functional neurosurgery 1331 stereotactic functional surgery 604 stereotactic guidance, accuracy 443 stereotactic head frame (Da-Jie Jiang) 130 stereotactic hemiarc 480
stereotactic image acquisition 942 stereotactic imaging 1095 – postoperative 275 – system 1089 stereotactic implantation, cluster headache 2519 stereotactic instrument – for radiosurgery 67 – production of the first 59 stereotactic irradiation – fractionated 515 – surgery, open 623 stereotactic localization, procedures 1089 stereotactic localizing box 255, 257 stereotactic microsurgery 478 stereotactic MRI scan 275, 276 stereotactic neurosurgery, atlas-assisted 421 stereotactic operating room – CT 627 – evolution 626 stereotactic pallidotomy 115 stereotactic pontine nucleotractotomy 2120 stereotactic procedure, functional 515 stereotactic radiosurgery (SRS) 140, 608, 742, 839, 867, 929, 2704 – brain metastases, studies 1146 – definition 885 – facts 977 – forms 978 – heavy particle radiotherapy 890 – linear accelerator 868 – quantitative parameters 867 – radiobiology 853 – spinal 883 – with MRI 608 stereotactic reference planes 351 stereotactic registration 335 – methods 443 stereotactic retractor 685 stereotactic societies, history 35 stereotactic space 1097 stereotactic surgery – ansotomy 1472 – birth of human 5 – brain, morphologic 620 – frame based 619 – frameless 600 – frameless CT- based 625 – history 3, 1155 – history in Brazil 197 – history in Canada 113 – history in China 125 – history in France 97 – history in Germany 53 – history in Great Britain 77 – history in India 155 – history in Italy 193
Subject index
– history in Japan 59 – history in Korea 171 – history in Spain 179 – history in Switzerland 73 – history in the Nordic Countries 65 – history in US 45 – imaging 249 – Madras Medical College 158 – pain 105 – pallidotomy 1472 – physiological localization 147 – subthalamic region 1472 – target localization 1473 – thalamotomy 1472 – tools 490 – volumetric 696 stereotactic system – of Roberts 700 – electromagnetic, advantages 540 stereotactic target 1094 stereotactic targeting 345, 464 stereotactic technique 2090 stereotactic thalamotomy 81, 194 stereotactic thermocoagulation for central pain 147 stereotactic translation 685 stereotactic treatment of movement disorders 104 stereotactic trigeminal nucleotractotomy 2106 – results and complications 2110 stereotactic ventriculography 1610 stereotaxic atlas 354 stereotaxic neurosurgery 567 stereotaxis 77 – electromagnetic 536 – frameless, stereotactic biopsy 525 – functional 516 – miniframe frameless 525 – morphological 604 – nonfunctional 514 – volumetric 695 – volumetric, advantage 696 stereotaxy 97, 179, 815 sternocleidomastoid muscle denervation 1887, 1900 steroid medication 737 sterotactic biopsy surgery 1337 StimLoc lead anchoring burr hole system 1588 STIMuGRIP 3056 stimulation – central gray 2054 – complications 1595 – deep brain 2034 – dorsal collum 2054 – effects on autonomic function 1300 – effects on consciousness 1300 – effects on mood 1300 – electrical 2021
– hippocampus 2839 – mapping 2653, 2692 – motor cortex 2039 – of deep structures, placement of electrodes 521 – PAG-PVG 2039 – physiology, functional neurosurgery 1383 – procedure 1274 – protocols 1413 – response location 1270 – seizure focus 2839 – subcortical 1476 – technology 1420 – technology, functional neurosurgery 1401 – thalamic 2035–2040 – VPM/VPL nuclei 2058 stimulator – implanted 1349 – lesions 1349 stimulus – intensity 1290 – router system 3049 – noxious visceral 2000 STIMuSTEP 3053 Stoffel’s technique 198 storage reflex 3002 Streptococcus 769 striatal output, disinhibition 1784 striatum, structural lesions associated with dystonia 1783 strip electrode 1258 structure searching 398 studies – neurophysiological 2050 – PET 2031 Sturge-Weber disease 2746 subcallosal cingulate gyrus (SCG) 2954 subcortical mapping with microelectrode recording 1288 subdural strip electrode 2688 subependymal giant cell astrocytoma (SEGA) 3209 subgenual cingulate cortex (SCC) 2927 subnucleus – caudalis 2098, 2100, 2102, 2108, 2109, 2115, 2120 – interpolaris 2098 – oralis 2098 substantia nigra complex 1606 subthalamic area 1300 subthalamic nucleus 274, 269, 1383, 1477 – anatomy, functional 1603 – anatomy, photomicrographs 1605 – chemoarchitecture 1604 – cytoarchitecture 1604 – deep brain stimulation 1569 – deep brain stimulation, effects 1389 – electrophysiology 1605 – implantation of electrodes, surgical procedure 1610 – lesion 1570
3281
3282
Subject index
– stimulation, Parkinson’s disease 1603 – stimulation, Parkinson’s disease, bad prognosis factors 1609 – stimulation, Parkinson’s disease, complications 1618 – stimulation, Parkinson’s disease, complications 1621 – stimulation, Parkinson’s disease, contraindications 1608 – stimulation, Parkinson’s disease, costs 1621 – stimulation, Parkinson’s disease, hardware improvements 1626 – stimulation, Parkinson’s disease, indications 1608 – stimulation, Parkinson’s disease, physiology improvements 1626 – stimulation, Parkinson’s disease, side effects 1618 – stimulation, Parkinson’s disease, side effects 1621 – stimulation, Parkinson’s disease, software improvements 1626 – target coordinates 1613 – topography 1604 subthalamic nucleus procedure 1261 subthalamic nucleus stimulation 1353 – clinical outcome 1248 subthalamic nucleus, impedance profile 1328 subthalamotomy 1438 – complications 1574 – technical aspects 1570 subthalamus 1672 subtotal obliteration of AVM (SOAVM) 1065 sufentanil 1339 suicidal ideation, management 2935 suicide 1490 – gene therapy 760 sumatriptan 2491, 2498 supercomputer 3132 superconducting magnets, cylindrical 821 superconducting quantum interference device (SQUID) 2661 superficial cerebral hemosiderosis (SCH) 2742 superficial petrous nerve, section 2526 superposition algorithm 919 superposition of beams 897 supplementary motor area 2643 support, atlas-assisted intraoperative 410, 425 supraorbital nerve stimulation 2528 surface-based image registration 338 surface/edge matching registration 337 surface electrodes 3048 – schematic of the application 637 surface matching 818 surgery 517, 2061 – chronic pain 164 – cytoreductive 738
– functional 88 – movement disorders 137 – transsphenoidal 602 surgical cordotomy – for pain 2161 – technique 2163 surgical field 726 – real-time view 726 surgical instruments 472 surgical navigation 600 – device 552 surgical planning 708 – system 483 Surgical Planning Laboratory (SPL) 381 surgical procedure 681 surgical robot 587 surgical targeting, transsphenoidal 709 surgical technique, accuracy 1248 SurgiPlan 480, 483, 622, 1550 SurgiScope operating microscope system 3138 surviving fraction 858 Sweet, William H. 1360 Switzerland 73 Symmetrel 1516 sympathectomy, for pain 2297 sympatholysis 2298 symptomatic trigeminal neuralgia 2361 – causes 2362 symptomatogenic zone 2549 Synchrony Respiratory Tracking System 954 Synergy Experience 557 Synergy S 1210 system control 556, 561 system, medial lemniscus 1998
T tachycardia 2564 tactile area 1288 Tae Joon Moon 174 Taira’s method 1896, 1898 Talairach 350, 364 Talairach stereotactic frame 496, 499 – modified 505 – new 506 – setup 498 Talairach grid 397 Talairach landmarks 407 Talairach stereotactic system 495 – data computation 501 – data processing 500 – description 496 – routine procedure 499 Talairach transformation 422 – fast 422 Talairach, J. 101, 103, 2878, 2879 Talairach’s space 366
Subject index
tamoxifen 754 tap response 1555 Tarceva 753, 3220 tardive dystonia 106 target coordinates, determination 516 target determination 1819 – Schaltenbrand atlas 1808 – Talairach atlas 1809 target immobilization 1207 target localization 1207, 1473 – conventional radiographic 475 target position, transposing into stereotactic space 1242 target shape 1096 target trajectory 462 target volume 1116 – determination 1233 target, ventriculographic determination 507 targeting – accurate, peri-operative confirmation 1244 – CT image 605 – errors 1244 – frame-based 571 – frameless 575 – left Vim 605 – on multiple orientations 424 – PFA-determined accuracy 423 task block 289 Tasker, Ron 119, 358 Tasmar 1513 taxane 994 Teixeira, Manoel Jacobsen 211, 212 telecommunication, future 3139 telemetric stimulation Telles Ribeiro, Carlos Roberto 204 temozolamide 744, 991 temperature 1432 temporal bone, surgery 826 temporal lobe epilepsy 2633, 2677 – assessments for surgical candidacy 2684 – epidemiology 2677 – lesional – mesial (mTLE) 2680, 2698 – neocortical (nTLE) 2682, 2699 – neuroimaging 2684 – subtypes 2680 – surgery, goals 2679 – surgery, indication 2679 – surgical strategies 2698 tenascin 757 TENS unit 18 Terminologia Anatomica 396 test microinjection 1286 tetrabenazine 1776, 1866 tetracycline transactivator (tet) system 1720 TEW Electrode 1371 T-field 1378
thalamic deep brain stimulation, Benabid’s application 88 thalamic electrical activity 1564 thalamic neuronal activity 1553 thalamic procedure 1268 thalamic stimulation 105, 1761, 1763, 1875 – anti-PD medication usage 1641 – completion of DBS implantation 1636 – headframe application 1632 – imaging 1632 – intra-operative confirmation of the target 1633 – long-term efficacy 1640 – neuropsychological outcomes 1641, 1645 – neuropsychological performances 1644 – operative setting 1636, 1637 – outcomes 1639 – Parkinson’s disease 1631 – Parkinson’s disease, patient selection 1631 – Parkinson’s disease, surgical procedure 1631 – patient selection 1638 – programming the impulse generator 1637 – side-effects 1641 – stimulation parameters 1639 – stimulator settings 1641 – surgical complications 1640 – surgical planning 1632 – surgical technique 1632 thalamic subnuclei 274 thalamocortical dysrhythmia 2088 thalamotomy 11, 115, 183, 348, 1028, 1472, 1749, 1760, 1810, 1840, 1841, 1874 – central lateral, clinical results 2091 – central lateral, complications 2092 – efficacy 611 – gamma 1557 – gamma, after stereotactic thalamotomy 1563 – gamma, indication 1558 – gamma, operation 1558 – pain 1308 – radiosurgical 1193, 1751 – safety 611 – selective 1549 – technical improvement 1550 – thalamic deep brain stimulation 1310 – tremor 1310 – ventral 1351 – ventro-lateral 181 thalamus 354, 357, 360, 369, 1288, 1327, 1549, 1559, 1789, 2003 – anatomy 1746 – clinical neurophysiology 1309 – human 2084 – neuronal activity changes 1498 – overlapping functional representation 1790 – physiology 1746
3283
3284
Subject index
– sagittal section 2778 – sensory, electrode trajectory 1289 – somesthetic 358 – target zone for tremor surgery 1556 therapeutic ratio 860 therapy, local 739 thermocapsulotomy 2901 thermoreceptor 1990 3-D (three-dimensional) anatomy 725 3-D (three-dimensional) proportional system 364 3-D (three-dimensional) visualization 398 tic douloureux, landmarks in the surgical treatment 2401 TIC Kit electrodes 1371 tics 1458, 2860 – de Gilles de la Tourette 106 – etiological classification 1459 tinnitus 1152 tip geometry 1368 tip temperature 1365 tissue – damage 1411 – impedance 632 – safety 1409 – specimen 646 – transplantation 14, 1691 – transplantation, Parkinson’s disease 1691 – transplantation, Parkinson’s disease, double-blind randomized controlled trials 1696 – transplantation, Parkinson’s disease, open-label trials 1693 Todd, Edwin M. 453, 454 Todd-Wells apparatus 10, 453, 456 – conceptual schematic 455 Todd-Wells frame 131, 600 Todd-Wells system 457 Togaviridae 3101 tolcapone 1513 tolerance dose 860 tomography 633 tomotherapy 877, 970 – efficacy 878 – performance features 877 – quality 878 – treatment plan 880 – unit 3148 TomoTherapy Hi-Art system 1209 tonic seizure 2779 topiramate 2673 topography, optic 2597 topotecan 993 Toronto Western Spasmodic rating Scale 1891 torticollis 1857 – non-dystonic causes 1858 – spasmodic 1769 total dose 862 touch sensation 2105
Tourette’s syndrome 1458, 2860 – characteristics 2860 – deep brain stimulation 2964 – deep brain stimulation, patient selection 2966 – deep brain stimulation, perioperative evaluation 2968 – deep brain stimulation, post-operative evaluation 2968 – deep brain stimulation, surgical procedure 2968 – surgical procedures 2963 – treatment 2964 Tournoux 364 toxicity 979 Toxoplasma 673 toxoplasmosis, cerebral 673 tracking, intra-operative 555, 560 tractography 611, 716 tractotomy – mesencephalic 1309 – stereotactic subcaudate (SCT) 2944 – subcaudate 2879, 2880, 2900 Traditional Chinese Medicine (TCM) 126 TRAIL 3083 trajectory 463, 809, 1243 – determination 1819 – overview 573 – planning 709 – problem 464 transaction, subpial 2715 transcranial magnetic stimulation 2595 transcutaneous electric nerve stimulation (TENS) 1403, 2307 transcutaneous nerve stimulator (TNS) 1403 transfection efficiency 3092 transferrin receptor 3093 transferrin-CRM 757 transform matrix 458 transformation – matrix 445 – non-rigid transforming growth factor (TGF) 756, 3166 transmitter, electromagnetic 536, 537 transplantation 1477, 3161 – appropriate patients 1697 – improving transplanted cell survival 1700 – optimal targets 1698 transplanted cells, identification 3174 transurethral resection of the prostate (TURP) 592 trauma 2641 treatment – delivery 968 – delivery space 1097 – evaluation tools 905 – export 905 – field 915 – optimization system 1099
Subject index
– options 2068 – plan 2068 – plan optimization 1099 – planning 967 – planning system 967 – planning tool 905 – planning, future 3151 – planning, inverse 864 – planning, radiosurgery 863 – resistance 994 – multi-session, rationale 1116, 1117 tremor 11, 1028, 1449, 1531 – assessment 1760 – at rest 1497 – augmentation 1256 – cells 1293 – clinical classification 1757 – differential diagnosis 1449 – dystonic 1449 – essential (ET) 1449 – etiology 1758 – management 1757 – parkinsonian, microinjection of lidocaine 1294 – parkinsonian, thalamic microstimulation 1294 – parkinsonian, thalamic stimulus parameter 1301 – post-traumatic 1763 – reduction by electrical simulation 1293 – reduction by lidocaine microinjection 1293 – reduction, gamma thalamotomy 1560 – stereotactic target 1763 – surgical treatment 1760 – thalamic stimulation 1761, 1763 – thalamotomy 1760, 1763 trephine craniotomy 521, 684 tricyclic antidepressant 2071, 2495, 2925 tricyclics 2207 trigeminal cistern 2433, 2435, 2436, 2440 trigeminal ganglion 2527 trigeminal nerve 2103, 2126 – intracranial portion 2376 – preservation 1013 – root entry zone, radiosurgery 2527 – sheath tumor 1024 – three-dimensional time-of-flight (3D-TOF) image 2468 – vascular anatomy 2377 – vascular anatomy, anatomical studies and surgical observation 2378 – vascular anatomy, imaging studies 2381 – vascular anatomy, neurovascular anatomic studies 2379 trigeminal neuralgia 15, 1072, 1911, 2429, 2465, 2476 – advanced imaging techniques 2405 – anatomy of the REZ 2374 – atypical 2476 – characteristics 2457
– – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – –
classification 2359 clinical features 2421 clinical manifestations 1912 compression versus decompression 2400 course-onset 2399 demyelinated nerve fibres 2368 diagnostic criteria 2476 differential diagnosis 2421 distribution 2399 DREZ 2405 epidemiology 2363 Gamma Knife surgery 2475 genetics 2363 hereditary factors 2364 history of surgery 2400 history of the REZ 2374 hypermyelinisation 2371 indication for glycerol rhizolysis 2430 investigation 1913 key clinical features 2398 lessons from medical treatment 2403 lessons from surgical treatment 2400 microvascular decompression 1911, 2465 microvascular decompression, complications 1917, 2472 microvascular decompression, investigation 2467 microvascular decompression, management of recurrent cases 2472 microvascular decompression, operative technique 2467 microvascular decompression, patient selection 2466 microvascular decompression, surgical outcomes 2470 migration 2399 multiple sclerosis 2371, 2373, 2522 natural history 2398 neural transplantation 2406 neurophysiology 2391 neurophysiology, abnormal mechanisms, primarily central 2393 neurophysiology, abnormal mechanisms, primarily peripheral 2391 neurophysiology, clinical electrophysiology 2395 neurophysiology, ectopic impulse generation 2392 neurophysiology, ignition hypothesis 2397 neurophysiology, normal physiology 2391 neurophysiology, pharmacological agents that relieve pain 2395 neurophysiology, sensory loss in TN 2396 pain-relieving drugs 2405 pathophysiology 1911, 2359 patient selection for balloon compression 2457
3285
3286
Subject index
– – – – – – – –
pharmacotherapy 2422 physiological background 2429 plexiform microneuroma 2367 pre-trigeminal neuralgia 2398 radiation 2399 radiofrequency rhizotomy 2421 radiosurgery 1027 rationale of percutaneous radiofrequency rhizotomy 2423 – spontaneous remissions 2399, 2407 – stimulation techniques to treat pain 2402 – surgery, intraoperative findings 1916 – surgery, microvascular decompression 1916 – surgery, selection of patients 1913 – surgical technique 1914 – surgical treatment 2422 – technique of retrogasserian glycerol injection 2431 – the future 2404 – trauma to the trigeminal nerve 2407 – trigeminal root transition zone 2369 – trigeminal rootlets 2370 – trigeminal sensory REZ 2373 – ultrastructural examinations of the REZ 2375 trigeminal nuclear structures, anatomophysiological aspects 2097 trigeminal nucleotomy 2101 trigeminal nucleotractotomy – surgical techniques 2120 – technique 2097 – with cysternography 2106 – with stereotomography 2107, 2108 trigeminal pathway, interruption 2526 trigeminal rhizolysis – histological data of mechanisms of glycerol 2449 – mechanisms of glycerol 2448 – neurophysiological data of mechanisms of glycerol 2450 trigeminal rhizotomy 2103 trigeminal root 2098 trigeminal system – functional anatomy 2382 – normal anatomy 2366 – normal and pathological anatomy 2365 – pathological anatomy, brainstem 2369 – pathological anatomy, ganglion 2366 – pathological anatomy, peripheral nerve 2366 – pathological anatomy, roots 2366 trigeminal tract 2100, 2126 trigeminal tractotomy 2101 – complications 2103, 2104 – problems trigeminal vascular system 2488 trigeminothalamic tract (TrT) 1295
trihexyphenidyl 1516, 1776, 1865 Trilogy device 874, 875 trunnion system 903 – treatment ranges 925 Tsui score 1892 tuberous sclerosis complex (TSC) 3209 tumor – biopsy 808, 809 – biopsy intraventricular 811 – biopsy kit 513 – cell survival 1115 – control probability (TCP) 1115 – cyst aspiration 791 – decompression 737 – growth control 1013 – in the pineal region 1058 – intraventricular 694 – metastatic 690, 692 – monitor 730 – nasopharyngeal 1128 – non-neurosurgical 1127 – pituitary 719 – progression, radiographic 1215 – radiosurgery 1012 – recurrent 317 – resection 712 – resection slices 729 – resection of intra-axial 570 – suppressor gene 3087 – surgery 299 – type, imaging 705 Turcot’s syndrome 3211 Turnbull, Frank 121 twist-drill craniotomy set 477 tyrosine kinase inhibitor 3220
U ultrasonography 706 ultrasound 820 – intraoperative 731 – navigational 3-D, intraoperative 302 – triangulation 24 Unified Parkinsons’ disease rating scale (UPDRS) 139 Unimation PUMA (Programmable Universal Machine for Assembly) 592 United States 45 University of Sa˜o Paulo 199 urinary bladder function 2999 urinary disturbances 1518 urine free cortisol (UFC) 1176 urodynamic assessment 3004 urodynamic parameters 3004 urogenital tract – central pathways 3001 – innervation 3000 uveal melanoma 1059
Subject index
V vaccinia virus 3099 vagus nerve stimulation 240, 1336, 2801, 3188 – clinical utility 2810 – complications 2809 – device components 2803 – lead 1409 – lead removal 2809 – off-label use 2814 – patient selection 2813 – refractory epilepsy 1405 – relevant anatomy 2806 – resistant major depression 1405 – revision 2809 – schematic representation 2803 – surgical insertion 2805 vagus nerve trunk, helical electrodes 2808 valproic acid 2496, 2673 van Buren 357 variability map 359 variability profile 355 Varian Trilogy image-guided radiotherapy 874 – efficacy 876 – performance features 874 – quality 876 Varma’s foramen ovale thalamotom’ 162 vascular brain disorders, Gamma Knife radiosurgery 1009 vascular collision 503 vascular compression (VC) 1911 – of trigeminal nerve 2459 vascular endothelial growth factor (VEGF) 746, 751, 753, 997, 3166 – receptor (VEGFR) 752 vascular injury 502, 2690 vascular lesion, peri-rolandic 2620 vascular malformation 693, 1059, 2642 – radiosurgery 1007 vatalanib 754, 998 Vector Vision 703 – mobile optical tracking unit 569 vegetative state patients, deep brain stimulation 2982 Vein of Galen malformation (VGM) 1070 venous angioma 2642 venous compression, endoscopy view 1920 ventral intermediate nucleus (Vim) 1488, 1746 – kinesthetic neurons 1553 – neurons 1554 – stimulation 1351, 1757 – surgery for lesioning 1749 – target planning 1633 – thalamic deep brain stimulation, essential tremor 1405 – thalamic deep brain stimulation, Parkinsonian tremor 1405 – thalamotomy 1757
– thalamotomy, selective, complications 1557 – thalamus (Vim) – tremor neurons 1553 ventral striatum 2929 ventricle 707 ventricular fluid 2057 ventricular landmarks, determination 516 ventriculogram 1611 ventriculograph 22 ventriculography 198, 251, 1809 ventriculostomy catheter 809 ventrolateral column 2001 ventromedial mesencephalic lesion 1665 verapamil 2499 verbal memory 2694 Verelan 2499 Vernier scale 547 vesicular monoamine transporter (VMAT) 3166 vesicular stomatitis virus (VSV) 3101 vestibular pathway 1298 vicious cycle 2067, 2210 video camera 24, 726, 728 video display 704, 819 video EEG 2687 – recording 2796 videotactic surgery 731 Viewscope 703, 709 Vilela Filho, Osvaldo 222 vincristine 743 viral vector 759, 1719, 1720, 3062 virtual fluoroscopy 822 virtual reality 725 – image guidance, benefits 731 – image guidance, development 730 – image guided surgery 730 – surgery 725 visual aura 2563 visual evoked potential (VEP) 1255 visual loss 1183 visual pathway 1298 visual prosthesis 3009 visualization 555, 561 visuotopic map 3014 vocal cord paralysis 2809 vocal tics 1458 voiding symptom 3003 voltammetry, fast-scan cyclic (FSCV) 641 volume effects 862 volume in space 680 voluntary cells 1292 voluntary contraction, thalamic stimulation 1302 Von Hippel Landau protein 997 Von Hippel Lindau disease 3210 voxel size places 819 voxel-based similarity registration 339 Voxim 1612
3287
3288
Subject index
W Wada test 1337, 2578, 2685, 2686 Wahren 363 Walk Aide foot drop stimulator 3052 Walker pedunculotomy 11 Walker’s parcellation 396 Walsh, Lawrence 158 warping 396 water 308 Watkins 354 Watts, James 2871 Wells, Trent 453 , 454, 460, 462 Wernicke’s area 717 white matter impedance 1326 whole body radiosurgery (WBRS) 1203 whole brain radiotherapy (WBRT) 833, 834, 840, 841, 1022, 1055, 1124, 1140, 1143 – radiosurgery vs surgery 1141 Wilcoxon signed rank test 3004 Williams, Dennis 158 Wilson’s disease 1805, 1836 workstation, neurosurgical 429 World Federation of Neurosurgical Societies (WFNS) 35, 37 WSSFN 36–39, 43, 44, 120, 135, 1421 World Society for Stereotactic and Functional Neurosurgery (WSSFN) 9, 37, 135, 210 – presidents prior to 1993 39 worn ear level process 3030 wound infection 2755 Wycis, Henry T. 5, 35, 36, 46, 348
X X-irradiation, effects 1112 XKnife 869
– efficacy 870 – performance features 870, 872 – quality 870 – treatment plan 872 X-ray 853 – angiogram 503 – angiolocalizer 505 – camera 1208 – setup 503 X sight spine tumor 953 XZ-I to XZ-V brain stereotactic apparatus 131
Y Yale-Brown Obsessiv Compulsive Scale (YBOCS) 1197 Yale Global Tic Severity Scale (YGTSS) 1459 Yi-Cheng Zhao 127, 128 Yonsei epilepsy protocol 175 Young’s modulus 637 Yu Fu 125
Z Zaclis, Jose´ 208, 209 Zamorano-Dujovny arc 487 Zamorano-Dujovny system 488 – fixation of the aiming bow to the base ring 491 – open base ring 492 – set-up 491 Zeiss MKM – stereotactically guided microscope 591 – system 703 Zelapar Zhong-Cheng Wang 137 zonisamide 1744, 2673